2 *************************************************************************
4 * Copyright (c) 1996-2010, International Business Machines Corporation and
5 * others. All Rights Reserved.
6 *************************************************************************
9 #include "unicode/utypes.h"
11 #if !UCONFIG_NO_NORMALIZATION
13 #include "unicode/uniset.h"
14 #include "unicode/unistr.h"
15 #include "unicode/chariter.h"
16 #include "unicode/schriter.h"
17 #include "unicode/uchriter.h"
18 #include "unicode/normlzr.h"
20 #include "normalizer2impl.h"
21 #include "uprops.h" // for uniset_getUnicode32Instance()
25 UOBJECT_DEFINE_RTTI_IMPLEMENTATION(Normalizer)
27 //-------------------------------------------------------------------------
28 // Constructors and other boilerplate
29 //-------------------------------------------------------------------------
31 Normalizer::Normalizer(const UnicodeString& str, UNormalizationMode mode) :
32 UObject(), fFilteredNorm2(NULL), fNorm2(NULL), fUMode(mode), fOptions(0),
33 text(new StringCharacterIterator(str)),
34 currentIndex(0), nextIndex(0),
35 buffer(), bufferPos(0)
40 Normalizer::Normalizer(const UChar *str, int32_t length, UNormalizationMode mode) :
41 UObject(), fFilteredNorm2(NULL), fNorm2(NULL), fUMode(mode), fOptions(0),
42 text(new UCharCharacterIterator(str, length)),
43 currentIndex(0), nextIndex(0),
44 buffer(), bufferPos(0)
49 Normalizer::Normalizer(const CharacterIterator& iter, UNormalizationMode mode) :
50 UObject(), fFilteredNorm2(NULL), fNorm2(NULL), fUMode(mode), fOptions(0),
52 currentIndex(0), nextIndex(0),
53 buffer(), bufferPos(0)
58 Normalizer::Normalizer(const Normalizer ©) :
59 UObject(copy), fFilteredNorm2(NULL), fNorm2(NULL), fUMode(copy.fUMode), fOptions(copy.fOptions),
60 text(copy.text->clone()),
61 currentIndex(copy.currentIndex), nextIndex(copy.nextIndex),
62 buffer(copy.buffer), bufferPos(copy.bufferPos)
67 static const UChar _NUL=0;
71 UErrorCode errorCode=U_ZERO_ERROR;
72 fNorm2=Normalizer2Factory::getInstance(fUMode, errorCode);
73 if(fOptions&UNORM_UNICODE_3_2) {
74 delete fFilteredNorm2;
75 fNorm2=fFilteredNorm2=
76 new FilteredNormalizer2(*fNorm2, *uniset_getUnicode32Instance(errorCode));
78 if(U_FAILURE(errorCode)) {
79 errorCode=U_ZERO_ERROR;
80 fNorm2=Normalizer2Factory::getNoopInstance(errorCode);
84 Normalizer::~Normalizer()
86 delete fFilteredNorm2;
91 Normalizer::clone() const
93 return new Normalizer(*this);
97 * Generates a hash code for this iterator.
99 int32_t Normalizer::hashCode() const
101 return text->hashCode() + fUMode + fOptions + buffer.hashCode() + bufferPos + currentIndex + nextIndex;
104 UBool Normalizer::operator==(const Normalizer& that) const
108 (fUMode==that.fUMode &&
109 fOptions==that.fOptions &&
111 buffer==that.buffer &&
112 bufferPos==that.bufferPos &&
113 nextIndex==that.nextIndex);
116 //-------------------------------------------------------------------------
117 // Static utility methods
118 //-------------------------------------------------------------------------
121 Normalizer::normalize(const UnicodeString& source,
122 UNormalizationMode mode, int32_t options,
123 UnicodeString& result,
124 UErrorCode &status) {
125 if(source.isBogus() || U_FAILURE(status)) {
127 if(U_SUCCESS(status)) {
128 status=U_ILLEGAL_ARGUMENT_ERROR;
131 UnicodeString localDest;
134 if(&source!=&result) {
137 // the source and result strings are the same object, use a temporary one
140 const Normalizer2 *n2=Normalizer2Factory::getInstance(mode, status);
141 if(U_SUCCESS(status)) {
142 if(options&UNORM_UNICODE_3_2) {
143 FilteredNormalizer2(*n2, *uniset_getUnicode32Instance(status)).
144 normalize(source, *dest, status);
146 n2->normalize(source, *dest, status);
149 if(dest==&localDest && U_SUCCESS(status)) {
156 Normalizer::compose(const UnicodeString& source,
157 UBool compat, int32_t options,
158 UnicodeString& result,
159 UErrorCode &status) {
160 normalize(source, compat ? UNORM_NFKC : UNORM_NFC, options, result, status);
164 Normalizer::decompose(const UnicodeString& source,
165 UBool compat, int32_t options,
166 UnicodeString& result,
167 UErrorCode &status) {
168 normalize(source, compat ? UNORM_NFKD : UNORM_NFD, options, result, status);
171 UNormalizationCheckResult
172 Normalizer::quickCheck(const UnicodeString& source,
173 UNormalizationMode mode, int32_t options,
174 UErrorCode &status) {
175 const Normalizer2 *n2=Normalizer2Factory::getInstance(mode, status);
176 if(U_SUCCESS(status)) {
177 if(options&UNORM_UNICODE_3_2) {
178 return FilteredNormalizer2(*n2, *uniset_getUnicode32Instance(status)).
179 quickCheck(source, status);
181 return n2->quickCheck(source, status);
189 Normalizer::isNormalized(const UnicodeString& source,
190 UNormalizationMode mode, int32_t options,
191 UErrorCode &status) {
192 const Normalizer2 *n2=Normalizer2Factory::getInstance(mode, status);
193 if(U_SUCCESS(status)) {
194 if(options&UNORM_UNICODE_3_2) {
195 return FilteredNormalizer2(*n2, *uniset_getUnicode32Instance(status)).
196 isNormalized(source, status);
198 return n2->isNormalized(source, status);
205 UnicodeString & U_EXPORT2
206 Normalizer::concatenate(UnicodeString &left, UnicodeString &right,
207 UnicodeString &result,
208 UNormalizationMode mode, int32_t options,
209 UErrorCode &errorCode) {
210 if(left.isBogus() || right.isBogus() || U_FAILURE(errorCode)) {
212 if(U_SUCCESS(errorCode)) {
213 errorCode=U_ILLEGAL_ARGUMENT_ERROR;
216 UnicodeString localDest;
219 if(&right!=&result) {
222 // the right and result strings are the same object, use a temporary one
226 const Normalizer2 *n2=Normalizer2Factory::getInstance(mode, errorCode);
227 if(U_SUCCESS(errorCode)) {
228 if(options&UNORM_UNICODE_3_2) {
229 FilteredNormalizer2(*n2, *uniset_getUnicode32Instance(errorCode)).
230 append(*dest, right, errorCode);
232 n2->append(*dest, right, errorCode);
235 if(dest==&localDest && U_SUCCESS(errorCode)) {
242 //-------------------------------------------------------------------------
244 //-------------------------------------------------------------------------
247 * Return the current character in the normalized text.
249 UChar32 Normalizer::current() {
250 if(bufferPos<buffer.length() || nextNormalize()) {
251 return buffer.char32At(bufferPos);
258 * Return the next character in the normalized text and advance
259 * the iteration position by one. If the end
260 * of the text has already been reached, {@link #DONE} is returned.
262 UChar32 Normalizer::next() {
263 if(bufferPos<buffer.length() || nextNormalize()) {
264 UChar32 c=buffer.char32At(bufferPos);
265 bufferPos+=UTF_CHAR_LENGTH(c);
273 * Return the previous character in the normalized text and decrement
274 * the iteration position by one. If the beginning
275 * of the text has already been reached, {@link #DONE} is returned.
277 UChar32 Normalizer::previous() {
278 if(bufferPos>0 || previousNormalize()) {
279 UChar32 c=buffer.char32At(bufferPos-1);
280 bufferPos-=UTF_CHAR_LENGTH(c);
287 void Normalizer::reset() {
288 currentIndex=nextIndex=text->setToStart();
293 Normalizer::setIndexOnly(int32_t index) {
294 text->setIndex(index); // pins index
295 currentIndex=nextIndex=text->getIndex();
300 * Return the first character in the normalized text. This resets
301 * the <tt>Normalizer's</tt> position to the beginning of the text.
303 UChar32 Normalizer::first() {
309 * Return the last character in the normalized text. This resets
310 * the <tt>Normalizer's</tt> position to be just before the
311 * the input text corresponding to that normalized character.
313 UChar32 Normalizer::last() {
314 currentIndex=nextIndex=text->setToEnd();
320 * Retrieve the current iteration position in the input text that is
321 * being normalized. This method is useful in applications such as
322 * searching, where you need to be able to determine the position in
323 * the input text that corresponds to a given normalized output character.
325 * <b>Note:</b> This method sets the position in the <em>input</em>, while
326 * {@link #next} and {@link #previous} iterate through characters in the
327 * <em>output</em>. This means that there is not necessarily a one-to-one
328 * correspondence between characters returned by <tt>next</tt> and
329 * <tt>previous</tt> and the indices passed to and returned from
330 * <tt>setIndex</tt> and {@link #getIndex}.
333 int32_t Normalizer::getIndex() const {
334 if(bufferPos<buffer.length()) {
342 * Retrieve the index of the start of the input text. This is the begin index
343 * of the <tt>CharacterIterator</tt> or the start (i.e. 0) of the <tt>String</tt>
344 * over which this <tt>Normalizer</tt> is iterating
346 int32_t Normalizer::startIndex() const {
347 return text->startIndex();
351 * Retrieve the index of the end of the input text. This is the end index
352 * of the <tt>CharacterIterator</tt> or the length of the <tt>String</tt>
353 * over which this <tt>Normalizer</tt> is iterating
355 int32_t Normalizer::endIndex() const {
356 return text->endIndex();
359 //-------------------------------------------------------------------------
360 // Property access methods
361 //-------------------------------------------------------------------------
364 Normalizer::setMode(UNormalizationMode newMode)
371 Normalizer::getUMode() const
377 Normalizer::setOption(int32_t option,
383 fOptions &= (~option);
389 Normalizer::getOption(int32_t option) const
391 return (fOptions & option) != 0;
395 * Set the input text over which this <tt>Normalizer</tt> will iterate.
396 * The iteration position is set to the beginning of the input text.
399 Normalizer::setText(const UnicodeString& newText,
402 if (U_FAILURE(status)) {
405 CharacterIterator *newIter = new StringCharacterIterator(newText);
406 if (newIter == NULL) {
407 status = U_MEMORY_ALLOCATION_ERROR;
416 * Set the input text over which this <tt>Normalizer</tt> will iterate.
417 * The iteration position is set to the beginning of the string.
420 Normalizer::setText(const CharacterIterator& newText,
423 if (U_FAILURE(status)) {
426 CharacterIterator *newIter = newText.clone();
427 if (newIter == NULL) {
428 status = U_MEMORY_ALLOCATION_ERROR;
437 Normalizer::setText(const UChar* newText,
441 if (U_FAILURE(status)) {
444 CharacterIterator *newIter = new UCharCharacterIterator(newText, length);
445 if (newIter == NULL) {
446 status = U_MEMORY_ALLOCATION_ERROR;
455 * Copies the text under iteration into the UnicodeString referred to by "result".
456 * @param result Receives a copy of the text under iteration.
459 Normalizer::getText(UnicodeString& result)
461 text->getText(result);
464 //-------------------------------------------------------------------------
465 // Private utility methods
466 //-------------------------------------------------------------------------
468 void Normalizer::clearBuffer() {
474 Normalizer::nextNormalize() {
476 currentIndex=nextIndex;
477 text->setIndex(nextIndex);
478 if(!text->hasNext()) {
481 // Skip at least one character so we make progress.
482 UnicodeString segment(text->next32PostInc());
483 while(text->hasNext()) {
485 if(fNorm2->hasBoundaryBefore(c=text->next32PostInc())) {
486 text->move32(-1, CharacterIterator::kCurrent);
491 nextIndex=text->getIndex();
492 UErrorCode errorCode=U_ZERO_ERROR;
493 fNorm2->normalize(segment, buffer, errorCode);
494 return U_SUCCESS(errorCode) && !buffer.isEmpty();
498 Normalizer::previousNormalize() {
500 nextIndex=currentIndex;
501 text->setIndex(currentIndex);
502 if(!text->hasPrevious()) {
505 UnicodeString segment;
506 while(text->hasPrevious()) {
507 UChar32 c=text->previous32();
508 segment.insert(0, c);
509 if(fNorm2->hasBoundaryBefore(c)) {
513 currentIndex=text->getIndex();
514 UErrorCode errorCode=U_ZERO_ERROR;
515 fNorm2->normalize(segment, buffer, errorCode);
516 bufferPos=buffer.length();
517 return U_SUCCESS(errorCode) && !buffer.isEmpty();
522 #endif /* #if !UCONFIG_NO_NORMALIZATION */