1 /******************************************************************** 2 * COPYRIGHT: 3 * Copyright (c) 2012-2014, International Business Machines Corporation 4 * and others. All Rights Reserved. 5 ********************************************************************/ 6 // 7 // file: alphaindextst.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 #include "cmemory.h" 17 18 #include "unicode/alphaindex.h" 19 #include "unicode/coll.h" 20 #include "unicode/localpointer.h" 21 #include "unicode/tblcoll.h" 22 #include "unicode/uniset.h" 23 24 #if !UCONFIG_NO_COLLATION && !UCONFIG_NO_NORMALIZATION 25 26 // #include <string> 27 // #include <iostream> 28 29 namespace { 30 31 UnicodeString joinLabelsAndAppend(AlphabeticIndex::ImmutableIndex &index, UnicodeString &dest) { 32 int32_t oldLength = dest.length(); 33 const AlphabeticIndex::Bucket *bucket; 34 for (int32_t i = 0; (bucket = index.getBucket(i)) != NULL; ++i) { 35 if (dest.length() > oldLength) { 36 dest.append((UChar)0x3A); // ':' 37 } 38 dest.append(bucket->getLabel()); 39 } 40 return dest; 41 } 42 43 } // namespace 44 45 AlphabeticIndexTest::AlphabeticIndexTest() { 46 } 47 48 AlphabeticIndexTest::~AlphabeticIndexTest() { 49 } 50 51 void AlphabeticIndexTest::runIndexedTest( int32_t index, UBool exec, const char* &name, char* /*par*/ ) 52 { 53 if (exec) logln("TestSuite AlphabeticIndex: "); 54 TESTCASE_AUTO_BEGIN; 55 TESTCASE_AUTO(APITest); 56 TESTCASE_AUTO(ManyLocalesTest); 57 TESTCASE_AUTO(HackPinyinTest); 58 TESTCASE_AUTO(TestBug9009); 59 TESTCASE_AUTO(TestIndexCharactersList); 60 TESTCASE_AUTO(TestHaniFirst); 61 TESTCASE_AUTO(TestPinyinFirst); 62 TESTCASE_AUTO(TestSchSt); 63 TESTCASE_AUTO(TestNoLabels); 64 TESTCASE_AUTO(TestChineseZhuyin); 65 TESTCASE_AUTO(TestJapaneseKanji); 66 TESTCASE_AUTO(TestChineseUnihan); 67 TESTCASE_AUTO_END; 68 } 69 70 #define TEST_CHECK_STATUS {if (U_FAILURE(status)) {dataerrln("%s:%d: Test failure. status=%s", \ 71 __FILE__, __LINE__, u_errorName(status)); return;}} 72 73 #define TEST_ASSERT(expr) {if ((expr)==FALSE) {errln("%s:%d: Test failure \n", __FILE__, __LINE__);};} 74 75 // 76 // APITest. Invoke every function at least once, and check that it does something. 77 // Does not attempt to check complete functionality. 78 // 79 void AlphabeticIndexTest::APITest() { 80 // 81 // Simple constructor and destructor, getBucketCount() 82 // 83 UErrorCode status = U_ZERO_ERROR; 84 int32_t lc = 0; 85 int32_t i = 0; 86 AlphabeticIndex *index = new AlphabeticIndex(Locale::getEnglish(), status); 87 TEST_CHECK_STATUS; 88 lc = index->getBucketCount(status); 89 TEST_CHECK_STATUS; 90 TEST_ASSERT(28 == lc); // 26 letters plus two under/overflow labels. 91 //printf("getBucketCount() == %d\n", lc); 92 delete index; 93 94 // Constructor from a Collator 95 // 96 status = U_ZERO_ERROR; 97 RuleBasedCollator *coll = dynamic_cast<RuleBasedCollator *>( 98 Collator::createInstance(Locale::getGerman(), status)); 99 TEST_CHECK_STATUS; 100 TEST_ASSERT(coll != NULL); 101 index = new AlphabeticIndex(coll, status); 102 TEST_CHECK_STATUS; 103 TEST_ASSERT(coll == &index->getCollator()); 104 assertEquals("only the underflow label in an index built from a collator", 105 1, index->getBucketCount(status)); 106 TEST_CHECK_STATUS; 107 delete index; 108 109 110 // addLabels() 111 112 status = U_ZERO_ERROR; 113 index = new AlphabeticIndex(Locale::getEnglish(), status); 114 TEST_CHECK_STATUS; 115 UnicodeSet additions; 116 additions.add((UChar32)0x410).add((UChar32)0x415); // A couple of Cyrillic letters 117 index->addLabels(additions, status); 118 TEST_CHECK_STATUS; 119 lc = index->getBucketCount(status); 120 TEST_CHECK_STATUS; 121 assertEquals("underflow, A-Z, inflow, 2 Cyrillic, overflow", 122 31, index->getBucketCount(status)); 123 // std::cout << lc << std::endl; 124 delete index; 125 126 127 // addLabels(Locale) 128 129 status = U_ZERO_ERROR; 130 index = new AlphabeticIndex(Locale::getEnglish(), status); 131 TEST_CHECK_STATUS; 132 AlphabeticIndex &aip = index->addLabels(Locale::getJapanese(), status); 133 TEST_ASSERT(&aip == index); 134 TEST_CHECK_STATUS; 135 lc = index->getBucketCount(status); 136 TEST_CHECK_STATUS; 137 TEST_ASSERT(35 < lc); // Japanese should add a bunch. Don't rely on the exact value. 138 delete index; 139 140 // GetCollator(), Get under/in/over flow labels 141 142 status = U_ZERO_ERROR; 143 index = new AlphabeticIndex(Locale::getGerman(), status); 144 TEST_CHECK_STATUS; 145 Collator *germanCol = Collator::createInstance(Locale::getGerman(), status); 146 TEST_CHECK_STATUS; 147 const RuleBasedCollator &indexCol = index->getCollator(); 148 TEST_ASSERT(*germanCol == indexCol); 149 delete germanCol; 150 151 UnicodeString ELLIPSIS; ELLIPSIS.append((UChar32)0x2026); 152 UnicodeString s = index->getUnderflowLabel(); 153 TEST_ASSERT(ELLIPSIS == s); 154 s = index->getOverflowLabel(); 155 TEST_ASSERT(ELLIPSIS == s); 156 s = index->getInflowLabel(); 157 TEST_ASSERT(ELLIPSIS == s); 158 index->setOverflowLabel(UNICODE_STRING_SIMPLE("O"), status); 159 index->setUnderflowLabel(UNICODE_STRING_SIMPLE("U"), status).setInflowLabel(UNICODE_STRING_SIMPLE("I"), status); 160 s = index->getUnderflowLabel(); 161 TEST_ASSERT(UNICODE_STRING_SIMPLE("U") == s); 162 s = index->getOverflowLabel(); 163 TEST_ASSERT(UNICODE_STRING_SIMPLE("O") == s); 164 s = index->getInflowLabel(); 165 TEST_ASSERT(UNICODE_STRING_SIMPLE("I") == s); 166 167 168 169 170 delete index; 171 172 173 174 const UnicodeString adam = UNICODE_STRING_SIMPLE("Adam"); 175 const UnicodeString baker = UNICODE_STRING_SIMPLE("Baker"); 176 const UnicodeString charlie = UNICODE_STRING_SIMPLE("Charlie"); 177 const UnicodeString chad = UNICODE_STRING_SIMPLE("Chad"); 178 const UnicodeString zed = UNICODE_STRING_SIMPLE("Zed"); 179 const UnicodeString Cyrillic = UNICODE_STRING_SIMPLE("\\u0410\\u0443\\u0435").unescape(); 180 181 // addRecord(), verify that it comes back out. 182 // 183 status = U_ZERO_ERROR; 184 index = new AlphabeticIndex(Locale::getEnglish(), status); 185 TEST_CHECK_STATUS; 186 index->addRecord(UnicodeString("Adam"), this, status); 187 UBool b; 188 TEST_CHECK_STATUS; 189 index->resetBucketIterator(status); 190 TEST_CHECK_STATUS; 191 index->nextBucket(status); // Move to underflow label 192 index->nextBucket(status); // Move to "A" 193 TEST_CHECK_STATUS; 194 const UnicodeString &label2 = index->getBucketLabel(); 195 UnicodeString A_STR = UNICODE_STRING_SIMPLE("A"); 196 TEST_ASSERT(A_STR == label2); 197 198 b = index->nextRecord(status); 199 TEST_CHECK_STATUS; 200 TEST_ASSERT(b); 201 const UnicodeString &itemName = index->getRecordName(); 202 TEST_ASSERT(adam == itemName); 203 204 const void *itemContext = index->getRecordData(); 205 TEST_ASSERT(itemContext == this); 206 207 delete index; 208 209 // clearRecords, addRecord(), Iteration 210 211 status = U_ZERO_ERROR; 212 index = new AlphabeticIndex(Locale::getEnglish(), status); 213 TEST_CHECK_STATUS; 214 while (index->nextBucket(status)) { 215 TEST_CHECK_STATUS; 216 while (index->nextRecord(status)) { 217 TEST_CHECK_STATUS; 218 TEST_ASSERT(FALSE); // No items have been added. 219 } 220 TEST_CHECK_STATUS; 221 } 222 223 index->addRecord(adam, NULL, status); 224 index->addRecord(baker, NULL, status); 225 index->addRecord(charlie, NULL, status); 226 index->addRecord(chad, NULL, status); 227 TEST_CHECK_STATUS; 228 int itemCount = 0; 229 index->resetBucketIterator(status); 230 while (index->nextBucket(status)) { 231 TEST_CHECK_STATUS; 232 while (index->nextRecord(status)) { 233 TEST_CHECK_STATUS; 234 ++itemCount; 235 } 236 } 237 TEST_CHECK_STATUS; 238 TEST_ASSERT(itemCount == 4); 239 240 TEST_ASSERT(index->nextBucket(status) == FALSE); 241 index->resetBucketIterator(status); 242 TEST_CHECK_STATUS; 243 TEST_ASSERT(index->nextBucket(status) == TRUE); 244 245 index->clearRecords(status); 246 TEST_CHECK_STATUS; 247 index->resetBucketIterator(status); 248 while (index->nextBucket(status)) { 249 TEST_CHECK_STATUS; 250 while (index->nextRecord(status)) { 251 TEST_ASSERT(FALSE); // No items have been added. 252 } 253 } 254 TEST_CHECK_STATUS; 255 delete index; 256 257 // getBucketLabel(), getBucketType() 258 259 status = U_ZERO_ERROR; 260 index = new AlphabeticIndex(Locale::getEnglish(), status); 261 TEST_CHECK_STATUS; 262 index->setUnderflowLabel(adam, status).setOverflowLabel(charlie, status); 263 TEST_CHECK_STATUS; 264 for (i=0; index->nextBucket(status); i++) { 265 TEST_CHECK_STATUS; 266 UnicodeString label = index->getBucketLabel(); 267 UAlphabeticIndexLabelType type = index->getBucketLabelType(); 268 if (i == 0) { 269 TEST_ASSERT(type == U_ALPHAINDEX_UNDERFLOW); 270 TEST_ASSERT(label == adam); 271 } else if (i <= 26) { 272 // Labels A - Z for English locale 273 TEST_ASSERT(type == U_ALPHAINDEX_NORMAL); 274 UnicodeString expectedLabel((UChar)(0x40 + i)); 275 TEST_ASSERT(expectedLabel == label); 276 } else if (i == 27) { 277 TEST_ASSERT(type == U_ALPHAINDEX_OVERFLOW); 278 TEST_ASSERT(label == charlie); 279 } else { 280 TEST_ASSERT(FALSE); 281 } 282 } 283 TEST_ASSERT(i==28); 284 delete index; 285 286 // getBucketIndex() 287 288 status = U_ZERO_ERROR; 289 index = new AlphabeticIndex(Locale::getEnglish(), status); 290 TEST_CHECK_STATUS; 291 int32_t n = index->getBucketIndex(adam, status); 292 TEST_CHECK_STATUS; 293 TEST_ASSERT(n == 1); /* Label #0 is underflow, 1 is A, etc. */ 294 n = index->getBucketIndex(baker, status); 295 TEST_ASSERT(n == 2); 296 n = index->getBucketIndex(Cyrillic, status); 297 TEST_ASSERT(n == 27); // Overflow label 298 n = index->getBucketIndex(zed, status); 299 TEST_ASSERT(n == 26); 300 301 for (i=0; index->nextBucket(status); i++) { 302 n = index->getBucketIndex(); 303 TEST_ASSERT(n == i); 304 UnicodeString label = index->getBucketLabel(); 305 TEST_ASSERT(n == i); 306 } 307 TEST_ASSERT(i == 28); 308 309 delete index; 310 index = new AlphabeticIndex(Locale::createFromName("ru"), status); 311 TEST_CHECK_STATUS; 312 assertEquals("Russian index.getBucketCount()", 32, index->getBucketCount(status)); 313 // Latin-script names should go into the underflow label (0) 314 // if the Russian collation does not use script reordering, 315 // but into the overflow label (getBucketCount()-1) 316 // if Russian sorts Cyrillic first. 317 int32_t reorderCodes[20]; 318 int32_t expectedLatinIndex = 0; 319 if (index->getCollator().getReorderCodes(reorderCodes, UPRV_LENGTHOF(reorderCodes), status) > 0) { 320 expectedLatinIndex = index->getBucketCount(status) - 1; 321 } 322 n = index->getBucketIndex(adam, status); 323 TEST_CHECK_STATUS; 324 assertEquals("Russian index.getBucketIndex(adam)", expectedLatinIndex, n); 325 n = index->getBucketIndex(baker, status); 326 assertEquals("Russian index.getBucketIndex(baker)", expectedLatinIndex, n); 327 n = index->getBucketIndex(Cyrillic, status); 328 assertEquals("Russian index.getBucketIndex(Cyrillic)", 1, n); 329 n = index->getBucketIndex(zed, status); 330 assertEquals("Russian index.getBucketIndex(zed)", expectedLatinIndex, n); 331 332 delete index; 333 334 } 335 336 337 static const char * KEY_LOCALES[] = { 338 "en", "es", "de", "fr", "ja", "it", "tr", "pt", "zh", "nl", 339 "pl", "ar", "ru", "zh_Hant", "ko", "th", "sv", "fi", "da", 340 "he", "nb", "el", "hr", "bg", "sk", "lt", "vi", "lv", "sr", 341 "pt_PT", "ro", "hu", "cs", "id", "sl", "fil", "fa", "uk", 342 "ca", "hi", "et", "eu", "is", "sw", "ms", "bn", "am", "ta", 343 "te", "mr", "ur", "ml", "kn", "gu", "or", ""}; 344 345 346 void AlphabeticIndexTest::ManyLocalesTest() { 347 UErrorCode status = U_ZERO_ERROR; 348 int32_t lc = 0; 349 350 for (int i=0; ; ++i) { 351 status = U_ZERO_ERROR; 352 const char *localeName = KEY_LOCALES[i]; 353 if (localeName[0] == 0) { 354 break; 355 } 356 // std::cout << localeName << " "; 357 Locale loc = Locale::createFromName(localeName); 358 AlphabeticIndex index(loc, status); 359 TEST_CHECK_STATUS; 360 lc = index.getBucketCount(status); 361 TEST_CHECK_STATUS; 362 // std::cout << "getBucketCount() == " << lc << std::endl; 363 364 LocalPointer<AlphabeticIndex::ImmutableIndex> immIndex(index.buildImmutableIndex(status)); 365 TEST_CHECK_STATUS; 366 TEST_ASSERT(lc == immIndex->getBucketCount()); 367 368 assertEquals("initial bucket index", -1, index.getBucketIndex()); 369 int32_t bucketIndex = 0; 370 while (index.nextBucket(status)) { 371 TEST_CHECK_STATUS; 372 assertEquals("bucket index", bucketIndex, index.getBucketIndex()); 373 const UnicodeString &label = index.getBucketLabel(); 374 TEST_ASSERT(label.length()>0); 375 // std::string ss; 376 // std::cout << ":" << label.toUTF8String(ss); 377 const AlphabeticIndex::Bucket *bucket = immIndex->getBucket(bucketIndex); 378 TEST_ASSERT(bucket != NULL); 379 assertEquals("bucket label vs. immutable: locale=" + UnicodeString(localeName) + 380 " index=" + bucketIndex, 381 label, bucket->getLabel()); 382 TEST_ASSERT(&label != &bucket->getLabel()); // not the same pointers 383 UAlphabeticIndexLabelType labelType = index.getBucketLabelType(); 384 TEST_ASSERT(labelType == bucket->getLabelType()); 385 ++bucketIndex; 386 } 387 // std::cout << ":" << std::endl; 388 389 TEST_ASSERT(immIndex->getBucketCount() == bucketIndex); 390 TEST_ASSERT(immIndex->getBucket(-1) == NULL); 391 TEST_ASSERT(immIndex->getBucket(bucketIndex) == NULL); 392 } 393 } 394 395 396 // Test data for Pinyin based indexes. 397 // The Chinese characters should be distributed under latin labels in 398 // an index. 399 400 static const char *pinyinTestData[] = { 401 "\\u0101", "\\u5416", "\\u58ba", // 402 "b", "\\u516b", "\\u62d4", "\\u8500", // 403 "c", "\\u5693", "\\u7938", "\\u9e7e", // 404 "d", "\\u5491", "\\u8fcf", "\\u964a", // 405 "\\u0113","\\u59b8", "\\u92e8", "\\u834b", // 406 "f", "\\u53d1", "\\u9197", "\\u99a5", // 407 "g", "\\u7324", "\\u91d3", "\\u8142", // 408 "h", "\\u598e", "\\u927f", "\\u593b", // 409 "j", "\\u4e0c", "\\u6785", "\\u9d58", // 410 "k", "\\u5494", "\\u958b", "\\u7a52", // 411 "l", "\\u5783", "\\u62c9", "\\u9ba5", // 412 "m", "\\u5638", "\\u9ebb", "\\u65c0", // 413 "n", "\\u62ff", "\\u80ad", "\\u685b", // 414 "\\u014D", "\\u5662", "\\u6bee", "\\u8bb4", // 415 "p", "\\u5991", "\\u8019", "\\u8c31", // 416 "q", "\\u4e03", "\\u6053", "\\u7f56", // 417 "r", "\\u5465", "\\u72aa", "\\u6e03", // 418 "s", "\\u4ee8", "\\u9491", "\\u93c1", // 419 "t", "\\u4ed6", "\\u9248", "\\u67dd", // 420 "w", "\\u5c72", "\\u5558", "\\u5a7a", // 421 "x", "\\u5915", "\\u5438", "\\u6bbe", // 422 "y", "\\u4e2b", "\\u82bd", "\\u8574", // 423 "z", "\\u5e00", "\\u707d", "\\u5c0a", 424 NULL 425 }; 426 427 void AlphabeticIndexTest::HackPinyinTest() { 428 UErrorCode status = U_ZERO_ERROR; 429 AlphabeticIndex aindex(Locale::createFromName("zh"), status); 430 TEST_CHECK_STATUS; 431 432 UnicodeString names[sizeof(pinyinTestData) / sizeof(pinyinTestData[0])]; 433 int32_t nameCount; 434 for (nameCount=0; pinyinTestData[nameCount] != NULL; nameCount++) { 435 names[nameCount] = UnicodeString(pinyinTestData[nameCount], -1, UnicodeString::kInvariant).unescape(); 436 aindex.addRecord(names[nameCount], &names[nameCount], status); 437 TEST_CHECK_STATUS; 438 if (U_FAILURE(status)) { 439 return; 440 } 441 } 442 TEST_ASSERT(nameCount == aindex.getRecordCount(status)); 443 444 // Weak checking: make sure that none of the Chinese names landed in the overflow bucket 445 // of the index, and that the names are distributed among several buckets. 446 // (Exact expected data would be subject to change with evolution of the collation rules.) 447 448 int32_t bucketCount = 0; 449 int32_t filledBucketCount = 0; 450 while (aindex.nextBucket(status)) { 451 bucketCount++; 452 UnicodeString label = aindex.getBucketLabel(); 453 // std::string s; 454 // std::cout << label.toUTF8String(s) << ": "; 455 456 UBool bucketHasContents = FALSE; 457 while (aindex.nextRecord(status)) { 458 bucketHasContents = TRUE; 459 UnicodeString name = aindex.getRecordName(); 460 if (aindex.getBucketLabelType() != U_ALPHAINDEX_NORMAL) { 461 errln("File %s, Line %d, Name \"\\u%x\" is in an under or overflow bucket.", 462 __FILE__, __LINE__, name.char32At(0)); 463 } 464 // s.clear(); 465 // std::cout << aindex.getRecordName().toUTF8String(s) << " "; 466 } 467 if (bucketHasContents) { 468 filledBucketCount++; 469 } 470 // std::cout << std::endl; 471 } 472 TEST_ASSERT(bucketCount > 25); 473 TEST_ASSERT(filledBucketCount > 15); 474 } 475 476 477 void AlphabeticIndexTest::TestBug9009() { 478 UErrorCode status = U_ZERO_ERROR; 479 Locale loc("root"); 480 AlphabeticIndex aindex(loc, status); 481 TEST_CHECK_STATUS; 482 aindex.nextBucket(status); // Crash here before bug was fixed. 483 TEST_CHECK_STATUS; 484 } 485 486 static const char *localeAndIndexCharactersLists[][2] = { 487 /* 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"}, 488 /* 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"}, 489 /* 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"}, 490 /* 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"}, 491 /* 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"}, 492 /* 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"}, 493 /* 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"}, 494 /* 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"}, 495 /* 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"}, 496 /* 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"}, 497 /* 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"}, 498 /* 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"}, 499 /* 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"}, 500 /* 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"}, 501 /* Hebrew*/ {"he", "\\u05D0:\\u05D1:\\u05D2:\\u05D3:\\u05D4:\\u05D5:\\u05D6:\\u05D7:\\u05D8:\\u05D9:\\u05DB:\\u05DC:\\u05DE:\\u05E0:\\u05E1:\\u05E2:\\u05E4:\\u05E6:\\u05E7:\\u05E8:\\u05E9:\\u05EA"}, 502 /* 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"}, 503 /* 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"}, 504 /* Japanese*/ {"ja", "\\u3042:\\u304B:\\u3055:\\u305F:\\u306A:\\u306F:\\u307E:\\u3084:\\u3089:\\u308F"}, 505 /* Korean*/ {"ko", "\\u3131:\\u3134:\\u3137:\\u3139:\\u3141:\\u3142:\\u3145:\\u3147:\\u3148:\\u314A:\\u314B:\\u314C:\\u314D:\\u314E"}, 506 /* 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"}, 507 /* 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"}, 508 /* 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"}, 509 /* 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"}, 510 /* 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"}, 511 /* 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"}, 512 /* 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"}, 513 /* 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"}, 514 /* 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"}, 515 /* 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"}, 516 /* 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"}, 517 /* 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"}, 518 /* 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"}, 519 /* 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"}, 520 /* 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"}, 521 /* 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"}, 522 /* 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"}, 523 }; 524 525 void AlphabeticIndexTest::TestIndexCharactersList() { 526 UErrorCode status = U_ZERO_ERROR; 527 for (int32_t i = 0; i < UPRV_LENGTHOF(localeAndIndexCharactersLists); ++i) { 528 const char *(&localeAndIndexCharacters)[2] = localeAndIndexCharactersLists[i]; 529 const char *locale = localeAndIndexCharacters[0]; 530 UnicodeString expectedIndexCharacters 531 = (UnicodeString("\\u2026:") + localeAndIndexCharacters[1] + ":\\u2026").unescape(); 532 AlphabeticIndex index(locale, status); 533 TEST_CHECK_STATUS; 534 LocalPointer<AlphabeticIndex::ImmutableIndex> immIndex(index.buildImmutableIndex(status)); 535 TEST_CHECK_STATUS; 536 537 // Join the elements of the list to a string with delimiter ":" 538 UnicodeString actualIndexCharacters; 539 assertEquals(locale, 540 expectedIndexCharacters, 541 joinLabelsAndAppend(*immIndex, actualIndexCharacters)); 542 logln(locale + UnicodeString(": ") + actualIndexCharacters); 543 } 544 } 545 546 void AlphabeticIndexTest::TestHaniFirst() { 547 UErrorCode status = U_ZERO_ERROR; 548 LocalPointer<RuleBasedCollator> coll( 549 static_cast<RuleBasedCollator *>(Collator::createInstance(Locale::getRoot(), status))); 550 551 if (U_FAILURE(status)) { 552 dataerrln("Failed Collator::createInstance call - %s", u_errorName(status)); 553 return; 554 } 555 int32_t reorderCodes[] = { USCRIPT_HAN }; 556 coll->setReorderCodes(reorderCodes, UPRV_LENGTHOF(reorderCodes), status); 557 TEST_CHECK_STATUS; 558 AlphabeticIndex index(coll.orphan(), status); 559 TEST_CHECK_STATUS; 560 assertEquals("getBucketCount()", 1, index.getBucketCount(status)); // ... (underflow only) 561 index.addLabels(Locale::getEnglish(), status); 562 assertEquals("getBucketCount()", 28, index.getBucketCount(status)); // ... A-Z ... 563 int32_t bucketIndex = index.getBucketIndex(UnicodeString((UChar)0x897f), status); 564 assertEquals("getBucketIndex(U+897F)", 0, bucketIndex); // underflow bucket 565 bucketIndex = index.getBucketIndex("i", status); 566 assertEquals("getBucketIndex(i)", 9, bucketIndex); 567 bucketIndex = index.getBucketIndex(UnicodeString((UChar)0x03B1), status); 568 assertEquals("getBucketIndex(Greek alpha)", 27, bucketIndex); 569 // U+50005 is an unassigned code point which sorts at the end, independent of the Hani group. 570 bucketIndex = index.getBucketIndex(UnicodeString(0x50005), status); 571 assertEquals("getBucketIndex(U+50005)", 27, bucketIndex); 572 bucketIndex = index.getBucketIndex(UnicodeString((UChar)0xFFFF), status); 573 assertEquals("getBucketIndex(U+FFFF)", 27, bucketIndex); 574 } 575 576 void AlphabeticIndexTest::TestPinyinFirst() { 577 UErrorCode status = U_ZERO_ERROR; 578 LocalPointer<RuleBasedCollator> coll( 579 static_cast<RuleBasedCollator *>(Collator::createInstance(Locale::getChinese(), status))); 580 if (U_FAILURE(status)) { 581 dataerrln("Failed Collator::createInstance call - %s", u_errorName(status)); 582 return; 583 } 584 int32_t reorderCodes[] = { USCRIPT_HAN }; 585 coll->setReorderCodes(reorderCodes, UPRV_LENGTHOF(reorderCodes), status); 586 TEST_CHECK_STATUS; 587 AlphabeticIndex index(coll.orphan(), status); 588 TEST_CHECK_STATUS; 589 assertEquals("getBucketCount()", 28, index.getBucketCount(status)); // ... A-Z ... 590 index.addLabels(Locale::getChinese(), status); 591 assertEquals("getBucketCount()", 28, index.getBucketCount(status)); // ... A-Z ... 592 int32_t bucketIndex = index.getBucketIndex(UnicodeString((UChar)0x897f), status); 593 assertEquals("getBucketIndex(U+897F)", (int32_t)((UChar)0x0058/*X*/ - (UChar)0x0041/*A*/ + 1), bucketIndex); 594 bucketIndex = index.getBucketIndex("i", status); 595 assertEquals("getBucketIndex(i)", 9, bucketIndex); 596 bucketIndex = index.getBucketIndex(UnicodeString((UChar)0x03B1), status); 597 assertEquals("getBucketIndex(Greek alpha)", (int32_t)27, bucketIndex); 598 // U+50005 is an unassigned code point which sorts at the end, independent of the Hani group. 599 bucketIndex = index.getBucketIndex(UnicodeString(0x50005), status); 600 assertEquals("getBucketIndex(U+50005)", 27, bucketIndex); 601 bucketIndex = index.getBucketIndex(UnicodeString((UChar)0xFFFF), status); 602 assertEquals("getBucketIndex(U+FFFF)", 27, bucketIndex); 603 } 604 605 void AlphabeticIndexTest::TestSchSt() { 606 UErrorCode status = U_ZERO_ERROR; 607 AlphabeticIndex index(Locale::getGerman(), status); 608 index.addLabels(UnicodeSet("[\\u00C6{Sch*}{St*}]", status), status); 609 TEST_CHECK_STATUS; 610 // ... A AE-ligature B-R S Sch St T-Z ... 611 LocalPointer<AlphabeticIndex::ImmutableIndex> immIndex(index.buildImmutableIndex(status)); 612 TEST_CHECK_STATUS; 613 assertEquals("getBucketCount()", 31, index.getBucketCount(status)); 614 assertEquals("immutable getBucketCount()", 31, immIndex->getBucketCount()); 615 static const struct TestCase { 616 const char *name; 617 int32_t bucketIndex; 618 const char *bucketLabel; 619 } testCases[] = { 620 // name, bucket index, bucket label 621 { "Adelbert", 1, "A" }, 622 { "Afrika", 1, "A" }, 623 { "\\u00C6sculap", 2, "\\u00C6" }, 624 { "Aesthet", 2, "\\u00C6" }, 625 { "Berlin", 3, "B" }, 626 { "Rilke", 19, "R" }, 627 { "Sacher", 20, "S" }, 628 { "Seiler", 20, "S" }, 629 { "Sultan", 20, "S" }, 630 { "Schiller", 21, "Sch" }, 631 { "Steiff", 22, "St" }, 632 { "Thomas", 23, "T" } 633 }; 634 for (int32_t i = 0; i < UPRV_LENGTHOF(testCases); ++i) { 635 const TestCase &testCase = testCases[i]; 636 UnicodeString name = UnicodeString(testCase.name).unescape(); 637 UnicodeString label = UnicodeString(testCase.bucketLabel).unescape(); 638 char msg[100]; 639 sprintf(msg, "getBucketIndex(%s)", testCase.name); 640 assertEquals(msg, testCase.bucketIndex, index.getBucketIndex(name, status)); 641 sprintf(msg, "immutable getBucketIndex(%s)", testCase.name); 642 assertEquals(msg, testCase.bucketIndex, immIndex->getBucketIndex(name, status)); 643 sprintf(msg, "immutable bucket label (%s)", testCase.name); 644 assertEquals(msg, label, immIndex->getBucket(testCase.bucketIndex)->getLabel()); 645 } 646 } 647 648 void AlphabeticIndexTest::TestNoLabels() { 649 UErrorCode status = U_ZERO_ERROR; 650 LocalPointer<RuleBasedCollator> coll( 651 static_cast<RuleBasedCollator *>(Collator::createInstance(Locale::getRoot(), status))); 652 TEST_CHECK_STATUS; 653 AlphabeticIndex index(coll.orphan(), status); 654 TEST_CHECK_STATUS; 655 index.addRecord(UnicodeString((UChar)0x897f), NULL, status); 656 index.addRecord("i", NULL, status); 657 index.addRecord(UnicodeString((UChar)0x03B1), NULL, status); 658 assertEquals("getBucketCount()", 1, index.getBucketCount(status)); // ... 659 TEST_ASSERT(index.nextBucket(status)); 660 assertEquals("underflow label type", (int32_t)U_ALPHAINDEX_UNDERFLOW, index.getBucketLabelType()); 661 assertEquals("all records in the underflow bucket", (int32_t)3, index.getBucketRecordCount()); 662 } 663 664 void AlphabeticIndexTest::TestChineseZhuyin() { 665 UErrorCode status = U_ZERO_ERROR; 666 char loc[100]; 667 uloc_forLanguageTag("zh-u-co-zhuyin", loc, UPRV_LENGTHOF(loc), NULL, &status); 668 AlphabeticIndex index(loc, status); 669 LocalPointer<AlphabeticIndex::ImmutableIndex> immIndex(index.buildImmutableIndex(status)); 670 TEST_CHECK_STATUS; 671 assertEquals("getBucketCount()", 38, immIndex->getBucketCount()); 672 assertEquals("label 1", UnicodeString((UChar)0x3105), immIndex->getBucket(1)->getLabel()); 673 assertEquals("label 2", UnicodeString((UChar)0x3106), immIndex->getBucket(2)->getLabel()); 674 assertEquals("label 3", UnicodeString((UChar)0x3107), immIndex->getBucket(3)->getLabel()); 675 assertEquals("label 4", UnicodeString((UChar)0x3108), immIndex->getBucket(4)->getLabel()); 676 assertEquals("label 5", UnicodeString((UChar)0x3109), immIndex->getBucket(5)->getLabel()); 677 } 678 679 void AlphabeticIndexTest::TestJapaneseKanji() { 680 UErrorCode status = U_ZERO_ERROR; 681 AlphabeticIndex index(Locale::getJapanese(), status); 682 LocalPointer<AlphabeticIndex::ImmutableIndex> immIndex(index.buildImmutableIndex(status)); 683 TEST_CHECK_STATUS; 684 // There are no index characters for Kanji in the Japanese standard collator. 685 // They should all go into the overflow bucket. 686 static const UChar32 kanji[] = { 0x4E9C, 0x95C7, 0x4E00, 0x58F1 }; 687 int32_t overflowIndex = immIndex->getBucketCount() - 1; 688 for(int32_t i = 0; i < UPRV_LENGTHOF(kanji); ++i) { 689 char msg[40]; 690 sprintf(msg, "kanji[%d]=U+%04lX in overflow bucket", (int)i, (long)kanji[i]); 691 assertEquals(msg, overflowIndex, immIndex->getBucketIndex(UnicodeString(kanji[i]), status)); 692 TEST_CHECK_STATUS; 693 } 694 } 695 696 void AlphabeticIndexTest::TestChineseUnihan() { 697 UErrorCode status = U_ZERO_ERROR; 698 AlphabeticIndex index("zh-u-co-unihan", status); 699 if(U_FAILURE(status)) { 700 dataerrln("unable create an AlphabeticIndex for Chinese/unihan: %s", u_errorName(status)); 701 return; 702 } 703 index.setMaxLabelCount(500, status); // ICU 54 default is 99. 704 LocalPointer<AlphabeticIndex::ImmutableIndex> immIndex(index.buildImmutableIndex(status)); 705 TEST_CHECK_STATUS; 706 int32_t bucketCount = immIndex->getBucketCount(); 707 if(bucketCount < 216) { 708 // There should be at least an underflow and overflow label, 709 // and one for each of 214 radicals, 710 // and maybe additional labels for simplified radicals. 711 dataerrln("too few buckets/labels for Chinese/unihan: %d (is zh/unihan data available?)", 712 bucketCount); 713 return; 714 } else { 715 logln("Chinese/unihan has %d buckets/labels", bucketCount); 716 } 717 // bucketIndex = radical number, adjusted for simplified radicals in lower buckets. 718 int32_t bucketIndex = index.getBucketIndex(UnicodeString((UChar)0x4e5d), status); 719 assertEquals("getBucketIndex(U+4E5D)", 5, bucketIndex); 720 bucketIndex = index.getBucketIndex(UnicodeString((UChar)0x7527), status); 721 assertEquals("getBucketIndex(U+7527)", 100, bucketIndex); 722 } 723 724 #endif 725
