1 /******************************************************************** 2 * Copyright (c) 1997-2010 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 35 static void TestAttribute(void); 36 static void TestDefault(void); 37 static void TestDefaultKeyword(void); 38 int TestBufferSize(); /* defined in "colutil.c" */ 39 40 41 42 43 /* next two function is modified from "i18n/ucol.cpp" to avoid include "ucol_imp.h" */ 44 static void uprv_appendByteToHexString(char *dst, uint8_t val) { 45 uint32_t len = (uint32_t)strlen(dst); 46 sprintf(dst+len, "%02X", val); 47 } 48 49 static char* U_EXPORT2 ucol_sortKeyToString(const UCollator *coll, const uint8_t *sortkey, char *buffer, uint32_t *len) { 50 int32_t strength = UCOL_PRIMARY; 51 uint32_t res_size = 0; 52 UBool doneCase = FALSE; 53 54 char *current = buffer; 55 const uint8_t *currentSk = sortkey; 56 57 UErrorCode error_code = U_ZERO_ERROR; 58 59 strcpy(current, "["); 60 61 while(strength <= UCOL_QUATERNARY && strength <= ucol_getAttribute(coll,UCOL_STRENGTH, &error_code)) { 62 if(U_FAILURE(error_code)) { 63 log_err("ucol_getAttribute returned error: %s\n", u_errorName(error_code)); 64 } 65 if(strength > UCOL_PRIMARY) { 66 strcat(current, " . "); 67 } 68 while(*currentSk != 0x01 && *currentSk != 0x00) { /* print a level */ 69 uprv_appendByteToHexString(current, *currentSk++); 70 strcat(current, " "); 71 } 72 if(ucol_getAttribute(coll,UCOL_CASE_LEVEL, &error_code) == UCOL_ON && strength == UCOL_SECONDARY && doneCase == FALSE) { 73 doneCase = TRUE; 74 } else if(ucol_getAttribute(coll,UCOL_CASE_LEVEL, &error_code) == UCOL_OFF || doneCase == TRUE || strength != UCOL_SECONDARY) { 75 strength ++; 76 } 77 if(U_FAILURE(error_code)) { 78 log_err("ucol_getAttribute returned error: %s\n", u_errorName(error_code)); 79 } 80 uprv_appendByteToHexString(current, *currentSk++); /* This should print '01' */ 81 if(strength == UCOL_QUATERNARY && ucol_getAttribute(coll,UCOL_ALTERNATE_HANDLING, &error_code) == UCOL_NON_IGNORABLE) { 82 break; 83 } 84 } 85 86 if(ucol_getAttribute(coll,UCOL_STRENGTH, &error_code) == UCOL_IDENTICAL) { 87 strcat(current, " . "); 88 while(*currentSk != 0) { 89 uprv_appendByteToHexString(current, *currentSk++); 90 strcat(current, " "); 91 } 92 93 uprv_appendByteToHexString(current, *currentSk++); 94 } 95 if(U_FAILURE(error_code)) { 96 log_err("ucol_getAttribute returned error: %s\n", u_errorName(error_code)); 97 } 98 strcat(current, "]"); 99 100 if(res_size > *len) { 101 return NULL; 102 } 103 104 return buffer; 105 } 106 /* end of avoid include "ucol_imp.h" */ 107 108 109 void addCollAPITest(TestNode** root) 110 { 111 /* WEIVTODO: return tests here */ 112 addTest(root, &TestProperty, "tscoll/capitst/TestProperty"); 113 addTest(root, &TestRuleBasedColl, "tscoll/capitst/TestRuleBasedColl"); 114 addTest(root, &TestCompare, "tscoll/capitst/TestCompare"); 115 addTest(root, &TestSortKey, "tscoll/capitst/TestSortKey"); 116 addTest(root, &TestHashCode, "tscoll/capitst/TestHashCode"); 117 addTest(root, &TestElemIter, "tscoll/capitst/TestElemIter"); 118 addTest(root, &TestGetAll, "tscoll/capitst/TestGetAll"); 119 /*addTest(root, &TestGetDefaultRules, "tscoll/capitst/TestGetDefaultRules");*/ 120 addTest(root, &TestDecomposition, "tscoll/capitst/TestDecomposition"); 121 addTest(root, &TestSafeClone, "tscoll/capitst/TestSafeClone"); 122 addTest(root, &TestCloneBinary, "tscoll/capitst/TestCloneBinary"); 123 addTest(root, &TestGetSetAttr, "tscoll/capitst/TestGetSetAttr"); 124 addTest(root, &TestBounds, "tscoll/capitst/TestBounds"); 125 addTest(root, &TestGetLocale, "tscoll/capitst/TestGetLocale"); 126 addTest(root, &TestSortKeyBufferOverrun, "tscoll/capitst/TestSortKeyBufferOverrun"); 127 addTest(root, &TestAttribute, "tscoll/capitst/TestAttribute"); 128 addTest(root, &TestGetTailoredSet, "tscoll/capitst/TestGetTailoredSet"); 129 addTest(root, &TestMergeSortKeys, "tscoll/capitst/TestMergeSortKeys"); 130 addTest(root, &TestShortString, "tscoll/capitst/TestShortString"); 131 addTest(root, &TestGetContractionsAndUnsafes, "tscoll/capitst/TestGetContractionsAndUnsafes"); 132 addTest(root, &TestOpenBinary, "tscoll/capitst/TestOpenBinary"); 133 addTest(root, &TestDefault, "tscoll/capitst/TestDefault"); 134 addTest(root, &TestDefaultKeyword, "tscoll/capitst/TestDefaultKeyword"); 135 addTest(root, &TestOpenVsOpenRules, "tscoll/capitst/TestOpenVsOpenRules"); 136 addTest(root, &TestGetKeywordValuesForLocale, "tscoll/capitst/TestGetKeywordValuesForLocale"); 137 } 138 139 void TestGetSetAttr(void) { 140 UErrorCode status = U_ZERO_ERROR; 141 UCollator *coll = ucol_open(NULL, &status); 142 struct attrTest { 143 UColAttribute att; 144 UColAttributeValue val[5]; 145 uint32_t valueSize; 146 UColAttributeValue nonValue; 147 } attrs[] = { 148 {UCOL_FRENCH_COLLATION, {UCOL_ON, UCOL_OFF}, 2, UCOL_SHIFTED}, 149 {UCOL_ALTERNATE_HANDLING, {UCOL_NON_IGNORABLE, UCOL_SHIFTED}, 2, UCOL_OFF},/* attribute for handling variable elements*/ 150 {UCOL_CASE_FIRST, {UCOL_OFF, UCOL_LOWER_FIRST, UCOL_UPPER_FIRST}, 3, UCOL_SHIFTED},/* who goes first, lower case or uppercase */ 151 {UCOL_CASE_LEVEL, {UCOL_ON, UCOL_OFF}, 2, UCOL_SHIFTED},/* do we have an extra case level */ 152 {UCOL_NORMALIZATION_MODE, {UCOL_ON, UCOL_OFF}, 2, UCOL_SHIFTED},/* attribute for normalization */ 153 {UCOL_DECOMPOSITION_MODE, {UCOL_ON, UCOL_OFF}, 2, UCOL_SHIFTED}, 154 {UCOL_STRENGTH, {UCOL_PRIMARY, UCOL_SECONDARY, UCOL_TERTIARY, UCOL_QUATERNARY, UCOL_IDENTICAL}, 5, UCOL_SHIFTED},/* attribute for strength */ 155 {UCOL_HIRAGANA_QUATERNARY_MODE, {UCOL_ON, UCOL_OFF}, 2, UCOL_SHIFTED},/* when turned on, this attribute */ 156 }; 157 UColAttribute currAttr; 158 UColAttributeValue value; 159 uint32_t i = 0, j = 0; 160 161 if (coll == NULL) { 162 log_err_status(status, "Unable to open collator. %s\n", u_errorName(status)); 163 return; 164 } 165 for(i = 0; i<sizeof(attrs)/sizeof(attrs[0]); i++) { 166 currAttr = attrs[i].att; 167 ucol_setAttribute(coll, currAttr, UCOL_DEFAULT, &status); 168 if(U_FAILURE(status)) { 169 log_err_status(status, "ucol_setAttribute with the default value returned error: %s\n", u_errorName(status)); 170 break; 171 } 172 value = ucol_getAttribute(coll, currAttr, &status); 173 if(U_FAILURE(status)) { 174 log_err("ucol_getAttribute returned error: %s\n", u_errorName(status)); 175 break; 176 } 177 for(j = 0; j<attrs[i].valueSize; j++) { 178 ucol_setAttribute(coll, currAttr, attrs[i].val[j], &status); 179 if(U_FAILURE(status)) { 180 log_err("ucol_setAttribute with the value %i returned error: %s\n", attrs[i].val[j], u_errorName(status)); 181 break; 182 } 183 } 184 status = U_ZERO_ERROR; 185 ucol_setAttribute(coll, currAttr, attrs[i].nonValue, &status); 186 if(U_SUCCESS(status)) { 187 log_err("ucol_setAttribute with the bad value didn't return an error\n"); 188 break; 189 } 190 status = U_ZERO_ERROR; 191 192 ucol_setAttribute(coll, currAttr, value, &status); 193 if(U_FAILURE(status)) { 194 log_err("ucol_setAttribute with the default valuereturned error: %s\n", u_errorName(status)); 195 break; 196 } 197 } 198 status = U_ZERO_ERROR; 199 value = ucol_getAttribute(coll, UCOL_ATTRIBUTE_COUNT, &status); 200 if(U_SUCCESS(status)) { 201 log_err("ucol_getAttribute for UCOL_ATTRIBUTE_COUNT didn't return an error\n"); 202 } 203 status = U_ZERO_ERROR; 204 ucol_setAttribute(coll, UCOL_ATTRIBUTE_COUNT, UCOL_DEFAULT, &status); 205 if(U_SUCCESS(status)) { 206 log_err("ucol_setAttribute for UCOL_ATTRIBUTE_COUNT didn't return an error\n"); 207 } 208 status = U_ZERO_ERROR; 209 ucol_close(coll); 210 } 211 212 213 static void doAssert(int condition, const char *message) 214 { 215 if (condition==0) { 216 log_err("ERROR : %s\n", message); 217 } 218 } 219 220 #if 0 221 /* We don't have default rules, at least not in the previous sense */ 222 void TestGetDefaultRules(){ 223 uint32_t size=0; 224 UErrorCode status=U_ZERO_ERROR; 225 UCollator *coll=NULL; 226 int32_t len1 = 0, len2=0; 227 uint8_t *binColData = NULL; 228 229 UResourceBundle *res = NULL; 230 UResourceBundle *binColl = NULL; 231 uint8_t *binResult = NULL; 232 233 234 const UChar * defaultRulesArray=ucol_getDefaultRulesArray(&size); 235 log_verbose("Test the function ucol_getDefaultRulesArray()\n"); 236 237 coll = ucol_openRules(defaultRulesArray, size, UCOL_ON, UCOL_PRIMARY, &status); 238 if(U_SUCCESS(status) && coll !=NULL) { 239 binColData = (uint8_t*)ucol_cloneRuleData(coll, &len1, &status); 240 241 } 242 243 244 status=U_ZERO_ERROR; 245 res=ures_open(NULL, "root", &status); 246 if(U_FAILURE(status)){ 247 log_err("ERROR: Failed to get resource for \"root Locale\" with %s", myErrorName(status)); 248 return; 249 } 250 binColl=ures_getByKey(res, "%%Collation", binColl, &status); 251 if(U_SUCCESS(status)){ 252 binResult=(uint8_t*)ures_getBinary(binColl, &len2, &status); 253 if(U_FAILURE(status)){ 254 log_err("ERROR: ures_getBinary() failed\n"); 255 } 256 }else{ 257 log_err("ERROR: ures_getByKey(locale(default), %%Collation) failed"); 258 } 259 260 261 if(len1 != len2){ 262 log_err("Error: ucol_getDefaultRulesArray() failed to return the correct length.\n"); 263 } 264 if(memcmp(binColData, binResult, len1) != 0){ 265 log_err("Error: ucol_getDefaultRulesArray() failed\n"); 266 } 267 268 free(binColData); 269 ures_close(binColl); 270 ures_close(res); 271 ucol_close(coll); 272 273 } 274 #endif 275 276 /* Collator Properties 277 ucol_open, ucol_strcoll, getStrength/setStrength 278 getDecomposition/setDecomposition, getDisplayName*/ 279 void TestProperty() 280 { 281 UCollator *col, *ruled; 282 UChar *disName; 283 int32_t len = 0; 284 UChar *source, *target; 285 int32_t tempLength; 286 UErrorCode status = U_ZERO_ERROR; 287 /* 288 * Expected version of the English collator. 289 * Currently, the major/minor version numbers change when the builder code 290 * changes, 291 * number 2 is from the tailoring data version and 292 * number 3 is the UCA version. 293 * This changes with every UCA version change, and the expected value 294 * needs to be adjusted. 295 * Same in intltest/apicoll.cpp. 296 */ 297 UVersionInfo currVersionArray = {0x31, 0xC0, 0x05, 0x2A}; /* from ICU 4.4/UCA 5.2 */ 298 UVersionInfo versionArray = {0, 0, 0, 0}; 299 UVersionInfo versionUCAArray = {0, 0, 0, 0}; 300 UVersionInfo versionUCDArray = {0, 0, 0, 0}; 301 302 log_verbose("The property tests begin : \n"); 303 log_verbose("Test ucol_strcoll : \n"); 304 col = ucol_open("en_US", &status); 305 if (U_FAILURE(status)) { 306 log_err_status(status, "Default Collator creation failed.: %s\n", myErrorName(status)); 307 return; 308 } 309 310 ucol_getVersion(col, versionArray); 311 /* Check for a version greater than some value rather than equality 312 * so that we need not update the expected version each time. */ 313 if (uprv_memcmp(versionArray, currVersionArray, 4)<0) { 314 log_err("Testing ucol_getVersion() - unexpected result: %02x.%02x.%02x.%02x\n", 315 versionArray[0], versionArray[1], versionArray[2], versionArray[3]); 316 } else { 317 log_verbose("ucol_getVersion() result: %02x.%02x.%02x.%02x\n", 318 versionArray[0], versionArray[1], versionArray[2], versionArray[3]); 319 } 320 321 /* Assume that the UCD and UCA versions are the same, 322 * rather than hardcoding (and updating each time) a particular UCA version. */ 323 u_getUnicodeVersion(versionUCDArray); 324 ucol_getUCAVersion(col, versionUCAArray); 325 if (0!=uprv_memcmp(versionUCAArray, versionUCDArray, 4)) { 326 log_err("Testing ucol_getUCAVersion() - unexpected result: %hu.%hu.%hu.%hu\n", 327 versionUCAArray[0], versionUCAArray[1], versionUCAArray[2], versionUCAArray[3]); 328 } 329 330 source=(UChar*)malloc(sizeof(UChar) * 12); 331 target=(UChar*)malloc(sizeof(UChar) * 12); 332 333 334 u_uastrcpy(source, "ab"); 335 u_uastrcpy(target, "abc"); 336 337 doAssert((ucol_strcoll(col, source, u_strlen(source), target, u_strlen(target)) == UCOL_LESS), "ab < abc comparison failed"); 338 339 u_uastrcpy(source, "ab"); 340 u_uastrcpy(target, "AB"); 341 342 doAssert((ucol_strcoll(col, source, u_strlen(source), target, u_strlen(target)) == UCOL_LESS), "ab < AB comparison failed"); 343 /* u_uastrcpy(source, "black-bird"); 344 u_uastrcpy(target, "blackbird"); */ 345 u_uastrcpy(target, "black-bird"); 346 u_uastrcpy(source, "blackbird"); 347 348 doAssert((ucol_strcoll(col, source, u_strlen(source), target, u_strlen(target)) == UCOL_GREATER), 349 "black-bird > blackbird comparison failed"); 350 u_uastrcpy(source, "black bird"); 351 u_uastrcpy(target, "black-bird"); 352 doAssert((ucol_strcoll(col, source, u_strlen(source), target, u_strlen(target)) == UCOL_LESS), 353 "black bird < black-bird comparison failed"); 354 u_uastrcpy(source, "Hello"); 355 u_uastrcpy(target, "hello"); 356 357 doAssert((ucol_strcoll(col, source, u_strlen(source), target, u_strlen(target)) == UCOL_GREATER), 358 "Hello > hello comparison failed"); 359 free(source); 360 free(target); 361 log_verbose("Test ucol_strcoll ends.\n"); 362 363 log_verbose("testing ucol_getStrength() method ...\n"); 364 doAssert( (ucol_getStrength(col) == UCOL_TERTIARY), "collation object has the wrong strength"); 365 doAssert( (ucol_getStrength(col) != UCOL_PRIMARY), "collation object's strength is primary difference"); 366 367 log_verbose("testing ucol_setStrength() method ...\n"); 368 ucol_setStrength(col, UCOL_SECONDARY); 369 doAssert( (ucol_getStrength(col) != UCOL_TERTIARY), "collation object's strength is secondary difference"); 370 doAssert( (ucol_getStrength(col) != UCOL_PRIMARY), "collation object's strength is primary difference"); 371 doAssert( (ucol_getStrength(col) == UCOL_SECONDARY), "collation object has the wrong strength"); 372 373 374 log_verbose("Get display name for the default collation in German : \n"); 375 376 len=ucol_getDisplayName("en_US", "de_DE", NULL, 0, &status); 377 if(status==U_BUFFER_OVERFLOW_ERROR){ 378 status=U_ZERO_ERROR; 379 disName=(UChar*)malloc(sizeof(UChar) * (len+1)); 380 ucol_getDisplayName("en_US", "de_DE", disName, len+1, &status); 381 log_verbose("the display name for default collation in german: %s\n", austrdup(disName) ); 382 free(disName); 383 } 384 if(U_FAILURE(status)){ 385 log_err("ERROR: in getDisplayName: %s\n", myErrorName(status)); 386 return; 387 } 388 log_verbose("Default collation getDisplayName ended.\n"); 389 390 ruled = ucol_open("da_DK", &status); 391 log_verbose("ucol_getRules() testing ...\n"); 392 ucol_getRules(ruled, &tempLength); 393 doAssert( tempLength != 0, "getRules() result incorrect" ); 394 log_verbose("getRules tests end.\n"); 395 { 396 UChar *buffer = (UChar *)malloc(200000*sizeof(UChar)); 397 int32_t bufLen = 200000; 398 buffer[0] = '\0'; 399 log_verbose("ucol_getRulesEx() testing ...\n"); 400 tempLength = ucol_getRulesEx(col,UCOL_TAILORING_ONLY,buffer,bufLen ); 401 doAssert( tempLength == 0x00, "getRulesEx() result incorrect" ); 402 log_verbose("getRules tests end.\n"); 403 404 log_verbose("ucol_getRulesEx() testing ...\n"); 405 tempLength=ucol_getRulesEx(col,UCOL_FULL_RULES,buffer,bufLen ); 406 doAssert( tempLength != 0, "getRulesEx() result incorrect" ); 407 log_verbose("getRules tests end.\n"); 408 free(buffer); 409 } 410 ucol_close(ruled); 411 ucol_close(col); 412 413 log_verbose("open an collator for french locale"); 414 col = ucol_open("fr_FR", &status); 415 if (U_FAILURE(status)) { 416 log_err("ERROR: Creating French collation failed.: %s\n", myErrorName(status)); 417 return; 418 } 419 ucol_setStrength(col, UCOL_PRIMARY); 420 log_verbose("testing ucol_getStrength() method again ...\n"); 421 doAssert( (ucol_getStrength(col) != UCOL_TERTIARY), "collation object has the wrong strength"); 422 doAssert( (ucol_getStrength(col) == UCOL_PRIMARY), "collation object's strength is not primary difference"); 423 424 log_verbose("testing French ucol_setStrength() method ...\n"); 425 ucol_setStrength(col, UCOL_TERTIARY); 426 doAssert( (ucol_getStrength(col) == UCOL_TERTIARY), "collation object's strength is not tertiary difference"); 427 doAssert( (ucol_getStrength(col) != UCOL_PRIMARY), "collation object's strength is primary difference"); 428 doAssert( (ucol_getStrength(col) != UCOL_SECONDARY), "collation object's strength is secondary difference"); 429 ucol_close(col); 430 431 log_verbose("Get display name for the french collation in english : \n"); 432 len=ucol_getDisplayName("fr_FR", "en_US", NULL, 0, &status); 433 if(status==U_BUFFER_OVERFLOW_ERROR){ 434 status=U_ZERO_ERROR; 435 disName=(UChar*)malloc(sizeof(UChar) * (len+1)); 436 ucol_getDisplayName("fr_FR", "en_US", disName, len+1, &status); 437 log_verbose("the display name for french collation in english: %s\n", austrdup(disName) ); 438 free(disName); 439 } 440 if(U_FAILURE(status)){ 441 log_err("ERROR: in getDisplayName: %s\n", myErrorName(status)); 442 return; 443 } 444 log_verbose("Default collation getDisplayName ended.\n"); 445 446 } 447 448 /* Test RuleBasedCollator and getRules*/ 449 void TestRuleBasedColl() 450 { 451 UCollator *col1, *col2, *col3, *col4; 452 UCollationElements *iter1, *iter2; 453 UChar ruleset1[60]; 454 UChar ruleset2[50]; 455 UChar teststr[10]; 456 UChar teststr2[10]; 457 const UChar *rule1, *rule2, *rule3, *rule4; 458 int32_t tempLength; 459 UErrorCode status = U_ZERO_ERROR; 460 u_uastrcpy(ruleset1, "&9 < a, A < b, B < c, C; ch, cH, Ch, CH < d, D, e, E"); 461 u_uastrcpy(ruleset2, "&9 < a, A < b, B < c, C < d, D, e, E"); 462 463 464 col1 = ucol_openRules(ruleset1, u_strlen(ruleset1), UCOL_DEFAULT, UCOL_DEFAULT_STRENGTH, NULL,&status); 465 if (U_FAILURE(status)) { 466 log_err_status(status, "RuleBased Collator creation failed.: %s\n", myErrorName(status)); 467 return; 468 } 469 else 470 log_verbose("PASS: RuleBased Collator creation passed\n"); 471 472 status = U_ZERO_ERROR; 473 col2 = ucol_openRules(ruleset2, u_strlen(ruleset2), UCOL_DEFAULT, UCOL_DEFAULT_STRENGTH, NULL, &status); 474 if (U_FAILURE(status)) { 475 log_err("RuleBased Collator creation failed.: %s\n", myErrorName(status)); 476 return; 477 } 478 else 479 log_verbose("PASS: RuleBased Collator creation passed\n"); 480 481 482 status = U_ZERO_ERROR; 483 col3= ucol_open(NULL, &status); 484 if (U_FAILURE(status)) { 485 log_err("Default Collator creation failed.: %s\n", myErrorName(status)); 486 return; 487 } 488 else 489 log_verbose("PASS: Default Collator creation passed\n"); 490 491 rule1 = ucol_getRules(col1, &tempLength); 492 rule2 = ucol_getRules(col2, &tempLength); 493 rule3 = ucol_getRules(col3, &tempLength); 494 495 doAssert((u_strcmp(rule1, rule2) != 0), "Default collator getRules failed"); 496 doAssert((u_strcmp(rule2, rule3) != 0), "Default collator getRules failed"); 497 doAssert((u_strcmp(rule1, rule3) != 0), "Default collator getRules failed"); 498 499 col4=ucol_openRules(rule2, u_strlen(rule2), UCOL_DEFAULT, UCOL_DEFAULT_STRENGTH, NULL, &status); 500 if (U_FAILURE(status)) { 501 log_err("RuleBased Collator creation failed.: %s\n", myErrorName(status)); 502 return; 503 } 504 rule4= ucol_getRules(col4, &tempLength); 505 doAssert((u_strcmp(rule2, rule4) == 0), "Default collator getRules failed"); 506 507 ucol_close(col1); 508 ucol_close(col2); 509 ucol_close(col3); 510 ucol_close(col4); 511 512 /* tests that modifier ! is always ignored */ 513 u_uastrcpy(ruleset1, "!&a<b"); 514 teststr[0] = 0x0e40; 515 teststr[1] = 0x0e01; 516 teststr[2] = 0x0e2d; 517 col1 = ucol_openRules(ruleset1, u_strlen(ruleset1), UCOL_DEFAULT, UCOL_DEFAULT_STRENGTH, NULL, &status); 518 if (U_FAILURE(status)) { 519 log_err("RuleBased Collator creation failed.: %s\n", myErrorName(status)); 520 return; 521 } 522 col2 = ucol_open("en_US", &status); 523 if (U_FAILURE(status)) { 524 log_err("en_US Collator creation failed.: %s\n", myErrorName(status)); 525 return; 526 } 527 iter1 = ucol_openElements(col1, teststr, 3, &status); 528 iter2 = ucol_openElements(col2, teststr, 3, &status); 529 if(U_FAILURE(status)) { 530 log_err("ERROR: CollationElement iterator creation failed.: %s\n", myErrorName(status)); 531 return; 532 } 533 while (TRUE) { 534 /* testing with en since thai has its own tailoring */ 535 uint32_t ce = ucol_next(iter1, &status); 536 uint32_t ce2 = ucol_next(iter2, &status); 537 if(U_FAILURE(status)) { 538 log_err("ERROR: CollationElement iterator creation failed.: %s\n", myErrorName(status)); 539 return; 540 } 541 if (ce2 != ce) { 542 log_err("! modifier test failed"); 543 } 544 if (ce == UCOL_NULLORDER) { 545 break; 546 } 547 } 548 ucol_closeElements(iter1); 549 ucol_closeElements(iter2); 550 ucol_close(col1); 551 ucol_close(col2); 552 /* test that we can start a rule without a & or < */ 553 u_uastrcpy(ruleset1, "< z < a"); 554 col1 = ucol_openRules(ruleset1, u_strlen(ruleset1), UCOL_DEFAULT, UCOL_DEFAULT_STRENGTH, NULL, &status); 555 if (U_FAILURE(status)) { 556 log_err("RuleBased Collator creation failed.: %s\n", myErrorName(status)); 557 return; 558 } 559 u_uastrcpy(teststr, "z"); 560 u_uastrcpy(teststr2, "a"); 561 if (ucol_greaterOrEqual(col1, teststr, 1, teststr2, 1)) { 562 log_err("Rule \"z < a\" fails"); 563 } 564 ucol_close(col1); 565 } 566 567 void TestCompare() 568 { 569 UErrorCode status = U_ZERO_ERROR; 570 UCollator *col; 571 UChar* test1; 572 UChar* test2; 573 574 log_verbose("The compare tests begin : \n"); 575 status=U_ZERO_ERROR; 576 col = ucol_open("en_US", &status); 577 if(U_FAILURE(status)) { 578 log_err_status(status, "ucal_open() collation creation failed.: %s\n", myErrorName(status)); 579 return; 580 } 581 test1=(UChar*)malloc(sizeof(UChar) * 6); 582 test2=(UChar*)malloc(sizeof(UChar) * 6); 583 u_uastrcpy(test1, "Abcda"); 584 u_uastrcpy(test2, "abcda"); 585 586 log_verbose("Use tertiary comparison level testing ....\n"); 587 588 doAssert( (!ucol_equal(col, test1, u_strlen(test1), test2, u_strlen(test2))), "Result should be \"Abcda\" != \"abcda\" "); 589 doAssert( (ucol_greater(col, test1, u_strlen(test1), test2, u_strlen(test2))), "Result should be \"Abcda\" >>> \"abcda\" "); 590 doAssert( (ucol_greaterOrEqual(col, test1, u_strlen(test1), test2, u_strlen(test2))), "Result should be \"Abcda\" >>> \"abcda\""); 591 592 ucol_setStrength(col, UCOL_SECONDARY); 593 log_verbose("Use secondary comparison level testing ....\n"); 594 595 doAssert( (ucol_equal(col, test1, u_strlen(test1), test2, u_strlen(test2) )), "Result should be \"Abcda\" == \"abcda\""); 596 doAssert( (!ucol_greater(col, test1, u_strlen(test1), test2, u_strlen(test2))), "Result should be \"Abcda\" == \"abcda\""); 597 doAssert( (ucol_greaterOrEqual(col, test1, u_strlen(test1), test2, u_strlen(test2) )), "Result should be \"Abcda\" == \"abcda\""); 598 599 ucol_setStrength(col, UCOL_PRIMARY); 600 log_verbose("Use primary comparison level testing ....\n"); 601 602 doAssert( (ucol_equal(col, test1, u_strlen(test1), test2, u_strlen(test2))), "Result should be \"Abcda\" == \"abcda\""); 603 doAssert( (!ucol_greater(col, test1, u_strlen(test1), test2, u_strlen(test2))), "Result should be \"Abcda\" == \"abcda\""); 604 doAssert( (ucol_greaterOrEqual(col, test1, u_strlen(test1), test2, u_strlen(test2))), "Result should be \"Abcda\" == \"abcda\""); 605 606 607 log_verbose("The compare tests end.\n"); 608 ucol_close(col); 609 free(test1); 610 free(test2); 611 612 } 613 /* 614 --------------------------------------------- 615 tests decomposition setting 616 */ 617 void TestDecomposition() { 618 UErrorCode status = U_ZERO_ERROR; 619 UCollator *en_US, *el_GR, *vi_VN; 620 en_US = ucol_open("en_US", &status); 621 el_GR = ucol_open("el_GR", &status); 622 vi_VN = ucol_open("vi_VN", &status); 623 624 if (U_FAILURE(status)) { 625 log_err_status(status, "ERROR: collation creation failed.: %s\n", myErrorName(status)); 626 return; 627 } 628 629 if (ucol_getAttribute(vi_VN, UCOL_NORMALIZATION_MODE, &status) != UCOL_ON || 630 U_FAILURE(status)) 631 { 632 log_err("ERROR: vi_VN collation did not have cannonical decomposition for normalization!\n"); 633 } 634 635 status = U_ZERO_ERROR; 636 if (ucol_getAttribute(el_GR, UCOL_NORMALIZATION_MODE, &status) != UCOL_ON || 637 U_FAILURE(status)) 638 { 639 log_err("ERROR: el_GR collation did not have cannonical decomposition for normalization!\n"); 640 } 641 642 status = U_ZERO_ERROR; 643 if (ucol_getAttribute(en_US, UCOL_NORMALIZATION_MODE, &status) != UCOL_OFF || 644 U_FAILURE(status)) 645 { 646 log_err("ERROR: en_US collation had cannonical decomposition for normalization!\n"); 647 } 648 649 ucol_close(en_US); 650 ucol_close(el_GR); 651 ucol_close(vi_VN); 652 } 653 654 #define CLONETEST_COLLATOR_COUNT 4 655 656 void TestSafeClone() { 657 UChar test1[6]; 658 UChar test2[6]; 659 static const UChar umlautUStr[] = {0x00DC, 0}; 660 static const UChar oeStr[] = {0x0055, 0x0045, 0}; 661 UCollator * someCollators [CLONETEST_COLLATOR_COUNT]; 662 UCollator * someClonedCollators [CLONETEST_COLLATOR_COUNT]; 663 UCollator * col; 664 UErrorCode err = U_ZERO_ERROR; 665 int8_t index = 6; /* Leave this here to test buffer alingment in memory*/ 666 uint8_t buffer [CLONETEST_COLLATOR_COUNT] [U_COL_SAFECLONE_BUFFERSIZE]; 667 int32_t bufferSize = U_COL_SAFECLONE_BUFFERSIZE; 668 const char sampleRuleChars[] = "&Z < CH"; 669 UChar sampleRule[sizeof(sampleRuleChars)]; 670 671 if (TestBufferSize()) { 672 log_err("U_COL_SAFECLONE_BUFFERSIZE should be larger than sizeof(UCollator)\n"); 673 return; 674 } 675 676 u_uastrcpy(test1, "abCda"); 677 u_uastrcpy(test2, "abcda"); 678 u_uastrcpy(sampleRule, sampleRuleChars); 679 680 /* one default collator & two complex ones */ 681 someCollators[0] = ucol_open("en_US", &err); 682 someCollators[1] = ucol_open("ko", &err); 683 someCollators[2] = ucol_open("ja_JP", &err); 684 someCollators[3] = ucol_openRules(sampleRule, -1, UCOL_ON, UCOL_TERTIARY, NULL, &err); 685 if(U_FAILURE(err)) { 686 for (index = 0; index < CLONETEST_COLLATOR_COUNT; index++) { 687 ucol_close(someCollators[index]); 688 } 689 log_data_err("Couldn't open one or more collators\n"); 690 return; 691 } 692 693 /* Check the various error & informational states: */ 694 695 /* Null status - just returns NULL */ 696 if (0 != ucol_safeClone(someCollators[0], buffer[0], &bufferSize, 0)) 697 { 698 log_err("FAIL: Cloned Collator failed to deal correctly with null status\n"); 699 } 700 /* error status - should return 0 & keep error the same */ 701 err = U_MEMORY_ALLOCATION_ERROR; 702 if (0 != ucol_safeClone(someCollators[0], buffer[0], &bufferSize, &err) || err != U_MEMORY_ALLOCATION_ERROR) 703 { 704 log_err("FAIL: Cloned Collator failed to deal correctly with incoming error status\n"); 705 } 706 err = U_ZERO_ERROR; 707 708 /* Null buffer size pointer - just returns NULL & set error to U_ILLEGAL_ARGUMENT_ERROR*/ 709 if (0 != ucol_safeClone(someCollators[0], buffer[0], 0, &err) || err != U_ILLEGAL_ARGUMENT_ERROR) 710 { 711 log_err("FAIL: Cloned Collator failed to deal correctly with null bufferSize pointer\n"); 712 } 713 err = U_ZERO_ERROR; 714 715 /* buffer size pointer is 0 - fill in pbufferSize with a size */ 716 bufferSize = 0; 717 if (0 != ucol_safeClone(someCollators[0], buffer[0], &bufferSize, &err) || U_FAILURE(err) || bufferSize <= 0) 718 { 719 log_err("FAIL: Cloned Collator failed a sizing request ('preflighting')\n"); 720 } 721 /* Verify our define is large enough */ 722 if (U_COL_SAFECLONE_BUFFERSIZE < bufferSize) 723 { 724 log_err("FAIL: Pre-calculated buffer size is too small\n"); 725 } 726 /* Verify we can use this run-time calculated size */ 727 if (0 == (col = ucol_safeClone(someCollators[0], buffer[0], &bufferSize, &err)) || U_FAILURE(err)) 728 { 729 log_err("FAIL: Collator can't be cloned with run-time size\n"); 730 } 731 if (col) ucol_close(col); 732 /* size one byte too small - should allocate & let us know */ 733 --bufferSize; 734 if (0 == (col = ucol_safeClone(someCollators[0], 0, &bufferSize, &err)) || err != U_SAFECLONE_ALLOCATED_WARNING) 735 { 736 log_err("FAIL: Cloned Collator failed to deal correctly with too-small buffer size\n"); 737 } 738 if (col) ucol_close(col); 739 err = U_ZERO_ERROR; 740 bufferSize = U_COL_SAFECLONE_BUFFERSIZE; 741 742 743 /* Null buffer pointer - return Collator & set error to U_SAFECLONE_ALLOCATED_ERROR */ 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 null buffer pointer\n"); 747 } 748 if (col) ucol_close(col); 749 err = U_ZERO_ERROR; 750 751 /* Null Collator - return NULL & set U_ILLEGAL_ARGUMENT_ERROR */ 752 if (0 != ucol_safeClone(0, buffer[0], &bufferSize, &err) || err != U_ILLEGAL_ARGUMENT_ERROR) 753 { 754 log_err("FAIL: Cloned Collator failed to deal correctly with null Collator pointer\n"); 755 } 756 757 err = U_ZERO_ERROR; 758 759 /* Test that a cloned collator doesn't accidentally use UCA. */ 760 col=ucol_open("de@collation=phonebook", &err); 761 bufferSize = U_COL_SAFECLONE_BUFFERSIZE; 762 someClonedCollators[0] = ucol_safeClone(col, buffer[0], &bufferSize, &err); 763 doAssert( (ucol_greater(col, umlautUStr, u_strlen(umlautUStr), oeStr, u_strlen(oeStr))), "Original German phonebook collation sorts differently than expected"); 764 doAssert( (ucol_greater(someClonedCollators[0], umlautUStr, u_strlen(umlautUStr), oeStr, u_strlen(oeStr))), "Cloned German phonebook collation sorts differently than expected"); 765 if (!ucol_equals(someClonedCollators[0], col)) { 766 log_err("FAIL: Cloned German phonebook collator is not equal to original.\n"); 767 } 768 ucol_close(col); 769 ucol_close(someClonedCollators[0]); 770 771 err = U_ZERO_ERROR; 772 773 /* change orig & clone & make sure they are independent */ 774 775 for (index = 0; index < CLONETEST_COLLATOR_COUNT; index++) 776 { 777 ucol_setStrength(someCollators[index], UCOL_IDENTICAL); 778 bufferSize = 1; 779 err = U_ZERO_ERROR; 780 ucol_close(ucol_safeClone(someCollators[index], buffer[index], &bufferSize, &err)); 781 if (err != U_SAFECLONE_ALLOCATED_WARNING) { 782 log_err("FAIL: collator number %d was not allocated.\n", index); 783 log_err("FAIL: status of Collator[%d] is %d (hex: %x).\n", index, err, err); 784 } 785 786 bufferSize = U_COL_SAFECLONE_BUFFERSIZE; 787 err = U_ZERO_ERROR; 788 someClonedCollators[index] = ucol_safeClone(someCollators[index], buffer[index], &bufferSize, &err); 789 if (someClonedCollators[index] == NULL 790 || someClonedCollators[index] < (UCollator *)buffer[index] 791 || someClonedCollators[index] > (UCollator *)(buffer[index]+(U_COL_SAFECLONE_BUFFERSIZE-1))) 792 { 793 log_err("FAIL: Cloned collator didn't use provided buffer.\n"); 794 return; 795 } 796 if (!ucol_equals(someClonedCollators[index], someCollators[index])) { 797 log_err("FAIL: Cloned collator is not equal to original at index = %d.\n", index); 798 } 799 800 /* Check the usability */ 801 ucol_setStrength(someCollators[index], UCOL_PRIMARY); 802 ucol_setAttribute(someCollators[index], UCOL_CASE_LEVEL, UCOL_OFF, &err); 803 804 doAssert( (ucol_equal(someCollators[index], test1, u_strlen(test1), test2, u_strlen(test2))), "Result should be \"abcda\" == \"abCda\""); 805 806 /* Close the original to make sure that the clone is usable. */ 807 ucol_close(someCollators[index]); 808 809 ucol_setStrength(someClonedCollators[index], UCOL_TERTIARY); 810 ucol_setAttribute(someClonedCollators[index], UCOL_CASE_LEVEL, UCOL_OFF, &err); 811 doAssert( (ucol_greater(someClonedCollators[index], test1, u_strlen(test1), test2, u_strlen(test2))), "Result should be \"abCda\" >>> \"abcda\" "); 812 813 ucol_close(someClonedCollators[index]); 814 } 815 } 816 817 void TestCloneBinary(){ 818 UErrorCode err = U_ZERO_ERROR; 819 UCollator * col = ucol_open("en_US", &err); 820 UCollator * c; 821 int32_t size; 822 uint8_t * buffer; 823 824 if (U_FAILURE(err)) { 825 log_data_err("Couldn't open collator. Error: %s\n", u_errorName(err)); 826 return; 827 } 828 829 size = ucol_cloneBinary(col, NULL, 0, &err); 830 if(size==0 || err!=U_BUFFER_OVERFLOW_ERROR) { 831 log_err("ucol_cloneBinary - couldn't check size. Error: %s\n", u_errorName(err)); 832 return; 833 } 834 err = U_ZERO_ERROR; 835 836 buffer = (uint8_t *) malloc(size); 837 ucol_cloneBinary(col, buffer, size, &err); 838 if(U_FAILURE(err)) { 839 log_err("ucol_cloneBinary - couldn't clone.. Error: %s\n", u_errorName(err)); 840 free(buffer); 841 return; 842 } 843 844 /* how to check binary result ? */ 845 846 c = ucol_openBinary(buffer, size, col, &err); 847 if(U_FAILURE(err)) { 848 log_err("ucol_openBinary failed. Error: %s\n", u_errorName(err)); 849 } else { 850 UChar t[] = {0x41, 0x42, 0x43, 0}; /* ABC */ 851 uint8_t *k1, *k2; 852 int l1, l2; 853 l1 = ucol_getSortKey(col, t, -1, NULL,0); 854 l2 = ucol_getSortKey(c, t, -1, NULL,0); 855 k1 = (uint8_t *) malloc(sizeof(uint8_t) * l1); 856 k2 = (uint8_t *) malloc(sizeof(uint8_t) * l2); 857 ucol_getSortKey(col, t, -1, k1, l1); 858 ucol_getSortKey(col, t, -1, k2, l2); 859 if (strcmp((char *)k1,(char *)k2) != 0){ 860 log_err("ucol_openBinary - new collator should equal to old one\n"); 861 }; 862 free(k1); 863 free(k2); 864 } 865 free(buffer); 866 ucol_close(c); 867 ucol_close(col); 868 } 869 /* 870 TestOpenVsOpenRules ensures that collators from ucol_open and ucol_openRules 871 will generate identical sort keys 872 */ 873 void TestOpenVsOpenRules(){ 874 875 /* create an array of all the locales */ 876 int32_t numLocales = uloc_countAvailable(); 877 int32_t sizeOfStdSet; 878 uint32_t adder; 879 UChar str[41]; /* create an array of UChar of size maximum strSize + 1 */ 880 USet *stdSet; 881 char* curLoc; 882 UCollator * c1; 883 UCollator * c2; 884 const UChar* rules; 885 int32_t rulesLength; 886 int32_t sortKeyLen1, sortKeyLen2; 887 uint8_t *sortKey1 = NULL, *sortKey2 = NULL; 888 ULocaleData *uld; 889 int32_t x, y, z; 890 USet *eSet; 891 int32_t eSize; 892 int strSize; 893 894 UErrorCode err = U_ZERO_ERROR; 895 896 /* create a set of standard characters that aren't very interesting... 897 and then we can find some interesting ones later */ 898 899 stdSet = uset_open(0x61, 0x7A); 900 uset_addRange(stdSet, 0x41, 0x5A); 901 uset_addRange(stdSet, 0x30, 0x39); 902 sizeOfStdSet = uset_size(stdSet); 903 904 adder = 1; 905 if(getTestOption(QUICK_OPTION)) 906 { 907 adder = 10; 908 } 909 910 for(x = 0; x < numLocales; x+=adder){ 911 curLoc = (char *)uloc_getAvailable(x); 912 log_verbose("Processing %s\n", curLoc); 913 914 /* create a collator the normal API way */ 915 c1 = ucol_open(curLoc, &err); 916 if (U_FAILURE(err)) { 917 log_err("ERROR: Normal collation creation failed with locale: %s : %s\n", curLoc, myErrorName(err)); 918 return; 919 } 920 921 /* grab the rules */ 922 rules = ucol_getRules(c1, &rulesLength); 923 924 /* use those rules to create a collator from rules */ 925 c2 = ucol_openRules(rules, rulesLength, UCOL_DEFAULT, UCOL_DEFAULT_STRENGTH, NULL, &err); 926 if (U_FAILURE(err)) { 927 log_err("ERROR: Creating collator from rules failed with locale: %s : %s\n", curLoc, myErrorName(err)); 928 return; 929 } 930 931 uld = ulocdata_open(curLoc, &err); 932 933 /*now that we have some collators, we get several strings */ 934 935 for(y = 0; y < 5; y++){ 936 937 /* get a set of ALL the characters in this locale */ 938 eSet = ulocdata_getExemplarSet(uld, NULL, 0, ULOCDATA_ES_STANDARD, &err); 939 eSize = uset_size(eSet); 940 941 /* make a string with these characters in it */ 942 strSize = (rand()%40) + 1; 943 944 for(z = 0; z < strSize; z++){ 945 str[z] = uset_charAt(eSet, rand()%eSize); 946 } 947 948 /* change the set to only include 'abnormal' characters (not A-Z, a-z, 0-9 */ 949 uset_removeAll(eSet, stdSet); 950 eSize = uset_size(eSet); 951 952 /* if there are some non-normal characters left, put a few into the string, just to make sure we have some */ 953 if(eSize > 0){ 954 str[2%strSize] = uset_charAt(eSet, rand()%eSize); 955 str[3%strSize] = uset_charAt(eSet, rand()%eSize); 956 str[5%strSize] = uset_charAt(eSet, rand()%eSize); 957 str[10%strSize] = uset_charAt(eSet, rand()%eSize); 958 str[13%strSize] = uset_charAt(eSet, rand()%eSize); 959 } 960 /* terminate the string */ 961 str[strSize-1] = '\0'; 962 log_verbose("String used: %S\n", str); 963 964 /* get sort keys for both of them, and check that the keys are identicle */ 965 sortKeyLen1 = ucol_getSortKey(c1, str, u_strlen(str), NULL, 0); 966 sortKey1 = (uint8_t*)malloc(sizeof(uint8_t) * (sortKeyLen1 + 1)); 967 /*memset(sortKey1, 0xFE, sortKeyLen1);*/ 968 ucol_getSortKey(c1, str, u_strlen(str), sortKey1, sortKeyLen1 + 1); 969 970 sortKeyLen2 = ucol_getSortKey(c2, str, u_strlen(str), NULL, 0); 971 sortKey2 = (uint8_t*)malloc(sizeof(uint8_t) * (sortKeyLen2 + 1)); 972 /*memset(sortKey2, 0xFE, sortKeyLen2);*/ 973 ucol_getSortKey(c2, str, u_strlen(str), sortKey2, sortKeyLen2 + 1); 974 975 /* Check that the lengths are the same */ 976 doAssert((sortKeyLen1 == sortKeyLen2), "Sort key lengths do not match."); 977 978 /* check that the keys are the same */ 979 doAssert((memcmp(sortKey1, sortKey2, sortKeyLen1) == 0), "Keys are not equivalent"); 980 981 /* clean up after each string */ 982 free(sortKey1); 983 free(sortKey2); 984 uset_close(eSet); 985 } 986 /* clean up after each locale */ 987 ulocdata_close(uld); 988 ucol_close(c1); 989 ucol_close(c2); 990 } 991 /* final clean up */ 992 uset_close(stdSet); 993 } 994 /* 995 ---------------------------------------------------------------------------- 996 ctor -- Tests the getSortKey 997 */ 998 void TestSortKey() 999 { 1000 uint8_t *sortk1 = NULL, *sortk2 = NULL, *sortk3 = NULL, *sortkEmpty = NULL; 1001 int32_t sortklen, osortklen; 1002 uint32_t toStringLen=0; 1003 UCollator *col; 1004 UChar *test1, *test2, *test3; 1005 UErrorCode status = U_ZERO_ERROR; 1006 char toStringBuffer[256], *resultP; 1007 1008 1009 uint8_t s1[] = { 0x9f, 0x00 }; 1010 uint8_t s2[] = { 0x61, 0x00 }; 1011 int strcmpResult; 1012 1013 strcmpResult = strcmp((const char *)s1, (const char *)s2); 1014 log_verbose("strcmp(0x9f..., 0x61...) = %d\n", strcmpResult); 1015 1016 if(strcmpResult <= 0) { 1017 log_err("ERR: expected strcmp(\"9f 00\", \"61 00\") to be >=0 (GREATER).. got %d. Calling strcmp() for sortkeys may not work! \n", 1018 strcmpResult); 1019 } 1020 1021 1022 log_verbose("testing SortKey begins...\n"); 1023 /* this is supposed to open default date format, but later on it treats it like it is "en_US" 1024 - very bad if you try to run the tests on machine where default locale is NOT "en_US" */ 1025 /* col = ucol_open(NULL, &status); */ 1026 col = ucol_open("en_US", &status); 1027 if (U_FAILURE(status)) { 1028 log_err_status(status, "ERROR: Default collation creation failed.: %s\n", myErrorName(status)); 1029 return; 1030 } 1031 1032 1033 if(ucol_getStrength(col) != UCOL_DEFAULT_STRENGTH) 1034 { 1035 log_err("ERROR: default collation did not have UCOL_DEFAULT_STRENGTH !\n"); 1036 } 1037 /* Need to use identical strength */ 1038 ucol_setAttribute(col, UCOL_STRENGTH, UCOL_IDENTICAL, &status); 1039 1040 test1=(UChar*)malloc(sizeof(UChar) * 6); 1041 test2=(UChar*)malloc(sizeof(UChar) * 6); 1042 test3=(UChar*)malloc(sizeof(UChar) * 6); 1043 1044 memset(test1,0xFE, sizeof(UChar)*6); 1045 memset(test2,0xFE, sizeof(UChar)*6); 1046 memset(test3,0xFE, sizeof(UChar)*6); 1047 1048 1049 u_uastrcpy(test1, "Abcda"); 1050 u_uastrcpy(test2, "abcda"); 1051 u_uastrcpy(test3, "abcda"); 1052 1053 log_verbose("Use tertiary comparison level testing ....\n"); 1054 1055 sortklen=ucol_getSortKey(col, test1, u_strlen(test1), NULL, 0); 1056 sortk1=(uint8_t*)malloc(sizeof(uint8_t) * (sortklen+1)); 1057 memset(sortk1,0xFE, sortklen); 1058 ucol_getSortKey(col, test1, u_strlen(test1), sortk1, sortklen+1); 1059 1060 sortklen=ucol_getSortKey(col, test2, u_strlen(test2), NULL, 0); 1061 sortk2=(uint8_t*)malloc(sizeof(uint8_t) * (sortklen+1)); 1062 memset(sortk2,0xFE, sortklen); 1063 ucol_getSortKey(col, test2, u_strlen(test2), sortk2, sortklen+1); 1064 1065 osortklen = sortklen; 1066 sortklen=ucol_getSortKey(col, test2, u_strlen(test3), NULL, 0); 1067 sortk3=(uint8_t*)malloc(sizeof(uint8_t) * (sortklen+1)); 1068 memset(sortk3,0xFE, sortklen); 1069 ucol_getSortKey(col, test2, u_strlen(test2), sortk3, sortklen+1); 1070 1071 doAssert( (sortklen == osortklen), "Sortkey length should be the same (abcda, abcda)"); 1072 1073 doAssert( (memcmp(sortk1, sortk2, sortklen) > 0), "Result should be \"Abcda\" > \"abcda\""); 1074 doAssert( (memcmp(sortk2, sortk1, sortklen) < 0), "Result should be \"abcda\" < \"Abcda\""); 1075 doAssert( (memcmp(sortk2, sortk3, sortklen) == 0), "Result should be \"abcda\" == \"abcda\""); 1076 1077 resultP = ucol_sortKeyToString(col, sortk3, toStringBuffer, &toStringLen); 1078 doAssert( (resultP != 0), "sortKeyToString failed!"); 1079 1080 #if 1 /* verobse log of sortkeys */ 1081 { 1082 char junk2[1000]; 1083 char junk3[1000]; 1084 int i; 1085 1086 strcpy(junk2, "abcda[2] "); 1087 strcpy(junk3, " abcda[3] "); 1088 1089 for(i=0;i<sortklen;i++) 1090 { 1091 sprintf(junk2+strlen(junk2), "%02X ",(int)( 0xFF & sortk2[i])); 1092 sprintf(junk3+strlen(junk3), "%02X ",(int)( 0xFF & sortk3[i])); 1093 } 1094 1095 log_verbose("%s\n", junk2); 1096 log_verbose("%s\n", junk3); 1097 } 1098 #endif 1099 1100 free(sortk1); 1101 free(sortk2); 1102 free(sortk3); 1103 1104 log_verbose("Use secondary comparision level testing ...\n"); 1105 ucol_setStrength(col, UCOL_SECONDARY); 1106 sortklen=ucol_getSortKey(col, test1, u_strlen(test1), NULL, 0); 1107 sortk1=(uint8_t*)malloc(sizeof(uint8_t) * (sortklen+1)); 1108 ucol_getSortKey(col, test1, u_strlen(test1), sortk1, sortklen+1); 1109 sortklen=ucol_getSortKey(col, test2, u_strlen(test2), NULL, 0); 1110 sortk2=(uint8_t*)malloc(sizeof(uint8_t) * (sortklen+1)); 1111 ucol_getSortKey(col, test2, u_strlen(test2), sortk2, sortklen+1); 1112 1113 doAssert( !(memcmp(sortk1, sortk2, sortklen) > 0), "Result should be \"Abcda\" == \"abcda\""); 1114 doAssert( !(memcmp(sortk2, sortk1, sortklen) < 0), "Result should be \"abcda\" == \"Abcda\""); 1115 doAssert( (memcmp(sortk1, sortk2, sortklen) == 0), "Result should be \"abcda\" == \"abcda\""); 1116 1117 log_verbose("getting sortkey for an empty string\n"); 1118 ucol_setAttribute(col, UCOL_STRENGTH, UCOL_TERTIARY, &status); 1119 sortklen = ucol_getSortKey(col, test1, 0, NULL, 0); 1120 sortkEmpty = (uint8_t*)malloc(sizeof(uint8_t) * sortklen+1); 1121 sortklen = ucol_getSortKey(col, test1, 0, sortkEmpty, sortklen+1); 1122 if(sortklen != 3 || sortkEmpty[0] != 1 || sortkEmpty[0] != 1 || sortkEmpty[2] != 0) { 1123 log_err("Empty string generated wrong sortkey!\n"); 1124 } 1125 free(sortkEmpty); 1126 1127 log_verbose("testing passing invalid string\n"); 1128 sortklen = ucol_getSortKey(col, NULL, 0, NULL, 0); 1129 if(sortklen != 0) { 1130 log_err("Invalid string didn't return sortkey size of 0\n"); 1131 } 1132 1133 1134 log_verbose("testing sortkey ends...\n"); 1135 ucol_close(col); 1136 free(test1); 1137 free(test2); 1138 free(test3); 1139 free(sortk1); 1140 free(sortk2); 1141 1142 } 1143 void TestHashCode() 1144 { 1145 uint8_t *sortk1, *sortk2, *sortk3; 1146 int32_t sortk1len, sortk2len, sortk3len; 1147 UCollator *col; 1148 UChar *test1, *test2, *test3; 1149 UErrorCode status = U_ZERO_ERROR; 1150 log_verbose("testing getHashCode begins...\n"); 1151 col = ucol_open("en_US", &status); 1152 if (U_FAILURE(status)) { 1153 log_err_status(status, "ERROR: Default collation creation failed.: %s\n", myErrorName(status)); 1154 return; 1155 } 1156 test1=(UChar*)malloc(sizeof(UChar) * 6); 1157 test2=(UChar*)malloc(sizeof(UChar) * 6); 1158 test3=(UChar*)malloc(sizeof(UChar) * 6); 1159 u_uastrcpy(test1, "Abcda"); 1160 u_uastrcpy(test2, "abcda"); 1161 u_uastrcpy(test3, "abcda"); 1162 1163 log_verbose("Use tertiary comparison level testing ....\n"); 1164 sortk1len=ucol_getSortKey(col, test1, u_strlen(test1), NULL, 0); 1165 sortk1=(uint8_t*)malloc(sizeof(uint8_t) * (sortk1len+1)); 1166 ucol_getSortKey(col, test1, u_strlen(test1), sortk1, sortk1len+1); 1167 sortk2len=ucol_getSortKey(col, test2, u_strlen(test2), NULL, 0); 1168 sortk2=(uint8_t*)malloc(sizeof(uint8_t) * (sortk2len+1)); 1169 ucol_getSortKey(col, test2, u_strlen(test2), sortk2, sortk2len+1); 1170 sortk3len=ucol_getSortKey(col, test2, u_strlen(test3), NULL, 0); 1171 sortk3=(uint8_t*)malloc(sizeof(uint8_t) * (sortk3len+1)); 1172 ucol_getSortKey(col, test2, u_strlen(test2), sortk3, sortk3len+1); 1173 1174 1175 log_verbose("ucol_hashCode() testing ...\n"); 1176 1177 doAssert( ucol_keyHashCode(sortk1, sortk1len) != ucol_keyHashCode(sortk2, sortk2len), "Hash test1 result incorrect" ); 1178 doAssert( !(ucol_keyHashCode(sortk1, sortk1len) == ucol_keyHashCode(sortk2, sortk2len)), "Hash test2 result incorrect" ); 1179 doAssert( ucol_keyHashCode(sortk2, sortk2len) == ucol_keyHashCode(sortk3, sortk3len), "Hash result not equal" ); 1180 1181 log_verbose("hashCode tests end.\n"); 1182 ucol_close(col); 1183 free(sortk1); 1184 free(sortk2); 1185 free(sortk3); 1186 free(test1); 1187 free(test2); 1188 free(test3); 1189 1190 1191 } 1192 /* 1193 *---------------------------------------------------------------------------- 1194 * Tests the UCollatorElements API. 1195 * 1196 */ 1197 void TestElemIter() 1198 { 1199 int32_t offset; 1200 int32_t order1, order2, order3; 1201 UChar *testString1, *testString2; 1202 UCollator *col; 1203 UCollationElements *iterator1, *iterator2, *iterator3; 1204 UErrorCode status = U_ZERO_ERROR; 1205 log_verbose("testing UCollatorElements begins...\n"); 1206 col = ucol_open("en_US", &status); 1207 ucol_setAttribute(col, UCOL_NORMALIZATION_MODE, UCOL_OFF, &status); 1208 if (U_FAILURE(status)) { 1209 log_err_status(status, "ERROR: Default collation creation failed.: %s\n", myErrorName(status)); 1210 return; 1211 } 1212 1213 testString1=(UChar*)malloc(sizeof(UChar) * 150); 1214 testString2=(UChar*)malloc(sizeof(UChar) * 150); 1215 u_uastrcpy(testString1, "XFILE What subset of all possible test cases has the highest probability of detecting the most errors?"); 1216 u_uastrcpy(testString2, "Xf_ile What subset of all possible test cases has the lowest probability of detecting the least errors?"); 1217 1218 log_verbose("Constructors and comparison testing....\n"); 1219 1220 iterator1 = ucol_openElements(col, testString1, u_strlen(testString1), &status); 1221 if(U_FAILURE(status)) { 1222 log_err("ERROR: Default collationElement iterator creation failed.: %s\n", myErrorName(status)); 1223 ucol_close(col); 1224 return; 1225 } 1226 else{ log_verbose("PASS: Default collationElement iterator1 creation passed\n");} 1227 1228 iterator2 = ucol_openElements(col, testString1, u_strlen(testString1), &status); 1229 if(U_FAILURE(status)) { 1230 log_err("ERROR: Default collationElement iterator creation failed.: %s\n", myErrorName(status)); 1231 ucol_close(col); 1232 return; 1233 } 1234 else{ log_verbose("PASS: Default collationElement iterator2 creation passed\n");} 1235 1236 iterator3 = ucol_openElements(col, testString2, u_strlen(testString2), &status); 1237 if(U_FAILURE(status)) { 1238 log_err("ERROR: Default collationElement iterator creation failed.: %s\n", myErrorName(status)); 1239 ucol_close(col); 1240 return; 1241 } 1242 else{ log_verbose("PASS: Default collationElement iterator3 creation passed\n");} 1243 1244 offset=ucol_getOffset(iterator1); 1245 ucol_setOffset(iterator1, 6, &status); 1246 if (U_FAILURE(status)) { 1247 log_err("Error in setOffset for UCollatorElements iterator.: %s\n", myErrorName(status)); 1248 return; 1249 } 1250 if(ucol_getOffset(iterator1)==6) 1251 log_verbose("setOffset and getOffset working fine\n"); 1252 else{ 1253 log_err("error in set and get Offset got %d instead of 6\n", ucol_getOffset(iterator1)); 1254 } 1255 1256 ucol_setOffset(iterator1, 0, &status); 1257 order1 = ucol_next(iterator1, &status); 1258 if (U_FAILURE(status)) { 1259 log_err("Somehow ran out of memory stepping through the iterator1.: %s\n", myErrorName(status)); 1260 return; 1261 } 1262 order2=ucol_getOffset(iterator2); 1263 doAssert((order1 != order2), "The first iterator advance failed"); 1264 order2 = ucol_next(iterator2, &status); 1265 if (U_FAILURE(status)) { 1266 log_err("Somehow ran out of memory stepping through the iterator2.: %s\n", myErrorName(status)); 1267 return; 1268 } 1269 order3 = ucol_next(iterator3, &status); 1270 if (U_FAILURE(status)) { 1271 log_err("Somehow ran out of memory stepping through the iterator3.: %s\n", myErrorName(status)); 1272 return; 1273 } 1274 1275 doAssert((order1 == order2), "The second iterator advance failed should be the same as first one"); 1276 1277 doAssert( (ucol_primaryOrder(order1) == ucol_primaryOrder(order3)), "The primary orders should be identical"); 1278 doAssert( (ucol_secondaryOrder(order1) == ucol_secondaryOrder(order3)), "The secondary orders should be identical"); 1279 doAssert( (ucol_tertiaryOrder(order1) == ucol_tertiaryOrder(order3)), "The tertiary orders should be identical"); 1280 1281 order1=ucol_next(iterator1, &status); 1282 if (U_FAILURE(status)) { 1283 log_err("Somehow ran out of memory stepping through the iterator2.: %s\n", myErrorName(status)); 1284 return; 1285 } 1286 order3=ucol_next(iterator3, &status); 1287 if (U_FAILURE(status)) { 1288 log_err("Somehow ran out of memory stepping through the iterator2.: %s\n", myErrorName(status)); 1289 return; 1290 } 1291 doAssert( (ucol_primaryOrder(order1) == ucol_primaryOrder(order3)), "The primary orders should be identical"); 1292 doAssert( (ucol_tertiaryOrder(order1) != ucol_tertiaryOrder(order3)), "The tertiary orders should be different"); 1293 1294 order1=ucol_next(iterator1, &status); 1295 if (U_FAILURE(status)) { 1296 log_err("Somehow ran out of memory stepping through the iterator2.: %s\n", myErrorName(status)); 1297 return; 1298 } 1299 order3=ucol_next(iterator3, &status); 1300 if (U_FAILURE(status)) { 1301 log_err("Somehow ran out of memory stepping through the iterator2.: %s\n", myErrorName(status)); 1302 return; 1303 } 1304 /* this here, my friends, is either pure lunacy or something so obsolete that even it's mother 1305 * doesn't care about it. Essentialy, this test complains if secondary values for 'I' and '_' 1306 * are the same. According to the UCA, this is not true. Therefore, remove the test. 1307 * Besides, if primary strengths for two code points are different, it doesn't matter one bit 1308 * what is the relation between secondary or any other strengths. 1309 * killed by weiv 06/11/2002. 1310 */ 1311 /* 1312 doAssert( ((order1 & UCOL_SECONDARYMASK) != (order3 & UCOL_SECONDARYMASK)), "The secondary orders should be different"); 1313 */ 1314 doAssert( (order1 != UCOL_NULLORDER), "Unexpected end of iterator reached"); 1315 1316 free(testString1); 1317 free(testString2); 1318 ucol_closeElements(iterator1); 1319 ucol_closeElements(iterator2); 1320 ucol_closeElements(iterator3); 1321 ucol_close(col); 1322 1323 log_verbose("testing CollationElementIterator ends...\n"); 1324 } 1325 1326 void TestGetLocale() { 1327 UErrorCode status = U_ZERO_ERROR; 1328 const char *rules = "&a<x<y<z"; 1329 UChar rlz[256] = {0}; 1330 uint32_t rlzLen = u_unescape(rules, rlz, 256); 1331 1332 UCollator *coll = NULL; 1333 const char *locale = NULL; 1334 1335 int32_t i = 0; 1336 1337 /* Now that the collation tree is separate, actual==valid at all times. [alan] */ 1338 static const struct { 1339 const char* requestedLocale; 1340 const char* validLocale; 1341 const char* actualLocale; 1342 } testStruct[] = { 1343 { "sr_RS", "sr_Cyrl_RS", "sr" }, 1344 { "sh_YU", "sr_Latn_RS", "hr" }, /* this used to be sh, but now sh collation aliases hr */ 1345 { "en_BE_FOO", "en_BE", "root" }, 1346 { "de_DE_NONEXISTANT", "de_DE", "de" } 1347 }; 1348 1349 /* test opening collators for different locales */ 1350 for(i = 0; i<sizeof(testStruct)/sizeof(testStruct[0]); i++) { 1351 status = U_ZERO_ERROR; 1352 coll = ucol_open(testStruct[i].requestedLocale, &status); 1353 if(U_FAILURE(status)) { 1354 log_err_status(status, "Failed to open collator for %s with %s\n", testStruct[i].requestedLocale, u_errorName(status)); 1355 ucol_close(coll); 1356 continue; 1357 } 1358 locale = ucol_getLocaleByType(coll, ULOC_REQUESTED_LOCALE, &status); 1359 if(strcmp(locale, testStruct[i].requestedLocale) != 0) { 1360 log_err("[Coll %s]: Error in requested locale, expected %s, got %s\n", testStruct[i].requestedLocale, testStruct[i].requestedLocale, locale); 1361 } 1362 locale = ucol_getLocaleByType(coll, ULOC_VALID_LOCALE, &status); 1363 if(strcmp(locale, testStruct[i].validLocale) != 0) { 1364 log_err("[Coll %s]: Error in valid locale, expected %s, got %s\n", testStruct[i].requestedLocale, testStruct[i].validLocale, locale); 1365 } 1366 locale = ucol_getLocaleByType(coll, ULOC_ACTUAL_LOCALE, &status); 1367 if(strcmp(locale, testStruct[i].actualLocale) != 0) { 1368 log_err("[Coll %s]: Error in actual locale, expected %s, got %s\n", testStruct[i].requestedLocale, testStruct[i].actualLocale, locale); 1369 } 1370 ucol_close(coll); 1371 } 1372 1373 /* completely non-existant locale for collator should get a default collator */ 1374 { 1375 UCollator *defaultColl = ucol_open(NULL, &status); 1376 coll = ucol_open("blahaha", &status); 1377 if(U_SUCCESS(status)) { 1378 if(strcmp(ucol_getLocaleByType(coll, ULOC_REQUESTED_LOCALE, &status), "blahaha")) { 1379 log_err("Nonexisting locale didn't preserve the requested locale\n"); 1380 } 1381 if(strcmp(ucol_getLocaleByType(coll, ULOC_VALID_LOCALE, &status), 1382 ucol_getLocaleByType(defaultColl, ULOC_VALID_LOCALE, &status))) { 1383 log_err("Valid locale for nonexisting locale locale collator differs " 1384 "from valid locale for default collator\n"); 1385 } 1386 if(strcmp(ucol_getLocaleByType(coll, ULOC_ACTUAL_LOCALE, &status), 1387 ucol_getLocaleByType(defaultColl, ULOC_ACTUAL_LOCALE, &status))) { 1388 log_err("Actual locale for nonexisting locale locale collator differs " 1389 "from actual locale for default collator\n"); 1390 } 1391 ucol_close(coll); 1392 ucol_close(defaultColl); 1393 } else { 1394 log_data_err("Couldn't open collators\n"); 1395 } 1396 } 1397 1398 1399 1400 /* collator instantiated from rules should have all three locales NULL */ 1401 coll = ucol_openRules(rlz, rlzLen, UCOL_DEFAULT, UCOL_DEFAULT, NULL, &status); 1402 locale = ucol_getLocaleByType(coll, ULOC_REQUESTED_LOCALE, &status); 1403 if(locale != NULL) { 1404 log_err("For collator instantiated from rules, requested locale returned %s instead of NULL\n", locale); 1405 } 1406 locale = ucol_getLocaleByType(coll, ULOC_VALID_LOCALE, &status); 1407 if(locale != NULL) { 1408 log_err("For collator instantiated from rules, valid locale returned %s instead of NULL\n", locale); 1409 } 1410 locale = ucol_getLocaleByType(coll, ULOC_ACTUAL_LOCALE, &status); 1411 if(locale != NULL) { 1412 log_err("For collator instantiated from rules, actual locale returned %s instead of NULL\n", locale); 1413 } 1414 ucol_close(coll); 1415 1416 } 1417 1418 1419 void TestGetAll() 1420 { 1421 int32_t i, count; 1422 count=ucol_countAvailable(); 1423 /* use something sensible w/o hardcoding the count */ 1424 if(count < 0){ 1425 log_err("Error in countAvailable(), it returned %d\n", count); 1426 } 1427 else{ 1428 log_verbose("PASS: countAvailable() successful, it returned %d\n", count); 1429 } 1430 for(i=0;i<count;i++) 1431 log_verbose("%s\n", ucol_getAvailable(i)); 1432 1433 1434 } 1435 1436 1437 struct teststruct { 1438 const char *original; 1439 uint8_t key[256]; 1440 } ; 1441 1442 static int compare_teststruct(const void *string1, const void *string2) { 1443 return(strcmp((const char *)((struct teststruct *)string1)->key, (const char *)((struct teststruct *)string2)->key)); 1444 } 1445 1446 void TestBounds() { 1447 UErrorCode status = U_ZERO_ERROR; 1448 1449 UCollator *coll = ucol_open("sh", &status); 1450 1451 uint8_t sortkey[512], lower[512], upper[512]; 1452 UChar buffer[512]; 1453 1454 static const char * const test[] = { 1455 "John Smith", 1456 "JOHN SMITH", 1457 "john SMITH", 1458 "j\\u00F6hn sm\\u00EFth", 1459 "J\\u00F6hn Sm\\u00EFth", 1460 "J\\u00D6HN SM\\u00CFTH", 1461 "john smithsonian", 1462 "John Smithsonian", 1463 }; 1464 1465 struct teststruct tests[] = { 1466 {"\\u010CAKI MIHALJ" } , 1467 {"\\u010CAKI MIHALJ" } , 1468 {"\\u010CAKI PIRO\\u0160KA" }, 1469 {"\\u010CABAI ANDRIJA" } , 1470 {"\\u010CABAI LAJO\\u0160" } , 1471 {"\\u010CABAI MARIJA" } , 1472 {"\\u010CABAI STEVAN" } , 1473 {"\\u010CABAI STEVAN" } , 1474 {"\\u010CABARKAPA BRANKO" } , 1475 {"\\u010CABARKAPA MILENKO" } , 1476 {"\\u010CABARKAPA MIROSLAV" } , 1477 {"\\u010CABARKAPA SIMO" } , 1478 {"\\u010CABARKAPA STANKO" } , 1479 {"\\u010CABARKAPA TAMARA" } , 1480 {"\\u010CABARKAPA TOMA\\u0160" } , 1481 {"\\u010CABDARI\\u0106 NIKOLA" } , 1482 {"\\u010CABDARI\\u0106 ZORICA" } , 1483 {"\\u010CABI NANDOR" } , 1484 {"\\u010CABOVI\\u0106 MILAN" } , 1485 {"\\u010CABRADI AGNEZIJA" } , 1486 {"\\u010CABRADI IVAN" } , 1487 {"\\u010CABRADI JELENA" } , 1488 {"\\u010CABRADI LJUBICA" } , 1489 {"\\u010CABRADI STEVAN" } , 1490 {"\\u010CABRDA MARTIN" } , 1491 {"\\u010CABRILO BOGDAN" } , 1492 {"\\u010CABRILO BRANISLAV" } , 1493 {"\\u010CABRILO LAZAR" } , 1494 {"\\u010CABRILO LJUBICA" } , 1495 {"\\u010CABRILO SPASOJA" } , 1496 {"\\u010CADE\\u0160 ZDENKA" } , 1497 {"\\u010CADESKI BLAGOJE" } , 1498 {"\\u010CADOVSKI VLADIMIR" } , 1499 {"\\u010CAGLJEVI\\u0106 TOMA" } , 1500 {"\\u010CAGOROVI\\u0106 VLADIMIR" } , 1501 {"\\u010CAJA VANKA" } , 1502 {"\\u010CAJI\\u0106 BOGOLJUB" } , 1503 {"\\u010CAJI\\u0106 BORISLAV" } , 1504 {"\\u010CAJI\\u0106 RADOSLAV" } , 1505 {"\\u010CAK\\u0160IRAN MILADIN" } , 1506 {"\\u010CAKAN EUGEN" } , 1507 {"\\u010CAKAN EVGENIJE" } , 1508 {"\\u010CAKAN IVAN" } , 1509 {"\\u010CAKAN JULIJAN" } , 1510 {"\\u010CAKAN MIHAJLO" } , 1511 {"\\u010CAKAN STEVAN" } , 1512 {"\\u010CAKAN VLADIMIR" } , 1513 {"\\u010CAKAN VLADIMIR" } , 1514 {"\\u010CAKAN VLADIMIR" } , 1515 {"\\u010CAKARA ANA" } , 1516 {"\\u010CAKAREVI\\u0106 MOMIR" } , 1517 {"\\u010CAKAREVI\\u0106 NEDELJKO" } , 1518 {"\\u010CAKI \\u0160ANDOR" } , 1519 {"\\u010CAKI AMALIJA" } , 1520 {"\\u010CAKI ANDRA\\u0160" } , 1521 {"\\u010CAKI LADISLAV" } , 1522 {"\\u010CAKI LAJO\\u0160" } , 1523 {"\\u010CAKI LASLO" } , 1524 }; 1525 1526 1527 1528 int32_t i = 0, j = 0, k = 0, buffSize = 0, skSize = 0, lowerSize = 0, upperSize = 0; 1529 int32_t arraySize = sizeof(tests)/sizeof(tests[0]); 1530 1531 if(U_SUCCESS(status) && coll) { 1532 for(i = 0; i<arraySize; i++) { 1533 buffSize = u_unescape(tests[i].original, buffer, 512); 1534 skSize = ucol_getSortKey(coll, buffer, buffSize, tests[i].key, 512); 1535 } 1536 1537 qsort(tests, arraySize, sizeof(struct teststruct), compare_teststruct); 1538 1539 for(i = 0; i < arraySize-1; i++) { 1540 for(j = i+1; j < arraySize; j++) { 1541 lowerSize = ucol_getBound(tests[i].key, -1, UCOL_BOUND_LOWER, 1, lower, 512, &status); 1542 upperSize = ucol_getBound(tests[j].key, -1, UCOL_BOUND_UPPER, 1, upper, 512, &status); 1543 for(k = i; k <= j; k++) { 1544 if(strcmp((const char *)lower, (const char *)tests[k].key) > 0) { 1545 log_err("Problem with lower! j = %i (%s vs %s)\n", k, tests[k].original, tests[i].original); 1546 } 1547 if(strcmp((const char *)upper, (const char *)tests[k].key) <= 0) { 1548 log_err("Problem with upper! j = %i (%s vs %s)\n", k, tests[k].original, tests[j].original); 1549 } 1550 } 1551 } 1552 } 1553 1554 1555 #if 0 1556 for(i = 0; i < 1000; i++) { 1557 lowerRND = (rand()/(RAND_MAX/arraySize)); 1558 upperRND = lowerRND + (rand()/(RAND_MAX/(arraySize-lowerRND))); 1559 1560 lowerSize = ucol_getBound(tests[lowerRND].key, -1, UCOL_BOUND_LOWER, 1, lower, 512, &status); 1561 upperSize = ucol_getBound(tests[upperRND].key, -1, UCOL_BOUND_UPPER_LONG, 1, upper, 512, &status); 1562 1563 for(j = lowerRND; j<=upperRND; j++) { 1564 if(strcmp(lower, tests[j].key) > 0) { 1565 log_err("Problem with lower! j = %i (%s vs %s)\n", j, tests[j].original, tests[lowerRND].original); 1566 } 1567 if(strcmp(upper, tests[j].key) <= 0) { 1568 log_err("Problem with upper! j = %i (%s vs %s)\n", j, tests[j].original, tests[upperRND].original); 1569 } 1570 } 1571 } 1572 #endif 1573 1574 1575 1576 1577 1578 for(i = 0; i<sizeof(test)/sizeof(test[0]); i++) { 1579 buffSize = u_unescape(test[i], buffer, 512); 1580 skSize = ucol_getSortKey(coll, buffer, buffSize, sortkey, 512); 1581 lowerSize = ucol_getBound(sortkey, skSize, UCOL_BOUND_LOWER, 1, lower, 512, &status); 1582 upperSize = ucol_getBound(sortkey, skSize, UCOL_BOUND_UPPER_LONG, 1, upper, 512, &status); 1583 for(j = i+1; j<sizeof(test)/sizeof(test[0]); j++) { 1584 buffSize = u_unescape(test[j], buffer, 512); 1585 skSize = ucol_getSortKey(coll, buffer, buffSize, sortkey, 512); 1586 if(strcmp((const char *)lower, (const char *)sortkey) > 0) { 1587 log_err("Problem with lower! i = %i, j = %i (%s vs %s)\n", i, j, test[i], test[j]); 1588 } 1589 if(strcmp((const char *)upper, (const char *)sortkey) <= 0) { 1590 log_err("Problem with upper! i = %i, j = %i (%s vs %s)\n", i, j, test[i], test[j]); 1591 } 1592 } 1593 } 1594 ucol_close(coll); 1595 } else { 1596 log_data_err("Couldn't open collator\n"); 1597 } 1598 1599 } 1600 1601 static void doOverrunTest(UCollator *coll, const UChar *uString, int32_t strLen) { 1602 int32_t skLen = 0, skLen2 = 0; 1603 uint8_t sortKey[256]; 1604 int32_t i, j; 1605 uint8_t filler = 0xFF; 1606 1607 skLen = ucol_getSortKey(coll, uString, strLen, NULL, 0); 1608 1609 for(i = 0; i < skLen; i++) { 1610 memset(sortKey, filler, 256); 1611 skLen2 = ucol_getSortKey(coll, uString, strLen, sortKey, i); 1612 if(skLen != skLen2) { 1613 log_err("For buffer size %i, got different sortkey length. Expected %i got %i\n", i, skLen, skLen2); 1614 } 1615 for(j = i; j < 256; j++) { 1616 if(sortKey[j] != filler) { 1617 log_err("Something run over index %i\n", j); 1618 break; 1619 } 1620 } 1621 } 1622 } 1623 1624 /* j1865 reports that if a shorter buffer is passed to 1625 * to get sort key, a buffer overrun happens in some 1626 * cases. This test tries to check this. 1627 */ 1628 void TestSortKeyBufferOverrun(void) { 1629 UErrorCode status = U_ZERO_ERROR; 1630 const char* cString = "A very Merry liTTle-lamB.."; 1631 UChar uString[256]; 1632 int32_t strLen = 0; 1633 UCollator *coll = ucol_open("root", &status); 1634 strLen = u_unescape(cString, uString, 256); 1635 1636 if(U_SUCCESS(status)) { 1637 log_verbose("testing non ignorable\n"); 1638 ucol_setAttribute(coll, UCOL_ALTERNATE_HANDLING, UCOL_NON_IGNORABLE, &status); 1639 doOverrunTest(coll, uString, strLen); 1640 1641 log_verbose("testing shifted\n"); 1642 ucol_setAttribute(coll, UCOL_ALTERNATE_HANDLING, UCOL_SHIFTED, &status); 1643 doOverrunTest(coll, uString, strLen); 1644 1645 log_verbose("testing shifted quaternary\n"); 1646 ucol_setAttribute(coll, UCOL_STRENGTH, UCOL_QUATERNARY, &status); 1647 doOverrunTest(coll, uString, strLen); 1648 1649 log_verbose("testing with french secondaries\n"); 1650 ucol_setAttribute(coll, UCOL_FRENCH_COLLATION, UCOL_ON, &status); 1651 ucol_setAttribute(coll, UCOL_STRENGTH, UCOL_TERTIARY, &status); 1652 ucol_setAttribute(coll, UCOL_ALTERNATE_HANDLING, UCOL_NON_IGNORABLE, &status); 1653 doOverrunTest(coll, uString, strLen); 1654 1655 } 1656 ucol_close(coll); 1657 } 1658 1659 static void TestAttribute() 1660 { 1661 UErrorCode error = U_ZERO_ERROR; 1662 UCollator *coll = ucol_open(NULL, &error); 1663 1664 if (U_FAILURE(error)) { 1665 log_err_status(error, "Creation of default collator failed\n"); 1666 return; 1667 } 1668 1669 ucol_setAttribute(coll, UCOL_FRENCH_COLLATION, UCOL_OFF, &error); 1670 if (ucol_getAttribute(coll, UCOL_FRENCH_COLLATION, &error) != UCOL_OFF || 1671 U_FAILURE(error)) { 1672 log_err_status(error, "Setting and retrieving of the french collation failed\n"); 1673 } 1674 1675 ucol_setAttribute(coll, UCOL_FRENCH_COLLATION, UCOL_ON, &error); 1676 if (ucol_getAttribute(coll, UCOL_FRENCH_COLLATION, &error) != UCOL_ON || 1677 U_FAILURE(error)) { 1678 log_err_status(error, "Setting and retrieving of the french collation failed\n"); 1679 } 1680 1681 ucol_setAttribute(coll, UCOL_ALTERNATE_HANDLING, UCOL_SHIFTED, &error); 1682 if (ucol_getAttribute(coll, UCOL_ALTERNATE_HANDLING, &error) != UCOL_SHIFTED || 1683 U_FAILURE(error)) { 1684 log_err_status(error, "Setting and retrieving of the alternate handling failed\n"); 1685 } 1686 1687 ucol_setAttribute(coll, UCOL_ALTERNATE_HANDLING, UCOL_NON_IGNORABLE, &error); 1688 if (ucol_getAttribute(coll, UCOL_ALTERNATE_HANDLING, &error) != UCOL_NON_IGNORABLE || 1689 U_FAILURE(error)) { 1690 log_err_status(error, "Setting and retrieving of the alternate handling failed\n"); 1691 } 1692 1693 ucol_setAttribute(coll, UCOL_CASE_FIRST, UCOL_LOWER_FIRST, &error); 1694 if (ucol_getAttribute(coll, UCOL_CASE_FIRST, &error) != UCOL_LOWER_FIRST || 1695 U_FAILURE(error)) { 1696 log_err_status(error, "Setting and retrieving of the case first attribute failed\n"); 1697 } 1698 1699 ucol_setAttribute(coll, UCOL_CASE_FIRST, UCOL_UPPER_FIRST, &error); 1700 if (ucol_getAttribute(coll, UCOL_CASE_FIRST, &error) != UCOL_UPPER_FIRST || 1701 U_FAILURE(error)) { 1702 log_err_status(error, "Setting and retrieving of the case first attribute failed\n"); 1703 } 1704 1705 ucol_setAttribute(coll, UCOL_CASE_LEVEL, UCOL_ON, &error); 1706 if (ucol_getAttribute(coll, UCOL_CASE_LEVEL, &error) != UCOL_ON || 1707 U_FAILURE(error)) { 1708 log_err_status(error, "Setting and retrieving of the case level attribute failed\n"); 1709 } 1710 1711 ucol_setAttribute(coll, UCOL_CASE_LEVEL, UCOL_OFF, &error); 1712 if (ucol_getAttribute(coll, UCOL_CASE_LEVEL, &error) != UCOL_OFF || 1713 U_FAILURE(error)) { 1714 log_err_status(error, "Setting and retrieving of the case level attribute failed\n"); 1715 } 1716 1717 ucol_setAttribute(coll, UCOL_NORMALIZATION_MODE, UCOL_ON, &error); 1718 if (ucol_getAttribute(coll, UCOL_NORMALIZATION_MODE, &error) != UCOL_ON || 1719 U_FAILURE(error)) { 1720 log_err_status(error, "Setting and retrieving of the normalization on/off attribute failed\n"); 1721 } 1722 1723 ucol_setAttribute(coll, UCOL_NORMALIZATION_MODE, UCOL_OFF, &error); 1724 if (ucol_getAttribute(coll, UCOL_NORMALIZATION_MODE, &error) != UCOL_OFF || 1725 U_FAILURE(error)) { 1726 log_err_status(error, "Setting and retrieving of the normalization on/off attribute failed\n"); 1727 } 1728 1729 ucol_setAttribute(coll, UCOL_STRENGTH, UCOL_PRIMARY, &error); 1730 if (ucol_getAttribute(coll, UCOL_STRENGTH, &error) != UCOL_PRIMARY || 1731 U_FAILURE(error)) { 1732 log_err_status(error, "Setting and retrieving of the collation strength failed\n"); 1733 } 1734 1735 ucol_setAttribute(coll, UCOL_STRENGTH, UCOL_SECONDARY, &error); 1736 if (ucol_getAttribute(coll, UCOL_STRENGTH, &error) != UCOL_SECONDARY || 1737 U_FAILURE(error)) { 1738 log_err_status(error, "Setting and retrieving of the collation strength failed\n"); 1739 } 1740 1741 ucol_setAttribute(coll, UCOL_STRENGTH, UCOL_TERTIARY, &error); 1742 if (ucol_getAttribute(coll, UCOL_STRENGTH, &error) != UCOL_TERTIARY || 1743 U_FAILURE(error)) { 1744 log_err_status(error, "Setting and retrieving of the collation strength failed\n"); 1745 } 1746 1747 ucol_setAttribute(coll, UCOL_STRENGTH, UCOL_QUATERNARY, &error); 1748 if (ucol_getAttribute(coll, UCOL_STRENGTH, &error) != UCOL_QUATERNARY || 1749 U_FAILURE(error)) { 1750 log_err_status(error, "Setting and retrieving of the collation strength failed\n"); 1751 } 1752 1753 ucol_setAttribute(coll, UCOL_STRENGTH, UCOL_IDENTICAL, &error); 1754 if (ucol_getAttribute(coll, UCOL_STRENGTH, &error) != UCOL_IDENTICAL || 1755 U_FAILURE(error)) { 1756 log_err_status(error, "Setting and retrieving of the collation strength failed\n"); 1757 } 1758 1759 ucol_close(coll); 1760 } 1761 1762 void TestGetTailoredSet() { 1763 struct { 1764 const char *rules; 1765 const char *tests[20]; 1766 int32_t testsize; 1767 } setTest[] = { 1768 { "&a < \\u212b", { "\\u212b", "A\\u030a", "\\u00c5" }, 3}, 1769 { "& S < \\u0161 <<< \\u0160", { "\\u0161", "s\\u030C", "\\u0160", "S\\u030C" }, 4} 1770 }; 1771 1772 int32_t i = 0, j = 0; 1773 UErrorCode status = U_ZERO_ERROR; 1774 UParseError pError; 1775 1776 UCollator *coll = NULL; 1777 UChar buff[1024]; 1778 int32_t buffLen = 0; 1779 USet *set = NULL; 1780 1781 for(i = 0; i < sizeof(setTest)/sizeof(setTest[0]); i++) { 1782 buffLen = u_unescape(setTest[i].rules, buff, 1024); 1783 coll = ucol_openRules(buff, buffLen, UCOL_DEFAULT, UCOL_DEFAULT, &pError, &status); 1784 if(U_SUCCESS(status)) { 1785 set = ucol_getTailoredSet(coll, &status); 1786 if(uset_size(set) != setTest[i].testsize) { 1787 log_err("Tailored set size different (%d) than expected (%d)\n", uset_size(set), setTest[i].testsize); 1788 } 1789 for(j = 0; j < setTest[i].testsize; j++) { 1790 buffLen = u_unescape(setTest[i].tests[j], buff, 1024); 1791 if(!uset_containsString(set, buff, buffLen)) { 1792 log_err("Tailored set doesn't contain %s... It should\n", setTest[i].tests[j]); 1793 } 1794 } 1795 uset_close(set); 1796 } else { 1797 log_err_status(status, "Couldn't open collator with rules %s\n", setTest[i].rules); 1798 } 1799 ucol_close(coll); 1800 } 1801 } 1802 1803 static int tMemCmp(const uint8_t *first, const uint8_t *second) { 1804 int32_t firstLen = (int32_t)strlen((const char *)first); 1805 int32_t secondLen = (int32_t)strlen((const char *)second); 1806 return memcmp(first, second, uprv_min(firstLen, secondLen)); 1807 } 1808 static const char * strengthsC[] = { 1809 "UCOL_PRIMARY", 1810 "UCOL_SECONDARY", 1811 "UCOL_TERTIARY", 1812 "UCOL_QUATERNARY", 1813 "UCOL_IDENTICAL" 1814 }; 1815 1816 void TestMergeSortKeys(void) { 1817 UErrorCode status = U_ZERO_ERROR; 1818 UCollator *coll = ucol_open("en", &status); 1819 if(U_SUCCESS(status)) { 1820 1821 const char* cases[] = { 1822 "abc", 1823 "abcd", 1824 "abcde" 1825 }; 1826 uint32_t casesSize = sizeof(cases)/sizeof(cases[0]); 1827 const char* prefix = "foo"; 1828 const char* suffix = "egg"; 1829 char outBuff1[256], outBuff2[256]; 1830 1831 uint8_t **sortkeys = (uint8_t **)malloc(casesSize*sizeof(uint8_t *)); 1832 uint8_t **mergedPrefixkeys = (uint8_t **)malloc(casesSize*sizeof(uint8_t *)); 1833 uint8_t **mergedSuffixkeys = (uint8_t **)malloc(casesSize*sizeof(uint8_t *)); 1834 uint32_t *sortKeysLen = (uint32_t *)malloc(casesSize*sizeof(uint32_t)); 1835 uint8_t prefixKey[256], suffixKey[256]; 1836 uint32_t prefixKeyLen = 0, suffixKeyLen = 0, i = 0; 1837 UChar buffer[256]; 1838 uint32_t unescapedLen = 0, l1 = 0, l2 = 0; 1839 UColAttributeValue strength; 1840 1841 log_verbose("ucol_mergeSortkeys test\n"); 1842 log_verbose("Testing order of the test cases\n"); 1843 genericLocaleStarter("en", cases, casesSize); 1844 1845 for(i = 0; i<casesSize; i++) { 1846 sortkeys[i] = (uint8_t *)malloc(256*sizeof(uint8_t)); 1847 mergedPrefixkeys[i] = (uint8_t *)malloc(256*sizeof(uint8_t)); 1848 mergedSuffixkeys[i] = (uint8_t *)malloc(256*sizeof(uint8_t)); 1849 } 1850 1851 unescapedLen = u_unescape(prefix, buffer, 256); 1852 prefixKeyLen = ucol_getSortKey(coll, buffer, unescapedLen, prefixKey, 256); 1853 1854 unescapedLen = u_unescape(suffix, buffer, 256); 1855 suffixKeyLen = ucol_getSortKey(coll, buffer, unescapedLen, suffixKey, 256); 1856 1857 log_verbose("Massaging data with prefixes and different strengths\n"); 1858 strength = UCOL_PRIMARY; 1859 while(strength <= UCOL_IDENTICAL) { 1860 log_verbose("Strength %s\n", strengthsC[strength<=UCOL_QUATERNARY?strength:4]); 1861 ucol_setAttribute(coll, UCOL_STRENGTH, strength, &status); 1862 for(i = 0; i<casesSize; i++) { 1863 unescapedLen = u_unescape(cases[i], buffer, 256); 1864 sortKeysLen[i] = ucol_getSortKey(coll, buffer, unescapedLen, sortkeys[i], 256); 1865 ucol_mergeSortkeys(prefixKey, prefixKeyLen, sortkeys[i], sortKeysLen[i], mergedPrefixkeys[i], 256); 1866 ucol_mergeSortkeys(sortkeys[i], sortKeysLen[i], suffixKey, suffixKeyLen, mergedSuffixkeys[i], 256); 1867 if(i>0) { 1868 if(tMemCmp(mergedPrefixkeys[i-1], mergedPrefixkeys[i]) >= 0) { 1869 log_err("Error while comparing prefixed keys @ strength %s:\n", strengthsC[strength<=UCOL_QUATERNARY?strength:4]); 1870 log_err("%s\n%s\n", 1871 ucol_sortKeyToString(coll, mergedPrefixkeys[i-1], outBuff1, &l1), 1872 ucol_sortKeyToString(coll, mergedPrefixkeys[i], outBuff2, &l2)); 1873 } 1874 if(tMemCmp(mergedSuffixkeys[i-1], mergedSuffixkeys[i]) >= 0) { 1875 log_err("Error while comparing suffixed keys @ strength %s:\n", strengthsC[strength<=UCOL_QUATERNARY?strength:4]); 1876 log_err("%s\n%s\n", 1877 ucol_sortKeyToString(coll, mergedSuffixkeys[i-1], outBuff1, &l1), 1878 ucol_sortKeyToString(coll, mergedSuffixkeys[i], outBuff2, &l2)); 1879 } 1880 } 1881 } 1882 if(strength == UCOL_QUATERNARY) { 1883 strength = UCOL_IDENTICAL; 1884 } else { 1885 strength++; 1886 } 1887 } 1888 1889 { 1890 uint8_t smallBuf[3]; 1891 uint32_t reqLen = 0; 1892 log_verbose("testing buffer overflow\n"); 1893 reqLen = ucol_mergeSortkeys(prefixKey, prefixKeyLen, suffixKey, suffixKeyLen, smallBuf, 3); 1894 if(reqLen != (prefixKeyLen+suffixKeyLen-1)) { 1895 log_err("Wrong preflight size for merged sortkey\n"); 1896 } 1897 } 1898 1899 { 1900 UChar empty = 0; 1901 uint8_t emptyKey[20], abcKey[50], mergedKey[100]; 1902 int32_t emptyKeyLen = 0, abcKeyLen = 0, mergedKeyLen = 0; 1903 1904 log_verbose("testing merging with sortkeys generated for empty strings\n"); 1905 emptyKeyLen = ucol_getSortKey(coll, &empty, 0, emptyKey, 20); 1906 unescapedLen = u_unescape(cases[0], buffer, 256); 1907 abcKeyLen = ucol_getSortKey(coll, buffer, unescapedLen, abcKey, 50); 1908 mergedKeyLen = ucol_mergeSortkeys(emptyKey, emptyKeyLen, abcKey, abcKeyLen, mergedKey, 100); 1909 if(mergedKey[0] != 2) { 1910 log_err("Empty sortkey didn't produce a level separator\n"); 1911 } 1912 /* try with zeros */ 1913 mergedKeyLen = ucol_mergeSortkeys(emptyKey, 0, abcKey, abcKeyLen, mergedKey, 100); 1914 if(mergedKeyLen != 0 || mergedKey[0] != 0) { 1915 log_err("Empty key didn't produce null mergedKey\n"); 1916 } 1917 mergedKeyLen = ucol_mergeSortkeys(abcKey, abcKeyLen, emptyKey, 0, mergedKey, 100); 1918 if(mergedKeyLen != 0 || mergedKey[0] != 0) { 1919 log_err("Empty key didn't produce null mergedKey\n"); 1920 } 1921 1922 } 1923 1924 for(i = 0; i<casesSize; i++) { 1925 free(sortkeys[i]); 1926 free(mergedPrefixkeys[i]); 1927 free(mergedSuffixkeys[i]); 1928 } 1929 free(sortkeys); 1930 free(mergedPrefixkeys); 1931 free(mergedSuffixkeys); 1932 free(sortKeysLen); 1933 ucol_close(coll); 1934 /* need to finish this up */ 1935 } else { 1936 log_data_err("Couldn't open collator"); 1937 } 1938 } 1939 static void TestShortString(void) 1940 { 1941 struct { 1942 const char *input; 1943 const char *expectedOutput; 1944 const char *locale; 1945 UErrorCode expectedStatus; 1946 int32_t expectedOffset; 1947 uint32_t expectedIdentifier; 1948 } testCases[] = { 1949 /* 1950 * The following expectedOutput contains a collation weight (2700 from UCA 6.0) 1951 * which is the primary weight for the T character (U+0041) in the input. 1952 * When that character gets a different weight in FractionalUCA.txt, 1953 * the expectedOutput needs to be adjusted. 1954 * That is, when we upgrade to a new UCA version or change collation 1955 * in such a way that the absolute weight for 'A' changes, 1956 * we will get a test failure here and need to adjust the test case. 1957 */ 1958 {"LDE_RDE_KPHONEBOOK_T0041_ZLATN","B2700_KPHONEBOOK_LDE", "de@collation=phonebook", U_USING_FALLBACK_WARNING, 0, 0 }, 1959 1960 {"LEN_RUS_NO_AS_S4","AS_LROOT_NO_S4", NULL, U_USING_DEFAULT_WARNING, 0, 0 }, 1961 {"LDE_VPHONEBOOK_EO_SI","EO_KPHONEBOOK_LDE_SI", "de@collation=phonebook", U_ZERO_ERROR, 0, 0 }, 1962 {"LDE_Kphonebook","KPHONEBOOK_LDE", "de@collation=phonebook", U_ZERO_ERROR, 0, 0 }, 1963 {"Xqde_DE@collation=phonebookq_S3_EX","KPHONEBOOK_LDE", "de@collation=phonebook", U_USING_FALLBACK_WARNING, 0, 0 }, 1964 {"LFR_FO", "FO_LROOT", NULL, U_USING_DEFAULT_WARNING, 0, 0 }, 1965 {"SO_LX_AS", "", NULL, U_ILLEGAL_ARGUMENT_ERROR, 8, 0 }, 1966 {"S3_ASS_MMM", "", NULL, U_ILLEGAL_ARGUMENT_ERROR, 5, 0 } 1967 }; 1968 1969 int32_t i = 0; 1970 UCollator *coll = NULL, *fromNormalized = NULL; 1971 UParseError parseError; 1972 UErrorCode status = U_ZERO_ERROR; 1973 char fromShortBuffer[256], normalizedBuffer[256], fromNormalizedBuffer[256]; 1974 const char* locale = NULL; 1975 1976 1977 for(i = 0; i < sizeof(testCases)/sizeof(testCases[0]); i++) { 1978 status = U_ZERO_ERROR; 1979 if(testCases[i].locale) { 1980 locale = testCases[i].locale; 1981 } else { 1982 locale = NULL; 1983 } 1984 1985 coll = ucol_openFromShortString(testCases[i].input, FALSE, &parseError, &status); 1986 if(status != testCases[i].expectedStatus) { 1987 log_err_status(status, "Got status '%s' that is different from expected '%s' for '%s'\n", 1988 u_errorName(status), u_errorName(testCases[i].expectedStatus), testCases[i].input); 1989 continue; 1990 } 1991 1992 if(U_SUCCESS(status)) { 1993 ucol_getShortDefinitionString(coll, locale, fromShortBuffer, 256, &status); 1994 1995 if(strcmp(fromShortBuffer, testCases[i].expectedOutput)) { 1996 log_err("Got short string '%s' from the collator. Expected '%s' for input '%s'\n", 1997 fromShortBuffer, testCases[i].expectedOutput, testCases[i].input); 1998 } 1999 2000 ucol_normalizeShortDefinitionString(testCases[i].input, normalizedBuffer, 256, &parseError, &status); 2001 fromNormalized = ucol_openFromShortString(normalizedBuffer, FALSE, &parseError, &status); 2002 ucol_getShortDefinitionString(fromNormalized, locale, fromNormalizedBuffer, 256, &status); 2003 2004 if(strcmp(fromShortBuffer, fromNormalizedBuffer)) { 2005 log_err("Strings obtained from collators instantiated by short string ('%s') and from normalized string ('%s') differ\n", 2006 fromShortBuffer, fromNormalizedBuffer); 2007 } 2008 2009 2010 if(!ucol_equals(coll, fromNormalized)) { 2011 log_err("Collator from short string ('%s') differs from one obtained through a normalized version ('%s')\n", 2012 testCases[i].input, normalizedBuffer); 2013 } 2014 2015 ucol_close(fromNormalized); 2016 ucol_close(coll); 2017 2018 } else { 2019 if(parseError.offset != testCases[i].expectedOffset) { 2020 log_err("Got parse error offset %i, but expected %i instead for '%s'\n", 2021 parseError.offset, testCases[i].expectedOffset, testCases[i].input); 2022 } 2023 } 2024 } 2025 2026 } 2027 2028 static void 2029 doSetsTest(const char *locale, const USet *ref, USet *set, const char* inSet, const char* outSet, UErrorCode *status) { 2030 UChar buffer[512]; 2031 int32_t bufLen; 2032 2033 uset_clear(set); 2034 bufLen = u_unescape(inSet, buffer, 512); 2035 uset_applyPattern(set, buffer, bufLen, 0, status); 2036 if(U_FAILURE(*status)) { 2037 log_err("%s: Failure setting pattern %s\n", locale, u_errorName(*status)); 2038 } 2039 2040 if(!uset_containsAll(ref, set)) { 2041 log_err("%s: Some stuff from %s is not present in the set\n", locale, inSet); 2042 } 2043 2044 uset_clear(set); 2045 bufLen = u_unescape(outSet, buffer, 512); 2046 uset_applyPattern(set, buffer, bufLen, 0, status); 2047 if(U_FAILURE(*status)) { 2048 log_err("%s: Failure setting pattern %s\n", locale, u_errorName(*status)); 2049 } 2050 2051 if(!uset_containsNone(ref, set)) { 2052 log_err("%s: Some stuff from %s is present in the set\n", locale, outSet); 2053 } 2054 } 2055 2056 2057 2058 2059 static void 2060 TestGetContractionsAndUnsafes(void) 2061 { 2062 static struct { 2063 const char* locale; 2064 const char* inConts; 2065 const char* outConts; 2066 const char* inExp; 2067 const char* outExp; 2068 const char* unsafeCodeUnits; 2069 const char* safeCodeUnits; 2070 } tests[] = { 2071 { "ru", 2072 "[{\\u0418\\u0306}{\\u0438\\u0306}]", 2073 "[\\u0439\\u0457]", 2074 "[\\u00e6]", 2075 "[ae]", 2076 "[\\u0418\\u0438]", 2077 "[aAbB\\u0430\\u0410\\u0433\\u0413]" 2078 }, 2079 { "uk", 2080 "[{\\u0406\\u0308}{\\u0456\\u0308}{\\u0418\\u0306}{\\u0438\\u0306}]", 2081 "[\\u0407\\u0419\\u0439\\u0457]", 2082 "[\\u00e6]", 2083 "[ae]", 2084 "[\\u0406\\u0456\\u0418\\u0438]", 2085 "[aAbBxv]", 2086 }, 2087 { "sh", 2088 "[{C\\u0301}{C\\u030C}{C\\u0341}{DZ\\u030C}{Dz\\u030C}{D\\u017D}{D\\u017E}{lj}{nj}]", 2089 "[{\\u309d\\u3099}{\\u30fd\\u3099}]", 2090 "[\\u00e6]", 2091 "[a]", 2092 "[nlcdzNLCDZ]", 2093 "[jabv]" 2094 }, 2095 { "ja", 2096 "[{\\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}]", 2097 "[{\\u30FD\\u3099}{\\u309D\\u3099}{\\u3053\\u3099}{\\u30B3\\u3099}{lj}{nj}]", 2098 "[\\u30FE\\u00e6]", 2099 "[a]", 2100 "[\\u3099]", 2101 "[]" 2102 } 2103 }; 2104 2105 2106 2107 2108 UErrorCode status = U_ZERO_ERROR; 2109 UCollator *coll = NULL; 2110 int32_t i = 0; 2111 int32_t noConts = 0; 2112 USet *conts = uset_open(0,0); 2113 USet *exp = uset_open(0, 0); 2114 USet *set = uset_open(0,0); 2115 int32_t setBufferLen = 65536; 2116 UChar buffer[65536]; 2117 int32_t setLen = 0; 2118 2119 for(i = 0; i < sizeof(tests)/sizeof(tests[0]); i++) { 2120 log_verbose("Testing locale: %s\n", tests[i].locale); 2121 coll = ucol_open(tests[i].locale, &status); 2122 if (coll == NULL || U_FAILURE(status)) { 2123 log_err_status(status, "Unable to open collator for locale %s ==> %s\n", tests[i].locale, u_errorName(status)); 2124 continue; 2125 } 2126 ucol_getContractionsAndExpansions(coll, conts, exp, TRUE, &status); 2127 doSetsTest(tests[i].locale, conts, set, tests[i].inConts, tests[i].outConts, &status); 2128 setLen = uset_toPattern(conts, buffer, setBufferLen, TRUE, &status); 2129 if(U_SUCCESS(status)) { 2130 /*log_verbose("Contractions %i: %s\n", uset_getItemCount(conts), aescstrdup(buffer, setLen));*/ 2131 } else { 2132 log_err("error %s. %i\n", u_errorName(status), setLen); 2133 status = U_ZERO_ERROR; 2134 } 2135 doSetsTest(tests[i].locale, exp, set, tests[i].inExp, tests[i].outExp, &status); 2136 setLen = uset_toPattern(exp, buffer, setBufferLen, TRUE, &status); 2137 if(U_SUCCESS(status)) { 2138 /*log_verbose("Expansions %i: %s\n", uset_getItemCount(exp), aescstrdup(buffer, setLen));*/ 2139 } else { 2140 log_err("error %s. %i\n", u_errorName(status), setLen); 2141 status = U_ZERO_ERROR; 2142 } 2143 2144 noConts = ucol_getUnsafeSet(coll, conts, &status); 2145 doSetsTest(tests[i].locale, conts, set, tests[i].unsafeCodeUnits, tests[i].safeCodeUnits, &status); 2146 setLen = uset_toPattern(conts, buffer, setBufferLen, TRUE, &status); 2147 if(U_SUCCESS(status)) { 2148 log_verbose("Unsafe %i: %s\n", uset_getItemCount(exp), aescstrdup(buffer, setLen)); 2149 } else { 2150 log_err("error %s. %i\n", u_errorName(status), setLen); 2151 status = U_ZERO_ERROR; 2152 } 2153 2154 ucol_close(coll); 2155 } 2156 2157 2158 uset_close(conts); 2159 uset_close(exp); 2160 uset_close(set); 2161 } 2162 2163 static void 2164 TestOpenBinary(void) 2165 { 2166 UErrorCode status = U_ZERO_ERROR; 2167 /* 2168 char rule[] = "&h < d < c < b"; 2169 char *wUCA[] = { "a", "h", "d", "c", "b", "i" }; 2170 char *noUCA[] = {"d", "c", "b", "a", "h", "i" }; 2171 */ 2172 /* we have to use Cyrillic letters because latin-1 always gets copied */ 2173 const char rule[] = "&\\u0452 < \\u0434 < \\u0433 < \\u0432"; /* &dje < d < g < v */ 2174 const char *wUCA[] = { "\\u0430", "\\u0452", "\\u0434", "\\u0433", "\\u0432", "\\u0435" }; /* a, dje, d, g, v, e */ 2175 const char *noUCA[] = {"\\u0434", "\\u0433", "\\u0432", "\\u0430", "\\u0435", "\\u0452" }; /* d, g, v, a, e, dje */ 2176 2177 UChar uRules[256]; 2178 int32_t uRulesLen = u_unescape(rule, uRules, 256); 2179 2180 UCollator *coll = ucol_openRules(uRules, uRulesLen, UCOL_DEFAULT, UCOL_DEFAULT, NULL, &status); 2181 UCollator *UCA = NULL; 2182 UCollator *cloneNOUCA = NULL, *cloneWUCA = NULL; 2183 2184 uint8_t imageBuffer[32768]; 2185 uint8_t *image = imageBuffer; 2186 int32_t imageBufferCapacity = 32768; 2187 2188 int32_t imageSize; 2189 2190 if((coll==NULL)||(U_FAILURE(status))) { 2191 log_data_err("could not load collators or error occured: %s\n", 2192 u_errorName(status)); 2193 return; 2194 } 2195 UCA = ucol_open("root", &status); 2196 if((UCA==NULL)||(U_FAILURE(status))) { 2197 log_data_err("could not load UCA collator or error occured: %s\n", 2198 u_errorName(status)); 2199 return; 2200 } 2201 imageSize = ucol_cloneBinary(coll, image, imageBufferCapacity, &status); 2202 if(U_FAILURE(status)) { 2203 image = (uint8_t *)malloc(imageSize*sizeof(uint8_t)); 2204 status = U_ZERO_ERROR; 2205 imageSize = ucol_cloneBinary(coll, imageBuffer, imageSize, &status); 2206 } 2207 2208 2209 cloneWUCA = ucol_openBinary(image, imageSize, UCA, &status); 2210 cloneNOUCA = ucol_openBinary(image, imageSize, NULL, &status); 2211 2212 genericOrderingTest(coll, wUCA, sizeof(wUCA)/sizeof(wUCA[0])); 2213 2214 genericOrderingTest(cloneWUCA, wUCA, sizeof(wUCA)/sizeof(wUCA[0])); 2215 genericOrderingTest(cloneNOUCA, noUCA, sizeof(noUCA)/sizeof(noUCA[0])); 2216 2217 if(image != imageBuffer) { 2218 free(image); 2219 } 2220 ucol_close(coll); 2221 ucol_close(cloneNOUCA); 2222 ucol_close(cloneWUCA); 2223 ucol_close(UCA); 2224 } 2225 2226 static void TestDefault(void) { 2227 /* Tests for code coverage. */ 2228 UErrorCode status = U_ZERO_ERROR; 2229 UCollator *coll = ucol_open("es@collation=pinyin", &status); 2230 if (coll == NULL || status == U_FILE_ACCESS_ERROR) { 2231 log_data_err("Unable to open collator es@collation=pinyin\n"); 2232 return; 2233 } 2234 if (status != U_USING_DEFAULT_WARNING) { 2235 /* What do you mean that you know about using pinyin collation in Spanish!? This should be in the zh locale. */ 2236 log_err("es@collation=pinyin should return U_USING_DEFAULT_WARNING, but returned %s\n", u_errorName(status)); 2237 } 2238 ucol_close(coll); 2239 if (ucol_getKeywordValues("funky", &status) != NULL) { 2240 log_err("Collators should not know about the funky keyword.\n"); 2241 } 2242 if (status != U_ILLEGAL_ARGUMENT_ERROR) { 2243 log_err("funky keyword didn't fail as expected %s\n", u_errorName(status)); 2244 } 2245 if (ucol_getKeywordValues("collation", &status) != NULL) { 2246 log_err("ucol_getKeywordValues should not work when given a bad status.\n"); 2247 } 2248 } 2249 2250 static void TestDefaultKeyword(void) { 2251 /* Tests for code coverage. */ 2252 UErrorCode status = U_ZERO_ERROR; 2253 const char *loc = "zh_TW@collation=default"; 2254 UCollator *coll = ucol_open(loc, &status); 2255 if(U_FAILURE(status)) { 2256 log_info("Warning: ucol_open(%s, ...) returned %s, at least it didn't crash.\n", loc, u_errorName(status)); 2257 } else if (status != U_USING_FALLBACK_WARNING) { 2258 /* Hmm, skip the following test for CLDR 1.9 data and/or ICU 4.6, no longer seems to apply */ 2259 #if 0 2260 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)); 2261 #endif 2262 } 2263 ucol_close(coll); 2264 } 2265 2266 static void TestGetKeywordValuesForLocale(void) { 2267 #define INCLUDE_UNIHAN_COLLATION 0 2268 #define PREFERRED_SIZE 16 2269 #define MAX_NUMBER_OF_KEYWORDS 8 2270 const char *PREFERRED[PREFERRED_SIZE][MAX_NUMBER_OF_KEYWORDS+1] = { 2271 { "und", "standard", "ducet", "search", NULL, NULL, NULL, NULL, NULL }, 2272 { "en_US", "standard", "ducet", "search", NULL, NULL, NULL, NULL, NULL }, 2273 { "en_029", "standard", "ducet", "search", NULL, NULL, NULL, NULL, NULL }, 2274 { "de_DE", "standard", "phonebook", "search", "ducet", NULL, NULL, NULL, NULL }, 2275 { "de_Latn_DE", "standard", "phonebook", "search", "ducet", NULL, NULL, NULL, NULL }, 2276 #if INCLUDE_UNIHAN_COLLATION 2277 { "zh", "pinyin", "big5han", "gb2312han", "standard", "stroke", "unihan", "ducet", "search" }, 2278 { "zh_Hans", "pinyin", "big5han", "gb2312han", "standard", "stroke", "unihan", "ducet", "search" }, 2279 { "zh_CN", "pinyin", "big5han", "gb2312han", "standard", "stroke", "unihan", "ducet", "search" }, 2280 { "zh_Hant", "stroke", "big5han", "gb2312han", "pinyin", "standard", "unihan", "ducet", "search" }, 2281 { "zh_TW", "stroke", "big5han", "gb2312han", "pinyin", "standard", "unihan", "ducet", "search" }, 2282 { "zh__PINYIN", "pinyin", "big5han", "gb2312han", "standard", "stroke", "unihan", "ducet", "search" }, 2283 #else 2284 { "zh", "pinyin", "big5han", "gb2312han", "standard", "stroke", "ducet", "search", NULL }, 2285 { "zh_Hans", "pinyin", "big5han", "gb2312han", "standard", "stroke", "ducet", "search", NULL }, 2286 { "zh_CN", "pinyin", "big5han", "gb2312han", "standard", "stroke", "ducet", "search", NULL }, 2287 { "zh_Hant", "stroke", "big5han", "gb2312han", "pinyin", "standard", "ducet", "search", NULL }, 2288 { "zh_TW", "stroke", "big5han", "gb2312han", "pinyin", "standard", "ducet", "search", NULL }, 2289 { "zh__PINYIN", "pinyin", "big5han", "gb2312han", "standard", "stroke", "ducet", "search", NULL }, 2290 #endif 2291 { "es_ES", "standard", "search", "traditional", "ducet", NULL, NULL, NULL, NULL }, 2292 { "es__TRADITIONAL","traditional", "search", "standard", "ducet", NULL, NULL, NULL, NULL }, 2293 { "und@collation=phonebook", "standard", "ducet", "search", NULL, NULL, NULL, NULL, NULL }, 2294 { "de_DE@collation=big5han", "standard", "phonebook", "search", "ducet", NULL, NULL, NULL, NULL }, 2295 { "zzz@collation=xxx", "standard", "ducet", "search", NULL, NULL, NULL, NULL, NULL } 2296 }; 2297 #if INCLUDE_UNIHAN_COLLATION 2298 const int32_t expectedLength[PREFERRED_SIZE] = { 3, 3, 3, 4, 4, 8, 8, 8, 8, 8, 8, 4, 4, 3, 4, 3 }; 2299 #else 2300 const int32_t expectedLength[PREFERRED_SIZE] = { 3, 3, 3, 4, 4, 7, 7, 7, 7, 7, 7, 4, 4, 3, 4, 3 }; 2301 #endif 2302 2303 UErrorCode status = U_ZERO_ERROR; 2304 UEnumeration *keywordValues = NULL; 2305 int32_t i, n, size, valueLength; 2306 const char *locale = NULL, *value = NULL; 2307 UBool errorOccurred = FALSE; 2308 2309 for (i = 0; i < PREFERRED_SIZE; i++) { 2310 locale = PREFERRED[i][0]; 2311 value = NULL; 2312 valueLength = 0; 2313 size = 0; 2314 2315 keywordValues = ucol_getKeywordValuesForLocale("collation", locale, TRUE, &status); 2316 if (keywordValues == NULL || U_FAILURE(status)) { 2317 log_err_status(status, "Error getting keyword values: %s\n", u_errorName(status)); 2318 break; 2319 } 2320 size = uenum_count(keywordValues, &status); 2321 2322 if (size == expectedLength[i]) { 2323 for (n = 0; n < expectedLength[i]; n++) { 2324 if ((value = uenum_next(keywordValues, &valueLength, &status)) != NULL && U_SUCCESS(status)) { 2325 if (uprv_strcmp(value, PREFERRED[i][n+1]) != 0) { 2326 log_err("Keyword values differ: Got [%s] Expected [%s] for locale: %s\n", value, PREFERRED[i][n+1], locale); 2327 errorOccurred = TRUE; 2328 break; 2329 } 2330 2331 } else { 2332 log_err("While getting keyword value from locale: %s got this error: %s\n", locale, u_errorName(status)); 2333 errorOccurred = TRUE; 2334 break; 2335 } 2336 } 2337 if (errorOccurred) { 2338 break; 2339 } 2340 } else { 2341 log_err("Number of keywords (%d) does not match expected size (%d) for locale: %s\n", size, expectedLength[i], locale); 2342 break; 2343 } 2344 uenum_close(keywordValues); 2345 keywordValues = NULL; 2346 } 2347 if (keywordValues != NULL) { 2348 uenum_close(keywordValues); 2349 } 2350 } 2351 2352 2353 #endif /* #if !UCONFIG_NO_COLLATION */ 2354