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) 2005-2012, International Business Machines |
7 | * Corporation and others. All Rights Reserved. |
8 | * |
9 | ******************************************************************************* |
10 | * file name: ucasemap.h |
11 | * encoding: UTF-8 |
12 | * tab size: 8 (not used) |
13 | * indentation:4 |
14 | * |
15 | * created on: 2005may06 |
16 | * created by: Markus W. Scherer |
17 | * |
18 | * Case mapping service object and functions using it. |
19 | */ |
20 | |
21 | #ifndef __UCASEMAP_H__ |
22 | #define __UCASEMAP_H__ |
23 | |
24 | #include "unicode/utypes.h" |
25 | #include "unicode/stringoptions.h" |
26 | #include "unicode/ustring.h" |
27 | |
28 | #if U_SHOW_CPLUSPLUS_API |
29 | #include "unicode/localpointer.h" |
30 | #endif // U_SHOW_CPLUSPLUS_API |
31 | |
32 | /** |
33 | * \file |
34 | * \brief C API: Unicode case mapping functions using a UCaseMap service object. |
35 | * |
36 | * The service object takes care of memory allocations, data loading, and setup |
37 | * for the attributes, as usual. |
38 | * |
39 | * Currently, the functionality provided here does not overlap with uchar.h |
40 | * and ustring.h, except for ucasemap_toTitle(). |
41 | * |
42 | * ucasemap_utf8XYZ() functions operate directly on UTF-8 strings. |
43 | */ |
44 | |
45 | /** |
46 | * UCaseMap is an opaque service object for newer ICU case mapping functions. |
47 | * Older functions did not use a service object. |
48 | * @stable ICU 3.4 |
49 | */ |
50 | struct UCaseMap; |
51 | typedef struct UCaseMap UCaseMap; /**< C typedef for struct UCaseMap. @stable ICU 3.4 */ |
52 | |
53 | /** |
54 | * Open a UCaseMap service object for a locale and a set of options. |
55 | * The locale ID and options are preprocessed so that functions using the |
56 | * service object need not process them in each call. |
57 | * |
58 | * @param locale ICU locale ID, used for language-dependent |
59 | * upper-/lower-/title-casing according to the Unicode standard. |
60 | * Usual semantics: ""=root, NULL=default locale, etc. |
61 | * @param options Options bit set, used for case folding and string comparisons. |
62 | * Same flags as for u_foldCase(), u_strFoldCase(), |
63 | * u_strCaseCompare(), etc. |
64 | * Use 0 or U_FOLD_CASE_DEFAULT for default behavior. |
65 | * @param pErrorCode Must be a valid pointer to an error code value, |
66 | * which must not indicate a failure before the function call. |
67 | * @return Pointer to a UCaseMap service object, if successful. |
68 | * |
69 | * @see U_FOLD_CASE_DEFAULT |
70 | * @see U_FOLD_CASE_EXCLUDE_SPECIAL_I |
71 | * @see U_TITLECASE_NO_LOWERCASE |
72 | * @see U_TITLECASE_NO_BREAK_ADJUSTMENT |
73 | * @stable ICU 3.4 |
74 | */ |
75 | U_CAPI UCaseMap * U_EXPORT2 |
76 | ucasemap_open(const char *locale, uint32_t options, UErrorCode *pErrorCode); |
77 | |
78 | /** |
79 | * Close a UCaseMap service object. |
80 | * @param csm Object to be closed. |
81 | * @stable ICU 3.4 |
82 | */ |
83 | U_CAPI void U_EXPORT2 |
84 | ucasemap_close(UCaseMap *csm); |
85 | |
86 | #if U_SHOW_CPLUSPLUS_API |
87 | |
88 | U_NAMESPACE_BEGIN |
89 | |
90 | /** |
91 | * \class LocalUCaseMapPointer |
92 | * "Smart pointer" class, closes a UCaseMap via ucasemap_close(). |
93 | * For most methods see the LocalPointerBase base class. |
94 | * |
95 | * @see LocalPointerBase |
96 | * @see LocalPointer |
97 | * @stable ICU 4.4 |
98 | */ |
99 | U_DEFINE_LOCAL_OPEN_POINTER(LocalUCaseMapPointer, UCaseMap, ucasemap_close); |
100 | |
101 | U_NAMESPACE_END |
102 | |
103 | #endif |
104 | |
105 | /** |
106 | * Get the locale ID that is used for language-dependent case mappings. |
107 | * @param csm UCaseMap service object. |
108 | * @return locale ID |
109 | * @stable ICU 3.4 |
110 | */ |
111 | U_CAPI const char * U_EXPORT2 |
112 | ucasemap_getLocale(const UCaseMap *csm); |
113 | |
114 | /** |
115 | * Get the options bit set that is used for case folding and string comparisons. |
116 | * @param csm UCaseMap service object. |
117 | * @return options bit set |
118 | * @stable ICU 3.4 |
119 | */ |
120 | U_CAPI uint32_t U_EXPORT2 |
121 | ucasemap_getOptions(const UCaseMap *csm); |
122 | |
123 | /** |
124 | * Set the locale ID that is used for language-dependent case mappings. |
125 | * |
126 | * @param csm UCaseMap service object. |
127 | * @param locale Locale ID, see ucasemap_open(). |
128 | * @param pErrorCode Must be a valid pointer to an error code value, |
129 | * which must not indicate a failure before the function call. |
130 | * |
131 | * @see ucasemap_open |
132 | * @stable ICU 3.4 |
133 | */ |
134 | U_CAPI void U_EXPORT2 |
135 | ucasemap_setLocale(UCaseMap *csm, const char *locale, UErrorCode *pErrorCode); |
136 | |
137 | /** |
138 | * Set the options bit set that is used for case folding and string comparisons. |
139 | * |
140 | * @param csm UCaseMap service object. |
141 | * @param options Options bit set, see ucasemap_open(). |
142 | * @param pErrorCode Must be a valid pointer to an error code value, |
143 | * which must not indicate a failure before the function call. |
144 | * |
145 | * @see ucasemap_open |
146 | * @stable ICU 3.4 |
147 | */ |
148 | U_CAPI void U_EXPORT2 |
149 | ucasemap_setOptions(UCaseMap *csm, uint32_t options, UErrorCode *pErrorCode); |
150 | |
151 | #if !UCONFIG_NO_BREAK_ITERATION |
152 | |
153 | /** |
154 | * Get the break iterator that is used for titlecasing. |
155 | * Do not modify the returned break iterator. |
156 | * @param csm UCaseMap service object. |
157 | * @return titlecasing break iterator |
158 | * @stable ICU 3.8 |
159 | */ |
160 | U_CAPI const UBreakIterator * U_EXPORT2 |
161 | ucasemap_getBreakIterator(const UCaseMap *csm); |
162 | |
163 | /** |
164 | * Set the break iterator that is used for titlecasing. |
165 | * The UCaseMap service object releases a previously set break iterator |
166 | * and "adopts" this new one, taking ownership of it. |
167 | * It will be released in a subsequent call to ucasemap_setBreakIterator() |
168 | * or ucasemap_close(). |
169 | * |
170 | * Break iterator operations are not thread-safe. Therefore, titlecasing |
171 | * functions use non-const UCaseMap objects. It is not possible to titlecase |
172 | * strings concurrently using the same UCaseMap. |
173 | * |
174 | * @param csm UCaseMap service object. |
175 | * @param iterToAdopt Break iterator to be adopted for titlecasing. |
176 | * @param pErrorCode Must be a valid pointer to an error code value, |
177 | * which must not indicate a failure before the function call. |
178 | * |
179 | * @see ucasemap_toTitle |
180 | * @see ucasemap_utf8ToTitle |
181 | * @stable ICU 3.8 |
182 | */ |
183 | U_CAPI void U_EXPORT2 |
184 | ucasemap_setBreakIterator(UCaseMap *csm, UBreakIterator *iterToAdopt, UErrorCode *pErrorCode); |
185 | |
186 | /** |
187 | * Titlecase a UTF-16 string. This function is almost a duplicate of u_strToTitle(), |
188 | * except that it takes ucasemap_setOptions() into account and has performance |
189 | * advantages from being able to use a UCaseMap object for multiple case mapping |
190 | * operations, saving setup time. |
191 | * |
192 | * Casing is locale-dependent and context-sensitive. |
193 | * Titlecasing uses a break iterator to find the first characters of words |
194 | * that are to be titlecased. It titlecases those characters and lowercases |
195 | * all others. (This can be modified with ucasemap_setOptions().) |
196 | * |
197 | * Note: This function takes a non-const UCaseMap pointer because it will |
198 | * open a default break iterator if no break iterator was set yet, |
199 | * and effectively call ucasemap_setBreakIterator(); |
200 | * also because the break iterator is stateful and will be modified during |
201 | * the iteration. |
202 | * |
203 | * The titlecase break iterator can be provided to customize for arbitrary |
204 | * styles, using rules and dictionaries beyond the standard iterators. |
205 | * The standard titlecase iterator for the root locale implements the |
206 | * algorithm of Unicode TR 21. |
207 | * |
208 | * This function uses only the setText(), first() and next() methods of the |
209 | * provided break iterator. |
210 | * |
211 | * The result may be longer or shorter than the original. |
212 | * The source string and the destination buffer must not overlap. |
213 | * |
214 | * @param csm UCaseMap service object. This pointer is non-const! |
215 | * See the note above for details. |
216 | * @param dest A buffer for the result string. The result will be NUL-terminated if |
217 | * the buffer is large enough. |
218 | * The contents is undefined in case of failure. |
219 | * @param destCapacity The size of the buffer (number of UChars). If it is 0, then |
220 | * dest may be NULL and the function will only return the length of the result |
221 | * without writing any of the result string. |
222 | * @param src The original string. |
223 | * @param srcLength The length of the original string. If -1, then src must be NUL-terminated. |
224 | * @param pErrorCode Must be a valid pointer to an error code value, |
225 | * which must not indicate a failure before the function call. |
226 | * @return The length of the result string, if successful - or in case of a buffer overflow, |
227 | * in which case it will be greater than destCapacity. |
228 | * |
229 | * @see u_strToTitle |
230 | * @stable ICU 3.8 |
231 | */ |
232 | U_CAPI int32_t U_EXPORT2 |
233 | ucasemap_toTitle(UCaseMap *csm, |
234 | UChar *dest, int32_t destCapacity, |
235 | const UChar *src, int32_t srcLength, |
236 | UErrorCode *pErrorCode); |
237 | |
238 | #endif // UCONFIG_NO_BREAK_ITERATION |
239 | |
240 | /** |
241 | * Lowercase the characters in a UTF-8 string. |
242 | * Casing is locale-dependent and context-sensitive. |
243 | * The result may be longer or shorter than the original. |
244 | * The source string and the destination buffer must not overlap. |
245 | * |
246 | * @param csm UCaseMap service object. |
247 | * @param dest A buffer for the result string. The result will be NUL-terminated if |
248 | * the buffer is large enough. |
249 | * The contents is undefined in case of failure. |
250 | * @param destCapacity The size of the buffer (number of bytes). If it is 0, then |
251 | * dest may be NULL and the function will only return the length of the result |
252 | * without writing any of the result string. |
253 | * @param src The original string. |
254 | * @param srcLength The length of the original string. If -1, then src must be NUL-terminated. |
255 | * @param pErrorCode Must be a valid pointer to an error code value, |
256 | * which must not indicate a failure before the function call. |
257 | * @return The length of the result string, if successful - or in case of a buffer overflow, |
258 | * in which case it will be greater than destCapacity. |
259 | * |
260 | * @see u_strToLower |
261 | * @stable ICU 3.4 |
262 | */ |
263 | U_CAPI int32_t U_EXPORT2 |
264 | ucasemap_utf8ToLower(const UCaseMap *csm, |
265 | char *dest, int32_t destCapacity, |
266 | const char *src, int32_t srcLength, |
267 | UErrorCode *pErrorCode); |
268 | |
269 | /** |
270 | * Uppercase the characters in a UTF-8 string. |
271 | * Casing is locale-dependent and context-sensitive. |
272 | * The result may be longer or shorter than the original. |
273 | * The source string and the destination buffer must not overlap. |
274 | * |
275 | * @param csm UCaseMap service object. |
276 | * @param dest A buffer for the result string. The result will be NUL-terminated if |
277 | * the buffer is large enough. |
278 | * The contents is undefined in case of failure. |
279 | * @param destCapacity The size of the buffer (number of bytes). If it is 0, then |
280 | * dest may be NULL and the function will only return the length of the result |
281 | * without writing any of the result string. |
282 | * @param src The original string. |
283 | * @param srcLength The length of the original string. If -1, then src must be NUL-terminated. |
284 | * @param pErrorCode Must be a valid pointer to an error code value, |
285 | * which must not indicate a failure before the function call. |
286 | * @return The length of the result string, if successful - or in case of a buffer overflow, |
287 | * in which case it will be greater than destCapacity. |
288 | * |
289 | * @see u_strToUpper |
290 | * @stable ICU 3.4 |
291 | */ |
292 | U_CAPI int32_t U_EXPORT2 |
293 | ucasemap_utf8ToUpper(const UCaseMap *csm, |
294 | char *dest, int32_t destCapacity, |
295 | const char *src, int32_t srcLength, |
296 | UErrorCode *pErrorCode); |
297 | |
298 | #if !UCONFIG_NO_BREAK_ITERATION |
299 | |
300 | /** |
301 | * Titlecase a UTF-8 string. |
302 | * Casing is locale-dependent and context-sensitive. |
303 | * Titlecasing uses a break iterator to find the first characters of words |
304 | * that are to be titlecased. It titlecases those characters and lowercases |
305 | * all others. (This can be modified with ucasemap_setOptions().) |
306 | * |
307 | * Note: This function takes a non-const UCaseMap pointer because it will |
308 | * open a default break iterator if no break iterator was set yet, |
309 | * and effectively call ucasemap_setBreakIterator(); |
310 | * also because the break iterator is stateful and will be modified during |
311 | * the iteration. |
312 | * |
313 | * The titlecase break iterator can be provided to customize for arbitrary |
314 | * styles, using rules and dictionaries beyond the standard iterators. |
315 | * The standard titlecase iterator for the root locale implements the |
316 | * algorithm of Unicode TR 21. |
317 | * |
318 | * This function uses only the setUText(), first(), next() and close() methods of the |
319 | * provided break iterator. |
320 | * |
321 | * The result may be longer or shorter than the original. |
322 | * The source string and the destination buffer must not overlap. |
323 | * |
324 | * @param csm UCaseMap service object. This pointer is non-const! |
325 | * See the note above for details. |
326 | * @param dest A buffer for the result string. The result will be NUL-terminated if |
327 | * the buffer is large enough. |
328 | * The contents is undefined in case of failure. |
329 | * @param destCapacity The size of the buffer (number of bytes). If it is 0, then |
330 | * dest may be NULL and the function will only return the length of the result |
331 | * without writing any of the result string. |
332 | * @param src The original string. |
333 | * @param srcLength The length of the original string. If -1, then src must be NUL-terminated. |
334 | * @param pErrorCode Must be a valid pointer to an error code value, |
335 | * which must not indicate a failure before the function call. |
336 | * @return The length of the result string, if successful - or in case of a buffer overflow, |
337 | * in which case it will be greater than destCapacity. |
338 | * |
339 | * @see u_strToTitle |
340 | * @see U_TITLECASE_NO_LOWERCASE |
341 | * @see U_TITLECASE_NO_BREAK_ADJUSTMENT |
342 | * @stable ICU 3.8 |
343 | */ |
344 | U_CAPI int32_t U_EXPORT2 |
345 | ucasemap_utf8ToTitle(UCaseMap *csm, |
346 | char *dest, int32_t destCapacity, |
347 | const char *src, int32_t srcLength, |
348 | UErrorCode *pErrorCode); |
349 | |
350 | #endif |
351 | |
352 | /** |
353 | * Case-folds the characters in a UTF-8 string. |
354 | * |
355 | * Case-folding is locale-independent and not context-sensitive, |
356 | * but there is an option for whether to include or exclude mappings for dotted I |
357 | * and dotless i that are marked with 'T' in CaseFolding.txt. |
358 | * |
359 | * The result may be longer or shorter than the original. |
360 | * The source string and the destination buffer must not overlap. |
361 | * |
362 | * @param csm UCaseMap service object. |
363 | * @param dest A buffer for the result string. The result will be NUL-terminated if |
364 | * the buffer is large enough. |
365 | * The contents is undefined in case of failure. |
366 | * @param destCapacity The size of the buffer (number of bytes). If it is 0, then |
367 | * dest may be NULL and the function will only return the length of the result |
368 | * without writing any of the result string. |
369 | * @param src The original string. |
370 | * @param srcLength The length of the original string. If -1, then src must be NUL-terminated. |
371 | * @param pErrorCode Must be a valid pointer to an error code value, |
372 | * which must not indicate a failure before the function call. |
373 | * @return The length of the result string, if successful - or in case of a buffer overflow, |
374 | * in which case it will be greater than destCapacity. |
375 | * |
376 | * @see u_strFoldCase |
377 | * @see ucasemap_setOptions |
378 | * @see U_FOLD_CASE_DEFAULT |
379 | * @see U_FOLD_CASE_EXCLUDE_SPECIAL_I |
380 | * @stable ICU 3.8 |
381 | */ |
382 | U_CAPI int32_t U_EXPORT2 |
383 | ucasemap_utf8FoldCase(const UCaseMap *csm, |
384 | char *dest, int32_t destCapacity, |
385 | const char *src, int32_t srcLength, |
386 | UErrorCode *pErrorCode); |
387 | |
388 | #endif |
389 | |