1 /******************************************************************** 2 * Copyright (c) 1997-2009 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 "cstring.h" 33 34 static void TestAttribute(void); 35 static void TestDefault(void); 36 static void TestDefaultKeyword(void); 37 int TestBufferSize(); /* defined in "colutil.c" */ 38 39 40 41 42 /* next two function is modified from "i18n/ucol.cpp" to avoid include "ucol_imp.h" */ 43 static void uprv_appendByteToHexString(char *dst, uint8_t val) { 44 uint32_t len = (uint32_t)strlen(dst); 45 sprintf(dst+len, "%02X", val); 46 } 47 48 static char* U_EXPORT2 ucol_sortKeyToString(const UCollator *coll, const uint8_t *sortkey, char *buffer, uint32_t *len) { 49 int32_t strength = UCOL_PRIMARY; 50 uint32_t res_size = 0; 51 UBool doneCase = FALSE; 52 53 char *current = buffer; 54 const uint8_t *currentSk = sortkey; 55 56 UErrorCode error_code = U_ZERO_ERROR; 57 58 strcpy(current, "["); 59 60 while(strength <= UCOL_QUATERNARY && strength <= ucol_getAttribute(coll,UCOL_STRENGTH, &error_code)) { 61 if(U_FAILURE(error_code)) { 62 log_err("ucol_getAttribute returned error: %s\n", u_errorName(error_code)); 63 } 64 if(strength > UCOL_PRIMARY) { 65 strcat(current, " . "); 66 } 67 while(*currentSk != 0x01 && *currentSk != 0x00) { /* print a level */ 68 uprv_appendByteToHexString(current, *currentSk++); 69 strcat(current, " "); 70 } 71 if(ucol_getAttribute(coll,UCOL_CASE_LEVEL, &error_code) == UCOL_ON && strength == UCOL_SECONDARY && doneCase == FALSE) { 72 doneCase = TRUE; 73 } else if(ucol_getAttribute(coll,UCOL_CASE_LEVEL, &error_code) == UCOL_OFF || doneCase == TRUE || strength != UCOL_SECONDARY) { 74 strength ++; 75 } 76 if(U_FAILURE(error_code)) { 77 log_err("ucol_getAttribute returned error: %s\n", u_errorName(error_code)); 78 } 79 uprv_appendByteToHexString(current, *currentSk++); /* This should print '01' */ 80 if(strength == UCOL_QUATERNARY && ucol_getAttribute(coll,UCOL_ALTERNATE_HANDLING, &error_code) == UCOL_NON_IGNORABLE) { 81 break; 82 } 83 } 84 85 if(ucol_getAttribute(coll,UCOL_STRENGTH, &error_code) == UCOL_IDENTICAL) { 86 strcat(current, " . "); 87 while(*currentSk != 0) { 88 uprv_appendByteToHexString(current, *currentSk++); 89 strcat(current, " "); 90 } 91 92 uprv_appendByteToHexString(current, *currentSk++); 93 } 94 if(U_FAILURE(error_code)) { 95 log_err("ucol_getAttribute returned error: %s\n", u_errorName(error_code)); 96 } 97 strcat(current, "]"); 98 99 if(res_size > *len) { 100 return NULL; 101 } 102 103 return buffer; 104 } 105 /* end of avoid include "ucol_imp.h" */ 106 107 108 void addCollAPITest(TestNode** root) 109 { 110 /* WEIVTODO: return tests here */ 111 addTest(root, &TestProperty, "tscoll/capitst/TestProperty"); 112 addTest(root, &TestRuleBasedColl, "tscoll/capitst/TestRuleBasedColl"); 113 addTest(root, &TestCompare, "tscoll/capitst/TestCompare"); 114 addTest(root, &TestSortKey, "tscoll/capitst/TestSortKey"); 115 addTest(root, &TestHashCode, "tscoll/capitst/TestHashCode"); 116 addTest(root, &TestElemIter, "tscoll/capitst/TestElemIter"); 117 addTest(root, &TestGetAll, "tscoll/capitst/TestGetAll"); 118 /*addTest(root, &TestGetDefaultRules, "tscoll/capitst/TestGetDefaultRules");*/ 119 addTest(root, &TestDecomposition, "tscoll/capitst/TestDecomposition"); 120 addTest(root, &TestSafeClone, "tscoll/capitst/TestSafeClone"); 121 addTest(root, &TestCloneBinary, "tscoll/capitst/TestCloneBinary"); 122 addTest(root, &TestGetSetAttr, "tscoll/capitst/TestGetSetAttr"); 123 addTest(root, &TestBounds, "tscoll/capitst/TestBounds"); 124 addTest(root, &TestGetLocale, "tscoll/capitst/TestGetLocale"); 125 addTest(root, &TestSortKeyBufferOverrun, "tscoll/capitst/TestSortKeyBufferOverrun"); 126 addTest(root, &TestAttribute, "tscoll/capitst/TestAttribute"); 127 addTest(root, &TestGetTailoredSet, "tscoll/capitst/TestGetTailoredSet"); 128 addTest(root, &TestMergeSortKeys, "tscoll/capitst/TestMergeSortKeys"); 129 addTest(root, &TestShortString, "tscoll/capitst/TestShortString"); 130 /* BEGIN android-removed 131 To save space, Android does not include the collation tailoring rules. 132 We skip the tailing tests for collations. */ 133 /* addTest(root, &TestGetContractionsAndUnsafes, "tscoll/capitst/TestGetContractionsAndUnsafes"); */ 134 /* END android-removed */ 135 addTest(root, &TestOpenBinary, "tscoll/capitst/TestOpenBinary"); 136 addTest(root, &TestDefault, "tscoll/capitst/TestDefault"); 137 addTest(root, &TestDefaultKeyword, "tscoll/capitst/TestDefaultKeyword"); 138 /* BEGIN android-removed 139 To save space, Android does not build full collation tables and tailing rules. 140 We skip the related tests. */ 141 /* addTest(root, &TestOpenVsOpenRules, "tscoll/capitst/TestOpenVsOpenRules"); */ 142 /* addTest(root, &TestGetKeywordValuesForLocale, "tscoll/capitst/TestGetKeywordValuesForLocale"); */ 143 /* END android-removed */ 144 } 145 146 void TestGetSetAttr(void) { 147 UErrorCode status = U_ZERO_ERROR; 148 UCollator *coll = ucol_open(NULL, &status); 149 struct attrTest { 150 UColAttribute att; 151 UColAttributeValue val[5]; 152 uint32_t valueSize; 153 UColAttributeValue nonValue; 154 } attrs[] = { 155 {UCOL_FRENCH_COLLATION, {UCOL_ON, UCOL_OFF}, 2, UCOL_SHIFTED}, 156 {UCOL_ALTERNATE_HANDLING, {UCOL_NON_IGNORABLE, UCOL_SHIFTED}, 2, UCOL_OFF},/* attribute for handling variable elements*/ 157 {UCOL_CASE_FIRST, {UCOL_OFF, UCOL_LOWER_FIRST, UCOL_UPPER_FIRST}, 3, UCOL_SHIFTED},/* who goes first, lower case or uppercase */ 158 {UCOL_CASE_LEVEL, {UCOL_ON, UCOL_OFF}, 2, UCOL_SHIFTED},/* do we have an extra case level */ 159 {UCOL_NORMALIZATION_MODE, {UCOL_ON, UCOL_OFF}, 2, UCOL_SHIFTED},/* attribute for normalization */ 160 {UCOL_DECOMPOSITION_MODE, {UCOL_ON, UCOL_OFF}, 2, UCOL_SHIFTED}, 161 {UCOL_STRENGTH, {UCOL_PRIMARY, UCOL_SECONDARY, UCOL_TERTIARY, UCOL_QUATERNARY, UCOL_IDENTICAL}, 5, UCOL_SHIFTED},/* attribute for strength */ 162 {UCOL_HIRAGANA_QUATERNARY_MODE, {UCOL_ON, UCOL_OFF}, 2, UCOL_SHIFTED},/* when turned on, this attribute */ 163 }; 164 UColAttribute currAttr; 165 UColAttributeValue value; 166 uint32_t i = 0, j = 0; 167 168 if (coll == NULL) { 169 log_err_status(status, "Unable to open collator. %s\n", u_errorName(status)); 170 return; 171 } 172 for(i = 0; i<sizeof(attrs)/sizeof(attrs[0]); i++) { 173 currAttr = attrs[i].att; 174 ucol_setAttribute(coll, currAttr, UCOL_DEFAULT, &status); 175 if(U_FAILURE(status)) { 176 log_err_status(status, "ucol_setAttribute with the default value returned error: %s\n", u_errorName(status)); 177 break; 178 } 179 value = ucol_getAttribute(coll, currAttr, &status); 180 if(U_FAILURE(status)) { 181 log_err("ucol_getAttribute returned error: %s\n", u_errorName(status)); 182 break; 183 } 184 for(j = 0; j<attrs[i].valueSize; j++) { 185 ucol_setAttribute(coll, currAttr, attrs[i].val[j], &status); 186 if(U_FAILURE(status)) { 187 log_err("ucol_setAttribute with the value %i returned error: %s\n", attrs[i].val[j], u_errorName(status)); 188 break; 189 } 190 } 191 status = U_ZERO_ERROR; 192 ucol_setAttribute(coll, currAttr, attrs[i].nonValue, &status); 193 if(U_SUCCESS(status)) { 194 log_err("ucol_setAttribute with the bad value didn't return an error\n"); 195 break; 196 } 197 status = U_ZERO_ERROR; 198 199 ucol_setAttribute(coll, currAttr, value, &status); 200 if(U_FAILURE(status)) { 201 log_err("ucol_setAttribute with the default valuereturned error: %s\n", u_errorName(status)); 202 break; 203 } 204 } 205 status = U_ZERO_ERROR; 206 value = ucol_getAttribute(coll, UCOL_ATTRIBUTE_COUNT, &status); 207 if(U_SUCCESS(status)) { 208 log_err("ucol_getAttribute for UCOL_ATTRIBUTE_COUNT didn't return an error\n"); 209 } 210 status = U_ZERO_ERROR; 211 ucol_setAttribute(coll, UCOL_ATTRIBUTE_COUNT, UCOL_DEFAULT, &status); 212 if(U_SUCCESS(status)) { 213 log_err("ucol_setAttribute for UCOL_ATTRIBUTE_COUNT didn't return an error\n"); 214 } 215 status = U_ZERO_ERROR; 216 ucol_close(coll); 217 } 218 219 220 static void doAssert(int condition, const char *message) 221 { 222 if (condition==0) { 223 log_err("ERROR : %s\n", message); 224 } 225 } 226 227 #if 0 228 /* We don't have default rules, at least not in the previous sense */ 229 void TestGetDefaultRules(){ 230 uint32_t size=0; 231 UErrorCode status=U_ZERO_ERROR; 232 UCollator *coll=NULL; 233 int32_t len1 = 0, len2=0; 234 uint8_t *binColData = NULL; 235 236 UResourceBundle *res = NULL; 237 UResourceBundle *binColl = NULL; 238 uint8_t *binResult = NULL; 239 240 241 const UChar * defaultRulesArray=ucol_getDefaultRulesArray(&size); 242 log_verbose("Test the function ucol_getDefaultRulesArray()\n"); 243 244 coll = ucol_openRules(defaultRulesArray, size, UCOL_ON, UCOL_PRIMARY, &status); 245 if(U_SUCCESS(status) && coll !=NULL) { 246 binColData = (uint8_t*)ucol_cloneRuleData(coll, &len1, &status); 247 248 } 249 250 251 status=U_ZERO_ERROR; 252 res=ures_open(NULL, "root", &status); 253 if(U_FAILURE(status)){ 254 log_err("ERROR: Failed to get resource for \"root Locale\" with %s", myErrorName(status)); 255 return; 256 } 257 binColl=ures_getByKey(res, "%%Collation", binColl, &status); 258 if(U_SUCCESS(status)){ 259 binResult=(uint8_t*)ures_getBinary(binColl, &len2, &status); 260 if(U_FAILURE(status)){ 261 log_err("ERROR: ures_getBinary() failed\n"); 262 } 263 }else{ 264 log_err("ERROR: ures_getByKey(locale(default), %%Collation) failed"); 265 } 266 267 268 if(len1 != len2){ 269 log_err("Error: ucol_getDefaultRulesArray() failed to return the correct length.\n"); 270 } 271 if(memcmp(binColData, binResult, len1) != 0){ 272 log_err("Error: ucol_getDefaultRulesArray() failed\n"); 273 } 274 275 free(binColData); 276 ures_close(binColl); 277 ures_close(res); 278 ucol_close(coll); 279 280 } 281 #endif 282 283 /* Collator Properties 284 ucol_open, ucol_strcoll, getStrength/setStrength 285 getDecomposition/setDecomposition, getDisplayName*/ 286 void TestProperty() 287 { 288 UCollator *col, *ruled; 289 UChar *disName; 290 int32_t len = 0, i = 0; 291 UChar *source, *target; 292 int32_t tempLength; 293 UErrorCode status = U_ZERO_ERROR; 294 /* 295 * Expected version of the English collator. 296 * Currently, the major/minor version numbers change when the builder code 297 * changes, 298 * number 2 is from the tailoring data version and 299 * number 3 is the UCA version. 300 * This changes with every UCA version change, and the expected value 301 * needs to be adjusted. 302 * Same in intltest/apicoll.cpp. 303 */ 304 UVersionInfo currVersionArray = {0x31, 0xC0, 0x00, 0x2A}; 305 UVersionInfo currUCAVersionArray = {5, 2, 0, 0}; 306 UVersionInfo versionArray = {0, 0, 0, 0}; 307 UVersionInfo versionUCAArray = {0, 0, 0, 0}; 308 309 log_verbose("The property tests begin : \n"); 310 log_verbose("Test ucol_strcoll : \n"); 311 col = ucol_open("en_US", &status); 312 if (U_FAILURE(status)) { 313 log_err_status(status, "Default Collator creation failed.: %s\n", myErrorName(status)); 314 return; 315 } 316 317 ucol_getVersion(col, versionArray); 318 for (i=0; i<4; ++i) { 319 if (versionArray[i] != currVersionArray[i]) { 320 log_err("Testing ucol_getVersion() - unexpected result: %hu.%hu.%hu.%hu\n", 321 versionArray[0], versionArray[1], versionArray[2], versionArray[3]); 322 break; 323 } 324 } 325 326 ucol_getUCAVersion(col, versionUCAArray); 327 for (i=0; i<4; ++i) { 328 if (versionUCAArray[i] != currUCAVersionArray[i]) { 329 log_err("Testing ucol_getUCAVersion() - unexpected result: %hu.%hu.%hu.%hu\n", 330 versionUCAArray[0], versionUCAArray[1], versionUCAArray[2], versionUCAArray[3]); 331 break; 332 } 333 } 334 335 source=(UChar*)malloc(sizeof(UChar) * 12); 336 target=(UChar*)malloc(sizeof(UChar) * 12); 337 338 339 u_uastrcpy(source, "ab"); 340 u_uastrcpy(target, "abc"); 341 342 doAssert((ucol_strcoll(col, source, u_strlen(source), target, u_strlen(target)) == UCOL_LESS), "ab < abc comparison failed"); 343 344 u_uastrcpy(source, "ab"); 345 u_uastrcpy(target, "AB"); 346 347 doAssert((ucol_strcoll(col, source, u_strlen(source), target, u_strlen(target)) == UCOL_LESS), "ab < AB comparison failed"); 348 /* u_uastrcpy(source, "black-bird"); 349 u_uastrcpy(target, "blackbird"); */ 350 u_uastrcpy(target, "black-bird"); 351 u_uastrcpy(source, "blackbird"); 352 353 doAssert((ucol_strcoll(col, source, u_strlen(source), target, u_strlen(target)) == UCOL_GREATER), 354 "black-bird > blackbird comparison failed"); 355 u_uastrcpy(source, "black bird"); 356 u_uastrcpy(target, "black-bird"); 357 doAssert((ucol_strcoll(col, source, u_strlen(source), target, u_strlen(target)) == UCOL_LESS), 358 "black bird < black-bird comparison failed"); 359 u_uastrcpy(source, "Hello"); 360 u_uastrcpy(target, "hello"); 361 362 doAssert((ucol_strcoll(col, source, u_strlen(source), target, u_strlen(target)) == UCOL_GREATER), 363 "Hello > hello comparison failed"); 364 free(source); 365 free(target); 366 log_verbose("Test ucol_strcoll ends.\n"); 367 368 log_verbose("testing ucol_getStrength() method ...\n"); 369 doAssert( (ucol_getStrength(col) == UCOL_TERTIARY), "collation object has the wrong strength"); 370 doAssert( (ucol_getStrength(col) != UCOL_PRIMARY), "collation object's strength is primary difference"); 371 372 log_verbose("testing ucol_setStrength() method ...\n"); 373 ucol_setStrength(col, UCOL_SECONDARY); 374 doAssert( (ucol_getStrength(col) != UCOL_TERTIARY), "collation object's strength is secondary difference"); 375 doAssert( (ucol_getStrength(col) != UCOL_PRIMARY), "collation object's strength is primary difference"); 376 doAssert( (ucol_getStrength(col) == UCOL_SECONDARY), "collation object has the wrong strength"); 377 378 379 log_verbose("Get display name for the default collation in German : \n"); 380 381 len=ucol_getDisplayName("en_US", "de_DE", NULL, 0, &status); 382 if(status==U_BUFFER_OVERFLOW_ERROR){ 383 status=U_ZERO_ERROR; 384 disName=(UChar*)malloc(sizeof(UChar) * (len+1)); 385 ucol_getDisplayName("en_US", "de_DE", disName, len+1, &status); 386 log_verbose("the display name for default collation in german: %s\n", austrdup(disName) ); 387 free(disName); 388 } 389 if(U_FAILURE(status)){ 390 log_err("ERROR: in getDisplayName: %s\n", myErrorName(status)); 391 return; 392 } 393 log_verbose("Default collation getDisplayName ended.\n"); 394 395 /* BEGIN android-removed 396 To save space, Android does not include the collation tailoring rules. 397 Skip the related tests. 398 ruled = ucol_open("da_DK", &status); 399 log_verbose("ucol_getRules() testing ...\n"); 400 ucol_getRules(ruled, &tempLength); 401 doAssert( tempLength != 0, "getRules() result incorrect" ); 402 log_verbose("getRules tests end.\n"); 403 { 404 UChar *buffer = (UChar *)malloc(200000*sizeof(UChar)); 405 int32_t bufLen = 200000; 406 buffer[0] = '\0'; 407 log_verbose("ucol_getRulesEx() testing ...\n"); 408 tempLength = ucol_getRulesEx(col,UCOL_TAILORING_ONLY,buffer,bufLen ); 409 doAssert( tempLength == 0x0a, "getRulesEx() result incorrect" ); 410 log_verbose("getRules tests end.\n"); 411 412 log_verbose("ucol_getRulesEx() testing ...\n"); 413 tempLength=ucol_getRulesEx(col,UCOL_FULL_RULES,buffer,bufLen ); 414 doAssert( tempLength != 0, "getRulesEx() result incorrect" ); 415 log_verbose("getRules tests end.\n"); 416 free(buffer); 417 } 418 ucol_close(ruled); 419 ucol_close(col); 420 421 END android-removed */ 422 423 log_verbose("open an collator for french locale"); 424 col = ucol_open("fr_FR", &status); 425 if (U_FAILURE(status)) { 426 log_err("ERROR: Creating French collation failed.: %s\n", myErrorName(status)); 427 return; 428 } 429 ucol_setStrength(col, UCOL_PRIMARY); 430 log_verbose("testing ucol_getStrength() method again ...\n"); 431 doAssert( (ucol_getStrength(col) != UCOL_TERTIARY), "collation object has the wrong strength"); 432 doAssert( (ucol_getStrength(col) == UCOL_PRIMARY), "collation object's strength is not primary difference"); 433 434 log_verbose("testing French ucol_setStrength() method ...\n"); 435 ucol_setStrength(col, UCOL_TERTIARY); 436 doAssert( (ucol_getStrength(col) == UCOL_TERTIARY), "collation object's strength is not tertiary difference"); 437 doAssert( (ucol_getStrength(col) != UCOL_PRIMARY), "collation object's strength is primary difference"); 438 doAssert( (ucol_getStrength(col) != UCOL_SECONDARY), "collation object's strength is secondary difference"); 439 ucol_close(col); 440 441 log_verbose("Get display name for the french collation in english : \n"); 442 len=ucol_getDisplayName("fr_FR", "en_US", NULL, 0, &status); 443 if(status==U_BUFFER_OVERFLOW_ERROR){ 444 status=U_ZERO_ERROR; 445 disName=(UChar*)malloc(sizeof(UChar) * (len+1)); 446 ucol_getDisplayName("fr_FR", "en_US", disName, len+1, &status); 447 log_verbose("the display name for french collation in english: %s\n", austrdup(disName) ); 448 free(disName); 449 } 450 if(U_FAILURE(status)){ 451 log_err("ERROR: in getDisplayName: %s\n", myErrorName(status)); 452 return; 453 } 454 log_verbose("Default collation getDisplayName ended.\n"); 455 456 } 457 458 /* Test RuleBasedCollator and getRules*/ 459 void TestRuleBasedColl() 460 { 461 UCollator *col1, *col2, *col3, *col4; 462 UCollationElements *iter1, *iter2; 463 UChar ruleset1[60]; 464 UChar ruleset2[50]; 465 UChar teststr[10]; 466 UChar teststr2[10]; 467 const UChar *rule1, *rule2, *rule3, *rule4; 468 int32_t tempLength; 469 UErrorCode status = U_ZERO_ERROR; 470 u_uastrcpy(ruleset1, "&9 < a, A < b, B < c, C; ch, cH, Ch, CH < d, D, e, E"); 471 u_uastrcpy(ruleset2, "&9 < a, A < b, B < c, C < d, D, e, E"); 472 473 474 col1 = ucol_openRules(ruleset1, u_strlen(ruleset1), UCOL_DEFAULT, UCOL_DEFAULT_STRENGTH, NULL,&status); 475 if (U_FAILURE(status)) { 476 log_err_status(status, "RuleBased Collator creation failed.: %s\n", myErrorName(status)); 477 return; 478 } 479 else 480 log_verbose("PASS: RuleBased Collator creation passed\n"); 481 482 status = U_ZERO_ERROR; 483 col2 = ucol_openRules(ruleset2, u_strlen(ruleset2), UCOL_DEFAULT, UCOL_DEFAULT_STRENGTH, NULL, &status); 484 if (U_FAILURE(status)) { 485 log_err("RuleBased Collator creation failed.: %s\n", myErrorName(status)); 486 return; 487 } 488 else 489 log_verbose("PASS: RuleBased Collator creation passed\n"); 490 491 492 status = U_ZERO_ERROR; 493 col3= ucol_open(NULL, &status); 494 if (U_FAILURE(status)) { 495 log_err("Default Collator creation failed.: %s\n", myErrorName(status)); 496 return; 497 } 498 else 499 log_verbose("PASS: Default Collator creation passed\n"); 500 501 rule1 = ucol_getRules(col1, &tempLength); 502 rule2 = ucol_getRules(col2, &tempLength); 503 rule3 = ucol_getRules(col3, &tempLength); 504 505 doAssert((u_strcmp(rule1, rule2) != 0), "Default collator getRules failed"); 506 doAssert((u_strcmp(rule2, rule3) != 0), "Default collator getRules failed"); 507 doAssert((u_strcmp(rule1, rule3) != 0), "Default collator getRules failed"); 508 509 col4=ucol_openRules(rule2, u_strlen(rule2), UCOL_DEFAULT, UCOL_DEFAULT_STRENGTH, NULL, &status); 510 if (U_FAILURE(status)) { 511 log_err("RuleBased Collator creation failed.: %s\n", myErrorName(status)); 512 return; 513 } 514 rule4= ucol_getRules(col4, &tempLength); 515 doAssert((u_strcmp(rule2, rule4) == 0), "Default collator getRules failed"); 516 517 ucol_close(col1); 518 ucol_close(col2); 519 ucol_close(col3); 520 ucol_close(col4); 521 522 /* tests that modifier ! is always ignored */ 523 u_uastrcpy(ruleset1, "!&a<b"); 524 teststr[0] = 0x0e40; 525 teststr[1] = 0x0e01; 526 teststr[2] = 0x0e2d; 527 col1 = ucol_openRules(ruleset1, u_strlen(ruleset1), UCOL_DEFAULT, UCOL_DEFAULT_STRENGTH, NULL, &status); 528 if (U_FAILURE(status)) { 529 log_err("RuleBased Collator creation failed.: %s\n", myErrorName(status)); 530 return; 531 } 532 col2 = ucol_open("en_US", &status); 533 if (U_FAILURE(status)) { 534 log_err("en_US Collator creation failed.: %s\n", myErrorName(status)); 535 return; 536 } 537 iter1 = ucol_openElements(col1, teststr, 3, &status); 538 iter2 = ucol_openElements(col2, teststr, 3, &status); 539 if(U_FAILURE(status)) { 540 log_err("ERROR: CollationElement iterator creation failed.: %s\n", myErrorName(status)); 541 return; 542 } 543 while (TRUE) { 544 /* testing with en since thai has its own tailoring */ 545 uint32_t ce = ucol_next(iter1, &status); 546 uint32_t ce2 = ucol_next(iter2, &status); 547 if(U_FAILURE(status)) { 548 log_err("ERROR: CollationElement iterator creation failed.: %s\n", myErrorName(status)); 549 return; 550 } 551 if (ce2 != ce) { 552 log_err("! modifier test failed"); 553 } 554 if (ce == UCOL_NULLORDER) { 555 break; 556 } 557 } 558 ucol_closeElements(iter1); 559 ucol_closeElements(iter2); 560 ucol_close(col1); 561 ucol_close(col2); 562 /* test that we can start a rule without a & or < */ 563 u_uastrcpy(ruleset1, "< z < a"); 564 col1 = ucol_openRules(ruleset1, u_strlen(ruleset1), UCOL_DEFAULT, UCOL_DEFAULT_STRENGTH, NULL, &status); 565 if (U_FAILURE(status)) { 566 log_err("RuleBased Collator creation failed.: %s\n", myErrorName(status)); 567 return; 568 } 569 u_uastrcpy(teststr, "z"); 570 u_uastrcpy(teststr2, "a"); 571 if (ucol_greaterOrEqual(col1, teststr, 1, teststr2, 1)) { 572 log_err("Rule \"z < a\" fails"); 573 } 574 ucol_close(col1); 575 } 576 577 void TestCompare() 578 { 579 UErrorCode status = U_ZERO_ERROR; 580 UCollator *col; 581 UChar* test1; 582 UChar* test2; 583 584 log_verbose("The compare tests begin : \n"); 585 status=U_ZERO_ERROR; 586 col = ucol_open("en_US", &status); 587 if(U_FAILURE(status)) { 588 log_err_status(status, "ucal_open() collation creation failed.: %s\n", myErrorName(status)); 589 return; 590 } 591 test1=(UChar*)malloc(sizeof(UChar) * 6); 592 test2=(UChar*)malloc(sizeof(UChar) * 6); 593 u_uastrcpy(test1, "Abcda"); 594 u_uastrcpy(test2, "abcda"); 595 596 log_verbose("Use tertiary comparison level testing ....\n"); 597 598 doAssert( (!ucol_equal(col, test1, u_strlen(test1), test2, u_strlen(test2))), "Result should be \"Abcda\" != \"abcda\" "); 599 doAssert( (ucol_greater(col, test1, u_strlen(test1), test2, u_strlen(test2))), "Result should be \"Abcda\" >>> \"abcda\" "); 600 doAssert( (ucol_greaterOrEqual(col, test1, u_strlen(test1), test2, u_strlen(test2))), "Result should be \"Abcda\" >>> \"abcda\""); 601 602 ucol_setStrength(col, UCOL_SECONDARY); 603 log_verbose("Use secondary comparison level testing ....\n"); 604 605 doAssert( (ucol_equal(col, test1, u_strlen(test1), test2, u_strlen(test2) )), "Result should be \"Abcda\" == \"abcda\""); 606 doAssert( (!ucol_greater(col, test1, u_strlen(test1), test2, u_strlen(test2))), "Result should be \"Abcda\" == \"abcda\""); 607 doAssert( (ucol_greaterOrEqual(col, test1, u_strlen(test1), test2, u_strlen(test2) )), "Result should be \"Abcda\" == \"abcda\""); 608 609 ucol_setStrength(col, UCOL_PRIMARY); 610 log_verbose("Use primary comparison level testing ....\n"); 611 612 doAssert( (ucol_equal(col, test1, u_strlen(test1), test2, u_strlen(test2))), "Result should be \"Abcda\" == \"abcda\""); 613 doAssert( (!ucol_greater(col, test1, u_strlen(test1), test2, u_strlen(test2))), "Result should be \"Abcda\" == \"abcda\""); 614 doAssert( (ucol_greaterOrEqual(col, test1, u_strlen(test1), test2, u_strlen(test2))), "Result should be \"Abcda\" == \"abcda\""); 615 616 617 log_verbose("The compare tests end.\n"); 618 ucol_close(col); 619 free(test1); 620 free(test2); 621 622 } 623 /* 624 --------------------------------------------- 625 tests decomposition setting 626 */ 627 void TestDecomposition() { 628 UErrorCode status = U_ZERO_ERROR; 629 UCollator *en_US, *el_GR, *vi_VN; 630 en_US = ucol_open("en_US", &status); 631 el_GR = ucol_open("el_GR", &status); 632 vi_VN = ucol_open("vi_VN", &status); 633 634 if (U_FAILURE(status)) { 635 log_err_status(status, "ERROR: collation creation failed.: %s\n", myErrorName(status)); 636 return; 637 } 638 639 if (ucol_getAttribute(vi_VN, UCOL_NORMALIZATION_MODE, &status) != UCOL_ON || 640 U_FAILURE(status)) 641 { 642 log_err("ERROR: vi_VN collation did not have cannonical decomposition for normalization!\n"); 643 } 644 645 status = U_ZERO_ERROR; 646 if (ucol_getAttribute(el_GR, UCOL_NORMALIZATION_MODE, &status) != UCOL_ON || 647 U_FAILURE(status)) 648 { 649 log_err("ERROR: el_GR collation did not have cannonical decomposition for normalization!\n"); 650 } 651 652 status = U_ZERO_ERROR; 653 if (ucol_getAttribute(en_US, UCOL_NORMALIZATION_MODE, &status) != UCOL_OFF || 654 U_FAILURE(status)) 655 { 656 log_err("ERROR: en_US collation had cannonical decomposition for normalization!\n"); 657 } 658 659 ucol_close(en_US); 660 ucol_close(el_GR); 661 ucol_close(vi_VN); 662 } 663 664 #define CLONETEST_COLLATOR_COUNT 4 665 666 void TestSafeClone() { 667 UChar test1[6]; 668 UChar test2[6]; 669 static const UChar umlautUStr[] = {0x00DC, 0}; 670 static const UChar oeStr[] = {0x0055, 0x0045, 0}; 671 UCollator * someCollators [CLONETEST_COLLATOR_COUNT]; 672 UCollator * someClonedCollators [CLONETEST_COLLATOR_COUNT]; 673 UCollator * col; 674 UErrorCode err = U_ZERO_ERROR; 675 int8_t index = 6; /* Leave this here to test buffer alingment in memory*/ 676 uint8_t buffer [CLONETEST_COLLATOR_COUNT] [U_COL_SAFECLONE_BUFFERSIZE]; 677 int32_t bufferSize = U_COL_SAFECLONE_BUFFERSIZE; 678 const char sampleRuleChars[] = "&Z < CH"; 679 UChar sampleRule[sizeof(sampleRuleChars)]; 680 681 if (TestBufferSize()) { 682 log_err("U_COL_SAFECLONE_BUFFERSIZE should be larger than sizeof(UCollator)\n"); 683 return; 684 } 685 686 u_uastrcpy(test1, "abCda"); 687 u_uastrcpy(test2, "abcda"); 688 u_uastrcpy(sampleRule, sampleRuleChars); 689 690 /* one default collator & two complex ones */ 691 someCollators[0] = ucol_open("en_US", &err); 692 someCollators[1] = ucol_open("ko", &err); 693 someCollators[2] = ucol_open("ja_JP", &err); 694 someCollators[3] = ucol_openRules(sampleRule, -1, UCOL_ON, UCOL_TERTIARY, NULL, &err); 695 if(U_FAILURE(err)) { 696 for (index = 0; index < CLONETEST_COLLATOR_COUNT; index++) { 697 ucol_close(someCollators[index]); 698 } 699 log_data_err("Couldn't open one or more collators\n"); 700 return; 701 } 702 703 /* Check the various error & informational states: */ 704 705 /* Null status - just returns NULL */ 706 if (0 != ucol_safeClone(someCollators[0], buffer[0], &bufferSize, 0)) 707 { 708 log_err("FAIL: Cloned Collator failed to deal correctly with null status\n"); 709 } 710 /* error status - should return 0 & keep error the same */ 711 err = U_MEMORY_ALLOCATION_ERROR; 712 if (0 != ucol_safeClone(someCollators[0], buffer[0], &bufferSize, &err) || err != U_MEMORY_ALLOCATION_ERROR) 713 { 714 log_err("FAIL: Cloned Collator failed to deal correctly with incoming error status\n"); 715 } 716 err = U_ZERO_ERROR; 717 718 /* Null buffer size pointer - just returns NULL & set error to U_ILLEGAL_ARGUMENT_ERROR*/ 719 if (0 != ucol_safeClone(someCollators[0], buffer[0], 0, &err) || err != U_ILLEGAL_ARGUMENT_ERROR) 720 { 721 log_err("FAIL: Cloned Collator failed to deal correctly with null bufferSize pointer\n"); 722 } 723 err = U_ZERO_ERROR; 724 725 /* buffer size pointer is 0 - fill in pbufferSize with a size */ 726 bufferSize = 0; 727 if (0 != ucol_safeClone(someCollators[0], buffer[0], &bufferSize, &err) || U_FAILURE(err) || bufferSize <= 0) 728 { 729 log_err("FAIL: Cloned Collator failed a sizing request ('preflighting')\n"); 730 } 731 /* Verify our define is large enough */ 732 if (U_COL_SAFECLONE_BUFFERSIZE < bufferSize) 733 { 734 log_err("FAIL: Pre-calculated buffer size is too small\n"); 735 } 736 /* Verify we can use this run-time calculated size */ 737 if (0 == (col = ucol_safeClone(someCollators[0], buffer[0], &bufferSize, &err)) || U_FAILURE(err)) 738 { 739 log_err("FAIL: Collator can't be cloned with run-time size\n"); 740 } 741 if (col) ucol_close(col); 742 /* size one byte too small - should allocate & let us know */ 743 --bufferSize; 744 if (0 == (col = ucol_safeClone(someCollators[0], 0, &bufferSize, &err)) || err != U_SAFECLONE_ALLOCATED_WARNING) 745 { 746 log_err("FAIL: Cloned Collator failed to deal correctly with too-small buffer size\n"); 747 } 748 if (col) ucol_close(col); 749 err = U_ZERO_ERROR; 750 bufferSize = U_COL_SAFECLONE_BUFFERSIZE; 751 752 753 /* Null buffer pointer - return Collator & set error to U_SAFECLONE_ALLOCATED_ERROR */ 754 if (0 == (col = ucol_safeClone(someCollators[0], 0, &bufferSize, &err)) || err != U_SAFECLONE_ALLOCATED_WARNING) 755 { 756 log_err("FAIL: Cloned Collator failed to deal correctly with null buffer pointer\n"); 757 } 758 if (col) ucol_close(col); 759 err = U_ZERO_ERROR; 760 761 /* Null Collator - return NULL & set U_ILLEGAL_ARGUMENT_ERROR */ 762 if (0 != ucol_safeClone(0, buffer[0], &bufferSize, &err) || err != U_ILLEGAL_ARGUMENT_ERROR) 763 { 764 log_err("FAIL: Cloned Collator failed to deal correctly with null Collator pointer\n"); 765 } 766 767 err = U_ZERO_ERROR; 768 769 /* Test that a cloned collator doesn't accidentally use UCA. */ 770 col=ucol_open("de@collation=phonebook", &err); 771 bufferSize = U_COL_SAFECLONE_BUFFERSIZE; 772 someClonedCollators[0] = ucol_safeClone(col, buffer[0], &bufferSize, &err); 773 doAssert( (ucol_greater(col, umlautUStr, u_strlen(umlautUStr), oeStr, u_strlen(oeStr))), "Original German phonebook collation sorts differently than expected"); 774 doAssert( (ucol_greater(someClonedCollators[0], umlautUStr, u_strlen(umlautUStr), oeStr, u_strlen(oeStr))), "Cloned German phonebook collation sorts differently than expected"); 775 if (!ucol_equals(someClonedCollators[0], col)) { 776 log_err("FAIL: Cloned German phonebook collator is not equal to original.\n"); 777 } 778 ucol_close(col); 779 ucol_close(someClonedCollators[0]); 780 781 err = U_ZERO_ERROR; 782 783 /* change orig & clone & make sure they are independent */ 784 785 for (index = 0; index < CLONETEST_COLLATOR_COUNT; index++) 786 { 787 ucol_setStrength(someCollators[index], UCOL_IDENTICAL); 788 bufferSize = 1; 789 err = U_ZERO_ERROR; 790 ucol_close(ucol_safeClone(someCollators[index], buffer[index], &bufferSize, &err)); 791 if (err != U_SAFECLONE_ALLOCATED_WARNING) { 792 log_err("FAIL: collator number %d was not allocated.\n", index); 793 log_err("FAIL: status of Collator[%d] is %d (hex: %x).\n", index, err, err); 794 } 795 796 bufferSize = U_COL_SAFECLONE_BUFFERSIZE; 797 err = U_ZERO_ERROR; 798 someClonedCollators[index] = ucol_safeClone(someCollators[index], buffer[index], &bufferSize, &err); 799 if (someClonedCollators[index] == NULL 800 || someClonedCollators[index] < (UCollator *)buffer[index] 801 || someClonedCollators[index] > (UCollator *)(buffer[index]+(U_COL_SAFECLONE_BUFFERSIZE-1))) 802 { 803 log_err("FAIL: Cloned collator didn't use provided buffer.\n"); 804 return; 805 } 806 if (!ucol_equals(someClonedCollators[index], someCollators[index])) { 807 log_err("FAIL: Cloned collator is not equal to original at index = %d.\n", index); 808 } 809 810 /* Check the usability */ 811 ucol_setStrength(someCollators[index], UCOL_PRIMARY); 812 ucol_setAttribute(someCollators[index], UCOL_CASE_LEVEL, UCOL_OFF, &err); 813 814 doAssert( (ucol_equal(someCollators[index], test1, u_strlen(test1), test2, u_strlen(test2))), "Result should be \"abcda\" == \"abCda\""); 815 816 /* Close the original to make sure that the clone is usable. */ 817 ucol_close(someCollators[index]); 818 819 ucol_setStrength(someClonedCollators[index], UCOL_TERTIARY); 820 ucol_setAttribute(someClonedCollators[index], UCOL_CASE_LEVEL, UCOL_OFF, &err); 821 doAssert( (ucol_greater(someClonedCollators[index], test1, u_strlen(test1), test2, u_strlen(test2))), "Result should be \"abCda\" >>> \"abcda\" "); 822 823 ucol_close(someClonedCollators[index]); 824 } 825 } 826 827 void TestCloneBinary(){ 828 UErrorCode err = U_ZERO_ERROR; 829 UCollator * col = ucol_open("en_US", &err); 830 UCollator * c; 831 int32_t size; 832 uint8_t * buffer; 833 834 if (U_FAILURE(err)) { 835 log_data_err("Couldn't open collator. Error: %s\n", u_errorName(err)); 836 return; 837 } 838 839 size = ucol_cloneBinary(col, NULL, 0, &err); 840 if(size==0 || err!=U_BUFFER_OVERFLOW_ERROR) { 841 log_err("ucol_cloneBinary - couldn't check size. Error: %s\n", u_errorName(err)); 842 return; 843 } 844 err = U_ZERO_ERROR; 845 846 buffer = (uint8_t *) malloc(size); 847 ucol_cloneBinary(col, buffer, size, &err); 848 if(U_FAILURE(err)) { 849 log_err("ucol_cloneBinary - couldn't clone.. Error: %s\n", u_errorName(err)); 850 free(buffer); 851 return; 852 } 853 854 /* how to check binary result ? */ 855 856 c = ucol_openBinary(buffer, size, col, &err); 857 if(U_FAILURE(err)) { 858 log_err("ucol_openBinary failed. Error: %s\n", u_errorName(err)); 859 } else { 860 UChar t[] = {0x41, 0x42, 0x43, 0}; /* ABC */ 861 uint8_t *k1, *k2; 862 int l1, l2; 863 l1 = ucol_getSortKey(col, t, -1, NULL,0); 864 l2 = ucol_getSortKey(c, t, -1, NULL,0); 865 k1 = (uint8_t *) malloc(sizeof(uint8_t) * l1); 866 k2 = (uint8_t *) malloc(sizeof(uint8_t) * l2); 867 ucol_getSortKey(col, t, -1, k1, l1); 868 ucol_getSortKey(col, t, -1, k2, l2); 869 if (strcmp((char *)k1,(char *)k2) != 0){ 870 log_err("ucol_openBinary - new collator should equal to old one\n"); 871 }; 872 free(k1); 873 free(k2); 874 } 875 free(buffer); 876 ucol_close(c); 877 ucol_close(col); 878 } 879 /* 880 TestOpenVsOpenRules ensures that collators from ucol_open and ucol_openRules 881 will generate identical sort keys 882 */ 883 void TestOpenVsOpenRules(){ 884 885 /* create an array of all the locales */ 886 int32_t numLocales = uloc_countAvailable(); 887 int32_t sizeOfStdSet; 888 uint32_t adder; 889 UChar str[41]; /* create an array of UChar of size maximum strSize + 1 */ 890 USet *stdSet; 891 char* curLoc; 892 UCollator * c1; 893 UCollator * c2; 894 const UChar* rules; 895 int32_t rulesLength; 896 int32_t sortKeyLen1, sortKeyLen2; 897 uint8_t *sortKey1 = NULL, *sortKey2 = NULL; 898 ULocaleData *uld; 899 int32_t x, y, z; 900 USet *eSet; 901 int32_t eSize; 902 int strSize; 903 904 UErrorCode err = U_ZERO_ERROR; 905 906 /* create a set of standard characters that aren't very interesting... 907 and then we can find some interesting ones later */ 908 909 stdSet = uset_open(0x61, 0x7A); 910 uset_addRange(stdSet, 0x41, 0x5A); 911 uset_addRange(stdSet, 0x30, 0x39); 912 sizeOfStdSet = uset_size(stdSet); 913 914 adder = 1; 915 if(QUICK) 916 { 917 adder = 10; 918 } 919 920 for(x = 0; x < numLocales; x+=adder){ 921 curLoc = (char *)uloc_getAvailable(x); 922 log_verbose("Processing %s\n", curLoc); 923 924 /* create a collator the normal API way */ 925 c1 = ucol_open(curLoc, &err); 926 if (U_FAILURE(err)) { 927 log_err("ERROR: Normal collation creation failed with locale: %s : %s\n", curLoc, myErrorName(err)); 928 return; 929 } 930 931 /* grab the rules */ 932 rules = ucol_getRules(c1, &rulesLength); 933 934 /* use those rules to create a collator from rules */ 935 c2 = ucol_openRules(rules, rulesLength, UCOL_DEFAULT, UCOL_DEFAULT_STRENGTH, NULL, &err); 936 if (U_FAILURE(err)) { 937 log_err("ERROR: Creating collator from rules failed with locale: %s : %s\n", curLoc, myErrorName(err)); 938 return; 939 } 940 941 uld = ulocdata_open(curLoc, &err); 942 943 /*now that we have some collators, we get several strings */ 944 945 for(y = 0; y < 5; y++){ 946 947 /* get a set of ALL the characters in this locale */ 948 eSet = ulocdata_getExemplarSet(uld, NULL, 0, ULOCDATA_ES_STANDARD, &err); 949 eSize = uset_size(eSet); 950 951 /* make a string with these characters in it */ 952 strSize = (rand()%40) + 1; 953 954 for(z = 0; z < strSize; z++){ 955 str[z] = uset_charAt(eSet, rand()%eSize); 956 } 957 958 /* change the set to only include 'abnormal' characters (not A-Z, a-z, 0-9 */ 959 uset_removeAll(eSet, stdSet); 960 eSize = uset_size(eSet); 961 962 /* if there are some non-normal characters left, put a few into the string, just to make sure we have some */ 963 if(eSize > 0){ 964 str[2%strSize] = uset_charAt(eSet, rand()%eSize); 965 str[3%strSize] = uset_charAt(eSet, rand()%eSize); 966 str[5%strSize] = uset_charAt(eSet, rand()%eSize); 967 str[10%strSize] = uset_charAt(eSet, rand()%eSize); 968 str[13%strSize] = uset_charAt(eSet, rand()%eSize); 969 } 970 /* terminate the string */ 971 str[strSize-1] = '\0'; 972 log_verbose("String used: %S\n", str); 973 974 /* get sort keys for both of them, and check that the keys are identicle */ 975 sortKeyLen1 = ucol_getSortKey(c1, str, u_strlen(str), NULL, 0); 976 sortKey1 = (uint8_t*)malloc(sizeof(uint8_t) * (sortKeyLen1 + 1)); 977 /*memset(sortKey1, 0xFE, sortKeyLen1);*/ 978 ucol_getSortKey(c1, str, u_strlen(str), sortKey1, sortKeyLen1 + 1); 979 980 sortKeyLen2 = ucol_getSortKey(c2, str, u_strlen(str), NULL, 0); 981 sortKey2 = (uint8_t*)malloc(sizeof(uint8_t) * (sortKeyLen2 + 1)); 982 /*memset(sortKey2, 0xFE, sortKeyLen2);*/ 983 ucol_getSortKey(c2, str, u_strlen(str), sortKey2, sortKeyLen2 + 1); 984 985 /* Check that the lengths are the same */ 986 doAssert((sortKeyLen1 == sortKeyLen2), "Sort key lengths do not match."); 987 988 /* check that the keys are the same */ 989 doAssert((memcmp(sortKey1, sortKey2, sortKeyLen1) == 0), "Keys are not equivalent"); 990 991 /* clean up after each string */ 992 free(sortKey1); 993 free(sortKey2); 994 uset_close(eSet); 995 } 996 /* clean up after each locale */ 997 ulocdata_close(uld); 998 ucol_close(c1); 999 ucol_close(c2); 1000 } 1001 /* final clean up */ 1002 uset_close(stdSet); 1003 } 1004 /* 1005 ---------------------------------------------------------------------------- 1006 ctor -- Tests the getSortKey 1007 */ 1008 void TestSortKey() 1009 { 1010 uint8_t *sortk1 = NULL, *sortk2 = NULL, *sortk3 = NULL, *sortkEmpty = NULL; 1011 int32_t sortklen, osortklen; 1012 uint32_t toStringLen=0; 1013 UCollator *col; 1014 UChar *test1, *test2, *test3; 1015 UErrorCode status = U_ZERO_ERROR; 1016 char toStringBuffer[256], *resultP; 1017 1018 1019 uint8_t s1[] = { 0x9f, 0x00 }; 1020 uint8_t s2[] = { 0x61, 0x00 }; 1021 int strcmpResult; 1022 1023 strcmpResult = strcmp((const char *)s1, (const char *)s2); 1024 log_verbose("strcmp(0x9f..., 0x61...) = %d\n", strcmpResult); 1025 1026 if(strcmpResult <= 0) { 1027 log_err("ERR: expected strcmp(\"9f 00\", \"61 00\") to be >=0 (GREATER).. got %d. Calling strcmp() for sortkeys may not work! \n", 1028 strcmpResult); 1029 } 1030 1031 1032 log_verbose("testing SortKey begins...\n"); 1033 /* this is supposed to open default date format, but later on it treats it like it is "en_US" 1034 - very bad if you try to run the tests on machine where default locale is NOT "en_US" */ 1035 /* col = ucol_open(NULL, &status); */ 1036 col = ucol_open("en_US", &status); 1037 if (U_FAILURE(status)) { 1038 log_err_status(status, "ERROR: Default collation creation failed.: %s\n", myErrorName(status)); 1039 return; 1040 } 1041 1042 1043 if(ucol_getStrength(col) != UCOL_DEFAULT_STRENGTH) 1044 { 1045 log_err("ERROR: default collation did not have UCOL_DEFAULT_STRENGTH !\n"); 1046 } 1047 /* Need to use identical strength */ 1048 ucol_setAttribute(col, UCOL_STRENGTH, UCOL_IDENTICAL, &status); 1049 1050 test1=(UChar*)malloc(sizeof(UChar) * 6); 1051 test2=(UChar*)malloc(sizeof(UChar) * 6); 1052 test3=(UChar*)malloc(sizeof(UChar) * 6); 1053 1054 memset(test1,0xFE, sizeof(UChar)*6); 1055 memset(test2,0xFE, sizeof(UChar)*6); 1056 memset(test3,0xFE, sizeof(UChar)*6); 1057 1058 1059 u_uastrcpy(test1, "Abcda"); 1060 u_uastrcpy(test2, "abcda"); 1061 u_uastrcpy(test3, "abcda"); 1062 1063 log_verbose("Use tertiary comparison level testing ....\n"); 1064 1065 sortklen=ucol_getSortKey(col, test1, u_strlen(test1), NULL, 0); 1066 sortk1=(uint8_t*)malloc(sizeof(uint8_t) * (sortklen+1)); 1067 memset(sortk1,0xFE, sortklen); 1068 ucol_getSortKey(col, test1, u_strlen(test1), sortk1, sortklen+1); 1069 1070 sortklen=ucol_getSortKey(col, test2, u_strlen(test2), NULL, 0); 1071 sortk2=(uint8_t*)malloc(sizeof(uint8_t) * (sortklen+1)); 1072 memset(sortk2,0xFE, sortklen); 1073 ucol_getSortKey(col, test2, u_strlen(test2), sortk2, sortklen+1); 1074 1075 osortklen = sortklen; 1076 sortklen=ucol_getSortKey(col, test2, u_strlen(test3), NULL, 0); 1077 sortk3=(uint8_t*)malloc(sizeof(uint8_t) * (sortklen+1)); 1078 memset(sortk3,0xFE, sortklen); 1079 ucol_getSortKey(col, test2, u_strlen(test2), sortk3, sortklen+1); 1080 1081 doAssert( (sortklen == osortklen), "Sortkey length should be the same (abcda, abcda)"); 1082 1083 doAssert( (memcmp(sortk1, sortk2, sortklen) > 0), "Result should be \"Abcda\" > \"abcda\""); 1084 doAssert( (memcmp(sortk2, sortk1, sortklen) < 0), "Result should be \"abcda\" < \"Abcda\""); 1085 doAssert( (memcmp(sortk2, sortk3, sortklen) == 0), "Result should be \"abcda\" == \"abcda\""); 1086 1087 resultP = ucol_sortKeyToString(col, sortk3, toStringBuffer, &toStringLen); 1088 doAssert( (resultP != 0), "sortKeyToString failed!"); 1089 1090 #if 1 /* verobse log of sortkeys */ 1091 { 1092 char junk2[1000]; 1093 char junk3[1000]; 1094 int i; 1095 1096 strcpy(junk2, "abcda[2] "); 1097 strcpy(junk3, " abcda[3] "); 1098 1099 for(i=0;i<sortklen;i++) 1100 { 1101 sprintf(junk2+strlen(junk2), "%02X ",(int)( 0xFF & sortk2[i])); 1102 sprintf(junk3+strlen(junk3), "%02X ",(int)( 0xFF & sortk3[i])); 1103 } 1104 1105 log_verbose("%s\n", junk2); 1106 log_verbose("%s\n", junk3); 1107 } 1108 #endif 1109 1110 free(sortk1); 1111 free(sortk2); 1112 free(sortk3); 1113 1114 log_verbose("Use secondary comparision level testing ...\n"); 1115 ucol_setStrength(col, UCOL_SECONDARY); 1116 sortklen=ucol_getSortKey(col, test1, u_strlen(test1), NULL, 0); 1117 sortk1=(uint8_t*)malloc(sizeof(uint8_t) * (sortklen+1)); 1118 ucol_getSortKey(col, test1, u_strlen(test1), sortk1, sortklen+1); 1119 sortklen=ucol_getSortKey(col, test2, u_strlen(test2), NULL, 0); 1120 sortk2=(uint8_t*)malloc(sizeof(uint8_t) * (sortklen+1)); 1121 ucol_getSortKey(col, test2, u_strlen(test2), sortk2, sortklen+1); 1122 1123 doAssert( !(memcmp(sortk1, sortk2, sortklen) > 0), "Result should be \"Abcda\" == \"abcda\""); 1124 doAssert( !(memcmp(sortk2, sortk1, sortklen) < 0), "Result should be \"abcda\" == \"Abcda\""); 1125 doAssert( (memcmp(sortk1, sortk2, sortklen) == 0), "Result should be \"abcda\" == \"abcda\""); 1126 1127 log_verbose("getting sortkey for an empty string\n"); 1128 ucol_setAttribute(col, UCOL_STRENGTH, UCOL_TERTIARY, &status); 1129 sortklen = ucol_getSortKey(col, test1, 0, NULL, 0); 1130 sortkEmpty = (uint8_t*)malloc(sizeof(uint8_t) * sortklen+1); 1131 sortklen = ucol_getSortKey(col, test1, 0, sortkEmpty, sortklen+1); 1132 if(sortklen != 3 || sortkEmpty[0] != 1 || sortkEmpty[0] != 1 || sortkEmpty[2] != 0) { 1133 log_err("Empty string generated wrong sortkey!\n"); 1134 } 1135 free(sortkEmpty); 1136 1137 log_verbose("testing passing invalid string\n"); 1138 sortklen = ucol_getSortKey(col, NULL, 0, NULL, 0); 1139 if(sortklen != 0) { 1140 log_err("Invalid string didn't return sortkey size of 0\n"); 1141 } 1142 1143 1144 log_verbose("testing sortkey ends...\n"); 1145 ucol_close(col); 1146 free(test1); 1147 free(test2); 1148 free(test3); 1149 free(sortk1); 1150 free(sortk2); 1151 1152 } 1153 void TestHashCode() 1154 { 1155 uint8_t *sortk1, *sortk2, *sortk3; 1156 int32_t sortk1len, sortk2len, sortk3len; 1157 UCollator *col; 1158 UChar *test1, *test2, *test3; 1159 UErrorCode status = U_ZERO_ERROR; 1160 log_verbose("testing getHashCode begins...\n"); 1161 col = ucol_open("en_US", &status); 1162 if (U_FAILURE(status)) { 1163 log_err_status(status, "ERROR: Default collation creation failed.: %s\n", myErrorName(status)); 1164 return; 1165 } 1166 test1=(UChar*)malloc(sizeof(UChar) * 6); 1167 test2=(UChar*)malloc(sizeof(UChar) * 6); 1168 test3=(UChar*)malloc(sizeof(UChar) * 6); 1169 u_uastrcpy(test1, "Abcda"); 1170 u_uastrcpy(test2, "abcda"); 1171 u_uastrcpy(test3, "abcda"); 1172 1173 log_verbose("Use tertiary comparison level testing ....\n"); 1174 sortk1len=ucol_getSortKey(col, test1, u_strlen(test1), NULL, 0); 1175 sortk1=(uint8_t*)malloc(sizeof(uint8_t) * (sortk1len+1)); 1176 ucol_getSortKey(col, test1, u_strlen(test1), sortk1, sortk1len+1); 1177 sortk2len=ucol_getSortKey(col, test2, u_strlen(test2), NULL, 0); 1178 sortk2=(uint8_t*)malloc(sizeof(uint8_t) * (sortk2len+1)); 1179 ucol_getSortKey(col, test2, u_strlen(test2), sortk2, sortk2len+1); 1180 sortk3len=ucol_getSortKey(col, test2, u_strlen(test3), NULL, 0); 1181 sortk3=(uint8_t*)malloc(sizeof(uint8_t) * (sortk3len+1)); 1182 ucol_getSortKey(col, test2, u_strlen(test2), sortk3, sortk3len+1); 1183 1184 1185 log_verbose("ucol_hashCode() testing ...\n"); 1186 1187 doAssert( ucol_keyHashCode(sortk1, sortk1len) != ucol_keyHashCode(sortk2, sortk2len), "Hash test1 result incorrect" ); 1188 doAssert( !(ucol_keyHashCode(sortk1, sortk1len) == ucol_keyHashCode(sortk2, sortk2len)), "Hash test2 result incorrect" ); 1189 doAssert( ucol_keyHashCode(sortk2, sortk2len) == ucol_keyHashCode(sortk3, sortk3len), "Hash result not equal" ); 1190 1191 log_verbose("hashCode tests end.\n"); 1192 ucol_close(col); 1193 free(sortk1); 1194 free(sortk2); 1195 free(sortk3); 1196 free(test1); 1197 free(test2); 1198 free(test3); 1199 1200 1201 } 1202 /* 1203 *---------------------------------------------------------------------------- 1204 * Tests the UCollatorElements API. 1205 * 1206 */ 1207 void TestElemIter() 1208 { 1209 int32_t offset; 1210 int32_t order1, order2, order3; 1211 UChar *testString1, *testString2; 1212 UCollator *col; 1213 UCollationElements *iterator1, *iterator2, *iterator3; 1214 UErrorCode status = U_ZERO_ERROR; 1215 log_verbose("testing UCollatorElements begins...\n"); 1216 col = ucol_open("en_US", &status); 1217 ucol_setAttribute(col, UCOL_NORMALIZATION_MODE, UCOL_OFF, &status); 1218 if (U_FAILURE(status)) { 1219 log_err_status(status, "ERROR: Default collation creation failed.: %s\n", myErrorName(status)); 1220 return; 1221 } 1222 1223 testString1=(UChar*)malloc(sizeof(UChar) * 150); 1224 testString2=(UChar*)malloc(sizeof(UChar) * 150); 1225 u_uastrcpy(testString1, "XFILE What subset of all possible test cases has the highest probability of detecting the most errors?"); 1226 u_uastrcpy(testString2, "Xf_ile What subset of all possible test cases has the lowest probability of detecting the least errors?"); 1227 1228 log_verbose("Constructors and comparison testing....\n"); 1229 1230 iterator1 = ucol_openElements(col, testString1, u_strlen(testString1), &status); 1231 if(U_FAILURE(status)) { 1232 log_err("ERROR: Default collationElement iterator creation failed.: %s\n", myErrorName(status)); 1233 ucol_close(col); 1234 return; 1235 } 1236 else{ log_verbose("PASS: Default collationElement iterator1 creation passed\n");} 1237 1238 iterator2 = ucol_openElements(col, testString1, u_strlen(testString1), &status); 1239 if(U_FAILURE(status)) { 1240 log_err("ERROR: Default collationElement iterator creation failed.: %s\n", myErrorName(status)); 1241 ucol_close(col); 1242 return; 1243 } 1244 else{ log_verbose("PASS: Default collationElement iterator2 creation passed\n");} 1245 1246 iterator3 = ucol_openElements(col, testString2, u_strlen(testString2), &status); 1247 if(U_FAILURE(status)) { 1248 log_err("ERROR: Default collationElement iterator creation failed.: %s\n", myErrorName(status)); 1249 ucol_close(col); 1250 return; 1251 } 1252 else{ log_verbose("PASS: Default collationElement iterator3 creation passed\n");} 1253 1254 offset=ucol_getOffset(iterator1); 1255 ucol_setOffset(iterator1, 6, &status); 1256 if (U_FAILURE(status)) { 1257 log_err("Error in setOffset for UCollatorElements iterator.: %s\n", myErrorName(status)); 1258 return; 1259 } 1260 if(ucol_getOffset(iterator1)==6) 1261 log_verbose("setOffset and getOffset working fine\n"); 1262 else{ 1263 log_err("error in set and get Offset got %d instead of 6\n", ucol_getOffset(iterator1)); 1264 } 1265 1266 ucol_setOffset(iterator1, 0, &status); 1267 order1 = ucol_next(iterator1, &status); 1268 if (U_FAILURE(status)) { 1269 log_err("Somehow ran out of memory stepping through the iterator1.: %s\n", myErrorName(status)); 1270 return; 1271 } 1272 order2=ucol_getOffset(iterator2); 1273 doAssert((order1 != order2), "The first iterator advance failed"); 1274 order2 = ucol_next(iterator2, &status); 1275 if (U_FAILURE(status)) { 1276 log_err("Somehow ran out of memory stepping through the iterator2.: %s\n", myErrorName(status)); 1277 return; 1278 } 1279 order3 = ucol_next(iterator3, &status); 1280 if (U_FAILURE(status)) { 1281 log_err("Somehow ran out of memory stepping through the iterator3.: %s\n", myErrorName(status)); 1282 return; 1283 } 1284 1285 doAssert((order1 == order2), "The second iterator advance failed should be the same as first one"); 1286 1287 doAssert( (ucol_primaryOrder(order1) == ucol_primaryOrder(order3)), "The primary orders should be identical"); 1288 doAssert( (ucol_secondaryOrder(order1) == ucol_secondaryOrder(order3)), "The secondary orders should be identical"); 1289 doAssert( (ucol_tertiaryOrder(order1) == ucol_tertiaryOrder(order3)), "The tertiary orders should be identical"); 1290 1291 order1=ucol_next(iterator1, &status); 1292 if (U_FAILURE(status)) { 1293 log_err("Somehow ran out of memory stepping through the iterator2.: %s\n", myErrorName(status)); 1294 return; 1295 } 1296 order3=ucol_next(iterator3, &status); 1297 if (U_FAILURE(status)) { 1298 log_err("Somehow ran out of memory stepping through the iterator2.: %s\n", myErrorName(status)); 1299 return; 1300 } 1301 doAssert( (ucol_primaryOrder(order1) == ucol_primaryOrder(order3)), "The primary orders should be identical"); 1302 doAssert( (ucol_tertiaryOrder(order1) != ucol_tertiaryOrder(order3)), "The tertiary orders should be different"); 1303 1304 order1=ucol_next(iterator1, &status); 1305 if (U_FAILURE(status)) { 1306 log_err("Somehow ran out of memory stepping through the iterator2.: %s\n", myErrorName(status)); 1307 return; 1308 } 1309 order3=ucol_next(iterator3, &status); 1310 if (U_FAILURE(status)) { 1311 log_err("Somehow ran out of memory stepping through the iterator2.: %s\n", myErrorName(status)); 1312 return; 1313 } 1314 /* this here, my friends, is either pure lunacy or something so obsolete that even it's mother 1315 * doesn't care about it. Essentialy, this test complains if secondary values for 'I' and '_' 1316 * are the same. According to the UCA, this is not true. Therefore, remove the test. 1317 * Besides, if primary strengths for two code points are different, it doesn't matter one bit 1318 * what is the relation between secondary or any other strengths. 1319 * killed by weiv 06/11/2002. 1320 */ 1321 /* 1322 doAssert( ((order1 & UCOL_SECONDARYMASK) != (order3 & UCOL_SECONDARYMASK)), "The secondary orders should be different"); 1323 */ 1324 doAssert( (order1 != UCOL_NULLORDER), "Unexpected end of iterator reached"); 1325 1326 free(testString1); 1327 free(testString2); 1328 ucol_closeElements(iterator1); 1329 ucol_closeElements(iterator2); 1330 ucol_closeElements(iterator3); 1331 ucol_close(col); 1332 1333 log_verbose("testing CollationElementIterator ends...\n"); 1334 } 1335 1336 void TestGetLocale() { 1337 UErrorCode status = U_ZERO_ERROR; 1338 const char *rules = "&a<x<y<z"; 1339 UChar rlz[256] = {0}; 1340 uint32_t rlzLen = u_unescape(rules, rlz, 256); 1341 1342 UCollator *coll = NULL; 1343 const char *locale = NULL; 1344 1345 int32_t i = 0; 1346 1347 /* Now that the collation tree is separate, actual==valid at all times. [alan] */ 1348 static const struct { 1349 const char* requestedLocale; 1350 const char* validLocale; 1351 const char* actualLocale; 1352 } testStruct[] = { 1353 { "sr_RS", "sr_Cyrl_RS", "ru" }, 1354 { "sh_YU", "sr_Latn_RS", "hr" }, /* this used to be sh, but now sh collation aliases hr */ 1355 { "en_BE_FOO", "en_BE", "en_BE" }, 1356 { "fr_FR_NONEXISTANT", "fr_FR", "fr" } 1357 }; 1358 1359 /* test opening collators for different locales */ 1360 for(i = 0; i<sizeof(testStruct)/sizeof(testStruct[0]); i++) { 1361 status = U_ZERO_ERROR; 1362 coll = ucol_open(testStruct[i].requestedLocale, &status); 1363 if(U_FAILURE(status)) { 1364 log_err_status(status, "Failed to open collator for %s with %s\n", testStruct[i].requestedLocale, u_errorName(status)); 1365 ucol_close(coll); 1366 continue; 1367 } 1368 locale = ucol_getLocaleByType(coll, ULOC_REQUESTED_LOCALE, &status); 1369 if(strcmp(locale, testStruct[i].requestedLocale) != 0) { 1370 log_err("[Coll %s]: Error in requested locale, expected %s, got %s\n", testStruct[i].requestedLocale, testStruct[i].requestedLocale, locale); 1371 } 1372 locale = ucol_getLocaleByType(coll, ULOC_VALID_LOCALE, &status); 1373 if(strcmp(locale, testStruct[i].validLocale) != 0) { 1374 log_err("[Coll %s]: Error in valid locale, expected %s, got %s\n", testStruct[i].requestedLocale, testStruct[i].validLocale, locale); 1375 } 1376 locale = ucol_getLocaleByType(coll, ULOC_ACTUAL_LOCALE, &status); 1377 if(strcmp(locale, testStruct[i].actualLocale) != 0) { 1378 log_err("[Coll %s]: Error in actual locale, expected %s, got %s\n", testStruct[i].requestedLocale, testStruct[i].actualLocale, locale); 1379 } 1380 ucol_close(coll); 1381 } 1382 1383 /* completely non-existant locale for collator should get a default collator */ 1384 { 1385 UCollator *defaultColl = ucol_open(NULL, &status); 1386 coll = ucol_open("blahaha", &status); 1387 if(U_SUCCESS(status)) { 1388 if(strcmp(ucol_getLocaleByType(coll, ULOC_REQUESTED_LOCALE, &status), "blahaha")) { 1389 log_err("Nonexisting locale didn't preserve the requested locale\n"); 1390 } 1391 if(strcmp(ucol_getLocaleByType(coll, ULOC_VALID_LOCALE, &status), 1392 ucol_getLocaleByType(defaultColl, ULOC_VALID_LOCALE, &status))) { 1393 log_err("Valid locale for nonexisting locale locale collator differs " 1394 "from valid locale for default collator\n"); 1395 } 1396 if(strcmp(ucol_getLocaleByType(coll, ULOC_ACTUAL_LOCALE, &status), 1397 ucol_getLocaleByType(defaultColl, ULOC_ACTUAL_LOCALE, &status))) { 1398 log_err("Actual locale for nonexisting locale locale collator differs " 1399 "from actual locale for default collator\n"); 1400 } 1401 ucol_close(coll); 1402 ucol_close(defaultColl); 1403 } else { 1404 log_data_err("Couldn't open collators\n"); 1405 } 1406 } 1407 1408 1409 1410 /* collator instantiated from rules should have all three locales NULL */ 1411 coll = ucol_openRules(rlz, rlzLen, UCOL_DEFAULT, UCOL_DEFAULT, NULL, &status); 1412 locale = ucol_getLocaleByType(coll, ULOC_REQUESTED_LOCALE, &status); 1413 if(locale != NULL) { 1414 log_err("For collator instantiated from rules, requested locale returned %s instead of NULL\n", locale); 1415 } 1416 locale = ucol_getLocaleByType(coll, ULOC_VALID_LOCALE, &status); 1417 if(locale != NULL) { 1418 log_err("For collator instantiated from rules, valid locale returned %s instead of NULL\n", locale); 1419 } 1420 locale = ucol_getLocaleByType(coll, ULOC_ACTUAL_LOCALE, &status); 1421 if(locale != NULL) { 1422 log_err("For collator instantiated from rules, actual locale returned %s instead of NULL\n", locale); 1423 } 1424 ucol_close(coll); 1425 1426 } 1427 1428 1429 void TestGetAll() 1430 { 1431 int32_t i, count; 1432 count=ucol_countAvailable(); 1433 /* use something sensible w/o hardcoding the count */ 1434 if(count < 0){ 1435 log_err("Error in countAvailable(), it returned %d\n", count); 1436 } 1437 else{ 1438 log_verbose("PASS: countAvailable() successful, it returned %d\n", count); 1439 } 1440 for(i=0;i<count;i++) 1441 log_verbose("%s\n", ucol_getAvailable(i)); 1442 1443 1444 } 1445 1446 1447 struct teststruct { 1448 const char *original; 1449 uint8_t key[256]; 1450 } ; 1451 1452 static int compare_teststruct(const void *string1, const void *string2) { 1453 return(strcmp((const char *)((struct teststruct *)string1)->key, (const char *)((struct teststruct *)string2)->key)); 1454 } 1455 1456 void TestBounds() { 1457 UErrorCode status = U_ZERO_ERROR; 1458 1459 UCollator *coll = ucol_open("sh", &status); 1460 1461 uint8_t sortkey[512], lower[512], upper[512]; 1462 UChar buffer[512]; 1463 1464 static const char * const test[] = { 1465 "John Smith", 1466 "JOHN SMITH", 1467 "john SMITH", 1468 "j\\u00F6hn sm\\u00EFth", 1469 "J\\u00F6hn Sm\\u00EFth", 1470 "J\\u00D6HN SM\\u00CFTH", 1471 "john smithsonian", 1472 "John Smithsonian", 1473 }; 1474 1475 struct teststruct tests[] = { 1476 {"\\u010CAKI MIHALJ" } , 1477 {"\\u010CAKI MIHALJ" } , 1478 {"\\u010CAKI PIRO\\u0160KA" }, 1479 {"\\u010CABAI ANDRIJA" } , 1480 {"\\u010CABAI LAJO\\u0160" } , 1481 {"\\u010CABAI MARIJA" } , 1482 {"\\u010CABAI STEVAN" } , 1483 {"\\u010CABAI STEVAN" } , 1484 {"\\u010CABARKAPA BRANKO" } , 1485 {"\\u010CABARKAPA MILENKO" } , 1486 {"\\u010CABARKAPA MIROSLAV" } , 1487 {"\\u010CABARKAPA SIMO" } , 1488 {"\\u010CABARKAPA STANKO" } , 1489 {"\\u010CABARKAPA TAMARA" } , 1490 {"\\u010CABARKAPA TOMA\\u0160" } , 1491 {"\\u010CABDARI\\u0106 NIKOLA" } , 1492 {"\\u010CABDARI\\u0106 ZORICA" } , 1493 {"\\u010CABI NANDOR" } , 1494 {"\\u010CABOVI\\u0106 MILAN" } , 1495 {"\\u010CABRADI AGNEZIJA" } , 1496 {"\\u010CABRADI IVAN" } , 1497 {"\\u010CABRADI JELENA" } , 1498 {"\\u010CABRADI LJUBICA" } , 1499 {"\\u010CABRADI STEVAN" } , 1500 {"\\u010CABRDA MARTIN" } , 1501 {"\\u010CABRILO BOGDAN" } , 1502 {"\\u010CABRILO BRANISLAV" } , 1503 {"\\u010CABRILO LAZAR" } , 1504 {"\\u010CABRILO LJUBICA" } , 1505 {"\\u010CABRILO SPASOJA" } , 1506 {"\\u010CADE\\u0160 ZDENKA" } , 1507 {"\\u010CADESKI BLAGOJE" } , 1508 {"\\u010CADOVSKI VLADIMIR" } , 1509 {"\\u010CAGLJEVI\\u0106 TOMA" } , 1510 {"\\u010CAGOROVI\\u0106 VLADIMIR" } , 1511 {"\\u010CAJA VANKA" } , 1512 {"\\u010CAJI\\u0106 BOGOLJUB" } , 1513 {"\\u010CAJI\\u0106 BORISLAV" } , 1514 {"\\u010CAJI\\u0106 RADOSLAV" } , 1515 {"\\u010CAK\\u0160IRAN MILADIN" } , 1516 {"\\u010CAKAN EUGEN" } , 1517 {"\\u010CAKAN EVGENIJE" } , 1518 {"\\u010CAKAN IVAN" } , 1519 {"\\u010CAKAN JULIJAN" } , 1520 {"\\u010CAKAN MIHAJLO" } , 1521 {"\\u010CAKAN STEVAN" } , 1522 {"\\u010CAKAN VLADIMIR" } , 1523 {"\\u010CAKAN VLADIMIR" } , 1524 {"\\u010CAKAN VLADIMIR" } , 1525 {"\\u010CAKARA ANA" } , 1526 {"\\u010CAKAREVI\\u0106 MOMIR" } , 1527 {"\\u010CAKAREVI\\u0106 NEDELJKO" } , 1528 {"\\u010CAKI \\u0160ANDOR" } , 1529 {"\\u010CAKI AMALIJA" } , 1530 {"\\u010CAKI ANDRA\\u0160" } , 1531 {"\\u010CAKI LADISLAV" } , 1532 {"\\u010CAKI LAJO\\u0160" } , 1533 {"\\u010CAKI LASLO" } , 1534 }; 1535 1536 1537 1538 int32_t i = 0, j = 0, k = 0, buffSize = 0, skSize = 0, lowerSize = 0, upperSize = 0; 1539 int32_t arraySize = sizeof(tests)/sizeof(tests[0]); 1540 1541 if(U_SUCCESS(status) && coll) { 1542 for(i = 0; i<arraySize; i++) { 1543 buffSize = u_unescape(tests[i].original, buffer, 512); 1544 skSize = ucol_getSortKey(coll, buffer, buffSize, tests[i].key, 512); 1545 } 1546 1547 qsort(tests, arraySize, sizeof(struct teststruct), compare_teststruct); 1548 1549 for(i = 0; i < arraySize-1; i++) { 1550 for(j = i+1; j < arraySize; j++) { 1551 lowerSize = ucol_getBound(tests[i].key, -1, UCOL_BOUND_LOWER, 1, lower, 512, &status); 1552 upperSize = ucol_getBound(tests[j].key, -1, UCOL_BOUND_UPPER, 1, upper, 512, &status); 1553 for(k = i; k <= j; k++) { 1554 if(strcmp((const char *)lower, (const char *)tests[k].key) > 0) { 1555 log_err("Problem with lower! j = %i (%s vs %s)\n", k, tests[k].original, tests[i].original); 1556 } 1557 if(strcmp((const char *)upper, (const char *)tests[k].key) <= 0) { 1558 log_err("Problem with upper! j = %i (%s vs %s)\n", k, tests[k].original, tests[j].original); 1559 } 1560 } 1561 } 1562 } 1563 1564 1565 #if 0 1566 for(i = 0; i < 1000; i++) { 1567 lowerRND = (rand()/(RAND_MAX/arraySize)); 1568 upperRND = lowerRND + (rand()/(RAND_MAX/(arraySize-lowerRND))); 1569 1570 lowerSize = ucol_getBound(tests[lowerRND].key, -1, UCOL_BOUND_LOWER, 1, lower, 512, &status); 1571 upperSize = ucol_getBound(tests[upperRND].key, -1, UCOL_BOUND_UPPER_LONG, 1, upper, 512, &status); 1572 1573 for(j = lowerRND; j<=upperRND; j++) { 1574 if(strcmp(lower, tests[j].key) > 0) { 1575 log_err("Problem with lower! j = %i (%s vs %s)\n", j, tests[j].original, tests[lowerRND].original); 1576 } 1577 if(strcmp(upper, tests[j].key) <= 0) { 1578 log_err("Problem with upper! j = %i (%s vs %s)\n", j, tests[j].original, tests[upperRND].original); 1579 } 1580 } 1581 } 1582 #endif 1583 1584 1585 1586 1587 1588 for(i = 0; i<sizeof(test)/sizeof(test[0]); i++) { 1589 buffSize = u_unescape(test[i], buffer, 512); 1590 skSize = ucol_getSortKey(coll, buffer, buffSize, sortkey, 512); 1591 lowerSize = ucol_getBound(sortkey, skSize, UCOL_BOUND_LOWER, 1, lower, 512, &status); 1592 upperSize = ucol_getBound(sortkey, skSize, UCOL_BOUND_UPPER_LONG, 1, upper, 512, &status); 1593 for(j = i+1; j<sizeof(test)/sizeof(test[0]); j++) { 1594 buffSize = u_unescape(test[j], buffer, 512); 1595 skSize = ucol_getSortKey(coll, buffer, buffSize, sortkey, 512); 1596 if(strcmp((const char *)lower, (const char *)sortkey) > 0) { 1597 log_err("Problem with lower! i = %i, j = %i (%s vs %s)\n", i, j, test[i], test[j]); 1598 } 1599 if(strcmp((const char *)upper, (const char *)sortkey) <= 0) { 1600 log_err("Problem with upper! i = %i, j = %i (%s vs %s)\n", i, j, test[i], test[j]); 1601 } 1602 } 1603 } 1604 ucol_close(coll); 1605 } else { 1606 log_data_err("Couldn't open collator\n"); 1607 } 1608 1609 } 1610 1611 static void doOverrunTest(UCollator *coll, const UChar *uString, int32_t strLen) { 1612 int32_t skLen = 0, skLen2 = 0; 1613 uint8_t sortKey[256]; 1614 int32_t i, j; 1615 uint8_t filler = 0xFF; 1616 1617 skLen = ucol_getSortKey(coll, uString, strLen, NULL, 0); 1618 1619 for(i = 0; i < skLen; i++) { 1620 memset(sortKey, filler, 256); 1621 skLen2 = ucol_getSortKey(coll, uString, strLen, sortKey, i); 1622 if(skLen != skLen2) { 1623 log_err("For buffer size %i, got different sortkey length. Expected %i got %i\n", i, skLen, skLen2); 1624 } 1625 for(j = i; j < 256; j++) { 1626 if(sortKey[j] != filler) { 1627 log_err("Something run over index %i\n", j); 1628 break; 1629 } 1630 } 1631 } 1632 } 1633 1634 /* j1865 reports that if a shorter buffer is passed to 1635 * to get sort key, a buffer overrun happens in some 1636 * cases. This test tries to check this. 1637 */ 1638 void TestSortKeyBufferOverrun(void) { 1639 UErrorCode status = U_ZERO_ERROR; 1640 const char* cString = "A very Merry liTTle-lamB.."; 1641 UChar uString[256]; 1642 int32_t strLen = 0; 1643 UCollator *coll = ucol_open("root", &status); 1644 strLen = u_unescape(cString, uString, 256); 1645 1646 if(U_SUCCESS(status)) { 1647 log_verbose("testing non ignorable\n"); 1648 ucol_setAttribute(coll, UCOL_ALTERNATE_HANDLING, UCOL_NON_IGNORABLE, &status); 1649 doOverrunTest(coll, uString, strLen); 1650 1651 log_verbose("testing shifted\n"); 1652 ucol_setAttribute(coll, UCOL_ALTERNATE_HANDLING, UCOL_SHIFTED, &status); 1653 doOverrunTest(coll, uString, strLen); 1654 1655 log_verbose("testing shifted quaternary\n"); 1656 ucol_setAttribute(coll, UCOL_STRENGTH, UCOL_QUATERNARY, &status); 1657 doOverrunTest(coll, uString, strLen); 1658 1659 log_verbose("testing with french secondaries\n"); 1660 ucol_setAttribute(coll, UCOL_FRENCH_COLLATION, UCOL_ON, &status); 1661 ucol_setAttribute(coll, UCOL_STRENGTH, UCOL_TERTIARY, &status); 1662 ucol_setAttribute(coll, UCOL_ALTERNATE_HANDLING, UCOL_NON_IGNORABLE, &status); 1663 doOverrunTest(coll, uString, strLen); 1664 1665 } 1666 ucol_close(coll); 1667 } 1668 1669 static void TestAttribute() 1670 { 1671 UErrorCode error = U_ZERO_ERROR; 1672 UCollator *coll = ucol_open(NULL, &error); 1673 1674 if (U_FAILURE(error)) { 1675 log_err_status(error, "Creation of default collator failed\n"); 1676 return; 1677 } 1678 1679 ucol_setAttribute(coll, UCOL_FRENCH_COLLATION, UCOL_OFF, &error); 1680 if (ucol_getAttribute(coll, UCOL_FRENCH_COLLATION, &error) != UCOL_OFF || 1681 U_FAILURE(error)) { 1682 log_err_status(error, "Setting and retrieving of the french collation failed\n"); 1683 } 1684 1685 ucol_setAttribute(coll, UCOL_FRENCH_COLLATION, UCOL_ON, &error); 1686 if (ucol_getAttribute(coll, UCOL_FRENCH_COLLATION, &error) != UCOL_ON || 1687 U_FAILURE(error)) { 1688 log_err_status(error, "Setting and retrieving of the french collation failed\n"); 1689 } 1690 1691 ucol_setAttribute(coll, UCOL_ALTERNATE_HANDLING, UCOL_SHIFTED, &error); 1692 if (ucol_getAttribute(coll, UCOL_ALTERNATE_HANDLING, &error) != UCOL_SHIFTED || 1693 U_FAILURE(error)) { 1694 log_err_status(error, "Setting and retrieving of the alternate handling failed\n"); 1695 } 1696 1697 ucol_setAttribute(coll, UCOL_ALTERNATE_HANDLING, UCOL_NON_IGNORABLE, &error); 1698 if (ucol_getAttribute(coll, UCOL_ALTERNATE_HANDLING, &error) != UCOL_NON_IGNORABLE || 1699 U_FAILURE(error)) { 1700 log_err_status(error, "Setting and retrieving of the alternate handling failed\n"); 1701 } 1702 1703 ucol_setAttribute(coll, UCOL_CASE_FIRST, UCOL_LOWER_FIRST, &error); 1704 if (ucol_getAttribute(coll, UCOL_CASE_FIRST, &error) != UCOL_LOWER_FIRST || 1705 U_FAILURE(error)) { 1706 log_err_status(error, "Setting and retrieving of the case first attribute failed\n"); 1707 } 1708 1709 ucol_setAttribute(coll, UCOL_CASE_FIRST, UCOL_UPPER_FIRST, &error); 1710 if (ucol_getAttribute(coll, UCOL_CASE_FIRST, &error) != UCOL_UPPER_FIRST || 1711 U_FAILURE(error)) { 1712 log_err_status(error, "Setting and retrieving of the case first attribute failed\n"); 1713 } 1714 1715 ucol_setAttribute(coll, UCOL_CASE_LEVEL, UCOL_ON, &error); 1716 if (ucol_getAttribute(coll, UCOL_CASE_LEVEL, &error) != UCOL_ON || 1717 U_FAILURE(error)) { 1718 log_err_status(error, "Setting and retrieving of the case level attribute failed\n"); 1719 } 1720 1721 ucol_setAttribute(coll, UCOL_CASE_LEVEL, UCOL_OFF, &error); 1722 if (ucol_getAttribute(coll, UCOL_CASE_LEVEL, &error) != UCOL_OFF || 1723 U_FAILURE(error)) { 1724 log_err_status(error, "Setting and retrieving of the case level attribute failed\n"); 1725 } 1726 1727 ucol_setAttribute(coll, UCOL_NORMALIZATION_MODE, UCOL_ON, &error); 1728 if (ucol_getAttribute(coll, UCOL_NORMALIZATION_MODE, &error) != UCOL_ON || 1729 U_FAILURE(error)) { 1730 log_err_status(error, "Setting and retrieving of the normalization on/off attribute failed\n"); 1731 } 1732 1733 ucol_setAttribute(coll, UCOL_NORMALIZATION_MODE, UCOL_OFF, &error); 1734 if (ucol_getAttribute(coll, UCOL_NORMALIZATION_MODE, &error) != UCOL_OFF || 1735 U_FAILURE(error)) { 1736 log_err_status(error, "Setting and retrieving of the normalization on/off attribute failed\n"); 1737 } 1738 1739 ucol_setAttribute(coll, UCOL_STRENGTH, UCOL_PRIMARY, &error); 1740 if (ucol_getAttribute(coll, UCOL_STRENGTH, &error) != UCOL_PRIMARY || 1741 U_FAILURE(error)) { 1742 log_err_status(error, "Setting and retrieving of the collation strength failed\n"); 1743 } 1744 1745 ucol_setAttribute(coll, UCOL_STRENGTH, UCOL_SECONDARY, &error); 1746 if (ucol_getAttribute(coll, UCOL_STRENGTH, &error) != UCOL_SECONDARY || 1747 U_FAILURE(error)) { 1748 log_err_status(error, "Setting and retrieving of the collation strength failed\n"); 1749 } 1750 1751 ucol_setAttribute(coll, UCOL_STRENGTH, UCOL_TERTIARY, &error); 1752 if (ucol_getAttribute(coll, UCOL_STRENGTH, &error) != UCOL_TERTIARY || 1753 U_FAILURE(error)) { 1754 log_err_status(error, "Setting and retrieving of the collation strength failed\n"); 1755 } 1756 1757 ucol_setAttribute(coll, UCOL_STRENGTH, UCOL_QUATERNARY, &error); 1758 if (ucol_getAttribute(coll, UCOL_STRENGTH, &error) != UCOL_QUATERNARY || 1759 U_FAILURE(error)) { 1760 log_err_status(error, "Setting and retrieving of the collation strength failed\n"); 1761 } 1762 1763 ucol_setAttribute(coll, UCOL_STRENGTH, UCOL_IDENTICAL, &error); 1764 if (ucol_getAttribute(coll, UCOL_STRENGTH, &error) != UCOL_IDENTICAL || 1765 U_FAILURE(error)) { 1766 log_err_status(error, "Setting and retrieving of the collation strength failed\n"); 1767 } 1768 1769 ucol_close(coll); 1770 } 1771 1772 void TestGetTailoredSet() { 1773 struct { 1774 const char *rules; 1775 const char *tests[20]; 1776 int32_t testsize; 1777 } setTest[] = { 1778 { "&a < \\u212b", { "\\u212b", "A\\u030a", "\\u00c5" }, 3}, 1779 { "& S < \\u0161 <<< \\u0160", { "\\u0161", "s\\u030C", "\\u0160", "S\\u030C" }, 4} 1780 }; 1781 1782 int32_t i = 0, j = 0; 1783 UErrorCode status = U_ZERO_ERROR; 1784 UParseError pError; 1785 1786 UCollator *coll = NULL; 1787 UChar buff[1024]; 1788 int32_t buffLen = 0; 1789 USet *set = NULL; 1790 1791 for(i = 0; i < sizeof(setTest)/sizeof(setTest[0]); i++) { 1792 buffLen = u_unescape(setTest[i].rules, buff, 1024); 1793 coll = ucol_openRules(buff, buffLen, UCOL_DEFAULT, UCOL_DEFAULT, &pError, &status); 1794 if(U_SUCCESS(status)) { 1795 set = ucol_getTailoredSet(coll, &status); 1796 if(uset_size(set) != setTest[i].testsize) { 1797 log_err("Tailored set size different (%d) than expected (%d)\n", uset_size(set), setTest[i].testsize); 1798 } 1799 for(j = 0; j < setTest[i].testsize; j++) { 1800 buffLen = u_unescape(setTest[i].tests[j], buff, 1024); 1801 if(!uset_containsString(set, buff, buffLen)) { 1802 log_err("Tailored set doesn't contain %s... It should\n", setTest[i].tests[j]); 1803 } 1804 } 1805 uset_close(set); 1806 } else { 1807 log_err_status(status, "Couldn't open collator with rules %s\n", setTest[i].rules); 1808 } 1809 ucol_close(coll); 1810 } 1811 } 1812 1813 static int tMemCmp(const uint8_t *first, const uint8_t *second) { 1814 int32_t firstLen = (int32_t)strlen((const char *)first); 1815 int32_t secondLen = (int32_t)strlen((const char *)second); 1816 return memcmp(first, second, uprv_min(firstLen, secondLen)); 1817 } 1818 static const char * strengthsC[] = { 1819 "UCOL_PRIMARY", 1820 "UCOL_SECONDARY", 1821 "UCOL_TERTIARY", 1822 "UCOL_QUATERNARY", 1823 "UCOL_IDENTICAL" 1824 }; 1825 1826 void TestMergeSortKeys(void) { 1827 UErrorCode status = U_ZERO_ERROR; 1828 UCollator *coll = ucol_open("en", &status); 1829 if(U_SUCCESS(status)) { 1830 1831 const char* cases[] = { 1832 "abc", 1833 "abcd", 1834 "abcde" 1835 }; 1836 uint32_t casesSize = sizeof(cases)/sizeof(cases[0]); 1837 const char* prefix = "foo"; 1838 const char* suffix = "egg"; 1839 char outBuff1[256], outBuff2[256]; 1840 1841 uint8_t **sortkeys = (uint8_t **)malloc(casesSize*sizeof(uint8_t *)); 1842 uint8_t **mergedPrefixkeys = (uint8_t **)malloc(casesSize*sizeof(uint8_t *)); 1843 uint8_t **mergedSuffixkeys = (uint8_t **)malloc(casesSize*sizeof(uint8_t *)); 1844 uint32_t *sortKeysLen = (uint32_t *)malloc(casesSize*sizeof(uint32_t)); 1845 uint8_t prefixKey[256], suffixKey[256]; 1846 uint32_t prefixKeyLen = 0, suffixKeyLen = 0, i = 0; 1847 UChar buffer[256]; 1848 uint32_t unescapedLen = 0, l1 = 0, l2 = 0; 1849 UColAttributeValue strength; 1850 1851 log_verbose("ucol_mergeSortkeys test\n"); 1852 log_verbose("Testing order of the test cases\n"); 1853 genericLocaleStarter("en", cases, casesSize); 1854 1855 for(i = 0; i<casesSize; i++) { 1856 sortkeys[i] = (uint8_t *)malloc(256*sizeof(uint8_t)); 1857 mergedPrefixkeys[i] = (uint8_t *)malloc(256*sizeof(uint8_t)); 1858 mergedSuffixkeys[i] = (uint8_t *)malloc(256*sizeof(uint8_t)); 1859 } 1860 1861 unescapedLen = u_unescape(prefix, buffer, 256); 1862 prefixKeyLen = ucol_getSortKey(coll, buffer, unescapedLen, prefixKey, 256); 1863 1864 unescapedLen = u_unescape(suffix, buffer, 256); 1865 suffixKeyLen = ucol_getSortKey(coll, buffer, unescapedLen, suffixKey, 256); 1866 1867 log_verbose("Massaging data with prefixes and different strengths\n"); 1868 strength = UCOL_PRIMARY; 1869 while(strength <= UCOL_IDENTICAL) { 1870 log_verbose("Strength %s\n", strengthsC[strength<=UCOL_QUATERNARY?strength:4]); 1871 ucol_setAttribute(coll, UCOL_STRENGTH, strength, &status); 1872 for(i = 0; i<casesSize; i++) { 1873 unescapedLen = u_unescape(cases[i], buffer, 256); 1874 sortKeysLen[i] = ucol_getSortKey(coll, buffer, unescapedLen, sortkeys[i], 256); 1875 ucol_mergeSortkeys(prefixKey, prefixKeyLen, sortkeys[i], sortKeysLen[i], mergedPrefixkeys[i], 256); 1876 ucol_mergeSortkeys(sortkeys[i], sortKeysLen[i], suffixKey, suffixKeyLen, mergedSuffixkeys[i], 256); 1877 if(i>0) { 1878 if(tMemCmp(mergedPrefixkeys[i-1], mergedPrefixkeys[i]) >= 0) { 1879 log_err("Error while comparing prefixed keys @ strength %s:\n", strengthsC[strength<=UCOL_QUATERNARY?strength:4]); 1880 log_err("%s\n%s\n", 1881 ucol_sortKeyToString(coll, mergedPrefixkeys[i-1], outBuff1, &l1), 1882 ucol_sortKeyToString(coll, mergedPrefixkeys[i], outBuff2, &l2)); 1883 } 1884 if(tMemCmp(mergedSuffixkeys[i-1], mergedSuffixkeys[i]) >= 0) { 1885 log_err("Error while comparing suffixed keys @ strength %s:\n", strengthsC[strength<=UCOL_QUATERNARY?strength:4]); 1886 log_err("%s\n%s\n", 1887 ucol_sortKeyToString(coll, mergedSuffixkeys[i-1], outBuff1, &l1), 1888 ucol_sortKeyToString(coll, mergedSuffixkeys[i], outBuff2, &l2)); 1889 } 1890 } 1891 } 1892 if(strength == UCOL_QUATERNARY) { 1893 strength = UCOL_IDENTICAL; 1894 } else { 1895 strength++; 1896 } 1897 } 1898 1899 { 1900 uint8_t smallBuf[3]; 1901 uint32_t reqLen = 0; 1902 log_verbose("testing buffer overflow\n"); 1903 reqLen = ucol_mergeSortkeys(prefixKey, prefixKeyLen, suffixKey, suffixKeyLen, smallBuf, 3); 1904 if(reqLen != (prefixKeyLen+suffixKeyLen-1)) { 1905 log_err("Wrong preflight size for merged sortkey\n"); 1906 } 1907 } 1908 1909 { 1910 UChar empty = 0; 1911 uint8_t emptyKey[20], abcKey[50], mergedKey[100]; 1912 int32_t emptyKeyLen = 0, abcKeyLen = 0, mergedKeyLen = 0; 1913 1914 log_verbose("testing merging with sortkeys generated for empty strings\n"); 1915 emptyKeyLen = ucol_getSortKey(coll, &empty, 0, emptyKey, 20); 1916 unescapedLen = u_unescape(cases[0], buffer, 256); 1917 abcKeyLen = ucol_getSortKey(coll, buffer, unescapedLen, abcKey, 50); 1918 mergedKeyLen = ucol_mergeSortkeys(emptyKey, emptyKeyLen, abcKey, abcKeyLen, mergedKey, 100); 1919 if(mergedKey[0] != 2) { 1920 log_err("Empty sortkey didn't produce a level separator\n"); 1921 } 1922 /* try with zeros */ 1923 mergedKeyLen = ucol_mergeSortkeys(emptyKey, 0, abcKey, abcKeyLen, mergedKey, 100); 1924 if(mergedKeyLen != 0 || mergedKey[0] != 0) { 1925 log_err("Empty key didn't produce null mergedKey\n"); 1926 } 1927 mergedKeyLen = ucol_mergeSortkeys(abcKey, abcKeyLen, emptyKey, 0, mergedKey, 100); 1928 if(mergedKeyLen != 0 || mergedKey[0] != 0) { 1929 log_err("Empty key didn't produce null mergedKey\n"); 1930 } 1931 1932 } 1933 1934 for(i = 0; i<casesSize; i++) { 1935 free(sortkeys[i]); 1936 free(mergedPrefixkeys[i]); 1937 free(mergedSuffixkeys[i]); 1938 } 1939 free(sortkeys); 1940 free(mergedPrefixkeys); 1941 free(mergedSuffixkeys); 1942 free(sortKeysLen); 1943 ucol_close(coll); 1944 /* need to finish this up */ 1945 } else { 1946 log_data_err("Couldn't open collator"); 1947 } 1948 } 1949 static void TestShortString(void) 1950 { 1951 struct { 1952 const char *input; 1953 const char *expectedOutput; 1954 const char *locale; 1955 UErrorCode expectedStatus; 1956 int32_t expectedOffset; 1957 uint32_t expectedIdentifier; 1958 } testCases[] = { 1959 /* 1960 * The following expectedOutput contains a collation weight (2D00 from UCA 5.2) 1961 * which is the primary weight for the T character (U+0041) in the input. 1962 * When that character gets a different weight in FractionalUCA.txt, 1963 * the expectedOutput needs to be adjusted. 1964 * That is, when we upgrade to a new UCA version or change collation 1965 * in such a way that the absolute weight for 'A' changes, 1966 * we will get a test failure here and need to adjust the test case. 1967 */ 1968 {"LDE_RDE_KPHONEBOOK_T0041_ZLATN","B2D00_KPHONEBOOK_LDE", "de@collation=phonebook", U_USING_FALLBACK_WARNING, 0, 0 }, 1969 1970 {"LEN_RUS_NO_AS_S4","AS_LEN_NO_S4", NULL, U_USING_FALLBACK_WARNING, 0, 0 }, 1971 {"LDE_VPHONEBOOK_EO_SI","EO_KPHONEBOOK_LDE_SI", "de@collation=phonebook", U_ZERO_ERROR, 0, 0 }, 1972 {"LDE_Kphonebook","KPHONEBOOK_LDE", "de@collation=phonebook", U_ZERO_ERROR, 0, 0 }, 1973 {"Xqde_DE@collation=phonebookq_S3_EX","KPHONEBOOK_LDE", "de@collation=phonebook", U_USING_FALLBACK_WARNING, 0, 0 }, 1974 {"LFR_FO", "LFR", NULL, U_ZERO_ERROR, 0, 0 }, 1975 {"SO_LX_AS", "", NULL, U_ILLEGAL_ARGUMENT_ERROR, 8, 0 }, 1976 {"S3_ASS_MMM", "", NULL, U_ILLEGAL_ARGUMENT_ERROR, 5, 0 } 1977 }; 1978 1979 int32_t i = 0; 1980 UCollator *coll = NULL, *fromNormalized = NULL; 1981 UParseError parseError; 1982 UErrorCode status = U_ZERO_ERROR; 1983 char fromShortBuffer[256], normalizedBuffer[256], fromNormalizedBuffer[256]; 1984 const char* locale = NULL; 1985 1986 1987 for(i = 0; i < sizeof(testCases)/sizeof(testCases[0]); i++) { 1988 status = U_ZERO_ERROR; 1989 if(testCases[i].locale) { 1990 locale = testCases[i].locale; 1991 } else { 1992 locale = NULL; 1993 } 1994 1995 coll = ucol_openFromShortString(testCases[i].input, FALSE, &parseError, &status); 1996 if(status != testCases[i].expectedStatus) { 1997 log_err_status(status, "Got status '%s' that is different from expected '%s' for '%s'\n", 1998 u_errorName(status), u_errorName(testCases[i].expectedStatus), testCases[i].input); 1999 continue; 2000 } 2001 2002 if(U_SUCCESS(status)) { 2003 ucol_getShortDefinitionString(coll, locale, fromShortBuffer, 256, &status); 2004 2005 if(strcmp(fromShortBuffer, testCases[i].expectedOutput)) { 2006 log_err("Got short string '%s' from the collator. Expected '%s' for input '%s'\n", 2007 fromShortBuffer, testCases[i].expectedOutput, testCases[i].input); 2008 } 2009 2010 ucol_normalizeShortDefinitionString(testCases[i].input, normalizedBuffer, 256, &parseError, &status); 2011 fromNormalized = ucol_openFromShortString(normalizedBuffer, FALSE, &parseError, &status); 2012 ucol_getShortDefinitionString(fromNormalized, locale, fromNormalizedBuffer, 256, &status); 2013 2014 if(strcmp(fromShortBuffer, fromNormalizedBuffer)) { 2015 log_err("Strings obtained from collators instantiated by short string ('%s') and from normalized string ('%s') differ\n", 2016 fromShortBuffer, fromNormalizedBuffer); 2017 } 2018 2019 2020 if(!ucol_equals(coll, fromNormalized)) { 2021 log_err("Collator from short string ('%s') differs from one obtained through a normalized version ('%s')\n", 2022 testCases[i].input, normalizedBuffer); 2023 } 2024 2025 ucol_close(fromNormalized); 2026 ucol_close(coll); 2027 2028 } else { 2029 if(parseError.offset != testCases[i].expectedOffset) { 2030 log_err("Got parse error offset %i, but expected %i instead for '%s'\n", 2031 parseError.offset, testCases[i].expectedOffset, testCases[i].input); 2032 } 2033 } 2034 } 2035 2036 } 2037 2038 static void 2039 doSetsTest(const char *locale, const USet *ref, USet *set, const char* inSet, const char* outSet, UErrorCode *status) { 2040 UChar buffer[512]; 2041 int32_t bufLen; 2042 2043 uset_clear(set); 2044 bufLen = u_unescape(inSet, buffer, 512); 2045 uset_applyPattern(set, buffer, bufLen, 0, status); 2046 if(U_FAILURE(*status)) { 2047 log_err("%s: Failure setting pattern %s\n", locale, u_errorName(*status)); 2048 } 2049 2050 if(!uset_containsAll(ref, set)) { 2051 log_err("%s: Some stuff from %s is not present in the set\n", locale, inSet); 2052 } 2053 2054 uset_clear(set); 2055 bufLen = u_unescape(outSet, buffer, 512); 2056 uset_applyPattern(set, buffer, bufLen, 0, status); 2057 if(U_FAILURE(*status)) { 2058 log_err("%s: Failure setting pattern %s\n", locale, u_errorName(*status)); 2059 } 2060 2061 if(!uset_containsNone(ref, set)) { 2062 log_err("%s: Some stuff from %s is present in the set\n", locale, outSet); 2063 } 2064 } 2065 2066 2067 2068 2069 static void 2070 TestGetContractionsAndUnsafes(void) 2071 { 2072 static struct { 2073 const char* locale; 2074 const char* inConts; 2075 const char* outConts; 2076 const char* inExp; 2077 const char* outExp; 2078 const char* unsafeCodeUnits; 2079 const char* safeCodeUnits; 2080 } tests[] = { 2081 { "ru", 2082 "[{\\u0474\\u030F}{\\u0475\\u030F}{\\u04D8\\u0308}{\\u04D9\\u0308}{\\u04E8\\u0308}{\\u04E9\\u0308}]", 2083 "[{\\u0430\\u0306}{\\u0410\\u0306}{\\u0430\\u0308}{\\u0410\\u0306}{\\u0433\\u0301}{\\u0413\\u0301}]", 2084 "[\\u00e6]", 2085 "[a]", 2086 "[\\u0474\\u0475\\u04d8\\u04d9\\u04e8\\u04e9]", 2087 "[aAbB\\u0430\\u0410\\u0433\\u0413]" 2088 }, 2089 { "uk", 2090 "[{\\u0474\\u030F}{\\u0475\\u030F}{\\u04D8\\u0308}{\\u04D9\\u0308}{\\u04E8\\u0308}{\\u04E9\\u0308}" 2091 "{\\u0430\\u0306}{\\u0410\\u0306}{\\u0430\\u0308}{\\u0410\\u0306}{\\u0433\\u0301}{\\u0413\\u0301}]", 2092 "[]", 2093 "[\\u00e6]", 2094 "[a]", 2095 "[\\u0474\\u0475\\u04D8\\u04D9\\u04E8\\u04E9\\u0430\\u0410\\u0433\\u0413]", 2096 "[aAbBxv]", 2097 }, 2098 { "sh", 2099 "[{C\\u0301}{C\\u030C}{C\\u0341}{DZ\\u030C}{Dz\\u030C}{D\\u017D}{D\\u017E}{lj}{nj}]", 2100 "[{\\u309d\\u3099}{\\u30fd\\u3099}]", 2101 "[\\u00e6]", 2102 "[a]", 2103 "[nlcdzNLCDZ]", 2104 "[jabv]" 2105 }, 2106 { "ja", 2107 "[{\\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}]", 2108 "[{\\u30FD\\u3099}{\\u309D\\u3099}{\\u3053\\u3099}{\\u30B3\\u3099}{lj}{nj}]", 2109 "[\\u30FE\\u00e6]", 2110 "[a]", 2111 "[\\u3099]", 2112 "[]" 2113 } 2114 }; 2115 2116 2117 2118 2119 UErrorCode status = U_ZERO_ERROR; 2120 UCollator *coll = NULL; 2121 int32_t i = 0; 2122 int32_t noConts = 0; 2123 USet *conts = uset_open(0,0); 2124 USet *exp = uset_open(0, 0); 2125 USet *set = uset_open(0,0); 2126 int32_t setBufferLen = 65536; 2127 UChar buffer[65536]; 2128 int32_t setLen = 0; 2129 2130 for(i = 0; i < sizeof(tests)/sizeof(tests[0]); i++) { 2131 log_verbose("Testing locale: %s\n", tests[i].locale); 2132 coll = ucol_open(tests[i].locale, &status); 2133 if (coll == NULL || U_FAILURE(status)) { 2134 log_err_status(status, "Unable to open collator for locale %s ==> %s\n", tests[i].locale, u_errorName(status)); 2135 continue; 2136 } 2137 ucol_getContractionsAndExpansions(coll, conts, exp, TRUE, &status); 2138 doSetsTest(tests[i].locale, conts, set, tests[i].inConts, tests[i].outConts, &status); 2139 setLen = uset_toPattern(conts, buffer, setBufferLen, TRUE, &status); 2140 if(U_SUCCESS(status)) { 2141 /*log_verbose("Contractions %i: %s\n", uset_getItemCount(conts), aescstrdup(buffer, setLen));*/ 2142 } else { 2143 log_err("error %s. %i\n", u_errorName(status), setLen); 2144 status = U_ZERO_ERROR; 2145 } 2146 doSetsTest(tests[i].locale, exp, set, tests[i].inExp, tests[i].outExp, &status); 2147 setLen = uset_toPattern(exp, buffer, setBufferLen, TRUE, &status); 2148 if(U_SUCCESS(status)) { 2149 /*log_verbose("Expansions %i: %s\n", uset_getItemCount(exp), aescstrdup(buffer, setLen));*/ 2150 } else { 2151 log_err("error %s. %i\n", u_errorName(status), setLen); 2152 status = U_ZERO_ERROR; 2153 } 2154 2155 noConts = ucol_getUnsafeSet(coll, conts, &status); 2156 doSetsTest(tests[i].locale, conts, set, tests[i].unsafeCodeUnits, tests[i].safeCodeUnits, &status); 2157 setLen = uset_toPattern(conts, buffer, setBufferLen, TRUE, &status); 2158 if(U_SUCCESS(status)) { 2159 log_verbose("Unsafe %i: %s\n", uset_getItemCount(exp), aescstrdup(buffer, setLen)); 2160 } else { 2161 log_err("error %s. %i\n", u_errorName(status), setLen); 2162 status = U_ZERO_ERROR; 2163 } 2164 2165 ucol_close(coll); 2166 } 2167 2168 2169 uset_close(conts); 2170 uset_close(exp); 2171 uset_close(set); 2172 } 2173 2174 static void 2175 TestOpenBinary(void) 2176 { 2177 UErrorCode status = U_ZERO_ERROR; 2178 /* 2179 char rule[] = "&h < d < c < b"; 2180 char *wUCA[] = { "a", "h", "d", "c", "b", "i" }; 2181 char *noUCA[] = {"d", "c", "b", "a", "h", "i" }; 2182 */ 2183 /* we have to use Cyrillic letters because latin-1 always gets copied */ 2184 const char rule[] = "&\\u0452 < \\u0434 < \\u0433 < \\u0432"; /* &dje < d < g < v */ 2185 const char *wUCA[] = { "\\u0430", "\\u0452", "\\u0434", "\\u0433", "\\u0432", "\\u0435" }; /* a, dje, d, g, v, e */ 2186 const char *noUCA[] = {"\\u0434", "\\u0433", "\\u0432", "\\u0430", "\\u0435", "\\u0452" }; /* d, g, v, a, e, dje */ 2187 2188 UChar uRules[256]; 2189 int32_t uRulesLen = u_unescape(rule, uRules, 256); 2190 2191 UCollator *coll = ucol_openRules(uRules, uRulesLen, UCOL_DEFAULT, UCOL_DEFAULT, NULL, &status); 2192 UCollator *UCA = NULL; 2193 UCollator *cloneNOUCA = NULL, *cloneWUCA = NULL; 2194 2195 uint8_t imageBuffer[32768]; 2196 uint8_t *image = imageBuffer; 2197 int32_t imageBufferCapacity = 32768; 2198 2199 int32_t imageSize; 2200 2201 if((coll==NULL)||(U_FAILURE(status))) { 2202 log_data_err("could not load collators or error occured: %s\n", 2203 u_errorName(status)); 2204 return; 2205 } 2206 UCA = ucol_open("root", &status); 2207 if((UCA==NULL)||(U_FAILURE(status))) { 2208 log_data_err("could not load UCA collator or error occured: %s\n", 2209 u_errorName(status)); 2210 return; 2211 } 2212 imageSize = ucol_cloneBinary(coll, image, imageBufferCapacity, &status); 2213 if(U_FAILURE(status)) { 2214 image = (uint8_t *)malloc(imageSize*sizeof(uint8_t)); 2215 status = U_ZERO_ERROR; 2216 imageSize = ucol_cloneBinary(coll, imageBuffer, imageSize, &status); 2217 } 2218 2219 2220 cloneWUCA = ucol_openBinary(image, imageSize, UCA, &status); 2221 cloneNOUCA = ucol_openBinary(image, imageSize, NULL, &status); 2222 2223 genericOrderingTest(coll, wUCA, sizeof(wUCA)/sizeof(wUCA[0])); 2224 2225 genericOrderingTest(cloneWUCA, wUCA, sizeof(wUCA)/sizeof(wUCA[0])); 2226 genericOrderingTest(cloneNOUCA, noUCA, sizeof(noUCA)/sizeof(noUCA[0])); 2227 2228 if(image != imageBuffer) { 2229 free(image); 2230 } 2231 ucol_close(coll); 2232 ucol_close(cloneNOUCA); 2233 ucol_close(cloneWUCA); 2234 ucol_close(UCA); 2235 } 2236 2237 static void TestDefault(void) { 2238 /* Tests for code coverage. */ 2239 UErrorCode status = U_ZERO_ERROR; 2240 UCollator *coll = ucol_open("es@collation=pinyin", &status); 2241 if (coll == NULL || status == U_FILE_ACCESS_ERROR) { 2242 log_data_err("Unable to open collator es@collation=pinyin\n"); 2243 return; 2244 } 2245 if (status != U_USING_DEFAULT_WARNING) { 2246 /* What do you mean that you know about using pinyin collation in Spanish!? This should be in the zh locale. */ 2247 log_err("es@collation=pinyin should return U_USING_DEFAULT_WARNING, but returned %s\n", u_errorName(status)); 2248 } 2249 ucol_close(coll); 2250 if (ucol_getKeywordValues("funky", &status) != NULL) { 2251 log_err("Collators should not know about the funky keyword.\n"); 2252 } 2253 if (status != U_ILLEGAL_ARGUMENT_ERROR) { 2254 log_err("funky keyword didn't fail as expected %s\n", u_errorName(status)); 2255 } 2256 if (ucol_getKeywordValues("collation", &status) != NULL) { 2257 log_err("ucol_getKeywordValues should not work when given a bad status.\n"); 2258 } 2259 } 2260 2261 static void TestDefaultKeyword(void) { 2262 /* Tests for code coverage. */ 2263 UErrorCode status = U_ZERO_ERROR; 2264 const char *loc = "zh_TW@collation=default"; 2265 UCollator *coll = ucol_open(loc, &status); 2266 if(U_FAILURE(status)) { 2267 log_info("Warning: ucol_open(%s, ...) returned %s, at least it didn't crash.\n", loc, u_errorName(status)); 2268 } else if (status != U_USING_FALLBACK_WARNING) { 2269 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)); 2270 } 2271 ucol_close(coll); 2272 } 2273 2274 static void TestGetKeywordValuesForLocale(void) { 2275 #define PREFERRED_SIZE 16 2276 #define MAX_NUMBER_OF_KEYWORDS 6 2277 const char *PREFERRED[PREFERRED_SIZE][MAX_NUMBER_OF_KEYWORDS+1] = { 2278 { "und", "standard", NULL, NULL, NULL, NULL, NULL }, 2279 { "en_US", "standard", NULL, NULL, NULL, NULL, NULL }, 2280 { "en_029", "standard", NULL, NULL, NULL, NULL, NULL }, 2281 { "de_DE", "standard", "phonebook", NULL, NULL, NULL, NULL }, 2282 { "de_Latn_DE", "standard", "phonebook", NULL, NULL, NULL, NULL }, 2283 { "zh", "pinyin", "big5han", "gb2312han", "standard", "stroke", "unihan" }, 2284 { "zh_Hans", "pinyin", "big5han", "gb2312han", "standard", "stroke", "unihan" }, 2285 { "zh_CN", "pinyin", "big5han", "gb2312han", "standard", "stroke", "unihan" }, 2286 { "zh_Hant", "stroke", "big5han", "gb2312han", "pinyin", "standard", "unihan" }, 2287 { "zh_TW", "stroke", "big5han", "gb2312han", "pinyin", "standard", "unihan" }, 2288 { "zh__PINYIN", "pinyin", "big5han", "gb2312han", "standard", "stroke", "unihan" }, 2289 { "es_ES", "standard", "traditional", NULL, NULL, NULL, NULL }, 2290 { "es__TRADITIONAL","traditional", "standard", NULL, NULL, NULL, NULL }, 2291 { "und@collation=phonebook", "standard", NULL, NULL, NULL, NULL, NULL }, 2292 { "de_DE@collation=big5han", "standard", "phonebook", NULL, NULL, NULL, NULL }, 2293 { "zzz@collation=xxx", "standard", NULL, NULL, NULL, NULL, NULL } 2294 }; 2295 const int32_t expectedLength[PREFERRED_SIZE] = { 1, 1, 1, 2, 2, 6, 6, 6, 6, 6, 6, 2, 2, 1, 2, 1 }; 2296 2297 UErrorCode status = U_ZERO_ERROR; 2298 UEnumeration *keywordValues = NULL; 2299 int32_t i, n, size, valueLength; 2300 const char *locale = NULL, *value = NULL; 2301 UBool errorOccurred = FALSE; 2302 2303 for (i = 0; i < PREFERRED_SIZE; i++) { 2304 locale = PREFERRED[i][0]; 2305 value = NULL; 2306 valueLength = 0; 2307 size = 0; 2308 2309 keywordValues = ucol_getKeywordValuesForLocale("collation", locale, TRUE, &status); 2310 if (keywordValues == NULL || U_FAILURE(status)) { 2311 log_err_status(status, "Error getting keyword values: %s\n", u_errorName(status)); 2312 break; 2313 } 2314 size = uenum_count(keywordValues, &status); 2315 2316 if (size == expectedLength[i]) { 2317 for (n = 0; n < expectedLength[i]; n++) { 2318 if ((value = uenum_next(keywordValues, &valueLength, &status)) != NULL && U_SUCCESS(status)) { 2319 if (uprv_strcmp(value, PREFERRED[i][n+1]) != 0) { 2320 log_err("Keyword values differ: Got [%s] Expected [%s] for locale: %s\n", value, PREFERRED[i][n+1], locale); 2321 errorOccurred = TRUE; 2322 break; 2323 } 2324 2325 } else { 2326 log_err("While getting keyword value from locale: %s got this error: %s\n", locale, u_errorName(status)); 2327 errorOccurred = TRUE; 2328 break; 2329 } 2330 } 2331 if (errorOccurred) { 2332 break; 2333 } 2334 } else { 2335 log_err("Number of keywords (%d) does not match expected size (%d) for locale: %s\n", size, expectedLength[i], locale); 2336 break; 2337 } 2338 uenum_close(keywordValues); 2339 keywordValues = NULL; 2340 } 2341 if (keywordValues != NULL) { 2342 uenum_close(keywordValues); 2343 } 2344 } 2345 2346 2347 #endif /* #if !UCONFIG_NO_COLLATION */ 2348