icuSources/common/unicode/ubrk.h

   1 // © 2016 and later: Unicode, Inc. and others.
   2 // License & terms of use: http://www.unicode.org/copyright.html
   3 /*
   4 ******************************************************************************
   5 * Copyright (C) 1996-2015, International Business Machines Corporation and others.
   6 * All Rights Reserved.
   7 ******************************************************************************
   8 */
   9
  10 #ifndef UBRK_H
  11 #define UBRK_H
  12
  13 #include "unicode/utypes.h"
  14 #include "unicode/uloc.h"
  15 #include "unicode/utext.h"
  16 #include "unicode/localpointer.h"
  17
  18 /**
  19  * A text-break iterator.
  20  *  For usage in C programs.
  21  */
  22 #ifndef UBRK_TYPEDEF_UBREAK_ITERATOR
  23 #   define UBRK_TYPEDEF_UBREAK_ITERATOR
  24     /**
  25      *  Opaque type representing an ICU Break iterator object.
  26      *  @stable ICU 2.0
  27      */
  28     typedef struct UBreakIterator UBreakIterator;
  29 #endif
  30
  31 #if !UCONFIG_NO_BREAK_ITERATION
  32
  33 #include "unicode/parseerr.h"
  34
  35 /**
  36  * \file
  37  * \brief C API: BreakIterator
  38  *
  39  * <h2> BreakIterator C API </h2>
  40  *
  41  * The BreakIterator C API defines  methods for finding the location
  42  * of boundaries in text. Pointer to a UBreakIterator maintain a
  43  * current position and scan over text returning the index of characters
  44  * where boundaries occur.
  45  * <p>
  46  * Line boundary analysis determines where a text string can be broken
  47  * when line-wrapping. The mechanism correctly handles punctuation and
  48  * hyphenated words.
  49  * <p>
  50  * Note: The locale keyword "lb" can be used to modify line break
  51  * behavior according to the CSS level 3 line-break options, see
  52  * <http://dev.w3.org/csswg/css-text/#line-breaking>. For example:
  53  * "ja@lb=strict", "zh@lb=loose".
  54  * <p>
  55  * Sentence boundary analysis allows selection with correct
  56  * interpretation of periods within numbers and abbreviations, and
  57  * trailing punctuation marks such as quotation marks and parentheses.
  58  * <p>
  59  * Note: The locale keyword "ss" can be used to enable use of
  60  * segmentation suppression data (preventing breaks in English after
  61  * abbreviations such as "Mr." or "Est.", for example), as follows:
  62  * "en@ss=standard".
  63  * <p>
  64  * Word boundary analysis is used by search and replace functions, as
  65  * well as within text editing applications that allow the user to
  66  * select words with a double click. Word selection provides correct
  67  * interpretation of punctuation marks within and following
  68  * words. Characters that are not part of a word, such as symbols or
  69  * punctuation marks, have word-breaks on both sides.
  70  * <p>
  71  * Character boundary analysis identifies the boundaries of
  72  * "Extended Grapheme Clusters", which are groupings of codepoints
  73  * that should be treated as character-like units for many text operations.
  74  * Please see Unicode Standard Annex #29, Unicode Text Segmentation,
  75  * http://www.unicode.org/reports/tr29/ for additional information
  76  * on grapheme clusters and guidelines on their use.
  77  * <p>
  78  * Title boundary analysis locates all positions,
  79  * typically starts of words, that should be set to Title Case
  80  * when title casing the text.
  81  * <p>
  82  * The text boundary positions are found according to the rules
  83  * described in Unicode Standard Annex #29, Text Boundaries, and
  84  * Unicode Standard Annex #14, Line Breaking Properties.  These
  85  * are available at http://www.unicode.org/reports/tr14/ and
  86  * http://www.unicode.org/reports/tr29/.
  87  * <p>
  88  * In addition to the plain C API defined in this header file, an
  89  * object oriented C++ API with equivalent functionality is defined in the
  90  * file brkiter.h.
  91  * <p>
  92  * Code snippets illustrating the use of the Break Iterator APIs
  93  * are available in the ICU User Guide,
  94  * http://icu-project.org/userguide/boundaryAnalysis.html
  95  * and in the sample program icu/source/samples/break/break.cpp
  96  */
  97
  98 /** The possible types of text boundaries.  @stable ICU 2.0 */
  99 typedef enum UBreakIteratorType {
 100   /** Character breaks  @stable ICU 2.0 */
 101   UBRK_CHARACTER = 0,
 102   /** Word breaks @stable ICU 2.0 */
 103   UBRK_WORD = 1,
 104   /** Line breaks @stable ICU 2.0 */
 105   UBRK_LINE = 2,
 106   /** Sentence breaks @stable ICU 2.0 */
 107   UBRK_SENTENCE = 3,
 108
 109 #ifndef U_HIDE_DEPRECATED_API
 110   /**
 111    * Title Case breaks
 112    * The iterator created using this type locates title boundaries as described for
 113    * Unicode 3.2 only. For Unicode 4.0 and above title boundary iteration,
 114    * please use Word Boundary iterator.
 115    *
 116    * @deprecated ICU 2.8 Use the word break iterator for titlecasing for Unicode 4 and later.
 117    */
 118   UBRK_TITLE = 4,
 119     /**
 120      * One more than the highest normal UBreakIteratorType value.
 121      * @deprecated ICU 58 The numeric value may change over time, see ICU ticket #12420.
 122      */
 123     UBRK_COUNT = 5
 124 #endif  // U_HIDE_DEPRECATED_API
 125 } UBreakIteratorType;
 126
 127 /** Value indicating all text boundaries have been returned.
 128  *  @stable ICU 2.0
 129  */
 130 #define UBRK_DONE ((int32_t) -1)
 131
 132
 133 /**
 134  *  Enum constants for the word break tags returned by
 135  *  getRuleStatus().  A range of values is defined for each category of
 136  *  word, to allow for further subdivisions of a category in future releases.
 137  *  Applications should check for tag values falling within the range, rather
 138  *  than for single individual values.
 139  *
 140  * The numeric values of all of these constants are stable (will not change).
 141  *
 142  * @stable ICU 2.2
 143 */
 144 typedef enum UWordBreak {
 145     /** Tag value for "words" that do not fit into any of other categories.
 146      *  Includes spaces and most punctuation. */
 147     UBRK_WORD_NONE           = 0,
 148     /** Upper bound for tags for uncategorized words. */
 149     UBRK_WORD_NONE_LIMIT     = 100,
 150     /** Tag value for words that appear to be numbers, lower limit.    */
 151     UBRK_WORD_NUMBER         = 100,
 152     /** Tag value for words that appear to be numbers, upper limit.    */
 153     UBRK_WORD_NUMBER_LIMIT   = 200,
 154     /** Tag value for words that contain letters, excluding
 155      *  hiragana, katakana or ideographic characters, lower limit.    */
 156     UBRK_WORD_LETTER         = 200,
 157     /** Tag value for words containing letters, upper limit  */
 158     UBRK_WORD_LETTER_LIMIT   = 300,
 159     /** Tag value for words containing kana characters, lower limit */
 160     UBRK_WORD_KANA           = 300,
 161     /** Tag value for words containing kana characters, upper limit */
 162     UBRK_WORD_KANA_LIMIT     = 400,
 163     /** Tag value for words containing ideographic characters, lower limit */
 164     UBRK_WORD_IDEO           = 400,
 165     /** Tag value for words containing ideographic characters, upper limit */
 166     UBRK_WORD_IDEO_LIMIT     = 500
 167 } UWordBreak;
 168
 169 /**
 170  *  Enum constants for the line break tags returned by getRuleStatus().
 171  *  A range of values is defined for each category of
 172  *  word, to allow for further subdivisions of a category in future releases.
 173  *  Applications should check for tag values falling within the range, rather
 174  *  than for single individual values.
 175  *
 176  * The numeric values of all of these constants are stable (will not change).
 177  *
 178  * @stable ICU 2.8
 179 */
 180 typedef enum ULineBreakTag {
 181     /** Tag value for soft line breaks, positions at which a line break
 182       *  is acceptable but not required                */
 183     UBRK_LINE_SOFT            = 0,
 184     /** Upper bound for soft line breaks.              */
 185     UBRK_LINE_SOFT_LIMIT      = 100,
 186     /** Tag value for a hard, or mandatory line break  */
 187     UBRK_LINE_HARD            = 100,
 188     /** Upper bound for hard line breaks.              */
 189     UBRK_LINE_HARD_LIMIT      = 200
 190 } ULineBreakTag;
 191
 192
 193
 194 /**
 195  *  Enum constants for the sentence break tags returned by getRuleStatus().
 196  *  A range of values is defined for each category of
 197  *  sentence, to allow for further subdivisions of a category in future releases.
 198  *  Applications should check for tag values falling within the range, rather
 199  *  than for single individual values.
 200  *
 201  * The numeric values of all of these constants are stable (will not change).
 202  *
 203  * @stable ICU 2.8
 204 */
 205 typedef enum USentenceBreakTag {
 206     /** Tag value for for sentences  ending with a sentence terminator
 207       * ('.', '?', '!', etc.) character, possibly followed by a
 208       * hard separator (CR, LF, PS, etc.)
 209       */
 210     UBRK_SENTENCE_TERM       = 0,
 211     /** Upper bound for tags for sentences ended by sentence terminators.    */
 212     UBRK_SENTENCE_TERM_LIMIT = 100,
 213     /** Tag value for for sentences that do not contain an ending
 214       * sentence terminator ('.', '?', '!', etc.) character, but
 215       * are ended only by a hard separator (CR, LF, PS, etc.) or end of input.
 216       */
 217     UBRK_SENTENCE_SEP        = 100,
 218     /** Upper bound for tags for sentences ended by a separator.              */
 219     UBRK_SENTENCE_SEP_LIMIT  = 200
 220     /** Tag value for a hard, or mandatory line break  */
 221 } USentenceBreakTag;
 222
 223
 224 /**
 225  * Open a new UBreakIterator for locating text boundaries for a specified locale.
 226  * A UBreakIterator may be used for detecting character, line, word,
 227  * and sentence breaks in text.
 228  * @param type The type of UBreakIterator to open: one of UBRK_CHARACTER, UBRK_WORD,
 229  * UBRK_LINE, UBRK_SENTENCE
 230  * @param locale The locale specifying the text-breaking conventions. Note that
 231  * locale keys such as "lb" and "ss" may be used to modify text break behavior,
 232  * see general discussion of BreakIterator C API.
 233  * @param text The text to be iterated over.
 234  * @param textLength The number of characters in text, or -1 if null-terminated.
 235  * @param status A UErrorCode to receive any errors.
 236  * @return A UBreakIterator for the specified locale.
 237  * @see ubrk_openRules
 238  * @stable ICU 2.0
 239  */
 240 U_STABLE UBreakIterator* U_EXPORT2
 241 ubrk_open(UBreakIteratorType type,
 242       const char *locale,
 243       const UChar *text,
 244       int32_t textLength,
 245       UErrorCode *status);
 246
 247 /**
 248  * Open a new UBreakIterator for locating text boundaries using specified breaking rules.
 249  * The rule syntax is ... (TBD)
 250  * @param rules A set of rules specifying the text breaking conventions.
 251  * @param rulesLength The number of characters in rules, or -1 if null-terminated.
 252  * @param text The text to be iterated over.  May be null, in which case ubrk_setText() is
 253  *        used to specify the text to be iterated.
 254  * @param textLength The number of characters in text, or -1 if null-terminated.
 255  * @param parseErr   Receives position and context information for any syntax errors
 256  *                   detected while parsing the rules.
 257  * @param status A UErrorCode to receive any errors.
 258  * @return A UBreakIterator for the specified rules.
 259  * @see ubrk_open
 260  * @stable ICU 2.2
 261  */
 262 U_STABLE UBreakIterator* U_EXPORT2
 263 ubrk_openRules(const UChar     *rules,
 264                int32_t         rulesLength,
 265                const UChar     *text,
 266                int32_t          textLength,
 267                UParseError     *parseErr,
 268                UErrorCode      *status);
 269
 270 #ifndef U_HIDE_DRAFT_API
 271 /**
 272  * Open a new UBreakIterator for locating text boundaries using precompiled binary rules.
 273  * Opening a UBreakIterator this way is substantially faster than using ubrk_openRules.
 274  * Binary rules may be obtained using ubrk_getBinaryRules. The compiled rules are not
 275  * compatible across different major versions of ICU, nor across platforms of different
 276  * endianness or different base character set family (ASCII vs EBCDIC).
 277  * @param binaryRules A set of compiled binary rules specifying the text breaking
 278  *                    conventions. Ownership of the storage containing the compiled
 279  *                    rules remains with the caller of this function. The compiled
 280  *                    rules must not be modified or deleted during the life of the
 281  *                    break iterator.
 282  * @param rulesLength The length of binaryRules in bytes; must be >= 0.
 283  * @param text        The text to be iterated over.  May be null, in which case
 284  *                    ubrk_setText() is used to specify the text to be iterated.
 285  * @param textLength  The number of characters in text, or -1 if null-terminated.
 286  * @param status      Pointer to UErrorCode to receive any errors.
 287  * @return            UBreakIterator for the specified rules.
 288  * @see ubrk_getBinaryRules
 289  * @draft ICU 59
 290  */
 291 U_DRAFT UBreakIterator* U_EXPORT2
 292 ubrk_openBinaryRules(const uint8_t *binaryRules, int32_t rulesLength,
 293                      const UChar *  text, int32_t textLength,
 294                      UErrorCode *   status);
 295
 296 #endif  /* U_HIDE_DRAFT_API */
 297
 298 /**
 299  * Thread safe cloning operation
 300  * @param bi iterator to be cloned
 301  * @param stackBuffer <em>Deprecated functionality as of ICU 52, use NULL.</em><br>
 302  *  user allocated space for the new clone. If NULL new memory will be allocated.
 303  *  If buffer is not large enough, new memory will be allocated.
 304  *  Clients can use the U_BRK_SAFECLONE_BUFFERSIZE.
 305  * @param pBufferSize <em>Deprecated functionality as of ICU 52, use NULL or 1.</em><br>
 306  *  pointer to size of allocated space.
 307  *  If *pBufferSize == 0, a sufficient size for use in cloning will
 308  *  be returned ('pre-flighting')
 309  *  If *pBufferSize is not enough for a stack-based safe clone,
 310  *  new memory will be allocated.
 311  * @param status to indicate whether the operation went on smoothly or there were errors
 312  *  An informational status value, U_SAFECLONE_ALLOCATED_ERROR, is used if any allocations were necessary.
 313  * @return pointer to the new clone
 314  * @stable ICU 2.0
 315  */
 316 U_STABLE UBreakIterator * U_EXPORT2
 317 ubrk_safeClone(
 318           const UBreakIterator *bi,
 319           void *stackBuffer,
 320           int32_t *pBufferSize,
 321           UErrorCode *status);
 322
 323 #ifndef U_HIDE_DEPRECATED_API
 324
 325 /**
 326   * A recommended size (in bytes) for the memory buffer to be passed to ubrk_saveClone().
 327   * @deprecated ICU 52. Do not rely on ubrk_safeClone() cloning into any provided buffer.
 328   */
 329 #define U_BRK_SAFECLONE_BUFFERSIZE 1
 330
 331 #endif /* U_HIDE_DEPRECATED_API */
 332
 333 /**
 334 * Close a UBreakIterator.
 335 * Once closed, a UBreakIterator may no longer be used.
 336 * @param bi The break iterator to close.
 337  * @stable ICU 2.0
 338 */
 339 U_STABLE void U_EXPORT2
 340 ubrk_close(UBreakIterator *bi);
 341
 342 #if U_SHOW_CPLUSPLUS_API
 343
 344 U_NAMESPACE_BEGIN
 345
 346 /**
 347  * \class LocalUBreakIteratorPointer
 348  * "Smart pointer" class, closes a UBreakIterator via ubrk_close().
 349  * For most methods see the LocalPointerBase base class.
 350  *
 351  * @see LocalPointerBase
 352  * @see LocalPointer
 353  * @stable ICU 4.4
 354  */
 355 U_DEFINE_LOCAL_OPEN_POINTER(LocalUBreakIteratorPointer, UBreakIterator, ubrk_close);
 356
 357 U_NAMESPACE_END
 358
 359 #endif // U_SHOW_CPLUSPLUS_API
 360
 361 /**
 362  * Sets an existing iterator to point to a new piece of text.
 363  * The break iterator retains a pointer to the supplied text.
 364  * The caller must not modify or delete the text while the BreakIterator
 365  * retains the reference.
 366  *
 367  * @param bi The iterator to use
 368  * @param text The text to be set
 369  * @param textLength The length of the text
 370  * @param status The error code
 371  * @stable ICU 2.0
 372  */
 373 U_STABLE void U_EXPORT2
 374 ubrk_setText(UBreakIterator* bi,
 375              const UChar*    text,
 376              int32_t         textLength,
 377              UErrorCode*     status);
 378
 379
 380 /**
 381  * Sets an existing iterator to point to a new piece of text.
 382  *
 383  * All index positions returned by break iterator functions are
 384  * native indices from the UText. For example, when breaking UTF-8
 385  * encoded text, the break positions returned by \ref ubrk_next, \ref ubrk_previous, etc.
 386  * will be UTF-8 string indices, not UTF-16 positions.
 387  *
 388  * @param bi The iterator to use
 389  * @param text The text to be set.
 390  *             This function makes a shallow clone of the supplied UText.  This means
 391  *             that the caller is free to immediately close or otherwise reuse the
 392  *             UText that was passed as a parameter, but that the underlying text itself
 393  *             must not be altered while being referenced by the break iterator.
 394  * @param status The error code
 395  * @stable ICU 3.4
 396  */
 397 U_STABLE void U_EXPORT2
 398 ubrk_setUText(UBreakIterator* bi,
 399              UText*          text,
 400              UErrorCode*     status);
 401
 402
 403
 404 /**
 405  * Determine the most recently-returned text boundary.
 406  *
 407  * @param bi The break iterator to use.
 408  * @return The character index most recently returned by \ref ubrk_next, \ref ubrk_previous,
 409  * \ref ubrk_first, or \ref ubrk_last.
 410  * @stable ICU 2.0
 411  */
 412 U_STABLE int32_t U_EXPORT2
 413 ubrk_current(const UBreakIterator *bi);
 414
 415 /**
 416  * Advance the iterator to the boundary following the current boundary.
 417  *
 418  * @param bi The break iterator to use.
 419  * @return The character index of the next text boundary, or UBRK_DONE
 420  * if all text boundaries have been returned.
 421  * @see ubrk_previous
 422  * @stable ICU 2.0
 423  */
 424 U_STABLE int32_t U_EXPORT2
 425 ubrk_next(UBreakIterator *bi);
 426
 427 /**
 428  * Set the iterator position to the boundary preceding the current boundary.
 429  *
 430  * @param bi The break iterator to use.
 431  * @return The character index of the preceding text boundary, or UBRK_DONE
 432  * if all text boundaries have been returned.
 433  * @see ubrk_next
 434  * @stable ICU 2.0
 435  */
 436 U_STABLE int32_t U_EXPORT2
 437 ubrk_previous(UBreakIterator *bi);
 438
 439 /**
 440  * Set the iterator position to zero, the start of the text being scanned.
 441  * @param bi The break iterator to use.
 442  * @return The new iterator position (zero).
 443  * @see ubrk_last
 444  * @stable ICU 2.0
 445  */
 446 U_STABLE int32_t U_EXPORT2
 447 ubrk_first(UBreakIterator *bi);
 448
 449 /**
 450  * Set the iterator position to the index immediately <EM>beyond</EM> the last character in the text being scanned.
 451  * This is not the same as the last character.
 452  * @param bi The break iterator to use.
 453  * @return The character offset immediately <EM>beyond</EM> the last character in the
 454  * text being scanned.
 455  * @see ubrk_first
 456  * @stable ICU 2.0
 457  */
 458 U_STABLE int32_t U_EXPORT2
 459 ubrk_last(UBreakIterator *bi);
 460
 461 /**
 462  * Set the iterator position to the first boundary preceding the specified offset.
 463  * The new position is always smaller than offset, or UBRK_DONE.
 464  * @param bi The break iterator to use.
 465  * @param offset The offset to begin scanning.
 466  * @return The text boundary preceding offset, or UBRK_DONE.
 467  * @see ubrk_following
 468  * @stable ICU 2.0
 469  */
 470 U_STABLE int32_t U_EXPORT2
 471 ubrk_preceding(UBreakIterator *bi,
 472            int32_t offset);
 473
 474 /**
 475  * Advance the iterator to the first boundary following the specified offset.
 476  * The value returned is always greater than offset, or UBRK_DONE.
 477  * @param bi The break iterator to use.
 478  * @param offset The offset to begin scanning.
 479  * @return The text boundary following offset, or UBRK_DONE.
 480  * @see ubrk_preceding
 481  * @stable ICU 2.0
 482  */
 483 U_STABLE int32_t U_EXPORT2
 484 ubrk_following(UBreakIterator *bi,
 485            int32_t offset);
 486
 487 /**
 488 * Get a locale for which text breaking information is available.
 489 * A UBreakIterator in a locale returned by this function will perform the correct
 490 * text breaking for the locale.
 491 * @param index The index of the desired locale.
 492 * @return A locale for which number text breaking information is available, or 0 if none.
 493 * @see ubrk_countAvailable
 494 * @stable ICU 2.0
 495 */
 496 U_STABLE const char* U_EXPORT2
 497 ubrk_getAvailable(int32_t index);
 498
 499 /**
 500 * Determine how many locales have text breaking information available.
 501 * This function is most useful as determining the loop ending condition for
 502 * calls to \ref ubrk_getAvailable.
 503 * @return The number of locales for which text breaking information is available.
 504 * @see ubrk_getAvailable
 505 * @stable ICU 2.0
 506 */
 507 U_STABLE int32_t U_EXPORT2
 508 ubrk_countAvailable(void);
 509
 510
 511 /**
 512 * Returns true if the specfied position is a boundary position.  As a side
 513 * effect, leaves the iterator pointing to the first boundary position at
 514 * or after "offset".
 515 * @param bi The break iterator to use.
 516 * @param offset the offset to check.
 517 * @return True if "offset" is a boundary position.
 518 * @stable ICU 2.0
 519 */
 520 U_STABLE  UBool U_EXPORT2
 521 ubrk_isBoundary(UBreakIterator *bi, int32_t offset);
 522
 523 /**
 524  * Return the status from the break rule that determined the most recently
 525  * returned break position.  The values appear in the rule source
 526  * within brackets, {123}, for example.  For rules that do not specify a
 527  * status, a default value of 0 is returned.
 528  * <p>
 529  * For word break iterators, the possible values are defined in enum UWordBreak.
 530  * @stable ICU 2.2
 531  */
 532 U_STABLE  int32_t U_EXPORT2
 533 ubrk_getRuleStatus(UBreakIterator *bi);
 534
 535 /**
 536  * Get the statuses from the break rules that determined the most recently
 537  * returned break position.  The values appear in the rule source
 538  * within brackets, {123}, for example.  The default status value for rules
 539  * that do not explicitly provide one is zero.
 540  * <p>
 541  * For word break iterators, the possible values are defined in enum UWordBreak.
 542  * @param bi        The break iterator to use
 543  * @param fillInVec an array to be filled in with the status values.
 544  * @param capacity  the length of the supplied vector.  A length of zero causes
 545  *                  the function to return the number of status values, in the
 546  *                  normal way, without attemtping to store any values.
 547  * @param status    receives error codes.
 548  * @return          The number of rule status values from rules that determined
 549  *                  the most recent boundary returned by the break iterator.
 550  * @stable ICU 3.0
 551  */
 552 U_STABLE  int32_t U_EXPORT2
 553 ubrk_getRuleStatusVec(UBreakIterator *bi, int32_t *fillInVec, int32_t capacity, UErrorCode *status);
 554
 555 /**
 556  * Return the locale of the break iterator. You can choose between the valid and
 557  * the actual locale.
 558  * @param bi break iterator
 559  * @param type locale type (valid or actual)
 560  * @param status error code
 561  * @return locale string
 562  * @stable ICU 2.8
 563  */
 564 U_STABLE const char* U_EXPORT2
 565 ubrk_getLocaleByType(const UBreakIterator *bi, ULocDataLocaleType type, UErrorCode* status);
 566
 567 /**
 568   *  Set the subject text string upon which the break iterator is operating
 569   *  without changing any other aspect of the state.
 570   *  The new and previous text strings must have the same content.
 571   *
 572   *  This function is intended for use in environments where ICU is operating on
 573   *  strings that may move around in memory.  It provides a mechanism for notifying
 574   *  ICU that the string has been relocated, and providing a new UText to access the
 575   *  string in its new position.
 576   *
 577   *  Note that the break iterator never copies the underlying text
 578   *  of a string being processed, but always operates directly on the original text
 579   *  provided by the user. Refreshing simply drops the references to the old text
 580   *  and replaces them with references to the new.
 581   *
 582   *  Caution:  this function is normally used only by very specialized
 583   *            system-level code.   One example use case is with garbage collection
 584   *            that moves the text in memory.
 585   *
 586   * @param bi         The break iterator.
 587   * @param text       The new (moved) text string.
 588   * @param status     Receives errors detected by this function.
 589   *
 590   * @stable ICU 49
 591   */
 592 U_STABLE void U_EXPORT2
 593 ubrk_refreshUText(UBreakIterator *bi,
 594                        UText          *text,
 595                        UErrorCode     *status);
 596
 597
 598 #ifndef U_HIDE_DRAFT_API
 599 /**
 600  * Get a compiled binary version of the rules specifying the behavior of a UBreakIterator.
 601  * The binary rules may be used with ubrk_openBinaryRules to open a new UBreakIterator
 602  * more quickly than using ubrk_openRules. The compiled rules are not compatible across
 603  * different major versions of ICU, nor across platforms of different endianness or
 604  * different base character set family (ASCII vs EBCDIC). Supports preflighting (with
 605  * binaryRules=NULL and rulesCapacity=0) to get the rules length without copying them to
 606  * the binaryRules buffer. However, whether preflighting or not, if the actual length
 607  * is greater than INT32_MAX, then the function returns 0 and sets *status to
 608  * U_INDEX_OUTOFBOUNDS_ERROR.
 609
 610  * @param bi            The break iterator to use.
 611  * @param binaryRules   Buffer to receive the compiled binary rules; set to NULL for
 612  *                      preflighting.
 613  * @param rulesCapacity Capacity (in bytes) of the binaryRules buffer; set to 0 for
 614  *                      preflighting. Must be >= 0.
 615  * @param status        Pointer to UErrorCode to receive any errors, such as
 616  *                      U_BUFFER_OVERFLOW_ERROR, U_INDEX_OUTOFBOUNDS_ERROR, or
 617  *                      U_ILLEGAL_ARGUMENT_ERROR.
 618  * @return              The actual byte length of the binary rules, if <= INT32_MAX;
 619  *                      otherwise 0. If not preflighting and this is larger than
 620  *                      rulesCapacity, *status will be set to an error.
 621  * @see ubrk_openBinaryRules
 622  * @draft ICU 59
 623  */
 624 U_DRAFT int32_t U_EXPORT2
 625 ubrk_getBinaryRules(UBreakIterator *bi,
 626                     uint8_t *       binaryRules, int32_t rulesCapacity,
 627                     UErrorCode *    status);
 628
 629 #endif  /* U_HIDE_DRAFT_API */
 630
 631 #endif /* #if !UCONFIG_NO_BREAK_ITERATION */
 632
 633 #endif