icuSources/common/unicode/uset.h

   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) 2002-2014, International Business Machines
   7 *   Corporation and others.  All Rights Reserved.
   8 *
   9 *******************************************************************************
  10 *   file name:  uset.h
  11 *   encoding:   UTF-8
  12 *   tab size:   8 (not used)
  13 *   indentation:4
  14 *
  15 *   created on: 2002mar07
  16 *   created by: Markus W. Scherer
  17 *
  18 *   C version of UnicodeSet.
  19 */
  20
  21
  22 /**
  23  * \file
  24  * \brief C API: Unicode Set
  25  *
  26  * <p>This is a C wrapper around the C++ UnicodeSet class.</p>
  27  */
  28
  29 #ifndef __USET_H__
  30 #define __USET_H__
  31
  32 #include "unicode/utypes.h"
  33 #include "unicode/uchar.h"
  34 #include "unicode/localpointer.h"
  35
  36 #ifndef USET_DEFINED
  37
  38 #ifndef U_IN_DOXYGEN
  39 #define USET_DEFINED
  40 #endif
  41 /**
  42  * USet is the C API type corresponding to C++ class UnicodeSet.
  43  * Use the uset_* API to manipulate.  Create with
  44  * uset_open*, and destroy with uset_close.
  45  * @stable ICU 2.4
  46  */
  47 typedef struct USet USet;
  48 #endif
  49
  50 /**
  51  * Bitmask values to be passed to uset_openPatternOptions() or
  52  * uset_applyPattern() taking an option parameter.
  53  * @stable ICU 2.4
  54  */
  55 enum {
  56     /**
  57      * Ignore white space within patterns unless quoted or escaped.
  58      * @stable ICU 2.4
  59      */
  60     USET_IGNORE_SPACE = 1,
  61
  62     /**
  63      * Enable case insensitive matching.  E.g., "[ab]" with this flag
  64      * will match 'a', 'A', 'b', and 'B'.  "[^ab]" with this flag will
  65      * match all except 'a', 'A', 'b', and 'B'. This performs a full
  66      * closure over case mappings, e.g. U+017F for s.
  67      *
  68      * The resulting set is a superset of the input for the code points but
  69      * not for the strings.
  70      * It performs a case mapping closure of the code points and adds
  71      * full case folding strings for the code points, and reduces strings of
  72      * the original set to their full case folding equivalents.
  73      *
  74      * This is designed for case-insensitive matches, for example
  75      * in regular expressions. The full code point case closure allows checking of
  76      * an input character directly against the closure set.
  77      * Strings are matched by comparing the case-folded form from the closure
  78      * set with an incremental case folding of the string in question.
  79      *
  80      * The closure set will also contain single code points if the original
  81      * set contained case-equivalent strings (like U+00DF for "ss" or "Ss" etc.).
  82      * This is not necessary (that is, redundant) for the above matching method
  83      * but results in the same closure sets regardless of whether the original
  84      * set contained the code point or a string.
  85      *
  86      * @stable ICU 2.4
  87      */
  88     USET_CASE_INSENSITIVE = 2,
  89
  90     /**
  91      * Enable case insensitive matching.  E.g., "[ab]" with this flag
  92      * will match 'a', 'A', 'b', and 'B'.  "[^ab]" with this flag will
  93      * match all except 'a', 'A', 'b', and 'B'. This adds the lower-,
  94      * title-, and uppercase mappings as well as the case folding
  95      * of each existing element in the set.
  96      * @stable ICU 3.2
  97      */
  98     USET_ADD_CASE_MAPPINGS = 4
  99 };
 100
 101 /**
 102  * Argument values for whether span() and similar functions continue while
 103  * the current character is contained vs. not contained in the set.
 104  *
 105  * The functionality is straightforward for sets with only single code points,
 106  * without strings (which is the common case):
 107  * - USET_SPAN_CONTAINED and USET_SPAN_SIMPLE work the same.
 108  * - USET_SPAN_CONTAINED and USET_SPAN_SIMPLE are inverses of USET_SPAN_NOT_CONTAINED.
 109  * - span() and spanBack() partition any string the same way when
 110  *   alternating between span(USET_SPAN_NOT_CONTAINED) and
 111  *   span(either "contained" condition).
 112  * - Using a complemented (inverted) set and the opposite span conditions
 113  *   yields the same results.
 114  *
 115  * When a set contains multi-code point strings, then these statements may not
 116  * be true, depending on the strings in the set (for example, whether they
 117  * overlap with each other) and the string that is processed.
 118  * For a set with strings:
 119  * - The complement of the set contains the opposite set of code points,
 120  *   but the same set of strings.
 121  *   Therefore, complementing both the set and the span conditions
 122  *   may yield different results.
 123  * - When starting spans at different positions in a string
 124  *   (span(s, ...) vs. span(s+1, ...)) the ends of the spans may be different
 125  *   because a set string may start before the later position.
 126  * - span(USET_SPAN_SIMPLE) may be shorter than
 127  *   span(USET_SPAN_CONTAINED) because it will not recursively try
 128  *   all possible paths.
 129  *   For example, with a set which contains the three strings "xy", "xya" and "ax",
 130  *   span("xyax", USET_SPAN_CONTAINED) will return 4 but
 131  *   span("xyax", USET_SPAN_SIMPLE) will return 3.
 132  *   span(USET_SPAN_SIMPLE) will never be longer than
 133  *   span(USET_SPAN_CONTAINED).
 134  * - With either "contained" condition, span() and spanBack() may partition
 135  *   a string in different ways.
 136  *   For example, with a set which contains the two strings "ab" and "ba",
 137  *   and when processing the string "aba",
 138  *   span() will yield contained/not-contained boundaries of { 0, 2, 3 }
 139  *   while spanBack() will yield boundaries of { 0, 1, 3 }.
 140  *
 141  * Note: If it is important to get the same boundaries whether iterating forward
 142  * or backward through a string, then either only span() should be used and
 143  * the boundaries cached for backward operation, or an ICU BreakIterator
 144  * could be used.
 145  *
 146  * Note: Unpaired surrogates are treated like surrogate code points.
 147  * Similarly, set strings match only on code point boundaries,
 148  * never in the middle of a surrogate pair.
 149  * Illegal UTF-8 sequences are treated like U+FFFD.
 150  * When processing UTF-8 strings, malformed set strings
 151  * (strings with unpaired surrogates which cannot be converted to UTF-8)
 152  * are ignored.
 153  *
 154  * @stable ICU 3.8
 155  */
 156 typedef enum USetSpanCondition {
 157     /**
 158      * Continues a span() while there is no set element at the current position.
 159      * Increments by one code point at a time.
 160      * Stops before the first set element (character or string).
 161      * (For code points only, this is like while contains(current)==FALSE).
 162      *
 163      * When span() returns, the substring between where it started and the position
 164      * it returned consists only of characters that are not in the set,
 165      * and none of its strings overlap with the span.
 166      *
 167      * @stable ICU 3.8
 168      */
 169     USET_SPAN_NOT_CONTAINED = 0,
 170     /**
 171      * Spans the longest substring that is a concatenation of set elements (characters or strings).
 172      * (For characters only, this is like while contains(current)==TRUE).
 173      *
 174      * When span() returns, the substring between where it started and the position
 175      * it returned consists only of set elements (characters or strings) that are in the set.
 176      *
 177      * If a set contains strings, then the span will be the longest substring for which there
 178      * exists at least one non-overlapping concatenation of set elements (characters or strings).
 179      * This is equivalent to a POSIX regular expression for <code>(OR of each set element)*</code>.
 180      * (Java/ICU/Perl regex stops at the first match of an OR.)
 181      *
 182      * @stable ICU 3.8
 183      */
 184     USET_SPAN_CONTAINED = 1,
 185     /**
 186      * Continues a span() while there is a set element at the current position.
 187      * Increments by the longest matching element at each position.
 188      * (For characters only, this is like while contains(current)==TRUE).
 189      *
 190      * When span() returns, the substring between where it started and the position
 191      * it returned consists only of set elements (characters or strings) that are in the set.
 192      *
 193      * If a set only contains single characters, then this is the same
 194      * as USET_SPAN_CONTAINED.
 195      *
 196      * If a set contains strings, then the span will be the longest substring
 197      * with a match at each position with the longest single set element (character or string).
 198      *
 199      * Use this span condition together with other longest-match algorithms,
 200      * such as ICU converters (ucnv_getUnicodeSet()).
 201      *
 202      * @stable ICU 3.8
 203      */
 204     USET_SPAN_SIMPLE = 2,
 205 #ifndef U_HIDE_DEPRECATED_API
 206     /**
 207      * One more than the last span condition.
 208      * @deprecated ICU 58 The numeric value may change over time, see ICU ticket #12420.
 209      */
 210     USET_SPAN_CONDITION_COUNT
 211 #endif  // U_HIDE_DEPRECATED_API
 212 } USetSpanCondition;
 213
 214 enum {
 215     /**
 216      * Capacity of USerializedSet::staticArray.
 217      * Enough for any single-code point set.
 218      * Also provides padding for nice sizeof(USerializedSet).
 219      * @stable ICU 2.4
 220      */
 221     USET_SERIALIZED_STATIC_ARRAY_CAPACITY=8
 222 };
 223
 224 /**
 225  * A serialized form of a Unicode set.  Limited manipulations are
 226  * possible directly on a serialized set.  See below.
 227  * @stable ICU 2.4
 228  */
 229 typedef struct USerializedSet {
 230     /**
 231      * The serialized Unicode Set.
 232      * @stable ICU 2.4
 233      */
 234     const uint16_t *array;
 235     /**
 236      * The length of the array that contains BMP characters.
 237      * @stable ICU 2.4
 238      */
 239     int32_t bmpLength;
 240     /**
 241      * The total length of the array.
 242      * @stable ICU 2.4
 243      */
 244     int32_t length;
 245     /**
 246      * A small buffer for the array to reduce memory allocations.
 247      * @stable ICU 2.4
 248      */
 249     uint16_t staticArray[USET_SERIALIZED_STATIC_ARRAY_CAPACITY];
 250 } USerializedSet;
 251
 252 /*********************************************************************
 253  * USet API
 254  *********************************************************************/
 255
 256 /**
 257  * Create an empty USet object.
 258  * Equivalent to uset_open(1, 0).
 259  * @return a newly created USet.  The caller must call uset_close() on
 260  * it when done.
 261  * @stable ICU 4.2
 262  */
 263 U_STABLE USet* U_EXPORT2
 264 uset_openEmpty(void);
 265
 266 /**
 267  * Creates a USet object that contains the range of characters
 268  * start..end, inclusive.  If <code>start > end</code>
 269  * then an empty set is created (same as using uset_openEmpty()).
 270  * @param start first character of the range, inclusive
 271  * @param end last character of the range, inclusive
 272  * @return a newly created USet.  The caller must call uset_close() on
 273  * it when done.
 274  * @stable ICU 2.4
 275  */
 276 U_STABLE USet* U_EXPORT2
 277 uset_open(UChar32 start, UChar32 end);
 278
 279 /**
 280  * Creates a set from the given pattern.  See the UnicodeSet class
 281  * description for the syntax of the pattern language.
 282  * @param pattern a string specifying what characters are in the set
 283  * @param patternLength the length of the pattern, or -1 if null
 284  * terminated
 285  * @param ec the error code
 286  * @stable ICU 2.4
 287  */
 288 U_STABLE USet* U_EXPORT2
 289 uset_openPattern(const UChar* pattern, int32_t patternLength,
 290                  UErrorCode* ec);
 291
 292 /**
 293  * Creates a set from the given pattern.  See the UnicodeSet class
 294  * description for the syntax of the pattern language.
 295  * @param pattern a string specifying what characters are in the set
 296  * @param patternLength the length of the pattern, or -1 if null
 297  * terminated
 298  * @param options bitmask for options to apply to the pattern.
 299  * Valid options are USET_IGNORE_SPACE and USET_CASE_INSENSITIVE.
 300  * @param ec the error code
 301  * @stable ICU 2.4
 302  */
 303 U_STABLE USet* U_EXPORT2
 304 uset_openPatternOptions(const UChar* pattern, int32_t patternLength,
 305                  uint32_t options,
 306                  UErrorCode* ec);
 307
 308 /**
 309  * Disposes of the storage used by a USet object.  This function should
 310  * be called exactly once for objects returned by uset_open().
 311  * @param set the object to dispose of
 312  * @stable ICU 2.4
 313  */
 314 U_STABLE void U_EXPORT2
 315 uset_close(USet* set);
 316
 317 #if U_SHOW_CPLUSPLUS_API
 318
 319 U_NAMESPACE_BEGIN
 320
 321 /**
 322  * \class LocalUSetPointer
 323  * "Smart pointer" class, closes a USet via uset_close().
 324  * For most methods see the LocalPointerBase base class.
 325  *
 326  * @see LocalPointerBase
 327  * @see LocalPointer
 328  * @stable ICU 4.4
 329  */
 330 U_DEFINE_LOCAL_OPEN_POINTER(LocalUSetPointer, USet, uset_close);
 331
 332 U_NAMESPACE_END
 333
 334 #endif // U_SHOW_CPLUSPLUS_API
 335
 336 /**
 337  * Returns a copy of this object.
 338  * If this set is frozen, then the clone will be frozen as well.
 339  * Use uset_cloneAsThawed() for a mutable clone of a frozen set.
 340  * @param set the original set
 341  * @return the newly allocated copy of the set
 342  * @see uset_cloneAsThawed
 343  * @stable ICU 3.8
 344  */
 345 U_STABLE USet * U_EXPORT2
 346 uset_clone(const USet *set);
 347
 348 /**
 349  * Determines whether the set has been frozen (made immutable) or not.
 350  * See the ICU4J Freezable interface for details.
 351  * @param set the set
 352  * @return TRUE/FALSE for whether the set has been frozen
 353  * @see uset_freeze
 354  * @see uset_cloneAsThawed
 355  * @stable ICU 3.8
 356  */
 357 U_STABLE UBool U_EXPORT2
 358 uset_isFrozen(const USet *set);
 359
 360 /**
 361  * Freeze the set (make it immutable).
 362  * Once frozen, it cannot be unfrozen and is therefore thread-safe
 363  * until it is deleted.
 364  * See the ICU4J Freezable interface for details.
 365  * Freezing the set may also make some operations faster, for example
 366  * uset_contains() and uset_span().
 367  * A frozen set will not be modified. (It remains frozen.)
 368  * @param set the set
 369  * @return the same set, now frozen
 370  * @see uset_isFrozen
 371  * @see uset_cloneAsThawed
 372  * @stable ICU 3.8
 373  */
 374 U_STABLE void U_EXPORT2
 375 uset_freeze(USet *set);
 376
 377 /**
 378  * Clone the set and make the clone mutable.
 379  * See the ICU4J Freezable interface for details.
 380  * @param set the set
 381  * @return the mutable clone
 382  * @see uset_freeze
 383  * @see uset_isFrozen
 384  * @see uset_clone
 385  * @stable ICU 3.8
 386  */
 387 U_STABLE USet * U_EXPORT2
 388 uset_cloneAsThawed(const USet *set);
 389
 390 /**
 391  * Causes the USet object to represent the range <code>start - end</code>.
 392  * If <code>start > end</code> then this USet is set to an empty range.
 393  * A frozen set will not be modified.
 394  * @param set the object to set to the given range
 395  * @param start first character in the set, inclusive
 396  * @param end last character in the set, inclusive
 397  * @stable ICU 3.2
 398  */
 399 U_STABLE void U_EXPORT2
 400 uset_set(USet* set,
 401          UChar32 start, UChar32 end);
 402
 403 /**
 404  * Modifies the set to represent the set specified by the given
 405  * pattern. See the UnicodeSet class description for the syntax of
 406  * the pattern language. See also the User Guide chapter about UnicodeSet.
 407  * <em>Empties the set passed before applying the pattern.</em>
 408  * A frozen set will not be modified.
 409  * @param set               The set to which the pattern is to be applied.
 410  * @param pattern           A pointer to UChar string specifying what characters are in the set.
 411  *                          The character at pattern[0] must be a '['.
 412  * @param patternLength     The length of the UChar string. -1 if NUL terminated.
 413  * @param options           A bitmask for options to apply to the pattern.
 414  *                          Valid options are USET_IGNORE_SPACE and USET_CASE_INSENSITIVE.
 415  * @param status            Returns an error if the pattern cannot be parsed.
 416  * @return                  Upon successful parse, the value is either
 417  *                          the index of the character after the closing ']'
 418  *                          of the parsed pattern.
 419  *                          If the status code indicates failure, then the return value
 420  *                          is the index of the error in the source.
 421  *
 422  * @stable ICU 2.8
 423  */
 424 U_STABLE int32_t U_EXPORT2
 425 uset_applyPattern(USet *set,
 426                   const UChar *pattern, int32_t patternLength,
 427                   uint32_t options,
 428                   UErrorCode *status);
 429
 430 /**
 431  * Modifies the set to contain those code points which have the given value
 432  * for the given binary or enumerated property, as returned by
 433  * u_getIntPropertyValue.  Prior contents of this set are lost.
 434  * A frozen set will not be modified.
 435  *
 436  * @param set the object to contain the code points defined by the property
 437  *
 438  * @param prop a property in the range UCHAR_BIN_START..UCHAR_BIN_LIMIT-1
 439  * or UCHAR_INT_START..UCHAR_INT_LIMIT-1
 440  * or UCHAR_MASK_START..UCHAR_MASK_LIMIT-1.
 441  *
 442  * @param value a value in the range u_getIntPropertyMinValue(prop)..
 443  * u_getIntPropertyMaxValue(prop), with one exception.  If prop is
 444  * UCHAR_GENERAL_CATEGORY_MASK, then value should not be a UCharCategory, but
 445  * rather a mask value produced by U_GET_GC_MASK().  This allows grouped
 446  * categories such as [:L:] to be represented.
 447  *
 448  * @param ec error code input/output parameter
 449  *
 450  * @stable ICU 3.2
 451  */
 452 U_STABLE void U_EXPORT2
 453 uset_applyIntPropertyValue(USet* set,
 454                            UProperty prop, int32_t value, UErrorCode* ec);
 455
 456 /**
 457  * Modifies the set to contain those code points which have the
 458  * given value for the given property.  Prior contents of this
 459  * set are lost.
 460  * A frozen set will not be modified.
 461  *
 462  * @param set the object to contain the code points defined by the given
 463  * property and value alias
 464  *
 465  * @param prop a string specifying a property alias, either short or long.
 466  * The name is matched loosely.  See PropertyAliases.txt for names and a
 467  * description of loose matching.  If the value string is empty, then this
 468  * string is interpreted as either a General_Category value alias, a Script
 469  * value alias, a binary property alias, or a special ID.  Special IDs are
 470  * matched loosely and correspond to the following sets:
 471  *
 472  * "ANY" = [\\u0000-\\U0010FFFF],
 473  * "ASCII" = [\\u0000-\\u007F],
 474  * "Assigned" = [:^Cn:].
 475  *
 476  * @param propLength the length of the prop, or -1 if NULL
 477  *
 478  * @param value a string specifying a value alias, either short or long.
 479  * The name is matched loosely.  See PropertyValueAliases.txt for names
 480  * and a description of loose matching.  In addition to aliases listed,
 481  * numeric values and canonical combining classes may be expressed
 482  * numerically, e.g., ("nv", "0.5") or ("ccc", "220").  The value string
 483  * may also be empty.
 484  *
 485  * @param valueLength the length of the value, or -1 if NULL
 486  *
 487  * @param ec error code input/output parameter
 488  *
 489  * @stable ICU 3.2
 490  */
 491 U_STABLE void U_EXPORT2
 492 uset_applyPropertyAlias(USet* set,
 493                         const UChar *prop, int32_t propLength,
 494                         const UChar *value, int32_t valueLength,
 495                         UErrorCode* ec);
 496
 497 /**
 498  * Return true if the given position, in the given pattern, appears
 499  * to be the start of a UnicodeSet pattern.
 500  *
 501  * @param pattern a string specifying the pattern
 502  * @param patternLength the length of the pattern, or -1 if NULL
 503  * @param pos the given position
 504  * @stable ICU 3.2
 505  */
 506 U_STABLE UBool U_EXPORT2
 507 uset_resemblesPattern(const UChar *pattern, int32_t patternLength,
 508                       int32_t pos);
 509
 510 /**
 511  * Returns a string representation of this set.  If the result of
 512  * calling this function is passed to a uset_openPattern(), it
 513  * will produce another set that is equal to this one.
 514  * @param set the set
 515  * @param result the string to receive the rules, may be NULL
 516  * @param resultCapacity the capacity of result, may be 0 if result is NULL
 517  * @param escapeUnprintable if TRUE then convert unprintable
 518  * character to their hex escape representations, \\uxxxx or
 519  * \\Uxxxxxxxx.  Unprintable characters are those other than
 520  * U+000A, U+0020..U+007E.
 521  * @param ec error code.
 522  * @return length of string, possibly larger than resultCapacity
 523  * @stable ICU 2.4
 524  */
 525 U_STABLE int32_t U_EXPORT2
 526 uset_toPattern(const USet* set,
 527                UChar* result, int32_t resultCapacity,
 528                UBool escapeUnprintable,
 529                UErrorCode* ec);
 530
 531 /**
 532  * Adds the given character to the given USet.  After this call,
 533  * uset_contains(set, c) will return TRUE.
 534  * A frozen set will not be modified.
 535  * @param set the object to which to add the character
 536  * @param c the character to add
 537  * @stable ICU 2.4
 538  */
 539 U_STABLE void U_EXPORT2
 540 uset_add(USet* set, UChar32 c);
 541
 542 /**
 543  * Adds all of the elements in the specified set to this set if
 544  * they're not already present.  This operation effectively
 545  * modifies this set so that its value is the <i>union</i> of the two
 546  * sets.  The behavior of this operation is unspecified if the specified
 547  * collection is modified while the operation is in progress.
 548  * A frozen set will not be modified.
 549  *
 550  * @param set the object to which to add the set
 551  * @param additionalSet the source set whose elements are to be added to this set.
 552  * @stable ICU 2.6
 553  */
 554 U_STABLE void U_EXPORT2
 555 uset_addAll(USet* set, const USet *additionalSet);
 556
 557 /**
 558  * Adds the given range of characters to the given USet.  After this call,
 559  * uset_contains(set, start, end) will return TRUE.
 560  * A frozen set will not be modified.
 561  * @param set the object to which to add the character
 562  * @param start the first character of the range to add, inclusive
 563  * @param end the last character of the range to add, inclusive
 564  * @stable ICU 2.2
 565  */
 566 U_STABLE void U_EXPORT2
 567 uset_addRange(USet* set, UChar32 start, UChar32 end);
 568
 569 /**
 570  * Adds the given string to the given USet.  After this call,
 571  * uset_containsString(set, str, strLen) will return TRUE.
 572  * A frozen set will not be modified.
 573  * @param set the object to which to add the character
 574  * @param str the string to add
 575  * @param strLen the length of the string or -1 if null terminated.
 576  * @stable ICU 2.4
 577  */
 578 U_STABLE void U_EXPORT2
 579 uset_addString(USet* set, const UChar* str, int32_t strLen);
 580
 581 /**
 582  * Adds each of the characters in this string to the set. Thus "ch" => {"c", "h"}
 583  * If this set already any particular character, it has no effect on that character.
 584  * A frozen set will not be modified.
 585  * @param set the object to which to add the character
 586  * @param str the source string
 587  * @param strLen the length of the string or -1 if null terminated.
 588  * @stable ICU 3.4
 589  */
 590 U_STABLE void U_EXPORT2
 591 uset_addAllCodePoints(USet* set, const UChar *str, int32_t strLen);
 592
 593 /**
 594  * Removes the given character from the given USet.  After this call,
 595  * uset_contains(set, c) will return FALSE.
 596  * A frozen set will not be modified.
 597  * @param set the object from which to remove the character
 598  * @param c the character to remove
 599  * @stable ICU 2.4
 600  */
 601 U_STABLE void U_EXPORT2
 602 uset_remove(USet* set, UChar32 c);
 603
 604 /**
 605  * Removes the given range of characters from the given USet.  After this call,
 606  * uset_contains(set, start, end) will return FALSE.
 607  * A frozen set will not be modified.
 608  * @param set the object to which to add the character
 609  * @param start the first character of the range to remove, inclusive
 610  * @param end the last character of the range to remove, inclusive
 611  * @stable ICU 2.2
 612  */
 613 U_STABLE void U_EXPORT2
 614 uset_removeRange(USet* set, UChar32 start, UChar32 end);
 615
 616 /**
 617  * Removes the given string to the given USet.  After this call,
 618  * uset_containsString(set, str, strLen) will return FALSE.
 619  * A frozen set will not be modified.
 620  * @param set the object to which to add the character
 621  * @param str the string to remove
 622  * @param strLen the length of the string or -1 if null terminated.
 623  * @stable ICU 2.4
 624  */
 625 U_STABLE void U_EXPORT2
 626 uset_removeString(USet* set, const UChar* str, int32_t strLen);
 627
 628 /**
 629  * Removes from this set all of its elements that are contained in the
 630  * specified set.  This operation effectively modifies this
 631  * set so that its value is the <i>asymmetric set difference</i> of
 632  * the two sets.
 633  * A frozen set will not be modified.
 634  * @param set the object from which the elements are to be removed
 635  * @param removeSet the object that defines which elements will be
 636  * removed from this set
 637  * @stable ICU 3.2
 638  */
 639 U_STABLE void U_EXPORT2
 640 uset_removeAll(USet* set, const USet* removeSet);
 641
 642 /**
 643  * Retain only the elements in this set that are contained in the
 644  * specified range.  If <code>start > end</code> then an empty range is
 645  * retained, leaving the set empty.  This is equivalent to
 646  * a boolean logic AND, or a set INTERSECTION.
 647  * A frozen set will not be modified.
 648  *
 649  * @param set the object for which to retain only the specified range
 650  * @param start first character, inclusive, of range to be retained
 651  * to this set.
 652  * @param end last character, inclusive, of range to be retained
 653  * to this set.
 654  * @stable ICU 3.2
 655  */
 656 U_STABLE void U_EXPORT2
 657 uset_retain(USet* set, UChar32 start, UChar32 end);
 658
 659 /**
 660  * Retains only the elements in this set that are contained in the
 661  * specified set.  In other words, removes from this set all of
 662  * its elements that are not contained in the specified set.  This
 663  * operation effectively modifies this set so that its value is
 664  * the <i>intersection</i> of the two sets.
 665  * A frozen set will not be modified.
 666  *
 667  * @param set the object on which to perform the retain
 668  * @param retain set that defines which elements this set will retain
 669  * @stable ICU 3.2
 670  */
 671 U_STABLE void U_EXPORT2
 672 uset_retainAll(USet* set, const USet* retain);
 673
 674 /**
 675  * Reallocate this objects internal structures to take up the least
 676  * possible space, without changing this object's value.
 677  * A frozen set will not be modified.
 678  *
 679  * @param set the object on which to perfrom the compact
 680  * @stable ICU 3.2
 681  */
 682 U_STABLE void U_EXPORT2
 683 uset_compact(USet* set);
 684
 685 /**
 686  * Inverts this set.  This operation modifies this set so that
 687  * its value is its complement.  This operation does not affect
 688  * the multicharacter strings, if any.
 689  * A frozen set will not be modified.
 690  * @param set the set
 691  * @stable ICU 2.4
 692  */
 693 U_STABLE void U_EXPORT2
 694 uset_complement(USet* set);
 695
 696 /**
 697  * Complements in this set all elements contained in the specified
 698  * set.  Any character in the other set will be removed if it is
 699  * in this set, or will be added if it is not in this set.
 700  * A frozen set will not be modified.
 701  *
 702  * @param set the set with which to complement
 703  * @param complement set that defines which elements will be xor'ed
 704  * from this set.
 705  * @stable ICU 3.2
 706  */
 707 U_STABLE void U_EXPORT2
 708 uset_complementAll(USet* set, const USet* complement);
 709
 710 /**
 711  * Removes all of the elements from this set.  This set will be
 712  * empty after this call returns.
 713  * A frozen set will not be modified.
 714  * @param set the set
 715  * @stable ICU 2.4
 716  */
 717 U_STABLE void U_EXPORT2
 718 uset_clear(USet* set);
 719
 720 /**
 721  * Close this set over the given attribute.  For the attribute
 722  * USET_CASE, the result is to modify this set so that:
 723  *
 724  * 1. For each character or string 'a' in this set, all strings or
 725  * characters 'b' such that foldCase(a) == foldCase(b) are added
 726  * to this set.
 727  *
 728  * 2. For each string 'e' in the resulting set, if e !=
 729  * foldCase(e), 'e' will be removed.
 730  *
 731  * Example: [aq\\u00DF{Bc}{bC}{Fi}] => [aAqQ\\u00DF\\uFB01{ss}{bc}{fi}]
 732  *
 733  * (Here foldCase(x) refers to the operation u_strFoldCase, and a
 734  * == b denotes that the contents are the same, not pointer
 735  * comparison.)
 736  *
 737  * A frozen set will not be modified.
 738  *
 739  * @param set the set
 740  *
 741  * @param attributes bitmask for attributes to close over.
 742  * Currently only the USET_CASE bit is supported.  Any undefined bits
 743  * are ignored.
 744  * @stable ICU 4.2
 745  */
 746 U_STABLE void U_EXPORT2
 747 uset_closeOver(USet* set, int32_t attributes);
 748
 749 /**
 750  * Remove all strings from this set.
 751  *
 752  * @param set the set
 753  * @stable ICU 4.2
 754  */
 755 U_STABLE void U_EXPORT2
 756 uset_removeAllStrings(USet* set);
 757
 758 /**
 759  * Returns TRUE if the given USet contains no characters and no
 760  * strings.
 761  * @param set the set
 762  * @return true if set is empty
 763  * @stable ICU 2.4
 764  */
 765 U_STABLE UBool U_EXPORT2
 766 uset_isEmpty(const USet* set);
 767
 768 /**
 769  * Returns TRUE if the given USet contains the given character.
 770  * This function works faster with a frozen set.
 771  * @param set the set
 772  * @param c The codepoint to check for within the set
 773  * @return true if set contains c
 774  * @stable ICU 2.4
 775  */
 776 U_STABLE UBool U_EXPORT2
 777 uset_contains(const USet* set, UChar32 c);
 778
 779 /**
 780  * Returns TRUE if the given USet contains all characters c
 781  * where start <= c && c <= end.
 782  * @param set the set
 783  * @param start the first character of the range to test, inclusive
 784  * @param end the last character of the range to test, inclusive
 785  * @return TRUE if set contains the range
 786  * @stable ICU 2.2
 787  */
 788 U_STABLE UBool U_EXPORT2
 789 uset_containsRange(const USet* set, UChar32 start, UChar32 end);
 790
 791 /**
 792  * Returns TRUE if the given USet contains the given string.
 793  * @param set the set
 794  * @param str the string
 795  * @param strLen the length of the string or -1 if null terminated.
 796  * @return true if set contains str
 797  * @stable ICU 2.4
 798  */
 799 U_STABLE UBool U_EXPORT2
 800 uset_containsString(const USet* set, const UChar* str, int32_t strLen);
 801
 802 /**
 803  * Returns the index of the given character within this set, where
 804  * the set is ordered by ascending code point.  If the character
 805  * is not in this set, return -1.  The inverse of this method is
 806  * <code>charAt()</code>.
 807  * @param set the set
 808  * @param c the character to obtain the index for
 809  * @return an index from 0..size()-1, or -1
 810  * @stable ICU 3.2
 811  */
 812 U_STABLE int32_t U_EXPORT2
 813 uset_indexOf(const USet* set, UChar32 c);
 814
 815 /**
 816  * Returns the character at the given index within this set, where
 817  * the set is ordered by ascending code point.  If the index is
 818  * out of range, return (UChar32)-1.  The inverse of this method is
 819  * <code>indexOf()</code>.
 820  * @param set the set
 821  * @param charIndex an index from 0..size()-1 to obtain the char for
 822  * @return the character at the given index, or (UChar32)-1.
 823  * @stable ICU 3.2
 824  */
 825 U_STABLE UChar32 U_EXPORT2
 826 uset_charAt(const USet* set, int32_t charIndex);
 827
 828 /**
 829  * Returns the number of characters and strings contained in the given
 830  * USet.
 831  * @param set the set
 832  * @return a non-negative integer counting the characters and strings
 833  * contained in set
 834  * @stable ICU 2.4
 835  */
 836 U_STABLE int32_t U_EXPORT2
 837 uset_size(const USet* set);
 838
 839 /**
 840  * Returns the number of items in this set.  An item is either a range
 841  * of characters or a single multicharacter string.
 842  * @param set the set
 843  * @return a non-negative integer counting the character ranges
 844  * and/or strings contained in set
 845  * @stable ICU 2.4
 846  */
 847 U_STABLE int32_t U_EXPORT2
 848 uset_getItemCount(const USet* set);
 849
 850 /**
 851  * Returns an item of this set.  An item is either a range of
 852  * characters or a single multicharacter string.
 853  * @param set the set
 854  * @param itemIndex a non-negative integer in the range 0..
 855  * uset_getItemCount(set)-1
 856  * @param start pointer to variable to receive first character
 857  * in range, inclusive
 858  * @param end pointer to variable to receive last character in range,
 859  * inclusive
 860  * @param str buffer to receive the string, may be NULL
 861  * @param strCapacity capacity of str, or 0 if str is NULL
 862  * @param ec error code
 863  * @return the length of the string (>= 2), or 0 if the item is a
 864  * range, in which case it is the range *start..*end, or -1 if
 865  * itemIndex is out of range
 866  * @stable ICU 2.4
 867  */
 868 U_STABLE int32_t U_EXPORT2
 869 uset_getItem(const USet* set, int32_t itemIndex,
 870              UChar32* start, UChar32* end,
 871              UChar* str, int32_t strCapacity,
 872              UErrorCode* ec);
 873
 874 /**
 875  * Returns true if set1 contains all the characters and strings
 876  * of set2. It answers the question, 'Is set1 a superset of set2?'
 877  * @param set1 set to be checked for containment
 878  * @param set2 set to be checked for containment
 879  * @return true if the test condition is met
 880  * @stable ICU 3.2
 881  */
 882 U_STABLE UBool U_EXPORT2
 883 uset_containsAll(const USet* set1, const USet* set2);
 884
 885 /**
 886  * Returns true if this set contains all the characters
 887  * of the given string. This is does not check containment of grapheme
 888  * clusters, like uset_containsString.
 889  * @param set set of characters to be checked for containment
 890  * @param str string containing codepoints to be checked for containment
 891  * @param strLen the length of the string or -1 if null terminated.
 892  * @return true if the test condition is met
 893  * @stable ICU 3.4
 894  */
 895 U_STABLE UBool U_EXPORT2
 896 uset_containsAllCodePoints(const USet* set, const UChar *str, int32_t strLen);
 897
 898 /**
 899  * Returns true if set1 contains none of the characters and strings
 900  * of set2. It answers the question, 'Is set1 a disjoint set of set2?'
 901  * @param set1 set to be checked for containment
 902  * @param set2 set to be checked for containment
 903  * @return true if the test condition is met
 904  * @stable ICU 3.2
 905  */
 906 U_STABLE UBool U_EXPORT2
 907 uset_containsNone(const USet* set1, const USet* set2);
 908
 909 /**
 910  * Returns true if set1 contains some of the characters and strings
 911  * of set2. It answers the question, 'Does set1 and set2 have an intersection?'
 912  * @param set1 set to be checked for containment
 913  * @param set2 set to be checked for containment
 914  * @return true if the test condition is met
 915  * @stable ICU 3.2
 916  */
 917 U_STABLE UBool U_EXPORT2
 918 uset_containsSome(const USet* set1, const USet* set2);
 919
 920 /**
 921  * Returns the length of the initial substring of the input string which
 922  * consists only of characters and strings that are contained in this set
 923  * (USET_SPAN_CONTAINED, USET_SPAN_SIMPLE),
 924  * or only of characters and strings that are not contained
 925  * in this set (USET_SPAN_NOT_CONTAINED).
 926  * See USetSpanCondition for details.
 927  * Similar to the strspn() C library function.
 928  * Unpaired surrogates are treated according to contains() of their surrogate code points.
 929  * This function works faster with a frozen set and with a non-negative string length argument.
 930  * @param set the set
 931  * @param s start of the string
 932  * @param length of the string; can be -1 for NUL-terminated
 933  * @param spanCondition specifies the containment condition
 934  * @return the length of the initial substring according to the spanCondition;
 935  *         0 if the start of the string does not fit the spanCondition
 936  * @stable ICU 3.8
 937  * @see USetSpanCondition
 938  */
 939 U_STABLE int32_t U_EXPORT2
 940 uset_span(const USet *set, const UChar *s, int32_t length, USetSpanCondition spanCondition);
 941
 942 /**
 943  * Returns the start of the trailing substring of the input string which
 944  * consists only of characters and strings that are contained in this set
 945  * (USET_SPAN_CONTAINED, USET_SPAN_SIMPLE),
 946  * or only of characters and strings that are not contained
 947  * in this set (USET_SPAN_NOT_CONTAINED).
 948  * See USetSpanCondition for details.
 949  * Unpaired surrogates are treated according to contains() of their surrogate code points.
 950  * This function works faster with a frozen set and with a non-negative string length argument.
 951  * @param set the set
 952  * @param s start of the string
 953  * @param length of the string; can be -1 for NUL-terminated
 954  * @param spanCondition specifies the containment condition
 955  * @return the start of the trailing substring according to the spanCondition;
 956  *         the string length if the end of the string does not fit the spanCondition
 957  * @stable ICU 3.8
 958  * @see USetSpanCondition
 959  */
 960 U_STABLE int32_t U_EXPORT2
 961 uset_spanBack(const USet *set, const UChar *s, int32_t length, USetSpanCondition spanCondition);
 962
 963 /**
 964  * Returns the length of the initial substring of the input string which
 965  * consists only of characters and strings that are contained in this set
 966  * (USET_SPAN_CONTAINED, USET_SPAN_SIMPLE),
 967  * or only of characters and strings that are not contained
 968  * in this set (USET_SPAN_NOT_CONTAINED).
 969  * See USetSpanCondition for details.
 970  * Similar to the strspn() C library function.
 971  * Malformed byte sequences are treated according to contains(0xfffd).
 972  * This function works faster with a frozen set and with a non-negative string length argument.
 973  * @param set the set
 974  * @param s start of the string (UTF-8)
 975  * @param length of the string; can be -1 for NUL-terminated
 976  * @param spanCondition specifies the containment condition
 977  * @return the length of the initial substring according to the spanCondition;
 978  *         0 if the start of the string does not fit the spanCondition
 979  * @stable ICU 3.8
 980  * @see USetSpanCondition
 981  */
 982 U_STABLE int32_t U_EXPORT2
 983 uset_spanUTF8(const USet *set, const char *s, int32_t length, USetSpanCondition spanCondition);
 984
 985 /**
 986  * Returns the start of the trailing substring of the input string which
 987  * consists only of characters and strings that are contained in this set
 988  * (USET_SPAN_CONTAINED, USET_SPAN_SIMPLE),
 989  * or only of characters and strings that are not contained
 990  * in this set (USET_SPAN_NOT_CONTAINED).
 991  * See USetSpanCondition for details.
 992  * Malformed byte sequences are treated according to contains(0xfffd).
 993  * This function works faster with a frozen set and with a non-negative string length argument.
 994  * @param set the set
 995  * @param s start of the string (UTF-8)
 996  * @param length of the string; can be -1 for NUL-terminated
 997  * @param spanCondition specifies the containment condition
 998  * @return the start of the trailing substring according to the spanCondition;
 999  *         the string length if the end of the string does not fit the spanCondition
1000  * @stable ICU 3.8
1001  * @see USetSpanCondition
1002  */
1003 U_STABLE int32_t U_EXPORT2
1004 uset_spanBackUTF8(const USet *set, const char *s, int32_t length, USetSpanCondition spanCondition);
1005
1006 /**
1007  * Returns true if set1 contains all of the characters and strings
1008  * of set2, and vis versa. It answers the question, 'Is set1 equal to set2?'
1009  * @param set1 set to be checked for containment
1010  * @param set2 set to be checked for containment
1011  * @return true if the test condition is met
1012  * @stable ICU 3.2
1013  */
1014 U_STABLE UBool U_EXPORT2
1015 uset_equals(const USet* set1, const USet* set2);
1016
1017 /*********************************************************************
1018  * Serialized set API
1019  *********************************************************************/
1020
1021 /**
1022  * Serializes this set into an array of 16-bit integers.  Serialization
1023  * (currently) only records the characters in the set; multicharacter
1024  * strings are ignored.
1025  *
1026  * The array
1027  * has following format (each line is one 16-bit integer):
1028  *
1029  *  length     = (n+2*m) | (m!=0?0x8000:0)
1030  *  bmpLength  = n; present if m!=0
1031  *  bmp[0]
1032  *  bmp[1]
1033  *  ...
1034  *  bmp[n-1]
1035  *  supp-high[0]
1036  *  supp-low[0]
1037  *  supp-high[1]
1038  *  supp-low[1]
1039  *  ...
1040  *  supp-high[m-1]
1041  *  supp-low[m-1]
1042  *
1043  * The array starts with a header.  After the header are n bmp
1044  * code points, then m supplementary code points.  Either n or m
1045  * or both may be zero.  n+2*m is always <= 0x7FFF.
1046  *
1047  * If there are no supplementary characters (if m==0) then the
1048  * header is one 16-bit integer, 'length', with value n.
1049  *
1050  * If there are supplementary characters (if m!=0) then the header
1051  * is two 16-bit integers.  The first, 'length', has value
1052  * (n+2*m)|0x8000.  The second, 'bmpLength', has value n.
1053  *
1054  * After the header the code points are stored in ascending order.
1055  * Supplementary code points are stored as most significant 16
1056  * bits followed by least significant 16 bits.
1057  *
1058  * @param set the set
1059  * @param dest pointer to buffer of destCapacity 16-bit integers.
1060  * May be NULL only if destCapacity is zero.
1061  * @param destCapacity size of dest, or zero.  Must not be negative.
1062  * @param pErrorCode pointer to the error code.  Will be set to
1063  * U_INDEX_OUTOFBOUNDS_ERROR if n+2*m > 0x7FFF.  Will be set to
1064  * U_BUFFER_OVERFLOW_ERROR if n+2*m+(m!=0?2:1) > destCapacity.
1065  * @return the total length of the serialized format, including
1066  * the header, that is, n+2*m+(m!=0?2:1), or 0 on error other
1067  * than U_BUFFER_OVERFLOW_ERROR.
1068  * @stable ICU 2.4
1069  */
1070 U_STABLE int32_t U_EXPORT2
1071 uset_serialize(const USet* set, uint16_t* dest, int32_t destCapacity, UErrorCode* pErrorCode);
1072
1073 /**
1074  * Given a serialized array, fill in the given serialized set object.
1075  * @param fillSet pointer to result
1076  * @param src pointer to start of array
1077  * @param srcLength length of array
1078  * @return true if the given array is valid, otherwise false
1079  * @stable ICU 2.4
1080  */
1081 U_STABLE UBool U_EXPORT2
1082 uset_getSerializedSet(USerializedSet* fillSet, const uint16_t* src, int32_t srcLength);
1083
1084 /**
1085  * Set the USerializedSet to contain the given character (and nothing
1086  * else).
1087  * @param fillSet pointer to result
1088  * @param c The codepoint to set
1089  * @stable ICU 2.4
1090  */
1091 U_STABLE void U_EXPORT2
1092 uset_setSerializedToOne(USerializedSet* fillSet, UChar32 c);
1093
1094 /**
1095  * Returns TRUE if the given USerializedSet contains the given
1096  * character.
1097  * @param set the serialized set
1098  * @param c The codepoint to check for within the set
1099  * @return true if set contains c
1100  * @stable ICU 2.4
1101  */
1102 U_STABLE UBool U_EXPORT2
1103 uset_serializedContains(const USerializedSet* set, UChar32 c);
1104
1105 /**
1106  * Returns the number of disjoint ranges of characters contained in
1107  * the given serialized set.  Ignores any strings contained in the
1108  * set.
1109  * @param set the serialized set
1110  * @return a non-negative integer counting the character ranges
1111  * contained in set
1112  * @stable ICU 2.4
1113  */
1114 U_STABLE int32_t U_EXPORT2
1115 uset_getSerializedRangeCount(const USerializedSet* set);
1116
1117 /**
1118  * Returns a range of characters contained in the given serialized
1119  * set.
1120  * @param set the serialized set
1121  * @param rangeIndex a non-negative integer in the range 0..
1122  * uset_getSerializedRangeCount(set)-1
1123  * @param pStart pointer to variable to receive first character
1124  * in range, inclusive
1125  * @param pEnd pointer to variable to receive last character in range,
1126  * inclusive
1127  * @return true if rangeIndex is valid, otherwise false
1128  * @stable ICU 2.4
1129  */
1130 U_STABLE UBool U_EXPORT2
1131 uset_getSerializedRange(const USerializedSet* set, int32_t rangeIndex,
1132                         UChar32* pStart, UChar32* pEnd);
1133
1134 #endif