1 | // © 2016 and later: Unicode, Inc. and others. |
2 | // License & terms of use: http://www.unicode.org/copyright.html |
3 | /* |
4 | ******************************************************************************* |
5 | * |
6 | * Copyright (C) 2009-2015, International Business Machines |
7 | * Corporation and others. All Rights Reserved. |
8 | * |
9 | ******************************************************************************* |
10 | * file name: unorm2.h |
11 | * encoding: UTF-8 |
12 | * tab size: 8 (not used) |
13 | * indentation:4 |
14 | * |
15 | * created on: 2009dec15 |
16 | * created by: Markus W. Scherer |
17 | */ |
18 | |
19 | #ifndef __UNORM2_H__ |
20 | #define __UNORM2_H__ |
21 | |
22 | /** |
23 | * \file |
24 | * \brief C API: New API for Unicode Normalization. |
25 | * |
26 | * Unicode normalization functionality for standard Unicode normalization or |
27 | * for using custom mapping tables. |
28 | * All instances of UNormalizer2 are unmodifiable/immutable. |
29 | * Instances returned by unorm2_getInstance() are singletons that must not be deleted by the caller. |
30 | * For more details see the Normalizer2 C++ class. |
31 | */ |
32 | |
33 | #include "unicode/utypes.h" |
34 | #include "unicode/stringoptions.h" |
35 | #include "unicode/uset.h" |
36 | |
37 | #if U_SHOW_CPLUSPLUS_API |
38 | #include "unicode/localpointer.h" |
39 | #endif // U_SHOW_CPLUSPLUS_API |
40 | |
41 | /** |
42 | * Constants for normalization modes. |
43 | * For details about standard Unicode normalization forms |
44 | * and about the algorithms which are also used with custom mapping tables |
45 | * see http://www.unicode.org/unicode/reports/tr15/ |
46 | * @stable ICU 4.4 |
47 | */ |
48 | typedef enum { |
49 | /** |
50 | * Decomposition followed by composition. |
51 | * Same as standard NFC when using an "nfc" instance. |
52 | * Same as standard NFKC when using an "nfkc" instance. |
53 | * For details about standard Unicode normalization forms |
54 | * see http://www.unicode.org/unicode/reports/tr15/ |
55 | * @stable ICU 4.4 |
56 | */ |
57 | UNORM2_COMPOSE, |
58 | /** |
59 | * Map, and reorder canonically. |
60 | * Same as standard NFD when using an "nfc" instance. |
61 | * Same as standard NFKD when using an "nfkc" instance. |
62 | * For details about standard Unicode normalization forms |
63 | * see http://www.unicode.org/unicode/reports/tr15/ |
64 | * @stable ICU 4.4 |
65 | */ |
66 | UNORM2_DECOMPOSE, |
67 | /** |
68 | * "Fast C or D" form. |
69 | * If a string is in this form, then further decomposition <i>without reordering</i> |
70 | * would yield the same form as DECOMPOSE. |
71 | * Text in "Fast C or D" form can be processed efficiently with data tables |
72 | * that are "canonically closed", that is, that provide equivalent data for |
73 | * equivalent text, without having to be fully normalized. |
74 | * Not a standard Unicode normalization form. |
75 | * Not a unique form: Different FCD strings can be canonically equivalent. |
76 | * For details see http://www.unicode.org/notes/tn5/#FCD |
77 | * @stable ICU 4.4 |
78 | */ |
79 | UNORM2_FCD, |
80 | /** |
81 | * Compose only contiguously. |
82 | * Also known as "FCC" or "Fast C Contiguous". |
83 | * The result will often but not always be in NFC. |
84 | * The result will conform to FCD which is useful for processing. |
85 | * Not a standard Unicode normalization form. |
86 | * For details see http://www.unicode.org/notes/tn5/#FCC |
87 | * @stable ICU 4.4 |
88 | */ |
89 | UNORM2_COMPOSE_CONTIGUOUS |
90 | } UNormalization2Mode; |
91 | |
92 | /** |
93 | * Result values for normalization quick check functions. |
94 | * For details see http://www.unicode.org/reports/tr15/#Detecting_Normalization_Forms |
95 | * @stable ICU 2.0 |
96 | */ |
97 | typedef enum UNormalizationCheckResult { |
98 | /** |
99 | * The input string is not in the normalization form. |
100 | * @stable ICU 2.0 |
101 | */ |
102 | UNORM_NO, |
103 | /** |
104 | * The input string is in the normalization form. |
105 | * @stable ICU 2.0 |
106 | */ |
107 | UNORM_YES, |
108 | /** |
109 | * The input string may or may not be in the normalization form. |
110 | * This value is only returned for composition forms like NFC and FCC, |
111 | * when a backward-combining character is found for which the surrounding text |
112 | * would have to be analyzed further. |
113 | * @stable ICU 2.0 |
114 | */ |
115 | UNORM_MAYBE |
116 | } UNormalizationCheckResult; |
117 | |
118 | /** |
119 | * Opaque C service object type for the new normalization API. |
120 | * @stable ICU 4.4 |
121 | */ |
122 | struct UNormalizer2; |
123 | typedef struct UNormalizer2 UNormalizer2; /**< C typedef for struct UNormalizer2. @stable ICU 4.4 */ |
124 | |
125 | #if !UCONFIG_NO_NORMALIZATION |
126 | |
127 | /** |
128 | * Returns a UNormalizer2 instance for Unicode NFC normalization. |
129 | * Same as unorm2_getInstance(NULL, "nfc", UNORM2_COMPOSE, pErrorCode). |
130 | * Returns an unmodifiable singleton instance. Do not delete it. |
131 | * @param pErrorCode Standard ICU error code. Its input value must |
132 | * pass the U_SUCCESS() test, or else the function returns |
133 | * immediately. Check for U_FAILURE() on output or use with |
134 | * function chaining. (See User Guide for details.) |
135 | * @return the requested Normalizer2, if successful |
136 | * @stable ICU 49 |
137 | */ |
138 | U_CAPI const UNormalizer2 * U_EXPORT2 |
139 | unorm2_getNFCInstance(UErrorCode *pErrorCode); |
140 | |
141 | /** |
142 | * Returns a UNormalizer2 instance for Unicode NFD normalization. |
143 | * Same as unorm2_getInstance(NULL, "nfc", UNORM2_DECOMPOSE, pErrorCode). |
144 | * Returns an unmodifiable singleton instance. Do not delete it. |
145 | * @param pErrorCode Standard ICU error code. Its input value must |
146 | * pass the U_SUCCESS() test, or else the function returns |
147 | * immediately. Check for U_FAILURE() on output or use with |
148 | * function chaining. (See User Guide for details.) |
149 | * @return the requested Normalizer2, if successful |
150 | * @stable ICU 49 |
151 | */ |
152 | U_CAPI const UNormalizer2 * U_EXPORT2 |
153 | unorm2_getNFDInstance(UErrorCode *pErrorCode); |
154 | |
155 | /** |
156 | * Returns a UNormalizer2 instance for Unicode NFKC normalization. |
157 | * Same as unorm2_getInstance(NULL, "nfkc", UNORM2_COMPOSE, pErrorCode). |
158 | * Returns an unmodifiable singleton instance. Do not delete it. |
159 | * @param pErrorCode Standard ICU error code. Its input value must |
160 | * pass the U_SUCCESS() test, or else the function returns |
161 | * immediately. Check for U_FAILURE() on output or use with |
162 | * function chaining. (See User Guide for details.) |
163 | * @return the requested Normalizer2, if successful |
164 | * @stable ICU 49 |
165 | */ |
166 | U_CAPI const UNormalizer2 * U_EXPORT2 |
167 | unorm2_getNFKCInstance(UErrorCode *pErrorCode); |
168 | |
169 | /** |
170 | * Returns a UNormalizer2 instance for Unicode NFKD normalization. |
171 | * Same as unorm2_getInstance(NULL, "nfkc", UNORM2_DECOMPOSE, pErrorCode). |
172 | * Returns an unmodifiable singleton instance. Do not delete it. |
173 | * @param pErrorCode Standard ICU error code. Its input value must |
174 | * pass the U_SUCCESS() test, or else the function returns |
175 | * immediately. Check for U_FAILURE() on output or use with |
176 | * function chaining. (See User Guide for details.) |
177 | * @return the requested Normalizer2, if successful |
178 | * @stable ICU 49 |
179 | */ |
180 | U_CAPI const UNormalizer2 * U_EXPORT2 |
181 | unorm2_getNFKDInstance(UErrorCode *pErrorCode); |
182 | |
183 | /** |
184 | * Returns a UNormalizer2 instance for Unicode NFKC_Casefold normalization. |
185 | * Same as unorm2_getInstance(NULL, "nfkc_cf", UNORM2_COMPOSE, pErrorCode). |
186 | * Returns an unmodifiable singleton instance. Do not delete it. |
187 | * @param pErrorCode Standard ICU error code. Its input value must |
188 | * pass the U_SUCCESS() test, or else the function returns |
189 | * immediately. Check for U_FAILURE() on output or use with |
190 | * function chaining. (See User Guide for details.) |
191 | * @return the requested Normalizer2, if successful |
192 | * @stable ICU 49 |
193 | */ |
194 | U_CAPI const UNormalizer2 * U_EXPORT2 |
195 | unorm2_getNFKCCasefoldInstance(UErrorCode *pErrorCode); |
196 | |
197 | /** |
198 | * Returns a UNormalizer2 instance which uses the specified data file |
199 | * (packageName/name similar to ucnv_openPackage() and ures_open()/ResourceBundle) |
200 | * and which composes or decomposes text according to the specified mode. |
201 | * Returns an unmodifiable singleton instance. Do not delete it. |
202 | * |
203 | * Use packageName=NULL for data files that are part of ICU's own data. |
204 | * Use name="nfc" and UNORM2_COMPOSE/UNORM2_DECOMPOSE for Unicode standard NFC/NFD. |
205 | * Use name="nfkc" and UNORM2_COMPOSE/UNORM2_DECOMPOSE for Unicode standard NFKC/NFKD. |
206 | * Use name="nfkc_cf" and UNORM2_COMPOSE for Unicode standard NFKC_CF=NFKC_Casefold. |
207 | * |
208 | * @param packageName NULL for ICU built-in data, otherwise application data package name |
209 | * @param name "nfc" or "nfkc" or "nfkc_cf" or name of custom data file |
210 | * @param mode normalization mode (compose or decompose etc.) |
211 | * @param pErrorCode Standard ICU error code. Its input value must |
212 | * pass the U_SUCCESS() test, or else the function returns |
213 | * immediately. Check for U_FAILURE() on output or use with |
214 | * function chaining. (See User Guide for details.) |
215 | * @return the requested UNormalizer2, if successful |
216 | * @stable ICU 4.4 |
217 | */ |
218 | U_CAPI const UNormalizer2 * U_EXPORT2 |
219 | unorm2_getInstance(const char *packageName, |
220 | const char *name, |
221 | UNormalization2Mode mode, |
222 | UErrorCode *pErrorCode); |
223 | |
224 | /** |
225 | * Constructs a filtered normalizer wrapping any UNormalizer2 instance |
226 | * and a filter set. |
227 | * Both are aliased and must not be modified or deleted while this object |
228 | * is used. |
229 | * The filter set should be frozen; otherwise the performance will suffer greatly. |
230 | * @param norm2 wrapped UNormalizer2 instance |
231 | * @param filterSet USet which determines the characters to be normalized |
232 | * @param pErrorCode Standard ICU error code. Its input value must |
233 | * pass the U_SUCCESS() test, or else the function returns |
234 | * immediately. Check for U_FAILURE() on output or use with |
235 | * function chaining. (See User Guide for details.) |
236 | * @return the requested UNormalizer2, if successful |
237 | * @stable ICU 4.4 |
238 | */ |
239 | U_CAPI UNormalizer2 * U_EXPORT2 |
240 | unorm2_openFiltered(const UNormalizer2 *norm2, const USet *filterSet, UErrorCode *pErrorCode); |
241 | |
242 | /** |
243 | * Closes a UNormalizer2 instance from unorm2_openFiltered(). |
244 | * Do not close instances from unorm2_getInstance()! |
245 | * @param norm2 UNormalizer2 instance to be closed |
246 | * @stable ICU 4.4 |
247 | */ |
248 | U_CAPI void U_EXPORT2 |
249 | unorm2_close(UNormalizer2 *norm2); |
250 | |
251 | #if U_SHOW_CPLUSPLUS_API |
252 | |
253 | U_NAMESPACE_BEGIN |
254 | |
255 | /** |
256 | * \class LocalUNormalizer2Pointer |
257 | * "Smart pointer" class, closes a UNormalizer2 via unorm2_close(). |
258 | * For most methods see the LocalPointerBase base class. |
259 | * |
260 | * @see LocalPointerBase |
261 | * @see LocalPointer |
262 | * @stable ICU 4.4 |
263 | */ |
264 | U_DEFINE_LOCAL_OPEN_POINTER(LocalUNormalizer2Pointer, UNormalizer2, unorm2_close); |
265 | |
266 | U_NAMESPACE_END |
267 | |
268 | #endif |
269 | |
270 | /** |
271 | * Writes the normalized form of the source string to the destination string |
272 | * (replacing its contents) and returns the length of the destination string. |
273 | * The source and destination strings must be different buffers. |
274 | * @param norm2 UNormalizer2 instance |
275 | * @param src source string |
276 | * @param length length of the source string, or -1 if NUL-terminated |
277 | * @param dest destination string; its contents is replaced with normalized src |
278 | * @param capacity number of UChars that can be written to dest |
279 | * @param pErrorCode Standard ICU error code. Its input value must |
280 | * pass the U_SUCCESS() test, or else the function returns |
281 | * immediately. Check for U_FAILURE() on output or use with |
282 | * function chaining. (See User Guide for details.) |
283 | * @return dest |
284 | * @stable ICU 4.4 |
285 | */ |
286 | U_CAPI int32_t U_EXPORT2 |
287 | unorm2_normalize(const UNormalizer2 *norm2, |
288 | const UChar *src, int32_t length, |
289 | UChar *dest, int32_t capacity, |
290 | UErrorCode *pErrorCode); |
291 | /** |
292 | * Appends the normalized form of the second string to the first string |
293 | * (merging them at the boundary) and returns the length of the first string. |
294 | * The result is normalized if the first string was normalized. |
295 | * The first and second strings must be different buffers. |
296 | * @param norm2 UNormalizer2 instance |
297 | * @param first string, should be normalized |
298 | * @param firstLength length of the first string, or -1 if NUL-terminated |
299 | * @param firstCapacity number of UChars that can be written to first |
300 | * @param second string, will be normalized |
301 | * @param secondLength length of the source string, or -1 if NUL-terminated |
302 | * @param pErrorCode Standard ICU error code. Its input value must |
303 | * pass the U_SUCCESS() test, or else the function returns |
304 | * immediately. Check for U_FAILURE() on output or use with |
305 | * function chaining. (See User Guide for details.) |
306 | * @return first |
307 | * @stable ICU 4.4 |
308 | */ |
309 | U_CAPI int32_t U_EXPORT2 |
310 | unorm2_normalizeSecondAndAppend(const UNormalizer2 *norm2, |
311 | UChar *first, int32_t firstLength, int32_t firstCapacity, |
312 | const UChar *second, int32_t secondLength, |
313 | UErrorCode *pErrorCode); |
314 | /** |
315 | * Appends the second string to the first string |
316 | * (merging them at the boundary) and returns the length of the first string. |
317 | * The result is normalized if both the strings were normalized. |
318 | * The first and second strings must be different buffers. |
319 | * @param norm2 UNormalizer2 instance |
320 | * @param first string, should be normalized |
321 | * @param firstLength length of the first string, or -1 if NUL-terminated |
322 | * @param firstCapacity number of UChars that can be written to first |
323 | * @param second string, should be normalized |
324 | * @param secondLength length of the source string, or -1 if NUL-terminated |
325 | * @param pErrorCode Standard ICU error code. Its input value must |
326 | * pass the U_SUCCESS() test, or else the function returns |
327 | * immediately. Check for U_FAILURE() on output or use with |
328 | * function chaining. (See User Guide for details.) |
329 | * @return first |
330 | * @stable ICU 4.4 |
331 | */ |
332 | U_CAPI int32_t U_EXPORT2 |
333 | unorm2_append(const UNormalizer2 *norm2, |
334 | UChar *first, int32_t firstLength, int32_t firstCapacity, |
335 | const UChar *second, int32_t secondLength, |
336 | UErrorCode *pErrorCode); |
337 | |
338 | /** |
339 | * Gets the decomposition mapping of c. |
340 | * Roughly equivalent to normalizing the String form of c |
341 | * on a UNORM2_DECOMPOSE UNormalizer2 instance, but much faster, and except that this function |
342 | * returns a negative value and does not write a string |
343 | * if c does not have a decomposition mapping in this instance's data. |
344 | * This function is independent of the mode of the UNormalizer2. |
345 | * @param norm2 UNormalizer2 instance |
346 | * @param c code point |
347 | * @param decomposition String buffer which will be set to c's |
348 | * decomposition mapping, if there is one. |
349 | * @param capacity number of UChars that can be written to decomposition |
350 | * @param pErrorCode Standard ICU error code. Its input value must |
351 | * pass the U_SUCCESS() test, or else the function returns |
352 | * immediately. Check for U_FAILURE() on output or use with |
353 | * function chaining. (See User Guide for details.) |
354 | * @return the non-negative length of c's decomposition, if there is one; otherwise a negative value |
355 | * @stable ICU 4.6 |
356 | */ |
357 | U_CAPI int32_t U_EXPORT2 |
358 | unorm2_getDecomposition(const UNormalizer2 *norm2, |
359 | UChar32 c, UChar *decomposition, int32_t capacity, |
360 | UErrorCode *pErrorCode); |
361 | |
362 | /** |
363 | * Gets the raw decomposition mapping of c. |
364 | * |
365 | * This is similar to the unorm2_getDecomposition() function but returns the |
366 | * raw decomposition mapping as specified in UnicodeData.txt or |
367 | * (for custom data) in the mapping files processed by the gennorm2 tool. |
368 | * By contrast, unorm2_getDecomposition() returns the processed, |
369 | * recursively-decomposed version of this mapping. |
370 | * |
371 | * When used on a standard NFKC Normalizer2 instance, |
372 | * unorm2_getRawDecomposition() returns the Unicode Decomposition_Mapping (dm) property. |
373 | * |
374 | * When used on a standard NFC Normalizer2 instance, |
375 | * it returns the Decomposition_Mapping only if the Decomposition_Type (dt) is Canonical (Can); |
376 | * in this case, the result contains either one or two code points (=1..4 UChars). |
377 | * |
378 | * This function is independent of the mode of the UNormalizer2. |
379 | * @param norm2 UNormalizer2 instance |
380 | * @param c code point |
381 | * @param decomposition String buffer which will be set to c's |
382 | * raw decomposition mapping, if there is one. |
383 | * @param capacity number of UChars that can be written to decomposition |
384 | * @param pErrorCode Standard ICU error code. Its input value must |
385 | * pass the U_SUCCESS() test, or else the function returns |
386 | * immediately. Check for U_FAILURE() on output or use with |
387 | * function chaining. (See User Guide for details.) |
388 | * @return the non-negative length of c's raw decomposition, if there is one; otherwise a negative value |
389 | * @stable ICU 49 |
390 | */ |
391 | U_CAPI int32_t U_EXPORT2 |
392 | unorm2_getRawDecomposition(const UNormalizer2 *norm2, |
393 | UChar32 c, UChar *decomposition, int32_t capacity, |
394 | UErrorCode *pErrorCode); |
395 | |
396 | /** |
397 | * Performs pairwise composition of a & b and returns the composite if there is one. |
398 | * |
399 | * Returns a composite code point c only if c has a two-way mapping to a+b. |
400 | * In standard Unicode normalization, this means that |
401 | * c has a canonical decomposition to a+b |
402 | * and c does not have the Full_Composition_Exclusion property. |
403 | * |
404 | * This function is independent of the mode of the UNormalizer2. |
405 | * @param norm2 UNormalizer2 instance |
406 | * @param a A (normalization starter) code point. |
407 | * @param b Another code point. |
408 | * @return The non-negative composite code point if there is one; otherwise a negative value. |
409 | * @stable ICU 49 |
410 | */ |
411 | U_CAPI UChar32 U_EXPORT2 |
412 | unorm2_composePair(const UNormalizer2 *norm2, UChar32 a, UChar32 b); |
413 | |
414 | /** |
415 | * Gets the combining class of c. |
416 | * The default implementation returns 0 |
417 | * but all standard implementations return the Unicode Canonical_Combining_Class value. |
418 | * @param norm2 UNormalizer2 instance |
419 | * @param c code point |
420 | * @return c's combining class |
421 | * @stable ICU 49 |
422 | */ |
423 | U_CAPI uint8_t U_EXPORT2 |
424 | unorm2_getCombiningClass(const UNormalizer2 *norm2, UChar32 c); |
425 | |
426 | /** |
427 | * Tests if the string is normalized. |
428 | * Internally, in cases where the quickCheck() method would return "maybe" |
429 | * (which is only possible for the two COMPOSE modes) this method |
430 | * resolves to "yes" or "no" to provide a definitive result, |
431 | * at the cost of doing more work in those cases. |
432 | * @param norm2 UNormalizer2 instance |
433 | * @param s input string |
434 | * @param length length of the string, or -1 if NUL-terminated |
435 | * @param pErrorCode Standard ICU error code. Its input value must |
436 | * pass the U_SUCCESS() test, or else the function returns |
437 | * immediately. Check for U_FAILURE() on output or use with |
438 | * function chaining. (See User Guide for details.) |
439 | * @return true if s is normalized |
440 | * @stable ICU 4.4 |
441 | */ |
442 | U_CAPI UBool U_EXPORT2 |
443 | unorm2_isNormalized(const UNormalizer2 *norm2, |
444 | const UChar *s, int32_t length, |
445 | UErrorCode *pErrorCode); |
446 | |
447 | /** |
448 | * Tests if the string is normalized. |
449 | * For the two COMPOSE modes, the result could be "maybe" in cases that |
450 | * would take a little more work to resolve definitively. |
451 | * Use spanQuickCheckYes() and normalizeSecondAndAppend() for a faster |
452 | * combination of quick check + normalization, to avoid |
453 | * re-checking the "yes" prefix. |
454 | * @param norm2 UNormalizer2 instance |
455 | * @param s input string |
456 | * @param length length of the string, or -1 if NUL-terminated |
457 | * @param pErrorCode Standard ICU error code. Its input value must |
458 | * pass the U_SUCCESS() test, or else the function returns |
459 | * immediately. Check for U_FAILURE() on output or use with |
460 | * function chaining. (See User Guide for details.) |
461 | * @return UNormalizationCheckResult |
462 | * @stable ICU 4.4 |
463 | */ |
464 | U_CAPI UNormalizationCheckResult U_EXPORT2 |
465 | unorm2_quickCheck(const UNormalizer2 *norm2, |
466 | const UChar *s, int32_t length, |
467 | UErrorCode *pErrorCode); |
468 | |
469 | /** |
470 | * Returns the end of the normalized substring of the input string. |
471 | * In other words, with <code>end=spanQuickCheckYes(s, ec);</code> |
472 | * the substring <code>UnicodeString(s, 0, end)</code> |
473 | * will pass the quick check with a "yes" result. |
474 | * |
475 | * The returned end index is usually one or more characters before the |
476 | * "no" or "maybe" character: The end index is at a normalization boundary. |
477 | * (See the class documentation for more about normalization boundaries.) |
478 | * |
479 | * When the goal is a normalized string and most input strings are expected |
480 | * to be normalized already, then call this method, |
481 | * and if it returns a prefix shorter than the input string, |
482 | * copy that prefix and use normalizeSecondAndAppend() for the remainder. |
483 | * @param norm2 UNormalizer2 instance |
484 | * @param s input string |
485 | * @param length length of the string, or -1 if NUL-terminated |
486 | * @param pErrorCode Standard ICU error code. Its input value must |
487 | * pass the U_SUCCESS() test, or else the function returns |
488 | * immediately. Check for U_FAILURE() on output or use with |
489 | * function chaining. (See User Guide for details.) |
490 | * @return "yes" span end index |
491 | * @stable ICU 4.4 |
492 | */ |
493 | U_CAPI int32_t U_EXPORT2 |
494 | unorm2_spanQuickCheckYes(const UNormalizer2 *norm2, |
495 | const UChar *s, int32_t length, |
496 | UErrorCode *pErrorCode); |
497 | |
498 | /** |
499 | * Tests if the character always has a normalization boundary before it, |
500 | * regardless of context. |
501 | * For details see the Normalizer2 base class documentation. |
502 | * @param norm2 UNormalizer2 instance |
503 | * @param c character to test |
504 | * @return true if c has a normalization boundary before it |
505 | * @stable ICU 4.4 |
506 | */ |
507 | U_CAPI UBool U_EXPORT2 |
508 | unorm2_hasBoundaryBefore(const UNormalizer2 *norm2, UChar32 c); |
509 | |
510 | /** |
511 | * Tests if the character always has a normalization boundary after it, |
512 | * regardless of context. |
513 | * For details see the Normalizer2 base class documentation. |
514 | * @param norm2 UNormalizer2 instance |
515 | * @param c character to test |
516 | * @return true if c has a normalization boundary after it |
517 | * @stable ICU 4.4 |
518 | */ |
519 | U_CAPI UBool U_EXPORT2 |
520 | unorm2_hasBoundaryAfter(const UNormalizer2 *norm2, UChar32 c); |
521 | |
522 | /** |
523 | * Tests if the character is normalization-inert. |
524 | * For details see the Normalizer2 base class documentation. |
525 | * @param norm2 UNormalizer2 instance |
526 | * @param c character to test |
527 | * @return true if c is normalization-inert |
528 | * @stable ICU 4.4 |
529 | */ |
530 | U_CAPI UBool U_EXPORT2 |
531 | unorm2_isInert(const UNormalizer2 *norm2, UChar32 c); |
532 | |
533 | /** |
534 | * Compares two strings for canonical equivalence. |
535 | * Further options include case-insensitive comparison and |
536 | * code point order (as opposed to code unit order). |
537 | * |
538 | * Canonical equivalence between two strings is defined as their normalized |
539 | * forms (NFD or NFC) being identical. |
540 | * This function compares strings incrementally instead of normalizing |
541 | * (and optionally case-folding) both strings entirely, |
542 | * improving performance significantly. |
543 | * |
544 | * Bulk normalization is only necessary if the strings do not fulfill the FCD |
545 | * conditions. Only in this case, and only if the strings are relatively long, |
546 | * is memory allocated temporarily. |
547 | * For FCD strings and short non-FCD strings there is no memory allocation. |
548 | * |
549 | * Semantically, this is equivalent to |
550 | * strcmp[CodePointOrder](NFD(foldCase(NFD(s1))), NFD(foldCase(NFD(s2)))) |
551 | * where code point order and foldCase are all optional. |
552 | * |
553 | * UAX 21 2.5 Caseless Matching specifies that for a canonical caseless match |
554 | * the case folding must be performed first, then the normalization. |
555 | * |
556 | * @param s1 First source string. |
557 | * @param length1 Length of first source string, or -1 if NUL-terminated. |
558 | * |
559 | * @param s2 Second source string. |
560 | * @param length2 Length of second source string, or -1 if NUL-terminated. |
561 | * |
562 | * @param options A bit set of options: |
563 | * - U_FOLD_CASE_DEFAULT or 0 is used for default options: |
564 | * Case-sensitive comparison in code unit order, and the input strings |
565 | * are quick-checked for FCD. |
566 | * |
567 | * - UNORM_INPUT_IS_FCD |
568 | * Set if the caller knows that both s1 and s2 fulfill the FCD conditions. |
569 | * If not set, the function will quickCheck for FCD |
570 | * and normalize if necessary. |
571 | * |
572 | * - U_COMPARE_CODE_POINT_ORDER |
573 | * Set to choose code point order instead of code unit order |
574 | * (see u_strCompare for details). |
575 | * |
576 | * - U_COMPARE_IGNORE_CASE |
577 | * Set to compare strings case-insensitively using case folding, |
578 | * instead of case-sensitively. |
579 | * If set, then the following case folding options are used. |
580 | * |
581 | * - Options as used with case-insensitive comparisons, currently: |
582 | * |
583 | * - U_FOLD_CASE_EXCLUDE_SPECIAL_I |
584 | * (see u_strCaseCompare for details) |
585 | * |
586 | * - regular normalization options shifted left by UNORM_COMPARE_NORM_OPTIONS_SHIFT |
587 | * |
588 | * @param pErrorCode ICU error code in/out parameter. |
589 | * Must fulfill U_SUCCESS before the function call. |
590 | * @return <0 or 0 or >0 as usual for string comparisons |
591 | * |
592 | * @see unorm_normalize |
593 | * @see UNORM_FCD |
594 | * @see u_strCompare |
595 | * @see u_strCaseCompare |
596 | * |
597 | * @stable ICU 2.2 |
598 | */ |
599 | U_CAPI int32_t U_EXPORT2 |
600 | unorm_compare(const UChar *s1, int32_t length1, |
601 | const UChar *s2, int32_t length2, |
602 | uint32_t options, |
603 | UErrorCode *pErrorCode); |
604 | |
605 | #endif /* !UCONFIG_NO_NORMALIZATION */ |
606 | #endif /* __UNORM2_H__ */ |
607 | |