rbbiapts.cpp revision 4fceb0aeb072e9c6879c37fbcdcef2c4286c4719
1/******************************************************************** 2 * Copyright (c) 1999-2011, International Business Machines 3 * Corporation and others. All Rights Reserved. 4 ******************************************************************** 5 * Date Name Description 6 * 12/14/99 Madhu Creation. 7 * 01/12/2000 Madhu updated for changed API 8 ********************************************************************/ 9 10#include "unicode/utypes.h" 11 12#if !UCONFIG_NO_BREAK_ITERATION 13 14#include "unicode/uchar.h" 15#include "intltest.h" 16#include "unicode/rbbi.h" 17#include "unicode/schriter.h" 18#include "rbbiapts.h" 19#include "rbbidata.h" 20#include "cstring.h" 21#include "ubrkimpl.h" 22#include "unicode/locid.h" 23#include "unicode/ustring.h" 24#include "unicode/utext.h" 25#include "cmemory.h" 26 27/** 28 * API Test the RuleBasedBreakIterator class 29 */ 30 31 32#define TEST_ASSERT_SUCCESS(status) {if (U_FAILURE(status)) {\ 33errln("Failure at file %s, line %d, error = %s", __FILE__, __LINE__, u_errorName(status));}} 34 35#define TEST_ASSERT(expr) {if ((expr) == FALSE) { \ 36 errln("Test Failure at file %s, line %d: \"%s\" is false.\n", __FILE__, __LINE__, #expr);};} 37 38void RBBIAPITest::TestCloneEquals() 39{ 40 41 UErrorCode status=U_ZERO_ERROR; 42 RuleBasedBreakIterator* bi1 = (RuleBasedBreakIterator*)RuleBasedBreakIterator::createCharacterInstance(Locale::getDefault(), status); 43 RuleBasedBreakIterator* biequal = (RuleBasedBreakIterator*)RuleBasedBreakIterator::createCharacterInstance(Locale::getDefault(), status); 44 RuleBasedBreakIterator* bi3 = (RuleBasedBreakIterator*)RuleBasedBreakIterator::createCharacterInstance(Locale::getDefault(), status); 45 RuleBasedBreakIterator* bi2 = (RuleBasedBreakIterator*)RuleBasedBreakIterator::createWordInstance(Locale::getDefault(), status); 46 if(U_FAILURE(status)){ 47 errcheckln(status, "Fail : in construction - %s", u_errorName(status)); 48 return; 49 } 50 51 52 UnicodeString testString="Testing word break iterators's clone() and equals()"; 53 bi1->setText(testString); 54 bi2->setText(testString); 55 biequal->setText(testString); 56 57 bi3->setText("hello"); 58 59 logln((UnicodeString)"Testing equals()"); 60 61 logln((UnicodeString)"Testing == and !="); 62 UBool b = (*bi1 != *biequal); 63 b |= *bi1 == *bi2; 64 b |= *bi1 == *bi3; 65 if (b) { 66 errln((UnicodeString)"ERROR:1 RBBI's == and != operator failed."); 67 } 68 69 if(*bi2 == *biequal || *bi2 == *bi1 || *biequal == *bi3) 70 errln((UnicodeString)"ERROR:2 RBBI's == and != operator failed."); 71 72 73 // Quick test of RulesBasedBreakIterator assignment - 74 // Check that 75 // two different iterators are != 76 // they are == after assignment 77 // source and dest iterator produce the same next() after assignment. 78 // deleting one doesn't disable the other. 79 logln("Testing assignment"); 80 RuleBasedBreakIterator *bix = (RuleBasedBreakIterator *)BreakIterator::createLineInstance(Locale::getDefault(), status); 81 if(U_FAILURE(status)){ 82 errcheckln(status, "Fail : in construction - %s", u_errorName(status)); 83 return; 84 } 85 86 RuleBasedBreakIterator biDefault, biDefault2; 87 if(U_FAILURE(status)){ 88 errln((UnicodeString)"FAIL : in construction of default iterator"); 89 return; 90 } 91 if (biDefault == *bix) { 92 errln((UnicodeString)"ERROR: iterators should not compare =="); 93 return; 94 } 95 if (biDefault != biDefault2) { 96 errln((UnicodeString)"ERROR: iterators should compare =="); 97 return; 98 } 99 100 101 UnicodeString HelloString("Hello Kitty"); 102 bix->setText(HelloString); 103 if (*bix == *bi2) { 104 errln(UnicodeString("ERROR: strings should not be equal before assignment.")); 105 } 106 *bix = *bi2; 107 if (*bix != *bi2) { 108 errln(UnicodeString("ERROR: strings should be equal before assignment.")); 109 } 110 111 int bixnext = bix->next(); 112 int bi2next = bi2->next(); 113 if (! (bixnext == bi2next && bixnext == 7)) { 114 errln(UnicodeString("ERROR: iterators behaved differently after assignment.")); 115 } 116 delete bix; 117 if (bi2->next() != 8) { 118 errln(UnicodeString("ERROR: iterator.next() failed after deleting copy.")); 119 } 120 121 122 123 logln((UnicodeString)"Testing clone()"); 124 RuleBasedBreakIterator* bi1clone=(RuleBasedBreakIterator*)bi1->clone(); 125 RuleBasedBreakIterator* bi2clone=(RuleBasedBreakIterator*)bi2->clone(); 126 127 if(*bi1clone != *bi1 || *bi1clone != *biequal || 128 *bi1clone == *bi3 || *bi1clone == *bi2) 129 errln((UnicodeString)"ERROR:1 RBBI's clone() method failed"); 130 131 if(*bi2clone == *bi1 || *bi2clone == *biequal || 132 *bi2clone == *bi3 || *bi2clone != *bi2) 133 errln((UnicodeString)"ERROR:2 RBBI's clone() method failed"); 134 135 if(bi1->getText() != bi1clone->getText() || 136 bi2clone->getText() != bi2->getText() || 137 *bi2clone == *bi1clone ) 138 errln((UnicodeString)"ERROR: RBBI's clone() method failed"); 139 140 delete bi1clone; 141 delete bi2clone; 142 delete bi1; 143 delete bi3; 144 delete bi2; 145 delete biequal; 146} 147 148void RBBIAPITest::TestBoilerPlate() 149{ 150 UErrorCode status = U_ZERO_ERROR; 151 BreakIterator* a = BreakIterator::createWordInstance(Locale("hi"), status); 152 BreakIterator* b = BreakIterator::createWordInstance(Locale("hi_IN"),status); 153 if (U_FAILURE(status)) { 154 errcheckln(status, "Creation of break iterator failed %s", u_errorName(status)); 155 return; 156 } 157 if(*a!=*b){ 158 errln("Failed: boilerplate method operator!= does not return correct results"); 159 } 160 BreakIterator* c = BreakIterator::createWordInstance(Locale("ja"),status); 161 if(a && c){ 162 if(*c==*a){ 163 errln("Failed: boilerplate method opertator== does not return correct results"); 164 } 165 }else{ 166 errln("creation of break iterator failed"); 167 } 168 delete a; 169 delete b; 170 delete c; 171} 172 173void RBBIAPITest::TestgetRules() 174{ 175 UErrorCode status=U_ZERO_ERROR; 176 177 RuleBasedBreakIterator* bi1=(RuleBasedBreakIterator*)RuleBasedBreakIterator::createCharacterInstance(Locale::getDefault(), status); 178 RuleBasedBreakIterator* bi2=(RuleBasedBreakIterator*)RuleBasedBreakIterator::createWordInstance(Locale::getDefault(), status); 179 if(U_FAILURE(status)){ 180 errcheckln(status, "FAIL: in construction - %s", u_errorName(status)); 181 delete bi1; 182 delete bi2; 183 return; 184 } 185 186 187 188 logln((UnicodeString)"Testing toString()"); 189 190 bi1->setText((UnicodeString)"Hello there"); 191 192 RuleBasedBreakIterator* bi3 =(RuleBasedBreakIterator*)bi1->clone(); 193 194 UnicodeString temp=bi1->getRules(); 195 UnicodeString temp2=bi2->getRules(); 196 UnicodeString temp3=bi3->getRules(); 197 if( temp2.compare(temp3) ==0 || temp.compare(temp2) == 0 || temp.compare(temp3) != 0) 198 errln((UnicodeString)"ERROR: error in getRules() method"); 199 200 delete bi1; 201 delete bi2; 202 delete bi3; 203} 204void RBBIAPITest::TestHashCode() 205{ 206 UErrorCode status=U_ZERO_ERROR; 207 RuleBasedBreakIterator* bi1 = (RuleBasedBreakIterator*)RuleBasedBreakIterator::createCharacterInstance(Locale::getDefault(), status); 208 RuleBasedBreakIterator* bi3 = (RuleBasedBreakIterator*)RuleBasedBreakIterator::createCharacterInstance(Locale::getDefault(), status); 209 RuleBasedBreakIterator* bi2 = (RuleBasedBreakIterator*)RuleBasedBreakIterator::createWordInstance(Locale::getDefault(), status); 210 if(U_FAILURE(status)){ 211 errcheckln(status, "Fail : in construction - %s", u_errorName(status)); 212 delete bi1; 213 delete bi2; 214 delete bi3; 215 return; 216 } 217 218 219 logln((UnicodeString)"Testing hashCode()"); 220 221 bi1->setText((UnicodeString)"Hash code"); 222 bi2->setText((UnicodeString)"Hash code"); 223 bi3->setText((UnicodeString)"Hash code"); 224 225 RuleBasedBreakIterator* bi1clone= (RuleBasedBreakIterator*)bi1->clone(); 226 RuleBasedBreakIterator* bi2clone= (RuleBasedBreakIterator*)bi2->clone(); 227 228 if(bi1->hashCode() != bi1clone->hashCode() || bi1->hashCode() != bi3->hashCode() || 229 bi1clone->hashCode() != bi3->hashCode() || bi2->hashCode() != bi2clone->hashCode()) 230 errln((UnicodeString)"ERROR: identical objects have different hashcodes"); 231 232 if(bi1->hashCode() == bi2->hashCode() || bi2->hashCode() == bi3->hashCode() || 233 bi1clone->hashCode() == bi2clone->hashCode() || bi1clone->hashCode() == bi2->hashCode()) 234 errln((UnicodeString)"ERROR: different objects have same hashcodes"); 235 236 delete bi1clone; 237 delete bi2clone; 238 delete bi1; 239 delete bi2; 240 delete bi3; 241 242} 243void RBBIAPITest::TestGetSetAdoptText() 244{ 245 logln((UnicodeString)"Testing getText setText "); 246 IcuTestErrorCode status(*this, "TestGetSetAdoptText"); 247 UnicodeString str1="first string."; 248 UnicodeString str2="Second string."; 249 LocalPointer<RuleBasedBreakIterator> charIter1((RuleBasedBreakIterator*)RuleBasedBreakIterator::createCharacterInstance(Locale::getDefault(), status)); 250 LocalPointer<RuleBasedBreakIterator> wordIter1((RuleBasedBreakIterator*)RuleBasedBreakIterator::createWordInstance(Locale::getDefault(), status)); 251 if(status.isFailure()){ 252 errcheckln(status, "Fail : in construction - %s", status.errorName()); 253 return; 254 } 255 256 257 CharacterIterator* text1= new StringCharacterIterator(str1); 258 CharacterIterator* text1Clone = text1->clone(); 259 CharacterIterator* text2= new StringCharacterIterator(str2); 260 CharacterIterator* text3= new StringCharacterIterator(str2, 3, 10, 3); // "ond str" 261 262 wordIter1->setText(str1); 263 CharacterIterator *tci = &wordIter1->getText(); 264 UnicodeString tstr; 265 tci->getText(tstr); 266 TEST_ASSERT(tstr == str1); 267 if(wordIter1->current() != 0) 268 errln((UnicodeString)"ERROR:1 setText did not set the iteration position to the beginning of the text, it is" + wordIter1->current() + (UnicodeString)"\n"); 269 270 wordIter1->next(2); 271 272 wordIter1->setText(str2); 273 if(wordIter1->current() != 0) 274 errln((UnicodeString)"ERROR:2 setText did not reset the iteration position to the beginning of the text, it is" + wordIter1->current() + (UnicodeString)"\n"); 275 276 277 charIter1->adoptText(text1Clone); 278 TEST_ASSERT(wordIter1->getText() != charIter1->getText()); 279 tci = &wordIter1->getText(); 280 tci->getText(tstr); 281 TEST_ASSERT(tstr == str2); 282 tci = &charIter1->getText(); 283 tci->getText(tstr); 284 TEST_ASSERT(tstr == str1); 285 286 287 LocalPointer<RuleBasedBreakIterator> rb((RuleBasedBreakIterator*)wordIter1->clone()); 288 rb->adoptText(text1); 289 if(rb->getText() != *text1) 290 errln((UnicodeString)"ERROR:1 error in adoptText "); 291 rb->adoptText(text2); 292 if(rb->getText() != *text2) 293 errln((UnicodeString)"ERROR:2 error in adoptText "); 294 295 // Adopt where iterator range is less than the entire orignal source string. 296 // (With the change of the break engine to working with UText internally, 297 // CharacterIterators starting at positions other than zero are not supported) 298 rb->adoptText(text3); 299 TEST_ASSERT(rb->preceding(2) == 0); 300 TEST_ASSERT(rb->following(11) == BreakIterator::DONE); 301 //if(rb->preceding(2) != 3) { 302 // errln((UnicodeString)"ERROR:3 error in adoptText "); 303 //} 304 //if(rb->following(11) != BreakIterator::DONE) { 305 // errln((UnicodeString)"ERROR:4 error in adoptText "); 306 //} 307 308 // UText API 309 // 310 // Quick test to see if UText is working at all. 311 // 312 const char *s1 = "\x68\x65\x6C\x6C\x6F\x20\x77\x6F\x72\x6C\x64"; /* "hello world" in UTF-8 */ 313 const char *s2 = "\x73\x65\x65\x20\x79\x61"; /* "see ya" in UTF-8 */ 314 // 012345678901 315 316 status.reset(); 317 LocalUTextPointer ut(utext_openUTF8(NULL, s1, -1, status)); 318 wordIter1->setText(ut.getAlias(), status); 319 TEST_ASSERT_SUCCESS(status); 320 321 int32_t pos; 322 pos = wordIter1->first(); 323 TEST_ASSERT(pos==0); 324 pos = wordIter1->next(); 325 TEST_ASSERT(pos==5); 326 pos = wordIter1->next(); 327 TEST_ASSERT(pos==6); 328 pos = wordIter1->next(); 329 TEST_ASSERT(pos==11); 330 pos = wordIter1->next(); 331 TEST_ASSERT(pos==UBRK_DONE); 332 333 status.reset(); 334 LocalUTextPointer ut2(utext_openUTF8(NULL, s2, -1, status)); 335 TEST_ASSERT_SUCCESS(status); 336 wordIter1->setText(ut2.getAlias(), status); 337 TEST_ASSERT_SUCCESS(status); 338 339 pos = wordIter1->first(); 340 TEST_ASSERT(pos==0); 341 pos = wordIter1->next(); 342 TEST_ASSERT(pos==3); 343 pos = wordIter1->next(); 344 TEST_ASSERT(pos==4); 345 346 pos = wordIter1->last(); 347 TEST_ASSERT(pos==6); 348 pos = wordIter1->previous(); 349 TEST_ASSERT(pos==4); 350 pos = wordIter1->previous(); 351 TEST_ASSERT(pos==3); 352 pos = wordIter1->previous(); 353 TEST_ASSERT(pos==0); 354 pos = wordIter1->previous(); 355 TEST_ASSERT(pos==UBRK_DONE); 356 357 status.reset(); 358 UnicodeString sEmpty; 359 LocalUTextPointer gut2(utext_openUnicodeString(NULL, &sEmpty, status)); 360 wordIter1->getUText(gut2.getAlias(), status); 361 TEST_ASSERT_SUCCESS(status); 362 status.reset(); 363} 364 365 366void RBBIAPITest::TestIteration() 367{ 368 // This test just verifies that the API is present. 369 // Testing for correct operation of the break rules happens elsewhere. 370 371 UErrorCode status=U_ZERO_ERROR; 372 RuleBasedBreakIterator* bi = (RuleBasedBreakIterator*)RuleBasedBreakIterator::createCharacterInstance(Locale::getDefault(), status); 373 if (U_FAILURE(status) || bi == NULL) { 374 errcheckln(status, "Failure creating character break iterator. Status = %s", u_errorName(status)); 375 } 376 delete bi; 377 378 status=U_ZERO_ERROR; 379 bi = (RuleBasedBreakIterator*)RuleBasedBreakIterator::createWordInstance(Locale::getDefault(), status); 380 if (U_FAILURE(status) || bi == NULL) { 381 errcheckln(status, "Failure creating Word break iterator. Status = %s", u_errorName(status)); 382 } 383 delete bi; 384 385 status=U_ZERO_ERROR; 386 bi = (RuleBasedBreakIterator*)RuleBasedBreakIterator::createLineInstance(Locale::getDefault(), status); 387 if (U_FAILURE(status) || bi == NULL) { 388 errcheckln(status, "Failure creating Line break iterator. Status = %s", u_errorName(status)); 389 } 390 delete bi; 391 392 status=U_ZERO_ERROR; 393 bi = (RuleBasedBreakIterator*)RuleBasedBreakIterator::createSentenceInstance(Locale::getDefault(), status); 394 if (U_FAILURE(status) || bi == NULL) { 395 errcheckln(status, "Failure creating Sentence break iterator. Status = %s", u_errorName(status)); 396 } 397 delete bi; 398 399 status=U_ZERO_ERROR; 400 bi = (RuleBasedBreakIterator*)RuleBasedBreakIterator::createTitleInstance(Locale::getDefault(), status); 401 if (U_FAILURE(status) || bi == NULL) { 402 errcheckln(status, "Failure creating Title break iterator. Status = %s", u_errorName(status)); 403 } 404 delete bi; 405 406 status=U_ZERO_ERROR; 407 bi = (RuleBasedBreakIterator*)RuleBasedBreakIterator::createCharacterInstance(Locale::getDefault(), status); 408 if (U_FAILURE(status) || bi == NULL) { 409 errcheckln(status, "Failure creating character break iterator. Status = %s", u_errorName(status)); 410 return; // Skip the rest of these tests. 411 } 412 413 414 UnicodeString testString="0123456789"; 415 bi->setText(testString); 416 417 int32_t i; 418 i = bi->first(); 419 if (i != 0) { 420 errln("Incorrect value from bi->first(). Expected 0, got %d.", i); 421 } 422 423 i = bi->last(); 424 if (i != 10) { 425 errln("Incorrect value from bi->last(). Expected 10, got %d", i); 426 } 427 428 // 429 // Previous 430 // 431 bi->last(); 432 i = bi->previous(); 433 if (i != 9) { 434 errln("Incorrect value from bi->last() at line %d. Expected 9, got %d", __LINE__, i); 435 } 436 437 438 bi->first(); 439 i = bi->previous(); 440 if (i != BreakIterator::DONE) { 441 errln("Incorrect value from bi->previous() at line %d. Expected DONE, got %d", __LINE__, i); 442 } 443 444 // 445 // next() 446 // 447 bi->first(); 448 i = bi->next(); 449 if (i != 1) { 450 errln("Incorrect value from bi->next() at line %d. Expected 1, got %d", __LINE__, i); 451 } 452 453 bi->last(); 454 i = bi->next(); 455 if (i != BreakIterator::DONE) { 456 errln("Incorrect value from bi->next() at line %d. Expected DONE, got %d", __LINE__, i); 457 } 458 459 460 // 461 // current() 462 // 463 bi->first(); 464 i = bi->current(); 465 if (i != 0) { 466 errln("Incorrect value from bi->previous() at line %d. Expected 0, got %d", __LINE__, i); 467 } 468 469 bi->next(); 470 i = bi->current(); 471 if (i != 1) { 472 errln("Incorrect value from bi->previous() at line %d. Expected 1, got %d", __LINE__, i); 473 } 474 475 bi->last(); 476 bi->next(); 477 i = bi->current(); 478 if (i != 10) { 479 errln("Incorrect value from bi->previous() at line %d. Expected 10, got %d", __LINE__, i); 480 } 481 482 bi->first(); 483 bi->previous(); 484 i = bi->current(); 485 if (i != 0) { 486 errln("Incorrect value from bi->previous() at line %d. Expected 0, got %d", __LINE__, i); 487 } 488 489 490 // 491 // Following() 492 // 493 i = bi->following(4); 494 if (i != 5) { 495 errln("Incorrect value from bi->following() at line %d. Expected 5, got %d", __LINE__, i); 496 } 497 498 i = bi->following(9); 499 if (i != 10) { 500 errln("Incorrect value from bi->following() at line %d. Expected 10, got %d", __LINE__, i); 501 } 502 503 i = bi->following(10); 504 if (i != BreakIterator::DONE) { 505 errln("Incorrect value from bi->following() at line %d. Expected DONE, got %d", __LINE__, i); 506 } 507 508 509 // 510 // Preceding 511 // 512 i = bi->preceding(4); 513 if (i != 3) { 514 errln("Incorrect value from bi->preceding() at line %d. Expected 3, got %d", __LINE__, i); 515 } 516 517 i = bi->preceding(10); 518 if (i != 9) { 519 errln("Incorrect value from bi->preceding() at line %d. Expected 9, got %d", __LINE__, i); 520 } 521 522 i = bi->preceding(1); 523 if (i != 0) { 524 errln("Incorrect value from bi->preceding() at line %d. Expected 0, got %d", __LINE__, i); 525 } 526 527 i = bi->preceding(0); 528 if (i != BreakIterator::DONE) { 529 errln("Incorrect value from bi->preceding() at line %d. Expected DONE, got %d", __LINE__, i); 530 } 531 532 533 // 534 // isBoundary() 535 // 536 bi->first(); 537 if (bi->isBoundary(3) != TRUE) { 538 errln("Incorrect value from bi->isBoudary() at line %d. Expected TRUE, got FALSE", __LINE__, i); 539 } 540 i = bi->current(); 541 if (i != 3) { 542 errln("Incorrect value from bi->current() at line %d. Expected 3, got %d", __LINE__, i); 543 } 544 545 546 if (bi->isBoundary(11) != FALSE) { 547 errln("Incorrect value from bi->isBoudary() at line %d. Expected FALSE, got TRUE", __LINE__, i); 548 } 549 i = bi->current(); 550 if (i != 10) { 551 errln("Incorrect value from bi->current() at line %d. Expected 10, got %d", __LINE__, i); 552 } 553 554 // 555 // next(n) 556 // 557 bi->first(); 558 i = bi->next(4); 559 if (i != 4) { 560 errln("Incorrect value from bi->next() at line %d. Expected 4, got %d", __LINE__, i); 561 } 562 563 i = bi->next(6); 564 if (i != 10) { 565 errln("Incorrect value from bi->next() at line %d. Expected 10, got %d", __LINE__, i); 566 } 567 568 bi->first(); 569 i = bi->next(11); 570 if (i != BreakIterator::DONE) { 571 errln("Incorrect value from bi->next() at line %d. Expected BreakIterator::DONE, got %d", __LINE__, i); 572 } 573 574 delete bi; 575 576} 577 578 579 580 581 582 583void RBBIAPITest::TestBuilder() { 584 UnicodeString rulesString1 = "$Letters = [:L:];\n" 585 "$Numbers = [:N:];\n" 586 "$Letters+;\n" 587 "$Numbers+;\n" 588 "[^$Letters $Numbers];\n" 589 "!.*;\n"; 590 UnicodeString testString1 = "abc123..abc"; 591 // 01234567890 592 int32_t bounds1[] = {0, 3, 6, 7, 8, 11}; 593 UErrorCode status=U_ZERO_ERROR; 594 UParseError parseError; 595 596 RuleBasedBreakIterator *bi = new RuleBasedBreakIterator(rulesString1, parseError, status); 597 if(U_FAILURE(status)) { 598 dataerrln("Fail : in construction - %s", u_errorName(status)); 599 } else { 600 bi->setText(testString1); 601 doBoundaryTest(*bi, testString1, bounds1); 602 } 603 delete bi; 604} 605 606 607// 608// TestQuoteGrouping 609// Single quotes within rules imply a grouping, so that a modifier 610// following the quoted text (* or +) applies to all of the quoted chars. 611// 612void RBBIAPITest::TestQuoteGrouping() { 613 UnicodeString rulesString1 = "#Here comes the rule...\n" 614 "'$@!'*;\n" // (\$\@\!)* 615 ".;\n"; 616 617 UnicodeString testString1 = "$@!$@!X$@!!X"; 618 // 0123456789012 619 int32_t bounds1[] = {0, 6, 7, 10, 11, 12}; 620 UErrorCode status=U_ZERO_ERROR; 621 UParseError parseError; 622 623 RuleBasedBreakIterator *bi = new RuleBasedBreakIterator(rulesString1, parseError, status); 624 if(U_FAILURE(status)) { 625 dataerrln("Fail : in construction - %s", u_errorName(status)); 626 } else { 627 bi->setText(testString1); 628 doBoundaryTest(*bi, testString1, bounds1); 629 } 630 delete bi; 631} 632 633// 634// TestRuleStatus 635// Test word break rule status constants. 636// 637void RBBIAPITest::TestRuleStatus() { 638 UChar str[30]; 639 u_unescape("plain word 123.45 \\u9160\\u9161 \\u30a1\\u30a2 \\u3041\\u3094", 640 // 012345678901234567 8 9 0 1 2 3 4 5 6 641 // Ideographic Katakana Hiragana 642 str, 30); 643 UnicodeString testString1(str); 644 int32_t bounds1[] = {0, 5, 6, 10, 11, 17, 18, 19, 20, 21, 23, 24, 25, 26}; 645 int32_t tag_lo[] = {UBRK_WORD_NONE, UBRK_WORD_LETTER, UBRK_WORD_NONE, UBRK_WORD_LETTER, 646 UBRK_WORD_NONE, UBRK_WORD_NUMBER, UBRK_WORD_NONE, 647 UBRK_WORD_IDEO, UBRK_WORD_IDEO, UBRK_WORD_NONE, 648 UBRK_WORD_KANA, UBRK_WORD_NONE, UBRK_WORD_KANA, UBRK_WORD_KANA}; 649 650 int32_t tag_hi[] = {UBRK_WORD_NONE_LIMIT, UBRK_WORD_LETTER_LIMIT, UBRK_WORD_NONE_LIMIT, UBRK_WORD_LETTER_LIMIT, 651 UBRK_WORD_NONE_LIMIT, UBRK_WORD_NUMBER_LIMIT, UBRK_WORD_NONE_LIMIT, 652 UBRK_WORD_IDEO_LIMIT, UBRK_WORD_IDEO_LIMIT, UBRK_WORD_NONE_LIMIT, 653 UBRK_WORD_KANA_LIMIT, UBRK_WORD_NONE_LIMIT, UBRK_WORD_KANA_LIMIT, UBRK_WORD_KANA_LIMIT}; 654 655 UErrorCode status=U_ZERO_ERROR; 656 657 RuleBasedBreakIterator *bi = (RuleBasedBreakIterator *)BreakIterator::createWordInstance(Locale::getEnglish(), status); 658 if(U_FAILURE(status)) { 659 errcheckln(status, "Fail : in construction - %s", u_errorName(status)); 660 } else { 661 bi->setText(testString1); 662 // First test that the breaks are in the right spots. 663 doBoundaryTest(*bi, testString1, bounds1); 664 665 // Then go back and check tag values 666 int32_t i = 0; 667 int32_t pos, tag; 668 for (pos = bi->first(); pos != BreakIterator::DONE; pos = bi->next(), i++) { 669 if (pos != bounds1[i]) { 670 errln("FAIL: unexpected word break at postion %d", pos); 671 break; 672 } 673 tag = bi->getRuleStatus(); 674 if (tag < tag_lo[i] || tag >= tag_hi[i]) { 675 errln("FAIL: incorrect tag value %d at position %d", tag, pos); 676 break; 677 } 678 679 // Check that we get the same tag values from getRuleStatusVec() 680 int32_t vec[10]; 681 int t = bi->getRuleStatusVec(vec, 10, status); 682 TEST_ASSERT_SUCCESS(status); 683 TEST_ASSERT(t==1); 684 TEST_ASSERT(vec[0] == tag); 685 } 686 } 687 delete bi; 688 689 // Now test line break status. This test mostly is to confirm that the status constants 690 // are correctly declared in the header. 691 testString1 = "test line. \n"; 692 // break type s s h 693 694 bi = (RuleBasedBreakIterator *) 695 BreakIterator::createLineInstance(Locale::getEnglish(), status); 696 if(U_FAILURE(status)) { 697 errcheckln(status, "failed to create word break iterator. - %s", u_errorName(status)); 698 } else { 699 int32_t i = 0; 700 int32_t pos, tag; 701 UBool success; 702 703 bi->setText(testString1); 704 pos = bi->current(); 705 tag = bi->getRuleStatus(); 706 for (i=0; i<3; i++) { 707 switch (i) { 708 case 0: 709 success = pos==0 && tag==UBRK_LINE_SOFT; break; 710 case 1: 711 success = pos==5 && tag==UBRK_LINE_SOFT; break; 712 case 2: 713 success = pos==12 && tag==UBRK_LINE_HARD; break; 714 default: 715 success = FALSE; break; 716 } 717 if (success == FALSE) { 718 errln("Fail: incorrect word break status or position. i=%d, pos=%d, tag=%d", 719 i, pos, tag); 720 break; 721 } 722 pos = bi->next(); 723 tag = bi->getRuleStatus(); 724 } 725 if (UBRK_LINE_SOFT >= UBRK_LINE_SOFT_LIMIT || 726 UBRK_LINE_HARD >= UBRK_LINE_HARD_LIMIT || 727 (UBRK_LINE_HARD > UBRK_LINE_SOFT && UBRK_LINE_HARD < UBRK_LINE_SOFT_LIMIT)) { 728 errln("UBRK_LINE_* constants from header are inconsistent."); 729 } 730 } 731 delete bi; 732 733} 734 735 736// 737// TestRuleStatusVec 738// Test the vector form of break rule status. 739// 740void RBBIAPITest::TestRuleStatusVec() { 741 UnicodeString rulesString( "[A-N]{100}; \n" 742 "[a-w]{200}; \n" 743 "[\\p{L}]{300}; \n" 744 "[\\p{N}]{400}; \n" 745 "[0-5]{500}; \n" 746 "!.*;\n", -1, US_INV); 747 UnicodeString testString1 = "Aapz5?"; 748 int32_t statusVals[10]; 749 int32_t numStatuses; 750 int32_t pos; 751 752 UErrorCode status=U_ZERO_ERROR; 753 UParseError parseError; 754 755 RuleBasedBreakIterator *bi = new RuleBasedBreakIterator(rulesString, parseError, status); 756 if (U_FAILURE(status)) { 757 dataerrln("Failure at file %s, line %d, error = %s", __FILE__, __LINE__, u_errorName(status)); 758 } else { 759 bi->setText(testString1); 760 761 // A 762 pos = bi->next(); 763 TEST_ASSERT(pos==1); 764 numStatuses = bi->getRuleStatusVec(statusVals, 10, status); 765 TEST_ASSERT_SUCCESS(status); 766 TEST_ASSERT(numStatuses == 2); 767 TEST_ASSERT(statusVals[0] == 100); 768 TEST_ASSERT(statusVals[1] == 300); 769 770 // a 771 pos = bi->next(); 772 TEST_ASSERT(pos==2); 773 numStatuses = bi->getRuleStatusVec(statusVals, 10, status); 774 TEST_ASSERT_SUCCESS(status); 775 TEST_ASSERT(numStatuses == 2); 776 TEST_ASSERT(statusVals[0] == 200); 777 TEST_ASSERT(statusVals[1] == 300); 778 779 // p 780 pos = bi->next(); 781 TEST_ASSERT(pos==3); 782 numStatuses = bi->getRuleStatusVec(statusVals, 10, status); 783 TEST_ASSERT_SUCCESS(status); 784 TEST_ASSERT(numStatuses == 2); 785 TEST_ASSERT(statusVals[0] == 200); 786 TEST_ASSERT(statusVals[1] == 300); 787 788 // z 789 pos = bi->next(); 790 TEST_ASSERT(pos==4); 791 numStatuses = bi->getRuleStatusVec(statusVals, 10, status); 792 TEST_ASSERT_SUCCESS(status); 793 TEST_ASSERT(numStatuses == 1); 794 TEST_ASSERT(statusVals[0] == 300); 795 796 // 5 797 pos = bi->next(); 798 TEST_ASSERT(pos==5); 799 numStatuses = bi->getRuleStatusVec(statusVals, 10, status); 800 TEST_ASSERT_SUCCESS(status); 801 TEST_ASSERT(numStatuses == 2); 802 TEST_ASSERT(statusVals[0] == 400); 803 TEST_ASSERT(statusVals[1] == 500); 804 805 // ? 806 pos = bi->next(); 807 TEST_ASSERT(pos==6); 808 numStatuses = bi->getRuleStatusVec(statusVals, 10, status); 809 TEST_ASSERT_SUCCESS(status); 810 TEST_ASSERT(numStatuses == 1); 811 TEST_ASSERT(statusVals[0] == 0); 812 813 // 814 // Check buffer overflow error handling. Char == A 815 // 816 bi->first(); 817 pos = bi->next(); 818 TEST_ASSERT(pos==1); 819 memset(statusVals, -1, sizeof(statusVals)); 820 numStatuses = bi->getRuleStatusVec(statusVals, 0, status); 821 TEST_ASSERT(status == U_BUFFER_OVERFLOW_ERROR); 822 TEST_ASSERT(numStatuses == 2); 823 TEST_ASSERT(statusVals[0] == -1); 824 825 status = U_ZERO_ERROR; 826 memset(statusVals, -1, sizeof(statusVals)); 827 numStatuses = bi->getRuleStatusVec(statusVals, 1, status); 828 TEST_ASSERT(status == U_BUFFER_OVERFLOW_ERROR); 829 TEST_ASSERT(numStatuses == 2); 830 TEST_ASSERT(statusVals[0] == 100); 831 TEST_ASSERT(statusVals[1] == -1); 832 833 status = U_ZERO_ERROR; 834 memset(statusVals, -1, sizeof(statusVals)); 835 numStatuses = bi->getRuleStatusVec(statusVals, 2, status); 836 TEST_ASSERT_SUCCESS(status); 837 TEST_ASSERT(numStatuses == 2); 838 TEST_ASSERT(statusVals[0] == 100); 839 TEST_ASSERT(statusVals[1] == 300); 840 TEST_ASSERT(statusVals[2] == -1); 841 } 842 delete bi; 843 844} 845 846// 847// Bug 2190 Regression test. Builder crash on rule consisting of only a 848// $variable reference 849void RBBIAPITest::TestBug2190() { 850 UnicodeString rulesString1 = "$aaa = abcd;\n" 851 "$bbb = $aaa;\n" 852 "$bbb;\n"; 853 UnicodeString testString1 = "abcdabcd"; 854 // 01234567890 855 int32_t bounds1[] = {0, 4, 8}; 856 UErrorCode status=U_ZERO_ERROR; 857 UParseError parseError; 858 859 RuleBasedBreakIterator *bi = new RuleBasedBreakIterator(rulesString1, parseError, status); 860 if(U_FAILURE(status)) { 861 dataerrln("Fail : in construction - %s", u_errorName(status)); 862 } else { 863 bi->setText(testString1); 864 doBoundaryTest(*bi, testString1, bounds1); 865 } 866 delete bi; 867} 868 869 870void RBBIAPITest::TestRegistration() { 871#if !UCONFIG_NO_SERVICE 872 UErrorCode status = U_ZERO_ERROR; 873 BreakIterator* ja_word = BreakIterator::createWordInstance("ja_JP", status); 874 875 // ok to not delete these if we exit because of error? 876 BreakIterator* ja_char = BreakIterator::createCharacterInstance("ja_JP", status); 877 BreakIterator* root_word = BreakIterator::createWordInstance("", status); 878 BreakIterator* root_char = BreakIterator::createCharacterInstance("", status); 879 880 if (status == U_MISSING_RESOURCE_ERROR || status == U_FILE_ACCESS_ERROR) { 881 dataerrln("Error creating instances of break interactors - %s", u_errorName(status)); 882 delete ja_word; 883 delete ja_char; 884 delete root_word; 885 delete root_char; 886 887 return; 888 } 889 890 URegistryKey key = BreakIterator::registerInstance(ja_word, "xx", UBRK_WORD, status); 891 { 892 if (ja_word && *ja_word == *root_word) { 893 errln("japan not different from root"); 894 } 895 } 896 897 { 898 BreakIterator* result = BreakIterator::createWordInstance("xx_XX", status); 899 UBool fail = TRUE; 900 if(result){ 901 fail = *result != *ja_word; 902 } 903 delete result; 904 if (fail) { 905 errln("bad result for xx_XX/word"); 906 } 907 } 908 909 { 910 BreakIterator* result = BreakIterator::createCharacterInstance("ja_JP", status); 911 UBool fail = TRUE; 912 if(result){ 913 fail = *result != *ja_char; 914 } 915 delete result; 916 if (fail) { 917 errln("bad result for ja_JP/char"); 918 } 919 } 920 921 { 922 BreakIterator* result = BreakIterator::createCharacterInstance("xx_XX", status); 923 UBool fail = TRUE; 924 if(result){ 925 fail = *result != *root_char; 926 } 927 delete result; 928 if (fail) { 929 errln("bad result for xx_XX/char"); 930 } 931 } 932 933 { 934 StringEnumeration* avail = BreakIterator::getAvailableLocales(); 935 UBool found = FALSE; 936 const UnicodeString* p; 937 while ((p = avail->snext(status))) { 938 if (p->compare("xx") == 0) { 939 found = TRUE; 940 break; 941 } 942 } 943 delete avail; 944 if (!found) { 945 errln("did not find test locale"); 946 } 947 } 948 949 { 950 UBool unreg = BreakIterator::unregister(key, status); 951 if (!unreg) { 952 errln("unable to unregister"); 953 } 954 } 955 956 { 957 BreakIterator* result = BreakIterator::createWordInstance("en_US", status); 958 BreakIterator* root = BreakIterator::createWordInstance("", status); 959 UBool fail = TRUE; 960 if(root){ 961 fail = *root != *result; 962 } 963 delete root; 964 delete result; 965 if (fail) { 966 errln("did not get root break"); 967 } 968 } 969 970 { 971 StringEnumeration* avail = BreakIterator::getAvailableLocales(); 972 UBool found = FALSE; 973 const UnicodeString* p; 974 while ((p = avail->snext(status))) { 975 if (p->compare("xx") == 0) { 976 found = TRUE; 977 break; 978 } 979 } 980 delete avail; 981 if (found) { 982 errln("found test locale"); 983 } 984 } 985 986 { 987 int32_t count; 988 UBool foundLocale = FALSE; 989 const Locale *avail = BreakIterator::getAvailableLocales(count); 990 for (int i=0; i<count; i++) { 991 if (avail[i] == Locale::getEnglish()) { 992 foundLocale = TRUE; 993 break; 994 } 995 } 996 if (foundLocale == FALSE) { 997 errln("BreakIterator::getAvailableLocales(&count), failed to find EN."); 998 } 999 } 1000 1001 1002 // ja_word was adopted by factory 1003 delete ja_char; 1004 delete root_word; 1005 delete root_char; 1006#endif 1007} 1008 1009void RBBIAPITest::RoundtripRule(const char *dataFile) { 1010 UErrorCode status = U_ZERO_ERROR; 1011 UParseError parseError; 1012 parseError.line = 0; 1013 parseError.offset = 0; 1014 LocalUDataMemoryPointer data(udata_open(U_ICUDATA_BRKITR, "brk", dataFile, &status)); 1015 uint32_t length; 1016 const UChar *builtSource; 1017 const uint8_t *rbbiRules; 1018 const uint8_t *builtRules; 1019 1020 if (U_FAILURE(status)) { 1021 errcheckln(status, "Can't open \"%s\" - %s", dataFile, u_errorName(status)); 1022 return; 1023 } 1024 1025 builtRules = (const uint8_t *)udata_getMemory(data.getAlias()); 1026 builtSource = (const UChar *)(builtRules + ((RBBIDataHeader*)builtRules)->fRuleSource); 1027 RuleBasedBreakIterator *brkItr = new RuleBasedBreakIterator(builtSource, parseError, status); 1028 if (U_FAILURE(status)) { 1029 errln("createRuleBasedBreakIterator: ICU Error \"%s\" at line %d, column %d\n", 1030 u_errorName(status), parseError.line, parseError.offset); 1031 return; 1032 }; 1033 rbbiRules = brkItr->getBinaryRules(length); 1034 logln("Comparing \"%s\" len=%d", dataFile, length); 1035 if (memcmp(builtRules, rbbiRules, (int32_t)length) != 0) { 1036 errln("Built rules and rebuilt rules are different %s", dataFile); 1037 return; 1038 } 1039 delete brkItr; 1040} 1041 1042void RBBIAPITest::TestRoundtripRules() { 1043 RoundtripRule("word"); 1044 RoundtripRule("title"); 1045 RoundtripRule("sent"); 1046 RoundtripRule("line"); 1047 RoundtripRule("char"); 1048 if (!quick) { 1049 RoundtripRule("word_ja"); 1050 RoundtripRule("word_POSIX"); 1051 } 1052} 1053 1054// Try out the RuleBasedBreakIterator constructors that take RBBIDataHeader* 1055// (these are protected so we access them via a local class RBBIWithProtectedFunctions). 1056// This is just a sanity check, not a thorough test (e.g. we don't check that the 1057// first delete actually frees rulesCopy). 1058void RBBIAPITest::TestCreateFromRBBIData() { 1059 // Get some handy RBBIData 1060 const char *brkName = "word"; // or "sent", "line", "char", etc. 1061 UErrorCode status = U_ZERO_ERROR; 1062 LocalUDataMemoryPointer data(udata_open(U_ICUDATA_BRKITR, "brk", brkName, &status)); 1063 if ( U_SUCCESS(status) ) { 1064 const RBBIDataHeader * builtRules = (const RBBIDataHeader *)udata_getMemory(data.getAlias()); 1065 uint32_t length = builtRules->fLength; 1066 RBBIWithProtectedFunctions * brkItr; 1067 1068 // Try the memory-adopting constructor, need to copy the data first 1069 RBBIDataHeader * rulesCopy = (RBBIDataHeader *) uprv_malloc(length); 1070 if ( rulesCopy ) { 1071 uprv_memcpy( rulesCopy, builtRules, length ); 1072 1073 brkItr = new RBBIWithProtectedFunctions(rulesCopy, status); 1074 if ( U_SUCCESS(status) ) { 1075 delete brkItr; // this should free rulesCopy 1076 } else { 1077 errln("create RuleBasedBreakIterator from RBBIData (adopted): ICU Error \"%s\"\n", u_errorName(status) ); 1078 status = U_ZERO_ERROR;// reset for the next test 1079 uprv_free( rulesCopy ); 1080 } 1081 } 1082 1083 // Now try the non-adopting constructor 1084 brkItr = new RBBIWithProtectedFunctions(builtRules, RBBIWithProtectedFunctions::kDontAdopt, status); 1085 if ( U_SUCCESS(status) ) { 1086 delete brkItr; // this should NOT attempt to free builtRules 1087 if (builtRules->fLength != length) { // sanity check 1088 errln("create RuleBasedBreakIterator from RBBIData (non-adopted): delete affects data\n" ); 1089 } 1090 } else { 1091 errln("create RuleBasedBreakIterator from RBBIData (non-adopted): ICU Error \"%s\"\n", u_errorName(status) ); 1092 } 1093 } 1094 1095 // getBinaryRules() and RuleBasedBreakIterator(uint8_t binaryRules, ...) 1096 // 1097 status = U_ZERO_ERROR; 1098 RuleBasedBreakIterator *rb = (RuleBasedBreakIterator *)BreakIterator::createWordInstance(Locale::getEnglish(), status); 1099 if (rb == NULL || U_FAILURE(status)) { 1100 dataerrln("Unable to create BreakIterator::createWordInstance (Locale::getEnglish) - %s", u_errorName(status)); 1101 } else { 1102 uint32_t length; 1103 const uint8_t *rules = rb->getBinaryRules(length); 1104 RuleBasedBreakIterator *rb2 = new RuleBasedBreakIterator(rules, length, status); 1105 TEST_ASSERT_SUCCESS(status); 1106 TEST_ASSERT(*rb == *rb2); 1107 UnicodeString words = "one two three "; 1108 rb2->setText(words); 1109 int wordCounter = 0; 1110 while (rb2->next() != UBRK_DONE) { 1111 wordCounter++; 1112 } 1113 TEST_ASSERT(wordCounter == 6); 1114 1115 status = U_ZERO_ERROR; 1116 RuleBasedBreakIterator *rb3 = new RuleBasedBreakIterator(rules, length-1, status); 1117 TEST_ASSERT(status == U_ILLEGAL_ARGUMENT_ERROR); 1118 1119 delete rb; 1120 delete rb2; 1121 delete rb3; 1122 } 1123} 1124 1125 1126void RBBIAPITest::TestRefreshInputText() { 1127 /* 1128 * RefreshInput changes out the input of a Break Iterator without 1129 * changing anything else in the iterator's state. Used with Java JNI, 1130 * when Java moves the underlying string storage. This test 1131 * runs BreakIterator::next() repeatedly, moving the text in the middle of the sequence. 1132 * The right set of boundaries should still be found. 1133 */ 1134 UChar testStr[] = {0x20, 0x41, 0x20, 0x42, 0x20, 0x43, 0x20, 0x44, 0x0}; /* = " A B C D" */ 1135 UChar movedStr[] = {0x20, 0x20, 0x20, 0x20, 0x20, 0x20, 0x20, 0x20, 0}; 1136 UErrorCode status = U_ZERO_ERROR; 1137 UText ut1 = UTEXT_INITIALIZER; 1138 UText ut2 = UTEXT_INITIALIZER; 1139 RuleBasedBreakIterator *bi = (RuleBasedBreakIterator *)BreakIterator::createLineInstance(Locale::getEnglish(), status); 1140 TEST_ASSERT_SUCCESS(status); 1141 1142 utext_openUChars(&ut1, testStr, -1, &status); 1143 TEST_ASSERT_SUCCESS(status); 1144 bi->setText(&ut1, status); 1145 TEST_ASSERT_SUCCESS(status); 1146 1147 /* Line boundaries will occur before each letter in the original string */ 1148 TEST_ASSERT(1 == bi->next()); 1149 TEST_ASSERT(3 == bi->next()); 1150 1151 /* Move the string, kill the original string. */ 1152 u_strcpy(movedStr, testStr); 1153 u_memset(testStr, 0x20, u_strlen(testStr)); 1154 utext_openUChars(&ut2, movedStr, -1, &status); 1155 TEST_ASSERT_SUCCESS(status); 1156 RuleBasedBreakIterator *returnedBI = &bi->refreshInputText(&ut2, status); 1157 TEST_ASSERT_SUCCESS(status); 1158 TEST_ASSERT(bi == returnedBI); 1159 1160 /* Find the following matches, now working in the moved string. */ 1161 TEST_ASSERT(5 == bi->next()); 1162 TEST_ASSERT(7 == bi->next()); 1163 TEST_ASSERT(8 == bi->next()); 1164 TEST_ASSERT(UBRK_DONE == bi->next()); 1165 1166 delete bi; 1167 utext_close(&ut1); 1168 utext_close(&ut2); 1169 1170} 1171 1172 1173//--------------------------------------------- 1174// runIndexedTest 1175//--------------------------------------------- 1176 1177void RBBIAPITest::runIndexedTest( int32_t index, UBool exec, const char* &name, char* /*par*/ ) 1178{ 1179 if (exec) logln((UnicodeString)"TestSuite RuleBasedBreakIterator API "); 1180 switch (index) { 1181 // case 0: name = "TestConstruction"; if (exec) TestConstruction(); break; 1182#if !UCONFIG_NO_FILE_IO 1183 case 0: name = "TestCloneEquals"; if (exec) TestCloneEquals(); break; 1184 case 1: name = "TestgetRules"; if (exec) TestgetRules(); break; 1185 case 2: name = "TestHashCode"; if (exec) TestHashCode(); break; 1186 case 3: name = "TestGetSetAdoptText"; if (exec) TestGetSetAdoptText(); break; 1187 case 4: name = "TestIteration"; if (exec) TestIteration(); break; 1188#else 1189 case 0: case 1: case 2: case 3: case 4: name = "skip"; break; 1190#endif 1191 case 5: name = "TestBuilder"; if (exec) TestBuilder(); break; 1192 case 6: name = "TestQuoteGrouping"; if (exec) TestQuoteGrouping(); break; 1193 case 7: name = "TestRuleStatusVec"; if (exec) TestRuleStatusVec(); break; 1194 case 8: name = "TestBug2190"; if (exec) TestBug2190(); break; 1195#if !UCONFIG_NO_FILE_IO 1196 case 9: name = "TestRegistration"; if (exec) TestRegistration(); break; 1197 case 10: name = "TestBoilerPlate"; if (exec) TestBoilerPlate(); break; 1198 case 11: name = "TestRuleStatus"; if (exec) TestRuleStatus(); break; 1199 case 12: name = "TestRoundtripRules"; if (exec) TestRoundtripRules(); break; 1200 case 13: name = "TestCreateFromRBBIData"; if (exec) TestCreateFromRBBIData(); break; 1201#else 1202 case 9: case 10: case 11: case 12: case 13: name = "skip"; break; 1203#endif 1204 case 14: name = "TestRefreshInputText"; if (exec) TestRefreshInputText(); break; 1205 1206 default: name = ""; break; // needed to end loop 1207 } 1208} 1209 1210//--------------------------------------------- 1211//Internal subroutines 1212//--------------------------------------------- 1213 1214void RBBIAPITest::doBoundaryTest(RuleBasedBreakIterator& bi, UnicodeString& text, int32_t *boundaries){ 1215 logln((UnicodeString)"testIsBoundary():"); 1216 int32_t p = 0; 1217 UBool isB; 1218 for (int32_t i = 0; i < text.length(); i++) { 1219 isB = bi.isBoundary(i); 1220 logln((UnicodeString)"bi.isBoundary(" + i + ") -> " + isB); 1221 1222 if (i == boundaries[p]) { 1223 if (!isB) 1224 errln((UnicodeString)"Wrong result from isBoundary() for " + i + (UnicodeString)": expected true, got false"); 1225 p++; 1226 } 1227 else { 1228 if (isB) 1229 errln((UnicodeString)"Wrong result from isBoundary() for " + i + (UnicodeString)": expected false, got true"); 1230 } 1231 } 1232} 1233void RBBIAPITest::doTest(UnicodeString& testString, int32_t start, int32_t gotoffset, int32_t expectedOffset, const char* expectedString){ 1234 UnicodeString selected; 1235 UnicodeString expected=CharsToUnicodeString(expectedString); 1236 1237 if(gotoffset != expectedOffset) 1238 errln((UnicodeString)"ERROR:****returned #" + gotoffset + (UnicodeString)" instead of #" + expectedOffset); 1239 if(start <= gotoffset){ 1240 testString.extractBetween(start, gotoffset, selected); 1241 } 1242 else{ 1243 testString.extractBetween(gotoffset, start, selected); 1244 } 1245 if(selected.compare(expected) != 0) 1246 errln(prettify((UnicodeString)"ERROR:****selected \"" + selected + "\" instead of \"" + expected + "\"")); 1247 else 1248 logln(prettify("****selected \"" + selected + "\"")); 1249} 1250 1251//--------------------------------------------- 1252//RBBIWithProtectedFunctions class functions 1253//--------------------------------------------- 1254 1255RBBIWithProtectedFunctions::RBBIWithProtectedFunctions(RBBIDataHeader* data, UErrorCode &status) 1256 : RuleBasedBreakIterator(data, status) 1257{ 1258} 1259 1260RBBIWithProtectedFunctions::RBBIWithProtectedFunctions(const RBBIDataHeader* data, enum EDontAdopt, UErrorCode &status) 1261 : RuleBasedBreakIterator(data, RuleBasedBreakIterator::kDontAdopt, status) 1262{ 1263} 1264 1265#endif /* #if !UCONFIG_NO_BREAK_ITERATION */ 1266