1/******************************************************************** 2 * Copyright (c) 1997-2013, International Business Machines 3 * Corporation and others. All Rights Reserved. 4 ********************************************************************/ 5/***************************************************************************** 6* 7* File CAPITEST.C 8* 9* Modification History: 10* Name Description 11* Madhu Katragadda Ported for C API 12* Brian Rower Added TestOpenVsOpenRules 13****************************************************************************** 14*//* C API TEST For COLLATOR */ 15 16#include "unicode/utypes.h" 17 18#if !UCONFIG_NO_COLLATION 19 20#include <stdio.h> 21#include <stdlib.h> 22#include <string.h> 23#include "unicode/uloc.h" 24#include "unicode/ulocdata.h" 25#include "unicode/ustring.h" 26#include "unicode/ures.h" 27#include "unicode/ucoleitr.h" 28#include "cintltst.h" 29#include "capitst.h" 30#include "ccolltst.h" 31#include "putilimp.h" 32#include "cmemory.h" 33#include "cstring.h" 34 35static void TestAttribute(void); 36static void TestDefault(void); 37static void TestDefaultKeyword(void); 38static void TestBengaliSortKey(void); 39 int TestBufferSize(); /* defined in "colutil.c" */ 40 41 42static char* U_EXPORT2 ucol_sortKeyToString(const UCollator *coll, const uint8_t *sortkey, char *buffer, uint32_t len) { 43 uint32_t position = 0; 44 uint8_t b; 45 46 if (position + 1 < len) 47 position += sprintf(buffer + position, "["); 48 while ((b = *sortkey++) != 0) { 49 if (b == 1 && position + 5 < len) { 50 position += sprintf(buffer + position, "%02X . ", b); 51 } else if (b != 1 && position + 3 < len) { 52 position += sprintf(buffer + position, "%02X ", b); 53 } 54 } 55 if (position + 3 < len) 56 position += sprintf(buffer + position, "%02X]", b); 57 return buffer; 58} 59 60void addCollAPITest(TestNode** root) 61{ 62 /* WEIVTODO: return tests here */ 63 addTest(root, &TestProperty, "tscoll/capitst/TestProperty"); 64 addTest(root, &TestRuleBasedColl, "tscoll/capitst/TestRuleBasedColl"); 65 addTest(root, &TestCompare, "tscoll/capitst/TestCompare"); 66 addTest(root, &TestSortKey, "tscoll/capitst/TestSortKey"); 67 addTest(root, &TestHashCode, "tscoll/capitst/TestHashCode"); 68 addTest(root, &TestElemIter, "tscoll/capitst/TestElemIter"); 69 addTest(root, &TestGetAll, "tscoll/capitst/TestGetAll"); 70 /*addTest(root, &TestGetDefaultRules, "tscoll/capitst/TestGetDefaultRules");*/ 71 addTest(root, &TestDecomposition, "tscoll/capitst/TestDecomposition"); 72 addTest(root, &TestSafeClone, "tscoll/capitst/TestSafeClone"); 73 addTest(root, &TestCloneBinary, "tscoll/capitst/TestCloneBinary"); 74 addTest(root, &TestGetSetAttr, "tscoll/capitst/TestGetSetAttr"); 75 addTest(root, &TestBounds, "tscoll/capitst/TestBounds"); 76 addTest(root, &TestGetLocale, "tscoll/capitst/TestGetLocale"); 77 addTest(root, &TestSortKeyBufferOverrun, "tscoll/capitst/TestSortKeyBufferOverrun"); 78 addTest(root, &TestAttribute, "tscoll/capitst/TestAttribute"); 79 addTest(root, &TestGetTailoredSet, "tscoll/capitst/TestGetTailoredSet"); 80 addTest(root, &TestMergeSortKeys, "tscoll/capitst/TestMergeSortKeys"); 81 addTest(root, &TestShortString, "tscoll/capitst/TestShortString"); 82 /* BEGIN android-removed 83 To save space, Android does not include the collation tailoring rules. 84 We skip the tailing tests for collations. */ 85 /* addTest(root, &TestGetContractionsAndUnsafes, "tscoll/capitst/TestGetContractionsAndUnsafes"); */ 86 /* END android-removed */ 87 addTest(root, &TestOpenBinary, "tscoll/capitst/TestOpenBinary"); 88 addTest(root, &TestDefault, "tscoll/capitst/TestDefault"); 89 addTest(root, &TestDefaultKeyword, "tscoll/capitst/TestDefaultKeyword"); 90 /* BEGIN android-removed 91 To save space, Android does not build full collation tables and tailing rules. 92 We skip the related tests. */ 93 /* addTest(root, &TestOpenVsOpenRules, "tscoll/capitst/TestOpenVsOpenRules"); */ 94 /* addTest(root, &TestGetKeywordValuesForLocale, "tscoll/capitst/TestGetKeywordValuesForLocale"); */ 95 /* END android-removed */ 96 addTest(root, &TestBengaliSortKey, "tscoll/capitst/TestBengaliSortKey"); 97 addTest(root, &TestGetKeywordValuesForLocale, "tscoll/capitst/TestGetKeywordValuesForLocale"); 98} 99 100void TestGetSetAttr(void) { 101 UErrorCode status = U_ZERO_ERROR; 102 UCollator *coll = ucol_open(NULL, &status); 103 struct attrTest { 104 UColAttribute att; 105 UColAttributeValue val[5]; 106 uint32_t valueSize; 107 UColAttributeValue nonValue; 108 } attrs[] = { 109 {UCOL_FRENCH_COLLATION, {UCOL_ON, UCOL_OFF}, 2, UCOL_SHIFTED}, 110 {UCOL_ALTERNATE_HANDLING, {UCOL_NON_IGNORABLE, UCOL_SHIFTED}, 2, UCOL_OFF},/* attribute for handling variable elements*/ 111 {UCOL_CASE_FIRST, {UCOL_OFF, UCOL_LOWER_FIRST, UCOL_UPPER_FIRST}, 3, UCOL_SHIFTED},/* who goes first, lower case or uppercase */ 112 {UCOL_CASE_LEVEL, {UCOL_ON, UCOL_OFF}, 2, UCOL_SHIFTED},/* do we have an extra case level */ 113 {UCOL_NORMALIZATION_MODE, {UCOL_ON, UCOL_OFF}, 2, UCOL_SHIFTED},/* attribute for normalization */ 114 {UCOL_DECOMPOSITION_MODE, {UCOL_ON, UCOL_OFF}, 2, UCOL_SHIFTED}, 115 {UCOL_STRENGTH, {UCOL_PRIMARY, UCOL_SECONDARY, UCOL_TERTIARY, UCOL_QUATERNARY, UCOL_IDENTICAL}, 5, UCOL_SHIFTED},/* attribute for strength */ 116 {UCOL_HIRAGANA_QUATERNARY_MODE, {UCOL_ON, UCOL_OFF}, 2, UCOL_SHIFTED},/* when turned on, this attribute */ 117 }; 118 UColAttribute currAttr; 119 UColAttributeValue value; 120 uint32_t i = 0, j = 0; 121 122 if (coll == NULL) { 123 log_err_status(status, "Unable to open collator. %s\n", u_errorName(status)); 124 return; 125 } 126 for(i = 0; i<sizeof(attrs)/sizeof(attrs[0]); i++) { 127 currAttr = attrs[i].att; 128 ucol_setAttribute(coll, currAttr, UCOL_DEFAULT, &status); 129 if(U_FAILURE(status)) { 130 log_err_status(status, "ucol_setAttribute with the default value returned error: %s\n", u_errorName(status)); 131 break; 132 } 133 value = ucol_getAttribute(coll, currAttr, &status); 134 if(U_FAILURE(status)) { 135 log_err("ucol_getAttribute returned error: %s\n", u_errorName(status)); 136 break; 137 } 138 for(j = 0; j<attrs[i].valueSize; j++) { 139 ucol_setAttribute(coll, currAttr, attrs[i].val[j], &status); 140 if(U_FAILURE(status)) { 141 log_err("ucol_setAttribute with the value %i returned error: %s\n", attrs[i].val[j], u_errorName(status)); 142 break; 143 } 144 } 145 status = U_ZERO_ERROR; 146 ucol_setAttribute(coll, currAttr, attrs[i].nonValue, &status); 147 if(U_SUCCESS(status)) { 148 log_err("ucol_setAttribute with the bad value didn't return an error\n"); 149 break; 150 } 151 status = U_ZERO_ERROR; 152 153 ucol_setAttribute(coll, currAttr, value, &status); 154 if(U_FAILURE(status)) { 155 log_err("ucol_setAttribute with the default valuereturned error: %s\n", u_errorName(status)); 156 break; 157 } 158 } 159 status = U_ZERO_ERROR; 160 value = ucol_getAttribute(coll, UCOL_ATTRIBUTE_COUNT, &status); 161 if(U_SUCCESS(status)) { 162 log_err("ucol_getAttribute for UCOL_ATTRIBUTE_COUNT didn't return an error\n"); 163 } 164 status = U_ZERO_ERROR; 165 ucol_setAttribute(coll, UCOL_ATTRIBUTE_COUNT, UCOL_DEFAULT, &status); 166 if(U_SUCCESS(status)) { 167 log_err("ucol_setAttribute for UCOL_ATTRIBUTE_COUNT didn't return an error\n"); 168 } 169 status = U_ZERO_ERROR; 170 ucol_close(coll); 171} 172 173 174static void doAssert(int condition, const char *message) 175{ 176 if (condition==0) { 177 log_err("ERROR : %s\n", message); 178 } 179} 180 181#define UTF8_BUF_SIZE 128 182 183static void doStrcoll(const UCollator* coll, const UChar* src, int32_t srcLen, const UChar* tgt, int32_t tgtLen, 184 UCollationResult expected, const char *message) { 185 UErrorCode err = U_ZERO_ERROR; 186 char srcU8[UTF8_BUF_SIZE], tgtU8[UTF8_BUF_SIZE]; 187 int32_t srcU8Len = -1, tgtU8Len = -1; 188 int32_t len = 0; 189 190 if (ucol_strcoll(coll, src, srcLen, tgt, tgtLen) != expected) { 191 log_err("ERROR : %s\n", message); 192 } 193 194 u_strToUTF8(srcU8, UTF8_BUF_SIZE, &len, src, srcLen, &err); 195 if (U_FAILURE(err) || len >= UTF8_BUF_SIZE) { 196 log_err("ERROR : UTF-8 conversion error\n"); 197 return; 198 } 199 if (srcLen >= 0) { 200 srcU8Len = len; 201 } 202 u_strToUTF8(tgtU8, UTF8_BUF_SIZE, &len, tgt, tgtLen, &err); 203 if (U_FAILURE(err) || len >= UTF8_BUF_SIZE) { 204 log_err("ERROR : UTF-8 conversion error\n"); 205 return; 206 } 207 if (tgtLen >= 0) { 208 tgtU8Len = len; 209 } 210 211 if (ucol_strcollUTF8(coll, srcU8, srcU8Len, tgtU8, tgtU8Len, &err) != expected 212 || U_FAILURE(err)) { 213 log_err("ERROR: %s (strcollUTF8)\n", message); 214 } 215} 216 217#if 0 218/* We don't have default rules, at least not in the previous sense */ 219void TestGetDefaultRules(){ 220 uint32_t size=0; 221 UErrorCode status=U_ZERO_ERROR; 222 UCollator *coll=NULL; 223 int32_t len1 = 0, len2=0; 224 uint8_t *binColData = NULL; 225 226 UResourceBundle *res = NULL; 227 UResourceBundle *binColl = NULL; 228 uint8_t *binResult = NULL; 229 230 231 const UChar * defaultRulesArray=ucol_getDefaultRulesArray(&size); 232 log_verbose("Test the function ucol_getDefaultRulesArray()\n"); 233 234 coll = ucol_openRules(defaultRulesArray, size, UCOL_ON, UCOL_PRIMARY, &status); 235 if(U_SUCCESS(status) && coll !=NULL) { 236 binColData = (uint8_t*)ucol_cloneRuleData(coll, &len1, &status); 237 238 } 239 240 241 status=U_ZERO_ERROR; 242 res=ures_open(NULL, "root", &status); 243 if(U_FAILURE(status)){ 244 log_err("ERROR: Failed to get resource for \"root Locale\" with %s", myErrorName(status)); 245 return; 246 } 247 binColl=ures_getByKey(res, "%%Collation", binColl, &status); 248 if(U_SUCCESS(status)){ 249 binResult=(uint8_t*)ures_getBinary(binColl, &len2, &status); 250 if(U_FAILURE(status)){ 251 log_err("ERROR: ures_getBinary() failed\n"); 252 } 253 }else{ 254 log_err("ERROR: ures_getByKey(locale(default), %%Collation) failed"); 255 } 256 257 258 if(len1 != len2){ 259 log_err("Error: ucol_getDefaultRulesArray() failed to return the correct length.\n"); 260 } 261 if(memcmp(binColData, binResult, len1) != 0){ 262 log_err("Error: ucol_getDefaultRulesArray() failed\n"); 263 } 264 265 free(binColData); 266 ures_close(binColl); 267 ures_close(res); 268 ucol_close(coll); 269 270} 271#endif 272 273/* Collator Properties 274 ucol_open, ucol_strcoll, getStrength/setStrength 275 getDecomposition/setDecomposition, getDisplayName*/ 276void TestProperty() 277{ 278 UCollator *col, *ruled; 279 UChar *disName; 280 int32_t len = 0; 281 UChar source[12], target[12]; 282 int32_t tempLength; 283 UErrorCode status = U_ZERO_ERROR; 284 /* 285 * Expected version of the English collator. 286 * Currently, the major/minor version numbers change when the builder code 287 * changes, 288 * number 2 is from the tailoring data version and 289 * number 3 is the UCA version. 290 * This changes with every UCA version change, and the expected value 291 * needs to be adjusted. 292 * Same in intltest/apicoll.cpp. 293 */ 294 UVersionInfo currVersionArray = {0x31, 0xC0, 0x05, 0x2A}; /* from ICU 4.4/UCA 5.2 */ 295 UVersionInfo versionArray = {0, 0, 0, 0}; 296 UVersionInfo versionUCAArray = {0, 0, 0, 0}; 297 UVersionInfo versionUCDArray = {0, 0, 0, 0}; 298 299 log_verbose("The property tests begin : \n"); 300 log_verbose("Test ucol_strcoll : \n"); 301 col = ucol_open("en_US", &status); 302 if (U_FAILURE(status)) { 303 log_err_status(status, "Default Collator creation failed.: %s\n", myErrorName(status)); 304 return; 305 } 306 307 ucol_getVersion(col, versionArray); 308 /* Check for a version greater than some value rather than equality 309 * so that we need not update the expected version each time. */ 310 if (uprv_memcmp(versionArray, currVersionArray, 4)<0) { 311 log_err("Testing ucol_getVersion() - unexpected result: %02x.%02x.%02x.%02x\n", 312 versionArray[0], versionArray[1], versionArray[2], versionArray[3]); 313 } else { 314 log_verbose("ucol_getVersion() result: %02x.%02x.%02x.%02x\n", 315 versionArray[0], versionArray[1], versionArray[2], versionArray[3]); 316 } 317 318 /* Assume that the UCD and UCA versions are the same, 319 * rather than hardcoding (and updating each time) a particular UCA version. */ 320 u_getUnicodeVersion(versionUCDArray); 321 ucol_getUCAVersion(col, versionUCAArray); 322 if (0!=uprv_memcmp(versionUCAArray, versionUCDArray, 4)) { 323 log_err("Testing ucol_getUCAVersion() - unexpected result: %hu.%hu.%hu.%hu\n", 324 versionUCAArray[0], versionUCAArray[1], versionUCAArray[2], versionUCAArray[3]); 325 } 326 327 u_uastrcpy(source, "ab"); 328 u_uastrcpy(target, "abc"); 329 330 doStrcoll(col, source, u_strlen(source), target, u_strlen(target), UCOL_LESS, "ab < abc comparison failed"); 331 332 u_uastrcpy(source, "ab"); 333 u_uastrcpy(target, "AB"); 334 335 doStrcoll(col, source, u_strlen(source), target, u_strlen(target), UCOL_LESS, "ab < AB comparison failed"); 336 337 u_uastrcpy(source, "blackbird"); 338 u_uastrcpy(target, "black-bird"); 339 340 doStrcoll(col, source, u_strlen(source), target, u_strlen(target), UCOL_GREATER, "black-bird > blackbird comparison failed"); 341 342 u_uastrcpy(source, "black bird"); 343 u_uastrcpy(target, "black-bird"); 344 345 doStrcoll(col, source, u_strlen(source), target, u_strlen(target), UCOL_LESS, "black bird < black-bird comparison failed"); 346 347 u_uastrcpy(source, "Hello"); 348 u_uastrcpy(target, "hello"); 349 350 doStrcoll(col, source, u_strlen(source), target, u_strlen(target), UCOL_GREATER, "Hello > hello comparison failed"); 351 352 log_verbose("Test ucol_strcoll ends.\n"); 353 354 log_verbose("testing ucol_getStrength() method ...\n"); 355 doAssert( (ucol_getStrength(col) == UCOL_TERTIARY), "collation object has the wrong strength"); 356 doAssert( (ucol_getStrength(col) != UCOL_PRIMARY), "collation object's strength is primary difference"); 357 358 log_verbose("testing ucol_setStrength() method ...\n"); 359 ucol_setStrength(col, UCOL_SECONDARY); 360 doAssert( (ucol_getStrength(col) != UCOL_TERTIARY), "collation object's strength is secondary difference"); 361 doAssert( (ucol_getStrength(col) != UCOL_PRIMARY), "collation object's strength is primary difference"); 362 doAssert( (ucol_getStrength(col) == UCOL_SECONDARY), "collation object has the wrong strength"); 363 364 365 log_verbose("Get display name for the default collation in German : \n"); 366 367 len=ucol_getDisplayName("en_US", "de_DE", NULL, 0, &status); 368 if(status==U_BUFFER_OVERFLOW_ERROR){ 369 status=U_ZERO_ERROR; 370 disName=(UChar*)malloc(sizeof(UChar) * (len+1)); 371 ucol_getDisplayName("en_US", "de_DE", disName, len+1, &status); 372 log_verbose("the display name for default collation in german: %s\n", austrdup(disName) ); 373 free(disName); 374 } 375 if(U_FAILURE(status)){ 376 log_err("ERROR: in getDisplayName: %s\n", myErrorName(status)); 377 return; 378 } 379 log_verbose("Default collation getDisplayName ended.\n"); 380 381 /* BEGIN android-removed 382 To save space, Android does not include the collation tailoring rules. 383 Skip the related tests. 384 385 ruled = ucol_open("da_DK", &status); 386 log_verbose("ucol_getRules() testing ...\n"); 387 ucol_getRules(ruled, &tempLength); 388 doAssert( tempLength != 0, "getRules() result incorrect" ); 389 log_verbose("getRules tests end.\n"); 390 { 391 UChar *buffer = (UChar *)malloc(200000*sizeof(UChar)); 392 int32_t bufLen = 200000; 393 buffer[0] = '\0'; 394 log_verbose("ucol_getRulesEx() testing ...\n"); 395 tempLength = ucol_getRulesEx(col,UCOL_TAILORING_ONLY,buffer,bufLen ); 396 doAssert( tempLength == 0x00, "getRulesEx() result incorrect" ); 397 log_verbose("getRules tests end.\n"); 398 399 log_verbose("ucol_getRulesEx() testing ...\n"); 400 tempLength=ucol_getRulesEx(col,UCOL_FULL_RULES,buffer,bufLen ); 401 doAssert( tempLength != 0, "getRulesEx() result incorrect" ); 402 log_verbose("getRules tests end.\n"); 403 free(buffer); 404 } 405 ucol_close(ruled); 406 ucol_close(col); 407 408 END android-removed */ 409 410 log_verbose("open an collator for french locale"); 411 col = ucol_open("fr_FR", &status); 412 if (U_FAILURE(status)) { 413 log_err("ERROR: Creating French collation failed.: %s\n", myErrorName(status)); 414 return; 415 } 416 ucol_setStrength(col, UCOL_PRIMARY); 417 log_verbose("testing ucol_getStrength() method again ...\n"); 418 doAssert( (ucol_getStrength(col) != UCOL_TERTIARY), "collation object has the wrong strength"); 419 doAssert( (ucol_getStrength(col) == UCOL_PRIMARY), "collation object's strength is not primary difference"); 420 421 log_verbose("testing French ucol_setStrength() method ...\n"); 422 ucol_setStrength(col, UCOL_TERTIARY); 423 doAssert( (ucol_getStrength(col) == UCOL_TERTIARY), "collation object's strength is not tertiary difference"); 424 doAssert( (ucol_getStrength(col) != UCOL_PRIMARY), "collation object's strength is primary difference"); 425 doAssert( (ucol_getStrength(col) != UCOL_SECONDARY), "collation object's strength is secondary difference"); 426 ucol_close(col); 427 428 log_verbose("Get display name for the french collation in english : \n"); 429 len=ucol_getDisplayName("fr_FR", "en_US", NULL, 0, &status); 430 if(status==U_BUFFER_OVERFLOW_ERROR){ 431 status=U_ZERO_ERROR; 432 disName=(UChar*)malloc(sizeof(UChar) * (len+1)); 433 ucol_getDisplayName("fr_FR", "en_US", disName, len+1, &status); 434 log_verbose("the display name for french collation in english: %s\n", austrdup(disName) ); 435 free(disName); 436 } 437 if(U_FAILURE(status)){ 438 log_err("ERROR: in getDisplayName: %s\n", myErrorName(status)); 439 return; 440 } 441 log_verbose("Default collation getDisplayName ended.\n"); 442 443} 444 445/* Test RuleBasedCollator and getRules*/ 446void TestRuleBasedColl() 447{ 448 UCollator *col1, *col2, *col3, *col4; 449 UCollationElements *iter1, *iter2; 450 UChar ruleset1[60]; 451 UChar ruleset2[50]; 452 UChar teststr[10]; 453 UChar teststr2[10]; 454 const UChar *rule1, *rule2, *rule3, *rule4; 455 int32_t tempLength; 456 UErrorCode status = U_ZERO_ERROR; 457 u_uastrcpy(ruleset1, "&9 < a, A < b, B < c, C; ch, cH, Ch, CH < d, D, e, E"); 458 u_uastrcpy(ruleset2, "&9 < a, A < b, B < c, C < d, D, e, E"); 459 460 461 col1 = ucol_openRules(ruleset1, u_strlen(ruleset1), UCOL_DEFAULT, UCOL_DEFAULT_STRENGTH, NULL,&status); 462 if (U_FAILURE(status)) { 463 log_err_status(status, "RuleBased Collator creation failed.: %s\n", myErrorName(status)); 464 return; 465 } 466 else 467 log_verbose("PASS: RuleBased Collator creation passed\n"); 468 469 status = U_ZERO_ERROR; 470 col2 = ucol_openRules(ruleset2, u_strlen(ruleset2), UCOL_DEFAULT, UCOL_DEFAULT_STRENGTH, NULL, &status); 471 if (U_FAILURE(status)) { 472 log_err("RuleBased Collator creation failed.: %s\n", myErrorName(status)); 473 return; 474 } 475 else 476 log_verbose("PASS: RuleBased Collator creation passed\n"); 477 478 479 status = U_ZERO_ERROR; 480 col3= ucol_open(NULL, &status); 481 if (U_FAILURE(status)) { 482 log_err("Default Collator creation failed.: %s\n", myErrorName(status)); 483 return; 484 } 485 else 486 log_verbose("PASS: Default Collator creation passed\n"); 487 488 rule1 = ucol_getRules(col1, &tempLength); 489 rule2 = ucol_getRules(col2, &tempLength); 490 rule3 = ucol_getRules(col3, &tempLength); 491 492 doAssert((u_strcmp(rule1, rule2) != 0), "Default collator getRules failed"); 493 doAssert((u_strcmp(rule2, rule3) != 0), "Default collator getRules failed"); 494 doAssert((u_strcmp(rule1, rule3) != 0), "Default collator getRules failed"); 495 496 col4=ucol_openRules(rule2, u_strlen(rule2), UCOL_DEFAULT, UCOL_DEFAULT_STRENGTH, NULL, &status); 497 if (U_FAILURE(status)) { 498 log_err("RuleBased Collator creation failed.: %s\n", myErrorName(status)); 499 return; 500 } 501 rule4= ucol_getRules(col4, &tempLength); 502 doAssert((u_strcmp(rule2, rule4) == 0), "Default collator getRules failed"); 503 504 ucol_close(col1); 505 ucol_close(col2); 506 ucol_close(col3); 507 ucol_close(col4); 508 509 /* tests that modifier ! is always ignored */ 510 u_uastrcpy(ruleset1, "!&a<b"); 511 teststr[0] = 0x0e40; 512 teststr[1] = 0x0e01; 513 teststr[2] = 0x0e2d; 514 col1 = ucol_openRules(ruleset1, u_strlen(ruleset1), UCOL_DEFAULT, UCOL_DEFAULT_STRENGTH, NULL, &status); 515 if (U_FAILURE(status)) { 516 log_err("RuleBased Collator creation failed.: %s\n", myErrorName(status)); 517 return; 518 } 519 col2 = ucol_open("en_US", &status); 520 if (U_FAILURE(status)) { 521 log_err("en_US Collator creation failed.: %s\n", myErrorName(status)); 522 return; 523 } 524 iter1 = ucol_openElements(col1, teststr, 3, &status); 525 iter2 = ucol_openElements(col2, teststr, 3, &status); 526 if(U_FAILURE(status)) { 527 log_err("ERROR: CollationElement iterator creation failed.: %s\n", myErrorName(status)); 528 return; 529 } 530 while (TRUE) { 531 /* testing with en since thai has its own tailoring */ 532 uint32_t ce = ucol_next(iter1, &status); 533 uint32_t ce2 = ucol_next(iter2, &status); 534 if(U_FAILURE(status)) { 535 log_err("ERROR: CollationElement iterator creation failed.: %s\n", myErrorName(status)); 536 return; 537 } 538 if (ce2 != ce) { 539 log_err("! modifier test failed"); 540 } 541 if (ce == UCOL_NULLORDER) { 542 break; 543 } 544 } 545 ucol_closeElements(iter1); 546 ucol_closeElements(iter2); 547 ucol_close(col1); 548 ucol_close(col2); 549 /* test that we can start a rule without a & or < */ 550 u_uastrcpy(ruleset1, "< z < a"); 551 col1 = ucol_openRules(ruleset1, u_strlen(ruleset1), UCOL_DEFAULT, UCOL_DEFAULT_STRENGTH, NULL, &status); 552 if (U_FAILURE(status)) { 553 log_err("RuleBased Collator creation failed.: %s\n", myErrorName(status)); 554 return; 555 } 556 u_uastrcpy(teststr, "z"); 557 u_uastrcpy(teststr2, "a"); 558 if (ucol_greaterOrEqual(col1, teststr, 1, teststr2, 1)) { 559 log_err("Rule \"z < a\" fails"); 560 } 561 ucol_close(col1); 562} 563 564void TestCompare() 565{ 566 UErrorCode status = U_ZERO_ERROR; 567 UCollator *col; 568 UChar* test1; 569 UChar* test2; 570 571 log_verbose("The compare tests begin : \n"); 572 status=U_ZERO_ERROR; 573 col = ucol_open("en_US", &status); 574 if(U_FAILURE(status)) { 575 log_err_status(status, "ucal_open() collation creation failed.: %s\n", myErrorName(status)); 576 return; 577 } 578 test1=(UChar*)malloc(sizeof(UChar) * 6); 579 test2=(UChar*)malloc(sizeof(UChar) * 6); 580 u_uastrcpy(test1, "Abcda"); 581 u_uastrcpy(test2, "abcda"); 582 583 log_verbose("Use tertiary comparison level testing ....\n"); 584 585 doAssert( (!ucol_equal(col, test1, u_strlen(test1), test2, u_strlen(test2))), "Result should be \"Abcda\" != \"abcda\" "); 586 doAssert( (ucol_greater(col, test1, u_strlen(test1), test2, u_strlen(test2))), "Result should be \"Abcda\" >>> \"abcda\" "); 587 doAssert( (ucol_greaterOrEqual(col, test1, u_strlen(test1), test2, u_strlen(test2))), "Result should be \"Abcda\" >>> \"abcda\""); 588 589 ucol_setStrength(col, UCOL_SECONDARY); 590 log_verbose("Use secondary comparison level testing ....\n"); 591 592 doAssert( (ucol_equal(col, test1, u_strlen(test1), test2, u_strlen(test2) )), "Result should be \"Abcda\" == \"abcda\""); 593 doAssert( (!ucol_greater(col, test1, u_strlen(test1), test2, u_strlen(test2))), "Result should be \"Abcda\" == \"abcda\""); 594 doAssert( (ucol_greaterOrEqual(col, test1, u_strlen(test1), test2, u_strlen(test2) )), "Result should be \"Abcda\" == \"abcda\""); 595 596 ucol_setStrength(col, UCOL_PRIMARY); 597 log_verbose("Use primary comparison level testing ....\n"); 598 599 doAssert( (ucol_equal(col, test1, u_strlen(test1), test2, u_strlen(test2))), "Result should be \"Abcda\" == \"abcda\""); 600 doAssert( (!ucol_greater(col, test1, u_strlen(test1), test2, u_strlen(test2))), "Result should be \"Abcda\" == \"abcda\""); 601 doAssert( (ucol_greaterOrEqual(col, test1, u_strlen(test1), test2, u_strlen(test2))), "Result should be \"Abcda\" == \"abcda\""); 602 603 604 log_verbose("The compare tests end.\n"); 605 ucol_close(col); 606 free(test1); 607 free(test2); 608 609} 610/* 611--------------------------------------------- 612 tests decomposition setting 613*/ 614void TestDecomposition() { 615 UErrorCode status = U_ZERO_ERROR; 616 UCollator *en_US, *el_GR, *vi_VN; 617 en_US = ucol_open("en_US", &status); 618 el_GR = ucol_open("el_GR", &status); 619 vi_VN = ucol_open("vi_VN", &status); 620 621 if (U_FAILURE(status)) { 622 log_err_status(status, "ERROR: collation creation failed.: %s\n", myErrorName(status)); 623 return; 624 } 625 626 if (ucol_getAttribute(vi_VN, UCOL_NORMALIZATION_MODE, &status) != UCOL_ON || 627 U_FAILURE(status)) 628 { 629 log_err("ERROR: vi_VN collation did not have canonical decomposition for normalization!\n"); 630 } 631 632 status = U_ZERO_ERROR; 633 if (ucol_getAttribute(el_GR, UCOL_NORMALIZATION_MODE, &status) != UCOL_ON || 634 U_FAILURE(status)) 635 { 636 log_err("ERROR: el_GR collation did not have canonical decomposition for normalization!\n"); 637 } 638 639 status = U_ZERO_ERROR; 640 if (ucol_getAttribute(en_US, UCOL_NORMALIZATION_MODE, &status) != UCOL_OFF || 641 U_FAILURE(status)) 642 { 643 log_err("ERROR: en_US collation had canonical decomposition for normalization!\n"); 644 } 645 646 ucol_close(en_US); 647 ucol_close(el_GR); 648 ucol_close(vi_VN); 649} 650 651#define CLONETEST_COLLATOR_COUNT 4 652 653void TestSafeClone() { 654 UChar test1[6]; 655 UChar test2[6]; 656 static const UChar umlautUStr[] = {0x00DC, 0}; 657 static const UChar oeStr[] = {0x0055, 0x0045, 0}; 658 UCollator * someCollators [CLONETEST_COLLATOR_COUNT]; 659 UCollator * someClonedCollators [CLONETEST_COLLATOR_COUNT]; 660 UCollator * col; 661 UErrorCode err = U_ZERO_ERROR; 662 int8_t idx = 6; /* Leave this here to test buffer alingment in memory*/ 663 uint8_t buffer [CLONETEST_COLLATOR_COUNT] [U_COL_SAFECLONE_BUFFERSIZE]; 664 int32_t bufferSize = U_COL_SAFECLONE_BUFFERSIZE; 665 const char sampleRuleChars[] = "&Z < CH"; 666 UChar sampleRule[sizeof(sampleRuleChars)]; 667 668 if (TestBufferSize()) { 669 log_err("U_COL_SAFECLONE_BUFFERSIZE should be larger than sizeof(UCollator)\n"); 670 return; 671 } 672 673 u_uastrcpy(test1, "abCda"); 674 u_uastrcpy(test2, "abcda"); 675 u_uastrcpy(sampleRule, sampleRuleChars); 676 677 /* one default collator & two complex ones */ 678 someCollators[0] = ucol_open("en_US", &err); 679 someCollators[1] = ucol_open("ko", &err); 680 someCollators[2] = ucol_open("ja_JP", &err); 681 someCollators[3] = ucol_openRules(sampleRule, -1, UCOL_ON, UCOL_TERTIARY, NULL, &err); 682 if(U_FAILURE(err)) { 683 for (idx = 0; idx < CLONETEST_COLLATOR_COUNT; idx++) { 684 ucol_close(someCollators[idx]); 685 } 686 log_data_err("Couldn't open one or more collators\n"); 687 return; 688 } 689 690 /* Check the various error & informational states: */ 691 692 /* Null status - just returns NULL */ 693 if (0 != ucol_safeClone(someCollators[0], buffer[0], &bufferSize, 0)) 694 { 695 log_err("FAIL: Cloned Collator failed to deal correctly with null status\n"); 696 } 697 /* error status - should return 0 & keep error the same */ 698 err = U_MEMORY_ALLOCATION_ERROR; 699 if (0 != ucol_safeClone(someCollators[0], buffer[0], &bufferSize, &err) || err != U_MEMORY_ALLOCATION_ERROR) 700 { 701 log_err("FAIL: Cloned Collator failed to deal correctly with incoming error status\n"); 702 } 703 err = U_ZERO_ERROR; 704 705 /* Null buffer size pointer - just returns NULL & set error to U_ILLEGAL_ARGUMENT_ERROR*/ 706 if (0 != ucol_safeClone(someCollators[0], buffer[0], 0, &err) || err != U_ILLEGAL_ARGUMENT_ERROR) 707 { 708 log_err("FAIL: Cloned Collator failed to deal correctly with null bufferSize pointer\n"); 709 } 710 err = U_ZERO_ERROR; 711 712 /* buffer size pointer is 0 - fill in pbufferSize with a size */ 713 bufferSize = 0; 714 if (0 != ucol_safeClone(someCollators[0], buffer[0], &bufferSize, &err) || U_FAILURE(err) || bufferSize <= 0) 715 { 716 log_err("FAIL: Cloned Collator failed a sizing request ('preflighting')\n"); 717 } 718 /* Verify our define is large enough */ 719 if (U_COL_SAFECLONE_BUFFERSIZE < bufferSize) 720 { 721 log_err("FAIL: Pre-calculated buffer size is too small\n"); 722 } 723 /* Verify we can use this run-time calculated size */ 724 if (0 == (col = ucol_safeClone(someCollators[0], buffer[0], &bufferSize, &err)) || U_FAILURE(err)) 725 { 726 log_err("FAIL: Collator can't be cloned with run-time size\n"); 727 } 728 if (col) ucol_close(col); 729 /* size one byte too small - should allocate & let us know */ 730 --bufferSize; 731 if (0 == (col = ucol_safeClone(someCollators[0], 0, &bufferSize, &err)) || err != U_SAFECLONE_ALLOCATED_WARNING) 732 { 733 log_err("FAIL: Cloned Collator failed to deal correctly with too-small buffer size\n"); 734 } 735 if (col) ucol_close(col); 736 err = U_ZERO_ERROR; 737 bufferSize = U_COL_SAFECLONE_BUFFERSIZE; 738 739 740 /* Null buffer pointer - return Collator & set error to U_SAFECLONE_ALLOCATED_ERROR */ 741 if (0 == (col = ucol_safeClone(someCollators[0], 0, &bufferSize, &err)) || err != U_SAFECLONE_ALLOCATED_WARNING) 742 { 743 log_err("FAIL: Cloned Collator failed to deal correctly with null buffer pointer\n"); 744 } 745 if (col) ucol_close(col); 746 err = U_ZERO_ERROR; 747 748 /* Null Collator - return NULL & set U_ILLEGAL_ARGUMENT_ERROR */ 749 if (0 != ucol_safeClone(0, buffer[0], &bufferSize, &err) || err != U_ILLEGAL_ARGUMENT_ERROR) 750 { 751 log_err("FAIL: Cloned Collator failed to deal correctly with null Collator pointer\n"); 752 } 753 754 err = U_ZERO_ERROR; 755 756 /* Test that a cloned collator doesn't accidentally use UCA. */ 757 col=ucol_open("de@collation=phonebook", &err); 758 bufferSize = U_COL_SAFECLONE_BUFFERSIZE; 759 someClonedCollators[0] = ucol_safeClone(col, buffer[0], &bufferSize, &err); 760 doAssert( (ucol_greater(col, umlautUStr, u_strlen(umlautUStr), oeStr, u_strlen(oeStr))), "Original German phonebook collation sorts differently than expected"); 761 doAssert( (ucol_greater(someClonedCollators[0], umlautUStr, u_strlen(umlautUStr), oeStr, u_strlen(oeStr))), "Cloned German phonebook collation sorts differently than expected"); 762 if (!ucol_equals(someClonedCollators[0], col)) { 763 log_err("FAIL: Cloned German phonebook collator is not equal to original.\n"); 764 } 765 ucol_close(col); 766 ucol_close(someClonedCollators[0]); 767 768 err = U_ZERO_ERROR; 769 770 /* change orig & clone & make sure they are independent */ 771 772 for (idx = 0; idx < CLONETEST_COLLATOR_COUNT; idx++) 773 { 774 ucol_setStrength(someCollators[idx], UCOL_IDENTICAL); 775 bufferSize = 1; 776 err = U_ZERO_ERROR; 777 ucol_close(ucol_safeClone(someCollators[idx], buffer[idx], &bufferSize, &err)); 778 if (err != U_SAFECLONE_ALLOCATED_WARNING) { 779 log_err("FAIL: collator number %d was not allocated.\n", idx); 780 log_err("FAIL: status of Collator[%d] is %d (hex: %x).\n", idx, err, err); 781 } 782 783 bufferSize = U_COL_SAFECLONE_BUFFERSIZE; 784 err = U_ZERO_ERROR; 785 someClonedCollators[idx] = ucol_safeClone(someCollators[idx], buffer[idx], &bufferSize, &err); 786 if (someClonedCollators[idx] == NULL 787 || someClonedCollators[idx] < (UCollator *)buffer[idx] 788 || someClonedCollators[idx] > (UCollator *)(buffer[idx]+(U_COL_SAFECLONE_BUFFERSIZE-1))) 789 { 790 /* TODO: The use of U_COL_SAFECLONE_BUFFERSIZE will be deprecated per #9932. 791 In the meantime, just turn the following former error into a log message. */ 792 log_verbose("NOTE: Cloned collator did not use provided buffer, index %d, status %s, clone NULL? %d\n", 793 idx, myErrorName(err), someClonedCollators[idx] == NULL); 794 } 795 if (!ucol_equals(someClonedCollators[idx], someCollators[idx])) { 796 log_err("FAIL: Cloned collator is not equal to original at index = %d.\n", idx); 797 } 798 799 /* Check the usability */ 800 ucol_setStrength(someCollators[idx], UCOL_PRIMARY); 801 ucol_setAttribute(someCollators[idx], UCOL_CASE_LEVEL, UCOL_OFF, &err); 802 803 doAssert( (ucol_equal(someCollators[idx], test1, u_strlen(test1), test2, u_strlen(test2))), "Result should be \"abcda\" == \"abCda\""); 804 805 /* Close the original to make sure that the clone is usable. */ 806 ucol_close(someCollators[idx]); 807 808 ucol_setStrength(someClonedCollators[idx], UCOL_TERTIARY); 809 ucol_setAttribute(someClonedCollators[idx], UCOL_CASE_LEVEL, UCOL_OFF, &err); 810 doAssert( (ucol_greater(someClonedCollators[idx], test1, u_strlen(test1), test2, u_strlen(test2))), "Result should be \"abCda\" >>> \"abcda\" "); 811 812 ucol_close(someClonedCollators[idx]); 813 } 814} 815 816void TestCloneBinary(){ 817 UErrorCode err = U_ZERO_ERROR; 818 UCollator * col = ucol_open("en_US", &err); 819 UCollator * c; 820 int32_t size; 821 uint8_t * buffer; 822 823 if (U_FAILURE(err)) { 824 log_data_err("Couldn't open collator. Error: %s\n", u_errorName(err)); 825 return; 826 } 827 828 size = ucol_cloneBinary(col, NULL, 0, &err); 829 if(size==0 || err!=U_BUFFER_OVERFLOW_ERROR) { 830 log_err("ucol_cloneBinary - couldn't check size. Error: %s\n", u_errorName(err)); 831 return; 832 } 833 err = U_ZERO_ERROR; 834 835 buffer = (uint8_t *) malloc(size); 836 ucol_cloneBinary(col, buffer, size, &err); 837 if(U_FAILURE(err)) { 838 log_err("ucol_cloneBinary - couldn't clone.. Error: %s\n", u_errorName(err)); 839 free(buffer); 840 return; 841 } 842 843 /* how to check binary result ? */ 844 845 c = ucol_openBinary(buffer, size, col, &err); 846 if(U_FAILURE(err)) { 847 log_err("ucol_openBinary failed. Error: %s\n", u_errorName(err)); 848 } else { 849 UChar t[] = {0x41, 0x42, 0x43, 0}; /* ABC */ 850 uint8_t *k1, *k2; 851 int l1, l2; 852 l1 = ucol_getSortKey(col, t, -1, NULL,0); 853 l2 = ucol_getSortKey(c, t, -1, NULL,0); 854 k1 = (uint8_t *) malloc(sizeof(uint8_t) * l1); 855 k2 = (uint8_t *) malloc(sizeof(uint8_t) * l2); 856 ucol_getSortKey(col, t, -1, k1, l1); 857 ucol_getSortKey(col, t, -1, k2, l2); 858 if (strcmp((char *)k1,(char *)k2) != 0){ 859 log_err("ucol_openBinary - new collator should equal to old one\n"); 860 }; 861 free(k1); 862 free(k2); 863 } 864 free(buffer); 865 ucol_close(c); 866 ucol_close(col); 867} 868 869 870static void TestBengaliSortKey(void) 871{ 872 const char *curLoc = "bn"; 873 UChar str1[] = { 0x09BE, 0 }; 874 UChar str2[] = { 0x0B70, 0 }; 875 UCollator *c2 = NULL; 876 const UChar *rules; 877 int32_t rulesLength=-1; 878 uint8_t *sortKey1; 879 int32_t sortKeyLen1 = 0; 880 uint8_t *sortKey2; 881 int32_t sortKeyLen2 = 0; 882 UErrorCode status = U_ZERO_ERROR; 883 char sortKeyStr1[2048]; 884 uint32_t sortKeyStrLen1 = sizeof(sortKeyStr1)/sizeof(sortKeyStr1[0]); 885 char sortKeyStr2[2048]; 886 uint32_t sortKeyStrLen2 = sizeof(sortKeyStr2)/sizeof(sortKeyStr2[0]); 887 UCollationResult result; 888 889 static UChar preRules[41] = { 0x26, 0x9fa, 0x3c, 0x98c, 0x3c, 0x9e1, 0x3c, 0x98f, 0x3c, 0x990, 0x3c, 0x993, 0x3c, 0x994, 0x3c, 0x9bc, 0x3c, 0x982, 0x3c, 0x983, 0x3c, 0x981, 0x3c, 0x9b0, 0x3c, 0x9b8, 0x3c, 0x9b9, 0x3c, 0x9bd, 0x3c, 0x9be, 0x3c, 0x9bf, 0x3c, 0x9c8, 0x3c, 0x9cb, 0x3d, 0x9cb , 0}; 890 891 rules = preRules; 892 893 log_verbose("Rules: %s\n", aescstrdup(rules, rulesLength)); 894 895 c2 = ucol_openRules(rules, rulesLength, UCOL_DEFAULT, UCOL_DEFAULT_STRENGTH, NULL, &status); 896 if (U_FAILURE(status)) { 897 log_data_err("ERROR: Creating collator from rules failed with locale: %s : %s\n", curLoc, myErrorName(status)); 898 return; 899 } 900 901 sortKeyLen1 = ucol_getSortKey(c2, str1, -1, NULL, 0); 902 sortKey1 = (uint8_t*)malloc(sortKeyLen1+1); 903 ucol_getSortKey(c2,str1,-1,sortKey1, sortKeyLen1+1); 904 ucol_sortKeyToString(c2, sortKey1, sortKeyStr1, sortKeyStrLen1); 905 906 907 sortKeyLen2 = ucol_getSortKey(c2, str2, -1, NULL, 0); 908 sortKey2 = (uint8_t*)malloc(sortKeyLen2+1); 909 ucol_getSortKey(c2,str2,-1,sortKey2, sortKeyLen2+1); 910 911 ucol_sortKeyToString(c2, sortKey2, sortKeyStr2, sortKeyStrLen2); 912 913 914 915 result=ucol_strcoll(c2, str1, -1, str2, -1); 916 if(result!=UCOL_LESS) { 917 log_err("Error: %s was not less than %s: result=%d.\n", aescstrdup(str1,-1), aescstrdup(str2,-1), result); 918 log_info("[%s] -> %s (%d, from rule)\n", aescstrdup(str1,-1), sortKeyStr1, sortKeyLen1); 919 log_info("[%s] -> %s (%d, from rule)\n", aescstrdup(str2,-1), sortKeyStr2, sortKeyLen2); 920 } else { 921 log_verbose("OK: %s was less than %s: result=%d.\n", aescstrdup(str1,-1), aescstrdup(str2,-1), result); 922 log_verbose("[%s] -> %s (%d, from rule)\n", aescstrdup(str1,-1), sortKeyStr1, sortKeyLen1); 923 log_verbose("[%s] -> %s (%d, from rule)\n", aescstrdup(str2,-1), sortKeyStr2, sortKeyLen2); 924 } 925 926 free(sortKey1); 927 free(sortKey2); 928 ucol_close(c2); 929 930} 931 932/* 933 TestOpenVsOpenRules ensures that collators from ucol_open and ucol_openRules 934 will generate identical sort keys 935*/ 936void TestOpenVsOpenRules(){ 937 938 /* create an array of all the locales */ 939 int32_t numLocales = uloc_countAvailable(); 940 int32_t sizeOfStdSet; 941 uint32_t adder; 942 UChar str[41]; /* create an array of UChar of size maximum strSize + 1 */ 943 USet *stdSet; 944 char* curLoc; 945 UCollator * c1; 946 UCollator * c2; 947 const UChar* rules; 948 int32_t rulesLength; 949 int32_t sortKeyLen1, sortKeyLen2; 950 uint8_t *sortKey1 = NULL, *sortKey2 = NULL; 951 char sortKeyStr1[512], sortKeyStr2[512]; 952 uint32_t sortKeyStrLen1 = sizeof(sortKeyStr1) / sizeof(sortKeyStr1[0]), 953 sortKeyStrLen2 = sizeof(sortKeyStr2) / sizeof(sortKeyStr2[0]); 954 ULocaleData *uld; 955 int32_t x, y, z; 956 USet *eSet; 957 int32_t eSize; 958 int strSize; 959 960 UErrorCode err = U_ZERO_ERROR; 961 962 /* create a set of standard characters that aren't very interesting... 963 and then we can find some interesting ones later */ 964 965 stdSet = uset_open(0x61, 0x7A); 966 uset_addRange(stdSet, 0x41, 0x5A); 967 uset_addRange(stdSet, 0x30, 0x39); 968 sizeOfStdSet = uset_size(stdSet); 969 970 adder = 1; 971 if(getTestOption(QUICK_OPTION)) 972 { 973 adder = 10; 974 } 975 976 for(x = 0; x < numLocales; x+=adder){ 977 curLoc = (char *)uloc_getAvailable(x); 978 log_verbose("Processing %s\n", curLoc); 979 980 /* create a collator the normal API way */ 981 c1 = ucol_open(curLoc, &err); 982 if (U_FAILURE(err)) { 983 log_err("ERROR: Normal collation creation failed with locale: %s : %s\n", curLoc, myErrorName(err)); 984 return; 985 } 986 987 /* grab the rules */ 988 rules = ucol_getRules(c1, &rulesLength); 989 990 /* use those rules to create a collator from rules */ 991 c2 = ucol_openRules(rules, rulesLength, UCOL_DEFAULT, UCOL_DEFAULT_STRENGTH, NULL, &err); 992 if (U_FAILURE(err)) { 993 log_err("ERROR: Creating collator from rules failed with locale: %s : %s\n", curLoc, myErrorName(err)); 994 return; 995 } 996 997 uld = ulocdata_open(curLoc, &err); 998 999 /*now that we have some collators, we get several strings */ 1000 1001 for(y = 0; y < 5; y++){ 1002 1003 /* get a set of ALL the characters in this locale */ 1004 eSet = ulocdata_getExemplarSet(uld, NULL, 0, ULOCDATA_ES_STANDARD, &err); 1005 eSize = uset_size(eSet); 1006 1007 /* make a string with these characters in it */ 1008 strSize = (rand()%40) + 1; 1009 1010 for(z = 0; z < strSize; z++){ 1011 str[z] = uset_charAt(eSet, rand()%eSize); 1012 } 1013 1014 /* change the set to only include 'abnormal' characters (not A-Z, a-z, 0-9 */ 1015 uset_removeAll(eSet, stdSet); 1016 eSize = uset_size(eSet); 1017 1018 /* if there are some non-normal characters left, put a few into the string, just to make sure we have some */ 1019 if(eSize > 0){ 1020 str[2%strSize] = uset_charAt(eSet, rand()%eSize); 1021 str[3%strSize] = uset_charAt(eSet, rand()%eSize); 1022 str[5%strSize] = uset_charAt(eSet, rand()%eSize); 1023 str[10%strSize] = uset_charAt(eSet, rand()%eSize); 1024 str[13%strSize] = uset_charAt(eSet, rand()%eSize); 1025 } 1026 /* terminate the string */ 1027 str[strSize-1] = '\0'; 1028 log_verbose("String used: %S\n", str); 1029 1030 /* get sort keys for both of them, and check that the keys are identicle */ 1031 sortKeyLen1 = ucol_getSortKey(c1, str, u_strlen(str), NULL, 0); 1032 sortKey1 = (uint8_t*)malloc(sizeof(uint8_t) * (sortKeyLen1 + 1)); 1033 /*memset(sortKey1, 0xFE, sortKeyLen1);*/ 1034 ucol_getSortKey(c1, str, u_strlen(str), sortKey1, sortKeyLen1 + 1); 1035 ucol_sortKeyToString(c1, sortKey1, sortKeyStr1, sortKeyStrLen1); 1036 1037 sortKeyLen2 = ucol_getSortKey(c2, str, u_strlen(str), NULL, 0); 1038 sortKey2 = (uint8_t*)malloc(sizeof(uint8_t) * (sortKeyLen2 + 1)); 1039 /*memset(sortKey2, 0xFE, sortKeyLen2);*/ 1040 ucol_getSortKey(c2, str, u_strlen(str), sortKey2, sortKeyLen2 + 1); 1041 ucol_sortKeyToString(c2, sortKey2, sortKeyStr2, sortKeyStrLen2); 1042 1043 /* Check that the lengths are the same */ 1044 if (sortKeyLen1 != sortKeyLen2) { 1045 log_err("ERROR : Sort key lengths %d and %d for text '%s' in locale '%s' do not match.\n", 1046 sortKeyLen1, sortKeyLen2, str, curLoc); 1047 } 1048 1049 /* check that the keys are the same */ 1050 if (memcmp(sortKey1, sortKey2, sortKeyLen1) != 0) { 1051 log_err("ERROR : Sort keys '%s' and '%s' for text '%s' in locale '%s' are not equivalent.\n", 1052 sortKeyStr1, sortKeyStr2, str, curLoc); 1053 } 1054 1055 /* clean up after each string */ 1056 free(sortKey1); 1057 free(sortKey2); 1058 uset_close(eSet); 1059 } 1060 /* clean up after each locale */ 1061 ulocdata_close(uld); 1062 ucol_close(c1); 1063 ucol_close(c2); 1064 } 1065 /* final clean up */ 1066 uset_close(stdSet); 1067} 1068/* 1069---------------------------------------------------------------------------- 1070 ctor -- Tests the getSortKey 1071*/ 1072void TestSortKey() 1073{ 1074 uint8_t *sortk1 = NULL, *sortk2 = NULL, *sortk3 = NULL, *sortkEmpty = NULL; 1075 int32_t sortklen, osortklen; 1076 UCollator *col; 1077 UChar *test1, *test2, *test3; 1078 UErrorCode status = U_ZERO_ERROR; 1079 char toStringBuffer[256], *resultP; 1080 uint32_t toStringLen=sizeof(toStringBuffer)/sizeof(toStringBuffer[0]); 1081 1082 1083 uint8_t s1[] = { 0x9f, 0x00 }; 1084 uint8_t s2[] = { 0x61, 0x00 }; 1085 int strcmpResult; 1086 1087 strcmpResult = strcmp((const char *)s1, (const char *)s2); 1088 log_verbose("strcmp(0x9f..., 0x61...) = %d\n", strcmpResult); 1089 1090 if(strcmpResult <= 0) { 1091 log_err("ERR: expected strcmp(\"9f 00\", \"61 00\") to be >=0 (GREATER).. got %d. Calling strcmp() for sortkeys may not work! \n", 1092 strcmpResult); 1093 } 1094 1095 1096 log_verbose("testing SortKey begins...\n"); 1097 /* this is supposed to open default date format, but later on it treats it like it is "en_US" 1098 - very bad if you try to run the tests on machine where default locale is NOT "en_US" */ 1099 /* col = ucol_open(NULL, &status); */ 1100 col = ucol_open("en_US", &status); 1101 if (U_FAILURE(status)) { 1102 log_err_status(status, "ERROR: Default collation creation failed.: %s\n", myErrorName(status)); 1103 return; 1104 } 1105 1106 1107 if(ucol_getStrength(col) != UCOL_DEFAULT_STRENGTH) 1108 { 1109 log_err("ERROR: default collation did not have UCOL_DEFAULT_STRENGTH !\n"); 1110 } 1111 /* Need to use identical strength */ 1112 ucol_setAttribute(col, UCOL_STRENGTH, UCOL_IDENTICAL, &status); 1113 1114 test1=(UChar*)malloc(sizeof(UChar) * 6); 1115 test2=(UChar*)malloc(sizeof(UChar) * 6); 1116 test3=(UChar*)malloc(sizeof(UChar) * 6); 1117 1118 memset(test1,0xFE, sizeof(UChar)*6); 1119 memset(test2,0xFE, sizeof(UChar)*6); 1120 memset(test3,0xFE, sizeof(UChar)*6); 1121 1122 1123 u_uastrcpy(test1, "Abcda"); 1124 u_uastrcpy(test2, "abcda"); 1125 u_uastrcpy(test3, "abcda"); 1126 1127 log_verbose("Use tertiary comparison level testing ....\n"); 1128 1129 sortklen=ucol_getSortKey(col, test1, u_strlen(test1), NULL, 0); 1130 sortk1=(uint8_t*)malloc(sizeof(uint8_t) * (sortklen+1)); 1131 memset(sortk1,0xFE, sortklen); 1132 ucol_getSortKey(col, test1, u_strlen(test1), sortk1, sortklen+1); 1133 1134 sortklen=ucol_getSortKey(col, test2, u_strlen(test2), NULL, 0); 1135 sortk2=(uint8_t*)malloc(sizeof(uint8_t) * (sortklen+1)); 1136 memset(sortk2,0xFE, sortklen); 1137 ucol_getSortKey(col, test2, u_strlen(test2), sortk2, sortklen+1); 1138 1139 osortklen = sortklen; 1140 sortklen=ucol_getSortKey(col, test2, u_strlen(test3), NULL, 0); 1141 sortk3=(uint8_t*)malloc(sizeof(uint8_t) * (sortklen+1)); 1142 memset(sortk3,0xFE, sortklen); 1143 ucol_getSortKey(col, test2, u_strlen(test2), sortk3, sortklen+1); 1144 1145 doAssert( (sortklen == osortklen), "Sortkey length should be the same (abcda, abcda)"); 1146 1147 doAssert( (memcmp(sortk1, sortk2, sortklen) > 0), "Result should be \"Abcda\" > \"abcda\""); 1148 doAssert( (memcmp(sortk2, sortk1, sortklen) < 0), "Result should be \"abcda\" < \"Abcda\""); 1149 doAssert( (memcmp(sortk2, sortk3, sortklen) == 0), "Result should be \"abcda\" == \"abcda\""); 1150 1151 resultP = ucol_sortKeyToString(col, sortk3, toStringBuffer, toStringLen); 1152 doAssert( (resultP != 0), "sortKeyToString failed!"); 1153 1154#if 1 /* verobse log of sortkeys */ 1155 { 1156 char junk2[1000]; 1157 char junk3[1000]; 1158 int i; 1159 1160 strcpy(junk2, "abcda[2] "); 1161 strcpy(junk3, " abcda[3] "); 1162 1163 for(i=0;i<sortklen;i++) 1164 { 1165 sprintf(junk2+strlen(junk2), "%02X ",(int)( 0xFF & sortk2[i])); 1166 sprintf(junk3+strlen(junk3), "%02X ",(int)( 0xFF & sortk3[i])); 1167 } 1168 1169 log_verbose("%s\n", junk2); 1170 log_verbose("%s\n", junk3); 1171 } 1172#endif 1173 1174 free(sortk1); 1175 free(sortk2); 1176 free(sortk3); 1177 1178 log_verbose("Use secondary comparision level testing ...\n"); 1179 ucol_setStrength(col, UCOL_SECONDARY); 1180 sortklen=ucol_getSortKey(col, test1, u_strlen(test1), NULL, 0); 1181 sortk1=(uint8_t*)malloc(sizeof(uint8_t) * (sortklen+1)); 1182 ucol_getSortKey(col, test1, u_strlen(test1), sortk1, sortklen+1); 1183 sortklen=ucol_getSortKey(col, test2, u_strlen(test2), NULL, 0); 1184 sortk2=(uint8_t*)malloc(sizeof(uint8_t) * (sortklen+1)); 1185 ucol_getSortKey(col, test2, u_strlen(test2), sortk2, sortklen+1); 1186 1187 doAssert( !(memcmp(sortk1, sortk2, sortklen) > 0), "Result should be \"Abcda\" == \"abcda\""); 1188 doAssert( !(memcmp(sortk2, sortk1, sortklen) < 0), "Result should be \"abcda\" == \"Abcda\""); 1189 doAssert( (memcmp(sortk1, sortk2, sortklen) == 0), "Result should be \"abcda\" == \"abcda\""); 1190 1191 log_verbose("getting sortkey for an empty string\n"); 1192 ucol_setAttribute(col, UCOL_STRENGTH, UCOL_TERTIARY, &status); 1193 sortklen = ucol_getSortKey(col, test1, 0, NULL, 0); 1194 sortkEmpty = (uint8_t*)malloc(sizeof(uint8_t) * sortklen+1); 1195 sortklen = ucol_getSortKey(col, test1, 0, sortkEmpty, sortklen+1); 1196 if(sortklen != 3 || sortkEmpty[0] != 1 || sortkEmpty[0] != 1 || sortkEmpty[2] != 0) { 1197 log_err("Empty string generated wrong sortkey!\n"); 1198 } 1199 free(sortkEmpty); 1200 1201 log_verbose("testing passing invalid string\n"); 1202 sortklen = ucol_getSortKey(col, NULL, 0, NULL, 0); 1203 if(sortklen != 0) { 1204 log_err("Invalid string didn't return sortkey size of 0\n"); 1205 } 1206 1207 1208 log_verbose("testing sortkey ends...\n"); 1209 ucol_close(col); 1210 free(test1); 1211 free(test2); 1212 free(test3); 1213 free(sortk1); 1214 free(sortk2); 1215 1216} 1217void TestHashCode() 1218{ 1219 uint8_t *sortk1, *sortk2, *sortk3; 1220 int32_t sortk1len, sortk2len, sortk3len; 1221 UCollator *col; 1222 UChar *test1, *test2, *test3; 1223 UErrorCode status = U_ZERO_ERROR; 1224 log_verbose("testing getHashCode begins...\n"); 1225 col = ucol_open("en_US", &status); 1226 if (U_FAILURE(status)) { 1227 log_err_status(status, "ERROR: Default collation creation failed.: %s\n", myErrorName(status)); 1228 return; 1229 } 1230 test1=(UChar*)malloc(sizeof(UChar) * 6); 1231 test2=(UChar*)malloc(sizeof(UChar) * 6); 1232 test3=(UChar*)malloc(sizeof(UChar) * 6); 1233 u_uastrcpy(test1, "Abcda"); 1234 u_uastrcpy(test2, "abcda"); 1235 u_uastrcpy(test3, "abcda"); 1236 1237 log_verbose("Use tertiary comparison level testing ....\n"); 1238 sortk1len=ucol_getSortKey(col, test1, u_strlen(test1), NULL, 0); 1239 sortk1=(uint8_t*)malloc(sizeof(uint8_t) * (sortk1len+1)); 1240 ucol_getSortKey(col, test1, u_strlen(test1), sortk1, sortk1len+1); 1241 sortk2len=ucol_getSortKey(col, test2, u_strlen(test2), NULL, 0); 1242 sortk2=(uint8_t*)malloc(sizeof(uint8_t) * (sortk2len+1)); 1243 ucol_getSortKey(col, test2, u_strlen(test2), sortk2, sortk2len+1); 1244 sortk3len=ucol_getSortKey(col, test2, u_strlen(test3), NULL, 0); 1245 sortk3=(uint8_t*)malloc(sizeof(uint8_t) * (sortk3len+1)); 1246 ucol_getSortKey(col, test2, u_strlen(test2), sortk3, sortk3len+1); 1247 1248 1249 log_verbose("ucol_hashCode() testing ...\n"); 1250 1251 doAssert( ucol_keyHashCode(sortk1, sortk1len) != ucol_keyHashCode(sortk2, sortk2len), "Hash test1 result incorrect" ); 1252 doAssert( !(ucol_keyHashCode(sortk1, sortk1len) == ucol_keyHashCode(sortk2, sortk2len)), "Hash test2 result incorrect" ); 1253 doAssert( ucol_keyHashCode(sortk2, sortk2len) == ucol_keyHashCode(sortk3, sortk3len), "Hash result not equal" ); 1254 1255 log_verbose("hashCode tests end.\n"); 1256 ucol_close(col); 1257 free(sortk1); 1258 free(sortk2); 1259 free(sortk3); 1260 free(test1); 1261 free(test2); 1262 free(test3); 1263 1264 1265} 1266/* 1267 *---------------------------------------------------------------------------- 1268 * Tests the UCollatorElements API. 1269 * 1270 */ 1271void TestElemIter() 1272{ 1273 int32_t offset; 1274 int32_t order1, order2, order3; 1275 UChar *testString1, *testString2; 1276 UCollator *col; 1277 UCollationElements *iterator1, *iterator2, *iterator3; 1278 UErrorCode status = U_ZERO_ERROR; 1279 log_verbose("testing UCollatorElements begins...\n"); 1280 col = ucol_open("en_US", &status); 1281 ucol_setAttribute(col, UCOL_NORMALIZATION_MODE, UCOL_OFF, &status); 1282 if (U_FAILURE(status)) { 1283 log_err_status(status, "ERROR: Default collation creation failed.: %s\n", myErrorName(status)); 1284 return; 1285 } 1286 1287 testString1=(UChar*)malloc(sizeof(UChar) * 150); 1288 testString2=(UChar*)malloc(sizeof(UChar) * 150); 1289 u_uastrcpy(testString1, "XFILE What subset of all possible test cases has the highest probability of detecting the most errors?"); 1290 u_uastrcpy(testString2, "Xf_ile What subset of all possible test cases has the lowest probability of detecting the least errors?"); 1291 1292 log_verbose("Constructors and comparison testing....\n"); 1293 1294 iterator1 = ucol_openElements(col, testString1, u_strlen(testString1), &status); 1295 if(U_FAILURE(status)) { 1296 log_err("ERROR: Default collationElement iterator creation failed.: %s\n", myErrorName(status)); 1297 ucol_close(col); 1298 return; 1299 } 1300 else{ log_verbose("PASS: Default collationElement iterator1 creation passed\n");} 1301 1302 iterator2 = ucol_openElements(col, testString1, u_strlen(testString1), &status); 1303 if(U_FAILURE(status)) { 1304 log_err("ERROR: Default collationElement iterator creation failed.: %s\n", myErrorName(status)); 1305 ucol_close(col); 1306 return; 1307 } 1308 else{ log_verbose("PASS: Default collationElement iterator2 creation passed\n");} 1309 1310 iterator3 = ucol_openElements(col, testString2, u_strlen(testString2), &status); 1311 if(U_FAILURE(status)) { 1312 log_err("ERROR: Default collationElement iterator creation failed.: %s\n", myErrorName(status)); 1313 ucol_close(col); 1314 return; 1315 } 1316 else{ log_verbose("PASS: Default collationElement iterator3 creation passed\n");} 1317 1318 offset=ucol_getOffset(iterator1); 1319 ucol_setOffset(iterator1, 6, &status); 1320 if (U_FAILURE(status)) { 1321 log_err("Error in setOffset for UCollatorElements iterator.: %s\n", myErrorName(status)); 1322 return; 1323 } 1324 if(ucol_getOffset(iterator1)==6) 1325 log_verbose("setOffset and getOffset working fine\n"); 1326 else{ 1327 log_err("error in set and get Offset got %d instead of 6\n", ucol_getOffset(iterator1)); 1328 } 1329 1330 ucol_setOffset(iterator1, 0, &status); 1331 order1 = ucol_next(iterator1, &status); 1332 if (U_FAILURE(status)) { 1333 log_err("Somehow ran out of memory stepping through the iterator1.: %s\n", myErrorName(status)); 1334 return; 1335 } 1336 order2=ucol_getOffset(iterator2); 1337 doAssert((order1 != order2), "The first iterator advance failed"); 1338 order2 = ucol_next(iterator2, &status); 1339 if (U_FAILURE(status)) { 1340 log_err("Somehow ran out of memory stepping through the iterator2.: %s\n", myErrorName(status)); 1341 return; 1342 } 1343 order3 = ucol_next(iterator3, &status); 1344 if (U_FAILURE(status)) { 1345 log_err("Somehow ran out of memory stepping through the iterator3.: %s\n", myErrorName(status)); 1346 return; 1347 } 1348 1349 doAssert((order1 == order2), "The second iterator advance failed should be the same as first one"); 1350 1351doAssert( (ucol_primaryOrder(order1) == ucol_primaryOrder(order3)), "The primary orders should be identical"); 1352doAssert( (ucol_secondaryOrder(order1) == ucol_secondaryOrder(order3)), "The secondary orders should be identical"); 1353doAssert( (ucol_tertiaryOrder(order1) == ucol_tertiaryOrder(order3)), "The tertiary orders should be identical"); 1354 1355 order1=ucol_next(iterator1, &status); 1356 if (U_FAILURE(status)) { 1357 log_err("Somehow ran out of memory stepping through the iterator2.: %s\n", myErrorName(status)); 1358 return; 1359 } 1360 order3=ucol_next(iterator3, &status); 1361 if (U_FAILURE(status)) { 1362 log_err("Somehow ran out of memory stepping through the iterator2.: %s\n", myErrorName(status)); 1363 return; 1364 } 1365doAssert( (ucol_primaryOrder(order1) == ucol_primaryOrder(order3)), "The primary orders should be identical"); 1366doAssert( (ucol_tertiaryOrder(order1) != ucol_tertiaryOrder(order3)), "The tertiary orders should be different"); 1367 1368 order1=ucol_next(iterator1, &status); 1369 if (U_FAILURE(status)) { 1370 log_err("Somehow ran out of memory stepping through the iterator2.: %s\n", myErrorName(status)); 1371 return; 1372 } 1373 order3=ucol_next(iterator3, &status); 1374 if (U_FAILURE(status)) { 1375 log_err("Somehow ran out of memory stepping through the iterator2.: %s\n", myErrorName(status)); 1376 return; 1377 } 1378 /* this here, my friends, is either pure lunacy or something so obsolete that even it's mother 1379 * doesn't care about it. Essentialy, this test complains if secondary values for 'I' and '_' 1380 * are the same. According to the UCA, this is not true. Therefore, remove the test. 1381 * Besides, if primary strengths for two code points are different, it doesn't matter one bit 1382 * what is the relation between secondary or any other strengths. 1383 * killed by weiv 06/11/2002. 1384 */ 1385 /* 1386 doAssert( ((order1 & UCOL_SECONDARYMASK) != (order3 & UCOL_SECONDARYMASK)), "The secondary orders should be different"); 1387 */ 1388 doAssert( (order1 != UCOL_NULLORDER), "Unexpected end of iterator reached"); 1389 1390 free(testString1); 1391 free(testString2); 1392 ucol_closeElements(iterator1); 1393 ucol_closeElements(iterator2); 1394 ucol_closeElements(iterator3); 1395 ucol_close(col); 1396 1397 log_verbose("testing CollationElementIterator ends...\n"); 1398} 1399 1400void TestGetLocale() { 1401 UErrorCode status = U_ZERO_ERROR; 1402 const char *rules = "&a<x<y<z"; 1403 UChar rlz[256] = {0}; 1404 uint32_t rlzLen = u_unescape(rules, rlz, 256); 1405 1406 UCollator *coll = NULL; 1407 const char *locale = NULL; 1408 1409 int32_t i = 0; 1410 1411 /* Now that the collation tree is separate, actual==valid at all times. [alan] */ 1412 static const struct { 1413 const char* requestedLocale; 1414 const char* validLocale; 1415 const char* actualLocale; 1416 } testStruct[] = { 1417 { "sr_RS", "sr_Cyrl_RS", "sr" }, 1418 { "sh_YU", "sr_Latn_RS", "sr_Latn" }, /* was sh, then aliased to hr, now sr_Latn via import per cldrbug 5647: */ 1419 { "en_BE_FOO", "en_BE", "root" }, 1420 { "de_DE_NONEXISTANT", "de_DE", "de" } 1421 }; 1422 1423 /* test opening collators for different locales */ 1424 for(i = 0; i<sizeof(testStruct)/sizeof(testStruct[0]); i++) { 1425 status = U_ZERO_ERROR; 1426 coll = ucol_open(testStruct[i].requestedLocale, &status); 1427 if(U_FAILURE(status)) { 1428 log_err_status(status, "Failed to open collator for %s with %s\n", testStruct[i].requestedLocale, u_errorName(status)); 1429 ucol_close(coll); 1430 continue; 1431 } 1432 locale = ucol_getLocaleByType(coll, ULOC_REQUESTED_LOCALE, &status); 1433 if(strcmp(locale, testStruct[i].requestedLocale) != 0) { 1434 log_err("[Coll %s]: Error in requested locale, expected %s, got %s\n", testStruct[i].requestedLocale, testStruct[i].requestedLocale, locale); 1435 } 1436 locale = ucol_getLocaleByType(coll, ULOC_VALID_LOCALE, &status); 1437 if(strcmp(locale, testStruct[i].validLocale) != 0) { 1438 log_err("[Coll %s]: Error in valid locale, expected %s, got %s\n", testStruct[i].requestedLocale, testStruct[i].validLocale, locale); 1439 } 1440 locale = ucol_getLocaleByType(coll, ULOC_ACTUAL_LOCALE, &status); 1441 if(strcmp(locale, testStruct[i].actualLocale) != 0) { 1442 log_err("[Coll %s]: Error in actual locale, expected %s, got %s\n", testStruct[i].requestedLocale, testStruct[i].actualLocale, locale); 1443 } 1444 ucol_close(coll); 1445 } 1446 1447 /* completely non-existant locale for collator should get a default collator */ 1448 { 1449 UCollator *defaultColl = ucol_open(NULL, &status); 1450 coll = ucol_open("blahaha", &status); 1451 if(U_SUCCESS(status)) { 1452 if(strcmp(ucol_getLocaleByType(coll, ULOC_REQUESTED_LOCALE, &status), "blahaha")) { 1453 log_err("Nonexisting locale didn't preserve the requested locale\n"); 1454 } 1455 if(strcmp(ucol_getLocaleByType(coll, ULOC_VALID_LOCALE, &status), 1456 ucol_getLocaleByType(defaultColl, ULOC_VALID_LOCALE, &status))) { 1457 log_err("Valid locale for nonexisting locale locale collator differs " 1458 "from valid locale for default collator\n"); 1459 } 1460 if(strcmp(ucol_getLocaleByType(coll, ULOC_ACTUAL_LOCALE, &status), 1461 ucol_getLocaleByType(defaultColl, ULOC_ACTUAL_LOCALE, &status))) { 1462 log_err("Actual locale for nonexisting locale locale collator differs " 1463 "from actual locale for default collator\n"); 1464 } 1465 ucol_close(coll); 1466 ucol_close(defaultColl); 1467 } else { 1468 log_data_err("Couldn't open collators\n"); 1469 } 1470 } 1471 1472 1473 1474 /* collator instantiated from rules should have all three locales NULL */ 1475 coll = ucol_openRules(rlz, rlzLen, UCOL_DEFAULT, UCOL_DEFAULT, NULL, &status); 1476 locale = ucol_getLocaleByType(coll, ULOC_REQUESTED_LOCALE, &status); 1477 if(locale != NULL) { 1478 log_err("For collator instantiated from rules, requested locale returned %s instead of NULL\n", locale); 1479 } 1480 locale = ucol_getLocaleByType(coll, ULOC_VALID_LOCALE, &status); 1481 if(locale != NULL) { 1482 log_err("For collator instantiated from rules, valid locale returned %s instead of NULL\n", locale); 1483 } 1484 locale = ucol_getLocaleByType(coll, ULOC_ACTUAL_LOCALE, &status); 1485 if(locale != NULL) { 1486 log_err("For collator instantiated from rules, actual locale returned %s instead of NULL\n", locale); 1487 } 1488 ucol_close(coll); 1489 1490} 1491 1492 1493void TestGetAll() 1494{ 1495 int32_t i, count; 1496 count=ucol_countAvailable(); 1497 /* use something sensible w/o hardcoding the count */ 1498 if(count < 0){ 1499 log_err("Error in countAvailable(), it returned %d\n", count); 1500 } 1501 else{ 1502 log_verbose("PASS: countAvailable() successful, it returned %d\n", count); 1503 } 1504 for(i=0;i<count;i++) 1505 log_verbose("%s\n", ucol_getAvailable(i)); 1506 1507 1508} 1509 1510 1511struct teststruct { 1512 const char *original; 1513 uint8_t key[256]; 1514} ; 1515 1516static int compare_teststruct(const void *string1, const void *string2) { 1517 return(strcmp((const char *)((struct teststruct *)string1)->key, (const char *)((struct teststruct *)string2)->key)); 1518} 1519 1520void TestBounds() { 1521 UErrorCode status = U_ZERO_ERROR; 1522 1523 UCollator *coll = ucol_open("sh", &status); 1524 1525 uint8_t sortkey[512], lower[512], upper[512]; 1526 UChar buffer[512]; 1527 1528 static const char * const test[] = { 1529 "John Smith", 1530 "JOHN SMITH", 1531 "john SMITH", 1532 "j\\u00F6hn sm\\u00EFth", 1533 "J\\u00F6hn Sm\\u00EFth", 1534 "J\\u00D6HN SM\\u00CFTH", 1535 "john smithsonian", 1536 "John Smithsonian", 1537 }; 1538 1539 struct teststruct tests[] = { 1540 {"\\u010CAKI MIHALJ" } , 1541 {"\\u010CAKI MIHALJ" } , 1542 {"\\u010CAKI PIRO\\u0160KA" }, 1543 {"\\u010CABAI ANDRIJA" } , 1544 {"\\u010CABAI LAJO\\u0160" } , 1545 {"\\u010CABAI MARIJA" } , 1546 {"\\u010CABAI STEVAN" } , 1547 {"\\u010CABAI STEVAN" } , 1548 {"\\u010CABARKAPA BRANKO" } , 1549 {"\\u010CABARKAPA MILENKO" } , 1550 {"\\u010CABARKAPA MIROSLAV" } , 1551 {"\\u010CABARKAPA SIMO" } , 1552 {"\\u010CABARKAPA STANKO" } , 1553 {"\\u010CABARKAPA TAMARA" } , 1554 {"\\u010CABARKAPA TOMA\\u0160" } , 1555 {"\\u010CABDARI\\u0106 NIKOLA" } , 1556 {"\\u010CABDARI\\u0106 ZORICA" } , 1557 {"\\u010CABI NANDOR" } , 1558 {"\\u010CABOVI\\u0106 MILAN" } , 1559 {"\\u010CABRADI AGNEZIJA" } , 1560 {"\\u010CABRADI IVAN" } , 1561 {"\\u010CABRADI JELENA" } , 1562 {"\\u010CABRADI LJUBICA" } , 1563 {"\\u010CABRADI STEVAN" } , 1564 {"\\u010CABRDA MARTIN" } , 1565 {"\\u010CABRILO BOGDAN" } , 1566 {"\\u010CABRILO BRANISLAV" } , 1567 {"\\u010CABRILO LAZAR" } , 1568 {"\\u010CABRILO LJUBICA" } , 1569 {"\\u010CABRILO SPASOJA" } , 1570 {"\\u010CADE\\u0160 ZDENKA" } , 1571 {"\\u010CADESKI BLAGOJE" } , 1572 {"\\u010CADOVSKI VLADIMIR" } , 1573 {"\\u010CAGLJEVI\\u0106 TOMA" } , 1574 {"\\u010CAGOROVI\\u0106 VLADIMIR" } , 1575 {"\\u010CAJA VANKA" } , 1576 {"\\u010CAJI\\u0106 BOGOLJUB" } , 1577 {"\\u010CAJI\\u0106 BORISLAV" } , 1578 {"\\u010CAJI\\u0106 RADOSLAV" } , 1579 {"\\u010CAK\\u0160IRAN MILADIN" } , 1580 {"\\u010CAKAN EUGEN" } , 1581 {"\\u010CAKAN EVGENIJE" } , 1582 {"\\u010CAKAN IVAN" } , 1583 {"\\u010CAKAN JULIJAN" } , 1584 {"\\u010CAKAN MIHAJLO" } , 1585 {"\\u010CAKAN STEVAN" } , 1586 {"\\u010CAKAN VLADIMIR" } , 1587 {"\\u010CAKAN VLADIMIR" } , 1588 {"\\u010CAKAN VLADIMIR" } , 1589 {"\\u010CAKARA ANA" } , 1590 {"\\u010CAKAREVI\\u0106 MOMIR" } , 1591 {"\\u010CAKAREVI\\u0106 NEDELJKO" } , 1592 {"\\u010CAKI \\u0160ANDOR" } , 1593 {"\\u010CAKI AMALIJA" } , 1594 {"\\u010CAKI ANDRA\\u0160" } , 1595 {"\\u010CAKI LADISLAV" } , 1596 {"\\u010CAKI LAJO\\u0160" } , 1597 {"\\u010CAKI LASLO" } , 1598 }; 1599 1600 1601 1602 int32_t i = 0, j = 0, k = 0, buffSize = 0, skSize = 0, lowerSize = 0, upperSize = 0; 1603 int32_t arraySize = sizeof(tests)/sizeof(tests[0]); 1604 1605 if(U_SUCCESS(status) && coll) { 1606 for(i = 0; i<arraySize; i++) { 1607 buffSize = u_unescape(tests[i].original, buffer, 512); 1608 skSize = ucol_getSortKey(coll, buffer, buffSize, tests[i].key, 512); 1609 } 1610 1611 qsort(tests, arraySize, sizeof(struct teststruct), compare_teststruct); 1612 1613 for(i = 0; i < arraySize-1; i++) { 1614 for(j = i+1; j < arraySize; j++) { 1615 lowerSize = ucol_getBound(tests[i].key, -1, UCOL_BOUND_LOWER, 1, lower, 512, &status); 1616 upperSize = ucol_getBound(tests[j].key, -1, UCOL_BOUND_UPPER, 1, upper, 512, &status); 1617 for(k = i; k <= j; k++) { 1618 if(strcmp((const char *)lower, (const char *)tests[k].key) > 0) { 1619 log_err("Problem with lower! j = %i (%s vs %s)\n", k, tests[k].original, tests[i].original); 1620 } 1621 if(strcmp((const char *)upper, (const char *)tests[k].key) <= 0) { 1622 log_err("Problem with upper! j = %i (%s vs %s)\n", k, tests[k].original, tests[j].original); 1623 } 1624 } 1625 } 1626 } 1627 1628 1629#if 0 1630 for(i = 0; i < 1000; i++) { 1631 lowerRND = (rand()/(RAND_MAX/arraySize)); 1632 upperRND = lowerRND + (rand()/(RAND_MAX/(arraySize-lowerRND))); 1633 1634 lowerSize = ucol_getBound(tests[lowerRND].key, -1, UCOL_BOUND_LOWER, 1, lower, 512, &status); 1635 upperSize = ucol_getBound(tests[upperRND].key, -1, UCOL_BOUND_UPPER_LONG, 1, upper, 512, &status); 1636 1637 for(j = lowerRND; j<=upperRND; j++) { 1638 if(strcmp(lower, tests[j].key) > 0) { 1639 log_err("Problem with lower! j = %i (%s vs %s)\n", j, tests[j].original, tests[lowerRND].original); 1640 } 1641 if(strcmp(upper, tests[j].key) <= 0) { 1642 log_err("Problem with upper! j = %i (%s vs %s)\n", j, tests[j].original, tests[upperRND].original); 1643 } 1644 } 1645 } 1646#endif 1647 1648 1649 1650 1651 1652 for(i = 0; i<sizeof(test)/sizeof(test[0]); i++) { 1653 buffSize = u_unescape(test[i], buffer, 512); 1654 skSize = ucol_getSortKey(coll, buffer, buffSize, sortkey, 512); 1655 lowerSize = ucol_getBound(sortkey, skSize, UCOL_BOUND_LOWER, 1, lower, 512, &status); 1656 upperSize = ucol_getBound(sortkey, skSize, UCOL_BOUND_UPPER_LONG, 1, upper, 512, &status); 1657 for(j = i+1; j<sizeof(test)/sizeof(test[0]); j++) { 1658 buffSize = u_unescape(test[j], buffer, 512); 1659 skSize = ucol_getSortKey(coll, buffer, buffSize, sortkey, 512); 1660 if(strcmp((const char *)lower, (const char *)sortkey) > 0) { 1661 log_err("Problem with lower! i = %i, j = %i (%s vs %s)\n", i, j, test[i], test[j]); 1662 } 1663 if(strcmp((const char *)upper, (const char *)sortkey) <= 0) { 1664 log_err("Problem with upper! i = %i, j = %i (%s vs %s)\n", i, j, test[i], test[j]); 1665 } 1666 } 1667 } 1668 ucol_close(coll); 1669 } else { 1670 log_data_err("Couldn't open collator\n"); 1671 } 1672 1673} 1674 1675static void doOverrunTest(UCollator *coll, const UChar *uString, int32_t strLen) { 1676 int32_t skLen = 0, skLen2 = 0; 1677 uint8_t sortKey[256]; 1678 int32_t i, j; 1679 uint8_t filler = 0xFF; 1680 1681 skLen = ucol_getSortKey(coll, uString, strLen, NULL, 0); 1682 1683 for(i = 0; i < skLen; i++) { 1684 memset(sortKey, filler, 256); 1685 skLen2 = ucol_getSortKey(coll, uString, strLen, sortKey, i); 1686 if(skLen != skLen2) { 1687 log_err("For buffer size %i, got different sortkey length. Expected %i got %i\n", i, skLen, skLen2); 1688 } 1689 for(j = i; j < 256; j++) { 1690 if(sortKey[j] != filler) { 1691 log_err("Something run over index %i\n", j); 1692 break; 1693 } 1694 } 1695 } 1696} 1697 1698/* j1865 reports that if a shorter buffer is passed to 1699* to get sort key, a buffer overrun happens in some 1700* cases. This test tries to check this. 1701*/ 1702void TestSortKeyBufferOverrun(void) { 1703 UErrorCode status = U_ZERO_ERROR; 1704 const char* cString = "A very Merry liTTle-lamB.."; 1705 UChar uString[256]; 1706 int32_t strLen = 0; 1707 UCollator *coll = ucol_open("root", &status); 1708 strLen = u_unescape(cString, uString, 256); 1709 1710 if(U_SUCCESS(status)) { 1711 log_verbose("testing non ignorable\n"); 1712 ucol_setAttribute(coll, UCOL_ALTERNATE_HANDLING, UCOL_NON_IGNORABLE, &status); 1713 doOverrunTest(coll, uString, strLen); 1714 1715 log_verbose("testing shifted\n"); 1716 ucol_setAttribute(coll, UCOL_ALTERNATE_HANDLING, UCOL_SHIFTED, &status); 1717 doOverrunTest(coll, uString, strLen); 1718 1719 log_verbose("testing shifted quaternary\n"); 1720 ucol_setAttribute(coll, UCOL_STRENGTH, UCOL_QUATERNARY, &status); 1721 doOverrunTest(coll, uString, strLen); 1722 1723 log_verbose("testing with french secondaries\n"); 1724 ucol_setAttribute(coll, UCOL_FRENCH_COLLATION, UCOL_ON, &status); 1725 ucol_setAttribute(coll, UCOL_STRENGTH, UCOL_TERTIARY, &status); 1726 ucol_setAttribute(coll, UCOL_ALTERNATE_HANDLING, UCOL_NON_IGNORABLE, &status); 1727 doOverrunTest(coll, uString, strLen); 1728 1729 } 1730 ucol_close(coll); 1731} 1732 1733static void TestAttribute() 1734{ 1735 UErrorCode error = U_ZERO_ERROR; 1736 UCollator *coll = ucol_open(NULL, &error); 1737 1738 if (U_FAILURE(error)) { 1739 log_err_status(error, "Creation of default collator failed\n"); 1740 return; 1741 } 1742 1743 ucol_setAttribute(coll, UCOL_FRENCH_COLLATION, UCOL_OFF, &error); 1744 if (ucol_getAttribute(coll, UCOL_FRENCH_COLLATION, &error) != UCOL_OFF || 1745 U_FAILURE(error)) { 1746 log_err_status(error, "Setting and retrieving of the french collation failed\n"); 1747 } 1748 1749 ucol_setAttribute(coll, UCOL_FRENCH_COLLATION, UCOL_ON, &error); 1750 if (ucol_getAttribute(coll, UCOL_FRENCH_COLLATION, &error) != UCOL_ON || 1751 U_FAILURE(error)) { 1752 log_err_status(error, "Setting and retrieving of the french collation failed\n"); 1753 } 1754 1755 ucol_setAttribute(coll, UCOL_ALTERNATE_HANDLING, UCOL_SHIFTED, &error); 1756 if (ucol_getAttribute(coll, UCOL_ALTERNATE_HANDLING, &error) != UCOL_SHIFTED || 1757 U_FAILURE(error)) { 1758 log_err_status(error, "Setting and retrieving of the alternate handling failed\n"); 1759 } 1760 1761 ucol_setAttribute(coll, UCOL_ALTERNATE_HANDLING, UCOL_NON_IGNORABLE, &error); 1762 if (ucol_getAttribute(coll, UCOL_ALTERNATE_HANDLING, &error) != UCOL_NON_IGNORABLE || 1763 U_FAILURE(error)) { 1764 log_err_status(error, "Setting and retrieving of the alternate handling failed\n"); 1765 } 1766 1767 ucol_setAttribute(coll, UCOL_CASE_FIRST, UCOL_LOWER_FIRST, &error); 1768 if (ucol_getAttribute(coll, UCOL_CASE_FIRST, &error) != UCOL_LOWER_FIRST || 1769 U_FAILURE(error)) { 1770 log_err_status(error, "Setting and retrieving of the case first attribute failed\n"); 1771 } 1772 1773 ucol_setAttribute(coll, UCOL_CASE_FIRST, UCOL_UPPER_FIRST, &error); 1774 if (ucol_getAttribute(coll, UCOL_CASE_FIRST, &error) != UCOL_UPPER_FIRST || 1775 U_FAILURE(error)) { 1776 log_err_status(error, "Setting and retrieving of the case first attribute failed\n"); 1777 } 1778 1779 ucol_setAttribute(coll, UCOL_CASE_LEVEL, UCOL_ON, &error); 1780 if (ucol_getAttribute(coll, UCOL_CASE_LEVEL, &error) != UCOL_ON || 1781 U_FAILURE(error)) { 1782 log_err_status(error, "Setting and retrieving of the case level attribute failed\n"); 1783 } 1784 1785 ucol_setAttribute(coll, UCOL_CASE_LEVEL, UCOL_OFF, &error); 1786 if (ucol_getAttribute(coll, UCOL_CASE_LEVEL, &error) != UCOL_OFF || 1787 U_FAILURE(error)) { 1788 log_err_status(error, "Setting and retrieving of the case level attribute failed\n"); 1789 } 1790 1791 ucol_setAttribute(coll, UCOL_NORMALIZATION_MODE, UCOL_ON, &error); 1792 if (ucol_getAttribute(coll, UCOL_NORMALIZATION_MODE, &error) != UCOL_ON || 1793 U_FAILURE(error)) { 1794 log_err_status(error, "Setting and retrieving of the normalization on/off attribute failed\n"); 1795 } 1796 1797 ucol_setAttribute(coll, UCOL_NORMALIZATION_MODE, UCOL_OFF, &error); 1798 if (ucol_getAttribute(coll, UCOL_NORMALIZATION_MODE, &error) != UCOL_OFF || 1799 U_FAILURE(error)) { 1800 log_err_status(error, "Setting and retrieving of the normalization on/off attribute failed\n"); 1801 } 1802 1803 ucol_setAttribute(coll, UCOL_STRENGTH, UCOL_PRIMARY, &error); 1804 if (ucol_getAttribute(coll, UCOL_STRENGTH, &error) != UCOL_PRIMARY || 1805 U_FAILURE(error)) { 1806 log_err_status(error, "Setting and retrieving of the collation strength failed\n"); 1807 } 1808 1809 ucol_setAttribute(coll, UCOL_STRENGTH, UCOL_SECONDARY, &error); 1810 if (ucol_getAttribute(coll, UCOL_STRENGTH, &error) != UCOL_SECONDARY || 1811 U_FAILURE(error)) { 1812 log_err_status(error, "Setting and retrieving of the collation strength failed\n"); 1813 } 1814 1815 ucol_setAttribute(coll, UCOL_STRENGTH, UCOL_TERTIARY, &error); 1816 if (ucol_getAttribute(coll, UCOL_STRENGTH, &error) != UCOL_TERTIARY || 1817 U_FAILURE(error)) { 1818 log_err_status(error, "Setting and retrieving of the collation strength failed\n"); 1819 } 1820 1821 ucol_setAttribute(coll, UCOL_STRENGTH, UCOL_QUATERNARY, &error); 1822 if (ucol_getAttribute(coll, UCOL_STRENGTH, &error) != UCOL_QUATERNARY || 1823 U_FAILURE(error)) { 1824 log_err_status(error, "Setting and retrieving of the collation strength failed\n"); 1825 } 1826 1827 ucol_setAttribute(coll, UCOL_STRENGTH, UCOL_IDENTICAL, &error); 1828 if (ucol_getAttribute(coll, UCOL_STRENGTH, &error) != UCOL_IDENTICAL || 1829 U_FAILURE(error)) { 1830 log_err_status(error, "Setting and retrieving of the collation strength failed\n"); 1831 } 1832 1833 ucol_close(coll); 1834} 1835 1836void TestGetTailoredSet() { 1837 struct { 1838 const char *rules; 1839 const char *tests[20]; 1840 int32_t testsize; 1841 } setTest[] = { 1842 { "&a < \\u212b", { "\\u212b", "A\\u030a", "\\u00c5" }, 3}, 1843 { "& S < \\u0161 <<< \\u0160", { "\\u0161", "s\\u030C", "\\u0160", "S\\u030C" }, 4} 1844 }; 1845 1846 int32_t i = 0, j = 0; 1847 UErrorCode status = U_ZERO_ERROR; 1848 UParseError pError; 1849 1850 UCollator *coll = NULL; 1851 UChar buff[1024]; 1852 int32_t buffLen = 0; 1853 USet *set = NULL; 1854 1855 for(i = 0; i < sizeof(setTest)/sizeof(setTest[0]); i++) { 1856 buffLen = u_unescape(setTest[i].rules, buff, 1024); 1857 coll = ucol_openRules(buff, buffLen, UCOL_DEFAULT, UCOL_DEFAULT, &pError, &status); 1858 if(U_SUCCESS(status)) { 1859 set = ucol_getTailoredSet(coll, &status); 1860 if(uset_size(set) != setTest[i].testsize) { 1861 log_err("Tailored set size different (%d) than expected (%d)\n", uset_size(set), setTest[i].testsize); 1862 } 1863 for(j = 0; j < setTest[i].testsize; j++) { 1864 buffLen = u_unescape(setTest[i].tests[j], buff, 1024); 1865 if(!uset_containsString(set, buff, buffLen)) { 1866 log_err("Tailored set doesn't contain %s... It should\n", setTest[i].tests[j]); 1867 } 1868 } 1869 uset_close(set); 1870 } else { 1871 log_err_status(status, "Couldn't open collator with rules %s\n", setTest[i].rules); 1872 } 1873 ucol_close(coll); 1874 } 1875} 1876 1877static int tMemCmp(const uint8_t *first, const uint8_t *second) { 1878 int32_t firstLen = (int32_t)strlen((const char *)first); 1879 int32_t secondLen = (int32_t)strlen((const char *)second); 1880 return memcmp(first, second, uprv_min(firstLen, secondLen)); 1881} 1882static const char * strengthsC[] = { 1883 "UCOL_PRIMARY", 1884 "UCOL_SECONDARY", 1885 "UCOL_TERTIARY", 1886 "UCOL_QUATERNARY", 1887 "UCOL_IDENTICAL" 1888}; 1889 1890void TestMergeSortKeys(void) { 1891 UErrorCode status = U_ZERO_ERROR; 1892 UCollator *coll = ucol_open("en", &status); 1893 if(U_SUCCESS(status)) { 1894 1895 const char* cases[] = { 1896 "abc", 1897 "abcd", 1898 "abcde" 1899 }; 1900 uint32_t casesSize = sizeof(cases)/sizeof(cases[0]); 1901 const char* prefix = "foo"; 1902 const char* suffix = "egg"; 1903 char outBuff1[256], outBuff2[256]; 1904 1905 uint8_t **sortkeys = (uint8_t **)malloc(casesSize*sizeof(uint8_t *)); 1906 uint8_t **mergedPrefixkeys = (uint8_t **)malloc(casesSize*sizeof(uint8_t *)); 1907 uint8_t **mergedSuffixkeys = (uint8_t **)malloc(casesSize*sizeof(uint8_t *)); 1908 uint32_t *sortKeysLen = (uint32_t *)malloc(casesSize*sizeof(uint32_t)); 1909 uint8_t prefixKey[256], suffixKey[256]; 1910 uint32_t prefixKeyLen = 0, suffixKeyLen = 0, i = 0; 1911 UChar buffer[256]; 1912 uint32_t unescapedLen = 0, l1 = 0, l2 = 0; 1913 UColAttributeValue strength; 1914 1915 log_verbose("ucol_mergeSortkeys test\n"); 1916 log_verbose("Testing order of the test cases\n"); 1917 genericLocaleStarter("en", cases, casesSize); 1918 1919 for(i = 0; i<casesSize; i++) { 1920 sortkeys[i] = (uint8_t *)malloc(256*sizeof(uint8_t)); 1921 mergedPrefixkeys[i] = (uint8_t *)malloc(256*sizeof(uint8_t)); 1922 mergedSuffixkeys[i] = (uint8_t *)malloc(256*sizeof(uint8_t)); 1923 } 1924 1925 unescapedLen = u_unescape(prefix, buffer, 256); 1926 prefixKeyLen = ucol_getSortKey(coll, buffer, unescapedLen, prefixKey, 256); 1927 1928 unescapedLen = u_unescape(suffix, buffer, 256); 1929 suffixKeyLen = ucol_getSortKey(coll, buffer, unescapedLen, suffixKey, 256); 1930 1931 log_verbose("Massaging data with prefixes and different strengths\n"); 1932 strength = UCOL_PRIMARY; 1933 while(strength <= UCOL_IDENTICAL) { 1934 log_verbose("Strength %s\n", strengthsC[strength<=UCOL_QUATERNARY?strength:4]); 1935 ucol_setAttribute(coll, UCOL_STRENGTH, strength, &status); 1936 for(i = 0; i<casesSize; i++) { 1937 unescapedLen = u_unescape(cases[i], buffer, 256); 1938 sortKeysLen[i] = ucol_getSortKey(coll, buffer, unescapedLen, sortkeys[i], 256); 1939 ucol_mergeSortkeys(prefixKey, prefixKeyLen, sortkeys[i], sortKeysLen[i], mergedPrefixkeys[i], 256); 1940 ucol_mergeSortkeys(sortkeys[i], sortKeysLen[i], suffixKey, suffixKeyLen, mergedSuffixkeys[i], 256); 1941 if(i>0) { 1942 if(tMemCmp(mergedPrefixkeys[i-1], mergedPrefixkeys[i]) >= 0) { 1943 log_err("Error while comparing prefixed keys @ strength %s:\n", strengthsC[strength<=UCOL_QUATERNARY?strength:4]); 1944 log_err("%s\n%s\n", 1945 ucol_sortKeyToString(coll, mergedPrefixkeys[i-1], outBuff1, l1), 1946 ucol_sortKeyToString(coll, mergedPrefixkeys[i], outBuff2, l2)); 1947 } 1948 if(tMemCmp(mergedSuffixkeys[i-1], mergedSuffixkeys[i]) >= 0) { 1949 log_err("Error while comparing suffixed keys @ strength %s:\n", strengthsC[strength<=UCOL_QUATERNARY?strength:4]); 1950 log_err("%s\n%s\n", 1951 ucol_sortKeyToString(coll, mergedSuffixkeys[i-1], outBuff1, l1), 1952 ucol_sortKeyToString(coll, mergedSuffixkeys[i], outBuff2, l2)); 1953 } 1954 } 1955 } 1956 if(strength == UCOL_QUATERNARY) { 1957 strength = UCOL_IDENTICAL; 1958 } else { 1959 strength++; 1960 } 1961 } 1962 1963 { 1964 uint8_t smallBuf[3]; 1965 uint32_t reqLen = 0; 1966 log_verbose("testing buffer overflow\n"); 1967 reqLen = ucol_mergeSortkeys(prefixKey, prefixKeyLen, suffixKey, suffixKeyLen, smallBuf, 3); 1968 if(reqLen != (prefixKeyLen+suffixKeyLen)) { 1969 log_err("Wrong preflight size for merged sortkey\n"); 1970 } 1971 } 1972 1973 { 1974 UChar empty = 0; 1975 uint8_t emptyKey[20], abcKey[50], mergedKey[100]; 1976 int32_t emptyKeyLen = 0, abcKeyLen = 0, mergedKeyLen = 0; 1977 1978 log_verbose("testing merging with sortkeys generated for empty strings\n"); 1979 emptyKeyLen = ucol_getSortKey(coll, &empty, 0, emptyKey, 20); 1980 unescapedLen = u_unescape(cases[0], buffer, 256); 1981 abcKeyLen = ucol_getSortKey(coll, buffer, unescapedLen, abcKey, 50); 1982 mergedKeyLen = ucol_mergeSortkeys(emptyKey, emptyKeyLen, abcKey, abcKeyLen, mergedKey, 100); 1983 if(mergedKey[0] != 2) { 1984 log_err("Empty sortkey didn't produce a level separator\n"); 1985 } 1986 /* try with zeros */ 1987 mergedKeyLen = ucol_mergeSortkeys(emptyKey, 0, abcKey, abcKeyLen, mergedKey, 100); 1988 if(mergedKeyLen != 0 || mergedKey[0] != 0) { 1989 log_err("Empty key didn't produce null mergedKey\n"); 1990 } 1991 mergedKeyLen = ucol_mergeSortkeys(abcKey, abcKeyLen, emptyKey, 0, mergedKey, 100); 1992 if(mergedKeyLen != 0 || mergedKey[0] != 0) { 1993 log_err("Empty key didn't produce null mergedKey\n"); 1994 } 1995 1996 } 1997 1998 for(i = 0; i<casesSize; i++) { 1999 free(sortkeys[i]); 2000 free(mergedPrefixkeys[i]); 2001 free(mergedSuffixkeys[i]); 2002 } 2003 free(sortkeys); 2004 free(mergedPrefixkeys); 2005 free(mergedSuffixkeys); 2006 free(sortKeysLen); 2007 ucol_close(coll); 2008 /* need to finish this up */ 2009 } else { 2010 log_data_err("Couldn't open collator"); 2011 } 2012} 2013static void TestShortString(void) 2014{ 2015 struct { 2016 const char *input; 2017 const char *expectedOutput; 2018 const char *locale; 2019 UErrorCode expectedStatus; 2020 int32_t expectedOffset; 2021 uint32_t expectedIdentifier; 2022 } testCases[] = { 2023 /* 2024 * The following expectedOutput contains a collation weight (2700 from UCA 6.0) 2025 * which is the primary weight for the T character (U+0041) in the input. 2026 * When that character gets a different weight in FractionalUCA.txt, 2027 * the expectedOutput needs to be adjusted. 2028 * That is, when we upgrade to a new UCA version or change collation 2029 * in such a way that the absolute weight for 'A' changes, 2030 * we will get a test failure here and need to adjust the test case. 2031 */ 2032 {"LDE_RDE_KPHONEBOOK_T0041_ZLATN","B2700_KPHONEBOOK_LDE", "de@collation=phonebook", U_USING_FALLBACK_WARNING, 0, 0 }, 2033 2034 {"LEN_RUS_NO_AS_S4","AS_LROOT_NO_S4", NULL, U_USING_DEFAULT_WARNING, 0, 0 }, 2035 {"LDE_VPHONEBOOK_EO_SI","EO_KPHONEBOOK_LDE_SI", "de@collation=phonebook", U_ZERO_ERROR, 0, 0 }, 2036 {"LDE_Kphonebook","KPHONEBOOK_LDE", "de@collation=phonebook", U_ZERO_ERROR, 0, 0 }, 2037 {"Xqde_DE@collation=phonebookq_S3_EX","KPHONEBOOK_LDE", "de@collation=phonebook", U_USING_FALLBACK_WARNING, 0, 0 }, 2038 {"LFR_FO", "FO_LROOT", NULL, U_USING_DEFAULT_WARNING, 0, 0 }, 2039 {"SO_LX_AS", "", NULL, U_ILLEGAL_ARGUMENT_ERROR, 8, 0 }, 2040 {"S3_ASS_MMM", "", NULL, U_ILLEGAL_ARGUMENT_ERROR, 5, 0 } 2041 }; 2042 2043 int32_t i = 0; 2044 UCollator *coll = NULL, *fromNormalized = NULL; 2045 UParseError parseError; 2046 UErrorCode status = U_ZERO_ERROR; 2047 char fromShortBuffer[256], normalizedBuffer[256], fromNormalizedBuffer[256]; 2048 const char* locale = NULL; 2049 2050 2051 for(i = 0; i < sizeof(testCases)/sizeof(testCases[0]); i++) { 2052 status = U_ZERO_ERROR; 2053 if(testCases[i].locale) { 2054 locale = testCases[i].locale; 2055 } else { 2056 locale = NULL; 2057 } 2058 2059 coll = ucol_openFromShortString(testCases[i].input, FALSE, &parseError, &status); 2060 if(status != testCases[i].expectedStatus) { 2061 log_err_status(status, "Got status '%s' that is different from expected '%s' for '%s'\n", 2062 u_errorName(status), u_errorName(testCases[i].expectedStatus), testCases[i].input); 2063 continue; 2064 } 2065 2066 if(U_SUCCESS(status)) { 2067 ucol_getShortDefinitionString(coll, locale, fromShortBuffer, 256, &status); 2068 2069 if(strcmp(fromShortBuffer, testCases[i].expectedOutput)) { 2070 log_err("Got short string '%s' from the collator. Expected '%s' for input '%s'\n", 2071 fromShortBuffer, testCases[i].expectedOutput, testCases[i].input); 2072 } 2073 2074 ucol_normalizeShortDefinitionString(testCases[i].input, normalizedBuffer, 256, &parseError, &status); 2075 fromNormalized = ucol_openFromShortString(normalizedBuffer, FALSE, &parseError, &status); 2076 ucol_getShortDefinitionString(fromNormalized, locale, fromNormalizedBuffer, 256, &status); 2077 2078 if(strcmp(fromShortBuffer, fromNormalizedBuffer)) { 2079 log_err("Strings obtained from collators instantiated by short string ('%s') and from normalized string ('%s') differ\n", 2080 fromShortBuffer, fromNormalizedBuffer); 2081 } 2082 2083 2084 if(!ucol_equals(coll, fromNormalized)) { 2085 log_err("Collator from short string ('%s') differs from one obtained through a normalized version ('%s')\n", 2086 testCases[i].input, normalizedBuffer); 2087 } 2088 2089 ucol_close(fromNormalized); 2090 ucol_close(coll); 2091 2092 } else { 2093 if(parseError.offset != testCases[i].expectedOffset) { 2094 log_err("Got parse error offset %i, but expected %i instead for '%s'\n", 2095 parseError.offset, testCases[i].expectedOffset, testCases[i].input); 2096 } 2097 } 2098 } 2099 2100} 2101 2102static void 2103doSetsTest(const char *locale, const USet *ref, USet *set, const char* inSet, const char* outSet, UErrorCode *status) { 2104 UChar buffer[512]; 2105 int32_t bufLen; 2106 2107 uset_clear(set); 2108 bufLen = u_unescape(inSet, buffer, 512); 2109 uset_applyPattern(set, buffer, bufLen, 0, status); 2110 if(U_FAILURE(*status)) { 2111 log_err("%s: Failure setting pattern %s\n", locale, u_errorName(*status)); 2112 } 2113 2114 if(!uset_containsAll(ref, set)) { 2115 log_err("%s: Some stuff from %s is not present in the set\n", locale, inSet); 2116 } 2117 2118 uset_clear(set); 2119 bufLen = u_unescape(outSet, buffer, 512); 2120 uset_applyPattern(set, buffer, bufLen, 0, status); 2121 if(U_FAILURE(*status)) { 2122 log_err("%s: Failure setting pattern %s\n", locale, u_errorName(*status)); 2123 } 2124 2125 if(!uset_containsNone(ref, set)) { 2126 log_err("%s: Some stuff from %s is present in the set\n", locale, outSet); 2127 } 2128} 2129 2130 2131 2132 2133static void 2134TestGetContractionsAndUnsafes(void) 2135{ 2136 static struct { 2137 const char* locale; 2138 const char* inConts; 2139 const char* outConts; 2140 const char* inExp; 2141 const char* outExp; 2142 const char* unsafeCodeUnits; 2143 const char* safeCodeUnits; 2144 } tests[] = { 2145 { "ru", 2146 "[{\\u0418\\u0306}{\\u0438\\u0306}]", 2147 "[\\u0439\\u0457]", 2148 "[\\u00e6]", 2149 "[ae]", 2150 "[\\u0418\\u0438]", 2151 "[aAbB\\u0430\\u0410\\u0433\\u0413]" 2152 }, 2153 { "uk", 2154 "[{\\u0406\\u0308}{\\u0456\\u0308}{\\u0418\\u0306}{\\u0438\\u0306}]", 2155 "[\\u0407\\u0419\\u0439\\u0457]", 2156 "[\\u00e6]", 2157 "[ae]", 2158 "[\\u0406\\u0456\\u0418\\u0438]", 2159 "[aAbBxv]", 2160 }, 2161 { "sh", 2162 "[{C\\u0301}{C\\u030C}{C\\u0341}{DZ\\u030C}{Dz\\u030C}{D\\u017D}{D\\u017E}{lj}{nj}]", 2163 "[{\\u309d\\u3099}{\\u30fd\\u3099}]", 2164 "[\\u00e6]", 2165 "[a]", 2166 "[nlcdzNLCDZ]", 2167 "[jabv]" 2168 }, 2169 { "ja", 2170 "[{\\u3053\\u3099\\u309D}{\\u3053\\u3099\\u309D\\u3099}{\\u3053\\u3099\\u309E}{\\u3053\\u3099\\u30FC}{\\u3053\\u309D}{\\u3053\\u309D\\u3099}{\\u3053\\u309E}{\\u3053\\u30FC}{\\u30B3\\u3099\\u30FC}{\\u30B3\\u3099\\u30FD}{\\u30B3\\u3099\\u30FD\\u3099}{\\u30B3\\u3099\\u30FE}{\\u30B3\\u30FC}{\\u30B3\\u30FD}{\\u30B3\\u30FD\\u3099}{\\u30B3\\u30FE}]", 2171 "[{\\u30FD\\u3099}{\\u309D\\u3099}{\\u3053\\u3099}{\\u30B3\\u3099}{lj}{nj}]", 2172 "[\\u30FE\\u00e6]", 2173 "[a]", 2174 "[\\u3099]", 2175 "[]" 2176 } 2177 }; 2178 2179 2180 2181 2182 UErrorCode status = U_ZERO_ERROR; 2183 UCollator *coll = NULL; 2184 int32_t i = 0; 2185 int32_t noConts = 0; 2186 USet *conts = uset_open(0,0); 2187 USet *exp = uset_open(0, 0); 2188 USet *set = uset_open(0,0); 2189 int32_t setBufferLen = 65536; 2190 UChar buffer[65536]; 2191 int32_t setLen = 0; 2192 2193 for(i = 0; i < sizeof(tests)/sizeof(tests[0]); i++) { 2194 log_verbose("Testing locale: %s\n", tests[i].locale); 2195 coll = ucol_open(tests[i].locale, &status); 2196 if (coll == NULL || U_FAILURE(status)) { 2197 log_err_status(status, "Unable to open collator for locale %s ==> %s\n", tests[i].locale, u_errorName(status)); 2198 continue; 2199 } 2200 ucol_getContractionsAndExpansions(coll, conts, exp, TRUE, &status); 2201 doSetsTest(tests[i].locale, conts, set, tests[i].inConts, tests[i].outConts, &status); 2202 setLen = uset_toPattern(conts, buffer, setBufferLen, TRUE, &status); 2203 if(U_SUCCESS(status)) { 2204 /*log_verbose("Contractions %i: %s\n", uset_getItemCount(conts), aescstrdup(buffer, setLen));*/ 2205 } else { 2206 log_err("error %s. %i\n", u_errorName(status), setLen); 2207 status = U_ZERO_ERROR; 2208 } 2209 doSetsTest(tests[i].locale, exp, set, tests[i].inExp, tests[i].outExp, &status); 2210 setLen = uset_toPattern(exp, buffer, setBufferLen, TRUE, &status); 2211 if(U_SUCCESS(status)) { 2212 /*log_verbose("Expansions %i: %s\n", uset_getItemCount(exp), aescstrdup(buffer, setLen));*/ 2213 } else { 2214 log_err("error %s. %i\n", u_errorName(status), setLen); 2215 status = U_ZERO_ERROR; 2216 } 2217 2218 noConts = ucol_getUnsafeSet(coll, conts, &status); 2219 doSetsTest(tests[i].locale, conts, set, tests[i].unsafeCodeUnits, tests[i].safeCodeUnits, &status); 2220 setLen = uset_toPattern(conts, buffer, setBufferLen, TRUE, &status); 2221 if(U_SUCCESS(status)) { 2222 log_verbose("Unsafe %i: %s\n", uset_getItemCount(exp), aescstrdup(buffer, setLen)); 2223 } else { 2224 log_err("error %s. %i\n", u_errorName(status), setLen); 2225 status = U_ZERO_ERROR; 2226 } 2227 2228 ucol_close(coll); 2229 } 2230 2231 2232 uset_close(conts); 2233 uset_close(exp); 2234 uset_close(set); 2235} 2236 2237static void 2238TestOpenBinary(void) 2239{ 2240 UErrorCode status = U_ZERO_ERROR; 2241 /* 2242 char rule[] = "&h < d < c < b"; 2243 char *wUCA[] = { "a", "h", "d", "c", "b", "i" }; 2244 char *noUCA[] = {"d", "c", "b", "a", "h", "i" }; 2245 */ 2246 /* we have to use Cyrillic letters because latin-1 always gets copied */ 2247 const char rule[] = "&\\u0452 < \\u0434 < \\u0433 < \\u0432"; /* &dje < d < g < v */ 2248 const char *wUCA[] = { "\\u0430", "\\u0452", "\\u0434", "\\u0433", "\\u0432", "\\u0435" }; /* a, dje, d, g, v, e */ 2249 const char *noUCA[] = {"\\u0434", "\\u0433", "\\u0432", "\\u0430", "\\u0435", "\\u0452" }; /* d, g, v, a, e, dje */ 2250 2251 UChar uRules[256]; 2252 int32_t uRulesLen = u_unescape(rule, uRules, 256); 2253 2254 UCollator *coll = ucol_openRules(uRules, uRulesLen, UCOL_DEFAULT, UCOL_DEFAULT, NULL, &status); 2255 UCollator *UCA = NULL; 2256 UCollator *cloneNOUCA = NULL, *cloneWUCA = NULL; 2257 2258 uint8_t imageBuffer[32768]; 2259 uint8_t *image = imageBuffer; 2260 int32_t imageBufferCapacity = 32768; 2261 2262 int32_t imageSize; 2263 2264 if((coll==NULL)||(U_FAILURE(status))) { 2265 log_data_err("could not load collators or error occured: %s\n", 2266 u_errorName(status)); 2267 return; 2268 } 2269 UCA = ucol_open("root", &status); 2270 if((UCA==NULL)||(U_FAILURE(status))) { 2271 log_data_err("could not load UCA collator or error occured: %s\n", 2272 u_errorName(status)); 2273 return; 2274 } 2275 imageSize = ucol_cloneBinary(coll, image, imageBufferCapacity, &status); 2276 if(U_FAILURE(status)) { 2277 image = (uint8_t *)malloc(imageSize*sizeof(uint8_t)); 2278 status = U_ZERO_ERROR; 2279 imageSize = ucol_cloneBinary(coll, imageBuffer, imageSize, &status); 2280 } 2281 2282 2283 cloneWUCA = ucol_openBinary(image, imageSize, UCA, &status); 2284 cloneNOUCA = ucol_openBinary(image, imageSize, NULL, &status); 2285 2286 genericOrderingTest(coll, wUCA, sizeof(wUCA)/sizeof(wUCA[0])); 2287 2288 genericOrderingTest(cloneWUCA, wUCA, sizeof(wUCA)/sizeof(wUCA[0])); 2289 genericOrderingTest(cloneNOUCA, noUCA, sizeof(noUCA)/sizeof(noUCA[0])); 2290 2291 if(image != imageBuffer) { 2292 free(image); 2293 } 2294 ucol_close(coll); 2295 ucol_close(cloneNOUCA); 2296 ucol_close(cloneWUCA); 2297 ucol_close(UCA); 2298} 2299 2300static void TestDefault(void) { 2301 /* Tests for code coverage. */ 2302 UErrorCode status = U_ZERO_ERROR; 2303 UCollator *coll = ucol_open("es@collation=pinyin", &status); 2304 if (coll == NULL || status == U_FILE_ACCESS_ERROR) { 2305 log_data_err("Unable to open collator es@collation=pinyin\n"); 2306 return; 2307 } 2308 if (status != U_USING_DEFAULT_WARNING) { 2309 /* What do you mean that you know about using pinyin collation in Spanish!? This should be in the zh locale. */ 2310 log_err("es@collation=pinyin should return U_USING_DEFAULT_WARNING, but returned %s\n", u_errorName(status)); 2311 } 2312 ucol_close(coll); 2313 if (ucol_getKeywordValues("funky", &status) != NULL) { 2314 log_err("Collators should not know about the funky keyword.\n"); 2315 } 2316 if (status != U_ILLEGAL_ARGUMENT_ERROR) { 2317 log_err("funky keyword didn't fail as expected %s\n", u_errorName(status)); 2318 } 2319 if (ucol_getKeywordValues("collation", &status) != NULL) { 2320 log_err("ucol_getKeywordValues should not work when given a bad status.\n"); 2321 } 2322} 2323 2324static void TestDefaultKeyword(void) { 2325 /* Tests for code coverage. */ 2326 UErrorCode status = U_ZERO_ERROR; 2327 const char *loc = "zh_TW@collation=default"; 2328 UCollator *coll = ucol_open(loc, &status); 2329 if(U_FAILURE(status)) { 2330 log_info("Warning: ucol_open(%s, ...) returned %s, at least it didn't crash.\n", loc, u_errorName(status)); 2331 } else if (status != U_USING_FALLBACK_WARNING) { 2332 /* Hmm, skip the following test for CLDR 1.9 data and/or ICU 4.6, no longer seems to apply */ 2333 #if 0 2334 log_err("ucol_open(%s, ...) should return an error or some sort of U_USING_FALLBACK_WARNING, but returned %s\n", loc, u_errorName(status)); 2335 #endif 2336 } 2337 ucol_close(coll); 2338} 2339 2340static void TestGetKeywordValuesForLocale(void) { 2341#define INCLUDE_UNIHAN_COLLATION 0 2342#define PREFERRED_SIZE 16 2343#define MAX_NUMBER_OF_KEYWORDS 8 2344 const char *PREFERRED[PREFERRED_SIZE][MAX_NUMBER_OF_KEYWORDS+1] = { 2345 { "und", "standard", "search", NULL, NULL, NULL, NULL, NULL, NULL }, 2346 { "en_US", "standard", "search", NULL, NULL, NULL, NULL, NULL, NULL }, 2347 { "en_029", "standard", "search", NULL, NULL, NULL, NULL, NULL, NULL }, 2348 { "de_DE", "standard", "phonebook", "search", NULL, NULL, NULL, NULL, NULL }, 2349 { "de_Latn_DE", "standard", "phonebook", "search", NULL, NULL, NULL, NULL, NULL }, 2350#if INCLUDE_UNIHAN_COLLATION 2351 { "zh", "pinyin", "big5han", "gb2312han", "stroke", "zhuyin", "unihan", "search", "standard" }, 2352 { "zh_Hans", "pinyin", "big5han", "gb2312han", "stroke", "zhuyin", "unihan", "search", "standard" }, 2353 { "zh_CN", "pinyin", "big5han", "gb2312han", "stroke", "zhuyin", "unihan", "search", "standard" }, 2354 { "zh_Hant", "stroke", "big5han", "gb2312han", "pinyin", "zhuyin", "unihan", "search", "standard" }, 2355 { "zh_TW", "stroke", "big5han", "gb2312han", "pinyin", "zhuyin", "unihan", "search", "standard" }, 2356 { "zh__PINYIN", "pinyin", "big5han", "gb2312han", "stroke", "zhuyin", "unihan", "search", "standard" }, 2357#else 2358 // BEGIN android-changed. No big5han, gb2312han, or zhuyin in Android 2359 { "zh", "pinyin", "stroke", "search", "standard", NULL, NULL, NULL, NULL }, 2360 { "zh_Hans", "pinyin", "stroke", "search", "standard", NULL, NULL, NULL, NULL }, 2361 { "zh_CN", "pinyin", "stroke", "search", "standard", NULL, NULL, NULL, NULL }, 2362 { "zh_Hant", "stroke", "pinyin", "search", "standard", NULL, NULL, NULL, NULL }, 2363 { "zh_TW", "stroke", "pinyin", "search", "standard", NULL, NULL, NULL, NULL }, 2364 { "zh__PINYIN", "pinyin", "stroke", "search", "standard", NULL, NULL, NULL, NULL }, 2365 // END android-changed 2366#endif 2367 { "es_ES", "standard", "search", "traditional", NULL, NULL, NULL, NULL, NULL }, 2368 { "es__TRADITIONAL","traditional", "search", "standard", NULL, NULL, NULL, NULL, NULL }, 2369 { "und@collation=phonebook", "standard", "search", NULL, NULL, NULL, NULL, NULL, NULL }, 2370 { "de_DE@collation=big5han", "standard", "phonebook", "search", NULL, NULL, NULL, NULL, NULL}, 2371 2372 { "zzz@collation=xxx", "standard", "search", NULL, NULL, NULL, NULL, NULL, NULL } 2373 }; 2374#if INCLUDE_UNIHAN_COLLATION 2375 const int32_t expectedLength[PREFERRED_SIZE] = { 2, 2, 2, 3, 3, 8, 8, 8, 8, 8, 8, 3, 3, 2, 3, 2 }; 2376#else 2377 // BEGIN android-change 2378 // Fewer cases are expected for "zh*" since Android removes collation sequences to save space. 2379 const int32_t expectedLength[PREFERRED_SIZE] = { 2, 2, 2, 3, 3, 4, 4, 4, 4, 4, 4, 3, 3, 2, 3, 2 }; 2380 // END android-change 2381#endif 2382 2383 UErrorCode status = U_ZERO_ERROR; 2384 UEnumeration *keywordValues = NULL; 2385 int32_t i, n, size, valueLength; 2386 const char *locale = NULL, *value = NULL; 2387 UBool errorOccurred = FALSE; 2388 2389 for (i = 0; i < PREFERRED_SIZE; i++) { 2390 locale = PREFERRED[i][0]; 2391 value = NULL; 2392 valueLength = 0; 2393 size = 0; 2394 2395 keywordValues = ucol_getKeywordValuesForLocale("collation", locale, TRUE, &status); 2396 if (keywordValues == NULL || U_FAILURE(status)) { 2397 log_err_status(status, "Error getting keyword values: %s\n", u_errorName(status)); 2398 break; 2399 } 2400 size = uenum_count(keywordValues, &status); 2401 2402 if (size == expectedLength[i]) { 2403 for (n = 0; n < expectedLength[i]; n++) { 2404 if ((value = uenum_next(keywordValues, &valueLength, &status)) != NULL && U_SUCCESS(status)) { 2405 if (uprv_strcmp(value, PREFERRED[i][n+1]) != 0) { 2406 log_err("Keyword values differ: Got [%s] Expected [%s] for locale: %s\n", value, PREFERRED[i][n+1], locale); 2407 errorOccurred = TRUE; 2408 break; 2409 } 2410 2411 } else { 2412 log_err("While getting keyword value from locale: %s got this error: %s\n", locale, u_errorName(status)); 2413 errorOccurred = TRUE; 2414 break; 2415 } 2416 } 2417 if (errorOccurred) { 2418 break; 2419 } 2420 } else { 2421 log_err("Number of keywords (%d) does not match expected size (%d) for locale: %s\n", size, expectedLength[i], locale); 2422 break; 2423 } 2424 uenum_close(keywordValues); 2425 keywordValues = NULL; 2426 } 2427 if (keywordValues != NULL) { 2428 uenum_close(keywordValues); 2429 } 2430} 2431 2432 2433#endif /* #if !UCONFIG_NO_COLLATION */ 2434