]> git.saurik.com Git - apple/icu.git/blame - icuSources/test/intltest/transrt.cpp
projects / apple / icu.git / blame
? search:
summary | shortlog | log | commit | commitdiff | tree
blob | blame (incremental) | history | HEAD
ICU-3.13.tar.gz
[apple/icu.git] / icuSources / test / intltest / transrt.cpp
CommitLineData
b75a7d8f
A		 1/*
2**********************************************************************
3* Copyright (C) 2000-2003, International Business Machines
4* Corporation and others. All Rights Reserved.
5**********************************************************************
6* Date Name Description
7* 05/23/00 aliu Creation.
8**********************************************************************
9*/
10
11#include "unicode/utypes.h"
12
13#if !UCONFIG_NO_TRANSLITERATION
14
15#include "unicode/translit.h"
16#include "rbt.h"
17#include "unicode/uniset.h"
18#include "unicode/uchar.h"
19#include "unicode/normlzr.h"
20#include "unicode/uchar.h"
21#include "unicode/parseerr.h"
22#include "unicode/usetiter.h"
23#include "unicode/putil.h"
24#include "unicode/uversion.h"
25#include "cmemory.h"
26#include "transrt.h"
27#include "testutil.h"
28#include <string.h>
29
30#define CASE(id,test) case id: \
31 name = #test; \
32 if (exec) { \
33 logln(#test "---"); \
34 logln((UnicodeString)""); \
35 int32_t t = uprv_getUTCtime(); \
36 test(); \
37 t = uprv_getUTCtime() - t; \
38 logln((UnicodeString)#test " took " + t + " seconds"); \
39 } \
40 break
41
42#define EXHAUSTIVE(id,test) case id: \
43 if(quick==FALSE){ \
44 name = #test; \
45 if (exec){ \
46 logln(#test "---"); \
47 logln((UnicodeString)""); \
48 test(); \
49 } \
50 }else{ \
51 name=""; \
52 } \
53 break
54void
55TransliteratorRoundTripTest::runIndexedTest(int32_t index, UBool exec,
56 const char* &name, char* /*par*/) {
57 switch (index) {
58 CASE(0, TestCyrillic);
59 // CASE(0,TestKana);
60 CASE(1,TestHiragana);
61 CASE(2,TestKatakana);
62 CASE(3,TestJamo);
63 CASE(4,TestHangul);
64 CASE(5,TestGreek);
65 CASE(6,TestGreekUNGEGN);
66 CASE(7,Testel);
67 CASE(8,TestDevanagariLatin);
68 CASE(9,TestInterIndic);
69 default: name = ""; break;
70 }
71}
72
73//--------------------------------------------------------------------
74// Time bomb - allows temporary behavior that expires at a given
75// release
76//--------------------------------------------------------------------
77
78static const UVersionInfo ICU_30 = {3,0,0,0};
79
80static UBool isICUVersionAtLeast(const UVersionInfo x) {
81 UVersionInfo v;
82 u_getVersion(v);
83 return (uprv_memcmp(v, x, U_MAX_VERSION_LENGTH) >= 0);
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 */
95class TransliteratorPointer {
96 Transliterator* t;
97 // disallowed:
98 TransliteratorPointer(const TransliteratorPointer& rhs);
99 TransliteratorPointer& operator=(const TransliteratorPointer& rhs);
100public:
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
116class Legal {
117public:
118 Legal() {}
119 virtual ~Legal() {}
120 virtual UBool is(const UnicodeString& /*sourceString*/) const {return TRUE;}
121};
122
123class 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)
127public:
128 LegalJamo() {}
129 virtual ~LegalJamo() {}
130 virtual UBool is(const UnicodeString& sourceString) const;
131 int getType(UChar c) const;
132};
133
134UBool 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
158int 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
168class LegalGreek : public Legal {
169 UBool full;
170public:
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
181UBool 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
234UBool 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
255UBool 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
266class AbbreviatedUnicodeSetIterator : public UnicodeSetIterator {
267public :
268
269 AbbreviatedUnicodeSetIterator();
270 virtual ~AbbreviatedUnicodeSetIterator();
271 void reset(UnicodeSet& set, UBool abb = FALSE, int32_t density = 100);
272
273 /**
274 * ICU "poor man's RTTI", returns a UClassID for the actual class.
275 */
276 virtual inline UClassID getDynamicClassID() const { return getStaticClassID(); }
277
278 /**
279 * ICU "poor man's RTTI", returns a UClassID for this class.
280 */
281 static inline UClassID getStaticClassID() { return (UClassID)&fgClassID; }
282
283private :
284 UBool abbreviated;
285 int32_t perRange;
286 virtual void loadRange(int32_t range);
287
288 /**
289 * The address of this static class variable serves as this class's ID
290 * for ICU "poor man's RTTI".
291 */
292 static const char fgClassID;
293};
294
295// AbbreviatedUnicodeSetIterator Implementation ---------------------------------------
296
297const char AbbreviatedUnicodeSetIterator::fgClassID=0;
298
299AbbreviatedUnicodeSetIterator::AbbreviatedUnicodeSetIterator() :
300 UnicodeSetIterator(), abbreviated(FALSE) {
301}
302
303AbbreviatedUnicodeSetIterator::~AbbreviatedUnicodeSetIterator() {
304}
305
306void AbbreviatedUnicodeSetIterator::reset(UnicodeSet& newSet, UBool abb, int32_t density) {
307 UnicodeSetIterator::reset(newSet);
308 abbreviated = abb;
309 perRange = newSet.getRangeCount();
310 if (perRange != 0) {
311 perRange = density / perRange;
312 }
313}
314
315void AbbreviatedUnicodeSetIterator::loadRange(int32_t myRange) {
316 UnicodeSetIterator::loadRange(myRange);
317 if (abbreviated && (endElement > nextElement + perRange)) {
318 endElement = nextElement + perRange;
319 }
320}
321
322//--------------------------------------------------------------------
323// RTTest Interface
324//--------------------------------------------------------------------
325
326class RTTest : public IntlTest {
327
328 // PrintWriter out;
329
330 UnicodeString transliteratorID;
331 int32_t errorLimit;
332 int32_t errorCount;
333 int32_t pairLimit;
334 UnicodeSet sourceRange;
335 UnicodeSet targetRange;
336 UnicodeSet toSource;
337 UnicodeSet toTarget;
338 UnicodeSet roundtripExclusionsSet;
339 IntlTest* parent;
340 Legal* legalSource; // NOT owned
341 UnicodeSet badCharacters;
342
343public:
344
345 /*
346 * create a test for the given script transliterator.
347 */
348 RTTest(const UnicodeString& transliteratorIDStr);
349
350 virtual ~RTTest();
351
352 void setErrorLimit(int32_t limit);
353
354 void setPairLimit(int32_t limit);
355
356 void test(const UnicodeString& sourceRange,
357 const UnicodeString& targetRange,
358 const char* roundtripExclusions,
359 IntlTest* parent,
360 UBool quick,
361 Legal* adoptedLegal,
362 int32_t density = 100);
363
364private:
365
366 // Added to do better equality check.
367
368 static UBool isSame(const UnicodeString& a, const UnicodeString& b);
369
370 static UBool isCamel(const UnicodeString& a);
371
372 UBool checkIrrelevants(Transliterator *t, const UnicodeString& irrelevants);
373
374 void test2(UBool quick, int32_t density);
375
376 void logWrongScript(const UnicodeString& label,
377 const UnicodeString& from,
378 const UnicodeString& to);
379
380 void logNotCanonical(const UnicodeString& label,
381 const UnicodeString& from,
382 const UnicodeString& to,
383 const UnicodeString& fromCan,
384 const UnicodeString& toCan);
385
386 void logFails(const UnicodeString& label);
387
388 void logToRulesFails(const UnicodeString& label,
389 const UnicodeString& from,
390 const UnicodeString& to,
391 const UnicodeString& toCan);
392
393 void logRoundTripFailure(const UnicodeString& from,
394 const UnicodeString& toID,
395 const UnicodeString& to,
396 const UnicodeString& backID,
397 const UnicodeString& back);
398};
399
400//--------------------------------------------------------------------
401// RTTest Implementation
402//--------------------------------------------------------------------
403
404/*
405 * create a test for the given script transliterator.
406 */
407RTTest::RTTest(const UnicodeString& transliteratorIDStr) {
408 transliteratorID = transliteratorIDStr;
409 errorLimit = 500;
410 errorCount = 0;
411 pairLimit = 0x10000;
412}
413
414RTTest::~RTTest() {
415}
416
417void RTTest::setErrorLimit(int32_t limit) {
418 errorLimit = limit;
419}
420
421void RTTest::setPairLimit(int32_t limit) {
422 pairLimit = limit;
423}
424
425UBool RTTest::isSame(const UnicodeString& a, const UnicodeString& b) {
426 if (a == b) return TRUE;
427 if (a.caseCompare(b, U_FOLD_CASE_DEFAULT)==0 && isCamel(a)) return TRUE;
428 UnicodeString aa, bb;
429 UErrorCode ec = U_ZERO_ERROR;
430 Normalizer::decompose(a, FALSE, 0, aa, ec);
431 Normalizer::decompose(b, FALSE, 0, bb, ec);
432 if (aa == bb) return TRUE;
433 if (aa.caseCompare(bb, U_FOLD_CASE_DEFAULT)==0 && isCamel(aa)) return TRUE;
434 return FALSE;
435}
436
437UBool RTTest::isCamel(const UnicodeString& a) {
438 // see if string is of the form aB; e.g. lower, then upper or title
439 UChar32 cp;
440 UBool haveLower = FALSE;
441 for (int32_t i = 0; i < a.length(); i += UTF_CHAR_LENGTH(cp)) {
442 cp = a.char32At(i);
443 int8_t t = u_charType(cp);
444 switch (t) {
445 case U_UPPERCASE_LETTER:
446 if (haveLower) return TRUE;
447 break;
448 case U_TITLECASE_LETTER:
449 if (haveLower) return TRUE;
450 // drop through, since second letter is lower.
451 case U_LOWERCASE_LETTER:
452 haveLower = TRUE;
453 break;
454 }
455 }
456 return FALSE;
457}
458
459void RTTest::test(const UnicodeString& sourceRangeVal,
460 const UnicodeString& targetRangeVal,
461 const char* roundtripExclusions,
462 IntlTest* logVal, UBool quickRt,
463 Legal* adoptedLegal,
464 int32_t density)
465{
466
467 UErrorCode status = U_ZERO_ERROR;
468
469 this->parent = logVal;
470 this->legalSource = adoptedLegal;
471
472 UnicodeSet neverOk("[:Other:]", status);
473 UnicodeSet okAnyway("[^[:Letter:]]", status);
474
475 if (U_FAILURE(status)) {
476 parent->errln("FAIL: Initializing UnicodeSet with [:Other:] or [^[:Letter:]]");
477 return;
478 }
479
480 this->sourceRange.clear();
481 this->sourceRange.applyPattern(sourceRangeVal, status);
482 if (U_FAILURE(status)) {
483 parent->errln("FAIL: UnicodeSet::applyPattern(" +
484 sourceRangeVal + ")");
485 return;
486 }
487 this->sourceRange.removeAll(neverOk);
488
489 this->targetRange.clear();
490 this->targetRange.applyPattern(targetRangeVal, status);
491 if (U_FAILURE(status)) {
492 parent->errln("FAIL: UnicodeSet::applyPattern(" +
493 targetRangeVal + ")");
494 return;
495 }
496 this->targetRange.removeAll(neverOk);
497
498 this->toSource.clear();
499 this->toSource.applyPattern(sourceRangeVal, status);
500 if (U_FAILURE(status)) {
501 parent->errln("FAIL: UnicodeSet::applyPattern(" +
502 sourceRangeVal + ")");
503 return;
504 }
505 this->toSource.addAll(okAnyway);
506
507 this->toTarget.clear();
508 this->toTarget.applyPattern(targetRangeVal, status);
509 if (U_FAILURE(status)) {
510 parent->errln("FAIL: UnicodeSet::applyPattern(" +
511 targetRangeVal + ")");
512 return;
513 }
514 this->toTarget.addAll(okAnyway);
515
516 this->roundtripExclusionsSet.clear();
517 if (roundtripExclusions != NULL && strlen(roundtripExclusions) > 0) {
518 this->roundtripExclusionsSet.applyPattern(roundtripExclusions, status);
519 if (U_FAILURE(status)) {
520 parent->errln("FAIL: UnicodeSet::applyPattern(%s)", roundtripExclusions);
521 return;
522 }
523 }
524
525 badCharacters.clear();
526 badCharacters.applyPattern("[:Other:]", status);
527 if (U_FAILURE(status)) {
528 parent->errln("FAIL: UnicodeSet::applyPattern([:Other:])");
529 return;
530 }
531
532 test2(quickRt, density);
533
534 if (errorCount > 0) {
535 char str[100];
536 int32_t length = transliteratorID.extract(str, 100, NULL, status);
537 str[length] = 0;
538 parent->errln("FAIL: %s errors: %d %s", str, errorCount, (errorCount > errorLimit ? " (at least!)" : " ")); // + ", see " + logFileName);
539 } else {
540 char str[100];
541 int32_t length = transliteratorID.extract(str, 100, NULL, status);
542 str[length] = 0;
543 parent->logln("%s ok", str);
544 }
545}
546
547UBool RTTest::checkIrrelevants(Transliterator *t,
548 const UnicodeString& irrelevants) {
549 for (int i = 0; i < irrelevants.length(); ++i) {
550 UChar c = irrelevants.charAt(i);
551 UnicodeString srcStr(c);
552 UnicodeString targ = srcStr;
553 t->transliterate(targ);
554 if (srcStr == targ) return TRUE;
555 }
556 return FALSE;
557}
558
559void RTTest::test2(UBool quickRt, int32_t density) {
560
561 UnicodeString srcStr, targ, reverse;
562 UErrorCode status = U_ZERO_ERROR;
563 UParseError parseError ;
564 TransliteratorPointer sourceToTarget(
565 Transliterator::createInstance(transliteratorID, UTRANS_FORWARD, parseError,
566 status));
567 if (sourceToTarget == NULL) {
568 parent->errln("FAIL: createInstance(" + transliteratorID +
569 ") returned NULL. Error: " + u_errorName(status)
570 + "\n\tpreContext : " + prettify(parseError.preContext)
571 + "\n\tpostContext : " + prettify(parseError.postContext));
572
573 return;
574 }
575 TransliteratorPointer targetToSource(sourceToTarget->createInverse(status));
576 if (targetToSource == NULL) {
577 parent->errln("FAIL: " + transliteratorID +
578 ".createInverse() returned NULL. Error:" + u_errorName(status)
579 + "\n\tpreContext : " + prettify(parseError.preContext)
580 + "\n\tpostContext : " + prettify(parseError.postContext));
581 return;
582 }
583
584 AbbreviatedUnicodeSetIterator usi;
585 AbbreviatedUnicodeSetIterator usi2;
586
587 parent->logln("Checking that at least one irrelevant character is not NFC'ed");
588 // string is from NFC_NO in the UCD
589 UnicodeString irrelevants = CharsToUnicodeString("\\u2000\\u2001\\u2126\\u212A\\u212B\\u2329");
590
591 if (checkIrrelevants(sourceToTarget, irrelevants) == FALSE) {
592 logFails("Source-Target, irrelevants");
593 }
594 if (checkIrrelevants(targetToSource, irrelevants) == FALSE) {
595 logFails("Target-Source, irrelevants");
596 }
597
598 if (!quickRt){
599 parent->logln("Checking that toRules works");
600 UnicodeString rules = "";
601
602 UParseError parseError;
603 rules = sourceToTarget->toRules(rules, TRUE);
604 // parent->logln((UnicodeString)"toRules => " + rules);
605 TransliteratorPointer sourceToTarget2(Transliterator::createFromRules(
606 "s2t2", rules,
607 UTRANS_FORWARD,
608 parseError, status));
609 if (U_FAILURE(status)) {
610 parent->errln("FAIL: createFromRules %s\n", u_errorName(status));
611 return;
612 }
613
614 rules = targetToSource->toRules(rules, FALSE);
615 TransliteratorPointer targetToSource2(Transliterator::createFromRules(
616 "t2s2", rules,
617 UTRANS_FORWARD,
618 parseError, status));
619 if (U_FAILURE(status)) {
620 parent->errln("FAIL: createFromRules %s\n", u_errorName(status));
621 return;
622 }
623
624 usi.reset(sourceRange);
625 for (;;) {
626 if (!usi.next() || usi.isString()) break;
627 UChar32 c = usi.getCodepoint();
628
629 UnicodeString srcStr((UChar32)c);
630 UnicodeString targ = srcStr;
631 sourceToTarget->transliterate(targ);
632 UnicodeString targ2 = srcStr;
633 sourceToTarget2->transliterate(targ2);
634 if (targ != targ2) {
635 logToRulesFails("Source-Target, toRules", srcStr, targ, targ2);
636 }
637 }
638
639 usi.reset(targetRange);
640 for (;;) {
641 if (!usi.next() || usi.isString()) break;
642 UChar32 c = usi.getCodepoint();
643
644 UnicodeString srcStr((UChar32)c);
645 UnicodeString targ = srcStr;
646 targetToSource->transliterate(targ);
647 UnicodeString targ2 = srcStr;
648 targetToSource2->transliterate(targ2);
649 if (targ != targ2) {
650 logToRulesFails("Target-Source, toRules", srcStr, targ, targ2);
651 }
652 }
653 }
654
655 parent->logln("Checking that all source characters convert to target - Singles");
656
657 UnicodeSet failSourceTarg;
658 usi.reset(sourceRange);
659 for (;;) {
660 if (!usi.next() || usi.isString()) break;
661 UChar32 c = usi.getCodepoint();
662
663 UnicodeString srcStr((UChar32)c);
664 UnicodeString targ = srcStr;
665 sourceToTarget->transliterate(targ);
666 if (toTarget.containsAll(targ) == FALSE
667 || badCharacters.containsSome(targ) == TRUE) {
668 UnicodeString targD;
669 Normalizer::decompose(targ, FALSE, 0, targD, status);
670 if (U_FAILURE(status)) {
671 parent->errln("FAIL: Internal error during decomposition %s\n", u_errorName(status));
672 return;
673 }
674 if (toTarget.containsAll(targD) == FALSE ||
675 badCharacters.containsSome(targD) == TRUE) {
676 logWrongScript("Source-Target", srcStr, targ);
677 failSourceTarg.add(c);
678 continue;
679 }
680 }
681
682 UnicodeString cs2;
683 Normalizer::decompose(srcStr, FALSE, 0, cs2, status);
684 if (U_FAILURE(status)) {
685 parent->errln("FAIL: Internal error during decomposition %s\n", u_errorName(status));
686 return;
687 }
688 UnicodeString targ2 = cs2;
689 sourceToTarget->transliterate(targ2);
690 if (targ != targ2) {
691 logNotCanonical("Source-Target", srcStr, targ,cs2, targ2);
692 }
693 }
694
695 parent->logln("Checking that all source characters convert to target - Doubles");
696
697 UnicodeSet sourceRangeMinusFailures(sourceRange);
698 sourceRangeMinusFailures.removeAll(failSourceTarg);
699
700 usi.reset(sourceRangeMinusFailures, quickRt, density);
701 for (;;) {
702 if (!usi.next() || usi.isString()) break;
703 UChar32 c = usi.getCodepoint();
704
705 usi2.reset(sourceRangeMinusFailures, quickRt, density);
706 for (;;) {
707 if (!usi2.next() || usi2.isString()) break;
708 UChar32 d = usi2.getCodepoint();
709
710 UnicodeString srcStr;
711 srcStr += (UChar32)c;
712 srcStr += (UChar32)d;
713 UnicodeString targ = srcStr;
714 sourceToTarget->transliterate(targ);
715 if (toTarget.containsAll(targ) == FALSE ||
716 badCharacters.containsSome(targ) == TRUE)
717 {
718 UnicodeString targD;
719 Normalizer::decompose(targ, FALSE, 0, targD, status);
720 if (U_FAILURE(status)) {
721 parent->errln("FAIL: Internal error during decomposition %s\n", u_errorName(status));
722 return;
723 }
724 if (toTarget.containsAll(targD) == FALSE ||
725 badCharacters.containsSome(targD) == TRUE) {
726 logWrongScript("Source-Target", srcStr, targ);
727 continue;
728 }
729 }
730 UnicodeString cs2;
731 Normalizer::decompose(srcStr, FALSE, 0, cs2, status);
732 if (U_FAILURE(status)) {
733 parent->errln("FAIL: Internal error during decomposition %s\n", u_errorName(status));
734 return;
735 }
736 UnicodeString targ2 = cs2;
737 sourceToTarget->transliterate(targ2);
738 if (targ != targ2) {
739 logNotCanonical("Source-Target", srcStr, targ, cs2,targ2);
740 }
741 }
742 }
743
744 parent->logln("Checking that target characters convert to source and back - Singles");
745
746 UnicodeSet failTargSource;
747 UnicodeSet failRound;
748
749 usi.reset(targetRange);
750 for (;;) {
751 if (!usi.next()) break;
752
753 if(usi.isString()){
754 srcStr = usi.getString();
755 }else{
756 srcStr = (UnicodeString)usi.getCodepoint();
757 }
758
759 UChar32 c = srcStr.char32At(0);
760
761 targ = srcStr;
762 targetToSource->transliterate(targ);
763 reverse = targ;
764 sourceToTarget->transliterate(reverse);
765
766 if (toSource.containsAll(targ) == FALSE ||
767 badCharacters.containsSome(targ) == TRUE) {
768 UnicodeString targD;
769 Normalizer::decompose(targ, FALSE, 0, targD, status);
770 if (U_FAILURE(status)) {
771 parent->errln("FAIL: Internal error during decomposition%s\n", u_errorName(status));
772 return;
773 }
774 if (toSource.containsAll(targD) == FALSE ||
775 badCharacters.containsSome(targD) == TRUE) {
776 logWrongScript("Target-Source", srcStr, targ);
777 failTargSource.add(c);
778 continue;
779 }
780 }
781 if (isSame(srcStr, reverse) == FALSE &&
782 roundtripExclusionsSet.contains(c) == FALSE
783 && roundtripExclusionsSet.contains(srcStr)==FALSE) {
784 logRoundTripFailure(srcStr,targetToSource->getID(), targ,sourceToTarget->getID(), reverse);
785 failRound.add(c);
786 continue;
787 }
788
789 UnicodeString targ2;
790 Normalizer::decompose(targ, FALSE, 0, targ2, status);
791 if (U_FAILURE(status)) {
792 parent->errln("FAIL: Internal error during decomposition%s\n", u_errorName(status));
793 return;
794 }
795 UnicodeString reverse2 = targ2;
796 sourceToTarget->transliterate(reverse2);
797 if (reverse != reverse2) {
798 logNotCanonical("Target-Source", targ, reverse, targ2, reverse2);
799 }
800 }
801
802 parent->logln("Checking that target characters convert to source and back - Doubles");
803 int32_t count = 0;
804
805 UnicodeSet targetRangeMinusFailures(targetRange);
806 targetRangeMinusFailures.removeAll(failTargSource);
807 targetRangeMinusFailures.removeAll(failRound);
808
809 usi.reset(targetRangeMinusFailures, quickRt, density);
810 UnicodeString targ2;
811 UnicodeString reverse2;
812 UnicodeString targD;
813 for (;;) {
814 if (!usi.next() || usi.isString()) break;
815 UChar32 c = usi.getCodepoint();
816 if (++count > pairLimit) {
817 //throw new TestTruncated("Test truncated at " + pairLimit + " x 64k pairs");
818 parent->logln("");
819 parent->logln((UnicodeString)"Test truncated at " + pairLimit + " x 64k pairs");
820 return;
821 }
822
823 usi2.reset(targetRangeMinusFailures, quickRt, density);
824 for (;;) {
825 if (!usi2.next() || usi2.isString())
826 break;
827 UChar32 d = usi2.getCodepoint();
828 srcStr.truncate(0); // empty the variable without construction/destruction
829 srcStr += c;
830 srcStr += d;
831
832 targ = srcStr;
833 targetToSource->transliterate(targ);
834 reverse = targ;
835 sourceToTarget->transliterate(reverse);
836
837 if (toSource.containsAll(targ) == FALSE ||
838 badCharacters.containsSome(targ) == TRUE)
839 {
840 targD.truncate(0); // empty the variable without construction/destruction
841 Normalizer::decompose(targ, FALSE, 0, targD, status);
842 if (U_FAILURE(status)) {
843 parent->errln("FAIL: Internal error during decomposition%s\n",
844 u_errorName(status));
845 return;
846 }
847 if (toSource.containsAll(targD) == FALSE
848 || badCharacters.containsSome(targD) == TRUE)
849 {
850 logWrongScript("Target-Source", srcStr, targ);
851 continue;
852 }
853 }
854 if (isSame(srcStr, reverse) == FALSE &&
855 roundtripExclusionsSet.contains(c) == FALSE&&
856 roundtripExclusionsSet.contains(d) == FALSE &&
857 roundtripExclusionsSet.contains(srcStr)== FALSE)
858 {
859 logRoundTripFailure(srcStr,targetToSource->getID(), targ, sourceToTarget->getID(),reverse);
860 continue;
861 }
862
863 targ2.truncate(0); // empty the variable without construction/destruction
864 Normalizer::decompose(targ, FALSE, 0, targ2, status);
865 if (U_FAILURE(status)) {
866 parent->errln("FAIL: Internal error during decomposition%s\n", u_errorName(status));
867 return;
868 }
869 reverse2 = targ2;
870 sourceToTarget->transliterate(reverse2);
871 if (reverse != reverse2) {
872 logNotCanonical("Target-Source", targ,reverse, targ2, reverse2);
873 }
874 }
875 }
876 parent->logln("");
877}
878
879void RTTest::logWrongScript(const UnicodeString& label,
880 const UnicodeString& from,
881 const UnicodeString& to) {
882 parent->errln((UnicodeString)"FAIL " +
883 label + ": " +
884 from + "(" + TestUtility::hex(from) + ") => " +
885 to + "(" + TestUtility::hex(to) + ")");
886 ++errorCount;
887}
888
889void RTTest::logNotCanonical(const UnicodeString& label,
890 const UnicodeString& from,
891 const UnicodeString& to,
892 const UnicodeString& fromCan,
893 const UnicodeString& toCan) {
894 parent->errln((UnicodeString)"FAIL (can.equiv)" +
895 label + ": " +
896 from + "(" + TestUtility::hex(from) + ") => " +
897 to + "(" + TestUtility::hex(to) + ")" +
898 fromCan + "(" + TestUtility::hex(fromCan) + ") => " +
899 toCan + " (" +
900 TestUtility::hex(toCan) + ")"
901 );
902 ++errorCount;
903}
904
905void RTTest::logFails(const UnicodeString& label) {
906 parent->errln((UnicodeString)"<br>FAIL " + label);
907 ++errorCount;
908}
909
910void RTTest::logToRulesFails(const UnicodeString& label,
911 const UnicodeString& from,
912 const UnicodeString& to,
913 const UnicodeString& otherTo)
914{
915 parent->errln((UnicodeString)"FAIL: " +
916 label + ": " +
917 from + "(" + TestUtility::hex(from) + ") => " +
918 to + "(" + TestUtility::hex(to) + ")" +
919 "!=" +
920 otherTo + " (" +
921 TestUtility::hex(otherTo) + ")"
922 );
923 ++errorCount;
924}
925
926
927void RTTest::logRoundTripFailure(const UnicodeString& from,
928 const UnicodeString& toID,
929 const UnicodeString& to,
930 const UnicodeString& backID,
931 const UnicodeString& back) {
932 if (legalSource->is(from) == FALSE) return; // skip illegals
933
934 parent->errln((UnicodeString)"FAIL Roundtrip: " +
935 from + "(" + TestUtility::hex(from) + ") => " +
936 to + "(" + TestUtility::hex(to) + ") "+toID+" => " +
937 back + "(" + TestUtility::hex(back) + ") "+backID+" => ");
938 ++errorCount;
939}
940
941//--------------------------------------------------------------------
942// Specific Tests
943//--------------------------------------------------------------------
944
945 /*
946 Note: Unicode 3.2 added new Hiragana/Katakana characters:
947
9483095..3096 ; 3.2 # [2] HIRAGANA LETTER SMALL KA..HIRAGANA LETTER SMALL KE
949309F..30A0 ; 3.2 # [2] HIRAGANA DIGRAPH YORI..KATAKANA-HIRAGANA DOUBLE HYPHEN
95030FF ; 3.2 # KATAKANA DIGRAPH KOTO
95131F0..31FF ; 3.2 # [16] KATAKANA LETTER SMALL KU..KATAKANA LETTER SMALL RO
952
953 We will not add them to the rules until they are more supported (e.g. in fonts on Windows)
954 A bug has been filed to remind us to do this: #1979.
955 */
956
957static const char KATAKANA[] = "[[[:katakana:][\\u30A1-\\u30FA\\u30FC]]-[\\u30FF\\u31F0-\\u31FF]]";
958static const char HIRAGANA[] = "[[[:hiragana:][\\u3040-\\u3094]]-[\\u3095-\\u3096\\u309F-\\u30A0]]";
959static const char LENGTH[] = "[\\u30FC]";
960static const char HALFWIDTH_KATAKANA[] = "[\\uFF65-\\uFF9D]";
961static const char KATAKANA_ITERATION[] = "[\\u30FD\\u30FE]";
962static const char HIRAGANA_ITERATION[] = "[\\u309D\\u309E]";
963static const int32_t TEMP_MAX=256;
964
965void TransliteratorRoundTripTest::TestKana() {
966 RTTest test("Katakana-Hiragana");
967 Legal *legal = new Legal();
968 char temp[TEMP_MAX];
969 strcpy(temp, "[");
970 strcat(temp, HALFWIDTH_KATAKANA);
971 strcat(temp, LENGTH);
972 strcat(temp, "]");
973 test.test(KATAKANA, UnicodeString("[") + HIRAGANA + LENGTH + UnicodeString("]"),
974 temp,
975 this, quick, legal);
976 delete legal;
977}
978
979void TransliteratorRoundTripTest::TestHiragana() {
980 RTTest test("Latin-Hiragana");
981 Legal *legal = new Legal();
982 test.test(UnicodeString("[a-zA-Z]", ""),
983 HIRAGANA,
984 HIRAGANA_ITERATION, this, quick, legal);
985 delete legal;
986}
987
988void TransliteratorRoundTripTest::TestKatakana() {
989 RTTest test("Latin-Katakana");
990 Legal *legal = new Legal();
991 char temp[TEMP_MAX];
992 strcpy(temp, "[");
993 strcat(temp, KATAKANA_ITERATION);
994 strcat(temp, HALFWIDTH_KATAKANA);
995 strcat(temp, "]");
996 test.test(UnicodeString("[a-zA-Z]", ""),
997 KATAKANA,
998 temp,
999 this, quick, legal);
1000 delete legal;
1001}
1002
1003void TransliteratorRoundTripTest::TestJamo() {
1004 RTTest t("Latin-Jamo");
1005 Legal *legal = new LegalJamo();
1006 t.test(UnicodeString("[a-zA-Z]", ""),
1007 UnicodeString("[\\u1100-\\u1112 \\u1161-\\u1175 \\u11A8-\\u11C2]",
1008 ""),
1009 NULL, this, quick, legal);
1010 delete legal;
1011}
1012
1013void TransliteratorRoundTripTest::TestHangul() {
1014 RTTest t("Latin-Hangul");
1015 Legal *legal = new Legal();
1016 if (quick) t.setPairLimit(1000);
1017 t.test(UnicodeString("[a-zA-Z]", ""),
1018 UnicodeString("[\\uAC00-\\uD7A4]", ""),
1019 NULL, this, quick, legal, 1);
1020 delete legal;
1021}
1022
1023void TransliteratorRoundTripTest::TestGreek() {
1024 if (isICUVersionAtLeast(ICU_30)) {
1025 // We temporarily filter against Unicode 3.2, but we only do this
1026 // before version 3.0.
1027 errln("FAIL: TestGreek needs to be updated to remove Unicode 3.2 filter");
1028 return;
1029 } else {
1030 logln("Warning: TestGreek needs to be updated to remove Unicode 3.2 filter");
1031 }
1032 RTTest test("Latin-Greek");
1033 LegalGreek *legal = new LegalGreek(TRUE);
1034
1035 test.test(UnicodeString("[a-zA-Z]", ""),
1036 UnicodeString("[[\\u003B\\u00B7[:Greek:]-[\\u03D7-\\u03EF]]&[:Age=3.2:]]",
1037 ""),
1038 "[\\u00B5\\u037A\\u03D0-\\u03F5]", /* exclusions */
1039 this, quick, legal, 50);
1040
1041
1042 delete legal;
1043}
1044
1045
1046void TransliteratorRoundTripTest::TestGreekUNGEGN() {
1047 if (isICUVersionAtLeast(ICU_30)) {
1048 // We temporarily filter against Unicode 3.2, but we only do this
1049 // before version 3.0.
1050 errln("FAIL: TestGreekUNGEGN needs to be updated to remove Unicode 3.2 filter");
1051 return;
1052 } else {
1053 logln("Warning: TestGreekUNGEGN needs to be updated to remove Unicode 3.2 filter");
1054 }
1055 RTTest test("Latin-Greek/UNGEGN");
1056 LegalGreek *legal = new LegalGreek(FALSE);
1057
1058 test.test(UnicodeString("[a-zA-Z]", ""),
1059 UnicodeString("[[\\u003B\\u00B7[:Greek:]-[\\u03D7-\\u03EF]]&[:Age=3.2:]]",
1060 ""),
1061 "[\\u00B5\\u037A\\u03D0-\\uFFFF {\\u039C\\u03C0}]", /* roundtrip exclusions */
1062 this, quick, legal);
1063
1064 delete legal;
1065}
1066
1067void TransliteratorRoundTripTest::Testel() {
1068 if (isICUVersionAtLeast(ICU_30)) {
1069 // We temporarily filter against Unicode 3.2, but we only do this
1070 // before version 3.0.
1071 errln("FAIL: Testel needs to be updated to remove Unicode 3.2 filter");
1072 return;
1073 } else {
1074 logln("Warning: Testel needs to be updated to remove Unicode 3.2 filter");
1075 }
1076 RTTest test("Latin-el");
1077 LegalGreek *legal = new LegalGreek(FALSE);
1078
1079 test.test(UnicodeString("[a-zA-Z]", ""),
1080 UnicodeString("[[\\u003B\\u00B7[:Greek:]-[\\u03D7-\\u03EF]]&[:Age=3.2:]]",
1081 ""),
1082 "[\\u00B5\\u037A\\u03D0-\\uFFFF {\\u039C\\u03C0}]", /* exclusions */
1083 this, quick, legal);
1084
1085
1086 delete legal;
1087}
1088
1089void TransliteratorRoundTripTest::TestCyrillic() {
1090 RTTest test("Latin-Cyrillic");
1091 Legal *legal = new Legal();
1092
1093 test.test(UnicodeString("[a-zA-Z\\u0110\\u0111\\u02BA\\u02B9]", ""),
1094 UnicodeString("[[\\u0400-\\u045F] & [:Age=3.2:]]", ""), NULL, this, quick,
1095 legal);
1096
1097 delete legal;
1098}
1099
1100
1101// Inter-Indic Tests ----------------------------------
1102class LegalIndic :public Legal{
1103 UnicodeSet vowelSignSet;
1104 UnicodeSet avagraha;
1105 UnicodeSet nukta;
1106 UnicodeSet virama;
1107 UnicodeSet sanskritStressSigns;
1108 UnicodeSet chandrabindu;
1109
1110public:
1111 LegalIndic(){
1112 UErrorCode status = U_ZERO_ERROR;
1113 vowelSignSet.addAll( UnicodeSet("[\\u0902\\u0903\\u093e-\\u094c\\u0962\\u0963]",status));/* Devanagari */
1114 vowelSignSet.addAll( UnicodeSet("[\\u0982\\u0983\\u09be-\\u09cc\\u09e2\\u09e3\\u09D7]",status));/* Bengali */
1115 vowelSignSet.addAll( UnicodeSet("[\\u0a02\\u0a03\\u0a3e-\\u0a4c\\u0a62\\u0a63\\u0a70\\u0a71]",status));/* Gurmukhi */
1116 vowelSignSet.addAll( UnicodeSet("[\\u0a82\\u0a83\\u0abe-\\u0acc\\u0ae2\\u0ae3]",status));/* Gujarati */
1117 vowelSignSet.addAll( UnicodeSet("[\\u0b02\\u0b03\\u0b3e-\\u0b4c\\u0b62\\u0b63\\u0b56\\u0b57]",status));/* Oriya */
1118 vowelSignSet.addAll( UnicodeSet("[\\u0b82\\u0b83\\u0bbe-\\u0bcc\\u0be2\\u0be3\\u0bd7]",status));/* Tamil */
1119 vowelSignSet.addAll( UnicodeSet("[\\u0c02\\u0c03\\u0c3e-\\u0c4c\\u0c62\\u0c63\\u0c55\\u0c56]",status));/* Telugu */
1120 vowelSignSet.addAll( UnicodeSet("[\\u0c82\\u0c83\\u0cbe-\\u0ccc\\u0ce2\\u0ce3\\u0cd5\\u0cd6]",status));/* Kannada */
1121 vowelSignSet.addAll( UnicodeSet("[\\u0d02\\u0d03\\u0d3e-\\u0d4c\\u0d62\\u0d63\\u0d57]",status));/* Malayalam */
1122
1123 avagraha.addAll(UnicodeSet("[\\u093d\\u0abd\\u0b3d]",status));
1124 nukta.addAll(UnicodeSet("[\\u093c\\u09bc\\u0a3c\\u0abc\\u0b3c]",status));
1125 virama.addAll(UnicodeSet("[\\u094d\\u09cd\\u0a4d\\u0acd\\u0b4d\\u0bcd\\u0c4d\\u0ccd\\u0d4d]",status));
1126 sanskritStressSigns.addAll(UnicodeSet("[\\u0951\\u0952\\u0953\\u0954]",status));
1127 chandrabindu.addAll(UnicodeSet("[\\u0901\\u0981\\u0A81\\u0b01\\u0c01]",status));
1128
1129 }
1130 virtual UBool is(const UnicodeString& sourceString) const;
1131 virtual ~LegalIndic() {};
1132};
1133UBool LegalIndic::is(const UnicodeString& sourceString) const{
1134 int cp=sourceString.charAt(0);
1135
1136 // A vowel sign cannot be the first char
1137 if(vowelSignSet.contains(cp)){
1138 return FALSE;
1139 }else if(avagraha.contains(cp)){
1140 return FALSE;
1141 }else if(virama.contains(cp)){
1142 return FALSE;
1143 }else if(nukta.contains(cp)){
1144 return FALSE;
1145 }else if(sanskritStressSigns.contains(cp)){
1146 return FALSE;
1147 }else if(chandrabindu.contains(cp) &&
1148 ((sourceString.length()>1) &&
1149 vowelSignSet.contains(sourceString.charAt(1)))){
1150 return FALSE;
1151 }
1152 return TRUE;
1153}
1154
1155static const char latinForIndic[] = "[['.0-9A-Za-z~\\u00C0-\\u00C5\\u00C7-\\u00CF\\u00D1-\\u00D6\\u00D9-\\u00DD"
1156 "\\u00E0-\\u00E5\\u00E7-\\u00EF\\u00F1-\\u00F6\\u00F9-\\u00FD\\u00FF-\\u010F"
1157 "\\u0112-\\u0125\\u0128-\\u0130\\u0134-\\u0137\\u0139-\\u013E\\u0143-\\u0148"
1158 "\\u014C-\\u0151\\u0154-\\u0165\\u0168-\\u017E\\u01A0-\\u01A1\\u01AF-\\u01B0"
1159 "\\u01CD-\\u01DC\\u01DE-\\u01E3\\u01E6-\\u01ED\\u01F0\\u01F4-\\u01F5\\u01F8-\\u01FB"
1160 "\\u0200-\\u021B\\u021E-\\u021F\\u0226-\\u0233\\u0303-\\u0304\\u0306\\u0314-\\u0315"
1161 "\\u0325\\u040E\\u0419\\u0439\\u045E\\u04C1-\\u04C2\\u04D0-\\u04D1\\u04D6-\\u04D7"
1162 "\\u04E2-\\u04E3\\u04EE-\\u04EF\\u1E00-\\u1E99\\u1EA0-\\u1EF9\\u1F01\\u1F03\\u1F05"
1163 "\\u1F07\\u1F09\\u1F0B\\u1F0D\\u1F0F\\u1F11\\u1F13\\u1F15\\u1F19\\u1F1B\\u1F1D\\u1F21"
1164 "\\u1F23\\u1F25\\u1F27\\u1F29\\u1F2B\\u1F2D\\u1F2F\\u1F31\\u1F33\\u1F35\\u1F37\\u1F39"
1165 "\\u1F3B\\u1F3D\\u1F3F\\u1F41\\u1F43\\u1F45\\u1F49\\u1F4B\\u1F4D\\u1F51\\u1F53\\u1F55"
1166 "\\u1F57\\u1F59\\u1F5B\\u1F5D\\u1F5F\\u1F61\\u1F63\\u1F65\\u1F67\\u1F69\\u1F6B\\u1F6D"
1167 "\\u1F6F\\u1F81\\u1F83\\u1F85\\u1F87\\u1F89\\u1F8B\\u1F8D\\u1F8F\\u1F91\\u1F93\\u1F95"
1168 "\\u1F97\\u1F99\\u1F9B\\u1F9D\\u1F9F\\u1FA1\\u1FA3\\u1FA5\\u1FA7\\u1FA9\\u1FAB\\u1FAD"
1169 "\\u1FAF-\\u1FB1\\u1FB8-\\u1FB9\\u1FD0-\\u1FD1\\u1FD8-\\u1FD9\\u1FE0-\\u1FE1\\u1FE5"
1170 "\\u1FE8-\\u1FE9\\u1FEC\\u212A-\\u212B\\uE04D\\uE064]"
1171 "-[\\uE000-\\uE080 \\u01E2\\u01E3]& [[:latin:][:mark:]]]";
1172
1173void TransliteratorRoundTripTest::TestDevanagariLatin() {
1174 {
1175 UErrorCode status = U_ZERO_ERROR;
1176 TransliteratorPointer t1(Transliterator::createInstance("[\\u0000-\\u00FE \\u0982\\u0983 [:Bengali:][:nonspacing mark:]];NFD;Bengali-InterIndic;InterIndic-Gujarati;NFC;( [ \\u0000-\\u00FE [:Gujarati:][[:nonspacing mark:]])",UTRANS_FORWARD, status));
1177 if(t1){
1178 TransliteratorPointer t2(t1->createInverse(status));
1179 if(U_FAILURE(status)){
1180 errln("FAIL: could not create the Inverse:-( \n");
1181 }
1182 }
1183 }
1184 RTTest test("Latin-Devanagari");
1185 Legal *legal = new LegalIndic();
1186
1187#if (U_ICU_VERSION_MAJOR_NUM==2 && U_ICU_VERSION_MINOR_NUM==6)
1188 test.test(UnicodeString(latinForIndic, ""),
1189 UnicodeString("[[:Devanagari:]&[:Age=3.2:]]", ""), NULL, this, quick,
1190 legal, 50);
1191#else
1192 test.test(UnicodeString(latinForIndic, ""),
1193 UnicodeString("[:Devanagari:]", ""), NULL, this, quick,
1194 legal, 50);
1195#endif
1196 delete legal;
1197}
1198
1199/* Defined this way for HP/UX11CC :-( */
1200static const int32_t INTER_INDIC_ARRAY_WIDTH = 4;
1201static const char * const interIndicArray[] = {
1202 "BENGALI-DEVANAGARI", "[:BENGALI:]", "[:Devanagari:]",
1203 "[\\u0951-\\u0954\\u0943-\\u0949\\u094a\\u0962\\u0963\\u090D\\u090e\\u0911\\u0912\\u0929\\u0933\\u0934\\u0935\\u093d\\u0950\\u0958\\u0959\\u095a\\u095b\\u095e\\u09f0\\u09f1]", /*roundtrip exclusions*/
1204
1205 "DEVANAGARI-BENGALI", "[:Devanagari:]", "[:BENGALI:]",
1206 "[\\u0951-\\u0954\\u09D7\\u090D\\u090e\\u0911\\u0912\\u0929\\u0933\\u0934\\u0935\\u093d\\u0950\\u0958\\u0959\\u095a\\u095b\\u095e\\u09f0\\u09f1]", /*roundtrip exclusions*/
1207
1208 "GURMUKHI-DEVANAGARI", "[:GURMUKHI:]", "[:Devanagari:]",
1209 "[\\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\\u0a72\\u0a73\\u0a74]", /*roundtrip exclusions*/
1210
1211 "DEVANAGARI-GURMUKHI", "[:Devanagari:]", "[:GURMUKHI:]",
1212 "[\\u0A02\\u0946\\u0A5C\\u0951-\\u0954\\u0A70\\u0A71\\u090B\\u090C\\u090D\\u090e\\u0911\\u0912\\u0934\\u0937\\u093D\\u0950\\u0960\\u0961\\u0a72\\u0a73\\u0a74]", /*roundtrip exclusions*/
1213
1214 "GUJARATI-DEVANAGARI", "[:GUJARATI:]", "[:Devanagari:]",
1215 "[\\u0946\\u094A\\u0962\\u0963\\u0951-\\u0954\\u0961\\u090c\\u090e\\u0912]", /*roundtrip exclusions*/
1216
1217 "DEVANAGARI-GUJARATI", "[:Devanagari:]", "[:GUJARATI:]",
1218 "[\\u0951-\\u0954\\u0961\\u090c\\u090e\\u0912]", /*roundtrip exclusions*/
1219
1220 "ORIYA-DEVANAGARI", "[:ORIYA:]", "[:Devanagari:]",
1221 "[\\u0943-\\u094a\\u0962\\u0963\\u0951-\\u0954\\u0950\\u090D\\u090e\\u0912\\u0911\\u0931\\u0935]", /*roundtrip exclusions*/
1222
1223 "DEVANAGARI-ORIYA", "[:Devanagari:]", "[:ORIYA:]",
1224 "[\\u0b5f\\u0b56\\u0b57\\u0950\\u090D\\u090e\\u0912\\u0911\\u0931\\u0935]", /*roundtrip exclusions*/
1225
1226 "Tamil-DEVANAGARI", "[:tamil:]", "[:Devanagari:]",
1227 "[\\u0901\\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]]", /*roundtrip exclusions*/
1228
1229 "DEVANAGARI-Tamil", "[:Devanagari:]", "[:tamil:]",
1230 "[\\u0bd7]", /*roundtrip exclusions*/
1231
1232 "Telugu-DEVANAGARI", "[:telugu:]", "[:Devanagari:]",
1233 "[\\u093c\\u0950\\u0945\\u0949\\u0951-\\u0954\\u0962\\u0963\\u090D\\u0911\\u093d\\u0929\\u0934[\\u0958-\\u095f]]", /*roundtrip exclusions*/
1234
1235 "DEVANAGARI-TELUGU", "[:Devanagari:]", "[:TELUGU:]",
1236 "[\\u0c55\\u0c56\\u0950\\u090D\\u0911\\u093d\\u0929\\u0934[\\u0958-\\u095f]]", /*roundtrip exclusions*/
1237
1238 "KANNADA-DEVANAGARI", "[:KANNADA:]", "[:Devanagari:]",
1239 "[\\u0901\\u0946\\u093c\\u0950\\u0945\\u0949\\u0951-\\u0954\\u0962\\u0963\\u0950\\u090D\\u0911\\u093d\\u0929\\u0934[\\u0958-\\u095f]]", /*roundtrip exclusions*/
1240
1241 "DEVANAGARI-KANNADA", "[:Devanagari:]", "[:KANNADA:]",
1242 "[\\u0cde\\u0cd5\\u0cd6\\u0950\\u090D\\u0911\\u093d\\u0929\\u0934[\\u0958-\\u095f]]", /*roundtrip exclusions*/
1243
1244 "MALAYALAM-DEVANAGARI", "[:MALAYALAM:]", "[:Devanagari:]",
1245 "[\\u0901\\u094a\\u094b\\u094c\\u093c\\u0950\\u0944\\u0945\\u0949\\u0951-\\u0954\\u0962\\u0963\\u090D\\u0911\\u093d\\u0929\\u0934[\\u0958-\\u095f]]", /*roundtrip exclusions*/
1246
1247 "DEVANAGARI-MALAYALAM", "[:Devanagari:]", "[:MALAYALAM:]",
1248 "[\\u0d4c\\u0d57\\u0950\\u090D\\u0911\\u093d\\u0929\\u0934[\\u0958-\\u095f]]", /*roundtrip exclusions*/
1249
1250 "GURMUKHI-BENGALI", "[:GURMUKHI:]", "[:BENGALI:]",
1251 "[\\u0981\\u0982\\u09b6\\u09e2\\u09e3\\u09c3\\u09c4\\u09d7\\u098B\\u098C\\u09B7\\u09E0\\u09E1\\u09F0\\u09F1]", /*roundtrip exclusions*/
1252
1253 "BENGALI-GURMUKHI", "[:BENGALI:]", "[:GURMUKHI:]",
1254 "[\\u0A02\\u0a5c\\u0a47\\u0a70\\u0a71\\u0A33\\u0A35\\u0A59\\u0A5A\\u0A5B\\u0A5E\\u0A72\\u0A73\\u0A74]", /*roundtrip exclusions*/
1255
1256 "GUJARATI-BENGALI", "[:GUJARATI:]", "[:BENGALI:]",
1257 "[\\u09d7\\u09e2\\u09e3\\u098c\\u09e1\\u09f0\\u09f1]", /*roundtrip exclusions*/
1258
1259 "BENGALI-GUJARATI", "[:BENGALI:]", "[:GUJARATI:]",
1260 "[\\u0A82\\u0a83\\u0Ac9\\u0Ac5\\u0ac7\\u0A8D\\u0A91\\u0AB3\\u0AB5\\u0ABD\\u0AD0]", /*roundtrip exclusions*/
1261
1262 "ORIYA-BENGALI", "[:ORIYA:]", "[:BENGALI:]",
1263 "[\\u09c4\\u09e2\\u09e3\\u09f0\\u09f1]", /*roundtrip exclusions*/
1264
1265 "BENGALI-ORIYA", "[:BENGALI:]", "[:ORIYA:]",
1266 "[\\u0b5f\\u0b56\\u0b33\\u0b3d]", /*roundtrip exclusions*/
1267
1268 "Tamil-BENGALI", "[:tamil:]", "[:BENGALI:]",
1269 "[\\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]", /*roundtrip exclusions*/
1270
1271 "BENGALI-Tamil", "[:BENGALI:]", "[:tamil:]",
1272 "[\\u0bc6\\u0bc7\\u0bca\\u0B8E\\u0B92\\u0BA9\\u0BB1\\u0BB3\\u0BB4\\u0BB5]", /*roundtrip exclusions*/
1273
1274 "Telugu-BENGALI", "[:telugu:]", "[:BENGALI:]",
1275 "[\\u09e2\\u09e3\\u09bc\\u09d7\\u09f0\\u09f1\\u09dc\\u09dd\\u09df]", /*roundtrip exclusions*/
1276
1277 "BENGALI-TELUGU", "[:BENGALI:]", "[:TELUGU:]",
1278 "[\\u0c55\\u0c56\\u0c47\\u0c46\\u0c4a\\u0C0E\\u0C12\\u0C31\\u0C33\\u0C35]", /*roundtrip exclusions*/
1279
1280 "KANNADA-BENGALI", "[:KANNADA:]", "[:BENGALI:]",
1281 "[\\u0981\\u09e2\\u09e3\\u09bc\\u09d7\\u09f0\\u09f1\\u09dc\\u09dd\\u09df]", /*roundtrip exclusions*/
1282
1283 "BENGALI-KANNADA", "[:BENGALI:]", "[:KANNADA:]",
1284 "[\\u0cc6\\u0cca\\u0cd5\\u0cd6\\u0cc7\\u0C8E\\u0C92\\u0CB1\\u0cb3\\u0cb5\\u0cde]", /*roundtrip exclusions*/
1285
1286 "MALAYALAM-BENGALI", "[:MALAYALAM:]", "[:BENGALI:]",
1287 "[\\u0981\\u09e2\\u09e3\\u09bc\\u09c4\\u09f0\\u09f1\\u09dc\\u09dd\\u09df]", /*roundtrip exclusions*/
1288
1289 "BENGALI-MALAYALAM", "[:BENGALI:]", "[:MALAYALAM:]",
1290 "[\\u0d46\\u0d4a\\u0d47\\u0d31-\\u0d35\\u0d0e\\u0d12]", /*roundtrip exclusions*/
1291
1292 "GUJARATI-GURMUKHI", "[:GUJARATI:]", "[:GURMUKHI:]",
1293 "[\\u0A02\\u0ab3\\u0ab6\\u0A70\\u0a71\\u0a82\\u0a83\\u0ac3\\u0ac4\\u0ac5\\u0ac9\\u0a5c\\u0a72\\u0a73\\u0a74\\u0a8b\\u0a8d\\u0a91\\u0abd]", /*roundtrip exclusions*/
1294
1295 "GURMUKHI-GUJARATI", "[:GURMUKHI:]", "[:GUJARATI:]",
1296 "[\\u0A81\\u0A82\\u0ab3\\u0ab6\\u0A70\\u0a71\\u0a82\\u0a83\\u0ac3\\u0ac4\\u0ac5\\u0ac9\\u0a5c\\u0a72\\u0a73\\u0a74\\u0a8b\\u0a8d\\u0a91\\u0ab7\\u0abd\\u0ad0\\u0ae0]", /*roundtrip exclusions*/
1297
1298 "ORIYA-GURMUKHI", "[:ORIYA:]", "[:GURMUKHI:]",
1299 "[\\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*/
1300
1301 "GURMUKHI-ORIYA", "[:GURMUKHI:]", "[:ORIYA:]",
1302 "[\\u0b01\\u0b02\\u0b03\\u0b33\\u0b36\\u0b43\\u0b56\\u0b57\\u0B0B\\u0B0C\\u0B37\\u0B3D\\u0B5F\\u0B60\\u0B61]", /*roundtrip exclusions*/
1303
1304 "TAMIL-GURMUKHI", "[:TAMIL:]", "[:GURMUKHI:]",
1305 "[\\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*/
1306
1307 "GURMUKHI-TAMIL", "[:GURMUKHI:]", "[:TAMIL:]",
1308 "[\\u0b82\\u0bc6\\u0bca\\u0bd7\\u0bb7\\u0bb3\\u0b83\\u0B8E\\u0B92\\u0BA9\\u0BB1\\u0BB4]", /*roundtrip exclusions*/
1309
1310 "TELUGU-GURMUKHI", "[:TELUGU:]", "[:GURMUKHI:]",
1311 "[\\u0A02\\u0a33\\u0a36\\u0a3c\\u0a70\\u0a71\\u0A59\\u0A5A\\u0A5B\\u0A5C\\u0A5E\\u0A72\\u0A73\\u0A74]", /*roundtrip exclusions*/
1312
1313 "GURMUKHI-TELUGU", "[:GURMUKHI:]", "[:TELUGU:]",
1314 "[\\u0c01\\u0c02\\u0c03\\u0c33\\u0c36\\u0c44\\u0c43\\u0c46\\u0c4a\\u0c56\\u0c55\\u0C0B\\u0C0C\\u0C0E\\u0C12\\u0C31\\u0C37\\u0C60\\u0C61]", /*roundtrip exclusions*/
1315
1316 "KANNADA-GURMUKHI", "[:KANNADA:]", "[:GURMUKHI:]",
1317 "[\\u0A02\\u0a33\\u0a36\\u0a3c\\u0a70\\u0a71\\u0A59\\u0A5A\\u0A5B\\u0A5C\\u0A5E\\u0A72\\u0A73\\u0A74]", /*roundtrip exclusions*/
1318
1319 "GURMUKHI-KANNADA", "[:GURMUKHI:]", "[:KANNADA:]",
1320 "[\\u0c82\\u0c83\\u0cb3\\u0cb6\\u0cc4\\u0cc3\\u0cc6\\u0cca\\u0cd5\\u0cd6\\u0C8B\\u0C8C\\u0C8E\\u0C92\\u0CB1\\u0CB7\\u0CE0\\u0CE1]", /*roundtrip exclusions*/
1321
1322 "MALAYALAM-GURMUKHI", "[:MALAYALAM:]", "[:GURMUKHI:]",
1323 "[\\u0A02\\u0a4b\\u0a4c\\u0a33\\u0a36\\u0a3c\\u0a70\\u0a71\\u0A59\\u0A5A\\u0A5B\\u0A5C\\u0A5E\\u0A72\\u0A73\\u0A74]", /*roundtrip exclusions*/
1324
1325 "GURMUKHI-MALAYALAM", "[:GURMUKHI:]", "[:MALAYALAM:]",
1326 "[\\u0d02\\u0d03\\u0d33\\u0d36\\u0d43\\u0d46\\u0d4a\\u0d4c\\u0d57\\u0D0B\\u0D0C\\u0D0E\\u0D12\\u0D31\\u0D34\\u0D37\\u0D60\\u0D61]", /*roundtrip exclusions*/
1327
1328 "GUJARATI-ORIYA", "[:GUJARATI:]", "[:ORIYA:]",
1329 "[\\u0b56\\u0b57\\u0B0C\\u0B5F\\u0B61]", /*roundtrip exclusions*/
1330
1331 "ORIYA-GUJARATI", "[:ORIYA:]", "[:GUJARATI:]",
1332 "[\\u0Ac4\\u0Ac5\\u0Ac9\\u0Ac7\\u0A8D\\u0A91\\u0AB5\\u0Ad0]", /*roundtrip exclusions*/
1333
1334 "TAMIL-GUJARATI", "[:TAMIL:]", "[:GUJARATI:]",
1335 "[\\u0A81\\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]", /*roundtrip exclusions*/
1336
1337 "GUJARATI-TAMIL", "[:GUJARATI:]", "[:TAMIL:]",
1338 "[\\u0Bc6\\u0Bca\\u0Bd7\\u0B8E\\u0B92\\u0BA9\\u0BB1\\u0BB4]", /*roundtrip exclusions*/
1339
1340 "TELUGU-GUJARATI", "[:TELUGU:]", "[:GUJARATI:]",
1341 "[\\u0abc\\u0Ac5\\u0Ac9\\u0A8D\\u0A91\\u0ABD\\u0Ad0]", /*roundtrip exclusions*/
1342
1343 "GUJARATI-TELUGU", "[:GUJARATI:]", "[:TELUGU:]",
1344 "[\\u0c46\\u0c4a\\u0c55\\u0c56\\u0C0C\\u0C0E\\u0C12\\u0C31\\u0C61]", /*roundtrip exclusions*/
1345
1346 "KANNADA-GUJARATI", "[:KANNADA:]", "[:GUJARATI:]",
1347 "[\\u0A81\\u0abc\\u0Ac5\\u0Ac9\\u0A8D\\u0A91\\u0ABD\\u0Ad0]", /*roundtrip exclusions*/
1348
1349 "GUJARATI-KANNADA", "[:GUJARATI:]", "[:KANNADA:]",
1350 "[\\u0cc6\\u0cca\\u0cd5\\u0cd6\\u0C8C\\u0C8E\\u0C92\\u0CB1\\u0CDE\\u0CE1]", /*roundtrip exclusions*/
1351
1352 "MALAYALAM-GUJARATI", "[:MALAYALAM:]", "[:GUJARATI:]",
1353 "[\\u0A81\\u0ac4\\u0acb\\u0acc\\u0abc\\u0Ac5\\u0Ac9\\u0A8D\\u0A91\\u0ABD\\u0Ad0]", /*roundtrip exclusions*/
1354
1355 "GUJARATI-MALAYALAM", "[:GUJARATI:]", "[:MALAYALAM:]",
1356 "[\\u0d46\\u0d4a\\u0d4c\\u0d55\\u0d57\\u0D0C\\u0D0E\\u0D12\\u0D31\\u0D34\\u0D61]", /*roundtrip exclusions*/
1357
1358 "TAMIL-ORIYA", "[:TAMIL:]", "[:ORIYA:]",
1359 "[\\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]", /*roundtrip exclusions*/
1360
1361 "ORIYA-TAMIL", "[:ORIYA:]", "[:TAMIL:]",
1362 "[\\u0bc6\\u0bca\\u0bc7\\u0B8E\\u0B92\\u0BA9\\u0BB1\\u0BB4\\u0BB5]", /*roundtrip exclusions*/
1363
1364 "TELUGU-ORIYA", "[:TELUGU:]", "[:ORIYA:]",
1365 "[\\u0b3c\\u0b57\\u0b56\\u0B3D\\u0B5C\\u0B5D\\u0B5F]", /*roundtrip exclusions*/
1366
1367 "ORIYA-TELUGU", "[:ORIYA:]", "[:TELUGU:]",
1368 "[\\u0c44\\u0c46\\u0c4a\\u0c55\\u0c47\\u0C0E\\u0C12\\u0C31\\u0C35]", /*roundtrip exclusions*/
1369
1370 "KANNADA-ORIYA", "[:KANNADA:]", "[:ORIYA:]",
1371 "[\\u0B01\\u0b3c\\u0b57\\u0B3D\\u0B5C\\u0B5D\\u0B5F]", /*roundtrip exclusions*/
1372
1373 "ORIYA-KANNADA", "[:ORIYA:]", "[:KANNADA:]",
1374 "[\\u0cc4\\u0cc6\\u0cca\\u0cd5\\u0cc7\\u0C8E\\u0C92\\u0CB1\\u0CB5\\u0CDE]", /*roundtrip exclusions*/
1375
1376 "MALAYALAM-ORIYA", "[:MALAYALAM:]", "[:ORIYA:]",
1377 "[\\u0B01\\u0b3c\\u0b56\\u0B3D\\u0B5C\\u0B5D\\u0B5F]", /*roundtrip exclusions*/
1378
1379 "ORIYA-MALAYALAM", "[:ORIYA:]", "[:MALAYALAM:]",
1380 "[\\u0D47\\u0D46\\u0D4a\\u0D0E\\u0D12\\u0D31\\u0D34\\u0D35]", /*roundtrip exclusions*/
1381
1382 "TELUGU-TAMIL", "[:TELUGU:]", "[:TAMIL:]",
1383 "[\\u0bd7\\u0ba9\\u0bb4]", /*roundtrip exclusions*/
1384
1385 "TAMIL-TELUGU", "[:TAMIL:]", "[:TELUGU:]",
1386 "[\\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*/
1387
1388 "KANNADA-TAMIL", "[:KANNADA:]", "[:TAMIL:]",
1389 "[\\u0bd7\\u0bc6\\u0ba9\\u0bb4]", /*roundtrip exclusions*/
1390
1391 "TAMIL-KANNADA", "[:TAMIL:]", "[:KANNADA:]",
1392 "[\\u0cc3\\u0cc4\\u0cc6\\u0cc7\\u0cd5\\u0cd6\\u0C8B\\u0C8C\\u0C96\\u0C97\\u0C98\\u0C9B\\u0C9D\\u0CA0\\u0CA1\\u0CA2\\u0CA5\\u0CA6\\u0CA7\\u0CAB\\u0CAC\\u0CAD\\u0CB6\\u0CDE\\u0CE0\\u0CE1]", /*roundtrip exclusions*/
1393
1394 "MALAYALAM-TAMIL", "[:MALAYALAM:]", "[:TAMIL:]",
1395 "[\\u0ba9]", /*roundtrip exclusions*/
1396
1397 "TAMIL-MALAYALAM", "[:TAMIL:]", "[:MALAYALAM:]",
1398 "[\\u0d43\\u0d12\\u0D0B\\u0D0C\\u0D16\\u0D17\\u0D18\\u0D1B\\u0D1D\\u0D20\\u0D21\\u0D22\\u0D25\\u0D26\\u0D27\\u0D2B\\u0D2C\\u0D2D\\u0D36\\u0D60\\u0D61]", /*roundtrip exclusions*/
1399
1400 "KANNADA-TELUGU", "[:KANNADA:]", "[:TELUGU:]",
1401 "[\\u0C01\\u0c3f\\u0c46\\u0c48\\u0c4a]", /*roundtrip exclusions*/
1402
1403 "TELUGU-KANNADA", "[:TELUGU:]", "[:KANNADA:]",
1404 "[\\u0cc8\\u0cd5\\u0cd6\\u0CDE]", /*roundtrip exclusions*/
1405
1406 "MALAYALAM-TELUGU", "[:MALAYALAM:]", "[:TELUGU:]",
1407 "[\\u0C01\\u0c44\\u0c4a\\u0c4c\\u0c4b\\u0c55\\u0c56]", /*roundtrip exclusions*/
1408
1409 "TELUGU-MALAYALAM", "[:TELUGU:]", "[:MALAYALAM:]",
1410 "[\\u0d4c\\u0d57\\u0D34]", /*roundtrip exclusions*/
1411
1412 "MALAYALAM-KANNADA", "[:MALAYALAM:]", "[:KANNADA:]",
1413 "[\\u0cc4\\u0cc6\\u0cca\\u0ccc\\u0ccb\\u0cd5\\u0cd6\\u0cDe]", /*roundtrip exclusions*/
1414
1415 "KANNADA-MALAYALAM", "[:KANNADA:]", "[:MALAYALAM:]",
1416 "[\\u0d4c\\u0d57\\u0d46\\u0D34]", /*roundtrip exclusions*/
1417
1418 "Latin-Bengali",latinForIndic, "[[:Bengali:][\\u0964\\u0965]]",
1419 "[\\u0965\\u09f0\\u09f1]" /*roundtrip exclusions*/ ,
1420
1421 "Latin-Gurmukhi", latinForIndic, "[[:Gurmukhi:][\\u0964\\u0965]]",
1422 "[\\u0965\\u0a02\\u0a72\\u0a73\\u0a74]" /*roundtrip exclusions*/,
1423
1424 "Latin-Gujarati",latinForIndic, "[[:Gujarati:][\\u0964\\u0965]]",
1425 "[\\u0965]" /*roundtrip exclusions*/,
1426
1427 "Latin-Oriya",latinForIndic, "[[:Oriya:][\\u0964\\u0965]]",
1428 "[\\u0965]" /*roundtrip exclusions*/,
1429
1430 "Latin-Tamil",latinForIndic, "[:Tamil:]",
1431 NULL /*roundtrip exclusions*/,
1432
1433 "Latin-Telugu",latinForIndic, "[:Telugu:]",
1434 NULL /*roundtrip exclusions*/,
1435
1436 "Latin-Kannada",latinForIndic, "[:Kannada:]",
1437 NULL /*roundtrip exclusions*/,
1438
1439 "Latin-Malayalam",latinForIndic, "[:Malayalam:]",
1440 NULL /*roundtrip exclusions*/
1441
1442};
1443
1444void TransliteratorRoundTripTest::TestDebug(const char* name,const char fromSet[],
1445 const char* toSet,const char* exclusions){
1446
1447 RTTest test(name);
1448 Legal *legal = new LegalIndic();
1449 test.test(UnicodeString(fromSet,""),UnicodeString(toSet,""),exclusions,this,quick,legal);
1450}
1451
1452void TransliteratorRoundTripTest::TestInterIndic() {
1453 //TestDebug("Latin-Gurmukhi", latinForIndic, "[:Gurmukhi:]","[\\u0965\\u0a02\\u0a72\\u0a73\\u0a74]",TRUE);
1454 int32_t num = (int32_t)(sizeof(interIndicArray)/(INTER_INDIC_ARRAY_WIDTH*sizeof(char*)));
1455 if(quick){
1456 logln("Testing only 5 of %i. Skipping rest (use -e for exhaustive)",num);
1457 num = 5;
1458 }
1459 for(int i = 0; i < num;i++){
1460 RTTest test(interIndicArray[i*INTER_INDIC_ARRAY_WIDTH + 0]);
1461 Legal *legal = new LegalIndic();
1462#if (U_ICU_VERSION_MAJOR_NUM==2 && U_ICU_VERSION_MINOR_NUM==6)
1463 UnicodeString temp1 = "[";
1464 temp1.append(interIndicArray[i*INTER_INDIC_ARRAY_WIDTH + 1]);
1465 temp1.append("& [:Age=3.2:]]");
1466 UnicodeString temp2 = "[";
1467 temp2.append(interIndicArray[i*INTER_INDIC_ARRAY_WIDTH + 2]);
1468 temp2.append("& [:Age=3.2:]]");
1469#else
1470 UnicodeString temp1 = interIndicArray[i*INTER_INDIC_ARRAY_WIDTH + 1];
1471 UnicodeString temp2 = interIndicArray[i*INTER_INDIC_ARRAY_WIDTH + 2];
1472#endif
1473 test.test(temp1,
1474 temp2,
1475 interIndicArray[i*INTER_INDIC_ARRAY_WIDTH + 3], // roundtrip exclusions
1476 this, quick, legal, 50);
1477 delete legal;
1478 }
1479
1480}
1481
1482// end indic tests ----------------------------------------------------------
1483
1484#endif /* #if !UCONFIG_NO_TRANSLITERATION */
mirror of ICU