1 // © 2016 and later: Unicode, Inc. and others.
2 // License & terms of use: http://www.unicode.org/copyright.html
3 /********************************************************************
4 * COPYRIGHT:
5 * Copyright (c) 1997-2016, International Business Machines Corporation and
6 * others. All Rights Reserved.
7 ********************************************************************/
8 /********************************************************************************
9 *
10 * File CBIAPTS.C
11 *
12 * Modification History:
13 * Name Description
14 * Madhu Katragadda Creation
15 *********************************************************************************/
16 /*C API TEST FOR BREAKITERATOR */
17 /**
18 * This is an API test. It doesn't test very many cases, and doesn't
19 * try to test the full functionality. It just calls each function in the class and
20 * verifies that it works on a basic level.
21 **/
22
23 #include "unicode/utypes.h"
24
25 #if !UCONFIG_NO_BREAK_ITERATION
26
27 #include <stdlib.h>
28 #include <string.h>
29 #include "unicode/uloc.h"
30 #include "unicode/ubrk.h"
31 #include "unicode/ustring.h"
32 #include "unicode/ucnv.h"
33 #include "unicode/utext.h"
34 #include "cintltst.h"
35 #include "cbiapts.h"
36 #include "cmemory.h"
37
38 #define TEST_ASSERT_SUCCESS(status) UPRV_BLOCK_MACRO_BEGIN { \
39 if (U_FAILURE(status)) { \
40 log_data_err("Failure at file %s, line %d, error = %s (Are you missing data?)\n", __FILE__, __LINE__, u_errorName(status)); \
41 } \
42 } UPRV_BLOCK_MACRO_END
43
44 #define TEST_ASSERT(expr) UPRV_BLOCK_MACRO_BEGIN { \
45 if ((expr)==FALSE) { \
46 log_data_err("Test Failure at file %s, line %d (Are you missing data?)\n", __FILE__, __LINE__); \
47 } \
48 } UPRV_BLOCK_MACRO_END
49
50 #if !UCONFIG_NO_FILE_IO
51 static void TestBreakIteratorSafeClone(void);
52 #endif
53 static void TestBreakIteratorRules(void);
54 static void TestBreakIteratorRuleError(void);
55 static void TestBreakIteratorStatusVec(void);
56 static void TestBreakIteratorUText(void);
57 static void TestBreakIteratorTailoring(void);
58 static void TestBreakIteratorRefresh(void);
59 static void TestBug11665(void);
60 static void TestBreakIteratorSuppressions(void);
61
62 void addBrkIterAPITest(TestNode** root);
63
addBrkIterAPITest(TestNode ** root)64 void addBrkIterAPITest(TestNode** root)
65 {
66 #if !UCONFIG_NO_FILE_IO
67 addTest(root, &TestBreakIteratorCAPI, "tstxtbd/cbiapts/TestBreakIteratorCAPI");
68 addTest(root, &TestBreakIteratorSafeClone, "tstxtbd/cbiapts/TestBreakIteratorSafeClone");
69 addTest(root, &TestBreakIteratorUText, "tstxtbd/cbiapts/TestBreakIteratorUText");
70 #endif
71 addTest(root, &TestBreakIteratorRules, "tstxtbd/cbiapts/TestBreakIteratorRules");
72 addTest(root, &TestBreakIteratorRuleError, "tstxtbd/cbiapts/TestBreakIteratorRuleError");
73 addTest(root, &TestBreakIteratorStatusVec, "tstxtbd/cbiapts/TestBreakIteratorStatusVec");
74 addTest(root, &TestBreakIteratorTailoring, "tstxtbd/cbiapts/TestBreakIteratorTailoring");
75 addTest(root, &TestBreakIteratorRefresh, "tstxtbd/cbiapts/TestBreakIteratorRefresh");
76 addTest(root, &TestBug11665, "tstxtbd/cbiapts/TestBug11665");
77 #if !UCONFIG_NO_FILTERED_BREAK_ITERATION
78 addTest(root, &TestBreakIteratorSuppressions, "tstxtbd/cbiapts/TestBreakIteratorSuppressions");
79 #endif
80 }
81
82 #define CLONETEST_ITERATOR_COUNT 2
83
84 /*
85 * Utility function for converting char * to UChar * strings, to
86 * simplify the test code. Converted strings are put in heap allocated
87 * storage. A hook (probably a local in the caller's code) allows all
88 * strings converted with that hook to be freed with a single call.
89 */
90 typedef struct StringStruct {
91 struct StringStruct *link;
92 UChar str[1];
93 } StringStruct;
94
95
toUChar(const char * src,void ** freeHook)96 static UChar* toUChar(const char *src, void **freeHook) {
97 /* Structure of the memory that we allocate on the heap */
98
99 int32_t numUChars;
100 int32_t destSize;
101 UChar stackBuf[2000 + sizeof(void *)/sizeof(UChar)];
102 StringStruct *dest;
103 UConverter *cnv;
104
105 UErrorCode status = U_ZERO_ERROR;
106 if (src == NULL) {
107 return NULL;
108 }
109
110 cnv = ucnv_open(NULL, &status);
111 if(U_FAILURE(status) || cnv == NULL) {
112 return NULL;
113 }
114 ucnv_reset(cnv);
115 numUChars = ucnv_toUChars(cnv,
116 stackBuf,
117 2000,
118 src, -1,
119 &status);
120
121 destSize = (numUChars+1) * sizeof(UChar) + sizeof(struct StringStruct);
122 dest = (StringStruct *)malloc(destSize);
123 if (dest != NULL) {
124 if (status == U_BUFFER_OVERFLOW_ERROR || status == U_STRING_NOT_TERMINATED_WARNING) {
125 ucnv_toUChars(cnv, dest->str, numUChars+1, src, -1, &status);
126 } else if (status == U_ZERO_ERROR) {
127 u_strcpy(dest->str, stackBuf);
128 } else {
129 free(dest);
130 dest = NULL;
131 }
132 }
133
134 ucnv_reset(cnv); /* be good citizens */
135 ucnv_close(cnv);
136 if (dest == NULL) {
137 return NULL;
138 }
139
140 dest->link = (StringStruct*)(*freeHook);
141 *freeHook = dest;
142 return dest->str;
143 }
144
freeToUCharStrings(void ** hook)145 static void freeToUCharStrings(void **hook) {
146 StringStruct *s = *(StringStruct **)hook;
147 while (s != NULL) {
148 StringStruct *next = s->link;
149 free(s);
150 s = next;
151 }
152 }
153
154
155 #if !UCONFIG_NO_FILE_IO
TestBreakIteratorCAPI()156 static void TestBreakIteratorCAPI()
157 {
158 UErrorCode status = U_ZERO_ERROR;
159 UBreakIterator *word, *sentence, *line, *character, *b, *bogus;
160 int32_t start,pos,end,to;
161 int32_t i;
162 int32_t count = 0;
163
164 UChar text[50];
165
166 /* Note: the adjacent "" are concatenating strings, not adding a \" to the
167 string, which is probably what whoever wrote this intended. Don't fix,
168 because it would throw off the hard coded break positions in the following
169 tests. */
170 u_uastrcpy(text, "He's from Africa. ""Mr. Livingston, I presume?"" Yeah");
171
172
173 /*test ubrk_open()*/
174 log_verbose("\nTesting BreakIterator open functions\n");
175
176 /* Use french for fun */
177 word = ubrk_open(UBRK_WORD, "en_US", text, u_strlen(text), &status);
178 if(status == U_FILE_ACCESS_ERROR) {
179 log_data_err("Check your data - it doesn't seem to be around\n");
180 return;
181 } else if(U_FAILURE(status)){
182 log_err_status(status, "FAIL: Error in ubrk_open() for word breakiterator: %s\n", myErrorName(status));
183 }
184 else{
185 log_verbose("PASS: Successfully opened word breakiterator\n");
186 }
187
188 sentence = ubrk_open(UBRK_SENTENCE, "en_US", text, u_strlen(text), &status);
189 if(U_FAILURE(status)){
190 log_err_status(status, "FAIL: Error in ubrk_open() for sentence breakiterator: %s\n", myErrorName(status));
191 return;
192 }
193 else{
194 log_verbose("PASS: Successfully opened sentence breakiterator\n");
195 }
196
197 line = ubrk_open(UBRK_LINE, "en_US", text, u_strlen(text), &status);
198 if(U_FAILURE(status)){
199 log_err("FAIL: Error in ubrk_open() for line breakiterator: %s\n", myErrorName(status));
200 return;
201 }
202 else{
203 log_verbose("PASS: Successfully opened line breakiterator\n");
204 }
205
206 character = ubrk_open(UBRK_CHARACTER, "en_US", text, u_strlen(text), &status);
207 if(U_FAILURE(status)){
208 log_err("FAIL: Error in ubrk_open() for character breakiterator: %s\n", myErrorName(status));
209 return;
210 }
211 else{
212 log_verbose("PASS: Successfully opened character breakiterator\n");
213 }
214 /*trying to open an illegal iterator*/
215 bogus = ubrk_open((UBreakIteratorType)5, "en_US", text, u_strlen(text), &status);
216 if(bogus != NULL) {
217 log_err("FAIL: expected NULL from opening an invalid break iterator.\n");
218 }
219 if(U_SUCCESS(status)){
220 log_err("FAIL: Error in ubrk_open() for BOGUS breakiterator. Expected U_ILLEGAL_ARGUMENT_ERROR\n");
221 }
222 if(U_FAILURE(status)){
223 if(status != U_ILLEGAL_ARGUMENT_ERROR){
224 log_err("FAIL: Error in ubrk_open() for BOGUS breakiterator. Expected U_ILLEGAL_ARGUMENT_ERROR\n Got %s\n", myErrorName(status));
225 }
226 }
227 status=U_ZERO_ERROR;
228
229
230 /* ======= Test ubrk_countAvialable() and ubrk_getAvialable() */
231
232 log_verbose("\nTesting ubrk_countAvailable() and ubrk_getAvailable()\n");
233 count=ubrk_countAvailable();
234 /* use something sensible w/o hardcoding the count */
235 if(count < 0){
236 log_err("FAIL: Error in ubrk_countAvialable() returned %d\n", count);
237 }
238 else{
239 log_verbose("PASS: ubrk_countAvialable() successful returned %d\n", count);
240 }
241 for(i=0;i<count;i++)
242 {
243 log_verbose("%s\n", ubrk_getAvailable(i));
244 if (ubrk_getAvailable(i) == 0)
245 log_err("No locale for which breakiterator is applicable\n");
246 else
247 log_verbose("A locale %s for which breakiterator is applicable\n",ubrk_getAvailable(i));
248 }
249
250 /*========Test ubrk_first(), ubrk_last()...... and other functions*/
251
252 log_verbose("\nTesting the functions for word\n");
253 start = ubrk_first(word);
254 if(start!=0)
255 log_err("error ubrk_start(word) did not return 0\n");
256 log_verbose("first (word = %d\n", (int32_t)start);
257 pos=ubrk_next(word);
258 if(pos!=4)
259 log_err("error ubrk_next(word) did not return 4\n");
260 log_verbose("next (word = %d\n", (int32_t)pos);
261 pos=ubrk_following(word, 4);
262 if(pos!=5)
263 log_err("error ubrl_following(word,4) did not return 6\n");
264 log_verbose("next (word = %d\n", (int32_t)pos);
265 end=ubrk_last(word);
266 if(end!=49)
267 log_err("error ubrk_last(word) did not return 49\n");
268 log_verbose("last (word = %d\n", (int32_t)end);
269
270 pos=ubrk_previous(word);
271 log_verbose("%d %d\n", end, pos);
272
273 pos=ubrk_previous(word);
274 log_verbose("%d \n", pos);
275
276 if (ubrk_isBoundary(word, 2) != FALSE) {
277 log_err("error ubrk_isBoundary(word, 2) did not return FALSE\n");
278 }
279 pos=ubrk_current(word);
280 if (pos != 4) {
281 log_err("error ubrk_current() != 4 after ubrk_isBoundary(word, 2)\n");
282 }
283 if (ubrk_isBoundary(word, 4) != TRUE) {
284 log_err("error ubrk_isBoundary(word, 4) did not return TRUE\n");
285 }
286
287
288
289 log_verbose("\nTesting the functions for character\n");
290 ubrk_first(character);
291 pos = ubrk_following(character, 5);
292 if(pos!=6)
293 log_err("error ubrk_following(character,5) did not return 6\n");
294 log_verbose("Following (character,5) = %d\n", (int32_t)pos);
295 pos=ubrk_following(character, 18);
296 if(pos!=19)
297 log_err("error ubrk_following(character,18) did not return 19\n");
298 log_verbose("Followingcharacter,18) = %d\n", (int32_t)pos);
299 pos=ubrk_preceding(character, 22);
300 if(pos!=21)
301 log_err("error ubrk_preceding(character,22) did not return 21\n");
302 log_verbose("preceding(character,22) = %d\n", (int32_t)pos);
303
304
305 log_verbose("\nTesting the functions for line\n");
306 pos=ubrk_first(line);
307 if(pos != 0)
308 log_err("error ubrk_first(line) returned %d, expected 0\n", (int32_t)pos);
309 pos = ubrk_next(line);
310 pos=ubrk_following(line, 18);
311 if(pos!=22)
312 log_err("error ubrk_following(line) did not return 22\n");
313 log_verbose("following (line) = %d\n", (int32_t)pos);
314
315
316 log_verbose("\nTesting the functions for sentence\n");
317 pos = ubrk_first(sentence);
318 pos = ubrk_current(sentence);
319 log_verbose("Current(sentence) = %d\n", (int32_t)pos);
320 pos = ubrk_last(sentence);
321 if(pos!=49)
322 log_err("error ubrk_last for sentence did not return 49\n");
323 log_verbose("Last (sentence) = %d\n", (int32_t)pos);
324 pos = ubrk_first(sentence);
325 to = ubrk_following( sentence, 0 );
326 if (to == 0) log_err("ubrk_following returned 0\n");
327 to = ubrk_preceding( sentence, to );
328 if (to != 0) log_err("ubrk_preceding didn't return 0\n");
329 if (ubrk_first(sentence)!=ubrk_current(sentence)) {
330 log_err("error in ubrk_first() or ubrk_current()\n");
331 }
332
333
334 /*---- */
335 /*Testing ubrk_open and ubrk_close()*/
336 log_verbose("\nTesting open and close for us locale\n");
337 b = ubrk_open(UBRK_WORD, "fr_FR", text, u_strlen(text), &status);
338 if (U_FAILURE(status)) {
339 log_err("ubrk_open for word returned NULL: %s\n", myErrorName(status));
340 }
341 ubrk_close(b);
342
343 /* Test setText and setUText */
344 {
345 UChar s1[] = {0x41, 0x42, 0x20, 0};
346 UChar s2[] = {0x41, 0x42, 0x43, 0x44, 0x45, 0};
347 UText *ut = NULL;
348 UBreakIterator *bb;
349 int j;
350
351 log_verbose("\nTesting ubrk_setText() and ubrk_setUText()\n");
352 status = U_ZERO_ERROR;
353 bb = ubrk_open(UBRK_WORD, "en_US", NULL, 0, &status);
354 TEST_ASSERT_SUCCESS(status);
355 ubrk_setText(bb, s1, -1, &status);
356 TEST_ASSERT_SUCCESS(status);
357 ubrk_first(bb);
358 j = ubrk_next(bb);
359 TEST_ASSERT(j == 2);
360 ut = utext_openUChars(ut, s2, -1, &status);
361 ubrk_setUText(bb, ut, &status);
362 TEST_ASSERT_SUCCESS(status);
363 j = ubrk_next(bb);
364 TEST_ASSERT(j == 5);
365
366 ubrk_close(bb);
367 utext_close(ut);
368 }
369
370 ubrk_close(word);
371 ubrk_close(sentence);
372 ubrk_close(line);
373 ubrk_close(character);
374 }
375
TestBreakIteratorSafeClone(void)376 static void TestBreakIteratorSafeClone(void)
377 {
378 UChar text[51]; /* Keep this odd to test for 64-bit memory alignment */
379 /* NOTE: This doesn't reliably force mis-alignment of following items. */
380 uint8_t buffer [CLONETEST_ITERATOR_COUNT] [U_BRK_SAFECLONE_BUFFERSIZE];
381 int32_t bufferSize = U_BRK_SAFECLONE_BUFFERSIZE;
382
383 UBreakIterator * someIterators [CLONETEST_ITERATOR_COUNT];
384 UBreakIterator * someClonedIterators [CLONETEST_ITERATOR_COUNT];
385
386 UBreakIterator * brk;
387 UErrorCode status = U_ZERO_ERROR;
388 int32_t start,pos;
389 int32_t i;
390
391 /*Testing ubrk_safeClone */
392
393 /* Note: the adjacent "" are concatenating strings, not adding a \" to the
394 string, which is probably what whoever wrote this intended. Don't fix,
395 because it would throw off the hard coded break positions in the following
396 tests. */
397 u_uastrcpy(text, "He's from Africa. ""Mr. Livingston, I presume?"" Yeah");
398
399 /* US & Thai - rule-based & dictionary based */
400 someIterators[0] = ubrk_open(UBRK_WORD, "en_US", text, u_strlen(text), &status);
401 if(!someIterators[0] || U_FAILURE(status)) {
402 log_data_err("Couldn't open en_US word break iterator - %s\n", u_errorName(status));
403 return;
404 }
405
406 someIterators[1] = ubrk_open(UBRK_WORD, "th_TH", text, u_strlen(text), &status);
407 if(!someIterators[1] || U_FAILURE(status)) {
408 log_data_err("Couldn't open th_TH word break iterator - %s\n", u_errorName(status));
409 return;
410 }
411
412 /* test each type of iterator */
413 for (i = 0; i < CLONETEST_ITERATOR_COUNT; i++)
414 {
415
416 /* Check the various error & informational states */
417
418 /* Null status - just returns NULL */
419 if (NULL != ubrk_safeClone(someIterators[i], buffer[i], &bufferSize, NULL))
420 {
421 log_err("FAIL: Cloned Iterator failed to deal correctly with null status\n");
422 }
423 /* error status - should return 0 & keep error the same */
424 status = U_MEMORY_ALLOCATION_ERROR;
425 if (NULL != ubrk_safeClone(someIterators[i], buffer[i], &bufferSize, &status) || status != U_MEMORY_ALLOCATION_ERROR)
426 {
427 log_err("FAIL: Cloned Iterator failed to deal correctly with incoming error status\n");
428 }
429 status = U_ZERO_ERROR;
430
431 /* Null buffer size pointer is ok */
432 if (NULL == (brk = ubrk_safeClone(someIterators[i], buffer[i], NULL, &status)) || U_FAILURE(status))
433 {
434 log_err("FAIL: Cloned Iterator failed to deal correctly with null bufferSize pointer\n");
435 }
436 ubrk_close(brk);
437 status = U_ZERO_ERROR;
438
439 /* buffer size pointer is 0 - fill in pbufferSize with a size */
440 bufferSize = 0;
441 if (NULL != ubrk_safeClone(someIterators[i], buffer[i], &bufferSize, &status) ||
442 U_FAILURE(status) || bufferSize <= 0)
443 {
444 log_err("FAIL: Cloned Iterator failed a sizing request ('preflighting')\n");
445 }
446 /* Verify our define is large enough */
447 if (U_BRK_SAFECLONE_BUFFERSIZE < bufferSize)
448 {
449 log_err("FAIL: Pre-calculated buffer size is too small - %d but needed %d\n", U_BRK_SAFECLONE_BUFFERSIZE, bufferSize);
450 }
451 /* Verify we can use this run-time calculated size */
452 if (NULL == (brk = ubrk_safeClone(someIterators[i], buffer[i], &bufferSize, &status)) || U_FAILURE(status))
453 {
454 log_err("FAIL: Iterator can't be cloned with run-time size\n");
455 }
456 if (brk)
457 ubrk_close(brk);
458 /* size one byte too small - should allocate & let us know */
459 if (bufferSize > 1) {
460 --bufferSize;
461 }
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 too-small buffer size\n");
465 }
466 if (brk)
467 ubrk_close(brk);
468 status = U_ZERO_ERROR;
469 bufferSize = U_BRK_SAFECLONE_BUFFERSIZE;
470
471 /* Null buffer pointer - return Iterator & set error to U_SAFECLONE_ALLOCATED_ERROR */
472 if (NULL == (brk = ubrk_safeClone(someIterators[i], NULL, &bufferSize, &status)) || status != U_SAFECLONE_ALLOCATED_WARNING)
473 {
474 log_err("FAIL: Cloned Iterator failed to deal correctly with null buffer pointer\n");
475 }
476 if (brk)
477 ubrk_close(brk);
478 status = U_ZERO_ERROR;
479
480 /* Mis-aligned buffer pointer. */
481 {
482 char stackBuf[U_BRK_SAFECLONE_BUFFERSIZE+sizeof(void *)];
483
484 brk = ubrk_safeClone(someIterators[i], &stackBuf[1], &bufferSize, &status);
485 if (U_FAILURE(status) || brk == NULL) {
486 log_err("FAIL: Cloned Iterator failed with misaligned buffer pointer\n");
487 }
488 if (status == U_SAFECLONE_ALLOCATED_WARNING) {
489 log_verbose("Cloned Iterator allocated when using a mis-aligned buffer.\n");
490 }
491 if (brk)
492 ubrk_close(brk);
493 }
494
495
496 /* Null Iterator - return NULL & set U_ILLEGAL_ARGUMENT_ERROR */
497 if (NULL != ubrk_safeClone(NULL, buffer[i], &bufferSize, &status) || status != U_ILLEGAL_ARGUMENT_ERROR)
498 {
499 log_err("FAIL: Cloned Iterator failed to deal correctly with null Iterator pointer\n");
500 }
501 status = U_ZERO_ERROR;
502
503 /* Do these cloned Iterators work at all - make a first & next call */
504 bufferSize = U_BRK_SAFECLONE_BUFFERSIZE;
505 someClonedIterators[i] = ubrk_safeClone(someIterators[i], buffer[i], &bufferSize, &status);
506
507 start = ubrk_first(someClonedIterators[i]);
508 if(start!=0)
509 log_err("error ubrk_start(clone) did not return 0\n");
510 pos=ubrk_next(someClonedIterators[i]);
511 if(pos!=4)
512 log_err("error ubrk_next(clone) did not return 4\n");
513
514 ubrk_close(someClonedIterators[i]);
515 ubrk_close(someIterators[i]);
516 }
517 }
518 #endif
519
520
521 /*
522 // Open a break iterator from char * rules. Take care of conversion
523 // of the rules and error checking.
524 */
testOpenRules(char * rules)525 static UBreakIterator * testOpenRules(char *rules) {
526 UErrorCode status = U_ZERO_ERROR;
527 UChar *ruleSourceU = NULL;
528 void *strCleanUp = NULL;
529 UParseError parseErr;
530 UBreakIterator *bi;
531
532 ruleSourceU = toUChar(rules, &strCleanUp);
533
534 bi = ubrk_openRules(ruleSourceU, -1, /* The rules */
535 NULL, -1, /* The text to be iterated over. */
536 &parseErr, &status);
537
538 if (U_FAILURE(status)) {
539 log_data_err("FAIL: ubrk_openRules: ICU Error \"%s\" (Are you missing data?)\n", u_errorName(status));
540 bi = 0;
541 }
542 freeToUCharStrings(&strCleanUp);
543 return bi;
544
545 }
546
547 /*
548 * TestBreakIteratorRules - Verify that a break iterator can be created from
549 * a set of source rules.
550 */
TestBreakIteratorRules()551 static void TestBreakIteratorRules() {
552 /* Rules will keep together any run of letters not including 'a', OR
553 * keep together 'abc', but only when followed by 'def', OTHERWISE
554 * just return one char at a time.
555 */
556 char rules[] = "abc/def{666};\n [\\p{L} - [a]]* {2}; . {1};";
557 /* 0123456789012345678 */
558 char data[] = "abcdex abcdefgh-def"; /* the test data string */
559 char breaks[] = "** ** * ** *"; /* * the expected break positions */
560 char tags[] = "01 21 6 21 2"; /* expected tag values at break positions */
561 int32_t tagMap[] = {0, 1, 2, 3, 4, 5, 666};
562
563 UChar *uData;
564 void *freeHook = NULL;
565 UErrorCode status = U_ZERO_ERROR;
566 int32_t pos;
567 int i;
568
569 UBreakIterator *bi = testOpenRules(rules);
570 if (bi == NULL) {return;}
571 uData = toUChar(data, &freeHook);
572 ubrk_setText(bi, uData, -1, &status);
573
574 pos = ubrk_first(bi);
575 for (i=0; i<(int)sizeof(breaks); i++) {
576 if (pos == i && breaks[i] != '*') {
577 log_err("FAIL: unexpected break at position %d found\n", pos);
578 break;
579 }
580 if (pos != i && breaks[i] == '*') {
581 log_err("FAIL: expected break at position %d not found.\n", i);
582 break;
583 }
584 if (pos == i) {
585 int32_t tag, expectedTag;
586 tag = ubrk_getRuleStatus(bi);
587 expectedTag = tagMap[tags[i]&0xf];
588 if (tag != expectedTag) {
589 log_err("FAIL: incorrect tag value. Position = %d; expected tag %d, got %d",
590 pos, expectedTag, tag);
591 break;
592 }
593 pos = ubrk_next(bi);
594 }
595 }
596
597 /* #12914 add basic sanity test for ubrk_getBinaryRules, ubrk_openBinaryRules */
598 /* Underlying functionality checked in C++ rbbiapts.cpp TestRoundtripRules */
599 status = U_ZERO_ERROR;
600 int32_t rulesLength = ubrk_getBinaryRules(bi, NULL, 0, &status); /* preflight */
601 if (U_FAILURE(status)) {
602 log_err("FAIL: ubrk_getBinaryRules preflight err: %s", u_errorName(status));
603 } else {
604 uint8_t* binaryRules = (uint8_t*)uprv_malloc(rulesLength);
605 if (binaryRules == NULL) {
606 log_err("FAIL: unable to malloc rules buffer, size %u", rulesLength);
607 } else {
608 rulesLength = ubrk_getBinaryRules(bi, binaryRules, rulesLength, &status);
609 if (U_FAILURE(status)) {
610 log_err("FAIL: ubrk_getBinaryRules err: %s", u_errorName(status));
611 } else {
612 UBreakIterator* bi2 = ubrk_openBinaryRules(binaryRules, rulesLength, uData, -1, &status);
613 if (U_FAILURE(status)) {
614 log_err("FAIL: ubrk_openBinaryRules err: %s", u_errorName(status));
615 } else {
616 int32_t maxCount = sizeof(breaks); /* fail-safe test limit */
617 int32_t pos2 = ubrk_first(bi2);
618 pos = ubrk_first(bi);
619 do {
620 if (pos2 != pos) {
621 log_err("FAIL: interator from ubrk_openBinaryRules does not match original, get pos = %d instead of %d", pos2, pos);
622 }
623 pos2 = ubrk_next(bi2);
624 pos = ubrk_next(bi);
625 } while ((pos != UBRK_DONE || pos2 != UBRK_DONE) && maxCount-- > 0);
626
627 ubrk_close(bi2);
628 }
629 }
630 uprv_free(binaryRules);
631 }
632 }
633
634 freeToUCharStrings(&freeHook);
635 ubrk_close(bi);
636 }
637
TestBreakIteratorRuleError()638 static void TestBreakIteratorRuleError() {
639 /*
640 * TestBreakIteratorRuleError - Try to create a BI from rules with syntax errors,
641 * check that the error is reported correctly.
642 */
643 char rules[] = " # This is a rule comment on line 1\n"
644 "[:L:]; # this rule is OK.\n"
645 "abcdefg); # Error, mismatched parens\n";
646 UChar *uRules;
647 void *freeHook = NULL;
648 UErrorCode status = U_ZERO_ERROR;
649 UParseError parseErr;
650 UBreakIterator *bi;
651
652 uRules = toUChar(rules, &freeHook);
653 bi = ubrk_openRules(uRules, -1, /* The rules */
654 NULL, -1, /* The text to be iterated over. */
655 &parseErr, &status);
656 if (U_SUCCESS(status)) {
657 log_err("FAIL: construction of break iterator succeeded when it should have failed.\n");
658 ubrk_close(bi);
659 } else {
660 if (parseErr.line != 3 || parseErr.offset != 8) {
661 log_data_err("FAIL: incorrect error position reported. Got line %d, char %d, expected line 3, char 7 (Are you missing data?)\n",
662 parseErr.line, parseErr.offset);
663 }
664 }
665 freeToUCharStrings(&freeHook);
666 }
667
668
669 /*
670 * TestsBreakIteratorStatusVals() Test the ubrk_getRuleStatusVec() funciton
671 */
TestBreakIteratorStatusVec()672 static void TestBreakIteratorStatusVec() {
673 #define RULE_STRING_LENGTH 200
674 UChar rules[RULE_STRING_LENGTH];
675
676 #define TEST_STRING_LENGTH 25
677 UChar testString[TEST_STRING_LENGTH];
678 UBreakIterator *bi = NULL;
679 int32_t pos = 0;
680 int32_t vals[10];
681 int32_t numVals;
682 UErrorCode status = U_ZERO_ERROR;
683
684 u_uastrncpy(rules, "[A-N]{100}; \n"
685 "[a-w]{200}; \n"
686 "[\\p{L}]{300}; \n"
687 "[\\p{N}]{400}; \n"
688 "[0-5]{500}; \n"
689 "!.*;\n", RULE_STRING_LENGTH);
690 u_uastrncpy(testString, "ABC", TEST_STRING_LENGTH);
691
692
693 bi = ubrk_openRules(rules, -1, testString, -1, NULL, &status);
694 TEST_ASSERT_SUCCESS(status);
695 TEST_ASSERT(bi != NULL);
696
697 /* The TEST_ASSERT above should change too... */
698 if (bi != NULL) {
699 pos = ubrk_next(bi);
700 TEST_ASSERT(pos == 1);
701
702 memset(vals, -1, sizeof(vals));
703 numVals = ubrk_getRuleStatusVec(bi, vals, 10, &status);
704 TEST_ASSERT_SUCCESS(status);
705 TEST_ASSERT(numVals == 2);
706 TEST_ASSERT(vals[0] == 100);
707 TEST_ASSERT(vals[1] == 300);
708 TEST_ASSERT(vals[2] == -1);
709
710 numVals = ubrk_getRuleStatusVec(bi, vals, 0, &status);
711 TEST_ASSERT(status == U_BUFFER_OVERFLOW_ERROR);
712 TEST_ASSERT(numVals == 2);
713 }
714
715 ubrk_close(bi);
716 }
717
718
719 /*
720 * static void TestBreakIteratorUText(void);
721 *
722 * Test that ubrk_setUText() is present and works for a simple case.
723 */
TestBreakIteratorUText(void)724 static void TestBreakIteratorUText(void) {
725 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 */
726 /* 0 1 2 34567890 */
727
728 UErrorCode status = U_ZERO_ERROR;
729 UBreakIterator *bi = NULL;
730 int32_t pos = 0;
731
732
733 UText *ut = utext_openUTF8(NULL, UTF8Str, -1, &status);
734 TEST_ASSERT_SUCCESS(status);
735
736 bi = ubrk_open(UBRK_WORD, "en_US", NULL, 0, &status);
737 if (U_FAILURE(status)) {
738 log_err_status(status, "Failure at file %s, line %d, error = %s\n", __FILE__, __LINE__, u_errorName(status));
739 utext_close(ut);
740 return;
741 }
742
743 ubrk_setUText(bi, ut, &status);
744 if (U_FAILURE(status)) {
745 log_err("Failure at file %s, line %d, error = %s\n", __FILE__, __LINE__, u_errorName(status));
746 ubrk_close(bi);
747 utext_close(ut);
748 return;
749 }
750
751 pos = ubrk_first(bi);
752 TEST_ASSERT(pos == 0);
753
754 pos = ubrk_next(bi);
755 TEST_ASSERT(pos == 4);
756
757 pos = ubrk_next(bi);
758 TEST_ASSERT(pos == 5);
759
760 pos = ubrk_next(bi);
761 TEST_ASSERT(pos == 10);
762
763 pos = ubrk_next(bi);
764 TEST_ASSERT(pos == UBRK_DONE);
765 ubrk_close(bi);
766 utext_close(ut);
767 }
768
769 /*
770 * static void TestBreakIteratorTailoring(void);
771 *
772 * Test break iterator tailorings from CLDR data.
773 */
774
775 /* Thai/Lao grapheme break tailoring */
776 static const UChar thTest[] = { 0x0020, 0x0E40, 0x0E01, 0x0020,
777 0x0E01, 0x0E30, 0x0020, 0x0E01, 0x0E33, 0x0020, 0 };
778 /*in Unicode 6.1 en should behave just like th for this*/
779 /*static const int32_t thTestOffs_enFwd[] = { 1, 3, 4, 6, 7, 9, 10 };*/
780 static const int32_t thTestOffs_thFwd[] = { 1, 2, 3, 4, 5, 6, 7, 9, 10 };
781 /*static const int32_t thTestOffs_enRev[] = { 9, 7, 6, 4, 3, 1, 0 };*/
782 static const int32_t thTestOffs_thRev[] = { 9, 7, 6, 5, 4, 3, 2, 1, 0 };
783
784 /* Hebrew line break tailoring, for cldrbug 3028 */
785 static const UChar heTest[] = { 0x0020, 0x002D, 0x0031, 0x0032, 0x0020,
786 0x0061, 0x002D, 0x006B, 0x0020,
787 0x0061, 0x0300, 0x2010, 0x006B, 0x0020,
788 0x05DE, 0x05D4, 0x002D, 0x0069, 0x0020,
789 0x05D1, 0x05BC, 0x2010, 0x0047, 0x0020, 0 };
790 /*in Unicode 6.1 en should behave just like he for this*/
791 /*static const int32_t heTestOffs_enFwd[] = { 1, 5, 7, 9, 12, 14, 17, 19, 22, 24 };*/
792 static const int32_t heTestOffs_heFwd[] = { 1, 5, 7, 9, 12, 14, 19, 24 };
793 /*static const int32_t heTestOffs_enRev[] = { 22, 19, 17, 14, 12, 9, 7, 5, 1, 0 };*/
794 static const int32_t heTestOffs_heRev[] = { 19, 14, 12, 9, 7, 5, 1, 0 };
795
796 /* Finnish line break tailoring, for cldrbug 3029.
797 * As of ICU 63, Finnish tailoring moved to root, Finnish and English should be the same. */
798 static const UChar fiTest[] = { /* 00 */ 0x0020, 0x002D, 0x0031, 0x0032, 0x0020,
799 /* 05 */ 0x0061, 0x002D, 0x006B, 0x0020,
800 /* 09 */ 0x0061, 0x0300, 0x2010, 0x006B, 0x0020,
801 /* 14 */ 0x0061, 0x0020, 0x002D, 0x006B, 0x0020,
802 /* 19 */ 0x0061, 0x0300, 0x0020, 0x2010, 0x006B, 0x0020, 0 };
803 //static const int32_t fiTestOffs_enFwd[] = { 1, 5, 7, 9, 12, 14, 16, 17, 19, 22, 23, 25 };
804 static const int32_t fiTestOffs_enFwd[] = { 1, 5, 7, 9, 12, 14, 16, 19, 22, 25 };
805 static const int32_t fiTestOffs_fiFwd[] = { 1, 5, 7, 9, 12, 14, 16, 19, 22, 25 };
806 //static const int32_t fiTestOffs_enRev[] = { 23, 22, 19, 17, 16, 14, 12, 9, 7, 5, 1, 0 };
807 static const int32_t fiTestOffs_enRev[] = { 22, 19, 16, 14, 12, 9, 7, 5, 1, 0 };
808 static const int32_t fiTestOffs_fiRev[] = { 22, 19, 16, 14, 12, 9, 7, 5, 1, 0 };
809
810 /* Khmer dictionary-based work break, for ICU ticket #8329 */
811 static const UChar kmTest[] = { /* 00 */ 0x179F, 0x17BC, 0x1798, 0x1785, 0x17C6, 0x178E, 0x17B6, 0x1799, 0x1796, 0x17C1,
812 /* 10 */ 0x179B, 0x1794, 0x1793, 0x17D2, 0x178F, 0x17B7, 0x1785, 0x178A, 0x17BE, 0x1798,
813 /* 20 */ 0x17D2, 0x1794, 0x17B8, 0x17A2, 0x1792, 0x17B7, 0x179F, 0x17D2, 0x178B, 0x17B6,
814 /* 30 */ 0x1793, 0x17A2, 0x179A, 0x1796, 0x17D2, 0x179A, 0x17C7, 0x1782, 0x17BB, 0x178E,
815 /* 40 */ 0x178A, 0x179B, 0x17CB, 0x1796, 0x17D2, 0x179A, 0x17C7, 0x17A2, 0x1784, 0x17D2,
816 /* 50 */ 0x1782, 0 };
817 static const int32_t kmTestOffs_kmFwd[] = { 3, /*8,*/ 11, 17, 23, 31, /*33,*/ 40, 43, 51 }; /* TODO: Investigate failure to break at offset 8 */
818 static const int32_t kmTestOffs_kmRev[] = { 43, 40, /*33,*/ 31, 23, 17, 11, /*8,*/ 3, 0 };
819
820 typedef struct {
821 const char * locale;
822 UBreakIteratorType type;
823 const UChar * test;
824 const int32_t * offsFwd;
825 const int32_t * offsRev;
826 int32_t numOffsets;
827 } RBBITailoringTest;
828
829 static const RBBITailoringTest tailoringTests[] = {
830 { "en", UBRK_CHARACTER, thTest, thTestOffs_thFwd, thTestOffs_thRev, UPRV_LENGTHOF(thTestOffs_thFwd) },
831 { "en_US_POSIX", UBRK_CHARACTER, thTest, thTestOffs_thFwd, thTestOffs_thRev, UPRV_LENGTHOF(thTestOffs_thFwd) },
832 { "en", UBRK_LINE, heTest, heTestOffs_heFwd, heTestOffs_heRev, UPRV_LENGTHOF(heTestOffs_heFwd) },
833 { "he", UBRK_LINE, heTest, heTestOffs_heFwd, heTestOffs_heRev, UPRV_LENGTHOF(heTestOffs_heFwd) },
834 { "en", UBRK_LINE, fiTest, fiTestOffs_enFwd, fiTestOffs_enRev, UPRV_LENGTHOF(fiTestOffs_enFwd) },
835 { "fi", UBRK_LINE, fiTest, fiTestOffs_fiFwd, fiTestOffs_fiRev, UPRV_LENGTHOF(fiTestOffs_fiFwd) },
836 { "km", UBRK_WORD, kmTest, kmTestOffs_kmFwd, kmTestOffs_kmRev, UPRV_LENGTHOF(kmTestOffs_kmFwd) },
837 { NULL, 0, NULL, NULL, NULL, 0 },
838 };
839
TestBreakIteratorTailoring(void)840 static void TestBreakIteratorTailoring(void) {
841 const RBBITailoringTest * testPtr;
842 for (testPtr = tailoringTests; testPtr->locale != NULL; ++testPtr) {
843 UErrorCode status = U_ZERO_ERROR;
844 UBreakIterator* ubrkiter = ubrk_open(testPtr->type, testPtr->locale, testPtr->test, -1, &status);
845 if ( U_SUCCESS(status) ) {
846 int32_t offset, offsindx;
847 UBool foundError;
848
849 foundError = FALSE;
850 for (offsindx = 0; (offset = ubrk_next(ubrkiter)) != UBRK_DONE; ++offsindx) {
851 if (!foundError && offsindx >= testPtr->numOffsets) {
852 log_err("FAIL: locale %s, break type %d, ubrk_next expected UBRK_DONE, got %d\n",
853 testPtr->locale, testPtr->type, offset);
854 foundError = TRUE;
855 } else if (!foundError && offset != testPtr->offsFwd[offsindx]) {
856 log_err("FAIL: locale %s, break type %d, ubrk_next expected %d, got %d\n",
857 testPtr->locale, testPtr->type, testPtr->offsFwd[offsindx], offset);
858 foundError = TRUE;
859 }
860 }
861 if (!foundError && offsindx < testPtr->numOffsets) {
862 log_err("FAIL: locale %s, break type %d, ubrk_next expected %d, got UBRK_DONE\n",
863 testPtr->locale, testPtr->type, testPtr->offsFwd[offsindx]);
864 }
865
866 foundError = FALSE;
867 for (offsindx = 0; (offset = ubrk_previous(ubrkiter)) != UBRK_DONE; ++offsindx) {
868 if (!foundError && offsindx >= testPtr->numOffsets) {
869 log_err("FAIL: locale %s, break type %d, ubrk_previous expected UBRK_DONE, got %d\n",
870 testPtr->locale, testPtr->type, offset);
871 foundError = TRUE;
872 } else if (!foundError && offset != testPtr->offsRev[offsindx]) {
873 log_err("FAIL: locale %s, break type %d, ubrk_previous expected %d, got %d\n",
874 testPtr->locale, testPtr->type, testPtr->offsRev[offsindx], offset);
875 foundError = TRUE;
876 }
877 }
878 if (!foundError && offsindx < testPtr->numOffsets) {
879 log_err("FAIL: locale %s, break type %d, ubrk_previous expected %d, got UBRK_DONE\n",
880 testPtr->locale, testPtr->type, testPtr->offsRev[offsindx]);
881 }
882
883 ubrk_close(ubrkiter);
884 } else {
885 log_err_status(status, "FAIL: locale %s, break type %d, ubrk_open status: %s\n", testPtr->locale, testPtr->type, u_errorName(status));
886 }
887 }
888 }
889
890
TestBreakIteratorRefresh(void)891 static void TestBreakIteratorRefresh(void) {
892 /*
893 * RefreshInput changes out the input of a Break Iterator without
894 * changing anything else in the iterator's state. Used with Java JNI,
895 * when Java moves the underlying string storage. This test
896 * runs a ubrk_next() repeatedly, moving the text in the middle of the sequence.
897 * The right set of boundaries should still be found.
898 */
899 UChar testStr[] = {0x20, 0x41, 0x20, 0x42, 0x20, 0x43, 0x20, 0x44, 0x0}; /* = " A B C D" */
900 UChar movedStr[] = {0x20, 0x20, 0x20, 0x20, 0x20, 0x20, 0x20, 0x20, 0};
901 UErrorCode status = U_ZERO_ERROR;
902 UBreakIterator *bi;
903 UText ut1 = UTEXT_INITIALIZER;
904 UText ut2 = UTEXT_INITIALIZER;
905
906 bi = ubrk_open(UBRK_LINE, "en_US", NULL, 0, &status);
907 TEST_ASSERT_SUCCESS(status);
908 if (U_FAILURE(status)) {
909 return;
910 }
911
912 utext_openUChars(&ut1, testStr, -1, &status);
913 TEST_ASSERT_SUCCESS(status);
914 ubrk_setUText(bi, &ut1, &status);
915 TEST_ASSERT_SUCCESS(status);
916
917 if (U_SUCCESS(status)) {
918 /* Line boundaries will occur before each letter in the original string */
919 TEST_ASSERT(1 == ubrk_next(bi));
920 TEST_ASSERT(3 == ubrk_next(bi));
921
922 /* Move the string, kill the original string. */
923 u_strcpy(movedStr, testStr);
924 u_memset(testStr, 0x20, u_strlen(testStr));
925 utext_openUChars(&ut2, movedStr, -1, &status);
926 TEST_ASSERT_SUCCESS(status);
927 ubrk_refreshUText(bi, &ut2, &status);
928 TEST_ASSERT_SUCCESS(status);
929
930 /* Find the following matches, now working in the moved string. */
931 TEST_ASSERT(5 == ubrk_next(bi));
932 TEST_ASSERT(7 == ubrk_next(bi));
933 TEST_ASSERT(8 == ubrk_next(bi));
934 TEST_ASSERT(UBRK_DONE == ubrk_next(bi));
935 TEST_ASSERT_SUCCESS(status);
936
937 utext_close(&ut1);
938 utext_close(&ut2);
939 }
940 ubrk_close(bi);
941 }
942
943
TestBug11665(void)944 static void TestBug11665(void) {
945 // The problem was with the incorrect breaking of Japanese text beginning
946 // with Katakana characters when no prior Japanese or Chinese text had been
947 // encountered.
948 //
949 // Tested here in cintltst, rather than in intltest, because only cintltst
950 // tests have the ability to reset ICU, which is needed to get the bug
951 // to manifest itself.
952
953 static UChar japaneseText[] = {0x30A2, 0x30EC, 0x30EB, 0x30AE, 0x30FC, 0x6027, 0x7D50, 0x819C, 0x708E};
954 int32_t boundaries[10] = {0};
955 UBreakIterator *bi = NULL;
956 int32_t brk;
957 int32_t brkIdx = 0;
958 int32_t totalBreaks = 0;
959 UErrorCode status = U_ZERO_ERROR;
960
961 ctest_resetICU();
962 bi = ubrk_open(UBRK_WORD, "en_US", japaneseText, UPRV_LENGTHOF(japaneseText), &status);
963 TEST_ASSERT_SUCCESS(status);
964 if (!bi) {
965 return;
966 }
967 for (brk=ubrk_first(bi); brk != UBRK_DONE; brk=ubrk_next(bi)) {
968 boundaries[brkIdx] = brk;
969 if (++brkIdx >= UPRV_LENGTHOF(boundaries) - 1) {
970 break;
971 }
972 }
973 if (brkIdx <= 2 || brkIdx >= UPRV_LENGTHOF(boundaries)) {
974 log_err("%s:%d too few or many breaks found.\n", __FILE__, __LINE__);
975 } else {
976 totalBreaks = brkIdx;
977 brkIdx = 0;
978 for (brk=ubrk_first(bi); brk != UBRK_DONE; brk=ubrk_next(bi)) {
979 if (brk != boundaries[brkIdx]) {
980 log_err("%s:%d Break #%d differs between first and second iteration.\n", __FILE__, __LINE__, brkIdx);
981 break;
982 }
983 if (++brkIdx >= UPRV_LENGTHOF(boundaries) - 1) {
984 log_err("%s:%d Too many breaks.\n", __FILE__, __LINE__);
985 break;
986 }
987 }
988 if (totalBreaks != brkIdx) {
989 log_err("%s:%d Number of breaks differ between first and second iteration.\n", __FILE__, __LINE__);
990 }
991 }
992 ubrk_close(bi);
993 }
994
995 /*
996 * expOffset is the set of expected offsets, ending with '-1'.
997 * "Expected expOffset -1" means "expected the end of the offsets"
998 */
999
1000 static const char testSentenceSuppressionsEn[] = "Mr. Jones comes home. Dr. Smith Ph.D. is out. In the U.S.A. it is hot.";
1001 static const int32_t testSentSuppFwdOffsetsEn[] = { 22, 26, 46, 70, -1 }; /* With suppressions, currently not handling Dr. */
1002 static const int32_t testSentFwdOffsetsEn[] = { 4, 22, 26, 46, 70, -1 }; /* Without suppressions */
1003 static const int32_t testSentSuppRevOffsetsEn[] = { 46, 26, 22, 0, -1 }; /* With suppressions, currently not handling Dr. */
1004 static const int32_t testSentRevOffsetsEn[] = { 46, 26, 22, 4, 0, -1 }; /* Without suppressions */
1005
1006 static const char testSentenceSuppressionsDe[] = "Wenn ich schon h\\u00F6re zu Guttenberg kommt evtl. zur\\u00FCck.";
1007 static const int32_t testSentSuppFwdOffsetsDe[] = { 53, -1 }; /* With suppressions */
1008 static const int32_t testSentFwdOffsetsDe[] = { 53, -1 }; /* Without suppressions; no break in evtl. zur due to casing */
1009 static const int32_t testSentSuppRevOffsetsDe[] = { 0, -1 }; /* With suppressions */
1010 static const int32_t testSentRevOffsetsDe[] = { 0, -1 }; /* Without suppressions */
1011
1012 static const char testSentenceSuppressionsEs[] = "Te esperamos todos los miercoles en Bravo 416, Col. El Pueblo a las 7 PM.";
1013 static const int32_t testSentSuppFwdOffsetsEs[] = { 73, -1 }; /* With suppressions */
1014 static const int32_t testSentFwdOffsetsEs[] = { 52, 73, -1 }; /* Without suppressions */
1015 static const int32_t testSentSuppRevOffsetsEs[] = { 0, -1 }; /* With suppressions */
1016 static const int32_t testSentRevOffsetsEs[] = { 52, 0, -1 }; /* Without suppressions */
1017
1018 enum { kTextULenMax = 128 };
1019
1020 typedef struct {
1021 const char * locale;
1022 const char * text;
1023 const int32_t * expFwdOffsets;
1024 const int32_t * expRevOffsets;
1025 } TestBISuppressionsItem;
1026
1027 static const TestBISuppressionsItem testBISuppressionsItems[] = {
1028 { "en@ss=standard", testSentenceSuppressionsEn, testSentSuppFwdOffsetsEn, testSentSuppRevOffsetsEn },
1029 { "en", testSentenceSuppressionsEn, testSentFwdOffsetsEn, testSentRevOffsetsEn },
1030 { "en_CA", testSentenceSuppressionsEn, testSentFwdOffsetsEn, testSentRevOffsetsEn },
1031 { "en_CA@ss=standard", testSentenceSuppressionsEn, testSentSuppFwdOffsetsEn, testSentSuppRevOffsetsEn },
1032 { "fr@ss=standard", testSentenceSuppressionsEn, testSentFwdOffsetsEn, testSentRevOffsetsEn },
1033 { "af@ss=standard", testSentenceSuppressionsEn, testSentFwdOffsetsEn, testSentRevOffsetsEn }, /* no brkiter data => nosuppressions? */
1034 { "af_ZA@ss=standard", testSentenceSuppressionsEn, testSentFwdOffsetsEn, testSentRevOffsetsEn }, /* no brkiter data => nosuppressions? */
1035 { "zh@ss=standard", testSentenceSuppressionsEn, testSentFwdOffsetsEn, testSentRevOffsetsEn }, /* brkiter data, no suppressions data => no suppressions */
1036 { "zh_Hant@ss=standard", testSentenceSuppressionsEn, testSentFwdOffsetsEn, testSentRevOffsetsEn }, /* brkiter data, no suppressions data => no suppressions */
1037 { "fi@ss=standard", testSentenceSuppressionsEn, testSentFwdOffsetsEn, testSentRevOffsetsEn }, /* brkiter data, no suppressions data => no suppressions */
1038 { "ja@ss=standard", testSentenceSuppressionsEn, testSentFwdOffsetsEn, testSentRevOffsetsEn }, /* brkiter data, no suppressions data => no suppressions */
1039 { "de@ss=standard", testSentenceSuppressionsDe, testSentSuppFwdOffsetsDe, testSentSuppRevOffsetsDe },
1040 { "de", testSentenceSuppressionsDe, testSentFwdOffsetsDe, testSentRevOffsetsDe },
1041 { "es@ss=standard", testSentenceSuppressionsEs, testSentSuppFwdOffsetsEs, testSentSuppRevOffsetsEs },
1042 { "es", testSentenceSuppressionsEs, testSentFwdOffsetsEs, testSentRevOffsetsEs },
1043 { NULL, NULL, NULL, NULL }
1044 };
1045
TestBreakIteratorSuppressions(void)1046 static void TestBreakIteratorSuppressions(void) {
1047 const TestBISuppressionsItem * itemPtr;
1048
1049 for (itemPtr = testBISuppressionsItems; itemPtr->locale != NULL; itemPtr++) {
1050 UChar textU[kTextULenMax];
1051 int32_t textULen = u_unescape(itemPtr->text, textU, kTextULenMax);
1052 UErrorCode status = U_ZERO_ERROR;
1053 UBreakIterator *bi = ubrk_open(UBRK_SENTENCE, itemPtr->locale, textU, textULen, &status);
1054 log_verbose("#%d: %s\n", (itemPtr-testBISuppressionsItems), itemPtr->locale);
1055 if (U_SUCCESS(status)) {
1056 int32_t offset, start;
1057 const int32_t * expOffsetPtr;
1058 const int32_t * expOffsetStart;
1059
1060 expOffsetStart = expOffsetPtr = itemPtr->expFwdOffsets;
1061 ubrk_first(bi);
1062 for (; (offset = ubrk_next(bi)) != UBRK_DONE && *expOffsetPtr >= 0; expOffsetPtr++) {
1063 if (offset != *expOffsetPtr) {
1064 log_err("FAIL: ubrk_next loc \"%s\", expected %d, got %d\n", itemPtr->locale, *expOffsetPtr, offset);
1065 }
1066 }
1067 if (offset != UBRK_DONE || *expOffsetPtr >= 0) {
1068 log_err("FAIL: ubrk_next loc \"%s\", expected UBRK_DONE & expOffset -1, got %d and %d\n", itemPtr->locale, offset, *expOffsetPtr);
1069 }
1070
1071 expOffsetStart = expOffsetPtr = itemPtr->expFwdOffsets;
1072 start = ubrk_first(bi) + 1;
1073 for (; (offset = ubrk_following(bi, start)) != UBRK_DONE && *expOffsetPtr >= 0; expOffsetPtr++) {
1074 if (offset != *expOffsetPtr) {
1075 log_err("FAIL: ubrk_following(%d) loc \"%s\", expected %d, got %d\n", start, itemPtr->locale, *expOffsetPtr, offset);
1076 }
1077 start = *expOffsetPtr + 1;
1078 }
1079 if (offset != UBRK_DONE || *expOffsetPtr >= 0) {
1080 log_err("FAIL: ubrk_following(%d) loc \"%s\", expected UBRK_DONE & expOffset -1, got %d and %d\n", start, itemPtr->locale, offset, *expOffsetPtr);
1081 }
1082
1083 expOffsetStart = expOffsetPtr = itemPtr->expRevOffsets;
1084 offset = ubrk_last(bi);
1085 log_verbose("___ @%d ubrk_last\n", offset);
1086 if(offset == 0) {
1087 log_err("FAIL: ubrk_last loc \"%s\" unexpected %d\n", itemPtr->locale, offset);
1088 }
1089 for (; (offset = ubrk_previous(bi)) != UBRK_DONE && *expOffsetPtr >= 0; expOffsetPtr++) {
1090 if (offset != *expOffsetPtr) {
1091 log_err("FAIL: ubrk_previous loc \"%s\", expected %d, got %d\n", itemPtr->locale, *expOffsetPtr, offset);
1092 } else {
1093 log_verbose("[%d] @%d ubrk_previous()\n", (expOffsetPtr - expOffsetStart), offset);
1094 }
1095 }
1096 if (offset != UBRK_DONE || *expOffsetPtr >= 0) {
1097 log_err("FAIL: ubrk_previous loc \"%s\", expected UBRK_DONE & expOffset[%d] -1, got %d and %d\n", itemPtr->locale,
1098 expOffsetPtr - expOffsetStart,
1099 offset, *expOffsetPtr);
1100 }
1101
1102 expOffsetStart = expOffsetPtr = itemPtr->expRevOffsets;
1103 start = ubrk_last(bi) - 1;
1104 for (; (offset = ubrk_preceding(bi, start)) != UBRK_DONE && *expOffsetPtr >= 0; expOffsetPtr++) {
1105 if (offset != *expOffsetPtr) {
1106 log_err("FAIL: ubrk_preceding(%d) loc \"%s\", expected %d, got %d\n", start, itemPtr->locale, *expOffsetPtr, offset);
1107 }
1108 start = *expOffsetPtr - 1;
1109 }
1110 if (start >=0 && (offset != UBRK_DONE || *expOffsetPtr >= 0)) {
1111 log_err("FAIL: ubrk_preceding loc(%d) \"%s\", expected UBRK_DONE & expOffset -1, got %d and %d\n", start, itemPtr->locale, offset, *expOffsetPtr);
1112 }
1113
1114 ubrk_close(bi);
1115 } else {
1116 log_data_err("FAIL: ubrk_open(UBRK_SENTENCE, \"%s\", ...) status %s (Are you missing data?)\n", itemPtr->locale, u_errorName(status));
1117 }
1118 }
1119 }
1120
1121
1122 #endif /* #if !UCONFIG_NO_BREAK_ITERATION */
1123