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
|
diff -ur icu.org/source/common/dictbe.cpp icu/source/common/dictbe.cpp
--- icu.org/source/common/dictbe.cpp 2018-10-02 00:39:56.000000000 +0200
+++ icu/source/common/dictbe.cpp 2018-10-20 00:14:46.462039038 +0200
@@ -29,7 +29,19 @@
******************************************************************
*/
-DictionaryBreakEngine::DictionaryBreakEngine() {
+DictionaryBreakEngine::DictionaryBreakEngine()
+ : fTypes(0), clusterLimit(0) {
+}
+
+DictionaryBreakEngine::DictionaryBreakEngine(uint32_t breakTypes)
+ : fTypes(breakTypes), clusterLimit(3) {
+ UErrorCode status = U_ZERO_ERROR;
+ fViramaSet.applyPattern(UNICODE_STRING_SIMPLE("[[:ccc=VR:]]"), status);
+
+ // note Skip Sets contain fIgnoreSet characters too.
+ fSkipStartSet.applyPattern(UNICODE_STRING_SIMPLE("[[:lb=OP:][:lb=QU:]\\u200C\\u200D\\u2060]"), status);
+ fSkipEndSet.applyPattern(UNICODE_STRING_SIMPLE("[[:lb=CP:][:lb=QU:][:lb=EX:][:lb=CL:]\\u200C\\u200D\\u2060]"), status);
+ fNBeforeSet.applyPattern(UNICODE_STRING_SIMPLE("[[:lb=CR:][:lb=LF:][:lb=NL:][:lb=SP:][:lb=ZW:][:lb=IS:][:lb=BA:][:lb=NS:]]"), status);
}
DictionaryBreakEngine::~DictionaryBreakEngine() {
@@ -76,6 +88,169 @@
fSet.compact();
}
+bool
+DictionaryBreakEngine::scanBeforeStart(UText *text, int32_t& start, bool &doBreak) const {
+ UErrorCode status = U_ZERO_ERROR;
+ UText* ut = utext_clone(NULL, text, false, true, &status);
+ utext_setNativeIndex(ut, start);
+ UChar32 c = utext_current32(ut);
+ bool res = false;
+ doBreak = true;
+ while (start >= 0) {
+ if (!fSkipStartSet.contains(c)) {
+ res = (c == ZWSP);
+ break;
+ }
+ --start;
+ c = utext_previous32(ut);
+ doBreak = false;
+ }
+ utext_close(ut);
+ return res;
+}
+
+bool
+DictionaryBreakEngine::scanAfterEnd(UText *text, int32_t textEnd, int32_t& end, bool &doBreak) const {
+ UErrorCode status = U_ZERO_ERROR;
+ UText* ut = utext_clone(NULL, text, false, true, &status);
+ utext_setNativeIndex(ut, end);
+ UChar32 c = utext_current32(ut);
+ bool res = false;
+ doBreak = !fNBeforeSet.contains(c);
+ while (end < textEnd) {
+ if (!fSkipEndSet.contains(c)) {
+ res = (c == ZWSP);
+ break;
+ }
+ ++end;
+ c = utext_next32(ut);
+ doBreak = false;
+ }
+ utext_close(ut);
+ return res;
+}
+
+void
+DictionaryBreakEngine::scanBackClusters(UText *text, int32_t textStart, int32_t& start) const {
+ UChar32 c = 0;
+ start = utext_getNativeIndex(text);
+ while (start > textStart) {
+ c = utext_previous32(text);
+ --start;
+ if (!fSkipEndSet.contains(c))
+ break;
+ }
+ for (int i = 0; i < clusterLimit; ++i) { // scan backwards clusterLimit clusters
+ while (start > textStart) {
+ while (fIgnoreSet.contains(c))
+ c = utext_previous32(text);
+ if (!fMarkSet.contains(c)) {
+ if (fBaseSet.contains(c)) {
+ c = utext_previous32(text);
+ if (!fViramaSet.contains(c)) { // Virama (e.g. coeng) preceding base. Treat sequence as a mark
+ utext_next32(text);
+ c = utext_current32(text);
+ break;
+ } else {
+ --start;
+ }
+ } else {
+ break;
+ }
+ }
+ c = utext_previous32(text);
+ --start;
+ }
+ if (!fBaseSet.contains(c) || start < textStart) { // not a cluster start so finish
+ break;
+ }
+ c = utext_previous32(text);
+ --start; // go round again
+ } // ignore hitting previous inhibitor since scanning for it should have found us!
+ ++start; // counteract --before
+}
+
+void
+DictionaryBreakEngine::scanFwdClusters(UText *text, int32_t textEnd, int32_t& end) const {
+ UChar32 c = utext_current32(text);
+ end = utext_getNativeIndex(text);
+ while (end < textEnd) {
+ if (!fSkipStartSet.contains(c))
+ break;
+ utext_next32(text);
+ c = utext_current32(text);
+ ++end;
+ }
+ for (int i = 0; i < clusterLimit; ++i) { // scan forwards clusterLimit clusters
+ while (fIgnoreSet.contains(c)) {
+ utext_next32(text);
+ c = utext_current32(text);
+ }
+ if (fBaseSet.contains(c)) {
+ while (end < textEnd) {
+ utext_next32(text);
+ c = utext_current32(text);
+ ++end;
+ if (!fMarkSet.contains(c))
+ break;
+ else if (fViramaSet.contains(c)) { // handle coeng + base as mark
+ utext_next32(text);
+ c = utext_current32(text);
+ ++end;
+ if (!fBaseSet.contains(c))
+ break;
+ }
+ }
+ } else {
+ --end; // bad char so break after char before it
+ break;
+ }
+ }
+}
+
+bool
+DictionaryBreakEngine::scanWJ(UText *text, int32_t &start, int32_t end, int32_t &before, int32_t &after) const {
+ UErrorCode status = U_ZERO_ERROR;
+ UText* ut = utext_clone(NULL, text, false, true, &status);
+ int32_t nat = start;
+ utext_setNativeIndex(ut, nat);
+ bool foundFirst = true;
+ int32_t curr = start;
+ while (nat < end) {
+ UChar32 c = utext_current32(ut);
+ if (c == ZWSP || c == WJ) {
+ curr = nat + 1;
+ if (foundFirst) // only scan backwards for first inhibitor
+ scanBackClusters(ut, start, before);
+ foundFirst = false; // don't scan backwards if we go around again. Also marks found something
+
+ utext_next32(ut);
+ scanFwdClusters(ut, end, after);
+ nat = after + 1;
+
+ if (c == ZWSP || c == WJ) { // did we hit another one?
+ continue;
+ } else {
+ break;
+ }
+ }
+
+ ++nat; // keep hunting
+ utext_next32(ut);
+ }
+
+ utext_close(ut);
+
+ if (nat >= end && foundFirst) {
+ start = before = after = nat;
+ return false; // failed to find anything
+ }
+ else {
+ start = curr;
+ }
+ return true; // yup hit one
+}
+
/*
******************************************************************
* PossibleWord
@@ -103,35 +278,35 @@
public:
PossibleWord() : count(0), prefix(0), offset(-1), mark(0), current(0) {};
~PossibleWord() {};
-
+
// Fill the list of candidates if needed, select the longest, and return the number found
- int32_t candidates( UText *text, DictionaryMatcher *dict, int32_t rangeEnd );
-
+ int32_t candidates( UText *text, DictionaryMatcher *dict, int32_t rangeEnd, UnicodeSet const *ignoreSet = NULL, int32_t minLength = 0 );
+
// Select the currently marked candidate, point after it in the text, and invalidate self
int32_t acceptMarked( UText *text );
-
+
// Back up from the current candidate to the next shorter one; return TRUE if that exists
// and point the text after it
UBool backUp( UText *text );
-
+
// Return the longest prefix this candidate location shares with a dictionary word
// Return value is in code points.
int32_t longestPrefix() { return prefix; };
-
+
// Mark the current candidate as the one we like
void markCurrent() { mark = current; };
-
+
// Get length in code points of the marked word.
int32_t markedCPLength() { return cpLengths[mark]; };
};
-int32_t PossibleWord::candidates( UText *text, DictionaryMatcher *dict, int32_t rangeEnd ) {
+int32_t PossibleWord::candidates( UText *text, DictionaryMatcher *dict, int32_t rangeEnd, UnicodeSet const *ignoreSet, int32_t minLength) {
// TODO: If getIndex is too slow, use offset < 0 and add discardAll()
int32_t start = (int32_t)utext_getNativeIndex(text);
if (start != offset) {
offset = start;
- count = dict->matches(text, rangeEnd-start, UPRV_LENGTHOF(cuLengths), cuLengths, cpLengths, NULL, &prefix);
+ count = dict->matches(text, rangeEnd-start, UPRV_LENGTHOF(cuLengths), cuLengths, cpLengths, NULL, &prefix, ignoreSet, minLength);
// Dictionary leaves text after longest prefix, not longest word. Back up.
if (count <= 0) {
utext_setNativeIndex(text, start);
@@ -803,51 +978,28 @@
* KhmerBreakEngine
*/
-// How many words in a row are "good enough"?
-static const int32_t KHMER_LOOKAHEAD = 3;
-
-// Will not combine a non-word with a preceding dictionary word longer than this
-static const int32_t KHMER_ROOT_COMBINE_THRESHOLD = 3;
-
-// Will not combine a non-word that shares at least this much prefix with a
-// dictionary word, with a preceding word
-static const int32_t KHMER_PREFIX_COMBINE_THRESHOLD = 3;
-
-// Minimum word size
-static const int32_t KHMER_MIN_WORD = 2;
-
-// Minimum number of characters for two words
-static const int32_t KHMER_MIN_WORD_SPAN = KHMER_MIN_WORD * 2;
-
KhmerBreakEngine::KhmerBreakEngine(DictionaryMatcher *adoptDictionary, UErrorCode &status)
- : DictionaryBreakEngine(),
+ : DictionaryBreakEngine((1 << UBRK_WORD) | (1 << UBRK_LINE)),
fDictionary(adoptDictionary)
{
- fKhmerWordSet.applyPattern(UNICODE_STRING_SIMPLE("[[:Khmr:]&[:LineBreak=SA:]]"), status);
+
+ clusterLimit = 3;
+
+ fKhmerWordSet.applyPattern(UNICODE_STRING_SIMPLE("[[:Khmr:]\\u2060\\u200C\\u200D]"), status);
if (U_SUCCESS(status)) {
setCharacters(fKhmerWordSet);
}
fMarkSet.applyPattern(UNICODE_STRING_SIMPLE("[[:Khmr:]&[:LineBreak=SA:]&[:M:]]"), status);
- fMarkSet.add(0x0020);
- fEndWordSet = fKhmerWordSet;
- fBeginWordSet.add(0x1780, 0x17B3);
- //fBeginWordSet.add(0x17A3, 0x17A4); // deprecated vowels
- //fEndWordSet.remove(0x17A5, 0x17A9); // Khmer independent vowels that can't end a word
- //fEndWordSet.remove(0x17B2); // Khmer independent vowel that can't end a word
- fEndWordSet.remove(0x17D2); // KHMER SIGN COENG that combines some following characters
- //fEndWordSet.remove(0x17B6, 0x17C5); // Remove dependent vowels
-// fEndWordSet.remove(0x0E31); // MAI HAN-AKAT
-// fEndWordSet.remove(0x0E40, 0x0E44); // SARA E through SARA AI MAIMALAI
-// fBeginWordSet.add(0x0E01, 0x0E2E); // KO KAI through HO NOKHUK
-// fBeginWordSet.add(0x0E40, 0x0E44); // SARA E through SARA AI MAIMALAI
-// fSuffixSet.add(THAI_PAIYANNOI);
-// fSuffixSet.add(THAI_MAIYAMOK);
+ fIgnoreSet.add(0x2060); // WJ
+ fIgnoreSet.add(0x200C, 0x200D); // ZWJ, ZWNJ
+ fBaseSet.applyPattern(UNICODE_STRING_SIMPLE("[[:Khmr:]&[:lb=SA:]&[:^M:]]"), status);
+ fPuncSet.applyPattern(UNICODE_STRING_SIMPLE("[\\u17D4\\u17D5\\u17D6\\u17D7\\u17D9:]"), status);
// Compact for caching.
fMarkSet.compact();
- fEndWordSet.compact();
- fBeginWordSet.compact();
-// fSuffixSet.compact();
+ fIgnoreSet.compact();
+ fBaseSet.compact();
+ fPuncSet.compact();
}
KhmerBreakEngine::~KhmerBreakEngine() {
@@ -859,180 +1011,204 @@
int32_t rangeStart,
int32_t rangeEnd,
UVector32 &foundBreaks ) const {
- if ((rangeEnd - rangeStart) < KHMER_MIN_WORD_SPAN) {
- return 0; // Not enough characters for two words
- }
-
- uint32_t wordsFound = 0;
- int32_t cpWordLength = 0;
- int32_t cuWordLength = 0;
- int32_t current;
+ uint32_t wordsFound = foundBreaks.size();
UErrorCode status = U_ZERO_ERROR;
- PossibleWord words[KHMER_LOOKAHEAD];
-
+ int32_t before = 0;
+ int32_t after = 0;
+ int32_t finalBefore = 0;
+ int32_t initAfter = 0;
+ int32_t scanStart = rangeStart;
+ int32_t scanEnd = rangeEnd;
+
+ bool startZwsp = false;
+ bool breakStart = false;
+ bool breakEnd = false;
+
+ if (rangeStart > 0) {
+ --scanStart;
+ startZwsp = scanBeforeStart(text, scanStart, breakStart);
+ }
utext_setNativeIndex(text, rangeStart);
+ scanFwdClusters(text, rangeEnd, initAfter);
+ bool endZwsp = scanAfterEnd(text, utext_nativeLength(text), scanEnd, breakEnd);
+ utext_setNativeIndex(text, rangeEnd - 1);
+ scanBackClusters(text, rangeStart, finalBefore);
+ if (finalBefore < initAfter) { // the whole run is tented so no breaks
+ if (breakStart || fTypes < UBRK_LINE)
+ foundBreaks.push(rangeStart, status);
+ if (breakEnd || fTypes < UBRK_LINE)
+ foundBreaks.push(rangeEnd, status);
+ return foundBreaks.size() - wordsFound;
+ }
- while (U_SUCCESS(status) && (current = (int32_t)utext_getNativeIndex(text)) < rangeEnd) {
- cuWordLength = 0;
- cpWordLength = 0;
-
- // Look for candidate words at the current position
- int32_t candidates = words[wordsFound%KHMER_LOOKAHEAD].candidates(text, fDictionary, rangeEnd);
-
- // If we found exactly one, use that
- if (candidates == 1) {
- cuWordLength = words[wordsFound % KHMER_LOOKAHEAD].acceptMarked(text);
- cpWordLength = words[wordsFound % KHMER_LOOKAHEAD].markedCPLength();
- wordsFound += 1;
- }
+ scanStart = rangeStart;
+ scanWJ(text, scanStart, rangeEnd, before, after);
+ if (startZwsp || initAfter >= before) {
+ after = initAfter;
+ before = 0;
+ }
+ if (!endZwsp && after > finalBefore && after < rangeEnd)
+ endZwsp = true;
+ if (endZwsp && before > finalBefore)
+ before = finalBefore;
- // If there was more than one, see which one can take us forward the most words
- else if (candidates > 1) {
- // If we're already at the end of the range, we're done
- if ((int32_t)utext_getNativeIndex(text) >= rangeEnd) {
- goto foundBest;
- }
- do {
- int32_t wordsMatched = 1;
- if (words[(wordsFound + 1) % KHMER_LOOKAHEAD].candidates(text, fDictionary, rangeEnd) > 0) {
- if (wordsMatched < 2) {
- // Followed by another dictionary word; mark first word as a good candidate
- words[wordsFound % KHMER_LOOKAHEAD].markCurrent();
- wordsMatched = 2;
- }
+ utext_setNativeIndex(text, rangeStart);
+ int32_t numCodePts = rangeEnd - rangeStart;
+ // bestSnlp[i] is the snlp of the best segmentation of the first i
+ // code points in the range to be matched.
+ UVector32 bestSnlp(numCodePts + 1, status);
+ bestSnlp.addElement(0, status);
+ for(int32_t i = 1; i <= numCodePts; i++) {
+ bestSnlp.addElement(kuint32max, status);
+ }
- // If we're already at the end of the range, we're done
- if ((int32_t)utext_getNativeIndex(text) >= rangeEnd) {
- goto foundBest;
- }
+ // prev[i] is the index of the last code point in the previous word in
+ // the best segmentation of the first i characters. Note negative implies
+ // that the code point is part of an unknown word.
+ UVector32 prev(numCodePts + 1, status);
+ for(int32_t i = 0; i <= numCodePts; i++) {
+ prev.addElement(kuint32max, status);
+ }
- // See if any of the possible second words is followed by a third word
- do {
- // If we find a third word, stop right away
- if (words[(wordsFound + 2) % KHMER_LOOKAHEAD].candidates(text, fDictionary, rangeEnd)) {
- words[wordsFound % KHMER_LOOKAHEAD].markCurrent();
- goto foundBest;
- }
- }
- while (words[(wordsFound + 1) % KHMER_LOOKAHEAD].backUp(text));
- }
+ const int32_t maxWordSize = 20;
+ UVector32 values(maxWordSize, status);
+ values.setSize(maxWordSize);
+ UVector32 lengths(maxWordSize, status);
+ lengths.setSize(maxWordSize);
+
+ // Dynamic programming to find the best segmentation.
+
+ // In outer loop, i is the code point index,
+ // ix is the corresponding string (code unit) index.
+ // They differ when the string contains supplementary characters.
+ int32_t ix = rangeStart;
+ for (int32_t i = 0; i < numCodePts; ++i, utext_setNativeIndex(text, ++ix)) {
+ if ((uint32_t)bestSnlp.elementAti(i) == kuint32max) {
+ continue;
+ }
+
+ int32_t count;
+ count = fDictionary->matches(text, numCodePts - i, maxWordSize,
+ NULL, lengths.getBuffer(), values.getBuffer(), NULL, &fIgnoreSet, 2);
+ // Note: lengths is filled with code point lengths
+ // The NULL parameter is the ignored code unit lengths.
+
+ for (int32_t j = 0; j < count; j++) {
+ int32_t ln = lengths.elementAti(j);
+ if (ln + i >= numCodePts)
+ continue;
+ utext_setNativeIndex(text, ln+ix);
+ int32_t c = utext_current32(text);
+ if (fMarkSet.contains(c) || c == 0x17D2) { // Coeng
+ lengths.removeElementAt(j);
+ values.removeElementAt(j);
+ --j;
+ --count;
}
- while (words[wordsFound % KHMER_LOOKAHEAD].backUp(text));
-foundBest:
- cuWordLength = words[wordsFound % KHMER_LOOKAHEAD].acceptMarked(text);
- cpWordLength = words[wordsFound % KHMER_LOOKAHEAD].markedCPLength();
- wordsFound += 1;
}
-
- // We come here after having either found a word or not. We look ahead to the
- // next word. If it's not a dictionary word, we will combine it with the word we
- // just found (if there is one), but only if the preceding word does not exceed
- // the threshold.
- // The text iterator should now be positioned at the end of the word we found.
- if ((int32_t)utext_getNativeIndex(text) < rangeEnd && cpWordLength < KHMER_ROOT_COMBINE_THRESHOLD) {
- // if it is a dictionary word, do nothing. If it isn't, then if there is
- // no preceding word, or the non-word shares less than the minimum threshold
- // of characters with a dictionary word, then scan to resynchronize
- if (words[wordsFound % KHMER_LOOKAHEAD].candidates(text, fDictionary, rangeEnd) <= 0
- && (cuWordLength == 0
- || words[wordsFound % KHMER_LOOKAHEAD].longestPrefix() < KHMER_PREFIX_COMBINE_THRESHOLD)) {
- // Look for a plausible word boundary
- int32_t remaining = rangeEnd - (current+cuWordLength);
- UChar32 pc;
- UChar32 uc;
- int32_t chars = 0;
- for (;;) {
- int32_t pcIndex = (int32_t)utext_getNativeIndex(text);
- pc = utext_next32(text);
- int32_t pcSize = (int32_t)utext_getNativeIndex(text) - pcIndex;
- chars += pcSize;
- remaining -= pcSize;
- if (remaining <= 0) {
+ if (count == 0) {
+ utext_setNativeIndex(text, ix);
+ int32_t c = utext_current32(text);
+ if (fPuncSet.contains(c) || fIgnoreSet.contains(c) || c == ZWSP) {
+ values.setElementAt(0, count);
+ lengths.setElementAt(1, count++);
+ } else if (fBaseSet.contains(c)) {
+ int32_t currix = utext_getNativeIndex(text);
+ do {
+ utext_next32(text);
+ c = utext_current32(text);
+ if (utext_getNativeIndex(text) >= rangeEnd)
break;
- }
- uc = utext_current32(text);
- if (fEndWordSet.contains(pc) && fBeginWordSet.contains(uc)) {
- // Maybe. See if it's in the dictionary.
- int32_t num_candidates = words[(wordsFound + 1) % KHMER_LOOKAHEAD].candidates(text, fDictionary, rangeEnd);
- utext_setNativeIndex(text, current+cuWordLength+chars);
- if (num_candidates > 0) {
+ if (c == 0x17D2) { // Coeng
+ utext_next32(text);
+ c = utext_current32(text);
+ if (!fBaseSet.contains(c) || utext_getNativeIndex(text) >= rangeEnd) {
break;
+ } else {
+ utext_next32(text);
+ c = utext_current32(text);
+ if (utext_getNativeIndex(text) >= rangeEnd)
+ break;
}
}
- }
-
- // Bump the word count if there wasn't already one
- if (cuWordLength <= 0) {
- wordsFound += 1;
- }
+ } while (fMarkSet.contains(c) || fIgnoreSet.contains(c));
+ values.setElementAt(BADSNLP, count);
+ lengths.setElementAt(utext_getNativeIndex(text) - currix, count++);
+ } else {
+ values.setElementAt(BADSNLP, count);
+ lengths.setElementAt(1, count++);
+ }
+ }
- // Update the length with the passed-over characters
- cuWordLength += chars;
+ for (int32_t j = 0; j < count; j++) {
+ uint32_t v = values.elementAti(j);
+ int32_t newSnlp = bestSnlp.elementAti(i) + v;
+ int32_t ln = lengths.elementAti(j);
+ utext_setNativeIndex(text, ln+ix);
+ int32_t c = utext_current32(text);
+ while ((fPuncSet.contains(c) || fIgnoreSet.contains(c)) && ln + i < numCodePts) {
+ ++ln;
+ utext_next32(text);
+ c = utext_current32(text);
}
- else {
- // Back up to where we were for next iteration
- utext_setNativeIndex(text, current+cuWordLength);
+ int32_t ln_j_i = ln + i; // yes really i!
+ if (newSnlp < bestSnlp.elementAti(ln_j_i)) {
+ if (v == BADSNLP) {
+ int32_t p = prev.elementAti(i);
+ if (p < 0)
+ prev.setElementAt(p, ln_j_i);
+ else
+ prev.setElementAt(-i, ln_j_i);
+ }
+ else
+ prev.setElementAt(i, ln_j_i);
+ bestSnlp.setElementAt(newSnlp, ln_j_i);
}
}
-
- // Never stop before a combining mark.
- int32_t currPos;
- while ((currPos = (int32_t)utext_getNativeIndex(text)) < rangeEnd && fMarkSet.contains(utext_current32(text))) {
- utext_next32(text);
- cuWordLength += (int32_t)utext_getNativeIndex(text) - currPos;
+ }
+ // Start pushing the optimal offset index into t_boundary (t for tentative).
+ // prev[numCodePts] is guaranteed to be meaningful.
+ // We'll first push in the reverse order, i.e.,
+ // t_boundary[0] = numCodePts, and afterwards do a swap.
+ UVector32 t_boundary(numCodePts+1, status);
+
+ int32_t numBreaks = 0;
+ // No segmentation found, set boundary to end of range
+ while (numCodePts >= 0 && (uint32_t)bestSnlp.elementAti(numCodePts) == kuint32max) {
+ --numCodePts;
+ }
+ if (numCodePts < 0) {
+ t_boundary.addElement(numCodePts, status);
+ numBreaks++;
+ } else {
+ for (int32_t i = numCodePts; (uint32_t)i != kuint32max; i = prev.elementAti(i)) {
+ if (i < 0) i = -i;
+ t_boundary.addElement(i, status);
+ numBreaks++;
}
+ U_ASSERT(prev.elementAti(t_boundary.elementAti(numBreaks - 1)) == 0);
+ }
- // Look ahead for possible suffixes if a dictionary word does not follow.
- // We do this in code rather than using a rule so that the heuristic
- // resynch continues to function. For example, one of the suffix characters
- // could be a typo in the middle of a word.
-// if ((int32_t)utext_getNativeIndex(text) < rangeEnd && wordLength > 0) {
-// if (words[wordsFound%KHMER_LOOKAHEAD].candidates(text, fDictionary, rangeEnd) <= 0
-// && fSuffixSet.contains(uc = utext_current32(text))) {
-// if (uc == KHMER_PAIYANNOI) {
-// if (!fSuffixSet.contains(utext_previous32(text))) {
-// // Skip over previous end and PAIYANNOI
-// utext_next32(text);
-// utext_next32(text);
-// wordLength += 1; // Add PAIYANNOI to word
-// uc = utext_current32(text); // Fetch next character
-// }
-// else {
-// // Restore prior position
-// utext_next32(text);
-// }
-// }
-// if (uc == KHMER_MAIYAMOK) {
-// if (utext_previous32(text) != KHMER_MAIYAMOK) {
-// // Skip over previous end and MAIYAMOK
-// utext_next32(text);
-// utext_next32(text);
-// wordLength += 1; // Add MAIYAMOK to word
-// }
-// else {
-// // Restore prior position
-// utext_next32(text);
-// }
-// }
-// }
-// else {
-// utext_setNativeIndex(text, current+wordLength);
-// }
-// }
-
- // Did we find a word on this iteration? If so, push it on the break stack
- if (cuWordLength > 0) {
- foundBreaks.push((current+cuWordLength), status);
+ // Now that we're done, convert positions in t_boundary[] (indices in
+ // the normalized input string) back to indices in the original input UText
+ // while reversing t_boundary and pushing values to foundBreaks.
+ for (int32_t i = numBreaks-1; i >= 0; i--) {
+ int32_t cpPos = t_boundary.elementAti(i);
+ if (cpPos == 0 && !breakStart && fTypes >= UBRK_LINE) continue;
+ int32_t utextPos = cpPos + rangeStart;
+ while (utextPos > after && scanWJ(text, utextPos, scanEnd, before, after));
+ if (utextPos < before) {
+ // Boundaries are added to foundBreaks output in ascending order.
+ U_ASSERT(foundBreaks.size() == 0 ||foundBreaks.peeki() < utextPos);
+ foundBreaks.push(utextPos, status);
}
}
-
+
// Don't return a break for the end of the dictionary range if there is one there.
- if (foundBreaks.peeki() >= rangeEnd) {
+ if (!breakEnd && fTypes >= UBRK_LINE && foundBreaks.peeki() >= rangeEnd) {
(void) foundBreaks.popi();
- wordsFound -= 1;
}
-
- return wordsFound;
+ return foundBreaks.size() - wordsFound;
}
#if !UCONFIG_NO_NORMALIZATION
diff -ur icu.org/source/common/dictbe.h icu/source/common/dictbe.h
--- icu.org/source/common/dictbe.h 2018-09-29 02:34:41.000000000 +0200
+++ icu/source/common/dictbe.h 2018-10-19 14:21:00.339942804 +0200
@@ -34,7 +34,8 @@
* threads without synchronization.</p>
*/
class DictionaryBreakEngine : public LanguageBreakEngine {
- private:
+ protected:
+
/**
* The set of characters handled by this engine
* @internal
@@ -42,14 +43,84 @@
UnicodeSet fSet;
+ const int32_t WJ = 0x2060;
+ const int32_t ZWSP = 0x200B;
+
+ /**
+ * The break types it was constructed with
+ * @internal
+ */
+ uint32_t fTypes;
+
+ /**
+ * A Unicode set of all viramas
+ * @internal
+ */
+ UnicodeSet fViramaSet;
+
+ /**
+ * A Unicode set of all base characters
+ * @internal
+ */
+ UnicodeSet fBaseSet;
+
+ /**
+ * A Unicode set of all marks
+ * @internal
+ */
+ UnicodeSet fMarkSet;
+
+ /**
+ * A Unicode set of all characters ignored ignored in dictionary matching
+ * @internal
+ */
+ UnicodeSet fIgnoreSet;
+
+ /**
+ * A Unicode set of all characters ignored ignored in dictionary matching
+ * @internal
+ */
+ UnicodeSet fSkipStartSet;
+
+ /**
+ * A Unicode set of all characters ignored ignored in dictionary matching
+ * @internal
+ */
+ UnicodeSet fSkipEndSet;
+
+ /**
+ * A Unicode set of all characters that should not be broken before
+ * @internal
+ */
+ UnicodeSet fNBeforeSet;
+
+ /**
+ * The number of clusters within which breaks are inhibited
+ * @internal
+ */
+ int32_t clusterLimit;
+
+ bool scanWJ(UText *text, int32_t &start, int32_t end, int32_t &before, int32_t &after) const;
+
+ bool scanBeforeStart(UText *text, int32_t& start, bool &doBreak) const;
+ bool scanAfterEnd(UText *text, int32_t rangeEnd, int32_t& end, bool &doBreak) const;
+ void scanBackClusters(UText *text, int32_t textStart, int32_t& start) const;
+ void scanFwdClusters(UText *text, int32_t textEnd, int32_t& end) const;
+
public:
/**
- * <p>Constructor </p>
+ * <p>Default constructor.</p>
+ *
*/
DictionaryBreakEngine();
/**
+ * <p>Constructor with break types.</p>
+ */
+ explicit DictionaryBreakEngine(uint32_t breakTypes);
+
+ /**
* <p>Virtual destructor.</p>
*/
virtual ~DictionaryBreakEngine();
@@ -68,7 +139,7 @@
* <p>Find any breaks within a run in the supplied text.</p>
*
* @param text A UText representing the text. The iterator is left at
- * the end of the run of characters which the engine is capable of handling
+ * the end of the run of characters which the engine is capable of handling
* that starts from the first character in the range.
* @param startPos The start of the run within the supplied text.
* @param endPos The end of the run within the supplied text.
@@ -218,118 +289,120 @@
};
-/*******************************************************************
- * BurmeseBreakEngine
- */
-
-/**
- * <p>BurmeseBreakEngine is a kind of DictionaryBreakEngine that uses a
- * DictionaryMatcher and heuristics to determine Burmese-specific breaks.</p>
- *
- * <p>After it is constructed a BurmeseBreakEngine may be shared between
- * threads without synchronization.</p>
- */
-class BurmeseBreakEngine : public DictionaryBreakEngine {
- private:
- /**
- * The set of characters handled by this engine
- * @internal
- */
-
- UnicodeSet fBurmeseWordSet;
- UnicodeSet fEndWordSet;
- UnicodeSet fBeginWordSet;
- UnicodeSet fMarkSet;
- DictionaryMatcher *fDictionary;
-
- public:
-
- /**
- * <p>Default constructor.</p>
- *
- * @param adoptDictionary A DictionaryMatcher to adopt. Deleted when the
- * engine is deleted.
- */
- BurmeseBreakEngine(DictionaryMatcher *adoptDictionary, UErrorCode &status);
-
- /**
- * <p>Virtual destructor.</p>
- */
- virtual ~BurmeseBreakEngine();
-
- protected:
- /**
- * <p>Divide up a range of known dictionary characters.</p>
- *
- * @param text A UText representing the text
- * @param rangeStart The start of the range of dictionary characters
- * @param rangeEnd The end of the range of dictionary characters
- * @param foundBreaks Output of C array of int32_t break positions, or 0
- * @return The number of breaks found
- */
- virtual int32_t divideUpDictionaryRange( UText *text,
- int32_t rangeStart,
- int32_t rangeEnd,
- UVector32 &foundBreaks ) const;
-
-};
-
-/*******************************************************************
- * KhmerBreakEngine
- */
-
-/**
- * <p>KhmerBreakEngine is a kind of DictionaryBreakEngine that uses a
- * DictionaryMatcher and heuristics to determine Khmer-specific breaks.</p>
- *
- * <p>After it is constructed a KhmerBreakEngine may be shared between
- * threads without synchronization.</p>
- */
-class KhmerBreakEngine : public DictionaryBreakEngine {
- private:
- /**
- * The set of characters handled by this engine
- * @internal
- */
-
- UnicodeSet fKhmerWordSet;
- UnicodeSet fEndWordSet;
- UnicodeSet fBeginWordSet;
- UnicodeSet fMarkSet;
- DictionaryMatcher *fDictionary;
-
- public:
-
- /**
- * <p>Default constructor.</p>
- *
- * @param adoptDictionary A DictionaryMatcher to adopt. Deleted when the
- * engine is deleted.
- */
- KhmerBreakEngine(DictionaryMatcher *adoptDictionary, UErrorCode &status);
-
- /**
- * <p>Virtual destructor.</p>
- */
- virtual ~KhmerBreakEngine();
-
- protected:
- /**
- * <p>Divide up a range of known dictionary characters.</p>
- *
- * @param text A UText representing the text
- * @param rangeStart The start of the range of dictionary characters
- * @param rangeEnd The end of the range of dictionary characters
- * @param foundBreaks Output of C array of int32_t break positions, or 0
- * @return The number of breaks found
- */
- virtual int32_t divideUpDictionaryRange( UText *text,
- int32_t rangeStart,
- int32_t rangeEnd,
- UVector32 &foundBreaks ) const;
-
-};
-
+/*******************************************************************
+ * BurmeseBreakEngine
+ */
+
+/**
+ * <p>BurmeseBreakEngine is a kind of DictionaryBreakEngine that uses a
+ * DictionaryMatcher and heuristics to determine Burmese-specific breaks.</p>
+ *
+ * <p>After it is constructed a BurmeseBreakEngine may be shared between
+ * threads without synchronization.</p>
+ */
+class BurmeseBreakEngine : public DictionaryBreakEngine {
+ private:
+ /**
+ * The set of characters handled by this engine
+ * @internal
+ */
+
+ UnicodeSet fBurmeseWordSet;
+ UnicodeSet fEndWordSet;
+ UnicodeSet fBeginWordSet;
+ UnicodeSet fMarkSet;
+ DictionaryMatcher *fDictionary;
+
+ public:
+
+ /**
+ * <p>Default constructor.</p>
+ *
+ * @param adoptDictionary A DictionaryMatcher to adopt. Deleted when the
+ * engine is deleted.
+ */
+ BurmeseBreakEngine(DictionaryMatcher *adoptDictionary, UErrorCode &status);
+
+ /**
+ * <p>Virtual destructor.</p>
+ */
+ virtual ~BurmeseBreakEngine();
+
+ protected:
+ /**
+ * <p>Divide up a range of known dictionary characters.</p>
+ *
+ * @param text A UText representing the text
+ * @param rangeStart The start of the range of dictionary characters
+ * @param rangeEnd The end of the range of dictionary characters
+ * @param foundBreaks Output of C array of int32_t break positions, or 0
+ * @return The number of breaks found
+ */
+ virtual int32_t divideUpDictionaryRange( UText *text,
+ int32_t rangeStart,
+ int32_t rangeEnd,
+ UVector32 &foundBreaks ) const;
+
+};
+
+/*******************************************************************
+ * KhmerBreakEngine
+ */
+
+/**
+ * <p>KhmerBreakEngine is a kind of DictionaryBreakEngine that uses a
+ * DictionaryMatcher and heuristics to determine Khmer-specific breaks.</p>
+ *
+ * <p>After it is constructed a KhmerBreakEngine may be shared between
+ * threads without synchronization.</p>
+ */
+class KhmerBreakEngine : public DictionaryBreakEngine {
+ private:
+ /**
+ * The set of characters handled by this engine
+ * @internal
+ */
+
+ UnicodeSet fKhmerWordSet;
+ UnicodeSet fBeginWordSet;
+ UnicodeSet fPuncSet;
+ DictionaryMatcher *fDictionary;
+
+ const uint32_t BADSNLP = 256 * 20;
+ const uint32_t kuint32max = 0x7FFFFFFF;
+
+ public:
+
+ /**
+ * <p>Default constructor.</p>
+ *
+ * @param adoptDictionary A DictionaryMatcher to adopt. Deleted when the
+ * engine is deleted.
+ */
+ KhmerBreakEngine(DictionaryMatcher *adoptDictionary, UErrorCode &status);
+
+ /**
+ * <p>Virtual destructor.</p>
+ */
+ virtual ~KhmerBreakEngine();
+
+ protected:
+ /**
+ * <p>Divide up a range of known dictionary characters.</p>
+ *
+ * @param text A UText representing the text
+ * @param rangeStart The start of the range of dictionary characters
+ * @param rangeEnd The end of the range of dictionary characters
+ * @param foundBreaks Output of C array of int32_t break positions, or 0
+ * @return The number of breaks found
+ */
+ virtual int32_t divideUpDictionaryRange( UText *text,
+ int32_t rangeStart,
+ int32_t rangeEnd,
+ UVector32 &foundBreaks ) const;
+
+};
+
#if !UCONFIG_NO_NORMALIZATION
/*******************************************************************
diff -ur icu.org/source/common/dictionarydata.cpp icu/source/common/dictionarydata.cpp
--- icu.org/source/common/dictionarydata.cpp 2018-09-29 02:34:41.000000000 +0200
+++ icu/source/common/dictionarydata.cpp 2018-10-19 14:21:00.339942804 +0200
@@ -44,7 +44,7 @@
int32_t UCharsDictionaryMatcher::matches(UText *text, int32_t maxLength, int32_t limit,
int32_t *lengths, int32_t *cpLengths, int32_t *values,
- int32_t *prefix) const {
+ int32_t *prefix, UnicodeSet const* ignoreSet, int32_t minLength) const {
UCharsTrie uct(characters);
int32_t startingTextIndex = (int32_t)utext_getNativeIndex(text);
@@ -55,7 +55,13 @@
UStringTrieResult result = (codePointsMatched == 0) ? uct.first(c) : uct.next(c);
int32_t lengthMatched = (int32_t)utext_getNativeIndex(text) - startingTextIndex;
codePointsMatched += 1;
+ if (ignoreSet != NULL && ignoreSet->contains(c)) {
+ continue;
+ }
if (USTRINGTRIE_HAS_VALUE(result)) {
+ if (codePointsMatched < minLength) {
+ continue;
+ }
if (wordCount < limit) {
if (values != NULL) {
values[wordCount] = uct.getValue();
@@ -112,7 +118,7 @@
int32_t BytesDictionaryMatcher::matches(UText *text, int32_t maxLength, int32_t limit,
int32_t *lengths, int32_t *cpLengths, int32_t *values,
- int32_t *prefix) const {
+ int32_t *prefix, UnicodeSet const* ignoreSet, int32_t minLength) const {
BytesTrie bt(characters);
int32_t startingTextIndex = (int32_t)utext_getNativeIndex(text);
int32_t wordCount = 0;
@@ -122,7 +128,13 @@
UStringTrieResult result = (codePointsMatched == 0) ? bt.first(transform(c)) : bt.next(transform(c));
int32_t lengthMatched = (int32_t)utext_getNativeIndex(text) - startingTextIndex;
codePointsMatched += 1;
+ if (ignoreSet != NULL && ignoreSet->contains(c)) {
+ continue;
+ }
if (USTRINGTRIE_HAS_VALUE(result)) {
+ if (codePointsMatched < minLength) {
+ continue;
+ }
if (wordCount < limit) {
if (values != NULL) {
values[wordCount] = bt.getValue();
diff -ur icu.org/source/common/dictionarydata.h icu/source/common/dictionarydata.h
--- icu.org/source/common/dictionarydata.h 2018-09-29 02:34:41.000000000 +0200
+++ icu/source/common/dictionarydata.h 2018-10-19 14:21:00.339942804 +0200
@@ -21,6 +21,7 @@
#include "unicode/utext.h"
#include "unicode/udata.h"
#include "udataswp.h"
+#include "unicode/uniset.h"
#include "unicode/uobject.h"
#include "unicode/ustringtrie.h"
@@ -92,7 +93,7 @@
*/
virtual int32_t matches(UText *text, int32_t maxLength, int32_t limit,
int32_t *lengths, int32_t *cpLengths, int32_t *values,
- int32_t *prefix) const = 0;
+ int32_t *prefix, UnicodeSet const* ignoreSet = NULL, int32_t minLength = 0) const = 0;
/** @return DictionaryData::TRIE_TYPE_XYZ */
virtual int32_t getType() const = 0;
@@ -107,7 +108,7 @@
virtual ~UCharsDictionaryMatcher();
virtual int32_t matches(UText *text, int32_t maxLength, int32_t limit,
int32_t *lengths, int32_t *cpLengths, int32_t *values,
- int32_t *prefix) const;
+ int32_t *prefix, UnicodeSet const* ignoreSet = NULL, int32_t minLength = 0) const;
virtual int32_t getType() const;
private:
const UChar *characters;
@@ -125,7 +126,7 @@
virtual ~BytesDictionaryMatcher();
virtual int32_t matches(UText *text, int32_t maxLength, int32_t limit,
int32_t *lengths, int32_t *cpLengths, int32_t *values,
- int32_t *prefix) const;
+ int32_t *prefix, UnicodeSet const* ignoreSet = NULL, int32_t minLength = 0) const;
virtual int32_t getType() const;
private:
UChar32 transform(UChar32 c) const;
diff -ur icu.org/source/data/Makefile.in icu/source/data/Makefile.in
--- icu.org/source/data/Makefile.in 2018-10-19 14:20:11.031075856 +0200
+++ icu/source/data/Makefile.in 2018-10-19 14:21:00.339942804 +0200
@@ -183,7 +183,7 @@
endif
endif
-packagedata: icupkg.inc $(PKGDATA_LIST) build-local
+packagedata: icupkg.inc $(PKGDATA_LIST) build-local $(MAINBUILDDIR)/khmerdict.stamp
ifneq ($(ENABLE_STATIC),)
ifeq ($(PKGDATA_MODE),dll)
$(PKGDATA_INVOKE) $(PKGDATA) -e $(ICUDATA_ENTRY_POINT) -T $(OUTTMPDIR) -p $(ICUDATA_NAME) $(PKGDATA_LIBSTATICNAME) -m static $(PKGDATA_VERSIONING) $(PKGDATA_LIST)
@@ -567,8 +567,14 @@
$(INVOKE) $(TOOLBINDIR)/gendict --bytes --transform offset-0x1000 -c -i $(BUILDDIR) $(DICTSRCDIR)/burmesedict.txt $(BRKBLDDIR)/burmesedict.dict
# TODO: figure out why combining characters are here?
-$(BRKBLDDIR)/khmerdict.dict: $(TOOLBINDIR)/gendict$(TOOLEXEEXT) $(DAT_FILES)
- $(INVOKE) $(TOOLBINDIR)/gendict --bytes --transform offset-0x1780 -c -i $(BUILDDIR) $(DICTSRCDIR)/khmerdict.txt $(BRKBLDDIR)/khmerdict.dict
+#$(BRKBLDDIR)/khmerdict.dict: $(TOOLBINDIR)/gendict$(TOOLEXEEXT) $(DAT_FILES)
+# $(INVOKE) $(TOOLBINDIR)/gendict --bytes --transform offset-0x1780 -c -i $(BUILDDIR) $(DICTSRCDIR)/khmerdict.txt $(BRKBLDDIR)/khmerdict.dict
+
+#$(MAINBUILDDIR)/khmerdict.stamp: $(TOOLBINDIR)/gendict$(TOOLEXEEXT) $(BRKSRCDIR)/khmerdict.txt build-local
+# $(INVOKE) $(TOOLBINDIR)/gendict --bytes --transform offset-0x1780 -c -i $(BUILDDIR) $(DICTSRCDIR)/khmerdict.txt $(BRKBLDDIR)/khmerdict.dict
+$(MAINBUILDDIR)/khmerdict.stamp: $(BRKSRCDIR)/khmerdict.dict build-local
+ cp $< $(BRKBLDDIR)
+ echo "timestamp" > $@
#################################################### CFU
# CFU FILES
|