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 <stdbool.h>
28 #include <stdlib.h>
29 #include <string.h>
30 #include "unicode/uloc.h"
31 #include "unicode/ubrk.h"
32 #include "unicode/ustring.h"
33 #include "unicode/ucnv.h"
34 #include "unicode/utext.h"
35 #include "cintltst.h"
36 #include "cbiapts.h"
37 #include "cmemory.h"
38
39 #define TEST_ASSERT_SUCCESS(status) UPRV_BLOCK_MACRO_BEGIN { \
40 if (U_FAILURE(status)) { \
41 log_data_err("Failure at file %s, line %d, error = %s (Are you missing data?)\n", __FILE__, __LINE__, u_errorName(status)); \
42 } \
43 } UPRV_BLOCK_MACRO_END
44
45 #define TEST_ASSERT(expr) UPRV_BLOCK_MACRO_BEGIN { \
46 if ((expr)==false) { \
47 log_data_err("Test Failure at file %s, line %d (Are you missing data?)\n", __FILE__, __LINE__); \
48 } \
49 } UPRV_BLOCK_MACRO_END
50
51 #if !UCONFIG_NO_FILE_IO
52 static void TestBreakIteratorSafeClone(void);
53 static void TestBreakIteratorClone(void);
54 #endif
55 static void TestBreakIteratorRules(void);
56 static void TestBreakIteratorRuleError(void);
57 static void TestBreakIteratorStatusVec(void);
58 static void TestBreakIteratorUText(void);
59 static void TestBreakIteratorTailoring(void);
60 static void TestBreakIteratorRefresh(void);
61 static void TestBug11665(void);
62 static void TestBreakIteratorSuppressions(void);
63
64 void addBrkIterAPITest(TestNode** root);
65
addBrkIterAPITest(TestNode ** root)66 void addBrkIterAPITest(TestNode** root)
67 {
68 #if !UCONFIG_NO_FILE_IO
69 addTest(root, &TestBreakIteratorCAPI, "tstxtbd/cbiapts/TestBreakIteratorCAPI");
70 addTest(root, &TestBreakIteratorSafeClone, "tstxtbd/cbiapts/TestBreakIteratorSafeClone");
71 addTest(root, &TestBreakIteratorClone, "tstxtbd/cbiapts/TestBreakIteratorClone");
72 addTest(root, &TestBreakIteratorUText, "tstxtbd/cbiapts/TestBreakIteratorUText");
73 #endif
74 addTest(root, &TestBreakIteratorRules, "tstxtbd/cbiapts/TestBreakIteratorRules");
75 addTest(root, &TestBreakIteratorRuleError, "tstxtbd/cbiapts/TestBreakIteratorRuleError");
76 addTest(root, &TestBreakIteratorStatusVec, "tstxtbd/cbiapts/TestBreakIteratorStatusVec");
77 addTest(root, &TestBreakIteratorTailoring, "tstxtbd/cbiapts/TestBreakIteratorTailoring");
78 addTest(root, &TestBreakIteratorRefresh, "tstxtbd/cbiapts/TestBreakIteratorRefresh");
79 addTest(root, &TestBug11665, "tstxtbd/cbiapts/TestBug11665");
80 #if !UCONFIG_NO_FILTERED_BREAK_ITERATION
81 addTest(root, &TestBreakIteratorSuppressions, "tstxtbd/cbiapts/TestBreakIteratorSuppressions");
82 #endif
83 }
84
85 #define CLONETEST_ITERATOR_COUNT 2
86
87 /*
88 * Utility function for converting char * to UChar * strings, to
89 * simplify the test code. Converted strings are put in heap allocated
90 * storage. A hook (probably a local in the caller's code) allows all
91 * strings converted with that hook to be freed with a single call.
92 */
93 typedef struct StringStruct {
94 struct StringStruct *link;
95 UChar str[1];
96 } StringStruct;
97
98
toUChar(const char * src,void ** freeHook)99 static UChar* toUChar(const char *src, void **freeHook) {
100 /* Structure of the memory that we allocate on the heap */
101
102 int32_t numUChars;
103 int32_t destSize;
104 UChar stackBuf[2000 + sizeof(void *)/sizeof(UChar)];
105 StringStruct *dest;
106 UConverter *cnv;
107
108 UErrorCode status = U_ZERO_ERROR;
109 if (src == NULL) {
110 return NULL;
111 }
112
113 cnv = ucnv_open(NULL, &status);
114 if(U_FAILURE(status) || cnv == NULL) {
115 return NULL;
116 }
117 ucnv_reset(cnv);
118 numUChars = ucnv_toUChars(cnv,
119 stackBuf,
120 2000,
121 src, -1,
122 &status);
123
124 destSize = (numUChars+1) * sizeof(UChar) + sizeof(struct StringStruct);
125 dest = (StringStruct *)malloc(destSize);
126 if (dest != NULL) {
127 if (status == U_BUFFER_OVERFLOW_ERROR || status == U_STRING_NOT_TERMINATED_WARNING) {
128 ucnv_toUChars(cnv, dest->str, numUChars+1, src, -1, &status);
129 } else if (status == U_ZERO_ERROR) {
130 u_strcpy(dest->str, stackBuf);
131 } else {
132 free(dest);
133 dest = NULL;
134 }
135 }
136
137 ucnv_reset(cnv); /* be good citizens */
138 ucnv_close(cnv);
139 if (dest == NULL) {
140 return NULL;
141 }
142
143 dest->link = (StringStruct*)(*freeHook);
144 *freeHook = dest;
145 return dest->str;
146 }
147
freeToUCharStrings(void ** hook)148 static void freeToUCharStrings(void **hook) {
149 StringStruct *s = *(StringStruct **)hook;
150 while (s != NULL) {
151 StringStruct *next = s->link;
152 free(s);
153 s = next;
154 }
155 }
156
157
158 #if !UCONFIG_NO_FILE_IO
TestBreakIteratorCAPI()159 static void TestBreakIteratorCAPI()
160 {
161 UErrorCode status = U_ZERO_ERROR;
162 UBreakIterator *word, *sentence, *line, *character, *b, *bogus;
163 int32_t start,pos,end,to;
164 int32_t i;
165 int32_t count = 0;
166
167 UChar text[50];
168
169 /* Note: the adjacent "" are concatenating strings, not adding a \" to the
170 string, which is probably what whoever wrote this intended. Don't fix,
171 because it would throw off the hard coded break positions in the following
172 tests. */
173 u_uastrcpy(text, "He's from Africa. ""Mr. Livingston, I presume?"" Yeah");
174
175
176 /*test ubrk_open()*/
177 log_verbose("\nTesting BreakIterator open functions\n");
178
179 /* Use french for fun */
180 word = ubrk_open(UBRK_WORD, "en_US", text, u_strlen(text), &status);
181 if(status == U_FILE_ACCESS_ERROR) {
182 log_data_err("Check your data - it doesn't seem to be around\n");
183 return;
184 } else if(U_FAILURE(status)){
185 log_err_status(status, "FAIL: Error in ubrk_open() for word breakiterator: %s\n", myErrorName(status));
186 }
187 else{
188 log_verbose("PASS: Successfully opened word breakiterator\n");
189 }
190
191 sentence = ubrk_open(UBRK_SENTENCE, "en_US", text, u_strlen(text), &status);
192 if(U_FAILURE(status)){
193 log_err_status(status, "FAIL: Error in ubrk_open() for sentence breakiterator: %s\n", myErrorName(status));
194 return;
195 }
196 else{
197 log_verbose("PASS: Successfully opened sentence breakiterator\n");
198 }
199
200 line = ubrk_open(UBRK_LINE, "en_US", text, u_strlen(text), &status);
201 if(U_FAILURE(status)){
202 log_err("FAIL: Error in ubrk_open() for line breakiterator: %s\n", myErrorName(status));
203 return;
204 }
205 else{
206 log_verbose("PASS: Successfully opened line breakiterator\n");
207 }
208
209 character = ubrk_open(UBRK_CHARACTER, "en_US", text, u_strlen(text), &status);
210 if(U_FAILURE(status)){
211 log_err("FAIL: Error in ubrk_open() for character breakiterator: %s\n", myErrorName(status));
212 return;
213 }
214 else{
215 log_verbose("PASS: Successfully opened character breakiterator\n");
216 }
217 /*trying to open an illegal iterator*/
218 bogus = ubrk_open((UBreakIteratorType)5, "en_US", text, u_strlen(text), &status);
219 if(bogus != NULL) {
220 log_err("FAIL: expected NULL from opening an invalid break iterator.\n");
221 }
222 if(U_SUCCESS(status)){
223 log_err("FAIL: Error in ubrk_open() for BOGUS breakiterator. Expected U_ILLEGAL_ARGUMENT_ERROR\n");
224 }
225 if(U_FAILURE(status)){
226 if(status != U_ILLEGAL_ARGUMENT_ERROR){
227 log_err("FAIL: Error in ubrk_open() for BOGUS breakiterator. Expected U_ILLEGAL_ARGUMENT_ERROR\n Got %s\n", myErrorName(status));
228 }
229 }
230 status=U_ZERO_ERROR;
231
232
233 /* ======= Test ubrk_countAvailable() and ubrk_getAvailable() */
234
235 log_verbose("\nTesting ubrk_countAvailable() and ubrk_getAvailable()\n");
236 count=ubrk_countAvailable();
237 /* use something sensible w/o hardcoding the count */
238 if(count < 0){
239 log_err("FAIL: Error in ubrk_countAvailable() returned %d\n", count);
240 }
241 else{
242 log_verbose("PASS: ubrk_countAvailable() successful returned %d\n", count);
243 }
244 for(i=0;i<count;i++)
245 {
246 log_verbose("%s\n", ubrk_getAvailable(i));
247 if (ubrk_getAvailable(i) == 0)
248 log_err("No locale for which breakiterator is applicable\n");
249 else
250 log_verbose("A locale %s for which breakiterator is applicable\n",ubrk_getAvailable(i));
251 }
252
253 /*========Test ubrk_first(), ubrk_last()...... and other functions*/
254
255 log_verbose("\nTesting the functions for word\n");
256 start = ubrk_first(word);
257 if(start!=0)
258 log_err("error ubrk_start(word) did not return 0\n");
259 log_verbose("first (word = %d\n", (int32_t)start);
260 pos=ubrk_next(word);
261 if(pos!=4)
262 log_err("error ubrk_next(word) did not return 4\n");
263 log_verbose("next (word = %d\n", (int32_t)pos);
264 pos=ubrk_following(word, 4);
265 if(pos!=5)
266 log_err("error ubrl_following(word,4) did not return 6\n");
267 log_verbose("next (word = %d\n", (int32_t)pos);
268 end=ubrk_last(word);
269 if(end!=49)
270 log_err("error ubrk_last(word) did not return 49\n");
271 log_verbose("last (word = %d\n", (int32_t)end);
272
273 pos=ubrk_previous(word);
274 log_verbose("%d %d\n", end, pos);
275
276 pos=ubrk_previous(word);
277 log_verbose("%d \n", pos);
278
279 if (ubrk_isBoundary(word, 2) != false) {
280 log_err("error ubrk_isBoundary(word, 2) did not return false\n");
281 }
282 pos=ubrk_current(word);
283 if (pos != 4) {
284 log_err("error ubrk_current() != 4 after ubrk_isBoundary(word, 2)\n");
285 }
286 if (ubrk_isBoundary(word, 4) != true) {
287 log_err("error ubrk_isBoundary(word, 4) did not return true\n");
288 }
289
290
291
292 log_verbose("\nTesting the functions for character\n");
293 ubrk_first(character);
294 pos = ubrk_following(character, 5);
295 if(pos!=6)
296 log_err("error ubrk_following(character,5) did not return 6\n");
297 log_verbose("Following (character,5) = %d\n", (int32_t)pos);
298 pos=ubrk_following(character, 18);
299 if(pos!=19)
300 log_err("error ubrk_following(character,18) did not return 19\n");
301 log_verbose("Followingcharacter,18) = %d\n", (int32_t)pos);
302 pos=ubrk_preceding(character, 22);
303 if(pos!=21)
304 log_err("error ubrk_preceding(character,22) did not return 21\n");
305 log_verbose("preceding(character,22) = %d\n", (int32_t)pos);
306
307
308 log_verbose("\nTesting the functions for line\n");
309 pos=ubrk_first(line);
310 if(pos != 0)
311 log_err("error ubrk_first(line) returned %d, expected 0\n", (int32_t)pos);
312 pos = ubrk_next(line);
313 pos=ubrk_following(line, 18);
314 if(pos!=22)
315 log_err("error ubrk_following(line) did not return 22\n");
316 log_verbose("following (line) = %d\n", (int32_t)pos);
317
318
319 log_verbose("\nTesting the functions for sentence\n");
320 pos = ubrk_first(sentence);
321 pos = ubrk_current(sentence);
322 log_verbose("Current(sentence) = %d\n", (int32_t)pos);
323 pos = ubrk_last(sentence);
324 if(pos!=49)
325 log_err("error ubrk_last for sentence did not return 49\n");
326 log_verbose("Last (sentence) = %d\n", (int32_t)pos);
327 pos = ubrk_first(sentence);
328 to = ubrk_following( sentence, 0 );
329 if (to == 0) log_err("ubrk_following returned 0\n");
330 to = ubrk_preceding( sentence, to );
331 if (to != 0) log_err("ubrk_preceding didn't return 0\n");
332 if (ubrk_first(sentence)!=ubrk_current(sentence)) {
333 log_err("error in ubrk_first() or ubrk_current()\n");
334 }
335
336
337 /*---- */
338 /*Testing ubrk_open and ubrk_close()*/
339 log_verbose("\nTesting open and close for us locale\n");
340 b = ubrk_open(UBRK_WORD, "fr_FR", text, u_strlen(text), &status);
341 if (U_FAILURE(status)) {
342 log_err("ubrk_open for word returned NULL: %s\n", myErrorName(status));
343 }
344 ubrk_close(b);
345
346 /* Test setText and setUText */
347 {
348 UChar s1[] = {0x41, 0x42, 0x20, 0};
349 UChar s2[] = {0x41, 0x42, 0x43, 0x44, 0x45, 0};
350 UText *ut = NULL;
351 UBreakIterator *bb;
352 int j;
353
354 log_verbose("\nTesting ubrk_setText() and ubrk_setUText()\n");
355 status = U_ZERO_ERROR;
356 bb = ubrk_open(UBRK_WORD, "en_US", NULL, 0, &status);
357 TEST_ASSERT_SUCCESS(status);
358 ubrk_setText(bb, s1, -1, &status);
359 TEST_ASSERT_SUCCESS(status);
360 ubrk_first(bb);
361 j = ubrk_next(bb);
362 TEST_ASSERT(j == 2);
363 ut = utext_openUChars(ut, s2, -1, &status);
364 ubrk_setUText(bb, ut, &status);
365 TEST_ASSERT_SUCCESS(status);
366 j = ubrk_next(bb);
367 TEST_ASSERT(j == 5);
368
369 ubrk_close(bb);
370 utext_close(ut);
371 }
372
373 ubrk_close(word);
374 ubrk_close(sentence);
375 ubrk_close(line);
376 ubrk_close(character);
377 }
378
TestBreakIteratorSafeClone(void)379 static void TestBreakIteratorSafeClone(void)
380 {
381 UChar text[51]; /* Keep this odd to test for 64-bit memory alignment */
382 /* NOTE: This doesn't reliably force misalignment of following items. */
383 uint8_t buffer [CLONETEST_ITERATOR_COUNT] [U_BRK_SAFECLONE_BUFFERSIZE];
384 int32_t bufferSize = U_BRK_SAFECLONE_BUFFERSIZE;
385
386 UBreakIterator * someIterators [CLONETEST_ITERATOR_COUNT];
387 UBreakIterator * someClonedIterators [CLONETEST_ITERATOR_COUNT];
388
389 UBreakIterator * brk;
390 UErrorCode status = U_ZERO_ERROR;
391 int32_t start,pos;
392 int32_t i;
393
394 /*Testing ubrk_safeClone */
395
396 /* Note: the adjacent "" are concatenating strings, not adding a \" to the
397 string, which is probably what whoever wrote this intended. Don't fix,
398 because it would throw off the hard coded break positions in the following
399 tests. */
400 u_uastrcpy(text, "He's from Africa. ""Mr. Livingston, I presume?"" Yeah");
401
402 /* US & Thai - rule-based & dictionary based */
403 someIterators[0] = ubrk_open(UBRK_WORD, "en_US", text, u_strlen(text), &status);
404 if(!someIterators[0] || U_FAILURE(status)) {
405 log_data_err("Couldn't open en_US word break iterator - %s\n", u_errorName(status));
406 return;
407 }
408
409 someIterators[1] = ubrk_open(UBRK_WORD, "th_TH", text, u_strlen(text), &status);
410 if(!someIterators[1] || U_FAILURE(status)) {
411 log_data_err("Couldn't open th_TH word break iterator - %s\n", u_errorName(status));
412 return;
413 }
414
415 /* test each type of iterator */
416 for (i = 0; i < CLONETEST_ITERATOR_COUNT; i++)
417 {
418
419 /* Check the various error & informational states */
420
421 /* Null status - just returns NULL */
422 if (NULL != ubrk_safeClone(someIterators[i], buffer[i], &bufferSize, NULL))
423 {
424 log_err("FAIL: Cloned Iterator failed to deal correctly with null status\n");
425 }
426 /* error status - should return 0 & keep error the same */
427 status = U_MEMORY_ALLOCATION_ERROR;
428 if (NULL != ubrk_safeClone(someIterators[i], buffer[i], &bufferSize, &status) || status != U_MEMORY_ALLOCATION_ERROR)
429 {
430 log_err("FAIL: Cloned Iterator failed to deal correctly with incoming error status\n");
431 }
432 status = U_ZERO_ERROR;
433
434 /* Null buffer size pointer is ok */
435 if (NULL == (brk = ubrk_safeClone(someIterators[i], buffer[i], NULL, &status)) || U_FAILURE(status))
436 {
437 log_err("FAIL: Cloned Iterator failed to deal correctly with null bufferSize pointer\n");
438 }
439 ubrk_close(brk);
440 status = U_ZERO_ERROR;
441
442 /* buffer size pointer is 0 - fill in pbufferSize with a size */
443 bufferSize = 0;
444 if (NULL != ubrk_safeClone(someIterators[i], buffer[i], &bufferSize, &status) ||
445 U_FAILURE(status) || bufferSize <= 0)
446 {
447 log_err("FAIL: Cloned Iterator failed a sizing request ('preflighting')\n");
448 }
449 /* Verify our define is large enough */
450 if (U_BRK_SAFECLONE_BUFFERSIZE < bufferSize)
451 {
452 log_err("FAIL: Pre-calculated buffer size is too small - %d but needed %d\n", U_BRK_SAFECLONE_BUFFERSIZE, bufferSize);
453 }
454 /* Verify we can use this run-time calculated size */
455 if (NULL == (brk = ubrk_safeClone(someIterators[i], buffer[i], &bufferSize, &status)) || U_FAILURE(status))
456 {
457 log_err("FAIL: Iterator can't be cloned with run-time size\n");
458 }
459 if (brk)
460 ubrk_close(brk);
461 /* size one byte too small - should allocate & let us know */
462 if (bufferSize > 1) {
463 --bufferSize;
464 }
465 if (NULL == (brk = ubrk_safeClone(someIterators[i], NULL, &bufferSize, &status)) || status != U_SAFECLONE_ALLOCATED_WARNING)
466 {
467 log_err("FAIL: Cloned Iterator failed to deal correctly with too-small buffer size\n");
468 }
469 if (brk)
470 ubrk_close(brk);
471 status = U_ZERO_ERROR;
472 bufferSize = U_BRK_SAFECLONE_BUFFERSIZE;
473
474 /* Null buffer pointer - return Iterator & set error to U_SAFECLONE_ALLOCATED_ERROR */
475 if (NULL == (brk = ubrk_safeClone(someIterators[i], NULL, &bufferSize, &status)) || status != U_SAFECLONE_ALLOCATED_WARNING)
476 {
477 log_err("FAIL: Cloned Iterator failed to deal correctly with null buffer pointer\n");
478 }
479 if (brk)
480 ubrk_close(brk);
481 status = U_ZERO_ERROR;
482
483 /* Mis-aligned buffer pointer. */
484 {
485 char stackBuf[U_BRK_SAFECLONE_BUFFERSIZE+sizeof(void *)];
486
487 brk = ubrk_safeClone(someIterators[i], &stackBuf[1], &bufferSize, &status);
488 if (U_FAILURE(status) || brk == NULL) {
489 log_err("FAIL: Cloned Iterator failed with misaligned buffer pointer\n");
490 }
491 if (status == U_SAFECLONE_ALLOCATED_WARNING) {
492 log_verbose("Cloned Iterator allocated when using a mis-aligned buffer.\n");
493 }
494 if (brk)
495 ubrk_close(brk);
496 }
497
498
499 /* Null Iterator - return NULL & set U_ILLEGAL_ARGUMENT_ERROR */
500 if (NULL != ubrk_safeClone(NULL, buffer[i], &bufferSize, &status) || status != U_ILLEGAL_ARGUMENT_ERROR)
501 {
502 log_err("FAIL: Cloned Iterator failed to deal correctly with null Iterator pointer\n");
503 }
504 status = U_ZERO_ERROR;
505
506 /* Do these cloned Iterators work at all - make a first & next call */
507 bufferSize = U_BRK_SAFECLONE_BUFFERSIZE;
508 someClonedIterators[i] = ubrk_safeClone(someIterators[i], buffer[i], &bufferSize, &status);
509
510 start = ubrk_first(someClonedIterators[i]);
511 if(start!=0)
512 log_err("error ubrk_start(clone) did not return 0\n");
513 pos=ubrk_next(someClonedIterators[i]);
514 if(pos!=4)
515 log_err("error ubrk_next(clone) did not return 4\n");
516
517 ubrk_close(someClonedIterators[i]);
518 ubrk_close(someIterators[i]);
519 }
520 }
521
TestBreakIteratorClone(void)522 static void TestBreakIteratorClone(void)
523 {
524 const UChar text[] = u"He's from Africa. Mr. Livingston, I presume? Yeah";
525 UBreakIterator * someIterators [CLONETEST_ITERATOR_COUNT];
526
527 UBreakIterator * brk;
528 UErrorCode status = U_ZERO_ERROR;
529 int32_t start,pos;
530 int32_t i;
531
532 /*Testing ubrk_clone */
533
534 /* US & Thai - rule-based & dictionary based */
535 someIterators[0] = ubrk_open(UBRK_WORD, "en_US", text, u_strlen(text), &status);
536 if(!someIterators[0] || U_FAILURE(status)) {
537 log_data_err("Couldn't open en_US word break iterator - %s\n", u_errorName(status));
538 return;
539 }
540
541 someIterators[1] = ubrk_open(UBRK_WORD, "th_TH", text, u_strlen(text), &status);
542 if(!someIterators[1] || U_FAILURE(status)) {
543 log_data_err("Couldn't open th_TH word break iterator - %s\n", u_errorName(status));
544 return;
545 }
546
547 /* test each type of iterator */
548 for (i = 0; i < CLONETEST_ITERATOR_COUNT; i++)
549 {
550 /* error status - should return 0 & keep error the same */
551 status = U_MEMORY_ALLOCATION_ERROR;
552 if (NULL != ubrk_clone(someIterators[i], &status) || status != U_MEMORY_ALLOCATION_ERROR)
553 {
554 log_err("FAIL: Cloned Iterator failed to deal correctly with incoming error status\n");
555 }
556
557 status = U_ZERO_ERROR;
558
559 /* Do these cloned Iterators work at all - make a first & next call */
560 brk = ubrk_clone(someIterators[i], &status);
561
562 start = ubrk_first(brk);
563 if(start!=0)
564 log_err("error ubrk_start(clone) did not return 0, but %i\n", start);
565 pos=ubrk_next(brk);
566 if(pos!=4)
567 log_err("error ubrk_next(clone) did not return 4, but %i\n", pos);
568
569 ubrk_close(brk);
570
571 pos = ubrk_next(someIterators[i]);
572 if (pos != 4) {
573 log_err("error ubrk_next(iter) did not return 4, but %i\n", pos);
574 }
575
576 brk = ubrk_clone(someIterators[i], &status);
577 // The text position should be kept in the new clone.
578 start = ubrk_current(brk);
579 if (start != 4) {
580 log_err("error ubrk_current(clone) did not return 4, but %i\n", start);
581 }
582
583 pos = ubrk_next(brk);
584 if (pos != 5) {
585 log_err("error ubrk_next(clone) did not return 5, but %i\n", pos);
586 }
587 start = ubrk_current(brk);
588 if (start != 5) {
589 log_err("error ubrk_current(clone) did not return 5, but %i\n", start);
590 }
591
592 start = ubrk_current(someIterators[i]);
593 if (start != 4) {
594 log_err("error ubrk_current(iter) did not keep the same position of 4,"
595 " but %i after advancing the position in its clone.\n", start);
596 }
597
598 ubrk_close(brk);
599
600 ubrk_close(someIterators[i]);
601 }
602 }
603 #endif
604
605
606 /*
607 // Open a break iterator from char * rules. Take care of conversion
608 // of the rules and error checking.
609 */
testOpenRules(char * rules)610 static UBreakIterator * testOpenRules(char *rules) {
611 UErrorCode status = U_ZERO_ERROR;
612 UChar *ruleSourceU = NULL;
613 void *strCleanUp = NULL;
614 UParseError parseErr;
615 UBreakIterator *bi;
616
617 ruleSourceU = toUChar(rules, &strCleanUp);
618
619 bi = ubrk_openRules(ruleSourceU, -1, /* The rules */
620 NULL, -1, /* The text to be iterated over. */
621 &parseErr, &status);
622
623 if (U_FAILURE(status)) {
624 log_data_err("FAIL: ubrk_openRules: ICU Error \"%s\" (Are you missing data?)\n", u_errorName(status));
625 bi = 0;
626 }
627 freeToUCharStrings(&strCleanUp);
628 return bi;
629
630 }
631
632 /*
633 * TestBreakIteratorRules - Verify that a break iterator can be created from
634 * a set of source rules.
635 */
TestBreakIteratorRules()636 static void TestBreakIteratorRules() {
637 /* Rules will keep together any run of letters not including 'a', OR
638 * keep together 'abc', but only when followed by 'def', OTHERWISE
639 * just return one char at a time.
640 */
641 char rules[] = "abc/def{666};\n [\\p{L} - [a]]* {2}; . {1};";
642 /* 0123456789012345678 */
643 char data[] = "abcdex abcdefgh-def"; /* the test data string */
644 char breaks[] = "** ** * ** *"; /* * the expected break positions */
645 char tags[] = "01 21 6 21 2"; /* expected tag values at break positions */
646 int32_t tagMap[] = {0, 1, 2, 3, 4, 5, 666};
647
648 UChar *uData;
649 void *freeHook = NULL;
650 UErrorCode status = U_ZERO_ERROR;
651 int32_t pos;
652 int i;
653
654 UBreakIterator *bi = testOpenRules(rules);
655 if (bi == NULL) {return;}
656 uData = toUChar(data, &freeHook);
657 ubrk_setText(bi, uData, -1, &status);
658
659 pos = ubrk_first(bi);
660 for (i=0; i<(int)sizeof(breaks); i++) {
661 if (pos == i && breaks[i] != '*') {
662 log_err("FAIL: unexpected break at position %d found\n", pos);
663 break;
664 }
665 if (pos != i && breaks[i] == '*') {
666 log_err("FAIL: expected break at position %d not found.\n", i);
667 break;
668 }
669 if (pos == i) {
670 int32_t tag, expectedTag;
671 tag = ubrk_getRuleStatus(bi);
672 expectedTag = tagMap[tags[i]&0xf];
673 if (tag != expectedTag) {
674 log_err("FAIL: incorrect tag value. Position = %d; expected tag %d, got %d",
675 pos, expectedTag, tag);
676 break;
677 }
678 pos = ubrk_next(bi);
679 }
680 }
681
682 /* #12914 add basic sanity test for ubrk_getBinaryRules, ubrk_openBinaryRules */
683 /* Underlying functionality checked in C++ rbbiapts.cpp TestRoundtripRules */
684 status = U_ZERO_ERROR;
685 int32_t rulesLength = ubrk_getBinaryRules(bi, NULL, 0, &status); /* preflight */
686 if (U_FAILURE(status)) {
687 log_err("FAIL: ubrk_getBinaryRules preflight err: %s", u_errorName(status));
688 } else {
689 uint8_t* binaryRules = (uint8_t*)uprv_malloc(rulesLength);
690 if (binaryRules == NULL) {
691 log_err("FAIL: unable to malloc rules buffer, size %u", rulesLength);
692 } else {
693 rulesLength = ubrk_getBinaryRules(bi, binaryRules, rulesLength, &status);
694 if (U_FAILURE(status)) {
695 log_err("FAIL: ubrk_getBinaryRules err: %s", u_errorName(status));
696 } else {
697 UBreakIterator* bi2 = ubrk_openBinaryRules(binaryRules, rulesLength, uData, -1, &status);
698 if (U_FAILURE(status)) {
699 log_err("FAIL: ubrk_openBinaryRules err: %s", u_errorName(status));
700 } else {
701 int32_t maxCount = sizeof(breaks); /* fail-safe test limit */
702 int32_t pos2 = ubrk_first(bi2);
703 pos = ubrk_first(bi);
704 do {
705 if (pos2 != pos) {
706 log_err("FAIL: iterator from ubrk_openBinaryRules does not match original, get pos = %d instead of %d", pos2, pos);
707 }
708 pos2 = ubrk_next(bi2);
709 pos = ubrk_next(bi);
710 } while ((pos != UBRK_DONE || pos2 != UBRK_DONE) && maxCount-- > 0);
711
712 ubrk_close(bi2);
713 }
714 }
715 uprv_free(binaryRules);
716 }
717 }
718
719 freeToUCharStrings(&freeHook);
720 ubrk_close(bi);
721 }
722
TestBreakIteratorRuleError()723 static void TestBreakIteratorRuleError() {
724 /*
725 * TestBreakIteratorRuleError - Try to create a BI from rules with syntax errors,
726 * check that the error is reported correctly.
727 */
728 char rules[] = " # This is a rule comment on line 1\n"
729 "[:L:]; # this rule is OK.\n"
730 "abcdefg); # Error, mismatched parens\n";
731 UChar *uRules;
732 void *freeHook = NULL;
733 UErrorCode status = U_ZERO_ERROR;
734 UParseError parseErr;
735 UBreakIterator *bi;
736
737 uRules = toUChar(rules, &freeHook);
738 bi = ubrk_openRules(uRules, -1, /* The rules */
739 NULL, -1, /* The text to be iterated over. */
740 &parseErr, &status);
741 if (U_SUCCESS(status)) {
742 log_err("FAIL: construction of break iterator succeeded when it should have failed.\n");
743 ubrk_close(bi);
744 } else {
745 if (parseErr.line != 3 || parseErr.offset != 8) {
746 log_data_err("FAIL: incorrect error position reported. Got line %d, char %d, expected line 3, char 7 (Are you missing data?)\n",
747 parseErr.line, parseErr.offset);
748 }
749 }
750 freeToUCharStrings(&freeHook);
751 }
752
753
754 /*
755 * TestsBreakIteratorStatusVals() Test the ubrk_getRuleStatusVec() function
756 */
TestBreakIteratorStatusVec()757 static void TestBreakIteratorStatusVec() {
758 #define RULE_STRING_LENGTH 200
759 UChar rules[RULE_STRING_LENGTH];
760
761 #define TEST_STRING_LENGTH 25
762 UChar testString[TEST_STRING_LENGTH];
763 UBreakIterator *bi = NULL;
764 int32_t pos = 0;
765 int32_t vals[10];
766 int32_t numVals;
767 UErrorCode status = U_ZERO_ERROR;
768
769 u_uastrncpy(rules, "[A-N]{100}; \n"
770 "[a-w]{200}; \n"
771 "[\\p{L}]{300}; \n"
772 "[\\p{N}]{400}; \n"
773 "[0-5]{500}; \n"
774 "!.*;\n", RULE_STRING_LENGTH);
775 u_uastrncpy(testString, "ABC", TEST_STRING_LENGTH);
776
777
778 bi = ubrk_openRules(rules, -1, testString, -1, NULL, &status);
779 TEST_ASSERT_SUCCESS(status);
780 TEST_ASSERT(bi != NULL);
781
782 /* The TEST_ASSERT above should change too... */
783 if (bi != NULL) {
784 pos = ubrk_next(bi);
785 TEST_ASSERT(pos == 1);
786
787 memset(vals, -1, sizeof(vals));
788 numVals = ubrk_getRuleStatusVec(bi, vals, 10, &status);
789 TEST_ASSERT_SUCCESS(status);
790 TEST_ASSERT(numVals == 2);
791 TEST_ASSERT(vals[0] == 100);
792 TEST_ASSERT(vals[1] == 300);
793 TEST_ASSERT(vals[2] == -1);
794
795 numVals = ubrk_getRuleStatusVec(bi, vals, 0, &status);
796 TEST_ASSERT(status == U_BUFFER_OVERFLOW_ERROR);
797 TEST_ASSERT(numVals == 2);
798 }
799
800 ubrk_close(bi);
801 }
802
803
804 /*
805 * static void TestBreakIteratorUText(void);
806 *
807 * Test that ubrk_setUText() is present and works for a simple case.
808 */
TestBreakIteratorUText(void)809 static void TestBreakIteratorUText(void) {
810 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 */
811 /* 0 1 2 34567890 */
812
813 UErrorCode status = U_ZERO_ERROR;
814 UBreakIterator *bi = NULL;
815 int32_t pos = 0;
816
817
818 UText *ut = utext_openUTF8(NULL, UTF8Str, -1, &status);
819 TEST_ASSERT_SUCCESS(status);
820
821 bi = ubrk_open(UBRK_WORD, "en_US", NULL, 0, &status);
822 if (U_FAILURE(status)) {
823 log_err_status(status, "Failure at file %s, line %d, error = %s\n", __FILE__, __LINE__, u_errorName(status));
824 utext_close(ut);
825 return;
826 }
827
828 ubrk_setUText(bi, ut, &status);
829 if (U_FAILURE(status)) {
830 log_err("Failure at file %s, line %d, error = %s\n", __FILE__, __LINE__, u_errorName(status));
831 ubrk_close(bi);
832 utext_close(ut);
833 return;
834 }
835
836 pos = ubrk_first(bi);
837 TEST_ASSERT(pos == 0);
838
839 pos = ubrk_next(bi);
840 TEST_ASSERT(pos == 4);
841
842 pos = ubrk_next(bi);
843 TEST_ASSERT(pos == 5);
844
845 pos = ubrk_next(bi);
846 TEST_ASSERT(pos == 10);
847
848 pos = ubrk_next(bi);
849 TEST_ASSERT(pos == UBRK_DONE);
850 ubrk_close(bi);
851 utext_close(ut);
852 }
853
854 /*
855 * static void TestBreakIteratorTailoring(void);
856 *
857 * Test break iterator tailorings from CLDR data.
858 */
859
860 /* Thai/Lao grapheme break tailoring */
861 static const UChar thTest[] = { 0x0020, 0x0E40, 0x0E01, 0x0020,
862 0x0E01, 0x0E30, 0x0020, 0x0E01, 0x0E33, 0x0020, 0 };
863 /*in Unicode 6.1 en should behave just like th for this*/
864 /*static const int32_t thTestOffs_enFwd[] = { 1, 3, 4, 6, 7, 9, 10 };*/
865 static const int32_t thTestOffs_thFwd[] = { 1, 2, 3, 4, 5, 6, 7, 9, 10 };
866 /*static const int32_t thTestOffs_enRev[] = { 9, 7, 6, 4, 3, 1, 0 };*/
867 static const int32_t thTestOffs_thRev[] = { 9, 7, 6, 5, 4, 3, 2, 1, 0 };
868
869 /* Hebrew line break tailoring, for cldrbug 3028 */
870 static const UChar heTest[] = { 0x0020, 0x002D, 0x0031, 0x0032, 0x0020,
871 0x0061, 0x002D, 0x006B, 0x0020,
872 0x0061, 0x0300, 0x2010, 0x006B, 0x0020,
873 0x05DE, 0x05D4, 0x002D, 0x0069, 0x0020,
874 0x05D1, 0x05BC, 0x2010, 0x0047, 0x0020, 0 };
875 /*in Unicode 6.1 en should behave just like he for this*/
876 /*static const int32_t heTestOffs_enFwd[] = { 1, 5, 7, 9, 12, 14, 17, 19, 22, 24 };*/
877 static const int32_t heTestOffs_heFwd[] = { 1, 5, 7, 9, 12, 14, 19, 24 };
878 /*static const int32_t heTestOffs_enRev[] = { 22, 19, 17, 14, 12, 9, 7, 5, 1, 0 };*/
879 static const int32_t heTestOffs_heRev[] = { 19, 14, 12, 9, 7, 5, 1, 0 };
880
881 /* Finnish line break tailoring, for cldrbug 3029.
882 * As of ICU 63, Finnish tailoring moved to root, Finnish and English should be the same. */
883 static const UChar fiTest[] = { /* 00 */ 0x0020, 0x002D, 0x0031, 0x0032, 0x0020,
884 /* 05 */ 0x0061, 0x002D, 0x006B, 0x0020,
885 /* 09 */ 0x0061, 0x0300, 0x2010, 0x006B, 0x0020,
886 /* 14 */ 0x0061, 0x0020, 0x002D, 0x006B, 0x0020,
887 /* 19 */ 0x0061, 0x0300, 0x0020, 0x2010, 0x006B, 0x0020, 0 };
888 //static const int32_t fiTestOffs_enFwd[] = { 1, 5, 7, 9, 12, 14, 16, 17, 19, 22, 23, 25 };
889 static const int32_t fiTestOffs_enFwd[] = { 1, 5, 7, 9, 12, 14, 16, 19, 22, 25 };
890 static const int32_t fiTestOffs_fiFwd[] = { 1, 5, 7, 9, 12, 14, 16, 19, 22, 25 };
891 //static const int32_t fiTestOffs_enRev[] = { 23, 22, 19, 17, 16, 14, 12, 9, 7, 5, 1, 0 };
892 static const int32_t fiTestOffs_enRev[] = { 22, 19, 16, 14, 12, 9, 7, 5, 1, 0 };
893 static const int32_t fiTestOffs_fiRev[] = { 22, 19, 16, 14, 12, 9, 7, 5, 1, 0 };
894
895 /* Khmer dictionary-based work break, for ICU ticket #8329 */
896 static const UChar kmTest[] = { /* 00 */ 0x179F, 0x17BC, 0x1798, 0x1785, 0x17C6, 0x178E, 0x17B6, 0x1799, 0x1796, 0x17C1,
897 /* 10 */ 0x179B, 0x1794, 0x1793, 0x17D2, 0x178F, 0x17B7, 0x1785, 0x178A, 0x17BE, 0x1798,
898 /* 20 */ 0x17D2, 0x1794, 0x17B8, 0x17A2, 0x1792, 0x17B7, 0x179F, 0x17D2, 0x178B, 0x17B6,
899 /* 30 */ 0x1793, 0x17A2, 0x179A, 0x1796, 0x17D2, 0x179A, 0x17C7, 0x1782, 0x17BB, 0x178E,
900 /* 40 */ 0x178A, 0x179B, 0x17CB, 0x1796, 0x17D2, 0x179A, 0x17C7, 0x17A2, 0x1784, 0x17D2,
901 /* 50 */ 0x1782, 0 };
902 static const int32_t kmTestOffs_kmFwd[] = { 3, /*8,*/ 11, 17, 23, 31, /*33,*/ 40, 43, 51 }; /* TODO: Investigate failure to break at offset 8 */
903 static const int32_t kmTestOffs_kmRev[] = { 43, 40, /*33,*/ 31, 23, 17, 11, /*8,*/ 3, 0 };
904
905 typedef struct {
906 const char * locale;
907 UBreakIteratorType type;
908 const UChar * test;
909 const int32_t * offsFwd;
910 const int32_t * offsRev;
911 int32_t numOffsets;
912 } RBBITailoringTest;
913
914 static const RBBITailoringTest tailoringTests[] = {
915 { "en", UBRK_CHARACTER, thTest, thTestOffs_thFwd, thTestOffs_thRev, UPRV_LENGTHOF(thTestOffs_thFwd) },
916 { "en_US_POSIX", UBRK_CHARACTER, thTest, thTestOffs_thFwd, thTestOffs_thRev, UPRV_LENGTHOF(thTestOffs_thFwd) },
917 { "en", UBRK_LINE, heTest, heTestOffs_heFwd, heTestOffs_heRev, UPRV_LENGTHOF(heTestOffs_heFwd) },
918 { "he", UBRK_LINE, heTest, heTestOffs_heFwd, heTestOffs_heRev, UPRV_LENGTHOF(heTestOffs_heFwd) },
919 { "en", UBRK_LINE, fiTest, fiTestOffs_enFwd, fiTestOffs_enRev, UPRV_LENGTHOF(fiTestOffs_enFwd) },
920 { "fi", UBRK_LINE, fiTest, fiTestOffs_fiFwd, fiTestOffs_fiRev, UPRV_LENGTHOF(fiTestOffs_fiFwd) },
921 { "km", UBRK_WORD, kmTest, kmTestOffs_kmFwd, kmTestOffs_kmRev, UPRV_LENGTHOF(kmTestOffs_kmFwd) },
922 { NULL, 0, NULL, NULL, NULL, 0 },
923 };
924
TestBreakIteratorTailoring(void)925 static void TestBreakIteratorTailoring(void) {
926 const RBBITailoringTest * testPtr;
927 for (testPtr = tailoringTests; testPtr->locale != NULL; ++testPtr) {
928 UErrorCode status = U_ZERO_ERROR;
929 UBreakIterator* ubrkiter = ubrk_open(testPtr->type, testPtr->locale, testPtr->test, -1, &status);
930 if ( U_SUCCESS(status) ) {
931 int32_t offset, offsindx;
932 UBool foundError;
933
934 foundError = false;
935 for (offsindx = 0; (offset = ubrk_next(ubrkiter)) != UBRK_DONE; ++offsindx) {
936 if (!foundError && offsindx >= testPtr->numOffsets) {
937 log_err("FAIL: locale %s, break type %d, ubrk_next expected UBRK_DONE, got %d\n",
938 testPtr->locale, testPtr->type, offset);
939 foundError = true;
940 } else if (!foundError && offset != testPtr->offsFwd[offsindx]) {
941 log_err("FAIL: locale %s, break type %d, ubrk_next expected %d, got %d\n",
942 testPtr->locale, testPtr->type, testPtr->offsFwd[offsindx], offset);
943 foundError = true;
944 }
945 }
946 if (!foundError && offsindx < testPtr->numOffsets) {
947 log_err("FAIL: locale %s, break type %d, ubrk_next expected %d, got UBRK_DONE\n",
948 testPtr->locale, testPtr->type, testPtr->offsFwd[offsindx]);
949 }
950
951 foundError = false;
952 for (offsindx = 0; (offset = ubrk_previous(ubrkiter)) != UBRK_DONE; ++offsindx) {
953 if (!foundError && offsindx >= testPtr->numOffsets) {
954 log_err("FAIL: locale %s, break type %d, ubrk_previous expected UBRK_DONE, got %d\n",
955 testPtr->locale, testPtr->type, offset);
956 foundError = true;
957 } else if (!foundError && offset != testPtr->offsRev[offsindx]) {
958 log_err("FAIL: locale %s, break type %d, ubrk_previous expected %d, got %d\n",
959 testPtr->locale, testPtr->type, testPtr->offsRev[offsindx], offset);
960 foundError = true;
961 }
962 }
963 if (!foundError && offsindx < testPtr->numOffsets) {
964 log_err("FAIL: locale %s, break type %d, ubrk_previous expected %d, got UBRK_DONE\n",
965 testPtr->locale, testPtr->type, testPtr->offsRev[offsindx]);
966 }
967
968 ubrk_close(ubrkiter);
969 } else {
970 log_err_status(status, "FAIL: locale %s, break type %d, ubrk_open status: %s\n", testPtr->locale, testPtr->type, u_errorName(status));
971 }
972 }
973 }
974
975
TestBreakIteratorRefresh(void)976 static void TestBreakIteratorRefresh(void) {
977 /*
978 * RefreshInput changes out the input of a Break Iterator without
979 * changing anything else in the iterator's state. Used with Java JNI,
980 * when Java moves the underlying string storage. This test
981 * runs a ubrk_next() repeatedly, moving the text in the middle of the sequence.
982 * The right set of boundaries should still be found.
983 */
984 UChar testStr[] = {0x20, 0x41, 0x20, 0x42, 0x20, 0x43, 0x20, 0x44, 0x0}; /* = " A B C D" */
985 UChar movedStr[] = {0x20, 0x20, 0x20, 0x20, 0x20, 0x20, 0x20, 0x20, 0};
986 UErrorCode status = U_ZERO_ERROR;
987 UBreakIterator *bi;
988 UText ut1 = UTEXT_INITIALIZER;
989 UText ut2 = UTEXT_INITIALIZER;
990
991 bi = ubrk_open(UBRK_LINE, "en_US", NULL, 0, &status);
992 TEST_ASSERT_SUCCESS(status);
993 if (U_FAILURE(status)) {
994 return;
995 }
996
997 utext_openUChars(&ut1, testStr, -1, &status);
998 TEST_ASSERT_SUCCESS(status);
999 ubrk_setUText(bi, &ut1, &status);
1000 TEST_ASSERT_SUCCESS(status);
1001
1002 if (U_SUCCESS(status)) {
1003 /* Line boundaries will occur before each letter in the original string */
1004 TEST_ASSERT(1 == ubrk_next(bi));
1005 TEST_ASSERT(3 == ubrk_next(bi));
1006
1007 /* Move the string, kill the original string. */
1008 u_strcpy(movedStr, testStr);
1009 u_memset(testStr, 0x20, u_strlen(testStr));
1010 utext_openUChars(&ut2, movedStr, -1, &status);
1011 TEST_ASSERT_SUCCESS(status);
1012 ubrk_refreshUText(bi, &ut2, &status);
1013 TEST_ASSERT_SUCCESS(status);
1014
1015 /* Find the following matches, now working in the moved string. */
1016 TEST_ASSERT(5 == ubrk_next(bi));
1017 TEST_ASSERT(7 == ubrk_next(bi));
1018 TEST_ASSERT(8 == ubrk_next(bi));
1019 TEST_ASSERT(UBRK_DONE == ubrk_next(bi));
1020 TEST_ASSERT_SUCCESS(status);
1021
1022 utext_close(&ut1);
1023 utext_close(&ut2);
1024 }
1025 ubrk_close(bi);
1026 }
1027
1028
TestBug11665(void)1029 static void TestBug11665(void) {
1030 // The problem was with the incorrect breaking of Japanese text beginning
1031 // with Katakana characters when no prior Japanese or Chinese text had been
1032 // encountered.
1033 //
1034 // Tested here in cintltst, rather than in intltest, because only cintltst
1035 // tests have the ability to reset ICU, which is needed to get the bug
1036 // to manifest itself.
1037
1038 static UChar japaneseText[] = {0x30A2, 0x30EC, 0x30EB, 0x30AE, 0x30FC, 0x6027, 0x7D50, 0x819C, 0x708E};
1039 int32_t boundaries[10] = {0};
1040 UBreakIterator *bi = NULL;
1041 int32_t brk;
1042 int32_t brkIdx = 0;
1043 int32_t totalBreaks = 0;
1044 UErrorCode status = U_ZERO_ERROR;
1045
1046 ctest_resetICU();
1047 bi = ubrk_open(UBRK_WORD, "en_US", japaneseText, UPRV_LENGTHOF(japaneseText), &status);
1048 TEST_ASSERT_SUCCESS(status);
1049 if (!bi) {
1050 return;
1051 }
1052 for (brk=ubrk_first(bi); brk != UBRK_DONE; brk=ubrk_next(bi)) {
1053 boundaries[brkIdx] = brk;
1054 if (++brkIdx >= UPRV_LENGTHOF(boundaries) - 1) {
1055 break;
1056 }
1057 }
1058 if (brkIdx <= 2 || brkIdx >= UPRV_LENGTHOF(boundaries)) {
1059 log_err("%s:%d too few or many breaks found.\n", __FILE__, __LINE__);
1060 } else {
1061 totalBreaks = brkIdx;
1062 brkIdx = 0;
1063 for (brk=ubrk_first(bi); brk != UBRK_DONE; brk=ubrk_next(bi)) {
1064 if (brk != boundaries[brkIdx]) {
1065 log_err("%s:%d Break #%d differs between first and second iteration.\n", __FILE__, __LINE__, brkIdx);
1066 break;
1067 }
1068 if (++brkIdx >= UPRV_LENGTHOF(boundaries) - 1) {
1069 log_err("%s:%d Too many breaks.\n", __FILE__, __LINE__);
1070 break;
1071 }
1072 }
1073 if (totalBreaks != brkIdx) {
1074 log_err("%s:%d Number of breaks differ between first and second iteration.\n", __FILE__, __LINE__);
1075 }
1076 }
1077 ubrk_close(bi);
1078 }
1079
1080 /*
1081 * expOffset is the set of expected offsets, ending with '-1'.
1082 * "Expected expOffset -1" means "expected the end of the offsets"
1083 */
1084
1085 static const char testSentenceSuppressionsEn[] = "Mr. Jones comes home. Dr. Smith Ph.D. is out. In the U.S.A. it is hot.";
1086 static const int32_t testSentSuppFwdOffsetsEn[] = { 22, 26, 46, 70, -1 }; /* With suppressions, currently not handling Dr. */
1087 static const int32_t testSentFwdOffsetsEn[] = { 4, 22, 26, 46, 70, -1 }; /* Without suppressions */
1088 static const int32_t testSentSuppRevOffsetsEn[] = { 46, 26, 22, 0, -1 }; /* With suppressions, currently not handling Dr. */
1089 static const int32_t testSentRevOffsetsEn[] = { 46, 26, 22, 4, 0, -1 }; /* Without suppressions */
1090
1091 static const char testSentenceSuppressionsDe[] = "Wenn ich schon h\\u00F6re zu Guttenberg kommt evtl. zur\\u00FCck.";
1092 static const int32_t testSentSuppFwdOffsetsDe[] = { 53, -1 }; /* With suppressions */
1093 static const int32_t testSentFwdOffsetsDe[] = { 53, -1 }; /* Without suppressions; no break in evtl. zur due to casing */
1094 static const int32_t testSentSuppRevOffsetsDe[] = { 0, -1 }; /* With suppressions */
1095 static const int32_t testSentRevOffsetsDe[] = { 0, -1 }; /* Without suppressions */
1096
1097 static const char testSentenceSuppressionsEs[] = "Te esperamos todos los miercoles en Bravo 416, Col. El Pueblo a las 7 PM.";
1098 static const int32_t testSentSuppFwdOffsetsEs[] = { 73, -1 }; /* With suppressions */
1099 static const int32_t testSentFwdOffsetsEs[] = { 52, 73, -1 }; /* Without suppressions */
1100 static const int32_t testSentSuppRevOffsetsEs[] = { 0, -1 }; /* With suppressions */
1101 static const int32_t testSentRevOffsetsEs[] = { 52, 0, -1 }; /* Without suppressions */
1102
1103 enum { kTextULenMax = 128 };
1104
1105 typedef struct {
1106 const char * locale;
1107 const char * text;
1108 const int32_t * expFwdOffsets;
1109 const int32_t * expRevOffsets;
1110 } TestBISuppressionsItem;
1111
1112 static const TestBISuppressionsItem testBISuppressionsItems[] = {
1113 { "en@ss=standard", testSentenceSuppressionsEn, testSentSuppFwdOffsetsEn, testSentSuppRevOffsetsEn },
1114 { "en", testSentenceSuppressionsEn, testSentFwdOffsetsEn, testSentRevOffsetsEn },
1115 { "en_CA", testSentenceSuppressionsEn, testSentFwdOffsetsEn, testSentRevOffsetsEn },
1116 { "en_CA@ss=standard", testSentenceSuppressionsEn, testSentSuppFwdOffsetsEn, testSentSuppRevOffsetsEn },
1117 { "fr@ss=standard", testSentenceSuppressionsEn, testSentFwdOffsetsEn, testSentRevOffsetsEn },
1118 { "af@ss=standard", testSentenceSuppressionsEn, testSentFwdOffsetsEn, testSentRevOffsetsEn }, /* no brkiter data => nosuppressions? */
1119 { "af_ZA@ss=standard", testSentenceSuppressionsEn, testSentFwdOffsetsEn, testSentRevOffsetsEn }, /* no brkiter data => nosuppressions? */
1120 { "zh@ss=standard", testSentenceSuppressionsEn, testSentFwdOffsetsEn, testSentRevOffsetsEn }, /* brkiter data, no suppressions data => no suppressions */
1121 { "zh_Hant@ss=standard", testSentenceSuppressionsEn, testSentFwdOffsetsEn, testSentRevOffsetsEn }, /* brkiter data, no suppressions data => no suppressions */
1122 { "fi@ss=standard", testSentenceSuppressionsEn, testSentFwdOffsetsEn, testSentRevOffsetsEn }, /* brkiter data, no suppressions data => no suppressions */
1123 { "ja@ss=standard", testSentenceSuppressionsEn, testSentFwdOffsetsEn, testSentRevOffsetsEn }, /* brkiter data, no suppressions data => no suppressions */
1124 { "de@ss=standard", testSentenceSuppressionsDe, testSentSuppFwdOffsetsDe, testSentSuppRevOffsetsDe },
1125 { "de", testSentenceSuppressionsDe, testSentFwdOffsetsDe, testSentRevOffsetsDe },
1126 { "es@ss=standard", testSentenceSuppressionsEs, testSentSuppFwdOffsetsEs, testSentSuppRevOffsetsEs },
1127 { "es", testSentenceSuppressionsEs, testSentFwdOffsetsEs, testSentRevOffsetsEs },
1128 { NULL, NULL, NULL, NULL }
1129 };
1130
TestBreakIteratorSuppressions(void)1131 static void TestBreakIteratorSuppressions(void) {
1132 const TestBISuppressionsItem * itemPtr;
1133
1134 for (itemPtr = testBISuppressionsItems; itemPtr->locale != NULL; itemPtr++) {
1135 UChar textU[kTextULenMax];
1136 int32_t textULen = u_unescape(itemPtr->text, textU, kTextULenMax);
1137 UErrorCode status = U_ZERO_ERROR;
1138 UBreakIterator *bi = ubrk_open(UBRK_SENTENCE, itemPtr->locale, textU, textULen, &status);
1139 log_verbose("#%d: %s\n", (itemPtr-testBISuppressionsItems), itemPtr->locale);
1140 if (U_SUCCESS(status)) {
1141 int32_t offset, start;
1142 const int32_t * expOffsetPtr;
1143 const int32_t * expOffsetStart;
1144
1145 expOffsetStart = expOffsetPtr = itemPtr->expFwdOffsets;
1146 ubrk_first(bi);
1147 for (; (offset = ubrk_next(bi)) != UBRK_DONE && *expOffsetPtr >= 0; expOffsetPtr++) {
1148 if (offset != *expOffsetPtr) {
1149 log_err("FAIL: ubrk_next loc \"%s\", expected %d, got %d\n", itemPtr->locale, *expOffsetPtr, offset);
1150 }
1151 }
1152 if (offset != UBRK_DONE || *expOffsetPtr >= 0) {
1153 log_err("FAIL: ubrk_next loc \"%s\", expected UBRK_DONE & expOffset -1, got %d and %d\n", itemPtr->locale, offset, *expOffsetPtr);
1154 }
1155
1156 expOffsetStart = expOffsetPtr = itemPtr->expFwdOffsets;
1157 start = ubrk_first(bi) + 1;
1158 for (; (offset = ubrk_following(bi, start)) != UBRK_DONE && *expOffsetPtr >= 0; expOffsetPtr++) {
1159 if (offset != *expOffsetPtr) {
1160 log_err("FAIL: ubrk_following(%d) loc \"%s\", expected %d, got %d\n", start, itemPtr->locale, *expOffsetPtr, offset);
1161 }
1162 start = *expOffsetPtr + 1;
1163 }
1164 if (offset != UBRK_DONE || *expOffsetPtr >= 0) {
1165 log_err("FAIL: ubrk_following(%d) loc \"%s\", expected UBRK_DONE & expOffset -1, got %d and %d\n", start, itemPtr->locale, offset, *expOffsetPtr);
1166 }
1167
1168 expOffsetStart = expOffsetPtr = itemPtr->expRevOffsets;
1169 offset = ubrk_last(bi);
1170 log_verbose("___ @%d ubrk_last\n", offset);
1171 if(offset == 0) {
1172 log_err("FAIL: ubrk_last loc \"%s\" unexpected %d\n", itemPtr->locale, offset);
1173 }
1174 for (; (offset = ubrk_previous(bi)) != UBRK_DONE && *expOffsetPtr >= 0; expOffsetPtr++) {
1175 if (offset != *expOffsetPtr) {
1176 log_err("FAIL: ubrk_previous loc \"%s\", expected %d, got %d\n", itemPtr->locale, *expOffsetPtr, offset);
1177 } else {
1178 log_verbose("[%d] @%d ubrk_previous()\n", (expOffsetPtr - expOffsetStart), offset);
1179 }
1180 }
1181 if (offset != UBRK_DONE || *expOffsetPtr >= 0) {
1182 log_err("FAIL: ubrk_previous loc \"%s\", expected UBRK_DONE & expOffset[%d] -1, got %d and %d\n", itemPtr->locale,
1183 expOffsetPtr - expOffsetStart,
1184 offset, *expOffsetPtr);
1185 }
1186
1187 expOffsetStart = expOffsetPtr = itemPtr->expRevOffsets;
1188 start = ubrk_last(bi) - 1;
1189 for (; (offset = ubrk_preceding(bi, start)) != UBRK_DONE && *expOffsetPtr >= 0; expOffsetPtr++) {
1190 if (offset != *expOffsetPtr) {
1191 log_err("FAIL: ubrk_preceding(%d) loc \"%s\", expected %d, got %d\n", start, itemPtr->locale, *expOffsetPtr, offset);
1192 }
1193 start = *expOffsetPtr - 1;
1194 }
1195 if (start >=0 && (offset != UBRK_DONE || *expOffsetPtr >= 0)) {
1196 log_err("FAIL: ubrk_preceding loc(%d) \"%s\", expected UBRK_DONE & expOffset -1, got %d and %d\n", start, itemPtr->locale, offset, *expOffsetPtr);
1197 }
1198
1199 ubrk_close(bi);
1200 } else {
1201 log_data_err("FAIL: ubrk_open(UBRK_SENTENCE, \"%s\", ...) status %s (Are you missing data?)\n", itemPtr->locale, u_errorName(status));
1202 }
1203 }
1204 }
1205
1206
1207 #endif /* #if !UCONFIG_NO_BREAK_ITERATION */
1208