1 /********************************************************************
2 * COPYRIGHT:
3 * Copyright (c) 2005-2010, International Business Machines Corporation and
4 * others. All Rights Reserved.
5 ********************************************************************/
6 /************************************************************************
7 * Tests for the UText and UTextIterator text abstraction classses
8 *
9 ************************************************************************/
10
11 #include <string.h>
12 #include <stdio.h>
13 #include <stdlib.h>
14 #include "unicode/utypes.h"
15 #include "unicode/utext.h"
16 #include "unicode/utf8.h"
17 #include "unicode/ustring.h"
18 #include "unicode/uchriter.h"
19 #include "utxttest.h"
20
21 static UBool gFailed = FALSE;
22 static int gTestNum = 0;
23
24 // Forward decl
25 UText *openFragmentedUnicodeString(UText *ut, UnicodeString *s, UErrorCode *status);
26
27 #define TEST_ASSERT(x) \
28 { if ((x)==FALSE) {errln("Test #%d failure in file %s at line %d\n", gTestNum, __FILE__, __LINE__);\
29 gFailed = TRUE;\
30 }}
31
32
33 #define TEST_SUCCESS(status) \
34 { if (U_FAILURE(status)) {errln("Test #%d failure in file %s at line %d. Error = \"%s\"\n", \
35 gTestNum, __FILE__, __LINE__, u_errorName(status)); \
36 gFailed = TRUE;\
37 }}
38
UTextTest()39 UTextTest::UTextTest() {
40 }
41
~UTextTest()42 UTextTest::~UTextTest() {
43 }
44
45
46 void
runIndexedTest(int32_t index,UBool exec,const char * & name,char *)47 UTextTest::runIndexedTest(int32_t index, UBool exec,
48 const char* &name, char* /*par*/) {
49 switch (index) {
50 case 0: name = "TextTest";
51 if (exec) TextTest(); break;
52 case 1: name = "ErrorTest";
53 if (exec) ErrorTest(); break;
54 case 2: name = "FreezeTest";
55 if (exec) FreezeTest(); break;
56 case 3: name = "Ticket5560";
57 if (exec) Ticket5560(); break;
58 case 4: name = "Ticket6847";
59 if (exec) Ticket6847(); break;
60 case 5: name = "ComparisonTest";
61 if (exec) ComparisonTest(); break;
62 default: name = ""; break;
63 }
64 }
65
66 //
67 // Quick and dirty random number generator.
68 // (don't use library so that results are portable.
69 static uint32_t m_seed = 1;
m_rand()70 static uint32_t m_rand()
71 {
72 m_seed = m_seed * 1103515245 + 12345;
73 return (uint32_t)(m_seed/65536) % 32768;
74 }
75
76
77 //
78 // TextTest()
79 //
80 // Top Level function for UText testing.
81 // Specifies the strings to be tested, with the acutal testing itself
82 // being carried out in another function, TestString().
83 //
TextTest()84 void UTextTest::TextTest() {
85 int32_t i, j;
86
87 TestString("abcd\\U00010001xyz");
88 TestString("");
89
90 // Supplementary chars at start or end
91 TestString("\\U00010001");
92 TestString("abc\\U00010001");
93 TestString("\\U00010001abc");
94
95 // Test simple strings of lengths 1 to 60, looking for glitches at buffer boundaries
96 UnicodeString s;
97 for (i=1; i<60; i++) {
98 s.truncate(0);
99 for (j=0; j<i; j++) {
100 if (j+0x30 == 0x5c) {
101 // backslash. Needs to be escaped
102 s.append((UChar)0x5c);
103 }
104 s.append(UChar(j+0x30));
105 }
106 TestString(s);
107 }
108
109 // Test strings with odd-aligned supplementary chars,
110 // looking for glitches at buffer boundaries
111 for (i=1; i<60; i++) {
112 s.truncate(0);
113 s.append((UChar)0x41);
114 for (j=0; j<i; j++) {
115 s.append(UChar32(j+0x11000));
116 }
117 TestString(s);
118 }
119
120 // String of chars of randomly varying size in utf-8 representation.
121 // Exercise the mapping, and the varying sized buffer.
122 //
123 s.truncate(0);
124 UChar32 c1 = 0;
125 UChar32 c2 = 0x100;
126 UChar32 c3 = 0xa000;
127 UChar32 c4 = 0x11000;
128 for (i=0; i<1000; i++) {
129 int len8 = m_rand()%4 + 1;
130 switch (len8) {
131 case 1:
132 c1 = (c1+1)%0x80;
133 // don't put 0 into string (0 terminated strings for some tests)
134 // don't put '\', will cause unescape() to fail.
135 if (c1==0x5c || c1==0) {
136 c1++;
137 }
138 s.append(c1);
139 break;
140 case 2:
141 s.append(c2++);
142 break;
143 case 3:
144 s.append(c3++);
145 break;
146 case 4:
147 s.append(c4++);
148 break;
149 }
150 }
151 TestString(s);
152 }
153
154
155 //
156 // TestString() Run a suite of UText tests on a string.
157 // The test string is unescaped before use.
158 //
TestString(const UnicodeString & s)159 void UTextTest::TestString(const UnicodeString &s) {
160 int32_t i;
161 int32_t j;
162 UChar32 c;
163 int32_t cpCount = 0;
164 UErrorCode status = U_ZERO_ERROR;
165 UText *ut = NULL;
166 int32_t saLen;
167
168 UnicodeString sa = s.unescape();
169 saLen = sa.length();
170
171 //
172 // Build up a mapping between code points and UTF-16 code unit indexes.
173 //
174 m *cpMap = new m[sa.length() + 1];
175 j = 0;
176 for (i=0; i<sa.length(); i=sa.moveIndex32(i, 1)) {
177 c = sa.char32At(i);
178 cpMap[j].nativeIdx = i;
179 cpMap[j].cp = c;
180 j++;
181 cpCount++;
182 }
183 cpMap[j].nativeIdx = i; // position following the last char in utf-16 string.
184
185
186 // UChar * test, null terminated
187 status = U_ZERO_ERROR;
188 UChar *buf = new UChar[saLen+1];
189 sa.extract(buf, saLen+1, status);
190 TEST_SUCCESS(status);
191 ut = utext_openUChars(NULL, buf, -1, &status);
192 TEST_SUCCESS(status);
193 TestAccess(sa, ut, cpCount, cpMap);
194 utext_close(ut);
195 delete [] buf;
196
197 // UChar * test, with length
198 status = U_ZERO_ERROR;
199 buf = new UChar[saLen+1];
200 sa.extract(buf, saLen+1, status);
201 TEST_SUCCESS(status);
202 ut = utext_openUChars(NULL, buf, saLen, &status);
203 TEST_SUCCESS(status);
204 TestAccess(sa, ut, cpCount, cpMap);
205 utext_close(ut);
206 delete [] buf;
207
208
209 // UnicodeString test
210 status = U_ZERO_ERROR;
211 ut = utext_openUnicodeString(NULL, &sa, &status);
212 TEST_SUCCESS(status);
213 TestAccess(sa, ut, cpCount, cpMap);
214 TestCMR(sa, ut, cpCount, cpMap, cpMap);
215 utext_close(ut);
216
217
218 // Const UnicodeString test
219 status = U_ZERO_ERROR;
220 ut = utext_openConstUnicodeString(NULL, &sa, &status);
221 TEST_SUCCESS(status);
222 TestAccess(sa, ut, cpCount, cpMap);
223 utext_close(ut);
224
225
226 // Replaceable test. (UnicodeString inherits Replaceable)
227 status = U_ZERO_ERROR;
228 ut = utext_openReplaceable(NULL, &sa, &status);
229 TEST_SUCCESS(status);
230 TestAccess(sa, ut, cpCount, cpMap);
231 TestCMR(sa, ut, cpCount, cpMap, cpMap);
232 utext_close(ut);
233
234 // Character Iterator Tests
235 status = U_ZERO_ERROR;
236 const UChar *cbuf = sa.getBuffer();
237 CharacterIterator *ci = new UCharCharacterIterator(cbuf, saLen, status);
238 TEST_SUCCESS(status);
239 ut = utext_openCharacterIterator(NULL, ci, &status);
240 TEST_SUCCESS(status);
241 TestAccess(sa, ut, cpCount, cpMap);
242 utext_close(ut);
243 delete ci;
244
245
246 // Fragmented UnicodeString (Chunk size of one)
247 //
248 status = U_ZERO_ERROR;
249 ut = openFragmentedUnicodeString(NULL, &sa, &status);
250 TEST_SUCCESS(status);
251 TestAccess(sa, ut, cpCount, cpMap);
252 utext_close(ut);
253
254 //
255 // UTF-8 test
256 //
257
258 // Convert the test string from UnicodeString to (char *) in utf-8 format
259 int32_t u8Len = sa.extract(0, sa.length(), NULL, 0, "utf-8");
260 char *u8String = new char[u8Len + 1];
261 sa.extract(0, sa.length(), u8String, u8Len+1, "utf-8");
262
263 // Build up the map of code point indices in the utf-8 string
264 m * u8Map = new m[sa.length() + 1];
265 i = 0; // native utf-8 index
266 for (j=0; j<cpCount ; j++) { // code point number
267 u8Map[j].nativeIdx = i;
268 U8_NEXT(u8String, i, u8Len, c)
269 u8Map[j].cp = c;
270 }
271 u8Map[cpCount].nativeIdx = u8Len; // position following the last char in utf-8 string.
272
273 // Do the test itself
274 status = U_ZERO_ERROR;
275 ut = utext_openUTF8(NULL, u8String, -1, &status);
276 TEST_SUCCESS(status);
277 TestAccess(sa, ut, cpCount, u8Map);
278 utext_close(ut);
279
280
281
282 delete []cpMap;
283 delete []u8Map;
284 delete []u8String;
285 }
286
287 // TestCMR test Copy, Move and Replace operations.
288 // us UnicodeString containing the test text.
289 // ut UText containing the same test text.
290 // cpCount number of code points in the test text.
291 // nativeMap Mapping from code points to native indexes for the UText.
292 // u16Map Mapping from code points to UTF-16 indexes, for use with the UnicodeString.
293 //
294 // This function runs a whole series of opertions on each incoming UText.
295 // The UText is deep-cloned prior to each operation, so that the original UText remains unchanged.
296 //
TestCMR(const UnicodeString & us,UText * ut,int cpCount,m * nativeMap,m * u16Map)297 void UTextTest::TestCMR(const UnicodeString &us, UText *ut, int cpCount, m *nativeMap, m *u16Map) {
298 TEST_ASSERT(utext_isWritable(ut) == TRUE);
299
300 int srcLengthType; // Loop variables for selecting the postion and length
301 int srcPosType; // of the block to operate on within the source text.
302 int destPosType;
303
304 int srcIndex = 0; // Code Point indexes of the block to operate on for
305 int srcLength = 0; // a specific test.
306
307 int destIndex = 0; // Code point index of the destination for a copy/move test.
308
309 int32_t nativeStart = 0; // Native unit indexes for a test.
310 int32_t nativeLimit = 0;
311 int32_t nativeDest = 0;
312
313 int32_t u16Start = 0; // UTF-16 indexes for a test.
314 int32_t u16Limit = 0; // used when performing the same operation in a Unicode String
315 int32_t u16Dest = 0;
316
317 // Iterate over a whole series of source index, length and a target indexes.
318 // This is done with code point indexes; these will be later translated to native
319 // indexes using the cpMap.
320 for (srcLengthType=1; srcLengthType<=3; srcLengthType++) {
321 switch (srcLengthType) {
322 case 1: srcLength = 1; break;
323 case 2: srcLength = 5; break;
324 case 3: srcLength = cpCount / 3;
325 }
326 for (srcPosType=1; srcPosType<=5; srcPosType++) {
327 switch (srcPosType) {
328 case 1: srcIndex = 0; break;
329 case 2: srcIndex = 1; break;
330 case 3: srcIndex = cpCount - srcLength; break;
331 case 4: srcIndex = cpCount - srcLength - 1; break;
332 case 5: srcIndex = cpCount / 2; break;
333 }
334 if (srcIndex < 0 || srcIndex + srcLength > cpCount) {
335 // filter out bogus test cases -
336 // those with a source range that falls of an edge of the string.
337 continue;
338 }
339
340 //
341 // Copy and move tests.
342 // iterate over a variety of destination positions.
343 //
344 for (destPosType=1; destPosType<=4; destPosType++) {
345 switch (destPosType) {
346 case 1: destIndex = 0; break;
347 case 2: destIndex = 1; break;
348 case 3: destIndex = srcIndex - 1; break;
349 case 4: destIndex = srcIndex + srcLength + 1; break;
350 case 5: destIndex = cpCount-1; break;
351 case 6: destIndex = cpCount; break;
352 }
353 if (destIndex<0 || destIndex>cpCount) {
354 // filter out bogus test cases.
355 continue;
356 }
357
358 nativeStart = nativeMap[srcIndex].nativeIdx;
359 nativeLimit = nativeMap[srcIndex+srcLength].nativeIdx;
360 nativeDest = nativeMap[destIndex].nativeIdx;
361
362 u16Start = u16Map[srcIndex].nativeIdx;
363 u16Limit = u16Map[srcIndex+srcLength].nativeIdx;
364 u16Dest = u16Map[destIndex].nativeIdx;
365
366 gFailed = FALSE;
367 TestCopyMove(us, ut, FALSE,
368 nativeStart, nativeLimit, nativeDest,
369 u16Start, u16Limit, u16Dest);
370
371 TestCopyMove(us, ut, TRUE,
372 nativeStart, nativeLimit, nativeDest,
373 u16Start, u16Limit, u16Dest);
374
375 if (gFailed) {
376 return;
377 }
378 }
379
380 //
381 // Replace tests.
382 //
383 UnicodeString fullRepString("This is an arbitrary string that will be used as replacement text");
384 for (int32_t replStrLen=0; replStrLen<20; replStrLen++) {
385 UnicodeString repStr(fullRepString, 0, replStrLen);
386 TestReplace(us, ut,
387 nativeStart, nativeLimit,
388 u16Start, u16Limit,
389 repStr);
390 if (gFailed) {
391 return;
392 }
393 }
394
395 }
396 }
397
398 }
399
400 //
401 // TestCopyMove run a single test case for utext_copy.
402 // Test cases are created in TestCMR and dispatched here for execution.
403 //
TestCopyMove(const UnicodeString & us,UText * ut,UBool move,int32_t nativeStart,int32_t nativeLimit,int32_t nativeDest,int32_t u16Start,int32_t u16Limit,int32_t u16Dest)404 void UTextTest::TestCopyMove(const UnicodeString &us, UText *ut, UBool move,
405 int32_t nativeStart, int32_t nativeLimit, int32_t nativeDest,
406 int32_t u16Start, int32_t u16Limit, int32_t u16Dest)
407 {
408 UErrorCode status = U_ZERO_ERROR;
409 UText *targetUT = NULL;
410 gTestNum++;
411 gFailed = FALSE;
412
413 //
414 // clone the UText. The test will be run in the cloned copy
415 // so that we don't alter the original.
416 //
417 targetUT = utext_clone(NULL, ut, TRUE, FALSE, &status);
418 TEST_SUCCESS(status);
419 UnicodeString targetUS(us); // And copy the reference string.
420
421 // do the test operation first in the reference
422 targetUS.copy(u16Start, u16Limit, u16Dest);
423 if (move) {
424 // delete out the source range.
425 if (u16Limit < u16Dest) {
426 targetUS.removeBetween(u16Start, u16Limit);
427 } else {
428 int32_t amtCopied = u16Limit - u16Start;
429 targetUS.removeBetween(u16Start+amtCopied, u16Limit+amtCopied);
430 }
431 }
432
433 // Do the same operation in the UText under test
434 utext_copy(targetUT, nativeStart, nativeLimit, nativeDest, move, &status);
435 if (nativeDest > nativeStart && nativeDest < nativeLimit) {
436 TEST_ASSERT(status == U_INDEX_OUTOFBOUNDS_ERROR);
437 } else {
438 TEST_SUCCESS(status);
439
440 // Compare the results of the two parallel tests
441 int32_t usi = 0; // UnicodeString postion, utf-16 index.
442 int64_t uti = 0; // UText position, native index.
443 int32_t cpi; // char32 position (code point index)
444 UChar32 usc; // code point from Unicode String
445 UChar32 utc; // code point from UText
446 utext_setNativeIndex(targetUT, 0);
447 for (cpi=0; ; cpi++) {
448 usc = targetUS.char32At(usi);
449 utc = utext_next32(targetUT);
450 if (utc < 0) {
451 break;
452 }
453 TEST_ASSERT(uti == usi);
454 TEST_ASSERT(utc == usc);
455 usi = targetUS.moveIndex32(usi, 1);
456 uti = utext_getNativeIndex(targetUT);
457 if (gFailed) {
458 goto cleanupAndReturn;
459 }
460 }
461 int64_t expectedNativeLength = utext_nativeLength(ut);
462 if (move == FALSE) {
463 expectedNativeLength += nativeLimit - nativeStart;
464 }
465 uti = utext_getNativeIndex(targetUT);
466 TEST_ASSERT(uti == expectedNativeLength);
467 }
468
469 cleanupAndReturn:
470 utext_close(targetUT);
471 }
472
473
474 //
475 // TestReplace Test a single Replace operation.
476 //
TestReplace(const UnicodeString & us,UText * ut,int32_t nativeStart,int32_t nativeLimit,int32_t u16Start,int32_t u16Limit,const UnicodeString & repStr)477 void UTextTest::TestReplace(
478 const UnicodeString &us, // reference UnicodeString in which to do the replace
479 UText *ut, // UnicodeText object under test.
480 int32_t nativeStart, // Range to be replaced, in UText native units.
481 int32_t nativeLimit,
482 int32_t u16Start, // Range to be replaced, in UTF-16 units
483 int32_t u16Limit, // for use in the reference UnicodeString.
484 const UnicodeString &repStr) // The replacement string
485 {
486 UErrorCode status = U_ZERO_ERROR;
487 UText *targetUT = NULL;
488 gTestNum++;
489 gFailed = FALSE;
490
491 //
492 // clone the target UText. The test will be run in the cloned copy
493 // so that we don't alter the original.
494 //
495 targetUT = utext_clone(NULL, ut, TRUE, FALSE, &status);
496 TEST_SUCCESS(status);
497 UnicodeString targetUS(us); // And copy the reference string.
498
499 //
500 // Do the replace operation in the Unicode String, to
501 // produce a reference result.
502 //
503 targetUS.replace(u16Start, u16Limit-u16Start, repStr);
504
505 //
506 // Do the replace on the UText under test
507 //
508 const UChar *rs = repStr.getBuffer();
509 int32_t rsLen = repStr.length();
510 int32_t actualDelta = utext_replace(targetUT, nativeStart, nativeLimit, rs, rsLen, &status);
511 int32_t expectedDelta = repStr.length() - (nativeLimit - nativeStart);
512 TEST_ASSERT(actualDelta == expectedDelta);
513
514 //
515 // Compare the results
516 //
517 int32_t usi = 0; // UnicodeString postion, utf-16 index.
518 int64_t uti = 0; // UText position, native index.
519 int32_t cpi; // char32 position (code point index)
520 UChar32 usc; // code point from Unicode String
521 UChar32 utc; // code point from UText
522 int64_t expectedNativeLength = 0;
523 utext_setNativeIndex(targetUT, 0);
524 for (cpi=0; ; cpi++) {
525 usc = targetUS.char32At(usi);
526 utc = utext_next32(targetUT);
527 if (utc < 0) {
528 break;
529 }
530 TEST_ASSERT(uti == usi);
531 TEST_ASSERT(utc == usc);
532 usi = targetUS.moveIndex32(usi, 1);
533 uti = utext_getNativeIndex(targetUT);
534 if (gFailed) {
535 goto cleanupAndReturn;
536 }
537 }
538 expectedNativeLength = utext_nativeLength(ut) + expectedDelta;
539 uti = utext_getNativeIndex(targetUT);
540 TEST_ASSERT(uti == expectedNativeLength);
541
542 cleanupAndReturn:
543 utext_close(targetUT);
544 }
545
546 //
547 // TestAccess Test the read only access functions on a UText, including cloning.
548 // The text is accessed in a variety of ways, and compared with
549 // the reference UnicodeString.
550 //
TestAccess(const UnicodeString & us,UText * ut,int cpCount,m * cpMap)551 void UTextTest::TestAccess(const UnicodeString &us, UText *ut, int cpCount, m *cpMap) {
552 // Run the standard tests on the caller-supplied UText.
553 TestAccessNoClone(us, ut, cpCount, cpMap);
554
555 // Re-run tests on a shallow clone.
556 utext_setNativeIndex(ut, 0);
557 UErrorCode status = U_ZERO_ERROR;
558 UText *shallowClone = utext_clone(NULL, ut, FALSE /*deep*/, FALSE /*readOnly*/, &status);
559 TEST_SUCCESS(status);
560 TestAccessNoClone(us, shallowClone, cpCount, cpMap);
561
562 //
563 // Rerun again on a deep clone.
564 // Note that text providers are not required to provide deep cloning,
565 // so unsupported errors are ignored.
566 //
567 status = U_ZERO_ERROR;
568 utext_setNativeIndex(shallowClone, 0);
569 UText *deepClone = utext_clone(NULL, shallowClone, TRUE, FALSE, &status);
570 utext_close(shallowClone);
571 if (status != U_UNSUPPORTED_ERROR) {
572 TEST_SUCCESS(status);
573 TestAccessNoClone(us, deepClone, cpCount, cpMap);
574 }
575 utext_close(deepClone);
576 }
577
578
579 //
580 // TestAccessNoClone() Test the read only access functions on a UText.
581 // The text is accessed in a variety of ways, and compared with
582 // the reference UnicodeString.
583 //
TestAccessNoClone(const UnicodeString & us,UText * ut,int cpCount,m * cpMap)584 void UTextTest::TestAccessNoClone(const UnicodeString &us, UText *ut, int cpCount, m *cpMap) {
585 UErrorCode status = U_ZERO_ERROR;
586 gTestNum++;
587
588 //
589 // Check the length from the UText
590 //
591 int64_t expectedLen = cpMap[cpCount].nativeIdx;
592 int64_t utlen = utext_nativeLength(ut);
593 TEST_ASSERT(expectedLen == utlen);
594
595 //
596 // Iterate forwards, verify that we get the correct code points
597 // at the correct native offsets.
598 //
599 int i = 0;
600 int64_t index;
601 int64_t expectedIndex = 0;
602 int64_t foundIndex = 0;
603 UChar32 expectedC;
604 UChar32 foundC;
605 int64_t len;
606
607 for (i=0; i<cpCount; i++) {
608 expectedIndex = cpMap[i].nativeIdx;
609 foundIndex = utext_getNativeIndex(ut);
610 TEST_ASSERT(expectedIndex == foundIndex);
611 expectedC = cpMap[i].cp;
612 foundC = utext_next32(ut);
613 TEST_ASSERT(expectedC == foundC);
614 foundIndex = utext_getPreviousNativeIndex(ut);
615 TEST_ASSERT(expectedIndex == foundIndex);
616 if (gFailed) {
617 return;
618 }
619 }
620 foundC = utext_next32(ut);
621 TEST_ASSERT(foundC == U_SENTINEL);
622
623 // Repeat above, using macros
624 utext_setNativeIndex(ut, 0);
625 for (i=0; i<cpCount; i++) {
626 expectedIndex = cpMap[i].nativeIdx;
627 foundIndex = UTEXT_GETNATIVEINDEX(ut);
628 TEST_ASSERT(expectedIndex == foundIndex);
629 expectedC = cpMap[i].cp;
630 foundC = UTEXT_NEXT32(ut);
631 TEST_ASSERT(expectedC == foundC);
632 if (gFailed) {
633 return;
634 }
635 }
636 foundC = UTEXT_NEXT32(ut);
637 TEST_ASSERT(foundC == U_SENTINEL);
638
639 //
640 // Forward iteration (above) should have left index at the
641 // end of the input, which should == length().
642 //
643 len = utext_nativeLength(ut);
644 foundIndex = utext_getNativeIndex(ut);
645 TEST_ASSERT(len == foundIndex);
646
647 //
648 // Iterate backwards over entire test string
649 //
650 len = utext_getNativeIndex(ut);
651 utext_setNativeIndex(ut, len);
652 for (i=cpCount-1; i>=0; i--) {
653 expectedC = cpMap[i].cp;
654 expectedIndex = cpMap[i].nativeIdx;
655 int64_t prevIndex = utext_getPreviousNativeIndex(ut);
656 foundC = utext_previous32(ut);
657 foundIndex = utext_getNativeIndex(ut);
658 TEST_ASSERT(expectedIndex == foundIndex);
659 TEST_ASSERT(expectedC == foundC);
660 TEST_ASSERT(prevIndex == foundIndex);
661 if (gFailed) {
662 return;
663 }
664 }
665
666 //
667 // Backwards iteration, above, should have left our iterator
668 // position at zero, and continued backwards iterationshould fail.
669 //
670 foundIndex = utext_getNativeIndex(ut);
671 TEST_ASSERT(foundIndex == 0);
672 foundIndex = utext_getPreviousNativeIndex(ut);
673 TEST_ASSERT(foundIndex == 0);
674
675
676 foundC = utext_previous32(ut);
677 TEST_ASSERT(foundC == U_SENTINEL);
678 foundIndex = utext_getNativeIndex(ut);
679 TEST_ASSERT(foundIndex == 0);
680 foundIndex = utext_getPreviousNativeIndex(ut);
681 TEST_ASSERT(foundIndex == 0);
682
683
684 // And again, with the macros
685 utext_setNativeIndex(ut, len);
686 for (i=cpCount-1; i>=0; i--) {
687 expectedC = cpMap[i].cp;
688 expectedIndex = cpMap[i].nativeIdx;
689 foundC = UTEXT_PREVIOUS32(ut);
690 foundIndex = UTEXT_GETNATIVEINDEX(ut);
691 TEST_ASSERT(expectedIndex == foundIndex);
692 TEST_ASSERT(expectedC == foundC);
693 if (gFailed) {
694 return;
695 }
696 }
697
698 //
699 // Backwards iteration, above, should have left our iterator
700 // position at zero, and continued backwards iterationshould fail.
701 //
702 foundIndex = UTEXT_GETNATIVEINDEX(ut);
703 TEST_ASSERT(foundIndex == 0);
704
705 foundC = UTEXT_PREVIOUS32(ut);
706 TEST_ASSERT(foundC == U_SENTINEL);
707 foundIndex = UTEXT_GETNATIVEINDEX(ut);
708 TEST_ASSERT(foundIndex == 0);
709 if (gFailed) {
710 return;
711 }
712
713 //
714 // next32From(), prevous32From(), Iterate in a somewhat random order.
715 //
716 int cpIndex = 0;
717 for (i=0; i<cpCount; i++) {
718 cpIndex = (cpIndex + 9973) % cpCount;
719 index = cpMap[cpIndex].nativeIdx;
720 expectedC = cpMap[cpIndex].cp;
721 foundC = utext_next32From(ut, index);
722 TEST_ASSERT(expectedC == foundC);
723 if (gFailed) {
724 return;
725 }
726 }
727
728 cpIndex = 0;
729 for (i=0; i<cpCount; i++) {
730 cpIndex = (cpIndex + 9973) % cpCount;
731 index = cpMap[cpIndex+1].nativeIdx;
732 expectedC = cpMap[cpIndex].cp;
733 foundC = utext_previous32From(ut, index);
734 TEST_ASSERT(expectedC == foundC);
735 if (gFailed) {
736 return;
737 }
738 }
739
740
741 //
742 // moveIndex(int32_t delta);
743 //
744
745 // Walk through frontwards, incrementing by one
746 utext_setNativeIndex(ut, 0);
747 for (i=1; i<=cpCount; i++) {
748 utext_moveIndex32(ut, 1);
749 index = utext_getNativeIndex(ut);
750 expectedIndex = cpMap[i].nativeIdx;
751 TEST_ASSERT(expectedIndex == index);
752 index = UTEXT_GETNATIVEINDEX(ut);
753 TEST_ASSERT(expectedIndex == index);
754 }
755
756 // Walk through frontwards, incrementing by two
757 utext_setNativeIndex(ut, 0);
758 for (i=2; i<cpCount; i+=2) {
759 utext_moveIndex32(ut, 2);
760 index = utext_getNativeIndex(ut);
761 expectedIndex = cpMap[i].nativeIdx;
762 TEST_ASSERT(expectedIndex == index);
763 index = UTEXT_GETNATIVEINDEX(ut);
764 TEST_ASSERT(expectedIndex == index);
765 }
766
767 // walk through the string backwards, decrementing by one.
768 i = cpMap[cpCount].nativeIdx;
769 utext_setNativeIndex(ut, i);
770 for (i=cpCount; i>=0; i--) {
771 expectedIndex = cpMap[i].nativeIdx;
772 index = utext_getNativeIndex(ut);
773 TEST_ASSERT(expectedIndex == index);
774 index = UTEXT_GETNATIVEINDEX(ut);
775 TEST_ASSERT(expectedIndex == index);
776 utext_moveIndex32(ut, -1);
777 }
778
779
780 // walk through backwards, decrementing by three
781 i = cpMap[cpCount].nativeIdx;
782 utext_setNativeIndex(ut, i);
783 for (i=cpCount; i>=0; i-=3) {
784 expectedIndex = cpMap[i].nativeIdx;
785 index = utext_getNativeIndex(ut);
786 TEST_ASSERT(expectedIndex == index);
787 index = UTEXT_GETNATIVEINDEX(ut);
788 TEST_ASSERT(expectedIndex == index);
789 utext_moveIndex32(ut, -3);
790 }
791
792
793 //
794 // Extract
795 //
796 int bufSize = us.length() + 10;
797 UChar *buf = new UChar[bufSize];
798 status = U_ZERO_ERROR;
799 expectedLen = us.length();
800 len = utext_extract(ut, 0, utlen, buf, bufSize, &status);
801 TEST_SUCCESS(status);
802 TEST_ASSERT(len == expectedLen);
803 int compareResult = us.compare(buf, -1);
804 TEST_ASSERT(compareResult == 0);
805
806 status = U_ZERO_ERROR;
807 len = utext_extract(ut, 0, utlen, NULL, 0, &status);
808 if (utlen == 0) {
809 TEST_ASSERT(status == U_STRING_NOT_TERMINATED_WARNING);
810 } else {
811 TEST_ASSERT(status == U_BUFFER_OVERFLOW_ERROR);
812 }
813 TEST_ASSERT(len == expectedLen);
814
815 status = U_ZERO_ERROR;
816 u_memset(buf, 0x5555, bufSize);
817 len = utext_extract(ut, 0, utlen, buf, 1, &status);
818 if (us.length() == 0) {
819 TEST_SUCCESS(status);
820 TEST_ASSERT(buf[0] == 0);
821 } else {
822 // Buf len == 1, extracting a single 16 bit value.
823 // If the data char is supplementary, it doesn't matter whether the buffer remains unchanged,
824 // or whether the lead surrogate of the pair is extracted.
825 // It's a buffer overflow error in either case.
826 TEST_ASSERT(buf[0] == us.charAt(0) ||
827 (buf[0] == 0x5555 && U_IS_SUPPLEMENTARY(us.char32At(0))));
828 TEST_ASSERT(buf[1] == 0x5555);
829 if (us.length() == 1) {
830 TEST_ASSERT(status == U_STRING_NOT_TERMINATED_WARNING);
831 } else {
832 TEST_ASSERT(status == U_BUFFER_OVERFLOW_ERROR);
833 }
834 }
835
836 delete []buf;
837 }
838
839
840 //
841 // ComparisonTest() Check the string comparison functions. Based on UnicodeStringTest::TestCompare()
842 //
ComparisonTest()843 void UTextTest::ComparisonTest()
844 {
845 UErrorCode status = U_ZERO_ERROR;
846 UnicodeString test1Str("this is a test");
847 UnicodeString test2Str("this is a test");
848 UnicodeString test3Str("this is a test of the emergency broadcast system");
849 UnicodeString test4Str("never say, \"this is a test\"!!");
850
851 UText test1 = UTEXT_INITIALIZER;
852 UText test2 = UTEXT_INITIALIZER;
853 UText test3 = UTEXT_INITIALIZER;
854 UText test4 = UTEXT_INITIALIZER;
855
856 UChar uniChars[] = { 0x74, 0x68, 0x69, 0x73, 0x20, 0x69, 0x73,
857 0x20, 0x61, 0x20, 0x74, 0x65, 0x73, 0x74, 0 };
858 char chars[] = { 0x74, 0x68, 0x69, 0x73, 0x20, 0x69, 0x73,
859 0x20, 0x61, 0x20, 0x74, 0x65, 0x73, 0x74, 0 };
860
861 UText uniCharText = UTEXT_INITIALIZER;
862 UText charText = UTEXT_INITIALIZER;
863
864 utext_openUnicodeString(&test1, &test1Str, &status);
865 utext_openUnicodeString(&test2, &test2Str, &status);
866 utext_openUnicodeString(&test3, &test3Str, &status);
867 utext_openUnicodeString(&test4, &test4Str, &status);
868
869 utext_openUChars(&uniCharText, uniChars, -1, &status);
870 utext_openUTF8(&charText, chars, -1, &status);
871
872 TEST_SUCCESS(status);
873
874 // test utext_compare(), simple
875 UTEXT_SETNATIVEINDEX(&test1, 0);
876 UTEXT_SETNATIVEINDEX(&test2, 0);
877 if (utext_compare(&test1, -1, &test2, -1) != 0) errln("utext_compare() failed, simple setup");
878 UTEXT_SETNATIVEINDEX(&test1, 0);
879 UTEXT_SETNATIVEINDEX(&test3, 0);
880 if (utext_compare(&test1, -1, &test3, -1) >= 0) errln("utext_compare() failed, simple setup");
881 UTEXT_SETNATIVEINDEX(&test1, 0);
882 UTEXT_SETNATIVEINDEX(&test4, 0);
883 if (utext_compare(&test1, -1, &test4, -1) <= 0) errln("utext_compare() failed, simple setup");
884
885 // test utext_compareNativeLimit(), simple
886 UTEXT_SETNATIVEINDEX(&test1, 0);
887 UTEXT_SETNATIVEINDEX(&test2, 0);
888 if (utext_compareNativeLimit(&test1, -1, &test2, -1) != 0) errln("utext_compareNativeLimit() failed, simple setup");
889 UTEXT_SETNATIVEINDEX(&test1, 0);
890 UTEXT_SETNATIVEINDEX(&test3, 0);
891 if (utext_compareNativeLimit(&test1, -1, &test3, -1) >= 0) errln("utext_compareNativeLimit() failed, simple setup");
892 UTEXT_SETNATIVEINDEX(&test1, 0);
893 UTEXT_SETNATIVEINDEX(&test4, 0);
894 if (utext_compareNativeLimit(&test1, -1, &test4, -1) <= 0) errln("utext_compareNativeLimit() failed, simple setup");
895
896 // test utext_compare(), one explicit length
897 UTEXT_SETNATIVEINDEX(&test1, 0);
898 UTEXT_SETNATIVEINDEX(&test2, 0);
899 if (utext_compare(&test1, 14, &test2, -1) != 0) errln("utext_compare() failed, one explicit length");
900 UTEXT_SETNATIVEINDEX(&test2, 0);
901 UTEXT_SETNATIVEINDEX(&test3, 0);
902 if (utext_compare(&test3, 14, &test2, -1) != 0) errln("utext_compare() failed, one explicit length");
903 UTEXT_SETNATIVEINDEX(&test2, 0);
904 UTEXT_SETNATIVEINDEX(&test4, 12);
905 if (utext_compare(&test4, 14, &test2, -1) != 0) errln("utext_compare() failed, one explicit length and offset");
906 UTEXT_SETNATIVEINDEX(&test1, 0);
907 UTEXT_SETNATIVEINDEX(&test3, 0);
908 if (utext_compare(&test3, 18, &test2, -1) <= 0) errln("utext_compare() failed, one explicit length");
909
910 // test utext_compareNativeLimit(), one explicit length
911 UTEXT_SETNATIVEINDEX(&test1, 0);
912 UTEXT_SETNATIVEINDEX(&test2, 0);
913 if (utext_compareNativeLimit(&test1, 14, &test2, -1) != 0) errln("utext_compareNativeLimit() failed, one explicit length");
914 UTEXT_SETNATIVEINDEX(&test2, 0);
915 UTEXT_SETNATIVEINDEX(&test3, 0);
916 if (utext_compareNativeLimit(&test3, 14, &test2, -1) != 0) errln("utext_compareNativeLimit() failed, one explicit length");
917 UTEXT_SETNATIVEINDEX(&test2, 0);
918 UTEXT_SETNATIVEINDEX(&test4, 12);
919 if (utext_compareNativeLimit(&test4, 26, &test2, -1) != 0) errln("utext_compareNativeLimit() failed, one explicit length and limit");
920 UTEXT_SETNATIVEINDEX(&test1, 0);
921 UTEXT_SETNATIVEINDEX(&test3, 0);
922 if (utext_compareNativeLimit(&test3, 18, &test2, -1) <= 0) errln("utext_compareNativeLimit() failed, one explicit length");
923
924 // test utext_compare(), UChar-based UText
925 UTEXT_SETNATIVEINDEX(&uniCharText, 0);
926 UTEXT_SETNATIVEINDEX(&test2, 0);
927 if (utext_compare(&test2, -1, &uniCharText, -1) != 0) errln("utext_compare() failed, UChar-based UText");
928 UTEXT_SETNATIVEINDEX(&uniCharText, 0);
929 UTEXT_SETNATIVEINDEX(&test3, 0);
930 if (utext_compare(&test3, -1, &uniCharText, -1) <= 0) errln("utext_compare() failed, UChar-based UText");
931 UTEXT_SETNATIVEINDEX(&uniCharText, 0);
932 UTEXT_SETNATIVEINDEX(&test4, 0);
933 if (utext_compare(&test4, -1, &uniCharText, -1) >= 0) errln("utext_compare() failed, UChar-based UText");
934
935 // test utext_compareNativeLimit(), UChar-based UText
936 UTEXT_SETNATIVEINDEX(&uniCharText, 0);
937 UTEXT_SETNATIVEINDEX(&test2, 0);
938 if (utext_compareNativeLimit(&test2, -1, &uniCharText, -1) != 0) errln("utext_compareNativeLimit() failed, UChar-based UText");
939 UTEXT_SETNATIVEINDEX(&uniCharText, 0);
940 UTEXT_SETNATIVEINDEX(&test3, 0);
941 if (utext_compareNativeLimit(&test3, -1, &uniCharText, -1) <= 0) errln("utext_compareNativeLimit() failed, UChar-based UText");
942 UTEXT_SETNATIVEINDEX(&uniCharText, 0);
943 UTEXT_SETNATIVEINDEX(&test4, 0);
944 if (utext_compareNativeLimit(&test4, -1, &uniCharText, -1) >= 0) errln("utext_compareNativeLimit() failed, UChar-based UText");
945
946 // test utext_compare(), UTF8-based UText
947 UTEXT_SETNATIVEINDEX(&charText, 0);
948 UTEXT_SETNATIVEINDEX(&test2, 0);
949 if (utext_compare(&test2, -1, &charText, -1) != 0) errln("utext_compare() failed, UTF8-based UText");
950 UTEXT_SETNATIVEINDEX(&charText, 0);
951 UTEXT_SETNATIVEINDEX(&test3, 0);
952 if (utext_compare(&test3, -1, &charText, -1) <= 0) errln("utext_compare() failed, UTF8-based UText");
953 UTEXT_SETNATIVEINDEX(&charText, 0);
954 UTEXT_SETNATIVEINDEX(&test4, 0);
955 if (utext_compare(&test4, -1, &charText, -1) >= 0) errln("utext_compare() failed, UTF8-based UText");
956
957 // test utext_compareNativeLimit(), UTF8-based UText
958 UTEXT_SETNATIVEINDEX(&charText, 0);
959 UTEXT_SETNATIVEINDEX(&test2, 0);
960 if (utext_compareNativeLimit(&test2, -1, &charText, -1) != 0) errln("utext_compareNativeLimit() failed, UTF8-based UText");
961 UTEXT_SETNATIVEINDEX(&charText, 0);
962 UTEXT_SETNATIVEINDEX(&test3, 0);
963 if (utext_compareNativeLimit(&test3, -1, &charText, -1) <= 0) errln("utext_compareNativeLimit() failed, UTF8-based UText");
964 UTEXT_SETNATIVEINDEX(&charText, 0);
965 UTEXT_SETNATIVEINDEX(&test4, 0);
966 if (utext_compareNativeLimit(&test4, -1, &charText, -1) >= 0) errln("utext_compareNativeLimit() failed, UTF8-based UText");
967
968 // test utext_compare(), length
969 UTEXT_SETNATIVEINDEX(&test1, 0);
970 UTEXT_SETNATIVEINDEX(&test2, 0);
971 if (utext_compare(&test1, -1, &test2, 4) != 0) errln("utext_compare() failed, one length");
972 UTEXT_SETNATIVEINDEX(&test1, 0);
973 UTEXT_SETNATIVEINDEX(&test2, 0);
974 if (utext_compare(&test1, 5, &test2, 4) <= 0) errln("utext_compare() failed, both lengths");
975
976 // test utext_compareNativeLimit(), limit
977 UTEXT_SETNATIVEINDEX(&test1, 0);
978 UTEXT_SETNATIVEINDEX(&test2, 0);
979 if (utext_compareNativeLimit(&test1, -1, &test2, 4) != 0) errln("utext_compareNativeLimit() failed, one limit");
980 UTEXT_SETNATIVEINDEX(&test1, 0);
981 UTEXT_SETNATIVEINDEX(&test2, 0);
982 if (utext_compareNativeLimit(&test1, 5, &test2, 4) <= 0) errln("utext_compareNativeLimit() failed, both limits");
983
984 // test utext_compare(), both explicit offsets and lengths
985 UTEXT_SETNATIVEINDEX(&test1, 0);
986 UTEXT_SETNATIVEINDEX(&test2, 0);
987 if (utext_compare(&test1, 14, &test2, 14) != 0) errln("utext_compare() failed, both explicit offsets and lengths");
988 UTEXT_SETNATIVEINDEX(&test1, 0);
989 UTEXT_SETNATIVEINDEX(&test3, 0);
990 if (utext_compare(&test1, 14, &test3, 14) != 0) errln("utext_compare() failed, both explicit offsets and lengths");
991 UTEXT_SETNATIVEINDEX(&test1, 0);
992 UTEXT_SETNATIVEINDEX(&test4, 12);
993 if (utext_compare(&test1, 14, &test4, 14) != 0) errln("utext_compare() failed, both explicit offsets and lengths");
994 UTEXT_SETNATIVEINDEX(&test1, 10);
995 UTEXT_SETNATIVEINDEX(&test2, 0);
996 if (utext_compare(&test1, 4, &test2, 4) >= 0) errln("utext_compare() failed, both explicit offsets and lengths");
997 UTEXT_SETNATIVEINDEX(&test1, 10);
998 UTEXT_SETNATIVEINDEX(&test3, 22);
999 if (utext_compare(&test1, 4, &test3, 9) <= 0) errln("utext_compare() failed, both explicit offsets and lengths");
1000 UTEXT_SETNATIVEINDEX(&test1, 10);
1001 UTEXT_SETNATIVEINDEX(&test4, 22);
1002 if (utext_compare(&test1, 4, &test4, 4) != 0) errln("utext_compare() failed, both explicit offsets and lengths");
1003
1004 // test utext_compareNativeLimit(), both explicit offsets and limits
1005 UTEXT_SETNATIVEINDEX(&test1, 0);
1006 UTEXT_SETNATIVEINDEX(&test2, 0);
1007 if (utext_compareNativeLimit(&test1, 14, &test2, 14) != 0) errln("utext_compareNativeLimit() failed, both explicit offsets and limits");
1008 UTEXT_SETNATIVEINDEX(&test1, 0);
1009 UTEXT_SETNATIVEINDEX(&test3, 0);
1010 if (utext_compareNativeLimit(&test1, 14, &test3, 14) != 0) errln("utext_compareNativeLimit() failed, both explicit offsets and limits");
1011 UTEXT_SETNATIVEINDEX(&test1, 0);
1012 UTEXT_SETNATIVEINDEX(&test4, 12);
1013 if (utext_compareNativeLimit(&test1, 14, &test4, 26) != 0) errln("utext_compareNativeLimit() failed, both explicit offsets and limits");
1014 UTEXT_SETNATIVEINDEX(&test1, 10);
1015 UTEXT_SETNATIVEINDEX(&test2, 0);
1016 if (utext_compareNativeLimit(&test1, 14, &test2, 4) >= 0) errln("utext_compareNativeLimit() failed, both explicit offsets and limits");
1017 UTEXT_SETNATIVEINDEX(&test1, 10);
1018 UTEXT_SETNATIVEINDEX(&test3, 22);
1019 if (utext_compareNativeLimit(&test1, 14, &test3, 31) <= 0) errln("utext_compareNativeLimit() failed, both explicit offsets and limits");
1020 UTEXT_SETNATIVEINDEX(&test1, 10);
1021 UTEXT_SETNATIVEINDEX(&test4, 22);
1022 if (utext_compareNativeLimit(&test1, 14, &test4, 26) != 0) errln("utext_compareNativeLimit() failed, both explicit offsets and limits");
1023
1024 /* test caseCompare() */
1025 {
1026 static const UChar
1027 _mixed[]= { 0x61, 0x42, 0x131, 0x3a3, 0xdf, 0x130, 0x49, 0xfb03, 0xd93f, 0xdfff, 0 },
1028 _otherDefault[]= { 0x41, 0x62, 0x131, 0x3c3, 0x73, 0x53, 0x69, 0x307, 0x69, 0x46, 0x66, 0x49, 0xd93f, 0xdfff, 0 },
1029 _otherExcludeSpecialI[]={ 0x41, 0x62, 0x131, 0x3c3, 0x53, 0x73, 0x69, 0x131, 0x66, 0x46, 0x69, 0xd93f, 0xdfff, 0 },
1030 _different[]= { 0x41, 0x62, 0x131, 0x3c3, 0x73, 0x53, 0x130, 0x49, 0x46, 0x66, 0x49, 0xd93f, 0xdffd, 0 };
1031
1032 UText
1033 mixed = UTEXT_INITIALIZER,
1034 otherDefault = UTEXT_INITIALIZER,
1035 otherExcludeSpecialI = UTEXT_INITIALIZER,
1036 different = UTEXT_INITIALIZER;
1037
1038 utext_openUChars(&mixed, _mixed, -1, &status);
1039 utext_openUChars(&otherDefault, _otherDefault, -1, &status);
1040 utext_openUChars(&otherExcludeSpecialI, _otherExcludeSpecialI, -1, &status);
1041 utext_openUChars(&different, _different, -1, &status);
1042
1043 TEST_SUCCESS(status);
1044
1045 int32_t result;
1046
1047 /* test default options */
1048 UTEXT_SETNATIVEINDEX(&mixed, 0);
1049 UTEXT_SETNATIVEINDEX(&otherDefault, 0);
1050 result = utext_caseCompare(&mixed, -1, &otherDefault, -1, U_FOLD_CASE_DEFAULT, &status);
1051 if (0 != result || U_FAILURE(status)) {
1052 errln("error: utext_caseCompare (other, default) gives %ld (should be 0) (%s)\n", result, u_errorName(status));
1053 }
1054 UTEXT_SETNATIVEINDEX(&mixed, 0);
1055 UTEXT_SETNATIVEINDEX(&otherDefault, 0);
1056 result = utext_caseCompareNativeLimit(&mixed, -1, &otherDefault, -1, U_FOLD_CASE_DEFAULT, &status);
1057 if (0 != result || U_FAILURE(status)) {
1058 errln("error: utext_caseCompareNativeLimit (other, default) gives %ld (should be 0) (%s)\n", result, u_errorName(status));
1059 }
1060
1061 /* test excluding special I */
1062 UTEXT_SETNATIVEINDEX(&mixed, 0);
1063 UTEXT_SETNATIVEINDEX(&otherExcludeSpecialI, 0);
1064 result = utext_caseCompare(&mixed, -1, &otherExcludeSpecialI, -1, U_FOLD_CASE_EXCLUDE_SPECIAL_I, &status);
1065 if (0 != result || U_FAILURE(status)) {
1066 errln("error: utext_caseCompare (otherExcludeSpecialI, U_FOLD_CASE_EXCLUDE_SPECIAL_I) gives %ld (should be 0) (%s)\n", result, u_errorName(status));
1067 }
1068 UTEXT_SETNATIVEINDEX(&mixed, 0);
1069 UTEXT_SETNATIVEINDEX(&otherExcludeSpecialI, 0);
1070 result = utext_caseCompareNativeLimit(&mixed, -1, &otherExcludeSpecialI, -1, U_FOLD_CASE_EXCLUDE_SPECIAL_I, &status);
1071 if (0 != result || U_FAILURE(status)) {
1072 errln("error: utext_caseCompareNativeLimit (otherExcludeSpecialI, U_FOLD_CASE_EXCLUDE_SPECIAL_I) gives %ld (should be 0) (%s)\n", result, u_errorName(status));
1073 }
1074 UTEXT_SETNATIVEINDEX(&mixed, 0);
1075 UTEXT_SETNATIVEINDEX(&otherDefault, 0);
1076 result = utext_caseCompare(&mixed, -1, &otherDefault, -1, U_FOLD_CASE_EXCLUDE_SPECIAL_I, &status);
1077 if (0 == result || U_FAILURE(status)) {
1078 errln("error: utext_caseCompare (other, U_FOLD_CASE_EXCLUDE_SPECIAL_I) gives %ld (should be nonzero) (%s)\n", result, u_errorName(status));
1079 }
1080 UTEXT_SETNATIVEINDEX(&mixed, 0);
1081 UTEXT_SETNATIVEINDEX(&otherDefault, 0);
1082 result = utext_caseCompareNativeLimit(&mixed, -1, &otherDefault, -1, U_FOLD_CASE_EXCLUDE_SPECIAL_I, &status);
1083 if (0 == result || U_FAILURE(status)) {
1084 errln("error: utext_caseCompareNativeLimit (other, U_FOLD_CASE_EXCLUDE_SPECIAL_I) gives %ld (should be nonzero) (%s)\n", result, u_errorName(status));
1085 }
1086
1087 /* test against different string */
1088 UTEXT_SETNATIVEINDEX(&mixed, 0);
1089 UTEXT_SETNATIVEINDEX(&different, 0);
1090 result = utext_caseCompare(&mixed, -1, &different, -1, U_FOLD_CASE_DEFAULT, &status);
1091 if (0 >= result || U_FAILURE(status)) {
1092 errln("error: utext_caseCompare (different, default) gives %ld (should be positive) (%s)\n", result, u_errorName(status));
1093 }
1094 UTEXT_SETNATIVEINDEX(&mixed, 0);
1095 UTEXT_SETNATIVEINDEX(&different, 0);
1096 result = utext_caseCompareNativeLimit(&mixed, -1, &different, -1, U_FOLD_CASE_DEFAULT, &status);
1097 if (0 >= result || U_FAILURE(status)) {
1098 errln("error: utext_caseCompareNativeLimit (different, default) gives %ld (should be positive) (%s)\n", result, u_errorName(status));
1099 }
1100
1101 /* test caseCompare() - include the folded sharp s (U+00df) with different lengths */
1102 UTEXT_SETNATIVEINDEX(&mixed, 1);
1103 UTEXT_SETNATIVEINDEX(&different, 1);
1104 result = utext_caseCompare(&mixed, 4, &different, 5, U_FOLD_CASE_DEFAULT, &status);
1105 if (0 != result || U_FAILURE(status)) {
1106 errln("error: utext_caseCompare (mixed[1-5), different[1-6), default) gives %ld (should be 0) (%s)\n", result, u_errorName(status));
1107 }
1108 UTEXT_SETNATIVEINDEX(&mixed, 1);
1109 UTEXT_SETNATIVEINDEX(&different, 1);
1110 result = utext_caseCompareNativeLimit(&mixed, 5, &different, 6, U_FOLD_CASE_DEFAULT, &status);
1111 if (0 != result || U_FAILURE(status)) {
1112 errln("error: utext_caseCompareNativeLimit (mixed[1-5), different[1-6), default) gives %ld (should be 0) (%s)\n", result, u_errorName(status));
1113 }
1114
1115 /* test caseCompare() - stop in the middle of the sharp s (U+00df) */
1116 UTEXT_SETNATIVEINDEX(&mixed, 1);
1117 UTEXT_SETNATIVEINDEX(&different, 1);
1118 result = utext_caseCompare(&mixed, 4, &different, 4, U_FOLD_CASE_DEFAULT, &status);
1119 if (0 >= result || U_FAILURE(status)) {
1120 errln("error: utext_caseCompare (mixed[1-5), different[1-5), default) gives %ld (should be positive) (%s)\n", result, u_errorName(status));
1121 }
1122 UTEXT_SETNATIVEINDEX(&mixed, 1);
1123 UTEXT_SETNATIVEINDEX(&different, 1);
1124 result = utext_caseCompareNativeLimit(&mixed, 5, &different, 5, U_FOLD_CASE_DEFAULT, &status);
1125 if (0 >= result || U_FAILURE(status)) {
1126 errln("error: utext_caseCompareNativeLimit (mixed[1-5), different[1-5), default) gives %ld (should be positive) (%s)\n", result, u_errorName(status));
1127 }
1128 }
1129
1130 /* test surrogates in comparison */
1131 {
1132 static const UChar
1133 _before[] = { 0x65, 0xd800, 0xd800, 0xdc01, 0x65, 0x00 },
1134 _after[] = { 0x65, 0xd800, 0xdc00, 0x65, 0x00 };
1135
1136 UText
1137 before = UTEXT_INITIALIZER,
1138 after = UTEXT_INITIALIZER;
1139
1140 utext_openUChars(&before, _before, -1, &status);
1141 utext_openUChars(&after, _after, -1, &status);
1142
1143 TEST_SUCCESS(status);
1144 int32_t result;
1145
1146 UTEXT_SETNATIVEINDEX(&before, 1);
1147 UTEXT_SETNATIVEINDEX(&after, 1);
1148 result = utext_compare(&before, -1, &after, -1);
1149 if (0 <= result || U_FAILURE(status)) {
1150 errln("error: utext_compare ({ 65, d800, 10001, 65 }, { 65, 10000, 65 }) gives %ld (should be negative) (%s)\n", result, u_errorName(status));
1151 }
1152
1153 UTEXT_SETNATIVEINDEX(&before, 1);
1154 UTEXT_SETNATIVEINDEX(&after, 1);
1155 result = utext_compare(&before, 3, &after, 3);
1156 if (0 <= result || U_FAILURE(status)) {
1157 errln("error: utext_compare with lengths ({ 65, d800, 10001, 65 }, { 65, 10000, 65 }) gives %ld (should be negative) (%s)\n", result, u_errorName(status));
1158 }
1159
1160 UTEXT_SETNATIVEINDEX(&before, 1);
1161 UTEXT_SETNATIVEINDEX(&after, 1);
1162 result = utext_caseCompare(&before, -1, &after, -1, U_FOLD_CASE_DEFAULT, &status);
1163 if (0 <= result || U_FAILURE(status)) {
1164 errln("error: utext_caseCompare ({ 65, d800, 10001, 65 }, { 65, 10000, 65 }) gives %ld (should be negative) (%s)\n", result, u_errorName(status));
1165 }
1166
1167 UTEXT_SETNATIVEINDEX(&before, 1);
1168 UTEXT_SETNATIVEINDEX(&after, 1);
1169 result = utext_caseCompare(&before, 3, &after, 3, U_FOLD_CASE_DEFAULT, &status);
1170 if (0 <= result || U_FAILURE(status)) {
1171 errln("error: utext_caseCompare with lengths ({ 65, d800, 10001, 65 }, { 65, 10000, 65 }) gives %ld (should be negative) (%s)\n", result, u_errorName(status));
1172 }
1173
1174 utext_close(&before);
1175 utext_close(&after);
1176 }
1177
1178 /* test surrogates at end of string */
1179 {
1180 static const UChar
1181 _before[] = { 0x65, 0xd800, 0xd800, 0xdc01, 0x00 },
1182 _after[] = { 0x65, 0xd800, 0xdc00, 0x00 };
1183
1184 UText
1185 before = UTEXT_INITIALIZER,
1186 after = UTEXT_INITIALIZER;
1187
1188 utext_openUChars(&before, _before, -1, &status);
1189 utext_openUChars(&after, _after, -1, &status);
1190
1191 TEST_SUCCESS(status);
1192 int32_t result;
1193
1194 UTEXT_SETNATIVEINDEX(&before, 1);
1195 UTEXT_SETNATIVEINDEX(&after, 1);
1196 result = utext_compare(&before, -1, &after, -1);
1197 if (0 <= result || U_FAILURE(status)) {
1198 errln("error: utext_compare ({ 65, d800, 10001 }, { 65, 10000 }) gives %ld (should be negative) (%s)\n", result, u_errorName(status));
1199 }
1200
1201 UTEXT_SETNATIVEINDEX(&before, 1);
1202 UTEXT_SETNATIVEINDEX(&after, 1);
1203 result = utext_caseCompare(&before, -1, &after, -1, U_FOLD_CASE_DEFAULT, &status);
1204 if (0 <= result || U_FAILURE(status)) {
1205 errln("error: utext_caseCompare ({ 65, d800, 10001 }, { 65, 10000 }) gives %ld (should be negative) (%s)\n", result, u_errorName(status));
1206 }
1207
1208 utext_close(&before);
1209 utext_close(&after);
1210 }
1211
1212 /* test empty strings */
1213 {
1214 UChar zero16 = 0;
1215 char zero8 = 0;
1216 UText emptyUChar = UTEXT_INITIALIZER;
1217 UText emptyUTF8 = UTEXT_INITIALIZER;
1218 UText nullUChar = UTEXT_INITIALIZER;
1219 UText nullUTF8 = UTEXT_INITIALIZER;
1220
1221 utext_openUChars(&emptyUChar, &zero16, -1, &status);
1222 utext_openUTF8(&emptyUTF8, &zero8, -1, &status);
1223 utext_openUChars(&nullUChar, NULL, 0, &status);
1224 utext_openUTF8(&nullUTF8, NULL, 0, &status);
1225
1226 if (utext_compare(&emptyUChar, -1, &emptyUTF8, -1) != 0) {
1227 errln("error: utext_compare(&emptyUChar, -1, &emptyUTF8, -1) != 0");
1228 }
1229 if (utext_compare(&emptyUChar, -1, &nullUChar, -1) != 0) {
1230 errln("error: utext_compare(&emptyUChar, -1, &nullUChar, -1) != 0");
1231 }
1232 if (utext_compare(&emptyUChar, -1, &nullUTF8, -1) != 0) {
1233 errln("error: utext_compare(&emptyUChar, -1, &nullUTF8, -1) != 0");
1234 }
1235 if (utext_compare(&emptyUTF8, -1, &nullUChar, -1) != 0) {
1236 errln("error: utext_compare(&emptyUTF8, -1, &nullUChar, -1) != 0");
1237 }
1238 if (utext_compare(&emptyUTF8, -1, &nullUTF8, -1) != 0) {
1239 errln("error: utext_compare(&emptyUTF8, -1, &nullUTF8, -1) != 0");
1240 }
1241 if (utext_compare(&nullUChar, -1, &nullUTF8, -1) != 0) {
1242 errln("error: utext_compare(&nullUChar, -1, &nullUTF8, -1) != 0");
1243 }
1244
1245 if (utext_compareNativeLimit(&emptyUChar, -1, &emptyUTF8, -1) != 0) {
1246 errln("error: utext_compareNativeLimit(&emptyUChar, -1, &emptyUTF8, -1) != 0");
1247 }
1248 if (utext_compareNativeLimit(&emptyUChar, -1, &nullUChar, -1) != 0) {
1249 errln("error: utext_compareNativeLimit(&emptyUChar, -1, &nullUChar, -1) != 0");
1250 }
1251 if (utext_compareNativeLimit(&emptyUChar, -1, &nullUTF8, -1) != 0) {
1252 errln("error: utext_compareNativeLimit(&emptyUChar, -1, &nullUTF8, -1) != 0");
1253 }
1254 if (utext_compareNativeLimit(&emptyUTF8, -1, &nullUChar, -1) != 0) {
1255 errln("error: utext_compareNativeLimit(&emptyUTF8, -1, &nullUChar, -1) != 0");
1256 }
1257 if (utext_compareNativeLimit(&emptyUTF8, -1, &nullUTF8, -1) != 0) {
1258 errln("error: utext_compareNativeLimit(&emptyUTF8, -1, &nullUTF8, -1) != 0");
1259 }
1260 if (utext_compareNativeLimit(&nullUChar, -1, &nullUTF8, -1) != 0) {
1261 errln("error: utext_compareNativeLimit(&nullUChar, -1, &nullUTF8, -1) != 0");
1262 }
1263
1264 if (utext_caseCompare(&emptyUChar, -1, &emptyUTF8, -1, 0, &status) != 0) {
1265 errln("error: utext_caseCompare(&emptyUChar, -1, &emptyUTF8, -1, 0, &status) != 0");
1266 }
1267 if (utext_caseCompare(&emptyUChar, -1, &nullUChar, -1, 0, &status) != 0) {
1268 errln("error: utext_caseCompare(&emptyUChar, -1, &nullUChar, -1, 0, &status) != 0");
1269 }
1270 if (utext_caseCompare(&emptyUChar, -1, &nullUTF8, -1, 0, &status) != 0) {
1271 errln("error: utext_caseCompare(&emptyUChar, -1, &nullUTF8, -1, 0, &status) != 0");
1272 }
1273 if (utext_caseCompare(&emptyUTF8, -1, &nullUChar, -1, 0, &status) != 0) {
1274 errln("error: utext_caseCompare(&emptyUTF8, -1, &nullUChar, -1, 0, &status) != 0");
1275 }
1276 if (utext_caseCompare(&emptyUTF8, -1, &nullUTF8, -1, 0, &status) != 0) {
1277 errln("error: utext_caseCompare(&emptyUTF8, -1, &nullUTF8, -1, 0, &status) != 0");
1278 }
1279 if (utext_caseCompare(&nullUChar, -1, &nullUTF8, -1, 0, &status) != 0) {
1280 errln("error: utext_caseCompare(&nullUChar, -1, &nullUTF8, -1, 0, &status) != 0");
1281 }
1282
1283 if (utext_caseCompareNativeLimit(&emptyUChar, -1, &emptyUTF8, -1, 0, &status) != 0) {
1284 errln("error: utext_caseCompareNativeLimit(&emptyUChar, -1, &emptyUTF8, -1, 0, &status) != 0");
1285 }
1286 if (utext_caseCompareNativeLimit(&emptyUChar, -1, &nullUChar, -1, 0, &status) != 0) {
1287 errln("error: utext_caseCompareNativeLimit(&emptyUChar, -1, &nullUChar, -1, 0, &status) != 0");
1288 }
1289 if (utext_caseCompareNativeLimit(&emptyUChar, -1, &nullUTF8, -1, 0, &status) != 0) {
1290 errln("error: utext_caseCompareNativeLimit(&emptyUChar, -1, &nullUTF8, -1, 0, &status) != 0");
1291 }
1292 if (utext_caseCompareNativeLimit(&emptyUTF8, -1, &nullUChar, -1, 0, &status) != 0) {
1293 errln("error: utext_caseCompareNativeLimit(&emptyUTF8, -1, &nullUChar, -1, 0, &status) != 0");
1294 }
1295 if (utext_caseCompareNativeLimit(&emptyUTF8, -1, &nullUTF8, -1, 0, &status) != 0) {
1296 errln("error: utext_caseCompareNativeLimit(&emptyUTF8, -1, &nullUTF8, -1, 0, &status) != 0");
1297 }
1298 if (utext_caseCompareNativeLimit(&nullUChar, -1, &nullUTF8, -1, 0, &status) != 0) {
1299 errln("error: utext_caseCompareNativeLimit(&nullUChar, -1, &nullUTF8, -1, 0, &status) != 0");
1300 }
1301
1302 utext_close(&emptyUChar);
1303 utext_close(&emptyUTF8);
1304 utext_close(&nullUChar);
1305 utext_close(&nullUTF8);
1306 utext_close(&charText);
1307 utext_close(&uniCharText);
1308 }
1309 }
1310
1311
1312
1313 //
1314 // ErrorTest() Check various error and edge cases.
1315 //
ErrorTest()1316 void UTextTest::ErrorTest()
1317 {
1318 // Close of an unitialized UText. Shouldn't blow up.
1319 {
1320 UText ut;
1321 memset(&ut, 0, sizeof(UText));
1322 utext_close(&ut);
1323 utext_close(NULL);
1324 }
1325
1326 // Double-close of a UText. Shouldn't blow up. UText should still be usable.
1327 {
1328 UErrorCode status = U_ZERO_ERROR;
1329 UText ut = UTEXT_INITIALIZER;
1330 UnicodeString s("Hello, World");
1331 UText *ut2 = utext_openUnicodeString(&ut, &s, &status);
1332 TEST_SUCCESS(status);
1333 TEST_ASSERT(ut2 == &ut);
1334
1335 UText *ut3 = utext_close(&ut);
1336 TEST_ASSERT(ut3 == &ut);
1337
1338 UText *ut4 = utext_close(&ut);
1339 TEST_ASSERT(ut4 == &ut);
1340
1341 utext_openUnicodeString(&ut, &s, &status);
1342 TEST_SUCCESS(status);
1343 utext_close(&ut);
1344 }
1345
1346 // Re-use of a UText, chaining through each of the types of UText
1347 // (If it doesn't blow up, and doesn't leak, it's probably working fine)
1348 {
1349 UErrorCode status = U_ZERO_ERROR;
1350 UText ut = UTEXT_INITIALIZER;
1351 UText *utp;
1352 UnicodeString s1("Hello, World");
1353 UChar s2[] = {(UChar)0x41, (UChar)0x42, (UChar)0};
1354 const char *s3 = "\x66\x67\x68";
1355
1356 utp = utext_openUnicodeString(&ut, &s1, &status);
1357 TEST_SUCCESS(status);
1358 TEST_ASSERT(utp == &ut);
1359
1360 utp = utext_openConstUnicodeString(&ut, &s1, &status);
1361 TEST_SUCCESS(status);
1362 TEST_ASSERT(utp == &ut);
1363
1364 utp = utext_openUTF8(&ut, s3, -1, &status);
1365 TEST_SUCCESS(status);
1366 TEST_ASSERT(utp == &ut);
1367
1368 utp = utext_openUChars(&ut, s2, -1, &status);
1369 TEST_SUCCESS(status);
1370 TEST_ASSERT(utp == &ut);
1371
1372 utp = utext_close(&ut);
1373 TEST_ASSERT(utp == &ut);
1374
1375 utp = utext_openUnicodeString(&ut, &s1, &status);
1376 TEST_SUCCESS(status);
1377 TEST_ASSERT(utp == &ut);
1378 }
1379
1380 // Invalid parameters on open
1381 //
1382 {
1383 UErrorCode status = U_ZERO_ERROR;
1384 UText ut = UTEXT_INITIALIZER;
1385
1386 utext_openUChars(&ut, NULL, 5, &status);
1387 TEST_ASSERT(status == U_ILLEGAL_ARGUMENT_ERROR);
1388
1389 status = U_ZERO_ERROR;
1390 utext_openUChars(&ut, NULL, -1, &status);
1391 TEST_ASSERT(status == U_ILLEGAL_ARGUMENT_ERROR);
1392
1393 status = U_ZERO_ERROR;
1394 utext_openUTF8(&ut, NULL, 4, &status);
1395 TEST_ASSERT(status == U_ILLEGAL_ARGUMENT_ERROR);
1396
1397 status = U_ZERO_ERROR;
1398 utext_openUTF8(&ut, NULL, -1, &status);
1399 TEST_ASSERT(status == U_ILLEGAL_ARGUMENT_ERROR);
1400 }
1401
1402 //
1403 // UTF-8 with malformed sequences.
1404 // These should come through as the Unicode replacement char, \ufffd
1405 //
1406 {
1407 UErrorCode status = U_ZERO_ERROR;
1408 UText *ut = NULL;
1409 const char *badUTF8 = "\x41\x81\x42\xf0\x81\x81\x43";
1410 UChar32 c;
1411
1412 ut = utext_openUTF8(NULL, badUTF8, -1, &status);
1413 TEST_SUCCESS(status);
1414 c = utext_char32At(ut, 1);
1415 TEST_ASSERT(c == 0xfffd);
1416 c = utext_char32At(ut, 3);
1417 TEST_ASSERT(c == 0xfffd);
1418 c = utext_char32At(ut, 5);
1419 TEST_ASSERT(c == 0xfffd);
1420 c = utext_char32At(ut, 6);
1421 TEST_ASSERT(c == 0x43);
1422
1423 UChar buf[10];
1424 int n = utext_extract(ut, 0, 9, buf, 10, &status);
1425 TEST_SUCCESS(status);
1426 TEST_ASSERT(n==5);
1427 TEST_ASSERT(buf[1] == 0xfffd);
1428 TEST_ASSERT(buf[3] == 0xfffd);
1429 TEST_ASSERT(buf[2] == 0x42);
1430 utext_close(ut);
1431 }
1432
1433
1434 //
1435 // isLengthExpensive - does it make the exptected transitions after
1436 // getting the length of a nul terminated string?
1437 //
1438 {
1439 UErrorCode status = U_ZERO_ERROR;
1440 UnicodeString sa("Hello, this is a string");
1441 UBool isExpensive;
1442
1443 UChar sb[100];
1444 memset(sb, 0x20, sizeof(sb));
1445 sb[99] = 0;
1446
1447 UText *uta = utext_openUnicodeString(NULL, &sa, &status);
1448 TEST_SUCCESS(status);
1449 isExpensive = utext_isLengthExpensive(uta);
1450 TEST_ASSERT(isExpensive == FALSE);
1451 utext_close(uta);
1452
1453 UText *utb = utext_openUChars(NULL, sb, -1, &status);
1454 TEST_SUCCESS(status);
1455 isExpensive = utext_isLengthExpensive(utb);
1456 TEST_ASSERT(isExpensive == TRUE);
1457 int64_t len = utext_nativeLength(utb);
1458 TEST_ASSERT(len == 99);
1459 isExpensive = utext_isLengthExpensive(utb);
1460 TEST_ASSERT(isExpensive == FALSE);
1461 utext_close(utb);
1462 }
1463
1464 //
1465 // Index to positions not on code point boundaries.
1466 //
1467 {
1468 const char *u8str = "\xc8\x81\xe1\x82\x83\xf1\x84\x85\x86";
1469 int32_t startMap[] = { 0, 0, 2, 2, 2, 5, 5, 5, 5, 9, 9};
1470 int32_t nextMap[] = { 2, 2, 5, 5, 5, 9, 9, 9, 9, 9, 9};
1471 int32_t prevMap[] = { 0, 0, 0, 0, 0, 2, 2, 2, 2, 5, 5};
1472 UChar32 c32Map[] = {0x201, 0x201, 0x1083, 0x1083, 0x1083, 0x044146, 0x044146, 0x044146, 0x044146, -1, -1};
1473 UChar32 pr32Map[] = { -1, -1, 0x201, 0x201, 0x201, 0x1083, 0x1083, 0x1083, 0x1083, 0x044146, 0x044146};
1474
1475 // extractLen is the size, in UChars, of what will be extracted between index and index+1.
1476 // is zero when both index positions lie within the same code point.
1477 int32_t exLen[] = { 0, 1, 0, 0, 1, 0, 0, 0, 2, 0, 0};
1478
1479
1480 UErrorCode status = U_ZERO_ERROR;
1481 UText *ut = utext_openUTF8(NULL, u8str, -1, &status);
1482 TEST_SUCCESS(status);
1483
1484 // Check setIndex
1485 int32_t i;
1486 int32_t startMapLimit = sizeof(startMap) / sizeof(int32_t);
1487 for (i=0; i<startMapLimit; i++) {
1488 utext_setNativeIndex(ut, i);
1489 int64_t cpIndex = utext_getNativeIndex(ut);
1490 TEST_ASSERT(cpIndex == startMap[i]);
1491 cpIndex = UTEXT_GETNATIVEINDEX(ut);
1492 TEST_ASSERT(cpIndex == startMap[i]);
1493 }
1494
1495 // Check char32At
1496 for (i=0; i<startMapLimit; i++) {
1497 UChar32 c32 = utext_char32At(ut, i);
1498 TEST_ASSERT(c32 == c32Map[i]);
1499 int64_t cpIndex = utext_getNativeIndex(ut);
1500 TEST_ASSERT(cpIndex == startMap[i]);
1501 }
1502
1503 // Check utext_next32From
1504 for (i=0; i<startMapLimit; i++) {
1505 UChar32 c32 = utext_next32From(ut, i);
1506 TEST_ASSERT(c32 == c32Map[i]);
1507 int64_t cpIndex = utext_getNativeIndex(ut);
1508 TEST_ASSERT(cpIndex == nextMap[i]);
1509 }
1510
1511 // check utext_previous32From
1512 for (i=0; i<startMapLimit; i++) {
1513 gTestNum++;
1514 UChar32 c32 = utext_previous32From(ut, i);
1515 TEST_ASSERT(c32 == pr32Map[i]);
1516 int64_t cpIndex = utext_getNativeIndex(ut);
1517 TEST_ASSERT(cpIndex == prevMap[i]);
1518 }
1519
1520 // check Extract
1521 // Extract from i to i+1, which may be zero or one code points,
1522 // depending on whether the indices straddle a cp boundary.
1523 for (i=0; i<startMapLimit; i++) {
1524 UChar buf[3];
1525 status = U_ZERO_ERROR;
1526 int32_t extractedLen = utext_extract(ut, i, i+1, buf, 3, &status);
1527 TEST_SUCCESS(status);
1528 TEST_ASSERT(extractedLen == exLen[i]);
1529 if (extractedLen > 0) {
1530 UChar32 c32;
1531 /* extractedLen-extractedLen == 0 is used to get around a compiler warning. */
1532 U16_GET(buf, 0, extractedLen-extractedLen, extractedLen, c32);
1533 TEST_ASSERT(c32 == c32Map[i]);
1534 }
1535 }
1536
1537 utext_close(ut);
1538 }
1539
1540
1541 { // Similar test, with utf16 instead of utf8
1542 // TODO: merge the common parts of these tests.
1543
1544 UnicodeString u16str("\\u1000\\U00011000\\u2000\\U00022000", -1, US_INV);
1545 int32_t startMap[] ={ 0, 1, 1, 3, 4, 4, 6, 6};
1546 int32_t nextMap[] = { 1, 3, 3, 4, 6, 6, 6, 6};
1547 int32_t prevMap[] = { 0, 0, 0, 1, 3, 3, 4, 4};
1548 UChar32 c32Map[] = {0x1000, 0x11000, 0x11000, 0x2000, 0x22000, 0x22000, -1, -1};
1549 UChar32 pr32Map[] = { -1, 0x1000, 0x1000, 0x11000, 0x2000, 0x2000, 0x22000, 0x22000};
1550 int32_t exLen[] = { 1, 0, 2, 1, 0, 2, 0, 0,};
1551
1552 u16str = u16str.unescape();
1553 UErrorCode status = U_ZERO_ERROR;
1554 UText *ut = utext_openUnicodeString(NULL, &u16str, &status);
1555 TEST_SUCCESS(status);
1556
1557 int32_t startMapLimit = sizeof(startMap) / sizeof(int32_t);
1558 int i;
1559 for (i=0; i<startMapLimit; i++) {
1560 utext_setNativeIndex(ut, i);
1561 int64_t cpIndex = utext_getNativeIndex(ut);
1562 TEST_ASSERT(cpIndex == startMap[i]);
1563 }
1564
1565 // Check char32At
1566 for (i=0; i<startMapLimit; i++) {
1567 UChar32 c32 = utext_char32At(ut, i);
1568 TEST_ASSERT(c32 == c32Map[i]);
1569 int64_t cpIndex = utext_getNativeIndex(ut);
1570 TEST_ASSERT(cpIndex == startMap[i]);
1571 }
1572
1573 // Check utext_next32From
1574 for (i=0; i<startMapLimit; i++) {
1575 UChar32 c32 = utext_next32From(ut, i);
1576 TEST_ASSERT(c32 == c32Map[i]);
1577 int64_t cpIndex = utext_getNativeIndex(ut);
1578 TEST_ASSERT(cpIndex == nextMap[i]);
1579 }
1580
1581 // check utext_previous32From
1582 for (i=0; i<startMapLimit; i++) {
1583 UChar32 c32 = utext_previous32From(ut, i);
1584 TEST_ASSERT(c32 == pr32Map[i]);
1585 int64_t cpIndex = utext_getNativeIndex(ut);
1586 TEST_ASSERT(cpIndex == prevMap[i]);
1587 }
1588
1589 // check Extract
1590 // Extract from i to i+1, which may be zero or one code points,
1591 // depending on whether the indices straddle a cp boundary.
1592 for (i=0; i<startMapLimit; i++) {
1593 UChar buf[3];
1594 status = U_ZERO_ERROR;
1595 int32_t extractedLen = utext_extract(ut, i, i+1, buf, 3, &status);
1596 TEST_SUCCESS(status);
1597 TEST_ASSERT(extractedLen == exLen[i]);
1598 if (extractedLen > 0) {
1599 UChar32 c32;
1600 /* extractedLen-extractedLen == 0 is used to get around a compiler warning. */
1601 U16_GET(buf, 0, extractedLen-extractedLen, extractedLen, c32);
1602 TEST_ASSERT(c32 == c32Map[i]);
1603 }
1604 }
1605
1606 utext_close(ut);
1607 }
1608
1609 { // Similar test, with UText over Replaceable
1610 // TODO: merge the common parts of these tests.
1611
1612 UnicodeString u16str("\\u1000\\U00011000\\u2000\\U00022000", -1, US_INV);
1613 int32_t startMap[] ={ 0, 1, 1, 3, 4, 4, 6, 6};
1614 int32_t nextMap[] = { 1, 3, 3, 4, 6, 6, 6, 6};
1615 int32_t prevMap[] = { 0, 0, 0, 1, 3, 3, 4, 4};
1616 UChar32 c32Map[] = {0x1000, 0x11000, 0x11000, 0x2000, 0x22000, 0x22000, -1, -1};
1617 UChar32 pr32Map[] = { -1, 0x1000, 0x1000, 0x11000, 0x2000, 0x2000, 0x22000, 0x22000};
1618 int32_t exLen[] = { 1, 0, 2, 1, 0, 2, 0, 0,};
1619
1620 u16str = u16str.unescape();
1621 UErrorCode status = U_ZERO_ERROR;
1622 UText *ut = utext_openReplaceable(NULL, &u16str, &status);
1623 TEST_SUCCESS(status);
1624
1625 int32_t startMapLimit = sizeof(startMap) / sizeof(int32_t);
1626 int i;
1627 for (i=0; i<startMapLimit; i++) {
1628 utext_setNativeIndex(ut, i);
1629 int64_t cpIndex = utext_getNativeIndex(ut);
1630 TEST_ASSERT(cpIndex == startMap[i]);
1631 }
1632
1633 // Check char32At
1634 for (i=0; i<startMapLimit; i++) {
1635 UChar32 c32 = utext_char32At(ut, i);
1636 TEST_ASSERT(c32 == c32Map[i]);
1637 int64_t cpIndex = utext_getNativeIndex(ut);
1638 TEST_ASSERT(cpIndex == startMap[i]);
1639 }
1640
1641 // Check utext_next32From
1642 for (i=0; i<startMapLimit; i++) {
1643 UChar32 c32 = utext_next32From(ut, i);
1644 TEST_ASSERT(c32 == c32Map[i]);
1645 int64_t cpIndex = utext_getNativeIndex(ut);
1646 TEST_ASSERT(cpIndex == nextMap[i]);
1647 }
1648
1649 // check utext_previous32From
1650 for (i=0; i<startMapLimit; i++) {
1651 UChar32 c32 = utext_previous32From(ut, i);
1652 TEST_ASSERT(c32 == pr32Map[i]);
1653 int64_t cpIndex = utext_getNativeIndex(ut);
1654 TEST_ASSERT(cpIndex == prevMap[i]);
1655 }
1656
1657 // check Extract
1658 // Extract from i to i+1, which may be zero or one code points,
1659 // depending on whether the indices straddle a cp boundary.
1660 for (i=0; i<startMapLimit; i++) {
1661 UChar buf[3];
1662 status = U_ZERO_ERROR;
1663 int32_t extractedLen = utext_extract(ut, i, i+1, buf, 3, &status);
1664 TEST_SUCCESS(status);
1665 TEST_ASSERT(extractedLen == exLen[i]);
1666 if (extractedLen > 0) {
1667 UChar32 c32;
1668 /* extractedLen-extractedLen == 0 is used to get around a compiler warning. */
1669 U16_GET(buf, 0, extractedLen-extractedLen, extractedLen, c32);
1670 TEST_ASSERT(c32 == c32Map[i]);
1671 }
1672 }
1673
1674 utext_close(ut);
1675 }
1676 }
1677
1678
FreezeTest()1679 void UTextTest::FreezeTest() {
1680 // Check isWritable() and freeze() behavior.
1681 //
1682
1683 UnicodeString ustr("Hello, World.");
1684 const char u8str[] = {char(0x31), (char)0x32, (char)0x33, 0};
1685 const UChar u16str[] = {(UChar)0x31, (UChar)0x32, (UChar)0x44, 0};
1686
1687 UErrorCode status = U_ZERO_ERROR;
1688 UText *ut = NULL;
1689 UText *ut2 = NULL;
1690
1691 ut = utext_openUTF8(ut, u8str, -1, &status);
1692 TEST_SUCCESS(status);
1693 UBool writable = utext_isWritable(ut);
1694 TEST_ASSERT(writable == FALSE);
1695 utext_copy(ut, 1, 2, 0, TRUE, &status);
1696 TEST_ASSERT(status == U_NO_WRITE_PERMISSION);
1697
1698 status = U_ZERO_ERROR;
1699 ut = utext_openUChars(ut, u16str, -1, &status);
1700 TEST_SUCCESS(status);
1701 writable = utext_isWritable(ut);
1702 TEST_ASSERT(writable == FALSE);
1703 utext_copy(ut, 1, 2, 0, TRUE, &status);
1704 TEST_ASSERT(status == U_NO_WRITE_PERMISSION);
1705
1706 status = U_ZERO_ERROR;
1707 ut = utext_openUnicodeString(ut, &ustr, &status);
1708 TEST_SUCCESS(status);
1709 writable = utext_isWritable(ut);
1710 TEST_ASSERT(writable == TRUE);
1711 utext_freeze(ut);
1712 writable = utext_isWritable(ut);
1713 TEST_ASSERT(writable == FALSE);
1714 utext_copy(ut, 1, 2, 0, TRUE, &status);
1715 TEST_ASSERT(status == U_NO_WRITE_PERMISSION);
1716
1717 status = U_ZERO_ERROR;
1718 ut = utext_openUnicodeString(ut, &ustr, &status);
1719 TEST_SUCCESS(status);
1720 ut2 = utext_clone(ut2, ut, FALSE, FALSE, &status); // clone with readonly = false
1721 TEST_SUCCESS(status);
1722 writable = utext_isWritable(ut2);
1723 TEST_ASSERT(writable == TRUE);
1724 ut2 = utext_clone(ut2, ut, FALSE, TRUE, &status); // clone with readonly = true
1725 TEST_SUCCESS(status);
1726 writable = utext_isWritable(ut2);
1727 TEST_ASSERT(writable == FALSE);
1728 utext_copy(ut2, 1, 2, 0, TRUE, &status);
1729 TEST_ASSERT(status == U_NO_WRITE_PERMISSION);
1730
1731 status = U_ZERO_ERROR;
1732 ut = utext_openConstUnicodeString(ut, (const UnicodeString *)&ustr, &status);
1733 TEST_SUCCESS(status);
1734 writable = utext_isWritable(ut);
1735 TEST_ASSERT(writable == FALSE);
1736 utext_copy(ut, 1, 2, 0, TRUE, &status);
1737 TEST_ASSERT(status == U_NO_WRITE_PERMISSION);
1738
1739 // Deep Clone of a frozen UText should re-enable writing in the copy.
1740 status = U_ZERO_ERROR;
1741 ut = utext_openUnicodeString(ut, &ustr, &status);
1742 TEST_SUCCESS(status);
1743 utext_freeze(ut);
1744 ut2 = utext_clone(ut2, ut, TRUE, FALSE, &status); // deep clone
1745 TEST_SUCCESS(status);
1746 writable = utext_isWritable(ut2);
1747 TEST_ASSERT(writable == TRUE);
1748
1749
1750 // Deep clone of a frozen UText, where the base type is intrinsically non-writable,
1751 // should NOT enable writing in the copy.
1752 status = U_ZERO_ERROR;
1753 ut = utext_openUChars(ut, u16str, -1, &status);
1754 TEST_SUCCESS(status);
1755 utext_freeze(ut);
1756 ut2 = utext_clone(ut2, ut, TRUE, FALSE, &status); // deep clone
1757 TEST_SUCCESS(status);
1758 writable = utext_isWritable(ut2);
1759 TEST_ASSERT(writable == FALSE);
1760
1761 // cleanup
1762 utext_close(ut);
1763 utext_close(ut2);
1764 }
1765
1766
1767 //
1768 // Fragmented UText
1769 // A UText type that works with a chunk size of 1.
1770 // Intended to test for edge cases.
1771 // Input comes from a UnicodeString.
1772 //
1773 // ut.b the character. Put into both halves.
1774 //
1775
1776 U_CDECL_BEGIN
1777 static UBool U_CALLCONV
fragTextAccess(UText * ut,int64_t index,UBool forward)1778 fragTextAccess(UText *ut, int64_t index, UBool forward) {
1779 const UnicodeString *us = (const UnicodeString *)ut->context;
1780 UChar c;
1781 int32_t length = us->length();
1782 if (forward && index>=0 && index<length) {
1783 c = us->charAt((int32_t)index);
1784 ut->b = c | c<<16;
1785 ut->chunkOffset = 0;
1786 ut->chunkLength = 1;
1787 ut->chunkNativeStart = index;
1788 ut->chunkNativeLimit = index+1;
1789 return true;
1790 }
1791 if (!forward && index>0 && index <=length) {
1792 c = us->charAt((int32_t)index-1);
1793 ut->b = c | c<<16;
1794 ut->chunkOffset = 1;
1795 ut->chunkLength = 1;
1796 ut->chunkNativeStart = index-1;
1797 ut->chunkNativeLimit = index;
1798 return true;
1799 }
1800 ut->b = 0;
1801 ut->chunkOffset = 0;
1802 ut->chunkLength = 0;
1803 if (index <= 0) {
1804 ut->chunkNativeStart = 0;
1805 ut->chunkNativeLimit = 0;
1806 } else {
1807 ut->chunkNativeStart = length;
1808 ut->chunkNativeLimit = length;
1809 }
1810 return false;
1811 }
1812
1813 // Function table to be used with this fragmented text provider.
1814 // Initialized in the open function.
1815 static UTextFuncs fragmentFuncs;
1816
1817 // Clone function for fragmented text provider.
1818 // Didn't really want to provide this, but it's easier to provide it than to keep it
1819 // out of the tests.
1820 //
1821 UText *
cloneFragmentedUnicodeString(UText * dest,const UText * src,UBool deep,UErrorCode * status)1822 cloneFragmentedUnicodeString(UText *dest, const UText *src, UBool deep, UErrorCode *status) {
1823 if (U_FAILURE(*status)) {
1824 return NULL;
1825 }
1826 if (deep) {
1827 *status = U_UNSUPPORTED_ERROR;
1828 return NULL;
1829 }
1830 dest = utext_openUnicodeString(dest, (UnicodeString *)src->context, status);
1831 utext_setNativeIndex(dest, utext_getNativeIndex(src));
1832 return dest;
1833 }
1834
1835 U_CDECL_END
1836
1837 // Open function for the fragmented text provider.
1838 UText *
openFragmentedUnicodeString(UText * ut,UnicodeString * s,UErrorCode * status)1839 openFragmentedUnicodeString(UText *ut, UnicodeString *s, UErrorCode *status) {
1840 ut = utext_openUnicodeString(ut, s, status);
1841 if (U_FAILURE(*status)) {
1842 return ut;
1843 }
1844
1845 // Copy of the function table from the stock UnicodeString UText,
1846 // and replace the entry for the access function.
1847 memcpy(&fragmentFuncs, ut->pFuncs, sizeof(fragmentFuncs));
1848 fragmentFuncs.access = fragTextAccess;
1849 fragmentFuncs.clone = cloneFragmentedUnicodeString;
1850 ut->pFuncs = &fragmentFuncs;
1851
1852 ut->chunkContents = (UChar *)&ut->b;
1853 ut->pFuncs->access(ut, 0, TRUE);
1854 return ut;
1855 }
1856
1857 // Regression test for Ticket 5560
1858 // Clone fails to update chunkContentPointer in the cloned copy.
1859 // This is only an issue for UText types that work in a local buffer,
1860 // (UTF-8 wrapper, for example)
1861 //
1862 // The test:
1863 // 1. Create an inital UText
1864 // 2. Deep clone it. Contents should match original.
1865 // 3. Reset original to something different.
1866 // 4. Check that clone contents did not change.
1867 //
Ticket5560()1868 void UTextTest::Ticket5560() {
1869 /* The following two strings are in UTF-8 even on EBCDIC platforms. */
1870 static const char s1[] = {0x41,0x42,0x43,0x44,0x45,0x46,0}; /* "ABCDEF" */
1871 static const char s2[] = {0x31,0x32,0x33,0x34,0x35,0x36,0}; /* "123456" */
1872 UErrorCode status = U_ZERO_ERROR;
1873
1874 UText ut1 = UTEXT_INITIALIZER;
1875 UText ut2 = UTEXT_INITIALIZER;
1876
1877 utext_openUTF8(&ut1, s1, -1, &status);
1878 UChar c = utext_next32(&ut1);
1879 TEST_ASSERT(c == 0x41); // c == 'A'
1880
1881 utext_clone(&ut2, &ut1, TRUE, FALSE, &status);
1882 TEST_SUCCESS(status);
1883 c = utext_next32(&ut2);
1884 TEST_ASSERT(c == 0x42); // c == 'B'
1885 c = utext_next32(&ut1);
1886 TEST_ASSERT(c == 0x42); // c == 'B'
1887
1888 utext_openUTF8(&ut1, s2, -1, &status);
1889 c = utext_next32(&ut1);
1890 TEST_ASSERT(c == 0x31); // c == '1'
1891 c = utext_next32(&ut2);
1892 TEST_ASSERT(c == 0x43); // c == 'C'
1893
1894 utext_close(&ut1);
1895 utext_close(&ut2);
1896 }
1897
1898
1899 // Test for Ticket 6847
1900 //
Ticket6847()1901 void UTextTest::Ticket6847() {
1902 const int STRLEN = 90;
1903 UChar s[STRLEN+1];
1904 u_memset(s, 0x41, STRLEN);
1905 s[STRLEN] = 0;
1906
1907 UErrorCode status = U_ZERO_ERROR;
1908 UText *ut = utext_openUChars(NULL, s, -1, &status);
1909
1910 utext_setNativeIndex(ut, 0);
1911 int32_t count = 0;
1912 UChar32 c = 0;
1913 int32_t nativeIndex = UTEXT_GETNATIVEINDEX(ut);
1914 TEST_ASSERT(nativeIndex == 0);
1915 while ((c = utext_next32(ut)) != U_SENTINEL) {
1916 TEST_ASSERT(c == 0x41);
1917 TEST_ASSERT(count < STRLEN);
1918 if (count >= STRLEN) {
1919 break;
1920 }
1921 count++;
1922 nativeIndex = UTEXT_GETNATIVEINDEX(ut);
1923 TEST_ASSERT(nativeIndex == count);
1924 }
1925 TEST_ASSERT(count == STRLEN);
1926 nativeIndex = UTEXT_GETNATIVEINDEX(ut);
1927 TEST_ASSERT(nativeIndex == STRLEN);
1928 utext_close(ut);
1929 }
1930
1931