forked from rust-lang/rust
-
Notifications
You must be signed in to change notification settings - Fork 0
/
rope.rs
1467 lines (1351 loc) · 41.7 KB
/
rope.rs
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
// Copyright 2012 The Rust Project Developers. See the COPYRIGHT
// file at the top-level directory of this distribution and at
// http://rust-lang.org/COPYRIGHT.
//
// Licensed under the Apache License, Version 2.0 <LICENSE-APACHE or
// http://www.apache.org/licenses/LICENSE-2.0> or the MIT license
// <LICENSE-MIT or http://opensource.org/licenses/MIT>, at your
// option. This file may not be copied, modified, or distributed
// except according to those terms.
/*!
* High-level text containers.
*
* Ropes are a high-level representation of text that offers
* much better performance than strings for common operations,
* and generally reduce memory allocations and copies, while only
* entailing a small degradation of less common operations.
*
* More precisely, where a string is represented as a memory buffer,
* a rope is a tree structure whose leaves are slices of immutable
* strings. Therefore, concatenation, appending, prepending, substrings,
* etc. are operations that require only trivial tree manipulation,
* generally without having to copy memory. In addition, the tree
* structure of ropes makes them suitable as a form of index to speed-up
* access to Unicode characters by index in long chunks of text.
*
* The following operations are algorithmically faster in ropes:
*
* * extracting a subrope is logarithmic (linear in strings);
* * appending/prepending is near-constant time (linear in strings);
* * concatenation is near-constant time (linear in strings);
* * char length is constant-time (linear in strings);
* * access to a character by index is logarithmic (linear in strings);
*/
#[forbid(deprecated_mode)];
use core::cast;
use core::char;
use core::option;
use core::prelude::*;
use core::str;
use core::uint;
use core::vec;
/// The type of ropes.
pub type Rope = node::Root;
/*
Section: Creating a rope
*/
/// Create an empty rope
pub pure fn empty() -> Rope {
return node::Empty;
}
/**
* Adopt a string as a rope.
*
* # Arguments
*
* * str - A valid string.
*
* # Return value
*
* A rope representing the same string as `str`. Depending of the length
* of `str`, this rope may be empty, flat or complex.
*
* # Performance notes
*
* * this operation does not copy the string;
* * the function runs in linear time.
*/
pub fn of_str(str: @~str) -> Rope {
return of_substr(str, 0u, str::len(*str));
}
/**
* As `of_str` but for a substring.
*
* # Arguments
* * byte_offset - The offset of `str` at which the rope starts.
* * byte_len - The number of bytes of `str` to use.
*
* # Return value
*
* A rope representing the same string as `str::substr(str, byte_offset,
* byte_len)`. Depending on `byte_len`, this rope may be empty, flat or
* complex.
*
* # Performance note
*
* This operation does not copy the substring.
*
* # Safety notes
*
* * this function does _not_ check the validity of the substring;
* * this function fails if `byte_offset` or `byte_len` do not match `str`.
*/
pub fn of_substr(str: @~str, byte_offset: uint, byte_len: uint) -> Rope {
if byte_len == 0u { return node::Empty; }
if byte_offset + byte_len > str::len(*str) { fail; }
return node::Content(node::of_substr(str, byte_offset, byte_len));
}
/*
Section: Adding things to a rope
*/
/**
* Add one char to the end of the rope
*
* # Performance note
*
* * this function executes in near-constant time
*/
pub fn append_char(rope: Rope, char: char) -> Rope {
return append_str(rope, @str::from_chars(~[char]));
}
/**
* Add one string to the end of the rope
*
* # Performance note
*
* * this function executes in near-linear time
*/
pub fn append_str(rope: Rope, str: @~str) -> Rope {
return append_rope(rope, of_str(str))
}
/**
* Add one char to the beginning of the rope
*
* # Performance note
* * this function executes in near-constant time
*/
pub fn prepend_char(rope: Rope, char: char) -> Rope {
return prepend_str(rope, @str::from_chars(~[char]));
}
/**
* Add one string to the beginning of the rope
*
* # Performance note
* * this function executes in near-linear time
*/
pub fn prepend_str(rope: Rope, str: @~str) -> Rope {
return append_rope(of_str(str), rope)
}
/// Concatenate two ropes
pub fn append_rope(left: Rope, right: Rope) -> Rope {
match (left) {
node::Empty => return right,
node::Content(left_content) => {
match (right) {
node::Empty => return left,
node::Content(right_content) => {
return node::Content(node::concat2(left_content, right_content));
}
}
}
}
}
/**
* Concatenate many ropes.
*
* If the ropes are balanced initially and have the same height, the resulting
* rope remains balanced. However, this function does not take any further
* measure to ensure that the result is balanced.
*/
pub fn concat(v: ~[Rope]) -> Rope {
//Copy `v` into a mut vector
let mut len = vec::len(v);
if len == 0u { return node::Empty; }
let ropes = vec::cast_to_mut(vec::from_elem(len, v[0]));
for uint::range(1u, len) |i| {
ropes[i] = v[i];
}
//Merge progresively
while len > 1u {
for uint::range(0u, len/2u) |i| {
ropes[i] = append_rope(ropes[2u*i], ropes[2u*i+1u]);
}
if len%2u != 0u {
ropes[len/2u] = ropes[len - 1u];
len = len/2u + 1u;
} else {
len = len/2u;
}
}
//Return final rope
return ropes[0];
}
/*
Section: Keeping ropes healthy
*/
/**
* Balance a rope.
*
* # Return value
*
* A copy of the rope in which small nodes have been grouped in memory,
* and with a reduced height.
*
* If you perform numerous rope concatenations, it is generally a good idea
* to rebalance your rope at some point, before using it for other purposes.
*/
pub fn bal(rope:Rope) -> Rope {
match (rope) {
node::Empty => return rope,
node::Content(x) => match (node::bal(x)) {
option::None => rope,
option::Some(y) => node::Content(y)
}
}
}
/*
Section: Transforming ropes
*/
/**
* Extract a subrope from a rope.
*
* # Performance note
*
* * on a balanced rope, this operation takes algorithmic time;
* * this operation does not involve any copying
*
* # Safety note
*
* * this function fails if char_offset/char_len do not represent
* valid positions in rope
*/
pub fn sub_chars(rope: Rope, char_offset: uint, char_len: uint) -> Rope {
if char_len == 0u { return node::Empty; }
match (rope) {
node::Empty => fail,
node::Content(node) => if char_len > node::char_len(node) {
fail
} else {
return node::Content(node::sub_chars(node, char_offset, char_len))
}
}
}
/**
* Extract a subrope from a rope.
*
* # Performance note
*
* * on a balanced rope, this operation takes algorithmic time;
* * this operation does not involve any copying
*
* # Safety note
*
* * this function fails if byte_offset/byte_len do not represent
* valid positions in rope
*/
pub fn sub_bytes(rope: Rope, byte_offset: uint, byte_len: uint) -> Rope {
if byte_len == 0u { return node::Empty; }
match (rope) {
node::Empty => fail,
node::Content(node) =>if byte_len > node::byte_len(node) {
fail
} else {
return node::Content(node::sub_bytes(node, byte_offset, byte_len))
}
}
}
/*
Section: Comparing ropes
*/
/**
* Compare two ropes by Unicode lexicographical order.
*
* This function compares only the contents of the rope, not their structure.
*
* # Return value
*
* A negative value if `left < right`, 0 if eq(left, right) or a positive
* value if `left > right`
*/
pub fn cmp(left: Rope, right: Rope) -> int {
match ((left, right)) {
(node::Empty, node::Empty) => return 0,
(node::Empty, _) => return -1,
(_, node::Empty) => return 1,
(node::Content(a), node::Content(b)) => {
return node::cmp(a, b);
}
}
}
/**
* Returns `true` if both ropes have the same content (regardless of
* their structure), `false` otherwise
*/
pub fn eq(left: Rope, right: Rope) -> bool {
return cmp(left, right) == 0;
}
/**
* # Arguments
*
* * left - an arbitrary rope
* * right - an arbitrary rope
*
* # Return value
*
* `true` if `left <= right` in lexicographical order (regardless of their
* structure), `false` otherwise
*/
pub fn le(left: Rope, right: Rope) -> bool {
return cmp(left, right) <= 0;
}
/**
* # Arguments
*
* * left - an arbitrary rope
* * right - an arbitrary rope
*
* # Return value
*
* `true` if `left < right` in lexicographical order (regardless of their
* structure), `false` otherwise
*/
pub fn lt(left: Rope, right: Rope) -> bool {
return cmp(left, right) < 0;
}
/**
* # Arguments
*
* * left - an arbitrary rope
* * right - an arbitrary rope
*
* # Return value
*
* `true` if `left >= right` in lexicographical order (regardless of their
* structure), `false` otherwise
*/
pub fn ge(left: Rope, right: Rope) -> bool {
return cmp(left, right) >= 0;
}
/**
* # Arguments
*
* * left - an arbitrary rope
* * right - an arbitrary rope
*
* # Return value
*
* `true` if `left > right` in lexicographical order (regardless of their
* structure), `false` otherwise
*/
pub fn gt(left: Rope, right: Rope) -> bool {
return cmp(left, right) > 0;
}
/*
Section: Iterating
*/
/**
* Loop through a rope, char by char
*
* While other mechanisms are available, this is generally the best manner
* of looping through the contents of a rope char by char. If you prefer a
* loop that iterates through the contents string by string (e.g. to print
* the contents of the rope or output it to the system), however,
* you should rather use `traverse_components`.
*
* # Arguments
*
* * rope - A rope to traverse. It may be empty.
* * it - A block to execute with each consecutive character of the rope.
* Return `true` to continue, `false` to stop.
*
* # Return value
*
* `true` If execution proceeded correctly, `false` if it was interrupted,
* that is if `it` returned `false` at any point.
*/
pub fn loop_chars(rope: Rope, it: fn(c: char) -> bool) -> bool {
match (rope) {
node::Empty => return true,
node::Content(x) => return node::loop_chars(x, it)
}
}
/**
* Loop through a rope, char by char, until the end.
*
* # Arguments
* * rope - A rope to traverse. It may be empty
* * it - A block to execute with each consecutive character of the rope.
*/
pub fn iter_chars(rope: Rope, it: fn(char)) {
do loop_chars(rope) |x| {
it(x);
true
};
}
/**
* Loop through a rope, string by string
*
* While other mechanisms are available, this is generally the best manner of
* looping through the contents of a rope string by string, which may be
* useful e.g. to print strings as you see them (without having to copy their
* contents into a new string), to send them to then network, to write them to
* a file, etc.. If you prefer a loop that iterates through the contents
* char by char (e.g. to search for a char), however, you should rather
* use `traverse`.
*
* # Arguments
*
* * rope - A rope to traverse. It may be empty
* * it - A block to execute with each consecutive string component of the
* rope. Return `true` to continue, `false` to stop
*
* # Return value
*
* `true` If execution proceeded correctly, `false` if it was interrupted,
* that is if `it` returned `false` at any point.
*/
pub fn loop_leaves(rope: Rope, it: fn(node::Leaf) -> bool) -> bool{
match (rope) {
node::Empty => return true,
node::Content(x) => return node::loop_leaves(x, it)
}
}
pub mod iterator {
pub mod leaf {
use rope::{Rope, node};
use core::prelude::*;
pub fn start(rope: Rope) -> node::leaf_iterator::T {
match (rope) {
node::Empty => return node::leaf_iterator::empty(),
node::Content(x) => return node::leaf_iterator::start(x)
}
}
pub fn next(it: &node::leaf_iterator::T) -> Option<node::Leaf> {
return node::leaf_iterator::next(it);
}
}
pub mod char {
use rope::{Rope, node};
use core::prelude::*;
pub fn start(rope: Rope) -> node::char_iterator::T {
match (rope) {
node::Empty => return node::char_iterator::empty(),
node::Content(x) => return node::char_iterator::start(x)
}
}
pub fn next(it: &node::char_iterator::T) -> Option<char> {
return node::char_iterator::next(it)
}
}
}
/*
Section: Rope properties
*/
/**
* Returns the height of the rope.
*
* The height of the rope is a bound on the number of operations which
* must be performed during a character access before finding the leaf in
* which a character is contained.
*
* # Performance note
*
* Constant time.
*/
pub pure fn height(rope: Rope) -> uint {
match (rope) {
node::Empty => return 0u,
node::Content(x) => return node::height(x)
}
}
/**
* The number of character in the rope
*
* # Performance note
*
* Constant time.
*/
pub pure fn char_len(rope: Rope) -> uint {
match (rope) {
node::Empty => return 0u,
node::Content(x) => return node::char_len(x)
}
}
/**
* The number of bytes in the rope
*
* # Performance note
*
* Constant time.
*/
pub pure fn byte_len(rope: Rope) -> uint {
match (rope) {
node::Empty => return 0u,
node::Content(x) => return node::byte_len(x)
}
}
/**
* The character at position `pos`
*
* # Arguments
*
* * pos - A position in the rope
*
* # Safety notes
*
* The function will fail if `pos` is not a valid position in the rope.
*
* # Performance note
*
* This function executes in a time proportional to the height of the
* rope + the (bounded) length of the largest leaf.
*/
pub fn char_at(rope: Rope, pos: uint) -> char {
match (rope) {
node::Empty => fail,
node::Content(x) => return node::char_at(x, pos)
}
}
/*
Section: Implementation
*/
pub mod node {
use rope::node;
use core::cast;
use core::char;
use core::option;
use core::prelude::*;
use core::str;
use core::uint;
use core::vec;
/// Implementation of type `rope`
pub enum Root {
/// An empty rope
Empty,
/// A non-empty rope
Content(@Node),
}
/**
* A text component in a rope.
*
* This is actually a slice in a rope, so as to ensure maximal sharing.
*
* # Fields
*
* * byte_offset = The number of bytes skippen in `content`
* * byte_len - The number of bytes of `content` to use
* * char_len - The number of chars in the leaf.
* * content - Contents of the leaf.
*
* Note that we can have `char_len < str::char_len(content)`, if
* this leaf is only a subset of the string. Also note that the
* string can be shared between several ropes, e.g. for indexing
* purposes.
*/
pub struct Leaf {
byte_offset: uint,
byte_len: uint,
char_len: uint,
content: @~str,
}
/**
* A node obtained from the concatenation of two other nodes
*
* # Fields
*
* * left - The node containing the beginning of the text.
* * right - The node containing the end of the text.
* * char_len - The number of chars contained in all leaves of this node.
* * byte_len - The number of bytes in the subrope.
*
* Used to pre-allocate the correct amount of storage for
* serialization.
*
* * height - Height of the subrope.
*
* Used for rebalancing and to allocate stacks for traversals.
*/
pub struct Concat {
//FIXME (#2744): Perhaps a `vec` instead of `left`/`right`
left: @Node,
right: @Node,
char_len: uint,
byte_len: uint,
height: uint,
}
pub enum Node {
/// A leaf consisting in a `str`
Leaf(Leaf),
/// The concatenation of two ropes
Concat(Concat),
}
/**
* The maximal number of chars that _should_ be permitted in a single node
*
* This is not a strict value
*/
pub const hint_max_leaf_char_len: uint = 256u;
/**
* The maximal height that _should_ be permitted in a tree.
*
* This is not a strict value
*/
pub const hint_max_node_height: uint = 16u;
/**
* Adopt a string as a node.
*
* If the string is longer than `max_leaf_char_len`, it is
* logically split between as many leaves as necessary. Regardless,
* the string itself is not copied.
*
* Performance note: The complexity of this function is linear in
* the length of `str`.
*/
pub fn of_str(str: @~str) -> @Node {
return of_substr(str, 0u, str::len(*str));
}
/**
* Adopt a slice of a string as a node.
*
* If the slice is longer than `max_leaf_char_len`, it is logically split
* between as many leaves as necessary. Regardless, the string itself
* is not copied
*
* # Arguments
*
* * byte_start - The byte offset where the slice of `str` starts.
* * byte_len - The number of bytes from `str` to use.
*
* # Safety note
*
* Behavior is undefined if `byte_start` or `byte_len` do not represent
* valid positions in `str`
*/
pub fn of_substr(str: @~str, byte_start: uint, byte_len: uint) -> @Node {
return of_substr_unsafer(str, byte_start, byte_len,
str::count_chars(*str, byte_start, byte_len));
}
/**
* Adopt a slice of a string as a node.
*
* If the slice is longer than `max_leaf_char_len`, it is logically split
* between as many leaves as necessary. Regardless, the string itself
* is not copied
*
* # Arguments
*
* * byte_start - The byte offset where the slice of `str` starts.
* * byte_len - The number of bytes from `str` to use.
* * char_len - The number of chars in `str` in the interval
* [byte_start, byte_start+byte_len)
*
* # Safety notes
*
* * Behavior is undefined if `byte_start` or `byte_len` do not represent
* valid positions in `str`
* * Behavior is undefined if `char_len` does not accurately represent the
* number of chars between byte_start and byte_start+byte_len
*/
pub fn of_substr_unsafer(str: @~str, byte_start: uint, byte_len: uint,
char_len: uint) -> @Node {
assert(byte_start + byte_len <= str::len(*str));
let candidate = @Leaf(Leaf {
byte_offset: byte_start,
byte_len: byte_len,
char_len: char_len,
content: str,
});
if char_len <= hint_max_leaf_char_len {
return candidate;
} else {
//Firstly, split `str` in slices of hint_max_leaf_char_len
let mut leaves = uint::div_ceil(char_len, hint_max_leaf_char_len);
//Number of leaves
let nodes = vec::cast_to_mut(vec::from_elem(leaves, candidate));
let mut i = 0u;
let mut offset = byte_start;
let first_leaf_char_len =
if char_len%hint_max_leaf_char_len == 0u {
hint_max_leaf_char_len
} else {
char_len%hint_max_leaf_char_len
};
while i < leaves {
let chunk_char_len: uint =
if i == 0u { first_leaf_char_len }
else { hint_max_leaf_char_len };
let chunk_byte_len =
str::count_bytes(*str, offset, chunk_char_len);
nodes[i] = @Leaf(Leaf {
byte_offset: offset,
byte_len: chunk_byte_len,
char_len: chunk_char_len,
content: str,
});
offset += chunk_byte_len;
i += 1u;
}
//Then, build a tree from these slices by collapsing them
while leaves > 1u {
i = 0u;
while i < leaves - 1u {//Concat nodes 0 with 1, 2 with 3 etc.
nodes[i/2u] = concat2(nodes[i], nodes[i + 1u]);
i += 2u;
}
if i == leaves - 1u {
//And don't forget the last node if it is in even position
nodes[i/2u] = nodes[i];
}
leaves = uint::div_ceil(leaves, 2u);
}
return nodes[0u];
}
}
pub pure fn byte_len(node: @Node) -> uint {
//FIXME (#2744): Could we do this without the pattern-matching?
match (*node) {
Leaf(y) => y.byte_len,
Concat(ref y) => y.byte_len
}
}
pub pure fn char_len(node: @Node) -> uint {
match (*node) {
Leaf(y) => y.char_len,
Concat(ref y) => y.char_len
}
}
/**
* Concatenate a forest of nodes into one tree.
*
* # Arguments
*
* * forest - The forest. This vector is progressively rewritten during
* execution and should be discarded as meaningless afterwards.
*/
pub fn tree_from_forest_destructive(forest: &[mut @Node]) -> @Node {
let mut i;
let mut len = vec::len(forest);
while len > 1u {
i = 0u;
while i < len - 1u {//Concat nodes 0 with 1, 2 with 3 etc.
let mut left = forest[i];
let mut right = forest[i+1u];
let left_len = char_len(left);
let right_len= char_len(right);
let mut left_height= height(left);
let mut right_height=height(right);
if left_len + right_len > hint_max_leaf_char_len {
if left_len <= hint_max_leaf_char_len {
left = flatten(left);
left_height = height(left);
}
if right_len <= hint_max_leaf_char_len {
right = flatten(right);
right_height = height(right);
}
}
if left_height >= hint_max_node_height {
left = of_substr_unsafer(@serialize_node(left),
0u,byte_len(left),
left_len);
}
if right_height >= hint_max_node_height {
right = of_substr_unsafer(@serialize_node(right),
0u,byte_len(right),
right_len);
}
forest[i/2u] = concat2(left, right);
i += 2u;
}
if i == len - 1u {
//And don't forget the last node if it is in even position
forest[i/2u] = forest[i];
}
len = uint::div_ceil(len, 2u);
}
return forest[0];
}
pub fn serialize_node(node: @Node) -> ~str {
unsafe {
let mut buf = vec::cast_to_mut(vec::from_elem(byte_len(node), 0));
let mut offset = 0u;//Current position in the buffer
let it = leaf_iterator::start(node);
loop {
match (leaf_iterator::next(&it)) {
option::None => break,
option::Some(x) => {
//FIXME (#2744): Replace with memcpy or something similar
let mut local_buf: ~[u8] =
cast::reinterpret_cast(&*x.content);
let mut i = x.byte_offset;
while i < x.byte_len {
buf[offset] = local_buf[i];
offset += 1u;
i += 1u;
}
cast::forget(move local_buf);
}
}
}
return cast::transmute(move buf);
}
}
/**
* Replace a subtree by a single leaf with the same contents.
*
* * Performance note
*
* This function executes in linear time.
*/
pub fn flatten(node: @Node) -> @Node {
unsafe {
match (*node) {
Leaf(_) => node,
Concat(ref x) => {
@Leaf(Leaf {
byte_offset: 0u,
byte_len: x.byte_len,
char_len: x.char_len,
content: @serialize_node(node),
})
}
}
}
}
/**
* Balance a node.
*
* # Algorithm
*
* * if the node height is smaller than `hint_max_node_height`, do nothing
* * otherwise, gather all leaves as a forest, rebuild a balanced node,
* concatenating small leaves along the way
*
* # Return value
*
* * `option::None` if no transformation happened
* * `option::Some(x)` otherwise, in which case `x` has the same contents
* as `node` bot lower height and/or fragmentation.
*/
pub fn bal(node: @Node) -> Option<@Node> {
if height(node) < hint_max_node_height { return option::None; }
//1. Gather all leaves as a forest
let mut forest = ~[];
let it = leaf_iterator::start(node);
loop {
match (leaf_iterator::next(&it)) {
option::None => break,
option::Some(x) => forest.push(@Leaf(x))
}
}
//2. Rebuild tree from forest
let root = @*tree_from_forest_destructive(forest);
return option::Some(root);
}
/**
* Compute the subnode of a node.
*
* # Arguments
*
* * node - A node
* * byte_offset - A byte offset in `node`
* * byte_len - The number of bytes to return
*
* # Performance notes
*
* * this function performs no copying;
* * this function executes in a time proportional to the height of `node`
*
* # Safety notes
*
* This function fails if `byte_offset` or `byte_len` do not represent
* valid positions in `node`.
*/
pub fn sub_bytes(node: @Node, byte_offset: uint,
byte_len: uint) -> @Node {
let mut node = node;
let mut byte_offset = byte_offset;
loop {
if byte_offset == 0u && byte_len == node::byte_len(node) {
return node;
}
match (*node) {
node::Leaf(x) => {
let char_len =
str::count_chars(*x.content, byte_offset, byte_len);
return @Leaf(Leaf {
byte_offset: byte_offset,
byte_len: byte_len,
char_len: char_len,
content: x.content,
});
}
node::Concat(ref x) => {
let left_len: uint = node::byte_len(x.left);
if byte_offset <= left_len {
if byte_offset + byte_len <= left_len {
//Case 1: Everything fits in x.left, tail-call
node = x.left;
} else {
//Case 2: A (non-empty, possibly full) suffix
//of x.left and a (non-empty, possibly full) prefix
//of x.right
let left_result =
sub_bytes(x.left, byte_offset, left_len);
let right_result =
sub_bytes(x.right, 0u, left_len - byte_offset);
return concat2(left_result, right_result);
}
} else {
//Case 3: Everything fits in x.right
byte_offset -= left_len;
node = x.right;
}
}
}
};
}
/**
* Compute the subnode of a node.
*
* # Arguments
*
* * node - A node
* * char_offset - A char offset in `node`
* * char_len - The number of chars to return
*
* # Performance notes
*
* * this function performs no copying;
* * this function executes in a time proportional to the height of `node`
*
* # Safety notes
*
* This function fails if `char_offset` or `char_len` do not represent
* valid positions in `node`.
*/
pub fn sub_chars(node: @Node, char_offset: uint,
char_len: uint) -> @Node {