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 static void TestBreakIteratorClone(void);
53 #endif
54 static void TestBreakIteratorRules(void);
55 static void TestBreakIteratorRuleError(void);
56 static void TestBreakIteratorStatusVec(void);
57 static void TestBreakIteratorUText(void);
58 static void TestBreakIteratorTailoring(void);
59 static void TestBreakIteratorRefresh(void);
60 static void TestBug11665(void);
61 static void TestBreakIteratorSuppressions(void);
62
63 void addBrkIterAPITest(TestNode** root);
64
addBrkIterAPITest(TestNode ** root)65 void addBrkIterAPITest(TestNode** root)
66 {
67 #if !UCONFIG_NO_FILE_IO
68 addTest(root, &TestBreakIteratorCAPI, "tstxtbd/cbiapts/TestBreakIteratorCAPI");
69 addTest(root, &TestBreakIteratorSafeClone, "tstxtbd/cbiapts/TestBreakIteratorSafeClone");
70 addTest(root, &TestBreakIteratorClone, "tstxtbd/cbiapts/TestBreakIteratorClone");
71 addTest(root, &TestBreakIteratorUText, "tstxtbd/cbiapts/TestBreakIteratorUText");
72 #endif
73 addTest(root, &TestBreakIteratorRules, "tstxtbd/cbiapts/TestBreakIteratorRules");
74 addTest(root, &TestBreakIteratorRuleError, "tstxtbd/cbiapts/TestBreakIteratorRuleError");
75 addTest(root, &TestBreakIteratorStatusVec, "tstxtbd/cbiapts/TestBreakIteratorStatusVec");
76 addTest(root, &TestBreakIteratorTailoring, "tstxtbd/cbiapts/TestBreakIteratorTailoring");
77 addTest(root, &TestBreakIteratorRefresh, "tstxtbd/cbiapts/TestBreakIteratorRefresh");
78 addTest(root, &TestBug11665, "tstxtbd/cbiapts/TestBug11665");
79 #if !UCONFIG_NO_FILTERED_BREAK_ITERATION
80 addTest(root, &TestBreakIteratorSuppressions, "tstxtbd/cbiapts/TestBreakIteratorSuppressions");
81 #endif
82 }
83
84 #define CLONETEST_ITERATOR_COUNT 2
85
86 /*
87 * Utility function for converting char * to UChar * strings, to
88 * simplify the test code. Converted strings are put in heap allocated
89 * storage. A hook (probably a local in the caller's code) allows all
90 * strings converted with that hook to be freed with a single call.
91 */
92 typedef struct StringStruct {
93 struct StringStruct *link;
94 UChar str[1];
95 } StringStruct;
96
97
toUChar(const char * src,void ** freeHook)98 static UChar* toUChar(const char *src, void **freeHook) {
99 /* Structure of the memory that we allocate on the heap */
100
101 int32_t numUChars;
102 int32_t destSize;
103 UChar stackBuf[2000 + sizeof(void *)/sizeof(UChar)];
104 StringStruct *dest;
105 UConverter *cnv;
106
107 UErrorCode status = U_ZERO_ERROR;
108 if (src == NULL) {
109 return NULL;
110 }
111
112 cnv = ucnv_open(NULL, &status);
113 if(U_FAILURE(status) || cnv == NULL) {
114 return NULL;
115 }
116 ucnv_reset(cnv);
117 numUChars = ucnv_toUChars(cnv,
118 stackBuf,
119 2000,
120 src, -1,
121 &status);
122
123 destSize = (numUChars+1) * sizeof(UChar) + sizeof(struct StringStruct);
124 dest = (StringStruct *)malloc(destSize);
125 if (dest != NULL) {
126 if (status == U_BUFFER_OVERFLOW_ERROR || status == U_STRING_NOT_TERMINATED_WARNING) {
127 ucnv_toUChars(cnv, dest->str, numUChars+1, src, -1, &status);
128 } else if (status == U_ZERO_ERROR) {
129 u_strcpy(dest->str, stackBuf);
130 } else {
131 free(dest);
132 dest = NULL;
133 }
134 }
135
136 ucnv_reset(cnv); /* be good citizens */
137 ucnv_close(cnv);
138 if (dest == NULL) {
139 return NULL;
140 }
141
142 dest->link = (StringStruct*)(*freeHook);
143 *freeHook = dest;
144 return dest->str;
145 }
146
freeToUCharStrings(void ** hook)147 static void freeToUCharStrings(void **hook) {
148 StringStruct *s = *(StringStruct **)hook;
149 while (s != NULL) {
150 StringStruct *next = s->link;
151 free(s);
152 s = next;
153 }
154 }
155
156
157 #if !UCONFIG_NO_FILE_IO
TestBreakIteratorCAPI()158 static void TestBreakIteratorCAPI()
159 {
160 UErrorCode status = U_ZERO_ERROR;
161 UBreakIterator *word, *sentence, *line, *character, *b, *bogus;
162 int32_t start,pos,end,to;
163 int32_t i;
164 int32_t count = 0;
165
166 UChar text[50];
167
168 /* Note: the adjacent "" are concatenating strings, not adding a \" to the
169 string, which is probably what whoever wrote this intended. Don't fix,
170 because it would throw off the hard coded break positions in the following
171 tests. */
172 u_uastrcpy(text, "He's from Africa. ""Mr. Livingston, I presume?"" Yeah");
173
174
175 /*test ubrk_open()*/
176 log_verbose("\nTesting BreakIterator open functions\n");
177
178 /* Use french for fun */
179 word = ubrk_open(UBRK_WORD, "en_US", text, u_strlen(text), &status);
180 if(status == U_FILE_ACCESS_ERROR) {
181 log_data_err("Check your data - it doesn't seem to be around\n");
182 return;
183 } else if(U_FAILURE(status)){
184 log_err_status(status, "FAIL: Error in ubrk_open() for word breakiterator: %s\n", myErrorName(status));
185 }
186 else{
187 log_verbose("PASS: Successfully opened word breakiterator\n");
188 }
189
190 sentence = ubrk_open(UBRK_SENTENCE, "en_US", text, u_strlen(text), &status);
191 if(U_FAILURE(status)){
192 log_err_status(status, "FAIL: Error in ubrk_open() for sentence breakiterator: %s\n", myErrorName(status));
193 return;
194 }
195 else{
196 log_verbose("PASS: Successfully opened sentence breakiterator\n");
197 }
198
199 line = ubrk_open(UBRK_LINE, "en_US", text, u_strlen(text), &status);
200 if(U_FAILURE(status)){
201 log_err("FAIL: Error in ubrk_open() for line breakiterator: %s\n", myErrorName(status));
202 return;
203 }
204 else{
205 log_verbose("PASS: Successfully opened line breakiterator\n");
206 }
207
208 character = ubrk_open(UBRK_CHARACTER, "en_US", text, u_strlen(text), &status);
209 if(U_FAILURE(status)){
210 log_err("FAIL: Error in ubrk_open() for character breakiterator: %s\n", myErrorName(status));
211 return;
212 }
213 else{
214 log_verbose("PASS: Successfully opened character breakiterator\n");
215 }
216 /*trying to open an illegal iterator*/
217 bogus = ubrk_open((UBreakIteratorType)5, "en_US", text, u_strlen(text), &status);
218 if(bogus != NULL) {
219 log_err("FAIL: expected NULL from opening an invalid break iterator.\n");
220 }
221 if(U_SUCCESS(status)){
222 log_err("FAIL: Error in ubrk_open() for BOGUS breakiterator. Expected U_ILLEGAL_ARGUMENT_ERROR\n");
223 }
224 if(U_FAILURE(status)){
225 if(status != U_ILLEGAL_ARGUMENT_ERROR){
226 log_err("FAIL: Error in ubrk_open() for BOGUS breakiterator. Expected U_ILLEGAL_ARGUMENT_ERROR\n Got %s\n", myErrorName(status));
227 }
228 }
229 status=U_ZERO_ERROR;
230
231
232 /* ======= Test ubrk_countAvialable() and ubrk_getAvialable() */
233
234 log_verbose("\nTesting ubrk_countAvailable() and ubrk_getAvailable()\n");
235 count=ubrk_countAvailable();
236 /* use something sensible w/o hardcoding the count */
237 if(count < 0){
238 log_err("FAIL: Error in ubrk_countAvialable() returned %d\n", count);
239 }
240 else{
241 log_verbose("PASS: ubrk_countAvialable() successful returned %d\n", count);
242 }
243 for(i=0;i<count;i++)
244 {
245 log_verbose("%s\n", ubrk_getAvailable(i));
246 if (ubrk_getAvailable(i) == 0)
247 log_err("No locale for which breakiterator is applicable\n");
248 else
249 log_verbose("A locale %s for which breakiterator is applicable\n",ubrk_getAvailable(i));
250 }
251
252 /*========Test ubrk_first(), ubrk_last()...... and other functions*/
253
254 log_verbose("\nTesting the functions for word\n");
255 start = ubrk_first(word);
256 if(start!=0)
257 log_err("error ubrk_start(word) did not return 0\n");
258 log_verbose("first (word = %d\n", (int32_t)start);
259 pos=ubrk_next(word);
260 if(pos!=4)
261 log_err("error ubrk_next(word) did not return 4\n");
262 log_verbose("next (word = %d\n", (int32_t)pos);
263 pos=ubrk_following(word, 4);
264 if(pos!=5)
265 log_err("error ubrl_following(word,4) did not return 6\n");
266 log_verbose("next (word = %d\n", (int32_t)pos);
267 end=ubrk_last(word);
268 if(end!=49)
269 log_err("error ubrk_last(word) did not return 49\n");
270 log_verbose("last (word = %d\n", (int32_t)end);
271
272 pos=ubrk_previous(word);
273 log_verbose("%d %d\n", end, pos);
274
275 pos=ubrk_previous(word);
276 log_verbose("%d \n", pos);
277
278 if (ubrk_isBoundary(word, 2) != FALSE) {
279 log_err("error ubrk_isBoundary(word, 2) did not return FALSE\n");
280 }
281 pos=ubrk_current(word);
282 if (pos != 4) {
283 log_err("error ubrk_current() != 4 after ubrk_isBoundary(word, 2)\n");
284 }
285 if (ubrk_isBoundary(word, 4) != TRUE) {
286 log_err("error ubrk_isBoundary(word, 4) did not return TRUE\n");
287 }
288
289
290
291 log_verbose("\nTesting the functions for character\n");
292 ubrk_first(character);
293 pos = ubrk_following(character, 5);
294 if(pos!=6)
295 log_err("error ubrk_following(character,5) did not return 6\n");
296 log_verbose("Following (character,5) = %d\n", (int32_t)pos);
297 pos=ubrk_following(character, 18);
298 if(pos!=19)
299 log_err("error ubrk_following(character,18) did not return 19\n");
300 log_verbose("Followingcharacter,18) = %d\n", (int32_t)pos);
301 pos=ubrk_preceding(character, 22);
302 if(pos!=21)
303 log_err("error ubrk_preceding(character,22) did not return 21\n");
304 log_verbose("preceding(character,22) = %d\n", (int32_t)pos);
305
306
307 log_verbose("\nTesting the functions for line\n");
308 pos=ubrk_first(line);
309 if(pos != 0)
310 log_err("error ubrk_first(line) returned %d, expected 0\n", (int32_t)pos);
311 pos = ubrk_next(line);
312 pos=ubrk_following(line, 18);
313 if(pos!=22)
314 log_err("error ubrk_following(line) did not return 22\n");
315 log_verbose("following (line) = %d\n", (int32_t)pos);
316
317
318 log_verbose("\nTesting the functions for sentence\n");
319 pos = ubrk_first(sentence);
320 pos = ubrk_current(sentence);
321 log_verbose("Current(sentence) = %d\n", (int32_t)pos);
322 pos = ubrk_last(sentence);
323 if(pos!=49)
324 log_err("error ubrk_last for sentence did not return 49\n");
325 log_verbose("Last (sentence) = %d\n", (int32_t)pos);
326 pos = ubrk_first(sentence);
327 to = ubrk_following( sentence, 0 );
328 if (to == 0) log_err("ubrk_following returned 0\n");
329 to = ubrk_preceding( sentence, to );
330 if (to != 0) log_err("ubrk_preceding didn't return 0\n");
331 if (ubrk_first(sentence)!=ubrk_current(sentence)) {
332 log_err("error in ubrk_first() or ubrk_current()\n");
333 }
334
335
336 /*---- */
337 /*Testing ubrk_open and ubrk_close()*/
338 log_verbose("\nTesting open and close for us locale\n");
339 b = ubrk_open(UBRK_WORD, "fr_FR", text, u_strlen(text), &status);
340 if (U_FAILURE(status)) {
341 log_err("ubrk_open for word returned NULL: %s\n", myErrorName(status));
342 }
343 ubrk_close(b);
344
345 /* Test setText and setUText */
346 {
347 UChar s1[] = {0x41, 0x42, 0x20, 0};
348 UChar s2[] = {0x41, 0x42, 0x43, 0x44, 0x45, 0};
349 UText *ut = NULL;
350 UBreakIterator *bb;
351 int j;
352
353 log_verbose("\nTesting ubrk_setText() and ubrk_setUText()\n");
354 status = U_ZERO_ERROR;
355 bb = ubrk_open(UBRK_WORD, "en_US", NULL, 0, &status);
356 TEST_ASSERT_SUCCESS(status);
357 ubrk_setText(bb, s1, -1, &status);
358 TEST_ASSERT_SUCCESS(status);
359 ubrk_first(bb);
360 j = ubrk_next(bb);
361 TEST_ASSERT(j == 2);
362 ut = utext_openUChars(ut, s2, -1, &status);
363 ubrk_setUText(bb, ut, &status);
364 TEST_ASSERT_SUCCESS(status);
365 j = ubrk_next(bb);
366 TEST_ASSERT(j == 5);
367
368 ubrk_close(bb);
369 utext_close(ut);
370 }
371
372 ubrk_close(word);
373 ubrk_close(sentence);
374 ubrk_close(line);
375 ubrk_close(character);
376 }
377
TestBreakIteratorSafeClone(void)378 static void TestBreakIteratorSafeClone(void)
379 {
380 UChar text[51]; /* Keep this odd to test for 64-bit memory alignment */
381 /* NOTE: This doesn't reliably force mis-alignment of following items. */
382 uint8_t buffer [CLONETEST_ITERATOR_COUNT] [U_BRK_SAFECLONE_BUFFERSIZE];
383 int32_t bufferSize = U_BRK_SAFECLONE_BUFFERSIZE;
384
385 UBreakIterator * someIterators [CLONETEST_ITERATOR_COUNT];
386 UBreakIterator * someClonedIterators [CLONETEST_ITERATOR_COUNT];
387
388 UBreakIterator * brk;
389 UErrorCode status = U_ZERO_ERROR;
390 int32_t start,pos;
391 int32_t i;
392
393 /*Testing ubrk_safeClone */
394
395 /* Note: the adjacent "" are concatenating strings, not adding a \" to the
396 string, which is probably what whoever wrote this intended. Don't fix,
397 because it would throw off the hard coded break positions in the following
398 tests. */
399 u_uastrcpy(text, "He's from Africa. ""Mr. Livingston, I presume?"" Yeah");
400
401 /* US & Thai - rule-based & dictionary based */
402 someIterators[0] = ubrk_open(UBRK_WORD, "en_US", text, u_strlen(text), &status);
403 if(!someIterators[0] || U_FAILURE(status)) {
404 log_data_err("Couldn't open en_US word break iterator - %s\n", u_errorName(status));
405 return;
406 }
407
408 someIterators[1] = ubrk_open(UBRK_WORD, "th_TH", text, u_strlen(text), &status);
409 if(!someIterators[1] || U_FAILURE(status)) {
410 log_data_err("Couldn't open th_TH word break iterator - %s\n", u_errorName(status));
411 return;
412 }
413
414 /* test each type of iterator */
415 for (i = 0; i < CLONETEST_ITERATOR_COUNT; i++)
416 {
417
418 /* Check the various error & informational states */
419
420 /* Null status - just returns NULL */
421 if (NULL != ubrk_safeClone(someIterators[i], buffer[i], &bufferSize, NULL))
422 {
423 log_err("FAIL: Cloned Iterator failed to deal correctly with null status\n");
424 }
425 /* error status - should return 0 & keep error the same */
426 status = U_MEMORY_ALLOCATION_ERROR;
427 if (NULL != ubrk_safeClone(someIterators[i], buffer[i], &bufferSize, &status) || status != U_MEMORY_ALLOCATION_ERROR)
428 {
429 log_err("FAIL: Cloned Iterator failed to deal correctly with incoming error status\n");
430 }
431 status = U_ZERO_ERROR;
432
433 /* Null buffer size pointer is ok */
434 if (NULL == (brk = ubrk_safeClone(someIterators[i], buffer[i], NULL, &status)) || U_FAILURE(status))
435 {
436 log_err("FAIL: Cloned Iterator failed to deal correctly with null bufferSize pointer\n");
437 }
438 ubrk_close(brk);
439 status = U_ZERO_ERROR;
440
441 /* buffer size pointer is 0 - fill in pbufferSize with a size */
442 bufferSize = 0;
443 if (NULL != ubrk_safeClone(someIterators[i], buffer[i], &bufferSize, &status) ||
444 U_FAILURE(status) || bufferSize <= 0)
445 {
446 log_err("FAIL: Cloned Iterator failed a sizing request ('preflighting')\n");
447 }
448 /* Verify our define is large enough */
449 if (U_BRK_SAFECLONE_BUFFERSIZE < bufferSize)
450 {
451 log_err("FAIL: Pre-calculated buffer size is too small - %d but needed %d\n", U_BRK_SAFECLONE_BUFFERSIZE, bufferSize);
452 }
453 /* Verify we can use this run-time calculated size */
454 if (NULL == (brk = ubrk_safeClone(someIterators[i], buffer[i], &bufferSize, &status)) || U_FAILURE(status))
455 {
456 log_err("FAIL: Iterator can't be cloned with run-time size\n");
457 }
458 if (brk)
459 ubrk_close(brk);
460 /* size one byte too small - should allocate & let us know */
461 if (bufferSize > 1) {
462 --bufferSize;
463 }
464 if (NULL == (brk = ubrk_safeClone(someIterators[i], NULL, &bufferSize, &status)) || status != U_SAFECLONE_ALLOCATED_WARNING)
465 {
466 log_err("FAIL: Cloned Iterator failed to deal correctly with too-small buffer size\n");
467 }
468 if (brk)
469 ubrk_close(brk);
470 status = U_ZERO_ERROR;
471 bufferSize = U_BRK_SAFECLONE_BUFFERSIZE;
472
473 /* Null buffer pointer - return Iterator & set error to U_SAFECLONE_ALLOCATED_ERROR */
474 if (NULL == (brk = ubrk_safeClone(someIterators[i], NULL, &bufferSize, &status)) || status != U_SAFECLONE_ALLOCATED_WARNING)
475 {
476 log_err("FAIL: Cloned Iterator failed to deal correctly with null buffer pointer\n");
477 }
478 if (brk)
479 ubrk_close(brk);
480 status = U_ZERO_ERROR;
481
482 /* Mis-aligned buffer pointer. */
483 {
484 char stackBuf[U_BRK_SAFECLONE_BUFFERSIZE+sizeof(void *)];
485
486 brk = ubrk_safeClone(someIterators[i], &stackBuf[1], &bufferSize, &status);
487 if (U_FAILURE(status) || brk == NULL) {
488 log_err("FAIL: Cloned Iterator failed with misaligned buffer pointer\n");
489 }
490 if (status == U_SAFECLONE_ALLOCATED_WARNING) {
491 log_verbose("Cloned Iterator allocated when using a mis-aligned buffer.\n");
492 }
493 if (brk)
494 ubrk_close(brk);
495 }
496
497
498 /* Null Iterator - return NULL & set U_ILLEGAL_ARGUMENT_ERROR */
499 if (NULL != ubrk_safeClone(NULL, buffer[i], &bufferSize, &status) || status != U_ILLEGAL_ARGUMENT_ERROR)
500 {
501 log_err("FAIL: Cloned Iterator failed to deal correctly with null Iterator pointer\n");
502 }
503 status = U_ZERO_ERROR;
504
505 /* Do these cloned Iterators work at all - make a first & next call */
506 bufferSize = U_BRK_SAFECLONE_BUFFERSIZE;
507 someClonedIterators[i] = ubrk_safeClone(someIterators[i], buffer[i], &bufferSize, &status);
508
509 start = ubrk_first(someClonedIterators[i]);
510 if(start!=0)
511 log_err("error ubrk_start(clone) did not return 0\n");
512 pos=ubrk_next(someClonedIterators[i]);
513 if(pos!=4)
514 log_err("error ubrk_next(clone) did not return 4\n");
515
516 ubrk_close(someClonedIterators[i]);
517 ubrk_close(someIterators[i]);
518 }
519 }
520
TestBreakIteratorClone(void)521 static void TestBreakIteratorClone(void)
522 {
523 const UChar text[] = u"He's from Africa. Mr. Livingston, I presume? Yeah";
524 UBreakIterator * someIterators [CLONETEST_ITERATOR_COUNT];
525
526 UBreakIterator * brk;
527 UErrorCode status = U_ZERO_ERROR;
528 int32_t start,pos;
529 int32_t i;
530
531 /*Testing ubrk_clone */
532
533 /* US & Thai - rule-based & dictionary based */
534 someIterators[0] = ubrk_open(UBRK_WORD, "en_US", text, u_strlen(text), &status);
535 if(!someIterators[0] || U_FAILURE(status)) {
536 log_data_err("Couldn't open en_US word break iterator - %s\n", u_errorName(status));
537 return;
538 }
539
540 someIterators[1] = ubrk_open(UBRK_WORD, "th_TH", text, u_strlen(text), &status);
541 if(!someIterators[1] || U_FAILURE(status)) {
542 log_data_err("Couldn't open th_TH word break iterator - %s\n", u_errorName(status));
543 return;
544 }
545
546 /* test each type of iterator */
547 for (i = 0; i < CLONETEST_ITERATOR_COUNT; i++)
548 {
549 /* error status - should return 0 & keep error the same */
550 status = U_MEMORY_ALLOCATION_ERROR;
551 if (NULL != ubrk_clone(someIterators[i], &status) || status != U_MEMORY_ALLOCATION_ERROR)
552 {
553 log_err("FAIL: Cloned Iterator failed to deal correctly with incoming error status\n");
554 }
555
556 status = U_ZERO_ERROR;
557
558 /* Do these cloned Iterators work at all - make a first & next call */
559 brk = ubrk_clone(someIterators[i], &status);
560
561 start = ubrk_first(brk);
562 if(start!=0)
563 log_err("error ubrk_start(clone) did not return 0, but %i\n", start);
564 pos=ubrk_next(brk);
565 if(pos!=4)
566 log_err("error ubrk_next(clone) did not return 4, but %i\n", pos);
567
568 ubrk_close(brk);
569
570 pos = ubrk_next(someIterators[i]);
571 if (pos != 4) {
572 log_err("error ubrk_next(iter) did not return 4, but %i\n", pos);
573 }
574
575 brk = ubrk_clone(someIterators[i], &status);
576 // The text position should be kept in the new clone.
577 start = ubrk_current(brk);
578 if (start != 4) {
579 log_err("error ubrk_current(clone) did not return 4, but %i\n", start);
580 }
581
582 pos = ubrk_next(brk);
583 if (pos != 5) {
584 log_err("error ubrk_next(clone) did not return 5, but %i\n", pos);
585 }
586 start = ubrk_current(brk);
587 if (start != 5) {
588 log_err("error ubrk_current(clone) did not return 5, but %i\n", start);
589 }
590
591 start = ubrk_current(someIterators[i]);
592 if (start != 4) {
593 log_err("error ubrk_current(iter) did not keep the same position of 4,"
594 " but %i after advancing the position in its clone.\n", start);
595 }
596
597 ubrk_close(brk);
598
599 ubrk_close(someIterators[i]);
600 }
601 }
602 #endif
603
604
605 /*
606 // Open a break iterator from char * rules. Take care of conversion
607 // of the rules and error checking.
608 */
testOpenRules(char * rules)609 static UBreakIterator * testOpenRules(char *rules) {
610 UErrorCode status = U_ZERO_ERROR;
611 UChar *ruleSourceU = NULL;
612 void *strCleanUp = NULL;
613 UParseError parseErr;
614 UBreakIterator *bi;
615
616 ruleSourceU = toUChar(rules, &strCleanUp);
617
618 bi = ubrk_openRules(ruleSourceU, -1, /* The rules */
619 NULL, -1, /* The text to be iterated over. */
620 &parseErr, &status);
621
622 if (U_FAILURE(status)) {
623 log_data_err("FAIL: ubrk_openRules: ICU Error \"%s\" (Are you missing data?)\n", u_errorName(status));
624 bi = 0;
625 }
626 freeToUCharStrings(&strCleanUp);
627 return bi;
628
629 }
630
631 /*
632 * TestBreakIteratorRules - Verify that a break iterator can be created from
633 * a set of source rules.
634 */
TestBreakIteratorRules()635 static void TestBreakIteratorRules() {
636 /* Rules will keep together any run of letters not including 'a', OR
637 * keep together 'abc', but only when followed by 'def', OTHERWISE
638 * just return one char at a time.
639 */
640 char rules[] = "abc/def{666};\n [\\p{L} - [a]]* {2}; . {1};";
641 /* 0123456789012345678 */
642 char data[] = "abcdex abcdefgh-def"; /* the test data string */
643 char breaks[] = "** ** * ** *"; /* * the expected break positions */
644 char tags[] = "01 21 6 21 2"; /* expected tag values at break positions */
645 int32_t tagMap[] = {0, 1, 2, 3, 4, 5, 666};
646
647 UChar *uData;
648 void *freeHook = NULL;
649 UErrorCode status = U_ZERO_ERROR;
650 int32_t pos;
651 int i;
652
653 UBreakIterator *bi = testOpenRules(rules);
654 if (bi == NULL) {return;}
655 uData = toUChar(data, &freeHook);
656 ubrk_setText(bi, uData, -1, &status);
657
658 pos = ubrk_first(bi);
659 for (i=0; i<(int)sizeof(breaks); i++) {
660 if (pos == i && breaks[i] != '*') {
661 log_err("FAIL: unexpected break at position %d found\n", pos);
662 break;
663 }
664 if (pos != i && breaks[i] == '*') {
665 log_err("FAIL: expected break at position %d not found.\n", i);
666 break;
667 }
668 if (pos == i) {
669 int32_t tag, expectedTag;
670 tag = ubrk_getRuleStatus(bi);
671 expectedTag = tagMap[tags[i]&0xf];
672 if (tag != expectedTag) {
673 log_err("FAIL: incorrect tag value. Position = %d; expected tag %d, got %d",
674 pos, expectedTag, tag);
675 break;
676 }
677 pos = ubrk_next(bi);
678 }
679 }
680
681 /* #12914 add basic sanity test for ubrk_getBinaryRules, ubrk_openBinaryRules */
682 /* Underlying functionality checked in C++ rbbiapts.cpp TestRoundtripRules */
683 status = U_ZERO_ERROR;
684 int32_t rulesLength = ubrk_getBinaryRules(bi, NULL, 0, &status); /* preflight */
685 if (U_FAILURE(status)) {
686 log_err("FAIL: ubrk_getBinaryRules preflight err: %s", u_errorName(status));
687 } else {
688 uint8_t* binaryRules = (uint8_t*)uprv_malloc(rulesLength);
689 if (binaryRules == NULL) {
690 log_err("FAIL: unable to malloc rules buffer, size %u", rulesLength);
691 } else {
692 rulesLength = ubrk_getBinaryRules(bi, binaryRules, rulesLength, &status);
693 if (U_FAILURE(status)) {
694 log_err("FAIL: ubrk_getBinaryRules err: %s", u_errorName(status));
695 } else {
696 UBreakIterator* bi2 = ubrk_openBinaryRules(binaryRules, rulesLength, uData, -1, &status);
697 if (U_FAILURE(status)) {
698 log_err("FAIL: ubrk_openBinaryRules err: %s", u_errorName(status));
699 } else {
700 int32_t maxCount = sizeof(breaks); /* fail-safe test limit */
701 int32_t pos2 = ubrk_first(bi2);
702 pos = ubrk_first(bi);
703 do {
704 if (pos2 != pos) {
705 log_err("FAIL: interator from ubrk_openBinaryRules does not match original, get pos = %d instead of %d", pos2, pos);
706 }
707 pos2 = ubrk_next(bi2);
708 pos = ubrk_next(bi);
709 } while ((pos != UBRK_DONE || pos2 != UBRK_DONE) && maxCount-- > 0);
710
711 ubrk_close(bi2);
712 }
713 }
714 uprv_free(binaryRules);
715 }
716 }
717
718 freeToUCharStrings(&freeHook);
719 ubrk_close(bi);
720 }
721
TestBreakIteratorRuleError()722 static void TestBreakIteratorRuleError() {
723 /*
724 * TestBreakIteratorRuleError - Try to create a BI from rules with syntax errors,
725 * check that the error is reported correctly.
726 */
727 char rules[] = " # This is a rule comment on line 1\n"
728 "[:L:]; # this rule is OK.\n"
729 "abcdefg); # Error, mismatched parens\n";
730 UChar *uRules;
731 void *freeHook = NULL;
732 UErrorCode status = U_ZERO_ERROR;
733 UParseError parseErr;
734 UBreakIterator *bi;
735
736 uRules = toUChar(rules, &freeHook);
737 bi = ubrk_openRules(uRules, -1, /* The rules */
738 NULL, -1, /* The text to be iterated over. */
739 &parseErr, &status);
740 if (U_SUCCESS(status)) {
741 log_err("FAIL: construction of break iterator succeeded when it should have failed.\n");
742 ubrk_close(bi);
743 } else {
744 if (parseErr.line != 3 || parseErr.offset != 8) {
745 log_data_err("FAIL: incorrect error position reported. Got line %d, char %d, expected line 3, char 7 (Are you missing data?)\n",
746 parseErr.line, parseErr.offset);
747 }
748 }
749 freeToUCharStrings(&freeHook);
750 }
751
752
753 /*
754 * TestsBreakIteratorStatusVals() Test the ubrk_getRuleStatusVec() funciton
755 */
TestBreakIteratorStatusVec()756 static void TestBreakIteratorStatusVec() {
757 #define RULE_STRING_LENGTH 200
758 UChar rules[RULE_STRING_LENGTH];
759
760 #define TEST_STRING_LENGTH 25
761 UChar testString[TEST_STRING_LENGTH];
762 UBreakIterator *bi = NULL;
763 int32_t pos = 0;
764 int32_t vals[10];
765 int32_t numVals;
766 UErrorCode status = U_ZERO_ERROR;
767
768 u_uastrncpy(rules, "[A-N]{100}; \n"
769 "[a-w]{200}; \n"
770 "[\\p{L}]{300}; \n"
771 "[\\p{N}]{400}; \n"
772 "[0-5]{500}; \n"
773 "!.*;\n", RULE_STRING_LENGTH);
774 u_uastrncpy(testString, "ABC", TEST_STRING_LENGTH);
775
776
777 bi = ubrk_openRules(rules, -1, testString, -1, NULL, &status);
778 TEST_ASSERT_SUCCESS(status);
779 TEST_ASSERT(bi != NULL);
780
781 /* The TEST_ASSERT above should change too... */
782 if (bi != NULL) {
783 pos = ubrk_next(bi);
784 TEST_ASSERT(pos == 1);
785
786 memset(vals, -1, sizeof(vals));
787 numVals = ubrk_getRuleStatusVec(bi, vals, 10, &status);
788 TEST_ASSERT_SUCCESS(status);
789 TEST_ASSERT(numVals == 2);
790 TEST_ASSERT(vals[0] == 100);
791 TEST_ASSERT(vals[1] == 300);
792 TEST_ASSERT(vals[2] == -1);
793
794 numVals = ubrk_getRuleStatusVec(bi, vals, 0, &status);
795 TEST_ASSERT(status == U_BUFFER_OVERFLOW_ERROR);
796 TEST_ASSERT(numVals == 2);
797 }
798
799 ubrk_close(bi);
800 }
801
802
803 /*
804 * static void TestBreakIteratorUText(void);
805 *
806 * Test that ubrk_setUText() is present and works for a simple case.
807 */
TestBreakIteratorUText(void)808 static void TestBreakIteratorUText(void) {
809 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 */
810 /* 0 1 2 34567890 */
811
812 UErrorCode status = U_ZERO_ERROR;
813 UBreakIterator *bi = NULL;
814 int32_t pos = 0;
815
816
817 UText *ut = utext_openUTF8(NULL, UTF8Str, -1, &status);
818 TEST_ASSERT_SUCCESS(status);
819
820 bi = ubrk_open(UBRK_WORD, "en_US", NULL, 0, &status);
821 if (U_FAILURE(status)) {
822 log_err_status(status, "Failure at file %s, line %d, error = %s\n", __FILE__, __LINE__, u_errorName(status));
823 utext_close(ut);
824 return;
825 }
826
827 ubrk_setUText(bi, ut, &status);
828 if (U_FAILURE(status)) {
829 log_err("Failure at file %s, line %d, error = %s\n", __FILE__, __LINE__, u_errorName(status));
830 ubrk_close(bi);
831 utext_close(ut);
832 return;
833 }
834
835 pos = ubrk_first(bi);
836 TEST_ASSERT(pos == 0);
837
838 pos = ubrk_next(bi);
839 TEST_ASSERT(pos == 4);
840
841 pos = ubrk_next(bi);
842 TEST_ASSERT(pos == 5);
843
844 pos = ubrk_next(bi);
845 TEST_ASSERT(pos == 10);
846
847 pos = ubrk_next(bi);
848 TEST_ASSERT(pos == UBRK_DONE);
849 ubrk_close(bi);
850 utext_close(ut);
851 }
852
853 /*
854 * static void TestBreakIteratorTailoring(void);
855 *
856 * Test break iterator tailorings from CLDR data.
857 */
858
859 /* Thai/Lao grapheme break tailoring */
860 static const UChar thTest[] = { 0x0020, 0x0E40, 0x0E01, 0x0020,
861 0x0E01, 0x0E30, 0x0020, 0x0E01, 0x0E33, 0x0020, 0 };
862 /*in Unicode 6.1 en should behave just like th for this*/
863 /*static const int32_t thTestOffs_enFwd[] = { 1, 3, 4, 6, 7, 9, 10 };*/
864 static const int32_t thTestOffs_thFwd[] = { 1, 2, 3, 4, 5, 6, 7, 9, 10 };
865 /*static const int32_t thTestOffs_enRev[] = { 9, 7, 6, 4, 3, 1, 0 };*/
866 static const int32_t thTestOffs_thRev[] = { 9, 7, 6, 5, 4, 3, 2, 1, 0 };
867
868 /* Hebrew line break tailoring, for cldrbug 3028 */
869 static const UChar heTest[] = { 0x0020, 0x002D, 0x0031, 0x0032, 0x0020,
870 0x0061, 0x002D, 0x006B, 0x0020,
871 0x0061, 0x0300, 0x2010, 0x006B, 0x0020,
872 0x05DE, 0x05D4, 0x002D, 0x0069, 0x0020,
873 0x05D1, 0x05BC, 0x2010, 0x0047, 0x0020, 0 };
874 /*in Unicode 6.1 en should behave just like he for this*/
875 /*static const int32_t heTestOffs_enFwd[] = { 1, 5, 7, 9, 12, 14, 17, 19, 22, 24 };*/
876 static const int32_t heTestOffs_heFwd[] = { 1, 5, 7, 9, 12, 14, 19, 24 };
877 /*static const int32_t heTestOffs_enRev[] = { 22, 19, 17, 14, 12, 9, 7, 5, 1, 0 };*/
878 static const int32_t heTestOffs_heRev[] = { 19, 14, 12, 9, 7, 5, 1, 0 };
879
880 /* Finnish line break tailoring, for cldrbug 3029.
881 * As of ICU 63, Finnish tailoring moved to root, Finnish and English should be the same. */
882 static const UChar fiTest[] = { /* 00 */ 0x0020, 0x002D, 0x0031, 0x0032, 0x0020,
883 /* 05 */ 0x0061, 0x002D, 0x006B, 0x0020,
884 /* 09 */ 0x0061, 0x0300, 0x2010, 0x006B, 0x0020,
885 /* 14 */ 0x0061, 0x0020, 0x002D, 0x006B, 0x0020,
886 /* 19 */ 0x0061, 0x0300, 0x0020, 0x2010, 0x006B, 0x0020, 0 };
887 //static const int32_t fiTestOffs_enFwd[] = { 1, 5, 7, 9, 12, 14, 16, 17, 19, 22, 23, 25 };
888 static const int32_t fiTestOffs_enFwd[] = { 1, 5, 7, 9, 12, 14, 16, 19, 22, 25 };
889 static const int32_t fiTestOffs_fiFwd[] = { 1, 5, 7, 9, 12, 14, 16, 19, 22, 25 };
890 //static const int32_t fiTestOffs_enRev[] = { 23, 22, 19, 17, 16, 14, 12, 9, 7, 5, 1, 0 };
891 static const int32_t fiTestOffs_enRev[] = { 22, 19, 16, 14, 12, 9, 7, 5, 1, 0 };
892 static const int32_t fiTestOffs_fiRev[] = { 22, 19, 16, 14, 12, 9, 7, 5, 1, 0 };
893
894 /* Khmer dictionary-based work break, for ICU ticket #8329 */
895 static const UChar kmTest[] = { /* 00 */ 0x179F, 0x17BC, 0x1798, 0x1785, 0x17C6, 0x178E, 0x17B6, 0x1799, 0x1796, 0x17C1,
896 /* 10 */ 0x179B, 0x1794, 0x1793, 0x17D2, 0x178F, 0x17B7, 0x1785, 0x178A, 0x17BE, 0x1798,
897 /* 20 */ 0x17D2, 0x1794, 0x17B8, 0x17A2, 0x1792, 0x17B7, 0x179F, 0x17D2, 0x178B, 0x17B6,
898 /* 30 */ 0x1793, 0x17A2, 0x179A, 0x1796, 0x17D2, 0x179A, 0x17C7, 0x1782, 0x17BB, 0x178E,
899 /* 40 */ 0x178A, 0x179B, 0x17CB, 0x1796, 0x17D2, 0x179A, 0x17C7, 0x17A2, 0x1784, 0x17D2,
900 /* 50 */ 0x1782, 0 };
901 static const int32_t kmTestOffs_kmFwd[] = { 3, /*8,*/ 11, 17, 23, 31, /*33,*/ 40, 43, 51 }; /* TODO: Investigate failure to break at offset 8 */
902 static const int32_t kmTestOffs_kmRev[] = { 43, 40, /*33,*/ 31, 23, 17, 11, /*8,*/ 3, 0 };
903
904 typedef struct {
905 const char * locale;
906 UBreakIteratorType type;
907 const UChar * test;
908 const int32_t * offsFwd;
909 const int32_t * offsRev;
910 int32_t numOffsets;
911 } RBBITailoringTest;
912
913 static const RBBITailoringTest tailoringTests[] = {
914 { "en", UBRK_CHARACTER, thTest, thTestOffs_thFwd, thTestOffs_thRev, UPRV_LENGTHOF(thTestOffs_thFwd) },
915 { "en_US_POSIX", UBRK_CHARACTER, thTest, thTestOffs_thFwd, thTestOffs_thRev, UPRV_LENGTHOF(thTestOffs_thFwd) },
916 { "en", UBRK_LINE, heTest, heTestOffs_heFwd, heTestOffs_heRev, UPRV_LENGTHOF(heTestOffs_heFwd) },
917 { "he", UBRK_LINE, heTest, heTestOffs_heFwd, heTestOffs_heRev, UPRV_LENGTHOF(heTestOffs_heFwd) },
918 { "en", UBRK_LINE, fiTest, fiTestOffs_enFwd, fiTestOffs_enRev, UPRV_LENGTHOF(fiTestOffs_enFwd) },
919 { "fi", UBRK_LINE, fiTest, fiTestOffs_fiFwd, fiTestOffs_fiRev, UPRV_LENGTHOF(fiTestOffs_fiFwd) },
920 { "km", UBRK_WORD, kmTest, kmTestOffs_kmFwd, kmTestOffs_kmRev, UPRV_LENGTHOF(kmTestOffs_kmFwd) },
921 { NULL, 0, NULL, NULL, NULL, 0 },
922 };
923
TestBreakIteratorTailoring(void)924 static void TestBreakIteratorTailoring(void) {
925 const RBBITailoringTest * testPtr;
926 for (testPtr = tailoringTests; testPtr->locale != NULL; ++testPtr) {
927 UErrorCode status = U_ZERO_ERROR;
928 UBreakIterator* ubrkiter = ubrk_open(testPtr->type, testPtr->locale, testPtr->test, -1, &status);
929 if ( U_SUCCESS(status) ) {
930 int32_t offset, offsindx;
931 UBool foundError;
932
933 foundError = FALSE;
934 for (offsindx = 0; (offset = ubrk_next(ubrkiter)) != UBRK_DONE; ++offsindx) {
935 if (!foundError && offsindx >= testPtr->numOffsets) {
936 log_err("FAIL: locale %s, break type %d, ubrk_next expected UBRK_DONE, got %d\n",
937 testPtr->locale, testPtr->type, offset);
938 foundError = TRUE;
939 } else if (!foundError && offset != testPtr->offsFwd[offsindx]) {
940 log_err("FAIL: locale %s, break type %d, ubrk_next expected %d, got %d\n",
941 testPtr->locale, testPtr->type, testPtr->offsFwd[offsindx], offset);
942 foundError = TRUE;
943 }
944 }
945 if (!foundError && offsindx < testPtr->numOffsets) {
946 log_err("FAIL: locale %s, break type %d, ubrk_next expected %d, got UBRK_DONE\n",
947 testPtr->locale, testPtr->type, testPtr->offsFwd[offsindx]);
948 }
949
950 foundError = FALSE;
951 for (offsindx = 0; (offset = ubrk_previous(ubrkiter)) != UBRK_DONE; ++offsindx) {
952 if (!foundError && offsindx >= testPtr->numOffsets) {
953 log_err("FAIL: locale %s, break type %d, ubrk_previous expected UBRK_DONE, got %d\n",
954 testPtr->locale, testPtr->type, offset);
955 foundError = TRUE;
956 } else if (!foundError && offset != testPtr->offsRev[offsindx]) {
957 log_err("FAIL: locale %s, break type %d, ubrk_previous expected %d, got %d\n",
958 testPtr->locale, testPtr->type, testPtr->offsRev[offsindx], offset);
959 foundError = TRUE;
960 }
961 }
962 if (!foundError && offsindx < testPtr->numOffsets) {
963 log_err("FAIL: locale %s, break type %d, ubrk_previous expected %d, got UBRK_DONE\n",
964 testPtr->locale, testPtr->type, testPtr->offsRev[offsindx]);
965 }
966
967 ubrk_close(ubrkiter);
968 } else {
969 log_err_status(status, "FAIL: locale %s, break type %d, ubrk_open status: %s\n", testPtr->locale, testPtr->type, u_errorName(status));
970 }
971 }
972 }
973
974
TestBreakIteratorRefresh(void)975 static void TestBreakIteratorRefresh(void) {
976 /*
977 * RefreshInput changes out the input of a Break Iterator without
978 * changing anything else in the iterator's state. Used with Java JNI,
979 * when Java moves the underlying string storage. This test
980 * runs a ubrk_next() repeatedly, moving the text in the middle of the sequence.
981 * The right set of boundaries should still be found.
982 */
983 UChar testStr[] = {0x20, 0x41, 0x20, 0x42, 0x20, 0x43, 0x20, 0x44, 0x0}; /* = " A B C D" */
984 UChar movedStr[] = {0x20, 0x20, 0x20, 0x20, 0x20, 0x20, 0x20, 0x20, 0};
985 UErrorCode status = U_ZERO_ERROR;
986 UBreakIterator *bi;
987 UText ut1 = UTEXT_INITIALIZER;
988 UText ut2 = UTEXT_INITIALIZER;
989
990 bi = ubrk_open(UBRK_LINE, "en_US", NULL, 0, &status);
991 TEST_ASSERT_SUCCESS(status);
992 if (U_FAILURE(status)) {
993 return;
994 }
995
996 utext_openUChars(&ut1, testStr, -1, &status);
997 TEST_ASSERT_SUCCESS(status);
998 ubrk_setUText(bi, &ut1, &status);
999 TEST_ASSERT_SUCCESS(status);
1000
1001 if (U_SUCCESS(status)) {
1002 /* Line boundaries will occur before each letter in the original string */
1003 TEST_ASSERT(1 == ubrk_next(bi));
1004 TEST_ASSERT(3 == ubrk_next(bi));
1005
1006 /* Move the string, kill the original string. */
1007 u_strcpy(movedStr, testStr);
1008 u_memset(testStr, 0x20, u_strlen(testStr));
1009 utext_openUChars(&ut2, movedStr, -1, &status);
1010 TEST_ASSERT_SUCCESS(status);
1011 ubrk_refreshUText(bi, &ut2, &status);
1012 TEST_ASSERT_SUCCESS(status);
1013
1014 /* Find the following matches, now working in the moved string. */
1015 TEST_ASSERT(5 == ubrk_next(bi));
1016 TEST_ASSERT(7 == ubrk_next(bi));
1017 TEST_ASSERT(8 == ubrk_next(bi));
1018 TEST_ASSERT(UBRK_DONE == ubrk_next(bi));
1019 TEST_ASSERT_SUCCESS(status);
1020
1021 utext_close(&ut1);
1022 utext_close(&ut2);
1023 }
1024 ubrk_close(bi);
1025 }
1026
1027
TestBug11665(void)1028 static void TestBug11665(void) {
1029 // The problem was with the incorrect breaking of Japanese text beginning
1030 // with Katakana characters when no prior Japanese or Chinese text had been
1031 // encountered.
1032 //
1033 // Tested here in cintltst, rather than in intltest, because only cintltst
1034 // tests have the ability to reset ICU, which is needed to get the bug
1035 // to manifest itself.
1036
1037 static UChar japaneseText[] = {0x30A2, 0x30EC, 0x30EB, 0x30AE, 0x30FC, 0x6027, 0x7D50, 0x819C, 0x708E};
1038 int32_t boundaries[10] = {0};
1039 UBreakIterator *bi = NULL;
1040 int32_t brk;
1041 int32_t brkIdx = 0;
1042 int32_t totalBreaks = 0;
1043 UErrorCode status = U_ZERO_ERROR;
1044
1045 ctest_resetICU();
1046 bi = ubrk_open(UBRK_WORD, "en_US", japaneseText, UPRV_LENGTHOF(japaneseText), &status);
1047 TEST_ASSERT_SUCCESS(status);
1048 if (!bi) {
1049 return;
1050 }
1051 for (brk=ubrk_first(bi); brk != UBRK_DONE; brk=ubrk_next(bi)) {
1052 boundaries[brkIdx] = brk;
1053 if (++brkIdx >= UPRV_LENGTHOF(boundaries) - 1) {
1054 break;
1055 }
1056 }
1057 if (brkIdx <= 2 || brkIdx >= UPRV_LENGTHOF(boundaries)) {
1058 log_err("%s:%d too few or many breaks found.\n", __FILE__, __LINE__);
1059 } else {
1060 totalBreaks = brkIdx;
1061 brkIdx = 0;
1062 for (brk=ubrk_first(bi); brk != UBRK_DONE; brk=ubrk_next(bi)) {
1063 if (brk != boundaries[brkIdx]) {
1064 log_err("%s:%d Break #%d differs between first and second iteration.\n", __FILE__, __LINE__, brkIdx);
1065 break;
1066 }
1067 if (++brkIdx >= UPRV_LENGTHOF(boundaries) - 1) {
1068 log_err("%s:%d Too many breaks.\n", __FILE__, __LINE__);
1069 break;
1070 }
1071 }
1072 if (totalBreaks != brkIdx) {
1073 log_err("%s:%d Number of breaks differ between first and second iteration.\n", __FILE__, __LINE__);
1074 }
1075 }
1076 ubrk_close(bi);
1077 }
1078
1079 /*
1080 * expOffset is the set of expected offsets, ending with '-1'.
1081 * "Expected expOffset -1" means "expected the end of the offsets"
1082 */
1083
1084 static const char testSentenceSuppressionsEn[] = "Mr. Jones comes home. Dr. Smith Ph.D. is out. In the U.S.A. it is hot.";
1085 static const int32_t testSentSuppFwdOffsetsEn[] = { 22, 26, 46, 70, -1 }; /* With suppressions, currently not handling Dr. */
1086 static const int32_t testSentFwdOffsetsEn[] = { 4, 22, 26, 46, 70, -1 }; /* Without suppressions */
1087 static const int32_t testSentSuppRevOffsetsEn[] = { 46, 26, 22, 0, -1 }; /* With suppressions, currently not handling Dr. */
1088 static const int32_t testSentRevOffsetsEn[] = { 46, 26, 22, 4, 0, -1 }; /* Without suppressions */
1089
1090 static const char testSentenceSuppressionsDe[] = "Wenn ich schon h\\u00F6re zu Guttenberg kommt evtl. zur\\u00FCck.";
1091 static const int32_t testSentSuppFwdOffsetsDe[] = { 53, -1 }; /* With suppressions */
1092 static const int32_t testSentFwdOffsetsDe[] = { 53, -1 }; /* Without suppressions; no break in evtl. zur due to casing */
1093 static const int32_t testSentSuppRevOffsetsDe[] = { 0, -1 }; /* With suppressions */
1094 static const int32_t testSentRevOffsetsDe[] = { 0, -1 }; /* Without suppressions */
1095
1096 static const char testSentenceSuppressionsEs[] = "Te esperamos todos los miercoles en Bravo 416, Col. El Pueblo a las 7 PM.";
1097 static const int32_t testSentSuppFwdOffsetsEs[] = { 73, -1 }; /* With suppressions */
1098 static const int32_t testSentFwdOffsetsEs[] = { 52, 73, -1 }; /* Without suppressions */
1099 static const int32_t testSentSuppRevOffsetsEs[] = { 0, -1 }; /* With suppressions */
1100 static const int32_t testSentRevOffsetsEs[] = { 52, 0, -1 }; /* Without suppressions */
1101
1102 enum { kTextULenMax = 128 };
1103
1104 typedef struct {
1105 const char * locale;
1106 const char * text;
1107 const int32_t * expFwdOffsets;
1108 const int32_t * expRevOffsets;
1109 } TestBISuppressionsItem;
1110
1111 static const TestBISuppressionsItem testBISuppressionsItems[] = {
1112 { "en@ss=standard", testSentenceSuppressionsEn, testSentSuppFwdOffsetsEn, testSentSuppRevOffsetsEn },
1113 { "en", testSentenceSuppressionsEn, testSentFwdOffsetsEn, testSentRevOffsetsEn },
1114 { "en_CA", testSentenceSuppressionsEn, testSentFwdOffsetsEn, testSentRevOffsetsEn },
1115 { "en_CA@ss=standard", testSentenceSuppressionsEn, testSentSuppFwdOffsetsEn, testSentSuppRevOffsetsEn },
1116 { "fr@ss=standard", testSentenceSuppressionsEn, testSentFwdOffsetsEn, testSentRevOffsetsEn },
1117 { "af@ss=standard", testSentenceSuppressionsEn, testSentFwdOffsetsEn, testSentRevOffsetsEn }, /* no brkiter data => nosuppressions? */
1118 { "af_ZA@ss=standard", testSentenceSuppressionsEn, testSentFwdOffsetsEn, testSentRevOffsetsEn }, /* no brkiter data => nosuppressions? */
1119 { "zh@ss=standard", testSentenceSuppressionsEn, testSentFwdOffsetsEn, testSentRevOffsetsEn }, /* brkiter data, no suppressions data => no suppressions */
1120 { "zh_Hant@ss=standard", testSentenceSuppressionsEn, testSentFwdOffsetsEn, testSentRevOffsetsEn }, /* brkiter data, no suppressions data => no suppressions */
1121 { "fi@ss=standard", testSentenceSuppressionsEn, testSentFwdOffsetsEn, testSentRevOffsetsEn }, /* brkiter data, no suppressions data => no suppressions */
1122 { "ja@ss=standard", testSentenceSuppressionsEn, testSentFwdOffsetsEn, testSentRevOffsetsEn }, /* brkiter data, no suppressions data => no suppressions */
1123 { "de@ss=standard", testSentenceSuppressionsDe, testSentSuppFwdOffsetsDe, testSentSuppRevOffsetsDe },
1124 { "de", testSentenceSuppressionsDe, testSentFwdOffsetsDe, testSentRevOffsetsDe },
1125 { "es@ss=standard", testSentenceSuppressionsEs, testSentSuppFwdOffsetsEs, testSentSuppRevOffsetsEs },
1126 { "es", testSentenceSuppressionsEs, testSentFwdOffsetsEs, testSentRevOffsetsEs },
1127 { NULL, NULL, NULL, NULL }
1128 };
1129
TestBreakIteratorSuppressions(void)1130 static void TestBreakIteratorSuppressions(void) {
1131 const TestBISuppressionsItem * itemPtr;
1132
1133 for (itemPtr = testBISuppressionsItems; itemPtr->locale != NULL; itemPtr++) {
1134 UChar textU[kTextULenMax];
1135 int32_t textULen = u_unescape(itemPtr->text, textU, kTextULenMax);
1136 UErrorCode status = U_ZERO_ERROR;
1137 UBreakIterator *bi = ubrk_open(UBRK_SENTENCE, itemPtr->locale, textU, textULen, &status);
1138 log_verbose("#%d: %s\n", (itemPtr-testBISuppressionsItems), itemPtr->locale);
1139 if (U_SUCCESS(status)) {
1140 int32_t offset, start;
1141 const int32_t * expOffsetPtr;
1142 const int32_t * expOffsetStart;
1143
1144 expOffsetStart = expOffsetPtr = itemPtr->expFwdOffsets;
1145 ubrk_first(bi);
1146 for (; (offset = ubrk_next(bi)) != UBRK_DONE && *expOffsetPtr >= 0; expOffsetPtr++) {
1147 if (offset != *expOffsetPtr) {
1148 log_err("FAIL: ubrk_next loc \"%s\", expected %d, got %d\n", itemPtr->locale, *expOffsetPtr, offset);
1149 }
1150 }
1151 if (offset != UBRK_DONE || *expOffsetPtr >= 0) {
1152 log_err("FAIL: ubrk_next loc \"%s\", expected UBRK_DONE & expOffset -1, got %d and %d\n", itemPtr->locale, offset, *expOffsetPtr);
1153 }
1154
1155 expOffsetStart = expOffsetPtr = itemPtr->expFwdOffsets;
1156 start = ubrk_first(bi) + 1;
1157 for (; (offset = ubrk_following(bi, start)) != UBRK_DONE && *expOffsetPtr >= 0; expOffsetPtr++) {
1158 if (offset != *expOffsetPtr) {
1159 log_err("FAIL: ubrk_following(%d) loc \"%s\", expected %d, got %d\n", start, itemPtr->locale, *expOffsetPtr, offset);
1160 }
1161 start = *expOffsetPtr + 1;
1162 }
1163 if (offset != UBRK_DONE || *expOffsetPtr >= 0) {
1164 log_err("FAIL: ubrk_following(%d) loc \"%s\", expected UBRK_DONE & expOffset -1, got %d and %d\n", start, itemPtr->locale, offset, *expOffsetPtr);
1165 }
1166
1167 expOffsetStart = expOffsetPtr = itemPtr->expRevOffsets;
1168 offset = ubrk_last(bi);
1169 log_verbose("___ @%d ubrk_last\n", offset);
1170 if(offset == 0) {
1171 log_err("FAIL: ubrk_last loc \"%s\" unexpected %d\n", itemPtr->locale, offset);
1172 }
1173 for (; (offset = ubrk_previous(bi)) != UBRK_DONE && *expOffsetPtr >= 0; expOffsetPtr++) {
1174 if (offset != *expOffsetPtr) {
1175 log_err("FAIL: ubrk_previous loc \"%s\", expected %d, got %d\n", itemPtr->locale, *expOffsetPtr, offset);
1176 } else {
1177 log_verbose("[%d] @%d ubrk_previous()\n", (expOffsetPtr - expOffsetStart), offset);
1178 }
1179 }
1180 if (offset != UBRK_DONE || *expOffsetPtr >= 0) {
1181 log_err("FAIL: ubrk_previous loc \"%s\", expected UBRK_DONE & expOffset[%d] -1, got %d and %d\n", itemPtr->locale,
1182 expOffsetPtr - expOffsetStart,
1183 offset, *expOffsetPtr);
1184 }
1185
1186 expOffsetStart = expOffsetPtr = itemPtr->expRevOffsets;
1187 start = ubrk_last(bi) - 1;
1188 for (; (offset = ubrk_preceding(bi, start)) != UBRK_DONE && *expOffsetPtr >= 0; expOffsetPtr++) {
1189 if (offset != *expOffsetPtr) {
1190 log_err("FAIL: ubrk_preceding(%d) loc \"%s\", expected %d, got %d\n", start, itemPtr->locale, *expOffsetPtr, offset);
1191 }
1192 start = *expOffsetPtr - 1;
1193 }
1194 if (start >=0 && (offset != UBRK_DONE || *expOffsetPtr >= 0)) {
1195 log_err("FAIL: ubrk_preceding loc(%d) \"%s\", expected UBRK_DONE & expOffset -1, got %d and %d\n", start, itemPtr->locale, offset, *expOffsetPtr);
1196 }
1197
1198 ubrk_close(bi);
1199 } else {
1200 log_data_err("FAIL: ubrk_open(UBRK_SENTENCE, \"%s\", ...) status %s (Are you missing data?)\n", itemPtr->locale, u_errorName(status));
1201 }
1202 }
1203 }
1204
1205
1206 #endif /* #if !UCONFIG_NO_BREAK_ITERATION */
1207