1.1 --- /dev/null Thu Jan 01 00:00:00 1970 +0000
1.2 +++ b/intl/icu/source/common/uset.cpp Wed Dec 31 06:09:35 2014 +0100
1.3 @@ -0,0 +1,639 @@
1.4 +/*
1.5 +*******************************************************************************
1.6 +*
1.7 +* Copyright (C) 2002-2011, International Business Machines
1.8 +* Corporation and others. All Rights Reserved.
1.9 +*
1.10 +*******************************************************************************
1.11 +* file name: uset.cpp
1.12 +* encoding: US-ASCII
1.13 +* tab size: 8 (not used)
1.14 +* indentation:4
1.15 +*
1.16 +* created on: 2002mar07
1.17 +* created by: Markus W. Scherer
1.18 +*
1.19 +* There are functions to efficiently serialize a USet into an array of uint16_t
1.20 +* and functions to use such a serialized form efficiently without
1.21 +* instantiating a new USet.
1.22 +*/
1.23 +
1.24 +#include "unicode/utypes.h"
1.25 +#include "unicode/uobject.h"
1.26 +#include "unicode/uset.h"
1.27 +#include "unicode/uniset.h"
1.28 +#include "cmemory.h"
1.29 +#include "unicode/ustring.h"
1.30 +#include "unicode/parsepos.h"
1.31 +
1.32 +U_NAMESPACE_USE
1.33 +
1.34 +U_CAPI USet* U_EXPORT2
1.35 +uset_openEmpty() {
1.36 + return (USet*) new UnicodeSet();
1.37 +}
1.38 +
1.39 +U_CAPI USet* U_EXPORT2
1.40 +uset_open(UChar32 start, UChar32 end) {
1.41 + return (USet*) new UnicodeSet(start, end);
1.42 +}
1.43 +
1.44 +U_CAPI void U_EXPORT2
1.45 +uset_close(USet* set) {
1.46 + delete (UnicodeSet*) set;
1.47 +}
1.48 +
1.49 +U_CAPI USet * U_EXPORT2
1.50 +uset_clone(const USet *set) {
1.51 + return (USet*) (((UnicodeSet*) set)->UnicodeSet::clone());
1.52 +}
1.53 +
1.54 +U_CAPI UBool U_EXPORT2
1.55 +uset_isFrozen(const USet *set) {
1.56 + return ((UnicodeSet*) set)->UnicodeSet::isFrozen();
1.57 +}
1.58 +
1.59 +U_CAPI void U_EXPORT2
1.60 +uset_freeze(USet *set) {
1.61 + ((UnicodeSet*) set)->UnicodeSet::freeze();
1.62 +}
1.63 +
1.64 +U_CAPI USet * U_EXPORT2
1.65 +uset_cloneAsThawed(const USet *set) {
1.66 + return (USet*) (((UnicodeSet*) set)->UnicodeSet::cloneAsThawed());
1.67 +}
1.68 +
1.69 +U_CAPI void U_EXPORT2
1.70 +uset_set(USet* set,
1.71 + UChar32 start, UChar32 end) {
1.72 + ((UnicodeSet*) set)->UnicodeSet::set(start, end);
1.73 +}
1.74 +
1.75 +U_CAPI void U_EXPORT2
1.76 +uset_addAll(USet* set, const USet *additionalSet) {
1.77 + ((UnicodeSet*) set)->UnicodeSet::addAll(*((const UnicodeSet*)additionalSet));
1.78 +}
1.79 +
1.80 +U_CAPI void U_EXPORT2
1.81 +uset_add(USet* set, UChar32 c) {
1.82 + ((UnicodeSet*) set)->UnicodeSet::add(c);
1.83 +}
1.84 +
1.85 +U_CAPI void U_EXPORT2
1.86 +uset_addRange(USet* set, UChar32 start, UChar32 end) {
1.87 + ((UnicodeSet*) set)->UnicodeSet::add(start, end);
1.88 +}
1.89 +
1.90 +U_CAPI void U_EXPORT2
1.91 +uset_addString(USet* set, const UChar* str, int32_t strLen) {
1.92 + // UnicodeString handles -1 for strLen
1.93 + UnicodeString s(strLen<0, str, strLen);
1.94 + ((UnicodeSet*) set)->UnicodeSet::add(s);
1.95 +}
1.96 +
1.97 +U_CAPI void U_EXPORT2
1.98 +uset_addAllCodePoints(USet* set, const UChar *str, int32_t strLen) {
1.99 + // UnicodeString handles -1 for strLen
1.100 + UnicodeString s(str, strLen);
1.101 + ((UnicodeSet*) set)->UnicodeSet::addAll(s);
1.102 +}
1.103 +
1.104 +U_CAPI void U_EXPORT2
1.105 +uset_remove(USet* set, UChar32 c) {
1.106 + ((UnicodeSet*) set)->UnicodeSet::remove(c);
1.107 +}
1.108 +
1.109 +U_CAPI void U_EXPORT2
1.110 +uset_removeRange(USet* set, UChar32 start, UChar32 end) {
1.111 + ((UnicodeSet*) set)->UnicodeSet::remove(start, end);
1.112 +}
1.113 +
1.114 +U_CAPI void U_EXPORT2
1.115 +uset_removeString(USet* set, const UChar* str, int32_t strLen) {
1.116 + UnicodeString s(strLen==-1, str, strLen);
1.117 + ((UnicodeSet*) set)->UnicodeSet::remove(s);
1.118 +}
1.119 +
1.120 +U_CAPI void U_EXPORT2
1.121 +uset_removeAll(USet* set, const USet* remove) {
1.122 + ((UnicodeSet*) set)->UnicodeSet::removeAll(*(const UnicodeSet*)remove);
1.123 +}
1.124 +
1.125 +U_CAPI void U_EXPORT2
1.126 +uset_retain(USet* set, UChar32 start, UChar32 end) {
1.127 + ((UnicodeSet*) set)->UnicodeSet::retain(start, end);
1.128 +}
1.129 +
1.130 +U_CAPI void U_EXPORT2
1.131 +uset_retainAll(USet* set, const USet* retain) {
1.132 + ((UnicodeSet*) set)->UnicodeSet::retainAll(*(const UnicodeSet*)retain);
1.133 +}
1.134 +
1.135 +U_CAPI void U_EXPORT2
1.136 +uset_compact(USet* set) {
1.137 + ((UnicodeSet*) set)->UnicodeSet::compact();
1.138 +}
1.139 +
1.140 +U_CAPI void U_EXPORT2
1.141 +uset_complement(USet* set) {
1.142 + ((UnicodeSet*) set)->UnicodeSet::complement();
1.143 +}
1.144 +
1.145 +U_CAPI void U_EXPORT2
1.146 +uset_complementAll(USet* set, const USet* complement) {
1.147 + ((UnicodeSet*) set)->UnicodeSet::complementAll(*(const UnicodeSet*)complement);
1.148 +}
1.149 +
1.150 +U_CAPI void U_EXPORT2
1.151 +uset_clear(USet* set) {
1.152 + ((UnicodeSet*) set)->UnicodeSet::clear();
1.153 +}
1.154 +
1.155 +U_CAPI void U_EXPORT2
1.156 +uset_removeAllStrings(USet* set) {
1.157 + ((UnicodeSet*) set)->UnicodeSet::removeAllStrings();
1.158 +}
1.159 +
1.160 +U_CAPI UBool U_EXPORT2
1.161 +uset_isEmpty(const USet* set) {
1.162 + return ((const UnicodeSet*) set)->UnicodeSet::isEmpty();
1.163 +}
1.164 +
1.165 +U_CAPI UBool U_EXPORT2
1.166 +uset_contains(const USet* set, UChar32 c) {
1.167 + return ((const UnicodeSet*) set)->UnicodeSet::contains(c);
1.168 +}
1.169 +
1.170 +U_CAPI UBool U_EXPORT2
1.171 +uset_containsRange(const USet* set, UChar32 start, UChar32 end) {
1.172 + return ((const UnicodeSet*) set)->UnicodeSet::contains(start, end);
1.173 +}
1.174 +
1.175 +U_CAPI UBool U_EXPORT2
1.176 +uset_containsString(const USet* set, const UChar* str, int32_t strLen) {
1.177 + UnicodeString s(strLen==-1, str, strLen);
1.178 + return ((const UnicodeSet*) set)->UnicodeSet::contains(s);
1.179 +}
1.180 +
1.181 +U_CAPI UBool U_EXPORT2
1.182 +uset_containsAll(const USet* set1, const USet* set2) {
1.183 + return ((const UnicodeSet*) set1)->UnicodeSet::containsAll(* (const UnicodeSet*) set2);
1.184 +}
1.185 +
1.186 +U_CAPI UBool U_EXPORT2
1.187 +uset_containsAllCodePoints(const USet* set, const UChar *str, int32_t strLen) {
1.188 + // Create a string alias, since nothing is being added to the set.
1.189 + UnicodeString s(strLen==-1, str, strLen);
1.190 + return ((const UnicodeSet*) set)->UnicodeSet::containsAll(s);
1.191 +}
1.192 +
1.193 +U_CAPI UBool U_EXPORT2
1.194 +uset_containsNone(const USet* set1, const USet* set2) {
1.195 + return ((const UnicodeSet*) set1)->UnicodeSet::containsNone(* (const UnicodeSet*) set2);
1.196 +}
1.197 +
1.198 +U_CAPI UBool U_EXPORT2
1.199 +uset_containsSome(const USet* set1, const USet* set2) {
1.200 + return ((const UnicodeSet*) set1)->UnicodeSet::containsSome(* (const UnicodeSet*) set2);
1.201 +}
1.202 +
1.203 +U_CAPI int32_t U_EXPORT2
1.204 +uset_span(const USet *set, const UChar *s, int32_t length, USetSpanCondition spanCondition) {
1.205 + return ((UnicodeSet*) set)->UnicodeSet::span(s, length, spanCondition);
1.206 +}
1.207 +
1.208 +U_CAPI int32_t U_EXPORT2
1.209 +uset_spanBack(const USet *set, const UChar *s, int32_t length, USetSpanCondition spanCondition) {
1.210 + return ((UnicodeSet*) set)->UnicodeSet::spanBack(s, length, spanCondition);
1.211 +}
1.212 +
1.213 +U_CAPI int32_t U_EXPORT2
1.214 +uset_spanUTF8(const USet *set, const char *s, int32_t length, USetSpanCondition spanCondition) {
1.215 + return ((UnicodeSet*) set)->UnicodeSet::spanUTF8(s, length, spanCondition);
1.216 +}
1.217 +
1.218 +U_CAPI int32_t U_EXPORT2
1.219 +uset_spanBackUTF8(const USet *set, const char *s, int32_t length, USetSpanCondition spanCondition) {
1.220 + return ((UnicodeSet*) set)->UnicodeSet::spanBackUTF8(s, length, spanCondition);
1.221 +}
1.222 +
1.223 +U_CAPI UBool U_EXPORT2
1.224 +uset_equals(const USet* set1, const USet* set2) {
1.225 + return *(const UnicodeSet*)set1 == *(const UnicodeSet*)set2;
1.226 +}
1.227 +
1.228 +U_CAPI int32_t U_EXPORT2
1.229 +uset_indexOf(const USet* set, UChar32 c) {
1.230 + return ((UnicodeSet*) set)->UnicodeSet::indexOf(c);
1.231 +}
1.232 +
1.233 +U_CAPI UChar32 U_EXPORT2
1.234 +uset_charAt(const USet* set, int32_t index) {
1.235 + return ((UnicodeSet*) set)->UnicodeSet::charAt(index);
1.236 +}
1.237 +
1.238 +U_CAPI int32_t U_EXPORT2
1.239 +uset_size(const USet* set) {
1.240 + return ((const UnicodeSet*) set)->UnicodeSet::size();
1.241 +}
1.242 +
1.243 +U_NAMESPACE_BEGIN
1.244 +/**
1.245 + * This class only exists to provide access to the UnicodeSet private
1.246 + * USet support API. Declaring a class a friend is more portable than
1.247 + * trying to declare extern "C" functions as friends.
1.248 + */
1.249 +class USetAccess /* not : public UObject because all methods are static */ {
1.250 +public:
1.251 + /* Try to have the compiler inline these*/
1.252 + inline static int32_t getStringCount(const UnicodeSet& set) {
1.253 + return set.getStringCount();
1.254 + }
1.255 + inline static const UnicodeString* getString(const UnicodeSet& set,
1.256 + int32_t i) {
1.257 + return set.getString(i);
1.258 + }
1.259 +private:
1.260 + /* do not instantiate*/
1.261 + USetAccess();
1.262 +};
1.263 +U_NAMESPACE_END
1.264 +
1.265 +U_CAPI int32_t U_EXPORT2
1.266 +uset_getItemCount(const USet* uset) {
1.267 + const UnicodeSet& set = *(const UnicodeSet*)uset;
1.268 + return set.getRangeCount() + USetAccess::getStringCount(set);
1.269 +}
1.270 +
1.271 +U_CAPI int32_t U_EXPORT2
1.272 +uset_getItem(const USet* uset, int32_t itemIndex,
1.273 + UChar32* start, UChar32* end,
1.274 + UChar* str, int32_t strCapacity,
1.275 + UErrorCode* ec) {
1.276 + if (U_FAILURE(*ec)) return 0;
1.277 + const UnicodeSet& set = *(const UnicodeSet*)uset;
1.278 + int32_t rangeCount;
1.279 +
1.280 + if (itemIndex < 0) {
1.281 + *ec = U_ILLEGAL_ARGUMENT_ERROR;
1.282 + return -1;
1.283 + } else if (itemIndex < (rangeCount = set.getRangeCount())) {
1.284 + *start = set.getRangeStart(itemIndex);
1.285 + *end = set.getRangeEnd(itemIndex);
1.286 + return 0;
1.287 + } else {
1.288 + itemIndex -= rangeCount;
1.289 + if (itemIndex < USetAccess::getStringCount(set)) {
1.290 + const UnicodeString* s = USetAccess::getString(set, itemIndex);
1.291 + return s->extract(str, strCapacity, *ec);
1.292 + } else {
1.293 + *ec = U_INDEX_OUTOFBOUNDS_ERROR;
1.294 + return -1;
1.295 + }
1.296 + }
1.297 +}
1.298 +
1.299 +//U_CAPI int32_t U_EXPORT2
1.300 +//uset_getRangeCount(const USet* set) {
1.301 +// return ((const UnicodeSet*) set)->getRangeCount();
1.302 +//}
1.303 +//
1.304 +//U_CAPI UBool U_EXPORT2
1.305 +//uset_getRange(const USet* set, int32_t rangeIndex,
1.306 +// UChar32* pStart, UChar32* pEnd) {
1.307 +// if ((uint32_t) rangeIndex >= (uint32_t) uset_getRangeCount(set)) {
1.308 +// return FALSE;
1.309 +// }
1.310 +// const UnicodeSet* us = (const UnicodeSet*) set;
1.311 +// *pStart = us->getRangeStart(rangeIndex);
1.312 +// *pEnd = us->getRangeEnd(rangeIndex);
1.313 +// return TRUE;
1.314 +//}
1.315 +
1.316 +/*
1.317 + * Serialize a USet into 16-bit units.
1.318 + * Store BMP code points as themselves with one 16-bit unit each.
1.319 + *
1.320 + * Important: the code points in the array are in ascending order,
1.321 + * therefore all BMP code points precede all supplementary code points.
1.322 + *
1.323 + * Store each supplementary code point in 2 16-bit units,
1.324 + * simply with higher-then-lower 16-bit halfs.
1.325 + *
1.326 + * Precede the entire list with the length.
1.327 + * If there are supplementary code points, then set bit 15 in the length
1.328 + * and add the bmpLength between it and the array.
1.329 + *
1.330 + * In other words:
1.331 + * - all BMP: (length=bmpLength) BMP, .., BMP
1.332 + * - some supplementary: (length|0x8000) (bmpLength<length) BMP, .., BMP, supp-high, supp-low, ..
1.333 + */
1.334 +U_CAPI int32_t U_EXPORT2
1.335 +uset_serialize(const USet* set, uint16_t* dest, int32_t destCapacity, UErrorCode* ec) {
1.336 + if (ec==NULL || U_FAILURE(*ec)) {
1.337 + return 0;
1.338 + }
1.339 +
1.340 + return ((const UnicodeSet*) set)->UnicodeSet::serialize(dest, destCapacity,* ec);
1.341 +}
1.342 +
1.343 +U_CAPI UBool U_EXPORT2
1.344 +uset_getSerializedSet(USerializedSet* fillSet, const uint16_t* src, int32_t srcLength) {
1.345 + int32_t length;
1.346 +
1.347 + if(fillSet==NULL) {
1.348 + return FALSE;
1.349 + }
1.350 + if(src==NULL || srcLength<=0) {
1.351 + fillSet->length=fillSet->bmpLength=0;
1.352 + return FALSE;
1.353 + }
1.354 +
1.355 + length=*src++;
1.356 + if(length&0x8000) {
1.357 + /* there are supplementary values */
1.358 + length&=0x7fff;
1.359 + if(srcLength<(2+length)) {
1.360 + fillSet->length=fillSet->bmpLength=0;
1.361 + return FALSE;
1.362 + }
1.363 + fillSet->bmpLength=*src++;
1.364 + } else {
1.365 + /* only BMP values */
1.366 + if(srcLength<(1+length)) {
1.367 + fillSet->length=fillSet->bmpLength=0;
1.368 + return FALSE;
1.369 + }
1.370 + fillSet->bmpLength=length;
1.371 + }
1.372 + fillSet->array=src;
1.373 + fillSet->length=length;
1.374 + return TRUE;
1.375 +}
1.376 +
1.377 +U_CAPI void U_EXPORT2
1.378 +uset_setSerializedToOne(USerializedSet* fillSet, UChar32 c) {
1.379 + if(fillSet==NULL || (uint32_t)c>0x10ffff) {
1.380 + return;
1.381 + }
1.382 +
1.383 + fillSet->array=fillSet->staticArray;
1.384 + if(c<0xffff) {
1.385 + fillSet->bmpLength=fillSet->length=2;
1.386 + fillSet->staticArray[0]=(uint16_t)c;
1.387 + fillSet->staticArray[1]=(uint16_t)c+1;
1.388 + } else if(c==0xffff) {
1.389 + fillSet->bmpLength=1;
1.390 + fillSet->length=3;
1.391 + fillSet->staticArray[0]=0xffff;
1.392 + fillSet->staticArray[1]=1;
1.393 + fillSet->staticArray[2]=0;
1.394 + } else if(c<0x10ffff) {
1.395 + fillSet->bmpLength=0;
1.396 + fillSet->length=4;
1.397 + fillSet->staticArray[0]=(uint16_t)(c>>16);
1.398 + fillSet->staticArray[1]=(uint16_t)c;
1.399 + ++c;
1.400 + fillSet->staticArray[2]=(uint16_t)(c>>16);
1.401 + fillSet->staticArray[3]=(uint16_t)c;
1.402 + } else /* c==0x10ffff */ {
1.403 + fillSet->bmpLength=0;
1.404 + fillSet->length=2;
1.405 + fillSet->staticArray[0]=0x10;
1.406 + fillSet->staticArray[1]=0xffff;
1.407 + }
1.408 +}
1.409 +
1.410 +U_CAPI UBool U_EXPORT2
1.411 +uset_serializedContains(const USerializedSet* set, UChar32 c) {
1.412 + const uint16_t* array;
1.413 +
1.414 + if(set==NULL || (uint32_t)c>0x10ffff) {
1.415 + return FALSE;
1.416 + }
1.417 +
1.418 + array=set->array;
1.419 + if(c<=0xffff) {
1.420 + /* find c in the BMP part */
1.421 + int32_t lo = 0;
1.422 + int32_t hi = set->bmpLength-1;
1.423 + if (c < array[0]) {
1.424 + hi = 0;
1.425 + } else if (c < array[hi]) {
1.426 + for(;;) {
1.427 + int32_t i = (lo + hi) >> 1;
1.428 + if (i == lo) {
1.429 + break; // Done!
1.430 + } else if (c < array[i]) {
1.431 + hi = i;
1.432 + } else {
1.433 + lo = i;
1.434 + }
1.435 + }
1.436 + } else {
1.437 + hi += 1;
1.438 + }
1.439 + return (UBool)(hi&1);
1.440 + } else {
1.441 + /* find c in the supplementary part */
1.442 + uint16_t high=(uint16_t)(c>>16), low=(uint16_t)c;
1.443 + int32_t base = set->bmpLength;
1.444 + int32_t lo = 0;
1.445 + int32_t hi = set->length - 2 - base;
1.446 + if (high < array[base] || (high==array[base] && low<array[base+1])) {
1.447 + hi = 0;
1.448 + } else if (high < array[base+hi] || (high==array[base+hi] && low<array[base+hi+1])) {
1.449 + for (;;) {
1.450 + int32_t i = ((lo + hi) >> 1) & ~1; // Guarantee even result
1.451 + int32_t iabs = i + base;
1.452 + if (i == lo) {
1.453 + break; // Done!
1.454 + } else if (high < array[iabs] || (high==array[iabs] && low<array[iabs+1])) {
1.455 + hi = i;
1.456 + } else {
1.457 + lo = i;
1.458 + }
1.459 + }
1.460 + } else {
1.461 + hi += 2;
1.462 + }
1.463 + /* count pairs of 16-bit units even per BMP and check if the number of pairs is odd */
1.464 + return (UBool)(((hi+(base<<1))&2)!=0);
1.465 + }
1.466 +}
1.467 +
1.468 +U_CAPI int32_t U_EXPORT2
1.469 +uset_getSerializedRangeCount(const USerializedSet* set) {
1.470 + if(set==NULL) {
1.471 + return 0;
1.472 + }
1.473 +
1.474 + return (set->bmpLength+(set->length-set->bmpLength)/2+1)/2;
1.475 +}
1.476 +
1.477 +U_CAPI UBool U_EXPORT2
1.478 +uset_getSerializedRange(const USerializedSet* set, int32_t rangeIndex,
1.479 + UChar32* pStart, UChar32* pEnd) {
1.480 + const uint16_t* array;
1.481 + int32_t bmpLength, length;
1.482 +
1.483 + if(set==NULL || rangeIndex<0 || pStart==NULL || pEnd==NULL) {
1.484 + return FALSE;
1.485 + }
1.486 +
1.487 + array=set->array;
1.488 + length=set->length;
1.489 + bmpLength=set->bmpLength;
1.490 +
1.491 + rangeIndex*=2; /* address start/limit pairs */
1.492 + if(rangeIndex<bmpLength) {
1.493 + *pStart=array[rangeIndex++];
1.494 + if(rangeIndex<bmpLength) {
1.495 + *pEnd=array[rangeIndex]-1;
1.496 + } else if(rangeIndex<length) {
1.497 + *pEnd=((((int32_t)array[rangeIndex])<<16)|array[rangeIndex+1])-1;
1.498 + } else {
1.499 + *pEnd=0x10ffff;
1.500 + }
1.501 + return TRUE;
1.502 + } else {
1.503 + rangeIndex-=bmpLength;
1.504 + rangeIndex*=2; /* address pairs of pairs of units */
1.505 + length-=bmpLength;
1.506 + if(rangeIndex<length) {
1.507 + array+=bmpLength;
1.508 + *pStart=(((int32_t)array[rangeIndex])<<16)|array[rangeIndex+1];
1.509 + rangeIndex+=2;
1.510 + if(rangeIndex<length) {
1.511 + *pEnd=((((int32_t)array[rangeIndex])<<16)|array[rangeIndex+1])-1;
1.512 + } else {
1.513 + *pEnd=0x10ffff;
1.514 + }
1.515 + return TRUE;
1.516 + } else {
1.517 + return FALSE;
1.518 + }
1.519 + }
1.520 +}
1.521 +
1.522 +// TODO The old, internal uset.c had an efficient uset_containsOne function.
1.523 +// Returned the one and only code point, or else -1 or something.
1.524 +// Consider adding such a function to both C and C++ UnicodeSet/uset.
1.525 +// See tools/gennorm/store.c for usage, now usetContainsOne there.
1.526 +
1.527 +// TODO Investigate incorporating this code into UnicodeSet to improve
1.528 +// efficiency.
1.529 +// ---
1.530 +// #define USET_GROW_DELTA 20
1.531 +//
1.532 +// static int32_t
1.533 +// findChar(const UChar32* array, int32_t length, UChar32 c) {
1.534 +// int32_t i;
1.535 +//
1.536 +// /* check the last range limit first for more efficient appending */
1.537 +// if(length>0) {
1.538 +// if(c>=array[length-1]) {
1.539 +// return length;
1.540 +// }
1.541 +//
1.542 +// /* do not check the last range limit again in the loop below */
1.543 +// --length;
1.544 +// }
1.545 +//
1.546 +// for(i=0; i<length && c>=array[i]; ++i) {}
1.547 +// return i;
1.548 +// }
1.549 +//
1.550 +// static UBool
1.551 +// addRemove(USet* set, UChar32 c, int32_t doRemove) {
1.552 +// int32_t i, length, more;
1.553 +//
1.554 +// if(set==NULL || (uint32_t)c>0x10ffff) {
1.555 +// return FALSE;
1.556 +// }
1.557 +//
1.558 +// length=set->length;
1.559 +// i=findChar(set->array, length, c);
1.560 +// if((i&1)^doRemove) {
1.561 +// /* c is already in the set */
1.562 +// return TRUE;
1.563 +// }
1.564 +//
1.565 +// /* how many more array items do we need? */
1.566 +// if(i<length && (c+1)==set->array[i]) {
1.567 +// /* c is just before the following range, extend that in-place by one */
1.568 +// set->array[i]=c;
1.569 +// if(i>0) {
1.570 +// --i;
1.571 +// if(c==set->array[i]) {
1.572 +// /* the previous range collapsed, remove it */
1.573 +// set->length=length-=2;
1.574 +// if(i<length) {
1.575 +// uprv_memmove(set->array+i, set->array+i+2, (length-i)*4);
1.576 +// }
1.577 +// }
1.578 +// }
1.579 +// return TRUE;
1.580 +// } else if(i>0 && c==set->array[i-1]) {
1.581 +// /* c is just after the previous range, extend that in-place by one */
1.582 +// if(++c<=0x10ffff) {
1.583 +// set->array[i-1]=c;
1.584 +// if(i<length && c==set->array[i]) {
1.585 +// /* the following range collapsed, remove it */
1.586 +// --i;
1.587 +// set->length=length-=2;
1.588 +// if(i<length) {
1.589 +// uprv_memmove(set->array+i, set->array+i+2, (length-i)*4);
1.590 +// }
1.591 +// }
1.592 +// } else {
1.593 +// /* extend the previous range (had limit 0x10ffff) to the end of Unicode */
1.594 +// set->length=i-1;
1.595 +// }
1.596 +// return TRUE;
1.597 +// } else if(i==length && c==0x10ffff) {
1.598 +// /* insert one range limit c */
1.599 +// more=1;
1.600 +// } else {
1.601 +// /* insert two range limits c, c+1 */
1.602 +// more=2;
1.603 +// }
1.604 +//
1.605 +// /* insert <more> range limits */
1.606 +// if(length+more>set->capacity) {
1.607 +// /* reallocate */
1.608 +// int32_t newCapacity=set->capacity+set->capacity/2+USET_GROW_DELTA;
1.609 +// UChar32* newArray=(UChar32* )uprv_malloc(newCapacity*4);
1.610 +// if(newArray==NULL) {
1.611 +// return FALSE;
1.612 +// }
1.613 +// set->capacity=newCapacity;
1.614 +// uprv_memcpy(newArray, set->array, length*4);
1.615 +//
1.616 +// if(set->array!=set->staticBuffer) {
1.617 +// uprv_free(set->array);
1.618 +// }
1.619 +// set->array=newArray;
1.620 +// }
1.621 +//
1.622 +// if(i<length) {
1.623 +// uprv_memmove(set->array+i+more, set->array+i, (length-i)*4);
1.624 +// }
1.625 +// set->array[i]=c;
1.626 +// if(more==2) {
1.627 +// set->array[i+1]=c+1;
1.628 +// }
1.629 +// set->length+=more;
1.630 +//
1.631 +// return TRUE;
1.632 +// }
1.633 +//
1.634 +// U_CAPI UBool U_EXPORT2
1.635 +// uset_add(USet* set, UChar32 c) {
1.636 +// return addRemove(set, c, 0);
1.637 +// }
1.638 +//
1.639 +// U_CAPI void U_EXPORT2
1.640 +// uset_remove(USet* set, UChar32 c) {
1.641 +// addRemove(set, c, 1);
1.642 +// }
