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
|
// © 2016 and later: Unicode, Inc. and others.
// License & terms of use: http://www.unicode.org/copyright.html
/*
*******************************************************************************
* Copyright (C) 2010-2014, International Business Machines
* Corporation and others. All Rights Reserved.
*******************************************************************************
* utf16collationiterator.cpp
*
* created on: 2010oct27
* created by: Markus W. Scherer
*/
#include "unicode/utypes.h"
#if !UCONFIG_NO_COLLATION
#include "charstr.h"
#include "cmemory.h"
#include "collation.h"
#include "collationdata.h"
#include "collationfcd.h"
#include "collationiterator.h"
#include "normalizer2impl.h"
#include "uassert.h"
#include "utf16collationiterator.h"
U_NAMESPACE_BEGIN
UTF16CollationIterator::UTF16CollationIterator(const UTF16CollationIterator &other,
const UChar *newText)
: CollationIterator(other),
start(newText),
pos(newText + (other.pos - other.start)),
limit(other.limit == NULL ? NULL : newText + (other.limit - other.start)) {
}
UTF16CollationIterator::~UTF16CollationIterator() {}
UBool
UTF16CollationIterator::operator==(const CollationIterator &other) const {
if(!CollationIterator::operator==(other)) { return FALSE; }
const UTF16CollationIterator &o = static_cast<const UTF16CollationIterator &>(other);
// Compare the iterator state but not the text: Assume that the caller does that.
return (pos - start) == (o.pos - o.start);
}
void
UTF16CollationIterator::resetToOffset(int32_t newOffset) {
reset();
pos = start + newOffset;
}
int32_t
UTF16CollationIterator::getOffset() const {
return (int32_t)(pos - start);
}
uint32_t
UTF16CollationIterator::handleNextCE32(UChar32 &c, UErrorCode & /*errorCode*/) {
if(pos == limit) {
c = U_SENTINEL;
return Collation::FALLBACK_CE32;
}
c = *pos++;
return UTRIE2_GET32_FROM_U16_SINGLE_LEAD(trie, c);
}
UChar
UTF16CollationIterator::handleGetTrailSurrogate() {
if(pos == limit) { return 0; }
UChar trail;
if(U16_IS_TRAIL(trail = *pos)) { ++pos; }
return trail;
}
UBool
UTF16CollationIterator::foundNULTerminator() {
if(limit == NULL) {
limit = --pos;
return TRUE;
} else {
return FALSE;
}
}
UChar32
UTF16CollationIterator::nextCodePoint(UErrorCode & /*errorCode*/) {
if(pos == limit) {
return U_SENTINEL;
}
UChar32 c = *pos;
if(c == 0 && limit == NULL) {
limit = pos;
return U_SENTINEL;
}
++pos;
UChar trail;
if(U16_IS_LEAD(c) && pos != limit && U16_IS_TRAIL(trail = *pos)) {
++pos;
return U16_GET_SUPPLEMENTARY(c, trail);
} else {
return c;
}
}
UChar32
UTF16CollationIterator::previousCodePoint(UErrorCode & /*errorCode*/) {
if(pos == start) {
return U_SENTINEL;
}
UChar32 c = *--pos;
UChar lead;
if(U16_IS_TRAIL(c) && pos != start && U16_IS_LEAD(lead = *(pos - 1))) {
--pos;
return U16_GET_SUPPLEMENTARY(lead, c);
} else {
return c;
}
}
void
UTF16CollationIterator::forwardNumCodePoints(int32_t num, UErrorCode & /*errorCode*/) {
while(num > 0 && pos != limit) {
UChar32 c = *pos;
if(c == 0 && limit == NULL) {
limit = pos;
break;
}
++pos;
--num;
if(U16_IS_LEAD(c) && pos != limit && U16_IS_TRAIL(*pos)) {
++pos;
}
}
}
void
UTF16CollationIterator::backwardNumCodePoints(int32_t num, UErrorCode & /*errorCode*/) {
while(num > 0 && pos != start) {
UChar32 c = *--pos;
--num;
if(U16_IS_TRAIL(c) && pos != start && U16_IS_LEAD(*(pos-1))) {
--pos;
}
}
}
// FCDUTF16CollationIterator ----------------------------------------------- ***
FCDUTF16CollationIterator::FCDUTF16CollationIterator(const FCDUTF16CollationIterator &other,
const UChar *newText)
: UTF16CollationIterator(other),
rawStart(newText),
segmentStart(newText + (other.segmentStart - other.rawStart)),
segmentLimit(other.segmentLimit == NULL ? NULL : newText + (other.segmentLimit - other.rawStart)),
rawLimit(other.rawLimit == NULL ? NULL : newText + (other.rawLimit - other.rawStart)),
nfcImpl(other.nfcImpl),
normalized(other.normalized),
checkDir(other.checkDir) {
if(checkDir != 0 || other.start == other.segmentStart) {
start = newText + (other.start - other.rawStart);
pos = newText + (other.pos - other.rawStart);
limit = other.limit == NULL ? NULL : newText + (other.limit - other.rawStart);
} else {
start = normalized.getBuffer();
pos = start + (other.pos - other.start);
limit = start + normalized.length();
}
}
FCDUTF16CollationIterator::~FCDUTF16CollationIterator() {}
UBool
FCDUTF16CollationIterator::operator==(const CollationIterator &other) const {
// Skip the UTF16CollationIterator and call its parent.
if(!CollationIterator::operator==(other)) { return FALSE; }
const FCDUTF16CollationIterator &o = static_cast<const FCDUTF16CollationIterator &>(other);
// Compare the iterator state but not the text: Assume that the caller does that.
if(checkDir != o.checkDir) { return FALSE; }
if(checkDir == 0 && (start == segmentStart) != (o.start == o.segmentStart)) { return FALSE; }
if(checkDir != 0 || start == segmentStart) {
return (pos - rawStart) == (o.pos - o.rawStart);
} else {
return (segmentStart - rawStart) == (o.segmentStart - o.rawStart) &&
(pos - start) == (o.pos - o.start);
}
}
void
FCDUTF16CollationIterator::resetToOffset(int32_t newOffset) {
reset();
start = segmentStart = pos = rawStart + newOffset;
limit = rawLimit;
checkDir = 1;
}
int32_t
FCDUTF16CollationIterator::getOffset() const {
if(checkDir != 0 || start == segmentStart) {
return (int32_t)(pos - rawStart);
} else if(pos == start) {
return (int32_t)(segmentStart - rawStart);
} else {
return (int32_t)(segmentLimit - rawStart);
}
}
uint32_t
FCDUTF16CollationIterator::handleNextCE32(UChar32 &c, UErrorCode &errorCode) {
for(;;) {
if(checkDir > 0) {
if(pos == limit) {
c = U_SENTINEL;
return Collation::FALLBACK_CE32;
}
c = *pos++;
if(CollationFCD::hasTccc(c)) {
if(CollationFCD::maybeTibetanCompositeVowel(c) ||
(pos != limit && CollationFCD::hasLccc(*pos))) {
--pos;
if(!nextSegment(errorCode)) {
c = U_SENTINEL;
return Collation::FALLBACK_CE32;
}
c = *pos++;
}
}
break;
} else if(checkDir == 0 && pos != limit) {
c = *pos++;
break;
} else {
switchToForward();
}
}
return UTRIE2_GET32_FROM_U16_SINGLE_LEAD(trie, c);
}
UBool
FCDUTF16CollationIterator::foundNULTerminator() {
if(limit == NULL) {
limit = rawLimit = --pos;
return TRUE;
} else {
return FALSE;
}
}
UChar32
FCDUTF16CollationIterator::nextCodePoint(UErrorCode &errorCode) {
UChar32 c;
for(;;) {
if(checkDir > 0) {
if(pos == limit) {
return U_SENTINEL;
}
c = *pos++;
if(CollationFCD::hasTccc(c)) {
if(CollationFCD::maybeTibetanCompositeVowel(c) ||
(pos != limit && CollationFCD::hasLccc(*pos))) {
--pos;
if(!nextSegment(errorCode)) {
return U_SENTINEL;
}
c = *pos++;
}
} else if(c == 0 && limit == NULL) {
limit = rawLimit = --pos;
return U_SENTINEL;
}
break;
} else if(checkDir == 0 && pos != limit) {
c = *pos++;
break;
} else {
switchToForward();
}
}
UChar trail;
if(U16_IS_LEAD(c) && pos != limit && U16_IS_TRAIL(trail = *pos)) {
++pos;
return U16_GET_SUPPLEMENTARY(c, trail);
} else {
return c;
}
}
UChar32
FCDUTF16CollationIterator::previousCodePoint(UErrorCode &errorCode) {
UChar32 c;
for(;;) {
if(checkDir < 0) {
if(pos == start) {
return U_SENTINEL;
}
c = *--pos;
if(CollationFCD::hasLccc(c)) {
if(CollationFCD::maybeTibetanCompositeVowel(c) ||
(pos != start && CollationFCD::hasTccc(*(pos - 1)))) {
++pos;
if(!previousSegment(errorCode)) {
return U_SENTINEL;
}
c = *--pos;
}
}
break;
} else if(checkDir == 0 && pos != start) {
c = *--pos;
break;
} else {
switchToBackward();
}
}
UChar lead;
if(U16_IS_TRAIL(c) && pos != start && U16_IS_LEAD(lead = *(pos - 1))) {
--pos;
return U16_GET_SUPPLEMENTARY(lead, c);
} else {
return c;
}
}
void
FCDUTF16CollationIterator::forwardNumCodePoints(int32_t num, UErrorCode &errorCode) {
// Specify the class to avoid a virtual-function indirection.
// In Java, we would declare this class final.
while(num > 0 && FCDUTF16CollationIterator::nextCodePoint(errorCode) >= 0) {
--num;
}
}
void
FCDUTF16CollationIterator::backwardNumCodePoints(int32_t num, UErrorCode &errorCode) {
// Specify the class to avoid a virtual-function indirection.
// In Java, we would declare this class final.
while(num > 0 && FCDUTF16CollationIterator::previousCodePoint(errorCode) >= 0) {
--num;
}
}
void
FCDUTF16CollationIterator::switchToForward() {
U_ASSERT(checkDir < 0 || (checkDir == 0 && pos == limit));
if(checkDir < 0) {
// Turn around from backward checking.
start = segmentStart = pos;
if(pos == segmentLimit) {
limit = rawLimit;
checkDir = 1; // Check forward.
} else { // pos < segmentLimit
checkDir = 0; // Stay in FCD segment.
}
} else {
// Reached the end of the FCD segment.
if(start == segmentStart) {
// The input text segment is FCD, extend it forward.
} else {
// The input text segment needed to be normalized.
// Switch to checking forward from it.
pos = start = segmentStart = segmentLimit;
// Note: If this segment is at the end of the input text,
// then it might help to return FALSE to indicate that, so that
// we do not have to re-check and normalize when we turn around and go backwards.
// However, that would complicate the call sites for an optimization of an unusual case.
}
limit = rawLimit;
checkDir = 1;
}
}
UBool
FCDUTF16CollationIterator::nextSegment(UErrorCode &errorCode) {
if(U_FAILURE(errorCode)) { return FALSE; }
U_ASSERT(checkDir > 0 && pos != limit);
// The input text [segmentStart..pos[ passes the FCD check.
const UChar *p = pos;
uint8_t prevCC = 0;
for(;;) {
// Fetch the next character's fcd16 value.
const UChar *q = p;
uint16_t fcd16 = nfcImpl.nextFCD16(p, rawLimit);
uint8_t leadCC = (uint8_t)(fcd16 >> 8);
if(leadCC == 0 && q != pos) {
// FCD boundary before the [q, p[ character.
limit = segmentLimit = q;
break;
}
if(leadCC != 0 && (prevCC > leadCC || CollationFCD::isFCD16OfTibetanCompositeVowel(fcd16))) {
// Fails FCD check. Find the next FCD boundary and normalize.
do {
q = p;
} while(p != rawLimit && nfcImpl.nextFCD16(p, rawLimit) > 0xff);
if(!normalize(pos, q, errorCode)) { return FALSE; }
pos = start;
break;
}
prevCC = (uint8_t)fcd16;
if(p == rawLimit || prevCC == 0) {
// FCD boundary after the last character.
limit = segmentLimit = p;
break;
}
}
U_ASSERT(pos != limit);
checkDir = 0;
return TRUE;
}
void
FCDUTF16CollationIterator::switchToBackward() {
U_ASSERT(checkDir > 0 || (checkDir == 0 && pos == start));
if(checkDir > 0) {
// Turn around from forward checking.
limit = segmentLimit = pos;
if(pos == segmentStart) {
start = rawStart;
checkDir = -1; // Check backward.
} else { // pos > segmentStart
checkDir = 0; // Stay in FCD segment.
}
} else {
// Reached the start of the FCD segment.
if(start == segmentStart) {
// The input text segment is FCD, extend it backward.
} else {
// The input text segment needed to be normalized.
// Switch to checking backward from it.
pos = limit = segmentLimit = segmentStart;
}
start = rawStart;
checkDir = -1;
}
}
UBool
FCDUTF16CollationIterator::previousSegment(UErrorCode &errorCode) {
if(U_FAILURE(errorCode)) { return FALSE; }
U_ASSERT(checkDir < 0 && pos != start);
// The input text [pos..segmentLimit[ passes the FCD check.
const UChar *p = pos;
uint8_t nextCC = 0;
for(;;) {
// Fetch the previous character's fcd16 value.
const UChar *q = p;
uint16_t fcd16 = nfcImpl.previousFCD16(rawStart, p);
uint8_t trailCC = (uint8_t)fcd16;
if(trailCC == 0 && q != pos) {
// FCD boundary after the [p, q[ character.
start = segmentStart = q;
break;
}
if(trailCC != 0 && ((nextCC != 0 && trailCC > nextCC) ||
CollationFCD::isFCD16OfTibetanCompositeVowel(fcd16))) {
// Fails FCD check. Find the previous FCD boundary and normalize.
do {
q = p;
} while(fcd16 > 0xff && p != rawStart &&
(fcd16 = nfcImpl.previousFCD16(rawStart, p)) != 0);
if(!normalize(q, pos, errorCode)) { return FALSE; }
pos = limit;
break;
}
nextCC = (uint8_t)(fcd16 >> 8);
if(p == rawStart || nextCC == 0) {
// FCD boundary before the following character.
start = segmentStart = p;
break;
}
}
U_ASSERT(pos != start);
checkDir = 0;
return TRUE;
}
UBool
FCDUTF16CollationIterator::normalize(const UChar *from, const UChar *to, UErrorCode &errorCode) {
// NFD without argument checking.
U_ASSERT(U_SUCCESS(errorCode));
nfcImpl.decompose(from, to, normalized, (int32_t)(to - from), errorCode);
if(U_FAILURE(errorCode)) { return FALSE; }
// Switch collation processing into the FCD buffer
// with the result of normalizing [segmentStart, segmentLimit[.
segmentStart = from;
segmentLimit = to;
start = normalized.getBuffer();
limit = start + normalized.length();
return TRUE;
}
U_NAMESPACE_END
#endif // !UCONFIG_NO_COLLATION
|