diff.go
41.1 KB
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
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
1088
1089
1090
1091
1092
1093
1094
1095
1096
1097
1098
1099
1100
1101
1102
1103
1104
1105
1106
1107
1108
1109
1110
1111
1112
1113
1114
1115
1116
1117
1118
1119
1120
1121
1122
1123
1124
1125
1126
1127
1128
1129
1130
1131
1132
1133
1134
1135
1136
1137
1138
1139
1140
1141
1142
1143
1144
1145
1146
1147
1148
1149
1150
1151
1152
1153
1154
1155
1156
1157
1158
1159
1160
1161
1162
1163
1164
1165
1166
1167
1168
1169
1170
1171
1172
1173
1174
1175
1176
1177
1178
1179
1180
1181
1182
1183
1184
1185
1186
1187
1188
1189
1190
1191
1192
1193
1194
1195
1196
1197
1198
1199
1200
1201
1202
1203
1204
1205
1206
1207
1208
1209
1210
1211
1212
1213
1214
1215
1216
1217
1218
1219
1220
1221
1222
1223
1224
1225
1226
1227
1228
1229
1230
1231
1232
1233
1234
1235
1236
1237
1238
1239
1240
1241
1242
1243
1244
1245
1246
1247
1248
1249
1250
1251
1252
1253
1254
1255
1256
1257
1258
1259
1260
1261
1262
1263
1264
1265
1266
1267
1268
1269
1270
1271
1272
1273
1274
1275
1276
1277
1278
1279
1280
1281
1282
1283
1284
1285
1286
1287
1288
1289
1290
1291
1292
1293
1294
1295
1296
1297
1298
1299
1300
1301
1302
1303
1304
1305
1306
1307
1308
1309
1310
1311
1312
1313
1314
1315
1316
1317
1318
1319
1320
1321
1322
1323
1324
1325
1326
1327
1328
1329
1330
1331
1332
1333
1334
1335
1336
1337
1338
1339
1340
1341
1342
1343
1344
1345
// Copyright (c) 2012-2016 The go-diff authors. All rights reserved.
// https://github.com/sergi/go-diff
// See the included LICENSE file for license details.
//
// go-diff is a Go implementation of Google's Diff, Match, and Patch library
// Original library is Copyright (c) 2006 Google Inc.
// http://code.google.com/p/google-diff-match-patch/
package diffmatchpatch
import (
"bytes"
"errors"
"fmt"
"html"
"math"
"net/url"
"regexp"
"strconv"
"strings"
"time"
"unicode/utf8"
)
// Operation defines the operation of a diff item.
type Operation int8
//go:generate stringer -type=Operation -trimprefix=Diff
const (
// DiffDelete item represents a delete diff.
DiffDelete Operation = -1
// DiffInsert item represents an insert diff.
DiffInsert Operation = 1
// DiffEqual item represents an equal diff.
DiffEqual Operation = 0
)
// Diff represents one diff operation
type Diff struct {
Type Operation
Text string
}
// splice removes amount elements from slice at index index, replacing them with elements.
func splice(slice []Diff, index int, amount int, elements ...Diff) []Diff {
if len(elements) == amount {
// Easy case: overwrite the relevant items.
copy(slice[index:], elements)
return slice
}
if len(elements) < amount {
// Fewer new items than old.
// Copy in the new items.
copy(slice[index:], elements)
// Shift the remaining items left.
copy(slice[index+len(elements):], slice[index+amount:])
// Calculate the new end of the slice.
end := len(slice) - amount + len(elements)
// Zero stranded elements at end so that they can be garbage collected.
tail := slice[end:]
for i := range tail {
tail[i] = Diff{}
}
return slice[:end]
}
// More new items than old.
// Make room in slice for new elements.
// There's probably an even more efficient way to do this,
// but this is simple and clear.
need := len(slice) - amount + len(elements)
for len(slice) < need {
slice = append(slice, Diff{})
}
// Shift slice elements right to make room for new elements.
copy(slice[index+len(elements):], slice[index+amount:])
// Copy in new elements.
copy(slice[index:], elements)
return slice
}
// DiffMain finds the differences between two texts.
// If an invalid UTF-8 sequence is encountered, it will be replaced by the Unicode replacement character.
func (dmp *DiffMatchPatch) DiffMain(text1, text2 string, checklines bool) []Diff {
return dmp.DiffMainRunes([]rune(text1), []rune(text2), checklines)
}
// DiffMainRunes finds the differences between two rune sequences.
// If an invalid UTF-8 sequence is encountered, it will be replaced by the Unicode replacement character.
func (dmp *DiffMatchPatch) DiffMainRunes(text1, text2 []rune, checklines bool) []Diff {
var deadline time.Time
if dmp.DiffTimeout > 0 {
deadline = time.Now().Add(dmp.DiffTimeout)
}
return dmp.diffMainRunes(text1, text2, checklines, deadline)
}
func (dmp *DiffMatchPatch) diffMainRunes(text1, text2 []rune, checklines bool, deadline time.Time) []Diff {
if runesEqual(text1, text2) {
var diffs []Diff
if len(text1) > 0 {
diffs = append(diffs, Diff{DiffEqual, string(text1)})
}
return diffs
}
// Trim off common prefix (speedup).
commonlength := commonPrefixLength(text1, text2)
commonprefix := text1[:commonlength]
text1 = text1[commonlength:]
text2 = text2[commonlength:]
// Trim off common suffix (speedup).
commonlength = commonSuffixLength(text1, text2)
commonsuffix := text1[len(text1)-commonlength:]
text1 = text1[:len(text1)-commonlength]
text2 = text2[:len(text2)-commonlength]
// Compute the diff on the middle block.
diffs := dmp.diffCompute(text1, text2, checklines, deadline)
// Restore the prefix and suffix.
if len(commonprefix) != 0 {
diffs = append([]Diff{Diff{DiffEqual, string(commonprefix)}}, diffs...)
}
if len(commonsuffix) != 0 {
diffs = append(diffs, Diff{DiffEqual, string(commonsuffix)})
}
return dmp.DiffCleanupMerge(diffs)
}
// diffCompute finds the differences between two rune slices. Assumes that the texts do not have any common prefix or suffix.
func (dmp *DiffMatchPatch) diffCompute(text1, text2 []rune, checklines bool, deadline time.Time) []Diff {
diffs := []Diff{}
if len(text1) == 0 {
// Just add some text (speedup).
return append(diffs, Diff{DiffInsert, string(text2)})
} else if len(text2) == 0 {
// Just delete some text (speedup).
return append(diffs, Diff{DiffDelete, string(text1)})
}
var longtext, shorttext []rune
if len(text1) > len(text2) {
longtext = text1
shorttext = text2
} else {
longtext = text2
shorttext = text1
}
if i := runesIndex(longtext, shorttext); i != -1 {
op := DiffInsert
// Swap insertions for deletions if diff is reversed.
if len(text1) > len(text2) {
op = DiffDelete
}
// Shorter text is inside the longer text (speedup).
return []Diff{
Diff{op, string(longtext[:i])},
Diff{DiffEqual, string(shorttext)},
Diff{op, string(longtext[i+len(shorttext):])},
}
} else if len(shorttext) == 1 {
// Single character string.
// After the previous speedup, the character can't be an equality.
return []Diff{
Diff{DiffDelete, string(text1)},
Diff{DiffInsert, string(text2)},
}
// Check to see if the problem can be split in two.
} else if hm := dmp.diffHalfMatch(text1, text2); hm != nil {
// A half-match was found, sort out the return data.
text1A := hm[0]
text1B := hm[1]
text2A := hm[2]
text2B := hm[3]
midCommon := hm[4]
// Send both pairs off for separate processing.
diffsA := dmp.diffMainRunes(text1A, text2A, checklines, deadline)
diffsB := dmp.diffMainRunes(text1B, text2B, checklines, deadline)
// Merge the results.
diffs := diffsA
diffs = append(diffs, Diff{DiffEqual, string(midCommon)})
diffs = append(diffs, diffsB...)
return diffs
} else if checklines && len(text1) > 100 && len(text2) > 100 {
return dmp.diffLineMode(text1, text2, deadline)
}
return dmp.diffBisect(text1, text2, deadline)
}
// diffLineMode does a quick line-level diff on both []runes, then rediff the parts for greater accuracy. This speedup can produce non-minimal diffs.
func (dmp *DiffMatchPatch) diffLineMode(text1, text2 []rune, deadline time.Time) []Diff {
// Scan the text on a line-by-line basis first.
text1, text2, linearray := dmp.diffLinesToRunes(text1, text2)
diffs := dmp.diffMainRunes(text1, text2, false, deadline)
// Convert the diff back to original text.
diffs = dmp.DiffCharsToLines(diffs, linearray)
// Eliminate freak matches (e.g. blank lines)
diffs = dmp.DiffCleanupSemantic(diffs)
// Rediff any replacement blocks, this time character-by-character.
// Add a dummy entry at the end.
diffs = append(diffs, Diff{DiffEqual, ""})
pointer := 0
countDelete := 0
countInsert := 0
// NOTE: Rune slices are slower than using strings in this case.
textDelete := ""
textInsert := ""
for pointer < len(diffs) {
switch diffs[pointer].Type {
case DiffInsert:
countInsert++
textInsert += diffs[pointer].Text
case DiffDelete:
countDelete++
textDelete += diffs[pointer].Text
case DiffEqual:
// Upon reaching an equality, check for prior redundancies.
if countDelete >= 1 && countInsert >= 1 {
// Delete the offending records and add the merged ones.
diffs = splice(diffs, pointer-countDelete-countInsert,
countDelete+countInsert)
pointer = pointer - countDelete - countInsert
a := dmp.diffMainRunes([]rune(textDelete), []rune(textInsert), false, deadline)
for j := len(a) - 1; j >= 0; j-- {
diffs = splice(diffs, pointer, 0, a[j])
}
pointer = pointer + len(a)
}
countInsert = 0
countDelete = 0
textDelete = ""
textInsert = ""
}
pointer++
}
return diffs[:len(diffs)-1] // Remove the dummy entry at the end.
}
// DiffBisect finds the 'middle snake' of a diff, split the problem in two and return the recursively constructed diff.
// If an invalid UTF-8 sequence is encountered, it will be replaced by the Unicode replacement character.
// See Myers 1986 paper: An O(ND) Difference Algorithm and Its Variations.
func (dmp *DiffMatchPatch) DiffBisect(text1, text2 string, deadline time.Time) []Diff {
// Unused in this code, but retained for interface compatibility.
return dmp.diffBisect([]rune(text1), []rune(text2), deadline)
}
// diffBisect finds the 'middle snake' of a diff, splits the problem in two and returns the recursively constructed diff.
// See Myers's 1986 paper: An O(ND) Difference Algorithm and Its Variations.
func (dmp *DiffMatchPatch) diffBisect(runes1, runes2 []rune, deadline time.Time) []Diff {
// Cache the text lengths to prevent multiple calls.
runes1Len, runes2Len := len(runes1), len(runes2)
maxD := (runes1Len + runes2Len + 1) / 2
vOffset := maxD
vLength := 2 * maxD
v1 := make([]int, vLength)
v2 := make([]int, vLength)
for i := range v1 {
v1[i] = -1
v2[i] = -1
}
v1[vOffset+1] = 0
v2[vOffset+1] = 0
delta := runes1Len - runes2Len
// If the total number of characters is odd, then the front path will collide with the reverse path.
front := (delta%2 != 0)
// Offsets for start and end of k loop. Prevents mapping of space beyond the grid.
k1start := 0
k1end := 0
k2start := 0
k2end := 0
for d := 0; d < maxD; d++ {
// Bail out if deadline is reached.
if !deadline.IsZero() && d%16 == 0 && time.Now().After(deadline) {
break
}
// Walk the front path one step.
for k1 := -d + k1start; k1 <= d-k1end; k1 += 2 {
k1Offset := vOffset + k1
var x1 int
if k1 == -d || (k1 != d && v1[k1Offset-1] < v1[k1Offset+1]) {
x1 = v1[k1Offset+1]
} else {
x1 = v1[k1Offset-1] + 1
}
y1 := x1 - k1
for x1 < runes1Len && y1 < runes2Len {
if runes1[x1] != runes2[y1] {
break
}
x1++
y1++
}
v1[k1Offset] = x1
if x1 > runes1Len {
// Ran off the right of the graph.
k1end += 2
} else if y1 > runes2Len {
// Ran off the bottom of the graph.
k1start += 2
} else if front {
k2Offset := vOffset + delta - k1
if k2Offset >= 0 && k2Offset < vLength && v2[k2Offset] != -1 {
// Mirror x2 onto top-left coordinate system.
x2 := runes1Len - v2[k2Offset]
if x1 >= x2 {
// Overlap detected.
return dmp.diffBisectSplit(runes1, runes2, x1, y1, deadline)
}
}
}
}
// Walk the reverse path one step.
for k2 := -d + k2start; k2 <= d-k2end; k2 += 2 {
k2Offset := vOffset + k2
var x2 int
if k2 == -d || (k2 != d && v2[k2Offset-1] < v2[k2Offset+1]) {
x2 = v2[k2Offset+1]
} else {
x2 = v2[k2Offset-1] + 1
}
var y2 = x2 - k2
for x2 < runes1Len && y2 < runes2Len {
if runes1[runes1Len-x2-1] != runes2[runes2Len-y2-1] {
break
}
x2++
y2++
}
v2[k2Offset] = x2
if x2 > runes1Len {
// Ran off the left of the graph.
k2end += 2
} else if y2 > runes2Len {
// Ran off the top of the graph.
k2start += 2
} else if !front {
k1Offset := vOffset + delta - k2
if k1Offset >= 0 && k1Offset < vLength && v1[k1Offset] != -1 {
x1 := v1[k1Offset]
y1 := vOffset + x1 - k1Offset
// Mirror x2 onto top-left coordinate system.
x2 = runes1Len - x2
if x1 >= x2 {
// Overlap detected.
return dmp.diffBisectSplit(runes1, runes2, x1, y1, deadline)
}
}
}
}
}
// Diff took too long and hit the deadline or number of diffs equals number of characters, no commonality at all.
return []Diff{
Diff{DiffDelete, string(runes1)},
Diff{DiffInsert, string(runes2)},
}
}
func (dmp *DiffMatchPatch) diffBisectSplit(runes1, runes2 []rune, x, y int,
deadline time.Time) []Diff {
runes1a := runes1[:x]
runes2a := runes2[:y]
runes1b := runes1[x:]
runes2b := runes2[y:]
// Compute both diffs serially.
diffs := dmp.diffMainRunes(runes1a, runes2a, false, deadline)
diffsb := dmp.diffMainRunes(runes1b, runes2b, false, deadline)
return append(diffs, diffsb...)
}
// DiffLinesToChars splits two texts into a list of strings, and educes the texts to a string of hashes where each Unicode character represents one line.
// It's slightly faster to call DiffLinesToRunes first, followed by DiffMainRunes.
func (dmp *DiffMatchPatch) DiffLinesToChars(text1, text2 string) (string, string, []string) {
chars1, chars2, lineArray := dmp.DiffLinesToRunes(text1, text2)
return string(chars1), string(chars2), lineArray
}
// DiffLinesToRunes splits two texts into a list of runes. Each rune represents one line.
func (dmp *DiffMatchPatch) DiffLinesToRunes(text1, text2 string) ([]rune, []rune, []string) {
// '\x00' is a valid character, but various debuggers don't like it. So we'll insert a junk entry to avoid generating a null character.
lineArray := []string{""} // e.g. lineArray[4] == 'Hello\n'
lineHash := map[string]int{} // e.g. lineHash['Hello\n'] == 4
chars1 := dmp.diffLinesToRunesMunge(text1, &lineArray, lineHash)
chars2 := dmp.diffLinesToRunesMunge(text2, &lineArray, lineHash)
return chars1, chars2, lineArray
}
func (dmp *DiffMatchPatch) diffLinesToRunes(text1, text2 []rune) ([]rune, []rune, []string) {
return dmp.DiffLinesToRunes(string(text1), string(text2))
}
// diffLinesToRunesMunge splits a text into an array of strings, and reduces the texts to a []rune where each Unicode character represents one line.
// We use strings instead of []runes as input mainly because you can't use []rune as a map key.
func (dmp *DiffMatchPatch) diffLinesToRunesMunge(text string, lineArray *[]string, lineHash map[string]int) []rune {
// Walk the text, pulling out a substring for each line. text.split('\n') would would temporarily double our memory footprint. Modifying text would create many large strings to garbage collect.
lineStart := 0
lineEnd := -1
runes := []rune{}
for lineEnd < len(text)-1 {
lineEnd = indexOf(text, "\n", lineStart)
if lineEnd == -1 {
lineEnd = len(text) - 1
}
line := text[lineStart : lineEnd+1]
lineStart = lineEnd + 1
lineValue, ok := lineHash[line]
if ok {
runes = append(runes, rune(lineValue))
} else {
*lineArray = append(*lineArray, line)
lineHash[line] = len(*lineArray) - 1
runes = append(runes, rune(len(*lineArray)-1))
}
}
return runes
}
// DiffCharsToLines rehydrates the text in a diff from a string of line hashes to real lines of text.
func (dmp *DiffMatchPatch) DiffCharsToLines(diffs []Diff, lineArray []string) []Diff {
hydrated := make([]Diff, 0, len(diffs))
for _, aDiff := range diffs {
chars := aDiff.Text
text := make([]string, len(chars))
for i, r := range chars {
text[i] = lineArray[r]
}
aDiff.Text = strings.Join(text, "")
hydrated = append(hydrated, aDiff)
}
return hydrated
}
// DiffCommonPrefix determines the common prefix length of two strings.
func (dmp *DiffMatchPatch) DiffCommonPrefix(text1, text2 string) int {
// Unused in this code, but retained for interface compatibility.
return commonPrefixLength([]rune(text1), []rune(text2))
}
// DiffCommonSuffix determines the common suffix length of two strings.
func (dmp *DiffMatchPatch) DiffCommonSuffix(text1, text2 string) int {
// Unused in this code, but retained for interface compatibility.
return commonSuffixLength([]rune(text1), []rune(text2))
}
// commonPrefixLength returns the length of the common prefix of two rune slices.
func commonPrefixLength(text1, text2 []rune) int {
// Linear search. See comment in commonSuffixLength.
n := 0
for ; n < len(text1) && n < len(text2); n++ {
if text1[n] != text2[n] {
return n
}
}
return n
}
// commonSuffixLength returns the length of the common suffix of two rune slices.
func commonSuffixLength(text1, text2 []rune) int {
// Use linear search rather than the binary search discussed at https://neil.fraser.name/news/2007/10/09/.
// See discussion at https://github.com/sergi/go-diff/issues/54.
i1 := len(text1)
i2 := len(text2)
for n := 0; ; n++ {
i1--
i2--
if i1 < 0 || i2 < 0 || text1[i1] != text2[i2] {
return n
}
}
}
// DiffCommonOverlap determines if the suffix of one string is the prefix of another.
func (dmp *DiffMatchPatch) DiffCommonOverlap(text1 string, text2 string) int {
// Cache the text lengths to prevent multiple calls.
text1Length := len(text1)
text2Length := len(text2)
// Eliminate the null case.
if text1Length == 0 || text2Length == 0 {
return 0
}
// Truncate the longer string.
if text1Length > text2Length {
text1 = text1[text1Length-text2Length:]
} else if text1Length < text2Length {
text2 = text2[0:text1Length]
}
textLength := int(math.Min(float64(text1Length), float64(text2Length)))
// Quick check for the worst case.
if text1 == text2 {
return textLength
}
// Start by looking for a single character match and increase length until no match is found. Performance analysis: http://neil.fraser.name/news/2010/11/04/
best := 0
length := 1
for {
pattern := text1[textLength-length:]
found := strings.Index(text2, pattern)
if found == -1 {
break
}
length += found
if found == 0 || text1[textLength-length:] == text2[0:length] {
best = length
length++
}
}
return best
}
// DiffHalfMatch checks whether the two texts share a substring which is at least half the length of the longer text. This speedup can produce non-minimal diffs.
func (dmp *DiffMatchPatch) DiffHalfMatch(text1, text2 string) []string {
// Unused in this code, but retained for interface compatibility.
runeSlices := dmp.diffHalfMatch([]rune(text1), []rune(text2))
if runeSlices == nil {
return nil
}
result := make([]string, len(runeSlices))
for i, r := range runeSlices {
result[i] = string(r)
}
return result
}
func (dmp *DiffMatchPatch) diffHalfMatch(text1, text2 []rune) [][]rune {
if dmp.DiffTimeout <= 0 {
// Don't risk returning a non-optimal diff if we have unlimited time.
return nil
}
var longtext, shorttext []rune
if len(text1) > len(text2) {
longtext = text1
shorttext = text2
} else {
longtext = text2
shorttext = text1
}
if len(longtext) < 4 || len(shorttext)*2 < len(longtext) {
return nil // Pointless.
}
// First check if the second quarter is the seed for a half-match.
hm1 := dmp.diffHalfMatchI(longtext, shorttext, int(float64(len(longtext)+3)/4))
// Check again based on the third quarter.
hm2 := dmp.diffHalfMatchI(longtext, shorttext, int(float64(len(longtext)+1)/2))
hm := [][]rune{}
if hm1 == nil && hm2 == nil {
return nil
} else if hm2 == nil {
hm = hm1
} else if hm1 == nil {
hm = hm2
} else {
// Both matched. Select the longest.
if len(hm1[4]) > len(hm2[4]) {
hm = hm1
} else {
hm = hm2
}
}
// A half-match was found, sort out the return data.
if len(text1) > len(text2) {
return hm
}
return [][]rune{hm[2], hm[3], hm[0], hm[1], hm[4]}
}
// diffHalfMatchI checks if a substring of shorttext exist within longtext such that the substring is at least half the length of longtext?
// Returns a slice containing the prefix of longtext, the suffix of longtext, the prefix of shorttext, the suffix of shorttext and the common middle, or null if there was no match.
func (dmp *DiffMatchPatch) diffHalfMatchI(l, s []rune, i int) [][]rune {
var bestCommonA []rune
var bestCommonB []rune
var bestCommonLen int
var bestLongtextA []rune
var bestLongtextB []rune
var bestShorttextA []rune
var bestShorttextB []rune
// Start with a 1/4 length substring at position i as a seed.
seed := l[i : i+len(l)/4]
for j := runesIndexOf(s, seed, 0); j != -1; j = runesIndexOf(s, seed, j+1) {
prefixLength := commonPrefixLength(l[i:], s[j:])
suffixLength := commonSuffixLength(l[:i], s[:j])
if bestCommonLen < suffixLength+prefixLength {
bestCommonA = s[j-suffixLength : j]
bestCommonB = s[j : j+prefixLength]
bestCommonLen = len(bestCommonA) + len(bestCommonB)
bestLongtextA = l[:i-suffixLength]
bestLongtextB = l[i+prefixLength:]
bestShorttextA = s[:j-suffixLength]
bestShorttextB = s[j+prefixLength:]
}
}
if bestCommonLen*2 < len(l) {
return nil
}
return [][]rune{
bestLongtextA,
bestLongtextB,
bestShorttextA,
bestShorttextB,
append(bestCommonA, bestCommonB...),
}
}
// DiffCleanupSemantic reduces the number of edits by eliminating semantically trivial equalities.
func (dmp *DiffMatchPatch) DiffCleanupSemantic(diffs []Diff) []Diff {
changes := false
// Stack of indices where equalities are found.
equalities := make([]int, 0, len(diffs))
var lastequality string
// Always equal to diffs[equalities[equalitiesLength - 1]][1]
var pointer int // Index of current position.
// Number of characters that changed prior to the equality.
var lengthInsertions1, lengthDeletions1 int
// Number of characters that changed after the equality.
var lengthInsertions2, lengthDeletions2 int
for pointer < len(diffs) {
if diffs[pointer].Type == DiffEqual {
// Equality found.
equalities = append(equalities, pointer)
lengthInsertions1 = lengthInsertions2
lengthDeletions1 = lengthDeletions2
lengthInsertions2 = 0
lengthDeletions2 = 0
lastequality = diffs[pointer].Text
} else {
// An insertion or deletion.
if diffs[pointer].Type == DiffInsert {
lengthInsertions2 += len(diffs[pointer].Text)
} else {
lengthDeletions2 += len(diffs[pointer].Text)
}
// Eliminate an equality that is smaller or equal to the edits on both sides of it.
difference1 := int(math.Max(float64(lengthInsertions1), float64(lengthDeletions1)))
difference2 := int(math.Max(float64(lengthInsertions2), float64(lengthDeletions2)))
if len(lastequality) > 0 &&
(len(lastequality) <= difference1) &&
(len(lastequality) <= difference2) {
// Duplicate record.
insPoint := equalities[len(equalities)-1]
diffs = splice(diffs, insPoint, 0, Diff{DiffDelete, lastequality})
// Change second copy to insert.
diffs[insPoint+1].Type = DiffInsert
// Throw away the equality we just deleted.
equalities = equalities[:len(equalities)-1]
if len(equalities) > 0 {
equalities = equalities[:len(equalities)-1]
}
pointer = -1
if len(equalities) > 0 {
pointer = equalities[len(equalities)-1]
}
lengthInsertions1 = 0 // Reset the counters.
lengthDeletions1 = 0
lengthInsertions2 = 0
lengthDeletions2 = 0
lastequality = ""
changes = true
}
}
pointer++
}
// Normalize the diff.
if changes {
diffs = dmp.DiffCleanupMerge(diffs)
}
diffs = dmp.DiffCleanupSemanticLossless(diffs)
// Find any overlaps between deletions and insertions.
// e.g: <del>abcxxx</del><ins>xxxdef</ins>
// -> <del>abc</del>xxx<ins>def</ins>
// e.g: <del>xxxabc</del><ins>defxxx</ins>
// -> <ins>def</ins>xxx<del>abc</del>
// Only extract an overlap if it is as big as the edit ahead or behind it.
pointer = 1
for pointer < len(diffs) {
if diffs[pointer-1].Type == DiffDelete &&
diffs[pointer].Type == DiffInsert {
deletion := diffs[pointer-1].Text
insertion := diffs[pointer].Text
overlapLength1 := dmp.DiffCommonOverlap(deletion, insertion)
overlapLength2 := dmp.DiffCommonOverlap(insertion, deletion)
if overlapLength1 >= overlapLength2 {
if float64(overlapLength1) >= float64(len(deletion))/2 ||
float64(overlapLength1) >= float64(len(insertion))/2 {
// Overlap found. Insert an equality and trim the surrounding edits.
diffs = splice(diffs, pointer, 0, Diff{DiffEqual, insertion[:overlapLength1]})
diffs[pointer-1].Text =
deletion[0 : len(deletion)-overlapLength1]
diffs[pointer+1].Text = insertion[overlapLength1:]
pointer++
}
} else {
if float64(overlapLength2) >= float64(len(deletion))/2 ||
float64(overlapLength2) >= float64(len(insertion))/2 {
// Reverse overlap found. Insert an equality and swap and trim the surrounding edits.
overlap := Diff{DiffEqual, deletion[:overlapLength2]}
diffs = splice(diffs, pointer, 0, overlap)
diffs[pointer-1].Type = DiffInsert
diffs[pointer-1].Text = insertion[0 : len(insertion)-overlapLength2]
diffs[pointer+1].Type = DiffDelete
diffs[pointer+1].Text = deletion[overlapLength2:]
pointer++
}
}
pointer++
}
pointer++
}
return diffs
}
// Define some regex patterns for matching boundaries.
var (
nonAlphaNumericRegex = regexp.MustCompile(`[^a-zA-Z0-9]`)
whitespaceRegex = regexp.MustCompile(`\s`)
linebreakRegex = regexp.MustCompile(`[\r\n]`)
blanklineEndRegex = regexp.MustCompile(`\n\r?\n$`)
blanklineStartRegex = regexp.MustCompile(`^\r?\n\r?\n`)
)
// diffCleanupSemanticScore computes a score representing whether the internal boundary falls on logical boundaries.
// Scores range from 6 (best) to 0 (worst). Closure, but does not reference any external variables.
func diffCleanupSemanticScore(one, two string) int {
if len(one) == 0 || len(two) == 0 {
// Edges are the best.
return 6
}
// Each port of this function behaves slightly differently due to subtle differences in each language's definition of things like 'whitespace'. Since this function's purpose is largely cosmetic, the choice has been made to use each language's native features rather than force total conformity.
rune1, _ := utf8.DecodeLastRuneInString(one)
rune2, _ := utf8.DecodeRuneInString(two)
char1 := string(rune1)
char2 := string(rune2)
nonAlphaNumeric1 := nonAlphaNumericRegex.MatchString(char1)
nonAlphaNumeric2 := nonAlphaNumericRegex.MatchString(char2)
whitespace1 := nonAlphaNumeric1 && whitespaceRegex.MatchString(char1)
whitespace2 := nonAlphaNumeric2 && whitespaceRegex.MatchString(char2)
lineBreak1 := whitespace1 && linebreakRegex.MatchString(char1)
lineBreak2 := whitespace2 && linebreakRegex.MatchString(char2)
blankLine1 := lineBreak1 && blanklineEndRegex.MatchString(one)
blankLine2 := lineBreak2 && blanklineEndRegex.MatchString(two)
if blankLine1 || blankLine2 {
// Five points for blank lines.
return 5
} else if lineBreak1 || lineBreak2 {
// Four points for line breaks.
return 4
} else if nonAlphaNumeric1 && !whitespace1 && whitespace2 {
// Three points for end of sentences.
return 3
} else if whitespace1 || whitespace2 {
// Two points for whitespace.
return 2
} else if nonAlphaNumeric1 || nonAlphaNumeric2 {
// One point for non-alphanumeric.
return 1
}
return 0
}
// DiffCleanupSemanticLossless looks for single edits surrounded on both sides by equalities which can be shifted sideways to align the edit to a word boundary.
// E.g: The c<ins>at c</ins>ame. -> The <ins>cat </ins>came.
func (dmp *DiffMatchPatch) DiffCleanupSemanticLossless(diffs []Diff) []Diff {
pointer := 1
// Intentionally ignore the first and last element (don't need checking).
for pointer < len(diffs)-1 {
if diffs[pointer-1].Type == DiffEqual &&
diffs[pointer+1].Type == DiffEqual {
// This is a single edit surrounded by equalities.
equality1 := diffs[pointer-1].Text
edit := diffs[pointer].Text
equality2 := diffs[pointer+1].Text
// First, shift the edit as far left as possible.
commonOffset := dmp.DiffCommonSuffix(equality1, edit)
if commonOffset > 0 {
commonString := edit[len(edit)-commonOffset:]
equality1 = equality1[0 : len(equality1)-commonOffset]
edit = commonString + edit[:len(edit)-commonOffset]
equality2 = commonString + equality2
}
// Second, step character by character right, looking for the best fit.
bestEquality1 := equality1
bestEdit := edit
bestEquality2 := equality2
bestScore := diffCleanupSemanticScore(equality1, edit) +
diffCleanupSemanticScore(edit, equality2)
for len(edit) != 0 && len(equality2) != 0 {
_, sz := utf8.DecodeRuneInString(edit)
if len(equality2) < sz || edit[:sz] != equality2[:sz] {
break
}
equality1 += edit[:sz]
edit = edit[sz:] + equality2[:sz]
equality2 = equality2[sz:]
score := diffCleanupSemanticScore(equality1, edit) +
diffCleanupSemanticScore(edit, equality2)
// The >= encourages trailing rather than leading whitespace on edits.
if score >= bestScore {
bestScore = score
bestEquality1 = equality1
bestEdit = edit
bestEquality2 = equality2
}
}
if diffs[pointer-1].Text != bestEquality1 {
// We have an improvement, save it back to the diff.
if len(bestEquality1) != 0 {
diffs[pointer-1].Text = bestEquality1
} else {
diffs = splice(diffs, pointer-1, 1)
pointer--
}
diffs[pointer].Text = bestEdit
if len(bestEquality2) != 0 {
diffs[pointer+1].Text = bestEquality2
} else {
diffs = append(diffs[:pointer+1], diffs[pointer+2:]...)
pointer--
}
}
}
pointer++
}
return diffs
}
// DiffCleanupEfficiency reduces the number of edits by eliminating operationally trivial equalities.
func (dmp *DiffMatchPatch) DiffCleanupEfficiency(diffs []Diff) []Diff {
changes := false
// Stack of indices where equalities are found.
type equality struct {
data int
next *equality
}
var equalities *equality
// Always equal to equalities[equalitiesLength-1][1]
lastequality := ""
pointer := 0 // Index of current position.
// Is there an insertion operation before the last equality.
preIns := false
// Is there a deletion operation before the last equality.
preDel := false
// Is there an insertion operation after the last equality.
postIns := false
// Is there a deletion operation after the last equality.
postDel := false
for pointer < len(diffs) {
if diffs[pointer].Type == DiffEqual { // Equality found.
if len(diffs[pointer].Text) < dmp.DiffEditCost &&
(postIns || postDel) {
// Candidate found.
equalities = &equality{
data: pointer,
next: equalities,
}
preIns = postIns
preDel = postDel
lastequality = diffs[pointer].Text
} else {
// Not a candidate, and can never become one.
equalities = nil
lastequality = ""
}
postIns = false
postDel = false
} else { // An insertion or deletion.
if diffs[pointer].Type == DiffDelete {
postDel = true
} else {
postIns = true
}
// Five types to be split:
// <ins>A</ins><del>B</del>XY<ins>C</ins><del>D</del>
// <ins>A</ins>X<ins>C</ins><del>D</del>
// <ins>A</ins><del>B</del>X<ins>C</ins>
// <ins>A</del>X<ins>C</ins><del>D</del>
// <ins>A</ins><del>B</del>X<del>C</del>
var sumPres int
if preIns {
sumPres++
}
if preDel {
sumPres++
}
if postIns {
sumPres++
}
if postDel {
sumPres++
}
if len(lastequality) > 0 &&
((preIns && preDel && postIns && postDel) ||
((len(lastequality) < dmp.DiffEditCost/2) && sumPres == 3)) {
insPoint := equalities.data
// Duplicate record.
diffs = splice(diffs, insPoint, 0, Diff{DiffDelete, lastequality})
// Change second copy to insert.
diffs[insPoint+1].Type = DiffInsert
// Throw away the equality we just deleted.
equalities = equalities.next
lastequality = ""
if preIns && preDel {
// No changes made which could affect previous entry, keep going.
postIns = true
postDel = true
equalities = nil
} else {
if equalities != nil {
equalities = equalities.next
}
if equalities != nil {
pointer = equalities.data
} else {
pointer = -1
}
postIns = false
postDel = false
}
changes = true
}
}
pointer++
}
if changes {
diffs = dmp.DiffCleanupMerge(diffs)
}
return diffs
}
// DiffCleanupMerge reorders and merges like edit sections. Merge equalities.
// Any edit section can move as long as it doesn't cross an equality.
func (dmp *DiffMatchPatch) DiffCleanupMerge(diffs []Diff) []Diff {
// Add a dummy entry at the end.
diffs = append(diffs, Diff{DiffEqual, ""})
pointer := 0
countDelete := 0
countInsert := 0
commonlength := 0
textDelete := []rune(nil)
textInsert := []rune(nil)
for pointer < len(diffs) {
switch diffs[pointer].Type {
case DiffInsert:
countInsert++
textInsert = append(textInsert, []rune(diffs[pointer].Text)...)
pointer++
break
case DiffDelete:
countDelete++
textDelete = append(textDelete, []rune(diffs[pointer].Text)...)
pointer++
break
case DiffEqual:
// Upon reaching an equality, check for prior redundancies.
if countDelete+countInsert > 1 {
if countDelete != 0 && countInsert != 0 {
// Factor out any common prefixies.
commonlength = commonPrefixLength(textInsert, textDelete)
if commonlength != 0 {
x := pointer - countDelete - countInsert
if x > 0 && diffs[x-1].Type == DiffEqual {
diffs[x-1].Text += string(textInsert[:commonlength])
} else {
diffs = append([]Diff{Diff{DiffEqual, string(textInsert[:commonlength])}}, diffs...)
pointer++
}
textInsert = textInsert[commonlength:]
textDelete = textDelete[commonlength:]
}
// Factor out any common suffixies.
commonlength = commonSuffixLength(textInsert, textDelete)
if commonlength != 0 {
insertIndex := len(textInsert) - commonlength
deleteIndex := len(textDelete) - commonlength
diffs[pointer].Text = string(textInsert[insertIndex:]) + diffs[pointer].Text
textInsert = textInsert[:insertIndex]
textDelete = textDelete[:deleteIndex]
}
}
// Delete the offending records and add the merged ones.
if countDelete == 0 {
diffs = splice(diffs, pointer-countInsert,
countDelete+countInsert,
Diff{DiffInsert, string(textInsert)})
} else if countInsert == 0 {
diffs = splice(diffs, pointer-countDelete,
countDelete+countInsert,
Diff{DiffDelete, string(textDelete)})
} else {
diffs = splice(diffs, pointer-countDelete-countInsert,
countDelete+countInsert,
Diff{DiffDelete, string(textDelete)},
Diff{DiffInsert, string(textInsert)})
}
pointer = pointer - countDelete - countInsert + 1
if countDelete != 0 {
pointer++
}
if countInsert != 0 {
pointer++
}
} else if pointer != 0 && diffs[pointer-1].Type == DiffEqual {
// Merge this equality with the previous one.
diffs[pointer-1].Text += diffs[pointer].Text
diffs = append(diffs[:pointer], diffs[pointer+1:]...)
} else {
pointer++
}
countInsert = 0
countDelete = 0
textDelete = nil
textInsert = nil
break
}
}
if len(diffs[len(diffs)-1].Text) == 0 {
diffs = diffs[0 : len(diffs)-1] // Remove the dummy entry at the end.
}
// Second pass: look for single edits surrounded on both sides by equalities which can be shifted sideways to eliminate an equality. E.g: A<ins>BA</ins>C -> <ins>AB</ins>AC
changes := false
pointer = 1
// Intentionally ignore the first and last element (don't need checking).
for pointer < (len(diffs) - 1) {
if diffs[pointer-1].Type == DiffEqual &&
diffs[pointer+1].Type == DiffEqual {
// This is a single edit surrounded by equalities.
if strings.HasSuffix(diffs[pointer].Text, diffs[pointer-1].Text) {
// Shift the edit over the previous equality.
diffs[pointer].Text = diffs[pointer-1].Text +
diffs[pointer].Text[:len(diffs[pointer].Text)-len(diffs[pointer-1].Text)]
diffs[pointer+1].Text = diffs[pointer-1].Text + diffs[pointer+1].Text
diffs = splice(diffs, pointer-1, 1)
changes = true
} else if strings.HasPrefix(diffs[pointer].Text, diffs[pointer+1].Text) {
// Shift the edit over the next equality.
diffs[pointer-1].Text += diffs[pointer+1].Text
diffs[pointer].Text =
diffs[pointer].Text[len(diffs[pointer+1].Text):] + diffs[pointer+1].Text
diffs = splice(diffs, pointer+1, 1)
changes = true
}
}
pointer++
}
// If shifts were made, the diff needs reordering and another shift sweep.
if changes {
diffs = dmp.DiffCleanupMerge(diffs)
}
return diffs
}
// DiffXIndex returns the equivalent location in s2.
func (dmp *DiffMatchPatch) DiffXIndex(diffs []Diff, loc int) int {
chars1 := 0
chars2 := 0
lastChars1 := 0
lastChars2 := 0
lastDiff := Diff{}
for i := 0; i < len(diffs); i++ {
aDiff := diffs[i]
if aDiff.Type != DiffInsert {
// Equality or deletion.
chars1 += len(aDiff.Text)
}
if aDiff.Type != DiffDelete {
// Equality or insertion.
chars2 += len(aDiff.Text)
}
if chars1 > loc {
// Overshot the location.
lastDiff = aDiff
break
}
lastChars1 = chars1
lastChars2 = chars2
}
if lastDiff.Type == DiffDelete {
// The location was deleted.
return lastChars2
}
// Add the remaining character length.
return lastChars2 + (loc - lastChars1)
}
// DiffPrettyHtml converts a []Diff into a pretty HTML report.
// It is intended as an example from which to write one's own display functions.
func (dmp *DiffMatchPatch) DiffPrettyHtml(diffs []Diff) string {
var buff bytes.Buffer
for _, diff := range diffs {
text := strings.Replace(html.EscapeString(diff.Text), "\n", "¶<br>", -1)
switch diff.Type {
case DiffInsert:
_, _ = buff.WriteString("<ins style=\"background:#e6ffe6;\">")
_, _ = buff.WriteString(text)
_, _ = buff.WriteString("</ins>")
case DiffDelete:
_, _ = buff.WriteString("<del style=\"background:#ffe6e6;\">")
_, _ = buff.WriteString(text)
_, _ = buff.WriteString("</del>")
case DiffEqual:
_, _ = buff.WriteString("<span>")
_, _ = buff.WriteString(text)
_, _ = buff.WriteString("</span>")
}
}
return buff.String()
}
// DiffPrettyText converts a []Diff into a colored text report.
func (dmp *DiffMatchPatch) DiffPrettyText(diffs []Diff) string {
var buff bytes.Buffer
for _, diff := range diffs {
text := diff.Text
switch diff.Type {
case DiffInsert:
_, _ = buff.WriteString("\x1b[32m")
_, _ = buff.WriteString(text)
_, _ = buff.WriteString("\x1b[0m")
case DiffDelete:
_, _ = buff.WriteString("\x1b[31m")
_, _ = buff.WriteString(text)
_, _ = buff.WriteString("\x1b[0m")
case DiffEqual:
_, _ = buff.WriteString(text)
}
}
return buff.String()
}
// DiffText1 computes and returns the source text (all equalities and deletions).
func (dmp *DiffMatchPatch) DiffText1(diffs []Diff) string {
//StringBuilder text = new StringBuilder()
var text bytes.Buffer
for _, aDiff := range diffs {
if aDiff.Type != DiffInsert {
_, _ = text.WriteString(aDiff.Text)
}
}
return text.String()
}
// DiffText2 computes and returns the destination text (all equalities and insertions).
func (dmp *DiffMatchPatch) DiffText2(diffs []Diff) string {
var text bytes.Buffer
for _, aDiff := range diffs {
if aDiff.Type != DiffDelete {
_, _ = text.WriteString(aDiff.Text)
}
}
return text.String()
}
// DiffLevenshtein computes the Levenshtein distance that is the number of inserted, deleted or substituted characters.
func (dmp *DiffMatchPatch) DiffLevenshtein(diffs []Diff) int {
levenshtein := 0
insertions := 0
deletions := 0
for _, aDiff := range diffs {
switch aDiff.Type {
case DiffInsert:
insertions += utf8.RuneCountInString(aDiff.Text)
case DiffDelete:
deletions += utf8.RuneCountInString(aDiff.Text)
case DiffEqual:
// A deletion and an insertion is one substitution.
levenshtein += max(insertions, deletions)
insertions = 0
deletions = 0
}
}
levenshtein += max(insertions, deletions)
return levenshtein
}
// DiffToDelta crushes the diff into an encoded string which describes the operations required to transform text1 into text2.
// E.g. =3\t-2\t+ing -> Keep 3 chars, delete 2 chars, insert 'ing'. Operations are tab-separated. Inserted text is escaped using %xx notation.
func (dmp *DiffMatchPatch) DiffToDelta(diffs []Diff) string {
var text bytes.Buffer
for _, aDiff := range diffs {
switch aDiff.Type {
case DiffInsert:
_, _ = text.WriteString("+")
_, _ = text.WriteString(strings.Replace(url.QueryEscape(aDiff.Text), "+", " ", -1))
_, _ = text.WriteString("\t")
break
case DiffDelete:
_, _ = text.WriteString("-")
_, _ = text.WriteString(strconv.Itoa(utf8.RuneCountInString(aDiff.Text)))
_, _ = text.WriteString("\t")
break
case DiffEqual:
_, _ = text.WriteString("=")
_, _ = text.WriteString(strconv.Itoa(utf8.RuneCountInString(aDiff.Text)))
_, _ = text.WriteString("\t")
break
}
}
delta := text.String()
if len(delta) != 0 {
// Strip off trailing tab character.
delta = delta[0 : utf8.RuneCountInString(delta)-1]
delta = unescaper.Replace(delta)
}
return delta
}
// DiffFromDelta given the original text1, and an encoded string which describes the operations required to transform text1 into text2, comAdde the full diff.
func (dmp *DiffMatchPatch) DiffFromDelta(text1 string, delta string) (diffs []Diff, err error) {
i := 0
runes := []rune(text1)
for _, token := range strings.Split(delta, "\t") {
if len(token) == 0 {
// Blank tokens are ok (from a trailing \t).
continue
}
// Each token begins with a one character parameter which specifies the operation of this token (delete, insert, equality).
param := token[1:]
switch op := token[0]; op {
case '+':
// Decode would Diff all "+" to " "
param = strings.Replace(param, "+", "%2b", -1)
param, err = url.QueryUnescape(param)
if err != nil {
return nil, err
}
if !utf8.ValidString(param) {
return nil, fmt.Errorf("invalid UTF-8 token: %q", param)
}
diffs = append(diffs, Diff{DiffInsert, param})
case '=', '-':
n, err := strconv.ParseInt(param, 10, 0)
if err != nil {
return nil, err
} else if n < 0 {
return nil, errors.New("Negative number in DiffFromDelta: " + param)
}
i += int(n)
// Break out if we are out of bounds, go1.6 can't handle this very well
if i > len(runes) {
break
}
// Remember that string slicing is by byte - we want by rune here.
text := string(runes[i-int(n) : i])
if op == '=' {
diffs = append(diffs, Diff{DiffEqual, text})
} else {
diffs = append(diffs, Diff{DiffDelete, text})
}
default:
// Anything else is an error.
return nil, errors.New("Invalid diff operation in DiffFromDelta: " + string(token[0]))
}
}
if i != len(runes) {
return nil, fmt.Errorf("Delta length (%v) is different from source text length (%v)", i, len(text1))
}
return diffs, nil
}