1 | // © 2016 and later: Unicode, Inc. and others. |
2 | // License & terms of use: http://www.unicode.org/copyright.html |
3 | /* |
4 | ************************************************************************* |
5 | * COPYRIGHT: |
6 | * Copyright (c) 1996-2012, International Business Machines Corporation and |
7 | * others. All Rights Reserved. |
8 | ************************************************************************* |
9 | */ |
10 | |
11 | #include "unicode/utypes.h" |
12 | |
13 | #if !UCONFIG_NO_NORMALIZATION |
14 | |
15 | #include "unicode/uniset.h" |
16 | #include "unicode/unistr.h" |
17 | #include "unicode/chariter.h" |
18 | #include "unicode/schriter.h" |
19 | #include "unicode/uchriter.h" |
20 | #include "unicode/normlzr.h" |
21 | #include "unicode/utf16.h" |
22 | #include "cmemory.h" |
23 | #include "normalizer2impl.h" |
24 | #include "uprops.h" // for uniset_getUnicode32Instance() |
25 | |
26 | #if defined(move32) |
27 | // System can define move32 intrinsics, but the char iters define move32 method |
28 | // using same undef trick in headers, so undef here to re-enable the method. |
29 | #undef move32 |
30 | #endif |
31 | |
32 | U_NAMESPACE_BEGIN |
33 | |
34 | UOBJECT_DEFINE_RTTI_IMPLEMENTATION(Normalizer) |
35 | |
36 | //------------------------------------------------------------------------- |
37 | // Constructors and other boilerplate |
38 | //------------------------------------------------------------------------- |
39 | |
40 | Normalizer::Normalizer(const UnicodeString& str, UNormalizationMode mode) : |
41 | UObject(), fFilteredNorm2(nullptr), fNorm2(nullptr), fUMode(mode), fOptions(0), |
42 | text(new StringCharacterIterator(str)), |
43 | currentIndex(0), nextIndex(0), |
44 | buffer(), bufferPos(0) |
45 | { |
46 | init(); |
47 | } |
48 | |
49 | Normalizer::Normalizer(ConstChar16Ptr str, int32_t length, UNormalizationMode mode) : |
50 | UObject(), fFilteredNorm2(nullptr), fNorm2(nullptr), fUMode(mode), fOptions(0), |
51 | text(new UCharCharacterIterator(str, length)), |
52 | currentIndex(0), nextIndex(0), |
53 | buffer(), bufferPos(0) |
54 | { |
55 | init(); |
56 | } |
57 | |
58 | Normalizer::Normalizer(const CharacterIterator& iter, UNormalizationMode mode) : |
59 | UObject(), fFilteredNorm2(nullptr), fNorm2(nullptr), fUMode(mode), fOptions(0), |
60 | text(iter.clone()), |
61 | currentIndex(0), nextIndex(0), |
62 | buffer(), bufferPos(0) |
63 | { |
64 | init(); |
65 | } |
66 | |
67 | Normalizer::Normalizer(const Normalizer ©) : |
68 | UObject(copy), fFilteredNorm2(nullptr), fNorm2(nullptr), fUMode(copy.fUMode), fOptions(copy.fOptions), |
69 | text(copy.text->clone()), |
70 | currentIndex(copy.currentIndex), nextIndex(copy.nextIndex), |
71 | buffer(copy.buffer), bufferPos(copy.bufferPos) |
72 | { |
73 | init(); |
74 | } |
75 | |
76 | void |
77 | Normalizer::init() { |
78 | UErrorCode errorCode=U_ZERO_ERROR; |
79 | fNorm2=Normalizer2Factory::getInstance(fUMode, errorCode); |
80 | if(fOptions&UNORM_UNICODE_3_2) { |
81 | delete fFilteredNorm2; |
82 | fNorm2=fFilteredNorm2= |
83 | new FilteredNormalizer2(*fNorm2, *uniset_getUnicode32Instance(errorCode)); |
84 | } |
85 | if(U_FAILURE(errorCode)) { |
86 | errorCode=U_ZERO_ERROR; |
87 | fNorm2=Normalizer2Factory::getNoopInstance(errorCode); |
88 | } |
89 | } |
90 | |
91 | Normalizer::~Normalizer() |
92 | { |
93 | delete fFilteredNorm2; |
94 | delete text; |
95 | } |
96 | |
97 | Normalizer* |
98 | Normalizer::clone() const |
99 | { |
100 | return new Normalizer(*this); |
101 | } |
102 | |
103 | /** |
104 | * Generates a hash code for this iterator. |
105 | */ |
106 | int32_t Normalizer::hashCode() const |
107 | { |
108 | return text->hashCode() + fUMode + fOptions + buffer.hashCode() + bufferPos + currentIndex + nextIndex; |
109 | } |
110 | |
111 | bool Normalizer::operator==(const Normalizer& that) const |
112 | { |
113 | return |
114 | this==&that || |
115 | (fUMode==that.fUMode && |
116 | fOptions==that.fOptions && |
117 | *text==*that.text && |
118 | buffer==that.buffer && |
119 | bufferPos==that.bufferPos && |
120 | nextIndex==that.nextIndex); |
121 | } |
122 | |
123 | //------------------------------------------------------------------------- |
124 | // Static utility methods |
125 | //------------------------------------------------------------------------- |
126 | |
127 | void U_EXPORT2 |
128 | Normalizer::normalize(const UnicodeString& source, |
129 | UNormalizationMode mode, int32_t options, |
130 | UnicodeString& result, |
131 | UErrorCode &status) { |
132 | if(source.isBogus() || U_FAILURE(status)) { |
133 | result.setToBogus(); |
134 | if(U_SUCCESS(status)) { |
135 | status=U_ILLEGAL_ARGUMENT_ERROR; |
136 | } |
137 | } else { |
138 | UnicodeString localDest; |
139 | UnicodeString *dest; |
140 | |
141 | if(&source!=&result) { |
142 | dest=&result; |
143 | } else { |
144 | // the source and result strings are the same object, use a temporary one |
145 | dest=&localDest; |
146 | } |
147 | const Normalizer2 *n2=Normalizer2Factory::getInstance(mode, status); |
148 | if(U_SUCCESS(status)) { |
149 | if(options&UNORM_UNICODE_3_2) { |
150 | FilteredNormalizer2(*n2, *uniset_getUnicode32Instance(status)). |
151 | normalize(source, *dest, status); |
152 | } else { |
153 | n2->normalize(source, *dest, status); |
154 | } |
155 | } |
156 | if(dest==&localDest && U_SUCCESS(status)) { |
157 | result=*dest; |
158 | } |
159 | } |
160 | } |
161 | |
162 | void U_EXPORT2 |
163 | Normalizer::compose(const UnicodeString& source, |
164 | UBool compat, int32_t options, |
165 | UnicodeString& result, |
166 | UErrorCode &status) { |
167 | normalize(source, compat ? UNORM_NFKC : UNORM_NFC, options, result, status); |
168 | } |
169 | |
170 | void U_EXPORT2 |
171 | Normalizer::decompose(const UnicodeString& source, |
172 | UBool compat, int32_t options, |
173 | UnicodeString& result, |
174 | UErrorCode &status) { |
175 | normalize(source, compat ? UNORM_NFKD : UNORM_NFD, options, result, status); |
176 | } |
177 | |
178 | UNormalizationCheckResult |
179 | Normalizer::quickCheck(const UnicodeString& source, |
180 | UNormalizationMode mode, int32_t options, |
181 | UErrorCode &status) { |
182 | const Normalizer2 *n2=Normalizer2Factory::getInstance(mode, status); |
183 | if(U_SUCCESS(status)) { |
184 | if(options&UNORM_UNICODE_3_2) { |
185 | return FilteredNormalizer2(*n2, *uniset_getUnicode32Instance(status)). |
186 | quickCheck(source, status); |
187 | } else { |
188 | return n2->quickCheck(source, status); |
189 | } |
190 | } else { |
191 | return UNORM_MAYBE; |
192 | } |
193 | } |
194 | |
195 | UBool |
196 | Normalizer::isNormalized(const UnicodeString& source, |
197 | UNormalizationMode mode, int32_t options, |
198 | UErrorCode &status) { |
199 | const Normalizer2 *n2=Normalizer2Factory::getInstance(mode, status); |
200 | if(U_SUCCESS(status)) { |
201 | if(options&UNORM_UNICODE_3_2) { |
202 | return FilteredNormalizer2(*n2, *uniset_getUnicode32Instance(status)). |
203 | isNormalized(source, status); |
204 | } else { |
205 | return n2->isNormalized(source, status); |
206 | } |
207 | } else { |
208 | return false; |
209 | } |
210 | } |
211 | |
212 | UnicodeString & U_EXPORT2 |
213 | Normalizer::concatenate(const UnicodeString &left, const UnicodeString &right, |
214 | UnicodeString &result, |
215 | UNormalizationMode mode, int32_t options, |
216 | UErrorCode &errorCode) { |
217 | if(left.isBogus() || right.isBogus() || U_FAILURE(errorCode)) { |
218 | result.setToBogus(); |
219 | if(U_SUCCESS(errorCode)) { |
220 | errorCode=U_ILLEGAL_ARGUMENT_ERROR; |
221 | } |
222 | } else { |
223 | UnicodeString localDest; |
224 | UnicodeString *dest; |
225 | |
226 | if(&right!=&result) { |
227 | dest=&result; |
228 | } else { |
229 | // the right and result strings are the same object, use a temporary one |
230 | dest=&localDest; |
231 | } |
232 | *dest=left; |
233 | const Normalizer2 *n2=Normalizer2Factory::getInstance(mode, errorCode); |
234 | if(U_SUCCESS(errorCode)) { |
235 | if(options&UNORM_UNICODE_3_2) { |
236 | FilteredNormalizer2(*n2, *uniset_getUnicode32Instance(errorCode)). |
237 | append(*dest, right, errorCode); |
238 | } else { |
239 | n2->append(*dest, right, errorCode); |
240 | } |
241 | } |
242 | if(dest==&localDest && U_SUCCESS(errorCode)) { |
243 | result=*dest; |
244 | } |
245 | } |
246 | return result; |
247 | } |
248 | |
249 | //------------------------------------------------------------------------- |
250 | // Iteration API |
251 | //------------------------------------------------------------------------- |
252 | |
253 | /** |
254 | * Return the current character in the normalized text. |
255 | */ |
256 | UChar32 Normalizer::current() { |
257 | if(bufferPos<buffer.length() || nextNormalize()) { |
258 | return buffer.char32At(bufferPos); |
259 | } else { |
260 | return DONE; |
261 | } |
262 | } |
263 | |
264 | /** |
265 | * Return the next character in the normalized text and advance |
266 | * the iteration position by one. If the end |
267 | * of the text has already been reached, {@link #DONE} is returned. |
268 | */ |
269 | UChar32 Normalizer::next() { |
270 | if(bufferPos<buffer.length() || nextNormalize()) { |
271 | UChar32 c=buffer.char32At(bufferPos); |
272 | bufferPos+=U16_LENGTH(c); |
273 | return c; |
274 | } else { |
275 | return DONE; |
276 | } |
277 | } |
278 | |
279 | /** |
280 | * Return the previous character in the normalized text and decrement |
281 | * the iteration position by one. If the beginning |
282 | * of the text has already been reached, {@link #DONE} is returned. |
283 | */ |
284 | UChar32 Normalizer::previous() { |
285 | if(bufferPos>0 || previousNormalize()) { |
286 | UChar32 c=buffer.char32At(bufferPos-1); |
287 | bufferPos-=U16_LENGTH(c); |
288 | return c; |
289 | } else { |
290 | return DONE; |
291 | } |
292 | } |
293 | |
294 | void Normalizer::reset() { |
295 | currentIndex=nextIndex=text->setToStart(); |
296 | clearBuffer(); |
297 | } |
298 | |
299 | void |
300 | Normalizer::setIndexOnly(int32_t index) { |
301 | text->setIndex(index); // pins index |
302 | currentIndex=nextIndex=text->getIndex(); |
303 | clearBuffer(); |
304 | } |
305 | |
306 | /** |
307 | * Return the first character in the normalized text. This resets |
308 | * the <tt>Normalizer's</tt> position to the beginning of the text. |
309 | */ |
310 | UChar32 Normalizer::first() { |
311 | reset(); |
312 | return next(); |
313 | } |
314 | |
315 | /** |
316 | * Return the last character in the normalized text. This resets |
317 | * the <tt>Normalizer's</tt> position to be just before the |
318 | * the input text corresponding to that normalized character. |
319 | */ |
320 | UChar32 Normalizer::last() { |
321 | currentIndex=nextIndex=text->setToEnd(); |
322 | clearBuffer(); |
323 | return previous(); |
324 | } |
325 | |
326 | /** |
327 | * Retrieve the current iteration position in the input text that is |
328 | * being normalized. This method is useful in applications such as |
329 | * searching, where you need to be able to determine the position in |
330 | * the input text that corresponds to a given normalized output character. |
331 | * <p> |
332 | * <b>Note:</b> This method sets the position in the <em>input</em>, while |
333 | * {@link #next} and {@link #previous} iterate through characters in the |
334 | * <em>output</em>. This means that there is not necessarily a one-to-one |
335 | * correspondence between characters returned by <tt>next</tt> and |
336 | * <tt>previous</tt> and the indices passed to and returned from |
337 | * <tt>setIndex</tt> and {@link #getIndex}. |
338 | * |
339 | */ |
340 | int32_t Normalizer::getIndex() const { |
341 | if(bufferPos<buffer.length()) { |
342 | return currentIndex; |
343 | } else { |
344 | return nextIndex; |
345 | } |
346 | } |
347 | |
348 | /** |
349 | * Retrieve the index of the start of the input text. This is the begin index |
350 | * of the <tt>CharacterIterator</tt> or the start (i.e. 0) of the <tt>String</tt> |
351 | * over which this <tt>Normalizer</tt> is iterating |
352 | */ |
353 | int32_t Normalizer::startIndex() const { |
354 | return text->startIndex(); |
355 | } |
356 | |
357 | /** |
358 | * Retrieve the index of the end of the input text. This is the end index |
359 | * of the <tt>CharacterIterator</tt> or the length of the <tt>String</tt> |
360 | * over which this <tt>Normalizer</tt> is iterating |
361 | */ |
362 | int32_t Normalizer::endIndex() const { |
363 | return text->endIndex(); |
364 | } |
365 | |
366 | //------------------------------------------------------------------------- |
367 | // Property access methods |
368 | //------------------------------------------------------------------------- |
369 | |
370 | void |
371 | Normalizer::setMode(UNormalizationMode newMode) |
372 | { |
373 | fUMode = newMode; |
374 | init(); |
375 | } |
376 | |
377 | UNormalizationMode |
378 | Normalizer::getUMode() const |
379 | { |
380 | return fUMode; |
381 | } |
382 | |
383 | void |
384 | Normalizer::setOption(int32_t option, |
385 | UBool value) |
386 | { |
387 | if (value) { |
388 | fOptions |= option; |
389 | } else { |
390 | fOptions &= (~option); |
391 | } |
392 | init(); |
393 | } |
394 | |
395 | UBool |
396 | Normalizer::getOption(int32_t option) const |
397 | { |
398 | return (fOptions & option) != 0; |
399 | } |
400 | |
401 | /** |
402 | * Set the input text over which this <tt>Normalizer</tt> will iterate. |
403 | * The iteration position is set to the beginning of the input text. |
404 | */ |
405 | void |
406 | Normalizer::setText(const UnicodeString& newText, |
407 | UErrorCode &status) |
408 | { |
409 | if (U_FAILURE(status)) { |
410 | return; |
411 | } |
412 | CharacterIterator *newIter = new StringCharacterIterator(newText); |
413 | if (newIter == nullptr) { |
414 | status = U_MEMORY_ALLOCATION_ERROR; |
415 | return; |
416 | } |
417 | delete text; |
418 | text = newIter; |
419 | reset(); |
420 | } |
421 | |
422 | /** |
423 | * Set the input text over which this <tt>Normalizer</tt> will iterate. |
424 | * The iteration position is set to the beginning of the string. |
425 | */ |
426 | void |
427 | Normalizer::setText(const CharacterIterator& newText, |
428 | UErrorCode &status) |
429 | { |
430 | if (U_FAILURE(status)) { |
431 | return; |
432 | } |
433 | CharacterIterator *newIter = newText.clone(); |
434 | if (newIter == nullptr) { |
435 | status = U_MEMORY_ALLOCATION_ERROR; |
436 | return; |
437 | } |
438 | delete text; |
439 | text = newIter; |
440 | reset(); |
441 | } |
442 | |
443 | void |
444 | Normalizer::setText(ConstChar16Ptr newText, |
445 | int32_t length, |
446 | UErrorCode &status) |
447 | { |
448 | if (U_FAILURE(status)) { |
449 | return; |
450 | } |
451 | CharacterIterator *newIter = new UCharCharacterIterator(newText, length); |
452 | if (newIter == nullptr) { |
453 | status = U_MEMORY_ALLOCATION_ERROR; |
454 | return; |
455 | } |
456 | delete text; |
457 | text = newIter; |
458 | reset(); |
459 | } |
460 | |
461 | /** |
462 | * Copies the text under iteration into the UnicodeString referred to by "result". |
463 | * @param result Receives a copy of the text under iteration. |
464 | */ |
465 | void |
466 | Normalizer::getText(UnicodeString& result) |
467 | { |
468 | text->getText(result); |
469 | } |
470 | |
471 | //------------------------------------------------------------------------- |
472 | // Private utility methods |
473 | //------------------------------------------------------------------------- |
474 | |
475 | void Normalizer::clearBuffer() { |
476 | buffer.remove(); |
477 | bufferPos=0; |
478 | } |
479 | |
480 | UBool |
481 | Normalizer::nextNormalize() { |
482 | clearBuffer(); |
483 | currentIndex=nextIndex; |
484 | text->setIndex(nextIndex); |
485 | if(!text->hasNext()) { |
486 | return false; |
487 | } |
488 | // Skip at least one character so we make progress. |
489 | UnicodeString segment(text->next32PostInc()); |
490 | while(text->hasNext()) { |
491 | UChar32 c; |
492 | if(fNorm2->hasBoundaryBefore(c=text->next32PostInc())) { |
493 | text->move32(-1, CharacterIterator::kCurrent); |
494 | break; |
495 | } |
496 | segment.append(c); |
497 | } |
498 | nextIndex=text->getIndex(); |
499 | UErrorCode errorCode=U_ZERO_ERROR; |
500 | fNorm2->normalize(segment, buffer, errorCode); |
501 | return U_SUCCESS(errorCode) && !buffer.isEmpty(); |
502 | } |
503 | |
504 | UBool |
505 | Normalizer::previousNormalize() { |
506 | clearBuffer(); |
507 | nextIndex=currentIndex; |
508 | text->setIndex(currentIndex); |
509 | if(!text->hasPrevious()) { |
510 | return false; |
511 | } |
512 | UnicodeString segment; |
513 | while(text->hasPrevious()) { |
514 | UChar32 c=text->previous32(); |
515 | segment.insert(0, c); |
516 | if(fNorm2->hasBoundaryBefore(c)) { |
517 | break; |
518 | } |
519 | } |
520 | currentIndex=text->getIndex(); |
521 | UErrorCode errorCode=U_ZERO_ERROR; |
522 | fNorm2->normalize(segment, buffer, errorCode); |
523 | bufferPos=buffer.length(); |
524 | return U_SUCCESS(errorCode) && !buffer.isEmpty(); |
525 | } |
526 | |
527 | U_NAMESPACE_END |
528 | |
529 | #endif /* #if !UCONFIG_NO_NORMALIZATION */ |
530 | |