2
******************************************************************************
3
* Copyright (C) 1996-2010, International Business Machines Corporation and others.
5
******************************************************************************
11
#include "unicode/utypes.h"
12
#include "unicode/uloc.h"
13
#include "unicode/utext.h"
14
#include "unicode/localpointer.h"
17
* A text-break iterator.
18
* For usage in C programs.
20
#ifndef UBRK_TYPEDEF_UBREAK_ITERATOR
21
# define UBRK_TYPEDEF_UBREAK_ITERATOR
23
* Opaque type representing an ICU Break iterator object.
26
typedef struct UBreakIterator UBreakIterator;
29
#if !UCONFIG_NO_BREAK_ITERATION
31
#include "unicode/parseerr.h"
35
* \brief C API: BreakIterator
37
* <h2> BreakIterator C API </h2>
39
* The BreakIterator C API defines methods for finding the location
40
* of boundaries in text. Pointer to a UBreakIterator maintain a
41
* current position and scan over text returning the index of characters
42
* where boundaries occur.
44
* Line boundary analysis determines where a text string can be broken
45
* when line-wrapping. The mechanism correctly handles punctuation and
48
* Sentence boundary analysis allows selection with correct
49
* interpretation of periods within numbers and abbreviations, and
50
* trailing punctuation marks such as quotation marks and parentheses.
52
* Word boundary analysis is used by search and replace functions, as
53
* well as within text editing applications that allow the user to
54
* select words with a double click. Word selection provides correct
55
* interpretation of punctuation marks within and following
56
* words. Characters that are not part of a word, such as symbols or
57
* punctuation marks, have word-breaks on both sides.
59
* Character boundary analysis identifies the boundaries of
60
* "Extended Grapheme Clusters", which are groupings of codepoints
61
* that should be treated as character-like units for many text operations.
62
* Please see Unicode Standard Annex #29, Unicode Text Segmentation,
63
* http://www.unicode.org/reports/tr29/ for additional information
64
* on grapheme clusters and guidelines on their use.
66
* Title boundary analysis locates all positions,
67
* typically starts of words, that should be set to Title Case
68
* when title casing the text.
70
* The text boundary positions are found according to the rules
71
* described in Unicode Standard Annex #29, Text Boundaries, and
72
* Unicode Standard Annex #14, Line Breaking Properties. These
73
* are available at http://www.unicode.org/reports/tr14/ and
74
* http://www.unicode.org/reports/tr29/.
76
* In addition to the plain C API defined in this header file, an
77
* object oriented C++ API with equivalent functionality is defined in the
80
* Code snippets illustrating the use of the Break Iterator APIs
81
* are available in the ICU User Guide,
82
* http://icu-project.org/userguide/boundaryAnalysis.html
83
* and in the sample program icu/source/samples/break/break.cpp
86
/** The possible types of text boundaries. @stable ICU 2.0 */
87
typedef enum UBreakIteratorType {
88
/** Character breaks @stable ICU 2.0 */
90
/** Word breaks @stable ICU 2.0 */
92
/** Line breaks @stable ICU 2.0 */
94
/** Sentence breaks @stable ICU 2.0 */
97
#ifndef U_HIDE_DEPRECATED_API
100
* The iterator created using this type locates title boundaries as described for
101
* Unicode 3.2 only. For Unicode 4.0 and above title boundary iteration,
102
* please use Word Boundary iterator.
104
* @deprecated ICU 2.8 Use the word break iterator for titlecasing for Unicode 4 and later.
107
#endif /* U_HIDE_DEPRECATED_API */
109
} UBreakIteratorType;
111
/** Value indicating all text boundaries have been returned.
114
#define UBRK_DONE ((int32_t) -1)
118
* Enum constants for the word break tags returned by
119
* getRuleStatus(). A range of values is defined for each category of
120
* word, to allow for further subdivisions of a category in future releases.
121
* Applications should check for tag values falling within the range, rather
122
* than for single individual values.
125
typedef enum UWordBreak {
126
/** Tag value for "words" that do not fit into any of other categories.
127
* Includes spaces and most punctuation. */
129
/** Upper bound for tags for uncategorized words. */
130
UBRK_WORD_NONE_LIMIT = 100,
131
/** Tag value for words that appear to be numbers, lower limit. */
132
UBRK_WORD_NUMBER = 100,
133
/** Tag value for words that appear to be numbers, upper limit. */
134
UBRK_WORD_NUMBER_LIMIT = 200,
135
/** Tag value for words that contain letters, excluding
136
* hiragana, katakana or ideographic characters, lower limit. */
137
UBRK_WORD_LETTER = 200,
138
/** Tag value for words containing letters, upper limit */
139
UBRK_WORD_LETTER_LIMIT = 300,
140
/** Tag value for words containing kana characters, lower limit */
141
UBRK_WORD_KANA = 300,
142
/** Tag value for words containing kana characters, upper limit */
143
UBRK_WORD_KANA_LIMIT = 400,
144
/** Tag value for words containing ideographic characters, lower limit */
145
UBRK_WORD_IDEO = 400,
146
/** Tag value for words containing ideographic characters, upper limit */
147
UBRK_WORD_IDEO_LIMIT = 500
151
* Enum constants for the line break tags returned by getRuleStatus().
152
* A range of values is defined for each category of
153
* word, to allow for further subdivisions of a category in future releases.
154
* Applications should check for tag values falling within the range, rather
155
* than for single individual values.
158
typedef enum ULineBreakTag {
159
/** Tag value for soft line breaks, positions at which a line break
160
* is acceptable but not required */
162
/** Upper bound for soft line breaks. */
163
UBRK_LINE_SOFT_LIMIT = 100,
164
/** Tag value for a hard, or mandatory line break */
165
UBRK_LINE_HARD = 100,
166
/** Upper bound for hard line breaks. */
167
UBRK_LINE_HARD_LIMIT = 200
173
* Enum constants for the sentence break tags returned by getRuleStatus().
174
* A range of values is defined for each category of
175
* sentence, to allow for further subdivisions of a category in future releases.
176
* Applications should check for tag values falling within the range, rather
177
* than for single individual values.
180
typedef enum USentenceBreakTag {
181
/** Tag value for for sentences ending with a sentence terminator
182
* ('.', '?', '!', etc.) character, possibly followed by a
183
* hard separator (CR, LF, PS, etc.)
185
UBRK_SENTENCE_TERM = 0,
186
/** Upper bound for tags for sentences ended by sentence terminators. */
187
UBRK_SENTENCE_TERM_LIMIT = 100,
188
/** Tag value for for sentences that do not contain an ending
189
* sentence terminator ('.', '?', '!', etc.) character, but
190
* are ended only by a hard separator (CR, LF, PS, etc.) or end of input.
192
UBRK_SENTENCE_SEP = 100,
193
/** Upper bound for tags for sentences ended by a separator. */
194
UBRK_SENTENCE_SEP_LIMIT = 200
195
/** Tag value for a hard, or mandatory line break */
200
* Open a new UBreakIterator for locating text boundaries for a specified locale.
201
* A UBreakIterator may be used for detecting character, line, word,
202
* and sentence breaks in text.
203
* @param type The type of UBreakIterator to open: one of UBRK_CHARACTER, UBRK_WORD,
204
* UBRK_LINE, UBRK_SENTENCE
205
* @param locale The locale specifying the text-breaking conventions.
206
* @param text The text to be iterated over.
207
* @param textLength The number of characters in text, or -1 if null-terminated.
208
* @param status A UErrorCode to receive any errors.
209
* @return A UBreakIterator for the specified locale.
210
* @see ubrk_openRules
213
U_STABLE UBreakIterator* U_EXPORT2
214
ubrk_open(UBreakIteratorType type,
221
* Open a new UBreakIterator for locating text boundaries using specified breaking rules.
222
* The rule syntax is ... (TBD)
223
* @param rules A set of rules specifying the text breaking conventions.
224
* @param rulesLength The number of characters in rules, or -1 if null-terminated.
225
* @param text The text to be iterated over. May be null, in which case ubrk_setText() is
226
* used to specify the text to be iterated.
227
* @param textLength The number of characters in text, or -1 if null-terminated.
228
* @param parseErr Receives position and context information for any syntax errors
229
* detected while parsing the rules.
230
* @param status A UErrorCode to receive any errors.
231
* @return A UBreakIterator for the specified rules.
235
U_STABLE UBreakIterator* U_EXPORT2
236
ubrk_openRules(const UChar *rules,
240
UParseError *parseErr,
244
* Thread safe cloning operation
245
* @param bi iterator to be cloned
246
* @param stackBuffer user allocated space for the new clone. If NULL new memory will be allocated.
247
* If buffer is not large enough, new memory will be allocated.
248
* Clients can use the U_BRK_SAFECLONE_BUFFERSIZE. This will probably be enough to avoid memory allocations.
249
* @param pBufferSize pointer to size of allocated space.
250
* If *pBufferSize == 0, a sufficient size for use in cloning will
251
* be returned ('pre-flighting')
252
* If *pBufferSize is not enough for a stack-based safe clone,
253
* new memory will be allocated.
254
* @param status to indicate whether the operation went on smoothly or there were errors
255
* An informational status value, U_SAFECLONE_ALLOCATED_ERROR, is used if any allocations were necessary.
256
* @return pointer to the new clone
259
U_STABLE UBreakIterator * U_EXPORT2
261
const UBreakIterator *bi,
263
int32_t *pBufferSize,
267
* A recommended size (in bytes) for the memory buffer to be passed to ubrk_saveClone().
270
#define U_BRK_SAFECLONE_BUFFERSIZE 512
273
* Close a UBreakIterator.
274
* Once closed, a UBreakIterator may no longer be used.
275
* @param bi The break iterator to close.
278
U_STABLE void U_EXPORT2
279
ubrk_close(UBreakIterator *bi);
281
#if U_SHOW_CPLUSPLUS_API
286
* \class LocalUBreakIteratorPointer
287
* "Smart pointer" class, closes a UBreakIterator via ubrk_close().
288
* For most methods see the LocalPointerBase base class.
290
* @see LocalPointerBase
294
U_DEFINE_LOCAL_OPEN_POINTER(LocalUBreakIteratorPointer, UBreakIterator, ubrk_close);
301
* Sets an existing iterator to point to a new piece of text
302
* @param bi The iterator to use
303
* @param text The text to be set
304
* @param textLength The length of the text
305
* @param status The error code
308
U_STABLE void U_EXPORT2
309
ubrk_setText(UBreakIterator* bi,
316
* Sets an existing iterator to point to a new piece of text
317
* @param bi The iterator to use
318
* @param text The text to be set.
319
* This function makes a shallow clone of the supplied UText. This means
320
* that the caller is free to immediately close or otherwise reuse the
321
* UText that was passed as a parameter, but that the underlying text itself
322
* must not be altered while being referenced by the break iterator.
323
* @param status The error code
326
U_STABLE void U_EXPORT2
327
ubrk_setUText(UBreakIterator* bi,
334
* Determine the most recently-returned text boundary.
336
* @param bi The break iterator to use.
337
* @return The character index most recently returned by \ref ubrk_next, \ref ubrk_previous,
338
* \ref ubrk_first, or \ref ubrk_last.
341
U_STABLE int32_t U_EXPORT2
342
ubrk_current(const UBreakIterator *bi);
345
* Determine the text boundary following the current text boundary.
347
* @param bi The break iterator to use.
348
* @return The character index of the next text boundary, or UBRK_DONE
349
* if all text boundaries have been returned.
353
U_STABLE int32_t U_EXPORT2
354
ubrk_next(UBreakIterator *bi);
357
* Determine the text boundary preceding the current text boundary.
359
* @param bi The break iterator to use.
360
* @return The character index of the preceding text boundary, or UBRK_DONE
361
* if all text boundaries have been returned.
365
U_STABLE int32_t U_EXPORT2
366
ubrk_previous(UBreakIterator *bi);
369
* Determine the index of the first character in the text being scanned.
370
* This is not always the same as index 0 of the text.
371
* @param bi The break iterator to use.
372
* @return The character index of the first character in the text being scanned.
376
U_STABLE int32_t U_EXPORT2
377
ubrk_first(UBreakIterator *bi);
380
* Determine the index immediately <EM>beyond</EM> the last character in the text being
382
* This is not the same as the last character.
383
* @param bi The break iterator to use.
384
* @return The character offset immediately <EM>beyond</EM> the last character in the
385
* text being scanned.
389
U_STABLE int32_t U_EXPORT2
390
ubrk_last(UBreakIterator *bi);
393
* Determine the text boundary preceding the specified offset.
394
* The value returned is always smaller than offset, or UBRK_DONE.
395
* @param bi The break iterator to use.
396
* @param offset The offset to begin scanning.
397
* @return The text boundary preceding offset, or UBRK_DONE.
398
* @see ubrk_following
401
U_STABLE int32_t U_EXPORT2
402
ubrk_preceding(UBreakIterator *bi,
406
* Determine the text boundary following the specified offset.
407
* The value returned is always greater than offset, or UBRK_DONE.
408
* @param bi The break iterator to use.
409
* @param offset The offset to begin scanning.
410
* @return The text boundary following offset, or UBRK_DONE.
411
* @see ubrk_preceding
414
U_STABLE int32_t U_EXPORT2
415
ubrk_following(UBreakIterator *bi,
419
* Get a locale for which text breaking information is available.
420
* A UBreakIterator in a locale returned by this function will perform the correct
421
* text breaking for the locale.
422
* @param index The index of the desired locale.
423
* @return A locale for which number text breaking information is available, or 0 if none.
424
* @see ubrk_countAvailable
427
U_STABLE const char* U_EXPORT2
428
ubrk_getAvailable(int32_t index);
431
* Determine how many locales have text breaking information available.
432
* This function is most useful as determining the loop ending condition for
433
* calls to \ref ubrk_getAvailable.
434
* @return The number of locales for which text breaking information is available.
435
* @see ubrk_getAvailable
438
U_STABLE int32_t U_EXPORT2
439
ubrk_countAvailable(void);
443
* Returns true if the specfied position is a boundary position. As a side
444
* effect, leaves the iterator pointing to the first boundary position at
446
* @param bi The break iterator to use.
447
* @param offset the offset to check.
448
* @return True if "offset" is a boundary position.
451
U_STABLE UBool U_EXPORT2
452
ubrk_isBoundary(UBreakIterator *bi, int32_t offset);
455
* Return the status from the break rule that determined the most recently
456
* returned break position. The values appear in the rule source
457
* within brackets, {123}, for example. For rules that do not specify a
458
* status, a default value of 0 is returned.
460
* For word break iterators, the possible values are defined in enum UWordBreak.
463
U_STABLE int32_t U_EXPORT2
464
ubrk_getRuleStatus(UBreakIterator *bi);
467
* Get the statuses from the break rules that determined the most recently
468
* returned break position. The values appear in the rule source
469
* within brackets, {123}, for example. The default status value for rules
470
* that do not explicitly provide one is zero.
472
* For word break iterators, the possible values are defined in enum UWordBreak.
473
* @param bi The break iterator to use
474
* @param fillInVec an array to be filled in with the status values.
475
* @param capacity the length of the supplied vector. A length of zero causes
476
* the function to return the number of status values, in the
477
* normal way, without attemtping to store any values.
478
* @param status receives error codes.
479
* @return The number of rule status values from rules that determined
480
* the most recent boundary returned by the break iterator.
483
U_STABLE int32_t U_EXPORT2
484
ubrk_getRuleStatusVec(UBreakIterator *bi, int32_t *fillInVec, int32_t capacity, UErrorCode *status);
487
* Return the locale of the break iterator. You can choose between the valid and
489
* @param bi break iterator
490
* @param type locale type (valid or actual)
491
* @param status error code
492
* @return locale string
495
U_STABLE const char* U_EXPORT2
496
ubrk_getLocaleByType(const UBreakIterator *bi, ULocDataLocaleType type, UErrorCode* status);
499
#endif /* #if !UCONFIG_NO_BREAK_ITERATION */