icuSources/common/unicode/ubrk.h

   1 /*
   2 ******************************************************************************
   3 * Copyright (C) 1996-2006, 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
  15 /**
  16  * A text-break iterator.
  17  *  For usage in C programs.
  18  */
  19 #ifndef UBRK_TYPEDEF_UBREAK_ITERATOR
  20 #   define UBRK_TYPEDEF_UBREAK_ITERATOR
  21     /**
  22      *  Opaque type representing an ICU Break iterator object.
  23      *  @stable ICU 2.0
  24      */
  25     typedef void UBreakIterator;
  26 #endif
  27
  28 #if !UCONFIG_NO_BREAK_ITERATION
  29
  30 #include "unicode/parseerr.h"
  31
  32 /**
  33  * \file
  34  * \brief C API: BreakIterator
  35  *
  36  * <h2> BreakIterator C API </h2>
  37  *
  38  * The BreakIterator C API defines  methods for finding the location
  39  * of boundaries in text. Pointer to a UBreakIterator maintain a
  40  * current position and scan over text returning the index of characters
  41  * where boundaries occur.
  42  * <p>
  43  * Line boundary analysis determines where a text string can be broken
  44  * when line-wrapping. The mechanism correctly handles punctuation and
  45  * hyphenated words.
  46  * <p>
  47  * Sentence boundary analysis allows selection with correct
  48  * interpretation of periods within numbers and abbreviations, and
  49  * trailing punctuation marks such as quotation marks and parentheses.
  50  * <p>
  51  * Word boundary analysis is used by search and replace functions, as
  52  * well as within text editing applications that allow the user to
  53  * select words with a double click. Word selection provides correct
  54  * interpretation of punctuation marks within and following
  55  * words. Characters that are not part of a word, such as symbols or
  56  * punctuation marks, have word-breaks on both sides.
  57  * <p>
  58  * Character boundary analysis allows users to interact with
  59  * characters as they expect to, for example, when moving the cursor
  60  * through a text string. Character boundary analysis provides correct
  61  * navigation of through character strings, regardless of how the
  62  * character is stored.  For example, an accented character might be
  63  * stored as a base character and a diacritical mark. What users
  64  * consider to be a character can differ between languages.
  65  * <p>
  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.
  69  * <p>
  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/.
  75  * <p>
  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
  78  * file brkiter.h.
  79  * <p>
  80  * Code snippits illustrating the use of the Break Iterator APIs
  81  * are available in the ICU User Guide,
  82  * http://icu.sourceforge.net/userguide/boundaryAnalysis.html
  83  * and in the sample program icu/source/samples/break/break.cpp"
  84  */
  85
  86 /** The possible types of text boundaries.  @stable ICU 2.0 */
  87 typedef enum UBreakIteratorType {
  88   /** Character breaks  @stable ICU 2.0 */
  89   UBRK_CHARACTER = 0,
  90   /** Word breaks @stable ICU 2.0 */
  91   UBRK_WORD = 1,
  92   /** Line breaks @stable ICU 2.0 */
  93   UBRK_LINE = 2,
  94   /** Sentence breaks @stable ICU 2.0 */
  95   UBRK_SENTENCE = 3,
  96
  97 #ifndef U_HIDE_DEPRECATED_API
  98   /**
  99    * Title Case breaks
 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.
 103    *
 104    * @deprecated ICU 2.8 Use the word break iterator for titlecasing for Unicode 4 and later.
 105    */
 106   UBRK_TITLE = 4,
 107 #endif /* U_HIDE_DEPRECATED_API */
 108   UBRK_COUNT = 5
 109 } UBreakIteratorType;
 110
 111 /** Value indicating all text boundaries have been returned.
 112  *  @stable ICU 2.0
 113  */
 114 #define UBRK_DONE ((int32_t) -1)
 115
 116
 117 /**
 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.
 123  *  @stable ICU 2.2
 124 */
 125 typedef enum UWordBreak {
 126     /** Tag value for "words" that do not fit into any of other categories.
 127      *  Includes spaces and most punctuation. */
 128     UBRK_WORD_NONE           = 0,
 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
 148 } UWordBreak;
 149
 150 /**
 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.
 156  *  @stable ICU 2.8
 157 */
 158 typedef enum ULineBreakTag {
 159     /** Tag value for soft line breaks, positions at which a line break
 160       *  is acceptable but not required                */
 161     UBRK_LINE_SOFT            = 0,
 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
 168 } ULineBreakTag;
 169
 170
 171
 172 /**
 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.
 178  *  @stable ICU 2.8
 179 */
 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.)
 184       */
 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.
 191       */
 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  */
 196 } USentenceBreakTag;
 197
 198
 199 /**
 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
 211  * @stable ICU 2.0
 212  */
 213 U_STABLE UBreakIterator* U_EXPORT2
 214 ubrk_open(UBreakIteratorType type,
 215       const char *locale,
 216       const UChar *text,
 217       int32_t textLength,
 218       UErrorCode *status);
 219
 220 /**
 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.
 232  * @see ubrk_open
 233  * @stable ICU 2.2
 234  */
 235 U_STABLE UBreakIterator* U_EXPORT2
 236 ubrk_openRules(const UChar     *rules,
 237                int32_t         rulesLength,
 238                const UChar     *text,
 239                int32_t          textLength,
 240                UParseError     *parseErr,
 241                UErrorCode      *status);
 242
 243 /**
 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
 257  * @stable ICU 2.0
 258  */
 259 U_STABLE UBreakIterator * U_EXPORT2
 260 ubrk_safeClone(
 261           const UBreakIterator *bi,
 262           void *stackBuffer,
 263           int32_t *pBufferSize,
 264           UErrorCode *status);
 265
 266 /**
 267   * A recommended size (in bytes) for the memory buffer to be passed to ubrk_saveClone().
 268   * @stable ICU 2.0
 269   */
 270 #define U_BRK_SAFECLONE_BUFFERSIZE 512
 271
 272 /**
 273 * Close a UBreakIterator.
 274 * Once closed, a UBreakIterator may no longer be used.
 275 * @param bi The break iterator to close.
 276  * @stable ICU 2.0
 277 */
 278 U_STABLE void U_EXPORT2
 279 ubrk_close(UBreakIterator *bi);
 280
 281 /**
 282  * Sets an existing iterator to point to a new piece of text
 283  * @param bi The iterator to use
 284  * @param text The text to be set
 285  * @param textLength The length of the text
 286  * @param status The error code
 287  * @stable ICU 2.0
 288  */
 289 U_STABLE void U_EXPORT2
 290 ubrk_setText(UBreakIterator* bi,
 291              const UChar*    text,
 292              int32_t         textLength,
 293              UErrorCode*     status);
 294
 295
 296 /**
 297  * Sets an existing iterator to point to a new piece of text
 298  * @param bi The iterator to use
 299  * @param text The text to be set
 300  * @param status The error code
 301  * @draft ICU 3.4
 302  */
 303 U_DRAFT void U_EXPORT2
 304 ubrk_setUText(UBreakIterator* bi,
 305              UText*          text,
 306              UErrorCode*     status);
 307
 308
 309
 310 /**
 311  * Determine the most recently-returned text boundary.
 312  *
 313  * @param bi The break iterator to use.
 314  * @return The character index most recently returned by \ref ubrk_next, \ref ubrk_previous,
 315  * \ref ubrk_first, or \ref ubrk_last.
 316  * @stable ICU 2.0
 317  */
 318 U_STABLE int32_t U_EXPORT2
 319 ubrk_current(const UBreakIterator *bi);
 320
 321 /**
 322  * Determine the text boundary following the current text boundary.
 323  *
 324  * @param bi The break iterator to use.
 325  * @return The character index of the next text boundary, or UBRK_DONE
 326  * if all text boundaries have been returned.
 327  * @see ubrk_previous
 328  * @stable ICU 2.0
 329  */
 330 U_STABLE int32_t U_EXPORT2
 331 ubrk_next(UBreakIterator *bi);
 332
 333 /**
 334  * Determine the text boundary preceding the current text boundary.
 335  *
 336  * @param bi The break iterator to use.
 337  * @return The character index of the preceding text boundary, or UBRK_DONE
 338  * if all text boundaries have been returned.
 339  * @see ubrk_next
 340  * @stable ICU 2.0
 341  */
 342 U_STABLE int32_t U_EXPORT2
 343 ubrk_previous(UBreakIterator *bi);
 344
 345 /**
 346  * Determine the index of the first character in the text being scanned.
 347  * This is not always the same as index 0 of the text.
 348  * @param bi The break iterator to use.
 349  * @return The character index of the first character in the text being scanned.
 350  * @see ubrk_last
 351  * @stable ICU 2.0
 352  */
 353 U_STABLE int32_t U_EXPORT2
 354 ubrk_first(UBreakIterator *bi);
 355
 356 /**
 357  * Determine the index immediately <EM>beyond</EM> the last character in the text being
 358  * scanned.
 359  * This is not the same as the last character.
 360  * @param bi The break iterator to use.
 361  * @return The character offset immediately <EM>beyond</EM> the last character in the
 362  * text being scanned.
 363  * @see ubrk_first
 364  * @stable ICU 2.0
 365  */
 366 U_STABLE int32_t U_EXPORT2
 367 ubrk_last(UBreakIterator *bi);
 368
 369 /**
 370  * Determine the text boundary preceding the specified offset.
 371  * The value returned is always smaller than offset, or UBRK_DONE.
 372  * @param bi The break iterator to use.
 373  * @param offset The offset to begin scanning.
 374  * @return The text boundary preceding offset, or UBRK_DONE.
 375  * @see ubrk_following
 376  * @stable ICU 2.0
 377  */
 378 U_STABLE int32_t U_EXPORT2
 379 ubrk_preceding(UBreakIterator *bi,
 380            int32_t offset);
 381
 382 /**
 383  * Determine the text boundary following the specified offset.
 384  * The value returned is always greater than offset, or UBRK_DONE.
 385  * @param bi The break iterator to use.
 386  * @param offset The offset to begin scanning.
 387  * @return The text boundary following offset, or UBRK_DONE.
 388  * @see ubrk_preceding
 389  * @stable ICU 2.0
 390  */
 391 U_STABLE int32_t U_EXPORT2
 392 ubrk_following(UBreakIterator *bi,
 393            int32_t offset);
 394
 395 /**
 396 * Get a locale for which text breaking information is available.
 397 * A UBreakIterator in a locale returned by this function will perform the correct
 398 * text breaking for the locale.
 399 * @param index The index of the desired locale.
 400 * @return A locale for which number text breaking information is available, or 0 if none.
 401 * @see ubrk_countAvailable
 402 * @stable ICU 2.0
 403 */
 404 U_STABLE const char* U_EXPORT2
 405 ubrk_getAvailable(int32_t index);
 406
 407 /**
 408 * Determine how many locales have text breaking information available.
 409 * This function is most useful as determining the loop ending condition for
 410 * calls to \ref ubrk_getAvailable.
 411 * @return The number of locales for which text breaking information is available.
 412 * @see ubrk_getAvailable
 413 * @stable ICU 2.0
 414 */
 415 U_STABLE int32_t U_EXPORT2
 416 ubrk_countAvailable(void);
 417
 418
 419 /**
 420 * Returns true if the specfied position is a boundary position.  As a side
 421 * effect, leaves the iterator pointing to the first boundary position at
 422 * or after "offset".
 423 * @param bi The break iterator to use.
 424 * @param offset the offset to check.
 425 * @return True if "offset" is a boundary position.
 426 * @stable ICU 2.0
 427 */
 428 U_STABLE  UBool U_EXPORT2
 429 ubrk_isBoundary(UBreakIterator *bi, int32_t offset);
 430
 431 /**
 432  * Return the status from the break rule that determined the most recently
 433  * returned break position.  The values appear in the rule source
 434  * within brackets, {123}, for example.  For rules that do not specify a
 435  * status, a default value of 0 is returned.
 436  * <p>
 437  * For word break iterators, the possible values are defined in enum UWordBreak.
 438  * @stable ICU 2.2
 439  */
 440 U_STABLE  int32_t U_EXPORT2
 441 ubrk_getRuleStatus(UBreakIterator *bi);
 442
 443 /**
 444  * Get the statuses from the break rules that determined the most recently
 445  * returned break position.  The values appear in the rule source
 446  * within brackets, {123}, for example.  The default status value for rules
 447  * that do not explicitly provide one is zero.
 448  * <p>
 449  * For word break iterators, the possible values are defined in enum UWordBreak.
 450  * @param bi        The break iterator to use
 451  * @param fillInVec an array to be filled in with the status values.
 452  * @param capacity  the length of the supplied vector.  A length of zero causes
 453  *                  the function to return the number of status values, in the
 454  *                  normal way, without attemtping to store any values.
 455  * @param status    receives error codes.
 456  * @return          The number of rule status values from rules that determined
 457  *                  the most recent boundary returned by the break iterator.
 458  * @stable ICU 3.0
 459  */
 460 U_STABLE  int32_t U_EXPORT2
 461 ubrk_getRuleStatusVec(UBreakIterator *bi, int32_t *fillInVec, int32_t capacity, UErrorCode *status);
 462
 463 /**
 464  * Return the locale of the break iterator. You can choose between the valid and
 465  * the actual locale.
 466  * @param bi break iterator
 467  * @param type locale type (valid or actual)
 468  * @param status error code
 469  * @return locale string
 470  * @stable ICU 2.8
 471  */
 472 U_STABLE const char* U_EXPORT2
 473 ubrk_getLocaleByType(const UBreakIterator *bi, ULocDataLocaleType type, UErrorCode* status);
 474
 475
 476 #endif /* #if !UCONFIG_NO_BREAK_ITERATION */
 477
 478 #endif