]> git.saurik.com Git - apple/icu.git/blame - icuSources/test/cintltst/cbiapts.c
ICU-59180.0.1.tar.gz
[apple/icu.git] / icuSources / test / cintltst / cbiapts.c
CommitLineData
f3c0d7a5
A
1// © 2016 and later: Unicode, Inc. and others.
2// License & terms of use: http://www.unicode.org/copyright.html
b75a7d8f 3/********************************************************************
f3c0d7a5 4 * COPYRIGHT:
2ca993e8 5 * Copyright (c) 1997-2016, International Business Machines Corporation and
b75a7d8f
A
6 * others. All Rights Reserved.
7 ********************************************************************/
8/********************************************************************************
9*
10* File CBIAPTS.C
11*
12* Modification History:
2ca993e8 13* Name Description
b75a7d8f
A
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>
374ca955 28#include <string.h>
b75a7d8f
A
29#include "unicode/uloc.h"
30#include "unicode/ubrk.h"
31#include "unicode/ustring.h"
32#include "unicode/ucnv.h"
73c04bcf 33#include "unicode/utext.h"
b75a7d8f
A
34#include "cintltst.h"
35#include "cbiapts.h"
2ca993e8 36#include "cmemory.h"
b75a7d8f 37
73c04bcf 38#define TEST_ASSERT_SUCCESS(status) {if (U_FAILURE(status)) { \
729e4ab9 39log_data_err("Failure at file %s, line %d, error = %s (Are you missing data?)\n", __FILE__, __LINE__, u_errorName(status));}}
374ca955
A
40
41#define TEST_ASSERT(expr) {if ((expr)==FALSE) { \
729e4ab9 42log_data_err("Test Failure at file %s, line %d (Are you missing data?)\n", __FILE__, __LINE__);}}
374ca955 43
f3c0d7a5
A
44#define APPLE_ADDITIONS 1
45
51004dcb 46#if !UCONFIG_NO_FILE_IO
b75a7d8f 47static void TestBreakIteratorSafeClone(void);
51004dcb 48#endif
b75a7d8f
A
49static void TestBreakIteratorRules(void);
50static void TestBreakIteratorRuleError(void);
374ca955 51static void TestBreakIteratorStatusVec(void);
73c04bcf 52static void TestBreakIteratorUText(void);
729e4ab9 53static void TestBreakIteratorTailoring(void);
4388f060 54static void TestBreakIteratorRefresh(void);
2ca993e8
A
55static void TestBug11665(void);
56static void TestBreakIteratorSuppressions(void);
f3c0d7a5
A
57#if APPLE_ADDITIONS
58static void TestRuleBasedTokenizer(void);
59#endif
b75a7d8f
A
60
61void addBrkIterAPITest(TestNode** root);
62
63void addBrkIterAPITest(TestNode** root)
64{
729e4ab9 65#if !UCONFIG_NO_FILE_IO
b75a7d8f
A
66 addTest(root, &TestBreakIteratorCAPI, "tstxtbd/cbiapts/TestBreakIteratorCAPI");
67 addTest(root, &TestBreakIteratorSafeClone, "tstxtbd/cbiapts/TestBreakIteratorSafeClone");
729e4ab9
A
68 addTest(root, &TestBreakIteratorUText, "tstxtbd/cbiapts/TestBreakIteratorUText");
69#endif
b75a7d8f
A
70 addTest(root, &TestBreakIteratorRules, "tstxtbd/cbiapts/TestBreakIteratorRules");
71 addTest(root, &TestBreakIteratorRuleError, "tstxtbd/cbiapts/TestBreakIteratorRuleError");
374ca955 72 addTest(root, &TestBreakIteratorStatusVec, "tstxtbd/cbiapts/TestBreakIteratorStatusVec");
729e4ab9 73 addTest(root, &TestBreakIteratorTailoring, "tstxtbd/cbiapts/TestBreakIteratorTailoring");
4388f060 74 addTest(root, &TestBreakIteratorRefresh, "tstxtbd/cbiapts/TestBreakIteratorRefresh");
2ca993e8 75 addTest(root, &TestBug11665, "tstxtbd/cbiapts/TestBug11665");
f3c0d7a5 76#if !UCONFIG_NO_FILTERED_BREAK_ITERATION
b331163b 77 addTest(root, &TestBreakIteratorSuppressions, "tstxtbd/cbiapts/TestBreakIteratorSuppressions");
f3c0d7a5
A
78#endif
79#if APPLE_ADDITIONS
80 addTest(root, &TestRuleBasedTokenizer, "tstxtbd/cbiapts/TestRuleBasedTokenizer");
81#endif
b75a7d8f
A
82}
83
84#define CLONETEST_ITERATOR_COUNT 2
85
86/*
87 * Utility function for converting char * to UChar * strings, to
88 * simplify the test code. Converted strings are put in heap allocated
89 * storage. A hook (probably a local in the caller's code) allows all
90 * strings converted with that hook to be freed with a single call.
91 */
92typedef struct StringStruct {
93 struct StringStruct *link;
94 UChar str[1];
95 } StringStruct;
96
97
98static UChar* toUChar(const char *src, void **freeHook) {
99 /* Structure of the memory that we allocate on the heap */
100
101 int32_t numUChars;
102 int32_t destSize;
103 UChar stackBuf[2000 + sizeof(void *)/sizeof(UChar)];
104 StringStruct *dest;
105 UConverter *cnv;
106
107 UErrorCode status = U_ZERO_ERROR;
108 if (src == NULL) {
109 return NULL;
110 };
111
374ca955 112 cnv = ucnv_open(NULL, &status);
b75a7d8f
A
113 if(U_FAILURE(status) || cnv == NULL) {
114 return NULL;
115 }
116 ucnv_reset(cnv);
117 numUChars = ucnv_toUChars(cnv,
118 stackBuf,
119 2000,
120 src, -1,
121 &status);
122
123 destSize = (numUChars+1) * sizeof(UChar) + sizeof(struct StringStruct);
124 dest = (StringStruct *)malloc(destSize);
125 if (dest != NULL) {
126 if (status == U_BUFFER_OVERFLOW_ERROR || status == U_STRING_NOT_TERMINATED_WARNING) {
127 ucnv_toUChars(cnv, dest->str, numUChars+1, src, -1, &status);
128 } else if (status == U_ZERO_ERROR) {
129 u_strcpy(dest->str, stackBuf);
130 } else {
131 free(dest);
132 dest = NULL;
133 }
134 }
135
136 ucnv_reset(cnv); /* be good citizens */
137 ucnv_close(cnv);
138 if (dest == NULL) {
139 return NULL;
140 }
2ca993e8 141
b75a7d8f
A
142 dest->link = (StringStruct*)(*freeHook);
143 *freeHook = dest;
144 return dest->str;
145}
146
147static void freeToUCharStrings(void **hook) {
148 StringStruct *s = *(StringStruct **)hook;
149 while (s != NULL) {
150 StringStruct *next = s->link;
151 free(s);
152 s = next;
153 }
154}
155
156
51004dcb 157#if !UCONFIG_NO_FILE_IO
b75a7d8f
A
158static void TestBreakIteratorCAPI()
159{
160 UErrorCode status = U_ZERO_ERROR;
161 UBreakIterator *word, *sentence, *line, *character, *b, *bogus;
162 int32_t start,pos,end,to;
163 int32_t i;
164 int32_t count = 0;
165
166 UChar text[50];
167
168 /* Note: the adjacent "" are concatenating strings, not adding a \" to the
169 string, which is probably what whoever wrote this intended. Don't fix,
170 because it would throw off the hard coded break positions in the following
171 tests. */
172 u_uastrcpy(text, "He's from Africa. ""Mr. Livingston, I presume?"" Yeah");
173
174
175/*test ubrk_open()*/
176 log_verbose("\nTesting BreakIterator open functions\n");
2ca993e8 177
b75a7d8f
A
178 /* Use french for fun */
179 word = ubrk_open(UBRK_WORD, "en_US", text, u_strlen(text), &status);
180 if(status == U_FILE_ACCESS_ERROR) {
374ca955
A
181 log_data_err("Check your data - it doesn't seem to be around\n");
182 return;
b75a7d8f 183 } else if(U_FAILURE(status)){
729e4ab9 184 log_err_status(status, "FAIL: Error in ubrk_open() for word breakiterator: %s\n", myErrorName(status));
b75a7d8f
A
185 }
186 else{
187 log_verbose("PASS: Successfully opened word breakiterator\n");
188 }
2ca993e8 189
b75a7d8f
A
190 sentence = ubrk_open(UBRK_SENTENCE, "en_US", text, u_strlen(text), &status);
191 if(U_FAILURE(status)){
729e4ab9 192 log_err_status(status, "FAIL: Error in ubrk_open() for sentence breakiterator: %s\n", myErrorName(status));
374ca955 193 return;
b75a7d8f
A
194 }
195 else{
196 log_verbose("PASS: Successfully opened sentence breakiterator\n");
197 }
2ca993e8 198
b75a7d8f
A
199 line = ubrk_open(UBRK_LINE, "en_US", text, u_strlen(text), &status);
200 if(U_FAILURE(status)){
201 log_err("FAIL: Error in ubrk_open() for line breakiterator: %s\n", myErrorName(status));
374ca955 202 return;
b75a7d8f
A
203 }
204 else{
205 log_verbose("PASS: Successfully opened line breakiterator\n");
206 }
2ca993e8 207
b75a7d8f
A
208 character = ubrk_open(UBRK_CHARACTER, "en_US", text, u_strlen(text), &status);
209 if(U_FAILURE(status)){
210 log_err("FAIL: Error in ubrk_open() for character breakiterator: %s\n", myErrorName(status));
374ca955 211 return;
b75a7d8f
A
212 }
213 else{
214 log_verbose("PASS: Successfully opened character breakiterator\n");
215 }
216 /*trying to open an illegal iterator*/
217 bogus = ubrk_open((UBreakIteratorType)5, "en_US", text, u_strlen(text), &status);
57a6839d
A
218 if(bogus != NULL) {
219 log_err("FAIL: expected NULL from opening an invalid break iterator.\n");
220 }
b75a7d8f 221 if(U_SUCCESS(status)){
73c04bcf 222 log_err("FAIL: Error in ubrk_open() for BOGUS breakiterator. Expected U_ILLEGAL_ARGUMENT_ERROR\n");
b75a7d8f
A
223 }
224 if(U_FAILURE(status)){
73c04bcf
A
225 if(status != U_ILLEGAL_ARGUMENT_ERROR){
226 log_err("FAIL: Error in ubrk_open() for BOGUS breakiterator. Expected U_ILLEGAL_ARGUMENT_ERROR\n Got %s\n", myErrorName(status));
b75a7d8f
A
227 }
228 }
229 status=U_ZERO_ERROR;
230
231
232/* ======= Test ubrk_countAvialable() and ubrk_getAvialable() */
233
234 log_verbose("\nTesting ubrk_countAvailable() and ubrk_getAvailable()\n");
235 count=ubrk_countAvailable();
236 /* use something sensible w/o hardcoding the count */
237 if(count < 0){
238 log_err("FAIL: Error in ubrk_countAvialable() returned %d\n", count);
239 }
240 else{
241 log_verbose("PASS: ubrk_countAvialable() successful returned %d\n", count);
242 }
243 for(i=0;i<count;i++)
244 {
2ca993e8 245 log_verbose("%s\n", ubrk_getAvailable(i));
b75a7d8f
A
246 if (ubrk_getAvailable(i) == 0)
247 log_err("No locale for which breakiterator is applicable\n");
2ca993e8 248 else
b75a7d8f
A
249 log_verbose("A locale %s for which breakiterator is applicable\n",ubrk_getAvailable(i));
250 }
251
252/*========Test ubrk_first(), ubrk_last()...... and other functions*/
253
254 log_verbose("\nTesting the functions for word\n");
255 start = ubrk_first(word);
256 if(start!=0)
257 log_err("error ubrk_start(word) did not return 0\n");
258 log_verbose("first (word = %d\n", (int32_t)start);
259 pos=ubrk_next(word);
260 if(pos!=4)
261 log_err("error ubrk_next(word) did not return 4\n");
262 log_verbose("next (word = %d\n", (int32_t)pos);
263 pos=ubrk_following(word, 4);
264 if(pos!=5)
265 log_err("error ubrl_following(word,4) did not return 6\n");
266 log_verbose("next (word = %d\n", (int32_t)pos);
267 end=ubrk_last(word);
268 if(end!=49)
269 log_err("error ubrk_last(word) did not return 49\n");
270 log_verbose("last (word = %d\n", (int32_t)end);
2ca993e8 271
b75a7d8f
A
272 pos=ubrk_previous(word);
273 log_verbose("%d %d\n", end, pos);
2ca993e8 274
b75a7d8f
A
275 pos=ubrk_previous(word);
276 log_verbose("%d \n", pos);
277
278 if (ubrk_isBoundary(word, 2) != FALSE) {
279 log_err("error ubrk_isBoundary(word, 2) did not return FALSE\n");
280 }
281 pos=ubrk_current(word);
282 if (pos != 4) {
283 log_err("error ubrk_current() != 4 after ubrk_isBoundary(word, 2)\n");
284 }
285 if (ubrk_isBoundary(word, 4) != TRUE) {
286 log_err("error ubrk_isBoundary(word, 4) did not return TRUE\n");
287 }
288
289
2ca993e8 290
b75a7d8f
A
291 log_verbose("\nTesting the functions for character\n");
292 ubrk_first(character);
293 pos = ubrk_following(character, 5);
294 if(pos!=6)
295 log_err("error ubrk_following(character,5) did not return 6\n");
296 log_verbose("Following (character,5) = %d\n", (int32_t)pos);
297 pos=ubrk_following(character, 18);
298 if(pos!=19)
299 log_err("error ubrk_following(character,18) did not return 19\n");
300 log_verbose("Followingcharacter,18) = %d\n", (int32_t)pos);
301 pos=ubrk_preceding(character, 22);
302 if(pos!=21)
303 log_err("error ubrk_preceding(character,22) did not return 21\n");
304 log_verbose("preceding(character,22) = %d\n", (int32_t)pos);
2ca993e8 305
b75a7d8f
A
306
307 log_verbose("\nTesting the functions for line\n");
308 pos=ubrk_first(line);
309 if(pos != 0)
310 log_err("error ubrk_first(line) returned %d, expected 0\n", (int32_t)pos);
311 pos = ubrk_next(line);
312 pos=ubrk_following(line, 18);
313 if(pos!=22)
314 log_err("error ubrk_following(line) did not return 22\n");
315 log_verbose("following (line) = %d\n", (int32_t)pos);
316
2ca993e8 317
b75a7d8f
A
318 log_verbose("\nTesting the functions for sentence\n");
319 ubrk_first(sentence);
320 pos = ubrk_current(sentence);
321 log_verbose("Current(sentence) = %d\n", (int32_t)pos);
322 pos = ubrk_last(sentence);
323 if(pos!=49)
324 log_err("error ubrk_last for sentence did not return 49\n");
325 log_verbose("Last (sentence) = %d\n", (int32_t)pos);
326 ubrk_first(sentence);
327 to = ubrk_following( sentence, 0 );
328 if (to == 0) log_err("ubrk_following returned 0\n");
329 to = ubrk_preceding( sentence, to );
330 if (to != 0) log_err("ubrk_preceding didn't return 0\n");
331 if (ubrk_first(sentence)!=ubrk_current(sentence)) {
332 log_err("error in ubrk_first() or ubrk_current()\n");
333 }
2ca993e8
A
334
335
b75a7d8f 336 /*---- */
73c04bcf 337 /*Testing ubrk_open and ubrk_close()*/
b75a7d8f
A
338 log_verbose("\nTesting open and close for us locale\n");
339 b = ubrk_open(UBRK_WORD, "fr_FR", text, u_strlen(text), &status);
340 if (U_FAILURE(status)) {
341 log_err("ubrk_open for word returned NULL: %s\n", myErrorName(status));
342 }
343 ubrk_close(b);
344
73c04bcf
A
345 /* Test setText and setUText */
346 {
347 UChar s1[] = {0x41, 0x42, 0x20, 0};
348 UChar s2[] = {0x41, 0x42, 0x43, 0x44, 0x45, 0};
349 UText *ut = NULL;
350 UBreakIterator *bb;
351 int j;
352
353 log_verbose("\nTesting ubrk_setText() and ubrk_setUText()\n");
354 status = U_ZERO_ERROR;
355 bb = ubrk_open(UBRK_WORD, "en_US", NULL, 0, &status);
356 TEST_ASSERT_SUCCESS(status);
357 ubrk_setText(bb, s1, -1, &status);
358 TEST_ASSERT_SUCCESS(status);
359 ubrk_first(bb);
360 j = ubrk_next(bb);
361 TEST_ASSERT(j == 2);
362 ut = utext_openUChars(ut, s2, -1, &status);
363 ubrk_setUText(bb, ut, &status);
364 TEST_ASSERT_SUCCESS(status);
365 j = ubrk_next(bb);
366 TEST_ASSERT(j == 5);
367
368 ubrk_close(bb);
369 utext_close(ut);
370 }
371
b75a7d8f
A
372 ubrk_close(word);
373 ubrk_close(sentence);
374 ubrk_close(line);
375 ubrk_close(character);
376}
377
378static void TestBreakIteratorSafeClone(void)
379{
380 UChar text[51]; /* Keep this odd to test for 64-bit memory alignment */
2ca993e8 381 /* NOTE: This doesn't reliably force mis-alignment of following items. */
b75a7d8f
A
382 uint8_t buffer [CLONETEST_ITERATOR_COUNT] [U_BRK_SAFECLONE_BUFFERSIZE];
383 int32_t bufferSize = U_BRK_SAFECLONE_BUFFERSIZE;
384
385 UBreakIterator * someIterators [CLONETEST_ITERATOR_COUNT];
386 UBreakIterator * someClonedIterators [CLONETEST_ITERATOR_COUNT];
387
388 UBreakIterator * brk;
389 UErrorCode status = U_ZERO_ERROR;
390 int32_t start,pos;
391 int32_t i;
392
393 /*Testing ubrk_safeClone */
394
395 /* Note: the adjacent "" are concatenating strings, not adding a \" to the
396 string, which is probably what whoever wrote this intended. Don't fix,
397 because it would throw off the hard coded break positions in the following
398 tests. */
399 u_uastrcpy(text, "He's from Africa. ""Mr. Livingston, I presume?"" Yeah");
400
401 /* US & Thai - rule-based & dictionary based */
402 someIterators[0] = ubrk_open(UBRK_WORD, "en_US", text, u_strlen(text), &status);
403 if(!someIterators[0] || U_FAILURE(status)) {
404 log_data_err("Couldn't open en_US word break iterator - %s\n", u_errorName(status));
405 return;
406 }
407
408 someIterators[1] = ubrk_open(UBRK_WORD, "th_TH", text, u_strlen(text), &status);
409 if(!someIterators[1] || U_FAILURE(status)) {
410 log_data_err("Couldn't open th_TH word break iterator - %s\n", u_errorName(status));
411 return;
412 }
413
414 /* test each type of iterator */
415 for (i = 0; i < CLONETEST_ITERATOR_COUNT; i++)
416 {
417
418 /* Check the various error & informational states */
419
420 /* Null status - just returns NULL */
57a6839d 421 if (NULL != ubrk_safeClone(someIterators[i], buffer[i], &bufferSize, NULL))
b75a7d8f
A
422 {
423 log_err("FAIL: Cloned Iterator failed to deal correctly with null status\n");
424 }
425 /* error status - should return 0 & keep error the same */
426 status = U_MEMORY_ALLOCATION_ERROR;
57a6839d 427 if (NULL != ubrk_safeClone(someIterators[i], buffer[i], &bufferSize, &status) || status != U_MEMORY_ALLOCATION_ERROR)
b75a7d8f
A
428 {
429 log_err("FAIL: Cloned Iterator failed to deal correctly with incoming error status\n");
430 }
431 status = U_ZERO_ERROR;
432
57a6839d
A
433 /* Null buffer size pointer is ok */
434 if (NULL == (brk = ubrk_safeClone(someIterators[i], buffer[i], NULL, &status)) || U_FAILURE(status))
b75a7d8f
A
435 {
436 log_err("FAIL: Cloned Iterator failed to deal correctly with null bufferSize pointer\n");
437 }
57a6839d 438 ubrk_close(brk);
b75a7d8f 439 status = U_ZERO_ERROR;
57a6839d 440
b75a7d8f
A
441 /* buffer size pointer is 0 - fill in pbufferSize with a size */
442 bufferSize = 0;
57a6839d
A
443 if (NULL != ubrk_safeClone(someIterators[i], buffer[i], &bufferSize, &status) ||
444 U_FAILURE(status) || bufferSize <= 0)
b75a7d8f
A
445 {
446 log_err("FAIL: Cloned Iterator failed a sizing request ('preflighting')\n");
447 }
448 /* Verify our define is large enough */
449 if (U_BRK_SAFECLONE_BUFFERSIZE < bufferSize)
450 {
4388f060 451 log_err("FAIL: Pre-calculated buffer size is too small - %d but needed %d\n", U_BRK_SAFECLONE_BUFFERSIZE, bufferSize);
b75a7d8f
A
452 }
453 /* Verify we can use this run-time calculated size */
57a6839d 454 if (NULL == (brk = ubrk_safeClone(someIterators[i], buffer[i], &bufferSize, &status)) || U_FAILURE(status))
b75a7d8f
A
455 {
456 log_err("FAIL: Iterator can't be cloned with run-time size\n");
457 }
458 if (brk)
459 ubrk_close(brk);
460 /* size one byte too small - should allocate & let us know */
57a6839d
A
461 if (bufferSize > 1) {
462 --bufferSize;
463 }
464 if (NULL == (brk = ubrk_safeClone(someIterators[i], NULL, &bufferSize, &status)) || status != U_SAFECLONE_ALLOCATED_WARNING)
b75a7d8f
A
465 {
466 log_err("FAIL: Cloned Iterator failed to deal correctly with too-small buffer size\n");
467 }
468 if (brk)
469 ubrk_close(brk);
470 status = U_ZERO_ERROR;
471 bufferSize = U_BRK_SAFECLONE_BUFFERSIZE;
472
473 /* Null buffer pointer - return Iterator & set error to U_SAFECLONE_ALLOCATED_ERROR */
57a6839d 474 if (NULL == (brk = ubrk_safeClone(someIterators[i], NULL, &bufferSize, &status)) || status != U_SAFECLONE_ALLOCATED_WARNING)
b75a7d8f
A
475 {
476 log_err("FAIL: Cloned Iterator failed to deal correctly with null buffer pointer\n");
477 }
478 if (brk)
479 ubrk_close(brk);
480 status = U_ZERO_ERROR;
481
482 /* Mis-aligned buffer pointer. */
483 {
484 char stackBuf[U_BRK_SAFECLONE_BUFFERSIZE+sizeof(void *)];
b75a7d8f
A
485
486 brk = ubrk_safeClone(someIterators[i], &stackBuf[1], &bufferSize, &status);
57a6839d 487 if (U_FAILURE(status) || brk == NULL) {
b75a7d8f
A
488 log_err("FAIL: Cloned Iterator failed with misaligned buffer pointer\n");
489 }
490 if (status == U_SAFECLONE_ALLOCATED_WARNING) {
57a6839d 491 log_verbose("Cloned Iterator allocated when using a mis-aligned buffer.\n");
b75a7d8f
A
492 }
493 if (brk)
494 ubrk_close(brk);
495 }
496
497
498 /* Null Iterator - return NULL & set U_ILLEGAL_ARGUMENT_ERROR */
57a6839d 499 if (NULL != ubrk_safeClone(NULL, buffer[i], &bufferSize, &status) || status != U_ILLEGAL_ARGUMENT_ERROR)
b75a7d8f
A
500 {
501 log_err("FAIL: Cloned Iterator failed to deal correctly with null Iterator pointer\n");
502 }
503 status = U_ZERO_ERROR;
504
505 /* Do these cloned Iterators work at all - make a first & next call */
506 bufferSize = U_BRK_SAFECLONE_BUFFERSIZE;
507 someClonedIterators[i] = ubrk_safeClone(someIterators[i], buffer[i], &bufferSize, &status);
508
509 start = ubrk_first(someClonedIterators[i]);
510 if(start!=0)
511 log_err("error ubrk_start(clone) did not return 0\n");
512 pos=ubrk_next(someClonedIterators[i]);
513 if(pos!=4)
514 log_err("error ubrk_next(clone) did not return 4\n");
515
516 ubrk_close(someClonedIterators[i]);
517 ubrk_close(someIterators[i]);
518 }
519}
51004dcb 520#endif
b75a7d8f
A
521
522
523/*
524// Open a break iterator from char * rules. Take care of conversion
525// of the rules and error checking.
526*/
527static UBreakIterator * testOpenRules(char *rules) {
528 UErrorCode status = U_ZERO_ERROR;
529 UChar *ruleSourceU = NULL;
530 void *strCleanUp = NULL;
531 UParseError parseErr;
532 UBreakIterator *bi;
533
534 ruleSourceU = toUChar(rules, &strCleanUp);
535
536 bi = ubrk_openRules(ruleSourceU, -1, /* The rules */
537 NULL, -1, /* The text to be iterated over. */
538 &parseErr, &status);
2ca993e8 539
b75a7d8f 540 if (U_FAILURE(status)) {
729e4ab9 541 log_data_err("FAIL: ubrk_openRules: ICU Error \"%s\" (Are you missing data?)\n", u_errorName(status));
b75a7d8f
A
542 bi = 0;
543 };
544 freeToUCharStrings(&strCleanUp);
545 return bi;
546
547}
548
549/*
550 * TestBreakIteratorRules - Verify that a break iterator can be created from
551 * a set of source rules.
552 */
553static void TestBreakIteratorRules() {
554 /* Rules will keep together any run of letters not including 'a', OR
555 * keep together 'abc', but only when followed by 'def', OTHERWISE
556 * just return one char at a time.
557 */
2ca993e8 558 char rules[] = "abc/def{666};\n [\\p{L} - [a]]* {2}; . {1};";
b75a7d8f
A
559 /* 0123456789012345678 */
560 char data[] = "abcdex abcdefgh-def"; /* the test data string */
561 char breaks[] = "** ** * ** *"; /* * the expected break positions */
562 char tags[] = "01 21 6 21 2"; /* expected tag values at break positions */
563 int32_t tagMap[] = {0, 1, 2, 3, 4, 5, 666};
564
565 UChar *uData;
566 void *freeHook = NULL;
567 UErrorCode status = U_ZERO_ERROR;
568 int32_t pos;
569 int i;
570
571 UBreakIterator *bi = testOpenRules(rules);
572 if (bi == NULL) {return;}
573 uData = toUChar(data, &freeHook);
574 ubrk_setText(bi, uData, -1, &status);
575
576 pos = ubrk_first(bi);
577 for (i=0; i<sizeof(breaks); i++) {
578 if (pos == i && breaks[i] != '*') {
579 log_err("FAIL: unexpected break at position %d found\n", pos);
580 break;
581 }
582 if (pos != i && breaks[i] == '*') {
583 log_err("FAIL: expected break at position %d not found.\n", i);
584 break;
585 }
586 if (pos == i) {
587 int32_t tag, expectedTag;
588 tag = ubrk_getRuleStatus(bi);
589 expectedTag = tagMap[tags[i]&0xf];
590 if (tag != expectedTag) {
591 log_err("FAIL: incorrect tag value. Position = %d; expected tag %d, got %d",
592 pos, expectedTag, tag);
593 break;
594 }
595 pos = ubrk_next(bi);
596 }
597 }
598
f3c0d7a5
A
599 /* #12914 add basic sanity test for ubrk_getBinaryRules, ubrk_openBinaryRules */
600 /* Underlying functionality checked in C++ rbbiapts.cpp TestRoundtripRules */
601 status = U_ZERO_ERROR;
602 int32_t rulesLength = ubrk_getBinaryRules(bi, NULL, 0, &status); /* preflight */
603 if (U_FAILURE(status)) {
604 log_err("FAIL: ubrk_getBinaryRules preflight err: %s", u_errorName(status));
605 } else {
606 uint8_t* binaryRules = (uint8_t*)uprv_malloc(rulesLength);
607 if (binaryRules == NULL) {
608 log_err("FAIL: unable to malloc rules buffer, size %u", rulesLength);
609 } else {
610 rulesLength = ubrk_getBinaryRules(bi, binaryRules, rulesLength, &status);
611 if (U_FAILURE(status)) {
612 log_err("FAIL: ubrk_getBinaryRules err: %s", u_errorName(status));
613 } else {
614 UBreakIterator* bi2 = ubrk_openBinaryRules(binaryRules, rulesLength, uData, -1, &status);
615 if (U_FAILURE(status)) {
616 log_err("FAIL: ubrk_openBinaryRules err: %s", u_errorName(status));
617 } else {
618 int32_t maxCount = sizeof(breaks); /* fail-safe test limit */
619 int32_t pos2 = ubrk_first(bi2);
620 pos = ubrk_first(bi);
621 do {
622 if (pos2 != pos) {
623 log_err("FAIL: interator from ubrk_openBinaryRules does not match original, get pos = %d instead of %d", pos2, pos);
624 }
625 pos2 = ubrk_next(bi2);
626 pos = ubrk_next(bi);
627 } while ((pos != UBRK_DONE || pos2 != UBRK_DONE) && maxCount-- > 0);
628
629 ubrk_close(bi2);
630 }
631 }
632 uprv_free(binaryRules);
633 }
634 }
635
b75a7d8f
A
636 freeToUCharStrings(&freeHook);
637 ubrk_close(bi);
638}
639
640static void TestBreakIteratorRuleError() {
641/*
642 * TestBreakIteratorRuleError - Try to create a BI from rules with syntax errors,
643 * check that the error is reported correctly.
644 */
645 char rules[] = " # This is a rule comment on line 1\n"
646 "[:L:]; # this rule is OK.\n"
647 "abcdefg); # Error, mismatched parens\n";
648 UChar *uRules;
649 void *freeHook = NULL;
650 UErrorCode status = U_ZERO_ERROR;
651 UParseError parseErr;
652 UBreakIterator *bi;
653
654 uRules = toUChar(rules, &freeHook);
655 bi = ubrk_openRules(uRules, -1, /* The rules */
656 NULL, -1, /* The text to be iterated over. */
657 &parseErr, &status);
658 if (U_SUCCESS(status)) {
659 log_err("FAIL: construction of break iterator succeeded when it should have failed.\n");
660 ubrk_close(bi);
661 } else {
662 if (parseErr.line != 3 || parseErr.offset != 8) {
729e4ab9 663 log_data_err("FAIL: incorrect error position reported. Got line %d, char %d, expected line 3, char 7 (Are you missing data?)\n",
b75a7d8f
A
664 parseErr.line, parseErr.offset);
665 }
666 }
667 freeToUCharStrings(&freeHook);
668}
669
374ca955
A
670
671/*
672* TestsBreakIteratorStatusVals() Test the ubrk_getRuleStatusVec() funciton
673*/
674static void TestBreakIteratorStatusVec() {
675 #define RULE_STRING_LENGTH 200
676 UChar rules[RULE_STRING_LENGTH];
677
678 #define TEST_STRING_LENGTH 25
679 UChar testString[TEST_STRING_LENGTH];
680 UBreakIterator *bi = NULL;
681 int32_t pos = 0;
682 int32_t vals[10];
683 int32_t numVals;
684 UErrorCode status = U_ZERO_ERROR;
685
686 u_uastrncpy(rules, "[A-N]{100}; \n"
687 "[a-w]{200}; \n"
688 "[\\p{L}]{300}; \n"
689 "[\\p{N}]{400}; \n"
690 "[0-5]{500}; \n"
691 "!.*;\n", RULE_STRING_LENGTH);
692 u_uastrncpy(testString, "ABC", TEST_STRING_LENGTH);
693
694
695 bi = ubrk_openRules(rules, -1, testString, -1, NULL, &status);
73c04bcf 696 TEST_ASSERT_SUCCESS(status);
374ca955
A
697 TEST_ASSERT(bi != NULL);
698
73c04bcf
A
699 /* The TEST_ASSERT above should change too... */
700 if (bi != NULL) {
701 pos = ubrk_next(bi);
702 TEST_ASSERT(pos == 1);
703
704 memset(vals, -1, sizeof(vals));
705 numVals = ubrk_getRuleStatusVec(bi, vals, 10, &status);
706 TEST_ASSERT_SUCCESS(status);
707 TEST_ASSERT(numVals == 2);
708 TEST_ASSERT(vals[0] == 100);
709 TEST_ASSERT(vals[1] == 300);
710 TEST_ASSERT(vals[2] == -1);
711
712 numVals = ubrk_getRuleStatusVec(bi, vals, 0, &status);
713 TEST_ASSERT(status == U_BUFFER_OVERFLOW_ERROR);
714 TEST_ASSERT(numVals == 2);
715 }
716
717 ubrk_close(bi);
718}
719
720
721/*
722 * static void TestBreakIteratorUText(void);
723 *
724 * Test that ubrk_setUText() is present and works for a simple case.
725 */
726static void TestBreakIteratorUText(void) {
727 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 */
728 /* 0 1 2 34567890 */
729
730 UErrorCode status = U_ZERO_ERROR;
731 UBreakIterator *bi = NULL;
732 int32_t pos = 0;
733
734
735 UText *ut = utext_openUTF8(NULL, UTF8Str, -1, &status);
736 TEST_ASSERT_SUCCESS(status);
737
738 bi = ubrk_open(UBRK_WORD, "en_US", NULL, 0, &status);
739 if (U_FAILURE(status)) {
729e4ab9 740 log_err_status(status, "Failure at file %s, line %d, error = %s\n", __FILE__, __LINE__, u_errorName(status));
73c04bcf
A
741 return;
742 }
743
744 ubrk_setUText(bi, ut, &status);
745 if (U_FAILURE(status)) {
746 log_err("Failure at file %s, line %d, error = %s\n", __FILE__, __LINE__, u_errorName(status));
747 return;
748 }
749
750 pos = ubrk_first(bi);
751 TEST_ASSERT(pos == 0);
752
374ca955 753 pos = ubrk_next(bi);
73c04bcf 754 TEST_ASSERT(pos == 4);
374ca955 755
73c04bcf
A
756 pos = ubrk_next(bi);
757 TEST_ASSERT(pos == 5);
374ca955 758
73c04bcf
A
759 pos = ubrk_next(bi);
760 TEST_ASSERT(pos == 10);
374ca955 761
73c04bcf
A
762 pos = ubrk_next(bi);
763 TEST_ASSERT(pos == UBRK_DONE);
374ca955 764 ubrk_close(bi);
73c04bcf 765 utext_close(ut);
374ca955
A
766}
767
729e4ab9
A
768/*
769 * static void TestBreakIteratorTailoring(void);
770 *
771 * Test break iterator tailorings from CLDR data.
772 */
773
774/* Thai/Lao grapheme break tailoring */
775static const UChar thTest[] = { 0x0020, 0x0E40, 0x0E01, 0x0020,
776 0x0E01, 0x0E30, 0x0020, 0x0E01, 0x0E33, 0x0020, 0 };
4388f060
A
777/*in Unicode 6.1 en should behave just like th for this*/
778/*static const int32_t thTestOffs_enFwd[] = { 1, 3, 4, 6, 7, 9, 10 };*/
729e4ab9 779static const int32_t thTestOffs_thFwd[] = { 1, 2, 3, 4, 5, 6, 7, 9, 10 };
4388f060 780/*static const int32_t thTestOffs_enRev[] = { 9, 7, 6, 4, 3, 1, 0 };*/
729e4ab9
A
781static const int32_t thTestOffs_thRev[] = { 9, 7, 6, 5, 4, 3, 2, 1, 0 };
782
783/* Hebrew line break tailoring, for cldrbug 3028 */
784static const UChar heTest[] = { 0x0020, 0x002D, 0x0031, 0x0032, 0x0020,
785 0x0061, 0x002D, 0x006B, 0x0020,
786 0x0061, 0x0300, 0x2010, 0x006B, 0x0020,
787 0x05DE, 0x05D4, 0x002D, 0x0069, 0x0020,
788 0x05D1, 0x05BC, 0x2010, 0x0047, 0x0020, 0 };
4388f060
A
789/*in Unicode 6.1 en should behave just like he for this*/
790/*static const int32_t heTestOffs_enFwd[] = { 1, 5, 7, 9, 12, 14, 17, 19, 22, 24 };*/
729e4ab9 791static const int32_t heTestOffs_heFwd[] = { 1, 5, 7, 9, 12, 14, 19, 24 };
4388f060 792/*static const int32_t heTestOffs_enRev[] = { 22, 19, 17, 14, 12, 9, 7, 5, 1, 0 };*/
729e4ab9
A
793static const int32_t heTestOffs_heRev[] = { 19, 14, 12, 9, 7, 5, 1, 0 };
794
795/* Finnish line break tailoring, for cldrbug 3029 */
796static const UChar fiTest[] = { /* 00 */ 0x0020, 0x002D, 0x0031, 0x0032, 0x0020,
797 /* 05 */ 0x0061, 0x002D, 0x006B, 0x0020,
798 /* 09 */ 0x0061, 0x0300, 0x2010, 0x006B, 0x0020,
799 /* 14 */ 0x0061, 0x0020, 0x002D, 0x006B, 0x0020,
800 /* 19 */ 0x0061, 0x0300, 0x0020, 0x2010, 0x006B, 0x0020, 0 };
801static const int32_t fiTestOffs_enFwd[] = { 1, 5, 7, 9, 12, 14, 16, 17, 19, 22, 23, 25 };
802static const int32_t fiTestOffs_fiFwd[] = { 1, 5, 7, 9, 12, 14, 16, 19, 22, 25 };
803static const int32_t fiTestOffs_enRev[] = { 23, 22, 19, 17, 16, 14, 12, 9, 7, 5, 1, 0 };
804static const int32_t fiTestOffs_fiRev[] = { 22, 19, 16, 14, 12, 9, 7, 5, 1, 0 };
805
4388f060
A
806/* Khmer dictionary-based work break, for ICU ticket #8329 */
807static const UChar kmTest[] = { /* 00 */ 0x179F, 0x17BC, 0x1798, 0x1785, 0x17C6, 0x178E, 0x17B6, 0x1799, 0x1796, 0x17C1,
808 /* 10 */ 0x179B, 0x1794, 0x1793, 0x17D2, 0x178F, 0x17B7, 0x1785, 0x178A, 0x17BE, 0x1798,
809 /* 20 */ 0x17D2, 0x1794, 0x17B8, 0x17A2, 0x1792, 0x17B7, 0x179F, 0x17D2, 0x178B, 0x17B6,
810 /* 30 */ 0x1793, 0x17A2, 0x179A, 0x1796, 0x17D2, 0x179A, 0x17C7, 0x1782, 0x17BB, 0x178E,
811 /* 40 */ 0x178A, 0x179B, 0x17CB, 0x1796, 0x17D2, 0x179A, 0x17C7, 0x17A2, 0x1784, 0x17D2,
812 /* 50 */ 0x1782, 0 };
813static const int32_t kmTestOffs_kmFwd[] = { 3, /*8,*/ 11, 17, 23, 31, /*33,*/ 40, 43, 51 }; /* TODO: Investigate failure to break at offset 8 */
814static const int32_t kmTestOffs_kmRev[] = { 43, 40, /*33,*/ 31, 23, 17, 11, /*8,*/ 3, 0 };
815
f3c0d7a5
A
816
817/* Korean keepAll vs Normal */
818static const UChar koTest[] = { /* 00 */ 0xBAA8, 0xB4E0, 0x0020, 0xC778, 0xB958, 0x0020, 0xAD6C, 0xC131, 0xC6D0, 0xC758,
819 /* 10 */ 0x0020, 0xCC9C, 0xBD80, 0xC758, 0x0020, 0xC874, 0xC5C4, 0xC131, 0xACFC, 0x0020,
820 /* 20 */ 0xB3D9, 0xB4F1, 0xD558, 0xACE0, 0x0020, 0xC591, 0xB3C4, 0xD560, 0 };
821static const int32_t koTestOffs_koKeepFwd[] = { 3, 6, 11, 15, 20, 25, 28 };
822static const int32_t koTestOffs_koKeepRev[] = { 25, 20, 15, 11, 6, 3, 0 };
823static const int32_t koTestOffs_koNormFwd[] = { 1, 3, 4, 6, 7, 8, 9, 11, 12, 13, 15, 16, 17, 18, 20, 21, 22, 23, 25, 26, 27, 28 };
824static const int32_t koTestOffs_koNormRev[] = { 27, 26, 25, 23, 22, 21, 20, 18, 17, 16, 15, 13, 12, 11, 9, 8, 7, 6, 4, 3, 1, 0 };
825
729e4ab9
A
826typedef struct {
827 const char * locale;
828 UBreakIteratorType type;
829 const UChar * test;
830 const int32_t * offsFwd;
831 const int32_t * offsRev;
832 int32_t numOffsets;
833} RBBITailoringTest;
834
835static const RBBITailoringTest tailoringTests[] = {
f3c0d7a5
A
836 { "en", UBRK_CHARACTER, thTest, thTestOffs_thFwd, thTestOffs_thRev, UPRV_LENGTHOF(thTestOffs_thFwd) },
837 { "en_US_POSIX", UBRK_CHARACTER, thTest, thTestOffs_thFwd, thTestOffs_thRev, UPRV_LENGTHOF(thTestOffs_thFwd) },
838 { "en", UBRK_LINE, heTest, heTestOffs_heFwd, heTestOffs_heRev, UPRV_LENGTHOF(heTestOffs_heFwd) },
839 { "he", UBRK_LINE, heTest, heTestOffs_heFwd, heTestOffs_heRev, UPRV_LENGTHOF(heTestOffs_heFwd) },
840 { "en", UBRK_LINE, fiTest, fiTestOffs_enFwd, fiTestOffs_enRev, UPRV_LENGTHOF(fiTestOffs_enFwd) },
841 { "fi", UBRK_LINE, fiTest, fiTestOffs_fiFwd, fiTestOffs_fiRev, UPRV_LENGTHOF(fiTestOffs_fiFwd) },
842 { "km", UBRK_WORD, kmTest, kmTestOffs_kmFwd, kmTestOffs_kmRev, UPRV_LENGTHOF(kmTestOffs_kmFwd) },
9f1b1155 843 { "ko", UBRK_LINE, koTest, koTestOffs_koNormFwd, koTestOffs_koNormRev, UPRV_LENGTHOF(koTestOffs_koNormFwd) },
f3c0d7a5
A
844 { "ko@lw=keepall", UBRK_LINE, koTest, koTestOffs_koKeepFwd, koTestOffs_koKeepRev, UPRV_LENGTHOF(koTestOffs_koKeepFwd) },
845 { "ko@lw=normal", UBRK_LINE, koTest, koTestOffs_koNormFwd, koTestOffs_koNormRev, UPRV_LENGTHOF(koTestOffs_koNormFwd) },
729e4ab9
A
846 { NULL, 0, NULL, NULL, NULL, 0 },
847};
848
849static void TestBreakIteratorTailoring(void) {
850 const RBBITailoringTest * testPtr;
851 for (testPtr = tailoringTests; testPtr->locale != NULL; ++testPtr) {
852 UErrorCode status = U_ZERO_ERROR;
853 UBreakIterator* ubrkiter = ubrk_open(testPtr->type, testPtr->locale, testPtr->test, -1, &status);
854 if ( U_SUCCESS(status) ) {
855 int32_t offset, offsindx;
856 UBool foundError;
857
858 foundError = FALSE;
f3c0d7a5 859 ubrk_first(ubrkiter);
729e4ab9
A
860 for (offsindx = 0; (offset = ubrk_next(ubrkiter)) != UBRK_DONE; ++offsindx) {
861 if (!foundError && offsindx >= testPtr->numOffsets) {
862 log_err("FAIL: locale %s, break type %d, ubrk_next expected UBRK_DONE, got %d\n",
863 testPtr->locale, testPtr->type, offset);
864 foundError = TRUE;
865 } else if (!foundError && offset != testPtr->offsFwd[offsindx]) {
866 log_err("FAIL: locale %s, break type %d, ubrk_next expected %d, got %d\n",
867 testPtr->locale, testPtr->type, testPtr->offsFwd[offsindx], offset);
868 foundError = TRUE;
869 }
870 }
871 if (!foundError && offsindx < testPtr->numOffsets) {
872 log_err("FAIL: locale %s, break type %d, ubrk_next expected %d, got UBRK_DONE\n",
f3c0d7a5 873 testPtr->locale, testPtr->type, testPtr->offsFwd[offsindx]);
729e4ab9
A
874 }
875
876 foundError = FALSE;
f3c0d7a5 877 ubrk_last(ubrkiter);
729e4ab9
A
878 for (offsindx = 0; (offset = ubrk_previous(ubrkiter)) != UBRK_DONE; ++offsindx) {
879 if (!foundError && offsindx >= testPtr->numOffsets) {
880 log_err("FAIL: locale %s, break type %d, ubrk_previous expected UBRK_DONE, got %d\n",
881 testPtr->locale, testPtr->type, offset);
882 foundError = TRUE;
883 } else if (!foundError && offset != testPtr->offsRev[offsindx]) {
884 log_err("FAIL: locale %s, break type %d, ubrk_previous expected %d, got %d\n",
885 testPtr->locale, testPtr->type, testPtr->offsRev[offsindx], offset);
886 foundError = TRUE;
887 }
888 }
889 if (!foundError && offsindx < testPtr->numOffsets) {
890 log_err("FAIL: locale %s, break type %d, ubrk_previous expected %d, got UBRK_DONE\n",
f3c0d7a5 891 testPtr->locale, testPtr->type, testPtr->offsRev[offsindx]);
729e4ab9 892 }
374ca955 893
729e4ab9
A
894 ubrk_close(ubrkiter);
895 } else {
896 log_err_status(status, "FAIL: locale %s, break type %d, ubrk_open status: %s\n", testPtr->locale, testPtr->type, u_errorName(status));
897 }
898 }
899}
73c04bcf 900
4388f060
A
901
902static void TestBreakIteratorRefresh(void) {
903 /*
904 * RefreshInput changes out the input of a Break Iterator without
905 * changing anything else in the iterator's state. Used with Java JNI,
906 * when Java moves the underlying string storage. This test
907 * runs a ubrk_next() repeatedly, moving the text in the middle of the sequence.
908 * The right set of boundaries should still be found.
909 */
910 UChar testStr[] = {0x20, 0x41, 0x20, 0x42, 0x20, 0x43, 0x20, 0x44, 0x0}; /* = " A B C D" */
911 UChar movedStr[] = {0x20, 0x20, 0x20, 0x20, 0x20, 0x20, 0x20, 0x20, 0};
912 UErrorCode status = U_ZERO_ERROR;
913 UBreakIterator *bi;
914 UText ut1 = UTEXT_INITIALIZER;
915 UText ut2 = UTEXT_INITIALIZER;
2ca993e8 916
4388f060
A
917 bi = ubrk_open(UBRK_LINE, "en_US", NULL, 0, &status);
918 TEST_ASSERT_SUCCESS(status);
2ca993e8
A
919 if (U_FAILURE(status)) {
920 return;
921 }
4388f060
A
922
923 utext_openUChars(&ut1, testStr, -1, &status);
924 TEST_ASSERT_SUCCESS(status);
925 ubrk_setUText(bi, &ut1, &status);
926 TEST_ASSERT_SUCCESS(status);
927
928 if (U_SUCCESS(status)) {
929 /* Line boundaries will occur before each letter in the original string */
930 TEST_ASSERT(1 == ubrk_next(bi));
931 TEST_ASSERT(3 == ubrk_next(bi));
932
933 /* Move the string, kill the original string. */
934 u_strcpy(movedStr, testStr);
935 u_memset(testStr, 0x20, u_strlen(testStr));
936 utext_openUChars(&ut2, movedStr, -1, &status);
937 TEST_ASSERT_SUCCESS(status);
938 ubrk_refreshUText(bi, &ut2, &status);
939 TEST_ASSERT_SUCCESS(status);
2ca993e8 940
4388f060
A
941 /* Find the following matches, now working in the moved string. */
942 TEST_ASSERT(5 == ubrk_next(bi));
943 TEST_ASSERT(7 == ubrk_next(bi));
944 TEST_ASSERT(8 == ubrk_next(bi));
945 TEST_ASSERT(UBRK_DONE == ubrk_next(bi));
946 TEST_ASSERT_SUCCESS(status);
947
948 utext_close(&ut1);
949 utext_close(&ut2);
950 }
951 ubrk_close(bi);
952}
953
2ca993e8
A
954
955static void TestBug11665(void) {
956 // The problem was with the incorrect breaking of Japanese text beginning
957 // with Katakana characters when no prior Japanese or Chinese text had been
958 // encountered.
959 //
960 // Tested here in cintltst, rather than in intltest, because only cintltst
961 // tests have the ability to reset ICU, which is needed to get the bug
962 // to manifest itself.
963
964 static UChar japaneseText[] = {0x30A2, 0x30EC, 0x30EB, 0x30AE, 0x30FC, 0x6027, 0x7D50, 0x819C, 0x708E};
965 int32_t boundaries[10] = {0};
966 UBreakIterator *bi = NULL;
967 int32_t brk;
968 int32_t brkIdx = 0;
969 int32_t totalBreaks = 0;
970 UErrorCode status = U_ZERO_ERROR;
971
972 ctest_resetICU();
973 bi = ubrk_open(UBRK_WORD, "en_US", japaneseText, UPRV_LENGTHOF(japaneseText), &status);
974 TEST_ASSERT_SUCCESS(status);
975 if (!bi) {
976 return;
977 }
978 for (brk=ubrk_first(bi); brk != UBRK_DONE; brk=ubrk_next(bi)) {
979 boundaries[brkIdx] = brk;
980 if (++brkIdx >= UPRV_LENGTHOF(boundaries) - 1) {
981 break;
982 }
983 }
984 if (brkIdx <= 2 || brkIdx >= UPRV_LENGTHOF(boundaries)) {
985 log_err("%s:%d too few or many breaks found.\n", __FILE__, __LINE__);
986 } else {
987 totalBreaks = brkIdx;
988 brkIdx = 0;
989 for (brk=ubrk_first(bi); brk != UBRK_DONE; brk=ubrk_next(bi)) {
990 if (brk != boundaries[brkIdx]) {
991 log_err("%s:%d Break #%d differs between first and second iteration.\n", __FILE__, __LINE__, brkIdx);
992 break;
993 }
994 if (++brkIdx >= UPRV_LENGTHOF(boundaries) - 1) {
995 log_err("%s:%d Too many breaks.\n", __FILE__, __LINE__);
996 break;
997 }
998 }
999 if (totalBreaks != brkIdx) {
1000 log_err("%s:%d Number of breaks differ between first and second iteration.\n", __FILE__, __LINE__);
1001 }
1002 }
1003 ubrk_close(bi);
1004}
1005
1006/*
1007 * expOffset is the set of expected offsets, ending with '-1'.
1008 * "Expected expOffset -1" means "expected the end of the offsets"
1009 */
1010
b331163b 1011static const char testSentenceSuppressionsEn[] = "Mr. Jones comes home. Dr. Smith Ph.D. is out. In the U.S.A. it is hot.";
2ca993e8
A
1012static const int32_t testSentSuppFwdOffsetsEn[] = { 22, 46, 70, -1 }; /* With suppressions */
1013static const int32_t testSentFwdOffsetsEn[] = { 4, 22, 26, 46, 70, -1 }; /* Without suppressions */
1014static const int32_t testSentSuppRevOffsetsEn[] = { 46, 22, 0, -1 }; /* With suppressions */
1015static const int32_t testSentRevOffsetsEn[] = { 46, 26, 22, 4, 0, -1 }; /* Without suppressions */
b331163b
A
1016
1017static const char testSentenceSuppressionsDe[] = "Wenn ich schon h\\u00F6re zu Guttenberg kommt evtl. zur\\u00FCck.";
1018// "Wenn ich schon höre zu Guttenberg kommt evtl. zurück."
1019static const int32_t testSentSuppFwdOffsetsDe[] = { 53, -1 }; /* With suppressions */
1020static const int32_t testSentFwdOffsetsDe[] = { 53, -1 }; /* Without suppressions; no break in evtl. zur due to casing */
1021static const int32_t testSentSuppRevOffsetsDe[] = { 0, -1 }; /* With suppressions */
1022static const int32_t testSentRevOffsetsDe[] = { 0, -1 }; /* Without suppressions */
1023
1024static const char testSentenceSuppressionsEs[] = "Te esperamos todos los miercoles en Bravo 416, Col. El Pueblo a las 7 PM.";
1025static const int32_t testSentSuppFwdOffsetsEs[] = { 73, -1 }; /* With suppressions */
1026static const int32_t testSentFwdOffsetsEs[] = { 52, 73, -1 }; /* Without suppressions */
1027static const int32_t testSentSuppRevOffsetsEs[] = { 0, -1 }; /* With suppressions */
1028static const int32_t testSentRevOffsetsEs[] = { 52, 0, -1 }; /* Without suppressions */
1029
2ca993e8
A
1030static const char testSentenceSuppressionsE1[] = "Add or detract. The world will little note.";
1031static const char testSentenceSuppressionsE1u[] = "ADD OR DETRACT. THE WORLD WILL LITTLE NOTE.";
1032static const int32_t testSentFwdOffsetsE1[] = { 16, 43, -1 }; /* Suppressions and case should make no difference */
1033static const int32_t testSentRevOffsetsE1[] = { 16, 0, -1 }; /* Suppressions and case should make no difference */
1034
1035static const char testSentenceSuppressionsE2[] = "Coming up, the sprints at NCAA. Are you watching?";
1036static const char testSentenceSuppressionsE2u[] = "COMING UP, THE SPRINTS AT NCAA. ARE YOU WATCHING?";
1037static const int32_t testSentFwdOffsetsE2[] = { 32, 49, -1 }; /* Suppressions and case should make no difference */
1038static const int32_t testSentRevOffsetsE2[] = { 32, 0, -1 }; /* Suppressions and case should make no difference */
1039
1040static const char testSentenceSuppressionsFr[] = "Tr\\u00E8s bonne prise de parole de M. Junod, municipal \\u00E0 la culture de Lausanne.";
1041// "Très bonne prise de parole de M. Junod, municipal à la culture de Lausanne."
1042static const int32_t testSentFwdOffsetsFr[] = { 33, 75, -1 }; /* Without suppressions */
1043static const int32_t testSentSuppFwdOffsetsFr[] = { 75, -1 }; /* With suppressions */
1044static const int32_t testSentRevOffsetsFr[] = { 33, 0, -1 }; /* Without suppressions */
1045static const int32_t testSentSuppRevOffsetsFr[] = { 0, -1 }; /* With suppressions */
1046
1047static const char testSentenceSuppressionsE3[] = "G8 countries e.g. U.K., Japan. Sanctions i.e. restrictions. Test E. Xx G. Xx I. Xx.";
1048static const char testSentenceSuppressionsE3u[] = "G8 COUNTRIES E.G. U.K., JAPAN. SANCTIONS I.E. RESTRICTIONS. TEST E. XX G. XX I. XX.";
1049static const int32_t testSentSuppFwdOffsetsE3[] = { 31, 60, 83, -1 }; /* With suppressions */
1050static const int32_t testSentSuppRevOffsetsE3[] = { 60, 31, 0, -1 }; /* With suppressions */
1051static const int32_t testSentFwdOffsetsE3[] = { 18, 31, 60, 68, 74, 80, 83, -1 }; /* Without suppressions */
1052static const int32_t testSentRevOffsetsE3[] = { 80, 74, 68, 60, 31, 18, 0, -1 }; /* Without suppressions */
1053static const int32_t testSentFwdOffsetsE3u[] = { 18, 31, 46, 60, 68, 74, 80, 83, -1 }; /* Without suppressions */
1054static const int32_t testSentRevOffsetsE3u[] = { 80, 74, 68, 60, 46, 31, 18, 0, -1 }; /* Without suppressions */
1055
1056enum { kTextULenMax = 128, kTextBLenMax = 192 };
b331163b
A
1057
1058typedef struct {
1059 const char * locale;
1060 const char * text;
1061 const int32_t * expFwdOffsets;
1062 const int32_t * expRevOffsets;
1063} TestBISuppressionsItem;
1064
1065static const TestBISuppressionsItem testBISuppressionsItems[] = {
1066 { "en@ss=standard", testSentenceSuppressionsEn, testSentSuppFwdOffsetsEn, testSentSuppRevOffsetsEn },
1067 { "en", testSentenceSuppressionsEn, testSentFwdOffsetsEn, testSentRevOffsetsEn },
f3c0d7a5
A
1068 { "en_CA", testSentenceSuppressionsEn, testSentFwdOffsetsEn, testSentRevOffsetsEn },
1069 { "en_CA@ss=standard", testSentenceSuppressionsEn, testSentSuppFwdOffsetsEn, testSentSuppRevOffsetsEn },
b331163b 1070 { "fr@ss=standard", testSentenceSuppressionsEn, testSentFwdOffsetsEn, testSentRevOffsetsEn },
f3c0d7a5
A
1071 { "af@ss=standard", testSentenceSuppressionsEn, testSentFwdOffsetsEn, testSentRevOffsetsEn }, /* no brkiter data => nosuppressions? */
1072 { "af_ZA@ss=standard", testSentenceSuppressionsEn, testSentFwdOffsetsEn, testSentRevOffsetsEn }, /* no brkiter data => nosuppressions? */
b331163b
A
1073 { "zh@ss=standard", testSentenceSuppressionsEn, testSentFwdOffsetsEn, testSentRevOffsetsEn }, /* brkiter data, no suppressions data => no suppressions */
1074 { "zh_Hant@ss=standard", testSentenceSuppressionsEn, testSentFwdOffsetsEn, testSentRevOffsetsEn }, /* brkiter data, no suppressions data => no suppressions */
1075 { "fi@ss=standard", testSentenceSuppressionsEn, testSentFwdOffsetsEn, testSentRevOffsetsEn }, /* brkiter data, no suppressions data => no suppressions */
1076 { "ja@ss=standard", testSentenceSuppressionsEn, testSentFwdOffsetsEn, testSentRevOffsetsEn }, /* brkiter data, no suppressions data => no suppressions */
1077 { "de@ss=standard", testSentenceSuppressionsDe, testSentSuppFwdOffsetsDe, testSentSuppRevOffsetsDe },
1078 { "de", testSentenceSuppressionsDe, testSentFwdOffsetsDe, testSentRevOffsetsDe },
1079 { "es@ss=standard", testSentenceSuppressionsEs, testSentSuppFwdOffsetsEs, testSentSuppRevOffsetsEs },
1080 { "es", testSentenceSuppressionsEs, testSentFwdOffsetsEs, testSentRevOffsetsEs },
2ca993e8
A
1081 { "en", testSentenceSuppressionsE1, testSentFwdOffsetsE1, testSentRevOffsetsE1 },
1082 { "en@ss=standard", testSentenceSuppressionsE1, testSentFwdOffsetsE1, testSentRevOffsetsE1 },
1083 { "en", testSentenceSuppressionsE1u, testSentFwdOffsetsE1, testSentRevOffsetsE1 },
1084 { "en@ss=standard", testSentenceSuppressionsE1u, testSentFwdOffsetsE1, testSentRevOffsetsE1 },
1085 { "en", testSentenceSuppressionsE2, testSentFwdOffsetsE2, testSentRevOffsetsE2 },
1086 { "en@ss=standard", testSentenceSuppressionsE2, testSentFwdOffsetsE2, testSentRevOffsetsE2 },
1087 { "en", testSentenceSuppressionsE2u, testSentFwdOffsetsE2, testSentRevOffsetsE2 },
1088 { "en@ss=standard", testSentenceSuppressionsE2u, testSentFwdOffsetsE2, testSentRevOffsetsE2 },
1089 { "fr", testSentenceSuppressionsFr, testSentFwdOffsetsFr, testSentRevOffsetsFr },
1090 { "fr@ss=standard", testSentenceSuppressionsFr, testSentSuppFwdOffsetsFr, testSentSuppRevOffsetsFr },
1091 { "en@ss=standard", testSentenceSuppressionsE3, testSentSuppFwdOffsetsE3, testSentSuppRevOffsetsE3 },
1092 { "en", testSentenceSuppressionsE3, testSentFwdOffsetsE3, testSentRevOffsetsE3 },
1093 { "en@ss=standard", testSentenceSuppressionsE3u, testSentSuppFwdOffsetsE3, testSentSuppRevOffsetsE3 },
1094 { "en", testSentenceSuppressionsE3u, testSentFwdOffsetsE3u, testSentRevOffsetsE3u },
b331163b
A
1095 { NULL, NULL, NULL }
1096};
1097
1098static void TestBreakIteratorSuppressions(void) {
1099 const TestBISuppressionsItem * itemPtr;
2ca993e8 1100
b331163b 1101 for (itemPtr = testBISuppressionsItems; itemPtr->locale != NULL; itemPtr++) {
2ca993e8
A
1102 UChar textU[kTextULenMax + 1];
1103 char textB[kTextBLenMax];
b331163b 1104 int32_t textULen = u_unescape(itemPtr->text, textU, kTextULenMax);
2ca993e8 1105 textU[kTextULenMax] = 0; // ensure zero termination
b331163b
A
1106 UErrorCode status = U_ZERO_ERROR;
1107 UBreakIterator *bi = ubrk_open(UBRK_SENTENCE, itemPtr->locale, textU, textULen, &status);
2ca993e8 1108 log_verbose("#%d: %s\n", (itemPtr-testBISuppressionsItems), itemPtr->locale);
b331163b
A
1109 if (U_SUCCESS(status)) {
1110 int32_t offset, start;
1111 const int32_t * expOffsetPtr;
2ca993e8
A
1112 const int32_t * expOffsetStart;
1113 u_austrcpy(textB, textU);
b331163b 1114
2ca993e8 1115 expOffsetStart = expOffsetPtr = itemPtr->expFwdOffsets;
b331163b
A
1116 ubrk_first(bi);
1117 for (; (offset = ubrk_next(bi)) != UBRK_DONE && *expOffsetPtr >= 0; expOffsetPtr++) {
1118 if (offset != *expOffsetPtr) {
2ca993e8
A
1119 log_err("FAIL: ubrk_next loc \"%s\", expected %d, got %d, text \"%s\"\n",
1120 itemPtr->locale, *expOffsetPtr, offset, textB);
b331163b
A
1121 }
1122 }
1123 if (offset != UBRK_DONE || *expOffsetPtr >= 0) {
2ca993e8
A
1124 log_err("FAIL: ubrk_next loc \"%s\", expected UBRK_DONE & expOffset -1, got %d and %d, text \"%s\"\n",
1125 itemPtr->locale, offset, *expOffsetPtr, textB);
b331163b
A
1126 }
1127
2ca993e8 1128 expOffsetStart = expOffsetPtr = itemPtr->expFwdOffsets;
b331163b
A
1129 start = ubrk_first(bi) + 1;
1130 for (; (offset = ubrk_following(bi, start)) != UBRK_DONE && *expOffsetPtr >= 0; expOffsetPtr++) {
1131 if (offset != *expOffsetPtr) {
2ca993e8
A
1132 log_err("FAIL: ubrk_following(%d) loc \"%s\", expected %d, got %d, text \"%s\"\n",
1133 start, itemPtr->locale, *expOffsetPtr, offset, textB);
b331163b
A
1134 }
1135 start = *expOffsetPtr + 1;
1136 }
1137 if (offset != UBRK_DONE || *expOffsetPtr >= 0) {
2ca993e8
A
1138 log_err("FAIL: ubrk_following(%d) loc \"%s\", expected UBRK_DONE & expOffset -1, got %d and %d, text \"%s\"\n",
1139 start, itemPtr->locale, offset, *expOffsetPtr, textB);
b331163b
A
1140 }
1141
2ca993e8
A
1142 expOffsetStart = expOffsetPtr = itemPtr->expRevOffsets;
1143 offset = ubrk_last(bi);
1144 log_verbose("___ @%d ubrk_last\n", offset);
1145 if(offset == 0) {
1146 log_err("FAIL: ubrk_last loc \"%s\" unexpected %d\n", itemPtr->locale, offset);
1147 }
b331163b
A
1148 for (; (offset = ubrk_previous(bi)) != UBRK_DONE && *expOffsetPtr >= 0; expOffsetPtr++) {
1149 if (offset != *expOffsetPtr) {
2ca993e8
A
1150 log_err("FAIL: ubrk_previous loc \"%s\", expected %d, got %d, text \"%s\"\n",
1151 itemPtr->locale, *expOffsetPtr, offset, textB);
1152 } else {
1153 log_verbose("[%d] @%d ubrk_previous()\n", (expOffsetPtr - expOffsetStart), offset);
b331163b
A
1154 }
1155 }
1156 if (offset != UBRK_DONE || *expOffsetPtr >= 0) {
2ca993e8
A
1157 log_err("FAIL: ubrk_previous loc \"%s\", expected UBRK_DONE & expOffset[%d] -1, got %d and %d, text \"%s\"\n",
1158 itemPtr->locale, expOffsetPtr - expOffsetStart, offset, *expOffsetPtr, textB);
b331163b
A
1159 }
1160
2ca993e8 1161 expOffsetStart = expOffsetPtr = itemPtr->expRevOffsets;
b331163b
A
1162 start = ubrk_last(bi) - 1;
1163 for (; (offset = ubrk_preceding(bi, start)) != UBRK_DONE && *expOffsetPtr >= 0; expOffsetPtr++) {
1164 if (offset != *expOffsetPtr) {
2ca993e8
A
1165 log_err("FAIL: ubrk_preceding(%d) loc \"%s\", expected %d, got %d, text \"%s\"\n",
1166 start, itemPtr->locale, *expOffsetPtr, offset, textB);
b331163b
A
1167 }
1168 start = *expOffsetPtr - 1;
1169 }
1170 if (start >=0 && (offset != UBRK_DONE || *expOffsetPtr >= 0)) {
2ca993e8
A
1171 log_err("FAIL: ubrk_preceding loc(%d) \"%s\", expected UBRK_DONE & expOffset -1, got %d and %d, text \"%s\"\n",
1172 start, itemPtr->locale, offset, *expOffsetPtr, textB);
b331163b
A
1173 }
1174
1175 ubrk_close(bi);
1176 } else {
2ca993e8
A
1177 log_data_err("FAIL: ubrk_open(UBRK_SENTENCE, \"%s\", ...) status %s (Are you missing data?)\n",
1178 itemPtr->locale, u_errorName(status));
b331163b
A
1179 }
1180 }
1181}
1182
f3c0d7a5
A
1183#if APPLE_ADDITIONS
1184#include <stdio.h>
9f1b1155
A
1185#include <unistd.h>
1186#include <mach/mach_time.h>
f3c0d7a5
A
1187#include "unicode/urbtok.h"
1188#include "cstring.h"
1189
1190static const char testRulesFilePath[] = "../testdata/word_urbTokTest.txt";
1191static const UChar textToTokenize[] = {
1192/*
1193"Short phrase! Another (with parens); done.\n
1194At 4:00, tea-time.\n
1195He wouldn't've wanted y'all to ... come at 3:30pm for $3 coffee @funman :)\n
1196x3:30 -- x1.0"
1197*/
1198 0x53,0x68,0x6F,0x72,0x74,0x20,0x70,0x68,0x72,0x61,0x73,0x65,0x21,0x20,
1199 0x41,0x6E,0x6F,0x74,0x68,0x65,0x72,0x20,0x28,0x77,0x69,0x74,0x68,0x20,0x70,0x61,0x72,0x65,0x6E,0x73,0x29,0x3B,0x20,0x64,0x6F,0x6E,0x65,0x2E,0x0A,
1200 0x41,0x74,0x20,0x34,0x3A,0x30,0x30,0x2C,0x20,0x74,0x65,0x61,0x2D,0x74,0x69,0x6D,0x65,0x2E,0x0A,
1201 0x48,0x65,0x20,0x77,0x6F,0x75,0x6C,0x64,0x6E,0x27,0x74,0x27,0x76,0x65,0x20,0x77,0x61,0x6E,0x74,0x65,0x64,0x20,
1202 0x79,0x27,0x61,0x6C,0x6C,0x20,0x74,0x6F,0x20,0x2E,0x2E,0x2E,0x20, 0x63,0x6F,0x6D,0x65,0x20,0x61,0x74,0x20,
1203 0x33,0x3A,0x33,0x30,0x70,0x6D,0x20,0x66,0x6F,0x72,0x20,0x24,0x33,0x20,0x63,0x6F,0x66,0x66,0x65,0x65,0x20,
1204 0x40,0x66,0x75,0x6E,0x6D,0x61,0x6E,0x20,0x3A,0x29,0x0A,
1205 0x78,0x33,0x3A,0x33,0x30,0x20,0x2D,0x2D,0x20,0x78,0x31,0x2E,0x30,0
1206};
1207typedef struct {
1208 RuleBasedTokenRange token;
1209 unsigned long flags;
1210} RBTokResult;
1211static const RBTokResult expectedResults[] = { // 66 tokens
1212 { { 0, 5 }, 0xC8 }, // Short
1213 { { 5, 1 }, 0x01 }, // _sp_
1214 { { 6, 6 }, 0xC8 }, // phrase
1215 { { 12, 1 }, 0x00 }, // !
1216 { { 13, 1 }, 0x01 }, // _sp_
1217 { { 14, 7 }, 0xC8 }, // Another
1218 { { 21, 1 }, 0x01 }, // _sp_
1219 { { 22, 1 }, 0x00 }, // (
1220 { { 23, 4 }, 0xC8 }, // with
1221 { { 27, 1 }, 0x01 }, // _sp_
1222 { { 28, 6 }, 0xC8 }, // parens
1223 { { 34, 1 }, 0x00 }, // )
1224 { { 35, 1 }, 0x00 }, // ;
1225 { { 36, 1 }, 0x01 }, // _sp_
1226 { { 37, 4 }, 0xC8 }, // done
1227 { { 41, 1 }, 0x14 }, // .
1228 { { 42, 1 }, 0x00 }, // _nl_
1229
1230 { { 43, 2 }, 0xC8 }, // At
1231 { { 45, 1 }, 0x01 }, // _sp_
1232 { { 46, 4 }, 0x76 }, // 4:00 ** here RBBI has x64
1233 { { 50, 1 }, 0x00 }, // ,
1234 { { 51, 1 }, 0x01 }, // _sp_
1235 { { 52, 3 }, 0xC8 }, // tea
1236 { { 55, 1 }, 0x15 }, // -
1237 { { 56, 4 }, 0xC8 }, // time
1238 { { 60, 1 }, 0x14 }, // .
1239 { { 61, 1 }, 0x00 }, // _nl_
1240
1241 { { 62, 2 }, 0xC8 }, // He
1242 { { 64, 1 }, 0x01 }, // _sp_
1243 { { 65, 8 }, 0xCA }, // wouldn't
1244 { { 73, 1 }, 0x16 }, // '
1245 { { 74, 2 }, 0xC8 }, // ve
1246 { { 76, 1 }, 0x01 }, // _sp_
1247 { { 77, 6 }, 0xC8 }, // wanted
1248 { { 83, 1 }, 0x01 }, // _sp_
1249 { { 84, 5 }, 0xCA }, // y'all
1250 { { 89, 1 }, 0x01 }, // _sp_
1251 { { 90, 2 }, 0xC8 }, // to
1252 { { 92, 1 }, 0x01 }, // _sp_
1253 { { 93, 3 }, 0x3C }, // ... ** here RBBI has 0x28
1254 { { 96, 1 }, 0x01 }, // _sp_
1255 { { 97, 4 }, 0xC8 }, // come
1256 { { 101, 1 }, 0x01 }, // _sp_
1257 { { 102, 2 }, 0xC8 }, // at
1258 { { 104, 1 }, 0x01 }, // _sp_
1259 { { 105, 6 }, 0xC8 }, // 3:30pm
1260 { { 111, 1 }, 0x01 }, // _sp_
1261 { { 112, 3 }, 0xC8 }, // for
1262 { { 115, 1 }, 0x01 }, // _sp_
1263 { { 116, 1 }, 0x00 }, // $
1264 { { 117, 1 }, 0x64 }, // 3
1265 { { 118, 1 }, 0x01 }, // _sp_
1266 { { 119, 6 }, 0xC8 }, // coffee
1267 { { 125, 1 }, 0x01 }, // _sp_
1268 { { 126, 7 }, 0xDF }, // @funman ** here RBBI has 0xC8
1269 { { 133, 1 }, 0x01 }, // _sp_
1270 { { 134, 2 }, 0x20 }, // :)
1271 { { 136, 1 }, 0x00 }, // _nl_
1272
1273 // ** incorrect ranges (and flags) currently produced by RBTok
1274 { { 137, 2 }, 0xEC }, // x3
1275 { { 139, 1 }, 0x00 }, // :
1276 { { 140, 2 }, 0x64 }, // 30
1277 // ** for the above, RBBI has
1278 //{ { 137, 1 }, 0x64 }, // x
1279 //{ { 138, 4 }, 0x64 }, // 3:30
1280 //
1281 { { 142, 1 }, 0x01 }, // _sp_
1282 { { 143, 2 }, 0x3D }, // -- ** here RBBI has 0x28
1283 { { 145, 1 }, 0x01 }, // _sp_
1284 { { 146, 2 }, 0xEC }, // x1 ** here RBBI has 0xC8
1285 { { 148, 1 }, 0x14 }, // .
1286 { { 149, 1 }, 0x64 }, // 0
1287};
1288enum {
1289 kNumTokensExpected = UPRV_LENGTHOF(expectedResults), // 66
1290 kMaxTokens = 96
1291};
1292
1293static void TestRuleBasedTokenizer(void) {
1294 FILE * testRulesFile;
1295 char * testRulesUTF8Buf;
1296 UChar* testRulesUTF16Buf = NULL;
1297 long testRulesFileSize, testRulesFileRead = 0;
1298 long testRulesUTF8Offset = 0;
1299 int32_t testRulesUTF16Size;
1300 UErrorCode status = U_ZERO_ERROR;
1301
1302 testRulesFile = fopen(testRulesFilePath, "r");
1303 if (testRulesFile == NULL) {
1304 log_data_err("FAIL: fopen fails for: %s\n", testRulesFilePath);
1305 return;
1306 }
1307 fseek(testRulesFile, 0, SEEK_END);
1308 testRulesFileSize = ftell(testRulesFile);
1309 rewind(testRulesFile);
1310
1311 testRulesUTF8Buf = (char *)uprv_malloc(testRulesFileSize);
1312 if (testRulesUTF8Buf != NULL) {
1313 testRulesFileRead = fread(testRulesUTF8Buf, 1, testRulesFileSize, testRulesFile);
1314 }
1315 fclose(testRulesFile);
1316 if (testRulesUTF8Buf == NULL) {
1317 log_data_err("FAIL: uprv_malloc fails for testRulesUTF8Buf[%ld]\n", testRulesFileSize);
1318 return;
1319 }
1320 if (testRulesFileRead < testRulesFileSize) {
1321 log_data_err("FAIL: fread fails for %s, read %ld of %ld\n", testRulesFile, testRulesFileRead, testRulesFileSize);
1322 uprv_free(testRulesUTF8Buf);
1323 return;
1324 }
1325 /* done with file, UTF8 rules in testRulesUTF8Buf. Handle UTF8 BOM: */
1326 if (uprv_strncmp(testRulesUTF8Buf, "\xEF\xBB\xBF", 3) == 0) {
1327 testRulesUTF8Offset = 3;
1328 testRulesFileSize -= testRulesUTF8Offset;
1329 }
1330
1331 u_strFromUTF8(NULL, 0, &testRulesUTF16Size, testRulesUTF8Buf+testRulesUTF8Offset, testRulesFileSize, &status); /* preflight */
1332 if (status == U_BUFFER_OVERFLOW_ERROR) { /* expected for preflight */
1333 status = U_ZERO_ERROR;
1334 }
1335 if (U_FAILURE(status)) {
1336 log_data_err("FAIL: u_strFromUTF8 preflight fails: %s\n", u_errorName(status));
1337 } else {
1338 testRulesUTF16Buf = (UChar *)uprv_malloc(testRulesUTF16Size*sizeof(UChar));
1339 if (testRulesUTF16Buf == NULL) {
1340 log_data_err("FAIL: uprv_malloc fails for testRulesUTF16Buf[%ld]\n", testRulesUTF16Size*sizeof(UChar));
1341 } else {
1342 u_strFromUTF8(testRulesUTF16Buf, testRulesUTF16Size, &testRulesUTF16Size, testRulesUTF8Buf+testRulesUTF8Offset, testRulesFileSize, &status);
1343 }
1344 }
1345 uprv_free(testRulesUTF8Buf);
1346 if (testRulesUTF16Buf == NULL) {
1347 return;
1348 }
1349 if (U_FAILURE(status)) {
1350 log_data_err("FAIL: u_strFromUTF8 fails: %s\n", u_errorName(status));
1351 } else {
9f1b1155
A
1352 mach_timebase_info_data_t info;
1353 uint64_t start, duration;
f3c0d7a5 1354 UParseError parseErr;
9f1b1155
A
1355 UBreakIterator *brkFromRules;
1356
1357 mach_timebase_info(&info);
1358
1359 start = mach_absolute_time();
1360 brkFromRules = urbtok_openRules(testRulesUTF16Buf, testRulesUTF16Size, &parseErr, &status);
1361 duration = ((mach_absolute_time() - start) * info.numer)/info.denom;
f3c0d7a5 1362 if (U_FAILURE(status)) {
9f1b1155 1363 log_err("FAIL: urbtok_openRules status: %s, line %d, col %d\n", u_errorName(status), parseErr.line, parseErr.offset);
f3c0d7a5
A
1364 } else {
1365 uint8_t *rulesBinaryBuf;
1366 uint32_t rulesBinarySize;
9f1b1155 1367 log_info("urbtok_openRules nsec %llu\n", duration);
f3c0d7a5
A
1368 rulesBinarySize = urbtok_getBinaryRules(brkFromRules, NULL, 0, &status);
1369 if (U_FAILURE(status)) {
1370 log_err("FAIL: urbtok_getBinaryRules preflight status: %s, rulesBinarySize %u\n", u_errorName(status), rulesBinarySize);
1371 } else {
1372 rulesBinaryBuf = (uint8_t *)uprv_malloc(rulesBinarySize);
1373 if (rulesBinaryBuf == NULL) {
1374 log_data_err("FAIL: uprv_malloc fails for rulesBinaryBuf[%ld]\n", rulesBinarySize);
1375 } else {
9f1b1155 1376 start = mach_absolute_time();
f3c0d7a5 1377 rulesBinarySize = urbtok_getBinaryRules(brkFromRules, rulesBinaryBuf, rulesBinarySize, &status);
9f1b1155 1378 duration = ((mach_absolute_time() - start) * info.numer)/info.denom;
f3c0d7a5
A
1379 if (U_FAILURE(status)) {
1380 log_err("FAIL: urbtok_getBinaryRules status: %s, rulesBinarySize %u\n", u_errorName(status), rulesBinarySize);
1381 } else {
9f1b1155
A
1382 UBreakIterator *brkFromBinary;
1383 log_info("urbtok_getBinaryRules nsec %llu\n", duration);
1384 start = mach_absolute_time();
1385 brkFromBinary = urbtok_openBinaryRules(rulesBinaryBuf, &status);
1386 duration = ((mach_absolute_time() - start) * info.numer)/info.denom;
f3c0d7a5
A
1387 if (U_FAILURE(status)) {
1388 log_err("FAIL: urbtok_openBinaryRules status: %s\n", u_errorName(status));
1389 } else {
1390 RuleBasedTokenRange tokens[kMaxTokens];
1391 unsigned long flags[kMaxTokens];
1392 int32_t iToken, numTokens = 0;
1393
9f1b1155
A
1394 log_info("urbtok_openBinaryRules nsec %llu\n", duration);
1395
f3c0d7a5
A
1396 status = U_ZERO_ERROR;
1397 ubrk_setText(brkFromRules, textToTokenize, -1, &status);
1398 if (U_FAILURE(status)) {
1399 log_err("FAIL: ubrk_setText brkFromRules status: %s\n", u_errorName(status));
1400 } else {
9f1b1155 1401 start = mach_absolute_time();
f3c0d7a5 1402 numTokens = urbtok_tokenize(brkFromRules, kMaxTokens, tokens, flags);
9f1b1155 1403 duration = ((mach_absolute_time() - start) * info.numer)/info.denom;
f3c0d7a5
A
1404 UBool fail = (numTokens != kNumTokensExpected);
1405 for (iToken = 0; !fail && iToken < numTokens; iToken++) {
1406 if ( tokens[iToken].location != expectedResults[iToken].token.location ||
1407 tokens[iToken].length != expectedResults[iToken].token.length ||
1408 flags[iToken] != expectedResults[iToken].flags ) {
1409 fail = TRUE;
1410 }
1411 }
1412 if (fail) {
9f1b1155 1413 log_err("FAIL: urbtok_tokenize bulk brkFromRules expected %d tokens, got %d\n", kNumTokensExpected, numTokens);
f3c0d7a5
A
1414 printf("# expect get\n");
1415 printf("# loc len flags loc len flags\n");
1416 int32_t maxTokens = (numTokens >= kNumTokensExpected)? numTokens: kNumTokensExpected;
1417 for (iToken = 0; iToken < maxTokens; iToken++) {
1418 if (iToken < kNumTokensExpected) {
1419 printf(" %3ld %3ld 0x%03lX", expectedResults[iToken].token.location,
1420 expectedResults[iToken].token.length, expectedResults[iToken].flags);
1421 } else {
1422 printf(" ");
1423 }
1424 if (iToken < numTokens) {
1425 printf(" %3ld %3ld 0x%03lX\n", tokens[iToken].location, tokens[iToken].length, flags[iToken] );
1426 } else {
1427 printf("\n");
1428 }
1429 }
9f1b1155
A
1430 } else {
1431 log_info("Latn urbtok_tokenize bulk brkFromRules nsec %llu\n", duration);
f3c0d7a5
A
1432 }
1433 }
1434
1435 status = U_ZERO_ERROR;
1436 ubrk_setText(brkFromBinary, textToTokenize, -1, &status);
1437 if (U_FAILURE(status)) {
1438 log_err("FAIL: ubrk_setText brkFromBinary status: %s\n", u_errorName(status));
1439 } else {
9f1b1155 1440 start = mach_absolute_time();
f3c0d7a5 1441 numTokens = urbtok_tokenize(brkFromBinary, kMaxTokens, tokens, flags);
9f1b1155 1442 duration = ((mach_absolute_time() - start) * info.numer)/info.denom;
f3c0d7a5
A
1443 UBool fail = (numTokens != kNumTokensExpected);
1444 for (iToken = 0; !fail && iToken < numTokens; iToken++) {
1445 if ( tokens[iToken].location != expectedResults[iToken].token.location ||
1446 tokens[iToken].length != expectedResults[iToken].token.length ||
1447 flags[iToken] != expectedResults[iToken].flags ) {
1448 fail = TRUE;
1449 }
1450 }
1451 if (fail) {
9f1b1155 1452 log_err("FAIL: urbtok_tokenize bulk brkFromBinary expected %d tokens, got %d\n", kNumTokensExpected, numTokens);
f3c0d7a5
A
1453 printf("# expect get\n");
1454 printf("# loc len flags loc len flags\n");
1455 int32_t maxTokens = (numTokens >= kNumTokensExpected)? numTokens: kNumTokensExpected;
1456 for (iToken = 0; iToken < maxTokens; iToken++) {
1457 if (iToken < kNumTokensExpected) {
1458 printf(" %3ld %3ld 0x%03lX", expectedResults[iToken].token.location,
1459 expectedResults[iToken].token.length, expectedResults[iToken].flags);
1460 } else {
1461 printf(" ");
1462 }
1463 if (iToken < numTokens) {
1464 printf(" %3ld %3ld 0x%03lX\n", tokens[iToken].location, tokens[iToken].length, flags[iToken] );
1465 } else {
1466 printf("\n");
1467 }
1468 }
9f1b1155
A
1469 } else {
1470 log_info("Latn urbtok_tokenize bulk brkFromBinary nsec %llu\n", duration);
1471 }
1472 }
1473
1474 status = U_ZERO_ERROR;
1475 ubrk_setText(brkFromBinary, textToTokenize, -1, &status);
1476 if (U_FAILURE(status)) {
1477 log_err("FAIL: ubrk_setText brkFromBinary status: %s\n", u_errorName(status));
1478 } else {
1479 RuleBasedTokenRange *tokenLimit = tokens + kMaxTokens;
1480 RuleBasedTokenRange *tokenP;
1481 unsigned long *flagsP;
1482
1483 start = mach_absolute_time();
1484 for (tokenP = tokens, flagsP = flags; tokenP < tokenLimit && urbtok_tokenize(brkFromBinary, 1, tokenP, flagsP) == 1; tokenP++, flagsP++) {
1485 ;
1486 }
1487 numTokens = tokenP - tokens;
1488 duration = ((mach_absolute_time() - start) * info.numer)/info.denom;
1489 UBool fail = (numTokens != kNumTokensExpected);
1490 for (iToken = 0; !fail && iToken < numTokens; iToken++) {
1491 if ( tokens[iToken].location != expectedResults[iToken].token.location ||
1492 tokens[iToken].length != expectedResults[iToken].token.length ||
1493 flags[iToken] != expectedResults[iToken].flags ) {
1494 fail = TRUE;
1495 }
f3c0d7a5 1496 }
9f1b1155
A
1497 if (fail) {
1498 log_err("FAIL: urbtok_tokenize loop brkFromBinary expected %d tokens, got %d\n", kNumTokensExpected, numTokens);
1499 printf("# expect get\n");
1500 printf("# loc len flags loc len flags\n");
1501 int32_t maxTokens = (numTokens >= kNumTokensExpected)? numTokens: kNumTokensExpected;
1502 for (iToken = 0; iToken < maxTokens; iToken++) {
1503 if (iToken < kNumTokensExpected) {
1504 printf(" %3ld %3ld 0x%03lX", expectedResults[iToken].token.location,
1505 expectedResults[iToken].token.length, expectedResults[iToken].flags);
1506 } else {
1507 printf(" ");
1508 }
1509 if (iToken < numTokens) {
1510 printf(" %3ld %3ld 0x%03lX\n", tokens[iToken].location, tokens[iToken].length, flags[iToken] );
1511 } else {
1512 printf("\n");
1513 }
1514 }
1515 } else {
1516 log_info("Latn urbtok_tokenize loop brkFromBinary nsec %llu\n", duration);
1517 }
f3c0d7a5 1518 }
9f1b1155 1519
f3c0d7a5
A
1520 ubrk_close(brkFromBinary);
1521 }
1522 }
1523 uprv_free(rulesBinaryBuf);
1524 }
1525 }
1526 ubrk_close(brkFromRules);
1527 }
1528 }
1529 uprv_free(testRulesUTF16Buf);
1530}
1531#endif
1532
2ca993e8 1533
b75a7d8f 1534#endif /* #if !UCONFIG_NO_BREAK_ITERATION */