1 /******************************************************************** 2 * COPYRIGHT: 3 * Copyright (c) 2012-2013, International Business Machines Corporation 4 * and others. All Rights Reserved. 5 ********************************************************************/ 6 // 7 // file: alphaindex.cpp 8 // Alphabetic Index Tests. 9 // 10 // Note: please... no character literals cast to UChars.. use (UChar)0xZZZZ 11 12 #include <stdio.h> // for sprintf 13 14 #include "intltest.h" 15 #include "alphaindextst.h" 16 17 #include "unicode/alphaindex.h" 18 #include "unicode/coll.h" 19 #include "unicode/localpointer.h" 20 #include "unicode/tblcoll.h" 21 #include "unicode/uniset.h" 22 23 #if !UCONFIG_NO_COLLATION && !UCONFIG_NO_NORMALIZATION 24 25 // #include <string> 26 // #include <iostream> 27 28 #define LENGTHOF(array) (int32_t)(sizeof(array)/sizeof((array)[0])) 29 30 namespace { 31 32 UnicodeString joinLabelsAndAppend(AlphabeticIndex::ImmutableIndex &index, UnicodeString &dest) { 33 int32_t oldLength = dest.length(); 34 const AlphabeticIndex::Bucket *bucket; 35 for (int32_t i = 0; (bucket = index.getBucket(i)) != NULL; ++i) { 36 if (dest.length() > oldLength) { 37 dest.append((UChar)0x3A); // ':' 38 } 39 dest.append(bucket->getLabel()); 40 } 41 return dest; 42 } 43 44 } // namespace 45 46 AlphabeticIndexTest::AlphabeticIndexTest() { 47 } 48 49 AlphabeticIndexTest::~AlphabeticIndexTest() { 50 } 51 52 void AlphabeticIndexTest::runIndexedTest( int32_t index, UBool exec, const char* &name, char* /*par*/ ) 53 { 54 if (exec) logln("TestSuite AlphabeticIndex: "); 55 TESTCASE_AUTO_BEGIN; 56 TESTCASE_AUTO(APITest); 57 TESTCASE_AUTO(ManyLocalesTest); 58 TESTCASE_AUTO(HackPinyinTest); 59 TESTCASE_AUTO(TestBug9009); 60 TESTCASE_AUTO(TestIndexCharactersList); 61 TESTCASE_AUTO(TestHaniFirst); 62 TESTCASE_AUTO(TestPinyinFirst); 63 TESTCASE_AUTO(TestSchSt); 64 TESTCASE_AUTO(TestNoLabels); 65 TESTCASE_AUTO(TestChineseZhuyin); 66 TESTCASE_AUTO_END; 67 } 68 69 #define TEST_CHECK_STATUS {if (U_FAILURE(status)) {dataerrln("%s:%d: Test failure. status=%s", \ 70 __FILE__, __LINE__, u_errorName(status)); return;}} 71 72 #define TEST_ASSERT(expr) {if ((expr)==FALSE) {errln("%s:%d: Test failure \n", __FILE__, __LINE__);};} 73 74 // 75 // APITest. Invoke every function at least once, and check that it does something. 76 // Does not attempt to check complete functionality. 77 // 78 void AlphabeticIndexTest::APITest() { 79 // 80 // Simple constructor and destructor, getBucketCount() 81 // 82 UErrorCode status = U_ZERO_ERROR; 83 int32_t lc = 0; 84 int32_t i = 0; 85 AlphabeticIndex *index = new AlphabeticIndex(Locale::getEnglish(), status); 86 TEST_CHECK_STATUS; 87 lc = index->getBucketCount(status); 88 TEST_CHECK_STATUS; 89 TEST_ASSERT(28 == lc); // 26 letters plus two under/overflow labels. 90 //printf("getBucketCount() == %d\n", lc); 91 delete index; 92 93 // Constructor from a Collator 94 // 95 status = U_ZERO_ERROR; 96 RuleBasedCollator *coll = dynamic_cast<RuleBasedCollator *>(Collator::createInstance(Locale::getChinese(), status)); 97 TEST_CHECK_STATUS; 98 TEST_ASSERT(coll != NULL); 99 index = new AlphabeticIndex(coll, status); 100 TEST_CHECK_STATUS; 101 TEST_ASSERT(coll == &index->getCollator()); 102 assertEquals("only the underflow label in an index built from a collator", 103 1, index->getBucketCount(status)); 104 TEST_CHECK_STATUS; 105 delete index; 106 107 108 // addLabels() 109 110 status = U_ZERO_ERROR; 111 index = new AlphabeticIndex(Locale::getEnglish(), status); 112 TEST_CHECK_STATUS; 113 UnicodeSet additions; 114 additions.add((UChar32)0x410).add((UChar32)0x415); // A couple of Cyrillic letters 115 index->addLabels(additions, status); 116 TEST_CHECK_STATUS; 117 lc = index->getBucketCount(status); 118 TEST_CHECK_STATUS; 119 assertEquals("underflow, A-Z, inflow, 2 Cyrillic, overflow", 120 31, index->getBucketCount(status)); 121 // std::cout << lc << std::endl; 122 delete index; 123 124 125 // addLabels(Locale) 126 127 status = U_ZERO_ERROR; 128 index = new AlphabeticIndex(Locale::getEnglish(), status); 129 TEST_CHECK_STATUS; 130 AlphabeticIndex &aip = index->addLabels(Locale::getJapanese(), status); 131 TEST_ASSERT(&aip == index); 132 TEST_CHECK_STATUS; 133 lc = index->getBucketCount(status); 134 TEST_CHECK_STATUS; 135 TEST_ASSERT(35 < lc); // Japanese should add a bunch. Don't rely on the exact value. 136 delete index; 137 138 // GetCollator(), Get under/in/over flow labels 139 140 status = U_ZERO_ERROR; 141 index = new AlphabeticIndex(Locale::getGerman(), status); 142 TEST_CHECK_STATUS; 143 Collator *germanCol = Collator::createInstance(Locale::getGerman(), status); 144 TEST_CHECK_STATUS; 145 const RuleBasedCollator &indexCol = index->getCollator(); 146 TEST_ASSERT(*germanCol == indexCol); 147 delete germanCol; 148 149 UnicodeString ELLIPSIS; ELLIPSIS.append((UChar32)0x2026); 150 UnicodeString s = index->getUnderflowLabel(); 151 TEST_ASSERT(ELLIPSIS == s); 152 s = index->getOverflowLabel(); 153 TEST_ASSERT(ELLIPSIS == s); 154 s = index->getInflowLabel(); 155 TEST_ASSERT(ELLIPSIS == s); 156 index->setOverflowLabel(UNICODE_STRING_SIMPLE("O"), status); 157 index->setUnderflowLabel(UNICODE_STRING_SIMPLE("U"), status).setInflowLabel(UNICODE_STRING_SIMPLE("I"), status); 158 s = index->getUnderflowLabel(); 159 TEST_ASSERT(UNICODE_STRING_SIMPLE("U") == s); 160 s = index->getOverflowLabel(); 161 TEST_ASSERT(UNICODE_STRING_SIMPLE("O") == s); 162 s = index->getInflowLabel(); 163 TEST_ASSERT(UNICODE_STRING_SIMPLE("I") == s); 164 165 166 167 168 delete index; 169 170 171 172 const UnicodeString adam = UNICODE_STRING_SIMPLE("Adam"); 173 const UnicodeString baker = UNICODE_STRING_SIMPLE("Baker"); 174 const UnicodeString charlie = UNICODE_STRING_SIMPLE("Charlie"); 175 const UnicodeString chad = UNICODE_STRING_SIMPLE("Chad"); 176 const UnicodeString zed = UNICODE_STRING_SIMPLE("Zed"); 177 const UnicodeString Cyrillic = UNICODE_STRING_SIMPLE("\\u0410\\u0443\\u0435").unescape(); 178 179 // addRecord(), verify that it comes back out. 180 // 181 status = U_ZERO_ERROR; 182 index = new AlphabeticIndex(Locale::getEnglish(), status); 183 TEST_CHECK_STATUS; 184 index->addRecord(UnicodeString("Adam"), this, status); 185 UBool b; 186 TEST_CHECK_STATUS; 187 index->resetBucketIterator(status); 188 TEST_CHECK_STATUS; 189 index->nextBucket(status); // Move to underflow label 190 index->nextBucket(status); // Move to "A" 191 TEST_CHECK_STATUS; 192 const UnicodeString &label2 = index->getBucketLabel(); 193 UnicodeString A_STR = UNICODE_STRING_SIMPLE("A"); 194 TEST_ASSERT(A_STR == label2); 195 196 b = index->nextRecord(status); 197 TEST_CHECK_STATUS; 198 TEST_ASSERT(b); 199 const UnicodeString &itemName = index->getRecordName(); 200 TEST_ASSERT(adam == itemName); 201 202 const void *itemContext = index->getRecordData(); 203 TEST_ASSERT(itemContext == this); 204 205 delete index; 206 207 // clearRecords, addRecord(), Iteration 208 209 status = U_ZERO_ERROR; 210 index = new AlphabeticIndex(Locale::getEnglish(), status); 211 TEST_CHECK_STATUS; 212 while (index->nextBucket(status)) { 213 TEST_CHECK_STATUS; 214 while (index->nextRecord(status)) { 215 TEST_CHECK_STATUS; 216 TEST_ASSERT(FALSE); // No items have been added. 217 } 218 TEST_CHECK_STATUS; 219 } 220 221 index->addRecord(adam, NULL, status); 222 index->addRecord(baker, NULL, status); 223 index->addRecord(charlie, NULL, status); 224 index->addRecord(chad, NULL, status); 225 TEST_CHECK_STATUS; 226 int itemCount = 0; 227 index->resetBucketIterator(status); 228 while (index->nextBucket(status)) { 229 TEST_CHECK_STATUS; 230 while (index->nextRecord(status)) { 231 TEST_CHECK_STATUS; 232 ++itemCount; 233 } 234 } 235 TEST_CHECK_STATUS; 236 TEST_ASSERT(itemCount == 4); 237 238 TEST_ASSERT(index->nextBucket(status) == FALSE); 239 index->resetBucketIterator(status); 240 TEST_CHECK_STATUS; 241 TEST_ASSERT(index->nextBucket(status) == TRUE); 242 243 index->clearRecords(status); 244 TEST_CHECK_STATUS; 245 index->resetBucketIterator(status); 246 while (index->nextBucket(status)) { 247 TEST_CHECK_STATUS; 248 while (index->nextRecord(status)) { 249 TEST_ASSERT(FALSE); // No items have been added. 250 } 251 } 252 TEST_CHECK_STATUS; 253 delete index; 254 255 // getBucketLabel(), getBucketType() 256 257 status = U_ZERO_ERROR; 258 index = new AlphabeticIndex(Locale::getEnglish(), status); 259 TEST_CHECK_STATUS; 260 index->setUnderflowLabel(adam, status).setOverflowLabel(charlie, status); 261 TEST_CHECK_STATUS; 262 for (i=0; index->nextBucket(status); i++) { 263 TEST_CHECK_STATUS; 264 UnicodeString label = index->getBucketLabel(); 265 UAlphabeticIndexLabelType type = index->getBucketLabelType(); 266 if (i == 0) { 267 TEST_ASSERT(type == U_ALPHAINDEX_UNDERFLOW); 268 TEST_ASSERT(label == adam); 269 } else if (i <= 26) { 270 // Labels A - Z for English locale 271 TEST_ASSERT(type == U_ALPHAINDEX_NORMAL); 272 UnicodeString expectedLabel((UChar)(0x40 + i)); 273 TEST_ASSERT(expectedLabel == label); 274 } else if (i == 27) { 275 TEST_ASSERT(type == U_ALPHAINDEX_OVERFLOW); 276 TEST_ASSERT(label == charlie); 277 } else { 278 TEST_ASSERT(FALSE); 279 } 280 } 281 TEST_ASSERT(i==28); 282 delete index; 283 284 // getBucketIndex() 285 286 status = U_ZERO_ERROR; 287 index = new AlphabeticIndex(Locale::getEnglish(), status); 288 TEST_CHECK_STATUS; 289 int32_t n = index->getBucketIndex(adam, status); 290 TEST_CHECK_STATUS; 291 TEST_ASSERT(n == 1); /* Label #0 is underflow, 1 is A, etc. */ 292 n = index->getBucketIndex(baker, status); 293 TEST_ASSERT(n == 2); 294 n = index->getBucketIndex(Cyrillic, status); 295 TEST_ASSERT(n == 27); // Overflow label 296 n = index->getBucketIndex(zed, status); 297 TEST_ASSERT(n == 26); 298 299 for (i=0; index->nextBucket(status); i++) { 300 n = index->getBucketIndex(); 301 TEST_ASSERT(n == i); 302 UnicodeString label = index->getBucketLabel(); 303 TEST_ASSERT(n == i); 304 } 305 TEST_ASSERT(i == 28); 306 307 delete index; 308 index = new AlphabeticIndex(Locale::createFromName("ru"), status); 309 TEST_CHECK_STATUS; 310 assertEquals("Russian index.getBucketCount()", 32, index->getBucketCount(status)); 311 // Latin-script names should go into the underflow label (0) 312 // if the Russian collation does not use script reordering, 313 // but into the overflow label (getBucketCount()-1) 314 // if Russian sorts Cyrillic first. 315 int32_t reorderCodes[20]; 316 int32_t expectedLatinIndex = 0; 317 if (index->getCollator().getReorderCodes(reorderCodes, LENGTHOF(reorderCodes), status) > 0) { 318 expectedLatinIndex = index->getBucketCount(status) - 1; 319 } 320 n = index->getBucketIndex(adam, status); 321 TEST_CHECK_STATUS; 322 assertEquals("Russian index.getBucketIndex(adam)", expectedLatinIndex, n); 323 n = index->getBucketIndex(baker, status); 324 assertEquals("Russian index.getBucketIndex(baker)", expectedLatinIndex, n); 325 n = index->getBucketIndex(Cyrillic, status); 326 assertEquals("Russian index.getBucketIndex(Cyrillic)", 1, n); 327 n = index->getBucketIndex(zed, status); 328 assertEquals("Russian index.getBucketIndex(zed)", expectedLatinIndex, n); 329 330 delete index; 331 332 } 333 334 335 static const char * KEY_LOCALES[] = { 336 "en", "es", "de", "fr", "ja", "it", "tr", "pt", "zh", "nl", 337 "pl", "ar", "ru", "zh_Hant", "ko", "th", "sv", "fi", "da", 338 "he", "nb", "el", "hr", "bg", "sk", "lt", "vi", "lv", "sr", 339 "pt_PT", "ro", "hu", "cs", "id", "sl", "fil", "fa", "uk", 340 "ca", "hi", "et", "eu", "is", "sw", "ms", "bn", "am", "ta", 341 "te", "mr", "ur", "ml", "kn", "gu", "or", ""}; 342 343 344 void AlphabeticIndexTest::ManyLocalesTest() { 345 UErrorCode status = U_ZERO_ERROR; 346 int32_t lc = 0; 347 348 for (int i=0; ; ++i) { 349 status = U_ZERO_ERROR; 350 const char *localeName = KEY_LOCALES[i]; 351 if (localeName[0] == 0) { 352 break; 353 } 354 // std::cout << localeName << " "; 355 Locale loc = Locale::createFromName(localeName); 356 AlphabeticIndex index(loc, status); 357 TEST_CHECK_STATUS; 358 lc = index.getBucketCount(status); 359 TEST_CHECK_STATUS; 360 // std::cout << "getBucketCount() == " << lc << std::endl; 361 362 LocalPointer<AlphabeticIndex::ImmutableIndex> immIndex(index.buildImmutableIndex(status)); 363 TEST_CHECK_STATUS; 364 TEST_ASSERT(lc == immIndex->getBucketCount()); 365 366 assertEquals("initial bucket index", -1, index.getBucketIndex()); 367 int32_t bucketIndex = 0; 368 while (index.nextBucket(status)) { 369 TEST_CHECK_STATUS; 370 assertEquals("bucket index", bucketIndex, index.getBucketIndex()); 371 const UnicodeString &label = index.getBucketLabel(); 372 TEST_ASSERT(label.length()>0); 373 // std::string ss; 374 // std::cout << ":" << label.toUTF8String(ss); 375 const AlphabeticIndex::Bucket *bucket = immIndex->getBucket(bucketIndex); 376 TEST_ASSERT(bucket != NULL); 377 assertEquals("bucket label vs. immutable: locale=" + UnicodeString(localeName) + 378 " index=" + bucketIndex, 379 label, bucket->getLabel()); 380 TEST_ASSERT(&label != &bucket->getLabel()); // not the same pointers 381 UAlphabeticIndexLabelType labelType = index.getBucketLabelType(); 382 TEST_ASSERT(labelType == bucket->getLabelType()); 383 ++bucketIndex; 384 } 385 // std::cout << ":" << std::endl; 386 387 TEST_ASSERT(immIndex->getBucketCount() == bucketIndex); 388 TEST_ASSERT(immIndex->getBucket(-1) == NULL); 389 TEST_ASSERT(immIndex->getBucket(bucketIndex) == NULL); 390 } 391 } 392 393 394 // Test data for Pinyin based indexes. 395 // The Chinese characters should be distributed under latin labels in 396 // an index. 397 398 static const char *pinyinTestData[] = { 399 "\\u0101", "\\u5416", "\\u58ba", // 400 "b", "\\u516b", "\\u62d4", "\\u8500", // 401 "c", "\\u5693", "\\u7938", "\\u9e7e", // 402 "d", "\\u5491", "\\u8fcf", "\\u964a", // 403 "\\u0113","\\u59b8", "\\u92e8", "\\u834b", // 404 "f", "\\u53d1", "\\u9197", "\\u99a5", // 405 "g", "\\u7324", "\\u91d3", "\\u8142", // 406 "h", "\\u598e", "\\u927f", "\\u593b", // 407 "j", "\\u4e0c", "\\u6785", "\\u9d58", // 408 "k", "\\u5494", "\\u958b", "\\u7a52", // 409 "l", "\\u5783", "\\u62c9", "\\u9ba5", // 410 "m", "\\u5638", "\\u9ebb", "\\u65c0", // 411 "n", "\\u62ff", "\\u80ad", "\\u685b", // 412 "\\u014D", "\\u5662", "\\u6bee", "\\u8bb4", // 413 "p", "\\u5991", "\\u8019", "\\u8c31", // 414 "q", "\\u4e03", "\\u6053", "\\u7f56", // 415 "r", "\\u5465", "\\u72aa", "\\u6e03", // 416 "s", "\\u4ee8", "\\u9491", "\\u93c1", // 417 "t", "\\u4ed6", "\\u9248", "\\u67dd", // 418 "w", "\\u5c72", "\\u5558", "\\u5a7a", // 419 "x", "\\u5915", "\\u5438", "\\u6bbe", // 420 "y", "\\u4e2b", "\\u82bd", "\\u8574", // 421 "z", "\\u5e00", "\\u707d", "\\u5c0a", 422 NULL 423 }; 424 425 void AlphabeticIndexTest::HackPinyinTest() { 426 UErrorCode status = U_ZERO_ERROR; 427 AlphabeticIndex aindex(Locale::createFromName("zh"), status); 428 TEST_CHECK_STATUS; 429 430 UnicodeString names[sizeof(pinyinTestData) / sizeof(pinyinTestData[0])]; 431 int32_t nameCount; 432 for (nameCount=0; pinyinTestData[nameCount] != NULL; nameCount++) { 433 names[nameCount] = UnicodeString(pinyinTestData[nameCount], -1, UnicodeString::kInvariant).unescape(); 434 aindex.addRecord(names[nameCount], &names[nameCount], status); 435 TEST_CHECK_STATUS; 436 if (U_FAILURE(status)) { 437 return; 438 } 439 } 440 TEST_ASSERT(nameCount == aindex.getRecordCount(status)); 441 442 // Weak checking: make sure that none of the Chinese names landed in the overflow bucket 443 // of the index, and that the names are distributed among several buckets. 444 // (Exact expected data would be subject to change with evolution of the collation rules.) 445 446 int32_t bucketCount = 0; 447 int32_t filledBucketCount = 0; 448 while (aindex.nextBucket(status)) { 449 bucketCount++; 450 UnicodeString label = aindex.getBucketLabel(); 451 // std::string s; 452 // std::cout << label.toUTF8String(s) << ": "; 453 454 UBool bucketHasContents = FALSE; 455 while (aindex.nextRecord(status)) { 456 bucketHasContents = TRUE; 457 UnicodeString name = aindex.getRecordName(); 458 if (aindex.getBucketLabelType() != U_ALPHAINDEX_NORMAL) { 459 errln("File %s, Line %d, Name \"\\u%x\" is in an under or overflow bucket.", 460 __FILE__, __LINE__, name.char32At(0)); 461 } 462 // s.clear(); 463 // std::cout << aindex.getRecordName().toUTF8String(s) << " "; 464 } 465 if (bucketHasContents) { 466 filledBucketCount++; 467 } 468 // std::cout << std::endl; 469 } 470 TEST_ASSERT(bucketCount > 25); 471 TEST_ASSERT(filledBucketCount > 15); 472 } 473 474 475 void AlphabeticIndexTest::TestBug9009() { 476 UErrorCode status = U_ZERO_ERROR; 477 Locale loc("root"); 478 AlphabeticIndex aindex(loc, status); 479 TEST_CHECK_STATUS; 480 aindex.nextBucket(status); // Crash here before bug was fixed. 481 TEST_CHECK_STATUS; 482 } 483 484 static const char *localeAndIndexCharactersLists[][2] = { 485 /* Arabic*/ {"ar", "\\u0627:\\u0628:\\u062A:\\u062B:\\u062C:\\u062D:\\u062E:\\u062F:\\u0630:\\u0631:\\u0632:\\u0633:\\u0634:\\u0635:\\u0636:\\u0637:\\u0638:\\u0639:\\u063A:\\u0641:\\u0642:\\u0643:\\u0644:\\u0645:\\u0646:\\u0647:\\u0648:\\u064A"}, 486 /* Bulgarian*/ {"bg", "\\u0410:\\u0411:\\u0412:\\u0413:\\u0414:\\u0415:\\u0416:\\u0417:\\u0418:\\u0419:\\u041A:\\u041B:\\u041C:\\u041D:\\u041E:\\u041F:\\u0420:\\u0421:\\u0422:\\u0423:\\u0424:\\u0425:\\u0426:\\u0427:\\u0428:\\u0429:\\u042E:\\u042F"}, 487 /* Catalan*/ {"ca", "A:B:C:D:E:F:G:H:I:J:K:L:M:N:O:P:Q:R:S:T:U:V:W:X:Y:Z"}, 488 /* Czech*/ {"cs", "A:B:C:\\u010C:D:E:F:G:H:CH:I:J:K:L:M:N:O:P:Q:R:\\u0158:S:\\u0160:T:U:V:W:X:Y:Z:\\u017D"}, 489 /* Danish*/ {"da", "A:B:C:D:E:F:G:H:I:J:K:L:M:N:O:P:Q:R:S:T:U:V:W:X:Y:Z:\\u00C6:\\u00D8:\\u00C5"}, 490 /* German*/ {"de", "A:B:C:D:E:F:G:H:I:J:K:L:M:N:O:P:Q:R:S:T:U:V:W:X:Y:Z"}, 491 /* Greek*/ {"el", "\\u0391:\\u0392:\\u0393:\\u0394:\\u0395:\\u0396:\\u0397:\\u0398:\\u0399:\\u039A:\\u039B:\\u039C:\\u039D:\\u039E:\\u039F:\\u03A0:\\u03A1:\\u03A3:\\u03A4:\\u03A5:\\u03A6:\\u03A7:\\u03A8:\\u03A9"}, 492 /* English*/ {"en", "A:B:C:D:E:F:G:H:I:J:K:L:M:N:O:P:Q:R:S:T:U:V:W:X:Y:Z"}, 493 /* Spanish*/ {"es", "A:B:C:D:E:F:G:H:I:J:K:L:M:N:\\u00D1:O:P:Q:R:S:T:U:V:W:X:Y:Z"}, 494 /* Estonian*/ {"et", "A:B:C:D:E:F:G:H:I:J:K:L:M:N:O:P:Q:R:S:\\u0160:Z:\\u017D:T:U:V:\\u00D5:\\u00C4:\\u00D6:\\u00DC:X:Y"}, 495 /* Finnish*/ {"fi", "A:B:C:D:E:F:G:H:I:J:K:L:M:N:O:P:Q:R:S:T:U:V:W:X:Y:Z:\\u00C5:\\u00C4:\\u00D6"}, 496 /* Filipino*/ {"fil", "A:B:C:D:E:F:G:H:I:J:K:L:M:N:O:P:Q:R:S:T:U:V:W:X:Y:Z"}, 497 /* French*/ {"fr", "A:B:C:D:E:F:G:H:I:J:K:L:M:N:O:P:Q:R:S:T:U:V:W:X:Y:Z"}, 498 /* Hebrew*/ {"he", "\\u05D0:\\u05D1:\\u05D2:\\u05D3:\\u05D4:\\u05D5:\\u05D6:\\u05D7:\\u05D8:\\u05D9:\\u05DB:\\u05DC:\\u05DE:\\u05E0:\\u05E1:\\u05E2:\\u05E4:\\u05E6:\\u05E7:\\u05E8:\\u05E9:\\u05EA"}, 499 /* Icelandic*/ {"is", "A:\\u00C1:B:C:D:\\u00D0:E:\\u00C9:F:G:H:I:\\u00CD:J:K:L:M:N:O:\\u00D3:P:Q:R:S:T:U:\\u00DA:V:W:X:Y:\\u00DD:Z:\\u00DE:\\u00C6:\\u00D6"}, 500 /* Italian*/ {"it", "A:B:C:D:E:F:G:H:I:J:K:L:M:N:O:P:Q:R:S:T:U:V:W:X:Y:Z"}, 501 /* Japanese*/ {"ja", "\\u3042:\\u304B:\\u3055:\\u305F:\\u306A:\\u306F:\\u307E:\\u3084:\\u3089:\\u308F"}, 502 /* Korean*/ {"ko", "\\u3131:\\u3134:\\u3137:\\u3139:\\u3141:\\u3142:\\u3145:\\u3147:\\u3148:\\u314A:\\u314B:\\u314C:\\u314D:\\u314E"}, 503 /* Lithuanian*/ {"lt", "A:B:C:\\u010C:D:E:F:G:H:I:J:K:L:M:N:O:P:R:S:\\u0160:T:U:V:Z:\\u017D"}, 504 /* Latvian*/ {"lv", "A:B:C:\\u010C:D:E:F:G:\\u0122:H:I:J:K:\\u0136:L:\\u013B:M:N:\\u0145:O:P:Q:R:S:\\u0160:T:U:V:W:X:Z:\\u017D"}, 505 /* Norwegian Bokm\\u00E5l*/ {"nb", "A:B:C:D:E:F:G:H:I:J:K:L:M:N:O:P:Q:R:S:T:U:V:W:X:Y:Z:\\u00C6:\\u00D8:\\u00C5"}, 506 /* Dutch*/ {"nl", "A:B:C:D:E:F:G:H:I:J:K:L:M:N:O:P:Q:R:S:T:U:V:W:X:Y:Z"}, 507 /* Polish*/ {"pl", "A:\\u0104:B:C:\\u0106:D:E:\\u0118:F:G:H:I:J:K:L:\\u0141:M:N:\\u0143:O:\\u00D3:P:Q:R:S:\\u015A:T:U:V:W:X:Y:Z:\\u0179:\\u017B"}, 508 /* Portuguese*/ {"pt", "A:B:C:D:E:F:G:H:I:J:K:L:M:N:O:P:Q:R:S:T:U:V:W:X:Y:Z"}, 509 /* Romanian*/ {"ro", "A:\\u0102:\\u00C2:B:C:D:E:F:G:H:I:\\u00CE:J:K:L:M:N:O:P:Q:R:S:\\u0218:T:\\u021A:U:V:W:X:Y:Z"}, 510 /* Russian*/ {"ru", "\\u0410:\\u0411:\\u0412:\\u0413:\\u0414:\\u0415:\\u0416:\\u0417:\\u0418:\\u0419:\\u041A:\\u041B:\\u041C:\\u041D:\\u041E:\\u041F:\\u0420:\\u0421:\\u0422:\\u0423:\\u0424:\\u0425:\\u0426:\\u0427:\\u0428:\\u0429:\\u042B:\\u042D:\\u042E:\\u042F"}, 511 /* Slovak*/ {"sk", "A:\\u00C4:B:C:\\u010C:D:E:F:G:H:CH:I:J:K:L:M:N:O:\\u00D4:P:Q:R:S:\\u0160:T:U:V:W:X:Y:Z:\\u017D"}, 512 /* Slovenian*/ {"sl", "A:B:C:\\u010C:\\u0106:D:\\u0110:E:F:G:H:I:J:K:L:M:N:O:P:Q:R:S:\\u0160:T:U:V:W:X:Y:Z:\\u017D"}, 513 /* Serbian*/ {"sr", "\\u0410:\\u0411:\\u0412:\\u0413:\\u0414:\\u0402:\\u0415:\\u0416:\\u0417:\\u0418:\\u0408:\\u041A:\\u041B:\\u0409:\\u041C:\\u041D:\\u040A:\\u041E:\\u041F:\\u0420:\\u0421:\\u0422:\\u040B:\\u0423:\\u0424:\\u0425:\\u0426:\\u0427:\\u040F:\\u0428"}, 514 /* Swedish*/ {"sv", "A:B:C:D:E:F:G:H:I:J:K:L:M:N:O:P:Q:R:S:T:U:V:W:X:Y:Z:\\u00C5:\\u00C4:\\u00D6"}, 515 /* Turkish*/ {"tr", "A:B:C:\\u00C7:D:E:F:G:H:I:\\u0130:J:K:L:M:N:O:\\u00D6:P:Q:R:S:\\u015E:T:U:\\u00DC:V:W:X:Y:Z"}, 516 /* Ukrainian*/ {"uk", "\\u0410:\\u0411:\\u0412:\\u0413:\\u0490:\\u0414:\\u0415:\\u0404:\\u0416:\\u0417:\\u0418:\\u0406:\\u0407:\\u0419:\\u041A:\\u041B:\\u041C:\\u041D:\\u041E:\\u041F:\\u0420:\\u0421:\\u0422:\\u0423:\\u0424:\\u0425:\\u0426:\\u0427:\\u0428:\\u0429:\\u042E:\\u042F"}, 517 /* Vietnamese*/ {"vi", "A:\\u0102:\\u00C2:B:C:D:\\u0110:E:\\u00CA:F:G:H:I:J:K:L:M:N:O:\\u00D4:\\u01A0:P:Q:R:S:T:U:\\u01AF:V:W:X:Y:Z"}, 518 /* Chinese*/ {"zh", "A:B:C:D:E:F:G:H:I:J:K:L:M:N:O:P:Q:R:S:T:U:V:W:X:Y:Z"}, 519 /* Chinese (Traditional Han)*/ {"zh_Hant", "1\\u5283:2\\u5283:3\\u5283:4\\u5283:5\\u5283:6\\u5283:7\\u5283:8\\u5283:9\\u5283:10\\u5283:11\\u5283:12\\u5283:13\\u5283:14\\u5283:15\\u5283:16\\u5283:17\\u5283:18\\u5283:19\\u5283:20\\u5283:21\\u5283:22\\u5283:23\\u5283:24\\u5283:25\\u5283:26\\u5283:27\\u5283:28\\u5283:29\\u5283:30\\u5283:31\\u5283:32\\u5283:33\\u5283:35\\u5283:36\\u5283:39\\u5283:48\\u5283"}, 520 521 // As of ICU 52, ICU does not have collation data for the following language. 522 // Therefore, constructing an AlphabeticIndex for it 523 // ends up with a collator for the default locale 524 // which makes the test unreliable. (see ticket #10277) 525 // It exposes a bigger problem in that it may not be desirable for collation 526 // to fall back to the default locale. 527 528 // /* Basque*/ {"eu", "A:B:C:D:E:F:G:H:I:J:K:L:M:N:O:P:Q:R:S:T:U:V:W:X:Y:Z"}, 529 }; 530 531 void AlphabeticIndexTest::TestIndexCharactersList() { 532 UErrorCode status = U_ZERO_ERROR; 533 for (int32_t i = 0; i < LENGTHOF(localeAndIndexCharactersLists); ++i) { 534 const char *(&localeAndIndexCharacters)[2] = localeAndIndexCharactersLists[i]; 535 const char *locale = localeAndIndexCharacters[0]; 536 UnicodeString expectedIndexCharacters 537 = (UnicodeString("\\u2026:") + localeAndIndexCharacters[1] + ":\\u2026").unescape(); 538 AlphabeticIndex index(locale, status); 539 TEST_CHECK_STATUS; 540 LocalPointer<AlphabeticIndex::ImmutableIndex> immIndex(index.buildImmutableIndex(status)); 541 TEST_CHECK_STATUS; 542 543 // Join the elements of the list to a string with delimiter ":" 544 UnicodeString actualIndexCharacters; 545 assertEquals(locale, 546 expectedIndexCharacters, 547 joinLabelsAndAppend(*immIndex, actualIndexCharacters)); 548 logln(locale + UnicodeString(": ") + actualIndexCharacters); 549 } 550 } 551 552 void AlphabeticIndexTest::TestHaniFirst() { 553 UErrorCode status = U_ZERO_ERROR; 554 LocalPointer<RuleBasedCollator> coll( 555 static_cast<RuleBasedCollator *>(Collator::createInstance(Locale::getRoot(), status))); 556 557 if (U_FAILURE(status)) { 558 dataerrln("Failed Collator::createInstance call - %s", u_errorName(status)); 559 return; 560 } 561 int32_t reorderCodes[] = { USCRIPT_HAN }; 562 coll->setReorderCodes(reorderCodes, LENGTHOF(reorderCodes), status); 563 TEST_CHECK_STATUS; 564 AlphabeticIndex index(coll.orphan(), status); 565 TEST_CHECK_STATUS; 566 assertEquals("getBucketCount()", 1, index.getBucketCount(status)); // ... (underflow only) 567 index.addLabels(Locale::getEnglish(), status); 568 assertEquals("getBucketCount()", 28, index.getBucketCount(status)); // ... A-Z ... 569 int32_t bucketIndex = index.getBucketIndex(UnicodeString((UChar)0x897f), status); 570 assertEquals("getBucketIndex(U+897F)", 0, bucketIndex); // underflow bucket 571 bucketIndex = index.getBucketIndex("i", status); 572 assertEquals("getBucketIndex(i)", 9, bucketIndex); 573 bucketIndex = index.getBucketIndex(UnicodeString((UChar)0x03B1), status); 574 assertEquals("getBucketIndex(Greek alpha)", 27, bucketIndex); 575 // TODO: Test with an unassigned code point (not just U+FFFF) 576 // when unassigned code points are not in the Hani reordering group any more. 577 // String unassigned = UTF16.valueOf(0x50005); 578 bucketIndex = index.getBucketIndex(UnicodeString((UChar)0xFFFF), status); 579 assertEquals("getBucketIndex(U+FFFF)", 27, bucketIndex); 580 } 581 582 void AlphabeticIndexTest::TestPinyinFirst() { 583 UErrorCode status = U_ZERO_ERROR; 584 LocalPointer<RuleBasedCollator> coll( 585 static_cast<RuleBasedCollator *>(Collator::createInstance(Locale::getChinese(), status))); 586 if (U_FAILURE(status)) { 587 dataerrln("Failed Collator::createInstance call - %s", u_errorName(status)); 588 return; 589 } 590 int32_t reorderCodes[] = { USCRIPT_HAN }; 591 coll->setReorderCodes(reorderCodes, LENGTHOF(reorderCodes), status); 592 TEST_CHECK_STATUS; 593 AlphabeticIndex index(coll.orphan(), status); 594 TEST_CHECK_STATUS; 595 assertEquals("getBucketCount()", 1, index.getBucketCount(status)); // ... (underflow only) 596 index.addLabels(Locale::getChinese(), status); 597 assertEquals("getBucketCount()", 28, index.getBucketCount(status)); // ... A-Z ... 598 int32_t bucketIndex = index.getBucketIndex(UnicodeString((UChar)0x897f), status); 599 assertEquals("getBucketIndex(U+897F)", (int32_t)((UChar)0x0058/*X*/ - (UChar)0x0041/*A*/ + 1), (int32_t)bucketIndex); 600 bucketIndex = index.getBucketIndex("i", status); 601 assertEquals("getBucketIndex(i)", 9, bucketIndex); 602 bucketIndex = index.getBucketIndex(UnicodeString((UChar)0x03B1), status); 603 assertEquals("getBucketIndex(Greek alpha)", (int32_t)27, bucketIndex); 604 // TODO: Test with an unassigned code point (not just U+FFFF) 605 // when unassigned code points are not in the Hani reordering group any more. 606 // String unassigned = UTF16.valueOf(0x50005); 607 bucketIndex = index.getBucketIndex(UnicodeString((UChar)0xFFFF), status); 608 assertEquals("getBucketIndex(U+FFFF)", 27, bucketIndex); 609 } 610 611 void AlphabeticIndexTest::TestSchSt() { 612 UErrorCode status = U_ZERO_ERROR; 613 AlphabeticIndex index(Locale::getGerman(), status); 614 index.addLabels(UnicodeSet("[\\u00C6{Sch*}{St*}]", status), status); 615 TEST_CHECK_STATUS; 616 // ... A AE-ligature B-R S Sch St T-Z ... 617 LocalPointer<AlphabeticIndex::ImmutableIndex> immIndex(index.buildImmutableIndex(status)); 618 TEST_CHECK_STATUS; 619 assertEquals("getBucketCount()", 31, index.getBucketCount(status)); 620 assertEquals("immutable getBucketCount()", 31, immIndex->getBucketCount()); 621 static const struct TestCase { 622 const char *name; 623 int32_t bucketIndex; 624 const char *bucketLabel; 625 } testCases[] = { 626 // name, bucket index, bucket label 627 { "Adelbert", 1, "A" }, 628 { "Afrika", 1, "A" }, 629 { "\\u00C6sculap", 2, "\\u00C6" }, 630 { "Aesthet", 2, "\\u00C6" }, 631 { "Berlin", 3, "B" }, 632 { "Rilke", 19, "R" }, 633 { "Sacher", 20, "S" }, 634 { "Seiler", 20, "S" }, 635 { "Sultan", 20, "S" }, 636 { "Schiller", 21, "Sch" }, 637 { "Steiff", 22, "St" }, 638 { "Thomas", 23, "T" } 639 }; 640 for (int32_t i = 0; i < LENGTHOF(testCases); ++i) { 641 const TestCase &testCase = testCases[i]; 642 UnicodeString name = UnicodeString(testCase.name).unescape(); 643 UnicodeString label = UnicodeString(testCase.bucketLabel).unescape(); 644 char msg[100]; 645 sprintf(msg, "getBucketIndex(%s)", testCase.name); 646 assertEquals(msg, testCase.bucketIndex, index.getBucketIndex(name, status)); 647 sprintf(msg, "immutable getBucketIndex(%s)", testCase.name); 648 assertEquals(msg, testCase.bucketIndex, immIndex->getBucketIndex(name, status)); 649 sprintf(msg, "immutable bucket label (%s)", testCase.name); 650 assertEquals(msg, label, immIndex->getBucket(testCase.bucketIndex)->getLabel()); 651 } 652 } 653 654 void AlphabeticIndexTest::TestNoLabels() { 655 UErrorCode status = U_ZERO_ERROR; 656 LocalPointer<RuleBasedCollator> coll( 657 static_cast<RuleBasedCollator *>(Collator::createInstance(Locale::getRoot(), status))); 658 TEST_CHECK_STATUS; 659 AlphabeticIndex index(coll.orphan(), status); 660 TEST_CHECK_STATUS; 661 index.addRecord(UnicodeString((UChar)0x897f), NULL, status); 662 index.addRecord("i", NULL, status); 663 index.addRecord(UnicodeString((UChar)0x03B1), NULL, status); 664 assertEquals("getBucketCount()", 1, index.getBucketCount(status)); // ... 665 TEST_ASSERT(index.nextBucket(status)); 666 assertEquals("underflow label type", (int32_t)U_ALPHAINDEX_UNDERFLOW, index.getBucketLabelType()); 667 assertEquals("all records in the underflow bucket", (int32_t)3, index.getBucketRecordCount()); 668 } 669 670 void AlphabeticIndexTest::TestChineseZhuyin() { 671 UErrorCode status = U_ZERO_ERROR; 672 char loc[100]; 673 uloc_forLanguageTag("zh-u-co-zhuyin", loc, LENGTHOF(loc), NULL, &status); 674 AlphabeticIndex index(loc, status); 675 LocalPointer<AlphabeticIndex::ImmutableIndex> immIndex(index.buildImmutableIndex(status)); 676 TEST_CHECK_STATUS; 677 assertEquals("getBucketCount()", 38, immIndex->getBucketCount()); 678 assertEquals("label 1", UnicodeString((UChar)0x3105), immIndex->getBucket(1)->getLabel()); 679 assertEquals("label 2", UnicodeString((UChar)0x3106), immIndex->getBucket(2)->getLabel()); 680 assertEquals("label 3", UnicodeString((UChar)0x3107), immIndex->getBucket(3)->getLabel()); 681 assertEquals("label 4", UnicodeString((UChar)0x3108), immIndex->getBucket(4)->getLabel()); 682 assertEquals("label 5", UnicodeString((UChar)0x3109), immIndex->getBucket(5)->getLabel()); 683 } 684 685 #endif 686
