-
Notifications
You must be signed in to change notification settings - Fork 2
/
conference-log
6704 lines (6532 loc) · 514 KB
/
conference-log
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
AOSCC 2018 — 主群组
From 2018-07-17 11:08:55 to 2018-07-30 03:40:32, total 5971
[2018-07-17 11:08:55] gumblex >>> [chat_add_user]
[2018-07-17 12:22:42] Mingcong Bai >>> [chat_change_photo]
[2018-07-17 12:22:58] Mingcong Bai >>> [chat_change_photo]
[2018-07-17 14:32:14] 希瑀 蔡 >>> [chat_add_user]
[2018-07-17 14:33:11] 迫真 正版 >>> [chat_add_user]
[2018-07-17 15:01:02] LAX VPS >>> [chat_add_user]
[2018-07-17 21:14:50] LAX VPS [Re: 10] >>> get
[2018-07-17 21:21:43] LAX VPS >>> @JeffBai need to invite @aoscc2018_bot
[2018-07-17 21:24:55] Mingcong Bai >>> 我的 iOS App bug 了
[2018-07-17 21:25:01] Mingcong Bai >>> 点不动
[2018-07-17 21:25:11] LAX VPS >>> 噗
[2018-07-17 21:27:11] Mingcong Bai >>> [chat_add_user]
[2018-07-17 21:27:14] Mingcong Bai >>> 搞定
[2018-07-17 21:28:02] AOSCC Relay bot >>> [Lily-Ayta] ojbk 不过irc这边的mod flag(
[2018-07-18 09:19:04] Mingcong Bai [Re: 19] >>> @Icenowy
[2018-07-18 09:19:13] 冰淇琳 >>> ?
[2018-07-18 09:19:37] Mingcong Bai >>> mod flag?
[2018-07-18 09:19:51] Mingcong Bai >>> 还是我理解错了什么
[2018-07-18 11:53:29] 井生 羽衣 >>> [chat_add_user]
[2018-07-19 12:52:18] 猫晴 弥柚 >>> [chat_add_user]
[2018-07-19 12:53:27] Sakamoto Tanmy >>> [chat_add_user]
[2018-07-19 12:55:56] Luke Yue >>> [chat_add_user]
[2018-07-19 12:57:16] 无聊怪 >>> [chat_add_user]
[2018-07-19 12:58:34] Ted Cynx >>> [chat_add_user]
[2018-07-19 13:00:14] Lion [Re: 24] >>> 野兽前辈瞩目
[2018-07-19 13:02:39] Mingcong Bai >>> 欢迎来到 AOSCC 2018 主群组!本群组和 IRC 频道 #aoscc-2018 同步。
— 本群组仅限 AOSCC 议程内使用,在议程规定时间之外请不要在此处进行任何讨论。
— 本群组禁止一切与议程中话题无关的讨论。
— 本群组禁止使用 Sticker。
集会日程: https://github.com/AOSC-Dev/aoscc/blob/master/2018/README.md
AOSCC 主群组: https://t.me/joinchat/BMnG90FZDQlSnazLvYCWpA
———————
**AOSCC 2018 已开始收集代号提名和壁纸投稿**
本年度 Core 6 代号将以 F 开头,但是:
— 代号不得包含政治意味和该方面敏感内容
— 代号不得包含任何粗俗和侮辱性语言
— 代号不得为 F 单字母
— 代号不得包含不可自由使用的商标
本年度收集默认壁纸投稿以及壁纸集投稿,其中默认壁纸投稿要求:
— 默认壁纸为原创且可跟随 AOSC OS 重分发的作品
— 默认壁纸为计算机平面或各类辅助设计,不接受任何未经风格化处理的摄影作品
— 默认壁纸不得包含政治意味和该方面敏感内容
— 默认壁纸限定每人投一种设计,但不限定同样设计下不同风格组成的集合
而壁纸集投稿的话只要求原创、可分发及禁止政治意味和该方面敏感内容。
代号提名(UTC 时间 7 月 21 日前)及壁纸投稿(UTC 时间 7 月 28 日前)请发送到 aoscc2018@aosc.io 。 [webpage]
[2018-07-19 13:02:46] Mingcong Bai [Re: 31] >>> pinned the message
[2018-07-19 13:03:53] Catten Linger >>> [chat_add_user]
[2018-07-19 13:03:59] Catten Linger >>> \ o /
[2018-07-19 13:04:29] Mingcong Bai >>> 我清理下消息(见 pin msg
[2018-07-19 13:05:49] Mingcong Bai >>> **AOSCC 2018 已开始收集代号提名和壁纸投稿**
本年度 Core 6 代号将以 F 开头,但是:
— 代号不得包含政治意味和该方面敏感内容
— 代号不得包含任何粗俗和侮辱性语言
— 代号不得为 F 单字母
— 代号不得包含不可自由使用的商标
本年度收集默认壁纸投稿以及壁纸集投稿,其中默认壁纸投稿要求:
— 默认壁纸为原创且可跟随 AOSC OS 重分发的作品
— 默认壁纸为计算机平面或各类辅助设计,不接受任何未经风格化处理的摄影作品
— 默认壁纸不得包含政治意味和该方面敏感内容
— 默认壁纸限定每人投一种设计,但不限定同样设计下不同风格组成的集合
而壁纸集投稿的话只要求原创、可分发及禁止政治意味和该方面敏感内容。
代号提名(UTC 时间 7 月 21 日前)及壁纸投稿(UTC 时间 7 月 28 日前)请发送到 aoscc2018@aosc.io 。
[2018-07-19 17:08:16] Estel >>> [chat_add_user]
[2018-07-19 17:19:01] liushuyu >>> @JeffBai 我需要更改主题,应该是 "AOSC OS: Contiguous Integration of System Delivery"
[2018-07-19 18:15:53] OriginCode >>> [chat_add_user]
[2018-07-19 23:12:10] FlyGoat >>> [chat_add_user]
[2018-07-20 00:17:54] 御坂0x4e21 看到她灌水一律赶她下线 >>> [chat_add_user]
[2018-07-20 09:34:15] CRT >>> [chat_add_user]
[2018-07-20 10:24:50] Hung-I Wang >>> [chat_add_user]
[2018-07-20 11:25:55] NeoTimesharingSystem >>> [chat_add_user]
[2018-07-20 11:26:18] Fung Gwo >>> [chat_add_user]
[2018-07-20 11:35:27] 艾颖初 (。•́︿•̀。) | Systemd 真好玩 >>> [chat_add_user]
[2018-07-20 12:28:20] 櫻川 Y00 | 为什么男装这么无聊 >>> [chat_add_user]
[2018-07-20 13:08:28] 迫真 正版 >>> 你们选代号的时候我这边早上8点的样子
[2018-07-20 13:08:34] 迫真 正版 >>> 好方(x
[2018-07-20 13:09:41] Lion [Re: 49] >>> 嘘,会议时间还没到🌚
[2018-07-20 13:10:07] 迫真 正版 [Re: 50] >>> 反正日程都出来了(小声
[2018-07-20 13:12:26] Lion [Re: 51] >>> 看第一条消息第一小条(゜-゜)
[2018-07-20 13:12:47] 迫真 正版 >>> ao
[2018-07-20 13:20:38] AOSCC Relay bot >>> [YJSNPI] IRC Channel 這邊 Topic 都沒設置嘛
[2018-07-20 17:12:21] fr >>> [chat_add_user]
[2018-07-20 21:02:52] jactry >>> [chat_add_user]
[2018-07-20 21:16:47] AOSCC Relay bot >>> [mingcongbai] test
[2018-07-20 21:16:59] Mingcong Bai >>> test
[2018-07-20 21:55:37] 千 qian >>> [chat_add_user]
[2018-07-20 21:55:40] 𝓦𝓑𝓒𝓛 >>> [chat_add_user]
[2018-07-20 22:42:46] 白雾 纱织 >>> [chat_add_user]
[2018-07-21 00:38:36] Mingcong Bai [Fwd: Mingcong Bai] >>> 本年度 Core 6 代号将以 F 开头,但是:
— 代号不得包含政治意味和该方面敏感内容
— 代号不得包含任何粗俗和侮辱性语言
— 代号不得为 F 单字母
— 代号不得包含不可自由使用的商标
代号提名将于北京时间周六早八点(UTC 零时)截止。有意提名者请抓紧时间发送代号到 aoscc2018@aosc.io 。
[2018-07-21 05:02:56] Mingcong Bai >>> [Forwarded from Mingcong Bai]
本年度 Core 6 代号将以 F 开头,但是:
— 代号不得包含政治意味和该方面敏感内容
— 代号不得包含任何粗俗和侮辱性语言
— 代号不得为 F 单字母
— 代号不得包含不可自由使用的商标
代号提名将于北京时间周六早八点(UTC 零时)截止。有意提名者请抓紧时间发送代号到 aoscc2018@aosc.io 。
[2018-07-21 05:08:47] cth451 ǝɔuǝƃɹǝʌᴉp‾ɥʇɔ >>> 每天滚动播送消息(
[2018-07-21 07:31:33] Mingcong Bai >>> 各位早上好,今日议程还有一个半小时就开始了。不过今天早上九点到九点半是给各位自由活动的时间(现场 AOSCC 模拟器 2018),接下来今天早上有两个讨论话题:
— 9:30 - 10:30,调整 AOSC OS 的更新周期和里程碑计划集成(本人主持)
— 11:00 - 12:00,AOSC OS 打包工作流和工具链版本管控(@Lionium 主持)
而后今晚有三个议程:
— 21:00 - 22:00,关于 AOSC OS 安全更新工作流程的介绍和安全更新维护者召集(@l2dy0 主持)
— 22:15 - 23:15,关于 AOSC OS 软件包二进制优化 overlay 的实现计划(@lmy441900 主持)
— 23:30 - 0:00,Core 6 代号投票
[2018-07-21 07:32:57] Mingcong Bai >>> 每个话题开始前我将 pin 上话题标题并通知所有成员。
[2018-07-21 07:40:16] Leway Colin >>> 👌
[2018-07-21 08:00:15] Mingcong Bai >>> 倒计时一小时,代号征集截止
[2018-07-21 08:01:34] Mingcong Bai >>> 将有 30 个代号加入今晚的投票(因为数量繁多且 @vote 无法处理超过 10 个选项,所以分开了三个小组;将进行多轮投票)。
[2018-07-21 08:03:26] OriginCode >>> 了解
[2018-07-21 08:20:31] AOSCC Relay bot >>> [mingcongbai] Pre-assembly IRC test
[2018-07-21 08:20:39] Mingcong Bai >>> Pre-assembly Telegram test
[2018-07-21 08:20:42] Mingcong Bai >>> Good.
[2018-07-21 08:49:50] liushuyu >>> 按照章程,第一场是预热/闲聊对吧
[2018-07-21 08:50:59] Mingcong Bai [Re: 65] >>> 嗯,九点到九点半,见上
[2018-07-21 08:51:18] Mingcong Bai >>> (今年第一次最大化 Telegram Desktop(x
[2018-07-21 08:55:57] Mingcong Bai >>> 五分钟 | Five Minutes | Пять минут
[2018-07-21 08:57:19] Zero King >>> 5分鐘 | 五分 | 5 분
[2018-07-21 08:58:18] Mingcong Bai >>> năm phút
[2018-07-21 08:58:53] OriginCode >>> 2分钟倒计时(
[2018-07-21 08:59:52] liushuyu >>> 10 秒
[2018-07-21 09:00:07] liushuyu >>> 好了
[2018-07-21 09:00:07] Mingcong Bai >>> #TOPIC 9:00 - 9:30 — 热场,闲聊
[2018-07-21 09:00:08] Lion >>> 喵
[2018-07-21 09:00:12] Mingcong Bai [Re: 83] >>> pinned the message
[2018-07-21 09:00:26] 迫真 正版 >>> 你是一个
[2018-07-21 09:00:28] liushuyu [Re: 84] >>> 狮子今天起这么早
[2018-07-21 09:00:29] 迫真 正版 >>> 开发者
[2018-07-21 09:00:44] Mingcong Bai >>> 各位早(
[2018-07-21 09:00:44] Lion [Re: 87] >>> 待会第二个就到我了🌝
[2018-07-21 09:00:49] liushuyu >>> 有点违反常理吧(
[2018-07-21 09:00:50] Mingcong Bai >>> 最后罗嗦一句
[2018-07-21 09:00:52] 迫真 正版 >>> 话说还有honeydrive这样的发行版啊
[2018-07-21 09:00:59] 迫真 正版 >>> 🌚
[2018-07-21 09:01:00] Mingcong Bai >>> 接下来今天早上有两个讨论话题:
— 9:30 - 10:30,调整 AOSC OS 的更新周期和里程碑计划集成(本人主持)
— 11:00 - 12:00,AOSC OS 打包工作流和工具链版本管控(@Lionium 主持)
而后今晚有三个议程:
— 21:00 - 22:00,关于 AOSC OS 安全更新工作流程的介绍和安全更新维护者召集(@l2dy0 主持)
— 22:15 - 23:15,关于 AOSC OS 软件包二进制优化 overlay 的实现计划(@lmy441900 主持)
— 23:30 - 0:00,Core 6 代号投票
[2018-07-21 09:01:00] OriginCode >>> 开始闲聊(
[2018-07-21 09:01:01] liushuyu [Re: 89] >>> 晚上好(
[2018-07-21 09:01:07] Mingcong Bai [Re: 93] >>> 这啥
[2018-07-21 09:01:11] Mingcong Bai [Re: 97] >>> 晚上好(
[2018-07-21 09:01:14] 迫真 正版 [Re: 98] >>> 蜜罐发行版
[2018-07-21 09:01:30] Mingcong Bai [Re: 100] >>> 最近还听说有个 Freenode 组织者做的发行版
[2018-07-21 09:01:30] Zero King >>> honeypot(
[2018-07-21 09:01:35] OriginCode >>> (咱还有半小时左右到晚饭场地)
[2018-07-21 09:01:36] 迫真 正版 >>> Honeypots in a box!
[2018-07-21 09:01:38] Mingcong Bai >>> “为了保存我的一些特殊爱好”
[2018-07-21 09:01:41] Lion [Re: 100] >>> 从头到尾安“充满恶意”的发行版??
[2018-07-21 09:01:58] Mingcong Bai >>> https://www.dailydot.com/debug/linux-exherbo-developer-debunked/ [webpage]
[2018-07-21 09:01:59] OriginCode [Re: 101] >>> 哦?
[2018-07-21 09:02:00] liushuyu [Re: 106] >>> 吸引狮子么(
[2018-07-21 09:02:03] Mingcong Bai [Re: 107] >>> 噢?
[2018-07-21 09:02:11] Lion [Re: 109] >>> 装爆
[2018-07-21 09:02:11] Mingcong Bai >>> 我上次看见的就是这个……
[2018-07-21 09:02:12] 猫晴 弥柚 >>> 早
[2018-07-21 09:02:17] Mingcong Bai [Re: 113] >>> 早上好
[2018-07-21 09:02:20] liushuyu [Re: 113] >>> 早
[2018-07-21 09:02:23] Mingcong Bai >>> (能专心聊天感觉真好
[2018-07-21 09:02:26] 迫真 正版 [Re: 106] >>> 是的,因为3年没更新,基于xubuntu 12.04…
[2018-07-21 09:02:26] 猫晴 弥柚 >>> utctime 1:00
[2018-07-21 09:02:27] OriginCode [Re: 113] >>> 早
[2018-07-21 09:02:29] 井生 羽衣 >>> 早喵~
[2018-07-21 09:02:34] Mingcong Bai [Re: 118] >>> 哇你在 GMT?
[2018-07-21 09:02:35] liushuyu >>> 早
[2018-07-21 09:02:38] Mingcong Bai [Re: 120] >>> 早
[2018-07-21 09:02:48] Lion [Re: 117] >>> 故意不更新还行
[2018-07-21 09:02:49] Mingcong Bai >>> 可惜不是现场(
[2018-07-21 09:02:50] OriginCode >>> 18点(
[2018-07-21 09:02:53] liushuyu >>> OpenJDK 8 的字体系统真的很先进(
[2018-07-21 09:02:55] Mingcong Bai [Re: 126] >>> 20:02
[2018-07-21 09:03:18] liushuyu [Re: 127] >>> 昨天调查清楚了(
[2018-07-21 09:03:23] Mingcong Bai [Re: 129] >>> 唔
[2018-07-21 09:03:28] OriginCode >>> 明天早上起来还得开会(
[2018-07-21 09:03:29] 猫晴 弥柚 >>> 单纯的用utctime来协调俺和俺的诸多盆友时区不同步的问题
[2018-07-21 09:03:43] Mingcong Bai >>> 目前状态( [photo]
[2018-07-21 09:03:48] Mingcong Bai [Re: 132] >>> 聪明
[2018-07-21 09:03:49] 迫真 正版 >>> 我的azure好像显示UTC的
[2018-07-21 09:03:54] 迫真 正版 >>> 233
[2018-07-21 09:03:56] liushuyu [Re: 130] >>> OpenJDK 会给不同框架用不同的方法
[2018-07-21 09:04:08] Mingcong Bai >>> 可怜的 @Icenowy 昨天似乎半夜才到家
[2018-07-21 09:04:10] Lion [Re: 133] >>> 开放括号不关
[2018-07-21 09:04:17] Mingcong Bai >>> 所以尽早我管 IRC(
[2018-07-21 09:04:20] Mingcong Bai >>> 今早 *
[2018-07-21 09:04:20] Lion >>> 爱右括号组织表示……
[2018-07-21 09:04:21] 迫真 正版 [Re: 134] >>> 我把西雅图和国内时差刻在DNA里了(没有
[2018-07-21 09:04:26] Mingcong Bai [Re: 142] >>> Dayum.
[2018-07-21 09:04:30] 猫晴 弥柚 >>> 有人在过utc-5,有人在过utc-6,有人在过utc+8,+5,+1s
[2018-07-21 09:04:38] Mingcong Bai [Re: 143] >>> [photo]
[2018-07-21 09:04:38] 死 肥 宅 鹹魚 [Re: 139] >>> 是原罪
[2018-07-21 09:04:40] OriginCode >>> 目前状态( [photo]
[2018-07-21 09:04:43] liushuyu >>> 1. Awt 使用自己的 t2k 实现
2. JavaFX 使用 fc
[2018-07-21 09:04:50] AOSCC Relay bot >>> [Umbelost] woc YJSNPI 是谁
[2018-07-21 09:04:55] 迫真 正版 [Re: 146] >>> 我一般看到时间直接算((
[2018-07-21 09:05:00] Mingcong Bai [Re: 150] >>> 不清楚诶……
[2018-07-21 09:05:02] liushuyu [Re: 142] >>> )))))) 怎么样
[2018-07-21 09:05:08] OriginCode >>> )
[2018-07-21 09:05:18] 迫真 正版 >>> 热爱Linux的先辈是鉴
[2018-07-21 09:05:18] Mingcong Bai >>> IRC 上 YJSNPI 是谁?
[2018-07-21 09:05:22] Lion [Re: 153] >>> 爱左括号组织……
[2018-07-21 09:05:23] 死 肥 宅 鹹魚 [Re: 150] >>> 我
[2018-07-21 09:05:28] Mingcong Bai [Re: 158] >>> hhhhhh
[2018-07-21 09:05:32] Mingcong Bai [Re: 157] >>> 你看你看(
[2018-07-21 09:05:32] 井生 羽衣 [Re: 158] >>> hhhhhhh
[2018-07-21 09:05:51] liushuyu [Re: 157] >>> 快去学习 Lisp(
[2018-07-21 09:05:52] Mingcong Bai >>> 对了如果各位想用语音发点东西也是可以接受的
[2018-07-21 09:05:52] Lion [Re: 160] >>> 不用括号一身轻🌝
[2018-07-21 09:05:52] 迫真 正版 >>> 😁
[2018-07-21 09:06:02] Mingcong Bai >>> 我会帮忙转文字
[2018-07-21 09:06:07] Lion [Re: 162] >>> Nooooo
[2018-07-21 09:06:14] liushuyu [Re: 166] >>> 人工语音识别(
[2018-07-21 09:06:16] 死 肥 宅 鹹魚 >>> 我另外還有個比利海靈頓(
[2018-07-21 09:06:24] Mingcong Bai >>> 发日语的话帮不上忙(
[2018-07-21 09:06:24] 迫真 正版 >>> 我有jinkela
[2018-07-21 09:06:26] 死 肥 宅 鹹魚 >>> 就是太久沒登錄了不知道還能不能用
[2018-07-21 09:06:28] 迫真 正版 >>> 🌚
[2018-07-21 09:06:32] Zero King >>> 目前状态( [photo]
[2018-07-21 09:06:37] Mingcong Bai [Re: 171] >>> 说起来代号里面真有金坷垃
[2018-07-21 09:06:42] 死 肥 宅 鹹魚 [Re: 170] >>> 君日本語學習可以
[2018-07-21 09:06:53] 迫真 正版 [Re: 175] >>> 肥料(小声
[2018-07-21 09:06:56] liushuyu [Re: 170] >>> 识别系统可以使用 zh_CN en_* ru_RU (
[2018-07-21 09:06:59] Mingcong Bai [Re: 176] >>> 我大屁眼子…… 呸
[2018-07-21 09:07:02] Mingcong Bai >>> 脑袋受不了
[2018-07-21 09:07:10] Mingcong Bai [Re: 178] >>> de_DE 也可以,识别率比较低(
[2018-07-21 09:07:15] 迫真 正版 >>> 特首要学114种语言
[2018-07-21 09:07:15] Mingcong Bai >>> zh_YUE 也可以
[2018-07-21 09:07:20] 井生 羽衣 [Re: 175] >>> 大概是我提的(
[2018-07-21 09:07:29] Mingcong Bai [Re: 182] >>> 以及 514 种非洲方言
[2018-07-21 09:07:29] 猫晴 弥柚 [Re: 182] >>> 羡慕114种语言
[2018-07-21 09:07:34] OriginCode >>> 呜这里只有我一个用手机开会的么(
[2018-07-21 09:07:34] Mingcong Bai [Re: 184] >>> 好像是(
[2018-07-21 09:07:41] 死 肥 宅 鹹魚 [Re: 185] >>> hhhhh
[2018-07-21 09:07:44] liushuyu [Re: 183] >>> 不是 yue_YUE 么(政治正确?
[2018-07-21 09:07:45] Lion >>> 学习一个一个一个语言
[2018-07-21 09:07:46] Luke Yue >>> 手机路过(
[2018-07-21 09:07:51] Mingcong Bai [Re: 174] >>> 我真的不敢肯定是 macOS 还是 *nix(
[2018-07-21 09:07:56] OriginCode [Re: 185] >>> 特首要学所有中国方言(
[2018-07-21 09:08:01] Luke Yue >>> 在驾校围观开会(
[2018-07-21 09:08:06] 死 肥 宅 鹹魚 [Re: 193] >>> 賭五毛 macOS
[2018-07-21 09:08:07] Mingcong Bai [Re: 190] >>> 好好(另外看群规,最上面一条消息
[2018-07-21 09:08:13] Mingcong Bai [Re: 196] >>> 影子不像?
[2018-07-21 09:08:16] Zero King [Re: 193] >>> 两个不同的Telegram当然是macOS
[2018-07-21 09:08:20] Luke Yue >>> 看字体是 macOS(
[2018-07-21 09:08:22] 猫晴 弥柚 >>> 手机当然是用来开小差去撩小姐姐的啦
[2018-07-21 09:08:29] Mingcong Bai [Re: 199] >>> 有道理(
[2018-07-21 09:08:31] liushuyu [Re: 197] >>> 抱歉
[2018-07-21 09:08:38] OriginCode >>> 看字体+字体渲染是 macOS
[2018-07-21 09:08:47] OriginCode >>> 实锤(
[2018-07-21 09:08:54] 迫真 正版 >>> lol
[2018-07-21 09:08:58] Mingcong Bai >>> 说起来现在我在电脑上看见 Noto 以外的字体我都开始觉得别扭了
[2018-07-21 09:09:07] Mingcong Bai >>> iOS 上的引号我也看不惯
[2018-07-21 09:09:14] liushuyu >>> 现在水量应该超过任何水群
[2018-07-21 09:09:15] 迫真 正版 [Re: 207] >>> 装个Windows(逃
[2018-07-21 09:09:15] OriginCode [Re: 207] >>> Source Han Sans(
[2018-07-21 09:09:23] Mingcong Bai [Re: 209] >>> Drawn this channel! (
[2018-07-21 09:09:25] Lion >>> 说起来这里的图片可以转发去 IRC 的吗?
[2018-07-21 09:09:30] Mingcong Bai [Re: 211] >>> 其实就一个
[2018-07-21 09:09:32] 死 肥 宅 鹹魚 >>> San Francisco, San Francisco
[2018-07-21 09:09:33] Mingcong Bai [Re: 213] >>> 可以
[2018-07-21 09:09:41] 死 肥 宅 鹹魚 >>> 蘋方,蘋方
[2018-07-21 09:09:42] Lion >>> 吼啊
[2018-07-21 09:09:42] Mingcong Bai >>> [photo]
[2018-07-21 09:09:58] OriginCode [Re: 217] >>> 在 Arch 上渲染成一坨
[2018-07-21 09:10:03] OriginCode >>> 不能看(
[2018-07-21 09:10:10] 迫真 正版 >>> 233
[2018-07-21 09:10:19] Mingcong Bai >>> 一个 1600x1200 用户的故事 [photo]
[2018-07-21 09:10:23] 死 肥 宅 鹹魚 >>> 我這邊是 Firefox 海星,Chrome 上爆炸了
[2018-07-21 09:10:30] Mingcong Bai >>> Telegram 100% 看得难受,但是 125% 又太大了
[2018-07-21 09:10:32] 井生 羽衣 [Re: 219] >>> [photo]
[2018-07-21 09:10:34] 死 肥 宅 鹹魚 [Re: 223] >>> 羨慕了
[2018-07-21 09:10:44] 迫真 正版 [Re: 223] >>> 羡慕死了
[2018-07-21 09:10:46] Mingcong Bai >>> (而且因为经常把窗口放到右上角所以老看不见
[2018-07-21 09:10:47] liushuyu [Re: 219] >>> 噢,对了 OpenJDK 强行指定 NotoSansCJK 它会选择日语字形渲染简体中文
[2018-07-21 09:10:50] 死 肥 宅 鹹魚 >>> 果凍 IRC 客戶端用的啥
[2018-07-21 09:10:50] OriginCode [Re: 224] >>> Noto Sans 吼啊
[2018-07-21 09:10:53] Mingcong Bai [Re: 230] >>> 肯定……
[2018-07-21 09:10:58] Mingcong Bai [Re: 231] >>> Polari
[2018-07-21 09:11:04] 迫真 正版 >>> 我还是1440x900
[2018-07-21 09:11:06] Mingcong Bai >>> (一看就不是个天天用 IRC 的
[2018-07-21 09:11:10] OriginCode [Re: 223] >>> xmdys
[2018-07-21 09:11:11] 猫晴 弥柚 >>> 压得跟手机端一样的工作区域(方便干别的 [photo]
[2018-07-21 09:11:18] 迫真 正版 [Re: 236] >>> 我每次登录都去开vacation
[2018-07-21 09:11:19] OriginCode [Re: 236] >>> 曾经是(
[2018-07-21 09:11:19] Mingcong Bai [Re: 238] >>> 其实我一般情况下就这样
[2018-07-21 09:11:24] 迫真 正版 >>> 懂吗(x
[2018-07-21 09:11:25] liushuyu [Re: 233] >>> 因为不支持 ttc(
[2018-07-21 09:11:35] Mingcong Bai [Re: 242] >>> ?
[2018-07-21 09:11:37] Mingcong Bai >>> 不懂
[2018-07-21 09:11:38] cth451 ǝɔuǝƃɹǝʌᴉp‾ɥʇɔ [Re: 223] >>> 一个 3840 x 1080 用户的故事 [photo]
[2018-07-21 09:11:42] OriginCode [Re: 241] >>> i3 排成这样(x
[2018-07-21 09:11:43] 迫真 正版 [Re: 244] >>> 几个月上一次
[2018-07-21 09:11:45] Mingcong Bai [Re: 246] >>> Woohoo 32:9
[2018-07-21 09:11:48] liushuyu >>> 代码里面是自动选择 ttc 里面第一个字体
[2018-07-21 09:11:52] OriginCode [Re: 246] >>> xmsl
[2018-07-21 09:11:54] 死 肥 宅 鹹魚 [Re: 246] >>> #RICH
[2018-07-21 09:11:55] 猫晴 弥柚 >>> 羡慕高分屏
[2018-07-21 09:11:56] Mingcong Bai >>> 怀疑 @cth451 有三个脑子
[2018-07-21 09:12:00] 死 肥 宅 鹹魚 >>> 羨慕死了.webp
[2018-07-21 09:12:08] 死 肥 宅 鹹魚 >>> 三個腦子hhhhhh
[2018-07-21 09:12:08] OriginCode >>> 等我回家也是 3840*1080(
[2018-07-21 09:12:14] Mingcong Bai >>> 说起来我们来 po 一下面前的工作区域吧(
[2018-07-21 09:12:16] 死 肥 宅 鹹魚 [Re: 257] >>> #RICH 真好
[2018-07-21 09:12:16] liushuyu [Re: 254] >>> CTH Brain Patch rev. 3.0 (
[2018-07-21 09:12:19] 迫真 正版 >>> 有三个脑子也打不了osu std模式
[2018-07-21 09:12:24] cth451 ǝɔuǝƃɹǝʌᴉp‾ɥʇɔ [Re: 254] >>> 只有一个
[2018-07-21 09:12:38] Lion [Re: 262] >>> 一个一个一个
[2018-07-21 09:12:38] liushuyu [Re: 262] >>> 1核 3 线程?
[2018-07-21 09:12:42] 迫真 正版 [Re: 260] >>> 大脑升级
[2018-07-21 09:12:43] Mingcong Bai >>> [photo]
[2018-07-21 09:12:45] OriginCode [Re: 259] >>> 从家里堆的一堆东西里面翻出来了个戴尔的全新显示器
[2018-07-21 09:12:45] 死 肥 宅 鹹魚 [Re: 258] >>> P50 * 1,沒了(
[2018-07-21 09:12:54] 死 肥 宅 鹹魚 [Re: 267] >>> 羨慕了
[2018-07-21 09:12:55] Mingcong Bai [Re: 268] >>> 不不来照片(
[2018-07-21 09:13:01] Mingcong Bai [Re: 266] >>> ^
[2018-07-21 09:13:07] liushuyu [Re: 265] >>> 如何获取更新(
[2018-07-21 09:13:10] 迫真 正版 >>> [photo]
[2018-07-21 09:13:13] OriginCode [Re: 258] >>> HUAWEI P10 Plus(Bootloader Unlocked)没了(
[2018-07-21 09:13:21] Mingcong Bai [Re: 272] >>> 想起 Borat 里面一个故事
[2018-07-21 09:13:31] 迫真 正版 [Re: 273] >>> 特首请不要盯着贴纸看
[2018-07-21 09:13:35] 迫真 正版 >>> 🌚
[2018-07-21 09:13:45] Luke Yue >>> [photo]
[2018-07-21 09:13:47] Luke Yue >>> 工作区域
[2018-07-21 09:13:48] Luke Yue >>> [document]
[2018-07-21 09:13:59] Mingcong Bai [Re: 272] >>> "My brother he got headache, my sister opened his head, and put some tears in it... He now retard."
[2018-07-21 09:14:02] 迫真 正版 >>> 特首有权删sticker吗(小声
[2018-07-21 09:14:14] Mingcong Bai [Re: 282] >>> Of course - I dictator
[2018-07-21 09:14:18] Luke Yue >>> 忘了(
[2018-07-21 09:14:18] liushuyu [Re: 281] >>> 咦?(
[2018-07-21 09:14:29] cth451 ǝɔuǝƃɹǝʌᴉp‾ɥʇɔ >>> [photo]
[2018-07-21 09:14:31] Mingcong Bai [Re: 285] >>> 网搜 Borat
[2018-07-21 09:14:41] 迫真 正版 [Re: 283] >>> Heil
[2018-07-21 09:14:45] 迫真 正版 >>> (不
[2018-07-21 09:14:50] Mingcong Bai >>> 说起来有人想做 admin 吗
[2018-07-21 09:14:55] Mingcong Bai >>> 或者说“愿意”
[2018-07-21 09:15:00] 迫真 正版 >>> 用户名admin
密码admin
[2018-07-21 09:15:04] 迫真 正版 >>> (无端联想
[2018-07-21 09:15:14] Zero King [Re: 290] >>> 职责?
[2018-07-21 09:15:35] Mingcong Bai [Re: 294] >>> 看第一条消息并保障消息符合上面的规定
[2018-07-21 09:15:37] 猫晴 弥柚 >>> 对了我黄油离0.1版不远了(好像一个月前也这么说 [photo]
[2018-07-21 09:15:43] OriginCode [Re: 290] >>> 职责(
[2018-07-21 09:15:48] Mingcong Bai [Fwd: Mingcong Bai] >>> 看第一条消息并保障消息符合上面的规定
[2018-07-21 09:15:56] OriginCode [Re: 298] >>> 了解(
[2018-07-21 09:15:57] Mingcong Bai >>> 看见 sticker 删,看见政治内容删
[2018-07-21 09:16:01] Mingcong Bai >>> 然后提醒那个人
[2018-07-21 09:16:14] Mingcong Bai >>> 除此之外的话,我们这个群是公开群
[2018-07-21 09:16:30] OriginCode >>> 像咱这种不能实时在会场的可以吗?
[2018-07-21 09:16:34] Mingcong Bai >>> 所以如果有 сука блять 进来就删
[2018-07-21 09:16:37] Mingcong Bai [Re: 303] >>> 可以
[2018-07-21 09:16:42] 迫真 正版 [Re: 304] >>> 哈哈哈哈
[2018-07-21 09:16:45] 迫真 正版 >>> 正确用途(不是
[2018-07-21 09:16:56] 迫真 正版 >>> 字面意义上的porn spam
[2018-07-21 09:17:06] Mingcong Bai >>> @OriginCode Promoted as admin
[2018-07-21 09:17:15] OriginCode [Re: 305] >>> 那咱要一个吧(伸手
[2018-07-21 09:17:21] Mingcong Bai [Re: 310] >>> Preempted (
[2018-07-21 09:17:35] OriginCode >>> 啊美国的漫游好慢(
[2018-07-21 09:17:35] Zero King [Re: 295] >>> IRC admin呢?
[2018-07-21 09:17:47] 迫真 正版 >>> [photo]
[2018-07-21 09:17:51] Mingcong Bai [Re: 313] >>> 是 @Icenowy ,但是她现在说太累了,所以今早我先顶着
[2018-07-21 09:17:52] OriginCode [Re: 313] >>> IRC 没法删信息啊
[2018-07-21 09:17:52] 迫真 正版 >>> 泡一壶有助睡眠
[2018-07-21 09:18:03] Mingcong Bai [Re: 316] >>> 踢人和设定 topic
[2018-07-21 09:18:09] Mingcong Bai >>> 所以活其实不多
[2018-07-21 09:18:11] 迫真 正版 >>> 可以警告三次然后开机票
[2018-07-21 09:18:15] OriginCode >>> 嗯
[2018-07-21 09:18:15] 迫真 正版 >>> 😁
[2018-07-21 09:18:32] Zero King [Re: 318] >>> 给我个op用来kick
[2018-07-21 09:18:42] Mingcong Bai [Re: 323] >>> 我看看我能不能做……
[2018-07-21 09:18:59] OriginCode [Re: 324] >>> chanserv 能做(
[2018-07-21 09:19:02] Mingcong Bai >>> 你的 IRC id 是什么
[2018-07-21 09:19:04] Mingcong Bai >>> @l2dy0
[2018-07-21 09:19:25] Mingcong Bai >>> l2dy?
[2018-07-21 09:19:31] Zero King >>> Yes.
[2018-07-21 09:19:55] 死 肥 宅 鹹魚 [Re: 270] >>> 吸毒 [photo]
[2018-07-21 09:20:16] Mingcong Bai [Re: 329] >>> 没有 mode 命令是什么情况
[2018-07-21 09:20:42] 迫真 正版 [Re: 330] >>> 我现在都有点不习惯镜面“自拍”屏了
[2018-07-21 09:20:46] Zero King [Re: 331] >>> 要看客户端吧
[2018-07-21 09:21:04] OriginCode [Re: 330] >>> 今天咱喝了一肚子水了
[2018-07-21 09:21:07] 死 肥 宅 鹹魚 [Re: 332] >>> 我這個算鏡面麼...
[2018-07-21 09:21:22] OriginCode [Re: 331] >>> /msg chanserv mode?
[2018-07-21 09:21:32] 死 肥 宅 鹹魚 >>> 這屏幕是屑(
[2018-07-21 09:21:38] OriginCode >>> 特首有管理员权限么(
[2018-07-21 09:21:51] Mingcong Bai [Re: 338] >>> 我目前 +o
[2018-07-21 09:21:59] 死 肥 宅 鹹魚 >>> 亮度不行色域不行還老反光(
[2018-07-21 09:22:00] OriginCode >>> 了解
[2018-07-21 09:22:07] 迫真 正版 [Re: 335] >>> 话说你这个是触屏吗(
[2018-07-21 09:22:12] 迫真 正版 >>> 好像还是电磁屏来着?
[2018-07-21 09:22:16] Mingcong Bai [Re: 341] >>> 看起来我没权限
[2018-07-21 09:22:17] 死 肥 宅 鹹魚 [Re: 342] >>> 是,還能用筆
[2018-07-21 09:22:18] Mingcong Bai >>> @lilyayta
[2018-07-21 09:22:33] OriginCode [Re: 345] >>> 羡慕(
[2018-07-21 09:22:34] Mingcong Bai >>> P50 真是好文明
[2018-07-21 09:22:35] 井生 羽衣 >>> [photo]
[2018-07-21 09:22:46] 迫真 正版 [Re: 348] >>> 电池(呕
[2018-07-21 09:23:00] Mingcong Bai [Re: 350] >>> True...
[2018-07-21 09:23:02] 迫真 正版 >>> 真的是T410s既视感
[2018-07-21 09:23:06] OriginCode [Re: 345] >>> 有 GNOME 体验吗?
[2018-07-21 09:23:07] liushuyu >>> 现在这个群非常考验你的大脑的任务调度能力(
[2018-07-21 09:23:21] 迫真 正版 [Re: 354] >>> 装个EXEC内核就可以了
[2018-07-21 09:23:22] 无聊怪 [Re: 349] >>> 好大一块下巴
[2018-07-21 09:23:23] Mingcong Bai [Re: 349] >>> MATE 用户这里有没有(
[2018-07-21 09:23:26] 无聊怪 >>> 😂
[2018-07-21 09:23:30] liushuyu [Re: 357] >>> 我!(
[2018-07-21 09:23:33] OriginCode [Re: 354] >>> Overflowed
[2018-07-21 09:23:36] 迫真 正版 >>> KDE用户这里有没有
[2018-07-21 09:23:38] 死 肥 宅 鹹魚 [Re: 353] >>> Hail KDE
[2018-07-21 09:23:44] liushuyu [Re: 361] >>> 有(
[2018-07-21 09:23:45] Mingcong Bai >>> Hail KDE (
[2018-07-21 09:23:51] 迫真 正版 [Fwd: Mingcong Bai] >>> Hail KDE (
[2018-07-21 09:23:56] 迫真 正版 >>> 自从换了KDE后
[2018-07-21 09:24:00] liushuyu >>> 复读机时间?
[2018-07-21 09:24:01] 井生 羽衣 [Re: 357] >>> xfce(
[2018-07-21 09:24:03] 迫真 正版 >>> ……算了广告不打了
[2018-07-21 09:24:05] 死 肥 宅 鹹魚 [Re: 363] >>> 你到底用啥(
[2018-07-21 09:24:12] 无聊怪 >>> /me MATE / KDE 用户
[2018-07-21 09:24:15] OriginCode [Re: 361] >>> AOSC OS GNOME,Arch Linux KDE
[2018-07-21 09:24:16] liushuyu [Re: 370] >>> 两台设备么
[2018-07-21 09:24:18] 死 肥 宅 鹹魚 [Re: 366] >>> 腰不疼了腿不酸了
[2018-07-21 09:24:22] Mingcong Bai [Re: 371] >>> This, is a wise man
[2018-07-21 09:24:31] 猫晴 弥柚 [Re: 361] >>> 有啊
[2018-07-21 09:24:43] liushuyu [Re: 374] >>> 头不疼了,就连心脏也不跳了(
[2018-07-21 09:24:54] Mingcong Bai [Re: 377] >>> 内核也 tickless 了
[2018-07-21 09:25:01] 死 肥 宅 鹹魚 [Re: 377] >>> Walking dead
[2018-07-21 09:25:13] Mingcong Bai >>> 下个议程还有 5 分钟(
[2018-07-21 09:25:21] OriginCode >>> Kernel Panic(
[2018-07-21 09:25:24] Lion [Re: 361] >>> 我呀😽
[2018-07-21 09:25:33] Mingcong Bai >>> 问个问题,我这里准备的资料是一大段的,主要是提了四五个建议
[2018-07-21 09:25:40] OriginCode >>> 凉透了
[2018-07-21 09:25:46] Mingcong Bai >>> 你们希望我一下全发出来还是一个一个来讨论?
[2018-07-21 09:26:22] 猫晴 弥柚 >>> 一下子发出来,然后一个一个讨论(
[2018-07-21 09:26:23] OriginCode >>> 一个一个讨论?
[2018-07-21 09:26:32] 迫真 正版 >>> 特首趁此机会练练打字吧(x
[2018-07-21 09:26:33] OriginCode [Fwd: 猫晴 弥柚] >>> 一下子发出来,然后一个一个讨论(
[2018-07-21 09:26:36] Mingcong Bai [Re: 386] >>> 啊这是前者
[2018-07-21 09:26:56] 死 肥 宅 鹹魚 [Re: 388] >>> 人肉 OCR(x
[2018-07-21 09:27:03] Mingcong Bai [Re: 388] >>> 啥(
[2018-07-21 09:27:12] 迫真 正版 [Re: 392] >>> 抄一遍(x
[2018-07-21 09:27:17] OriginCode [Re: 388] >>> 特首应该打好了(
[2018-07-21 09:27:28] liushuyu >>> 还有三分钟
[2018-07-21 09:27:28] Mingcong Bai [Re: 394] >>> 嗯我打好了(
[2018-07-21 09:27:36] OriginCode >>> 特首可以使用古老的手抄(
[2018-07-21 09:27:43] 迫真 正版 >>> “抄”送到tg文本框
[2018-07-21 09:27:47] 猫晴 弥柚 >>> 然后扫图上传
[2018-07-21 09:27:52] Mingcong Bai >>> @liushuyu You've gotta fix that... [photo]
[2018-07-21 09:27:58] Mingcong Bai [Re: 398] >>> hhhh
[2018-07-21 09:28:01] Mingcong Bai >>> 还没那么快(
[2018-07-21 09:28:11] OriginCode >>> 特首字如何?
[2018-07-21 09:28:33] Mingcong Bai [Re: 403] >>> 英文和俄语是连笔
[2018-07-21 09:28:36] Mingcong Bai >>> 中文一般
[2018-07-21 09:28:37] 迫真 正版 >>> 当成画画就行了(x
[2018-07-21 09:28:37] Zero King [Re: 385] >>> 全发的话pin吗?
[2018-07-21 09:28:46] liushuyu [Re: 400] >>> 善
[2018-07-21 09:28:49] Mingcong Bai [Re: 407] >>> 可以
[2018-07-21 09:28:49] 猫晴 弥柚 >>> 俄文手写体还是算了吧
[2018-07-21 09:28:53] OriginCode [Re: 405] >>> 求图(逃
[2018-07-21 09:28:54] 猫晴 弥柚 >>> 太恐怖了
[2018-07-21 09:29:01] 迫真 正版 >>> 我都没听说过pluma(
[2018-07-21 09:29:07] OriginCode [Fwd: 迫真 正版] >>> 我都没听说过pluma(
[2018-07-21 09:29:20] 迫真 正版 >>> 原来是mate的
[2018-07-21 09:29:23] 迫真 正版 >>> 行吧
[2018-07-21 09:29:25] 井生 羽衣 [Re: 389] >>> +1
[2018-07-21 09:29:28] 死 肥 宅 鹹魚 >>> 要對一個果凍進行處決(x [photo]
[2018-07-21 09:29:42] OriginCode >>> [document]
[2018-07-21 09:29:52] OriginCode >>> emm
[2018-07-21 09:29:55] OriginCode >>> 等等(
[2018-07-21 09:29:55] Mingcong Bai >>> @OriginCode 你这个狗管理怎么(
[2018-07-21 09:29:56] 迫真 正版 >>> 我现在在Windows下用geany了
[2018-07-21 09:29:58] 井生 羽衣 >>> 啊忘了買果凍了
[2018-07-21 09:29:59] OriginCode >>> 手贱了(
[2018-07-21 09:30:10] 迫真 正版 >>> 特首:反了
[2018-07-21 09:30:13] OriginCode [Re: 422] >>> 抱歉(
[2018-07-21 09:30:13] Mingcong Bai >>> [photo]
[2018-07-21 09:30:19] Mingcong Bai >>> Shall we begin?
[2018-07-21 09:30:20] 猫晴 弥柚 [Fwd: 井生 羽衣] >>> 啊忘了買果凍了
[2018-07-21 09:30:32] 井生 羽衣 >>> Ready
[2018-07-21 09:30:34] OriginCode [Re: 429] >>> y
[2018-07-21 09:30:36] Mingcong Bai >>> 那我全发出来了
[2018-07-21 09:31:01] Mingcong Bai >>> #TOPIC 9:30 - 10:30 — 调整 AOSC OS 的更新周期和里程碑计划集成
去年这个时候我提出让 AOSC OS 转入半滚动更新模型,提供 testing 和 master/bugfix 两个更新分支,并且采用一个基于月度的更新日程。很明显,目前虽然这个模型本身保持住了,但是“月度”更新的目标却远远没有达到(上次更新是四月底),而且这个时间上的拖延还导致了一些依赖更大范围更新的 bug 修复没能及时推送给非 testing 用户。最典型的例子是 Qt 和 GLVND 的 bug,导致 Plasma Desktop 和 Telegram Desktop 部分透明效果不可用,这个问题是五月底提出的,目前的计划是在 25 日推送的更新中修复。
简单总结就是下面几点:
— 我们的人力不允许月度更新。这不只是个“如何把握更新量”的问题,而是我们作为业余维护者时常能遇到的时间安排问题(比如我本人目前因为学业和校内工作导致的时间和精力不足)导致很多时候计划虽然能提出来,却很难能保证准时完成。
— 每个周期中缺乏一系列类似“里程碑”的大目标和主题,绝大多数工作都放到了追赶更新上而(和前两年一样)缺乏对现有发行版的修缮。
— 稳定和 testing 分支缺乏交互,后者提供的修复不能及时集成到稳定分支——而目前的时间周期问题导致了稳定用户并不能享受到太多的实际稳定性/可用性。
于是提出下面几点建议:
— 将更新周期拉长到一个季度,并且预留最后一个月用于测试和处理架构之间同步。
— 进一步扩大更新例外列表( https://wiki.aosc.io/developers/aosc-os/monthly-exceptions )的范围,尤其是 patch release 级别和 LTS 系列的更新(比如 kde-workspace 从 5.13.3 更新到 5.13.4)。
— 在每个周期中计划好软件更新的目标分支(KDE Plasma 大约每个季度会发布一个 LTS 分支,可以考虑让我们的更新日程结尾对齐到这种时间点,于是稳定用户可以跟着这个分支获取更新。
— 使用文档/Wiki 记录我们的更新周期日程和上述的判定过程(可以直接将现在的 Exceptions 页扩展为这个内容)。 [webpage]
[2018-07-21 09:31:05] Mingcong Bai [Re: 434] >>> pinned the message
[2018-07-21 09:31:35] Mingcong Bai >>> 总体来说这是我的设想,里面举了几个近两个月发现的问题,看看大家怎么想……
[2018-07-21 09:32:08] OriginCode >>> Telegram 的透明问题貌似现在还有?
[2018-07-21 09:32:18] Mingcong Bai [Re: 437] >>> 是的,这是目前 testing 里面才解决了的问题
[2018-07-21 09:32:22] Mingcong Bai >>> 通过更新 Qt
[2018-07-21 09:32:45] Mingcong Bai >>> 实际上如果 bugfix 里面推送了 Qt 5.10 系列的 patch release 也许能解决问题
[2018-07-21 09:32:49] OriginCode [Re: 438] >>> 了解(
[2018-07-21 09:32:56] Mingcong Bai >>> 就是这方面的做法还没有具体定义过
[2018-07-21 09:34:09] OriginCode >>> 以及还有些 Stable 源里面的打包失误(
[2018-07-21 09:34:21] 迫真 正版 >>> 季度还是不错的
[2018-07-21 09:34:35] Mingcong Bai [Re: 443] >>> 是的,我感觉目前 stable 出来之后维护是不够的
[2018-07-21 09:34:48] Mingcong Bai >>> 一个是用户完全得不到除了安全更新之外的任何更新
[2018-07-21 09:34:55] 死 肥 宅 鹹魚 >>> 資辭
[2018-07-21 09:35:08] Mingcong Bai >>> 一个是他们遇到问题修复起来也慢,或者像上面说的那样拖延很久
[2018-07-21 09:35:21] OriginCode >>> 嗯
[2018-07-21 09:35:54] Zero King >>> 那么我们应该偏好使用LTS版本?例如把node降到8。
[2018-07-21 09:36:04] Mingcong Bai [Re: 450] >>> 是的,这点我得请教下……
[2018-07-21 09:36:07] 猫晴 弥柚 >>> 我对‘更新’的过程不是很理解,无法实现自动化反馈的方案吗
[2018-07-21 09:36:26] Mingcong Bai >>> 我上面提到,如果我们用季度更新,那么是否应该在每个季度周期前计划好对齐到什么版本?
[2018-07-21 09:36:44] Mingcong Bai >>> “在每个周期中计划好软件更新的目标分支(KDE Plasma 大约每个季度会发布一个 LTS 分支,可以考虑让我们的更新日程结尾对齐到这种时间点,于是稳定用户可以跟着这个分支获取更新。”
[2018-07-21 09:36:50] Mingcong Bai [Re: 452] >>> 能稍微解释下吗?
[2018-07-21 09:36:54] OriginCode >>> 最近 LTS 版本?
[2018-07-21 09:37:09] Mingcong Bai [Re: 456] >>> 嗯比如我假设三个月后 Plasma 要发布 5.15,是个 LTS
[2018-07-21 09:37:15] Mingcong Bai >>> 那么我们应该努力对齐那个版本分支
[2018-07-21 09:37:24] OriginCode >>> 对
[2018-07-21 09:37:29] Mingcong Bai >>> 然后 bugfix 在接下来一个周期都会得到 5.15.x 的更新
[2018-07-21 09:37:32] Zero King >>> +1
[2018-07-21 09:37:38] liushuyu [Re: 454] >>> 如果某个软件提供 mainline 和 LTS 我觉得需要看情况
[2018-07-21 09:37:38] Lion [Re: 452] >>> 下一节我的部分会有详细描述。目前主要靠人力、GitHub 项目半自动、repology 网站辅助,等。
[2018-07-21 09:37:44] 死 肥 宅 鹹魚 >>> 資辭(
[2018-07-21 09:37:46] liushuyu >>> 不一定要用 LTS
[2018-07-21 09:37:54] Mingcong Bai [Re: 465] >>> 这怎么说?
[2018-07-21 09:38:03] OriginCode [Re: 458] >>> 但是不同组件的 LTS 版本释出周期不同如何解决?
[2018-07-21 09:38:09] liushuyu [Re: 466] >>> 比如 Java
[2018-07-21 09:38:10] Mingcong Bai [Re: 467] >>> 这确实是个问题
[2018-07-21 09:38:24] Mingcong Bai >>> 大的项目一般都会有个固定的 LTS 周期
[2018-07-21 09:38:25] liushuyu >>> 还有 LMMS 之流
[2018-07-21 09:38:40] liushuyu >>> LTS 远比 mainline 老的软件
[2018-07-21 09:38:43] Mingcong Bai >>> 而且我有点怀疑这么做的话很可能会制约 testing 方面的更新(因为不敢跳到下个分支
[2018-07-21 09:39:06] 无聊怪 >>> 那就需要分成两个分支了
[2018-07-21 09:39:08] Mingcong Bai [Re: 472] >>> 特例肯定是有的
[2018-07-21 09:39:15] Mingcong Bai [Re: 474] >>> 嗯?
[2018-07-21 09:39:19] 无聊怪 >>> stable 提供 lts 软件包
[2018-07-21 09:39:33] Zamir SUN >>> 我赞成季度(甚至半年)一个周期
[2018-07-21 09:39:36] 无聊怪 >>> 另一个分支提供 mainline 的
[2018-07-21 09:39:38] OriginCode >>> 干脆变成 AOSC OS LTS (
[2018-07-21 09:39:51] Lion >>> “这不只是个如何把握更新量的问题,而是我们作为业余维护者时常遇到的时间安排问题导致很多时候计划虽然能提出来,却很难保证准时完成” 这一部分我感觉有点争议
[2018-07-21 09:39:56] Uncle Lin >>> [chat_add_user]
[2018-07-21 09:39:57] Sakamoto Tanmy >>> 插嘴:没有直播不开心(
[2018-07-21 09:39:58] Mingcong Bai [Re: 479] >>> LTS, Mainline, Testing?
[2018-07-21 09:40:02] 死 肥 宅 鹹魚 >>> AOSC OS Leap, AOSC OS Tumbleweed(x
[2018-07-21 09:40:03] 井生 羽衣 >>> 以及萬一某個軟件的 LTS 版本咕了(推遲發佈了)怎麼辦
[2018-07-21 09:40:04] Mingcong Bai [Re: 483] >>> 哪来的直播
[2018-07-21 09:40:12] OriginCode [Re: 485] >>> Nice(
[2018-07-21 09:40:27] 无聊怪 [Re: 484] >>> 其实我也不太清楚 Testing 跟 mainline 怎么处理比较好
[2018-07-21 09:40:32] OriginCode >>> 还有已经消失的 Factory
[2018-07-21 09:40:34] Mingcong Bai [Re: 484] >>> 我对这种设定原则上没有什么反对,但是…… 工作量会很大
[2018-07-21 09:40:36] Zero King >>> 我提到node主要是node的安全更新如果是非LTS就很可能要升minor release了。
[2018-07-21 09:40:42] Zamir SUN [Re: 478] >>> 但觉得 bugfix 应该尽量及时
[2018-07-21 09:40:45] Mingcong Bai [Re: 492] >>> 嗯这个是
[2018-07-21 09:40:50] 无聊怪 >>> 现在的想法是 testing 是用于测试 lts 软件包的
[2018-07-21 09:41:02] Mingcong Bai [Re: 493] >>> 我觉得现在 bugfix 是一个做好就不更新的分支
[2018-07-21 09:41:10] OriginCode [Re: 495] >>> 这样 testing 太滞后了吧(
[2018-07-21 09:41:14] Mingcong Bai >>> 我们应该更积极维护
[2018-07-21 09:41:14] 无聊怪 >>> mainline 则是现在的 testing
[2018-07-21 09:41:34] liushuyu [Re: 477] >>> 这个我觉得可以,还有 mainline 和 LTS 不兼容的问题,Java 11 有 arm 的 hs 但是不完全向下兼容 Java 8
[2018-07-21 09:41:36] Mingcong Bai [Re: 481] >>> 具体说说?
[2018-07-21 09:41:46] Lion [Re: 481] >>> 我觉得最直接的问题是我们没有把冻结真正做到位。一个仅仅和时间挂钩的周期,应该是尽可能无视上游更新的量,而仅仅在每个月有限的时间内做能做的事情。
[2018-07-21 09:42:01] Mingcong Bai [Re: 502] >>> 嗯这个我上面提到了
[2018-07-21 09:42:03] Zamir SUN >>> 我不是指 bugfix分支,而是说,对于 bug 的修复(而不是更新版本)要仍然有保障(例如一个月。?)
[2018-07-21 09:42:11] Zero King [Re: 500] >>> Java可以打多个包
[2018-07-21 09:42:20] Lion >>> 换句话说,“以清空更新” 为目的,不可能真正落实周期制度
[2018-07-21 09:42:30] liushuyu [Re: 505] >>> 对,就是这么想的
[2018-07-21 09:42:34] Zamir SUN [Re: 502] >>> 赞同狮子
[2018-07-21 09:42:40] Mingcong Bai [Re: 502] >>> 我这个观点主要说的是周期太短很容易由于时间安排上的突发事件会造成问题
[2018-07-21 09:42:50] Lion >>> 要划定时间的强制限制,允许堆积未完成的任务
[2018-07-21 09:43:09] Mingcong Bai >>> 因为如果一直赶着要做完你是不可能上周期的
[2018-07-21 09:43:15] Lion [Re: 509] >>> 哪怕有这些突发的出入,我觉得也是可以接受的
[2018-07-21 09:43:33] Mingcong Bai [Re: 512] >>> 我假设一下
[2018-07-21 09:43:38] liushuyu >>> 所以需要 stage 变更
[2018-07-21 09:43:40] OriginCode >>> 这样?:
KDE LTS 组件更新 - Bugfix... - GNOME LTS 组件更新 - Bugfix
[2018-07-21 09:43:48] Mingcong Bai >>> 我从现状说,我和 @colin4124 两个人是主要的打包者
[2018-07-21 09:44:03] Mingcong Bai >>> 如果我在月初要准备 paper,两周没法干太多事情
[2018-07-21 09:44:10] Mingcong Bai >>> @colin4124 刚好又身体不舒服(假设
[2018-07-21 09:44:11] liushuyu >>> 比方说 staging 在 freeze 之后,如果有东西需要更新
[2018-07-21 09:44:17] Mingcong Bai >>> 那么我们实际上剩下的时间也不多了
[2018-07-21 09:44:30] Mingcong Bai >>> (当然这和车祸指数也有关系
[2018-07-21 09:44:35] Lion [Re: 520] >>> 我觉得这种“闲月”是可以接受的
[2018-07-21 09:44:41] Mingcong Bai [Re: 522] >>> 嗯?
[2018-07-21 09:45:04] Mingcong Bai [Re: 514] >>> 我还是担心三层更新会造成工作量飙升
[2018-07-21 09:45:17] Lion >>> 相反,如果因为车祸指数太低了,导致时间波动而一直放大时间的弹性,这个周期就被破坏了
[2018-07-21 09:45:20] Mingcong Bai >>> 这样的话实际上需要三份系统同时测试
[2018-07-21 09:45:43] Mingcong Bai [Re: 525] >>> 车祸指数在这里说的是主要维护者/技能比较完全者的数量之少
[2018-07-21 09:45:48] Lion >>> 我觉得周期不能动才是关键,哪怕是每月或是每两月
[2018-07-21 09:46:24] Lion [Re: 527] >>> y,我知道。我指的是人少所以每个人有自己的事情的概率会很大程度影响到最终效果
[2018-07-21 09:46:44] OriginCode >>> 要不要去宣传一波 AOSC 招点人力?(逃
[2018-07-21 09:46:55] Mingcong Bai [Re: 530] >>> 我们先暂时不讨论这个问题……
[2018-07-21 09:47:25] Lion [Re: 529] >>> 噢,不是概率,是波动
[2018-07-21 09:48:08] Mingcong Bai >>> 定个具体时间固然是好,但是我们需要允许测试时间不足而导致的跳票情况?
[2018-07-21 09:48:57] Lion [Re: 533] >>> 是,这本来就是周期所暗示的结果
[2018-07-21 09:49:49] Lion >>> “很不幸,这个xx包这周期不能进了,但其他已经打好的包我们将要 ship 了。这个包下周期进。”
[2018-07-21 09:50:58] Lion >>> 如果用 “不跳票” 来限制我们自己,相反拖延了其他的包,让一堆的包跟着拖延…我感觉某种意义上这才是大规模的跳票吧
[2018-07-21 09:51:19] Mingcong Bai >>> 嗯
[2018-07-21 09:51:30] Mingcong Bai >>> 我先整理下上面我们集到的点子
[2018-07-21 09:53:27] Mingcong Bai >>> — 总体赞成延长周期到季度
— 固定每个周期的截止时间并且尽量地去保证准时
— 考虑增加一个新的测试级别(LTS,长期稳定,仅接受 LTS 和 patch release 更新;Testing,测试 LTS 分支的更新;Mainline,对应现在的 staging 分支);那么,要如何去计划这方面的工作,如何去符合季度周期? cc @sakiiily
[2018-07-21 09:53:36] Mingcong Bai >>> 上述几点是否准确?
[2018-07-21 09:54:54] 无聊怪 >>> 对
[2018-07-21 09:55:03] Mingcong Bai >>> @Lionium
[2018-07-21 09:55:37] Mingcong Bai >>> 那么假设我们需要这么做,可以考虑让 Mainline 在三个月内自由追赶更新,Testing 做 LTS 分支服务,而 LTS 只接受 Merge?
[2018-07-21 09:55:37] Lion >>> 第二点总结得没问题
[2018-07-21 09:56:11] Mingcong Bai >>> 那么也就是说 Mainline 达到新的分支之前,Testing 分支去跟 patch release
[2018-07-21 09:56:17] Mingcong Bai >>> 而 LTS 接受 Testing 测试过的包
[2018-07-21 09:56:33] Mingcong Bai >>> ……那么 Mainline 还有没有必要接受季度周期的制约?
[2018-07-21 09:57:06] 井生 羽衣 >>> 第三點我有些沒聽明白……就是說 Mainline 是否還專門測試?
[2018-07-21 09:57:22] Mingcong Bai [Re: 548] >>> Mainline 是个无主之地
[2018-07-21 09:57:47] Mingcong Bai >>> Testing 接受 Mainline 收到的新分支更新
[2018-07-21 09:58:00] Lion >>> [document]
[2018-07-21 09:58:00] Mingcong Bai >>> 但是具体怎么处理这两个分支里面的 patch release 更新,我没什么头绪……
[2018-07-21 09:58:30] Mingcong Bai >>> 我尝试画个图
[2018-07-21 09:59:07] 井生 羽衣 >>> 啊我沒看到 mainline 對應當前的 staging
[2018-07-21 09:59:18] 井生 羽衣 >>> 這樣就理解了
[2018-07-21 09:59:21] 死 肥 宅 鹹魚 >>> 還有個相關的問題,源上面的目錄結構能否做到像 Debian 那樣在鏈接結尾加上類別名稱就能添加對應類別
比如 https://mirrors.blablabla.tld/debian main contrib 什麼的這樣
[2018-07-21 09:59:44] 死 肥 宅 鹹魚 >>> 現在這樣要添加源相對麻煩一些
[2018-07-21 10:00:10] Mingcong Bai >>> (画不好
[2018-07-21 10:00:17] Mingcong Bai >>> 我稍微总结下
[2018-07-21 10:01:06] Lion [Re: 556] >>> 应该是可以的,之前主群有过讨论。唯一的问题是怎么过渡比较平滑
[2018-07-21 10:01:27] Lion >>> “非扁平源结构”
[2018-07-21 10:01:29] 井生 羽衣 >>> 主要是 mainline 這個名字很難讓人一下子反應過來它比 testing 更不「穩定」
[2018-07-21 10:01:39] Lion [Re: 562] >>> true
[2018-07-21 10:01:40] Mingcong Bai >>> — Mainline 是现在的 staging,不受周期制约(?),追赶一切更新
— Testing 接受 staging 的 LTS 分支更新(但是没有 LTS 定义的上游怎么处理?)
— LTS 接受测试过的 LTS 分支更新,并且接受 LTS 分支的 patch release 更新;Testing 在 LTS 接受某个分支后可以考虑继续接受 Mainline 的分支更新
[2018-07-21 10:01:53] Mingcong Bai [Re: 563] >>> 嗯,我们先看看这个逻辑能不能行得通吧
[2018-07-21 10:01:55] OriginCode [Re: 562] >>> 那 beta,alpha?
[2018-07-21 10:02:26] Mingcong Bai [Re: 564] >>> 现在的一个空洞在于 Testing 怎么判断接受哪部分的更新
[2018-07-21 10:02:57] 死 肥 宅 鹹魚 [Re: 562] >>> Unstable(學習 Debiantai
[2018-07-21 10:03:32] Lion >>> Debian 似乎有个叫做 Sid 的?
[2018-07-21 10:03:36] Zero King >>> 以及mainline不受制约怎么保证下个周期的LTS?
[2018-07-21 10:03:48] Lion >>> 有人简单介绍一下 Debian 的前沿版本之间的关系吗
[2018-07-21 10:04:10] 死 肥 宅 鹹魚 [Re: 569] >>> 也叫 Unstable(似乎
[2018-07-21 10:04:26] Mingcong Bai [Re: 570] >>> 是,不好把握
[2018-07-21 10:04:36] Mingcong Bai >>> 有 LTS 分支的上游很明显
[2018-07-21 10:04:43] Mingcong Bai >>> 但是没有的话就很难了
[2018-07-21 10:05:02] Mingcong Bai >>> 或者说我感觉只能通过人为判定
[2018-07-21 10:05:47] 死 肥 宅 鹹魚 >>> Debiantai 的命名 [photo]
[2018-07-21 10:06:31] Lion [Re: 575] >>> 我感觉会遇到很棘手的版本的问题,特别是如果我们没有相当的 backport 技术能力的话
[2018-07-21 10:06:56] 无聊怪 >>> 无 LTS 上游可以参考 debian 什么的版本号?
[2018-07-21 10:07:08] Mingcong Bai [Re: 578] >>> 这个问题会在什么情况出现?
[2018-07-21 10:07:20] Mingcong Bai >>> (类似 Qt 半透明的问题?
[2018-07-21 10:07:54] Lion >>> 我举个例子,有包 A 1-lts,有包 B 1,B 1 依赖 A 1。
[2018-07-21 10:08:34] Lion >>> 后来上游 B 更新为 B 3,我们的 LTS 线基本不动它
[2018-07-21 10:08:53] Lion >>> 但万一有一天出现安全漏洞,必须将 B 1 升级为 B 3+
[2018-07-21 10:09:19] Lion >>> 而 B 3 以后的版本已经不能再依赖 A 1-lts 了,至少要 A 2
[2018-07-21 10:09:52] Lion >>> 那么就有软件包要被 “拖出” lts 版本了
[2018-07-21 10:10:05] Mingcong Bai >>> 其实这种情况,安全更新我们不是 100% 要保证推送的……
[2018-07-21 10:10:14] Mingcong Bai >>> 比如现在的 liblouis 和 ffmpeg 安全漏洞
[2018-07-21 10:10:24] Mingcong Bai >>> 我们根本不会考虑推送到稳定
[2018-07-21 10:10:50] Mingcong Bai >>> ¹ https://github.com/AOSC-Dev/aosc-os-abbs/issues/1208
² https://github.com/AOSC-Dev/aosc-os-abbs/issues/1293 [webpage]
[2018-07-21 10:11:08] Mingcong Bai >>> cc @量
[2018-07-21 10:11:40] Lion >>> 所以我感觉 LTS 的维护,主要就是固定版本后,backport。那么如果我们又 backport 不成,就只好按需推进版本。
[2018-07-21 10:12:55] Lion >>> 当然发行版打包本身的问题,是直接修修就好了。主要是怕上游的 bug 修复很难轻巧地做进 LTS 线里。
[2018-07-21 10:13:09] Mingcong Bai >>> 我感觉这种案例其实不多
[2018-07-21 10:13:17] Mingcong Bai >>> 遇到了我们也可以集中去解决一下
[2018-07-21 10:14:47] Lion >>> 因为上面说的,lts 包评判标准不确定。那么就有可能对包版本集合的“前沿”判断不准确,会不会有什么影响。是这样的一个疑惑
[2018-07-21 10:14:59] Mingcong Bai >>> 啊你是这个意思……
[2018-07-21 10:15:16] Mingcong Bai >>> 虽然我必须说,需要掐分支的时候,这种风险基本不可能避免……
[2018-07-21 10:15:34] Mingcong Bai >>> 就算我们的知识体系很完善,比如我们知道 GNOME 3.28 是当前稳定而 3.29 是很明显不稳定的版本
[2018-07-21 10:15:50] Mingcong Bai >>> 我们依然不能避免 GNOME 后来出个漏洞但是只在 3.29 上修复
[2018-07-21 10:16:15] Mingcong Bai >>> 其实这个情况在 liblouis 这种不用 patch release(也就是 x.y.z 里面 z)更新的包最常见
[2018-07-21 10:16:20] Mingcong Bai >>> 他们每次发布新版本都会出安全问题
[2018-07-21 10:16:35] Mingcong Bai >>> 而每次安全问题是包含在一个 ABI 甚至 API 不兼容的包里面的
[2018-07-21 10:16:57] Mingcong Bai [Fwd: fpsNoooob] >>> 🌚🔫但是 testing 没人权
[2018-07-21 10:16:57] Mingcong Bai [Fwd: fpsNoooob] >>> 比如现在的 shadowsocks-libev 🤣🤣🤣
[2018-07-21 10:16:57] Mingcong Bai [Fwd: Robobo] >>> Package: shadowsocks-libev
old-bpo: 2.6.3+ds-3+deb9u1~bpo8+1
old-bpo-sl: 3.2.0+ds-3~bpo8+1
stable: 2.6.3+ds-3+deb9u1
stable-sec: 2.6.3+ds-3+deb9u1
stable-bpo: 3.2.0+ds-3~bpo9+1
unstable: 3.2.0+ds-3
[2018-07-21 10:17:01] Mingcong Bai >>> 刚好看见我就转过来了
[2018-07-21 10:17:15] Mingcong Bai >>> Debian 的分支管理很严格,代价也很重
[2018-07-21 10:17:47] Lion >>> 感觉像 RedHat 系之类的充满 LTS 味的发行版就基本靠 backport 来维护安全
[2018-07-21 10:18:30] Lion >>> Arch Gentoo 这些日滚夜滚(指更新频率高)的,直接靠跟上去最新版本来维护安全
[2018-07-21 10:18:33] Mingcong Bai >>> 对
[2018-07-21 10:18:51] Mingcong Bai >>> 我们刚好处于一个又追新又想保障分支稳定的……
[2018-07-21 10:18:53] Mingcong Bai >>> 很尴尬
[2018-07-21 10:19:27] Mingcong Bai [Re: 606] >>> 不过这个例子里面我指的是挑选分支的时候的一种后果(不过我们有 Exception 所以应该会好一点
[2018-07-21 10:20:40] 死 肥 宅 鹹魚 [Re: 610] >>> Gentoo 的穩定分支差不多是跟 Debian 一個級別的(老)
[2018-07-21 10:21:11] Mingcong Bai >>> 另外一个选择是,让 Mainline 只能跟 Mainline 而不能去跟明显的测试和不稳定分支
[2018-07-21 10:21:20] Mingcong Bai >>> 包括 Node 也只能在 LTS 之间跳
[2018-07-21 10:21:28] Mingcong Bai >>> GNOME 不能上 3.29 或者任何单数 minor
[2018-07-21 10:21:56] Mingcong Bai >>> 但是其他无明显分支项目比如 Coreutils
[2018-07-21 10:22:11] Mingcong Bai >>> 那么就根据 Testing 分支的时间需要来引入更新,然后充分测试
[2018-07-21 10:22:21] Mingcong Bai >>> LTS 分支不接受任何无明显分支界定的更新
[2018-07-21 10:22:37] Mingcong Bai >>> 只在 Testing 每季度结尾测试充分后并入到 LTS
[2018-07-21 10:23:48] Lion >>> (我们能暂时换一个准确的名字来讨论这个方案吗)
[2018-07-21 10:24:00] Mingcong Bai [Re: 623] >>> “三分支模型”?
[2018-07-21 10:24:08] Mingcong Bai >>> 还是说每个分支的名字
[2018-07-21 10:24:12] Lion >>> 比如 testing, lts-testing, lts?
[2018-07-21 10:24:41] Mingcong Bai >>> 我暂时这么叫吧,先不讨论名字:Chimera 不稳定分支,Testing 稳定测试分支,Stable 稳定分支
[2018-07-21 10:24:57] Mingcong Bai >>> Chimera 是三头六身怪兽,不详细解释了
[2018-07-21 10:26:44] Lion >>> 现在谈的是一个新的可选方案对吧……暂时用准确的贴近当前定义的名字,避免修辞吧。
[2018-07-21 10:26:49] Mingcong Bai >>> 重新整理下就是这样:
— Chimera 不接受不稳定和测试版本/分支(无明显分支界定的不能打任何 beta/rc,不接受 GNOME 3.29 这类明显的不稳定分支,Node 这类明显需要保障分支内安全更新的在 LTS 分支之间跳),但是不接受季度时间限制
— Testing 在季度周期(假设两个月更新,一个月测试)内引入新的 LTS 和稳定分支,无明显分支的也引入,但是要在第二个月结尾截止,然后一个月测试
— Stable 只在三个月内提供当前 LTS 分支和稳定分支内的 patch release 更新(比如 GNOME 3.28.1 到 3.28.3
[2018-07-21 10:27:01] Mingcong Bai >>> 大概是这么一回事
[2018-07-21 10:27:15] 井生 羽衣 >>> 所以說新的方案除了週期長度和 exception 之外,和 debian 模型最主要有哪些差別?
[2018-07-21 10:27:33] Mingcong Bai [Re: 632] >>> 具体测试方法不同
[2018-07-21 10:27:40] Mingcong Bai >>> 除此之外确实没有别的不同
[2018-07-21 10:28:01] Lion >>> 目前讨论的有多少个可选方案?
[2018-07-21 10:28:07] Mingcong Bai [Re: 630] >>> 另外暂时提出 Exception 针对 Stable 分支,基于现有的列表看
[2018-07-21 10:28:14] Mingcong Bai [Re: 635] >>> 就这一个
[2018-07-21 10:28:17] Lion >>> 原始方案扩展为季度的方案,以及这个?
[2018-07-21 10:28:19] Mingcong Bai >>> 时间周期是季度,三个分支
[2018-07-21 10:28:20] Mingcong Bai >>> 对
[2018-07-21 10:28:37] Lion [Re: 637] >>> wut,刚才说的那个那么快就采纳了吗
[2018-07-21 10:28:49] Mingcong Bai >>> 我没说采纳
[2018-07-21 10:28:57] Mingcong Bai >>> 而是目前没人提出另一个方案
[2018-07-21 10:29:13] Mingcong Bai >>> 所以我就在努力去填满现有的方案了
[2018-07-21 10:29:43] Mingcong Bai >>> 另外提醒下下个议程开始时间还有半小时,假设半小时内没有结论就用后面的空位继续讨论
[2018-07-21 10:30:35] Mingcong Bai >>> @Lionium 愿意继续跟着这个思路走吗,还是我们应该重新讨论一次?
[2018-07-21 10:30:42] cth451 ǝɔuǝƃɹǝʌᴉp‾ɥʇɔ [Re: 636] >>> +1
[2018-07-21 10:30:42] 死 肥 宅 鹹魚 >>> 我覺得海星,就是工作量怕是又增加了...
[2018-07-21 10:30:57] cth451 ǝɔuǝƃɹǝʌᴉp‾ɥʇɔ >>> 测试可能需要大规模部署 burnable vm
[2018-07-21 10:31:00] Mingcong Bai [Re: 648] >>> 工作量增加肯定是的了……
[2018-07-21 10:31:07] Mingcong Bai >>> 不过主要还是测试上
[2018-07-21 10:31:10] cth451 ǝɔuǝƃɹǝʌᴉp‾ɥʇɔ >>> (LVM 无所畏惧
[2018-07-21 10:31:12] Mingcong Bai >>> 更新软件包实际上不需要多少时间的
[2018-07-21 10:31:31] 死 肥 宅 鹹魚 >>> 人太少(
[2018-07-21 10:31:35] cth451 ǝɔuǝƃɹǝʌᴉp‾ɥʇɔ >>> 但是 nvidia 之类直接访问硬件的软件会受局限
[2018-07-21 10:31:59] Mingcong Bai [Re: 655] >>> 这种肯定需要实地测试的
[2018-07-21 10:31:59] 猿渡 美晴 >>> [chat_add_user]
[2018-07-21 10:32:29] Staph. aureus | In your GI Tract >>> SITS 有 NVIDIA 顯示卡,如果需要測試可以聯絡我
[2018-07-21 10:32:47] Lion [Re: 646] >>> 没问题,只是这个规模感觉开展起来比较艰难
[2018-07-21 10:32:57] cth451 ǝɔuǝƃɹǝʌᴉp‾ɥʇɔ [Re: 658] >>> 最佳方案是 burnable vm,测试需要的硬件的时候 pass thru
[2018-07-21 10:33:11] Mingcong Bai [Re: 659] >>> 不过依然要考虑时间上
[2018-07-21 10:33:14] cth451 ǝɔuǝƃɹǝʌᴉp‾ɥʇɔ >>> 不止是 nv
[2018-07-21 10:33:15] Lion [Re: 651] >>> 不是,不仅仅在测试上
[2018-07-21 10:33:21] Mingcong Bai >>> 做不完的事情要收手
[2018-07-21 10:33:32] Mingcong Bai >>> 这直接影响我们到底能有多新
[2018-07-21 10:33:42] Lion >>> 如果这样的话,我们的仓库拓扑可能又要变了
[2018-07-21 10:33:51] Mingcong Bai >>> 嗯,三个仓库了
[2018-07-21 10:33:55] Mingcong Bai >>> 三个分支
[2018-07-21 10:34:01] Zero King >>> 提议提高安全更新的优先级
[2018-07-21 10:34:08] Zero King >>> 尽量不跳安全更新
[2018-07-21 10:34:26] Mingcong Bai [Re: 670] >>> 以 Stable 上其他任何更新为代价去保障安全更新,是这样吗
[2018-07-21 10:34:37] Zero King [Fwd: Mingcong Bai] >>> 其实这种情况,安全更新我们不是 100% 要保证推送的……
[2018-07-21 10:34:37] Zero King [Fwd: Mingcong Bai] >>> 比如现在的 liblouis 和 ffmpeg 安全漏洞
[2018-07-21 10:34:37] Zero King [Fwd: Mingcong Bai] >>> 我们根本不会考虑推送到稳定
[2018-07-21 10:35:08] Zero King >>> 就是不能直接在bugfix上修复的
[2018-07-21 10:35:18] Zero King >>> 优先在staging修复
[2018-07-21 10:37:24] cth451 ǝɔuǝƃɹǝʌᴉp‾ɥʇɔ [Re: 689] >>> 注意
[2018-07-21 10:37:29] Mingcong Bai [Re: 686] >>> 层叠
[2018-07-21 10:37:34] Mingcong Bai >>> 版本必定有别
[2018-07-21 10:37:44] Mingcong Bai >>> Stable 也不会比 Testing 新,以此类推
[2018-07-21 10:37:45] 井生 羽衣 [Re: 689] >>> 噢
[2018-07-21 10:37:48] cth451 ǝɔuǝƃɹǝʌᴉp‾ɥʇɔ >>> 这将会导致不同 branch 有相同版本号的同一个包
[2018-07-21 10:38:16] Mingcong Bai [Re: 696] >>> 这个情况需要设定源优先级或者像现在的情况下,bump REL
[2018-07-21 10:38:16] cth451 ǝɔuǝƃɹǝʌᴉp‾ɥʇɔ >>> 需要刻意在 faster ring 上 bump rel
[2018-07-21 10:38:22] Mingcong Bai >>> 对,需要做这件事情
[2018-07-21 10:38:53] Lion [Re: 696] >>> 这个情况要避免的,解决方法是加 rel,至于自动化检测则是另外的话题了
[2018-07-21 10:39:02] Lion >>> 🌝
[2018-07-21 10:39:11] Mingcong Bai >>> 我再总结一次?
[2018-07-21 10:40:34] Leway Colin >>> 嗯
[2018-07-21 10:40:51] Staph. aureus | In your GI Tract [Re: 696] >>> 好奇下這樣會出問題嗎?
[2018-07-21 10:40:52] 井生 羽衣 [Re: 689] >>> 這麼說來這也是和 debian 模型的一個不同之處了?
[2018-07-21 10:41:27] Staph. aureus | In your GI Tract >>> 如果不同 Branch 有同一個版本的同一個包,應該包內容是一致的吧?
[2018-07-21 10:41:33] Lion >>> 唔。如果说周期工作一定要给一些包开例外的话,也应该保证最基本的一点空闲时间是绝对不允许例外的。类似于 “绝对不应期” 的数天或者一周。
[2018-07-21 10:41:53] Mingcong Bai >>> 关于分支定义:
— Chimera 不接受不稳定和测试版本/分支(无明显分支界定的不能打任何 beta/rc,不接受 GNOME 3.29 这类明显的不稳定分支,Node 这类明显需要保障分支内安全更新的在 LTS 分支之间跳),但是不接受季度时间限制
— Testing 在季度周期(假设两个月更新,一个月测试)内引入新的 LTS 和稳定分支,无明显分支的也引入,但是要在第二个月结尾截止,然后一个月测试
— Stable 只在三个月内提供当前 LTS 分支和稳定分支内的 patch release 更新(比如 GNOME 3.28.1 到 3.28.3)
安全更新和 Exception:
— 安全更新首要保证 Stable,假设无法通过任何手段保障 Stable,则向稳定性更低的方向走(Testing,Testing 不行就到 Chimera)
— Exception 更新推送到 Stable(当然,需要测试)
— 如果 Stable 版本因为安全或 Exception 比任何其他分支的版本高则需要 cherry-pick 到稳定性更低的分支,并且 bump REL 以保障其他分支的更新
[2018-07-21 10:42:09] Mingcong Bai [Re: 707] >>> 这个我们可能要开第二次讨论了
[2018-07-21 10:42:15] Mingcong Bai >>> 不过我们概念上可以确定下
[2018-07-21 10:42:33] Lion [Re: 707] >>> 这最后的一点作为周转和这些例外的紧急测试时间
[2018-07-21 10:42:35] Zero King [Re: 678] >>> 浪费太多精力就不值得了。我这个提议是针对“做不完的事情要收手”。
[2018-07-21 10:42:48] Mingcong Bai >>> 假设我们三个月的周期,两个月打包,第三个月 freeze Testing 和 Stable,则这个时间段里面不接受 Exceptions 更新
[2018-07-21 10:43:25] cth451 ǝɔuǝƃɹǝʌᴉp‾ɥʇɔ [Re: 706] >>> 不一致
[2018-07-21 10:43:30] Mingcong Bai [Re: 712] >>> 知会
[2018-07-21 10:43:40] Mingcong Bai [Re: 706] >>> 每分支重构
[2018-07-21 10:43:44] cth451 ǝɔuǝƃɹǝʌᴉp‾ɥʇɔ [Re: 706] >>> 下面可能 ld 不同的 so version.
[2018-07-21 10:43:49] Lion [Re: 706] >>> 可能不一致,但首先就不能这么打
[2018-07-21 10:44:03] Mingcong Bai >>> 我们需要保证越不稳定的分支里面版本越新
[2018-07-21 10:44:17] cth451 ǝɔuǝƃɹǝʌᴉp‾ɥʇɔ >>> 大批考古在处理不同源的行为不太明朗
[2018-07-21 10:44:19] cth451 ǝɔuǝƃɹǝʌᴉp‾ɥʇɔ >>> * dpkg
[2018-07-21 10:44:33] cth451 ǝɔuǝƃɹǝʌᴉp‾ɥʇɔ >>> 或者说是 apt
[2018-07-21 10:44:40] Lion [Re: 706] >>> 我在主群跟你说说包的各种唯一性问题吧
[2018-07-21 10:44:48] Mingcong Bai >>> 假设 Stable 里面有 Thunderbird 52.9.2,那么 cherry-pick 到另外两个分支的时候则需要让 Testing 得到 52.9.2-1 而 Chimera 得到 52.9.2-2
[2018-07-21 10:45:03] Mingcong Bai [Re: 724] >>> @cth451 正确?
[2018-07-21 10:45:04] 死 肥 宅 鹹魚 [Re: 720] >>> (大批考古 hhhh)
[2018-07-21 10:45:07] cth451 ǝɔuǝƃɹǝʌᴉp‾ɥʇɔ [Re: 725] >>> Yep
[2018-07-21 10:45:11] Mingcong Bai >>> K
[2018-07-21 10:45:21] cth451 ǝɔuǝƃɹǝʌᴉp‾ɥʇɔ >>> 否则没人知道 apt 会选择哪个
[2018-07-21 10:46:54] Mingcong Bai >>> 关于分支定义:
— Chimera 不接受不稳定和测试版本/分支(无明显分支界定的不能打任何 beta/rc,不接受 GNOME 3.29 这类明显的不稳定分支,Node 这类明显需要保障分支内安全更新的在 LTS 分支之间跳),但是不接受季度时间限制
— Testing 在季度周期(假设两个月更新,一个月测试)内引入新的 LTS 和稳定分支,无明显分支的也引入,但是要在第二个月结尾截止,然后一个月测试
— Stable 只在三个月内提供当前 LTS 分支和稳定分支内的 patch release 更新(比如 GNOME 3.28.1 到 3.28.3)
安全更新和 Exception:
— 安全更新首要保证 Stable,假设无法通过任何手段保障 Stable,则向稳定性更低的方向走(Testing,Testing 不行就到 Chimera)
— Exception 更新推送到 Stable(当然,需要测试)
— 如果 Stable 版本因为安全或 Exception 比任何其他分支的版本高则需要 cherry-pick 到稳定性更低的分支,并且 bump REL 以保障其他分支的更新(假设 Stable 里面有 Thunderbird 52.9.2,那么 cherry-pick 到另外两个分支的时候则需要让 Testing 得到 52.9.2-1 而 Chimera 得到 52.9.2-2)
周期集成:
— Chimera 不接受周期管制
— Testing 和 Stable 接受周期管制,而且有冻结期(假设第二个月后季度结余时间作为冻结期)
— 冻结期内不接受任何 Exception 类更新
— Stable 在冻结期内接受安全更新
[2018-07-21 10:47:10] Mingcong Bai >>> @cth451 @Lionium @l2dy0 请验证内容准确性
[2018-07-21 10:47:49] Mingcong Bai >>> 各位感觉这样的一个设计可否接受?
[2018-07-21 10:47:58] cth451 ǝɔuǝƃɹǝʌᴉp‾ɥʇɔ >>> OK
[2018-07-21 10:48:16] Zero King >>> Merge时还是用rebase?
[2018-07-21 10:48:25] Mingcong Bai [Re: 734] >>> Hmm 这个我们还没讨论到
[2018-07-21 10:48:30] Mingcong Bai >>> 先排除到问题之外吧
[2018-07-21 10:48:37] Mingcong Bai >>> (肯定会有第二次讨论的,不要担心
[2018-07-21 10:49:08] Mingcong Bai >>> 等 at 内人员都验证后我就先关闭这次讨论了,我会再安排个时间(现在还有空位,大概是明天
[2018-07-21 10:50:35] Zero King >>> OK,就是工作量比较大
[2018-07-21 10:50:42] Zero King >>> > bump REL 以保障其他分支的更新
[2018-07-21 10:50:54] Lion >>> Chimera 那个……措辞改一下吧
[2018-07-21 10:50:59] Mingcong Bai [Re: 741] >>> 嗯这个会做
[2018-07-21 10:51:01] Mingcong Bai >>> 先看思路吧
[2018-07-21 10:51:06] Lion >>> 把“但是”提到前面来
[2018-07-21 10:51:13] Mingcong Bai [Re: 744] >>> 哪个
[2018-07-21 10:51:49] Mingcong Bai >>> 噢我知道了,Chimera 那段?
[2018-07-21 10:52:05] Lion >>> 不是 “但是不接受季度限制”,而是 “它不受季度周期限制,但不更新标注为 rc、nightly 之类的上游版本”
[2018-07-21 10:52:34] Mingcong Bai [Re: 747] >>> 已修改
[2018-07-21 10:52:35] 井生 羽衣 >>> 安全更新 >「假设无法通过任何手段保障 Stable」建議寫明常見的原因
[2018-07-21 10:52:44] Mingcong Bai [Re: 749] >>> 好
[2018-07-21 10:52:52] Lion >>> 下面的 周期集成 一段里也是不接受→不受
[2018-07-21 10:53:36] Zero King >>> 好像没有提到安全更新在Chimera的优先级?
[2018-07-21 10:53:50] Mingcong Bai [Re: 752] >>> 那么再确定下
[2018-07-21 10:54:09] Mingcong Bai >>> 也就是说 Stable 接受安全更新之外,其他分支也需要接受安全更新而且优先完成?
[2018-07-21 10:54:30] cth451 ǝɔuǝƃɹǝʌᴉp‾ɥʇɔ [Re: 752] >>> 核心就是如果下面更了
[2018-07-21 10:54:34] cth451 ǝɔuǝƃɹǝʌᴉp‾ɥʇɔ >>> 上面也得跳
[2018-07-21 10:54:47] cth451 ǝɔuǝƃɹǝʌᴉp‾ɥʇɔ >>> 但是破坏稳定的时候
[2018-07-21 10:54:49] Mingcong Bai [Re: 756] >>> 下面更新的幅度不一定会超过上面,这个也要注意
[2018-07-21 10:54:51] cth451 ǝɔuǝƃɹǝʌᴉp‾ɥʇɔ >>> 下面可以不懂
[2018-07-21 10:54:56] cth451 ǝɔuǝƃɹǝʌᴉp‾ɥʇɔ >>> *动
[2018-07-21 10:55:01] cth451 ǝɔuǝƃɹǝʌᴉp‾ɥʇɔ >>> 上面跳
[2018-07-21 10:55:08] Zero King >>> 我之前提到的是Stable无法修复的情况下
[2018-07-21 10:55:13] Mingcong Bai >>> 噢知道了
[2018-07-21 10:55:19] Mingcong Bai >>> 我加上优先
[2018-07-21 10:55:36] Zero King >>> 尽量避免因为其他更新跳票
[2018-07-21 10:55:58] Zero King >>> 因为一个周期就是一个季度
[2018-07-21 10:56:11] Mingcong Bai >>> “安全更新首要保证 Stable,假设无法通过任何手段保障 Stable(如需要大幅度更新或当前人力无法保障 backport),则根据上述更新规则向稳定性更低的方向走(Testing,Testing 不行就到 Chimera),任何最后接受安全更新的分支需要优先保证安全更新”
[2018-07-21 10:56:13] Mingcong Bai >>> @l2dy0
[2018-07-21 10:56:51] Mingcong Bai >>> 还是说和你的意思是反的
[2018-07-21 10:57:32] Mingcong Bai >>> @Lionium 其他地方还有问题吗
[2018-07-21 10:57:35] Zero King >>> 三个分支的merge关系是怎样的?
[2018-07-21 10:57:52] Mingcong Bai [Re: 771] >>> 暂时不提新问题了,下个议程时间到了,二次讨论的时候再提一下?
[2018-07-21 10:58:32] Zero King >>> 按原分支是保证bugfix,无法保证则在staging分支中优先处理
[2018-07-21 10:58:41] Mingcong Bai [Re: 773] >>> 嗯对
[2018-07-21 10:58:53] Zero King >>> OK
[2018-07-21 10:58:56] Mingcong Bai >>> 好的
[2018-07-21 10:59:03] Mingcong Bai >>> 剩下 @Lionium 确定一下
[2018-07-21 10:59:19] Lion [Re: 770] >>> 安全更新方面我没什么看法,其他部分感觉可以吧
[2018-07-21 10:59:24] Mingcong Bai >>> 好
[2018-07-21 10:59:28] Mingcong Bai >>> 这个话题结束了
[2018-07-21 10:59:37] Lion >>> 噢别
[2018-07-21 10:59:53] Mingcong Bai >>> 还有二次讨论,我稍后就给时间……
[2018-07-21 10:59:53] Lion >>> 我还没喘过一口气就到我了吗…
[2018-07-21 11:00:00] Lion >>> 11 点了
[2018-07-21 11:00:04] Mingcong Bai >>> 那么推到 11:15?
[2018-07-21 11:00:21] Lion >>> 好的
[2018-07-21 11:00:34] Mingcong Bai >>> #TOPIC 11:15 - 12:15 — AOSC OS 打包工作流和工具链版本管控
[2018-07-21 11:00:38] Mingcong Bai [Re: 787] >>> pinned the message
[2018-07-21 11:00:53] Mingcong Bai [Re: 786] >>> 超过时限就二次讨论吧,我们还有空余时间的,可以看看 README
[2018-07-21 11:01:05] Mingcong Bai >>> 另外我开了个公告板,刚刚确定的内容会发到那里
[2018-07-21 11:01:10] Mingcong Bai >>> https://t.me/aoscc2018bulletin [webpage]
[2018-07-21 11:05:49] AOSCC Relay bot >>> [mingcongbai] IRC 用户可参考此处 https://pastebin.aosc.io/paste/y7hBIrWbvSav2SL9AAacTA [webpage]
[2018-07-21 11:11:55] Mingcong Bai >>> 上述讨论第二议程将于 7 月 22 日 23:00 - 0:00 进行二次讨论。
详见: https://github.com/AOSC-Dev/aoscc/blob/master/2018/README.md [webpage]
[2018-07-21 11:14:53] cth451 ǝɔuǝƃɹǝʌᴉp‾ɥʇɔ [Re: 791] >>> 我日这个壁纸疯狂 spoiler
[2018-07-21 11:15:05] Piung Zjeuh >>> [chat_add_user]
[2018-07-21 11:15:09] Mingcong Bai [Re: 794] >>> LOL
[2018-07-21 11:15:12] cth451 ǝɔuǝƃɹǝʌᴉp‾ɥʇɔ >>> 虽然早就已经露出去了(
[2018-07-21 11:15:17] Mingcong Bai >>> @Lionium 上场啦
[2018-07-21 11:15:39] Lion >>> 114514
[2018-07-21 11:15:43] Mingcong Bai >>> 514
[2018-07-21 11:16:08] Lion >>> AOSC OS 打包工作流和工具链版本管控
[2018-07-21 11:16:22] Mingcong Bai >>> "Are we Live?"
[2018-07-21 11:16:32] Lion >>> 我分段来开展这个话题
[2018-07-21 11:16:34] AOSCC Relay bot >>> [YJSNPI] 噫喲
[2018-07-21 11:16:47] Lion >>> 大家在中途可以适当插嘴
[2018-07-21 11:17:10] Lion >>> 先是对打包一个简单的介绍。简单粗暴地说,打包无非是从代码生成 .deb 文件。
[2018-07-21 11:17:22] Lion >>> 对于一个包来说,很是简单。但之后可以看到,为整个系统打包实际上面临着重重困难。
[2018-07-21 11:17:49] Mingcong Bai >>> 插嘴(雾,我不敢了
[2018-07-21 11:19:08] Lion >>> 在做(看)的各位里面有打过包的有没打过的,特别是也有在其他系统上有打包经验的人。在这里介绍一下 AOSC OS 打包的常见操作
[2018-07-21 11:19:14] Lion >>> 从实际中常见的一个打包的例子来介绍:
[2018-07-21 11:19:32] Lion >>> #栗子 某特首,打开一箩筐网页,订阅邮件,日理万机,观察各种软件的更新动态。
[2018-07-21 11:19:52] Lion >>> 它注意到有一个叫做 systemd 的包,从 239 升级到了 240。于是它输入:
[2018-07-21 11:20:08] Lion >>> cd /bootroots/ciel/testing
nano TREE/extra-admin/systemd/spec #(并把 PKGVER=239 改为了 240,Ctrl+S 保存, Ctrl+X 退出)
ciel rollback -i amd64 #(amd64 是一个容器实例的名字)
ciel build -i amd64 systemd
[2018-07-21 11:20:28] Mingcong Bai >>> Nano fuck yeah
[2018-07-21 11:20:44] Lion >>> 好了,某特首继续阅读其他的新闻。
[2018-07-21 11:20:57] Lion >>> 电脑一热,新鲜的包子自动出现在了 OUTPUT/os-amd64/os3-dpkg/s/systemd_240-0_amd64.deb
[2018-07-21 11:21:04] Lion >>> 狭义上,打包的任务就结束了。
[2018-07-21 11:21:29] Lion >>> (没有政治指代)
[2018-07-21 11:21:42] Lion >>> 在随后的几天里,OUTPUT 文件夹里的包子一个一个多了起来。
[2018-07-21 11:21:48] 无聊怪 >>> 只要改版本号,中间的步骤 ciel 都完成了吗?
[2018-07-21 11:21:48] 死 肥 宅 鹹魚 [Re: 818] >>> hhhhhh
[2018-07-21 11:21:56] Lion [Re: 820] >>> 是的
[2018-07-21 11:22:07] Lion [Re: 819] >>> 文件夹就像膀胱,总有憋不住的那一天,某特首进行了一定的操作
[2018-07-21 11:22:16] Mingcong Bai [Re: 823] >>> *cough*
[2018-07-21 11:22:17] Lion >>> scp -r -P **** os-amd64 ****@repo.aosc.io:/mirror/os-amd64/testing
[2018-07-21 11:22:28] Lion >>> 这一大笼包就进了 testing 测试源。打包从输入到输出的全过程就结束了。
[2018-07-21 11:23:06] Sakamoto Tanmy >>> [document]
[2018-07-21 11:23:19] 死 肥 宅 鹹魚 [Re: 820] >>> 但是會有那種喪心病狂的老改構建系統的東西(就不能只改版本號了(x
[2018-07-21 11:23:20] Mingcong Bai >>> @skmt_tanmy 本群不能用 sticker
[2018-07-21 11:23:23] Lion >>> 顺带一提,实际上,此时 testing 源的用户没法立刻使用这些包。
[2018-07-21 11:23:40] Mingcong Bai [Re: 828] >>> 改版本号是最最基本的情况,有复杂很多的情况,@Lionium 一会会介绍
[2018-07-21 11:23:43] Sakamoto Tanmy [Re: 829] >>> sticker
[2018-07-21 11:23:45] Lion >>> 位于台湾的 repo.aosc.io 源服务器收到了这些包,按照东八区时间,每日 0 4 8 12 16 20 时的时候,自动扫描一次 /mirror/os-amd64 等文件夹,在其下生成 Packages 和 InRelease 文件。Packages 记录有新的包的信息,意味着从此时开始,用户已经可以在 apt update 后看到新的包了。
[2018-07-21 11:24:09] Lion >>> ↑从 scp/rsync 推上包到真正可用的细节。
[2018-07-21 11:24:49] Lion >>> 其实大家看到,更新一个包并不是特别的复杂
[2018-07-21 11:25:03] Mingcong Bai >>> 在最简单的例子里 *
[2018-07-21 11:25:30] Lion >>> 当然,如果 ciel build -i amd64 systemd 这一行满屏的消息滑过去之后突然报错,就是另一回事了
[2018-07-21 11:25:57] Lion >>> 上述,更新一个包,最顺利的流程就是这样的。
[2018-07-21 11:26:40] Sakamoto Tanmy >>> 插嘴:什么时候有从Ubuntu的功能呢(
[2018-07-21 11:26:47] Lion >>> 但是打包碰壁远远不只更新一下版本号就完成这么简单(尽管大多数情况下确实只需要改一下版本号)。
[2018-07-21 11:26:49] Mingcong Bai [Re: 839] >>> 比如?
[2018-07-21 11:26:53] 艾颖初 (。•́︿•̀。) | Systemd 真好玩 [Re: 822] >>> 所以为什么不搞个 nvchecker 呢
[2018-07-21 11:27:08] Lion [Re: 842] >>> 嘿,有了
[2018-07-21 11:27:09] Sakamoto Tanmy [Re: 841] >>> 输入法抽风没打完(
[2018-07-21 11:27:09] Mingcong Bai [Re: 842] >>> 其实现在有实现,参考 AOSC-Dev/scriptlets
[2018-07-21 11:27:19] Mingcong Bai [Re: 844] >>> 还是没明白
[2018-07-21 11:27:35] Sakamoto Tanmy [Re: 846] >>> 换源,dist upgrade(
[2018-07-21 11:27:48] Mingcong Bai [Re: 847] >>> 不可能,完全不是一个依赖树
[2018-07-21 11:27:53] Mingcong Bai >>> 以后也不可能
[2018-07-21 11:27:53] 艾颖初 (。•́︿•̀。) | Systemd 真好玩 [Re: 843] >>> 那快和隔壁arch自动化程度差不多了(不成
出问题的话
[2018-07-21 11:27:54] Staph. aureus | In your GI Tract [Re: 846] >>> (這句意思是叫你先 standby )
[2018-07-21 11:28:10] Lion [Re: 840] >>> 那么,现在我们引入一下各种打包的情况的一个整体观。
[2018-07-21 11:28:26] Lion >>> 刚才的例子里,是一个“更新”包的操作,打包所需要的脚本早已在 TREE/extra-admin/systemd 里的 spec 和 autobuild 文件夹里写好了,
[2018-07-21 11:28:29] Mingcong Bai [Re: 850] >>> 个人感觉自动化更新方面程度是够了,测试就还差距挺大
[2018-07-21 11:28:48] Lion [Re: 853] >>> 除了最常见的更新,新增软件包也是需要打包的。
[2018-07-21 11:28:55] Lion >>> 于是有人在电脑上实验和测试,最后手写出 spec 和 autobuild 各个文件,最后交给工具链打出最后的成品。
[2018-07-21 11:28:58] 艾颖初 (。•́︿•̀。) | Systemd 真好玩 [Re: 854] >>> 所以说不出问题的话
[2018-07-21 11:29:20] Lion >>> 但这是可以说最难的一个部分
[2018-07-21 11:29:29] Mingcong Bai [Re: 857] >>> 打包不能一味依靠这种猜测…… 无论风险多大都应该有测试架构,这个是 Arch 做得很好的一部分……
[2018-07-21 11:29:49] Lion >>> 参考数据库管理系统的理论,可以把打包视为“去变更”一个叫做 repo.aosc.io 的特殊数据库。
[2018-07-21 11:30:11] Lion >>> 数据库的操作又一个基本的概念叫做 CURD,指创建、更新、替换、删除四种常见操作
[2018-07-21 11:30:37] 艾颖初 (。•́︿•̀。) | Systemd 真好玩 [Re: 859] >>> ciel 内部现在依赖检测和安装重演做的怎么样
[2018-07-21 11:31:14] 艾颖初 (。•́︿•̀。) | Systemd 真好玩 >>> 比如说上次给人打到 /usr 下了?
[2018-07-21 11:31:28] Mingcong Bai [Re: 862] >>> 这方面功能是 acbs/autobuild 实现的,说实话进展不大
[2018-07-21 11:32:06] Lion [Re: 862] >>> // (中途偏题回复) 可以,因为只要 rollback 得干净,缺乏依赖是不能构建的。但是这是间接的、依靠错误来检测构建依赖
[2018-07-21 11:32:19] Lion [Re: 861] >>> CURD,先从创建一个包开始看
[2018-07-21 11:32:47] 艾颖初 (。•́︿•̀。) | Systemd 真好玩 [Re: 865] >>> 所以无法检测多余依赖?)好像插嘴插多了
[2018-07-21 11:33:00] Mingcong Bai [Re: 867] >>> 对
[2018-07-21 11:33:04] Lion [Re: 867] >>> (去主群继续)