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