1 /******************************************************************** 2 * Copyright (c) 1997-2014, 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 #include "ucol_imp.h" 35 36 #define LENGTHOF(array) (int32_t)(sizeof(array)/sizeof((array)[0])) 37 38 static void TestAttribute(void); 39 static void TestDefault(void); 40 static void TestDefaultKeyword(void); 41 static void TestBengaliSortKey(void); 42 43 44 static char* U_EXPORT2 ucol_sortKeyToString(const UCollator *coll, const uint8_t *sortkey, char *buffer, uint32_t len) { 45 uint32_t position = 0; 46 uint8_t b; 47 48 if (position + 1 < len) 49 position += sprintf(buffer + position, "["); 50 while ((b = *sortkey++) != 0) { 51 if (b == 1 && position + 5 < len) { 52 position += sprintf(buffer + position, "%02X . ", b); 53 } else if (b != 1 && position + 3 < len) { 54 position += sprintf(buffer + position, "%02X ", b); 55 } 56 } 57 if (position + 3 < len) 58 position += sprintf(buffer + position, "%02X]", b); 59 return buffer; 60 } 61 62 void addCollAPITest(TestNode** root) 63 { 64 /* WEIVTODO: return tests here */ 65 addTest(root, &TestProperty, "tscoll/capitst/TestProperty"); 66 addTest(root, &TestRuleBasedColl, "tscoll/capitst/TestRuleBasedColl"); 67 addTest(root, &TestCompare, "tscoll/capitst/TestCompare"); 68 addTest(root, &TestSortKey, "tscoll/capitst/TestSortKey"); 69 addTest(root, &TestHashCode, "tscoll/capitst/TestHashCode"); 70 addTest(root, &TestElemIter, "tscoll/capitst/TestElemIter"); 71 addTest(root, &TestGetAll, "tscoll/capitst/TestGetAll"); 72 /*addTest(root, &TestGetDefaultRules, "tscoll/capitst/TestGetDefaultRules");*/ 73 addTest(root, &TestDecomposition, "tscoll/capitst/TestDecomposition"); 74 addTest(root, &TestSafeClone, "tscoll/capitst/TestSafeClone"); 75 addTest(root, &TestCloneBinary, "tscoll/capitst/TestCloneBinary"); 76 addTest(root, &TestGetSetAttr, "tscoll/capitst/TestGetSetAttr"); 77 addTest(root, &TestBounds, "tscoll/capitst/TestBounds"); 78 addTest(root, &TestGetLocale, "tscoll/capitst/TestGetLocale"); 79 addTest(root, &TestSortKeyBufferOverrun, "tscoll/capitst/TestSortKeyBufferOverrun"); 80 addTest(root, &TestAttribute, "tscoll/capitst/TestAttribute"); 81 addTest(root, &TestGetTailoredSet, "tscoll/capitst/TestGetTailoredSet"); 82 addTest(root, &TestMergeSortKeys, "tscoll/capitst/TestMergeSortKeys"); 83 addTest(root, &TestShortString, "tscoll/capitst/TestShortString"); 84 // android-changed (no rule strings) -- addTest(root, &TestGetContractionsAndUnsafes, "tscoll/capitst/TestGetContractionsAndUnsafes"); 85 addTest(root, &TestOpenBinary, "tscoll/capitst/TestOpenBinary"); 86 addTest(root, &TestDefault, "tscoll/capitst/TestDefault"); 87 addTest(root, &TestDefaultKeyword, "tscoll/capitst/TestDefaultKeyword"); 88 // android-changed(no rule strings) -- addTest(root, &TestOpenVsOpenRules, "tscoll/capitst/TestOpenVsOpenRules"); 89 addTest(root, &TestBengaliSortKey, "tscoll/capitst/TestBengaliSortKey"); 90 addTest(root, &TestGetKeywordValuesForLocale, "tscoll/capitst/TestGetKeywordValuesForLocale"); 91 addTest(root, &TestStrcollNull, "tscoll/capitst/TestStrcollNull"); 92 } 93 94 void TestGetSetAttr(void) { 95 UErrorCode status = U_ZERO_ERROR; 96 UCollator *coll = ucol_open(NULL, &status); 97 struct attrTest { 98 UColAttribute att; 99 UColAttributeValue val[5]; 100 uint32_t valueSize; 101 UColAttributeValue nonValue; 102 } attrs[] = { 103 {UCOL_FRENCH_COLLATION, {UCOL_ON, UCOL_OFF}, 2, UCOL_SHIFTED}, 104 {UCOL_ALTERNATE_HANDLING, {UCOL_NON_IGNORABLE, UCOL_SHIFTED}, 2, UCOL_OFF},/* attribute for handling variable elements*/ 105 {UCOL_CASE_FIRST, {UCOL_OFF, UCOL_LOWER_FIRST, UCOL_UPPER_FIRST}, 3, UCOL_SHIFTED},/* who goes first, lower case or uppercase */ 106 {UCOL_CASE_LEVEL, {UCOL_ON, UCOL_OFF}, 2, UCOL_SHIFTED},/* do we have an extra case level */ 107 {UCOL_NORMALIZATION_MODE, {UCOL_ON, UCOL_OFF}, 2, UCOL_SHIFTED},/* attribute for normalization */ 108 {UCOL_DECOMPOSITION_MODE, {UCOL_ON, UCOL_OFF}, 2, UCOL_SHIFTED}, 109 {UCOL_STRENGTH, {UCOL_PRIMARY, UCOL_SECONDARY, UCOL_TERTIARY, UCOL_QUATERNARY, UCOL_IDENTICAL}, 5, UCOL_SHIFTED},/* attribute for strength */ 110 {UCOL_HIRAGANA_QUATERNARY_MODE, {UCOL_ON, UCOL_OFF}, 2, UCOL_SHIFTED},/* when turned on, this attribute */ 111 }; 112 UColAttribute currAttr; 113 UColAttributeValue value; 114 uint32_t i = 0, j = 0; 115 116 if (coll == NULL) { 117 log_err_status(status, "Unable to open collator. %s\n", u_errorName(status)); 118 return; 119 } 120 for(i = 0; i<sizeof(attrs)/sizeof(attrs[0]); i++) { 121 currAttr = attrs[i].att; 122 ucol_setAttribute(coll, currAttr, UCOL_DEFAULT, &status); 123 if(U_FAILURE(status)) { 124 log_err_status(status, "ucol_setAttribute with the default value returned error: %s\n", u_errorName(status)); 125 break; 126 } 127 value = ucol_getAttribute(coll, currAttr, &status); 128 if(U_FAILURE(status)) { 129 log_err("ucol_getAttribute returned error: %s\n", u_errorName(status)); 130 break; 131 } 132 for(j = 0; j<attrs[i].valueSize; j++) { 133 ucol_setAttribute(coll, currAttr, attrs[i].val[j], &status); 134 if(U_FAILURE(status)) { 135 log_err("ucol_setAttribute with the value %i returned error: %s\n", attrs[i].val[j], u_errorName(status)); 136 break; 137 } 138 } 139 status = U_ZERO_ERROR; 140 ucol_setAttribute(coll, currAttr, attrs[i].nonValue, &status); 141 if(U_SUCCESS(status)) { 142 log_err("ucol_setAttribute with the bad value didn't return an error\n"); 143 break; 144 } 145 status = U_ZERO_ERROR; 146 147 ucol_setAttribute(coll, currAttr, value, &status); 148 if(U_FAILURE(status)) { 149 log_err("ucol_setAttribute with the default valuereturned error: %s\n", u_errorName(status)); 150 break; 151 } 152 } 153 status = U_ZERO_ERROR; 154 value = ucol_getAttribute(coll, UCOL_ATTRIBUTE_COUNT, &status); 155 if(U_SUCCESS(status)) { 156 log_err("ucol_getAttribute for UCOL_ATTRIBUTE_COUNT didn't return an error\n"); 157 } 158 status = U_ZERO_ERROR; 159 ucol_setAttribute(coll, UCOL_ATTRIBUTE_COUNT, UCOL_DEFAULT, &status); 160 if(U_SUCCESS(status)) { 161 log_err("ucol_setAttribute for UCOL_ATTRIBUTE_COUNT didn't return an error\n"); 162 } 163 status = U_ZERO_ERROR; 164 ucol_close(coll); 165 } 166 167 168 static void doAssert(int condition, const char *message) 169 { 170 if (condition==0) { 171 log_err("ERROR : %s\n", message); 172 } 173 } 174 175 #define UTF8_BUF_SIZE 128 176 177 static void doStrcoll(const UCollator* coll, const UChar* src, int32_t srcLen, const UChar* tgt, int32_t tgtLen, 178 UCollationResult expected, const char *message) { 179 UErrorCode err = U_ZERO_ERROR; 180 char srcU8[UTF8_BUF_SIZE], tgtU8[UTF8_BUF_SIZE]; 181 int32_t srcU8Len = -1, tgtU8Len = -1; 182 int32_t len = 0; 183 184 if (ucol_strcoll(coll, src, srcLen, tgt, tgtLen) != expected) { 185 log_err("ERROR : %s\n", message); 186 } 187 188 u_strToUTF8(srcU8, UTF8_BUF_SIZE, &len, src, srcLen, &err); 189 if (U_FAILURE(err) || len >= UTF8_BUF_SIZE) { 190 log_err("ERROR : UTF-8 conversion error\n"); 191 return; 192 } 193 if (srcLen >= 0) { 194 srcU8Len = len; 195 } 196 u_strToUTF8(tgtU8, UTF8_BUF_SIZE, &len, tgt, tgtLen, &err); 197 if (U_FAILURE(err) || len >= UTF8_BUF_SIZE) { 198 log_err("ERROR : UTF-8 conversion error\n"); 199 return; 200 } 201 if (tgtLen >= 0) { 202 tgtU8Len = len; 203 } 204 205 if (ucol_strcollUTF8(coll, srcU8, srcU8Len, tgtU8, tgtU8Len, &err) != expected 206 || U_FAILURE(err)) { 207 log_err("ERROR: %s (strcollUTF8)\n", message); 208 } 209 } 210 211 #if 0 212 /* We don't have default rules, at least not in the previous sense */ 213 void TestGetDefaultRules(){ 214 uint32_t size=0; 215 UErrorCode status=U_ZERO_ERROR; 216 UCollator *coll=NULL; 217 int32_t len1 = 0, len2=0; 218 uint8_t *binColData = NULL; 219 220 UResourceBundle *res = NULL; 221 UResourceBundle *binColl = NULL; 222 uint8_t *binResult = NULL; 223 224 225 const UChar * defaultRulesArray=ucol_getDefaultRulesArray(&size); 226 log_verbose("Test the function ucol_getDefaultRulesArray()\n"); 227 228 coll = ucol_openRules(defaultRulesArray, size, UCOL_ON, UCOL_PRIMARY, &status); 229 if(U_SUCCESS(status) && coll !=NULL) { 230 binColData = (uint8_t*)ucol_cloneRuleData(coll, &len1, &status); 231 232 } 233 234 235 status=U_ZERO_ERROR; 236 res=ures_open(NULL, "root", &status); 237 if(U_FAILURE(status)){ 238 log_err("ERROR: Failed to get resource for \"root Locale\" with %s", myErrorName(status)); 239 return; 240 } 241 binColl=ures_getByKey(res, "%%Collation", binColl, &status); 242 if(U_SUCCESS(status)){ 243 binResult=(uint8_t*)ures_getBinary(binColl, &len2, &status); 244 if(U_FAILURE(status)){ 245 log_err("ERROR: ures_getBinary() failed\n"); 246 } 247 }else{ 248 log_err("ERROR: ures_getByKey(locale(default), %%Collation) failed"); 249 } 250 251 252 if(len1 != len2){ 253 log_err("Error: ucol_getDefaultRulesArray() failed to return the correct length.\n"); 254 } 255 if(memcmp(binColData, binResult, len1) != 0){ 256 log_err("Error: ucol_getDefaultRulesArray() failed\n"); 257 } 258 259 free(binColData); 260 ures_close(binColl); 261 ures_close(res); 262 ucol_close(coll); 263 264 } 265 #endif 266 267 /* Collator Properties 268 ucol_open, ucol_strcoll, getStrength/setStrength 269 getDecomposition/setDecomposition, getDisplayName*/ 270 void TestProperty() 271 { 272 UCollator *col, *ruled; 273 UChar *disName; 274 int32_t len = 0; 275 UChar source[12], target[12]; 276 int32_t tempLength; 277 UErrorCode status = U_ZERO_ERROR; 278 /* 279 * Expected version of the English collator. 280 * Currently, the major/minor version numbers change when the builder code 281 * changes, 282 * number 2 is from the tailoring data version and 283 * number 3 is the UCA version. 284 * This changes with every UCA version change, and the expected value 285 * needs to be adjusted. 286 * Same in intltest/apicoll.cpp. 287 */ 288 UVersionInfo currVersionArray = {0x31, 0xC0, 0x05, 0x2A}; /* from ICU 4.4/UCA 5.2 */ 289 UVersionInfo versionArray = {0, 0, 0, 0}; 290 UVersionInfo versionUCAArray = {0, 0, 0, 0}; 291 UVersionInfo versionUCDArray = {0, 0, 0, 0}; 292 293 log_verbose("The property tests begin : \n"); 294 log_verbose("Test ucol_strcoll : \n"); 295 col = ucol_open("en_US", &status); 296 if (U_FAILURE(status)) { 297 log_err_status(status, "Default Collator creation failed.: %s\n", myErrorName(status)); 298 return; 299 } 300 301 ucol_getVersion(col, versionArray); 302 /* Check for a version greater than some value rather than equality 303 * so that we need not update the expected version each time. */ 304 if (uprv_memcmp(versionArray, currVersionArray, 4)<0) { 305 log_err("Testing ucol_getVersion() - unexpected result: %02x.%02x.%02x.%02x\n", 306 versionArray[0], versionArray[1], versionArray[2], versionArray[3]); 307 } else { 308 log_verbose("ucol_getVersion() result: %02x.%02x.%02x.%02x\n", 309 versionArray[0], versionArray[1], versionArray[2], versionArray[3]); 310 } 311 312 /* Assume that the UCD and UCA versions are the same, 313 * rather than hardcoding (and updating each time) a particular UCA version. */ 314 u_getUnicodeVersion(versionUCDArray); 315 ucol_getUCAVersion(col, versionUCAArray); 316 if (0!=uprv_memcmp(versionUCAArray, versionUCDArray, 4)) { 317 log_err("Testing ucol_getUCAVersion() - unexpected result: %hu.%hu.%hu.%hu\n", 318 versionUCAArray[0], versionUCAArray[1], versionUCAArray[2], versionUCAArray[3]); 319 } 320 321 u_uastrcpy(source, "ab"); 322 u_uastrcpy(target, "abc"); 323 324 doStrcoll(col, source, u_strlen(source), target, u_strlen(target), UCOL_LESS, "ab < abc comparison failed"); 325 326 u_uastrcpy(source, "ab"); 327 u_uastrcpy(target, "AB"); 328 329 doStrcoll(col, source, u_strlen(source), target, u_strlen(target), UCOL_LESS, "ab < AB comparison failed"); 330 331 u_uastrcpy(source, "blackbird"); 332 u_uastrcpy(target, "black-bird"); 333 334 doStrcoll(col, source, u_strlen(source), target, u_strlen(target), UCOL_GREATER, "black-bird > blackbird comparison failed"); 335 336 u_uastrcpy(source, "black bird"); 337 u_uastrcpy(target, "black-bird"); 338 339 doStrcoll(col, source, u_strlen(source), target, u_strlen(target), UCOL_LESS, "black bird < black-bird comparison failed"); 340 341 u_uastrcpy(source, "Hello"); 342 u_uastrcpy(target, "hello"); 343 344 doStrcoll(col, source, u_strlen(source), target, u_strlen(target), UCOL_GREATER, "Hello > hello comparison failed"); 345 346 log_verbose("Test ucol_strcoll ends.\n"); 347 348 log_verbose("testing ucol_getStrength() method ...\n"); 349 doAssert( (ucol_getStrength(col) == UCOL_TERTIARY), "collation object has the wrong strength"); 350 doAssert( (ucol_getStrength(col) != UCOL_PRIMARY), "collation object's strength is primary difference"); 351 352 log_verbose("testing ucol_setStrength() method ...\n"); 353 ucol_setStrength(col, UCOL_SECONDARY); 354 doAssert( (ucol_getStrength(col) != UCOL_TERTIARY), "collation object's strength is secondary difference"); 355 doAssert( (ucol_getStrength(col) != UCOL_PRIMARY), "collation object's strength is primary difference"); 356 doAssert( (ucol_getStrength(col) == UCOL_SECONDARY), "collation object has the wrong strength"); 357 358 359 log_verbose("Get display name for the default collation in German : \n"); 360 361 len=ucol_getDisplayName("en_US", "de_DE", NULL, 0, &status); 362 if(status==U_BUFFER_OVERFLOW_ERROR){ 363 status=U_ZERO_ERROR; 364 disName=(UChar*)malloc(sizeof(UChar) * (len+1)); 365 ucol_getDisplayName("en_US", "de_DE", disName, len+1, &status); 366 log_verbose("the display name for default collation in german: %s\n", austrdup(disName) ); 367 free(disName); 368 } 369 if(U_FAILURE(status)){ 370 log_err("ERROR: in getDisplayName: %s\n", myErrorName(status)); 371 return; 372 } 373 log_verbose("Default collation getDisplayName ended.\n"); 374 375 ruled = ucol_open("da_DK", &status); 376 log_verbose("ucol_getRules() testing ...\n"); 377 ucol_getRules(ruled, &tempLength); 378 // android-changed (no rule strings) -- doAssert( tempLength != 0, "getRules() result incorrect" ); 379 log_verbose("getRules tests end.\n"); 380 { 381 UChar *buffer = (UChar *)malloc(200000*sizeof(UChar)); 382 int32_t bufLen = 200000; 383 buffer[0] = '\0'; 384 log_verbose("ucol_getRulesEx() testing ...\n"); 385 tempLength = ucol_getRulesEx(col,UCOL_TAILORING_ONLY,buffer,bufLen ); 386 // android-changed (no rule strings) -- doAssert( tempLength == 0x00, "getRulesEx() result incorrect" ); 387 log_verbose("getRules tests end.\n"); 388 389 log_verbose("ucol_getRulesEx() testing ...\n"); 390 tempLength=ucol_getRulesEx(col,UCOL_FULL_RULES,buffer,bufLen ); 391 // android-changed (no rule strings) -- doAssert( tempLength != 0, "getRulesEx() result incorrect" ); 392 log_verbose("getRules tests end.\n"); 393 free(buffer); 394 } 395 ucol_close(ruled); 396 ucol_close(col); 397 398 log_verbose("open an collator for french locale"); 399 col = ucol_open("fr_FR", &status); 400 if (U_FAILURE(status)) { 401 log_err("ERROR: Creating French collation failed.: %s\n", myErrorName(status)); 402 return; 403 } 404 ucol_setStrength(col, UCOL_PRIMARY); 405 log_verbose("testing ucol_getStrength() method again ...\n"); 406 doAssert( (ucol_getStrength(col) != UCOL_TERTIARY), "collation object has the wrong strength"); 407 doAssert( (ucol_getStrength(col) == UCOL_PRIMARY), "collation object's strength is not primary difference"); 408 409 log_verbose("testing French ucol_setStrength() method ...\n"); 410 ucol_setStrength(col, UCOL_TERTIARY); 411 doAssert( (ucol_getStrength(col) == UCOL_TERTIARY), "collation object's strength is not tertiary difference"); 412 doAssert( (ucol_getStrength(col) != UCOL_PRIMARY), "collation object's strength is primary difference"); 413 doAssert( (ucol_getStrength(col) != UCOL_SECONDARY), "collation object's strength is secondary difference"); 414 ucol_close(col); 415 416 log_verbose("Get display name for the french collation in english : \n"); 417 len=ucol_getDisplayName("fr_FR", "en_US", NULL, 0, &status); 418 if(status==U_BUFFER_OVERFLOW_ERROR){ 419 status=U_ZERO_ERROR; 420 disName=(UChar*)malloc(sizeof(UChar) * (len+1)); 421 ucol_getDisplayName("fr_FR", "en_US", disName, len+1, &status); 422 log_verbose("the display name for french collation in english: %s\n", austrdup(disName) ); 423 free(disName); 424 } 425 if(U_FAILURE(status)){ 426 log_err("ERROR: in getDisplayName: %s\n", myErrorName(status)); 427 return; 428 } 429 log_verbose("Default collation getDisplayName ended.\n"); 430 431 } 432 433 /* Test RuleBasedCollator and getRules*/ 434 void TestRuleBasedColl() 435 { 436 UCollator *col1, *col2, *col3, *col4; 437 UCollationElements *iter1, *iter2; 438 UChar ruleset1[60]; 439 UChar ruleset2[50]; 440 UChar teststr[10]; 441 const UChar *rule1, *rule2, *rule3, *rule4; 442 int32_t tempLength; 443 UErrorCode status = U_ZERO_ERROR; 444 u_uastrcpy(ruleset1, "&9 < a, A < b, B < c, C; ch, cH, Ch, CH < d, D, e, E"); 445 u_uastrcpy(ruleset2, "&9 < a, A < b, B < c, C < d, D, e, E"); 446 447 448 col1 = ucol_openRules(ruleset1, u_strlen(ruleset1), UCOL_DEFAULT, UCOL_DEFAULT_STRENGTH, NULL,&status); 449 if (U_FAILURE(status)) { 450 log_err_status(status, "RuleBased Collator creation failed.: %s\n", myErrorName(status)); 451 return; 452 } 453 else 454 log_verbose("PASS: RuleBased Collator creation passed\n"); 455 456 status = U_ZERO_ERROR; 457 col2 = ucol_openRules(ruleset2, u_strlen(ruleset2), UCOL_DEFAULT, UCOL_DEFAULT_STRENGTH, NULL, &status); 458 if (U_FAILURE(status)) { 459 log_err("RuleBased Collator creation failed.: %s\n", myErrorName(status)); 460 return; 461 } 462 else 463 log_verbose("PASS: RuleBased Collator creation passed\n"); 464 465 466 status = U_ZERO_ERROR; 467 col3= ucol_open(NULL, &status); 468 if (U_FAILURE(status)) { 469 log_err("Default Collator creation failed.: %s\n", myErrorName(status)); 470 return; 471 } 472 else 473 log_verbose("PASS: Default Collator creation passed\n"); 474 475 rule1 = ucol_getRules(col1, &tempLength); 476 rule2 = ucol_getRules(col2, &tempLength); 477 rule3 = ucol_getRules(col3, &tempLength); 478 479 doAssert((u_strcmp(rule1, rule2) != 0), "Default collator getRules failed"); 480 doAssert((u_strcmp(rule2, rule3) != 0), "Default collator getRules failed"); 481 doAssert((u_strcmp(rule1, rule3) != 0), "Default collator getRules failed"); 482 483 col4=ucol_openRules(rule2, u_strlen(rule2), 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 rule4= ucol_getRules(col4, &tempLength); 489 doAssert((u_strcmp(rule2, rule4) == 0), "Default collator getRules failed"); 490 491 ucol_close(col1); 492 ucol_close(col2); 493 ucol_close(col3); 494 ucol_close(col4); 495 496 /* tests that modifier ! is always ignored */ 497 u_uastrcpy(ruleset1, "!&a<b"); 498 teststr[0] = 0x0e40; 499 teststr[1] = 0x0e01; 500 teststr[2] = 0x0e2d; 501 col1 = ucol_openRules(ruleset1, u_strlen(ruleset1), UCOL_DEFAULT, UCOL_DEFAULT_STRENGTH, NULL, &status); 502 if (U_FAILURE(status)) { 503 log_err("RuleBased Collator creation failed.: %s\n", myErrorName(status)); 504 return; 505 } 506 col2 = ucol_open("en_US", &status); 507 if (U_FAILURE(status)) { 508 log_err("en_US Collator creation failed.: %s\n", myErrorName(status)); 509 return; 510 } 511 iter1 = ucol_openElements(col1, teststr, 3, &status); 512 iter2 = ucol_openElements(col2, teststr, 3, &status); 513 if(U_FAILURE(status)) { 514 log_err("ERROR: CollationElement iterator creation failed.: %s\n", myErrorName(status)); 515 return; 516 } 517 while (TRUE) { 518 /* testing with en since thai has its own tailoring */ 519 uint32_t ce = ucol_next(iter1, &status); 520 uint32_t ce2 = ucol_next(iter2, &status); 521 if(U_FAILURE(status)) { 522 log_err("ERROR: CollationElement iterator creation failed.: %s\n", myErrorName(status)); 523 return; 524 } 525 if (ce2 != ce) { 526 log_err("! modifier test failed"); 527 } 528 if (ce == UCOL_NULLORDER) { 529 break; 530 } 531 } 532 ucol_closeElements(iter1); 533 ucol_closeElements(iter2); 534 ucol_close(col1); 535 ucol_close(col2); 536 /* CLDR 24+ requires a reset before the first relation */ 537 u_uastrcpy(ruleset1, "< z < a"); 538 col1 = ucol_openRules(ruleset1, u_strlen(ruleset1), UCOL_DEFAULT, UCOL_DEFAULT_STRENGTH, NULL, &status); 539 if (status != U_PARSE_ERROR && status != U_INVALID_FORMAT_ERROR) { 540 log_err("ucol_openRules(without initial reset: '< z < a') " 541 "should fail with U_PARSE_ERROR or U_INVALID_FORMAT_ERROR but yielded %s\n", 542 myErrorName(status)); 543 } 544 ucol_close(col1); 545 } 546 547 void TestCompare() 548 { 549 UErrorCode status = U_ZERO_ERROR; 550 UCollator *col; 551 UChar* test1; 552 UChar* test2; 553 554 log_verbose("The compare tests begin : \n"); 555 status=U_ZERO_ERROR; 556 col = ucol_open("en_US", &status); 557 if(U_FAILURE(status)) { 558 log_err_status(status, "ucal_open() collation creation failed.: %s\n", myErrorName(status)); 559 return; 560 } 561 test1=(UChar*)malloc(sizeof(UChar) * 6); 562 test2=(UChar*)malloc(sizeof(UChar) * 6); 563 u_uastrcpy(test1, "Abcda"); 564 u_uastrcpy(test2, "abcda"); 565 566 log_verbose("Use tertiary comparison level testing ....\n"); 567 568 doAssert( (!ucol_equal(col, test1, u_strlen(test1), test2, u_strlen(test2))), "Result should be \"Abcda\" != \"abcda\" "); 569 doAssert( (ucol_greater(col, test1, u_strlen(test1), test2, u_strlen(test2))), "Result should be \"Abcda\" >>> \"abcda\" "); 570 doAssert( (ucol_greaterOrEqual(col, test1, u_strlen(test1), test2, u_strlen(test2))), "Result should be \"Abcda\" >>> \"abcda\""); 571 572 ucol_setStrength(col, UCOL_SECONDARY); 573 log_verbose("Use secondary comparison level testing ....\n"); 574 575 doAssert( (ucol_equal(col, test1, u_strlen(test1), test2, u_strlen(test2) )), "Result should be \"Abcda\" == \"abcda\""); 576 doAssert( (!ucol_greater(col, test1, u_strlen(test1), test2, u_strlen(test2))), "Result should be \"Abcda\" == \"abcda\""); 577 doAssert( (ucol_greaterOrEqual(col, test1, u_strlen(test1), test2, u_strlen(test2) )), "Result should be \"Abcda\" == \"abcda\""); 578 579 ucol_setStrength(col, UCOL_PRIMARY); 580 log_verbose("Use primary comparison level testing ....\n"); 581 582 doAssert( (ucol_equal(col, test1, u_strlen(test1), test2, u_strlen(test2))), "Result should be \"Abcda\" == \"abcda\""); 583 doAssert( (!ucol_greater(col, test1, u_strlen(test1), test2, u_strlen(test2))), "Result should be \"Abcda\" == \"abcda\""); 584 doAssert( (ucol_greaterOrEqual(col, test1, u_strlen(test1), test2, u_strlen(test2))), "Result should be \"Abcda\" == \"abcda\""); 585 586 587 log_verbose("The compare tests end.\n"); 588 ucol_close(col); 589 free(test1); 590 free(test2); 591 592 } 593 /* 594 --------------------------------------------- 595 tests decomposition setting 596 */ 597 void TestDecomposition() { 598 UErrorCode status = U_ZERO_ERROR; 599 UCollator *en_US, *el_GR, *vi_VN; 600 en_US = ucol_open("en_US", &status); 601 el_GR = ucol_open("el_GR", &status); 602 vi_VN = ucol_open("vi_VN", &status); 603 604 if (U_FAILURE(status)) { 605 log_err_status(status, "ERROR: collation creation failed.: %s\n", myErrorName(status)); 606 return; 607 } 608 609 if (ucol_getAttribute(vi_VN, UCOL_NORMALIZATION_MODE, &status) != UCOL_ON || 610 U_FAILURE(status)) 611 { 612 log_err("ERROR: vi_VN collation did not have canonical decomposition for normalization!\n"); 613 } 614 615 status = U_ZERO_ERROR; 616 if (ucol_getAttribute(el_GR, UCOL_NORMALIZATION_MODE, &status) != UCOL_ON || 617 U_FAILURE(status)) 618 { 619 log_err("ERROR: el_GR collation did not have canonical decomposition for normalization!\n"); 620 } 621 622 status = U_ZERO_ERROR; 623 if (ucol_getAttribute(en_US, UCOL_NORMALIZATION_MODE, &status) != UCOL_OFF || 624 U_FAILURE(status)) 625 { 626 log_err("ERROR: en_US collation had canonical decomposition for normalization!\n"); 627 } 628 629 ucol_close(en_US); 630 ucol_close(el_GR); 631 ucol_close(vi_VN); 632 } 633 634 #define CLONETEST_COLLATOR_COUNT 4 635 636 void TestSafeClone() { 637 UChar test1[6]; 638 UChar test2[6]; 639 static const UChar umlautUStr[] = {0x00DC, 0}; 640 static const UChar oeStr[] = {0x0055, 0x0045, 0}; 641 UCollator * someCollators [CLONETEST_COLLATOR_COUNT]; 642 UCollator * someClonedCollators [CLONETEST_COLLATOR_COUNT]; 643 UCollator * col; 644 UErrorCode err = U_ZERO_ERROR; 645 int8_t idx = 6; /* Leave this here to test buffer alingment in memory*/ 646 uint8_t buffer [CLONETEST_COLLATOR_COUNT] [U_COL_SAFECLONE_BUFFERSIZE]; 647 int32_t bufferSize = U_COL_SAFECLONE_BUFFERSIZE; 648 const char sampleRuleChars[] = "&Z < CH"; 649 UChar sampleRule[sizeof(sampleRuleChars)]; 650 651 u_uastrcpy(test1, "abCda"); 652 u_uastrcpy(test2, "abcda"); 653 u_uastrcpy(sampleRule, sampleRuleChars); 654 655 /* one default collator & two complex ones */ 656 someCollators[0] = ucol_open("en_US", &err); 657 someCollators[1] = ucol_open("ko", &err); 658 someCollators[2] = ucol_open("ja_JP", &err); 659 someCollators[3] = ucol_openRules(sampleRule, -1, UCOL_ON, UCOL_TERTIARY, NULL, &err); 660 if(U_FAILURE(err)) { 661 for (idx = 0; idx < CLONETEST_COLLATOR_COUNT; idx++) { 662 ucol_close(someCollators[idx]); 663 } 664 log_data_err("Couldn't open one or more collators\n"); 665 return; 666 } 667 668 /* Check the various error & informational states: */ 669 670 /* Null status - just returns NULL */ 671 if (NULL != ucol_safeClone(someCollators[0], buffer[0], &bufferSize, NULL)) 672 { 673 log_err("FAIL: Cloned Collator failed to deal correctly with null status\n"); 674 } 675 /* error status - should return 0 & keep error the same */ 676 err = U_MEMORY_ALLOCATION_ERROR; 677 if (NULL != ucol_safeClone(someCollators[0], buffer[0], &bufferSize, &err) || err != U_MEMORY_ALLOCATION_ERROR) 678 { 679 log_err("FAIL: Cloned Collator failed to deal correctly with incoming error status\n"); 680 } 681 err = U_ZERO_ERROR; 682 683 /* Null buffer size pointer is ok */ 684 if (NULL == (col = ucol_safeClone(someCollators[0], buffer[0], NULL, &err)) || U_FAILURE(err)) 685 { 686 log_err("FAIL: Cloned Collator failed to deal correctly with null bufferSize pointer\n"); 687 } 688 ucol_close(col); 689 err = U_ZERO_ERROR; 690 691 /* buffer size pointer is 0 - fill in pbufferSize with a size */ 692 bufferSize = 0; 693 if (NULL != ucol_safeClone(someCollators[0], buffer[0], &bufferSize, &err) || 694 U_FAILURE(err) || bufferSize <= 0) 695 { 696 log_err("FAIL: Cloned Collator failed a sizing request ('preflighting')\n"); 697 } 698 /* Verify our define is large enough */ 699 if (U_COL_SAFECLONE_BUFFERSIZE < bufferSize) 700 { 701 log_err("FAIL: Pre-calculated buffer size is too small\n"); 702 } 703 /* Verify we can use this run-time calculated size */ 704 if (NULL == (col = ucol_safeClone(someCollators[0], buffer[0], &bufferSize, &err)) || U_FAILURE(err)) 705 { 706 log_err("FAIL: Collator can't be cloned with run-time size\n"); 707 } 708 if (col) ucol_close(col); 709 /* size one byte too small - should allocate & let us know */ 710 if (bufferSize > 1) { 711 --bufferSize; 712 } 713 if (NULL == (col = ucol_safeClone(someCollators[0], 0, &bufferSize, &err)) || err != U_SAFECLONE_ALLOCATED_WARNING) 714 { 715 log_err("FAIL: Cloned Collator failed to deal correctly with too-small buffer size\n"); 716 } 717 if (col) ucol_close(col); 718 err = U_ZERO_ERROR; 719 bufferSize = U_COL_SAFECLONE_BUFFERSIZE; 720 721 722 /* Null buffer pointer - return Collator & set error to U_SAFECLONE_ALLOCATED_ERROR */ 723 if (NULL == (col = ucol_safeClone(someCollators[0], 0, &bufferSize, &err)) || err != U_SAFECLONE_ALLOCATED_WARNING) 724 { 725 log_err("FAIL: Cloned Collator failed to deal correctly with null buffer pointer\n"); 726 } 727 if (col) ucol_close(col); 728 err = U_ZERO_ERROR; 729 730 /* Null Collator - return NULL & set U_ILLEGAL_ARGUMENT_ERROR */ 731 if (NULL != ucol_safeClone(NULL, buffer[0], &bufferSize, &err) || err != U_ILLEGAL_ARGUMENT_ERROR) 732 { 733 log_err("FAIL: Cloned Collator failed to deal correctly with null Collator pointer\n"); 734 } 735 736 err = U_ZERO_ERROR; 737 738 /* Test that a cloned collator doesn't accidentally use UCA. */ 739 col=ucol_open("de@collation=phonebook", &err); 740 bufferSize = U_COL_SAFECLONE_BUFFERSIZE; 741 someClonedCollators[0] = ucol_safeClone(col, buffer[0], &bufferSize, &err); 742 doAssert( (ucol_greater(col, umlautUStr, u_strlen(umlautUStr), oeStr, u_strlen(oeStr))), "Original German phonebook collation sorts differently than expected"); 743 doAssert( (ucol_greater(someClonedCollators[0], umlautUStr, u_strlen(umlautUStr), oeStr, u_strlen(oeStr))), "Cloned German phonebook collation sorts differently than expected"); 744 if (!ucol_equals(someClonedCollators[0], col)) { 745 log_err("FAIL: Cloned German phonebook collator is not equal to original.\n"); 746 } 747 ucol_close(col); 748 ucol_close(someClonedCollators[0]); 749 750 err = U_ZERO_ERROR; 751 752 /* change orig & clone & make sure they are independent */ 753 754 for (idx = 0; idx < CLONETEST_COLLATOR_COUNT; idx++) 755 { 756 ucol_setStrength(someCollators[idx], UCOL_IDENTICAL); 757 bufferSize = 1; 758 err = U_ZERO_ERROR; 759 ucol_close(ucol_safeClone(someCollators[idx], buffer[idx], &bufferSize, &err)); 760 if (err != U_SAFECLONE_ALLOCATED_WARNING) { 761 log_err("FAIL: collator number %d was not allocated.\n", idx); 762 log_err("FAIL: status of Collator[%d] is %d (hex: %x).\n", idx, err, err); 763 } 764 765 bufferSize = U_COL_SAFECLONE_BUFFERSIZE; 766 err = U_ZERO_ERROR; 767 someClonedCollators[idx] = ucol_safeClone(someCollators[idx], buffer[idx], &bufferSize, &err); 768 if (U_FAILURE(err)) { 769 log_err("FAIL: Unable to clone collator %d - %s\n", idx, u_errorName(err)); 770 continue; 771 } 772 if (!ucol_equals(someClonedCollators[idx], someCollators[idx])) { 773 log_err("FAIL: Cloned collator is not equal to original at index = %d.\n", idx); 774 } 775 776 /* Check the usability */ 777 ucol_setStrength(someCollators[idx], UCOL_PRIMARY); 778 ucol_setAttribute(someCollators[idx], UCOL_CASE_LEVEL, UCOL_OFF, &err); 779 780 doAssert( (ucol_equal(someCollators[idx], test1, u_strlen(test1), test2, u_strlen(test2))), "Result should be \"abcda\" == \"abCda\""); 781 782 /* Close the original to make sure that the clone is usable. */ 783 ucol_close(someCollators[idx]); 784 785 ucol_setStrength(someClonedCollators[idx], UCOL_TERTIARY); 786 ucol_setAttribute(someClonedCollators[idx], UCOL_CASE_LEVEL, UCOL_OFF, &err); 787 doAssert( (ucol_greater(someClonedCollators[idx], test1, u_strlen(test1), test2, u_strlen(test2))), "Result should be \"abCda\" >>> \"abcda\" "); 788 789 ucol_close(someClonedCollators[idx]); 790 } 791 } 792 793 void TestCloneBinary(){ 794 UErrorCode err = U_ZERO_ERROR; 795 UCollator * col = ucol_open("en_US", &err); 796 UCollator * c; 797 int32_t size; 798 uint8_t * buffer; 799 800 if (U_FAILURE(err)) { 801 log_data_err("Couldn't open collator. Error: %s\n", u_errorName(err)); 802 return; 803 } 804 805 size = ucol_cloneBinary(col, NULL, 0, &err); 806 if(size==0 || err!=U_BUFFER_OVERFLOW_ERROR) { 807 log_err("ucol_cloneBinary - couldn't check size. Error: %s\n", u_errorName(err)); 808 return; 809 } 810 err = U_ZERO_ERROR; 811 812 buffer = (uint8_t *) malloc(size); 813 ucol_cloneBinary(col, buffer, size, &err); 814 if(U_FAILURE(err)) { 815 log_err("ucol_cloneBinary - couldn't clone.. Error: %s\n", u_errorName(err)); 816 free(buffer); 817 return; 818 } 819 820 /* how to check binary result ? */ 821 822 c = ucol_openBinary(buffer, size, col, &err); 823 if(U_FAILURE(err)) { 824 log_err("ucol_openBinary failed. Error: %s\n", u_errorName(err)); 825 } else { 826 UChar t[] = {0x41, 0x42, 0x43, 0}; /* ABC */ 827 uint8_t *k1, *k2; 828 int l1, l2; 829 l1 = ucol_getSortKey(col, t, -1, NULL,0); 830 l2 = ucol_getSortKey(c, t, -1, NULL,0); 831 k1 = (uint8_t *) malloc(sizeof(uint8_t) * l1); 832 k2 = (uint8_t *) malloc(sizeof(uint8_t) * l2); 833 ucol_getSortKey(col, t, -1, k1, l1); 834 ucol_getSortKey(col, t, -1, k2, l2); 835 if (strcmp((char *)k1,(char *)k2) != 0){ 836 log_err("ucol_openBinary - new collator should equal to old one\n"); 837 }; 838 free(k1); 839 free(k2); 840 } 841 free(buffer); 842 ucol_close(c); 843 ucol_close(col); 844 } 845 846 847 static void TestBengaliSortKey(void) 848 { 849 const char *curLoc = "bn"; 850 UChar str1[] = { 0x09BE, 0 }; 851 UChar str2[] = { 0x0B70, 0 }; 852 UCollator *c2 = NULL; 853 const UChar *rules; 854 int32_t rulesLength=-1; 855 uint8_t *sortKey1; 856 int32_t sortKeyLen1 = 0; 857 uint8_t *sortKey2; 858 int32_t sortKeyLen2 = 0; 859 UErrorCode status = U_ZERO_ERROR; 860 char sortKeyStr1[2048]; 861 uint32_t sortKeyStrLen1 = sizeof(sortKeyStr1)/sizeof(sortKeyStr1[0]); 862 char sortKeyStr2[2048]; 863 uint32_t sortKeyStrLen2 = sizeof(sortKeyStr2)/sizeof(sortKeyStr2[0]); 864 UCollationResult result; 865 866 static UChar preRules[41] = { 0x26, 0x9fa, 0x3c, 0x98c, 0x3c, 0x9e1, 0x3c, 0x98f, 0x3c, 0x990, 0x3c, 0x993, 0x3c, 0x994, 0x3c, 0x9bc, 0x3c, 0x982, 0x3c, 0x983, 0x3c, 0x981, 0x3c, 0x9b0, 0x3c, 0x9b8, 0x3c, 0x9b9, 0x3c, 0x9bd, 0x3c, 0x9be, 0x3c, 0x9bf, 0x3c, 0x9c8, 0x3c, 0x9cb, 0x3d, 0x9cb , 0}; 867 868 rules = preRules; 869 870 log_verbose("Rules: %s\n", aescstrdup(rules, rulesLength)); 871 872 c2 = ucol_openRules(rules, rulesLength, UCOL_DEFAULT, UCOL_DEFAULT_STRENGTH, NULL, &status); 873 if (U_FAILURE(status)) { 874 log_data_err("ERROR: Creating collator from rules failed with locale: %s : %s\n", curLoc, myErrorName(status)); 875 return; 876 } 877 878 sortKeyLen1 = ucol_getSortKey(c2, str1, -1, NULL, 0); 879 sortKey1 = (uint8_t*)malloc(sortKeyLen1+1); 880 ucol_getSortKey(c2,str1,-1,sortKey1, sortKeyLen1+1); 881 ucol_sortKeyToString(c2, sortKey1, sortKeyStr1, sortKeyStrLen1); 882 883 884 sortKeyLen2 = ucol_getSortKey(c2, str2, -1, NULL, 0); 885 sortKey2 = (uint8_t*)malloc(sortKeyLen2+1); 886 ucol_getSortKey(c2,str2,-1,sortKey2, sortKeyLen2+1); 887 888 ucol_sortKeyToString(c2, sortKey2, sortKeyStr2, sortKeyStrLen2); 889 890 891 892 result=ucol_strcoll(c2, str1, -1, str2, -1); 893 if(result!=UCOL_LESS) { 894 log_err("Error: %s was not less than %s: result=%d.\n", aescstrdup(str1,-1), aescstrdup(str2,-1), result); 895 log_info("[%s] -> %s (%d, from rule)\n", aescstrdup(str1,-1), sortKeyStr1, sortKeyLen1); 896 log_info("[%s] -> %s (%d, from rule)\n", aescstrdup(str2,-1), sortKeyStr2, sortKeyLen2); 897 } else { 898 log_verbose("OK: %s was less than %s: result=%d.\n", aescstrdup(str1,-1), aescstrdup(str2,-1), result); 899 log_verbose("[%s] -> %s (%d, from rule)\n", aescstrdup(str1,-1), sortKeyStr1, sortKeyLen1); 900 log_verbose("[%s] -> %s (%d, from rule)\n", aescstrdup(str2,-1), sortKeyStr2, sortKeyLen2); 901 } 902 903 free(sortKey1); 904 free(sortKey2); 905 ucol_close(c2); 906 907 } 908 909 /* 910 TestOpenVsOpenRules ensures that collators from ucol_open and ucol_openRules 911 will generate identical sort keys 912 */ 913 void TestOpenVsOpenRules(){ 914 915 /* create an array of all the locales */ 916 int32_t numLocales = uloc_countAvailable(); 917 int32_t sizeOfStdSet; 918 uint32_t adder; 919 UChar str[41]; /* create an array of UChar of size maximum strSize + 1 */ 920 USet *stdSet; 921 char* curLoc; 922 UCollator * c1; 923 UCollator * c2; 924 const UChar* rules; 925 int32_t rulesLength; 926 int32_t sortKeyLen1, sortKeyLen2; 927 uint8_t *sortKey1 = NULL, *sortKey2 = NULL; 928 char sortKeyStr1[512], sortKeyStr2[512]; 929 uint32_t sortKeyStrLen1 = sizeof(sortKeyStr1) / sizeof(sortKeyStr1[0]), 930 sortKeyStrLen2 = sizeof(sortKeyStr2) / sizeof(sortKeyStr2[0]); 931 ULocaleData *uld; 932 int32_t x, y, z; 933 USet *eSet; 934 int32_t eSize; 935 int strSize; 936 937 UErrorCode err = U_ZERO_ERROR; 938 939 /* create a set of standard characters that aren't very interesting... 940 and then we can find some interesting ones later */ 941 942 stdSet = uset_open(0x61, 0x7A); 943 uset_addRange(stdSet, 0x41, 0x5A); 944 uset_addRange(stdSet, 0x30, 0x39); 945 sizeOfStdSet = uset_size(stdSet); 946 (void)sizeOfStdSet; /* Suppress set but not used warning. */ 947 948 adder = 1; 949 if(getTestOption(QUICK_OPTION)) 950 { 951 adder = 10; 952 } 953 954 for(x = 0; x < numLocales; x+=adder){ 955 curLoc = (char *)uloc_getAvailable(x); 956 log_verbose("Processing %s\n", curLoc); 957 958 /* create a collator the normal API way */ 959 c1 = ucol_open(curLoc, &err); 960 if (U_FAILURE(err)) { 961 log_err("ERROR: Normal collation creation failed with locale: %s : %s\n", curLoc, myErrorName(err)); 962 return; 963 } 964 965 /* grab the rules */ 966 rules = ucol_getRules(c1, &rulesLength); 967 968 /* use those rules to create a collator from rules */ 969 c2 = ucol_openRules(rules, rulesLength, UCOL_DEFAULT, UCOL_DEFAULT_STRENGTH, NULL, &err); 970 if (U_FAILURE(err)) { 971 log_err("ERROR: Creating collator from rules failed with locale: %s : %s\n", curLoc, myErrorName(err)); 972 return; 973 } 974 975 uld = ulocdata_open(curLoc, &err); 976 977 /*now that we have some collators, we get several strings */ 978 979 for(y = 0; y < 5; y++){ 980 981 /* get a set of ALL the characters in this locale */ 982 eSet = ulocdata_getExemplarSet(uld, NULL, 0, ULOCDATA_ES_STANDARD, &err); 983 eSize = uset_size(eSet); 984 985 /* make a string with these characters in it */ 986 strSize = (rand()%40) + 1; 987 988 for(z = 0; z < strSize; z++){ 989 str[z] = uset_charAt(eSet, rand()%eSize); 990 } 991 992 /* change the set to only include 'abnormal' characters (not A-Z, a-z, 0-9 */ 993 uset_removeAll(eSet, stdSet); 994 eSize = uset_size(eSet); 995 996 /* if there are some non-normal characters left, put a few into the string, just to make sure we have some */ 997 if(eSize > 0){ 998 str[2%strSize] = uset_charAt(eSet, rand()%eSize); 999 str[3%strSize] = uset_charAt(eSet, rand()%eSize); 1000 str[5%strSize] = uset_charAt(eSet, rand()%eSize); 1001 str[10%strSize] = uset_charAt(eSet, rand()%eSize); 1002 str[13%strSize] = uset_charAt(eSet, rand()%eSize); 1003 } 1004 /* terminate the string */ 1005 str[strSize-1] = '\0'; 1006 log_verbose("String used: %S\n", str); 1007 1008 /* get sort keys for both of them, and check that the keys are identicle */ 1009 sortKeyLen1 = ucol_getSortKey(c1, str, u_strlen(str), NULL, 0); 1010 sortKey1 = (uint8_t*)malloc(sizeof(uint8_t) * (sortKeyLen1 + 1)); 1011 /*memset(sortKey1, 0xFE, sortKeyLen1);*/ 1012 ucol_getSortKey(c1, str, u_strlen(str), sortKey1, sortKeyLen1 + 1); 1013 ucol_sortKeyToString(c1, sortKey1, sortKeyStr1, sortKeyStrLen1); 1014 1015 sortKeyLen2 = ucol_getSortKey(c2, str, u_strlen(str), NULL, 0); 1016 sortKey2 = (uint8_t*)malloc(sizeof(uint8_t) * (sortKeyLen2 + 1)); 1017 /*memset(sortKey2, 0xFE, sortKeyLen2);*/ 1018 ucol_getSortKey(c2, str, u_strlen(str), sortKey2, sortKeyLen2 + 1); 1019 ucol_sortKeyToString(c2, sortKey2, sortKeyStr2, sortKeyStrLen2); 1020 1021 /* Check that the lengths are the same */ 1022 if (sortKeyLen1 != sortKeyLen2) { 1023 log_err("ERROR : Sort key lengths %d and %d for text '%s' in locale '%s' do not match.\n", 1024 sortKeyLen1, sortKeyLen2, str, curLoc); 1025 } 1026 1027 /* check that the keys are the same */ 1028 if (memcmp(sortKey1, sortKey2, sortKeyLen1) != 0) { 1029 log_err("ERROR : Sort keys '%s' and '%s' for text '%s' in locale '%s' are not equivalent.\n", 1030 sortKeyStr1, sortKeyStr2, str, curLoc); 1031 } 1032 1033 /* clean up after each string */ 1034 free(sortKey1); 1035 free(sortKey2); 1036 uset_close(eSet); 1037 } 1038 /* clean up after each locale */ 1039 ulocdata_close(uld); 1040 ucol_close(c1); 1041 ucol_close(c2); 1042 } 1043 /* final clean up */ 1044 uset_close(stdSet); 1045 } 1046 /* 1047 ---------------------------------------------------------------------------- 1048 ctor -- Tests the getSortKey 1049 */ 1050 void TestSortKey() 1051 { 1052 uint8_t *sortk1 = NULL, *sortk2 = NULL, *sortk3 = NULL, *sortkEmpty = NULL; 1053 int32_t sortklen, osortklen; 1054 UCollator *col; 1055 UChar *test1, *test2, *test3; 1056 UErrorCode status = U_ZERO_ERROR; 1057 char toStringBuffer[256], *resultP; 1058 uint32_t toStringLen=sizeof(toStringBuffer)/sizeof(toStringBuffer[0]); 1059 1060 1061 uint8_t s1[] = { 0x9f, 0x00 }; 1062 uint8_t s2[] = { 0x61, 0x00 }; 1063 int strcmpResult; 1064 1065 strcmpResult = strcmp((const char *)s1, (const char *)s2); 1066 log_verbose("strcmp(0x9f..., 0x61...) = %d\n", strcmpResult); 1067 1068 if(strcmpResult <= 0) { 1069 log_err("ERR: expected strcmp(\"9f 00\", \"61 00\") to be >=0 (GREATER).. got %d. Calling strcmp() for sortkeys may not work! \n", 1070 strcmpResult); 1071 } 1072 1073 1074 log_verbose("testing SortKey begins...\n"); 1075 /* this is supposed to open default date format, but later on it treats it like it is "en_US" 1076 - very bad if you try to run the tests on machine where default locale is NOT "en_US" */ 1077 /* col = ucol_open(NULL, &status); */ 1078 col = ucol_open("en_US", &status); 1079 if (U_FAILURE(status)) { 1080 log_err_status(status, "ERROR: Default collation creation failed.: %s\n", myErrorName(status)); 1081 return; 1082 } 1083 1084 1085 if(ucol_getStrength(col) != UCOL_DEFAULT_STRENGTH) 1086 { 1087 log_err("ERROR: default collation did not have UCOL_DEFAULT_STRENGTH !\n"); 1088 } 1089 /* Need to use identical strength */ 1090 ucol_setAttribute(col, UCOL_STRENGTH, UCOL_IDENTICAL, &status); 1091 1092 test1=(UChar*)malloc(sizeof(UChar) * 6); 1093 test2=(UChar*)malloc(sizeof(UChar) * 6); 1094 test3=(UChar*)malloc(sizeof(UChar) * 6); 1095 1096 memset(test1,0xFE, sizeof(UChar)*6); 1097 memset(test2,0xFE, sizeof(UChar)*6); 1098 memset(test3,0xFE, sizeof(UChar)*6); 1099 1100 1101 u_uastrcpy(test1, "Abcda"); 1102 u_uastrcpy(test2, "abcda"); 1103 u_uastrcpy(test3, "abcda"); 1104 1105 log_verbose("Use tertiary comparison level testing ....\n"); 1106 1107 sortklen=ucol_getSortKey(col, test1, u_strlen(test1), NULL, 0); 1108 sortk1=(uint8_t*)malloc(sizeof(uint8_t) * (sortklen+1)); 1109 memset(sortk1,0xFE, sortklen); 1110 ucol_getSortKey(col, test1, u_strlen(test1), sortk1, sortklen+1); 1111 1112 sortklen=ucol_getSortKey(col, test2, u_strlen(test2), NULL, 0); 1113 sortk2=(uint8_t*)malloc(sizeof(uint8_t) * (sortklen+1)); 1114 memset(sortk2,0xFE, sortklen); 1115 ucol_getSortKey(col, test2, u_strlen(test2), sortk2, sortklen+1); 1116 1117 osortklen = sortklen; 1118 sortklen=ucol_getSortKey(col, test2, u_strlen(test3), NULL, 0); 1119 sortk3=(uint8_t*)malloc(sizeof(uint8_t) * (sortklen+1)); 1120 memset(sortk3,0xFE, sortklen); 1121 ucol_getSortKey(col, test2, u_strlen(test2), sortk3, sortklen+1); 1122 1123 doAssert( (sortklen == osortklen), "Sortkey length should be the same (abcda, abcda)"); 1124 1125 doAssert( (memcmp(sortk1, sortk2, sortklen) > 0), "Result should be \"Abcda\" > \"abcda\""); 1126 doAssert( (memcmp(sortk2, sortk1, sortklen) < 0), "Result should be \"abcda\" < \"Abcda\""); 1127 doAssert( (memcmp(sortk2, sortk3, sortklen) == 0), "Result should be \"abcda\" == \"abcda\""); 1128 1129 resultP = ucol_sortKeyToString(col, sortk3, toStringBuffer, toStringLen); 1130 doAssert( (resultP != 0), "sortKeyToString failed!"); 1131 1132 #if 1 /* verobse log of sortkeys */ 1133 { 1134 char junk2[1000]; 1135 char junk3[1000]; 1136 int i; 1137 1138 strcpy(junk2, "abcda[2] "); 1139 strcpy(junk3, " abcda[3] "); 1140 1141 for(i=0;i<sortklen;i++) 1142 { 1143 sprintf(junk2+strlen(junk2), "%02X ",(int)( 0xFF & sortk2[i])); 1144 sprintf(junk3+strlen(junk3), "%02X ",(int)( 0xFF & sortk3[i])); 1145 } 1146 1147 log_verbose("%s\n", junk2); 1148 log_verbose("%s\n", junk3); 1149 } 1150 #endif 1151 1152 free(sortk1); 1153 free(sortk2); 1154 free(sortk3); 1155 1156 log_verbose("Use secondary comparision level testing ...\n"); 1157 ucol_setStrength(col, UCOL_SECONDARY); 1158 sortklen=ucol_getSortKey(col, test1, u_strlen(test1), NULL, 0); 1159 sortk1=(uint8_t*)malloc(sizeof(uint8_t) * (sortklen+1)); 1160 ucol_getSortKey(col, test1, u_strlen(test1), sortk1, sortklen+1); 1161 sortklen=ucol_getSortKey(col, test2, u_strlen(test2), NULL, 0); 1162 sortk2=(uint8_t*)malloc(sizeof(uint8_t) * (sortklen+1)); 1163 ucol_getSortKey(col, test2, u_strlen(test2), sortk2, sortklen+1); 1164 1165 doAssert( !(memcmp(sortk1, sortk2, sortklen) > 0), "Result should be \"Abcda\" == \"abcda\""); 1166 doAssert( !(memcmp(sortk2, sortk1, sortklen) < 0), "Result should be \"abcda\" == \"Abcda\""); 1167 doAssert( (memcmp(sortk1, sortk2, sortklen) == 0), "Result should be \"abcda\" == \"abcda\""); 1168 1169 log_verbose("getting sortkey for an empty string\n"); 1170 ucol_setAttribute(col, UCOL_STRENGTH, UCOL_TERTIARY, &status); 1171 sortklen = ucol_getSortKey(col, test1, 0, NULL, 0); 1172 sortkEmpty = (uint8_t*)malloc(sizeof(uint8_t) * sortklen+1); 1173 sortklen = ucol_getSortKey(col, test1, 0, sortkEmpty, sortklen+1); 1174 if(sortklen != 3 || sortkEmpty[0] != 1 || sortkEmpty[0] != 1 || sortkEmpty[2] != 0) { 1175 log_err("Empty string generated wrong sortkey!\n"); 1176 } 1177 free(sortkEmpty); 1178 1179 log_verbose("testing passing invalid string\n"); 1180 sortklen = ucol_getSortKey(col, NULL, 10, NULL, 0); 1181 if(sortklen != 0) { 1182 log_err("Invalid string didn't return sortkey size of 0\n"); 1183 } 1184 1185 1186 log_verbose("testing sortkey ends...\n"); 1187 ucol_close(col); 1188 free(test1); 1189 free(test2); 1190 free(test3); 1191 free(sortk1); 1192 free(sortk2); 1193 1194 } 1195 void TestHashCode() 1196 { 1197 uint8_t *sortk1, *sortk2, *sortk3; 1198 int32_t sortk1len, sortk2len, sortk3len; 1199 UCollator *col; 1200 UChar *test1, *test2, *test3; 1201 UErrorCode status = U_ZERO_ERROR; 1202 log_verbose("testing getHashCode begins...\n"); 1203 col = ucol_open("en_US", &status); 1204 if (U_FAILURE(status)) { 1205 log_err_status(status, "ERROR: Default collation creation failed.: %s\n", myErrorName(status)); 1206 return; 1207 } 1208 test1=(UChar*)malloc(sizeof(UChar) * 6); 1209 test2=(UChar*)malloc(sizeof(UChar) * 6); 1210 test3=(UChar*)malloc(sizeof(UChar) * 6); 1211 u_uastrcpy(test1, "Abcda"); 1212 u_uastrcpy(test2, "abcda"); 1213 u_uastrcpy(test3, "abcda"); 1214 1215 log_verbose("Use tertiary comparison level testing ....\n"); 1216 sortk1len=ucol_getSortKey(col, test1, u_strlen(test1), NULL, 0); 1217 sortk1=(uint8_t*)malloc(sizeof(uint8_t) * (sortk1len+1)); 1218 ucol_getSortKey(col, test1, u_strlen(test1), sortk1, sortk1len+1); 1219 sortk2len=ucol_getSortKey(col, test2, u_strlen(test2), NULL, 0); 1220 sortk2=(uint8_t*)malloc(sizeof(uint8_t) * (sortk2len+1)); 1221 ucol_getSortKey(col, test2, u_strlen(test2), sortk2, sortk2len+1); 1222 sortk3len=ucol_getSortKey(col, test2, u_strlen(test3), NULL, 0); 1223 sortk3=(uint8_t*)malloc(sizeof(uint8_t) * (sortk3len+1)); 1224 ucol_getSortKey(col, test2, u_strlen(test2), sortk3, sortk3len+1); 1225 1226 1227 log_verbose("ucol_hashCode() testing ...\n"); 1228 1229 doAssert( ucol_keyHashCode(sortk1, sortk1len) != ucol_keyHashCode(sortk2, sortk2len), "Hash test1 result incorrect" ); 1230 doAssert( !(ucol_keyHashCode(sortk1, sortk1len) == ucol_keyHashCode(sortk2, sortk2len)), "Hash test2 result incorrect" ); 1231 doAssert( ucol_keyHashCode(sortk2, sortk2len) == ucol_keyHashCode(sortk3, sortk3len), "Hash result not equal" ); 1232 1233 log_verbose("hashCode tests end.\n"); 1234 ucol_close(col); 1235 free(sortk1); 1236 free(sortk2); 1237 free(sortk3); 1238 free(test1); 1239 free(test2); 1240 free(test3); 1241 1242 1243 } 1244 /* 1245 *---------------------------------------------------------------------------- 1246 * Tests the UCollatorElements API. 1247 * 1248 */ 1249 void TestElemIter() 1250 { 1251 int32_t offset; 1252 int32_t order1, order2, order3; 1253 UChar *testString1, *testString2; 1254 UCollator *col; 1255 UCollationElements *iterator1, *iterator2, *iterator3; 1256 UErrorCode status = U_ZERO_ERROR; 1257 log_verbose("testing UCollatorElements begins...\n"); 1258 col = ucol_open("en_US", &status); 1259 ucol_setAttribute(col, UCOL_NORMALIZATION_MODE, UCOL_OFF, &status); 1260 if (U_FAILURE(status)) { 1261 log_err_status(status, "ERROR: Default collation creation failed.: %s\n", myErrorName(status)); 1262 return; 1263 } 1264 1265 testString1=(UChar*)malloc(sizeof(UChar) * 150); 1266 testString2=(UChar*)malloc(sizeof(UChar) * 150); 1267 u_uastrcpy(testString1, "XFILE What subset of all possible test cases has the highest probability of detecting the most errors?"); 1268 u_uastrcpy(testString2, "Xf_ile What subset of all possible test cases has the lowest probability of detecting the least errors?"); 1269 1270 log_verbose("Constructors and comparison testing....\n"); 1271 1272 iterator1 = ucol_openElements(col, testString1, u_strlen(testString1), &status); 1273 if(U_FAILURE(status)) { 1274 log_err("ERROR: Default collationElement iterator creation failed.: %s\n", myErrorName(status)); 1275 ucol_close(col); 1276 return; 1277 } 1278 else{ log_verbose("PASS: Default collationElement iterator1 creation passed\n");} 1279 1280 iterator2 = ucol_openElements(col, testString1, u_strlen(testString1), &status); 1281 if(U_FAILURE(status)) { 1282 log_err("ERROR: Default collationElement iterator creation failed.: %s\n", myErrorName(status)); 1283 ucol_close(col); 1284 return; 1285 } 1286 else{ log_verbose("PASS: Default collationElement iterator2 creation passed\n");} 1287 1288 iterator3 = ucol_openElements(col, testString2, u_strlen(testString2), &status); 1289 if(U_FAILURE(status)) { 1290 log_err("ERROR: Default collationElement iterator creation failed.: %s\n", myErrorName(status)); 1291 ucol_close(col); 1292 return; 1293 } 1294 else{ log_verbose("PASS: Default collationElement iterator3 creation passed\n");} 1295 1296 offset=ucol_getOffset(iterator1); 1297 (void)offset; /* Suppress set but not used warning. */ 1298 ucol_setOffset(iterator1, 6, &status); 1299 if (U_FAILURE(status)) { 1300 log_err("Error in setOffset for UCollatorElements iterator.: %s\n", myErrorName(status)); 1301 return; 1302 } 1303 if(ucol_getOffset(iterator1)==6) 1304 log_verbose("setOffset and getOffset working fine\n"); 1305 else{ 1306 log_err("error in set and get Offset got %d instead of 6\n", ucol_getOffset(iterator1)); 1307 } 1308 1309 ucol_setOffset(iterator1, 0, &status); 1310 order1 = ucol_next(iterator1, &status); 1311 if (U_FAILURE(status)) { 1312 log_err("Somehow ran out of memory stepping through the iterator1.: %s\n", myErrorName(status)); 1313 return; 1314 } 1315 order2=ucol_getOffset(iterator2); 1316 doAssert((order1 != order2), "The first iterator advance failed"); 1317 order2 = ucol_next(iterator2, &status); 1318 if (U_FAILURE(status)) { 1319 log_err("Somehow ran out of memory stepping through the iterator2.: %s\n", myErrorName(status)); 1320 return; 1321 } 1322 order3 = ucol_next(iterator3, &status); 1323 if (U_FAILURE(status)) { 1324 log_err("Somehow ran out of memory stepping through the iterator3.: %s\n", myErrorName(status)); 1325 return; 1326 } 1327 1328 doAssert((order1 == order2), "The second iterator advance failed should be the same as first one"); 1329 1330 doAssert( (ucol_primaryOrder(order1) == ucol_primaryOrder(order3)), "The primary orders should be identical"); 1331 doAssert( (ucol_secondaryOrder(order1) == ucol_secondaryOrder(order3)), "The secondary orders should be identical"); 1332 doAssert( (ucol_tertiaryOrder(order1) == ucol_tertiaryOrder(order3)), "The tertiary orders should be identical"); 1333 1334 order1=ucol_next(iterator1, &status); 1335 if (U_FAILURE(status)) { 1336 log_err("Somehow ran out of memory stepping through the iterator2.: %s\n", myErrorName(status)); 1337 return; 1338 } 1339 order3=ucol_next(iterator3, &status); 1340 if (U_FAILURE(status)) { 1341 log_err("Somehow ran out of memory stepping through the iterator2.: %s\n", myErrorName(status)); 1342 return; 1343 } 1344 doAssert( (ucol_primaryOrder(order1) == ucol_primaryOrder(order3)), "The primary orders should be identical"); 1345 doAssert( (ucol_tertiaryOrder(order1) != ucol_tertiaryOrder(order3)), "The tertiary orders should be different"); 1346 1347 order1=ucol_next(iterator1, &status); 1348 if (U_FAILURE(status)) { 1349 log_err("Somehow ran out of memory stepping through the iterator2.: %s\n", myErrorName(status)); 1350 return; 1351 } 1352 order3=ucol_next(iterator3, &status); 1353 if (U_FAILURE(status)) { 1354 log_err("Somehow ran out of memory stepping through the iterator2.: %s\n", myErrorName(status)); 1355 return; 1356 } 1357 /* this here, my friends, is either pure lunacy or something so obsolete that even it's mother 1358 * doesn't care about it. Essentialy, this test complains if secondary values for 'I' and '_' 1359 * are the same. According to the UCA, this is not true. Therefore, remove the test. 1360 * Besides, if primary strengths for two code points are different, it doesn't matter one bit 1361 * what is the relation between secondary or any other strengths. 1362 * killed by weiv 06/11/2002. 1363 */ 1364 /* 1365 doAssert( ((order1 & UCOL_SECONDARYMASK) != (order3 & UCOL_SECONDARYMASK)), "The secondary orders should be different"); 1366 */ 1367 doAssert( (order1 != UCOL_NULLORDER), "Unexpected end of iterator reached"); 1368 1369 free(testString1); 1370 free(testString2); 1371 ucol_closeElements(iterator1); 1372 ucol_closeElements(iterator2); 1373 ucol_closeElements(iterator3); 1374 ucol_close(col); 1375 1376 log_verbose("testing CollationElementIterator ends...\n"); 1377 } 1378 1379 void TestGetLocale() { 1380 UErrorCode status = U_ZERO_ERROR; 1381 const char *rules = "&a<x<y<z"; 1382 UChar rlz[256] = {0}; 1383 uint32_t rlzLen = u_unescape(rules, rlz, 256); 1384 1385 UCollator *coll = NULL; 1386 const char *locale = NULL; 1387 1388 int32_t i = 0; 1389 1390 static const struct { 1391 const char* requestedLocale; 1392 const char* validLocale; 1393 const char* actualLocale; 1394 } testStruct[] = { 1395 { "sr_RS", "sr_Cyrl_RS", "sr" }, 1396 { "sh_YU", "sr_Latn_RS", "sr_Latn" }, /* was sh, then aliased to hr, now sr_Latn via import per cldrbug 5647: */ 1397 { "en_BE_FOO", "en", "root" }, 1398 { "sv_SE_NONEXISTANT", "sv", "sv" } 1399 }; 1400 1401 /* test opening collators for different locales */ 1402 for(i = 0; i<sizeof(testStruct)/sizeof(testStruct[0]); i++) { 1403 status = U_ZERO_ERROR; 1404 coll = ucol_open(testStruct[i].requestedLocale, &status); 1405 if(U_FAILURE(status)) { 1406 log_err_status(status, "Failed to open collator for %s with %s\n", testStruct[i].requestedLocale, u_errorName(status)); 1407 ucol_close(coll); 1408 continue; 1409 } 1410 /* 1411 * The requested locale may be the same as the valid locale, 1412 * or may not be supported at all. See ticket #10477. 1413 */ 1414 locale = ucol_getLocaleByType(coll, ULOC_REQUESTED_LOCALE, &status); 1415 if(strcmp(locale, testStruct[i].requestedLocale) != 0 && strcmp(locale, testStruct[i].validLocale) != 0) { 1416 log_err("[Coll %s]: Error in requested locale, expected %s, got %s\n", testStruct[i].requestedLocale, testStruct[i].requestedLocale, locale); 1417 } 1418 locale = ucol_getLocaleByType(coll, ULOC_VALID_LOCALE, &status); 1419 if(strcmp(locale, testStruct[i].validLocale) != 0) { 1420 log_err("[Coll %s]: Error in valid locale, expected %s, got %s\n", testStruct[i].requestedLocale, testStruct[i].validLocale, locale); 1421 } 1422 locale = ucol_getLocaleByType(coll, ULOC_ACTUAL_LOCALE, &status); 1423 if(strcmp(locale, testStruct[i].actualLocale) != 0) { 1424 log_err("[Coll %s]: Error in actual locale, expected %s, got %s\n", testStruct[i].requestedLocale, testStruct[i].actualLocale, locale); 1425 } 1426 ucol_close(coll); 1427 } 1428 1429 /* completely non-existant locale for collator should get a default collator */ 1430 { 1431 UCollator *defaultColl = ucol_open(NULL, &status); 1432 coll = ucol_open("blahaha", &status); 1433 if(U_SUCCESS(status)) { 1434 /* See comment above about ticket #10477. 1435 if(strcmp(ucol_getLocaleByType(coll, ULOC_REQUESTED_LOCALE, &status), "blahaha")) { 1436 log_err("Nonexisting locale didn't preserve the requested locale\n"); 1437 } */ 1438 if(strcmp(ucol_getLocaleByType(coll, ULOC_VALID_LOCALE, &status), 1439 ucol_getLocaleByType(defaultColl, ULOC_VALID_LOCALE, &status))) { 1440 log_err("Valid locale for nonexisting locale locale collator differs " 1441 "from valid locale for default collator\n"); 1442 } 1443 if(strcmp(ucol_getLocaleByType(coll, ULOC_ACTUAL_LOCALE, &status), 1444 ucol_getLocaleByType(defaultColl, ULOC_ACTUAL_LOCALE, &status))) { 1445 log_err("Actual locale for nonexisting locale locale collator differs " 1446 "from actual locale for default collator\n"); 1447 } 1448 ucol_close(coll); 1449 ucol_close(defaultColl); 1450 } else { 1451 log_data_err("Couldn't open collators\n"); 1452 } 1453 } 1454 1455 1456 1457 /* collator instantiated from rules should have all three locales NULL */ 1458 coll = ucol_openRules(rlz, rlzLen, UCOL_DEFAULT, UCOL_DEFAULT, NULL, &status); 1459 locale = ucol_getLocaleByType(coll, ULOC_REQUESTED_LOCALE, &status); 1460 if(locale != NULL) { 1461 log_err("For collator instantiated from rules, requested locale returned %s instead of NULL\n", locale); 1462 } 1463 locale = ucol_getLocaleByType(coll, ULOC_VALID_LOCALE, &status); 1464 if(locale != NULL) { 1465 log_err("For collator instantiated from rules, valid locale returned %s instead of NULL\n", locale); 1466 } 1467 locale = ucol_getLocaleByType(coll, ULOC_ACTUAL_LOCALE, &status); 1468 if(locale != NULL) { 1469 log_err("For collator instantiated from rules, actual locale returned %s instead of NULL\n", locale); 1470 } 1471 ucol_close(coll); 1472 1473 } 1474 1475 1476 void TestGetAll() 1477 { 1478 int32_t i, count; 1479 count=ucol_countAvailable(); 1480 /* use something sensible w/o hardcoding the count */ 1481 if(count < 0){ 1482 log_err("Error in countAvailable(), it returned %d\n", count); 1483 } 1484 else{ 1485 log_verbose("PASS: countAvailable() successful, it returned %d\n", count); 1486 } 1487 for(i=0;i<count;i++) 1488 log_verbose("%s\n", ucol_getAvailable(i)); 1489 1490 1491 } 1492 1493 1494 struct teststruct { 1495 const char *original; 1496 uint8_t key[256]; 1497 } ; 1498 1499 static int compare_teststruct(const void *string1, const void *string2) { 1500 return(strcmp((const char *)((struct teststruct *)string1)->key, (const char *)((struct teststruct *)string2)->key)); 1501 } 1502 1503 void TestBounds() { 1504 UErrorCode status = U_ZERO_ERROR; 1505 1506 UCollator *coll = ucol_open("sh", &status); 1507 1508 uint8_t sortkey[512], lower[512], upper[512]; 1509 UChar buffer[512]; 1510 1511 static const char * const test[] = { 1512 "John Smith", 1513 "JOHN SMITH", 1514 "john SMITH", 1515 "j\\u00F6hn sm\\u00EFth", 1516 "J\\u00F6hn Sm\\u00EFth", 1517 "J\\u00D6HN SM\\u00CFTH", 1518 "john smithsonian", 1519 "John Smithsonian", 1520 }; 1521 1522 struct teststruct tests[] = { 1523 {"\\u010CAKI MIHALJ" } , 1524 {"\\u010CAKI MIHALJ" } , 1525 {"\\u010CAKI PIRO\\u0160KA" }, 1526 {"\\u010CABAI ANDRIJA" } , 1527 {"\\u010CABAI LAJO\\u0160" } , 1528 {"\\u010CABAI MARIJA" } , 1529 {"\\u010CABAI STEVAN" } , 1530 {"\\u010CABAI STEVAN" } , 1531 {"\\u010CABARKAPA BRANKO" } , 1532 {"\\u010CABARKAPA MILENKO" } , 1533 {"\\u010CABARKAPA MIROSLAV" } , 1534 {"\\u010CABARKAPA SIMO" } , 1535 {"\\u010CABARKAPA STANKO" } , 1536 {"\\u010CABARKAPA TAMARA" } , 1537 {"\\u010CABARKAPA TOMA\\u0160" } , 1538 {"\\u010CABDARI\\u0106 NIKOLA" } , 1539 {"\\u010CABDARI\\u0106 ZORICA" } , 1540 {"\\u010CABI NANDOR" } , 1541 {"\\u010CABOVI\\u0106 MILAN" } , 1542 {"\\u010CABRADI AGNEZIJA" } , 1543 {"\\u010CABRADI IVAN" } , 1544 {"\\u010CABRADI JELENA" } , 1545 {"\\u010CABRADI LJUBICA" } , 1546 {"\\u010CABRADI STEVAN" } , 1547 {"\\u010CABRDA MARTIN" } , 1548 {"\\u010CABRILO BOGDAN" } , 1549 {"\\u010CABRILO BRANISLAV" } , 1550 {"\\u010CABRILO LAZAR" } , 1551 {"\\u010CABRILO LJUBICA" } , 1552 {"\\u010CABRILO SPASOJA" } , 1553 {"\\u010CADE\\u0160 ZDENKA" } , 1554 {"\\u010CADESKI BLAGOJE" } , 1555 {"\\u010CADOVSKI VLADIMIR" } , 1556 {"\\u010CAGLJEVI\\u0106 TOMA" } , 1557 {"\\u010CAGOROVI\\u0106 VLADIMIR" } , 1558 {"\\u010CAJA VANKA" } , 1559 {"\\u010CAJI\\u0106 BOGOLJUB" } , 1560 {"\\u010CAJI\\u0106 BORISLAV" } , 1561 {"\\u010CAJI\\u0106 RADOSLAV" } , 1562 {"\\u010CAK\\u0160IRAN MILADIN" } , 1563 {"\\u010CAKAN EUGEN" } , 1564 {"\\u010CAKAN EVGENIJE" } , 1565 {"\\u010CAKAN IVAN" } , 1566 {"\\u010CAKAN JULIJAN" } , 1567 {"\\u010CAKAN MIHAJLO" } , 1568 {"\\u010CAKAN STEVAN" } , 1569 {"\\u010CAKAN VLADIMIR" } , 1570 {"\\u010CAKAN VLADIMIR" } , 1571 {"\\u010CAKAN VLADIMIR" } , 1572 {"\\u010CAKARA ANA" } , 1573 {"\\u010CAKAREVI\\u0106 MOMIR" } , 1574 {"\\u010CAKAREVI\\u0106 NEDELJKO" } , 1575 {"\\u010CAKI \\u0160ANDOR" } , 1576 {"\\u010CAKI AMALIJA" } , 1577 {"\\u010CAKI ANDRA\\u0160" } , 1578 {"\\u010CAKI LADISLAV" } , 1579 {"\\u010CAKI LAJO\\u0160" } , 1580 {"\\u010CAKI LASLO" } , 1581 }; 1582 1583 1584 1585 int32_t i = 0, j = 0, k = 0, buffSize = 0, skSize = 0, lowerSize = 0, upperSize = 0; 1586 int32_t arraySize = sizeof(tests)/sizeof(tests[0]); 1587 1588 if(U_SUCCESS(status) && coll) { 1589 for(i = 0; i<arraySize; i++) { 1590 buffSize = u_unescape(tests[i].original, buffer, 512); 1591 skSize = ucol_getSortKey(coll, buffer, buffSize, tests[i].key, 512); 1592 } 1593 1594 qsort(tests, arraySize, sizeof(struct teststruct), compare_teststruct); 1595 1596 for(i = 0; i < arraySize-1; i++) { 1597 for(j = i+1; j < arraySize; j++) { 1598 lowerSize = ucol_getBound(tests[i].key, -1, UCOL_BOUND_LOWER, 1, lower, 512, &status); 1599 upperSize = ucol_getBound(tests[j].key, -1, UCOL_BOUND_UPPER, 1, upper, 512, &status); 1600 (void)lowerSize; /* Suppress set but not used warning. */ 1601 (void)upperSize; 1602 for(k = i; k <= j; k++) { 1603 if(strcmp((const char *)lower, (const char *)tests[k].key) > 0) { 1604 log_err("Problem with lower! j = %i (%s vs %s)\n", k, tests[k].original, tests[i].original); 1605 } 1606 if(strcmp((const char *)upper, (const char *)tests[k].key) <= 0) { 1607 log_err("Problem with upper! j = %i (%s vs %s)\n", k, tests[k].original, tests[j].original); 1608 } 1609 } 1610 } 1611 } 1612 1613 1614 #if 0 1615 for(i = 0; i < 1000; i++) { 1616 lowerRND = (rand()/(RAND_MAX/arraySize)); 1617 upperRND = lowerRND + (rand()/(RAND_MAX/(arraySize-lowerRND))); 1618 1619 lowerSize = ucol_getBound(tests[lowerRND].key, -1, UCOL_BOUND_LOWER, 1, lower, 512, &status); 1620 upperSize = ucol_getBound(tests[upperRND].key, -1, UCOL_BOUND_UPPER_LONG, 1, upper, 512, &status); 1621 1622 for(j = lowerRND; j<=upperRND; j++) { 1623 if(strcmp(lower, tests[j].key) > 0) { 1624 log_err("Problem with lower! j = %i (%s vs %s)\n", j, tests[j].original, tests[lowerRND].original); 1625 } 1626 if(strcmp(upper, tests[j].key) <= 0) { 1627 log_err("Problem with upper! j = %i (%s vs %s)\n", j, tests[j].original, tests[upperRND].original); 1628 } 1629 } 1630 } 1631 #endif 1632 1633 1634 1635 1636 1637 for(i = 0; i<sizeof(test)/sizeof(test[0]); i++) { 1638 buffSize = u_unescape(test[i], buffer, 512); 1639 skSize = ucol_getSortKey(coll, buffer, buffSize, sortkey, 512); 1640 lowerSize = ucol_getBound(sortkey, skSize, UCOL_BOUND_LOWER, 1, lower, 512, &status); 1641 upperSize = ucol_getBound(sortkey, skSize, UCOL_BOUND_UPPER_LONG, 1, upper, 512, &status); 1642 for(j = i+1; j<sizeof(test)/sizeof(test[0]); j++) { 1643 buffSize = u_unescape(test[j], buffer, 512); 1644 skSize = ucol_getSortKey(coll, buffer, buffSize, sortkey, 512); 1645 if(strcmp((const char *)lower, (const char *)sortkey) > 0) { 1646 log_err("Problem with lower! i = %i, j = %i (%s vs %s)\n", i, j, test[i], test[j]); 1647 } 1648 if(strcmp((const char *)upper, (const char *)sortkey) <= 0) { 1649 log_err("Problem with upper! i = %i, j = %i (%s vs %s)\n", i, j, test[i], test[j]); 1650 } 1651 } 1652 } 1653 ucol_close(coll); 1654 } else { 1655 log_data_err("Couldn't open collator\n"); 1656 } 1657 1658 } 1659 1660 static void doOverrunTest(UCollator *coll, const UChar *uString, int32_t strLen) { 1661 int32_t skLen = 0, skLen2 = 0; 1662 uint8_t sortKey[256]; 1663 int32_t i, j; 1664 uint8_t filler = 0xFF; 1665 1666 skLen = ucol_getSortKey(coll, uString, strLen, NULL, 0); 1667 1668 for(i = 0; i < skLen; i++) { 1669 memset(sortKey, filler, 256); 1670 skLen2 = ucol_getSortKey(coll, uString, strLen, sortKey, i); 1671 if(skLen != skLen2) { 1672 log_err("For buffer size %i, got different sortkey length. Expected %i got %i\n", i, skLen, skLen2); 1673 } 1674 for(j = i; j < 256; j++) { 1675 if(sortKey[j] != filler) { 1676 log_err("Something run over index %i\n", j); 1677 break; 1678 } 1679 } 1680 } 1681 } 1682 1683 /* j1865 reports that if a shorter buffer is passed to 1684 * to get sort key, a buffer overrun happens in some 1685 * cases. This test tries to check this. 1686 */ 1687 void TestSortKeyBufferOverrun(void) { 1688 UErrorCode status = U_ZERO_ERROR; 1689 const char* cString = "A very Merry liTTle-lamB.."; 1690 UChar uString[256]; 1691 int32_t strLen = 0; 1692 UCollator *coll = ucol_open("root", &status); 1693 strLen = u_unescape(cString, uString, 256); 1694 1695 if(U_SUCCESS(status)) { 1696 log_verbose("testing non ignorable\n"); 1697 ucol_setAttribute(coll, UCOL_ALTERNATE_HANDLING, UCOL_NON_IGNORABLE, &status); 1698 doOverrunTest(coll, uString, strLen); 1699 1700 log_verbose("testing shifted\n"); 1701 ucol_setAttribute(coll, UCOL_ALTERNATE_HANDLING, UCOL_SHIFTED, &status); 1702 doOverrunTest(coll, uString, strLen); 1703 1704 log_verbose("testing shifted quaternary\n"); 1705 ucol_setAttribute(coll, UCOL_STRENGTH, UCOL_QUATERNARY, &status); 1706 doOverrunTest(coll, uString, strLen); 1707 1708 log_verbose("testing with french secondaries\n"); 1709 ucol_setAttribute(coll, UCOL_FRENCH_COLLATION, UCOL_ON, &status); 1710 ucol_setAttribute(coll, UCOL_STRENGTH, UCOL_TERTIARY, &status); 1711 ucol_setAttribute(coll, UCOL_ALTERNATE_HANDLING, UCOL_NON_IGNORABLE, &status); 1712 doOverrunTest(coll, uString, strLen); 1713 1714 } 1715 ucol_close(coll); 1716 } 1717 1718 static void TestAttribute() 1719 { 1720 UErrorCode error = U_ZERO_ERROR; 1721 UCollator *coll = ucol_open(NULL, &error); 1722 1723 if (U_FAILURE(error)) { 1724 log_err_status(error, "Creation of default collator failed\n"); 1725 return; 1726 } 1727 1728 ucol_setAttribute(coll, UCOL_FRENCH_COLLATION, UCOL_OFF, &error); 1729 if (ucol_getAttribute(coll, UCOL_FRENCH_COLLATION, &error) != UCOL_OFF || 1730 U_FAILURE(error)) { 1731 log_err_status(error, "Setting and retrieving of the french collation failed\n"); 1732 } 1733 1734 ucol_setAttribute(coll, UCOL_FRENCH_COLLATION, UCOL_ON, &error); 1735 if (ucol_getAttribute(coll, UCOL_FRENCH_COLLATION, &error) != UCOL_ON || 1736 U_FAILURE(error)) { 1737 log_err_status(error, "Setting and retrieving of the french collation failed\n"); 1738 } 1739 1740 ucol_setAttribute(coll, UCOL_ALTERNATE_HANDLING, UCOL_SHIFTED, &error); 1741 if (ucol_getAttribute(coll, UCOL_ALTERNATE_HANDLING, &error) != UCOL_SHIFTED || 1742 U_FAILURE(error)) { 1743 log_err_status(error, "Setting and retrieving of the alternate handling failed\n"); 1744 } 1745 1746 ucol_setAttribute(coll, UCOL_ALTERNATE_HANDLING, UCOL_NON_IGNORABLE, &error); 1747 if (ucol_getAttribute(coll, UCOL_ALTERNATE_HANDLING, &error) != UCOL_NON_IGNORABLE || 1748 U_FAILURE(error)) { 1749 log_err_status(error, "Setting and retrieving of the alternate handling failed\n"); 1750 } 1751 1752 ucol_setAttribute(coll, UCOL_CASE_FIRST, UCOL_LOWER_FIRST, &error); 1753 if (ucol_getAttribute(coll, UCOL_CASE_FIRST, &error) != UCOL_LOWER_FIRST || 1754 U_FAILURE(error)) { 1755 log_err_status(error, "Setting and retrieving of the case first attribute failed\n"); 1756 } 1757 1758 ucol_setAttribute(coll, UCOL_CASE_FIRST, UCOL_UPPER_FIRST, &error); 1759 if (ucol_getAttribute(coll, UCOL_CASE_FIRST, &error) != UCOL_UPPER_FIRST || 1760 U_FAILURE(error)) { 1761 log_err_status(error, "Setting and retrieving of the case first attribute failed\n"); 1762 } 1763 1764 ucol_setAttribute(coll, UCOL_CASE_LEVEL, UCOL_ON, &error); 1765 if (ucol_getAttribute(coll, UCOL_CASE_LEVEL, &error) != UCOL_ON || 1766 U_FAILURE(error)) { 1767 log_err_status(error, "Setting and retrieving of the case level attribute failed\n"); 1768 } 1769 1770 ucol_setAttribute(coll, UCOL_CASE_LEVEL, UCOL_OFF, &error); 1771 if (ucol_getAttribute(coll, UCOL_CASE_LEVEL, &error) != UCOL_OFF || 1772 U_FAILURE(error)) { 1773 log_err_status(error, "Setting and retrieving of the case level attribute failed\n"); 1774 } 1775 1776 ucol_setAttribute(coll, UCOL_NORMALIZATION_MODE, UCOL_ON, &error); 1777 if (ucol_getAttribute(coll, UCOL_NORMALIZATION_MODE, &error) != UCOL_ON || 1778 U_FAILURE(error)) { 1779 log_err_status(error, "Setting and retrieving of the normalization on/off attribute failed\n"); 1780 } 1781 1782 ucol_setAttribute(coll, UCOL_NORMALIZATION_MODE, UCOL_OFF, &error); 1783 if (ucol_getAttribute(coll, UCOL_NORMALIZATION_MODE, &error) != UCOL_OFF || 1784 U_FAILURE(error)) { 1785 log_err_status(error, "Setting and retrieving of the normalization on/off attribute failed\n"); 1786 } 1787 1788 ucol_setAttribute(coll, UCOL_STRENGTH, UCOL_PRIMARY, &error); 1789 if (ucol_getAttribute(coll, UCOL_STRENGTH, &error) != UCOL_PRIMARY || 1790 U_FAILURE(error)) { 1791 log_err_status(error, "Setting and retrieving of the collation strength failed\n"); 1792 } 1793 1794 ucol_setAttribute(coll, UCOL_STRENGTH, UCOL_SECONDARY, &error); 1795 if (ucol_getAttribute(coll, UCOL_STRENGTH, &error) != UCOL_SECONDARY || 1796 U_FAILURE(error)) { 1797 log_err_status(error, "Setting and retrieving of the collation strength failed\n"); 1798 } 1799 1800 ucol_setAttribute(coll, UCOL_STRENGTH, UCOL_TERTIARY, &error); 1801 if (ucol_getAttribute(coll, UCOL_STRENGTH, &error) != UCOL_TERTIARY || 1802 U_FAILURE(error)) { 1803 log_err_status(error, "Setting and retrieving of the collation strength failed\n"); 1804 } 1805 1806 ucol_setAttribute(coll, UCOL_STRENGTH, UCOL_QUATERNARY, &error); 1807 if (ucol_getAttribute(coll, UCOL_STRENGTH, &error) != UCOL_QUATERNARY || 1808 U_FAILURE(error)) { 1809 log_err_status(error, "Setting and retrieving of the collation strength failed\n"); 1810 } 1811 1812 ucol_setAttribute(coll, UCOL_STRENGTH, UCOL_IDENTICAL, &error); 1813 if (ucol_getAttribute(coll, UCOL_STRENGTH, &error) != UCOL_IDENTICAL || 1814 U_FAILURE(error)) { 1815 log_err_status(error, "Setting and retrieving of the collation strength failed\n"); 1816 } 1817 1818 ucol_close(coll); 1819 } 1820 1821 void TestGetTailoredSet() { 1822 struct { 1823 const char *rules; 1824 const char *tests[20]; 1825 int32_t testsize; 1826 } setTest[] = { 1827 { "&a < \\u212b", { "\\u212b", "A\\u030a", "\\u00c5" }, 3}, 1828 { "& S < \\u0161 <<< \\u0160", { "\\u0161", "s\\u030C", "\\u0160", "S\\u030C" }, 4} 1829 }; 1830 1831 int32_t i = 0, j = 0; 1832 UErrorCode status = U_ZERO_ERROR; 1833 UParseError pError; 1834 1835 UCollator *coll = NULL; 1836 UChar buff[1024]; 1837 int32_t buffLen = 0; 1838 USet *set = NULL; 1839 1840 for(i = 0; i < LENGTHOF(setTest); i++) { 1841 buffLen = u_unescape(setTest[i].rules, buff, 1024); 1842 coll = ucol_openRules(buff, buffLen, UCOL_DEFAULT, UCOL_DEFAULT, &pError, &status); 1843 if(U_SUCCESS(status)) { 1844 set = ucol_getTailoredSet(coll, &status); 1845 if(uset_size(set) < setTest[i].testsize) { 1846 log_err("Tailored set size smaller (%d) than expected (%d)\n", uset_size(set), setTest[i].testsize); 1847 } 1848 for(j = 0; j < setTest[i].testsize; j++) { 1849 buffLen = u_unescape(setTest[i].tests[j], buff, 1024); 1850 if(!uset_containsString(set, buff, buffLen)) { 1851 log_err("Tailored set doesn't contain %s... It should\n", setTest[i].tests[j]); 1852 } 1853 } 1854 uset_close(set); 1855 } else { 1856 log_err_status(status, "Couldn't open collator with rules %s\n", setTest[i].rules); 1857 } 1858 ucol_close(coll); 1859 } 1860 } 1861 1862 static int tMemCmp(const uint8_t *first, const uint8_t *second) { 1863 int32_t firstLen = (int32_t)strlen((const char *)first); 1864 int32_t secondLen = (int32_t)strlen((const char *)second); 1865 return memcmp(first, second, uprv_min(firstLen, secondLen)); 1866 } 1867 static const char * strengthsC[] = { 1868 "UCOL_PRIMARY", 1869 "UCOL_SECONDARY", 1870 "UCOL_TERTIARY", 1871 "UCOL_QUATERNARY", 1872 "UCOL_IDENTICAL" 1873 }; 1874 1875 void TestMergeSortKeys(void) { 1876 UErrorCode status = U_ZERO_ERROR; 1877 UCollator *coll = ucol_open("en", &status); 1878 if(U_SUCCESS(status)) { 1879 1880 const char* cases[] = { 1881 "abc", 1882 "abcd", 1883 "abcde" 1884 }; 1885 uint32_t casesSize = sizeof(cases)/sizeof(cases[0]); 1886 const char* prefix = "foo"; 1887 const char* suffix = "egg"; 1888 char outBuff1[256], outBuff2[256]; 1889 1890 uint8_t **sortkeys = (uint8_t **)malloc(casesSize*sizeof(uint8_t *)); 1891 uint8_t **mergedPrefixkeys = (uint8_t **)malloc(casesSize*sizeof(uint8_t *)); 1892 uint8_t **mergedSuffixkeys = (uint8_t **)malloc(casesSize*sizeof(uint8_t *)); 1893 uint32_t *sortKeysLen = (uint32_t *)malloc(casesSize*sizeof(uint32_t)); 1894 uint8_t prefixKey[256], suffixKey[256]; 1895 uint32_t prefixKeyLen = 0, suffixKeyLen = 0, i = 0; 1896 UChar buffer[256]; 1897 uint32_t unescapedLen = 0, l1 = 0, l2 = 0; 1898 UColAttributeValue strength; 1899 1900 log_verbose("ucol_mergeSortkeys test\n"); 1901 log_verbose("Testing order of the test cases\n"); 1902 genericLocaleStarter("en", cases, casesSize); 1903 1904 for(i = 0; i<casesSize; i++) { 1905 sortkeys[i] = (uint8_t *)malloc(256*sizeof(uint8_t)); 1906 mergedPrefixkeys[i] = (uint8_t *)malloc(256*sizeof(uint8_t)); 1907 mergedSuffixkeys[i] = (uint8_t *)malloc(256*sizeof(uint8_t)); 1908 } 1909 1910 unescapedLen = u_unescape(prefix, buffer, 256); 1911 prefixKeyLen = ucol_getSortKey(coll, buffer, unescapedLen, prefixKey, 256); 1912 1913 unescapedLen = u_unescape(suffix, buffer, 256); 1914 suffixKeyLen = ucol_getSortKey(coll, buffer, unescapedLen, suffixKey, 256); 1915 1916 log_verbose("Massaging data with prefixes and different strengths\n"); 1917 strength = UCOL_PRIMARY; 1918 while(strength <= UCOL_IDENTICAL) { 1919 log_verbose("Strength %s\n", strengthsC[strength<=UCOL_QUATERNARY?strength:4]); 1920 ucol_setAttribute(coll, UCOL_STRENGTH, strength, &status); 1921 for(i = 0; i<casesSize; i++) { 1922 unescapedLen = u_unescape(cases[i], buffer, 256); 1923 sortKeysLen[i] = ucol_getSortKey(coll, buffer, unescapedLen, sortkeys[i], 256); 1924 ucol_mergeSortkeys(prefixKey, prefixKeyLen, sortkeys[i], sortKeysLen[i], mergedPrefixkeys[i], 256); 1925 ucol_mergeSortkeys(sortkeys[i], sortKeysLen[i], suffixKey, suffixKeyLen, mergedSuffixkeys[i], 256); 1926 if(i>0) { 1927 if(tMemCmp(mergedPrefixkeys[i-1], mergedPrefixkeys[i]) >= 0) { 1928 log_err("Error while comparing prefixed keys @ strength %s:\n", strengthsC[strength<=UCOL_QUATERNARY?strength:4]); 1929 log_err("%s\n%s\n", 1930 ucol_sortKeyToString(coll, mergedPrefixkeys[i-1], outBuff1, l1), 1931 ucol_sortKeyToString(coll, mergedPrefixkeys[i], outBuff2, l2)); 1932 } 1933 if(tMemCmp(mergedSuffixkeys[i-1], mergedSuffixkeys[i]) >= 0) { 1934 log_err("Error while comparing suffixed keys @ strength %s:\n", strengthsC[strength<=UCOL_QUATERNARY?strength:4]); 1935 log_err("%s\n%s\n", 1936 ucol_sortKeyToString(coll, mergedSuffixkeys[i-1], outBuff1, l1), 1937 ucol_sortKeyToString(coll, mergedSuffixkeys[i], outBuff2, l2)); 1938 } 1939 } 1940 } 1941 if(strength == UCOL_QUATERNARY) { 1942 strength = UCOL_IDENTICAL; 1943 } else { 1944 strength++; 1945 } 1946 } 1947 1948 { 1949 uint8_t smallBuf[3]; 1950 uint32_t reqLen = 0; 1951 log_verbose("testing buffer overflow\n"); 1952 reqLen = ucol_mergeSortkeys(prefixKey, prefixKeyLen, suffixKey, suffixKeyLen, smallBuf, 3); 1953 if(reqLen != (prefixKeyLen+suffixKeyLen)) { 1954 log_err("Wrong preflight size for merged sortkey\n"); 1955 } 1956 } 1957 1958 { 1959 UChar empty = 0; 1960 uint8_t emptyKey[20], abcKey[50], mergedKey[100]; 1961 int32_t emptyKeyLen = 0, abcKeyLen = 0, mergedKeyLen = 0; 1962 1963 log_verbose("testing merging with sortkeys generated for empty strings\n"); 1964 emptyKeyLen = ucol_getSortKey(coll, &empty, 0, emptyKey, 20); 1965 unescapedLen = u_unescape(cases[0], buffer, 256); 1966 abcKeyLen = ucol_getSortKey(coll, buffer, unescapedLen, abcKey, 50); 1967 mergedKeyLen = ucol_mergeSortkeys(emptyKey, emptyKeyLen, abcKey, abcKeyLen, mergedKey, 100); 1968 if(mergedKey[0] != 2) { 1969 log_err("Empty sortkey didn't produce a level separator\n"); 1970 } 1971 /* try with zeros */ 1972 mergedKeyLen = ucol_mergeSortkeys(emptyKey, 0, abcKey, abcKeyLen, mergedKey, 100); 1973 if(mergedKeyLen != 0 || mergedKey[0] != 0) { 1974 log_err("Empty key didn't produce null mergedKey\n"); 1975 } 1976 mergedKeyLen = ucol_mergeSortkeys(abcKey, abcKeyLen, emptyKey, 0, mergedKey, 100); 1977 if(mergedKeyLen != 0 || mergedKey[0] != 0) { 1978 log_err("Empty key didn't produce null mergedKey\n"); 1979 } 1980 1981 } 1982 1983 for(i = 0; i<casesSize; i++) { 1984 free(sortkeys[i]); 1985 free(mergedPrefixkeys[i]); 1986 free(mergedSuffixkeys[i]); 1987 } 1988 free(sortkeys); 1989 free(mergedPrefixkeys); 1990 free(mergedSuffixkeys); 1991 free(sortKeysLen); 1992 ucol_close(coll); 1993 /* need to finish this up */ 1994 } else { 1995 log_data_err("Couldn't open collator"); 1996 } 1997 } 1998 static void TestShortString(void) 1999 { 2000 struct { 2001 const char *input; 2002 const char *expectedOutput; 2003 const char *locale; 2004 UErrorCode expectedStatus; 2005 int32_t expectedOffset; 2006 uint32_t expectedIdentifier; 2007 } testCases[] = { 2008 /* 2009 * Note: The first test case sets variableTop to the dollar sign '$'. 2010 * We have agreed to drop support for variableTop in ucol_getShortDefinitionString(), 2011 * related to ticket #10372 "deprecate collation APIs for short definition strings", 2012 * and because it did not work for most spaces/punctuation/symbols, 2013 * as documented in ticket #10386 "collation short definition strings issues": 2014 * The old code wrote only 3 hex digits for primary weights below 0x0FFF, 2015 * which is a syntax error, and then failed to normalize the result. 2016 * 2017 * The "B2700" was removed from the expected result ("B2700_KPHONEBOOK_LDE"). 2018 * 2019 * Previously, this test had to be adjusted for root collator changes because the 2020 * primary weight of the variable top character naturally changed 2021 * but was baked into the expected result. 2022 */ 2023 {"LDE_RDE_KPHONEBOOK_T0024_ZLATN","KPHONEBOOK_LDE", "de@collation=phonebook", U_USING_FALLBACK_WARNING, 0, 0 }, 2024 2025 {"LEN_RUS_NO_AS_S4","AS_LROOT_NO_S4", NULL, U_USING_DEFAULT_WARNING, 0, 0 }, 2026 {"LDE_VPHONEBOOK_EO_SI","EO_KPHONEBOOK_LDE_SI", "de@collation=phonebook", U_ZERO_ERROR, 0, 0 }, 2027 {"LDE_Kphonebook","KPHONEBOOK_LDE", "de@collation=phonebook", U_ZERO_ERROR, 0, 0 }, 2028 {"Xqde_DE@collation=phonebookq_S3_EX","KPHONEBOOK_LDE", "de@collation=phonebook", U_USING_FALLBACK_WARNING, 0, 0 }, 2029 {"LFR_FO", "FO_LROOT", NULL, U_USING_DEFAULT_WARNING, 0, 0 }, 2030 {"SO_LX_AS", "", NULL, U_ILLEGAL_ARGUMENT_ERROR, 8, 0 }, 2031 {"S3_ASS_MMM", "", NULL, U_ILLEGAL_ARGUMENT_ERROR, 5, 0 } 2032 }; 2033 2034 int32_t i = 0; 2035 UCollator *coll = NULL, *fromNormalized = NULL; 2036 UParseError parseError; 2037 UErrorCode status = U_ZERO_ERROR; 2038 char fromShortBuffer[256], normalizedBuffer[256], fromNormalizedBuffer[256]; 2039 const char* locale = NULL; 2040 2041 2042 for(i = 0; i < sizeof(testCases)/sizeof(testCases[0]); i++) { 2043 status = U_ZERO_ERROR; 2044 if(testCases[i].locale) { 2045 locale = testCases[i].locale; 2046 } else { 2047 locale = NULL; 2048 } 2049 2050 coll = ucol_openFromShortString(testCases[i].input, FALSE, &parseError, &status); 2051 if(status != testCases[i].expectedStatus) { 2052 log_err_status(status, "Got status '%s' that is different from expected '%s' for '%s'\n", 2053 u_errorName(status), u_errorName(testCases[i].expectedStatus), testCases[i].input); 2054 continue; 2055 } 2056 2057 if(U_SUCCESS(status)) { 2058 ucol_getShortDefinitionString(coll, locale, fromShortBuffer, 256, &status); 2059 2060 if(strcmp(fromShortBuffer, testCases[i].expectedOutput)) { 2061 log_err("Got short string '%s' from the collator. Expected '%s' for input '%s'\n", 2062 fromShortBuffer, testCases[i].expectedOutput, testCases[i].input); 2063 } 2064 2065 ucol_normalizeShortDefinitionString(testCases[i].input, normalizedBuffer, 256, &parseError, &status); 2066 fromNormalized = ucol_openFromShortString(normalizedBuffer, FALSE, &parseError, &status); 2067 ucol_getShortDefinitionString(fromNormalized, locale, fromNormalizedBuffer, 256, &status); 2068 2069 if(strcmp(fromShortBuffer, fromNormalizedBuffer)) { 2070 log_err("Strings obtained from collators instantiated by short string ('%s') and from normalized string ('%s') differ\n", 2071 fromShortBuffer, fromNormalizedBuffer); 2072 } 2073 2074 2075 if(!ucol_equals(coll, fromNormalized)) { 2076 log_err("Collator from short string ('%s') differs from one obtained through a normalized version ('%s')\n", 2077 testCases[i].input, normalizedBuffer); 2078 } 2079 2080 ucol_close(fromNormalized); 2081 ucol_close(coll); 2082 2083 } else { 2084 if(parseError.offset != testCases[i].expectedOffset) { 2085 log_err("Got parse error offset %i, but expected %i instead for '%s'\n", 2086 parseError.offset, testCases[i].expectedOffset, testCases[i].input); 2087 } 2088 } 2089 } 2090 2091 } 2092 2093 static void 2094 doSetsTest(const char *locale, const USet *ref, USet *set, const char* inSet, const char* outSet, UErrorCode *status) { 2095 UChar buffer[65536]; 2096 int32_t bufLen; 2097 2098 uset_clear(set); 2099 bufLen = u_unescape(inSet, buffer, 512); 2100 uset_applyPattern(set, buffer, bufLen, 0, status); 2101 if(U_FAILURE(*status)) { 2102 log_err("%s: Failure setting pattern %s\n", locale, u_errorName(*status)); 2103 } 2104 2105 if(!uset_containsAll(ref, set)) { 2106 log_err("%s: Some stuff from %s is not present in the set\n", locale, inSet); 2107 uset_removeAll(set, ref); 2108 bufLen = uset_toPattern(set, buffer, LENGTHOF(buffer), TRUE, status); 2109 log_info(" missing: %s\n", aescstrdup(buffer, bufLen)); 2110 bufLen = uset_toPattern(ref, buffer, LENGTHOF(buffer), TRUE, status); 2111 log_info(" total: size=%i %s\n", uset_getItemCount(ref), aescstrdup(buffer, bufLen)); 2112 } 2113 2114 uset_clear(set); 2115 bufLen = u_unescape(outSet, buffer, 512); 2116 uset_applyPattern(set, buffer, bufLen, 0, status); 2117 if(U_FAILURE(*status)) { 2118 log_err("%s: Failure setting pattern %s\n", locale, u_errorName(*status)); 2119 } 2120 2121 if(!uset_containsNone(ref, set)) { 2122 log_err("%s: Some stuff from %s is present in the set\n", locale, outSet); 2123 } 2124 } 2125 2126 2127 2128 2129 static void 2130 TestGetContractionsAndUnsafes(void) 2131 { 2132 static struct { 2133 const char* locale; 2134 const char* inConts; 2135 const char* outConts; 2136 const char* inExp; 2137 const char* outExp; 2138 const char* unsafeCodeUnits; 2139 const char* safeCodeUnits; 2140 } tests[] = { 2141 { "ru", 2142 "[{\\u0418\\u0306}{\\u0438\\u0306}]", 2143 "[\\u0439\\u0457]", 2144 "[\\u00e6]", 2145 "[ae]", 2146 "[\\u0418\\u0438]", 2147 "[aAbB\\u0430\\u0410\\u0433\\u0413]" 2148 }, 2149 { "uk", 2150 "[{\\u0406\\u0308}{\\u0456\\u0308}{\\u0418\\u0306}{\\u0438\\u0306}]", 2151 "[\\u0407\\u0419\\u0439\\u0457]", 2152 "[\\u00e6]", 2153 "[ae]", 2154 "[\\u0406\\u0456\\u0418\\u0438]", 2155 "[aAbBxv]", 2156 }, 2157 { "sh", 2158 "[{C\\u0301}{C\\u030C}{C\\u0341}{DZ\\u030C}{Dz\\u030C}{D\\u017D}{D\\u017E}{lj}{nj}]", 2159 "[{\\u309d\\u3099}{\\u30fd\\u3099}]", 2160 "[\\u00e6]", 2161 "[a]", 2162 "[nlcdzNLCDZ]", 2163 "[jabv]" 2164 }, 2165 { "ja", 2166 /* 2167 * The "collv2" builder omits mappings if the collator maps their 2168 * character sequences to the same CEs. 2169 * For example, it omits Japanese contractions for NFD forms 2170 * of the voiced iteration mark (U+309E = U+309D + U+3099), such as 2171 * {\\u3053\\u3099\\u309D\\u3099}{\\u3053\\u309D\\u3099} 2172 * {\\u30B3\\u3099\\u30FD\\u3099}{\\u30B3\\u30FD\\u3099}. 2173 * It does add mappings for the precomposed forms. 2174 */ 2175 "[{\\u3053\\u3099\\u309D}{\\u3053\\u3099\\u309E}{\\u3053\\u3099\\u30FC}" 2176 "{\\u3053\\u309D}{\\u3053\\u309E}{\\u3053\\u30FC}" 2177 "{\\u30B3\\u3099\\u30FC}{\\u30B3\\u3099\\u30FD}{\\u30B3\\u3099\\u30FE}" 2178 "{\\u30B3\\u30FC}{\\u30B3\\u30FD}{\\u30B3\\u30FE}]", 2179 "[{\\u30FD\\u3099}{\\u309D\\u3099}{\\u3053\\u3099}{\\u30B3\\u3099}{lj}{nj}]", 2180 "[\\u30FE\\u00e6]", 2181 "[a]", 2182 "[\\u3099]", 2183 "[]" 2184 } 2185 }; 2186 2187 UErrorCode status = U_ZERO_ERROR; 2188 UCollator *coll = NULL; 2189 int32_t i = 0; 2190 int32_t noConts = 0; 2191 USet *conts = uset_open(0,0); 2192 USet *exp = uset_open(0, 0); 2193 USet *set = uset_open(0,0); 2194 int32_t setBufferLen = 65536; 2195 UChar buffer[65536]; 2196 int32_t setLen = 0; 2197 2198 for(i = 0; i < sizeof(tests)/sizeof(tests[0]); i++) { 2199 log_verbose("Testing locale: %s\n", tests[i].locale); 2200 coll = ucol_open(tests[i].locale, &status); 2201 if (coll == NULL || U_FAILURE(status)) { 2202 log_err_status(status, "Unable to open collator for locale %s ==> %s\n", tests[i].locale, u_errorName(status)); 2203 continue; 2204 } 2205 ucol_getContractionsAndExpansions(coll, conts, exp, TRUE, &status); 2206 doSetsTest(tests[i].locale, conts, set, tests[i].inConts, tests[i].outConts, &status); 2207 setLen = uset_toPattern(conts, buffer, setBufferLen, TRUE, &status); 2208 if(U_SUCCESS(status)) { 2209 /*log_verbose("Contractions %i: %s\n", uset_getItemCount(conts), aescstrdup(buffer, setLen));*/ 2210 } else { 2211 log_err("error %s. %i\n", u_errorName(status), setLen); 2212 status = U_ZERO_ERROR; 2213 } 2214 doSetsTest(tests[i].locale, exp, set, tests[i].inExp, tests[i].outExp, &status); 2215 setLen = uset_toPattern(exp, buffer, setBufferLen, TRUE, &status); 2216 if(U_SUCCESS(status)) { 2217 /*log_verbose("Expansions %i: %s\n", uset_getItemCount(exp), aescstrdup(buffer, setLen));*/ 2218 } else { 2219 log_err("error %s. %i\n", u_errorName(status), setLen); 2220 status = U_ZERO_ERROR; 2221 } 2222 2223 noConts = ucol_getUnsafeSet(coll, conts, &status); 2224 (void)noConts; /* Suppress set but not used warning */ 2225 doSetsTest(tests[i].locale, conts, set, tests[i].unsafeCodeUnits, tests[i].safeCodeUnits, &status); 2226 setLen = uset_toPattern(conts, buffer, setBufferLen, TRUE, &status); 2227 if(U_SUCCESS(status)) { 2228 log_verbose("Unsafe %i: %s\n", uset_getItemCount(exp), aescstrdup(buffer, setLen)); 2229 } else { 2230 log_err("error %s. %i\n", u_errorName(status), setLen); 2231 status = U_ZERO_ERROR; 2232 } 2233 2234 ucol_close(coll); 2235 } 2236 2237 2238 uset_close(conts); 2239 uset_close(exp); 2240 uset_close(set); 2241 } 2242 2243 static void 2244 TestOpenBinary(void) 2245 { 2246 /* 2247 * ucol_openBinary() documents: 2248 * "The API also takes a base collator which usually should be UCA." 2249 * and 2250 * "Currently it cannot be NULL." 2251 * 2252 * However, the check for NULL was commented out in ICU 3.4 (r18149). 2253 * Ticket #4355 requested "Make collation work with minimal data. 2254 * Optionally without UCA, with relevant parts of UCA copied into the tailoring table." 2255 * 2256 * The ICU team agreed with ticket #10517 "require base collator in ucol_openBinary() etc." 2257 * to require base!=NULL again. 2258 */ 2259 #define OPEN_BINARY_ACCEPTS_NULL_BASE 0 2260 UErrorCode status = U_ZERO_ERROR; 2261 /* 2262 char rule[] = "&h < d < c < b"; 2263 char *wUCA[] = { "a", "h", "d", "c", "b", "i" }; 2264 char *noUCA[] = {"d", "c", "b", "a", "h", "i" }; 2265 */ 2266 /* we have to use Cyrillic letters because latin-1 always gets copied */ 2267 const char rule[] = "&\\u0452 < \\u0434 < \\u0433 < \\u0432"; /* &dje < d < g < v */ 2268 const char *wUCA[] = { "\\u0430", "\\u0452", "\\u0434", "\\u0433", "\\u0432", "\\u0435" }; /* a, dje, d, g, v, e */ 2269 #if OPEN_BINARY_ACCEPTS_NULL_BASE 2270 const char *noUCA[] = {"\\u0434", "\\u0433", "\\u0432", "\\u0430", "\\u0435", "\\u0452" }; /* d, g, v, a, e, dje */ 2271 #endif 2272 2273 UChar uRules[256]; 2274 int32_t uRulesLen = u_unescape(rule, uRules, 256); 2275 2276 UCollator *coll = ucol_openRules(uRules, uRulesLen, UCOL_DEFAULT, UCOL_DEFAULT, NULL, &status); 2277 UCollator *UCA = NULL; 2278 UCollator *cloneNOUCA = NULL, *cloneWUCA = NULL; 2279 2280 uint8_t imageBuffer[32768]; 2281 uint8_t *image = imageBuffer; 2282 int32_t imageBufferCapacity = 32768; 2283 2284 int32_t imageSize; 2285 2286 if((coll==NULL)||(U_FAILURE(status))) { 2287 log_data_err("could not load collators or error occured: %s\n", 2288 u_errorName(status)); 2289 return; 2290 } 2291 UCA = ucol_open("root", &status); 2292 if((UCA==NULL)||(U_FAILURE(status))) { 2293 log_data_err("could not load UCA collator or error occured: %s\n", 2294 u_errorName(status)); 2295 return; 2296 } 2297 imageSize = ucol_cloneBinary(coll, image, imageBufferCapacity, &status); 2298 if(U_FAILURE(status)) { 2299 image = (uint8_t *)malloc(imageSize*sizeof(uint8_t)); 2300 status = U_ZERO_ERROR; 2301 imageSize = ucol_cloneBinary(coll, imageBuffer, imageSize, &status); 2302 } 2303 2304 2305 cloneWUCA = ucol_openBinary(image, imageSize, UCA, &status); 2306 cloneNOUCA = ucol_openBinary(image, imageSize, NULL, &status); 2307 #if !OPEN_BINARY_ACCEPTS_NULL_BASE 2308 if(status != U_ILLEGAL_ARGUMENT_ERROR) { 2309 log_err("ucol_openBinary(base=NULL) unexpectedly did not fail - %s\n", u_errorName(status)); 2310 } 2311 #endif 2312 2313 genericOrderingTest(coll, wUCA, sizeof(wUCA)/sizeof(wUCA[0])); 2314 2315 genericOrderingTest(cloneWUCA, wUCA, sizeof(wUCA)/sizeof(wUCA[0])); 2316 #if OPEN_BINARY_ACCEPTS_NULL_BASE 2317 genericOrderingTest(cloneNOUCA, noUCA, sizeof(noUCA)/sizeof(noUCA[0])); 2318 #endif 2319 2320 if(image != imageBuffer) { 2321 free(image); 2322 } 2323 ucol_close(coll); 2324 ucol_close(cloneNOUCA); 2325 ucol_close(cloneWUCA); 2326 ucol_close(UCA); 2327 } 2328 2329 static void TestDefault(void) { 2330 /* Tests for code coverage. */ 2331 UErrorCode status = U_ZERO_ERROR; 2332 UCollator *coll = ucol_open("es@collation=pinyin", &status); 2333 if (coll == NULL || status == U_FILE_ACCESS_ERROR) { 2334 log_data_err("Unable to open collator es@collation=pinyin\n"); 2335 return; 2336 } 2337 if (status != U_USING_DEFAULT_WARNING) { 2338 /* What do you mean that you know about using pinyin collation in Spanish!? This should be in the zh locale. */ 2339 log_err("es@collation=pinyin should return U_USING_DEFAULT_WARNING, but returned %s\n", u_errorName(status)); 2340 } 2341 ucol_close(coll); 2342 if (ucol_getKeywordValues("funky", &status) != NULL) { 2343 log_err("Collators should not know about the funky keyword.\n"); 2344 } 2345 if (status != U_ILLEGAL_ARGUMENT_ERROR) { 2346 log_err("funky keyword didn't fail as expected %s\n", u_errorName(status)); 2347 } 2348 if (ucol_getKeywordValues("collation", &status) != NULL) { 2349 log_err("ucol_getKeywordValues should not work when given a bad status.\n"); 2350 } 2351 } 2352 2353 static void TestDefaultKeyword(void) { 2354 /* Tests for code coverage. */ 2355 UErrorCode status = U_ZERO_ERROR; 2356 const char *loc = "zh_TW@collation=default"; 2357 UCollator *coll = ucol_open(loc, &status); 2358 if(U_FAILURE(status)) { 2359 log_info("Warning: ucol_open(%s, ...) returned %s, at least it didn't crash.\n", loc, u_errorName(status)); 2360 } else if (status != U_USING_FALLBACK_WARNING) { 2361 /* Hmm, skip the following test for CLDR 1.9 data and/or ICU 4.6, no longer seems to apply */ 2362 #if 0 2363 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)); 2364 #endif 2365 } 2366 ucol_close(coll); 2367 } 2368 2369 static void TestGetKeywordValuesForLocale(void) { 2370 #define INCLUDE_UNIHAN_COLLATION 0 2371 #define PREFERRED_SIZE 16 2372 #define MAX_NUMBER_OF_KEYWORDS 9 2373 const char *PREFERRED[PREFERRED_SIZE][MAX_NUMBER_OF_KEYWORDS+1] = { 2374 { "und", "standard", "eor", "search", NULL, NULL, NULL, NULL, NULL, NULL }, 2375 { "en_US", "standard", "eor", "search", NULL, NULL, NULL, NULL, NULL, NULL }, 2376 { "en_029", "standard", "eor", "search", NULL, NULL, NULL, NULL, NULL, NULL }, 2377 { "de_DE", "standard", "phonebook", "search", "eor", NULL, NULL, NULL, NULL, NULL }, 2378 { "de_Latn_DE", "standard", "phonebook", "search", "eor", NULL, NULL, NULL, NULL, NULL }, 2379 #if INCLUDE_UNIHAN_COLLATION 2380 { "zh", "pinyin", "big5han", "gb2312han", "stroke", "unihan", "zhuyin", "eor", "search", "standard" }, 2381 { "zh_Hans", "pinyin", "big5han", "gb2312han", "stroke", "unihan", "zhuyin", "eor", "search", "standard" }, 2382 { "zh_CN", "pinyin", "big5han", "gb2312han", "stroke", "unihan", "zhuyin", "eor", "search", "standard" }, 2383 { "zh_Hant", "stroke", "big5han", "gb2312han", "pinyin", "unihan", "zhuyin", "eor", "search", "standard" }, 2384 { "zh_TW", "stroke", "big5han", "gb2312han", "pinyin", "unihan", "zhuyin", "eor", "search", "standard" }, 2385 { "zh__PINYIN", "pinyin", "big5han", "gb2312han", "stroke", "unihan", "zhuyin", "eor", "search", "standard" }, 2386 #else 2387 // { "zh", "pinyin", "big5han", "gb2312han", "stroke", "zhuyin", "eor", "search", "standard", NULL }, 2388 // { "zh_Hans", "pinyin", "big5han", "gb2312han", "stroke", "zhuyin", "eor", "search", "standard", NULL }, 2389 // { "zh_CN", "pinyin", "big5han", "gb2312han", "stroke", "zhuyin", "eor", "search", "standard", NULL }, 2390 // { "zh_Hant", "stroke", "big5han", "gb2312han", "pinyin", "zhuyin", "eor", "search", "standard", NULL }, 2391 // { "zh_TW", "stroke", "big5han", "gb2312han", "pinyin", "zhuyin", "eor", "search", "standard", NULL }, 2392 // { "zh__PINYIN", "pinyin", "big5han", "gb2312han", "stroke", "zhuyin", "eor", "search", "standard", NULL }, 2393 { "zh", "pinyin", "stroke", "zhuyin", "eor", "search", "standard", NULL, NULL, NULL }, // android-changed 2394 { "zh_Hans", "pinyin", "stroke", "zhuyin", "eor", "search", "standard", NULL, NULL, NULL }, // android-changed 2395 { "zh_CN", "pinyin", "stroke", "zhuyin", "eor", "search", "standard", NULL, NULL, NULL }, // android-changed 2396 { "zh_Hant", "stroke", "pinyin", "zhuyin", "eor", "search", "standard", NULL, NULL, NULL }, // android-changed 2397 { "zh_TW", "stroke", "pinyin", "zhuyin", "eor", "search", "standard", NULL, NULL, NULL }, // android-changed 2398 { "zh__PINYIN", "pinyin", "stroke", "zhuyin", "eor", "search", "standard", NULL, NULL, NULL }, // android-changed 2399 #endif 2400 { "es_ES", "standard", "search", "traditional", "eor", NULL, NULL, NULL, NULL, NULL }, 2401 { "es__TRADITIONAL","traditional", "search", "standard", "eor", NULL, NULL, NULL, NULL, NULL }, 2402 { "und@collation=phonebook", "standard", "eor", "search", NULL, NULL, NULL, NULL, NULL, NULL }, 2403 { "de_DE@collation=big5han", "standard", "phonebook", "search", "eor", NULL, NULL, NULL, NULL, NULL }, 2404 { "zzz@collation=xxx", "standard", "eor", "search", NULL, NULL, NULL, NULL, NULL, NULL } 2405 }; 2406 #if INCLUDE_UNIHAN_COLLATION 2407 const int32_t expectedLength[PREFERRED_SIZE] = { 3, 3, 3, 4, 4, 9, 9, 9, 9, 9, 9, 4, 4, 3, 4, 3 }; 2408 #else 2409 // const int32_t expectedLength[PREFERRED_SIZE] = { 3, 3, 3, 4, 4, 8, 8, 8, 8, 8, 8, 4, 4, 3, 4, 3 }; 2410 const int32_t expectedLength[PREFERRED_SIZE] = { 3, 3, 3, 4, 4, 6, 6, 6, 6, 6, 6, 4, 4, 3, 4, 3 }; // android-changed 2411 #endif 2412 2413 UErrorCode status = U_ZERO_ERROR; 2414 UEnumeration *keywordValues = NULL; 2415 int32_t i, n, size, valueLength; 2416 const char *locale = NULL, *value = NULL; 2417 UBool errorOccurred = FALSE; 2418 2419 for (i = 0; i < PREFERRED_SIZE; i++) { 2420 locale = PREFERRED[i][0]; 2421 value = NULL; 2422 valueLength = 0; 2423 size = 0; 2424 2425 keywordValues = ucol_getKeywordValuesForLocale("collation", locale, TRUE, &status); 2426 if (keywordValues == NULL || U_FAILURE(status)) { 2427 log_err_status(status, "Error getting keyword values: %s\n", u_errorName(status)); 2428 break; 2429 } 2430 size = uenum_count(keywordValues, &status); 2431 2432 if (size == expectedLength[i]) { 2433 for (n = 0; n < expectedLength[i]; n++) { 2434 if ((value = uenum_next(keywordValues, &valueLength, &status)) != NULL && U_SUCCESS(status)) { 2435 if (uprv_strcmp(value, PREFERRED[i][n+1]) != 0) { 2436 log_err("Keyword values differ: Got [%s] Expected [%s] for locale: %s\n", value, PREFERRED[i][n+1], locale); 2437 errorOccurred = TRUE; 2438 break; 2439 } 2440 2441 } else { 2442 log_err("While getting keyword value from locale: %s got this error: %s\n", locale, u_errorName(status)); 2443 errorOccurred = TRUE; 2444 break; 2445 } 2446 } 2447 if (errorOccurred) { 2448 break; 2449 } 2450 } else { 2451 log_err("Number of keywords (%d) does not match expected size (%d) for locale: %s\n", size, expectedLength[i], locale); 2452 break; 2453 } 2454 uenum_close(keywordValues); 2455 keywordValues = NULL; 2456 } 2457 if (keywordValues != NULL) { 2458 uenum_close(keywordValues); 2459 } 2460 } 2461 2462 static void TestStrcollNull(void) { 2463 UErrorCode status = U_ZERO_ERROR; 2464 UCollator *coll; 2465 2466 const UChar u16asc[] = {0x0049, 0x0042, 0x004D, 0}; 2467 const int32_t u16ascLen = 3; 2468 2469 const UChar u16han[] = {0x5c71, 0x5ddd, 0}; 2470 const int32_t u16hanLen = 2; 2471 2472 const char *u8asc = "\x49\x42\x4D"; 2473 const int32_t u8ascLen = 3; 2474 2475 const char *u8han = "\xE5\xB1\xB1\xE5\xB7\x9D"; 2476 const int32_t u8hanLen = 6; 2477 2478 coll = ucol_open(NULL, &status); 2479 if (U_FAILURE(status)) { 2480 log_err_status(status, "Default Collator creation failed.: %s\n", myErrorName(status)); 2481 return; 2482 } 2483 2484 /* UChar API */ 2485 if (ucol_strcoll(coll, NULL, 0, NULL, 0) != 0) { 2486 log_err("ERROR : ucol_strcoll NULL/0 and NULL/0"); 2487 } 2488 2489 if (ucol_strcoll(coll, NULL, -1, NULL, 0) != 0) { 2490 /* No error arg, should return equal without crash */ 2491 log_err("ERROR : ucol_strcoll NULL/-1 and NULL/0"); 2492 } 2493 2494 if (ucol_strcoll(coll, u16asc, -1, NULL, 10) != 0) { 2495 /* No error arg, should return equal without crash */ 2496 log_err("ERROR : ucol_strcoll u16asc/u16ascLen and NULL/10"); 2497 } 2498 2499 if (ucol_strcoll(coll, u16asc, -1, NULL, 0) <= 0) { 2500 log_err("ERROR : ucol_strcoll u16asc/-1 and NULL/0"); 2501 } 2502 if (ucol_strcoll(coll, NULL, 0, u16asc, -1) >= 0) { 2503 log_err("ERROR : ucol_strcoll NULL/0 and u16asc/-1"); 2504 } 2505 if (ucol_strcoll(coll, u16asc, u16ascLen, NULL, 0) <= 0) { 2506 log_err("ERROR : ucol_strcoll u16asc/u16ascLen and NULL/0"); 2507 } 2508 2509 if (ucol_strcoll(coll, u16han, -1, NULL, 0) <= 0) { 2510 log_err("ERROR : ucol_strcoll u16han/-1 and NULL/0"); 2511 } 2512 if (ucol_strcoll(coll, NULL, 0, u16han, -1) >= 0) { 2513 log_err("ERROR : ucol_strcoll NULL/0 and u16han/-1"); 2514 } 2515 if (ucol_strcoll(coll, NULL, 0, u16han, u16hanLen) >= 0) { 2516 log_err("ERROR : ucol_strcoll NULL/0 and u16han/u16hanLen"); 2517 } 2518 2519 /* UTF-8 API */ 2520 status = U_ZERO_ERROR; 2521 if (ucol_strcollUTF8(coll, NULL, 0, NULL, 0, &status) != 0 || U_FAILURE(status)) { 2522 log_err("ERROR : ucol_strcollUTF8 NULL/0 and NULL/0"); 2523 } 2524 status = U_ZERO_ERROR; 2525 ucol_strcollUTF8(coll, NULL, -1, NULL, 0, &status); 2526 if (status != U_ILLEGAL_ARGUMENT_ERROR) { 2527 log_err("ERROR: ucol_strcollUTF8 NULL/-1 and NULL/0, should return U_ILLEGAL_ARGUMENT_ERROR"); 2528 } 2529 status = U_ZERO_ERROR; 2530 ucol_strcollUTF8(coll, u8asc, u8ascLen, NULL, 10, &status); 2531 if (status != U_ILLEGAL_ARGUMENT_ERROR) { 2532 log_err("ERROR: ucol_strcollUTF8 u8asc/u8ascLen and NULL/10, should return U_ILLEGAL_ARGUMENT_ERROR"); 2533 } 2534 2535 status = U_ZERO_ERROR; 2536 if (ucol_strcollUTF8(coll, u8asc, -1, NULL, 0, &status) <= 0 || U_FAILURE(status)) { 2537 log_err("ERROR : ucol_strcollUTF8 u8asc/-1 and NULL/0"); 2538 } 2539 status = U_ZERO_ERROR; 2540 if (ucol_strcollUTF8(coll, NULL, 0, u8asc, -1, &status) >= 0 || U_FAILURE(status)) { 2541 log_err("ERROR : ucol_strcollUTF8 NULL/0 and u8asc/-1"); 2542 } 2543 status = U_ZERO_ERROR; 2544 if (ucol_strcollUTF8(coll, u8asc, u8ascLen, NULL, 0, &status) <= 0 || U_FAILURE(status)) { 2545 log_err("ERROR : ucol_strcollUTF8 u8asc/u8ascLen and NULL/0"); 2546 } 2547 2548 status = U_ZERO_ERROR; 2549 if (ucol_strcollUTF8(coll, u8han, -1, NULL, 0, &status) <= 0 || U_FAILURE(status)) { 2550 log_err("ERROR : ucol_strcollUTF8 u8han/-1 and NULL/0"); 2551 } 2552 status = U_ZERO_ERROR; 2553 if (ucol_strcollUTF8(coll, NULL, 0, u8han, -1, &status) >= 0 || U_FAILURE(status)) { 2554 log_err("ERROR : ucol_strcollUTF8 NULL/0 and u8han/-1"); 2555 } 2556 status = U_ZERO_ERROR; 2557 if (ucol_strcollUTF8(coll, NULL, 0, u8han, u8hanLen, &status) >= 0 || U_FAILURE(status)) { 2558 log_err("ERROR : ucol_strcollUTF8 NULL/0 and u8han/u8hanLen"); 2559 } 2560 2561 ucol_close(coll); 2562 } 2563 2564 #endif /* #if !UCONFIG_NO_COLLATION */ 2565