/
dump_file_test
1000 lines (1000 loc) · 949 KB
/
dump_file_test
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
35|Navigator does not free preference hash table when exit.|Created by (weitsang@cs.cornell.edu) on Monday, April 6, 1998 6:37:03 PM PDT NEWLINE Additional Details : NEWLINE PREF_Cleanup is not called when Navigator exits NEWLINE (File->Exit), causing memory leaks. NEWLINE Updated by Sarah Wilson (swilson@netscape.com) on Tuesday, April 7, 1998 6:48:16 PM PDT NEWLINE Updated by Sarah Wilson (swilson@netscape.com) on Tuesday, April 7, 1998 6:49:27 PM PDT
36|Floating Point Exception on Execution|Created by J. Daniel Powell (dan@java-linux.org) on Monday, April 6, 1998 7:04:03 PM PDT NEWLINE Additional Details : NEWLINE After compilation on a Linux 2.0.33 Alpha-Linux machine with NEWLINE Motif-1.2 and gcc-2.8.1, Mozilla immediately halts with a NEWLINE floating point exception.
37|Preference Dialog does not show|Created by Chen Ronghua (chenrh@usa.net) on Monday, April 6, 1998 7:20:01 PM PDT NEWLINE Additional Details : NEWLINE When I complied the code and ran it, the preference NEWLINE dialog can show the first time. Then I never shown. NEWLINE Every time I fire edit/preference, it gives me NEWLINE "Unable to load the preference DLLs, Please re-installed NEWLINE Mozilla Communicator" NEWLINE Updated by (blythe@netscape.com) on Tuesday, June 9, 1998 4:10:04 PM PDT NEWLINE Additional Details : NEWLINE I fixed this. NEWLINE NEWLINE The problem is that the pref DLLs have relative paths. When switching the NEWLINE default directory, as saving a file would, then you couldn't load. NEWLINE NEWLINE I simply added a function FEU_CoCreateInstance which will switch to the NEWLINE mozilla.exe directory before creating the preference object. NEWLINE NEWLINE garrett
38|Close Mozilla lead to a Assert Failed|Created by Chen Ronghua (chenrh@usa.net) on Monday, April 6, 1998 7:30:03 PM PDT NEWLINE Additional Details : NEWLINE When we exit the Mozilla in normal state, it gives a NEWLINE Assert Failed, on NS\\LIB\\LIBNET\\MKGETURL.c Ln 5855, NEWLINE NEWLINE Additionally, I use NT4.0 Workstation Simplify Chinese NEWLINE Version. NEWLINE Updated by (gagan@netscape.com) on Wednesday, May 6, 1998 8:38:06 PM PDT NEWLINE Additional Details : NEWLINE Fix checked in.
39|Bookmark properties leads to an Assert failed|Created by Chen Ronghua (chenrh@usa.net) on Monday, April 6, 1998 7:34:14 PM PDT NEWLINE Additional Details : NEWLINE Show the Navigator Center, choose Bookmark Button, NEWLINE select one item in it, right click and choose "properties" NEWLINE menu item, it gives an NEWLINE Assert Failed, on NS\\MODULES\\RDF\\SRC\\ht.c Ln 3265. NEWLINE NEWLINE Additionally, I use NT4.0 Workstation Simplify Chinese NEWLINE Version.
41|Multiple Netscape windows cannot recieve data at the same time|Created by Chris Kennedy (chris@groovy.org) on Monday, April 6, 1998 10:05:25 PM PDT NEWLINE Additional Details : NEWLINE I am using Linux 2.0.33 libc 5.4.33 and motif 2.0, also used NEWLINE motif 1.2 with build. Used -O for optimization and also NEWLINE both statically/dynamically linked Motif with both NEWLINE moz-export and mozilla-export. When using two netscape NEWLINE windows or one netscape window and a download window the NEWLINE browser will not click urls properly if downloading of data NEWLINE is taking place in the other window. It strangely will work NEWLINE if I click the link with the right mouse button instead of NEWLINE the left which will only highlight the link like it has been NEWLINE visited and wait untill the other is done downloading or NEWLINE browsing. The menu still pops up for the right button but NEWLINE it will work when the left one will not. Also if I click NEWLINE the middle button(3 button mouse) it cancels the link NEWLINE highlight mentioned above when the left button was clicked NEWLINE on the link.
42|navigator redraw after initial startup|Created by Stephan Nagy (steph8@flash.net) on Monday, April 6, 1998 10:42:04 PM PDT NEWLINE Additional Details : NEWLINE Using Lesstiff .83 and gcc 2.8.1 built on a FreeBSD 2.2.6 NEWLINE box. Navigator doesn''t draw on the top 1/4 of the window. NEWLINE This is probably due to lesstif,but thought i''d report it NEWLINE anyway.
43|Just testing|Created by Aleksandar Totic (atotic@netscape.com) on Monday, April 6, 1998 11:53:01 PM PDT NEWLINE Additional Details : NEWLINE Test by atotic
51|Resizing communicator changes display format of the <PRE></PRE> tag|Created by Svein Erik Brostigen (sbrostig@no.oracle.com) on Tuesday, April 7, 1998 1:22:31 AM PDT NEWLINE Additional Details : NEWLINE When resizing the communicator window after loading the NEWLINE unxbuild.htm document, the part of the document within the NEWLINE <PRE></PRE> tags are drawn on just 2 lines. NEWLINE A reload of the document will draw the doscument correctly. NEWLINE Updated by null (null) on Wednesday, April 22, 1998 9:18:04 PM PDT NEWLINE Additional Details : NEWLINE Am assigning this to toshok. I think you already fixed this problem, right? NEWLINE Updated by (toshok@netscape.com) on Monday, May 18, 1998 11:44:35 AM PDT NEWLINE Additional Details : NEWLINE nope, this is a different. bug. the one i fixed was PRE regions that would get NEWLINE larger and larger (more newlines inserted) on every resize. NEWLINE NEWLINE will take a look. NEWLINE Updated by (toshok@netscape.com) on Monday, May 18, 1998 11:47:30 AM PDT NEWLINE Additional Details : NEWLINE changing os/platform since it happens on at least another one (freebsd) NEWLINE Updated by (toshok@netscape.com) on Tuesday, June 30, 1998 12:34:26 PM PDT NEWLINE Additional Details : NEWLINE So, this bug has been fixed for a while -- just closing the bug now.
61|Navigator shutdown set-zero-context bug|Created by Jukka Santala (donwulff@iki.fi) on Tuesday, April 7, 1998 4:56:14 AM PDT NEWLINE Additional Details : NEWLINE The call stack backtrace tells more than a thousand words : NEWLINE (It''s an assert instead of prot. fault because this has NEWLINE the below patch compiled in; it doesn''t affect the actual NEWLINE bug, just lets me recover from the error gracefully) NEWLINE NEWLINE XP_AssertAtLine(char * 0x00866778, int 63) line 2649 NEWLINE CFormButton::SetContext(CAbstractCX * 0x00000000) line 63 + NEWLINE 36 bytes NEWLINE CFormElement::GetFormElement(CAbstractCX * 0x00000000, NEWLINE LO_FormElementData_struct * 0x00a5c8c0) line 135 NEWLINE FE_FreeFormElement(MWContext_ * 0x009ebd30, NEWLINE LO_FormElementData_struct * 0x00a5c8c0) line 720 + 11 bytes NEWLINE lo_FreeDocumentFormListData(MWContext_ * 0x009ebd30, NEWLINE lo_SavedFormListData_struct * 0x00a2d5e0) line 858 + 19 NEWLINE bytes NEWLINE LO_FreeDocumentFormListData(MWContext_ * 0x009ebd30, void * NEWLINE 0x00a2d5e0) line 6387 + 13 bytes NEWLINE SHIST_FreeHistoryEntry(MWContext_ * 0x009ebd30, NEWLINE _History_entry * 0x00a29c30) line 101 + 16 bytes NEWLINE SHIST_EndSession(MWContext_ * 0x009ebd30) line 438 + 13 NEWLINE bytes NEWLINE CAbstractCX::DestroyContext() line 116 + 740 bytes NEWLINE CDCCX::DestroyContext() line 186 NEWLINE CPaneCX::DestroyContext() line 304 NEWLINE CWinCX::DestroyContext() line 619 NEWLINE CGenericView::FrameClosing() line 215 NEWLINE CMainFrame::OnClose() line 660 NEWLINE CWnd::OnWndMsg(unsigned int 16, unsigned int 0, long 0, long NEWLINE * 0x0012fd3c) line 1815 NEWLINE CWnd::WindowProc(unsigned int 16, unsigned int 0, long 0) NEWLINE line 1575 + 30 bytes NEWLINE AfxCallWndProc(CWnd * 0x00399ff0 {CMainFrame NEWLINE hWnd=0x0009035a}, HWND__ * 0x0009035a, unsigned int 16, NEWLINE unsigned int 0, long 0) line 217 + 26 bytes NEWLINE AfxWndProc(HWND__ * 0x0009035a, unsigned int 16, unsigned NEWLINE int 0, long 0) line 371 NEWLINE AfxWndProcBase(HWND__ * 0x0009035a, unsigned int 16, NEWLINE unsigned int 0, long 0) line 203 + 21 bytes NEWLINE USER32! 77e72cc4() NEWLINE USER32! 77e72ce2() NEWLINE CNetscapeStatusBar::CParentSubclass::ParentSubclassProc(HWND NEWLINE __ * 0x0009035a, unsigned int 16, unsigned int 0, long 0) NEWLINE line 1077 + 29 bytes NEWLINE USER32! 77e713ed() NEWLINE NEWLINE Temporary patch (Munched by the form): NEWLINE NEWLINE diff --unified --recursive mozorg/ns/cmd/winfe/fmbutton.cpp NEWLINE mozilla/ns/cmd/winfe/fmbutton.cpp NEWLINE --- mozorg/ns/cmd/winfe/fmbutton.cpp Mon Mar 30 17:57:41 NEWLINE 1998 NEWLINE +++ mozilla/ns/cmd/winfe/fmbutton.cpp Mon Apr 06 18:42:09 NEWLINE 1998 NEWLINE @@ -60,8 +60,9 @@ NEWLINE // Call the base. NEWLINE CFormElement::SetContext(pCX); NEWLINE NEWLINE + XP_ASSERT(pCX); NEWLINE //Update the widget if present for correct callbacks. NEWLINE - if(m_pWidget) { NEWLINE + if(m_pWidget&&pCX) { NEWLINE m_pWidget->RegisterContext(pCX->GetContext()); NEWLINE } NEWLINE } NEWLINE NEWLINE This will solve the immediate problem, but as to how and NEWLINE why pCX gets to be null is another story.
62|Composer ROWSPAN problem|Created by Jukka Santala (donwulff@iki.fi) on Tuesday, April 7, 1998 5:08:08 AM PDT NEWLINE Additional Details : NEWLINE Sometimes when the ROWSPAN attribute in a table has been NEWLINE set past the number of rows in a table, Composer will end NEWLINE up with memory corruption. This is due to a function using NEWLINE the number of rows to initialize an array. NEWLINE NEWLINE In particular, the following HTML code will demonstrate: NEWLINE NEWLINE <TABLE BORDER=0 CELLPADDING=0 CELLSPACING=0 WIDTH="100%"> NEWLINE <TR><TD BGCOLOR="#000000" VALIGN=TOP ROWSPAN=2> NEWLINE <IMG SRC="images/mozilla-banner.jpg" NEWLINE ALT="" BORDER=0 USEMAP="#banner" NEWLINE WIDTH=600 HEIGHT=50 VSPACE=0 HSPACE=0> NEWLINE </TD></TR></TABLE> NEWLINE NEWLINE From a netscape rep. suggested to use a re-sizable array. NEWLINE From personal view over the code, it seems that the extra NEWLINE array positions arne''t actually needed since the spanned NEWLINE rows don''t get visible, though, so ending the loop once NEWLINE the existing rows have been processed is another NEWLINE possibility. It''s also within possibility that the row NEWLINE counter method needs fixing; need to dig deeper into the NEWLINE code. NEWLINE NEWLINE The following patch solved the problem for debugging: NEWLINE NEWLINE diff --unified --recursive mozorg/ns/lib/layout/edtbuf.cpp NEWLINE mozilla/ns/lib/layout/edtbuf.cpp NEWLINE --- mozorg/ns/lib/layout/edtbuf.cpp Mon Mar 30 17:55:37 NEWLINE 1998 NEWLINE +++ mozilla/ns/lib/layout/edtbuf.cpp Sat Apr 04 21:35:37 NEWLINE 1998 NEWLINE @@ -1860,8 +1860,11 @@ NEWLINE // then it NEWLINE will cause extra columns in following row(s) NEWLINE if( iRowSpan > 1 ) NEWLINE { NEWLINE - for( intn j = 1; j < iRowSpan; NEWLINE j++ ) NEWLINE - ExtraColumns[iRow+j] += NEWLINE iColSpan; NEWLINE + for( intn j = 1; j < iRowSpan; NEWLINE j++ ) { NEWLINE + NEWLINE if((iRow+j)>=iRows) XP_TRACE(("CEditBuffer::FixupTableData() NEWLINE inconsistency!")); NEWLINE + NEWLINE else ExtraColumns[iRow+j] += iColSpan; NEWLINE + NEWLINE XP_TRACE(("Did stuff")); NEWLINE + } NEWLINE } NEWLINE // Save actual location and size NEWLINE data NEWLINE EDT_TableCellData *pCellData = NEWLINE pEdCell->GetData(); NEWLINE NEWLINE This patch is incomplete, as it doesn''t deal with the NEWLINE side-effects of the loop which may do other nasty stuff, NEWLINE however it''s useful in debugging the problem for me. NEWLINE Updated by (brade@netscape.com) on Thursday, April 16, 1998 8:52:10 PM PDT NEWLINE Additional Details : NEWLINE Reassign bug to myself since I'm the owner of this file. NEWLINE I'm guessing that this is really cross-platform (ALL) but I haven't tested it on NEWLINE any platform to confirm yet.
63|NavCenter closing up garbage-resource-string|Created by Jukka Santala (donwulff@iki.fi) on Tuesday, April 7, 1998 5:13:38 AM PDT NEWLINE Additional Details : NEWLINE Sometimes (don NEWLINE Updated by Mike Pinkerton (pinkerton@netscape.com) on Tuesday, April 7, 1998 8:40:37 AM PDT NEWLINE Additional Details : NEWLINE This is probably cross platform. I have seen that on my mac builds as well. NEWLINE Changing platform/OS to "ALL". NEWLINE Updated by Jukka Santala (donwulff@iki.fi) on Wednesday, April 8, 1998 6:00:55 PM PDT NEWLINE Additional Details : NEWLINE For some reason, adding the previous comment ate most of the original problem NEWLINE description; consider that a bug report for the Bugzilla code. In short though, NEWLINE /modules/rdf/src/nlcstore.c has at position 210 rows into itself code, where NEWLINE during Navigator shutdown resourceID(u) (which expands to ->url macro) sometimes NEWLINE returns "garbage". At first I thought it was just a NULL pointer, but after NEWLINE adding XP_ASSERT()''s for this case, Navigator crashed again, this time with NEWLINE resourceID(u)==0xdddddddd
71|assertion failed in mozilla.c|Created by Svein Erik Brostigen (sbrostig@no.oracle.com) on Tuesday, April 7, 1998 5:50:41 AM PDT NEWLINE Additional Details : NEWLINE When running Mozilla the assertion in line 997 in mozilla.c NEWLINE fails. The variable "spinning_wildly" is trigging an NEWLINE assertion failed when spinning_wildly > 12. I have done NEWLINE some debugging here and found that a value of 30 is too NEWLINE little. This is not a serious bug though. NEWLINE Btw, we need an IRIX 6.4 OS version entry here. NEWLINE Updated by Terry Weissman (terry@netscape.com) on Tuesday, April 7, 1998 8:52:01 AM PDT NEWLINE Additional Details : NEWLINE IRIX 6.4 now exists in the database; this bug has been changed to use it.
81|Navigator crashes when loading a page with BORDER-WIDTH and BORDER CSS|Created by Marcos Hauer (mhauer@iis.com.br) on Tuesday, April 7, 1998 6:48:04 AM PDT NEWLINE Updated by null (null) on Wednesday, April 22, 1998 9:20:26 PM PDT NEWLINE Additional Details : NEWLINE Assigning to David as he is in charge of CSS stabilization
82|Navigator crashes when loading a page with BORDER-WIDTH and WIDTH CSS|Created by Marcos Hauer (mhauer@iis.com.br) on Tuesday, April 7, 1998 6:57:17 AM PDT NEWLINE Additional Details : NEWLINE Both Win95 and Linux versions crash ( GPF and Seg Fault ) NEWLINE when loading a page that contains one CSS entity with BORDER NEWLINE and BORDER[-top,-left,-right,-bottom]-WIDTH and at the same NEWLINE time, like the example below: NEWLINE NEWLINE #borderblock { NEWLINE ... NEWLINE border-width: 2; NEWLINE ... NEWLINE width: 300; NEWLINE } NEWLINE NEWLINE I NEWLINE Updated by Marcos Hauer (mhauer@iis.com.br) on Tuesday, April 7, 1998 7:10:00 AM PDT NEWLINE Additional Details : NEWLINE Confirming, the error occurs with WIDTH and BORDER-WIDTH in the same entity, and NEWLINE not with BORDER and BORDER-WIDTH as said in the text above ( the example is NEWLINE right ). NEWLINE Updated by Marcos Hauer (mhauer@iis.com.br) on Wednesday, April 8, 1998 1:00:00 PM PDT NEWLINE Additional Details : NEWLINE I have found references linking WIDTH and BORDER-WIDTH as a ''shorthand'' NEWLINE variable in the files ns/lib/libstyle/csstab.c and ns/lib/libstyle/css.y, but NEWLINE I''m not sure if it''s related with the error. NEWLINE Updated by Nisheeth Ranjan (nisheeth@netscape.com) on Wednesday, April 22, 1998 9:31:15 PM PDT NEWLINE Additional Details : NEWLINE Assigning to djw as he is in charge of CSS stabilization.
91|document properties cannot be listed|Created by Till Krech (till@berlin.snafu.de) on Tuesday, April 7, 1998 9:05:23 AM PDT NEWLINE Additional Details : NEWLINE the JavaScript "for in" statement does not work on the NEWLINE document object. At least not in the Linux version.
92|Incorrect Link Underlining|Created by Till Krech (till@berlin.snafu.de) on Tuesday, April 7, 1998 9:09:24 AM PDT NEWLINE Additional Details : NEWLINE Linux X11 version: NEWLINE Sometimes, links aren''t underlined but "overlined" i.e. the NEWLINE line appears obove the link. NEWLINE Updated by Nisheeth Ranjan (nisheeth@netscape.com) on Wednesday, April 22, 1998 9:35:17 PM PDT NEWLINE Additional Details : NEWLINE Assigning this to toshok as he was looking at this for XFE. Till, can you NEWLINE please give us some more information so that we have a good test case to debug NEWLINE with? Thanks. NEWLINE NEWLINE Nisheeth NEWLINE Updated by (toshok@netscape.com) on Monday, May 18, 1998 11:46:06 AM PDT NEWLINE Additional Details : NEWLINE changing platform to X-windows so I don't get confused. NEWLINE Updated by (toshok@netscape.com) on Monday, May 18, 1998 1:45:30 PM PDT NEWLINE Additional Details : NEWLINE just checked in a fix for this in cmd/xfe/lay.c. please update and see if the NEWLINE problem is still there. if so, please reopen this bug. thanks. NEWLINE Updated by Felix (thehelix@yahoo.com) on Friday, August 14, 1998 7:09:49 AM PDT NEWLINE Additional Details : NEWLINE Bug is contained on the last Mozilla release as well (downloaded 12. Aug. 98). NEWLINE I'm on a SPARC running Solaris 2.5.1. NEWLINE Updated by Felix (thehelix@yahoo.com) on Friday, August 14, 1998 7:10:02 AM PDT NEWLINE Additional Details : NEWLINE Bug is contained on the last Mozilla release as well (downloaded 12. Aug. 98). NEWLINE I'm on a SPARC running Solaris 2.5.1. NEWLINE Updated by Felix (thehelix@yahoo.com) on Friday, August 14, 1998 7:10:27 AM PDT NEWLINE Additional Details : NEWLINE Bug is contained on the last Mozilla release as well (downloaded 12. Aug. 98). NEWLINE I'm on a SPARC running Solaris 2.5.1.
101|Communicator crashes when window is opened, early into page loading.|Created by Peter Gregg (gregg@linkup.net) on Tuesday, April 7, 1998 9:54:50 AM PDT NEWLINE Additional Details : NEWLINE Communicator give me error message: NEWLINE NEWLINE NETSCAPE caused an invalid page fault in NEWLINE module NETSCAPE.EXE at 0137:00436709. NEWLINE Registers: NEWLINE EAX=00000038 CS=0137 EIP=00436709 EFLGS=00010202 NEWLINE EBX=00000000 SS=013f ESP=009ff5b0 EBP=009ff5d8 NEWLINE ECX=7803c3c8 DS=013f ESI=00a70198 FS=117f NEWLINE EDX=00000001 ES=013f EDI=00bf4a90 GS=0000 NEWLINE Bytes at CS:EIP: NEWLINE f6 40 01 04 75 35 8d 6e 38 8b 45 00 85 c0 74 2b NEWLINE Stack dump: NEWLINE 009ff5d8 00a70198 00a70dbc 00bf4a90 0045ed18 00bf4a90 NEWLINE 00a70198 00000000 00c48c20 00c48c20 009ff610 00445e65 NEWLINE 00bf4a90 00c48c20 00a70dbc 00bf4a90 NEWLINE NEWLINE Or something similar. Exact stack dump, etc, are different NEWLINE each time. Causes complete closing of all Netscape windows NEWLINE open, even if only one page is loading.
102|Alt-w doesn''t close view-source window|Created by Darrell Kindred (dkindred=mozilla@cs.cmu.edu) on Tuesday, April 7, 1998 10:57:25 AM PDT NEWLINE Additional Details : NEWLINE Pressing Alt-w seems to have no effect in the Page Source NEWLINE window. NEWLINE Updated by Steve Lamm (slamm@netscape.com) on Thursday, April 30, 1998 10:26:04 AM PDT NEWLINE Additional Details : NEWLINE I have this on my internal buglist too, so I will take it.
103|layout bug: table cell overflows containing cell|Created by Darrell Kindred (dkindred=mozilla@cs.cmu.edu) on Tuesday, April 7, 1998 11:43:26 AM PDT NEWLINE Additional Details : NEWLINE The nested table here is laid out wrong: NEWLINE NEWLINE http://www.cs.cmu.edu/People/dkindred/bugs/table-layout.html NEWLINE NEWLINE A table cell extends outside its containing cell. NEWLINE Possible contributing factor: WIDTH=300 on outer table, NEWLINE WIDTH=100% on TD within it. NEWLINE NEWLINE [Linux, Motif 1.2, Helvetica fonts] NEWLINE Updated by Nisheeth Ranjan (nisheeth@netscape.com) on Wednesday, April 22, 1998 9:41:35 PM PDT NEWLINE Additional Details : NEWLINE Thanks Darrell, for providing a great test case. Hopefully, we'll have a public NEWLINE CVS repository soon and you can update to get this fix.
104|table saved as text: missing inter-column space|Created by Darrell Kindred (dkindred=mozilla@cs.cmu.edu) on Tuesday, April 7, 1998 12:16:02 PM PDT NEWLINE Additional Details : NEWLINE Visit this document (which contains a table), NEWLINE and save as "text": NEWLINE NEWLINE http://www.cs.cmu.edu/People/dkindred/bugs/table-text-bug.html NEWLINE NEWLINE The original table (in html) looks like this: NEWLINE NEWLINE <TABLE> NEWLINE <TR><TH>foo</TH><TH>bar</TH><TH>baz</TH><TH>qux</TH></TR> NEWLINE <TR><TD>abc</TD><TD>def</TD><TD>ghi</TD><TD>jkl</TD></TR> NEWLINE <TR><TD>abc</TD><TD>def</TD><TD>ghi</TD><TD>jkl</TD></TR> NEWLINE <TR><TD>abc</TD><TD>def</TD><TD>ghi</TD><TD>jkl</TD></TR> NEWLINE </TABLE> NEWLINE NEWLINE The table in the text file looks like this: NEWLINE NEWLINE foobar baz qux NEWLINE abcdef ghi jkl NEWLINE abcdef ghi jkl NEWLINE abcdef ghi jkl NEWLINE NEWLINE (Note the missing space between the first two columns.)
105|nested <TABLE>s: bgcolor of inner table not recognized|Created by (msuencks@marcant.de) on Tuesday, April 7, 1998 12:54:41 PM PDT NEWLINE Additional Details : NEWLINE The following HTML Code is not properley drawn: NEWLINE NEWLINE <body bgcolor=white> NEWLINE <table bgcolor=blue> NEWLINE <tr><td> NEWLINE <table bgcolor=yellow> NEWLINE <tr><td>hello NEWLINE </table> NEWLINE </table> NEWLINE </body> NEWLINE Updated by Nisheeth Ranjan (nisheeth@netscape.com) on Wednesday, April 22, 1998 9:55:02 PM PDT NEWLINE Additional Details : NEWLINE This bug is fixed. Hopefully we'll have a public CVS repository soon and you NEWLINE can update your tree to get the fix. Thanks for the bug report and the great NEWLINE test case.
106|X selection fails in presence of PRE|Created by Darrell Kindred (dkindred=mozilla@cs.cmu.edu) on Tuesday, April 7, 1998 1:40:36 PM PDT NEWLINE Additional Details : NEWLINE In the following document, I can''t make an X selection NEWLINE by dragging button1. If I remove the PRE section, NEWLINE it works. NEWLINE NEWLINE http://www.cs.cmu.edu/People/dkindred/bugs/x-select.html NEWLINE NEWLINE [Linux, Motif 1.2.4] NEWLINE Updated by Nisheeth Ranjan (nisheeth@netscape.com) on Wednesday, April 22, 1998 10:01:07 PM PDT NEWLINE Additional Details : NEWLINE Am assigning this to myself for further investigation...
107|OnMouseOver window.status setting -- url appears first|Created by Darrell Kindred (dkindred=mozilla@cs.cmu.edu) on Tuesday, April 7, 1998 2:24:59 PM PDT NEWLINE Additional Details : NEWLINE If an OnMouseOver handler on an anchor sets window.status, NEWLINE the URL appears first for an instant, and is then NEWLINE overwritten by the message the handler installs. You can NEWLINE see this at NEWLINE http://www.cs.cmu.edu/People/dkindred/bugs/mouseover.html NEWLINE NEWLINE This happens in 4.05 also, but not in 3.04.
108|compile problem, error is in Grid.c|null NEWLINE Updated by Justin Burket (zorton@thepentagon.com) on Wednesday, April 8, 1998 4:37:28 PM PDT NEWLINE Additional Details : NEWLINE NEWLINE Solved (thanks Marcus Butler), change was to just include Xm/DropSMgr.h in the NEWLINE file
109|TESTING PLEASE IGNORE|Created by Sarah Wilson (swilson@netscape.com) on Tuesday, April 7, 1998 6:24:44 PM PDT NEWLINE Additional Details : NEWLINE just a test for QA purposes. NEWLINE Updated by Sarah Wilson (swilson@netscape.com) on Tuesday, April 7, 1998 6:28:12 PM PDT NEWLINE Additional Details : NEWLINE more testing....blah blah NEWLINE Updated by Sarah Wilson (swilson@netscape.com) on Tuesday, April 7, 1998 6:51:53 PM PDT NEWLINE Updated by Arun Ramachandran (arunr@netscape.com) on Tuesday, April 7, 1998 6:58:16 PM PDT NEWLINE Updated by Arun Ramachandran (arunr@netscape.com) on Tuesday, April 7, 1998 6:59:09 PM PDT NEWLINE Updated by Arun Ramachandran (arunr@netscape.com) on Tuesday, April 7, 1998 7:00:08 PM PDT NEWLINE Additional Details : NEWLINE test NEWLINE Updated by Arun Ramachandran (arunr@netscape.com) on Sunday, May 17, 1998 4:50:18 PM PDT NEWLINE Additional Details :
110|"Copy link location" doesn''t|Created by Darrell Kindred (dkindred=mozilla@cs.cmu.edu) on Tuesday, April 7, 1998 10:58:46 PM PDT NEWLINE Additional Details : NEWLINE The "Copy link location" option in the pop-up NEWLINE menu that appears when you press button3 on a NEWLINE link seems to have no effect. (The X selection NEWLINE is unchanged.) This works fine in 4.05. NEWLINE Updated by Steve Lamm (slamm@netscape.com) on Thursday, April 30, 1998 10:33:46 AM PDT NEWLINE Additional Details : NEWLINE My current version doesn't even give the "Copy link location" option when NEWLINE clicking on a link. I will look into this. NEWLINE Note that the X selection should remain unchanged. The "Copy link locaiton" NEWLINE command should copy the link into the copy buffer. Try pasting with Edit->Paste NEWLINE (Alt+V).
111|Add <Q> tag support to Mozilla|Created by Brian Grunkemeyer (bg2k@cmu.edu) on Tuesday, April 7, 1998 11:55:18 PM PDT NEWLINE Additional Details : NEWLINE The quote tag, <q>..</q> isn NEWLINE Updated by Brian Grunkemeyer (bg2k@cmu.edu) on Wednesday, April 8, 1998 9:15:13 PM PDT NEWLINE Additional Details : NEWLINE This was the inline quote tag, Q. Didn''t realize it would NEWLINE munge the HTML so poorly. And look at the actual URL for the NEWLINE HTML spec, don''t read what I have quoted since the tags NEWLINE won''t show up right. NEWLINE Updated by Nisheeth Ranjan (nisheeth@netscape.com) on Sunday, April 26, 1998 6:24:39 PM PDT NEWLINE Additional Details : NEWLINE <Q> is a new tag in HTML 4.0 for marking inline quotations (<BLOCKQUOTE> marks NEWLINE block level quotations), support for which has not been added to Mozilla yet. NEWLINE It should be pretty simple to add, however. NEWLINE NEWLINE Brian, thanks a lots for reporting this bug. Do you want to take a crack at NEWLINE fixing it? If so, I'll email you details of how to start and assign the bug to NEWLINE you. NEWLINE Updated by Nisheeth Ranjan (nisheeth@netscape.com) on Sunday, April 26, 1998 6:25:00 PM PDT NEWLINE Additional Details : NEWLINE <Q> is a new tag in HTML 4.0 for marking inline quotations (<BLOCKQUOTE> marks NEWLINE block level quotations), support for which has not been added to Mozilla yet. NEWLINE It should be pretty simple to add, however. NEWLINE NEWLINE Brian, thanks a lots for reporting this bug. Do you want to take a crack at NEWLINE fixing it? If so, I'll email you details of how to start and assign the bug to NEWLINE you. NEWLINE Updated by Nisheeth Ranjan (nisheeth@netscape.com) on Sunday, April 26, 1998 6:25:24 PM PDT NEWLINE Additional Details : NEWLINE <Q> is a new tag in HTML 4.0 for marking inline quotations (<BLOCKQUOTE> marks NEWLINE block level quotations), support for which has not been added to Mozilla yet. NEWLINE It should be pretty simple to add, however. NEWLINE NEWLINE Brian, thanks a lots for reporting this bug. Do you want to take a crack at NEWLINE fixing it? If so, I'll email you details of how to start and assign the bug to NEWLINE you. NEWLINE Updated by Brian Grunkemeyer (bg2k@cmu.edu) on Thursday, April 30, 1998 11:41:20 PM PDT NEWLINE Additional Details : NEWLINE I think I now have the free time to work on this. But would I be better off NEWLINE fixing this in the current HTML renderer, or in Raptor?
112|make crashes when building zlib looking for PR3240.LIB|Created by Darren Starr (dstarr@geoffreys.com) on Wednesday, April 8, 1998 3:21:31 AM PDT NEWLINE Additional Details : NEWLINE I ran the nmake client.mak, I feel I have everything working NEWLINE properly so far, however, I can''t seem to track down this NEWLINE error after trying to find out where PR3240.LIB is even NEWLINE being created. I''ll continue searching, however I have a NEWLINE detailed make log showing what happened. Please mail me an NEWLINE address where to send this log, it should be helpful in NEWLINE figuring out the build problem
113|Error messages when starting; Composer starts after closing these messages|Created by Thorsten Heit (thorsten@wurzelausix.cs.uni-sb.de) on Wednesday, April 8, 1998 3:29:42 AM PDT NEWLINE Additional Details : NEWLINE When starting the Navigator, I always get two error NEWLINE messages: NEWLINE * One that says something like "cannot open NEWLINE home.netscape.com" or so NEWLINE (shows three URL NEWLINE Updated by Thorsten Heit (thorsten@wurzelausix.cs.uni-sb.de) on Wednesday, April 8, 1998 3:45:15 AM PDT NEWLINE Additional Details : NEWLINE I forgot: NEWLINE This phenomenon can only be seen once: each time when starting the Navigator NEWLINE (from the command line or others).
114|Toolbars cannot be minimized correctly|Created by Thorsten Heit (thorsten@wurzelausix.cs.uni-sb.de) on Wednesday, April 8, 1998 3:37:25 AM PDT NEWLINE Additional Details : NEWLINE Minimizing the toolbar that contains the back, forward, NEWLINE reload ... buttons (and an animated picture on its right NEWLINE side) results in having the animated picture on one of NEWLINE the two other toolbars, perhaps a little bit smaller to NEWLINE have the correct size. Restoring the first toolbar, and NEWLINE I get the original animated logo on its correct position. NEWLINE NEWLINE I cannot minimize all of the three toolbars; I always have NEWLINE one at the top of my Navigator window visible. NEWLINE NEWLINE NEWLINE My system: NEWLINE RedHat Linux 5.0 running on an AMD K6/233 with 128 MB NEWLINE RAM, 4 MB graphics card, 2 GB SCSI hard drive
115|Navigator doesn't build|Created by Thorsten Heit (thorsten@wurzelausix.cs.uni-sb.de) on Wednesday, April 8, 1998 5:02:34 AM PDT NEWLINE Additional Details : NEWLINE The patches in ns/nsprpub/pr/include/md/_linux.* are NEWLINE taken from the SunOS4 files and seem to work: NEWLINE NEWLINE --- ns/nsprpub/pr/include/md/_linux.h.orig Tue Mar 31 NEWLINE 04:05:25 1998 NEWLINE +++ ns/nsprpub/pr/include/md/_linux.h Wed Apr 8 NEWLINE 13:41:18 1998 NEWLINE @@ -29,6 +29,8 @@ NEWLINE #define _PR_SI_ARCHITECTURE "ppc" NEWLINE #elif defined(__alpha) NEWLINE #define _PR_SI_ARCHITECTURE "alpha" NEWLINE +#elif defined(__sparc__) NEWLINE +#define _PR_SI_ARCHITECTURE "sparc" NEWLINE #else NEWLINE #define _PR_SI_ARCHITECTURE "x86" NEWLINE #endif NEWLINE @@ -92,6 +94,12 @@ NEWLINE NEWLINE /* XXX not sure if this is correct, or maybe it should be NEWLINE 17? */ NEWLINE #define PR_NUM_GCREGS 9 NEWLINE + NEWLINE + NEWLINE +#elif defined(__sparc__) NEWLINE +/* Sparc based Linux */ NEWLINE +#define _MD_GET_SP(_t) NEWLINE (_t)->md.context[0].__jmpbuf[0].__fp NEWLINE +#define PR_NUM_GCREGS _JBLEN NEWLINE NEWLINE #else NEWLINE /* Intel based Linux */ NEWLINE --- ns/nsprpub/pr/include/md/_linux.cfg.orig Tue Mar 31 NEWLINE 04:05:25 1998 NEWLINE +++ ns/nsprpub/pr/include/md/_linux.cfg Tue Apr 7 NEWLINE 15:04:35 1998 NEWLINE @@ -116,6 +116,55 @@ NEWLINE #define PR_BYTES_PER_WORD_LOG2 3 NEWLINE #define PR_BYTES_PER_DWORD_LOG2 3 NEWLINE NEWLINE +#elif defined(__sparc__) NEWLINE + NEWLINE +#undef IS_LITTLE_ENDIAN NEWLINE +#define IS_BIG_ENDIAN 1 NEWLINE +#undef HAVE_LONG_LONG NEWLINE +#define HAVE_ALIGNED_DOUBLES NEWLINE +#undef HAVE_ALIGNED_LONGLONGS NEWLINE + NEWLINE +#define PR_BYTES_PER_BYTE 1 NEWLINE +#define PR_BYTES_PER_SHORT 2 NEWLINE +#define PR_BYTES_PER_INT 4 NEWLINE +#define PR_BYTES_PER_INT64 8 NEWLINE +#define PR_BYTES_PER_LONG 4 NEWLINE +#define PR_BYTES_PER_FLOAT 4 NEWLINE +#define PR_BYTES_PER_DOUBLE 8 NEWLINE +#define PR_BYTES_PER_WORD 4 NEWLINE +#define PR_BYTES_PER_DWORD 8 NEWLINE +#define PR_BYTES_PER_WORD_LOG2 2 NEWLINE +#define PR_BYTES_PER_DWORD_LOG2 3 NEWLINE + NEWLINE +#define PR_BITS_PER_BYTE 8 NEWLINE +#define PR_BITS_PER_SHORT 16 NEWLINE +#define PR_BITS_PER_INT 32 NEWLINE +#define PR_BITS_PER_INT64 64 NEWLINE +#define PR_BITS_PER_LONG 32 NEWLINE +#define PR_BITS_PER_FLOAT 32 NEWLINE +#define PR_BITS_PER_DOUBLE 64 NEWLINE +#define PR_BITS_PER_WORD 32 NEWLINE + NEWLINE +#define PR_BITS_PER_BYTE_LOG2 3 NEWLINE +#define PR_BITS_PER_SHORT_LOG2 4 NEWLINE +#define PR_BITS_PER_INT_LOG2 5 NEWLINE +#define PR_BITS_PER_INT64_LOG2 6 NEWLINE +#define PR_BITS_PER_LONG_LOG2 5 NEWLINE +#define PR_BITS_PER_FLOAT_LOG2 5 NEWLINE +#define PR_BITS_PER_DOUBLE_LOG2 6 NEWLINE +#define PR_BITS_PER_WORD_LOG2 5 NEWLINE + NEWLINE +#define PR_ALIGN_OF_SHORT 2 NEWLINE +#define PR_ALIGN_OF_INT 4 NEWLINE +#define PR_ALIGN_OF_LONG 4 NEWLINE +#define PR_ALIGN_OF_INT64 8 NEWLINE +#define PR_ALIGN_OF_FLOAT 4 NEWLINE +#define PR_ALIGN_OF_DOUBLE 8 NEWLINE +#define PR_ALIGN_OF_POINTER 4 NEWLINE + NEWLINE +#define BYTES_PER_WORD_LOG2 PR_BYTES_PER_WORD_LOG2 NEWLINE +#define BYTES_PER_DWORD_LOG2 PR_BYTES_PER_DWORD_LOG2 NEWLINE + NEWLINE #else NEWLINE NEWLINE #define IS_LITTLE_ENDIAN 1 NEWLINE NEWLINE NEWLINE NEWLINE NEWLINE NEWLINE The ns/config/config.guess file file should be changed NEWLINE because all Linux supported platforms are working the NEWLINE same way: NEWLINE NEWLINE --- ns/config/config.guess.orig Wed Apr 8 13:51:27 1998 NEWLINE +++ ns/config/config.guess Wed Apr 8 13:53:26 1998 NEWLINE @@ -196,13 +196,7 @@ NEWLINE i[3456]86:NetBSD:*:*) NEWLINE echo ${UNAME_MACHINE}-unknown-netbsd${UNAME_RELEASE} NEWLINE exit 0 ;; NEWLINE - alpha:Linux:*:*) NEWLINE - echo ${UNAME_MACHINE}-unknown-linux NEWLINE - exit 0 ;; NEWLINE - i[3456]86:Linux:*:*) NEWLINE - echo ${UNAME_MACHINE}-unknown-linux NEWLINE - exit 0 ;; NEWLINE - ppc:Linux:*:*) NEWLINE + *:Linux:*:*) NEWLINE echo ${UNAME_MACHINE}-unknown-linux NEWLINE exit 0 ;; NEWLINE i[3456]86:UNIX_SV:4.*:* \| i[3456]86:SYSTEM_V:4.*:*) NEWLINE Updated by Wan-Teh Chang (wtc@netscape.com) on Friday, April 24, 1998 12:05:36 PM PDT NEWLINE Additional Details : NEWLINE Added bug to myself. I will merge the patches to ns/nsprpub. NEWLINE I will email Brian Ostrom <briano@netscape.com> about the patch to NEWLINE ns/config/config.guess. NEWLINE Updated by Wan-Teh Chang (wtc@netscape.com) on Tuesday, June 16, 1998 8:21:46 PM PDT NEWLINE Additional Details : NEWLINE I believed that Mozilla builds and runs on NEWLINE Sparc Linux now. I ported nsprpub, and NEWLINE Chris McAfee ported the rest of Mozilla. NEWLINE NEWLINE Marked the bug resolved. NEWLINE Updated by Wan-Teh Chang (wtc@netscape.com) on Tuesday, June 16, 1998 8:26:25 PM PDT NEWLINE Additional Details : NEWLINE Closed the bug.
116|Can''t describe an xml file|Created by Asakura Hiroshi (hirosh-a@is.aist-nara.ac.jp) on Wednesday, April 8, 1998 5:38:22 AM PDT NEWLINE Additional Details : NEWLINE If Communicator reads an xml file, it be done with NEWLINE Segmentation fault. NEWLINE Any xml file make Communicator do.
117|Compiling Navigator on SPARC/Linux doesn't compile the c files with "-fPIC" so a .so triggers an ICE|Created by Thorsten Heit (thorsten@wurzelausix.cs.uni-sb.de) on Wednesday, April 8, 1998 6:34:28 AM PDT NEWLINE Additional Details : NEWLINE gcc doesn''t compile the c files with "-fpic" so the NEWLINE creation of a shared library results in an internal NEWLINE compiler error NEWLINE (only seen on Sparc/Linux; not on x86/Linux) NEWLINE NEWLINE Fix: NEWLINE NEWLINE --- ns/config/Linux.mk.orig Tue Mar 31 04:06:52 1998 NEWLINE +++ ns/config/Linux.mk Wed Apr 8 15:16:42 1998 NEWLINE @@ -43,7 +43,7 @@ NEWLINE PORT_FLAGS = -D_POSIX_SOURCE -D_BSD_SOURCE NEWLINE -DSW_THREADS -DNEED_ENDIAN_H -DNEED_GETOPT_H -DNEED_IOCTL_H NEWLINE -DUSE_NODL_TABS -DHAVE_SIGNED_CHAR -DNEED_SYS_TIME_H NEWLINE -DHAVE_SYS_BITYPES_H -DNEED_UINT_T NEWLINE PDJAVA_FLAGS = -mx128m NEWLINE NEWLINE -OS_CFLAGS = $(PLATFORM_FLAGS) $(PORT_FLAGS) NEWLINE $(MOVEMAIL_FLAGS) NEWLINE +OS_CFLAGS = $(DSO_CFLAGS) $(PLATFORM_FLAGS) NEWLINE $(PORT_FLAGS) $(MOVEMAIL_FLAGS) NEWLINE NEWLINE LOCALE_MAP = $(DEPTH)/cmd/xfe/intl/linux.lm NEWLINE EN_LOCALE = C NEWLINE Updated by Wan-Teh Chang (wtc@netscape.com) on Friday, April 24, 1998 12:13:29 PM PDT NEWLINE Additional Details : NEWLINE Assigned bug to Brian Ostrom <briano@netscape.com>. NEWLINE Added myself to the CC list. NEWLINE NEWLINE Brian: we should use -fPIC instead of -fpic. NEWLINE ns/nsprpub/config/Linux.mk uses -fPIC. -fPIC allows NEWLINE you to have large-displacement branches. NEWLINE Updated by Brian Ostrom (briano@netscape.com) on Tuesday, July 14, 1998 5:42:04 PM PDT NEWLINE Additional Details : NEWLINE I think this is fixed now. I added -fPIC to the compile NEWLINE line for all Linux processor-variants. It doesn't seem NEWLINE to hurt anything to do so. We'll see.
118|Problem with little endian / big endian on Sparc/Linux|Created by Thorsten Heit (thorsten@wurzelausix.cs.uni-sb.de) on Wednesday, April 8, 1998 7:09:13 AM PDT NEWLINE Additional Details : NEWLINE I do not know what problem could arise, but before NEWLINE trying it I installed the following patch to remove NEWLINE some compiler warning about redefining symbols: NEWLINE NEWLINE (only seen on Sparc/Linux which is a big endian machine, NEWLINE not on x86/Linux) NEWLINE NEWLINE --- ns/include/mcom_db.h.orig Wed Apr 8 16:01:41 1998 NEWLINE +++ ns/include/mcom_db.h Wed Apr 8 16:02:09 1998 NEWLINE @@ -420,8 +420,10 @@ NEWLINE /* NEWLINE * Why you would want to override the system defines? NEWLINE */ NEWLINE +#if !defined(__sparc__) NEWLINE #if BYTE_ORDER != LITTLE_ENDIAN NEWLINE #define BYTE_ORDER LITTLE_ENDIAN NEWLINE +#endif NEWLINE #endif NEWLINE #endif NEWLINE --- ns/dbm/include/mcom_db.h.orig Tue Mar 31 04:05:30 NEWLINE 1998 NEWLINE +++ ns/dbm/include/mcom_db.h Wed Apr 8 16:06:07 1998 NEWLINE @@ -398,8 +398,10 @@ NEWLINE __END_DECLS NEWLINE NEWLINE #ifdef linux NEWLINE +#if !defined(__sparc__) NEWLINE #if BYTE_ORDER != LITTLE_ENDIAN NEWLINE #define BYTE_ORDER LITTLE_ENDIAN NEWLINE +#endif NEWLINE #endif NEWLINE #endif
119|[PP]User Break: Unable to translate Unix file path /dev/ to Mac path|Created by steve orens (sorens@acm.org) on Wednesday, April 8, 1998 10:28:12 AM PDT NEWLINE Additional Details : NEWLINE mozilla for the mainctosh throws out the above user break on NEWLINE startup. this prevents it from getting to its preferences NEWLINE correctly, i believe which is also preventing things like NEWLINE cookies from working. Also, every time you startup, it asks NEWLINE you which user you are again. NEWLINE NEWLINE i built this version of mozilla on a different hard drive NEWLINE than the boot drive (hence the /dev/). however, when i moved NEWLINE the client app to the boot drive, it didn NEWLINE Updated by Mike Pinkerton (pinkerton@netscape.com) on Monday, April 13, 1998 11:28:13 AM PDT NEWLINE Additional Details : NEWLINE This is a duplicate of bug #163. The asking about user profile at startup is NEWLINE hardwired for now, and isn''t really a bug. See bug #163 for a patch which I will NEWLINE integrate into the tree at first chance. NEWLINE Updated by Wan-Teh Chang (wtc@netscape.com) on Friday, April 24, 1998 12:19:16 PM PDT NEWLINE Additional Details : NEWLINE Assigned bug to myself. NEWLINE NEWLINE This bug has been fixed as Mike Pinkerton suggested. NEWLINE The fix is checked into ns/nsprpub/pr/src/md/macio.c, NEWLINE revision 3.2. NEWLINE NEWLINE Closed the bug.
120|Navigator crashes if external viewer is registered with WWW_RegisterViewer|Created by Stan Bobrovskiy (stanb@real.com) on Wednesday, April 8, 1998 12:08:13 PM PDT NEWLINE Additional Details : NEWLINE After configuring an external viewer with WWW_RegisterViewer NEWLINE Navigator 4.04 and 5.0 crashes when you click on a link that NEWLINE belongs to that registered viewer. The crash is - Access NEWLINE Violation, using a pointer without checking if it is not NEWLINE NULL. NEWLINE The same thing happens if using OLE to register external NEWLINE viewer. NEWLINE The reason of the crash is a new piece of code in NEWLINE ns\\cmd\\winfe\\cxsave.cpp in CSaveCX::CanCreate() method: NEWLINE NEWLINE // If this file exists, then we must attempt another temp NEWLINE file. NEWLINE if(-1 != _access(pLocalName, 0)) { NEWLINE int type = NET_URL_Type(pUrl->address); NEWLINE // bug 63751 for Mail/News attachment NEWLINE NEWLINE pUrl here is NULL, so simple change fixes the cresh: NEWLINE NEWLINE // If this file exists, then we must attempt another temp NEWLINE file. NEWLINE if(-1 != _access(pLocalName, 0)) { NEWLINE if(pUrl) NEWLINE { NEWLINE int type = NET_URL_Type(pUrl->address); NEWLINE // bug 63751 for Mail/News attachment NEWLINE ... NEWLINE } NEWLINE Updated by Garrett Arch Blythe (blythe@netscape.com) on Monday, August 10, 1998 3:14:06 PM PDT NEWLINE Additional Details : NEWLINE Don't know who bugzilla@mozilla.org would be, or why they would desire to fix NEWLINE the bug. Taking the bug over as clearly a windows issue. NEWLINE Updated by Garrett Arch Blythe (blythe@netscape.com) on Monday, August 10, 1998 3:57:40 PM PDT NEWLINE Additional Details : NEWLINE A fix has been checked into the mozilla code base.
121|Memory leaks in read_font_spec|Created by (weitsang@cs.cornell.edu) on Wednesday, April 8, 1998 3:29:12 PM PDT NEWLINE Additional Details : NEWLINE read_font_spec should free font_spec before returning.
122|BUILD: Mozilla does not compile on NetBSD.|Created by Frank van der Linden (frank@wins.uva.nl) on Wednesday, April 8, 1998 3:46:34 PM PDT NEWLINE Additional Details : NEWLINE Mozilla does not compile on NetBSD. I made it compile and NEWLINE run on NetBSD/i386. The diffs are at: NEWLINE NEWLINE ftp://ftp.netbsd.org/pub/NetBSD/misc/fvdl/mozilla/diff.001 NEWLINE NEWLINE These diffs mainly deal with getting things to work on NEWLINE NetBSD. This first patch is only tested for the i386 NEWLINE platform, other platforms may need some small modifications NEWLINE in the MD parts of the code, so more diffs are likely to NEWLINE follow. I also split out some of the *BSD defines, to not NEWLINE make them depend on eachother. NEWLINE NEWLINE In further bugreports for NetBSD, I will probably be using NEWLINE the directory mentioned above to make the patches available NEWLINE to you. NEWLINE NEWLINE - Frank NEWLINE NEWLINE P.S. NEWLINE I selected the "PC" platform because it allowed me to NEWLINE specify the "other" category :-) But it is by no means NEWLINE the only NetBSD platform. NEWLINE Updated by Wan-Teh Chang (wtc@netscape.com) on Wednesday, April 22, 1998 12:58:23 PM PDT NEWLINE Additional Details : NEWLINE Added myself to the CC list. NEWLINE NEWLINE I will look at the diffs concerning ns/nsprpub. NEWLINE Updated by Frank van der Linden (frank@wins.uva.nl) on Sunday, May 24, 1998 4:43:14 PM PDT NEWLINE Additional Details : NEWLINE The updated changes for the 1998-04-29 version can be found in NEWLINE ftp://ftp.netbsd.org/ftp/pub/NetBSD/misc/fvdl/mozilla/diff-19980429.001 NEWLINE NEWLINE There weren't many changes; just a few conflicts with changes made in the main NEWLINE source tree were resolved. NEWLINE Updated by Frank van der Linden (frank@wins.uva.nl) on Sunday, May 24, 1998 4:45:42 PM PDT NEWLINE Additional Details : NEWLINE Just to be a bit clearer, I updated the version & platform in the bugreport NEWLINE (I hope the update for that works in bugzilla) NEWLINE Updated by Frank van der Linden (frank@wins.uva.nl) on Tuesday, August 4, 1998 12:02:11 PM PDT NEWLINE Additional Details : NEWLINE The remainder of the patches (i.e. the ones that are left now that the nsprpub NEWLINE part has been integrated) for the 1998-07-28 release can be found in NEWLINE ftp://ftp.netbsd.org/pub/NetBSD/misc/fvdl/mozilla/diff-19980728.001 NEWLINE NEWLINE This diff does contain 2 additions to files that already were patched; one NEWLINE because the name of a jmpbuf index was changed in the arm32 specific files NEWLINE of NetBSD, and another that makes the compile always use -fPIC.
123|Form menus not wide enough|Created by Darrell Kindred (dkindred=mozilla@cs.cmu.edu) on Wednesday, April 8, 1998 5:18:59 PM PDT NEWLINE Additional Details : NEWLINE Using Motif 1.2.4 under Linux, menus in HTML forms are not NEWLINE wide enough for both the text of the longest item and the NEWLINE little arrow button on the right. It looks like the width NEWLINE of the arrow is not taken into account. For an example with NEWLINE screenshot, see NEWLINE http://www.cs.cmu.edu/People/dkindred/bugs/menu-width.html NEWLINE Updated by Ramiro Estrugo (ramiro@netscape.com) on Wednesday, April 29, 1998 2:46:32 AM PDT NEWLINE Additional Details : NEWLINE Assign to ramiro. NEWLINE NEWLINE Yes the combo box sucks and has lots of bugs. Thanks.
124|Navigator word-wrap endless loop|Created by Jukka Santala (donwulff@iki.fi) on Wednesday, April 8, 1998 5:46:05 PM PDT NEWLINE Additional Details : NEWLINE This problem is already known to Netscape, since quick NEWLINE debugging turned up: NEWLINE NEWLINE /* NEWLINE * BUG BUG: We''re restoring the break state to the beginning NEWLINE of the buffer - ie NEWLINE * to before the space we skipped above. We need to fix the NEWLINE space skipping mechanism NEWLINE * to remove this case (we can go into an infinite loop here NEWLINE if there''s not enough space NEWLINE * for the first word). NEWLINE * NEWLINE * Should be able to have a new flag "canSkipSpace" but then NEWLINE actually don''t skip it. Then NEWLINE * if the line does fit and it''s at the beginning, we can NEWLINE skip the space. lo_FindLineBreak NEWLINE * should probably be the one to do this work so that the NEWLINE width we get back is correct. NEWLINE */ NEWLINE NEWLINE This text appears in LO_Layout_Text_Block() of NEWLINE lib\\layout\\leytext.c, 6486 lines into it to be exact. A NEWLINE quick way to skip it for stable build is to quote out the NEWLINE "continue" statement at the end of the function, altough NEWLINE this will cause at the very least bad formattign and NEWLINE potential loss of readability/text in the browser. However, NEWLINE for debugging and testing etc. I find this more desirable NEWLINE than an all-out crash while waiting for a correct patch, NEWLINE which I assume by the remark in the source-code Netscape is NEWLINE working on if hasn''t already finished. NEWLINE NEWLINE To actually see this bug in action (Well, as much as there''s NEWLINE to see in a hanging up browser) try the page NEWLINE http://www.cnet.com/Content/Builder/Authoring/Xml20/ss04.htm NEWLINE l - warning, make sure you have nothing important unsaved/in NEWLINE process in Mozilla, because there''s no graceful way out of NEWLINE this. NEWLINE Updated by Nisheeth Ranjan (nisheeth@netscape.com) on Sunday, April 26, 1998 6:31:54 PM PDT NEWLINE Additional Details : NEWLINE Actually, Jukka, its good that you reported this problem because the engineer NEWLINE who wrote this comment was a contractor and is no longer at Netscape. So nobody NEWLINE at Netscape was dealing with this problem. NEWLINE NEWLINE Am assigning this bug to myself... Thanks again, Jukka.
125|some links are "overlined" instead of underlined in xfe|Created by Jacob Berkman (jberk+@cmu.edu) on Wednesday, April 8, 1998 10:31:37 PM PDT NEWLINE Additional Details : NEWLINE some web pages have "overlined" links. NEWLINE One site that usually does for me is http://freshmeat.net NEWLINE Updated by Nisheeth Ranjan (nisheeth@netscape.com) on Sunday, April 26, 1998 6:33:51 PM PDT NEWLINE Additional Details : NEWLINE Thanks Jacob. NEWLINE NEWLINE Toshok is working on this problem already. Am assigning the bug to him... NEWLINE Updated by (toshok@netscape.com) on Monday, May 18, 1998 11:46:36 AM PDT NEWLINE Additional Details : NEWLINE changing platform to X-windows NEWLINE Updated by (toshok@netscape.com) on Monday, May 18, 1998 1:48:19 PM PDT NEWLINE Additional Details : NEWLINE just checked in a fix for this in cmd/xfe/lay.c. please update that file and NEWLINE try to reproduce the problem. if you can, please reopen this bug. i've already NEWLINE tried freshmeat.net, and it seems to be ok. NEWLINE NEWLINE chris
126|Navigator crashes on http://wwwis.win.tue.nl/~debra/orientatie/index1.html|Created by Paul De Bra (debra@win.tue.nl) on Thursday, April 9, 1998 5:41:11 AM PDT NEWLINE Additional Details : NEWLINE Load the mentioned page. It shows two frames. NEWLINE The left frame shows some kind of index. NEWLINE Click on one of the words in the index. NEWLINE Navigator crashes. NEWLINE NEWLINE This bug was already present in Navigator 4.04 and 4.05 and NEWLINE occurs on all unix versions but not on Windows''95. NEWLINE (The bug never showed up in the list of known issues.)
127|Fails to compile|Created by Edward Newman (Edward_newman@ml.com) on Thursday, April 9, 1998 8:01:37 AM PDT NEWLINE Additional Details : NEWLINE Compiling under NT 4 & VC++ V5 fails with uuid2.lib not NEWLINE found errors NEWLINE Updated by (warren@netscape.com) on Wednesday, April 15, 1998 5:24:42 PM PDT NEWLINE Additional Details : NEWLINE I compile with vc5 by setting the environment variable _MSC_VER to 1100 before NEWLINE running mozilla.mak. This is for the netscape commercial product tree -- I NEWLINE haven't tried this with the mozilla free source tree, but I assume it works the NEWLINE same.
128|"No move" drag''n''drop in outline == death|Created by Ben Laurie (ben@algroup.co.uk) on Thursday, April 9, 1998 10:47:14 AM PDT NEWLINE Additional Details : NEWLINE In the outliner, drag a bookmark until the line is just NEWLINE below the current location of the bookmark. Bang! NEWLINE NEWLINE I''ve traced through and the problem seems to be that the old NEWLINE location is removed before it is used to construct the new NEWLINE one. Or something like that. Unfortunately I can''t see a fix NEWLINE because you never seem to have quite all the information NEWLINE needed to detect the condition at once.
129|Missing variable initialisation in ns\\lib\\layout\\layimage.c|Created by Ben Laurie (ben@algroup.co.uk) on Thursday, April 9, 1998 10:57:38 AM PDT NEWLINE Additional Details : NEWLINE Although I haven''t actually seen this cause a problem, it NEWLINE seems clearly wrong: NEWLINE NEWLINE ns\\lib\\layout\\layimage.c line 2861: NEWLINE NEWLINE LO_ImageStruct *lo_image; NEWLINE NEWLINE should be: NEWLINE NEWLINE LO_ImageStruct *lo_image=NULL; NEWLINE Updated by Wan-Teh Chang (wtc@netscape.com) on Friday, April 24, 1998 12:24:12 PM PDT NEWLINE Additional Details : NEWLINE Added the file name to "Summary Description". NEWLINE Assigned bug to owner of the layout module, NEWLINE David Williams <djw@netscape.com>. NEWLINE Updated by Nisheeth Ranjan (nisheeth@netscape.com) on Sunday, April 26, 1998 6:37:29 PM PDT NEWLINE Additional Details : NEWLINE Hmm... Wan-Teh's bug assignment to djw didn't go through properly because the NEWLINE assigned to field was still bugzilla@mozilla.org. NEWLINE NEWLINE Anyway, am assigning the bug to myself...
130|paste into forms with middle mouse doesnt always work|Created by Brendan White (bmwt@bullwinkle.pas.rochester.edu) on Thursday, April 9, 1998 11:39:48 AM PDT NEWLINE Additional Details : NEWLINE When selecting text from external application and pasting NEWLINE into a text entry field, the middle mouse button (paste) NEWLINE doesnt always work. For example, pasting to the "enter a NEWLINE bug ID" on NEWLINE http://casetracker.netscape.com/Bugzilla/navframe.html NEWLINE NEWLINE doesnt work. NEWLINE Updated by Steve Lamm (slamm@netscape.com) on Thursday, April 30, 1998 10:38:29 AM PDT NEWLINE Additional Details : NEWLINE I am the one most familiar with this code. I add it to my plate of bugs.
131|Add config file support for Linux 2.1.x|Created by Rick Niles (niles@axp745.gsfc.nasa.gov) on Thursday, April 9, 1998 1:19:24 PM PDT NEWLINE Additional Details : NEWLINE Copy ns/nsprpub/config/LinuxELF2.0.mk to NEWLINE ns/nsprpub/config/LinuxELF2.1.mk and NEWLINE then add the follow hacked patch (there NEWLINE should be real LINUX2_2 flags in the NEWLINE source later I guess): NEWLINE NEWLINE *** Linux.mk Thu Apr 9 16:18:45 1998 NEWLINE --- Linux.mk.orig Thu Apr 9 16:18:20 1998 NEWLINE *************** ifeq ($(OS_RELEASE),2.0) NEWLINE *** 77,86 **** NEWLINE PLATFORM_FLAGS += -DLINUX2_0 NEWLINE endif NEWLINE NEWLINE - ifeq ($(OS_RELEASE),2.1) NEWLINE - PLATFORM_FLAGS += -DLINUX2_0 NEWLINE - endif NEWLINE - NEWLINE # NEWLINE # Linux ppc and 2.0 have shared libraries. NEWLINE # NEWLINE --- 77,82 ---- NEWLINE Updated by Wan-Teh Chang (wtc@netscape.com) on Wednesday, April 22, 1998 12:56:37 PM PDT NEWLINE Additional Details : NEWLINE Assigned the bug to myself. NEWLINE Updated by Wan-Teh Chang (wtc@netscape.com) on Tuesday, June 16, 1998 8:25:19 PM PDT NEWLINE Additional Details : NEWLINE I changed mozilla/nsprpub/config/Linux.mk NEWLINE so that it does not define version-specific NEWLINE macros. Instead, now we use feature-based NEWLINE macros when possible. NEWLINE NEWLINE So it should build just fine on Linux 2.1.x NEWLINE now. NEWLINE NEWLINE Marked the bug closed.
132|odd cnet site induced hang (CSS?)|Created by chris tacy (chris@enginered.com) on Thursday, April 9, 1998 4:53:50 PM PDT NEWLINE Additional Details : NEWLINE http://www.cnet.com/Content/Builder/Business/Shafer/040698/? NEWLINE st.bl.fd.ts2.mbcol NEWLINE NEWLINE this is a build of the newest source (from today). it''s the NEWLINE debugging build. NT4.0, sp3. VC++ w/sp. NEWLINE communicator completely hangs (can''t do anything with it, NEWLINE can''t get a trace, nothing). have to kill it. NEWLINE this happens every time i go to the above URL. NEWLINE NEWLINE HTML looks standard except maybe the CSS stuff.
133|Navigator draws entities like <&> as ''?''s when KOI-8 encoding is set.|Created by Boris Tobotras (boris@xtalk.msk.su) on Thursday, April 9, 1998 9:41:16 PM PDT NEWLINE Additional Details : NEWLINE To reproduce is easy: point navigator to the HTML document: NEWLINE <HTML><HEAD></HEAD><BODY><P>Can you see this & NEWLINE that?</HTML> NEWLINE NEWLINE Fix seems to be easy as well: NEWLINE NEWLINE --- mozilla-orig/ns/lib/libi18n/ugendata.c Tue Mar 31 NEWLINE 06:04:14 1998 NEWLINE +++ mozilla/ns/lib/libi18n/ugendata.c Wed Apr 1 19:04:18 NEWLINE 1998 NEWLINE @@ -251,6 +251,14 @@ NEWLINE {CS_DEFAULT,0xFF,0x00}, NEWLINE {CS_DEFAULT,0xFF,0x00} NEWLINE }}, NEWLINE + /*+ [BT] +*/ NEWLINE + { CS_KOI8_R, { NEWLINE + {CS_KOI8_R,0x80,0xFE}, NEWLINE + {CS_ASCII,0x00,0x7E}, NEWLINE + {CS_DEFAULT,0xFF,0x00}, NEWLINE + {CS_DEFAULT,0xFF,0x00} NEWLINE + }}, NEWLINE + /*- [BT] -*/ NEWLINE { CS_CNS_8BIT, { NEWLINE {CS_CNS11643_1,0xA1,0xFE}, NEWLINE {CS_CNS11643_2,0x8E,0x8E}, NEWLINE NEWLINE NEWLINE This was introduced in 4.x and is _*very*_ annoying and NEWLINE disturbing for KOI8 users.
134|Navigator crashes w/ javascript setting textarea value|Created by Darrell Kindred (dkindred=mozilla@cs.cmu.edu) on Friday, April 10, 1998 12:59:33 AM PDT NEWLINE Additional Details : NEWLINE I get a reproducible seg fault from the 19980408 release NEWLINE when visiting this document: NEWLINE NEWLINE http://www.cs.cmu.edu/People/dkindred/bugs/moz-crash-0410.html NEWLINE NEWLINE The problem seems to be caused by javascript code setting NEWLINE the default value of a TEXTAREA field. NEWLINE NEWLINE (Linux, Motif 1.2.4) NEWLINE Updated by Darrell Kindred (dkindred=mozilla@cs.cmu.edu) on Friday, April 10, 1998 1:03:58 AM PDT NEWLINE Additional Details : NEWLINE Backtrace is available here: NEWLINE http://www.cs.cmu.edu/People/dkindred/bugs/moz-crash-0410-bt.txt NEWLINE Updated by (toshok@netscape.com) on Monday, May 18, 1998 11:28:19 AM PDT NEWLINE Additional Details : NEWLINE Just checked in a fix submitted by waider@waider.ie (Ronan Waide) for this NEWLINE problem.
135|layout error: no linebreak before UL bullet in table|Created by Darrell Kindred (dkindred=mozilla@cs.cmu.edu) on Friday, April 10, 1998 2:15:25 AM PDT NEWLINE Additional Details : NEWLINE Mozilla version 19980408 lays out the UL list here badly, NEWLINE using 14-point Helvetica fonts under Linux. NEWLINE NEWLINE There''s no linebreak before the second bulleted item, and NEWLINE the bullet overlaps the text for the previous item. NEWLINE A screenshot is included. NEWLINE NEWLINE http://www.cs.cmu.edu/People/dkindred/bugs/bullet-glitch.html
136|Compilation : missing include|Created by Hubert Figuiere (Hubert.Figuiere@solsoft.com) on Friday, April 10, 1998 5:41:40 AM PDT NEWLINE Additional Details : NEWLINE In ''ns/cmd/xfe/Microline3.0/XmL/Grid.c'', you should insert NEWLINE a ''#include <Xm/DropSMgr.h>'' to be able to compile it with NEWLINE LessTif 0.83 under Linux. Otherwise NEWLINE ''XmDropProcCallbackStruct'' is undefined. NEWLINE NEWLINE This happens with both 03-31 and 04-08 snapshots. NEWLINE NEWLINE NEWLINE Hub
137|Compilation : wrong library Path|Created by Hubert Figuiere (Hubert.Figuiere@solsoft.com) on Friday, April 10, 1998 5:45:25 AM PDT NEWLINE Additional Details : NEWLINE At link-time, the Makefile has Motif libXm.a hardcoded to be NEWLINE in /usr/X11R6/lib. On my system it is /usr/local/lib NEWLINE NEWLINE I patched the makefile but a configurable way would be NEWLINE better. NEWLINE This happens in both 03-31 and 04-08 releases. NEWLINE NEWLINE Hub
138|word broken across lines after first char|Created by Darrell Kindred (dkindred=mozilla@cs.cmu.edu) on Friday, April 10, 1998 11:00:06 AM PDT NEWLINE Additional Details : NEWLINE In versions 19980331 and 19980408, line breaks are NEWLINE sometimes improperly inserted after the first character NEWLINE in a word. You can see the effects of this bug here: NEWLINE NEWLINE http://www.cs.cmu.edu/People/dkindred/bugs/word-wrap.html NEWLINE NEWLINE I submitted a patch for this bug to mozilla.patches: NEWLINE NEWLINE news://news.mozilla.org/352C662E.641C6967@cmu.edu
139|nsrpub/pr/src/md/unix.c error for BSD/OS 3.1|Created by (jason@ackley.net) on Friday, April 10, 1998 1:16:31 PM PDT NEWLINE Additional Details : NEWLINE In case it has not been reported: NEWLINE NEWLINE Line 2319 of unix.c has a if defined(FREEBSD) to set osflags NEWLINE to O_SYNC or O_FSYNC, it should include BSDI (at least 3.1, NEWLINE I have not tested on other revs) NEWLINE NEWLINE To fix: NEWLINE NEWLINE in unix.c , I added under some of the #include''s NEWLINE NEWLINE #if defined(BSDI) NEWLINE #include <sys/param.h> /* For _BSDI_VERSION */ NEWLINE #endif NEWLINE NEWLINE Then at the if defined(FREEBSD), I changed it to NEWLINE NEWLINE #if defined(FREEBSD) \|\| defined(BSDI) && _BSDI_VERSION >= \\ NEWLINE 199701 NEWLINE NEWLINE etc etc NEWLINE Updated by Wan-Teh Chang (wtc@netscape.com) on Wednesday, April 22, 1998 12:52:59 PM PDT NEWLINE Additional Details : NEWLINE Assigned the bug to myself. NEWLINE NEWLINE We already fixed this problem in our internal development version. NEWLINE Our fix is similar to yours, except that we do not check the BSDI NEWLINE version, i.e., we just say NEWLINE #if defined(FREEBSD) \|\| defined(BSDI) NEWLINE etc etc NEWLINE NEWLINE Perhaps we should make it more OS-independent, like this: NEWLINE #if defined(O_SYNC) NEWLINE osflags \|= O_SYNC; NEWLINE #elif defined(O_FSYNC) NEWLINE osflags \|= O_FSYNC; NEWLINE #else NEWLINE #error "Neither O_SYNC nor O_FSYNC is defined" NEWLINE #endif NEWLINE NEWLINE I will test this fix. NEWLINE NEWLINE Thanks for reporting the problem. NEWLINE Updated by (jason@ackley.net) on Wednesday, April 22, 1998 10:19:41 PM PDT NEWLINE Additional Details : NEWLINE I was able to locate my BSDi 2.0 CD, and check its /usr/include/sys//fcntl.h NEWLINE against the current 3.1 fcntl.h, only differences are BSDi RCS versions, so NEWLINE it should be OK not to check the version number for it. NEWLINE Updated by Wan-Teh Chang (wtc@netscape.com) on Friday, April 24, 1998 12:31:04 PM PDT NEWLINE Additional Details : NEWLINE Thanks for checking BSDi 2.0 for me. NEWLINE NEWLINE I checked in the fix that conditionalizes on the NEWLINE O_SYNC/O_FSYNC macros instead of the platform NEWLINE macros like FREEBSD/BSDI. I tested the fix NEWLINE on BSDi 2.0 and 3.0. NEWLINE NEWLINE The fix is in ns/nsprpub/pr/src/md/unix/unix.c, NEWLINE revision 3.4. NEWLINE NEWLINE Marked the bug closed.
140|''&'' not correctly displated in doc title|Created by Ben Laurie (ben@algroup.co.uk) on Friday, April 10, 1998 2:12:58 PM PDT NEWLINE Additional Details : NEWLINE OK, it''s a small thing, and it''s MS''s fault, but it has NEWLINE always annoyed me. Ampersands don''t get displayed correctly NEWLINE in the title bar on Windows. This fixes it. Give me CVS NEWLINE access and I''ll fix it everywhere :-) NEWLINE NEWLINE In cmd/winfe/genchrom.cpp: NEWLINE NEWLINE void CGenericChrome::SetDocumentTitle(const char *lpszText) NEWLINE NEWLINE and NEWLINE NEWLINE void CGenericChrome::SetWindowTitle(const char *lpszText) NEWLINE NEWLINE (so why doesn''t one of these call the other?) should look NEWLINE like this: NEWLINE NEWLINE void CGenericChrome::SetWindowTitle(const char *lpszText) NEWLINE { NEWLINE m_csWindowTitle = lpszText; NEWLINE NEWLINE CString cs; NEWLINE cs = "Ben " + m_csDocTitle; NEWLINE if (!m_csDocTitle.IsEmpty()) { NEWLINE cs += " - "; NEWLINE } NEWLINE cs += m_csWindowTitle; NEWLINE #if 1 NEWLINE // Ben fix ''&'' in title NEWLINE char *buf=(char *)alloca(cs.GetLength()*2+1); NEWLINE int s,d; NEWLINE for(s=d=0 ; s < cs.GetLength() ; ++d,++s) NEWLINE { NEWLINE if(cs[s] == ''&'') NEWLINE buf[d++]=''&''; NEWLINE buf[d]=cs[s]; NEWLINE } NEWLINE buf[d]=''\\0''; NEWLINE m_pParent->SetWindowText(buf); NEWLINE #else NEWLINE m_pParent->SetWindowText(cs); NEWLINE #endif NEWLINE }
141|BUILD: Unable to compile Navigator/Communicator 5.0|Created by Dwight Chang (dcrs6000@pacbell.net) on Friday, April 10, 1998 6:35:05 PM PDT NEWLINE Additional Details : NEWLINE Hardware: IBM RS/6000 NEWLINE Operating System: AIX 4.2.1 NEWLINE NEWLINE Problem: NEWLINE I noticed that h_page.c and hash.c did not compile NEWLINE because: NEWLINE BYTE_ORDER BIG_ENDIAN NEWLINE were not defined. There seems to be a problem with the NEWLINE mcom_db.h include file. Not sure, but I think there are NEWLINE some unmatched "endif". The result of which is that the: NEWLINE #include <sys/machine.h> NEWLINE statement is not activated. Which result in the undefined NEWLINE references. NEWLINE NEWLINE Question: NEWLINE Is there a way to get the "gmake -ek" command to use NEWLINE the GNU gcc, g++, c++ compliers that are in my NEWLINE /usr/local/bin directory? Everytime I executed NEWLINE gmake -ek command I ended up using AIX''s "c" compiler. NEWLINE I do not have a "c++" compiler from IBM. The only NEWLINE "c++" compiler I have is the GNU one which resides NEWLINE in the /usr/local/bin directory.
142|Unable to compile Navigator/Communicator 5.0|Created by Dwight Chang (dcrs6000@pacbell.net) on Friday, April 10, 1998 6:35:06 PM PDT NEWLINE Additional Details : NEWLINE Hardware: IBM RS/6000 NEWLINE Operating System: AIX 4.2.1 NEWLINE NEWLINE Problem: NEWLINE I noticed that h_page.c and hash.c did not compile NEWLINE because: NEWLINE BYTE_ORDER BIG_ENDIAN NEWLINE were not defined. There seems to be a problem with the NEWLINE mcom_db.h include file. Not sure, but I think there are NEWLINE some unmatched "endif". The result of which is that the: NEWLINE #include <sys/machine.h> NEWLINE statement is not activated. Which result in the undefined NEWLINE references. NEWLINE NEWLINE Question: NEWLINE Is there a way to get the "gmake -ek" command to use NEWLINE the GNU gcc, g++, c++ compliers that are in my NEWLINE /usr/local/bin directory? Everytime I executed NEWLINE gmake -ek command I ended up using AIX''s "c" compiler. NEWLINE I do not have a "c++" compiler from IBM. The only NEWLINE "c++" compiler I have is the GNU one which resides NEWLINE in the /usr/local/bin directory.
143|Mozilla compile failure...looking for uuid2.lib|Created by Ken Camann (camannk@hotmail.com) on Friday, April 10, 1998 8:31:00 PM PDT NEWLINE Additional Details : NEWLINE When I try to compile CLIENT.MAK, I (eventually) get to a NEWLINE point where GMAKE reports it cannot find the library file NEWLINE "uuid2.lib". I have looked and I don NEWLINE Updated by Ludovic Dubost (ludovic@netscape.com) on Tuesday, April 14, 1998 2:18:02 AM PDT NEWLINE Additional Details : NEWLINE NEWLINE RTFM.. In the build instructions is indicated NEWLINE NEWLINE if you are running VC++ 5.0 or higher: set _MSC_VER=1100 NEWLINE NEWLINE In MSVC 5.0 the uuid2.lib has been merged into uuid.lib
144|ToolBar tabs get messed up when closed and reopened much.|Created by Lars Clausen (larsrc@irisa.fr) on Saturday, April 11, 1998 4:21:01 AM PDT NEWLINE Updated by Lars Clausen (larsrc@irisa.fr) on Saturday, April 11, 1998 4:25:35 AM PDT NEWLINE Additional Details : NEWLINE Close two of the toolbar items, then open them right-most first. Or close NEWLINE them all and open them again. Severe visual defects follow (toolbar children NEWLINE whited out, page view disappears, noise in status box, incorrect height NEWLINE calculation for toolbox or other. This may be one or more actual bugs. NEWLINE Updated by Steve Lamm (slamm@netscape.com) on Thursday, April 30, 1998 10:28:58 AM PDT NEWLINE Additional Details : NEWLINE Ramiro has been working on this. Reassigning to him.
145|linebreaks lost in PRE within fixed-width table|Created by Darrell Kindred (dkindred=mozilla@cs.cmu.edu) on Saturday, April 11, 1998 12:31:29 PM PDT NEWLINE Additional Details : NEWLINE In Navigator 19980408, in a table with WIDTH specified, NEWLINE PRE blocks don''t have their linebreaks preserved. NEWLINE You can see this bug here: NEWLINE NEWLINE http://www.cs.cmu.edu/People/dkindred/bugs/moz/table-pre.html
146|prscanf.c(184) : unary minus operator applied to unsigned type|Created by Rick Osborne (rick@rixsoft.com) on Saturday, April 11, 1998 8:38:25 PM PDT NEWLINE Additional Details : NEWLINE The function _pr_strtoull is for Uint64 but gets called in a NEWLINE signed context. Just need to cast as signed when applying NEWLINE negative. Trivial, but it eliminates one more compiler NEWLINE warning. NEWLINE NEWLINE --- prscanf.c Sat Apr 11 23:16:19 1998 NEWLINE +++ prscanf.old.c Wed Apr 08 17:55:01 1998 NEWLINE @@ -181,7 +181,7 @@ NEWLINE } NEWLINE NEWLINE if (negative) { NEWLINE - LL_NEG((PRInt64)x, (PRInt64)x); NEWLINE + LL_NEG(x, x); NEWLINE } NEWLINE NEWLINE if (endptr) { NEWLINE Updated by Wan-Teh Chang (wtc@netscape.com) on Friday, April 24, 1998 12:32:08 PM PDT NEWLINE Additional Details : NEWLINE Assigned bug to myself.
147|warning : ntinrval.c(77) : unreference variable count|Created by Rick Osborne (rick@rixsoft.com) on Saturday, April 11, 1998 9:04:15 PM PDT NEWLINE Additional Details : NEWLINE (In the continuing quest to eliminate even compiler warnings NEWLINE ... ) NEWLINE NEWLINE I don NEWLINE Updated by Wan-Teh Chang (wtc@netscape.com) on Monday, April 13, 1998 6:08:14 PM PDT NEWLINE Additional Details : NEWLINE This has been fixed in revision 3.2 of NEWLINE ns/nsprpub/pr/src/md/windows/ntinrval.c. NEWLINE NEWLINE Thanks for reporting this problem and NEWLINE providing a fix. NEWLINE NEWLINE Mark the bug closed. NEWLINE Updated by Wan-Teh Chang (wtc@netscape.com) on Friday, April 24, 1998 12:32:59 PM PDT NEWLINE Additional Details : NEWLINE Assigned the bug to myself and closed the bug. NEWLINE It's fixed.
148|warning fix : ns/nsprpub/pr/src/threads/combined/prucv.c|Created by Rick Osborne (rick@rixsoft.com) on Saturday, April 11, 1998 9:37:24 PM PDT NEWLINE Additional Details : NEWLINE (In the continuing quest to eliminate even compiler warnings NEWLINE ... ) NEWLINE NEWLINE Just a couple of cases of compiler directives mucking with NEWLINE variable scopes. The last one was a bit tricky, but I NEWLINE figured it out. It boils down to the fact that the compiler NEWLINE just eliminates the call because it simplifies to if(!1). NEWLINE So, you take the variable out of the picture when that case NEWLINE occurs (global threads). NEWLINE NEWLINE --- prucv.c Sun Apr 12 00:33:55 1998 NEWLINE +++ prucv.old.c Wed Apr 08 17:55:09 1998 NEWLINE @@ -157,9 +157,7 @@ void _PR_NotifyLockedThread (PRThread *t NEWLINE PRStatus _PR_WaitCondVar( NEWLINE PRThread *thread, PRCondVar *cvar, PRLock *lock, NEWLINE PRIntervalTime timeout) NEWLINE { NEWLINE -#ifndef _PR_GLOBAL_THREADS_ONLY /* is isn''t used in that NEWLINE context */ NEWLINE intn is; NEWLINE -#endif NEWLINE PRStatus rv = PR_SUCCESS; NEWLINE NEWLINE PR_ASSERT(thread == _PR_MD_CURRENT_THREAD()); NEWLINE @@ -530,10 +528,8 @@ PR_IMPLEMENT(PRStatus) NEWLINE PR_NotifyCondVar( NEWLINE */ NEWLINE PR_IMPLEMENT(PRStatus) PR_NotifyAllCondVar(PRCondVar *cvar) NEWLINE { NEWLINE -#ifndef _PR_GLOBAL_THREADS_ONLY /* is and q not used in NEWLINE that context */ NEWLINE PRCList *q; NEWLINE PRIntn is; NEWLINE -#endif NEWLINE PRThread *me = _PR_MD_CURRENT_THREAD(); NEWLINE NEWLINE PR_ASSERT(cvar->lock->owner == me); NEWLINE @@ -618,11 +614,7 @@ PR_IMPLEMENT(PRStatus) NEWLINE PRP_NakedNotify(P NEWLINE PR_IMPLEMENT(PRStatus) PRP_NakedBroadcast(PRCondVar *cvar) NEWLINE { NEWLINE PRCList *q; NEWLINE -#ifndef _PR_GLOBAL_THREADS_ONLY NEWLINE - /* since this is all macro''ed, we get a warning in the NEWLINE above case due to NEWLINE - the compiler throwing out the call and thinking the NEWLINE var goes unrefed */ NEWLINE PRIntn is; NEWLINE -#endif NEWLINE PRThread *me = _PR_MD_CURRENT_THREAD(); NEWLINE PR_ASSERT(_PR_NAKED_CV_LOCK == cvar->lock); NEWLINE Updated by Wan-Teh Chang (wtc@netscape.com) on Friday, April 24, 1998 12:33:41 PM PDT NEWLINE Additional Details : NEWLINE Assigned bug to myself.
149|warning fix : ns/nsprpub/pr/src/threads/combined/prulock.c|Created by Rick Osborne (rick@rixsoft.com) on Saturday, April 11, 1998 9:50:36 PM PDT NEWLINE Additional Details : NEWLINE (In the continuing quest to eliminate even compiler warnings NEWLINE ... ) NEWLINE NEWLINE Ever more compiler directives mucking with variable scopes. NEWLINE Didn''t even have to think about this one. NEWLINE NEWLINE --- prulock.c Sun Apr 12 00:48:17 1998 NEWLINE +++ prulock.old.c Wed Apr 08 17:55:09 1998 NEWLINE @@ -206,11 +206,9 @@ extern PRThread *suspendAllThread; NEWLINE PR_IMPLEMENT(void) PR_Lock(PRLock *lock) NEWLINE { NEWLINE PRThread *me = _PR_MD_CURRENT_THREAD(); NEWLINE -#ifndef _PR_GLOBAL_THREADS_ONLY /* vars not used in that NEWLINE context */ NEWLINE PRIntn is; NEWLINE PRThread *t; NEWLINE PRCList *q; NEWLINE -#endif NEWLINE NEWLINE PR_ASSERT(me != suspendAllThread); NEWLINE #if !defined(XP_MAC) NEWLINE @@ -305,11 +303,9 @@ PR_IMPLEMENT(void) PR_Lock(PRLock NEWLINE *lock) NEWLINE */ NEWLINE PR_IMPLEMENT(PRStatus) PR_Unlock(PRLock *lock) NEWLINE { NEWLINE -#ifndef _PR_GLOBAL_THREADS_ONLY /* vars not used in that NEWLINE context */ NEWLINE PRCList *q; NEWLINE PRThreadPriority pri, boost; NEWLINE PRIntn is; NEWLINE -#endif NEWLINE PRThread *me = _PR_MD_CURRENT_THREAD(); NEWLINE NEWLINE PR_ASSERT(lock->owner == me); NEWLINE Updated by Wan-Teh Chang (wtc@netscape.com) on Friday, April 24, 1998 12:34:20 PM PDT NEWLINE Additional Details : NEWLINE Assigned bug to myself.
150|warning fix : ns/nsprpub/pr/src/threads/combined/pruthr.c|Created by Rick Osborne (rick@rixsoft.com) on Saturday, April 11, 1998 10:36:56 PM PDT NEWLINE Additional Details : NEWLINE (In the continuing quest to eliminate even compiler warnings NEWLINE ... ) NEWLINE NEWLINE Ever more compiler directives mucking with variable scopes. NEWLINE NEWLINE --- pruthr.c Sun Apr 12 01:39:09 1998 NEWLINE +++ pruthr.old.c Wed Apr 08 17:55:09 1998 NEWLINE @@ -440,5 +440,3 @@ void _PR_UserRunThread(void) NEWLINE PRThread *thread = _PR_MD_CURRENT_THREAD(); NEWLINE -#ifndef _PR_GLOBAL_THREADS_ONLY /* calls to var get thrown NEWLINE out */ NEWLINE PRIntn is; NEWLINE -#endif NEWLINE NEWLINE @@ -522,5 +520,3 @@ void _PR_UserRunThread(void) NEWLINE { NEWLINE -#ifndef _PR_GLOBAL_THREADS_ONLY /* calls to var get thrown NEWLINE out */ NEWLINE PRInt32 is; NEWLINE -#endif NEWLINE _PR_INTSOFF(is); NEWLINE @@ -536,8 +532,3 @@ void _PR_SetThreadPriority(PRThread *thr NEWLINE PRThread *me = _PR_MD_CURRENT_THREAD(); NEWLINE -#ifndef _PR_GLOBAL_THREADS_ONLY NEWLINE - /* since this is all macro''ed, we get a warning in the NEWLINE - above case due to the compiler throwing out the call NEWLINE - and thinking the var goes unreferenced */ NEWLINE PRIntn is; NEWLINE -#endif NEWLINE NEWLINE @@ -605,5 +596,3 @@ static void _PR_Suspend(PRThread *thread NEWLINE { NEWLINE -#ifndef _PR_GLOBAL_THREADS_ONLY NEWLINE PRIntn is; NEWLINE -#endif NEWLINE PRThread *me = _PR_MD_CURRENT_THREAD(); NEWLINE @@ -663,5 +652,3 @@ static void _PR_Resume(PRThread *thread) NEWLINE PRThreadPriority pri; NEWLINE -#ifndef _PR_GLOBAL_THREADS_ONLY NEWLINE PRIntn is; NEWLINE -#endif NEWLINE PRThread *me = _PR_MD_CURRENT_THREAD(); NEWLINE @@ -1054,5 +1041,3 @@ PR_IMPLEMENT(PRThread*) NEWLINE _PR_CreateThread NEWLINE char *top; NEWLINE -#ifndef _PR_GLOBAL_THREADS_ONLY NEWLINE PRIntn is; NEWLINE -#endif NEWLINE PRIntn native = 0; NEWLINE @@ -1374,5 +1359,3 @@ PR_IMPLEMENT(PRThread*) NEWLINE _PRI_AttachThrea NEWLINE if (thread) { NEWLINE -#ifndef _PR_GLOBAL_THREADS_ONLY /* calls to var get thrown NEWLINE out */ NEWLINE PRIntn is; NEWLINE -#endif NEWLINE NEWLINE @@ -1437,5 +1420,3 @@ PR_IMPLEMENT(void) NEWLINE PR_DetachThread(void) NEWLINE { NEWLINE -#ifndef _PR_GLOBAL_THREADS_ONLY /* calls to var get thrown NEWLINE out */ NEWLINE PRIntn is; NEWLINE -#endif NEWLINE PRThread *me = _PR_MD_CURRENT_THREAD(); NEWLINE @@ -1473,5 +1454,3 @@ PR_IMPLEMENT(PRStatus) NEWLINE PR_JoinThread(PRT NEWLINE { NEWLINE -#ifndef _PR_GLOBAL_THREADS_ONLY /* calls to var get thrown NEWLINE out */ NEWLINE PRIntn is; NEWLINE -#endif NEWLINE PRCondVar *term; NEWLINE Updated by Wan-Teh Chang (wtc@netscape.com) on Friday, April 24, 1998 12:34:52 PM PDT NEWLINE Additional Details : NEWLINE Assigned bug to myself.
151|warning fix : ns/nsprpub/pr/src/threads/prdump.c|Created by Rick Osborne (rick@rixsoft.com) on Saturday, April 11, 1998 10:45:46 PM PDT NEWLINE Additional Details : NEWLINE (In the continuing quest to eliminate even compiler warnings NEWLINE ... ) NEWLINE NEWLINE Ever more compiler directives mucking with variable scopes. NEWLINE NEWLINE --- prdump.c Sun Apr 12 01:47:43 1998 NEWLINE +++ prdump.old.c Wed Apr 08 17:55:09 1998 NEWLINE @@ -108,5 +108,3 @@ void PR_ShowStatus(void) NEWLINE { NEWLINE -#ifndef _PR_GLOBAL_THREADS_ONLY /* vars not used in this NEWLINE context */ NEWLINE PRIntn is; NEWLINE -#endif NEWLINE NEWLINE Updated by Wan-Teh Chang (wtc@netscape.com) on Friday, April 24, 1998 12:35:29 PM PDT NEWLINE Additional Details : NEWLINE Assigned bug to myself.
152|warning fix : ns/nsprpub/pr/src/threads/prcthr.c|Created by Rick Osborne (rick@rixsoft.com) on Saturday, April 11, 1998 10:52:00 PM PDT NEWLINE Additional Details : NEWLINE (In the continuing quest to eliminate even compiler warnings NEWLINE ... ) NEWLINE NEWLINE Ever more compiler directives mucking with variable scopes. NEWLINE NEWLINE --- prcthr.c Sun Apr 12 01:54:07 1998 NEWLINE +++ prcthr.old.c Wed Apr 08 17:55:09 1998 NEWLINE @@ -94,5 +94,3 @@ PR_IMPLEMENT(PRStatus) PR_Sleep(PRInterv NEWLINE */ NEWLINE -#ifndef _PR_GLOBAL_THREADS_ONLY /* vars not used in this NEWLINE context */ NEWLINE PRIntn is; NEWLINE -#endif NEWLINE PRThread *me = PR_GetCurrentThread(); NEWLINE @@ -242,5 +240,3 @@ PR_IMPLEMENT(void) PR_ClearInterrupt() NEWLINE { NEWLINE -#ifndef _PR_GLOBAL_THREADS_ONLY /* vars not used in this NEWLINE context */ NEWLINE PRIntn is; NEWLINE -#endif NEWLINE PRThread *me = _PR_MD_CURRENT_THREAD(); NEWLINE Updated by Wan-Teh Chang (wtc@netscape.com) on Friday, April 24, 1998 12:35:57 PM PDT NEWLINE Additional Details : NEWLINE Assigned bug to myself.
153|Assertion failed: expected_pixel == color.pixel, file colors.c, line 903|Created by Kartik Subbarao (subbarao@computer.org) on Sunday, April 12, 1998 2:08:10 PM PDT NEWLINE Additional Details : NEWLINE When trying to open the home page, the debug version of NEWLINE Mozilla dies with this assertion failure on HP-UX 10.20: NEWLINE NEWLINE Assertion failed: expected_pixel == color.pixel, file NEWLINE colors.c, line 903 NEWLINE NEWLINE Running it under the debugger: NEWLINE expected_pixel = 0 NEWLINE color.pixel = 1 NEWLINE NEWLINE color = struct XColor { NEWLINE pixel = 1; NEWLINE red = 0; NEWLINE green = 0; NEWLINE blue = 0; NEWLINE flags = ''^@'' (\\000); NEWLINE pad = ''^@'' (\\000); NEWLINE } NEWLINE NEWLINE Looks like it''s trying to allocate a color for black, and it NEWLINE expects to get assigned pixel 0, but is instead being NEWLINE assigned pixel 1. At least this is what it looks like at a NEWLINE superficial level.
154|warning fix : ns/nsprpub/pr/src/threads/prtpd.c|Created by Rick Osborne (rick@rixsoft.com) on Sunday, April 12, 1998 6:14:28 PM PDT NEWLINE Additional Details : NEWLINE (In the continuing quest to eliminate even compiler warnings NEWLINE ... ) NEWLINE NEWLINE This was a pleasant switch. Just some nice *normal* NEWLINE unreferenced vars, instead of these no-fun compiler NEWLINE directive problems. Oh well. NEWLINE NEWLINE --- prtpd.c Sun Apr 12 02:01:03 1998 NEWLINE +++ prtpd.old.c Wed Apr 08 17:55:09 1998 NEWLINE @@ -126,4 +126,3 @@ PR_IMPLEMENT(PRStatus) PR_NewThreadPriva NEWLINE PRThreadPrivateDTOR *old = NULL; NEWLINE - /* unreferenced - why is this here? */ NEWLINE - /* PRIntn _is; */ NEWLINE + PRIntn _is; NEWLINE NEWLINE @@ -194,4 +193,3 @@ PR_IMPLEMENT(PRStatus) PR_SetThreadPriva NEWLINE { NEWLINE - /* Why is this here? */ NEWLINE - /* PRIntn _is; */ NEWLINE + PRIntn _is; NEWLINE NEWLINE Updated by Wan-Teh Chang (wtc@netscape.com) on Friday, April 24, 1998 12:40:07 PM PDT NEWLINE Additional Details : NEWLINE Assigned bug to myself. NEWLINE NEWLINE The _is variables in this file may still be referenced on NEWLINE some platforms. See the definition of these two macros at NEWLINE the beginning of the file prtpd.c: NEWLINE #define _PR_LOCK_TPINDEX() _PR_INTSOFF(_is) NEWLINE #define _PR_UNLOCK_TPINDEX() _PR_INTSON (_is)
155|warning fix : ns/nsprpub/pr/tests/join.c|Created by Rick Osborne (rick@rixsoft.com) on Sunday, April 12, 1998 6:19:29 PM PDT NEWLINE Additional Details : NEWLINE (In the continuing quest to eliminate even compiler warnings NEWLINE ... ) NEWLINE NEWLINE The compiler was complaining that it wanted it to return NEWLINE something. I cast NEWLINE it as void to shut the compiler up. NEWLINE NEWLINE --- join.c Sun Apr 12 02:22:33 1998 NEWLINE +++ join.old.c Wed Apr 08 17:55:11 1998 NEWLINE @@ -69,4 +69,3 @@ NEWLINE NEWLINE -/* Cast as void to eliminate compiler warnings */ NEWLINE -static void Test_Result (int result) NEWLINE +static Test_Result (int result) NEWLINE { NEWLINE Updated by Wan-Teh Chang (wtc@netscape.com) on Friday, April 24, 1998 12:41:25 PM PDT NEWLINE Additional Details : NEWLINE Assigned bug to myself and closed the bug. NEWLINE This is fixed in ns/nsprpub/pr/tests/join.c, NEWLINE revision 3.2.
156|warning fix : ns/nsprpub/pr/tests/parent.c|Created by Rick Osborne (rick@rixsoft.com) on Sunday, April 12, 1998 6:20:34 PM PDT NEWLINE Additional Details : NEWLINE (In the continuing quest to eliminate even compiler warnings NEWLINE ... ) NEWLINE NEWLINE Yet another ''I want to return something'' warning. NEWLINE NEWLINE Also, me in my stupor, I''ve had my diffs backwards all night NEWLINE and didn''t realize it. Oops. It''s what I get for NEWLINE programming at this time of night. NEWLINE NEWLINE --- parent.old.c Wed Apr 08 17:55:13 1998 NEWLINE +++ parent.c Sun Apr 12 02:37:27 1998 NEWLINE @@ -36,3 +36,4 @@ NEWLINE NEWLINE -static PrintUsage(void) NEWLINE +/* Cast as void to eliminate compiler warnings about no NEWLINE return */ NEWLINE +static void PrintUsage(void) NEWLINE { NEWLINE Updated by Wan-Teh Chang (wtc@netscape.com) on Friday, April 24, 1998 12:42:25 PM PDT NEWLINE Additional Details : NEWLINE Assigned the bug to myself and closed the bug. NEWLINE This is fixed in ns/nsprpub/pr/tests/parent.c, NEWLINE revision 3.2.
157|warning fix : ns/nsprpub/pr/tests/select2.c|Created by Rick Osborne (rick@rixsoft.com) on Sunday, April 12, 1998 6:21:26 PM PDT NEWLINE Additional Details : NEWLINE (In the continuing quest to eliminate even compiler warnings NEWLINE ... ) NEWLINE NEWLINE Yet another ''I want to return something'' warning, and a NEWLINE couple of unreferenced vars to boot. Gee, Wally, this is NEWLINE getting exciting! NEWLINE NEWLINE --- select2.old.c Wed Apr 08 17:55:15 1998 NEWLINE +++ select2.c Sun Apr 12 02:46:35 1998 NEWLINE @@ -66,3 +66,4 @@ NEWLINE NEWLINE -static Test_Result (int result) NEWLINE +/* Cast as void to avoid ''no return value'' warnings */ NEWLINE +static void Test_Result (int result) NEWLINE { NEWLINE @@ -106,3 +107,4 @@ NEWLINE PRNetAddr serverAddr; NEWLINE - PRThread *WorkerThread; NEWLINE + /* We never use this, so why do we have it here? */ NEWLINE + /* PRThread *WorkerThread; */ NEWLINE NEWLINE @@ -187,3 +189,4 @@ NEWLINE PRNetAddr serverAddr; NEWLINE - PRThread *WorkerThread; NEWLINE + /* Never use this one, either */ NEWLINE + /* PRThread *WorkerThread; */ NEWLINE NEWLINE Updated by Wan-Teh Chang (wtc@netscape.com) on Friday, April 24, 1998 12:43:24 PM PDT NEWLINE Additional Details : NEWLINE Assigned bug to myself. Closed the bug. NEWLINE This is fixed in ns/nsprpub/pr/tests/select2.c, NEWLINE revision 3.2.
158|warning fix : ns/nsprpub/pr/tests/thruput.c|Created by Rick Osborne (rick@rixsoft.com) on Sunday, April 12, 1998 6:22:17 PM PDT NEWLINE Additional Details : NEWLINE (In the continuing quest to eliminate even compiler warnings NEWLINE ... ) NEWLINE NEWLINE Simple fix. Initial streams is always positive and some NEWLINE equality tests were done agains another unsigned, so I NEWLINE changed it to unsigend to stave off the warnings. Other NEWLINE code only increments or converts from a command-line value. NEWLINE (You can''t have negative streams, anyway, unless maybe NEWLINE you''re in Bizarro-nix.) NEWLINE NEWLINE --- thruput.old.c Wed Apr 08 17:55:15 1998 NEWLINE +++ thruput.c Sun Apr 12 02:56:23 1998 NEWLINE @@ -53,3 +53,3 @@ NEWLINE static PRInt32 xport_buffer = -1; NEWLINE -static PRInt32 initial_streams = 1; NEWLINE +static PRUint32 initial_streams = 1; /* no negative, NEWLINE eliminate a warning */ NEWLINE static PRInt32 buffer_size = BUFFER_SIZE; NEWLINE Updated by Wan-Teh Chang (wtc@netscape.com) on Friday, April 24, 1998 12:47:44 PM PDT NEWLINE Additional Details : NEWLINE Assigned the bug to myself. Closed the bug. NEWLINE This is fixed in ns/nsprpub/pr/tests/thruput.c, NEWLINE revision 3.2.
159|warning fix : ns/js/jsj/jsStubs.c|Created by Rick Osborne (rick@rixsoft.com) on Sunday, April 12, 1998 6:23:11 PM PDT NEWLINE Additional Details : NEWLINE (In the continuing quest to eliminate even compiler warnings NEWLINE ... ) NEWLINE NEWLINE We''re back to compiler directives mucking with var scope. NEWLINE The fun. NEWLINE NEWLINE --- jsstubs.old.c Wed Apr 08 17:52:13 1998 NEWLINE +++ jsstubs.c Sun Apr 12 03:16:21 1998 NEWLINE @@ -304,3 +304,5 @@ NEWLINE JSSavedState saved; NEWLINE +#ifdef JAVA /* only referenced if defined */ NEWLINE const char *cstr; NEWLINE +#endif NEWLINE NEWLINE @@ -324,3 +326,5 @@ NEWLINE NEWLINE +#ifdef JAVA /* only referenced if defined */ NEWLINE do_exit: NEWLINE +#endif NEWLINE exitJS(env, self, cx, jso, &saved); NEWLINE @@ -496,4 +500,6 @@ NEWLINE JSSavedState saved; NEWLINE +#ifdef JAVA /* only referenced if defined */ NEWLINE JSString *jsstr; NEWLINE char *cstr; NEWLINE +#endif NEWLINE struct java_lang_String *ret; NEWLINE Updated by Mike McCabe (mccabe@netscape.com) on Wednesday, April 29, 1998 8:42:20 AM PDT NEWLINE Additional Details : NEWLINE I poked in the changes... they followed the same pattern as many existing NEWLINE #ifdefs in the file.
160|fatal error during creating WIN32_D.OBJsdlg32.dll|Created by shu zhang (zhang@cs.uri.edu) on Sunday, April 12, 1998 6:26:28 PM PDT NEWLINE Additional Details : NEWLINE can not open input file "uuid2.lib" NEWLINE I can not find this library file.
163|[PP]NSPR: pathnames that start with /dev, /bin, /etc are not allowed on the Mac|Created by Hisakuni Fujimoto (hisa@imasy.or.jp) on Monday, April 13, 1998 12:07:40 AM PDT NEWLINE Additional Details : NEWLINE In function ConvertUnixPathToMacPath() of ./ns/nsprpub/pr/ NEWLINE src/md/mac/macio.c. NEWLINE NEWLINE both 03-31, 04-08 version. NEWLINE NEWLINE If one of HDD volumes name starts with some specified NEWLINE strings (dev, bin, etc) , display debug message "Unable to NEWLINE translate Unix file path %s to Mac path" and immediately NEWLINE exit. NEWLINE NEWLINE patch for my Macintosh environment only bellow... NEWLINE NEWLINE --- macio.c.orig Mon Mar 30 18:15:06 1998 NEWLINE +++ macio.c Thu Apr 9 16:42:46 1998 NEWLINE @@ -1258,3 +1258,3 @@ NEWLINE NEWLINE - else if (strncmp(unixPath, "/dev", strlen("/dev")) = NEWLINE = 0) { NEWLINE + else if (strcmp(unixPath, "/dev") == 0) { NEWLINE dprintf("Unable to translate Unix file path %s NEWLINE to Mac path\\n", unixPath); NEWLINE @@ -1731,3 +1731,3 @@ NEWLINE NEWLINE - else if (strncmp(unixPath, "/dev", strlen("/dev")) = NEWLINE = 0) { NEWLINE + else if (strcmp(unixPath, "/dev") == 0) { NEWLINE dprintf("Unable to translate Unix file path %s NEWLINE to Mac path\\n", unixPath); NEWLINE NEWLINE NEWLINE Updated by Mike Pinkerton (pinkerton@netscape.com) on Monday, April 13, 1998 12:20:40 PM PDT NEWLINE Additional Details : NEWLINE The fix we are implementing is to remove the lines that check for /dev /bin /etc NEWLINE because they are pointless. I should be able to have a volume called "dev" if I NEWLINE want to. NEWLINE Updated by Wan-Teh Chang (wtc@netscape.com) on Friday, April 24, 1998 12:49:58 PM PDT NEWLINE Additional Details : NEWLINE Assigned bug to myself. NEWLINE NEWLINE This has been fixed in ns/nsprpub/pr/src/md/mac/macio.c, NEWLINE revision 3.2, as Mike Pinkerton outlined above. NEWLINE NEWLINE Closed the bug. NEWLINE Updated by Wan-Teh Chang (wtc@netscape.com) on Friday, April 24, 1998 12:52:20 PM PDT NEWLINE Additional Details : NEWLINE Made the "Summary Description" more descriptive.
173|Results of POST''s (almost?) always expired from cache|Created by Dan McGuirk (mcguirk@indirect.com) on Monday, April 13, 1998 3:22:46 AM PDT NEWLINE Additional Details : NEWLINE Pages that are generated as a result of POST''s seem to NEWLINE always or almost always be expired from the cache when you NEWLINE try to go back to them. E.g., pick a bug from the Bugzilla NEWLINE bugs list, then hit the back button. With 4.05 the page is NEWLINE still there but with 5.0 it''s expired.
174|URL: mailto:'s -> Failed assert on exit at mkgeturl.c:5855 (net_CleanupMailtoStub)|Created by Dan McGuirk (mcguirk@indirect.com) on Monday, April 13, 1998 3:25:18 AM PDT NEWLINE Additional Details : NEWLINE Exiting the browser triggers an XP_ASSERT(0) on the NEWLINE unimplemented function net_CleanupMailtoStub.
175|Unix NavCenter window title still refers to ''Netscape''|Created by Dan McGuirk (mcguirk@indirect.com) on Monday, April 13, 1998 3:29:21 AM PDT NEWLINE Additional Details : NEWLINE ''Netscape'' is the title bar of the Unix NavCenter window. NEWLINE NEWLINE Should these lines in cmd/xfe/name.h be changed? NEWLINE NEWLINE #define XFE_NAME Netscape NEWLINE #define XFE_PROGNAME netscape NEWLINE #define XFE_PROGCLASS Netscape NEWLINE NEWLINE Also, the copyright notice in cmd/xfe/name.h should NEWLINE include 1998.
176|Closing last window sometimes doesn''t kill process|Created by Dan McGuirk (mcguirk@indirect.com) on Monday, April 13, 1998 3:36:04 AM PDT NEWLINE Additional Details : NEWLINE Sometimes, closing all of the Mozilla windows is not enough NEWLINE to get the process to die. Attaching to the process reveals NEWLINE that it is in _PR_CPU_Idle. Only kill -9 will kill the NEWLINE process. NEWLINE Updated by Ramiro Estrugo (ramiro@netscape.com) on Wednesday, April 29, 1998 2:36:19 AM PDT NEWLINE Additional Details : NEWLINE This problem happened when you opened the nav center frame. It was fixed by NEWLINE slamm@netscape.com yesterday. NEWLINE NEWLINE Updated by Ramiro Estrugo (ramiro@netscape.com) on Wednesday, April 29, 1998 2:37:04 AM PDT NEWLINE Additional Details : NEWLINE This problem happened when you opened the nav center frame. It was fixed by NEWLINE slamm@netscape.com yesterday.
177|Running NavCenter when another browser has lock gives asserts|Created by Dan McGuirk (mcguirk@indirect.com) on Monday, April 13, 1998 3:43:34 AM PDT NEWLINE Additional Details : NEWLINE If you run the NavCenter when another browser has the lock, NEWLINE you get a lot of asserts from nlcstore.c (line 262 mostly, NEWLINE and 294, and possibly one other.) NEWLINE NEWLINE Maybe this is not really a bug, but shouldn''t asserts only NEWLINE fail on things that should never, ever happen, not normal NEWLINE error conditions? NEWLINE Updated by Steve Lamm (slamm@netscape.com) on Thursday, April 30, 1998 10:45:00 AM PDT NEWLINE Additional Details : NEWLINE We need to add some logic that only allows you to run one Browser at a time, or NEWLINE that makes the addition Browsers read-only. NEWLINE NEWLINE If you do need to run two at once, you can create a second home directory. I NEWLINE usually test in a extra home directory. For instance, NEWLINE NEWLINE setenv HOME /u/slamm/.testhome NEWLINE NEWLINE For starters, we could modify the code that pops up the lock dialog to popup a NEWLINE dialog explaining the situation and then quiting. We should probably allow NEWLINE people to force the lock to go away too.
178|"Back" after following internal link is broken|Created by Dan McGuirk (mcguirk@indirect.com) on Monday, April 13, 1998 4:05:19 AM PDT NEWLINE Additional Details : NEWLINE Follow a link to an anchor within the same document, then NEWLINE hit the back button--you wind up at the previous document, NEWLINE not the same one but before going to the anchor. Also, NEWLINE things get screwed up after this--if you hit forward to go NEWLINE back to the original document, you subsequently can''t click NEWLINE on any of the links.
183|Row Cells crash for et_HandleEvent_EvalAck at http://www.zdnet.com/zdnn|Created by Rick Niles (niles@axp745.gsfc.nasa.gov) on Monday, April 13, 1998 11:29:15 AM PDT NEWLINE Additional Details : NEWLINE Mozilla crashs when loading: NEWLINE NEWLINE http://www.zdnet.com/zdnn NEWLINE NEWLINE Happens for all platforms (tested on Win95, Linux). NEWLINE NEWLINE Here''s the traceback: NEWLINE NEWLINE Program received signal SIGSEGV, Segmentation fault. NEWLINE 0x82825ea in lo_cell_relayout_pass (context=<error type>, NEWLINE state=<error type>, NEWLINE table=<error type>, cell_array=0x8b02fc0, NEWLINE blank_cell=<error type>, cell_pad={type = 0, NEWLINE size = <incomplete type>, creationTime = NEWLINE PR_TRANSMITFILE_KEEP_OPEN, NEWLINE modifyTime = 138268048}, rowspan_pass=0xbfffd512, NEWLINE relayout=0xffd55400 <abuf.42+3214434576>) at NEWLINE laytable.c:6462 NEWLINE 6462 row_max = row_max->next; NEWLINE (gdb) where NEWLINE #0 0x82825ea in lo_cell_relayout_pass (context=<error NEWLINE type>, state=<error type>, NEWLINE table=<error type>, cell_array=0x8b02fc0, NEWLINE blank_cell=<error type>, cell_pad={type = 0, NEWLINE size = <incomplete type>, creationTime = NEWLINE PR_TRANSMITFILE_KEEP_OPEN, NEWLINE modifyTime = 138268048}, rowspan_pass=0xbfffd512, NEWLINE relayout=0xffd55400 <abuf.42+3214434576>) at NEWLINE laytable.c:6462 NEWLINE #1 0x82836b9 in lo_EndTable (context=<error type>, NEWLINE state=<error type>, table=<error type>, NEWLINE relayout=0xffd57c00 <abuf.42+3214444816>) at NEWLINE laytable.c:7178 NEWLINE #2 0x8287ceb in lo_CloseTable (context=<error type>, NEWLINE state=<error type>) at laytags.c:1951 NEWLINE #3 0x8287ec5 in lo_process_table_tag (context=<error type>, NEWLINE state=<error type>, NEWLINE tag=<error type>) at laytags.c:2053 NEWLINE #4 0x828ec9d in lo_LayoutTag (context=<error type>, NEWLINE state=<error type>, tag=<error type>) NEWLINE at laytags.c:5922 NEWLINE #5 0x826a487 in lo_FlushBlockage (context=<error type>, NEWLINE state=<error type>, NEWLINE main_doc_state=<error type>) at layout.c:2551 NEWLINE #6 0x82f8c63 in lo_UnblockLayout (context=<error type>, NEWLINE top_state=<error type>) NEWLINE at layscrip.c:588 NEWLINE #7 0x82f8941 in lo_unblock_script_tag (context=<error NEWLINE type>, NEWLINE messWithParser=0xffd7d800 <abuf.42+3214599440>) at NEWLINE layscrip.c:390 NEWLINE #8 0x82f8be3 in lo_StyleEvalExitFn (data={void ( NEWLINE <unknown type>)} 0xbfffd7e0 <abuf.42+2143479024>, NEWLINE str=<incomplete type>, len=0, NEWLINE wysiwyg_url=<incomplete type>, base_href=<incomplete NEWLINE type>, NEWLINE valid=0xffd7fc01 <abuf.42+3214608657>) at layscrip.c:575 NEWLINE #9 0x83dcddb in et_HandleEvent_EvalAck (e=0x8ad0280) at NEWLINE et_moz.c:1762 NEWLINE #10 0x8430ef7 in PL_HandleEvent (self=0x8ad0280) at NEWLINE plevent.c:398 NEWLINE #11 0x8430e0d in PL_ProcessPendingEvents (self=0x874b200) at NEWLINE plevent.c:360 NEWLINE #12 0x80a5535 in fe_event_processor_callback (closure=0x0, NEWLINE fd=0x89710b4, id=0xbfffd864) NEWLINE at mozilla.c:1756 NEWLINE #13 0x4017a94d in XtAppProcessEvent () NEWLINE #14 0x80a45db in fe_EventLoop () at mozilla.c:1075 NEWLINE #15 0x80a74b1 in main (argc=1, argv=0xbffff404) at NEWLINE mozilla.c:3323 NEWLINE NEWLINE ----------- NEWLINE NEWLINE Obviously, we are moving past the end of the linked-list NEWLINE row_list. NEWLINE NEWLINE However, everything looked foobar''ed back to NEWLINE et_HandleEvent_EvalAck NEWLINE (et_moz.c) with those <error type> for the values. It think NEWLINE there''s NEWLINE something wrong with the event itself. NEWLINE NEWLINE Thanks, NEWLINE Rick Niles.
184|Unable to compile with Motif 2.0|Created by Kevin Brannen (kevin@inetspace.com) on Monday, April 13, 1998 12:07:35 PM PDT NEWLINE Additional Details : NEWLINE Unable to compile ns/cmd/xfe/XfeWidgets/Xfe/ because NEWLINE StringDefs.h has XmNlist inside: NEWLINE #if XmVersion < 2000 NEWLINE when it should be outside. There is no XmNlist define in NEWLINE any of the Motif headers. NEWLINE NEWLINE Platform: Linux (Slakware 3.4) NEWLINE Kernel: 2.0.30 NEWLINE X: XFree 3.3.1 (X11R6) NEWLINE Motif: 2.0.0 (from Metrolink) NEWLINE compiler: gcc 2.7.2.1 NEWLINE NEWLINE diff -c: NEWLINE NEWLINE *** StringDefs.h Wed Apr 1 21:26:38 1998 NEWLINE --- StringDefs.h.orig Wed Apr 1 21:27:56 1998 NEWLINE *************** NEWLINE *** 498,507 **** NEWLINE NEWLINE /*----------------------------------------------------------------------*/ NEWLINE #if XmVersion < 2000 NEWLINE #define XmNdragStartCallback NEWLINE "dragStartCallback" NEWLINE #define NEWLINE XmNcomboBoxType NEWLINE "comboBoxType" NEWLINE #define NEWLINE XmCComboBoxType NEWLINE "ComboBoxType" NEWLINE #define NEWLINE XmRComboBoxType NEWLINE "ComboBoxType" NEWLINE #endif NEWLINE - #define NEWLINE XmNlist NEWLINE "list" NEWLINE NEWLINE NEWLINE #endif NEWLINE /* end StringDefs.h */ NEWLINE --- 498,507 ---- NEWLINE NEWLINE /*----------------------------------------------------------------------*/ NEWLINE #if XmVersion < 2000 NEWLINE #define XmNdragStartCallback NEWLINE "dragStartCallback" NEWLINE + #define NEWLINE XmNlist NEWLINE "list" NEWLINE #define NEWLINE XmNcomboBoxType NEWLINE "comboBoxType" NEWLINE #define NEWLINE XmCComboBoxType NEWLINE "ComboBoxType" NEWLINE #define NEWLINE XmRComboBoxType NEWLINE "ComboBoxType" NEWLINE #endif NEWLINE NEWLINE NEWLINE #endif NEWLINE /* end StringDefs.h */
193|Non-Latin-1 strings in frameListmenu are not displayed properly|Created by Jungshik Shin (jshin@pantheon.yale.edu) on Monday, April 13, 1998 4:19:32 PM PDT NEWLINE Additional Details : NEWLINE Open multiple browser windows and make each of them display NEWLINE web pages NEWLINE with non-Latin-1 strings as their titles. FrameList menu NEWLINE display NEWLINE titles of those pages in Compound text encoding. This is NEWLINE always the NEWLINE case in Communicator/Navigator up to 4.05. In Mozilla NEWLINE (19980331), NEWLINE it''s not consistently the case. For some pages, Frame List NEWLINE menu NEWLINE displays the title correctly while for other pages, Frame NEWLINE List menu NEWLINE displays the title in Compound text encoding. NEWLINE ns/cmd/xfe/src/FrameListmenu.cpp NEWLINE gets the title of the frame using NEWLINE XFE_Frame->getTitle().(defined in Frame.cpp). NEWLINE The way it returns the title of the frame is inconsistent. NEWLINE Sometimes, it NEWLINE returns it in locale-dependent encoding and othertimes it NEWLINE returns it NEWLINE in Compound Text encoding of Xlib. In the former case, a NEWLINE subsequent call NEWLINE to convert locale-dependent encoding to Compound String of NEWLINE Motif leads NEWLINE to the correct result while in the latter case, it doesn''t. NEWLINE Updated by Ramiro Estrugo (ramiro@netscape.com) on Friday, April 24, 1998 9:29:47 PM PDT NEWLINE Additional Details : NEWLINE Assigned to ramiro@netscape.com NEWLINE NEWLINE I think we can fix this by not using XFE_Frame::getTitle() which as you indicate NEWLINE seems to be flaky. NEWLINE NEWLINE Instead we can store the title when the back end calls XFE_SetDocTitle() and NEWLINE used that stored string in the frame. NEWLINE NEWLINE So that when XFE_FrameListMenu iterates through the frames, it gets what the NEWLINE backend thinks should be the right thing for the title. This we know will work, NEWLINE since titles for documents are not broken in the BE as far as we can tell.
194|Client Side Image Maps Take a "Long Time" to Load|Created by Shawn Steele (shawn@terra.org) on Monday, April 13, 1998 5:09:31 PM PDT NEWLINE Additional Details : NEWLINE Large client side image maps take a long time to be NEWLINE processed. (Compared to say Netscape 2.x, where it was much NEWLINE faster). In addition, the user doesn NEWLINE Updated by Shawn Steele (shawn@terra.org) on Tuesday, April 14, 1998 1:07:56 PM PDT NEWLINE Additional Details : NEWLINE Oops, I seem to have made a mistake. I don''t think this is a problem. I had NEWLINE noticed this in 4.0x, and thought I had repro''d it here, but I think I was NEWLINE getting confused by the lack of Java. Sorry.
195|warning fix : ns/lib/liblayer/src/cl_comp.c|Created by Rick Osborne (rick@rixsoft.com) on Monday, April 13, 1998 5:34:45 PM PDT NEWLINE Additional Details : NEWLINE (In the continuing quest to eliminate even compiler warnings NEWLINE ... ) NEWLINE NEWLINE Updated by Rick Osborne (rick@rixsoft.com) on Monday, April 13, 1998 5:37:31 PM PDT NEWLINE Additional Details : NEWLINE (Fixed subject, wrong path) NEWLINE Updated by (vidur@netscape.com) on Friday, April 17, 1998 11:18:42 AM PDT NEWLINE Additional Details : NEWLINE Fixed. Thanks, Rick!
196|warning fix : ns/lib/layout/editor.c|Created by Rick Osborne (rick@rixsoft.com) on Monday, April 13, 1998 5:54:31 PM PDT NEWLINE Additional Details : NEWLINE (In the continuing quest to eliminate even compiler warnings NEWLINE ... ) NEWLINE NEWLINE warning : ''IsPositioned'': member function definition looks NEWLINE like a ctor, but name does not match enclosing class NEWLINE NEWLINE This function is only called once in editor.cpp and is in an NEWLINE if() with a subclass member that returns and int, so I NEWLINE figured an int return type would be the most appropriate. I NEWLINE guess VC5 doesn''t like member functions with no return types NEWLINE if they aren''t ctors? NEWLINE NEWLINE --- editor.old.h Wed Apr 08 17:52:21 1998 NEWLINE +++ editor.h Mon Apr 13 20:49:51 1998 NEWLINE @@ -1990,3 +1990,3 @@ public: NEWLINE CEditElement* Element() { return m_pElement; } NEWLINE - IsPositioned(){ return m_pElement != 0; } NEWLINE + int IsPositioned(){ return m_pElement != 0; } NEWLINE }; NEWLINE Updated by (brade@netscape.com) on Thursday, April 16, 1998 8:44:04 PM PDT NEWLINE Additional Details : NEWLINE Reassign bug to myself since I'm the owner of this file.
197|warning fix : ns/lib/layout/edtbuf.c|Created by Rick Osborne (rick@rixsoft.com) on Monday, April 13, 1998 6:10:03 PM PDT NEWLINE Additional Details : NEWLINE (In the continuing quest to eliminate even compiler warnings NEWLINE ... ) NEWLINE NEWLINE (14345) : warning : ''formatting'' : unreferenced local NEWLINE variable NEWLINE (14345) : warning : ''i'' : unreferenced local variable NEWLINE (14451) : warning : ''i'' : unreferenced local variable NEWLINE NEWLINE Basic stuff. We didn''t get an warning for selection, but NEWLINE that''s just because it''s an object instead of a primitive. NEWLINE We don''t use it anyway. (Look ma, I saved an object NEWLINE creation/deletion!) NEWLINE NEWLINE --- edtbuf.old.cpp Wed Apr 08 17:52:23 1998 NEWLINE +++ edtbuf.cpp Mon Apr 13 21:10:17 1998 NEWLINE @@ -14320,7 +14320,9 @@ CEditBuffer::ReplaceOnce( char NEWLINE *pReplace NEWLINE you want replaced has been selected already. */ NEWLINE NEWLINE +#ifdef FORMATING_FIXED /* these vars use only in this NEWLINE context */ NEWLINE CEditSelection selection; NEWLINE int i; NEWLINE EDT_CharacterData *formatting; /* This will hold the NEWLINE formatting of the selected text */ NEWLINE +#endif NEWLINE NEWLINE #ifdef FORMATING_FIXED NEWLINE @@ -14352,5 +14354,6 @@ CEditBuffer::ReplaceLoop(char NEWLINE *pReplaceT NEWLINE XP_Bool bBackward, XP_Bool bDoWrap NEWLINE ) NEWLINE { NEWLINE - int i; NEWLINE + /* We don''t need this, but I''m sure it was a good idea NEWLINE at the time */ NEWLINE + /* int i; */ NEWLINE LO_Element *start_ele_loc, *end_ele_loc, NEWLINE *original_start_ele_loc, NEWLINE *original_end_ele_loc; NEWLINE Updated by (brade@netscape.com) on Thursday, April 16, 1998 8:45:03 PM PDT NEWLINE Additional Details : NEWLINE Reassign bug to myself since I'm the owner of this file.
198|warning fix : ns/lib/layout/layedit.c|Created by Rick Osborne (rick@rixsoft.com) on Monday, April 13, 1998 6:44:09 PM PDT NEWLINE Additional Details : NEWLINE (In the continuing quest to eliminate even compiler warnings NEWLINE ... ) NEWLINE NEWLINE Buckets o'' fun! NEWLINE NEWLINE (1075) : warning C4101: ''pTagElement'' : unreferenced local NEWLINE variable NEWLINE (1075) : warning C4101: ''bItemComplete'' : unreferenced local NEWLINE variable NEWLINE (1075) : warning C4101: ''element'' : unreferenced local NEWLINE variable NEWLINE (1075) : warning C4101: ''elementType'' : unreferenced local NEWLINE variable NEWLINE (2049) : warning C4101: ''doc_id'' : unreferenced local NEWLINE variable NEWLINE (2094) : warning C4101: ''end'' : unreferenced local variable NEWLINE (2094) : warning C4101: ''array'' : unreferenced local NEWLINE variable NEWLINE (2094) : warning C4101: ''start'' : unreferenced local NEWLINE variable NEWLINE (2094) : warning C4101: ''eptr'' : unreferenced local variable NEWLINE NEWLINE --- layedit.old.c Wed Apr 08 17:52:23 1998 NEWLINE +++ layedit.c Mon Apr 13 21:45:19 1998 NEWLINE @@ -881,8 +881,9 @@ void LO_EditorReflow(MWContext *context, NEWLINE int32 iEndLine = -1; NEWLINE LO_Element *leadingMquotes = NULL; NEWLINE - ED_Element *pTagElement; NEWLINE - Bool bItemComplete; NEWLINE - LO_Element * element; NEWLINE - ED_ElementType elementType; NEWLINE + /* Unused */ NEWLINE + /* ED_Element *pTagElement; */ NEWLINE + /* Bool bItemComplete; */ NEWLINE + /* LO_Element * element; */ NEWLINE + /* ED_ElementType elementType; */ NEWLINE LO_Element * startElement; NEWLINE LO_Element * endElement; NEWLINE @@ -2059,9 +2060,10 @@ lo_VerifyLayoutImplementation( NEWLINE MWContext NEWLINE int32 doc_id; NEWLINE lo_TopState *top_state; NEWLINE - lo_DocState *state; NEWLINE - LO_Element **array; NEWLINE - LO_Element *eptr; NEWLINE - LO_Element *start; NEWLINE - LO_Element *end; NEWLINE + lo_DocState *state; NEWLINE + /* Unused vars */ NEWLINE + /* LO_Element **array; */ NEWLINE + /* LO_Element *eptr; */ NEWLINE + /* LO_Element *start; */ NEWLINE + /* LO_Element *end; */ NEWLINE #ifdef XP_WIN16 NEWLINE XP_Block *larray_array; NEWLINE Updated by (brade@netscape.com) on Thursday, April 16, 1998 8:45:32 PM PDT NEWLINE Additional Details : NEWLINE Reassign bug to myself since I'm the owner of this file.
199|warning fix : ns/lib/layout/layblock.c|Created by Rick Osborne (rick@rixsoft.com) on Monday, April 13, 1998 6:59:11 PM PDT NEWLINE Additional Details : NEWLINE (In the continuing quest to eliminate even compiler warnings NEWLINE ... ) NEWLINE NEWLINE Var i is only used in an unsigned context, and is compared NEWLINE against an unsigned struct member later. Therefore, should NEWLINE be unsigned. NEWLINE NEWLINE (1316) : warning C4018: ''<'' : signed/unsigned mismatch NEWLINE NEWLINE --- layblock.old.c Wed Apr 08 17:52:23 1998 NEWLINE +++ layblock.c Mon Apr 13 22:01:25 1998 NEWLINE @@ -1271,5 +1271,5 @@ lo_insert_suppress_tags(MWContext NEWLINE *conte NEWLINE { NEWLINE PA_Tag *end_tag; NEWLINE - int32 i; NEWLINE + uint32 i; NEWLINE NEWLINE /*
200|warning fix : ns/lib/layout/laycols.c|Created by Rick Osborne (rick@rixsoft.com) on Monday, April 13, 1998 7:09:13 PM PDT NEWLINE Additional Details : NEWLINE (In the continuing quest to eliminate even compiler warnings NEWLINE ... ) NEWLINE NEWLINE (843) : warning C4101: ''ele'' : unreferenced local variable NEWLINE NEWLINE --- laycols.old.c Wed Apr 08 17:52:23 1998 NEWLINE +++ laycols.c Mon Apr 13 22:11:39 1998 NEWLINE @@ -606,5 +606,6 @@ lo_EndMulticolumn(MWContext *context, lo NEWLINE LO_Element *cell_ele; NEWLINE LO_TableStruct *table_ele; NEWLINE - LO_Element *ele; NEWLINE + /* Unused */ NEWLINE + /* LO_Element *ele; */ NEWLINE NEWLINE cell_ele = NULL;
201|CRASH when scroll HTML window (mouse and/or PAGEup/down)|Created by Matthew Stevens (mstevens@naviant.com) on Monday, April 13, 1998 8:25:17 PM PDT NEWLINE Additional Details : NEWLINE I built on Windows NT 4.0 SP3 w/ latest hotfixes NEWLINE I have MSVC 5.0 with SP3 NEWLINE I followed build procedures perfectly until I ran into BUG NEWLINE 143 so I hacked the .MAK files in a few places to avoid the NEWLINE UUID2.LIB. I built Mozilla.exe and it works great except NEWLINE when I use the scroll bar or hit page down. I will try to NEWLINE build DEBUG and find the bug. matt NEWLINE Updated by Matthew Stevens (mstevens@naviant.com) on Tuesday, April 14, 1998 8:15:35 AM PDT NEWLINE Additional Details : NEWLINE I have rebuilt the baseline with MOZ_DEBUG=1 and I DO NOT experience the problem NEWLINE anymore. I will continue to use and test with the DEBUG version for now and NEWLINE move to the release build later,
202|warning fix : ns/lib/layout/layscrip.c|Created by Rick Osborne (rick@rixsoft.com) on Monday, April 13, 1998 9:03:49 PM PDT NEWLINE Additional Details : NEWLINE (In the continuing quest to eliminate even compiler warnings NEWLINE ... ) NEWLINE NEWLINE Just a couple of #ifdef statements mucking with variable NEWLINE scope and label usage. NEWLINE NEWLINE (538) : warning C4102: ''out'' : unreferenced label NEWLINE (538) : warning C4101: ''principals'' : unreferenced local NEWLINE variable NEWLINE (538) : warning C4101: ''name'' : unreferenced local variable NEWLINE NEWLINE --- layscrip.old.c Wed Apr 08 17:52:27 1998 NEWLINE +++ layscrip.c Tue Apr 14 00:03:13 1998 NEWLINE @@ -452,4 +452,6 @@ lo_script_archive_exit_fn(URL_Struct *ur NEWLINE ScriptData *data = NULL; NEWLINE +#ifdef JAVA /* Vars only used in JAVA context */ NEWLINE char *name; NEWLINE JSPrincipals *principals; NEWLINE +#endif NEWLINE ETEvalStuff * stuff; NEWLINE @@ -534,3 +536,5 @@ lo_script_archive_exit_fn(URL_Struct *ur NEWLINE NEWLINE +#ifdef JAVA /* Label only used in this context */ NEWLINE out: NEWLINE +#endif NEWLINE /* Always free (or drop a ref on) the url_struct before NEWLINE returning. */
203|warning fix : ns/lib/layout/laysel.c|Created by Rick Osborne (rick@rixsoft.com) on Monday, April 13, 1998 9:11:38 PM PDT NEWLINE Additional Details : NEWLINE (In the continuing quest to eliminate even compiler warnings NEWLINE ... ) NEWLINE NEWLINE Ever more unreferenced vars. NEWLINE NEWLINE (4785) : warning C4101: ''endOfDocument'' : unreferenced local NEWLINE variable NEWLINE (4785) : warning C4101: ''startOfDocument'' : unreferenced NEWLINE local variable NEWLINE NEWLINE --- laysel.old.c Wed Apr 08 17:52:27 1998 NEWLINE +++ laysel.c Tue Apr 14 00:13:31 1998 NEWLINE @@ -4715,3 +4715,4 @@ void LO_Hit(MWContext *context, int32 NEWLINE x, NEWLINE { NEWLINE - LO_Position endOfDocument; NEWLINE + /* Unused */ NEWLINE + /* LO_Position endOfDocument; */ NEWLINE int32 endY; NEWLINE @@ -4733,3 +4734,4 @@ void LO_Hit(MWContext *context, int32 NEWLINE x, NEWLINE { NEWLINE - LO_Position startOfDocument; NEWLINE + /* Unused */ NEWLINE + /* LO_Position startOfDocument; */ NEWLINE int32 startY;
204|warning fix : ns/lib/layout/laytable.c|Created by Rick Osborne (rick@rixsoft.com) on Monday, April 13, 1998 9:38:21 PM PDT NEWLINE Additional Details : NEWLINE (In the continuing quest to eliminate even compiler warnings NEWLINE ... ) NEWLINE NEWLINE These were fun. Apparently the table code was massively NEWLINE overhauled and so someone #ifdef-ed and commented out a NEWLINE whole bunch of stuff, including the code that used these NEWLINE vars. NEWLINE NEWLINE (3246) : warning C4101: ''save_state_pushes'' : unreferenced NEWLINE local variable NEWLINE (3246) : warning C4101: ''tag_end_ptr'' : unreferenced local NEWLINE variable NEWLINE (3246) : warning C4101: ''save_state_pops'' : unreferenced NEWLINE local variable NEWLINE (3246) : warning C4101: ''tag_ptr'' : unreferenced local NEWLINE variable NEWLINE (3246) : warning C4101: ''save_diff_state'' : unreferenced NEWLINE local variable NEWLINE (3246) : warning C4101: ''tag_list'' : unreferenced local NEWLINE variable NEWLINE (3774) : warning C4101: ''save_state_pushes'' : unreferenced NEWLINE local variable NEWLINE (3774) : warning C4101: ''tag_end_ptr'' : unreferenced local NEWLINE variable NEWLINE (3774) : warning C4101: ''save_state_pops'' : unreferenced NEWLINE local variable NEWLINE (3774) : warning C4101: ''tag_ptr'' : unreferenced local NEWLINE variable NEWLINE (3774) : warning C4101: ''elem_list'' : unreferenced local NEWLINE variable NEWLINE (3774) : warning C4101: ''save_diff_state'' : unreferenced NEWLINE local variable NEWLINE (3774) : warning C4101: ''tag_list'' : unreferenced local NEWLINE variable NEWLINE NEWLINE --- laytable.old.c Wed Apr 08 17:52:27 1998 NEWLINE +++ laytable.c Tue Apr 14 00:39:51 1998 NEWLINE @@ -2630,5 +2630,5 @@ lo_RelayoutCaptionSubdoc(MWContext NEWLINE *cont NEWLINE lo_DocState *new_state; NEWLINE - PA_Tag *tag_list; NEWLINE - PA_Tag *tag_ptr; NEWLINE - PA_Tag *tag_end_ptr; NEWLINE + /* ifdef-ed out of service: PA_Tag *tag_list; */ NEWLINE + /* Commented and ifdef-ed out of service: PA_Tag NEWLINE *tag_ptr; */ NEWLINE + /* Commented out of service: PA_Tag *tag_end_ptr; */ NEWLINE int32 doc_id; NEWLINE @@ -2646,5 +2646,5 @@ lo_RelayoutCaptionSubdoc(MWContext NEWLINE *cont NEWLINE NET_ReloadMethod save_force; NEWLINE - Bool save_diff_state; NEWLINE - int32 save_state_pushes; NEWLINE - int32 save_state_pops; NEWLINE + /* ifdef-ed out of service: Bool save_diff_state; */ NEWLINE + /* ifdef-ed out of service: int32 save_state_pushes; NEWLINE */ NEWLINE + /* ifdef-ed out of service: int32 save_state_pops; NEWLINE */ NEWLINE lo_DocLists *doc_lists; NEWLINE @@ -3259,5 +3259,5 @@ lo_RelayoutCell(MWContext *context, NEWLINE lo_D NEWLINE lo_DocState *new_state; NEWLINE - PA_Tag *tag_list; NEWLINE - PA_Tag *tag_ptr; NEWLINE - PA_Tag *tag_end_ptr; NEWLINE + /* Unused: PA_Tag *tag_list; */ NEWLINE + /* COmmented out of service: PA_Tag *tag_ptr; */ NEWLINE + /* Commented out of service: PA_Tag *tag_end_ptr; */ NEWLINE int32 doc_id; NEWLINE @@ -3275,5 +3275,5 @@ lo_RelayoutCell(MWContext *context, NEWLINE lo_D NEWLINE NET_ReloadMethod save_force; NEWLINE - Bool save_diff_state; NEWLINE - int32 save_state_pushes; NEWLINE - int32 save_state_pops; NEWLINE + /* Unused: Bool save_diff_state; */ NEWLINE + /* Unused: int32 save_state_pushes; */ NEWLINE + /* Unused: int32 save_state_pops; */ NEWLINE lo_DocLists *doc_lists; NEWLINE @@ -3283,3 +3283,3 @@ lo_RelayoutCell(MWContext *context, NEWLINE lo_D NEWLINE int32 hold_current_layer_num; NEWLINE - LO_Element * elem_list; NEWLINE + /* Commented out of service: LO_Element * elem_list; NEWLINE */
205|slashdot.org doesn''t display properly|Created by Dan McGuirk (mcguirk@indirect.com) on Tuesday, April 14, 1998 12:30:29 AM PDT NEWLINE Additional Details : NEWLINE The main body of http://slashdot.org/ doesn''t show up NEWLINE correctly when you first go to the page. If you hit back NEWLINE and then forward, it''s almost but not quite completely NEWLINE correct.
213|showDocument("javascript:..."); broken|Created by Holger Hippenstiel (lynxx@uni.de) on Tuesday, April 14, 1998 7:06:16 AM PDT NEWLINE Additional Details : NEWLINE Since Version 4.02 of Communicator "Life" generating of NEWLINE Javascript doesn''t work anymore. NEWLINE NEWLINE The following example works fine in Netscape Navigator NEWLINE 3.01++ or Netscape Communicator 4.01 (Win 95/Win NT/IRIX NEWLINE 5.3, 6.2/AIX 4/HP-UX 10). NEWLINE NEWLINE It "creates" a new Html (with Javascript [also tried NEWLINE "mocha:" and "livescript:"]) and the browser should display NEWLINE "Test". NEWLINE NEWLINE AppletContext is there. NEWLINE The URL is created. NEWLINE No Exception is raised. NEWLINE NEWLINE What''s the problem !?!?! NEWLINE NEWLINE import java.awt.*; NEWLINE import java.applet.*; NEWLINE import java.net.*; NEWLINE NEWLINE public class HtmlCreate extends Applet { NEWLINE public void init() { NEWLINE try { NEWLINE getAppletContext().showDocument(new NEWLINE URL("javascript:''<html><body>Test</body></html>''"), NEWLINE "_self"); NEWLINE System.out.println("Html created"); NEWLINE } catch (Exception e) { NEWLINE System.out.println("Error: "+e); NEWLINE } NEWLINE } NEWLINE } NEWLINE NEWLINE Also tried to give another destination: "_top", and also NEWLINE showDocument(URL) without destination, nothing works !
214|Navigator doesn''t highlight the URL when the URL is given focus using the mouse|Created by Mark Lehmoine (majornut@mindspring.com) on Tuesday, April 14, 1998 7:46:17 AM PDT NEWLINE Additional Details : NEWLINE To reproduce the problem, just give the URL edit the focus NEWLINE using the mouse.
215|When Navigator is closed using the ''exit'' menuitem, it leaves 3 threads running|Created by Mark Lehmoine (majornut@mindspring.com) on Tuesday, April 14, 1998 7:51:36 AM PDT NEWLINE Additional Details : NEWLINE When Navigator closes, it leaves 3 threads running in the NEWLINE system. To completely stop it, I need to kill it using the NEWLINE task manager. NEWLINE NEWLINE To reproduce, just start it up and then shut it down.
216|HTML Tables not drawn properly (bleed between cells)|Created by Matthew Stevens (mstevens@naviant.com) on Tuesday, April 14, 1998 8:26:51 AM PDT NEWLINE Additional Details : NEWLINE When loading an HTML Table, the text in many cells bleed NEWLINE across cells when cell text is long; the text is not wrapped NEWLINE and/or the cells is not streched. NEWLINE NEWLINE matt NEWLINE Updated by Matthew Stevens (mstevens@naviant.com) on Tuesday, April 14, 1998 8:28:28 AM PDT NEWLINE Additional Details : NEWLINE Forgot to mention that this is a DEBUG build. NEWLINE NEWLINE Also, have noticed that there is bleeding of many HTML constructs (including NEWLINE this BUG REPORT page I am looking at now) NEWLINE Updated by Suresh Duddi (dp@netscape.com) on Thursday, April 23, 1998 12:09:17 PM PDT NEWLINE Additional Details : NEWLINE In general, a test case will be really helpful. In this case, I havent yet tried NEWLINE to reproduce it.
217|Netscape Compile Error|Created by David Nebinger (dnebinger@synertech.highmark.com) on Tuesday, April 14, 1998 11:20:12 AM PDT NEWLINE Additional Details : NEWLINE Netscape compile fails on line #124 of NEWLINE ns/modules/libpref/src/prefapi.c. Changing the line from: NEWLINE XP_FREEIF((void*)he->key); NEWLINE to: NEWLINE XP_FREEIF(he->key); NEWLINE NEWLINE fixes the problem. If other compilers complain about the NEWLINE missing typecast, potential resolution would be to create a NEWLINE new macro that takes the type as an argument. Instead of NEWLINE using the macro: NEWLINE #define XP_FREEIF(obj) do { if(obj) { XP_FREE(obj); obj = 0; NEWLINE }} while(0) NEWLINE NEWLINE try the macro: NEWLINE #define XP_FREEIFTYPE(obj,typ) do { if (obj) { XP_FREE(obj); NEWLINE obj = (typ)0;}} while(0) NEWLINE with line #124 as NEWLINE XP_FREEIFTYPE(he->key,void*); NEWLINE NEWLINE Thanks
218|Compile Warning|Created by David Nebinger (dnebinger@synertech.highmark.com) on Tuesday, April 14, 1998 11:31:33 AM PDT NEWLINE Additional Details : NEWLINE When compiling for HP/UX using the ''gmake -k'' command, a NEWLINE compiler warning is generated about the HAVE_STRERROR being NEWLINE re-defined. To remove this compiler warning, change the NEWLINE ns/nsprpub/pr/include/md/_hpux.h file as follows. Surround NEWLINE line # 44 to become: NEWLINE NEWLINE #ifndef HAVE_STRERROR NEWLINE #define HAVE_STRERROR NEWLINE #endif NEWLINE NEWLINE Thanks NEWLINE Updated by Wan-Teh Chang (wtc@netscape.com) on Wednesday, May 6, 1998 7:55:39 PM PDT NEWLINE Additional Details : NEWLINE Assigned bug to myself. NEWLINE Updated by Wan-Teh Chang (wtc@netscape.com) on Wednesday, May 6, 1998 11:38:35 PM PDT NEWLINE Additional Details : NEWLINE Fixed as you suggested, in mozilla/nsprpub/pr/include/md/_hpux.h, NEWLINE revision 3.2. NEWLINE NEWLINE Closed the bug.
219|BUILD: Compiler Error|Created by David Nebinger (dnebinger@synertech.highmark.com) on Tuesday, April 14, 1998 11:43:46 AM PDT NEWLINE Additional Details : NEWLINE When compiling Netscape using HP''s CC, the following error NEWLINE is generated: NEWLINE NEWLINE CC -Aa +a1 -o HP-UXB.10.20_DBG.OBJ/editor.o -c -DXP_UNIX -g NEWLINE +Z -DHPUX -Dhpux -Dhppa -DHPUX10 -Dhpux10 -DHPUX10_20 -D_H NEWLINE PUX_SOURCE -DSW_THREADS -DNO_SIGNED -DNO_FNDELAY NEWLINE -DHAVE_ODD_SELECT -DNO_CDEFS_H -DNO_LONG_LONG -DNEED_IOCTL_H NEWLINE -DNEED_MAT NEWLINE H_H -DUSE_NODL_TABS -DMITSHM -DNEED_SYS_WAIT_H NEWLINE -DHAVE_INT32_T -DNEED_UINT_T -DNEED_H_ERRNO NEWLINE -DRW_NO_OVERLOAD_SCHAR -DHAVE_MODEL_H -DHAVE_STRERROR NEWLINE -DDEBUG -UNDEBUG -DDEBUG_dnebing -DTRACING -DNSPR20 NEWLINE -DNETSCAPE -DOSTYPE=\\"HP-UXB.10\\" -DMOZILL NEWLINE A_CLIENT -DNO_SECURITY -DEDITOR -DMOZ_COMMUNICATOR_IIDS NEWLINE -DEDITOR -DEDITOR_UI -DLAYERS -DUNIX_EMBED -DX_PLUGINS NEWLINE -DUNIX_LD NEWLINE AP -DNSPR -DMOCHA -DUNIX_ASYNC_DNS -DDEVELOPER_DEBUG NEWLINE -I../../dist/public/parse -I../../dist/public/style NEWLINE -I../../dist/p NEWLINE ublic/js -I../../dist/public/layer NEWLINE -I../../dist/public/applet -I../../dist/public/nspr NEWLINE -I../../dist/public/security -I../../dist/public/dbm NEWLINE -I../../dist/public/img -I../../dist/public/util NEWLINE -I../../dist/public/jtools -I../../dist/public/edtp NEWLINE lug -I../../dist/public/java -I../../dist/public/hook NEWLINE -I../../dist/public/pref -I../../dist/public/libfont NEWLINE -I../../dist/ NEWLINE public/lay -I../../dist/public/rdf -I../../include NEWLINE -I../../dist/HP-UXB.10.20_DBG.OBJ/include editor.cpp NEWLINE CC: "../../include/msgcom.h", line 2015: error: long NEWLINE enumerator (1179) NEWLINE NEWLINE I''ve been able to compile by changing line 2015 to use the NEWLINE constant -1 (but not -1L); I don''t know yet if this really NEWLINE fixes the problem or not (I''ll let you know if I have any NEWLINE success with this). NEWLINE NEWLINE Dave
220|Incorrect display of CP-1250 pages in Unix version|Created by Tomas Brodsky (brodsky@cfar.umd.edu) on Tuesday, April 14, 1998 11:54:25 AM PDT NEWLINE Additional Details : NEWLINE Mozilla running in X with Latin 2 fonts installed NEWLINE uses Latin 1 fonts to display pages with win-cp-1250 NEWLINE encoding that advertize (correctly, I suppose) NEWLINE the encoding with a meta tag NEWLINE <META http-equiv="Content-Type" content="text/html; NEWLINE charset=windows-1250"> NEWLINE NEWLINE See for example NEWLINE NEWLINE http://svet.namodro.cz (the text is in Czech). NEWLINE NEWLINE I posted a one line fix for the problem: NEWLINE NEWLINE news://news.mozilla.org/3533A0E4.6EE6@cfar.umd.edu
221|testing my quote|Created by Tim Kaiser (tkaiser@netscape.com) on Tuesday, April 14, 1998 2:09:16 PM PDT NEWLINE Additional Details : NEWLINE testing please ignroe a'asd'fasdflasd;fl asd NEWLINE NEWLINE asdfl;asjd fasdf asa;ldfj sadf sda NEWLINE f NEWLINE asdf NEWLINE asd NEWLINE f NEWLINE <html> NEWLINE <body> NEWLINE <table> NEWLINE <tr> NEWLINE <td>asdf</td> NEWLINE <td>1111</td> NEWLINE </tr> NEWLINE </table> NEWLINE </body> NEWLINE </html>
222|testing again pleae ignore|Created by Tim Kaiser (tkaiser@netscape.com) on Tuesday, April 14, 1998 2:11:07 PM PDT NEWLINE Additional Details : NEWLINE testing please ingore NEWLINE Updated by Tim Kaiser (tkaiser@netscape.com) on Tuesday, April 14, 1998 2:12:00 PM PDT NEWLINE Additional Details : NEWLINE NEWLINE Content-type: text/plain NEWLINE NEWLINE NEWLINE Testing please ingore.. Working NEWLINE Updated by Arun Ramachandran (arunr@netscape.com) on Monday, May 18, 1998 1:20:56 PM PDT NEWLINE Additional Details : NEWLINE Updated by Arun Ramachandran (arunr@netscape.com) on Monday, May 18, 1998 1:21:15 PM PDT NEWLINE Additional Details :
223|Linux 2.1.x, NSPR: duplicate function already exists by default on the system|Created by Luc Stepniewski (lstep@mail.dotcom.fr) on Tuesday, April 14, 1998 3:12:29 PM PDT NEWLINE Additional Details : NEWLINE I'm using a debian Linux, glibc, with kernel 2.1.91. NEWLINE The compilation aborts because the file NEWLINE ns/nsprpub/pr/src/misc/prnetdb.c contains a function NEWLINE declaration that conflicts with one of the predefined NEWLINE function in the kernel. NEWLINE It is the function getprotobynumber_r(PRInt32 number) NEWLINE which already exists and is used for 'reentrant' network NEWLINE use. NEWLINE NEWLINE Here is the full error message, if this can help: NEWLINE gcc -o Linux2.1.91_x86_DBG.OBJ/prnetdb.o -c -g -fPIC -ansi NEWLINE -Wall -DLINUX -Dlinux -mno-486 -DLINUX1_2 -Di386 NEWLINE -D_POSIX_SOURCE -D_BSD_SOURCE -DHAVE_STRERROR -DFORCE_PR_LOG NEWLINE -DXP_UNIX -DDEBUG -UNDEBUG -DDEBUG_enlight NEWLINE -D_PR_LOCAL_THREADS_ONLY NEWLINE -I../../../../dist/Linux2.1.91_x86_DBG.OBJ/include/private NEWLINE -I../../../../dist/Linux2.1.91_x86_DBG.OBJ/include NEWLINE prnetdb.c NEWLINE prnetdb.c:383: conflicting types for `getprotobyname_r' NEWLINE /usr/include/netdb.h:330: previous declaration of NEWLINE `getprotobyname_r' NEWLINE prnetdb.c:388: conflicting types for `getprotobynumber_r' NEWLINE /usr/include/netdb.h:337: previous declaration of NEWLINE `getprotobynumber_r' NEWLINE Updated by Wan-Teh Chang (wtc@netscape.com) on Friday, April 24, 1998 12:59:47 PM PDT NEWLINE Additional Details : NEWLINE You can try one of the two fixes: NEWLINE 1. ns/nsprpub/config/Linux.mk: pretend that you are NEWLINE on a 2.0.x system, i.e., add this NEWLINE ifeq ($(basename $(OS_RELEASE)),2.0) NEWLINE PLATFORM_FLAGS += -DLINUX2_0 NEWLINE endif NEWLINE NEWLINE 2. ns/nsprpub/pr/src/misc/prnetdb.c: change the 'LINUX2_0' NEWLINE macro to 'LINUX'. NEWLINE NEWLINE I think solution 2 alone should fix this problem. But NEWLINE in the meantime, it's probably a good idea to define the NEWLINE LINUX2_0 macro on 2.1.x systems. NEWLINE Updated by Wan-Teh Chang (wtc@netscape.com) on Tuesday, June 16, 1998 8:30:53 PM PDT NEWLINE Additional Details : NEWLINE I checked in solution 2 (checking 'LINUX' NEWLINE instead of 'LINUX2_0' in prnetdb.c), so NEWLINE it should work now. NEWLINE NEWLINE Closed the bug. NEWLINE NEWLINE By the way, I had a typo in my previous message. NEWLINE Solution 1 should read: NEWLINE ifeq ($(basename $(OS_RELEASE)),2.1) NEWLINE PLATFORM_FLAGS += -DLINUX2_0 NEWLINE endif
224|It will not compile!|Created by Mike Machado (mike@innercite.com) on Tuesday, April 14, 1998 4:49:39 PM PDT NEWLINE Additional Details : NEWLINE Here here an error I get: I am using the lesstif headers, NEWLINE but I thought they were supposed to contain the same NEWLINE fucntions as the motif ones... NEWLINE NEWLINE gcc -o Linux2.0.33_x86_DBG.OBJ/Folder.o -c -DXP_UNIX -g NEWLINE -ansi -Wall -pipe -DLINU NEWLINE X -Dlinux -mno-486 -DLINUX1_2 -Di386 -DLINUX2_0 NEWLINE -D_POSIX_SOURCE -D_BSD_SOURCE -D NEWLINE SW_THREADS -DNEED_ENDIAN_H -DNEED_GETOPT_H -DNEED_IOCTL_H NEWLINE -DUSE_NODL_TABS -DHAVE NEWLINE _SIGNED_CHAR -DNEED_SYS_TIME_H -DHAVE_SYS_BITYPES_H NEWLINE -DNEED_UINT_T -DNEED_TIME_R NEWLINE -DMITSHM -D_XOPEN_SOURCE -DNO_INT64_T NEWLINE -D_PR_LOCAL_THREADS_ONLY -DHAVE_STRERROR NEWLINE -DDEBUG -UNDEBUG -DDEBUG_root -DTRACING -DNSPR20 -DNETSCAPE NEWLINE -DOSTYPE=\\"Linux2.0\\ NEWLINE " -DMOZILLA_CLIENT -DNO_SECURITY -DEDITOR NEWLINE -DMOZ_COMMUNICATOR_IIDS -DEDITOR -DEDI NEWLINE TOR_UI -DLAYERS -DUNIX_EMBED -DX_PLUGINS -DUNIX_LDAP -DNSPR NEWLINE -DMOCHA -DUNIX_ASYNC NEWLINE _DNS -DDEVELOPER_DEBUG -I../../../../dist/public/Microline NEWLINE -I../../../../includ NEWLINE e -I/usr/X11R6/include NEWLINE -I../../../../dist/Linux2.0.33_x86_DBG.OBJ/include Fo NEWLINE lder.c NEWLINE gcc -o Linux2.0.33_x86_DBG.OBJ/Grid.o -c -DXP_UNIX -g -ansi NEWLINE -Wall -pipe -DLINUX NEWLINE -Dlinux -mno-486 -DLINUX1_2 -Di386 -DLINUX2_0 NEWLINE -D_POSIX_SOURCE -D_BSD_SOURCE -DSW NEWLINE _THREADS -DNEED_ENDIAN_H -DNEED_GETOPT_H -DNEED_IOCTL_H NEWLINE -DUSE_NODL_TABS -DHAVE_S NEWLINE IGNED_CHAR -DNEED_SYS_TIME_H -DHAVE_SYS_BITYPES_H NEWLINE -DNEED_UINT_T -DNEED_TIME_R -D NEWLINE MITSHM -D_XOPEN_SOURCE -DNO_INT64_T -D_PR_LOCAL_THREADS_ONLY NEWLINE -DHAVE_STRERROR -D NEWLINE DEBUG -UNDEBUG -DDEBUG_root -DTRACING -DNSPR20 -DNETSCAPE NEWLINE -DOSTYPE=\\"Linux2.0\\" NEWLINE -DMOZILLA_CLIENT -DNO_SECURITY -DEDITOR NEWLINE -DMOZ_COMMUNICATOR_IIDS -DEDITOR -DEDITO NEWLINE R_UI -DLAYERS -DUNIX_EMBED -DX_PLUGINS -DUNIX_LDAP -DNSPR NEWLINE -DMOCHA -DUNIX_ASYNC_D NEWLINE NS -DDEVELOPER_DEBUG -I../../../../dist/public/Microline NEWLINE -I../../../../include NEWLINE -I/usr/X11R6/include NEWLINE -I../../../../dist/Linux2.0.33_x86_DBG.OBJ/include Grid NEWLINE .c NEWLINE Grid.c: In function `DropRegister': NEWLINE Grid.c:7132: warning: implicit declaration of function NEWLINE `XmDropSiteRegister' NEWLINE Grid.c:7135: warning: implicit declaration of function NEWLINE `XmDropSiteUnregister' NEWLINE Grid.c: In function `DropStart': NEWLINE Grid.c:7146: `XmDropProcCallbackStruct' undeclared (first NEWLINE use this function) NEWLINE Grid.c:7146: (Each undeclared identifier is reported only NEWLINE once NEWLINE Grid.c:7146: for each function it appears in.) NEWLINE Grid.c:7146: `cbs' undeclared (first use this function) NEWLINE Grid.c:7146: warning: statement with no effect NEWLINE Grid.c:7147: parse error before `te' NEWLINE Grid.c:7153: parse error before `)' NEWLINE Grid.c:7156: `XmINVALID_DROP_SITE' undeclared (first use NEWLINE this function) NEWLINE Grid.c:7159: `valid' undeclared (first use this function) NEWLINE Grid.c:7160: `row' undeclared (first use this function) NEWLINE Grid.c:7160: `col' undeclared (first use this function) NEWLINE Grid.c:7164: `exportTargets' undeclared (first use this NEWLINE function) NEWLINE Grid.c:7165: `n' undeclared (first use this function) NEWLINE Grid.c:7167: `i' undeclared (first use this function) NEWLINE Grid.c:7175: `args' undeclared (first use this function) NEWLINE Grid.c:7175: warning: left-hand operand of comma expression NEWLINE has no effect NEWLINE Grid.c:7176: warning: left-hand operand of comma expression NEWLINE has no effect NEWLINE Grid.c:7183: `te' undeclared (first use this function) NEWLINE Grid.c:7185: warning: left-hand operand of comma expression NEWLINE has no effect NEWLINE Grid.c:7186: warning: left-hand operand of comma expression NEWLINE has no effect NEWLINE Grid.c:7187: warning: left-hand operand of comma expression NEWLINE has no effect NEWLINE make[3]: *** [Linux2.0.33_x86_DBG.OBJ/Grid.o] Error 1 NEWLINE make[3]: Leaving directory NEWLINE `/home/apps/netscape-source/ns/cmd/xfe/Microline3.0/X NEWLINE mL' NEWLINE make[2]: *** [libs] Error 2 NEWLINE make[2]: Leaving directory NEWLINE `/home/apps/netscape-source/ns/cmd/xfe' NEWLINE make[1]: *** [libs] Error 2 NEWLINE make[1]: Leaving directory NEWLINE `/home/apps/netscape-source/ns/cmd' NEWLINE make: *** [libs] Error 2 NEWLINE [root@linus ns]# NEWLINE NEWLINE NEWLINE NEWLINE please let me know if it is my lesstif setup or what. NEWLINE I have version 0.82 NEWLINE Updated by null (null) on Friday, April 24, 1998 9:40:23 PM PDT NEWLINE Additional Details : NEWLINE This problem gets fixed when you use the latest lesstif from lesstif.org. NEWLINE NEWLINE Marking resolved.
225|warning fix : ns/cmd/winfe/cxdc.cpp|Created by Rick Osborne (rick@rixsoft.com) on Tuesday, April 14, 1998 6:38:47 PM PDT NEWLINE Additional Details : NEWLINE (In the continuing quest to eliminate even compiler warnings NEWLINE ... ) NEWLINE NEWLINE Simple unused var. NEWLINE NEWLINE (1729) : warning : 'nStrokeWidth' : unreferenced local NEWLINE variable NEWLINE NEWLINE --- cxdc.old.cpp Wed Apr 08 17:46:31 1998 NEWLINE +++ cxdc.cpp Tue Apr 14 19:52:43 1998 NEWLINE @@ -1621,3 +1621,3 @@ CDCCX::DisplayTableBorder(LTRB& Rect, NEWLINE LO NEWLINE HDC hDC; NEWLINE - int nStrokeWidth; NEWLINE + /* Unused: int nStrokeWidth; */ NEWLINE RECT r = {CASTINT(Rect.left), NEWLINE CASTINT(Rect.top), CASTINT(Rect.right), NEWLINE CASTINT(Rect.bottom)}; NEWLINE NEWLINE Updated by Rick Osborne (rick@rixsoft.com) on Tuesday, April 14, 1998 6:43:55 PM PDT NEWLINE Additional Details : NEWLINE Wrong file specified in subject. Oops. NEWLINE Updated by (blythe@netscape.com) on Wednesday, April 15, 1998 4:38:21 PM PDT NEWLINE Additional Details : NEWLINE I merged in the diff. NEWLINE Thanks Rick, next source drop will have it!
226|warning fix : ns/cmd/winfe/cxprint.cpp|Created by Rick Osborne (rick@rixsoft.com) on Tuesday, April 14, 1998 6:42:26 PM PDT NEWLINE Additional Details : NEWLINE (In the continuing quest to eliminate even compiler warnings NEWLINE ... ) NEWLINE NEWLINE Simple unreferenced var. NEWLINE NEWLINE (2013) : warning : 'hOldPal' : unreferenced local variable NEWLINE NEWLINE --- cxprint.old.cpp Wed Apr 08 17:46:33 1998 NEWLINE +++ cxprint.cpp Tue Apr 14 20:01:01 1998 NEWLINE @@ -1962,3 +1962,3 @@ void NEWLINE CPrintCX::CopyOffscreenBitmap(NI_Pi NEWLINE lpBmi->biBitCount = nBitCount; NEWLINE - HPALETTE hOldPal; NEWLINE + /* Unreferenced: HPALETTE hOldPal; */ NEWLINE // Ask the driver to tell us the number of NEWLINE bits we need to allocate NEWLINE NEWLINE Updated by (blythe@netscape.com) on Wednesday, April 15, 1998 4:40:09 PM PDT NEWLINE Additional Details : NEWLINE I merged in the diff. NEWLINE Thanks Rick, next source drop will have it!
227|warning fix : ns/cmd/winfe/dde.cpp|Created by Rick Osborne (rick@rixsoft.com) on Tuesday, April 14, 1998 6:46:30 PM PDT NEWLINE Additional Details : NEWLINE (In the continuing quest to eliminate even compiler warnings NEWLINE ... ) NEWLINE NEWLINE Simple unreferenced var. NEWLINE NEWLINE (3648) : warning : 'iTraverseIndex' : unreferenced local NEWLINE variable NEWLINE NEWLINE --- dde.old.cpp Wed Apr 08 17:46:35 1998 NEWLINE +++ dde.cpp Tue Apr 14 21:33:41 1998 NEWLINE @@ -3625,3 +3625,3 @@ HDDEDATA NEWLINE CDDEWrapper::CancelProgress(HSZ NEWLINE // transaction ID. NEWLINE - int iTraverseIndex; NEWLINE + /* Unreferenced: int iTraverseIndex; */ NEWLINE MWContext *pTraverseContext = NULL; NEWLINE Updated by (blythe@netscape.com) on Wednesday, April 15, 1998 4:41:12 PM PDT NEWLINE Additional Details : NEWLINE I merged in the diff. NEWLINE Thanks Rick, next source drop will have it!
228|Navigator spawn a lot of error messages in infinite loop.|Created by Denis Pershin (dyp@inetlab.com) on Tuesday, April 14, 1998 11:30:38 PM PDT NEWLINE Additional Details : NEWLINE when I try to see page: NEWLINE http://wwwkoi.nsk.su/Information/ NEWLINE (It is in russian but it does not matter) NEWLINE Navigator opens a lot of message boxes (I wait until 120th) NEWLINE Updated by Denis Pershin (dyp@inetlab.com) on Friday, June 5, 1998 1:58:34 PM PDT NEWLINE Additional Details : NEWLINE Fixed in last build... NEWLINE Updated by Mike McCabe (mccabe@netscape.com) on Monday, June 15, 1998 5:22:33 PM PDT NEWLINE Additional Details : NEWLINE Denis Pershin assures me that this was fixed in a recent build.
229|HTML comments are not closed in the right place|Created by Wilson Afonso (wilson@zaz.com.br) on Wednesday, April 15, 1998 6:19:19 AM PDT NEWLINE Additional Details : NEWLINE Mozilla is failing to recognize te closing of a comment tag NEWLINE under certain circunstances. For example, the following NEWLINE code: <br> NEWLINE <table><br> NEWLINE <tr><td>Comment 1 <!-- X --> NEWLINE ok</td></tr><br> NEWLINE <tr><td>Comment 2 <!--X--> NEWLINE ok</td></tr><br> NEWLINE <tr><td>Comment 3 <!-- X--> NEWLINE ok</td></tr><br> NEWLINE <tr><td>Comment 4 <!--X --> NEWLINE ok</td></tr><br> NEWLINE <tr><td>Comment 5 <!-- X ---> NEWLINE ok</td></tr><br> NEWLINE <tr><td>Comment NEWLINE 6<!--X-->ok</td></tr><br> NEWLINE </table><br> NEWLINE NEWLINE Show up as:<br> NEWLINE NEWLINE Comment 1<br> NEWLINE Comment 2<br> NEWLINE Comment 3<br> NEWLINE Comment 4<br> NEWLINE Comment 5<br> NEWLINE Comment 6ok<br>
230|HTML comments are not closed in the right place (resub 229)|Created by Wilson Afonso (wilson@zaz.com.br) on Wednesday, April 15, 1998 6:28:24 AM PDT NEWLINE Additional Details : NEWLINE I assumed this field was going to be displayed as NEWLINE HTML... so, resubmitting bug 229 in order to make it NEWLINE readable. NEWLINE NEWLINE Mozilla is failing to recognize the closing of a comment tag NEWLINE under certain circunstances. For example, the following NEWLINE code: NEWLINE <table> NEWLINE <tr><td>Comment 1 <!-- X --> ok</td></tr> NEWLINE <tr><td>Comment 2 <!--X--> ok</td></tr> NEWLINE <tr><td>Comment 3 <!-- X--> ok</td></tr> NEWLINE <tr><td>Comment 4 <!--X --> ok</td></tr> NEWLINE <tr><td>Comment 5 <!-- X ---> ok</td></tr> NEWLINE <tr><td>Comment 6<!--X-->ok</td></tr> NEWLINE </table> NEWLINE NEWLINE Shows up as: NEWLINE NEWLINE Comment 1 NEWLINE Comment 2 NEWLINE Comment 3 NEWLINE Comment 4 NEWLINE Comment 5 NEWLINE Comment 6ok
231|Compiler Error|Created by David Nebinger (dnebinger@synertech.highmark.com) on Wednesday, April 15, 1998 6:40:44 AM PDT NEWLINE Additional Details : NEWLINE Lines 2314 and 2315 of ns/lib/libmocha/et_moz.c generate NEWLINE errors due to the cast included in the macro call to NEWLINE PR_FREEIF. NEWLINE NEWLINE Instead of using: NEWLINE PR_FREEIF((char*) e->wysiwyg_url); NEWLINE PR_FREEIF((char*) e->base_href); NEWLINE NEWLINE Try using: NEWLINE PR_FREEIF(e->wysiwyg_url); NEWLINE PR_FREEIF(e->base_href);
232|Compiler Error|Created by David Nebinger (dnebinger@synertech.highmark.com) on Wednesday, April 15, 1998 7:43:33 AM PDT NEWLINE Additional Details : NEWLINE HP's compiler craps out on line 844 of file NEWLINE ns/cmd/xfe/icons/mkicons.c, and issues the error: NEWLINE cc: line 844: internal error 5172: Backend Assert ** Cause NEWLINE of inconsistency unknown. (5172) NEWLINE NEWLINE Looking at line 844, there are a series of trigraphs linked NEWLINE together that I think is causing the problem. By taking a NEWLINE less sophisticated approach (code follows below), the NEWLINE compile completed without complaint. NEWLINE NEWLINE Code replacement follows: NEWLINE /* Create a stream and decode the image. */ NEWLINE #ifndef HPUX NEWLINE stream = IL_NewStream (FO_PRESENT, NEWLINE (strstr (file, ".gif") ? (void *) NEWLINE IL_GIF : NEWLINE strstr (file, ".jpg") ? (void *) NEWLINE IL_JPEG : NEWLINE strstr (file, ".jpeg") ? (void *) NEWLINE IL_JPEG : NEWLINE strstr (file, ".xbm") ? (void *) NEWLINE IL_XBM : NEWLINE (void *) IL_GIF), NEWLINE url, &cx); NEWLINE #else NEWLINE { NEWLINE void* il_type_tmp=(void*)IL_GIF; NEWLINE NEWLINE if ((strstr(file,".jpg"))\|\|(strstr(file,".jpeg"))) NEWLINE il_type_tmp=(void*)IL_JPEG; NEWLINE else if (strstr(file,".xbm")) il_type_tmp=(void*)IL_XBM; NEWLINE NEWLINE stream=IL_NewStream(FO_PRESENT,il_type_tmp,url,&cx); NEWLINE } NEWLINE #endif
233|cannot go through proxy server|Created by Sid Liu (sidliu@yahoo.com) on Wednesday, April 15, 1998 1:10:54 PM PDT NEWLINE Additional Details : NEWLINE cannot go through proxy server on NT server 4.0. NEWLINE It will stop at Contacting ... NEWLINE Then shows a TCP/IP error. NEWLINE Updated by (blythe@netscape.com) on Wednesday, April 15, 1998 4:44:51 PM PDT NEWLINE Additional Details : NEWLINE Netlib level bug when dealing with proxy's. NEWLINE Handing over to owner of netlib. NEWLINE Updated by (gagan@netscape.com) on Friday, May 1, 1998 6:17:38 PM PDT NEWLINE Additional Details : NEWLINE Need more details. Have contacted Sid Liu.
234|Scheduler debug emits warnings|Created by (aoki@netscape.com) on Wednesday, April 15, 1998 2:30:27 PM PDT NEWLINE Additional Details : NEWLINE Originally reported by: TenThumbs <tenthumbs@cybernex.net> NEWLINE NEWLINE modules/schedulr/src/schedulr.c generates these messages NEWLINE NEWLINE schedulr.c:869: warning: implicit declaration of function NEWLINE `sprintf' NEWLINE schedulr.c:877: warning: implicit declaration of function NEWLINE `strcat' NEWLINE schedulr.c:892: warning: implicit declaration of function NEWLINE `strncpy'
235|PRNetAddr doesn't match sockaddr_in|Created by Bert Driehuis (driehuis@playbeing.org) on Wednesday, April 15, 1998 4:01:59 PM PDT NEWLINE Additional Details : NEWLINE Building Mozilla 1998-04-08 fails on BSD/OS 3.0. On NEWLINE investigation, the definition of PRNetAddr (which is cast NEWLINE into a sockaddr_in in ns/nsprpub/src/md/unix/unix.c in the NEWLINE MD_connect routine) turns out to use a short in host order NEWLINE for the sin_family field. BSD/OS 3.0 (and I'd imagine all NEWLINE 4.4lite2 derived unixes, but I haven't checked) define this NEWLINE field as NEWLINE u_char sa_len; NEWLINE u_char sa_family; NEWLINE which happens to be byte-swapped w.r.t. the assumptions in NEWLINE ns/nsprpub/include/prio.h, where these two bytes are NEWLINE addressed as one PRUint16. NEWLINE NEWLINE This needs to be addressed, probably in NEWLINE ns/nsprpub/src/md/unix/unix.c, by filling in a sockaddr_in NEWLINE from the PRNetAddr structure, before performing the connect. NEWLINE NEWLINE I haven't provided a fix, since I'm not positive about the NEWLINE proper way of fixing it. I'll see if I can come up with a NEWLINE solution that works with BSD/OS 3.0 without obviously NEWLINE breaking ports on which casting to sockaddr_in works. NEWLINE Updated by Bert Driehuis (driehuis@playbeing.org) on Wednesday, April 15, 1998 4:35:06 PM PDT NEWLINE Additional Details : NEWLINE This patch gets the BSDI 3.0 version working: NEWLINE NEWLINE *** nsprpub/pr/src/md/unix/unix.c.dist Thu Apr 9 02:52:30 1998 NEWLINE --- nsprpub/pr/src/md/unix/unix.c Thu Apr 16 01:21:47 1998 NEWLINE *************** NEWLINE *** 1062,1065 **** NEWLINE --- 1062,1072 ---- NEWLINE PRThread *me = _PR_MD_CURRENT_THREAD(); NEWLINE PRInt32 osfd = fd->secret->md.osfd; NEWLINE + #ifdef __bsdi__ NEWLINE + struct sockaddr the_addr; NEWLINE + the_addr = *((struct sockaddr *)addr); NEWLINE + the_addr.sa_len = sizeof(the_addr); NEWLINE + the_addr.sa_family = addr->inet.family; NEWLINE + addr = (PRNetAddr *) &the_addr; NEWLINE + #endif NEWLINE #ifdef IRIX NEWLINE extern PRInt32 _MD_irix_connect( NEWLINE *************** NEWLINE *** 2317,2321 **** NEWLINE osflags \|= O_TRUNC; NEWLINE if (flags & PR_SYNC) { NEWLINE ! #if defined(FREEBSD) NEWLINE osflags \|= O_FSYNC; NEWLINE #else NEWLINE --- 2324,2328 ---- NEWLINE osflags \|= O_TRUNC; NEWLINE if (flags & PR_SYNC) { NEWLINE ! #if defined(FREEBSD) \|\| defined(__bsdi__) NEWLINE osflags \|= O_FSYNC; NEWLINE #else NEWLINE NEWLINE (the FreeBSD/BSDI fix could probably have been avoided by conditionalizing on BS NEWLINE D44LITE2 or something like that, that would help NetBSD as well, I guess). NEWLINE NEWLINE Are there any systems that Mozilla runs on that don't define sa_len? NEWLINE NEWLINE I could unfortunately not update my report using Mozilla 1998-04-08 because of NEWLINE apparent JavaScript lossage ("try again: you didn't specify a short NEWLINE description). Fortunately, 4.04 works just fine :-) NEWLINE Updated by Wan-Teh Chang (wtc@netscape.com) on Wednesday, May 6, 1998 8:02:09 PM PDT NEWLINE Additional Details : NEWLINE Assigned bug to myself. NEWLINE NEWLINE The O_FSYNC bug has been fixed. NEWLINE NEWLINE I will look at the sockaddr_in.sa_len bug. NEWLINE Updated by Wan-Teh Chang (wtc@netscape.com) on Thursday, May 7, 1998 8:56:24 AM PDT NEWLINE Additional Details : NEWLINE I just fixed the sockaddr_in.sa_len problem in NEWLINE mozilla/nsprpub/pr/include/md/_bsdi.h, revision NEWLINE 3.2, in conjunction with a prior fix to NEWLINE mozilla/nsprpub/pr/src/md/unix.c. NEWLINE NEWLINE I define the macro _PR_HAVE_SOCKADDR_LEN in _bsdi.h. NEWLINE (This macro is also defined for _aix.h, _freebsd.h, NEWLINE and _rhapsody.h, indicating their BSD4.4 heritage.) NEWLINE The code in unix.c checks the _PR_HAVE_SOCKADDR_LEN NEWLINE macro and does the appropriate things. NEWLINE NEWLINE I ran a few socket tests (in mozilla/nsprpub/pr/tests) NEWLINE on our BSD/OS 2.1 and 3.0 machines and they all pased. NEWLINE NEWLINE Closed the bug. NEWLINE NEWLINE PS: To answer your question, yes, there are platforms NEWLINE that do not have a length field in their sockaddr, e.g., NEWLINE Solaris.
236|Local directory listing layouts messed up|Created by Dan McGuirk (mcguirk@indirect.com) on Wednesday, April 15, 1998 11:51:18 PM PDT NEWLINE Additional Details : NEWLINE The layout for local directory listings is fairly NEWLINE broken--different lines are overwriting each other, etc.
237|Resizing www.pricewatch.com breaks frames|Created by Dan McGuirk (mcguirk@indirect.com) on Wednesday, April 15, 1998 11:54:21 PM PDT NEWLINE Additional Details : NEWLINE Go to www.pricewatch.com, then resize the window. The NEWLINE frames show up in the wrong places.
238|Unix Aurora: must click directly on text|Created by Dan McGuirk (mcguirk@indirect.com) on Thursday, April 16, 1998 12:01:03 AM PDT NEWLINE Additional Details : NEWLINE Some might not consider this a bug, but in the Unix NEWLINE NavCenter left-hand pane you have to click directly on the NEWLINE text for a line to be selected. I think you should be able NEWLINE to click anywhere on the line. NEWLINE Updated by Steve Lamm (slamm@netscape.com) on Friday, May 8, 1998 5:34:11 PM PDT NEWLINE Additional Details : NEWLINE I thought that was a problem too. This should not be an issue when we switch to NEWLINE icons. However, I will pass this along to ramiro since he is working on that NEWLINE widget.
239|Must click twice in Unix Aurora|Created by Dan McGuirk (mcguirk@indirect.com) on Thursday, April 16, 1998 12:02:59 AM PDT NEWLINE Additional Details : NEWLINE In Unix Aurora, for resources that require downloading an NEWLINE .rdf file, the first time you click on the name the .rdf NEWLINE file is downloaded but the tree doesn't show up on the NEWLINE right-hand side. You have to click again before it does. NEWLINE Updated by Steve Lamm (slamm@netscape.com) on Friday, May 8, 1998 5:36:06 PM PDT NEWLINE Additional Details : NEWLINE This one is mine.
240|XP_GetBuiltinString error messages missing|Created by Dan McGuirk (mcguirk@indirect.com) on Thursday, April 16, 1998 12:09:50 AM PDT NEWLINE Additional Details : NEWLINE Some error messages from XP_GetBuiltinString that come up NEWLINE when I think either 'host unreachable' or 'network NEWLINE unreachable' or maybe just 'no response' messages are NEWLINE supposed to be displayed are not there. Specifically I've NEWLINE seen 1006 and 1020 reported as missing.
241|Bookmarks don't update in Aurora|Created by Dan McGuirk (mcguirk@indirect.com) on Thursday, April 16, 1998 12:28:16 AM PDT NEWLINE Additional Details : NEWLINE The display of my bookmarks in Aurora never updates unless I NEWLINE delete the .db files in the NavCen directory. NEWLINE Updated by Steve Lamm (slamm@netscape.com) on Friday, May 8, 1998 5:31:25 PM PDT NEWLINE Additional Details : NEWLINE I need to fix this. It happens because Aurora is only set up to read in your NEWLINE bookmarks at the beginning. From there on out, you should edit your bookmarks NEWLINE in Aurora and the old interface will go away. We are not there yet, but I have NEWLINE changes in the works to make it happen.
242|Unix Aurora "search" and "site tools" don't do anything|Created by Dan McGuirk (mcguirk@indirect.com) on Thursday, April 16, 1998 12:33:50 AM PDT NEWLINE Additional Details : NEWLINE "Search" and "Site Tools" under Unix Aurora don't seem to do NEWLINE anything.
243|More local directories problems: links and scrollbars|Created by Dan McGuirk (mcguirk@indirect.com) on Thursday, April 16, 1998 12:37:34 AM PDT NEWLINE Additional Details : NEWLINE When viewing a local directory whose file listing goes NEWLINE beyond the bottom of the window, the scrollbar doesn't show NEWLINE up. If you resize the window, the scrollbar shows up, but NEWLINE afterward none of the links are clickable.
244|Busted -MDupdate|Created by Keith Adams (kma@cs.brown.edu) on Thursday, April 16, 1998 2:14:02 AM PDT NEWLINE Additional Details : NEWLINE gcc 2.7.2.3 doesn't support -MDupdate apparently; is this NEWLINE a sun thing? In any event, the following diff worked... NEWLINE NEWLINE 113,114d112 NEWLINE < NO_MDUPDATE = 1 NEWLINE < NEWLINE 117a116 NEWLINE > NO_MDUPDATE = 1 NEWLINE 138c137 NEWLINE < MDUPDATE_FLAGS = -MD $(DEPENDENCIES) NEWLINE --- NEWLINE > MDUPDATE_FLAGS = -MDupdate $(DEPENDENCIES) NEWLINE Updated by (briano@netscape.com) on Monday, July 6, 1998 7:55:57 PM PDT NEWLINE Additional Details : NEWLINE NO_MDUPDATE is one of the environment variables you need NEWLINE to set before building on all platforms except IRIX (using NEWLINE the native compiler). This is described in the Unix build NEWLINE README.
245|Assertion failure at nlcstore.c: 144|Created by Keith Adams (kma@cs.brown.edu) on Thursday, April 16, 1998 2:17:50 AM PDT NEWLINE Additional Details : NEWLINE DBM_OpenDBMStore doesn't account for the possibility that NEWLINE CallDBOpenUsingFileURL fails because the database is locked NEWLINE (e.g., there was already a .netscape/lock file when we NEWLINE started up). Nothing tricky to this patch to NEWLINE ns/modules/rdf/src/nlcstore.c, and it seems to work :) NEWLINE NEWLINE 113,114d112 NEWLINE < NO_MDUPDATE = 1 NEWLINE < NEWLINE 117a116 NEWLINE > NO_MDUPDATE = 1 NEWLINE 138c137 NEWLINE < MDUPDATE_FLAGS = -MD $(DEPENDENCIES) NEWLINE --- NEWLINE > MDUPDATE_FLAGS = -MDupdate $(DEPENDENCIES) NEWLINE NEWLINE Updated by Keith Adams (kma@cs.brown.edu) on Thursday, April 16, 1998 2:21:13 AM PDT NEWLINE Additional Details : NEWLINE DBM_OpenDBMStore doesn't account for the possibility that NEWLINE CallDBOpenUsingFileURL fails because the database is locked NEWLINE (e.g., there was already a .netscape/lock file when we NEWLINE started up). Nothing tricky to this patch to NEWLINE ns/modules/rdf/src/nlcstore.c, and it seems to work :) NEWLINE NEWLINE 144,148c144 NEWLINE < /* if the lock file is present, the store can't be opened. we should just NEWLINE < * return gracefully ... */ NEWLINE < if (! store->nameDB) NEWLINE < return; NEWLINE < NEWLINE --- NEWLINE > CHECK_VAR1(store->nameDB); NEWLINE Updated by Keith Adams (kma@cs.brown.edu) on Thursday, April 16, 1998 2:22:20 AM PDT NEWLINE Additional Details : NEWLINE Sorry, I'm a big loser. The point of the last update NEWLINE was that I pasted in the wrong patch :( NEWLINE NEWLINE So, yeah. Take the second patch. NEWLINE Updated by Steve Lamm (slamm@netscape.com) on Friday, May 8, 1998 5:39:51 PM PDT NEWLINE Additional Details : NEWLINE Robert, can you take a look?
246|bookmarks not correctly saved in frames|Created by Richard Fellner (r.fellner@xpoint.at) on Thursday, April 16, 1998 3:58:33 AM PDT NEWLINE Additional Details : NEWLINE within frames, if you right-click a link NEWLINE and select "Add bookmark", the main URL, NEWLINE and not the URL of the selected link will be saved.
247|back button clears hidden fields|Created by lenny karpel (lenny@opentext.com) on Thursday, April 16, 1998 9:57:02 AM PDT NEWLINE Additional Details : NEWLINE back button clears hidden fields .. why NEWLINE Updated by lenny karpel (lenny@opentext.com) on Thursday, April 16, 1998 9:58:26 AM PDT NEWLINE Additional Details : NEWLINE hidden fields are reset when you go 'back' to a screen .. all others are not .. NEWLINE why are hidden fields reset ??
248|Proxy server access does not work|Created by Dan Kirkpatrick (dan.kirkpatrick@int.com) on Thursday, April 16, 1998 10:09:47 AM PDT NEWLINE Additional Details : NEWLINE I set up the proxy server preferences identical to those in NEWLINE V4.0. Netscape cannot see beyond the firewall (I can see NEWLINE internal web pages fine, as are behind the firewall). NEWLINE Updated by (aoki@netscape.com) on Monday, August 17, 1998 9:25:30 AM PDT NEWLINE Additional Details : NEWLINE Bug 268 has been marked a duplicate of this bug.
249|incorrect include path for endian.h|Created by Armen Nakashian (armen@unx.dec.com) on Thursday, April 16, 1998 12:41:28 PM PDT NEWLINE Additional Details : NEWLINE On DEC Unix 4.0D (and I'm guessing all other 4.0 releases), NEWLINE endian.h lives in /usr/include/alpha/. The include-path NEWLINE doesn't seem to include the alpha/ dir, and several files NEWLINE break when compiling. NEWLINE Updated by Wan-Teh Chang (wtc@netscape.com) on Wednesday, May 6, 1998 8:05:45 PM PDT NEWLINE Additional Details : NEWLINE That #include should probably be written as: NEWLINE #ifdef OSF1 NEWLINE #include <machine/endian.h> NEWLINE #endif NEWLINE NEWLINE That is what we do in mozilla/nsprpub/pr/include/prinet.h. NEWLINE Updated by Brian Ostrom (briano@netscape.com) on Monday, July 13, 1998 6:01:29 PM PDT NEWLINE Additional Details : NEWLINE I have a potential fix for this problem. Just a simple NEWLINE change to mozilla/config/OSF1.mk: NEWLINE NEWLINE OS_INCLUDES = -I/usr/include -I/usr/include/sys -I/usr/include/machine NEWLINE NEWLINE I haven't tested this yet, and I'm certainly open to any NEWLINE alternatives. (I threw in the first two simply to be safe, NEWLINE and because I'm paranoid.... ;-) NEWLINE Updated by Brian Ostrom (briano@netscape.com) on Tuesday, July 14, 1998 4:51:58 PM PDT NEWLINE Additional Details : NEWLINE The fix is in place (mozilla/config/OSF1.mk:3.4).
250|Test bug -- please ignore.|Created by Terry Weissman (terry@netscape.com) on Thursday, April 16, 1998 7:37:42 PM PDT NEWLINE Additional Details : NEWLINE I'm sorry; I just be testing the bug system here...
251|UnixWare: Mouse DisarmLink() appears not to be called (UnixWare 2.1)|Created by Björn Smith (smith@compound.se) on Friday, April 17, 1998 1:15:00 AM PDT NEWLINE Additional Details : NEWLINE When clicking on a link, the following message is written to NEWLINE stdout: NEWLINE mozilla-export: ArmLink() invoked twice without intervening NEWLINE DisarmLink()? NEWLINE NEWLINE Another problem is that some frame URL's leave one of the NEWLINE frame halves empty. NEWLINE NEWLINE It is built on UnixWare 2.1 (actually 2.1.2).
252|failure to get https site/page|Created by syd low (syd@who.net) on Friday, April 17, 1998 4:56:48 AM PDT NEWLINE Additional Details : NEWLINE Page/site causing NEWLINE problem NEWLINE NEWLINE https://www.qantas.com.au/fflyer_fr.html
253|Messenger Crashes if Imap-Index Files are not correct.|Created by (nsa@simmcomm.ch) on Friday, April 17, 1998 9:45:04 AM PDT NEWLINE Additional Details : NEWLINE If you change user-name in preferences and re-start the NEWLINE messenger, it will "look for mailboxes". Since the index NEWLINE files in the ns_imap directory are from the user before (not NEWLINE comparing to the ones that should be there), messenger NEWLINE exits. NEWLINE NEWLINE This problem also happens if index-files are killed, NEWLINE "remember password option" is activated and you are starting NEWLINE the messenger. Messenger thinks that index files are there, NEWLINE connects the imap server with password in preferences.js, NEWLINE but crashes. If you disable "remember password", messenger NEWLINE will ask for pw, after correct entered pw, messenger NEWLINE re-creates missing index files. Messenger should not crash NEWLINE in any of the above situations. NEWLINE Updated by (nsa@simmcomm.ch) on Friday, April 17, 1998 9:50:39 AM PDT NEWLINE Additional Details : NEWLINE I have tested this with: NEWLINE NEWLINE Linux 2.x, X-Windows and Windows 95.
254|MOZ_LITE configuration will not build|Created by (tor@cs.brown.edu) on Friday, April 17, 1998 12:40:41 PM PDT NEWLINE Additional Details : NEWLINE This patch fixes the MOZ_LITE configuration in config.mk and NEWLINE lib/layout/Makefile. It also changes references to "EDITOR" NEWLINE in the build process to "MOZ_EDITOR" to avoid problems on NEWLINE unix where EDITOR is commonly set in a user's environment. NEWLINE NEWLINE The patch is a little large to include here; grab it from: NEWLINE http://www.cs.brown.edu/people/tor/mozilla/lite.patch NEWLINE Updated by (tor@cs.brown.edu) on Wednesday, April 29, 1998 11:14:27 PM PDT NEWLINE Additional Details : NEWLINE Patch updated for 19980429 release. Same place and name.
255|Voicemail with Beatnik plugin, cannot play voicemail message.|Created by wendy price (chloezzz@total.net) on Saturday, April 18, 1998 12:20:51 PM PDT NEWLINE Additional Details : NEWLINE After downloading & installing the plugin Beatnik (one of NEWLINE the options (under plugins in NS),I'm not able to play back NEWLINE any voice mail. I can record, and playback what I've NEWLINE recorded, but when receiving voicemail, Beatnik becomes the NEWLINE default option once installed. NEWLINE Mentioned under the voicemail message is the usual message NEWLINE :voicemail requires NS-communicator PR3 or higher for proper NEWLINE operation. NEWLINE I have Communicator 4.05. I had 4.04 and did have all the NEWLINE componnents for voicemail & conference.
256|FIX: Extra periods in filname changed to underscores|Created by Brian Wiles (brian@inforum.net) on Sunday, April 19, 1998 3:06:03 PM PDT NEWLINE Additional Details : NEWLINE I wrote my own fix to not change extra periods to NEWLINE underscores when downloading a file. Feel free to NEWLINE incorporate this into your version of Mozilla. Forgive me NEWLINE if this has been addressed previously. I did not see it in NEWLINE the Bugzilla query. NEWLINE NEWLINE This is a context diff in function fe_URLtoLocalName in NEWLINE ns\\cmd\\winfe\\fenet.cpp: NEWLINE NEWLINE *** c:\\temp\\fenet.cpp Mon Mar 30 17:57:42 1998 NEWLINE --- fenet.cpp Sun Apr 19 14:45:30 1998 NEWLINE *************** NEWLINE *** 268,277 **** NEWLINE #endif NEWLINE char *name = csName.GetBuffer(0); NEWLINE NEWLINE ! // replace extra periods in 8 character name with NEWLINE underscores NEWLINE for (idx =0 ; idx < csName.GetLength(); idx++) { NEWLINE if ((name[idx] == '.')\|\|(name[idx] == ':')) NEWLINE name[idx] = '_'; NEWLINE } NEWLINE csName.ReleaseBuffer(-1); NEWLINE NEWLINE *csURL = csName + csExt; NEWLINE --- 268,285 ---- NEWLINE #endif NEWLINE char *name = csName.GetBuffer(0); NEWLINE NEWLINE ! #ifdef XP_WIN16 NEWLINE ! // replace extra periods and colons in 8 character NEWLINE name with underscores NEWLINE NEWLINE for (idx =0 ; idx < csName.GetLength(); idx++) { NEWLINE if ((name[idx] == '.')\|\|(name[idx] == ':')) NEWLINE name[idx] = '_'; NEWLINE } NEWLINE + #else NEWLINE + // replace extra colons in 8 character name with NEWLINE underscores NEWLINE + // This fixes the filename mangling problem in Win32 NEWLINE since it can handle multiple periods. -BCW 04/01/1998 NEWLINE + for (idx =0 ; idx < csName.GetLength(); idx++) { NEWLINE + if (name[idx] == ':') name[idx] = '_'; NEWLINE + } NEWLINE + #endif NEWLINE csName.ReleaseBuffer(-1); NEWLINE NEWLINE *csURL = csName + csExt; NEWLINE Updated by Garrett Arch Blythe (blythe@netscape.com) on Thursday, June 11, 1998 4:24:09 PM PDT NEWLINE Additional Details : NEWLINE Checked a modified fix into Mozilla tree. Thanks Brian!
257|FIX: Extra periods in filname changed to underscores|Created by Brian Wiles (brian@inforum.net) on Sunday, April 19, 1998 3:07:03 PM PDT NEWLINE Additional Details : NEWLINE I wrote my own fix to not change extra periods to NEWLINE underscores when downloading a file. Feel free to NEWLINE incorporate this into your version of Mozilla. Forgive me NEWLINE if this has been addressed previously. I did not see it in NEWLINE the Bugzilla query. NEWLINE NEWLINE This is a context diff in function fe_URLtoLocalName in NEWLINE ns\\cmd\\winfe\\fenet.cpp: NEWLINE NEWLINE *** c:\\temp\\fenet.cpp Mon Mar 30 17:57:42 1998 NEWLINE --- fenet.cpp Sun Apr 19 14:45:30 1998 NEWLINE *************** NEWLINE *** 268,277 **** NEWLINE #endif NEWLINE char *name = csName.GetBuffer(0); NEWLINE NEWLINE ! // replace extra periods in 8 character name with NEWLINE underscores NEWLINE for (idx =0 ; idx < csName.GetLength(); idx++) { NEWLINE if ((name[idx] == '.')\|\|(name[idx] == ':')) NEWLINE name[idx] = '_'; NEWLINE } NEWLINE csName.ReleaseBuffer(-1); NEWLINE NEWLINE *csURL = csName + csExt; NEWLINE --- 268,285 ---- NEWLINE #endif NEWLINE char *name = csName.GetBuffer(0); NEWLINE NEWLINE ! #ifdef XP_WIN16 NEWLINE ! // replace extra periods and colons in 8 character NEWLINE name with underscores NEWLINE NEWLINE for (idx =0 ; idx < csName.GetLength(); idx++) { NEWLINE if ((name[idx] == '.')\|\|(name[idx] == ':')) NEWLINE name[idx] = '_'; NEWLINE } NEWLINE + #else NEWLINE + // replace extra colons in 8 character name with NEWLINE underscores NEWLINE + // This fixes the filename mangling problem in Win32 NEWLINE since it can handle multiple periods. -BCW 04/01/1998 NEWLINE + for (idx =0 ; idx < csName.GetLength(); idx++) { NEWLINE + if (name[idx] == ':') name[idx] = '_'; NEWLINE + } NEWLINE + #endif NEWLINE csName.ReleaseBuffer(-1); NEWLINE NEWLINE *csURL = csName + csExt; NEWLINE Updated by Brian Wiles (brian@inforum.net) on Sunday, April 19, 1998 3:12:46 PM PDT NEWLINE Additional Details : NEWLINE This is a duplicate of Bug # 256. I told Communicator to save the frame, and it NEWLINE reloaded it from the server, thus causing a repost of the same bug. -BCW NEWLINE Updated by (blythe@netscape.com) on Tuesday, June 9, 1998 4:11:53 PM PDT NEWLINE Additional Details : NEWLINE Assuming this is a duplicate of bug 256
258|CSS spec. for XML is never reloaded|Created by Jan Kroken (jankr@nntp.ifi.uio.no) on Monday, April 20, 1998 6:52:10 AM PDT NEWLINE Additional Details : NEWLINE An CSS file specified in an XML document like NEWLINE <?xml:style type="text/css2" src="test.css"?> NEWLINE is never reloaded.
259|Edit/Preferences Dialog fails|Created by Ralf Hildebrandt (R.Hildebrandt@tu-bs.de) on Monday, April 20, 1998 11:45:54 AM PDT NEWLINE Additional Details : NEWLINE Warning: NEWLINE Name: historyBox NEWLINE Class: XmForm NEWLINE Bailed out of edge synchronization after 10,000 iterations. NEWLINE Check for contradictory constraints on the children of this NEWLINE form. NEWLINE NEWLINE Warning: homePageBox NEWLINE Class: XmForm NEWLINE Bailed out of edge synchronization after 10,000 iterations. NEWLINE Check for contradictory constraints on the children of this NEWLINE form. NEWLINE Updated by Ralf Hildebrandt (R.Hildebrandt@tu-bs.de) on Wednesday, May 6, 1998 8:38:19 AM PDT NEWLINE Additional Details : NEWLINE fixed in 19980429 release
260|HTML pages or images don't get diplayed|Created by Ralf Hildebrandt (R.Hildebrandt@tu-bs.de) on Monday, April 20, 1998 11:52:30 AM PDT NEWLINE Additional Details : NEWLINE HTML pages or images don't get diplayed: NEWLINE NEWLINE Reason: NEWLINE The webserver returns: NEWLINE NEWLINE Forbidden NEWLINE You don't have permission to access "URL" NEWLINE NEWLINE A "reload" or "shift Reload" fix the problem only partially, NEWLINE some elements just don't get displayed. NEWLINE Error in HTTP code ?
261|testing " <table> testing|Created by Arun Ramachandran (arunr@netscape.com) on Monday, April 20, 1998 12:45:10 PM PDT NEWLINE Additional Details : NEWLINE Testing NEWLINE NEWLINE <table> NEWLINE <tr> NEWLINE <td>aasdfasdf NEWLINE </tr> NEWLINE </table> NEWLINE Updated by Arun Ramachandran (arunr@netscape.com) on Thursday, April 23, 1998 3:34:35 PM PDT NEWLINE Additional Details : NEWLINE testing again
262|testing case - 1|Created by Arun Ramachandran (T1) (arunr@test.com) on Monday, April 20, 1998 1:07:31 PM PDT NEWLINE Additional Details : NEWLINE testing case - 1 NEWLINE Updated by Arun Ramachandran (T1) (arunr@test.com) on Monday, April 20, 1998 1:07:55 PM PDT NEWLINE Additional Details : NEWLINE testing again NEWLINE Updated by Arun Ramachandran (T1) (arunr@test.com) on Monday, April 20, 1998 1:10:59 PM PDT NEWLINE Additional Details : NEWLINE test NEWLINE Updated by Arun Ramachandran (T1) (arunr@test.com) on Monday, April 20, 1998 2:09:04 PM PDT NEWLINE Additional Details : NEWLINE testing again NEWLINE Updated by Arun Ramachandran (T1) (arunr@test.com) on Sunday, May 17, 1998 5:28:41 PM PDT NEWLINE Additional Details : NEWLINE testing
263|Mozilla does not compile on NetBSD/arm32|Created by Frank van der Linden (frank@wins.uva.nl) on Tuesday, April 21, 1998 4:39:48 PM PDT NEWLINE Additional Details : NEWLINE Mozilla does not compile on NetBSD for the ARM processor NEWLINE (NetBSD/arm32). The patches mentioned in bugreport #122 make NEWLINE it work on some, but not all architectures that NetBSD runs NEWLINE on. The following patch fixes this, and can be applied on NEWLINE a source tree that has the patches in report #122 already NEWLINE applied: NEWLINE ftp://ftp.netbsd.org/pub/NetBSD/misc/fvdl/mozilla/diff.002 NEWLINE Updated by Frank van der Linden (frank@wins.uva.nl) on Sunday, May 24, 1998 4:47:35 PM PDT NEWLINE Additional Details : NEWLINE The updated changes for the 1998-04-29 version can be found in NEWLINE ftp://ftp.netbsd.org/ftp/pub/NetBSD/misc/fvdl/mozilla/diff-19980429.001 NEWLINE NEWLINE This patch combines the patch in report #122 with the one for this report, NEWLINE and updates it for the 1998-04-29 version. NEWLINE NEWLINE There weren't many changes; just a few conflicts with changes made in the main NEWLINE source tree were resolved. NEWLINE Updated by Brian Ostrom (briano@netscape.com) on Monday, July 13, 1998 3:49:20 PM PDT NEWLINE Additional Details : NEWLINE Well, that worked well.... I changed the assigned-to field and added a comment, NEWLINE and submitted it..., and only the assigned-to entry got updated. Swell. NEWLINE NEWLINE Anyway, Wan-Teh, I reassigned this to you because most of the remaining NEWLINE changes are in NSPR. The most recent diff file is in: NEWLINE NEWLINE ftp://ftp.netbsd.org/pub/NetBSD/misc/fvdl/mozilla/ NEWLINE Updated by Wan-Teh Chang (wtc@netscape.com) on Monday, July 13, 1998 3:54:09 PM PDT NEWLINE Additional Details : NEWLINE OK. Accepted the bug. NEWLINE NEWLINE Chris Toshok also gave me some NetBSD files NEWLINE on June 30, 1998. NEWLINE NEWLINE Are we under a code freeze right now? NEWLINE Updated by Wan-Teh Chang (wtc@netscape.com) on Monday, July 13, 1998 5:50:00 PM PDT NEWLINE Additional Details : NEWLINE I merged into my private tree the two files Chris Toshok NEWLINE gave me and also diff-19980408.001, diff-19980408.002, NEWLINE diff-19980429.001, and diff-19980603.001 in NEWLINE ftp://ftp.netbsd.org/pub/NetBSD/misc/fvdl/mozilla/. NEWLINE NEWLINE I will check it in when I find out whether there is a NEWLINE code freeze. NEWLINE Updated by Wan-Teh Chang (wtc@netscape.com) on Friday, July 17, 1998 4:33:00 PM PDT NEWLINE Additional Details : NEWLINE I just checked in the NetBSD port of NSPR. NEWLINE NEWLINE Frank, after you verify that Mozilla builds on NEWLINE NetBSD, let us know so that we can close this NEWLINE bug. NEWLINE Updated by Frank van der Linden (frank@wins.uva.nl) on Tuesday, August 4, 1998 12:04:09 PM PDT NEWLINE Additional Details : NEWLINE The remainder of the patches (i.e. the ones that are left now that the nsprpub NEWLINE part has been integrated) for the 1998-07-28 release can be found in NEWLINE ftp://ftp.netbsd.org/pub/NetBSD/misc/fvdl/mozilla/diff-19980728.001 NEWLINE NEWLINE This diff does contain 2 additions to files that already were patched; one NEWLINE because the name of a jmpbuf index was changed in the arm32 specific files NEWLINE of NetBSD, and another that makes the compile always use -fPIC.
264|Fails to complie due to having wrong include|Created by Paul Kanz (pkanz@conix.com) on Wednesday, April 22, 1998 1:42:24 PM PDT NEWLINE Additional Details : NEWLINE Need to change xpcom/src/nsRepository.cpp:#include NEWLINE <iostream.h> to <sys/stream.h> to build under Solaris 2.5.1 NEWLINE Updated by (briano@netscape.com) on Monday, July 6, 1998 8:02:06 PM PDT NEWLINE Additional Details : NEWLINE I don't know if this is still a problem, but I'll let NEWLINE Will make that call.
265|Use Solaris 'as' not GNU 'as' (gas)|Created by Paul Kanz (pkanz@conix.com) on Wednesday, April 22, 1998 1:47:44 PM PDT NEWLINE Additional Details : NEWLINE Use '/usr/ccs/bin/as' not GNU 'as' - the '-P' option is not NEWLINE valid with gas. 'as -v': GNU assembler version 2.7 NEWLINE (sparc-sun-solaris2.5), using BFD version 2.7 NEWLINE NEWLINE This should be well known, but.... NEWLINE Updated by Wan-Teh Chang (wtc@netscape.com) on Wednesday, May 6, 1998 11:30:39 PM PDT NEWLINE Additional Details : NEWLINE Fixed as you suggested, in mozilla/nsprpub/pr/src/md/unix/Makefile, NEWLINE revision 3.7. NEWLINE NEWLINE Assigned the bug to myself. NEWLINE NEWLINE Closed the bug.
266|makeNewId is busted|Created by John Giannandrea (jg@netscape.com) on Wednesday, April 22, 1998 5:57:32 PM PDT NEWLINE Additional Details : NEWLINE makeNewId contains a bug in the LONG_LONG case which NEWLINE causes it to generate endless hash collisions. NEWLINE NEWLINE in fact the routine should be rewritten to NEWLINE a/ not have two implementations NEWLINE b/ not call PR_Now() or printf if at all possible. NEWLINE NEWLINE -jg
267|Page Composer auto-starts|Created by David Johnson (dlj0@lehigh.edu) on Wednesday, April 22, 1998 7:22:22 PM PDT NEWLINE Additional Details : NEWLINE When starting Mozilla, after the initial page loads (seen NEWLINE with a blank initial page, or with about:mozilla as the NEWLINE initial page), the Composer window pops up without being NEWLINE asked. NEWLINE NEWLINE It's no biggie, but it is a bit annoying. NEWLINE Updated by Ramiro Estrugo (ramiro@netscape.com) on Wednesday, April 29, 1998 2:40:47 AM PDT NEWLINE Additional Details : NEWLINE This problem has been fixed.
268|Navigator ignores proxy server/firewall settings|Created by Edward Newman (Edward_newman@ml.com) on Thursday, April 23, 1998 1:17:57 AM PDT NEWLINE Additional Details : NEWLINE Although a proxy server is defined in the manual proxy NEWLINE settings, it is ignored and it is not possible to connect to NEWLINE external web servers using Navigator V5.0. NEWLINE NEWLINE Is this facility enabled in this release? NEWLINE Updated by (aoki@netscape.com) on Monday, August 17, 1998 9:25:02 AM PDT NEWLINE Additional Details : NEWLINE Duplicate of bug 248
269|ns/module/rdf/ht.c ? 1755 line|Created by Kyung-chan Jin (jkc@strc.kyungpook.ac.kr) on Thursday, April 23, 1998 5:29:15 PM PDT NEWLINE Additional Details : NEWLINE (1) After starting, click personal bar with right button, NEWLINE the error message is displayed. NEWLINE NEWLINE (2) Sometimes, The html message was not aligned and NEWLINE the left and right message was overlapped and no linefeed.
270|Binhex archive unreadable|Created by (jah@alien.bt.co.uk) on Friday, April 24, 1998 4:07:55 AM PDT NEWLINE Additional Details : NEWLINE I don't believe it but... NEWLINE NEWLINE I cannot decode mac_19980408.sit.bin using Stuffit Expander. NEWLINE All the other archives are ok. Expander doesn't give an NEWLINE error, it simply exits without doing anything. NEWLINE NEWLINE I've tried both my local mirror and direct from mozilla.org NEWLINE on a number of machines with the same results. Memory is not NEWLINE the problem. NEWLINE Updated by (jah@alien.bt.co.uk) on Tuesday, May 5, 1998 2:40:41 AM PDT NEWLINE Additional Details : NEWLINE This appears to be fixed in mac_19980429.sit.bin
271|Proxy user can't fetch RDF sitemap from rdf.netscape.com|Created by (meltsner@hotmail.com) on Friday, April 24, 1998 8:06:15 AM PDT NEWLINE Additional Details : NEWLINE A minute or so after starting Mozilla, the system times out NEWLINE the retrieval of the sitemap (for NavCenter/Aurora, I NEWLINE believe) from rdf.netscape.com. The proxy setting is NEWLINE probably being ignored. Without looking at the code, I NEWLINE suspect it's fetching the info directly, rather than going NEWLINE through the standard NetLib method. NEWLINE NEWLINE Both the Manual Proxy setting and the autoproxy.pac files NEWLINE are ignored. NEWLINE NEWLINE There are other odd things happening around the proxy NEWLINE settings, but this is quite reproducible. NEWLINE Updated by (meltsner@hotmail.com) on Wednesday, May 6, 1998 6:26:44 AM PDT NEWLINE Additional Details : NEWLINE It still happens with 4-29 release. NEWLINE NEWLINE I suspect it's that the default NavCenter RDF file is fetched without looking at NEWLINE the proxy settings, or it could be the other problem I've noticed, which is that NEWLINE frequently the proxy settings don't seem to be read upon startup -- you have to NEWLINE change them back and forth to enable application proxy access.
272|Navigator toolbars are redrawn in the wrong place.|Created by (shaman@nlc.net.au) on Friday, April 24, 1998 4:03:48 PM PDT NEWLINE Additional Details : NEWLINE If the three collapsable toolbars are all closed by clicking NEWLINE on the little tabs on their left, then opened again, the NEWLINE first toolbar is redrawn in the wrong position, and NEWLINE overwrites a line of pixels above it. (And leaves a thick NEWLINE black line below it.) NEWLINE Updated by (blythe@netscape.com) on Tuesday, June 9, 1998 4:12:57 PM PDT NEWLINE Additional Details : NEWLINE David, you are the toolbar god
273|Submenu open delay wrong on Windows|Created by (shaman@nlc.net.au) on Friday, April 24, 1998 4:07:26 PM PDT NEWLINE Additional Details : NEWLINE There is a little known setting in the Windows95 registry NEWLINE which sets the time it takes a menu to open up after the NEWLINE mouse has moved over it's root node. It's really NEWLINE disconcerting when the normal menus are nice and snappy, but NEWLINE I have to sit there and wait for the submenus in the NEWLINE bookmark tree to open up. NEWLINE Updated by (blythe@netscape.com) on Tuesday, June 9, 1998 4:14:28 PM PDT NEWLINE Additional Details : NEWLINE Bill, this would be a nice OS integration enhancement. NEWLINE Updated by (blythe@netscape.com) on Tuesday, June 9, 1998 4:14:37 PM PDT NEWLINE Additional Details : NEWLINE Bill, this would be a nice OS integration enhancement.
274|Hardcoded string needs to be resourced|Created by Rogelio Andreo (rba@murcia.net) on Friday, April 24, 1998 10:59:58 PM PDT NEWLINE Additional Details : NEWLINE ns\\modules\\rdf\\src\\columns.c NEWLINE Function ColumnsGetSlotValue has a bunch of strings that NEWLINE should be moved out of the code (to xpstr or something like NEWLINE that).
275|Externalize "http://rdf.netscape.com/navcntr.rdf"?|Created by Rogelio Andreo (rba@murcia.net) on Friday, April 24, 1998 11:05:11 PM PDT NEWLINE Additional Details : NEWLINE ns\\modules\\rdf\\src\\rdfht.c NEWLINE Function RDF_Init calls NEWLINE PREF_SetDefaultCharPref("browser.NavCenter", NEWLINE "http://rdf.netscape.com/navcntr.rdf"); NEWLINE NEWLINE The URL should be moved out of the code (to xpstr I guess) - NEWLINE for instance for intl. versions. NEWLINE Updated by Rogelio Andreo (rba@murcia.net) on Friday, April 24, 1998 11:18:12 PM PDT NEWLINE Additional Details : NEWLINE Or the user pref default file should include the "browser.NavCenter" entry by NEWLINE default.
276|Hardcoded strings in navbar.cpp|Created by Rogelio Andreo (rba@murcia.net) on Friday, April 24, 1998 11:09:34 PM PDT NEWLINE Additional Details : NEWLINE ns\\cmd\\winfe\\navbar.cpp NEWLINE CNavMenuBar::OnCreate calls m_pMenuButton->Create using NEWLINE hardcoded strings that should be "resourced" ("No view NEWLINE selected.", "Click here to view ..." ... ) NEWLINE NEWLINE Updated by (blythe@netscape.com) on Tuesday, June 9, 1998 4:15:45 PM PDT NEWLINE Additional Details : NEWLINE David, move these to the resource file.
277|Hardcoded strings in nscmdwinfeavfram.cpp|Created by Rogelio Andreo (rba@murcia.net) on Friday, April 24, 1998 11:25:23 PM PDT NEWLINE Additional Details : NEWLINE File ns\\cmd\\winfe\\navfram.cpp NEWLINE NEWLINE CNSNavFrame::UpdateTitleBar and NEWLINE CNSNavFrame::CreateNewNavCenter contain strings that should NEWLINE be externalized from the code. NEWLINE Updated by (blythe@netscape.com) on Tuesday, June 9, 1998 4:16:22 PM PDT NEWLINE Additional Details : NEWLINE David, please externalize these strings.
278|.exp problem w. CWPro3 when building NSRuntime|Created by flip phillips (flip@tvl.psy.ohio-state.edu) on Monday, April 27, 1998 9:46:44 AM PDT NEWLINE Additional Details : NEWLINE The exported symbol names in the runtime library for NEWLINE NSRuntime.mcp are incorrect for CWPro3. Apparently the NEWLINE generated names differ between the two distributions of the NEWLINE MSL.
279|Garbage collector incorrectly called during script execution|Created by Andrew Hodgkinson (ahodgkin@acorn.co.uk) on Tuesday, April 28, 1998 3:40:12 AM PDT NEWLINE Additional Details : NEWLINE JS_MaybeGC() will, now and again, call the garbage NEWLINE collector. Unfortunately it does this through js_GC(), NEWLINE rather than JS_GC(), so nothing actually gets NEWLINE garbage collected. NEWLINE NEWLINE A discussion about this went on in mozilla.general. NEWLINE See article <353FBFE6.CA026B4C@netscape.com> and the NEWLINE subsequent reply for details of the performance impacts NEWLINE of the fix (or URL NEWLINE NEWLINE http://www.acorn.com/~ahodgkin/javascript/js_speed.html NEWLINE NEWLINE if you don't want to dig out the news articles). NEWLINE Updated by Mike McCabe (mccabe@netscape.com) on Tuesday, May 26, 1998 10:53:37 PM PDT NEWLINE Additional Details : NEWLINE Looks like this was fixed a while ago. I've confirmed that JS_MaybeGC in NEWLINE jsapi.c calls JS_GC() rather than js_GC(). NEWLINE NEWLINE Thanks for catching this.
280|Version Checking regarding uuid2.dll|Created by (yardley@uiuc.edu) on Tuesday, April 28, 1998 12:08:31 PM PDT NEWLINE Additional Details : NEWLINE The version (vd version) checking in the makefile.win under NEWLINE \\ns\\cmd\\winfe\\prefs\\nsdlg checks for version 1100. I have NEWLINE vc++ 5.0 professional with the latest service pack, and it NEWLINE has a different version, thus the linking is messed up NEWLINE because uuid[2,3].lib have been consoladated to uuid.lib in NEWLINE ver 5.0. NEWLINE Updated by (yardley@uiuc.edu) on Tuesday, April 28, 1998 12:12:24 PM PDT NEWLINE Additional Details : NEWLINE \\ns\\cmd\\winfe\\prefs\\nsprefui\\src as well, and I am still checking for more... I NEWLINE imagine there are more NEWLINE Updated by (yardley@uiuc.edu) on Tuesday, April 28, 1998 12:18:17 PM PDT NEWLINE Additional Details : NEWLINE also: (still building more - will update later) NEWLINE \\ns\\cmd\\winfe\\prefs\\brpref\\src NEWLINE \\ns\\cmd\\winfe\\prefs\\edpref\\src NEWLINE Updated by (yardley@uiuc.edu) on Tuesday, April 28, 1998 1:51:19 PM PDT NEWLINE Additional Details : NEWLINE last one: NEWLINE \\ns\\cmd\\winfe\\mkfiles32\\mozilla.mak NEWLINE NEWLINE specific line: NEWLINE !if "$(_MSC_VER)" != "1100" NEWLINE LLIBS32 = $(LLIBS32) uuid2.lib NEWLINE !endif NEWLINE NEWLINE maybe change that to: NEWLINE !if "$(_MSC_VER)" >= "1100" NEWLINE Updated by (yardley@uiuc.edu) on Tuesday, April 28, 1998 1:54:58 PM PDT NEWLINE Additional Details : NEWLINE my bad... make that: NEWLINE NEWLINE !if "$(_MSC_VER)" <= "1100" NEWLINE LLIBS32 = $(LLIBS32) uuid2.lib NEWLINE !endif NEWLINE Updated by Brian Ostrom (briano@netscape.com) on Monday, July 6, 1998 8:28:46 PM PDT NEWLINE Additional Details : NEWLINE Reassigning to Garrett. NEWLINE Updated by Garrett Arch Blythe (blythe@netscape.com) on Tuesday, July 7, 1998 2:58:23 PM PDT NEWLINE Additional Details : NEWLINE Sure, we can do that. NEWLINE Note that _MSC_VER is set manually by you in your environment. NEWLINE You can set it to 1100 if you need to just to avoid linking this. NEWLINE Send a diff to the patches group, and we can take it, otherwise just set your NEWLINE environment to 1100 and avoid the problem. NEWLINE NEWLINE garrett
281|Build Problem|Created by (yardley@uiuc.edu) on Tuesday, April 28, 1998 2:16:18 PM PDT NEWLINE Additional Details : NEWLINE NMAKE : fatal error U1073: don't know how to make NEWLINE '..\\..\\..\\dist\\WIN32_O.OBJ\\bin NEWLINE \\patcher.exe' NEWLINE Stop. NEWLINE NMAKE : fatal error U1077: '"C:\\Program NEWLINE Files\\DevStudio\\VC\\BIN\\NMAKE.EXE"' : ret NEWLINE urn code '0x2' NEWLINE Stop. NEWLINE NEWLINE umm... no clue as to what causes this one or how to fix it. NEWLINE Updated by (yardley@uiuc.edu) on Tuesday, April 28, 1998 2:24:14 PM PDT NEWLINE Additional Details : NEWLINE DIRS = xfe patcher NEWLINE dir listing of /ns/cmd/: NEWLINE 04/28/98 04:10p <DIR> . NEWLINE 04/28/98 04:10p <DIR> .. NEWLINE 04/28/98 12:49p <DIR> CVS NEWLINE 04/28/98 12:50p <DIR> macfe NEWLINE 04/08/98 05:55p 753 Makefile NEWLINE 04/08/98 05:46p 1,557 makefile.win NEWLINE 04/28/98 12:50p <DIR> stubfe NEWLINE 04/28/98 04:10p <DIR> wincom NEWLINE 04/28/98 04:10p <DIR> winfe NEWLINE 04/28/98 12:52p <DIR> xfe NEWLINE 10 File(s) 2,310 bytes NEWLINE NEWLINE As you can see... patcher does not exist. That a problem? NEWLINE Updated by Brian Ostrom (briano@netscape.com) on Monday, July 6, 1998 8:11:05 PM PDT NEWLINE Additional Details : NEWLINE Garrett, any idea? I thought non-existent directories in NEWLINE the DIRS macro list were simply ignored. NEWLINE Updated by Garrett Arch Blythe (blythe@netscape.com) on Tuesday, July 7, 1998 3:01:02 PM PDT NEWLINE Additional Details : NEWLINE This doesn't make sense. Have you gotten past this problem yet? Lots of people NEWLINE build without seeing this error. Reopen if still an issue for you.
282|XFE overrides color preferences after viewing frames|Created by (ksw@dna.lth.se) on Wednesday, April 29, 1998 7:07:31 AM PDT NEWLINE Additional Details : NEWLINE All X Windows versions have a bug in the front end in which NEWLINE the user preferences for colors are permanently overridden NEWLINE after having viewed certain frames. Upon displaying a frame NEWLINE the code attempts to set the _default_ colors, instead of NEWLINE those of the document. NEWLINE NEWLINE Patch available.
283|SINIX: Assertion Failure in uxwrap.c|Created by Wan-Teh Chang (wtc@netscape.com) on Wednesday, April 29, 1998 11:59:54 AM PDT NEWLINE Additional Details : NEWLINE This bug is reported by Sanjay Gupta <gupta@informix.com>. NEWLINE NEWLINE He gets an assertion failure in NEWLINE nsprpub/pr/src/md/unix/uxwrap.c at line 283 NEWLINE PR_ASSERT(nbits > 0); NEWLINE when he runs "mozilla-export -version". The NEWLINE pd->out_flags has the value 0x10 (POLLHUP), NEWLINE meaning that the file descriptor hung up. NEWLINE NEWLINE The stack trace at the assertion failure is: NEWLINE debug> stack NEWLINE Stack Trace for p2, Program (dns NEWLINE [9] 0x180362f4 _kill() NEWLINE [8] 0x1802a658 abort() NEWLINE [7] 0x00a77d40 PR_Assert(s="nbits > 0", file="uxwrap.c", NEWLINE ln=289) NEWLINE [prlog.c@415] NEWLINE [6] 0x00ab3fb8 select(width=1024, rd=0x7ffed860, wr=0x0, NEWLINE ex=0x0, NEWLINE tv=0x0) [uxwrap.c@289] NEWLINE [5] 0x006d00fc dns_driver_main_loop(in_fd=0, out_fd=1) NEWLINE [unix-dns.c@858] NEWLINE [4] 0x006cfbfc dns_driver(argc=1, argv=0x7ffef8a4, in_fd=0, NEWLINE out_fd=1) NEWLINE [unix-dns.c@720] NEWLINE [3] 0x006d064c DNS_SpawnProcess(argc=1, argv=0x7ffef8a4) NEWLINE [unix-dns.c@1021] NEWLINE [2] 0x004f2454 XFE_InitDNS_Early(argc=1, argv=0x7ffef8a4) NEWLINE [xfe-dns.c@58] NEWLINE [1] 0x00489870 main(argc=1, argv=0x7ffef8a4) NEWLINE [mozilla.c@2051] NEWLINE [0] 0x0040ffe4 _start() NEWLINE debug> NEWLINE NEWLINE If he runs without the -version option, it seems that NEWLINE the application forks and the process p1 gets a SIGALRM NEWLINE with the following stack trace: NEWLINE > debug> run NEWLINE SIGNALED 14 (alrm) in p1 [_sigprocmask] NEWLINE 0x18041b94(_sigprocmask+20:) beq NEWLINE a3,zero,0x18041ba8 NEWLINE 0x18041b98(_sigprocmask+24:) lw NEWLINE at,-32468(gp) NEWLINE debug> stack NEWLINE Stack Trace for p1, Program mozilla-export NEWLINE [4] 0x18041b94 _sigprocmask() NEWLINE [3] 0x00aab3f8 _MD_UnblockClockInterrupts() NEWLINE [unix.c@2274] NEWLINE [2] 0x00a87c54 PR_UnblockClockInterrupts() NEWLINE [prinit.c@184] NEWLINE [1] 0x0048a7d8 main(argc=1, argv=0x7ffef89c) NEWLINE [mozilla.c@2430] NEWLINE [0] 0x0040ffe4 _start() NEWLINE NEWLINE If he makes p2 runnable, it hangs forever in poll(), NEWLINE the stack trace at that point is NEWLINE NEWLINE debug> set %proc p2 NEWLINE Current process is now p2, program mozilla-export NEWLINE debug> run NEWLINE Process p2 exec'd; new executable is (dns, program (dns NEWLINE HALTED p2 [main in mozilla.c] NEWLINE 2015: <no source text available> NEWLINE debug> run NEWLINE NEWLINE NEWLINE debug> NEWLINE HALTED p2 [_poll] NEWLINE 0x18046534(_poll+20:) beq a3,zero,0x18046554 NEWLINE 0x18046538(_poll+24:) li NEWLINE at,91 NEWLINE debug> stack NEWLINE Stack Trace for p2, Program (dns NEWLINE [4] 0x18046534 _poll() NEWLINE [3] 0x00aaa550 _MD_PauseCPU(ticks=4294967295) NEWLINE [unix.c@1854] NEWLINE [2] 0x00a9bbac _PR_CPU_Idle(_cpu=0xd8b000) NEWLINE [prucpu.c@5025504] NEWLINE [1] 0x00aa06d8 _PR_UserRunThread() [pruthr.c@467] NEWLINE [0] 0x00aa06d8 _PR_UserRunThread() [pruthr.c@467] NEWLINE debug> NEWLINE NEWLINE If he runs outside the debugger, it seems to NEWLINE be getting the info from the URL, like it says NEWLINE that contacted, transferring data and so on and NEWLINE suddenly it crashes with the same assertion failure. NEWLINE NEWLINE The output of 'uname' on his machine is: NEWLINE SINIX-Y dagobert 5.43 C2003 RM600 4/512 R10000 NEWLINE NEWLINE He is building with the native cc and CC.
284|Unconfirmed Memory Leak|Created by Daniel Hable (dthable@bigfoot.com) on Wednesday, April 29, 1998 7:09:23 PM PDT NEWLINE Additional Details : NEWLINE We've been offering support for UW-Madison students and NEWLINE have seen some major problems with Netscape while running NEWLINE the OS in question. After watching the NT Task Manager, we NEWLINE have found that the process netscape.exe keeps requesting NEWLINE memory until about 12000K at which point Netscape crashes ( NEWLINE using 32 MB of RAM.) Numerous messages occur with the lack NEWLINE of Virutal Memory. NEWLINE We are looking for a confirmation on this finding. I NEWLINE already looked at developer.netscape.com and NEWLINE home.netscape.com without finding any whitepapers on the NEWLINE topic. NEWLINE Updated by (blythe@netscape.com) on Tuesday, June 9, 1998 4:19:17 PM PDT NEWLINE Additional Details : NEWLINE Will assign to danm@netscape.com as soon as they add to AssignTo field.
285|missing a mk file for Linux 2.1 machines....|Created by Jauder Ho (jauderho@transmeta.com) on Wednesday, April 29, 1998 11:44:14 PM PDT NEWLINE Additional Details : NEWLINE mozilla/nsprpub/config/LinuxELF2.1.mk needs to be a soft NEWLINE link to mozilla/nsprpub/config/LinuxELF2.0.mk to enable NEWLINE mozilla to compile for 2.1.x people. NEWLINE NEWLINE --Jauder NEWLINE Updated by Wan-Teh Chang (wtc@netscape.com) on Wednesday, May 6, 1998 8:12:03 PM PDT NEWLINE Additional Details : NEWLINE Please get the latest source code (either the tar file or NEWLINE from cvs). This has been fixed.
286|iostream.h|Created by Nayan Hajratwala (nayan@walanet.com) on Thursday, April 30, 1998 3:47:34 PM PDT NEWLINE Additional Details : NEWLINE The build procedure doesn't seem able to find my iostream.h NEWLINE NEWLINE I have it located in NEWLINE /opt/FSFlibg++/lib/g++-include/iostream.h NEWLINE NEWLINE Including a "-I" on the SunOS5.mk variable G++INCLUDES NEWLINE didn't help. I had to add it on the OS_INCLUDES variable.
287|pageup/pagedown messages|Created by Nayan Hajratwala (nayan@walanet.com) on Thursday, April 30, 1998 3:50:27 PM PDT NEWLINE Additional Details : NEWLINE when starting up mozilla, I get the following messages on my NEWLINE console: NEWLINE NEWLINE Warning: Actions not found: PageUp PageDown NEWLINE Warning: Actions not found: PageDown, LineUp, LineDown, NEWLINE PageUp, PageUp, PageDown NEWLINE NEWLINE This is repeated 8 times.
288|Where is "ToolFrontEnd" CW Plug-in ?|Created by Alexandre Parenteau (alexp@strata3d.com) on Friday, May 1, 1998 11:04:21 AM PDT NEWLINE Additional Details : NEWLINE Where is "ToolFrontEnd" CW Plug-in ? It is used in NEWLINE "NavStringLibPPC.mcp" and is mapped on the file NEWLINE "xpstring.xps". NEWLINE Updated by Alexandre Parenteau (alexp@strata3d.com) on Friday, May 1, 1998 1:20:05 PM PDT NEWLINE Additional Details : NEWLINE For get to tell. I accessed the source code thru cvs on Macintosh. So it seems NEWLINE to be a cvs tree related problem NEWLINE Updated by Alexandre Parenteau (alexp@strata3d.com) on Friday, May 1, 1998 2:28:58 PM PDT NEWLINE Additional Details : NEWLINE Forget it, I've seen the build infos on www.mozilla.org. Could be nice to have NEWLINE these notes in the cvs tree.
289|Cache database changed but files are not (19980429 as well)|Created by (tenthumbs@cybernex.net) on Friday, May 1, 1998 12:43:40 PM PDT NEWLINE Additional Details : NEWLINE It looks like the problem's in NET_CleanupCacheDirectory in NEWLINE lib/libnet/mkcache.c. NEWLINE I've added some instrumentation and now I seen the following NEWLINE when I run mozilla NEWLINE for a little bit. NEWLINE NEWLINE NEWLINE NET_CleanupCacheDirectory cleaning NEWLINE "/home/10thumbs/.mozilla/cache" with prefix "cache" NEWLINE cleaning up cache directory of unknown cache files NEWLINE Got filename: 0B/cache354A190B0003806 from DBT NEWLINE Got filename: 0D/cache354A190D0013806.gif from DBT NEWLINE Got filename: 0E/cache354A190E0033806 from DBT NEWLINE Got filename: 0E/cache354A190E0043806.dat from DBT NEWLINE Got filename: 13/cache354A19130053806.html from DBT NEWLINE Got filename: 18/cache354A19180063806.html from DBT NEWLINE Got filename: 07/cache354A19270083806.htm from DBT NEWLINE Got filename: 0E/cache354A190E0023806.jpg from DBT NEWLINE Got filename: 19/cache354A19190073806.jpg from DBT NEWLINE XP_FileRemove fails on /19/cache354A19190073806.jpg; NEWLINE result=-1; errno=2 NEWLINE Unknown cache file /19/cache354A19190073806.jpg found! -- NEWLINE deleteing... NEWLINE XP_FileRemove fails on /0B/cache354A190B0003806; NEWLINE result=-1; errno=2 NEWLINE Unknown cache file /0B/cache354A190B0003806 found! -- NEWLINE deleteing... NEWLINE XP_FileRemove fails on /18/cache354A19180063806.html; NEWLINE result=-1; errno=2 NEWLINE Unknown cache file /18/cache354A19180063806.html found! -- NEWLINE deleteing... NEWLINE XP_FileRemove fails on /0D/cache354A190D0013806.gif; NEWLINE result=-1; errno=2 NEWLINE Unknown cache file /0D/cache354A190D0013806.gif found! -- NEWLINE deleteing... NEWLINE XP_FileRemove fails on /13/cache354A19130053806.html; NEWLINE result=-1; errno=2 NEWLINE Unknown cache file /13/cache354A19130053806.html found! -- NEWLINE deleteing... NEWLINE XP_FileRemove fails on /0E/cache354A190E0023806.jpg; NEWLINE result=-1; errno=2 NEWLINE Unknown cache file /0E/cache354A190E0023806.jpg found! -- NEWLINE deleteing... NEWLINE XP_FileRemove fails on /07/cache354A19270083806.htm; NEWLINE result=-1; errno=2 NEWLINE Unknown cache file /07/cache354A19270083806.htm found! -- NEWLINE deleteing... NEWLINE XP_FileRemove fails on /0E/cache354A190E0043806.dat; NEWLINE result=-1; errno=2 NEWLINE Unknown cache file /0E/cache354A190E0043806.dat found! -- NEWLINE deleteing... NEWLINE XP_FileRemove fails on /0E/cache354A190E0033806; NEWLINE result=-1; errno=2 NEWLINE Unknown cache file /0E/cache354A190E0033806 found! -- NEWLINE deleteing... NEWLINE Deleting 3 unneeded cache entries NEWLINE Reseting number of files in cache to: 6 NEWLINE NEWLINE NET_CleanupCacheDirectory asks net_GetFilenameInCacheDBT NEWLINE which returns a correct NEWLINE relative pathname. NET_CleanupCacheDirectory then mangles it NEWLINE somewhere by prepending NEWLINE a '/' which causes XP_FileRemove to fail. This wouldn't be NEWLINE so bad but NEWLINE NET_CleanupCacheDirectory ignores the return value of NEWLINE XP_FileRemove and removes the NEWLINE filename from the cache database. NEWLINE NEWLINE Updated by (tenthumbs@cybernex.net) on Tuesday, May 19, 1998 4:54:41 AM PDT NEWLINE Additional Details : NEWLINE The internal function net_cache_recursive_file_finder doesn't work as advertised NEWLINE if the base_dir parameter does not end with a "/". This causes all teh problems NEWLINE seen later.
290|Crash on PageDown|Created by (yardley@uiuc.edu) on Saturday, May 2, 1998 11:34:21 AM PDT NEWLINE Additional Details : NEWLINE The Instruction at 0x00662495 referenced memory at NEWLINE 0xffffffff. The memory could not be read. NEWLINE NEWLINE That is the "typical" message... although it sometimes NEWLINE caries. It occurs when one scrolls down on any page, be it NEWLINE by using the scroll bar or by hitting page-down. NEWLINE Updated by (yardley@uiuc.edu) on Saturday, May 2, 1998 11:43:46 AM PDT NEWLINE Additional Details : NEWLINE More specifically... NEWLINE NEWLINE on pagedown: NEWLINE The Instruction at 0x0068A270 referenced memory at 0x00000000. The memory could NEWLINE not be written. NEWLINE - : NEWLINE 0068A270 add byte ptr [eax],al NEWLINE NEWLINE NEWLINE on scroll bar move: NEWLINE The Instruction at 0x00662490 referenced memory at 0x00000000. The memory could NEWLINE not be written. NEWLINE - on debug: NEWLINE Unhandled Exception 0xC0000005. Access Violation NEWLINE - code segment: NEWLINE 00662490 add byte ptr [eax],al NEWLINE Updated by (blythe@netscape.com) on Tuesday, June 9, 1998 4:20:30 PM PDT NEWLINE Additional Details : NEWLINE Bill, I believe you fixed this. NEWLINE The bug only happened with a VC5 release build, right?
291|<align=right> != working|Created by Jason Hall (jhall@united-railway.com) on Sunday, May 3, 1998 12:01:30 AM PDT NEWLINE Additional Details : NEWLINE Environment: NEWLINE Linux 2.0.33 NEWLINE Motif 2.0 (sorry don't know vendor) NEWLINE Mozilla 19980429 NEWLINE (note: Encountered One error in build had to comment NEWLINE remove -lXp, and add -lXpm) NEWLINE NEWLINE Code to reproduce bug: NEWLINE NEWLINE <html> NEWLINE <head><title>Foo</title></head> NEWLINE <table width=100%> NEWLINE <tr><td align=right>Hello World!</td></tr> NEWLINE </table> NEWLINE </html> NEWLINE NEWLINE Effects: NEWLINE Hello World! Appears on left hand side of navigator NEWLINE window. Appears on right hand side in Communicator 4.05.
292|Bookmarks -> '---' item Crashes Mozilla|Created by Jason Hall (jhall@united-railway.com) on Sunday, May 3, 1998 12:13:56 AM PDT NEWLINE Additional Details : NEWLINE Environment: NEWLINE Linux 2.0.33 NEWLINE Motif 2.0 -sorry dont know the vendor =( NEWLINE Mozilla 19980429 NEWLINE NEWLINE Description: NEWLINE Motif 2.0 makes these menu items that look like '-----' NEWLINE Like Tk does. In Tk they make a Window out of the menu, I'm NEWLINE assuming in Motif 2.0 they do the same. When you use that NEWLINE menu item (from the bookmarks menu) it makes mozilla NEWLINE segfault. I've also noticed other problems with the pulldown NEWLINE menus using Motif 2.0 -- I haven't been able to replicate NEWLINE them as of yet, expect another report in the next few days NEWLINE when I get more time to play with it. NEWLINE NEWLINE I have also noticed Motif 1.2.4 doesn't have this item. I'm NEWLINE assuming that this is the reason it hasnt been caught up NEWLINE until now. It would be nice to get Mozilla up an running NEWLINE fully with Motif 2.0 (I really like the pulldown selection NEWLINE boxes w/scrollbars!) NEWLINE NEWLINE Thanks.
293|Source redefines PI in dom/public/coreDom/nsIDOMNode.h|Created by Brian Grunkemeyer (bg2k@cmu.edu) on Sunday, May 3, 1998 2:02:32 AM PDT NEWLINE Additional Details : NEWLINE PI is used in an enum internally in NEWLINE dom/public/coreDom/nsIDOMNode.h. It's set to some number NEWLINE (4), but since the preprocessor expands it before it can be NEWLINE assigned, you end up with something approximating this: NEWLINE NEWLINE enum { 3.14159L = 4 }; NEWLINE NEWLINE That isn't right. I got around it by undefining PI in the NEWLINE file immediately before this, but that isn't the right thing NEWLINE to do either. Change the name from PI to something other NEWLINE than a commonly used mathematical constant. Besides, I have NEWLINE no idea what a "PI" node is from the name - it's a crappy NEWLINE name or abbreviation. NEWLINE NEWLINE The errant code is around line 44: NEWLINE NEWLINE class nsIDOMNode : public nsISupports { NEWLINE public: NEWLINE // NodeType NEWLINE enum NodeType { NEWLINE DOCUMENT = 1, NEWLINE ELEMENT = 2, NEWLINE ATTRIBUTE = 3, NEWLINE PI = 4, <<-------------- NEWLINE COMMENT = 5, NEWLINE TEXT = 6 NEWLINE }; NEWLINE NEWLINE ... NEWLINE }; NEWLINE Updated by (toshok@netscape.com) on Monday, May 18, 1998 11:38:43 AM PDT NEWLINE Additional Details : NEWLINE vidur?
294|Undisplayable web page|Created by Brian Grunkemeyer (bg2k@cmu.edu) on Sunday, May 3, 1998 12:40:31 PM PDT NEWLINE Additional Details : NEWLINE Mozilla gives up when trying to display this web page: NEWLINE http://www.geocities.com/ResearchTriangle/Lab/8720/overview.htm NEWLINE NEWLINE Displays an all-white screen and just sits there. No text NEWLINE whatsoever. Viewing the page's source HTML shows that there NEWLINE should be lots of text, and this page looks, at least in a NEWLINE cursory scan, like valid HTML. This is about the 5th page NEWLINE on the net I've seen this way, so it's definitely not a NEWLINE really obscure phenomenon. NEWLINE NEWLINE (Please don't judge the content then dismiss this bug)
295|[xfe] selection forms not remembering their state|Created by Tomas Brodsky (brodsky@cfar.umd.edu) on Sunday, May 3, 1998 2:18:17 PM PDT NEWLINE Additional Details : NEWLINE When I open a page containing selection forms, make some NEWLINE selections, then open another page and go back, the select NEWLINE forms are reset to the default values. NEWLINE NEWLINE The cause is probably in cmd/xfe/forms.c, function NEWLINE select_create_widget, where the selected fields are NEWLINE always initialized from the defaults. NEWLINE NEWLINE forms.c, line 1651 in 1998-04-29 source: NEWLINE NEWLINE selected_p [i] = !!d2->def_selected; NEWLINE d2->selected = d2->def_selected;
296|Build Failed - Undefined symbols|Created by Dwight Chang (dcrs6000@pacbell.net) on Monday, May 4, 1998 6:01:36 PM PDT NEWLINE Additional Details : NEWLINE Environment: NEWLINE IBM Rs/6000 Power 2(MCA) System NEWLINE IBM AIX 4.2.1 NEWLINE CDE NEWLINE X11 R5 NEWLINE IBM xlC Ver. 3.1.4.8, xlC Ver. 4.1 and GNU C/C++ Ver. NEWLINE 2.8.1 NEWLINE GNU make Ver. 3.76 NEWLINE Source Code Version: unix-19980429.tar.gz NEWLINE NEWLINE Problem: (Solved) NEWLINE ENDIAN definition problems NEWLINE h_page.c and hash.c had to be modified because of a NEWLINE recursive #include of the NEWLINE header file "mcom_db.h". It seems that the header file NEWLINE hash.h has an include NEWLINE for the "mcom_db.h" file. Not sure why, but when the NEWLINE compiler performs all the NEWLINE includes the ENDIAN definitions do not get defined. NEWLINE This occures when I use gcc, IBM C Ver. 3 and 4. NEWLINE NEWLINE Problem: (Solved) NEWLINE Due to "Undefined Symbol" problems I modified the "ld" NEWLINE command by adding all NEWLINE "-L" and "-l" directives to allow the build to occur. NEWLINE NEWLINE gmake of /tmp/GNU/ns/mozilla/js/src NEWLINE NEWLINE ld -brtl -bM:SRE -bnoentry NEWLINE -bexpall \\ NEWLINE -o AIX4.2_OPT.OBJ/libjs.so AIX4.2_OPT.OBJ/jsapi.o NEWLINE AIX4.2_OPT.OBJ/jsarray.o \\ NEWLINE AIX4.2_OPT.OBJ/jsatom.o AIX4.2_OPT.OBJ/jsbool.o NEWLINE AIX4.2_OPT.OBJ/jscntxt.o \\ NEWLINE AIX4.2_OPT.OBJ/jsdate.o AIX4.2_OPT.OBJ/jsdbgapi.o NEWLINE AIX4.2_OPT.OBJ/jsemit.o \\ NEWLINE AIX4.2_OPT.OBJ/jsfun.o AIX4.2_OPT.OBJ/jsgc.o NEWLINE AIX4.2_OPT.OBJ/jsinterp.o \\ NEWLINE AIX4.2_OPT.OBJ/jsmath.o AIX4.2_OPT.OBJ/jsnum.o NEWLINE AIX4.2_OPT.OBJ/jsobj.o \\ NEWLINE AIX4.2_OPT.OBJ/jsopcode.o AIX4.2_OPT.OBJ/jsparse.o NEWLINE AIX4.2_OPT.OBJ/jsregexp.o\\ NEWLINE AIX4.2_OPT.OBJ/jsscan.o AIX4.2_OPT.OBJ/jsscope.o NEWLINE AIX4.2_OPT.OBJ/jsscript.o \\ NEWLINE AIX4.2_OPT.OBJ/jsstr.o AIX4.2_OPT.OBJ/jslock.o NEWLINE AIX4.2_OPT.OBJ/jsxdrapi.o \\ NEWLINE AIX4.2_OPT.OBJ/prmjtime.o NEWLINE AIX4.2_OPT.OBJ/prarena.o \\ NEWLINE -L /tmp/GNU/ns/mozilla/dist/AIX4.2_OPT.OBJ/lib NEWLINE -lplds21 \\ NEWLINE -L /tmp/GNU/ns/mozilla/dist/AIX4.2_OPT.OBJ/lib NEWLINE -lnspr21 \\ NEWLINE -L /usr/lib -lm -lc -lc_r NEWLINE NEWLINE Problem: (UNSOLVED - UNABLE TO DETERMINE PROBLEM!) NEWLINE Due to "Undefined Symbol" problems I modified the "ld" NEWLINE command by adding all NEWLINE "-L" and "-l" directives to allow the build to occur. NEWLINE NEWLINE gmake of /tmp/GNU/ns/mozilla/xpcom/src NEWLINE NEWLINE ld -brtl -bM:SRE -bnoentry NEWLINE -bexpall \\ NEWLINE -o AIX4.2_OPT.OBJ/libxpcom.so NEWLINE AIX4.2_OPT.OBJ/nsDebug.o \\ NEWLINE AIX4.2_OPT.OBJ/nsHashtable.o NEWLINE AIX4.2_OPT.OBJ/nsID.o \\ NEWLINE NEWLINE AIX4.2_OPT.OBJ/nsRepository.o NEWLINE \\ NEWLINE -L /tmp/GNU/ns/mozilla/dist/AIX4.2_OPT.OBJ/lib NEWLINE -lnspr21 \\ NEWLINE -L /tmp/GNU/ns/mozilla/dist/AIX4.2_OPT.OBJ/lib NEWLINE -lplds21 \\ NEWLINE -L /tmp/GNU/ns/mozilla/dist/AIX4.2_OPT.OBJ/lib NEWLINE -lplc21 \\ NEWLINE -L /usr/local/lib/gcc-lib/rs6000-ibm-aix4.2.1.0/2.8.1 NEWLINE -lgcc \\ NEWLINE -lc NEWLINE NEWLINE Problem: NEWLINE Due to "Undefined Symbol" problems I modified the "ld" NEWLINE command by adding all NEWLINE "-L" and "-l" directives to allow the build to occur. NEWLINE However; in this case I was unable to find some Undefined NEWLINE Symbols in any NEWLINE of the libraries that had been created already. NEWLINE NEWLINE gmake of /tmp/GNU/ns/mozilla/modules/libhook/src NEWLINE NEWLINE ld -brtl -bM:SRE -bnoentry NEWLINE -bexpall \\ NEWLINE -o AIX4.2_OPT.OBJ/libhook.so AIX4.2_OPT.OBJ/hk_conf.o NEWLINE AIX4.2_OPT.OBJ/hk_file.o \\ NEWLINE AIX4.2_OPT.OBJ/hk_hook.o AIX4.2_OPT.OBJ/hk_init.o NEWLINE AIX4.2_OPT.OBJ/hk_tag.o \\ NEWLINE -L /tmp/GNU/ns/mozilla/dist/AIX4.2_OPT.OBJ/lib NEWLINE -ljs \\ NEWLINE -L /tmp/GNU/ns/mozilla/dist/AIX4.2_OPT.OBJ/lib NEWLINE -lplds21 \\ NEWLINE -L /tmp/GNU/ns/mozilla/dist/AIX4.2_OPT.OBJ/lib NEWLINE -lnspr21 \\ NEWLINE -L /usr/lib -lm -lc -lc_r NEWLINE NEWLINE ld: 0711-224 WARNING: Duplicate symbol: .PL_ArenaGrow NEWLINE ld: 0711-224 WARNING: Duplicate symbol: PL_ArenaGrow NEWLINE ld: 0711-224 WARNING: Duplicate symbol: .PL_ArenaRelease NEWLINE ld: 0711-224 WARNING: Duplicate symbol: .PL_ArenaAllocate NEWLINE ld: 0711-224 WARNING: Duplicate symbol: .PL_ArenaFinish NEWLINE ld: 0711-224 WARNING: Duplicate symbol: .PL_CompactArenaPool NEWLINE ld: 0711-224 WARNING: Duplicate symbol: .PL_FinishArenaPool NEWLINE ld: 0711-224 WARNING: Duplicate symbol: .PL_FreeArenaPool NEWLINE ld: 0711-224 WARNING: Duplicate symbol: .PL_InitArenaPool NEWLINE ld: 0711-224 WARNING: Duplicate symbol: PL_ArenaAllocate NEWLINE ld: 0711-224 WARNING: Duplicate symbol: PL_ArenaRelease NEWLINE ld: 0711-224 WARNING: Duplicate symbol: PL_FinishArenaPool NEWLINE ld: 0711-224 WARNING: Duplicate symbol: PL_InitArenaPool NEWLINE ld: 0711-224 WARNING: Duplicate symbol: PL_ArenaFinish NEWLINE ld: 0711-224 WARNING: Duplicate symbol: PL_FreeArenaPool NEWLINE ld: 0711-345 Use the -bloadmap or -bnoquiet option to obtain NEWLINE more information. NEWLINE ld: 0711-317 ERROR: Undefined symbol: .PA_TagIndex NEWLINE ld: 0711-317 ERROR: Undefined symbol: .PA_TagString NEWLINE ld: 0711-317 ERROR: Undefined symbol: .PREF_GetConfigContext NEWLINE ld: 0711-317 ERROR: Undefined symbol: NEWLINE .PREF_GetGlobalConfigObject NEWLINE ld: 0711-317 ERROR: Undefined symbol: .xp_toupper NEWLINE ld: 0711-317 ERROR: Undefined symbol: .FE_Alert NEWLINE ld: 0711-317 ERROR: Undefined symbol: .XP_Stat NEWLINE ld: 0711-317 ERROR: Undefined symbol: .XP_FileOpen NEWLINE ld: 0711-317 ERROR: Undefined symbol: .XP_FileClose NEWLINE NEWLINE Updated by Brian Ostrom (briano@netscape.com) on Monday, July 13, 1998 6:07:49 PM PDT NEWLINE Additional Details : NEWLINE Is this still a problem with the latest source release?
297|Incomplete download of cgi page|Created by Giuseppe Fedon (gfedon@ing.unico.it) on Tuesday, May 5, 1998 6:10:43 AM PDT NEWLINE Additional Details : NEWLINE I make a page width a cgi program and the download is NEWLINE incomplate width communicator 4.0x. NEWLINE The same page is viewable width explorer 4.x and netscape NEWLINE 3.x. NEWLINE I have Server OS WinNT width Web Server IIS4 and the same NEWLINE repons width Omnihttpd2.a7 and apache 1.3b3 NEWLINE NEWLINE The test url is NEWLINE http://corsionline.como.polimi.it:80/AutoWeb/corsionline.exe NEWLINE ?Comp=4,1,1,0+Elem=0,7,0,0+Web=21,63,63,64+ElRef=7,0,0,0+Col NEWLINE l=0,0,13,14+Mode=0,0,1,1+Code=2495007159+Frame=2+Part=19
298|Missing Personal Toolbar Bookmark folder == instant crash|Created by Mark LTC (mark@zang.com) on Tuesday, May 5, 1998 12:35:02 PM PDT NEWLINE Additional Details : NEWLINE My platform: SPARC Solaris 2.5 gcc 2.7.2.3 NEWLINE Version: Mozilla unix-19980429 + cvs as of 4th May NEWLINE NEWLINE NEWLINE Well cut and paste to this blows so check out the prob at: NEWLINE NEWLINE http://sunsite.unc.edu/mark/mozilla/crash.bug.html
299|Build fails on wrong include path|Created by Berkley Shands (berkley@cs.wustl.edu) on Wednesday, May 6, 1998 6:28:08 AM PDT NEWLINE Additional Details : NEWLINE /usr/sys/include/arch/alpha/endian.h is not in the NEWLINE standard include path on this 433AU with NO GNU installed. NEWLINE Several include files in mozilla/19980429/include NEWLINE reference this file. Either copy it to the include directory NEWLINE (this is what I did) or explicitly add it to the Makefile. NEWLINE NEWLINE berkley NEWLINE Updated by Wan-Teh Chang (wtc@netscape.com) on Wednesday, May 6, 1998 8:17:56 PM PDT NEWLINE Additional Details : NEWLINE This one should be #include <machine/endian.h>. NEWLINE Please see mozilla/nsprpub/pr/include/prinet.h NEWLINE for an example. NEWLINE Updated by Brian Ostrom (briano@netscape.com) on Monday, July 13, 1998 5:56:26 PM PDT NEWLINE Additional Details : NEWLINE This is a duplicate of bug#249.
300|CXX V6.0-20 +ANSI requires explicit cast|Created by Berkley Shands (berkley@cs.wustl.edu) on Wednesday, May 6, 1998 6:32:48 AM PDT NEWLINE Additional Details : NEWLINE Lines 374 and 408 have "NULL" as arguments to constructors. NEWLINE I use "#define NULL 0L" to get around this cast problem, NEWLINE but for now I edited those lines to have explicit casts NEWLINE (const char*) NULL, and (ns<something> *). NEWLINE NEWLINE berkley
301|CXX V6.0-20 "NULL" explicit cast needed|Created by Berkley Shands (berkley@cs.wustl.edu) on Wednesday, May 6, 1998 6:51:18 AM PDT NEWLINE Additional Details : NEWLINE cxx: Error: fb.cpp, line 57: no instance of constructor NEWLINE "wfList::wfList" matches the argument list NEWLINE argument types are: (long) NEWLINE FontBrokerObject() : fpPeers(NULL), fpPeersFromCatalog(NULL) NEWLINE -----------------------------^ NEWLINE NEWLINE This is similar to the other CXX complaint. GNU just doesn't NEWLINE care about arguments. the CXX compiler is VERY picky. NEWLINE the fix is to add a constructor to wfList.h supporting NEWLINE an argument of long... NEWLINE at line 66 NEWLINE public: NEWLINE // Constructor NEWLINE wfList(WF_FREE_FUNC freeFunc); NEWLINE #if defined(OSF1) NEWLINE wfList(long); NEWLINE #endif NEWLINE NEWLINE and add code in wfList.cpp NEWLINE at line 49 NEWLINE #if defined(OSF1) NEWLINE wfList::wfList(WF_FREE_FUNC freeFunc) NEWLINE : head(NULL), tail(NULL), freeItemFunc(NULL) NEWLINE { NEWLINE } NEWLINE FYI - In my multiplatform code I've collected these... NEWLINE /* The following pre-defined symbols are provided by the NEWLINE compiler */ NEWLINE NEWLINE /* _MSC_VER - Visual C++ version ID 4.2E */ NEWLINE /* __BCPLUSPLUS__ - Borland C++ V5.0 */ NEWLINE /* __SUNPRO_CC - Sun Sparcworks Pro V3.1 */ NEWLINE /* __DECCXX - Digital CXX V5.6 */ NEWLINE /* __DECCXX_VER = 60030001 first F/T of V6.0 */ NEWLINE /* __DECCXX_VER = 50660098 -oldcxx (V5.6-098) */ NEWLINE /* __UNIX All flavors of real unix */ NEWLINE /* WIN32 Both flavors of Intel/MS NT4.0 */ NEWLINE /* __GNUG__ GNU G++ */ NEWLINE /* __NetBSD__ NetBSD unix */ NEWLINE /* __i386 used with NetBSD for Intel NEWLINE platforms */ NEWLINE /* __mips64 SGI 7.x Compilers (not pre-defined) NEWLINE */ NEWLINE /* __mips SGI systems (GNU) */ NEWLINE /* __linux__ Linux running on an Intel Platform NEWLINE */ NEWLINE NEWLINE #endif
302|Another NULL cast issue in wffpPeer.cpp|Created by Berkley Shands (berkley@cs.wustl.edu) on Wednesday, May 6, 1998 7:08:54 AM PDT NEWLINE Additional Details : NEWLINE cxx: Error: wffpPeer.cpp, line 43: no instance of NEWLINE constructor "wfDlm::wfDlm" matches the argument list NEWLINE argument types are: (long) NEWLINE : dlm(NULL), deleted(0), disabled(0), native(-1), NEWLINE --------------^ NEWLINE cxx: Error: wffpPeer.cpp, line 76: no instance of NEWLINE constructor "wfMimeList::wfMimeList" matches the argument NEWLINE list NEWLINE argument types are: (long) NEWLINE : dlm(dlmName), mimeList(NULL), deleted(0), NEWLINE disabled(0), native(-1), NEWLINE ---------------------------------^ NEWLINE cxx: Error: wffpPeer.cpp, line 125: no instance of NEWLINE constructor "wfDlm::wfDlm" matches the argument list NEWLINE argument types are: (long) NEWLINE : dlm(NULL), deleted(0), disabled(0), native(-1), NEWLINE --------------^ NEWLINE at line 45 change to NEWLINE FontDisplayerPeerObject(struct nffp *fp) NEWLINE : dlm((const char*) NULL, (const char*) NULL), deleted(0), NEWLINE disabled(0), native(-1), NEWLINE streamCount(0), fhList(NULL), unloadTimerId(NULL) NEWLINE NEWLINE same at line 76 (add explicit cast on NULL) NEWLINE same at line 125 (add explicit args and casts) NEWLINE NEWLINE berkley
303|More NULL casts in wffpCat.cpp|Created by Berkley Shands (berkley@cs.wustl.edu) on Wednesday, May 6, 1998 7:12:19 AM PDT NEWLINE Additional Details : NEWLINE cxx: Error: wffpCat.cpp, line 289: no instance of overloaded NEWLINE function "FontDisplayerCatalogObject::addFmi" matches the NEWLINE argument list NEWLINE argument types are: (jint, jint, long) NEWLINE addFmi(rcMajorType, NEWLINE rcMinorType, NULL); NEWLINE ----------------------------------------^ NEWLINE 1 error detected in the compilation of "wffpCat.cpp". NEWLINE at line 289 add(struct nffmi*) NULL. NEWLINE NEWLINE berkley
304|BUILD: catch-up with class change for "readonly" keyword|Created by Berkley Shands (berkley@cs.wustl.edu) on Wednesday, May 6, 1998 8:59:23 AM PDT NEWLINE Additional Details : NEWLINE cc: Error: layform.c, line 989: Identifier expected but not NEWLINE found. NEWLINE form_data->readonly = FALSE; NEWLINE -------------------^ NEWLINE cc: Error: layform.c, line 996: Identifier expected but not NEWLINE found. NEWLINE form_data->readonly = TRUE; NEWLINE -----------------------------------^ NEWLINE cc: Error: layform.c, line 1149: Identifier expected but not NEWLINE found. NEWLINE form_data->readonly = FALSE; NEWLINE -------------------^ NEWLINE cc: Error: layform.c, line 1154: Identifier expected but not NEWLINE found. NEWLINE form_data->readonly = TRUE; NEWLINE ---------------------------^ NEWLINE NEWLINE this was changed to "readonly2" to get around a keyword NEWLINE problem submitted as an earier bug. NEWLINE NEWLINE berkley
305|edtbuf.cpp SelectTable overload needs explicit type cast|Created by Berkley Shands (berkley@cs.wustl.edu) on Wednesday, May 6, 1998 9:24:53 AM PDT NEWLINE Additional Details : NEWLINE cxx: Error: edtbuf.cpp, line 14446: no instance of NEWLINE overloaded function "CEditBuffer::SelectTable" matches the NEWLINE argument list NEWLINE argument types are: (int, long, NEWLINE CEditTableElement *) NEWLINE SelectTable(FALSE, NULL, NEWLINE (CEditTableElement*)pElement); NEWLINE ----------------^ NEWLINE 1 error detected in the compilation of "edtbuf.cpp". NEWLINE make: *** [OSF1V4.0_DBG.OBJ/edtbuf.o] Error 1 NEWLINE mozilla/lib/layout> NEWLINE NEWLINE this is in the 19980429 edition. NEWLINE NEWLINE NEWLINE if( pElement->IsTable() ) NEWLINE { NEWLINE SelectTable(FALSE, (LO_TableStruct*) NULL, NEWLINE (CEditTableElement*)pElement); NEWLINE return; NEWLINE } NEWLINE else if( pElement->IsTableCell() ) NEWLINE { NEWLINE SelectCell(FALSE, (LO_CellStruct*) NULL, NEWLINE (CEditTableCellElement*)pElement); NEWLINE return; NEWLINE } NEWLINE NEWLINE berkley
306|explict type cast needed in edtele.cpp lines 5955 & 10668|Created by Berkley Shands (berkley@cs.wustl.edu) on Wednesday, May 6, 1998 9:33:58 AM PDT NEWLINE Additional Details : NEWLINE from mozilla/19980429/mozilla/lib/layout: NEWLINE NEWLINE cxx: Error: edtele.cpp, line 5955: no instance of NEWLINE constructor NEWLINE "CEditEndContainerElement::CEditEndContainerElement" matches NEWLINE the argument list NEWLINE argument types are: (long) NEWLINE CEditEndContainerElement* pEndContainer = new NEWLINE CEditEndContainerElement(NULL); NEWLINE -------------------------------------------------------------------------------^ NEWLINE cxx: Error: edtele.cpp, line 10668: no instance of NEWLINE constructor "CEditContainerElement::CEditContainerElement" NEWLINE matches the argument list NEWLINE argument types are: (CEditElement *, long, int, NEWLINE ED_Alignment) NEWLINE CEditContainerElement(pParent, NULL, 0 /* Never used NEWLINE for an end container */, ED_ALIGN_LEFT) NEWLINE ------------------------------^ NEWLINE the fixes - NEWLINE CEditEndContainerElement::CEditEndContainerElement(CEditElement NEWLINE *pParent) : NEWLINE CEditContainerElement(pParent, (PA_Tag*) NULL, 0 /* NEWLINE Never used for an end container */, ED_ALIGN_LEFT) NEWLINE NEWLINE and NEWLINE NEWLINE CEditEndContainerElement* pEndContainer = new NEWLINE CEditEndContainerElement((CEditElement *)NULL); NEWLINE NEWLINE Geez, I need to figure out you patch system... NEWLINE berkley
307|another explicit cast in edtutil.cpp|Created by Berkley Shands (berkley@cs.wustl.edu) on Wednesday, May 6, 1998 9:36:57 AM PDT NEWLINE Additional Details : NEWLINE cxx: Warning: edtutil.cpp, line 4902: variable "iHeight" is NEWLINE used before its value is set NEWLINE iHeight = (iHeight * 100) / m_iViewHeight; NEWLINE -----------------------^ NEWLINE cxx: Error: edtutil.cpp, line 5380: no instance of NEWLINE overloaded function "CEditLinkManager::AdjustLink" matches NEWLINE the argument list NEWLINE argument types are: (ED_LinkId, char *, long, NEWLINE long) NEWLINE pEditBuffer->linkManager.AdjustLink(linkId, NEWLINE pBaseURL, NULL, NULL); NEWLINE -------------------------------------^ NEWLINE NEWLINE fix is NEWLINE pEditBuffer->linkManager.AdjustLink(linkId, NEWLINE pBaseURL, NEWLINE (char*) NULL, (ED_HREFList *) NULL); NEWLINE NEWLINE berkley
308|19980429/mozilla/lib/mariner/src/xpform.c keyword change|Created by Berkley Shands (berkley@cs.wustl.edu) on Wednesday, May 6, 1998 10:01:41 AM PDT NEWLINE Additional Details : NEWLINE with the change of "readonly" to readonly2" this file NEWLINE needs a quick update. NEWLINE NEWLINE cc: Error: xpform.c, line 192: Identifier expected but not NEWLINE found. NEWLINE return form->ele_text.readonly; NEWLINE --------------------------------^ NEWLINE cc: Error: xpform.c, line 194: Identifier expected but not NEWLINE found. NEWLINE return form->ele_textarea.readonly; NEWLINE ------------------------------------^ NEWLINE make: *** [OSF1V4.0_DBG.OBJ/xpform.o] Error 1 NEWLINE NEWLINE case FORM_TYPE_READONLY: NEWLINE return form->ele_text.readonly2; NEWLINE case FORM_TYPE_TEXTAREA: NEWLINE return form->ele_textarea.readonly2; NEWLINE default: NEWLINE NEWLINE berkley NEWLINE NEWLINE Updated by (toshok@netscape.com) on Monday, May 18, 1998 11:14:02 AM PDT NEWLINE Additional Details : NEWLINE oh my. is this a reserved word on dec's compiler? NEWLINE NEWLINE ok, i'll change it. NEWLINE Updated by (toshok@netscape.com) on Monday, May 18, 1998 2:56:51 PM PDT NEWLINE Additional Details : NEWLINE ok, as soon as my build finishes I'll be checking in a fix for this. instead of NEWLINE readonly2, I've named it read_only.
309|explicit cast needed in 19980429/mozilla/cmd/xfe/src/ Button.cpp|Created by Berkley Shands (berkley@cs.wustl.edu) on Wednesday, May 6, 1998 10:17:18 AM PDT NEWLINE Additional Details : NEWLINE cxx: Warning: Command.h, line 242: "void NEWLINE XFE_ObjectIsCommand::doCommand(XFE_View *, XFE_CommandInfo NEWLINE *)" does not match "XFE_ViewCommand::doCommand" -- virtual NEWLINE function override intended? NEWLINE void doCommand(XFE_View* view, NEWLINE XFE_CommandInfo* info); NEWLINE ----------------------^ NEWLINE cxx: Error: Button.cpp, line 304: no instance of constructor NEWLINE "XFE_CommandInfo::XFE_CommandInfo" matches the argument list NEWLINE argument types are: (XFE_CommandEventType, NEWLINE Widget, XEvent *, long, int) NEWLINE XFE_CommandInfo NEWLINE e_info(XFE_COMMAND_BUTTON_ACTIVATE, NEWLINE -----------------------------------------------^ NEWLINE cxx: Error: Button.cpp, line 327: no instance of constructor NEWLINE "XFE_CommandInfo::XFE_CommandInfo" matches the argument list NEWLINE argument types are: (XFE_CommandEventType, NEWLINE Widget, XEvent *, long, int) NEWLINE XFE_CommandInfo NEWLINE e_info(XFE_COMMAND_BUTTON_ACTIVATE, NEWLINE -----------------------------------------------^ NEWLINE 2 errors detected in the compilation of "Button.cpp". NEWLINE NEWLINE the needed cast is (String*) NEWLINE XFE_CommandInfo e_info(XFE_COMMAND_BUTTON_ACTIVATE, NEWLINE w,cbs->event,(String*) NULL,0); NEWLINE NEWLINE berkley
310|19980429/mozilla/cmd/xfe/src/BookmarkFrame.cpp explicit typecast|Created by Berkley Shands (berkley@cs.wustl.edu) on Wednesday, May 6, 1998 10:23:40 AM PDT NEWLINE Additional Details : NEWLINE cxx: Error: BookmarkFrame.cpp, line 136: no instance of NEWLINE constructor "XFE_BookmarkView::XFE_BookmarkView" matches the NEWLINE argument list NEWLINE argument types are: (XFE_BookmarkFrame *, NEWLINE Widget, long, MWContext *) NEWLINE XFE_BookmarkView *view = new XFE_BookmarkView(this, NEWLINE ------------------------------------------------^ NEWLINE 1 error detected in the compilation of "BookmarkFrame.cpp". NEWLINE NEWLINE the fix is to add NEWLINE XFE_BookmarkView *view = new XFE_BookmarkView(this, NEWLINE getChromeParent(), NEWLINE (XFE_View *) NULL, NEWLINE m_context); NEWLINE at line 140 NEWLINE NEWLINE berkley
311|explict type cast needed|Created by Berkley Shands (berkley@cs.wustl.edu) on Wednesday, May 6, 1998 10:27:14 AM PDT NEWLINE Additional Details : NEWLINE cxx: Warning: Command.h, line 242: "void NEWLINE XFE_ObjectIsCommand::doCommand(XFE_View *, XFE_CommandInfo NEWLINE *)" does not match "XFE_ViewCommand::doCommand" -- virtual NEWLINE function override intended? NEWLINE void doCommand(XFE_View* view, NEWLINE XFE_CommandInfo* info); NEWLINE ----------------------^ NEWLINE cxx: Error: BookmarkView.cpp, line 1842: no instance of NEWLINE constructor "XFE_PopupMenu::XFE_PopupMenu" matches the NEWLINE argument list NEWLINE argument types are: (char [6], XFE_Frame *, NEWLINE Widget, long) NEWLINE m_popup = new XFE_PopupMenu("popup", NEWLINE ------------------------------^ NEWLINE 1 error detected in the compilation of "BookmarkView.cpp". NEWLINE the fix at line 1845 is NEWLINE m_popup = new XFE_PopupMenu("popup", NEWLINE (XFE_Frame*)m_toplevel, // XXXXXXX NEWLINE widget, NEWLINE (MenuSpec *) NULL); NEWLINE NEWLINE berkley
312|explicit cast needed (ANSI C++)|Created by Berkley Shands (berkley@cs.wustl.edu) on Wednesday, May 6, 1998 10:31:18 AM PDT NEWLINE Additional Details : NEWLINE 19980429/mozilla/cmd/xfe/src/BrowserDrop.cpp NEWLINE NEWLINE cxx: Warning: Command.h, line 242: "void NEWLINE XFE_ObjectIsCommand::doCommand(XFE_View *, XFE_CommandInfo NEWLINE *)" does not match "XFE_ViewCommand::doCommand" -- virtual NEWLINE function override intended? NEWLINE void doCommand(XFE_View* view, NEWLINE XFE_CommandInfo* info); NEWLINE ----------------------^ NEWLINE cxx: Error: BrowserDrop.cpp, line 219: no instance of NEWLINE constructor "XFE_BrowserFrame::XFE_BrowserFrame" matches the NEWLINE argument list NEWLINE argument types are: (Widget, XFE_BrowserFrame *, NEWLINE long) NEWLINE XFE_BrowserFrame *newFrame=new NEWLINE XFE_BrowserFrame(XtParent(_browserFrame->getBaseWidget()), NEWLINE _browserFrame, NULL); NEWLINE --------------------------------------------------------^ NEWLINE NEWLINE the fix is at line 220 NEWLINE XFE_BrowserFrame *newFrame=new NEWLINE NEWLINE XFE_BrowserFrame(XtParent(_browserFrame->getBaseWidget()), NEWLINE _browserFrame, (Chrome *) NULL); NEWLINE NEWLINE berkley
313|explicit typecast needed in BrowserFrame.cpp|Created by Berkley Shands (berkley@cs.wustl.edu) on Wednesday, May 6, 1998 10:37:47 AM PDT NEWLINE Additional Details : NEWLINE cxx: Warning: Command.h, line 192: overloaded virtual NEWLINE function "XFE_AbstractCommand::doCommand" is only partially NEWLINE overridden in class "XFE_FrameCommand" NEWLINE class XFE_FrameCommand : public XFE_Command NEWLINE ------^ NEWLINE cxx: Warning: Command.h, line 237: overloaded virtual NEWLINE function "XFE_AbstractCommand::doCommand" is only partially NEWLINE overridden in class "XFE_ObjectIsCommand" NEWLINE class XFE_ObjectIsCommand : public XFE_ViewCommand NEWLINE ------^ NEWLINE cxx: Warning: Command.h, line 242: "void NEWLINE XFE_ObjectIsCommand::doCommand(XFE_View *, XFE_CommandInfo NEWLINE *)" does not match "XFE_ViewCommand::doCommand" -- virtual NEWLINE function override intended? NEWLINE void doCommand(XFE_View* view, NEWLINE XFE_CommandInfo* info); NEWLINE ----------------------^ NEWLINE cxx: Error: BrowserFrame.cpp, line 232: no instance of NEWLINE constructor "XFE_HTMLView::XFE_HTMLView" matches the NEWLINE argument list NEWLINE argument types are: (XFE_BrowserFrame *, Widget, NEWLINE long, MWContext *) NEWLINE htmlview = new XFE_HTMLView(this, getChromeParent(), NULL, NEWLINE m_context); NEWLINE NEWLINE the fix is NEWLINE htmlview = new XFE_HTMLView(this, getChromeParent(), NEWLINE (XFE_View *) NULL, m_context); NEWLINE NEWLINE berkley NEWLINE NEWLINE ------------------------------^
314|another explicit cast in DownloadFrame.cpp|Created by Berkley Shands (berkley@cs.wustl.edu) on Wednesday, May 6, 1998 10:45:58 AM PDT NEWLINE Additional Details : NEWLINE cxx: Error: DownloadFrame.cpp, line 262: no instance of NEWLINE constructor "XFE_Frame::XFE_Frame" matches the argument list NEWLINE argument types are: (char [9], Widget, XFE_Frame NEWLINE *, EFrameType, long, int, int, int, NEWLINE int, int) NEWLINE : XFE_Frame("Download", // name NEWLINE --------------------^ NEWLINE cxx: Error: DownloadFrame.cpp, line 278: no instance of NEWLINE constructor "XFE_DownloadView::XFE_DownloadView" matches the NEWLINE argument list NEWLINE argument types are: (XFE_DownloadFrame *, NEWLINE Widget, long, MWContext *) NEWLINE XFE_DownloadView *v = new XFE_DownloadView(this, NEWLINE ---------------------------------------------------^ NEWLINE NEWLINE fix at line 266 NEWLINE NEWLINE (Chrome *) NULL, // chrome NEWLINE NEWLINE and at 280 NEWLINE NEWLINE (XFE_View *) NULL, NEWLINE NEWLINE berkley
315|explicit casts in Frame.cpp|Created by Berkley Shands (berkley@cs.wustl.edu) on Wednesday, May 6, 1998 10:59:30 AM PDT NEWLINE Additional Details : NEWLINE cxx: Error: Frame.cpp, line 904: no instance of constructor NEWLINE "XFE_MenuBar::XFE_MenuBar" matches the argument list NEWLINE argument types are: (XFE_Frame *, long) NEWLINE m_menubar = new XFE_MenuBar(this, NEWLINE NULL); NEWLINE ----------------------------------------------------^ NEWLINE cxx: Error: Frame.cpp, line 988: no instance of constructor NEWLINE "XFE_Toolbar::XFE_Toolbar" matches the argument list NEWLINE argument types are: (XFE_Frame *, XFE_Toolbox *, NEWLINE long) NEWLINE m_toolbar = new NEWLINE XFE_Toolbar(this,m_toolbox,NULL); NEWLINE --------------------------------------------^ NEWLINE cxx: Warning: Frame.cpp, line 3787: conversion from pointer NEWLINE to smaller integer NEWLINE XP_Bool busy = (XP_Bool)(int)clientData; NEWLINE --------------------------------^ NEWLINE cxx: Error: Frame.cpp, line 3878: no instance of overloaded NEWLINE function "XFE_Command::getDocString" matches the argument NEWLINE list NEWLINE argument types are: (XFE_Frame *, long) NEWLINE doc_string = handler->getDocString(this, NEWLINE NULL); NEWLINE --------------------------------------^ NEWLINE 3 errors detected in the compilation of "Frame.cpp". NEWLINE NEWLINE fix @ 90 m_menubar = new XFE_MenuBar(this, (MenuSpec NEWLINE *)NULL); NEWLINE 4 NEWLINE fix at 991 NEWLINE NEWLINE m_toolbar = new XFE_Toolbar(this, NEWLINE m_toolbox, NEWLINE (ToolbarSpec *) NULL); NEWLINE NEWLINE fix at 3878 NEWLINE NEWLINE doc_string = handler->getDocString(this, NEWLINE (XFE_CommandInfo*) NULL); NEWLINE NEWLINE NEWLINE berkley
316|explixit type cast needed in HTMLDialogFrame.cpp|Created by Berkley Shands (berkley@cs.wustl.edu) on Wednesday, May 6, 1998 11:03:24 AM PDT NEWLINE Additional Details : NEWLINE cxx: Error: HTMLDialogFrame.cpp, line 54: no instance of NEWLINE constructor "XFE_HTMLView::XFE_HTMLView" matches the NEWLINE argument list NEWLINE argument types are: (XFE_HTMLDialogFrame *, NEWLINE Widget, long, MWContext *) NEWLINE htmlview = new XFE_HTMLView(this, getChromeParent(), NEWLINE NULL, m_context); NEWLINE ------------------------------------^ NEWLINE 1 error detected in the compilation of NEWLINE "HTMLDialogFrame.cpp". NEWLINE Dec Unix 4.0D + CXX V6.0-20 (ANSI C++) NEWLINE on 19980429/mozilla/cmd/xfe/src NEWLINE the fix at line 54 is NEWLINE NEWLINE htmlview = new XFE_HTMLView(this, getChromeParent(), NEWLINE (XFE_View *) NULL, m_context); NEWLINE NEWLINE berkley
317|explicit typecast in HistoryFrame.cpp|Created by Berkley Shands (berkley@cs.wustl.edu) on Wednesday, May 6, 1998 11:05:38 AM PDT NEWLINE Additional Details : NEWLINE mozilla/19980429/mozilla/cmd/xfe/src/HistoryFrame.cpp NEWLINE cxx: Error: HistoryFrame.cpp, line 113: no instance of NEWLINE constructor "XFE_HistoryView::XFE_HistoryView" matches the NEWLINE argument list NEWLINE argument types are: (XFE_HistoryFrame *, Widget, NEWLINE long, MWContext *) NEWLINE XFE_HistoryView *view = new XFE_HistoryView(this, NEWLINE ----------------------------------------------^ NEWLINE NEWLINE the fix at line 115 is NEWLINE (XFE_View *) NEWLINE NULL, NEWLINE berkley
318|explicit cast in Menu.cpp|Created by Berkley Shands (berkley@cs.wustl.edu) on Wednesday, May 6, 1998 11:12:27 AM PDT NEWLINE Additional Details : NEWLINE cxx: Error: Menu.cpp, line 176: no instance of constructor NEWLINE "XFE_CommandInfo::XFE_CommandInfo" matches the argument list NEWLINE argument types are: (XFE_CommandEventType, NEWLINE Widget, long, char **) NEWLINE XFE_CommandInfo info(XFE_COMMAND_BUTTON_ACTIVATE, NEWLINE -----------------------^ NEWLINE cxx: Error: Menu.cpp, line 361: no instance of constructor NEWLINE "XFE_CommandInfo::XFE_CommandInfo" matches the argument list NEWLINE argument types are: (XFE_CommandEventType, NEWLINE Widget, long) NEWLINE XFE_CommandInfo info(XFE_COMMAND_BUTTON_ACTIVATE, NEWLINE -----------------------^ NEWLINE 2 errors detected in the compilation of "Menu.cpp". NEWLINE NEWLINE the fix in both places is (XEvent*) NULL. NEWLINE NEWLINE berkley
319|explicit cast in MozillaApp.cpp|Created by Berkley Shands (berkley@cs.wustl.edu) on Wednesday, May 6, 1998 11:17:27 AM PDT NEWLINE Additional Details : NEWLINE cxx: Error: MozillaApp.cpp, line 944: no instance of NEWLINE constructor "XFE_CommandInfo::XFE_CommandInfo" matches the NEWLINE argument list NEWLINE argument types are: (XFE_CommandEventType, NEWLINE Widget, XEvent *, long, int) NEWLINE XFE_CommandInfo info(XFE_COMMAND_EVENT_ACTION, w, event, NEWLINE NULL, 0); NEWLINE -----------------------^ NEWLINE 1 error detected in the compilation of "MozillaApp.cpp". NEWLINE NEWLINE the fix is to add (String*) NULL at line 944
320|explicit cast in NavCenterFrame.cpp|Created by Berkley Shands (berkley@cs.wustl.edu) on Wednesday, May 6, 1998 11:19:28 AM PDT NEWLINE Additional Details : NEWLINE mozilla/19980429/mozilla/cmd/xfe/src/NavCenterFrame.cpp NEWLINE cxx: Error: NavCenterFrame.cpp, line 65: no instance of NEWLINE constructor "XFE_NavCenterView::XFE_NavCenterView" matches NEWLINE the argument list NEWLINE argument types are: (XFE_NavCenterFrame *, NEWLINE Widget, long, MWContext *) NEWLINE XFE_View *view = new XFE_NavCenterView(this, NEWLINE getChromeParent(), NEWLINE -----------------------------------------^ NEWLINE 1 error detected in the compilation of "NavCenterFrame.cpp". NEWLINE NEWLINE the fix is NEWLINE XFE_View *view = new XFE_NavCenterView(this, NEWLINE getChromeParent(), NEWLINE (XFE_View *) NULL, NEWLINE m_context); NEWLINE NEWLINE berkley
321|explicit cast needed in NavCenterView.cpp|Created by Berkley Shands (berkley@cs.wustl.edu) on Wednesday, May 6, 1998 11:22:45 AM PDT NEWLINE Additional Details : NEWLINE mozilla/19980429/mozilla/cmd/xfe/src/NavCenterView.cpp NEWLINE cxx: Error: NavCenterView.cpp, line 91: no instance of NEWLINE constructor "XFE_RDFView::XFE_RDFView" matches the argument NEWLINE list NEWLINE argument types are: (XFE_NavCenterView *, NEWLINE Widget, long, MWContext *, HT_View) NEWLINE m_rdfview = new XFE_RDFView(this, rdf_parent, NEWLINE ------------------------------^ NEWLINE NEWLINE the fix is NEWLINE m_rdfview = new XFE_RDFView(this, rdf_parent, NEWLINE (XFE_View *) NULL, context, NEWLINE m_htview); NEWLINE NEWLINE berkley
322|explicit cast needed in PopupMenu.cpp|Created by Berkley Shands (berkley@cs.wustl.edu) on Wednesday, May 6, 1998 11:27:57 AM PDT NEWLINE Additional Details : NEWLINE mozilla/19980429/mozilla/cmd/xfe/src/PopupMenu.cpp NEWLINE cxx: Error: PopupMenu.cpp, line 54: no instance of NEWLINE constructor "XFE_Menu::XFE_Menu" matches the argument list NEWLINE argument types are: (XFE_Frame *, long, long) NEWLINE XFE_Menu(parent_frame, NEWLINE -----------------^ NEWLINE NEWLINE the fix at line 55 is NEWLINE (MenuSpec *) NULL, NEWLINE (Widget) NULL) NEWLINE NEWLINE berkley
323|explicit cast needed in URLBar.cpp|Created by Berkley Shands (berkley@cs.wustl.edu) on Wednesday, May 6, 1998 11:42:14 AM PDT NEWLINE Additional Details : NEWLINE cxx: Error: URLBar.cpp, line 809: no instance of constructor NEWLINE "XFE_PopupMenu::XFE_PopupMenu" matches the argument list NEWLINE argument types are: (char [6], XFE_Frame *, NEWLINE Widget, long) NEWLINE m_popup = new NEWLINE XFE_PopupMenu("popup",(XFE_Frame*)m_toplevel, widget, NULL); NEWLINE ----------------------------------^ NEWLINE NEWLINE the fix at line 811 is NEWLINE m_popup = new NEWLINE XFE_PopupMenu("popup",(XFE_Frame*)m_toplevel, NEWLINE widget, (MenuSpec *) NULL); NEWLINE NEWLINE berkley
324|explicit casts needed in EditorFrame.cpp|Created by Berkley Shands (berkley@cs.wustl.edu) on Wednesday, May 6, 1998 11:47:15 AM PDT NEWLINE Additional Details : NEWLINE 19980429/mozilla/cmd/xfe/src/EditorFrame.cppcxx: Error: NEWLINE EditorFrame.cpp, line 523: no instance of constructor NEWLINE "XFE_PopupMenu::XFE_PopupMenu" matches the argument list NEWLINE argument types are: (char [6], XFE_Frame *, NEWLINE Widget, long) NEWLINE popup = new XFE_PopupMenu("popup",frame, parent, NEWLINE NULL); NEWLINE ----------------------------------^ NEWLINE cxx: Error: EditorFrame.cpp, line 715: no instance of NEWLINE constructor "XFE_EditorView::XFE_EditorView" matches the NEWLINE argument list NEWLINE argument types are: (XFE_EditorFrame *, Widget, NEWLINE long, MWContext *) NEWLINE XFE_EditorView *editorview = new XFE_EditorView(this, NEWLINE --------------------------------------------------^ NEWLINE 2 errors detected in the compilation of "EditorFrame.cpp". NEWLINE NEWLINE NEWLINE the fixes at line 523 NEWLINE NEWLINE popup = new XFE_PopupMenu("popup",frame, parent, (MenuSpec NEWLINE *) NULL); NEWLINE NEWLINE NEWLINE the fixes at line 715 NEWLINE XFE_EditorView *editorview = new XFE_EditorView(this, NEWLINE getChromeParent(), NEWLINE (XFE_View *) NULL, NEWLINE m_context); NEWLINE NEWLINE berkley
325|explicit cast in EditorToolbar.cpp|Created by Berkley Shands (berkley@cs.wustl.edu) on Wednesday, May 6, 1998 12:03:07 PM PDT NEWLINE Additional Details : NEWLINE cxx: Error: EditorToolbar.cpp, line 375: no instance of NEWLINE constructor "XFE_EditorToolbarItem::XFE_EditorToolbarItem" NEWLINE matches the argument list NEWLINE argument types are: (XFE_Component *, long) NEWLINE XFE_EditorToolbarItem(tb, NULL) NEWLINE ------------------------------^ NEWLINE cxx: Error: EditorToolbar.cpp, line 555: no instance of NEWLINE constructor "XFE_CommandInfo::XFE_CommandInfo" matches the NEWLINE argument list NEWLINE argument types are: (XFE_CommandEventType, NEWLINE Widget, long, String [2], int) NEWLINE XFE_CommandInfo e_info(XFE_COMMAND_BUTTON_ACTIVATE, NEWLINE -------------------------------^ NEWLINE cxx: Error: EditorToolbar.cpp, line 813: no instance of NEWLINE constructor "XFE_CommandInfo::XFE_CommandInfo" matches the NEWLINE argument list NEWLINE argument types are: (XFE_CommandEventType, NEWLINE Widget, long, String [2], int) NEWLINE XFE_CommandInfo e_info(XFE_COMMAND_BUTTON_ACTIVATE, NEWLINE -------------------------------^ NEWLINE 3 errors detected in the compilation of "EditorToolbar.cpp". NEWLINE NEWLINE NEWLINE the fixes are at line 375 NEWLINE XFE_EditorToolbarItem(tb, (ToolbarSpec*) NULL) NEWLINE NEWLINE line 557 NEWLINE (XEvent*) NULL, NEWLINE NEWLINE line 813 NEWLINE (XEvent*) NULL, NEWLINE NEWLINE berkley
326|Suggestions for Makefile on DUnix|Created by Berkley Shands (berkley@cs.wustl.edu) on Wednesday, May 6, 1998 12:05:08 PM PDT NEWLINE Additional Details : NEWLINE A suggestion for later model DEC alpha (OSF1) Dec Unix NEWLINE platforms; NEWLINE Don't use GCC/G++/GNU - it leads to code problems since that NEWLINE compiler is not NEWLINE very picky about ANSI C++ standards on callings and NEWLINE arguments. NEWLINE I have made a few ports to this 64-bit platform, here are a NEWLINE few useful things I found NEWLINE NEWLINE /* NEWLINE +========================================================================+ NEWLINE */ NEWLINE /* \| Playground, Copyright 1996 Washington University All NEWLINE Rights Reserved. \| */ NEWLINE /* NEWLINE \|------------------------------------------------------------------------\| NEWLINE */ NEWLINE /* \| FILE: CompilerDefs.hh */ NEWLINE /* NEWLINE \|------------------------------------------------------------------------\| NEWLINE */ NEWLINE /* \| Who \| Date \| Implementation Modification */ NEWLINE /* NEWLINE \|------------------------------------------------------------------------\| NEWLINE */ NEWLINE /* \| jyp \| Feb 01, 97 \| Initial Code Generation */ NEWLINE /* NEWLINE +========================================================================+ NEWLINE */ NEWLINE NEWLINE #ifndef CompilerDefs_h_ NEWLINE #define CompilerDefs_h_ NEWLINE /* */ NEWLINE /* list of weird definitions and other tricks needed to NEWLINE work around */ NEWLINE /* various compilers' idiosyncrasies */ NEWLINE /* */ NEWLINE NEWLINE /* The following pre-defined symbols are provided by the NEWLINE compiler */ NEWLINE NEWLINE /* _MSC_VER - Visual C++ version ID 4.2E */ NEWLINE /* __BCPLUSPLUS__ - Borland C++ V5.0 */ NEWLINE /* __SUNPRO_CC - Sun Sparcworks Pro V3.1 */ NEWLINE /* __DECCXX - Digital CXX V5.6 */ NEWLINE /* __DECCXX_VER = 60030001 first F/T of V6.0 */ NEWLINE /* __DECCXX_VER = 50660098 -oldcxx (V5.6-098) */ NEWLINE /* __UNIX All flavors of real unix */ NEWLINE /* WIN32 Both flavors of Intel/MS NT4.0 */ NEWLINE /* __GNUG__ GNU G++ */ NEWLINE /* __NetBSD__ NetBSD unix */ NEWLINE /* __i386 used with NetBSD for Intel NEWLINE platforms */ NEWLINE /* __mips64 SGI 7.x Compilers (not pre-defined) NEWLINE */ NEWLINE /* __mips SGI systems (GNU) */ NEWLINE /* __linux__ Linux running on an Intel Platform NEWLINE */ NEWLINE NEWLINE /* Microsft C++ doesn't like throw() in template NEWLINE declarations! */ NEWLINE /* gcc is still not too good at dealing with exception NEWLINE signatures */ NEWLINE NEWLINE #if defined (_MSC_VER) \|\| defined (__GNUG__) \|\| defined NEWLINE (__BCPLUSPLUS__) NEWLINE # define NOTHROW NEWLINE #else NEWLINE # define NOTHROW throw() NEWLINE #endif NEWLINE NEWLINE /* dealing with naming conventions for include files */ NEWLINE /* */ NEWLINE NEWLINE #if defined (__cplusplus) NEWLINE #include <iostream.h> NEWLINE #else NEWLINE #include <stdio.h> NEWLINE #endif NEWLINE NEWLINE /* after iostream, all hell breaks loose with g++ NEWLINE otherwise! */ NEWLINE NEWLINE #include "bool.h" NEWLINE NEWLINE /* dealing with various representations of 64 bit longs */ NEWLINE /* and 32 bit integers to make sure all use the same amount NEWLINE of storage */ NEWLINE NEWLINE #if defined(__i386) \|\| defined(__i486) \|\| defined(__i586) \|\| NEWLINE defined(__i686) NEWLINE # define IS_X86_BOX 1 NEWLINE #endif NEWLINE NEWLINE #if defined (__alpha) \|\| defined (__LP64) \|\| defined NEWLINE (__mips64) NEWLINE # define PGLONG64 long NEWLINE # define PGINT32 int NEWLINE # define PGUINT32 unsigned int NEWLINE #elif defined (__sparc) \|\| defined (__mips) NEWLINE # define PGLONG64 long long NEWLINE # define PGINT32 long NEWLINE # define PGUINT32 unsigned long NEWLINE #elif defined(IS_X86_BOX) && (defined (__NetBSD__) \|\| NEWLINE defined(__linux__)) NEWLINE # define PGLONG64 long long NEWLINE # define PGINT32 long NEWLINE # define PGUINT32 unsigned long NEWLINE NEWLINE /* defined in veneer/MissingTmpl.cxx */ NEWLINE ostream & operator << (ostream &, const PGLONG64 &); NEWLINE NEWLINE #elif defined (_M_IX86) NEWLINE # if defined (_MSC_VER) NEWLINE # define PGLONG64 __int64 NEWLINE # else NEWLINE # define PGLONG64 long NEWLINE # endif NEWLINE # define PGINT32 long NEWLINE # define PGUINT32 unsigned long NEWLINE NEWLINE /* defined in veneer/MissingTmpl.cxx */ NEWLINE ostream & operator << (ostream &, const PGLONG64 &); NEWLINE NEWLINE #else NEWLINE # error "CompilerDefs.hh; unknown compilation machine" NEWLINE #endif NEWLINE NEWLINE /* for gcc on Ix86 boxes, the << operator can't get the cast NEWLINE from */ NEWLINE /* PGint to PGINT64 right, so we must explicitly cast it. */ NEWLINE NEWLINE #if defined(__NetBSD__) \|\| defined(__linux__) NEWLINE #define INT64FIX (PGLONG64) NEWLINE #else NEWLINE #define INT64FIX NEWLINE #endif NEWLINE NEWLINE #endif /* CompilerDefs_h_ */ NEWLINE NEWLINE You can case statement the various CXX compilers based in NEWLINE its version number NEWLINE (V5.X vs V6.X the ANSI compiler) NEWLINE NEWLINE A few make flags (I use GNU make) NEWLINE NEWLINE # 111-D statement is unreachable NEWLINE # 177-D variable "foo" was declared but never referenced NEWLINE # 193-D zero used for undefined preprocessing identifier NEWLINE # 236-D controlling expression is constant NEWLINE # 340-D value copied to temporary, reference to temporary NEWLINE used NEWLINE # 402-D class "Dispatcher" has no accessible constructor NEWLINE # 835-D function parameter "foo" was declared but never NEWLINE referenced NEWLINE NEWLINE DWARNINGS = -msg_display_number -msg_display_tag \\ NEWLINE -msg_disable 111,177,193,236,340,402,835 NEWLINE OWARNINGS = -msg_disable 111,177,193,236,340,402,835 NEWLINE NEWLINE MAINFLAGS = -ieee -verbose -use_ld_input -ttimestamp NEWLINE -nocompress NEWLINE #DEBUGFLAGS = -g -O0 -w0 $(DWARNINGS) -gall -gall_pattern NEWLINE PGint.cxx NEWLINE DEBUGFLAGS = -g -O0 -w0 $(DWARNINGS) NEWLINE OPTIMFLAGS = -O4 -w1 $(OWARNINGS) NEWLINE NEWLINE TMPLFLAGS = -ptr $(PGVENEERDIR)/Templates.DB -pts NEWLINE LIBFLAGS = -shared -o NEWLINE LINKFLAGS = $(CCFLAGS) NEWLINE NEWLINE # if cxx starts whining about code and data overlapping one NEWLINE another NEWLINE #LIBFLAGS = -shared -T 0x20000000 -D 0x40000000 -o NEWLINE NEWLINE #OTHEROBJS = $(PGVENEERDIR)/Templates.DB/*.$(OBJEXT) NEWLINE NEWLINE OTHEROBJS = NEWLINE NEWLINE the -verbose leads to the choices being displayed when an NEWLINE overload conflict is detected. NEWLINE NEWLINE Our software targets SlowLaris 2.5, Windows-NT 4.0 and 5.0, NEWLINE SGI IRIX 6.2/6.4, OSF1 4.X, netBSD, Linux NEWLINE (redhat). The compiler errors I see are very typical of the NEWLINE ANSI changes. NEWLINE NEWLINE overall, very nice stuff. You have a good product. NEWLINE NEWLINE berkley
327|testing|Created by Arun Ramachandran (arunr@netscape.com) on Wednesday, May 6, 1998 5:22:23 PM PDT NEWLINE Additional Details : NEWLINE testing please ignore
328|testing|Created by Arun Ramachandran (arunr@netscape.com) on Wednesday, May 6, 1998 5:23:01 PM PDT NEWLINE Additional Details : NEWLINE testing please ignore
329|testing again|Created by Arun Ramachandran (arunr@netscape.com) on Wednesday, May 6, 1998 5:24:41 PM PDT NEWLINE Additional Details : NEWLINE testing please ignore NEWLINE Updated by Arun Ramachandran (arunr@netscape.com) on Friday, May 15, 1998 5:02:20 PM PDT NEWLINE Additional Details : NEWLINE testing again NEWLINE Updated by Arun Ramachandran (arunr@netscape.com) on Friday, May 15, 1998 5:02:38 PM PDT NEWLINE Additional Details : NEWLINE update again -1 NEWLINE Updated by Arun Ramachandran (arunr@netscape.com) on Tuesday, May 19, 1998 2:17:04 AM PDT NEWLINE Additional Details : NEWLINE testig
330|Superfluous warnings and cleanup|Created by (null) on Wednesday, May 6, 1998 10:34:23 PM PDT NEWLINE Additional Details : NEWLINE Reported by Brian Grunkemeyer (bg2k@andrew.cmu.edu) NEWLINE NEWLINE Added a default case to a switch to handle some unhandled NEWLINE enum values, but I don't know if that ever comes up, and I NEWLINE don't know if it's a bad thing. So right now it just prints NEWLINE out that it fell through the switch at that file & line NEWLINE number. Maybe there should be an assertion saying someone NEWLINE called the function with the wrong parameters, if these are NEWLINE wrong. Have someone who understands that file check this NEWLINE and do something intelligent. NEWLINE NEWLINE Got rid of some unused variables as well. NEWLINE NEWLINE Apply this patch in modules/libpref/src/. NEWLINE NEWLINE ------------------------------------------------------- NEWLINE ----------------- NEWLINE Index: prefapi.c NEWLINE ======= NEWLINE ========================================================= NEWLINE === NEWLINE RCS file: /cvsroot/mozilla/modules/libpref/src/ NEWLINE prefapi.c,v NEWLINE retrieving revision 3.1 NEWLINE diff -u -r3.1 prefapi.c NEWLINE --- prefapi.c 1998/03/28 03:35:10 3.1 NEWLINE +++ prefapi.c 1998/05/03 00:27:26 NEWLINE @@ -1454,6 +1454,10 @@ NEWLINE case PREF_BOOL: NEWLINE changed = oldValue.boolVal != NEWLINE newValue.boolVal; NEWLINE break; NEWLINE + default: NEWLINE + /* This doesn't handle NEWLINE PREF_LOCKED, PREF_USERSET, NEWLINE + PREF_CONFIG, nor PREF_LILOCAL. NEWLINE Is this bad? */ NEWLINE + fprintf(stderr, "Fell through NEWLINE case in %s at line %d. Is this bad?\\n", NEWLINE __FILE__, __LINE__); NEWLINE } NEWLINE return changed; NEWLINE } NEWLINE @@ -1655,9 +1659,8 @@ NEWLINE JSBool PR_CALLBACK pref_NativeGetPref NEWLINE (JSContext *cx, JSObject *obj, unsigned int argc, NEWLINE jsval *argv, jsval *rval) NEWLINE { NEWLINE - void* value = NULL; NEWLINE PrefNode* pref; NEWLINE - XP_Bool prefExists = TRUE; NEWLINE + /* Unused: XP_Bool prefExists = TRUE; */ NEWLINE NEWLINE if (argc >= 1 && JSVAL_IS_STRING(argv[0])) NEWLINE { NEWLINE @@ -2042,7 +2045,6 @@ NEWLINE { NEWLINE char *last; NEWLINE NEWLINE - int i, j, k, n; NEWLINE const char *s, *t; NEWLINE NEWLINE last = PR_sprintf_append(0, "An error occurred NEWLINE reading the startup NEWLINE configuration file. "
331|Superfluous warnings and cleanup|Created by (aoki@netscape.com) on Wednesday, May 6, 1998 10:34:23 PM PDT NEWLINE Additional Details : NEWLINE Reported by Brian Grunkemeyer (bg2k@andrew.cmu.edu) NEWLINE NEWLINE Added a default case to a switch to handle some unhandled NEWLINE enum values, but I don't know if that ever comes up, and I NEWLINE don't know if it's a bad thing. So right now it just prints NEWLINE out that it fell through the switch at that file & line NEWLINE number. Maybe there should be an assertion saying someone NEWLINE called the function with the wrong parameters, if these are NEWLINE wrong. Have someone who understands that file check this NEWLINE and do something intelligent. NEWLINE NEWLINE Got rid of some unused variables as well. NEWLINE NEWLINE Apply this patch in modules/libpref/src/. NEWLINE NEWLINE ------------------------------------------------------- NEWLINE ----------------- NEWLINE Index: prefapi.c NEWLINE ======= NEWLINE ========================================================= NEWLINE === NEWLINE RCS file: /cvsroot/mozilla/modules/libpref/src/ NEWLINE prefapi.c,v NEWLINE retrieving revision 3.1 NEWLINE diff -u -r3.1 prefapi.c NEWLINE --- prefapi.c 1998/03/28 03:35:10 3.1 NEWLINE +++ prefapi.c 1998/05/03 00:27:26 NEWLINE @@ -1454,6 +1454,10 @@ NEWLINE case PREF_BOOL: NEWLINE changed = oldValue.boolVal != NEWLINE newValue.boolVal; NEWLINE break; NEWLINE + default: NEWLINE + /* This doesn't handle NEWLINE PREF_LOCKED, PREF_USERSET, NEWLINE + PREF_CONFIG, nor PREF_LILOCAL. NEWLINE Is this bad? */ NEWLINE + fprintf(stderr, "Fell through NEWLINE case in %s at line %d. Is this bad?\\n", NEWLINE __FILE__, __LINE__); NEWLINE } NEWLINE return changed; NEWLINE } NEWLINE @@ -1655,9 +1659,8 @@ NEWLINE JSBool PR_CALLBACK pref_NativeGetPref NEWLINE (JSContext *cx, JSObject *obj, unsigned int argc, NEWLINE jsval *argv, jsval *rval) NEWLINE { NEWLINE - void* value = NULL; NEWLINE PrefNode* pref; NEWLINE - XP_Bool prefExists = TRUE; NEWLINE + /* Unused: XP_Bool prefExists = TRUE; */ NEWLINE NEWLINE if (argc >= 1 && JSVAL_IS_STRING(argv[0])) NEWLINE { NEWLINE @@ -2042,7 +2045,6 @@ NEWLINE { NEWLINE char *last; NEWLINE NEWLINE - int i, j, k, n; NEWLINE const char *s, *t; NEWLINE NEWLINE last = PR_sprintf_append(0, "An error occurred NEWLINE reading the startup NEWLINE configuration file. " NEWLINE NEWLINE NEWLINE NEWLINE NEWLINE Updated by (aoki@netscape.com) on Monday, August 17, 1998 9:35:11 AM PDT NEWLINE Additional Details : NEWLINE Duplicate of bug #330
332|Superfluous warnings and code cleanup|Created by (aoki@netscape.com) on Wednesday, May 6, 1998 10:42:10 PM PDT NEWLINE Additional Details : NEWLINE Reported by Grian Grunkemeyer (bg2k@andrew.cmu.edu) NEWLINE [comments from his patch appended below] NEWLINE NEWLINE Added a default case to switch to handle some unhandled enum NEWLINE values, but I don't know if that ever comes up, and I don't NEWLINE know if it's a bad thing. So right now it just prints out NEWLINE that it fell through teh switch at that file & line number. NEWLINE Maybe there should be an assertion saying someone called the NEWLINE function with the wrong parameters, if these are wrong. Have NEWLINE someone who understands that file check this and do NEWLINE something intelligent. NEWLINE NEWLINE Got rid of some unused variables as well. NEWLINE NEWLINE [patch submitted to newsgroup]. NEWLINE Updated by (aoki@netscape.com) on Monday, August 17, 1998 9:34:42 AM PDT NEWLINE Additional Details : NEWLINE Duplicate of bug #330
333|Floating point exeption, core dump|Created by Jörgen Jonsson (jorgen@saitek.se) on Thursday, May 7, 1998 5:01:39 AM PDT NEWLINE Additional Details : NEWLINE I compiled the code from 1998-04-29 and when I try to start NEWLINE mozilla-export it just core dumps. NEWLINE Jörgen Jonsson NEWLINE Saitek AB NEWLINE Sollentuna NEWLINE Sweden NEWLINE jorgen@saitek.se NEWLINE Updated by Wan-Teh Chang (wtc@netscape.com) on Thursday, May 7, 1998 11:08:43 PM PDT NEWLINE Additional Details : NEWLINE This may be due to the SIGFPE signal generated by jspai.c. NEWLINE The Javascript people are aware of the SIGFPE problem. NEWLINE There is a SIGFPE handler in mozilla/nsprpub/pr/src/md/unix/hpux.c, NEWLINE but apparently it doesn't work sometimes. NEWLINE NEWLINE Added myself to the CC list. NEWLINE Updated by Brian Ostrom (briano@netscape.com) on Monday, July 13, 1998 3:25:08 PM PDT NEWLINE Additional Details : NEWLINE Is this still a problem? If so, can you give me a traceback NEWLINE from gdb (or whatever)? Then I can forward this bug to the NEWLINE right person. Thanks.
334|ANSI disallows enum casts - make explicit in EditorView.cpp|Created by Berkley Shands (berkley@cs.wustl.edu) on Thursday, May 7, 1998 11:15:42 AM PDT NEWLINE Additional Details : NEWLINE diff mozilla/cmd/xfe/src/EditorView.cpp NEWLINE /hog/sources/mozilla/19980429/mozilla/cmd/xfe/src/EditorView.cpp NEWLINE 786c786 NEWLINE < ED_FontSize size = NEWLINE (ED_FontSize)(int)set_font_size_params[i].data; NEWLINE --- NEWLINE > ED_FontSize size = NEWLINE (ED_FontSize)set_font_size_params[i].data; NEWLINE 811c811 NEWLINE < ED_FontSize match = NEWLINE (ED_FontSize)(int)set_font_size_params[i].data; NEWLINE --- NEWLINE > ED_FontSize match = NEWLINE (ED_FontSize)set_font_size_params[i].data; NEWLINE NEWLINE this is for cxx V6.0-20 NEWLINE berkley
335|ANSI disallows enum casts - make explicit in HistoryView.cpp|Created by Berkley Shands (berkley@cs.wustl.edu) on Thursday, May 7, 1998 11:17:21 AM PDT NEWLINE Additional Details : NEWLINE diff mozilla/cmd/xfe/src/HistoryView.cpp NEWLINE /hog/sources/mozilla/19980429/mozilla/cmd/xfe/src/HistoryView.cpp NEWLINE 274c274 NEWLINE < resort((enHistSortCol) (int)calldata, False); NEWLINE --- NEWLINE > resort((enHistSortCol)calldata, False); NEWLINE 279c279 NEWLINE < resort((enHistSortCol) (int)calldata, True); NEWLINE --- NEWLINE > resort((enHistSortCol)calldata, True); NEWLINE 427c427 NEWLINE < return m_sortBy == (enHistSortCol) (int)calldata; NEWLINE --- NEWLINE > return m_sortBy == (enHistSortCol)calldata;
336|build changes successful OSF1 4.0D + CXX V6.0-20|Created by Berkley Shands (berkley@cs.wustl.edu) on Thursday, May 7, 1998 11:25:38 AM PDT NEWLINE Additional Details : NEWLINE This bug is being submitted using the optimized native NEWLINE version NEWLINE of Mozilla 5.0! A list of all altered files... NEWLINE ./mozilla/cmd/xfe/src/MozillaApp.cpp NEWLINE ./mozilla/cmd/xfe/src/HistoryFrame.cpp NEWLINE ./mozilla/cmd/xfe/src/Frame.cpp NEWLINE ./mozilla/cmd/xfe/src/NavCenterFrame.cpp NEWLINE ./mozilla/cmd/xfe/src/EditorFrame.cpp NEWLINE ./mozilla/cmd/xfe/src/HistoryView.cpp * NEWLINE ./mozilla/cmd/xfe/src/Menu.cpp NEWLINE ./mozilla/cmd/xfe/src/EditorToolbar.cpp NEWLINE ./mozilla/cmd/xfe/src/NavCenterView.cpp NEWLINE ./mozilla/cmd/xfe/src/PopupMenu.cpp NEWLINE ./mozilla/cmd/xfe/src/EditorView.cpp * NEWLINE ./mozilla/cmd/xfe/src/URLBar.cpp NEWLINE ./mozilla/cmd/xfe/src/HTMLDialogFrame.cpp NEWLINE ./mozilla/cmd/xfe/cxxlink-driver.sh NEWLINE ./mozilla/nsprpub/pr/include/md/_osf1.cfg NEWLINE ./mozilla/nsprpub/config/OSF1.mk NEWLINE ./mozilla/nsprpub/config/UNIX.mk NEWLINE ./mozilla/config/OSF1.mk NEWLINE NEWLINE I should have submitted a change for each file, all explicit NEWLINE type casts, NEWLINE except for the OSF1.mk and cxxlink-driver.sh NEWLINE NEWLINE In the configs, I removed "-taso" everywhere (not needed, NEWLINE and it makes things slower) NEWLINE NEWLINE diff mozilla/cmd/xfe/cxxlink-driver.sh NEWLINE $X/mozilla/cmd/xfe/cxxlink-driver.sh NEWLINE 79,82c79 NEWLINE < # HEAD="$LD_DIR/ld -G 8 -g2 -call_shared -nocount NEWLINE $LD_DIR/crt0.o \\ NEWLINE < # $CXX_DIR/_main.o -count -taso" NEWLINE < HEAD="$LD_DIR/ld -G 8 -g2 -call_shared -nocount NEWLINE $LD_DIR/crt0.o \\ NEWLINE < $CXX_DIR/_main.o -count " NEWLINE --- NEWLINE > HEAD="$LD_DIR/ld -G 8 -g2 -call_shared -nocount NEWLINE $LD_DIR/crt0.o $CXX_DIR/_main.o -count -taso" NEWLINE NEWLINE diff mozilla/nsprpub/config/OSF1.mk NEWLINE $X/mozilla/nsprpub/config/OSF1.mk NEWLINE 64c64 NEWLINE < CC = cc $(NON_LD_FLAGS) -std1 -w1 NEWLINE -readonly_strings NEWLINE --- NEWLINE > CC = cc $(NON_LD_FLAGS) -std1 -w0 NEWLINE -readonly_strings NEWLINE 66,70c66,67 NEWLINE < #CCC = cxx NEWLINE < CCC = cxx -ieee_with_inexact -x cxx -w1 NEWLINE -verbose \\ NEWLINE < -use_ld_input -ttimestamp NEWLINE -nocompress \\ NEWLINE < -msg_disable NEWLINE 111,177,193,236,340,402,610,611,835,846 \\ NEWLINE < -msg_display_number NEWLINE --- NEWLINE > #CCC = cxx -w0 NEWLINE > CCC = cxx -verbose NEWLINE 84c81 NEWLINE < OS_CFLAGS = -DOSF1 -D_REENTRANT -w1 NEWLINE --- NEWLINE > OS_CFLAGS = -DOSF1 -D_REENTRANT -w0 NEWLINE NEWLINE diff mozilla/config/OSF1.mk $X/mozilla/config/OSF1.mk NEWLINE 84c84 NEWLINE < CC = cc -ieee_with_inexact -std1 -w1 NEWLINE --- NEWLINE > CC = cc -ieee_with_inexact -std1 -w0 NEWLINE 86,89c86 NEWLINE < CCC = cxx -ieee_with_inexact -x cxx -w1 NEWLINE -verbose \\ NEWLINE < -use_ld_input -ttimestamp NEWLINE -nocompress \\ NEWLINE < -msg_disable NEWLINE 111,177,193,236,340,402,610,611,835,846 \\ NEWLINE < -msg_display_number NEWLINE --- NEWLINE > CCC = cxx -ieee_with_inexact -x cxx -w0 NEWLINE -verbose -use_ld_input -ttimestamp -nocompress NEWLINE NEWLINE this copy works MUCH better than the released 4.0.5 NEWLINE navigator, no java errors, NEWLINE no .so count errors, etc. NEWLINE NEWLINE I used the 19980429 sources (your web pages doesn't list NEWLINE this one :-) NEWLINE NEWLINE berkley NEWLINE NEWLINE Updated by Wan-Teh Chang (wtc@netscape.com) on Thursday, May 7, 1998 11:32:48 PM PDT NEWLINE Additional Details : NEWLINE The -taso linker option (force to use only the lower 32 bits NEWLINE of the address space) is necessary to build a complete Netscape NEWLINE Communicator because Sun's Java code stores a pointer in a 32-bit NEWLINE int. Sigh. NEWLINE NEWLINE For Mozilla, which doesn't have Sun's Java code, -taso is not NEWLINE necessary. So I think it is fine to remove -taso from Mozilla's NEWLINE cxxlink-driver.sh. Brian, we still need -taso in Netscape's NEWLINE commercial build as long as we are using Sun Java. NEWLINE NEWLINE By the way, the reason Mozilla doesn't have Java errors is NEWLINE that it doesn't have Java :-) NEWLINE Updated by Berkley Shands (berkley@cs.wustl.edu) on Wednesday, May 13, 1998 6:00:26 AM PDT NEWLINE Additional Details : NEWLINE You can mix -taso and 64 bit code, you just cannot pass pointers around :-)
337|Not displaying text properly|Created by Chris Hodson (cmh@fpk.hp.com) on Thursday, May 7, 1998 1:09:13 PM PDT NEWLINE Additional Details : NEWLINE On a significant portion of all links on all pages, the NEWLINE underlines for links appears over the word, or, sometimes, NEWLINE through the middle of the word. NEWLINE NEWLINE Note that I am running HPUX 11.0, which uses Motif2.1, if NEWLINE that is important. NEWLINE Updated by Chris Hodson (cmh@fpk.hp.com) on Wednesday, May 13, 1998 7:57:20 AM PDT NEWLINE Additional Details : NEWLINE Also, it seems that a number of text objects(some links, some not) are not NEWLINE being displayed at all. This is more that the annoyance it first seemed.
338|Unable to copy to clipboard|Created by (bob@guiduck.com) on Thursday, May 7, 1998 3:54:51 PM PDT NEWLINE Additional Details : NEWLINE When a page is first opened unable to highlight with mouse NEWLINE text to be copied to clipboard.. However if you expand the NEWLINE navigation center then collapse tree by clicking any button NEWLINE twice you will be able to highlight text. Another variation NEWLINE of the same problem is when a page if first opened using NEWLINE 'select all' NEWLINE option from edit menu causes mozilla to crash. However if NEWLINE you expand the navigation center then collapse tree by NEWLINE clicking any button you will be able 'select all' from edit NEWLINE menu. NEWLINE Updated by (blythe@netscape.com) on Tuesday, June 9, 1998 4:21:41 PM PDT NEWLINE Additional Details : NEWLINE David, bad interaction with Aurora NEWLINE Updated by David Hyatt (hyatt@netscape.com) on Tuesday, June 9, 1998 4:51:08 PM PDT NEWLINE Additional Details : NEWLINE Reassigning to Nisheeth. This has to do with resize, not Aurora.
339|Table captions not centered|Created by Brian Grunkemeyer (bg2k@cmu.edu) on Thursday, May 7, 1998 11:07:19 PM PDT NEWLINE Additional Details : NEWLINE Table captions should be centered to look nice. They're NEWLINE currently flush with the left-hand side of the table. NEWLINE NEWLINE For an example, look at: NEWLINE http://www.cs.cmu.edu/afs/andrew/usr/bg2k/www/Contact.html
340|Bugzilla text missing in Mozilla|Created by Brian Grunkemeyer (bg2k@cmu.edu) on Friday, May 8, 1998 12:09:42 AM PDT NEWLINE Additional Details : NEWLINE After posting Bug 339 w/ Mozilla, I reviewed the bug. The NEWLINE following lines marked w/ a * were missing from the bug NEWLINE report: NEWLINE NEWLINE Created by Brian Grunkemeyer (bg2k@cmu.edu) on Thursday, May NEWLINE 7, 1998 11:07:19 PM PDT NEWLINE * Additional Details : NEWLINE * Table captions should be centered to look nice. They're NEWLINE * currently flush with the left-hand side of the table. NEWLINE * NEWLINE * For an example, look at: NEWLINE http://www.cs.cmu.edu/afs/andrew/usr/bg2k/www/Contact.html NEWLINE NEWLINE It displays correctly in Communicator 4.04. I don't know if NEWLINE this is a bug in Mozilla, Bugzilla, or both. But a browser NEWLINE that doesn't display text right is definitely wrong.
341|submit results returned to wrong screen|Created by Chris Hodson (cmh@fpk.hp.com) on Friday, May 8, 1998 1:20:20 PM PDT NEWLINE Additional Details : NEWLINE After opening a second window to edit a bug on bugzilla and NEWLINE clicking submit, the results went to my first window, not NEWLINE the one I had clicked submit in. NEWLINE NEWLINE This is under HPUX 11 using Motif2.1 if it matters.
342|Browser crashes when user selects View->Page Source|Created by Carl Fongheiser (cmf@netins.net) on Friday, May 8, 1998 6:53:19 PM PDT NEWLINE Additional Details : NEWLINE In cmd:macfe:gui:CHTMLView.cp (in the routine CHTMLView::ObeyCommand) there is the NEWLINE following code: NEWLINE NEWLINE case cmd_ViewSource: NEWLINE { NEWLINE URL_Struct* url = NET_CreateURLStruct(mContext->GetCurrentURL(), NET_DONT_RELOAD); NEWLINE mContext->ImmediateLoadURL(url, FO_VIEW_SOURCE); NEWLINE NET_FreeURLStruct ( url ); NEWLINE cmdHandled = true; NEWLINE break; NEWLINE } NEWLINE NEWLINE In this case, url is getting freed prematurely, and the browser crashes when it tries to NEWLINE access a URL_Struct which is now filled with garbage. Commenting out the call to NEWLINE NET_FreeURLStruct() prevents the crash.
343|libreg/src/reg.c compile error|Created by (sjm@nwc.dk) on Saturday, May 9, 1998 2:29:37 AM PDT NEWLINE Additional Details : NEWLINE reg.c will not compile since "REG" is a reserved keyword in NEWLINE gcc-2.8+ hence REGOFF causes a parse error. NEWLINE Updated by (sjm@nwc.dk) on Saturday, May 9, 1998 2:35:41 AM PDT NEWLINE Additional Details : NEWLINE Actually, its "REGOFF" that causes the problem. If "REGOFF" is substituted with NEWLINE something else it compiles OK. NEWLINE Updated by (sjm@nwc.dk) on Saturday, May 9, 1998 2:42:12 AM PDT NEWLINE Additional Details : NEWLINE Platform=i586-pc-solaris2.6 (Solaris 2.6_X86) NEWLINE GCC 2.8.1 NEWLINE Updated by (sjm@nwc.dk) on Tuesday, May 12, 1998 12:22:55 AM PDT NEWLINE Additional Details : NEWLINE It is /usr/include/sys/reg.h on a Solaris2.6_X86 platform that contains "#define NEWLINE REGOFF MINFRAME". This define breaks reg.h and reg.c NEWLINE Updated by (sjm@nwc.dk) on Sunday, May 24, 1998 12:42:09 AM PDT NEWLINE Additional Details : NEWLINE This bug only applies to SunOS5.6_x86. NEWLINE There is a patch for this problem. Waiting for wtc@netscape.com to approve. NEWLINE Updated by Soren Juul Moller (sjm@nwc.dk) on Thursday, July 9, 1998 10:53:43 AM PDT NEWLINE Additional Details : NEWLINE Please refer to NEWLINE ftp://ftp.nwc.dk/pub/patches/mozilla-SunOS5.6_x86-CVS1998JUN09.txt NEWLINE Updated by Soren Juul Moller (sjm@nwc.dk) on Tuesday, July 14, 1998 11:29:51 AM PDT NEWLINE Additional Details : NEWLINE The patch URL is now NEWLINE ftp://ftp.nwc.dk/pub/patches/mozilla/SunOS5.6_x86-current.txt
344|Cached sockets closing unexpectedly cause 0-length error|Created by (tenthumbs@cybernex.net) on Saturday, May 9, 1998 10:12:47 AM PDT NEWLINE Additional Details : NEWLINE If the browser reuses a cached socket and the socket closes NEWLINE unexpectedly, net_parse_first_http_line returns the NEWLINE "Document contains no data" error. NEWLINE NEWLINE If the browser knows how long the socket will remain NEWLINE available, it should respect that. If the socket doesn't NEWLINE know, it should not just give up but rather try a news NEWLINE socket. NEWLINE NEWLINE I have more data available.
345|mozilla source fails to compile in Xfe, missing XfeBm directory|Created by Simon. Laule (laule@mbi-berlin.de) on Sunday, May 10, 1998 12:37:58 PM PDT NEWLINE Additional Details : NEWLINE Based on source updated today mozilla fails to compile NEWLINE on Solaris 2.6: NEWLINE NEWLINE ... NEWLINE Menu.cpp:48: Xfe/BmButton.h: No such file or directory NEWLINE Menu.cpp:49: Xfe/BmCascade.h: No such file or directory NEWLINE gmake[3]: *** [SunOS5.6_sparc_DBG.OBJ/Menu.o] Error 1 NEWLINE gmake[3]: Target `install' not remade because of errors. NEWLINE gmake[3]: Leaving directory NEWLINE `/users/z/laule/mozilla/mozilla/cmd/xfe/src' NEWLINE gmake[2]: *** [install] Error 2 NEWLINE cd XfeWidgets/XfeBm; gmake NEWLINE gmake[3]: Entering directory NEWLINE `/users/z/laule/mozilla/mozilla/cmd/xfe/XfeWidgets/XfeBm' NEWLINE gmake[3]: *** No targets specified and no makefile found. NEWLINE Stop. NEWLINE gmake[3]: Leaving directory NEWLINE `/users/z/laule/mozilla/mozilla/cmd/xfe/XfeWidgets/XfeBm' NEWLINE gmake[2]: *** NEWLINE [../../dist/SunOS5.6_sparc_DBG.OBJ/lib/libXfeBm.a] Error 2 NEWLINE gmake[2]: Target `install' not remade because of errors. NEWLINE gmake[2]: Leaving directory NEWLINE `/users/z/laule/mozilla/mozilla/cmd/xfe' NEWLINE gmake[1]: *** [install] Error 2 NEWLINE gmake[1]: Leaving directory NEWLINE `/users/z/laule/mozilla/mozilla/cmd' NEWLINE gmake: *** [install] Error 2 NEWLINE gmake: Target `all' not remade because of errors. NEWLINE mail laule 52 (.../mozilla/mozilla):
346|Missing "&" in parameter, cmd/xfe/forms.c:2116|Created by Ronan Waide (waider@waider.ie) on Monday, May 11, 1998 10:03:09 AM PDT NEWLINE Additional Details : NEWLINE Hiho, NEWLINE NEWLINE this fixes a warning and a bug of unknown severity: NEWLINE NEWLINE diff -c -r3.2 forms.c NEWLINE *** forms.c 1998/05/01 21:40:36 3.2 NEWLINE --- forms.c 1998/05/11 17:03:05 NEWLINE *************** NEWLINE *** 2100,2106 **** NEWLINE text_attr->charset, NEWLINE fe_font, NEWLINE type, NEWLINE ! fontlist); NEWLINE } NEWLINE else NEWLINE { NEWLINE --- 2113,2119 ---- NEWLINE text_attr->charset, NEWLINE fe_font, NEWLINE type, NEWLINE ! &fontlist); NEWLINE Updated by (toshok@netscape.com) on Monday, May 18, 1998 11:29:41 AM PDT NEWLINE Additional Details : NEWLINE I just checked in your fix for this bug, as well as bug 134. Thanks alot for NEWLINE the fixes.
347|scroll bar bug(no down)|Created by tongse Jung (tongse@tcom.tit.ac.kr) on Monday, May 11, 1998 1:00:39 PM PDT NEWLINE Additional Details : NEWLINE Memory read NEWLINE Updated by (blythe@netscape.com) on Tuesday, June 9, 1998 4:22:29 PM PDT NEWLINE Additional Details : NEWLINE Bill, this is a duplicate of a bug I already handed over to you.
348|History display in Aurora appears to be bogus|Created by Ronan Waide (waider@waider.ie) on Tuesday, May 12, 1998 8:09:58 AM PDT NEWLINE Additional Details : NEWLINE History -> By Date seems to have totally scrambled dates. NEWLINE This may well be contributed to by the fact that I NEWLINE originally copied my entire .netscape directory into NEWLINE mozilla's feeding ground, however. NEWLINE NEWLINE History -> By Site only has one page entry per site. NEWLINE NEWLINE I'm looking at the date one at the moment. NEWLINE Updated by Ronan Waide (waider@waider.ie) on Tuesday, May 12, 1998 9:37:08 AM PDT NEWLINE Additional Details : NEWLINE Date munging appears to be a big-endian/little-endian problem - the datablock NEWLINE retrieved from the history db file is in sparc's natural byte order, but the NEWLINE COPY_INT32 seems to be expecting the opposite. Not really sure why thisis NEWLINE happening, though. NEWLINE Updated by Ronan Waide (waider@waider.ie) on Tuesday, May 12, 1998 10:22:05 AM PDT NEWLINE Additional Details : NEWLINE Okay, it looks like the problem is that Netscape 4.0 has the date field arranged NEWLINE one way, and RDF is expecting it to be the other. Aurora is opening NEWLINE ~/.netscape/history.db despite the fact that I'm using jwz's (HOME=~/.mozilla; NEWLINE mozilla) hack to avoid breaking my Netscape 4 stuff. NEWLINE NEWLINE The one page per site seems to have been fixed in the meantime. I think I NEWLINE changed a type somewhere that might have fixed this. diff will sort it out. NEWLINE Updated by Ronan Waide (waider@waider.ie) on Tuesday, May 19, 1998 2:12:22 AM PDT NEWLINE Additional Details : NEWLINE Duh. It was looking in my ~/.netscape dir because prefs.js points there. Don't NEWLINE forget to change prefs.js, boys and girls! NEWLINE Updated by Ronan Waide (waider@waider.ie) on Thursday, May 21, 1998 2:39:20 AM PDT NEWLINE Additional Details : NEWLINE Problem found. glhist.c defines COPY_INT32 as byteswapping if you're a NEWLINE macintosh, while hist2rdf.h defines COPY_INT32 as byteswapping if you're on on a NEWLINE BIG_ENDIAN machine. NEWLINE NEWLINE Here's a patch to get it working: NEWLINE Index: modules/rdf/src/hist2rdf.c NEWLINE =================================================================== NEWLINE RCS file: /cvsroot/mozilla/modules/rdf/src/hist2rdf.c,v NEWLINE retrieving revision 3.3 NEWLINE diff -c -r3.3 hist2rdf.c NEWLINE *** hist2rdf.c 1998/04/25 00:07:01 3.3 NEWLINE --- hist2rdf.c 1998/05/21 09:40:19 NEWLINE *************** NEWLINE *** 1,4 **** NEWLINE ! /* -*- Mode: C; tab-width: 4; indent-tabs-mode: nil; c-basic-offset: 2 -*- NEWLINE * NEWLINE * The contents of this file are subject to the Netscape Public License NEWLINE * Version 1.0 (the "NPL"); you may not use this file except in NEWLINE --- 1,4 ---- NEWLINE ! /* -*- Mode: C; tab-width: 8; indent-tabs-mode: nil; c-basic-offset: 2 -*- NEWLINE * NEWLINE * The contents of this file are subject to the Netscape Public License NEWLINE * Version 1.0 (the "NPL"); you may not use this file except in NEWLINE *************** NEWLINE *** 70,78 **** NEWLINE { NEWLINE HASHINFO hash = { 4*1024, 0, 0, 0, 0, 0}; NEWLINE DBT key, data; NEWLINE ! time_t last,first,numaccess; NEWLINE PRBool firstOne = 0; NEWLINE ! DB* db = CallDBOpenUsingFileURL(gGlobalHistoryURL, O_RDONLY ,0600, NEWLINE DB_HASH, &hash); NEWLINE grdf = r; NEWLINE if (db != NULL) { NEWLINE --- 70,82 ---- NEWLINE { NEWLINE HASHINFO hash = { 4*1024, 0, 0, 0, 0, 0}; NEWLINE DBT key, data; NEWLINE ! time_t last,first; NEWLINE ! uint32 numaccess; NEWLINE PRBool firstOne = 0; NEWLINE ! NEWLINE ! NEWLINE ! NEWLINE ! DB* db = CallDBOpenUsingFileURL(gGlobalHistoryURL, O_RDONLY, 0600, NEWLINE DB_HASH, &hash); NEWLINE grdf = r; NEWLINE if (db != NULL) { NEWLINE *************** NEWLINE *** 82,102 **** NEWLINE while (0 == (*db->seq)(db, &key, &data, (firstOne ? R_NEXT : R_FIRST))) NEWLINE { NEWLINE char* title = ((char*)data.data + 16); /* title */ NEWLINE char* url = (char*)key.data; /* url */ NEWLINE ! int32 flag = (int32)*((char*)data.data + 3*sizeof(int32)); NEWLINE firstOne = 1; NEWLINE #ifdef XP_UNIX NEWLINE if ((/*1 == flag &&*/ displayHistoryItem((char*)key.data))) { NEWLINE #else NEWLINE if (1 == flag && displayHistoryItem((char*)key.data)) { NEWLINE #endif NEWLINE ! COPY_INT32(&last, (time_t *)((char *)data.data)); NEWLINE ! COPY_INT32(&first, (time_t *)((char *)data.data + 4)); NEWLINE ! COPY_INT32(&numaccess, (time_t *)((char *)data.data + 8)); NEWLINE NEWLINE ! collateOneHist(r, u,url,title, last, first, numaccess, byDateFlag); NEWLINE } NEWLINE } NEWLINE ! (*db->close)(db); NEWLINE } NEWLINE } NEWLINE NEWLINE --- 86,109 ---- NEWLINE while (0 == (*db->seq)(db, &key, &data, (firstOne ? R_NEXT : R_FIRST))) NEWLINE { NEWLINE char* title = ((char*)data.data + 16); /* title */ NEWLINE char* url = (char*)key.data; /* url */ NEWLINE ! /* int32 flag = (int32)*((char*)data.data + 3*sizeof(int32));*/ NEWLINE ! int32 flag; NEWLINE ! HIST_COPY_INT32( &flag, (int8 *) data.data + 3 * sizeof( int32 )); NEWLINE firstOne = 1; NEWLINE #ifdef XP_UNIX NEWLINE if ((/*1 == flag &&*/ displayHistoryItem((char*)key.data))) { NEWLINE #else NEWLINE if (1 == flag && displayHistoryItem((char*)key.data)) { NEWLINE #endif NEWLINE ! NEWLINE ! HIST_COPY_INT32(&last, (int8 *)data.data ); NEWLINE ! HIST_COPY_INT32(&first, (int8 *)data.data + sizeof( int32 )); NEWLINE ! HIST_COPY_INT32(&numaccess, (int8 *)data.data + 2 *sizeof( int32 )); NEWLINE NEWLINE ! collateOneHist(r, u,url,title, last, first, numaccess, byDateFlag); NEWLINE } NEWLINE } NEWLINE ! (*db->close)(db); NEWLINE } NEWLINE } NEWLINE NEWLINE *************** NEWLINE *** 345,353 **** NEWLINE int32 flg = (int32)*((char*)data->data + 3*sizeof(int32)); NEWLINE if (!displayHistoryItem((char*)key->data)) return; NEWLINE if (grdf != NULL) { NEWLINE ! COPY_INT32(&last, (time_t *)((char *)data->data)); NEWLINE ! COPY_INT32(&first, (time_t *)((char *)data->data + 4)); NEWLINE ! COPY_INT32(&numaccess, (time_t *)((char *)data->data + 8)); NEWLINE NEWLINE if (hostHash) collateOneHist(grdf, gNavCenter->RDF_HistoryBySite, NEWLINE (char*)key->data, /* url */ NEWLINE --- 352,360 ---- NEWLINE int32 flg = (int32)*((char*)data->data + 3*sizeof(int32)); NEWLINE if (!displayHistoryItem((char*)key->data)) return; NEWLINE if (grdf != NULL) { NEWLINE ! HIST_COPY_INT32(&last, (time_t *)((char *)data->data)); NEWLINE ! HIST_COPY_INT32(&first, (time_t *)((char *)data->data + 4)); NEWLINE ! HIST_COPY_INT32(&numaccess, (time_t *)((char *)data->data + 8)); NEWLINE NEWLINE if (hostHash) collateOneHist(grdf, gNavCenter->RDF_HistoryBySite, NEWLINE (char*)key->data, /* url */ NEWLINE Index: modules/rdf/src/hist2rdf.h NEWLINE =================================================================== NEWLINE RCS file: /cvsroot/mozilla/modules/rdf/src/hist2rdf.h,v NEWLINE retrieving revision 3.3 NEWLINE diff -c -r3.3 hist2rdf.h NEWLINE *** hist2rdf.h 1998/04/25 00:07:03 3.3 NEWLINE --- hist2rdf.h 1998/05/21 09:40:19 NEWLINE *************** NEWLINE *** 43,48 **** NEWLINE --- 43,53 ---- NEWLINE #error Must have a byte order NEWLINE #endif NEWLINE NEWLINE + /* XXX glhist.c does this */ NEWLINE + #if !defined(XP_MAC) NEWLINE + #define HIST_COPY_INT32(_a,_b) XP_MEMCPY(_a, _b, sizeof(int32)); NEWLINE + #endif NEWLINE + NEWLINE #ifdef IS_LITTLE_ENDIAN NEWLINE #define COPY_INT32(_a,_b) XP_MEMCPY(_a, _b, sizeof(int32)); NEWLINE #else NEWLINE Updated by Robert Churchill (rjc@netscape.com) on Tuesday, May 26, 1998 6:36:48 PM PDT NEWLINE Additional Details : NEWLINE Fix from Ronan Waide (waider@waider.ie) checked into the RDF_BRANCH. NEWLINE Updated by Robert Churchill (rjc@netscape.com) on Tuesday, May 26, 1998 6:37:00 PM PDT NEWLINE Additional Details :
349|Segmentation fault if can't get current directory|Created by Michael Gordon (mfg@ee.ed.ac.uk) on Wednesday, May 13, 1998 6:15:17 AM PDT NEWLINE Additional Details : NEWLINE There's a segmentation fault in strlen() at line 4915 NEWLINE of cmd/xfe/mozilla.c if the previous getcwd() fails. NEWLINE Is setting 'string' to '.' if getcwd() fails acceptable?
350|PR_ExplodeTime() works only if given a PRTime argument between year 1901-2099|Created by Wan-Teh Chang (wtc@netscape.com) on Wednesday, May 13, 1998 6:20:07 PM PDT NEWLINE Additional Details : NEWLINE This problem is first reported by Christopher Shaulis NEWLINE <cjs2895@la-paz.org> in the netscape.public.mozilla.general NEWLINE newsgroup. He noticed a comment in NEWLINE mozilla/nsprpub/pr/src/misc/prtime.c saying that NEWLINE the leap-year calculation used in ComputeGMT() NEWLINE is only correct for 1901-2099. ComputeGMT() is NEWLINE an internal function used by PR_ExplodeTIme(). NEWLINE NEWLINE This means that PR_ExplodeTime returns an incorrect NEWLINE result if given a PRTime value that represents a NEWLINE date before 1900 or after 2100. NEWLINE Updated by Wan-Teh Chang (wtc@netscape.com) on Wednesday, May 13, 1998 6:21:33 PM PDT NEWLINE Additional Details : NEWLINE Marked the bug assigned with priority P5.
351|[PP]Syntax error in a header file when compiled in 68K mode|Created by Romain Vignes (rvignes@cal.fr) on Thursday, May 14, 1998 8:20:05 AM PDT NEWLINE Additional Details : NEWLINE In the file 'mozilla:nsprpub:pr:include:md:_macos.h', the NEWLINE following code (line 174) does not compile in 68K Mode : NEWLINE NEWLINE #define _MD_GET_PC(_t) (*((PRUint32 *)((_t)->>md.jb) + 6)) NEWLINE #define _MD_GET_SP(_t) (*((PRUint32 *)((_t)->>md.jb) + 12)) NEWLINE NEWLINE In fact, it is due to a syntax error in pointer NEWLINE dereferencing. The fixed code is below: NEWLINE NEWLINE #define _MD_GET_PC(_t) (*((PRUint32 *)((_t)->md.jb) + 6)) NEWLINE #define _MD_GET_SP(_t) (*((PRUint32 *)((_t)->md.jb) + 12)) NEWLINE NEWLINE Updated by Wan-Teh Chang (wtc@netscape.com) on Monday, June 8, 1998 6:13:51 PM PDT NEWLINE Additional Details : NEWLINE Fixed in mozilla/nsprpub/pr/include/md/_macos.h, revision 3.4. NEWLINE Thanks a lot for pointing out this error. NEWLINE NEWLINE Closed the bug.
352|Mozilla-CVS19980505-0513 (Linux,Glibc,Lesstif) crashes at the beginning|Created by Aleksey Nogin (ayn2@cornell.edu) on Thursday, May 14, 1998 10:51:17 AM PDT NEWLINE Additional Details : NEWLINE Hi, NEWLINE NEWLINE The new bug seem to be introduced after Mozilla-19980429. NEWLINE I've compiled the Mozilla from the public CVS server image NEWLINE 05/05 and 05/13 and both had the same problem - Mozilla NEWLINE segfaults after loading plugins and before opening any NEWLINE windows. NEWLINE NEWLINE On 05/13: NEWLINE NEWLINE (gdb) bt NEWLINE #0 0x845bdbb in _MD_WAIT (thread=0x50025ec0, NEWLINE ticks=4294967295) at linux.c:72 NEWLINE #1 0x844a2f5 in _PR_WaitCondVar (thread=0x50025ec0, NEWLINE cvar=0x885f820, lock=0x885f800, NEWLINE timeout=4294967295) at prucv.c:232 NEWLINE #2 0x844915c in PR_Wait (mon=0x885f7e0, ticks=4294967295) NEWLINE at prmon.c:155 NEWLINE #3 0x83e044a in et_SubEventLoop (qse=0x8853e60) at NEWLINE et_mocha.c:2255 NEWLINE #4 0x83e04f7 in lm_wait_for_events (pB=0x0) at NEWLINE et_mocha.c:2296 NEWLINE #5 0x844cf38 in _PR_UserRunThread () at pruthr.c:471 NEWLINE #6 0x844e77d in _PR_CreateThread (type=141101824, NEWLINE start=0xbfffe538 <_pr_pageSize+3080215300>, NEWLINE arg=0x4033eda0, priority=128, scope=1073783288, NEWLINE state=-1073743736, stackSize=134558480, NEWLINE flags=3221218640) at pruthr.c:1263 NEWLINE NEWLINE RedHat 5.0, Linux 2.0.32, glibc-2.0.7-7, gcc-2.7.2.3-8, NEWLINE lesstif-current of May 1. NEWLINE Updated by Aleksey Nogin (ayn2@cornell.edu) on Thursday, May 14, 1998 3:29:13 PM PDT NEWLINE Additional Details : NEWLINE When I downloaded and compiled the lastest (05/14) Mozilla update from the NEWLINE public CVS server and the lastest (05/14) lesstif-current, the problem NEWLINE disappeared.
353|Pressing ESC in font selection menu crashes Mozilla|Created by Aleksey Nogin (ayn2@cornell.edu) on Thursday, May 14, 1998 3:36:58 PM PDT NEWLINE Additional Details : NEWLINE Hi, NEWLINE NEWLINE When I go to Preferences\|Font menu, choose some encoding, NEWLINE click to open the list of availiable fonts and instead of NEWLINE selecting some font, press ESC, Mozilla segfaults (I tried NEWLINE it 3 times, it crashed all 3 times): NEWLINE NEWLINE (gdb) bt NEWLINE #0 0x40138780 in ?? () from /usr/X11R6/lib/libXt.so.6 NEWLINE #1 0x8677d1c in fe_FontCharSets () NEWLINE #2 0x817cb43 in _ComboBoxKbdCancel (w=0x8938400, NEWLINE event=0xbfffe504, params=0x0, NEWLINE num_params=0x8935868) at ComboBox.c:777 NEWLINE #3 0x4014b20f in ?? () from /usr/X11R6/lib/libXt.so.6 NEWLINE #4 0x4014b699 in ?? () from /usr/X11R6/lib/libXt.so.6 NEWLINE #5 0x4014bb90 in ?? () from /usr/X11R6/lib/libXt.so.6 NEWLINE #6 0x4012c70a in ?? () from /usr/X11R6/lib/libXt.so.6 NEWLINE #7 0x4012cfcf in ?? () from /usr/X11R6/lib/libXt.so.6 NEWLINE #8 0x4012d1ea in ?? () from /usr/X11R6/lib/libXt.so.6 NEWLINE #9 0x809ec47 in fe_EventLoop () at mozilla.c:1004 NEWLINE #10 0x80a1901 in main (argc=1, argv=0xbffff898) at NEWLINE mozilla.c:3373 NEWLINE NEWLINE Versions: NEWLINE Mozilla: CVS-19980514 (todays lates update from the public NEWLINE CVS server) NEWLINE Redhat 5.0, Linux 2.0.32, glibc-2.0.7-7, gcc-2.7.2.3-8, NEWLINE lesstif-current of May 14. NEWLINE Updated by Aleksey Nogin (ayn2@cornell.edu) on Sunday, August 16, 1998 10:30:33 PM PDT NEWLINE Additional Details : NEWLINE The problem still exists with NEWLINE Communicator CVS sources as of Sun Aug 16 NEWLINE RedHat Linux 5.1 - kernel 2.0.35 NEWLINE glibc-2.0.7-19 NEWLINE gcc-2.7.2.3-11 NEWLINE egcs-c++-1.0.2-8 NEWLINE lesstif-current-1998.08.14-1 NEWLINE NEWLINE I was told that this should be a Lesstif problem. NEWLINE Updated by Felix (thehelix@yahoo.com) on Tuesday, August 18, 1998 3:17:22 PM PDT NEWLINE Additional Details : NEWLINE I've hunted this bug for some hours today as well, using Purify. NEWLINE The weird thing about this is as follows: NEWLINE NEWLINE * It only happens in the DtComboBoxWidgets. These are only used in the NEWLINE preferences section, under Fonts. Pressing ESC while in the first ComboBox NEWLINE (labeled "For the Encoding:") works fine. Pressing ESC in *any* of the other NEWLINE ComboBoxes crashes Mozilla, and also Netscape Communicator 4.0.5 (haven't tried NEWLINE it with other versions, but I suspect it's the same). NEWLINE NEWLINE Could anyone help me out on this one? I'm not sure whether this would be a Motif NEWLINE problem, since Aleksey is using Lesstif. Perhaps something in the initialization NEWLINE is set wrong that causes all ComboBoxes after the first one to have this bug? NEWLINE NEWLINE I'm using Sun's C compiler under Solaris 2.5.1. I've managed to track down the NEWLINE following: NEWLINE NEWLINE cmd/xfe/DtWidgets/ComboBox.c NEWLINE NEWLINE (around line 780): NEWLINE NEWLINE /* NEWLINE * osfCancel virtual key hit. NEWLINE */ NEWLINE static void NEWLINE _ComboBoxKbdCancel(Widget w, NEWLINE XEvent *event, NEWLINE String *params, NEWLINE Cardinal *num_params) NEWLINE { NEWLINE DtComboBoxWidget combo; NEWLINE DtComboBoxPart *combo_p; NEWLINE XtPointer data; NEWLINE Arg args[1]; NEWLINE NEWLINE /* Get combo-box off list data */ NEWLINE XtSetArg(args[0], XmNuserData, &data); NEWLINE XtGetValues(w, args, 1); NEWLINE NEWLINE NEWLINE combo = (DtComboBoxWidget)data; NEWLINE NEWLINE /*********** combo is not getting assigned correctly here!!!!!! */ NEWLINE combo_p = (DtComboBoxPart*)&(combo->combo_box); NEWLINE NEWLINE combo_p->popped_up = FALSE; NEWLINE /* FELIX */ NEWLINE if (!(combo_p->shell)) { NEWLINE fprintf (stderr, "WARNING: SHELL IS ZERO!!!!\\n"); NEWLINE fprintf (stderr, "combo_p->type = %d\\n", combo_p->type); NEWLINE fflush (stderr); NEWLINE return; NEWLINE } NEWLINE /* END FELIX */ NEWLINE XtPopdown(combo_p->shell);
354|TEXTAREA form fields are missing|Created by Aleksey Nogin (ayn2@cornell.edu) on Thursday, May 14, 1998 3:46:01 PM PDT NEWLINE Additional Details : NEWLINE All form TEXTINPUT fields are missing. They were missing in NEWLINE Mozilla 19980429 and they are still missing in Mozilla NEWLINE CVS-19980514. NEWLINE NEWLINE RedHat 5.0, Linux 2.0.32, glibc-2.0.7-7, gcc-2.7.2.3-8, NEWLINE lesstiff-current of May 14. NEWLINE Updated by Aleksey Nogin (ayn2@cornell.edu) on Friday, June 19, 1998 10:31:29 AM PDT NEWLINE Additional Details : NEWLINE This was a Lesstif problem. It was fixed in the lastest version of Lesstif.
355|Tries to go through proxy before asking password|Created by Chris Hodson (cmh@fpk.hp.com) on Friday, May 15, 1998 7:23:06 AM PDT NEWLINE Additional Details : NEWLINE We have a proxy that requires a username and password. At NEWLINE startup, I get the 2 warnings saying: NEWLINE NEWLINE unable to locate rdf.netscape.com NEWLINE NEWLINE and NEWLINE NEWLINE warning the following hosts are unknown. blah blah NEWLINE perhaps there is a problem with your name server. NEWLINE NEWLINE As soon as I go to the first site through the proxy server, NEWLINE it asks for my name and password and everything is ok. NEWLINE NEWLINE Looks like its not doing the checks in the right order.
356|Missing -lresolv for SunOS5.5.1+|Created by (sjm@nwc.dk) on Friday, May 15, 1998 3:04:51 PM PDT NEWLINE Additional Details : NEWLINE Platform: SunOS5.6_x86 NEWLINE NEWLINE config/SunOS5.mk OS_LIBS laks -lresolv. "_res" and NEWLINE "res_init()" is undefined in libnetwork.so. Patch: NEWLINE NEWLINE RCS file: /cvsroot/mozilla/config/SunOS5.mk,v NEWLINE retrieving revision 3.1 NEWLINE diff -r3.1 SunOS5.mk NEWLINE 158c158 NEWLINE < OS_LIBS += -lsocket -lnsl -ldl NEWLINE --- NEWLINE > OS_LIBS += -lresolv -lsocket -lnsl -ldl NEWLINE Updated by Soren Juul Moller (sjm@nwc.dk) on Tuesday, June 9, 1998 1:52:34 PM PDT NEWLINE Additional Details : NEWLINE -lresolv is also needed on SunOS5.5.1_x86 NEWLINE Updated by Soren Juul Moller (sjm@nwc.dk) on Thursday, July 9, 1998 10:51:43 AM PDT NEWLINE Additional Details : NEWLINE Please refer to patch at NEWLINE ftp://ftp.nwc.dk/pub/patches/mozilla-SunOS5.6_x86-CVS1998JUN09.txt NEWLINE for at better solution. NEWLINE This fix is needed for sparc SunOS5.5.1, sparc SunOS5.6, x86 SunOS5.5.1 and x86 NEWLINE SunOS5.6. NEWLINE Updated by Brian Ostrom (briano@netscape.com) on Monday, July 13, 1998 5:13:35 PM PDT NEWLINE Additional Details : NEWLINE I think I've fixed the problem. If you pull the latest version of NEWLINE mozilla/config/SunOS5.mk, you should be happy. NEWLINE NEWLINE I couldn't get to your machine (ftp.nwc.dk) to look at the latest NEWLINE patch, so please let me know what I've missed. NEWLINE NEWLINE Thanks, NEWLINE briano.
357|test -1|Created by Arun Ramachandran (arunr@netscape.com) on Friday, May 15, 1998 5:02:59 PM PDT NEWLINE Additional Details : NEWLINE atestr again NEWLINE Updated by Arun Ramachandran (arunr@netscape.com) on Friday, May 15, 1998 5:03:21 PM PDT NEWLINE Additional Details :
358|testing - 1|Created by Arun Ramachandran (arunr@netscape.com) on Friday, May 15, 1998 5:35:44 PM PDT NEWLINE Additional Details : NEWLINE testing NEWLINE Updated by Arun Ramachandran (arunr@netscape.com) on Friday, May 15, 1998 5:36:12 PM PDT NEWLINE Additional Details : NEWLINE afdsfasdf
359|testing - 1|Created by Arun Ramachandran (arunr@netscape.com) on Saturday, May 16, 1998 3:59:00 PM PDT NEWLINE Additional Details : NEWLINE <b> testing NEWLINE <xmp> NEWLINE NEWLINE <table> NEWLINE <tr> NEWLINE <td>col1 NEWLINE <td>col2 NEWLINE </table> NEWLINE NEWLINE <xmp> NEWLINE Updated by Arun Ramachandran (arunr@netscape.com) on Saturday, May 16, 1998 3:59:46 PM PDT NEWLINE Additional Details : NEWLINE testing agian NEWLINE </xmp> NEWLINE testing agina NEWLINE Updated by Arun Ramachandran (arunr@netscape.com) on Monday, May 18, 1998 5:16:11 PM PDT NEWLINE Additional Details : NEWLINE testing again
360|Testing Bugzilla -> BugSplat stuff.|Created by Terry Weissman (terry@netscape.com) on Monday, May 18, 1998 9:02:03 AM PDT NEWLINE Additional Details : NEWLINE Like, does it work? NEWLINE NEWLINE Oh, and what happens if I say <PRE> here? And what's the NEWLINE handling of apostrophes these days? NEWLINE Updated by Terry Weissman (terry@netscape.com) on Monday, May 18, 1998 9:05:49 AM PDT NEWLINE Additional Details :
361|Fail to parse some PICS labels|Created by Paul Chek (pchek@ziplink.net) on Monday, May 18, 1998 12:39:24 PM PDT NEWLINE Additional Details : NEWLINE 1.The "extension" option must be the very first option in a NEWLINE PICS label for it to be parsed correctly. If it is preceded NEWLINE by any of the other label options then the parse for NEWLINE the extension and the label fails. I tracked this down to NEWLINE the fact NEWLINE that the single label object is closed when the NEWLINE processing for any option completes. The problem was in NEWLINE CSLABEL.C, about line 245 in the table definition for NEWLINE SingleLabel_stateTokens element "option value". The next NEWLINE to last specification containing the command was NEWLINE Command_CLOSE and it should be Command_NONE. NEWLINE 2.Multiple ratings in a single label are parsed without NEWLINE errors but the structures to iterate through the multiple NEWLINE single NEWLINE labels are not set up correctly. This prevents Mozilla NEWLINE from creating standalone labels from a single label NEWLINE containing multiple ratings. Creating standalone labels NEWLINE is needed for processing of signed labels. I found this in NEWLINE the C reference code, not the Mozilla code, but Mozilla NEWLINE probably has the problem also. It does not affect the trust NEWLINE label processing because that does not depend on the NEWLINE ability to iterate through the parsing structures. NEWLINE 8.If you have an extension option, the parser does not NEWLINE correctly finish the parse on the label when it sees the NEWLINE closing paren after ratings. If you remove the NEWLINE extension option the parser detects the closing paren NEWLINE correctly and NEWLINE returns _done status.
369|Testing please ignore|testing new field target fix version NEWLINE ------- Additional Comments From arunr 05/18/98 16:20 ------- NEWLINE updateing again
370|Cant parse PICS labels in HTTP headers - fix available|Created by Paul Chek (pchek@ziplink.net) on Monday, May 18, 1998 6:58:36 PM PDT NEWLINE Additional Details : NEWLINE NEWLINE CSParse_targetParser where I found: NEWLINE NEWLINE /* changed by montulli@netscape.com 11/29/97 NEWLINE NEWLINE * if (HTChunk_size(pCSParse->token)) { NEWLINE NEWLINE * HTChunk_terminate(pCSParse->token); NEWLINE NEWLINE * token = HTChunk_data(pCSParse->token); NEWLINE NEWLINE * } NEWLINE NEWLINE */ NEWLINE NEWLINE if(HTChunk_data(pCSParse->token)) NEWLINE NEWLINE { NEWLINE NEWLINE HTChunk_terminate(pCSParse->token); NEWLINE NEWLINE token = HTChunk_data(pCSParse->token); NEWLINE NEWLINE } NEWLINE NEWLINE At the closing paren for the label the call to NEWLINE HTChunk_data returns true and sets the token to an EMPTY NEWLINE string with a size of 1. So the code winds through the NEWLINE parsing again, gets confused and throws an error. At NEWLINE the same point HTChunk_size returns false, so a new NEWLINE EMPTY token is not set and it successfully parses the NEWLINE PICS label. After restoring the original code the PICS NEWLINE label parses correctly in the HTTP header.
371|dead code and erroneous code in PICS label processing|Created by Paul Chek (pchek@ziplink.net) on Monday, May 18, 1998 7:00:39 PM PDT NEWLINE Additional Details : NEWLINE In PICAPI.C, function target_callback "if(target == NEWLINE CSLLTC_SINGLE)" will never be executed because the NEWLINE variable cd is always 0 when. "(target == NEWLINE CSLLTC_SINGLE)". Furthermore, the test NEWLINE "if(lo->generic.state)" NEWLINE is incorrect, it should be ‘if(BVal_value(&lo->generic) NEWLINE )".
372|user pref callback not called while processing user.js|Created by Paul Chek (pchek@ziplink.net) on Monday, May 18, 1998 7:02:54 PM PDT NEWLINE Additional Details : NEWLINE In NET_RegisterCookiePrefCallbacks when PREF_GetCharPref is NEWLINE called to get the name of the Java script NEWLINE to use for filtering cookies NEWLINE "network.cookie.filterName" it returns NULL meaning it NEWLINE couldn’t find the NEWLINE preference. This is despite the fact that it is in NEWLINE prefs.js and liprefs.js was removed. When I looked into this NEWLINE I NEWLINE found that when the user preference scripts are NEWLINE processed by pref_OpenFile it calls NEWLINE PREF_EvaluateConfigScript with bCallbacks equal FALSE. NEWLINE This keeps NET_CookieBehaviorPrefChanged NEWLINE and NET_CookieScriptPrefChanged from being called to NEWLINE record the changed user preference. If I change NEWLINE bCallbacks to TRUE then the callback functions are NEWLINE called and preferences work as designed. The current NEWLINE work around is to call PREF_GetIntPref instead of NEWLINE relying on the saved state of the preference, which is set NEWLINE by the callback.
373|"Edit \| Select All" crashes application|Created by Rogelio Andreo (rba@murcia.net) on Tuesday, May 19, 1998 12:39:08 AM PDT NEWLINE Additional Details : NEWLINE Sometimes, doing a "Edit \| Select All" from the browser NEWLINE window crashes the application. You can bet it will crash if NEWLINE you try to select some text with the mouse but nothing NEWLINE gets selected. I haven't been able to find yet a pattern to NEWLINE consistently reproduce the bug. NEWLINE Updated by (blythe@netscape.com) on Tuesday, June 9, 1998 4:23:23 PM PDT NEWLINE Additional Details : NEWLINE Heard of this one. NEWLINE Handing over to layout expert, as crash happens in LO_SelectAll.
374|testing Bugzilla Case - 1|Created by Arun Ramachandran (arunr@netscape.com) on Tuesday, May 19, 1998 1:37:58 AM PDT NEWLINE Additional Details : NEWLINE Testing Bugzilla Case -1 NEWLINE ------- Additional Comments From arunr 05/19/98 03:31 ------- NEWLINE Update again testing
375|test case -2|Created by Arun Ramachandran (arunr@netscape.com) on Tuesday, May 19, 1998 1:42:46 AM PDT NEWLINE Additional Details : NEWLINE test case -2 NEWLINE Updated by Arun Ramachandran (arunr@netscape.com) on Tuesday, May 19, 1998 3:28:25 AM PDT NEWLINE Additional Details : NEWLINE testing - 1 NEWLINE ------- Additional Comments From arunr 05/19/98 03:31 ------- NEWLINE Update again testing NEWLINE Updated by Arun Ramachandran (arunr@netscape.com) on Wednesday, May 27, 1998 10:19:35 AM PDT NEWLINE Additional Details : NEWLINE adfasdf
381|testing|Created by Arun Ramachandran (arunr@netscape.com) on Tuesday, May 19, 1998 2:27:41 AM PDT NEWLINE Additional Details : NEWLINE testing NEWLINE Updated by Arun Ramachandran (arunr@netscape.com) on Tuesday, May 19, 1998 2:34:49 AM PDT NEWLINE Additional Details : NEWLINE testing again -1 NEWLINE ------- Additional Comments From arunr 05/19/98 03:31 ------- NEWLINE Update again testing
382|No not shared library support in Makefiles|Created by Kazushi Marukawa (jam@pobox.com) on Tuesday, May 19, 1998 3:02:14 AM PDT NEWLINE Additional Details : NEWLINE If I decide to use not shared libraries only, some makefiles NEWLINE under nsprpub directory made errors. I made patches for NEWLINE Makefiles to fix them. But I couldn't paste them here. NEWLINE NEWLINE I put my patch at NEWLINE "http://www.io.com/~kazushi/unix/netscape.noshared-patch" NEWLINE Updated by Kazushi Marukawa (jam@pobox.com) on Friday, June 12, 1998 11:42:40 PM PDT NEWLINE Additional Details : NEWLINE I decide to paste above my patch because I want to delete it. NEWLINE NEWLINE Index: nsprpub/config/rules.mk NEWLINE =================================================================== NEWLINE RCS file: /cvsroot/mozilla/nsprpub/config/rules.mk,v NEWLINE retrieving revision 3.5 NEWLINE diff -u -r3.5 rules.mk NEWLINE --- rules.mk 1998/05/07 07:23:49 3.5 NEWLINE +++ rules.mk 1998/05/19 05:47:19 NEWLINE @@ -84,10 +84,12 @@ NEWLINE else NEWLINE NEWLINE LIBRARY = NEWLINE $(OBJDIR)/lib$(LIBRARY_NAME)$(LIBRARY_VERSION).$(LIB_SUFFIX) NEWLINE +ifdef MKSHLIB NEWLINE ifeq ($(OS_ARCH)$(OS_RELEASE), AIX4.1) NEWLINE SHARED_LIBRARY = $(OBJDIR)/lib$(LIBRARY_NAME)$(LIBRARY_VERSION)_shr.a NEWLINE else NEWLINE SHARED_LIBRARY = $(OBJDIR)/lib$(LIBRARY_NAME)$(LIBRARY_VERSION).$(DLL_SUFFIX) NEWLINE +endif NEWLINE endif NEWLINE ifdef HAVE_PURIFY NEWLINE ifdef DSO_BACKEND NEWLINE Index: nsprpub/lib/ds/Makefile NEWLINE =================================================================== NEWLINE RCS file: /cvsroot/mozilla/nsprpub/lib/ds/Makefile,v NEWLINE retrieving revision 3.2 NEWLINE diff -u -r3.2 Makefile NEWLINE --- Makefile 1998/04/30 19:36:45 3.2 NEWLINE +++ Makefile 1998/05/19 05:47:20 NEWLINE @@ -97,7 +97,9 @@ NEWLINE export:: $(TARGETS) NEWLINE $(INSTALL) -m 444 $(HEADERS) $(DIST)/include NEWLINE $(INSTALL) -m 444 $(TARGETS) $(DIST)/lib NEWLINE +ifdef SHARED_LIBRARY NEWLINE $(INSTALL) -m 444 $(SHARED_LIBRARY) $(DIST)/bin NEWLINE +endif NEWLINE ifeq ($(MOZ_BITS),16) NEWLINE $(INSTALL) -m 444 $(HEADERS) $(MOZ_INCL) NEWLINE $(INSTALL) -m 444 $(TARGETS) $(MOZ_DIST)/lib NEWLINE Index: nsprpub/lib/libc/src/Makefile NEWLINE =================================================================== NEWLINE RCS file: /cvsroot/mozilla/nsprpub/lib/libc/src/Makefile,v NEWLINE retrieving revision 3.2 NEWLINE diff -u -r3.2 Makefile NEWLINE --- Makefile 1998/04/30 19:36:46 3.2 NEWLINE +++ Makefile 1998/05/19 05:47:22 NEWLINE @@ -95,7 +95,9 @@ NEWLINE NEWLINE export:: $(TARGETS) NEWLINE $(INSTALL) -m 444 $(TARGETS) $(DIST)/lib NEWLINE +ifdef SHARED_LIBRARY NEWLINE $(INSTALL) -m 444 $(SHARED_LIBRARY) $(DIST)/bin NEWLINE +endif NEWLINE ifeq ($(MOZ_BITS),16) NEWLINE $(INSTALL) -m 444 $(TARGETS) $(MOZ_DIST)/lib NEWLINE $(INSTALL) -m 444 $(TARGETS) $(MOZ_DIST)/bin NEWLINE Index: nsprpub/lib/msgc/src/Makefile NEWLINE =================================================================== NEWLINE RCS file: /cvsroot/mozilla/nsprpub/lib/msgc/src/Makefile,v NEWLINE retrieving revision 3.2 NEWLINE diff -u -r3.2 Makefile NEWLINE --- Makefile 1998/04/30 19:36:47 3.2 NEWLINE +++ Makefile 1998/05/19 05:47:25 NEWLINE @@ -98,7 +98,9 @@ NEWLINE NEWLINE export:: $(TARGETS) NEWLINE $(INSTALL) -m 444 $(TARGETS) $(DIST)/lib NEWLINE +ifdef SHARED_LIBRARY NEWLINE $(INSTALL) -m 444 $(SHARED_LIBRARY) $(DIST)/bin NEWLINE +endif NEWLINE ifeq ($(MOZ_BITS),16) NEWLINE $(INSTALL) -m 444 $(TARGETS) $(MOZ_DIST)/lib NEWLINE $(INSTALL) -m 444 $(TARGETS) $(MOZ_DIST)/bin NEWLINE Index: nsprpub/pr/src/Makefile NEWLINE =================================================================== NEWLINE RCS file: /cvsroot/mozilla/nsprpub/pr/src/Makefile,v NEWLINE retrieving revision 3.3 NEWLINE diff -u -r3.3 Makefile NEWLINE --- Makefile 1998/04/21 17:21:49 3.3 NEWLINE +++ Makefile 1998/05/19 05:47:35 NEWLINE @@ -296,7 +296,9 @@ NEWLINE NEWLINE export:: $(TARGETS) NEWLINE $(INSTALL) -m 444 $(TARGETS) $(DIST)/lib NEWLINE +ifdef SHARED_LIBRARY NEWLINE $(INSTALL) -m 444 $(SHARED_LIBRARY) $(DIST)/bin NEWLINE +endif NEWLINE ifeq ($(MOZ_BITS),16) NEWLINE $(INSTALL) -m 444 $(TARGETS) $(MOZ_DIST)/lib NEWLINE $(INSTALL) -m 444 $(TARGETS) $(MOZ_DIST)/bin NEWLINE Index: nsprpub/pr/tests/dll/Makefile NEWLINE =================================================================== NEWLINE RCS file: /cvsroot/mozilla/nsprpub/pr/tests/dll/Makefile,v NEWLINE retrieving revision 3.3 NEWLINE diff -u -r3.3 Makefile NEWLINE --- Makefile 1998/05/08 18:52:44 3.3 NEWLINE +++ Makefile 1998/05/19 05:47:56 NEWLINE @@ -58,7 +58,9 @@ NEWLINE TARGETS = $(SHARED_LIBRARY) $(IMPORT_LIBRARY) NEWLINE endif NEWLINE else NEWLINE +ifdef MKSHLIB NEWLINE SHARED_LIBRARY = $(OBJDIR)/libmy.$(DLL_SUFFIX) NEWLINE +endif NEWLINE TARGETS = $(SHARED_LIBRARY) NEWLINE endif NEWLINE NEWLINE @@ -82,9 +84,13 @@ NEWLINE endif NEWLINE NEWLINE export:: $(TARGETS) NEWLINE +ifdef SHARED_LIBRARY NEWLINE $(NSINSTALL) $(TARGETS) ../$(OBJDIR)/dll NEWLINE +endif NEWLINE NEWLINE install:: export NEWLINE NEWLINE clean:: NEWLINE +ifdef SHARED_LIBRARY NEWLINE rm -rf $(TARGETS) NEWLINE +endif NEWLINE Updated by Wan-Teh Chang (wtc@netscape.com) on Tuesday, June 16, 1998 8:33:05 PM PDT NEWLINE Additional Details : NEWLINE Accepted the bug. NEWLINE NEWLINE Thanks for the patch. I will review it NEWLINE and check it in. NEWLINE NEWLINE I am curious: which Unix platform do you NEWLINE use that doesn't have shared libraries? NEWLINE Updated by Wan-Teh Chang (wtc@netscape.com) on Monday, July 13, 1998 2:55:01 PM PDT NEWLINE Additional Details : NEWLINE Your fix has been checked into our internal NEWLINE development version of nspr. I used pretty NEWLINE much the patch you provided. NEWLINE NEWLINE The public Mozilla tree is in a stability NEWLINE code freeze right now. I will check in the NEWLINE fix when the code freeze is over.
383|test -11|Created by Arun Ramachandran (arunr@netscape.com) on Tuesday, May 19, 1998 3:30:37 AM PDT NEWLINE Additional Details : NEWLINE test -11
384|testing accented char ŕáâăäĺćçčéęëěíîďđńňóôőö÷řůúűüýţ˙|testing NEWLINE ------- Additional Comments From arunr 05/19/98 14:27 ------- NEWLINE testing again please ignore
385|GH_InitGlobalHistory called too soon|Created by Ronan Waide (waider@waider.ie) on Wednesday, May 20, 1998 4:05:03 AM PDT NEWLINE Additional Details : NEWLINE GH_InitGlobalHistory is called before FE_GlobalHist is set NEWLINE from the preferences file, meaning that gh_open_database NEWLINE doesn't have a filename to work with. Here's a patch to NEWLINE main()/mozilla-main() to fix this: NEWLINE NEWLINE Index: cmd/xfe/mozilla.c NEWLINE =================================================================== NEWLINE RCS file: /cvsroot/mozilla/cmd/xfe/mozilla.c,v NEWLINE retrieving revision 3.5 NEWLINE diff -c -r3.5 mozilla.c NEWLINE *** mozilla.c 1998/05/14 09:18:26 3.5 NEWLINE --- mozilla.c 1998/05/20 11:02:04 NEWLINE *************** NEWLINE *** 2799,2812 **** NEWLINE PREF_SetDefaultCharPref("profile.directory", NEWLINE fe_config_dir); NEWLINE PREF_SetDefaultIntPref("profile.numprofiles", 1); NEWLINE NEWLINE - GH_InitGlobalHistory(); NEWLINE - NEWLINE /* SECNAV_INIT needs this defined, but NEWLINE build_user_agent_string cannot NEWLINE * be called until after SECNAV_INIT, so call this NEWLINE simplified version. NEWLINE */ NEWLINE build_simple_user_agent_string(versionLocale); NEWLINE NEWLINE fe_InstallPreferences (0); NEWLINE NEWLINE /* NEWLINE ** Initialize the security library. NEWLINE --- 2799,2812 ---- NEWLINE PREF_SetDefaultCharPref("profile.directory", NEWLINE fe_config_dir); NEWLINE PREF_SetDefaultIntPref("profile.numprofiles", 1); NEWLINE NEWLINE /* SECNAV_INIT needs this defined, but NEWLINE build_user_agent_string cannot NEWLINE * be called until after SECNAV_INIT, so call this NEWLINE simplified version. NEWLINE */ NEWLINE build_simple_user_agent_string(versionLocale); NEWLINE NEWLINE fe_InstallPreferences (0); NEWLINE + NEWLINE + GH_InitGlobalHistory(); NEWLINE NEWLINE /* NEWLINE ** Initialize the security library. NEWLINE Updated by Steve Lamm (slamm@netscape.com) on Wednesday, May 20, 1998 6:43:36 PM PDT NEWLINE Additional Details : NEWLINE The patch looks good. NEWLINE I checked it in. NEWLINE Updated by Steve Lamm (slamm@netscape.com) on Wednesday, May 20, 1998 6:44:18 PM PDT NEWLINE Additional Details : NEWLINE Marking resolved.
386|Navigator Compile fails in lib/libmocha/et_mocha.c and et_moz.c.|Created by Chris Lahey (clahey@mediaone.net) on Wednesday, May 20, 1998 8:35:35 PM PDT NEWLINE Additional Details : NEWLINE Using CVS checkout on May 20, 1998, compilation fails with NEWLINE the message: NEWLINE et_mocha.c:42: netcache.h: No such file or directory NEWLINE NEWLINE Doing find -name netcache.h at the root of the tree reports NEWLINE that no such file exists. NEWLINE NEWLINE Removing the line NEWLINE #include "netcache.h" NEWLINE makes et_mocha.c compile with no errors. This is also a NEWLINE problem in et_moz.c. NEWLINE Chris
387|warnings when compiling jsmath.c in javascript ref|Created by (t92mas@student.tdb.uu.se) on Thursday, May 21, 1998 6:38:11 AM PDT NEWLINE Additional Details : NEWLINE Warnings: NEWLINE NEWLINE gcc -Wall -Wno-format -o Linux_All_DBG.OBJ/jsmath.o -c -g NEWLINE -DXP_UNIX -DSVR4 -DSYSV -D_BSD_SOURCE -DPOSIX_SOURCE NEWLINE -DLINUX -DDEBUG -DDEBUG_magnus -ILinux_All_DBG.OBJ NEWLINE -DJSFILE jsmath.c NEWLINE jsmath.c: In function `random_init': NEWLINE jsmath.c:256: warning: left shift count >= width of type NEWLINE jsmath.c:265: warning: left shift count >= width of type NEWLINE jsmath.c:265: warning: left shift count >= width of type NEWLINE jsmath.c:265: warning: right shift count is negative NEWLINE jsmath.c:269: warning: left shift count >= width of type NEWLINE jsmath.c:269: warning: left shift count >= width of type NEWLINE jsmath.c:269: warning: right shift count is negative NEWLINE NEWLINE NEWLINE Updated by Mike McCabe (mccabe@netscape.com) on Tuesday, May 26, 1998 10:46:16 PM PDT NEWLINE Additional Details : NEWLINE Thanks for looking at this. Keeping the js engine as warning-free as possible NEWLINE is a goal. This particular warning came up a while ago... I don't remember the NEWLINE details, but I do remember that we decided to ignore it. NEWLINE NEWLINE gcc is complaining about dead code here - the compiler warning mechanism kicks NEWLINE in before the dead code elimination does. Brendan had a fix that removed the NEWLINE warning at the cost of a runtime check, but decided that it was really more of a NEWLINE problem with gcc than with javascript, and didn't justify the workaround. NEWLINE Esthetic arguments should be directed to brendan@mozilla.org.
388|Core dump when NavCenter is open and|Created by Ronan Waide (waider@waider.ie) on Thursday, May 21, 1998 6:42:32 AM PDT NEWLINE Additional Details : NEWLINE Try this: NEWLINE Sparc/Solaris 2.5.1 NEWLINE Fire up mozilla (current version) NEWLINE Open NavCenter NEWLINE Open a history view to the level of showing individual pages NEWLINE In the main browser window, go to one of the displayed pages NEWLINE If your history view is by site, you should get a crash NEWLINE If your history view is by date, press back. *boom*. NEWLINE NEWLINE As far as I can tell, the problem is due to some sort of NEWLINE data mismatch between hist2rdf and ht. It's a bit hard to NEWLINE track as there's a lot of jump-table stuff going on. NEWLINE Stack trace: NEWLINE #0 0xef2a3dc0 in strlen () NEWLINE #1 0xef2d613c in _strdup () NEWLINE #2 0x36f7f4 in resynchItem (node=0xc71180, token=0xb87820, NEWLINE data=0x35642af9, NEWLINE assertAction=1) at ht.c:1871 NEWLINE #3 0x36ac00 in htrdfNotifFunc (ns=0xbf9420, pdata=0xbfce80) NEWLINE at ht.c:281 NEWLINE #4 0x369a48 in assertNotify (rdf=0xbf9220, not=0xbf9400, NEWLINE u=0xc6eb60, NEWLINE s=0xb87820, v=0x35642af9, type=2, tv=1, ds=0xb7ef40 NEWLINE "rdf:remoteStore") NEWLINE at mcf.c:849 NEWLINE #5 0x36a0d0 in sendNotifications (rdf=0xbf9220, opType=1, NEWLINE u=0xc6eb60, NEWLINE s=0xb87820, v=0x35642af9, type=2, tv=1, ds=0xb7ef40 NEWLINE "rdf:remoteStore") NEWLINE at mcf.c:944 NEWLINE #6 0x369f84 in sendNotifications2 (r=0xb81a80, opType=1, NEWLINE u=0xc6eb60, NEWLINE s=0xb87820, v=0x35642af9, type=2, tv=1) at mcf.c:926 NEWLINE #7 0x35ec00 in remoteStoreAdd (mcf=0xb81a80, u=0xc6eb60, NEWLINE s=0xb87820, NEWLINE v=0x35642af9, type=2, tv=1) at remstore.c:215 NEWLINE #8 0x38a500 in collateOneHist (r=0xb8a0c0, u=0xb84f40, NEWLINE url=0xb8b300 NEWLINE "http://www.cognotec.com/~waider/emacs/index.html", NEWLINE title=0xb8c2d0 "ehacks", lastAccessDate=895757049, NEWLINE firstAccessDate=895662101, numAccesses=3, byDateFlag=0) NEWLINE at hist2rdf.c:134 NEWLINE #9 0x38b4f8 in updateNewHistItem (key=0xefffc358, NEWLINE data=0xefffc348) NEWLINE at hist2rdf.c:359 NEWLINE #10 0x571218 in GH_UpdateURLTitle (pUrl=0xbcf600, NEWLINE pszTitle=0xb8c240 "ehacks", NEWLINE bFrameCell=0 '\\000') at glhist.c:2211 NEWLINE #11 0x59194c in SHIST_SetTitleOfCurrentDoc (hist=0xb9140c, NEWLINE title=0xbe4600 "ehacks") at shist.c:853 NEWLINE #12 0x114b00 in XFE_SetDocTitle (context=0xb91400, NEWLINE title=0xbe4600 "ehacks") NEWLINE at lay.c:2079 NEWLINE #13 0x423ba8 in lo_process_title_tag (context=0xb91400, NEWLINE state=0xbcf200, NEWLINE tag=0xb8c200) at laytags.c:824 NEWLINE #14 0x42bfb8 in lo_LayoutTag (context=0xb91400, NEWLINE state=0xbcf200, tag=0xb8c200) NEWLINE at laytags.c:4272 NEWLINE #15 0x3fedb4 in LO_ProcessTag (data_object=0xc71a00, NEWLINE tag=0xb8c200, status=0) NEWLINE at layout.c:4731 NEWLINE #16 0x449bcc in EDT_ProcessTag (data_object=0xc71a00, NEWLINE tag=0xb8c200, status=0) NEWLINE at editor.cpp:971 NEWLINE #17 0x535760 in PA_ParseBlock (stream=0xc74060, NEWLINE block=0xb3b000 "<!DOCTYPE HTML PUBLIC \\"-//W3C//DTD HTML NEWLINE 3.2//EN\\">\\n<HTML>\\n<HEAD>\\n <META NEWLINE HTTP-EQUIV=\\"Content-Type\\" CONTENT=\\"text/html; NEWLINE charset=iso-8859-1\\">\\n <META NAME=\\"GENERATOR\\" NEWLINE CONTENT=\\"Mozilla/4.0b2 (Win95; I) "..., NEWLINE block_len=4293) at pa_parse.c:1434 NEWLINE #18 0x53fff4 in net_AutoCharCodeConv (stream=0xc6dcc0, NEWLINE s=0xb3b000 "<!DOCTYPE HTML PUBLIC \\"-//W3C//DTD HTML NEWLINE 3.2//EN\\">\\n<HTML>\\n<HEAD>\\n <META NEWLINE HTTP-EQUIV=\\"Content-Type\\" CONTENT=\\"text/html; NEWLINE charset=iso-8859-1\\">\\n <META NAME=\\"GENERATOR\\" NEWLINE CONTENT=\\"Mozilla/4.0b2 (Win95; I) "..., NEWLINE l=4293) at net_junk.c:192 NEWLINE #19 0x336a2c in net_read_file_chunk (cur_entry=0xc5f980) at NEWLINE mkfile.c:882 NEWLINE #20 0x337760 in net_ProcessFile (cur_entry=0xc5f980) at NEWLINE mkfile.c:1233 NEWLINE #21 0x30d1b4 in NET_ProcessNet (ready_fd=0x0, fd_type=1) at NEWLINE mkgeturl.c:3158 NEWLINE #22 0x31bffc in net_process_net_timer_callback (closure=0x0) NEWLINE at mkselect.c:189 NEWLINE #23 0x1441ac in fe_do_timeout (p=0xc6df60, id=0xefffd93c) at NEWLINE xfe.c:2921 NEWLINE #24 0xef5bec2c in XtAppProcessEvent () NEWLINE #25 0x11d7d0 in fe_EventLoop () at mozilla.c:1067 NEWLINE Updated by Ronan Waide (waider@waider.ie) on Thursday, May 21, 1998 8:31:23 AM PDT NEWLINE Additional Details : NEWLINE FIX: It appears that columns.c converts some RDF_INT_TYPE items NEWLINE to HT_*_STRING types. This causes the above core. I've patched this out on my NEWLINE copy of the source, but I'm not sure as yet if it breaks anything to do this. NEWLINE Still, can't hurt to try. Here's the patch: NEWLINE NEWLINE Index: modules/rdf/src/columns.c NEWLINE =================================================================== NEWLINE RCS file: /cvsroot/mozilla/modules/rdf/src/columns.c,v NEWLINE retrieving revision 3.2 NEWLINE diff -c -r3.2 columns.c NEWLINE *** columns.c 1998/05/08 05:48:03 3.2 NEWLINE --- columns.c 1998/05/21 15:32:55 NEWLINE *************** NEWLINE *** 77,88 **** NEWLINE { NEWLINE if (u == gNavCenter->RDF_bookmarkAddDate \|\| NEWLINE u == gWebData->RDF_lastVisitDate \|\| NEWLINE ! u == gWebData->RDF_lastModifiedDate) NEWLINE { NEWLINE - val = (void *)HT_COLUMN_DATE_STRING; NEWLINE - } NEWLINE - else if (u == gWebData->RDF_firstVisitDate) NEWLINE - { NEWLINE val = (void *)HT_COLUMN_DATE_INT; NEWLINE } NEWLINE else if (u == gWebData->RDF_size \|\| NEWLINE --- 77,85 ---- NEWLINE { NEWLINE if (u == gNavCenter->RDF_bookmarkAddDate \|\| NEWLINE u == gWebData->RDF_lastVisitDate \|\| NEWLINE ! u == gWebData->RDF_lastModifiedDate \|\| NEWLINE ! u == gWebData->RDF_firstVisitDate) NEWLINE { NEWLINE val = (void *)HT_COLUMN_DATE_INT; NEWLINE } NEWLINE else if (u == gWebData->RDF_size \|\| NEWLINE *************** NEWLINE *** 93,99 **** NEWLINE else NEWLINE { NEWLINE /* default to string... XXX wrong thing to do */ NEWLINE ! val = (void *)HT_COLUMN_STRING; NEWLINE } NEWLINE } NEWLINE else if ((s == gNavCenter->RDF_ColumnWidth) && NEWLINE --- 90,96 ---- NEWLINE else NEWLINE { NEWLINE /* default to string... XXX wrong thing to do */ NEWLINE ! val = (void *)HT_COLUMN_INT; NEWLINE } NEWLINE } NEWLINE else if ((s == gNavCenter->RDF_ColumnWidth) && NEWLINE Updated by Robert Churchill (rjc@netscape.com) on Tuesday, May 26, 1998 6:13:32 PM PDT NEWLINE Additional Details : NEWLINE Don't apply the fix from Ronan Waide... its not the appropriate fix. NEWLINE NEWLINE There are really three (sigh, yes, 3) types of dates in RDF: NEWLINE NEWLINE Numeric 32-bit Integer representation of dates (such as 123456789) NEWLINE String representations of 32-bit Integer dates (such as "123456789") NEWLINE Strings dates (such as "May 26, 1998 12:02 PM") NEWLINE NEWLINE column.c wasn't "converting" RDF_INT_TYPE to a string. When asked "What kind of NEWLINE type is X?" (where X was one of the three types above) it would try and respond NEWLINE with a number (always an integer) that was its guess. NEWLINE NEWLINE Unfortunately, different sections of code in various spots in RDF would mark NEWLINE things differently... for example, Bookmark code would say that the NEWLINE RDF_FirstVisitDate entry was a "Numeric Date String" while History code would say NEWLINE that the RDF_FirstVisitDate entry was a "Numeric Date" (not a String). NEWLINE NEWLINE So, that's bad. The "right thing" to do, which I've just done, is to have every NEWLINE section of code in RDF use the same representation of dates... I chose STRINGs. NEWLINE NEWLINE Thanks for pointing this problem out! NEWLINE Robert NEWLINE rjc@netscape.com NEWLINE Updated by Robert Churchill (rjc@netscape.com) on Tuesday, May 26, 1998 6:14:01 PM PDT NEWLINE Additional Details :
389|help/about crashes due to bad return value from SECNAV_SSLCapabilities|Created by David Goodenough (dpg@ix.netcom.com) on Thursday, May 21, 1998 11:14:28 PM PDT NEWLINE Additional Details : NEWLINE This bug can be reproduced by selecting "about" from the NEWLINE "help" menu. NEWLINE NEWLINE The call to SECNAV_SSLCapabilities at line 2344 in NEWLINE winfe\\fegui.cpp expects a return value that is a malloc'ed NEWLINE string, such as would be returned from strdup(). Since the NEWLINE current SECNAV_SSLCapabilities at line 242 in navstubs.c NEWLINE returns NULL, this causes a crash at line 2346 in fegui.cpp. NEWLINE NEWLINE To fix this, alter the return statement in NEWLINE SECNAV_SSLCapabilities to be ``return(strdup(""));''
399|Compilation Error|Created by (niraj@aero.iisc.ernet.in) on Saturday, May 23, 1998 1:14:14 AM PDT NEWLINE Additional Details : NEWLINE make[3]: Entering directory NEWLINE `/var/tmp/install/19980429/mozilla/modules/libpref/src' NEWLINE cc -Ae -o HP-UXB.10.20_OPT.OBJ/prefapi.o -c -DXP_UNIX -O +Z NEWLINE -DHPUX -Dhpux -Dhppa -DHPUX10 -Dhpux10 -DHPUX10_20 NEWLINE -D_HPUX_SOURCE -DSW_THREADS NEWLINE -DNO_SIGNED -DNO_FNDELAY -DHAVE_ODD_SELECT -DNO_CDEFS_H NEWLINE -DNO_LONG_LONG -DNEED_IOCTL_H -DNEED_MATH_H -DUSE_NODL_TABS NEWLINE -DMITSHM -DNEED_SYS_WAIT NEWLINE _H -DHAVE_INT32_T -DNEED_UINT_T -DNEED_H_ERRNO NEWLINE -DRW_NO_OVERLOAD_SCHAR -DHAVE_MODEL_H -DHAVE_STRERROR NEWLINE -UDEBUG -DNDEBUG -DTRIMMED -DNSPR20 -D NEWLINE NETSCAPE -DOSTYPE=\\"HP-UXB.10\\" -DMOZILLA_CLIENT -DEDITOR NEWLINE -DMOZ_COMMUNICATOR_IIDS -DNO_SECURITY -DEDITOR -DEDITOR_UI NEWLINE -DLAYERS -DUNIX_EMBED - NEWLINE DX_PLUGINS -DUNIX_LDAP -DNSPR -DMOCHA -DUNIX_ASYNC_DNS NEWLINE -DDEVELOPER_DEBUG -I../../../dist/public/libxp NEWLINE -I../../../dist/public/js -I../../../ NEWLINE dist/public/pref -I../../../dist/public/nspr NEWLINE -I../../../dist/public/dbm -I../../../dist/public/security NEWLINE -I../../../dist/public/ldap -I../../ NEWLINE ../dist/public/java -I../../../dist/public/img NEWLINE -I../../../dist/public/layer -I../../../dist/public/util NEWLINE -I../../../include -I../../../dist NEWLINE /HP-UXB.10.20_OPT.OBJ/include -I../../../cmd/xfe/icons NEWLINE -Iunix prefapi.c NEWLINE cpp: "_hpux.h", line 43: warning 2001: Redefinition of macro NEWLINE HAVE_STRERROR. NEWLINE cc: "prefapi.c", line 124: warning 524: Cast (non-lvalue) NEWLINE appears on left-hand side of assignment. NEWLINE cc: "prefapi.c", line 124: error 1549: Modifiable lvalue NEWLINE required for assignment operator.
409|Seg. fault after visiting a page with plugin.|Created by (weitsang@cs.cornell.edu) on Sunday, May 24, 1998 9:49:53 PM PDT NEWLINE Additional Details : NEWLINE Visit http://www.geocities.com/Tokyo/Ginza/2878/index.html. NEWLINE When ask to download plugin, say "no". Then visit NEWLINE someplace else. Mozilla will crash. NEWLINE NEWLINE This seems to be a NULL pointer problem in FE_UnloadPlugin. NEWLINE We should check for possible NULL plugin->handle before NEWLINE accessing members of plugin->handle. NEWLINE Updated by Ramiro Estrugo (ramiro@netscape.com) on Tuesday, May 26, 1998 3:58:01 AM PDT NEWLINE Additional Details : NEWLINE Fixed
410|Crashes, with assert, when loading linked stylesheet|Created by gavin porter (gavin.porter@warwick.ac.uk) on Tuesday, May 26, 1998 3:25:10 AM PDT NEWLINE Additional Details : NEWLINE Whenever a page is loaded containg a <LINK> tag telling it NEWLINE to load a stylesheet. Appears to handle styles specified NEWLINE within the HTML document correctly. This error did not exist NEWLINE in the 19980331 version, and I didn't try the 19980408 NEWLINE version. NEWLINE NEWLINE Crashes with the following output, NEWLINE NEWLINE assert: line 626 in libstyle.c
411|testing|Created by Arun Ramachandran (arunr@netscape.com) on Wednesday, May 27, 1998 10:20:04 AM PDT NEWLINE Additional Details : NEWLINE sdfasdf NEWLINE Updated by Arun Ramachandran (arunr@netscape.com) on Wednesday, June 10, 1998 12:22:32 PM PDT NEWLINE Additional Details : NEWLINE Closing test call
412|Warnings in feselect.cpp|Created by David Goodenough (dpg@ix.netcom.com) on Wednesday, May 27, 1998 4:29:11 PM PDT NEWLINE Additional Details : NEWLINE Trying to compile feselect.cpp with MSVC V5.0, with level 3 NEWLINE warnings on (/W3) generates complaints about lines 64, 65, NEWLINE 99, 100, 126 & 127. Looking at the source, it appears that NEWLINE the problems are caused by the two #defines at lines 53 and NEWLINE 54. Parenthesize the RHS of these #defines and it works NEWLINE right. NEWLINE Updated by (blythe@netscape.com) on Tuesday, June 9, 1998 4:24:25 PM PDT NEWLINE Additional Details : NEWLINE Will properly handle this once we are beyond the current stability phase.
413|history.go(0) emptys hidden fields|Hidden fields are emptied by a call to history.go(0). NEWLINE Text field are not.. Users expect the behavior of hidden NEWLINE field to be the same as text field (execpt they are hidden) NEWLINE To reproduce: NEWLINE 1/ Enter a value in Text field source for "non-hidden" example NEWLINE 2/ Click transfer NEWLINE 3/ Click run 'history.go(0)' NEWLINE - the value is still here NEWLINE 1/ Enter a value in Text field source for "hidden" example NEWLINE 2/ Click transfer NEWLINE 3/ Click read to see the value in the hidden field NEWLINE 4/ Click run 'history.go(0)' NEWLINE 5/ Click read - The alert is empty.. the value is lost NEWLINE ------- Additional Comments From ludovic 04/08/98 12:21 ------- NEWLINE I love this free source thing.. I've debugged Mozilla and NEWLINE have a patch to propose.. NEWLINE Check http://champagne.mcom.com/dev/bugs/16/bug.html for NEWLINE the modified files... NEWLINE * NEWLINE * Differences to liblayout to not loose NEWLINE * the Client-JS modified hidden form values NEWLINE * when doing history.go(0) (Bug 113155) NEWLINE * NEWLINE * Disclaimer: I have no idea if these modification will break NEWLINE * or not other parts of the layout/js engine. NEWLINE In function void NEWLINE lo_CleanFormElementData(LO_FormElementData *element_data) NEWLINE layfree.c:240d239 NEWLINE < case FORM_TYPE_HIDDEN: NEWLINE layfree.c:258a258,271 NEWLINE > case FORM_TYPE_HIDDEN: NEWLINE > { NEWLINE > lo_FormElementMinimalData *form_data; NEWLINE > NEWLINE > form_data = (lo_FormElementMinimalData *) NEWLINE > element_data; NEWLINE > if (form_data->name != NULL) NEWLINE > { NEWLINE > PA_FREE(form_data->name); NEWLINE > form_data->name = NULL; NEWLINE > } NEWLINE > } NEWLINE > break; NEWLINE > NEWLINE In function static LO_FormElementStruct * NEWLINE lo_form_input_minimal(MWContext *context, lo_DocState *state, NEWLINE PA_Tag *tag, int32 type) NEWLINE layform.c:1276c1276,1278 NEWLINE < form_data->value = lo_FetchParamValue(context, tag, PARAM_VALUE); NEWLINE --- NEWLINE > if (form_data->type != FORM_TYPE_HIDDEN) NEWLINE > form_data->value = lo_FetchParamValue(context, tag, PARAM_VALUE); NEWLINE > NEWLINE ------- Additional Comments From ludovic 04/09/98 02:57 ------- NEWLINE oops.. The default value of the Hidden field is lost in NEWLINE the first layout with the modifications I did. NEWLINE This modification should be better NEWLINE @@ -1236,8 +1236,6 @@ NEWLINE if (form_data->type == FORM_TYPE_NONE) NEWLINE { NEWLINE form_data->FE_Data = NULL; NEWLINE - if (type==FORM_TYPE_HIDDEN) NEWLINE - form_data->value = lo_FetchParamValue(context, tag, PARAM_VALUE); NEWLINE }
414|on resizing browser A HREF= function() js function undefined|Created by Reba James (rjames@hotoffice.com) on Thursday, May 28, 1998 12:18:44 PM PDT NEWLINE Additional Details : NEWLINE Original Problem: NEWLINE NEWLINE Open Netscape. NEWLINE Load the url. NEWLINE Resize the browser. NEWLINE Mouse over any link. NEWLINE NEWLINE Would give you an error such as : NEWLINE JavaScript Error: NEWLINE http://208.128.121.67/asp/desktop.asp, NEWLINE NEWLINE goSubmit is not defined. NEWLINE NEWLINE Resolved this problem by replacing WriteToStatus(' '); with NEWLINE window.status= ' '; NEWLINE NEWLINE New Problem : NEWLINE NEWLINE Open Netscape. NEWLINE Load the url. NEWLINE Resize the browser. NEWLINE Click any link. NEWLINE NEWLINE Would give you an error such as : NEWLINE error on line 999 NEWLINE function xxx is not defined, xxx.inc NEWLINE NEWLINE NEWLINE Would welcome any solutions ???? NEWLINE Thank You. NEWLINE NEWLINE Reba James \\\\\|\|\|// NEWLINE HotOfficeTechnologies, Inc NEWLINE Software Engineer (.)-(.) NEWLINE 561-995-0005x223 NEWLINE ===ooOO=(_)=OOoo== NEWLINE rjames @ hotoffice.com NEWLINE NEWLINE NEWLINE NEWLINE Updated by Mike McCabe (mccabe@netscape.com) on Monday, June 1, 1998 2:39:56 PM PDT NEWLINE Additional Details : NEWLINE This bug is a duplicate of bugzilla bug 415. NEWLINE Updated by Mike McCabe (mccabe@netscape.com) on Monday, June 1, 1998 2:55:09 PM PDT NEWLINE Additional Details : NEWLINE This bug is a duplicate of bugzilla bug 415.
415|on resizing browser A HREF= function() js function undefined|Created by Reba James (rjames@hotoffice.com) on Thursday, May 28, 1998 12:22:52 PM PDT NEWLINE Additional Details : NEWLINE Original Problem: NEWLINE NEWLINE Open Netscape. NEWLINE Load the url. NEWLINE Resize the browser. NEWLINE Mouse over any link. NEWLINE NEWLINE Would give you an error such as : NEWLINE JavaScript Error: NEWLINE http://208.128.121.77/asp/desktop.asp, line 213: NEWLINE writeToStatus is not defined. NEWLINE NEWLINE Resolved this problem by replacing WriteToStatus(' '); with NEWLINE window.status= ' '; NEWLINE NEWLINE New Problem : NEWLINE NEWLINE Open Netscape. NEWLINE Load the url. NEWLINE Resize the browser. NEWLINE Click any link. NEWLINE NEWLINE Would give you an error such as : NEWLINE JavaScript Error: NEWLINE http://208.128.121.67/asp/desktop.asp, NEWLINE goSubmit is not defined. NEWLINE NEWLINE Would welcome any solutions ???? NEWLINE NEWLINE Release deadline June 1st. NEWLINE Thank You. NEWLINE NEWLINE Reba James \\\\\|\|\|// NEWLINE HotOfficeTechnologies, Inc. NEWLINE Software Engineer (.)-(.) NEWLINE 561-995-0005x223 NEWLINE ===ooOO=(_)=OOoo== NEWLINE rjames @ hotoffice.com NEWLINE NEWLINE NEWLINE NEWLINE Updated by Reba James (rjames@hotoffice.com) on Thursday, May 28, 1998 12:29:11 PM PDT NEWLINE Additional Details : NEWLINE Runing Communicator 4.05. NEWLINE 50% of our users are Netscape Clients. NEWLINE NEWLINE Deadline June 1St. NEWLINE This is a priority one issue. NEWLINE Need a solution fast. NEWLINE have called number after number, trying to purchase an incident but to no avail. NEWLINE Even the online registration bombs out. You are my final solution. NEWLINE please call me ASAP. THANK YOU IN ADVANCE NEWLINE NEWLINE Reba James \\\\\|\|\|// NEWLINE HotOfficeTechnologies, Inc. NEWLINE Software Engineer (.)-(.) NEWLINE 561-995-0005x223 NEWLINE ===ooOO=(_)=OOoo== NEWLINE rjames @ hotoffice.com NEWLINE Updated by Mike McCabe (mccabe@netscape.com) on Monday, June 1, 1998 3:19:44 PM PDT NEWLINE Additional Details : NEWLINE This doesn't seem to be a bug in the core javascript engine or Mozilla. NEWLINE NEWLINE Based on the resolution of the resolved bug that you mention, it seems like this NEWLINE is an area where the client functions exposed to javascript differ between NEWLINE Netscape and IE. 'goSubmit is not defined' may imply that the Netscape browser NEWLINE does not define the 'goSubmit' function, and there is another method of NEWLINE accomplishing the same thing. (May I assume that you developed the .asp page NEWLINE using the IE browser?) I would suggest that you use one of several javascript NEWLINE libraries that attempt to abstract the different browsers. NEWLINE NEWLINE Closing the bug...
416|Mozilla compile fails under AIX 4.1|Created by (jew@rtd.com) on Thursday, May 28, 1998 4:40:23 PM PDT NEWLINE Additional Details : NEWLINE It appears to be using IBM's C compiler, though I have gcc NEWLINE installed. Is this correct? This is a vrigin install, I NEWLINE have modified nothing. NEWLINE NEWLINE ...AIX4.1_OPT.OBJ/include/private NEWLINE -I../../../../dist/AIX4.1_OPT.OBJ/include prlink NEWLINE .c NEWLINE "prlink.c", line 32.10: 1506-296 (S) #include file <dlfcn.h> NEWLINE not found. NEWLINE "prlink.c", line 100.19: 1506-068 (S) Operation between NEWLINE types "const char*" and NEWLINE "int" is not allowed. NEWLINE "prlink.c", line 155.19: 1506-045 (S) Undeclared identifier NEWLINE RTLD_NOW. NEWLINE "prlink.c", line 622.32: 1506-045 (S) Undeclared identifier NEWLINE RTLD_NOW. NEWLINE "prlink.c", line 801.7: 1506-068 (S) Operation between types NEWLINE "void*" and "int" i NEWLINE s not allowed. NEWLINE gmake[4]: *** [AIX4.1_OPT.OBJ/prlink.o] Error 1 NEWLINE gmake[4]: Leaving directory NEWLINE `/.../ssd.sanjose.ibm.com/fs/home/jeward/src/1998042 NEWLINE 9/mozilla/nsprpub/pr/src/linking' NEWLINE gmake[3]: *** [export] Error 2 NEWLINE gmake[3]: Leaving directory NEWLINE `/.../ssd.sanjose.ibm.com/fs/home/jeward/src/1998042 NEWLINE 9/mozilla/nsprpub/pr/src' NEWLINE gmake[2]: *** [export] Error 2 NEWLINE gmake[2]: Leaving directory NEWLINE `/.../ssd.sanjose.ibm.com/fs/home/jeward/src/1998042 NEWLINE 9/mozilla/nsprpub/pr' NEWLINE gmake[1]: *** [export] Error 2 NEWLINE gmake[1]: Leaving directory NEWLINE `/.../ssd.sanjose.ibm.com/fs/home/jeward/src/1998042 NEWLINE 9/mozilla/nsprpub' NEWLINE gmake: *** [export] Error 2 NEWLINE Updated by Wan-Teh Chang (wtc@netscape.com) on Monday, June 8, 1998 6:20:28 PM PDT NEWLINE Additional Details : NEWLINE Assigned the bug to myself. NEWLINE Updated by Wan-Teh Chang (wtc@netscape.com) on Tuesday, June 16, 1998 8:45:41 PM PDT NEWLINE Additional Details : NEWLINE Our AIX 4.1.5.0 machine here has <dlfcn.h>. NEWLINE Can you run 'oslevel' on your machine and NEWLINE tell me what version it is? If it is not NEWLINE 4.1.5.0, can you upgrade it? NEWLINE NEWLINE Judging from the format of the error messages, NEWLINE I believe you are using IBM's C compiler, not gcc.
417|seg fault when ran without specifying a URL|Created by brian hourigan (digi@inreach.net) on Sunday, May 31, 1998 5:08:25 PM PDT NEWLINE Additional Details : NEWLINE when you run mozilla without passing a argument, it will NEWLINE sigsegv. NEWLINE NEWLINE eg; NEWLINE NEWLINE moz-export NEWLINE NEWLINE that would cause it to sigsegv.
418|when downloading a file, screen is all fubared|Created by brian hourigan (digi@inreach.net) on Sunday, May 31, 1998 5:11:26 PM PDT NEWLINE Additional Details : NEWLINE when you download a file, the default size is probably a lot NEWLINE larger than what desired, and the top of the contents NEWLINE resides at the bottom of the window
419|scroll bar is missing until you scroll down|Created by brian hourigan (digi@inreach.net) on Sunday, May 31, 1998 5:18:27 PM PDT NEWLINE Additional Details : NEWLINE i apologize for submitting to possibly the incorrect NEWLINE catagory, but when you load a document (eg; moz-export NEWLINE http://www.fooman.com/2pagedocument.html), at first the NEWLINE scrollbar is missing. but when you push a key to scroll, it NEWLINE appears
420|[NATIVE-WIDGET] a <select> doesnt show up as the document background color|Created by brian hourigan (digi@inreach.net) on Sunday, May 31, 1998 5:24:00 PM PDT NEWLINE Additional Details : NEWLINE when you have a <select> tag, it doesnt make the menu the NEWLINE background color
421|mkicons ld failure|Created by Keith Otis (Keith_Otis@fa.disney.com) on Monday, June 1, 1998 9:39:45 AM PDT NEWLINE Additional Details : NEWLINE cc -DXP_UNIX -g -DIRIX -multigot -Wl,-nltgot,170 -DIRIX6_2 NEWLINE -32 -fullwarn -xansi -DIRIX_STARTUP_SPEEDUPS -DSVR4 NEWLINE -DHAVE_LCHOWN -DHAVE_SIGNED_CHAR -DHAVE_FILIO_H -DHAS_PGNO_T NEWLINE -DMITSHM -DHAVE_WAITID -DNEED_VBASE -DNEED_SYS_TIME_H NEWLINE -DHAVE_SYSTEMINFO_H -DNO_JNI_STUBS -D_MIPS_SIM_ABI32 NEWLINE -DNO_UINT32_T -DNO_INT64_T -DNEED_BSD_TYPES -DSW_THREADS NEWLINE -DDEBUG -UNDEBUG -DDEBUG_root -DTRACING -DNSPR20 -DNETSCAPE NEWLINE -DOSTYPE=\\"IRIX6\\" -DMOZILLA_CLIENT -DEDITOR NEWLINE -DMOZ_COMMUNICATOR_IIDS -DNO_SECURITY -DEDITOR -DEDITOR_UI NEWLINE -DLAYERS -DUNIX_EMBED -DX_PLUGINS -DUNIX_LDAP -DNSPR -DMOCHA NEWLINE -DUNIX_ASYNC_DNS -DDEVELOPER_DEBUG -DXFE NEWLINE -I../../../dist/public/img -I../../../dist/public/java NEWLINE -I../../../dist/public/jtools -I../../../dist/public/layer NEWLINE -I../../../dist/public/nspr -I../../../dist/public/util NEWLINE -I../../../dist/public/xp -I../../../include -I. NEWLINE -IIRIX6.2_DBG.OBJ -I../../../dist/IRIX6.2_DBG.OBJ/include NEWLINE -o IRIX6.2_DBG.OBJ/mkicons -I../../../modules/libimg/src NEWLINE mkicons.c ../../../dist/IRIX6.2_DBG.OBJ/lib/libimg.a NEWLINE ../../../dist/IRIX6.2_DBG.OBJ/lib/libstubsj.a NEWLINE ../../../dist/IRIX6.2_DBG.OBJ/lib/libpng.a NEWLINE ../../../dist/IRIX6.2_DBG.OBJ/lib/libzlib.a /usr/lib/libm.a NEWLINE ../../../dist/IRIX6.2_DBG.OBJ/lib/libnspr21.a NEWLINE ../../../dist/IRIX6.2_DBG.OBJ/lib/libjpeg.a NEWLINE ../../../dist/IRIX6.2_DBG.OBJ/lib/libutil.a NEWLINE ../../../dist/IRIX6.2_DBG.OBJ/lib/libxp.a NEWLINE ld: ERROR 33: Unresolved text symbol "PL_strncpyz" -- 1st NEWLINE referenced by NEWLINE ../../../dist/IRIX6.2_DBG.OBJ/lib/libxp.a(xp_file.o). NEWLINE ld: ERROR 33: Unresolved text symbol "PREF_GetCharPref" -- NEWLINE 1st referenced by NEWLINE ../../../dist/IRIX6.2_DBG.OBJ/lib/libxp.a(xp_file.o). NEWLINE ld: ERROR 33: Unresolved text symbol "PL_strcasecmp" -- 1st NEWLINE referenced by NEWLINE ../../../dist/IRIX6.2_DBG.OBJ/lib/libxp.a(xp_file.o). NEWLINE ld: INFO 152: Output file removed because of error.
422|Mozilla crashs if PICS label missing closing right paren|Created by Paul Chek (pchek@ziplink.net) on Tuesday, June 2, 1998 8:16:10 AM PDT NEWLINE Additional Details : NEWLINE If a site returns a PICS label without the closing NEWLINE right paren Mozilla crashes in the PICS parser, file NEWLINE cslabel.c, function LabelTree_open. NEWLINE NEWLINE Sample label: NEWLINE (PICS-1.1 ""http://www.rsac.org/ratingsv01.html"" l gen true NEWLINE r (n 0 s 0 v 0 l 0) NEWLINE NEWLINE NOTE THE MISSING RIGHT PAREN AT THE END. NEWLINE NEWLINE Who sent such a obviously bad label. Hit the search NEWLINE button in mozilla.
423|[CORE} NavCenter tries to update after closing|Created by Ronan Waide (waider@waider.ie) on Thursday, June 4, 1998 10:01:25 AM PDT NEWLINE Additional Details : NEWLINE Open NavCenter, close it again, then go browsing. mozilla NEWLINE will shortly fall over with the core dump below. It appears NEWLINE to be attempting to update the closed navcenter window. NEWLINE NEWLINE #0 0xef5ba470 in XtIsSubclass () NEWLINE #1 0x281024 in WidgetToTree (w=0x0, funcname=0x780f78 NEWLINE "XmLTreeAddRows()") NEWLINE at Tree.c:1255 NEWLINE #2 0x2811a8 in XmLTreeAddRows (w=0x0, rows=0xefffb818, NEWLINE count=1, position=2) NEWLINE at Tree.c:1310 NEWLINE #3 0x28115c in XmLTreeAddRow (w=0x0, level=1, expands=0 NEWLINE '\\000', NEWLINE isExpanded=0 '\\000', position=2, pixmap=62915614, NEWLINE pixmask=62915616, NEWLINE string=0xbf7840 NEWLINE "ß\\200\\006#\\001\\tISO8859-1\\002\\026Antipatico's Home Page") NEWLINE at Tree.c:1296 NEWLINE #4 0x201020 in XFE_RDFView::add_row (this=0xc46900, NEWLINE node=0xc08200) NEWLINE at RDFView.cpp:625 NEWLINE #5 0x2009a4 in XFE_RDFView::notify (this=0xc46900, NEWLINE ns=0xc495e0, n=0xc08200, NEWLINE whatHappened=1) at RDFView.cpp:449 NEWLINE #6 0x1c8d04 in XFE_NavCenterView::notify (this=0xc56180, NEWLINE ns=0xc495e0, NEWLINE n=0xc08200, whatHappened=1) at NavCenterView.cpp:203 NEWLINE #7 0x1c8b8c in notify_cb (ns=0xc495e0, n=0xc08200, NEWLINE whatHappened=1) NEWLINE at NavCenterView.cpp:154 NEWLINE #8 0x3735b8 in sendNotification (node=0xc08200, NEWLINE whatHappened=1) at ht.c:1717 NEWLINE #9 0x370ee8 in refreshItemList (node=0xc08200, NEWLINE whatHappened=1) at ht.c:900 NEWLINE #10 0x374eac in addContainerItem (container=0xc2cb80, NEWLINE item=0xc044e0) NEWLINE at ht.c:2275 NEWLINE #11 0x36f798 in htrdfNotifFunc (ns=0xc49820, pdata=0xc2c380) NEWLINE at ht.c:444 NEWLINE #12 0x38f1d8 in assertNotify (rdf=0xc49640, not=0xc49800, NEWLINE u=0xc044e0, NEWLINE s=0xba2b40, v=0xba7c00, type=1, tv=1, ds=0xbaa6a0 NEWLINE "rdf:history") NEWLINE at mcf.c:980 NEWLINE #13 0x38f860 in sendNotifications (rdf=0xc49640, opType=1, NEWLINE u=0xc044e0, NEWLINE s=0xba2b40, v=0xba7c00, type=1, tv=1, ds=0xbaa6a0 NEWLINE "rdf:history") NEWLINE at mcf.c:1075 NEWLINE #14 0x38f714 in sendNotifications2 (r=0xba6740, opType=1, NEWLINE u=0xc044e0, NEWLINE s=0xba2b40, v=0xba7c00, type=1, tv=1) at mcf.c:1057 NEWLINE #15 0x39bde0 in histAddParent (child=0xc044e0, NEWLINE parent=0xba7c00) NEWLINE at hist2rdf.c:505 NEWLINE #16 0x39aab0 in hostUnitOfURL (r=0xba6740, top=0xba7c00, NEWLINE nu=0xc044e0, NEWLINE title=0xbf77d0 "Antipatico's Home Page") at NEWLINE hist2rdf.c:201 NEWLINE #17 0x39a4b4 in collateOneHist (r=0xba6740, u=0xba7c00, NEWLINE url=0xbf7740 NEWLINE "http://www.geocities.com/SunsetStrip/Backstage/4183/", NEWLINE title=0xbf77d0 "Antipatico's Home Page", NEWLINE lastAccessDate=896975968, NEWLINE firstAccessDate=896975968, numAccesses=1, byDateFlag=0) NEWLINE at hist2rdf.c:125 NEWLINE #18 0x39b630 in updateNewHistItem (key=0xefffc250, NEWLINE data=0xefffc240) NEWLINE at hist2rdf.c:376 NEWLINE #19 0x57f950 in GH_UpdateURLTitle (pUrl=0xbff800, NEWLINE pszTitle=0xbfc180 "Antipatico's Home Page", bFrameCell=0 NEWLINE '\\000') NEWLINE at glhist.c:2211 NEWLINE #20 0x5a0084 in SHIST_SetTitleOfCurrentDoc (hist=0xbcc40c, NEWLINE title=0xbf4800 "Antipatico's Home Page") at shist.c:853 NEWLINE #21 0x115e0c in XFE_SetDocTitle (context=0xbcc400, NEWLINE title=0xbf4800 "Antipatico's Home Page") at lay.c:2079 NEWLINE #22 0x4323a8 in lo_process_title_tag (context=0xbcc400, NEWLINE state=0xc27e00, NEWLINE tag=0xc044a0) at laytags.c:841 NEWLINE #23 0x43a7b8 in lo_LayoutTag (context=0xbcc400, NEWLINE state=0xc27e00, tag=0xc044a0) NEWLINE at laytags.c:4292 NEWLINE #24 0x40d5b4 in LO_ProcessTag (data_object=0xc95d80, NEWLINE tag=0xc044a0, status=0) NEWLINE at layout.c:4731 NEWLINE #25 0x4583d8 in EDT_ProcessTag (data_object=0xc95d80, NEWLINE tag=0xc044a0, status=0) NEWLINE at editor.cpp:971 NEWLINE #26 0x54645c in PA_ParseBlock (stream=0xc55d20, NEWLINE block=0x98d400 "tle>\\n<META HTTP-EQUIV=\\"Content-type\\" NEWLINE CONTENT=\\"text/html; charset=ISO-8859-1\\">\\n<META NEWLINE HTTP-EQUIV=\\"Content-language\\" CONTENT=\\"en-US\\">\\n<LINK NEWLINE REV=made href=\\"mailto:jamesc@s3dub.ie\\">\\n<META NEWLINE NAME=\\"keywords\\" C"..., block_len=900) at pa_parse.c:1434 NEWLINE #27 0x550cf0 in net_AutoCharCodeConv (stream=0xc55900, NEWLINE s=0x98d400 "tle>\\n<META HTTP-EQUIV=\\"Content-type\\" NEWLINE CONTENT=\\"text/html; charset=ISO-8859-1\\">\\n<META NEWLINE HTTP-EQUIV=\\"Content-language\\" CONTENT=\\"en-US\\">\\n<LINK NEWLINE REV=made href=\\"mailto:jamesc@s3dub.ie\\">\\n<META NEWLINE NAME=\\"keywords\\" C"..., NEWLINE l=900) at net_junk.c:192 NEWLINE #28 0x2ed3ec in net_CacheWrite (stream=0xc55da0, NEWLINE buffer=0x98d400 "tle>\\n<META HTTP-EQUIV=\\"Content-type\\" NEWLINE CONTENT=\\"text/html; charset=ISO-8859-1\\">\\n<META NEWLINE HTTP-EQUIV=\\"Content-language\\" CONTENT=\\"en-US\\">\\n<LINK NEWLINE REV=made href=\\"mailto:jamesc@s3dub.ie\\">\\n<META NEWLINE NAME=\\"keywords\\" C"..., len=900) at mkcache.c:1392 NEWLINE #29 0x3034c8 in net_ChunkedWrite (stream=0xc558e0, NEWLINE s=0xb58000 "tle>\\n<META HTTP-EQUIV=\\"Content-type\\" NEWLINE CONTENT=\\"text/html; charset=ISO-8859-1\\">\\n<META NEWLINE HTTP-EQUIV=\\"Content-language\\" CONTENT=\\"en-US\\">\\n<LINK NEWLINE REV=made href=\\"mailto:jamesc@s3dub.ie\\">\\n<META NEWLINE NAME=\\"keywords\\" C"..., NEWLINE l=900) at cvchunk.c:124 NEWLINE #30 0x358f0c in net_pull_http_data (ce=0xc2c100) at NEWLINE mkhttp.c:3108 NEWLINE #31 0x359d60 in net_ProcessHTTP (ce=0xc2c100) at NEWLINE mkhttp.c:3497 NEWLINE #32 0x31632c in NET_ProcessNet (ready_fd=0xc49360, NEWLINE fd_type=2) NEWLINE at mkgeturl.c:3172 NEWLINE #33 0x322c20 in net_process_net_timer_callback (closure=0x0) NEWLINE at mkselect.c:189 NEWLINE #34 0x1455c0 in fe_do_timeout (p=0xc44940, id=0xefffd93c) at NEWLINE xfe.c:2923 NEWLINE #35 0xef5bec2c in XtAppProcessEvent () NEWLINE #36 0x11eb10 in fe_EventLoop () at mozilla.c:1072 NEWLINE #37 0x123af8 in main (argc=1, argv=0xeffffa9c) at NEWLINE mozilla.c:3344 NEWLINE Updated by Steve Lamm (slamm@netscape.com) on Wednesday, June 17, 1998 1:31:30 PM PDT NEWLINE Additional Details : NEWLINE I have a fix for this. I will check it in as soon as I get approval. NEWLINE Updated by Steve Lamm (slamm@netscape.com) on Wednesday, June 17, 1998 2:38:37 PM PDT NEWLINE Additional Details : NEWLINE Ok, the fix is checked in.
424|Warning fix: -DBSD on BSD platforms|Created by Bert Driehuis (driehuis@playbeing.org) on Friday, June 5, 1998 12:38:10 PM PDT NEWLINE Additional Details : NEWLINE Systems which define BSD in one of the system header files NEWLINE cause spurious warning messages. In message NEWLINE snews://secnews.netscape.com/Pine.SUN.3.95q.980601001123.13327A-200000%40dewmoth NEWLINE I propose a workaround by renaming the instances of -DBSD NEWLINE actually used by Mozilla to GENERIC_BSD. NEWLINE Updated by Wan-Teh Chang (wtc@netscape.com) on Monday, June 8, 1998 6:22:27 PM PDT NEWLINE Additional Details : NEWLINE Reassigned the bug to Brian Ostrom.
425|Communicator crashes on startup|Created by Jauder Ho (jauderho@carumba.com) on Saturday, June 6, 1998 3:51:15 AM PDT NEWLINE Additional Details : NEWLINE This is from today's snapshot but I noticed that this has NEWLINE happened for a while. The snapshot actually gets further now NEWLINE before barfing. NEWLINE NEWLINE strace shows the following, it seems that it is not able to NEWLINE find the splash screen. NEWLINE NEWLINE read(10, "\\37\\200\\"\\1\\37\\200\\"\\1O\\0\\236\\10"..., 12) = 12 NEWLINE read(10, "\\1\\30\\372\\r\\0\\0\\0\\0T\\0\\0\\0\\0\\0\\0"..., 32) = 32 NEWLINE gettimeofday({897130230, 211362}, NULL) = 0 NEWLINE open("/@LOC_LANG@/netscape/splash", O_RDONLY) = -1 ENOENT NEWLINE (No such file or directory) NEWLINE brk(0x8ad6000) = 0x8ad6000 NEWLINE brk(0x8ad7000) = 0x8ad7000 NEWLINE --- SIGSEGV (Segmentation fault) --- NEWLINE +++ killed by SIGSEGV +++ NEWLINE NEWLINE NEWLINE --Jauder
426|Having problem compiling - suse 5.1, Motif 2.0|Created by M reddy (tekusa@io.com) on Saturday, June 6, 1998 8:25:07 AM PDT NEWLINE Additional Details : NEWLINE NEWLINE Unable to complie for the past 3 weeks. Help is appreciated. NEWLINE Has anyone successfully complied on SuSe distribution 5.1 NEWLINE using Motif 2.0 so far. NEWLINE NEWLINE Reddy
427|clicking on a link while printing dumps core|Created by Cyrille Chepelov (chepelov@rip.ens-cachan.fr) on Sunday, June 7, 1998 6:08:36 AM PDT NEWLINE Additional Details : NEWLINE open a page. NEWLINE Click on print, click "OK" in the print dialog NEWLINE *before* the page has been sent to the spooler (message NEWLINE dialog box from lp), click on a link. NEWLINE Watch the browser die. NEWLINE NEWLINE Done with : any version from the CVS (including last week's) NEWLINE NEWLINE compilator : egcs-1.02, libc is Sun's. (Solaris 2.5.1) NEWLINE real category : printing.
428|DialogShell widget crash|Created by Paul (pwilkins@wilkins.ne.mediaone.net) on Sunday, June 7, 1998 7:03:36 PM PDT NEWLINE Additional Details : NEWLINE Mozilla version 19980603 NEWLINE SUSE linux 5.1 NEWLINE Motif v1.24 from RedHat CDE distribution NEWLINE NEWLINE Mozilla prints this message: NEWLINE NEWLINE Error: DialogShell widget only supports one rectObj child NEWLINE NEWLINE Then exits when I do: NEWLINE NEWLINE Edit Menu / Prefrences / Colors NEWLINE then click on the button that lets you change the text NEWLINE color. NEWLINE NEWLINE -Paul
429|JB_SP Symbol not defined|Created by Bernd Retzlaff (bretzlaff@bb-data.de) on Monday, June 8, 1998 7:52:56 AM PDT NEWLINE Additional Details : NEWLINE During make Igot the following message: NEWLINE NEWLINE `JB_SP' undeclared (first use this function) NEWLINE NEWLINE How do I have to set this value ? NEWLINE NEWLINE Thanks NEWLINE Updated by Wan-Teh Chang (wtc@netscape.com) on Monday, June 8, 1998 6:26:53 PM PDT NEWLINE Additional Details : NEWLINE Can you tell me the version number of your Linux kernel? NEWLINE I am assuming you are using Alpha Linux? NEWLINE Can you also email me the files /usr/include/jmp_buf.h NEWLINE and /usr/include/alpha/jmp_buf.h? NEWLINE NEWLINE I need to know how 'sigjmpbuf' is defined on your system. NEWLINE Updated by Wan-Teh Chang (wtc@netscape.com) on Tuesday, June 16, 1998 8:35:40 PM PDT NEWLINE Additional Details : NEWLINE Accepted the bug. NEWLINE NEWLINE Still don't have time to look at the header NEWLINE files you sent me. Sorry :-) NEWLINE Updated by Wan-Teh Chang (wtc@netscape.com) on Monday, July 13, 1998 11:44:44 AM PDT NEWLINE Additional Details : NEWLINE Bernd Retzlaff supplied the following info: NEWLINE The machine is running RedHat Linux 4.1 for NEWLINE Alpha (kernel 2.0.30). NEWLINE NEWLINE The type __jmp_buf is defined as follows: NEWLINE typedef struct NEWLINE { NEWLINE ... NEWLINE long int *__pc, *__fp, *__sp; NEWLINE ... NEWLINE } __jmp_buf[1]; NEWLINE NEWLINE Bernd, can you send me the /usr/include/features.h NEWLINE file on your machine, so that I can determine the NEWLINE major and minor version numbers of your libc? NEWLINE NEWLINE On the RedHat 5.0 Linux for alpha machine we have NEWLINE here, its /usr/include/features.h defines the NEWLINE following macros: NEWLINE #define __GNU_LIBRARY__ 6 NEWLINE #define __GLIBC__ 2 NEWLINE #define __GLIBC_MINOR__ 0 NEWLINE NEWLINE I hope that we can programmatically distinguish your libc NEWLINE from the libc on our machine by testing these macros. NEWLINE If this doesn't work, I will teach you how to manually NEWLINE edit /mozilla/nsprpub/pr/include/md/_linux.h for your NEWLINE system.
430|Mozilla (1998-06-03) fails to initialise X and dies|Created by Peter Collinson (pc@hillside.co.uk) on Monday, June 8, 1998 5:27:46 PM PDT NEWLINE Additional Details : NEWLINE On Solaris 2.6 (Sparc) under CDE, Mozilla dies with a NEWLINE segmentation fault. The problem is in NEWLINE XtAppInitialize (line 2393 of mozilla.c). Using adb seems NEWLINE to show that the crash happens in GetDatabase, called NEWLINE from XrmGetStringDatabase, called from XmVirtKeysInitialize.
431|Crash while pressing the back button|Created by Patrick Scieller (scieller@bigfoot.com) on Tuesday, June 9, 1998 7:26:31 AM PDT NEWLINE Additional Details : NEWLINE Mozilla failed as I was pushing the [Back] button on the NEWLINE toolbar. NEWLINE A personal bookmark folder was opened on the left. NEWLINE Previously navigation went to Altavista, Lycos and Yahoo and NEWLINE I closed and NEWLINE opened left bookmark folders two or three times. NEWLINE NEWLINE ******** DEBUGGER INFORMATION ********* NEWLINE call stack NEWLINE ---------- NEWLINE lo_UpdateStateWhileFlushingLine(MWContext_ * 0x0099cb00, NEWLINE lo_DocState_struct * 0x01c22958) line 7498 + 9 bytes NEWLINE lo_FlushLineList(MWContext_ * 0x0099cb00, lo_DocState_struct NEWLINE * 0x01c22958, unsigned long 0, unsigned long 0, int 0) line NEWLINE 1083 + 13 bytes NEWLINE lo_EndLayer(MWContext_ * 0x0099cb00, lo_DocState_struct * NEWLINE 0x01c22958, int 1) line 2230 + 19 bytes NEWLINE lo_FinishLayout(MWContext_ * 0x0099cb00, lo_DocState_struct NEWLINE * 0x01c22958, long 524288) line 3676 + 15 bytes NEWLINE lo_FlushBlockage(MWContext_ * 0x0099cb00, lo_DocState_struct NEWLINE * 0x01c22958, lo_DocState_struct * 0x01c22958) line 2747 + NEWLINE 18 bytes NEWLINE lo_block_src_exit_fn(URL_Struct_ * 0x00c3cb10, int 1, NEWLINE MWContext_ * 0x0099cb00) line 471 + 17 bytes NEWLINE net_CallExitRoutine(void (URL_Struct_ *, int, MWContext_ *)* NEWLINE 0x004a8cb0 lo_block_src_exit_fn(URL_Struct_ *, int, NEWLINE MWContext_ *), URL_Struct_ * 0x00c3cb10, int 1, int 23, NEWLINE MWContext_ * 0x0099cb00) line 1053 + 15 bytes NEWLINE NET_ProcessNet(PRFileDesc * 0x00000000, int 1) line 3309 + NEWLINE 39 bytes NEWLINE net_process_slow_net_timer_callback(void * 0x00000000) line NEWLINE 216 + 9 bytes NEWLINE wfe_ProcessTimeouts(unsigned long 11872822) line 298 + 12 NEWLINE bytes NEWLINE FireTimeout(HWND__ * 0x001000ce, unsigned int 275, unsigned NEWLINE int 777, unsigned long 11872822) line 60 + 9 bytes NEWLINE USER32! 77e71373() NEWLINE CNetscapeApp::Run() line 1608 + 8 bytes NEWLINE AfxWinMain(HINSTANCE__ * 0x00400000, HINSTANCE__ * NEWLINE 0x00000000, char * 0x00141fea, int 1) line 52 + 11 bytes NEWLINE WinMain(HINSTANCE__ * 0x00400000, HINSTANCE__ * 0x00000000, NEWLINE char * 0x00141fea, int 1) line 33 + 21 bytes NEWLINE WinMainCRTStartup() line 330 + 57 bytes NEWLINE KERNEL32! 77f1b304() NEWLINE NEWLINE NEWLINE source code : layout.c NEWLINE ----------- NEWLINE NEWLINE ------------------------------------------------------------ NEWLINE ------------ NEWLINE The symbol ">>>>" show the line where it fails with NEWLINE state->list_stack -> NULL NEWLINE ------------------------------------------------------------ NEWLINE ------------ NEWLINE NEWLINE void lo_UpdateStateWhileFlushingLine( MWContext NEWLINE *context, lo_DocState *state ) NEWLINE { NEWLINE int32 justification_remainder=0; NEWLINE NEWLINE if (state->top_state->nothing_displayed != FALSE) NEWLINE { NEWLINE /* NEWLINE * If we are displaying elements we are NEWLINE * no longer in the HEAD section of the HTML NEWLINE * we are in the BODY section. NEWLINE */ NEWLINE state->top_state->in_head = FALSE; NEWLINE state->top_state->in_body = TRUE; NEWLINE NEWLINE lo_use_default_doc_background(context, NEWLINE state); NEWLINE state->top_state->nothing_displayed = FALSE; NEWLINE } NEWLINE NEWLINE /* NEWLINE * There is a break at the end of this line. NEWLINE * this may change min_width. NEWLINE */ NEWLINE { NEWLINE int32 new_break_holder; NEWLINE int32 min_width; NEWLINE int32 indent; NEWLINE NEWLINE new_break_holder = state->x; NEWLINE min_width = new_break_holder - NEWLINE state->break_holder; NEWLINE >>>> indent = state->list_stack->old_left_margin - NEWLINE state->win_left; NEWLINE min_width += indent; NEWLINE if (min_width > state->min_width) NEWLINE { NEWLINE state->min_width = min_width; NEWLINE }
432|PowerBuilder plugin causes random page fault error.|Created by Jean Campbell (jccampb@mitre.org) on Tuesday, June 9, 1998 11:36:22 AM PDT NEWLINE Additional Details : NEWLINE On opening html page with embedded object, a page fault NEWLINE error is generated (sometimes): NEWLINE NEWLINE "This program has performed an illegal operation and will be NEWLINE shut down. If the problem persists, contact the program NEWLINE vendor. NEWLINE NEWLINE NETSCAPE caused an invalid page fault in module PBVM60.DLL NEWLINE at 0137:1123f86b." NEWLINE NEWLINE The embed object references a child window in a PowerBuilder NEWLINE dynamic library. If the error is not generated, the NEWLINE application works as expected. If the error is generated, NEWLINE the child window is never loaded in Netscape.
433|Page fault error created when Netscape launched before Windows completes loading|Created by Jean Campbell (jccampb@mitre.org) on Tuesday, June 9, 1998 11:41:42 AM PDT NEWLINE Additional Details : NEWLINE Even when Windows 95 indicates that it is ready to launch NEWLINE applications (i.e., the hour-glass turns to arrow), if NEWLINE insufficient time has passed before launching Netscape, the NEWLINE following error will be generated every time: NEWLINE NEWLINE "This program has performed an illegal operation and will be NEWLINE shut down. If the problem persists, contact the program NEWLINE vendor. NEWLINE NEWLINE NETSCAPE caused an invalid page fault in module KERNEL32.DLL NEWLINE at 0137:bff780f4." NEWLINE NEWLINE The version of Netscape being used is Navigator Gold 3.01. NEWLINE Updated by Patrick Pollet (Patrick.pollet@insa-lyon.fr) on Thursday, July 2, 1998 9:51:11 AM PDT NEWLINE Additional Details : NEWLINE Hi, NEWLINE Never had it with Netscape 3.01 Gold. BUT NEWLINE NEWLINE I am getting a similar problem with communicator 4.05 Fr NEWLINE and the new release (4.05 english with JDK 1.1 support). NEWLINE NEWLINE Communicator is installed in a Netware file server and is NEWLINE launched by Netware Application Launcher . Makes no difference NEWLINE actually. I get the same when double-clicking on the EXE file. NEWLINE NEWLINE This GPF happens BEFORE the splash screen shows up. Thus it is NEWLINE not a problem of cache or bad netscape.hst or bad user prefs NEWLINE as I have red in NUGIES. NEWLINE NEWLINE If I just ignore the Windows dialog Box , Communicator DO load NEWLINE and run properly. I can then close the GPF dialog box,after Communictaor NEWLINE is fully loaded, and it keeps running ... and is functionnal... NEWLINE NEWLINE It looks like a GPF in a "side thread" launched at startup. NEWLINE I did try to use a lower screen resolution (down to 640*349 * 16 colors) NEWLINE since I initially though it was a problem with the video drivers NEWLINE (S2 trio V3), but no difference. NEWLINE NEWLINE This error only happens the first time a user launch Communicator NEWLINE after logging in. The second run is OK... but if we stops the PC or NEWLINE simply logout-login, we get the very same GPF at the first run !!! NEWLINE It tried Dr Watson, but I never got any message. I tried to NEWLINE launch a DLL spy program, but I never found any extra DLL in memory NEWLINE whether I got or not the GPF... NEWLINE NEWLINE I noticed that the mouse cursor always stays as "hour glass" ,until I NEWLINE press the mouse button, whether I get or not the GPF. NEWLINE NEWLINE If I simply copy the Communicator directory on the local disk (or install NEWLINE it there), I never get the problem... NEWLINE NEWLINE Looks like Windows is " getting impatient" about Communicator NEWLINE showing up and try to kill it ... why ? NEWLINE NEWLINE NEWLINE NEWLINE Here is what I get on a 16Mb RAM PC ( win 95B, with kernel update NEWLINE and service pack 1). Never had any problem on machines with 32 Mb NEWLINE of RAM , whatever Windows version we are using... NEWLINE NEWLINE NETSCAPE a causé une exception c0000006H dans le module KERNEL32.DLL à NEWLINE 0137:bff95db6. NEWLINE Registres : NEWLINE EAX=c00197e4 CS=0137 EIP=bff95db6 EFLGS=00000202 NEWLINE EBX=6000e180 SS=013f ESP=0117fd9c EBP=0117ff80 NEWLINE ECX=c00197e8 DS=013f ESI=81567ee0 FS=406f NEWLINE EDX=00000044 ES=013f EDI=815647b8 GS=0000 NEWLINE Octets à CS : EIP : NEWLINE 8b 73 04 8b 43 10 2b 33 8b 11 03 c6 50 ff 72 18 NEWLINE Etat de la pile : NEWLINE 00000008 815647b8 81567ee0 81567ee0 bff95ebc 81567ee0 6000e180 815647b8 81567ee0 NEWLINE bff7b942 81564b18 6000e180 00000002 00000001 81564b18 00000040 NEWLINE NEWLINE sometimes I get this one: NEWLINE NEWLINE NETSCAPE a causé une exception c0000006H dans le module JRT3240.DLL à NEWLINE 0137:600a8240. NEWLINE Registres : NEWLINE EAX=600a8240 CS=0137 EIP=600a8240 EFLGS=00000a82 NEWLINE EBX=81563834 SS=013f ESP=0128fdb8 EBP=0128ff80 NEWLINE ECX=c420405c DS=013f ESI=81563878 FS=4127 NEWLINE EDX=c00197ec ES=013f EDI=60070000 GS=0000 NEWLINE Octets à CS : EIP : NEWLINE 53 b8 01 00 00 00 8b 5c 24 0c 56 57 85 db 55 75 NEWLINE Etat de la pile : NEWLINE bff7b9b5 60070000 00000002 00000000 60070000 81563878 81563834 81560f4c 00000040 NEWLINE 81563834 c420405c c420405c bff78028 bff62482 01287000 bff7bab0
434|Communicator does not work with PowerBuilder plugin.|Created by Jean Campbell (jccampb@mitre.org) on Tuesday, June 9, 1998 11:47:20 AM PDT NEWLINE Additional Details : NEWLINE Navigator Gold 3.01 will work fine with PowerBuilder NEWLINE 6.0 plugin (usually), but the same plug-in will not work NEWLINE with Netscape Communicator, any version. We don't know if NEWLINE the problem resides on the PowerBuilder side or the NEWLINE Communicator side, or both.
435|snprintf prototype mismatch|Created by Soren Juul Moller (sjm@nwc.dk) on Tuesday, June 9, 1998 2:23:49 PM PDT NEWLINE Additional Details : NEWLINE On SunOS5.6 and SunOS5.6_x86 the snprintf prototype NEWLINE defined in mozilla/dbm/src/snprintf.c does not match NEWLINE the prototype defined in /usr/include/stdio.h NEWLINE Fix: Add #ifndef HAVE_SNPRINTF at top and bottom and NEWLINE and -DHAVE_SNPRINTF in mozilla/config/SunOS5.6*.mk NEWLINE Updated by Brian Ostrom (briano@netscape.com) on Monday, July 6, 1998 8:17:37 PM PDT NEWLINE Additional Details : NEWLINE The HAVE_SNPRINTF macro is now defined properly in NEWLINE mozilla/config/SunOS5.mk. Now snprintf.c needs to NEWLINE be dealt with. Reassigning to Lou. NEWLINE Updated by Soren Juul Moller (sjm@nwc.dk) on Tuesday, July 14, 1998 11:30:52 AM PDT NEWLINE Additional Details : NEWLINE Please refer to NEWLINE ftp://ftp.nwc.dk/pub/patches/mozilla/SunOS5.6_x86-current.txt
436|Doesn't display some GIFs|Created by (rob@gadgetguru.com) on Tuesday, June 9, 1998 2:40:25 PM PDT NEWLINE Additional Details : NEWLINE Check out http://www.gadgetguru.com/ - on the menu on the NEWLINE left hand side, the very top and bottom GIFs show the NEWLINE "broken image" icon, everything else displays correctly. NEWLINE It can't seem to find those two images, even though they NEWLINE display properly in other browsers. This has been present NEWLINE since the first source release. I'm currently using today's NEWLINE (06/08/98) build. NEWLINE Updated by (rob@gadgetguru.com) on Wednesday, June 10, 1998 11:14:04 AM PDT NEWLINE Additional Details : NEWLINE Additional investigation shows that Mozilla is having a problem with the NEWLINE redirections given by Apache's mod_speling module. The graphic mtop.gif was NEWLINE incorrectly referenced on the page as MTop.gif, mod_speling attempted to correct NEWLINE the error and redirect to the correct image, but Mozilla didn't take the NEWLINE redirection.
437|<DIV ALIGN=center> doesn't work|Created by (rob@gadgetguru.com) on Tuesday, June 9, 1998 2:42:28 PM PDT NEWLINE Additional Details : NEWLINE Check out http://www.gadgetguru.com/TOPSTORY.HTML - the NEWLINE picture in the story *should* be centered. Using today's NEWLINE (06/08/98) build.
438|Incorrect anonymous FTP login|Created by (chrisputtick@writeme.com) on Wednesday, June 10, 1998 12:06:46 PM PDT NEWLINE Additional Details : NEWLINE Mozilla sends mozilla@ as an anonymous FTP login, similar to NEWLINE IE which send IE@... It's a breach of netiquette, and, well NEWLINE configured (i.e. secure!) FTP servers will refuse entry if a NEWLINE email address is not sent. NEWLINE Updated by Garrett Arch Blythe (blythe@netscape.com) on Wednesday, June 10, 1998 1:48:30 PM PDT NEWLINE Additional Details : NEWLINE The problem is worse when dialing with a LITE version of the browser (browser NEWLINE only). There is no UI for entering the email ID of the person, but the pref is NEWLINE in the advanced pane....
439|Testing Bugzilla email option Please ignore|Created by Arun Ramachandran (arunr@netscape.com) on Wednesday, June 10, 1998 12:24:44 PM PDT NEWLINE Additional Details : NEWLINE Please ignore this bug. NEWLINE NEWLINE Testing if Bugzilla is sending mail when a new bug NEWLINE is created to both reporter and assigned to. NEWLINE NEWLINE Updated by Arun Ramachandran (arunr@netscape.com) on Wednesday, June 10, 1998 12:26:20 PM PDT NEWLINE Additional Details : NEWLINE Please ignore this again NEWLINE Updated by Arun Ramachandran (arunr@netscape.com) on Wednesday, June 10, 1998 1:41:14 PM PDT NEWLINE Additional Details : NEWLINE update again testing if its is closing
440|Testing -- Bugzilla testing|Created by Arun Ramachandran (arunr@netscape.com) on Wednesday, June 10, 1998 1:55:11 PM PDT NEWLINE Additional Details : NEWLINE TEsting email notification Pleaes ignore NEWLINE NEWLINE I will close this Bug
441|testing|Created by Arun Ramachandran Test -1 (arunr-test1@netscape.com) on Wednesday, June 10, 1998 2:03:38 PM PDT NEWLINE Additional Details : NEWLINE testing please ignore
442|testing|Created by Arun Ramachandran Test -1 (arunr-test1@netscape.com) on Wednesday, June 10, 1998 2:06:51 PM PDT NEWLINE Additional Details : NEWLINE testing if email notification is being received while NEWLINE creating new bugs NEWLINE Updated by Arun Ramachandran (T1) (arunr@test.com) on Wednesday, June 10, 1998 3:55:31 PM PDT NEWLINE Additional Details : NEWLINE testing agian from somebodyyyyyy NEWLINE Updated by Arun Ramachandran (T1) (arunr@test.com) on Wednesday, June 10, 1998 3:57:36 PM PDT NEWLINE Additional Details : NEWLINE Updated by Arun Ramachandran (arunr@netscape.com) on Thursday, June 11, 1998 12:02:42 PM PDT NEWLINE Additional Details : NEWLINE closing this bug
443|Bugzilla testing|Created by Arun Ramachandran Test -2 (arunr-test2@netscape.com) on Wednesday, June 10, 1998 2:09:53 PM PDT NEWLINE Additional Details : NEWLINE Bugzilla test NEWLINE Updated by Arun Ramachandran Test -2 (arunr-test2@netscape.com) on Wednesday, June 10, 1998 2:10:28 PM PDT NEWLINE Additional Details : NEWLINE closing this bug NEWLINE Updated by Arun Ramachandran Test -1 (arunr-test1@netscape.com) on Wednesday, June 10, 1998 2:13:27 PM PDT NEWLINE Additional Details : NEWLINE addtion additional comments to the bug NEWLINE Updated by Arun Ramachandran (T1) (arunr@test.com) on Wednesday, June 10, 1998 3:54:37 PM PDT NEWLINE Additional Details : NEWLINE Update by test NEWLINE Updated by Arun Ramachandran (T1) (arunr@test.com) on Wednesday, June 10, 1998 4:21:36 PM PDT NEWLINE Additional Details : NEWLINE addition comments added NEWLINE again NEWLINE pleaes ignore hti mail NEWLINE Updated by Arun Ramachandran (arunr@netscape.com) on Thursday, June 11, 1998 11:50:33 AM PDT NEWLINE Additional Details : NEWLINE addtion commets addt to NEWLINE check if the performance has increased. NEWLINE NEWLINE NEWLINE Arun
444|bugzilla testing|Created by Arun Ramachandran (T1) (arunr@test.com) on Wednesday, June 10, 1998 2:39:36 PM PDT NEWLINE Additional Details : NEWLINE Bugzilla priority testing NEWLINE Updated by Arun Ramachandran (T1) (arunr@test.com) on Wednesday, June 10, 1998 2:53:19 PM PDT NEWLINE Additional Details : NEWLINE addtion comments NEWLINE NEWLINE Testing NEWLINE Updated by Arun Ramachandran (arunr@netscape.com) on Thursday, June 11, 1998 12:01:09 PM PDT NEWLINE Additional Details : NEWLINE closing this bug
445|RDF_BRANCH_060398 xmlglue.c compile error|Created by Ronan Waide (waider@waider.ie) on Thursday, June 11, 1998 10:45:24 AM PDT NEWLINE Additional Details : NEWLINE xmlglue.h wasn't kept up to date with xmlglue.c, here's NEWLINE the patch: NEWLINE NEWLINE Index: modules/xml/glue/xmlglue.h NEWLINE =================================================================== NEWLINE RCS file: /cvsroot/mozilla/modules/xml/glue/xmlglue.h,v NEWLINE retrieving revision 1.6.4.1 NEWLINE diff -c -r1.6.4.1 xmlglue.h NEWLINE *** xmlglue.h 1998/06/11 02:41:16 1.6.4.1 NEWLINE --- xmlglue.h 1998/06/11 17:45:29 NEWLINE *************** NEWLINE *** 203,209 **** NEWLINE void xmlhtml_complete_int NEWLINE (XMLFile xml); NEWLINE void xmlhtml_GetUrlExitFunc NEWLINE (URL_Struct *urls, int status, MWContext *cx); NEWLINE void readHTML (char* url, NEWLINE XMLHTMLInclusion ss); NEWLINE ! NET_StreamClass * XML_ConverterInt(FO_Present_Types NEWLINE format_out, void *data_object, URL_Struct *URL_s, MWContext NEWLINE *window_id); NEWLINE NEWLINE PUBLIC NET_StreamClass * NEWLINE XML_XMLConverter(FO_Present_Types format_out, void NEWLINE *data_object, URL_Struct *URL_s, MWContext *window_id); NEWLINE PUBLIC NET_StreamClass * NEWLINE XML_CSSConverter(FO_Present_Types format_out, void NEWLINE *data_object, URL_Struct *URL_s, MWContext *window_id); NEWLINE --- 203,209 ---- NEWLINE void xmlhtml_complete_int NEWLINE (XMLFile xml); NEWLINE void xmlhtml_GetUrlExitFunc NEWLINE (URL_Struct *urls, int status, MWContext *cx); NEWLINE void readHTML (char* url, NEWLINE XMLHTMLInclusion ss); NEWLINE ! NET_StreamClass * XML_ConverterInt(FO_Present_Types NEWLINE format_out, void *data_object, URL_Struct *URL_s, MWContext NEWLINE *window_id, int16 fileType); NEWLINE NEWLINE PUBLIC NET_StreamClass * NEWLINE XML_XMLConverter(FO_Present_Types format_out, void NEWLINE *data_object, URL_Struct *URL_s, MWContext *window_id); NEWLINE PUBLIC NET_StreamClass * NEWLINE XML_CSSConverter(FO_Present_Types format_out, void NEWLINE *data_object, URL_Struct *URL_s, MWContext *window_id);
446|testing|Created by Arun Ramachandran (arunr@netscape.com) on Thursday, June 11, 1998 11:59:00 AM PDT NEWLINE Additional Details : NEWLINE testing NEWLINE again NEWLINE gagina NEWLINE gasfsdfasdf NEWLINE NEWLINE a NEWLINE sdfasdf NEWLINE Updated by Arun Ramachandran (arunr@netscape.com) on Thursday, June 11, 1998 12:00:07 PM PDT NEWLINE Additional Details : NEWLINE happy to see the speed
447|conversion trouble from int to XP_Bool|Created by Kazushi Marukawa (jam@pobox.com) on Friday, June 12, 1998 11:21:07 PM PDT NEWLINE Additional Details : NEWLINE This is a patch to fix the calculation of XP_Bool values. NEWLINE The size of XP_Bool might be equal to the size of char. It NEWLINE means that "XP_Bool bClearCellHeight = iMode & NEWLINE ED_MODE_NO_CELL_HEIGHT;" might be wrong because NEWLINE ED_MODE_NO_CELL_HEIGHT is equal to 0x0800. NEWLINE NEWLINE I changed all converting expressions from int to XP_Bool NEWLINE which I found put them into a following patch. And I NEWLINE suggest you to use "expr ? TRUE : FALSE;" form instead of NEWLINE using "(XP_Bool)int-value" form. NEWLINE NEWLINE I send a patch to netscape.public.mozilla.patches. The NEWLINE message-id is <6lt59a$kpm2@secnews.netscape.com>.
448|testing bugzilla ignore|Created by Prasad Ramachandran (ramachandran.prasad@roche.com) on Saturday, June 13, 1998 6:27:24 PM PDT NEWLINE Additional Details : NEWLINE This bugzilla bug inter NEWLINE Updated by Prasad Ramachandran (ramachandran.prasad@roche.com) on Saturday, June 13, 1998 6:29:28 PM PDT NEWLINE Additional Details : NEWLINE addition comments added by Ramachandran prasad NEWLINE NEWLINE Opps.. testing
449|Crash when loading http://www.bryant.edu/~ssl2/nscrash.html|Created by Steve Lawrance (slawrance@technologist.com) on Sunday, June 14, 1998 7:17:06 AM PDT NEWLINE Additional Details : NEWLINE Netscape crashes while loading the page at NEWLINE http://www.bryant.edu/~ssl2/nscrash.html NEWLINE This page was generated with Microsoft FrontPage '98 and NEWLINE uses Cascading StyleSheet properties in most of the tags. NEWLINE Works fine in Internet Explorer 4. NEWLINE I discovered this while I was improving my site at NEWLINE http://nt4sg1.bryant.edu/home. When I tested in Netscape NEWLINE Communicator, the "Contact Me..." item in the "Window" menu NEWLINE (implemented as a drop-down combo box at the top-right of NEWLINE the page) opens up http://nt4sg1.bryant.edu/home/mail.html NEWLINE where it crashes (this file was copied to www.bryant.edu as NEWLINE nscrash.html because nt4sg1 will be offline for the summer).
450|Kills disk that is saved on|Created by null (Doltish7@hotmail.com) on Monday, June 15, 1998 4:13:36 PM PDT NEWLINE Additional Details : NEWLINE I saved the file to my JAZ disk, it completely screwed up NEWLINE all the FAT files and corrupted all the data on my disk. It NEWLINE was a virus nest to put it plainly. Now I don't know much NEWLINE about all this computer stuff, but I know how to program NEWLINE with Visual Bsic pretty well, and I tell you, this is not NEWLINE normal. I have been able to reproduce the same situation, NEWLINE but I had to delete the file after it started going after my NEWLINE hard drive files. I downloaded it from Download.com.
451|Kills disk that is saved on|Created by Da Hacker (null) on Monday, June 15, 1998 4:13:36 PM PDT NEWLINE Additional Details : NEWLINE I saved the file to my JAZ disk, it completely screwed up NEWLINE all the FAT files and corrupted all the data on my disk. It NEWLINE was a virus nest to put it plainly. Now I don't know much NEWLINE about all this computer stuff, but I know how to program NEWLINE with Visual Bsic pretty well, and I tell you, this is not NEWLINE normal. I have been able to reproduce the same situation, NEWLINE but I had to delete the file after it started going after my NEWLINE hard drive files. I downloaded it from Download.com.
452|user entries in LOCAL_MACHINE instead of CURRENT_USER|Created by (fprefect@schiller.big.ac.at) on Tuesday, June 16, 1998 12:28:40 PM PDT NEWLINE Additional Details : NEWLINE I installed a windows NT network here, with one server and 4 NEWLINE clients. NEWLINE After installing netscape I noticed that netscape stores the NEWLINE specific user information under the LOCAL_MACHINE key in the NEWLINE registry. NEWLINE Since only the CURRENT_USER key gets saved per user this NEWLINE leads to some problems. NEWLINE For example: user x saves the profile under U:\\profiles NEWLINE u: being the mounted user-directory from the server. NEWLINE Now he logs out and another user starts netscape. NEWLINE Since he doesnt have u:\\profiles on his u: drive netscpae NEWLINE asks him to create a profile. He does it..and everything NEWLINE works. But this way you get a pretty large number of NEWLINE profiles on one machine and the worst part is, when the user NEWLINE goes to another machine he has to do it again. NEWLINE NEWLINE A workaround is to tell the users to create a profile with a NEWLINE special name "default" and store it under u:\\netscape. So NEWLINE netscape looks for it on the network drive and finds it. NEWLINE But that way it is impossible for a user to have more than NEWLINE one profile for himself.
453|OnKeyUp event is not stable, sometimes not fire|Created by Yuan-Fu Yu (QUSYUYU@am1.ericsson.se) on Friday, June 19, 1998 2:01:48 PM PDT NEWLINE Additional Details : NEWLINE I use following javascript code to trace OnKeyUp event on NEWLINE Communicator 4.03 sometimes work, sometimes not. NEWLINE NEWLINE <input type="text" name="text2" onKeyUp='alert("Link got an NEWLINE event: "+ event.type)'> NEWLINE NEWLINE or NEWLINE NEWLINE <textarea name="text2" onKeyUp='alert("Link got an event: "+ NEWLINE event.type)'></textarea>
454|Unix: 0x80-0x9F in cp1252 do not display correctly|Created by Tim Eliseo (tim@quiknet.com) on Friday, June 19, 1998 8:48:18 PM PDT NEWLINE Additional Details : NEWLINE Many Web pages use quote characters in the range 0x91-0x94 NEWLINE which are Microsoft codepage 1252 extensions. For X these NEWLINE are currently mapped to the ? character rather than normal NEWLINE quote characters. A patch follows to fix this. Note that NEWLINE sequences such as ‘ are currently mapped properly; this NEWLINE problem only shows up when the actual characters are in the NEWLINE file. NEWLINE NEWLINE --- mozilla/lib/libi18n/sbconvtb.c Sat May 9 03:57:48 1998 NEWLINE +++ mozilla/lib/libi18n/sbconvtb.c.new Fri Jun 19 20:28:19 NEWLINE 1998 NEWLINE @@ -71,7 +71,7 @@ NEWLINE /* Tables for Win CP1252 -> ISO 8859-1 */ NEWLINE PRIVATE unsigned char cp1252_to_iso8859_1[] = { NEWLINE /*8x*/ '?', '?', ',', 'f', '?', '?', '?', '?', '^', '?', NEWLINE 'S', '<', '?', '?', '?', '?', NEWLINE -/*9x*/ '?', '?', '?', '?', '?', '*', '-', '-', '~', '?', NEWLINE 's', '>', '?', '?', '?', 'Y', NEWLINE +/*9x*/ '?', '`', '\\'', '"', '"', '*', '-', '-', '~', '?', NEWLINE 's', '>', '?', '?', '?', 'Y', NEWLINE /*Ax*/ NEWLINE 0xA0,0xA1,0xA2,0xA3,0xA4,0xA5,0xA6,0xA7,0xA8,0xA9,0xAA,0xAB,0xAC,0xAD,0xAE,0xAF, NEWLINE /*Bx*/ NEWLINE 0xB0,0xB1,0xB2,0xB3,0xB4,0xB5,0xB6,0xB7,0xB8,0xB9,0xBA,0xBB,0xBC,0xBD,0xBE,0xBF, NEWLINE /*Cx*/ NEWLINE 0xC0,0xC1,0xC2,0xC3,0xC4,0xC5,0xC6,0xC7,0xC8,0xC9,0xCA,0xCB,0xCC,0xCD,0xCE,0xCF, NEWLINE NEWLINE NEWLINE For those of you like myself annoyed by this bug in the NEWLINE commercial Netscape version, here's a quick fix: NEWLINE NEWLINE adb -w netscape NEWLINE cp1252_to_iso8859_1+0x11?W 0x22222760 NEWLINE ^d NEWLINE NEWLINE This is correct for little-endian architectures.
455|<hr> within <td> doesn't display|Created by Kartik Subbarao (subbarao@computer.org) on Sunday, June 21, 1998 12:31:04 PM PDT NEWLINE Additional Details : NEWLINE Here's an HTML file that doesn't render properly: NEWLINE NEWLINE <html> NEWLINE <table> NEWLINE <tr> <td> <hr> </td> </tr> NEWLINE <tr> <td>test</td> NEWLINE <tr> <td> <hr> </td> </tr> NEWLINE </table> NEWLINE </html> NEWLINE NEWLINE Communicator 4.05 displays the horizontal rules properly, NEWLINE but Mozilla doesn't.
456|Too much whitespace when viewing http://www.mozilla.org using mozilla-1998-06-03|Created by Thorsten Heit (thorsten@wurzelausix.cs.uni-sb.de) on Monday, June 22, 1998 7:33:33 AM PDT NEWLINE Additional Details : NEWLINE Between the table containing the logo-star.gif and the NEWLINE Mozilla News table there's a distance of about 40 times NEWLINE the height of the upper table when using the 1998/06/03 NEWLINE version of Mozilla, started on a Pentium-based NEWLINE RedHat-Linux 5.1 system and displayed on a SGI Indy
457|Improper FTP download dialog box|Created by Thorsten Heit (thorsten@wurzelausix.cs.uni-sb.de) on Monday, June 22, 1998 7:59:07 AM PDT NEWLINE Additional Details : NEWLINE When downloading a file via Mozilla's FTP client a dialog NEWLINE box pops up which is NEWLINE a) much too big (initially 600x480 pixels) NEWLINE and NEWLINE b) doesn't display its contents properly: NEWLINE NEWLINE All the elements in the dialog are aligned at the bottom of NEWLINE the window instead at the top; furthermore the only visible NEWLINE elements are the "Stop" button, half of the text field NEWLINE containing the FTP source file and half of the animated NEWLINE image besides that field. NEWLINE The field containing where the downloaded file will be NEWLINE stored is not visible.
458|Netscape doesn't link: undefined reference to `_Xlcmbcurmax'|Created by Raphael Wegmann (wegmann@psi.co.at) on Monday, June 22, 1998 8:08:15 AM PDT NEWLINE Additional Details : NEWLINE ResConvert.o and Text.o has undefined references to NEWLINE `_Xlcmbcurmax'. This is because Xinsides Motif 2.0 needs NEWLINE libXintl for _Xlcmbcurmax. I fixed it in cmd/xfe/Makefile in NEWLINE line 1035: NEWLINE XTOOLLIB = -L/usr/X11R6/lib -lXt -lSM -lICE -lXintl NEWLINE NEWLINE This problem should at least be documented somewhere.
459|Popup menus lock up Mozilla (1998-06-03) for a while|Created by Thorsten Heit (thorsten@wurzelausix.cs.uni-sb.de) on Monday, June 22, 1998 8:24:15 AM PDT NEWLINE Additional Details : NEWLINE I was downloading a file via FTP in the background, and in NEWLINE the foreground I returned to www.mozilla.org (see Bug ID NEWLINE 456). Right-clicking on the logo-star.gif didn't work NEWLINE directly; the popup menu appeared only when right-clicking NEWLINE on a link on the menu on the left or when clicking somewhere NEWLINE in the menu bar. NEWLINE Then, the menu locks up the whole screen for a while, from NEWLINE two seconds up to more than 10 seconds :-(
460|Cosmetic problem when minimizing all of the three toolbars|Created by Thorsten Heit (thorsten@wurzelausix.cs.uni-sb.de) on Monday, June 22, 1998 8:48:58 AM PDT NEWLINE Additional Details : NEWLINE When minimizing all of the three toolbars I don't see any of NEWLINE the buttons when minimizing only one or two of the toolbars; NEWLINE instead I see the top 10 pixels (or whatever the size of NEWLINE that minimized part is) of the toolbar I minimized last...
461|Shared library conflict when compiling&using under a glibc-2 based Linux system|Created by Thorsten Heit (thorsten@wurzelausix.cs.uni-sb.de) on Monday, June 22, 1998 9:09:52 AM PDT NEWLINE Additional Details : NEWLINE When compiling Mozilla, a library called libutil.so will be NEWLINE created. A glibc-2 based Linux system already has a library NEWLINE with the same name Mozilla won't start because of missing NEWLINE symbols in the glibc-libutil.so. NEWLINE I renamed all references to "libutil.*" and "util" in all NEWLINE makefiles, let the make create a libnsutil.[a,so] instead of NEWLINE libutil.[a,so] and then it worked.
462|Navigator locks dead after connect and parse|Created by Gregory Block (gblock@netscape.com) on Tuesday, June 23, 1998 7:06:55 AM PDT NEWLINE Additional Details : NEWLINE Source pull as of 6/22 AM, go to http://www.suck.com/ NEWLINE NEWLINE Browser should stop dead.
463|Leaves process hanging with memory allocated|Created by Chris Alonso (calonso@cobank.com) on Wednesday, June 24, 1998 12:55:19 PM PDT NEWLINE Additional Details : NEWLINE Killing browser after using large Java applets causes NEWLINE applicatio to die, but proces lingers with memory allocated NEWLINE (often lots (15-20MB)). I believe this also happens in NEWLINE Windows95, with result being GPF's when other applications NEWLINE get allocated those blocks from OS. I have seen this for a NEWLINE while, (several versions), so it is probably a know bug... NEWLINE NEWLINE Chris Alonso NEWLINE IBM NEWLINE Updated by Garrett Arch Blythe (blythe@netscape.com) on Wednesday, June 24, 1998 2:21:32 PM PDT NEWLINE Additional Details : NEWLINE This appears to be a Netscape bug, not a Mozilla bug. NEWLINE Handing over to java people to decide.
464|Add -configdir patch|Created by (mcafee@netscape.com) on Monday, June 29, 1998 1:30:36 AM PDT NEWLINE Additional Details : NEWLINE This bug is to track the -configdir patch so we don't NEWLINE forget to check it in.
465|FTP PORT (active) command support|Created by Christopher Seawood (cls@seawood.org) on Tuesday, June 30, 1998 10:25:16 AM PDT NEWLINE Additional Details : NEWLINE I noticed the function call NET_UsePASV(bool) already exists NEWLINE in network/protocol/ftp/mkftp.c but nobody uses it. As my NEWLINE firewall does not let any random high port connection NEWLINE through, using the ftp feature of Netscape/Mozilla hangs. NEWLINE Updated by (aoki@netscape.com) on Monday, August 17, 1998 9:31:39 AM PDT NEWLINE Additional Details : NEWLINE This would be a netlib issue. Reassigning to gagan
466|"font stacks" and a parsing algorithm bug|Created by Jared Momose (jmomose@casefinder.com) on Tuesday, June 30, 1998 2:30:24 PM PDT NEWLINE Additional Details : NEWLINE There is not a separate section for "logical" bugs, so Ill NEWLINE present my case in this section. NEWLINE NEWLINE This bug is deals with text and attribute changing when a NEWLINE document is displayed on a screen. NEWLINE NEWLINE From what I have seen, when Netscape loads and displays NEWLINE documents on the screen, it sets up a "Font Stack" to do its NEWLINE attribute changing. Say you change the current text in a NEWLINE document to bold; Netscape pushes the current font onto a NEWLINE stack and uses that font + bold for the new font. The NEWLINE problem comes when you have a line like the following. NEWLINE NEWLINE text <B> text <I> text </B> text </I> text NEWLINE NEWLINE When the </B> tag appears, the font stack is popped, and a NEWLINE previous font is retrieved. However, this previous font NEWLINE does not have bold turned off, but rather has italics turned NEWLINE off! If you are following me, take it one step further. NEWLINE Try putting a <FONT COLOR="#mycolor"> in the above html line NEWLINE and modifying it such that a different attribute is turned NEWLINE off before the color is. NEWLINE NEWLINE Another question arises: "who would do such a thing as NEWLINE overlap font attributes?". Well....me! NEWLINE NEWLINE I would love to hear some input from mozilla developers, NEWLINE especially reasons *why* this "Font Stack" was implemented NEWLINE into the browser. (Am I missing something? Is it faster NEWLINE than other techniques for formatting and displaying a doc? NEWLINE Easier?) NEWLINE NEWLINE Jared Momose. NEWLINE Geronimo Developement Corp.
467|Undefined references in $CODEROOT/nsprpub/pr/src/linking/prlink.c|Created by Sean Chitwood (chaos_mage@bigfoot.com) on Tuesday, June 30, 1998 10:59:44 PM PDT NEWLINE Additional Details : NEWLINE I get the following errors while trying to make RegFactory NEWLINE in the xpcom/tests directory NEWLINE NEWLINE NEWLINE NEWLINE NEWLINE NEWLINE ../../dist/Linux2.0.30_ppc_DBG.OBJ/lib/ NEWLINE libnspr21.a(prlink.o): In function `DLLErrorInternal': NEWLINE /code/mozilla/nsprpub/pr/src/linking/prlink.c:103: undefined NEWLINE reference to `dlerror' NEWLINE ../../dist/Linux2.0.30_ppc_DBG.OBJ/lib/ NEWLINE libnspr21.a(prlink.o): In function `_PR_InitLinker': NEWLINE /code/mozilla/nsprpub/pr/src/linking/prlink.c:151: undefined NEWLINE reference to `dlopen' NEWLINE ../../dist/Linux2.0.30_ppc_DBG.OBJ/lib/ NEWLINE libnspr21.a(prlink.o): In function `PR_LoadLibrary': NEWLINE /code/mozilla/nsprpub/pr/src/linking/prlink.c:634: undefined NEWLINE reference to `dlopen' NEWLINE ../../dist/Linux2.0.30_ppc_DBG.OBJ/lib/ NEWLINE libnspr21.a(prlink.o): In function `PR_UnloadLibrary': NEWLINE /code/mozilla/nsprpub/pr/src/linking/prlink.c:711: undefined NEWLINE reference to `dlclose' NEWLINE ../../dist/Linux2.0.30_ppc_DBG.OBJ/lib/ NEWLINE libnspr21.a(prlink.o): In function `pr_FindSymbolInLib': NEWLINE /code/mozilla/nsprpub/pr/src/linking/prlink.c:822: undefined NEWLINE reference to `dlsym'
468|[PP]JavaScript sometimes thinks form text is empty|Created by (ocrat@ocrat.com) on Friday, July 3, 1998 4:20:55 PM PDT NEWLINE Additional Details : NEWLINE Mozilla version: CVS snapshot, July 3 1998. NEWLINE Clicking the Show button gives 0, not 5, for NEWLINE the length of text field string "hello". NEWLINE NEWLINE NEWLINE <BODY> NEWLINE <FORM onSubmit="return false;"> NEWLINE NEWLINE <INPUT TYPE="button" VALUE="Show" NEWLINE onClick='alert(this.form.cinput.value.length); return NEWLINE true;'> NEWLINE NEWLINE <INPUT TYPE="text" NAME="cinput" VALUE="hello" NEWLINE SIZE="16"> NEWLINE NEWLINE </FORM> NEWLINE </BODY> NEWLINE Updated by (ocrat@ocrat.com) on Friday, July 3, 1998 4:27:07 PM PDT NEWLINE Additional Details : NEWLINE In case it's somehow relevant, I'm using lesstif-0.85.3 NEWLINE Updated by Mike McCabe (mccabe@netscape.com) on Tuesday, July 7, 1998 3:18:55 PM PDT NEWLINE Additional Details : NEWLINE This seems to be a valid bug; I've been able to duplicate it in a recent Mozilla NEWLINE build. It doesn't seem to show up in 4.05. Thanks for the report! NEWLINE Updated by Mike McCabe (mccabe@netscape.com) on Tuesday, July 7, 1998 3:49:23 PM PDT NEWLINE Additional Details : NEWLINE mlm has confirmed that the bug doesn't show up in 4.06 either, and NEWLINE javascript/libmocha changes from 4.06 are being migrated to Mozilla as soon as NEWLINE we have a stability thaw, so it seems likely that this problem my go away with NEWLINE those changes. Until then, I'll keep an eye on it.
469|mcom_db.h is looking for endian.h in the wrong place|Created by Scott Butler (r-scott.butler@usa.dupont.com) on Friday, July 3, 1998 6:26:14 PM PDT NEWLINE Additional Details : NEWLINE In mozilla/include/mcom_db.h and NEWLINE mozilla/dbm/include/mcom_db.h under the conditional NEWLINE #ifdef __alpha NEWLINE NEWLINE the code is #include <endian.h> NEWLINE NEWLINE and (at least on my system) needs to be NEWLINE NEWLINE #include <machine/endian.h> NEWLINE NEWLINE Sorry if this is the wrong 'area'; I'm not sure where to put NEWLINE this. NEWLINE Updated by Brian Ostrom (briano@netscape.com) on Monday, July 13, 1998 5:54:34 PM PDT NEWLINE Additional Details : NEWLINE This is a duplicate of bug#249. I have a "fix", but I can't NEWLINE test it until we get our DEC licenses straightened out. NEWLINE What I propose is a single change to mozilla/config/OSF1.mk: NEWLINE NEWLINE OS_INCLUDES = -I/usr/include -I/usr/include/sys -I/usr/include/machine NEWLINE NEWLINE I don't know if the first two need to be there, but I wanted NEWLINE to be safe....
470|File|Created by Scott Butler (r-scott.butler@usa.dupont.com) on Friday, July 3, 1998 6:38:37 PM PDT NEWLINE Additional Details : NEWLINE An attempt to compile mozilla/modules/libpref/src/prefapi.c NEWLINE fails because it can't find sechash.h. There are a number of NEWLINE sec*.h files in ../../security/freenav. If you copy those NEWLINE into libpref/src you get a clean compile but I have no idea NEWLINE if they're the correct ones. NEWLINE Updated by (briano@netscape.com) on Monday, July 6, 1998 7:44:17 PM PDT NEWLINE Additional Details : NEWLINE You have to run 'make export' from the mozilla directory NEWLINE before most of the modules will build successfully.
471|More about bug 470|Created by Scott Butler (r-scott.butler@usa.dupont.com) on Friday, July 3, 1998 8:01:38 PM PDT NEWLINE Additional Details : NEWLINE I suspect that the problem is that the 'dist' directory NEWLINE doesn't have a 'security' directory in it. I noticed that NEWLINE when attempting to build NEWLINE NEWLINE mozilla/modules/libpref/src/prefapi.c NEWLINE NEWLINE the compile command has a -I../dist/security element but NEWLINE there is no corresponding directory in 'dist'. If I create NEWLINE one and copy the contents of .../module/security/freenav/ NEWLINE then things seem to compile much better. NEWLINE NEWLINE I have no way of knowing if those are the CORRECT .h files NEWLINE but they do make the error messages go away. NEWLINE Updated by (briano@netscape.com) on Monday, July 6, 1998 7:45:39 PM PDT NEWLINE Additional Details : NEWLINE make export NEWLINE NEWLINE See the comments for bug 470.
472|BUILD: g++ reports compile error in context_funcs.cpp|Created by Scott Butler (r-scott.butler@usa.dupont.com) on Saturday, July 4, 1998 1:05:12 PM PDT NEWLINE Additional Details : NEWLINE ... context_funcs.cpp NEWLINE NEWLINE /usr/include/sys/statvfs.h:75: previous declaration of `int NEWLINE statvfs(const char * NEWLINE , struct statvfs *)' with C++ linkage NEWLINE context_funcs.cpp:78: conflicts with new declaration with C NEWLINE linkage NEWLINE make[3]: *** [OSF1V4.0_DBG.OBJ/context_funcs.o] Error 1 NEWLINE NEWLINE ghidorah.es.dupont.com> gcc -v NEWLINE Reading specs from NEWLINE /usr/local/lib/gcc-lib/alpha-dec-osf4.0/2.7.2.3/specs NEWLINE gcc version 2.7.2.3 NEWLINE ghidorah.es.dupont.com> make -v NEWLINE GNU Make version 3.75, by Richard Stallman and Roland NEWLINE McGrath. NEWLINE Updated by (briano@netscape.com) on Monday, July 6, 1998 7:50:22 PM PDT NEWLINE Additional Details : NEWLINE This looks like an XFE problem. Reassigning to Ramiro.
473|Bogus "Out Of Memory" error on mail send when no NIS passwds|Created by James Beckett (jmb@ctxuk.citrix.com) on Tuesday, July 7, 1998 8:34:10 AM PDT NEWLINE Additional Details : NEWLINE System: RedHat Linux 5.0 (libc6) with libc5 shlibs for NEWLINE alleged compatibility. NIS passwords used. Netscape NEWLINE release 4.05, linked against libc5. NEWLINE NEWLINE Netscape refuses to send mail, claiming ".. is out of NEWLINE memory; close some applications blah blah blah". NEWLINE NEWLINE cmd/xfe/dialogs.c: fe_DefaultUserInfo() is calling NEWLINE getpwuid(geteuid) which is giving NULL, as libc5 does not NEWLINE support NIS (only local passwd entries are returned). NEWLINE NEWLINE cmd/xfe/xfe.c: FE_UsersRealMailAddress() is caching "" as NEWLINE the user's mail address (eg should be "jmb") NEWLINE NEWLINE lib/libnet/mksmtp.c: net_smtp_send_data_response(...) NEWLINE is passing (real_name == "") to MSG_MakeFullAddress() NEWLINE which returns NULL, and the next lines dislikes this NEWLINE and return with MK_OUT_OF_MEMORY set in CE_URL_S. NEWLINE NEWLINE Yes, I have found a version (RPM) linked against libc6 NEWLINE now, but the error message is definitely wrong for the NEWLINE circumstance. NEWLINE NEWLINE And it wasn't easy, when the messenger source hasn't been NEWLINE released yet - ie I couldn't build a debug version, I've NEWLINE been grovelling in the disassembly with gdb... NEWLINE NEWLINE Enjoy.
474|mozilla crashes at http://docs.sun.com/mozilla/disability.xml|Created by Michael Rohleder (mike@rohleder.de) on Friday, July 10, 1998 8:55:59 AM PDT NEWLINE Additional Details : NEWLINE #0 0x8336a5a in PA_CloneMDLTag (src=0x0) at pa_mdl.c:78 NEWLINE 78 tag->type = src->type; NEWLINE (gdb) bt NEWLINE #0 0x8336a5a in PA_CloneMDLTag (src=0x0) at pa_mdl.c:78 NEWLINE #1 0x8289e7d in lo_TeardownList (context=0x8efda00, NEWLINE state=0x8f04e00, tag=0x0) NEWLINE at laytags.c:2724 NEWLINE #2 0x8259dd1 in LO_PopStyleTagByIndex (context=0x8efda00, NEWLINE state=0xbfffdd34, NEWLINE tag_type=19 '\\023', index=0) at laystyle.c:736
475|[LAYER] Mozilla will not render all of http://slashdot.org|Created by Jeremy Sholovitz (jeremy@rowcure.ml.org) on Sunday, July 12, 1998 10:13:00 AM PDT NEWLINE Additional Details : NEWLINE Mozilla renders the top of the page fine, but nothing after NEWLINE that. It all seems to simply get lost. NEWLINE Updated by (kit@connectnet.com) on Thursday, July 16, 1998 1:17:50 AM PDT NEWLINE Additional Details : NEWLINE If you go back to "Netsite:" and then hit enter it is able to render the page NEWLINE correctly. Weblint comes up with 229 errors on this page. There are table tags NEWLINE opened and not closed etc. This does appear to be part of a general problem I NEWLINE see where a page with tables his hidden areas unless I go back to the page again NEWLINE or hit reload.
476|Core dump on startup of moz-export|Created by Torsten Rüger (torsten@ponton-hamburg.de) on Monday, July 13, 1998 4:56:09 AM PDT NEWLINE Additional Details : NEWLINE I'm getting a floating point exception in js/src/jsapi.c NEWLINE line 1184 because the check JSDOUBLE_IS_NaN does not NEWLINE recognise the variable as being Nan 0xfffffffffffff NEWLINE Updated by Mike McCabe (mccabe@netscape.com) on Monday, July 13, 1998 1:18:59 PM PDT NEWLINE Additional Details : NEWLINE Thanks for looking at this; other people have noticed the same problem with NEWLINE builds on alpha linux and bsd - platforms that raise SIGFPE on int casts of NEWLINE NaN values. NEWLINE NEWLINE I have a fix in hand, and it's checked into the javascript sources in js/ref NEWLINE (from which js/src is generated; see READMEs in js/ref and js/src.) The fix NEWLINE is to change JSDOUBLE_IS_INT(d, i) to conditionalize the (jsint) cast on d NEWLINE being non-NaN, plus some removed casts and restructuring. NEWLINE NEWLINE I'm putting off propagating it into js/src for the time being, because the NEWLINE client is currently in a stability freeze for the time being, and it's not a NEWLINE critical fix. I'll close this bug as soon as I can propagate the fixes to NEWLINE js/src, and confirm that they correct the crash.
477|Debug build on SunOS5_x86|Created by Soren Juul Moller (sjm@nwc.dk) on Tuesday, July 14, 1998 11:28:21 AM PDT NEWLINE Additional Details : NEWLINE One need to use `-Wa,-s -gstabs´ on Solaris 2.x X86 if using NEWLINE gcc as compiler and want to debug Mozilla. NEWLINE This is done by setting OPTIMIZER=-Wa,-s -gstabs NEWLINE Please refer to NEWLINE ftp://ftp.nwc.dk/pub/patches/mozilla/SunOS5.6_x86-current.tx NEWLINE t NEWLINE Updated by Brian Ostrom (briano@netscape.com) on Tuesday, July 14, 1998 4:27:43 PM PDT NEWLINE Additional Details : NEWLINE The OPTIMIZER change is now in place (mozilla/config/SunOS5.mk:3.7). NEWLINE I still can't get to ftp.nwc.dk. You're welcome to uuencode it and NEWLINE mail it to me, if you wish....
478|"Xm/Xm.h" Motif problems/does not compile|Created by David Quintero (dquinter@cs.ucr.edu) on Tuesday, July 14, 1998 2:22:55 PM PDT NEWLINE Additional Details : NEWLINE ********************************************************************* NEWLINE THESE ARE THE ERRORS I GET WHILE TRYING TO COMPILE NETSCAPE. NEWLINE NEWLINE I GET THEM IN THE DIRECTORY: NEWLINE mozilla/cmd/xfe/XfeWidgets/Xfe/ NEWLINE NEWLINE PLEASE LET ME KNOW WHAT IT IS THAT I AM MISSING. NEWLINE THANK YOU NEWLINE NEWLINE ********************************************************************* NEWLINE NEWLINE NEWLINE In file included from NEWLINE ../../../../dist/public/XfeWidgets/Xfe/Xfe.h:30, NEWLINE from NEWLINE ../../../../dist/public/XfeWidgets/Xfe/Arrow.h:30, NEWLINE from NEWLINE ../../../../dist/public/XfeWidgets/Xfe/ArrowP.h:30, NEWLINE from Arrow.c:27: NEWLINE ../../../../dist/public/XfeWidgets/Xfe/Xm/Xm.h:39: NEWLINE Xm/XmStrDefs.h: No such NEWLINE file or directory NEWLINE ../../../../dist/public/XfeWidgets/Xfe/Xm/Xm.h:40: NEWLINE Xm/VirtKeys.h: No such NEWLINE file or directory NEWLINE ../../../../dist/public/XfeWidgets/Xfe/Xm/Xm.h:1484: NEWLINE Xm/VendorS.h: No such NEWLINE file or directory NEWLINE In file included from NEWLINE ../../../../dist/public/XfeWidgets/Xfe/Xfe.h:37, NEWLINE from NEWLINE ../../../../dist/public/XfeWidgets/Xfe/Arrow.h:30, NEWLINE from NEWLINE ../../../../dist/public/XfeWidgets/Xfe/ArrowP.h:30, NEWLINE from Arrow.c:27: NEWLINE ../../../../dist/public/XfeWidgets/Xfe/Draw.h:30: Xm/Xm.h: NEWLINE No such file or NEWLINE directory NEWLINE In file included from NEWLINE ../../../../dist/public/XfeWidgets/Xfe/Xfe.h:38, NEWLINE from NEWLINE ../../../../dist/public/XfeWidgets/Xfe/Arrow.h:30, NEWLINE from NEWLINE ../../../../dist/public/XfeWidgets/Xfe/ArrowP.h:30, NEWLINE from Arrow.c:27: NEWLINE ../../../../dist/public/XfeWidgets/Xfe/Find.h:30: Xm/Xm.h: NEWLINE No such file or NEWLINE directory NEWLINE In file included from NEWLINE ../../../../dist/public/XfeWidgets/Xfe/Xfe.h:39, NEWLINE from NEWLINE ../../../../dist/public/XfeWidgets/Xfe/Arrow.h:30, NEWLINE from NEWLINE ../../../../dist/public/XfeWidgets/Xfe/ArrowP.h:30, NEWLINE from Arrow.c:27: NEWLINE ../../../../dist/public/XfeWidgets/Xfe/Geometry.h:30: NEWLINE Xm/Xm.h: No such NEWLINE file or directory NEWLINE In file included from NEWLINE ../../../../dist/public/XfeWidgets/Xfe/Xfe.h:40, NEWLINE from NEWLINE ../../../../dist/public/XfeWidgets/Xfe/Arrow.h:30, NEWLINE from NEWLINE ../../../../dist/public/XfeWidgets/Xfe/ArrowP.h:30, NEWLINE from Arrow.c:27: NEWLINE ../../../../dist/public/XfeWidgets/Xfe/ListUtil.h:30: NEWLINE Xm/Xm.h: No such NEWLINE file or directory NEWLINE In file included from NEWLINE ../../../../dist/public/XfeWidgets/Xfe/Xfe.h:41, NEWLINE from NEWLINE ../../../../dist/public/XfeWidgets/Xfe/Arrow.h:30, NEWLINE from NEWLINE ../../../../dist/public/XfeWidgets/Xfe/ArrowP.h:30, NEWLINE from Arrow.c:27: NEWLINE ../../../../dist/public/XfeWidgets/Xfe/MenuUtil.h:30: NEWLINE Xm/Xm.h: No such NEWLINE file or directory NEWLINE In file included from NEWLINE ../../../../dist/public/XfeWidgets/Xfe/Xfe.h:42, NEWLINE from NEWLINE ../../../../dist/public/XfeWidgets/Xfe/Arrow.h:30, NEWLINE from NEWLINE ../../../../dist/public/XfeWidgets/Xfe/ArrowP.h:30, NEWLINE from Arrow.c:27: NEWLINE ../../../../dist/public/XfeWidgets/Xfe/RepType.h:30: NEWLINE Xm/Xm.h: No such file NEWLINE or directory NEWLINE In file included from NEWLINE ../../../../dist/public/XfeWidgets/Xfe/Xfe.h:43, NEWLINE from NEWLINE ../../../../dist/public/XfeWidgets/Xfe/Arrow.h:30, NEWLINE from NEWLINE ../../../../dist/public/XfeWidgets/Xfe/ArrowP.h:30, NEWLINE from Arrow.c:27: NEWLINE ../../../../dist/public/XfeWidgets/Xfe/Resources.h:30: NEWLINE Xm/Xm.h: No such NEWLINE file or directory NEWLINE In file included from NEWLINE ../../../../dist/public/XfeWidgets/Xfe/Xfe.h:45, NEWLINE from NEWLINE ../../../../dist/public/XfeWidgets/Xfe/Arrow.h:30, NEWLINE from NEWLINE ../../../../dist/public/XfeWidgets/Xfe/ArrowP.h:30, NEWLINE from Arrow.c:27: NEWLINE ../../../../dist/public/XfeWidgets/Xfe/StringUtil.h:30: NEWLINE Xm/Xm.h: No such NEWLINE file or directory NEWLINE In file included from NEWLINE ../../../../dist/public/XfeWidgets/Xfe/PrimitiveP.h:30, NEWLINE from NEWLINE ../../../../dist/public/XfeWidgets/Xfe/LabelP.h:31, NEWLINE from NEWLINE ../../../../dist/public/XfeWidgets/Xfe/ButtonP.h:31, NEWLINE from NEWLINE ../../../../dist/public/XfeWidgets/Xfe/ArrowP.h:31, NEWLINE from Arrow.c:27: NEWLINE ../../../../dist/public/XfeWidgets/Xfe/PrimitiveP.h:34: NEWLINE Xm/DrawP.h: No NEWLINE such file or directory NEWLINE In file included from Arrow.c:28: NEWLINE ../../../../dist/public/XfeWidgets/Xfe/ManagerP.h:35: NEWLINE Xm/ManagerP.h: No NEWLINE such file or directory NEWLINE In file included from Arrow.c:29: NEWLINE Xm/RowColumnP.h:25: Xm/RowColumn.h: No such file or NEWLINE directory NEWLINE Xm/RowColumnP.h:26: Xm/ManagerP.h: No such file or directory NEWLINE In file included from NEWLINE ../../../../dist/public/XfeWidgets/Xfe/PrimitiveP.h:30, NEWLINE from NEWLINE ../../../../dist/public/XfeWidgets/Xfe/LabelP.h:31, NEWLINE from NEWLINE ../../../../dist/public/XfeWidgets/Xfe/ButtonP.h:31, NEWLINE from NEWLINE ../../../../dist/public/XfeWidgets/Xfe/ArrowP.h:31, NEWLINE from Arrow.c:27: NEWLINE ../../../../dist/public/XfeWidgets/Xfe/XfeP.h:519: parse NEWLINE error before NEWLINE `_XfeSyntheticSetResourceForChild' NEWLINE ../../../../dist/public/XfeWidgets/Xfe/XfeP.h:521: warning: NEWLINE data NEWLINE definition has no type or storage class NEWLINE In file included from NEWLINE ../../../../dist/public/XfeWidgets/Xfe/LabelP.h:31, NEWLINE from NEWLINE ../../../../dist/public/XfeWidgets/Xfe/ButtonP.h:31, NEWLINE from NEWLINE ../../../../dist/public/XfeWidgets/Xfe/ArrowP.h:31, NEWLINE from Arrow.c:27: NEWLINE Updated by dave quintero (dquinter@boris.ucr.edu) on Friday, July 24, 1998 11:38:43 AM PDT NEWLINE Additional Details : NEWLINE JUST DOWNLOAD AND INSTALL THE NEWEST BINARY VERSION OF LESSTIFF.
487|Multi-threading?|Created by Rick Downes (mozilla@radsoft.net) on Thursday, July 16, 1998 6:15:35 AM PDT NEWLINE Additional Details : NEWLINE Seems you guys haven't really gotten multithreading. NEWLINE Nowadays you've got so much going on in this mutating app NEWLINE that even at exit there are threads running, accessing NEWLINE things which other threads do their politest to clean up NEWLINE and destroy. NEWLINE With all these threads running you must find a way of NEWLINE synchronising the ExitProcess call, either directly or NEWLINE indirectly, by waiting on all running threads to exit, NEWLINE signalling them first of the shutdown. NEWLINE Several times your app has actually been running in the NEWLINE background, totally invisible, several seconds realtime NEWLINE after it disappeared from the screen. NEWLINE In this case if it is opened again, one of three things can NEWLINE result (so far): NEWLINE 1.) It just doesn't open. Period. NEWLINE 2.) It opens and disappears just as fast - like a flash on NEWLINE the screen. NEWLINE 3.) (Most often) - it crashes ALL OVER THE PLACE. These NEWLINE successive crashes, caused by the structured exception NEWLINE handling, show something is DIRELY wrong, A.) You have no NEWLINE build-in handler at all, B.) some thread is accessing NEWLINE something which has been removed from you process page NEWLINE tables subsequent to the DestroyWindow and its follow-ups. NEWLINE Updated by Garrett Arch Blythe (blythe@netscape.com) on Tuesday, July 21, 1998 2:48:12 PM PDT NEWLINE Additional Details : NEWLINE Yes, this sucks.
488|Mnemonics with form listboxes|Created by Rick Downes (mozilla@radsoft.net) on Thursday, July 16, 1998 6:18:26 AM PDT NEWLINE Additional Details : NEWLINE The mnemonmics with form listboxes and comboboxes, so NEWLINE necessary and funtional with Netscape 3 are gone. Hitting a NEWLINE keyboard key will not give ordinary Windows functionality, NEWLINE i.e. hitting a 'r' will scroll to the beginning of the r's NEWLINE in the list, etc. This is VERY unfortunate! (In your VM?) NEWLINE Updated by Garrett Arch Blythe (blythe@netscape.com) on Tuesday, July 21, 1998 2:52:47 PM PDT NEWLINE Additional Details : NEWLINE Dan, this is possibly a forms bug though the "correct" behavior is never NEWLINE specified in the report other than "be like NS version 3.0". NEWLINE NEWLINE Rick may impossibly be talking about some 4.X Java AWT problem (the mention of a NEWLINE VM makes me think this). More info may be needed, so Rick please be more NEWLINE specific if you are listening in. NEWLINE NEWLINE NEWLINE Terry, I'm actually assigning to you. I can't seem to assign bugs to NEWLINE danm@netscape.com who needs to handle this.
489|<img alt="a & b">/Tooltip-Ampersand-Display-Problem|Created by Klaus Malorny (Klaus.Malorny@knipp.de) on Thursday, July 16, 1998 10:42:06 AM PDT NEWLINE Additional Details : NEWLINE NEWLINE Tooltips, which show the contents of the 'alt'- NEWLINE attribute of an image tag do not show an ampersand (&) NEWLINE correctly (eg <img src="..." alt="this & that">) NEWLINE I tracked the bug down to the file NEWLINE NEWLINE /cmd/winfe/cxwin.cpp line 3945: NEWLINE m_ToolTip->Create(CWnd::FromHandle(GetPane()), NEWLINE TTS_ALWAYSTIP); NEWLINE NEWLINE in addition to TTS_ALWAYSTIP, the flag TTS_NOPREFIX NEWLINE is required to avoid interpretation of "&" as a hot key NEWLINE marker (like those in menues) NEWLINE NEWLINE NEWLINE Updated by Garrett Arch Blythe (blythe@netscape.com) on Tuesday, July 21, 1998 2:54:03 PM PDT NEWLINE Additional Details : NEWLINE Bill, a tooltip problem.
490|[Dogfood] Editor needs to insert formatting whitespace|Created by Klaus Malorny (Klaus.Malorny@knipp.de) on Thursday, July 16, 1998 11:39:10 AM PDT NEWLINE Additional Details : NEWLINE NEWLINE The Composer creates '\\n<BR>' for line breaks, e.g. NEWLINE entering NEWLINE NEWLINE Hello World! (RETURN) NEWLINE How are you today? (RETURN) NEWLINE NEWLINE generates NEWLINE ... NEWLINE <BODY> NEWLINE Hello World! NEWLINE <BR>How Are You Today NEWLINE <BR> NEWLINE ... NEWLINE NEWLINE The Linefeed before the <BR> tag creates additional NEWLINE white space, which is anoying in several circumstances NEWLINE (eg. you set the font size of the line to size=-2, then the NEWLINE '\\n' produces a white space of size=0, which changes the NEWLINE line height). A better solution would be '<BR>\\n', since NEWLINE whitespace is removed after a <BR>. NEWLINE NEWLINE You should review the following code for that: NEWLINE lib/layout/edtele.cpp, lines 6842, Method NEWLINE void CEditContainerElement::PrintOpen( CPrintState NEWLINE *pPrintState )
491|Some Makefiles use $(CCC) to link executables|Created by (Randy.Austin@digital.com) on Friday, July 17, 1998 11:01:10 AM PDT NEWLINE Additional Details : NEWLINE NEWLINE The three Makefiles NEWLINE xpcom/tests/Makefile NEWLINE nsprpub/lib/prstreams/tests/testprstrm/Makefile NEWLINE network/module/tests/Makefile NEWLINE NEWLINE link executables via $(CCC). NEWLINE NEWLINE On Digital UNIX, CCC includes the "-x cxx" option, which NEWLINE does not understand that .so files are not cxx files (it NEWLINE does understand .o and .a files). Therefore those makefiles NEWLINE fail on Digital UNIX. NEWLINE NEWLINE My quick fix after finding the first one was to find the NEWLINE rest and to add "-x none" to the "$(CCC) -o" command. NEWLINE However modifying the Makefiles to use the save targets that NEWLINE all other Makefiles use is the correct fix.
492|Multiple ALIGN="LEFT" tables don't render|Created by Michiel Trimpe (mtrimpe@dse.nl) on Friday, July 17, 1998 2:34:54 PM PDT NEWLINE Additional Details : NEWLINE IF you have two tables, both align=left , the one that's on NEWLINE the left, (the first one in the HTML code) renders only as NEWLINE far as the second table's height. It looks like this NEWLINE NEWLINE This is what it's suppossed to be: NEWLINE NEWLINE /-------\\ /---------------------\\ NEWLINE \| Menu1 \| \| Text Text Text \| NEWLINE \| Menu2 \| \\---------------------/ NEWLINE \| Menu3 \| NEWLINE \| Menu4 \| NEWLINE \\-------/ NEWLINE NEWLINE And this is what it looks like: NEWLINE NEWLINE /-------\\ /---------------------\\ NEWLINE \| Menu1 \| \| Text Text Text \| NEWLINE \| Menu2 \| \\---------------------/
497|Stop icon replaced by a padlock|Created by Daniel Bratell (bratell@lysator.liu.se) on Saturday, July 18, 1998 6:56:54 AM PDT NEWLINE Additional Details : NEWLINE I just did a fresh checkout of the source and compiled with NEWLINE VC++ 5.0 and found that the stop icon was replaced by a NEWLINE padlock. The same as used as a security icon I think. NEWLINE Even though it looks different it works as normal and the NEWLINE help text as shown when on leaves the pointer over it is the NEWLINE same as normal. NEWLINE Updated by Garrett Arch Blythe (blythe@netscape.com) on Tuesday, July 21, 1998 2:54:50 PM PDT NEWLINE Additional Details : NEWLINE David, the toolbar has been looking funny for a while...
507|Performance: Newsgroups pause during download|Created by Chris Waterson (waterson@netscape.com) on Monday, July 20, 1998 11:10:55 AM PDT NEWLINE Additional Details : NEWLINE Subscribe to a new (large) newsgroup. Every 200 messages or NEWLINE so, the download halts for 3-5 sec & all network traffic NEWLINE stops (witness while over a modem connection). What's going NEWLINE on? NEWLINE Updated by Chris Waterson (waterson@netscape.com) on Monday, July 20, 1998 11:14:20 AM PDT NEWLINE Additional Details : NEWLINE Reassigned to waterson@netsacpe. Initially reported by Zodd NEWLINE <locx@accessone.com>; trying to see if he can work on & fix it.
508|Default browser focus|Created by Garrett Arch Blythe (blythe@netscape.com) on Tuesday, July 21, 1998 2:59:20 PM PDT NEWLINE Additional Details : NEWLINE The focus on the "Take over these types" dialog on startup NEWLINE is never on the OK button, it is on the checkbox. NEWLINE NEWLINE I think it should be on the OK button.
509|Access Violation in JS3240.DLL during JS execution|Created by Anthony Karrer (akarrer@techempower.com) on Thursday, July 23, 1998 10:46:12 AM PDT NEWLINE Additional Details : NEWLINE Please contact one of the following persons at our NEWLINE development lab (310-338-2911): NEWLINE NEWLINE Anthony Karrer NEWLINE James S. Vitale NEWLINE Brian Hauer NEWLINE Alan Laser NEWLINE NEWLINE ----------------------------------------------------------- NEWLINE Back to the bug NEWLINE ----------------------------------------------------------- NEWLINE NEWLINE The above error continually occurs under Netscape Navigator NEWLINE 4.05 in a website we have under development. The site NEWLINE involves multiple-nested frames which make use of DHTML(div NEWLINE tags, etc), and heavy use of JavaScript data structures and NEWLINE function calls. Navigator continually breaks down under the NEWLINE load placed upon it by execution of JS code over multiple NEWLINE frames. NEWLINE NEWLINE The error dialog under Netscape appears as follows: NEWLINE NEWLINE unhandled exeception in netscape.exe 0xC0000005 NEWLINE access violation NEWLINE NEWLINE when this exeception is debugged under MS Dev Studio 5.0, NEWLINE the location of the error is within JS3240.DLL.
510|Netscape crash when running applets|Created by Suneetha Yerneni (suneetha@hilco.com) on Friday, July 24, 1998 9:26:23 AM PDT NEWLINE Additional Details : NEWLINE When tried running applets through netscape, it uses the NEWLINE system resources each time the applet is loaded and NEWLINE eventually the netscape crashes as all the system resources NEWLINE are used. Is there a way which works as a garbage collector NEWLINE and which releases the resources in Netscape? NEWLINE NEWLINE This problem doesnot occur when I use Internet explorer.
511|missing "libXm.a" library. <Error 1>|Created by dave quintero (dquinter@boris.ucr.edu) on Friday, July 24, 1998 11:45:21 AM PDT NEWLINE Additional Details : NEWLINE while running make in /mozilla/cmd/xfe it fails. NEWLINE NEWLINE gcc: /usr/X11R6/lib/libXm.a: No such file or directory NEWLINE gmake: [Linux2.1.103_x86_OPT.OBJ/moz-export] Error 1 NEWLINE gcc: /usr/X11R6/lib/libXm.a: No such file or directory NEWLINE gmake: [Linux2.1.103_x86_OPT.OBJ/mozilla-export] Error 1 NEWLINE NEWLINE p.s. i cannot find this file anywhere on the system. i NEWLINE thought it should have been included with the lesstif NEWLINE package i downloaded.
512|Improving Makefiles for Layout modules|Created by Kazushi Marukawa (jam@pobox.com) on Monday, July 27, 1998 11:36:26 AM PDT NEWLINE Additional Details : NEWLINE -l<library> style is better than lib<library>.a style for NEWLINE traditional UNIX. I think it is OK for Mac and Windows. So NEWLINE please change your style of Makefile. NEWLINE NEWLINE Here is my patch for today's Mozilla. NEWLINE NEWLINE Index: dom/src/build/Makefile NEWLINE =================================================================== NEWLINE RCS file: /cvsroot/mozilla/dom/src/build/Makefile,v NEWLINE retrieving revision 1.3 NEWLINE diff -u -r1.3 Makefile NEWLINE --- Makefile 1998/07/17 04:50:27 1.3 NEWLINE +++ Makefile 1998/07/27 18:01:10 NEWLINE @@ -42,19 +42,21 @@ NEWLINE NEWLINE EXTRA_DSO_LDOPTS = \\ NEWLINE $(LD_ALL) \\ NEWLINE - $(DIST)/lib/libjsdombase_s.a \\ NEWLINE - $(DIST)/lib/libjsdomcore_s.a \\ NEWLINE - $(DIST)/lib/libjsdomevents_s.a \\ NEWLINE - $(DIST)/lib/libjsdomhtml_s.a \\ NEWLINE + -L$(DIST)/lib \\ NEWLINE + -ljsdombase_s \\ NEWLINE + -ljsdomcore_s \\ NEWLINE + -ljsdomevents_s \\ NEWLINE + -ljsdomhtml_s \\ NEWLINE $(LD_NONE) \\ NEWLINE - $(DIST)/bin/libraptorbase.so \\ NEWLINE - $(DIST)/bin/libgmbaseunix.so \\ NEWLINE - $(DIST)/bin/libnetlib.so \\ NEWLINE - $(DIST)/bin/libjs.so \\ NEWLINE - $(DIST)/bin/libxpcom.so \\ NEWLINE - $(DIST)/bin/libnspr21.so \\ NEWLINE - $(DIST)/bin/libplds21.so \\ NEWLINE - $(DIST)/bin/libplc21.so \\ NEWLINE + -L$(DIST)/bin \\ NEWLINE + -lraptorbase \\ NEWLINE + -lgmbaseunix \\ NEWLINE + -lnetlib \\ NEWLINE + -ljs \\ NEWLINE + -lxpcom \\ NEWLINE + -lnspr21 \\ NEWLINE + -lplds21 \\ NEWLINE + -lplc21 \\ NEWLINE -lm NEWLINE NEWLINE include $(DEPTH)/config/rules.mk NEWLINE Index: gfx/src/motif/Makefile NEWLINE =================================================================== NEWLINE RCS file: /cvsroot/mozilla/gfx/src/motif/Makefile,v NEWLINE retrieving revision 1.9 NEWLINE diff -u -r1.9 Makefile NEWLINE --- Makefile 1998/07/15 01:08:34 1.9 NEWLINE +++ Makefile 1998/07/27 18:01:10 NEWLINE @@ -28,15 +28,17 @@ NEWLINE LCFLAGS+=-D_IMPL_NS_GFXONXP NEWLINE NEWLINE EXTRA_DSO_LDOPTS+=\\ NEWLINE - $(DIST)/bin/libxpcom.so \\ NEWLINE - $(DIST)/bin/libraptorgfx.so \\ NEWLINE - $(DIST)/bin/libraptorbase.so \\ NEWLINE - $(DIST)/bin/libreg.so \\ NEWLINE + -L$(DIST)/bin \\ NEWLINE + -lxpcom \\ NEWLINE + -lraptorgfx \\ NEWLINE + -lraptorbase \\ NEWLINE + -lreg \\ NEWLINE $(LIBNSPR)\\ NEWLINE - $(DIST)/lib/libplc21.a NEWLINE + -L$(DIST)/lib \\ NEWLINE + -lplc21 NEWLINE NEWLINE # This should really be done properly in 'config' NEWLINE -ifeq ($(OS_ARCH),Linux) NEWLINE +ifneq (,$(filter Linux NetBSD,$(OS_ARCH))) NEWLINE EXTRA_DSO_LDOPTS+=\\ NEWLINE -L/usr/X11R6/lib -lXt -lX11 -lXext -lm NEWLINE else NEWLINE NEWLINE Index: layout/build/Makefile NEWLINE =================================================================== NEWLINE RCS file: /cvsroot/mozilla/layout/build/Makefile,v NEWLINE retrieving revision 3.9 NEWLINE diff -u -r3.9 Makefile NEWLINE --- Makefile 1998/07/17 04:51:46 3.9 NEWLINE +++ Makefile 1998/07/27 18:01:15 NEWLINE @@ -41,27 +41,29 @@ NEWLINE NEWLINE EXTRA_DSO_LDOPTS = \\ NEWLINE $(LD_ALL) \\ NEWLINE - $(DIST)/lib/libraptorhtmldoc_s.a \\ NEWLINE - $(DIST)/lib/libnglhtmlcon_s.a \\ NEWLINE - $(DIST)/lib/libraptorhtmlforms_s.a \\ NEWLINE - $(DIST)/lib/libraptorhtmltable_s.a \\ NEWLINE - $(DIST)/lib/libraptorhtmlbase_s.a \\ NEWLINE - $(DIST)/lib/libnglcsslay_s.a \\ NEWLINE - $(DIST)/lib/libraptorhtmlstyle_s.a \\ NEWLINE - $(DIST)/lib/libraptorlayout_s.a \\ NEWLINE - $(DIST)/lib/libraptorevents_s.a \\ NEWLINE + -L$(DIST)/lib \\ NEWLINE + -lraptorhtmldoc_s \\ NEWLINE + -lnglhtmlcon_s \\ NEWLINE + -lraptorhtmlforms_s \\ NEWLINE + -lraptorhtmltable_s \\ NEWLINE + -lraptorhtmlbase_s \\ NEWLINE + -lnglcsslay_s \\ NEWLINE + -lraptorhtmlstyle_s \\ NEWLINE + -lraptorlayout_s \\ NEWLINE + -lraptorevents_s \\ NEWLINE $(LD_NONE) \\ NEWLINE - $(DIST)/bin/libraptorhtmlpars.so \\ NEWLINE - $(DIST)/bin/libraptorgfx.so \\ NEWLINE - $(DIST)/bin/libraptorbase.so \\ NEWLINE - $(DIST)/bin/libgmbaseunix.so \\ NEWLINE - $(DIST)/bin/libjsdom.so \\ NEWLINE - $(DIST)/bin/libjs.so \\ NEWLINE - $(DIST)/bin/libnetlib.so \\ NEWLINE - $(DIST)/bin/libxpcom.so \\ NEWLINE - $(DIST)/bin/libnspr21.so \\ NEWLINE - $(DIST)/bin/libplds21.so \\ NEWLINE - $(DIST)/bin/libplc21.so \\ NEWLINE + -L$(DIST)/bin \\ NEWLINE + -lraptorhtmlpars \\ NEWLINE + -lraptorgfx \\ NEWLINE + -lraptorbase \\ NEWLINE + -lgmbaseunix \\ NEWLINE + -ljsdom \\ NEWLINE + -ljs \\ NEWLINE + -lnetlib \\ NEWLINE + -lxpcom \\ NEWLINE + -lnspr21 \\ NEWLINE + -lplds21 \\ NEWLINE + -lplc21 \\ NEWLINE -lm NEWLINE NEWLINE include $(DEPTH)/config/rules.mk NEWLINE NEWLINE Index: webshell/tests/viewer/Makefile NEWLINE =================================================================== NEWLINE RCS file: /cvsroot/mozilla/webshell/tests/viewer/Makefile,v NEWLINE retrieving revision 1.37 NEWLINE diff -u -r1.37 Makefile NEWLINE --- Makefile 1998/07/27 05:59:30 1.37 NEWLINE +++ Makefile 1998/07/27 18:01:35 NEWLINE @@ -35,59 +35,60 @@ NEWLINE NEWLINE OBJS = $(CPPSRCS:.cpp=.o) NEWLINE NEWLINE -EX_LIBS = \\ NEWLINE - $(DIST)/bin/libraptorbase.so \\ NEWLINE - $(DIST)/bin/libpng.so \\ NEWLINE - $(DIST)/bin/libpref.so \\ NEWLINE - $(DIST)/bin/libraptorbase.so \\ NEWLINE - $(DIST)/bin/libwidgetunix.so \\ NEWLINE - $(DIST)/bin/libraptorgfx.so \\ NEWLINE - $(DIST)/bin/libgfxunix.so \\ NEWLINE - $(DIST)/bin/libraptorhtml.so \\ NEWLINE - $(DIST)/bin/libgmbaseunix.so \\ NEWLINE - $(DIST)/bin/libraptorhtmlpars.so \\ NEWLINE - $(DIST)/bin/libraptorview.so \\ NEWLINE - $(DIST)/bin/libreg.so \\ NEWLINE - $(DIST)/bin/libabouturl.so \\ NEWLINE - $(DIST)/bin/libfileurl.so \\ NEWLINE - $(DIST)/bin/libftpurl.so \\ NEWLINE - $(DIST)/bin/libgophurl.so \\ NEWLINE - $(DIST)/bin/libhttpurl.so \\ NEWLINE - $(DIST)/bin/libimg.so \\ NEWLINE - $(DIST)/bin/libjpeg.so \\ NEWLINE - $(DIST)/bin/libjs.so \\ NEWLINE - $(DIST)/bin/libjsdom.so \\ NEWLINE - $(DIST)/bin/libjsj.so \\ NEWLINE - $(DIST)/bin/libmimetype.so \\ NEWLINE - $(DIST)/bin/libmsgc21.so \\ NEWLINE - $(DIST)/bin/libnetcache.so \\ NEWLINE - $(DIST)/bin/libnetcnvts.so \\ NEWLINE - $(DIST)/bin/libnetlib.so \\ NEWLINE - $(DIST)/bin/libnetutil.so \\ NEWLINE - $(DIST)/bin/libnetwork.so \\ NEWLINE - $(DIST)/bin/libnspr21.so \\ NEWLINE - $(DIST)/bin/libplc21.so \\ NEWLINE - $(DIST)/bin/libplds21.so \\ NEWLINE - $(DIST)/bin/libraptorwebwidget.so \\ NEWLINE - $(DIST)/bin/libreg.so \\ NEWLINE - $(DIST)/bin/libremoturl.so \\ NEWLINE - $(DIST)/bin/libsecfree.so \\ NEWLINE - $(DIST)/bin/libstubnj.so \\ NEWLINE - $(DIST)/bin/libstubsj.so \\ NEWLINE - $(DIST)/bin/libtestdynamic.so \\ NEWLINE - $(DIST)/bin/libutil.so \\ NEWLINE - $(DIST)/bin/libxp.so \\ NEWLINE - $(DIST)/bin/libxpcom.so \\ NEWLINE - $(DIST)/bin/libzlib.so \\ NEWLINE - $(DIST)/bin/libutil.so \\ NEWLINE - $(DIST)/bin/libxp.so \\ NEWLINE - $(DIST)/bin/libxpcom.so \\ NEWLINE - $(DIST)/bin/libzlib.so \\ NEWLINE - $(DIST)/bin/libraptorplugin.so \\ NEWLINE - $(NULL) NEWLINE +EX_LDOPS = \\ NEWLINE + -L$(DIST)/bin \\ NEWLINE + -lraptorbase \\ NEWLINE + -lpng \\ NEWLINE + -lpref \\ NEWLINE + -lraptorbase \\ NEWLINE + -lwidgetunix \\ NEWLINE + -lraptorgfx \\ NEWLINE + -lgfxunix \\ NEWLINE + -lraptorhtml \\ NEWLINE + -lgmbaseunix \\ NEWLINE + -lraptorhtmlpars \\ NEWLINE + -lraptorview \\ NEWLINE + -lreg \\ NEWLINE + -labouturl \\ NEWLINE + -lfileurl \\ NEWLINE + -lftpurl \\ NEWLINE + -lgophurl \\ NEWLINE + -lhttpurl \\ NEWLINE + -limg \\ NEWLINE + -ljpeg \\ NEWLINE + -ljs \\ NEWLINE + -ljsdom \\ NEWLINE + -ljsj \\ NEWLINE + -lmimetype \\ NEWLINE + -lmsgc21 \\ NEWLINE + -lnetcache \\ NEWLINE + -lnetcnvts \\ NEWLINE + -lnetlib \\ NEWLINE + -lnetutil \\ NEWLINE + -lnetwork \\ NEWLINE + -lnspr21 \\ NEWLINE + -lplc21 \\ NEWLINE + -lplds21 \\ NEWLINE + -lraptorwebwidget \\ NEWLINE + -lreg \\ NEWLINE + -lremoturl \\ NEWLINE + -lsecfree \\ NEWLINE + -lstubnj \\ NEWLINE + -lstubsj \\ NEWLINE + -ltestdynamic \\ NEWLINE + -lutil \\ NEWLINE + -lxp \\ NEWLINE + -lxpcom \\ NEWLINE + -lzlib \\ NEWLINE + -lutil \\ NEWLINE + -lxp \\ NEWLINE + -lxpcom \\ NEWLINE + -lzlib \\ NEWLINE + -lraptorplugin \\ NEWLINE + $(NULL) NEWLINE NEWLINE ifeq ($(OS_ARCH),Linux) NEWLINE -EX_LIBS += \\ NEWLINE +EX_LDOPS += \\ NEWLINE $(DIST)/bin/libpwcac.so \\ NEWLINE $(DIST)/bin/libdbm.so \\ NEWLINE $(NULL) NEWLINE @@ -99,12 +100,16 @@ NEWLINE NEWLINE include $(DEPTH)/config/rules.mk NEWLINE NEWLINE -$(PROGS): $(OBJS) $(EX_LIBS) NEWLINE +$(PROGS): $(OBJS) NEWLINE @$(MAKE_OBJDIR) NEWLINE ifeq ($(OS_ARCH),Linux) NEWLINE - $(CCC) -rdynamic -o $@ $(OBJS) $(LDFLAGS) $(EX_LIBS) NEWLINE $(OS_LIBS) -L/usr/X11R6/lib -lXm -lXt -lX11 -lXp -lXext NEWLINE + $(CCC) -rdynamic -o $@ $(OBJS) $(LDFLAGS) NEWLINE $(EX_LDOPS) $(OS_LIBS) -L/usr/X11R6/lib -lXm -lXt -lX11 -lXp NEWLINE -lXext NEWLINE +else NEWLINE +ifeq ($(OS_ARCH),NetBSD) NEWLINE + $(CCC) -o $@ $(LDFLAGS) $(OBJS) $(EX_LDOPS) NEWLINE $(OS_LIBS) -L/usr/X11R6/lib -lXm -lXt -lX11 NEWLINE else NEWLINE - $(CCC) -o $@ -woff 84,85 $(LDFLAGS) $(OBJS) NEWLINE $(EX_LIBS) $(OS_LIBS) -lXm -lXt -lX11 NEWLINE + $(CCC) -o $@ -woff 84,85 $(LDFLAGS) $(OBJS) NEWLINE $(EX_LDOPS) $(OS_LIBS) -lXm -lXt -lX11 NEWLINE +endif NEWLINE endif NEWLINE NEWLINE export:: NEWLINE Index: widget/src/motif/Makefile NEWLINE =================================================================== NEWLINE RCS file: /cvsroot/mozilla/widget/src/motif/Makefile,v NEWLINE retrieving revision 1.20 NEWLINE diff -u -r1.20 Makefile NEWLINE --- Makefile 1998/06/23 21:18:41 1.20 NEWLINE +++ Makefile 1998/07/27 18:01:35 NEWLINE @@ -29,14 +29,16 @@ NEWLINE DEFINES = -D_IMPL_NS_WIDGET NEWLINE NEWLINE EXTRA_DSO_LDOPTS+=\\ NEWLINE - $(DIST)/bin/libxpcom.so \\ NEWLINE - $(DIST)/bin/libraptorbase.so \\ NEWLINE - $(DIST)/bin/libreg.so \\ NEWLINE + -L$(DIST)/bin \\ NEWLINE + -lxpcom \\ NEWLINE + -lraptorbase \\ NEWLINE + -lreg \\ NEWLINE $(LIBNSPR)\\ NEWLINE - $(DIST)/lib/libplc21.a NEWLINE + -L$(DIST)/lib \\ NEWLINE + -lplc21 NEWLINE NEWLINE # This should really be done properly in 'config' NEWLINE -ifeq ($(OS_ARCH),Linux) NEWLINE +ifneq (,$(filter Linux NetBSD,$(OS_ARCH))) NEWLINE EXTRA_DSO_LDOPTS+=\\ NEWLINE -L/usr/X11R6/lib -lX11 -lm NEWLINE else
513|Mozilla crashes when moving mouse in Swatches Color dialog box|Created by David Pax (dpax@wyse.com) on Friday, July 31, 1998 8:33:45 AM PDT NEWLINE Additional Details : NEWLINE Problem also exists on 7/28/98 release. NEWLINE Environment : Linux Slackware 2.0.30; Motif 1.2.4; X11R6 NEWLINE NEWLINE Mozilla built as MOZ_LITE without composer NEWLINE NEWLINE Open Preferences to the Color menu NEWLINE Click on one of the color boxes (text, bg, links, etc.) NEWLINE This brings up the Swatches Dialog box. NEWLINE Move cursor over the color boxes eventually mozilla will NEWLINE crash. 6/3/98 version crashed almost immediately while NEWLINE 7/28/98 version takes longer. Selection of a color is not NEWLINE necessary to cause crash. NEWLINE NEWLINE Updated by Aleksey Nogin (ayn2@cornell.edu) on Sunday, August 16, 1998 10:41:52 PM PDT NEWLINE Additional Details : NEWLINE I was not able to reproduce this bug with: NEWLINE Communicator CVS image as of Sun, Aug 16 NEWLINE RedHat Linux 5.1 (x86) - kernel 2.0.35 NEWLINE glibc-2.0.7-19 NEWLINE gcc-2.7.2.3-11 NEWLINE egcs-c++-1.0.2-8 NEWLINE lesstif-current-1998.08.14-1 NEWLINE NEWLINE Mozilla did not crash when I opened the "Netscape: Color Picker" window (BTW, NEWLINE why is it "Netscape" and not "Mozilla"?), but "OK" and "Cancel" buttons do not NEWLINE work (probably because of some Lesstif bug). NEWLINE Updated by (aoki@netscape.com) on Monday, August 17, 1998 9:32:12 AM PDT NEWLINE Additional Details : NEWLINE Sounds like an xfe issue; sending to mcafee NEWLINE Updated by (aoki@netscape.com) on Monday, August 17, 1998 9:32:25 AM PDT NEWLINE Additional Details : NEWLINE Sounds like an xfe issue; sending to ramiro
514|Tilesd Background image fails to fill entire background|Created by David Pax (dpax@wyse.com) on Friday, July 31, 1998 8:50:42 AM PDT NEWLINE Additional Details : NEWLINE Environment - Linux Slackware v2.0.30; Motif 1.2.4; X11R6 NEWLINE NEWLINE Mozilla built as MOZ_LITE without composer NEWLINE NEWLINE example page for problem - www.infoworld.com NEWLINE NEWLINE <body background="pageone/hedrgifs/bg_image.gif"> NEWLINE NEWLINE EXPECTED - background filled with yellow and white columns NEWLINE ACTUAL - Background image sometimes fills other times gray NEWLINE or black fill background. NEWLINE NEWLINE Additional problem appears after scrolling - click on right NEWLINE scroll bar to scroll up. NEWLINE NEWLINE EXPECTED - scrolling are to clear and repaint with nes data NEWLINE ACTUAL - clearing does not occur and new data is NEWLINE superimposed on existing data. Sometimes ackground will NEWLINE appear black, the same color as the foreground text thereby NEWLINE obscuring it. NEWLINE NEWLINE Got www.pcweek.com - similar problems occur NEWLINE NEWLINE NOTE: By selecting "Always use my colors, overriding NEWLINE document" in the Color Preference menu, the web pages with NEWLINE display and scroll correctly. Therefore, I believe the NEWLINE problem is centered around the "tiled background image".
515|MOZ_LITE build unresolved references|Created by David Pax (dpax@wyse.com) on Friday, July 31, 1998 9:36:28 AM PDT NEWLINE Additional Details : NEWLINE For the MOZ_LITE build the following modifications need to NEWLINE be made. NEWLINE NEWLINE in the file - mozilla/lib/layout/Makefile NEWLINE NEWLINE the following files need to be included when EDITOR is NEWLINE undefined. NEWLINE NEWLINE layfree.c NEWLINE layimage.c NEWLINE layout.c NEWLINE laysel.c NEWLINE laytable.c NEWLINE laytag.c NEWLINE laytext.c NEWLINE layutil.c NEWLINE NEWLINE So ifdef EDITOR needs to be moved to below these files since NEWLINE they are needed for both when EDITOR is and is not defined. NEWLINE NEWLINE Current file -------------------------------------- NEWLINE ifdef EDITOR NEWLINE CSRCS += layedit.c \\ NEWLINE layfree.c \\ NEWLINE layimage.c \\ NEWLINE layout.c \\ NEWLINE laysel.c \\ NEWLINE laytable.c \\ NEWLINE laytag.c \\ NEWLINE laytext.c \\ NEWLINE layutil.c \\ NEWLINE $(NULL) NEWLINE NEWLINE suggested changes to Makefile NEWLINE -------------------------------- NEWLINE CSRCS += layedit.c \\ NEWLINE layfree.c \\ NEWLINE layimage.c \\ NEWLINE layout.c \\ NEWLINE laysel.c \\ NEWLINE laytable.c \\ NEWLINE laytag.c \\ NEWLINE laytext.c \\ NEWLINE layutil.c \\ NEWLINE $(NULL) NEWLINE ifdef EDITOR
516|Toolbar/Toolbox does not work and looks like crap|Created by David Pax (dpax@wyse.com) on Friday, July 31, 1998 10:09:24 AM PDT NEWLINE Additional Details : NEWLINE Environment - Linux Slackware 2.0.30; Motif 1.2.4; X11R6 NEWLINE NEWLINE Default - Show toolbar As Pictures and Text NEWLINE NEWLINE Enter preferences and select Show toolbar as "pictures only" NEWLINE Exit preferences NEWLINE Re-enter preferences and select Show toolbar as "pictures NEWLINE and text" NEWLINE Exit Preferences NEWLINE NEWLINE Note that the Mozilla activity indicator on the toolbar is NEWLINE clipped in half. NEWLINE Updated by (aoki@netscape.com) on Monday, August 17, 1998 9:33:04 AM PDT NEWLINE Additional Details : NEWLINE XFE issue; sending to ramiro
517|Problem with http://user@passwd:hostname/path format|Created by Kazushi Marukawa (jam@pobox.com) on Friday, July 31, 1998 3:25:14 PM PDT NEWLINE Additional Details : NEWLINE Netscape and Mozilla hasn't work correctly with a NEWLINE http://user@passwd:hostname/path style URI after the NEWLINE Netscape 4.05. NEWLINE NEWLINE I'm not sure whether this URI is right or not. But few NEWLINE months ago we can use it. Is it a BUG or right behavior?
518|parallel make on SMP system fail 'make -j 3'|Created by Erez Strauss (erez@newplaces.com) on Sunday, August 2, 1998 4:16:28 PM PDT NEWLINE Additional Details : NEWLINE to reproduce the error message use gnu make NEWLINE run the command 'make -j 3' NEWLINE The command will run the build processes in parallel, so NEWLINE will use two CPU when available.
519|Sources in XFE don't compile because of misdeclaration|Created by Torsten Rüger (torsten@ponton-hamburg.de) on Thursday, August 6, 1998 2:24:07 AM PDT NEWLINE Additional Details : NEWLINE in directory cmd/xfe/src the following files dont compile NEWLINE because of the named functions. NEWLINE Frame.cpp in DisplayPixmap(..) NEWLINE RDFImage.cpp in RDFDisplayPixmap(..) NEWLINE In both cases the declaration and definition of the NEWLINE functions differ. That is jint is used equivalent to long NEWLINE int, and my compiler throws error. The fix is trivial, and NEWLINE it did compile when I changed the definition to match the NEWLINE declaration. NEWLINE I suppose the error only occurs on 64-bit machines ??
520|4.5b1 completely hoses NSClean software product|Created by Kevin McAleavey (kevinmca@nsclean.com) on Thursday, August 6, 1998 4:12:46 AM PDT NEWLINE Additional Details : NEWLINE I am enclosing copy of report to support@netscape: NEWLINE NEWLINE Thank you. Your bug report was successfully submitted. NEWLINE NEWLINE Here is a copy of what was sent: NEWLINE NEWLINE From: kevinmca@nsclean.com NEWLINE Reply-to: kevinmca@nsclean.com NEWLINE To: win95_cbug@netscape.com NEWLINE Subject: 4.5b1 - Windows 95 (95 and 98) - missing registry NEWLINE entries which hose up the NSClean product NEWLINE NEWLINE Submitter name: Kevin McAleavey NEWLINE Submitter email address: kevinmca@nsclean.com NEWLINE Product: Communicator NEWLINE Version: 4.5b1 NEWLINE NEWLINE Personal Edition: NEWLINE Personal NEWLINE Edition NEWLINE NEWLINE Operating system: Windows 95 NEWLINE OS version: 95 and 98 NEWLINE Browser_type: Windows 32-bit NEWLINE Language: NEWLINE Issue summary: missing registry entries NEWLINE which hose up the NSClean product NEWLINE NEWLINE Issue details: NEWLINE In all earlier versions of Communicator, there was an entry NEWLINE in NEWLINE NEWLINE HKEY_LOCAL_MACHINE\\SOFTWARE\\NETSCAPE\\NETSCAPE NEWLINE NEWLINE NAVIGATOR\\USERS pointing to the current user and subkeys for NEWLINE each user profile NEWLINE NEWLINE which contained an entry called "DirRoot" which in turn NEWLINE pointed to the path where NEWLINE NEWLINE prefs.js could be found permitting NSClean to locate NEWLINE prefs.js to locate the files which NEWLINE NEWLINE NSClean processes. NEWLINE NEWLINE NEWLINE NEWLINE We've since found "Biff" entries in HKCU but none point to NEWLINE the location of the \\users\\ NEWLINE NEWLINE folder under Program Files and as you know, people can place NEWLINE their Netscape NEWLINE NEWLINE anywhere on the drive and will often relocate the \\users\\ NEWLINE folders. This in turn has NEWLINE NEWLINE caused Privacy Software Corporation and its customers untold NEWLINE expense and hassle. NEWLINE NEWLINE NEWLINE NEWLINE As a result, we are presently informing our customers to NEWLINE AVOID the Netscape 4.5b1 NEWLINE NEWLINE product and are gravely concerned about how Netscape is NEWLINE going to resolve this NEWLINE NEWLINE missing data our product depends on. NEWLINE NEWLINE NEWLINE NEWLINE Are there any plans to correct this and provide a registry NEWLINE pointer to where that folder NEWLINE NEWLINE is located or do we need to talk to Jim B about it? NEWLINE NEWLINE Severity: Other NEWLINE Can reproduce?: Yes, Always NEWLINE Try this URL: does not apply NEWLINE Additional computer info: Any customer using NEWLINE NSClean32 4.10 or NSClean16 3.35 products is affected. NEWLINE Acknowledgement checkbox: on NEWLINE NEWLINE This bug was submitted with Mozilla/4.04 [en] (Win95; I). NEWLINE NEWLINE NEWLINE Updated by Kevin McAleavey (kevinmca@nsclean.com) on Thursday, August 6, 1998 5:12:58 AM PDT NEWLINE Additional Details : NEWLINE Any documentation in C for parsing NSREG.DAT as contained in 4.5b1 and future NEWLINE betas?
521|Navigator 4.05 font oddity|Created by (Jazzmand@hotmail.com) on Friday, August 7, 1998 1:42:56 AM PDT NEWLINE Additional Details : NEWLINE I have found in a bug/oddity in Communicator 4.05. If you NEWLINE have loaded a NEWLINE page with text on, then hold down ALT whilst NEWLINE keying in 027, the text shrinks. Repeating this, you can get NEWLINE the text to NEWLINE become unreadable. NEWLINE NEWLINE Jasper Ditton NEWLINE J.G.Ditton@iti.salford.ac.uk
522|Missing includes and build problems on OSF1 4.0|Created by Bob McElrath (mcelrath@isp.nwu.edu) on Saturday, August 8, 1998 8:26:05 AM PDT NEWLINE Additional Details : NEWLINE include/mcom_db.h needs system include endian.h, which is in NEWLINE /usr/sys/include/arch/alpha. Can be fixed by adding the NEWLINE line: NEWLINE NEWLINE OS_INCLUDES = -I/usr/sys/include/arch/alpha NEWLINE NEWLINE to config/OSF1.mk NEWLINE NEWLINE also: NEWLINE NEWLINE running gmake in the dbm/src directory (after fixing the NEWLINE above problem): NEWLINE cc -ieee_with_inexact -std1 -o OSF1V4.0_OPT.OBJ/h_page.o -c NEWLINE -DXP_UNIX -O -Olimit 4000 -taso -D_ALPHA_ -DIS_64 -DOSF1 NEWLINE -DOSF1V4 -D_REENTRANT -DHAVE_LCHOWN -DNEED_CDEFS_H NEWLINE -DNTOHL_ENDIAN_H -DNEED_IOCTL_H -DMACHINE_ENDIAN_H NEWLINE -DHAVE_VA_LIST_STRUCT -DNEED_BYTE_ALIGNMENT -DMITSHM NEWLINE -DNEED_REALPATH -DHAVE_WAITID -DNEED_H_ERRNO NEWLINE -DNEED_SYS_TIME_H -DHAVE_SYSTEMINFO_H -DNEED_SYS_PARAM_H NEWLINE -DHAVE_INT32_T -DODD_VA_START -DHAVE_REMAINDER NEWLINE -DSW_THREADS -UDEBUG -DNDEBUG -DTRIMMED -DNSPR20 NEWLINE -DNETSCAPE -DOSTYPE=\\"OSF1V4\\" -DMOZILLA_CLIENT -DEDITOR NEWLINE -DMOZ_COMMUNICATOR_IIDS -DNO_SECURITY -DEDITOR -DEDITOR_UI NEWLINE -DLAYERS -DUNIX_EMBED -DX_PLUGINS -DUNIX_LDAP -DNSPR -DMOCHA NEWLINE -DUNIX_ASYNC_DNS -DDEVELOPER_DEBUG -DMEMMOVE NEWLINE -D__DBINTERFACE_PRIVATE -DNSPR20=1 -I../../include NEWLINE -I/usr/sys/include/arch/alpha NEWLINE -I../../dist/OSF1V4.0_OPT.OBJ/include -I../include NEWLINE h_page.c NEWLINE cc: Error: h_page.c, line 1160: In this statement, NEWLINE "SIG_BLOCK" is not declared. NEWLINE (void)sigprocmask(SIG_BLOCK, &set, &oset); NEWLINE --------------------------^ NEWLINE cc: Error: h_page.c, line 1171: In this statement, NEWLINE "SIG_SETMASK" is not declared. NEWLINE (void)sigprocmask(SIG_SETMASK, &oset, (sigset_t NEWLINE *)NULL); NEWLINE --------------------------^ NEWLINE gmake: *** [OSF1V4.0_OPT.OBJ/h_page.o] Error 1 NEWLINE NEWLINE I can clearly see that signal.h is included, and that NEWLINE SIG_BLOCK and SIG_SETMASK are defined. I suspect something NEWLINE odd is happening in the 1000-odd lines between #include NEWLINE <signal.h> and the lines on which the error occurs. NEWLINE NEWLINE running: NEWLINE cpp h_page.c -I../../include -I/usr/sys/include/arch/alpha NEWLINE -I../../dist/OSF1V4.0_OPT.OBJ/include -I../include \| grep NEWLINE SIG NEWLINE gives: NEWLINE (void)sigprocmask(SIG_BLOCK, &set, &oset); NEWLINE (void)sigprocmask(SIG_SETMASK, &oset, (sigset_t NEWLINE *)0L); NEWLINE So clearly the pre-processor is failing. (Fails for both NEWLINE DEC's cpp and gcc -E version 2.7.2.3)
523|A *.tar.gz file saved as *_tar.tar|Created by (tbielick@lynx.neu.edu) on Tuesday, August 11, 1998 6:54:09 AM PDT NEWLINE Additional Details : NEWLINE Saving of a *.tar.gz file may result in the saved file being NEWLINE named *_tar.tar Thus loseing the gz extention, as well as NEWLINE undersources replaceing dots in file name. NEWLINE Updated by Garrett Arch Blythe (blythe@netscape.com) on Tuesday, August 11, 1998 10:25:31 AM PDT NEWLINE Additional Details : NEWLINE This is fixed in the mozilla builds. You are using mozilla, right? NEWLINE FYI, if you are not, use Netscape tech support for Netscape brand bugs.
524|Resize of main browser window causes endless refresh loop|Created by Christopher Blizzard (blizzard@appliedtheory.com) on Tuesday, August 11, 1998 5:21:03 PM PDT NEWLINE Additional Details : NEWLINE There is a problem with the mozilla code as checked out NEWLINE on Aug 11, 1998. NEWLINE NEWLINE Resizing the main browser window causes and endless loop NEWLINE of trying to refresh the screen. This has been compiled NEWLINE on RedHat Linux 5.1 with Motif 2.1. A stack trace reveals NEWLINE a stack that has over 500 stack frames in less than a NEWLINE second. This was generated by pausing the browser in gdb, NEWLINE resizing the window and then restarting the browser and NEWLINE immediately stopping it again. Letting it run for over 2 NEWLINE seconds created a stack that had over 2000 frames. NEWLINE NEWLINE The stack trace for anyone who is interested is located at: NEWLINE NEWLINE http://odin.appliedtheory.com/big_stack_trace.txt NEWLINE NEWLINE It's a bit large to attach to a bug report. NEWLINE Updated by Christopher Blizzard (blizzard@appliedtheory.com) on Thursday, August 13, 1998 8:37:18 AM PDT NEWLINE Additional Details : NEWLINE I have verified that this bug has shown up in between the mozilla-19980603 and NEWLINE the mozilla-19980728 releases and affects the version that is in CVS now. NEWLINE NEWLINE I've built all three on the same machine. Version 06/03 works well and verions NEWLINE 07/28 exhibits the problem. I'll spend more time on it when I get home tonight NEWLINE and see if I can track it down.
525|cmdwinfepwplat.cpp wont compile|Created by Doug Needham (sheridan3003@yahoo.com) on Wednesday, August 12, 1998 5:23:08 AM PDT NEWLINE Additional Details : NEWLINE running NT Server 4.0 (sp3) NEWLINE MSVC 5.0 NEWLINE NEWLINE I run nmake /f cleint.mak NEWLINE and I get the following result. Am I missing something? NEWLINE NEWLINE ..\\..\\..\\lib\\plugin\\nsIPlug.h(368) : warning C4099: NEWLINE 'NPIPlugin' : type NEWLINE name first seen using 'struct' now seen using 'class' NEWLINE ..\\..\\..\\lib\\plugin\\nsIPlug.h(452) : warning C4099: NEWLINE 'NPIPlugin' : type NEWLINE name first seen using 'struct' now seen using 'class' NEWLINE E:\\DOUG\\MOZILLA\\MOZILLA\\CMD\\WINFE\\NPWPLAT.cpp(459) : error NEWLINE C2065: NEWLINE 'NS_IPLUGINMANAGER_IID' : undeclared identifier NEWLINE E:\\DOUG\\MOZILLA\\MOZILLA\\CMD\\WINFE\\NPWPLAT.cpp(459) : error NEWLINE C2440: NEWLINE 'initializing' : cannot convert from 'int' to 'const struct NEWLINE nsID' NEWLINE NEWLINE No NEWLINE constructor could take the source type, or constructor NEWLINE overload NEWLINE resolution was ambiguous NEWLINE E:\\DOUG\\MOZILLA\\MOZILLA\\CMD\\WINFE\\NPWPLAT.cpp(667) : error NEWLINE C2065: NEWLINE 'nsFactoryProc' : undeclared identifier NEWLINE E:\\DOUG\\MOZILLA\\MOZILLA\\CMD\\WINFE\\NPWPLAT.cpp(667) : error NEWLINE C2146: NEWLINE syntax error : missing ';' before identifier 'nsGetFactory' NEWLINE E:\\DOUG\\MOZILLA\\MOZILLA\\CMD\\WINFE\\NPWPLAT.cpp(667) : error NEWLINE C2065: NEWLINE 'nsGetFactory' : undeclared identifier NEWLINE E:\\DOUG\\MOZILLA\\MOZILLA\\CMD\\WINFE\\NPWPLAT.cpp(668) : error NEWLINE C2146: NEWLINE syntax error : missing ';' before identifier 'PR_FindSymbol' NEWLINE E:\\DOUG\\MOZILLA\\MOZILLA\\CMD\\WINFE\\NPWPLAT.cpp(670) : error NEWLINE C2065: NEWLINE 'NS_IPLUGIN_IID' : undeclared identifier NEWLINE E:\\DOUG\\MOZILLA\\MOZILLA\\CMD\\WINFE\\NPWPLAT.cpp(670) : error NEWLINE C2440: NEWLINE 'initializing' : cannot convert from 'int' to 'const struct NEWLINE nsID' NEWLINE NEWLINE No NEWLINE constructor could take the source type, or constructor NEWLINE overload NEWLINE resolution was ambiguous NEWLINE E:\\DOUG\\MOZILLA\\MOZILLA\\CMD\\WINFE\\NPWPLAT.cpp(671) : error NEWLINE C2065: NEWLINE 'nsIPlugin' : undeclared identifier NEWLINE E:\\DOUG\\MOZILLA\\MOZILLA\\CMD\\WINFE\\NPWPLAT.cpp(671) : error NEWLINE C2065: NEWLINE 'plugin' : undeclared identifier NEWLINE E:\\DOUG\\MOZILLA\\MOZILLA\\CMD\\WINFE\\NPWPLAT.cpp(671) : error NEWLINE C2106: '=' NEWLINE : left operand must be l-value NEWLINE E:\\DOUG\\MOZILLA\\MOZILLA\\CMD\\WINFE\\NPWPLAT.cpp(672) : error NEWLINE C2065: NEWLINE 'nsIFactory' : undeclared identifier NEWLINE E:\\DOUG\\MOZILLA\\MOZILLA\\CMD\\WINFE\\NPWPLAT.cpp(672) : error NEWLINE C2059: NEWLINE syntax error : ')' NEWLINE E:\\DOUG\\MOZILLA\\MOZILLA\\CMD\\WINFE\\NPWPLAT.cpp(675) : error NEWLINE C2227: left NEWLINE of '->Initialize' must point to class/struct/union NEWLINE E:\\DOUG\\MOZILLA\\MOZILLA\\CMD\\WINFE\\NPWPLAT.cpp(675) : error NEWLINE C2065: NEWLINE 'nsIPluginManager' : undeclared identifier NEWLINE E:\\DOUG\\MOZILLA\\MOZILLA\\CMD\\WINFE\\NPWPLAT.cpp(675) : error NEWLINE C2059: NEWLINE syntax error : ')' NEWLINE E:\\DOUG\\MOZILLA\\MOZILLA\\CMD\\WINFE\\NPWPLAT.cpp(690) : error NEWLINE C2143: NEWLINE syntax error : missing ';' before 'else' NEWLINE E:\\DOUG\\MOZILLA\\MOZILLA\\CMD\\WINFE\\NPWPLAT.cpp(698) : error NEWLINE C2143: NEWLINE syntax error : missing ';' before 'if' NEWLINE E:\\DOUG\\MOZILLA\\MOZILLA\\CMD\\WINFE\\NPWPLAT.cpp(701) : error NEWLINE C2065: NEWLINE 'pPathSave' : undeclared identifier NEWLINE E:\\DOUG\\MOZILLA\\MOZILLA\\CMD\\WINFE\\NPWPLAT.cpp(701) : error NEWLINE C2501: NEWLINE 'wfe_PopPath' : missing decl-specifiers NEWLINE E:\\DOUG\\MOZILLA\\MOZILLA\\CMD\\WINFE\\NPWPLAT.cpp(701) : error NEWLINE C2371: NEWLINE 'wfe_PopPath' : redefinition; different basic types NEWLINE E:\\DOUG\\MOZILLA\\MOZILLA\\CMD\\WINFE\\NPWPLAT.cpp(702) : error NEWLINE C2143: NEWLINE syntax error : missing ';' before 'return' NEWLINE E:\\DOUG\\MOZILLA\\MOZILLA\\CMD\\WINFE\\NPWPLAT.cpp(703) : error NEWLINE C2143: NEWLINE syntax error : missing ';' before '}' NEWLINE E:\\DOUG\\MOZILLA\\MOZILLA\\CMD\\WINFE\\NPWPLAT.cpp(708) : error NEWLINE C2501: NEWLINE 'pNPMgtBlock' : missing decl-specifiers NEWLINE E:\\DOUG\\MOZILLA\\MOZILLA\\CMD\\WINFE\\NPWPLAT.cpp(708) : error NEWLINE C2239: NEWLINE unexpected token '->' following declaration of 'pNPMgtBlock' NEWLINE E:\\DOUG\\MOZILLA\\MOZILLA\\CMD\\WINFE\\NPWPLAT.cpp(708) : error NEWLINE C2059: NEWLINE syntax error : '->' NEWLINE E:\\DOUG\\MOZILLA\\MOZILLA\\CMD\\WINFE\\NPWPLAT.cpp(709) : error NEWLINE C2501: NEWLINE 'pNPMgtBlock' : missing decl-specifiers NEWLINE E:\\DOUG\\MOZILLA\\MOZILLA\\CMD\\WINFE\\NPWPLAT.cpp(709) : error NEWLINE C2086: NEWLINE 'pNPMgtBlock' : redefinition NEWLINE E:\\DOUG\\MOZILLA\\MOZILLA\\CMD\\WINFE\\NPWPLAT.cpp(709) : error NEWLINE C2239: NEWLINE unexpected token '->' following declaration of 'pNPMgtBlock' NEWLINE E:\\DOUG\\MOZILLA\\MOZILLA\\CMD\\WINFE\\NPWPLAT.cpp(709) : error NEWLINE C2059: NEWLINE syntax error : '->' NEWLINE E:\\DOUG\\MOZILLA\\MOZILLA\\CMD\\WINFE\\NPWPLAT.cpp(710) : error NEWLINE C2143: NEWLINE syntax error : missing ';' before 'if' NEWLINE E:\\DOUG\\MOZILLA\\MOZILLA\\CMD\\WINFE\\NPWPLAT.cpp(713) : error NEWLINE C2501: NEWLINE 'wfe_PopPath' : missing decl-specifiers NEWLINE E:\\DOUG\\MOZILLA\\MOZILLA\\CMD\\WINFE\\NPWPLAT.cpp(713) : error NEWLINE C2371: NEWLINE 'wfe_PopPath' : redefinition; different basic types NEWLINE E:\\DOUG\\MOZILLA\\MOZILLA\\CMD\\WINFE\\NPWPLAT.cpp(714) : error NEWLINE C2143: NEWLINE syntax error : missing ';' before 'return' NEWLINE E:\\DOUG\\MOZILLA\\MOZILLA\\CMD\\WINFE\\NPWPLAT.cpp(715) : error NEWLINE C2143: NEWLINE syntax error : missing ';' before '}' NEWLINE E:\\DOUG\\MOZILLA\\MOZILLA\\CMD\\WINFE\\NPWPLAT.cpp(721) : error NEWLINE C2143: NEWLINE syntax error : missing ';' before 'delete' NEWLINE E:\\DOUG\\MOZILLA\\MOZILLA\\CMD\\WINFE\\NPWPLAT.cpp(722) : error NEWLINE C2501: NEWLINE 'pNPMgtBlock' : missing decl-specifiers NEWLINE E:\\DOUG\\MOZILLA\\MOZILLA\\CMD\\WINFE\\NPWPLAT.cpp(722) : error NEWLINE C2086: NEWLINE 'pNPMgtBlock' : redefinition NEWLINE E:\\DOUG\\MOZILLA\\MOZILLA\\CMD\\WINFE\\NPWPLAT.cpp(722) : error NEWLINE C2239: NEWLINE unexpected token '->' following declaration of 'pNPMgtBlock' NEWLINE E:\\DOUG\\MOZILLA\\MOZILLA\\CMD\\WINFE\\NPWPLAT.cpp(722) : error NEWLINE C2059: NEWLINE syntax error : '->' NEWLINE E:\\DOUG\\MOZILLA\\MOZILLA\\CMD\\WINFE\\NPWPLAT.cpp(723) : error NEWLINE C2501: NEWLINE 'wfe_PopPath' : missing decl-specifiers NEWLINE E:\\DOUG\\MOZILLA\\MOZILLA\\CMD\\WINFE\\NPWPLAT.cpp(723) : error NEWLINE C2371: NEWLINE 'wfe_PopPath' : redefinition; different basic types NEWLINE E:\\DOUG\\MOZILLA\\MOZILLA\\CMD\\WINFE\\NPWPLAT.cpp(724) : error NEWLINE C2143: NEWLINE syntax error : missing ';' before 'return' NEWLINE E:\\DOUG\\MOZILLA\\MOZILLA\\CMD\\WINFE\\NPWPLAT.cpp(725) : error NEWLINE C2143: NEWLINE syntax error : missing ';' before '}' NEWLINE E:\\DOUG\\MOZILLA\\MOZILLA\\CMD\\WINFE\\NPWPLAT.cpp(732) : error NEWLINE C2143: NEWLINE syntax error : missing ';' before 'delete' NEWLINE E:\\DOUG\\MOZILLA\\MOZILLA\\CMD\\WINFE\\NPWPLAT.cpp(733) : error NEWLINE C2501: NEWLINE 'pNPMgtBlock' : missing decl-specifiers NEWLINE E:\\DOUG\\MOZILLA\\MOZILLA\\CMD\\WINFE\\NPWPLAT.cpp(733) : error NEWLINE C2086: NEWLINE 'pNPMgtBlock' : redefinition NEWLINE E:\\DOUG\\MOZILLA\\MOZILLA\\CMD\\WINFE\\NPWPLAT.cpp(733) : error NEWLINE C2239: NEWLINE unexpected token '->' following declaration of 'pNPMgtBlock' NEWLINE E:\\DOUG\\MOZILLA\\MOZILLA\\CMD\\WINFE\\NPWPLAT.cpp(733) : error NEWLINE C2059: NEWLINE syntax error : '->' NEWLINE E:\\DOUG\\MOZILLA\\MOZILLA\\CMD\\WINFE\\NPWPLAT.cpp(734) : error NEWLINE C2501: NEWLINE 'wfe_PopPath' : missing decl-specifiers NEWLINE E:\\DOUG\\MOZILLA\\MOZILLA\\CMD\\WINFE\\NPWPLAT.cpp(734) : error NEWLINE C2371: NEWLINE 'wfe_PopPath' : redefinition; different basic types NEWLINE E:\\DOUG\\MOZILLA\\MOZILLA\\CMD\\WINFE\\NPWPLAT.cpp(735) : error NEWLINE C2143: NEWLINE syntax error : missing ';' before 'return' NEWLINE E:\\DOUG\\MOZILLA\\MOZILLA\\CMD\\WINFE\\NPWPLAT.cpp(736) : error NEWLINE C2143: NEWLINE syntax error : missing ';' before '}' NEWLINE E:\\DOUG\\MOZILLA\\MOZILLA\\CMD\\WINFE\\NPWPLAT.cpp(741) : error NEWLINE C2143: NEWLINE syntax error : missing ';' before 'if' NEWLINE E:\\DOUG\\MOZILLA\\MOZILLA\\CMD\\WINFE\\NPWPLAT.cpp(753) : error NEWLINE C2143: NEWLINE syntax error : missing ';' before 'if' NEWLINE E:\\DOUG\\MOZILLA\\MOZILLA\\CMD\\WINFE\\NPWPLAT.cpp(760) : error NEWLINE C2143: NEWLINE syntax error : missing ';' before '}' NEWLINE E:\\DOUG\\MOZILLA\\MOZILLA\\CMD\\WINFE\\NPWPLAT.cpp(765) : error NEWLINE C2143: NEWLINE syntax error : missing ';' before 'delete' NEWLINE E:\\DOUG\\MOZILLA\\MOZILLA\\CMD\\WINFE\\NPWPLAT.cpp(766) : error NEWLINE C2501: NEWLINE 'pNPMgtBlock' : missing decl-specifiers NEWLINE E:\\DOUG\\MOZILLA\\MOZILLA\\CMD\\WINFE\\NPWPLAT.cpp(766) : error NEWLINE C2086: NEWLINE 'pNPMgtBlock' : redefinition NEWLINE E:\\DOUG\\MOZILLA\\MOZILLA\\CMD\\WINFE\\NPWPLAT.cpp(766) : error NEWLINE C2239: NEWLINE unexpected token '->' following declaration of 'pNPMgtBlock' NEWLINE E:\\DOUG\\MOZILLA\\MOZILLA\\CMD\\WINFE\\NPWPLAT.cpp(766) : error NEWLINE C2059: NEWLINE syntax error : '->' NEWLINE E:\\DOUG\\MOZILLA\\MOZILLA\\CMD\\WINFE\\NPWPLAT.cpp(767) : error NEWLINE C2501: NEWLINE 'wfe_PopPath' : missing decl-specifiers NEWLINE E:\\DOUG\\MOZILLA\\MOZILLA\\CMD\\WINFE\\NPWPLAT.cpp(767) : error NEWLINE C2371: NEWLINE 'wfe_PopPath' : redefinition; different basic types NEWLINE E:\\DOUG\\MOZILLA\\MOZILLA\\CMD\\WINFE\\NPWPLAT.cpp(768) : error NEWLINE C2143: NEWLINE syntax error : missing ';' before 'return' NEWLINE E:\\DOUG\\MOZILLA\\MOZILLA\\CMD\\WINFE\\NPWPLAT.cpp(769) : error NEWLINE C2143: NEWLINE syntax error : missing ';' before '}' NEWLINE E:\\DOUG\\MOZILLA\\MOZILLA\\CMD\\WINFE\\NPWPLAT.cpp(773) : error NEWLINE C2143: NEWLINE syntax error : missing ';' before 'delete' NEWLINE E:\\DOUG\\MOZILLA\\MOZILLA\\CMD\\WINFE\\NPWPLAT.cpp(774) : error NEWLINE C2501: NEWLINE 'pNPMgtBlock' : missing decl-specifiers NEWLINE E:\\DOUG\\MOZILLA\\MOZILLA\\CMD\\WINFE\\NPWPLAT.cpp(774) : error NEWLINE C2086: NEWLINE 'pNPMgtBlock' : redefinition NEWLINE E:\\DOUG\\MOZILLA\\MOZILLA\\CMD\\WINFE\\NPWPLAT.cpp(774) : error NEWLINE C2239: NEWLINE unexpected token '->' following declaration of 'pNPMgtBlock' NEWLINE E:\\DOUG\\MOZILLA\\MOZILLA\\CMD\\WINFE\\NPWPLAT.cpp(774) : error NEWLINE C2059: NEWLINE syntax error : '->' NEWLINE E:\\DOUG\\MOZILLA\\MOZILLA\\CMD\\WINFE\\NPWPLAT.cpp(775) : error NEWLINE C2501: NEWLINE 'wfe_PopPath' : missing decl-specifiers NEWLINE E:\\DOUG\\MOZILLA\\MOZILLA\\CMD\\WINFE\\NPWPLAT.cpp(775) : error NEWLINE C2371: NEWLINE 'wfe_PopPath' : redefinition; different basic types NEWLINE E:\\DOUG\\MOZILLA\\MOZILLA\\CMD\\WINFE\\NPWPLAT.cpp(776) : error NEWLINE C2143: NEWLINE syntax error : missing ';' before 'return' NEWLINE E:\\DOUG\\MOZILLA\\MOZILLA\\CMD\\WINFE\\NPWPLAT.cpp(777) : error NEWLINE C2143: NEWLINE syntax error : missing ';' before '}' NEWLINE E:\\DOUG\\MOZILLA\\MOZILLA\\CMD\\WINFE\\NPWPLAT.cpp(783) : error NEWLINE C2501: NEWLINE 'wfe_PopPath' : missing decl-specifiers NEWLINE E:\\DOUG\\MOZILLA\\MOZILLA\\CMD\\WINFE\\NPWPLAT.cpp(783) : error NEWLINE C2371: NEWLINE 'wfe_PopPath' : redefinition; different basic types NEWLINE E:\\DOUG\\MOZILLA\\MOZILLA\\CMD\\WINFE\\NPWPLAT.cpp(784) : error NEWLINE C2143: NEWLINE syntax error : missing ';' before 'return' NEWLINE E:\\DOUG\\MOZILLA\\MOZILLA\\CMD\\WINFE\\NPWPLAT.cpp(785) : error NEWLINE C2143: NEWLINE syntax error : missing ';' before '}' NEWLINE E:\\DOUG\\MOZILLA\\MOZILLA\\CMD\\WINFE\\NPWPLAT.cpp(792) : error NEWLINE C2143: NEWLINE syntax error : missing ';' before 'if' NEWLINE E:\\DOUG\\MOZILLA\\MOZILLA\\CMD\\WINFE\\NPWPLAT.cpp(799) : error NEWLINE C2501: NEWLINE 'pNPMgtBlk' : missing decl-specifiers NEWLINE E:\\DOUG\\MOZILLA\\MOZILLA\\CMD\\WINFE\\NPWPLAT.cpp(799) : error NEWLINE C2239: NEWLINE unexpected token '->' following declaration of 'pNPMgtBlk' NEWLINE E:\\DOUG\\MOZILLA\\MOZILLA\\CMD\\WINFE\\NPWPLAT.cpp(799) : error NEWLINE C2059: NEWLINE syntax error : '->' NEWLINE E:\\DOUG\\MOZILLA\\MOZILLA\\CMD\\WINFE\\NPWPLAT.cpp(802) : error NEWLINE C2143: NEWLINE syntax error : missing ';' before 'if' NEWLINE E:\\DOUG\\MOZILLA\\MOZILLA\\CMD\\WINFE\\NPWPLAT.cpp(805) : error NEWLINE C2143: NEWLINE syntax error : missing ')' before '==' NEWLINE E:\\DOUG\\MOZILLA\\MOZILLA\\CMD\\WINFE\\NPWPLAT.cpp(811) : error NEWLINE C2143: NEWLINE syntax error : missing ';' before '}' NEWLINE E:\\DOUG\\MOZILLA\\MOZILLA\\CMD\\WINFE\\NPWPLAT.cpp(824) : error NEWLINE C2143: NEWLINE syntax error : missing ';' before 'if' NEWLINE E:\\DOUG\\MOZILLA\\MOZILLA\\CMD\\WINFE\\NPWPLAT.cpp(831) : error NEWLINE C2143: NEWLINE syntax error : missing ';' before '}' NEWLINE E:\\DOUG\\MOZILLA\\MOZILLA\\CMD\\WINFE\\NPWPLAT.cpp(835) : error NEWLINE C2143: NEWLINE syntax error : missing ';' before '}' NEWLINE E:\\DOUG\\MOZILLA\\MOZILLA\\CMD\\WINFE\\NPWPLAT.cpp(840) : error NEWLINE C2501: NEWLINE 'pNPMgtBlk' : missing decl-specifiers NEWLINE E:\\DOUG\\MOZILLA\\MOZILLA\\CMD\\WINFE\\NPWPLAT.cpp(840) : error NEWLINE C2086: NEWLINE 'pNPMgtBlk' : redefinition NEWLINE E:\\DOUG\\MOZILLA\\MOZILLA\\CMD\\WINFE\\NPWPLAT.cpp(840) : error NEWLINE C2239: NEWLINE unexpected token '->' following declaration of 'pNPMgtBlk' NEWLINE E:\\DOUG\\MOZILLA\\MOZILLA\\CMD\\WINFE\\NPWPLAT.cpp(840) : error NEWLINE C2059: NEWLINE syntax error : '->' NEWLINE E:\\DOUG\\MOZILLA\\MOZILLA\\CMD\\WINFE\\NPWPLAT.cpp(842) : error NEWLINE C2143: NEWLINE syntax error : missing ';' before 'if' NEWLINE E:\\DOUG\\MOZILLA\\MOZILLA\\CMD\\WINFE\\NPWPLAT.cpp(844) : error NEWLINE C2501: NEWLINE 'handle' : missing decl-specifiers NEWLINE E:\\DOUG\\MOZILLA\\MOZILLA\\CMD\\WINFE\\NPWPLAT.cpp(844) : error NEWLINE C2239: NEWLINE unexpected token '->' following declaration of 'handle' NEWLINE E:\\DOUG\\MOZILLA\\MOZILLA\\CMD\\WINFE\\NPWPLAT.cpp(844) : error NEWLINE C2059: NEWLINE syntax error : '->' NEWLINE E:\\DOUG\\MOZILLA\\MOZILLA\\CMD\\WINFE\\NPWPLAT.cpp(845) : error NEWLINE C2143: NEWLINE syntax error : missing ';' before '}' NEWLINE E:\\DOUG\\MOZILLA\\MOZILLA\\CMD\\WINFE\\NPWPLAT.cpp(848) : error NEWLINE C2143: NEWLINE syntax error : missing ';' before '}' NEWLINE E:\\DOUG\\MOZILLA\\MOZILLA\\CMD\\WINFE\\NPWPLAT.cpp(850) : error NEWLINE C2143: NEWLINE syntax error : missing ';' before '}' NEWLINE E:\\DOUG\\MOZILLA\\MOZILLA\\CMD\\WINFE\\NPWPLAT.cpp(858) : error NEWLINE C2143: NEWLINE syntax error : missing ';' before 'if' NEWLINE E:\\DOUG\\MOZILLA\\MOZILLA\\CMD\\WINFE\\NPWPLAT.cpp(861) : error NEWLINE C2143: NEWLINE syntax error : missing ';' before 'if' NEWLINE E:\\DOUG\\MOZILLA\\MOZILLA\\CMD\\WINFE\\NPWPLAT.cpp(861) : fatal NEWLINE error NEWLINE C1003: error count exceeds 100; stopping compilation
526|Won't Compile on Sparc Linux (Redhat 5.1)|Created by Alex Tomlins (apdt97@doc.ic.ac.uk) on Wednesday, August 12, 1998 6:59:24 AM PDT NEWLINE Additional Details : NEWLINE Compile stops with the following error. NEWLINE NEWLINE gcc -o Linux2.0.33_sparc_DBG.OBJ/linux.o -c -g -fPIC -ansi NEWLINE -Wall -pipe -DLINUX -Dlinux -D_POSIX_SOURCE -D_BSD_SOURCE NEWLINE -DHAVE_STRERROR -DFORCE_PR_LOG -DXP_UNIX -DDEBUG -UNDEBUG NEWLINE -DDEBUG_root -D_PR_LOCAL_THREADS_ONLY NEWLINE -I../../../../../dist/Linux2.0.33_sparc_DBG.OBJ/include/private NEWLINE -I../../../../../dist/Linux2.0.33_sparc_DBG.OBJ/include NEWLINE linux.c NEWLINE linux.c: In function `_MD_InitializeThread': NEWLINE linux.c:63: request for member `__fp' in something not a NEWLINE structure or union NEWLINE make[5]: *** [Linux2.0.33_sparc_DBG.OBJ/linux.o] Error 1 NEWLINE make[5]: Leaving directory NEWLINE `/mnt/d/rpm/BUILD/mozilla/nsprpub/pr/src/md/unix' NEWLINE make[4]: *** [export] Error 2 NEWLINE make[4]: Leaving directory NEWLINE `/mnt/d/rpm/BUILD/mozilla/nsprpub/pr/src/md' NEWLINE make[3]: *** [export] Error 2 NEWLINE make[3]: Leaving directory NEWLINE `/mnt/d/rpm/BUILD/mozilla/nsprpub/pr/src' NEWLINE make[2]: *** [export] Error 2 NEWLINE make[2]: Leaving directory NEWLINE `/mnt/d/rpm/BUILD/mozilla/nsprpub/pr' NEWLINE make[1]: *** [export] Error 2 NEWLINE make[1]: Leaving directory NEWLINE `/mnt/d/rpm/BUILD/mozilla/nsprpub' NEWLINE make: *** [export] Error 2 NEWLINE NEWLINE I am using lesstif version 0.85 NEWLINE Updated by Alex Tomlins (apdt97@doc.ic.ac.uk) on Friday, August 14, 1998 2:02:00 AM PDT NEWLINE Additional Details : NEWLINE This error has now been sorted, but it now comes up with an error, saying it NEWLINE cannot execute cc1plus NEWLINE NEWLINE I have GNU make 3.76.1 and GCC version egcs-2.90.27 (egcs-1.0.2 release) NEWLINE NEWLINE thanks NEWLINE Alex
527|NgLayout.mak needs Unix_Date; Not Found On Windows.|Created by Andrew Barfield (xtremesoftware@usa.net) on Wednesday, August 12, 1998 7:23:36 AM PDT NEWLINE Additional Details : NEWLINE A transcript of the MS_DOS build of NgLayout is as follows NEWLINE (the script encounters a need for unix_date and fails): NEWLINE ----------------------------------------------------------- NEWLINE NEWLINE [WINDOWS] C:\\raptor_source\\mozilla>dir/w/p NEWLINE NEWLINE Volume in drive C is Xtreme_Station_1 NEWLINE Volume Serial Number is ****-**** NEWLINE Directory of C:\\raptor_source\\mozilla NEWLINE NEWLINE [.] [..] [BASE] [CONFIG] [CVS] [DIST] [DOM] [GFX] NEWLINE [HTMLPA~1] [INCLUDE] [JPEG] [JS] [LAYOUT] [LIB] NEWLINE [MODULES] [NAV-JAVA] [NSPRPUB] [SUN-JAVA] [VIEW] NEWLINE [WEBSHELL] [WIDGET] [XPCOM] LEGAL LICENSE NGLAYOUT.MAK NEWLINE NGLAYOUT.MK TODAY.INC NEWLINE NEWLINE 5 file(s) 40,026 bytes NEWLINE 22 dir(s) 816,300,032 bytes free NEWLINE NEWLINE [WINDOWS] C:\\raptor_source\\mozilla>nmake /f nglayout.mak NEWLINE Microsoft (R) Program Maintenance Utility Version 1.62.7022 NEWLINE Copyright (C) Microsoft Corp 1988-1997. All rights reserved. NEWLINE NEWLINE NMAKE : fatal error U1081: NEWLINE 'D:\\DL\\wintools\\windows\\bin\\unix_date' NEWLINE : program not found Stop. NEWLINE NEWLINE [WINDOWS] C:\\raptor_source\\mozilla>NGLAYOUT_HELP!! NEWLINE Bad Command Or Filename
528|Toolbars absent + no fix for first bwsr session;|Created by (thecubic@somethingorother.com) on Wednesday, August 12, 1998 5:47:15 PM PDT NEWLINE Additional Details : NEWLINE no error messages, not toolbars present
529|mozilla-export crashes on startup|Created by Julien PLISSONNEAU DUQUENE (jplissd@atlantic-line.fr) on Friday, August 14, 1998 9:19:25 AM PDT NEWLINE Additional Details : NEWLINE NEWLINE I compiled mozilla-export (with a few problems, makefile, NEWLINE #include changes, file copies ...) with NEWLINE egcs-2.90.29 980515 NEWLINE glibc-2.0.6 NEWLINE lesstif (recent version) NEWLINE NEWLINE However, like communicator binary version I downloaded from NEWLINE Netscape site, it crashes when I try to launch it, after two NEWLINE warning. NEWLINE NEWLINE It still displays -help and -version. NEWLINE NEWLINE Unfortunately I can't investigate more now.
530|Navigator seg faults when resizing main browser window|Created by Kevin Kreamer (kreamer@hotmail.com) on Saturday, August 15, 1998 7:04:30 PM PDT NEWLINE Additional Details : NEWLINE It hangs for a second then crashes with "Segmentation NEWLINE Fault." It doesn't dump core, though.
531|Allow config directories different from ~/.netscape/|Created by Aleksey Nogin (ayn2@cornell.edu) on Sunday, August 16, 1998 12:12:23 PM PDT NEWLINE Additional Details : NEWLINE Currently Mozilla still uses ~/.netscape/ directory for its NEWLINE config files, although ~/.mozilla/ may seem more NEWLINE appropriate. Authors of different "brands" of Mozilla may NEWLINE want to have their programs to use separate directories and NEWLINE users may also want to use different versions of Mozilla NEWLINE with different config files. NEWLINE NEWLINE I wrote a patch (it is derived from the patch by Dan McGuirk NEWLINE <djm@indirect.com>) that adds a new run-time option NEWLINE -configdir <directory> that allows to specify a config NEWLINE subdirectory different from ".netscape" and also allows to NEWLINE change the compiled in default by changing one line of the NEWLINE ocode instead of hunting for all these hardcoded strings NEWLINE through all the sources. NEWLINE NEWLINE I am trying to keep the patch up-to-date with the current NEWLINE CVS sources and you should always be able to find them NEWLINE (divided into files according to modules) at NEWLINE ftp://ftp.cs.cornell.edu/pub/nogin/mozilla/
532|CViewerSecurity dialog doesn't save info properly.|Created by (shaman@nlc.net.au) on Friday, August 21, 1998 7:39:49 PM PDT NEWLINE Additional Details : NEWLINE When I click on a file that is opened by an external NEWLINE program and select "open it" as well as "don't show this NEWLINE again", it works for that instance of the program, but it NEWLINE forgets it the next time I load communicator.
533|Java applets force focus to it's parent window.|Created by (shaman@nlc.net.au) on Friday, August 21, 1998 7:43:42 PM PDT NEWLINE Additional Details : NEWLINE When applets finish loading, they force the focus to their NEWLINE parent page. This is increadibly annoying, and is downright NEWLINE dangerous. There have been several instances when I've tried NEWLINE to click on a link the instant before a Java applet switched NEWLINE to another navigator window, where my click resulted in a NEWLINE completely different effect. I have to turn Java off, NEWLINE because when I open 5-10 news sites, they visibly 'flicker' NEWLINE as the focus jumps from one window to another.
534|Saving bookmarks corrupts dir ownership|Created by Doug Schaapveld (djschaap@eng.iastate.edu) on Wednesday, August 26, 1998 7:29:27 AM PDT NEWLINE Additional Details : NEWLINE On IRIX 6.2/6.3, saving the bookmarks file as (for example) NEWLINE /home/djschaap results in the owner of /home/djschaap being NEWLINE changed to 270614016 and group being changed to 9857420 NEWLINE (obviously not correct). Admin must then go in & reset NEWLINE directory permissions for the affected users. This appears NEWLINE to occur with NIS and local users, NFS and local home NEWLINE directories. This "feature" has been around since at lest NEWLINE Navigator 4.02. Under Linux, the same process will save the NEWLINE bookmarks in a file called -0nnnnn in the specified NEWLINE directory, where nnnnn is the PID of the mozilla/netscape NEWLINE process. This appears to be the "correct" behavior in this NEWLINE case.
535|Can't do a MOZ_LITE build|It's not possible to do a MOZ_LITE build from the original 28th July 1998 NEWLINE Mozilla tarball. Yes, I know the recommended default is to do a MOZ_MEDIUM NEWLINE build, but some of us don't want Composer eating RAM thank you :-) NEWLINE NEWLINE I managed to do a MOZ_LITE build on HP-UX by changing NEWLINE lib/layout/Makefile by moving the following CSRCS source files NEWLINE outside of the ifdef EDITOR section and into the main NEWLINE section: NEWLINE NEWLINE layfree.c, layimage.c, layout.c, laysel.c, laytable.c, NEWLINE laytags.c, laytext.c, layutil.c NEWLINE NEWLINE (in other words, the only thing left in the CSRCS += line NEWLINE inside ifdef EDITOR was layedit.c and $(NULL)) NEWLINE NEWLINE The other change I had to make was to cmd/xfe/src/FindDialog.cpp. NEWLINE None of the EDT_* functions should be called when doing a MOZ_LITE build. NEWLINE Unfortunately, fe_replace_db() calls EDT_ReplaceText(), so NEWLINE #ifdef EDITOR was put around the whole fe_replace_db() function. This, of NEWLINE course, meant that #ifdef EDITOR was needed around the code that called this NEWLINE function too. Line 498 onwards of FindDialog.cpp now reads: NEWLINE NEWLINE #ifdef EDITOR NEWLINE if (replaceP) NEWLINE { NEWLINE ... NEWLINE down to line 550: NEWLINE ... NEWLINE else // not replaceP NEWLINE #endif NEWLINE { NEWLINE XtVaSetValues (find_text, NEWLINE ... NEWLINE NEWLINE One final note - is it wise to use the environmental variable NEWLINE EDITOR in the various Mozilla Makefiles to determine if NEWLINE Composer is built ? EDITOR is a standard UNIX env. var. to NEWLINE indicate your default text editor and is often defined in a NEWLINE user's environment ! It means that you have to remember to NEWLINE unset EDITOR prior to any MOZ_LITE build...ho hum... NEWLINE NEWLINE Note that I've fixed this (and other problems under HP-UX) via NEWLINE a patch available here: NEWLINE http://hpux.csc.liv.ac.uk/hppd/mozilla/mozpatch-5.00.tar.gz NEWLINE (this contains a file mozilla.patch, which is the context diffs for NEWLINE all HP-UX problems to date - documentation on the changes in the patch NEWLINE are in the HPUX.Install file)
536|Error if Mozilla started off-line|When starting up off-line (i.e. no Net connection, including no DNS), NEWLINE Mozilla pops up a warning: NEWLINE "A network error occurred: unable to connect to server NEWLINE (TCP Error: XP_GetBuiltinString: 4420 not found) NEWLINE The server may be down or unreachable" NEWLINE However, it's then quite happy to surf a Web server running on the localhost. NEWLINE Looking at modules/rdf/src/ht.c and modules/rdf/src/rdfht.c, it would appear NEWLINE that Mozilla bizarrely attempts to access this URL by default whenever it NEWLINE starts: http://rdf.netscape.com/rdf/navcntr.rdf NEWLINE NEWLINE This isn't good if you're not connected to the Net at the time ! Surely a NEWLINE local copy of this should be used and perhaps update this local copy when NEWLINE you're next connected to the Net "live" (assuming you can tell if you're NEWLINE online vs. offline), so that when you're next offline, it'll have an NEWLINE up-to-date local copy of the RDF file to use.
537|Can't close View->Page Source or View->Page Info windows|There is no button or pull-down menu to close either the NEWLINE View->Page Source or View->Page Info windows. The keyboard short-cut NEWLINE ALT-W, which used to work in Netscape 3.X, no longer works in Netscape 4.X NEWLINE or any version of Mozilla. In exasperation (after getting a lot of source/info NEWLINE windows left around in a session !), I eventually had to reconfigure my NEWLINE window manager to support "Kill Window", but it's an inconvenient NEWLINE way to close the offending windows. Not sure if this applies to HP-UX only, but NEWLINE it's annoying enough to stop me using Netscape 4.X or Mozilla ! NEWLINE NEWLINE BTW, I can't search the View->Page Source window, which is maddening ! NEWLINE The PC version can do this (with a bit of trickery...it's non-obvious on that NEWLINE platform though), but the UNIX one can't ! I end up having to save the HTML NEWLINE source to a file and grep'ing it (for links, tags etc), which is ridiculous.
538|Bugzilla can't search bug lists for a text match|Bugzilla currently has no way to type in a word or phrase and have it NEWLINE come back with a list of bugs that match that word/phrase. Without this NEWLINE facility, it meant that I've duplicate-reported some bugs by mistake, NEWLINE without realising they'd already been reported. NEWLINE NEWLINE To save on CPU cycles, perhaps only the Summary line needs to be searched, NEWLINE rather than the Description body ? NEWLINE NEWLINE Also, when submitting a bug, there appears to be no mechanism to avoid NEWLINE (or at least warning about the possibility of) duplicate bug reports. NEWLINE Perhaps a keyword system or something would have to be used - if your bug NEWLINE report matches keywords, list those matching bug reports (allowing the user to NEWLINE check them out) as a warning before accepting a new bug report. Or maybe NEWLINE you folks could figure something out better than this, I dunno NEWLINE NEWLINE BTW, why are there no "Bugzilla components" ? I'm forced to choose a component NEWLINE when reporting this bug, but they all look like Mozilla components to me !
539|enter_bug.cgi chokes if we have multiple products.|It gets the version and component field wrong, and therefore the default NEWLINE assigned-to field wrong. Oops! NEWLINE NEWLINE Hopefully, people will manually assign Bugzilla bugs to terry@mozilla.org until NEWLINE I get this fixed.
541|Sorting the initial query by "assignee" doesn't work.|Fixed by a trivial change to buglist.cgi.
542|Whine at people when they have bugs that have been NEW for longer than a week.|It's gotta be done; otherwise, bugs will just sit around and rot.
543|Bug list should not be titled "Mozilla bug list"|When I asked to give me a list of all Bugzilla bugs in Bugzilla, NEWLINE buglist.cgi gave me a list titled "Mozilla bug list". I think NEWLINE you should drop "Mozilla" from the title
7195|UMR in xptcall|reassigning to roger.
545|There should be a way to enter a version number for "other"|It would be nice if you could add some extra possible values to the "Version" NEWLINE field. NEWLINE I would suggest adding: NEWLINE - A "CVS image" with an extra field for the date/time of the last "cvs NEWLINE update" NEWLINE - An extra text field for the "Other" choice NEWLINE - An "Unknown" (with no extra fields).
546|Adding "CC" field does not work|When type any address (no matter, mine or some other) to the CC field and press NEWLINE "Commit", I get an error message. NEWLINE NEWLINE The error message from my attempt to update bug #513 follows. The error message NEWLINE from bug #464 was similar. NEWLINE NEWLINE <TITLE>Update Bug 513</TITLE> NEWLINE <B>Bug List:</B> (88 of 96) NEWLINE <A HREF="show_bug.cgi?id=239">First</A> NEWLINE <A HREF="show_bug.cgi?id=350">Last</A> NEWLINE <A HREF="show_bug.cgi?id=511">Prev</A> NEWLINE <A HREF="show_bug.cgi?id=515">Next</A> NEWLINE <A HREF="query.cgi">Query page</A> NEWLINE <HR> NEWLINE Error: mysqlsel/db server: parse error near NEWLINE ''assigned_to','ramiro@netscape.com','')' at line 1 NEWLINE while executing NEWLINE "mysqlsel $mysqlhandle $str" NEWLINE (procedure "SendSQL" line 3) NEWLINE invoked from within NEWLINE "SendSQL $q" NEWLINE invoked from within NEWLINE "if {![cequal $old $new]} { NEWLINE if {![info exists whoid]} { NEWLINE set whoid [DBNameToIdAndCheck $FORM(who)] NEWLINE SendSQL " ..." NEWLINE ("foreach" body line 4) NEWLINE invoked from within NEWLINE "foreach col $log_columns { NEWLINE set old [lvarpop oldvalues] NEWLINE set new [lvarpop newvalues] NEWLINE if {![cequal $old $new]} { NEWLINE if { ..." NEWLINE ("foreach" body line 48) NEWLINE invoked from within NEWLINE "foreach id $idlist { NEWLINE SendSQL "lock tables bugs write, bugs_activity write, cc write, profiles NEWLINE write" NEWLINE set oldvalues [SnapShotBug $id] NEWLINE NEWLINE set ..." NEWLINE (file "/opt/suitespot/docs/webtools/bugzilla/process_bug.cgi" line 226)
547|TextFields in Java are to "small" for Fonts|On Windows & Macintosh a TextField which is 20 Pixel height, is enough for a 12 NEWLINE Pixel Font. With X-Windows (HP/UX, AIX, anything !) the TextField must be 30 NEWLINE Pixel in height, else only 2 Pixels of the 12 Pixel-Font are visible. NEWLINE It seems the Insets of TextFields for X-Windows is much to big !? NEWLINE At the moment i must use System.getProperty("os.name"); to "fix" this bug, but NEWLINE this is not 100% Pure Java !
548|Restore the "show my bugs" from the old Bugzilla|The old Bugzilla had a nice feature - it had a link on the query page that NEWLINE allowed to see all "my" bugs (including those where I added myself to the CC NEWLINE list). It would be nice to have something similar in the new Bugzilla. A NEWLINE reasonable choice may be to add a checkbox to the query form with effect of NEWLINE showing only those bugs that have my e-mail somewhere (in CC list, in NEWLINE Description field - anywhere).
549|BUILD: mozilla-export uses shared -lplds21 -lplc21 -lmsgc21 -lnspr21 -lzlib -ljs -ljsj libraries.|My build of mozilla-export doesn't work anymore. make mozilla-export gives: NEWLINE ... NEWLINE ../../dist/Linux2.0.34_x86_DBG.OBJ/lib/libxpcom NEWLINE .a ../../dist/Linux2.0.34_x86_DBG.OBJ/lib/librdf.a -L../../dist/Linux2.0.34_x86 NEWLINE _DBG.OBJ/bin -lplds21 -lplc21 -lmsgc21 -lnspr21 -lzlib -ljs -ljsj -L/usr/X11 NEWLINE R6/lib -lXm -lXt -lXmu -lXpm -lX11 -lXext -lSM -lICE -lm -ldl NEWLINE ... NEWLINE (-lXm handhacked in cmd/xfe/Makefile) NEWLINE NEWLINE moz-export builds ok, but hasn't ever run here. NEWLINE NEWLINE I use gcc 2.7.2.2, lesstif-cvs, libc5.2.18, mozilla-cvs. NEWLINE NEWLINE Btw, a make with autoconf gives: NEWLINE ... NEWLINE cd nsprpub; make export NEWLINE make[1]: Entering directory NEWLINE `/fire/pub/cvs/moz/mozilla/obj-i586-pc-linux/nsprpub' NEWLINE cd config; make export NEWLINE make[2]: Entering directory NEWLINE `/fire/pub/cvs/moz/mozilla/obj-i586-pc-linux/nsprpub/config' NEWLINE gcc -o Linux2.0.34_x86_AC.OBJ/nsinstall -g -fPIC -DFORCE_PR_LOG -DXP_UNIX NEWLINE -DDEBUG -UNDEBUG -DDEBUG_jip -D_PR_LOCAL_THREADS_ONLY NEWLINE Linux2.0.34_x86_AC.OBJ/nsinstall.o Linux2.0.34_x86_AC.OBJ/pathsub.o NEWLINE Linux2.0.34_x86_AC.OBJ/nsinstall.o(.text+0xa85): undefined reference to `lchown' NEWLINE make[2]: *** [Linux2.0.34_x86_AC.OBJ/nsinstall] Error 1 NEWLINE make[2]: Leaving directory NEWLINE `/fire/pub/cvs/moz/mozilla/obj-i586-pc-linux/nsprpub/config' NEWLINE make[1]: *** [export] Error 2 NEWLINE make[1]: Leaving directory `/fire/pub/cvs/moz/mozilla/obj-i586-pc-linux/nsprpub' NEWLINE make: *** [export] Error 2 NEWLINE ... NEWLINE NEWLINE succes, NEWLINE Jip
550|BUILD: compile error with glibc 2.1|Mozilla will not compile under glibc 2.1 w/o a patch similiar to the one at NEWLINE http://gate.crashing.org/~trini/mozilla.diff This patch also cleans up some NEWLINE "problems" with Linux/PPC building (-DMKLINUX is ugly and very unneeded)
551|compile error with glibc 2.1|Mozilla will not compile under glibc 2.1 w/o a patch similiar to the one at NEWLINE http://gate.crashing.org/~trini/mozilla.diff This patch also cleans up some NEWLINE "problems" with Linux/PPC building (-DMKLINUX is ugly and very unneeded)
552|Resize stacks all pictures into upper right corner|Resizing the window stacks all pictures into upper right corner.
553|Need better querying of assigned_to and reported_by fields.|When you type in a user's e-mail address, it should check syntax for NEWLINE a valid address. There should also be ways of typing a substring NEWLINE or a regexp in for the address.
555|Need a version-numbering scheme for bugzilla source.|I think I should introduce a version number in the bugzilla source, and NEWLINE make tarballs. NEWLINE NEWLINE My current thoughts are: NEWLINE NEWLINE - Create a file called "version.tcl", which contains simply a line like NEWLINE set versionnumber "1.4" NEWLINE NEWLINE - Make the version number appear in the header or something. NEWLINE NEWLINE - Whenever I feel it's time to spin a new tarball, I'll up the version number, NEWLINE create the tarball, and then UP THE NUMBER AGAIN. That way, the running NEWLINE version will always have a different version number than the one in the NEWLINE tarball. So, I can still check in changes whenever I feel like it, and not NEWLINE worry that my thing is different than something out there wrapped in a NEWLINE tarball.
556|Bugzilla does not work with tcl 8.0p2, mysqltclsh 1.53, mysql 3.22.5alpha|I tried to use bugzilla with mentioned in summary software but get the following NEWLINE error in apache log. NEWLINE NEWLINE Seems it is some incompatibility.... NEWLINE NEWLINE -------- error.log ------- NEWLINE invalid command name "cequal" NEWLINE while executing NEWLINE "cequal $remaining """ NEWLINE (procedure "ProcessFormFields" line 6) NEWLINE invoked from within NEWLINE "ProcessFormFields $buffer" NEWLINE (file "CGI.tcl" line 91) NEWLINE invoked from within NEWLINE "source "CGI.tcl"" NEWLINE (file "/home/dyp/html/bugzilla/query.cgi" line 22) NEWLINE [Thu Sep 3 10:32:46 1998] [error] Premature end of script headers: /home/dyp/ht NEWLINE ml/bugzilla/query.cgi NEWLINE ------ end --------
557|shouldn't have to type http:// all the time|First official raptor bug! I'm tired of typing http:// all the time. Jud, please NEWLINE fix this. NEWLINE NEWLINE Just copied scullin and jevering on this to let them know that Raptor has been NEWLINE added has a "Product" in Scopus so bugs can be filed, queried, etc. The default NEWLINE version is 1.0.
558|two minor related links bugs|1. Should be called What's Related, if we want to match what 4.06/4.5 says. NEWLINE 2. If no sites are found, should say "Nothing Available". Right now it's just NEWLINE blank.
559|Scrolling kills resources on win95|Scrolling on win95 sucks resources dry (like a dollar h....) Even the first NEWLINE test0.html page sucks about 1Mb per scroll if you have it half minimized and NEWLINE scroll from top to down. Haven't noticed it on NT though.
560|Ugly residue between page loads on Win95|On my win95 machine, there is some ugly leftover residue from the previous page NEWLINE just as a new page loads. Happens between all page loads and it's annoying NEWLINE factor is proportional to the amount of color and graphics on the previous page. NEWLINE NEWLINE On WinNT, it doesn't really do this, though there is an annoying flashing. NEWLINE Course this may have to do with what my colors are set at, although it's 16-bit NEWLINE on both machines.
561|Can't navigate around pages|page up, page down, tab, arrows don't do anything
562|Mozilla doesn't redraw properly when Raptor is running.|The toolbars and such of Mozilla wouldn't redraw properly while Raptor was NEWLINE running. When I killed Raptor and refreshed Mozilla everything worked fine NEWLINE again. The promised lack of interaction appears to be broken somehow. NEWLINE NEWLINE I'm doing this on an HP Omnibook 800CT with a 7/1 version of Mozilla. NEWLINE NEWLINE I also just noticed that the little titles under my desktop icons had the same NEWLINE problem when they got covered over by Raptor. Perhaps it's not forwarding a NEWLINE message to invalidate screen areas after it gives them up?
563|Redirects don't work|We need to enable redirects in order to browse some of the top sites that use NEWLINE redirects. For example, almost all banner ads use a redirect (you click, it logs NEWLINE the ad, and redirects you to the advertister site).
564|ISO-8859-1 should be converted as CP 1252|There are several blank lines where there should be text in the article at NEWLINE http://www.microsoft.com/presspass/features/7-23realnet.htm
565|After setting bugsplat login, all pages crash|Goto http://scopus/bugsplat. Login with your username/password. You are sent to NEWLINE a confirmation page. All subsequent pages will crash.
566|Puke appears on screen in between page loads on Win95|1. Go to URL NEWLINE 2. Click on a link NEWLINE 3. (wait) NEWLINE 4. New page loads. NEWLINE NEWLINE During (3), you see what appears to be a random dump of memory painted in the NEWLINE webshell area. It looks like a scrambled cable TV station.
567|Images disappear from pages after the first time they're viewed|1. Go to http://www.yahoo.com/ NEWLINE 2. Notice the big Yahoo! logo at the top. NEWLINE 3. Click on a link to travel to a new page or enter a new URL NEWLINE 4. Press the back button or enter Yahoo URL again NEWLINE 5. Notice the big Yahoo! logo is no longer there. NEWLINE NEWLINE Also, instead of steps 3&4, you can just move the viewer window off the edge of NEWLINE the desktop and then move it back and you'll notice the Yahoo logo doesn't NEWLINE repaint.
568|NGLayout Crashes on this URL trying parse cookie date|Crashes every time... NEWLINE NEWLINE Here's a stack trace: NEWLINE MSVCRTD! 102406c4() NEWLINE NET_ParseDate(char * 0x01036c5b) line 577 + 20 bytes NEWLINE NET_SetCookieStringFromHttp(int 102, URL_Struct_ * 0x010c1a90, MWContext_ * NEWLINE 0x010c1d60, char * 0x010c1cc0, char * 0x01036c2c) line 2416 + 14 bytes NEWLINE NET_ParseMimeHeader(int 102, MWContext_ * 0x010c1d60, URL_Struct_ * 0x010c1a90, NEWLINE char * 0x01036c20, char * 0x01036c2c, int 1) line 1617 + 28 bytes NEWLINE net_parse_http_mime_headers(_ActiveEntry * 0x010c1ef0) line 1982 + 35 bytes NEWLINE net_ProcessHTTP(_ActiveEntry * 0x010c1ef0) line 3506 + 9 bytes NEWLINE NET_ProcessNet(PRFileDesc * 0x010c1380, int 2) line 3266 + 13 bytes NEWLINE NET_PollSockets() line 180 + 18 bytes NEWLINE nsNativeViewerApp::Run() line 68 NEWLINE main(int 1, char * * 0x00fe4af0) line 109 NEWLINE mainCRTStartup() line 338 + 17 bytes NEWLINE KERNEL32! 77f1b304()
569|Make the Beep Stop!|Hitting Return in the URL Bar causes a beep for those machines that are NEWLINE auditorially endowed. Make it stop, please.
570|Courier, Times New Roman fonts don't display in Test 0|Test reads: NEWLINE "This is a paragraph with font variations: Arial, Verdana, Courier, Times New NEWLINE Roman." NEWLINE NEWLINE Notice that Verdana is actually italicized Arial, Courier is Arial, and Times NEWLINE New Roman is Arial.
571|Crash when Copying to Clipboard|1. Load test 0. NEWLINE 2. Toggle on selection via debug menu NEWLINE 3. Highlight some text (Example 0, for instance) NEWLINE 4. From Edit menu, select "Copy." NEWLINE 5. Viewer crashes within a second or so.
572|Release Build Crashes on Mouseovers|When you mouseover any content element in a release build, it crashes.
573|LINK'd style sheets do not work|We don't seem to be loading style sheets via the LINK tag. Example: NEWLINE NEWLINE <LINK rel="stylesheet" type="text/css" href="linktest.css" title="Link1">
574|Whitespace is chopped after closing tags|Sample HTML: NEWLINE NEWLINE <html> NEWLINE <body> NEWLINE <H3>This sentence should show <STRONG>strong</STRONG> and <EM>emphasized</EM> NEWLINE text.</H3> NEWLINE </body> NEWLINE </html> NEWLINE NEWLINE Result: NEWLINE ... strongand ... emphasizedtext NEWLINE NEWLINE I think this is a whitespace issue (kipp?) but maybe it has to do with the NEWLINE parser (cc'ing rickg).
575|Underlined <TT>Foo</TT> content has low underline|Sample HTML: NEWLINE NEWLINE <html> NEWLINE <body> NEWLINE <u>Hello <tt>world,</tt> how are you today?</u> NEWLINE </body> NEWLINE </html> NEWLINE NEWLINE The content should have a single unbroken underline appearing at a consistent NEWLINE vertical location, but below the word "world" (inside a TT), the underline is a NEWLINE pixel too low, making the underline appear "broken."
576|<STRONG> doesn't work|Content inside the <STRONG> tag should be treated like it's inside a <B> tag. NEWLINE Currently we seem to ignore <STRONG>. Not sure if this is a style or parser NEWLINE issue.
577|Clicking on a link doesn't make it "ACTIVE"|When you mousedown on a link, the link should become "active" see sample html: NEWLINE NEWLINE <style type="text/css"> NEWLINE A { color: blue; } NEWLINE A:ACTIVE { color: orange; } NEWLINE </style> NEWLINE <body> NEWLINE <a href="http://www.netscape.com/">Netscape Netcenter</a> NEWLINE </body> NEWLINE NEWLINE Not sure if this is events (joki) stuff or style (peterl) stuff. NEWLINE NEWLINE ------- Additional Comments From joki 08/02/98 01:31 ------- NEWLINE NEWLINE Well the color it changes is style stuff but that actual changing is mine so I NEWLINE own this one.
578|CSS First-line selector doesn't work|I assume that we don't have APIs into the content model to get the first line NEWLINE (kipp) but this is a stlye thing too (peterl). NEWLINE NEWLINE <style type="text/css"> NEWLINE P:first-line {color: green;} NEWLINE </style> NEWLINE <body> NEWLINE <p>The first line only should be green. The rest should be default NEWLINE text color (black) bla bla bla bla bla bla bla bla bla bla NEWLINE hope this makes it to the second line bla bla bla. NEWLINE </body>
579|{feature} CSS First-letter selector isn't finished|See URL given for sample HTML and test case. Again (similar to 311767) I expect NEWLINE we don't have the api yet in the content model, but ultimately it's a style NEWLINE thing.
580|CSS !important doesn't work|See URL given for test case and sample HTML / CSS.
581|CSS "must have" fonts cursive and fantasy aren't supported|In order to comply with CSS1, we need to support "Cursive" and "Fantasy" fonts. NEWLINE NEWLINE See URL for test case and sample code. Is this a libfont issue (does raptor use NEWLINE libfont)? If so, please reassign to dp.
582|{css1} Monospace font inside a serif div doesn't work|Sample HTML: NEWLINE NEWLINE <head> NEWLINE <style type="text/css"> NEWLINE DIV.a {font-family: serif;} NEWLINE .five {font-family: monospace;} NEWLINE </style> NEWLINE </head> NEWLINE <body> NEWLINE <div class="a"> NEWLINE I should be serif. NEWLINE <P class="five"> NEWLINE I should be monospace. NEWLINE </P> NEWLINE </div> NEWLINE </body> NEWLINE NEWLINE Note if you replace "monospace" with "sans-serif," it works, so I think it has NEWLINE something to do with the monospace font in particular.
583|Oblique font style not supported|I think we're supposed to treat it like italics. See URL for test case and NEWLINE sample code.
584|Small-caps font variant not supported|See URL for test case and sample code. Again, not sure if this is a libfont NEWLINE thing or a style thing...
585|line-height property not supported|See URL for test case and sample code. line-height is a CSS property to control NEWLINE the vertical space between lines of text. I imagine this will involve the NEWLINE vertical spacing of elements in the content model (cc'ing kipp). NEWLINE NEWLINE NEWLINE ------- Additional Comments From angus 08/02/98 02:16 ------- NEWLINE NEWLINE Actually, another, better test case is here: NEWLINE http://www.w3.org/Style/CSS/Test/current/sec548.htm NEWLINE NEWLINE Unfortunately, this case crashes my viewer on win95.
586|Scrolling results in rendering weirdness on win95|To see bug: NEWLINE NEWLINE 1. Use win95 (I'm not sure if this bug exists on winNT; haven't tested there NEWLINE yet) NEWLINE 2. Load URL: NEWLINE http://www.w3.org/Style/CSS/Test/current/sec536.htm NEWLINE 3. Scroll NEWLINE NEWLINE This page has a CSS property called "background-position" that could be part of NEWLINE the problem, but at first glance it appears to be a rendering thing.
587|Text transforms don't work|Text transforms are things like toUpperCase toLowerCase etc. See URL for test NEWLINE case and sample code.
588|{feature} [TEXT] Full justification of text doesn't work|We don't support justified text yet. Does this fit in with the whitespace work? NEWLINE It's also a style issue. See URL for test case and sample code.
589|Mixed style lists aren't counted properly|When you have mixed style list items, they don't work right. Sample code: NEWLINE <head> NEWLINE <style type="text/css"> NEWLINE .letters {list-style-type: upper-alpha;} NEWLINE .numbers {list-style-type: decimal;} NEWLINE </style> NEWLINE </head> NEWLINE <body> NEWLINE <OL class="letters"> NEWLINE <LI>This list should feature... NEWLINE <LI>...letters for each item... NEWLINE <LI class="numbers">...except this one. NEWLINE </OL> NEWLINE </body> NEWLINE NEWLINE How it should appear: NEWLINE A. This list should feature... NEWLINE B. ... leters for each item... NEWLINE 3. ... except this one NEWLINE NEWLINE Instead the third item shows up as "2." not "3." like this: NEWLINE 2. ... except this one
590|List items don't support images as bullets|See URL for test case and sample code. I'm guessing that kipp is the right NEWLINE person to update the LI element to handle images for bullets, and peterl can NEWLINE hook up style to this.
591|top border is one pixel too short horizontally|See URL for test case and sample code. The problem looks like this (with a thick NEWLINE top border): NEWLINE NEWLINE ------ NEWLINE ------- NEWLINE \| foo \| NEWLINE -------
592|NGLayout crashes when loading this URL|Stack trace: NEWLINE NEWLINE NTDLL! 77f76148() NEWLINE nsDebug::NotYetImplemented(char * 0x0054c624, char * 0x0054c5f0, int 434) line NEWLINE 105 + 13 bytes NEWLINE CSSStyleRuleImpl::CalcLength(const nsCSSValue & {...}, const nsStyleFont * NEWLINE 0x014a7e28, nsIPresContext * 0x0106e430) line 434 + 21 bytes NEWLINE CSSStyleRuleImpl::MapStyleInto(nsIStyleContext * 0x01459b00, nsIPresContext * NEWLINE 0x0106e430) line 613 + 32 bytes NEWLINE MapStyleRule(nsISupports * 0x01046be0, void * 0x0012eb80) line 954 NEWLINE SupportsArrayImpl::EnumerateBackwards(SupportsArrayImpl * const 0x010c5350, int NEWLINE (nsISupports *, void *)* 0x004a5610 MapStyleRule(nsISupports *, void *), void * NEWLINE 0x0012eb80) line 320 + 20 bytes NEWLINE StyleContextImpl::RemapStyle(nsIPresContext * 0x0106e430) line 983 NEWLINE StyleContextImpl::StyleContextImpl(nsIStyleContext * 0x014a7df8, NEWLINE nsISupportsArray * 0x010c5350, nsIContent * 0x010c1510, nsIPresContext * NEWLINE 0x0106e430) line 663 NEWLINE NS_NewStyleContext(nsIStyleContext * * 0x0012ebfc, nsIStyleContext * 0x014a7df8, NEWLINE nsISupportsArray * 0x010c5350, nsIContent * 0x010c1510, nsIPresContext * NEWLINE 0x0106e430) line 1061 + 50 bytes NEWLINE StyleSetImpl::GetContext(nsIPresContext * 0x0106e430, nsIFrame * 0x0105bd50, NEWLINE nsIContent * 0x010c1510, nsIStyleContext * 0x014a7df8, nsISupportsArray * NEWLINE 0x010c5350, int 0) line 514 + 25 bytes NEWLINE StyleSetImpl::ResolveStyleFor(nsIPresContext * 0x0106e430, nsIContent * NEWLINE 0x010c1510, nsIFrame * 0x0105bd50, int 0) line 556 + 32 bytes NEWLINE nsPresContext::ResolveStyleContextFor(nsIContent * 0x010c1510, nsIFrame * NEWLINE 0x0105bd50, int 0) line 204 + 27 bytes NEWLINE nsHTMLBase::CreateFrame(nsIPresContext * 0x0106e430, nsIFrame * 0x0105bd50, NEWLINE nsIContent * 0x010c1510, nsIFrame * 0x00000000, nsIFrame * & 0x010c5290) line NEWLINE 146 + 21 bytes NEWLINE nsCSSBlockFrame::CreateNewFrames(nsIPresContext * 0x0106e430) line 1647 + 23 NEWLINE bytes NEWLINE nsCSSBlockFrame::FrameAppendedReflow(nsCSSBlockReflowState & {...}) line 1512 + NEWLINE 15 bytes NEWLINE nsCSSBlockFrame::ReflowAround(nsCSSBlockFrame * const 0x0105bda0, nsIPresContext NEWLINE & {...}, nsISpaceManager * 0x01054a70, nsReflowMetrics & {...}, const NEWLINE nsReflowState & {...}, nsRect & {...}, unsigned int & 0) line 1269 + 18 bytes NEWLINE nsBodyFrame::Reflow(nsBodyFrame * const 0x0104f570, nsIPresContext & {...}, NEWLINE nsReflowMetrics & {...}, const nsReflowState & {...}, unsigned int & 0) line 218 NEWLINE nsContainerFrame::ReflowChild(nsIFrame * 0x0104f570, nsIPresContext * NEWLINE 0x0106e430, nsReflowMetrics & {...}, const nsReflowState & {...}) line 498 NEWLINE RootContentFrame::Reflow(RootContentFrame * const 0x0104e550, nsIPresContext & NEWLINE {...}, nsReflowMetrics & {...}, const nsReflowState & {...}, unsigned int & 0) NEWLINE line 326 + 27 bytes NEWLINE nsContainerFrame::ReflowChild(nsIFrame * 0x0104e550, nsIPresContext * NEWLINE 0x0106e430, nsReflowMetrics & {...}, const nsReflowState & {...}) line 498 NEWLINE RootFrame::Reflow(RootFrame * const 0x0104eed0, nsIPresContext & {...}, NEWLINE nsReflowMetrics & {...}, const nsReflowState & {...}, unsigned int & 0) line 132 NEWLINE + 27 bytes NEWLINE nsHTMLReflowCommand::Dispatch(nsHTMLReflowCommand * const 0x010c51a0, NEWLINE nsIPresContext & {...}, nsReflowMetrics & {...}, const nsSize & {...}) line 133 NEWLINE PresShell::ProcessReflowCommands() line 567 NEWLINE PresShell::ExitReflowLock(PresShell * const 0x010c2920) line 371 NEWLINE PresShell::ContentAppended(PresShell * const 0x010c2924, nsIDocument * NEWLINE 0x0106d7b0, nsIContent * 0x01049e00) line 654 NEWLINE nsDocument::ContentAppended(nsIContent * 0x01049e00) line 504 NEWLINE nsHTMLContainer::AppendChild(nsHTMLContainer * const 0x01049e00, nsIContent * NEWLINE 0x010c1510, int 1) line 192 NEWLINE HTMLContentSink::AppendToCorrectParent(nsHTMLTag eHTMLTag_body, nsIHTMLContent * NEWLINE 0x01049e00, nsHTMLTag eHTMLTag_p, nsIHTMLContent * 0x010c1510, int 1) line 1429 NEWLINE HTMLContentSink::CloseContainer(HTMLContentSink * const 0x01467028, const NEWLINE nsIParserNode & {...}) line 891 NEWLINE CNavDTD::CloseContainer(const nsIParserNode & {...}, nsHTMLTag eHTMLTag_p, int NEWLINE 1) line 2332 + 22 bytes NEWLINE CNavDTD::CloseContainersTo(int 2, nsHTMLTag eHTMLTag_p, int 1) line 2366 + 26 NEWLINE bytes NEWLINE CNavDTD::CloseContainersTo(nsHTMLTag eHTMLTag_p, int 1) line 2387 + 20 bytes NEWLINE CNavDTD::HandleEndToken(CToken * 0x010702e0) line 749 + 17 bytes NEWLINE NavDispatchTokenHandler(CToken * 0x010702e0, nsIDTD * 0x014861c0) line 271 + 12 NEWLINE bytes NEWLINE CTokenHandler::operator()(CToken * 0x010702e0, nsIDTD * 0x014861c0) line 80 + 14 NEWLINE bytes NEWLINE CNavDTD::HandleToken(CNavDTD * const 0x014861c0, CToken * 0x010702e0) line 489 + NEWLINE 18 bytes NEWLINE nsParser::BuildModel() line 578 + 16 bytes NEWLINE nsParser::ResumeParse() line 526 NEWLINE nsParser::OnDataAvailable(nsParser * const 0x0106dcb4, nsIURL * 0x01064ef0, NEWLINE nsIInputStream * 0x0106c1a0, int 3704) line 757 + 15 bytes NEWLINE nsDocumentBindInfo::OnDataAvailable(nsDocumentBindInfo * const 0x01064e90, NEWLINE nsIURL * 0x01064ef0, nsIInputStream * 0x0106c1a0, int 3704) line 904 + 30 bytes NEWLINE stub_put_block(_NET_StreamClass * 0x0106c150, char * 0x00fea2c0, long 3704) line NEWLINE 558 + 36 bytes NEWLINE net_MemCacheWrite(_NET_StreamClass * 0x010c2fc0, char * 0x00fea2c0, long 3704) NEWLINE line 660 + 24 bytes NEWLINE net_pull_http_data(_ActiveEntry * 0x0106b560) line 3156 + 29 bytes NEWLINE net_ProcessHTTP(_ActiveEntry * 0x0106b560) line 3548 + 9 bytes NEWLINE NET_ProcessNet(PRFileDesc * 0x010ca380, int 2) line 3272 + 13 bytes NEWLINE NET_PollSockets() line 180 + 18 bytes NEWLINE nsNetlibService::NetPollSocketsCallback(nsITimer * 0x01035eb0, void * NEWLINE 0x01013e60) line 488 NEWLINE TimerImpl::Fire(unsigned long 435833205) line 319 + 17 bytes NEWLINE TimerImpl::ProcessTimeouts(unsigned long 435833205) line 197 NEWLINE FireTimeout(void * 0x00000000, unsigned int 275, unsigned int 14641, unsigned NEWLINE long 435833205) line 101 + 9 bytes NEWLINE USER32! 77e7128c() NEWLINE main(int 1, char * * 0x00ff51d0) line 95 NEWLINE mainCRTStartup() line 338 + 17 bytes NEWLINE KERNEL32! 77f1b304()
593|Forms with only one text field should submit on enter|In a form with only one text field, the form should be submitted when the user NEWLINE presses the enter key inside the text field. Sample HTML: NEWLINE NEWLINE <form action="foo.cgi" method="post"> NEWLINE <input type=text size=25 value="Submit me"> NEWLINE </form> NEWLINE NEWLINE Currently, I hear a "ding" when hitting enter but nothing happens. I assume this NEWLINE is a forms issue but will need something from the events world too.
594|Raptor Crashes whenever a form is submitted|To reproduce bug: NEWLINE 1. load a page with a simple form, like www.yahoo.com or www.netscape.com (for NEWLINE searches) NEWLINE 2. Enter in a search phrase, such as "search" NEWLINE 3. Press the submit button NEWLINE NEWLINE NGLayout crashes, it appears to be an event-related thing, but what do I know? NEWLINE NEWLINE Here's a stack trace: NEWLINE nsFrame::DisplaySelection(nsIPresContext & {...}) line 487 + 3 bytes NEWLINE nsFrame::HandleEvent(nsFrame * const 0x010678a0, nsIPresContext & {...}, NEWLINE nsGUIEvent * 0x0012f308, nsEventStatus & nsEventStatus_eIgnore) line 566 + 12 NEWLINE bytes NEWLINE nsContainerFrame::HandleEvent(nsContainerFrame * const 0x01067820, NEWLINE nsIPresContext & {...}, nsGUIEvent * 0x0012f308, nsEventStatus & NEWLINE nsEventStatus_eIgnore) line 341 NEWLINE nsView::HandleEvent(nsGUIEvent * 0x0012f308, unsigned int 28) line 666 NEWLINE nsScrollingView::HandleEvent(nsGUIEvent * 0x0012f308, unsigned int 28) line 900 NEWLINE + 16 bytes NEWLINE HandleEvent(nsGUIEvent * 0x0012f308) line 151 + 17 bytes NEWLINE nsWindow::DispatchEvent(nsGUIEvent * 0x0012f308) line 288 + 10 bytes NEWLINE nsWindow::DispatchStandardEvent(unsigned int 103) line 311 + 12 bytes NEWLINE nsWindow::DispatchFocus(unsigned int 103) line 1935 + 12 bytes NEWLINE nsWindow::ProcessMessage(unsigned int 7, unsigned int 2033148, long 0, long * NEWLINE 0x0012f458) line 1511 + 19 bytes NEWLINE nsWindow::WindowProc(void * 0x001405b6, unsigned int 7, unsigned int 2033148, NEWLINE long 0) line 337 + 27 bytes NEWLINE USER32! 77e71ab7() NEWLINE USER32! 77e71a77() NEWLINE NTDLL! 77f7624f() NEWLINE nsWindow::~nsWindow() line 391 NEWLINE nsButton::~nsButton() line 43 + 13 bytes NEWLINE nsButton::`scalar deleting destructor'() + 20 bytes NEWLINE nsObject::ReleaseObject() line 105 + 65 bytes NEWLINE nsObject::InnerSupport::Release(nsObject::InnerSupport * const 0x01140710) line NEWLINE 71 + 17 bytes NEWLINE nsView::~nsView() line 231 + 18 bytes NEWLINE nsView::`scalar deleting destructor'() + 20 bytes NEWLINE nsView::Destroy() line 368 + 38 bytes NEWLINE nsView::~nsView() line 186 + 11 bytes NEWLINE nsView::`scalar deleting destructor'() + 20 bytes NEWLINE nsView::Release(nsView * const 0x01067960) line 286 + 38 bytes NEWLINE nsFrame::~nsFrame() line 196 + 18 bytes NEWLINE nsSplittableFrame::~nsSplittableFrame() line 33 + 13 bytes NEWLINE nsContainerFrame::~nsContainerFrame() line 58 + 13 bytes NEWLINE RootContentFrame::~RootContentFrame() + 25 bytes NEWLINE RootContentFrame::`scalar deleting destructor'() + 20 bytes NEWLINE nsFrame::DeleteFrame(nsFrame * const 0x010678a0) line 254 + 38 bytes NEWLINE nsContainerFrame::~nsContainerFrame() line 56 NEWLINE RootFrame::~RootFrame() + 25 bytes NEWLINE RootFrame::`scalar deleting destructor'() + 20 bytes NEWLINE nsFrame::DeleteFrame(nsFrame * const 0x01067820) line 254 + 38 bytes NEWLINE PresShell::~PresShell() line 315 NEWLINE PresShell::`scalar deleting destructor'() + 20 bytes NEWLINE PresShell::Release(PresShell * const 0x010a63c0) line 279 + 32 bytes NEWLINE DocumentViewerImpl::~DocumentViewerImpl() line 190 + 18 bytes NEWLINE DocumentViewerImpl::`scalar deleting destructor'() + 20 bytes NEWLINE DocumentViewerImpl::Release(DocumentViewerImpl * const 0x010cb480) line 143 + NEWLINE 105 bytes NEWLINE nsWebShell::Embed(nsWebShell * const 0x01031be0, nsIContentViewer * 0x010cbd80, NEWLINE char * 0x00000000, nsISupports * 0x00000000) line 378 + 31 bytes NEWLINE nsDocumentBindInfo::OnStartBinding(nsDocumentBindInfo * const 0x010c0e20, nsIURL NEWLINE * 0x010c0e80, char * 0x010ca640) line 862 + 36 bytes NEWLINE NET_NGLayoutConverter(int 38, void * 0x00000000, URL_Struct_ * 0x010c05d0, NEWLINE MWContext_ * 0x010c21c0) line 660 + 32 bytes NEWLINE NET_StreamBuilder(int 38, URL_Struct_ * 0x010c05d0, MWContext_ * 0x010c21c0) NEWLINE line 239 + 24 bytes NEWLINE NET_CacheConverter(int 38, void * 0x00000000, URL_Struct_ * 0x010c05d0, NEWLINE MWContext_ * 0x010c21c0) line 1558 + 17 bytes NEWLINE NET_StreamBuilder(int 102, URL_Struct_ * 0x010c05d0, MWContext_ * 0x010c21c0) NEWLINE line 239 + 24 bytes NEWLINE net_setup_http_stream(_ActiveEntry * 0x010c4400) line 2878 + 22 bytes NEWLINE net_ProcessHTTP(_ActiveEntry * 0x010c4400) line 3536 + 9 bytes NEWLINE NET_ProcessNet(PRFileDesc * 0x010caec0, int 2) line 3272 + 13 bytes NEWLINE NET_PollSockets() line 180 + 18 bytes NEWLINE nsNetlibService::NetPollSocketsCallback(nsITimer * 0x010c8ca0, void * NEWLINE 0x01013e60) line 488 NEWLINE TimerImpl::Fire(unsigned long 436979873) line 319 + 17 bytes NEWLINE TimerImpl::ProcessTimeouts(unsigned long 436979873) line 197 NEWLINE FireTimeout(void * 0x00000000, unsigned int 275, unsigned int 13827, unsigned NEWLINE long 436979873) line 101 + 9 bytes NEWLINE USER32! 77e7128c() NEWLINE main(int 1, char * * 0x00ff51d0) line 95 NEWLINE mainCRTStartup() line 338 + 17 bytes NEWLINE KERNEL32! 77f1b304()
595|Clicking on links in a frame with target="_top" crash NGLayout|To reproduce bug: NEWLINE 1. Load http://warp/ NEWLINE 2. Click on "client" in left frame NEWLINE 3. Click on "Raptor" in right frame - notice it works - yippee! (no target=_top) NEWLINE 4. Click on "client" again in left frame NEWLINE 5. Click on "Nova" in right frame - crashes - (target=_top) is the only NEWLINE difference really. NEWLINE NEWLINE Here's a stack trace: NEWLINE nsFrame::DisplaySelection(nsIPresContext & {...}) line 487 + 3 bytes NEWLINE nsFrame::HandleEvent(nsFrame * const 0x010c7430, nsIPresContext & {...}, NEWLINE nsGUIEvent * 0x0012f308, nsEventStatus & nsEventStatus_eIgnore) line 566 + 12 NEWLINE bytes NEWLINE nsContainerFrame::HandleEvent(nsContainerFrame * const 0x010c73b0, NEWLINE nsIPresContext & {...}, nsGUIEvent * 0x0012f308, nsEventStatus & NEWLINE nsEventStatus_eIgnore) line 341 NEWLINE nsView::HandleEvent(nsGUIEvent * 0x0012f308, unsigned int 28) line 666 NEWLINE nsScrollingView::HandleEvent(nsGUIEvent * 0x0012f308, unsigned int 28) line 900 NEWLINE + 16 bytes NEWLINE HandleEvent(nsGUIEvent * 0x0012f308) line 151 + 17 bytes NEWLINE nsWindow::DispatchEvent(nsGUIEvent * 0x0012f308) line 288 + 10 bytes NEWLINE nsWindow::DispatchStandardEvent(unsigned int 103) line 311 + 12 bytes NEWLINE nsWindow::DispatchFocus(unsigned int 103) line 1935 + 12 bytes NEWLINE nsWindow::ProcessMessage(unsigned int 7, unsigned int 5637742, long 0, long * NEWLINE 0x0012f458) line 1511 + 19 bytes NEWLINE nsWindow::WindowProc(void * 0x0045065c, unsigned int 7, unsigned int 5637742, NEWLINE long 0) line 337 + 27 bytes NEWLINE USER32! 77e71ab7() NEWLINE USER32! 77e71a77() NEWLINE NTDLL! 77f7624f() NEWLINE nsWindow::~nsWindow() line 391 NEWLINE ChildWindow::~ChildWindow() + 25 bytes NEWLINE ChildWindow::`scalar deleting destructor'() + 20 bytes NEWLINE nsObject::ReleaseObject() line 105 + 65 bytes NEWLINE nsObject::InnerSupport::Release(nsObject::InnerSupport * const 0x01068ef0) line NEWLINE 71 + 17 bytes NEWLINE nsView::~nsView() line 231 + 18 bytes NEWLINE nsView::`scalar deleting destructor'() + 20 bytes NEWLINE nsView::Destroy() line 368 + 38 bytes NEWLINE nsView::~nsView() line 186 + 11 bytes NEWLINE nsView::`scalar deleting destructor'() + 20 bytes NEWLINE nsView::Release(nsView * const 0x010c74f0) line 286 + 38 bytes NEWLINE nsFrame::~nsFrame() line 196 + 18 bytes NEWLINE nsSplittableFrame::~nsSplittableFrame() line 33 + 13 bytes NEWLINE nsContainerFrame::~nsContainerFrame() line 58 + 13 bytes NEWLINE RootContentFrame::~RootContentFrame() + 25 bytes NEWLINE RootContentFrame::`scalar deleting destructor'() + 20 bytes NEWLINE nsFrame::DeleteFrame(nsFrame * const 0x010c7430) line 254 + 38 bytes NEWLINE nsContainerFrame::~nsContainerFrame() line 56 NEWLINE RootFrame::~RootFrame() + 25 bytes NEWLINE RootFrame::`scalar deleting destructor'() + 20 bytes NEWLINE nsFrame::DeleteFrame(nsFrame * const 0x010c73b0) line 254 + 38 bytes NEWLINE PresShell::~PresShell() line 315 NEWLINE PresShell::`scalar deleting destructor'() + 20 bytes NEWLINE PresShell::Release(PresShell * const 0x010bc820) line 279 + 32 bytes NEWLINE DocumentViewerImpl::~DocumentViewerImpl() line 190 + 18 bytes NEWLINE DocumentViewerImpl::`scalar deleting destructor'() + 20 bytes NEWLINE DocumentViewerImpl::Release(DocumentViewerImpl * const 0x010c9cb0) line 143 + NEWLINE 105 bytes NEWLINE nsWebShell::Embed(nsWebShell * const 0x01031ef0, nsIContentViewer * 0x0110dc00, NEWLINE char * 0x00000000, nsISupports * 0x00000000) line 378 + 31 bytes NEWLINE nsDocumentBindInfo::OnStartBinding(nsDocumentBindInfo * const 0x011052f0, nsIURL NEWLINE * 0x01105670, char * 0x01107320) line 862 + 36 bytes NEWLINE NET_NGLayoutConverter(int 38, void * 0x00000000, URL_Struct_ * 0x011069a0, NEWLINE MWContext_ * 0x01106c60) line 660 + 32 bytes NEWLINE NET_StreamBuilder(int 38, URL_Struct_ * 0x011069a0, MWContext_ * 0x01106c60) NEWLINE line 239 + 24 bytes NEWLINE NET_CacheConverter(int 38, void * 0x00000000, URL_Struct_ * 0x011069a0, NEWLINE MWContext_ * 0x01106c60) line 1558 + 17 bytes NEWLINE NET_StreamBuilder(int 102, URL_Struct_ * 0x011069a0, MWContext_ * 0x01106c60) NEWLINE line 239 + 24 bytes NEWLINE net_setup_http_stream(_ActiveEntry * 0x01106e30) line 2878 + 22 bytes NEWLINE net_ProcessHTTP(_ActiveEntry * 0x01106e30) line 3536 + 9 bytes NEWLINE NET_ProcessNet(PRFileDesc * 0x01125320, int 2) line 3272 + 13 bytes NEWLINE NET_PollSockets() line 180 + 18 bytes NEWLINE nsNetlibService::NetPollSocketsCallback(nsITimer * 0x011089c0, void * NEWLINE 0x01013e60) line 488 NEWLINE TimerImpl::Fire(unsigned long 437456989) line 319 + 17 bytes NEWLINE TimerImpl::ProcessTimeouts(unsigned long 437456989) line 197 NEWLINE FireTimeout(void * 0x00000000, unsigned int 275, unsigned int 3046, unsigned NEWLINE long 437456989) line 101 + 9 bytes NEWLINE USER32! 77e7128c() NEWLINE main(int 1, char * * 0x00ff51d0) line 95 NEWLINE mainCRTStartup() line 338 + 17 bytes NEWLINE KERNEL32! 77f1b304()
596|NGLayout crashes on this URL|See URL for test case. Here's a stack trace: NEWLINE NEWLINE NEWLINE nsCSSInlineLayout::ReflowFrame(nsIFrame * 0x0113c8a0, nsReflowMetrics & {...}, NEWLINE const nsReflowState & {...}, int & 2) line 299 NEWLINE nsCSSInlineLayout::ReflowAndPlaceFrame(nsIFrame * 0x0113c8a0) line 175 + 30 NEWLINE bytes NEWLINE nsCSSInlineFrame::ReflowMapped(nsCSSInlineReflowState & {...}, unsigned int & 0) NEWLINE line 573 + 15 bytes NEWLINE nsCSSInlineFrame::InitialReflow(nsCSSInlineReflowState & {...}) line 420 + 16 NEWLINE bytes NEWLINE nsCSSInlineFrame::InlineReflow(nsCSSInlineFrame * const 0x0113c4f0, NEWLINE nsCSSLineLayout & {...}, nsReflowMetrics & {...}, const nsReflowState & {...}) NEWLINE line 262 + 18 bytes NEWLINE nsCSSInlineLayout::ReflowFrame(nsIFrame * 0x0113c4a0, nsReflowMetrics & {...}, NEWLINE const nsReflowState & {...}, int & 1236220) line 292 + 26 bytes NEWLINE nsCSSInlineLayout::ReflowAndPlaceFrame(nsIFrame * 0x0113c4a0) line 175 + 30 NEWLINE bytes NEWLINE nsCSSBlockFrame::ReflowInlineFrame(nsCSSBlockReflowState & {...}, LineData * NEWLINE 0x0113c520, nsIFrame * 0x0113c4a0, unsigned int & 4) line 2421 + 18 bytes NEWLINE nsCSSBlockFrame::ReflowLine(nsCSSBlockReflowState & {...}, LineData * NEWLINE 0x0113c520, unsigned int & 4) line 2004 + 24 bytes NEWLINE nsCSSBlockFrame::ReflowLinesAt(nsCSSBlockReflowState & {...}, LineData * NEWLINE 0x0113c520) line 1879 + 20 bytes NEWLINE nsCSSBlockFrame::ResizeReflow(nsCSSBlockReflowState & {...}) line 1866 + 19 NEWLINE bytes NEWLINE nsCSSBlockFrame::InitialReflow(nsCSSBlockReflowState & {...}) line 1504 + 12 NEWLINE bytes NEWLINE nsCSSBlockFrame::ReflowAround(nsCSSBlockFrame * const 0x0113c390, nsIPresContext NEWLINE & {...}, nsISpaceManager * 0x0113cf20, nsReflowMetrics & {...}, const NEWLINE nsReflowState & {...}, nsRect & {...}, unsigned int & 0) line 1249 + 18 bytes NEWLINE nsBodyFrame::Reflow(nsBodyFrame * const 0x0113ce90, nsIPresContext & {...}, NEWLINE nsReflowMetrics & {...}, const nsReflowState & {...}, unsigned int & 0) line 218 NEWLINE nsContainerFrame::ReflowChild(nsIFrame * 0x0113ce90, nsIPresContext * NEWLINE 0x019e1120, nsReflowMetrics & {...}, const nsReflowState & {...}) line 498 NEWLINE nsTableCellFrame::Reflow(nsTableCellFrame * const 0x0113cdc0, nsIPresContext & NEWLINE {...}, nsReflowMetrics & {...}, const nsReflowState & {...}, unsigned int & 0) NEWLINE line 308 + 30 bytes NEWLINE nsContainerFrame::ReflowChild(nsIFrame * 0x0113cdc0, nsIPresContext * NEWLINE 0x019e1120, nsReflowMetrics & {...}, const nsReflowState & {...}) line 498 NEWLINE nsTableRowFrame::InitialReflow(nsIPresContext & {...}, RowReflowState & {...}, NEWLINE nsReflowMetrics & {...}) line 577 + 30 bytes NEWLINE nsTableRowFrame::Reflow(nsTableRowFrame * const 0x0113b280, nsIPresContext & NEWLINE {...}, nsReflowMetrics & {...}, const nsReflowState & {...}, unsigned int & 0) NEWLINE line 883 + 20 bytes NEWLINE nsContainerFrame::ReflowChild(nsIFrame * 0x0113b280, nsIPresContext * NEWLINE 0x019e1120, nsReflowMetrics & {...}, const nsReflowState & {...}) line 498 NEWLINE nsTableRowGroupFrame::ReflowUnmappedChildren(nsIPresContext * 0x019e1120, NEWLINE RowGroupReflowState & {...}, nsSize * 0x0012e85c) line 793 + 30 bytes NEWLINE nsTableRowGroupFrame::Reflow(nsTableRowGroupFrame * const 0x010c2f10, NEWLINE nsIPresContext & {...}, nsReflowMetrics & {...}, const nsReflowState & {...}, NEWLINE unsigned int & 0) line 1122 + 23 bytes NEWLINE nsContainerFrame::ReflowChild(nsIFrame * 0x010c2f10, nsIPresContext * NEWLINE 0x019e1120, nsReflowMetrics & {...}, const nsReflowState & {...}) line 498 NEWLINE nsTableFrame::ResizeReflowPass1(nsIPresContext * 0x019e1120, nsReflowMetrics & NEWLINE {...}, const nsReflowState & {...}, unsigned int & 0) line 1452 + 33 bytes NEWLINE nsTableFrame::Reflow(nsTableFrame * const 0x010c28a0, nsIPresContext & {...}, NEWLINE nsReflowMetrics & {...}, const nsReflowState & {...}, unsigned int & 0) line NEWLINE 1299 + 30 bytes NEWLINE nsContainerFrame::ReflowChild(nsIFrame * 0x010c28a0, nsIPresContext * NEWLINE 0x019e1120, nsReflowMetrics & {...}, const nsReflowState & {...}) line 498 NEWLINE nsTableOuterFrame::Reflow(nsTableOuterFrame * const 0x010bc620, nsIPresContext & NEWLINE {...}, nsReflowMetrics & {...}, const nsReflowState & {...}, unsigned int & 0) NEWLINE line 448 + 27 bytes NEWLINE nsCSSBlockFrame::ReflowBlockFrame(nsCSSBlockReflowState & {...}, LineData * NEWLINE 0x010bc6d0, nsIFrame * 0x010bc620, unsigned int & 16640) line 2237 + 37 bytes NEWLINE nsCSSBlockFrame::ReflowLine(nsCSSBlockReflowState & {...}, LineData * NEWLINE 0x010bc6d0, unsigned int & 16640) line 1999 + 24 bytes NEWLINE nsCSSBlockFrame::ReflowLinesAt(nsCSSBlockReflowState & {...}, LineData * NEWLINE 0x010bc6d0) line 1879 + 20 bytes NEWLINE nsCSSBlockFrame::FrameAppendedReflow(nsCSSBlockReflowState & {...}) line 1580 + NEWLINE 16 bytes NEWLINE nsCSSBlockFrame::ReflowAround(nsCSSBlockFrame * const 0x013bf2c0, nsIPresContext NEWLINE & {...}, nsISpaceManager * 0x013be780, nsReflowMetrics & {...}, const NEWLINE nsReflowState & {...}, nsRect & {...}, unsigned int & 0) line 1269 + 18 bytes NEWLINE nsBodyFrame::Reflow(nsBodyFrame * const 0x013be600, nsIPresContext & {...}, NEWLINE nsReflowMetrics & {...}, const nsReflowState & {...}, unsigned int & 0) line 218 NEWLINE nsContainerFrame::ReflowChild(nsIFrame * 0x013be600, nsIPresContext * NEWLINE 0x019e1120, nsReflowMetrics & {...}, const nsReflowState & {...}) line 498 NEWLINE RootContentFrame::Reflow(RootContentFrame * const 0x013b9950, nsIPresContext & NEWLINE {...}, nsReflowMetrics & {...}, const nsReflowState & {...}, unsigned int & 0) NEWLINE line 326 + 27 bytes NEWLINE nsContainerFrame::ReflowChild(nsIFrame * 0x013b9950, nsIPresContext * NEWLINE 0x019e1120, nsReflowMetrics & {...}, const nsReflowState & {...}) line 498 NEWLINE RootFrame::Reflow(RootFrame * const 0x013b9770, nsIPresContext & {...}, NEWLINE nsReflowMetrics & {...}, const nsReflowState & {...}, unsigned int & 0) line 132 NEWLINE + 27 bytes NEWLINE nsHTMLReflowCommand::Dispatch(nsHTMLReflowCommand * const 0x010bbdf0, NEWLINE nsIPresContext & {...}, nsReflowMetrics & {...}, const nsSize & {...}) line 133 NEWLINE PresShell::ProcessReflowCommands() line 567 NEWLINE PresShell::ExitReflowLock(PresShell * const 0x019bac10) line 371 NEWLINE PresShell::ContentAppended(PresShell * const 0x019bac14, nsIDocument * NEWLINE 0x019e0ef0, nsIContent * 0x013b9e20) line 654 NEWLINE nsDocument::ContentAppended(nsIContent * 0x013b9e20) line 504 NEWLINE nsHTMLContainer::AppendChild(nsHTMLContainer * const 0x013b9e20, nsIContent * NEWLINE 0x013ea600, int 1) line 192 NEWLINE HTMLContentSink::AppendToCorrectParent(nsHTMLTag eHTMLTag_body, nsIHTMLContent * NEWLINE 0x013b9e20, nsHTMLTag eHTMLTag_table, nsIHTMLContent * 0x013ea600, int 1) line NEWLINE 1429 NEWLINE HTMLContentSink::CloseContainer(HTMLContentSink * const 0x01452f50, const NEWLINE nsIParserNode & {...}) line 891 NEWLINE CNavDTD::CloseContainer(const nsIParserNode & {...}, nsHTMLTag eHTMLTag_table, NEWLINE int 1) line 2332 + 22 bytes NEWLINE CNavDTD::CloseContainersTo(int 2, nsHTMLTag eHTMLTag_table, int 1) line 2366 + NEWLINE 26 bytes NEWLINE CNavDTD::CloseContainersTo(nsHTMLTag eHTMLTag_table, int 1) line 2387 + 20 bytes NEWLINE CNavDTD::HandleEndToken(CToken * 0x01377400) line 749 + 17 bytes NEWLINE NavDispatchTokenHandler(CToken * 0x01377400, nsIDTD * 0x014ac640) line 271 + 12 NEWLINE bytes NEWLINE CTokenHandler::operator()(CToken * 0x01377400, nsIDTD * 0x014ac640) line 80 + 14 NEWLINE bytes NEWLINE CNavDTD::HandleToken(CNavDTD * const 0x014ac640, CToken * 0x01377400) line 489 + NEWLINE 18 bytes NEWLINE nsParser::BuildModel() line 578 + 16 bytes NEWLINE nsParser::ResumeParse() line 526 NEWLINE nsParser::OnDataAvailable(nsParser * const 0x019e1cc4, nsIURL * 0x019debe0, NEWLINE nsIInputStream * 0x019e03d0, int 7476) line 757 + 15 bytes NEWLINE nsDocumentBindInfo::OnDataAvailable(nsDocumentBindInfo * const 0x019deb80, NEWLINE nsIURL * 0x019debe0, nsIInputStream * 0x019e03d0, int 7476) line 904 + 30 bytes NEWLINE stub_put_block(_NET_StreamClass * 0x019e0380, char * 0x00fea2c0, long 7476) line NEWLINE 558 + 36 bytes NEWLINE net_MemCacheWrite(_NET_StreamClass * 0x019ba120, char * 0x00fea2c0, long 7476) NEWLINE line 660 + 24 bytes NEWLINE net_pull_http_data(_ActiveEntry * 0x019df300) line 3156 + 29 bytes NEWLINE net_ProcessHTTP(_ActiveEntry * 0x019df300) line 3548 + 9 bytes NEWLINE NET_ProcessNet(PRFileDesc * 0x010e15a0, int 2) line 3272 + 13 bytes NEWLINE NET_PollSockets() line 180 + 18 bytes NEWLINE nsNetlibService::NetPollSocketsCallback(nsITimer * 0x019e0c20, void * NEWLINE 0x01013e60) line 488 NEWLINE TimerImpl::Fire(unsigned long 438095708) line 319 + 17 bytes NEWLINE TimerImpl::ProcessTimeouts(unsigned long 438095708) line 197 NEWLINE FireTimeout(void * 0x00000000, unsigned int 275, unsigned int 4091, unsigned NEWLINE long 438095708) line 101 + 9 bytes NEWLINE USER32! 77e7128c() NEWLINE main(int 1, char * * 0x00ff51d0) line 95 NEWLINE mainCRTStartup() line 338 + 17 bytes NEWLINE KERNEL32! 77f1b304()
597|PageDown/PageUp key presses should scroll document|Not sure if this is a joki thing or not, but it would be nice if pressing pgdn NEWLINE and pgup scrolled the document.
598|Raptor hangs at www.cbs.com|Type in www.cbs.com, hit return, hangs immediately.
599|ss:FTP directories not displaying|ftp://ftp.netscape.com/pub doesn't show up, neither does a http directory, like NEWLINE http://slip/projects/test/install
600|No UI so can't authenticate with username/password|Can't get in to something like your e*trade account because you can't put in NEWLINE your username/password. Note: putting it in the url bar doesn't work either NEWLINE (does work in 4.06) so no workaround.
601|some GIF images are always dithering to 8bit color cube|This page has an example of GIF images with gray in them; these and other such NEWLINE images look bad in NGLayout. Photographic images, however, look rather good. The NEWLINE issue is that we sometimes assume 8bit when we actually might have a more NEWLINE capable display.
602|Frames "marginwidth" and "marginheight" not implemented yet|We should pass on the values of these attributes to the individual frame NEWLINE documents and have them treat it properly. Currently, the framed documents are NEWLINE assuming the standard 10 pixel over, 10 pixel down margins.
603|script src= doesn't work|See URL for test case and sample code. Most likely a known issue, but putting NEWLINE it into bugsplat anyway.
604|Viewer Crashes in form element|Viewer crashes when a form control gets focus and the document is switched. NEWLINE Can be reproduced by viewing test8 clicking inside a text field and then NEWLINE switching documents. Any fix should also be tested by viewing test9 (frameset NEWLINE test) and then switching to test1.
605|text appears in wrong place in doc|There is an <EMBED> of a quicktime movie, and to the *right* of it, the word NEWLINE "Alice" should appear. In Raptor it appears to the left of the plugin.
606|align=middle broken|the vertical alignment code needs fixing; align=middle is busted.
607|frameborder & border not yet implemented in <frameset>|<frameset border=.. frameborder=..> not yet implemented. www.stocksite.com NEWLINE is a good test case to use.
608|not implemented - frame borders responsive to mouse|The user cannot drag a frame border and have the frames resize.
609|Links don't work due to new selection code|Links have stopped working based on an if statement in some new selection code.
610|Editor mode and JS console need to move out of debug-only code|In the last revisions of webshell/browser window the hook up of editor mode and NEWLINE JS console to their menu items was moved to debug only code. They need to move NEWLINE back.
611|Clicking on links do not work.|This is too old to be useful.
612|Selection cursor appears when clicking on a link|This is too old to be useful (selection has been completely reworked)
613|ss:hitting submit button causes re-entrancy crash|I put a hack into nsPresShell.cpp that sidesteps the bug for the NEWLINE stability release; a larger fix is really needed that involves event NEWLINE handling and widgets...
614|when window is scrolled, resize causes widgets to go nuts|i introduced this and have a fix.
615|unpainted region at bottom of window|if the page is too tall to fit in the window (so it needs to be scrolled), NEWLINE clicking outside of the thumb area until the thumb reaches the bottom of the NEWLINE scrollbar will leave you with the bottom part of the window unpainted. i have a NEWLINE fix for this.
616|Crash in nsTextContent - deleting a stack varible|This is too old to be useful (the ode has been completely reworked)
617|Crash in Dispatch event when switch documents|This happends (on Wndows) because a focus event is coming through when the NEWLINE window has already been deleted. I think the best fix for now is to just disable NEWLINE the notification of focus events, these notifications are not currently being NEWLINE used. The fix is will be in nsWindow.cpp
618|row background painting not backwards compatible|today we paint row backgrounds as a solid block, with cell backgrounds painting NEWLINE over them. NEWLINE Nav4 and IE treat row backgrounds as being inherited by the cells. NEWLINE The visual difference is the space between cells gets the table background NEWLINE rather than the row background. NEWLINE CSS-2 seems to indicate we should do it the way we're doing it today. NEWLINE I've added code to do this the Nav4 way, making sample 3 look correct. NEWLINE We can easily add variant code to check which mode we're in and do the correct NEWLINE thing.
619|table layout broken with nscoord-width columns and colspans|whenever we have colspans, we have a rule that the space required by the NEWLINE spanning cell is proportionately distributed to the column it spans, with the NEWLINE percentage based on the required width of the columns as if colspans didn't NEWLINE exist. This rule has an exception: whenever the widths of all the columns NEWLINE involved are specified, the specified widths are used and the proportions are NEWLINE never computed.
620|images with no width attribute in tables don't work well|whenever we have an image that doesn't have a specified width, and that image NEWLINE is inside a table, the table will lay out incorrectly. This is because we are NEWLINE not computing the minimum required size of the cell containing the image NEWLINE correctly during the incremental reflow that is generated once the image width NEWLINE is known.
621|a document that ends in text doesn't layout properly|When you resize the window then it fixes it up. NEWLINE NEWLINE It's because the flushing of text when closing the document isn't NEWLINE trigger an appended reflow like it should.
622|propagation error for tables|backward propagation was not terminating at "gating" tags, such as <table>. NEWLINE This would cause content to be ejected from the table when a <tr> was omitted NEWLINE and the table was inside another container, such as a nested table or a table NEWLINE inside a <div>.
623|table assertion triggered when maxElementSize > desiredSize|a floating image on this page reports its desired size to be less than its NEWLINE maxElementSize. Table code asserts that maxElementSize <= desiredSize. I've NEWLINE fixed it so that the table cell always reserves at least the maxElementSize for NEWLINE the content.
624|child frames border settings override parent's|<HTML><BODY BGCOLOR="Black" TEXT="White" LINK="#336699" VLINK="#33CC33" NEWLINE BACKGROUND="http://www.westwood.com/assets/background2.jpg" TOPMARGIN="nnn"> NEWLINE <TABLE> NEWLINE <TR><TD>x</TD> NEWLINE <TD VALIGN=TOP><IMG src="http://www.westwood.com//assets2/d2box.jpg" NEWLINE ALIGN="RIGHT" BORDER=0 WIDTH=124 HEIGHT=141></TD></TR></TABLE> NEWLINE </BODY> NEWLINE </HTML> NEWLINE NEWLINE you won't be able to view the page due to a jpeg bug.
625|floating image inteferes with painting (layer involved)|see the bottom of the url for the problem, or look at this minimized test case: NEWLINE NEWLINE <HTML> NEWLINE <BODY BACKGROUND="/images/curve_aqua.gif" BGCOLOR="#FFFFFF" TEXT="#000000" NEWLINE LINK="#DD0000" VLINK="#999999" ALINK="#FF0000" ALT="wave"> NEWLINE NEWLINE <table width=640> NEWLINE <tr> NEWLINE <td width="468" colspan="2"> NEWLINE <TABLE BORDER="0" BGCOLOR="#66cccc" WIDTH="468"> NEWLINE <TR> NEWLINE <TD BGCOLOR="#66cccc" ALIGN="CENTER">A N N O U N C E M E N T S</TD> NEWLINE </TR> NEWLINE <TR> NEWLINE <TD ALIGN="CENTER" VALIGN="TOP"> NEWLINE <TABLE CELLSPACING="0" WIDTH="468" BORDER="0" VALIGN="TOP"> NEWLINE <TR> NEWLINE <TD VALIGN="TOP" WIDTH="68" BGCOLOR="#FFFFFF" ALIGN="LEFT"> NEWLINE <IMG src="http://www.DEVELOPER.com/images/check.gif" NEWLINE Align="left" Alt="Announcements" BORDER="0"> NEWLINE </TD> NEWLINE <TD VALIGN="TOP" BGCOLOR="#FFFFFF" ALIGN="LEFT" WIDTH="400" NEWLINE COLSPAN="4"> NEWLINE <IMG src="http://www.DEVELOPER.com/images/redear.gif" NEWLINE WIDTH="8" HEIGHT="8" BORDER="0">Dozens of new job openings at the <a NEWLINE href="/jobs/">Job Bank</a>.<BR> NEWLINE </TD> NEWLINE </TR> NEWLINE </TABLE> NEWLINE NEWLINE <TABLE CELLSPACING="0" WIDTH="468" BORDER="0" VALIGN="TOP"> NEWLINE <TR> NEWLINE <TD BGCOLOR="#66cccc" ALIGN="CENTER" WIDTH="234"> NEWLINE </TD> NEWLINE <td width="0" bgcolor="#66cccc"> NEWLINE <nowrap><img NEWLINE src="http://www.DEVELOPER.com/images/rule.gif" HEIGHT="10" WIDTH="1"> NEWLINE </td> NEWLINE <TD BGCOLOR="#66cccc" ALIGN="CENTER" WIDTH="234"> NEWLINE </TD> NEWLINE </TR> NEWLINE <tr> NEWLINE <TD VALIGN="TOP" WIDTH="234" BGCOLOR="#FFFFFF"> NEWLINE </TD> NEWLINE <td width="0" bgcolor="#66cccc"><nowrap><img NEWLINE src="http://www.DEVELOPER.com/images/dotclear.gif" HEIGHT="1" WIDTH="1"></td> NEWLINE <TD VALIGN="TOP" WIDTH="234" BGCOLOR="#FFFFFF"> NEWLINE </TD> NEWLINE </TR> NEWLINE </TABLE> NEWLINE </TD> NEWLINE </TR> NEWLINE </TABLE> NEWLINE </td> NEWLINE </tr> NEWLINE </table> NEWLINE </BODY></HTML>
626|Assert in CSS code|Hey Tom, NEWLINE I'm not sure if this goes to you or not. I hit an assert when I go to this site NEWLINE (debug build obviously). Here's the stack: NEWLINE NEWLINE nsDebug::Assertion(char * 0x00566018, char * 0x00566008, char * 0x00565fd0, int NEWLINE 2421) line 95 + 13 bytes NEWLINE nsCSSBlockFrame::ReflowBlockFrame(nsCSSBlockReflowState & {...}, LineData * NEWLINE 0x01128f50, nsIFrame * 0x01128ea0, unsigned int & 0) line 2421 + 39 bytes NEWLINE nsCSSBlockFrame::ReflowLine(nsCSSBlockReflowState & {...}, LineData * NEWLINE 0x01128f50, unsigned int & 0) line 2086 + 24 bytes NEWLINE nsCSSBlockFrame::ReflowLinesAt(nsCSSBlockReflowState & {...}, LineData * NEWLINE 0x01128f50) line 1958 + 20 bytes NEWLINE nsCSSBlockFrame::ResizeReflow(nsCSSBlockReflowState & {...}) line 1943 + 19 NEWLINE bytes NEWLINE nsCSSBlockFrame::ReflowAround(nsCSSBlockFrame * const 0x01128b60, nsIPresContext NEWLINE & {...}, nsISpaceManager * 0x01128990, nsReflowMetrics & {...}, const NEWLINE nsReflowState & {...}, nsRect & {...}, unsigned int & 0) line 1327 + 18 bytes NEWLINE nsBodyFrame::Reflow(nsBodyFrame * const 0x01128900, nsIPresContext & {...}, NEWLINE nsReflowMetrics & {...}, const nsReflowState & {...}, unsigned int & 0) line 218 NEWLINE nsContainerFrame::ReflowChild(nsIFrame * 0x01128900, nsIPresContext * NEWLINE 0x01090d50, nsReflowMetrics & {...}, const nsReflowState & {...}) line 499 NEWLINE nsTableCellFrame::Reflow(nsTableCellFrame * const 0x01127b30, nsIPresContext & NEWLINE {...}, nsReflowMetrics & {...}, const nsReflowState & {...}, unsigned int & 0) NEWLINE line 308 + 30 bytes NEWLINE nsContainerFrame::ReflowChild(nsIFrame * 0x01127b30, nsIPresContext * NEWLINE 0x01090d50, nsReflowMetrics & {...}, const nsReflowState & {...}) line 499 NEWLINE nsTableRowFrame::IncrementalReflow(nsIPresContext & {...}, RowReflowState & NEWLINE {...}, nsReflowMetrics & {...}) line 804 + 33 bytes NEWLINE nsTableRowFrame::Reflow(nsTableRowFrame * const 0x011258d0, nsIPresContext & NEWLINE {...}, nsReflowMetrics & {...}, const nsReflowState & {...}, unsigned int & 0) NEWLINE line 923 + 20 bytes NEWLINE nsContainerFrame::ReflowChild(nsIFrame * 0x011258d0, nsIPresContext * NEWLINE 0x01090d50, nsReflowMetrics & {...}, const nsReflowState & {...}) line 499 NEWLINE nsTableRowGroupFrame::Reflow(nsTableRowGroupFrame * const 0x01124af0, NEWLINE nsIPresContext & {...}, nsReflowMetrics & {...}, const nsReflowState & {...}, NEWLINE unsigned int & 5256331) line 1084 + 27 bytes NEWLINE nsContainerFrame::ReflowChild(nsIFrame * 0x01124af0, nsIPresContext * NEWLINE 0x01090d50, nsReflowMetrics & {...}, const nsReflowState & {...}) line 499 NEWLINE nsTableFrame::Reflow(nsTableFrame * const 0x01124980, nsIPresContext & {...}, NEWLINE nsReflowMetrics & {...}, const nsReflowState & {...}, unsigned int & 0) line NEWLINE 1268 + 30 bytes NEWLINE nsTableOuterFrame::IncrementalReflow(nsIPresContext * 0x01090d50, NEWLINE OuterTableReflowState & {...}, nsReflowMetrics & {...}, const nsReflowState & NEWLINE {...}, unsigned int & 0) line 288 NEWLINE nsTableOuterFrame::Reflow(nsTableOuterFrame * const 0x01122cd0, nsIPresContext & NEWLINE {...}, nsReflowMetrics & {...}, const nsReflowState & {...}, unsigned int & 0) NEWLINE line 400 NEWLINE nsCSSBlockFrame::ReflowBlockFrame(nsCSSBlockReflowState & {...}, LineData * NEWLINE 0x01122d80, nsIFrame * 0x01122cd0, unsigned int & 16640) line 2324 + 37 bytes NEWLINE nsCSSBlockFrame::ReflowLine(nsCSSBlockReflowState & {...}, LineData * NEWLINE 0x01122d80, unsigned int & 16640) line 2086 + 24 bytes NEWLINE nsCSSBlockFrame::ReflowLinesAt(nsCSSBlockReflowState & {...}, LineData * NEWLINE 0x01122d80) line 1958 + 20 bytes NEWLINE nsCSSBlockFrame::ChildIncrementalReflow(nsCSSBlockReflowState & {...}) line 1927 NEWLINE + 19 bytes NEWLINE nsCSSBlockFrame::ReflowAround(nsCSSBlockFrame * const 0x01121f10, nsIPresContext NEWLINE & {...}, nsISpaceManager * 0x011208a0, nsReflowMetrics & {...}, const NEWLINE nsReflowState & {...}, nsRect & {...}, unsigned int & 0) line 1322 + 18 bytes NEWLINE nsBodyFrame::Reflow(nsBodyFrame * const 0x01120810, nsIPresContext & {...}, NEWLINE nsReflowMetrics & {...}, const nsReflowState & {...}, unsigned int & 0) line 218 NEWLINE nsContainerFrame::ReflowChild(nsIFrame * 0x01120810, nsIPresContext * NEWLINE 0x01090d50, nsReflowMetrics & {...}, const nsReflowState & {...}) line 499 NEWLINE nsTableCellFrame::Reflow(nsTableCellFrame * const 0x0111f770, nsIPresContext & NEWLINE {...}, nsReflowMetrics & {...}, const nsReflowState & {...}, unsigned int & 0) NEWLINE line 308 + 30 bytes NEWLINE nsContainerFrame::ReflowChild(nsIFrame * 0x0111f770, nsIPresContext * NEWLINE 0x01090d50, nsReflowMetrics & {...}, const nsReflowState & {...}) line 499 NEWLINE nsTableRowFrame::IncrementalReflow(nsIPresContext & {...}, RowReflowState & NEWLINE {...}, nsReflowMetrics & {...}) line 795 + 33 bytes NEWLINE nsTableRowFrame::Reflow(nsTableRowFrame * const 0x0111ca50, nsIPresContext & NEWLINE {...}, nsReflowMetrics & {...}, const nsReflowState & {...}, unsigned int & 0) NEWLINE line 923 + 20 bytes NEWLINE nsContainerFrame::ReflowChild(nsIFrame * 0x0111ca50, nsIPresContext * NEWLINE 0x01090d50, nsReflowMetrics & {...}, const nsReflowState & {...}) line 499 NEWLINE nsTableRowGroupFrame::Reflow(nsTableRowGroupFrame * const 0x0111ba60, NEWLINE nsIPresContext & {...}, nsReflowMetrics & {...}, const nsReflowState & {...}, NEWLINE unsigned int & 5256331) line 1084 + 27 bytes NEWLINE nsContainerFrame::ReflowChild(nsIFrame * 0x0111ba60, nsIPresContext * NEWLINE 0x01090d50, nsReflowMetrics & {...}, const nsReflowState & {...}) line 499 NEWLINE nsTableFrame::Reflow(nsTableFrame * const 0x0110a560, nsIPresContext & {...}, NEWLINE nsReflowMetrics & {...}, const nsReflowState & {...}, unsigned int & 0) line NEWLINE 1268 + 30 bytes NEWLINE nsTableOuterFrame::IncrementalReflow(nsIPresContext * 0x01090d50, NEWLINE OuterTableReflowState & {...}, nsReflowMetrics & {...}, const nsReflowState & NEWLINE {...}, unsigned int & 0) line 288 NEWLINE nsTableOuterFrame::Reflow(nsTableOuterFrame * const 0x010fecc0, nsIPresContext & NEWLINE {...}, nsReflowMetrics & {...}, const nsReflowState & {...}, unsigned int & 0) NEWLINE line 400 NEWLINE nsCSSBlockFrame::ReflowBlockFrame(nsCSSBlockReflowState & {...}, LineData * NEWLINE 0x010fed70, nsIFrame * 0x010fecc0, unsigned int & 16640) line 2324 + 37 bytes NEWLINE nsCSSBlockFrame::ReflowLine(nsCSSBlockReflowState & {...}, LineData * NEWLINE 0x010fed70, unsigned int & 16640) line 2086 + 24 bytes NEWLINE nsCSSBlockFrame::ReflowLinesAt(nsCSSBlockReflowState & {...}, LineData * NEWLINE 0x010fed70) line 1958 + 20 bytes NEWLINE nsCSSBlockFrame::ChildIncrementalReflow(nsCSSBlockReflowState & {...}) line 1927 NEWLINE + 19 bytes NEWLINE nsCSSBlockFrame::ReflowAround(nsCSSBlockFrame * const 0x0106aec0, nsIPresContext NEWLINE & {...}, nsISpaceManager * 0x0112bee0, nsReflowMetrics & {...}, const NEWLINE nsReflowState & {...}, nsRect & {...}, unsigned int & 0) line 1322 + 18 bytes NEWLINE nsBodyFrame::Reflow(nsBodyFrame * const 0x0112be50, nsIPresContext & {...}, NEWLINE nsReflowMetrics & {...}, const nsReflowState & {...}, unsigned int & 0) line 218 NEWLINE nsContainerFrame::ReflowChild(nsIFrame * 0x0112be50, nsIPresContext * NEWLINE 0x01090d50, nsReflowMetrics & {...}, const nsReflowState & {...}) line 499 NEWLINE nsTableCellFrame::Reflow(nsTableCellFrame * const 0x0112bd80, nsIPresContext & NEWLINE {...}, nsReflowMetrics & {...}, const nsReflowState & {...}, unsigned int & 0) NEWLINE line 308 + 30 bytes NEWLINE nsContainerFrame::ReflowChild(nsIFrame * 0x0112bd80, nsIPresContext * NEWLINE 0x01090d50, nsReflowMetrics & {...}, const nsReflowState & {...}) line 499 NEWLINE nsTableRowFrame::IncrementalReflow(nsIPresContext & {...}, RowReflowState & NEWLINE {...}, nsReflowMetrics & {...}) line 795 + 33 bytes NEWLINE nsTableRowFrame::Reflow(nsTableRowFrame * const 0x0112bc20, nsIPresContext & NEWLINE {...}, nsReflowMetrics & {...}, const nsReflowState & {...}, unsigned int & 0) NEWLINE line 923 + 20 bytes NEWLINE nsContainerFrame::ReflowChild(nsIFrame * 0x0112bc20, nsIPresContext * NEWLINE 0x01090d50, nsReflowMetrics & {...}, const nsReflowState & {...}) line 499 NEWLINE nsTableRowGroupFrame::Reflow(nsTableRowGroupFrame * const 0x0112bad0, NEWLINE nsIPresContext & {...}, nsReflowMetrics & {...}, const nsReflowState & {...}, NEWLINE unsigned int & 5256331) line 1084 + 27 bytes NEWLINE nsContainerFrame::ReflowChild(nsIFrame * 0x0112bad0, nsIPresContext * NEWLINE 0x01090d50, nsReflowMetrics & {...}, const nsReflowState & {...}) line 499 NEWLINE nsTableFrame::Reflow(nsTableFrame * const 0x0112b960, nsIPresContext & {...}, NEWLINE nsReflowMetrics & {...}, const nsReflowState & {...}, unsigned int & 0) line NEWLINE 1268 + 30 bytes NEWLINE nsTableOuterFrame::IncrementalReflow(nsIPresContext * 0x01090d50, NEWLINE OuterTableReflowState & {...}, nsReflowMetrics & {...}, const nsReflowState & NEWLINE {...}, unsigned int & 0) line 288 NEWLINE nsTableOuterFrame::Reflow(nsTableOuterFrame * const 0x0112b740, nsIPresContext & NEWLINE {...}, nsReflowMetrics & {...}, const nsReflowState & {...}, unsigned int & 0) NEWLINE line 400 NEWLINE nsCSSBlockFrame::ReflowBlockFrame(nsCSSBlockReflowState & {...}, LineData * NEWLINE 0x0112b7f0, nsIFrame * 0x0112b740, unsigned int & 16640) line 2324 + 37 bytes NEWLINE nsCSSBlockFrame::ReflowLine(nsCSSBlockReflowState & {...}, LineData * NEWLINE 0x0112b7f0, unsigned int & 16640) line 2086 + 24 bytes NEWLINE nsCSSBlockFrame::ReflowLinesAt(nsCSSBlockReflowState & {...}, LineData * NEWLINE 0x0112b7f0) line 1958 + 20 bytes NEWLINE nsCSSBlockFrame::ChildIncrementalReflow(nsCSSBlockReflowState & {...}) line 1927 NEWLINE + 19 bytes NEWLINE nsCSSBlockFrame::ReflowAround(nsCSSBlockFrame * const 0x01066f20, nsIPresContext NEWLINE & {...}, nsISpaceManager * 0x01066da0, nsReflowMetrics & {...}, const NEWLINE nsReflowState & {...}, nsRect & {...}, unsigned int & 0) line 1322 + 18 bytes NEWLINE nsBodyFrame::Reflow(nsBodyFrame * const 0x01066d10, nsIPresContext & {...}, NEWLINE nsReflowMetrics & {...}, const nsReflowState & {...}, unsigned int & 0) line 218 NEWLINE nsContainerFrame::ReflowChild(nsIFrame * 0x01066d10, nsIPresContext * NEWLINE 0x01090d50, nsReflowMetrics & {...}, const nsReflowState & {...}) line 499 NEWLINE RootContentFrame::Reflow(RootContentFrame * const 0x01066a50, nsIPresContext & NEWLINE {...}, nsReflowMetrics & {...}, const nsReflowState & {...}, unsigned int & 0) NEWLINE line 326 + 27 bytes NEWLINE nsContainerFrame::ReflowChild(nsIFrame * 0x01066a50, nsIPresContext * NEWLINE 0x01090d50, nsReflowMetrics & {...}, const nsReflowState & {...}) line 499 NEWLINE RootFrame::Reflow(RootFrame * const 0x010669d0, nsIPresContext & {...}, NEWLINE nsReflowMetrics & {...}, const nsReflowState & {...}, unsigned int & 0) line 132 NEWLINE + 27 bytes NEWLINE nsHTMLReflowCommand::Dispatch(nsHTMLReflowCommand * const 0x01070f70, NEWLINE nsIPresContext & {...}, nsReflowMetrics & {...}, const nsSize & {...}) line 133 NEWLINE PresShell::ProcessReflowCommands() line 569 NEWLINE PresShell::ExitReflowLock(PresShell * const 0x010659e0) line 373 NEWLINE PresShell::ContentChanged(PresShell * const 0x010659e4, nsIDocument * NEWLINE 0x0104a8b0, nsIContent * 0x0108c530, nsISupports * 0x00000000) line 631 NEWLINE nsDocument::ContentChanged(nsIContent * 0x0108c530, nsISupports * 0x00000000) NEWLINE line 495 NEWLINE nsFrameImageLoader::ReflowFrame() line 344 NEWLINE nsFrameImageLoader::Notify(nsIImageRequest * 0x0112f5d0, nsIImage * 0x00000000, NEWLINE nsImageNotification nsImageNotification_kDimensions, int 24, int 16, void * NEWLINE 0x00000000) line 194 NEWLINE ns_observer_proc(void * 0x0112f740, long 2, void * 0x0012f430, void * NEWLINE 0x0112f5d0) line 240 NEWLINE XP_NotifyObservers(OpaqueObserverList * 0x0112f6d0, long 2, void * 0x0012f430) NEWLINE line 259 + 28 bytes NEWLINE il_dimensions_notify(il_container_struct * 0x0112f7d0, int 24, int 16) line 110 NEWLINE + 18 bytes NEWLINE il_size(il_container_struct * 0x0112f7d0) line 583 + 29 bytes NEWLINE il_gif_write(il_container_struct * 0x0112f7d0, unsigned char * 0x01456ef0, long NEWLINE 128) line 1264 + 9 bytes NEWLINE IL_StreamWrite(il_container_struct * 0x0112f7d0, unsigned char * 0x01456ef0, NEWLINE long 128) line 791 + 18 bytes NEWLINE NetReaderImpl::Write(unsigned char * 0x01456ef0, long 128) line 92 + 20 bytes NEWLINE ImageConsumer::OnDataAvailable(ImageConsumer * const 0x0112fe30, nsIURL * NEWLINE 0x0112fc10, nsIInputStream * 0x0106f8b0, int 230) line 225 NEWLINE stub_put_block(_NET_StreamClass * 0x0106f860, char * 0x00fea2c0, long 230) line NEWLINE 558 + 36 bytes NEWLINE net_MemCacheWrite(_NET_StreamClass * 0x01070720, char * 0x00fea2c0, long 230) NEWLINE line 661 + 24 bytes NEWLINE net_pull_http_data(_ActiveEntry * 0x01130600) line 3146 + 29 bytes NEWLINE net_ProcessHTTP(_ActiveEntry * 0x01130600) line 3532 + 9 bytes NEWLINE NET_ProcessNet(PRFileDesc * 0x011308d0, int 2) line 3283 + 13 bytes NEWLINE NET_PollSockets() line 180 + 18 bytes NEWLINE nsNetlibService::NetPollSocketsCallback(nsITimer * 0x0106fdc0, void * NEWLINE 0x01013e60) line 488 NEWLINE TimerImpl::Fire(unsigned long 5071983) line 319 + 17 bytes NEWLINE TimerImpl::ProcessTimeouts(unsigned long 5071983) line 197 NEWLINE FireTimeout(void * 0x00000000, unsigned int 275, unsigned int 17508, unsigned NEWLINE long 5071983) line 101 + 9 bytes NEWLINE USER32! 77e7128c() NEWLINE main(int 1, char * * 0x00ff5440) line 95 NEWLINE mainCRTStartup() line 338 + 17 bytes NEWLINE KERNEL32! 77f1b304()
627|floating table problems|floating tables seem to be the primary culprit on this page.
628|floating image problems|floating images seem to be the primary culprit on this page.
629|parser bug, illegal table not handled right|here is a minimized test case. In Nav4, the outer table is effectively thrown NEWLINE away. I've inserted comment tags where the <tr> and <td> are missing. If they NEWLINE are added properly, the table behaves as expected. NEWLINE NEWLINE <HTML> NEWLINE <body> NEWLINE NEWLINE <TABLE CELLSPACING=10 CELLPADDING=10 BORDER=10 bgcolor=yellow> NEWLINE <!-- missing_tr><!-- missing_td> NEWLINE <TABLE bgcolor=green> NEWLINE <TR> NEWLINE <TD VALIGN="TOP" width="110"> NEWLINE a NEWLINE </TD> NEWLINE <TD></TD> NEWLINE <TD></TD> NEWLINE <TD></TD> NEWLINE <TD></TD> NEWLINE <TD></TD> NEWLINE <TD VALIGN="TOP" width="500"> NEWLINE x y z NEWLINE </TD> NEWLINE </TR> NEWLINE </TABLE> NEWLINE </TABLE> NEWLINE NEWLINE </BODY> NEWLINE </HTML>
630|parser bug, <center> tags in table not handled correctly|here is a minimized test case. <html> NEWLINE <head><body> NEWLINE NEWLINE <center> NEWLINE <table border=0 cellspacing=0 cellpadding=0 width=210 bgcolor=red> NEWLINE <tr> NEWLINE <td> NEWLINE <center> NEWLINE <table width=201 cellspacing=1 cellpadding=1 border=0 columns=2> NEWLINE <tr><td height=6> NEWLINE </center> NEWLINE </table> NEWLINE </td> NEWLINE </tr> NEWLINE <tr> NEWLINE <td valign=bottom> NEWLINE x NEWLINE </td> NEWLINE </tr> NEWLINE </table> NEWLINE </center> NEWLINE NEWLINE </body></html>
632|Select All does turn on selection display selected|Select All needs to turn selection on in the document. NEWLINE NEWLINE This is a small and easy problem to fix.
633|The initial shift click in window crashes the viewer|The initial shift click (without first just clicking) in the viewer window NEWLINE crashes the viewer. (The fix has been checked into the tip)
635|BODY attributes "marginwidth" & "marginheight" not supported|Adding chris to the cc list as this affects Frames - you need to be able to NEWLINE specify these attributes using frame tags as well...
636|too much vertical space in table cells|Minor display problem. NEWLINE It looks like we're doing the "right" thing based on the content, but it's not NEWLINE what Nav4 does. Nav4 seems to ignore the trailing <br>'s in each cell, while we NEWLINE render them (use visual debugging to see what I mean)
637|tables do not yet support percent height attributes|height=x% on all table tags is currently ignored.
638|floating image align=right messes up table layout|below is a minimized test case. If you remove the align=right the table lays NEWLINE out correctly. NEWLINE NEWLINE <HTML> NEWLINE <BODY TEXT="#000000" LINK="#0000B3" VLINK="#7A7A7A" ALINK="#0c0c0c" NEWLINE BGCOLOR="#FFFFFF"> NEWLINE NEWLINE <TABLE BORDER=1 WIDTH=566 CELLPADDING="0" CELLSPACING="0"> NEWLINE <TR> NEWLINE <TD VALIGN=top width=110> NEWLINE <IMG SRC="http://www.hp.com/ghp/images/drivers.gif" ALIGN=TOP NEWLINE WIDTH="109" HEIGHT="165" USEMAP="#bottleftmap" BORDER="0"><br> NEWLINE </TD> NEWLINE <TD VALIGN=top > NEWLINE <IMG SRC="http://www.hp.com/ghp/images/spacer.gif" NAME="deInfo" NEWLINE ALIGN=left WIDTH="119" HEIGHT="146" BORDER="0" HSPACE="0" VSPACE="0"> NEWLINE <A HREF="/ghp/aquarium/"><IMG NEWLINE SRC="http://www.hp.com/ghp/images/nettle.jpg" ALIGN=RIGHT WIDTH="337" NEWLINE HEIGHT="201" BORDER="0" HSPACE="0"></A> NEWLINE <br clear=all> NEWLINE <BR><BR> NEWLINE <UL TYPE=CIRCLE><LI>Lew Platt, a really swell guy at NEWLINE HP</LI></UL> NEWLINE </TD> NEWLINE </TR> NEWLINE </TABLE> NEWLINE </BODY> NEWLINE </HTML>
639|we need minimization notifications|so that we don't bother doing a reflow to 0,0
640|floating table display error|see the URL or this handy mimimized test case: NEWLINE NEWLINE <HTML><BODY BGCOLOR="#FFFFFF"> NEWLINE NEWLINE <TABLE CELLPADDING="0" CELLSPACING="0" BORDER=0 WIDTH="442"> NEWLINE <TR> NEWLINE <TD WIDTH="442" VALIGN=TOP> NEWLINE <IMG SRC="http://java.sun.com/images/banners/pageheader2.2.gif" NEWLINE HEIGHT="44" WIDTH="425" BORDER="0"> NEWLINE <TABLE ALIGN=LEFT> NEWLINE <TR VALIGN=TOP> NEWLINE <TD WIDTH=430 HEIGHT=124> NEWLINE <center><A HREF="/features/1998/08/banking.html"><IMG NEWLINE SRC="http://java.sun.com/images/rollover/banking.cover.jpg" NEWLINE WIDTH=425 HEIGHT=120 NEWLINE BORDER="0" VSPACE="10" ALIGN=MIDDLE ALT="Banking on Java NEWLINE Technology"></A></center> NEWLINE </TD> NEWLINE </TR> NEWLINE </TABLE> NEWLINE <BR CLEAR=ALL> NEWLINE <P>Reliability, security, and overall performance are essentials in NEWLINE the world of online banking, and now Java technology is not only measuring up NEWLINE -- it is steadily empowering a financial services revolution. (August 4, NEWLINE 1998)</P><P> NEWLINE </TD> NEWLINE </TR> NEWLINE </TABLE> NEWLINE </BODY></HTML>
641|table height attribute doesn't work|Sample HTML, pixel value: NEWLINE <html> NEWLINE <body> NEWLINE <table bgcolor=red height=300 width=100%><tr><td> NEWLINE Foo NEWLINE </td></tr></table> NEWLINE </body> NEWLINE </html> NEWLINE NEWLINE Percent value: NEWLINE <html> NEWLINE <body> NEWLINE <table bgcolor=red height=100% width=100%><tr><td> NEWLINE Foo NEWLINE </td></tr></table> NEWLINE </body> NEWLINE </html> NEWLINE NEWLINE Neither case works; table height is instead set to height of content.
642|text field widget is wrong size, font|This text field should be 2 chars long, but it's renders as 3: NEWLINE <input type="text" size=2 value="1234"> NEWLINE NEWLINE This text field should be 20 chars long, but it renders as about 35: NEWLINE <font face="Times"> NEWLINE <input type="text" size=20 value="12345678901234567890"> NEWLINE </font> NEWLINE NEWLINE Not sure if this is a forms thing or a widget thing...
643|form widgets don't inherit style|Style attributes such as: NEWLINE NEWLINE background-color: NEWLINE background-image: NEWLINE color: NEWLINE NEWLINE etc. NEWLINE Don't work on widgets yet. NEWLINE NEWLINE Not sure if this is a style thing, a widget thing, or a forms thing...
644|Checking on named anchors (#foo) in page forces reload|Let's say you're on a page, such as: NEWLINE http://warp/java/oji/index.html NEWLINE NEWLINE And you click on the "What's New" link, whose URL is: NEWLINE http://warp/java/oji/index.html#new NEWLINE NEWLINE It should just scroll you down to the location of that anchor (#new), but NEWLINE instead, it reloads the page before scrolling down to the named anchor.
645|Crash in dom/js when clicking on links with js in framesets|Jevering, you had a site you get this repeatedly on right? I just have a local NEWLINE test case. I want to get it into the bug report so I can verify on tip and NEWLINE branch.
646|<noscript> tag not respected|we should not display any document content within the <noscript> tag when NEWLINE javascript is enabled. we should probably handle this like we would a comment, NEWLINE so we can recreate the original HTML if necessary (like an editor would need to)
647|mismatched style tags in nested tables closes outer table prematurely|this is a minimized test case derived from http://www.dogpile.com. Interesting NEWLINE behaviors: NEWLINE 1. if the begin/end tags inside the inner table are matched correctly, the table NEWLINE lays out correctly. NEWLINE 2. if the tags are left mismatched, but the outer <center> tag is removed, the NEWLINE table lays out correctly. NEWLINE NEWLINE test case: NEWLINE =============================== NEWLINE <HTML><BODY> NEWLINE <table border> NEWLINE <tr> NEWLINE <td valign=top> NEWLINE <center> NEWLINE <TABLE BORDER=1 WIDTH=100> NEWLINE <TR> NEWLINE <TD> NEWLINE <font size = "-2"> NEWLINE <center> NEWLINE inside NEWLINE </font> NEWLINE </center> NEWLINE </td> NEWLINE </tr> NEWLINE </table> NEWLINE outside table NEWLINE </center> NEWLINE </td> NEWLINE </tr> NEWLINE </table> NEWLINE </BODY></HTML>
648|vertical space incorrect inside a <font> tag|this is a minimized test case derived from http://www.slashdot.org. If you NEWLINE remove the <FONT color=000000>, everything spaces correctly. The <font> seems NEWLINE to be surpressing proper vertical spacing from the <p>. NEWLINE NEWLINE test case: NEWLINE =============================== NEWLINE <HTML><BODY> NEWLINE <FONT color=000000> NEWLINE <p>too NEWLINE <P>short! NEWLINE </FONT> NEWLINE <p>just NEWLINE <P>right! NEWLINE </BODY></HTML>
649|vertical centering|To center a table cell horizontally and vertically, I often use an approach like NEWLINE what you see at http://checkmate/aol/ -- we don't NEWLINE seem to vertically center the stuff properly yet, though.
650|resizing window horiz. changes table layout incorrectly|load up www.yahoo.com; wiggle around the right edge of the window. notice how NEWLINE the right most table cell bounces around...if NEWLINE you turn on visual debugging you can see the bug more clearly...
651|garbage form buttons|This url has two form buttons at the bottom that have the value: NEWLINE NEWLINE VALUE=' > ' & VALUE=' >> ' NEWLINE NEWLINE Currently these buttons are garbled with "gt" and "gt;gt;" respectively. The NEWLINE bottom of the 'g's looks wierd too.
652|Server Side Image Maps Do Not Work|Visit this URL for a sample server-side image map (it's a map of the south bay NEWLINE area) that does not work in NGLayout, but works in Navigator.
653|On documents without a title, the viewer title is wrong|If you're browsing a document with the title "Foo Bar" and you subsequently load NEWLINE a document that has no title, the viewer window title remains "Foo Bar." It NEWLINE should be set to "No Title" or the URL of the new page, etc.
654|Text does not display correctly|The ":" gets truncated, I think the text isn't being measured correctly. NEWLINE Here is the HTML: NEWLINE NEWLINE <BODY TEXT="#000000" BGCOLOR="#FFFFFF" LINK="#0000CC" VLINK="#800080" NEWLINE ALINK="#800000"> NEWLINE NEWLINE <P><BR><FONT FACE="Arial"><FONT COLOR="#000000"><FONT SIZE=-1>WASHINGTON, NEWLINE Aug 6 (Reuters) - A federal judge ordered Microsoft Corp. (Nasdaq:<A NEWLINE HREF="../../quotes/?symbol=MSFT">MSFT</A>) NEWLINE on Thursday to make available top executives, including Chairman Bill Gates, NEWLINE and key software code to government lawyers building an antitrust suit NEWLINE against the software giant.</FONT></FONT></FONT> NEWLINE </BODY> NEWLINE </HTML>
655|<DT> in <TD>|original note from Angus: NEWLINE NEWLINE Ok, I think I figured out the problem. It appears to have something to do with NEWLINE usage of the "DT" tag. NEWLINE NEWLINE I simpled down the beginning of this document. The first test case appears at NEWLINE the URL below, and has the "DT" tag on the text NEWLINE "Communicator 4.5 (Nova)," which appears in line 54 of the HTML source: NEWLINE http://checkmate/nglayout/bugs/client.html NEWLINE NEWLINE This URL is the exact same page, except the DT's surrounding the text NEWLINE "Communicator 4.5 (Nova)" have been removed on line NEWLINE 54: NEWLINE http://checkmate/nglayout/bugs/client-nodt.html NEWLINE NEWLINE Sounds like it might be a parser issue ... forward to Rick if you agree. NEWLINE -angus
656|Selection draw incorrectly at beginning of line|Simple, safe fix.
657|percent-width cells in auto-width tables don't work|not yet implemented: cell widths given as a percentage inside an auto-width NEWLINE table. NEWLINE NEWLINE the implementation will take place in several steps: NEWLINE straightforward tables (no colspans, no conflicting attributes) NEWLINE tables including colspans NEWLINE tables including conflicting attributes NEWLINE tables including illegal attributes (the sum of percent-widths > 100%, for NEWLINE example)
658|Copying HTML to AIM does not work under windows 95/NT|Copy HTML to AIM seems to work just fine under NT. However, under win 95 and win NEWLINE 98 after a copy is done, AIM doesn't seem to recognize that anything is on the NEWLINE pasteboard.
660|Javascript loaded images get drawn twice|There's javascript on this page load a random image, for some reason it gets NEWLINE called twice.
661|navigator.mimetypes array does not work (blocked on with(document))|I can't seem to get plug-ins to work on release builds on win95. Perhaps this is NEWLINE a known issue ... making note of it here anyway. NEWLINE NEWLINE What I *do* see in place of a plug-in is a rectangle that never repaints where NEWLINE the plug-in should be.
662|too much whitespace between some elements on aol.com|relatively speaking, the whitespace between the elements seem NEWLINE to be acceptable. however, i've found another problem on NEWLINE aol.com - not all the links show up. NEWLINE NEWLINE using the 11/30 build on NT/win98 NEWLINE 1) Go to the My News section NEWLINE 2) The link generated from the code below NEWLINE does NOT show up. only the text preceeding NEWLINE the link shows up. NEWLINE NEWLINE <font face="times, times new roman" size="2"> NEWLINE Get free custom news and weather in <a NEWLINE href="http://dynamic.aol.com/cgi/redir-complex?url=http%3a%2f%2fwww.aol.com%2fmy NEWLINE news%2fhome.adp&sid=stat3a">My News</a>.
663|Resizing image in a floating table doesn't reflow text properly|In this example, the first time you load the page, since I didn't specify the NEWLINE height/width of the image, NGlayout loads the document as if the image (which is NEWLINE contained in a table) has zero width/height. NEWLINE NEWLINE The HTML boils down to this test case: NEWLINE <table align=left><tr><td> NEWLINE <img src="foo.jpg"> NEWLINE </td></table> NEWLINE I am the text that appears to the right of the floating table. I will not reflow NEWLINE when the image foo.jpg comes in and resizes it's parent table. Instead, the NEWLINE foo.jpg image will appear over me.
664|Paragraphs with CSS margin attributes are corrupted|Perhaps corrupted isn't the proper adjective, but I couldn't come up with a NEWLINE better one... NEWLINE NEWLINE In this example, the only trickiness is a paragraph with left and right CSS NEWLINE margins defined in pixel widths. This seems to cause some wackiness with memory NEWLINE issues - I notice missing text at first, then as I resize the window, garbage NEWLINE characters appear and my system starts running low on memory. NEWLINE NEWLINE I'm seeing this on a release build on WinNT.
665|Drop-down lists are too large (should have scrollbar)|Two small fixes are needed, and they should effect anything else.
668|image submit controls not responsive|In some pages with an image submit control inside a table, the control will not NEWLINE get any mouse events, so the form will not be submitted. www.nbc.com/tvcentral NEWLINE has a GO image control that fails. www.sportszone.com has a search image control NEWLINE that fails. NEWLINE NEWLINE On other pages with an image submit, the submit needs to be clicked twice. test8 NEWLINE is an example of this. The problem here is that mouse enter is not being seen by NEWLINE the control.
669|Bad layout with image with ALIGN=LEFT inside UL tags|See the URL <http://opera/users/hecker/publish/dod-license.htm>. The page looks NEWLINE OK on Communicator 4.5 PR1 but with Raptor (8/12 mid-day build) the image laps NEWLINE over the text. The problem appears to be with the content being inside NEWLINE <UL>...</UL> tags (albeit not part of a <LI>...</LI> list item); if the <UL> and NEWLINE </UL> tags are removed then the page lays out properly in Raptor (identical to NEWLINE 4.5). NEWLINE NEWLINE The original page (from which the example above was extracted) is at NEWLINE <http://www.menk.com/dod_license/>. Note that the page was created and modified NEWLINE exclusively using Composer, and presumably Composer inserted the LI-less UL tags NEWLINE at some point for reasons of its own.
670|assert because we try to BalanceColumnWidths on a continuing frame|continuing frames always get their cell map and column width information from NEWLINE the first-in-flow. this must be enforced correctly throughout the table code.
671|layout incorrect in tables with 1 row and cells with colspans|colspans in tables with only a single row are meaningless. However, they are NEWLINE not consistently treated that way in the table code. The table frame code must NEWLINE be careful to check if there is only a single row and do the right thing with NEWLINE colspans when building the cell map, fluffing out column frames, etc.
672|yahoo margin problem|When browsing www.yahoo.com in the viewer the right margin is not set the same NEWLINE as in Navigator. The viewer has no margin, while Navigator has a margin 3/4 to NEWLINE 1 inch wide.
673|OnDataAvailable() and OnStartBinding() never get called.|I went to this page to test long documents. It is a simple page with a TON of NEWLINE plain text content. In Raptor, you get notified in the status bar that data is NEWLINE coming across, but it doesnt ever show up in OnDataAvailable() or NEWLINE OnStartBinding() in the parser.
674|NGLayout crashes on link from this URL|Go to jamie's page - about:jwz or www.jwz.org. NEWLINE Click on "Links" NEWLINE Notice NGLayout crashes. NEWLINE I think it has something to do with the code you turned on for window.top - NEWLINE here's a piece of the stack trace: NEWLINE NEWLINE nsWebShell::QueryInterface(nsWebShell * const 0x010412a0, const nsID & {...}, NEWLINE void * * 0x10001180) line 375 + 19 bytes NEWLINE GlobalWindowImpl::GetTop(GlobalWindowImpl * const 0x01073318, nsIDOMWindow * * NEWLINE 0x0012e668) line 324 + 20 bytes NEWLINE GetWindowProperty(JSContext * 0x0108acf0, JSObject * 0x014544f0, long -33, long NEWLINE * 0x0012ea44) line 242 + 16 bytes NEWLINE js_Interpret(JSContext * 0x0108acf0, long * 0x0012ec74) line 2090 + 550 bytes NEWLINE js_Execute(JSContext * 0x0108acf0, JSObject * 0x014544f0, JSScript * 0x010df450, NEWLINE JSFunction * 0x00000000, JSStackFrame * 0x00000000, int 0, long * 0x0012ec74) NEWLINE line 805 + 13 bytes NEWLINE JS_EvaluateUCScriptForPrincipals(JSContext * 0x0108acf0, JSObject * 0x014544f0, NEWLINE JSPrincipals * 0x00000000, unsigned short * 0x0145dec8, unsigned int 481, char * NEWLINE 0x0106e480, unsigned int 4, long * 0x0012ec74) line 2226 + 27 bytes NEWLINE JS_EvaluateUCScript(JSContext * 0x0108acf0, JSObject * 0x014544f0, unsigned NEWLINE short * 0x0145dec8, unsigned int 481, char * 0x0106e480, unsigned int 4, long * NEWLINE 0x0012ec74) line 2208 + 35 bytes NEWLINE JS_EvaluateScript(JSContext * 0x0108acf0, JSObject * 0x014544f0, char * NEWLINE 0x0147be98, unsigned int 481, char * 0x0106e480, unsigned int 4, long * NEWLINE 0x0012ec74) line 2175 + 33 bytes NEWLINE nsJSContext::EvaluateString(char * 0x0147be98, unsigned int 481, char * NEWLINE 0x0106e480, unsigned int 4, long * 0x0012ec74) line 76 + 50 bytes NEWLINE HTMLContentSink::ProcessSCRIPTTag(const nsIParserNode & {...}) line 1860 + 46 NEWLINE bytes NEWLINE HTMLContentSink::AddLeaf(HTMLContentSink * const 0x01489c28, const nsIParserNode NEWLINE & {...}) line 1240 NEWLINE CNavDTD::AddLeaf(const nsIParserNode & {...}) line 2748 + 22 bytes NEWLINE (snip)
675|%-width cells not layout out correctly|very close, but not quite nav4 compatible
676|Purify Reports Leak of Form Object|Purify reports a leak of the form object for this HTML: NEWLINE NEWLINE <HTML> NEWLINE <body> NEWLINE <FORM> NEWLINE <input type=submit name=file> NEWLINE </form> NEWLINE </body> NEWLINE </html> NEWLINE NEWLINE Purify output: NEWLINE NEWLINE [W] MLK: Memory leak of 144 bytes from 1 block allocated in nsForm::new(UINT) NEWLINE Distribution of leaked blocks NEWLINE Allocation location NEWLINE new(UINT) [new.cpp:23] NEWLINE nsForm::new(UINT) [nsForm.cpp:123] NEWLINE NS_NewHTMLForm(nsIFormManager * *,nsIAtom *) [nsForm.cpp:1224] NEWLINE HTMLContentSink::OpenForm(nsIParserNode const&) NEWLINE [nsHTMLContentSink.cpp:557] NEWLINE CNavDTD::OpenForm(nsIParserNode const&) [CNavDTD.cpp:2461] NEWLINE CNavDTD::HandleStartToken(CToken *) [CNavDTD.cpp:797] NEWLINE NavDispatchTokenHandler(CToken *,nsIDTD *) [CNavDTD.cpp:417] NEWLINE CTokenHandler::()(CToken *,nsIDTD *) [nsTokenHandler.cpp:80] NEWLINE CNavDTD::HandleToken(CToken *) [CNavDTD.cpp:679] NEWLINE nsParser::BuildModel(void) [nsParser.cpp:586] NEWLINE nsParser::ResumeParse(void) [nsParser.cpp:550] NEWLINE nsParser::OnDataAvailable(nsIURL *,nsIInputStream *,int) NEWLINE [nsParser.cpp:761] NEWLINE nsDocumentBindInfo::OnDataAvailable(nsIURL *,nsIInputStream *,int) NEWLINE [nsDocLoader.cpp:1056] NEWLINE stub_put_block(_NET_StreamClass *,char const*,long) NEWLINE [nsStubContext.cpp:566] NEWLINE net_read_file_chunk [mkfile.c:887]
677|Headings are positioned off the right edge of the page|If you load this page, the headings are all off the right edge of the page NEWLINE (their left edges are clipped). I'm using the release build.
678|text/html and text/plain OBJECTs don't work|This page contains several Object tag tests. I couldn't get any of them to NEWLINE work in Raptor. I'm using the release build.
680|crash trying to view tinderbox|code: NEWLINE HTMLContentSink::GetCurrentContainer(eHTMLTags* aType) NEWLINE { NEWLINE nsIHTMLContent* parent; NEWLINE if (mStackPos <= 2) { // assume HTML and BODY/FRAMESET are on the stack NEWLINE if (mBody) { NEWLINE parent = mBody; NEWLINE *aType = eHTMLTag_body; NEWLINE } else { NEWLINE parent = mFrameset; NEWLINE *aType = eHTMLTag_frameset; NEWLINE } NEWLINE } else { NEWLINE parent = mContainerStack[mStackPos - 1]; // <<---------- NEWLINE *aType = mNodeStack[mStackPos - 1]; NEWLINE } NEWLINE return parent; NEWLINE } NEWLINE NEWLINE NEWLINE mStackPos 17326001 NEWLINE NEWLINE NEWLINE NEWLINE crash stack: NEWLINE HTMLContentSink::GetCurrentContainer(nsHTMLTag * 0x0012f270) line 1212 + 12 NEWLINE bytes NEWLINE HTMLContentSink::AddLeaf(HTMLContentSink * const 0x01494cb8, const nsIParserNode NEWLINE & {...}) line 1263 + 12 bytes NEWLINE CNavDTD::AddLeaf(const nsIParserNode & {...}) line 2748 + 22 bytes NEWLINE CNavDTD::HandleDefaultStartToken(CToken * 0x0113f890, nsHTMLTag NEWLINE eHTMLTag_newline, nsIParserNode & {...}) line 735 + 12 bytes NEWLINE CNavDTD::HandleStartToken(CToken * 0x0113f890) line 853 + 23 bytes NEWLINE NavDispatchTokenHandler(CToken * 0x0113f890, nsIDTD * 0x01476890) line 427 + 12 NEWLINE bytes NEWLINE CTokenHandler::operator()(CToken * 0x0113f890, nsIDTD * 0x01476890) line 80 + 14 NEWLINE bytes NEWLINE CNavDTD::HandleToken(CNavDTD * const 0x01476890, CToken * 0x0113f890) line 679 + NEWLINE 18 bytes NEWLINE nsParser::BuildModel() line 586 + 16 bytes NEWLINE nsParser::ResumeParse() line 551 NEWLINE nsParser::OnDataAvailable(nsParser * const 0x01143484, nsIURL * 0x011318d0, NEWLINE nsIInputStream * 0x01140580, int 8066) line 761 + 15 bytes NEWLINE nsDocumentBindInfo::OnDataAvailable(nsDocumentBindInfo * const 0x01131870, NEWLINE nsIURL * 0x011318d0, nsIInputStream * 0x01140580, int 8066) line 977 + 30 bytes NEWLINE stub_put_block(_NET_StreamClass * 0x01140530, char * 0x00fe9278, long 8066) line NEWLINE 560 + 36 bytes NEWLINE net_MemCacheWrite(_NET_StreamClass * 0x011c7ce0, char * 0x00fe9278, long 8066) NEWLINE line 661 + 24 bytes NEWLINE net_pull_http_data(_ActiveEntry * 0x011313f0) line 3003 + 29 bytes NEWLINE net_ProcessHTTP(_ActiveEntry * 0x011313f0) line 3389 + 9 bytes NEWLINE NET_ProcessNet(PRFileDesc * 0x010b2fe0, int 2) line 3297 + 13 bytes NEWLINE NET_PollSockets() line 180 + 18 bytes NEWLINE nsNetlibService::NetPollSocketsCallback(nsITimer * 0x0112a300, void * NEWLINE 0x01013e60) line 502 NEWLINE TimerImpl::Fire(unsigned long 1626839301) line 321 + 17 bytes NEWLINE TimerImpl::ProcessTimeouts(unsigned long 1626839301) line 199 NEWLINE FireTimeout(void * 0x00000000, unsigned int 275, unsigned int 15129, unsigned NEWLINE long 1626839301) line 103 + 9 bytes NEWLINE USER32! 77e7128c() NEWLINE main(int 2, char * * 0x00ff4f20) line 95 NEWLINE mainCRTStartup() line 338 + 17 bytes NEWLINE KERNEL32! 77f1b304()
681|ugly link parses incorrectly, source gets put into content|minimized test case: NEWLINE <html><body> NEWLINE 123<A NEWLINE HREF="http://404.jodi.org/!!@#!@!#@!$!!$#!!$^!@^!#(~><><"target="_top">456</a> NEWLINE </body> NEWLINE </html>
682|form elements with nbsp don't give back correct measurements|the two cells below give the same max element size NEWLINE <td nowrap><input 1> <input 2> NEWLINE <td><input 1> <input 2> NEWLINE (where the two input elements are real legal form elements, of course) NEWLINE the first one should have a minimum width of the entire unbroken line. NEWLINE NEWLINE NEWLINE ============================================================================= NEWLINE here is a minimized test case. both cells claim to have the same minimum size: NEWLINE NEWLINE <html><body> NEWLINE NEWLINE <form method=post action="/esearch/isr/1,249,,00.html?st.cn.fd.nws.cn"> NEWLINE <table border="0" cellPadding="4" cellSpacing="0" width="20%" bgcolor="#FFCC00"> NEWLINE <tr> NEWLINE <td noWrap vAlign="top"> NEWLINE <input name="keyword" size="15"> <input name="search" NEWLINE type="submit" value="Go!"> NEWLINE </td> NEWLINE </tr> NEWLINE <tr> NEWLINE <td vAlign="top"> NEWLINE <input name="keyword" size="15"> <input name="search" NEWLINE type="submit" value="Go!"> NEWLINE </td> NEWLINE </tr> NEWLINE </table> NEWLINE </form> NEWLINE NEWLINE </body></html>
683|parser complains about null tag on www.stocksite.com|CNavDTD line 1680 is complaining about a null tag. This page loaded a few weeks NEWLINE ago. This is on the tip.
684|Image.setDOMAttribute("WIDTH",400); is broken|... this would be incredibly cool for demos ... NEWLINE NEWLINE Also, if I change the SRC of an image, I'd like a way to programmatically tell NEWLINE it be the correct size for the new SRC (without breaking old content that NEWLINE assumes we won't scale, of course.) I suppose the best way to do this is to NEWLINE createElement with the new image and then replaceChild on the old one.
685|document.documentElement.removeChild() doesn't work for IMGs|Here's the test case I'm using: NEWLINE NEWLINE <html> NEWLINE <head> NEWLINE <script> NEWLINE function init() { NEWLINE var foo = document.getElementsByTagName("IMG")[0]; NEWLINE alert(foo); // tells me I've got an HTMLImageElement NEWLINE document.documentElement.removeChild(foo); NEWLINE } NEWLINE </script> NEWLINE </head> NEWLINE <body onload="init();"> NEWLINE <img src="http://my.netscape.com/mys/images/nnc_lockup.gif"> NEWLINE </body> NEWLINE </html>
686|reflow of tables doesn't reposition caption|If you have a caption in a table that does not specify size, and the the table NEWLINE resizes itself as it loads the content in contains, the caption is not properly NEWLINE recentered above the table unless you force a reflow. NEWLINE NEWLINE The URL given is a good test case - load the url - notice the caption appears on NEWLINE the left side above the table. Now force a reflow by resizing. Notice the NEWLINE caption goes to the middle where it should be...
687|infinite loop|type http://w3 into the URL widget, hit return, and watch it loop.
688|crash destroying content when there is a font tag inside a select form element|load the URL, or the tiny test case provided below for your pleasure. Go to any NEWLINE other document. Crash in nsHTMLContainer destrucutor. Stack provided below. NEWLINE NEWLINE =============== test case =============== NEWLINE <HTML><BODY> NEWLINE <FORM ACTION="post"> NEWLINE <SELECT NAME=""> NEWLINE <font size=-2> NEWLINE <OPTION VALUE="Loves">Loves NEWLINE </font> NEWLINE </SELECT> NEWLINE </FORM> NEWLINE </BODY></HTML> NEWLINE NEWLINE ===================== stack ========================= NEWLINE nsHTMLContainer::~nsHTMLContainer() line 85 + 9 bytes NEWLINE nsHTMLContainer::`scalar deleting destructor'() + 20 bytes NEWLINE nsHTMLContent::Release(nsHTMLContent * const 0x0104f920) line 164 + 149 bytes NEWLINE nsHTMLTagContent::Release(nsHTMLTagContent * const 0x0104f920) line 185 + 9 NEWLINE bytes NEWLINE nsHTMLContainer::~nsHTMLContainer() line 85 + 12 bytes NEWLINE nsInput::~nsInput() line 75 + 24 bytes NEWLINE nsSelect::~nsSelect() line 413 + 13 bytes NEWLINE nsSelect::`scalar deleting destructor'() + 20 bytes NEWLINE nsInput::Release(nsInput * const 0x0104ee50) line 164 + 38 bytes NEWLINE nsHTMLContainer::~nsHTMLContainer() line 85 + 12 bytes NEWLINE BodyPart::~BodyPart() line 175 + 13 bytes NEWLINE BodyPart::`scalar deleting destructor'() + 20 bytes NEWLINE BodyPart::Release(BodyPart * const 0x0105f7f0) line 185 + 38 bytes NEWLINE nsHTMLContainer::~nsHTMLContainer() line 85 + 12 bytes NEWLINE RootPart::~RootPart() line 638 + 13 bytes NEWLINE RootPart::`scalar deleting destructor'() + 20 bytes NEWLINE nsHTMLContent::Release(nsHTMLContent * const 0x0105a490) line 164 + 149 bytes NEWLINE nsHTMLTagContent::Release(nsHTMLTagContent * const 0x0105a490) line 185 + 9 NEWLINE bytes NEWLINE nsDocument::~nsDocument() line 188 + 31 bytes NEWLINE nsMarkupDocument::~nsMarkupDocument() line 46 + 13 bytes NEWLINE nsHTMLDocument::~nsHTMLDocument() line 109 + 43 bytes NEWLINE nsHTMLDocument::`scalar deleting destructor'() + 20 bytes NEWLINE nsDocument::Release(nsDocument * const 0x010442c0) line 247 + 108 bytes NEWLINE nsHTMLDocument::Release(nsHTMLDocument * const 0x010442c0) line 143 + 9 bytes NEWLINE nsEventStateManager::~nsEventStateManager() line 36 + 31 bytes NEWLINE nsEventStateManager::`scalar deleting destructor'() + 20 bytes NEWLINE nsEventStateManager::Release(nsEventStateManager * const 0x01058ea0) line 41 + NEWLINE 102 bytes NEWLINE nsPresContext::~nsPresContext() line 92 + 40 bytes NEWLINE GalleyContext::~GalleyContext() line 40 + 13 bytes NEWLINE GalleyContext::`scalar deleting destructor'() + 20 bytes NEWLINE nsPresContext::Release(nsPresContext * const 0x0104d3d0) line 108 + 38 bytes NEWLINE PresShell::~PresShell() line 318 + 31 bytes NEWLINE PresShell::`scalar deleting destructor'() + 20 bytes NEWLINE PresShell::Release(PresShell * const 0x01054ac0) line 279 + 32 bytes NEWLINE DocumentViewerImpl::~DocumentViewerImpl() line 201 + 18 bytes NEWLINE DocumentViewerImpl::`scalar deleting destructor'() + 20 bytes NEWLINE DocumentViewerImpl::Release(DocumentViewerImpl * const 0x0104c850) line 143 + NEWLINE 105 bytes NEWLINE nsWebShell::Embed(nsWebShell * const 0x01036680, nsIContentViewer * 0x0104f3d0, NEWLINE char * 0x00000000, nsISupports * 0x00000000) line 484 + 31 bytes NEWLINE nsDocumentBindInfo::OnStartBinding(nsDocumentBindInfo * const 0x01049240, nsIURL NEWLINE * 0x01049120, char * 0x0104c200) line 957 + 36 bytes NEWLINE NET_NGLayoutConverter(int 38, void * 0x00000000, URL_Struct_ * 0x01059bf0, NEWLINE MWContext_ * 0x01058240) line 666 + 32 bytes NEWLINE NET_StreamBuilder(int 38, URL_Struct_ * 0x01059bf0, MWContext_ * 0x01058240) NEWLINE line 243 + 24 bytes NEWLINE net_setup_file_stream(_ActiveEntry * 0x01058a70) line 762 + 25 bytes NEWLINE net_ProcessFile(_ActiveEntry * 0x01058a70) line 1234 + 9 bytes NEWLINE NET_ProcessNet(PRFileDesc * 0x00000000, int 1) line 3302 + 13 bytes NEWLINE NET_PollSockets() line 162 + 9 bytes NEWLINE nsNetlibService::NetPollSocketsCallback(nsITimer * 0x01048d50, void * NEWLINE 0x010148e0) line 530 NEWLINE TimerImpl::Fire(unsigned long 2056265753) line 321 + 17 bytes NEWLINE TimerImpl::ProcessTimeouts(unsigned long 2056265753) line 199 NEWLINE FireTimeout(void * 0x00000000, unsigned int 275, unsigned int 13907, unsigned NEWLINE long 2056265753) line 103 + 9 bytes NEWLINE USER32! 77e7128c() NEWLINE main(int 2, char * * 0x00ff61d0) line 95 NEWLINE mainCRTStartup() line 338 + 17 bytes NEWLINE KERNEL32! 77f1b304()
689|table alignment not quite right in constrained layout case|min and max are fine, but when the table is in between the cells don't quite NEWLINE line up right.
690|Link text gets "smushed" into other text (bad spacing)|This test case shows that if you have an LI with a sufficiently long text node NEWLINE before a link, the linked text chops the whitespace before it. However, if there NEWLINE is not much text before the link, everything looks OK. It's very difficult to NEWLINE explain in words, but the test case is pretty good at showing the problem. NEWLINE NEWLINE This is a debug build I'm running.
691|removeChild() causes crash|See URL for test case. Partial stack trace: NEWLINE NEWLINE nsCSSBlockFrame::FrameDeletedReflow(nsCSSBlockReflowState & {...}) line 1889 + 3 NEWLINE bytes NEWLINE nsCSSBlockFrame::ReflowAround(nsCSSBlockFrame * const 0x010ba6a0, nsIPresContext NEWLINE & {...}, nsISpaceManager * 0x010b7d00, nsReflowMetrics & {...}, const NEWLINE nsReflowState & {...}, nsRect & {...}, unsigned int & 1073741824) line 1309 + 18 NEWLINE bytes NEWLINE nsCSSBlockFrame::ReflowBlockFrame(nsCSSBlockReflowState & {...}, LineData * NEWLINE 0x010ba6e0, nsIFrame * 0x010ba650, unsigned int & 16640) line 2313 + 47 bytes NEWLINE nsCSSBlockFrame::ReflowLine(nsCSSBlockReflowState & {...}, LineData * NEWLINE 0x010ba6e0, unsigned int & 16640) line 2087 + 24 bytes NEWLINE nsCSSBlockFrame::ReflowLinesAt(nsCSSBlockReflowState & {...}, LineData * NEWLINE 0x010ba6e0) line 1959 + 20 bytes NEWLINE nsCSSBlockFrame::ChildIncrementalReflow(nsCSSBlockReflowState & {...}) line 1928 NEWLINE + 19 bytes NEWLINE nsCSSBlockFrame::ReflowAround(nsCSSBlockFrame * const 0x010b7e80, nsIPresContext NEWLINE & {...}, nsISpaceManager * 0x010b7d00, nsReflowMetrics & {...}, const NEWLINE nsReflowState & {...}, nsRect & {...}, unsigned int & 0) line 1321 + 18 bytes NEWLINE nsBodyFrame::Reflow(nsBodyFrame * const 0x010b7c70, nsIPresContext & {...}, NEWLINE nsReflowMetrics & {...}, const nsReflowState & {...}, unsigned int & 0) line 222 NEWLINE nsContainerFrame::ReflowChild(nsIFrame * 0x010b7c70, nsIPresContext * NEWLINE 0x0109c3c0, nsReflowMetrics & {...}, const nsReflowState & {...}) line 495 NEWLINE RootContentFrame::Reflow(RootContentFrame * const 0x010b79e0, nsIPresContext & NEWLINE {...}, nsReflowMetrics & {...}, const nsReflowState & {...}, unsigned int & 0) NEWLINE line 324 + 27 bytes NEWLINE nsContainerFrame::ReflowChild(nsIFrame * 0x010b79e0, nsIPresContext * NEWLINE 0x0109c3c0, nsReflowMetrics & {...}, const nsReflowState & {...}) line 495 NEWLINE RootFrame::Reflow(RootFrame * const 0x010b7960, nsIPresContext & {...}, NEWLINE nsReflowMetrics & {...}, const nsReflowState & {...}, unsigned int & 0) line 132 NEWLINE + 27 bytes NEWLINE nsHTMLReflowCommand::Dispatch(nsHTMLReflowCommand * const 0x010cc320, NEWLINE nsIPresContext & {...}, nsReflowMetrics & {...}, const nsSize & {...}) line 133 NEWLINE PresShell::ProcessReflowCommands() line 569 NEWLINE PresShell::ContentHasBeenRemoved(PresShell * const 0x0109d864, nsIDocument * NEWLINE 0x0109b340, nsIContent * 0x010ba1b0, nsIContent * 0x010ba2f0, int 0) line 738 NEWLINE nsDocument::ContentHasBeenRemoved(nsIContent * 0x010ba1b0, nsIContent * NEWLINE 0x010ba2f0, int 0) line 558 NEWLINE nsHTMLContainer::RemoveChildAt(nsHTMLContainer * const 0x010ba1b0, int 0, int 1) NEWLINE line 222 NEWLINE nsHTMLContainer::RemoveChild(nsHTMLContainer * const 0x010ba1b4, nsIDOMNode * NEWLINE 0x010ba2f4, nsIDOMNode * * 0x0012e7b8) line 1068 + 22 bytes NEWLINE NodeRemoveChild(JSContext * 0x0109f420, JSObject * 0x01474970, unsigned int 1, NEWLINE long * 0x014700b4, long * 0x0012e84c) line 669 + 25 bytes NEWLINE js_Invoke(JSContext * 0x0109f420, unsigned int 1, int 0) line 630 + 26 bytes
692|"Below the fold" widgets are not rendered properly on Win95|Load test five on Win95 and scroll down a ways. You'll notice that text doesn't NEWLINE show up on buttons, etc.
693|We're parsing comments strangely inside a script tag|Load this URL in Raptor and you'll see a JS error on the command line saying NEWLINE that it doesn't understand "<<!--" on line 8. However, if you view source in NEWLINE Nav4, you'll see that line 8 has only one "<" -- it is: "<!--". This is a common NEWLINE first line in scripts used to hide JS from old, non-JS browsers. NEWLINE NEWLINE When the window *does* open, it's the wrong size I think, but that's a different NEWLINE bug altogether for joki.
694|ALT text on images can't handle entities|Example: NEWLINE <img src="bla.gif" width=200 height=200 alt="Hello, "Quotes" how are NEWLINE you?"> NEWLINE NEWLINE See Nav4 or IE4 for expected behavior...
695|Bottom and right borders of HRs are not visible|See test case in Nav4 and Raptor to see the problem. A one pixel border should NEWLINE surround the HR, but there is no bottom or right side border in Raptor right NEWLINE now.
696|TR tags don't expose their ID attribute to DOM|See URL for test case - watch stdout for the alerts. Currently it returns an NEWLINE empty string, rather than "mytr" (expected) or even null/undefined.
697|Raptor build status displays incorrectly|The build status for Raptor isn't displaying correctly. It's pretty obvious NEWLINE what's different
698|[LAYER] javascript doesn't work on this page|from nglayout newsgroup
699|table layout not correct|from the newsgroup: NEWLINE Problem: When viewing the index page of mozilla.org sometimes the rows NEWLINE in the main table were offset i.e. the borders were not flush with each NEWLINE other (occured several times, but not consistently). Furthermore in the NEWLINE new box and on the star logo, the left hand side of the table was not NEWLINE rendered, as well as the background color of the navagation bar towards NEWLINE the bottom. NEWLINE NEWLINE OS: Windows NT 4.0 Service Pack 3 and Post Admin Java Fix
700|can't draw unicode characters|from the newsgroup: NEWLINE ----------------------------------------------------------------------- NEWLINE Subject: NEWLINE NGLayout Chokes on ldquo and rdquo HTML Entities NEWLINE Date: NEWLINE Mon, 31 Aug 1998 15:01:22 -0400 NEWLINE From: NEWLINE "David Baron" <dbaron@netaxs.com> NEWLINE Organization: NEWLINE Another Netscape Collabra Server User NEWLINE Newsgroups: NEWLINE netscape.public.mozilla.layout NEWLINE NEWLINE NEWLINE NEWLINE NEWLINE Build: 8/25 build of NGLayout from NEWLINE http://members.tripod.com/~mozillabuilds/nglayout.html NEWLINE NEWLINE OS: Win '95 OSR2 NEWLINE NEWLINE Page: http://www.psych.upenn.edu/~baron/david/inet/out/Search.html NEWLINE NEWLINE Problem: There are four links in this page that contain the html entities NEWLINE “(left double quote) and ”(right double quote). The entire link NEWLINE that contans the entity fails to appear, intead leaving a blank space (2 are NEWLINE in the UL element at the top of the page, and then they are repeated in the NEWLINE headers below.) NEWLINE NEWLINE David Baron NEWLINE --------------------------------------------------------------------- NEWLINE L. David Baron \| Entering Freshman, Harvard NEWLINE dbaron@netaxs.com \| < http://www.psych.upenn.edu/~baron/david/ > NEWLINE dbaron@fas.harvard.edu after September 6 NEWLINE Webmaster, International Weather Satellite Imagery Center, etc. NEWLINE ---------------------------------------------------------------------
701|Multiple CSS classes aren't implemented yet|David Baron <dbaron@netaxs.com> wrote: NEWLINE NEWLINE The NGLayout viewer does not support the use of CSS with multiple classes <P NEWLINE CLASS="rbox gbox"> at all. I have written a simple test page for such use at NEWLINE http://www.psych.upenn.edu/~baron/david/csstest/twoclass.html .
702|vertical centering wrong in <p align="center">|the top frame of this page can be reduced to: NEWLINE <html><body topmargin="0" leftmargin="0" marginwidth="0" marginheight="0" > NEWLINE NEWLINE <table height="60" border=1> NEWLINE <tr height="40"> NEWLINE <td><p align="center">a</td> NEWLINE <td>b</td> NEWLINE </tr> NEWLINE </table> NEWLINE </body></html> NEWLINE NEWLINE "a" is top-aligned when it should be centered (as "b" is.) Tha table seems to NEWLINE be doing the right thing, it looks like a <p> problem.
703|align=middle fails on <h1>|<html> NEWLINE <BODY background="images/recbg.jpg"> NEWLINE <h1 align=center>Cascading Style Sheets, level 2</h1> NEWLINE The previous h1 should be centered, but isn't. :( NEWLINE </body> NEWLINE </html>
704|MFC42D.DLL required for install (nightly build)|The nightly Mozilla build is dependent upon the file MFC42D.DLL. NEWLINE NEWLINE I don't have that file, and so I am unable to test out the nightly build. I will NEWLINE go and try and harvest it from Microsoft's site, but it would be nice to have a NEWLINE link to this somewhere on the Binaries Download page. NEWLINE NEWLINE By the time someone reads this, I will probably have found a suitable URL. Get NEWLINE in touch with me, and I will give it to you. NEWLINE NEWLINE All this is contingent on the assumption that MFC42D.DLL is a NEWLINE freely-distributable file.
706|mfc42d.dll missing|mozilla.exe in 9/03/98 binary will not run. Error message says needs MFC42D.DLL NEWLINE to run. File not included in zip file and is not in Win 95 OSR2 which I am NEWLINE running.
707|CSS clear attribute not working|Clear attribute does not move text past floated HTML elements, NEWLINE i.e. div. As an example the following simple HTML fragment NEWLINE with two floated divs: NEWLINE NEWLINE <div style="float: left; width: 50%; color: #000000; background: NEWLINE #c0c0c0;"><p>Left</p></div> NEWLINE <div style="float: right; width: 50%; color: #ffffff; background: NEWLINE #000000;"><p>Right</p></div> NEWLINE <p style="clear: both; text-align: center;">New Line of Text</p>
708|Molila 5.0a1 on Redhat 5.0 does not show URL propertly.|When I use NEWLINE http://www.crn.com:80/dailies/weekending090498/sep02dig12.asp NEWLINE NEWLINE I see a mess of different pages in Mozilla 5.0a1 on Linux
709|<TR ALIGN="center"> Doesn't center cell data|I think the subject says it all... when you use a <TR ALIGN="center"> the data NEWLINE in the <TD>'s for that row are not automatically centered.
710|Pop up menu problems...|When I click right on the forward button, I can have a popup menu. NEWLINE If I select open URL, of course it goes wrong...
711|CSS color: statement ignored.|View this HTML page to reproduce. Please email directly w/ any questions; NEWLINE I filter Scopus email to trash. NEWLINE --- CUT HERE --- NEWLINE <!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 3.2//EN"> NEWLINE <HTML> NEWLINE <HEAD> NEWLINE NEWLINE <!-- * Internal HTML page title for indexing use only. Not shown on page. * NEWLINE --> NEWLINE <TITLE></TITLE> NEWLINE NEWLINE NEWLINE <!-- * If you want to add a CSS style sheet specific to this page, NEWLINE * uncomment the next line and add your formatting commands. * --> NEWLINE <STYLE TYPE="text/css"> NEWLINE /* First, set attributes of BODY element. NEWLINE These will be inherited by all elements inside BODY. NEWLINE Therefore, they will become default settings within document. NEWLINE */ NEWLINE NEWLINE BODY { NEWLINE color: red; NEWLINE background: white; NEWLINE } NEWLINE NEWLINE NEWLINE NEWLINE /* Set color for links and visited links. */ NEWLINE NEWLINE :link { color: #CC3300 } NEWLINE :visited { color: #666666 } NEWLINE :active { color: #000000 } NEWLINE NEWLINE NEWLINE NEWLINE /* Make sure that the table, row, and cell elements of NEWLINE the table which forms the skeleton of our presentation NEWLINE pages will have zero width margin and padding. NEWLINE */ NEWLINE NEWLINE .page_skeleton_table { NEWLINE margin: 0 0 0 0; NEWLINE border: 0 0 0 0; NEWLINE padding: 0 0 0 0; NEWLINE } NEWLINE NEWLINE NEWLINE NEWLINE H1 { NEWLINE font-size: 48 NEWLINE } NEWLINE NEWLINE P { NEWLINE font-size: 24 NEWLINE } NEWLINE NEWLINE LI { NEWLINE font-size: 24 NEWLINE } NEWLINE NEWLINE // show source code in a smaller font so more fits on the page NEWLINE NEWLINE PRE { NEWLINE font-size: 16 NEWLINE } NEWLINE NEWLINE // default size of text within tables NEWLINE NEWLINE TH { NEWLINE font-size: 24 NEWLINE } NEWLINE NEWLINE TD { NEWLINE font-size: 24 NEWLINE } NEWLINE NEWLINE </STYLE> NEWLINE NEWLINE NEWLINE <META NAME="GENERATOR" CONTENT="Mozilla/3.0Gold (Win95; I) [Netscape]"> NEWLINE </HEAD> NEWLINE NEWLINE <BODY> NEWLINE NEWLINE <TABLE CLASS=page_skeleton_table> NEWLINE <TR CLASS=page_skeleton_table> NEWLINE <TD CLASS=page_skeleton_table> NEWLINE NEWLINE <!-- This 1-pixel wide invisible GIF is necessary to ensure that --> NEWLINE <!-- the height of the table's top cell matches the height of the --> NEWLINE <!-- logo. Do not cut it or the top cell's height will change. --> NEWLINE NEWLINE <IMG SRC="basefile/head_inv.gif" HEIGHT=72 WIDTH=1 ALIGN=RIGHT> NEWLINE NEWLINE <!-- ******************************** --> NEWLINE <!-- * Place your header text here. * --> NEWLINE <!-- ******************************** --> NEWLINE NEWLINE <H1>All this text should be red.</H1> NEWLINE NEWLINE <!-- ******************************** --> NEWLINE <!-- * Header text ends here. * --> NEWLINE <!-- ******************************** --> NEWLINE NEWLINE </TD> NEWLINE </TR> NEWLINE NEWLINE <TR CLASS=page_skeleton_table> NEWLINE <TD CLASS=page_skeleton_table> NEWLINE NEWLINE <!-- ****************************** --> NEWLINE <!-- * Place your body text here. * --> NEWLINE <!-- ****************************** --> NEWLINE NEWLINE <P>What's wrong? NEWLINE NEWLINE NEWLINE NEWLINE NEWLINE <!-- ****************************** --> NEWLINE <!-- * Body text ends here. * --> NEWLINE <!-- ****************************** --> NEWLINE NEWLINE </TD> NEWLINE </TR> NEWLINE </TABLE> NEWLINE </BODY> NEWLINE </HTML>
712|CSS: font-family defaults don't work for cursive & fantasy fonts|When using 'class=' to call for the font-family for cursive and fantasy, the NEWLINE text is defaulting to a san serif font instead. NEWLINE Here is an abbreviated code for the above URL which will show this bug: NEWLINE NEWLINE <HTML> NEWLINE <HEAD> NEWLINE <TITLE>CSS1 Test Suite: font-family</TITLE> NEWLINE NEWLINE <STYLE type="text/css"> NEWLINE NEWLINE .three {font-family: cursive} NEWLINE .threeplus {font-family: Brush Script MT} NEWLINE .four {font-family: fantasy} NEWLINE .fourplus {font-family: wide latin} NEWLINE NEWLINE </STYLE> NEWLINE NEWLINE </HEAD> NEWLINE <BODY> NEWLINE NEWLINE <P class="three">This sentence should be in a cursive font. NEWLINE </P> NEWLINE NEWLINE <P class="threeplus">This sentence should be in Brush Script MT (a cursive NEWLINE font.) NEWLINE </P> NEWLINE NEWLINE <P class="four">This sentence should be in a fantasy font. NEWLINE </P> NEWLINE NEWLINE <P class="fourplus">This sentence should be in wide latin (a fantasy font.) NEWLINE </P> NEWLINE NEWLINE </BODY> NEWLINE </HTML>
713|CSS: Background positioning not working at all|The code for the URL: NEWLINE http://slip/projects/dogbert/test/web/html/stylesheets/cssstylesheets/torture_te NEWLINE sts/sec536.htm NEWLINE NEWLINE is this: NEWLINE NEWLINE BODY { background-image: url(bg.gif); NEWLINE background-position: right top; NEWLINE background-repeat: no-repeat; } NEWLINE .one { background-image: url(bg.gif); NEWLINE background-position: center top; } NEWLINE .two { background-image: url(bg.gif); NEWLINE background-position: top left; } NEWLINE .three { background-image: url(bg.gif); NEWLINE background-position: bottom right;} NEWLINE .four { background-image: url(bg.gif); NEWLINE background-position: 50% 50%; } NEWLINE NEWLINE Things that are wrong: NEWLINE - The yellow square which is in the upper left should be in the upper right NEWLINE - 2nd paragraph should be one yellow square at the center-top of the text. NEWLINE - 3rd paragraph should be one yellow square at the top left of the text. NEWLINE - 4th paragraph should be one yellow square at the bottom right of the text. NEWLINE - 5th paragraph should be one yellow square at the exact center of the NEWLINE paragraph. NEWLINE NEWLINE Instead of any of the above circumstances, the yellow square is used as a NEWLINE background of each paragraph.
714|CSS: using the float property makes the page look scrambled|I stole the test at the URL: NEWLINE http://slip/projects/dogbert/test/web/html/stylesheets/cssstylesheets/w3c_float_ NEWLINE test.html NEWLINE from the W3C site as it shows better than anything we currently had that the NEWLINE float attribute doesn't work. NEWLINE To see an image of what the page _should_ look like go to this URL: NEWLINE http://www.w3.org/Style/CSS/Test/w3c-l NEWLINE NEWLINE (reported at request of W3C)
715|CSS: align=center on HR causes text above it to be centered|Strange, but true the following code causes the text above the <HR> to be NEWLINE centered: NEWLINE NEWLINE <HTML><HEAD><TITLE>CSS complex tests</TITLE> NEWLINE <STYLE type="text/css"> NEWLINE HR {align: center} NEWLINE </STYLE></HEAD> NEWLINE <BODY> NEWLINE <BR>Horizontal Rule below this text. (This text should not be centered.) NEWLINE <HR> NEWLINE </BODY></HTML>
716|nested cssp bug|In most cases, you can't nest cssp elements. This is a major bug for developers, NEWLINE because it is absolutely necessary to nest elements/layers for creating certain NEWLINE D-HTML content. (note that you can nest cssp elements in IE without it breaking NEWLINE like it does in dogbert.) NEWLINE NEWLINE See the sample URL. Notice that the first nest of elements works fine, but then NEWLINE the following nests of the cssp layers somehow loose their cssp properties or NEWLINE something... NEWLINE NEWLINE Please fix this! NEWLINE NEWLINE If you have any questions, let me know. NEWLINE NEWLINE -Gary
717|CSS: border styles set to show individually all show as outset|- Go to the URL: NEWLINE http://slip/projects/marvin/css/sec5517.htm NEWLINE NEWLINE - border styles when combined to make each side a different style, instead all NEWLINE borders show as outset. NEWLINE NEWLINE - Here is that part of the code illustrating the bug for the above URL NEWLINE NEWLINE <HTML> NEWLINE <HEAD> NEWLINE <TITLE>CSS1 Test Suite: border-style</TITLE> NEWLINE <STYLE type="text/css"> NEWLINE .nine {border-style: double groove; border-color: red; border-width: thick} NEWLINE .ten {border-style: groove ridge double inset; border-color: orange; NEWLINE border-width: thick} NEWLINE </STYLE> NEWLINE </HEAD> NEWLINE <BODY> NEWLINE NEWLINE <P class="nine">This paragraph should have thick, double, top and bottom NEWLINE borders, and thick grooved side borders, and all should be red.</P> NEWLINE NEWLINE <P class="ten">This paragraph should have a grooved top border, a ridged right NEWLINE border, a double bottom border, and an inset left border. All should be thick NEWLINE and orange. Extra text, XXX XX X XXXXX XXX X X XXX XXX XXXX XXXX XXXXXX XX X NEWLINE XXXXX X XXXXXXX.</P> NEWLINE NEWLINE </BODY> NEWLINE </HTML>
718|Bad CSS formatting on page|During general surfing (I noticed most with http://www.abcnews.com) images are NEWLINE sometimes being drawn with white gaps in various places on the images.
719|TextTransform does not work on CSS for non-ASCII|Currently the CSS1 implementation does not handle TextTransform correctly for NEWLINE non-ASCII part. NEWLINE 1. For Any Single Byte charset, it only do the right thing for ASCII toupper NEWLINE and tolower, European character, Greek, NEWLINE Cyrillic characters do the wrong thing. NEWLINE 2. For any multibyte charset, TextTransform do NOTHING. NEWLINE NEWLINE Please list these as one of your CSS improvement.
720|Font size rounding bug|Peter Linss wrote: NEWLINE NEWLINE In working to make NGLayout's font handling match that of Nav's, I've uncovered NEWLINE a rounding error in the Win libfont code: NEWLINE NEWLINE modules/libfont/producers/win/src/winfp.cpp: line 875 NEWLINE NEWLINE // for passing to Windows, convert pointsize to pixel size and negative. NEWLINE if( pointSize > 0 ) NEWLINE pixelSize = (int) - pointSize * pPrmFont->YPixelPerInch / 72.0; NEWLINE NEWLINE pointSize is the input size (a double), the rounding error is introduced in the NEWLINE conversion to pixel size. NEWLINE The problem is a classic lack of parentheses, the int conversion is happening NEWLINE before the divide by 72.0. NEWLINE To do what was meant, the line should read: NEWLINE pixelSize = (int) - ((pointSize * pPrmFont->YPixelPerInch) / 72.0); NEWLINE However, this is also wrong. The proper (ie: Windows) way is to round to the NEWLINE rearest integer, not the smallest. According to the NEWLINE Windows docs the code should read: NEWLINE pixelSize = - MulDiv((int)pointSize, pPrmFont->YPixelPerInch, 72); NEWLINE That would give the correct rounding. NEWLINE NEWLINE This bug results in Nav being unable to render certain font sizes, ie: 8 point NEWLINE text comes out significantly smaller (more like 7 NEWLINE point). NEWLINE NEWLINE For more information about the way it looks checkout the above URL.
721|Fonts on toolbars not laying out correctly|The fonts on the toolbars not laying out correctly, the images -such as the NEWLINE reload button are viewable but "Reload" is halfway off the button NEWLINE NEWLINE Image at http://www.risecom.net/maasp/mozerr.jpg NEWLINE NEWLINE I am running windows 98 and have a large number of fonts on my computer.
722|Problem with priority langage|I have choose french has my priority langage (english in second position) NEWLINE And it makes this error : NEWLINE Program: C:\\SOFTS\\x86Dbg\\mozilla.exe NEWLINE File: y:\\nightly\\mozilla\\modules\\rdf\\src\\remstore.c NEWLINE Lien: 175
723|Browser Nukes if you delete prefs.js|deleteing prefs.js causes browser to give the following messages NEWLINE ver 9-4-98 nightly NEWLINE NEWLINE Microsoft Visual C++ Debug Library NEWLINE Program - mozilla.exe NEWLINE NEWLINE Damage: after Normal block (#54341) at 0x00EEE330 NEWLINE Damage: after Normal block (#54341) at 0x00EEFBD0 NEWLINE Damage: after Normal block (#54341) at 0x00EF5610 NEWLINE Damage: after Normal block (#54341) at 0x00EF5640 NEWLINE Damage: after Normal block (#54341) at 0x00EF63C0 NEWLINE Damage: after Normal block (#54341) at 0x00EF63F0 NEWLINE NEWLINE Hitting Ignore will bypass the screens, allowing the browser to operate NEWLINE recreating the prefs.js file NEWLINE NEWLINE MOZILLA caused an exception 03H in module MSVCRTD.DLL at 015f:10210c90. NEWLINE Registers: NEWLINE EAX=00000001 CS=015f EIP=10210c90 EFLGS=00000246 NEWLINE EBX=00b30000 SS=0167 ESP=00c3f314 EBP=00c3f324 NEWLINE ECX=80005968 DS=0167 ESI=00e98c60 FS=1357 NEWLINE EDX=80006768 ES=0167 EDI=00000000 GS=0000 NEWLINE Bytes at CS:EIP: NEWLINE 33 c0 85 c0 75 bd 8b 4d fc 83 79 14 03 75 6c 8b NEWLINE Stack dump: NEWLINE 00000000 00e98c60 00b30000 00ef1f70 00c3f334 10210a7a 00ef1f90 00000001 NEWLINE 00c3f344 5f428ac5 00ef1f90 00000001 00c3f444 006fa884 00ef1f90 00b30000
724|Referer: should not be sent from URL field|Load page A (say, a hotmail page that has user information encoded into the NEWLINE URL). NEWLINE NEWLINE Enter the URL of page B (say, http://www.evil-hackers.com/) and load the page. NEWLINE NEWLINE The URL of the previously displayed page gets sent to evil-hackers.com, a clear NEWLINE security hole. NEWLINE NEWLINE Solution: when a new URL is entered into the netsite: line and loaded, clear the NEWLINE Referer: field. After all, there is no logical relationship between the old NEWLINE page and the new one, so no reason to send the referer:
725|Problem with <table>|On the right side, there is a problem : NEWLINE the link "http://cedric.cnam.fr/cgi-bin/ABU/kwic3" is not accessible. NEWLINE This link is in a cell of the table.
726|Problem changing font on the fly.|When font is changed in preferences, the screen does not repaint properly after NEWLINE the change. The net result is text appears garbled until screen is refreshed. NEWLINE It is difficult to get a true redraw done.
727|[p4x] table layout problems|This bug occurs in the 9-2-98 build of NGLayout. NEWLINE NEWLINE If you look at the "search archive" and "older news" sections, you'll see that NEWLINE the older news section extends beyond the edge of the navigation bar. Both NEWLINE tables have exactly the same properties. Also, the <P> tags in the "older NEWLINE news" table do not render at all. NEWLINE NEWLINE If viewed in the 9-3-98 mozilla or Communicator 4.5, that table renders NEWLINE correctly - the text flows properly and the <P> tags render properly as well. NEWLINE NEWLINE That's all for now
728|Resizing the window also resizes an absolutely-sized frame|This bug occurs in the 9-2-98 build of NGLayout NEWLINE NEWLINE If you go to the URL, and resize the page after it's loaded by dragging the NEWLINE lower right corner, you'll notice that the upper frame, which is supposed to be NEWLINE sized absolutely at 32 pixels vertically, will resize to incorrect heights.
729| characters render with a dot|In the 9-2-98 build of NGLayout, characters render with a black filled NEWLINE circle, instead of blank.
730|This page fails to load completely, and doesn't render. Yet, adding a bunch of comments after the /HTML makes it renderable!|This bug also affects 4.06 but not Explorer (boo!) or 4.5. NEWLINE NEWLINE http://www.animeigo.com/products/videos/all/crisis3.html fails to load; view NEWLINE source shows the page to be truncates, last couple of hundred bytes gone. Page NEWLINE is being sent fine by the server. NEWLINE NEWLINE http://www.animeigo.com/products/videos/all/crisis.html loads fine. The NEWLINE difference is a couple hundred bytes of comments after the /HTML tag. They NEWLINE don't get truncated, btw! NEWLINE NEWLINE http://www.animeigo.com/products/videos/all/crisis2.html loads fine on 4.06 but NEWLINE not on the 9/4 Mac PPC build. The difference between it and crisis3.html is a NEWLINE single carriage return after the /HTML tag!
731|Page misrenders in Sept 7 1998 nightly Linux binary|Complicated graphics cause the layout engine to mess up. NEWLINE Graphics misplaced and don't scroll properly. NEWLINE Point the Sept 7 snapshot at http://www.wwc.edu/people. NEWLINE Works in the release version.
732|Title reads: Version %s-5009808600 - Mozilla...|The title for the about: page includes %s indicating there is a missing argument NEWLINE in some vsprintf() which handles it. The full title reads: NEWLINE NEWLINE Version %s-5009808600 - Mozilla
733|CSS :font-size declaration works for TD but ignored for TH.|View this HTML page to reproduce. NEWLINE --- CUT HERE --- NEWLINE <!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 3.2//EN"> NEWLINE <HTML> NEWLINE <HEAD> NEWLINE NEWLINE <!-- * Internal HTML page title for indexing use only. Not shown on page. * NEWLINE --> NEWLINE <TITLE></TITLE> NEWLINE NEWLINE NEWLINE <!-- * If you want to add a CSS style sheet specific to this page, NEWLINE * uncomment the next line and add your formatting commands. * --> NEWLINE <STYLE TYPE="text/css"> NEWLINE /* First, set attributes of BODY element. NEWLINE These will be inherited by all elements inside BODY. NEWLINE Therefore, they will become default settings within document. NEWLINE */ NEWLINE NEWLINE BODY { NEWLINE color: #000080; NEWLINE background: white; NEWLINE } NEWLINE NEWLINE NEWLINE NEWLINE /* Set color for links and visited links. */ NEWLINE NEWLINE :link { color: #CC3300 } NEWLINE :visited { color: #666666 } NEWLINE :active { color: #000000 } NEWLINE NEWLINE NEWLINE NEWLINE /* Make sure that the table, row, and cell elements of NEWLINE the table which forms the skeleton of our presentation NEWLINE pages will have zero width margin and padding. NEWLINE */ NEWLINE NEWLINE .page_skeleton_table { NEWLINE margin: 0 0 0 0; NEWLINE border: 0 0 0 0; NEWLINE padding: 0 0 0 0; NEWLINE } NEWLINE NEWLINE NEWLINE NEWLINE H1 { NEWLINE font-size: 48 NEWLINE } NEWLINE NEWLINE P { NEWLINE font-size: 24 NEWLINE } NEWLINE NEWLINE LI { NEWLINE font-size: 24 NEWLINE } NEWLINE NEWLINE // show source code in a smaller font so more fits on the page NEWLINE NEWLINE PRE { NEWLINE font-size: 16 NEWLINE } NEWLINE NEWLINE // default size of text within tables NEWLINE NEWLINE TH { NEWLINE font-size: 24 NEWLINE } NEWLINE NEWLINE TD { NEWLINE font-size: 24 NEWLINE } NEWLINE NEWLINE </STYLE> NEWLINE NEWLINE NEWLINE <META NAME="GENERATOR" CONTENT="Mozilla/3.0Gold (Win95; I) [Netscape]"> NEWLINE </HEAD> NEWLINE NEWLINE <BODY> NEWLINE NEWLINE <TABLE CLASS=page_skeleton_table> NEWLINE <TR CLASS=page_skeleton_table> NEWLINE <TD CLASS=page_skeleton_table> NEWLINE NEWLINE <!-- This 1-pixel wide invisible GIF is necessary to ensure that --> NEWLINE <!-- the height of the table's top cell matches the height of the --> NEWLINE <!-- logo. Do not cut it or the top cell's height will change. --> NEWLINE NEWLINE <IMG SRC="basefile/head_inv.gif" HEIGHT=72 WIDTH=1 ALIGN=RIGHT> NEWLINE NEWLINE <!-- ******************************** --> NEWLINE <!-- * Place your header text here. * --> NEWLINE <!-- ******************************** --> NEWLINE NEWLINE <H1>A slide with a table</H1> NEWLINE NEWLINE <!-- ******************************** --> NEWLINE <!-- * Header text ends here. * --> NEWLINE <!-- ******************************** --> NEWLINE NEWLINE </TD> NEWLINE </TR> NEWLINE NEWLINE <TR CLASS=page_skeleton_table> NEWLINE <TD CLASS=page_skeleton_table> NEWLINE NEWLINE <!-- ****************************** --> NEWLINE <!-- * Place your body text here. * --> NEWLINE <!-- ****************************** --> NEWLINE NEWLINE <P> NEWLINE NEWLINE <TABLE CELLPADDING=0 CELLSPACING=0> NEWLINE <TR> NEWLINE NEWLINE <TH> NEWLINE </TH> NEWLINE NEWLINE <TH>This should be... NEWLINE </TH> NEWLINE NEWLINE <TH>... the same size ... NEWLINE </TH> NEWLINE NEWLINE </TR> NEWLINE NEWLINE <TR> NEWLINE NEWLINE <TD> NEWLINE </TD> NEWLINE NEWLINE <TD>as NEWLINE </TD> NEWLINE NEWLINE <TD>this! NEWLINE </TD> NEWLINE NEWLINE </TR> NEWLINE NEWLINE <TR> NEWLINE NEWLINE <TD> NEWLINE </TD> NEWLINE NEWLINE <TD>What's NEWLINE </TD> NEWLINE NEWLINE <TD>wrong? NEWLINE </TD> NEWLINE NEWLINE </TR> NEWLINE </TABLE> NEWLINE NEWLINE NEWLINE <!-- ****************************** --> NEWLINE <!-- * Body text ends here. * --> NEWLINE <!-- ****************************** --> NEWLINE NEWLINE </TD> NEWLINE </TR> NEWLINE </TABLE> NEWLINE </BODY> NEWLINE </HTML>
734|Need to add system 8.x for macs.|Date: Sun, 6 Sep 1998 09:38:41 -0400 (06:38 PDT) NEWLINE To: terry@netscape.com NEWLINE From: "Robert J. Woodhead (AnimEigo)" <trebor@animeigo.com> NEWLINE Subject: Bugzilla NEWLINE NEWLINE hasn't heard about system 8.x on macs. It only goes up to 7.5
735|BUILD: ./configure fails with : configure: error: Could not find standard motif libraries.|./configure fails with : NEWLINE configure: error: Could not find standard motif libraries. NEWLINE NEWLINE Have been dowloading today (08/Sep/1998) via CVS. NEWLINE autoconf NEWLINE mkdir obj NEWLINE cd obj NEWLINE ../configure --prefix=/users2/local NEWLINE NEWLINE ..configure ran... NEWLINE Using FE: x. NEWLINE checking for XDrawLines in -lX11... yes NEWLINE checking for XextAddDisplay in -lXext... yes NEWLINE checking for IceFlush in -lICE... yes NEWLINE checking for SmcCloseConnection in -lSM... yes NEWLINE checking for XtFree in -lXt... yes NEWLINE checking for XmuLookupString in -lXmu... yes NEWLINE checking for XpmWriteFileFromImage in -lXpm... yes NEWLINE checking for XmStringCreate in -lXm... no NEWLINE configure: error: Could not find standard motif libraries. NEWLINE NEWLINE My flags: NEWLINE $CC = gcc NEWLINE $CFLAGS = -O -I/users2/local/include -I/usr/include/X11R6 NEWLINE -I/usr/include/Motif1.2 NEWLINE $LDFLAGS = -L/users2/local/lib -L/usr/lib/X11R6 -L/usr/lib/Motif1.2
736|Backgrounded images munged on some pages|Mozilla has trouble with the background images on a lot of pages, including, but NEWLINE not limited to, these: NEWLINE NEWLINE http://www.news.com/ NEWLINE http://www.castlewales.com/home.html NEWLINE NEWLINE The backgrounds will often not cover the whole page the way they're supposed to, NEWLINE or they may get redrawn in a slightly different location as the page is NEWLINE scrolled, causing them to cover the other information on the page, making it NEWLINE illegible. It appears that sometimes if you highlight the text which is being NEWLINE hidden, it becomes readable again. NEWLINE NEWLINE Also, I have noticed that if you go to each of the above URLs, in the above NEWLINE order, parts of www.news.com's background will show up in the other page. NEWLINE NEWLINE I noticed this problem on Data General DG/UX UNIX.
737|file:/// url navigation and generated directory|Bug report from Mathieu ARNOLD <arn@mygale.org>: NEWLINE NEWLINE I put file:/c:/ into the location toolbar of viewer.exe NEWLINE I clicked on Program Files NEWLINE so, the location became : file://c:/Program Files NEWLINE then i clicked on directx NEWLINE and the location became : file://c:/directx which doesn't exist... NEWLINE so the displayed page remained the same. NEWLINE then i passed over 'Up to higher level directory' and the status bar NEWLINE displayed file:/// which doesn't exist at all, so, i clicked on it, and NEWLINE it crashed...
738|Tool bars in Mozilla appear twice|Recent builds of Mozilla have started displaying an odd problem. On startup, NEWLINE Mozilla immediately resizes itself, moving the firebreathing lizard logo NEWLINE downwards, and then drawing the three toolbars at the top of the window a second NEWLINE time. NEWLINE NEWLINE The new set do not appear to be functional (when I push the buttons, nothing NEWLINE happens). The 'balloon help'-like floating tags (what are those things called NEWLINE anyway?) have debug info in them, instead of the names that appear next to the NEWLINE original button bars. NEWLINE NEWLINE This info is: NEWLINE NEWLINE For "Navigation Toolbar" it says "Debug: no tipString resource for widget Tab3 NEWLINE This message only appears in a DEBUG build" NEWLINE NEWLINE For "Location Toolbar" it says the same thing, only with "Tab4" in place of NEWLINE "Tab3". NEWLINE NEWLINE The third tool bar I usually keep minimized, and it doesn't have any "balloon NEWLINE help" here. NEWLINE NEWLINE I can minimize the extra toolbars, and they seem to pretty much go away, NEWLINE although they leave a slightly thick line between the 'real' toolbars and the NEWLINE main window. NEWLINE NEWLINE The platform where I'm seeing this is Data General DG/UX UNIX. I don't know if NEWLINE this problem shows up on other platforms or not.
739|HP-UX's cfront C++ ("CC") can't build 4th September 1998 Mozilla|28th July 1998 release of Mozilla built with HP's cfront C++ ("CC" command) NEWLINE with only minor changes. 4th September 1998 release of Mozilla has major, NEWLINE major problems building with "CC" on HP-UX 9.X or 10.X (I suspect this may NEWLINE well apply to all cfront-based C++'s, not just HP's). NEWLINE NEWLINE The list of files that gave me "aggro" (and this is up until I got really NEWLINE fred up, there may be more) are: NEWLINE NEWLINE Filename Problem NEWLINE base/src/nsAtomTable.cpp "new" operator params not treated properly NEWLINE build/build_number Constant too large ("L" on end might fix it) NEWLINE cmd/winfe/mozilla.cpp, xpcom/src/nsRepository.cpp, NEWLINE xpcom/tests/TestFactory.cpp, xpcom/tests/TestServMgr.cpp and NEWLINE xpcom/tests/dynamic/TestDynamic.cpp NEWLINE RegisterFactory() overload NEWLINE js/ref/os/hpux.h and nsprpub/pr/include/md/_hpux.cfg NEWLINE HAVE_LONG_LONG was originally undefined, probably NEWLINE because CC doesn't support long long's. Really it NEWLINE should be defined (HP's ANSI C and ANSI C++ NEWLINE both support long long, as does gcc/g++). Causes NEWLINE rippling problems in other sources files if not NEWLINE defined. NEWLINE modules/oji/public/nsIJVMManager.h NEWLINE NS_JVM_ERROR_BASE RHS constant is too large NEWLINE (perhaps cast to int ?) NEWLINE modules/oji/public/nsIJVMPlugin.h and modules/oji/src/jvmmgr.cpp NEWLINE GetPluginInstance() overload NEWLINE nsprpub/pr/include/prtypes.h NEWLINE CC doesn't support signed typedefs, so generates NEWLINE a (very frequently displayed) warning... NEWLINE nsHashtable.h CC-compiled code won't link - nsHashKey::nsHashKey NEWLINE code not visible NEWLINE NEWLINE For the moment, I'm switching to gcc/g++, but also will have a look at NEWLINE HP's ANSI C and HP's ANSI C++ ("aCC") in combo to see they work.
740|cmd/xfe/XfeWidgets/Xfe/Pane.c uses illegal cpp conditional|Many cpp's do not like "#ifdef 0" (it's illegal in strict ANSI C I suspect), NEWLINE including HP-UX's ANSI C. Line 1633 of cmd/xfe/XfeWidgets/Xfe/Pane.c uses NEWLINE "#ifdef 0" - it should be "#if 0" of course. HP's ANSI C refuses to compile NEWLINE this file without it being fixed.
741|Typo bug at line 451 of lib/layout/laybuiltin.c|I think there's a typo bug at line 451 of lib/layout/laybuiltin.c: NEWLINE builtin_width was supposed to be asigned to, but the author put builtin->width. NEWLINE Compiler whinges about an uninitialised variable in the original code without NEWLINE this being fixed.
742|NET_DisplayCookieInfoOfSiteAsHTML undefined (during MOZ_LITE build ?)|When linking the 4th September 1998 Mozilla after MOZ_LITE build, the code NEWLINE NET_DisplayCookieInfoOfSiteAsHTML is undefined. It looks like NEWLINE the end of network/protocol/http/mkaccess.c doesn't have a dummy NEWLINE NET_DisplayCookieInfoOfSiteAsHTML() function to satisfy the MOZ_LITE NEWLINE linking.
743|Makefile in xpcom/tests fails to build libreg first|The Makefile (or Makefile.in) in xpcom/tests should have NEWLINE $(DEPTH)/modules/libreg on the DIRS line because NEWLINE libreg.$(DLL_SUFFIX) is mentioned in EX_LIBS, NEWLINE but that librariy isn't actually built at this point NEWLINE in the build, so the make falls over !
744|No 1998-09-04 Mozilla mentioned in Bugzilla when I select the Mozilla product !|Bugzilla displays the various Mozilla products up to the 1998-07-28 release NEWLINE in its scrolling box, but *no* 1998-09-04 release, which is the latest one ! NEWLINE Please fix this - it's getting boring selecting "other" :-)
745|ss:url interrupts cause crash|http://w3 does two redirects, one right after the other. the redirect code calls NEWLINE the WebShell's urlLoad routine with the new url to load. the w3 test NEWLINE case simulates entering a url, hitting enter and entering another url and NEWLINE hitting enter before the first one had a chance to finish. the assert and NEWLINE subsequent crash occurs on the third url entered (i.e. the second NEWLINE interrupt/redirect) in the image library (see modules\\libimg\\src\\if.cpp NEWLINE il_frame_complete_notify() around line: 186). We never seem to be NEWLINE interrupting/aborting the streams we create during the url loads when they're NEWLINE interrupted with another top level load. Are we handling interruption yet? Does NEWLINE this bug look like an interruption issue?
746|missing CSS styles|we don't seem to apply the style to the various paragraphs defined at the top NEWLINE of the page.
747|Bulleted list items eat anchors|if a list item starts with an anchor, ie: NEWLINE <ol> NEWLINE <li><a href=foo>item</a> one</li> NEWLINE <li>item two</li> NEWLINE <li>item three</li> NEWLINE </ol> NEWLINE NEWLINE then the anchor gets treated like a bullet, and the item doesn't get a real NEWLINE bullet.
748|Bugzilla OS versions missing|The "OS" field lists only up to MacOS version 7.5. 8.1 is the current MacOS NEWLINE version. There were also 7.6 and 8.0 releases
749|nuke GetFontCache API's|hide it
750|assigning bugs to non-owners (bugsplat, but no bugzilla acct)|I got this error while trying to reassign a bug. How do we assign bugs to NEWLINE people not on the owners list? NEWLINE NEWLINE The name morse@netscape.com is not a valid username. Please hit the Back button NEWLINE and try again.
751|GetPosition in nsTextFrame is slow|Currently it does a linear scan forward. A simple demonstration of why this is NEWLINE very bad can be made by creating a text element with, say 10,000 characters in NEWLINE it, and then watching the selection speed suck. NEWLINE NEWLINE Something on the order of a binary search seeded assuming the x coordinate maps NEWLINE to 1/2 the max-width of the font will probably work much better
752|nsISelection needs to be XPCOM'd|And while you are at it, you should clean up SelectionRange and NEWLINE SelectionPoint. Since they need to be visible to nsISelection, NEWLINE they should be defined (as structs) in nsISelection and frozen NEWLINE forever. NEWLINE NEWLINE Second, the API's to get them should return const&'s and (for ranges and NEWLINE point's). NEWLINE NEWLINE If you want me to review it, I'll be glad to review it before you complete it.
753|Combined nsImage* & gfxImageFrame|Setting all current Open/Normal to M4.
754|CSS Selectors not targetting tags properly.|At this page: NEWLINE http://www.w3.org/Style/CSS/Test/current/sec15.htm NEWLINE NEWLINE in win95 osr2 with the august 2nd build, the second line that reads: "This NEWLINE sentence should be green." is not green. It is black.
755|rdf netchrome causes scroll bar flashing|Mozilla starts up, displays one page and then freezes. If I edit NEWLINE /etc/hosts and point rdf.netscape.com to some invalid ip address NEWLINE then mozilla will start up without freezing but i'm unable to NEWLINE follow links. A few minutes later a blank popup appears (presumably NEWLINE an error message) and two scrollbar sized strips along the right and NEWLINE bottom edges of the frame start flashing between gray and white. NEWLINE NEWLINE Looking in gdb, it looks like there's an infinite loop with the code NEWLINE thrashing around between doc_bbox_changed_func and scroller_resize. Every NEWLINE few iterations in doc_bbox_changed_func the bounding box has a height NEWLINE and width of zero.
756|ABBR not handled correctly|Not sure if this is a parser or a content model problem, but ABBR seems to hose NEWLINE things. NGLayout inserts a newline before any ABBR content, and may hose nearby NEWLINE inline markup.
757|"width" property not properly enforced|The "width" property is not properly enforced. The URL specifies a DIV with a NEWLINE width of 400px. But the width of this element is always sized to fit the NEWLINE browser window width.
758|Highlighting text in the viewer causes the app to crash|Highlighting text in the viewer (by clicking and dragging with the mouse) NEWLINE causes the application to crash with an Application Error in NEWLINE NetscapeDispatchWnd. Looks like you tried to read a zero pointer.
759|Mozilla inserts a few pixels of space between the border and the padding/content|The summary pretty much sums it up; see the URL provided for an example.
760|Problem with the background of a parent element|Refer to the example. The background specified for the outer DIV should show NEWLINE through in the margins of the inner DIV.
761|CSS "width" not interpreted correctly.|The "width" property should specify the "content width". This is the space NEWLINE *between* the left and right padding. So, if the width is 40, and the padding NEWLINE on each side is 10, the width of the [content width] + [padding] would be 60. NEWLINE Please refer to the example. It specifies a with of 400px, but Mozilla renders NEWLINE it with a width of only ~240 pixels.
762|Block element between left- and right-floated images is rendered incorrectly|Please refer to the example. Mozilla puts the images in the right place, but NEWLINE the content of the DIV should be *between* them. Mozilla has it rendered tall NEWLINE and skinny way off to the right.
763|Display "jumps" when resized|When resizing navigator window, layout of all pieces on screen jumps about ten NEWLINE pizels down and right. Unfortunately, I just noticed that text-entry fields will NEWLINE sometimes (on first resize?) completely lose their positioning as well, which NEWLINE turns out little more severe problem.
764|UTF8 strings in taskbar...|New bookmarks code sucks in taskbar elements, however for some reason this code NEWLINE doesn't handle UTF8 strings. Program entries like 'Forté Agent' cause a volley NEWLINE of ASSERT()'s and an eventual memory error.
765|Erased bookmark folder doesn't accept bookmarks|Well, here's an unsurprising bug... when you remove ("Cut") the bookmarks-folder NEWLINE from the toolbars, and subsequently try to create another bookmark, things crash NEWLINE since the bookmark-folder doesn't exist. Presumably, Mozilla shouldn't allow one NEWLINE to Cut off the active bookmark-folder (Like it doesn't allow you to do with new NEWLINE bookmarks-folder).
766|Empty title in 'history' list|I was unable to reproduce this as of yet, so there isn't much to tell. While NEWLINE trying to access a very slow-loading page I started browsing thru the pull-down NEWLINE menues, until Mozilla suddenly came down on an ASSERT(). NEWLINE NEWLINE CHistoryMenuBuilder::AddHistoryToMenu NEWLINE (HMENU__ * 0x000602ce, int 0, int 1, _History_entry * 0x00ac41c0, int 45) NEWLINE line 176 + 33 bytes NEWLINE CHistoryMenuBuilder::FillItems NEWLINE (HMENU__ * 0x000602ce, _XP_List * 0x00ac8da0, int 0, int 45, NEWLINE _History_entry * 0x00ac41c0) line 238 NEWLINE CHistoryMenuBuilder::Fill NEWLINE (HMENU__ * 0x000602ce, int 45, eFillEnum eFILLALL) line 106 NEWLINE CMainFrame::BuildHistoryMenu(CMenu * 0x00a61b50 {CMenu}) line 600 NEWLINE CMainFrame::OnInitMenuPopup NEWLINE (CMenu * 0x00a61b50 {CMenu}, unsigned int 3, int 0) line 540 NEWLINE CWnd::OnWndMsg NEWLINE (unsigned int 279, unsigned int 393934, long 3, long * 0x0012f6c4) line 1858 NEWLINE CWnd::WindowProc NEWLINE (unsigned int 279, unsigned int 393934, long 3) line 1575 + 30 bytes NEWLINE AfxCallWndProc NEWLINE (CWnd * 0x01be85e8 {CMainFrame hWnd=0x00050306}, HWND__ * 0x00050306, NEWLINE unsigned int 279, unsigned int 393934, long 3) line 217 + 26 bytes NEWLINE AfxWndProc NEWLINE (HWND__ * 0x00050306, unsigned int 279, unsigned int 393934, long 3) line 371 NEWLINE AfxWndProcBase NEWLINE (HWND__ * 0x00050306, unsigned int 279, unsigned int 393934, long 3) NEWLINE line 203 + 21 bytes NEWLINE ...
767|Messed up directory/FTP listings|I've seen this problem mostly on FTP listings, altough I seem to remember seeing NEWLINE it appear on local directory listings at times. Quite easy to spot, the FTP file NEWLINE descriptions - especially for unknown filetypes - seem to be going a little here NEWLINE and there...
768|Non-scrolling URL entry box|That ought to already cover it. Since the URL entry box can't be scrolled, NEWLINE entering longer URL's into it is impossible. Altough Ctrl-O always works, I NEWLINE somehow doubt this was the intention.
769|Spacers in tables know trouble...|I was playing around with a spacer-phenomenon using the following HTML snippet: NEWLINE foo<table border=1><tr><td height=3><spacer width=1></td></tr></table>bar It NEWLINE occurs occasionally at least with "block"-type spacers that the table ends up at NEWLINE an unreasonable height. Anyway, using the above piece, after a couple of NEWLINE refreshes: NEWLINE NEWLINE _free_dbg_lk(void * 0xdddddddd, int 1) line 1011 + 48 bytes NEWLINE _free_dbg(void * 0xdddddddd, int 1) line 970 + 13 bytes NEWLINE free(void * 0xdddddddd) line 926 + 11 bytes NEWLINE PA_FreeTag(PA_Tag_struct * 0x00ad0850) line 222 + 13 bytes NEWLINE lo_free_cell_record(MWContext_ * 0x00a55340, lo_DocState_struct * 0x01c17440, NEWLINE lo_TableCell_struct * 0x00ad04f0) line 6504 + 9 bytes NEWLINE lo_ScrapeElement(MWContext_ * 0x00a55340, LO_Element_struct * 0x01c4e0bc, int 1) NEWLINE line 693 + 20 bytes NEWLINE lo_RecycleElements(MWContext_ * 0x00a55340, lo_DocState_struct * 0x01c17440, NEWLINE LO_Element_struct * 0x01c4dea4) line 1946 + 15 bytes NEWLINE lo_InternalDiscardDocument(MWContext_ * 0x00a55340, lo_DocState_struct * NEWLINE 0x01c17440, pa_DocData_struct * 0x00acb430, int 1) line 5989 + 17 bytes NEWLINE lo_GetRecycleList(MWContext_ * 0x00a55340, long 3, pa_DocData_struct * NEWLINE 0x00acb430, LO_Element_struct * * 0x0012f818, lo_arena_struct * * 0x0012f814) NEWLINE line 3933 + 19 bytes NEWLINE LO_ProcessTag(void * 0x00acb430, PA_Tag_struct * 0x00acbdb0, int 0) line 4533 + NEWLINE 31 bytes NEWLINE EDT_ProcessTag(void * 0x00acb430, PA_Tag_struct * 0x00acbdb0, int 0) line 970 + NEWLINE 17 bytes NEWLINE PA_ParseBlock(_NET_StreamClass * 0x00acb760, char * 0x01be2e58, int 1960) line NEWLINE 1165 + 16 bytes NEWLINE net_AutoCharCodeConv(_NET_StreamClass * 0x00acbe40, char * 0x01be2e58, long NEWLINE 1960) line 180 + 24 bytes NEWLINE net_read_file_chunk(_ActiveEntry * 0x00acb820) line 904 + 27 bytes NEWLINE net_ProcessFile(_ActiveEntry * 0x00acb820) line 1264 + 9 bytes NEWLINE NET_ProcessNet(PRFileDesc * 0x00000000, int 1) line 3334 + 13 bytes NEWLINE net_process_slow_net_timer_callback(void * 0x00000000) line 216 + 9 bytes NEWLINE wfe_ProcessTimeouts(unsigned long 14242339) line 303 + 12 bytes NEWLINE FireTimeout(HWND__ * 0x0074036e, unsigned int 275, unsigned int 777, unsigned NEWLINE long 14242339) line 60 + 9 bytes NEWLINE USER32! 77e71373() NEWLINE CNetscapeApp::Run() line 1675 + 8 bytes NEWLINE AfxWinMain(HINSTANCE__ * 0x00400000, HINSTANCE__ * 0x00000000, char * NEWLINE 0x00142040, int 10) line 52 + 11 bytes NEWLINE WinMain(HINSTANCE__ * 0x00400000, HINSTANCE__ * 0x00000000, char * 0x00142040, NEWLINE int 10) line 34 NEWLINE WinMainCRTStartup() line 330 + 54 bytes NEWLINE KERNEL32! 77f1b304()
770|HTML entities don't work in DOM|I tested your DOM examples at NEWLINE NEWLINE http://www.mozilla.org/layout/demos/ NEWLINE NEWLINE and found a bug. The demo for changing the title and the NEWLINE text value don't work with HTML entities such as ö etc.
771|Improper free of memory causes an Assert on quitting app|Pretty simple - this was actually caught by your own memory manager, NEWLINE here's the MacsBug stdlog. Note it was with System 8.1, which was not NEWLINE an option in the popup list. NEWLINE NEWLINE NEWLINE MacsBug 6.5.4a4, Copyright Apple Computer, Inc. 1981-98 NEWLINE NEWLINE User break at 09AA0684 free+000A4 NEWLINE fastmem: attempt to dispose illegal block NEWLINE NEWLINE 9-Sep-1998 8:58:37 AM (since boot = 4 hours, 39 minutes) NEWLINE Current application is “NavigatorDebug” NEWLINE Machine = 510 (PowerMacG3), System $0810, sysu = $01008000 NEWLINE ROM version $077D, $40F2, $0001 (ROMBase $FFC00000) NEWLINE VM is on; paging is currently safe NEWLINE NIL^ = $FFC10000 NEWLINE Stack space used = +112349186 NEWLINE Address 09AA0684 is in VM file-mapped logical memory space NEWLINE The address is in a CFM fragment “MemoryAllocator” [non-write exec] NEWLINE It is 00001684 bytes from the start of the fragment NEWLINE PowerPC 740/750 Registers NEWLINE CR0 CR1 CR2 CR3 CR4 CR5 CR6 CR7 NEWLINE PC = 09AA0684 CR 1000 1100 0000 0000 0000 1000 1000 0000 NEWLINE LR = 09AA0684 <>=O XEVO NEWLINE CTR = FFD69AA8 NEWLINE MSR = 00000000 SOC Compare Count NEWLINE Int = 0 XER 001 00 00 NEWLINE NEWLINE R0 = FFD69AA8 R8 = 55534544 R16 = 07B8E520 R24 = 00000000 NEWLINE SP = 06733810 R9 = 3A006564 R17 = 079F5AFC R25 = 099FE359 NEWLINE TOC = 0002F1B8 R10 = 75736564 R18 = 00000000 R26 = 00000000 NEWLINE R3 = 09AA20A2 R11 = 09AA2057 R19 = 061321B8 R27 = 0640E118 NEWLINE R4 = 55534544 R12 = 09AA0684 R20 = 099F8147 R28 = 00000000 NEWLINE R5 = 46524545 R13 = 00000000 R21 = 0785ADC4 R29 = 0640E12C NEWLINE R6 = 00000000 R14 = 00000000 R22 = 07A053BC R30 = 0640E12C NEWLINE R7 = 55534544 R15 = 000000D0 R23 = 00000000 R31 = 0640E118 NEWLINE Disassembling PowerPC code from 09AA065C NEWLINE free NEWLINE +0007C 09AA065C cmplw r7,r8 \| NEWLINE 7C074040 NEWLINE +00080 09AA0660 bne free+00098 ; 0x09AA0678 \| NEWLINE 40820018 NEWLINE +00084 09AA0664 lwz r9,0x0000(r29) \| NEWLINE 813D0000 NEWLINE +00088 09AA0668 lis r10,0x7573 \| NEWLINE 3D407573 NEWLINE +0008C 09AA066C addi r10,r10,0x6564 \| NEWLINE 394A6564 NEWLINE +00090 09AA0670 cmplw r9,r10 \| NEWLINE 7C095040 NEWLINE +00094 09AA0674 beq free+000AC ; 0x09AA068C \| NEWLINE 41820018 NEWLINE +00098 09AA0678 lwz r11,0x00FC(RTOC) \| NEWLINE 816200FC NEWLINE +0009C 09AA067C addi r3,r11,0x004B \| NEWLINE 386B004B NEWLINE +000A0 09AA0680 bl DebugStr ; 0x09AA1E80 \| NEWLINE 48001801 NEWLINE +000A4 09AA0684 *lwz RTOC,0x0014(SP) \| NEWLINE 80410014 NEWLINE +000A8 09AA0688 b free+00160 ; 0x09AA0740 \| NEWLINE 480000B8 NEWLINE +000AC 09AA068C lwz r12,0x000C(r31) \| NEWLINE 819F000C NEWLINE +000B0 09AA0690 cmplwi r12,0x0000 \| NEWLINE 280C0000 NEWLINE +000B4 09AA0694 beq free+000C8 ; 0x09AA06A8 \| NEWLINE 41820014 NEWLINE +000B8 09AA0698 lwz r3,0x0010(r31) \| NEWLINE 807F0010 NEWLINE +000BC 09AA069C lwz r4,0x000C(r31) \| NEWLINE 809F000C NEWLINE +000C0 09AA06A0 stw r3,0x0010(r4) \| NEWLINE 90640010 NEWLINE +000C4 09AA06A4 b free+000D4 ; 0x09AA06B4 \| NEWLINE 48000010 NEWLINE +000C8 09AA06A8 lwz r5,0x0010(r31) \| NEWLINE 80BF0010 NEWLINE Heap zones NEWLINE #1 Mod 10725K 00002800 to 00A7BC1F SysZone^ NEWLINE #2 Mod 6K 000145C0 to 000160DF ROM read-only zone NEWLINE #3 Mod 48K 0003DAD0 to 00049ACF NEWLINE #4 Mod 187K 0073AC60 to 00769AFF NEWLINE #5 Mod 256K 00769B20 to 007A9B1F NEWLINE #6 Mod 115786K 00A7BC20 to 07B8E7CF Process Manager zone NEWLINE #7 Mod 6979K 060540F0 to 06724FEF “NavigatorDebug” ApplZone^ NEWLINE TheZone^ Target NEWLINE Zone NEWLINE #8 Mod 4679K 06735100 to 06BC6FFF “Acrobat™ Reader 3.01” NEWLINE #9 Mod 12153K 06BD7110 to 077B58DF “Netscape 4.0” NEWLINE #10 Mod 519K 077C5A00 to 078478FF “Telnet 2.7b4” NEWLINE #11 Mod 713K 07867E50 to 0791A54F “NetPresenz” NEWLINE #12 Mod 187K 0792AC80 to 079599BF “NotifyMail” NEWLINE #13 Mod 183K 079BE030 to 079EBC9F “File Sharing Extension” NEWLINE #14 Mod 896K 07A09FA0 to 07AEA29F “Finder” NEWLINE #15 Mod 66K 07B1A2A0 to 07B2AE8F “EPSON Launcher” NEWLINE #16 Mod 23K 07B53A20 to 07B596CF “DT Timer” NEWLINE Checking all heaps NEWLINE The System heap at 00002800 is ok NEWLINE The ROM read-only heap at 000145C0 is ok NEWLINE The heap at 0003DAD0 is ok NEWLINE The heap at 0073AC60 is ok NEWLINE The heap at 00769B20 is ok NEWLINE The Process Manager heap at 00A7BC20 is ok NEWLINE The “NavigatorDebug” heap at 060540F0 is ok NEWLINE Totaling the “NavigatorDebug” heap at 060540F0 NEWLINE Total Blocks Total of Block Sizes NEWLINE Free 0040 #64 000E96F0 #956144 NEWLINE Nonrelocatable 0097 #151 005AE70C #5957388 NEWLINE Relocatable 0132 #306 000390C0 #233664 NEWLINE Locked 0004 #4 00000300 #768 NEWLINE Purgeable and not locked 0039 #57 00003490 #13456 NEWLINE Heap size 0209 #521 006D0EBC #7147196 NEWLINE The target heap is the System heap at 00002800 NEWLINE Totaling the System heap at 00002800 NEWLINE Total Blocks Total of Block Sizes NEWLINE Free 0030 #48 00019810 #104464 NEWLINE Nonrelocatable 093C #2364 004C7F9C #5013404 NEWLINE Relocatable 0893 #2195 00597C30 #5864496 NEWLINE Locked 01B0 #432 003DA920 #4040992 NEWLINE Purgeable and not locked 0118 #280 000E6E90 #945808 NEWLINE Heap size 11FF #4607 00A793DC #10982364 NEWLINE The target heap is the “NavigatorDebug” heap at 060540F0 NEWLINE Displaying File Control Blocks NEWLINE fRef File Vol Type Fl Fork LEof NEWLINE 0002 System Macintosh … zsys dW rsrc #5958092 NEWLINE 0060 **** EXTENTS B-TREE Macintosh … •••• dw data #4194304 NEWLINE 00BE **** CATALOG B-TREE Macintosh … •••• dw data #8478720 NEWLINE 011C **** VOLUME BITMAP Macintosh … •••• dw data #976896 NEWLINE 017A VM Storage Macintosh … ZSYS dW data #135266304 NEWLINE 1CA6 System Macintosh … zsys dw data #966352 NEWLINE 1D04 Open Tpt AppleTalk Li… Macintosh … libr dw rsrc #541758 NEWLINE 1D62 Open Transport Library Macintosh … libr dw rsrc #587356 NEWLINE 1DC0 OpenTransportLib Macintosh … shlb dw data #563840 NEWLINE 1E1E Shared Library Manage… Macintosh … INIT dw rsrc #211694 NEWLINE 1E7C Open Transport Library Macintosh … libr dw rsrc #587356 NEWLINE 1EDA Open Tpt AppleTalk Li… Macintosh … libr dw rsrc #541758 NEWLINE 1F38 Open Tpt Internet Lib… Macintosh … libr dw rsrc #485159 NEWLINE 1F96 OpenTpt Remote Access Macintosh … libr dw rsrc #541432 NEWLINE 1FF4 OpenTpt Modem Macintosh … libr dw rsrc #85766 NEWLINE 2052 OpenTpt Remote Access Macintosh … libr dw rsrc #541432 NEWLINE 20B0 Remote Access Log Macintosh … lzlg dW data #245760 NEWLINE 210E Serial (Built-in) Macintosh … libr dw rsrc #63254 NEWLINE 216C Open Transport Library Macintosh … libr dw rsrc #587356 NEWLINE 21CA OpenTptAppleTalkLib Macintosh … shlb dw data #54706 NEWLINE 2228 OpenTpt Serial Arbitr… Macintosh … libr dw rsrc #7638 NEWLINE 2286 OpenTpt Serial Arbitr… Macintosh … libr dw rsrc #7638 NEWLINE 22E4 ATI Video Memory Mana… Macintosh … shlb dw data #18010 NEWLINE 2342 Appearance Extension Macintosh … INIT dw rsrc #633281 NEWLINE 23A0 Norton AntiVirus Libr… Macintosh … shlb dw data #416419 NEWLINE 23FE Users & Groups Data F… Macintosh … BTFL dW data #245760 NEWLINE 245C Appearance Extension Macintosh … INIT dw data #121624 NEWLINE 24BA Norton AntiVirus Macr… Macintosh … shlb dw data #410856 NEWLINE 2518 Appearance Extension Macintosh … INIT dw rsrc #633281 NEWLINE 2576 Contextual Menu Exten… Macintosh … INIT dw data #66103 NEWLINE 25D4 SOMobjects™ for Mac OS Macintosh … shlb dw data #128944 NEWLINE 2632 OT AutoPush Support Macintosh … INIT dw data #7392 NEWLINE 2690 QuickTime™ Macintosh … INIT dw data #298076 NEWLINE 26EE Mac OS Easy Open Macintosh … cdev dw rsrc #132666 NEWLINE 274C Speech Manager Macintosh … INIT dw data #4000 NEWLINE 27AA ~ATM™ Macintosh … cdev dw data #881790 NEWLINE 2866 DT Timer Macintosh … appe dW rsrc #15803 NEWLINE 28C4 EPSON Launcher Macintosh … appe dW rsrc #7631 NEWLINE 2922 Program Switcher CMM Macintosh … cmpi dw data #6481 NEWLINE 2980 Finder Macintosh … FNDR dW rsrc #502012 NEWLINE 29DE Finder Macintosh … FNDR dw data #3631000 NEWLINE 2A3C Apple Guide Macintosh … INIT dw data #351914 NEWLINE 2A9A Color Picker Macintosh … INIT dw data #6400 NEWLINE 2AF8 File Sharing Library Macintosh … shlb dw data #91214 NEWLINE 2B56 Mac OS Easy Open Macintosh … cdev dw data #8320 NEWLINE 2BB4 PrintingLib Macintosh … shlb dw data #908576 NEWLINE 2C12 File Sharing Library Macintosh … shlb dw rsrc #3483 NEWLINE 2C70 Finder Preferences Macintosh … pref dW rsrc #1186 NEWLINE 2CCE Desktop DB Macintosh … BTFL dW data #491520 NEWLINE 2D2C Desktop DF Macintosh … DTFL dW data #3454242 NEWLINE 2D8A OpenTptInternetLib Macintosh … shlb dw data #275054 NEWLINE 2DE8 Open Tpt Internet Lib… Macintosh … libr dw rsrc #485159 NEWLINE 2E46 DT Timer Macintosh … appe dw data #10354 NEWLINE 2EA4 File Sharing Extension Macintosh … INIT dW rsrc #194076 NEWLINE 2F02 Timbuktu Extension Macintosh … appe dw rsrc #788215 NEWLINE 2F60 Timbuktu Resources Macintosh … eTB2 dw rsrc #527 NEWLINE 2FBE Telnet 2.7b4 Macintosh … APPL dW rsrc #308391 NEWLINE 301C Timbuktu Pro Preferen… Macintosh … pref dW rsrc #4364 NEWLINE 307A AppleTalk Transport D… Macintosh … dTB2 dw rsrc #415119 NEWLINE 30D8 Chat DropIn Macintosh … dTB2 dw rsrc #244694 NEWLINE 3136 Dial Direct DropIn Macintosh … dTB2 dw rsrc #538963 NEWLINE 3194 Exchange Host DropIn Macintosh … dTB2 dw rsrc #117675 NEWLINE 31F2 FlashNotes DropIn Macintosh … dTB2 dw rsrc #517432 NEWLINE 3250 Intercom DropIn Macintosh … dTB2 dw rsrc #678507 NEWLINE 32AE Notify DropIn Macintosh … dTB2 dw rsrc #147967 NEWLINE 330C ScreenShare Host Drop… Macintosh … dTB2 dw rsrc #377199 NEWLINE 336A TCP/IP Transport Drop… Macintosh … dTB2 dw rsrc #181520 NEWLINE 33C8 Timbuktu Log Macintosh … TEXT dW data #31135 NEWLINE 3426 Telnet 2.7b4 Macintosh … APPL dw data #251521 NEWLINE 3484 NotifyMail Macintosh … APPL dW rsrc #164664 NEWLINE 34E2 NetPresenz Macintosh … APPL dW rsrc #250968 NEWLINE 3540 EPSON Launcher Macintosh … appe dw data #6529 NEWLINE 359E EPSON Printer Library Macintosh … shlb dw data #1202624 NEWLINE 35FC EPSON Printer Utility Macintosh … shlb dw data #83394 NEWLINE 365A Stylus COLOR 800 Libr… Macintosh … shlb dw data #34739 NEWLINE 36B8 NetPresenz Macintosh … APPL dw data #311905 NEWLINE 3716 NetPresenz Log Macintosh … TEXT dW data #2565815 NEWLINE 3774 NotifyMail Macintosh … APPL dw data #158393 NEWLINE 37D2 AppleScriptLib Macintosh … shlb dw data #22636 NEWLINE 3830 Users & Groups Data F… Macintosh … BTFL dW data #245760 NEWLINE 388E AppleShare PDS Macintosh … BTFL dW data #983040 NEWLINE 38EC NCSA Telnet Preferenc… Macintosh … pref dW rsrc #3594 NEWLINE 394A Netscape 4.0 Macintosh … APPL dW rsrc #769979 NEWLINE 39A8 Netscape 4.0 Macintosh … APPL dw data #5453612 NEWLINE 3A06 QuickTime™ PowerPlug Macintosh … INIT dw data #170185 NEWLINE 3A64 TrueDocDisplayer.dlm Macintosh … shlb dw data #366152 NEWLINE 3AC2 Netscape Resources Macintosh … NSPL dW rsrc #631671 NEWLINE 3B20 Global History Macintosh … DBMG dW data #684032 NEWLINE 3B7E CCache log Macintosh … DBMC dW data #102400 NEWLINE 3BDC Certificates7 Macintosh … CERT dW data #98304 NEWLINE 3C3A Key Database3 Macintosh … TEXT dW data #16384 NEWLINE 3C98 Security Macintosh … TEXT dW data #16384 NEWLINE 3CF6 JavaScript Debug Supp… Macintosh … shlb dw data #24737 NEWLINE 3D54 NavigatorDebug Macintosh … APPL dW rsrc #916759 NEWLINE 3DB2 NavigatorDebug Macintosh … APPL dw data #4622745 NEWLINE 3E10 Acrobat™ Reader 3.01 Macintosh … APPL dW rsrc #838740 NEWLINE 3E6E Acrobat™ Reader 3.01 Macintosh … APPL dw data #2055473 NEWLINE 3ECC Acrobat™ Reader Prefs Macintosh … PREF dW rsrc #4042 NEWLINE 3F2A Acrobat Temp00000 Macintosh … TEMP dW rsrc #486 NEWLINE 3F88 Acrobat™ WebLink Macintosh … XTND dw rsrc #101654 NEWLINE 3FE6 AcroForm Macintosh … XTND dw rsrc #327317 NEWLINE 4044 EWH Macintosh … XTND dw rsrc #56547 NEWLINE 40A2 HLS Macintosh … XTND dw rsrc #20728 NEWLINE 4100 Movie Macintosh … XTND dw rsrc #150639 NEWLINE 415E Acrobat™ WebLink Macintosh … XTND dw data #89966 NEWLINE 41BC AcroForm Macintosh … XTND dw data #366885 NEWLINE 421A EWH Macintosh … XTND dw data #78366 NEWLINE 4278 HLS Macintosh … XTND dw data #26251 NEWLINE 42D6 Movie Macintosh … XTND dw data #150664 NEWLINE 4334 Acrobat™ Weblink Prefs Macintosh … PREF dW data #0 NEWLINE 4392 DBMDebug.shlb Macintosh … shlb dw data #24651 NEWLINE 43F0 MemAllocatorDebug.shlb Macintosh … shlb dw data #14137 NEWLINE 444E NSRuntimeDebug.shlb Macintosh … shlb dw data #30490 NEWLINE 44AC NSPR20Debug.shlb Macintosh … shlb dw data #249914 NEWLINE 450A NSStdLibDebug.shlb Macintosh … shlb dw data #307182 NEWLINE 4568 LaserWriter 8 Macintosh … PRER dw data #732592 NEWLINE 45C6 MoreFilesDebug.shlb Macintosh … shlb dw data #58342 NEWLINE 4624 JavaRuntimeDebug.shlb Macintosh … shlb dw data #1574 NEWLINE 4682 JavaScriptDebug.shlb Macintosh … shlb dw data #410256 NEWLINE 46E0 JPEGDebug.shlb Macintosh … shlb dw data #64897 NEWLINE 473E NavJavaDebug.shlb Macintosh … shlb dw data #5780 NEWLINE 479C RDFDebug.shlb Macintosh … shlb dw data #244079 NEWLINE 47FA StringsDebug.shlb Macintosh … shlb dw data #117531 NEWLINE 4858 XMLDebug.shlb Macintosh … shlb dw data #117700 NEWLINE 48B6 zlibDebug.shlb Macintosh … shlb dw data #41910 NEWLINE 4914 xpcomDebug.shlb Macintosh … shlb dw data #22000 NEWLINE 4972 PowerPlantDebug.shlb Macintosh … shlb dw data #1226277 NEWLINE 49D0 LiveConnectDebug.shlb Macintosh … shlb dw data #86161 NEWLINE 4A2E Log.txt Macintosh … •••• dW data #34750 NEWLINE 4A8C Mozilla Resources Macintosh … NSPL dW rsrc #512564 NEWLINE 4AEA names.db Macintosh … BINA dW data #896 NEWLINE 4B48 child.db Macintosh … BINA dW data #16384 NEWLINE 4BA6 lstr.db Macintosh … BINA dW data #4096 NEWLINE 4C04 ilstr.db Macintosh … BINA dW data #65536 NEWLINE 4C62 CCache log Macintosh … DBMC dW data #16384 NEWLINE 4CC0 StdLog Macintosh … TEXT dW data #14422 NEWLINE #254 FCBs, #210 in use (including #74 fonts not listed), #44 free NEWLINE Displaying resource information: NEWLINE > Map $06054278, flags $0000, file $3D54 = NavigatorDebug NEWLINE Map $06137D58, flags $0000, file $4A8C = Mozilla Resources NEWLINE + Map $0000341C, flags $001A, file $2342 = Appearance Extension NEWLINE + Map $000032AC, flags $801E, file $0003 = •ROM resources that override NEWLINE System• NEWLINE S Map $00003330, flags $000D, file $0002 = System NEWLINE Map $005496F8, flags $0000, file $26EE = Mac OS Easy Open NEWLINE [Skipped $004A maps belonging to font files] NEWLINE Calling chain using A6/R1 links NEWLINE Back chain ISA Caller NEWLINE 00000000 PPC 0987AB2C XP_GetNavCenterContext+18858 NEWLINE 06734E30 PPC 096A3E7C main+00154 NEWLINE 06734DE0 PPC 0969E3E0 CFrontApp::~CFrontApp()+0022C NEWLINE 06734D60 PPC 09892080 NET_CleanupCacheDirectory+000DC NEWLINE 06733B40 PPC 09891DA0 net_cache_recursive_file_finder+000CC NEWLINE 06733AC0 PPC 09AC3F98 PR_OpenDir+00038 NEWLINE 06733A80 PPC 09AD941C _MD_OpenDir+000D0 NEWLINE 067338A0 PPC 09ACBF8C PR_Free+00014 NEWLINE Return addresses on the stack NEWLINE Stack Addr Frame Addr ISA Caller NEWLINE 06733BD8 PPC FFDB8914 NQDSetCCursor+03870 NEWLINE 06733BA8 PPC FFDDBD74 NQDStretch+006F4 NEWLINE 06733B48 PPC 09892080 NET_CleanupCacheDirectory+000DC NEWLINE 06733B18 68K 07A053BA NEWLINE 06733B0C 68K 061321B6 NEWLINE 06733AF8 68K 098091B6 XP_HashListNew+000A6 NEWLINE 06733AC8 06733AC0 PPC 09891DA0 net_cache_recursive_file_finder+000CC NEWLINE 06733A88 06733A80 PPC 09AC3F98 PR_OpenDir+00038 NEWLINE 06733A48 06733A40 PPC 09ACBEA0 PR_Malloc+00014 NEWLINE 067339D4 067339D0 68K 07A053BA NEWLINE 067339A8 PPC FFDDBD74 NQDStretch+006F4 NEWLINE 0673396C 68K 0673398E NEWLINE 06733910 0673390C 68K 0047C628 NEWLINE 06733908 06733900 PPC 098B2470 _MK_TraceMsg+00050 NEWLINE 067338C8 067338C0 PPC 002125C8 EmToNatEndMoveParams+00014 NEWLINE 067338A8 067338A0 PPC 09AD941C _MD_OpenDir+000D0 NEWLINE 06733878 68K 07A053BA NEWLINE 0673386C 68K 061321B6 NEWLINE 06733868 06733860 PPC 09ACBF8C PR_Free+00014 NEWLINE 06733838 06733830 PPC 00426E68 main+015C8 NEWLINE 0673382C 68K 0040CAAA NEWLINE 06733818 06733810 PPC FFDC80C0 NQDMapRgn+07384 NEWLINE Displaying memory from 0 NEWLINE 00000000 FFC1 0000 6F72 5379 6E63 4C66 006C 5268 •¡••orSyncLf•lRh NEWLINE 00000010 006C 526A 006C 526C 0068 9080 0068 9080 •lRj•lRl•hêÄ•hêÄ NEWLINE Closing log
772|Incorrect display of this page|I've just downloaded the latest Win32 build of Mozilla and was visiting NEWLINE the following link which rendered incorrectly. I thought sending this to NEWLINE you might help in debugging the problem. NEWLINE NEWLINE http://www.w3.org/RDF/Implementations/SiRPAC/
773|Chunked decoding can get stuck in STRIP_CRLF state|The state machine in net_ChunkedWrite in mozilla/network/cnvts/cvchunk.c assumes NEWLINE that obj->in_buf always contains a CRLF when in the STRIP_CRLF state. That's not NEWLINE always true so Mozilla gets stuck in an infinite loop. This seems to happen on NEWLINE certain cgi programs which spit out very small chunks. A patch is available.
774|error on exiting program|MacsBug 6.5.4a3c1, © Apple Computer, Inc. NEWLINE 1981-97 NEWLINE NEWLINE User break at 02200534 free+000A4 NEWLINE fastmem: attempt to dispose illegal block NEWLINE NEWLINE 9-Sep-98 10:09:04 AM (since boot = 55 NEWLINE minutes) NEWLINE Current application is “NavigatorDebug” NEWLINE Machine = 108 (PowerMac7200), System NEWLINE $0810, sysu = $01008000 NEWLINE ROM version $077D, $28F2, $0001 (ROMBase NEWLINE $FFC00000) NEWLINE VM is off NEWLINE NIL^ = $FFC10000 NEWLINE Stack space used = +39332890 NEWLINE Address 02200534 is in the Process Manager NEWLINE heap at 00951DA0 at free+000A4 NEWLINE The address is in a CFM fragment NEWLINE “MemoryAllocator” [non-write exec] NEWLINE It is 00001684 bytes from the start of the NEWLINE fragment NEWLINE It is 00001684 bytes into this heap block: NEWLINE Start Length Tag Mstr Ptr Lock Prg Type NEWLINE ID File Name NEWLINE • 021FEEB0 00003748+0C R 02583E28 L NEWLINE PowerPC 601 Registers NEWLINE CR0 CR1 CR2 CR3 CR4 CR5 NEWLINE CR6 CR7 NEWLINE PC = 02200534 CR 1000 1010 0000 NEWLINE 0000 0000 1000 1000 0000 NEWLINE LR = 02200534 <>=O XEVO NEWLINE CTR = FFD5AE38 NEWLINE MSR = 00000000 SOC Compare Count NEWLINE Int = 0 XER 001 00 00 NEWLINE NEWLINE R0 = FFD5AE38 R8 = 55534544 R16 = NEWLINE 02583E94 R24 = 00000000 NEWLINE SP = 02191410 R9 = 3A006564 R17 = NEWLINE 0224CB2C R25 = 01A35629 NEWLINE TOC = 0002ABF8 R10 = 75736564 R18 NEWLINE = 00000000 R26 = 00000000 NEWLINE R3 = 02201F52 R11 = 02201F07 R19 = NEWLINE 01B968D8 R27 = 01E72848 NEWLINE R4 = 55534544 R12 = 02200534 R20 NEWLINE = 01A2F417 R28 = 00000000 NEWLINE R5 = 46524545 R13 = 00000000 R21 NEWLINE = 021C5210 R29 = 01E7285C NEWLINE R6 = 00000000 R14 = 00000000 R22 NEWLINE = 022571C4 R30 = 01E7285C NEWLINE R7 = 55534544 R15 = 000000D0 R23 NEWLINE = 00000000 R31 = 01E72848 NEWLINE Disassembling PowerPC code from 0220050C NEWLINE free NEWLINE +0007C 0220050C cmplw r7,r8 NEWLINE \| 7C074040 NEWLINE +00080 02200510 bne free+00098 NEWLINE ; 0x02200528 \| 40820018 NEWLINE +00084 02200514 lwz NEWLINE r9,0x0000(r29) \| 813D0000 NEWLINE +00088 02200518 lis r10,0x7573 NEWLINE \| 3D407573 NEWLINE +0008C 0220051C addi NEWLINE r10,r10,0x6564 \| 394A6564 NEWLINE +00090 02200520 cmplw r9,r10 NEWLINE \| 7C095040 NEWLINE +00094 02200524 beq free+000AC NEWLINE ; 0x0220053C \| 41820018 NEWLINE +00098 02200528 lwz NEWLINE r11,0x00FC(RTOC) \| 816200FC NEWLINE +0009C 0220052C addi NEWLINE r3,r11,0x004B \| 386B004B NEWLINE +000A0 02200530 bl DebugStr NEWLINE ; 0x02201D30 \| 48001801 NEWLINE +000A4 02200534 *lwz NEWLINE RTOC,0x0014(SP) \| 80410014 NEWLINE +000A8 02200538 b free+00160 NEWLINE ; 0x022005F0 \| 480000B8 NEWLINE +000AC 0220053C lwz NEWLINE r12,0x000C(r31) \| 819F000C NEWLINE +000B0 02200540 cmplwi r12,0x0000 NEWLINE \| 280C0000 NEWLINE +000B4 02200544 beq free+000C8 NEWLINE ; 0x02200558 \| 41820014 NEWLINE +000B8 02200548 lwz NEWLINE r3,0x0010(r31) \| 807F0010 NEWLINE +000BC 0220054C lwz NEWLINE r4,0x000C(r31) \| 809F000C NEWLINE +000C0 02200550 stw NEWLINE r3,0x0010(r4) \| 90640010 NEWLINE +000C4 02200554 b free+000D4 NEWLINE ; 0x02200564 \| 48000010 NEWLINE +000C8 02200558 lwz NEWLINE r5,0x0010(r31) \| 80BF0010 NEWLINE Heap zones NEWLINE #1 Mod 9533K 00002800 to 00951D9F NEWLINE SysZone^ NEWLINE #2 Mod 6K 000092C0 to 0000AE5F NEWLINE ROM read-only zone NEWLINE #3 Mod 48K 0015DF90 to 00169F8F NEWLINE #4 Mod 28872K 00951DA0 to 0258401F NEWLINE Process Manager zone NEWLINE #5 Mod 11379K 01665F60 to NEWLINE 02182C0F “NavigatorDebug” ApplZone^ NEWLINE TheZone^ Target NEWLINE Zone NEWLINE #6 Mod 19K 021AF610 to 021B45CF NEWLINE #7 Mod 32K 021B45F0 to 021BC71F NEWLINE #8 Mod 24K 021BC740 to 021C2A9F NEWLINE #9 Mod 15K 021D3AE0 to 021D7ABF NEWLINE #10 Mod 16K 021D7AE0 to 021DBDFF NEWLINE #11 Mod 47K 021DBE20 to 021E7BCF NEWLINE #12 Mod 59K 021E7BF0 to 021F6ABF NEWLINE #13 Mod 4K 021F6AE0 to 021F7E1F NEWLINE #14 Mod 27K 021F7E40 to 021FED8F NEWLINE #15 Mod 950K 02482600 to 0257018F NEWLINE “Finder” NEWLINE Checking all heaps NEWLINE The System heap at 00002800 is ok NEWLINE The ROM read-only heap at 000092C0 is ok NEWLINE The heap at 0015DF90 is ok NEWLINE The Process Manager heap at 00951DA0 is ok NEWLINE The “NavigatorDebug” heap at 01665F60 is ok NEWLINE The heap at 021AF610 is ok NEWLINE The heap at 021B45F0 is ok NEWLINE The heap at 021BC740 is ok NEWLINE The heap at 021D3AE0 is ok NEWLINE The heap at 021D7AE0 is ok NEWLINE The heap at 021DBE20 is ok NEWLINE The heap at 021E7BF0 is ok NEWLINE The heap at 021F6AE0 is ok NEWLINE The heap at 021F7E40 is ok NEWLINE The “Finder” heap at 02482600 is ok NEWLINE Totaling the “NavigatorDebug” heap at NEWLINE 01665F60 NEWLINE Total Blocks Total of Block NEWLINE Sizes NEWLINE Free 009B #155 NEWLINE 000D2DF0 #863728 NEWLINE Nonrelocatable 009E #158 NEWLINE 009FCABC #10472124 NEWLINE Relocatable 0276 #630 NEWLINE 0004D3C0 #316352 NEWLINE Locked 0004 #4 00000300 NEWLINE #768 NEWLINE Purgeable and not locked 0119 #281 NEWLINE 00008470 #33904 NEWLINE Heap size 03AF #943 NEWLINE 00B1CC6C #11652204 NEWLINE The target heap is the System heap at 00002800 NEWLINE Totaling the System heap at 00002800 NEWLINE Total Blocks Total of Block NEWLINE Sizes NEWLINE Free 002D #45 00021880 NEWLINE #137344 NEWLINE Nonrelocatable 05F6 #1526 NEWLINE 0045289C #4532380 NEWLINE Relocatable 0677 #1655 NEWLINE 004DB440 #5092416 NEWLINE Locked 012D #301 NEWLINE 002E3880 #3029120 NEWLINE Purgeable and not locked 00F9 #249 NEWLINE 00156370 #1401712 NEWLINE Heap size 0C9A #3226 NEWLINE 0094F55C #9762140 NEWLINE The target heap is the “NavigatorDebug” heap at NEWLINE 01665F60 NEWLINE Displaying File Control Blocks NEWLINE fRef File Vol Type Fl Fork NEWLINE LEof NEWLINE 0002 System Macintosh … zsys dW NEWLINE rsrc #6823994 NEWLINE 0060 Macintosh … •••• dw data NEWLINE #1044480 NEWLINE 00BE Macintosh … •••• dw data NEWLINE #2088960 NEWLINE 1DC0 Open Tpt AppleTalk Li… Macintosh … libr NEWLINE dw rsrc #541750 NEWLINE 1E1E Open Transport Library Macintosh … libr NEWLINE dw rsrc #587348 NEWLINE 1E7C Shared Library Manage… Macintosh … NEWLINE INIT dw rsrc #211694 NEWLINE 1EDA Open Transport Library Macintosh … libr NEWLINE dw rsrc #587348 NEWLINE 1F38 Open Tpt AppleTalk Li… Macintosh … libr NEWLINE dw rsrc #541750 NEWLINE 1F96 Open Tpt Internet Lib… Macintosh … libr NEWLINE dw rsrc #485151 NEWLINE 1FF4 OpenTpt Remote Access Macintosh … libr NEWLINE dw rsrc #541432 NEWLINE 2052 OpenTpt Modem Macintosh … libr dw NEWLINE rsrc #85766 NEWLINE 20B0 OpenTpt Remote Access Macintosh … libr NEWLINE dw rsrc #541432 NEWLINE 210E Remote Access Log Macintosh … lzlg dW NEWLINE data #1427456 NEWLINE 216C Serial (Built-in) Macintosh … libr dw NEWLINE rsrc #63254 NEWLINE 21CA Open Transport Library Macintosh … libr NEWLINE dw rsrc #587348 NEWLINE 2228 OpenTpt Serial Arbitr… Macintosh … libr NEWLINE dw rsrc #7638 NEWLINE 2286 OpenTpt Serial Arbitr… Macintosh … libr NEWLINE dw rsrc #7638 NEWLINE 22E4 Appearance Extension Macintosh … INIT NEWLINE dw rsrc #633281 NEWLINE 2342 Appearance Extension Macintosh … INIT NEWLINE dw rsrc #633281 NEWLINE 23A0 Users & Groups Data F… Macintosh … NEWLINE BTFL dW data #69632 NEWLINE 23FE QuickTime™ Macintosh … INIT dw NEWLINE data #369778 NEWLINE 245C Mac OS Easy Open Macintosh … cdev dw NEWLINE rsrc #132666 NEWLINE 24BA Finder Macintosh … FNDR dW NEWLINE rsrc #502012 NEWLINE 2518 File Sharing Library Macintosh … shlb NEWLINE dw rsrc #3483 NEWLINE 2576 Finder Preferences Macintosh … pref NEWLINE dW rsrc #622 NEWLINE 25D4 Desktop DB Macintosh … BTFL dW NEWLINE data #452608 NEWLINE 2632 Desktop DF Macintosh … DTFL dW NEWLINE data #2985746 NEWLINE 2690 Open Tpt Internet Lib… Macintosh … libr NEWLINE dw rsrc #485151 NEWLINE 26EE NavigatorDebug Macintosh … APPL NEWLINE dW rsrc #917471 NEWLINE 274C Log.txt Macintosh … TEXT dW NEWLINE data #179308 NEWLINE 27AA OpenTpt Remote Access Macintosh … libr NEWLINE dw rsrc #541432 NEWLINE 2808 Mozilla Resources Macintosh … NSPL NEWLINE dW rsrc #512564 NEWLINE 2866 names.db Macintosh … BINA dW NEWLINE data #1152 NEWLINE 28C4 child.db Macintosh … BINA dW NEWLINE data #16384 NEWLINE 2922 lstr.db Macintosh … BINA dW NEWLINE data #4096 NEWLINE 2980 ilstr.db Macintosh … BINA dW NEWLINE data #65536 NEWLINE 29DE CCache log Macintosh … DBMC dW NEWLINE data #102400 NEWLINE 2A3C StdLog Macintosh … TEXT dW NEWLINE data #8337 NEWLINE #238 FCBs, #116 in use (including #78 fonts NEWLINE not listed), #122 free NEWLINE Displaying resource information: NEWLINE > Map $016660E8, flags $0000, file $26EE = NEWLINE NavigatorDebug NEWLINE Map $01B9C48C, flags $0000, file $2808 = NEWLINE Mozilla Resources NEWLINE + Map $0039ED30, flags $001E, file $22E4 NEWLINE = Appearance Extension NEWLINE + Map $00003320, flags $801E, file $0003 NEWLINE = •ROM resources that override System• NEWLINE S Map $00003374, flags $000D, file $0002 NEWLINE = System NEWLINE Map $0054BC58, flags $0000, file $245C = NEWLINE Mac OS Easy Open NEWLINE [Skipped $004E maps belonging to font files] NEWLINE Calling chain using A6/R1 links NEWLINE Back chain ISA Caller NEWLINE 00000000 PPC 018B1DFC NEWLINE XP_GetNavCenterContext+18858 NEWLINE 02192A30 PPC 016DB14C main+00154 NEWLINE 021929E0 PPC 016D56B0 NEWLINE CFrontApp::~CFrontApp()+0022C NEWLINE 02192960 PPC 018C9350 NEWLINE NET_CleanupCacheDirectory+000DC NEWLINE 02191740 PPC 018C9070 NEWLINE net_cache_recursive_file_finder+000CC NEWLINE 021916C0 PPC 015B28F8 PR_OpenDir+ NEWLINE 00038 NEWLINE 02191680 PPC 015C7D7C _MD_OpenDir+ NEWLINE 000D0 NEWLINE 021914A0 PPC 015BA8EC PR_Free+00014 NEWLINE Return addresses on the stack NEWLINE Stack Addr Frame Addr ISA Caller NEWLINE 021917E4 PPC 00184E14 NEWLINE NQDSetCCursor+03870 NEWLINE 0219176C 68K 0219158E NEWLINE 02191748 PPC 018C9350 NEWLINE NET_CleanupCacheDirectory+000DC NEWLINE 0219172C 68K 01B3A756 NEWLINE 0219170C 68K 01B968D6 NEWLINE 021916F8 68K 01840486 NEWLINE XP_HashListNew+000A6 NEWLINE 021916DC 68K 01B3A756 NEWLINE 021916C8 021916C0 PPC 018C9070 NEWLINE net_cache_recursive_file_finder+000CC NEWLINE 02191688 02191680 PPC 015B28F8 NEWLINE PR_OpenDir+00038 NEWLINE 02191648 02191640 PPC 015BA800 NEWLINE PR_Malloc+00014 NEWLINE 021915E8 68K 01B3A756 NEWLINE 021915A8 021915A0 PPC 01374FB0 NEWLINE LCommander::ProcessCommand(long, void*)+ NEWLINE 00058 NEWLINE 0219156C 68K 0219158E NEWLINE 02191508 02191500 PPC 018E9740 NEWLINE _MK_TraceMsg+00050 NEWLINE 021914C8 021914C0 PPC 001790C8 NEWLINE EmToNatEndMoveParams+00014 NEWLINE 021914A8 021914A0 PPC 015C7D7C NEWLINE _MD_OpenDir+000D0 NEWLINE 0219146C 68K 01B968D6 NEWLINE 02191468 02191460 PPC 015BA8EC NEWLINE PR_Free+00014 NEWLINE 02191438 68K 0137162A NEWLINE LArrayIterator::Current(void*)+00032 NEWLINE 02191418 02191410 PPC 01397150 NEWLINE LWindow::ObeyCommand(long, void*)+00038 NEWLINE Displaying memory from 0 NEWLINE 00000000 FFC1 0000 FFC1 0000 0067 NEWLINE 4E46 0067 4E48 •¡•••¡•••gNF•gNH NEWLINE 00000010 0067 4E4A 0067 4E4C FFC0 NEWLINE 31A8 FFC0 31AA •gNJ•gNL•¿1®•¿1™ NEWLINE Closing log
775|max-element-size not impacted by margins|(see mailnote to kipp and troy for minimized test case) NEWLINE An <img> tag with a non-0 HSPACE attribute reports incorrect size info. for NEWLINE example, given NS_UNCONSTRAINEDSIZE, the image on this page returns min=120, NEWLINE desired=150. Then, given 142 available width, the image reports back a desired NEWLINE width of 150.
776|MDEF|You are using a MDEf other that the standard MDEF. Either get a newer version or use the system MDEF. With Appearance, the system MDEF has most NEWLINE of the features you would need. (I think)
777|Pop-up message about saving passwords has no line-breaks|The first time a user gets the message about the "We can save your password..." NEWLINE feature, the thing is a mile long and runs way the hell off the screen. NEWLINE Just an annoyance.
778|[PP] Time code wrong?|At Mozilla startup, NetscapeDebug says it will expire October, 1994. Not a real NEWLINE problem, but disconcerting. If this is based on some system call, is all time NEWLINE related code in Mozilla for MacOS wrong?
779|Images not loading occasionally|I don't know if this only affects macs or not (or if this is the right category), NEWLINE but after 5 minutes of surfing or so, NavigatorDebug will stop loading images, NEWLINE displaying the broken image picture. This most frequently happens at NEWLINE slashdot.org.
780|Custom bar bookmark-properties dialog makes mouse unresponsive|That explanation pretty much explains it. Cursor moves, clicks anywhere - NEWLINE including the dialog - have no effect. Once I exit the dialog from keyboard,e NEWLINE verything returns to normal.
781|9-9-98, or 9-7-98 Mozilla Binaries do not load/run|Could not get the 9-9-98, or 9-7-98 Mozilla Binaries to load/run. Gives me an NEWLINE error that reads: NEWLINE NEWLINE "the MOZILLA.EXE file is linked to missing export MFCO42D.DLL:3300" NEWLINE NEWLINE I've tried both an older version of MFC42D.DLL, MFCO42D.DLL, MSVCIRTD.DLL, and NEWLINE MSVCRTD.DLL (from MSVC 4.x), and the newest versions (from Dev Studio 6.0). NEWLINE Neither allow me to load MOZILLA.EXE. NEWLINE NEWLINE I'd be happy to help test your code if I could only get it to run. NEWLINE NEWLINE -=STZ=-
782|New window - more toolbars added|When a new window is opened by any process, the toolbar state is not the same as NEWLINE the previous one was. Whether it is opening the browser for the first time or NEWLINE just opening a new window, the toolbars are reset to their default it seems.
783|Word Netscape in Sept 7th build.|Assorted dialog boxes still say "Netscape" or "Navigator". The lock file NEWLINE warning for multiple sessions is one of them. NEWLINE NEWLINE Another is when typing the URL to go to another page, the title of Mozilla NEWLINE before it loads the page is still "Netscape Navigator". NEWLINE NEWLINE Adding a #define for company/product name would prevent quite a few headaches NEWLINE later for people who may miss one or ten... NEWLINE NEWLINE As I said, it's trivial.
784|Mangled graphics in 9/7/98|Look at the right side of the "Features", "Quick Links" and "Books" tables. NEWLINE Normally they have a shadow graphic in it. Netscape 4.05 correctly renders the NEWLINE graphic while Mozilla 9/7/98 renders it as broken black fill. If I "shake" the NEWLINE vertical scroll bar up and down, I can change the pattern of the broken black NEWLINE fill. Looks pretty strange.
785|{sink} Handling frameset's with bodies|When dealing with a framset document mBody will be NULL, but DidBuildModel() NEWLINE is called anyway. Down at the bottom of the function there's some code that NEWLINE informs the document that content has been appended: NEWLINE NEWLINE mDocument->ContentAppended(mBody, mBodyChildCount); NEWLINE NEWLINE I added a NULL pointer check around the call (and an additional call that I NEWLINE added, which had been crashing), but we probably shouldn't be calling this NEWLINE function at all for framesets
786|Use of rowspan causes cursor to not change when over link, but link works|See the referenced URL; the first link is recognized as an active link and can NEWLINE be clicked on, but while the second is rendered properly it cannot be clicked on NEWLINE (mouseover of the link is apparently not recognized). This is with an NGLayout NEWLINE build as of 10 September 1998. NEWLINE NEWLINE The HTML to cause this bug (also at the referenced URL)is surprisingly simple: NEWLINE NEWLINE <TABLE> NEWLINE NEWLINE <TR> NEWLINE <TD ROWSPAN=2> NEWLINE NEWLINE <TABLE> NEWLINE <TR><TD><A HREF="01.html">01</A></TD></TR> NEWLINE <TR><TD><A HREF="02.html">02</A></TD></TR> NEWLINE </TABLE> NEWLINE NEWLINE </TD> NEWLINE </TR> NEWLINE NEWLINE <TR><TD>03</TD></TR> NEWLINE NEWLINE </TABLE> NEWLINE NEWLINE If either the ROWSPAN or the last table row is removed ("03") then the problem NEWLINE goes away. NEWLINE NEWLINE Note that this appears to be the root cause for why half the links on the NEWLINE mozilla.org page <http://www.mozilla.org/> do not work in the NGLayout viewer. NEWLINE (I extracted the above from the mozilla.org HTML source.)
787|Overlayed text in the Related Menu|text get overlayed when long discriptions cover the url names, causing garbled NEWLINE text. happens in the History and Related Menus NEWLINE NEWLINE this pic will show the problem in more detail NEWLINE NEWLINE http://www.risecom.net/~maasp/mes.jpg
788|Awful rendering, assert failure|Well, you wanted binary bug reports. :) NEWLINE NEWLINE 9/7/98 Linux build. NEWLINE NEWLINE http://ESPN.SportsZone.com/mlb/today/sco.html NEWLINE NEWLINE looks awful, as does NEWLINE NEWLINE http://ESPN.SportsZone.com/mlb/1998/980910/game_log/stlcin.html NEWLINE NEWLINE Clicking and dragging the "Bookmarks/History/Related" bar causes "assert: line NEWLINE 901, file BrowserFrame.cpp" also. NEWLINE NEWLINE Other assorted warnings I've had while browsing: NEWLINE NEWLINE malloc() warning: freeing free chunk at 8cd8d00 NEWLINE malloc() warning: freeing free chunk at 8cdce40 NEWLINE malloc() warning: freeing free chunk at 8cdec80 NEWLINE assert: line 84, file ToolboxItem.cpp NEWLINE assert: line 84, file ToolboxItem.cpp NEWLINE assert: line 84, file ToolboxItem.cpp NEWLINE malloc() warning: freeing free chunk at 8d6c920 NEWLINE malloc() warning: freeing free chunk at 8d89200 NEWLINE malloc() warning: freeing free chunk at 8d9f3a0 NEWLINE malloc() warning: freeing free chunk at 8da0d60 NEWLINE assert: line 84, file ToolboxItem.cpp NEWLINE assert: line 84, file ToolboxItem.cpp NEWLINE malloc() warning: freeing free chunk at 8ce49e0 NEWLINE malloc() warning: freeing free chunk at 8d836e0 NEWLINE malloc() warning: freeing free chunk at 8da01c0 NEWLINE malloc() warning: freeing free chunk at 8db8e80 NEWLINE NEWLINE Great work on Mozila.
789|about:blank leaks|Michael, NEWLINE NEWLINE Purify is reporting major leaks when bringing up the viewer with the NEWLINE about:blank URL and then quiting. NEWLINE NEWLINE [W] MLK: Memory leak of 80 bytes from 1 block allocated in NEWLINE nsGfxFactoryWin::CreateInstance(nsISupports *,nsID const&,void * *) NEWLINE [W] MLK: Memory leak of 256 bytes from 1 block allocated in NEWLINE DeviceContextImpl::DeviceContextImpl(void) NEWLINE [W] MLK: Memory leak of 24 bytes from 1 block allocated in NS_NewFontCache NEWLINE [W] MLK: Memory leak of 48 bytes from 1 block allocated in NEWLINE nsFontMetricsWin::new(UINT) NEWLINE [W] MLK: Memory leak of 28 bytes from 1 block allocated in NEWLINE nsFontMetricsWin::Init(nsFont const&,nsIDeviceContext *) NEWLINE [W] MLK: Memory leak of 34 bytes from 1 block allocated in NEWLINE nsString::EnsureCapacityFor(int) NEWLINE [W] MLK: Memory leak of 4 bytes from 1 block allocated in NEWLINE DeviceContextImpl::CreateFontAliasTable(void) NEWLINE [W] MLK: Memory leak of 32 bytes from 1 block allocated in PR_Malloc NEWLINE [W] MLK: Memory leak of 1024 bytes from 1 block allocated in PR_Malloc NEWLINE [W] MLK: Memory leak of 16 bytes from 1 block allocated in NEWLINE nsString::ToNewString(void)const NEWLINE [W] MLK: Memory leak of 34 bytes from 1 block allocated in NEWLINE nsString::EnsureCapacityFor(int) NEWLINE [W] MLK: Memory leak of 84 bytes from 1 block allocated in NEWLINE StringKey::Clone(void)const NEWLINE [W] MLK: Memory leak of 32 bytes from 2 blocks allocated in PR_Malloc NEWLINE [W] MLK: Memory leak of 16 bytes from 1 block allocated in NEWLINE nsString::ToNewString(void)const NEWLINE [W] MLK: Memory leak of 34 bytes from 1 block allocated in NEWLINE nsString::EnsureCapacityFor(int) NEWLINE [W] MLK: Memory leak of 84 bytes from 1 block allocated in NEWLINE StringKey::Clone(void)const NEWLINE [W] MLK: Memory leak of 32 bytes from 2 blocks allocated in PR_Malloc NEWLINE [W] MLK: Memory leak of 32 bytes from 1 block allocated in NEWLINE nsVoidArray::InsertElementAt(void *,int) NEWLINE [W] MLK: Memory leak of 80 bytes from 1 block allocated in NEWLINE nsGfxFactoryWin::CreateInstance(nsISupports *,nsID const&,void * *) NEWLINE [W] MLK: Memory leak of 256 bytes from 1 block allocated in NEWLINE DeviceContextImpl::DeviceContextImpl(void) NEWLINE [W] MLK: Memory leak of 24 bytes from 1 block allocated in NS_NewFontCache NEWLINE [W] MLK: Memory leak of 48 bytes from 1 block allocated in NEWLINE nsFontMetricsWin::new(UINT) NEWLINE [W] MLK: Memory leak of 28 bytes from 1 block allocated in NEWLINE nsFontMetricsWin::Init(nsFont const&,nsIDeviceContext *) NEWLINE [W] MLK: Memory leak of 34 bytes from 1 block allocated in NEWLINE nsString::EnsureCapacityFor(int) NEWLINE [W] MLK: Memory leak of 4 bytes from 1 block allocated in NEWLINE DeviceContextImpl::CreateFontAliasTable(void) NEWLINE [W] MLK: Memory leak of 32 bytes from 1 block allocated in PR_Malloc NEWLINE [W] MLK: Memory leak of 1024 bytes from 1 block allocated in PR_Malloc NEWLINE [W] MLK: Memory leak of 16 bytes from 1 block allocated in NEWLINE nsString::ToNewString(void)const NEWLINE [W] MLK: Memory leak of 34 bytes from 1 block allocated in NEWLINE nsString::EnsureCapacityFor(int) NEWLINE [W] MLK: Memory leak of 84 bytes from 1 block allocated in NEWLINE StringKey::Clone(void)const NEWLINE [W] MLK: Memory leak of 16 bytes from 1 block allocated in NEWLINE nsString::ToNewString(void)const NEWLINE [W] MLK: Memory leak of 34 bytes from 1 block allocated in NEWLINE nsString::EnsureCapacityFor(int) NEWLINE [W] MLK: Memory leak of 84 bytes from 1 block allocated in NEWLINE StringKey::Clone(void)const NEWLINE [W] MLK: Memory leak of 32 bytes from 1 block allocated in NEWLINE nsVoidArray::InsertElementAt(void *,int) NEWLINE [W] MLK: Memory leak of 80 bytes from 1 block allocated in NEWLINE nsGfxFactoryWin::CreateInstance(nsISupports *,nsID const&,void * *) NEWLINE [W] MLK: Memory leak of 256 bytes from 1 block allocated in NEWLINE DeviceContextImpl::DeviceContextImpl(void) NEWLINE [W] MLK: Memory leak of 24 bytes from 1 block allocated in NS_NewFontCache NEWLINE [W] MLK: Memory leak of 48 bytes from 1 block allocated in NEWLINE nsFontMetricsWin::new(UINT) NEWLINE [W] MLK: Memory leak of 28 bytes from 1 block allocated in NEWLINE nsFontMetricsWin::Init(nsFont const&,nsIDeviceContext *) NEWLINE [W] MLK: Memory leak of 34 bytes from 1 block allocated in NEWLINE nsString::EnsureCapacityFor(int) NEWLINE [W] MLK: Memory leak of 4 bytes from 1 block allocated in NEWLINE DeviceContextImpl::CreateFontAliasTable(void) NEWLINE [W] MLK: Memory leak of 32 bytes from 1 block allocated in PR_Malloc NEWLINE [W] MLK: Memory leak of 1024 bytes from 1 block allocated in PR_Malloc NEWLINE [W] MLK: Memory leak of 16 bytes from 1 block allocated in NEWLINE nsString::ToNewString(void)const NEWLINE [W] MLK: Memory leak of 34 bytes from 1 block allocated in NEWLINE nsString::EnsureCapacityFor(int) NEWLINE [W] MLK: Memory leak of 84 bytes from 1 block allocated in NEWLINE StringKey::Clone(void)const NEWLINE [W] MLK: Memory leak of 32 bytes from 2 blocks allocated in PR_Malloc NEWLINE [W] MLK: Memory leak of 16 bytes from 1 block allocated in NEWLINE nsString::ToNewString(void)const NEWLINE [W] MLK: Memory leak of 34 bytes from 1 block allocated in NEWLINE nsString::EnsureCapacityFor(int) NEWLINE [W] MLK: Memory leak of 84 bytes from 1 block allocated in NEWLINE StringKey::Clone(void)const NEWLINE [W] MLK: Memory leak of 32 bytes from 2 blocks allocated in PR_Malloc NEWLINE [W] MLK: Memory leak of 32 bytes from 1 block allocated in NEWLINE nsVoidArray::InsertElementAt(void *,int) NEWLINE [I] MPK: Potential memory leak of 136 bytes from 1 block allocated in NEWLINE nsWidgetFactory::CreateInstance(nsISupports *,nsID const&,void * *) NEWLINE [W] MLK: Memory leak of 80 bytes from 1 block allocated in NEWLINE nsGfxFactoryWin::CreateInstance(nsISupports *,nsID const&,void * *) NEWLINE [W] MLK: Memory leak of 256 bytes from 1 block allocated in NEWLINE DeviceContextImpl::DeviceContextImpl(void) NEWLINE [W] MLK: Memory leak of 24 bytes from 1 block allocated in NS_NewFontCache NEWLINE [W] MLK: Memory leak of 48 bytes from 1 block allocated in NEWLINE nsFontMetricsWin::new(UINT) NEWLINE [W] MLK: Memory leak of 28 bytes from 1 block allocated in NEWLINE nsFontMetricsWin::Init(nsFont const&,nsIDeviceContext *) NEWLINE [W] MLK: Memory leak of 34 bytes from 1 block allocated in NEWLINE nsString::EnsureCapacityFor(int) NEWLINE [W] MLK: Memory leak of 4 bytes from 1 block allocated in NEWLINE DeviceContextImpl::CreateFontAliasTable(void) NEWLINE [W] MLK: Memory leak of 32 bytes from 1 block allocated in PR_Malloc NEWLINE [W] MLK: Memory leak of 1024 bytes from 1 block allocated in PR_Malloc NEWLINE [W] MLK: Memory leak of 16 bytes from 1 block allocated in NEWLINE nsString::ToNewString(void)const NEWLINE [W] MLK: Memory leak of 34 bytes from 1 block allocated in NEWLINE nsString::EnsureCapacityFor(int) NEWLINE [W] MLK: Memory leak of 84 bytes from 1 block allocated in NEWLINE StringKey::Clone(void)const NEWLINE [W] MLK: Memory leak of 16 bytes from 1 block allocated in NEWLINE nsString::ToNewString(void)const NEWLINE [W] MLK: Memory leak of 34 bytes from 1 block allocated in NEWLINE nsString::EnsureCapacityFor(int) NEWLINE [W] MLK: Memory leak of 84 bytes from 1 block allocated in NEWLINE StringKey::Clone(void)const NEWLINE [W] MLK: Memory leak of 32 bytes from 1 block allocated in NEWLINE nsVoidArray::InsertElementAt(void *,int)
790|Mozilla crashes when looking history|When I try to sort history by site, Mozilla crashes. Last debug message is NEWLINE allways the same: NEWLINE assert: line 7188, file ht.c NEWLINE I can repeat this bug. My Mozilla build is Sep 11th 1998 nightly build.
791|ss:crash on startup|This is probably a known bug, so forgive me if you've seen this one a hundred NEWLINE times. NEWLINE NEWLINE Using the latest build of mozilla binary, it crashes on startup and a dialog box NEWLINE appears with this message: NEWLINE NEWLINE Assertion Failure NEWLINE y;\\nightly\\mozilla\\modules\\rdf\\src\\nlcstore.c NEWLINE Line 163.
792|Misaligned Text|In the 9/10/98 Linux Mozilla build, the file 'README' in the above FTP url NEWLINE listing is misaligned down one line. NEWLINE NEWLINE Also, clicking on the throbber still goes to http://www.netscape.com. Not sure NEWLINE if it is supposed to go to http://www.mozilla.org or not though. (I'd vote for NEWLINE http://slashdot.org personally.) :)
793|The Title bar is not gradianted (Win 98)|CVS build 09-11-98 NEWLINE NEWLINE Just to be a nit picker, the title bar in the Win32 version does not copy the NEWLINE image with the gradiant background from the default window provided. Instead NEWLINE most of the titlebar has the first color then immediately changes to the second NEWLINE color just before the minimize button.
794|Split alignment back into three buttons|This has driven me nuts ever since Communicator came out. Back with Netscape NEWLINE Gold the positioning was three seperate buttons. When Communicator came out NEWLINE they integrated it into one list button. I have wasted a long time just NEWLINE remembering to click on that stupid button to get the list of three other NEWLINE buttons so I can position my text. Could you please split it in to three NEWLINE seperate buttons on the button toolbar?
795|Improper layout of table|Mozilla inserts a *lot* of extra space on the right side of this table, for no NEWLINE apparent reason.
796|Table totally hosed|NGLayout really screws up on this table.
797|OBJECT element alternate content is not displayed|OBJECT element alternate content is not displayed. Instead, the rectangle the NEWLINE *primary* content would normally occupy is never redrawn.
798|BUILD: Link error with --enable-pthreads and mozilla-export|On my RedHat 5.1 x86 box (glibc-2.0.7-19) mozilla-export won't link when using NEWLINE pthreads. I have USE_PTHREADS set. moz-export links fine, and gnuz-export and NEWLINE gnuzilla-export both link w/ pthreads, but mozilla-export won't link and gives NEWLINE all the undef'ed refs to pthread_* in libnspr21.a
799|ilayer height intermittently set to 0|The slashdot page is composed of a banner ad centered at the top of the NEWLINE page followed by the rest of the page which is enclosed by a table. NEWLINE On the first visit to slashdot after starting mozilla, the page begins NEWLINE to load correctly. Text is laid out, and images fill in one by one. When NEWLINE the image for the banner ad at the top of the page loads, the table jumps NEWLINE up to the top of the page with the banner ad overlapping it. The area to NEWLINE the left of the banner ad becomes black. There's also some cruft along NEWLINE the right hand edge of the page. Pressing back and then forward results in NEWLINE the page laying out correctly. Pressing reload 2 or 3 times also works. It NEWLINE also seems like sometimes the banner image gets clipped resulting in a square NEWLINE image. NEWLINE NEWLINE It looks like the page author has gone to a lot of trouble to make something NEWLINE that loads with both navigator and msie as well as working around problems NEWLINE with the ilayer tag. NEWLINE NEWLINE It looks like the banner ad mozilla is displaying is part of a <NOLAYER> NEWLINE block. Isn't that a bug too? NEWLINE NEWLINE mozilla version: updated from cvs Saturday morning, Sept 12
800|seg fault on "add bookmark"|(code updated from cvs 9-12-98) NEWLINE NEWLINE adding a bookmark causes asserts and segfault here: NEWLINE assert: line 595, file cl_layer.c NEWLINE assert: line 595, file cl_layer.c NEWLINE NEWLINE Program received signal SIGSEGV, Segmentation fault. NEWLINE 0x8181a49 in XFE_DisplayTable (context=0x8ac9200, loc=1, ts=0x8c3a0a0) NEWLINE at lay.c:1399 NEWLINE 1399 long x = ts->x + ts->x_offset - CONTEXT_DATA (context)->document_x + NEWLINE NEWLINE --- NEWLINE I set a breakpoint to find out what's causing the assert. NEWLINE Some table cell's lo_image.layer is null. NEWLINE NEWLINE (gdb) break cl_layer.c:598 NEWLINE Breakpoint 1 at 0x84f669c: file cl_layer.c, line 598. NEWLINE (gdb) run NEWLINE The program being debugged has been started already. NEWLINE Start it from the beginning? (y or n) y NEWLINE Starting program: NEWLINE /usr/local/src/dawn/mine/mozilla/dist/Linux2.0.35_x86_DBG.OBJ/bin/. NEWLINE /mozilla-export NEWLINE assert: line 595, file cl_layer.c NEWLINE NEWLINE Breakpoint 1, CL_MoveLayer (layer=0x0, x=0, y=0) at cl_layer.c:598 NEWLINE 598 return; NEWLINE (gdb) where NEWLINE #0 CL_MoveLayer (layer=0x0, x=0, y=0) at cl_layer.c:598 NEWLINE #1 0x8379ba4 in lo_MoveElementLayers (eptr=0x8c3a180) at laycell.c:326 NEWLINE #2 0x8379d79 in lo_ShiftCell (cell=0x8c3a26c, dx=0, dy=0) at laycell.c:393 NEWLINE #3 0x83c001d in lo_EndTable (context=0x8ac9200, state=0x8ba1a00, NEWLINE table=0x8c09c00, relayout=0 '\\000') at laytable.c:7374 NEWLINE #4 0x83c3fb0 in lo_CloseTable (context=0x8ac9200, state=0x8ba1a00) NEWLINE at laytags.c:2096 NEWLINE #5 0x83c4199 in lo_process_table_tag (context=0x8ac9200, state=0x8ba1a00, NEWLINE tag=0x8bf71c0) at laytags.c:2198 NEWLINE #6 0x83cb9e9 in lo_LayoutTag (context=0x8ac9200, state=0x8ba1a00, NEWLINE tag=0x8bf71c0) at laytags.c:6138 NEWLINE #7 0x83a7915 in LO_ProcessTag (data_object=0x8b4dd00, tag=0x8bf71c0, status=0) NEWLINE at layout.c:4839 NEWLINE #8 0x83dedf9 in EDT_ProcessTag (data_object=0x8b4dd00, tag=0x8bf71c0, NEWLINE status=0) at editor.cpp:970 NEWLINE #9 0x8488e59 in PA_ParseBlock (stream=0x8bc86e0, NEWLINE block=0x8aad000 "<HTML>\\n<HEAD>\\n<TITLE>mozilla.org</TITLE>\\n</HEAD>\\n<BODY NEWLINE BGCOL NEWLINE OR=\\"#FFFFFF\\" TEXT=\\"#000000\\"\\nLINK=\\"#0000EE\\" VLINK=\\"#551A8B\\" NEWLINE ALINK=\\"#FF0000\\" NEWLINE \\nMARGINHEIGHT=0 MARGINWIDTH=0>\\n<MAP NAME=\\"banner\\">\\n<AREA SHAPE=REC"..., NEWLINE block_le NEWLINE n=8192) at pa_parse.c:1436 NEWLINE #10 0x848ffe0 in net_AutoCharCodeConv (stream=0x8b845a0, NEWLINE s=0x8aad000 "<HTML>\\n<HEAD>\\n<TITLE>mozilla.org</TITLE>\\n</HEAD>\\n<BODY NEWLINE BGCOLOR=\\ NEWLINE "#FFFFFF\\" TEXT=\\"#000000\\"\\nLINK=\\"#0000EE\\" VLINK=\\"#551A8B\\" NEWLINE ALINK=\\"#FF0000\\"\\nMA NEWLINE RGINHEIGHT=0 MARGINWIDTH=0>\\n<MAP NAME=\\"banner\\">\\n<AREA SHAPE=REC"..., l=8192) NEWLINE at n NEWLINE et_junk.c:192 NEWLINE #11 0x830b20e in net_read_file_chunk (cur_entry=0x8bb2000) at mkfile.c:904 NEWLINE #12 0x830bba9 in net_ProcessFile (cur_entry=0x8bb2000) at mkfile.c:1264 NEWLINE #13 0x82e9e03 in NET_ProcessNet (ready_fd=0x0, fd_type=1) at mkgeturl.c:3370 NEWLINE #14 0x82f5905 in net_process_net_timer_callback (closure=0x0) at mkselect.c:189 NEWLINE #15 0x81a0110 in fe_do_timeout (p=0x8b84540, id=0xbfffdfc8) at xfe.c:2965 NEWLINE #16 0x4009d8bd in XtAppProcessEvent () NEWLINE #17 0x818906a in fe_EventLoop () at mozilla.c:1143 NEWLINE #18 0x818c1f9 in main (argc=1, argv=0xbffff840) at mozilla.c:3460 NEWLINE NEWLINE (gdb) down NEWLINE #1 0x8379ba4 in lo_MoveElementLayers (eptr=0x8c3a180) at laycell.c:326 NEWLINE 326 CL_MoveLayer(eptr->lo_image.layer, NEWLINE (gdb) print eptr->lo_image NEWLINE $8 = {type = 4, x_offset = 0, ele_id = 3, x = 0, y = 0, y_offset = 0, NEWLINE width = 600, height = 58, line_height = 58, next = 0x8c3a224, prev = 0x0, NEWLINE edit_element = 0x0, edit_offset = 0, layer_id = 0, percent_width = 0, NEWLINE percent_height = 0, mocha_object = 0x0, layer = 0x0, valid_rect = {top = 0, NEWLINE left = 0, bottom = 0, right = 0}, image_req = 0x8b880c0, NEWLINE lowres_image_req = 0x0, is_icon = 0 '\\000', icon_number = 0, NEWLINE alt = 0x8c08da0, anchor_href = 0x0, image_url = 0x8c06580, NEWLINE lowres_image_url = 0x0, image_attr = 0x8bf7160, text_attr = 0x8ba0cc0, NEWLINE border_width = 0, border_vert_space = 0, border_horiz_space = 0, NEWLINE ele_attrmask = 0, image_status = 7, is_transparent = 0 '\\000', NEWLINE suppress_mode = LO_SUPPRESS_UNDEFINED, alt_len = 0, sel_start = -1, NEWLINE sel_end = -1, seq_num = 0, pending_mocha_event = 0 '\\000', next_image = 0x0} NEWLINE NEWLINE (gdb) print *cell NEWLINE $12 = {type = 9, x_offset = 0, ele_id = 5, x = -1, y = -1, y_offset = 0, NEWLINE width = 600, height = 58, line_height = 0, next = 0x0, prev = 0x0, NEWLINE edit_element = 0x0, edit_offset = 0, FE_Data = 0x0, cell_list = 0x8c3a180, NEWLINE cell_list_end = 0x8c3a224, cell_float_list = 0x0, backdrop = { NEWLINE bg_color = 0x8c08f80, url = 0x0, tile_mode = LO_TILE_BOTH}, NEWLINE border_width = 0, border_vert_space = 0, border_horiz_space = 0, NEWLINE ele_attrmask = 0, sel_start = -1, sel_end = -1, inter_cell_space = 0, NEWLINE cell_inflow_layer = 0x0, cell_bg_layer = 0x0, table_cell = 0x8c2f200, NEWLINE table_row = 0x8ba0380, table = 0x8c09c00, isCaption = 0 '\\000'}
801|Crashes|You should put in that GPF crash detector/reporter which is in Netscape 4.5b1 NEWLINE into Mozilla. Then you could set up a server side of the bug reporter which NEWLINE would delete multiple entries of the same crash.
802|{list} bullets positioned strangely around floats|The bulleted list items in column two are positioned too far to the left.
803|Floaters Aren't Resized Properly|Floaters with a percentage 'width' property aren't resized like they should be NEWLINE when the window is resized. NEWLINE NEWLINE This page demonstrates a two column display using a floated left DIV for column NEWLINE width. The width is 50%
804|Miami University page will not load.|The web page above causes the September 13th and 12th Windows builds to crash NEWLINE upon rendering.
805|the dom leaks memory. alot.|Here are some example lines from purify: NEWLINE NEWLINE [W] MLK: Memory leak of 24 bytes from 1 block allocated in JS_malloc NEWLINE [W] MLK: Memory leak of 24 bytes from 1 block allocated in js_alloc_atom
806|milestone branch reminder -> User agent implementation needs update|From: cyeh@netscape.com (Christopher Yeh) NEWLINE NEWLINE As you go about designing Raptor, would you please take a moment and decide NEWLINE a simple XP way of specifing the version and user agent string? I can't NEWLINE tell you how painful it is everytime we decide to bump either in the NEWLINE current 5.0 code base. You have to muck with each FE in a completely NEWLINE different way. NEWLINE NEWLINE If there was a single file that was used to config all platforms, that NEWLINE would be my dream. NEWLINE NEWLINE Thanks. NEWLINE NEWLINE Chris Yeh
807|'em' units don't work|In general, 'em' units don't work as length values. NEWLINE NEWLINE See issue #1 in the nglayout test document
808|padding and borders don't work on HTML and BODY elements|Setting padding and/or margins on HTML/BODY elements don't have NEWLINE any effect (with the exception of padding on BODY elements). NEWLINE NEWLINE For a test case, See issue #2 in the attached UR
809|block-level elements don't strech out to the right|For a test case, See issue #3 in the attached URL
810|borders around elements eat their way into text|For a test case, See issue #4 in the attached URL
811|default margin on DIV elements is non-zero|This is not a bug since User Agents are free to use any default NEWLINE style sheet. I do think margins values should be zero in the NEWLINE browser's default style sheet, though. NEWLINE NEWLINE For a test case, See issue #6 in the attached URL
812|no left padding|The left (or perhaps, horizontal) padding on block-level NEWLINE elements seem to be ignored. NEWLINE NEWLINE For a test case, See issue #4 in the attached URL
813|element widths|When elements are set to e.g. 50%, the textual content is constrained, NEWLINE but the background rectangle is unconstrained. NEWLINE NEWLINE For a test case, See issue #7 in the attached URL
814|overlapping floating elements - no!|Floating elements should not overlap when space is to tight. NEWLINE NEWLINE This is described in [1]: NEWLINE NEWLINE If the current box is left-floating, and there are any left NEWLINE floating boxes generated by elements earlier in the source document, NEWLINE then for each such earlier box, either the left outer edge of the NEWLINE current box must be to the right of the right outer edge of the NEWLINE earlier box, or its top must be lower than the bottom of the NEWLINE earlier box. Analogous rules hold for right-floating boxes. NEWLINE NEWLINE [1] http://www.w3.org/TR/REC-CSS2/visuren.html#floats NEWLINE NEWLINE I know this is different from how NS traditionally lays out NEWLINE floating images. I think the time has come to change that behavior NEWLINE -- or, alternatively -- let floating text elements behave NEWLINE differently from floating images.
815|We shouldn't process content inside <NOSCRIPT> tags|Currently we're processing content inside NOSCRIPT tags. This problem is most NEWLINE visible on various Netcenter pages, the first of which being the sign in page. NEWLINE The sign in page uses a meta-redirect inside a NOSCRIPT tag to avoid non script NEWLINE browsers. We need to get this fixed if we want to be able to test Netcenter in NEWLINE NGLayout.
816|HTML tags leaking into edit field|Go to the sign in page for Netcenter. One way to get there is via NEWLINE my.netscape.com, then click on 'I am already a My Netscape User. Find My NEWLINE Page!'. When the page comes up you'll note at '></TD>' inside the first text NEWLINE field. Unfortunately, the page will then immediately flash away due to bug NEWLINE #815, parsing content inside NOSCRIPT tags.
817|XP timebomb code needed for betas|Something else you probably haven't thought of is timebomb code. NEWLINE NEWLINE At some magical point, you'll be releasing betas, and you'll need a way to NEWLINE disable the engine after a particular date and time, or relative to the first NEWLINE launching of Raptor. NEWLINE NEWLINE This should be implemented in an XP way.
818|Internet Explorer can't handle server-push pages.|When querying the bugzilla database the cgi spits out lots of "-- NEWLINE ThisRandomString". Most of the time this is just annoying, but sometimes (if NEWLINE you try and get a list of all the bugs in NGLayout for instance) it puts it NEWLINE infront of the MIME header confusing the browser somewhat.
819|crashes on launch|I was working in netscape (Mozilla) when all of a sudden, it crashed, NEWLINE i dont know how much help this information will be, im new at debugging NEWLINE software: NEWLINE NEWLINE First this appeared: NEWLINE *Microsoft Visual C++ Debug Library* NEWLINE NEWLINE Debug Assertion Failed! NEWLINE Program: C:\\program files\\netscape\\program\\mozilla.exe NEWLINE File: y:\\nightly\\mozilla\\modules\\rdf\\src\\remstore.c NEWLINE Line: 189 NEWLINE NEWLINE then, after i pressed ignore a few times, this appeared: NEWLINE NEWLINE MOZILLA caused an invalid page fault in NEWLINE module MSVCRTD.DLL at 023f:10209748. NEWLINE Registers: NEWLINE EAX=5f4b3428 CS=023f EIP=10209748 EFLGS=00010202 NEWLINE EBX=00c4fb8c SS=0247 ESP=00b50000 EBP=00b50018 NEWLINE ECX=81628cec DS=0247 ESI=81628cec FS=35df NEWLINE EDX=81628cf4 ES=0247 EDI=00b501e0 GS=0000 NEWLINE Bytes at CS:EIP: NEWLINE 57 8b 45 08 8b 40 10 89 45 f8 8b 45 08 8b 40 0c NEWLINE Stack dump: NEWLINE 81628cec 00c4fb8c 00000000 00000000 1020d922 00000000 00b5004c 10209672 5f4b3428 NEWLINE 00000000 00000000 00b50040 00b50048 00b501e0 81628cec 00c4fb8c NEWLINE NEWLINE good luck!! NEWLINE NEWLINE -Joshua Pollack
820|Mozilla crashing bug in jpeg handling dll|Builds 1998-09-12, 1998-09-13, and 1998-09-14 will crash if you go to one of the NEWLINE urls below. The crash infomation was this: NEWLINE NEWLINE MOZILLA caused an invalid page fault in NEWLINE module JPEG3250.DLL at 015f:009ee701. NEWLINE Registers: NEWLINE EAX=00e70c50 CS=015f EIP=009ee701 EFLGS=00210282 NEWLINE EBX=00e6fc8c SS=0167 ESP=00c4f338 EBP=00c4f3bc NEWLINE ECX=00e6fc0c DS=0167 ESI=00e70c48 FS=19bf NEWLINE EDX=00e74ff8 ES=0167 EDI=00e74fb8 GS=0000 NEWLINE Bytes at CS:EIP: NEWLINE 0f 7f 6a 08 0f 7f f3 0f 60 f6 0f 60 f9 0f 60 f1 NEWLINE Stack dump: NEWLINE 00000000 81664020 00b40000 00cb7d54 00e6ea58 00000001 00800080 00800080 00000000 NEWLINE 00000000 00e6ea10 00e6e9d0 00e6ea50 00cb7124 00000000 00000000 NEWLINE NEWLINE and NEWLINE NEWLINE MOZILLA caused an invalid page fault in NEWLINE module MSVCRTD.DLL at 015f:102117d5. NEWLINE Registers: NEWLINE EAX=00e6eb60 CS=015f EIP=102117d5 EFLGS=00010206 NEWLINE EBX=81672d74 SS=0167 ESP=0213fb90 EBP=0213fba4 NEWLINE ECX=55ffff21 DS=0167 ESI=5f400000 FS=4e1f NEWLINE EDX=55ffff21 ES=0167 EDI=00000000 GS=0000 NEWLINE Bytes at CS:EIP: NEWLINE 8b 42 14 25 ff ff 00 00 85 c0 7c 66 8b 4d f8 8b NEWLINE Stack dump: NEWLINE 00000000 5f400000 81672d74 55ffff21 00000005 0213fbf0 10211dc2 0213fbbc 00000000 NEWLINE 5f400000 81672d74 00e6eb60 00000000 00000000 00000000 00000000 NEWLINE NEWLINE http://developer.netscape.com/images/pixel3.jpg is a 1x1x24bit jpg. Which is in NEWLINE the page for reasons I can only guess. Netscape Communicator 4.5p2 has no NEWLINE problem with it. NEWLINE NEWLINE http://developer.netscape.com/images/pixel3.jpg NEWLINE http://developer.netscape.com/source/intel.html
821|Tried to rename bookmark; crashed; would NOT restart|This was in version 980915. NEWLINE I went to a URL and dragged it into my bookmark list. I then pulled up NEWLINE its properties so I could rename it. When I hit OK (or whatever) Dr. Watson NEWLINE immediately reported a fatal application fault. Mozilla folded up and went NEWLINE away. When I try to restart it, I get this error: NEWLINE NEWLINE Debug Assertion Failed NEWLINE ... modules\\rdf\\src\\ht.c, line 2531
822|assert|open http://www.infoseek.com NEWLINE the assert happens immediately. NEWLINE NEWLINE code: NEWLINE NS_ASSERTION(!mBandList.IsEmpty(), "no bands"); NEWLINE NEWLINE stack: NEWLINE NTDLL! 77f76148() NEWLINE nsDebug::Assertion(char * 0x006d5294, char * 0x006d527c, char * 0x006d524c, int NEWLINE 250) line 95 + 13 bytes NEWLINE nsSpaceManager::GetBandData(int -1751486202, const nsSize & {width=1073741824 NEWLINE height=1073741824}, nsBandData & {...}) line 250 + 45 bytes NEWLINE nsCSSBlockReflowState::GetAvailableSpace() line 942 NEWLINE nsCSSBlockFrame::ReflowBlockFrame(nsCSSBlockReflowState & {...}, LineData * NEWLINE 0x0140c950, nsIFrame * 0x0140c8b0, unsigned int & 0) line 2735 NEWLINE nsCSSBlockFrame::ReflowLine(nsCSSBlockReflowState & {...}, LineData * NEWLINE 0x0140c950, unsigned int & 0) line 2373 + 24 bytes NEWLINE nsCSSBlockFrame::ReflowLinesAt(nsCSSBlockReflowState & {...}, LineData * NEWLINE 0x0140c950) line 2245 + 20 bytes NEWLINE nsCSSBlockFrame::ResizeReflow(nsCSSBlockReflowState & {...}) line 2230 + 19 NEWLINE bytes NEWLINE nsCSSBlockFrame::InitialReflow(nsCSSBlockReflowState & {...}) line 1851 + 12 NEWLINE bytes NEWLINE nsCSSBlockFrame::ReflowAround(nsCSSBlockFrame * const 0x0140cd40, nsIPresContext NEWLINE & {...}, nsISpaceManager * 0x0140cb70, nsReflowMetrics & {width=0 height=0}, NEWLINE const nsReflowState & {...}, nsRect & {x=24521740 y=0 width=0 height=0}, NEWLINE unsigned int & 0) line 1 NEWLINE nsBodyFrame::Reflow(nsBodyFrame * const 0x01409570, nsIPresContext & {...}, NEWLINE nsReflowMetrics & {width=0 height=0}, const nsReflowState & {...}, unsigned int NEWLINE & 0) line 264 NEWLINE nsContainerFrame::ReflowChild(nsIFrame * 0x01409570, nsIPresContext * NEWLINE 0x01558270, nsReflowMetrics & {width=0 height=0}, const nsReflowState & {...}) NEWLINE line 515 NEWLINE nsTableCellFrame::Reflow(nsTableCellFrame * const 0x01409450, nsIPresContext & NEWLINE {...}, nsReflowMetrics & {width=0 height=0}, const nsReflowState & {...}, NEWLINE unsigned int & 0) line 340 + 30 bytes NEWLINE nsContainerFrame::ReflowChild(nsIFrame * 0x01409450, nsIPresContext * NEWLINE 0x01558270, nsReflowMetrics & {width=0 height=0}, const nsReflowState & {...}) NEWLINE line 515 NEWLINE nsTableRowFrame::InitialReflow(nsIPresContext & {...}, RowReflowState & {...}, NEWLINE nsReflowMetrics & {width=0 height=0}) line 625 + 30 bytes NEWLINE nsTableRowFrame::Reflow(nsTableRowFrame * const 0x01408e00, nsIPresContext & NEWLINE {...}, nsReflowMetrics & {width=0 height=0}, const nsReflowState & {...}, NEWLINE unsigned int & 0) line 956 + 20 bytes NEWLINE nsContainerFrame::ReflowChild(nsIFrame * 0x01408e00, nsIPresContext * NEWLINE 0x01558270, nsReflowMetrics & {width=0 height=0}, const nsReflowState & {...}) NEWLINE line 515 NEWLINE nsTableRowGroupFrame::ReflowUnmappedChildren(nsIPresContext * 0x01558270, NEWLINE RowGroupReflowState & {...}, nsSize * 0x0012e288 {width=15 height=30}) line 800 NEWLINE + 30 bytes NEWLINE nsTableRowGroupFrame::Reflow(nsTableRowGroupFrame * const 0x013f1c10, NEWLINE nsIPresContext & {...}, nsReflowMetrics & {width=0 height=0}, const NEWLINE nsReflowState & {...}, unsigned int & 0) line 1131 + 23 bytes NEWLINE nsContainerFrame::ReflowChild(nsIFrame * 0x013f1c10, nsIPresContext * NEWLINE 0x01558270, nsReflowMetrics & {width=0 height=0}, const nsReflowState & {...}) NEWLINE line 515 NEWLINE nsTableFrame::ResizeReflowPass1(nsIPresContext * 0x01558270, nsReflowMetrics & NEWLINE {width=0 height=0}, const nsReflowState & {...}, unsigned int & 0) line 1576 + NEWLINE 33 bytes NEWLINE nsTableFrame::Reflow(nsTableFrame * const 0x013f1640, nsIPresContext & {...}, NEWLINE nsReflowMetrics & {width=0 height=0}, const nsReflowState & {...}, unsigned int NEWLINE & 0) line 1409 + 30 bytes NEWLINE nsContainerFrame::ReflowChild(nsIFrame * 0x013f1640, nsIPresContext * NEWLINE 0x01558270, nsReflowMetrics & {width=0 height=0}, const nsReflowState & {...}) NEWLINE line 515 NEWLINE nsTableOuterFrame::Reflow(nsTableOuterFrame * const 0x013b9210, nsIPresContext & NEWLINE {...}, nsReflowMetrics & {width=0 height=0}, const nsReflowState & {...}, NEWLINE unsigned int & 0) line 451 + 27 bytes NEWLINE nsCSSBlockFrame::ReflowBlockFrame(nsCSSBlockReflowState & {...}, LineData * NEWLINE 0x013b93d0, nsIFrame * 0x013b9210, unsigned int & 16640) line 2611 + 37 bytes NEWLINE nsCSSBlockFrame::ReflowLine(nsCSSBlockReflowState & {...}, LineData * NEWLINE 0x013b93d0, unsigned int & 16640) line 2373 + 24 bytes NEWLINE nsCSSBlockFrame::ReflowLinesAt(nsCSSBlockReflowState & {...}, LineData * NEWLINE 0x013b93d0) line 2245 + 20 bytes NEWLINE nsCSSBlockFrame::ResizeReflow(nsCSSBlockReflowState & {...}) line 2230 + 19 NEWLINE bytes NEWLINE nsCSSBlockFrame::InitialReflow(nsCSSBlockReflowState & {...}) line 1851 + 12 NEWLINE bytes NEWLINE nsCSSBlockFrame::ReflowAround(nsCSSBlockFrame * const 0x013b8cf0, nsIPresContext NEWLINE & {...}, nsISpaceManager * 0x012b1a70, nsReflowMetrics & {width=0 height=0}, NEWLINE const nsReflowState & {...}, nsRect & ...) line 1368 + 18 bytes NEWLINE nsCSSBlockFrame::ReflowBlockFrame(nsCSSBlockReflowState & {...}, LineData * NEWLINE 0x013bcde0, nsIFrame * 0x013b8ca0, unsigned int & 16640) line 2599 + 47 bytes NEWLINE nsCSSBlockFrame::ReflowLine(nsCSSBlockReflowState & {...}, LineData * NEWLINE 0x013bcde0, unsigned int & 16640) line 2373 + 24 bytes NEWLINE nsCSSBlockFrame::ReflowLinesAt(nsCSSBlockReflowState & {...}, LineData * NEWLINE 0x013bcde0) line 2245 + 20 bytes NEWLINE nsCSSBlockFrame::FrameAppendedReflow(nsCSSBlockReflowState & {...}) line 1941 + NEWLINE 16 bytes NEWLINE nsCSSBlockFrame::ReflowAround(nsCSSBlockFrame * const 0x012b1e20, nsIPresContext NEWLINE & {...}, nsISpaceManager * 0x012b1a70, nsReflowMetrics & {width=0 height=0}, NEWLINE const nsReflowState & {...}, nsRect & {x=268439936 y=270584775 width=270912216 NEWLINE height=0}, ...) li NEWLINE nsBodyFrame::Reflow(nsBodyFrame * const 0x012b19e0, nsIPresContext & {...}, NEWLINE nsReflowMetrics & {width=0 height=0}, const nsReflowState & {...}, unsigned int NEWLINE & 0) line 264 NEWLINE nsContainerFrame::ReflowChild(nsIFrame * 0x012b19e0, nsIPresContext * NEWLINE 0x01558270, nsReflowMetrics & {width=0 height=0}, const nsReflowState & {...}) NEWLINE line 515 NEWLINE RootContentFrame::Reflow(RootContentFrame * const 0x012b3500, nsIPresContext & NEWLINE {...}, nsReflowMetrics & {width=0 height=0}, const nsReflowState & {...}, NEWLINE unsigned int & 0) line 353 + 27 bytes NEWLINE nsContainerFrame::ReflowChild(nsIFrame * 0x012b3500, nsIPresContext * NEWLINE 0x01558270, nsReflowMetrics & {width=0 height=0}, const nsReflowState & {...}) NEWLINE line 515 NEWLINE RootFrame::Reflow(RootFrame * const 0x012b1f60, nsIPresContext & {...}, NEWLINE nsReflowMetrics & {width=0 height=0}, const nsReflowState & {...}, unsigned int NEWLINE & 0) line 156 + 27 bytes NEWLINE nsHTMLReflowCommand::Dispatch(nsHTMLReflowCommand * const 0x013bc5a0, NEWLINE nsIPresContext & {...}, nsReflowMetrics & {width=0 height=0}, const nsSize & NEWLINE {width=9120 height=4410}) line 133 NEWLINE PresShell::ProcessReflowCommands() line 662 NEWLINE PresShell::ExitReflowLock(PresShell * const 0x012ab690) line 415 NEWLINE PresShell::ContentAppended(PresShell * const 0x012ab698, nsIDocument * NEWLINE 0x015423c0, nsIContent * 0x012b0f2c, int 0) line 736 NEWLINE nsDocument::ContentAppended(nsIContent * 0x012b0f2c, int 0) line 525 NEWLINE HTMLContentSink::WillInterrupt(HTMLContentSink * const 0x015445e0) line 1398 NEWLINE CNavDTD::WillInterruptParse(CNavDTD * const 0x012ac400) line 3090 + 18 bytes NEWLINE nsParser::ResumeParse() line 567 NEWLINE nsParser::OnDataAvailable(nsParser * const 0x01544c94, nsIURL * 0x01412150, NEWLINE nsIInputStream * 0x015426e0, int 4542) line 774 + 15 bytes NEWLINE nsDocumentBindInfo::OnDataAvailable(nsDocumentBindInfo * const 0x014121b0, NEWLINE nsIURL * 0x01412150, nsIInputStream * 0x015426e0, int 4542) line 1058 + 30 bytes NEWLINE stub_put_block(_NET_StreamClass * 0x0151d800, char * 0x011e12f0, long 4542) line NEWLINE 574 + 36 bytes NEWLINE net_MemCacheWrite(_NET_StreamClass * 0x012ac780, char * 0x011e12f0, long 4542) NEWLINE line 665 + 24 bytes NEWLINE net_pull_http_data(_ActiveEntry * 0x014249a0) line 3015 + 29 bytes NEWLINE net_ProcessHTTP(_ActiveEntry * 0x014249a0) line 3400 + 9 bytes NEWLINE NET_ProcessNet(PRFileDesc * 0x012cbc80, int 2) line 3370 + 13 bytes NEWLINE NET_PollSockets() line 180 + 18 bytes NEWLINE nsNetlibService::NetPollSocketsCallback(nsITimer * 0x01333510, void * NEWLINE 0x011f1e10) line 565 NEWLINE TimerImpl::Fire(unsigned long 30525823) line 308 + 17 bytes NEWLINE TimerImpl::ProcessTimeouts(unsigned long 30525823) line 187 NEWLINE FireTimeout(void * 0x00000000, unsigned int 275, unsigned int 4960, unsigned NEWLINE long 30525823) line 101 + 9 bytes NEWLINE USER32! 77e7128c() NEWLINE main(int 6, char * * 0x01214540) line 96 NEWLINE mainCRTStartup() line 338 + 17 bytes NEWLINE KERNEL32! 77f1b304()
823|LISTING and PRE tag's need to eat the first newline|The very first newline following the tag needs to be eaten. NEWLINE NEWLINE I'm not sure who should do this, the parser or the content sink.
824|password filter is too picky|There can be no good reason for excluding the quote character from a password. NEWLINE we should allow all of !~!@#$%^&*()_+{}\|"<>?`-=[]\\;',./
825|Exceed hurts Mozilla under >256 colors|When using Exceed (a popular X server) on NT, with the monitor set to more than NEWLINE 256 colors, Mozilla does a redraw every time the user context changes back to NEWLINE Mozilla from Exceed. This can be very annoying.
826|Crashed when doing a query|On query screen, selected new, assigned, reopen, resolved and verified. NEWLINE did a search under mozilla NEWLINE result: output a lot of text NEWLINE NEWLINE when I did a search under bugzilla, the search resulted in files being shown
827|Crashed after adding comments to a bug.|I tried to add "additional comments" to bug 821, and it showed me the "Changes NEWLINE Submitted" page with this error: NEWLINE NEWLINE Semicolon seems to be missing at NEWLINE bug_form.tcl line 41 (#1) (W) A nearby syntax error was probably caused by a NEWLINE missing semicolon, or possibly some NEWLINE other missing operator, such as a comma. [Wed Sep 16 05:51:00 1998] NEWLINE process_bug.cgi: Semicolon seems to be NEWLINE missing at bug_form.tcl line 41. NEWLINE NEWLINE When I clicked the link "Back to Bug#821", the comments I entered were not NEWLINE there.
828|Form upload: default to "All Files"|Instead of having the form upload default to HTML Files, make it default to All NEWLINE Files... or at least remember the last mask you specified.
829|"Who Points To Me?" crashes on Slashdot|What's Related/Who Points To Me? crashes on http://slashdot.org with a failed NEWLINE assertion at modules\\rdf\\src\\remstore.c line 82.
830|Crash in mozilla/layout/html/base/src/nsBlockFrame.cpp|Click on the stocks link under "Daily Essentials..." NEWLINE NEWLINE It looks like aParentFrame->mLines can be NULL in some situations... NEWLINE Below is the code fragment where the crash happened... NEWLINE NEWLINE nsBlockFrame.cpp (rev 3.73 line 3319) NEWLINE --------------------------------------- NEWLINE // Insert/append the frame into flows line list at the right spot NEWLINE LineData* newLine; NEWLINE LineData* line = aParentFrame->mLines; NEWLINE if (nsnull == aPrevSibling) { NEWLINE // Insert new frame into the sibling list NEWLINE ==> aNewFrame->SetNextSibling(line->mFirstChild); NEWLINE NEWLINE NEWLINE Here's the Stack: NEWLINE ----------------- NEWLINE NEWLINE nsBlockFrame::InsertNewFrame(nsBlockFrame * 0x0118e1c0, nsIFrame * 0x011f1800, NEWLINE nsIFrame * 0x00000000) line 3324 + 3 bytes NEWLINE nsBlockFrame::ProcessInitialReflow(nsIPresContext * 0x01239ec0) line 1512 NEWLINE nsBlockFrame::InitialReflow(nsBlockReflowState & {...}) line 1825 + 15 bytes NEWLINE nsBlockFrame::ReflowAround(nsBlockFrame * const 0x0118e210, nsIPresContext & NEWLINE {...}, nsISpaceManager * 0x011a7910, nsReflowMetrics & {...}, const NEWLINE nsReflowState & {...}, nsRect & {...}, unsigned int & 1227036) line 1368 + 18 NEWLINE bytes NEWLINE nsBlockFrame::ReflowBlockFrame(nsBlockReflowState & {...}, LineData * NEWLINE 0x0118e2a0, nsIFrame * 0x0118e1c0, unsigned int & 16640) line 2600 + 47 bytes NEWLINE nsBlockFrame::ReflowLine(nsBlockReflowState & {...}, LineData * 0x0118e2a0, NEWLINE unsigned int & 16640) line 2374 + 24 bytes NEWLINE nsBlockFrame::ReflowLinesAt(nsBlockReflowState & {...}, LineData * 0x0118e2a0) NEWLINE line 2246 + 20 bytes NEWLINE nsBlockFrame::ResizeReflow(nsBlockReflowState & {...}) line 2232 NEWLINE nsBlockFrame::InitialReflow(nsBlockReflowState & {...}) line 1853 NEWLINE nsBlockFrame::ReflowAround(nsBlockFrame * const 0x0118eb20, nsIPresContext & NEWLINE {...}, nsISpaceManager * 0x011a7910, nsReflowMetrics & {...}, const NEWLINE nsReflowState & {...}, nsRect & {...}, unsigned int & 4969613) line 1368 + 18 NEWLINE bytes NEWLINE nsInlineLayout::ReflowFrame(nsIFrame * 0x0118ead0, nsReflowMetrics & {...}, NEWLINE const nsReflowState & {...}, int & 0) line 288 NEWLINE nsInlineLayout::ReflowAndPlaceFrame(nsIFrame * 0x0118ead0) line 175 + 30 bytes NEWLINE nsInlineFrame::ReflowMapped(nsInlineReflowState & {...}, unsigned int & 0) line NEWLINE 618 + 15 bytes NEWLINE nsInlineFrame::InitialReflow(nsInlineReflowState & {...}) line 465 + 16 bytes NEWLINE nsInlineFrame::InlineReflow(nsInlineFrame * const 0x011e0540, nsLineLayout & NEWLINE {...}, nsReflowMetrics & {...}, const nsReflowState & {...}) line 320 + 18 bytes NEWLINE nsInlineLayout::ReflowFrame(nsIFrame * 0x011e04f0, nsReflowMetrics & {...}, NEWLINE const nsReflowState & {...}, int & 5728269) line 299 + 26 bytes NEWLINE nsInlineLayout::ReflowAndPlaceFrame(nsIFrame * 0x011e04f0) line 175 + 30 bytes NEWLINE nsBlockFrame::ReflowInlineFrame(nsBlockReflowState & {...}, LineData * NEWLINE 0x011e4170, nsIFrame * 0x011e04f0, unsigned int & 17153) line 2809 + 18 bytes NEWLINE nsBlockFrame::ReflowLine(nsBlockReflowState & {...}, LineData * 0x011e4170, NEWLINE unsigned int & 17153) line 2379 + 24 bytes NEWLINE nsBlockFrame::ReflowLinesAt(nsBlockReflowState & {...}, LineData * 0x011e4170) NEWLINE line 2246 + 20 bytes NEWLINE nsBlockFrame::ResizeReflow(nsBlockReflowState & {...}) line 2232 NEWLINE nsBlockFrame::InitialReflow(nsBlockReflowState & {...}) line 1853 NEWLINE nsBlockFrame::ReflowAround(nsBlockFrame * const 0x011a7a10, nsIPresContext & NEWLINE {...}, nsISpaceManager * 0x011a7910, nsReflowMetrics & {...}, const NEWLINE nsReflowState & {...}, nsRect & {...}, unsigned int & 0) line 1368 + 18 bytes NEWLINE nsBodyFrame::Reflow(nsBodyFrame * const 0x011a7880, nsIPresContext & {...}, NEWLINE nsReflowMetrics & {...}, const nsReflowState & {...}, unsigned int & 0) line 264 NEWLINE nsContainerFrame::ReflowChild(nsIFrame * 0x011a7880, nsIPresContext * NEWLINE 0x01239ec0, nsReflowMetrics & {...}, const nsReflowState & {...}) line 515 NEWLINE nsTableCellFrame::Reflow(nsTableCellFrame * const 0x0118d5f0, nsIPresContext & NEWLINE {...}, nsReflowMetrics & {...}, const nsReflowState & {...}, unsigned int & 0) NEWLINE line 363 + 30 bytes NEWLINE nsContainerFrame::ReflowChild(nsIFrame * 0x0118d5f0, nsIPresContext * NEWLINE 0x01239ec0, nsReflowMetrics & {...}, const nsReflowState & {...}) line 515 NEWLINE nsTableRowFrame::InitialReflow(nsIPresContext & {...}, RowReflowState & {...}, NEWLINE nsReflowMetrics & {...}) line 768 + 30 bytes NEWLINE nsTableRowFrame::Reflow(nsTableRowFrame * const 0x0118d490, nsIPresContext & NEWLINE {...}, nsReflowMetrics & {...}, const nsReflowState & {...}, unsigned int & 0) NEWLINE line 1083 + 20 bytes NEWLINE nsContainerFrame::ReflowChild(nsIFrame * 0x0118d490, nsIPresContext * NEWLINE 0x01239ec0, nsReflowMetrics & {...}, const nsReflowState & {...}) line 515 NEWLINE nsTableRowGroupFrame::ReflowMappedChildren(nsIPresContext * 0x01239ec0, NEWLINE RowGroupReflowState & {...}, nsSize * 0x0012cb40) line 346 + 33 bytes NEWLINE nsTableRowGroupFrame::Reflow(nsTableRowGroupFrame * const 0x0118d340, NEWLINE nsIPresContext & {...}, nsReflowMetrics & {...}, const nsReflowState & {...}, NEWLINE unsigned int & 0) line 1155 + 23 bytes NEWLINE nsContainerFrame::ReflowChild(nsIFrame * 0x0118d340, nsIPresContext * NEWLINE 0x01239ec0, nsReflowMetrics & {...}, const nsReflowState & {...}) line 515 NEWLINE nsTableFrame::ResizeReflowPass1(nsIPresContext * 0x01239ec0, nsReflowMetrics & NEWLINE {...}, const nsReflowState & {...}, unsigned int & 0) line 1679 + 30 bytes NEWLINE nsTableFrame::Reflow(nsTableFrame * const 0x0118d1d0, nsIPresContext & {...}, NEWLINE nsReflowMetrics & {...}, const nsReflowState & {...}, unsigned int & 0) line NEWLINE 1462 + 30 bytes NEWLINE nsContainerFrame::ReflowChild(nsIFrame * 0x0118d1d0, nsIPresContext * NEWLINE 0x01239ec0, nsReflowMetrics & {...}, const nsReflowState & {...}) line 515 NEWLINE nsTableOuterFrame::Reflow(nsTableOuterFrame * const 0x0118d120, nsIPresContext & NEWLINE {...}, nsReflowMetrics & {...}, const nsReflowState & {...}, unsigned int & 0) NEWLINE line 468 + 27 bytes NEWLINE nsBlockFrame::ReflowBlockFrame(nsBlockReflowState & {...}, LineData * NEWLINE 0x011a7d40, nsIFrame * 0x0118d120, unsigned int & 16640) line 2612 + 37 bytes NEWLINE nsBlockFrame::ReflowLine(nsBlockReflowState & {...}, LineData * 0x011a7d40, NEWLINE unsigned int & 16640) line 2374 + 24 bytes NEWLINE nsBlockFrame::ReflowLinesAt(nsBlockReflowState & {...}, LineData * 0x011a7d40) NEWLINE line 2246 + 20 bytes NEWLINE nsBlockFrame::ResizeReflow(nsBlockReflowState & {...}) line 2232 NEWLINE nsBlockFrame::InitialReflow(nsBlockReflowState & {...}) line 1853 NEWLINE nsBlockFrame::ReflowAround(nsBlockFrame * const 0x01187ce0, nsIPresContext & NEWLINE {...}, nsISpaceManager * 0x011afe10, nsReflowMetrics & {...}, const NEWLINE nsReflowState & {...}, nsRect & {...}, unsigned int & 1234028) line 1368 + 18 NEWLINE bytes NEWLINE nsBlockFrame::ReflowBlockFrame(nsBlockReflowState & {...}, LineData * NEWLINE 0x011af550, nsIFrame * 0x01187c90, unsigned int & 16640) line 2600 + 47 bytes NEWLINE nsBlockFrame::ReflowLine(nsBlockReflowState & {...}, LineData * 0x011af550, NEWLINE unsigned int & 16640) line 2374 + 24 bytes NEWLINE nsBlockFrame::ReflowLinesAt(nsBlockReflowState & {...}, LineData * 0x011af550) NEWLINE line 2246 + 20 bytes NEWLINE nsBlockFrame::ResizeReflow(nsBlockReflowState & {...}) line 2232 NEWLINE nsBlockFrame::InitialReflow(nsBlockReflowState & {...}) line 1853 NEWLINE nsBlockFrame::ReflowAround(nsBlockFrame * const 0x011aff10, nsIPresContext & NEWLINE {...}, nsISpaceManager * 0x011afe10, nsReflowMetrics & {...}, const NEWLINE nsReflowState & {...}, nsRect & {...}, unsigned int & 0) line 1368 + 18 bytes NEWLINE nsBodyFrame::Reflow(nsBodyFrame * const 0x011afd80, nsIPresContext & {...}, NEWLINE nsReflowMetrics & {...}, const nsReflowState & {...}, unsigned int & 0) line 264 NEWLINE nsContainerFrame::ReflowChild(nsIFrame * 0x011afd80, nsIPresContext * NEWLINE 0x01239ec0, nsReflowMetrics & {...}, const nsReflowState & {...}) line 515 NEWLINE nsTableCellFrame::Reflow(nsTableCellFrame * const 0x0118aa70, nsIPresContext & NEWLINE {...}, nsReflowMetrics & {...}, const nsReflowState & {...}, unsigned int & 0) NEWLINE line 363 + 30 bytes NEWLINE nsContainerFrame::ReflowChild(nsIFrame * 0x0118aa70, nsIPresContext * NEWLINE 0x01239ec0, nsReflowMetrics & {...}, const nsReflowState & {...}) line 515 NEWLINE nsTableRowFrame::InitialReflow(nsIPresContext & {...}, RowReflowState & {...}, NEWLINE nsReflowMetrics & {...}) line 768 + 30 bytes NEWLINE nsTableRowFrame::Reflow(nsTableRowFrame * const 0x0118a8c0, nsIPresContext & NEWLINE {...}, nsReflowMetrics & {...}, const nsReflowState & {...}, unsigned int & 0) NEWLINE line 1083 + 20 bytes NEWLINE nsContainerFrame::ReflowChild(nsIFrame * 0x0118a8c0, nsIPresContext * NEWLINE 0x01239ec0, nsReflowMetrics & {...}, const nsReflowState & {...}) line 515 NEWLINE nsTableRowGroupFrame::ReflowMappedChildren(nsIPresContext * 0x01239ec0, NEWLINE RowGroupReflowState & {...}, nsSize * 0x0012def8) line 346 + 33 bytes NEWLINE nsTableRowGroupFrame::Reflow(nsTableRowGroupFrame * const 0x0117b8f0, NEWLINE nsIPresContext & {...}, nsReflowMetrics & {...}, const nsReflowState & {...}, NEWLINE unsigned int & 0) line 1155 + 23 bytes NEWLINE nsContainerFrame::ReflowChild(nsIFrame * 0x0117b8f0, nsIPresContext * NEWLINE 0x01239ec0, nsReflowMetrics & {...}, const nsReflowState & {...}) line 515 NEWLINE nsTableFrame::ResizeReflowPass1(nsIPresContext * 0x01239ec0, nsReflowMetrics & NEWLINE {...}, const nsReflowState & {...}, unsigned int & 0) line 1679 + 30 bytes NEWLINE nsTableFrame::Reflow(nsTableFrame * const 0x0117b7b0, nsIPresContext & {...}, NEWLINE nsReflowMetrics & {...}, const nsReflowState & {...}, unsigned int & 0) line NEWLINE 1462 + 30 bytes NEWLINE nsContainerFrame::ReflowChild(nsIFrame * 0x0117b7b0, nsIPresContext * NEWLINE 0x01239ec0, nsReflowMetrics & {...}, const nsReflowState & {...}) line 515 NEWLINE nsTableOuterFrame::Reflow(nsTableOuterFrame * const 0x0117b700, nsIPresContext & NEWLINE {...}, nsReflowMetrics & {...}, const nsReflowState & {...}, unsigned int & 0) NEWLINE line 468 + 27 bytes NEWLINE nsBlockFrame::ReflowBlockFrame(nsBlockReflowState & {...}, LineData * NEWLINE 0x011d8610, nsIFrame * 0x0117b700, unsigned int & 16640) line 2612 + 37 bytes NEWLINE nsBlockFrame::ReflowLine(nsBlockReflowState & {...}, LineData * 0x011d8610, NEWLINE unsigned int & 16640) line 2374 + 24 bytes NEWLINE nsBlockFrame::ReflowLinesAt(nsBlockReflowState & {...}, LineData * 0x011d8610) NEWLINE line 2246 + 20 bytes NEWLINE nsBlockFrame::ResizeReflow(nsBlockReflowState & {...}) line 2232 NEWLINE nsBlockFrame::InitialReflow(nsBlockReflowState & {...}) line 1853 NEWLINE nsBlockFrame::ReflowAround(nsBlockFrame * const 0x0117b560, nsIPresContext & NEWLINE {...}, nsISpaceManager * 0x01241b30, nsReflowMetrics & {...}, const NEWLINE nsReflowState & {...}, nsRect & {...}, unsigned int & 1239076) line 1368 + 18 NEWLINE bytes NEWLINE nsBlockFrame::ReflowBlockFrame(nsBlockReflowState & {...}, LineData * NEWLINE 0x011e0c80, nsIFrame * 0x0117b510, unsigned int & 16640) line 2600 + 47 bytes NEWLINE nsBlockFrame::ReflowLine(nsBlockReflowState & {...}, LineData * 0x011e0c80, NEWLINE unsigned int & 16640) line 2374 + 24 bytes NEWLINE nsBlockFrame::ReflowLinesAt(nsBlockReflowState & {...}, LineData * 0x011e0c80) NEWLINE line 2246 + 20 bytes NEWLINE nsBlockFrame::ResizeReflow(nsBlockReflowState & {...}) line 2232 NEWLINE nsBlockFrame::InitialReflow(nsBlockReflowState & {...}) line 1853 NEWLINE nsBlockFrame::ReflowAround(nsBlockFrame * const 0x0124a430, nsIPresContext & NEWLINE {...}, nsISpaceManager * 0x01241b30, nsReflowMetrics & {...}, const NEWLINE nsReflowState & {...}, nsRect & {...}, unsigned int & 1240436) line 1368 + 18 NEWLINE bytes NEWLINE nsBlockFrame::ReflowBlockFrame(nsBlockReflowState & {...}, LineData * NEWLINE 0x011e0170, nsIFrame * 0x0124a3e0, unsigned int & 16640) line 2600 + 47 bytes NEWLINE nsBlockFrame::ReflowLine(nsBlockReflowState & {...}, LineData * 0x011e0170, NEWLINE unsigned int & 16640) line 2374 + 24 bytes NEWLINE nsBlockFrame::ReflowLinesAt(nsBlockReflowState & {...}, LineData * 0x011e0170) NEWLINE line 2246 + 20 bytes NEWLINE nsBlockFrame::FrameAppendedReflow(nsBlockReflowState & {...}) line 1943 NEWLINE nsBlockFrame::ReflowAround(nsBlockFrame * const 0x012421f0, nsIPresContext & NEWLINE {...}, nsISpaceManager * 0x01241b30, nsReflowMetrics & {...}, const NEWLINE nsReflowState & {...}, nsRect & {...}, unsigned int & 0) line 1388 + 18 bytes NEWLINE nsBodyFrame::Reflow(nsBodyFrame * const 0x012419e0, nsIPresContext & {...}, NEWLINE nsReflowMetrics & {...}, const nsReflowState & {...}, unsigned int & 0) line 264 NEWLINE nsContainerFrame::ReflowChild(nsIFrame * 0x012419e0, nsIPresContext * NEWLINE 0x01239ec0, nsReflowMetrics & {...}, const nsReflowState & {...}) line 515 NEWLINE RootContentFrame::Reflow(RootContentFrame * const 0x01242400, nsIPresContext & NEWLINE {...}, nsReflowMetrics & {...}, const nsReflowState & {...}, unsigned int & 0) NEWLINE line 353 + 27 bytes NEWLINE nsContainerFrame::ReflowChild(nsIFrame * 0x01242400, nsIPresContext * NEWLINE 0x01239ec0, nsReflowMetrics & {...}, const nsReflowState & {...}) line 515 NEWLINE RootFrame::Reflow(RootFrame * const 0x012413d0, nsIPresContext & {...}, NEWLINE nsReflowMetrics & {...}, const nsReflowState & {...}, unsigned int & 0) line 156 NEWLINE + 27 bytes NEWLINE nsHTMLReflowCommand::Dispatch(nsHTMLReflowCommand * const 0x011e0f00, NEWLINE nsIPresContext & {...}, nsReflowMetrics & {...}, const nsSize & {...}) line 133 NEWLINE PresShell::ProcessReflowCommands() line 662 NEWLINE PresShell::ExitReflowLock(PresShell * const 0x0123aca0) line 415 NEWLINE PresShell::ContentAppended(PresShell * const 0x0123aca8, nsIDocument * NEWLINE 0x0126be70, nsIContent * 0x0123e8fc, int 0) line 736 NEWLINE nsDocument::ContentAppended(nsIContent * 0x0123e8fc, int 0) line 525 NEWLINE HTMLContentSink::WillInterrupt(HTMLContentSink * const 0x0126b560) line 1398 NEWLINE CNavDTD::WillInterruptParse(CNavDTD * const 0x0123b870) line 3090 + 18 bytes NEWLINE nsParser::ResumeParse() line 567 NEWLINE nsParser::OnDataAvailable(nsParser * const 0x0126b514, nsIURL * 0x01237990, NEWLINE nsIInputStream * 0x0126a8e0, int 2145) line 775 NEWLINE nsDocumentBindInfo::OnDataAvailable(nsDocumentBindInfo * const 0x01237930, NEWLINE nsIURL * 0x01237990, nsIInputStream * 0x0126a8e0, int 2145) line 1114 + 30 bytes NEWLINE stub_put_block(_NET_StreamClass * 0x0126a260, char * 0x010e12d0, long 2145) line NEWLINE 574 + 36 bytes NEWLINE net_MemCacheWrite(_NET_StreamClass * 0x0123a580, char * 0x010e12d0, long 2145) NEWLINE line 665 + 24 bytes NEWLINE net_pull_http_data(_ActiveEntry * 0x0124b1c0) line 3019 + 30 bytes NEWLINE net_ProcessHTTP(_ActiveEntry * 0x0124b1c0) line 3404 + 9 bytes NEWLINE NET_ProcessNet(PRFileDesc * 0x01191fc0, int 2) line 3379 + 13 bytes NEWLINE NET_PollSockets() line 180 + 18 bytes NEWLINE nsNetlibService::NetPollSocketsCallback(nsITimer * 0x01235fb0, void * NEWLINE 0x010f1e20) line 585 NEWLINE TimerImpl::Fire(unsigned long 60788669) line 308 + 17 bytes NEWLINE TimerImpl::ProcessTimeouts(unsigned long 60788669) line 187 NEWLINE FireTimeout(void * 0x00000000, unsigned int 275, unsigned int 32611, unsigned NEWLINE long 60788669) line 101 + 9 bytes NEWLINE USER32! 77e7128c() NEWLINE main(int 1, char * * 0x01114540) line 96 NEWLINE mainCRTStartup() line 338 + 17 bytes
831|Incorrect table layout|This is actually one of the frames displayed in http://www.legomindstorms.com NEWLINE (it makes heavy use of scripting). One of the images is missing and the NEWLINE background of one of the cells is not painted (hence the wrong background for NEWLINE the last image). At least part of the problem is table related.
832|Can't call method "prepare" without a package or object reference error|Running enter_bug.cgi from webserver get the following line in error.log: NEWLINE -------------- NEWLINE [Thu Sep 17 12:49:53 1998] [error] Premature end of script headers: /home/dyp/ht NEWLINE ml/bugzilla/enter_bug.cgi NEWLINE -------------- NEWLINE NEWLINE When I running it manually I get: NEWLINE NEWLINE -------------- NEWLINE bash-2.02$ ./enter_bug.cgi NEWLINE Can't call method "prepare" without a package or object reference at NEWLINE /usr/lib/perl5/Mysql.pm line 150 (#1) NEWLINE NEWLINE (F) You used the syntax of a method call, but the slot filled by the NEWLINE object reference or package name contains an expression that returns NEWLINE neither an object reference nor a package name. (Perhaps it's null?) NEWLINE Something like this will reproduce the error: NEWLINE NEWLINE $BADREF = undef; NEWLINE process $BADREF 1,2,3; NEWLINE $BADREF->process(1,2,3); NEWLINE NEWLINE Content-type: text/html NEWLINE NEWLINE <H1>Software error:</H1> NEWLINE <CODE>Can't call method "prepare" without a package or object reference at /usr/ NEWLINE lib/perl5/Mysql.pm line 150. NEWLINE </CODE> NEWLINE <P> NEWLINE Please send mail to this site's webmaster for help. NEWLINE [Thu Sep 17 12:51:47 1998] enter_bug.cgi: Can't call method "prepare" without a NEWLINE package or object reference at /usr/lib/perl5/Mysql.pm line 150. NEWLINE -------------
833|unable to browse web or ftp links on Win98|As of the 98/09/16 binary build, mozilla is unable to view pages originating on NEWLINE http or ftp protocols. It is able to load and view local documents, even those NEWLINE that include inline graphics that are remotely accessed by http, without NEWLINE problems. A debug event viewer shows the error "Warning: attempt to call NEWLINE Invoke with NULL m_lpDispatch!" logged twice when you try to visit a document NEWLINE via http.
834|Mozilla crashes when a javascript pop up window occurs|I'm using Windows 98 and VC 5.0 with 128MB of RAM. NEWLINE Here's the details of the error. NEWLINE NEWLINE >MOZILLA caused an invalid page fault in NEWLINE >module MOZILLA.EXE at 017f:00774b70. NEWLINE NEWLINE This is the binary build created on 9-17-98
835|META description causes assertion failure in the RDF module|When trying to browse some pages I got a lot of assertion failures in the rdf NEWLINE modules so I stripped the source code of those pages as much as possible and NEWLINE found that the tag: NEWLINE <meta name="description" content="Foo"> NEWLINE caused it but only when visiting certain domains. For instance, you get the NEWLINE failures at http://c245.ryd.student.liu.se/mozillabug1.html and at NEWLINE http://www.student.liu.se/ but not at http://www.lysator.liu.se/~bratell/ which NEWLINE has a copy of the page at c245.ryd.student.liu.se. NEWLINE NEWLINE What I get is assertion failures at /mozilla/modules/rdf/src/mcf.c line 321, NEWLINE /mozilla/modules/rdf/src/nlcstore.c line 328 NEWLINE NEWLINE or NEWLINE NEWLINE assertion failures at /mozilla/modules/rdf/src/nlcstore.c line 371, NEWLINE /mozilla/modules/rdf/src/mcf.c line 631 NEWLINE NEWLINE I didn't manage to get a stack trace since I was using a nightly build NEWLINE (19980917), but maybe you can reproduce it yourself. Since this might be caused NEWLINE by some I18n stuff, I should tell you that I have a Swedish system (character NEWLINE set latin-1)
836|french accents|The characters with french accents are not displayed correctly in the bookmarks NEWLINE list.
837|HTML page corruption, leaking from previous page.|Running Linux 2.1.121 on Redhat 5.0 with custom built dual Pentium 133, 64 MB NEWLINE RAM. Build was Mozilla 5.0a1 (5009808600), downloaded as the 980914 nightly NEWLINE binary. NEWLINE NEWLINE View the above Slashdot page, then click on the news.com site. If you scroll NEWLINE down to the bottom of the page with the scrollbar and then quickly back to the NEWLINE top, the screen will corrupt itself. Layout seemed the best fit though I don't NEWLINE know if it is X specific. An example of how it looks after the corruption is NEWLINE at: NEWLINE NEWLINE http://www.van.ml.org/~greerga/ack.jpg
838|Mozilla fails to build correctly on any non-gcc/g++ HP-UX compiler|Well, I'm back again with my exploits on HP-UX 10.20 and the by-now-surely- NEWLINE infamous 4th September 1998 Mozilla build - here's a progress report: NEWLINE NEWLINE * HP's cfront-based C++ ("CC" command) just can't build 4th Sept 98 Mozilla NEWLINE at all (bombs out early during the compile and many fixes later convinced NEWLINE me to give up). All previous Mozilla releases built fine with "CC", so there's NEWLINE a lot of "g++-isms" been added with the 4th Sept release I suspect. NEWLINE NEWLINE * HP's ANSI C++ ("aCC" command) can actually build Mozilla, but it core dumps NEWLINE during intiialisation. I didn't really fancy debugging this, so I concentrated NEWLINE on installing the latest gcc/libstdc++ and using that to build Mozilla NEWLINE instead. NEWLINE NEWLINE * gcc 2.8.1/libstdc++ 2.8.1.1 both builds Mozilla and manages to run Mozilla NEWLINE on HP-UX 10.20, but there's serious problems with background tiling - loads NEWLINE of graphical corruption on any pages with background tiles. NEWLINE NEWLINE So the advice is to stick with gcc/g++ for building Mozilla on HP-UX - all NEWLINE other compilers don't work. Even gcc/g++ produces that background tile problem NEWLINE though. Has anyone at Netscape actually managed to build the 4th Sept 98 NEWLINE Mozilla at all and get it to work sensibly ? NEWLINE NEWLINE BTW, there's a major omission in the build config for Mozilla - a "mid-way" NEWLINE build - i.e. one without optimisation *or* debugging. I ended up having to NEWLINE hack around in 3 or 4 places to get rid of the -O's (whilst keeping BUILD_OPT NEWLINE defined). I did this to rule out optimisation bugs in gcc/g++ (I ruled them out: NEWLINE same problem with a non-optimised build).
839|border widths can be negative|<STYLE> NEWLINE #negative { border-bottom-width: -10px; } NEWLINE </STYLE> NEWLINE <P ID=negative>Hi there! I should have no border. Go away! NEWLINE NEWLINE The above test doesn't work right in raptor because the style system isn't NEWLINE clamping the border value to 0. The CSS2 spec says that borders are not NEWLINE allowed to be negative.
840|libplds21.so: can't open file: errno=2|Not really sure who this goes to... but if you would please direct to the proper NEWLINE person it will be much appreciated. NEWLINE NEWLINE I am just a software engineer @ Raytheon and thought I would download the source NEWLINE and build it... well... built Mozilla on a Sun Ultra 2 Enterprise running NEWLINE Solaris 2.5.1, got a few warnings but when I went to run NEWLINE "<mozilla_dir>./dist/SunOS5.5.1_sparc_OPT.OBJ/bin/mozilla-export" NEWLINE NEWLINE I got this error: NEWLINE "ld.so.1: ./dist/SunOS5.5.1_sparc_OPT.OBJ/bin/mozilla-export: fatal: NEWLINE libplds21.so: can't open file: errno=2 NEWLINE Killed" NEWLINE NEWLINE (I also had to download gmake, got version 3.77, we are using gcc 2.7.2 - which NEWLINE you say that you support.) NEWLINE NEWLINE I saw some "sh:" errors in the make having to deal with xemacs.... didn't NEWLINE understand why... since you said on your web page that you didn't need xemacs. NEWLINE NEWLINE Soooo.... did I not compile correctly, or can it just not find a file?
841|[PP]changing visited link color crashes Mozilla|I just downloaded the 9/18/98 binary of Mozilla for Mac/PPC. I am running MacOS NEWLINE 8.1 on a PowerMac G3/300 with 128MB of RAM. NEWLINE NEWLINE I launched Mozilla and was successfully playing around with it when I tried to NEWLINE change the visited link color (to red) in the prefs dialog. Shortly after I hit NEWLINE the OK button, it crashed with a Type 3 error. NEWLINE NEWLINE After that, running the same user configuration caused Mozilla to crash while NEWLINE starting up. I was able to create a new user, and tried changing the visited NEWLINE link color again--it crashed again. I tried downloading a fresh copy of NEWLINE Mozilla, and the same thing occurred. In short, this bug seems very repeatable. NEWLINE NEWLINE Cordially, NEWLINE Steven G. Johnson
842|No MacOS 8 option in "OS" popup|The "OS" popup menu only lists "System 7" and "System 7.5". Besides the fact NEWLINE that it would be more correct (and clear) to write "MacOS" instead of "System," NEWLINE there is the obvious omission of no MacOS 8, 8.1, or 8.5 options. There is a NEWLINE cryptic "7.1.6" option; I have no clue what OS that is supposed to be. You're NEWLINE also missing Windows 98.
843|Colour controls not disabled when preference dialog opened|If the "Use Windows Colors" preference setting is turned on, and then the NEWLINE "Preferences" dialog is opened and the Appearance:Colours pane selected, the NEWLINE "Text" and "Background" colour selectors should be disabled and dimmed, but NEWLINE instead they are left enabled and visible. If you then click the "Use windows NEWLINE colours" checkbox twice so that it is turned off and then back on, the "Text" NEWLINE and "Background" colour selectors become diabled and dimmed as they should be NEWLINE when the dialog is opened.
844|whitespace at the end of text shouldn't be underlined|Probably shouldn't bother rendering it all!
845|CSS Parser ignoring first declaration in STYLE element surrounded by comments|When CSS in a STYLE element is enclosed in SGML comments (as described at the NEWLINE end of section 1.1 in http://www.w3.org/TR/REC-CSS1#containment-in-html ), the NEWLINE first declaration is ignored. See http://www.w3.org/TR/REC-CSS1#containment-in- NEWLINE html , where the P elements (about half of the page) should be red. NEWLINE NEWLINE (I didn't test other pages to see if this happens all the time.) NEWLINE NEWLINE (I also didn't know whether to file this under Parser or Style System.)‰
846|History Windows Does Nothing|Simple install of 4.5b2 over 4.5b1 - and suddenly the History windows does NEWLINE nothing. It's always completely empty.
847|IID collision in nsIConnectionPoint|nsIEnumConnections and nsIEnumConnectionPoints have the same IID. NEWLINE Don't ask me how or why I spotted that. I have no idea. NEWLINE It doesn't really matter since they're unused. NEWLINE Plus ConnectionPoints are the suckiest part of COM. Forget 'em.
848|Viewer App Reports incorrect HTTP_USER_AGENT version|NGLayout for Windows reports that is the WinNT version of Netscape 4.05. This NEWLINE is done on line 516 of /webshell/tests/viewer/nsViewerApp.cpp . The string is NEWLINE currently set to "4.05 [en] (WinNT;I)". It should probably be set to something NEWLINE like "5.5b0 NGLayout" and possibly contain a build date or platform. However, NEWLINE it is very important that it be changed to *anything else* because the reporting NEWLINE version 4.05 causes NGLayout to receive browser-sniffed versions of CSS for Win NEWLINE Navigator 4 that don't send the stuff that crashes Navigator 4 (lots of stuff). NEWLINE However, you may need to use a version number such as 5.5test or 6.0test for the NEWLINE benefit of other types of browser sniffing (I'm not sure what is out there). NEWLINE NEWLINE This is related to bug 806, but I think this should be changed even before a NEWLINE cross-platform way of changing it is made so that NGLayout can be tested NEWLINE properly.‰
849|Somtimes forgets scrollbar, sometimes draws blank page.|If you enter something into the advanced query box (e.g. fish and vitamins) and NEWLINE do a search, then the resulting page will not have the necessary vertical NEWLINE scrollbar. If you press the down arrow or drag the mouse down across the bottom NEWLINE of the screen, the scrollbar will appear. NEWLINE NEWLINE Jump to page 10 and click on a search result, then jump back. The page will be NEWLINE completely white. Refresh does not do anything, bu redraw does. This is fairly NEWLINE reproducable. Actually, I think just trying a few links on the first page NEWLINE should suffice instead of going to page 10.
850|CSS Universal Selector should be supported|In a style system where style sheets can cascade, it is important to support the NEWLINE CSS2 universal selector, "*", as described in the CSS2 spec at the URL above, NEWLINE and with priority rules as defined at NEWLINE http://www.w3.org/TR/REC-CSS2/cascade.html#specificity NEWLINE NEWLINE This is needed in case a user style sheet defines, for example, a color or NEWLINE background-color for B, EM, DT, or some other similar element. The page author, NEWLINE to override this, would either need to define his color and background-color on NEWLINE every single element, or use the universal selector. This is why it is so NEWLINE important (but is only useful in the context of *cascading* style sheets.) NEWLINE ‰
851|Profile menu exit assert...|I hate reporting in bugs whose description fits on the summary line. I assume NEWLINE not many people have a change of running into this one, since the compile NEWLINE instructions say to use 'mozilla -p"mozilla"'. Well, don't, and choose Exit NEWLINE instead. There's an immediate assert on trying to delete an unexistent panel, NEWLINE which is hopefully rather simple problem to fix: NEWLINE NEWLINE XP_AssertAtLine(char * 0x008cd218, int 2427) line 2687 NEWLINE HT_DeletePane(_HT_PaneStruct * 0x00000000) line 2427 + 37 bytes NEWLINE HT_Shutdown() line 123 + 11 bytes NEWLINE RDF_Shutdown() line 151 NEWLINE CNetscapeApp::ExitInstance() line 2242 NEWLINE AfxWinMain(HINSTANCE__ * 0x00400000, HINSTANCE__ * 0x00000000, char * NEWLINE 0x001418e7, int 10) line 47 + 11 bytes NEWLINE WinMain(HINSTANCE__ * 0x00400000, HINSTANCE__ * 0x00000000, char * 0x001418e7, NEWLINE int 10) line 34 NEWLINE WinMainCRTStartup() line 330 + 54 bytes NEWLINE KERNEL32! 77f1b304() NEWLINE NEWLINE I've started using Alexa's navigation-tool that interfaces into the browser as NEWLINE of this bugreport, though, which I originally thought was related to the bug, NEWLINE but now it would seem independent.
852|Remembering Username/password for entry field failed|Entered in email address and then click on 'Email me password'. NEWLINE Dialogue box popped up and asked to save password. NEWLINE Then asked to enable automatic insertion of l/p for this page. NEWLINE Upon clicking Okay to enable, mozilla crapped the bed. NEWLINE NEWLINE Stack Dump: NEWLINE MOZILLA caused an invalid page fault in NEWLINE module MSVCRTD.DLL at 0217:10216120. NEWLINE Registers: NEWLINE EAX=00000000 CS=0217 EIP=10216120 EFLGS=00010246 NEWLINE EBX=00c4fc46 SS=021f ESP=00c4faec EBP=00c4fb38 NEWLINE ECX=00000000 DS=021f ESI=00008c6a FS=46a7 NEWLINE EDX=800053d8 ES=021f EDI=00c4fbfc GS=0000 NEWLINE Bytes at CS:EIP: NEWLINE 8b 01 ba ff fe fe 7e 03 d0 83 f0 ff 33 c2 83 c1 NEWLINE Stack dump: NEWLINE 007a3718 00000000 816db6f8 00008c6a 1020b174 10257f50 00000000 102106f4 NEWLINE 00000009 00c4fb20 00000001 00e8d890 00000001 00c4fb38 007a4881 00e8d890 NEWLINE NEWLINE System: Gateway 2k p5-120(over clocked to 133)
853|padding-left property doesn't work|We do padding-right/top/bottom properly, but not padding-left.
854|{sink} content sink needs to support nested incremental reflow|This test case is a document with a <UL> followed by 500 <LI>s followed by a NEWLINE </UL>. NEWLINE NEWLINE If you load this URL in IE, you'll see that it lays out list items NEWLINE incrementally. NEWLINE NEWLINE We don't show anything until we hit the </UL> - you'll notice that the same set NEWLINE of list items broken up into 25 UL groups renders much faster (see test case at: NEWLINE http://www.mozilla.org/nglayout/testcases/stress/500in25uls.html ) NEWLINE NEWLINE We should lay out long lists incrementally. We'll need to do this in order to NEWLINE support an Aurora Front End in HTML.
855|animated gif not displaying|This url causes big problems with the 4.x browsers too. NEWLINE The image http://www.mathcs.emory.edu/~abell/sowanim.gif NEWLINE is not being re-displayed properly after being obscured NEWLINE by another window. On Win95 with 4.x browser the space where NEWLINE the image was still contains the same bits as the window that NEWLINE obscured it. Eventually on win95 netscape will crash after being NEWLINE manipulated a bit. NEWLINE NEWLINE With mozilla on Linux (redhat 5.0, 2.0.35) NEWLINE the redrawn image is overwritten by some kind of noise NEWLINE pattern. If do "view image" in linux a corrupted image shows up NEWLINE and i get a "Mozilla is out of memory" warning (I have 128mb ram NEWLINE and not much else running). NEWLINE NEWLINE Since sulaco.htm has a background image the linux build has the NEWLINE bug 514 problem but it seems to act a lot differently than i've NEWLINE seen with other pages. The left half of the screen is mostly ok NEWLINE but the right half either has no image or sometimes shows the NEWLINE bits from the previous page. The background image problem isn't NEWLINE as bad when the browser is first loaded but if I run it for a NEWLINE while until i've stumbled across a few asserts then the background NEWLINE gets pretty messed up. The page has 2 animated gifs other than the NEWLINE broken cow image. The images are centered. At the moment on my NEWLINE screen, areas to the left of them seem to have the background image NEWLINE applied correctly. Everything to the right of the left edge of the NEWLINE animations has a plain black background except the space directly NEWLINE to the right of the image. So far in this session i've had an assert NEWLINE at line 179 of xfe.c and in a previous with a more messed up background NEWLINE I also had an assert at line 953 of Browserframe.cpp. NEWLINE NEWLINE I last updated from cvs on sept 20.
856|Crashing on startup ....|Mozilla starts up, begins displaying content of www.mozilla.org, everything NEWLINE running perfect for a few seconds, and then there appears panel of the tool bar NEWLINE and mozilla crashes. The message is: NEWLINE NEWLINE Debug Assertion Failed NEWLINE Program: D:\\MOZILLA\\X86DBG\\MOZILLA.EXE NEWLINE File: Y:\\NIGHTLY\\MOZILLA\\CMD\\WINFE\\FENET.CPP NEWLINE Line: 199 NEWLINE NEWLINE If you press Retry or Ignore, you get error: NEWLINE NEWLINE MOZILLA caused an invalid page fault in NEWLINE module MSVCRTD.DLL at 0137:10216120. NEWLINE Registers: NEWLINE EAX=00000000 CS=0137 EIP=10216120 EFLGS=00010246 NEWLINE EBX=00b40000 SS=013f ESP=00c4de28 EBP=00c4de78 NEWLINE ECX=00000000 DS=013f ESI=00c4e720 FS=1137 NEWLINE EDX=80094df8 ES=013f EDI=00ca46b0 GS=0000 NEWLINE Bytes at CS:EIP: NEWLINE 8b 01 ba ff fe fe 7e 03 d0 83 f0 ff 33 c2 83 c1 NEWLINE Stack dump: NEWLINE 006ee6c1 00000000 00ca3cf0 00e72000 00c4de54 10210ab7 00eb1660 5f4a1914 00000000 NEWLINE 00000001 0000003c 0000002a 00000064 00000064 0000003c 0000002a NEWLINE NEWLINE Good luck !
857|<ins> and <del> tag parsed incorrectly|see the minimized test case below. The <INS> and <DEL> tags incorrectly NEWLINE terminate the table. NEWLINE NEWLINE ------------ Test case ----------------- NEWLINE NEWLINE <HTML><BODY> NEWLINE <TABLE BORDER=1> NEWLINE <TR><TD><DEL>...</DEL><TD>Normal <DEL>Deleted style</DEL> Normal NEWLINE <TR><TD>This should be in the table above</TD></TR> NEWLINE </table> NEWLINE <br><br> NEWLINE <table border> NEWLINE <TR><TD><TT><INS>...</INS></TT><TD>Normal <INS>Inserted style</INS> NEWLINE Normal NEWLINE <TR><TD>This should be in the table above</TD></TR> NEWLINE </TABLE> NEWLINE </BODY></HTML>
858|[Feature] JavaScript auto-disable per-domain RFE|This is a request for enhancement. It's overdue in mozilla, in the competition, NEWLINE and demanded by champions and users. NEWLINE NEWLINE When I or Chuck Simmons (chrlsim@futureone.com, netscape champion) or many NEWLINE others surf the wilds of the Internet, we turn JS off. Why? Paranoia, good NEWLINE sense, whatever -- it doesn't matter. Mozilla has not had its last security NEWLINE hole closed. This goes for JS, Java, HTML layout (remember the Danish NEWLINE form-type-change attack), netlib, etc. Then there are denial of service attacks NEWLINE to consider. NEWLINE NEWLINE OK, what can be done? This bug (really, RFE) asks that mozilla at least NEWLINE automate the disabling of executable content when surfing away from URLs that NEWLINE begin with host parts from a known-trustworthy set of fully-qualified domain NEWLINE names. NEWLINE NEWLINE That requires preference UI support, I suppose. Although with just the pref NEWLINE checking code in libmocha and the Java glue, and with a signed script that NEWLINE called navigator.preference, we (or anyone trusted) could construct a "set your NEWLINE shields-up preferences" page on mozilla.org, home.netscape.com, that acted as a NEWLINE web-server-based pref UI. NEWLINE NEWLINE So Mike, can you bug me about implementation, do the libmocha hacks (or find NEWLINE someone else to do them), then reassign this bug to raman for Java? We should NEWLINE figure out the pref syntax and value types first, make them common and NEWLINE extensible. After you and raman are done, we can give it to german for UI NEWLINE consideration -- but the web-based pref UI approach seems better to me. NEWLINE NEWLINE /be
859|crash with LI in table|see the minimized test case below. The <LI> in combination with the missing NEWLINE </TD> cause a TR to be inserted as a child of the LI. Frames to get created in NEWLINE an illegal sequence and this leads to a crash in table frames. A dump of the NEWLINE content at the time of the crash follows the test case. NEWLINE NEWLINE ------------ Test case ----------------- NEWLINE NEWLINE <HTML><BODY> NEWLINE <TABLE BORDER=1> NEWLINE <TR> NEWLINE <TD> NEWLINE <LI>Item 1 NEWLINE <LI>Item 2 NEWLINE <LI>Item 3 NEWLINE <!-- note the missing /td --> NEWLINE NEWLINE <tr> NEWLINE <td colspan=3>last thing in the table</td> NEWLINE </tr> NEWLINE </TABLE> NEWLINE NEWLINE </BODY></HTML> NEWLINE NEWLINE NEWLINE ------------ Content Dump -------------- NEWLINE HTML refcount=9< NEWLINE HEAD refcount=4< NEWLINE BODY refcount=7< NEWLINE Text refcount=3<\\n> NEWLINE TABLE BORDER=1 refcount=5< NEWLINE TBODY refcount=4< NEWLINE TR refcount=5< NEWLINE TD refcount=6< NEWLINE Text refcount=3<\\n > NEWLINE LI refcount=4< NEWLINE Bullet RefCnt=3<> NEWLINE Text refcount=3<Item 1\\n > NEWLINE LI refcount=3< NEWLINE Bullet RefCnt=3<> NEWLINE Text refcount=3<Item 2\\n > NEWLINE LI refcount=3< NEWLINE Bullet RefCnt=3<> NEWLINE Text refcount=3<Item 3\\n \\n\\n > NEWLINE TR refcount=4< NEWLINE TD COLSPAN=3 refcount=6< NEWLINE Text refcount=3<last thing in the table> NEWLINE Text refcount=3<\\n> NEWLINE COLGROUP refcount=4< NEWLINE COL refcount=4<
860|Viewer's "Forward" button doesn't work|I'm running the release build on win95
861|CSS parser eats style rule after malformed rule|If the parser encounters an illegal selector value, it discards the following NEWLINE style rule also. NEWLINE NEWLINE ie: NEWLINE <style> NEWLINE H2 { color: red } NEWLINE <> NEWLINE H1 { color: blue } NEWLINE </style> NEWLINE NEWLINE the H1 rule will get discarded
862|resetting a form destroys a textarea prematurely|In test8 (or the following), entering reset will cause the textarea to be NEWLINE destroyed. Entering it a 2nd time will cause the viewer to crash. NEWLINE NEWLINE <html> NEWLINE <body> NEWLINE <form> NEWLINE <textarea>foo</textarea> NEWLINE <input type=reset> NEWLINE </form> NEWLINE </body> NEWLINE </html>
863|we dont' support the hex notation for numeric character references|and we should
864|BUILD: fonts.c not up to date|The file cmd/xfe/fonts.c has not been updated after a change to include/csid.h. NEWLINE The array fe_CharSetInfoArray[] contains only 96 elements, while NEWLINE INTL_CHAR_SET_MAX is now defined as 97 (and all the rest of the file scans NEWLINE fe_CharSetInfoArray[] from 0 to INTL_CHAR_SET_MAX. NEWLINE NEWLINE Running a debug version built from the latest sources prints a couple of NEWLINE warning messages. NEWLINE NEWLINE Index: fonts.c NEWLINE =================================================================== NEWLINE RCS file: /cvsroot/mozilla/cmd/xfe/fonts.c,v NEWLINE retrieving revision 3.7 NEWLINE diff -r3.7 fonts.c NEWLINE 531a532 NEWLINE > { CS_T61 , FE_FONT_TYPE_NONE , FE_FONT_INFO_NONE },
865|[PP][NATIVE] textfield widgets have extra leading whitespace|Example: NEWLINE <input type=text size=10 value="foo"> NEWLINE NEWLINE Expected: NEWLINE [foo ] NEWLINE NEWLINE Actual: NEWLINE [ foo ] NEWLINE NEWLINE There's an extra "blank" space at the beginning of the textfield...
866|Cannot login to bugzilla using NGLayout|If you attempt to file a bug using bugzilla in NGlayout, I am first asked for my NEWLINE email address and password to log in. I enter valid data that works fine in NEWLINE Nav4, but apparently we're not sending the same string as Nav4, because bugzilla NEWLINE refuses my login in NGLayout, but accepts it in Nav4. NEWLINE NEWLINE I suspect the fact that the login form uses GET rather than POST may be part of NEWLINE the issue, but what do I know? :-)
867|Following link in window 2 loads doc in window 1|To reproduce: NEWLINE NEWLINE 1. Launch nglayout (window #1) NEWLINE 2. "File > New Window" (create window #2) NEWLINE 3. In window #2, go to http://www.mozilla.org/bugs/ NEWLINE 4. In window #2 on that page, click "Bugzilla" NEWLINE 5. Notice that the bugzilla page loads in window #1. NEWLINE NEWLINE Not sure if I've selected the right component ("viewer app")...
868|Following link to webmail goes nowhere|I assume (perhaps incorrectly) that the problem is due to the redirects we're NEWLINE doing on webmail.netscape.com. I'm using the release build on win95...
869|window.open() opens new windows but with wrong dimensions, attributes|Test case: NEWLINE NEWLINE <script> NEWLINE var NEWLINE win=window.open("http://home.netscape.com/misc/snf/popup_mkm4.html",'tCw','scrol NEWLINE lbars,width=315,height=320'); NEWLINE </script> NEWLINE NEWLINE In Nav4, this opens a new window that is 315x320 wide as measured in the content NEWLINE area. The window is quite larger if you measure the extra pixels used by the NEWLINE chrome and the scrollbar area. NEWLINE NEWLINE In NGLayout, this opens a window that is 315x320, including all chrome. NEWLINE NEWLINE Note that in Nav4, if scrollbars are specified (as in this example) there is NEWLINE space left to the right of the content area for scrollbars, but no scrollbars NEWLINE are there if they're not necessary (as in this example). NEWLINE NEWLINE Anyway, it's the whole innerWidth/outerWidth thing. NEWLINE NEWLINE Also, NGLayout appears to be trying to render a scrollbar, even though one is NEWLINE not necessary for this example.
870|XFE Mozilla crashes on exit in HT_|Program received signal SIGSEGV, Segmentation fault. NEWLINE 0x40345d96 in HT_GetViewFEData (view=0x0) at ht.c:8627 NEWLINE 8627 return view->pdata; NEWLINE (gdb) bt NEWLINE #0 0x40345d96 in HT_GetViewFEData (view=0x0) at ht.c:8627 NEWLINE #1 0x400cfe7b in XFE_RDFBase::~XFE_RDFBase (this=0x849ec18, __in_chrg=0) NEWLINE at RDFBase.cpp:49 NEWLINE #2 0x400d20ef in XFE_RDFMenuToolbarBase::~XFE_RDFMenuToolbarBase ( NEWLINE this=0x849ec00, __in_chrg=3) at RDFMenuToolbarBase.cpp:99 NEWLINE #3 0x400740c5 in XFE_BookmarkMenu::~XFE_BookmarkMenu (this=0x849ec00, NEWLINE __in_chrg=3) at BookmarkMenu.h:266 NEWLINE #4 0x40073c6b in XFE_BookmarkMenu::destroy_cb (client_data=0x849ec00) NEWLINE at BookmarkMenu.cpp:110 NEWLINE #5 0x408b98f1 in XtCallCallbackList () NEWLINE #6 0x408bfd5b in Phase2Callbacks () NEWLINE #7 0x408bfc97 in Recursive () NEWLINE #8 0x408bfc58 in Recursive () NEWLINE #9 0x408bfc58 in Recursive () NEWLINE #10 0x408bfc85 in Recursive () NEWLINE #11 0x408bfc58 in Recursive () NEWLINE #12 0x408bfc58 in Recursive () NEWLINE #13 0x408bfc58 in Recursive () NEWLINE #14 0x408bfc58 in Recursive () NEWLINE #15 0x408bfc58 in Recursive () NEWLINE #16 0x408c00b5 in XtPhase2Destroy () NEWLINE #17 0x408c0249 in _XtDoPhase2Destroy () NEWLINE #18 0x408c03c7 in XtDestroyWidget () NEWLINE ---Type <return> to continue, or q <return> to quit--- NEWLINE #19 0x4008a6f7 in XFE_Frame::really_delete (data=0x8482c00) at Frame.cpp:1671 NEWLINE #20 0x40733d04 in et_moz_removewindow_epilog (data=0x842b1c0) NEWLINE at et_mocha.c:1644 NEWLINE #21 0x407381c0 in et_HandleEvent_CallFunction (e=0x842b180) at et_moz.c:1984 NEWLINE #22 0x4084312f in PL_HandleEvent (self=0x842b180) at plevent.c:395 NEWLINE #23 0x40843049 in PL_ProcessPendingEvents (self=0x826e280) at plevent.c:357 NEWLINE #24 0x80befa8 in fe_event_processor_callback (closure=0x0, fd=0x8485a54, NEWLINE id=0xbfffd6a8) at mozilla.c:1822 NEWLINE #25 0x408cd93d in XtAppProcessEvent () NEWLINE #26 0x80be1be in fe_EventLoop () at mozilla.c:1151 NEWLINE #27 0x80c1461 in main (argc=1, argv=0xbfffef24) at mozilla.c:3458
871|CSS - text-align: right and text-align: center ignored|On my website, the headings should be right-aligned, and the text in the boxes NEWLINE centered. However, it doesn't happing in Mozilla. NEWLINE NEWLINE (nightly build for 22nd Sep) NEWLINE NEWLINE Rod.
872|Bogus "property is not defined" errors|The lookup code for the JavaScript property cache is wrong. False cache hits are possible when a property is cached as PROP_NOT_FOUND. To the user, the bug manifests itself as bogus "property is not defined" errors, in perfectly valid JavaScript programs. NEWLINE NEWLINE NEWLINE NEWLINE This is the buggy code, file jsinterp.h: NEWLINE NEWLINE NEWLINE NEWLINE #define PROPERTY_CACHE_TEST(cache, obj, id, prop) \\ NEWLINE NEWLINE PR_BEGIN_MACRO \\ NEWLINE NEWLINE uintN _hashIndex = (uintN)PROPERTY_CACHE_HASH(obj, id); \\ NEWLINE NEWLINE JSPropertyCacheEntry *_pce = &(cache)->table[_hashIndex]; \\ NEWLINE NEWLINE (cache)->tests++; \\ NEWLINE NEWLINE if (_pce->property && \\ NEWLINE NEWLINE (_pce->property == PROP_NOT_FOUND \|\| NEWLINE NEWLINE /* *** BUG! obj has to be compared every time! *** */ NEWLINE NEWLINE _pce->property->object == obj) && \\ NEWLINE NEWLINE _pce->symbolid == id) { \\ NEWLINE NEWLINE prop = _pce->property; \\ NEWLINE NEWLINE } else { \\ NEWLINE NEWLINE (cache)->misses++; \\ NEWLINE NEWLINE prop = NULL; \\ NEWLINE NEWLINE } \\ NEWLINE NEWLINE PR_END_MACRO NEWLINE NEWLINE NEWLINE NEWLINE The simplest fix is putting an extra field containing the object into the JSPropertyCacheEntry structure. That's what I did and it works. Here is the fixed section of jsinterp.h: NEWLINE NEWLINE NEWLINE NEWLINE typedef struct JSPropertyCacheEntry { NEWLINE NEWLINE JSObject *obj; /* weak link to object */ NEWLINE NEWLINE JSProperty *property; /* weak link to property */ NEWLINE NEWLINE jsval symbolid; /* strong link to name atom, or index jsval */ NEWLINE NEWLINE } JSPropertyCacheEntry; NEWLINE NEWLINE NEWLINE NEWLINE typedef struct JSPropertyCache { NEWLINE NEWLINE JSPropertyCacheEntry table[PROPERTY_CACHE_SIZE]; NEWLINE NEWLINE JSBool empty; NEWLINE NEWLINE uint32 fills; NEWLINE NEWLINE uint32 recycles; NEWLINE NEWLINE uint32 tests; NEWLINE NEWLINE uint32 misses; NEWLINE NEWLINE uint32 flushes; NEWLINE NEWLINE uint32 pflushes; NEWLINE NEWLINE } JSPropertyCache; NEWLINE NEWLINE NEWLINE NEWLINE #define PROP_NOT_FOUND ((JSProperty *)1) NEWLINE NEWLINE #define PROP_FOUND(prop) ((prword)(prop) & (prword)-2) NEWLINE NEWLINE NEWLINE NEWLINE #define PROPERTY_CACHE_FILL(cx, cache, obj, id, prop) \\ NEWLINE NEWLINE PR_BEGIN_MACRO \\ NEWLINE NEWLINE uintN _hashIndex = (uintN)PROPERTY_CACHE_HASH(obj, id); \\ NEWLINE NEWLINE JSPropertyCacheEntry *_pce = &(cache)->table[_hashIndex]; \\ NEWLINE NEWLINE if (_pce->property) { \\ NEWLINE NEWLINE (cache)->recycles++; \\ NEWLINE NEWLINE if (!JSVAL_IS_INT(_pce->symbolid)) \\ NEWLINE NEWLINE js_DropAtom(cx, (JSAtom *)_pce->symbolid); \\ NEWLINE NEWLINE } \\ NEWLINE NEWLINE _pce->obj = obj; \\ NEWLINE NEWLINE _pce->property = prop; \\ NEWLINE NEWLINE _pce->symbolid = id; \\ NEWLINE NEWLINE if (!JSVAL_IS_INT(id)) \\ NEWLINE NEWLINE js_HoldAtom(cx, (JSAtom *)id); \\ NEWLINE NEWLINE (cache)->empty = JS_FALSE; \\ NEWLINE NEWLINE (cache)->fills++; \\ NEWLINE NEWLINE PR_END_MACRO NEWLINE NEWLINE NEWLINE NEWLINE #define PROPERTY_CACHE_TEST(cache, obj, id, prop) \\ NEWLINE NEWLINE PR_BEGIN_MACRO \\ NEWLINE NEWLINE uintN _hashIndex = (uintN)PROPERTY_CACHE_HASH(obj, id); \\ NEWLINE NEWLINE JSPropertyCacheEntry *_pce = &(cache)->table[_hashIndex]; \\ NEWLINE NEWLINE (cache)->tests++; \\ NEWLINE NEWLINE if (_pce->obj == obj && _pce->symbolid == id) { \\ NEWLINE NEWLINE prop = _pce->property; \\ NEWLINE NEWLINE } else { \\ NEWLINE NEWLINE (cache)->misses++; \\ NEWLINE NEWLINE prop = NULL; \\ NEWLINE NEWLINE } \\ NEWLINE NEWLINE PR_END_MACRO
873|PR_..printf("%d %lld", LONG_MIN, LONGLONG_MIN) doesn't work|Subject says it all, more or less. Buggy source file is prprf.c, function dosprintf: NEWLINE NEWLINE if (u.l < 0) { NEWLINE u.l = -u.l; /* BUG: overflow if u.l equals LONG_MIN */ NEWLINE flags \|= _NEG; NEWLINE } NEWLINE NEWLINE Same problem for long longs, a few lines further down. NEWLINE NEWLINE Th
874|addrbk.c should be part of MOZ_MAIL_NEWS|Submitted by jgk@jgk.org (bugzilla won't let me cc that address). litestubs.c NEWLINE and menu.c are apparently also not needed.
875|Bogus "property is not defined" errors|The lookup code for the JavaScript property cache is wrong. False cache hits are possible when a property is cached as PROP_NOT_FOUND. To the user, the bug manifests itself as bogus "property is not defined" errors, in perfectly valid JavaScript programs. NEWLINE NEWLINE NEWLINE NEWLINE This is the buggy code, file jsinterp.h: NEWLINE NEWLINE NEWLINE NEWLINE #define PROPERTY_CACHE_TEST(cache, obj, id, prop) \\ NEWLINE NEWLINE PR_BEGIN_MACRO \\ NEWLINE NEWLINE uintN _hashIndex = (uintN)PROPERTY_CACHE_HASH(obj, id); \\ NEWLINE NEWLINE JSPropertyCacheEntry *_pce = &(cache)->table[_hashIndex]; \\ NEWLINE NEWLINE (cache)->tests++; \\ NEWLINE NEWLINE if (_pce->property && \\ NEWLINE NEWLINE (_pce->property == PROP_NOT_FOUND \|\| NEWLINE NEWLINE /* *** BUG! obj has to be compared every time! *** */ NEWLINE NEWLINE _pce->property->object == obj) && \\ NEWLINE NEWLINE _pce->symbolid == id) { \\ NEWLINE NEWLINE prop = _pce->property; \\ NEWLINE NEWLINE } else { \\ NEWLINE NEWLINE (cache)->misses++; \\ NEWLINE NEWLINE prop = NULL; \\ NEWLINE NEWLINE } \\ NEWLINE NEWLINE PR_END_MACRO NEWLINE NEWLINE NEWLINE NEWLINE The simplest fix is putting an extra field containing the object into the JSPropertyCacheEntry structure. That's what I did and it works. Here is the fixed section of jsinterp.h: NEWLINE NEWLINE NEWLINE NEWLINE typedef struct JSPropertyCacheEntry { NEWLINE NEWLINE JSObject *obj; /* weak link to object */ NEWLINE NEWLINE JSProperty *property; /* weak link to property */ NEWLINE NEWLINE jsval symbolid; /* strong link to name atom, or index jsval */ NEWLINE NEWLINE } JSPropertyCacheEntry; NEWLINE NEWLINE NEWLINE NEWLINE typedef struct JSPropertyCache { NEWLINE NEWLINE JSPropertyCacheEntry table[PROPERTY_CACHE_SIZE]; NEWLINE NEWLINE JSBool empty; NEWLINE NEWLINE uint32 fills; NEWLINE NEWLINE uint32 recycles; NEWLINE NEWLINE uint32 tests; NEWLINE NEWLINE uint32 misses; NEWLINE NEWLINE uint32 flushes; NEWLINE NEWLINE uint32 pflushes; NEWLINE NEWLINE } JSPropertyCache; NEWLINE NEWLINE NEWLINE NEWLINE #define PROP_NOT_FOUND ((JSProperty *)1) NEWLINE NEWLINE #define PROP_FOUND(prop) ((prword)(prop) & (prword)-2) NEWLINE NEWLINE NEWLINE NEWLINE #define PROPERTY_CACHE_FILL(cx, cache, obj, id, prop) \\ NEWLINE NEWLINE PR_BEGIN_MACRO \\ NEWLINE NEWLINE uintN _hashIndex = (uintN)PROPERTY_CACHE_HASH(obj, id); \\ NEWLINE NEWLINE JSPropertyCacheEntry *_pce = &(cache)->table[_hashIndex]; \\ NEWLINE NEWLINE if (_pce->property) { \\ NEWLINE NEWLINE (cache)->recycles++; \\ NEWLINE NEWLINE if (!JSVAL_IS_INT(_pce->symbolid)) \\ NEWLINE NEWLINE js_DropAtom(cx, (JSAtom *)_pce->symbolid); \\ NEWLINE NEWLINE } \\ NEWLINE NEWLINE _pce->obj = obj; \\ NEWLINE NEWLINE _pce->property = prop; \\ NEWLINE NEWLINE _pce->symbolid = id; \\ NEWLINE NEWLINE if (!JSVAL_IS_INT(id)) \\ NEWLINE NEWLINE js_HoldAtom(cx, (JSAtom *)id); \\ NEWLINE NEWLINE (cache)->empty = JS_FALSE; \\ NEWLINE NEWLINE (cache)->fills++; \\ NEWLINE NEWLINE PR_END_MACRO NEWLINE NEWLINE NEWLINE NEWLINE #define PROPERTY_CACHE_TEST(cache, obj, id, prop) \\ NEWLINE NEWLINE PR_BEGIN_MACRO \\ NEWLINE NEWLINE uintN _hashIndex = (uintN)PROPERTY_CACHE_HASH(obj, id); \\ NEWLINE NEWLINE JSPropertyCacheEntry *_pce = &(cache)->table[_hashIndex]; \\ NEWLINE NEWLINE (cache)->tests++; \\ NEWLINE NEWLINE if (_pce->obj == obj && _pce->symbolid == id) { \\ NEWLINE NEWLINE prop = _pce->property; \\ NEWLINE NEWLINE } else { \\ NEWLINE NEWLINE (cache)->misses++; \\ NEWLINE NEWLINE prop = NULL; \\ NEWLINE NEWLINE } \\ NEWLINE NEWLINE PR_END_MACRO
876|Artifacts on TextBox during resize|I resized the Bug Query page while it was loading, and the "Additional NEWLINE Comments" textbox ended up with big grey artifact bars that didn't go NEWLINE away until a redraw. (Windows problem?)
877|blank bug|Marking Verified/Invalid.
878|BUILD: modules/libimg/src/png_png.cpp and config'ed --with-png|When trying to compile the above file and config'ed with --with-png, various odd NEWLINE parse errors show up. This comes from something else includeing <setjmp.h>, NEWLINE which libpng 1.0.1 and up at least include in pngconf.h
879|BUILD: modules/libimg/src/png_png.cpp and config'ed --with-png|When trying to compile the above file and config'ed with --with-png, various odd NEWLINE parse errors show up. This comes from something else includeing <setjmp.h>, NEWLINE which libpng 1.0.1 and up at least include in pngconf.h
880|[PP]WaitForThisThread forever in ReadWriteProc|Version: 1998-09-04 and before NEWLINE Platform: Macintosh, MacOS 7.x, 8.x NEWLINE Module: NSPR, machine dependent code NEWLINE NEWLINE The following code works fine, most of the times: NEWLINE (http://cvs-mirror.mozilla.org/webtools/lxr/source/nsprpub/pr/src/md/mac/macio.c NEWLINE #266) NEWLINE NEWLINE 266 if (op == READ_ASYNC) NEWLINE 267 (void) PBReadAsync(&pbAsync); NEWLINE 268 else NEWLINE 269 (void) PBWriteAsync(&pbAsync); NEWLINE 270 NEWLINE 271 WaitOnThisThread(me, PR_INTERVAL_NO_TIMEOUT); NEWLINE NEWLINE Sometimes, however, it waits on this thread forever. NEWLINE Couldn't figger out why. Not yet. The following modifications NEWLINE work fine for me, but eliminate the context switch: NEWLINE NEWLINE 266 if (op == READ_ASYNC) NEWLINE 267 (void) PBReadSync(&pbAsync); /* now synchronous */ NEWLINE 268 else NEWLINE 269 (void) PBWriteSync(&pbAsync); /* now synchronous */ NEWLINE 270 NEWLINE 271 /* WaitOnThisThread(me, PR_INTERVAL_NO_TIMEOUT); not needed */ NEWLINE 272 me->io_pending = PR_FALSE; /* nothing pending, io done */ NEWLINE NEWLINE Cheers, NEWLINE Peter Hanappe NEWLINE hanappe@ircam.fr
881|BUILD: error building with glibc/gcc 2.8.1|The file lib/xp/xp_trace.c does not compiles when using glibc and gcc 2.8.1 NEWLINE NEWLINE The error is documented in glibc FAQ: NEWLINE NEWLINE -------------------------------------------------------------------------- NEWLINE 3.9. I get compiler messages "Initializer element not constant" with NEWLINE stdin/stdout/stderr. Why? NEWLINE NEWLINE {RM,AJ} Constructs like: NEWLINE static FILE *InPtr = stdin; NEWLINE NEWLINE lead to this message. This is correct behaviour with glibc since stdin is NEWLINE not a constant expression. Please note that a strict reading of ISO C does NEWLINE not allow above constructs. NEWLINE -------------------------------------------------------------------------- NEWLINE NEWLINE Since the variable real_stderr gets initialized in cmd/xfe/mozilla.c:main NEWLINE and in cmd/xfe/src/context_funcs.cpp, there is no point in setting its value NEWLINE here. NEWLINE NEWLINE NOTE: don't ask me why the same assignment in a C++ module (as NEWLINE context_funcs.cpp) works fine :( NEWLINE NEWLINE Index: xp_trace.c NEWLINE =================================================================== NEWLINE RCS file: /cvsroot/mozilla/lib/xp/xp_trace.c,v NEWLINE retrieving revision 3.1 NEWLINE diff -b -r3.1 xp_trace.c NEWLINE 65c65 NEWLINE < FILE *real_stderr = stderr; NEWLINE --- NEWLINE > FILE *real_stderr = NULL;
882|nsButton not initializing mRefCnt|Kevin, NEWLINE NEWLINE Purify is complaining that the nsButton AddRef() member function is making an NEWLINE uninitialized memory reference to mRefCnt. NEWLINE NEWLINE That's because the Windows version of nsButton.h uses NS_DECL_ISUPPORTS which NEWLINE gets a mRefCnt member data declared, but the constructor doesn't use NEWLINE NS_INIT_REFCNT() to initialize the member data NEWLINE NEWLINE I would have added that, but that isn't sufficient because nsBaseWidget also NEWLINE uses NS_DECL_ISUPPORTS which means that nsButton really has two mRefCnt member NEWLINE data. That's bad and needs to be fixed NEWLINE NEWLINE Troy
883|BUILD: Mozilla does not build on ARM Linux|Building Mozilla on an ARM Linux box fails (tested on Corel NetWinder). NEWLINE NEWLINE The file mozilla/nsprpub/pr/include/md/_linux.h does not contain the NEWLINE definitions for _MD_GET_SP, _MD_SET_FP, etc. NEWLINE NEWLINE Also, gcc on ARM Linux defines, by default, characters as unsigned, while NEWLINE an explicit manifest for all Linux platforms defines HAVE_SIGNED_CHAR. NEWLINE So, either the manifest must be undefined or an explicit compilation flag NEWLINE must be added. NEWLINE NEWLINE The enclosed file contains the changes to make the build work NEWLINE (I also added extra flags for C++ to overcome a current limitation of NetWinder NEWLINE standard libraries - they shouldn't hurt, assumed Mozilla does not use C++ NEWLINE exceptions and rtti, but they can be removed when these libraries get fixed). NEWLINE NEWLINE With these changes Mozilla correctly builds on NetWinder using LessTif. NEWLINE moz-export crashes while loading, but mozilla-export start and can be used NEWLINE for a while (then it crashes or hung the system).
884|BUILD: messages about OSF in Linux build|The module is supposed to be for Linux but contains references about OSF NEWLINE (purely cosmetic change) NEWLINE NEWLINE Index: nsprpub/pr/src/md/unix/linux.c NEWLINE =================================================================== NEWLINE RCS file: /cvsroot/mozilla/nsprpub/pr/src/md/unix/linux.c,v NEWLINE retrieving revision 3.3 NEWLINE diff -b -r3.3 linux.c NEWLINE 85c85 NEWLINE < /* These functions should not be called for OSF1 */ NEWLINE --- NEWLINE > /* These functions should not be called for Linux */ NEWLINE 89c89 NEWLINE < PR_NOT_REACHED("_MD_YIELD should not be called for OSF1."); NEWLINE --- NEWLINE > PR_NOT_REACHED("_MD_YIELD should not be called for Linux."); NEWLINE 101c101 NEWLINE < PR_NOT_REACHED("_MD_CREATE_THREAD should not be called for OSF1."); NEWLINE --- NEWLINE > PR_NOT_REACHED("_MD_CREATE_THREAD should not be called for Linux.");
885|BUILD: compiler warning|The compiler complains about assigning a double to an integer: NEWLINE NEWLINE Index: cmd/xfe/icons/mkicons.cpp NEWLINE =================================================================== NEWLINE RCS file: /cvsroot/mozilla/cmd/xfe/icons/mkicons.cpp,v NEWLINE retrieving revision 3.3 NEWLINE diff -b -r3.3 mkicons.cpp NEWLINE 445c445 NEWLINE < luminance = (0.299 * r) + (0.587 * g) + (0.114 * b); NEWLINE --- NEWLINE > luminance = (int) (0.299 * r) + (0.587 * g) + (0.114 * b);
886|BUILD: compiler warning for missing brakets|Actually, without the brakets the code is also quite hard to read: NEWLINE NEWLINE Index: cmd/xfe/colors.c NEWLINE =================================================================== NEWLINE RCS file: /cvsroot/mozilla/cmd/xfe/colors.c,v NEWLINE retrieving revision 3.4 NEWLINE diff -b -r3.4 colors.c NEWLINE 164a165 NEWLINE > { NEWLINE 165a167 NEWLINE > { NEWLINE 169a172 NEWLINE > } NEWLINE 171a175 NEWLINE > } NEWLINE 172a177 NEWLINE > { NEWLINE 176a182 NEWLINE > }
887|BUILD: moz-export does not build if motif static libraries are not available|If Motif static libraries are not available moz-export does not build and NEWLINE the make breaks (tested with LessTif on NetWinder). NEWLINE NEWLINE Since the configurator has set all the flags and variables to check if NEWLINE these libraries are available, it's quite easy to do the check and use the NEWLINE shared libraries instead.
888|URL unsupported protocol handling|Processing of URLs in the format protocol://uri where the protocol is not NEWLINE supported (i.e. https) is not correct or at least not as I would expect. NEWLINE NEWLINE If the URL is entered manually, a dialog appear with the error: NEWLINE NEWLINE Netscape is unable to locate the server: NEWLINE protocol: NEWLINE The server does not have a DNS entry. NEWLINE NEWLINE If the URL is referred from a page, the browser try to access to a URL NEWLINE formed from the base URL of that page and the "invalid" URL appended to it NEWLINE (i.e. a link from http://www.server.com/home/insecure.html to NEWLINE https://www.server.com/home/secure.html will be expanded into NEWLINE http://www.server.com/home/https://www.server.com/home/insecure.html) NEWLINE NEWLINE NOTE: the same happen with Netscape Communicator 4.04 for Linux/i386 NEWLINE NEWLINE IMHO this behavior generates confusion when dealing with invalid URL NEWLINE (i.e. mispelled http) but specially with a valid, but still unsopported, NEWLINE protocol like https. In this case an error dialog with a message like NEWLINE 'unsupported protocol' or 'protocol not yet supported' would be more NEWLINE appropriate and less confusing.
889|parser not handling illegal content within table|"front" and "back" content is lost. Third cell is placed after the table as a NEWLINE second table with a single cell. NEWLINE NEWLINE ---------------- test case ---------------- NEWLINE NEWLINE <TABLE border="2"> NEWLINE front NEWLINE <TR> NEWLINE <TD> NEWLINE <TABLE border="2"> NEWLINE <TR><TD>1</TD></TR> NEWLINE <P></P> NEWLINE </TABLE> NEWLINE <TABLE border="2"> NEWLINE <TR><TD>2</TD></TR> NEWLINE </TABLE> NEWLINE </TD> NEWLINE <TD> NEWLINE Lorem<BR>ipsum<BR>dolor NEWLINE </TD> NEWLINE </TR> NEWLINE back NEWLINE </TABLE>
890|buglist.cgi: apache changes boundary=ThisRandomString to boundary=thisrandomstring|...but does not, of course, change the printed boundary markers NEWLINE --ThisRandomString and thus Netscape, when receiving the full proper stream of NEWLINE data from the Apache (1.3.something, linux) never sees a boundary (its looking NEWLINE for --thisrandomstring instead of --ThisRandomString). NEWLINE NEWLINE The solution here was to change the printed --ThisRandomString's to NEWLINE --thisrandomstring. NEWLINE NEWLINE I don't know if this is an apache bug or feature or whatever, but I couldn't NEWLINE find it documented, and don't feel like scouring the apache sources when this NEWLINE fix works fine.
891|Segfault on exit.|Using current CVS source. Mozilla always segfaults on exiting (File -> Exit). NEWLINE Happens on both my builds and the mozilla.org builds.
892|Config directory ignored|Using current CVS source. Even if MOZ_USER_DIR is set to something like NEWLINE ".mozilla", it is still saving the preferences to ".netscape". Some of this NEWLINE appears to be hard-coded.
893|forced new line after </NOSCRIPT> when combined with margin-right|In the example pointed to by the URL, one can see an extra new line after the NEWLINE words "ionic concentration." This extra new-line is only present after the NEWLINE active second pair of NOSCRIPT-/NOSCRIPT tags (comment the second pair out and NEWLINE the new line will move down to the next pair). This behavior occurs only when NEWLINE the "ModuleText" class assigned to the DIV enveloping the entire text has the NEWLINE "margin-right" property set to some value.
894|"scroll wheel" on new mice has no functionality|Scroll wheel on newer MicroSoft and Logitech mice has no functionality. Works NEWLINE fine in most other win32 apps, where it manipulates the vertical scrollbar up NEWLINE and down. Am very used to it now and am missing it in Mozilla :)I am using the NEWLINE 9/17/98 daily build. -Matt Hargett (matt@darpanet.net) 9/24/98
895|Reword compilation warning on Win95 build|Under Windows 95, the following warning is issued: NEWLINE NEWLINE "Cannot recursively make on win95 using command.com, use w95make" NEWLINE NEWLINE The wording implies an error occured because w95make should be used but isn't. NEWLINE In truth, it is simply a warning indicating w95make is in use. NEWLINE NEWLINE This message is repeated scores of times during the make on win95, yet it NEWLINE seems to be irrelevant. It should either be removed or reworded to be more NEWLINE correct: "Using w95make instead of recursive make"
896|Default file association questioning|Every time Mozilla is launched, it asks if it should register the default file NEWLINE extension associations. No matter what one answers, the message always pops up, NEWLINE even if the check box is clicked to indicate it not do so. NEWLINE NEWLINE Note that in spite of this, the file associations seem to be set correctly.
897|Fail to load homepage|I have set my homepage preferences to NEWLINE http://www.neptune.net/~bryce/startup.html. This should cause this page to be NEWLINE loaded when Mozilla is first launched, and whenever the home button is pressed. NEWLINE NEWLINE However, Mozilla seems to get hung up between looking up the domain name and NEWLINE making contact with the server, and it never loads the homepage. However, if NEWLINE Stop is pressed and the home button pushed, it will load the homepage just fine.
898|Extra error message when PPP not established|When Mozilla is launched without an established PPP connection, it normally NEWLINE returns an error message stating that the startup page is not accessable. NEWLINE Sometimes, however, an additional error message pops up regarding an rdf server NEWLINE not being available. NEWLINE NEWLINE It would probably be best to combine these two error messages, or prevent the NEWLINE RDF error message from showing.
899|The New Bookmark dialog box does not respond to mouse input|When accessing the New Bookmark command listed in the context menu from the NEWLINE toolbar, a dialog titled "Mozilla User Prompt". NEWLINE NEWLINE This dialog does not respond to mouse signals in any way. The cursor does not NEWLINE even change shape as it passes over the edit box.
900|Listbox size overrun for font preferences|One of the fonts installed on my system is called SPW. This font consists of NEWLINE sheet music symbols, like treble cleff, half note, etc. etc. NEWLINE NEWLINE In the font preferences page, there are a pair of list boxes showing the NEWLINE different fonts as they actually appear. For some reason, SPW is listed as a NEWLINE font, and it appears to be forcing a resize of these controls. Both listboxes NEWLINE are increased height-wise, and this causes the top font listbox to slightly NEWLINE overlap the lower font listbox. NEWLINE NEWLINE It would look better if the spacing between the two listboxes was doubled, of if NEWLINE a better filter mechanism could be found to weed symbolic fonts out of the NEWLINE listing.
901|Auto-refresh page does not auto-refresh|This NASA page is supposed to auto-refresh every 30 seconds. It does in MSIE4, NEWLINE but *not* in NN4.06 or Moz (9/24/98 Win32 build, from MozillaZine.org)
902|Applet problem in Netscape|If I load an applet from Netscape Browser which is calling a Native method NEWLINE through a helper class, after the function is called if I refer the return NEWLINE value from Native call the browser crashes with "Bus Core" error. This NEWLINE behaviour is same in 4.x and 4.5. I checked the same code with "appletviewer" NEWLINE and as well as console application, in both the case it works fine. The same NEWLINE applet works fine if I load it from Netscape running on Windows where the NEWLINE native library is in the form of DLL. The Native library is not using any NEWLINE system call, it is a simple example which returns a "NewStringUTF". NEWLINE NEWLINE This is a potential bug with Netscape browsers. If required I can send the NEWLINE sample programs. NEWLINE NEWLINE Thanks, NEWLINE Swami.S
903|Font Performance Problems|The GetWidth() member functions of nsFontMetricsWin are causing serious NEWLINE performance problems, because each time they make the following calls: NEWLINE - GetDC() NEWLINE - ReleaseDC() NEWLINE - SelectObject() NEWLINE NEWLINE Plus the ftol overhead is a problem as well
904|Implement repeating, prioritised timers on Mac|The timer API and implementation doesn't support repeating timers, and this NEWLINE causes a large number of temporary timer objects to be created and NEWLINE destroyed; e.g., view mgr, throbber, image system services
905|nsFont embedded nsString object|nsFont objects tend to be used on the stack, e.g. by TextFrame in its Paint() NEWLINE member function, and because nsFont contains a nsString object that allocates NEWLINE space from the heap, we end up with a large number of temporary object NEWLINE generated. NEWLINE NEWLINE We either need to not have nsFont allocate the string from the heap, or re-think NEWLINE the usage pattern: e.g., TextFrame is defining a nsFont instance so it can NEWLINE override the decoration that's in the style information
906|CreateFontAliasTable() creates too many nsString objects|CreateFontAliasTable() and its companion AliasFont() end up generating NEWLINE a large number of temporary nsString objects. For example, with demo1 Purify NEWLINE reports 120 nsString objects created and destroyed
907|Too many Selector and SelectorList|There are too many temporary Selector and SelectorList objects created and NEWLINE destroyed during the load of a document. NEWLINE NEWLINE In the case of demo1, Purify reports 468 Selector objects and 174 SelectorList NEWLINE objects. See CSSParserImpl::ParseSelectorGroup(), CSSParserImpl::ParseRuleSet(), NEWLINE and CSSParserImpl::ParseSelectorList()
908|Lots of temporary SupportsArrayImpl objects|The style system is generating lots of temporary SupportsArrayImpl objects NEWLINE during the document load. NEWLINE NEWLINE In the case of demo1 we have 178 SupportsArrayImpl temporary objects allocated NEWLINE by StyleSetImpl::ResolvePseudoStyleFor(), and 101 SupportsArrayImpl temporary NEWLINE objects allocated by StyleSetImpl::ResolveStyleFor()
909|Confusing dialog box|Not sure if I got the right "Component"... NEWLINE NEWLINE Mozilla has a very nice feature (I think it's new...) where it saves the NEWLINE username and password for login sites. It works well, but one of the NEWLINE confirmation dialogs is worded in a confusing manner: NEWLINE NEWLINE ------------------------------------------------------------------------ NEWLINE Confirmation Dialog NEWLINE Do you want to save the username and password contained on this form? NEWLINE NEWLINE [X] If not, do you want to remember this decision? NEWLINE NEWLINE [ OK ] [ Cancel ] NEWLINE ------------------------------------------------------------------------ NEWLINE NEWLINE It's very unclear what's going on here. I have no idea what the check box is NEWLINE supposed to do. If I understand the purpose of this dialog box, might something NEWLINE like this be clearer: NEWLINE NEWLINE ------------------------------------------------------------------------ NEWLINE Confirmation Dialog NEWLINE Do you want to save the username and password for this form? NEWLINE NEWLINE [ ] Do not ask when viewing this page in the future. NEWLINE NEWLINE [ OK ] [ Cancel ] NEWLINE ------------------------------------------------------------------------ NEWLINE NEWLINE or even: NEWLINE NEWLINE ------------------------------------------------------------------------ NEWLINE Confirmation Dialog NEWLINE Do you want to save the username and password for this form? NEWLINE NEWLINE [ Yes ] [ Skip ] [ Never ] [ Help ] NEWLINE ------------------------------------------------------------------------
910|Leaking Bullet Frame|Under certain circumstances the nsBlockFrame code is leaking the bullet frame NEWLINE which means that the list item content object (and all its child content objects NEWLINE are leaking). NEWLINE NEWLINE I looked the bullet frame is handled differently if it's an outside bullet vs. NEWLINE an inside bullet so I'm leaving it to since you know best. NEWLINE NEWLINE Here's some simple HTML that demonstrates the problem: NEWLINE <html> NEWLINE <body> NEWLINE <ol> NEWLINE <li>List item #1</li> NEWLINE </ol> NEWLINE </body> NEWLINE </html> NEWLINE NEWLINE Here's some output from your nifty XP-COM ref count tracking additions: NEWLINE NEWLINE 0[d526a0]: AddRef: dc0c20: 0=>1 [nsHTMLLIElement::QueryInterface NEWLINE NS_NewHTMLLIElement NS_CreateHTMLElement SinkContext::OpenContainer NEWLINE SinkContext::OpenContainer HTMLContentSink::OpenContainer CNavDT NEWLINE 0[d526a0]: AddRef: dc0c2c: 1=>2 [nsGenericHTMLContainerElement::AppendChildTo NEWLINE nsHTMLOListElement::AppendChildTo SinkContext::CloseContainer NEWLINE HTMLContentSink::CloseContainer CNavDTD::CloseContainer C NEWLINE 0[d526a0]: Release: dc0c2c: 2=>1 [SinkContext::CloseContainer NEWLINE HTMLContentSink::CloseContainer CNavDTD::CloseContainer NEWLINE CNavDTD::CloseContainersTo CNavDTD::CloseContainersTo CNavDTD::HandleEndToken N NEWLINE 0[d526a0]: AddRef: dc0c2c: 1=>2 [nsGenericHTMLContainerElement::ChildAt NEWLINE nsHTMLOListElement::ChildAt HTMLStyleSheetImpl::ProcessChildren NEWLINE HTMLStyleSheetImpl::ConstructFrameByDisplayType HTMLStyleShee NEWLINE 0[d526a0]: AddRef: dc0c20: 2=>3 [nsHTMLLIElement::QueryInterface NEWLINE CSSStyleSheetImpl::RulesMatching StyleSetImpl::RulesMatching NEWLINE StyleSetImpl::ResolveStyleFor nsPresContext::ResolveStyleContextFor HTM NEWLINE 0[d526a0]: Release: dc0c2c: 3=>2 [CSSStyleSheetImpl::RulesMatching NEWLINE StyleSetImpl::RulesMatching StyleSetImpl::ResolveStyleFor NEWLINE nsPresContext::ResolveStyleContextFor HTMLStyleSheetImpl::ConstructFrame NEWLINE 0[d526a0]: AddRef: dc0c20: 2=>3 [nsHTMLLIElement::QueryInterface NEWLINE HTMLStyleSheetImpl::RulesMatching StyleSetImpl::RulesMatching NEWLINE StyleSetImpl::ResolveStyleFor nsPresContext::ResolveStyleContextFor HT NEWLINE 0[d526a0]: Release: dc0c2c: 3=>2 [HTMLStyleSheetImpl::RulesMatching NEWLINE StyleSetImpl::RulesMatching StyleSetImpl::ResolveStyleFor NEWLINE nsPresContext::ResolveStyleContextFor HTMLStyleSheetImpl::ConstructFram NEWLINE 0[d526a0]: AddRef: dc0c20: 2=>3 [nsHTMLLIElement::QueryInterface NEWLINE HTMLCSSStyleSheetImpl::RulesMatching StyleSetImpl::RulesMatching NEWLINE StyleSetImpl::ResolveStyleFor nsPresContext::ResolveStyleContextFor NEWLINE 0[d526a0]: Release: dc0c2c: 3=>2 [HTMLCSSStyleSheetImpl::RulesMatching NEWLINE StyleSetImpl::RulesMatching StyleSetImpl::ResolveStyleFor NEWLINE nsPresContext::ResolveStyleContextFor HTMLStyleSheetImpl::ConstructF NEWLINE 0[d526a0]: AddRef: dc0c2c: 2=>3 [nsFrame::nsFrame NEWLINE nsSplittableFrame::nsSplittableFrame nsContainerFrame::nsContainerFrame NEWLINE nsHTMLContainerFrame::nsHTMLContainerFrame nsBlockFrame::nsBlockFrame NS_Ne NEWLINE 0[d526a0]: AddRef: dc0c2c: 3=>4 [nsFrame::nsFrame BulletFrame::BulletFrame NEWLINE nsBlockFrame::Init HTMLStyleSheetImpl::ConstructFrameByDisplayType NEWLINE HTMLStyleSheetImpl::ConstructFrame HTMLStyleSheetImpl:: NEWLINE 0[d526a0]: Release: dc0c2c: 4=>3 [HTMLStyleSheetImpl::ProcessChildren NEWLINE HTMLStyleSheetImpl::ConstructFrameByDisplayType NEWLINE HTMLStyleSheetImpl::ConstructFrame HTMLStyleSheetImpl::ContentAppended StyleSet NEWLINE 0[d526a0]: AddRef: dc0c2c: 3=>4 [nsFrame::GetContent NEWLINE BulletFrame::SetListItemOrdinal nsBlockFrame::RenumberLists NEWLINE nsBlockFrame::ReflowAround nsInlineReflow::ReflowFrame NEWLINE nsInlineReflow::ReflowFrame n NEWLINE 0[d526a0]: AddRef: dc0c20: 4=>5 [nsHTMLLIElement::QueryInterface NEWLINE BulletFrame::SetListItemOrdinal nsBlockFrame::RenumberLists NEWLINE nsBlockFrame::ReflowAround nsInlineReflow::ReflowFrame nsInlineReflow::R NEWLINE 0[d526a0]: Release: dc0c2c: 5=>4 [BulletFrame::SetListItemOrdinal NEWLINE nsBlockFrame::RenumberLists nsBlockFrame::ReflowAround NEWLINE nsInlineReflow::ReflowFrame nsInlineReflow::ReflowFrame nsBlockFrame::Reflow NEWLINE 0[d526a0]: Release: dc0c2c: 4=>3 [BulletFrame::SetListItemOrdinal NEWLINE nsBlockFrame::RenumberLists nsBlockFrame::ReflowAround NEWLINE nsInlineReflow::ReflowFrame nsInlineReflow::ReflowFrame nsBlockFrame::Reflow NEWLINE 0[d526a0]: AddRef: dc0c20: 3=>4 [nsHTMLLIElement::QueryInterface NEWLINE nsBlockFrame::RenumberLists nsBlockFrame::ReflowAround NEWLINE nsInlineReflow::ReflowFrame nsInlineReflow::ReflowFrame nsBlockFrame::ReflowB NEWLINE 0[d526a0]: Release: dc0c2c: 4=>3 [nsBlockFrame::RenumberLists NEWLINE nsBlockFrame::ReflowAround nsInlineReflow::ReflowFrame NEWLINE nsInlineReflow::ReflowFrame nsBlockFrame::ReflowBlockFrame nsBlockFrame::ReflowL NEWLINE 0[d526a0]: Release: dc0c2c: 3=>2 [nsFrame::~nsFrame NEWLINE nsSplittableFrame::~nsSplittableFrame nsContainerFrame::~nsContainerFrame NEWLINE nsHTMLContainerFrame::~nsHTMLContainerFrame nsBlockFrame::~nsBlockFrame NEWLINE 0[d526a0]: Release: dc0c2c: 2=>1 NEWLINE [nsGenericHTMLContainerElement::~nsGenericHTMLContainerElement NEWLINE nsHTMLOListElement::~nsHTMLOListElement nsHTMLOListElement::`scalar deleting NEWLINE destructor' nsHTMLOListE NEWLINE NEWLINE If you whittle that done you'll see the unbalanced reference is here: NEWLINE NEWLINE 0[d526a0]: AddRef: dc0c2c: 3=>4 [nsFrame::nsFrame BulletFrame::BulletFrame NEWLINE nsBlockFrame::Init HTMLStyleSheetImpl::ConstructFrameByDisplayType NEWLINE HTMLStyleSheetImpl::ConstructFrame HTMLStyleSheetImpl::
911|Fix(?) for crash when quickly launching multiple windows|Not sure if I have the right component for this one... Pardon if not so. NEWLINE It might belong to the RDF or the Cookies folks. NEWLINE NEWLINE I found a bug that occurs when one opens a number of windows, which may NEWLINE be related to a low memory condition, and somehow to cookies. NEWLINE NEWLINE I've located the bug exactly and have a patch that seems to stop the NEWLINE bug. However, whatever condition happens when this bug crops up is now NEWLINE causing a bug in a different area of the program (I think) so this isn't NEWLINE a total fix. NEWLINE NEWLINE I'm running WinNT4.0 SR3, on a 166MHz Dell Pentium with 64MB RAM and NEWLINE 200MB swap. In addition to WinNT processes and mozilla, I am running NEWLINE cdplayer.exe, MSACCESS, TASKMGR, MSDEV, CRT, and a notepad-like editor. NEWLINE NEWLINE The crash occurs when 4-8 windows are opened in rapid succession (within NEWLINE 30 sec or so). The faster one opens windows, the sooner the error will NEWLINE occur. NEWLINE NEWLINE I was looking at nine web pages on CGI scripts, looking for the NEWLINE slashdot source code. I had run a search on "slashdot.cgi" and opened NEWLINE up a bunch of windows to view the results of the search. Seven of the NEWLINE pages loaded, the final two hadn't loaded up fully at the time of crash. NEWLINE Error was reported "Access violation". NEWLINE NEWLINE Here is the stack trace: NEWLINE NEWLINE ProcessCookiesAndTrustLabels(_ActiveEntry * 0x025ea8a0) line 3739 + 21 bytes NEWLINE net_ProcessFile(_ActiveEntry * 0x025ea8a0) line 1300 + 9 bytes NEWLINE NET_ProcessNet(PRFileDesc * 0x00000000, int 1) line 3334 + 13 bytes NEWLINE net_process_slow_net_timer_callback(void * 0x00000000) line 216 + 9 bytes NEWLINE wfe_ProcessTimeouts(unsigned long 13087538) line 303 + 12 bytes NEWLINE FireTimeout(HWND__ * 0x001201b2, unsigned int 275, unsigned int 777, unsigned NEWLINE long 13087538) line 60 + 9 bytes NEWLINE USER32! 77e7128c() NEWLINE CNetscapeApp::Run() line 1675 + 8 bytes NEWLINE AfxWinMain(HINSTANCE__ * 0x00400000, HINSTANCE__ * 0x00000000, char * NEWLINE 0x00142595, int 1) line 52 + 11 bytes NEWLINE WinMain(HINSTANCE__ * 0x00400000, HINSTANCE__ * 0x00000000, char * 0x00142595, NEWLINE int 1) line 33 + 21 bytes NEWLINE WinMainCRTStartup() line 330 + 57 bytes NEWLINE KERNEL32! 77f1b304() NEWLINE NEWLINE NEWLINE Here's the code snippit where MSVC says the crash occurred: NEWLINE NEWLINE void ProcessCookiesAndTrustLabels( ActiveEntry *ce ) NEWLINE { NEWLINE #define TEN_MINUTES (time_t)(10*60) /* 10 minutes in seconds */ NEWLINE unsigned int i; NEWLINE TrustLabel *ALabel; NEWLINE XP_List *TempTrustList; NEWLINE NEWLINE if ( IsTrustLabelsEnabled() && ce && ce->URL_s) { NEWLINE /* NEWLINE * if the trust label parsing is enabled then look at each cookie NEWLINE * and try to match it to a trust label on the trust list to see NEWLINE * if one matches the cookie NEWLINE */ NEWLINE for(i=0 ;i < ce->URL_s->all_headers.empty_index; i++) { NEWLINE /* look for a cookie field - allow Set-cookie: or Set-Cookie2: - NEWLINE CASE INSENSITIVE COMPARE */ NEWLINE /* >> */ if(!PL_strncasecmp(ce->URL_s->all_headers.key[i],"Set-Cookie", 10)) NEWLINE { NEWLINE NET_SetCookieStringFromHttp(CE_FORMAT_OUT, NEWLINE ce->URL_s,CE_WINDOW_ID, ce->URL_s->address, ce->URL_s->all_headers.value[i]); NEWLINE } NEWLINE } NEWLINE NEWLINE /* Snip */ NEWLINE NEWLINE It died on the if statement line. Here's what the variables were: NEWLINE i = 0 NEWLINE ce->URL_s->all_headers.empty_index = 39256272 NEWLINE ce->URL_s->all_headers.key = 0x001e71e3 NEWLINE NEWLINE .key should be an array, or at least, that's how it's being used in the NEWLINE above code snippit. But MSVC couldn't evaluate the pointer. NEWLINE NEWLINE MSVC crashed. Reloaded MSVC, then launched Mozilla in debug mode, did a search NEWLINE for slashdot.cgi again and started launching off the windows. Got five open NEWLINE before it crashed. NEWLINE NEWLINE Error was: "Unhandled exception in mozilla.exe: 0xC0000005: Access NEWLINE Violation." NEWLINE NEWLINE Crashed on the same line of the same function. This time variables are: NEWLINE i = 0 NEWLINE ce->URL_s->all_headers.empty_index = 3722304989 NEWLINE ce->URL_s->all_headers.key = 0xdddddddd NEWLINE NEWLINE Here's a few lines of relevant assembly code: NEWLINE NEWLINE 007dcd24 mov eax,dword ptr [eax+ecx*4] NEWLINE 007dcd27 push eax NEWLINE 007dcd28 call _PL_strncasecmp (00831644) NEWLINE 007dcd2d add esp,0000000c NEWLINE 007dcd30 test eax,eax NEWLINE 007dcd32 jne ProcessCookiesAndTrustLabels+000000af (007dcd6f) NEWLINE 3740: NET_SetCookieStringFromHttp(CE_FORMAT_OUT, ce->URL_s, NEWLINE CE_WINDOW_ID, ce->URL_s->address, ce->URL_s->all_headers.value[i]); NEWLINE NEWLINE The ce structure pretty much looks like its blank. There's a ton of NEWLINE fields, like "window_chrome", "referer", "username", "password", etc. NEWLINE etc. but nearly all of the fields are set to either 0xdddddddd "", NEWLINE -572662307, 3722304989, or 221. The few that are set to particular NEWLINE values: NEWLINE NEWLINE ce->status = 1 NEWLINE ce->bytes_received = 16534 NEWLINE ce->socket = 0x00000000 NEWLINE ce->con_sock = 0x00000000 NEWLINE ce->local_file = 1 NEWLINE ce->memory_file = 0 NEWLINE ce->protocol = 12 NEWLINE ce->proto_impl = 0x0097d470 NEWLINE ce->con_data = 0x00d45f30 NEWLINE ce->exit_routine = 0x00792cc0 il_netgeturldone(URL_Struct_ *, int, MWContext_ *) NEWLINE ce->window_id = 0x00c5a120 NEWLINE ce->format_out = 2 NEWLINE ce->save_stream = 0x00000000 NEWLINE ce->busy = 1 NEWLINE ce->proxy_conf = 0x00000000 NEWLINE ce->proxy_addr = 0x00000000 NEWLINE ce->socks_host = 0 NEWLINE ce->socks_port = 0 NEWLINE NEWLINE Some Debug output: NEWLINE Created rdf:ht4 NEWLINE www.hax0r.org error=0 h_name=1 task=12 NEWLINE www.kalifornia.com error=0 h_name=1 task=13 NEWLINE sunsite.unc.edu error=0 h_name=1 task=14 NEWLINE hax0r.org error=0 h_name=1 task=15 NEWLINE First-chance exception in mozilla.exe: 0xC0000005: Access Violation. NEWLINE NEWLINE NEWLINE My first impression of what's going on is: NEWLINE - Mozilla, while idle, polls the sockets. NEWLINE - Since net_calling_all_the_time_count != 0 (it's set to 5, in this NEWLINE case) NEWLINE - NET_ProcessNet is called, which allows multiple connections to be NEWLINE processed simultaneously. NEWLINE - ready_fd = NULL, so an attempt to find a socket is made NEWLINE - a bunch of code is run to set up the socket (I think...) NEWLINE - sockets ready for reading are processed one by one NEWLINE - tmpEntry->busy is false, so processing proceeds NEWLINE - ready_fd = 0, and since both tmpEntry->socket and tmpEntry->con_sock NEWLINE are NULL, the else if statement is executed. NEWLINE - The line NEWLINE rv = (*tmpEntry->proto_impl->process)(tmpEntry); NEWLINE evaluates to NEWLINE rv = net_ProcessFile(tmpEntry); NEWLINE and so net_ProcessFile is called. NEWLINE - net_ProcessFile is called for the file "M1AIR9QS.GIF", which is a NEWLINE picture of the USSR flag. NEWLINE - con_data->next_state = 15, which is NET_FILE_DONE NEWLINE - con_data->stream is non-zero, so the macro COMPLETE_STREAM is run. NEWLINE - con_data->next_state is set to NET_FILE_FREE NEWLINE - ProcessCookiesAndTrustLabels is called: NEWLINE - Checks are made: trust labels is enabled, ce is non-null, and NEWLINE ce->URL_s is non-null. All pass. NEWLINE - A loop is made through all the headers (I think?) NEWLINE - Loop runs from 0 to ce->URL_s->all_headers.empty_index, which equals NEWLINE 3722304989. Hmm. Here's the problem. NEWLINE NEWLINE Okay, now to figure out a solution. Obviously this cookie code NEWLINE shouldn't be called in some cases. NEWLINE NEWLINE I don't think it's the pages themselves that's causing the crash, but NEWLINE rather the strain of loading several heavy duty pages all at once. NEWLINE NEWLINE Here are the exact pages I loaded: NEWLINE http://harbor.ecn.purdue.edu/~jacoby/Slashdot_Mailer/ NEWLINE http://www.krazi.org/ NEWLINE http://www.stars.com/vlib/providers/cgi.html NEWLINE http://www.hax0r.org/ NEWLINE http://www.icemall.com/free/free_perl_scripts.html NEWLINE NEWLINE I'm going to load each one in turn and see if one page in particular is NEWLINE causing the problem... NEWLINE NEWLINE Nope. I loaded each page up in a single browser, with no crash. Then I NEWLINE went nuts loading links into new browsers. I loaded up half a dozen NEWLINE links out of the list on NEWLINE http://www.icemall.com/free/free_perl_scripts.html and got the same NEWLINE error that I've been having. NEWLINE NEWLINE Tried again, this time only got four windows opened. I was a tad NEWLINE slower in launching the windows this time. NEWLINE NEWLINE Here's a possible fix. Change the code to look like this: NEWLINE NEWLINE void ProcessCookiesAndTrustLabels( ActiveEntry *ce ) NEWLINE { NEWLINE #define TEN_MINUTES (time_t)(10*60) /* 10 minutes in seconds */ NEWLINE unsigned int i; NEWLINE TrustLabel *ALabel; NEWLINE XP_List *TempTrustList; NEWLINE NEWLINE if ( IsTrustLabelsEnabled() && ce && ce->URL_s NEWLINE && ce->URL_s->all_headers.empty_index != 0xdddddddd NEWLINE ) { NEWLINE NEWLINE The code under the if statement shouldn't really be executed when NEWLINE empty_index is set to such a large number. The program still crashes, NEWLINE but in a different location, and it seems to allow more web pages to NEWLINE load. I think this new way of crashing is not related to the fix I just NEWLINE made, but I'm not certain. I'll submit it as a separate bug report once NEWLINE I have more info on it.
912|we don't support background-position % values yet|among others. pixels works, but since the style context contains only an nscoord NEWLINE (from ye olde days), no other units are possible... NEWLINE NEWLINE The good news, of sorts, that at least the pixel values are scaled so I can fix NEWLINE up the rendering problem orthogonally...
913|X-windows access control problem|When starting Mozilla: NEWLINE NEWLINE Xlib: connection to ":0.0" refused by server NEWLINE Xlib: Client is not authorized to connect to Server NEWLINE Error: Can't open display: :0.0 NEWLINE NEWLINE All other programs can open from the local machine and user NEWLINE even when access control is on. 'xhost +' "solved" the problem. NEWLINE NEWLINE Version: Night buil 980924: mozilla-sparc-sun-solaris2.6.tar.gz NEWLINE NEWLINE Machine: NEWLINE Sun (Sun Microsystems) Ultra 1 NEWLINE Main Memory : 92 MB (and 200MB+ swap) NEWLINE Kernel Version : SunOS Release 5.6 Version Generic_105181-06 NEWLINE [UNIX(R) System V Release 4.0]
914|Opening/closing bookmarks (and history) window. Closing crashes mozilla|Version: Night build 980924: mozilla-sparc-sun-solaris2.6.tar.gz NEWLINE NEWLINE When opening history window or bookmark window first time: NEWLINE NEWLINE assert: line 179, file xfe.c NEWLINE NEWLINE After clicking on most frequent: NEWLINE assert: line 179, file xfe.c NEWLINE assert: line 189, file remstore.c NEWLINE assert: line 82, file remstore.c NEWLINE assert: line 328, file nlcstore.c NEWLINE assert: line 354, file remstore.c NEWLINE assert: line 328, file nlcstore.c NEWLINE assert: line 631, file mcf.c NEWLINE assert: line 7392, file ht.c NEWLINE NEWLINE When closing bookmark window: NEWLINE NEWLINE Bus Error NEWLINE (and Mozilla has crashed... :) NEWLINE NEWLINE Version: Night buil 980924: mozilla-sparc-sun-solaris2.6.tar.gz NEWLINE NEWLINE Machine: NEWLINE Sun (Sun Microsystems) Ultra 1 NEWLINE Main Memory : 92 MB (and 200MB+ swap) NEWLINE Kernel Version : SunOS Release 5.6 Version Generic_105181-06 NEWLINE [UNIX(R) System V Release 4.0]
916|Incorrect handling of 8-bit characters|Some ISO-8859-1 charcters are not displayed correcly if entered directly into NEWLINE the html file. They are however correctly displayed if replaced with å and NEWLINE friends.
917|A bug in the PNG parsing|A bug in the PNG parsing that make the picture look like RGB is seperated on NEWLINE individual rows. NEWLINE NEWLINE This applies to the release build from 20 september 1998.
918|BUILD: 'make clean' does not clean up correctly|A 'make clean' run AFTER changing the environment does not remove all the NEWLINE object files. NEWLINE NEWLINE I was trying to rebuild Mozilla after adding the environment variable NEWLINE USE_PTHREADS=1 and I got compilation errors (undefined symbols and such) NEWLINE on object files used by the internal threads (the previous build was using NEWLINE internal threads). NEWLINE NEWLINE I'm quite sure I run 'make clean' after setting the new environment. NEWLINE NEWLINE [I just find out there is a target 'realclean' - that hopefully removes NEWLINE ALL the object files, so maybe this is not a real problem, if documented]
919|BUILD: ptio.c always uses poll() and related macros|If building with USE_PTHREADS=1 the module nsprpub/pr/src/pthreads/ptio.c NEWLINE get compiled with a call to _MD_unix_map_poll_error (macro NEWLINE _PR_MD_MAP_POLL_ERROR) that is currently undefined for linux NEWLINE (see nsprpub/pr/include/md/_linux.h). NEWLINE NEWLINE The code should probably check for _PR_USE_POLL.
920|Problems with Plugins inside Layers in Navigator?|Hi! NEWLINE NEWLINE I recently sent the following mail to devinfo@netscape.com: NEWLINE NEWLINE Hi! NEWLINE I'm writing a window application in JS/DHTML that utilizes plugins within NEWLINE layers. NEWLINE I'm having problems with the stacking order of plugin windows inside layers. The NEWLINE plugins inside the layer is not placed NEWLINE within the layer container, instead it seems like the plugin window "floats" NEWLINE above, and when two layers with plugin NEWLINE windows are placed on top of eachother, the both plugin windows "fight" for the NEWLINE right to be viewed, that is, they flickers a lot, NEWLINE like 40 times per second when they trade places (in view order). NEWLINE I have tried several ways of using plugins (EMBED and OBJECT tag) in layers but NEWLINE it is does not work either way, as NEWLINE well as tried several different type of plugins but it seem to be a general NEWLINE error with the way Netscape handle plugin OLE NEWLINE windows. NEWLINE I've tried on different versions of Netscape Navigator on Windows 95/NT4.0 SP3, NEWLINE from version 4.02 to 4.06 and it is the same NEWLINE problem. NEWLINE Please let me know if there is any way to work around this problem. NEWLINE With best regards NEWLINE Per Andersson NEWLINE Cherry Casino Group NEWLINE NEWLINE And after that I received the following mail: NEWLINE NEWLINE Per, NEWLINE Thank you for your inquiry. NEWLINE To Submit Code Source Bugs please go to: http://www.mozilla.org/bugs/. NEWLINE For further assistance, please contact the Developer Response Center at NEWLINE 650.937.2986. NEWLINE Thanks, NEWLINE Lionel Edgar NEWLINE Developer Relations Representative NEWLINE Netscape Communications Corporation NEWLINE NEWLINE And here I am. I hope I'm at the right place. The bug report is as above, NEWLINE nothing more to add to it. Please let me know if this lead to anything (like a NEWLINE fix in version 2.5 or 4.6). NEWLINE NEWLINE With best regards NEWLINE Per Andersson NEWLINE Cherry Casino Group
921|BR tags must have zero height|which means that the way they work must be redone.
922|whitespace should not render|from Al Vining <al@heimdallr.u-net.com> on netscape.public.mozilla.layout NEWLINE <TABLE border="2"> NEWLINE <TR><TD>1</TD></TR> NEWLINE </TABLE> NEWLINE and: NEWLINE <TABLE border="2"> NEWLINE <TR> NEWLINE <TD> NEWLINE 1 NEWLINE </TD> NEWLINE </TR> NEWLINE </TABLE> NEWLINE don't look the same? If an author had wanted the space, they could NEWLINE have used NEWLINE NEWLINE ================================================================== NEWLINE NEWLINE from Angus: NEWLINE NEWLINE Whitespace shouldn't be significant in this case. This appears to be a bug in NEWLINE Nav. NEWLINE NEWLINE This area of Nav compatibility is not desirable for content developers (since NEWLINE content isn't depending on this bug) and should be deprecated. Actually, I get NEWLINE mail from content developers asking us to fix this bug in Nav all the time - NEWLINE people want to know why there is extra space in their tables, etc. NEWLINE NEWLINE Let's ignore the whitespace as HTML 4.0 intended; this is not a bug we should NEWLINE carry over from Navigator.
923|<font> reporting wrong metrics to table - 0 maxElementSize|content inside of table cells that are wrapped in a <FONT> report back 0 NEWLINE maxElementSize. This confuses tables horribly. NEWLINE NEWLINE see test case below. NEWLINE ================================================================= NEWLINE <TABLE BORDER WIDTH="70%" BGCOLOR="#F08080" > NEWLINE <TR> NEWLINE <TD><FONT SIZE=+2>Tag (Attribute)</FONT></TD> NEWLINE </TR> NEWLINE </table>
924|Crash after loading PNG test icon suite|Mozilla crashes while loading the page of PNG test icons located at: NEWLINE NEWLINE http://www.cdrom.com/pub/png/pngsuite.html
925|floating tables don't float|amny of the major head links (like "Recreation") off of yahoo! are broken. Here NEWLINE is a reduced test case that shows the pattern: NEWLINE NEWLINE ==================================================== NEWLINE <html><body> NEWLINE <table cellpadding=1 cellspacing=0 align=right border=0><tr><td> NEWLINE <tr><td align=center bgcolor="#eeeeee">here<br>is<br>some<br>text</td></tr> NEWLINE </table> NEWLINE <form method=get action="http://search.yahoo.com/bin/search"> NEWLINE <input type=submit value=Search> NEWLINE </form> NEWLINE </body></html>
926|The style attribute with a CSS "text-decoration" value is numeric|The application is not reflecting the string value for text decorations NEWLINE NEWLINE 1) Save off example 0 or any html document with a NEWLINE "text-decoration: underline", "text-decoration: line-through" or NEWLINE "text-decoration: overline" style value. NEWLINE 2) Notice that the values for attributes have "text-decoration: 1", NEWLINE "text-decoration: 2" or "text-decoration: 4"
927|crash in image group code|I get an intermittent code clicking around in http://www.wb.com, particularly NEWLINE when I get to http://www.wbmovies.com/main.html by clicking on the middle right NEWLINE image (today it says "Practical Magic".) Then click on "Feature Presentation", NEWLINE this opens a second window and I click around. Eventually I get the crash in NEWLINE image group. NEWLINE NEWLINE NEWLINE ======================================================================= NEWLINE stack: NEWLINE ReconnectHack(void * 0x0134aa40, nsIStreamListener * 0x013fca70) line 125 + 9 NEWLINE bytes NEWLINE ImageNetContextImpl::GetURL(ilIURL * 0x013fcca0, NET_ReloadMethod NEWLINE NET_DONT_RELOAD, ilINetReader * 0x013fca40) line 463 + 30 bytes NEWLINE il_image_complete(il_container_struct * 0x0135c1b0) line 1338 NEWLINE process_buffered_gif_input_data(gif_struct * 0x012a3120) line 637 + 9 bytes NEWLINE gif_delay_time_callback(void * 0x012a3120) line 657 + 9 bytes NEWLINE timer_callback(nsITimer * 0x014a2b60, void * 0x014a2bb0) line 70 + 12 bytes NEWLINE TimerImpl::Fire(unsigned long 16282523) line 308 + 17 bytes NEWLINE TimerImpl::ProcessTimeouts(unsigned long 16282523) line 187 NEWLINE FireTimeout(void * 0x00000000, unsigned int 275, unsigned int 26756, unsigned NEWLINE long 16282523) line 101 + 9 bytes NEWLINE USER32! 77e7128c() NEWLINE main(int 2, char * * 0x011f6570) line 96 NEWLINE mainCRTStartup() line 338 + 17 bytes NEWLINE KERNEL32! 77f1b304()
928|Multiple windows crash - fix|Don't know if NetLib is the right component or not... NEWLINE NEWLINE I'm running WinNT4.0 SR3, on a 166MHz Dell Pentium with 64MB RAM and NEWLINE 200MB swap. In addition to WinNT processes and mozilla, I am running NEWLINE cdplayer.exe, MSACCESS, TASKMGR, MSDEV, CRT, and a notepad-like editor. NEWLINE NEWLINE This bug occurs after multiple windows are opened. For example, opening 4-8 NEWLINE links in a new window. The bug seems to occur 15-30 sec after opening the last NEWLINE window. NEWLINE NEWLINE Here's the stack trace (sorry, kinda long): NEWLINE NEWLINE strlen() line 66 NEWLINE cvt_s(SprintfStateStr * 0x0012f328, char * 0xdddddddd, int 0x00000000, int NEWLINE 0x0000076c, int 0x00000001) line 369 + 19 bytes NEWLINE dosprintf(SprintfStateStr * 0x0012f328, char * 0x00952bc0, char * 0x0012f584) NEWLINE line 971 + 25 bytes NEWLINE PR_vsnprintf(char * 0x0012f368, unsigned int 0x00000200, char * 0x00952b68, char NEWLINE * 0x0012f574) line 1180 + 17 bytes NEWLINE _MK_TraceMsg(char * 0x00952b68) line 86 + 28 bytes NEWLINE NET_ProcessNet(PRFileDesc * 0x00000000, int 0x00000001) line 3451 + 82 bytes NEWLINE net_process_slow_net_timer_callback(void * 0x00000000) line 216 + 9 bytes NEWLINE wfe_ProcessTimeouts(unsigned long 0x05f4ea17) line 303 + 12 bytes NEWLINE FireTimeout(HWND__ * 0x000c035c, unsigned int 0x00000113, unsigned int NEWLINE 0x00000309, unsigned long 0x05f4ea17) line 60 + 9 bytes NEWLINE USER32! 77e71ab7() NEWLINE USER32! 77e71a77() NEWLINE NTDLL! 77f7624f() NEWLINE USER32! 77e7288d() NEWLINE USER32! 77e72918() NEWLINE CWnd::DefWindowProcA(unsigned int 0x00000112, unsigned int 0x0000f077, long NEWLINE 0x013603f4) line 938 + 32 bytes NEWLINE CWnd::WindowProc(unsigned int 0x00000112, unsigned int 0x0000f077, long NEWLINE 0x013603f4) line 1523 + 26 bytes NEWLINE AfxCallWndProc(CWnd * 0x00b51470 {CNetscapeView hWnd=0x000608ee}, HWND__ * NEWLINE 0x000608ee, unsigned int 0x00000112, unsigned int 0x0000f077, long 0x013603f4) NEWLINE line 210 + 26 bytes NEWLINE AfxWndProc(HWND__ * 0x000608ee, unsigned int 0x00000112, unsigned int NEWLINE 0x0000f077, long 0x013603f4) line 363 + 25 bytes NEWLINE AfxWndProcBase(HWND__ * 0x000608ee, unsigned int 0x00000112, unsigned int NEWLINE 0x0000f077, long 0x013603f4) line 203 + 21 bytes NEWLINE USER32! 77e7288d() NEWLINE USER32! 77e72918() NEWLINE PaneProc(HWND__ * 0x000608ee, unsigned int 0x00000112, unsigned int 0x0000f077, NEWLINE long 0x013603f4) line 1422 + 26 bytes NEWLINE USER32! 77e71ab7() NEWLINE USER32! 77e71a77() NEWLINE NTDLL! 77f7624f() NEWLINE USER32! 77e7288d() NEWLINE USER32! 77e72918() NEWLINE CWnd::DefWindowProcA(unsigned int 0x000000a1, unsigned int 0x00000007, long NEWLINE 0x013603f4) line 938 + 32 bytes NEWLINE CWnd::WindowProc(unsigned int 0x000000a1, unsigned int 0x00000007, long NEWLINE 0x013603f4) line 1523 + 26 bytes NEWLINE AfxCallWndProc(CWnd * 0x00b51470 {CNetscapeView hWnd=0x000608ee}, HWND__ * NEWLINE 0x000608ee, unsigned int 0x000000a1, unsigned int 0x00000007, long 0x013603f4) NEWLINE line 210 + 26 bytes NEWLINE AfxWndProc(HWND__ * 0x000608ee, unsigned int 0x000000a1, unsigned int NEWLINE 0x00000007, long 0x013603f4) line 363 + 25 bytes NEWLINE AfxWndProcBase(HWND__ * 0x000608ee, unsigned int 0x000000a1, unsigned int NEWLINE 0x00000007, long 0x013603f4) line 203 + 21 bytes NEWLINE USER32! 77e7288d() NEWLINE USER32! 77e72918() NEWLINE PaneProc(HWND__ * 0x000608ee, unsigned int 0x000000a1, unsigned int 0x00000007, NEWLINE long 0x013603f4) line 1422 + 26 bytes NEWLINE USER32! 77e71250() NEWLINE 013603f4() NEWLINE NEWLINE NEWLINE This bug seems to have its start in mkgeturl.c. Here's the code (line 3461): NEWLINE NEWLINE else NEWLINE { NEWLINE /* XP_OS2_FIX IBM-MAS: limit size of URL string to 100 to keep from NEWLINE blowing trace message buffer! */ NEWLINE TRACEMSG(("End of transfer, entry (soc=%d, con=%d) being removed from list NEWLINE with %d status: %-.1900s", NEWLINE tmpEntry->socket, tmpEntry->con_sock, tmpEntry->status, NEWLINE /* >> */ (tmpEntry->URL_s->address ? tmpEntry->URL_s->address : NEWLINE ""))); NEWLINE /* catch out of memory errors at the lowest NEWLINE * level since we don't do it at all the out NEWLINE * of memory condition spots NEWLINE */ NEWLINE if(tmpEntry->status == MK_OUT_OF_MEMORY NEWLINE && !tmpEntry->URL_s->error_msg) NEWLINE { NEWLINE NEWLINE The error seems to crop up as a result of the trace message, so try #ifdef'ing NEWLINE it out: NEWLINE NEWLINE else NEWLINE { NEWLINE /* XP_OS2_FIX IBM-MAS: limit size of URL string to 100 to keep from NEWLINE blowing trace message buffer! */ NEWLINE #ifdef DONTUSE NEWLINE TRACEMSG(("End of transfer, entry (soc=%d, con=%d) being removed from list NEWLINE with %d status: %-.1900s", NEWLINE tmpEntry->socket, tmpEntry->con_sock, tmpEntry->status, NEWLINE (tmpEntry->URL_s->address ? tmpEntry->URL_s->address : NEWLINE ""))); NEWLINE #endif NEWLINE /* catch out of memory errors at the lowest NEWLINE * level since we don't do it at all the out NEWLINE * of memory condition spots NEWLINE */ NEWLINE if(tmpEntry->status == MK_OUT_OF_MEMORY NEWLINE && !tmpEntry->URL_s->error_msg) NEWLINE {
929|Crash with Most Frequented Pages|While messing around with the (docked) Aurora history view, I had clicked on the NEWLINE [+] next to the Most Frequented Pages folder. First time through, I went to NEWLINE debug and got the info listed below. Second time I clicked "Ignore" (three NEWLINE times); eleven sites are listed under this topic. NEWLINE NEWLINE Cause: NEWLINE ------ NEWLINE Line 176 of remstore.c has: NEWLINE XP_ASSERT( (RDF_STRING_TYPE != type) \|\| ( IsUTF8String((const char* )v))); NEWLINE NEWLINE where at time of crash, type = 0x0003. NEWLINE NEWLINE In rdf.h is the define: NEWLINE #define RDF_STRING_TYPE ((RDF_ValueType)0x0003) NEWLINE NEWLINE NEWLINE Abbreviated stack trace: NEWLINE ------------------------ NEWLINE 1. XP_AssertAtLine(char * 0x00944ed4, int 0x000000af) line 2687 NEWLINE 2. remoteStoreAdd(RDF_TranslatorStruct * 0x009f8120, RDF_ResourceStruct * NEWLINE 0x00c02120, RDF_ResourceStruct * 0x009f8bf0, void * 0x00c063d0, unsigned short NEWLINE 0x0003, int 0x00000001) line 175 + 86 bytes NEWLINE 3. collateOneHist(RDF_TranslatorStruct * 0x00a42eb0, RDF_ResourceStruct * NEWLINE 0x00aaea20, char * 0x02294c6e, char * 0x02294c4e, long 0x360adc0b, long NEWLINE 0x360adc0b, unsigned long 0x00000001, int 0x00000000) line 124 + 93 bytes NEWLINE 4. collateHistory(RDF_TranslatorStruct * 0x00a42eb0, RDF_ResourceStruct * NEWLINE 0x00aaea20, int 0x00000000) line 99 + 37 bytes NEWLINE 5. HistPossiblyAccessFile(RDF_TranslatorStruct * 0x00a42eb0, RDF_ResourceStruct NEWLINE * 0x00ab1640, RDF_ResourceStruct * 0x009f8950, int 0x00000001) line 618 + 20 NEWLINE bytes NEWLINE <Snip> NEWLINE NEWLINE NEWLINE Debug output: NEWLINE ------------- NEWLINE www.thehistorynet.com error=0 h_name=1 task=31 NEWLINE Warning: no message line prompt for ID 0xF3A8. NEWLINE Warning: no message line prompt for ID 0xF3A8. NEWLINE Warning: no message line prompt for ID 0xF3A9. NEWLINE Warning: no message line prompt for ID 0xF3AA. NEWLINE NEWLINE Created rdf:ht6 NEWLINE Warning: Creating a pane with no CDocument. NEWLINE NEWLINE Created rdf:ht7 NEWLINE Warning: Creating a pane with no CDocument. NEWLINE NEWLINE Created rdf:ht8 NEWLINE Warning: Creating a pane with no CDocument. NEWLINE NEWLINE Created rdf:ht9 NEWLINE Warning: Creating a pane with no CDocument. NEWLINE NEWLINE NEWLINE System: NEWLINE ------- NEWLINE I'm running WinNT4.0 SR3, on a 166MHz Dell Pentium with 64MB RAM and NEWLINE 200MB swap. In addition to WinNT processes and mozilla, I am running NEWLINE cdplayer.exe, MSACCESS, TASKMGR, MSDEV, CRT, and a notepad-like editor.
930|Unordered lists markers do not render correctly when nested|I brought up our quick look test case for lists and the nested list markers are NEWLINE not displaying correctly. For example, I expected to see an unordered list NEWLINE display like this: NEWLINE disc NEWLINE square NEWLINE circle NEWLINE disc NEWLINE square NEWLINE circle NEWLINE NEWLINE in NGLayout it is displayed like this: NEWLINE disc NEWLINE circle NEWLINE square NEWLINE square NEWLINE square NEWLINE square
931|Core dump in HT_ posting a tooltip|I was trying the new rdf tooltips and i got the following core dump. I was NEWLINE mousing over the "Bookmarks For Ramiro Estrugo" button. It was the 2nd time i NEWLINE moused over theat button. It worked the first time. We should purify this NEWLINE code. NEWLINE NEWLINE #0 0x4033f483 in HT_TopNode (view=0xe85344ec) at ht.c:2654 NEWLINE ---Type <return> to continue, or q <return> to quit--- NEWLINE 2654 top = view->top; NEWLINE (gdb) where NEWLINE #0 0x4033f483 in HT_TopNode (view=0xe85344ec) at ht.c:2654 NEWLINE #1 0x400d9452 in XFE_RDFToolbar::tooltipCB (w=0x8615800, NEWLINE client_data=0x8616ea0, string_return=0xbfffd614, NEWLINE need_to_free_string=0xbfffd613 "") at RDFToolbar.cpp:588 NEWLINE #2 0x4020d9d0 in ItemGetTipString (w=0x8615800, xmstr_out=0xbfffd614, NEWLINE need_to_free_out=0xbfffd613 "") at ToolTip.c:618 NEWLINE #3 0x4020e15d in ItemPostToolTip (w=0x8615800) at ToolTip.c:879 NEWLINE #4 0x4020e2f4 in StageTwoTimeout (closure=0x8615800, id=0xbfffd65c) NEWLINE at ToolTip.c:972 NEWLINE #5 0x408dc8ad in XtAppProcessEvent () NEWLINE #6 0x80b9706 in fe_EventLoop () at mozilla.c:1143 NEWLINE #7 0x80bc9c5 in main (argc=1, argv=0xbfffeed4) at mozilla.c:3458 NEWLINE (gdb)
932|Ordered List marker - lower alpha is not displayed correctly|The lower alpha numbering scheme is displayed as upper alpha NEWLINE NEWLINE <OL TYPE=a> NEWLINE <LI>here is a ordered list with the lower alpha style</LI> NEWLINE <LI>here is the second line</LI> NEWLINE </OL>
933|Ordered List marker - lower roman is not displayed correctly|The lower roman numbering scheme is displayed as upper roman. NEWLINE NEWLINE <OL TYPE=i> NEWLINE <LI>here is a ordered list with the lower roman style</LI> NEWLINE <LI>here is the second line</LI> NEWLINE </OL>
934|Unordered lists markers do not render correctly when nested|I expected to see ordered lists displayed like this: NEWLINE arabic NEWLINE lower alpha NEWLINE upper alpha NEWLINE lower roman NEWLINE upper roman NEWLINE -- and then repeat NEWLINE NEWLINE in NGLayout arabic is the only marker displayed for nested lists
935|auto-closing DT/DD doesn't work|According to the HTML4 DTD, the end tags for DT and DD tags are optional. NEWLINE Therefore the NEWLINE following html should be legal: NEWLINE NEWLINE <HTML> NEWLINE <STYLE> NEWLINE </STYLE> NEWLINE <BODY> NEWLINE Before the DL NEWLINE <DL> NEWLINE <DT>Dee-tee NEWLINE <DD>Dee-dee NEWLINE <DT>Dee-tee #2 NEWLINE <DD>Dee-dee #2 NEWLINE </DL> NEWLINE After the DL NEWLINE </BODY> NEWLINE </HTML> NEWLINE NEWLINE However, your parser creates the following (incorrect) content tree: NEWLINE NEWLINE HTML refcount=4< NEWLINE HEAD refcount=2< NEWLINE > NEWLINE BODY refcount=4< NEWLINE Text refcount=3<\\nBefore the DL\\n> NEWLINE DL refcount=3< NEWLINE Text refcount=3<\\n > NEWLINE DT refcount=3< NEWLINE Text refcount=3<Dee-tee\\n > NEWLINE DD refcount=3< NEWLINE Text refcount=3<Dee-dee\\n > NEWLINE > NEWLINE DT refcount=3< NEWLINE Text refcount=3<Dee-tee #2\\n > NEWLINE DD refcount=3< NEWLINE Text refcount=3<Dee-dee #2\\n> NEWLINE > NEWLINE > NEWLINE > NEWLINE > NEWLINE Text refcount=3<\\nAfter the DL\\n\\n\\n> NEWLINE > NEWLINE >
936|Unclickable links|Links are unclickable with the latest cvs versions of the build. The cursor NEWLINE does not change when placed over a link. You can grab the url from the link to NEWLINE the location window. The problem occurs when using XFE but not GNOMEFE nor NEWLINE QTFE. It occurs with all webpages. NEWLINE NEWLINE Build enviornments: NEWLINE Redhat 5.1 Linux, Motif 2.1, egcs 1.0.3a c/c++ compilers, autoconf build NEWLINE plus native jpeg, png & zlib libraries, XFree86 3.3.2p3 NEWLINE Solaris 2.5.1, Motif 1.2.4, gcc 2.7.2.3 c/c++ compilers, autoconf build, NEWLINE X11R6.1
937|Support MacBin encoding for FILE uploads in FORMs|Hello I hope I found the right place for this bug and patch request. NEWLINE NEWLINE Operating system MAC (ie. iMac) NEWLINE Browsers ALL Netscape(3,4) (works perfectly on Mac IE4) NEWLINE NEWLINE On a Netscape / Mac, Http uploads are destroying files with both a data fork and NEWLINE resource fork. Files like a 120k simpletext application get sent up as 0K NEWLINE NEWLINE On IE4 / Mac everything is kept in tact. I think somehow they use MacBinary to NEWLINE their advantage. NEWLINE NEWLINE I would rather have myusers use Netscape. How do we go about getting a patch. NEWLINE I have researched this extensively with no luck, (except learning it is a NEWLINE Netscape error) NEWLINE NEWLINE NEWLINE Thanks NEWLINE NEWLINE Frank LLosa
938|"Bookmarks For Joe User" pane grows everytime its clicked|Click on the "Bookmarks For Joe User" button on the personal toolbar. Let go. NEWLINE Do it again and let go again. Keep doing this and notice how items keep getting NEWLINE added to the pane. The added items are an exact copy of the previous items. NEWLINE NEWLINE Maybe we are handling an unrelated ht event as if it were an add event...
939|Bugzilla login problem|Geez.. ^^^ That's confusing.. Anyway, I've noticed problems with all of the NEWLINE versions of Netscape Navigator that I've used thus far (up to 4.06) that when NEWLINE you do a "Save Link As" there is not only no status, but if you start more than NEWLINE one at a time they tend to crash -- You can't close the windows and they don't NEWLINE close themselves. But, the navigator window still remains available. NEWLINE I've experienced this problem on a PC running windows 95 with a 28.8 NEWLINE connection. NEWLINE NEWLINE Another thing I noticed is that in the logon for this thingamabob where I'm NEWLINE writing you this note, if you enter a valid E-Mail address and an incorrect NEWLINE password, you get a "Server Error". You also get this same error if you enter NEWLINE and Email address and password, and then click to have you mail the password. NEWLINE NEWLINE Hope that this can in some way be useful to you guys. Keep up all of the NEWLINE good work! :) NEWLINE NEWLINE - jim.
940|window.location.href: URL doesn't load second file|loading this URL runs a short animation, then should jump to NEWLINE http://www.cyberworks.net/discover-frame.html. NGL does not jump to the second NEWLINE URL, it just cycles the animation. NEWLINE This bug is important because cyberworks wants to imbed NGL in an upcoming NEWLINE product.
941|viewer won't scroll to end of page|it's as if the document isn't really telling the view the correct height. The NEWLINE viewer only scrolls about 80% of the way down the page.
942|layout is incorrect on this page|notice the top images, and the rightmost (solid black) column
943|Some strange layout behavior with img tags and framesets.|5.01a - 98092801 NEWLINE NEWLINE macos system 8.1 - 7200/90 NEWLINE NEWLINE See this url for tests that I've posted to explain: NEWLINE http://www.emeraldnet.net/~corvin/mozilla/ NEWLINE img_test/index.html NEWLINE NEWLINE The images start out being centered but are then NEWLINE bumped over to the left. NEWLINE NEWLINE Ones with longer alts are not bumped to the left as NEWLINE much. - Their centering seems effected by the alt NEWLINE tag. NEWLINE NEWLINE If put together in a frameset you can see the effect NEWLINE more clearly. - see the last goup of three links NEWLINE for this.
944|FIELDSET and LEGEND don't work|See URL for testcase. IE5 seems to implement FIELDSET and LEGEND properly. NEWLINE NEWLINE This is a necessary feature to do HTML-based dialogs for our XPFE.
945|LABEL doesn't work|See URL for testcase of explicit label associations with an input element. Note NEWLINE this test case does not test for implicit associations, of the fashion: NEWLINE NEWLINE <label> NEWLINE Yes NEWLINE <input type="radio" id="decision" value="yes"> NEWLINE </label> NEWLINE <label> NEWLINE No NEWLINE <input type="radio" id="decision" value="no"> NEWLINE </label> NEWLINE NEWLINE LABEL is required for, among other things, HTML dialogs in an XPFE.
946|non breaking space formats incorrectly in table w/anchors|the following snippet formats as NEWLINE NEWLINE test NEWLINE test 2 NEWLINE test NEWLINE test 2 NEWLINE NEWLINE <table> NEWLINE <tr> NEWLINE <td> NEWLINE <a href=foo> test</a><br> NEWLINE <a href=foo> test 2</a><br> NEWLINE <a href=foo> test</a><br> NEWLINE <a href=foo> test 2</a><br> NEWLINE </td> NEWLINE </tr> NEWLINE </table>
947|Table cell content overrides cell borders|View above URL, scroll to 2nd table NEWLINE Cell content in the first column overrides the border. It seems that the column NEWLINE width is acting like it is fixed and not relative to the content.
948|vertical margins should be collapse across generations|Vertical margins should collapse between block-level elements -- even when the NEWLINE elements aren't siblings. It's only if ther is a padding and/or border between NEWLINE them that there should be no collaping.
949|margins should be transparent|The margin areas of an elemetn is transparent and the underlying element should NEWLINE shine through. The example shows a slightly odd use of this. It seems like NEWLINE NGlayout adjustes the size of the enclosing DIV element to also include the NEWLINE margin of its child elements. This is incorrect.
950|ASSERT in CAbstractCX via mkgeturl.c|Access Violation crash while viewing CNN & Slashdot. I had just closed NEWLINE a window. I have a lot of programs loaded into memory. NEWLINE NEWLINE Call Stack: NEWLINE ----------- NEWLINE CAbstractCX::IsFrameContext() line 185 + 3 bytes NEWLINE GetFrame(MWContext_ * 0x00ba5710) line 112 + 24 bytes NEWLINE FE_URLEcho(URL_Struct_ * 0x00c26230, int 0x00000001, MWContext_ * 0x00ba5710) NEWLINE line 47 + 14 bytes NEWLINE net_CallExitRoutine(void (URL_Struct_ *, int, MWContext_ *)* 0x00792cc0 NEWLINE il_netgeturldone(URL_Struct_ *, int, MWContext_ *), URL_Struct_ * 0x00c26230, NEWLINE int 0x00000001, int 0x00000002, MWContext_ * 0x00ba5710) line 1082 + 17 bytes NEWLINE NET_ProcessNet(PRFileDesc * 0x00000000, int 0x00000001) line 3472 + 39 bytes NEWLINE net_process_slow_net_timer_callback(void * 0x00000000) line 216 + 9 bytes NEWLINE wfe_ProcessTimeouts(unsigned long 0x0b07a289) line 303 + 12 bytes NEWLINE FireTimeout(HWND__ * 0x006f031e, unsigned int 0x00000113, unsigned int NEWLINE 0x00000309, unsigned long 0x0b07a289) line 60 + 9 bytes NEWLINE USER32! 77e7128c() NEWLINE CNetscapeApp::Run() line 1675 + 8 bytes NEWLINE AfxWinMain(HINSTANCE__ * 0x00400000, HINSTANCE__ * 0x00000000, char * NEWLINE 0x00142595, int 0x00000001) line 52 + 11 bytes NEWLINE WinMain(HINSTANCE__ * 0x00400000, HINSTANCE__ * 0x00000000, char * 0x00142595, NEWLINE int 0x00000001) line 33 + 21 bytes NEWLINE WinMainCRTStartup() line 330 + 57 bytes NEWLINE NEWLINE Cause: NEWLINE ------ NEWLINE Here is the assert that failed, on line 194 of cxabstra.h, a member NEWLINE function of CAbstractCX: NEWLINE NEWLINE BOOL IsFrameContext() const { NEWLINE /* >> */ ASSERT(m_cxType); NEWLINE switch (m_cxType) { NEWLINE case Window: NEWLINE return TRUE; NEWLINE NEWLINE default: NEWLINE return FALSE; NEWLINE } NEWLINE } NEWLINE NEWLINE Tracing this back, an appropriate place to address this problem NEWLINE might be net_CallExitRoutine, in mkgeturl.c: NEWLINE NEWLINE #if defined(XP_WIN) \|\| defined (XP_MAC) \|\| defined (XP_OS2) NEWLINE FE_URLEcho(URL_s, status, window_id); NEWLINE #endif /* XP_WIN/MAC/OS2 */ NEWLINE NEWLINE I guess you'd want to add an if statement to only call the FE_URLEcho NEWLINE function if window_id is valid. In this case, the window_id pointer NEWLINE is okay (it equals 0x00ba5710), but its contents are all set to NEWLINE 0xdddddddd or some such. I'm not sure how you'd construct the if NEWLINE statement to address this... NEWLINE NEWLINE NEWLINE System: NEWLINE ------- NEWLINE I'm running WinNT4.0 SR3, on a 166MHz Dell Pentium with 64MB RAM and NEWLINE 200MB swap. In addition to WinNT processes and mozilla, I am running NEWLINE cdplayer.exe, MSACCESS, TASKMGR, MSDEV, CRT, PowerPoint, Netscape 4.06, NEWLINE and a notepad-like editor.
951|ASSERT while freeing memory in HT_DeleteCursor|My system was swapping pretty heavily because I was cutting and pasting NEWLINE large images between applications (a 23MB image in a word document). NEWLINE Had a bunch of software loaded, too. While waiting for swapping to NEWLINE finish, I switched over to Mozilla to read slashdot. Mozilla started NEWLINE having trouble. NEWLINE NEWLINE Debug messages: NEWLINE --------------- NEWLINE Created rdf:ht5 NEWLINE Remembering document position at element id 72 NEWLINE www.netlib.org error=0 h_name=1 task=18 NEWLINE Remembering document position at element id 35 NEWLINE Updating location bar for named anchor NEWLINE banner.surfree.com error=0 h_name=1 task=19 NEWLINE memory check error at 0x00D30D44 = 0x70, should be 0xFD. NEWLINE memory check error at 0x00D30D45 = 0x0D, should be 0xFD. NEWLINE memory check error at 0x00D30D46 = 0xD3, should be 0xFD. NEWLINE memory check error at 0x00D30D47 = 0x00, should be 0xFD. NEWLINE NEWLINE NEWLINE Call stack: NEWLINE ----------- NEWLINE 1. _free_dbg_lk(void * 0x00d30d90, int 0x00000001) line 1033 + 63 bytes NEWLINE 2. _free_dbg(void * 0x00d30d90, int 0x00000001) line 970 + 13 bytes NEWLINE 3. free(void * 0x00d30d90) line 926 + 11 bytes NEWLINE 4. PR_Free(void * 0x00d30d90) line 66 + 10 bytes NEWLINE 5. HT_DeleteCursor(_HT_CursorStruct * 0x00d30d90) line 3251 + 9 bytes NEWLINE 6. CRDFToolbar::WidthChanged(int 0x00000024) line 2130 + 9 bytes NEWLINE 7. CDragToolbar::ArrangeToolbar(int 0xffffffdc, int 0x0000002e) line 725 NEWLINE 8. CDragToolbar::OnSize(unsigned int 0x00000000, int 0x00000000, int NEWLINE 0x0000002e) line 432 NEWLINE 9. CWnd::OnWndMsg(unsigned int 0x00000005, unsigned int 0x00000000, long NEWLINE 0x002e0000, long * 0x0012e160) line 1763 NEWLINE NEWLINE NEWLINE Cause: NEWLINE ------ NEWLINE This appears to be a bug in Microsoft code, since the last four function NEWLINE calls were inside Windows/MFC. The place in the Mozilla code where this NEWLINE bug would need to be isolated is in ht.c, line 3244: NEWLINE NEWLINE PR_PUBLIC_API(HT_Error) NEWLINE HT_DeleteCursor (HT_Cursor cursor) NEWLINE { NEWLINE XP_ASSERT(cursor != NULL); NEWLINE NEWLINE if (cursor != NULL) NEWLINE { NEWLINE /* >> */ freeMem(cursor); NEWLINE } NEWLINE return(HT_NoErr); NEWLINE } NEWLINE NEWLINE Here is the code at the tail end, where the stoppage occurred: NEWLINE NEWLINE Line 1035 of msdev/crt/src/dbgheap.c... NEWLINE NEWLINE /* if we didn't already check entire heap, at least check this object */ NEWLINE if (!(_crtDbgFlag & _CRTDBG_CHECK_ALWAYS_DF)) NEWLINE { NEWLINE /* check no-mans-land gaps */ NEWLINE if (!CheckBytes(pHead->gap, _bNoMansLandFill, nNoMansLandSize)) NEWLINE _RPT3(_CRT_ERROR, "DAMAGE: before %hs block (#%d) at 0x%08X.\\n", NEWLINE szBlockUseName[_BLOCK_TYPE(pHead->nBlockUse)], NEWLINE pHead->lRequest, NEWLINE (BYTE *) pbData(pHead)); NEWLINE NEWLINE /* >> */ if (!CheckBytes(pbData(pHead) + pHead->nDataSize, _bNoMansLandFill, NEWLINE nNoMansLandSize)) NEWLINE _RPT3(_CRT_ERROR, "DAMAGE: after %hs block (#%d) at 0x%08X.\\n", NEWLINE szBlockUseName[_BLOCK_TYPE(pHead->nBlockUse)], NEWLINE pHead->lRequest, NEWLINE (BYTE *) pbData(pHead)); NEWLINE } NEWLINE NEWLINE NEWLINE I don't know what CheckBytes does, but apparently it didn't pass the second NEWLINE call. NEWLINE Here's the values of the variables for this call: NEWLINE NEWLINE pHead = 0x00d30d70 NEWLINE pHead->nDataSize = 0xffffffb4 NEWLINE _bNoMansLandFill = 0xfd NEWLINE nNoMansLandFill = ?? Couldn't tell. NEWLINE NEWLINE NEWLINE Is it possible that another application (Word, perhaps) tinkered with Mozilla's NEWLINE memory? NEWLINE NEWLINE NEWLINE System: NEWLINE ------- NEWLINE I'm running WinNT4.0 SR3, on a 166MHz Dell Pentium with 64MB RAM and NEWLINE 200MB swap. In addition to WinNT processes and mozilla, I am running NEWLINE cdplayer.exe, MSACCESS, TASKMGR, MSDEV, CRT, and a notepad-like editor.
952|XP_ASSERT in PA_DropDocData pa_parse.c|'Nother assert... This one while viewing http://sagan.earthspace.net/~esr/ecsl/ NEWLINE NEWLINE Line 541 of pa_parse.c: NEWLINE NEWLINE pa_DocData * NEWLINE PA_DropDocData(NET_StreamClass *stream) NEWLINE { NEWLINE pa_DocData *doc_data=(pa_DocData *)stream->data_object; NEWLINE XP_ASSERT(doc_data && NEWLINE /* >> */ (doc_data->stream_count >= 1)); NEWLINE if (!doc_data) NEWLINE return NULL; NEWLINE doc_data->stream_count--; NEWLINE return pa_check_doc_data_count(stream); NEWLINE } NEWLINE NEWLINE doc_data equals NULL. I've tried tracing it back, but I can't tell NEWLINE how the NULL enters the picture. NEWLINE NEWLINE Stack trace (trimmed): NEWLINE ---------------------- NEWLINE 1. XP_AssertAtLine(char * 0x00906cc8, int 0x00000213) line 2687 NEWLINE 2. PA_DropDocData(_NET_StreamClass * 0x0012fc54) line 531 + 78 bytes NEWLINE 3. lo_FlushBlockage(MWContext_ * 0x00c5a4a0, lo_DocState_struct * 0x01c98190, NEWLINE lo_DocState_struct * 0x01c98190) line 2742 + 9 bytes NEWLINE 4. lo_set_image_info(MWContext_ * 0x00c5a4a0, long 0x000003da, long 0x00000070, NEWLINE long 0x000000a4) line 5624 + 17 bytes NEWLINE 5. lo_process_deferred_image_info(void * 0x00000000) line 5724 + 32 bytes NEWLINE 6. wfe_ProcessTimeouts(unsigned long 0x01788359) line 303 + 12 bytes NEWLINE 7. FireTimeout(HWND__ * 0x00040272, unsigned int 0x00000113, unsigned int NEWLINE 0x00000309, unsigned long 0x01788359) line 60 + 9 bytes NEWLINE 8. USER32! 77e7128c() NEWLINE 9. CNetscapeApp::Run() line 1675 + 8 bytes NEWLINE <Snip> NEWLINE NEWLINE NEWLINE System: NEWLINE ------- NEWLINE I'm running WinNT4.0 SR3, on a 166MHz Dell Pentium with 64MB RAM and NEWLINE 200MB swap. In addition to WinNT processes and mozilla, I am running NEWLINE cdplayer.exe, MSACCESS, TASKMGR, MSDEV, CRT, and a notepad-like editor.
953|URL: news: protocol causes ASSERT|Ran into a PR_ASSERT when I clicked on a news link. NEWLINE NEWLINE Cause: NEWLINE ------ NEWLINE NEWLINE /* get a handle to a protocol implemenation NEWLINE */ NEWLINE NET_ProtoImpl * NEWLINE net_get_protocol_impl(int for_url_type) NEWLINE { NEWLINE int count=0; NEWLINE NEWLINE /* if we ever get around to doing dynamic protocol loading NEWLINE * this would be a good place to plug it in. NEWLINE * just load a DLL with the implementation and NEWLINE * return the handle. NEWLINE * The integer URL_TYPE would need to be replaced with NEWLINE * strings or some other identifier so that it can all be NEWLINE * handled dynamically NEWLINE */ NEWLINE NEWLINE for(; count < net_number_of_proto_impls; count++) NEWLINE { NEWLINE if(net_proto_impls[count].url_type == for_url_type) NEWLINE return net_proto_impls[count].impl; NEWLINE } NEWLINE NEWLINE /* >> */ PR_ASSERT(0); /* should always find one */ NEWLINE return NULL; NEWLINE } NEWLINE NEWLINE NEWLINE I hit the above assert line. NEWLINE NEWLINE I'm guessing this is hit because the news client isn't available, right? NEWLINE It'd be nice if instead of asserting, a dialog to this effect popped up. NEWLINE NEWLINE NEWLINE System: NEWLINE ------- NEWLINE I'm running WinNT4.0 SR3, on a 166MHz Dell Pentium with 64MB RAM and NEWLINE 200MB swap. In addition to WinNT processes and mozilla, I am running NEWLINE cdplayer.exe, MSACCESS, TASKMGR, MSDEV, CRT, PowerPoint, Netscape 4.06, NEWLINE and a notepad-like editor.
954|Percentage widths on text fields is not inherited properly|See URL for simple test case: NEWLINE NEWLINE <form> NEWLINE <div style="width:50%; background-color:yellow"> NEWLINE <input type="text" style="width:75%"> NEWLINE </div> NEWLINE </form> NEWLINE NEWLINE The input width should be 75% of the parent element's width (DIV) which is 50% NEWLINE of the width of the body. NEWLINE NEWLINE This is a requirement I have to do some constraints stuff in an XPFE.
955|Mozilla crashing in tooltip code|On IRIX (with display depth not equal to depth of the overlay plane?): NEWLINE Bring up a browser window, move the mouse over a button, and pause as though NEWLINE waiting for a tooltip to come up. Mozilla crashes with an assertion: NEWLINE NEWLINE Assertion failed: _pixmap_buffer_depth == _XfeDepth(w), file PixmapUtil.c, line NEWLINE 64 NEWLINE NEWLINE with stack trace: NEWLINE NEWLINE 3 __assert(0x7507dc, 0x750804, 0x7fff0340, 0x40) ["assert.c":59, 0xfa585f8] NEWLINE 4 _XfePixmapBufferAllocate(w = 0x10377000) NEWLINE ["/gromit/moz9.30/mozilla/cmd/xfe/XfeWidgets/Xfe/PixmapUtil.c":64, 0x731168] NEWLINE 5 BufferAllocate(w = 0x10377000) NEWLINE ["/gromit/moz9.30/mozilla/cmd/xfe/XfeWidgets/Xfe/Primitive.c":1147, 0x733408] NEWLINE 6 BufferUpdate(w = 0x10377000) NEWLINE ["/gromit/moz9.30/mozilla/cmd/xfe/XfeWidgets/Xfe/Primitive.c":1190, 0x7335b0] NEWLINE 7 InitializePostHook(rw = 0x7fff0690, nw = 0x10377000) NEWLINE ["/gromit/moz9.30/mozilla/cmd/xfe/XfeWidgets/Xfe/Primitive.c":989, 0x732e1c] NEWLINE 8 _XfePrimitiveChainInitialize(rw = 0x7fff0690, nw = 0x10377000, wc = NEWLINE 0x752758) ["/gromit/moz9.30/mozilla/cmd/xfe/XfeWidgets/Xfe/Primitive.c":1446, NEWLINE 0x733d58] NEWLINE 9 Initialize(rw = 0x7fff0690, nw = 0x10377000, args = 0x7fff0a88, nargs = NEWLINE 0x7fff0550) ["/gromit/moz9.30/mozilla/cmd/xfe/XfeWidgets/Xfe/Button.c":850, NEWLINE 0x70c5dc] NEWLINE 10 CallInitialize(0x752758, 0x7fff0690, 0x10377000, 0x7fff0a88) NEWLINE ["Create.c":229, 0xf64c128] NEWLINE 11 xtCreate(0x6d6d54, 0x0, 0x752758, 0x1036c600) ["Create.c":476, 0xf64ccac] NEWLINE 12 _XtCreateWidget(0x6d6d54, 0x752758, 0x1036c600, 0x7fff0a88) NEWLINE ["Create.c":634, 0xf64d220] NEWLINE 13 XtCreateManagedWidget(0x2d6c, 0x6, 0x7fff0340, 0x0) ["Create.c":694, NEWLINE 0xf640414] NEWLINE 14 LabelCreate(w = 0x1036c600) NEWLINE ["/gromit/moz9.30/mozilla/cmd/xfe/XfeWidgets/XfeToolTip/ToolTipShell.c":426, NEWLINE 0x6c597c] NEWLINE 15 Initialize(rw = 0x7fff0d20, nw = 0x1036c600, args = (nil), nargs = NEWLINE 0x7fff0be0) NEWLINE ["/gromit/moz9.30/mozilla/cmd/xfe/XfeWidgets/XfeToolTip/ToolTipShell.c":290, NEWLINE 0x6c5414] NEWLINE 16 InitializeLeafWrapper(0x7fff0d20, 0x1036c600, 0x0, 0x7fff0be0) NEWLINE ["BaseClass.c":2030, 0xb3a1964] NEWLINE 17 CallInitialize(0x6d6f34, 0x7fff0d20, 0x1036c600, 0x0) ["Create.c":229, NEWLINE 0xf64c128] NEWLINE 18 xtCreate(0x6d67f4, 0x0, 0x6d6f34, 0x100dbe00) ["Create.c":476, 0xf64ccac] NEWLINE 19 _XtCreatePopupShell(0x6d67f4, 0x6d6f34, 0x100dbe00, 0x0) ["Create.c":739, NEWLINE 0xf663d0c] NEWLINE 20 XtCreatePopupShell(0x2d6c, 0x6, 0x7fff0340, 0x0) ["Create.c":769, NEWLINE 0xf663b28] NEWLINE 21 XfeCreateToolTipShell(pw = 0x100dbe00, name = 0x6d67f4 = "ToolTipShell", NEWLINE av = (nil), ac = 0) NEWLINE ["/gromit/moz9.30/mozilla/cmd/xfe/XfeWidgets/XfeToolTip/ToolTipShell.c":446, NEWLINE 0x6c59f0]
956|Tooltips dont respond to tooltip pref|They just dont care...
957|BUTTON element not implemented|We haven't implemented the HTML 4.0 element "BUTTON." NEWLINE NEWLINE It works like this: NEWLINE <button> NEWLINE <img src="..."><br> NEWLINE Hello, World NEWLINE </button> NEWLINE NEWLINE The button can contain arbitrary HTML that will be rendered on the face of a NEWLINE button. For an example of how this works, check out IE 4/5.
958|form element disabled attribute DOM wiring not implemented|We need to support the "disabled" attribute on many HTML form elements, such as NEWLINE a button: NEWLINE NEWLINE <input type="button" value="Test" disabled="true"> NEWLINE NEWLINE - This button should appear disabled (grayed out) NEWLINE - clicking on it should NOT fire an event NEWLINE - the disabled attribute needs to be exposed to the DOM so it can be set through NEWLINE JavaScript. NEWLINE NEWLINE Obviously the semantics will vary slightly among elements (for example, a NEWLINE disabled text area may be grayed out and never get focus)
959|[FEATURE] Accesskey attribute not implemented yet for XUL|Several tricks here to implement... NEWLINE NEWLINE <input type="button" value="Test" accesskey="T"> NEWLINE NEWLINE - when the user presses alt-(accesskey), the appropriate element should get NEWLINE focus. If an accesskey is specified, it overrides any accesskey in the viewer NEWLINE application. For example, if the accesskey on a form element were NEWLINE "F," pressing "alt-F" would set focus on the element, not bring up the "File" NEWLINE menu. NEWLINE NEWLINE I assume joki owns this piece of the bug... NEWLINE NEWLINE - we should underline the first instance of the accesskey character. In the NEWLINE example above, the first "T" in the word "Test" would be underlined. NEWLINE NEWLINE I assume chris or kevin or kipp or someone owns this piece of the bug...
960|[FEATURE] window.onResize not implemented|I need to capture this event for some XPFE stuff I'm working on. I don't need to NEWLINE know much, other than the event happened, and the new dimensions of the window NEWLINE (which I'll file in a seperate bug report)
961|[BLOCKER] Document dimensions not exposed to JS yet|document.width NEWLINE window.innerWidth NEWLINE window.outerWidth NEWLINE NEWLINE ... not implemented yet. I need to get this value for some XPFE stuff I'm NEWLINE working on. I do not need set yet.
962|document.getElementByID() does not work|Test case: NEWLINE NEWLINE <p id="foo">Testing</p> NEWLINE <script> NEWLINE alert(document.getElementByID("foo")); NEWLINE </script> NEWLINE NEWLINE this should send out "[object HTMLParagraphElement]" NEWLINE NEWLINE currently, the only way I know to get an element is getElementByTagName(), which NEWLINE is tedious because then I need to index all of those and check each one's ID via NEWLINE getDOMAttribute (which should be changed to getAttribute some day, but that's NEWLINE another bug) NEWLINE NEWLINE This would be very helpful to the XPFE stuff I'm hacking together.
963|Table cells can overlap when shrinking window|Test case: NEWLINE NEWLINE <table> NEWLINE <tr> NEWLINE <td>Back</td> NEWLINE <td>Forward</td> NEWLINE <td>Reload</td> NEWLINE <td>Home</td> NEWLINE <td>Search</td> NEWLINE <td>My Netscape</td> NEWLINE <td>Print</td> NEWLINE <td>Security</td> NEWLINE <td>Stop</td> NEWLINE </tr> NEWLINE </table> NEWLINE NEWLINE Now, resize window to make it smaller. At some point, a horizontal scrollbar NEWLINE should appear and the table should stop shrinking. But instead, the cells just NEWLINE start smushing up over one another.
964|mozilla.exe - Application Error|This is in the nightly build test of mozilla. Version "%s - 5009808600" NEWLINE (Looks like something may be wrong there). I am using a Pentium Pro 200 with NEWLINE 64 Megs of ram (A micron system), my OS is NT 4.0 SP3 and IE4. I was trying to NEWLINE read the news at www.slashdot.org, and I recieved an Application Error. The NEWLINE contents of the box were : NEWLINE "The exception Breakpoint (Doah!) NEWLINE A breakpoint has been reached. NEWLINE (0x80000003)occurred in the application at location 0x77f76148. NEWLINE NEWLINE Click on OK to ... NEWLINE Click on CANCEL to ..." NEWLINE NEWLINE It appears as if someone left a breakpoint in the code, and needs to clear it... NEWLINE NEWLINE Also when I overlap mozilla onto another window, such as wordpad, then change NEWLINE focus to wordpad, I have a chunk of mozilla stuck on the wordpad window... NEWLINE NEWLINE And when I click on the "Email me a password" button without putting a name NEWLINE into the username section on the Bugzilla login screen, Mozilla asks if it NEWLINE wants me to remember the username and password, I click ok, and I get: NEWLINE "Debug Assertion Failed! NEWLINE NEWLINE Program: C:\\Downloads.....\\Mozilla.exe NEWLINE File: Y:\\Nightly\\Mozilla\\cmd\\winfe\\fenet.cpp NEWLINE Line: 199 NEWLINE NEWLINE For information on how your program can cause an assertion NEWLINE failure .... NEWLINE NEWLINE (Press Retry...)" NEWLINE NEWLINE Hope this helps.
965|Tables with border styles don't display properly|This table: NEWLINE <table style="border: thin solid red" border=2> NEWLINE <tr><td>One</td><td>Two</td></tr> NEWLINE </table> NEWLINE NEWLINE Should look just like this one: NEWLINE <table border=2> NEWLINE <tr><td>One</td><td>Two</td></tr> NEWLINE </table> NEWLINE ... but in in addition it should have a red border around it. Currently, we NEWLINE display the red border, but we disregard the "border=2" stuff that should be NEWLINE applied to TD's. NEWLINE NEWLINE Seperately, the border on TD elements is an inconsistent width. For example, see NEWLINE this testcase: NEWLINE NEWLINE <table border=2 style="border: thin solid red"> NEWLINE <tr> NEWLINE <Td style="border: thin solid green">one</td> NEWLINE <td style="border: thin solid red">two</td> NEWLINE <td style="border: thin solid blue">three</td> NEWLINE </tr> NEWLINE <tr> NEWLINE <td style="border: thin solid green">four</td> NEWLINE <td style="border: thin solid yellow">five</td> NEWLINE <td style="border: thin solid orange">six</td> NEWLINE </tr> NEWLINE </table> NEWLINE NEWLINE IE 5.0 displays this properly. But in raptor, the borders of some sides of some NEWLINE TDs are wider than other borders. Additionally, the red border surrounding the NEWLINE entire table is too wide on the right side.
966|border: none creates a 5 pixel border|Try this html and you will see a 5 pixel border... NEWLINE NEWLINE <HTML> NEWLINE <STYLE> NEWLINE BODY { NEWLINE color: black; NEWLINE background: white; NEWLINE margin: 2em; NEWLINE } NEWLINE DIV { NEWLINE border: none; NEWLINE padding: 0; NEWLINE margin: 0; NEWLINE } NEWLINE DIV.outside { NEWLINE color: white; NEWLINE background-image: url(ruler.gif); NEWLINE border: 2px solid black; NEWLINE } NEWLINE DIV.one { NEWLINE background: khaki; NEWLINE } NEWLINE DIV.two { NEWLINE background: transparent; NEWLINE } NEWLINE DIV P { NEWLINE margin: 50px; NEWLINE background-image: transparent; NEWLINE } NEWLINE </STYLE> NEWLINE <BODY> NEWLINE <DIV CLASS=outside> NEWLINE <DIV CLASS=one> NEWLINE <P>Paragraph inside DIV.one</P> NEWLINE </DIV> NEWLINE <DIV CLASS=two> NEWLINE <P>Paragraph inside DIV.two</P> NEWLINE </DIV> NEWLINE </DIV> NEWLINE </BODY> NEWLINE </HTML>
967|crash opening page|the URL opens a second window. I might have moused over the second window just NEWLINE as it opened, I'm not sure. NEWLINE NEWLINE in GlobalWindowImpl, NEWLINE newGlobalObject->SetOpenerWindow(this); //newGlobalObject is null!!! NEWLINE NEWLINE I got this stack: NEWLINE NEWLINE GlobalWindowImpl::Open(GlobalWindowImpl * const 0x0132e5c8, JSContext * NEWLINE 0x013f0e20, long * 0x01d35fcc, unsigned int 2, nsIDOMWindow * * 0x0012eb0c) line NEWLINE 1091 + 50 bytes NEWLINE WindowOpen(JSContext * 0x013f0e20, JSObject * 0x01bb4db8, unsigned int 2, long * NEWLINE 0x01d35fcc, long * 0x0012eba0) line 882 + 28 bytes NEWLINE js_Invoke(JSContext * 0x013f0e20, unsigned int 2, int 0) line 630 + 26 bytes NEWLINE js_Interpret(JSContext * 0x013f0e20, long * 0x0012f100) line 2122 + 15 bytes NEWLINE js_Invoke(JSContext * 0x013f0e20, unsigned int 0, int 0) line 646 + 13 bytes NEWLINE js_Interpret(JSContext * 0x013f0e20, long * 0x0012f6f8) line 2122 + 15 bytes NEWLINE js_Execute(JSContext * 0x013f0e20, JSObject * 0x01bb56e8, JSScript * 0x0141d410, NEWLINE JSFunction * 0x00000000, JSStackFrame * 0x00000000, int 0, long * 0x0012f6f8) NEWLINE line 804 + 13 bytes NEWLINE JS_EvaluateUCScriptForPrincipals(JSContext * 0x013f0e20, JSObject * 0x01bb56e8, NEWLINE JSPrincipals * 0x00000000, unsigned short * 0x01ce6050, unsigned int 1095, char NEWLINE * 0x01321c80, unsigned int 9, long * 0x0012f6f8) line 2226 + 27 bytes NEWLINE JS_EvaluateUCScript(JSContext * 0x013f0e20, JSObject * 0x01bb56e8, unsigned NEWLINE short * 0x01ce6050, unsigned int 1095, char * 0x01321c80, unsigned int 9, long * NEWLINE 0x0012f6f8) line 2208 + 35 bytes NEWLINE JS_EvaluateScript(JSContext * 0x013f0e20, JSObject * 0x01bb56e8, char * NEWLINE 0x01cff728, unsigned int 1095, char * 0x01321c80, unsigned int 9, long * NEWLINE 0x0012f6f8) line 2175 + 33 bytes NEWLINE nsJSContext::EvaluateString(char * 0x01cff728, unsigned int 1095, char * NEWLINE 0x01321c80, unsigned int 9, long * 0x0012f6f8) line 77 + 50 bytes NEWLINE HTMLContentSink::ProcessSCRIPTTag(const nsIParserNode & {...}) line 2191 + 46 NEWLINE bytes NEWLINE HTMLContentSink::AddLeaf(HTMLContentSink * const 0x013ea580, const nsIParserNode NEWLINE & {...}) line 1734 + 12 bytes NEWLINE CNavDTD::AddLeaf(const nsIParserNode & {...}) line 2521 + 22 bytes NEWLINE CNavDTD::HandleScriptToken(nsCParserNode & {...}) line 844 + 12 bytes NEWLINE CNavDTD::OpenContainer(const nsIParserNode & {...}, int 1) line 2354 + 12 bytes NEWLINE CNavDTD::HandleDefaultStartToken(CToken * 0x01288c20, nsHTMLTag eHTMLTag_script, NEWLINE nsIParserNode & {...}) line 576 + 14 bytes NEWLINE CNavDTD::HandleStartToken(CToken * 0x01288c20) line 661 + 23 bytes NEWLINE NavDispatchTokenHandler(CToken * 0x01288c20, nsIDTD * 0x0136ab80) line 293 + 12 NEWLINE bytes NEWLINE CTokenHandler::operator()(CToken * 0x01288c20, nsIDTD * 0x0136ab80) line 80 + 14 NEWLINE bytes NEWLINE CNavDTD::HandleToken(CNavDTD * const 0x0136ab80, CToken * 0x01288c20) line 524 + NEWLINE 18 bytes NEWLINE nsParser::BuildModel() line 599 + 16 bytes NEWLINE nsParser::ResumeParse() line 564 NEWLINE nsParser::OnDataAvailable(nsParser * const 0x013e6ba4, nsIURL * 0x01320d70, NEWLINE nsIInputStream * 0x013ab240, int 38) line 774 + 15 bytes NEWLINE nsDocumentBindInfo::OnDataAvailable(nsDocumentBindInfo * const 0x01320b90, NEWLINE nsIURL * 0x01320d70, nsIInputStream * 0x013ab240, int 38) line 1125 + 30 bytes NEWLINE OnDataAvailableProxyEvent::HandleEvent(OnDataAvailableProxyEvent * const NEWLINE 0x013c1440) line 572 + 45 bytes NEWLINE StreamListenerProxyEvent::HandlePLEvent(PLEvent * 0x013c1444) line 418 + 12 NEWLINE bytes NEWLINE PL_HandleEvent(PLEvent * 0x013c1444) line 395 + 10 bytes NEWLINE PL_ProcessPendingEvents(PLEventQueue * 0x01206540) line 357 + 9 bytes NEWLINE _md_EventReceiverProc(void * 0x00a00746, unsigned int 49403, unsigned int 0, NEWLINE long 18900288) line 675 + 9 bytes NEWLINE USER32! 77e71250() NEWLINE 01206540()
968|Viewer crashes when rendering the above url|on the 9/28 build, accessing the above page causes the viewer to crash. I get NEWLINE the following info from ht crash dialog: NEWLINE NEWLINE VIEWER caused an invalid page fault in NEWLINE module RAPTORHTML.DLL at 014f:006188cc. NEWLINE NEWLINE After looking at the code on this page, Germaine and I found that a form element NEWLINE start tag is within a table and the form element end tag is outside of the table NEWLINE element end tag. When we move the form element start tag to outside the table NEWLINE the page does not crash.
969|[BLOCK] min-width, max-width and min/max-height properties buggy|This is a requirement to do better contraints stuff in HTML. buster said it NEWLINE would be straightforward to hook this up for TD's; I'm filing on Kipp since NEWLINE there would be additional work involved to hook it up to all elements (DIVs, NEWLINE etc.) NEWLINE NEWLINE It grew out of a post I made to mozilla-layout, and David's reply: NEWLINE NEWLINE "L. David Baron" wrote: NEWLINE >From: angus@netscape.com (Angus Davis) NEWLINE >I need to be able to do something like: NEWLINE > NEWLINE ><td style="width: 100%; minimum-width: 100px;"> NEWLINE > NEWLINE >Thoughts? Is there some other way to accomplish this? (CSS geeks - speak NEWLINE >up!) NEWLINE >-angus NEWLINE NEWLINE This is part of the CSS2 spec, except the properties are called NEWLINE "min-width" and "max-width". It is described in detail in NEWLINE http://www.w3.org/TR/REC-CSS2/visudet.html#min-max-widths NEWLINE (That whole part of the spec is probably worth reading, as it has NEWLINE describes quite thoroughly how to handle some rather difficult NEWLINE situations.) NEWLINE - David Baron
970|CSS font-weight: Bold and Bolder look the same|- Go to URL NEWLINE http://www.mozilla.org/newlayout/testcases/css/sec525bolder.htm NEWLINE code: .two {font-weight: bolder} NEWLINE - Compare to URL NEWLINE http://www.mozilla.org/newlayout/testcases/css/sec525b.htm NEWLINE code: .one {font-weight: bold} NEWLINE NEWLINE - Note that both appear to be the same boldness.
971|floating button not placed correctly|the floating button is not placed correctly. It spills out of the table on the NEWLINE original page (an ad that changes frequently, so the URL is irrelevant.) The NEWLINE test case below shows the problem. NEWLINE NEWLINE <html><body> NEWLINE NEWLINE <form METHOD="POST" ACTION="http://www.neodata.com/cgi-bin/SFOrder"> NEWLINE NEWLINE <table width="290" units="pixels" BORDER=1 cellpadding="0" cellspacing="0"> NEWLINE <tr> NEWLINE <td colspan="2"> NEWLINE <input type="submit" VALUE="Order Now" border="0" align="middle"> NEWLINE Offer Good One Per Person, or while supply lasts. Void where NEWLINE prohibitted. Your mileage may vary. Who knows what your dog is doing right NEWLINE now? NEWLINE </td> NEWLINE </tr> NEWLINE </table> NEWLINE </form> NEWLINE NEWLINE </body> </html>
972|[FONT MAC] CSS font-weight: all font weights show as either normal or bold|Font weights normal, lighter, and 100 through 400 all appear as normal text. NEWLINE Font weights bold, bolder, and 500 through 900 all appear as bold text. NEWLINE (bold and bolder already reported on bug #970)
973|CSS font-size: "small" looks like normal font size|- Go to URL NEWLINE http://www.mozilla.org/newlayout/testcases/css/sec526s.htm NEWLINE - font size should be smaller than normal
974|CSS font-size: "ex" is showing as xxsmall ; any ex units are too small|- Go to the URL NEWLINE http://www.mozilla.org/newlayout/testcases/css/sec526ex.htm NEWLINE - The text should be 2ex high, instead it is more like xxsmall.
975|{css1} font-size: test for ".5in" is nearly an inch high|- Go to URL NEWLINE http://www.mozilla.org/newlayout/testcases/css/sec526in.htm NEWLINE - Style should show font size as .5in, but it appears as nearly an inch high.
976|{css1} font-size: test for "1cm" font size is 1.5 cm high|- Go to URL NEWLINE http://www.mozilla.org/newlayout/testcases/css/sec526cm.htm NEWLINE - Style should show font size as 1cm, but it's 1.5cm high.
977|CSS font: italics shows as normal text|- Go to URL NEWLINE http://www.mozilla.org/newlayout/testcases/css/sec527sty.htm NEWLINE - Text should appear as itallic, but shows as normal instead.
978|CSS font: shows as normal text instead of small-caps|- Go to URL NEWLINE http://www.mozilla.org/newlayout/testcases/css/sec527v.htm NEWLINE - text should show as small-caps, but shows as normal instead.
979|CSS font: font should show as bold, but shows as normal|- Go to URL NEWLINE http://www.mozilla.org/newlayout/testcases/css/sec527w.htm NEWLINE - Font should show as bold, but shows as normal
980|CSS font: font size should be xxsmall, but is normal instead|- Go to URL NEWLINE http://www.mozilla.org/newlayout/testcases/css/sec527siz.htm NEWLINE - font should show as xxsmall, but is normal instead.
981|CSS font: line height should be 200%, but shows as normal instead|- Go to URL NEWLINE http://www.mozilla.org/newlayout/testcases/css/sec527h.htm NEWLINE - line height should be 200%, but shows as normal instead
982|CSS font: sentence should be san-serif, but is serif (my default font)|- Go to URL NEWLINE http://www.mozilla.org/newlayout/testcases/css/sec527f.htm NEWLINE - sentence should be san-serif, but is serif (my default font)
983|CSS color: RGB colors are all showing as black|- Go to URL NEWLINE http://www.mozilla.org/newlayout/testcases/css/sec531rgb6aqua.htm NEWLINE - The text should be aqua, but shows as black instead.
984|CSS background color: 3 & 6 digit RGB values all show as normal|- Go to URL NEWLINE http://www.mozilla.org/newlayout/testcases/css/sec532rgb6aqua.htm NEWLINE - no background colors are working for any 3 or 6 digit RGB values
985|{perf} Performance problem with long directory listings|- Go to URL NEWLINE http://slip/projects/marvin/css/ NEWLINE (netscape internal location - behind firewall) NEWLINE - viewer starts to list a directory of all the files in this folder, then locks NEWLINE up and dies.
986|Unable to go to last half of links on page|- Go to URL NEWLINE http://www.mozilla.org/newlayout/testcases/css/index.html NEWLINE - Put the scroll bar about 1/2 way down and you can no longer access the links NEWLINE to the test cases. I stop being able to access after this test: sec541in.htm
987|background color for form elements is wrong|If you compare this page in navigator and IE you can see the difference. The NEWLINE text fields are inheriting (!) the background color of their container, which I NEWLINE believe is wrong.
988|the bullets in front of some of the items are positioned oddly|and don't forget to fix the vertical alignment with bullets bug... NEWLINE NEWLINE maybe I can factor the code so that the bullet is handled directly by the NEWLINE vertical alignment code? Put the frames into an array or something?
989|crash in js form handling stuff...|00000002() NEWLINE GetHTMLFormElementProperty(JSContext * 0x011cbb60, JSObject * 0x01785bc8, long NEWLINE 24662876, long * 0x0012fab4) line 257 + 29 bytes NEWLINE js_GetProperty(JSContext * 0x011cbb60, JSObject * 0x01785bc8, long 18836768, NEWLINE long * 0x0012fab4) line 1602 + 131 bytes NEWLINE js_Interpret(JSContext * 0x011cbb60, long * 0x0012fc2c) line 2089 + 912 bytes NEWLINE js_Invoke(JSContext * 0x011cbb60, unsigned int 1, int 0) line 646 + 13 bytes NEWLINE js_CallFunctionValue(JSContext * 0x011cbb60, JSObject * 0x01784228, long NEWLINE 24662856, unsigned int 1, long * 0x0012fd4c, long * 0x0012fd54) line 732 + 15 NEWLINE bytes NEWLINE JS_CallFunctionValue(JSContext * 0x011cbb60, JSObject * 0x01784228, long NEWLINE 24662856, unsigned int 1, long * 0x0012fd4c, long * 0x0012fd54) line 2257 + 29 NEWLINE bytes NEWLINE nsJSEventListener::ProcessEvent(nsIDOMEvent * 0x011c99a0) line 98 + 34 bytes NEWLINE nsEventListenerManager::HandleEvent(nsIPresContext & {...}, nsEvent * NEWLINE 0x0012fe20, nsIDOMEvent * * 0x0012fdf8, nsEventStatus & nsEventStatus_eIgnore) NEWLINE line 485 + 17 bytes NEWLINE GlobalWindowImpl::HandleDOMEvent(GlobalWindowImpl * const 0x011c35c4, NEWLINE nsIPresContext & {...}, nsEvent * 0x0012fe20, nsIDOMEvent * * 0x0012fdf8, NEWLINE unsigned int 1, nsEventStatus & nsEventStatus_eIgnore) line 1408 NEWLINE nsWebShell::OnConnectionsComplete(nsWebShell * const 0x011862b0) line 1658 + 34 NEWLINE bytes NEWLINE nsDocLoaderImpl::LoadURLComplete(nsISupports * 0x01196660) line 730 NEWLINE nsDocumentBindInfo::OnStopBinding(nsDocumentBindInfo * const 0x01196660, nsIURL NEWLINE * 0x011966c0, int 0, const nsString & {""}) line 1153 NEWLINE OnStopBindingProxyEvent::HandleEvent(OnStopBindingProxyEvent * const 0x0120d670) NEWLINE line 536 NEWLINE StreamListenerProxyEvent::HandlePLEvent(PLEvent * 0x0120d674) line 418 + 12 NEWLINE bytes NEWLINE PL_HandleEvent(PLEvent * 0x0120d674) line 395 + 10 bytes NEWLINE PL_ProcessPendingEvents(PLEventQueue * 0x01146720) line 357 + 9 bytes NEWLINE _md_EventReceiverProc(void * 0x0059037e, unsigned int 49277, unsigned int 0, NEWLINE long 18114336) line 675 + 9 bytes NEWLINE USER32! 77e71250()
990|floaters in the box above the weather area are wrong|the text ends up overlapping the floaters (or whatever they are)
991|[DOGFOOD][4.xP] font tags containing blocks are 'badly' parsed|Under the "Top stories" section we get different fonts than IE and NEWLINE navigator...I suspect its a face/attribute problem.
992|Contact (SM) is laying out wrong|the center web-search table has a super-script on it that is not laying out NEWLINE properly. In addition, the form elements layout wrong as well.
993|Unhandled exception in jsapi|Unhandled exception. This time in jsapi.c, line 829. NEWLINE (I'm guessing this is part of the javascript module.) NEWLINE NEWLINE I had about 8 windows open when the crash occurred. NEWLINE I'd been looking at BBS CGI scripts, and had just NEWLINE closed a window. NEWLINE NEWLINE This occurred on Win95 running Moz-980904. NEWLINE NEWLINE Cause: NEWLINE ------ NEWLINE The debugger indicates that the exception was NEWLINE somewhere inside of: NEWLINE NEWLINE JS_PUBLIC_API(void) NEWLINE JS_GC(JSContext *cx) NEWLINE { NEWLINE if (!cx->fp) NEWLINE PR_FinishArenaPool(&cx->stackPool); NEWLINE /* >> */ PR_FinishArenaPool(&cx->codePool); NEWLINE PR_FinishArenaPool(&cx->tempPool); NEWLINE js_ForceGC(cx); NEWLINE } NEWLINE NEWLINE The debugger only traces into PR_FinishArenaPool in the dissassembly window, NEWLINE which I can't follow. So I can't find the precise cause. NEWLINE NEWLINE I am guessing that the above code is supposed NEWLINE to be more like this: NEWLINE NEWLINE JS_PUBLIC_API(void) NEWLINE JS_GC(JSContext *cx) NEWLINE { NEWLINE if (!cx->fp) { NEWLINE PR_FinishArenaPool(&cx->stackPool); NEWLINE PR_FinishArenaPool(&cx->codePool); NEWLINE PR_FinishArenaPool(&cx->tempPool); NEWLINE js_ForceGC(cx); NEWLINE } NEWLINE } NEWLINE NEWLINE PLEASE let me know if this is correct or not, and if not, NEWLINE what a better solution might be. NEWLINE NEWLINE Stack Trace: NEWLINE ------------ NEWLINE KERNEL32! bff76693() NEWLINE FreeArenaList(PLArenaPool * 0x026c4158, PLArena * 0x026c4158, int 1) line 137 + NEWLINE 61 bytes NEWLINE PL_FinishArenaPool(PLArenaPool * 0x026c4158) line 185 + 15 bytes NEWLINE JS_GC(JSContext * 0x026c4120) line 829 + 12 bytes NEWLINE LM_ReleaseDocument(MWContext_ * 0x026c2830, int 0) line 1868 + 10 bytes NEWLINE et_releasedocument_handler(ReleaseDocStruct * 0x026723d0) line 1436 + 19 bytes NEWLINE PL_HandleEvent(PLEvent * 0x026723d0) line 395 + 10 bytes NEWLINE et_SubEventLoop(QueueStackElement * 0x00ed5570) line 2311 + 9 bytes NEWLINE lm_wait_for_events(void * 0x00000000) line 2366 + 11 bytes NEWLINE _PR_NativeRunThread(void * 0x00ed9280) line 381 + 13 bytes NEWLINE _threadstartex(void * 0x00ed9440) line 212 + 13 bytes NEWLINE KERNEL32! bff86917() NEWLINE KERNEL32! bff84327() NEWLINE 00000006()
994|OPTGROUP not implemented yet|We don't yet support HTML 4.0's OPTGROUP element. Here's how it should work: NEWLINE NEWLINE <SELECT name="ComOS"> NEWLINE <OPTGROUP label="Fruits"> NEWLINE <OPTION>Apple NEWLINE <OPTION>Orange NEWLINE <OPTION>Pineapple NEWLINE </OPTGROUP> NEWLINE <OPTGROUP label="Meats"> NEWLINE <OPTION>Beef NEWLINE <OPTION>Chicken NEWLINE </OPTGROUP> NEWLINE <OPTGROUP label="Deserts"> NEWLINE <OPTION>Ice Cream NEWLINE <OPTION>Cake NEWLINE </OPTGROUP> NEWLINE </SELECT> NEWLINE NEWLINE Should render as: NEWLINE NEWLINE [ Fruits ] NEWLINE [ Meats ] NEWLINE [ Deserts ] NEWLINE NEWLINE When you mouseover "Deserts" for example, you get: NEWLINE NEWLINE [ Fruits ] NEWLINE [ Meats ] NEWLINE [ Deserts > Ice Cream ] NEWLINE [ Cake ]
995|table cellspacing not displaying correctly|Go to the above URL NEWLINE The 2nd table uses cellpadding and cellspacing. The expected result is that the NEWLINE spacing between cells be transparent. NEWLINE NEWLINE Another example of rendering behavior is: http://webgroup/ - the top table NEWLINE should have a border down the center and along the bottom, but it doesn't. The NEWLINE bottom table, however, that has the who's who in it, does have the cell boarders NEWLINE display correctly.
996|ASSERT in histbld.cpp|I was checking links on one of my HTML pages by quickly clicking on each NEWLINE link and then using Alt-Lft to get back to the previous page. I did this NEWLINE for about half a dozen links and then got an Assert: NEWLINE NEWLINE Line 176 of histbld.cpp in winfe NEWLINE NEWLINE // Adds a history item to the passed in menu. Also attaches a numerica NEWLINE accelerator to the menu item NEWLINE // if nPosition is less than 10. NEWLINE BOOL CHistoryMenuBuilder::AddHistoryToMenu(HMENU hMenu, int nPosition, BOOL NEWLINE bCurrent, History_entry* pEntry, int nMaxMenuItemLength) NEWLINE { NEWLINE char* lpszText; NEWLINE /*>*/ ASSERT(pEntry->title); NEWLINE if (!pEntry->title) NEWLINE return FALSE; NEWLINE NEWLINE <Snip> NEWLINE NEWLINE pEntry->title is 0x00ee5810, and points to "" NEWLINE NEWLINE The title pointer is non-zero, so I don't understand why it is asserting. NEWLINE ASSERT causes a break only when the argument is equal to zero, right? NEWLINE NEWLINE pEntry->address = NEWLINE "http://www.genealogy.org/~bryce/Harrington/Benjamin/HarringtonRoyalDescent.txt" NEWLINE pEntry->referer = NEWLINE "http://www.genealogy.org/~bryce/Harrington/Benjamin/index.html" NEWLINE NEWLINE Stack Trace: NEWLINE ------------ NEWLINE 1. CHistoryMenuBuilder::AddHistoryToMenu(HMENU__ * 0x0000096c, int 0, int 1, NEWLINE _History_entry * 0x011719d0, int 45) line 176 + 41 bytes NEWLINE 2. CHistoryMenuBuilder::FillItems(HMENU__ * 0x0000096c, _XP_List * 0x01171bd0, NEWLINE int 0, int 45, _History_entry * 0x011719d0) line 238 NEWLINE 3. CHistoryMenuBuilder::Fill(HMENU__ * 0x0000096c, int 45, eFillEnum eFILLALL) NEWLINE line 106 NEWLINE 4. CMainFrame::BuildHistoryMenu(CMenu * 0x00ee39a0 {CMenu}) line 600 NEWLINE 5. CMainFrame::OnInitMenuPopup(CMenu * 0x00ee39a0 {CMenu}, unsigned int 3, int NEWLINE 0) line 540 NEWLINE 6. CWnd::OnWndMsg(unsigned int 279, unsigned int 2412, long 3, long * NEWLINE 0x00cbf95c) line 1786 NEWLINE 7. CWnd::WindowProc(unsigned int 279, unsigned int 2412, long 3) line 1522 + 30 NEWLINE bytes NEWLINE 8. AfxCallWndProc(CWnd * 0x00d6d1b4 {CMainFrame hWnd=0x00000ce8}, HWND__ * NEWLINE 0x00000ce8, unsigned int 279, unsigned int 2412, long 3) line 210 + 26 bytes NEWLINE 9. AfxWndProc(HWND__ * 0x00000ce8, unsigned int 279, unsigned int 2412, long 3) NEWLINE line 363 + 25 bytes NEWLINE 10. AfxWndProcBase(HWND__ * 0x00000ce8, unsigned int 279, unsigned int 2412, NEWLINE long 3) line 203 + 21 bytes NEWLINE 11. KERNEL32! bff93d4b() NEWLINE 12. KERNEL32! bff73663() NEWLINE 13. KERNEL32! bff928e0() NEWLINE 14. 00cb8ae4() NEWLINE 15. 058f64c9()
997|Nondeterministic layout|(This is based on observations from a Solaris/sun4u/native-compiler NEWLINE build from a 10/1 CVS pull) NEWLINE NEWLINE The layout engine is behaving nondeterministically. Two examples: NEWLINE NEWLINE http://www.news.com/ NEWLINE NEWLINE Load the page and hit "reload" a number of times. Sometimes NEWLINE the layout will be correct, but often table cells will be NEWLINE much too large vertically (as in orders of magnitude wrong). NEWLINE NEWLINE http://www.xfree86.org/ NEWLINE NEWLINE Again, load the page and try repeated reloading. Fairly NEWLINE often the page will be laid out so that the title graphic NEWLINE is overlapping the introduction text.
998|table with iframe not rendering properly|the above URL utilizes in-line frames. The in-line frame has several rendering NEWLINE issues: displays horizontal scrollbar, displays vertical scrollbar and is NEWLINE displaced to the far rignt of the window.
999|float does not work properly|See this page for a test case of wildly strange layout that uses the float NEWLINE property. A page that gives a screen shot of the expected layout is available NEWLINE from this URL: NEWLINE NEWLINE http://www.w3.org/Style/CSS/Test/float
1000|Core Dump in laysel.c|Goto the above url and click on the talkback graphics...BOOM... NEWLINE NEWLINE Core Dump in laysel.c NEWLINE NEWLINE Here is a stack trace, The problem is the begin is NULL and being NEWLINE dereferenced. There was no selection when i went to the above url. NEWLINE NEWLINE NEWLINE #0 0x403f891d in lo_HitLine (context=0x861fe00, state=0x862de00, x=393, y=15, NEWLINE requireCaret=1 '\\001', result=0xbfffd2c4) at laysel.c:4487 NEWLINE ---Type <return> to continue, or q <return> to quit--- NEWLINE 4487 if ( begin->type != LO_LINEFEED && end->type == LO_LINEFEED ) { NEWLINE NEWLINE (gdb) bt NEWLINE #0 0x403f891d in lo_HitLine (context=0x861fe00, state=0x862de00, x=393, y=15, NEWLINE requireCaret=1 '\\001', result=0xbfffd2c4) at laysel.c:4487 NEWLINE #1 0x403f9236 in LO_Hit (context=0x861fe00, x=393, y=15, NEWLINE requireCaret=1 '\\001', result=0xbfffd2c4, layer=0x8668b00) at laysel.c:4941 NEWLINE #2 0x403f93de in LO_Click (context=0x861fe00, x=393, y=535, NEWLINE requireCaret=1 '\\001', layer=0x8668b00) at laysel.c:5009 NEWLINE #3 0x403f2fab in LO_StartSelection (context=0x861fe00, x=393, y=535, NEWLINE layer=0x8668b00) at laysel.c:426 NEWLINE #4 0x80baefd in fe_arm_link_action_for_layer (context=0x861fe00, NEWLINE layer=0x8668b00, layer_event=0x854e600) at lay.c:3461 NEWLINE #5 0x80d8fe1 in FE_HandleLayerEvent (context=0x861fe00, layer=0x8668b00, NEWLINE layer_event=0x854e600) at xfe.c:3280 NEWLINE #6 0x403dc138 in lo_html_event_callback (pContext=0x861fe00, pEle=0x86a99e8, NEWLINE event=1, pObj=0x826e020, status=EVENT_OK) at laylayer.c:305 NEWLINE #7 0x406b5bce in et_HandleEvent_JsEventAck (e=0x8652500) at et_moz.c:1311 NEWLINE #8 0x407c612f in PL_HandleEvent (self=0x8652500) at plevent.c:395 NEWLINE #9 0x407c6049 in PL_ProcessPendingEvents (self=0x8271280) at plevent.c:357 NEWLINE #10 0x80bfc18 in fe_event_processor_callback (closure=0x0, fd=0x8495134, NEWLINE id=0xbfffd428) at mozilla.c:1812 NEWLINE #11 0x408e093d in XtAppProcessEvent () NEWLINE #12 0x80bee2e in fe_EventLoop () at mozilla.c:1141 NEWLINE #13 0x80c20d1 in main (argc=1, argv=0xbfffeca4) at mozilla.c:3421 NEWLINE (gdb) NEWLINE NEWLINE NEWLINE (gdb) p begin NEWLINE $2 = (LO_Element *) 0x0
1001|Build bombs because detect_motif.sh returns "/usr/include"|Since my Xm.h is /usr/include/Xm/Xm.h, mozilla/config/mkdetect/detect_motif.sh NEWLINE returns /usr/include, which is then put before the build specific includes which NEWLINE ends up including /usr/include/ndbm.h rather than mozilla/dbm/include/ndbm.h. NEWLINE NEWLINE I'm just getting started with Mozilla so I don't know The Right Way to fix this NEWLINE yet but I would suggest either making detect_motif.sh return anything but NEWLINE /usr/include or reordering the includes so this comes after the build specific NEWLINE includes. NEWLINE NEWLINE My kludge was just to make detect set MOTIF_INCLUDE_DIR="", which works for me. NEWLINE NEWLINE My last CVS update was 2 Oct 1998.
1002|Placeholder bug for SU support in 5.0|Wherein we document progress on SmartUpdate support for 5.0
1003|ASSERT in fenet.cpp when closing mozilla with a RAM running|Exited Mozilla while watching a RAM. Caused ASSERT: NEWLINE NEWLINE Line 373 of winfe\\fenet.cpp: NEWLINE NEWLINE void FE_DeleteTempFilesNow() NEWLINE { NEWLINE char *pBuffer = new char[8192]; NEWLINE theApp.GetPrivateProfileString("Temporary File URL Resolution", NULL, "", NEWLINE pBuffer, 8192, AfxGetApp()->m_pszProfileName); NEWLINE NEWLINE // We have a double null terminated list of file names here. NEWLINE // Go ahead and go through each one, deleteing them all. NEWLINE char *pTraverse = pBuffer; NEWLINE int iRemoveResult = 0; NEWLINE while(*pTraverse != '\\0') { NEWLINE // Remove this file. NEWLINE iRemoveResult = remove(pTraverse); NEWLINE /*>*/ ASSERT(!iRemoveResult); NEWLINE NEWLINE // Go on to the next entry. NEWLINE while(*pTraverse != '\\0') { NEWLINE pTraverse++; NEWLINE } NEWLINE pTraverse++; NEWLINE } NEWLINE NEWLINE NEWLINE The value of iRemoveResult was -1 NEWLINE pTraverse was "C:\\windows\\temp\\tnc_web98_netscape.ram", NEWLINE the RAM I was in the middle of watching. NEWLINE NEWLINE NEWLINE Stack trace: NEWLINE ------------ NEWLINE FE_DeleteTempFilesNow() line 373 + 42 bytes NEWLINE CNetscapeApp::ExitInstance() line 2333 NEWLINE CNetscapeApp::Run() line 1676 + 11 bytes NEWLINE AfxWinMain(HINSTANCE__ * 0x00400000, HINSTANCE__ * 0x00000000, char * NEWLINE 0x8168d907, int 1) line 52 + 11 bytes NEWLINE WinMain(HINSTANCE__ * 0x00400000, HINSTANCE__ * 0x00000000, char * 0x8168d907, NEWLINE int 1) line 33 + 21 bytes NEWLINE WinMainCRTStartup() line 330 + 57 bytes NEWLINE KERNEL32! bff88f75() NEWLINE KERNEL32! bff88e23() NEWLINE KERNEL32! bff8783f()
1004|should support CSS2 system colors|CSS2 provides a mechanism for the author to designate a color based on NEWLINE aliases to system color defaults, rather than an explicit color name or hex NEWLINE value. Specified values are outlined at the above URL. NEWLINE NEWLINE Support for this would be particularly helpful to those trying to build an NEWLINE appliction interface with the native look and feel using only HTML/CSS.
1005|Mozilla Crashes when there are pop up windows.|Mozilla 1998-10-03 crashed when viewing this NEWLINE page http://209.139.38.36/lucky/index.html which poped up an new browser window. NEWLINE At that time Norton Antivirus 5 and ICQ were running. NEWLINE My PC has an AMD k6 233MHz cpu and 64Mb Ram and a 4,3Gb Quantum Fireball SE. NEWLINE My OS is win98 ( windows 4.10- 1998)
1006|we don't support text-indent|fix that
1007|class=foo needs to be case insensitive, I think|according to the css spec, most rules are case insensitive (see 4.1.3 in the css NEWLINE spec) except where outside of control of the css2 spec (like html's id/class NEWLINE attributes). I dug around in the html4.0 spec and couldn't find a resolution on NEWLINE this so I guess you need to dig furthur :-( NEWLINE NEWLINE However, some test cases I've been seeing from the www.webstandards.org lead me NEWLINE to believe otherwise. You might be able to get some info out of Sue Sims NEWLINE <sjacct@worldnet.att.net>
1008|XPViewer Crashes on Start-up|I crash with an unhandled exception; my stack trace says: NEWLINE NEWLINE nsBrowserWindow::GetWebShell(nsBrowserWindow * const 0x012467b0, nsIWebShell * NEWLINE &) line 2155 + 9 bytes NEWLINE nsWebShell::OnProgress(nsWebShell * const 0x01246ef8, nsIURL * 0x01290200, int NEWLINE 0, int 3829) line 1859 NEWLINE nsDocumentBindInfo::OnProgress(nsDocumentBindInfo * const 0x01290340, nsIURL * NEWLINE 0x01290200, int 0, int 3829) line 1221 NEWLINE OnProgressProxyEvent::HandleEvent(OnProgressProxyEvent * const 0x0124ecf0) line NEWLINE 481 + 45 bytes NEWLINE StreamListenerProxyEvent::HandlePLEvent(PLEvent * 0x0124ecf4) line 418 + 12 NEWLINE bytes NEWLINE PL_HandleEvent(PLEvent * 0x0124ecf4) line 395 + 10 bytes NEWLINE PL_ProcessPendingEvents(PLEventQueue * 0x01205df0) line 357 + 9 bytes NEWLINE _md_EventReceiverProc(void * 0x000a010c, unsigned int 49402, unsigned int 0, NEWLINE long 18898416) line 675 + 9 bytes NEWLINE USER32! 77e71250() NEWLINE 01205df0()
1009|CSSDeclarationImpl Memory Leak|Purify reports seven CSSDeclarationImpl objects leaked when viewing demo1. NEWLINE NEWLINE I looked at it for a while, but I couldn't easily determine the problem. I did NEWLINE change some of the html\\style\\src code to make sure the NS_ADDREF(), NEWLINE NS_ADDREF_THIS(), NS_NEWXPCOM(), and NS_DELETEXPCOM() macros were used. This NEWLINE makes Kipp's ref counting code report more information NEWLINE NEWLINE I also added a missing NS_RELEAES(declaration) to NEWLINE CSSParserImpl::ParseDeclarations(), but that didn't fix it. NEWLINE NEWLINE Here's the Purify output: NEWLINE [W] MLK: Memory leak of 336 bytes from 7 blocks allocated in NEWLINE CSSDeclarationImpl::new(UINT) NEWLINE Distribution of leaked blocks NEWLINE Allocation location NEWLINE new(UINT) [new.cpp:23] NEWLINE CSSDeclarationImpl::new(UINT) [nsCSSDeclaration.cpp:832] NEWLINE NS_NewCSSDeclaration(nsICSSDeclaration * *) [nsCSSDeclaration.cpp:2493] NEWLINE CSSParserImpl::ParseDeclarationBlock(int *,int) [nsCSSParser.cpp:1032] NEWLINE } NEWLINE nsICSSDeclaration* declaration = nsnull; NEWLINE if (NS_OK == NS_NewCSSDeclaration(&declaration)) { NEWLINE => PRInt32 count = 0; NEWLINE for (;;) { NEWLINE PRInt32 hint; NEWLINE if (ParseDeclaration(aErrorCode, declaration, NEWLINE aCheckForBraces, &hint)) { NEWLINE CSSParserImpl::ParseDeclarations(nsString const&,nsIURL *,nsIStyleRule NEWLINE *&) [nsCSSParser.cpp:438] NEWLINE nsGenericHTMLElement::SetAttribute(nsIAtom *,nsString const&,int) NEWLINE [nsGenericHTMLElement.cpp:1153] NEWLINE nsHTMLSpanElement::SetAttribute(nsIAtom *,nsString const&,int) NEWLINE [nsHTMLSpanElement.cpp:61] NEWLINE AddAttributes [nsHTMLContentSink.cpp:448] NEWLINE SinkContext::OpenContainer(nsIParserNode const&) NEWLINE [nsHTMLContentSink.cpp:912] NEWLINE HTMLContentSink::OpenContainer(nsIParserNode const&) NEWLINE [nsHTMLContentSink.cpp:1687] NEWLINE CNavDTD::OpenContainer(nsIParserNode const&,int) [CNavDTD.cpp:2359] NEWLINE CNavDTD::HandleDefaultStartToken(CToken *,nsHTMLTag,nsIParserNode&) NEWLINE [CNavDTD.cpp:576] NEWLINE CNavDTD::HandleStartToken(CToken *) [CNavDTD.cpp:661] NEWLINE NavDispatchTokenHandler(CToken *,nsIDTD *) [CNavDTD.cpp:293] NEWLINE CTokenHandler::()(CToken *,nsIDTD *) [nsTokenHandler.cpp:80] NEWLINE [W] MLK: Memory leak of 228 bytes from 3 blocks allocated in NEWLINE CSSDeclarationImpl::AppendValue(int,nsCSSValue const&) NEWLINE [W] MLK: Memory leak of 36 bytes from 3 blocks allocated in NEWLINE CSSDeclarationImpl::AppendValue(int,nsCSSValue const&) NEWLINE [W] MLK: Memory leak of 64 bytes from 2 blocks allocated in NEWLINE nsVoidArray::InsertElementAt(void *,int) NEWLINE [W] MLK: Memory leak of 48 bytes from 4 blocks allocated in NEWLINE CSSDeclarationImpl::AppendValue(int,nsCSSValue const&) NEWLINE [W] MLK: Memory leak of 368 bytes from 4 blocks allocated in NEWLINE CSSDeclarationImpl::AppendValue(int,nsCSSValue const&) NEWLINE [W] MLK: Memory leak of 128 bytes from 4 blocks allocated in NEWLINE nsVoidArray::InsertElementAt(void *,int)
1010|Links in don't work inside rowspan'd cells|In the test case below, links only work for the top half of the list items. The NEWLINE links in the lower list items don't work (no mouseover feedback, etc.) I was NEWLINE able to isolate the problem to the "rowspan" attribute on the first TD. Removing NEWLINE either the rowspan attribute, or the second row of the table solves the problem. NEWLINE This is a simplified test case from the "real" page, which is the URL mentioned NEWLINE above. I've noticed this on some other tables pages, too. NEWLINE NEWLINE <TABLE BORDER=1 WIDTH="100%"> NEWLINE <tr> NEWLINE <TD VALIGN=TOP rowspan=2> NEWLINE <ul> NEWLINE <li><A href="500in1ul.html">500 list items in a single UL</a></li> NEWLINE <li><a href="500in25uls.html">500 list items in 25 top-level ULs</a></li> NEWLINE <li><a href="500in25nesteduls.html">500 list items in a 25-level nested NEWLINE UL</a></li> NEWLINE <li><a href="300textfields.html">300 text input form fields</a></li> NEWLINE <li><a href="test6.html">Massively nested tables</a></li> NEWLINE <li><a href="12nesteddivs.html">Massively nested DIVs with padding and NEWLINE borders</a></li> NEWLINE <li><a href="select.html">1000 OPTIONs in a single SELECT form element</a></li> NEWLINE <li><a href="wbtbltxt.html">10,000 cell table with text in each cell</a></li> NEWLINE <li><a href="wbtblclr.html">10,000 cell table with text and color in each NEWLINE cell</a></li> NEWLINE <li><a href="wblnks.html">A 713k HTML document packed full of links</a></li> NEWLINE <li><a href="wbbightm.html">Extraordinarily long HTML doc (approx 1 NEWLINE Meg)</a></li> NEWLINE <li><a href="wbclrtxt.html">Another large HTML doc, with colored text</a></li> NEWLINE <li><a href="wbhtmmix.html">A relatively large (150k) document with an average NEWLINE mix of HTML NEWLINE content</a></li> NEWLINE </ul> NEWLINE </TD> NEWLINE <td>foo 1</td> NEWLINE <td>foo 2</td> NEWLINE </TR> NEWLINE <TR> NEWLINE <td>Foo 3</td> NEWLINE <td>Foo 4</td> NEWLINE </TR> NEWLINE </TABLE> NEWLINE </BODY> NEWLINE </HTML>
1011|news.com columns are getting mushed to the left|Couldn't simplify to a test case yet, but it looks pretty nasty.
1012|crash on clicking on link|links finally work on autoconf builds but NEWLINE I got the following after clicking on a link today NEWLINE NEWLINE Program received signal SIGSEGV, Segmentation fault. NEWLINE 0x8239cf0 in lo_PopList () NEWLINE (gdb) info stack NEWLINE #0 0x8239cf0 in lo_PopList () NEWLINE #1 0x822840f in lo_ProcessBlockQuoteElement () NEWLINE #2 0x8211209 in lo_rl_InitRelayoutState () NEWLINE #3 0x8210d3a in lo_rl_InitRelayoutState () NEWLINE #4 0x8210682 in LO_RelayoutOnResize () NEWLINE #5 0x82107e2 in LO_RelayoutFromElement () NEWLINE #6 0x8219919 in LO_XYToElement () NEWLINE #7 0x812891a in INTL_Relayout () NEWLINE #8 0x4009e8ea in XtAppProcessEvent () NEWLINE #9 0x811af98 in fe_EventLoop () NEWLINE #10 0x811cd25 in main () NEWLINE #11 0x80ee37b in _start ()
1013|toolbar can't undo unwanted customizations|I accidently screwed up the toolbars by creating 2 url bars. I was also able to NEWLINE cut out the original url bar. Is there any way for me to remove the changes I NEWLINE made? my email address is shadow777@softhome.net
1014|Crash in laysel.c|At line 4487 on the source from CVS on 10/4. Function lo_HitLine. NEWLINE NEWLINE lo_GetLineEnds(context, state, line, & begin, & end); NEWLINE >> this returns begin == NULL NEWLINE /* lo_GetLineEnds returns the start of the next line for 'end' */ NEWLINE if ( end ) { NEWLINE end = end->lo_any.prev; NEWLINE } else { NEWLINE /* Last line. We know that the last line only has one element. */ NEWLINE end = begin; NEWLINE } NEWLINE /* Except for cases where the entire line is a line feed, don't select NEWLINE the end line-feed. */ NEWLINE >>> IT THEN CRASHES HERE NEWLINE if ( begin->type != LO_LINEFEED && end->type == LO_LINEFEED ) { NEWLINE NEWLINE HOW DID I GET THERE: NEWLINE With Mozilla running, I started the HTML WIZARED in office 97. NEWLINE It said there is an updated version of the wizarad, do I want to update. NEWLINE I said yes. I dont know the URL.
1015|navigator.mimeTypes not implemented|Setting all current Open/Normal to M4.
1016|nsIRenderingContext woes|1. It needs to live in gfx/public NEWLINE 2. It needs to be XPCOM'ized NEWLINE NEWLINE thanks
1017|Crash in DisplayPixmap()|Do: Edit->editToolbars ... BOOM...crash with the following stack trace: NEWLINE NEWLINE NEWLINE Program received signal SIGSEGV, Segmentation fault. NEWLINE 0x80b34de in _IMGCB_DisplayPixmap (img_cb=0x8599b60, op=12, dpy_cx=0x8597e00, NEWLINE image=0x8630680, mask=0x0, x=0, y=0, x_offset=0, y_offset=0, width=50, NEWLINE height=50, req_w=0, req_h=0) at images.c:667 NEWLINE 667 Drawable drawable = fe_drawable->xdrawable; NEWLINE NEWLINE #0 0x80b34de in _IMGCB_DisplayPixmap (img_cb=0x8599b60, op=12, NEWLINE dpy_cx=0x8597e00, image=0x8630680, mask=0x0, x=0, y=0, x_offset=0, NEWLINE y_offset=0, width=50, height=50, req_w=0, req_h=0) at images.c:667 NEWLINE #1 0x404c1aaa in IL_DisplaySubImage (image_req=0x8634c60, x=0, y=0, NEWLINE x_offset=0, y_offset=0, width=50, height=50) at if.cpp:2200 NEWLINE #2 0x403c8daa in lo_DisplayImageWithoutCompositor (context=0x8597e00, NEWLINE image=0x863b00c) at laydisp.c:429 NEWLINE #3 0x403c923f in lo_ClipImage (context=0x8597e00, image=0x863b00c, x=0, y=0, NEWLINE width=1073741823, height=1073741823) at laydisp.c:565 NEWLINE #4 0x403c97d3 in lo_DisplayElement (context=0x8597e00, tptr=0x863b00c, NEWLINE base_x=0, base_y=0, x=0, y=0, width=1073741823, height=1073741823) NEWLINE at laydisp.c:744 NEWLINE #5 0x403f6023 in lo_DisplayLine (context=0x8597e00, state=0x8636600, NEWLINE line_num=0, x=0, y=0, w=1073741823, h=1073741823) at layout.c:2879 NEWLINE #6 0x403fcb5f in lo_UpdateStateAfterFlushingLine (context=0x8597e00, NEWLINE state=0x8636600, linefeed=0x863b0b0, inRelayout=0 '\\000') at layout.c:8164 NEWLINE #7 0x403f4abf in lo_FlushLineList (context=0x8597e00, state=0x8636600, NEWLINE break_type=0, clear_type=0, breaking=0 '\\000') at layout.c:1611 NEWLINE #8 0x4041ff04 in lo_InsertLineBreak (context=0x8597e00, state=0x8636600, NEWLINE break_type=0, clear_type=0, breaking=0 '\\000') at laytext.c:673 NEWLINE #9 0x40420506 in lo_SetLineBreakState (context=0x8597e00, state=0x8636600, NEWLINE breaking=0 '\\000', break_type=0, linefeed_state=1, relayout=0 '\\000') NEWLINE at laytext.c:963 NEWLINE #10 0x403f659c in lo_CloseOutLayout (context=0x8597e00, state=0x8636600) NEWLINE at layout.c:3218 NEWLINE #11 0x403f7d4d in LO_ProcessTag (data_object=0x8553900, tag=0x0, status=1) NEWLINE at layout.c:4422 NEWLINE #12 0x4043765d in EDT_ProcessTag (data_object=0x8553900, tag=0x0, status=1) NEWLINE at editor.cpp:963 NEWLINE #13 0x404df414 in PA_MDLComplete (stream=0x86348a0) at pa_parse.c:1743 NEWLINE #14 0x404dddd9 in pa_check_doc_data_count (stream=0x86348a0) at pa_parse.c:390 NEWLINE #15 0x404de0c7 in PA_DropDocData (stream=0x86348a0) at pa_parse.c:538 NEWLINE #16 0x404de147 in pa_cleanup_after_netlib (stream=0x86348a0) at pa_parse.c:553 NEWLINE #17 0x404de169 in pa_netlib_stream_complete (stream=0x86348a0) NEWLINE at pa_parse.c:559 NEWLINE #18 0x404fee27 in net_CvtCharCodeComplete (stream=0x8634820) at net_junk.c:53 NEWLINE #19 0x404b9abc in IL_ViewStream (format_out=1, newshack=0x0, urls=0x85e6000, NEWLINE cx=0x8597e00) at external.cpp:310 NEWLINE #20 0x4029c775 in NET_StreamBuilder (format_out=1, URL_s=0x85e6000, NEWLINE context=0x8597e00) at mkstream.c:242 NEWLINE #21 0x402bc79f in NET_CacheConverter (format_out=1, converter_obj=0x0, NEWLINE URL_s=0x85e6000, window_id=0x8597e00) at mkcache.c:1688 NEWLINE #22 0x4029c775 in NET_StreamBuilder (format_out=65, URL_s=0x85e6000, NEWLINE context=0x8597e00) at mkstream.c:242 NEWLINE #23 0x4030a064 in net_setup_http_stream (ce=0x855b800) at mkhttp.c:2763 NEWLINE #24 0x4030b189 in net_ProcessHTTP (ce=0x855b800) at mkhttp.c:3404 NEWLINE #25 0x4029066f in NET_ProcessNet (ready_fd=0x8439e00, fd_type=2) NEWLINE at mkgeturl.c:3369 NEWLINE #26 0x4029b701 in net_process_net_timer_callback (closure=0x0) NEWLINE at mkselect.c:189 NEWLINE #27 0x80d86b8 in fe_do_timeout (p=0x85f4720, id=0xbfffd40c) at xfe.c:2946 NEWLINE #28 0x408ea8ad in XtAppProcessEvent () NEWLINE #29 0x80beed2 in fe_EventLoop () at mozilla.c:1133 NEWLINE #30 0x80c2191 in main (argc=1, argv=0xbfffec84) at mozilla.c:3421
1018|Crash on this URL|NGLayout crashes when attempting to load this URL. I'm not sure if it's a style NEWLINE thing or a layout thing. NEWLINE NEWLINE Stack trace: NEWLINE CSSStyleSheetImpl::~CSSStyleSheetImpl() line 710 + 17 bytes NEWLINE CSSStyleSheetImpl::`scalar deleting destructor'() + 20 bytes NEWLINE CSSStyleSheetImpl::Release(CSSStyleSheetImpl * const 0x01298970) line 715 + 146 NEWLINE bytes NEWLINE nsDocument::~nsDocument() line 201 + 12 bytes NEWLINE nsMarkupDocument::~nsMarkupDocument() line 48 + 13 bytes NEWLINE nsHTMLDocument::~nsHTMLDocument() line 386 + 28 bytes NEWLINE nsHTMLDocument::`scalar deleting destructor'() + 20 bytes NEWLINE nsDocument::Release(nsDocument * const 0x012b7830) line 264 + 108 bytes NEWLINE nsHTMLDocument::Release(nsHTMLDocument * const 0x012b7830) line 420 + 9 bytes NEWLINE PresShell::~PresShell() line 357 + 18 bytes NEWLINE PresShell::`scalar deleting destructor'() + 20 bytes NEWLINE PresShell::Release(PresShell * const 0x01260bd0) line 304 + 32 bytes NEWLINE DocumentViewerImpl::~DocumentViewerImpl() line 201 + 18 bytes NEWLINE DocumentViewerImpl::`scalar deleting destructor'() + 20 bytes NEWLINE DocumentViewerImpl::Release(DocumentViewerImpl * const 0x012be580) line 143 + NEWLINE 105 bytes NEWLINE nsWebShell::Embed(nsWebShell * const 0x01255490, nsIContentViewer * 0x012bd630, NEWLINE char * 0x00000000, nsISupports * 0x00000000) line 527 + 31 bytes NEWLINE nsDocumentBindInfo::OnStartBinding(nsDocumentBindInfo * const 0x012bda30, nsIURL NEWLINE * 0x012bda90, char * 0x012bc5e0) line 1285 + 36 bytes NEWLINE OnStartBindingProxyEvent::HandleEvent(OnStartBindingProxyEvent * const NEWLINE 0x012bc590) line 455 + 38 bytes NEWLINE StreamListenerProxyEvent::HandlePLEvent(PLEvent * 0x012bc594) line 421 + 12 NEWLINE bytes NEWLINE PL_HandleEvent(PLEvent * 0x012bc594) line 395 + 10 bytes NEWLINE PL_ProcessPendingEvents(PLEventQueue * 0x01215ea0) line 357 + 9 bytes NEWLINE _md_EventReceiverProc(void * 0x00430154, unsigned int 49326, unsigned int 0, NEWLINE long 18964128) line 675 + 9 bytes NEWLINE USER32! 77e71250()
1019|Text decorations break at container boundaries|Text decorations should span container boundaries. NEWLINE For: NEWLINE <u>underline <b>bold</b></u> NEWLINE the underline should be continuous.
1020|VC6 cannot compile nsdlg|makefile.win in cmd/winfe/prefs/nsdlg/src/ tries to use _MSC_VER to determine NEWLINE if VC6 is being used. _MSC_VER is not set by default in the enviroment. NEWLINE NEWLINE Either the makefile should be changed, or the requirement of setting _MSC_VER NEWLINE put into the winbuild documentation.
1021|[html.css] Widgets should use CSS2 system colors (and fonts)|The default background colors for most widgets except buttons are very dark. It NEWLINE would be nice if the widget library could set them to reasonable defaults and NEWLINE possibly even get the settings from the operating system. NEWLINE NEWLINE This can be tested by uncommenting the code in NEWLINE forms\\nsFormControlFrame::SetColors and switching to Standard mode in the NEWLINE viewer upon viewing test8.
1022|Only need to include olectl.h in VC6|The windows include file olectlid.h is included by NEWLINE cmd/wincom/dllcom/pub/lic/dlliface.h NEWLINE NEWLINE Under VC6, the olectlid file actually only includes olectl.h, and as such NEWLINE doesn't need to be separately included. If the file is included, a friendly NEWLINE warning message appears to this effect. NEWLINE NEWLINE This patch should clean this up. NEWLINE RCS file: /cvsroot/mozilla/cmd/wincom/dllcom/public/dlliface.h,v NEWLINE retrieving revision 3.1 NEWLINE diff -u -c -r3.1 dlliface.h NEWLINE *** dlliface.h 1998/03/28 03:16:01 3.1 NEWLINE --- dlliface.h 1998/10/08 01:29:47 NEWLINE *************** NEWLINE *** 30,35 **** NEWLINE --- 30,37 ---- NEWLINE #include <objbase.h> NEWLINE #endif NEWLINE #include <olectl.h> NEWLINE + #if (_MSC_VER < 1200) /* Not required in VC6 */ NEWLINE #include <olectlid.h> NEWLINE + #endif NEWLINE NEWLINE #endif // __DllInterfaces_H NEWLINE NEWLINE NEWLINE -dave
1023|Modifying textDecoration via CSS OM doesn't force reflow|I'm modifying the textDecoration style property on an element, but it doesn't NEWLINE appear. I find that by forcing a reflow, the right thing appears. Thus, I assume NEWLINE we need to wire something up to make this work like it should. Here's the test NEWLINE case: NEWLINE NEWLINE <html> NEWLINE <body> NEWLINE <a href="foo" NEWLINE onmouseover="this.style.textDecoration='underline'" NEWLINE onmouseout="this.style.textDecoration='none'">Foo</a> NEWLINE </body> NEWLINE </html> NEWLINE NEWLINE As you pass your mouse over the element, the underline should appear (it's on NEWLINE by default). If you pass your mouse out of the element, the underline should go NEWLINE away.
1024|Modifying style.top actually modifies style.left|Here's the test case: NEWLINE <a href="foo" NEWLINE onmousedown="this.style.position='relative'; this.style.top='10px'">Foo</a> NEWLINE NEWLINE It should move the word "foo" down by 10 pixels when you mousedown on the link. NEWLINE Instead, it moves the word "foo" to the right by ten pixels.
1025|NGLayout crashes on this URL|NGLayout loads the URL, renders the text very, very small, and then crashes. My NEWLINE stack trace makes it sound like something to do with the Intel image code NEWLINE optimizations, but what do I know? NEWLINE NEWLINE h2v2_merged_upsample_mmx(jpeg_decompress_struct * 0x0179d6f8, unsigned char * * NEWLINE * 0x01828210, unsigned int 0, unsigned char * * 0x0012fca0) line 606 NEWLINE h2v2_merged_upsample(jpeg_decompress_struct * 0x0179d6f8, unsigned char * * * NEWLINE 0x01828210, unsigned int 0, unsigned char * * 0x0012fca0) line 326 + 21 bytes NEWLINE merged_2v_upsample(jpeg_decompress_struct * 0x0179d6f8, unsigned char * * * NEWLINE 0x01828210, unsigned int * 0x0182823c, unsigned int 8, unsigned char * * NEWLINE 0x01826910, unsigned int * 0x0012fd08, unsigned int 1) line 194 + 24 bytes NEWLINE process_data_simple_main(jpeg_decompress_struct * 0x0179d6f8, unsigned char * * NEWLINE 0x01826910, unsigned int * 0x0012fd08, unsigned int 1) line 369 + 46 bytes NEWLINE jpeg_read_scanlines(jpeg_decompress_struct * 0x0179d6f8, unsigned char * * NEWLINE 0x01826910, unsigned int 1) line 173 + 28 bytes NEWLINE output_jpeg_scanlines(il_container_struct * 0x013581a0, int -2) line 596 + 18 NEWLINE bytes NEWLINE il_jpeg_write(il_container_struct * 0x013581a0, unsigned char * 0x0176d3b8, long NEWLINE 128) line 835 + 11 bytes NEWLINE IL_StreamWrite(il_container_struct * 0x013581a0, unsigned char * 0x0176d3b8, NEWLINE long 128) line 805 + 18 bytes NEWLINE NetReaderImpl::Write(unsigned char * 0x0176d3b8, long 128) line 92 + 20 bytes NEWLINE ImageConsumer::OnDataAvailable(ImageConsumer * const 0x01368d80, nsIURL * NEWLINE 0x013599c0, nsIInputStream * 0x01357920, int 31) line 234 NEWLINE nsDocumentBindInfo::OnDataAvailable(nsDocumentBindInfo * const 0x0125dfb0, NEWLINE nsIURL * 0x013599c0, nsIInputStream * 0x01357920, int 31) line 1331 + 30 bytes NEWLINE OnDataAvailableProxyEvent::HandleEvent(OnDataAvailableProxyEvent * const NEWLINE 0x0135ed10) line 575 + 45 bytes NEWLINE StreamListenerProxyEvent::HandlePLEvent(PLEvent * 0x0135ed14) line 421 + 12 NEWLINE bytes NEWLINE PL_HandleEvent(PLEvent * 0x0135ed14) line 395 + 10 bytes NEWLINE PL_ProcessPendingEvents(PLEventQueue * 0x01215ea0) line 357 + 9 bytes NEWLINE _md_EventReceiverProc(void * 0x00660244, unsigned int 49353, unsigned int 0, NEWLINE long 18964128) line 675 + 9 bytes NEWLINE USER32! 77e71250() NEWLINE 01215ea0()
1026|NGLayout does not incrementally display fixed width tables|When displaying a fixed width table, NGLayout downloads the entire table before NEWLINE rendering anything. Fixed width tables should render something immediately while NEWLINE the rest of the table is still downloading. IE5 seems to do the right thing with NEWLINE this URL. NEWLINE NEWLINE On a totally separate note, it seems painfully slow to resize the window with NEWLINE this large of a document, compared to IE at least.
1027|segfault traceback under redhat 5.1, XF86_SDD|just clicking 'back'. NEWLINE NEWLINE (gdb) where NEWLINE #0 0x400c35dc in ?? () from /usr/X11R6/lib/libXt.so.6 NEWLINE #1 0x400d746e in ?? () from /usr/X11R6/lib/libXt.so.6 NEWLINE #2 0x400e6c40 in ?? () from /usr/X11R6/lib/libXt.so.6 NEWLINE #3 0x8235150 in XfeToolTipShellSetString () NEWLINE #4 0x8234672 in ItemPostToolTip () NEWLINE #5 0x8234747 in StageTwoTimeout () NEWLINE #6 0x400cf8c9 in ?? () from /usr/X11R6/lib/libXt.so.6 NEWLINE #7 0x816e452 in fe_EventLoop () NEWLINE #8 0x81708a1 in main () NEWLINE NEWLINE Its certainly critical, try to test for null pointers etc, NEWLINE not just hand them off to the X libraries and get a crash there. NEWLINE NEWLINE It might not really be P1. but freaking netscape self combusts NEWLINE randomly too. linux doesnt have a real browser yet.
1028|Composer: insert link causes crash|When inserting a link into a page in Composer, Mozilla dies. NEWLINE NEWLINE To repeat: NEWLINE - start Mozilla NEWLINE - File->New->Blank Page NEWLINE - Click the link icon NEWLINE - Text: "foo", link to: "bar" NEWLINE -foom! NEWLINE NEWLINE Program received signal SIGSEGV, Segmentation fault. NEWLINE 0xc9b4cc1 in _XimLocalFilter () NEWLINE (gdb) bt NEWLINE #0 0xc9b4cc1 in _XimLocalFilter () NEWLINE #1 0xc98e53d in XFilterEvent () NEWLINE #2 0xc8f496e in _XtDefaultDispatcher () NEWLINE #3 0xc8f4bd3 in XtDispatchEvent () NEWLINE #4 0xc889d14 in RemoveAllPMgr () NEWLINE #5 0xc8ea3e3 in XtCallCallbackList () NEWLINE #6 0xc8f0cb3 in Phase2Callbacks () NEWLINE #7 0xc8f0bd1 in Recursive () NEWLINE #8 0xc8f0f72 in XtPhase2Destroy () NEWLINE #9 0xc8f115e in _XtDoPhase2Destroy () NEWLINE #10 0xc8f4bf4 in XtDispatchEvent () NEWLINE #11 0xc8fdc63 in XtAppProcessEvent () NEWLINE #12 0x808cfdc in fe_EventLoop () NEWLINE #13 0x808ef29 in main () NEWLINE #14 0x805ebc4 in __start () NEWLINE NEWLINE BSD/OS 4.0, X11R6, Motif 1.2.2, autoconf build, CVS as of 1998/10/07 NEWLINE Clean build with -O2 NEWLINE ../configure --with-nspr=/usr/local/nspr --enable-editor --enable-mail-compose NEWLINE --enable-ldap
1029|mozilla/cmd/xfe/icons/mkicons segfaults.|mozilla/cmd/xfe/icons/mkicons builds ok, but segfaults when run. NEWLINE linux 2.0.34, lesstiff-cvs, gcc 2.7.2, autoconf & classic build. NEWLINE NEWLINE ... NEWLINE NEWLINE NEWLINE ../../../config/./nsinstall -R -m 444 ./icon_extern.h NEWLINE /usr/home/jip/moz/mozilla/cmd/xfe/icons/icondata.h ../../../dist/public/xfeicons NEWLINE c++ -DXP_UNIX -g -O2 -DUSE_AUTOCONF=1 -DSTDC_HEADERS=1 -DHAVE_ST_BLKSIZE=1 NEWLINE -DHAVE_ST_RDEV=1 -DHAVE_TZNAME=1 -DHAVE_DIRENT_H=1 -DSTDC_HEADERS=1 NEWLINE -DHAVE_SYS_WAIT_H=1 -DTIME_WITH_SYS_TIME=1 -DHAVE_FCNTL_H=1 -DHAVE_LIMITS_H=1 NEWLINE -DHAVE_MALLOC_H=1 -DHAVE_PATHS_H=1 -DHAVE_STRINGS_H=1 -DHAVE_UNISTD_H=1 NEWLINE -DHAVE_SYS_FILE_H=1 -DHAVE_SYS_IOCTL_H=1 -DHAVE_SYS_TIME_H=1 -DHAVE_GETOPT_H=1 NEWLINE -DHAVE_SYS_CDEFS_H=1 -DHAVE_ALLOCA_H=1 -DHAVE_ALLOCA=1 -DHAVE_UNISTD_H=1 NEWLINE -DHAVE_GETPAGESIZE=1 -DHAVE_MMAP=1 -DRETSIGTYPE=void -DHAVE_STRCOLL=1 NEWLINE -DHAVE_STRFTIME=1 -DHAVE_UTIME_NULL=1 -DHAVE_VPRINTF=1 -DHAVE_FTIME=1 NEWLINE -DHAVE_GETCWD=1 -DHAVE_GETHOSTNAME=1 -DHAVE_GETWD=1 -DHAVE_MKDIR=1 NEWLINE -DHAVE_MKTIME=1 -DHAVE_PUTENV=1 -DHAVE_RMDIR=1 -DHAVE_SELECT=1 -DHAVE_SOCKET=1 NEWLINE -DHAVE_STRCSPN=1 -DHAVE_STRDUP=1 -DHAVE_STRERROR=1 -DHAVE_STRSPN=1 NEWLINE -DHAVE_STRSTR=1 -DHAVE_STRTOL=1 -DHAVE_STRTOUL=1 -DHAVE_UNAME=1 -DHAVE_QSORT=1 NEWLINE -DHAVE_SNPRINTF=1 -DHAVE_GETTIMEOFDAY=1 -DGETTIMEOFDAY_TWO_ARGS=1 NEWLINE -DHAVE_SYSERRLIST=1 -DHAVE_LIBC=1 -DHAVE_LIBM=1 -DHAVE_LIBDL=1 -DHAVE_LIBINTL=1 NEWLINE -DHAVE_LIBL=1 -DHAVE_LIBX11=1 -DHAVE_LIBXEXT=1 -DHAVE_LIBICE=1 -DHAVE_LIBSM=1 NEWLINE -DHAVE_LIBXT=1 -DHAVE_LIBXMU=1 -DHAVE_LIBXPM=1 -I/usr/local/nspr/include NEWLINE -UDEBUG -DNDEBUG -DTRIMMED -DNSPR20 -DNETSCAPE -DOSTYPE=\\"Linux2.0\\" NEWLINE -DMOZILLA_CLIENT -DCookieManagement -DNO_SECURITY -DLAYERS -DUNIX_EMBED NEWLINE -DX_PLUGINS -DNSPR -DJS_THREADSAFE -DUNIX_ASYNC_DNS -DMOZ_USER_DIR=\\".mozilla\\" NEWLINE -DXFE -I../../../dist/public/img -I../../../dist/public/java NEWLINE -I../../../dist/public/jtools -I../../../dist/public/layer NEWLINE -I../../../dist/public/nspr -I../../../dist/public/util NEWLINE -I../../../dist/public/xp -I../../../dist/public/xpcom NEWLINE -I../../../dist/public/jpeg -I../../../dist/public/png NEWLINE -I../../../dist/public/zlib -I/usr/home/jip/moz/mozilla/include NEWLINE -I/usr/home/jip/moz/mozilla/cmd/xfe/icons/. -I. -I/usr/include/g++ NEWLINE -I../../../dist/./include -o mkicons NEWLINE -I/usr/home/jip/moz/mozilla/modules/libimg/src NEWLINE /usr/home/jip/moz/mozilla/cmd/xfe/icons/mkicons.cpp NEWLINE ../../../dist/./lib/libimg.a ../../../dist/./lib/libstubsj.a NEWLINE ../../../dist/./lib/libutil.a ../../../dist/./lib/libxp.a NEWLINE ../../../dist/./lib/libxpcom.a ../../../dist/./lib/libpng.a NEWLINE ../../../dist/./lib/libzlib.a ../../../dist/./lib/libjpeg.a NEWLINE -L/usr/local/nspr/lib -lplds21 -lplc21 -lnspr21 -L/lib -ldl -lc -lm NEWLINE /usr/home/jip/moz/mozilla/cmd/xfe/icons/mkicons.cpp: In function `void NEWLINE _IMGCB_NewPixmap(struct IMGCB *, long int, void *, long int, long int, struct NEWLINE _NI_Pixmap *, struct _NI_Pixmap *)': NEWLINE /usr/home/jip/moz/mozilla/cmd/xfe/icons/mkicons.cpp:246: warning: ANSI C++ NEWLINE forbids implicit conversion from `void *' in initialization NEWLINE ./mkicons -no-xfe-define - < ./iconanim.list > icondata.c.$$ && mv icondata.c.$$ NEWLINE icondata.c NEWLINE make: *** [icondata.c] Error 139 NEWLINE NEWLINE ... NEWLINE NEWLINE ../../../config/./nsinstall -R -m 555 ./libxfe2.so ../../../dist/./bin NEWLINE make[3]: Leaving directory `/fire/tmp/moz/cmd/xfe/src' NEWLINE cd icons; make libs NEWLINE make[3]: Entering directory `/fire/tmp/moz/cmd/xfe/icons' NEWLINE ./mkicons -no-xfe-define - < ./iconanim.list > icondata.c.$$ && mv icondata.c.$$ NEWLINE icondata.c NEWLINE make[3]: *** [icondata.c] Error 139 NEWLINE make[3]: Leaving directory `/fire/tmp/moz/cmd/xfe/icons' NEWLINE make[2]: *** [libs] Error 2 NEWLINE make[2]: Leaving directory `/fire/tmp/moz/cmd/xfe' NEWLINE make[1]: *** [libs] Error 2 NEWLINE make[1]: Leaving directory `/fire/tmp/moz/cmd' NEWLINE make: *** [libs] Error 2 NEWLINE NEWLINE ... NEWLINE NEWLINE succes, NEWLINE jip
1030|Animated GIF that crashes NGLayout win32|Bugreport for NGLayout win32 nighly build 19981008. NEWLINE The GIF image on the page in the URL crashes with the following error displayed: NEWLINE NEWLINE VIEWER orsakade ett ogiltigt sidfel i NEWLINE modul RAPTORGFXWIN.DLL på adress 014f:00470fb8. NEWLINE Register: NEWLINE EAX=0125ef00 CS=014f EIP=00470fb8 EFLGS=00010246 NEWLINE EBX=00abfcbe SS=0157 ESP=00abfb98 EBP=00abfbbc NEWLINE ECX=dddddddd DS=0157 ESI=00008ce2 FS=4aa7 NEWLINE EDX=0125ef00 ES=0157 EDI=00abfc74 GS=769f NEWLINE Byte på CS:EIP: NEWLINE ff 51 10 89 45 f8 83 7d f8 00 75 23 8b 45 e8 3b NEWLINE Stackdump: NEWLINE 0125ef00 00abfba8 30019ab1 00000000 00008ce2 0125ae60 00000000 00000000 0125ef00 NEWLINE 00abfbe8 00471120 0125aec0 0125aa00 0125ef00 00001fcd 10257f50 NEWLINE NEWLINE Sorry for the Swedish, but most is universal. My computer is an AMD K6 with 64MB NEWLINE RAM running in 1024x768 with 24-bits truecolour.
1031|Animation stops|The animation displayed on the page should be rotating infinitly. It does not NEWLINE with the NGLayout win32 19981008 nigly build. NEWLINE NEWLINE Tested on my AMD K6 with 64MB RAM, 1024x768 in 24-bits true colour.
1032|Linux NGLayout webshell viewer coredumps in nsInlineReflow::nsInlineReflow|Any url causes the following core dump: NEWLINE NEWLINE #0 0x40291d44 in nsInlineReflow::nsInlineReflow (this=0xbfffd8a0, NEWLINE aLineLayout=@0x0, aOuterReflowState=@0xbfffdb34, aOuterFrame=0x818c800) NEWLINE at nsInlineReflow.cpp:47 NEWLINE 47 mSpaceManager = aLineLayout.mSpaceManager; NEWLINE #0 0x40291d44 in nsInlineReflow::nsInlineReflow (this=0xbfffd8a0, NEWLINE aLineLayout=@0x0, aOuterReflowState=@0xbfffdb34, aOuterFrame=0x818c800) NEWLINE at nsInlineReflow.cpp:47#1 0x4029074a in nsInlineFrame::Reflow NEWLINE (this=0x818c800, NEWLINE aPresContext=@0x817f900, aMetrics=@0xbfffde10, aReflowState=@0xbfffdc2c, NEWLINE aStatus=@0xbfffdeb8) at nsInlineFrame.cpp:342#2 0x402805ab in NEWLINE nsContainerFrame::ReflowChild (this=0x818ca00, aKidFrame=0x818c800, NEWLINE aPresContext=@0x817f900, aDesiredSize=@0xbfffde10, NEWLINE aReflowState=@0xbfffdc2c, aStatus=@0xbfffdeb8) at NEWLINE nsContainerFrame.cpp:305#3 0x4028cc6c in RootContentFrame::Reflow NEWLINE (this=0x818ca00, NEWLINE aPresContext=@0x817f900, aDesiredSize=@0xbfffde10, NEWLINE aReflowState=@0xbfffdde8, aStatus=@0xbfffdeb8) at nsHTMLFrame.cpp:474 NEWLINE #4 0x402805ab in nsContainerFrame::ReflowChild (this=0x818c780, NEWLINE aKidFrame=0x818ca00, aPresContext=@0x817f900, aDesiredSize=@0xbfffde10, NEWLINE aReflowState=@0xbfffdde8, aStatus=@0xbfffdeb8) at NEWLINE nsContainerFrame.cpp:305#5 0x4028bfff in RootFrame::Reflow (this=0x818c780, NEWLINE aPresContext=@0x817f900, NEWLINE aDesiredSize=@0xbfffde90, aReflowState=@0xbfffde60, aStatus=@0xbfffdeb8) NEWLINE at nsHTMLFrame.cpp:154 NEWLINE #6 0x402986b7 in PresShell::InitialReflow (this=0x8187440, aWidth=11907, NEWLINE aHeight=6491) at nsPresShell.cpp:512 NEWLINE #7 0x401ffa47 in HTMLContentSink::StartLayout (this=0x817f500) at NEWLINE nsHTMLContentSink.cpp:1767 NEWLINE #8 0x401ff0af in HTMLContentSink::OpenBody (this=0x817f500, NEWLINE aNode=@0xbfffe0d0) at nsHTMLContentSink.cpp:1536 NEWLINE #9 0x403a18ea in CNavDTD::OpenBody (this=0x817ff00, aNode=@0xbfffe0d0) at NEWLINE CNavDTD.cpp:2264 NEWLINE #10 0x403a1e96 in CNavDTD::OpenContainer (this=0x817ff00, aNode=@0xbfffe0d0, NEWLINE aUpdateStyleStack=1) at CNavDTD.cpp:2405 NEWLINE #11 0x4039ea5d in CNavDTD::HandleDefaultStartToken (this=0x817ff00, NEWLINE aToken=0x818c480, aChildTag=eHTMLTag_body, aNode=@0xbfffe0d0) NEWLINE at CNavDTD.cpp:609#12 0x4039ed72 in CNavDTD::HandleStartToken NEWLINE (this=0x817ff00, aToken=0x818c480) NEWLINE at CNavDTD.cpp:700#13 0x4039df7d in NavDispatchTokenHandler NEWLINE (aToken=0x818c480, aDTD=0x817ff00) NEWLINE at CNavDTD.cpp:293#14 0x403af1e4 in CTokenHandler::operator() () at NEWLINE nsTokenHandler.cpp:80 NEWLINE #15 0x4039e7ab in CNavDTD::HandleToken (this=0x817ff00, aToken=0x818c480) at NEWLINE CNavDTD.cpp:524 NEWLINE #16 0x403ac5db in nsParser::BuildModel (this=0x8187040) at nsParser.cpp:599#17 NEWLINE 0x403ac468 in nsParser::ResumeParse (this=0x8187040) at nsParser.cpp:563 NEWLINE #18 0x403aca7c in nsParser::OnDataAvailable (this=0x8187040, aURL=0x8174d00, NEWLINE pIStream=0x817d6c0, length=148) at nsParser.cpp:774 NEWLINE #19 0x40634224 in nsDocumentBindInfo::OnDataAvailable (this=0x8174cc0, NEWLINE aURL=0x8174d00, aStream=0x817d6c0, aLength=148) at nsDocLoader.cpp:1331 NEWLINE #20 0x4059e588 in stub_put_block (stream=0x817eda0, NEWLINE buffer=0x8177e00 "<html><head><title>Yahoo!</title><base NEWLINE href=http://www.yahoo.com/></head><body><center><form NEWLINE action=http://search.yahoo.com/bin/search><a href=/bin/", length=148) at NEWLINE nsStubContext.cpp:575#21 0x4057d941 in net_MemCacheWrite (stream=0x818b2a0, NEWLINE buffer=0x8177e00 "<html><head><title>Yahoo!</title><base NEWLINE href=http://www.yahoo.com/></head><body><center><form NEWLINE action=http://search.yahoo.com/bin/search><a href=/bin/", len=148) at NEWLINE mkmemcac.c:664#22 0x40407465 in net_setup_http_stream (ce=0x817b000) at NEWLINE mkhttp.c:2888 NEWLINE #23 0x40408329 in net_ProcessHTTP (ce=0x817b000) at mkhttp.c:3476#24 0x405ba41f NEWLINE in NET_ProcessNet (ready_fd=0x8175980, fd_type=2) NEWLINE at mkgeturl.c:3409#25 0x405c23b9 in net_process_net_timer_callback NEWLINE (closure=0x0) NEWLINE at mkselect.c:189#26 0x400b36c8 in timer_callback (aTimer=0x81731e0, NEWLINE aClosure=0x8173200) NEWLINE at nsImageSystemServices.cpp:70 NEWLINE #27 0x40388b3d in TimerImpl::FireTimeout (this=0x81731e0) at nsTimer.cpp:79 NEWLINE #28 0x40388f8e in nsTimerExpired (aCallData=0x81731e0) at nsTimer.cpp:172 NEWLINE #29 0x408d348b in DoOtherSources () NEWLINE #30 0x408d36c8 in XtAppNextEvent () NEWLINE #31 0x4006b99d in nsAppShell::Run (this=0x8085f60) at nsAppShell.cpp:67 NEWLINE #32 0x805692a in nsNativeViewerApp::Run (this=0x8069680) at nsUnixMain.cpp:70 NEWLINE #33 0x8056b5c in main (argc=2, argv=0xbfffec80) at nsUnixMain.cpp:118 NEWLINE (gdb)
1033|window.open ignores size tags|[from Bryce Nesbitt <bryce@nextbus.COM>] NEWLINE On Netscape 4.06 and 4.5B1 under Unix, but not Windows: NEWLINE NEWLINE <input type=button value="Send Message" NEWLINE name=message NEWLINE onclick="w = window.open( NEWLINE 'http://emery.nextbus.com/emery/private/message.html', NEWLINE 'messageWindow','width=500,height=250'); NEWLINE w.focus()"> NEWLINE NEWLINE The window gets opened to a DIFFERENT size the first time the button is NEWLINE hit. NEWLINE The SECOND time it respects the width= and height= tags.
1034|NetscapeDispatchWnd crashes on URL|Last line in \\bin\\viewer.exe: NEWLINE DocumentViewerImpl::BindToDocument NEWLINE NEWLINE Fatal error: NEWLINE NEWLINE The instruction at "0x00291048" referened memory at "0xdddddded". Memory could NEWLINE not be "read". NEWLINE NEWLINE ‰
1035|Scrollbars: Layout not knowing when scrollbars are present?|On numerious pages, including my own, Mozilla appearenly does not know that NEWLINE scrollbars are present, and does not align the text along the scollbar, NEWLINE instead along the screen edge, which causes about 5-6 pixels worth of text to be NEWLINE lost under the scollbars. After going in to the display control panel and NEWLINE changing the size of the scrollbars, mozilla started laying things out properly NEWLINE durring its current session. If mozilla is restared thought, the toolbars once NEWLINE again block content.
1036|Forward Button Broken|Aargh! The forward button stopped working in the 10-8 build. (Or else I'm NEWLINE hallucinating, which is unlikely.)‰
1037|border-color-left,right,top,bottom broken|The following renders all sides with the same color. It should have the NEWLINE appearance of an inset or outset (with adjacent color boundaries at 45%). NEWLINE NEWLINE <html> NEWLINE <body> NEWLINE <DIV style="width: 30px; height: 30px; border-color: NEWLINE red,black,red,black;"></DIV> NEWLINE <DIV style="width: 30px; height: 30px; border-color-left: red; border-color-top: NEWLINE black; border-color-right: red; border-color-bottom: black;"></DIV> NEWLINE </body> NEWLINE </html>
1038|fantasy and cursive fonts are treated as sans-serif and something else|Fantasy and cursive fonts (as declared in CSS) aren't quite treated as I would NEWLINE expect them to be. Fantasy is treated the same as sans-serif, and cursive is NEWLINE just a serif font that is a little different from Times. (This is what happens NEWLINE in Windows.)‰
1039|Negative CSS font-size should not be allowed|According to the spec's rules for error-handling, negative font sizes should not NEWLINE be allowed. See NEWLINE http://www.w3.org/TR/REC-CSS2/fonts.html#font-size-props‰
1040|CSS vertical-align problems; reflow strange|vertical-align: middle (bug 606, maybe), vertical-align: text-bottom (I'm not NEWLINE sure about this one), and vertical-align: 50% do not work.‰
1041|Percentage margins don't shrink when the browser window gets narrow|The percentage margins in the following pages don't shrink any more when the NEWLINE browser window is narrowed beyond a certain point (about half the width of my NEWLINE screen, I think): NEWLINE http://www.w3.org/Style/CSS/Test/current/sec5501.htm NEWLINE http://www.w3.org/Style/CSS/Test/current/sec5502.htm NEWLINE http://www.w3.org/Style/CSS/Test/current/sec5503.htm NEWLINE http://www.w3.org/Style/CSS/Test/current/sec5504.htm NEWLINE http://www.w3.org/Style/CSS/Test/current/sec5505.htm NEWLINE ‰
1042|CSS auto margins don't justify blocks|The setting of left, right, or both margins to auto should (respectively) right- NEWLINE justify, left-justify, or center a block whose width is less than 100%.‰
1043|DIV should be considered to terminate a P|A DIV should be considered to end a P element that does not have an end tag. NEWLINE See the URL for example and link to specs.‰
1044|inheritance problems in tables|tables don't inherit the properties of a surrounding DIV (and it's clear that NEWLINE the problem is *not* that a table is being (incorrectly) considered to be the NEWLINE end of a DIV). See the Example URL, where the text in the DIV shows up properly NEWLINE on both sides of the table, but incorrectly within the table.‰
1045|CSS background-attachment|It would be nice to support the CSS1 properties background-position and NEWLINE background-attachment. There seems to be some attempt at supporting background- NEWLINE position, since it sometimes seems to cause a background that is tiled only in NEWLINE one direction (using background-repeat) to run twice in that direction NEWLINE (parallel) rather than once. See the URL given for an example. See tests NEWLINE 5.3.4, 5.3.6, and 5.3.7 (link above) for examples of these properties.‰
1046|letter-spacing should apply on space characters too [INLINE]|It would be nice to support the CSS attributes word-spacing and letter-spacing. NEWLINE Although I would have to say I would put them among the lowest priorities in NEWLINE CSS1, I think they are still definitely worth implementing, especially if you NEWLINE want to claim complete CSS1 support.‰
1047|CSS text-align: justify doesn't work|The justify keyword for the text-align property doesn't do anything. It just NEWLINE acts as though it were text-align: left.‰
1048|CSS property text-indent not supported|text-indent is not supported. I think this is one of the more commonly used and NEWLINE better supported properties.‰
1049|{css1} CSS list-style-position not supported|The CSS1 property list-style-position is not supported. (Possible values are NEWLINE inside and outside. See the above URL).‰
1050|Phantom border with CSS border-style set to none|When the CSS property border-style is set to none, the border-width should be NEWLINE displayed as zero. Currently NGLayout is displaying extra padding filling up NEWLINE the area where that border width would be. See not only the above test case but NEWLINE the ffour before it, i.e., 5.5.11-5.5.15.‰
1051|Default CSS border-width=medium border-style=none|The default CSS border-width is medium and the default border-style is none. NEWLINE You currently seem to use (0,solid) (??) rather than (medium,none). This means NEWLINE that if border-style is set to solid, then a medium solid border should be NEWLINE drawn, but if border-width is set to something without setting border-style, NEWLINE then nothing should happen. There are two test cases for this, really (one each NEWLINE way): NEWLINE http://www.w3.org/Style/CSS/Test/current/sec5515.htm NEWLINE http://www.w3.org/Style/CSS/Test/current/sec5516.htm (the above URL)‰
1052|Top borders are a pixel too thin|Top borders seem to be a pixel too thin. This is especially noticeable in the NEWLINE thin border in the above URL.‰
1053|groove and ridge borders are reversed|Based on the light-from-the-upper-left model, which is pretty standard and which NEWLINE you use for inset and outset borders, groove and inset borders got switched.‰
1054|{css1} Dashed and dotted borders have ugly corners|The dashed and dotted borders, when thick, are rather ugly at the corners. I'm NEWLINE not sure what can be done to fix this, but Opera handles it a rather neat way by NEWLINE changing the gaps between the dots and the dashes.‰
1055|{css1} BORDER attribute incorrectly overriding CSS rules|The thin top/bottom borders are thinner inside the table than outside. Appears NEWLINE in NEWLINE top: http://www.w3.org/Style/CSS/Test/current/sec5518.htm (above URL) NEWLINE bottom: http://www.w3.org/Style/CSS/Test/current/sec5520.htm‰
1056|maxAscent/maxDescent not backed up|when a frame on a line ends up being pushed to the next line we don't backout NEWLINE that frames impact on: NEWLINE NEWLINE maxAscent NEWLINE maxDescent NEWLINE max-element-size NEWLINE NEWLINE etc. NEWLINE NEWLINE Start using the PerFrameData in nsInlineReflow and this can be easily fixed: NEWLINE change the code to compute the max values upon completion instead of on the fly.
1057|Percentage margins load incorrectly if window was resized - better on reflow|Percentage margins load incorrectly if window was resized in the past, I think, NEWLINE but they get better on reflow when the window is resized while the page is NEWLINE displayed.‰
1058|Margins not applied|In this page, for some reason, the percentage margins aren't applied in the part NEWLINE outside of the table. They work within the table.‰
1059|Floats within Tables|Floats within Tables are a mess!! Also the reflow is *very* slow on this page.‰
1060|Percentage width floats not resized|The percentage width floats in this page aren't resized when the viewer window NEWLINE size is changed.‰
1061|nested quotes not yet supported (<q> CSS quoting)|[This is just a reminder: peter, reassign this to kipp when you have implemented NEWLINE the css2 content attributes - thanks]
1062|Image misaligned in table (extends past cell boundary)|Nightly build October 10th Win 95 (but noticed in prior builds). NEWLINE NEWLINE A table cell containing two images: layout misaligns the top image so that the NEWLINE top image crosses into the next cell (to right hand side of "home" cell for the NEWLINE image). NEWLINE NEWLINE Occurs on initial load and subsequent reloads, but a reflow (forced by resizing NEWLINE the window) causes the "correct" layout to occur. NEWLINE NEWLINE Problem is dependent on the use of <br> between the images, and on whether the NEWLINE WIDTH and HEIGHT attributes are declared for the top image. NEWLINE NEWLINE Please see test cases at above URL NEWLINE NEWLINE Thanks, NEWLINE John
1063|width of form element not right|the drop-down list box with initial value "10" of left side of page is a big too NEWLINE large, pushing the layout of the whole page off a bit
1064|missing vertical space|there should be a little more vertical space between the image "Lycos acquires NEWLINE Wired Digital" and the text just underneath that image. NEWLINE Otherwise, the page looks perfect!
1065|image alignment|the webtracker image in the center of the page doesn't look like it's aligned NEWLINE correctly. I've copied off the page source below. NEWLINE ================================================================= NEWLINE <html> NEWLINE <META http-equiv="PICS-Label" content='(PICS-1.1 NEWLINE "http://www.rsac.org/ratingsv01.html" l gen true comment "RSACi North America NEWLINE Server" by "wyvern@netins.net" for "http://www.fxweb.holowww.com" on NEWLINE "1997.07.18T07:53-0800" r (n 0 s 0 v 0 l 0))'> NEWLINE <head><title>Welcome to FXWeb</title></head> NEWLINE NEWLINE <body bgcolor="#ffffff"> NEWLINE NEWLINE <CENTER> NEWLINE <TABLE width="620"><TR><TD> NEWLINE <TABLE cellpadding=0 cellspacing=0 border=0> NEWLINE <TR> NEWLINE <TD><IMG src="Assets/spc.gif" height="15" width="30"></TD> NEWLINE <TD><IMG src="Assets/spc.gif" height="15" width="15"></TD> NEWLINE <TD><IMG src="gifs/title.gif" height="50" width="460" vspace="0" NEWLINE hspace="0"></TD> NEWLINE <TD> </TD> NEWLINE <TD bgcolor="#333333" rowspan="20" valign="top" align="middle"> NEWLINE <!--- Left Navbar ---> NEWLINE <IMG src="Assets/spc.gif" height="1" width="100"><BR> NEWLINE <P><B><FONT color="white" face="Helvetica, Arial" NEWLINE size="-2">Our Services:<P> NEWLINE <A href="/tracker"><IMG src="Assets/wt_animated.gif" NEWLINE height="31" width="88" border=0"></A> NEWLINE <BR><BR> NEWLINE <TABLE width="88" align="center"> NEWLINE <TR> NEWLINE <TD><FONT color="white" face="Helvetica, NEWLINE Arial" size="-1"><B> NEWLINE <HR> NEWLINE <A href="/adinfo.shtml">Advertising NEWLINE Information</A><BR> NEWLINE <HR> NEWLINE <A href="/contact.shtml">How To Contact NEWLINE Us</A><BR> NEWLINE <HR> NEWLINE </TD> NEWLINE </TR> NEWLINE </TABLE> NEWLINE <BR><BR> NEWLINE <A href="http://www.rsac.org"><IMG NEWLINE src="Assets/rsacirated.gif" height="31" width="88" border="0"></A> NEWLINE <P>Best Viewed With:<BR> NEWLINE <A href="http://www.netscape.com"><IMG NEWLINE src="gifs/netscape.gif" height="31" width="88" border="0"></A> NEWLINE <!--- Navbar Ends ---> NEWLINE </TD> NEWLINE </TR> NEWLINE <TR> NEWLINE <TD height="15" bgcolor="#333333"> </TD> NEWLINE <TD height="15" valign="bottom" align="left"><IMG NEWLINE src="gifs/round-sw.gif" height="15" width="15"></TD> NEWLINE <TD height="15"><I><FONT size="-5" face="Helvetica, Arial">This NEWLINE site, including all text and artwork, is copyright© 1996, 1997 FXWeb Web NEWLINE Technologies</FONT></I></TD> NEWLINE <TD height="15" valign="bottom" align="right"><IMG NEWLINE src="gifs/round-se.gif" height="15" width="15"></TD> NEWLINE </TR> NEWLINE <TR> NEWLINE <TD height="15" bgcolor="#333333"> </TD> NEWLINE <TD colspan=3 bgcolor="#333333" valign="MIDDLE"> NEWLINE <A href="/"><IMG src="gifs/n-home.gif" border="0" vspace="3" NEWLINE hspace="1"></A> NEWLINE <A href="/tracker"><IMG src="gifs/n-webtracker.gif" border="0" NEWLINE vspace="3" hspace="1"></A> NEWLINE </TD> NEWLINE </TR> NEWLINE <TR> NEWLINE <TD bgcolor="#333333"> </TD> NEWLINE <TD height="15" valign="top" align="left"><IMG NEWLINE src="gifs/round-nw.gif" height="15" width="15"></TD> NEWLINE <TD align="center"> NEWLINE <FONT size="-2">Click below to visit our sponsor</FONT><BR> NEWLINE <!--- Banner Here ---> NEWLINE <A NEWLINE HREF="http://www.eads.com/ads/pageserv.pl?softshar05.cfg,CJ25035"> NEWLINE <IMG NEWLINE SRC="http://www.eads.com/graphics/ads/softshar05.gif?CJ25035" HEIGHT=55 NEWLINE WIDTH=460 ALT="Click Here to Visit Our Sponsor"></A> NEWLINE <!--- Banner Ends ---> NEWLINE </TD> NEWLINE <TD height="15" valign="top" align="right"><IMG NEWLINE src="gifs/round-ne.gif" height="15" width="15"></TD> NEWLINE </TR> NEWLINE <TR> NEWLINE <TD rowspan="10" bgcolor="#333333"> </TD> NEWLINE <TD> </TD> NEWLINE <TD> NEWLINE <FONT face="Helvetica, Arial"> NEWLINE <!--- Main Body Here ---> NEWLINE NEWLINE For almost a year, FXWeb has provided services to the internet NEWLINE free of charge. NEWLINE <P> NEWLINE <IMG src="Assets/wt_animated.gif" align="left" height="31" NEWLINE width="88"> NEWLINE <A href="/tracker">WebTracker</A> offers free statistic tracking NEWLINE for your web NEWLINE site, simply including a few lines of HTML in your page. You may NEWLINE have seen NEWLINE similar web counters before, but you've never seen anything like NEWLINE this. NEWLINE Learn more about WebTracker, or sign up today! NEWLINE <P> NEWLINE <!--- End Main Body ----> NEWLINE </TD> NEWLINE <TD></TD> NEWLINE </TR> NEWLINE </TABLE> NEWLINE </TD></TR></TABLE> NEWLINE </CENTER> NEWLINE NEWLINE NEWLINE </body> NEWLINE NEWLINE </html>
1066|form element incorrect width|the 2 edit controls in the middle of the page are too wide.
1067|layout wrong - wrong handling of badly nested table tags|this used to be a floating image problem, but we think floaters are returning NEWLINE correct info now. so take a look at it
1068|vertical layout wrong|might be a table issue, might not. Take a look.
1069|CRASH on any URL & none of the example seem to do anything|I just checked out and build raptor on my RH 5.1 linux box. NEWLINE NEWLINE (by the way, all that went smoothly, the documentation on mozilla.org is great!) NEWLINE NEWLINE when I tried to run it, none of the examples seem to do anything. NEWLINE NEWLINE The example it is configure to run on startup didn't appear to do anything NEWLINE either. NEWLINE NEWLINE all that seemed to be happening was that where the "throbbing N" is, there was NEWLINE the numbers 0 to 30 (I think) cycling over and over. NEWLINE NEWLINE When ever I try to open http://localhost:80 I crash every time. NEWLINE NEWLINE Here is the stack trace: NEWLINE NEWLINE #0 0x40295b18 in nsSize::nsSize (this=0xbfffe2cc) at nsInlineReflow.cpp:848 NEWLINE #1 0x4029428c in nsInlineFrame::Reflow () at nsInlineFrame.cpp:418 NEWLINE #2 0x40283f2b in nsContainerFrame::ReflowChild () at nsContainerFrame.cpp:228 NEWLINE #3 0x4029066c in RootContentFrame::Reflow () at nsHTMLFrame.cpp:516 NEWLINE #4 0x40283f2b in nsContainerFrame::ReflowChild () at nsContainerFrame.cpp:228 NEWLINE #5 0x4028f9ff in RootFrame::Reflow () at nsHTMLFrame.cpp:516 NEWLINE #6 0x4029cd2f in PresShell::InitialReflow () at nsPresShell.cpp:1203 NEWLINE #7 0x402033a7 in HTMLContentSink::StartLayout () at nsHTMLContentSink.cpp:803 NEWLINE #8 0x40202a0f in HTMLContentSink::OpenBody () at nsHTMLContentSink.cpp:803 NEWLINE #9 0x403a78ea in CNavDTD::OpenBody () at CNavDTD.cpp:1459 NEWLINE #10 0x403a7e96 in CNavDTD::OpenContainer () at CNavDTD.cpp:1459 NEWLINE #11 0x403a4a5d in CNavDTD::HandleDefaultStartToken () at CNavDTD.cpp:1459 NEWLINE #12 0x403a4d72 in CNavDTD::HandleStartToken () at CNavDTD.cpp:1459 NEWLINE #13 0x403a3f7d in NavDispatchTokenHandler () at CNavDTD.cpp:1459 NEWLINE #14 0x403b51e4 in CTokenHandler::operator() () at nsTokenHandler.cpp:83 NEWLINE #15 0x403a47ab in CNavDTD::HandleToken () at CNavDTD.cpp:1459 NEWLINE #16 0x403b25db in nsParser::BuildModel () at nsParser.cpp:624 NEWLINE #17 0x403b2468 in nsParser::ResumeParse () at nsParser.cpp:624 NEWLINE #18 0x403b2a7c in nsParser::OnDataAvailable () at nsParser.cpp:624 NEWLINE #19 0x40638334 in nsDocumentBindInfo::OnDataAvailable () at nsDocLoader.cpp:331 NEWLINE #20 0x405a3688 in stub_put_block () at nsStubContext.cpp:300 NEWLINE #21 0x40581941 in net_MemCacheWrite (stream=0x8360760, NEWLINE buffer=0x81e7800 "<!DOCTYPE HTML PUBLIC \\"-//W3C//DTD HTML 3.2 NEWLINE Final//EN\\">\\n<HTML>\\n <HEAD>\\n <TITLE>Test Page for Red Hat Linux's Apache NEWLINE Installation</TITLE>\\n </HEAD>\\n<!-- Background white, links blue (unvisited), NEWLINE navy ("..., NEWLINE len=1006) at mkmemcac.c:664 NEWLINE #22 0x4040c469 in net_setup_http_stream (ce=0x823d600) at mkhttp.c:2888 NEWLINE #23 0x4040d32d in net_ProcessHTTP (ce=0x823d600) at mkhttp.c:3476 NEWLINE #24 0x405bf41f in NET_ProcessNet () at mkgeturl.c:1912 NEWLINE #25 0x405c72fd in NET_PollSockets () at mkselect.c:125 NEWLINE #26 0x4059aff6 in nsNetlibService::NetPollSocketsCallback () NEWLINE at nsNetService.cpp:507 NEWLINE #27 0x4038eb3d in nsID::Equals (this=0x834e2e0, other=@0x4095f688) NEWLINE at nsTimer.cpp:144 NEWLINE #28 0x4038ef8e in nsTimerExpired () at nsTimer.cpp:144 NEWLINE #29 0x4093e47f in DoOtherSources () NEWLINE #30 0x4093e6bc in XtAppNextEvent () NEWLINE #31 0x4006aae9 in nsAppShell::Run () at nsAppShell.cpp:31 NEWLINE #32 0x8056a12 in nsNativeViewerApp::Run (this=0x806b680) at nsUnixMain.cpp:70 NEWLINE #33 0x8056c41 in main (argc=1, argv=0xbffff6b8) at nsUnixMain.cpp:118
1070|Crashes when viewing pages with too many nested tags.|A page with more than 199 <b>'s inside each other crashes NGlayout nightly build NEWLINE 981008 and generates the following error message: NEWLINE NEWLINE VIEWER orsakade ett ogiltigt sidfel i NEWLINE modul RAPTORHTMLPARS.DLL på adress 014f:007910a2. NEWLINE Register: NEWLINE EAX=00000007 CS=014f EIP=007910a2 EFLGS=00010202 NEWLINE EBX=00abfcbe SS=0157 ESP=00abfa1c EBP=00abfa20 NEWLINE ECX=00000007 DS=0157 ESI=00008ce2 FS=344f NEWLINE EDX=01248c70 ES=0157 EDI=00abfc74 GS=0000 NEWLINE Byte på CS:EIP: NEWLINE 8b 48 08 8b 55 fc 8b 45 08 89 44 8a 0c 8b 4d fc NEWLINE Stackdump: NEWLINE 00000007 00abfa38 00794fe3 00000007 00abfa38 01248c70 00000000 00abfa64 00794886 NEWLINE 00000007 00000007 fffffff8 01248c70 00793b59 00000007 007a8138 NEWLINE NEWLINE The computer is a AMD K6 with 64MB RAM.
1071|Image position incorect when inside <p></p>|Image overlap with text in the following code but should be on the next line: NEWLINE NEWLINE <br><p>The image should follow be on next line (not over this text).<img NEWLINE src="resource:/res/samples/raptor.jpg" align=left><br>This is correctly NEWLINE indented.</p> NEWLINE NEWLINE The same thing happens with headers (<h1>,<h2>,...). NEWLINE When removing the <p> </p>, the image position is correct.
1072|"Negative" characters|Characters over 128 are not displayed correctly. This seems to be because NEWLINE there is a 'signed char' somewhere in the code instead of an 'unsigned char'. NEWLINE NEWLINE Doing a "dump content" on this: éé NEWLINE Shows this: \\u00e9\\uffe9 NEWLINE The second character has been signed extended (but should not).
1073|navigator.preference() not wired up yet|I know we brought over libpref, but it doesn't appear to be exposed to NEWLINE JavaScript. NEWLINE NEWLINE navigator.preference(foo) should return value of pref foo NEWLINE navigator.preference(foo,bar) should set pref foo to value bar NEWLINE NEWLINE Obviously when we get security, we'll need to control access to these methods, NEWLINE as we do in Navigator. NEWLINE NEWLINE These methods are useful to me in building an XPFE prefs interface.
1074|CRASH: when I do "View \| There Ain't No Toolbar, Dude" and then mouse over the bookmarks in the personal toolbar|when I expand the bookmarks bar and the navigation bar (the only two bars that NEWLINE are visible in Mozilla right now) and then I go to "View \| There Ain't No NEWLINE Toolbar, Dude", and then I mouse over the links in the bookmarks bar, I still NEWLINE get this crash: NEWLINE NEWLINE #0 0x402d8981 in __kill () NEWLINE #1 0x402a7795 in raise (sig=142034949) NEWLINE #2 0x82d273c in StageTwoTimeout (closure=0x8b34a00, id=0xbfffdf60) NEWLINE at ToolTip.c:964 NEWLINE #3 0x4012034e in XtAppProcessEvent () NEWLINE #4 0x819423a in fe_EventLoop () at mozilla.c:1133 NEWLINE #5 0x819745d in main (argc=1, argv=0xbffff730) at mozilla.c:3421 NEWLINE #6 0x8141a8e in _start ()
1075|Resizing frameset window - frames shift|To reproduce, load this page (or any similar frameset), and reize the window. NEWLINE For some reason, as you're resizing, the individual frames grow/shrink/etc. NEWLINE rather than retaining nothing more/less than their originally specified NEWLINE width/height. Odd...
1076|[NATIVE]CSS borders don't work on input elements|Test case: NEWLINE <html> NEWLINE <body> NEWLINE <form> NEWLINE <input type="text" NEWLINE style="width: 200px; border: thin solid red; NEWLINE font-family: arial; NEWLINE color:blue; NEWLINE font-weight: bold;"> NEWLINE </form> NEWLINE </body> NEWLINE </html> NEWLINE NEWLINE In IE 4/5, instead of the traditional 3d border around the text field, the NEWLINE border becomes a thin solid red line. CSS spec says that supporting this is NEWLINE optional on elements that aren't "normal," such as forms.
1077|{inc} incremental reflow bug|The page jumps when you resize after loading
1078|Purpose of close button on dialog boxes|On "dialog" boxes it displays a close button on the upper right corner which NEWLINE does basically nothing. Can it be assigned as the Cancel command or be removed NEWLINE from the window that pops up?
1079|Window does not refresh properly when resized|If you are using a slow network connection i.e. 28.8 modem and a slow computer NEWLINE (i.e. P166 running NT), and maximize the window while the page is being loaded NEWLINE the window it will not render things properly.
1080|{vmargin} top/bottom block margins and floaters don't mix well|When a block is going to carry-out the top/bottom margins the floaters in it NEWLINE cannot be positioned properly becaues the top/bottom margin value is applied NEWLINE sometime later by some outer block frame. NEWLINE NEWLINE When this happens the floater needs to be repositioned.
1081|css9.4.2 bug|We apply the left/right margins regardless of how an inline frame is split. NEWLINE NEWLINE see j:/html/css9.4.2.html
1082|Mozilla loses default background color after viewing page with javascript|I normally have my background color set to white and my foreground set to black. NEWLINE After I view any page which contains Javascript code, on any page which doesn't NEWLINE specify a background, Mozilla forgets my default background color, and uses the NEWLINE standard gray backgound instead. NEWLINE NEWLINE Two URLs which show the problem (view in order): NEWLINE <http://www.egroups.com/list/freeqt/> NEWLINE <http://www.cygnus.com/ml/egcs/1998-Oct/> NEWLINE NEWLINE Selecting "View Source" on the mis-displayed page, closing the source window, NEWLINE and reloading the mis-displayed page makes it (and all subsequently visited NEWLINE pages) display correctly. NEWLINE NEWLINE This bug has been present in Navigator/Communicator since at least version 3. NEWLINE This bug report is relative to the Mozilla source as pulled down from CVS on NEWLINE 1998-10-09, built as MOZ_MEDIUM. NEWLINE NEWLINE This is Mozilla for SunOS 5.6, displaying remotely on an 8-bit XFree86 3.3.2 S3 NEWLINE server running on FreeBSD 2.2.7. (I can test other display setups as well. It NEWLINE *might* only be a problem with 8-bit displays.) NEWLINE NEWLINE In Navigator/Communicator, Java applets cause the same problem, but obviously NEWLINE this can't happen in Mozilla currently.
1083|password assignment mail had bad URL|When I signed up for my Bugzilla password, the mail I received contained NEWLINE the following text: NEWLINE NEWLINE To change your password, go to: NEWLINE [Param urlbase]changepassword.cgi NEWLINE NEWLINE Presumably "[Param urlbase]" was meant to have the actual urlbase substituted on NEWLINE the server side.
1084|Open 2nd browser window -> assert in ToolTip.c|I hit an assert in ToolTip.c, I think I opened a 2nd browser NEWLINE window. NEWLINE NEWLINE > ToolTip.c: NEWLINE > NEWLINE > if (_XfeIsAlive(_xfe_tt_stage_one_target)) NEWLINE > { NEWLINE > >> assert( w == _xfe_tt_stage_one_target); NEWLINE > NEWLINE > /* Reset the stage one target */ NEWLINE > _xfe_tt_stage_one_target = NULL; NEWLINE > NEWLINE > /* Set the stage two target */ NEWLINE > _xfe_tt_stage_two_target = w; NEWLINE > NEWLINE > ItemPostToolTip(w); NEWLINE > } NEWLINE >
1085|radio button does not trigger onMouseDown event|This bug was found while testing an application that uses the mouse position NEWLINE when selecting a radio button for positioning of a layer. NEWLINE NEWLINE The onMouseDown event was triggered in all 4.0 browsers tested; Mac, 95, 98, and NEWLINE NT. It was not triggered on the Unix version. NEWLINE NEWLINE A page was created for the demonstration of the bug at NEWLINE http://www.teale.ca.gov/bugs/radioMouse.html This page will show the mouses NEWLINE position when the onMouseDown event is triggered in the status bar. When the NEWLINE radio button is clicked... there is no event triggered.
1086|Swing Classes repaint problem with JTable & JTree|When clicking into a JTable or JTree, or when trying to scroll it, there is no NEWLINE repaint. After moving the Frame/Applet/Dialog outside of the visual part of the NEWLINE screen, the JTree/JTable is repainted. NEWLINE NEWLINE How to reproduce : NEWLINE Create a simple Applet/Frame or Dialog using the Swing-Classes V1.0.3. Place a NEWLINE simple JTable/JTree into it. Then start the Applet. Switch back to the Navigator NEWLINE Window and go to a different web-page. Then put the focus back into the applet NEWLINE and try to scroll/select the JTree/JTable. No repaint occurs anymore.
1087|Displays funny|Steve thinks the problem is because the page has floating tables
1088|Necko: Netlib performance ->URLImpl Locking Overhead|Currently URLImpl uses NSPR cached monitors in order to be thread safe. The NEWLINE problem is NSPR cached monitors have a very large performance overhead. NEWLINE NEWLINE Kipp pointed out that NSPR 2.0 has simple locks, see PR_NewLock() which on NEWLINE Windows are just a critical section. NEWLINE NEWLINE We should have one NSPR lock for all of the URLImpl objects which makes the NEWLINE resource overhead low
1089|Performance of style= attribute|Peter, Kipp nominated you for fixing this bug. NEWLINE NEWLINE The basic problem is that the overhead of using a "style" attribute is very NEWLINE high, because the nsGenericHTMLElement code creates a new CSS parser instance NEWLINE for each STYLE attribute. NEWLINE NEWLINE Each CSSParserImpl object as part of its ParseDeclarations() function creates NEWLINE a new nsString, Unichar input stream, and a CSS scanner which itself has NEWLINE additional overhead
1090|Scrollbar doesn't display|When clicking on a link that contains a fragment identifier (URL that ends NEWLINE with a '#' followed by an anchor identifier) the scrollbar doesn't display. NEWLINE NEWLINE Go to the URL above and once we scroll to the correct position in the NEWLINE document notice there's no scrollbar displayed. It's there and you can click NEWLINE on the line up button. Resizing the window also causes it to appear
1091|css overflow property and event handling, rendering, clipping|and cursor handling too NEWLINE NEWLINE we need to determine how to best handle children that are outside their parent NEWLINE for all othe above systems.
1092|window resize causes MARGINWIDTH & HEIGHT settings to be lost|I see that there was an added parameter in the body tag called MARGINWIDTH & NEWLINE HEIGHT. I don't particularly approve of adding a new parameter myself, but its NEWLINE going to be there for a while I guess. NEWLINE NEWLINE Anyway, in the URL I gave above, it has the MARGINWIDTH & HEIGHT set to 0 NEWLINE everything's fine and dandy until I maximize the window. When I do that the NEWLINE MARGINWIDTH & HEIGHT info seems to have disappeared. NEWLINE NEWLINE My suggested fix for that problem is to eliminate MARGINWIDTH & HEIGHT support NEWLINE actually and support the CSS-1 of setting BODY { padding: 0 }
1093|URLBar: needs watch-cursor status|entering an URL puts up the watch cursor, but only NEWLINE over the HTML area. cursor should "go watch" NEWLINE over the rest of the chrome too.
1094|FORM behaves like inline element|See the HTML below for an example. NEWLINE NEWLINE The problem is that FORM is behaving like an inline element rather than NEWLINE a block-level element like the HTML 4 specs indicates (both IE and NEWLINE Navigator handle this correctly). NEWLINE NEWLINE This may be a problem in the form code, I don't know. NEWLINE NEWLINE <body> NEWLINE <div style="border:2px blue solid"> NEWLINE This test is to see whether the following form is treated as block-level. NEWLINE <form> NEWLINE <input type="button" value="button"> NEWLINE </form> NEWLINE </div> NEWLINE </body>
1095|colspan with width distorts tables|doesn't display right in raptor because the iframes are used because we do them NEWLINE and they don't layout the same way that IE does them.
1096|Bookmarks don't remember # of column setting|What happened to the bookmarks? A while back they used to be OK. Now, there's NEWLINE no nice "Close" button in the corner like before, and it defaults to showing ALL NEWLINE COLUMNS of info for the bookmarks, which, needless to say, makes everything a NEWLINE big mess. So I move all but 1 column off the display, and close and reopen the NEWLINE bookmark window and it's gone back to display ALL the columns again! ??? This NEWLINE happens in win95 and NT, and in the past 2 optimized win32 builds from NEWLINE Mozillazine.
1097|crypto.signText() and Win16 problem.|I am testing signing function of Netscape Navigator - crypto.signText(). NEWLINE My application has to run on Windows 3.1 ( PC, Pentium 166 MMX, 64 MB RAM ) and NEWLINE I have met some serious problems. When I try to open this testing page NEWLINE NEWLINE <html> NEWLINE <head> NEWLINE <script> NEWLINE var foo = crypto.signText("Bill of Sale\\n--------------------", "ask"); NEWLINE </script> NEWLINE </head> NEWLINE <body> This is an HTML page<p> NEWLINE <script> document.write(foo); NEWLINE </script> NEWLINE </body> NEWLINE </html> NEWLINE NEWLINE browser reaction is following: NEWLINE NEWLINE - Netscape Navigator 4.04. ( WinNT ) .... everything is O.K. NEWLINE - Netscape Navigator 4.04. ( Win 3.1) .... browser crashes with GPF NEWLINE - Netscape Navigator 4.07. ( Win 3.1) .... appears EMPTY window, NEWLINE that refuses to be closed and that is all.
1098|window.open crashes pretty bad.|whenever i use window.open NGLayout receives an 'Application Error' and NEWLINE crashes.. NEWLINE NEWLINE example below: NEWLINE NEWLINE <HTML> NEWLINE <HEAD> NEWLINE <TITLE>just a test</TITLE> NEWLINE </HEAD> NEWLINE <BODY> NEWLINE <H1>this thing just opens another window and stuph</H2> NEWLINE </BODY> NEWLINE <SCRIPT> NEWLINE a = window.open("","laalaa"); NEWLINE a.document.open(); NEWLINE a.d = a.document; NEWLINE a.d.w = a.d.write; NEWLINE a.d.w("<HTML><HEAD><TITLE>laa</TITLE></HEAD><BODY><P> laa laa laa NEWLINE </P></BODY></HTML>"); NEWLINE a.d.close(); NEWLINE </SCRIPT> NEWLINE </HTML>
1099|negative left margins are broken|the vertical alignment code is to blame...
1100|Express mode should state tree open/closed state|I have to wait forever for Tinderbox to load over my modem. NEWLINE NEWLINE There are really only two relevant pieces of information I go to tinderbox for NEWLINE 95% of the time. NEWLINE NEWLINE 1) wether or not the tree is open NEWLINE 2) wether or not any of the trees are in flames (which is really wether or not NEWLINE the tree is open when it says open). NEWLINE NEWLINE NEWLINE These two pieces of information can be made known much sooner (as in a minute or NEWLINE more) if you seperated them from the massive table, and just had that NEWLINE information in a much smaller table right above and flush with the massive NEWLINE table. NEWLINE NEWLINE thx
1101|Starting up new profile mess-up|Using the latest version from CVS at the time logged, tried to create a new NEWLINE profile to use for testing things from "clean slate". Mozilla starts up with a NEWLINE small (100x100?) window reminiscent of the Java/script pop-up ads in the upper NEWLINE lefthand corner, menubar and main buttons only up before: NEWLINE NEWLINE _free_dbg_lk(void * 0x00acfb00, int 1) line 1033 + 60 bytes NEWLINE _free_dbg(void * 0x00acfb00, int 1) line 970 + 13 bytes NEWLINE operator delete(void * 0x00acfb00) line 282 + 12 bytes NEWLINE CRDFToolbar::LayoutButtons(int -1) line 1960 + 21 bytes NEWLINE CRDFToolbar::HandleEvent(_HT_NotificationStruct * 0x00a99700, _HT_ResourceStruct NEWLINE * 0x00acf610, unsigned long 1) line 1452 NEWLINE toolbarNotifyProcedure(_HT_NotificationStruct * 0x00a99700, _HT_ResourceStruct * NEWLINE 0x00acf610, unsigned long 1, void * 0x00000000, unsigned long 0) line 1397 NEWLINE sendNotification(_HT_ResourceStruct * 0x00acf610, unsigned long 1) line 2083 + NEWLINE 21 bytes NEWLINE refreshItemList(_HT_ResourceStruct * 0x00acf610, unsigned long 1) line 1000 + 13 NEWLINE bytes NEWLINE addContainerItem(_HT_ResourceStruct * 0x00acc120, RDF_ResourceStruct * NEWLINE 0x00acf480) line 2646 + 11 bytes NEWLINE htrdfNotifFunc(RDF_EventStruct * 0x00a9a330, void * 0x00a99730) line 506 + 15 NEWLINE bytes NEWLINE assertNotify(RDF_DBStruct * 0x00a99f20, RDF_NotificationStruct * 0x00a9a2f0, NEWLINE RDF_ResourceStruct * 0x00acf480, RDF_ResourceStruct * 0x00a72580, void * NEWLINE 0x00acbd70, unsigned short 1, int 1, char * 0x00a86700) line 1086 + 17 bytes NEWLINE sendNotifications(RDF_DBStruct * 0x00a99f20, unsigned long 1, RDF_ResourceStruct NEWLINE * 0x00acf480, RDF_ResourceStruct * 0x00a72580, void * 0x00acbd70, unsigned short NEWLINE 1, int 1, char * 0x00a86700) line 1187 + 38 bytes NEWLINE sendNotifications2(RDF_TranslatorStruct * 0x00a86690, unsigned long 1, NEWLINE RDF_ResourceStruct * 0x00acf480, RDF_ResourceStruct * 0x00a72580, void * NEWLINE 0x00acbd70, unsigned short 1, int 1) line 1169 + 43 bytes NEWLINE remoteStoreAdd(RDF_TranslatorStruct * 0x00a86690, RDF_ResourceStruct * NEWLINE 0x00acf480, RDF_ResourceStruct * 0x00a72580, void * 0x00acbd70, unsigned short NEWLINE 1, int 1) line 212 + 32 bytes NEWLINE remoteAssert3(RDF_FileStruct * 0x00a86820, RDF_TranslatorStruct * 0x00a86690, NEWLINE RDF_ResourceStruct * 0x00acf480, RDF_ResourceStruct * 0x00a72580, void * NEWLINE 0x00acbd70, unsigned short 1, int 1) line 111 + 30 bytes NEWLINE addSlotValue(RDF_FileStruct * 0x00a86820, RDF_ResourceStruct * 0x00acf480, NEWLINE RDF_ResourceStruct * 0x00a72580, void * 0x00acbd70, unsigned short 1, char * NEWLINE 0x00000000) line 560 + 44 bytes NEWLINE parseNextRDFToken(RDF_FileStruct * 0x00a86820, char * 0x01bf5fe8) line 376 + 63 NEWLINE bytes NEWLINE parseNextRDFXMLBlobInt(RDF_FileStruct * 0x00a86820, char * 0x01be2e58, long NEWLINE 1200) line 127 + 22 bytes NEWLINE parseNextRDFXMLBlob(_NET_StreamClass * 0x00acb8a0, char * 0x01be2e58, long 1200) NEWLINE line 145 + 17 bytes NEWLINE net_CacheWrite(_NET_StreamClass * 0x00acbcb0, char * 0x01be2e58, long 1200) line NEWLINE 1387 + 24 bytes NEWLINE net_pull_http_data(_ActiveEntry * 0x00a87810) line 3015 + 30 bytes NEWLINE net_ProcessHTTP(_ActiveEntry * 0x00a87810) line 3400 + 9 bytes NEWLINE NET_ProcessNet(PRFileDesc * 0x00a86440, int 2) line 3334 + 13 bytes NEWLINE net_process_slow_net_timer_callback(void * 0x00000000) line 216 + 9 bytes NEWLINE wfe_ProcessTimeouts(unsigned long 12203447) line 303 + 12 bytes NEWLINE FireTimeout(HWND__ * 0x0007054c, unsigned int 275, unsigned int 777, unsigned NEWLINE long 12203447) line 60 + 9 bytes
1102|bad box layout|see the test...
1103|Patch to use DBI module instead of Mysql.pm|use Mysql was failing for me because perl failed to link the library (multiple NEWLINE versions of Mysql.pm/DBI/DBD/etc I'm sure). But since we use DBI (1.02 with NEWLINE msql/mysql 2.0205) I redid globals.pl using DBI instead. And thanks for NEWLINE centralizing the DB routines, made this very easy :-) NEWLINE NEWLINE You can find the patch at NEWLINE http://www.stocknet.ml.org/~sstock/bugzilla/globals.pl.diff
1104|bugzilla uses non-standard sql|bugzilla cannot be easily port to non-mysql databases such as postgress or NEWLINE oracle. while things like bitint and tinytext are trivial to port; enum, NEWLINE timestamp, and datetime may not be easy. NEWLINE NEWLINE is there anyway to start moving the perl and sql parts towards sql92?
1105|new_comment.cgi allows blank lines and extraineous spaces|new_comment allows blank entries into the comment file. to forces something from NEWLINE the user, try this patch: (it also strings ^$ spaces). NEWLINE NEWLINE --- new_comment.cgi.ORIG Thu Oct 15 16:52:27 1998 NEWLINE +++ new_comment.cgi Thu Oct 15 18:01:45 1998 NEWLINE @@ -28,11 +28,20 @@ NEWLINE ($name, $value) = split(/=/, $pair); NEWLINE NEWLINE $value =~ tr/+/ /; NEWLINE + $value =~ s/^(\\s*)//s; NEWLINE + $value =~ s/(\\s*)$//s; NEWLINE $value =~ s/%([a-fA-F0-9][a-fA-F0-9])/pack("C", hex($1))/eg; NEWLINE $FORM{$name} = $value; NEWLINE } NEWLINE -open(COMMENTS, ">>data/comments"); NEWLINE $c=$FORM{"comment"}; NEWLINE +if ( (!defined $c) \|\| ($c eq '') ) { NEWLINE + print "Content-type: text/html\\n\\n"; NEWLINE + print "<TITLE>Nothing on your mind?</TITLE>"; NEWLINE + print "<H1>Does you mind draw a blank?</H1>"; NEWLINE + exit 0; NEWLINE +} NEWLINE + NEWLINE +open(COMMENTS, ">>data/comments"); NEWLINE print COMMENTS $FORM{"comment"} . "\\n"; NEWLINE close(COMMENTS); NEWLINE print "Content-type: text/html\\n\\n";
1106|clicking back button will dump core when start up page is about:blank|In XFE_BackForwardMenu::fillSubmenu (BackForwardMenu.cpp:210) NEWLINE there should be a check to see if current_node is NULL or not.
1107|Once text-decoration is set to blink in CSSOM, it's impossible to turn off|Once you set text-decoration to blink, it can't be set back.
1108|CSS OM changes to line-height do not work|I can't modify the line-height property via the CSS OM.
1109|Cursor doesn't change to hand when over a link|So special case here, it just never seems to change to a hand...
1110|Table layout looks funky|Sorry I haven't had time to dumb this case down yet, but I will try taking a NEWLINE look into it tonight. Visiting the URL in a 10/15/98 7:50pm optimized build NEWLINE yields less than ideal results.
1111|Resizing a page with floats causes weird painting glitch|Load this page, scroll down, size it wide enough to fit everything, and then NEWLINE start resizing horizontally. The big float on the right repaints itself in a NEWLINE jagged fashion - very difficult to explain using words. :-) see for yourself. NEWLINE NEWLINE btw, it is VERY COOL that we lay this page out properly!!!
1112|Can't submit a form|I fill in the search field and hit enter (or click the submit button). I notice NEWLINE my status bar changes with a new URL, but nothing happens - no search results NEWLINE are loaded...
1113|Scrollbar doesn't appear properly unless you resize window|On this page, when you initially load the document, the scrollbar will not NEWLINE appear properly unless you resize the window. If you reload the document after NEWLINE the resize, the same problem occurs. NEWLINE NEWLINE By "not appear properly" I mean that the scroll down button part of the scroll NEWLINE bar ("v") doesn't appear, and the area is painted either gray or white. NEWLINE NEWLINE I see this on release build 10/15/98/7:00pm
1114|ss:we need to stop eating the stuff inside of IFRAME's|they need to go into the content model too, just like for other things...
1115|JavaScript build error|When building on an ARM Linux box (i.e. NetWinder) the compilation of NEWLINE some JavaScript modules breaks because an error in js/src/jscpucfg.h NEWLINE (line 360: #error "linux cpu architecture not supported by jscpucfg.h") NEWLINE NEWLINE The following lines fix the problem: NEWLINE NEWLINE Index: jscpucfg.h NEWLINE =================================================================== NEWLINE RCS file: /cvsroot/mozilla/js/src/jscpucfg.h,v NEWLINE retrieving revision 3.2 NEWLINE diff -r3.2 jscpucfg.h NEWLINE 355a356,358 NEWLINE > #elif __arm__ NEWLINE > #define IS_LITTLE_ENDIAN 1 NEWLINE > #undef IS_BIG_ENDIAN NEWLINE NEWLINE Also, the code could be made more independent from the processor by NEWLINE using the definitions in /usr/include/endian.h - Something like the NEWLINE following should do it: NEWLINE NEWLINE #include <endian.h> NEWLINE #if BYTE_ORDER == LITTLE_ENDIAN NEWLINE #define IS_LITTLE_ENDIAN 1 NEWLINE #undef IS_BIG_ENDIAN NEWLINE #elif BYTE_ORDER == BIG_ENDIAN NEWLINE #define IS_BIG_ENDIAN 1 NEWLINE #undef IS_LITTLE_ENDIAN NEWLINE #else NEWLINE #error "linux cpu architecture not supported by jscpucfg.h" NEWLINE #endif
1116|Hanging while trying to display Cookie information|The backtrace is as follows: NEWLINE NEWLINE #0 strlen () at strlen.s:84 NEWLINE #1 0x980510 in NET_SACat () at xp_str.c:353 NEWLINE #2 0x71deec in net_DisplayCookieInfoAsHTML () at mkaccess.c:4930 NEWLINE #3 0x71e620 in NET_DisplayCookieInfoAsHTML () at mkaccess.c:5297 NEWLINE #4 0x5bd15c in XFE_Frame::doCommand () at Frame.cpp:3523 NEWLINE #5 0x5baa80 in xfe_ExecuteCommand () at Frame.cpp:3198 NEWLINE #6 0x5c9588 in pushb_activate_cb__8XFE_MenuSFP10_WidgetRecPvT2 () NEWLINE at Menu.cpp:870 NEWLINE #7 0xf6587b8 in XtCallCallbackList () at Callback.c:755 NEWLINE #8 0xe2c71b8 in BtnUp () at PushB.c:1346 NEWLINE #9 0xf671294 in HandleActions () at TMstate.c:2321 NEWLINE #10 0xf673060 in HandleSimpleState () at TMstate.c:2321 NEWLINE #11 0xf672afc in _XtTranslateEvent () at TMstate.c:2321 NEWLINE #12 0xf65f4ec in XtDispatchEventToWidget () at Event.c:1917 NEWLINE #13 0xf65e148 in _XtDefaultDispatcher () at Event.c:1917 NEWLINE #14 0xf65ea70 in XtDispatchEvent () at Event.c:1917 NEWLINE #15 0x56adec in fe_EventLoop () at mozilla.c:1047 NEWLINE #16 0x56e370 in main () at mozilla.c:3421 NEWLINE #17 0x51bd78 in __start () at crt1text.s:165 NEWLINE NEWLINE Depending on when the interuption took place, the line is strlen is different, NEWLINE but not the function stack. NEWLINE NEWLINE How to repeat: Start Mozilla, go to Edit Toolbars, then to Window/Privacy NEWLINE Tools/Cookies and it freezes.
1117|Hanging while trying to display Cookie information|The backtrace is as follows: NEWLINE NEWLINE #0 strlen () at strlen.s:84 NEWLINE #1 0x980510 in NET_SACat () at xp_str.c:353 NEWLINE #2 0x71deec in net_DisplayCookieInfoAsHTML () at mkaccess.c:4930 NEWLINE #3 0x71e620 in NET_DisplayCookieInfoAsHTML () at mkaccess.c:5297 NEWLINE #4 0x5bd15c in XFE_Frame::doCommand () at Frame.cpp:3523 NEWLINE #5 0x5baa80 in xfe_ExecuteCommand () at Frame.cpp:3198 NEWLINE #6 0x5c9588 in pushb_activate_cb__8XFE_MenuSFP10_WidgetRecPvT2 () NEWLINE at Menu.cpp:870 NEWLINE #7 0xf6587b8 in XtCallCallbackList () at Callback.c:755 NEWLINE #8 0xe2c71b8 in BtnUp () at PushB.c:1346 NEWLINE #9 0xf671294 in HandleActions () at TMstate.c:2321 NEWLINE #10 0xf673060 in HandleSimpleState () at TMstate.c:2321 NEWLINE #11 0xf672afc in _XtTranslateEvent () at TMstate.c:2321 NEWLINE #12 0xf65f4ec in XtDispatchEventToWidget () at Event.c:1917 NEWLINE #13 0xf65e148 in _XtDefaultDispatcher () at Event.c:1917 NEWLINE #14 0xf65ea70 in XtDispatchEvent () at Event.c:1917 NEWLINE #15 0x56adec in fe_EventLoop () at mozilla.c:1047 NEWLINE #16 0x56e370 in main () at mozilla.c:3421 NEWLINE #17 0x51bd78 in __start () at crt1text.s:165 NEWLINE NEWLINE Depending on when the interuption took place, the line is strlen is different, NEWLINE but not the function stack. NEWLINE NEWLINE How to repeat: Start Mozilla, go to Edit Toolbars, then to Window/Privacy NEWLINE Tools/Cookies and it freezes. NEWLINE NEWLINE Among the things I get in stderr (I think they are independant) NEWLINE Warning: Cannot convert string "Netscape" to type Widget NEWLINE Warning: Actions not found: BtnUp, BtnDown NEWLINE XFE_ToolbarItem::setBaseWidget(urlBar)
1118|Using content-type JS versioning is "sticky"|Reported by jay@kesoftware.com NEWLINE NEWLINE If you have the server return a content-type of application/x-javascript NEWLINE with a version parameter on it, the first time through, it works okay, NEWLINE but the second time through, trying to re-upgrade to the next version NEWLINE does not work (it still uses the old version instead.) NEWLINE NEWLINE A perl script is attached to demonstrate the behavior: NEWLINE NEWLINE #!/usr/local/bin/perl NEWLINE $test="Versioned Mime"; NEWLINE #$test="tag"; NEWLINE $version="1.3"; NEWLINE NEWLINE # this tests to see if a versioned mime type changes NEWLINE # the behaviour of a browser's javascript implementation NEWLINE NEWLINE if ($test eq "tag") NEWLINE { NEWLINE print "content-type: text/html\\n\\n <script NEWLINE language=Javascript$version>\\n\\n"; NEWLINE } NEWLINE else NEWLINE { NEWLINE print "content-type: application/x-javascript; version=$version\\n\\n"; NEWLINE } NEWLINE print <<EOF; NEWLINE NEWLINE alert ("Testing javascript actions: Attempting javascript $version code using NEWLINE $test"); NEWLINE NEWLINE NEWLINE if ("1" == 1 ) NEWLINE { NEWLINE alert("== is true; Javascript 1.3+ behaviour"); NEWLINE } NEWLINE else NEWLINE { NEWLINE alert("== is false, Javascript 1.2 behaviour"); NEWLINE } NEWLINE NEWLINE // this is true for javascript 1.0, 1.1 and 1.3+, false for 1.2 NEWLINE NEWLINE NEWLINE EOF NEWLINE if ($test eq "tag" ) { NEWLINE print "</script>\\n\\n"; NEWLINE } NEWLINE exit
1119|List items not indented|The list items in the URL above aren't indented. The problem started sometime NEWLINE between the 10-8 build and the 10-15 build. It may have to do with the CSS in NEWLINE the linked stylesheet, which thoroughly goes through anything that could indent NEWLINE a list, and sets all but one of these properties (margin-left on UL) to 0.‰
1120|top-margin of first element within a float is ignored|The top margin of the H2 in the left-floating first column in the above URL is NEWLINE ignored. The two columns should have even tops.‰
1121|build fails|../../../dist/public/js/jscpucfg.h:357: #error "linux cpu architecture not NEWLINE supported by jscpucfg.h" NEWLINE NEWLINE I have a patch for this posted on netscape.public.mozilla.patches. Please add NEWLINE this as soon as possible. If you need another copy of the patch, please feel NEWLINE free to mail me. NEWLINE NEWLINE I have a machine in the Mozilla-External section of Tinderbox, FYI. NEWLINE NEWLINE If anyone is interested in developing Mozilla on Linux Alpha, I can probably NEWLINE help, mail me.
1122|body margins increased maximize and restore|When I load my home page (you're getting pretty familiar with it now...) into NEWLINE NGLayout Viewer when the viewer is not maximized, maximize the window, and then NEWLINE restore it, the BODY-left-margin is increased and there is a stripe of the page NEWLINE from when it was maximized left over in the extra margin area. NEWLINE ‰
1123|8bit characters in TITLE tag, UTF8|Also see 835, the fixed META tag bug and 929 (duplicate 764, too). This problem NEWLINE falls in the same category. It drives me nuts, though, because today I don't NEWLINE seem to be able to reproduce it, and unfortunately I didn't log the stack NEWLINE backtrace. It was the usual IsUDF8String() assert (764) in trying to add the NEWLINE TITLE tag with 8 bit characters to the history entries. However, instead of NEWLINE failing assert when viewing that page, it now fails when browsing history (929) NEWLINE in what seems to be pretty much same place: NEWLINE NEWLINE XP_AssertAtLine(char * 0x008cff6c, int 0x00000051) line 2687 NEWLINE makeNewAssertion(RDF_TranslatorStruct * 0x0098ca70, RDF_ResourceStruct * NEWLINE 0x00aac8f0, RDF_ResourceStruct * 0x009919a0, void * 0x00aacca0, unsigned short NEWLINE 0x0003, int 0x00000001) line 81 + 72 bytes NEWLINE remoteStoreAdd(RDF_TranslatorStruct * 0x0098ca70, RDF_ResourceStruct * NEWLINE 0x00aac8f0, RDF_ResourceStruct * 0x009919a0, void * 0x00aacca0, unsigned short NEWLINE 0x0003, int 0x00000001) line 194 + 30 bytes NEWLINE collateOneHist(RDF_TranslatorStruct * 0x009c88a0, RDF_ResourceStruct * NEWLINE 0x009ac760, char * 0x01c6b6cb, char * 0x01c6b6b7, long 0x3628a540, long NEWLINE 0x3628a540, unsigned long 0x00000001, int 0x00000000) line 124 + 84 bytes NEWLINE collateHistory(RDF_TranslatorStruct * 0x009c88a0, RDF_ResourceStruct * NEWLINE 0x009ac760, int 0x00000000) line 99 + 37 bytes NEWLINE HistPossiblyAccessFile(RDF_TranslatorStruct * 0x009c88a0, RDF_ResourceStruct * NEWLINE 0x009ade10, RDF_ResourceStruct * 0x00990b40, int 0x00000001) line 618 + 21 bytes NEWLINE possiblyAccessFile(RDF_TranslatorStruct * 0x009c88a0, RDF_ResourceStruct * NEWLINE 0x009ade10, RDF_ResourceStruct * 0x00990b40, int 0x00000001) line 282 + 22 bytes NEWLINE remoteStoreGetSlotValuesInt(RDF_TranslatorStruct * 0x009c88a0, NEWLINE RDF_ResourceStruct * 0x009ade10, RDF_ResourceStruct * 0x00990b40, unsigned short NEWLINE 0x0001, int 0x00000001, int 0x00000001) line 358 + 21 bytes NEWLINE remoteStoreGetSlotValues(RDF_TranslatorStruct * 0x009c88a0, RDF_ResourceStruct * NEWLINE 0x009ade10, RDF_ResourceStruct * 0x00990b40, unsigned short 0x0001, int NEWLINE 0x00000001, int 0x00000001) line 379 + 30 bytes NEWLINE getSlotValues(RDF_DBStruct * 0x00a62500, RDF_ResourceStruct * 0x009ade10, NEWLINE RDF_ResourceStruct * 0x00990b40, unsigned short 0x0001, int 0x00000001, int NEWLINE 0x00000001) line 680 + 98 bytes NEWLINE RDF_GetSources(RDF_DBStruct * 0x00a62500, RDF_ResourceStruct * 0x009ade10, NEWLINE RDF_ResourceStruct * 0x00990b40, unsigned short 0x0001, int 0x00000001) line 706 NEWLINE + 28 bytes NEWLINE NEWLINE And: NEWLINE + (char*)v 0x00aacca0 "After a moment you´" NEWLINE (That's the page title)
1124|letters a-f after decimal character entity e.g. ' are eaten|The entities Ξ and ξ don't display in NGLayout [12/10/98 nightly build]. NEWLINE There's nothing wrong with the numerical values given by the parser, but NEWLINE since NGLayout displays, say, &x00e9; (as well as é) as é (é), I'd NEWLINE guess that ξ is being caught here. Is this really needed for compatibility? NEWLINE Neither Nav4 nor IE4/5 do it.
1125|</FONT size> problem|The HTML won't win any prizes, but the following URL - and several others on the NEWLINE same site - work on Nav4 and IE4/5, but crash the NGLayout viewer app [12/10/98 NEWLINE nightly build]: NEWLINE NEWLINE http://www.snerpa.is/net/biblia/opin.htm NEWLINE NEWLINE As far as I can guess, the problem is that </FONT size> doesn't close a <FONT NEWLINE size=n> tag. To test this, copy and paste this line about 100 times into a NEWLINE document: NEWLINE NEWLINE <FONT size=-3><SUP>1</SUP></FONT size>Xxxx. NEWLINE NEWLINE Debug:Dump Content, and watch things keep nesting deeper and deeper. Now try it NEWLINE about 200 times and the viewer crashes. (These figures for an 80Mb Win95 P120, NEWLINE probably different elsewhere). NEWLINE NEWLINE This is presumably an issue for <FONT color> and <FONT face> as well.
1126|Setting column width on COLGROUP not implemented yet|Test case is available at URL given. The syntax we don't support yet is: NEWLINE <colgroup span=5 width=100>
1127|SPAN attribute on COL element not implemented yet|Test case is available at URL given. The syntax we don't support yet is: NEWLINE <col span=5 width=100>
1128|Proportional Width Table Columns not implemented yet|Test case is available at URL given. The syntax we don't support yet is: NEWLINE <col width=50> NEWLINE <col width=2*> NEWLINE <col width=1*> NEWLINE <col width=3*>
1129|Can't apply style properties to table columns (not implemented yet)|Test case is available at URL given. We don't apply styles on COLs or NEWLINE COLGROUPs.
1130|Password INPUTs don't handle VALUE attribute correctly|<input type="password" value="foo" size=5> NEWLINE NEWLINE expected behavior: NEWLINE [*** ] NEWLINE NEWLINE actual behavior: NEWLINE [ ] NEWLINE NEWLINE The value doesn't appear, even in the form of asterisks.
1131|Text input fields have extra whitespace|I think I might have filed this bug before; if so, please mark as duplicate. NEWLINE NEWLINE <input type=text size=3 value="123"> NEWLINE NEWLINE Expected behavior (this is what I see in IE5 and Nav4.5): NEWLINE [123] NEWLINE NEWLINE Actual behavior: NEWLINE [ 1 ] NEWLINE NEWLINE There is blank space inside the left and right edges of the input field.
1132|We don't emulate this TEXTAREA quirk|Legacy browsers (IE5, Nav4.5) treat text inside a TEXTAREA like PRE by default, NEWLINE meaning newlines are honored, so: NEWLINE NEWLINE <textarea> NEWLINE 1 NEWLINE 2 NEWLINE 3 NEWLINE </textarea> NEWLINE NEWLINE ... will render on 3 seperate lines. NEWLINE NEWLINE However, all whitespace (including newlines) before the first non-whitespace NEWLINE character is ignored. Which means that these two examples should render NEWLINE identically: NEWLINE NEWLINE <textarea>1</textarea> NEWLINE <textarea> NEWLINE 1</textarea> NEWLINE NEWLINE In ngLayout, we render the first newline, which should be ignored.
1133|READONLY attribute not supported|The READONLY attribute applies to text input fields (text and password) and NEWLINE textareas only. NEWLINE NEWLINE A test case is available at the URL given, at the very bottom of the page. NEWLINE NEWLINE Behavior of READONLY controls is described in the html 4.0 spec, and they work NEWLINE in ie5 for testing.
1134|DISABLED attribute not supported on OPTION element|The disabled attribute works almost everywhere (terrific!) but not yet on the NEWLINE OPTION element. A test case is available at the URL given. It doesn't work right NEWLINE in ie5b1 either.
1135|DEL and INS need some default style|Currently, these HTML 4.0 elements are not visually rendered in a distinctive NEWLINE fashion. IE 5 uses this style, from what I can tell: NEWLINE NEWLINE INS { text-decoration: underline; } NEWLINE DEL { text-decoration: strike-through; }
1136|Title of viewer window changes as frames are loaded|As you load Test 9, the title of the viewer window changes as the frames load. NEWLINE The title of the viewer window should be the title of the _top frameset NEWLINE document.
1137|Selection is wrong color|The selection should be the default selection color for the user's environment NEWLINE (the OS tracks this somewhere), not black. NEWLINE NEWLINE When I make a selection, the background selection color flickers.
1138|HTML 4.0 "Q" (Quotation) tag not implemented yet|The URL given contains a test case for the <Q> tag, which is not implemented yet NEWLINE in NGLayout.
1139|DFN tag needs default style|The DFN tag needs a default style. IE uses (from what I can tell): NEWLINE NEWLINE DFN { font-style: italic; }
1140|XMP tag not implemented yet|Support for the XML tag has not been fully implemented yet. I remember you two NEWLINE (rickg,kipp) talking about this once, but I forgot the details. NEWLINE NEWLINE The URL given contains a testcase (at the very bottom).
1141|CSS inherit Value should be supported|The inherit value is very useful for authors to develop stylesheets in cascades NEWLINE where they want to inherit values from the parent element. I don't think it is NEWLINE quite as important as the universal selector (bug #850), but it is worth NEWLINE implementing. NEWLINE ‰
1142|Strange Border Bugs|The following test case should show a thin black double border, but instead, NEWLINE only a bottom border is shown: NEWLINE NEWLINE <style type=text/css> NEWLINE H2 { NEWLINE border-bottom: thin solid black; NEWLINE } NEWLINE </style> NEWLINE <body> NEWLINE <h2> NEWLINE Test test test test test NEWLINE </h2> NEWLINE </body> NEWLINE NEWLINE The following test should show an H1 with a bottom border, followed by an H2 NEWLINE surrounded by a double black border. Instead, the double border around the H2 NEWLINE appears extra thick (on top, outside and bottom, inside lines of the double NEWLINE border): NEWLINE NEWLINE <style type=text/css> NEWLINE H1 { NEWLINE border-bottom: thin solid black NEWLINE } NEWLINE NEWLINE H2 { NEWLINE border: double black; NEWLINE } NEWLINE </style> NEWLINE </head> NEWLINE <body> NEWLINE <h1>test</h1> NEWLINE <h2>test</h2> NEWLINE </body> NEWLINE NEWLINE If you remove the "thin" from the first border declaration, the problem goes NEWLINE away.
1143|Changing style on anchor in list causes layout glitch|Test case follows. Not sure where this one belongs... The weird behavior is: NEWLINE NEWLINE 1. When you mouseover a link for the first time, all of the bullets in the NEWLINE unordered list shrink in size slightly. NEWLINE NEWLINE 2. When you mouseout of a link and change the style of the anchor for the second NEWLINE time, additional "padding" (I don't know if it's really padding) appears around NEWLINE the text NEWLINE NEWLINE <html> NEWLINE <head> NEWLINE <script> NEWLINE function hotlink(e) { this.style.color="red"; } NEWLINE function dimlink(e) { this.style.color="blue"; } NEWLINE function init() { NEWLINE var ancs = document.getElementsByTagName("A"); NEWLINE for (var i=0; i<ancs.length; i++) { NEWLINE ancs[i].onmouseover=hotlink; NEWLINE ancs[i].onmouseout=dimlink; NEWLINE } NEWLINE } NEWLINE </script> NEWLINE <body onload="init()"> NEWLINE <ul> NEWLINE <li><a href="#">Foo</a> NEWLINE <li><a href="#">Foo</a> NEWLINE </ul> NEWLINE </body> NEWLINE </html>
1144|SMART_MAIL=1 needs to be on by default for Win32, Mac|I don't think this is valid in the seamonkey world. Please reopen when it NEWLINE becomes valid again.
1145|Need wrapper script for executable|Since we're shipping .so's now, we should probably NEWLINE include a run-mozilla script that sets LD_LIBRARY_PATH NEWLINE up for the user. I've created a first version of this NEWLINE (/u/mcafee/xp/nswrap), we should probably rename this NEWLINE to run-mozilla, check it into the xfe tree somewhere, NEWLINE and make this part of make-release.
1146|the DOM memory leak is enourmous...|When using the DOM way of accessing the document, i get _huge_ memory NEWLINE leaks of over 170 megabytes, but when the HTML is something called NEWLINE 'well formed' (XML term), it leaks less, but still it leaks unbelieavably much.
1147|Reflow of PRE text inside tables|There is a problem with reflow of PRE text inside tables, because the size of NEWLINE the table changes when you resize. A good example is viewing a diff using NEWLINE Bonsai.
1148|Test bug.|(This bug imported from BugSplat, Netscape's internal bugsystem. It NEWLINE was known there as bug #15984 NEWLINE http://scopus.netscape.com/bugsplat/show_bug.cgi?id=15984 NEWLINE Imported into Bugzilla on 10/19/98 11:40) NEWLINE NEWLINE Should end up with product being "Dogbert", and version being "4.0a1".
1149|Test bug.|(This bug imported from BugSplat, Netscape's internal bugsystem. It NEWLINE was known there as bug #15984 NEWLINE http://scopus.netscape.com/bugsplat/show_bug.cgi?id=15984 NEWLINE Imported into Bugzilla on 10/19/98 11:48) NEWLINE NEWLINE Should end up with product being "Dogbert", and version being "4.0a1".
1150|{inc} extra newlines triggering an extra incremental reflow during doc loading|Because we aren't yet doing a good job of getting rid of all the extra newlines NEWLINE in an HTML document we end up having an extra incremental reflow occur after all NEWLINE of the document has been loaded because the parser takes all the buffered NEWLINE newlines and appends them to the content before finishing. NEWLINE NEWLINE We need to work out a proper solution to this to avoid the reflow, and to NEWLINE lighten up the total document storage.
1151|We don't honor the "border" attribute on input type=image|Here's a simple test case, taken from a netcenter sign-in page: NEWLINE NEWLINE <html> NEWLINE <body> NEWLINE <form> NEWLINE <INPUT TYPE=image SRC="http://ureg.netscape.com/images/signin_button.gif" NEWLINE WIDTH=55 HEIGHT=23 BORDER=0> NEWLINE </form> NEWLINE </body> NEWLINE </html> NEWLINE NEWLINE Currently, we paint the default blue border around the image, despite the NEWLINE "border=0."
1152|Layout: ftp URL's are blank|Marked WONTFIX because NGLayout is going to replace the current layout engine.
1153|nsHTMLReflowState has incorrect percentage min values|The following has two problems (1) The width is constrained relative to the NEWLINE viewer screen, not the DIV and (2) The height is not being constrained at NEWLINE all (the nsHTMLReflowState being passed to the image's Reflow is saying that NEWLINE HaveConstrainedWidth() is false). Substituting other elements for <IMG> gives NEWLINE the same results. NEWLINE NEWLINE <DIV style="width: 300px; height: 300px; border: 2px solid black;"> NEWLINE <img src=raptor.jpg style="width: 50%; height: 50%;"> NEWLINE </DIV>
1154|body frame not handling constrained widths|nsFieldSetFrame has a body frame. When the nsFieldSetFrame is reflowed with a NEWLINE constrained width, it passes this along to the body's Reflow. However, the body NEWLINE frame is not honoring the minWidth. The example below illustrates this; the NEWLINE images should have been reflowed below one another. I did not try constrained NEWLINE heights. NEWLINE NEWLINE <body> NEWLINE <FORM> NEWLINE <fieldset style="display: inline; width: 75px; height: 200px"> NEWLINE <legend>hello</legend> NEWLINE <img src=raptor.jpg style="width:50px; height:50px;"> NEWLINE <img src=raptor.jpg style="width:50px; height:50px;"> NEWLINE </fieldset> NEWLINE </form> NEWLINE </body>
1155|OBJECTs do not degrade correctly|OBJECTs with unknown MIME types do not degrade correctly NEWLINE - i.e. the content of the OBJECT element is not displayed. NEWLINE For example: NEWLINE NEWLINE <object type="video/mpeg" data="test.mpg"> NEWLINE alternative text NEWLINE </object> NEWLINE NEWLINE The words "alternative text" should be rendered, if NGLayout NEWLINE cannot find a plugin for video/mpeg. Currently, they are not.
1156|OBJECTs without type attribute aren't rendered|OBJECTs without the *optional* type attribute are not NEWLINE rendered at all, even when the file format is a familiar NEWLINE one. They don't degrade correctly, either. NEWLINE NEWLINE Example: NEWLINE NEWLINE <object data="test.avi"> NEWLINE alternative text NEWLINE </object> NEWLINE NEWLINE Either the test.avi video or the words "alternative text" should be rendered. NEWLINE Currently, nothing is displayed.
1157|AVI and WAV OBJECTs crash again|(Note: I ran these tests with both the plugins from NEWLINE Navigator 4.5 PR2 and the old versions from Navigator 3.0. NEWLINE I changed the plugins by modifying the registry at NEWLINE "hkey_current_user/software/netscape/netscape NEWLINE navigator/main/install directory".) NEWLINE NEWLINE Trouble with AVI and WAV OBJECTs. The following NEWLINE problems occurred on any OBJECT with the MIME type NEWLINE "video/x-msvideo" or "audio/wav". NEWLINE NEWLINE The plugin gets loaded, but the video/audio clip won't play. NEWLINE NEWLINE When using the plugins from 4.5 PR2, the AVI object is just NEWLINE an empty space where the previous page shows through. NEWLINE Clicking right button on the space gives a NPAVI32 menu, but NEWLINE selecting "Play" or anything else does nothing. When I use NEWLINE the plugins from Nav3.0, NGLayout crashes immediately on NEWLINE loading the page! NEWLINE NEWLINE In the WAV tests, a LiveAudio plugin panel saying "Loading NEWLINE audio file..." is shown. Clicking on the panel a few times NEWLINE causes a crash. The behaviour is identical with both plugin NEWLINE versions. NEWLINE NEWLINE Here's a sample output from my NGLayout on a WAV object test: NEWLINE NEWLINE loaded plugin npaudio.dll for mime type audio/wav NEWLINE result of creating plugin adapter: 0 NEWLINE successfully created plugin instance 01253dd0 for NEWLINE npaudio.dll, mimetype audio/wav NEWLINE instance start called NEWLINE NEWLINE The output is similar in the AVI tests.
1158|OBJECTs in natively supported formats aren't rendered at all|GIF, JPEG, PNG, HTML and plain text OBJECTs are not NEWLINE displayed - even though NGLayout is able to display inline NEWLINE images in those 3 formats, and IFRAMEs in HTML or plain NEWLINE text. NGLayout says that it's "unable to find plugin to NEWLINE handle [the MIME type in question]". It shouldn't really NEWLINE need to look for plugins when it can handle the object NEWLINE natively. NEWLINE NEWLINE Example: NEWLINE NEWLINE <object type="image/gif" data="test.gif"> NEWLINE alternative text NEWLINE </object> NEWLINE NEWLINE Either a GIF image or "alternative text" should be displayed. Currently, NEWLINE there's just a grey box.
1159|browser buster not loading a page or two on 10/19 raptor bits|win95, laptop, ~ 35mb available on disk. NEWLINE several attempts at running xpviewer on a 30 sec. cycle got crashes NEWLINE on the first or second page load.
1160|Apprunner window paints incorrectly|when trying to run browser buster on say a 30 second cycle NEWLINE the upper lefthand frame does not correct contents. NEWLINE NEWLINE looks like some mush from the previous frame.
1161|This animiated GIF is flashing a lot - looks bad|It's almost as though we're not double buffering for some reason. The animated NEWLINE GIF is at the top of the page mentioned in the URL.
1162|Scrollbars never appear if you load this URL|Try loading that URL into the Raptor test viewer. I imagine this occurs on other NEWLINE pages where the anchor leads you somewhere "below the fold." The vertical NEWLINE scrollbar never appears, although you can scroll via selection or pgdn/pgup key NEWLINE presses.
1163|The bordercolor attribute is not being honored|Build 10-21-98 NEWLINE Win Nt 4. sp.3 NEWLINE NEWLINE Border Color tag is not supported in the following page. NEWLINE NEWLINE <!doctype html public "-//w3c//dtd html 4.0 transitional//en"> NEWLINE <html> NEWLINE <!-- testcase: border_color.htm --> NEWLINE <!-- Author: Michael Cobbley, Date: 10/19/98 --> NEWLINE <!-- CONFIDENTIAL AND PROPRIETARY TEST SCRIPT OF NETSCAPE COMMUNICATIONS NEWLINE CORPORATION NEWLINE Copyright © 1998 Netscape Communications Corporation. All Rights Reserved. Use NEWLINE of this Test Script is NEWLINE subject to the terms of the applicable license agreement from Netscape NEWLINE Communications Corporation. --> NEWLINE <!-- Purpose: HTML 3.2 and HTML 4.0 The purpose is to specify a specific border NEWLINE color. --> NEWLINE <!-- Expected result: In Communicator the border should be Blue.--> NEWLINE NEWLINE <head> NEWLINE <meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1"> NEWLINE <meta name="Author" content="Mike Cobbley"> NEWLINE <meta name="GENERATOR" content="Mozilla/4.5 [en]C-NSCP (WinNT; I) NEWLINE [Netscape]"> NEWLINE <title>Border_Color</title> NEWLINE </head> NEWLINE <body> NEWLINE NEWLINE <table border="3" BORDERCOLOR="#RRGGBB"> NEWLINE <caption>Testing BORDERCOLOR tag the border shood be blue.</caption> NEWLINE NEWLINE <tr> NEWLINE <td>mike</td> NEWLINE NEWLINE <td>was</td> NEWLINE NEWLINE <td>here</td> NEWLINE NEWLINE <td>mike</td> NEWLINE NEWLINE <td>was</td> NEWLINE NEWLINE <td>here</td> NEWLINE </tr> NEWLINE NEWLINE <tr> NEWLINE <td>mike</td> NEWLINE NEWLINE <td>was</td> NEWLINE NEWLINE <td>here</td> NEWLINE NEWLINE <td>mike</td> NEWLINE NEWLINE <td>was</td> NEWLINE NEWLINE <td>here</td> NEWLINE </tr> NEWLINE </table> NEWLINE NEWLINE </body> NEWLINE </html>
1164|ALIGN=MIDDLE and ALIGN=ABSMIDDLE values are not being supported|This tag works fine in 4.0x and 4.5 but is not supported now. NEWLINE NEWLINE NEWLINE <HTML> NEWLINE <head> NEWLINE <!-- testcase: table_abs.htm --> NEWLINE <!-- Author: Michael Cobbley, Date: 10/19/98 --> NEWLINE <!-- CONFIDENTIAL AND PROPRIETARY TEST SCRIPT OF NETSCAPE COMMUNICATIONS NEWLINE CORPORATION NEWLINE Copyright © 1998 Netscape Communications Corporation. All Rights Reserved. Use NEWLINE of this Test Script is NEWLINE subject to the terms of the applicable license agreement from Netscape NEWLINE Communications Corporation. --> NEWLINE <!-- Purpose: HTML 3.2 and HTML 4.0 To test the alignment of objects. --> NEWLINE <!-- Expected result: The ABS middle tag should center the nested table in the NEWLINE middle of cell one. This tag will not work in IE 4--> NEWLINE NEWLINE <meta name="Author" content="Mike Cobbley"> NEWLINE <meta name="GENERATOR" content="Mozilla/4.5 [en]C-NSCP (WinNT; I) NEWLINE [Netscape]"> NEWLINE </head> NEWLINE <body> NEWLINE NEWLINE <table BORDER COLS=2 WIDTH="50%" > NEWLINE <tr> NEWLINE <td><table ALIGN="ABSMIDDLE" BORDER COLS=2 WIDTH="5%" > NEWLINE <tr> NEWLINE <td>hello</td> NEWLINE NEWLINE <td>hello</td> NEWLINE </tr> NEWLINE NEWLINE <tr> NEWLINE <td>hello</td> NEWLINE NEWLINE <td>hello</td> NEWLINE </tr> NEWLINE </table></td> NEWLINE NEWLINE <td>hello</td> NEWLINE </tr> NEWLINE NEWLINE <tr> NEWLINE <td>hello</td> NEWLINE NEWLINE <td>hello</td> NEWLINE </tr> NEWLINE </table> NEWLINE NEWLINE </body> NEWLINE </html>
1165|margins on floating elements misbehave|Setting margin properties on floating elements (images, in this case) has NEWLINE effect on their placement, but not quite as expected. The enclosed URL has NEWLINE reference images. ‰
1166|<BUTTON> is confined to <FORM>|<BUTTON> needs to exist outside of a <FORM>. A button outside of a form can NEWLINE service javascript needs. The following should be allowed. NEWLINE NEWLINE <html> NEWLINE <body> NEWLINE <button>hello button</button> NEWLINE </body> NEWLINE </html>
1167|edit mode is broken|something in the content change-over doesn't work :-(
1168|can't log in to bugzilla using raptor|This page asks for an email address and password which I supply. I click Login NEWLINE and it seems to accept the values, but then returns the same page with the form NEWLINE fields emptied. I can't log in. I think this bug is relatively high priority NEWLINE because we want our testers to use Viewer as their everyday browser, and they'll NEWLINE need this.
6437|content2.xul: non-localizable XUL file|BULK MOVE: Changing component from XUL to XP Toolkit/Widgets: XUL. XUL NEWLINE component will be deleted.
1170|missing image|I can't tell if this is a layout problem or an imagelib problem. We're missing NEWLINE an image in the top banner (right side, next to "Microsoft" logo.) Nav4 shows NEWLINE the image simply as a black rectangle. IE shows the "Micorsoft Network" image. NEWLINE Maybe the site is vending different pages based on the browser type?
1171|The voucher field appears as a number when listing users.|We don't (as despots) have an easy way to match id's to email addresses, so it'd NEWLINE be nice to have the email address listed in the 'voucher' field, rather than id.
1172|frames dont work with javascript|When trying to output to a frame in javascript, nothing is printed..
1173|margin-top incorrect for inlines|The following does not render correctly. Similar results are obtained with other NEWLINE inlines besides <SPAN>. NEWLINE NEWLINE <html> NEWLINE <body> NEWLINE <SPAN style="border: 1px solid black;">SPAM</SPAN> NEWLINE <BR> NEWLINE <SPAN style="border: 1px solid black; margin-top: 10px">EGGS</SPAN> NEWLINE <BR> NEWLINE <SPAN style="border: 1px solid black;">SPAM</SPAN> NEWLINE </body> NEWLINE </html>
1174|floaters handled poorly|the left column of this page is one big table with align=left. NEWLINE If you remove the align=left attribute from this table, everything lays out well NEWLINE (though not as intended, obviously.)