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