]>
Commit | Line | Data |
---|---|---|
b75a7d8f A |
1 | /******************************************************************** |
2 | * COPYRIGHT: | |
374ca955 | 3 | * Copyright (c) 1997-2004, International Business Machines Corporation and |
b75a7d8f A |
4 | * others. All Rights Reserved. |
5 | ********************************************************************/ | |
6 | /******************************************************************************** | |
7 | * | |
8 | * File CAPITEST.C | |
9 | * | |
10 | * Modification History: | |
11 | * Name Description | |
12 | * Madhu Katragadda Ported for C API | |
13 | ********************************************************************************* | |
14 | *//* C API TEST For COLLATOR */ | |
15 | ||
16 | #include "unicode/utypes.h" | |
17 | ||
18 | #if !UCONFIG_NO_COLLATION | |
19 | ||
20 | #include <stdio.h> | |
21 | #include <stdlib.h> | |
22 | #include <string.h> | |
23 | #include "unicode/uloc.h" | |
24 | #include "unicode/ustring.h" | |
25 | #include "unicode/ures.h" | |
26 | #include "unicode/ucoleitr.h" | |
27 | #include "cintltst.h" | |
28 | #include "capitst.h" | |
29 | #include "ccolltst.h" | |
374ca955 | 30 | #include "putilimp.h" |
b75a7d8f A |
31 | |
32 | static void TestAttribute(void); | |
374ca955 | 33 | int TestBufferSize(); /* defined in "colutil.c" */ |
b75a7d8f A |
34 | |
35 | ||
374ca955 A |
36 | |
37 | ||
b75a7d8f A |
38 | /* next two function is modified from "i18n/ucol.cpp" to avoid include "ucol_imp.h" */ |
39 | static void uprv_appendByteToHexString(char *dst, uint8_t val) { | |
40 | uint32_t len = (uint32_t)strlen(dst); | |
41 | sprintf(dst+len, "%02X", val); | |
42 | } | |
43 | ||
44 | static char* U_EXPORT2 ucol_sortKeyToString(const UCollator *coll, const uint8_t *sortkey, char *buffer, uint32_t *len) { | |
45 | int32_t strength = UCOL_PRIMARY; | |
46 | uint32_t res_size = 0; | |
47 | UBool doneCase = FALSE; | |
48 | ||
49 | char *current = buffer; | |
50 | const uint8_t *currentSk = sortkey; | |
51 | ||
52 | UErrorCode error_code = U_ZERO_ERROR; | |
53 | ||
54 | strcpy(current, "["); | |
55 | ||
56 | while(strength <= UCOL_QUATERNARY && strength <= ucol_getAttribute(coll,UCOL_STRENGTH, &error_code)) { | |
57 | if(U_FAILURE(error_code)) { | |
58 | log_err("ucol_getAttribute returned error: %s\n", u_errorName(error_code)); | |
59 | } | |
60 | if(strength > UCOL_PRIMARY) { | |
61 | strcat(current, " . "); | |
62 | } | |
63 | while(*currentSk != 0x01 && *currentSk != 0x00) { /* print a level */ | |
64 | uprv_appendByteToHexString(current, *currentSk++); | |
65 | strcat(current, " "); | |
66 | } | |
67 | if(ucol_getAttribute(coll,UCOL_CASE_LEVEL, &error_code) == UCOL_ON && strength == UCOL_SECONDARY && doneCase == FALSE) { | |
68 | doneCase = TRUE; | |
69 | } else if(ucol_getAttribute(coll,UCOL_CASE_LEVEL, &error_code) == UCOL_OFF || doneCase == TRUE || strength != UCOL_SECONDARY) { | |
70 | strength ++; | |
71 | } | |
72 | if(U_FAILURE(error_code)) { | |
73 | log_err("ucol_getAttribute returned error: %s\n", u_errorName(error_code)); | |
74 | } | |
75 | uprv_appendByteToHexString(current, *currentSk++); /* This should print '01' */ | |
76 | if(strength == UCOL_QUATERNARY && ucol_getAttribute(coll,UCOL_ALTERNATE_HANDLING, &error_code) == UCOL_NON_IGNORABLE) { | |
77 | break; | |
78 | } | |
79 | } | |
80 | ||
81 | if(ucol_getAttribute(coll,UCOL_STRENGTH, &error_code) == UCOL_IDENTICAL) { | |
82 | strcat(current, " . "); | |
83 | while(*currentSk != 0) { | |
84 | uprv_appendByteToHexString(current, *currentSk++); | |
85 | strcat(current, " "); | |
86 | } | |
87 | ||
88 | uprv_appendByteToHexString(current, *currentSk++); | |
89 | } | |
90 | if(U_FAILURE(error_code)) { | |
91 | log_err("ucol_getAttribute returned error: %s\n", u_errorName(error_code)); | |
92 | } | |
93 | strcat(current, "]"); | |
94 | ||
95 | if(res_size > *len) { | |
96 | return NULL; | |
97 | } | |
98 | ||
99 | return buffer; | |
100 | } | |
101 | /* end of avoid include "ucol_imp.h" */ | |
102 | ||
103 | ||
104 | void addCollAPITest(TestNode** root) | |
105 | { | |
106 | /* WEIVTODO: return tests here */ | |
107 | addTest(root, &TestProperty, "tscoll/capitst/TestProperty"); | |
108 | addTest(root, &TestRuleBasedColl, "tscoll/capitst/TestRuleBasedColl"); | |
109 | addTest(root, &TestCompare, "tscoll/capitst/TestCompare"); | |
110 | addTest(root, &TestSortKey, "tscoll/capitst/TestSortKey"); | |
111 | addTest(root, &TestHashCode, "tscoll/capitst/TestHashCode"); | |
112 | addTest(root, &TestElemIter, "tscoll/capitst/TestElemIter"); | |
113 | addTest(root, &TestGetAll, "tscoll/capitst/TestGetAll"); | |
114 | /*addTest(root, &TestGetDefaultRules, "tscoll/capitst/TestGetDefaultRules");*/ | |
115 | addTest(root, &TestDecomposition, "tscoll/capitst/TestDecomposition"); | |
116 | addTest(root, &TestSafeClone, "tscoll/capitst/TestSafeClone"); | |
117 | addTest(root, &TestGetSetAttr, "tscoll/capitst/TestGetSetAttr"); | |
118 | addTest(root, &TestBounds, "tscoll/capitst/TestBounds"); | |
119 | addTest(root, &TestGetLocale, "tscoll/capitst/TestGetLocale"); | |
120 | addTest(root, &TestSortKeyBufferOverrun, "tscoll/capitst/TestSortKeyBufferOverrun"); | |
121 | addTest(root, &TestAttribute, "tscoll/capitst/TestAttribute"); | |
122 | addTest(root, &TestGetTailoredSet, "tscoll/capitst/TestGetTailoredSet"); | |
123 | addTest(root, &TestMergeSortKeys, "tscoll/capitst/TestMergeSortKeys"); | |
374ca955 A |
124 | addTest(root, &TestShortString, "tscoll/capitst/TestShortString"); |
125 | addTest(root, &TestGetContractionsAndUnsafes, "tscoll/capitst/TestGetContractionsAndUnsafes"); | |
b75a7d8f A |
126 | } |
127 | ||
128 | void TestGetSetAttr(void) { | |
129 | UErrorCode status = U_ZERO_ERROR; | |
130 | UCollator *coll = ucol_open(NULL, &status); | |
131 | struct attrTest { | |
132 | UColAttribute att; | |
133 | UColAttributeValue val[5]; | |
134 | uint32_t valueSize; | |
135 | UColAttributeValue nonValue; | |
136 | } attrs[] = { | |
137 | {UCOL_FRENCH_COLLATION, {UCOL_ON, UCOL_OFF}, 2, UCOL_SHIFTED}, | |
138 | {UCOL_ALTERNATE_HANDLING, {UCOL_NON_IGNORABLE, UCOL_SHIFTED}, 2, UCOL_OFF},/* attribute for handling variable elements*/ | |
139 | {UCOL_CASE_FIRST, {UCOL_OFF, UCOL_LOWER_FIRST, UCOL_UPPER_FIRST}, 3, UCOL_SHIFTED},/* who goes first, lower case or uppercase */ | |
140 | {UCOL_CASE_LEVEL, {UCOL_ON, UCOL_OFF}, 2, UCOL_SHIFTED},/* do we have an extra case level */ | |
141 | {UCOL_NORMALIZATION_MODE, {UCOL_ON, UCOL_OFF}, 2, UCOL_SHIFTED},/* attribute for normalization */ | |
142 | {UCOL_DECOMPOSITION_MODE, {UCOL_ON, UCOL_OFF}, 2, UCOL_SHIFTED}, | |
143 | {UCOL_STRENGTH, {UCOL_PRIMARY, UCOL_SECONDARY, UCOL_TERTIARY, UCOL_QUATERNARY, UCOL_IDENTICAL}, 5, UCOL_SHIFTED},/* attribute for strength */ | |
144 | {UCOL_HIRAGANA_QUATERNARY_MODE, {UCOL_ON, UCOL_OFF}, 2, UCOL_SHIFTED},/* when turned on, this attribute */ | |
145 | }; | |
146 | UColAttribute currAttr; | |
147 | UColAttributeValue value; | |
148 | uint32_t i = 0, j = 0; | |
149 | ||
150 | for(i = 0; i<sizeof(attrs)/sizeof(attrs[0]); i++) { | |
151 | currAttr = attrs[i].att; | |
152 | ucol_setAttribute(coll, currAttr, UCOL_DEFAULT, &status); | |
153 | if(U_FAILURE(status)) { | |
154 | log_err("ucol_setAttribute with the default value returned error: %s\n", u_errorName(status)); | |
155 | break; | |
156 | } | |
157 | value = ucol_getAttribute(coll, currAttr, &status); | |
158 | if(U_FAILURE(status)) { | |
159 | log_err("ucol_getAttribute returned error: %s\n", u_errorName(status)); | |
160 | break; | |
161 | } | |
162 | for(j = 0; j<attrs[i].valueSize; j++) { | |
163 | ucol_setAttribute(coll, currAttr, attrs[i].val[j], &status); | |
164 | if(U_FAILURE(status)) { | |
165 | log_err("ucol_setAttribute with the value %i returned error: %s\n", attrs[i].val[j], u_errorName(status)); | |
166 | break; | |
167 | } | |
168 | } | |
169 | status = U_ZERO_ERROR; | |
170 | ucol_setAttribute(coll, currAttr, attrs[i].nonValue, &status); | |
171 | if(U_SUCCESS(status)) { | |
172 | log_err("ucol_setAttribute with the bad value didn't return an error\n"); | |
173 | break; | |
174 | } | |
175 | status = U_ZERO_ERROR; | |
176 | ||
177 | ucol_setAttribute(coll, currAttr, value, &status); | |
178 | if(U_FAILURE(status)) { | |
179 | log_err("ucol_setAttribute with the default valuereturned error: %s\n", u_errorName(status)); | |
180 | break; | |
181 | } | |
182 | ||
183 | } | |
184 | status = U_ZERO_ERROR; | |
185 | value = ucol_getAttribute(coll, UCOL_ATTRIBUTE_COUNT, &status); | |
186 | if(U_SUCCESS(status)) { | |
187 | log_err("ucol_getAttribute for UCOL_ATTRIBUTE_COUNT didn't return an error\n"); | |
188 | } | |
189 | status = U_ZERO_ERROR; | |
190 | ucol_setAttribute(coll, UCOL_ATTRIBUTE_COUNT, UCOL_DEFAULT, &status); | |
191 | if(U_SUCCESS(status)) { | |
192 | log_err("ucol_setAttribute for UCOL_ATTRIBUTE_COUNT didn't return an error\n"); | |
193 | } | |
194 | status = U_ZERO_ERROR; | |
195 | ucol_close(coll); | |
196 | } | |
197 | ||
198 | ||
199 | static void doAssert(int condition, const char *message) | |
200 | { | |
201 | if (condition==0) { | |
202 | log_err("ERROR : %s\n", message); | |
203 | } | |
204 | } | |
205 | ||
206 | #if 0 | |
207 | /* We don't have default rules, at least not in the previous sense */ | |
208 | void TestGetDefaultRules(){ | |
209 | uint32_t size=0; | |
210 | UErrorCode status=U_ZERO_ERROR; | |
211 | UCollator *coll=NULL; | |
212 | int32_t len1 = 0, len2=0; | |
213 | uint8_t *binColData = NULL; | |
214 | ||
215 | UResourceBundle *res = NULL; | |
216 | UResourceBundle *binColl = NULL; | |
217 | uint8_t *binResult = NULL; | |
218 | ||
219 | ||
220 | const UChar * defaultRulesArray=ucol_getDefaultRulesArray(&size); | |
221 | log_verbose("Test the function ucol_getDefaultRulesArray()\n"); | |
222 | ||
223 | coll = ucol_openRules(defaultRulesArray, size, UCOL_ON, UCOL_PRIMARY, &status); | |
224 | if(U_SUCCESS(status) && coll !=NULL) { | |
225 | binColData = (uint8_t*)ucol_cloneRuleData(coll, &len1, &status); | |
226 | ||
227 | } | |
228 | ||
229 | ||
230 | status=U_ZERO_ERROR; | |
231 | res=ures_open(NULL, "root", &status); | |
232 | if(U_FAILURE(status)){ | |
233 | log_err("ERROR: Failed to get resource for \"root Locale\" with %s", myErrorName(status)); | |
234 | return; | |
235 | } | |
236 | binColl=ures_getByKey(res, "%%Collation", binColl, &status); | |
237 | if(U_SUCCESS(status)){ | |
238 | binResult=(uint8_t*)ures_getBinary(binColl, &len2, &status); | |
239 | if(U_FAILURE(status)){ | |
240 | log_err("ERROR: ures_getBinary() failed\n"); | |
241 | } | |
242 | }else{ | |
243 | log_err("ERROR: ures_getByKey(locale(default), %%Collation) failed"); | |
244 | } | |
245 | ||
246 | ||
247 | if(len1 != len2){ | |
248 | log_err("Error: ucol_getDefaultRulesArray() failed to return the correct length.\n"); | |
249 | } | |
250 | if(memcmp(binColData, binResult, len1) != 0){ | |
251 | log_err("Error: ucol_getDefaultRulesArray() failed\n"); | |
252 | } | |
253 | ||
254 | free(binColData); | |
255 | ures_close(binColl); | |
256 | ures_close(res); | |
257 | ucol_close(coll); | |
258 | ||
259 | } | |
260 | #endif | |
261 | ||
262 | /* Collator Properties | |
263 | ucol_open, ucol_strcoll, getStrength/setStrength | |
264 | getDecomposition/setDecomposition, getDisplayName*/ | |
265 | void TestProperty() | |
266 | { | |
267 | UCollator *col, *ruled; | |
268 | UChar *disName; | |
269 | int32_t len = 0, i = 0; | |
270 | UChar *source, *target; | |
271 | int32_t tempLength; | |
272 | UErrorCode status = U_ZERO_ERROR; | |
273 | /* | |
274 | All the collations have the same version in an ICU | |
275 | version. | |
276 | ICU 2.0 currVersionArray = {0x18, 0xC0, 0x02, 0x02}; | |
277 | ICU 2.1 currVersionArray = {0x19, 0x00, 0x03, 0x03}; | |
278 | ICU 2.2 currVersionArray = {0x21, 0x40, 0x04, 0x04}; | |
279 | ICU 2.4 currVersionArray = {0x21, 0x40, 0x04, 0x04}; | |
374ca955 | 280 | ICU 2.6 currVersionArray = {0x21, 0x40, 0x03, 0x03}; |
b75a7d8f | 281 | */ |
374ca955 A |
282 | UVersionInfo currVersionArray = {0x29, 0x80, 0x00, 0x04}; |
283 | UVersionInfo currUCAVersionArray = {4, 0, 0, 0}; | |
284 | UVersionInfo versionArray = {0, 0, 0, 0}; | |
285 | UVersionInfo versionUCAArray = {0, 0, 0, 0}; | |
b75a7d8f A |
286 | |
287 | log_verbose("The property tests begin : \n"); | |
288 | log_verbose("Test ucol_strcoll : \n"); | |
289 | col = ucol_open("en_US", &status); | |
290 | if (U_FAILURE(status)) { | |
291 | log_err("Default Collator creation failed.: %s\n", myErrorName(status)); | |
292 | return; | |
293 | } | |
294 | ||
295 | ucol_getVersion(col, versionArray); | |
296 | for (i=0; i<4; ++i) { | |
297 | if (versionArray[i] != currVersionArray[i]) { | |
298 | log_err("Testing ucol_getVersion() - unexpected result: %hu.%hu.%hu.%hu\n", | |
299 | versionArray[0], versionArray[1], versionArray[2], versionArray[3]); | |
300 | break; | |
301 | } | |
302 | } | |
303 | ||
374ca955 A |
304 | ucol_getUCAVersion(col, versionUCAArray); |
305 | for (i=0; i<4; ++i) { | |
306 | if (versionUCAArray[i] != currUCAVersionArray[i]) { | |
307 | log_err("Testing ucol_getUCAVersion() - unexpected result: %hu.%hu.%hu.%hu\n", | |
308 | versionUCAArray[0], versionUCAArray[1], versionUCAArray[2], versionUCAArray[3]); | |
309 | break; | |
310 | } | |
311 | } | |
312 | ||
b75a7d8f A |
313 | source=(UChar*)malloc(sizeof(UChar) * 12); |
314 | target=(UChar*)malloc(sizeof(UChar) * 12); | |
315 | ||
316 | ||
317 | u_uastrcpy(source, "ab"); | |
318 | u_uastrcpy(target, "abc"); | |
319 | ||
320 | doAssert((ucol_strcoll(col, source, u_strlen(source), target, u_strlen(target)) == UCOL_LESS), "ab < abc comparison failed"); | |
321 | ||
322 | u_uastrcpy(source, "ab"); | |
323 | u_uastrcpy(target, "AB"); | |
324 | ||
325 | doAssert((ucol_strcoll(col, source, u_strlen(source), target, u_strlen(target)) == UCOL_LESS), "ab < AB comparison failed"); | |
326 | /* u_uastrcpy(source, "black-bird"); | |
327 | u_uastrcpy(target, "blackbird"); */ | |
328 | u_uastrcpy(target, "black-bird"); | |
329 | u_uastrcpy(source, "blackbird"); | |
330 | ||
331 | doAssert((ucol_strcoll(col, source, u_strlen(source), target, u_strlen(target)) == UCOL_GREATER), | |
332 | "black-bird > blackbird comparison failed"); | |
333 | u_uastrcpy(source, "black bird"); | |
334 | u_uastrcpy(target, "black-bird"); | |
335 | doAssert((ucol_strcoll(col, source, u_strlen(source), target, u_strlen(target)) == UCOL_LESS), | |
336 | "black bird < black-bird comparison failed"); | |
337 | u_uastrcpy(source, "Hello"); | |
338 | u_uastrcpy(target, "hello"); | |
339 | ||
340 | doAssert((ucol_strcoll(col, source, u_strlen(source), target, u_strlen(target)) == UCOL_GREATER), | |
341 | "Hello > hello comparison failed"); | |
342 | free(source); | |
343 | free(target); | |
344 | log_verbose("Test ucol_strcoll ends.\n"); | |
345 | ||
346 | log_verbose("testing ucol_getStrength() method ...\n"); | |
347 | doAssert( (ucol_getStrength(col) == UCOL_TERTIARY), "collation object has the wrong strength"); | |
348 | doAssert( (ucol_getStrength(col) != UCOL_PRIMARY), "collation object's strength is primary difference"); | |
349 | ||
350 | log_verbose("testing ucol_setStrength() method ...\n"); | |
351 | ucol_setStrength(col, UCOL_SECONDARY); | |
352 | doAssert( (ucol_getStrength(col) != UCOL_TERTIARY), "collation object's strength is secondary difference"); | |
353 | doAssert( (ucol_getStrength(col) != UCOL_PRIMARY), "collation object's strength is primary difference"); | |
354 | doAssert( (ucol_getStrength(col) == UCOL_SECONDARY), "collation object has the wrong strength"); | |
355 | ||
356 | ||
357 | log_verbose("Get display name for the default collation in German : \n"); | |
358 | ||
359 | len=ucol_getDisplayName("en_US", "de_DE", NULL, 0, &status); | |
360 | if(status==U_BUFFER_OVERFLOW_ERROR){ | |
361 | status=U_ZERO_ERROR; | |
362 | disName=(UChar*)malloc(sizeof(UChar) * (len+1)); | |
363 | ucol_getDisplayName("en_US", "de_DE", disName, len+1, &status); | |
364 | log_verbose("the display name for default collation in german: %s\n", austrdup(disName) ); | |
365 | free(disName); | |
366 | } | |
367 | if(U_FAILURE(status)){ | |
368 | log_err("ERROR: in getDisplayName: %s\n", myErrorName(status)); | |
369 | return; | |
370 | } | |
371 | log_verbose("Default collation getDisplayName ended.\n"); | |
372 | ||
373 | ruled = ucol_open("da_DK", &status); | |
374 | log_verbose("ucol_getRules() testing ...\n"); | |
375 | ucol_getRules(ruled, &tempLength); | |
376 | doAssert( tempLength != 0, "getRules() result incorrect" ); | |
377 | log_verbose("getRules tests end.\n"); | |
378 | { | |
379 | UChar *buffer = (UChar *)malloc(200000*sizeof(UChar)); | |
380 | int32_t bufLen = 200000; | |
381 | buffer[0] = '\0'; | |
382 | log_verbose("ucol_getRulesEx() testing ...\n"); | |
383 | tempLength = ucol_getRulesEx(col,UCOL_TAILORING_ONLY,buffer,bufLen ); | |
374ca955 | 384 | doAssert( tempLength == 0x0a, "getRulesEx() result incorrect" ); |
b75a7d8f A |
385 | log_verbose("getRules tests end.\n"); |
386 | ||
387 | log_verbose("ucol_getRulesEx() testing ...\n"); | |
388 | tempLength=ucol_getRulesEx(col,UCOL_FULL_RULES,buffer,bufLen ); | |
389 | doAssert( tempLength != 0, "getRulesEx() result incorrect" ); | |
390 | log_verbose("getRules tests end.\n"); | |
391 | free(buffer); | |
392 | } | |
393 | ucol_close(ruled); | |
394 | ucol_close(col); | |
395 | ||
396 | log_verbose("open an collator for french locale"); | |
397 | col = ucol_open("fr_FR", &status); | |
398 | if (U_FAILURE(status)) { | |
399 | log_err("ERROR: Creating French collation failed.: %s\n", myErrorName(status)); | |
400 | return; | |
401 | } | |
402 | ucol_setStrength(col, UCOL_PRIMARY); | |
403 | log_verbose("testing ucol_getStrength() method again ...\n"); | |
404 | doAssert( (ucol_getStrength(col) != UCOL_TERTIARY), "collation object has the wrong strength"); | |
405 | doAssert( (ucol_getStrength(col) == UCOL_PRIMARY), "collation object's strength is not primary difference"); | |
406 | ||
407 | log_verbose("testing French ucol_setStrength() method ...\n"); | |
408 | ucol_setStrength(col, UCOL_TERTIARY); | |
409 | doAssert( (ucol_getStrength(col) == UCOL_TERTIARY), "collation object's strength is not tertiary difference"); | |
410 | doAssert( (ucol_getStrength(col) != UCOL_PRIMARY), "collation object's strength is primary difference"); | |
411 | doAssert( (ucol_getStrength(col) != UCOL_SECONDARY), "collation object's strength is secondary difference"); | |
412 | ucol_close(col); | |
413 | ||
414 | log_verbose("Get display name for the french collation in english : \n"); | |
415 | len=ucol_getDisplayName("fr_FR", "en_US", NULL, 0, &status); | |
416 | if(status==U_BUFFER_OVERFLOW_ERROR){ | |
417 | status=U_ZERO_ERROR; | |
418 | disName=(UChar*)malloc(sizeof(UChar) * (len+1)); | |
419 | ucol_getDisplayName("fr_FR", "en_US", disName, len+1, &status); | |
420 | log_verbose("the display name for french collation in english: %s\n", austrdup(disName) ); | |
421 | free(disName); | |
422 | } | |
423 | if(U_FAILURE(status)){ | |
424 | log_err("ERROR: in getDisplayName: %s\n", myErrorName(status)); | |
425 | return; | |
426 | } | |
427 | log_verbose("Default collation getDisplayName ended.\n"); | |
428 | ||
429 | } | |
430 | ||
431 | /* Test RuleBasedCollator and getRules*/ | |
432 | void TestRuleBasedColl() | |
433 | { | |
434 | UCollator *col1, *col2, *col3, *col4; | |
435 | UCollationElements *iter1, *iter2; | |
436 | UChar ruleset1[60]; | |
437 | UChar ruleset2[50]; | |
438 | UChar teststr[10]; | |
439 | UChar teststr2[10]; | |
440 | const UChar *rule1, *rule2, *rule3, *rule4; | |
441 | int32_t tempLength; | |
442 | UErrorCode status = U_ZERO_ERROR; | |
443 | u_uastrcpy(ruleset1, "&9 < a, A < b, B < c, C; ch, cH, Ch, CH < d, D, e, E"); | |
444 | u_uastrcpy(ruleset2, "&9 < a, A < b, B < c, C < d, D, e, E"); | |
445 | ||
374ca955 | 446 | |
b75a7d8f A |
447 | col1 = ucol_openRules(ruleset1, u_strlen(ruleset1), UCOL_DEFAULT, UCOL_DEFAULT_STRENGTH, NULL,&status); |
448 | if (U_FAILURE(status)) { | |
449 | log_err("RuleBased Collator creation failed.: %s\n", myErrorName(status)); | |
450 | return; | |
451 | } | |
452 | else | |
453 | log_verbose("PASS: RuleBased Collator creation passed\n"); | |
454 | ||
455 | status = U_ZERO_ERROR; | |
456 | col2 = ucol_openRules(ruleset2, u_strlen(ruleset2), UCOL_DEFAULT, UCOL_DEFAULT_STRENGTH, NULL, &status); | |
457 | if (U_FAILURE(status)) { | |
458 | log_err("RuleBased Collator creation failed.: %s\n", myErrorName(status)); | |
459 | return; | |
460 | } | |
461 | else | |
462 | log_verbose("PASS: RuleBased Collator creation passed\n"); | |
463 | ||
464 | ||
465 | status = U_ZERO_ERROR; | |
466 | col3= ucol_open(NULL, &status); | |
467 | if (U_FAILURE(status)) { | |
468 | log_err("Default Collator creation failed.: %s\n", myErrorName(status)); | |
469 | return; | |
470 | } | |
471 | else | |
472 | log_verbose("PASS: Default Collator creation passed\n"); | |
473 | ||
474 | rule1 = ucol_getRules(col1, &tempLength); | |
475 | rule2 = ucol_getRules(col2, &tempLength); | |
476 | rule3 = ucol_getRules(col3, &tempLength); | |
477 | ||
478 | doAssert((u_strcmp(rule1, rule2) != 0), "Default collator getRules failed"); | |
479 | doAssert((u_strcmp(rule2, rule3) != 0), "Default collator getRules failed"); | |
480 | doAssert((u_strcmp(rule1, rule3) != 0), "Default collator getRules failed"); | |
481 | ||
482 | col4=ucol_openRules(rule2, u_strlen(rule2), UCOL_DEFAULT, UCOL_DEFAULT_STRENGTH, NULL, &status); | |
483 | if (U_FAILURE(status)) { | |
484 | log_err("RuleBased Collator creation failed.: %s\n", myErrorName(status)); | |
485 | return; | |
486 | } | |
487 | rule4= ucol_getRules(col4, &tempLength); | |
488 | doAssert((u_strcmp(rule2, rule4) == 0), "Default collator getRules failed"); | |
489 | ||
490 | ucol_close(col1); | |
491 | ucol_close(col2); | |
492 | ucol_close(col3); | |
493 | ucol_close(col4); | |
494 | ||
495 | /* tests that modifier ! is always ignored */ | |
496 | u_uastrcpy(ruleset1, "!&a<b"); | |
497 | teststr[0] = 0x0e40; | |
498 | teststr[1] = 0x0e01; | |
499 | teststr[2] = 0x0e2d; | |
500 | col1 = ucol_openRules(ruleset1, u_strlen(ruleset1), UCOL_DEFAULT, UCOL_DEFAULT_STRENGTH, NULL, &status); | |
501 | if (U_FAILURE(status)) { | |
502 | log_err("RuleBased Collator creation failed.: %s\n", myErrorName(status)); | |
503 | return; | |
504 | } | |
505 | col2 = ucol_open("en_US", &status); | |
506 | if (U_FAILURE(status)) { | |
507 | log_err("en_US Collator creation failed.: %s\n", myErrorName(status)); | |
508 | return; | |
509 | } | |
510 | iter1 = ucol_openElements(col1, teststr, 3, &status); | |
511 | iter2 = ucol_openElements(col2, teststr, 3, &status); | |
512 | if(U_FAILURE(status)) { | |
513 | log_err("ERROR: CollationElement iterator creation failed.: %s\n", myErrorName(status)); | |
514 | return; | |
515 | } | |
516 | while (TRUE) { | |
517 | /* testing with en since thai has its own tailoring */ | |
518 | uint32_t ce = ucol_next(iter1, &status); | |
519 | uint32_t ce2 = ucol_next(iter2, &status); | |
374ca955 | 520 | if(U_FAILURE(status)) { |
b75a7d8f A |
521 | log_err("ERROR: CollationElement iterator creation failed.: %s\n", myErrorName(status)); |
522 | return; | |
374ca955 | 523 | } |
b75a7d8f A |
524 | if (ce2 != ce) { |
525 | log_err("! modifier test failed"); | |
526 | } | |
527 | if (ce == UCOL_NULLORDER) { | |
528 | break; | |
529 | } | |
530 | } | |
531 | ucol_closeElements(iter1); | |
532 | ucol_closeElements(iter2); | |
533 | ucol_close(col1); | |
534 | ucol_close(col2); | |
535 | /* test that we can start a rule without a & or < */ | |
536 | u_uastrcpy(ruleset1, "< z < a"); | |
537 | col1 = ucol_openRules(ruleset1, u_strlen(ruleset1), UCOL_DEFAULT, UCOL_DEFAULT_STRENGTH, NULL, &status); | |
538 | if (U_FAILURE(status)) { | |
539 | log_err("RuleBased Collator creation failed.: %s\n", myErrorName(status)); | |
540 | return; | |
541 | } | |
542 | u_uastrcpy(teststr, "z"); | |
543 | u_uastrcpy(teststr2, "a"); | |
544 | if (ucol_greaterOrEqual(col1, teststr, 1, teststr2, 1)) { | |
545 | log_err("Rule \"z < a\" fails"); | |
546 | } | |
547 | ucol_close(col1); | |
548 | } | |
549 | ||
550 | void TestCompare() | |
551 | { | |
552 | UErrorCode status = U_ZERO_ERROR; | |
553 | UCollator *col; | |
554 | UChar* test1; | |
555 | UChar* test2; | |
556 | ||
557 | log_verbose("The compare tests begin : \n"); | |
558 | status=U_ZERO_ERROR; | |
559 | col = ucol_open("en_US", &status); | |
560 | if(U_FAILURE(status)) { | |
561 | log_err("ucal_open() collation creation failed.: %s\n", myErrorName(status)); | |
562 | return; | |
563 | } | |
564 | test1=(UChar*)malloc(sizeof(UChar) * 6); | |
565 | test2=(UChar*)malloc(sizeof(UChar) * 6); | |
566 | u_uastrcpy(test1, "Abcda"); | |
567 | u_uastrcpy(test2, "abcda"); | |
568 | ||
569 | log_verbose("Use tertiary comparison level testing ....\n"); | |
570 | ||
571 | doAssert( (!ucol_equal(col, test1, u_strlen(test1), test2, u_strlen(test2))), "Result should be \"Abcda\" != \"abcda\" "); | |
572 | doAssert( (ucol_greater(col, test1, u_strlen(test1), test2, u_strlen(test2))), "Result should be \"Abcda\" >>> \"abcda\" "); | |
573 | doAssert( (ucol_greaterOrEqual(col, test1, u_strlen(test1), test2, u_strlen(test2))), "Result should be \"Abcda\" >>> \"abcda\""); | |
574 | ||
575 | ucol_setStrength(col, UCOL_SECONDARY); | |
576 | log_verbose("Use secondary comparison level testing ....\n"); | |
577 | ||
578 | doAssert( (ucol_equal(col, test1, u_strlen(test1), test2, u_strlen(test2) )), "Result should be \"Abcda\" == \"abcda\""); | |
579 | doAssert( (!ucol_greater(col, test1, u_strlen(test1), test2, u_strlen(test2))), "Result should be \"Abcda\" == \"abcda\""); | |
580 | doAssert( (ucol_greaterOrEqual(col, test1, u_strlen(test1), test2, u_strlen(test2) )), "Result should be \"Abcda\" == \"abcda\""); | |
581 | ||
582 | ucol_setStrength(col, UCOL_PRIMARY); | |
583 | log_verbose("Use primary comparison level testing ....\n"); | |
584 | ||
585 | doAssert( (ucol_equal(col, test1, u_strlen(test1), test2, u_strlen(test2))), "Result should be \"Abcda\" == \"abcda\""); | |
586 | doAssert( (!ucol_greater(col, test1, u_strlen(test1), test2, u_strlen(test2))), "Result should be \"Abcda\" == \"abcda\""); | |
587 | doAssert( (ucol_greaterOrEqual(col, test1, u_strlen(test1), test2, u_strlen(test2))), "Result should be \"Abcda\" == \"abcda\""); | |
588 | ||
589 | ||
590 | log_verbose("The compare tests end.\n"); | |
591 | ucol_close(col); | |
592 | free(test1); | |
593 | free(test2); | |
594 | ||
595 | } | |
596 | /* | |
597 | --------------------------------------------- | |
598 | tests decomposition setting | |
599 | */ | |
600 | void TestDecomposition() { | |
601 | UErrorCode status = U_ZERO_ERROR; | |
602 | UCollator *en_US, *el_GR, *vi_VN; | |
603 | en_US = ucol_open("en_US", &status); | |
604 | el_GR = ucol_open("el_GR", &status); | |
605 | vi_VN = ucol_open("vi_VN", &status); | |
606 | ||
607 | if (U_FAILURE(status)) { | |
608 | log_err("ERROR: collation creation failed.: %s\n", myErrorName(status)); | |
609 | return; | |
610 | } | |
611 | ||
612 | if (ucol_getAttribute(vi_VN, UCOL_NORMALIZATION_MODE, &status) != UCOL_ON || | |
613 | U_FAILURE(status)) | |
614 | { | |
615 | log_err("ERROR: vi_VN collation did not have cannonical decomposition for normalization!\n"); | |
616 | } | |
617 | ||
618 | status = U_ZERO_ERROR; | |
619 | if (ucol_getAttribute(el_GR, UCOL_NORMALIZATION_MODE, &status) != UCOL_ON || | |
620 | U_FAILURE(status)) | |
621 | { | |
622 | log_err("ERROR: el_GR collation did not have cannonical decomposition for normalization!\n"); | |
623 | } | |
624 | ||
625 | status = U_ZERO_ERROR; | |
626 | if (ucol_getAttribute(en_US, UCOL_NORMALIZATION_MODE, &status) != UCOL_OFF || | |
627 | U_FAILURE(status)) | |
628 | { | |
629 | log_err("ERROR: en_US collation had cannonical decomposition for normalization!\n"); | |
630 | } | |
631 | ||
632 | ucol_close(en_US); | |
633 | ucol_close(el_GR); | |
634 | ucol_close(vi_VN); | |
635 | } | |
636 | ||
637 | #define CLONETEST_COLLATOR_COUNT 3 | |
638 | ||
639 | void TestSafeClone() { | |
640 | UChar* test1; | |
641 | UChar* test2; | |
642 | UCollator * someCollators [CLONETEST_COLLATOR_COUNT]; | |
643 | UCollator * someClonedCollators [CLONETEST_COLLATOR_COUNT]; | |
644 | UCollator * col; | |
645 | UErrorCode err = U_ZERO_ERROR; | |
646 | int8_t testSize = 6; /* Leave this here to test buffer alingment in memory*/ | |
647 | uint8_t buffer [CLONETEST_COLLATOR_COUNT] [U_COL_SAFECLONE_BUFFERSIZE]; | |
648 | int32_t bufferSize = U_COL_SAFECLONE_BUFFERSIZE; | |
649 | int index; | |
650 | ||
651 | if (TestBufferSize()) { | |
652 | log_err("U_COL_SAFECLONE_BUFFERSIZE should be larger than sizeof(UCollator)\n"); | |
653 | return; | |
654 | } | |
655 | ||
656 | test1=(UChar*)malloc(sizeof(UChar) * testSize); | |
657 | test2=(UChar*)malloc(sizeof(UChar) * testSize); | |
658 | u_uastrcpy(test1, "abCda"); | |
659 | u_uastrcpy(test2, "abcda"); | |
660 | ||
661 | /* one default collator & two complex ones */ | |
662 | someCollators[0] = ucol_open("en_US", &err); | |
663 | someCollators[1] = ucol_open("ko", &err); | |
664 | someCollators[2] = ucol_open("ja_JP", &err); | |
665 | if(U_FAILURE(err)) { | |
666 | log_data_err("Couldn't open one or more collators\n"); | |
667 | return; | |
668 | } | |
669 | ||
670 | /* Check the various error & informational states: */ | |
671 | ||
672 | /* Null status - just returns NULL */ | |
673 | if (0 != ucol_safeClone(someCollators[0], buffer[0], &bufferSize, 0)) | |
674 | { | |
675 | log_err("FAIL: Cloned Collator failed to deal correctly with null status\n"); | |
676 | } | |
677 | /* error status - should return 0 & keep error the same */ | |
678 | err = U_MEMORY_ALLOCATION_ERROR; | |
679 | if (0 != ucol_safeClone(someCollators[0], buffer[0], &bufferSize, &err) || err != U_MEMORY_ALLOCATION_ERROR) | |
680 | { | |
681 | log_err("FAIL: Cloned Collator failed to deal correctly with incoming error status\n"); | |
682 | } | |
683 | err = U_ZERO_ERROR; | |
684 | ||
685 | /* Null buffer size pointer - just returns NULL & set error to U_ILLEGAL_ARGUMENT_ERROR*/ | |
686 | if (0 != ucol_safeClone(someCollators[0], buffer[0], 0, &err) || err != U_ILLEGAL_ARGUMENT_ERROR) | |
687 | { | |
688 | log_err("FAIL: Cloned Collator failed to deal correctly with null bufferSize pointer\n"); | |
689 | } | |
690 | err = U_ZERO_ERROR; | |
691 | ||
692 | /* buffer size pointer is 0 - fill in pbufferSize with a size */ | |
693 | bufferSize = 0; | |
694 | if (0 != ucol_safeClone(someCollators[0], buffer[0], &bufferSize, &err) || U_FAILURE(err) || bufferSize <= 0) | |
695 | { | |
696 | log_err("FAIL: Cloned Collator failed a sizing request ('preflighting')\n"); | |
697 | } | |
698 | /* Verify our define is large enough */ | |
699 | if (U_COL_SAFECLONE_BUFFERSIZE < bufferSize) | |
700 | { | |
701 | log_err("FAIL: Pre-calculated buffer size is too small\n"); | |
702 | } | |
703 | /* Verify we can use this run-time calculated size */ | |
704 | if (0 == (col = ucol_safeClone(someCollators[0], buffer[0], &bufferSize, &err)) || U_FAILURE(err)) | |
705 | { | |
706 | log_err("FAIL: Collator can't be cloned with run-time size\n"); | |
707 | } | |
708 | if (col) ucol_close(col); | |
709 | /* size one byte too small - should allocate & let us know */ | |
710 | --bufferSize; | |
711 | if (0 == (col = ucol_safeClone(someCollators[0], 0, &bufferSize, &err)) || err != U_SAFECLONE_ALLOCATED_WARNING) | |
712 | { | |
713 | log_err("FAIL: Cloned Collator failed to deal correctly with too-small buffer size\n"); | |
714 | } | |
715 | if (col) ucol_close(col); | |
716 | err = U_ZERO_ERROR; | |
717 | bufferSize = U_COL_SAFECLONE_BUFFERSIZE; | |
718 | ||
719 | ||
720 | /* Null buffer pointer - return Collator & set error to U_SAFECLONE_ALLOCATED_ERROR */ | |
721 | if (0 == (col = ucol_safeClone(someCollators[0], 0, &bufferSize, &err)) || err != U_SAFECLONE_ALLOCATED_WARNING) | |
722 | { | |
723 | log_err("FAIL: Cloned Collator failed to deal correctly with null buffer pointer\n"); | |
724 | } | |
725 | if (col) ucol_close(col); | |
726 | err = U_ZERO_ERROR; | |
727 | ||
728 | /* Null Collator - return NULL & set U_ILLEGAL_ARGUMENT_ERROR */ | |
729 | if (0 != ucol_safeClone(0, buffer[0], &bufferSize, &err) || err != U_ILLEGAL_ARGUMENT_ERROR) | |
730 | { | |
731 | log_err("FAIL: Cloned Collator failed to deal correctly with null Collator pointer\n"); | |
732 | } | |
733 | ||
734 | err = U_ZERO_ERROR; | |
735 | ||
736 | /* change orig & clone & make sure they are independent */ | |
737 | ||
738 | for (index = 0; index < CLONETEST_COLLATOR_COUNT; index++) | |
739 | { | |
740 | bufferSize = U_COL_SAFECLONE_BUFFERSIZE; | |
741 | someClonedCollators[index] = ucol_safeClone(someCollators[index], buffer[index], &bufferSize, &err); | |
742 | ||
743 | ucol_setStrength(someClonedCollators[index], UCOL_TERTIARY); | |
744 | ucol_setStrength(someCollators[index], UCOL_PRIMARY); | |
745 | ucol_setAttribute(someClonedCollators[index], UCOL_CASE_LEVEL, UCOL_OFF, &err); | |
746 | ucol_setAttribute(someCollators[index], UCOL_CASE_LEVEL, UCOL_OFF, &err); | |
747 | ||
748 | doAssert( (ucol_greater(someClonedCollators[index], test1, u_strlen(test1), test2, u_strlen(test2))), "Result should be \"abCda\" >>> \"abcda\" "); | |
749 | doAssert( (ucol_equal(someCollators[index], test1, u_strlen(test1), test2, u_strlen(test2))), "Result should be \"abcda\" == \"abCda\""); | |
750 | ||
751 | ucol_close(someClonedCollators[index]); | |
752 | ucol_close(someCollators[index]); | |
753 | } | |
754 | free(test1); | |
755 | free(test2); | |
756 | } | |
757 | ||
758 | /* | |
759 | ---------------------------------------------------------------------------- | |
760 | ctor -- Tests the getSortKey | |
761 | */ | |
762 | void TestSortKey() | |
763 | { | |
764 | uint8_t *sortk1 = NULL, *sortk2 = NULL, *sortk3 = NULL, *sortkEmpty = NULL; | |
765 | uint8_t sortk2_compat[] = { | |
374ca955 A |
766 | /* 2.6.1 key */ |
767 | 0x26, 0x28, 0x2A, 0x2C, 0x26, 0x01, | |
768 | 0x09, 0x01, 0x09, 0x01, 0x25, 0x01, | |
769 | 0x92, 0x93, 0x94, 0x95, 0x92, 0x00 | |
b75a7d8f | 770 | /* 2.2 key */ |
374ca955 | 771 | /*0x1D, 0x1F, 0x21, 0x23, 0x1D, 0x01, 0x09, 0x01, 0x09, 0x01, 0x1C, 0x01, 0x92, 0x93, 0x94, 0x95, 0x92, 0x00*/ |
b75a7d8f A |
772 | /* 2.0 key */ |
773 | /*0x19, 0x1B, 0x1D, 0x1F, 0x19, 0x01, 0x09, 0x01, 0x09, 0x01, 0x18, 0x01, 0x92, 0x93, 0x94, 0x95, 0x92, 0x00*/ | |
774 | /* 1.8.1 key.*/ | |
775 | /*0x19, 0x1B, 0x1D, 0x1F, 0x19, 0x01, 0x0A, 0x01, 0x0A, 0x01, 0x92, 0x93, 0x94, 0x95, 0x92, 0x00*/ | |
776 | /*this is a 1.8 sortkey */ | |
777 | /*0x17, 0x19, 0x1B, 0x1D, 0x17, 0x01, 0x08, 0x01, 0x08, 0x00*/ | |
778 | /*this is a 1.7 sortkey */ | |
779 | /*0x02, 0x54, 0x02, 0x55, 0x02, 0x56, 0x02, 0x57, 0x02, 0x54, 0x01, 0x02, 0x02, 0x02, 0x02, 0x02, 0x01, 0x02, 0x02, 0x02, 0x02, 0x02, 0x00*/ | |
780 | /* this is a 1.6 sortkey */ | |
781 | /*0x00, 0x53, 0x00, 0x54, 0x00, 0x55, 0x00, 0x56, 0x00, 0x53, 0x00, 0x00, 0x00, 0x01, 0x00, 0x01, 0x00, 0x01, 0x00, 0x01, 0x00, 0x01, 0x00, 0x00, 0x00, 0x01, 0x00, 0x01, 0x00, 0x01, 0x00, 0x01, 0x00, 0x01, 0x00, 0x00*/ | |
782 | }; | |
783 | ||
784 | int32_t sortklen, osortklen; | |
785 | uint32_t toStringLen=0; | |
786 | UCollator *col; | |
787 | UChar *test1, *test2, *test3; | |
788 | UErrorCode status = U_ZERO_ERROR; | |
789 | char toStringBuffer[256], *resultP; | |
790 | ||
791 | ||
792 | uint8_t s1[] = { 0x9f, 0x00 }; | |
793 | uint8_t s2[] = { 0x61, 0x00 }; | |
794 | int strcmpResult; | |
795 | ||
796 | strcmpResult = strcmp((const char *)s1, (const char *)s2); | |
797 | log_verbose("strcmp(0x9f..., 0x61...) = %d\n", strcmpResult); | |
798 | ||
799 | if(strcmpResult <= 0) { | |
800 | log_err("ERR: expected strcmp(\"9f 00\", \"61 00\") to be >=0 (GREATER).. got %d. Calling strcmp() for sortkeys may not work! \n", | |
801 | strcmpResult); | |
802 | } | |
803 | ||
804 | ||
805 | log_verbose("testing SortKey begins...\n"); | |
806 | /* this is supposed to open default date format, but later on it treats it like it is "en_US" | |
807 | - very bad if you try to run the tests on machine where default locale is NOT "en_US" */ | |
808 | /* col = ucol_open(NULL, &status); */ | |
809 | col = ucol_open("en_US", &status); | |
810 | if (U_FAILURE(status)) { | |
811 | log_err("ERROR: Default collation creation failed.: %s\n", myErrorName(status)); | |
812 | return; | |
813 | } | |
814 | ||
815 | ||
816 | if(ucol_getStrength(col) != UCOL_DEFAULT_STRENGTH) | |
817 | { | |
818 | log_err("ERROR: default collation did not have UCOL_DEFAULT_STRENGTH !\n"); | |
819 | } | |
820 | /* Need to use identical strength */ | |
821 | ucol_setAttribute(col, UCOL_STRENGTH, UCOL_IDENTICAL, &status); | |
822 | ||
823 | test1=(UChar*)malloc(sizeof(UChar) * 6); | |
824 | test2=(UChar*)malloc(sizeof(UChar) * 6); | |
825 | test3=(UChar*)malloc(sizeof(UChar) * 6); | |
826 | ||
827 | memset(test1,0xFE, sizeof(UChar)*6); | |
828 | memset(test2,0xFE, sizeof(UChar)*6); | |
829 | memset(test3,0xFE, sizeof(UChar)*6); | |
830 | ||
831 | ||
832 | u_uastrcpy(test1, "Abcda"); | |
833 | u_uastrcpy(test2, "abcda"); | |
834 | u_uastrcpy(test3, "abcda"); | |
835 | ||
836 | log_verbose("Use tertiary comparison level testing ....\n"); | |
837 | ||
838 | sortklen=ucol_getSortKey(col, test1, u_strlen(test1), NULL, 0); | |
839 | sortk1=(uint8_t*)malloc(sizeof(uint8_t) * (sortklen+1)); | |
840 | memset(sortk1,0xFE, sortklen); | |
841 | ucol_getSortKey(col, test1, u_strlen(test1), sortk1, sortklen+1); | |
842 | ||
843 | sortklen=ucol_getSortKey(col, test2, u_strlen(test2), NULL, 0); | |
844 | sortk2=(uint8_t*)malloc(sizeof(uint8_t) * (sortklen+1)); | |
845 | memset(sortk2,0xFE, sortklen); | |
846 | ucol_getSortKey(col, test2, u_strlen(test2), sortk2, sortklen+1); | |
847 | ||
848 | osortklen = sortklen; | |
849 | sortklen=ucol_getSortKey(col, test2, u_strlen(test3), NULL, 0); | |
850 | sortk3=(uint8_t*)malloc(sizeof(uint8_t) * (sortklen+1)); | |
851 | memset(sortk3,0xFE, sortklen); | |
852 | ucol_getSortKey(col, test2, u_strlen(test2), sortk3, sortklen+1); | |
853 | ||
854 | doAssert( (sortklen == osortklen), "Sortkey length should be the same (abcda, abcda)"); | |
855 | ||
856 | doAssert( (memcmp(sortk1, sortk2, sortklen) > 0), "Result should be \"Abcda\" > \"abcda\""); | |
857 | doAssert( (memcmp(sortk2, sortk1, sortklen) < 0), "Result should be \"abcda\" < \"Abcda\""); | |
858 | doAssert( (memcmp(sortk2, sortk3, sortklen) == 0), "Result should be \"abcda\" == \"abcda\""); | |
859 | ||
860 | doAssert( (memcmp(sortk2, sortk2_compat, sortklen) == 0), "Binary format for 'abcda' sortkey different!"); | |
861 | ||
862 | resultP = ucol_sortKeyToString(col, sortk2_compat, toStringBuffer, &toStringLen); | |
863 | doAssert( (resultP != 0), "sortKeyToString failed!"); | |
864 | ||
865 | #if 1 /* verobse log of sortkeys */ | |
866 | { | |
867 | char junk2[1000]; | |
868 | char junk3[1000]; | |
869 | int i; | |
870 | ||
871 | strcpy(junk2, "abcda[2] "); | |
872 | strcpy(junk3, " abcda[3] "); | |
873 | ||
874 | for(i=0;i<sortklen;i++) | |
875 | { | |
876 | sprintf(junk2+strlen(junk2), "%02X ",(int)( 0xFF & sortk2[i])); | |
877 | sprintf(junk3+strlen(junk3), "%02X ",(int)( 0xFF & sortk3[i])); | |
878 | } | |
879 | ||
880 | log_verbose("%s\n", junk2); | |
881 | log_verbose("%s\n", junk3); | |
882 | } | |
883 | #endif | |
884 | ||
885 | free(sortk1); | |
886 | free(sortk2); | |
887 | free(sortk3); | |
888 | ||
889 | log_verbose("Use secondary comparision level testing ...\n"); | |
890 | ucol_setStrength(col, UCOL_SECONDARY); | |
891 | sortklen=ucol_getSortKey(col, test1, u_strlen(test1), NULL, 0); | |
892 | sortk1=(uint8_t*)malloc(sizeof(uint8_t) * (sortklen+1)); | |
893 | ucol_getSortKey(col, test1, u_strlen(test1), sortk1, sortklen+1); | |
894 | sortklen=ucol_getSortKey(col, test2, u_strlen(test2), NULL, 0); | |
895 | sortk2=(uint8_t*)malloc(sizeof(uint8_t) * (sortklen+1)); | |
896 | ucol_getSortKey(col, test2, u_strlen(test2), sortk2, sortklen+1); | |
897 | ||
898 | doAssert( !(memcmp(sortk1, sortk2, sortklen) > 0), "Result should be \"Abcda\" == \"abcda\""); | |
899 | doAssert( !(memcmp(sortk2, sortk1, sortklen) < 0), "Result should be \"abcda\" == \"Abcda\""); | |
900 | doAssert( (memcmp(sortk1, sortk2, sortklen) == 0), "Result should be \"abcda\" == \"abcda\""); | |
901 | ||
902 | log_verbose("getting sortkey for an empty string\n"); | |
903 | ucol_setAttribute(col, UCOL_STRENGTH, UCOL_TERTIARY, &status); | |
904 | sortklen = ucol_getSortKey(col, test1, 0, NULL, 0); | |
905 | sortkEmpty = (uint8_t*)malloc(sizeof(uint8_t) * sortklen+1); | |
906 | sortklen = ucol_getSortKey(col, test1, 0, sortkEmpty, sortklen+1); | |
907 | if(sortklen != 3 || sortkEmpty[0] != 1 || sortkEmpty[0] != 1 || sortkEmpty[2] != 0) { | |
908 | log_err("Empty string generated wrong sortkey!\n"); | |
909 | } | |
910 | free(sortkEmpty); | |
911 | ||
912 | log_verbose("testing passing invalid string\n"); | |
913 | sortklen = ucol_getSortKey(col, NULL, 0, NULL, 0); | |
914 | if(sortklen != 0) { | |
915 | log_err("Invalid string didn't return sortkey size of 0\n"); | |
916 | } | |
917 | ||
918 | ||
919 | log_verbose("testing sortkey ends...\n"); | |
920 | ucol_close(col); | |
921 | free(test1); | |
922 | free(test2); | |
923 | free(test3); | |
924 | free(sortk1); | |
925 | free(sortk2); | |
926 | ||
927 | } | |
928 | void TestHashCode() | |
929 | { | |
930 | uint8_t *sortk1, *sortk2, *sortk3; | |
931 | int32_t sortk1len, sortk2len, sortk3len; | |
932 | UCollator *col; | |
933 | UChar *test1, *test2, *test3; | |
934 | UErrorCode status = U_ZERO_ERROR; | |
935 | log_verbose("testing getHashCode begins...\n"); | |
936 | col = ucol_open("en_US", &status); | |
937 | if (U_FAILURE(status)) { | |
938 | log_err("ERROR: Default collation creation failed.: %s\n", myErrorName(status)); | |
939 | return; | |
940 | } | |
941 | test1=(UChar*)malloc(sizeof(UChar) * 6); | |
942 | test2=(UChar*)malloc(sizeof(UChar) * 6); | |
943 | test3=(UChar*)malloc(sizeof(UChar) * 6); | |
944 | u_uastrcpy(test1, "Abcda"); | |
945 | u_uastrcpy(test2, "abcda"); | |
946 | u_uastrcpy(test3, "abcda"); | |
947 | ||
948 | log_verbose("Use tertiary comparison level testing ....\n"); | |
949 | sortk1len=ucol_getSortKey(col, test1, u_strlen(test1), NULL, 0); | |
950 | sortk1=(uint8_t*)malloc(sizeof(uint8_t) * (sortk1len+1)); | |
951 | ucol_getSortKey(col, test1, u_strlen(test1), sortk1, sortk1len+1); | |
952 | sortk2len=ucol_getSortKey(col, test2, u_strlen(test2), NULL, 0); | |
953 | sortk2=(uint8_t*)malloc(sizeof(uint8_t) * (sortk2len+1)); | |
954 | ucol_getSortKey(col, test2, u_strlen(test2), sortk2, sortk2len+1); | |
955 | sortk3len=ucol_getSortKey(col, test2, u_strlen(test3), NULL, 0); | |
956 | sortk3=(uint8_t*)malloc(sizeof(uint8_t) * (sortk3len+1)); | |
957 | ucol_getSortKey(col, test2, u_strlen(test2), sortk3, sortk3len+1); | |
958 | ||
959 | ||
960 | log_verbose("ucol_hashCode() testing ...\n"); | |
961 | ||
962 | doAssert( ucol_keyHashCode(sortk1, sortk1len) != ucol_keyHashCode(sortk2, sortk2len), "Hash test1 result incorrect" ); | |
963 | doAssert( !(ucol_keyHashCode(sortk1, sortk1len) == ucol_keyHashCode(sortk2, sortk2len)), "Hash test2 result incorrect" ); | |
964 | doAssert( ucol_keyHashCode(sortk2, sortk2len) == ucol_keyHashCode(sortk3, sortk3len), "Hash result not equal" ); | |
965 | ||
966 | log_verbose("hashCode tests end.\n"); | |
967 | ucol_close(col); | |
968 | free(sortk1); | |
969 | free(sortk2); | |
970 | free(sortk3); | |
971 | free(test1); | |
972 | free(test2); | |
973 | free(test3); | |
974 | ||
975 | ||
976 | } | |
977 | /* | |
978 | *---------------------------------------------------------------------------- | |
979 | * Tests the UCollatorElements API. | |
980 | * | |
981 | */ | |
982 | void TestElemIter() | |
983 | { | |
984 | int32_t offset; | |
985 | int32_t order1, order2, order3; | |
986 | UChar *testString1, *testString2; | |
987 | UCollator *col; | |
988 | UCollationElements *iterator1, *iterator2, *iterator3; | |
989 | UErrorCode status = U_ZERO_ERROR; | |
990 | log_verbose("testing UCollatorElements begins...\n"); | |
991 | col = ucol_open("en_US", &status); | |
992 | ucol_setAttribute(col, UCOL_NORMALIZATION_MODE, UCOL_OFF, &status); | |
993 | if (U_FAILURE(status)) { | |
994 | log_err("ERROR: Default collation creation failed.: %s\n", myErrorName(status)); | |
995 | return; | |
996 | } | |
997 | ||
998 | testString1=(UChar*)malloc(sizeof(UChar) * 150); | |
999 | testString2=(UChar*)malloc(sizeof(UChar) * 150); | |
1000 | u_uastrcpy(testString1, "XFILE What subset of all possible test cases has the highest probability of detecting the most errors?"); | |
1001 | u_uastrcpy(testString2, "Xf_ile What subset of all possible test cases has the lowest probability of detecting the least errors?"); | |
1002 | ||
1003 | log_verbose("Constructors and comparison testing....\n"); | |
1004 | ||
1005 | iterator1 = ucol_openElements(col, testString1, u_strlen(testString1), &status); | |
1006 | if(U_FAILURE(status)) { | |
1007 | log_err("ERROR: Default collationElement iterator creation failed.: %s\n", myErrorName(status)); | |
1008 | ucol_close(col); | |
1009 | return; | |
1010 | } | |
1011 | else{ log_verbose("PASS: Default collationElement iterator1 creation passed\n");} | |
1012 | ||
1013 | iterator2 = ucol_openElements(col, testString1, u_strlen(testString1), &status); | |
1014 | if(U_FAILURE(status)) { | |
1015 | log_err("ERROR: Default collationElement iterator creation failed.: %s\n", myErrorName(status)); | |
1016 | ucol_close(col); | |
1017 | return; | |
1018 | } | |
1019 | else{ log_verbose("PASS: Default collationElement iterator2 creation passed\n");} | |
1020 | ||
1021 | iterator3 = ucol_openElements(col, testString2, u_strlen(testString2), &status); | |
1022 | if(U_FAILURE(status)) { | |
1023 | log_err("ERROR: Default collationElement iterator creation failed.: %s\n", myErrorName(status)); | |
1024 | ucol_close(col); | |
1025 | return; | |
1026 | } | |
1027 | else{ log_verbose("PASS: Default collationElement iterator3 creation passed\n");} | |
1028 | ||
1029 | offset=ucol_getOffset(iterator1); | |
1030 | ucol_setOffset(iterator1, 6, &status); | |
1031 | if (U_FAILURE(status)) { | |
1032 | log_err("Error in setOffset for UCollatorElements iterator.: %s\n", myErrorName(status)); | |
1033 | return; | |
1034 | } | |
1035 | if(ucol_getOffset(iterator1)==6) | |
1036 | log_verbose("setOffset and getOffset working fine\n"); | |
1037 | else{ | |
1038 | log_err("error in set and get Offset got %d instead of 6\n", ucol_getOffset(iterator1)); | |
1039 | } | |
1040 | ||
1041 | ucol_setOffset(iterator1, 0, &status); | |
1042 | order1 = ucol_next(iterator1, &status); | |
1043 | if (U_FAILURE(status)) { | |
1044 | log_err("Somehow ran out of memory stepping through the iterator1.: %s\n", myErrorName(status)); | |
1045 | return; | |
1046 | } | |
1047 | order2=ucol_getOffset(iterator2); | |
1048 | doAssert((order1 != order2), "The first iterator advance failed"); | |
1049 | order2 = ucol_next(iterator2, &status); | |
1050 | if (U_FAILURE(status)) { | |
1051 | log_err("Somehow ran out of memory stepping through the iterator2.: %s\n", myErrorName(status)); | |
1052 | return; | |
1053 | } | |
1054 | order3 = ucol_next(iterator3, &status); | |
1055 | if (U_FAILURE(status)) { | |
1056 | log_err("Somehow ran out of memory stepping through the iterator3.: %s\n", myErrorName(status)); | |
1057 | return; | |
1058 | } | |
1059 | ||
1060 | doAssert((order1 == order2), "The second iterator advance failed should be the same as first one"); | |
1061 | ||
1062 | doAssert( (ucol_primaryOrder(order1) == ucol_primaryOrder(order3)), "The primary orders should be identical"); | |
1063 | doAssert( (ucol_secondaryOrder(order1) == ucol_secondaryOrder(order3)), "The secondary orders should be identical"); | |
1064 | doAssert( (ucol_tertiaryOrder(order1) == ucol_tertiaryOrder(order3)), "The tertiary orders should be identical"); | |
1065 | ||
1066 | order1=ucol_next(iterator1, &status); | |
1067 | if (U_FAILURE(status)) { | |
1068 | log_err("Somehow ran out of memory stepping through the iterator2.: %s\n", myErrorName(status)); | |
1069 | return; | |
1070 | } | |
1071 | order3=ucol_next(iterator3, &status); | |
1072 | if (U_FAILURE(status)) { | |
1073 | log_err("Somehow ran out of memory stepping through the iterator2.: %s\n", myErrorName(status)); | |
1074 | return; | |
1075 | } | |
1076 | doAssert( (ucol_primaryOrder(order1) == ucol_primaryOrder(order3)), "The primary orders should be identical"); | |
1077 | doAssert( (ucol_tertiaryOrder(order1) != ucol_tertiaryOrder(order3)), "The tertiary orders should be different"); | |
1078 | ||
1079 | order1=ucol_next(iterator1, &status); | |
1080 | if (U_FAILURE(status)) { | |
1081 | log_err("Somehow ran out of memory stepping through the iterator2.: %s\n", myErrorName(status)); | |
1082 | return; | |
1083 | } | |
1084 | order3=ucol_next(iterator3, &status); | |
1085 | if (U_FAILURE(status)) { | |
1086 | log_err("Somehow ran out of memory stepping through the iterator2.: %s\n", myErrorName(status)); | |
1087 | return; | |
1088 | } | |
1089 | /* this here, my friends, is either pure lunacy or something so obsolete that even it's mother | |
1090 | * doesn't care about it. Essentialy, this test complains if secondary values for 'I' and '_' | |
1091 | * are the same. According to the UCA, this is not true. Therefore, remove the test. | |
1092 | * Besides, if primary strengths for two code points are different, it doesn't matter one bit | |
1093 | * what is the relation between secondary or any other strengths. | |
1094 | * killed by weiv 06/11/2002. | |
1095 | */ | |
1096 | /* | |
1097 | doAssert( ((order1 & UCOL_SECONDARYMASK) != (order3 & UCOL_SECONDARYMASK)), "The secondary orders should be different"); | |
1098 | */ | |
1099 | doAssert( (order1 != UCOL_NULLORDER), "Unexpected end of iterator reached"); | |
1100 | ||
1101 | free(testString1); | |
1102 | free(testString2); | |
1103 | ucol_closeElements(iterator1); | |
1104 | ucol_closeElements(iterator2); | |
1105 | ucol_closeElements(iterator3); | |
1106 | ucol_close(col); | |
1107 | ||
1108 | log_verbose("testing CollationElementIterator ends...\n"); | |
1109 | } | |
1110 | ||
1111 | void TestGetLocale() { | |
1112 | UErrorCode status = U_ZERO_ERROR; | |
1113 | const char *rules = "&a<x<y<z"; | |
1114 | UChar rlz[256] = {0}; | |
1115 | uint32_t rlzLen = u_unescape(rules, rlz, 256); | |
1116 | ||
1117 | UCollator *coll = NULL; | |
1118 | const char *locale = NULL; | |
1119 | ||
1120 | int32_t i = 0; | |
1121 | ||
374ca955 | 1122 | /* Now that the collation tree is separate, actual==valid at all times. [alan] */ |
b75a7d8f A |
1123 | static const struct { |
1124 | const char* requestedLocale; | |
1125 | const char* validLocale; | |
1126 | const char* actualLocale; | |
1127 | } testStruct[] = { | |
1128 | { "sr_YU", "sr_YU", "ru" }, | |
374ca955 A |
1129 | { "sh_YU", "sh_YU", "hr" }, /* this used to be sh, but now sh collation aliases hr */ |
1130 | { "en_BE_FOO", "en_BE", "en_BE" }, | |
b75a7d8f A |
1131 | { "fr_FR_NONEXISTANT", "fr_FR", "fr" } |
1132 | }; | |
1133 | ||
1134 | /* test opening collators for different locales */ | |
1135 | for(i = 0; i<sizeof(testStruct)/sizeof(testStruct[0]); i++) { | |
1136 | status = U_ZERO_ERROR; | |
1137 | coll = ucol_open(testStruct[i].requestedLocale, &status); | |
1138 | if(U_FAILURE(status)) { | |
1139 | log_err("Failed to open collator for %s with %s\n", testStruct[i].requestedLocale, u_errorName(status)); | |
1140 | ucol_close(coll); | |
1141 | continue; | |
1142 | } | |
1143 | locale = ucol_getLocale(coll, ULOC_REQUESTED_LOCALE, &status); | |
1144 | if(strcmp(locale, testStruct[i].requestedLocale) != 0) { | |
1145 | log_err("[Coll %s]: Error in requested locale, expected %s, got %s\n", testStruct[i].requestedLocale, testStruct[i].requestedLocale, locale); | |
1146 | } | |
1147 | locale = ucol_getLocale(coll, ULOC_VALID_LOCALE, &status); | |
1148 | if(strcmp(locale, testStruct[i].validLocale) != 0) { | |
1149 | log_err("[Coll %s]: Error in valid locale, expected %s, got %s\n", testStruct[i].requestedLocale, testStruct[i].validLocale, locale); | |
1150 | } | |
1151 | locale = ucol_getLocale(coll, ULOC_ACTUAL_LOCALE, &status); | |
1152 | if(strcmp(locale, testStruct[i].actualLocale) != 0) { | |
1153 | log_err("[Coll %s]: Error in actual locale, expected %s, got %s\n", testStruct[i].requestedLocale, testStruct[i].actualLocale, locale); | |
1154 | } | |
1155 | ucol_close(coll); | |
1156 | } | |
1157 | ||
1158 | /* completely non-existant locale for collator should get a default collator */ | |
1159 | { | |
1160 | UCollator *defaultColl = ucol_open(NULL, &status); | |
1161 | coll = ucol_open("blahaha", &status); | |
1162 | if(U_SUCCESS(status)) { | |
1163 | if(strcmp(ucol_getLocale(coll, ULOC_REQUESTED_LOCALE, &status), "blahaha")) { | |
1164 | log_err("Nonexisting locale didn't preserve the requested locale\n"); | |
1165 | } | |
1166 | if(strcmp(ucol_getLocale(coll, ULOC_VALID_LOCALE, &status), | |
1167 | ucol_getLocale(defaultColl, ULOC_VALID_LOCALE, &status))) { | |
1168 | log_err("Valid locale for nonexisting locale locale collator differs " | |
1169 | "from valid locale for default collator\n"); | |
1170 | } | |
1171 | if(strcmp(ucol_getLocale(coll, ULOC_ACTUAL_LOCALE, &status), | |
1172 | ucol_getLocale(defaultColl, ULOC_ACTUAL_LOCALE, &status))) { | |
1173 | log_err("Actual locale for nonexisting locale locale collator differs " | |
1174 | "from actual locale for default collator\n"); | |
1175 | } | |
1176 | ucol_close(coll); | |
1177 | ucol_close(defaultColl); | |
1178 | } else { | |
1179 | log_data_err("Couldn't open collators\n"); | |
1180 | } | |
1181 | } | |
1182 | ||
1183 | ||
1184 | ||
1185 | /* collator instantiated from rules should have all three locales NULL */ | |
1186 | coll = ucol_openRules(rlz, rlzLen, UCOL_DEFAULT, UCOL_DEFAULT, NULL, &status); | |
1187 | locale = ucol_getLocale(coll, ULOC_REQUESTED_LOCALE, &status); | |
1188 | if(locale != NULL) { | |
1189 | log_err("For collator instantiated from rules, requested locale returned %s instead of NULL\n", locale); | |
1190 | } | |
1191 | locale = ucol_getLocale(coll, ULOC_VALID_LOCALE, &status); | |
1192 | if(locale != NULL) { | |
1193 | log_err("For collator instantiated from rules, valid locale returned %s instead of NULL\n", locale); | |
1194 | } | |
1195 | locale = ucol_getLocale(coll, ULOC_ACTUAL_LOCALE, &status); | |
1196 | if(locale != NULL) { | |
1197 | log_err("For collator instantiated from rules, actual locale returned %s instead of NULL\n", locale); | |
1198 | } | |
1199 | ucol_close(coll); | |
1200 | ||
1201 | } | |
1202 | ||
1203 | ||
1204 | void TestGetAll() | |
1205 | { | |
1206 | int32_t i, count; | |
1207 | count=ucol_countAvailable(); | |
1208 | /* use something sensible w/o hardcoding the count */ | |
1209 | if(count < 0){ | |
1210 | log_err("Error in countAvailable(), it returned %d\n", count); | |
1211 | } | |
1212 | else{ | |
1213 | log_verbose("PASS: countAvailable() successful, it returned %d\n", count); | |
1214 | } | |
1215 | for(i=0;i<count;i++) | |
1216 | log_verbose("%s\n", ucol_getAvailable(i)); | |
1217 | ||
1218 | ||
1219 | } | |
1220 | ||
1221 | ||
1222 | struct teststruct { | |
1223 | const char *original; | |
1224 | uint8_t key[256]; | |
1225 | } ; | |
1226 | ||
1227 | static int compare_teststruct(const void *string1, const void *string2) { | |
1228 | return(strcmp((const char *)((struct teststruct *)string1)->key, (const char *)((struct teststruct *)string2)->key)); | |
1229 | } | |
1230 | ||
1231 | void TestBounds() { | |
1232 | UErrorCode status = U_ZERO_ERROR; | |
1233 | ||
1234 | UCollator *coll = ucol_open("sh", &status); | |
1235 | ||
1236 | uint8_t sortkey[512], lower[512], upper[512]; | |
1237 | UChar buffer[512]; | |
1238 | ||
1239 | const char *test[] = { | |
1240 | "John Smith", | |
1241 | "JOHN SMITH", | |
1242 | "john SMITH", | |
1243 | "j\\u00F6hn sm\\u00EFth", | |
1244 | "J\\u00F6hn Sm\\u00EFth", | |
1245 | "J\\u00D6HN SM\\u00CFTH", | |
1246 | "john smithsonian", | |
1247 | "John Smithsonian", | |
1248 | }; | |
1249 | ||
1250 | static struct teststruct tests[] = { | |
1251 | {"\\u010CAKI MIHALJ" } , | |
1252 | {"\\u010CAKI MIHALJ" } , | |
1253 | {"\\u010CAKI PIRO\\u0160KA" }, | |
1254 | { "\\u010CABAI ANDRIJA" } , | |
1255 | {"\\u010CABAI LAJO\\u0160" } , | |
1256 | {"\\u010CABAI MARIJA" } , | |
1257 | {"\\u010CABAI STEVAN" } , | |
1258 | {"\\u010CABAI STEVAN" } , | |
1259 | {"\\u010CABARKAPA BRANKO" } , | |
1260 | {"\\u010CABARKAPA MILENKO" } , | |
1261 | {"\\u010CABARKAPA MIROSLAV" } , | |
1262 | {"\\u010CABARKAPA SIMO" } , | |
1263 | {"\\u010CABARKAPA STANKO" } , | |
1264 | {"\\u010CABARKAPA TAMARA" } , | |
1265 | {"\\u010CABARKAPA TOMA\\u0160" } , | |
1266 | {"\\u010CABDARI\\u0106 NIKOLA" } , | |
1267 | {"\\u010CABDARI\\u0106 ZORICA" } , | |
1268 | {"\\u010CABI NANDOR" } , | |
1269 | {"\\u010CABOVI\\u0106 MILAN" } , | |
1270 | {"\\u010CABRADI AGNEZIJA" } , | |
1271 | {"\\u010CABRADI IVAN" } , | |
1272 | {"\\u010CABRADI JELENA" } , | |
1273 | {"\\u010CABRADI LJUBICA" } , | |
1274 | {"\\u010CABRADI STEVAN" } , | |
1275 | {"\\u010CABRDA MARTIN" } , | |
1276 | {"\\u010CABRILO BOGDAN" } , | |
1277 | {"\\u010CABRILO BRANISLAV" } , | |
1278 | {"\\u010CABRILO LAZAR" } , | |
1279 | {"\\u010CABRILO LJUBICA" } , | |
1280 | {"\\u010CABRILO SPASOJA" } , | |
1281 | {"\\u010CADE\\u0160 ZDENKA" } , | |
1282 | {"\\u010CADESKI BLAGOJE" } , | |
1283 | {"\\u010CADOVSKI VLADIMIR" } , | |
1284 | {"\\u010CAGLJEVI\\u0106 TOMA" } , | |
1285 | {"\\u010CAGOROVI\\u0106 VLADIMIR" } , | |
1286 | {"\\u010CAJA VANKA" } , | |
1287 | {"\\u010CAJI\\u0106 BOGOLJUB" } , | |
1288 | {"\\u010CAJI\\u0106 BORISLAV" } , | |
1289 | {"\\u010CAJI\\u0106 RADOSLAV" } , | |
1290 | {"\\u010CAK\\u0160IRAN MILADIN" } , | |
1291 | {"\\u010CAKAN EUGEN" } , | |
1292 | {"\\u010CAKAN EVGENIJE" } , | |
1293 | {"\\u010CAKAN IVAN" } , | |
1294 | {"\\u010CAKAN JULIJAN" } , | |
1295 | {"\\u010CAKAN MIHAJLO" } , | |
1296 | {"\\u010CAKAN STEVAN" } , | |
1297 | {"\\u010CAKAN VLADIMIR" } , | |
1298 | {"\\u010CAKAN VLADIMIR" } , | |
1299 | {"\\u010CAKAN VLADIMIR" } , | |
1300 | {"\\u010CAKARA ANA" } , | |
1301 | {"\\u010CAKAREVI\\u0106 MOMIR" } , | |
1302 | {"\\u010CAKAREVI\\u0106 NEDELJKO" } , | |
1303 | {"\\u010CAKI \\u0160ANDOR" } , | |
1304 | {"\\u010CAKI AMALIJA" } , | |
1305 | {"\\u010CAKI ANDRA\\u0160" } , | |
1306 | {"\\u010CAKI LADISLAV" } , | |
1307 | {"\\u010CAKI LAJO\\u0160" } , | |
1308 | {"\\u010CAKI LASLO" } , | |
1309 | }; | |
1310 | ||
1311 | ||
1312 | ||
1313 | int32_t i = 0, j = 0, k = 0, buffSize = 0, skSize = 0, lowerSize = 0, upperSize = 0; | |
1314 | int32_t arraySize = sizeof(tests)/sizeof(tests[0]); | |
1315 | ||
1316 | if(U_SUCCESS(status) && coll) { | |
1317 | for(i = 0; i<arraySize; i++) { | |
1318 | buffSize = u_unescape(tests[i].original, buffer, 512); | |
1319 | skSize = ucol_getSortKey(coll, buffer, buffSize, tests[i].key, 512); | |
1320 | } | |
1321 | ||
1322 | qsort(tests, arraySize, sizeof(struct teststruct), compare_teststruct); | |
1323 | ||
1324 | for(i = 0; i < arraySize-1; i++) { | |
1325 | for(j = i+1; j < arraySize; j++) { | |
1326 | lowerSize = ucol_getBound(tests[i].key, -1, UCOL_BOUND_LOWER, 1, lower, 512, &status); | |
1327 | upperSize = ucol_getBound(tests[j].key, -1, UCOL_BOUND_UPPER, 1, upper, 512, &status); | |
1328 | for(k = i; k <= j; k++) { | |
1329 | if(strcmp((const char *)lower, (const char *)tests[k].key) > 0) { | |
1330 | log_err("Problem with lower! j = %i (%s vs %s)\n", k, tests[k].original, tests[i].original); | |
1331 | } | |
1332 | if(strcmp((const char *)upper, (const char *)tests[k].key) <= 0) { | |
1333 | log_err("Problem with upper! j = %i (%s vs %s)\n", k, tests[k].original, tests[j].original); | |
1334 | } | |
1335 | } | |
1336 | } | |
1337 | } | |
1338 | ||
1339 | ||
1340 | #if 0 | |
1341 | for(i = 0; i < 1000; i++) { | |
1342 | lowerRND = (rand()/(RAND_MAX/arraySize)); | |
1343 | upperRND = lowerRND + (rand()/(RAND_MAX/(arraySize-lowerRND))); | |
1344 | ||
1345 | lowerSize = ucol_getBound(tests[lowerRND].key, -1, UCOL_BOUND_LOWER, 1, lower, 512, &status); | |
1346 | upperSize = ucol_getBound(tests[upperRND].key, -1, UCOL_BOUND_UPPER_LONG, 1, upper, 512, &status); | |
1347 | ||
1348 | for(j = lowerRND; j<=upperRND; j++) { | |
1349 | if(strcmp(lower, tests[j].key) > 0) { | |
1350 | log_err("Problem with lower! j = %i (%s vs %s)\n", j, tests[j].original, tests[lowerRND].original); | |
1351 | } | |
1352 | if(strcmp(upper, tests[j].key) <= 0) { | |
1353 | log_err("Problem with upper! j = %i (%s vs %s)\n", j, tests[j].original, tests[upperRND].original); | |
1354 | } | |
1355 | } | |
1356 | } | |
1357 | #endif | |
1358 | ||
1359 | ||
1360 | ||
1361 | ||
1362 | ||
1363 | for(i = 0; i<sizeof(test)/sizeof(test[0]); i++) { | |
1364 | buffSize = u_unescape(test[i], buffer, 512); | |
1365 | skSize = ucol_getSortKey(coll, buffer, buffSize, sortkey, 512); | |
1366 | lowerSize = ucol_getBound(sortkey, skSize, UCOL_BOUND_LOWER, 1, lower, 512, &status); | |
1367 | upperSize = ucol_getBound(sortkey, skSize, UCOL_BOUND_UPPER_LONG, 1, upper, 512, &status); | |
1368 | for(j = i+1; j<sizeof(test)/sizeof(test[0]); j++) { | |
1369 | buffSize = u_unescape(test[j], buffer, 512); | |
1370 | skSize = ucol_getSortKey(coll, buffer, buffSize, sortkey, 512); | |
1371 | if(strcmp((const char *)lower, (const char *)sortkey) > 0) { | |
1372 | log_err("Problem with lower! i = %i, j = %i (%s vs %s)\n", i, j, test[i], test[j]); | |
1373 | } | |
1374 | if(strcmp((const char *)upper, (const char *)sortkey) <= 0) { | |
1375 | log_err("Problem with upper! i = %i, j = %i (%s vs %s)\n", i, j, test[i], test[j]); | |
1376 | } | |
1377 | } | |
1378 | } | |
1379 | ucol_close(coll); | |
1380 | } else { | |
1381 | log_data_err("Couldn't open collator\n"); | |
1382 | } | |
1383 | ||
1384 | } | |
1385 | ||
1386 | static void doOverrunTest(UCollator *coll, const UChar *uString, int32_t strLen) { | |
1387 | int32_t skLen = 0, skLen2 = 0; | |
1388 | uint8_t sortKey[256]; | |
1389 | int32_t i, j; | |
1390 | uint8_t filler = 0xFF; | |
1391 | ||
1392 | skLen = ucol_getSortKey(coll, uString, strLen, NULL, 0); | |
1393 | ||
1394 | for(i = 0; i < skLen; i++) { | |
1395 | memset(sortKey, filler, 256); | |
1396 | skLen2 = ucol_getSortKey(coll, uString, strLen, sortKey, i); | |
1397 | if(skLen != skLen2) { | |
1398 | log_err("For buffer size %i, got different sortkey length. Expected %i got %i\n", i, skLen, skLen2); | |
1399 | } | |
1400 | for(j = i; j < 256; j++) { | |
1401 | if(sortKey[j] != filler) { | |
1402 | log_err("Something run over index %i\n", j); | |
1403 | break; | |
1404 | } | |
1405 | } | |
1406 | } | |
1407 | } | |
1408 | ||
1409 | /* j1865 reports that if a shorter buffer is passed to | |
1410 | * to get sort key, a buffer overrun happens in some | |
1411 | * cases. This test tries to check this. | |
1412 | */ | |
1413 | void TestSortKeyBufferOverrun(void) { | |
1414 | UErrorCode status = U_ZERO_ERROR; | |
1415 | const char* cString = "A very Merry liTTle-lamB.."; | |
1416 | UChar uString[256]; | |
1417 | int32_t strLen = 0; | |
1418 | UCollator *coll = ucol_open("root", &status); | |
1419 | strLen = u_unescape(cString, uString, 256); | |
1420 | ||
1421 | if(U_SUCCESS(status)) { | |
1422 | log_verbose("testing non ignorable\n"); | |
1423 | ucol_setAttribute(coll, UCOL_ALTERNATE_HANDLING, UCOL_NON_IGNORABLE, &status); | |
1424 | doOverrunTest(coll, uString, strLen); | |
1425 | ||
1426 | log_verbose("testing shifted\n"); | |
1427 | ucol_setAttribute(coll, UCOL_ALTERNATE_HANDLING, UCOL_SHIFTED, &status); | |
1428 | doOverrunTest(coll, uString, strLen); | |
1429 | ||
1430 | log_verbose("testing shifted quaternary\n"); | |
1431 | ucol_setAttribute(coll, UCOL_STRENGTH, UCOL_QUATERNARY, &status); | |
1432 | doOverrunTest(coll, uString, strLen); | |
1433 | ||
1434 | log_verbose("testing with french secondaries\n"); | |
1435 | ucol_setAttribute(coll, UCOL_FRENCH_COLLATION, UCOL_ON, &status); | |
1436 | ucol_setAttribute(coll, UCOL_STRENGTH, UCOL_TERTIARY, &status); | |
1437 | ucol_setAttribute(coll, UCOL_ALTERNATE_HANDLING, UCOL_NON_IGNORABLE, &status); | |
1438 | doOverrunTest(coll, uString, strLen); | |
1439 | ||
1440 | } | |
1441 | ucol_close(coll); | |
1442 | } | |
1443 | ||
1444 | static void TestAttribute() | |
1445 | { | |
1446 | UErrorCode error = U_ZERO_ERROR; | |
1447 | UCollator *coll = ucol_open(NULL, &error); | |
1448 | ||
1449 | if (U_FAILURE(error)) { | |
1450 | log_err("Creation of default collator failed"); | |
1451 | return; | |
1452 | } | |
1453 | ||
1454 | ucol_setAttribute(coll, UCOL_FRENCH_COLLATION, UCOL_OFF, &error); | |
1455 | if (ucol_getAttribute(coll, UCOL_FRENCH_COLLATION, &error) != UCOL_OFF || | |
1456 | U_FAILURE(error)) { | |
1457 | log_err("Setting and retrieving of the french collation failed"); | |
1458 | } | |
1459 | ||
1460 | ucol_setAttribute(coll, UCOL_FRENCH_COLLATION, UCOL_ON, &error); | |
1461 | if (ucol_getAttribute(coll, UCOL_FRENCH_COLLATION, &error) != UCOL_ON || | |
1462 | U_FAILURE(error)) { | |
1463 | log_err("Setting and retrieving of the french collation failed"); | |
1464 | } | |
1465 | ||
1466 | ucol_setAttribute(coll, UCOL_ALTERNATE_HANDLING, UCOL_SHIFTED, &error); | |
1467 | if (ucol_getAttribute(coll, UCOL_ALTERNATE_HANDLING, &error) != UCOL_SHIFTED || | |
1468 | U_FAILURE(error)) { | |
1469 | log_err("Setting and retrieving of the alternate handling failed"); | |
1470 | } | |
1471 | ||
1472 | ucol_setAttribute(coll, UCOL_ALTERNATE_HANDLING, UCOL_NON_IGNORABLE, &error); | |
1473 | if (ucol_getAttribute(coll, UCOL_ALTERNATE_HANDLING, &error) != UCOL_NON_IGNORABLE || | |
1474 | U_FAILURE(error)) { | |
1475 | log_err("Setting and retrieving of the alternate handling failed"); | |
1476 | } | |
1477 | ||
1478 | ucol_setAttribute(coll, UCOL_CASE_FIRST, UCOL_LOWER_FIRST, &error); | |
1479 | if (ucol_getAttribute(coll, UCOL_CASE_FIRST, &error) != UCOL_LOWER_FIRST || | |
1480 | U_FAILURE(error)) { | |
1481 | log_err("Setting and retrieving of the case first attribute failed"); | |
1482 | } | |
1483 | ||
1484 | ucol_setAttribute(coll, UCOL_CASE_FIRST, UCOL_UPPER_FIRST, &error); | |
1485 | if (ucol_getAttribute(coll, UCOL_CASE_FIRST, &error) != UCOL_UPPER_FIRST || | |
1486 | U_FAILURE(error)) { | |
1487 | log_err("Setting and retrieving of the case first attribute failed"); | |
1488 | } | |
1489 | ||
1490 | ucol_setAttribute(coll, UCOL_CASE_LEVEL, UCOL_ON, &error); | |
1491 | if (ucol_getAttribute(coll, UCOL_CASE_LEVEL, &error) != UCOL_ON || | |
1492 | U_FAILURE(error)) { | |
1493 | log_err("Setting and retrieving of the case level attribute failed"); | |
1494 | } | |
1495 | ||
1496 | ucol_setAttribute(coll, UCOL_CASE_LEVEL, UCOL_OFF, &error); | |
1497 | if (ucol_getAttribute(coll, UCOL_CASE_LEVEL, &error) != UCOL_OFF || | |
1498 | U_FAILURE(error)) { | |
1499 | log_err("Setting and retrieving of the case level attribute failed"); | |
1500 | } | |
1501 | ||
1502 | ucol_setAttribute(coll, UCOL_NORMALIZATION_MODE, UCOL_ON, &error); | |
1503 | if (ucol_getAttribute(coll, UCOL_NORMALIZATION_MODE, &error) != UCOL_ON || | |
1504 | U_FAILURE(error)) { | |
1505 | log_err("Setting and retrieving of the normalization on/off attribute failed"); | |
1506 | } | |
1507 | ||
1508 | ucol_setAttribute(coll, UCOL_NORMALIZATION_MODE, UCOL_OFF, &error); | |
1509 | if (ucol_getAttribute(coll, UCOL_NORMALIZATION_MODE, &error) != UCOL_OFF || | |
1510 | U_FAILURE(error)) { | |
1511 | log_err("Setting and retrieving of the normalization on/off attribute failed"); | |
1512 | } | |
1513 | ||
1514 | ucol_setAttribute(coll, UCOL_STRENGTH, UCOL_PRIMARY, &error); | |
1515 | if (ucol_getAttribute(coll, UCOL_STRENGTH, &error) != UCOL_PRIMARY || | |
1516 | U_FAILURE(error)) { | |
1517 | log_err("Setting and retrieving of the collation strength failed"); | |
1518 | } | |
1519 | ||
1520 | ucol_setAttribute(coll, UCOL_STRENGTH, UCOL_SECONDARY, &error); | |
1521 | if (ucol_getAttribute(coll, UCOL_STRENGTH, &error) != UCOL_SECONDARY || | |
1522 | U_FAILURE(error)) { | |
1523 | log_err("Setting and retrieving of the collation strength failed"); | |
1524 | } | |
1525 | ||
1526 | ucol_setAttribute(coll, UCOL_STRENGTH, UCOL_TERTIARY, &error); | |
1527 | if (ucol_getAttribute(coll, UCOL_STRENGTH, &error) != UCOL_TERTIARY || | |
1528 | U_FAILURE(error)) { | |
1529 | log_err("Setting and retrieving of the collation strength failed"); | |
1530 | } | |
1531 | ||
1532 | ucol_setAttribute(coll, UCOL_STRENGTH, UCOL_QUATERNARY, &error); | |
1533 | if (ucol_getAttribute(coll, UCOL_STRENGTH, &error) != UCOL_QUATERNARY || | |
1534 | U_FAILURE(error)) { | |
1535 | log_err("Setting and retrieving of the collation strength failed"); | |
1536 | } | |
1537 | ||
1538 | ucol_setAttribute(coll, UCOL_STRENGTH, UCOL_IDENTICAL, &error); | |
1539 | if (ucol_getAttribute(coll, UCOL_STRENGTH, &error) != UCOL_IDENTICAL || | |
1540 | U_FAILURE(error)) { | |
1541 | log_err("Setting and retrieving of the collation strength failed"); | |
1542 | } | |
1543 | ||
1544 | ucol_close(coll); | |
1545 | } | |
1546 | ||
1547 | void TestGetTailoredSet() { | |
1548 | struct { | |
1549 | const char *rules; | |
1550 | const char *tests[20]; | |
1551 | int32_t testsize; | |
1552 | } setTest[] = { | |
1553 | { "&a < \\u212b", { "\\u212b", "A\\u030a", "\\u00c5" }, 3}, | |
1554 | { "& S < \\u0161 <<< \\u0160", { "\\u0161", "s\\u030C", "\\u0160", "S\\u030C" }, 4} | |
1555 | }; | |
1556 | ||
1557 | int32_t i = 0, j = 0; | |
1558 | UErrorCode status = U_ZERO_ERROR; | |
1559 | UParseError pError; | |
1560 | ||
1561 | UCollator *coll = NULL; | |
1562 | UChar buff[1024]; | |
1563 | int32_t buffLen = 0; | |
1564 | USet *set = NULL; | |
1565 | ||
1566 | for(i = 0; i < sizeof(setTest)/sizeof(setTest[0]); i++) { | |
1567 | buffLen = u_unescape(setTest[i].rules, buff, 1024); | |
1568 | coll = ucol_openRules(buff, buffLen, UCOL_DEFAULT, UCOL_DEFAULT, &pError, &status); | |
1569 | if(U_SUCCESS(status)) { | |
1570 | set = ucol_getTailoredSet(coll, &status); | |
1571 | if(uset_size(set) != setTest[i].testsize) { | |
1572 | log_err("Tailored set size different (%d) than expected (%d)\n", uset_size(set), setTest[i].testsize); | |
1573 | } | |
1574 | for(j = 0; j < setTest[i].testsize; j++) { | |
1575 | buffLen = u_unescape(setTest[i].tests[j], buff, 1024); | |
1576 | if(!uset_containsString(set, buff, buffLen)) { | |
1577 | log_err("Tailored set doesn't contain %s... It should\n", setTest[i].tests[j]); | |
1578 | } | |
1579 | } | |
1580 | uset_close(set); | |
1581 | } else { | |
1582 | log_err("Couldn't open collator with rules %s\n", setTest[i].rules); | |
1583 | } | |
1584 | ucol_close(coll); | |
1585 | } | |
1586 | } | |
1587 | ||
1588 | static int tMemCmp(const uint8_t *first, const uint8_t *second) { | |
374ca955 A |
1589 | int32_t firstLen = (int32_t)strlen((const char *)first); |
1590 | int32_t secondLen = (int32_t)strlen((const char *)second); | |
b75a7d8f A |
1591 | return memcmp(first, second, uprv_min(firstLen, secondLen)); |
1592 | } | |
1593 | static const char * strengthsC[] = { | |
1594 | "UCOL_PRIMARY", | |
1595 | "UCOL_SECONDARY", | |
1596 | "UCOL_TERTIARY", | |
1597 | "UCOL_QUATERNARY", | |
1598 | "UCOL_IDENTICAL" | |
1599 | }; | |
1600 | ||
1601 | void TestMergeSortKeys(void) { | |
1602 | UErrorCode status = U_ZERO_ERROR; | |
1603 | UCollator *coll = ucol_open("en", &status); | |
1604 | if(U_SUCCESS(status)) { | |
1605 | ||
1606 | const char* cases[] = { | |
1607 | "abc", | |
1608 | "abcd", | |
1609 | "abcde" | |
1610 | }; | |
1611 | uint32_t casesSize = sizeof(cases)/sizeof(cases[0]); | |
1612 | const char* prefix = "foo"; | |
1613 | const char* suffix = "egg"; | |
1614 | char outBuff1[256], outBuff2[256]; | |
1615 | ||
1616 | uint8_t **sortkeys = (uint8_t **)malloc(casesSize*sizeof(uint8_t *)); | |
1617 | uint8_t **mergedPrefixkeys = (uint8_t **)malloc(casesSize*sizeof(uint8_t *)); | |
1618 | uint8_t **mergedSuffixkeys = (uint8_t **)malloc(casesSize*sizeof(uint8_t *)); | |
1619 | uint32_t *sortKeysLen = (uint32_t *)malloc(casesSize*sizeof(uint32_t)); | |
1620 | uint8_t prefixKey[256], suffixKey[256]; | |
1621 | uint32_t prefixKeyLen = 0, suffixKeyLen = 0, i = 0; | |
1622 | UChar buffer[256]; | |
1623 | uint32_t unescapedLen = 0, l1 = 0, l2 = 0; | |
1624 | UColAttributeValue strength; | |
1625 | ||
1626 | log_verbose("ucol_mergeSortkeys test\n"); | |
1627 | log_verbose("Testing order of the test cases\n"); | |
1628 | genericLocaleStarter("en", cases, casesSize); | |
1629 | ||
1630 | for(i = 0; i<casesSize; i++) { | |
1631 | sortkeys[i] = (uint8_t *)malloc(256*sizeof(uint8_t)); | |
1632 | mergedPrefixkeys[i] = (uint8_t *)malloc(256*sizeof(uint8_t)); | |
1633 | mergedSuffixkeys[i] = (uint8_t *)malloc(256*sizeof(uint8_t)); | |
1634 | } | |
1635 | ||
1636 | unescapedLen = u_unescape(prefix, buffer, 256); | |
1637 | prefixKeyLen = ucol_getSortKey(coll, buffer, unescapedLen, prefixKey, 256); | |
1638 | ||
1639 | unescapedLen = u_unescape(suffix, buffer, 256); | |
1640 | suffixKeyLen = ucol_getSortKey(coll, buffer, unescapedLen, suffixKey, 256); | |
1641 | ||
1642 | log_verbose("Massaging data with prefixes and different strengths\n"); | |
1643 | strength = UCOL_PRIMARY; | |
1644 | while(strength <= UCOL_IDENTICAL) { | |
1645 | log_verbose("Strength %s\n", strengthsC[strength<=UCOL_QUATERNARY?strength:4]); | |
1646 | ucol_setAttribute(coll, UCOL_STRENGTH, strength, &status); | |
1647 | for(i = 0; i<casesSize; i++) { | |
1648 | unescapedLen = u_unescape(cases[i], buffer, 256); | |
1649 | sortKeysLen[i] = ucol_getSortKey(coll, buffer, unescapedLen, sortkeys[i], 256); | |
1650 | ucol_mergeSortkeys(prefixKey, prefixKeyLen, sortkeys[i], sortKeysLen[i], mergedPrefixkeys[i], 256); | |
1651 | ucol_mergeSortkeys(sortkeys[i], sortKeysLen[i], suffixKey, suffixKeyLen, mergedSuffixkeys[i], 256); | |
1652 | if(i>0) { | |
1653 | if(tMemCmp(mergedPrefixkeys[i-1], mergedPrefixkeys[i]) >= 0) { | |
1654 | log_err("Error while comparing prefixed keys @ strength %s:\n", strengthsC[strength<=UCOL_QUATERNARY?strength:4]); | |
1655 | log_err("%s\n%s\n", | |
1656 | ucol_sortKeyToString(coll, mergedPrefixkeys[i-1], outBuff1, &l1), | |
1657 | ucol_sortKeyToString(coll, mergedPrefixkeys[i], outBuff2, &l2)); | |
1658 | } | |
1659 | if(tMemCmp(mergedSuffixkeys[i-1], mergedSuffixkeys[i]) >= 0) { | |
1660 | log_err("Error while comparing suffixed keys @ strength %s:\n", strengthsC[strength<=UCOL_QUATERNARY?strength:4]); | |
1661 | log_err("%s\n%s\n", | |
1662 | ucol_sortKeyToString(coll, mergedSuffixkeys[i-1], outBuff1, &l1), | |
1663 | ucol_sortKeyToString(coll, mergedSuffixkeys[i], outBuff2, &l2)); | |
1664 | } | |
1665 | } | |
1666 | } | |
1667 | if(strength == UCOL_QUATERNARY) { | |
1668 | strength = UCOL_IDENTICAL; | |
1669 | } else { | |
1670 | strength++; | |
1671 | } | |
1672 | } | |
1673 | ||
1674 | { | |
1675 | uint8_t smallBuf[3]; | |
1676 | uint32_t reqLen = 0; | |
1677 | log_verbose("testing buffer overflow\n"); | |
1678 | reqLen = ucol_mergeSortkeys(prefixKey, prefixKeyLen, suffixKey, suffixKeyLen, smallBuf, 3); | |
1679 | if(reqLen != (prefixKeyLen+suffixKeyLen-1)) { | |
1680 | log_err("Wrong preflight size for merged sortkey\n"); | |
1681 | } | |
1682 | } | |
1683 | ||
1684 | { | |
1685 | UChar empty = 0; | |
1686 | uint8_t emptyKey[20], abcKey[50], mergedKey[100]; | |
1687 | int32_t emptyKeyLen = 0, abcKeyLen = 0, mergedKeyLen = 0; | |
1688 | ||
1689 | log_verbose("testing merging with sortkeys generated for empty strings\n"); | |
1690 | emptyKeyLen = ucol_getSortKey(coll, &empty, 0, emptyKey, 20); | |
1691 | unescapedLen = u_unescape(cases[0], buffer, 256); | |
1692 | abcKeyLen = ucol_getSortKey(coll, buffer, unescapedLen, abcKey, 50); | |
1693 | mergedKeyLen = ucol_mergeSortkeys(emptyKey, emptyKeyLen, abcKey, abcKeyLen, mergedKey, 100); | |
1694 | if(mergedKey[0] != 2) { | |
1695 | log_err("Empty sortkey didn't produce a level separator\n"); | |
1696 | } | |
1697 | /* try with zeros */ | |
1698 | mergedKeyLen = ucol_mergeSortkeys(emptyKey, 0, abcKey, abcKeyLen, mergedKey, 100); | |
1699 | if(mergedKeyLen != 0 || mergedKey[0] != 0) { | |
1700 | log_err("Empty key didn't produce null mergedKey\n"); | |
1701 | } | |
1702 | mergedKeyLen = ucol_mergeSortkeys(abcKey, abcKeyLen, emptyKey, 0, mergedKey, 100); | |
1703 | if(mergedKeyLen != 0 || mergedKey[0] != 0) { | |
1704 | log_err("Empty key didn't produce null mergedKey\n"); | |
1705 | } | |
1706 | ||
1707 | } | |
1708 | ||
1709 | for(i = 0; i<casesSize; i++) { | |
1710 | free(sortkeys[i]); | |
1711 | free(mergedPrefixkeys[i]); | |
1712 | free(mergedSuffixkeys[i]); | |
1713 | } | |
1714 | free(sortkeys); | |
1715 | free(mergedPrefixkeys); | |
1716 | free(mergedSuffixkeys); | |
1717 | free(sortKeysLen); | |
1718 | ucol_close(coll); | |
1719 | /* need to finish this up */ | |
1720 | } else { | |
1721 | log_data_err("Couldn't open collator"); | |
1722 | } | |
1723 | } | |
374ca955 A |
1724 | static void TestShortString(void) |
1725 | { | |
1726 | struct { | |
1727 | const char *input; | |
1728 | const char *expectedOutput; | |
1729 | const char *locale; | |
1730 | UErrorCode expectedStatus; | |
1731 | int32_t expectedOffset; | |
1732 | uint32_t expectedIdentifier; | |
1733 | } testCases[] = { | |
1734 | {"LDE_RDE_KPHONEBOOK_T0041_ZLATN","B2600_KPHONEBOOK_LDE", "de@collation=phonebook", U_USING_FALLBACK_WARNING, 0, 0 }, | |
1735 | {"LEN_RUS_NO_AS_S4","AS_LEN_NO_S4", NULL, U_USING_FALLBACK_WARNING, 0, 0 }, | |
1736 | {"LDE_VPHONEBOOK_EO_SI","EO_KPHONEBOOK_LDE_SI", "de@collation=phonebook", U_ZERO_ERROR, 0, 0 }, | |
1737 | {"LDE_Kphonebook","KPHONEBOOK_LDE", "de@collation=phonebook", U_ZERO_ERROR, 0, 0 }, | |
1738 | {"Xqde_DE@collation=phonebookq_S3_EX","KPHONEBOOK_LDE", "de@collation=phonebook", U_USING_FALLBACK_WARNING, 0, 0 }, | |
1739 | {"LFR_FO", "LFR", NULL, U_ZERO_ERROR, 0, 0 }, | |
1740 | {"SO_LX_AS", "", NULL, U_ILLEGAL_ARGUMENT_ERROR, 8, 0 }, | |
1741 | {"S3_ASS_MMM", "", NULL, U_ILLEGAL_ARGUMENT_ERROR, 5, 0 } | |
1742 | }; | |
1743 | ||
1744 | int32_t i = 0, j = 0; | |
1745 | UCollator *coll = NULL, *fromID = NULL, *fromNormalized = NULL; | |
1746 | UParseError parseError; | |
1747 | UErrorCode status = U_ZERO_ERROR; | |
1748 | char fromShortBuffer[256], fromIDBuffer[256], fromIDRoundtrip[256], normalizedBuffer[256], fromNormalizedBuffer[256]; | |
1749 | uint32_t identifier = 0, idFromSS = 0; | |
1750 | const char* locale = NULL; | |
1751 | ||
1752 | ||
1753 | for(i = 0; i < sizeof(testCases)/sizeof(testCases[0]); i++) { | |
1754 | status = U_ZERO_ERROR; | |
1755 | if(testCases[i].locale) { | |
1756 | locale = testCases[i].locale; | |
1757 | } else { | |
1758 | locale = NULL; | |
1759 | } | |
1760 | ||
1761 | coll = ucol_openFromShortString(testCases[i].input, FALSE, &parseError, &status); | |
1762 | if(status != testCases[i].expectedStatus) { | |
1763 | log_err("Got status '%s' that is different from expected '%s' for '%s'\n", | |
1764 | u_errorName(status), u_errorName(testCases[i].expectedStatus), testCases[i].input); | |
1765 | } | |
1766 | ||
1767 | if(U_SUCCESS(status)) { | |
1768 | ucol_getShortDefinitionString(coll, locale, fromShortBuffer, 256, &status); | |
1769 | ||
1770 | if(strcmp(fromShortBuffer, testCases[i].expectedOutput)) { | |
1771 | log_err("Got short string '%s' from the collator. Expected '%s' for input '%s'\n", | |
1772 | fromShortBuffer, testCases[i].expectedOutput, testCases[i].input); | |
1773 | } | |
1774 | ||
1775 | ucol_normalizeShortDefinitionString(testCases[i].input, normalizedBuffer, 256, &parseError, &status); | |
1776 | fromNormalized = ucol_openFromShortString(normalizedBuffer, FALSE, &parseError, &status); | |
1777 | ucol_getShortDefinitionString(fromNormalized, locale, fromNormalizedBuffer, 256, &status); | |
1778 | ||
1779 | if(strcmp(fromShortBuffer, fromNormalizedBuffer)) { | |
1780 | log_err("Strings obtained from collators instantiated by short string ('%s') and from normalized string ('%s') differ\n", | |
1781 | fromShortBuffer, fromNormalizedBuffer); | |
1782 | } | |
1783 | ||
1784 | ||
1785 | if(!ucol_equals(coll, fromNormalized)) { | |
1786 | log_err("Collator from short string ('%s') differs from one obtained through a normalized version ('%s')\n", | |
1787 | testCases[i].input, normalizedBuffer); | |
1788 | } | |
1789 | ||
1790 | /* test identifiers */ | |
1791 | identifier = ucol_collatorToIdentifier(coll, locale, &status); | |
1792 | if(identifier < UCOL_SIT_COLLATOR_NOT_ENCODABLE) { | |
1793 | ucol_identifierToShortString(identifier, fromIDBuffer, 256, FALSE, &status); | |
1794 | fromID = ucol_openFromIdentifier(identifier, FALSE, &status); | |
1795 | if(!ucol_equals(coll, fromID)) { | |
1796 | log_err("Collator from short string ('%s') differs from one obtained through an identifier ('%s')\n", | |
1797 | testCases[i].input, fromIDBuffer); | |
1798 | } | |
1799 | ucol_close(fromID); | |
1800 | } | |
1801 | ||
1802 | /* round-trip short string - identifier */ | |
1803 | for(j = 1; j < 2; j++) { | |
1804 | idFromSS = ucol_shortStringToIdentifier(testCases[i].input, (UBool)j, &status); | |
1805 | ucol_identifierToShortString(idFromSS, fromIDBuffer, 256, (UBool)j, &status); | |
1806 | identifier = ucol_shortStringToIdentifier(fromIDBuffer, (UBool)j, &status); | |
1807 | ucol_identifierToShortString(identifier, fromIDRoundtrip, 256, (UBool)j, &status); | |
1808 | ||
1809 | if(idFromSS != identifier) { | |
1810 | log_err("FD = %i, id didn't round trip. %08X vs %08X (%s)\n", | |
1811 | j, idFromSS, identifier, testCases[i]); | |
1812 | } | |
1813 | if(strcmp(fromIDBuffer, fromIDRoundtrip)) { | |
1814 | log_err("FD = %i, SS didn't round trip. %s vs %s (%s)\n", | |
1815 | j, fromIDBuffer, fromIDRoundtrip, testCases[i]); | |
1816 | } | |
1817 | } | |
1818 | ||
1819 | ucol_close(fromNormalized); | |
1820 | ucol_close(coll); | |
1821 | ||
1822 | } else { | |
1823 | if(parseError.offset != testCases[i].expectedOffset) { | |
1824 | log_err("Got parse error offset %i, but expected %i instead for '%s'\n", | |
1825 | parseError.offset, testCases[i].expectedOffset, testCases[i].input); | |
1826 | } | |
1827 | } | |
1828 | } | |
1829 | ||
1830 | } | |
1831 | ||
1832 | static void | |
1833 | doSetsTest(const USet *ref, USet *set, const char* inSet, const char* outSet, UErrorCode *status) { | |
1834 | UChar buffer[512]; | |
1835 | int32_t bufLen; | |
1836 | ||
1837 | uset_clear(set); | |
1838 | bufLen = u_unescape(inSet, buffer, 512); | |
1839 | uset_applyPattern(set, buffer, bufLen, 0, status); | |
1840 | if(U_FAILURE(*status)) { | |
1841 | log_err("Failure setting pattern %s\n", u_errorName(*status)); | |
1842 | } | |
1843 | ||
1844 | if(!uset_containsAll(ref, set)) { | |
1845 | log_err("Some stuff from %s is not present in the set\n", inSet); | |
1846 | } | |
1847 | ||
1848 | uset_clear(set); | |
1849 | bufLen = u_unescape(outSet, buffer, 512); | |
1850 | uset_applyPattern(set, buffer, bufLen, 0, status); | |
1851 | if(U_FAILURE(*status)) { | |
1852 | log_err("Failure setting pattern %s\n", u_errorName(*status)); | |
1853 | } | |
1854 | ||
1855 | if(!uset_containsNone(ref, set)) { | |
1856 | log_err("Some stuff from %s is present in the set\n", outSet); | |
1857 | } | |
1858 | } | |
1859 | ||
1860 | ||
1861 | ||
1862 | ||
1863 | static void | |
1864 | TestGetContractionsAndUnsafes(void) | |
1865 | { | |
1866 | static struct { | |
1867 | const char* locale; | |
1868 | const char* inConts; | |
1869 | const char* outConts; | |
1870 | const char* unsafeCodeUnits; | |
1871 | const char* safeCodeUnits; | |
1872 | } tests[] = { | |
1873 | { "ru", | |
1874 | "[{\\u0474\\u030F}{\\u0475\\u030F}{\\u04D8\\u0308}{\\u04D9\\u0308}{\\u04E8\\u0308}{\\u04E9\\u0308}]", | |
1875 | "[{\\u0430\\u0306}{\\u0410\\u0306}{\\u0430\\u0308}{\\u0410\\u0306}{\\u0433\\u0301}{\\u0413\\u0301}]", | |
1876 | "[\\u0474\\u0475\\u04d8\\u04d9\\u04e8\\u04e9]", | |
1877 | "[aAbB\\u0430\\u0410\\u0433\\u0413]" | |
1878 | }, | |
1879 | { "uk", | |
1880 | "[{\\u0474\\u030F}{\\u0475\\u030F}{\\u04D8\\u0308}{\\u04D9\\u0308}{\\u04E8\\u0308}{\\u04E9\\u0308}" | |
1881 | "{\\u0430\\u0306}{\\u0410\\u0306}{\\u0430\\u0308}{\\u0410\\u0306}{\\u0433\\u0301}{\\u0413\\u0301}]", | |
1882 | "[]", | |
1883 | "[\\u0474\\u0475\\u04D8\\u04D9\\u04E8\\u04E9\\u0430\\u0410\\u0433\\u0413]", | |
1884 | "[aAbBxv]", | |
1885 | }, | |
1886 | { "ja", | |
1887 | "[{\\u309d\\u3099}{\\u30fd\\u3099}]", | |
1888 | "[{lj}{nj}]", | |
1889 | "[\\u3099\\u309d\\u30fd]", | |
1890 | "[\\u30a6\\u3044\\uff73]" | |
1891 | }, | |
1892 | { "sh", | |
1893 | "[{C\\u0301}{C\\u030C}{C\\u0341}{DZ\\u030C}{Dz\\u030C}{D\\u017D}{D\\u017E}{lj}{nj}]", | |
1894 | "[{\\u309d\\u3099}{\\u30fd\\u3099}]", | |
1895 | "[nlcdzNLCDZ]", | |
1896 | "[jabv]" | |
1897 | } | |
1898 | }; | |
1899 | ||
1900 | ||
1901 | ||
1902 | ||
1903 | UErrorCode status = U_ZERO_ERROR; | |
1904 | UCollator *coll = NULL; | |
1905 | int32_t i = 0; | |
1906 | int32_t noConts = 0; | |
1907 | USet *conts = uset_open(0,0); | |
1908 | USet *set = uset_open(0,0); | |
1909 | UChar buffer[32768]; | |
1910 | int32_t setLen = 0; | |
1911 | ||
1912 | for(i = 0; i < sizeof(tests)/sizeof(tests[0]); i++) { | |
1913 | log_verbose("Testing locale: %s\n", tests[i].locale); | |
1914 | coll = ucol_open(tests[i].locale, &status); | |
1915 | noConts = ucol_getContractions(coll, conts, &status); | |
1916 | doSetsTest(conts, set, tests[i].inConts, tests[i].outConts, &status); | |
1917 | setLen = uset_toPattern(conts, buffer, 32768, TRUE, &status); | |
1918 | if(U_SUCCESS(status)) { | |
1919 | log_verbose("%i: %s\n", noConts, aescstrdup(buffer, setLen)); | |
1920 | } else { | |
1921 | log_err("error %s. %i\n", u_errorName(status), setLen); | |
1922 | } | |
1923 | ||
1924 | noConts = ucol_getUnsafeSet(coll, conts, &status); | |
1925 | doSetsTest(conts, set, tests[i].unsafeCodeUnits, tests[i].safeCodeUnits, &status); | |
1926 | ||
1927 | ucol_close(coll); | |
1928 | } | |
1929 | ||
1930 | ||
1931 | uset_close(conts); | |
1932 | uset_close(set); | |
1933 | } | |
1934 | ||
b75a7d8f | 1935 | #endif /* #if !UCONFIG_NO_COLLATION */ |