1 // 2016 and later: Unicode, Inc. and others. 2 // License & terms of use: http://www.unicode.org/copyright.html 3 /* 4 ********************************************************************** 5 * Copyright (C) 2000-2016, International Business Machines 6 * Corporation and others. All Rights Reserved. 7 ********************************************************************** 8 * Date Name Description 9 * 05/23/00 aliu Creation. 10 ********************************************************************** 11 */ 12 13 #include "unicode/utypes.h" 14 15 #if !UCONFIG_NO_TRANSLITERATION 16 17 #include "unicode/translit.h" 18 #include "rbt.h" 19 #include "unicode/calendar.h" 20 #include "unicode/uniset.h" 21 #include "unicode/uchar.h" 22 #include "unicode/normlzr.h" 23 #include "unicode/uchar.h" 24 #include "unicode/parseerr.h" 25 #include "unicode/usetiter.h" 26 #include "unicode/putil.h" 27 #include "unicode/uversion.h" 28 #include "unicode/locid.h" 29 #include "unicode/ulocdata.h" 30 #include "unicode/utf8.h" 31 #include "unicode/utf16.h" 32 #include "putilimp.h" 33 #include "cmemory.h" 34 #include "transrt.h" 35 #include "testutil.h" 36 #include <string.h> 37 #include <stdio.h> 38 39 #define CASE(id,test) case id: \ 40 name = #test; \ 41 if (exec) { \ 42 logln(#test "---"); \ 43 logln((UnicodeString)""); \ 44 UDate t = uprv_getUTCtime(); \ 45 test(); \ 46 t = uprv_getUTCtime() - t; \ 47 logln((UnicodeString)#test " took " + t/U_MILLIS_PER_DAY + " seconds"); \ 48 } \ 49 break 50 51 #define EXHAUSTIVE(id,test) case id: \ 52 if(quick==FALSE){ \ 53 name = #test; \ 54 if (exec){ \ 55 logln(#test "---"); \ 56 logln((UnicodeString)""); \ 57 test(); \ 58 } \ 59 }else{ \ 60 name=""; \ 61 } \ 62 break 63 void 64 TransliteratorRoundTripTest::runIndexedTest(int32_t index, UBool exec, 65 const char* &name, char* /*par*/) { 66 switch (index) { 67 CASE(0, TestCyrillic); 68 // CASE(0,TestKana); 69 CASE(1,TestHiragana); 70 CASE(2,TestKatakana); 71 CASE(3,TestJamo); 72 CASE(4,TestHangul); 73 CASE(5,TestGreek); 74 CASE(6,TestGreekUNGEGN); 75 CASE(7,Testel); 76 CASE(8,TestDevanagariLatin); 77 CASE(9,TestInterIndic); 78 CASE(10, TestHebrew); 79 CASE(11, TestArabic); 80 CASE(12, TestHan); 81 default: name = ""; break; 82 } 83 } 84 85 86 //-------------------------------------------------------------------- 87 // TransliteratorPointer 88 //-------------------------------------------------------------------- 89 90 /** 91 * A transliterator pointer wrapper that deletes the contained 92 * pointer automatically when the wrapper goes out of scope. 93 * Sometimes called a "janitor" or "smart pointer". 94 */ 95 class TransliteratorPointer { 96 Transliterator* t; 97 // disallowed: 98 TransliteratorPointer(const TransliteratorPointer& rhs); 99 TransliteratorPointer& operator=(const TransliteratorPointer& rhs); 100 public: 101 TransliteratorPointer(Transliterator* adopted) { 102 t = adopted; 103 } 104 ~TransliteratorPointer() { 105 delete t; 106 } 107 inline Transliterator* operator->() { return t; } 108 inline operator const Transliterator*() const { return t; } 109 inline operator Transliterator*() { return t; } 110 }; 111 112 //-------------------------------------------------------------------- 113 // Legal 114 //-------------------------------------------------------------------- 115 116 class Legal { 117 public: 118 Legal() {} 119 virtual ~Legal() {} 120 virtual UBool is(const UnicodeString& /*sourceString*/) const {return TRUE;} 121 }; 122 123 class LegalJamo : public Legal { 124 // any initial must be followed by a medial (or initial) 125 // any medial must follow an initial (or medial) 126 // any final must follow a medial (or final) 127 public: 128 LegalJamo() {} 129 virtual ~LegalJamo() {} 130 virtual UBool is(const UnicodeString& sourceString) const; 131 int getType(UChar c) const; 132 }; 133 134 UBool LegalJamo::is(const UnicodeString& sourceString) const { 135 int t; 136 UnicodeString decomp; 137 UErrorCode ec = U_ZERO_ERROR; 138 Normalizer::decompose(sourceString, FALSE, 0, decomp, ec); 139 if (U_FAILURE(ec)) { 140 return FALSE; 141 } 142 for (int i = 0; i < decomp.length(); ++i) { // don't worry about surrogates 143 switch (getType(decomp.charAt(i))) { 144 case 0: t = getType(decomp.charAt(i+1)); 145 if (t != 0 && t != 1) { return FALSE; } 146 break; 147 case 1: t = getType(decomp.charAt(i-1)); 148 if (t != 0 && t != 1) { return FALSE; } 149 break; 150 case 2: t = getType(decomp.charAt(i-1)); 151 if (t != 1 && t != 2) { return FALSE; } 152 break; 153 } 154 } 155 return TRUE; 156 } 157 158 int LegalJamo::getType(UChar c) const { 159 if (0x1100 <= c && c <= 0x1112) 160 return 0; 161 else if (0x1161 <= c && c <= 0x1175) 162 return 1; 163 else if (0x11A8 <= c && c <= 0x11C2) 164 return 2; 165 return -1; // other 166 } 167 168 class LegalGreek : public Legal { 169 UBool full; 170 public: 171 LegalGreek(UBool _full) { full = _full; } 172 virtual ~LegalGreek() {} 173 174 virtual UBool is(const UnicodeString& sourceString) const; 175 176 static UBool isVowel(UChar c); 177 178 static UBool isRho(UChar c); 179 }; 180 181 UBool LegalGreek::is(const UnicodeString& sourceString) const { 182 UnicodeString decomp; 183 UErrorCode ec = U_ZERO_ERROR; 184 Normalizer::decompose(sourceString, FALSE, 0, decomp, ec); 185 186 // modern is simpler: don't care about anything but a grave 187 if (full == FALSE) { 188 // A special case which is legal but should be 189 // excluded from round trip 190 // if (sourceString == UnicodeString("\\u039C\\u03C0", "")) { 191 // return FALSE; 192 // } 193 for (int32_t i = 0; i < decomp.length(); ++i) { 194 UChar c = decomp.charAt(i); 195 // exclude all the accents 196 if (c == 0x0313 || c == 0x0314 || c == 0x0300 || c == 0x0302 197 || c == 0x0342 || c == 0x0345 198 ) return FALSE; 199 } 200 return TRUE; 201 } 202 203 // Legal greek has breathing marks IFF there is a vowel or RHO at the start 204 // IF it has them, it has exactly one. 205 // IF it starts with a RHO, then the breathing mark must come before the second letter. 206 // Since there are no surrogates in greek, don't worry about them 207 UBool firstIsVowel = FALSE; 208 UBool firstIsRho = FALSE; 209 UBool noLetterYet = TRUE; 210 int32_t breathingCount = 0; 211 int32_t letterCount = 0; 212 for (int32_t i = 0; i < decomp.length(); ++i) { 213 UChar c = decomp.charAt(i); 214 if (u_isalpha(c)) { 215 ++letterCount; 216 if (noLetterYet) { 217 noLetterYet = FALSE; 218 firstIsVowel = isVowel(c); 219 firstIsRho = isRho(c); 220 } 221 if (firstIsRho && letterCount == 2 && breathingCount == 0) { 222 return FALSE; 223 } 224 } 225 if (c == 0x0313 || c == 0x0314) { 226 ++breathingCount; 227 } 228 } 229 230 if (firstIsVowel || firstIsRho) return breathingCount == 1; 231 return breathingCount == 0; 232 } 233 234 UBool LegalGreek::isVowel(UChar c) { 235 switch (c) { 236 case 0x03B1: 237 case 0x03B5: 238 case 0x03B7: 239 case 0x03B9: 240 case 0x03BF: 241 case 0x03C5: 242 case 0x03C9: 243 case 0x0391: 244 case 0x0395: 245 case 0x0397: 246 case 0x0399: 247 case 0x039F: 248 case 0x03A5: 249 case 0x03A9: 250 return TRUE; 251 } 252 return FALSE; 253 } 254 255 UBool LegalGreek::isRho(UChar c) { 256 switch (c) { 257 case 0x03C1: 258 case 0x03A1: 259 return TRUE; 260 } 261 return FALSE; 262 } 263 264 // AbbreviatedUnicodeSetIterator Interface --------------------------------------------- 265 // 266 // Iterate over a UnicodeSet, only returning a sampling of the contained code points. 267 // density is the approximate total number of code points to returned for the entire set. 268 // 269 270 class AbbreviatedUnicodeSetIterator : public UnicodeSetIterator { 271 public : 272 273 AbbreviatedUnicodeSetIterator(); 274 virtual ~AbbreviatedUnicodeSetIterator(); 275 void reset(UnicodeSet& set, UBool abb = FALSE, int32_t density = 100); 276 277 /** 278 * ICU "poor man's RTTI", returns a UClassID for this class. 279 */ 280 static inline UClassID getStaticClassID() { return (UClassID)&fgClassID; } 281 282 /** 283 * ICU "poor man's RTTI", returns a UClassID for the actual class. 284 */ 285 virtual inline UClassID getDynamicClassID() const { return getStaticClassID(); } 286 287 private : 288 UBool abbreviated; 289 int32_t perRange; // The maximum number of code points to be returned from each range 290 virtual void loadRange(int32_t range); 291 292 /** 293 * The address of this static class variable serves as this class's ID 294 * for ICU "poor man's RTTI". 295 */ 296 static const char fgClassID; 297 }; 298 299 // AbbreviatedUnicodeSetIterator Implementation --------------------------------------- 300 301 const char AbbreviatedUnicodeSetIterator::fgClassID=0; 302 303 AbbreviatedUnicodeSetIterator::AbbreviatedUnicodeSetIterator() : 304 UnicodeSetIterator(), abbreviated(FALSE) { 305 } 306 307 AbbreviatedUnicodeSetIterator::~AbbreviatedUnicodeSetIterator() { 308 } 309 310 void AbbreviatedUnicodeSetIterator::reset(UnicodeSet& newSet, UBool abb, int32_t density) { 311 UnicodeSetIterator::reset(newSet); 312 abbreviated = abb; 313 perRange = newSet.getRangeCount(); 314 if (perRange != 0) { 315 perRange = density / perRange; 316 } 317 } 318 319 void AbbreviatedUnicodeSetIterator::loadRange(int32_t myRange) { 320 UnicodeSetIterator::loadRange(myRange); 321 if (abbreviated && (endElement > nextElement + perRange)) { 322 endElement = nextElement + perRange; 323 } 324 } 325 326 //-------------------------------------------------------------------- 327 // RTTest Interface 328 //-------------------------------------------------------------------- 329 330 class RTTest : public IntlTest { 331 332 // PrintWriter out; 333 334 UnicodeString transliteratorID; 335 int32_t errorLimit; 336 int32_t errorCount; 337 int32_t pairLimit; 338 UnicodeSet sourceRange; 339 UnicodeSet targetRange; 340 UnicodeSet toSource; 341 UnicodeSet toTarget; 342 UnicodeSet roundtripExclusionsSet; 343 IntlTest* parent; 344 Legal* legalSource; // NOT owned 345 UnicodeSet badCharacters; 346 347 public: 348 349 /* 350 * create a test for the given script transliterator. 351 */ 352 RTTest(const UnicodeString& transliteratorIDStr); 353 354 virtual ~RTTest(); 355 356 void setErrorLimit(int32_t limit); 357 358 void setPairLimit(int32_t limit); 359 360 void test(const UnicodeString& sourceRange, 361 const UnicodeString& targetRange, 362 const char* roundtripExclusions, 363 IntlTest* parent, 364 UBool quick, 365 Legal* adoptedLegal, 366 int32_t density = 100); 367 368 private: 369 370 // Added to do better equality check. 371 372 static UBool isSame(const UnicodeString& a, const UnicodeString& b); 373 374 static UBool isCamel(const UnicodeString& a); 375 376 UBool checkIrrelevants(Transliterator *t, const UnicodeString& irrelevants); 377 378 void test2(UBool quick, int32_t density); 379 380 void logWrongScript(const UnicodeString& label, 381 const UnicodeString& from, 382 const UnicodeString& to); 383 384 void logNotCanonical(const UnicodeString& label, 385 const UnicodeString& from, 386 const UnicodeString& to, 387 const UnicodeString& fromCan, 388 const UnicodeString& toCan); 389 390 void logFails(const UnicodeString& label); 391 392 void logToRulesFails(const UnicodeString& label, 393 const UnicodeString& from, 394 const UnicodeString& to, 395 const UnicodeString& toCan); 396 397 void logRoundTripFailure(const UnicodeString& from, 398 const UnicodeString& toID, 399 const UnicodeString& to, 400 const UnicodeString& backID, 401 const UnicodeString& back); 402 }; 403 404 //-------------------------------------------------------------------- 405 // RTTest Implementation 406 //-------------------------------------------------------------------- 407 408 /* 409 * create a test for the given script transliterator. 410 */ 411 RTTest::RTTest(const UnicodeString& transliteratorIDStr) { 412 transliteratorID = transliteratorIDStr; 413 errorLimit = 500; 414 errorCount = 0; 415 pairLimit = 0x10000; 416 } 417 418 RTTest::~RTTest() { 419 } 420 421 void RTTest::setErrorLimit(int32_t limit) { 422 errorLimit = limit; 423 } 424 425 void RTTest::setPairLimit(int32_t limit) { 426 pairLimit = limit; 427 } 428 429 UBool RTTest::isSame(const UnicodeString& a, const UnicodeString& b) { 430 if (a == b) return TRUE; 431 if (a.caseCompare(b, U_FOLD_CASE_DEFAULT)==0 && isCamel(a)) return TRUE; 432 UnicodeString aa, bb; 433 UErrorCode ec = U_ZERO_ERROR; 434 Normalizer::decompose(a, FALSE, 0, aa, ec); 435 Normalizer::decompose(b, FALSE, 0, bb, ec); 436 if (aa == bb) return TRUE; 437 if (aa.caseCompare(bb, U_FOLD_CASE_DEFAULT)==0 && isCamel(aa)) return TRUE; 438 return FALSE; 439 } 440 441 UBool RTTest::isCamel(const UnicodeString& a) { 442 // see if string is of the form aB; e.g. lower, then upper or title 443 UChar32 cp; 444 UBool haveLower = FALSE; 445 for (int32_t i = 0; i < a.length(); i += U16_LENGTH(cp)) { 446 cp = a.char32At(i); 447 int8_t t = u_charType(cp); 448 switch (t) { 449 case U_UPPERCASE_LETTER: 450 if (haveLower) return TRUE; 451 break; 452 case U_TITLECASE_LETTER: 453 if (haveLower) return TRUE; 454 // fall through, since second letter is lower. 455 U_FALLTHROUGH; 456 case U_LOWERCASE_LETTER: 457 haveLower = TRUE; 458 break; 459 } 460 } 461 return FALSE; 462 } 463 464 void RTTest::test(const UnicodeString& sourceRangeVal, 465 const UnicodeString& targetRangeVal, 466 const char* roundtripExclusions, 467 IntlTest* logVal, UBool quickRt, 468 Legal* adoptedLegal, 469 int32_t density) 470 { 471 472 UErrorCode status = U_ZERO_ERROR; 473 474 this->parent = logVal; 475 this->legalSource = adoptedLegal; 476 477 UnicodeSet neverOk("[:Other:]", status); 478 UnicodeSet okAnyway("[^[:Letter:]]", status); 479 480 if (U_FAILURE(status)) { 481 parent->dataerrln("FAIL: Initializing UnicodeSet with [:Other:] or [^[:Letter:]] - Error: %s", u_errorName(status)); 482 return; 483 } 484 485 this->sourceRange.clear(); 486 this->sourceRange.applyPattern(sourceRangeVal, status); 487 if (U_FAILURE(status)) { 488 parent->errln("FAIL: UnicodeSet::applyPattern(" + 489 sourceRangeVal + ")"); 490 return; 491 } 492 this->sourceRange.removeAll(neverOk); 493 494 this->targetRange.clear(); 495 this->targetRange.applyPattern(targetRangeVal, status); 496 if (U_FAILURE(status)) { 497 parent->errln("FAIL: UnicodeSet::applyPattern(" + 498 targetRangeVal + ")"); 499 return; 500 } 501 this->targetRange.removeAll(neverOk); 502 503 this->toSource.clear(); 504 this->toSource.applyPattern(sourceRangeVal, status); 505 if (U_FAILURE(status)) { 506 parent->errln("FAIL: UnicodeSet::applyPattern(" + 507 sourceRangeVal + ")"); 508 return; 509 } 510 this->toSource.addAll(okAnyway); 511 512 this->toTarget.clear(); 513 this->toTarget.applyPattern(targetRangeVal, status); 514 if (U_FAILURE(status)) { 515 parent->errln("FAIL: UnicodeSet::applyPattern(" + 516 targetRangeVal + ")"); 517 return; 518 } 519 this->toTarget.addAll(okAnyway); 520 521 this->roundtripExclusionsSet.clear(); 522 if (roundtripExclusions != NULL && strlen(roundtripExclusions) > 0) { 523 this->roundtripExclusionsSet.applyPattern(UnicodeString(roundtripExclusions, -1, US_INV), status); 524 if (U_FAILURE(status)) { 525 parent->errln("FAIL: UnicodeSet::applyPattern(%s)", roundtripExclusions); 526 return; 527 } 528 } 529 530 badCharacters.clear(); 531 badCharacters.applyPattern("[:Other:]", status); 532 if (U_FAILURE(status)) { 533 parent->errln("FAIL: UnicodeSet::applyPattern([:Other:])"); 534 return; 535 } 536 537 test2(quickRt, density); 538 539 if (errorCount > 0) { 540 char str[100]; 541 int32_t length = transliteratorID.extract(str, 100, NULL, status); 542 str[length] = 0; 543 parent->errln("FAIL: %s errors: %d %s", str, errorCount, (errorCount > errorLimit ? " (at least!)" : " ")); // + ", see " + logFileName); 544 } else { 545 char str[100]; 546 int32_t length = transliteratorID.extract(str, 100, NULL, status); 547 str[length] = 0; 548 parent->logln("%s ok", str); 549 } 550 } 551 552 UBool RTTest::checkIrrelevants(Transliterator *t, 553 const UnicodeString& irrelevants) { 554 for (int i = 0; i < irrelevants.length(); ++i) { 555 UChar c = irrelevants.charAt(i); 556 UnicodeString srcStr(c); 557 UnicodeString targ = srcStr; 558 t->transliterate(targ); 559 if (srcStr == targ) return TRUE; 560 } 561 return FALSE; 562 } 563 564 void RTTest::test2(UBool quickRt, int32_t density) { 565 566 UnicodeString srcStr, targ, reverse; 567 UErrorCode status = U_ZERO_ERROR; 568 UParseError parseError ; 569 TransliteratorPointer sourceToTarget( 570 Transliterator::createInstance(transliteratorID, UTRANS_FORWARD, parseError, 571 status)); 572 if ((Transliterator *)sourceToTarget == NULL) { 573 parent->dataerrln("FAIL: createInstance(" + transliteratorID + 574 ") returned NULL. Error: " + u_errorName(status) 575 + "\n\tpreContext : " + prettify(parseError.preContext) 576 + "\n\tpostContext : " + prettify(parseError.postContext)); 577 578 return; 579 } 580 TransliteratorPointer targetToSource(sourceToTarget->createInverse(status)); 581 if ((Transliterator *)targetToSource == NULL) { 582 parent->errln("FAIL: " + transliteratorID + 583 ".createInverse() returned NULL. Error:" + u_errorName(status) 584 + "\n\tpreContext : " + prettify(parseError.preContext) 585 + "\n\tpostContext : " + prettify(parseError.postContext)); 586 return; 587 } 588 589 AbbreviatedUnicodeSetIterator usi; 590 AbbreviatedUnicodeSetIterator usi2; 591 592 parent->logln("Checking that at least one irrelevant character is not NFC'ed"); 593 // string is from NFC_NO in the UCD 594 UnicodeString irrelevants = CharsToUnicodeString("\\u2000\\u2001\\u2126\\u212A\\u212B\\u2329"); 595 596 if (checkIrrelevants(sourceToTarget, irrelevants) == FALSE) { 597 logFails("Source-Target, irrelevants"); 598 } 599 if (checkIrrelevants(targetToSource, irrelevants) == FALSE) { 600 logFails("Target-Source, irrelevants"); 601 } 602 603 if (!quickRt){ 604 parent->logln("Checking that toRules works"); 605 UnicodeString rules = ""; 606 607 UParseError parseError; 608 rules = sourceToTarget->toRules(rules, TRUE); 609 // parent->logln((UnicodeString)"toRules => " + rules); 610 TransliteratorPointer sourceToTarget2(Transliterator::createFromRules( 611 "s2t2", rules, 612 UTRANS_FORWARD, 613 parseError, status)); 614 if (U_FAILURE(status)) { 615 parent->errln("FAIL: createFromRules %s\n", u_errorName(status)); 616 return; 617 } 618 619 rules = targetToSource->toRules(rules, FALSE); 620 TransliteratorPointer targetToSource2(Transliterator::createFromRules( 621 "t2s2", rules, 622 UTRANS_FORWARD, 623 parseError, status)); 624 if (U_FAILURE(status)) { 625 parent->errln("FAIL: createFromRules %s\n", u_errorName(status)); 626 return; 627 } 628 629 usi.reset(sourceRange); 630 for (;;) { 631 if (!usi.next() || usi.isString()) break; 632 UChar32 c = usi.getCodepoint(); 633 634 UnicodeString srcStr((UChar32)c); 635 UnicodeString targ = srcStr; 636 sourceToTarget->transliterate(targ); 637 UnicodeString targ2 = srcStr; 638 sourceToTarget2->transliterate(targ2); 639 if (targ != targ2) { 640 logToRulesFails("Source-Target, toRules", srcStr, targ, targ2); 641 } 642 } 643 644 usi.reset(targetRange); 645 for (;;) { 646 if (!usi.next() || usi.isString()) break; 647 UChar32 c = usi.getCodepoint(); 648 649 UnicodeString srcStr((UChar32)c); 650 UnicodeString targ = srcStr; 651 targetToSource->transliterate(targ); 652 UnicodeString targ2 = srcStr; 653 targetToSource2->transliterate(targ2); 654 if (targ != targ2) { 655 logToRulesFails("Target-Source, toRules", srcStr, targ, targ2); 656 } 657 } 658 } 659 660 parent->logln("Checking that all source characters convert to target - Singles"); 661 662 UnicodeSet failSourceTarg; 663 usi.reset(sourceRange); 664 for (;;) { 665 if (!usi.next() || usi.isString()) break; 666 UChar32 c = usi.getCodepoint(); 667 668 UnicodeString srcStr((UChar32)c); 669 UnicodeString targ = srcStr; 670 sourceToTarget->transliterate(targ); 671 if (toTarget.containsAll(targ) == FALSE 672 || badCharacters.containsSome(targ) == TRUE) { 673 UnicodeString targD; 674 Normalizer::decompose(targ, FALSE, 0, targD, status); 675 if (U_FAILURE(status)) { 676 parent->errln("FAIL: Internal error during decomposition %s\n", u_errorName(status)); 677 return; 678 } 679 if (toTarget.containsAll(targD) == FALSE || 680 badCharacters.containsSome(targD) == TRUE) { 681 logWrongScript("Source-Target", srcStr, targ); 682 failSourceTarg.add(c); 683 continue; 684 } 685 } 686 687 UnicodeString cs2; 688 Normalizer::decompose(srcStr, FALSE, 0, cs2, status); 689 if (U_FAILURE(status)) { 690 parent->errln("FAIL: Internal error during decomposition %s\n", u_errorName(status)); 691 return; 692 } 693 UnicodeString targ2 = cs2; 694 sourceToTarget->transliterate(targ2); 695 if (targ != targ2) { 696 logNotCanonical("Source-Target", srcStr, targ,cs2, targ2); 697 } 698 } 699 700 parent->logln("Checking that all source characters convert to target - Doubles"); 701 702 UnicodeSet sourceRangeMinusFailures(sourceRange); 703 sourceRangeMinusFailures.removeAll(failSourceTarg); 704 705 usi.reset(sourceRangeMinusFailures, quickRt, density); 706 for (;;) { 707 if (!usi.next() || usi.isString()) break; 708 UChar32 c = usi.getCodepoint(); 709 710 usi2.reset(sourceRangeMinusFailures, quickRt, density); 711 for (;;) { 712 if (!usi2.next() || usi2.isString()) break; 713 UChar32 d = usi2.getCodepoint(); 714 715 UnicodeString srcStr; 716 srcStr += (UChar32)c; 717 srcStr += (UChar32)d; 718 UnicodeString targ = srcStr; 719 sourceToTarget->transliterate(targ); 720 if (toTarget.containsAll(targ) == FALSE || 721 badCharacters.containsSome(targ) == TRUE) 722 { 723 UnicodeString targD; 724 Normalizer::decompose(targ, FALSE, 0, targD, status); 725 if (U_FAILURE(status)) { 726 parent->errln("FAIL: Internal error during decomposition %s\n", u_errorName(status)); 727 return; 728 } 729 if (toTarget.containsAll(targD) == FALSE || 730 badCharacters.containsSome(targD) == TRUE) { 731 logWrongScript("Source-Target", srcStr, targ); 732 continue; 733 } 734 } 735 UnicodeString cs2; 736 Normalizer::decompose(srcStr, FALSE, 0, cs2, status); 737 if (U_FAILURE(status)) { 738 parent->errln("FAIL: Internal error during decomposition %s\n", u_errorName(status)); 739 return; 740 } 741 UnicodeString targ2 = cs2; 742 sourceToTarget->transliterate(targ2); 743 if (targ != targ2) { 744 logNotCanonical("Source-Target", srcStr, targ, cs2,targ2); 745 } 746 } 747 } 748 749 parent->logln("Checking that target characters convert to source and back - Singles"); 750 751 UnicodeSet failTargSource; 752 UnicodeSet failRound; 753 754 usi.reset(targetRange); 755 for (;;) { 756 if (!usi.next()) break; 757 758 if(usi.isString()){ 759 srcStr = usi.getString(); 760 }else{ 761 srcStr = (UnicodeString)usi.getCodepoint(); 762 } 763 764 UChar32 c = srcStr.char32At(0); 765 766 targ = srcStr; 767 targetToSource->transliterate(targ); 768 reverse = targ; 769 sourceToTarget->transliterate(reverse); 770 771 if (toSource.containsAll(targ) == FALSE || 772 badCharacters.containsSome(targ) == TRUE) { 773 UnicodeString targD; 774 Normalizer::decompose(targ, FALSE, 0, targD, status); 775 if (U_FAILURE(status)) { 776 parent->errln("FAIL: Internal error during decomposition%s\n", u_errorName(status)); 777 return; 778 } 779 if (toSource.containsAll(targD) == FALSE) { 780 logWrongScript("Target-Source", srcStr, targ); 781 failTargSource.add(c); 782 continue; 783 } 784 if (badCharacters.containsSome(targD) == TRUE) { 785 logWrongScript("Target-Source*", srcStr, targ); 786 failTargSource.add(c); 787 continue; 788 } 789 } 790 if (isSame(srcStr, reverse) == FALSE && 791 roundtripExclusionsSet.contains(c) == FALSE 792 && roundtripExclusionsSet.contains(srcStr)==FALSE) { 793 logRoundTripFailure(srcStr,targetToSource->getID(), targ,sourceToTarget->getID(), reverse); 794 failRound.add(c); 795 continue; 796 } 797 798 UnicodeString targ2; 799 Normalizer::decompose(targ, FALSE, 0, targ2, status); 800 if (U_FAILURE(status)) { 801 parent->errln("FAIL: Internal error during decomposition%s\n", u_errorName(status)); 802 return; 803 } 804 UnicodeString reverse2 = targ2; 805 sourceToTarget->transliterate(reverse2); 806 if (reverse != reverse2) { 807 logNotCanonical("Target-Source", targ, reverse, targ2, reverse2); 808 } 809 } 810 811 parent->logln("Checking that target characters convert to source and back - Doubles"); 812 int32_t count = 0; 813 814 UnicodeSet targetRangeMinusFailures(targetRange); 815 targetRangeMinusFailures.removeAll(failTargSource); 816 targetRangeMinusFailures.removeAll(failRound); 817 818 usi.reset(targetRangeMinusFailures, quickRt, density); 819 UnicodeString targ2; 820 UnicodeString reverse2; 821 UnicodeString targD; 822 for (;;) { 823 if (!usi.next() || usi.isString()) break; 824 UChar32 c = usi.getCodepoint(); 825 if (++count > pairLimit) { 826 //throw new TestTruncated("Test truncated at " + pairLimit + " x 64k pairs"); 827 parent->logln(""); 828 parent->logln((UnicodeString)"Test truncated at " + pairLimit + " x 64k pairs"); 829 return; 830 } 831 832 usi2.reset(targetRangeMinusFailures, quickRt, density); 833 for (;;) { 834 if (!usi2.next() || usi2.isString()) 835 break; 836 UChar32 d = usi2.getCodepoint(); 837 srcStr.truncate(0); // empty the variable without construction/destruction 838 srcStr += c; 839 srcStr += d; 840 841 targ = srcStr; 842 targetToSource->transliterate(targ); 843 reverse = targ; 844 sourceToTarget->transliterate(reverse); 845 846 if (toSource.containsAll(targ) == FALSE || 847 badCharacters.containsSome(targ) == TRUE) 848 { 849 targD.truncate(0); // empty the variable without construction/destruction 850 Normalizer::decompose(targ, FALSE, 0, targD, status); 851 if (U_FAILURE(status)) { 852 parent->errln("FAIL: Internal error during decomposition%s\n", 853 u_errorName(status)); 854 return; 855 } 856 if (toSource.containsAll(targD) == FALSE 857 || badCharacters.containsSome(targD) == TRUE) 858 { 859 logWrongScript("Target-Source", srcStr, targ); 860 continue; 861 } 862 } 863 if (isSame(srcStr, reverse) == FALSE && 864 roundtripExclusionsSet.contains(c) == FALSE&& 865 roundtripExclusionsSet.contains(d) == FALSE && 866 roundtripExclusionsSet.contains(srcStr)== FALSE) 867 { 868 logRoundTripFailure(srcStr,targetToSource->getID(), targ, sourceToTarget->getID(),reverse); 869 continue; 870 } 871 872 targ2.truncate(0); // empty the variable without construction/destruction 873 Normalizer::decompose(targ, FALSE, 0, targ2, status); 874 if (U_FAILURE(status)) { 875 parent->errln("FAIL: Internal error during decomposition%s\n", u_errorName(status)); 876 return; 877 } 878 reverse2 = targ2; 879 sourceToTarget->transliterate(reverse2); 880 if (reverse != reverse2) { 881 logNotCanonical("Target-Source", targ,reverse, targ2, reverse2); 882 } 883 } 884 } 885 parent->logln(""); 886 } 887 888 void RTTest::logWrongScript(const UnicodeString& label, 889 const UnicodeString& from, 890 const UnicodeString& to) { 891 parent->errln((UnicodeString)"FAIL " + 892 label + ": " + 893 from + "(" + TestUtility::hex(from) + ") => " + 894 to + "(" + TestUtility::hex(to) + ")"); 895 ++errorCount; 896 } 897 898 void RTTest::logNotCanonical(const UnicodeString& label, 899 const UnicodeString& from, 900 const UnicodeString& to, 901 const UnicodeString& fromCan, 902 const UnicodeString& toCan) { 903 parent->errln((UnicodeString)"FAIL (can.equiv)" + 904 label + ": " + 905 from + "(" + TestUtility::hex(from) + ") => " + 906 to + "(" + TestUtility::hex(to) + ")" + 907 fromCan + "(" + TestUtility::hex(fromCan) + ") => " + 908 toCan + " (" + 909 TestUtility::hex(toCan) + ")" 910 ); 911 ++errorCount; 912 } 913 914 void RTTest::logFails(const UnicodeString& label) { 915 parent->errln((UnicodeString)"<br>FAIL " + label); 916 ++errorCount; 917 } 918 919 void RTTest::logToRulesFails(const UnicodeString& label, 920 const UnicodeString& from, 921 const UnicodeString& to, 922 const UnicodeString& otherTo) 923 { 924 parent->errln((UnicodeString)"FAIL: " + 925 label + ": " + 926 from + "(" + TestUtility::hex(from) + ") => " + 927 to + "(" + TestUtility::hex(to) + ")" + 928 "!=" + 929 otherTo + " (" + 930 TestUtility::hex(otherTo) + ")" 931 ); 932 ++errorCount; 933 } 934 935 936 void RTTest::logRoundTripFailure(const UnicodeString& from, 937 const UnicodeString& toID, 938 const UnicodeString& to, 939 const UnicodeString& backID, 940 const UnicodeString& back) { 941 if (legalSource->is(from) == FALSE) return; // skip illegals 942 943 parent->errln((UnicodeString)"FAIL Roundtrip: " + 944 from + "(" + TestUtility::hex(from) + ") => " + 945 to + "(" + TestUtility::hex(to) + ") "+toID+" => " + 946 back + "(" + TestUtility::hex(back) + ") "+backID+" => "); 947 ++errorCount; 948 } 949 950 //-------------------------------------------------------------------- 951 // Specific Tests 952 //-------------------------------------------------------------------- 953 954 /* 955 Note: Unicode 3.2 added new Hiragana/Katakana characters: 956 957 3095..3096 ; 3.2 # [2] HIRAGANA LETTER SMALL KA..HIRAGANA LETTER SMALL KE 958 309F..30A0 ; 3.2 # [2] HIRAGANA DIGRAPH YORI..KATAKANA-HIRAGANA DOUBLE HYPHEN 959 30FF ; 3.2 # KATAKANA DIGRAPH KOTO 960 31F0..31FF ; 3.2 # [16] KATAKANA LETTER SMALL KU..KATAKANA LETTER SMALL RO 961 962 Unicode 5.2 added another Hiragana character: 963 1F200 ; 5.2 # SQUARE HIRAGANA HOKA 964 965 We will not add them to the rules until they are more supported (e.g. in fonts on Windows) 966 A bug has been filed to remind us to do this: #1979. 967 */ 968 969 static const char KATAKANA[] = "[[[:katakana:][\\u30A1-\\u30FA\\u30FC]]-[\\u30FF\\u31F0-\\u31FF]-[:^age=5.2:]]"; 970 static const char HIRAGANA[] = "[[[:hiragana:][\\u3040-\\u3094]]-[\\u3095-\\u3096\\u309F-\\u30A0\\U0001F200-\\U0001F2FF]-[:^age=5.2:]]"; 971 static const char LENGTH[] = "[\\u30FC]"; 972 static const char HALFWIDTH_KATAKANA[] = "[\\uFF65-\\uFF9D]"; 973 static const char KATAKANA_ITERATION[] = "[\\u30FD\\u30FE]"; 974 static const char HIRAGANA_ITERATION[] = "[\\u309D\\u309E]"; 975 static const int32_t TEMP_MAX=256; 976 977 void TransliteratorRoundTripTest::TestKana() { 978 RTTest test("Katakana-Hiragana"); 979 Legal *legal = new Legal(); 980 char temp[TEMP_MAX]; 981 strcpy(temp, "["); 982 strcat(temp, HALFWIDTH_KATAKANA); 983 strcat(temp, LENGTH); 984 strcat(temp, "]"); 985 test.test(KATAKANA, UnicodeString("[") + HIRAGANA + LENGTH + UnicodeString("]"), 986 temp, 987 this, quick, legal); 988 delete legal; 989 } 990 991 void TransliteratorRoundTripTest::TestHiragana() { 992 RTTest test("Latin-Hiragana"); 993 Legal *legal = new Legal(); 994 test.test(UnicodeString("[a-zA-Z]", ""), 995 UnicodeString(HIRAGANA, -1, US_INV), 996 HIRAGANA_ITERATION, this, quick, legal); 997 delete legal; 998 } 999 1000 void TransliteratorRoundTripTest::TestKatakana() { 1001 RTTest test("Latin-Katakana"); 1002 Legal *legal = new Legal(); 1003 char temp[TEMP_MAX]; 1004 strcpy(temp, "["); 1005 strcat(temp, KATAKANA_ITERATION); 1006 strcat(temp, HALFWIDTH_KATAKANA); 1007 strcat(temp, "]"); 1008 test.test(UnicodeString("[a-zA-Z]", ""), 1009 UnicodeString(KATAKANA, -1, US_INV), 1010 temp, 1011 this, quick, legal); 1012 delete legal; 1013 } 1014 1015 void TransliteratorRoundTripTest::TestJamo() { 1016 RTTest t("Latin-Jamo"); 1017 Legal *legal = new LegalJamo(); 1018 t.test(UnicodeString("[a-zA-Z]", ""), 1019 UnicodeString("[\\u1100-\\u1112 \\u1161-\\u1175 \\u11A8-\\u11C2]", 1020 ""), 1021 NULL, this, quick, legal); 1022 delete legal; 1023 } 1024 1025 void TransliteratorRoundTripTest::TestHangul() { 1026 RTTest t("Latin-Hangul"); 1027 Legal *legal = new Legal(); 1028 if (quick) t.setPairLimit(1000); 1029 t.test(UnicodeString("[a-zA-Z]", ""), 1030 UnicodeString("[\\uAC00-\\uD7A4]", ""), 1031 NULL, this, quick, legal, 1); 1032 delete legal; 1033 } 1034 1035 1036 #define ASSERT_SUCCESS(status) {if (U_FAILURE(status)) { \ 1037 errcheckln(status, "error at file %s, line %d, status = %s", __FILE__, __LINE__, \ 1038 u_errorName(status)); \ 1039 return;}} 1040 1041 1042 static void writeStringInU8(FILE *out, const UnicodeString &s) { 1043 int i; 1044 for (i=0; i<s.length(); i=s.moveIndex32(i, 1)) { 1045 UChar32 c = s.char32At(i); 1046 uint8_t bufForOneChar[10]; 1047 UBool isError = FALSE; 1048 int32_t destIdx = 0; 1049 U8_APPEND(bufForOneChar, destIdx, (int32_t)sizeof(bufForOneChar), c, isError); 1050 fwrite(bufForOneChar, 1, destIdx, out); 1051 } 1052 } 1053 1054 1055 1056 1057 void TransliteratorRoundTripTest::TestHan() { 1058 UErrorCode status = U_ZERO_ERROR; 1059 LocalULocaleDataPointer uld(ulocdata_open("zh",&status)); 1060 LocalUSetPointer USetExemplars(ulocdata_getExemplarSet(uld.getAlias(), uset_openEmpty(), 0, ULOCDATA_ES_STANDARD, &status)); 1061 ASSERT_SUCCESS(status); 1062 1063 UnicodeString source; 1064 UChar32 c; 1065 int i; 1066 for (i=0; ;i++) { 1067 // Add all of the Chinese exemplar chars to the string "source". 1068 c = uset_charAt(USetExemplars.getAlias(), i); 1069 if (c == (UChar32)-1) { 1070 break; 1071 } 1072 source.append(c); 1073 } 1074 1075 // transform with Han translit 1076 Transliterator *hanTL = Transliterator::createInstance("Han-Latin", UTRANS_FORWARD, status); 1077 ASSERT_SUCCESS(status); 1078 UnicodeString target=source; 1079 hanTL->transliterate(target); 1080 // now verify that there are no Han characters left 1081 UnicodeSet allHan("[:han:]", status); 1082 ASSERT_SUCCESS(status); 1083 if (allHan.containsSome(target)) { 1084 errln("file %s, line %d, No Han must be left after Han-Latin transliteration", 1085 __FILE__, __LINE__); 1086 } 1087 1088 // check the pinyin translit 1089 Transliterator *pn = Transliterator::createInstance("Latin-NumericPinyin", UTRANS_FORWARD, status); 1090 ASSERT_SUCCESS(status); 1091 UnicodeString target2 = target; 1092 pn->transliterate(target2); 1093 1094 // verify that there are no marks 1095 Transliterator *nfd = Transliterator::createInstance("nfd", UTRANS_FORWARD, status); 1096 ASSERT_SUCCESS(status); 1097 1098 UnicodeString nfded = target2; 1099 nfd->transliterate(nfded); 1100 UnicodeSet allMarks(UNICODE_STRING_SIMPLE("[\\u0304\\u0301\\u030C\\u0300\\u0306]"), status); // look only for Pinyin tone marks, not all marks (there are some others in there) 1101 ASSERT_SUCCESS(status); 1102 assertFalse("NumericPinyin must contain no marks", allMarks.containsSome(nfded)); 1103 1104 // verify roundtrip 1105 Transliterator *np = pn->createInverse(status); 1106 ASSERT_SUCCESS(status); 1107 UnicodeString target3 = target2; 1108 np->transliterate(target3); 1109 UBool roundtripOK = (target3.compare(target) == 0); 1110 assertTrue("NumericPinyin must roundtrip", roundtripOK); 1111 if (!roundtripOK) { 1112 const char *filename = "numeric-pinyin.log.txt"; 1113 FILE *out = fopen(filename, "w"); 1114 errln("Creating log file %s\n", filename); 1115 fprintf(out, "Pinyin: "); 1116 writeStringInU8(out, target); 1117 fprintf(out, "\nPinyin-Numeric-Pinyin: "); 1118 writeStringInU8(out, target2); 1119 fprintf(out, "\nNumeric-Pinyin-Pinyin: "); 1120 writeStringInU8(out, target3); 1121 fprintf(out, "\n"); 1122 fclose(out); 1123 } 1124 1125 delete hanTL; 1126 delete pn; 1127 delete nfd; 1128 delete np; 1129 } 1130 1131 1132 void TransliteratorRoundTripTest::TestGreek() { 1133 logKnownIssue( "cldrbug:1911"); 1134 // It is left in its current state as a regression test. 1135 1136 RTTest test("Latin-Greek"); 1137 LegalGreek *legal = new LegalGreek(TRUE); 1138 1139 test.test(UnicodeString("[a-zA-Z]", ""), 1140 UnicodeString("[\\u003B\\u00B7[[:Greek:]&[:Letter:]]-[" 1141 "\\u1D26-\\u1D2A" // L& [5] GREEK LETTER SMALL CAPITAL GAMMA..GREEK LETTER SMALL CAPITAL PSI 1142 "\\u1D5D-\\u1D61" // Lm [5] MODIFIER LETTER SMALL BETA..MODIFIER LETTER SMALL CHI 1143 "\\u1D66-\\u1D6A" // L& [5] GREEK SUBSCRIPT SMALL LETTER BETA..GREEK SUBSCRIPT SMALL LETTER CHI 1144 "\\u03D7-\\u03EF" // \N{GREEK KAI SYMBOL}..\N{COPTIC SMALL LETTER DEI} 1145 "] & [:Age=4.0:]]", 1146 1147 //UnicodeString("[[\\u003B\\u00B7[:Greek:]-[\\u0374\\u0385\\u1fcd\\u1fce\\u1fdd\\u1fde\\u1fed-\\u1fef\\u1ffd\\u03D7-\\u03EF]]&[:Age=3.2:]]", 1148 ""), 1149 "[\\u00B5\\u037A\\u03D0-\\u03F5\\u03f9]", /* exclusions */ 1150 this, quick, legal, 50); 1151 1152 1153 delete legal; 1154 } 1155 1156 1157 void TransliteratorRoundTripTest::TestGreekUNGEGN() { 1158 logKnownIssue( "cldrbug:1911"); 1159 // It is left in its current state as a regression test. 1160 1161 RTTest test("Latin-Greek/UNGEGN"); 1162 LegalGreek *legal = new LegalGreek(FALSE); 1163 1164 test.test(UnicodeString("[a-zA-Z]", ""), 1165 UnicodeString("[\\u003B\\u00B7[[:Greek:]&[:Letter:]]-[" 1166 "\\u1D26-\\u1D2A" // L& [5] GREEK LETTER SMALL CAPITAL GAMMA..GREEK LETTER SMALL CAPITAL PSI 1167 "\\u1D5D-\\u1D61" // Lm [5] MODIFIER LETTER SMALL BETA..MODIFIER LETTER SMALL CHI 1168 "\\u1D66-\\u1D6A" // L& [5] GREEK SUBSCRIPT SMALL LETTER BETA..GREEK SUBSCRIPT SMALL LETTER CHI 1169 "\\u03D7-\\u03EF" // \N{GREEK KAI SYMBOL}..\N{COPTIC SMALL LETTER DEI} 1170 "] & [:Age=4.0:]]", 1171 //UnicodeString("[[\\u003B\\u00B7[:Greek:]-[\\u0374\\u0385\\u1fce\\u1fde\\u03D7-\\u03EF]]&[:Age=3.2:]]", 1172 ""), 1173 "[\\u0385\\u00B5\\u037A\\u03D0-\\uFFFF {\\u039C\\u03C0}]", /* roundtrip exclusions */ 1174 this, quick, legal); 1175 1176 delete legal; 1177 } 1178 1179 void TransliteratorRoundTripTest::Testel() { 1180 logKnownIssue( "cldrbug:1911"); 1181 // It is left in its current state as a regression test. 1182 1183 RTTest test("Latin-el"); 1184 LegalGreek *legal = new LegalGreek(FALSE); 1185 1186 test.test(UnicodeString("[a-zA-Z]", ""), 1187 UnicodeString("[\\u003B\\u00B7[[:Greek:]&[:Letter:]]-[" 1188 "\\u1D26-\\u1D2A" // L& [5] GREEK LETTER SMALL CAPITAL GAMMA..GREEK LETTER SMALL CAPITAL PSI 1189 "\\u1D5D-\\u1D61" // Lm [5] MODIFIER LETTER SMALL BETA..MODIFIER LETTER SMALL CHI 1190 "\\u1D66-\\u1D6A" // L& [5] GREEK SUBSCRIPT SMALL LETTER BETA..GREEK SUBSCRIPT SMALL LETTER CHI 1191 "\\u03D7-\\u03EF" // \N{GREEK KAI SYMBOL}..\N{COPTIC SMALL LETTER DEI} 1192 "] & [:Age=4.0:]]", 1193 //UnicodeString("[[\\u003B\\u00B7[:Greek:]-[\\u0374\\u0385\\u1fce\\u1fde\\u03D7-\\u03EF]]&[:Age=3.2:]]", 1194 ""), 1195 "[\\u00B5\\u037A\\u03D0-\\uFFFF {\\u039C\\u03C0}]", /* exclusions */ 1196 this, quick, legal); 1197 1198 1199 delete legal; 1200 } 1201 1202 1203 void TransliteratorRoundTripTest::TestArabic() { 1204 UnicodeString ARABIC("[\\u060C\\u061B\\u061F\\u0621\\u0627-\\u063A\\u0641-\\u0655\\u0660-\\u066C\\u067E\\u0686\\u0698\\u06A4\\u06AD\\u06AF\\u06CB-\\u06CC\\u06F0-\\u06F9]", -1, US_INV); 1205 Legal *legal = new Legal(); 1206 RTTest test("Latin-Arabic"); 1207 test.test(UNICODE_STRING_SIMPLE("[a-zA-Z\\u02BE\\u02BF\\u207F]"), ARABIC, "[a-zA-Z\\u02BE\\u02BF\\u207F]",this, quick, legal); // 1208 delete legal; 1209 } 1210 class LegalHebrew : public Legal { 1211 private: 1212 UnicodeSet FINAL; 1213 UnicodeSet NON_FINAL; 1214 UnicodeSet LETTER; 1215 public: 1216 LegalHebrew(UErrorCode& error); 1217 virtual ~LegalHebrew() {} 1218 virtual UBool is(const UnicodeString& sourceString) const; 1219 }; 1220 1221 LegalHebrew::LegalHebrew(UErrorCode& error){ 1222 FINAL.applyPattern(UNICODE_STRING_SIMPLE("[\\u05DA\\u05DD\\u05DF\\u05E3\\u05E5]"), error); 1223 NON_FINAL.applyPattern(UNICODE_STRING_SIMPLE("[\\u05DB\\u05DE\\u05E0\\u05E4\\u05E6]"), error); 1224 LETTER.applyPattern("[:letter:]", error); 1225 } 1226 UBool LegalHebrew::is(const UnicodeString& sourceString)const{ 1227 1228 if (sourceString.length() == 0) return TRUE; 1229 // don't worry about surrogates. 1230 for (int i = 0; i < sourceString.length(); ++i) { 1231 UChar ch = sourceString.charAt(i); 1232 UChar next = i+1 == sourceString.length() ? 0x0000 : sourceString.charAt(i); 1233 if (FINAL.contains(ch)) { 1234 if (LETTER.contains(next)) return FALSE; 1235 } else if (NON_FINAL.contains(ch)) { 1236 if (!LETTER.contains(next)) return FALSE; 1237 } 1238 } 1239 return TRUE; 1240 } 1241 void TransliteratorRoundTripTest::TestHebrew() { 1242 logKnownIssue( "cldrbug:1911"); 1243 // It is left in its current state as a regression test. 1244 1245 //long start = System.currentTimeMillis(); 1246 UErrorCode error = U_ZERO_ERROR; 1247 LegalHebrew* legal = new LegalHebrew(error); 1248 if(U_FAILURE(error)){ 1249 dataerrln("Could not construct LegalHebrew object. Error: %s", u_errorName(error)); 1250 return; 1251 } 1252 RTTest test("Latin-Hebrew"); 1253 test.test(UNICODE_STRING_SIMPLE("[a-zA-Z\\u02BC\\u02BB]"), UNICODE_STRING_SIMPLE("[[[:hebrew:]-[\\u05BD\\uFB00-\\uFBFF]]&[:Age=4.0:]]"), "[\\u05F0\\u05F1\\u05F2]", this, quick, legal); 1254 1255 //showElapsed(start, "TestHebrew"); 1256 delete legal; 1257 } 1258 void TransliteratorRoundTripTest::TestCyrillic() { 1259 RTTest test("Latin-Cyrillic"); 1260 Legal *legal = new Legal(); 1261 1262 test.test(UnicodeString("[a-zA-Z\\u0110\\u0111\\u02BA\\u02B9]", ""), 1263 UnicodeString("[[\\u0400-\\u045F] & [:Age=3.2:]]", ""), NULL, this, quick, 1264 legal); 1265 1266 delete legal; 1267 } 1268 1269 1270 // Inter-Indic Tests ---------------------------------- 1271 class LegalIndic :public Legal{ 1272 UnicodeSet vowelSignSet; 1273 UnicodeSet avagraha; 1274 UnicodeSet nukta; 1275 UnicodeSet virama; 1276 UnicodeSet sanskritStressSigns; 1277 UnicodeSet chandrabindu; 1278 1279 public: 1280 LegalIndic(); 1281 virtual UBool is(const UnicodeString& sourceString) const; 1282 virtual ~LegalIndic() {}; 1283 }; 1284 UBool LegalIndic::is(const UnicodeString& sourceString) const{ 1285 int cp=sourceString.charAt(0); 1286 1287 // A vowel sign cannot be the first char 1288 if(vowelSignSet.contains(cp)){ 1289 return FALSE; 1290 }else if(avagraha.contains(cp)){ 1291 return FALSE; 1292 }else if(virama.contains(cp)){ 1293 return FALSE; 1294 }else if(nukta.contains(cp)){ 1295 return FALSE; 1296 }else if(sanskritStressSigns.contains(cp)){ 1297 return FALSE; 1298 }else if(chandrabindu.contains(cp) && 1299 ((sourceString.length()>1) && 1300 vowelSignSet.contains(sourceString.charAt(1)))){ 1301 return FALSE; 1302 } 1303 return TRUE; 1304 } 1305 LegalIndic::LegalIndic(){ 1306 UErrorCode status = U_ZERO_ERROR; 1307 vowelSignSet.addAll( UnicodeSet("[\\u0902\\u0903\\u0904\\u093e-\\u094c\\u0962\\u0963]",status));/* Devanagari */ 1308 vowelSignSet.addAll( UnicodeSet("[\\u0982\\u0983\\u09be-\\u09cc\\u09e2\\u09e3\\u09D7]",status));/* Bengali */ 1309 vowelSignSet.addAll( UnicodeSet("[\\u0a02\\u0a03\\u0a3e-\\u0a4c\\u0a62\\u0a63\\u0a70\\u0a71]",status));/* Gurmukhi */ 1310 vowelSignSet.addAll( UnicodeSet("[\\u0a82\\u0a83\\u0abe-\\u0acc\\u0ae2\\u0ae3]",status));/* Gujarati */ 1311 vowelSignSet.addAll( UnicodeSet("[\\u0b02\\u0b03\\u0b3e-\\u0b4c\\u0b62\\u0b63\\u0b56\\u0b57]",status));/* Oriya */ 1312 vowelSignSet.addAll( UnicodeSet("[\\u0b82\\u0b83\\u0bbe-\\u0bcc\\u0be2\\u0be3\\u0bd7]",status));/* Tamil */ 1313 vowelSignSet.addAll( UnicodeSet("[\\u0c02\\u0c03\\u0c3e-\\u0c4c\\u0c62\\u0c63\\u0c55\\u0c56]",status));/* Telugu */ 1314 vowelSignSet.addAll( UnicodeSet("[\\u0c82\\u0c83\\u0cbe-\\u0ccc\\u0ce2\\u0ce3\\u0cd5\\u0cd6]",status));/* Kannada */ 1315 vowelSignSet.addAll( UnicodeSet("[\\u0d02\\u0d03\\u0d3e-\\u0d4c\\u0d62\\u0d63\\u0d57]",status));/* Malayalam */ 1316 1317 avagraha.addAll(UnicodeSet("[\\u093d\\u09bd\\u0abd\\u0b3d\\u0cbd]",status)); 1318 nukta.addAll(UnicodeSet("[\\u093c\\u09bc\\u0a3c\\u0abc\\u0b3c\\u0cbc]",status)); 1319 virama.addAll(UnicodeSet("[\\u094d\\u09cd\\u0a4d\\u0acd\\u0b4d\\u0bcd\\u0c4d\\u0ccd\\u0d4d]",status)); 1320 sanskritStressSigns.addAll(UnicodeSet("[\\u0951\\u0952\\u0953\\u0954\\u097d]",status)); 1321 chandrabindu.addAll(UnicodeSet("[\\u0901\\u0981\\u0A81\\u0b01\\u0c01]",status)); 1322 1323 } 1324 1325 static const char latinForIndic[] = "[['.0-9A-Za-z~\\u00C0-\\u00C5\\u00C7-\\u00CF\\u00D1-\\u00D6\\u00D9-\\u00DD" 1326 "\\u00E0-\\u00E5\\u00E7-\\u00EF\\u00F1-\\u00F6\\u00F9-\\u00FD\\u00FF-\\u010F" 1327 "\\u0112-\\u0125\\u0128-\\u0130\\u0134-\\u0137\\u0139-\\u013E\\u0143-\\u0148" 1328 "\\u014C-\\u0151\\u0154-\\u0165\\u0168-\\u017E\\u01A0-\\u01A1\\u01AF-\\u01B0" 1329 "\\u01CD-\\u01DC\\u01DE-\\u01E3\\u01E6-\\u01ED\\u01F0\\u01F4-\\u01F5\\u01F8-\\u01FB" 1330 "\\u0200-\\u021B\\u021E-\\u021F\\u0226-\\u0233\\u0294\\u0303-\\u0304\\u0306\\u0314-\\u0315" 1331 "\\u0325\\u040E\\u0419\\u0439\\u045E\\u04C1-\\u04C2\\u04D0-\\u04D1\\u04D6-\\u04D7" 1332 "\\u04E2-\\u04E3\\u04EE-\\u04EF\\u1E00-\\u1E99\\u1EA0-\\u1EF9\\u1F01\\u1F03\\u1F05" 1333 "\\u1F07\\u1F09\\u1F0B\\u1F0D\\u1F0F\\u1F11\\u1F13\\u1F15\\u1F19\\u1F1B\\u1F1D\\u1F21" 1334 "\\u1F23\\u1F25\\u1F27\\u1F29\\u1F2B\\u1F2D\\u1F2F\\u1F31\\u1F33\\u1F35\\u1F37\\u1F39" 1335 "\\u1F3B\\u1F3D\\u1F3F\\u1F41\\u1F43\\u1F45\\u1F49\\u1F4B\\u1F4D\\u1F51\\u1F53\\u1F55" 1336 "\\u1F57\\u1F59\\u1F5B\\u1F5D\\u1F5F\\u1F61\\u1F63\\u1F65\\u1F67\\u1F69\\u1F6B\\u1F6D" 1337 "\\u1F6F\\u1F81\\u1F83\\u1F85\\u1F87\\u1F89\\u1F8B\\u1F8D\\u1F8F\\u1F91\\u1F93\\u1F95" 1338 "\\u1F97\\u1F99\\u1F9B\\u1F9D\\u1F9F\\u1FA1\\u1FA3\\u1FA5\\u1FA7\\u1FA9\\u1FAB\\u1FAD" 1339 "\\u1FAF-\\u1FB1\\u1FB8-\\u1FB9\\u1FD0-\\u1FD1\\u1FD8-\\u1FD9\\u1FE0-\\u1FE1\\u1FE5" 1340 "\\u1FE8-\\u1FE9\\u1FEC\\u212A-\\u212B\\uE04D\\uE064]" 1341 "-[\\uE000-\\uE080 \\u01E2\\u01E3]& [[:latin:][:mark:]]]"; 1342 1343 void TransliteratorRoundTripTest::TestDevanagariLatin() { 1344 { 1345 UErrorCode status = U_ZERO_ERROR; 1346 UParseError parseError; 1347 TransliteratorPointer t1(Transliterator::createInstance("[\\u0964-\\u0965\\u0981-\\u0983\\u0985-\\u098C\\u098F-\\u0990\\u0993-\\u09A8\\u09AA-\\u09B0\\u09B2\\u09B6-\\u09B9\\u09BC\\u09BE-\\u09C4\\u09C7-\\u09C8\\u09CB-\\u09CD\\u09D7\\u09DC-\\u09DD\\u09DF-\\u09E3\\u09E6-\\u09FA];NFD;Bengali-InterIndic;InterIndic-Gujarati;NFC;",UTRANS_FORWARD, parseError, status)); 1348 if((Transliterator *)t1 != NULL){ 1349 TransliteratorPointer t2(t1->createInverse(status)); 1350 if(U_FAILURE(status)){ 1351 errln("FAIL: could not create the Inverse:-( \n"); 1352 } 1353 }else { 1354 dataerrln("FAIL: could not create the transliterator. Error: %s\n", u_errorName(status)); 1355 } 1356 1357 } 1358 RTTest test("Latin-Devanagari"); 1359 Legal *legal = new LegalIndic(); 1360 logKnownIssue( "cldrbug:1911"); 1361 // It is left in its current state as a regression test. 1362 1363 test.test(UnicodeString(latinForIndic, ""), 1364 UnicodeString("[[[:Devanagari:][\\u094d][\\u0964\\u0965]]&[:Age=4.1:]-[\\u0970]]", ""), "[\\u0965\\u0904]", this, quick, 1365 legal, 50); 1366 1367 delete legal; 1368 } 1369 1370 /* Defined this way for HP/UX11CC :-( */ 1371 static const int32_t INTER_INDIC_ARRAY_WIDTH = 4; 1372 static const char * const interIndicArray[] = { 1373 1374 "BENGALI-DEVANAGARI", "[:BENGALI:]", "[[:Devanagari:]-[\\u0970]]", 1375 "[\\u0904\\u0951-\\u0954\\u0943-\\u0949\\u094a\\u0962\\u0963\\u090D\\u090e\\u0911\\u0912\\u0929\\u0933\\u0934\\u0935\\u093d\\u0950\\u0958\\u0959\\u095a\\u095b\\u095e\\u097d]", /*roundtrip exclusions*/ 1376 1377 "DEVANAGARI-BENGALI", "[[:Devanagari:]-[\\u0970]]", "[:BENGALI:]", 1378 "[\\u0951-\\u0954\\u0951-\\u0954\\u09D7\\u090D\\u090e\\u0911\\u0912\\u0929\\u0933\\u0934\\u0935\\u093d\\u0950\\u0958\\u0959\\u095a\\u095b\\u095e\\u09f0\\u09f1\\u09f2-\\u09fa\\u09ce]", /*roundtrip exclusions*/ 1379 1380 "GURMUKHI-DEVANAGARI", "[:GURMUKHI:]", "[[:Devanagari:]-[\\u0970]]", 1381 "[\\u0904\\u0901\\u0902\\u0936\\u0933\\u0951-\\u0954\\u0902\\u0903\\u0943-\\u0949\\u094a\\u0962\\u0963\\u090B\\u090C\\u090D\\u090e\\u0911\\u0912\\u0934\\u0937\\u093D\\u0950\\u0960\\u0961\\u097d]", /*roundtrip exclusions*/ 1382 1383 "DEVANAGARI-GURMUKHI", "[[:Devanagari:]-[\\u0970]]", "[:GURMUKHI:]", 1384 "[\\u0904\\u0A02\\u0946\\u0A5C\\u0951-\\u0954\\u0A70\\u0A71\\u090B\\u090C\\u090D\\u090e\\u0911\\u0912\\u0934\\u0937\\u093D\\u0950\\u0960\\u0961\\u0a72\\u0a73\\u0a74]", /*roundtrip exclusions*/ 1385 1386 "GUJARATI-DEVANAGARI", "[:GUJARATI:]", "[[:Devanagari:]-[\\u0970]]", 1387 "[\\u0946\\u094A\\u0962\\u0963\\u0951-\\u0954\\u0961\\u090c\\u090e\\u0912\\u097d]", /*roundtrip exclusions*/ 1388 1389 "DEVANAGARI-GUJARATI", "[[:Devanagari:]-[\\u0970]]", "[:GUJARATI:]", 1390 "[\\u0951-\\u0954\\u0961\\u090c\\u090e\\u0912]", /*roundtrip exclusions*/ 1391 1392 "ORIYA-DEVANAGARI", "[:ORIYA:]", "[[:Devanagari:]-[\\u0970]]", 1393 "[\\u0904\\u0943-\\u094a\\u0962\\u0963\\u0951-\\u0954\\u0950\\u090D\\u090e\\u0912\\u0911\\u0931\\u0935\\u097d]", /*roundtrip exclusions*/ 1394 1395 "DEVANAGARI-ORIYA", "[[:Devanagari:]-[\\u0970]]", "[:ORIYA:]", 1396 "[\\u0b5f\\u0b56\\u0b57\\u0b70\\u0b71\\u0950\\u090D\\u090e\\u0912\\u0911\\u0931]", /*roundtrip exclusions*/ 1397 1398 "Tamil-DEVANAGARI", "[:tamil:]", "[[:Devanagari:]-[\\u0970]]", 1399 "[\\u0901\\u0904\\u093c\\u0943-\\u094a\\u0951-\\u0954\\u0962\\u0963\\u090B\\u090C\\u090D\\u0911\\u0916\\u0917\\u0918\\u091B\\u091D\\u0920\\u0921\\u0922\\u0925\\u0926\\u0927\\u092B\\u092C\\u092D\\u0936\\u093d\\u0950[\\u0958-\\u0961]\\u097d]", /*roundtrip exclusions*/ 1400 1401 "DEVANAGARI-Tamil", "[[:Devanagari:]-[\\u0970]]", "[:tamil:]", 1402 "[\\u0bd7\\u0BF0\\u0BF1\\u0BF2]", /*roundtrip exclusions*/ 1403 1404 "Telugu-DEVANAGARI", "[:telugu:]", "[[:Devanagari:]-[\\u0970]]", 1405 "[\\u0904\\u093c\\u0950\\u0945\\u0949\\u0951-\\u0954\\u0962\\u0963\\u090D\\u0911\\u093d\\u0929\\u0934[\\u0958-\\u095f]\\u097d]", /*roundtrip exclusions*/ 1406 1407 "DEVANAGARI-TELUGU", "[[:Devanagari:]-[\\u0970]]", "[:TELUGU:]", 1408 "[\\u0c55\\u0c56\\u0950\\u090D\\u0911\\u093d\\u0929\\u0934[\\u0958-\\u095f]]", /*roundtrip exclusions*/ 1409 1410 "KANNADA-DEVANAGARI", "[:KANNADA:]", "[[:Devanagari:]-[\\u0970]]", 1411 "[\\u0901\\u0904\\u0946\\u093c\\u0950\\u0945\\u0949\\u0951-\\u0954\\u0962\\u0963\\u0950\\u090D\\u0911\\u093d\\u0929\\u0934[\\u0958-\\u095f]\\u097d]", /*roundtrip exclusions*/ 1412 1413 "DEVANAGARI-KANNADA", "[[:Devanagari:]-[\\u0970]]", "[:KANNADA:]", 1414 "[{\\u0cb0\\u0cbc}{\\u0cb3\\u0cbc}\\u0cde\\u0cd5\\u0cd6\\u0950\\u090D\\u0911\\u093d\\u0929\\u0934[\\u0958-\\u095f]]", /*roundtrip exclusions*/ 1415 1416 "MALAYALAM-DEVANAGARI", "[:MALAYALAM:]", "[[:Devanagari:]-[\\u0970]]", 1417 "[\\u0901\\u0904\\u094a\\u094b\\u094c\\u093c\\u0950\\u0944\\u0945\\u0949\\u0951-\\u0954\\u0962\\u0963\\u090D\\u0911\\u093d\\u0929\\u0934[\\u0958-\\u095f]\\u097d]", /*roundtrip exclusions*/ 1418 1419 "DEVANAGARI-MALAYALAM", "[[:Devanagari:]-[\\u0970]]", "[:MALAYALAM:]", 1420 "[\\u0d4c\\u0d57\\u0950\\u090D\\u0911\\u093d\\u0929\\u0934[\\u0958-\\u095f]]", /*roundtrip exclusions*/ 1421 1422 "GURMUKHI-BENGALI", "[:GURMUKHI:]", "[:BENGALI:]", 1423 "[\\u0981\\u0982\\u09b6\\u09e2\\u09e3\\u09c3\\u09c4\\u09d7\\u098B\\u098C\\u09B7\\u09E0\\u09E1\\u09F0\\u09F1\\u09f2-\\u09fa\\u09ce]", /*roundtrip exclusions*/ 1424 1425 "BENGALI-GURMUKHI", "[:BENGALI:]", "[:GURMUKHI:]", 1426 "[\\u0A02\\u0a5c\\u0a47\\u0a70\\u0a71\\u0A33\\u0A35\\u0A59\\u0A5A\\u0A5B\\u0A5E\\u0A72\\u0A73\\u0A74]", /*roundtrip exclusions*/ 1427 1428 "GUJARATI-BENGALI", "[:GUJARATI:]", "[:BENGALI:]", 1429 "[\\u09d7\\u09e2\\u09e3\\u098c\\u09e1\\u09f0\\u09f1\\u09f2-\\u09fa\\u09ce]", /*roundtrip exclusions*/ 1430 1431 "BENGALI-GUJARATI", "[:BENGALI:]", "[:GUJARATI:]", 1432 "[\\u0A82\\u0a83\\u0Ac9\\u0Ac5\\u0ac7\\u0A8D\\u0A91\\u0AB3\\u0AB5\\u0ABD\\u0AD0]", /*roundtrip exclusions*/ 1433 1434 "ORIYA-BENGALI", "[:ORIYA:]", "[:BENGALI:]", 1435 "[\\u09c4\\u09e2\\u09e3\\u09f0\\u09f1\\u09f2-\\u09fa\\u09ce]", /*roundtrip exclusions*/ 1436 1437 "BENGALI-ORIYA", "[:BENGALI:]", "[:ORIYA:]", 1438 "[\\u0b35\\u0b71\\u0b5f\\u0b56\\u0b33\\u0b3d]", /*roundtrip exclusions*/ 1439 1440 "Tamil-BENGALI", "[:tamil:]", "[:BENGALI:]", 1441 "[\\u0981\\u09bc\\u09c3\\u09c4\\u09e2\\u09e3\\u09f0\\u09f1\\u098B\\u098C\\u0996\\u0997\\u0998\\u099B\\u099D\\u09A0\\u09A1\\u09A2\\u09A5\\u09A6\\u09A7\\u09AB\\u09AC\\u09AD\\u09B6\\u09DC\\u09DD\\u09DF\\u09E0\\u09E1\\u09f2-\\u09fa\\u09ce]", /*roundtrip exclusions*/ 1442 1443 "BENGALI-Tamil", "[:BENGALI:]", "[:tamil:]", 1444 "[\\u0bc6\\u0bc7\\u0bca\\u0B8E\\u0B92\\u0BA9\\u0BB1\\u0BB3\\u0BB4\\u0BB5\\u0BF0\\u0BF1\\u0BF2]", /*roundtrip exclusions*/ 1445 1446 "Telugu-BENGALI", "[:telugu:]", "[:BENGALI:]", 1447 "[\\u09e2\\u09e3\\u09bc\\u09d7\\u09f0\\u09f1\\u09dc\\u09dd\\u09df\\u09f2-\\u09fa\\u09ce]", /*roundtrip exclusions*/ 1448 1449 "BENGALI-TELUGU", "[:BENGALI:]", "[:TELUGU:]", 1450 "[\\u0c55\\u0c56\\u0c47\\u0c46\\u0c4a\\u0C0E\\u0C12\\u0C31\\u0C33\\u0C35]", /*roundtrip exclusions*/ 1451 1452 "KANNADA-BENGALI", "[:KANNADA:]", "[:BENGALI:]", 1453 "[\\u0981\\u09e2\\u09e3\\u09bc\\u09d7\\u09dc\\u09dd\\u09df\\u09f0\\u09f1\\u09f2-\\u09fa\\u09ce]", /*roundtrip exclusions*/ 1454 1455 "BENGALI-KANNADA", "[:BENGALI:]", "[:KANNADA:]", 1456 "[{\\u0cb0\\u0cbc}{\\u0cb3\\u0cbc}\\u0cc6\\u0cca\\u0cd5\\u0cd6\\u0cc7\\u0C8E\\u0C92\\u0CB1\\u0cb3\\u0cb5\\u0cde]", /*roundtrip exclusions*/ 1457 1458 "MALAYALAM-BENGALI", "[:MALAYALAM:]", "[:BENGALI:]", 1459 "[\\u0981\\u09e2\\u09e3\\u09bc\\u09c4\\u09f0\\u09f1\\u09dc\\u09dd\\u09df\\u09dc\\u09dd\\u09df\\u09f2-\\u09fa\\u09ce]", /*roundtrip exclusions*/ 1460 1461 "BENGALI-MALAYALAM", "[:BENGALI:]", "[:MALAYALAM:]", 1462 "[\\u0d46\\u0d4a\\u0d47\\u0d31-\\u0d35\\u0d0e\\u0d12]", /*roundtrip exclusions*/ 1463 1464 "GUJARATI-GURMUKHI", "[:GUJARATI:]", "[:GURMUKHI:]", 1465 "[\\u0A02\\u0ab3\\u0ab6\\u0A70\\u0a71\\u0a82\\u0a83\\u0ac3\\u0ac4\\u0ac5\\u0ac9\\u0a5c\\u0a72\\u0a73\\u0a74\\u0a8b\\u0a8d\\u0a91\\u0abd]", /*roundtrip exclusions*/ 1466 1467 "GURMUKHI-GUJARATI", "[:GURMUKHI:]", "[:GUJARATI:]", 1468 "[\\u0a5c\\u0A70\\u0a71\\u0a72\\u0a73\\u0a74\\u0a82\\u0a83\\u0a8b\\u0a8c\\u0a8d\\u0a91\\u0ab3\\u0ab6\\u0ab7\\u0abd\\u0ac3\\u0ac4\\u0ac5\\u0ac9\\u0ad0\\u0ae0\\u0ae1]", /*roundtrip exclusions*/ 1469 1470 "ORIYA-GURMUKHI", "[:ORIYA:]", "[:GURMUKHI:]", 1471 "[\\u0A01\\u0A02\\u0a5c\\u0a21\\u0a47\\u0a71\\u0b02\\u0b03\\u0b33\\u0b36\\u0b43\\u0b56\\u0b57\\u0B0B\\u0B0C\\u0B37\\u0B3D\\u0B5F\\u0B60\\u0B61\\u0a35\\u0a72\\u0a73\\u0a74]", /*roundtrip exclusions*/ 1472 1473 "GURMUKHI-ORIYA", "[:GURMUKHI:]", "[:ORIYA:]", 1474 "[\\u0b01\\u0b02\\u0b03\\u0b33\\u0b36\\u0b43\\u0b56\\u0b57\\u0B0B\\u0B0C\\u0B37\\u0B3D\\u0B5F\\u0B60\\u0B61\\u0b70\\u0b71]", /*roundtrip exclusions*/ 1475 1476 "TAMIL-GURMUKHI", "[:TAMIL:]", "[:GURMUKHI:]", 1477 "[\\u0A01\\u0A02\\u0a33\\u0a36\\u0a3c\\u0a70\\u0a71\\u0a47\\u0A16\\u0A17\\u0A18\\u0A1B\\u0A1D\\u0A20\\u0A21\\u0A22\\u0A25\\u0A26\\u0A27\\u0A2B\\u0A2C\\u0A2D\\u0A59\\u0A5A\\u0A5B\\u0A5C\\u0A5E\\u0A72\\u0A73\\u0A74]", /*roundtrip exclusions*/ 1478 1479 "GURMUKHI-TAMIL", "[:GURMUKHI:]", "[:TAMIL:]", 1480 "[\\u0b82\\u0bc6\\u0bca\\u0bd7\\u0bb7\\u0bb3\\u0b83\\u0B8E\\u0B92\\u0BA9\\u0BB1\\u0BB4\\u0bb6\\u0BF0\\u0BF1\\u0BF2]", /*roundtrip exclusions*/ 1481 1482 "TELUGU-GURMUKHI", "[:TELUGU:]", "[:GURMUKHI:]", 1483 "[\\u0A02\\u0a33\\u0a36\\u0a3c\\u0a70\\u0a71\\u0A59\\u0A5A\\u0A5B\\u0A5C\\u0A5E\\u0A72\\u0A73\\u0A74]", /*roundtrip exclusions*/ 1484 1485 "GURMUKHI-TELUGU", "[:GURMUKHI:]", "[:TELUGU:]", 1486 "[\\u0c01\\u0c02\\u0c03\\u0c33\\u0c36\\u0c44\\u0c43\\u0c46\\u0c4a\\u0c56\\u0c55\\u0C0B\\u0C0C\\u0C0E\\u0C12\\u0C31\\u0C37\\u0C60\\u0C61]", /*roundtrip exclusions*/ 1487 1488 "KANNADA-GURMUKHI", "[:KANNADA:]", "[:GURMUKHI:]", 1489 "[\\u0A01\\u0A02\\u0a33\\u0a36\\u0a3c\\u0a70\\u0a71\\u0A59\\u0A5A\\u0A5B\\u0A5C\\u0A5E\\u0A72\\u0A73\\u0A74]", /*roundtrip exclusions*/ 1490 1491 "GURMUKHI-KANNADA", "[:GURMUKHI:]", "[:KANNADA:]", 1492 "[{\\u0cb0\\u0cbc}{\\u0cb3\\u0cbc}\\u0c82\\u0c83\\u0cb3\\u0cb6\\u0cc4\\u0cc3\\u0cc6\\u0cca\\u0cd5\\u0cd6\\u0C8B\\u0C8C\\u0C8E\\u0C92\\u0CB1\\u0CB7\\u0cbd\\u0CE0\\u0CE1\\u0cde]", /*roundtrip exclusions*/ 1493 1494 "MALAYALAM-GURMUKHI", "[:MALAYALAM:]", "[:GURMUKHI:]", 1495 "[\\u0A01\\u0A02\\u0a4b\\u0a4c\\u0a33\\u0a36\\u0a3c\\u0a70\\u0a71\\u0A59\\u0A5A\\u0A5B\\u0A5C\\u0A5E\\u0A72\\u0A73\\u0A74]", /*roundtrip exclusions*/ 1496 1497 "GURMUKHI-MALAYALAM", "[:GURMUKHI:]", "[:MALAYALAM:]", 1498 "[\\u0d02\\u0d03\\u0d33\\u0d36\\u0d43\\u0d46\\u0d4a\\u0d4c\\u0d57\\u0D0B\\u0D0C\\u0D0E\\u0D12\\u0D31\\u0D34\\u0D37\\u0D60\\u0D61]", /*roundtrip exclusions*/ 1499 1500 "GUJARATI-ORIYA", "[:GUJARATI:]", "[:ORIYA:]", 1501 "[\\u0b56\\u0b57\\u0B0C\\u0B5F\\u0B61\\u0b70\\u0b71]", /*roundtrip exclusions*/ 1502 1503 "ORIYA-GUJARATI", "[:ORIYA:]", "[:GUJARATI:]", 1504 "[\\u0Ac4\\u0Ac5\\u0Ac9\\u0Ac7\\u0A8D\\u0A91\\u0AB5\\u0Ad0]", /*roundtrip exclusions*/ 1505 1506 "TAMIL-GUJARATI", "[:TAMIL:]", "[:GUJARATI:]", 1507 "[\\u0A81\\u0a8c\\u0abc\\u0ac3\\u0Ac4\\u0Ac5\\u0Ac9\\u0Ac7\\u0A8B\\u0A8D\\u0A91\\u0A96\\u0A97\\u0A98\\u0A9B\\u0A9D\\u0AA0\\u0AA1\\u0AA2\\u0AA5\\u0AA6\\u0AA7\\u0AAB\\u0AAC\\u0AAD\\u0AB6\\u0ABD\\u0AD0\\u0AE0\\u0AE1]", /*roundtrip exclusions*/ 1508 1509 "GUJARATI-TAMIL", "[:GUJARATI:]", "[:TAMIL:]", 1510 "[\\u0Bc6\\u0Bca\\u0Bd7\\u0B8E\\u0B92\\u0BA9\\u0BB1\\u0BB4\\u0BF0\\u0BF1\\u0BF2]", /*roundtrip exclusions*/ 1511 1512 "TELUGU-GUJARATI", "[:TELUGU:]", "[:GUJARATI:]", 1513 "[\\u0abc\\u0Ac5\\u0Ac9\\u0A8D\\u0A91\\u0ABD\\u0Ad0]", /*roundtrip exclusions*/ 1514 1515 "GUJARATI-TELUGU", "[:GUJARATI:]", "[:TELUGU:]", 1516 "[\\u0c46\\u0c4a\\u0c55\\u0c56\\u0C0C\\u0C0E\\u0C12\\u0C31\\u0C61]", /*roundtrip exclusions*/ 1517 1518 "KANNADA-GUJARATI", "[:KANNADA:]", "[:GUJARATI:]", 1519 "[\\u0A81\\u0abc\\u0Ac5\\u0Ac9\\u0A8D\\u0A91\\u0ABD\\u0Ad0]", /*roundtrip exclusions*/ 1520 1521 "GUJARATI-KANNADA", "[:GUJARATI:]", "[:KANNADA:]", 1522 "[{\\u0cb0\\u0cbc}{\\u0cb3\\u0cbc}\\u0cc6\\u0cca\\u0cd5\\u0cd6\\u0C8C\\u0C8E\\u0C92\\u0CB1\\u0CDE\\u0CE1]", /*roundtrip exclusions*/ 1523 1524 "MALAYALAM-GUJARATI", "[:MALAYALAM:]", "[:GUJARATI:]", 1525 "[\\u0A81\\u0ac4\\u0acb\\u0acc\\u0abc\\u0Ac5\\u0Ac9\\u0A8D\\u0A91\\u0ABD\\u0Ad0]", /*roundtrip exclusions*/ 1526 1527 "GUJARATI-MALAYALAM", "[:GUJARATI:]", "[:MALAYALAM:]", 1528 "[\\u0d46\\u0d4a\\u0d4c\\u0d55\\u0d57\\u0D0C\\u0D0E\\u0D12\\u0D31\\u0D34\\u0D61]", /*roundtrip exclusions*/ 1529 1530 "TAMIL-ORIYA", "[:TAMIL:]", "[:ORIYA:]", 1531 "[\\u0B01\\u0b3c\\u0b43\\u0b56\\u0B0B\\u0B0C\\u0B16\\u0B17\\u0B18\\u0B1B\\u0B1D\\u0B20\\u0B21\\u0B22\\u0B25\\u0B26\\u0B27\\u0B2B\\u0B2C\\u0B2D\\u0B36\\u0B3D\\u0B5C\\u0B5D\\u0B5F\\u0B60\\u0B61\\u0b70\\u0b71]", /*roundtrip exclusions*/ 1532 1533 "ORIYA-TAMIL", "[:ORIYA:]", "[:TAMIL:]", 1534 "[\\u0bc6\\u0bca\\u0bc7\\u0B8E\\u0B92\\u0BA9\\u0BB1\\u0BB4\\u0BB5\\u0BF0\\u0BF1\\u0BF2]", /*roundtrip exclusions*/ 1535 1536 "TELUGU-ORIYA", "[:TELUGU:]", "[:ORIYA:]", 1537 "[\\u0b3c\\u0b57\\u0b56\\u0B3D\\u0B5C\\u0B5D\\u0B5F\\u0b70\\u0b71]", /*roundtrip exclusions*/ 1538 1539 "ORIYA-TELUGU", "[:ORIYA:]", "[:TELUGU:]", 1540 "[\\u0c44\\u0c46\\u0c4a\\u0c55\\u0c47\\u0C0E\\u0C12\\u0C31\\u0C35]", /*roundtrip exclusions*/ 1541 1542 "KANNADA-ORIYA", "[:KANNADA:]", "[:ORIYA:]", 1543 "[\\u0B01\\u0b3c\\u0b57\\u0B3D\\u0B5C\\u0B5D\\u0B5F\\u0b70\\u0b71]", /*roundtrip exclusions*/ 1544 1545 "ORIYA-KANNADA", "[:ORIYA:]", "[:KANNADA:]", 1546 "[{\\u0cb0\\u0cbc}{\\u0cb3\\u0cbc}\\u0cc4\\u0cc6\\u0cca\\u0cd5\\u0cc7\\u0C8E\\u0C92\\u0CB1\\u0CB5\\u0CDE]", /*roundtrip exclusions*/ 1547 1548 "MALAYALAM-ORIYA", "[:MALAYALAM:]", "[:ORIYA:]", 1549 "[\\u0B01\\u0b3c\\u0b56\\u0B3D\\u0B5C\\u0B5D\\u0B5F\\u0b70\\u0b71]", /*roundtrip exclusions*/ 1550 1551 "ORIYA-MALAYALAM", "[:ORIYA:]", "[:MALAYALAM:]", 1552 "[\\u0D47\\u0D46\\u0D4a\\u0D0E\\u0D12\\u0D31\\u0D34\\u0D35]", /*roundtrip exclusions*/ 1553 1554 "TELUGU-TAMIL", "[:TELUGU:]", "[:TAMIL:]", 1555 "[\\u0bd7\\u0ba9\\u0bb4\\u0BF0\\u0BF1\\u0BF2]", /*roundtrip exclusions*/ 1556 1557 "TAMIL-TELUGU", "[:TAMIL:]", "[:TELUGU:]", 1558 "[\\u0C01\\u0c43\\u0c44\\u0c46\\u0c47\\u0c55\\u0c56\\u0c66\\u0C0B\\u0C0C\\u0C16\\u0C17\\u0C18\\u0C1B\\u0C1D\\u0C20\\u0C21\\u0C22\\u0C25\\u0C26\\u0C27\\u0C2B\\u0C2C\\u0C2D\\u0C36\\u0C60\\u0C61]", /*roundtrip exclusions*/ 1559 1560 "KANNADA-TAMIL", "[:KANNADA:]", "[:TAMIL:]", 1561 "[\\u0bd7\\u0bc6\\u0ba9\\u0bb4\\u0BF0\\u0BF1\\u0BF2]", /*roundtrip exclusions*/ 1562 1563 "TAMIL-KANNADA", "[:TAMIL:]", "[:KANNADA:]", 1564 "[\\u0cc3\\u0cc4\\u0cc6\\u0cc7\\u0cd5\\u0cd6\\u0C8B\\u0C8C\\u0C96\\u0C97\\u0C98\\u0C9B\\u0C9D\\u0CA0\\u0CA1\\u0CA2\\u0CA5\\u0CA6\\u0CA7\\u0CAB\\u0CAC\\u0CAD\\u0CB6\\u0cbc\\u0cbd\\u0CDE\\u0CE0\\u0CE1]", /*roundtrip exclusions*/ 1565 1566 "MALAYALAM-TAMIL", "[:MALAYALAM:]", "[:TAMIL:]", 1567 "[\\u0ba9\\u0BF0\\u0BF1\\u0BF2]", /*roundtrip exclusions*/ 1568 1569 "TAMIL-MALAYALAM", "[:TAMIL:]", "[:MALAYALAM:]", 1570 "[\\u0d43\\u0d12\\u0D0B\\u0D0C\\u0D16\\u0D17\\u0D18\\u0D1B\\u0D1D\\u0D20\\u0D21\\u0D22\\u0D25\\u0D26\\u0D27\\u0D2B\\u0D2C\\u0D2D\\u0D36\\u0D60\\u0D61]", /*roundtrip exclusions*/ 1571 1572 "KANNADA-TELUGU", "[:KANNADA:]", "[:TELUGU:]", 1573 "[\\u0C01\\u0c3f\\u0c46\\u0c48\\u0c4a]", /*roundtrip exclusions*/ 1574 1575 "TELUGU-KANNADA", "[:TELUGU:]", "[:KANNADA:]", 1576 "[\\u0cc8\\u0cd5\\u0cd6\\u0cbc\\u0cbd\\u0CDE]", /*roundtrip exclusions*/ 1577 1578 "MALAYALAM-TELUGU", "[:MALAYALAM:]", "[:TELUGU:]", 1579 "[\\u0C01\\u0c44\\u0c4a\\u0c4c\\u0c4b\\u0c55\\u0c56]", /*roundtrip exclusions*/ 1580 1581 "TELUGU-MALAYALAM", "[:TELUGU:]", "[:MALAYALAM:]", 1582 "[\\u0d4c\\u0d57\\u0D34]", /*roundtrip exclusions*/ 1583 1584 "MALAYALAM-KANNADA", "[:MALAYALAM:]", "[:KANNADA:]", 1585 "[\\u0cbc\\u0cbd\\u0cc4\\u0cc6\\u0cca\\u0ccc\\u0ccb\\u0cd5\\u0cd6\\u0cDe]", /*roundtrip exclusions*/ 1586 1587 "KANNADA-MALAYALAM", "[:KANNADA:]", "[:MALAYALAM:]", 1588 "[\\u0d4c\\u0d57\\u0d46\\u0D34]", /*roundtrip exclusions*/ 1589 1590 "Latin-Bengali",latinForIndic, "[[:Bengali:][\\u0964\\u0965]]", 1591 "[\\u0965\\u09f0-\\u09fa\\u09ce]" /*roundtrip exclusions*/ , 1592 1593 "Latin-Gurmukhi", latinForIndic, "[[:Gurmukhi:][\\u0964\\u0965]]", 1594 "[\\u0a01\\u0965\\u0a02\\u0a72\\u0a73\\u0a74]" /*roundtrip exclusions*/, 1595 1596 "Latin-Gujarati",latinForIndic, "[[:Gujarati:][\\u0964\\u0965]]", 1597 "[\\u0965]" /*roundtrip exclusions*/, 1598 1599 "Latin-Oriya",latinForIndic, "[[:Oriya:][\\u0964\\u0965]]", 1600 "[\\u0965\\u0b70]" /*roundtrip exclusions*/, 1601 1602 "Latin-Tamil",latinForIndic, "[:Tamil:]", 1603 "[\\u0BF0\\u0BF1\\u0BF2]" /*roundtrip exclusions*/, 1604 1605 "Latin-Telugu",latinForIndic, "[:Telugu:]", 1606 NULL /*roundtrip exclusions*/, 1607 1608 "Latin-Kannada",latinForIndic, "[:Kannada:]", 1609 NULL /*roundtrip exclusions*/, 1610 1611 "Latin-Malayalam",latinForIndic, "[:Malayalam:]", 1612 NULL /*roundtrip exclusions*/ 1613 }; 1614 1615 void TransliteratorRoundTripTest::TestDebug(const char* name,const char fromSet[], 1616 const char* toSet,const char* exclusions){ 1617 1618 RTTest test(name); 1619 Legal *legal = new LegalIndic(); 1620 test.test(UnicodeString(fromSet,""),UnicodeString(toSet,""),exclusions,this,quick,legal); 1621 } 1622 1623 void TransliteratorRoundTripTest::TestInterIndic() { 1624 //TestDebug("Latin-Gurmukhi", latinForIndic, "[:Gurmukhi:]","[\\u0965\\u0a02\\u0a72\\u0a73\\u0a74]",TRUE); 1625 int32_t num = UPRV_LENGTHOF(interIndicArray)/INTER_INDIC_ARRAY_WIDTH; 1626 if(quick){ 1627 logln("Testing only 5 of %i. Skipping rest (use -e for exhaustive)",num); 1628 num = 5; 1629 } 1630 for(int i = 0; i < num;i++){ 1631 RTTest test(interIndicArray[i*INTER_INDIC_ARRAY_WIDTH + 0]); 1632 Legal *legal = new LegalIndic(); 1633 logln(UnicodeString("Stress testing ") + interIndicArray[i*INTER_INDIC_ARRAY_WIDTH + 0]); 1634 if( !logKnownIssue( "cldrbug:1911" ) ) { 1635 /* "full test" */ 1636 // CLDR bug #1911: This test should be moved into CLDR. 1637 test.test( interIndicArray[i*INTER_INDIC_ARRAY_WIDTH + 1], 1638 interIndicArray[i*INTER_INDIC_ARRAY_WIDTH + 2], 1639 interIndicArray[i*INTER_INDIC_ARRAY_WIDTH + 3], // roundtrip exclusions 1640 this, quick, legal, 50); 1641 } else { 1642 // It is left in its current state as a regression test. 1643 // CLDR should test, and remove the age filter. 1644 /* regression test - ""temporary"" until CLDR#1911 is fixed */ 1645 // start 1646 UnicodeString source("["); 1647 source.append(interIndicArray[i*INTER_INDIC_ARRAY_WIDTH + 1]); 1648 source.append(" & [:Age=4.1:]]"); 1649 UnicodeString target("["); 1650 target.append(interIndicArray[i*INTER_INDIC_ARRAY_WIDTH + 2]); 1651 target.append(" & [:Age=4.1:]]"); 1652 test.test( source, 1653 target, 1654 interIndicArray[i*INTER_INDIC_ARRAY_WIDTH + 3], // roundtrip exclusions 1655 this, quick, legal, 50); 1656 // end 1657 delete legal; 1658 } 1659 } 1660 } 1661 1662 // end indic tests ---------------------------------------------------------- 1663 1664 #endif /* #if !UCONFIG_NO_TRANSLITERATION */ 1665
