OLD | NEW |
| (Empty) |
1 /******************************************************************** | |
2 * COPYRIGHT: | |
3 * Copyright (c) 1997-2014, International Business Machines Corporation and | |
4 * others. All Rights Reserved. | |
5 ********************************************************************/ | |
6 /******************************************************************************* | |
7 * | |
8 * File CALLCOLL.C | |
9 * | |
10 * Modification History: | |
11 * Name Description | |
12 * Madhu Katragadda Ported for C API | |
13 ******************************************************************************** | |
14 */ | |
15 | |
16 /* | |
17 * Important: This file is included into intltest/allcoll.cpp so that the | |
18 * test data is shared. This makes it easier to maintain the test data, | |
19 * especially since the Unicode data must be portable and quoted character | |
20 * literals will not work. | |
21 * If it is included, then there will be a #define INCLUDE_CALLCOLL_C | |
22 * that must prevent the actual code in here from being part of the | |
23 * allcoll.cpp compilation. | |
24 */ | |
25 | |
26 /** | |
27 * CollationDummyTest is a third level test class. This tests creation of | |
28 * a customized collator object. For example, number 1 to be sorted | |
29 * equlivalent to word 'one'. | |
30 */ | |
31 | |
32 #include <string.h> | |
33 #include <stdlib.h> | |
34 | |
35 #include "unicode/utypes.h" | |
36 | |
37 #if !UCONFIG_NO_COLLATION | |
38 | |
39 #include "unicode/ucol.h" | |
40 #include "unicode/uloc.h" | |
41 #include "unicode/ures.h" | |
42 #include "unicode/udata.h" | |
43 #include "unicode/ucoleitr.h" | |
44 #include "unicode/ustring.h" | |
45 #include "unicode/uclean.h" | |
46 #include "unicode/putil.h" | |
47 #include "unicode/uenum.h" | |
48 | |
49 #include "cintltst.h" | |
50 #include "ccolltst.h" | |
51 #include "callcoll.h" | |
52 #include "calldata.h" | |
53 #include "cstring.h" | |
54 #include "cmemory.h" | |
55 | |
56 /* set to 1 to test offsets in backAndForth() */ | |
57 #define TEST_OFFSETS 0 | |
58 | |
59 /* perform test with strength PRIMARY */ | |
60 static void TestPrimary(void); | |
61 | |
62 /* perform test with strength SECONDARY */ | |
63 static void TestSecondary(void); | |
64 | |
65 /* perform test with strength tertiary */ | |
66 static void TestTertiary(void); | |
67 | |
68 /*perform tests with strength Identical */ | |
69 static void TestIdentical(void); | |
70 | |
71 /* perform extra tests */ | |
72 static void TestExtra(void); | |
73 | |
74 /* Test jitterbug 581 */ | |
75 static void TestJB581(void); | |
76 | |
77 /* Test jitterbug 1401 */ | |
78 static void TestJB1401(void); | |
79 | |
80 /* Test [variable top] in the rule syntax */ | |
81 static void TestVariableTop(void); | |
82 | |
83 /* Test surrogates */ | |
84 static void TestSurrogates(void); | |
85 | |
86 static void TestInvalidRules(void); | |
87 | |
88 static void TestJitterbug1098(void); | |
89 | |
90 static void TestFCDCrash(void); | |
91 | |
92 static void TestJ5298(void); | |
93 | |
94 const UCollationResult results[] = { | |
95 UCOL_LESS, | |
96 UCOL_LESS, /*UCOL_GREATER,*/ | |
97 UCOL_LESS, | |
98 UCOL_LESS, | |
99 UCOL_LESS, | |
100 UCOL_LESS, | |
101 UCOL_LESS, | |
102 UCOL_GREATER, | |
103 UCOL_GREATER, | |
104 UCOL_LESS, /* 10 */ | |
105 UCOL_GREATER, | |
106 UCOL_LESS, | |
107 UCOL_GREATER, | |
108 UCOL_GREATER, | |
109 UCOL_LESS, | |
110 UCOL_LESS, | |
111 UCOL_LESS, | |
112 /* test primary > 17 */ | |
113 UCOL_EQUAL, | |
114 UCOL_EQUAL, | |
115 UCOL_EQUAL, /* 20 */ | |
116 UCOL_LESS, | |
117 UCOL_LESS, | |
118 UCOL_EQUAL, | |
119 UCOL_EQUAL, | |
120 UCOL_EQUAL, | |
121 UCOL_LESS, | |
122 /* test secondary > 26 */ | |
123 UCOL_EQUAL, | |
124 UCOL_EQUAL, | |
125 UCOL_EQUAL, | |
126 UCOL_EQUAL, | |
127 UCOL_EQUAL, /* 30 */ | |
128 UCOL_EQUAL, | |
129 UCOL_LESS, | |
130 UCOL_EQUAL, /* 34 */ | |
131 UCOL_EQUAL, | |
132 UCOL_EQUAL, | |
133 UCOL_LESS /* 37 */ | |
134 }; | |
135 | |
136 | |
137 static | |
138 void uprv_appendByteToHexString(char *dst, uint8_t val) { | |
139 uint32_t len = (uint32_t)uprv_strlen(dst); | |
140 *(dst+len) = T_CString_itosOffset((val >> 4)); | |
141 *(dst+len+1) = T_CString_itosOffset((val & 0xF)); | |
142 *(dst+len+2) = 0; | |
143 } | |
144 | |
145 /* this function makes a string with representation of a sortkey */ | |
146 static char* U_EXPORT2 sortKeyToString(const UCollator *coll, const uint8_t *sor
tkey, char *buffer, uint32_t *len) { | |
147 int32_t strength = UCOL_PRIMARY; | |
148 uint32_t res_size = 0; | |
149 UBool doneCase = FALSE; | |
150 UErrorCode errorCode = U_ZERO_ERROR; | |
151 | |
152 char *current = buffer; | |
153 const uint8_t *currentSk = sortkey; | |
154 | |
155 uprv_strcpy(current, "["); | |
156 | |
157 while(strength <= UCOL_QUATERNARY && strength <= ucol_getStrength(coll)) { | |
158 if(strength > UCOL_PRIMARY) { | |
159 uprv_strcat(current, " . "); | |
160 } | |
161 while(*currentSk != 0x01 && *currentSk != 0x00) { /* print a level */ | |
162 uprv_appendByteToHexString(current, *currentSk++); | |
163 uprv_strcat(current, " "); | |
164 } | |
165 if(ucol_getAttribute(coll, UCOL_CASE_LEVEL, &errorCode) == UCOL_ON && st
rength == UCOL_SECONDARY && doneCase == FALSE) { | |
166 doneCase = TRUE; | |
167 } else if(ucol_getAttribute(coll, UCOL_CASE_LEVEL, &errorCode) == UCOL_O
FF || doneCase == TRUE || strength != UCOL_SECONDARY) { | |
168 strength ++; | |
169 } | |
170 if (*currentSk) { | |
171 uprv_appendByteToHexString(current, *currentSk++); /* This should pr
int '01' */ | |
172 } | |
173 if(strength == UCOL_QUATERNARY && ucol_getAttribute(coll, UCOL_ALTERNATE
_HANDLING, &errorCode) == UCOL_NON_IGNORABLE) { | |
174 break; | |
175 } | |
176 } | |
177 | |
178 if(ucol_getStrength(coll) == UCOL_IDENTICAL) { | |
179 uprv_strcat(current, " . "); | |
180 while(*currentSk != 0) { | |
181 uprv_appendByteToHexString(current, *currentSk++); | |
182 uprv_strcat(current, " "); | |
183 } | |
184 | |
185 uprv_appendByteToHexString(current, *currentSk++); | |
186 } | |
187 uprv_strcat(current, "]"); | |
188 | |
189 if(res_size > *len) { | |
190 return NULL; | |
191 } | |
192 | |
193 return buffer; | |
194 } | |
195 | |
196 void addAllCollTest(TestNode** root) | |
197 { | |
198 addTest(root, &TestPrimary, "tscoll/callcoll/TestPrimary"); | |
199 addTest(root, &TestSecondary, "tscoll/callcoll/TestSecondary"); | |
200 addTest(root, &TestTertiary, "tscoll/callcoll/TestTertiary"); | |
201 addTest(root, &TestIdentical, "tscoll/callcoll/TestIdentical"); | |
202 addTest(root, &TestExtra, "tscoll/callcoll/TestExtra"); | |
203 addTest(root, &TestJB581, "tscoll/callcoll/TestJB581"); | |
204 addTest(root, &TestVariableTop, "tscoll/callcoll/TestVariableTop"); | |
205 addTest(root, &TestSurrogates, "tscoll/callcoll/TestSurrogates"); | |
206 addTest(root, &TestInvalidRules, "tscoll/callcoll/TestInvalidRules"); | |
207 addTest(root, &TestJB1401, "tscoll/callcoll/TestJB1401"); | |
208 addTest(root, &TestJitterbug1098, "tscoll/callcoll/TestJitterbug1098"); | |
209 addTest(root, &TestFCDCrash, "tscoll/callcoll/TestFCDCrash"); | |
210 addTest(root, &TestJ5298, "tscoll/callcoll/TestJ5298"); | |
211 } | |
212 | |
213 UBool hasCollationElements(const char *locName) { | |
214 | |
215 UErrorCode status = U_ZERO_ERROR; | |
216 | |
217 UResourceBundle *loc = ures_open(U_ICUDATA_NAME U_TREE_SEPARATOR_STRING "coll"
, locName, &status);; | |
218 | |
219 if(U_SUCCESS(status)) { | |
220 status = U_ZERO_ERROR; | |
221 loc = ures_getByKey(loc, "collations", loc, &status); | |
222 ures_close(loc); | |
223 if(status == U_ZERO_ERROR) { /* do the test - there are real elements */ | |
224 return TRUE; | |
225 } | |
226 } | |
227 return FALSE; | |
228 } | |
229 | |
230 static UCollationResult compareUsingPartials(UCollator *coll, const UChar source
[], int32_t sLen, const UChar target[], int32_t tLen, int32_t pieceSize, UErrorC
ode *status) { | |
231 int32_t partialSKResult = 0; | |
232 UCharIterator sIter, tIter; | |
233 uint32_t sState[2], tState[2]; | |
234 int32_t sSize = pieceSize, tSize = pieceSize; | |
235 /*int32_t i = 0;*/ | |
236 uint8_t sBuf[16384], tBuf[16384]; | |
237 if(pieceSize > 16384) { | |
238 log_err("Partial sortkey size buffer too small. Please consider increasing t
he buffer!\n"); | |
239 *status = U_BUFFER_OVERFLOW_ERROR; | |
240 return UCOL_EQUAL; | |
241 } | |
242 *status = U_ZERO_ERROR; | |
243 sState[0] = 0; sState[1] = 0; | |
244 tState[0] = 0; tState[1] = 0; | |
245 while(sSize == pieceSize && tSize == pieceSize && partialSKResult == 0) { | |
246 uiter_setString(&sIter, source, sLen); | |
247 uiter_setString(&tIter, target, tLen); | |
248 sSize = ucol_nextSortKeyPart(coll, &sIter, sState, sBuf, pieceSize, status); | |
249 tSize = ucol_nextSortKeyPart(coll, &tIter, tState, tBuf, pieceSize, status); | |
250 | |
251 if(sState[0] != 0 || tState[0] != 0) { | |
252 /*log_verbose("State != 0 : %08X %08X\n", sState[0], tState[0]);*/ | |
253 } | |
254 /*log_verbose("%i ", i++);*/ | |
255 | |
256 partialSKResult = memcmp(sBuf, tBuf, pieceSize); | |
257 } | |
258 | |
259 if(partialSKResult < 0) { | |
260 return UCOL_LESS; | |
261 } else if(partialSKResult > 0) { | |
262 return UCOL_GREATER; | |
263 } else { | |
264 return UCOL_EQUAL; | |
265 } | |
266 } | |
267 | |
268 static void doTestVariant(UCollator* myCollation, const UChar source[], const UC
har target[], UCollationResult result) | |
269 { | |
270 int32_t sortklen1, sortklen2, sortklenmax, sortklenmin; | |
271 int temp=0, gSortklen1=0,gSortklen2=0; | |
272 UCollationResult compareResult, compareResulta, keyResult, compareResultIter
= result; | |
273 uint8_t *sortKey1, *sortKey2, *sortKey1a, *sortKey2a; | |
274 uint32_t sLen = u_strlen(source); | |
275 uint32_t tLen = u_strlen(target); | |
276 char buffer[256]; | |
277 uint32_t len; | |
278 UErrorCode status = U_ZERO_ERROR; | |
279 UColAttributeValue norm = ucol_getAttribute(myCollation, UCOL_NORMALIZATION_
MODE, &status); | |
280 | |
281 UCharIterator sIter, tIter; | |
282 | |
283 compareResult = ucol_strcoll(myCollation, source, sLen, target, tLen); | |
284 if (compareResult != result) { | |
285 log_err("ucol_strcoll with explicit length returned wrong result (%i exp
. %i): %s, %s\n", | |
286 compareResult, result, aescstrdup(source,-1), aescstrdup(target,-1))
; | |
287 } | |
288 compareResulta = ucol_strcoll(myCollation, source, -1, target, -1); | |
289 if (compareResulta != result) { | |
290 log_err("ucol_strcoll with null terminated strings returned wrong result
(%i exp. %i): %s, %s\n", | |
291 compareResult, result, aescstrdup(source,-1), aescstrdup(target,-1))
; | |
292 } | |
293 | |
294 uiter_setString(&sIter, source, sLen); | |
295 uiter_setString(&tIter, target, tLen); | |
296 compareResultIter = ucol_strcollIter(myCollation, &sIter, &tIter, &status); | |
297 if(compareResultIter != result) { | |
298 log_err("different results in iterative comparison for UTF-16 encoded st
rings. %s, %s\n", aescstrdup(source,-1), aescstrdup(target,-1)); | |
299 } | |
300 | |
301 /* convert the strings to UTF-8 and do try comparing with char iterator and
ucol_strcollUTF8 */ | |
302 { | |
303 char utf8Source[256], utf8Target[256]; | |
304 int32_t utf8SourceLen = 0, utf8TargetLen = 0; | |
305 | |
306 u_strToUTF8(utf8Source, 256, &utf8SourceLen, source, sLen, &status); | |
307 if(U_FAILURE(status)) { /* probably buffer is not big enough */ | |
308 log_verbose("Src UTF-8 buffer too small! Will not compare!\n"); | |
309 } else { | |
310 u_strToUTF8(utf8Target, 256, &utf8TargetLen, target, tLen, &status); | |
311 if(U_SUCCESS(status)) { | |
312 { | |
313 /* ucol_strcollUTF8 */ | |
314 compareResulta = ucol_strcollUTF8(myCollation, utf8Source, u
tf8SourceLen, utf8Target, utf8TargetLen, &status); | |
315 if (U_FAILURE(status)) { | |
316 log_err("Error in ucol_strcollUTF8 with explicit length\
n"); | |
317 status = U_ZERO_ERROR; | |
318 } else if (compareResulta != result) { | |
319 log_err("ucol_strcollUTF8 with explicit length returned
wrong result (%i exp. %i): %s, %s\n", | |
320 compareResulta, result, aescstrdup(source,-1), aescs
trdup(target,-1)); | |
321 } | |
322 compareResulta = ucol_strcollUTF8(myCollation, utf8Source, -
1, utf8Target, -1, &status); | |
323 if (U_FAILURE(status)) { | |
324 log_err("Error in ucol_strcollUTF8 with null terminated
strings\n"); | |
325 status = U_ZERO_ERROR; | |
326 } else if (compareResulta != result) { | |
327 log_err("ucol_strcollUTF8 with null terminated strings r
eturned wrong result (%i exp. %i): %s, %s\n", | |
328 compareResulta, result, aescstrdup(source,-1), aescs
trdup(target,-1)); | |
329 } | |
330 } | |
331 | |
332 { | |
333 /* char iterator over UTF8 */ | |
334 UCollationResult compareResultUTF8Iter = result, compareResu
ltUTF8IterNorm = result; | |
335 | |
336 uiter_setUTF8(&sIter, utf8Source, utf8SourceLen); | |
337 uiter_setUTF8(&tIter, utf8Target, utf8TargetLen); | |
338 compareResultUTF8Iter = ucol_strcollIter(myCollation, &sIter
, &tIter, &status); | |
339 | |
340 ucol_setAttribute(myCollation, UCOL_NORMALIZATION_MODE, UCOL
_ON, &status); | |
341 sIter.move(&sIter, 0, UITER_START); | |
342 tIter.move(&tIter, 0, UITER_START); | |
343 compareResultUTF8IterNorm = ucol_strcollIter(myCollation, &s
Iter, &tIter, &status); | |
344 | |
345 ucol_setAttribute(myCollation, UCOL_NORMALIZATION_MODE, norm
, &status); | |
346 if(compareResultUTF8Iter != compareResultIter) { | |
347 log_err("different results in iterative comparison for U
TF-16 and UTF-8 encoded strings. %s, %s\n", aescstrdup(source,-1), aescstrdup(ta
rget,-1)); | |
348 } | |
349 if(compareResultUTF8Iter != compareResultUTF8IterNorm) { | |
350 log_err("different results in iterative when normalizati
on is turned on with UTF-8 strings. %s, %s\n", aescstrdup(source,-1), aescstrdup
(target,-1)); | |
351 } | |
352 } | |
353 } else { | |
354 log_verbose("Target UTF-8 buffer too small! Did not compare!\n")
; | |
355 } | |
356 if(U_FAILURE(status)) { | |
357 log_verbose("UTF-8 strcoll failed! Ignoring result\n"); | |
358 } | |
359 } | |
360 } | |
361 | |
362 /* testing the partial sortkeys */ | |
363 if(1) { /*!QUICK*/ | |
364 int32_t i = 0; | |
365 int32_t partialSizes[] = { 3, 1, 2, 4, 8, 20, 80 }; /* just size 3 in the
quick mode */ | |
366 int32_t partialSizesSize = 1; | |
367 if(getTestOption(QUICK_OPTION) <= 0) { | |
368 partialSizesSize = 7; | |
369 } | |
370 /*log_verbose("partial sortkey test piecesize=");*/ | |
371 for(i = 0; i < partialSizesSize; i++) { | |
372 UCollationResult partialSKResult = result, partialNormalizedSKResult = r
esult; | |
373 /*log_verbose("%i ", partialSizes[i]);*/ | |
374 | |
375 partialSKResult = compareUsingPartials(myCollation, source, sLen, target
, tLen, partialSizes[i], &status); | |
376 if(partialSKResult != result) { | |
377 log_err("Partial sortkey comparison returned wrong result (%i exp. %i)
: %s, %s (size %i)\n", | |
378 partialSKResult, result, | |
379 aescstrdup(source,-1), aescstrdup(target,-1), partialSizes[i]); | |
380 } | |
381 | |
382 if(getTestOption(QUICK_OPTION) <= 0 && norm != UCOL_ON) { | |
383 /*log_verbose("N ");*/ | |
384 ucol_setAttribute(myCollation, UCOL_NORMALIZATION_MODE, UCOL_ON, &stat
us); | |
385 partialNormalizedSKResult = compareUsingPartials(myCollation, source,
sLen, target, tLen, partialSizes[i], &status); | |
386 ucol_setAttribute(myCollation, UCOL_NORMALIZATION_MODE, norm, &status)
; | |
387 if(partialSKResult != partialNormalizedSKResult) { | |
388 log_err("Partial sortkey comparison gets different result when norma
lization is on: %s, %s (size %i)\n", | |
389 aescstrdup(source,-1), aescstrdup(target,-1), partialSizes[i]); | |
390 } | |
391 } | |
392 } | |
393 /*log_verbose("\n");*/ | |
394 } | |
395 | |
396 sortklen1=ucol_getSortKey(myCollation, source, sLen, NULL, 0); | |
397 sortklen2=ucol_getSortKey(myCollation, target, tLen, NULL, 0); | |
398 | |
399 sortklenmax = (sortklen1>sortklen2?sortklen1:sortklen2); | |
400 sortklenmin = (sortklen1<sortklen2?sortklen1:sortklen2); | |
401 (void)sortklenmin; /* Suppress set but not used warning. */ | |
402 | |
403 sortKey1 =(uint8_t*)malloc(sizeof(uint8_t) * (sortklenmax+1)); | |
404 sortKey1a=(uint8_t*)malloc(sizeof(uint8_t) * (sortklenmax+1)); | |
405 ucol_getSortKey(myCollation, source, sLen, sortKey1, sortklen1+1); | |
406 ucol_getSortKey(myCollation, source, -1, sortKey1a, sortklen1+1); | |
407 | |
408 sortKey2 =(uint8_t*)malloc(sizeof(uint8_t) * (sortklenmax+1)); | |
409 sortKey2a=(uint8_t*)malloc(sizeof(uint8_t) * (sortklenmax+1)); | |
410 ucol_getSortKey(myCollation, target, tLen, sortKey2, sortklen2+1); | |
411 ucol_getSortKey(myCollation, target, -1, sortKey2a, sortklen2+1); | |
412 | |
413 /* Check that sort key generated with null terminated string is identical *
/ | |
414 /* to that generted with a length specified. *
/ | |
415 if (uprv_strcmp((const char *)sortKey1, (const char *)sortKey1a) != 0 || | |
416 uprv_strcmp((const char *)sortKey2, (const char *)sortKey2a) != 0 ) { | |
417 log_err("Sort Keys from null terminated and explicit length strings diff
er.\n"); | |
418 } | |
419 | |
420 /*memcmp(sortKey1, sortKey2,sortklenmax);*/ | |
421 temp= uprv_strcmp((const char *)sortKey1, (const char *)sortKey2); | |
422 gSortklen1 = uprv_strlen((const char *)sortKey1)+1; | |
423 gSortklen2 = uprv_strlen((const char *)sortKey2)+1; | |
424 if(sortklen1 != gSortklen1){ | |
425 log_err("SortKey length does not match Expected: %i Got: %i\n",sortklen1
, gSortklen1); | |
426 log_verbose("Generated sortkey: %s\n", sortKeyToString(myCollation, sort
Key1, buffer, &len)); | |
427 } | |
428 if(sortklen2!= gSortklen2){ | |
429 log_err("SortKey length does not match Expected: %i Got: %i\n", sortklen
2, gSortklen2); | |
430 log_verbose("Generated sortkey: %s\n", sortKeyToString(myCollation, sort
Key2, buffer, &len)); | |
431 } | |
432 | |
433 if(temp < 0) { | |
434 keyResult=UCOL_LESS; | |
435 } | |
436 else if(temp > 0) { | |
437 keyResult= UCOL_GREATER; | |
438 } | |
439 else { | |
440 keyResult = UCOL_EQUAL; | |
441 } | |
442 reportCResult( source, target, sortKey1, sortKey2, compareResult, keyResult,
compareResultIter, result ); | |
443 free(sortKey1); | |
444 free(sortKey2); | |
445 free(sortKey1a); | |
446 free(sortKey2a); | |
447 | |
448 } | |
449 | |
450 void doTest(UCollator* myCollation, const UChar source[], const UChar target[],
UCollationResult result) | |
451 { | |
452 if(myCollation) { | |
453 doTestVariant(myCollation, source, target, result); | |
454 if(result == UCOL_LESS) { | |
455 doTestVariant(myCollation, target, source, UCOL_GREATER); | |
456 } else if(result == UCOL_GREATER) { | |
457 doTestVariant(myCollation, target, source, UCOL_LESS); | |
458 } else { | |
459 doTestVariant(myCollation, target, source, UCOL_EQUAL); | |
460 } | |
461 } else { | |
462 log_data_err("No collator! Any data around?\n"); | |
463 } | |
464 } | |
465 | |
466 | |
467 /** | |
468 * Return an integer array containing all of the collation orders | |
469 * returned by calls to next on the specified iterator | |
470 */ | |
471 OrderAndOffset* getOrders(UCollationElements *iter, int32_t *orderLength) | |
472 { | |
473 UErrorCode status; | |
474 int32_t order; | |
475 int32_t maxSize = 100; | |
476 int32_t size = 0; | |
477 int32_t offset = ucol_getOffset(iter); | |
478 OrderAndOffset *temp; | |
479 OrderAndOffset *orders =(OrderAndOffset *)malloc(sizeof(OrderAndOffset) * ma
xSize); | |
480 status= U_ZERO_ERROR; | |
481 | |
482 | |
483 while ((order=ucol_next(iter, &status)) != UCOL_NULLORDER) | |
484 { | |
485 if (size == maxSize) | |
486 { | |
487 maxSize *= 2; | |
488 temp = (OrderAndOffset *)malloc(sizeof(OrderAndOffset) * maxSize); | |
489 | |
490 memcpy(temp, orders, size * sizeof(OrderAndOffset)); | |
491 free(orders); | |
492 orders = temp; | |
493 | |
494 } | |
495 | |
496 orders[size].order = order; | |
497 orders[size].offset = offset; | |
498 | |
499 offset = ucol_getOffset(iter); | |
500 size += 1; | |
501 } | |
502 | |
503 if (maxSize > size && size > 0) | |
504 { | |
505 temp = (OrderAndOffset *)malloc(sizeof(OrderAndOffset) * size); | |
506 | |
507 memcpy(temp, orders, size * sizeof(OrderAndOffset)); | |
508 free(orders); | |
509 orders = temp; | |
510 | |
511 | |
512 } | |
513 | |
514 *orderLength = size; | |
515 return orders; | |
516 } | |
517 | |
518 | |
519 void | |
520 backAndForth(UCollationElements *iter) | |
521 { | |
522 /* Run through the iterator forwards and stick it into an array */ | |
523 int32_t idx, o; | |
524 UErrorCode status = U_ZERO_ERROR; | |
525 int32_t orderLength = 0; | |
526 OrderAndOffset *orders = getOrders(iter, &orderLength); | |
527 | |
528 | |
529 /* Now go through it backwards and make sure we get the same values */ | |
530 idx = orderLength; | |
531 ucol_reset(iter); | |
532 | |
533 /* synwee : changed */ | |
534 while ((o = ucol_previous(iter, &status)) != UCOL_NULLORDER) { | |
535 #if TEST_OFFSETS | |
536 int32_t offset = | |
537 #endif | |
538 ucol_getOffset(iter); | |
539 | |
540 idx -= 1; | |
541 if (o != orders[idx].order) { | |
542 if (o == 0) | |
543 idx ++; | |
544 else { | |
545 while (idx > 0 && orders[-- idx].order == 0) { | |
546 /* nothing... */ | |
547 } | |
548 | |
549 if (o != orders[idx].order) { | |
550 log_err("Mismatched order at index %d: 0x%8.8X vs. 0x%8.8X\n", idx
, | |
551 orders[idx].order, o); | |
552 goto bail; | |
553 } | |
554 } | |
555 } | |
556 | |
557 #if TEST_OFFSETS | |
558 if (offset != orders[idx].offset) { | |
559 log_err("Mismatched offset at index %d: %d vs. %d\n", idx, | |
560 orders[idx].offset, offset); | |
561 goto bail; | |
562 } | |
563 #endif | |
564 | |
565 } | |
566 | |
567 while (idx != 0 && orders[idx - 1].order == 0) { | |
568 idx -= 1; | |
569 } | |
570 | |
571 if (idx != 0) { | |
572 log_err("Didn't get back to beginning - index is %d\n", idx); | |
573 | |
574 ucol_reset(iter); | |
575 log_err("\nnext: "); | |
576 | |
577 if ((o = ucol_next(iter, &status)) != UCOL_NULLORDER) { | |
578 log_err("Error at %x\n", o); | |
579 } | |
580 | |
581 log_err("\nprev: "); | |
582 | |
583 if ((o = ucol_previous(iter, &status)) != UCOL_NULLORDER) { | |
584 log_err("Error at %x\n", o); | |
585 } | |
586 | |
587 log_verbose("\n"); | |
588 } | |
589 | |
590 bail: | |
591 free(orders); | |
592 } | |
593 | |
594 void genericOrderingTestWithResult(UCollator *coll, const char * const s[], uint
32_t size, UCollationResult result) { | |
595 UChar t1[2048] = {0}; | |
596 UChar t2[2048] = {0}; | |
597 UCollationElements *iter; | |
598 UErrorCode status = U_ZERO_ERROR; | |
599 | |
600 uint32_t i = 0, j = 0; | |
601 log_verbose("testing sequence:\n"); | |
602 for(i = 0; i < size; i++) { | |
603 log_verbose("%s\n", s[i]); | |
604 } | |
605 | |
606 iter = ucol_openElements(coll, t1, u_strlen(t1), &status); | |
607 if (U_FAILURE(status)) { | |
608 log_err("Creation of iterator failed\n"); | |
609 } | |
610 for(i = 0; i < size-1; i++) { | |
611 for(j = i+1; j < size; j++) { | |
612 u_unescape(s[i], t1, 2048); | |
613 u_unescape(s[j], t2, 2048); | |
614 doTest(coll, t1, t2, result); | |
615 /* synwee : added collation element iterator test */ | |
616 ucol_setText(iter, t1, u_strlen(t1), &status); | |
617 backAndForth(iter); | |
618 ucol_setText(iter, t2, u_strlen(t2), &status); | |
619 backAndForth(iter); | |
620 } | |
621 } | |
622 ucol_closeElements(iter); | |
623 } | |
624 | |
625 void genericOrderingTest(UCollator *coll, const char * const s[], uint32_t size)
{ | |
626 genericOrderingTestWithResult(coll, s, size, UCOL_LESS); | |
627 } | |
628 | |
629 void genericLocaleStarter(const char *locale, const char * const s[], uint32_t s
ize) { | |
630 UErrorCode status = U_ZERO_ERROR; | |
631 UCollator *coll = ucol_open(locale, &status); | |
632 | |
633 log_verbose("Locale starter for %s\n", locale); | |
634 | |
635 if(U_SUCCESS(status)) { | |
636 genericOrderingTest(coll, s, size); | |
637 } else if(status == U_FILE_ACCESS_ERROR) { | |
638 log_data_err("Is your data around?\n"); | |
639 return; | |
640 } else { | |
641 log_err("Unable to open collator for locale %s\n", locale); | |
642 } | |
643 ucol_close(coll); | |
644 } | |
645 | |
646 void genericLocaleStarterWithResult(const char *locale, const char * const s[],
uint32_t size, UCollationResult result) { | |
647 UErrorCode status = U_ZERO_ERROR; | |
648 UCollator *coll = ucol_open(locale, &status); | |
649 | |
650 log_verbose("Locale starter for %s\n", locale); | |
651 | |
652 if(U_SUCCESS(status)) { | |
653 genericOrderingTestWithResult(coll, s, size, result); | |
654 } else if(status == U_FILE_ACCESS_ERROR) { | |
655 log_data_err("Is your data around?\n"); | |
656 return; | |
657 } else { | |
658 log_err("Unable to open collator for locale %s\n", locale); | |
659 } | |
660 ucol_close(coll); | |
661 } | |
662 | |
663 /* currently not used with options */ | |
664 void genericRulesStarterWithOptionsAndResult(const char *rules, const char * con
st s[], uint32_t size, const UColAttribute *attrs, const UColAttributeValue *val
ues, uint32_t attsize, UCollationResult result) { | |
665 UErrorCode status = U_ZERO_ERROR; | |
666 UChar rlz[RULE_BUFFER_LEN] = { 0 }; | |
667 uint32_t rlen = u_unescape(rules, rlz, RULE_BUFFER_LEN); | |
668 uint32_t i; | |
669 | |
670 UCollator *coll = ucol_openRules(rlz, rlen, UCOL_DEFAULT, UCOL_DEFAULT,NULL, &
status); | |
671 | |
672 log_verbose("Rules starter for %s\n", rules); | |
673 | |
674 if(U_SUCCESS(status)) { | |
675 log_verbose("Setting attributes\n"); | |
676 for(i = 0; i < attsize; i++) { | |
677 ucol_setAttribute(coll, attrs[i], values[i], &status); | |
678 } | |
679 | |
680 genericOrderingTestWithResult(coll, s, size, result); | |
681 } else { | |
682 log_err_status(status, "Unable to open collator with rules %s\n", rules); | |
683 } | |
684 ucol_close(coll); | |
685 } | |
686 | |
687 void genericLocaleStarterWithOptionsAndResult(const char *locale, const char * c
onst s[], uint32_t size, const UColAttribute *attrs, const UColAttributeValue *v
alues, uint32_t attsize, UCollationResult result) { | |
688 UErrorCode status = U_ZERO_ERROR; | |
689 uint32_t i; | |
690 | |
691 UCollator *coll = ucol_open(locale, &status); | |
692 | |
693 log_verbose("Locale starter for %s\n", locale); | |
694 | |
695 if(U_SUCCESS(status)) { | |
696 | |
697 log_verbose("Setting attributes\n"); | |
698 for(i = 0; i < attsize; i++) { | |
699 ucol_setAttribute(coll, attrs[i], values[i], &status); | |
700 } | |
701 | |
702 genericOrderingTestWithResult(coll, s, size, result); | |
703 } else { | |
704 log_err_status(status, "Unable to open collator for locale %s\n", locale); | |
705 } | |
706 ucol_close(coll); | |
707 } | |
708 | |
709 void genericLocaleStarterWithOptions(const char *locale, const char * const s[],
uint32_t size, const UColAttribute *attrs, const UColAttributeValue *values, ui
nt32_t attsize) { | |
710 genericLocaleStarterWithOptionsAndResult(locale, s, size, attrs, values, attsi
ze, UCOL_LESS); | |
711 } | |
712 | |
713 void genericRulesStarterWithResult(const char *rules, const char * const s[], ui
nt32_t size, UCollationResult result) { | |
714 UErrorCode status = U_ZERO_ERROR; | |
715 UChar rlz[RULE_BUFFER_LEN] = { 0 }; | |
716 uint32_t rlen = u_unescape(rules, rlz, RULE_BUFFER_LEN); | |
717 | |
718 UCollator *coll = NULL; | |
719 coll = ucol_openRules(rlz, rlen, UCOL_DEFAULT, UCOL_DEFAULT,NULL, &status); | |
720 log_verbose("Rules starter for %s\n", rules); | |
721 | |
722 if(U_SUCCESS(status)) { | |
723 genericOrderingTestWithResult(coll, s, size, result); | |
724 ucol_close(coll); | |
725 } else if(status == U_FILE_ACCESS_ERROR) { | |
726 log_data_err("Is your data around?\n"); | |
727 } else { | |
728 log_err("Unable to open collator with rules %s\n", rules); | |
729 } | |
730 } | |
731 | |
732 void genericRulesStarter(const char *rules, const char * const s[], uint32_t siz
e) { | |
733 genericRulesStarterWithResult(rules, s, size, UCOL_LESS); | |
734 } | |
735 | |
736 static void TestTertiary() | |
737 { | |
738 int32_t len,i; | |
739 UCollator *myCollation; | |
740 UErrorCode status=U_ZERO_ERROR; | |
741 static const char str[]="& C < ch, cH, Ch, CH & Five, 5 & Four, 4 & one, 1 &
Ampersand; '&' & Two, 2 "; | |
742 UChar rules[sizeof(str)]; | |
743 len = strlen(str); | |
744 u_uastrcpy(rules, str); | |
745 | |
746 myCollation=ucol_openRules(rules, len, UCOL_OFF, UCOL_DEFAULT_STRENGTH, NULL
, &status); | |
747 if(U_FAILURE(status)){ | |
748 log_err_status(status, "ERROR: in creation of rule based collator :%s\n"
, myErrorName(status)); | |
749 return; | |
750 } | |
751 | |
752 ucol_setStrength(myCollation, UCOL_TERTIARY); | |
753 for (i = 0; i < 17 ; i++) | |
754 { | |
755 doTest(myCollation, testSourceCases[i], testTargetCases[i], results[i]); | |
756 } | |
757 ucol_close(myCollation); | |
758 myCollation = 0; | |
759 } | |
760 | |
761 static void TestPrimary( ) | |
762 { | |
763 int32_t len,i; | |
764 UCollator *myCollation; | |
765 UErrorCode status=U_ZERO_ERROR; | |
766 static const char str[]="& C < ch, cH, Ch, CH & Five, 5 & Four, 4 & one, 1 &
Ampersand; '&' & Two, 2 "; | |
767 UChar rules[sizeof(str)]; | |
768 len = strlen(str); | |
769 u_uastrcpy(rules, str); | |
770 | |
771 myCollation=ucol_openRules(rules, len, UCOL_OFF, UCOL_DEFAULT_STRENGTH,NULL,
&status); | |
772 if(U_FAILURE(status)){ | |
773 log_err_status(status, "ERROR: in creation of rule based collator :%s\n"
, myErrorName(status)); | |
774 return; | |
775 } | |
776 ucol_setStrength(myCollation, UCOL_PRIMARY); | |
777 | |
778 for (i = 17; i < 26 ; i++) | |
779 { | |
780 | |
781 doTest(myCollation, testSourceCases[i], testTargetCases[i], results[i]); | |
782 } | |
783 ucol_close(myCollation); | |
784 myCollation = 0; | |
785 } | |
786 | |
787 static void TestSecondary() | |
788 { | |
789 int32_t i; | |
790 int32_t len; | |
791 UCollator *myCollation; | |
792 UErrorCode status=U_ZERO_ERROR; | |
793 static const char str[]="& C < ch, cH, Ch, CH & Five, 5 & Four, 4 & one, 1 &
Ampersand; '&' & Two, 2 "; | |
794 UChar rules[sizeof(str)]; | |
795 len = strlen(str); | |
796 u_uastrcpy(rules, str); | |
797 | |
798 myCollation=ucol_openRules(rules, len, UCOL_OFF, UCOL_DEFAULT_STRENGTH,NULL,
&status); | |
799 if(U_FAILURE(status)){ | |
800 log_err_status(status, "ERROR: in creation of rule based collator :%s\n"
, myErrorName(status)); | |
801 return; | |
802 } | |
803 ucol_setStrength(myCollation, UCOL_SECONDARY); | |
804 for (i = 26; i < 34 ; i++) | |
805 { | |
806 doTest(myCollation, testSourceCases[i], testTargetCases[i], results[i]); | |
807 } | |
808 ucol_close(myCollation); | |
809 myCollation = 0; | |
810 } | |
811 | |
812 static void TestIdentical() | |
813 { | |
814 int32_t i; | |
815 int32_t len; | |
816 UCollator *myCollation; | |
817 UErrorCode status=U_ZERO_ERROR; | |
818 static const char str[]="& C < ch, cH, Ch, CH & Five, 5 & Four, 4 & one, 1 &
Ampersand; '&' & Two, 2 "; | |
819 UChar rules[sizeof(str)]; | |
820 len = strlen(str); | |
821 u_uastrcpy(rules, str); | |
822 | |
823 myCollation=ucol_openRules(rules, len, UCOL_OFF, UCOL_IDENTICAL, NULL,&statu
s); | |
824 if(U_FAILURE(status)){ | |
825 log_err_status(status, "ERROR: in creation of rule based collator :%s\n"
, myErrorName(status)); | |
826 return; | |
827 } | |
828 for(i= 34; i<37; i++) | |
829 { | |
830 doTest(myCollation, testSourceCases[i], testTargetCases[i], results[i]); | |
831 } | |
832 ucol_close(myCollation); | |
833 myCollation = 0; | |
834 } | |
835 | |
836 static void TestExtra() | |
837 { | |
838 int32_t i, j; | |
839 int32_t len; | |
840 UCollator *myCollation; | |
841 UErrorCode status = U_ZERO_ERROR; | |
842 static const char str[]="& C < ch, cH, Ch, CH & Five, 5 & Four, 4 & one, 1 &
Ampersand; '&' & Two, 2 "; | |
843 UChar rules[sizeof(str)]; | |
844 len = strlen(str); | |
845 u_uastrcpy(rules, str); | |
846 | |
847 myCollation=ucol_openRules(rules, len, UCOL_OFF, UCOL_DEFAULT_STRENGTH,NULL,
&status); | |
848 if(U_FAILURE(status)){ | |
849 log_err_status(status, "ERROR: in creation of rule based collator :%s\n"
, myErrorName(status)); | |
850 return; | |
851 } | |
852 ucol_setStrength(myCollation, UCOL_TERTIARY); | |
853 for (i = 0; i < COUNT_TEST_CASES-1 ; i++) | |
854 { | |
855 for (j = i + 1; j < COUNT_TEST_CASES; j += 1) | |
856 { | |
857 | |
858 doTest(myCollation, testCases[i], testCases[j], UCOL_LESS); | |
859 } | |
860 } | |
861 ucol_close(myCollation); | |
862 myCollation = 0; | |
863 } | |
864 | |
865 static void TestJB581(void) | |
866 { | |
867 int32_t bufferLen = 0; | |
868 UChar source [100]; | |
869 UChar target [100]; | |
870 UCollationResult result = UCOL_EQUAL; | |
871 uint8_t sourceKeyArray [100]; | |
872 uint8_t targetKeyArray [100]; | |
873 int32_t sourceKeyOut = 0, | |
874 targetKeyOut = 0; | |
875 UCollator *myCollator = 0; | |
876 UErrorCode status = U_ZERO_ERROR; | |
877 | |
878 /*u_uastrcpy(source, "This is a test.");*/ | |
879 /*u_uastrcpy(target, "THISISATEST.");*/ | |
880 u_uastrcpy(source, "THISISATEST."); | |
881 u_uastrcpy(target, "Thisisatest."); | |
882 | |
883 myCollator = ucol_open("en_US", &status); | |
884 if (U_FAILURE(status)){ | |
885 log_err_status(status, "ERROR: Failed to create the collator : %s\n", u_
errorName(status)); | |
886 return; | |
887 } | |
888 result = ucol_strcoll(myCollator, source, -1, target, -1); | |
889 /* result is 1, secondary differences only for ignorable space characters*/ | |
890 if (result != 1) | |
891 { | |
892 log_err("Comparing two strings with only secondary differences in C fail
ed.\n"); | |
893 } | |
894 /* To compare them with just primary differences */ | |
895 ucol_setStrength(myCollator, UCOL_PRIMARY); | |
896 result = ucol_strcoll(myCollator, source, -1, target, -1); | |
897 /* result is 0 */ | |
898 if (result != 0) | |
899 { | |
900 log_err("Comparing two strings with no differences in C failed.\n"); | |
901 } | |
902 /* Now, do the same comparison with keys */ | |
903 sourceKeyOut = ucol_getSortKey(myCollator, source, -1, sourceKeyArray, 100); | |
904 (void)sourceKeyOut; /* Suppress set but not used warning. */ | |
905 targetKeyOut = ucol_getSortKey(myCollator, target, -1, targetKeyArray, 100); | |
906 bufferLen = ((targetKeyOut > 100) ? 100 : targetKeyOut); | |
907 if (memcmp(sourceKeyArray, targetKeyArray, bufferLen) != 0) | |
908 { | |
909 log_err("Comparing two strings with sort keys in C failed.\n"); | |
910 } | |
911 ucol_close(myCollator); | |
912 } | |
913 | |
914 static void TestJB1401(void) | |
915 { | |
916 UCollator *myCollator = 0; | |
917 UErrorCode status = U_ZERO_ERROR; | |
918 static UChar NFD_UnsafeStartChars[] = { | |
919 0x0f73, /* Tibetan Vowel Sign II */ | |
920 0x0f75, /* Tibetan Vowel Sign UU */ | |
921 0x0f81, /* Tibetan Vowel Sign Reversed II */ | |
922 0 | |
923 }; | |
924 int i; | |
925 | |
926 | |
927 myCollator = ucol_open("en_US", &status); | |
928 if (U_FAILURE(status)){ | |
929 log_err_status(status, "ERROR: Failed to create the collator : %s\n", u_
errorName(status)); | |
930 return; | |
931 } | |
932 ucol_setAttribute(myCollator, UCOL_NORMALIZATION_MODE, UCOL_ON, &status); | |
933 if (U_FAILURE(status)){ | |
934 log_err("ERROR: Failed to set normalization mode ON for collator.\n"); | |
935 return; | |
936 } | |
937 | |
938 for (i=0; ; i++) { | |
939 UChar c; | |
940 UChar X[4]; | |
941 UChar Y[20]; | |
942 UChar Z[20]; | |
943 | |
944 /* Get the next funny character to be tested, and set up the | |
945 * three test strings X, Y, Z, consisting of an A-grave + test char, | |
946 * in original form, NFD, and then NFC form. | |
947 */ | |
948 c = NFD_UnsafeStartChars[i]; | |
949 if (c==0) {break;} | |
950 | |
951 X[0]=0xC0; X[1]=c; X[2]=0; /* \u00C0 is A Grave*/ | |
952 | |
953 unorm_normalize(X, -1, UNORM_NFD, 0, Y, 20, &status); | |
954 unorm_normalize(Y, -1, UNORM_NFC, 0, Z, 20, &status); | |
955 if (U_FAILURE(status)){ | |
956 log_err("ERROR: Failed to normalize test of character %x\n", c); | |
957 return; | |
958 } | |
959 | |
960 /* Collation test. All three strings should be equal. | |
961 * doTest does both strcoll and sort keys, with params in both orders. | |
962 */ | |
963 doTest(myCollator, X, Y, UCOL_EQUAL); | |
964 doTest(myCollator, X, Z, UCOL_EQUAL); | |
965 doTest(myCollator, Y, Z, UCOL_EQUAL); | |
966 | |
967 /* Run collation element iterators over the three strings. Results shou
ld be same for each. | |
968 */ | |
969 { | |
970 UCollationElements *ceiX, *ceiY, *ceiZ; | |
971 int32_t ceX, ceY, ceZ; | |
972 int j; | |
973 | |
974 ceiX = ucol_openElements(myCollator, X, -1, &status); | |
975 ceiY = ucol_openElements(myCollator, Y, -1, &status); | |
976 ceiZ = ucol_openElements(myCollator, Z, -1, &status); | |
977 if (U_FAILURE(status)) { | |
978 log_err("ERROR: uucol_openElements failed.\n"); | |
979 return; | |
980 } | |
981 | |
982 for (j=0;; j++) { | |
983 ceX = ucol_next(ceiX, &status); | |
984 ceY = ucol_next(ceiY, &status); | |
985 ceZ = ucol_next(ceiZ, &status); | |
986 if (U_FAILURE(status)) { | |
987 log_err("ERROR: ucol_next failed for iteration #%d.\n", j); | |
988 break; | |
989 } | |
990 if (ceX != ceY || ceY != ceZ) { | |
991 log_err("ERROR: ucol_next failed for iteration #%d.\n", j); | |
992 break; | |
993 } | |
994 if (ceX == UCOL_NULLORDER) { | |
995 break; | |
996 } | |
997 } | |
998 ucol_closeElements(ceiX); | |
999 ucol_closeElements(ceiY); | |
1000 ucol_closeElements(ceiZ); | |
1001 } | |
1002 } | |
1003 ucol_close(myCollator); | |
1004 } | |
1005 | |
1006 | |
1007 | |
1008 /** | |
1009 * Tests the [variable top] tag in rule syntax. Since the default [alternate] | |
1010 * tag has the value shifted, any codepoints before [variable top] should give | |
1011 * a primary ce of 0. | |
1012 */ | |
1013 static void TestVariableTop(void) | |
1014 { | |
1015 #if 0 | |
1016 /* | |
1017 * Starting with ICU 53, setting the variable top via a pseudo relation stri
ng | |
1018 * is not supported any more. | |
1019 * It was replaced by the [maxVariable symbol] setting. | |
1020 * See ICU tickets #9958 and #8032. | |
1021 */ | |
1022 static const char str[] = "&z = [variable top]"; | |
1023 int len = strlen(str); | |
1024 UChar rules[sizeof(str)]; | |
1025 UCollator *myCollation; | |
1026 UCollator *enCollation; | |
1027 UErrorCode status = U_ZERO_ERROR; | |
1028 UChar source[1]; | |
1029 UChar ch; | |
1030 uint8_t result[20]; | |
1031 uint8_t expected[20]; | |
1032 | |
1033 u_uastrcpy(rules, str); | |
1034 | |
1035 enCollation = ucol_open("en_US", &status); | |
1036 if (U_FAILURE(status)) { | |
1037 log_err_status(status, "ERROR: in creation of collator :%s\n", | |
1038 myErrorName(status)); | |
1039 return; | |
1040 } | |
1041 myCollation = ucol_openRules(rules, len, UCOL_OFF, | |
1042 UCOL_PRIMARY,NULL, &status); | |
1043 if (U_FAILURE(status)) { | |
1044 ucol_close(enCollation); | |
1045 log_err("ERROR: in creation of rule based collator :%s\n", | |
1046 myErrorName(status)); | |
1047 return; | |
1048 } | |
1049 | |
1050 ucol_setStrength(enCollation, UCOL_PRIMARY); | |
1051 ucol_setAttribute(enCollation, UCOL_ALTERNATE_HANDLING, UCOL_SHIFTED, | |
1052 &status); | |
1053 ucol_setAttribute(myCollation, UCOL_ALTERNATE_HANDLING, UCOL_SHIFTED, | |
1054 &status); | |
1055 | |
1056 if (ucol_getAttribute(myCollation, UCOL_ALTERNATE_HANDLING, &status) != | |
1057 UCOL_SHIFTED || U_FAILURE(status)) { | |
1058 log_err("ERROR: ALTERNATE_HANDLING value can not be set to SHIFTED\n"); | |
1059 } | |
1060 | |
1061 uprv_memset(expected, 0, 20); | |
1062 | |
1063 /* space is supposed to be a variable */ | |
1064 source[0] = ' '; | |
1065 len = ucol_getSortKey(enCollation, source, 1, result, | |
1066 sizeof(result)); | |
1067 | |
1068 if (uprv_memcmp(expected, result, len) != 0) { | |
1069 log_err("ERROR: SHIFTED alternate does not return 0 for primary of space
\n"); | |
1070 } | |
1071 | |
1072 ch = 'a'; | |
1073 while (ch < 'z') { | |
1074 source[0] = ch; | |
1075 len = ucol_getSortKey(myCollation, source, 1, result, | |
1076 sizeof(result)); | |
1077 if (uprv_memcmp(expected, result, len) != 0) { | |
1078 log_err("ERROR: SHIFTED alternate does not return 0 for primary of %
c\n", | |
1079 ch); | |
1080 } | |
1081 ch ++; | |
1082 } | |
1083 | |
1084 ucol_close(enCollation); | |
1085 ucol_close(myCollation); | |
1086 enCollation = NULL; | |
1087 myCollation = NULL; | |
1088 #endif | |
1089 } | |
1090 | |
1091 /** | |
1092 * Tests surrogate support. | |
1093 * NOTE: This test used \\uD801\\uDC01 pair, which is now assigned to Desseret | |
1094 * Therefore, another (unassigned) code point was used for this test. | |
1095 */ | |
1096 static void TestSurrogates(void) | |
1097 { | |
1098 static const char str[] = | |
1099 "&z<'\\uD800\\uDC00'<'\\uD800\\uDC0A\\u0308'<A"; | |
1100 int len = strlen(str); | |
1101 int rlen = 0; | |
1102 UChar rules[sizeof(str)]; | |
1103 UCollator *myCollation; | |
1104 UCollator *enCollation; | |
1105 UErrorCode status = U_ZERO_ERROR; | |
1106 UChar source[][4] = | |
1107 {{'z', 0, 0}, {0xD800, 0xDC00, 0}, {0xD800, 0xDC0A, 0x0308, 0}, {0xD80
0, 0xDC02}}; | |
1108 UChar target[][4] = | |
1109 {{0xD800, 0xDC00, 0}, {0xD800, 0xDC0A, 0x0308, 0}, {'A', 0, 0}, {0xD80
0, 0xDC03}}; | |
1110 int count = 0; | |
1111 uint8_t enresult[20], myresult[20]; | |
1112 int enlen, mylen; | |
1113 | |
1114 /* tests for open rules with surrogate rules */ | |
1115 rlen = u_unescape(str, rules, len); | |
1116 | |
1117 enCollation = ucol_open("en_US", &status); | |
1118 if (U_FAILURE(status)) { | |
1119 log_err_status(status, "ERROR: in creation of collator :%s\n", | |
1120 myErrorName(status)); | |
1121 return; | |
1122 } | |
1123 myCollation = ucol_openRules(rules, rlen, UCOL_OFF, | |
1124 UCOL_TERTIARY,NULL, &status); | |
1125 if (U_FAILURE(status)) { | |
1126 ucol_close(enCollation); | |
1127 log_err("ERROR: in creation of rule based collator :%s\n", | |
1128 myErrorName(status)); | |
1129 return; | |
1130 } | |
1131 | |
1132 /* | |
1133 this test is to verify the supplementary sort key order in the english | |
1134 collator | |
1135 */ | |
1136 log_verbose("start of english collation supplementary characters test\n"); | |
1137 while (count < 2) { | |
1138 doTest(enCollation, source[count], target[count], UCOL_LESS); | |
1139 count ++; | |
1140 } | |
1141 doTest(enCollation, source[count], target[count], UCOL_GREATER); | |
1142 | |
1143 log_verbose("start of tailored collation supplementary characters test\n"); | |
1144 count = 0; | |
1145 /* tests getting collation elements for surrogates for tailored rules */ | |
1146 while (count < 4) { | |
1147 doTest(myCollation, source[count], target[count], UCOL_LESS); | |
1148 count ++; | |
1149 } | |
1150 | |
1151 /* tests that \uD800\uDC02 still has the same value, not changed */ | |
1152 enlen = ucol_getSortKey(enCollation, source[3], 2, enresult, 20); | |
1153 mylen = ucol_getSortKey(myCollation, source[3], 2, myresult, 20); | |
1154 if (enlen != mylen || | |
1155 uprv_memcmp(enresult, myresult, enlen) != 0) { | |
1156 log_verbose("Failed : non-tailored supplementary characters should have
the same value\n"); | |
1157 } | |
1158 | |
1159 ucol_close(enCollation); | |
1160 ucol_close(myCollation); | |
1161 enCollation = NULL; | |
1162 myCollation = NULL; | |
1163 } | |
1164 | |
1165 /* | |
1166 *### TODO: Add more invalid rules to test all different scenarios. | |
1167 * | |
1168 */ | |
1169 static void | |
1170 TestInvalidRules(){ | |
1171 #define MAX_ERROR_STATES 2 | |
1172 | |
1173 static const char* rulesArr[MAX_ERROR_STATES] = { | |
1174 "& C < ch, cH, Ch[this should fail]<d", | |
1175 "& C < ch, cH, & Ch[variable top]" | |
1176 }; | |
1177 static const char* preContextArr[MAX_ERROR_STATES] = { | |
1178 " C < ch, cH, Ch", | |
1179 "& C < ch, cH", | |
1180 | |
1181 }; | |
1182 static const char* postContextArr[MAX_ERROR_STATES] = { | |
1183 "[this should fa", | |
1184 ", & Ch[variable" | |
1185 }; | |
1186 int i; | |
1187 | |
1188 for(i = 0;i<MAX_ERROR_STATES;i++){ | |
1189 UChar rules[1000] = { '\0' }; | |
1190 UChar preContextExp[1000] = { '\0' }; | |
1191 UChar postContextExp[1000] = { '\0' }; | |
1192 UParseError parseError; | |
1193 UErrorCode status = U_ZERO_ERROR; | |
1194 UCollator* coll=0; | |
1195 u_charsToUChars(rulesArr[i],rules,uprv_strlen(rulesArr[i])+1); | |
1196 u_charsToUChars(preContextArr[i],preContextExp,uprv_strlen(preContextArr
[i])+1); | |
1197 u_charsToUChars(postContextArr[i],postContextExp,uprv_strlen(postContext
Arr[i])+1); | |
1198 /* clean up stuff in parseError */ | |
1199 u_memset(parseError.preContext,0x0000,U_PARSE_CONTEXT_LEN); | |
1200 u_memset(parseError.postContext,0x0000,U_PARSE_CONTEXT_LEN); | |
1201 /* open the rules and test */ | |
1202 coll = ucol_openRules(rules,u_strlen(rules),UCOL_OFF,UCOL_DEFAULT_STRENG
TH,&parseError,&status); | |
1203 (void)coll; /* Suppress set but not used warning. */ | |
1204 if(u_strcmp(parseError.preContext,preContextExp)!=0){ | |
1205 log_err_status(status, "preContext in UParseError for ucol_openRules
does not match: \"%s\"\n", | |
1206 aescstrdup(parseError.preContext, -1)); | |
1207 } | |
1208 if(u_strcmp(parseError.postContext,postContextExp)!=0){ | |
1209 log_err_status(status, "postContext in UParseError for ucol_openRule
s does not match: \"%s\"\n", | |
1210 aescstrdup(parseError.postContext, -1)); | |
1211 } | |
1212 } | |
1213 } | |
1214 | |
1215 static void | |
1216 TestJitterbug1098(){ | |
1217 UChar rule[1000]; | |
1218 UCollator* c1 = NULL; | |
1219 UErrorCode status = U_ZERO_ERROR; | |
1220 UParseError parseError; | |
1221 char preContext[200]={0}; | |
1222 char postContext[200]={0}; | |
1223 int i=0; | |
1224 const char* rules[] = { | |
1225 "&''<\\\\", | |
1226 "&\\'<\\\\", | |
1227 "&\\\"<'\\'", | |
1228 "&'\"'<\\'", | |
1229 NULL | |
1230 | |
1231 }; | |
1232 const UCollationResult results1098[] = { | |
1233 UCOL_LESS, | |
1234 UCOL_LESS, | |
1235 UCOL_LESS, | |
1236 UCOL_LESS, | |
1237 }; | |
1238 const UChar input[][2]= { | |
1239 {0x0027,0x005c}, | |
1240 {0x0027,0x005c}, | |
1241 {0x0022,0x005c}, | |
1242 {0x0022,0x0027}, | |
1243 }; | |
1244 UChar X[2] ={0}; | |
1245 UChar Y[2] ={0}; | |
1246 u_memset(parseError.preContext,0x0000,U_PARSE_CONTEXT_LEN); | |
1247 u_memset(parseError.postContext,0x0000,U_PARSE_CONTEXT_LEN); | |
1248 for(;rules[i]!=0;i++){ | |
1249 u_uastrcpy(rule, rules[i]); | |
1250 c1 = ucol_openRules(rule, u_strlen(rule), UCOL_OFF, UCOL_DEFAULT_STRENGT
H, &parseError, &status); | |
1251 if(U_FAILURE(status)){ | |
1252 log_err_status(status, "Could not parse the rules syntax. Error: %s\
n", u_errorName(status)); | |
1253 | |
1254 if (status == U_PARSE_ERROR) { | |
1255 u_UCharsToChars(parseError.preContext,preContext,20); | |
1256 u_UCharsToChars(parseError.postContext,postContext,20); | |
1257 log_verbose("\n\tPre-Context: %s \n\tPost-Context:%s \n",preCont
ext,postContext); | |
1258 } | |
1259 | |
1260 return; | |
1261 } | |
1262 X[0] = input[i][0]; | |
1263 Y[0] = input[i][1]; | |
1264 doTest(c1,X,Y,results1098[i]); | |
1265 ucol_close(c1); | |
1266 } | |
1267 } | |
1268 | |
1269 static void | |
1270 TestFCDCrash(void) { | |
1271 static const char *test[] = { | |
1272 "Gr\\u00F6\\u00DFe", | |
1273 "Grossist" | |
1274 }; | |
1275 | |
1276 UErrorCode status = U_ZERO_ERROR; | |
1277 UCollator *coll = ucol_open("es", &status); | |
1278 if(U_FAILURE(status)) { | |
1279 log_err_status(status, "Couldn't open collator -> %s\n", u_errorName(sta
tus)); | |
1280 return; | |
1281 } | |
1282 ucol_close(coll); | |
1283 coll = NULL; | |
1284 ctest_resetICU(); | |
1285 coll = ucol_open("de_DE", &status); | |
1286 if(U_FAILURE(status)) { | |
1287 log_err_status(status, "Couldn't open collator -> %s\n", u_errorName(sta
tus)); | |
1288 return; | |
1289 } | |
1290 ucol_setAttribute(coll, UCOL_NORMALIZATION_MODE, UCOL_ON, &status); | |
1291 genericOrderingTest(coll, test, 2); | |
1292 ucol_close(coll); | |
1293 } | |
1294 | |
1295 /*static UBool | |
1296 find(UEnumeration* list, const char* str, UErrorCode* status){ | |
1297 const char* value = NULL; | |
1298 int32_t length=0; | |
1299 if(U_FAILURE(*status)){ | |
1300 return FALSE; | |
1301 } | |
1302 uenum_reset(list, status); | |
1303 while( (value= uenum_next(list, &length, status))!=NULL){ | |
1304 if(strcmp(value, str)==0){ | |
1305 return TRUE; | |
1306 } | |
1307 } | |
1308 return FALSE; | |
1309 }*/ | |
1310 | |
1311 static void TestJ5298(void) | |
1312 { | |
1313 UErrorCode status = U_ZERO_ERROR; | |
1314 char input[256], output[256]; | |
1315 UBool isAvailable; | |
1316 int32_t i = 0; | |
1317 UEnumeration* values = NULL; | |
1318 const char *keywordValue = NULL; | |
1319 log_verbose("Number of collator locales returned : %i \n", ucol_countAvailab
le()); | |
1320 values = ucol_getKeywordValues("collation", &status); | |
1321 while ((keywordValue = uenum_next(values, NULL, &status)) != NULL) { | |
1322 if (strncmp(keywordValue, "private-", 8) == 0) { | |
1323 log_err("ucol_getKeywordValues() returns private collation keyword:
%s\n", keywordValue); | |
1324 } | |
1325 } | |
1326 for (i = 0; i < ucol_countAvailable(); i++) { | |
1327 uenum_reset(values, &status); | |
1328 while ((keywordValue = uenum_next(values, NULL, &status)) != NULL) { | |
1329 strcpy(input, ucol_getAvailable(i)); | |
1330 if (strcmp(keywordValue, "standard") != 0) { | |
1331 strcat(input, "@collation="); | |
1332 strcat(input, keywordValue); | |
1333 } | |
1334 | |
1335 ucol_getFunctionalEquivalent(output, 256, "collation", input, &isAva
ilable, &status); | |
1336 if (strcmp(input, output) == 0) { /* Unique locale, print it out */ | |
1337 log_verbose("%s, \n", output); | |
1338 } | |
1339 } | |
1340 } | |
1341 uenum_close(values); | |
1342 log_verbose("\n"); | |
1343 } | |
1344 #endif /* #if !UCONFIG_NO_COLLATION */ | |
OLD | NEW |