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) 2002-2011, International Business Machines |
7 | * Corporation and others. All Rights Reserved. |
8 | * |
9 | ******************************************************************************* |
10 | * file name: uset.cpp |
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 | * There are functions to efficiently serialize a USet into an array of uint16_t |
19 | * and functions to use such a serialized form efficiently without |
20 | * instantiating a new USet. |
21 | */ |
22 | |
23 | #include "unicode/utypes.h" |
24 | #include "unicode/uobject.h" |
25 | #include "unicode/uset.h" |
26 | #include "unicode/uniset.h" |
27 | #include "cmemory.h" |
28 | #include "unicode/ustring.h" |
29 | #include "unicode/parsepos.h" |
30 | |
31 | U_NAMESPACE_USE |
32 | |
33 | U_CAPI USet* U_EXPORT2 |
34 | uset_openEmpty() { |
35 | return (USet*) new UnicodeSet(); |
36 | } |
37 | |
38 | U_CAPI USet* U_EXPORT2 |
39 | uset_open(UChar32 start, UChar32 end) { |
40 | return (USet*) new UnicodeSet(start, end); |
41 | } |
42 | |
43 | U_CAPI void U_EXPORT2 |
44 | uset_close(USet* set) { |
45 | delete (UnicodeSet*) set; |
46 | } |
47 | |
48 | U_CAPI USet * U_EXPORT2 |
49 | uset_clone(const USet *set) { |
50 | return (USet*) (((UnicodeSet*) set)->UnicodeSet::clone()); |
51 | } |
52 | |
53 | U_CAPI UBool U_EXPORT2 |
54 | uset_isFrozen(const USet *set) { |
55 | return ((UnicodeSet*) set)->UnicodeSet::isFrozen(); |
56 | } |
57 | |
58 | U_CAPI void U_EXPORT2 |
59 | uset_freeze(USet *set) { |
60 | ((UnicodeSet*) set)->UnicodeSet::freeze(); |
61 | } |
62 | |
63 | U_CAPI USet * U_EXPORT2 |
64 | uset_cloneAsThawed(const USet *set) { |
65 | return (USet*) (((UnicodeSet*) set)->UnicodeSet::cloneAsThawed()); |
66 | } |
67 | |
68 | U_CAPI void U_EXPORT2 |
69 | uset_set(USet* set, |
70 | UChar32 start, UChar32 end) { |
71 | ((UnicodeSet*) set)->UnicodeSet::set(start, end); |
72 | } |
73 | |
74 | U_CAPI void U_EXPORT2 |
75 | uset_addAll(USet* set, const USet *additionalSet) { |
76 | ((UnicodeSet*) set)->UnicodeSet::addAll(*((const UnicodeSet*)additionalSet)); |
77 | } |
78 | |
79 | U_CAPI void U_EXPORT2 |
80 | uset_add(USet* set, UChar32 c) { |
81 | ((UnicodeSet*) set)->UnicodeSet::add(c); |
82 | } |
83 | |
84 | U_CAPI void U_EXPORT2 |
85 | uset_addRange(USet* set, UChar32 start, UChar32 end) { |
86 | ((UnicodeSet*) set)->UnicodeSet::add(start, end); |
87 | } |
88 | |
89 | U_CAPI void U_EXPORT2 |
90 | uset_addString(USet* set, const char16_t* str, int32_t strLen) { |
91 | // UnicodeString handles -1 for strLen |
92 | UnicodeString s(strLen<0, str, strLen); |
93 | ((UnicodeSet*) set)->UnicodeSet::add(s); |
94 | } |
95 | |
96 | U_CAPI void U_EXPORT2 |
97 | uset_addAllCodePoints(USet* set, const char16_t *str, int32_t strLen) { |
98 | // UnicodeString handles -1 for strLen |
99 | UnicodeString s(str, strLen); |
100 | ((UnicodeSet*) set)->UnicodeSet::addAll(s); |
101 | } |
102 | |
103 | U_CAPI void U_EXPORT2 |
104 | uset_remove(USet* set, UChar32 c) { |
105 | ((UnicodeSet*) set)->UnicodeSet::remove(c); |
106 | } |
107 | |
108 | U_CAPI void U_EXPORT2 |
109 | uset_removeRange(USet* set, UChar32 start, UChar32 end) { |
110 | ((UnicodeSet*) set)->UnicodeSet::remove(start, end); |
111 | } |
112 | |
113 | U_CAPI void U_EXPORT2 |
114 | uset_removeString(USet* set, const char16_t* str, int32_t strLen) { |
115 | UnicodeString s(strLen==-1, str, strLen); |
116 | ((UnicodeSet*) set)->UnicodeSet::remove(s); |
117 | } |
118 | |
119 | U_CAPI void U_EXPORT2 |
120 | uset_removeAllCodePoints(USet *set, const char16_t *str, int32_t length) { |
121 | UnicodeString s(length==-1, str, length); |
122 | ((UnicodeSet*) set)->UnicodeSet::removeAll(s); |
123 | } |
124 | |
125 | U_CAPI void U_EXPORT2 |
126 | uset_removeAll(USet* set, const USet* remove) { |
127 | ((UnicodeSet*) set)->UnicodeSet::removeAll(*(const UnicodeSet*)remove); |
128 | } |
129 | |
130 | U_CAPI void U_EXPORT2 |
131 | uset_retain(USet* set, UChar32 start, UChar32 end) { |
132 | ((UnicodeSet*) set)->UnicodeSet::retain(start, end); |
133 | } |
134 | |
135 | U_CAPI void U_EXPORT2 |
136 | uset_retainString(USet *set, const char16_t *str, int32_t length) { |
137 | UnicodeString s(length==-1, str, length); |
138 | ((UnicodeSet*) set)->UnicodeSet::retain(s); |
139 | } |
140 | |
141 | U_CAPI void U_EXPORT2 |
142 | uset_retainAllCodePoints(USet *set, const char16_t *str, int32_t length) { |
143 | UnicodeString s(length==-1, str, length); |
144 | ((UnicodeSet*) set)->UnicodeSet::retainAll(s); |
145 | } |
146 | |
147 | U_CAPI void U_EXPORT2 |
148 | uset_retainAll(USet* set, const USet* retain) { |
149 | ((UnicodeSet*) set)->UnicodeSet::retainAll(*(const UnicodeSet*)retain); |
150 | } |
151 | |
152 | U_CAPI void U_EXPORT2 |
153 | uset_compact(USet* set) { |
154 | ((UnicodeSet*) set)->UnicodeSet::compact(); |
155 | } |
156 | |
157 | U_CAPI void U_EXPORT2 |
158 | uset_complement(USet* set) { |
159 | ((UnicodeSet*) set)->UnicodeSet::complement(); |
160 | } |
161 | |
162 | U_CAPI void U_EXPORT2 |
163 | uset_complementRange(USet *set, UChar32 start, UChar32 end) { |
164 | ((UnicodeSet*) set)->UnicodeSet::complement(start, end); |
165 | } |
166 | |
167 | U_CAPI void U_EXPORT2 |
168 | uset_complementString(USet *set, const char16_t *str, int32_t length) { |
169 | UnicodeString s(length==-1, str, length); |
170 | ((UnicodeSet*) set)->UnicodeSet::complement(s); |
171 | } |
172 | |
173 | U_CAPI void U_EXPORT2 |
174 | uset_complementAllCodePoints(USet *set, const char16_t *str, int32_t length) { |
175 | UnicodeString s(length==-1, str, length); |
176 | ((UnicodeSet*) set)->UnicodeSet::complementAll(s); |
177 | } |
178 | |
179 | U_CAPI void U_EXPORT2 |
180 | uset_complementAll(USet* set, const USet* complement) { |
181 | ((UnicodeSet*) set)->UnicodeSet::complementAll(*(const UnicodeSet*)complement); |
182 | } |
183 | |
184 | U_CAPI void U_EXPORT2 |
185 | uset_clear(USet* set) { |
186 | ((UnicodeSet*) set)->UnicodeSet::clear(); |
187 | } |
188 | |
189 | U_CAPI void U_EXPORT2 |
190 | uset_removeAllStrings(USet* set) { |
191 | ((UnicodeSet*) set)->UnicodeSet::removeAllStrings(); |
192 | } |
193 | |
194 | U_CAPI UBool U_EXPORT2 |
195 | uset_isEmpty(const USet* set) { |
196 | return ((const UnicodeSet*) set)->UnicodeSet::isEmpty(); |
197 | } |
198 | |
199 | U_CAPI UBool U_EXPORT2 |
200 | uset_hasStrings(const USet* set) { |
201 | return ((const UnicodeSet*) set)->UnicodeSet::hasStrings(); |
202 | } |
203 | |
204 | U_CAPI UBool U_EXPORT2 |
205 | uset_contains(const USet* set, UChar32 c) { |
206 | return ((const UnicodeSet*) set)->UnicodeSet::contains(c); |
207 | } |
208 | |
209 | U_CAPI UBool U_EXPORT2 |
210 | uset_containsRange(const USet* set, UChar32 start, UChar32 end) { |
211 | return ((const UnicodeSet*) set)->UnicodeSet::contains(start, end); |
212 | } |
213 | |
214 | U_CAPI UBool U_EXPORT2 |
215 | uset_containsString(const USet* set, const char16_t* str, int32_t strLen) { |
216 | UnicodeString s(strLen==-1, str, strLen); |
217 | return ((const UnicodeSet*) set)->UnicodeSet::contains(s); |
218 | } |
219 | |
220 | U_CAPI UBool U_EXPORT2 |
221 | uset_containsAll(const USet* set1, const USet* set2) { |
222 | return ((const UnicodeSet*) set1)->UnicodeSet::containsAll(* (const UnicodeSet*) set2); |
223 | } |
224 | |
225 | U_CAPI UBool U_EXPORT2 |
226 | uset_containsAllCodePoints(const USet* set, const char16_t *str, int32_t strLen) { |
227 | // Create a string alias, since nothing is being added to the set. |
228 | UnicodeString s(strLen==-1, str, strLen); |
229 | return ((const UnicodeSet*) set)->UnicodeSet::containsAll(s); |
230 | } |
231 | |
232 | U_CAPI UBool U_EXPORT2 |
233 | uset_containsNone(const USet* set1, const USet* set2) { |
234 | return ((const UnicodeSet*) set1)->UnicodeSet::containsNone(* (const UnicodeSet*) set2); |
235 | } |
236 | |
237 | U_CAPI UBool U_EXPORT2 |
238 | uset_containsSome(const USet* set1, const USet* set2) { |
239 | return ((const UnicodeSet*) set1)->UnicodeSet::containsSome(* (const UnicodeSet*) set2); |
240 | } |
241 | |
242 | U_CAPI int32_t U_EXPORT2 |
243 | uset_span(const USet *set, const char16_t *s, int32_t length, USetSpanCondition spanCondition) { |
244 | return ((UnicodeSet*) set)->UnicodeSet::span(s, length, spanCondition); |
245 | } |
246 | |
247 | U_CAPI int32_t U_EXPORT2 |
248 | uset_spanBack(const USet *set, const char16_t *s, int32_t length, USetSpanCondition spanCondition) { |
249 | return ((UnicodeSet*) set)->UnicodeSet::spanBack(s, length, spanCondition); |
250 | } |
251 | |
252 | U_CAPI int32_t U_EXPORT2 |
253 | uset_spanUTF8(const USet *set, const char *s, int32_t length, USetSpanCondition spanCondition) { |
254 | return ((UnicodeSet*) set)->UnicodeSet::spanUTF8(s, length, spanCondition); |
255 | } |
256 | |
257 | U_CAPI int32_t U_EXPORT2 |
258 | uset_spanBackUTF8(const USet *set, const char *s, int32_t length, USetSpanCondition spanCondition) { |
259 | return ((UnicodeSet*) set)->UnicodeSet::spanBackUTF8(s, length, spanCondition); |
260 | } |
261 | |
262 | U_CAPI UBool U_EXPORT2 |
263 | uset_equals(const USet* set1, const USet* set2) { |
264 | return *(const UnicodeSet*)set1 == *(const UnicodeSet*)set2; |
265 | } |
266 | |
267 | U_CAPI int32_t U_EXPORT2 |
268 | uset_indexOf(const USet* set, UChar32 c) { |
269 | return ((UnicodeSet*) set)->UnicodeSet::indexOf(c); |
270 | } |
271 | |
272 | U_CAPI UChar32 U_EXPORT2 |
273 | uset_charAt(const USet* set, int32_t index) { |
274 | return ((UnicodeSet*) set)->UnicodeSet::charAt(index); |
275 | } |
276 | |
277 | U_CAPI int32_t U_EXPORT2 |
278 | uset_size(const USet* set) { |
279 | return ((const UnicodeSet*) set)->UnicodeSet::size(); |
280 | } |
281 | |
282 | U_NAMESPACE_BEGIN |
283 | /** |
284 | * This class only exists to provide access to the UnicodeSet private |
285 | * USet support API. Declaring a class a friend is more portable than |
286 | * trying to declare extern "C" functions as friends. |
287 | */ |
288 | class USetAccess /* not : public UObject because all methods are static */ { |
289 | public: |
290 | /* Try to have the compiler inline these*/ |
291 | inline static int32_t getStringCount(const UnicodeSet& set) { |
292 | return set.stringsSize(); |
293 | } |
294 | inline static const UnicodeString* getString(const UnicodeSet& set, |
295 | int32_t i) { |
296 | return set.getString(i); |
297 | } |
298 | private: |
299 | /* do not instantiate*/ |
300 | USetAccess(); |
301 | }; |
302 | U_NAMESPACE_END |
303 | |
304 | U_CAPI int32_t U_EXPORT2 |
305 | uset_getRangeCount(const USet *set) { |
306 | return ((const UnicodeSet *)set)->UnicodeSet::getRangeCount(); |
307 | } |
308 | |
309 | U_CAPI int32_t U_EXPORT2 |
310 | uset_getItemCount(const USet* uset) { |
311 | const UnicodeSet& set = *(const UnicodeSet*)uset; |
312 | return set.getRangeCount() + USetAccess::getStringCount(set); |
313 | } |
314 | |
315 | U_CAPI int32_t U_EXPORT2 |
316 | uset_getItem(const USet* uset, int32_t itemIndex, |
317 | UChar32* start, UChar32* end, |
318 | char16_t* str, int32_t strCapacity, |
319 | UErrorCode* ec) { |
320 | if (U_FAILURE(*ec)) return 0; |
321 | const UnicodeSet& set = *(const UnicodeSet*)uset; |
322 | int32_t rangeCount; |
323 | |
324 | if (itemIndex < 0) { |
325 | *ec = U_ILLEGAL_ARGUMENT_ERROR; |
326 | return -1; |
327 | } else if (itemIndex < (rangeCount = set.getRangeCount())) { |
328 | *start = set.getRangeStart(itemIndex); |
329 | *end = set.getRangeEnd(itemIndex); |
330 | return 0; |
331 | } else { |
332 | itemIndex -= rangeCount; |
333 | if (itemIndex < USetAccess::getStringCount(set)) { |
334 | const UnicodeString* s = USetAccess::getString(set, itemIndex); |
335 | return s->extract(str, strCapacity, *ec); |
336 | } else { |
337 | *ec = U_INDEX_OUTOFBOUNDS_ERROR; |
338 | return -1; |
339 | } |
340 | } |
341 | } |
342 | |
343 | //U_CAPI UBool U_EXPORT2 |
344 | //uset_getRange(const USet* set, int32_t rangeIndex, |
345 | // UChar32* pStart, UChar32* pEnd) { |
346 | // if ((uint32_t) rangeIndex >= (uint32_t) uset_getRangeCount(set)) { |
347 | // return false; |
348 | // } |
349 | // const UnicodeSet* us = (const UnicodeSet*) set; |
350 | // *pStart = us->getRangeStart(rangeIndex); |
351 | // *pEnd = us->getRangeEnd(rangeIndex); |
352 | // return true; |
353 | //} |
354 | |
355 | /* |
356 | * Serialize a USet into 16-bit units. |
357 | * Store BMP code points as themselves with one 16-bit unit each. |
358 | * |
359 | * Important: the code points in the array are in ascending order, |
360 | * therefore all BMP code points precede all supplementary code points. |
361 | * |
362 | * Store each supplementary code point in 2 16-bit units, |
363 | * simply with higher-then-lower 16-bit halves. |
364 | * |
365 | * Precede the entire list with the length. |
366 | * If there are supplementary code points, then set bit 15 in the length |
367 | * and add the bmpLength between it and the array. |
368 | * |
369 | * In other words: |
370 | * - all BMP: (length=bmpLength) BMP, .., BMP |
371 | * - some supplementary: (length|0x8000) (bmpLength<length) BMP, .., BMP, supp-high, supp-low, .. |
372 | */ |
373 | U_CAPI int32_t U_EXPORT2 |
374 | uset_serialize(const USet* set, uint16_t* dest, int32_t destCapacity, UErrorCode* ec) { |
375 | if (ec==nullptr || U_FAILURE(*ec)) { |
376 | return 0; |
377 | } |
378 | |
379 | return ((const UnicodeSet*) set)->UnicodeSet::serialize(dest, destCapacity,* ec); |
380 | } |
381 | |
382 | U_CAPI UBool U_EXPORT2 |
383 | uset_getSerializedSet(USerializedSet* fillSet, const uint16_t* src, int32_t srcLength) { |
384 | int32_t length; |
385 | |
386 | if(fillSet==nullptr) { |
387 | return false; |
388 | } |
389 | if(src==nullptr || srcLength<=0) { |
390 | fillSet->length=fillSet->bmpLength=0; |
391 | return false; |
392 | } |
393 | |
394 | length=*src++; |
395 | if(length&0x8000) { |
396 | /* there are supplementary values */ |
397 | length&=0x7fff; |
398 | if(srcLength<(2+length)) { |
399 | fillSet->length=fillSet->bmpLength=0; |
400 | return false; |
401 | } |
402 | fillSet->bmpLength=*src++; |
403 | } else { |
404 | /* only BMP values */ |
405 | if(srcLength<(1+length)) { |
406 | fillSet->length=fillSet->bmpLength=0; |
407 | return false; |
408 | } |
409 | fillSet->bmpLength=length; |
410 | } |
411 | fillSet->array=src; |
412 | fillSet->length=length; |
413 | return true; |
414 | } |
415 | |
416 | U_CAPI void U_EXPORT2 |
417 | uset_setSerializedToOne(USerializedSet* fillSet, UChar32 c) { |
418 | if(fillSet==nullptr || (uint32_t)c>0x10ffff) { |
419 | return; |
420 | } |
421 | |
422 | fillSet->array=fillSet->staticArray; |
423 | if(c<0xffff) { |
424 | fillSet->bmpLength=fillSet->length=2; |
425 | fillSet->staticArray[0]=(uint16_t)c; |
426 | fillSet->staticArray[1]=(uint16_t)c+1; |
427 | } else if(c==0xffff) { |
428 | fillSet->bmpLength=1; |
429 | fillSet->length=3; |
430 | fillSet->staticArray[0]=0xffff; |
431 | fillSet->staticArray[1]=1; |
432 | fillSet->staticArray[2]=0; |
433 | } else if(c<0x10ffff) { |
434 | fillSet->bmpLength=0; |
435 | fillSet->length=4; |
436 | fillSet->staticArray[0]=(uint16_t)(c>>16); |
437 | fillSet->staticArray[1]=(uint16_t)c; |
438 | ++c; |
439 | fillSet->staticArray[2]=(uint16_t)(c>>16); |
440 | fillSet->staticArray[3]=(uint16_t)c; |
441 | } else /* c==0x10ffff */ { |
442 | fillSet->bmpLength=0; |
443 | fillSet->length=2; |
444 | fillSet->staticArray[0]=0x10; |
445 | fillSet->staticArray[1]=0xffff; |
446 | } |
447 | } |
448 | |
449 | U_CAPI UBool U_EXPORT2 |
450 | uset_serializedContains(const USerializedSet* set, UChar32 c) { |
451 | const uint16_t* array; |
452 | |
453 | if(set==nullptr || (uint32_t)c>0x10ffff) { |
454 | return false; |
455 | } |
456 | |
457 | array=set->array; |
458 | if(c<=0xffff) { |
459 | /* find c in the BMP part */ |
460 | int32_t lo = 0; |
461 | int32_t hi = set->bmpLength-1; |
462 | if (c < array[0]) { |
463 | hi = 0; |
464 | } else if (c < array[hi]) { |
465 | for(;;) { |
466 | int32_t i = (lo + hi) >> 1; |
467 | if (i == lo) { |
468 | break; // Done! |
469 | } else if (c < array[i]) { |
470 | hi = i; |
471 | } else { |
472 | lo = i; |
473 | } |
474 | } |
475 | } else { |
476 | hi += 1; |
477 | } |
478 | return (UBool)(hi&1); |
479 | } else { |
480 | /* find c in the supplementary part */ |
481 | uint16_t high=(uint16_t)(c>>16), low=(uint16_t)c; |
482 | int32_t base = set->bmpLength; |
483 | int32_t lo = 0; |
484 | int32_t hi = set->length - 2 - base; |
485 | if (high < array[base] || (high==array[base] && low<array[base+1])) { |
486 | hi = 0; |
487 | } else if (high < array[base+hi] || (high==array[base+hi] && low<array[base+hi+1])) { |
488 | for (;;) { |
489 | int32_t i = ((lo + hi) >> 1) & ~1; // Guarantee even result |
490 | int32_t iabs = i + base; |
491 | if (i == lo) { |
492 | break; // Done! |
493 | } else if (high < array[iabs] || (high==array[iabs] && low<array[iabs+1])) { |
494 | hi = i; |
495 | } else { |
496 | lo = i; |
497 | } |
498 | } |
499 | } else { |
500 | hi += 2; |
501 | } |
502 | /* count pairs of 16-bit units even per BMP and check if the number of pairs is odd */ |
503 | return (UBool)(((hi+(base<<1))&2)!=0); |
504 | } |
505 | } |
506 | |
507 | U_CAPI int32_t U_EXPORT2 |
508 | uset_getSerializedRangeCount(const USerializedSet* set) { |
509 | if(set==nullptr) { |
510 | return 0; |
511 | } |
512 | |
513 | return (set->bmpLength+(set->length-set->bmpLength)/2+1)/2; |
514 | } |
515 | |
516 | U_CAPI UBool U_EXPORT2 |
517 | uset_getSerializedRange(const USerializedSet* set, int32_t rangeIndex, |
518 | UChar32* pStart, UChar32* pEnd) { |
519 | const uint16_t* array; |
520 | int32_t bmpLength, length; |
521 | |
522 | if(set==nullptr || rangeIndex<0 || pStart==nullptr || pEnd==nullptr) { |
523 | return false; |
524 | } |
525 | |
526 | array=set->array; |
527 | length=set->length; |
528 | bmpLength=set->bmpLength; |
529 | |
530 | rangeIndex*=2; /* address start/limit pairs */ |
531 | if(rangeIndex<bmpLength) { |
532 | *pStart=array[rangeIndex++]; |
533 | if(rangeIndex<bmpLength) { |
534 | *pEnd=array[rangeIndex]-1; |
535 | } else if(rangeIndex<length) { |
536 | *pEnd=((((int32_t)array[rangeIndex])<<16)|array[rangeIndex+1])-1; |
537 | } else { |
538 | *pEnd=0x10ffff; |
539 | } |
540 | return true; |
541 | } else { |
542 | rangeIndex-=bmpLength; |
543 | rangeIndex*=2; /* address pairs of pairs of units */ |
544 | length-=bmpLength; |
545 | if(rangeIndex<length) { |
546 | array+=bmpLength; |
547 | *pStart=(((int32_t)array[rangeIndex])<<16)|array[rangeIndex+1]; |
548 | rangeIndex+=2; |
549 | if(rangeIndex<length) { |
550 | *pEnd=((((int32_t)array[rangeIndex])<<16)|array[rangeIndex+1])-1; |
551 | } else { |
552 | *pEnd=0x10ffff; |
553 | } |
554 | return true; |
555 | } else { |
556 | return false; |
557 | } |
558 | } |
559 | } |
560 | |
561 | // TODO The old, internal uset.c had an efficient uset_containsOne function. |
562 | // Returned the one and only code point, or else -1 or something. |
563 | // Consider adding such a function to both C and C++ UnicodeSet/uset. |
564 | // See tools/gennorm/store.c for usage, now usetContainsOne there. |
565 | |
566 | // TODO Investigate incorporating this code into UnicodeSet to improve |
567 | // efficiency. |
568 | // --- |
569 | // #define USET_GROW_DELTA 20 |
570 | // |
571 | // static int32_t |
572 | // findChar(const UChar32* array, int32_t length, UChar32 c) { |
573 | // int32_t i; |
574 | // |
575 | // /* check the last range limit first for more efficient appending */ |
576 | // if(length>0) { |
577 | // if(c>=array[length-1]) { |
578 | // return length; |
579 | // } |
580 | // |
581 | // /* do not check the last range limit again in the loop below */ |
582 | // --length; |
583 | // } |
584 | // |
585 | // for(i=0; i<length && c>=array[i]; ++i) {} |
586 | // return i; |
587 | // } |
588 | // |
589 | // static UBool |
590 | // addRemove(USet* set, UChar32 c, int32_t doRemove) { |
591 | // int32_t i, length, more; |
592 | // |
593 | // if(set==nullptr || (uint32_t)c>0x10ffff) { |
594 | // return false; |
595 | // } |
596 | // |
597 | // length=set->length; |
598 | // i=findChar(set->array, length, c); |
599 | // if((i&1)^doRemove) { |
600 | // /* c is already in the set */ |
601 | // return true; |
602 | // } |
603 | // |
604 | // /* how many more array items do we need? */ |
605 | // if(i<length && (c+1)==set->array[i]) { |
606 | // /* c is just before the following range, extend that in-place by one */ |
607 | // set->array[i]=c; |
608 | // if(i>0) { |
609 | // --i; |
610 | // if(c==set->array[i]) { |
611 | // /* the previous range collapsed, remove it */ |
612 | // set->length=length-=2; |
613 | // if(i<length) { |
614 | // uprv_memmove(set->array+i, set->array+i+2, (length-i)*4); |
615 | // } |
616 | // } |
617 | // } |
618 | // return true; |
619 | // } else if(i>0 && c==set->array[i-1]) { |
620 | // /* c is just after the previous range, extend that in-place by one */ |
621 | // if(++c<=0x10ffff) { |
622 | // set->array[i-1]=c; |
623 | // if(i<length && c==set->array[i]) { |
624 | // /* the following range collapsed, remove it */ |
625 | // --i; |
626 | // set->length=length-=2; |
627 | // if(i<length) { |
628 | // uprv_memmove(set->array+i, set->array+i+2, (length-i)*4); |
629 | // } |
630 | // } |
631 | // } else { |
632 | // /* extend the previous range (had limit 0x10ffff) to the end of Unicode */ |
633 | // set->length=i-1; |
634 | // } |
635 | // return true; |
636 | // } else if(i==length && c==0x10ffff) { |
637 | // /* insert one range limit c */ |
638 | // more=1; |
639 | // } else { |
640 | // /* insert two range limits c, c+1 */ |
641 | // more=2; |
642 | // } |
643 | // |
644 | // /* insert <more> range limits */ |
645 | // if(length+more>set->capacity) { |
646 | // /* reallocate */ |
647 | // int32_t newCapacity=set->capacity+set->capacity/2+USET_GROW_DELTA; |
648 | // UChar32* newArray=(UChar32* )uprv_malloc(newCapacity*4); |
649 | // if(newArray==nullptr) { |
650 | // return false; |
651 | // } |
652 | // set->capacity=newCapacity; |
653 | // uprv_memcpy(newArray, set->array, length*4); |
654 | // |
655 | // if(set->array!=set->staticBuffer) { |
656 | // uprv_free(set->array); |
657 | // } |
658 | // set->array=newArray; |
659 | // } |
660 | // |
661 | // if(i<length) { |
662 | // uprv_memmove(set->array+i+more, set->array+i, (length-i)*4); |
663 | // } |
664 | // set->array[i]=c; |
665 | // if(more==2) { |
666 | // set->array[i+1]=c+1; |
667 | // } |
668 | // set->length+=more; |
669 | // |
670 | // return true; |
671 | // } |
672 | // |
673 | // U_CAPI UBool U_EXPORT2 |
674 | // uset_add(USet* set, UChar32 c) { |
675 | // return addRemove(set, c, 0); |
676 | // } |
677 | // |
678 | // U_CAPI void U_EXPORT2 |
679 | // uset_remove(USet* set, UChar32 c) { |
680 | // addRemove(set, c, 1); |
681 | // } |
682 | |