]> git.saurik.com Git - apple/icu.git/blob - icuSources/test/cintltst/cbiapts.c
projects / apple / icu.git / blob
? search:
summary | shortlog | log | commit | commitdiff | tree
blame | history | raw | HEAD
ICU-57132.0.1.tar.gz
[apple/icu.git] / icuSources / test / cintltst / cbiapts.c
1 /********************************************************************
2 * COPYRIGHT:
3 * Copyright (c) 1997-2016, International Business Machines Corporation and
4 * others. All Rights Reserved.
5 ********************************************************************/
6 /********************************************************************************
7 *
8 * File CBIAPTS.C
9 *
10 * Modification History:
11 * Name Description
12 * Madhu Katragadda Creation
13 *********************************************************************************/
14 /*C API TEST FOR BREAKITERATOR */
15 /**
16 * This is an API test. It doesn't test very many cases, and doesn't
17 * try to test the full functionality. It just calls each function in the class and
18 * verifies that it works on a basic level.
19 **/
20
21 #include "unicode/utypes.h"
22
23 #if !UCONFIG_NO_BREAK_ITERATION
24
25 #include <stdlib.h>
26 #include <string.h>
27 #include "unicode/uloc.h"
28 #include "unicode/ubrk.h"
29 #include "unicode/ustring.h"
30 #include "unicode/ucnv.h"
31 #include "unicode/utext.h"
32 #include "cintltst.h"
33 #include "cbiapts.h"
34 #include "cmemory.h"
35
36 #define TEST_ASSERT_SUCCESS(status) {if (U_FAILURE(status)) { \
37 log_data_err("Failure at file %s, line %d, error = %s (Are you missing data?)\n", __FILE__, __LINE__, u_errorName(status));}}
38
39 #define TEST_ASSERT(expr) {if ((expr)==FALSE) { \
40 log_data_err("Test Failure at file %s, line %d (Are you missing data?)\n", __FILE__, __LINE__);}}
41
42 #if !UCONFIG_NO_FILE_IO
43 static void TestBreakIteratorSafeClone(void);
44 #endif
45 static void TestBreakIteratorRules(void);
46 static void TestBreakIteratorRuleError(void);
47 static void TestBreakIteratorStatusVec(void);
48 static void TestBreakIteratorUText(void);
49 static void TestBreakIteratorTailoring(void);
50 static void TestBreakIteratorRefresh(void);
51 static void TestBug11665(void);
52 static void TestBreakIteratorSuppressions(void);
53
54 void addBrkIterAPITest(TestNode** root);
55
56 void addBrkIterAPITest(TestNode** root)
57 {
58 #if !UCONFIG_NO_FILE_IO
59 addTest(root, &TestBreakIteratorCAPI, "tstxtbd/cbiapts/TestBreakIteratorCAPI");
60 addTest(root, &TestBreakIteratorSafeClone, "tstxtbd/cbiapts/TestBreakIteratorSafeClone");
61 addTest(root, &TestBreakIteratorUText, "tstxtbd/cbiapts/TestBreakIteratorUText");
62 #endif
63 addTest(root, &TestBreakIteratorRules, "tstxtbd/cbiapts/TestBreakIteratorRules");
64 addTest(root, &TestBreakIteratorRuleError, "tstxtbd/cbiapts/TestBreakIteratorRuleError");
65 addTest(root, &TestBreakIteratorStatusVec, "tstxtbd/cbiapts/TestBreakIteratorStatusVec");
66 addTest(root, &TestBreakIteratorTailoring, "tstxtbd/cbiapts/TestBreakIteratorTailoring");
67 addTest(root, &TestBreakIteratorRefresh, "tstxtbd/cbiapts/TestBreakIteratorRefresh");
68 addTest(root, &TestBug11665, "tstxtbd/cbiapts/TestBug11665");
69 addTest(root, &TestBreakIteratorSuppressions, "tstxtbd/cbiapts/TestBreakIteratorSuppressions");
70 }
71
72 #define CLONETEST_ITERATOR_COUNT 2
73
74 /*
75 * Utility function for converting char * to UChar * strings, to
76 * simplify the test code. Converted strings are put in heap allocated
77 * storage. A hook (probably a local in the caller's code) allows all
78 * strings converted with that hook to be freed with a single call.
79 */
80 typedef struct StringStruct {
81 struct StringStruct *link;
82 UChar str[1];
83 } StringStruct;
84
85
86 static UChar* toUChar(const char *src, void **freeHook) {
87 /* Structure of the memory that we allocate on the heap */
88
89 int32_t numUChars;
90 int32_t destSize;
91 UChar stackBuf[2000 + sizeof(void *)/sizeof(UChar)];
92 StringStruct *dest;
93 UConverter *cnv;
94
95 UErrorCode status = U_ZERO_ERROR;
96 if (src == NULL) {
97 return NULL;
98 };
99
100 cnv = ucnv_open(NULL, &status);
101 if(U_FAILURE(status) || cnv == NULL) {
102 return NULL;
103 }
104 ucnv_reset(cnv);
105 numUChars = ucnv_toUChars(cnv,
106 stackBuf,
107 2000,
108 src, -1,
109 &status);
110
111 destSize = (numUChars+1) * sizeof(UChar) + sizeof(struct StringStruct);
112 dest = (StringStruct *)malloc(destSize);
113 if (dest != NULL) {
114 if (status == U_BUFFER_OVERFLOW_ERROR || status == U_STRING_NOT_TERMINATED_WARNING) {
115 ucnv_toUChars(cnv, dest->str, numUChars+1, src, -1, &status);
116 } else if (status == U_ZERO_ERROR) {
117 u_strcpy(dest->str, stackBuf);
118 } else {
119 free(dest);
120 dest = NULL;
121 }
122 }
123
124 ucnv_reset(cnv); /* be good citizens */
125 ucnv_close(cnv);
126 if (dest == NULL) {
127 return NULL;
128 }
129
130 dest->link = (StringStruct*)(*freeHook);
131 *freeHook = dest;
132 return dest->str;
133 }
134
135 static void freeToUCharStrings(void **hook) {
136 StringStruct *s = *(StringStruct **)hook;
137 while (s != NULL) {
138 StringStruct *next = s->link;
139 free(s);
140 s = next;
141 }
142 }
143
144
145 #if !UCONFIG_NO_FILE_IO
146 static void TestBreakIteratorCAPI()
147 {
148 UErrorCode status = U_ZERO_ERROR;
149 UBreakIterator *word, *sentence, *line, *character, *b, *bogus;
150 int32_t start,pos,end,to;
151 int32_t i;
152 int32_t count = 0;
153
154 UChar text[50];
155
156 /* Note: the adjacent "" are concatenating strings, not adding a \" to the
157 string, which is probably what whoever wrote this intended. Don't fix,
158 because it would throw off the hard coded break positions in the following
159 tests. */
160 u_uastrcpy(text, "He's from Africa. ""Mr. Livingston, I presume?"" Yeah");
161
162
163 /*test ubrk_open()*/
164 log_verbose("\nTesting BreakIterator open functions\n");
165
166 /* Use french for fun */
167 word = ubrk_open(UBRK_WORD, "en_US", text, u_strlen(text), &status);
168 if(status == U_FILE_ACCESS_ERROR) {
169 log_data_err("Check your data - it doesn't seem to be around\n");
170 return;
171 } else if(U_FAILURE(status)){
172 log_err_status(status, "FAIL: Error in ubrk_open() for word breakiterator: %s\n", myErrorName(status));
173 }
174 else{
175 log_verbose("PASS: Successfully opened word breakiterator\n");
176 }
177
178 sentence = ubrk_open(UBRK_SENTENCE, "en_US", text, u_strlen(text), &status);
179 if(U_FAILURE(status)){
180 log_err_status(status, "FAIL: Error in ubrk_open() for sentence breakiterator: %s\n", myErrorName(status));
181 return;
182 }
183 else{
184 log_verbose("PASS: Successfully opened sentence breakiterator\n");
185 }
186
187 line = ubrk_open(UBRK_LINE, "en_US", text, u_strlen(text), &status);
188 if(U_FAILURE(status)){
189 log_err("FAIL: Error in ubrk_open() for line breakiterator: %s\n", myErrorName(status));
190 return;
191 }
192 else{
193 log_verbose("PASS: Successfully opened line breakiterator\n");
194 }
195
196 character = ubrk_open(UBRK_CHARACTER, "en_US", text, u_strlen(text), &status);
197 if(U_FAILURE(status)){
198 log_err("FAIL: Error in ubrk_open() for character breakiterator: %s\n", myErrorName(status));
199 return;
200 }
201 else{
202 log_verbose("PASS: Successfully opened character breakiterator\n");
203 }
204 /*trying to open an illegal iterator*/
205 bogus = ubrk_open((UBreakIteratorType)5, "en_US", text, u_strlen(text), &status);
206 if(bogus != NULL) {
207 log_err("FAIL: expected NULL from opening an invalid break iterator.\n");
208 }
209 if(U_SUCCESS(status)){
210 log_err("FAIL: Error in ubrk_open() for BOGUS breakiterator. Expected U_ILLEGAL_ARGUMENT_ERROR\n");
211 }
212 if(U_FAILURE(status)){
213 if(status != U_ILLEGAL_ARGUMENT_ERROR){
214 log_err("FAIL: Error in ubrk_open() for BOGUS breakiterator. Expected U_ILLEGAL_ARGUMENT_ERROR\n Got %s\n", myErrorName(status));
215 }
216 }
217 status=U_ZERO_ERROR;
218
219
220 /* ======= Test ubrk_countAvialable() and ubrk_getAvialable() */
221
222 log_verbose("\nTesting ubrk_countAvailable() and ubrk_getAvailable()\n");
223 count=ubrk_countAvailable();
224 /* use something sensible w/o hardcoding the count */
225 if(count < 0){
226 log_err("FAIL: Error in ubrk_countAvialable() returned %d\n", count);
227 }
228 else{
229 log_verbose("PASS: ubrk_countAvialable() successful returned %d\n", count);
230 }
231 for(i=0;i<count;i++)
232 {
233 log_verbose("%s\n", ubrk_getAvailable(i));
234 if (ubrk_getAvailable(i) == 0)
235 log_err("No locale for which breakiterator is applicable\n");
236 else
237 log_verbose("A locale %s for which breakiterator is applicable\n",ubrk_getAvailable(i));
238 }
239
240 /*========Test ubrk_first(), ubrk_last()...... and other functions*/
241
242 log_verbose("\nTesting the functions for word\n");
243 start = ubrk_first(word);
244 if(start!=0)
245 log_err("error ubrk_start(word) did not return 0\n");
246 log_verbose("first (word = %d\n", (int32_t)start);
247 pos=ubrk_next(word);
248 if(pos!=4)
249 log_err("error ubrk_next(word) did not return 4\n");
250 log_verbose("next (word = %d\n", (int32_t)pos);
251 pos=ubrk_following(word, 4);
252 if(pos!=5)
253 log_err("error ubrl_following(word,4) did not return 6\n");
254 log_verbose("next (word = %d\n", (int32_t)pos);
255 end=ubrk_last(word);
256 if(end!=49)
257 log_err("error ubrk_last(word) did not return 49\n");
258 log_verbose("last (word = %d\n", (int32_t)end);
259
260 pos=ubrk_previous(word);
261 log_verbose("%d %d\n", end, pos);
262
263 pos=ubrk_previous(word);
264 log_verbose("%d \n", pos);
265
266 if (ubrk_isBoundary(word, 2) != FALSE) {
267 log_err("error ubrk_isBoundary(word, 2) did not return FALSE\n");
268 }
269 pos=ubrk_current(word);
270 if (pos != 4) {
271 log_err("error ubrk_current() != 4 after ubrk_isBoundary(word, 2)\n");
272 }
273 if (ubrk_isBoundary(word, 4) != TRUE) {
274 log_err("error ubrk_isBoundary(word, 4) did not return TRUE\n");
275 }
276
277
278
279 log_verbose("\nTesting the functions for character\n");
280 ubrk_first(character);
281 pos = ubrk_following(character, 5);
282 if(pos!=6)
283 log_err("error ubrk_following(character,5) did not return 6\n");
284 log_verbose("Following (character,5) = %d\n", (int32_t)pos);
285 pos=ubrk_following(character, 18);
286 if(pos!=19)
287 log_err("error ubrk_following(character,18) did not return 19\n");
288 log_verbose("Followingcharacter,18) = %d\n", (int32_t)pos);
289 pos=ubrk_preceding(character, 22);
290 if(pos!=21)
291 log_err("error ubrk_preceding(character,22) did not return 21\n");
292 log_verbose("preceding(character,22) = %d\n", (int32_t)pos);
293
294
295 log_verbose("\nTesting the functions for line\n");
296 pos=ubrk_first(line);
297 if(pos != 0)
298 log_err("error ubrk_first(line) returned %d, expected 0\n", (int32_t)pos);
299 pos = ubrk_next(line);
300 pos=ubrk_following(line, 18);
301 if(pos!=22)
302 log_err("error ubrk_following(line) did not return 22\n");
303 log_verbose("following (line) = %d\n", (int32_t)pos);
304
305
306 log_verbose("\nTesting the functions for sentence\n");
307 ubrk_first(sentence);
308 pos = ubrk_current(sentence);
309 log_verbose("Current(sentence) = %d\n", (int32_t)pos);
310 pos = ubrk_last(sentence);
311 if(pos!=49)
312 log_err("error ubrk_last for sentence did not return 49\n");
313 log_verbose("Last (sentence) = %d\n", (int32_t)pos);
314 ubrk_first(sentence);
315 to = ubrk_following( sentence, 0 );
316 if (to == 0) log_err("ubrk_following returned 0\n");
317 to = ubrk_preceding( sentence, to );
318 if (to != 0) log_err("ubrk_preceding didn't return 0\n");
319 if (ubrk_first(sentence)!=ubrk_current(sentence)) {
320 log_err("error in ubrk_first() or ubrk_current()\n");
321 }
322
323
324 /*---- */
325 /*Testing ubrk_open and ubrk_close()*/
326 log_verbose("\nTesting open and close for us locale\n");
327 b = ubrk_open(UBRK_WORD, "fr_FR", text, u_strlen(text), &status);
328 if (U_FAILURE(status)) {
329 log_err("ubrk_open for word returned NULL: %s\n", myErrorName(status));
330 }
331 ubrk_close(b);
332
333 /* Test setText and setUText */
334 {
335 UChar s1[] = {0x41, 0x42, 0x20, 0};
336 UChar s2[] = {0x41, 0x42, 0x43, 0x44, 0x45, 0};
337 UText *ut = NULL;
338 UBreakIterator *bb;
339 int j;
340
341 log_verbose("\nTesting ubrk_setText() and ubrk_setUText()\n");
342 status = U_ZERO_ERROR;
343 bb = ubrk_open(UBRK_WORD, "en_US", NULL, 0, &status);
344 TEST_ASSERT_SUCCESS(status);
345 ubrk_setText(bb, s1, -1, &status);
346 TEST_ASSERT_SUCCESS(status);
347 ubrk_first(bb);
348 j = ubrk_next(bb);
349 TEST_ASSERT(j == 2);
350 ut = utext_openUChars(ut, s2, -1, &status);
351 ubrk_setUText(bb, ut, &status);
352 TEST_ASSERT_SUCCESS(status);
353 j = ubrk_next(bb);
354 TEST_ASSERT(j == 5);
355
356 ubrk_close(bb);
357 utext_close(ut);
358 }
359
360 ubrk_close(word);
361 ubrk_close(sentence);
362 ubrk_close(line);
363 ubrk_close(character);
364 }
365
366 static void TestBreakIteratorSafeClone(void)
367 {
368 UChar text[51]; /* Keep this odd to test for 64-bit memory alignment */
369 /* NOTE: This doesn't reliably force mis-alignment of following items. */
370 uint8_t buffer [CLONETEST_ITERATOR_COUNT] [U_BRK_SAFECLONE_BUFFERSIZE];
371 int32_t bufferSize = U_BRK_SAFECLONE_BUFFERSIZE;
372
373 UBreakIterator * someIterators [CLONETEST_ITERATOR_COUNT];
374 UBreakIterator * someClonedIterators [CLONETEST_ITERATOR_COUNT];
375
376 UBreakIterator * brk;
377 UErrorCode status = U_ZERO_ERROR;
378 int32_t start,pos;
379 int32_t i;
380
381 /*Testing ubrk_safeClone */
382
383 /* Note: the adjacent "" are concatenating strings, not adding a \" to the
384 string, which is probably what whoever wrote this intended. Don't fix,
385 because it would throw off the hard coded break positions in the following
386 tests. */
387 u_uastrcpy(text, "He's from Africa. ""Mr. Livingston, I presume?"" Yeah");
388
389 /* US & Thai - rule-based & dictionary based */
390 someIterators[0] = ubrk_open(UBRK_WORD, "en_US", text, u_strlen(text), &status);
391 if(!someIterators[0] || U_FAILURE(status)) {
392 log_data_err("Couldn't open en_US word break iterator - %s\n", u_errorName(status));
393 return;
394 }
395
396 someIterators[1] = ubrk_open(UBRK_WORD, "th_TH", text, u_strlen(text), &status);
397 if(!someIterators[1] || U_FAILURE(status)) {
398 log_data_err("Couldn't open th_TH word break iterator - %s\n", u_errorName(status));
399 return;
400 }
401
402 /* test each type of iterator */
403 for (i = 0; i < CLONETEST_ITERATOR_COUNT; i++)
404 {
405
406 /* Check the various error & informational states */
407
408 /* Null status - just returns NULL */
409 if (NULL != ubrk_safeClone(someIterators[i], buffer[i], &bufferSize, NULL))
410 {
411 log_err("FAIL: Cloned Iterator failed to deal correctly with null status\n");
412 }
413 /* error status - should return 0 & keep error the same */
414 status = U_MEMORY_ALLOCATION_ERROR;
415 if (NULL != ubrk_safeClone(someIterators[i], buffer[i], &bufferSize, &status) || status != U_MEMORY_ALLOCATION_ERROR)
416 {
417 log_err("FAIL: Cloned Iterator failed to deal correctly with incoming error status\n");
418 }
419 status = U_ZERO_ERROR;
420
421 /* Null buffer size pointer is ok */
422 if (NULL == (brk = ubrk_safeClone(someIterators[i], buffer[i], NULL, &status)) || U_FAILURE(status))
423 {
424 log_err("FAIL: Cloned Iterator failed to deal correctly with null bufferSize pointer\n");
425 }
426 ubrk_close(brk);
427 status = U_ZERO_ERROR;
428
429 /* buffer size pointer is 0 - fill in pbufferSize with a size */
430 bufferSize = 0;
431 if (NULL != ubrk_safeClone(someIterators[i], buffer[i], &bufferSize, &status) ||
432 U_FAILURE(status) || bufferSize <= 0)
433 {
434 log_err("FAIL: Cloned Iterator failed a sizing request ('preflighting')\n");
435 }
436 /* Verify our define is large enough */
437 if (U_BRK_SAFECLONE_BUFFERSIZE < bufferSize)
438 {
439 log_err("FAIL: Pre-calculated buffer size is too small - %d but needed %d\n", U_BRK_SAFECLONE_BUFFERSIZE, bufferSize);
440 }
441 /* Verify we can use this run-time calculated size */
442 if (NULL == (brk = ubrk_safeClone(someIterators[i], buffer[i], &bufferSize, &status)) || U_FAILURE(status))
443 {
444 log_err("FAIL: Iterator can't be cloned with run-time size\n");
445 }
446 if (brk)
447 ubrk_close(brk);
448 /* size one byte too small - should allocate & let us know */
449 if (bufferSize > 1) {
450 --bufferSize;
451 }
452 if (NULL == (brk = ubrk_safeClone(someIterators[i], NULL, &bufferSize, &status)) || status != U_SAFECLONE_ALLOCATED_WARNING)
453 {
454 log_err("FAIL: Cloned Iterator failed to deal correctly with too-small buffer size\n");
455 }
456 if (brk)
457 ubrk_close(brk);
458 status = U_ZERO_ERROR;
459 bufferSize = U_BRK_SAFECLONE_BUFFERSIZE;
460
461 /* Null buffer pointer - return Iterator & set error to U_SAFECLONE_ALLOCATED_ERROR */
462 if (NULL == (brk = ubrk_safeClone(someIterators[i], NULL, &bufferSize, &status)) || status != U_SAFECLONE_ALLOCATED_WARNING)
463 {
464 log_err("FAIL: Cloned Iterator failed to deal correctly with null buffer pointer\n");
465 }
466 if (brk)
467 ubrk_close(brk);
468 status = U_ZERO_ERROR;
469
470 /* Mis-aligned buffer pointer. */
471 {
472 char stackBuf[U_BRK_SAFECLONE_BUFFERSIZE+sizeof(void *)];
473
474 brk = ubrk_safeClone(someIterators[i], &stackBuf[1], &bufferSize, &status);
475 if (U_FAILURE(status) || brk == NULL) {
476 log_err("FAIL: Cloned Iterator failed with misaligned buffer pointer\n");
477 }
478 if (status == U_SAFECLONE_ALLOCATED_WARNING) {
479 log_verbose("Cloned Iterator allocated when using a mis-aligned buffer.\n");
480 }
481 if (brk)
482 ubrk_close(brk);
483 }
484
485
486 /* Null Iterator - return NULL & set U_ILLEGAL_ARGUMENT_ERROR */
487 if (NULL != ubrk_safeClone(NULL, buffer[i], &bufferSize, &status) || status != U_ILLEGAL_ARGUMENT_ERROR)
488 {
489 log_err("FAIL: Cloned Iterator failed to deal correctly with null Iterator pointer\n");
490 }
491 status = U_ZERO_ERROR;
492
493 /* Do these cloned Iterators work at all - make a first & next call */
494 bufferSize = U_BRK_SAFECLONE_BUFFERSIZE;
495 someClonedIterators[i] = ubrk_safeClone(someIterators[i], buffer[i], &bufferSize, &status);
496
497 start = ubrk_first(someClonedIterators[i]);
498 if(start!=0)
499 log_err("error ubrk_start(clone) did not return 0\n");
500 pos=ubrk_next(someClonedIterators[i]);
501 if(pos!=4)
502 log_err("error ubrk_next(clone) did not return 4\n");
503
504 ubrk_close(someClonedIterators[i]);
505 ubrk_close(someIterators[i]);
506 }
507 }
508 #endif
509
510
511 /*
512 // Open a break iterator from char * rules. Take care of conversion
513 // of the rules and error checking.
514 */
515 static UBreakIterator * testOpenRules(char *rules) {
516 UErrorCode status = U_ZERO_ERROR;
517 UChar *ruleSourceU = NULL;
518 void *strCleanUp = NULL;
519 UParseError parseErr;
520 UBreakIterator *bi;
521
522 ruleSourceU = toUChar(rules, &strCleanUp);
523
524 bi = ubrk_openRules(ruleSourceU, -1, /* The rules */
525 NULL, -1, /* The text to be iterated over. */
526 &parseErr, &status);
527
528 if (U_FAILURE(status)) {
529 log_data_err("FAIL: ubrk_openRules: ICU Error \"%s\" (Are you missing data?)\n", u_errorName(status));
530 bi = 0;
531 };
532 freeToUCharStrings(&strCleanUp);
533 return bi;
534
535 }
536
537 /*
538 * TestBreakIteratorRules - Verify that a break iterator can be created from
539 * a set of source rules.
540 */
541 static void TestBreakIteratorRules() {
542 /* Rules will keep together any run of letters not including 'a', OR
543 * keep together 'abc', but only when followed by 'def', OTHERWISE
544 * just return one char at a time.
545 */
546 char rules[] = "abc/def{666};\n [\\p{L} - [a]]* {2}; . {1};";
547 /* 0123456789012345678 */
548 char data[] = "abcdex abcdefgh-def"; /* the test data string */
549 char breaks[] = "** ** * ** *"; /* * the expected break positions */
550 char tags[] = "01 21 6 21 2"; /* expected tag values at break positions */
551 int32_t tagMap[] = {0, 1, 2, 3, 4, 5, 666};
552
553 UChar *uData;
554 void *freeHook = NULL;
555 UErrorCode status = U_ZERO_ERROR;
556 int32_t pos;
557 int i;
558
559 UBreakIterator *bi = testOpenRules(rules);
560 if (bi == NULL) {return;}
561 uData = toUChar(data, &freeHook);
562 ubrk_setText(bi, uData, -1, &status);
563
564 pos = ubrk_first(bi);
565 for (i=0; i<sizeof(breaks); i++) {
566 if (pos == i && breaks[i] != '*') {
567 log_err("FAIL: unexpected break at position %d found\n", pos);
568 break;
569 }
570 if (pos != i && breaks[i] == '*') {
571 log_err("FAIL: expected break at position %d not found.\n", i);
572 break;
573 }
574 if (pos == i) {
575 int32_t tag, expectedTag;
576 tag = ubrk_getRuleStatus(bi);
577 expectedTag = tagMap[tags[i]&0xf];
578 if (tag != expectedTag) {
579 log_err("FAIL: incorrect tag value. Position = %d; expected tag %d, got %d",
580 pos, expectedTag, tag);
581 break;
582 }
583 pos = ubrk_next(bi);
584 }
585 }
586
587 freeToUCharStrings(&freeHook);
588 ubrk_close(bi);
589 }
590
591 static void TestBreakIteratorRuleError() {
592 /*
593 * TestBreakIteratorRuleError - Try to create a BI from rules with syntax errors,
594 * check that the error is reported correctly.
595 */
596 char rules[] = " # This is a rule comment on line 1\n"
597 "[:L:]; # this rule is OK.\n"
598 "abcdefg); # Error, mismatched parens\n";
599 UChar *uRules;
600 void *freeHook = NULL;
601 UErrorCode status = U_ZERO_ERROR;
602 UParseError parseErr;
603 UBreakIterator *bi;
604
605 uRules = toUChar(rules, &freeHook);
606 bi = ubrk_openRules(uRules, -1, /* The rules */
607 NULL, -1, /* The text to be iterated over. */
608 &parseErr, &status);
609 if (U_SUCCESS(status)) {
610 log_err("FAIL: construction of break iterator succeeded when it should have failed.\n");
611 ubrk_close(bi);
612 } else {
613 if (parseErr.line != 3 || parseErr.offset != 8) {
614 log_data_err("FAIL: incorrect error position reported. Got line %d, char %d, expected line 3, char 7 (Are you missing data?)\n",
615 parseErr.line, parseErr.offset);
616 }
617 }
618 freeToUCharStrings(&freeHook);
619 }
620
621
622 /*
623 * TestsBreakIteratorStatusVals() Test the ubrk_getRuleStatusVec() funciton
624 */
625 static void TestBreakIteratorStatusVec() {
626 #define RULE_STRING_LENGTH 200
627 UChar rules[RULE_STRING_LENGTH];
628
629 #define TEST_STRING_LENGTH 25
630 UChar testString[TEST_STRING_LENGTH];
631 UBreakIterator *bi = NULL;
632 int32_t pos = 0;
633 int32_t vals[10];
634 int32_t numVals;
635 UErrorCode status = U_ZERO_ERROR;
636
637 u_uastrncpy(rules, "[A-N]{100}; \n"
638 "[a-w]{200}; \n"
639 "[\\p{L}]{300}; \n"
640 "[\\p{N}]{400}; \n"
641 "[0-5]{500}; \n"
642 "!.*;\n", RULE_STRING_LENGTH);
643 u_uastrncpy(testString, "ABC", TEST_STRING_LENGTH);
644
645
646 bi = ubrk_openRules(rules, -1, testString, -1, NULL, &status);
647 TEST_ASSERT_SUCCESS(status);
648 TEST_ASSERT(bi != NULL);
649
650 /* The TEST_ASSERT above should change too... */
651 if (bi != NULL) {
652 pos = ubrk_next(bi);
653 TEST_ASSERT(pos == 1);
654
655 memset(vals, -1, sizeof(vals));
656 numVals = ubrk_getRuleStatusVec(bi, vals, 10, &status);
657 TEST_ASSERT_SUCCESS(status);
658 TEST_ASSERT(numVals == 2);
659 TEST_ASSERT(vals[0] == 100);
660 TEST_ASSERT(vals[1] == 300);
661 TEST_ASSERT(vals[2] == -1);
662
663 numVals = ubrk_getRuleStatusVec(bi, vals, 0, &status);
664 TEST_ASSERT(status == U_BUFFER_OVERFLOW_ERROR);
665 TEST_ASSERT(numVals == 2);
666 }
667
668 ubrk_close(bi);
669 }
670
671
672 /*
673 * static void TestBreakIteratorUText(void);
674 *
675 * Test that ubrk_setUText() is present and works for a simple case.
676 */
677 static void TestBreakIteratorUText(void) {
678 const char *UTF8Str = "\x41\xc3\x85\x5A\x20\x41\x52\x69\x6E\x67"; /* c3 85 is utf-8 for A with a ring on top */
679 /* 0 1 2 34567890 */
680
681 UErrorCode status = U_ZERO_ERROR;
682 UBreakIterator *bi = NULL;
683 int32_t pos = 0;
684
685
686 UText *ut = utext_openUTF8(NULL, UTF8Str, -1, &status);
687 TEST_ASSERT_SUCCESS(status);
688
689 bi = ubrk_open(UBRK_WORD, "en_US", NULL, 0, &status);
690 if (U_FAILURE(status)) {
691 log_err_status(status, "Failure at file %s, line %d, error = %s\n", __FILE__, __LINE__, u_errorName(status));
692 return;
693 }
694
695 ubrk_setUText(bi, ut, &status);
696 if (U_FAILURE(status)) {
697 log_err("Failure at file %s, line %d, error = %s\n", __FILE__, __LINE__, u_errorName(status));
698 return;
699 }
700
701 pos = ubrk_first(bi);
702 TEST_ASSERT(pos == 0);
703
704 pos = ubrk_next(bi);
705 TEST_ASSERT(pos == 4);
706
707 pos = ubrk_next(bi);
708 TEST_ASSERT(pos == 5);
709
710 pos = ubrk_next(bi);
711 TEST_ASSERT(pos == 10);
712
713 pos = ubrk_next(bi);
714 TEST_ASSERT(pos == UBRK_DONE);
715 ubrk_close(bi);
716 utext_close(ut);
717 }
718
719 /*
720 * static void TestBreakIteratorTailoring(void);
721 *
722 * Test break iterator tailorings from CLDR data.
723 */
724
725 /* Thai/Lao grapheme break tailoring */
726 static const UChar thTest[] = { 0x0020, 0x0E40, 0x0E01, 0x0020,
727 0x0E01, 0x0E30, 0x0020, 0x0E01, 0x0E33, 0x0020, 0 };
728 /*in Unicode 6.1 en should behave just like th for this*/
729 /*static const int32_t thTestOffs_enFwd[] = { 1, 3, 4, 6, 7, 9, 10 };*/
730 static const int32_t thTestOffs_thFwd[] = { 1, 2, 3, 4, 5, 6, 7, 9, 10 };
731 /*static const int32_t thTestOffs_enRev[] = { 9, 7, 6, 4, 3, 1, 0 };*/
732 static const int32_t thTestOffs_thRev[] = { 9, 7, 6, 5, 4, 3, 2, 1, 0 };
733
734 /* Hebrew line break tailoring, for cldrbug 3028 */
735 static const UChar heTest[] = { 0x0020, 0x002D, 0x0031, 0x0032, 0x0020,
736 0x0061, 0x002D, 0x006B, 0x0020,
737 0x0061, 0x0300, 0x2010, 0x006B, 0x0020,
738 0x05DE, 0x05D4, 0x002D, 0x0069, 0x0020,
739 0x05D1, 0x05BC, 0x2010, 0x0047, 0x0020, 0 };
740 /*in Unicode 6.1 en should behave just like he for this*/
741 /*static const int32_t heTestOffs_enFwd[] = { 1, 5, 7, 9, 12, 14, 17, 19, 22, 24 };*/
742 static const int32_t heTestOffs_heFwd[] = { 1, 5, 7, 9, 12, 14, 19, 24 };
743 /*static const int32_t heTestOffs_enRev[] = { 22, 19, 17, 14, 12, 9, 7, 5, 1, 0 };*/
744 static const int32_t heTestOffs_heRev[] = { 19, 14, 12, 9, 7, 5, 1, 0 };
745
746 /* Finnish line break tailoring, for cldrbug 3029 */
747 static const UChar fiTest[] = { /* 00 */ 0x0020, 0x002D, 0x0031, 0x0032, 0x0020,
748 /* 05 */ 0x0061, 0x002D, 0x006B, 0x0020,
749 /* 09 */ 0x0061, 0x0300, 0x2010, 0x006B, 0x0020,
750 /* 14 */ 0x0061, 0x0020, 0x002D, 0x006B, 0x0020,
751 /* 19 */ 0x0061, 0x0300, 0x0020, 0x2010, 0x006B, 0x0020, 0 };
752 static const int32_t fiTestOffs_enFwd[] = { 1, 5, 7, 9, 12, 14, 16, 17, 19, 22, 23, 25 };
753 static const int32_t fiTestOffs_fiFwd[] = { 1, 5, 7, 9, 12, 14, 16, 19, 22, 25 };
754 static const int32_t fiTestOffs_enRev[] = { 23, 22, 19, 17, 16, 14, 12, 9, 7, 5, 1, 0 };
755 static const int32_t fiTestOffs_fiRev[] = { 22, 19, 16, 14, 12, 9, 7, 5, 1, 0 };
756
757 /* Khmer dictionary-based work break, for ICU ticket #8329 */
758 static const UChar kmTest[] = { /* 00 */ 0x179F, 0x17BC, 0x1798, 0x1785, 0x17C6, 0x178E, 0x17B6, 0x1799, 0x1796, 0x17C1,
759 /* 10 */ 0x179B, 0x1794, 0x1793, 0x17D2, 0x178F, 0x17B7, 0x1785, 0x178A, 0x17BE, 0x1798,
760 /* 20 */ 0x17D2, 0x1794, 0x17B8, 0x17A2, 0x1792, 0x17B7, 0x179F, 0x17D2, 0x178B, 0x17B6,
761 /* 30 */ 0x1793, 0x17A2, 0x179A, 0x1796, 0x17D2, 0x179A, 0x17C7, 0x1782, 0x17BB, 0x178E,
762 /* 40 */ 0x178A, 0x179B, 0x17CB, 0x1796, 0x17D2, 0x179A, 0x17C7, 0x17A2, 0x1784, 0x17D2,
763 /* 50 */ 0x1782, 0 };
764 static const int32_t kmTestOffs_kmFwd[] = { 3, /*8,*/ 11, 17, 23, 31, /*33,*/ 40, 43, 51 }; /* TODO: Investigate failure to break at offset 8 */
765 static const int32_t kmTestOffs_kmRev[] = { 43, 40, /*33,*/ 31, 23, 17, 11, /*8,*/ 3, 0 };
766
767 typedef struct {
768 const char * locale;
769 UBreakIteratorType type;
770 const UChar * test;
771 const int32_t * offsFwd;
772 const int32_t * offsRev;
773 int32_t numOffsets;
774 } RBBITailoringTest;
775
776 static const RBBITailoringTest tailoringTests[] = {
777 { "en", UBRK_CHARACTER, thTest, thTestOffs_thFwd, thTestOffs_thRev, UPRV_LENGTHOF(thTestOffs_thFwd) },
778 { "en_US_POSIX", UBRK_CHARACTER, thTest, thTestOffs_thFwd, thTestOffs_thRev, UPRV_LENGTHOF(thTestOffs_thFwd) },
779 { "en", UBRK_LINE, heTest, heTestOffs_heFwd, heTestOffs_heRev, UPRV_LENGTHOF(heTestOffs_heFwd) },
780 { "he", UBRK_LINE, heTest, heTestOffs_heFwd, heTestOffs_heRev, UPRV_LENGTHOF(heTestOffs_heFwd) },
781 { "en", UBRK_LINE, fiTest, fiTestOffs_enFwd, fiTestOffs_enRev, UPRV_LENGTHOF(fiTestOffs_enFwd) },
782 { "fi", UBRK_LINE, fiTest, fiTestOffs_fiFwd, fiTestOffs_fiRev, UPRV_LENGTHOF(fiTestOffs_fiFwd) },
783 { "km", UBRK_WORD, kmTest, kmTestOffs_kmFwd, kmTestOffs_kmRev, UPRV_LENGTHOF(kmTestOffs_kmFwd) },
784 { NULL, 0, NULL, NULL, NULL, 0 },
785 };
786
787 static void TestBreakIteratorTailoring(void) {
788 const RBBITailoringTest * testPtr;
789 for (testPtr = tailoringTests; testPtr->locale != NULL; ++testPtr) {
790 UErrorCode status = U_ZERO_ERROR;
791 UBreakIterator* ubrkiter = ubrk_open(testPtr->type, testPtr->locale, testPtr->test, -1, &status);
792 if ( U_SUCCESS(status) ) {
793 int32_t offset, offsindx;
794 UBool foundError;
795
796 foundError = FALSE;
797 for (offsindx = 0; (offset = ubrk_next(ubrkiter)) != UBRK_DONE; ++offsindx) {
798 if (!foundError && offsindx >= testPtr->numOffsets) {
799 log_err("FAIL: locale %s, break type %d, ubrk_next expected UBRK_DONE, got %d\n",
800 testPtr->locale, testPtr->type, offset);
801 foundError = TRUE;
802 } else if (!foundError && offset != testPtr->offsFwd[offsindx]) {
803 log_err("FAIL: locale %s, break type %d, ubrk_next expected %d, got %d\n",
804 testPtr->locale, testPtr->type, testPtr->offsFwd[offsindx], offset);
805 foundError = TRUE;
806 }
807 }
808 if (!foundError && offsindx < testPtr->numOffsets) {
809 log_err("FAIL: locale %s, break type %d, ubrk_next expected %d, got UBRK_DONE\n",
810 testPtr->locale, testPtr->type, testPtr->offsFwd[offsindx]);
811 }
812
813 foundError = FALSE;
814 for (offsindx = 0; (offset = ubrk_previous(ubrkiter)) != UBRK_DONE; ++offsindx) {
815 if (!foundError && offsindx >= testPtr->numOffsets) {
816 log_err("FAIL: locale %s, break type %d, ubrk_previous expected UBRK_DONE, got %d\n",
817 testPtr->locale, testPtr->type, offset);
818 foundError = TRUE;
819 } else if (!foundError && offset != testPtr->offsRev[offsindx]) {
820 log_err("FAIL: locale %s, break type %d, ubrk_previous expected %d, got %d\n",
821 testPtr->locale, testPtr->type, testPtr->offsRev[offsindx], offset);
822 foundError = TRUE;
823 }
824 }
825 if (!foundError && offsindx < testPtr->numOffsets) {
826 log_err("FAIL: locale %s, break type %d, ubrk_previous expected %d, got UBRK_DONE\n",
827 testPtr->locale, testPtr->type, testPtr->offsRev[offsindx]);
828 }
829
830 ubrk_close(ubrkiter);
831 } else {
832 log_err_status(status, "FAIL: locale %s, break type %d, ubrk_open status: %s\n", testPtr->locale, testPtr->type, u_errorName(status));
833 }
834 }
835 }
836
837
838 static void TestBreakIteratorRefresh(void) {
839 /*
840 * RefreshInput changes out the input of a Break Iterator without
841 * changing anything else in the iterator's state. Used with Java JNI,
842 * when Java moves the underlying string storage. This test
843 * runs a ubrk_next() repeatedly, moving the text in the middle of the sequence.
844 * The right set of boundaries should still be found.
845 */
846 UChar testStr[] = {0x20, 0x41, 0x20, 0x42, 0x20, 0x43, 0x20, 0x44, 0x0}; /* = " A B C D" */
847 UChar movedStr[] = {0x20, 0x20, 0x20, 0x20, 0x20, 0x20, 0x20, 0x20, 0};
848 UErrorCode status = U_ZERO_ERROR;
849 UBreakIterator *bi;
850 UText ut1 = UTEXT_INITIALIZER;
851 UText ut2 = UTEXT_INITIALIZER;
852
853 bi = ubrk_open(UBRK_LINE, "en_US", NULL, 0, &status);
854 TEST_ASSERT_SUCCESS(status);
855 if (U_FAILURE(status)) {
856 return;
857 }
858
859 utext_openUChars(&ut1, testStr, -1, &status);
860 TEST_ASSERT_SUCCESS(status);
861 ubrk_setUText(bi, &ut1, &status);
862 TEST_ASSERT_SUCCESS(status);
863
864 if (U_SUCCESS(status)) {
865 /* Line boundaries will occur before each letter in the original string */
866 TEST_ASSERT(1 == ubrk_next(bi));
867 TEST_ASSERT(3 == ubrk_next(bi));
868
869 /* Move the string, kill the original string. */
870 u_strcpy(movedStr, testStr);
871 u_memset(testStr, 0x20, u_strlen(testStr));
872 utext_openUChars(&ut2, movedStr, -1, &status);
873 TEST_ASSERT_SUCCESS(status);
874 ubrk_refreshUText(bi, &ut2, &status);
875 TEST_ASSERT_SUCCESS(status);
876
877 /* Find the following matches, now working in the moved string. */
878 TEST_ASSERT(5 == ubrk_next(bi));
879 TEST_ASSERT(7 == ubrk_next(bi));
880 TEST_ASSERT(8 == ubrk_next(bi));
881 TEST_ASSERT(UBRK_DONE == ubrk_next(bi));
882 TEST_ASSERT_SUCCESS(status);
883
884 utext_close(&ut1);
885 utext_close(&ut2);
886 }
887 ubrk_close(bi);
888 }
889
890
891 static void TestBug11665(void) {
892 // The problem was with the incorrect breaking of Japanese text beginning
893 // with Katakana characters when no prior Japanese or Chinese text had been
894 // encountered.
895 //
896 // Tested here in cintltst, rather than in intltest, because only cintltst
897 // tests have the ability to reset ICU, which is needed to get the bug
898 // to manifest itself.
899
900 static UChar japaneseText[] = {0x30A2, 0x30EC, 0x30EB, 0x30AE, 0x30FC, 0x6027, 0x7D50, 0x819C, 0x708E};
901 int32_t boundaries[10] = {0};
902 UBreakIterator *bi = NULL;
903 int32_t brk;
904 int32_t brkIdx = 0;
905 int32_t totalBreaks = 0;
906 UErrorCode status = U_ZERO_ERROR;
907
908 ctest_resetICU();
909 bi = ubrk_open(UBRK_WORD, "en_US", japaneseText, UPRV_LENGTHOF(japaneseText), &status);
910 TEST_ASSERT_SUCCESS(status);
911 if (!bi) {
912 return;
913 }
914 for (brk=ubrk_first(bi); brk != UBRK_DONE; brk=ubrk_next(bi)) {
915 boundaries[brkIdx] = brk;
916 if (++brkIdx >= UPRV_LENGTHOF(boundaries) - 1) {
917 break;
918 }
919 }
920 if (brkIdx <= 2 || brkIdx >= UPRV_LENGTHOF(boundaries)) {
921 log_err("%s:%d too few or many breaks found.\n", __FILE__, __LINE__);
922 } else {
923 totalBreaks = brkIdx;
924 brkIdx = 0;
925 for (brk=ubrk_first(bi); brk != UBRK_DONE; brk=ubrk_next(bi)) {
926 if (brk != boundaries[brkIdx]) {
927 log_err("%s:%d Break #%d differs between first and second iteration.\n", __FILE__, __LINE__, brkIdx);
928 break;
929 }
930 if (++brkIdx >= UPRV_LENGTHOF(boundaries) - 1) {
931 log_err("%s:%d Too many breaks.\n", __FILE__, __LINE__);
932 break;
933 }
934 }
935 if (totalBreaks != brkIdx) {
936 log_err("%s:%d Number of breaks differ between first and second iteration.\n", __FILE__, __LINE__);
937 }
938 }
939 ubrk_close(bi);
940 }
941
942 /*
943 * expOffset is the set of expected offsets, ending with '-1'.
944 * "Expected expOffset -1" means "expected the end of the offsets"
945 */
946
947 static const char testSentenceSuppressionsEn[] = "Mr. Jones comes home. Dr. Smith Ph.D. is out. In the U.S.A. it is hot.";
948 static const int32_t testSentSuppFwdOffsetsEn[] = { 22, 46, 70, -1 }; /* With suppressions */
949 static const int32_t testSentFwdOffsetsEn[] = { 4, 22, 26, 46, 70, -1 }; /* Without suppressions */
950 static const int32_t testSentSuppRevOffsetsEn[] = { 46, 22, 0, -1 }; /* With suppressions */
951 static const int32_t testSentRevOffsetsEn[] = { 46, 26, 22, 4, 0, -1 }; /* Without suppressions */
952
953 static const char testSentenceSuppressionsDe[] = "Wenn ich schon h\\u00F6re zu Guttenberg kommt evtl. zur\\u00FCck.";
954 // "Wenn ich schon höre zu Guttenberg kommt evtl. zurück."
955 static const int32_t testSentSuppFwdOffsetsDe[] = { 53, -1 }; /* With suppressions */
956 static const int32_t testSentFwdOffsetsDe[] = { 53, -1 }; /* Without suppressions; no break in evtl. zur due to casing */
957 static const int32_t testSentSuppRevOffsetsDe[] = { 0, -1 }; /* With suppressions */
958 static const int32_t testSentRevOffsetsDe[] = { 0, -1 }; /* Without suppressions */
959
960 static const char testSentenceSuppressionsEs[] = "Te esperamos todos los miercoles en Bravo 416, Col. El Pueblo a las 7 PM.";
961 static const int32_t testSentSuppFwdOffsetsEs[] = { 73, -1 }; /* With suppressions */
962 static const int32_t testSentFwdOffsetsEs[] = { 52, 73, -1 }; /* Without suppressions */
963 static const int32_t testSentSuppRevOffsetsEs[] = { 0, -1 }; /* With suppressions */
964 static const int32_t testSentRevOffsetsEs[] = { 52, 0, -1 }; /* Without suppressions */
965
966 static const char testSentenceSuppressionsE1[] = "Add or detract. The world will little note.";
967 static const char testSentenceSuppressionsE1u[] = "ADD OR DETRACT. THE WORLD WILL LITTLE NOTE.";
968 static const int32_t testSentFwdOffsetsE1[] = { 16, 43, -1 }; /* Suppressions and case should make no difference */
969 static const int32_t testSentRevOffsetsE1[] = { 16, 0, -1 }; /* Suppressions and case should make no difference */
970
971 static const char testSentenceSuppressionsE2[] = "Coming up, the sprints at NCAA. Are you watching?";
972 static const char testSentenceSuppressionsE2u[] = "COMING UP, THE SPRINTS AT NCAA. ARE YOU WATCHING?";
973 static const int32_t testSentFwdOffsetsE2[] = { 32, 49, -1 }; /* Suppressions and case should make no difference */
974 static const int32_t testSentRevOffsetsE2[] = { 32, 0, -1 }; /* Suppressions and case should make no difference */
975
976 static const char testSentenceSuppressionsFr[] = "Tr\\u00E8s bonne prise de parole de M. Junod, municipal \\u00E0 la culture de Lausanne.";
977 // "Très bonne prise de parole de M. Junod, municipal à la culture de Lausanne."
978 static const int32_t testSentFwdOffsetsFr[] = { 33, 75, -1 }; /* Without suppressions */
979 static const int32_t testSentSuppFwdOffsetsFr[] = { 75, -1 }; /* With suppressions */
980 static const int32_t testSentRevOffsetsFr[] = { 33, 0, -1 }; /* Without suppressions */
981 static const int32_t testSentSuppRevOffsetsFr[] = { 0, -1 }; /* With suppressions */
982
983 static const char testSentenceSuppressionsE3[] = "G8 countries e.g. U.K., Japan. Sanctions i.e. restrictions. Test E. Xx G. Xx I. Xx.";
984 static const char testSentenceSuppressionsE3u[] = "G8 COUNTRIES E.G. U.K., JAPAN. SANCTIONS I.E. RESTRICTIONS. TEST E. XX G. XX I. XX.";
985 static const int32_t testSentSuppFwdOffsetsE3[] = { 31, 60, 83, -1 }; /* With suppressions */
986 static const int32_t testSentSuppRevOffsetsE3[] = { 60, 31, 0, -1 }; /* With suppressions */
987 static const int32_t testSentFwdOffsetsE3[] = { 18, 31, 60, 68, 74, 80, 83, -1 }; /* Without suppressions */
988 static const int32_t testSentRevOffsetsE3[] = { 80, 74, 68, 60, 31, 18, 0, -1 }; /* Without suppressions */
989 static const int32_t testSentFwdOffsetsE3u[] = { 18, 31, 46, 60, 68, 74, 80, 83, -1 }; /* Without suppressions */
990 static const int32_t testSentRevOffsetsE3u[] = { 80, 74, 68, 60, 46, 31, 18, 0, -1 }; /* Without suppressions */
991
992 enum { kTextULenMax = 128, kTextBLenMax = 192 };
993
994 typedef struct {
995 const char * locale;
996 const char * text;
997 const int32_t * expFwdOffsets;
998 const int32_t * expRevOffsets;
999 } TestBISuppressionsItem;
1000
1001 static const TestBISuppressionsItem testBISuppressionsItems[] = {
1002 { "en@ss=standard", testSentenceSuppressionsEn, testSentSuppFwdOffsetsEn, testSentSuppRevOffsetsEn },
1003 { "en", testSentenceSuppressionsEn, testSentFwdOffsetsEn, testSentRevOffsetsEn },
1004 { "fr@ss=standard", testSentenceSuppressionsEn, testSentFwdOffsetsEn, testSentRevOffsetsEn },
1005 { "af@ss=standard", testSentenceSuppressionsEn, testSentSuppFwdOffsetsEn, testSentSuppRevOffsetsEn }, /* no brkiter data => en suppressions? */
1006 { "zh@ss=standard", testSentenceSuppressionsEn, testSentFwdOffsetsEn, testSentRevOffsetsEn }, /* brkiter data, no suppressions data => no suppressions */
1007 { "zh_Hant@ss=standard", testSentenceSuppressionsEn, testSentFwdOffsetsEn, testSentRevOffsetsEn }, /* brkiter data, no suppressions data => no suppressions */
1008 { "fi@ss=standard", testSentenceSuppressionsEn, testSentFwdOffsetsEn, testSentRevOffsetsEn }, /* brkiter data, no suppressions data => no suppressions */
1009 { "ja@ss=standard", testSentenceSuppressionsEn, testSentFwdOffsetsEn, testSentRevOffsetsEn }, /* brkiter data, no suppressions data => no suppressions */
1010 { "de@ss=standard", testSentenceSuppressionsDe, testSentSuppFwdOffsetsDe, testSentSuppRevOffsetsDe },
1011 { "de", testSentenceSuppressionsDe, testSentFwdOffsetsDe, testSentRevOffsetsDe },
1012 { "es@ss=standard", testSentenceSuppressionsEs, testSentSuppFwdOffsetsEs, testSentSuppRevOffsetsEs },
1013 { "es", testSentenceSuppressionsEs, testSentFwdOffsetsEs, testSentRevOffsetsEs },
1014 { "en", testSentenceSuppressionsE1, testSentFwdOffsetsE1, testSentRevOffsetsE1 },
1015 { "en@ss=standard", testSentenceSuppressionsE1, testSentFwdOffsetsE1, testSentRevOffsetsE1 },
1016 { "en", testSentenceSuppressionsE1u, testSentFwdOffsetsE1, testSentRevOffsetsE1 },
1017 { "en@ss=standard", testSentenceSuppressionsE1u, testSentFwdOffsetsE1, testSentRevOffsetsE1 },
1018 { "en", testSentenceSuppressionsE2, testSentFwdOffsetsE2, testSentRevOffsetsE2 },
1019 { "en@ss=standard", testSentenceSuppressionsE2, testSentFwdOffsetsE2, testSentRevOffsetsE2 },
1020 { "en", testSentenceSuppressionsE2u, testSentFwdOffsetsE2, testSentRevOffsetsE2 },
1021 { "en@ss=standard", testSentenceSuppressionsE2u, testSentFwdOffsetsE2, testSentRevOffsetsE2 },
1022 { "fr", testSentenceSuppressionsFr, testSentFwdOffsetsFr, testSentRevOffsetsFr },
1023 { "fr@ss=standard", testSentenceSuppressionsFr, testSentSuppFwdOffsetsFr, testSentSuppRevOffsetsFr },
1024 { "en@ss=standard", testSentenceSuppressionsE3, testSentSuppFwdOffsetsE3, testSentSuppRevOffsetsE3 },
1025 { "en", testSentenceSuppressionsE3, testSentFwdOffsetsE3, testSentRevOffsetsE3 },
1026 { "en@ss=standard", testSentenceSuppressionsE3u, testSentSuppFwdOffsetsE3, testSentSuppRevOffsetsE3 },
1027 { "en", testSentenceSuppressionsE3u, testSentFwdOffsetsE3u, testSentRevOffsetsE3u },
1028 { NULL, NULL, NULL }
1029 };
1030
1031 static void TestBreakIteratorSuppressions(void) {
1032 const TestBISuppressionsItem * itemPtr;
1033
1034 for (itemPtr = testBISuppressionsItems; itemPtr->locale != NULL; itemPtr++) {
1035 UChar textU[kTextULenMax + 1];
1036 char textB[kTextBLenMax];
1037 int32_t textULen = u_unescape(itemPtr->text, textU, kTextULenMax);
1038 textU[kTextULenMax] = 0; // ensure zero termination
1039 UErrorCode status = U_ZERO_ERROR;
1040 UBreakIterator *bi = ubrk_open(UBRK_SENTENCE, itemPtr->locale, textU, textULen, &status);
1041 log_verbose("#%d: %s\n", (itemPtr-testBISuppressionsItems), itemPtr->locale);
1042 if (U_SUCCESS(status)) {
1043 int32_t offset, start;
1044 const int32_t * expOffsetPtr;
1045 const int32_t * expOffsetStart;
1046 u_austrcpy(textB, textU);
1047
1048 expOffsetStart = expOffsetPtr = itemPtr->expFwdOffsets;
1049 ubrk_first(bi);
1050 for (; (offset = ubrk_next(bi)) != UBRK_DONE && *expOffsetPtr >= 0; expOffsetPtr++) {
1051 if (offset != *expOffsetPtr) {
1052 log_err("FAIL: ubrk_next loc \"%s\", expected %d, got %d, text \"%s\"\n",
1053 itemPtr->locale, *expOffsetPtr, offset, textB);
1054 }
1055 }
1056 if (offset != UBRK_DONE || *expOffsetPtr >= 0) {
1057 log_err("FAIL: ubrk_next loc \"%s\", expected UBRK_DONE & expOffset -1, got %d and %d, text \"%s\"\n",
1058 itemPtr->locale, offset, *expOffsetPtr, textB);
1059 }
1060
1061 expOffsetStart = expOffsetPtr = itemPtr->expFwdOffsets;
1062 start = ubrk_first(bi) + 1;
1063 for (; (offset = ubrk_following(bi, start)) != UBRK_DONE && *expOffsetPtr >= 0; expOffsetPtr++) {
1064 if (offset != *expOffsetPtr) {
1065 log_err("FAIL: ubrk_following(%d) loc \"%s\", expected %d, got %d, text \"%s\"\n",
1066 start, itemPtr->locale, *expOffsetPtr, offset, textB);
1067 }
1068 start = *expOffsetPtr + 1;
1069 }
1070 if (offset != UBRK_DONE || *expOffsetPtr >= 0) {
1071 log_err("FAIL: ubrk_following(%d) loc \"%s\", expected UBRK_DONE & expOffset -1, got %d and %d, text \"%s\"\n",
1072 start, itemPtr->locale, offset, *expOffsetPtr, textB);
1073 }
1074
1075 expOffsetStart = expOffsetPtr = itemPtr->expRevOffsets;
1076 offset = ubrk_last(bi);
1077 log_verbose("___ @%d ubrk_last\n", offset);
1078 if(offset == 0) {
1079 log_err("FAIL: ubrk_last loc \"%s\" unexpected %d\n", itemPtr->locale, offset);
1080 }
1081 for (; (offset = ubrk_previous(bi)) != UBRK_DONE && *expOffsetPtr >= 0; expOffsetPtr++) {
1082 if (offset != *expOffsetPtr) {
1083 log_err("FAIL: ubrk_previous loc \"%s\", expected %d, got %d, text \"%s\"\n",
1084 itemPtr->locale, *expOffsetPtr, offset, textB);
1085 } else {
1086 log_verbose("[%d] @%d ubrk_previous()\n", (expOffsetPtr - expOffsetStart), offset);
1087 }
1088 }
1089 if (offset != UBRK_DONE || *expOffsetPtr >= 0) {
1090 log_err("FAIL: ubrk_previous loc \"%s\", expected UBRK_DONE & expOffset[%d] -1, got %d and %d, text \"%s\"\n",
1091 itemPtr->locale, expOffsetPtr - expOffsetStart, offset, *expOffsetPtr, textB);
1092 }
1093
1094 expOffsetStart = expOffsetPtr = itemPtr->expRevOffsets;
1095 start = ubrk_last(bi) - 1;
1096 for (; (offset = ubrk_preceding(bi, start)) != UBRK_DONE && *expOffsetPtr >= 0; expOffsetPtr++) {
1097 if (offset != *expOffsetPtr) {
1098 log_err("FAIL: ubrk_preceding(%d) loc \"%s\", expected %d, got %d, text \"%s\"\n",
1099 start, itemPtr->locale, *expOffsetPtr, offset, textB);
1100 }
1101 start = *expOffsetPtr - 1;
1102 }
1103 if (start >=0 && (offset != UBRK_DONE || *expOffsetPtr >= 0)) {
1104 log_err("FAIL: ubrk_preceding loc(%d) \"%s\", expected UBRK_DONE & expOffset -1, got %d and %d, text \"%s\"\n",
1105 start, itemPtr->locale, offset, *expOffsetPtr, textB);
1106 }
1107
1108 ubrk_close(bi);
1109 } else {
1110 log_data_err("FAIL: ubrk_open(UBRK_SENTENCE, \"%s\", ...) status %s (Are you missing data?)\n",
1111 itemPtr->locale, u_errorName(status));
1112 }
1113 }
1114 }
1115
1116
1117 #endif /* #if !UCONFIG_NO_BREAK_ITERATION */
mirror of ICU