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 "uassert.h"
37 #include <string.h>
38 #include <stdio.h>
39
40 #define CASE(id,test) case id: \
41 name = #test; \
42 if (exec) { \
43 logln(#test "---"); \
44 logln((UnicodeString)""); \
45 UDate t = uprv_getUTCtime(); \
46 test(); \
47 t = uprv_getUTCtime() - t; \
48 logln((UnicodeString)#test " took " + t/U_MILLIS_PER_DAY + " seconds"); \
49 } \
50 break
51
52 #define EXHAUSTIVE(id,test) case id: \
53 if(quick==FALSE){ \
54 name = #test; \
55 if (exec){ \
56 logln(#test "---"); \
57 logln((UnicodeString)""); \
58 test(); \
59 } \
60 }else{ \
61 name=""; \
62 } \
63 break
64 void
runIndexedTest(int32_t index,UBool exec,const char * & name,char *)65 TransliteratorRoundTripTest::runIndexedTest(int32_t index, UBool exec,
66 const char* &name, char* /*par*/) {
67 switch (index) {
68 CASE(0, TestCyrillic);
69 // CASE(0,TestKana);
70 CASE(1,TestHiragana);
71 CASE(2,TestKatakana);
72 CASE(3,TestJamo);
73 CASE(4,TestHangul);
74 CASE(5,TestGreek);
75 CASE(6,TestGreekUNGEGN);
76 CASE(7,Testel);
77 CASE(8,TestDevanagariLatin);
78 CASE(9,TestInterIndic);
79 CASE(10, TestHebrew);
80 CASE(11, TestArabic);
81 CASE(12, TestHan);
82 default: name = ""; break;
83 }
84 }
85
86
87 //--------------------------------------------------------------------
88 // TransliteratorPointer
89 //--------------------------------------------------------------------
90
91 /**
92 * A transliterator pointer wrapper that deletes the contained
93 * pointer automatically when the wrapper goes out of scope.
94 * Sometimes called a "janitor" or "smart pointer".
95 */
96 class TransliteratorPointer {
97 Transliterator* t;
98 // disallowed:
99 TransliteratorPointer(const TransliteratorPointer& rhs);
100 TransliteratorPointer& operator=(const TransliteratorPointer& rhs);
101 public:
TransliteratorPointer(Transliterator * adopted)102 TransliteratorPointer(Transliterator* adopted) {
103 t = adopted;
104 }
~TransliteratorPointer()105 ~TransliteratorPointer() {
106 delete t;
107 }
operator ->()108 inline Transliterator* operator->() { return t; }
operator const Transliterator*() const109 inline operator const Transliterator*() const { return t; }
operator Transliterator*()110 inline operator Transliterator*() { return t; }
111 };
112
113 //--------------------------------------------------------------------
114 // Legal
115 //--------------------------------------------------------------------
116
117 class Legal {
118 public:
Legal()119 Legal() {}
~Legal()120 virtual ~Legal() {}
is(const UnicodeString &) const121 virtual UBool is(const UnicodeString& /*sourceString*/) const {return TRUE;}
122 };
123
124 class LegalJamo : public Legal {
125 // any initial must be followed by a medial (or initial)
126 // any medial must follow an initial (or medial)
127 // any final must follow a medial (or final)
128 public:
LegalJamo()129 LegalJamo() {}
~LegalJamo()130 virtual ~LegalJamo() {}
131 virtual UBool is(const UnicodeString& sourceString) const;
132 int getType(UChar c) const;
133 };
134
is(const UnicodeString & sourceString) const135 UBool LegalJamo::is(const UnicodeString& sourceString) const {
136 int t;
137 UnicodeString decomp;
138 UErrorCode ec = U_ZERO_ERROR;
139 Normalizer::decompose(sourceString, FALSE, 0, decomp, ec);
140 if (U_FAILURE(ec)) {
141 return FALSE;
142 }
143 for (int i = 0; i < decomp.length(); ++i) { // don't worry about surrogates
144 switch (getType(decomp.charAt(i))) {
145 case 0: t = getType(decomp.charAt(i+1));
146 if (t != 0 && t != 1) { return FALSE; }
147 break;
148 case 1: t = getType(decomp.charAt(i-1));
149 if (t != 0 && t != 1) { return FALSE; }
150 break;
151 case 2: t = getType(decomp.charAt(i-1));
152 if (t != 1 && t != 2) { return FALSE; }
153 break;
154 }
155 }
156 return TRUE;
157 }
158
getType(UChar c) const159 int LegalJamo::getType(UChar c) const {
160 if (0x1100 <= c && c <= 0x1112)
161 return 0;
162 else if (0x1161 <= c && c <= 0x1175)
163 return 1;
164 else if (0x11A8 <= c && c <= 0x11C2)
165 return 2;
166 return -1; // other
167 }
168
169 class LegalGreek : public Legal {
170 UBool full;
171 public:
LegalGreek(UBool _full)172 LegalGreek(UBool _full) { full = _full; }
~LegalGreek()173 virtual ~LegalGreek() {}
174
175 virtual UBool is(const UnicodeString& sourceString) const;
176
177 static UBool isVowel(UChar c);
178
179 static UBool isRho(UChar c);
180 };
181
is(const UnicodeString & sourceString) const182 UBool LegalGreek::is(const UnicodeString& sourceString) const {
183 UnicodeString decomp;
184 UErrorCode ec = U_ZERO_ERROR;
185 Normalizer::decompose(sourceString, FALSE, 0, decomp, ec);
186
187 // modern is simpler: don't care about anything but a grave
188 if (full == FALSE) {
189 // A special case which is legal but should be
190 // excluded from round trip
191 // if (sourceString == UnicodeString("\\u039C\\u03C0", "")) {
192 // return FALSE;
193 // }
194 for (int32_t i = 0; i < decomp.length(); ++i) {
195 UChar c = decomp.charAt(i);
196 // exclude all the accents
197 if (c == 0x0313 || c == 0x0314 || c == 0x0300 || c == 0x0302
198 || c == 0x0342 || c == 0x0345
199 ) return FALSE;
200 }
201 return TRUE;
202 }
203
204 // Legal greek has breathing marks IFF there is a vowel or RHO at the start
205 // IF it has them, it has exactly one.
206 // IF it starts with a RHO, then the breathing mark must come before the second letter.
207 // Since there are no surrogates in greek, don't worry about them
208 UBool firstIsVowel = FALSE;
209 UBool firstIsRho = FALSE;
210 UBool noLetterYet = TRUE;
211 int32_t breathingCount = 0;
212 int32_t letterCount = 0;
213 for (int32_t i = 0; i < decomp.length(); ++i) {
214 UChar c = decomp.charAt(i);
215 if (u_isalpha(c)) {
216 ++letterCount;
217 if (noLetterYet) {
218 noLetterYet = FALSE;
219 firstIsVowel = isVowel(c);
220 firstIsRho = isRho(c);
221 }
222 if (firstIsRho && letterCount == 2 && breathingCount == 0) {
223 return FALSE;
224 }
225 }
226 if (c == 0x0313 || c == 0x0314) {
227 ++breathingCount;
228 }
229 }
230
231 if (firstIsVowel || firstIsRho) return breathingCount == 1;
232 return breathingCount == 0;
233 }
234
isVowel(UChar c)235 UBool LegalGreek::isVowel(UChar c) {
236 switch (c) {
237 case 0x03B1:
238 case 0x03B5:
239 case 0x03B7:
240 case 0x03B9:
241 case 0x03BF:
242 case 0x03C5:
243 case 0x03C9:
244 case 0x0391:
245 case 0x0395:
246 case 0x0397:
247 case 0x0399:
248 case 0x039F:
249 case 0x03A5:
250 case 0x03A9:
251 return TRUE;
252 }
253 return FALSE;
254 }
255
isRho(UChar c)256 UBool LegalGreek::isRho(UChar c) {
257 switch (c) {
258 case 0x03C1:
259 case 0x03A1:
260 return TRUE;
261 }
262 return FALSE;
263 }
264
265 // AbbreviatedUnicodeSetIterator Interface ---------------------------------------------
266 //
267 // Iterate over a UnicodeSet, only returning a sampling of the contained code points.
268 // density is the approximate total number of code points to returned for the entire set.
269 //
270
271 class AbbreviatedUnicodeSetIterator : public UnicodeSetIterator {
272 public :
273
274 AbbreviatedUnicodeSetIterator();
275 virtual ~AbbreviatedUnicodeSetIterator();
276 void reset(UnicodeSet& set, UBool abb = FALSE, int32_t density = 100);
277
278 /**
279 * ICU "poor man's RTTI", returns a UClassID for this class.
280 */
getStaticClassID()281 static inline UClassID getStaticClassID() { return (UClassID)&fgClassID; }
282
283 /**
284 * ICU "poor man's RTTI", returns a UClassID for the actual class.
285 */
getDynamicClassID() const286 virtual inline UClassID getDynamicClassID() const { return getStaticClassID(); }
287
288 private :
289 UBool abbreviated;
290 int32_t perRange; // The maximum number of code points to be returned from each range
291 virtual void loadRange(int32_t range);
292
293 /**
294 * The address of this static class variable serves as this class's ID
295 * for ICU "poor man's RTTI".
296 */
297 static const char fgClassID;
298 };
299
300 // AbbreviatedUnicodeSetIterator Implementation ---------------------------------------
301
302 const char AbbreviatedUnicodeSetIterator::fgClassID=0;
303
AbbreviatedUnicodeSetIterator()304 AbbreviatedUnicodeSetIterator::AbbreviatedUnicodeSetIterator() :
305 UnicodeSetIterator(), abbreviated(FALSE) {
306 }
307
~AbbreviatedUnicodeSetIterator()308 AbbreviatedUnicodeSetIterator::~AbbreviatedUnicodeSetIterator() {
309 }
310
reset(UnicodeSet & newSet,UBool abb,int32_t density)311 void AbbreviatedUnicodeSetIterator::reset(UnicodeSet& newSet, UBool abb, int32_t density) {
312 UnicodeSetIterator::reset(newSet);
313 abbreviated = abb;
314 perRange = newSet.getRangeCount();
315 if (perRange != 0) {
316 perRange = density / perRange;
317 }
318 }
319
loadRange(int32_t myRange)320 void AbbreviatedUnicodeSetIterator::loadRange(int32_t myRange) {
321 UnicodeSetIterator::loadRange(myRange);
322 if (abbreviated && (endElement > nextElement + perRange)) {
323 endElement = nextElement + perRange;
324 }
325 }
326
327 //--------------------------------------------------------------------
328 // RTTest Interface
329 //--------------------------------------------------------------------
330
331 class RTTest : public IntlTest {
332
333 // PrintWriter out;
334
335 UnicodeString transliteratorID;
336 int32_t errorLimit;
337 int32_t errorCount;
338 int32_t pairLimit;
339 UnicodeSet sourceRange;
340 UnicodeSet targetRange;
341 UnicodeSet toSource;
342 UnicodeSet toTarget;
343 UnicodeSet roundtripExclusionsSet;
344 IntlTest* parent;
345 Legal* legalSource; // NOT owned
346 UnicodeSet badCharacters;
347
348 public:
349
350 /*
351 * create a test for the given script transliterator.
352 */
353 RTTest(const UnicodeString& transliteratorIDStr);
354
355 virtual ~RTTest();
356
357 void setErrorLimit(int32_t limit);
358
359 void setPairLimit(int32_t limit);
360
361 void test(const UnicodeString& sourceRange,
362 const UnicodeString& targetRange,
363 const char* roundtripExclusions,
364 IntlTest* parent,
365 UBool quick,
366 Legal* adoptedLegal,
367 int32_t density = 100);
368
369 private:
370
371 // Added to do better equality check.
372
373 static UBool isSame(const UnicodeString& a, const UnicodeString& b);
374
375 static UBool isCamel(const UnicodeString& a);
376
377 UBool checkIrrelevants(Transliterator *t, const UnicodeString& irrelevants);
378
379 void test2(UBool quick, int32_t density);
380
381 void logWrongScript(const UnicodeString& label,
382 const UnicodeString& from,
383 const UnicodeString& to);
384
385 void logNotCanonical(const UnicodeString& label,
386 const UnicodeString& from,
387 const UnicodeString& to,
388 const UnicodeString& fromCan,
389 const UnicodeString& toCan);
390
391 void logFails(const UnicodeString& label);
392
393 void logToRulesFails(const UnicodeString& label,
394 const UnicodeString& from,
395 const UnicodeString& to,
396 const UnicodeString& toCan);
397
398 void logRoundTripFailure(const UnicodeString& from,
399 const UnicodeString& toID,
400 const UnicodeString& to,
401 const UnicodeString& backID,
402 const UnicodeString& back);
403 };
404
405 //--------------------------------------------------------------------
406 // RTTest Implementation
407 //--------------------------------------------------------------------
408
409 /*
410 * create a test for the given script transliterator.
411 */
RTTest(const UnicodeString & transliteratorIDStr)412 RTTest::RTTest(const UnicodeString& transliteratorIDStr) {
413 transliteratorID = transliteratorIDStr;
414 errorLimit = 500;
415 errorCount = 0;
416 pairLimit = 0x10000;
417 }
418
~RTTest()419 RTTest::~RTTest() {
420 }
421
setErrorLimit(int32_t limit)422 void RTTest::setErrorLimit(int32_t limit) {
423 errorLimit = limit;
424 }
425
setPairLimit(int32_t limit)426 void RTTest::setPairLimit(int32_t limit) {
427 pairLimit = limit;
428 }
429
isSame(const UnicodeString & a,const UnicodeString & b)430 UBool RTTest::isSame(const UnicodeString& a, const UnicodeString& b) {
431 if (a == b) return TRUE;
432 if (a.caseCompare(b, U_FOLD_CASE_DEFAULT)==0 && isCamel(a)) return TRUE;
433 UnicodeString aa, bb;
434 UErrorCode ec = U_ZERO_ERROR;
435 Normalizer::decompose(a, FALSE, 0, aa, ec);
436 Normalizer::decompose(b, FALSE, 0, bb, ec);
437 if (aa == bb) return TRUE;
438 if (aa.caseCompare(bb, U_FOLD_CASE_DEFAULT)==0 && isCamel(aa)) return TRUE;
439 return FALSE;
440 }
441
isCamel(const UnicodeString & a)442 UBool RTTest::isCamel(const UnicodeString& a) {
443 // see if string is of the form aB; e.g. lower, then upper or title
444 UChar32 cp;
445 UBool haveLower = FALSE;
446 for (int32_t i = 0; i < a.length(); i += U16_LENGTH(cp)) {
447 cp = a.char32At(i);
448 int8_t t = u_charType(cp);
449 switch (t) {
450 case U_UPPERCASE_LETTER:
451 if (haveLower) return TRUE;
452 break;
453 case U_TITLECASE_LETTER:
454 if (haveLower) return TRUE;
455 // fall through, since second letter is lower.
456 U_FALLTHROUGH;
457 case U_LOWERCASE_LETTER:
458 haveLower = TRUE;
459 break;
460 }
461 }
462 return FALSE;
463 }
464
test(const UnicodeString & sourceRangeVal,const UnicodeString & targetRangeVal,const char * roundtripExclusions,IntlTest * logVal,UBool quickRt,Legal * adoptedLegal,int32_t density)465 void RTTest::test(const UnicodeString& sourceRangeVal,
466 const UnicodeString& targetRangeVal,
467 const char* roundtripExclusions,
468 IntlTest* logVal, UBool quickRt,
469 Legal* adoptedLegal,
470 int32_t density)
471 {
472
473 UErrorCode status = U_ZERO_ERROR;
474
475 this->parent = logVal;
476 this->legalSource = adoptedLegal;
477
478 UnicodeSet neverOk("[:Other:]", status);
479 UnicodeSet okAnyway("[^[:Letter:]]", status);
480
481 if (U_FAILURE(status)) {
482 parent->dataerrln("FAIL: Initializing UnicodeSet with [:Other:] or [^[:Letter:]] - Error: %s", u_errorName(status));
483 return;
484 }
485
486 this->sourceRange.clear();
487 this->sourceRange.applyPattern(sourceRangeVal, status);
488 if (U_FAILURE(status)) {
489 parent->errln("FAIL: UnicodeSet::applyPattern(" +
490 sourceRangeVal + ")");
491 return;
492 }
493 this->sourceRange.removeAll(neverOk);
494
495 this->targetRange.clear();
496 this->targetRange.applyPattern(targetRangeVal, status);
497 if (U_FAILURE(status)) {
498 parent->errln("FAIL: UnicodeSet::applyPattern(" +
499 targetRangeVal + ")");
500 return;
501 }
502 this->targetRange.removeAll(neverOk);
503
504 this->toSource.clear();
505 this->toSource.applyPattern(sourceRangeVal, status);
506 if (U_FAILURE(status)) {
507 parent->errln("FAIL: UnicodeSet::applyPattern(" +
508 sourceRangeVal + ")");
509 return;
510 }
511 this->toSource.addAll(okAnyway);
512
513 this->toTarget.clear();
514 this->toTarget.applyPattern(targetRangeVal, status);
515 if (U_FAILURE(status)) {
516 parent->errln("FAIL: UnicodeSet::applyPattern(" +
517 targetRangeVal + ")");
518 return;
519 }
520 this->toTarget.addAll(okAnyway);
521
522 this->roundtripExclusionsSet.clear();
523 if (roundtripExclusions != NULL && strlen(roundtripExclusions) > 0) {
524 this->roundtripExclusionsSet.applyPattern(UnicodeString(roundtripExclusions, -1, US_INV), status);
525 if (U_FAILURE(status)) {
526 parent->errln("FAIL: UnicodeSet::applyPattern(%s)", roundtripExclusions);
527 return;
528 }
529 }
530
531 badCharacters.clear();
532 badCharacters.applyPattern("[:Other:]", status);
533 if (U_FAILURE(status)) {
534 parent->errln("FAIL: UnicodeSet::applyPattern([:Other:])");
535 return;
536 }
537
538 test2(quickRt, density);
539
540 if (errorCount > 0) {
541 char str[100];
542 int32_t length = transliteratorID.extract(str, 100, NULL, status);
543 str[length] = 0;
544 parent->errln("FAIL: %s errors: %d %s", str, errorCount, (errorCount > errorLimit ? " (at least!)" : " ")); // + ", see " + logFileName);
545 } else {
546 char str[100];
547 int32_t length = transliteratorID.extract(str, 100, NULL, status);
548 str[length] = 0;
549 parent->logln("%s ok", str);
550 }
551 }
552
checkIrrelevants(Transliterator * t,const UnicodeString & irrelevants)553 UBool RTTest::checkIrrelevants(Transliterator *t,
554 const UnicodeString& irrelevants) {
555 for (int i = 0; i < irrelevants.length(); ++i) {
556 UChar c = irrelevants.charAt(i);
557 UnicodeString srcStr(c);
558 UnicodeString targ = srcStr;
559 t->transliterate(targ);
560 if (srcStr == targ) return TRUE;
561 }
562 return FALSE;
563 }
564
test2(UBool quickRt,int32_t density)565 void RTTest::test2(UBool quickRt, int32_t density) {
566
567 UnicodeString srcStr, targ, reverse;
568 UErrorCode status = U_ZERO_ERROR;
569 UParseError parseError ;
570 TransliteratorPointer sourceToTarget(
571 Transliterator::createInstance(transliteratorID, UTRANS_FORWARD, parseError,
572 status));
573 if ((Transliterator *)sourceToTarget == NULL) {
574 parent->dataerrln("FAIL: createInstance(" + transliteratorID +
575 ") returned NULL. Error: " + u_errorName(status)
576 + "\n\tpreContext : " + prettify(parseError.preContext)
577 + "\n\tpostContext : " + prettify(parseError.postContext));
578
579 return;
580 }
581 TransliteratorPointer targetToSource(sourceToTarget->createInverse(status));
582 if ((Transliterator *)targetToSource == NULL) {
583 parent->errln("FAIL: " + transliteratorID +
584 ".createInverse() returned NULL. Error:" + u_errorName(status)
585 + "\n\tpreContext : " + prettify(parseError.preContext)
586 + "\n\tpostContext : " + prettify(parseError.postContext));
587 return;
588 }
589
590 AbbreviatedUnicodeSetIterator usi;
591 AbbreviatedUnicodeSetIterator usi2;
592
593 parent->logln("Checking that at least one irrelevant character is not NFC'ed");
594 // string is from NFC_NO in the UCD
595 UnicodeString irrelevants = CharsToUnicodeString("\\u2000\\u2001\\u2126\\u212A\\u212B\\u2329");
596
597 if (checkIrrelevants(sourceToTarget, irrelevants) == FALSE) {
598 logFails("Source-Target, irrelevants");
599 }
600 if (checkIrrelevants(targetToSource, irrelevants) == FALSE) {
601 logFails("Target-Source, irrelevants");
602 }
603
604 if (!quickRt){
605 parent->logln("Checking that toRules works");
606 UnicodeString rules = "";
607
608 UParseError parseError;
609 rules = sourceToTarget->toRules(rules, TRUE);
610 // parent->logln((UnicodeString)"toRules => " + rules);
611 TransliteratorPointer sourceToTarget2(Transliterator::createFromRules(
612 "s2t2", rules,
613 UTRANS_FORWARD,
614 parseError, status));
615 if (U_FAILURE(status)) {
616 parent->errln("FAIL: createFromRules %s\n", u_errorName(status));
617 return;
618 }
619
620 rules = targetToSource->toRules(rules, FALSE);
621 TransliteratorPointer targetToSource2(Transliterator::createFromRules(
622 "t2s2", rules,
623 UTRANS_FORWARD,
624 parseError, status));
625 if (U_FAILURE(status)) {
626 parent->errln("FAIL: createFromRules %s\n", u_errorName(status));
627 return;
628 }
629
630 usi.reset(sourceRange);
631 for (;;) {
632 if (!usi.next() || usi.isString()) break;
633 UChar32 c = usi.getCodepoint();
634
635 UnicodeString srcStr((UChar32)c);
636 UnicodeString targ = srcStr;
637 sourceToTarget->transliterate(targ);
638 UnicodeString targ2 = srcStr;
639 sourceToTarget2->transliterate(targ2);
640 if (targ != targ2) {
641 logToRulesFails("Source-Target, toRules", srcStr, targ, targ2);
642 }
643 }
644
645 usi.reset(targetRange);
646 for (;;) {
647 if (!usi.next() || usi.isString()) break;
648 UChar32 c = usi.getCodepoint();
649
650 UnicodeString srcStr((UChar32)c);
651 UnicodeString targ = srcStr;
652 targetToSource->transliterate(targ);
653 UnicodeString targ2 = srcStr;
654 targetToSource2->transliterate(targ2);
655 if (targ != targ2) {
656 logToRulesFails("Target-Source, toRules", srcStr, targ, targ2);
657 }
658 }
659 }
660
661 parent->logln("Checking that all source characters convert to target - Singles");
662
663 UnicodeSet failSourceTarg;
664 usi.reset(sourceRange);
665 for (;;) {
666 if (!usi.next() || usi.isString()) break;
667 UChar32 c = usi.getCodepoint();
668
669 UnicodeString srcStr((UChar32)c);
670 UnicodeString targ = srcStr;
671 sourceToTarget->transliterate(targ);
672 if (toTarget.containsAll(targ) == FALSE
673 || badCharacters.containsSome(targ) == TRUE) {
674 UnicodeString targD;
675 Normalizer::decompose(targ, FALSE, 0, targD, status);
676 if (U_FAILURE(status)) {
677 parent->errln("FAIL: Internal error during decomposition %s\n", u_errorName(status));
678 return;
679 }
680 if (toTarget.containsAll(targD) == FALSE ||
681 badCharacters.containsSome(targD) == TRUE) {
682 logWrongScript("Source-Target", srcStr, targ);
683 failSourceTarg.add(c);
684 continue;
685 }
686 }
687
688 UnicodeString cs2;
689 Normalizer::decompose(srcStr, FALSE, 0, cs2, status);
690 if (U_FAILURE(status)) {
691 parent->errln("FAIL: Internal error during decomposition %s\n", u_errorName(status));
692 return;
693 }
694 UnicodeString targ2 = cs2;
695 sourceToTarget->transliterate(targ2);
696 if (targ != targ2) {
697 logNotCanonical("Source-Target", srcStr, targ,cs2, targ2);
698 }
699 }
700
701 parent->logln("Checking that all source characters convert to target - Doubles");
702
703 UnicodeSet sourceRangeMinusFailures(sourceRange);
704 sourceRangeMinusFailures.removeAll(failSourceTarg);
705
706 usi.reset(sourceRangeMinusFailures, quickRt, density);
707 for (;;) {
708 if (!usi.next() || usi.isString()) break;
709 UChar32 c = usi.getCodepoint();
710
711 usi2.reset(sourceRangeMinusFailures, quickRt, density);
712 for (;;) {
713 if (!usi2.next() || usi2.isString()) break;
714 UChar32 d = usi2.getCodepoint();
715
716 UnicodeString srcStr;
717 srcStr += (UChar32)c;
718 srcStr += (UChar32)d;
719 UnicodeString targ = srcStr;
720 sourceToTarget->transliterate(targ);
721 if (toTarget.containsAll(targ) == FALSE ||
722 badCharacters.containsSome(targ) == TRUE)
723 {
724 UnicodeString targD;
725 Normalizer::decompose(targ, FALSE, 0, targD, status);
726 if (U_FAILURE(status)) {
727 parent->errln("FAIL: Internal error during decomposition %s\n", u_errorName(status));
728 return;
729 }
730 if (toTarget.containsAll(targD) == FALSE ||
731 badCharacters.containsSome(targD) == TRUE) {
732 logWrongScript("Source-Target", srcStr, targ);
733 continue;
734 }
735 }
736 UnicodeString cs2;
737 Normalizer::decompose(srcStr, FALSE, 0, cs2, status);
738 if (U_FAILURE(status)) {
739 parent->errln("FAIL: Internal error during decomposition %s\n", u_errorName(status));
740 return;
741 }
742 UnicodeString targ2 = cs2;
743 sourceToTarget->transliterate(targ2);
744 if (targ != targ2) {
745 logNotCanonical("Source-Target", srcStr, targ, cs2,targ2);
746 }
747 }
748 }
749
750 parent->logln("Checking that target characters convert to source and back - Singles");
751
752 UnicodeSet failTargSource;
753 UnicodeSet failRound;
754
755 usi.reset(targetRange);
756 for (;;) {
757 if (!usi.next()) break;
758
759 if(usi.isString()){
760 srcStr = usi.getString();
761 }else{
762 srcStr = (UnicodeString)usi.getCodepoint();
763 }
764
765 UChar32 c = srcStr.char32At(0);
766
767 targ = srcStr;
768 targetToSource->transliterate(targ);
769 reverse = targ;
770 sourceToTarget->transliterate(reverse);
771
772 if (toSource.containsAll(targ) == FALSE ||
773 badCharacters.containsSome(targ) == TRUE) {
774 UnicodeString targD;
775 Normalizer::decompose(targ, FALSE, 0, targD, status);
776 if (U_FAILURE(status)) {
777 parent->errln("FAIL: Internal error during decomposition%s\n", u_errorName(status));
778 return;
779 }
780 if (toSource.containsAll(targD) == FALSE) {
781 logWrongScript("Target-Source", srcStr, targ);
782 failTargSource.add(c);
783 continue;
784 }
785 if (badCharacters.containsSome(targD) == TRUE) {
786 logWrongScript("Target-Source*", srcStr, targ);
787 failTargSource.add(c);
788 continue;
789 }
790 }
791 if (isSame(srcStr, reverse) == FALSE &&
792 roundtripExclusionsSet.contains(c) == FALSE
793 && roundtripExclusionsSet.contains(srcStr)==FALSE) {
794 logRoundTripFailure(srcStr,targetToSource->getID(), targ,sourceToTarget->getID(), reverse);
795 failRound.add(c);
796 continue;
797 }
798
799 UnicodeString targ2;
800 Normalizer::decompose(targ, FALSE, 0, targ2, status);
801 if (U_FAILURE(status)) {
802 parent->errln("FAIL: Internal error during decomposition%s\n", u_errorName(status));
803 return;
804 }
805 UnicodeString reverse2 = targ2;
806 sourceToTarget->transliterate(reverse2);
807 if (reverse != reverse2) {
808 logNotCanonical("Target-Source", targ, reverse, targ2, reverse2);
809 }
810 }
811
812 parent->logln("Checking that target characters convert to source and back - Doubles");
813 int32_t count = 0;
814
815 UnicodeSet targetRangeMinusFailures(targetRange);
816 targetRangeMinusFailures.removeAll(failTargSource);
817 targetRangeMinusFailures.removeAll(failRound);
818
819 usi.reset(targetRangeMinusFailures, quickRt, density);
820 UnicodeString targ2;
821 UnicodeString reverse2;
822 UnicodeString targD;
823 for (;;) {
824 if (!usi.next() || usi.isString()) break;
825 UChar32 c = usi.getCodepoint();
826 if (++count > pairLimit) {
827 //throw new TestTruncated("Test truncated at " + pairLimit + " x 64k pairs");
828 parent->logln("");
829 parent->logln((UnicodeString)"Test truncated at " + pairLimit + " x 64k pairs");
830 return;
831 }
832
833 usi2.reset(targetRangeMinusFailures, quickRt, density);
834 for (;;) {
835 if (!usi2.next() || usi2.isString())
836 break;
837 UChar32 d = usi2.getCodepoint();
838 srcStr.truncate(0); // empty the variable without construction/destruction
839 srcStr += c;
840 srcStr += d;
841
842 targ = srcStr;
843 targetToSource->transliterate(targ);
844 reverse = targ;
845 sourceToTarget->transliterate(reverse);
846
847 if (toSource.containsAll(targ) == FALSE ||
848 badCharacters.containsSome(targ) == TRUE)
849 {
850 targD.truncate(0); // empty the variable without construction/destruction
851 Normalizer::decompose(targ, FALSE, 0, targD, status);
852 if (U_FAILURE(status)) {
853 parent->errln("FAIL: Internal error during decomposition%s\n",
854 u_errorName(status));
855 return;
856 }
857 if (toSource.containsAll(targD) == FALSE
858 || badCharacters.containsSome(targD) == TRUE)
859 {
860 logWrongScript("Target-Source", srcStr, targ);
861 continue;
862 }
863 }
864 if (isSame(srcStr, reverse) == FALSE &&
865 roundtripExclusionsSet.contains(c) == FALSE&&
866 roundtripExclusionsSet.contains(d) == FALSE &&
867 roundtripExclusionsSet.contains(srcStr)== FALSE)
868 {
869 logRoundTripFailure(srcStr,targetToSource->getID(), targ, sourceToTarget->getID(),reverse);
870 continue;
871 }
872
873 targ2.truncate(0); // empty the variable without construction/destruction
874 Normalizer::decompose(targ, FALSE, 0, targ2, status);
875 if (U_FAILURE(status)) {
876 parent->errln("FAIL: Internal error during decomposition%s\n", u_errorName(status));
877 return;
878 }
879 reverse2 = targ2;
880 sourceToTarget->transliterate(reverse2);
881 if (reverse != reverse2) {
882 logNotCanonical("Target-Source", targ,reverse, targ2, reverse2);
883 }
884 }
885 }
886 parent->logln("");
887 }
888
logWrongScript(const UnicodeString & label,const UnicodeString & from,const UnicodeString & to)889 void RTTest::logWrongScript(const UnicodeString& label,
890 const UnicodeString& from,
891 const UnicodeString& to) {
892 parent->errln((UnicodeString)"FAIL " +
893 label + ": " +
894 from + "(" + TestUtility::hex(from) + ") => " +
895 to + "(" + TestUtility::hex(to) + ")");
896 ++errorCount;
897 }
898
logNotCanonical(const UnicodeString & label,const UnicodeString & from,const UnicodeString & to,const UnicodeString & fromCan,const UnicodeString & toCan)899 void RTTest::logNotCanonical(const UnicodeString& label,
900 const UnicodeString& from,
901 const UnicodeString& to,
902 const UnicodeString& fromCan,
903 const UnicodeString& toCan) {
904 parent->errln((UnicodeString)"FAIL (can.equiv)" +
905 label + ": " +
906 from + "(" + TestUtility::hex(from) + ") => " +
907 to + "(" + TestUtility::hex(to) + ")" +
908 fromCan + "(" + TestUtility::hex(fromCan) + ") => " +
909 toCan + " (" +
910 TestUtility::hex(toCan) + ")"
911 );
912 ++errorCount;
913 }
914
logFails(const UnicodeString & label)915 void RTTest::logFails(const UnicodeString& label) {
916 parent->errln((UnicodeString)"<br>FAIL " + label);
917 ++errorCount;
918 }
919
logToRulesFails(const UnicodeString & label,const UnicodeString & from,const UnicodeString & to,const UnicodeString & otherTo)920 void RTTest::logToRulesFails(const UnicodeString& label,
921 const UnicodeString& from,
922 const UnicodeString& to,
923 const UnicodeString& otherTo)
924 {
925 parent->errln((UnicodeString)"FAIL: " +
926 label + ": " +
927 from + "(" + TestUtility::hex(from) + ") => " +
928 to + "(" + TestUtility::hex(to) + ")" +
929 "!=" +
930 otherTo + " (" +
931 TestUtility::hex(otherTo) + ")"
932 );
933 ++errorCount;
934 }
935
936
logRoundTripFailure(const UnicodeString & from,const UnicodeString & toID,const UnicodeString & to,const UnicodeString & backID,const UnicodeString & back)937 void RTTest::logRoundTripFailure(const UnicodeString& from,
938 const UnicodeString& toID,
939 const UnicodeString& to,
940 const UnicodeString& backID,
941 const UnicodeString& back) {
942 if (legalSource->is(from) == FALSE) return; // skip illegals
943
944 parent->errln((UnicodeString)"FAIL Roundtrip: " +
945 from + "(" + TestUtility::hex(from) + ") => " +
946 to + "(" + TestUtility::hex(to) + ") "+toID+" => " +
947 back + "(" + TestUtility::hex(back) + ") "+backID+" => ");
948 ++errorCount;
949 }
950
951 //--------------------------------------------------------------------
952 // Specific Tests
953 //--------------------------------------------------------------------
954
955 /*
956 Note: Unicode 3.2 added new Hiragana/Katakana characters:
957
958 3095..3096 ; 3.2 # [2] HIRAGANA LETTER SMALL KA..HIRAGANA LETTER SMALL KE
959 309F..30A0 ; 3.2 # [2] HIRAGANA DIGRAPH YORI..KATAKANA-HIRAGANA DOUBLE HYPHEN
960 30FF ; 3.2 # KATAKANA DIGRAPH KOTO
961 31F0..31FF ; 3.2 # [16] KATAKANA LETTER SMALL KU..KATAKANA LETTER SMALL RO
962
963 Unicode 5.2 added another Hiragana character:
964 1F200 ; 5.2 # SQUARE HIRAGANA HOKA
965
966 We will not add them to the rules until they are more supported (e.g. in fonts on Windows)
967 A bug has been filed to remind us to do this: #1979.
968 */
969
970 static const char KATAKANA[] = "[[[:katakana:][\\u30A1-\\u30FA\\u30FC]]-[\\u30FF\\u31F0-\\u31FF]-[:^age=5.2:]]";
971 static const char HIRAGANA[] = "[[[:hiragana:][\\u3040-\\u3094]]-[\\u3095-\\u3096\\u309F-\\u30A0\\U0001F200-\\U0001F2FF]-[:^age=5.2:]]";
972 static const char LENGTH[] = "[\\u30FC]";
973 static const char HALFWIDTH_KATAKANA[] = "[\\uFF65-\\uFF9D]";
974 static const char KATAKANA_ITERATION[] = "[\\u30FD\\u30FE]";
975 static const char HIRAGANA_ITERATION[] = "[\\u309D\\u309E]";
976 static const int32_t TEMP_MAX=256;
977
TestKana()978 void TransliteratorRoundTripTest::TestKana() {
979 RTTest test("Katakana-Hiragana");
980 Legal *legal = new Legal();
981 char temp[TEMP_MAX];
982 strcpy(temp, "[");
983 strcat(temp, HALFWIDTH_KATAKANA);
984 strcat(temp, LENGTH);
985 strcat(temp, "]");
986 test.test(KATAKANA, UnicodeString("[") + HIRAGANA + LENGTH + UnicodeString("]"),
987 temp,
988 this, quick, legal);
989 delete legal;
990 }
991
TestHiragana()992 void TransliteratorRoundTripTest::TestHiragana() {
993 RTTest test("Latin-Hiragana");
994 Legal *legal = new Legal();
995 test.test(UnicodeString("[a-zA-Z]", ""),
996 UnicodeString(HIRAGANA, -1, US_INV),
997 HIRAGANA_ITERATION, this, quick, legal);
998 delete legal;
999 }
1000
TestKatakana()1001 void TransliteratorRoundTripTest::TestKatakana() {
1002 RTTest test("Latin-Katakana");
1003 Legal *legal = new Legal();
1004 char temp[TEMP_MAX];
1005 strcpy(temp, "[");
1006 strcat(temp, KATAKANA_ITERATION);
1007 strcat(temp, HALFWIDTH_KATAKANA);
1008 strcat(temp, "]");
1009 test.test(UnicodeString("[a-zA-Z]", ""),
1010 UnicodeString(KATAKANA, -1, US_INV),
1011 temp,
1012 this, quick, legal);
1013 delete legal;
1014 }
1015
TestJamo()1016 void TransliteratorRoundTripTest::TestJamo() {
1017 RTTest t("Latin-Jamo");
1018 Legal *legal = new LegalJamo();
1019 t.test(UnicodeString("[a-zA-Z]", ""),
1020 UnicodeString("[\\u1100-\\u1112 \\u1161-\\u1175 \\u11A8-\\u11C2]",
1021 ""),
1022 NULL, this, quick, legal);
1023 delete legal;
1024 }
1025
TestHangul()1026 void TransliteratorRoundTripTest::TestHangul() {
1027 RTTest t("Latin-Hangul");
1028 Legal *legal = new Legal();
1029 if (quick) t.setPairLimit(1000);
1030 t.test(UnicodeString("[a-zA-Z]", ""),
1031 UnicodeString("[\\uAC00-\\uD7A4]", ""),
1032 NULL, this, quick, legal, 1);
1033 delete legal;
1034 }
1035
1036
1037 #define ASSERT_SUCCESS(status) UPRV_BLOCK_MACRO_BEGIN { \
1038 if (U_FAILURE(status)) { \
1039 errcheckln(status, "error at file %s, line %d, status = %s", __FILE__, __LINE__, \
1040 u_errorName(status)); \
1041 return; \
1042 } \
1043 } UPRV_BLOCK_MACRO_END
1044
1045
writeStringInU8(FILE * out,const UnicodeString & s)1046 static void writeStringInU8(FILE *out, const UnicodeString &s) {
1047 int i;
1048 for (i=0; i<s.length(); i=s.moveIndex32(i, 1)) {
1049 UChar32 c = s.char32At(i);
1050 uint8_t bufForOneChar[10];
1051 UBool isError = FALSE;
1052 int32_t destIdx = 0;
1053 U8_APPEND(bufForOneChar, destIdx, (int32_t)sizeof(bufForOneChar), c, isError);
1054 U_ASSERT(!isError);
1055 (void)isError;
1056 fwrite(bufForOneChar, 1, destIdx, out);
1057 }
1058 }
1059
1060
1061
1062
TestHan()1063 void TransliteratorRoundTripTest::TestHan() {
1064 UErrorCode status = U_ZERO_ERROR;
1065 LocalULocaleDataPointer uld(ulocdata_open("zh",&status));
1066 LocalUSetPointer USetExemplars(uset_openEmpty());
1067 assertTrue("", USetExemplars.isValid(), false, false, __FILE__, __LINE__);
1068 if (! USetExemplars.isValid()) return;
1069 ulocdata_getExemplarSet(uld.getAlias(), USetExemplars.getAlias(), 0, ULOCDATA_ES_STANDARD, &status);
1070 ASSERT_SUCCESS(status);
1071
1072 UnicodeString source;
1073 UChar32 c;
1074 int i;
1075 for (i=0; ;i++) {
1076 // Add all of the Chinese exemplar chars to the string "source".
1077 c = uset_charAt(USetExemplars.getAlias(), i);
1078 if (c == (UChar32)-1) {
1079 break;
1080 }
1081 source.append(c);
1082 }
1083
1084 // transform with Han translit
1085 Transliterator *hanTL = Transliterator::createInstance("Han-Latin", UTRANS_FORWARD, status);
1086 ASSERT_SUCCESS(status);
1087 UnicodeString target=source;
1088 hanTL->transliterate(target);
1089 // now verify that there are no Han characters left
1090 UnicodeSet allHan("[:han:]", status);
1091 ASSERT_SUCCESS(status);
1092 if (allHan.containsSome(target)) {
1093 errln("file %s, line %d, No Han must be left after Han-Latin transliteration",
1094 __FILE__, __LINE__);
1095 }
1096
1097 // check the pinyin translit
1098 Transliterator *pn = Transliterator::createInstance("Latin-NumericPinyin", UTRANS_FORWARD, status);
1099 ASSERT_SUCCESS(status);
1100 UnicodeString target2 = target;
1101 pn->transliterate(target2);
1102
1103 // verify that there are no marks
1104 Transliterator *nfd = Transliterator::createInstance("nfd", UTRANS_FORWARD, status);
1105 ASSERT_SUCCESS(status);
1106
1107 UnicodeString nfded = target2;
1108 nfd->transliterate(nfded);
1109 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)
1110 ASSERT_SUCCESS(status);
1111 assertFalse("NumericPinyin must contain no marks", allMarks.containsSome(nfded));
1112
1113 // verify roundtrip
1114 Transliterator *np = pn->createInverse(status);
1115 ASSERT_SUCCESS(status);
1116 UnicodeString target3 = target2;
1117 np->transliterate(target3);
1118 UBool roundtripOK = (target3.compare(target) == 0);
1119 assertTrue("NumericPinyin must roundtrip", roundtripOK);
1120 if (!roundtripOK) {
1121 const char *filename = "numeric-pinyin.log.txt";
1122 FILE *out = fopen(filename, "w");
1123 errln("Creating log file %s\n", filename);
1124 fprintf(out, "Pinyin: ");
1125 writeStringInU8(out, target);
1126 fprintf(out, "\nPinyin-Numeric-Pinyin: ");
1127 writeStringInU8(out, target2);
1128 fprintf(out, "\nNumeric-Pinyin-Pinyin: ");
1129 writeStringInU8(out, target3);
1130 fprintf(out, "\n");
1131 fclose(out);
1132 }
1133
1134 delete hanTL;
1135 delete pn;
1136 delete nfd;
1137 delete np;
1138 }
1139
1140
TestGreek()1141 void TransliteratorRoundTripTest::TestGreek() {
1142 logKnownIssue( "cldrbug:1911");
1143 // It is left in its current state as a regression test.
1144
1145 RTTest test("Latin-Greek");
1146 LegalGreek *legal = new LegalGreek(TRUE);
1147
1148 test.test(UnicodeString("[a-zA-Z]", ""),
1149 UnicodeString("[\\u003B\\u00B7[[:Greek:]&[:Letter:]]-["
1150 "\\u1D26-\\u1D2A" // L& [5] GREEK LETTER SMALL CAPITAL GAMMA..GREEK LETTER SMALL CAPITAL PSI
1151 "\\u1D5D-\\u1D61" // Lm [5] MODIFIER LETTER SMALL BETA..MODIFIER LETTER SMALL CHI
1152 "\\u1D66-\\u1D6A" // L& [5] GREEK SUBSCRIPT SMALL LETTER BETA..GREEK SUBSCRIPT SMALL LETTER CHI
1153 "\\u03D7-\\u03EF" // \N{GREEK KAI SYMBOL}..\N{COPTIC SMALL LETTER DEI}
1154 "] & [:Age=4.0:]]",
1155
1156 //UnicodeString("[[\\u003B\\u00B7[:Greek:]-[\\u0374\\u0385\\u1fcd\\u1fce\\u1fdd\\u1fde\\u1fed-\\u1fef\\u1ffd\\u03D7-\\u03EF]]&[:Age=3.2:]]",
1157 ""),
1158 "[\\u00B5\\u037A\\u03D0-\\u03F5\\u03f9]", /* exclusions */
1159 this, quick, legal, 50);
1160
1161
1162 delete legal;
1163 }
1164
1165
TestGreekUNGEGN()1166 void TransliteratorRoundTripTest::TestGreekUNGEGN() {
1167 logKnownIssue( "cldrbug:1911");
1168 // It is left in its current state as a regression test.
1169
1170 RTTest test("Latin-Greek/UNGEGN");
1171 LegalGreek *legal = new LegalGreek(FALSE);
1172
1173 test.test(UnicodeString("[a-zA-Z]", ""),
1174 UnicodeString("[\\u003B\\u00B7[[:Greek:]&[:Letter:]]-["
1175 "\\u1D26-\\u1D2A" // L& [5] GREEK LETTER SMALL CAPITAL GAMMA..GREEK LETTER SMALL CAPITAL PSI
1176 "\\u1D5D-\\u1D61" // Lm [5] MODIFIER LETTER SMALL BETA..MODIFIER LETTER SMALL CHI
1177 "\\u1D66-\\u1D6A" // L& [5] GREEK SUBSCRIPT SMALL LETTER BETA..GREEK SUBSCRIPT SMALL LETTER CHI
1178 "\\u03D7-\\u03EF" // \N{GREEK KAI SYMBOL}..\N{COPTIC SMALL LETTER DEI}
1179 "] & [:Age=4.0:]]",
1180 //UnicodeString("[[\\u003B\\u00B7[:Greek:]-[\\u0374\\u0385\\u1fce\\u1fde\\u03D7-\\u03EF]]&[:Age=3.2:]]",
1181 ""),
1182 "[\\u0385\\u00B5\\u037A\\u03D0-\\uFFFF {\\u039C\\u03C0}]", /* roundtrip exclusions */
1183 this, quick, legal);
1184
1185 delete legal;
1186 }
1187
Testel()1188 void TransliteratorRoundTripTest::Testel() {
1189 logKnownIssue( "cldrbug:1911");
1190 // It is left in its current state as a regression test.
1191
1192 RTTest test("Latin-el");
1193 LegalGreek *legal = new LegalGreek(FALSE);
1194
1195 test.test(UnicodeString("[a-zA-Z]", ""),
1196 UnicodeString("[\\u003B\\u00B7[[:Greek:]&[:Letter:]]-["
1197 "\\u1D26-\\u1D2A" // L& [5] GREEK LETTER SMALL CAPITAL GAMMA..GREEK LETTER SMALL CAPITAL PSI
1198 "\\u1D5D-\\u1D61" // Lm [5] MODIFIER LETTER SMALL BETA..MODIFIER LETTER SMALL CHI
1199 "\\u1D66-\\u1D6A" // L& [5] GREEK SUBSCRIPT SMALL LETTER BETA..GREEK SUBSCRIPT SMALL LETTER CHI
1200 "\\u03D7-\\u03EF" // \N{GREEK KAI SYMBOL}..\N{COPTIC SMALL LETTER DEI}
1201 "] & [:Age=4.0:]]",
1202 //UnicodeString("[[\\u003B\\u00B7[:Greek:]-[\\u0374\\u0385\\u1fce\\u1fde\\u03D7-\\u03EF]]&[:Age=3.2:]]",
1203 ""),
1204 "[\\u00B5\\u037A\\u03D0-\\uFFFF {\\u039C\\u03C0}]", /* exclusions */
1205 this, quick, legal);
1206
1207
1208 delete legal;
1209 }
1210
1211
TestArabic()1212 void TransliteratorRoundTripTest::TestArabic() {
1213 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);
1214 Legal *legal = new Legal();
1215 RTTest test("Latin-Arabic");
1216 test.test(UNICODE_STRING_SIMPLE("[a-zA-Z\\u02BE\\u02BF\\u207F]"), ARABIC, "[a-zA-Z\\u02BE\\u02BF\\u207F]",this, quick, legal); //
1217 delete legal;
1218 }
1219 class LegalHebrew : public Legal {
1220 private:
1221 UnicodeSet FINAL;
1222 UnicodeSet NON_FINAL;
1223 UnicodeSet LETTER;
1224 public:
1225 LegalHebrew(UErrorCode& error);
~LegalHebrew()1226 virtual ~LegalHebrew() {}
1227 virtual UBool is(const UnicodeString& sourceString) const;
1228 };
1229
LegalHebrew(UErrorCode & error)1230 LegalHebrew::LegalHebrew(UErrorCode& error){
1231 FINAL.applyPattern(UNICODE_STRING_SIMPLE("[\\u05DA\\u05DD\\u05DF\\u05E3\\u05E5]"), error);
1232 NON_FINAL.applyPattern(UNICODE_STRING_SIMPLE("[\\u05DB\\u05DE\\u05E0\\u05E4\\u05E6]"), error);
1233 LETTER.applyPattern("[:letter:]", error);
1234 }
is(const UnicodeString & sourceString) const1235 UBool LegalHebrew::is(const UnicodeString& sourceString)const{
1236
1237 if (sourceString.length() == 0) return TRUE;
1238 // don't worry about surrogates.
1239 for (int i = 0; i < sourceString.length(); ++i) {
1240 UChar ch = sourceString.charAt(i);
1241 UChar next = i+1 == sourceString.length() ? 0x0000 : sourceString.charAt(i);
1242 if (FINAL.contains(ch)) {
1243 if (LETTER.contains(next)) return FALSE;
1244 } else if (NON_FINAL.contains(ch)) {
1245 if (!LETTER.contains(next)) return FALSE;
1246 }
1247 }
1248 return TRUE;
1249 }
TestHebrew()1250 void TransliteratorRoundTripTest::TestHebrew() {
1251 logKnownIssue( "cldrbug:1911");
1252 // It is left in its current state as a regression test.
1253
1254 //long start = System.currentTimeMillis();
1255 UErrorCode error = U_ZERO_ERROR;
1256 LegalHebrew* legal = new LegalHebrew(error);
1257 if(U_FAILURE(error)){
1258 dataerrln("Could not construct LegalHebrew object. Error: %s", u_errorName(error));
1259 return;
1260 }
1261 RTTest test("Latin-Hebrew");
1262 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);
1263
1264 //showElapsed(start, "TestHebrew");
1265 delete legal;
1266 }
TestCyrillic()1267 void TransliteratorRoundTripTest::TestCyrillic() {
1268 RTTest test("Latin-Cyrillic");
1269 Legal *legal = new Legal();
1270
1271 test.test(UnicodeString("[a-zA-Z\\u0110\\u0111\\u02BA\\u02B9]", ""),
1272 UnicodeString("[[\\u0400-\\u045F] & [:Age=3.2:]]", ""), NULL, this, quick,
1273 legal);
1274
1275 delete legal;
1276 }
1277
1278
1279 // Inter-Indic Tests ----------------------------------
1280 class LegalIndic :public Legal{
1281 UnicodeSet vowelSignSet;
1282 UnicodeSet avagraha;
1283 UnicodeSet nukta;
1284 UnicodeSet virama;
1285 UnicodeSet sanskritStressSigns;
1286 UnicodeSet chandrabindu;
1287
1288 public:
1289 LegalIndic();
1290 virtual UBool is(const UnicodeString& sourceString) const;
~LegalIndic()1291 virtual ~LegalIndic() {}
1292 };
is(const UnicodeString & sourceString) const1293 UBool LegalIndic::is(const UnicodeString& sourceString) const{
1294 int cp=sourceString.charAt(0);
1295
1296 // A vowel sign cannot be the first char
1297 if(vowelSignSet.contains(cp)){
1298 return FALSE;
1299 }else if(avagraha.contains(cp)){
1300 return FALSE;
1301 }else if(virama.contains(cp)){
1302 return FALSE;
1303 }else if(nukta.contains(cp)){
1304 return FALSE;
1305 }else if(sanskritStressSigns.contains(cp)){
1306 return FALSE;
1307 }else if(chandrabindu.contains(cp) &&
1308 ((sourceString.length()>1) &&
1309 vowelSignSet.contains(sourceString.charAt(1)))){
1310 return FALSE;
1311 }
1312 return TRUE;
1313 }
LegalIndic()1314 LegalIndic::LegalIndic(){
1315 UErrorCode status = U_ZERO_ERROR;
1316 vowelSignSet.addAll( UnicodeSet("[\\u0902\\u0903\\u0904\\u093e-\\u094c\\u0962\\u0963]",status));/* Devanagari */
1317 vowelSignSet.addAll( UnicodeSet("[\\u0982\\u0983\\u09be-\\u09cc\\u09e2\\u09e3\\u09D7]",status));/* Bengali */
1318 vowelSignSet.addAll( UnicodeSet("[\\u0a02\\u0a03\\u0a3e-\\u0a4c\\u0a62\\u0a63\\u0a70\\u0a71]",status));/* Gurmukhi */
1319 vowelSignSet.addAll( UnicodeSet("[\\u0a82\\u0a83\\u0abe-\\u0acc\\u0ae2\\u0ae3]",status));/* Gujarati */
1320 vowelSignSet.addAll( UnicodeSet("[\\u0b02\\u0b03\\u0b3e-\\u0b4c\\u0b62\\u0b63\\u0b56\\u0b57]",status));/* Oriya */
1321 vowelSignSet.addAll( UnicodeSet("[\\u0b82\\u0b83\\u0bbe-\\u0bcc\\u0be2\\u0be3\\u0bd7]",status));/* Tamil */
1322 vowelSignSet.addAll( UnicodeSet("[\\u0c02\\u0c03\\u0c3e-\\u0c4c\\u0c62\\u0c63\\u0c55\\u0c56]",status));/* Telugu */
1323 vowelSignSet.addAll( UnicodeSet("[\\u0c82\\u0c83\\u0cbe-\\u0ccc\\u0ce2\\u0ce3\\u0cd5\\u0cd6]",status));/* Kannada */
1324 vowelSignSet.addAll( UnicodeSet("[\\u0d02\\u0d03\\u0d3e-\\u0d4c\\u0d62\\u0d63\\u0d57]",status));/* Malayalam */
1325
1326 avagraha.addAll(UnicodeSet("[\\u093d\\u09bd\\u0abd\\u0b3d\\u0cbd]",status));
1327 nukta.addAll(UnicodeSet("[\\u093c\\u09bc\\u0a3c\\u0abc\\u0b3c\\u0cbc]",status));
1328 virama.addAll(UnicodeSet("[\\u094d\\u09cd\\u0a4d\\u0acd\\u0b4d\\u0bcd\\u0c4d\\u0ccd\\u0d4d]",status));
1329 sanskritStressSigns.addAll(UnicodeSet("[\\u0951\\u0952\\u0953\\u0954\\u097d]",status));
1330 chandrabindu.addAll(UnicodeSet("[\\u0901\\u0981\\u0A81\\u0b01\\u0c01]",status));
1331
1332 }
1333
1334 static const char latinForIndic[] = "[['.0-9A-Za-z~\\u00C0-\\u00C5\\u00C7-\\u00CF\\u00D1-\\u00D6\\u00D9-\\u00DD"
1335 "\\u00E0-\\u00E5\\u00E7-\\u00EF\\u00F1-\\u00F6\\u00F9-\\u00FD\\u00FF-\\u010F"
1336 "\\u0112-\\u0125\\u0128-\\u0130\\u0134-\\u0137\\u0139-\\u013E\\u0143-\\u0148"
1337 "\\u014C-\\u0151\\u0154-\\u0165\\u0168-\\u017E\\u01A0-\\u01A1\\u01AF-\\u01B0"
1338 "\\u01CD-\\u01DC\\u01DE-\\u01E3\\u01E6-\\u01ED\\u01F0\\u01F4-\\u01F5\\u01F8-\\u01FB"
1339 "\\u0200-\\u021B\\u021E-\\u021F\\u0226-\\u0233\\u0294\\u0303-\\u0304\\u0306\\u0314-\\u0315"
1340 "\\u0325\\u040E\\u0419\\u0439\\u045E\\u04C1-\\u04C2\\u04D0-\\u04D1\\u04D6-\\u04D7"
1341 "\\u04E2-\\u04E3\\u04EE-\\u04EF\\u1E00-\\u1E99\\u1EA0-\\u1EF9\\u1F01\\u1F03\\u1F05"
1342 "\\u1F07\\u1F09\\u1F0B\\u1F0D\\u1F0F\\u1F11\\u1F13\\u1F15\\u1F19\\u1F1B\\u1F1D\\u1F21"
1343 "\\u1F23\\u1F25\\u1F27\\u1F29\\u1F2B\\u1F2D\\u1F2F\\u1F31\\u1F33\\u1F35\\u1F37\\u1F39"
1344 "\\u1F3B\\u1F3D\\u1F3F\\u1F41\\u1F43\\u1F45\\u1F49\\u1F4B\\u1F4D\\u1F51\\u1F53\\u1F55"
1345 "\\u1F57\\u1F59\\u1F5B\\u1F5D\\u1F5F\\u1F61\\u1F63\\u1F65\\u1F67\\u1F69\\u1F6B\\u1F6D"
1346 "\\u1F6F\\u1F81\\u1F83\\u1F85\\u1F87\\u1F89\\u1F8B\\u1F8D\\u1F8F\\u1F91\\u1F93\\u1F95"
1347 "\\u1F97\\u1F99\\u1F9B\\u1F9D\\u1F9F\\u1FA1\\u1FA3\\u1FA5\\u1FA7\\u1FA9\\u1FAB\\u1FAD"
1348 "\\u1FAF-\\u1FB1\\u1FB8-\\u1FB9\\u1FD0-\\u1FD1\\u1FD8-\\u1FD9\\u1FE0-\\u1FE1\\u1FE5"
1349 "\\u1FE8-\\u1FE9\\u1FEC\\u212A-\\u212B\\uE04D\\uE064]"
1350 "-[\\uE000-\\uE080 \\u01E2\\u01E3]& [[:latin:][:mark:]]]";
1351
TestDevanagariLatin()1352 void TransliteratorRoundTripTest::TestDevanagariLatin() {
1353 {
1354 UErrorCode status = U_ZERO_ERROR;
1355 UParseError parseError;
1356 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));
1357 if((Transliterator *)t1 != NULL){
1358 TransliteratorPointer t2(t1->createInverse(status));
1359 if(U_FAILURE(status)){
1360 errln("FAIL: could not create the Inverse:-( \n");
1361 }
1362 }else {
1363 dataerrln("FAIL: could not create the transliterator. Error: %s\n", u_errorName(status));
1364 }
1365
1366 }
1367 RTTest test("Latin-Devanagari");
1368 Legal *legal = new LegalIndic();
1369 logKnownIssue( "cldrbug:1911");
1370 // It is left in its current state as a regression test.
1371
1372 test.test(UnicodeString(latinForIndic, ""),
1373 UnicodeString("[[[:Devanagari:][\\u094d][\\u0964\\u0965]]&[:Age=4.1:]-[\\u0970]]", ""), "[\\u0965\\u0904]", this, quick,
1374 legal, 50);
1375
1376 delete legal;
1377 }
1378
1379 /* Defined this way for HP/UX11CC :-( */
1380 static const int32_t INTER_INDIC_ARRAY_WIDTH = 4;
1381 static const char * const interIndicArray[] = {
1382
1383 "BENGALI-DEVANAGARI", "[:BENGALI:]", "[[:Devanagari:]-[\\u0970]]",
1384 "[\\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*/
1385
1386 "DEVANAGARI-BENGALI", "[[:Devanagari:]-[\\u0970]]", "[:BENGALI:]",
1387 "[\\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*/
1388
1389 "GURMUKHI-DEVANAGARI", "[:GURMUKHI:]", "[[:Devanagari:]-[\\u0970]]",
1390 "[\\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*/
1391
1392 "DEVANAGARI-GURMUKHI", "[[:Devanagari:]-[\\u0970]]", "[:GURMUKHI:]",
1393 "[\\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*/
1394
1395 "GUJARATI-DEVANAGARI", "[:GUJARATI:]", "[[:Devanagari:]-[\\u0970]]",
1396 "[\\u0946\\u094A\\u0962\\u0963\\u0951-\\u0954\\u0961\\u090c\\u090e\\u0912\\u097d]", /*roundtrip exclusions*/
1397
1398 "DEVANAGARI-GUJARATI", "[[:Devanagari:]-[\\u0970]]", "[:GUJARATI:]",
1399 "[\\u0951-\\u0954\\u0961\\u090c\\u090e\\u0912]", /*roundtrip exclusions*/
1400
1401 "ORIYA-DEVANAGARI", "[:ORIYA:]", "[[:Devanagari:]-[\\u0970]]",
1402 "[\\u0904\\u0943-\\u094a\\u0962\\u0963\\u0951-\\u0954\\u0950\\u090D\\u090e\\u0912\\u0911\\u0931\\u0935\\u097d]", /*roundtrip exclusions*/
1403
1404 "DEVANAGARI-ORIYA", "[[:Devanagari:]-[\\u0970]]", "[:ORIYA:]",
1405 "[\\u0b5f\\u0b56\\u0b57\\u0b70\\u0b71\\u0950\\u090D\\u090e\\u0912\\u0911\\u0931]", /*roundtrip exclusions*/
1406
1407 "Tamil-DEVANAGARI", "[:tamil:]", "[[:Devanagari:]-[\\u0970]]",
1408 "[\\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*/
1409
1410 "DEVANAGARI-Tamil", "[[:Devanagari:]-[\\u0970]]", "[:tamil:]",
1411 "[\\u0bd7\\u0BF0\\u0BF1\\u0BF2]", /*roundtrip exclusions*/
1412
1413 "Telugu-DEVANAGARI", "[:telugu:]", "[[:Devanagari:]-[\\u0970]]",
1414 "[\\u0904\\u093c\\u0950\\u0945\\u0949\\u0951-\\u0954\\u0962\\u0963\\u090D\\u0911\\u093d\\u0929\\u0934[\\u0958-\\u095f]\\u097d]", /*roundtrip exclusions*/
1415
1416 "DEVANAGARI-TELUGU", "[[:Devanagari:]-[\\u0970]]", "[:TELUGU:]",
1417 "[\\u0c55\\u0c56\\u0950\\u090D\\u0911\\u093d\\u0929\\u0934[\\u0958-\\u095f]]", /*roundtrip exclusions*/
1418
1419 "KANNADA-DEVANAGARI", "[:KANNADA:]", "[[:Devanagari:]-[\\u0970]]",
1420 "[\\u0901\\u0904\\u0946\\u093c\\u0950\\u0945\\u0949\\u0951-\\u0954\\u0962\\u0963\\u0950\\u090D\\u0911\\u093d\\u0929\\u0934[\\u0958-\\u095f]\\u097d]", /*roundtrip exclusions*/
1421
1422 "DEVANAGARI-KANNADA", "[[:Devanagari:]-[\\u0970]]", "[:KANNADA:]",
1423 "[{\\u0cb0\\u0cbc}{\\u0cb3\\u0cbc}\\u0cde\\u0cd5\\u0cd6\\u0950\\u090D\\u0911\\u093d\\u0929\\u0934[\\u0958-\\u095f]]", /*roundtrip exclusions*/
1424
1425 "MALAYALAM-DEVANAGARI", "[:MALAYALAM:]", "[[:Devanagari:]-[\\u0970]]",
1426 "[\\u0901\\u0904\\u094a\\u094b\\u094c\\u093c\\u0950\\u0944\\u0945\\u0949\\u0951-\\u0954\\u0962\\u0963\\u090D\\u0911\\u093d\\u0929\\u0934[\\u0958-\\u095f]\\u097d]", /*roundtrip exclusions*/
1427
1428 "DEVANAGARI-MALAYALAM", "[[:Devanagari:]-[\\u0970]]", "[:MALAYALAM:]",
1429 "[\\u0d4c\\u0d57\\u0950\\u090D\\u0911\\u093d\\u0929\\u0934[\\u0958-\\u095f]]", /*roundtrip exclusions*/
1430
1431 "GURMUKHI-BENGALI", "[:GURMUKHI:]", "[:BENGALI:]",
1432 "[\\u0981\\u0982\\u09b6\\u09e2\\u09e3\\u09c3\\u09c4\\u09d7\\u098B\\u098C\\u09B7\\u09E0\\u09E1\\u09F0\\u09F1\\u09f2-\\u09fa\\u09ce]", /*roundtrip exclusions*/
1433
1434 "BENGALI-GURMUKHI", "[:BENGALI:]", "[:GURMUKHI:]",
1435 "[\\u0A02\\u0a5c\\u0a47\\u0a70\\u0a71\\u0A33\\u0A35\\u0A59\\u0A5A\\u0A5B\\u0A5E\\u0A72\\u0A73\\u0A74]", /*roundtrip exclusions*/
1436
1437 "GUJARATI-BENGALI", "[:GUJARATI:]", "[:BENGALI:]",
1438 "[\\u09d7\\u09e2\\u09e3\\u098c\\u09e1\\u09f0\\u09f1\\u09f2-\\u09fa\\u09ce]", /*roundtrip exclusions*/
1439
1440 "BENGALI-GUJARATI", "[:BENGALI:]", "[:GUJARATI:]",
1441 "[\\u0A82\\u0a83\\u0Ac9\\u0Ac5\\u0ac7\\u0A8D\\u0A91\\u0AB3\\u0AB5\\u0ABD\\u0AD0]", /*roundtrip exclusions*/
1442
1443 "ORIYA-BENGALI", "[:ORIYA:]", "[:BENGALI:]",
1444 "[\\u09c4\\u09e2\\u09e3\\u09f0\\u09f1\\u09f2-\\u09fa\\u09ce]", /*roundtrip exclusions*/
1445
1446 "BENGALI-ORIYA", "[:BENGALI:]", "[:ORIYA:]",
1447 "[\\u0b35\\u0b71\\u0b5f\\u0b56\\u0b33\\u0b3d]", /*roundtrip exclusions*/
1448
1449 "Tamil-BENGALI", "[:tamil:]", "[:BENGALI:]",
1450 "[\\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*/
1451
1452 "BENGALI-Tamil", "[:BENGALI:]", "[:tamil:]",
1453 "[\\u0bc6\\u0bc7\\u0bca\\u0B8E\\u0B92\\u0BA9\\u0BB1\\u0BB3\\u0BB4\\u0BB5\\u0BF0\\u0BF1\\u0BF2]", /*roundtrip exclusions*/
1454
1455 "Telugu-BENGALI", "[:telugu:]", "[:BENGALI:]",
1456 "[\\u09e2\\u09e3\\u09bc\\u09d7\\u09f0\\u09f1\\u09dc\\u09dd\\u09df\\u09f2-\\u09fa\\u09ce]", /*roundtrip exclusions*/
1457
1458 "BENGALI-TELUGU", "[:BENGALI:]", "[:TELUGU:]",
1459 "[\\u0c55\\u0c56\\u0c47\\u0c46\\u0c4a\\u0C0E\\u0C12\\u0C31\\u0C33\\u0C35]", /*roundtrip exclusions*/
1460
1461 "KANNADA-BENGALI", "[:KANNADA:]", "[:BENGALI:]",
1462 "[\\u0981\\u09e2\\u09e3\\u09bc\\u09d7\\u09dc\\u09dd\\u09df\\u09f0\\u09f1\\u09f2-\\u09fa\\u09ce]", /*roundtrip exclusions*/
1463
1464 "BENGALI-KANNADA", "[:BENGALI:]", "[:KANNADA:]",
1465 "[{\\u0cb0\\u0cbc}{\\u0cb3\\u0cbc}\\u0cc6\\u0cca\\u0cd5\\u0cd6\\u0cc7\\u0C8E\\u0C92\\u0CB1\\u0cb3\\u0cb5\\u0cde]", /*roundtrip exclusions*/
1466
1467 "MALAYALAM-BENGALI", "[:MALAYALAM:]", "[:BENGALI:]",
1468 "[\\u0981\\u09e2\\u09e3\\u09bc\\u09c4\\u09f0\\u09f1\\u09dc\\u09dd\\u09df\\u09dc\\u09dd\\u09df\\u09f2-\\u09fa\\u09ce]", /*roundtrip exclusions*/
1469
1470 "BENGALI-MALAYALAM", "[:BENGALI:]", "[:MALAYALAM:]",
1471 "[\\u0d46\\u0d4a\\u0d47\\u0d31-\\u0d35\\u0d0e\\u0d12]", /*roundtrip exclusions*/
1472
1473 "GUJARATI-GURMUKHI", "[:GUJARATI:]", "[:GURMUKHI:]",
1474 "[\\u0A02\\u0ab3\\u0ab6\\u0A70\\u0a71\\u0a82\\u0a83\\u0ac3\\u0ac4\\u0ac5\\u0ac9\\u0a5c\\u0a72\\u0a73\\u0a74\\u0a8b\\u0a8d\\u0a91\\u0abd]", /*roundtrip exclusions*/
1475
1476 "GURMUKHI-GUJARATI", "[:GURMUKHI:]", "[:GUJARATI:]",
1477 "[\\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*/
1478
1479 "ORIYA-GURMUKHI", "[:ORIYA:]", "[:GURMUKHI:]",
1480 "[\\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*/
1481
1482 "GURMUKHI-ORIYA", "[:GURMUKHI:]", "[:ORIYA:]",
1483 "[\\u0b01\\u0b02\\u0b03\\u0b33\\u0b36\\u0b43\\u0b56\\u0b57\\u0B0B\\u0B0C\\u0B37\\u0B3D\\u0B5F\\u0B60\\u0B61\\u0b70\\u0b71]", /*roundtrip exclusions*/
1484
1485 "TAMIL-GURMUKHI", "[:TAMIL:]", "[:GURMUKHI:]",
1486 "[\\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*/
1487
1488 "GURMUKHI-TAMIL", "[:GURMUKHI:]", "[:TAMIL:]",
1489 "[\\u0b82\\u0bc6\\u0bca\\u0bd7\\u0bb7\\u0bb3\\u0b83\\u0B8E\\u0B92\\u0BA9\\u0BB1\\u0BB4\\u0bb6\\u0BF0\\u0BF1\\u0BF2]", /*roundtrip exclusions*/
1490
1491 "TELUGU-GURMUKHI", "[:TELUGU:]", "[:GURMUKHI:]",
1492 "[\\u0A02\\u0a33\\u0a36\\u0a3c\\u0a70\\u0a71\\u0A59\\u0A5A\\u0A5B\\u0A5C\\u0A5E\\u0A72\\u0A73\\u0A74]", /*roundtrip exclusions*/
1493
1494 "GURMUKHI-TELUGU", "[:GURMUKHI:]", "[:TELUGU:]",
1495 "[\\u0c01\\u0c02\\u0c03\\u0c33\\u0c36\\u0c44\\u0c43\\u0c46\\u0c4a\\u0c56\\u0c55\\u0C0B\\u0C0C\\u0C0E\\u0C12\\u0C31\\u0C37\\u0C60\\u0C61]", /*roundtrip exclusions*/
1496
1497 "KANNADA-GURMUKHI", "[:KANNADA:]", "[:GURMUKHI:]",
1498 "[\\u0A01\\u0A02\\u0a33\\u0a36\\u0a3c\\u0a70\\u0a71\\u0A59\\u0A5A\\u0A5B\\u0A5C\\u0A5E\\u0A72\\u0A73\\u0A74]", /*roundtrip exclusions*/
1499
1500 "GURMUKHI-KANNADA", "[:GURMUKHI:]", "[:KANNADA:]",
1501 "[{\\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*/
1502
1503 "MALAYALAM-GURMUKHI", "[:MALAYALAM:]", "[:GURMUKHI:]",
1504 "[\\u0A01\\u0A02\\u0a4b\\u0a4c\\u0a33\\u0a36\\u0a3c\\u0a70\\u0a71\\u0A59\\u0A5A\\u0A5B\\u0A5C\\u0A5E\\u0A72\\u0A73\\u0A74]", /*roundtrip exclusions*/
1505
1506 "GURMUKHI-MALAYALAM", "[:GURMUKHI:]", "[:MALAYALAM:]",
1507 "[\\u0d02\\u0d03\\u0d33\\u0d36\\u0d43\\u0d46\\u0d4a\\u0d4c\\u0d57\\u0D0B\\u0D0C\\u0D0E\\u0D12\\u0D31\\u0D34\\u0D37\\u0D60\\u0D61]", /*roundtrip exclusions*/
1508
1509 "GUJARATI-ORIYA", "[:GUJARATI:]", "[:ORIYA:]",
1510 "[\\u0b56\\u0b57\\u0B0C\\u0B5F\\u0B61\\u0b70\\u0b71]", /*roundtrip exclusions*/
1511
1512 "ORIYA-GUJARATI", "[:ORIYA:]", "[:GUJARATI:]",
1513 "[\\u0Ac4\\u0Ac5\\u0Ac9\\u0Ac7\\u0A8D\\u0A91\\u0AB5\\u0Ad0]", /*roundtrip exclusions*/
1514
1515 "TAMIL-GUJARATI", "[:TAMIL:]", "[:GUJARATI:]",
1516 "[\\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*/
1517
1518 "GUJARATI-TAMIL", "[:GUJARATI:]", "[:TAMIL:]",
1519 "[\\u0Bc6\\u0Bca\\u0Bd7\\u0B8E\\u0B92\\u0BA9\\u0BB1\\u0BB4\\u0BF0\\u0BF1\\u0BF2]", /*roundtrip exclusions*/
1520
1521 "TELUGU-GUJARATI", "[:TELUGU:]", "[:GUJARATI:]",
1522 "[\\u0abc\\u0Ac5\\u0Ac9\\u0A8D\\u0A91\\u0ABD\\u0Ad0]", /*roundtrip exclusions*/
1523
1524 "GUJARATI-TELUGU", "[:GUJARATI:]", "[:TELUGU:]",
1525 "[\\u0c46\\u0c4a\\u0c55\\u0c56\\u0C0C\\u0C0E\\u0C12\\u0C31\\u0C61]", /*roundtrip exclusions*/
1526
1527 "KANNADA-GUJARATI", "[:KANNADA:]", "[:GUJARATI:]",
1528 "[\\u0A81\\u0abc\\u0Ac5\\u0Ac9\\u0A8D\\u0A91\\u0ABD\\u0Ad0]", /*roundtrip exclusions*/
1529
1530 "GUJARATI-KANNADA", "[:GUJARATI:]", "[:KANNADA:]",
1531 "[{\\u0cb0\\u0cbc}{\\u0cb3\\u0cbc}\\u0cc6\\u0cca\\u0cd5\\u0cd6\\u0C8C\\u0C8E\\u0C92\\u0CB1\\u0CDE\\u0CE1]", /*roundtrip exclusions*/
1532
1533 "MALAYALAM-GUJARATI", "[:MALAYALAM:]", "[:GUJARATI:]",
1534 "[\\u0A81\\u0ac4\\u0acb\\u0acc\\u0abc\\u0Ac5\\u0Ac9\\u0A8D\\u0A91\\u0ABD\\u0Ad0]", /*roundtrip exclusions*/
1535
1536 "GUJARATI-MALAYALAM", "[:GUJARATI:]", "[:MALAYALAM:]",
1537 "[\\u0d46\\u0d4a\\u0d4c\\u0d55\\u0d57\\u0D0C\\u0D0E\\u0D12\\u0D31\\u0D34\\u0D61]", /*roundtrip exclusions*/
1538
1539 "TAMIL-ORIYA", "[:TAMIL:]", "[:ORIYA:]",
1540 "[\\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*/
1541
1542 "ORIYA-TAMIL", "[:ORIYA:]", "[:TAMIL:]",
1543 "[\\u0bc6\\u0bca\\u0bc7\\u0B8E\\u0B92\\u0BA9\\u0BB1\\u0BB4\\u0BB5\\u0BF0\\u0BF1\\u0BF2]", /*roundtrip exclusions*/
1544
1545 "TELUGU-ORIYA", "[:TELUGU:]", "[:ORIYA:]",
1546 "[\\u0b3c\\u0b57\\u0b56\\u0B3D\\u0B5C\\u0B5D\\u0B5F\\u0b70\\u0b71]", /*roundtrip exclusions*/
1547
1548 "ORIYA-TELUGU", "[:ORIYA:]", "[:TELUGU:]",
1549 "[\\u0c44\\u0c46\\u0c4a\\u0c55\\u0c47\\u0C0E\\u0C12\\u0C31\\u0C35]", /*roundtrip exclusions*/
1550
1551 "KANNADA-ORIYA", "[:KANNADA:]", "[:ORIYA:]",
1552 "[\\u0B01\\u0b3c\\u0b57\\u0B3D\\u0B5C\\u0B5D\\u0B5F\\u0b70\\u0b71]", /*roundtrip exclusions*/
1553
1554 "ORIYA-KANNADA", "[:ORIYA:]", "[:KANNADA:]",
1555 "[{\\u0cb0\\u0cbc}{\\u0cb3\\u0cbc}\\u0cc4\\u0cc6\\u0cca\\u0cd5\\u0cc7\\u0C8E\\u0C92\\u0CB1\\u0CB5\\u0CDE]", /*roundtrip exclusions*/
1556
1557 "MALAYALAM-ORIYA", "[:MALAYALAM:]", "[:ORIYA:]",
1558 "[\\u0B01\\u0b3c\\u0b56\\u0B3D\\u0B5C\\u0B5D\\u0B5F\\u0b70\\u0b71]", /*roundtrip exclusions*/
1559
1560 "ORIYA-MALAYALAM", "[:ORIYA:]", "[:MALAYALAM:]",
1561 "[\\u0D47\\u0D46\\u0D4a\\u0D0E\\u0D12\\u0D31\\u0D34\\u0D35]", /*roundtrip exclusions*/
1562
1563 "TELUGU-TAMIL", "[:TELUGU:]", "[:TAMIL:]",
1564 "[\\u0bd7\\u0ba9\\u0bb4\\u0BF0\\u0BF1\\u0BF2]", /*roundtrip exclusions*/
1565
1566 "TAMIL-TELUGU", "[:TAMIL:]", "[:TELUGU:]",
1567 "[\\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*/
1568
1569 "KANNADA-TAMIL", "[:KANNADA:]", "[:TAMIL:]",
1570 "[\\u0bd7\\u0bc6\\u0ba9\\u0bb4\\u0BF0\\u0BF1\\u0BF2]", /*roundtrip exclusions*/
1571
1572 "TAMIL-KANNADA", "[:TAMIL:]", "[:KANNADA:]",
1573 "[\\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*/
1574
1575 "MALAYALAM-TAMIL", "[:MALAYALAM:]", "[:TAMIL:]",
1576 "[\\u0ba9\\u0BF0\\u0BF1\\u0BF2]", /*roundtrip exclusions*/
1577
1578 "TAMIL-MALAYALAM", "[:TAMIL:]", "[:MALAYALAM:]",
1579 "[\\u0d43\\u0d12\\u0D0B\\u0D0C\\u0D16\\u0D17\\u0D18\\u0D1B\\u0D1D\\u0D20\\u0D21\\u0D22\\u0D25\\u0D26\\u0D27\\u0D2B\\u0D2C\\u0D2D\\u0D36\\u0D60\\u0D61]", /*roundtrip exclusions*/
1580
1581 "KANNADA-TELUGU", "[:KANNADA:]", "[:TELUGU:]",
1582 "[\\u0C01\\u0c3f\\u0c46\\u0c48\\u0c4a]", /*roundtrip exclusions*/
1583
1584 "TELUGU-KANNADA", "[:TELUGU:]", "[:KANNADA:]",
1585 "[\\u0cc8\\u0cd5\\u0cd6\\u0cbc\\u0cbd\\u0CDE]", /*roundtrip exclusions*/
1586
1587 "MALAYALAM-TELUGU", "[:MALAYALAM:]", "[:TELUGU:]",
1588 "[\\u0C01\\u0c44\\u0c4a\\u0c4c\\u0c4b\\u0c55\\u0c56]", /*roundtrip exclusions*/
1589
1590 "TELUGU-MALAYALAM", "[:TELUGU:]", "[:MALAYALAM:]",
1591 "[\\u0d4c\\u0d57\\u0D34]", /*roundtrip exclusions*/
1592
1593 "MALAYALAM-KANNADA", "[:MALAYALAM:]", "[:KANNADA:]",
1594 "[\\u0cbc\\u0cbd\\u0cc4\\u0cc6\\u0cca\\u0ccc\\u0ccb\\u0cd5\\u0cd6\\u0cDe]", /*roundtrip exclusions*/
1595
1596 "KANNADA-MALAYALAM", "[:KANNADA:]", "[:MALAYALAM:]",
1597 "[\\u0d4c\\u0d57\\u0d46\\u0D34]", /*roundtrip exclusions*/
1598
1599 "Latin-Bengali",latinForIndic, "[[:Bengali:][\\u0964\\u0965]]",
1600 "[\\u0965\\u09f0-\\u09fa\\u09ce]" /*roundtrip exclusions*/ ,
1601
1602 "Latin-Gurmukhi", latinForIndic, "[[:Gurmukhi:][\\u0964\\u0965]]",
1603 "[\\u0a01\\u0965\\u0a02\\u0a72\\u0a73\\u0a74]" /*roundtrip exclusions*/,
1604
1605 "Latin-Gujarati",latinForIndic, "[[:Gujarati:][\\u0964\\u0965]]",
1606 "[\\u0965]" /*roundtrip exclusions*/,
1607
1608 "Latin-Oriya",latinForIndic, "[[:Oriya:][\\u0964\\u0965]]",
1609 "[\\u0965\\u0b70]" /*roundtrip exclusions*/,
1610
1611 "Latin-Tamil",latinForIndic, "[:Tamil:]",
1612 "[\\u0BF0\\u0BF1\\u0BF2]" /*roundtrip exclusions*/,
1613
1614 "Latin-Telugu",latinForIndic, "[:Telugu:]",
1615 NULL /*roundtrip exclusions*/,
1616
1617 "Latin-Kannada",latinForIndic, "[:Kannada:]",
1618 NULL /*roundtrip exclusions*/,
1619
1620 "Latin-Malayalam",latinForIndic, "[:Malayalam:]",
1621 NULL /*roundtrip exclusions*/
1622 };
1623
TestDebug(const char * name,const char fromSet[],const char * toSet,const char * exclusions)1624 void TransliteratorRoundTripTest::TestDebug(const char* name,const char fromSet[],
1625 const char* toSet,const char* exclusions){
1626
1627 RTTest test(name);
1628 Legal *legal = new LegalIndic();
1629 test.test(UnicodeString(fromSet,""),UnicodeString(toSet,""),exclusions,this,quick,legal);
1630 }
1631
TestInterIndic()1632 void TransliteratorRoundTripTest::TestInterIndic() {
1633 //TestDebug("Latin-Gurmukhi", latinForIndic, "[:Gurmukhi:]","[\\u0965\\u0a02\\u0a72\\u0a73\\u0a74]",TRUE);
1634 int32_t num = UPRV_LENGTHOF(interIndicArray)/INTER_INDIC_ARRAY_WIDTH;
1635 if(quick){
1636 logln("Testing only 5 of %i. Skipping rest (use -e for exhaustive)",num);
1637 num = 5;
1638 }
1639 for(int i = 0; i < num;i++){
1640 RTTest test(interIndicArray[i*INTER_INDIC_ARRAY_WIDTH + 0]);
1641 Legal *legal = new LegalIndic();
1642 logln(UnicodeString("Stress testing ") + interIndicArray[i*INTER_INDIC_ARRAY_WIDTH + 0]);
1643 if( !logKnownIssue( "cldrbug:1911" ) ) {
1644 /* "full test" */
1645 // CLDR bug #1911: This test should be moved into CLDR.
1646 test.test( interIndicArray[i*INTER_INDIC_ARRAY_WIDTH + 1],
1647 interIndicArray[i*INTER_INDIC_ARRAY_WIDTH + 2],
1648 interIndicArray[i*INTER_INDIC_ARRAY_WIDTH + 3], // roundtrip exclusions
1649 this, quick, legal, 50);
1650 } else {
1651 // It is left in its current state as a regression test.
1652 // CLDR should test, and remove the age filter.
1653 /* regression test - ""temporary"" until CLDR#1911 is fixed */
1654 // start
1655 UnicodeString source("[");
1656 source.append(interIndicArray[i*INTER_INDIC_ARRAY_WIDTH + 1]);
1657 source.append(" & [:Age=4.1:]]");
1658 UnicodeString target("[");
1659 target.append(interIndicArray[i*INTER_INDIC_ARRAY_WIDTH + 2]);
1660 target.append(" & [:Age=4.1:]]");
1661 test.test( source,
1662 target,
1663 interIndicArray[i*INTER_INDIC_ARRAY_WIDTH + 3], // roundtrip exclusions
1664 this, quick, legal, 50);
1665 // end
1666 delete legal;
1667 }
1668 }
1669 }
1670
1671 // end indic tests ----------------------------------------------------------
1672
1673 #endif /* #if !UCONFIG_NO_TRANSLITERATION */
1674