icuSources/common/unicode/ubrk.h

   1 /*
   2 ******************************************************************************
   3 * Copyright (C) 1996-2015, International Business Machines Corporation and others.
   4 * All Rights Reserved.
   5 ******************************************************************************
   6 */
   7
   8 #ifndef UBRK_H
   9 #define UBRK_H
  10
  11 #include "unicode/utypes.h"
  12 #include "unicode/uloc.h"
  13 #include "unicode/utext.h"
  14 #include "unicode/localpointer.h"
  15
  16 /**
  17  * A text-break iterator.
  18  *  For usage in C programs.
  19  */
  20 #ifndef UBRK_TYPEDEF_UBREAK_ITERATOR
  21 #   define UBRK_TYPEDEF_UBREAK_ITERATOR
  22     /**
  23      *  Opaque type representing an ICU Break iterator object.
  24      *  @stable ICU 2.0
  25      */
  26     typedef struct UBreakIterator UBreakIterator;
  27 #endif
  28
  29 #if !UCONFIG_NO_BREAK_ITERATION
  30
  31 #include "unicode/parseerr.h"
  32
  33 /**
  34  * \file
  35  * \brief C API: BreakIterator
  36  *
  37  * <h2> BreakIterator C API </h2>
  38  *
  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.
  43  * <p>
  44  * Line boundary analysis determines where a text string can be broken
  45  * when line-wrapping. The mechanism correctly handles punctuation and
  46  * hyphenated words.
  47  * <p>
  48  * Note: The locale keyword "lb" can be used to modify line break
  49  * behavior according to the CSS level 3 line-break options, see
  50  * <http://dev.w3.org/csswg/css-text/#line-breaking>. For example:
  51  * "ja@lb=strict", "zh@lb=loose".
  52  * <p>
  53  * Sentence boundary analysis allows selection with correct
  54  * interpretation of periods within numbers and abbreviations, and
  55  * trailing punctuation marks such as quotation marks and parentheses.
  56  * <p>
  57  * Note: The locale keyword "ss" can be used to enable use of
  58  * segmentation suppression data (preventing breaks in English after
  59  * abbreviations such as "Mr." or "Est.", for example), as follows:
  60  * "en@ss=standard".
  61  * <p>
  62  * Word boundary analysis is used by search and replace functions, as
  63  * well as within text editing applications that allow the user to
  64  * select words with a double click. Word selection provides correct
  65  * interpretation of punctuation marks within and following
  66  * words. Characters that are not part of a word, such as symbols or
  67  * punctuation marks, have word-breaks on both sides.
  68  * <p>
  69  * Character boundary analysis identifies the boundaries of
  70  * "Extended Grapheme Clusters", which are groupings of codepoints
  71  * that should be treated as character-like units for many text operations.
  72  * Please see Unicode Standard Annex #29, Unicode Text Segmentation,
  73  * http://www.unicode.org/reports/tr29/ for additional information
  74  * on grapheme clusters and guidelines on their use.
  75  * <p>
  76  * Title boundary analysis locates all positions,
  77  * typically starts of words, that should be set to Title Case
  78  * when title casing the text.
  79  * <p>
  80  * The text boundary positions are found according to the rules
  81  * described in Unicode Standard Annex #29, Text Boundaries, and
  82  * Unicode Standard Annex #14, Line Breaking Properties.  These
  83  * are available at http://www.unicode.org/reports/tr14/ and
  84  * http://www.unicode.org/reports/tr29/.
  85  * <p>
  86  * In addition to the plain C API defined in this header file, an
  87  * object oriented C++ API with equivalent functionality is defined in the
  88  * file brkiter.h.
  89  * <p>
  90  * Code snippets illustrating the use of the Break Iterator APIs
  91  * are available in the ICU User Guide,
  92  * http://icu-project.org/userguide/boundaryAnalysis.html
  93  * and in the sample program icu/source/samples/break/break.cpp
  94  */
  95
  96 /** The possible types of text boundaries.  @stable ICU 2.0 */
  97 typedef enum UBreakIteratorType {
  98   /** Character breaks  @stable ICU 2.0 */
  99   UBRK_CHARACTER = 0,
 100   /** Word breaks @stable ICU 2.0 */
 101   UBRK_WORD = 1,
 102   /** Line breaks @stable ICU 2.0 */
 103   UBRK_LINE = 2,
 104   /** Sentence breaks @stable ICU 2.0 */
 105   UBRK_SENTENCE = 3,
 106
 107 #ifndef U_HIDE_DEPRECATED_API
 108   /**
 109    * Title Case breaks
 110    * The iterator created using this type locates title boundaries as described for
 111    * Unicode 3.2 only. For Unicode 4.0 and above title boundary iteration,
 112    * please use Word Boundary iterator.
 113    *
 114    * @deprecated ICU 2.8 Use the word break iterator for titlecasing for Unicode 4 and later.
 115    */
 116   UBRK_TITLE = 4,
 117 #endif /* U_HIDE_DEPRECATED_API */
 118   UBRK_COUNT = 5
 119 } UBreakIteratorType;
 120
 121 /** Value indicating all text boundaries have been returned.
 122  *  @stable ICU 2.0
 123  */
 124 #define UBRK_DONE ((int32_t) -1)
 125
 126
 127 /**
 128  *  Enum constants for the word break tags returned by
 129  *  getRuleStatus().  A range of values is defined for each category of
 130  *  word, to allow for further subdivisions of a category in future releases.
 131  *  Applications should check for tag values falling within the range, rather
 132  *  than for single individual values.
 133  *  @stable ICU 2.2
 134 */
 135 typedef enum UWordBreak {
 136     /** Tag value for "words" that do not fit into any of other categories.
 137      *  Includes spaces and most punctuation. */
 138     UBRK_WORD_NONE           = 0,
 139     /** Upper bound for tags for uncategorized words. */
 140     UBRK_WORD_NONE_LIMIT     = 100,
 141     /** Tag value for words that appear to be numbers, lower limit.    */
 142     UBRK_WORD_NUMBER         = 100,
 143     /** Tag value for words that appear to be numbers, upper limit.    */
 144     UBRK_WORD_NUMBER_LIMIT   = 200,
 145     /** Tag value for words that contain letters, excluding
 146      *  hiragana, katakana or ideographic characters, lower limit.    */
 147     UBRK_WORD_LETTER         = 200,
 148     /** Tag value for words containing letters, upper limit  */
 149     UBRK_WORD_LETTER_LIMIT   = 300,
 150     /** Tag value for words containing kana characters, lower limit */
 151     UBRK_WORD_KANA           = 300,
 152     /** Tag value for words containing kana characters, upper limit */
 153     UBRK_WORD_KANA_LIMIT     = 400,
 154     /** Tag value for words containing ideographic characters, lower limit */
 155     UBRK_WORD_IDEO           = 400,
 156     /** Tag value for words containing ideographic characters, upper limit */
 157     UBRK_WORD_IDEO_LIMIT     = 500
 158 } UWordBreak;
 159
 160 /**
 161  *  Enum constants for the line break tags returned by getRuleStatus().
 162  *  A range of values is defined for each category of
 163  *  word, to allow for further subdivisions of a category in future releases.
 164  *  Applications should check for tag values falling within the range, rather
 165  *  than for single individual values.
 166  *  @stable ICU 2.8
 167 */
 168 typedef enum ULineBreakTag {
 169     /** Tag value for soft line breaks, positions at which a line break
 170       *  is acceptable but not required                */
 171     UBRK_LINE_SOFT            = 0,
 172     /** Upper bound for soft line breaks.              */
 173     UBRK_LINE_SOFT_LIMIT      = 100,
 174     /** Tag value for a hard, or mandatory line break  */
 175     UBRK_LINE_HARD            = 100,
 176     /** Upper bound for hard line breaks.              */
 177     UBRK_LINE_HARD_LIMIT      = 200
 178 } ULineBreakTag;
 179
 180
 181
 182 /**
 183  *  Enum constants for the sentence break tags returned by getRuleStatus().
 184  *  A range of values is defined for each category of
 185  *  sentence, to allow for further subdivisions of a category in future releases.
 186  *  Applications should check for tag values falling within the range, rather
 187  *  than for single individual values.
 188  *  @stable ICU 2.8
 189 */
 190 typedef enum USentenceBreakTag {
 191     /** Tag value for for sentences  ending with a sentence terminator
 192       * ('.', '?', '!', etc.) character, possibly followed by a
 193       * hard separator (CR, LF, PS, etc.)
 194       */
 195     UBRK_SENTENCE_TERM       = 0,
 196     /** Upper bound for tags for sentences ended by sentence terminators.    */
 197     UBRK_SENTENCE_TERM_LIMIT = 100,
 198     /** Tag value for for sentences that do not contain an ending
 199       * sentence terminator ('.', '?', '!', etc.) character, but
 200       * are ended only by a hard separator (CR, LF, PS, etc.) or end of input.
 201       */
 202     UBRK_SENTENCE_SEP        = 100,
 203     /** Upper bound for tags for sentences ended by a separator.              */
 204     UBRK_SENTENCE_SEP_LIMIT  = 200
 205     /** Tag value for a hard, or mandatory line break  */
 206 } USentenceBreakTag;
 207
 208
 209 /**
 210  * Open a new UBreakIterator for locating text boundaries for a specified locale.
 211  * A UBreakIterator may be used for detecting character, line, word,
 212  * and sentence breaks in text.
 213  * @param type The type of UBreakIterator to open: one of UBRK_CHARACTER, UBRK_WORD,
 214  * UBRK_LINE, UBRK_SENTENCE
 215  * @param locale The locale specifying the text-breaking conventions. Note that
 216  * locale keys such as "lb" and "ss" may be used to modify text break behavior,
 217  * see general discussion of BreakIterator C API.
 218  * @param text The text to be iterated over.
 219  * @param textLength The number of characters in text, or -1 if null-terminated.
 220  * @param status A UErrorCode to receive any errors.
 221  * @return A UBreakIterator for the specified locale.
 222  * @see ubrk_openRules
 223  * @stable ICU 2.0
 224  */
 225 U_STABLE UBreakIterator* U_EXPORT2
 226 ubrk_open(UBreakIteratorType type,
 227       const char *locale,
 228       const UChar *text,
 229       int32_t textLength,
 230       UErrorCode *status);
 231
 232 /**
 233  * Open a new UBreakIterator for locating text boundaries using specified breaking rules.
 234  * The rule syntax is ... (TBD)
 235  * @param rules A set of rules specifying the text breaking conventions.
 236  * @param rulesLength The number of characters in rules, or -1 if null-terminated.
 237  * @param text The text to be iterated over.  May be null, in which case ubrk_setText() is
 238  *        used to specify the text to be iterated.
 239  * @param textLength The number of characters in text, or -1 if null-terminated.
 240  * @param parseErr   Receives position and context information for any syntax errors
 241  *                   detected while parsing the rules.
 242  * @param status A UErrorCode to receive any errors.
 243  * @return A UBreakIterator for the specified rules.
 244  * @see ubrk_open
 245  * @stable ICU 2.2
 246  */
 247 U_STABLE UBreakIterator* U_EXPORT2
 248 ubrk_openRules(const UChar     *rules,
 249                int32_t         rulesLength,
 250                const UChar     *text,
 251                int32_t          textLength,
 252                UParseError     *parseErr,
 253                UErrorCode      *status);
 254
 255 /**
 256  * Thread safe cloning operation
 257  * @param bi iterator to be cloned
 258  * @param stackBuffer <em>Deprecated functionality as of ICU 52, use NULL.</em><br>
 259  *  user allocated space for the new clone. If NULL new memory will be allocated.
 260  *  If buffer is not large enough, new memory will be allocated.
 261  *  Clients can use the U_BRK_SAFECLONE_BUFFERSIZE.
 262  * @param pBufferSize <em>Deprecated functionality as of ICU 52, use NULL or 1.</em><br>
 263  *  pointer to size of allocated space.
 264  *  If *pBufferSize == 0, a sufficient size for use in cloning will
 265  *  be returned ('pre-flighting')
 266  *  If *pBufferSize is not enough for a stack-based safe clone,
 267  *  new memory will be allocated.
 268  * @param status to indicate whether the operation went on smoothly or there were errors
 269  *  An informational status value, U_SAFECLONE_ALLOCATED_ERROR, is used if any allocations were necessary.
 270  * @return pointer to the new clone
 271  * @stable ICU 2.0
 272  */
 273 U_STABLE UBreakIterator * U_EXPORT2
 274 ubrk_safeClone(
 275           const UBreakIterator *bi,
 276           void *stackBuffer,
 277           int32_t *pBufferSize,
 278           UErrorCode *status);
 279
 280 #ifndef U_HIDE_DEPRECATED_API
 281
 282 /**
 283   * A recommended size (in bytes) for the memory buffer to be passed to ubrk_saveClone().
 284   * @deprecated ICU 52. Do not rely on ubrk_safeClone() cloning into any provided buffer.
 285   */
 286 #define U_BRK_SAFECLONE_BUFFERSIZE 1
 287
 288 #endif /* U_HIDE_DEPRECATED_API */
 289
 290 /**
 291 * Close a UBreakIterator.
 292 * Once closed, a UBreakIterator may no longer be used.
 293 * @param bi The break iterator to close.
 294  * @stable ICU 2.0
 295 */
 296 U_STABLE void U_EXPORT2
 297 ubrk_close(UBreakIterator *bi);
 298
 299 #if U_SHOW_CPLUSPLUS_API
 300
 301 U_NAMESPACE_BEGIN
 302
 303 /**
 304  * \class LocalUBreakIteratorPointer
 305  * "Smart pointer" class, closes a UBreakIterator via ubrk_close().
 306  * For most methods see the LocalPointerBase base class.
 307  *
 308  * @see LocalPointerBase
 309  * @see LocalPointer
 310  * @stable ICU 4.4
 311  */
 312 U_DEFINE_LOCAL_OPEN_POINTER(LocalUBreakIteratorPointer, UBreakIterator, ubrk_close);
 313
 314 U_NAMESPACE_END
 315
 316 #endif
 317
 318 /**
 319  * Sets an existing iterator to point to a new piece of text
 320  * @param bi The iterator to use
 321  * @param text The text to be set
 322  * @param textLength The length of the text
 323  * @param status The error code
 324  * @stable ICU 2.0
 325  */
 326 U_STABLE void U_EXPORT2
 327 ubrk_setText(UBreakIterator* bi,
 328              const UChar*    text,
 329              int32_t         textLength,
 330              UErrorCode*     status);
 331
 332
 333 /**
 334  * Sets an existing iterator to point to a new piece of text.
 335  *
 336  * All index positions returned by break iterator functions are
 337  * native indices from the UText. For example, when breaking UTF-8
 338  * encoded text, the break positions returned by \ref ubrk_next, \ref ubrk_previous, etc.
 339  * will be UTF-8 string indices, not UTF-16 positions.
 340  *
 341  * @param bi The iterator to use
 342  * @param text The text to be set.
 343  *             This function makes a shallow clone of the supplied UText.  This means
 344  *             that the caller is free to immediately close or otherwise reuse the
 345  *             UText that was passed as a parameter, but that the underlying text itself
 346  *             must not be altered while being referenced by the break iterator.
 347  * @param status The error code
 348  * @stable ICU 3.4
 349  */
 350 U_STABLE void U_EXPORT2
 351 ubrk_setUText(UBreakIterator* bi,
 352              UText*          text,
 353              UErrorCode*     status);
 354
 355
 356
 357 /**
 358  * Determine the most recently-returned text boundary.
 359  *
 360  * @param bi The break iterator to use.
 361  * @return The character index most recently returned by \ref ubrk_next, \ref ubrk_previous,
 362  * \ref ubrk_first, or \ref ubrk_last.
 363  * @stable ICU 2.0
 364  */
 365 U_STABLE int32_t U_EXPORT2
 366 ubrk_current(const UBreakIterator *bi);
 367
 368 /**
 369  * Advance the iterator to the boundary following the current boundary.
 370  *
 371  * @param bi The break iterator to use.
 372  * @return The character index of the next text boundary, or UBRK_DONE
 373  * if all text boundaries have been returned.
 374  * @see ubrk_previous
 375  * @stable ICU 2.0
 376  */
 377 U_STABLE int32_t U_EXPORT2
 378 ubrk_next(UBreakIterator *bi);
 379
 380 /**
 381  * Set the iterator position to the boundary preceding the current boundary.
 382  *
 383  * @param bi The break iterator to use.
 384  * @return The character index of the preceding text boundary, or UBRK_DONE
 385  * if all text boundaries have been returned.
 386  * @see ubrk_next
 387  * @stable ICU 2.0
 388  */
 389 U_STABLE int32_t U_EXPORT2
 390 ubrk_previous(UBreakIterator *bi);
 391
 392 /**
 393  * Set the iterator position to zero, the start of the text being scanned.
 394  * @param bi The break iterator to use.
 395  * @return The new iterator position (zero).
 396  * @see ubrk_last
 397  * @stable ICU 2.0
 398  */
 399 U_STABLE int32_t U_EXPORT2
 400 ubrk_first(UBreakIterator *bi);
 401
 402 /**
 403  * Set the iterator position to the index immediately <EM>beyond</EM> the last character in the text being scanned.
 404  * This is not the same as the last character.
 405  * @param bi The break iterator to use.
 406  * @return The character offset immediately <EM>beyond</EM> the last character in the
 407  * text being scanned.
 408  * @see ubrk_first
 409  * @stable ICU 2.0
 410  */
 411 U_STABLE int32_t U_EXPORT2
 412 ubrk_last(UBreakIterator *bi);
 413
 414 /**
 415  * Set the iterator position to the first boundary preceding the specified offset.
 416  * The new position is always smaller than offset, or UBRK_DONE.
 417  * @param bi The break iterator to use.
 418  * @param offset The offset to begin scanning.
 419  * @return The text boundary preceding offset, or UBRK_DONE.
 420  * @see ubrk_following
 421  * @stable ICU 2.0
 422  */
 423 U_STABLE int32_t U_EXPORT2
 424 ubrk_preceding(UBreakIterator *bi,
 425            int32_t offset);
 426
 427 /**
 428  * Advance the iterator to the first boundary following the specified offset.
 429  * The value returned is always greater than offset, or UBRK_DONE.
 430  * @param bi The break iterator to use.
 431  * @param offset The offset to begin scanning.
 432  * @return The text boundary following offset, or UBRK_DONE.
 433  * @see ubrk_preceding
 434  * @stable ICU 2.0
 435  */
 436 U_STABLE int32_t U_EXPORT2
 437 ubrk_following(UBreakIterator *bi,
 438            int32_t offset);
 439
 440 /**
 441 * Get a locale for which text breaking information is available.
 442 * A UBreakIterator in a locale returned by this function will perform the correct
 443 * text breaking for the locale.
 444 * @param index The index of the desired locale.
 445 * @return A locale for which number text breaking information is available, or 0 if none.
 446 * @see ubrk_countAvailable
 447 * @stable ICU 2.0
 448 */
 449 U_STABLE const char* U_EXPORT2
 450 ubrk_getAvailable(int32_t index);
 451
 452 /**
 453 * Determine how many locales have text breaking information available.
 454 * This function is most useful as determining the loop ending condition for
 455 * calls to \ref ubrk_getAvailable.
 456 * @return The number of locales for which text breaking information is available.
 457 * @see ubrk_getAvailable
 458 * @stable ICU 2.0
 459 */
 460 U_STABLE int32_t U_EXPORT2
 461 ubrk_countAvailable(void);
 462
 463
 464 /**
 465 * Returns true if the specfied position is a boundary position.  As a side
 466 * effect, leaves the iterator pointing to the first boundary position at
 467 * or after "offset".
 468 * @param bi The break iterator to use.
 469 * @param offset the offset to check.
 470 * @return True if "offset" is a boundary position.
 471 * @stable ICU 2.0
 472 */
 473 U_STABLE  UBool U_EXPORT2
 474 ubrk_isBoundary(UBreakIterator *bi, int32_t offset);
 475
 476 /**
 477  * Return the status from the break rule that determined the most recently
 478  * returned break position.  The values appear in the rule source
 479  * within brackets, {123}, for example.  For rules that do not specify a
 480  * status, a default value of 0 is returned.
 481  * <p>
 482  * For word break iterators, the possible values are defined in enum UWordBreak.
 483  * @stable ICU 2.2
 484  */
 485 U_STABLE  int32_t U_EXPORT2
 486 ubrk_getRuleStatus(UBreakIterator *bi);
 487
 488 /**
 489  * Get the statuses from the break rules that determined the most recently
 490  * returned break position.  The values appear in the rule source
 491  * within brackets, {123}, for example.  The default status value for rules
 492  * that do not explicitly provide one is zero.
 493  * <p>
 494  * For word break iterators, the possible values are defined in enum UWordBreak.
 495  * @param bi        The break iterator to use
 496  * @param fillInVec an array to be filled in with the status values.
 497  * @param capacity  the length of the supplied vector.  A length of zero causes
 498  *                  the function to return the number of status values, in the
 499  *                  normal way, without attemtping to store any values.
 500  * @param status    receives error codes.
 501  * @return          The number of rule status values from rules that determined
 502  *                  the most recent boundary returned by the break iterator.
 503  * @stable ICU 3.0
 504  */
 505 U_STABLE  int32_t U_EXPORT2
 506 ubrk_getRuleStatusVec(UBreakIterator *bi, int32_t *fillInVec, int32_t capacity, UErrorCode *status);
 507
 508 /**
 509  * Return the locale of the break iterator. You can choose between the valid and
 510  * the actual locale.
 511  * @param bi break iterator
 512  * @param type locale type (valid or actual)
 513  * @param status error code
 514  * @return locale string
 515  * @stable ICU 2.8
 516  */
 517 U_STABLE const char* U_EXPORT2
 518 ubrk_getLocaleByType(const UBreakIterator *bi, ULocDataLocaleType type, UErrorCode* status);
 519
 520 /**
 521   *  Set the subject text string upon which the break iterator is operating
 522   *  without changing any other aspect of the state.
 523   *  The new and previous text strings must have the same content.
 524   *
 525   *  This function is intended for use in environments where ICU is operating on
 526   *  strings that may move around in memory.  It provides a mechanism for notifying
 527   *  ICU that the string has been relocated, and providing a new UText to access the
 528   *  string in its new position.
 529   *
 530   *  Note that the break iterator never copies the underlying text
 531   *  of a string being processed, but always operates directly on the original text
 532   *  provided by the user. Refreshing simply drops the references to the old text
 533   *  and replaces them with references to the new.
 534   *
 535   *  Caution:  this function is normally used only by very specialized
 536   *            system-level code.   One example use case is with garbage collection
 537   *            that moves the text in memory.
 538   *
 539   * @param bi         The break iterator.
 540   * @param text       The new (moved) text string.
 541   * @param status     Receives errors detected by this function.
 542   *
 543   * @stable ICU 49
 544   */
 545 U_STABLE void U_EXPORT2
 546 ubrk_refreshUText(UBreakIterator *bi,
 547                        UText          *text,
 548                        UErrorCode     *status);
 549
 550 #endif /* #if !UCONFIG_NO_BREAK_ITERATION */
 551
 552 #endif