1 // © 2016 and later: Unicode, Inc. and others.
2 // License & terms of use: http://www.unicode.org/copyright.html
3 /********************************************************************
4 * Copyright (c) 1997-2016, International Business Machines
5 * Corporation and others. All Rights Reserved.
6 ********************************************************************/
7 /*****************************************************************************
8 *
9 * File CAPITEST.C
10 *
11 * Modification History:
12 * Name Description
13 * Madhu Katragadda Ported for C API
14 * Brian Rower Added TestOpenVsOpenRules
15 ******************************************************************************
16 *//* C API TEST For COLLATOR */
17
18 #include "unicode/utypes.h"
19
20 #if !UCONFIG_NO_COLLATION
21
22 #include <stdio.h>
23 #include <stdlib.h>
24 #include <string.h>
25 #include "unicode/uloc.h"
26 #include "unicode/ulocdata.h"
27 #include "unicode/ustring.h"
28 #include "unicode/ures.h"
29 #include "unicode/ucoleitr.h"
30 #include "cintltst.h"
31 #include "capitst.h"
32 #include "ccolltst.h"
33 #include "putilimp.h"
34 #include "cmemory.h"
35 #include "cstring.h"
36 #include "ucol_imp.h"
37
38 static void TestAttribute(void);
39 static void TestDefault(void);
40 static void TestDefaultKeyword(void);
41 static void TestBengaliSortKey(void);
42
43
ucol_sortKeyToString(const UCollator * coll,const uint8_t * sortkey,char * buffer,uint32_t len)44 static char* U_EXPORT2 ucol_sortKeyToString(const UCollator *coll, const uint8_t *sortkey, char *buffer, uint32_t len) {
45 (void)coll; // suppress compiler warnings about unused variable
46 uint32_t position = 0;
47 uint8_t b;
48
49 if (position + 1 < len)
50 position += sprintf(buffer + position, "[");
51 while ((b = *sortkey++) != 0) {
52 if (b == 1 && position + 5 < len) {
53 position += sprintf(buffer + position, "%02X . ", b);
54 } else if (b != 1 && position + 3 < len) {
55 position += sprintf(buffer + position, "%02X ", b);
56 }
57 }
58 if (position + 3 < len)
59 position += sprintf(buffer + position, "%02X]", b);
60 return buffer;
61 }
62
addCollAPITest(TestNode ** root)63 void addCollAPITest(TestNode** root)
64 {
65 /* WEIVTODO: return tests here */
66 addTest(root, &TestProperty, "tscoll/capitst/TestProperty");
67 addTest(root, &TestRuleBasedColl, "tscoll/capitst/TestRuleBasedColl");
68 addTest(root, &TestCompare, "tscoll/capitst/TestCompare");
69 addTest(root, &TestSortKey, "tscoll/capitst/TestSortKey");
70 addTest(root, &TestHashCode, "tscoll/capitst/TestHashCode");
71 addTest(root, &TestElemIter, "tscoll/capitst/TestElemIter");
72 addTest(root, &TestGetAll, "tscoll/capitst/TestGetAll");
73 /*addTest(root, &TestGetDefaultRules, "tscoll/capitst/TestGetDefaultRules");*/
74 addTest(root, &TestDecomposition, "tscoll/capitst/TestDecomposition");
75 addTest(root, &TestSafeClone, "tscoll/capitst/TestSafeClone");
76 addTest(root, &TestCloneBinary, "tscoll/capitst/TestCloneBinary");
77 addTest(root, &TestGetSetAttr, "tscoll/capitst/TestGetSetAttr");
78 addTest(root, &TestBounds, "tscoll/capitst/TestBounds");
79 addTest(root, &TestGetLocale, "tscoll/capitst/TestGetLocale");
80 addTest(root, &TestSortKeyBufferOverrun, "tscoll/capitst/TestSortKeyBufferOverrun");
81 addTest(root, &TestAttribute, "tscoll/capitst/TestAttribute");
82 addTest(root, &TestGetTailoredSet, "tscoll/capitst/TestGetTailoredSet");
83 addTest(root, &TestMergeSortKeys, "tscoll/capitst/TestMergeSortKeys");
84 addTest(root, &TestShortString, "tscoll/capitst/TestShortString");
85 addTest(root, &TestGetContractionsAndUnsafes, "tscoll/capitst/TestGetContractionsAndUnsafes");
86 addTest(root, &TestOpenBinary, "tscoll/capitst/TestOpenBinary");
87 addTest(root, &TestDefault, "tscoll/capitst/TestDefault");
88 addTest(root, &TestDefaultKeyword, "tscoll/capitst/TestDefaultKeyword");
89 addTest(root, &TestOpenVsOpenRules, "tscoll/capitst/TestOpenVsOpenRules");
90 addTest(root, &TestBengaliSortKey, "tscoll/capitst/TestBengaliSortKey");
91 addTest(root, &TestGetKeywordValuesForLocale, "tscoll/capitst/TestGetKeywordValuesForLocale");
92 addTest(root, &TestStrcollNull, "tscoll/capitst/TestStrcollNull");
93 }
94
TestGetSetAttr(void)95 void TestGetSetAttr(void) {
96 UErrorCode status = U_ZERO_ERROR;
97 UCollator *coll = ucol_open(NULL, &status);
98 struct attrTest {
99 UColAttribute att;
100 UColAttributeValue val[5];
101 uint32_t valueSize;
102 UColAttributeValue nonValue;
103 } attrs[] = {
104 {UCOL_FRENCH_COLLATION, {UCOL_ON, UCOL_OFF}, 2, UCOL_SHIFTED},
105 {UCOL_ALTERNATE_HANDLING, {UCOL_NON_IGNORABLE, UCOL_SHIFTED}, 2, UCOL_OFF},/* attribute for handling variable elements*/
106 {UCOL_CASE_FIRST, {UCOL_OFF, UCOL_LOWER_FIRST, UCOL_UPPER_FIRST}, 3, UCOL_SHIFTED},/* who goes first, lower case or uppercase */
107 {UCOL_CASE_LEVEL, {UCOL_ON, UCOL_OFF}, 2, UCOL_SHIFTED},/* do we have an extra case level */
108 {UCOL_NORMALIZATION_MODE, {UCOL_ON, UCOL_OFF}, 2, UCOL_SHIFTED},/* attribute for normalization */
109 {UCOL_DECOMPOSITION_MODE, {UCOL_ON, UCOL_OFF}, 2, UCOL_SHIFTED},
110 {UCOL_STRENGTH, {UCOL_PRIMARY, UCOL_SECONDARY, UCOL_TERTIARY, UCOL_QUATERNARY, UCOL_IDENTICAL}, 5, UCOL_SHIFTED},/* attribute for strength */
111 {UCOL_HIRAGANA_QUATERNARY_MODE, {UCOL_ON, UCOL_OFF}, 2, UCOL_SHIFTED},/* when turned on, this attribute */
112 };
113 UColAttribute currAttr;
114 UColAttributeValue value;
115 uint32_t i = 0, j = 0;
116
117 if (coll == NULL) {
118 log_err_status(status, "Unable to open collator. %s\n", u_errorName(status));
119 return;
120 }
121 for(i = 0; i<UPRV_LENGTHOF(attrs); i++) {
122 currAttr = attrs[i].att;
123 ucol_setAttribute(coll, currAttr, UCOL_DEFAULT, &status);
124 if(U_FAILURE(status)) {
125 log_err_status(status, "ucol_setAttribute with the default value returned error: %s\n", u_errorName(status));
126 break;
127 }
128 value = ucol_getAttribute(coll, currAttr, &status);
129 if(U_FAILURE(status)) {
130 log_err("ucol_getAttribute returned error: %s\n", u_errorName(status));
131 break;
132 }
133 for(j = 0; j<attrs[i].valueSize; j++) {
134 ucol_setAttribute(coll, currAttr, attrs[i].val[j], &status);
135 if(U_FAILURE(status)) {
136 log_err("ucol_setAttribute with the value %i returned error: %s\n", attrs[i].val[j], u_errorName(status));
137 break;
138 }
139 }
140 status = U_ZERO_ERROR;
141 ucol_setAttribute(coll, currAttr, attrs[i].nonValue, &status);
142 if(U_SUCCESS(status)) {
143 log_err("ucol_setAttribute with the bad value didn't return an error\n");
144 break;
145 }
146 status = U_ZERO_ERROR;
147
148 ucol_setAttribute(coll, currAttr, value, &status);
149 if(U_FAILURE(status)) {
150 log_err("ucol_setAttribute with the default valuereturned error: %s\n", u_errorName(status));
151 break;
152 }
153 }
154 status = U_ZERO_ERROR;
155 value = ucol_getAttribute(coll, UCOL_ATTRIBUTE_COUNT, &status);
156 if(U_SUCCESS(status)) {
157 log_err("ucol_getAttribute for UCOL_ATTRIBUTE_COUNT didn't return an error\n");
158 }
159 status = U_ZERO_ERROR;
160 ucol_setAttribute(coll, UCOL_ATTRIBUTE_COUNT, UCOL_DEFAULT, &status);
161 if(U_SUCCESS(status)) {
162 log_err("ucol_setAttribute for UCOL_ATTRIBUTE_COUNT didn't return an error\n");
163 }
164 status = U_ZERO_ERROR;
165 ucol_close(coll);
166 }
167
168
doAssert(int condition,const char * message)169 static void doAssert(int condition, const char *message)
170 {
171 if (condition==0) {
172 log_err("ERROR : %s\n", message);
173 }
174 }
175
176 #define UTF8_BUF_SIZE 128
177
doStrcoll(const UCollator * coll,const UChar * src,int32_t srcLen,const UChar * tgt,int32_t tgtLen,UCollationResult expected,const char * message)178 static void doStrcoll(const UCollator* coll, const UChar* src, int32_t srcLen, const UChar* tgt, int32_t tgtLen,
179 UCollationResult expected, const char *message) {
180 UErrorCode err = U_ZERO_ERROR;
181 char srcU8[UTF8_BUF_SIZE], tgtU8[UTF8_BUF_SIZE];
182 int32_t srcU8Len = -1, tgtU8Len = -1;
183 int32_t len = 0;
184
185 if (ucol_strcoll(coll, src, srcLen, tgt, tgtLen) != expected) {
186 log_err("ERROR : %s\n", message);
187 }
188
189 u_strToUTF8(srcU8, UTF8_BUF_SIZE, &len, src, srcLen, &err);
190 if (U_FAILURE(err) || len >= UTF8_BUF_SIZE) {
191 log_err("ERROR : UTF-8 conversion error\n");
192 return;
193 }
194 if (srcLen >= 0) {
195 srcU8Len = len;
196 }
197 u_strToUTF8(tgtU8, UTF8_BUF_SIZE, &len, tgt, tgtLen, &err);
198 if (U_FAILURE(err) || len >= UTF8_BUF_SIZE) {
199 log_err("ERROR : UTF-8 conversion error\n");
200 return;
201 }
202 if (tgtLen >= 0) {
203 tgtU8Len = len;
204 }
205
206 if (ucol_strcollUTF8(coll, srcU8, srcU8Len, tgtU8, tgtU8Len, &err) != expected
207 || U_FAILURE(err)) {
208 log_err("ERROR: %s (strcollUTF8)\n", message);
209 }
210 }
211
212 #if 0
213 /* We don't have default rules, at least not in the previous sense */
214 void TestGetDefaultRules(){
215 uint32_t size=0;
216 UErrorCode status=U_ZERO_ERROR;
217 UCollator *coll=NULL;
218 int32_t len1 = 0, len2=0;
219 uint8_t *binColData = NULL;
220
221 UResourceBundle *res = NULL;
222 UResourceBundle *binColl = NULL;
223 uint8_t *binResult = NULL;
224
225
226 const UChar * defaultRulesArray=ucol_getDefaultRulesArray(&size);
227 log_verbose("Test the function ucol_getDefaultRulesArray()\n");
228
229 coll = ucol_openRules(defaultRulesArray, size, UCOL_ON, UCOL_PRIMARY, &status);
230 if(U_SUCCESS(status) && coll !=NULL) {
231 binColData = (uint8_t*)ucol_cloneRuleData(coll, &len1, &status);
232
233 }
234
235
236 status=U_ZERO_ERROR;
237 res=ures_open(NULL, "root", &status);
238 if(U_FAILURE(status)){
239 log_err("ERROR: Failed to get resource for \"root Locale\" with %s", myErrorName(status));
240 return;
241 }
242 binColl=ures_getByKey(res, "%%Collation", binColl, &status);
243 if(U_SUCCESS(status)){
244 binResult=(uint8_t*)ures_getBinary(binColl, &len2, &status);
245 if(U_FAILURE(status)){
246 log_err("ERROR: ures_getBinary() failed\n");
247 }
248 }else{
249 log_err("ERROR: ures_getByKey(locale(default), %%Collation) failed");
250 }
251
252
253 if(len1 != len2){
254 log_err("Error: ucol_getDefaultRulesArray() failed to return the correct length.\n");
255 }
256 if(memcmp(binColData, binResult, len1) != 0){
257 log_err("Error: ucol_getDefaultRulesArray() failed\n");
258 }
259
260 free(binColData);
261 ures_close(binColl);
262 ures_close(res);
263 ucol_close(coll);
264
265 }
266 #endif
267
268 /* Collator Properties
269 ucol_open, ucol_strcoll, getStrength/setStrength
270 getDecomposition/setDecomposition, getDisplayName*/
TestProperty()271 void TestProperty()
272 {
273 UCollator *col, *ruled;
274 const UChar *rules;
275 UChar *disName;
276 int32_t len = 0;
277 UChar source[12], target[12];
278 int32_t tempLength;
279 UErrorCode status = U_ZERO_ERROR;
280 /*
281 * Expected version of the English collator.
282 * Currently, the major/minor version numbers change when the builder code
283 * changes,
284 * number 2 is from the tailoring data version and
285 * number 3 is the UCA version.
286 * This changes with every UCA version change, and the expected value
287 * needs to be adjusted.
288 * Same in intltest/apicoll.cpp.
289 */
290 UVersionInfo currVersionArray = {0x31, 0xC0, 0x05, 0x2A}; /* from ICU 4.4/UCA 5.2 */
291 UVersionInfo versionArray = {0, 0, 0, 0};
292 UVersionInfo versionUCAArray = {0, 0, 0, 0};
293 UVersionInfo versionUCDArray = {0, 0, 0, 0};
294
295 log_verbose("The property tests begin : \n");
296 log_verbose("Test ucol_strcoll : \n");
297 col = ucol_open("en_US", &status);
298 if (U_FAILURE(status)) {
299 log_err_status(status, "Default Collator creation failed.: %s\n", myErrorName(status));
300 return;
301 }
302
303 ucol_getVersion(col, versionArray);
304 /* Check for a version greater than some value rather than equality
305 * so that we need not update the expected version each time. */
306 if (uprv_memcmp(versionArray, currVersionArray, 4)<0) {
307 log_err("Testing ucol_getVersion() - unexpected result: %02x.%02x.%02x.%02x\n",
308 versionArray[0], versionArray[1], versionArray[2], versionArray[3]);
309 } else {
310 log_verbose("ucol_getVersion() result: %02x.%02x.%02x.%02x\n",
311 versionArray[0], versionArray[1], versionArray[2], versionArray[3]);
312 }
313
314 /* Assume that the UCD and UCA versions are the same,
315 * rather than hardcoding (and updating each time) a particular UCA version. */
316 u_getUnicodeVersion(versionUCDArray);
317 ucol_getUCAVersion(col, versionUCAArray);
318 if (0!=uprv_memcmp(versionUCAArray, versionUCDArray, 4)) {
319 log_err("Testing ucol_getUCAVersion() - unexpected result: %hu.%hu.%hu.%hu\n",
320 versionUCAArray[0], versionUCAArray[1], versionUCAArray[2], versionUCAArray[3]);
321 }
322
323 u_uastrcpy(source, "ab");
324 u_uastrcpy(target, "abc");
325
326 doStrcoll(col, source, u_strlen(source), target, u_strlen(target), UCOL_LESS, "ab < abc comparison failed");
327
328 u_uastrcpy(source, "ab");
329 u_uastrcpy(target, "AB");
330
331 doStrcoll(col, source, u_strlen(source), target, u_strlen(target), UCOL_LESS, "ab < AB comparison failed");
332
333 u_uastrcpy(source, "blackbird");
334 u_uastrcpy(target, "black-bird");
335
336 doStrcoll(col, source, u_strlen(source), target, u_strlen(target), UCOL_GREATER, "black-bird > blackbird comparison failed");
337
338 u_uastrcpy(source, "black bird");
339 u_uastrcpy(target, "black-bird");
340
341 doStrcoll(col, source, u_strlen(source), target, u_strlen(target), UCOL_LESS, "black bird < black-bird comparison failed");
342
343 u_uastrcpy(source, "Hello");
344 u_uastrcpy(target, "hello");
345
346 doStrcoll(col, source, u_strlen(source), target, u_strlen(target), UCOL_GREATER, "Hello > hello comparison failed");
347
348 log_verbose("Test ucol_strcoll ends.\n");
349
350 log_verbose("testing ucol_getStrength() method ...\n");
351 doAssert( (ucol_getStrength(col) == UCOL_TERTIARY), "collation object has the wrong strength");
352 doAssert( (ucol_getStrength(col) != UCOL_PRIMARY), "collation object's strength is primary difference");
353
354 log_verbose("testing ucol_setStrength() method ...\n");
355 ucol_setStrength(col, UCOL_SECONDARY);
356 doAssert( (ucol_getStrength(col) != UCOL_TERTIARY), "collation object's strength is secondary difference");
357 doAssert( (ucol_getStrength(col) != UCOL_PRIMARY), "collation object's strength is primary difference");
358 doAssert( (ucol_getStrength(col) == UCOL_SECONDARY), "collation object has the wrong strength");
359
360
361 log_verbose("Get display name for the default collation in German : \n");
362
363 len=ucol_getDisplayName("en_US", "de_DE", NULL, 0, &status);
364 if(status==U_BUFFER_OVERFLOW_ERROR){
365 status=U_ZERO_ERROR;
366 disName=(UChar*)malloc(sizeof(UChar) * (len+1));
367 ucol_getDisplayName("en_US", "de_DE", disName, len+1, &status);
368 log_verbose("the display name for default collation in german: %s\n", austrdup(disName) );
369 free(disName);
370 }
371 if(U_FAILURE(status)){
372 log_err("ERROR: in getDisplayName: %s\n", myErrorName(status));
373 return;
374 }
375 log_verbose("Default collation getDisplayName ended.\n");
376
377 ruled = ucol_open("da_DK", &status);
378 if(U_FAILURE(status)) {
379 log_data_err("ucol_open(\"da_DK\") failed - %s\n", u_errorName(status));
380 ucol_close(col);
381 return;
382 }
383 log_verbose("ucol_getRules() testing ...\n");
384 rules = ucol_getRules(ruled, &tempLength);
385 if(tempLength == 0) {
386 log_data_err("missing da_DK tailoring rule string\n");
387 } else {
388 UChar aa[2] = { 0x61, 0x61 };
389 doAssert(u_strFindFirst(rules, tempLength, aa, 2) != NULL,
390 "da_DK rules do not contain 'aa'");
391 }
392 log_verbose("getRules tests end.\n");
393 {
394 UChar *buffer = (UChar *)malloc(200000*sizeof(UChar));
395 int32_t bufLen = 200000;
396 buffer[0] = '\0';
397 log_verbose("ucol_getRulesEx() testing ...\n");
398 tempLength = ucol_getRulesEx(col,UCOL_TAILORING_ONLY,buffer,bufLen );
399 doAssert( tempLength == 0x00, "getRulesEx() result incorrect" );
400 log_verbose("getRules tests end.\n");
401
402 log_verbose("ucol_getRulesEx() testing ...\n");
403 tempLength=ucol_getRulesEx(col,UCOL_FULL_RULES,buffer,bufLen );
404 if(tempLength == 0) {
405 log_data_err("missing *full* rule string\n");
406 }
407 log_verbose("getRulesEx tests end.\n");
408 free(buffer);
409 }
410 ucol_close(ruled);
411 ucol_close(col);
412
413 log_verbose("open an collator for french locale");
414 col = ucol_open("fr_FR", &status);
415 if (U_FAILURE(status)) {
416 log_err("ERROR: Creating French collation failed.: %s\n", myErrorName(status));
417 return;
418 }
419 ucol_setStrength(col, UCOL_PRIMARY);
420 log_verbose("testing ucol_getStrength() method again ...\n");
421 doAssert( (ucol_getStrength(col) != UCOL_TERTIARY), "collation object has the wrong strength");
422 doAssert( (ucol_getStrength(col) == UCOL_PRIMARY), "collation object's strength is not primary difference");
423
424 log_verbose("testing French ucol_setStrength() method ...\n");
425 ucol_setStrength(col, UCOL_TERTIARY);
426 doAssert( (ucol_getStrength(col) == UCOL_TERTIARY), "collation object's strength is not tertiary difference");
427 doAssert( (ucol_getStrength(col) != UCOL_PRIMARY), "collation object's strength is primary difference");
428 doAssert( (ucol_getStrength(col) != UCOL_SECONDARY), "collation object's strength is secondary difference");
429 ucol_close(col);
430
431 log_verbose("Get display name for the french collation in english : \n");
432 len=ucol_getDisplayName("fr_FR", "en_US", NULL, 0, &status);
433 if(status==U_BUFFER_OVERFLOW_ERROR){
434 status=U_ZERO_ERROR;
435 disName=(UChar*)malloc(sizeof(UChar) * (len+1));
436 ucol_getDisplayName("fr_FR", "en_US", disName, len+1, &status);
437 log_verbose("the display name for french collation in english: %s\n", austrdup(disName) );
438 free(disName);
439 }
440 if(U_FAILURE(status)){
441 log_err("ERROR: in getDisplayName: %s\n", myErrorName(status));
442 return;
443 }
444 log_verbose("Default collation getDisplayName ended.\n");
445
446 }
447
448 /* Test RuleBasedCollator and getRules*/
TestRuleBasedColl()449 void TestRuleBasedColl()
450 {
451 UCollator *col1, *col2, *col3, *col4;
452 UCollationElements *iter1, *iter2;
453 UChar ruleset1[60];
454 UChar ruleset2[50];
455 UChar teststr[10];
456 const UChar *rule1, *rule2, *rule3, *rule4;
457 int32_t tempLength;
458 UErrorCode status = U_ZERO_ERROR;
459 u_uastrcpy(ruleset1, "&9 < a, A < b, B < c, C; ch, cH, Ch, CH < d, D, e, E");
460 u_uastrcpy(ruleset2, "&9 < a, A < b, B < c, C < d, D, e, E");
461
462
463 col1 = ucol_openRules(ruleset1, u_strlen(ruleset1), UCOL_DEFAULT, UCOL_DEFAULT_STRENGTH, NULL,&status);
464 if (U_FAILURE(status)) {
465 log_err_status(status, "RuleBased Collator creation failed.: %s\n", myErrorName(status));
466 return;
467 }
468 else
469 log_verbose("PASS: RuleBased Collator creation passed\n");
470
471 status = U_ZERO_ERROR;
472 col2 = ucol_openRules(ruleset2, u_strlen(ruleset2), UCOL_DEFAULT, UCOL_DEFAULT_STRENGTH, NULL, &status);
473 if (U_FAILURE(status)) {
474 log_err("RuleBased Collator creation failed.: %s\n", myErrorName(status));
475 return;
476 }
477 else
478 log_verbose("PASS: RuleBased Collator creation passed\n");
479
480
481 status = U_ZERO_ERROR;
482 col3= ucol_open(NULL, &status);
483 if (U_FAILURE(status)) {
484 log_err("Default Collator creation failed.: %s\n", myErrorName(status));
485 return;
486 }
487 else
488 log_verbose("PASS: Default Collator creation passed\n");
489
490 rule1 = ucol_getRules(col1, &tempLength);
491 rule2 = ucol_getRules(col2, &tempLength);
492 rule3 = ucol_getRules(col3, &tempLength);
493
494 doAssert((u_strcmp(rule1, rule2) != 0), "Default collator getRules failed");
495 doAssert((u_strcmp(rule2, rule3) != 0), "Default collator getRules failed");
496 doAssert((u_strcmp(rule1, rule3) != 0), "Default collator getRules failed");
497
498 col4=ucol_openRules(rule2, u_strlen(rule2), UCOL_DEFAULT, UCOL_DEFAULT_STRENGTH, NULL, &status);
499 if (U_FAILURE(status)) {
500 log_err("RuleBased Collator creation failed.: %s\n", myErrorName(status));
501 return;
502 }
503 rule4= ucol_getRules(col4, &tempLength);
504 doAssert((u_strcmp(rule2, rule4) == 0), "Default collator getRules failed");
505
506 ucol_close(col1);
507 ucol_close(col2);
508 ucol_close(col3);
509 ucol_close(col4);
510
511 /* tests that modifier ! is always ignored */
512 u_uastrcpy(ruleset1, "!&a<b");
513 teststr[0] = 0x0e40;
514 teststr[1] = 0x0e01;
515 teststr[2] = 0x0e2d;
516 col1 = ucol_openRules(ruleset1, u_strlen(ruleset1), UCOL_DEFAULT, UCOL_DEFAULT_STRENGTH, NULL, &status);
517 if (U_FAILURE(status)) {
518 log_err("RuleBased Collator creation failed.: %s\n", myErrorName(status));
519 return;
520 }
521 col2 = ucol_open("en_US", &status);
522 if (U_FAILURE(status)) {
523 log_err("en_US Collator creation failed.: %s\n", myErrorName(status));
524 return;
525 }
526 iter1 = ucol_openElements(col1, teststr, 3, &status);
527 iter2 = ucol_openElements(col2, teststr, 3, &status);
528 if(U_FAILURE(status)) {
529 log_err("ERROR: CollationElement iterator creation failed.: %s\n", myErrorName(status));
530 return;
531 }
532 while (TRUE) {
533 /* testing with en since thai has its own tailoring */
534 int32_t ce = ucol_next(iter1, &status);
535 int32_t ce2 = ucol_next(iter2, &status);
536 if(U_FAILURE(status)) {
537 log_err("ERROR: CollationElement iterator creation failed.: %s\n", myErrorName(status));
538 return;
539 }
540 if (ce2 != ce) {
541 log_err("! modifier test failed");
542 }
543 if (ce == UCOL_NULLORDER) {
544 break;
545 }
546 }
547 ucol_closeElements(iter1);
548 ucol_closeElements(iter2);
549 ucol_close(col1);
550 ucol_close(col2);
551 /* CLDR 24+ requires a reset before the first relation */
552 u_uastrcpy(ruleset1, "< z < a");
553 col1 = ucol_openRules(ruleset1, u_strlen(ruleset1), UCOL_DEFAULT, UCOL_DEFAULT_STRENGTH, NULL, &status);
554 if (status != U_PARSE_ERROR && status != U_INVALID_FORMAT_ERROR) {
555 log_err("ucol_openRules(without initial reset: '< z < a') "
556 "should fail with U_PARSE_ERROR or U_INVALID_FORMAT_ERROR but yielded %s\n",
557 myErrorName(status));
558 }
559 ucol_close(col1);
560 }
561
TestCompare()562 void TestCompare()
563 {
564 UErrorCode status = U_ZERO_ERROR;
565 UCollator *col;
566 UChar* test1;
567 UChar* test2;
568
569 log_verbose("The compare tests begin : \n");
570 status=U_ZERO_ERROR;
571 col = ucol_open("en_US", &status);
572 if(U_FAILURE(status)) {
573 log_err_status(status, "ucal_open() collation creation failed.: %s\n", myErrorName(status));
574 return;
575 }
576 test1=(UChar*)malloc(sizeof(UChar) * 6);
577 test2=(UChar*)malloc(sizeof(UChar) * 6);
578 u_uastrcpy(test1, "Abcda");
579 u_uastrcpy(test2, "abcda");
580
581 log_verbose("Use tertiary comparison level testing ....\n");
582
583 doAssert( (!ucol_equal(col, test1, u_strlen(test1), test2, u_strlen(test2))), "Result should be \"Abcda\" != \"abcda\" ");
584 doAssert( (ucol_greater(col, test1, u_strlen(test1), test2, u_strlen(test2))), "Result should be \"Abcda\" >>> \"abcda\" ");
585 doAssert( (ucol_greaterOrEqual(col, test1, u_strlen(test1), test2, u_strlen(test2))), "Result should be \"Abcda\" >>> \"abcda\"");
586
587 ucol_setStrength(col, UCOL_SECONDARY);
588 log_verbose("Use secondary comparison level testing ....\n");
589
590 doAssert( (ucol_equal(col, test1, u_strlen(test1), test2, u_strlen(test2) )), "Result should be \"Abcda\" == \"abcda\"");
591 doAssert( (!ucol_greater(col, test1, u_strlen(test1), test2, u_strlen(test2))), "Result should be \"Abcda\" == \"abcda\"");
592 doAssert( (ucol_greaterOrEqual(col, test1, u_strlen(test1), test2, u_strlen(test2) )), "Result should be \"Abcda\" == \"abcda\"");
593
594 ucol_setStrength(col, UCOL_PRIMARY);
595 log_verbose("Use primary comparison level testing ....\n");
596
597 doAssert( (ucol_equal(col, test1, u_strlen(test1), test2, u_strlen(test2))), "Result should be \"Abcda\" == \"abcda\"");
598 doAssert( (!ucol_greater(col, test1, u_strlen(test1), test2, u_strlen(test2))), "Result should be \"Abcda\" == \"abcda\"");
599 doAssert( (ucol_greaterOrEqual(col, test1, u_strlen(test1), test2, u_strlen(test2))), "Result should be \"Abcda\" == \"abcda\"");
600
601
602 log_verbose("The compare tests end.\n");
603 ucol_close(col);
604 free(test1);
605 free(test2);
606
607 }
608 /*
609 ---------------------------------------------
610 tests decomposition setting
611 */
TestDecomposition()612 void TestDecomposition() {
613 UErrorCode status = U_ZERO_ERROR;
614 UCollator *en_US, *el_GR, *vi_VN;
615 en_US = ucol_open("en_US", &status);
616 el_GR = ucol_open("el_GR", &status);
617 vi_VN = ucol_open("vi_VN", &status);
618
619 if (U_FAILURE(status)) {
620 log_err_status(status, "ERROR: collation creation failed.: %s\n", myErrorName(status));
621 return;
622 }
623
624 if (ucol_getAttribute(vi_VN, UCOL_NORMALIZATION_MODE, &status) != UCOL_ON ||
625 U_FAILURE(status))
626 {
627 log_err("ERROR: vi_VN collation did not have canonical decomposition for normalization!\n");
628 }
629
630 status = U_ZERO_ERROR;
631 if (ucol_getAttribute(el_GR, UCOL_NORMALIZATION_MODE, &status) != UCOL_ON ||
632 U_FAILURE(status))
633 {
634 log_err("ERROR: el_GR collation did not have canonical decomposition for normalization!\n");
635 }
636
637 status = U_ZERO_ERROR;
638 if (ucol_getAttribute(en_US, UCOL_NORMALIZATION_MODE, &status) != UCOL_OFF ||
639 U_FAILURE(status))
640 {
641 log_err("ERROR: en_US collation had canonical decomposition for normalization!\n");
642 }
643
644 ucol_close(en_US);
645 ucol_close(el_GR);
646 ucol_close(vi_VN);
647 }
648
649 #define CLONETEST_COLLATOR_COUNT 4
650
TestSafeClone()651 void TestSafeClone() {
652 UChar test1[6];
653 UChar test2[6];
654 static const UChar umlautUStr[] = {0x00DC, 0};
655 static const UChar oeStr[] = {0x0055, 0x0045, 0};
656 UCollator * someCollators [CLONETEST_COLLATOR_COUNT];
657 UCollator * someClonedCollators [CLONETEST_COLLATOR_COUNT];
658 UCollator * col;
659 UErrorCode err = U_ZERO_ERROR;
660 int8_t idx = 6; /* Leave this here to test buffer alingment in memory*/
661 uint8_t buffer [CLONETEST_COLLATOR_COUNT] [U_COL_SAFECLONE_BUFFERSIZE];
662 int32_t bufferSize = U_COL_SAFECLONE_BUFFERSIZE;
663 const char sampleRuleChars[] = "&Z < CH";
664 UChar sampleRule[sizeof(sampleRuleChars)];
665
666 u_uastrcpy(test1, "abCda");
667 u_uastrcpy(test2, "abcda");
668 u_uastrcpy(sampleRule, sampleRuleChars);
669
670 /* one default collator & two complex ones */
671 someCollators[0] = ucol_open("en_US", &err);
672 someCollators[1] = ucol_open("ko", &err);
673 someCollators[2] = ucol_open("ja_JP", &err);
674 someCollators[3] = ucol_openRules(sampleRule, -1, UCOL_ON, UCOL_TERTIARY, NULL, &err);
675 if(U_FAILURE(err)) {
676 for (idx = 0; idx < CLONETEST_COLLATOR_COUNT; idx++) {
677 ucol_close(someCollators[idx]);
678 }
679 log_data_err("Couldn't open one or more collators\n");
680 return;
681 }
682
683 /* Check the various error & informational states: */
684
685 /* Null status - just returns NULL */
686 if (NULL != ucol_safeClone(someCollators[0], buffer[0], &bufferSize, NULL))
687 {
688 log_err("FAIL: Cloned Collator failed to deal correctly with null status\n");
689 }
690 /* error status - should return 0 & keep error the same */
691 err = U_MEMORY_ALLOCATION_ERROR;
692 if (NULL != ucol_safeClone(someCollators[0], buffer[0], &bufferSize, &err) || err != U_MEMORY_ALLOCATION_ERROR)
693 {
694 log_err("FAIL: Cloned Collator failed to deal correctly with incoming error status\n");
695 }
696 err = U_ZERO_ERROR;
697
698 /* Null buffer size pointer is ok */
699 if (NULL == (col = ucol_safeClone(someCollators[0], buffer[0], NULL, &err)) || U_FAILURE(err))
700 {
701 log_err("FAIL: Cloned Collator failed to deal correctly with null bufferSize pointer\n");
702 }
703 ucol_close(col);
704 err = U_ZERO_ERROR;
705
706 /* buffer size pointer is 0 - fill in pbufferSize with a size */
707 bufferSize = 0;
708 if (NULL != ucol_safeClone(someCollators[0], buffer[0], &bufferSize, &err) ||
709 U_FAILURE(err) || bufferSize <= 0)
710 {
711 log_err("FAIL: Cloned Collator failed a sizing request ('preflighting')\n");
712 }
713 /* Verify our define is large enough */
714 if (U_COL_SAFECLONE_BUFFERSIZE < bufferSize)
715 {
716 log_err("FAIL: Pre-calculated buffer size is too small\n");
717 }
718 /* Verify we can use this run-time calculated size */
719 if (NULL == (col = ucol_safeClone(someCollators[0], buffer[0], &bufferSize, &err)) || U_FAILURE(err))
720 {
721 log_err("FAIL: Collator can't be cloned with run-time size\n");
722 }
723 if (col) ucol_close(col);
724 /* size one byte too small - should allocate & let us know */
725 if (bufferSize > 1) {
726 --bufferSize;
727 }
728 if (NULL == (col = ucol_safeClone(someCollators[0], 0, &bufferSize, &err)) || err != U_SAFECLONE_ALLOCATED_WARNING)
729 {
730 log_err("FAIL: Cloned Collator failed to deal correctly with too-small buffer size\n");
731 }
732 if (col) ucol_close(col);
733 err = U_ZERO_ERROR;
734 bufferSize = U_COL_SAFECLONE_BUFFERSIZE;
735
736
737 /* Null buffer pointer - return Collator & set error to U_SAFECLONE_ALLOCATED_ERROR */
738 if (NULL == (col = ucol_safeClone(someCollators[0], 0, &bufferSize, &err)) || err != U_SAFECLONE_ALLOCATED_WARNING)
739 {
740 log_err("FAIL: Cloned Collator failed to deal correctly with null buffer pointer\n");
741 }
742 if (col) ucol_close(col);
743 err = U_ZERO_ERROR;
744
745 /* Null Collator - return NULL & set U_ILLEGAL_ARGUMENT_ERROR */
746 if (NULL != ucol_safeClone(NULL, buffer[0], &bufferSize, &err) || err != U_ILLEGAL_ARGUMENT_ERROR)
747 {
748 log_err("FAIL: Cloned Collator failed to deal correctly with null Collator pointer\n");
749 }
750
751 err = U_ZERO_ERROR;
752
753 /* Test that a cloned collator doesn't accidentally use UCA. */
754 col=ucol_open("de@collation=phonebook", &err);
755 bufferSize = U_COL_SAFECLONE_BUFFERSIZE;
756 someClonedCollators[0] = ucol_safeClone(col, buffer[0], &bufferSize, &err);
757 doAssert( (ucol_greater(col, umlautUStr, u_strlen(umlautUStr), oeStr, u_strlen(oeStr))), "Original German phonebook collation sorts differently than expected");
758 doAssert( (ucol_greater(someClonedCollators[0], umlautUStr, u_strlen(umlautUStr), oeStr, u_strlen(oeStr))), "Cloned German phonebook collation sorts differently than expected");
759 if (!ucol_equals(someClonedCollators[0], col)) {
760 log_err("FAIL: Cloned German phonebook collator is not equal to original.\n");
761 }
762 ucol_close(col);
763 ucol_close(someClonedCollators[0]);
764
765 err = U_ZERO_ERROR;
766
767 /* change orig & clone & make sure they are independent */
768
769 for (idx = 0; idx < CLONETEST_COLLATOR_COUNT; idx++)
770 {
771 ucol_setStrength(someCollators[idx], UCOL_IDENTICAL);
772 bufferSize = 1;
773 err = U_ZERO_ERROR;
774 ucol_close(ucol_safeClone(someCollators[idx], buffer[idx], &bufferSize, &err));
775 if (err != U_SAFECLONE_ALLOCATED_WARNING) {
776 log_err("FAIL: collator number %d was not allocated.\n", idx);
777 log_err("FAIL: status of Collator[%d] is %d (hex: %x).\n", idx, err, err);
778 }
779
780 bufferSize = U_COL_SAFECLONE_BUFFERSIZE;
781 err = U_ZERO_ERROR;
782 someClonedCollators[idx] = ucol_safeClone(someCollators[idx], buffer[idx], &bufferSize, &err);
783 if (U_FAILURE(err)) {
784 log_err("FAIL: Unable to clone collator %d - %s\n", idx, u_errorName(err));
785 continue;
786 }
787 if (!ucol_equals(someClonedCollators[idx], someCollators[idx])) {
788 log_err("FAIL: Cloned collator is not equal to original at index = %d.\n", idx);
789 }
790
791 /* Check the usability */
792 ucol_setStrength(someCollators[idx], UCOL_PRIMARY);
793 ucol_setAttribute(someCollators[idx], UCOL_CASE_LEVEL, UCOL_OFF, &err);
794
795 doAssert( (ucol_equal(someCollators[idx], test1, u_strlen(test1), test2, u_strlen(test2))), "Result should be \"abcda\" == \"abCda\"");
796
797 /* Close the original to make sure that the clone is usable. */
798 ucol_close(someCollators[idx]);
799
800 ucol_setStrength(someClonedCollators[idx], UCOL_TERTIARY);
801 ucol_setAttribute(someClonedCollators[idx], UCOL_CASE_LEVEL, UCOL_OFF, &err);
802 doAssert( (ucol_greater(someClonedCollators[idx], test1, u_strlen(test1), test2, u_strlen(test2))), "Result should be \"abCda\" >>> \"abcda\" ");
803
804 ucol_close(someClonedCollators[idx]);
805 }
806 }
807
TestCloneBinary()808 void TestCloneBinary(){
809 UErrorCode err = U_ZERO_ERROR;
810 UCollator * col = ucol_open("en_US", &err);
811 UCollator * c;
812 int32_t size;
813 uint8_t * buffer;
814
815 if (U_FAILURE(err)) {
816 log_data_err("Couldn't open collator. Error: %s\n", u_errorName(err));
817 return;
818 }
819
820 size = ucol_cloneBinary(col, NULL, 0, &err);
821 if(size==0 || err!=U_BUFFER_OVERFLOW_ERROR) {
822 log_err("ucol_cloneBinary - couldn't check size. Error: %s\n", u_errorName(err));
823 return;
824 }
825 err = U_ZERO_ERROR;
826
827 buffer = (uint8_t *) malloc(size);
828 ucol_cloneBinary(col, buffer, size, &err);
829 if(U_FAILURE(err)) {
830 log_err("ucol_cloneBinary - couldn't clone.. Error: %s\n", u_errorName(err));
831 free(buffer);
832 return;
833 }
834
835 /* how to check binary result ? */
836
837 c = ucol_openBinary(buffer, size, col, &err);
838 if(U_FAILURE(err)) {
839 log_err("ucol_openBinary failed. Error: %s\n", u_errorName(err));
840 } else {
841 UChar t[] = {0x41, 0x42, 0x43, 0}; /* ABC */
842 uint8_t *k1, *k2;
843 int l1, l2;
844 l1 = ucol_getSortKey(col, t, -1, NULL,0);
845 l2 = ucol_getSortKey(c, t, -1, NULL,0);
846 k1 = (uint8_t *) malloc(sizeof(uint8_t) * l1);
847 k2 = (uint8_t *) malloc(sizeof(uint8_t) * l2);
848 ucol_getSortKey(col, t, -1, k1, l1);
849 ucol_getSortKey(col, t, -1, k2, l2);
850 if (strcmp((char *)k1,(char *)k2) != 0){
851 log_err("ucol_openBinary - new collator should equal to old one\n");
852 }
853 free(k1);
854 free(k2);
855 }
856 free(buffer);
857 ucol_close(c);
858 ucol_close(col);
859 }
860
861
TestBengaliSortKey(void)862 static void TestBengaliSortKey(void)
863 {
864 const char *curLoc = "bn";
865 UChar str1[] = { 0x09BE, 0 };
866 UChar str2[] = { 0x0B70, 0 };
867 UCollator *c2 = NULL;
868 const UChar *rules;
869 int32_t rulesLength=-1;
870 uint8_t *sortKey1;
871 int32_t sortKeyLen1 = 0;
872 uint8_t *sortKey2;
873 int32_t sortKeyLen2 = 0;
874 UErrorCode status = U_ZERO_ERROR;
875 char sortKeyStr1[2048];
876 uint32_t sortKeyStrLen1 = UPRV_LENGTHOF(sortKeyStr1);
877 char sortKeyStr2[2048];
878 uint32_t sortKeyStrLen2 = UPRV_LENGTHOF(sortKeyStr2);
879 UCollationResult result;
880
881 static UChar preRules[41] = { 0x26, 0x9fa, 0x3c, 0x98c, 0x3c, 0x9e1, 0x3c, 0x98f, 0x3c, 0x990, 0x3c, 0x993, 0x3c, 0x994, 0x3c, 0x9bc, 0x3c, 0x982, 0x3c, 0x983, 0x3c, 0x981, 0x3c, 0x9b0, 0x3c, 0x9b8, 0x3c, 0x9b9, 0x3c, 0x9bd, 0x3c, 0x9be, 0x3c, 0x9bf, 0x3c, 0x9c8, 0x3c, 0x9cb, 0x3d, 0x9cb , 0};
882
883 rules = preRules;
884
885 log_verbose("Rules: %s\n", aescstrdup(rules, rulesLength));
886
887 c2 = ucol_openRules(rules, rulesLength, UCOL_DEFAULT, UCOL_DEFAULT_STRENGTH, NULL, &status);
888 if (U_FAILURE(status)) {
889 log_data_err("ERROR: Creating collator from rules failed with locale: %s : %s\n", curLoc, myErrorName(status));
890 return;
891 }
892
893 sortKeyLen1 = ucol_getSortKey(c2, str1, -1, NULL, 0);
894 sortKey1 = (uint8_t*)malloc(sortKeyLen1+1);
895 ucol_getSortKey(c2,str1,-1,sortKey1, sortKeyLen1+1);
896 ucol_sortKeyToString(c2, sortKey1, sortKeyStr1, sortKeyStrLen1);
897
898
899 sortKeyLen2 = ucol_getSortKey(c2, str2, -1, NULL, 0);
900 sortKey2 = (uint8_t*)malloc(sortKeyLen2+1);
901 ucol_getSortKey(c2,str2,-1,sortKey2, sortKeyLen2+1);
902
903 ucol_sortKeyToString(c2, sortKey2, sortKeyStr2, sortKeyStrLen2);
904
905
906
907 result=ucol_strcoll(c2, str1, -1, str2, -1);
908 if(result!=UCOL_LESS) {
909 log_err("Error: %s was not less than %s: result=%d.\n", aescstrdup(str1,-1), aescstrdup(str2,-1), result);
910 log_info("[%s] -> %s (%d, from rule)\n", aescstrdup(str1,-1), sortKeyStr1, sortKeyLen1);
911 log_info("[%s] -> %s (%d, from rule)\n", aescstrdup(str2,-1), sortKeyStr2, sortKeyLen2);
912 } else {
913 log_verbose("OK: %s was less than %s: result=%d.\n", aescstrdup(str1,-1), aescstrdup(str2,-1), result);
914 log_verbose("[%s] -> %s (%d, from rule)\n", aescstrdup(str1,-1), sortKeyStr1, sortKeyLen1);
915 log_verbose("[%s] -> %s (%d, from rule)\n", aescstrdup(str2,-1), sortKeyStr2, sortKeyLen2);
916 }
917
918 free(sortKey1);
919 free(sortKey2);
920 ucol_close(c2);
921
922 }
923
924 /*
925 TestOpenVsOpenRules ensures that collators from ucol_open and ucol_openRules
926 will generate identical sort keys
927 */
TestOpenVsOpenRules()928 void TestOpenVsOpenRules(){
929
930 /* create an array of all the locales */
931 int32_t numLocales = uloc_countAvailable();
932 int32_t sizeOfStdSet;
933 uint32_t adder;
934 UChar str[41]; /* create an array of UChar of size maximum strSize + 1 */
935 USet *stdSet;
936 char* curLoc;
937 UCollator * c1;
938 UCollator * c2;
939 const UChar* rules;
940 int32_t rulesLength;
941 int32_t sortKeyLen1, sortKeyLen2;
942 uint8_t *sortKey1 = NULL, *sortKey2 = NULL;
943 char sortKeyStr1[512], sortKeyStr2[512];
944 uint32_t sortKeyStrLen1 = UPRV_LENGTHOF(sortKeyStr1),
945 sortKeyStrLen2 = UPRV_LENGTHOF(sortKeyStr2);
946 ULocaleData *uld;
947 int32_t x, y, z;
948 USet *eSet;
949 int32_t eSize;
950 int strSize;
951
952 UErrorCode err = U_ZERO_ERROR;
953
954 /* create a set of standard characters that aren't very interesting...
955 and then we can find some interesting ones later */
956
957 stdSet = uset_open(0x61, 0x7A);
958 uset_addRange(stdSet, 0x41, 0x5A);
959 uset_addRange(stdSet, 0x30, 0x39);
960 sizeOfStdSet = uset_size(stdSet);
961 (void)sizeOfStdSet; /* Suppress set but not used warning. */
962
963 adder = 1;
964 if(getTestOption(QUICK_OPTION))
965 {
966 adder = 10;
967 }
968
969 for(x = 0; x < numLocales; x+=adder){
970 curLoc = (char *)uloc_getAvailable(x);
971 log_verbose("Processing %s\n", curLoc);
972
973 /* create a collator the normal API way */
974 c1 = ucol_open(curLoc, &err);
975 if (U_FAILURE(err)) {
976 log_err("ERROR: Normal collation creation failed with locale: %s : %s\n", curLoc, myErrorName(err));
977 return;
978 }
979
980 /* grab the rules */
981 rules = ucol_getRules(c1, &rulesLength);
982 if (rulesLength == 0) {
983 /* The optional tailoring rule string is either empty (boring) or missing. */
984 ucol_close(c1);
985 continue;
986 }
987
988 /* use those rules to create a collator from rules */
989 c2 = ucol_openRules(rules, rulesLength, UCOL_DEFAULT, UCOL_DEFAULT_STRENGTH, NULL, &err);
990 if (U_FAILURE(err)) {
991 log_err("ERROR: Creating collator from rules failed with locale: %s : %s\n", curLoc, myErrorName(err));
992 ucol_close(c1);
993 continue;
994 }
995
996 uld = ulocdata_open(curLoc, &err);
997
998 /*now that we have some collators, we get several strings */
999
1000 for(y = 0; y < 5; y++){
1001
1002 /* get a set of ALL the characters in this locale */
1003 eSet = ulocdata_getExemplarSet(uld, NULL, 0, ULOCDATA_ES_STANDARD, &err);
1004 eSize = uset_size(eSet);
1005
1006 /* make a string with these characters in it */
1007 strSize = (rand()%40) + 1;
1008
1009 for(z = 0; z < strSize; z++){
1010 str[z] = uset_charAt(eSet, rand()%eSize);
1011 }
1012
1013 /* change the set to only include 'abnormal' characters (not A-Z, a-z, 0-9 */
1014 uset_removeAll(eSet, stdSet);
1015 eSize = uset_size(eSet);
1016
1017 /* if there are some non-normal characters left, put a few into the string, just to make sure we have some */
1018 if(eSize > 0){
1019 str[2%strSize] = uset_charAt(eSet, rand()%eSize);
1020 str[3%strSize] = uset_charAt(eSet, rand()%eSize);
1021 str[5%strSize] = uset_charAt(eSet, rand()%eSize);
1022 str[10%strSize] = uset_charAt(eSet, rand()%eSize);
1023 str[13%strSize] = uset_charAt(eSet, rand()%eSize);
1024 }
1025 /* terminate the string */
1026 str[strSize-1] = '\0';
1027 log_verbose("String used: %S\n", str);
1028
1029 /* get sort keys for both of them, and check that the keys are identicle */
1030 sortKeyLen1 = ucol_getSortKey(c1, str, u_strlen(str), NULL, 0);
1031 sortKey1 = (uint8_t*)malloc(sizeof(uint8_t) * (sortKeyLen1 + 1));
1032 /*memset(sortKey1, 0xFE, sortKeyLen1);*/
1033 ucol_getSortKey(c1, str, u_strlen(str), sortKey1, sortKeyLen1 + 1);
1034 ucol_sortKeyToString(c1, sortKey1, sortKeyStr1, sortKeyStrLen1);
1035
1036 sortKeyLen2 = ucol_getSortKey(c2, str, u_strlen(str), NULL, 0);
1037 sortKey2 = (uint8_t*)malloc(sizeof(uint8_t) * (sortKeyLen2 + 1));
1038 /*memset(sortKey2, 0xFE, sortKeyLen2);*/
1039 ucol_getSortKey(c2, str, u_strlen(str), sortKey2, sortKeyLen2 + 1);
1040 ucol_sortKeyToString(c2, sortKey2, sortKeyStr2, sortKeyStrLen2);
1041
1042 /* Check that the lengths are the same */
1043 if (sortKeyLen1 != sortKeyLen2) {
1044 log_err("ERROR : Sort key lengths %d and %d for text '%s' in locale '%s' do not match.\n",
1045 sortKeyLen1, sortKeyLen2, str, curLoc);
1046 }
1047
1048 /* check that the keys are the same */
1049 if (memcmp(sortKey1, sortKey2, sortKeyLen1) != 0) {
1050 log_err("ERROR : Sort keys '%s' and '%s' for text '%s' in locale '%s' are not equivalent.\n",
1051 sortKeyStr1, sortKeyStr2, str, curLoc);
1052 }
1053
1054 /* clean up after each string */
1055 free(sortKey1);
1056 free(sortKey2);
1057 uset_close(eSet);
1058 }
1059 /* clean up after each locale */
1060 ulocdata_close(uld);
1061 ucol_close(c1);
1062 ucol_close(c2);
1063 }
1064 /* final clean up */
1065 uset_close(stdSet);
1066 }
1067 /*
1068 ----------------------------------------------------------------------------
1069 ctor -- Tests the getSortKey
1070 */
TestSortKey()1071 void TestSortKey()
1072 {
1073 uint8_t *sortk1 = NULL, *sortk2 = NULL, *sortk3 = NULL, *sortkEmpty = NULL;
1074 int32_t sortklen, osortklen;
1075 UCollator *col;
1076 UChar *test1, *test2, *test3;
1077 UErrorCode status = U_ZERO_ERROR;
1078 char toStringBuffer[256], *resultP;
1079 uint32_t toStringLen=UPRV_LENGTHOF(toStringBuffer);
1080
1081
1082 uint8_t s1[] = { 0x9f, 0x00 };
1083 uint8_t s2[] = { 0x61, 0x00 };
1084 int strcmpResult;
1085
1086 strcmpResult = strcmp((const char *)s1, (const char *)s2);
1087 log_verbose("strcmp(0x9f..., 0x61...) = %d\n", strcmpResult);
1088
1089 if(strcmpResult <= 0) {
1090 log_err("ERR: expected strcmp(\"9f 00\", \"61 00\") to be >=0 (GREATER).. got %d. Calling strcmp() for sortkeys may not work! \n",
1091 strcmpResult);
1092 }
1093
1094
1095 log_verbose("testing SortKey begins...\n");
1096 /* this is supposed to open default date format, but later on it treats it like it is "en_US"
1097 - very bad if you try to run the tests on machine where default locale is NOT "en_US" */
1098 /* col = ucol_open(NULL, &status); */
1099 col = ucol_open("en_US", &status);
1100 if (U_FAILURE(status)) {
1101 log_err_status(status, "ERROR: Default collation creation failed.: %s\n", myErrorName(status));
1102 return;
1103 }
1104
1105
1106 if(ucol_getStrength(col) != UCOL_DEFAULT_STRENGTH)
1107 {
1108 log_err("ERROR: default collation did not have UCOL_DEFAULT_STRENGTH !\n");
1109 }
1110 /* Need to use identical strength */
1111 ucol_setAttribute(col, UCOL_STRENGTH, UCOL_IDENTICAL, &status);
1112
1113 test1=(UChar*)malloc(sizeof(UChar) * 6);
1114 test2=(UChar*)malloc(sizeof(UChar) * 6);
1115 test3=(UChar*)malloc(sizeof(UChar) * 6);
1116
1117 memset(test1,0xFE, sizeof(UChar)*6);
1118 memset(test2,0xFE, sizeof(UChar)*6);
1119 memset(test3,0xFE, sizeof(UChar)*6);
1120
1121
1122 u_uastrcpy(test1, "Abcda");
1123 u_uastrcpy(test2, "abcda");
1124 u_uastrcpy(test3, "abcda");
1125
1126 log_verbose("Use tertiary comparison level testing ....\n");
1127
1128 sortklen=ucol_getSortKey(col, test1, u_strlen(test1), NULL, 0);
1129 sortk1=(uint8_t*)malloc(sizeof(uint8_t) * (sortklen+1));
1130 memset(sortk1,0xFE, sortklen);
1131 ucol_getSortKey(col, test1, u_strlen(test1), sortk1, sortklen+1);
1132
1133 sortklen=ucol_getSortKey(col, test2, u_strlen(test2), NULL, 0);
1134 sortk2=(uint8_t*)malloc(sizeof(uint8_t) * (sortklen+1));
1135 memset(sortk2,0xFE, sortklen);
1136 ucol_getSortKey(col, test2, u_strlen(test2), sortk2, sortklen+1);
1137
1138 osortklen = sortklen;
1139 sortklen=ucol_getSortKey(col, test2, u_strlen(test3), NULL, 0);
1140 sortk3=(uint8_t*)malloc(sizeof(uint8_t) * (sortklen+1));
1141 memset(sortk3,0xFE, sortklen);
1142 ucol_getSortKey(col, test2, u_strlen(test2), sortk3, sortklen+1);
1143
1144 doAssert( (sortklen == osortklen), "Sortkey length should be the same (abcda, abcda)");
1145
1146 doAssert( (memcmp(sortk1, sortk2, sortklen) > 0), "Result should be \"Abcda\" > \"abcda\"");
1147 doAssert( (memcmp(sortk2, sortk1, sortklen) < 0), "Result should be \"abcda\" < \"Abcda\"");
1148 doAssert( (memcmp(sortk2, sortk3, sortklen) == 0), "Result should be \"abcda\" == \"abcda\"");
1149
1150 resultP = ucol_sortKeyToString(col, sortk3, toStringBuffer, toStringLen);
1151 doAssert( (resultP != 0), "sortKeyToString failed!");
1152
1153 #if 1 /* verobse log of sortkeys */
1154 {
1155 char junk2[1000];
1156 char junk3[1000];
1157 int i;
1158
1159 strcpy(junk2, "abcda[2] ");
1160 strcpy(junk3, " abcda[3] ");
1161
1162 for(i=0;i<sortklen;i++)
1163 {
1164 sprintf(junk2+strlen(junk2), "%02X ",(int)( 0xFF & sortk2[i]));
1165 sprintf(junk3+strlen(junk3), "%02X ",(int)( 0xFF & sortk3[i]));
1166 }
1167
1168 log_verbose("%s\n", junk2);
1169 log_verbose("%s\n", junk3);
1170 }
1171 #endif
1172
1173 free(sortk1);
1174 free(sortk2);
1175 free(sortk3);
1176
1177 log_verbose("Use secondary comparision level testing ...\n");
1178 ucol_setStrength(col, UCOL_SECONDARY);
1179 sortklen=ucol_getSortKey(col, test1, u_strlen(test1), NULL, 0);
1180 sortk1=(uint8_t*)malloc(sizeof(uint8_t) * (sortklen+1));
1181 ucol_getSortKey(col, test1, u_strlen(test1), sortk1, sortklen+1);
1182 sortklen=ucol_getSortKey(col, test2, u_strlen(test2), NULL, 0);
1183 sortk2=(uint8_t*)malloc(sizeof(uint8_t) * (sortklen+1));
1184 ucol_getSortKey(col, test2, u_strlen(test2), sortk2, sortklen+1);
1185
1186 doAssert( !(memcmp(sortk1, sortk2, sortklen) > 0), "Result should be \"Abcda\" == \"abcda\"");
1187 doAssert( !(memcmp(sortk2, sortk1, sortklen) < 0), "Result should be \"abcda\" == \"Abcda\"");
1188 doAssert( (memcmp(sortk1, sortk2, sortklen) == 0), "Result should be \"abcda\" == \"abcda\"");
1189
1190 log_verbose("getting sortkey for an empty string\n");
1191 ucol_setAttribute(col, UCOL_STRENGTH, UCOL_TERTIARY, &status);
1192 sortklen = ucol_getSortKey(col, test1, 0, NULL, 0);
1193 sortkEmpty = (uint8_t*)malloc(sizeof(uint8_t) * sortklen+1);
1194 sortklen = ucol_getSortKey(col, test1, 0, sortkEmpty, sortklen+1);
1195 if(sortklen != 3 || sortkEmpty[0] != 1 || sortkEmpty[0] != 1 || sortkEmpty[2] != 0) {
1196 log_err("Empty string generated wrong sortkey!\n");
1197 }
1198 free(sortkEmpty);
1199
1200 log_verbose("testing passing invalid string\n");
1201 sortklen = ucol_getSortKey(col, NULL, 10, NULL, 0);
1202 if(sortklen != 0) {
1203 log_err("Invalid string didn't return sortkey size of 0\n");
1204 }
1205
1206
1207 log_verbose("testing sortkey ends...\n");
1208 ucol_close(col);
1209 free(test1);
1210 free(test2);
1211 free(test3);
1212 free(sortk1);
1213 free(sortk2);
1214
1215 }
TestHashCode()1216 void TestHashCode()
1217 {
1218 uint8_t *sortk1, *sortk2, *sortk3;
1219 int32_t sortk1len, sortk2len, sortk3len;
1220 UCollator *col;
1221 UChar *test1, *test2, *test3;
1222 UErrorCode status = U_ZERO_ERROR;
1223 log_verbose("testing getHashCode begins...\n");
1224 col = ucol_open("en_US", &status);
1225 if (U_FAILURE(status)) {
1226 log_err_status(status, "ERROR: Default collation creation failed.: %s\n", myErrorName(status));
1227 return;
1228 }
1229 test1=(UChar*)malloc(sizeof(UChar) * 6);
1230 test2=(UChar*)malloc(sizeof(UChar) * 6);
1231 test3=(UChar*)malloc(sizeof(UChar) * 6);
1232 u_uastrcpy(test1, "Abcda");
1233 u_uastrcpy(test2, "abcda");
1234 u_uastrcpy(test3, "abcda");
1235
1236 log_verbose("Use tertiary comparison level testing ....\n");
1237 sortk1len=ucol_getSortKey(col, test1, u_strlen(test1), NULL, 0);
1238 sortk1=(uint8_t*)malloc(sizeof(uint8_t) * (sortk1len+1));
1239 ucol_getSortKey(col, test1, u_strlen(test1), sortk1, sortk1len+1);
1240 sortk2len=ucol_getSortKey(col, test2, u_strlen(test2), NULL, 0);
1241 sortk2=(uint8_t*)malloc(sizeof(uint8_t) * (sortk2len+1));
1242 ucol_getSortKey(col, test2, u_strlen(test2), sortk2, sortk2len+1);
1243 sortk3len=ucol_getSortKey(col, test2, u_strlen(test3), NULL, 0);
1244 sortk3=(uint8_t*)malloc(sizeof(uint8_t) * (sortk3len+1));
1245 ucol_getSortKey(col, test2, u_strlen(test2), sortk3, sortk3len+1);
1246
1247
1248 log_verbose("ucol_hashCode() testing ...\n");
1249
1250 doAssert( ucol_keyHashCode(sortk1, sortk1len) != ucol_keyHashCode(sortk2, sortk2len), "Hash test1 result incorrect" );
1251 doAssert( !(ucol_keyHashCode(sortk1, sortk1len) == ucol_keyHashCode(sortk2, sortk2len)), "Hash test2 result incorrect" );
1252 doAssert( ucol_keyHashCode(sortk2, sortk2len) == ucol_keyHashCode(sortk3, sortk3len), "Hash result not equal" );
1253
1254 log_verbose("hashCode tests end.\n");
1255 ucol_close(col);
1256 free(sortk1);
1257 free(sortk2);
1258 free(sortk3);
1259 free(test1);
1260 free(test2);
1261 free(test3);
1262
1263
1264 }
1265 /*
1266 *----------------------------------------------------------------------------
1267 * Tests the UCollatorElements API.
1268 *
1269 */
TestElemIter()1270 void TestElemIter()
1271 {
1272 int32_t offset;
1273 int32_t order1, order2, order3;
1274 UChar *testString1, *testString2;
1275 UCollator *col;
1276 UCollationElements *iterator1, *iterator2, *iterator3;
1277 UErrorCode status = U_ZERO_ERROR;
1278 log_verbose("testing UCollatorElements begins...\n");
1279 col = ucol_open("en_US", &status);
1280 ucol_setAttribute(col, UCOL_NORMALIZATION_MODE, UCOL_OFF, &status);
1281 if (U_FAILURE(status)) {
1282 log_err_status(status, "ERROR: Default collation creation failed.: %s\n", myErrorName(status));
1283 return;
1284 }
1285
1286 testString1=(UChar*)malloc(sizeof(UChar) * 150);
1287 testString2=(UChar*)malloc(sizeof(UChar) * 150);
1288 u_uastrcpy(testString1, "XFILE What subset of all possible test cases has the highest probability of detecting the most errors?");
1289 u_uastrcpy(testString2, "Xf_ile What subset of all possible test cases has the lowest probability of detecting the least errors?");
1290
1291 log_verbose("Constructors and comparison testing....\n");
1292
1293 iterator1 = ucol_openElements(col, testString1, u_strlen(testString1), &status);
1294 if(U_FAILURE(status)) {
1295 log_err("ERROR: Default collationElement iterator creation failed.: %s\n", myErrorName(status));
1296 ucol_close(col);
1297 return;
1298 }
1299 else{ log_verbose("PASS: Default collationElement iterator1 creation passed\n");}
1300
1301 iterator2 = ucol_openElements(col, testString1, u_strlen(testString1), &status);
1302 if(U_FAILURE(status)) {
1303 log_err("ERROR: Default collationElement iterator creation failed.: %s\n", myErrorName(status));
1304 ucol_close(col);
1305 return;
1306 }
1307 else{ log_verbose("PASS: Default collationElement iterator2 creation passed\n");}
1308
1309 iterator3 = ucol_openElements(col, testString2, u_strlen(testString2), &status);
1310 if(U_FAILURE(status)) {
1311 log_err("ERROR: Default collationElement iterator creation failed.: %s\n", myErrorName(status));
1312 ucol_close(col);
1313 return;
1314 }
1315 else{ log_verbose("PASS: Default collationElement iterator3 creation passed\n");}
1316
1317 offset=ucol_getOffset(iterator1);
1318 (void)offset; /* Suppress set but not used warning. */
1319 ucol_setOffset(iterator1, 6, &status);
1320 if (U_FAILURE(status)) {
1321 log_err("Error in setOffset for UCollatorElements iterator.: %s\n", myErrorName(status));
1322 return;
1323 }
1324 if(ucol_getOffset(iterator1)==6)
1325 log_verbose("setOffset and getOffset working fine\n");
1326 else{
1327 log_err("error in set and get Offset got %d instead of 6\n", ucol_getOffset(iterator1));
1328 }
1329
1330 ucol_setOffset(iterator1, 0, &status);
1331 order1 = ucol_next(iterator1, &status);
1332 if (U_FAILURE(status)) {
1333 log_err("Somehow ran out of memory stepping through the iterator1.: %s\n", myErrorName(status));
1334 return;
1335 }
1336 order2=ucol_getOffset(iterator2);
1337 doAssert((order1 != order2), "The first iterator advance failed");
1338 order2 = ucol_next(iterator2, &status);
1339 if (U_FAILURE(status)) {
1340 log_err("Somehow ran out of memory stepping through the iterator2.: %s\n", myErrorName(status));
1341 return;
1342 }
1343 order3 = ucol_next(iterator3, &status);
1344 if (U_FAILURE(status)) {
1345 log_err("Somehow ran out of memory stepping through the iterator3.: %s\n", myErrorName(status));
1346 return;
1347 }
1348
1349 doAssert((order1 == order2), "The second iterator advance failed should be the same as first one");
1350
1351 doAssert( (ucol_primaryOrder(order1) == ucol_primaryOrder(order3)), "The primary orders should be identical");
1352 doAssert( (ucol_secondaryOrder(order1) == ucol_secondaryOrder(order3)), "The secondary orders should be identical");
1353 doAssert( (ucol_tertiaryOrder(order1) == ucol_tertiaryOrder(order3)), "The tertiary orders should be identical");
1354
1355 order1=ucol_next(iterator1, &status);
1356 if (U_FAILURE(status)) {
1357 log_err("Somehow ran out of memory stepping through the iterator2.: %s\n", myErrorName(status));
1358 return;
1359 }
1360 order3=ucol_next(iterator3, &status);
1361 if (U_FAILURE(status)) {
1362 log_err("Somehow ran out of memory stepping through the iterator2.: %s\n", myErrorName(status));
1363 return;
1364 }
1365 doAssert( (ucol_primaryOrder(order1) == ucol_primaryOrder(order3)), "The primary orders should be identical");
1366 doAssert( (ucol_tertiaryOrder(order1) != ucol_tertiaryOrder(order3)), "The tertiary orders should be different");
1367
1368 order1=ucol_next(iterator1, &status);
1369 if (U_FAILURE(status)) {
1370 log_err("Somehow ran out of memory stepping through the iterator2.: %s\n", myErrorName(status));
1371 return;
1372 }
1373 order3=ucol_next(iterator3, &status);
1374 if (U_FAILURE(status)) {
1375 log_err("Somehow ran out of memory stepping through the iterator2.: %s\n", myErrorName(status));
1376 return;
1377 }
1378 /* this here, my friends, is either pure lunacy or something so obsolete that even it's mother
1379 * doesn't care about it. Essentialy, this test complains if secondary values for 'I' and '_'
1380 * are the same. According to the UCA, this is not true. Therefore, remove the test.
1381 * Besides, if primary strengths for two code points are different, it doesn't matter one bit
1382 * what is the relation between secondary or any other strengths.
1383 * killed by weiv 06/11/2002.
1384 */
1385 /*
1386 doAssert( ((order1 & UCOL_SECONDARYMASK) != (order3 & UCOL_SECONDARYMASK)), "The secondary orders should be different");
1387 */
1388 doAssert( (order1 != UCOL_NULLORDER), "Unexpected end of iterator reached");
1389
1390 free(testString1);
1391 free(testString2);
1392 ucol_closeElements(iterator1);
1393 ucol_closeElements(iterator2);
1394 ucol_closeElements(iterator3);
1395 ucol_close(col);
1396
1397 log_verbose("testing CollationElementIterator ends...\n");
1398 }
1399
TestGetLocale()1400 void TestGetLocale() {
1401 UErrorCode status = U_ZERO_ERROR;
1402 const char *rules = "&a<x<y<z";
1403 UChar rlz[256] = {0};
1404 uint32_t rlzLen = u_unescape(rules, rlz, 256);
1405
1406 UCollator *coll = NULL;
1407 const char *locale = NULL;
1408
1409 int32_t i = 0;
1410
1411 static const struct {
1412 const char* requestedLocale;
1413 const char* validLocale;
1414 const char* actualLocale;
1415 } testStruct[] = {
1416 { "sr_RS", "sr_Cyrl_RS", "sr" },
1417 { "sh_YU", "sr_Latn_RS", "sr_Latn" }, /* was sh, then aliased to hr, now sr_Latn via import per cldrbug 5647: */
1418 { "en_BE_FOO", "en", "root" },
1419 { "sv_SE_NONEXISTANT", "sv", "sv" }
1420 };
1421
1422 /* test opening collators for different locales */
1423 for(i = 0; i<UPRV_LENGTHOF(testStruct); i++) {
1424 status = U_ZERO_ERROR;
1425 coll = ucol_open(testStruct[i].requestedLocale, &status);
1426 if(U_FAILURE(status)) {
1427 log_err_status(status, "Failed to open collator for %s with %s\n", testStruct[i].requestedLocale, u_errorName(status));
1428 ucol_close(coll);
1429 continue;
1430 }
1431 /*
1432 * The requested locale may be the same as the valid locale,
1433 * or may not be supported at all. See ticket #10477.
1434 */
1435 locale = ucol_getLocaleByType(coll, ULOC_REQUESTED_LOCALE, &status);
1436 if(U_SUCCESS(status) &&
1437 strcmp(locale, testStruct[i].requestedLocale) != 0 && strcmp(locale, testStruct[i].validLocale) != 0) {
1438 log_err("[Coll %s]: Error in requested locale, expected %s, got %s\n", testStruct[i].requestedLocale, testStruct[i].requestedLocale, locale);
1439 }
1440 status = U_ZERO_ERROR;
1441 locale = ucol_getLocaleByType(coll, ULOC_VALID_LOCALE, &status);
1442 if(strcmp(locale, testStruct[i].validLocale) != 0) {
1443 log_err("[Coll %s]: Error in valid locale, expected %s, got %s\n", testStruct[i].requestedLocale, testStruct[i].validLocale, locale);
1444 }
1445 locale = ucol_getLocaleByType(coll, ULOC_ACTUAL_LOCALE, &status);
1446 if(strcmp(locale, testStruct[i].actualLocale) != 0) {
1447 log_err("[Coll %s]: Error in actual locale, expected %s, got %s\n", testStruct[i].requestedLocale, testStruct[i].actualLocale, locale);
1448 }
1449 ucol_close(coll);
1450 }
1451
1452 /* completely non-existent locale for collator should get a root collator */
1453 {
1454 UCollator *defaultColl = ucol_open(NULL, &status);
1455 coll = ucol_open("blahaha", &status);
1456 if(U_SUCCESS(status)) {
1457 /* See comment above about ticket #10477.
1458 if(strcmp(ucol_getLocaleByType(coll, ULOC_REQUESTED_LOCALE, &status), "blahaha")) {
1459 log_err("Nonexisting locale didn't preserve the requested locale\n");
1460 } */
1461 const char *name = ucol_getLocaleByType(coll, ULOC_VALID_LOCALE, &status);
1462 if(*name != 0 && strcmp(name, "root") != 0) {
1463 log_err("Valid locale for nonexisting-locale collator is \"%s\" not root\n", name);
1464 }
1465 name = ucol_getLocaleByType(coll, ULOC_ACTUAL_LOCALE, &status);
1466 if(*name != 0 && strcmp(name, "root") != 0) {
1467 log_err("Actual locale for nonexisting-locale collator is \"%s\" not root\n", name);
1468 }
1469 ucol_close(coll);
1470 ucol_close(defaultColl);
1471 } else {
1472 log_data_err("Couldn't open collators\n");
1473 }
1474 }
1475
1476
1477
1478 /* collator instantiated from rules should have all three locales NULL */
1479 coll = ucol_openRules(rlz, rlzLen, UCOL_DEFAULT, UCOL_DEFAULT, NULL, &status);
1480 if (coll != NULL) {
1481 locale = ucol_getLocaleByType(coll, ULOC_REQUESTED_LOCALE, &status);
1482 if(U_SUCCESS(status) && locale != NULL) {
1483 log_err("For collator instantiated from rules, requested locale returned %s instead of NULL\n", locale);
1484 }
1485 status = U_ZERO_ERROR;
1486 locale = ucol_getLocaleByType(coll, ULOC_VALID_LOCALE, &status);
1487 if(locale != NULL) {
1488 log_err("For collator instantiated from rules, valid locale returned %s instead of NULL\n", locale);
1489 }
1490 locale = ucol_getLocaleByType(coll, ULOC_ACTUAL_LOCALE, &status);
1491 if(locale != NULL) {
1492 log_err("For collator instantiated from rules, actual locale returned %s instead of NULL\n", locale);
1493 }
1494 ucol_close(coll);
1495 } else {
1496 log_data_err("Couldn't get collator from ucol_openRules() - %s\n", u_errorName(status));
1497 }
1498 }
1499
1500
TestGetAll()1501 void TestGetAll()
1502 {
1503 int32_t i, count;
1504 count=ucol_countAvailable();
1505 /* use something sensible w/o hardcoding the count */
1506 if(count < 0){
1507 log_err("Error in countAvailable(), it returned %d\n", count);
1508 }
1509 else{
1510 log_verbose("PASS: countAvailable() successful, it returned %d\n", count);
1511 }
1512 for(i=0;i<count;i++)
1513 log_verbose("%s\n", ucol_getAvailable(i));
1514
1515
1516 }
1517
1518
1519 struct teststruct {
1520 const char *original;
1521 uint8_t key[256];
1522 } ;
1523
compare_teststruct(const void * string1,const void * string2)1524 static int compare_teststruct(const void *string1, const void *string2) {
1525 return(strcmp((const char *)((struct teststruct *)string1)->key, (const char *)((struct teststruct *)string2)->key));
1526 }
1527
TestBounds()1528 void TestBounds() {
1529 UErrorCode status = U_ZERO_ERROR;
1530
1531 UCollator *coll = ucol_open("sh", &status);
1532
1533 uint8_t sortkey[512], lower[512], upper[512];
1534 UChar buffer[512];
1535
1536 static const char * const test[] = {
1537 "John Smith",
1538 "JOHN SMITH",
1539 "john SMITH",
1540 "j\\u00F6hn sm\\u00EFth",
1541 "J\\u00F6hn Sm\\u00EFth",
1542 "J\\u00D6HN SM\\u00CFTH",
1543 "john smithsonian",
1544 "John Smithsonian",
1545 };
1546
1547 struct teststruct tests[] = {
1548 {"\\u010CAKI MIHALJ", {0}},
1549 {"\\u010CAKI MIHALJ", {0}},
1550 {"\\u010CAKI PIRO\\u0160KA", {0}},
1551 {"\\u010CABAI ANDRIJA", {0}},
1552 {"\\u010CABAI LAJO\\u0160", {0}},
1553 {"\\u010CABAI MARIJA", {0}},
1554 {"\\u010CABAI STEVAN", {0}},
1555 {"\\u010CABAI STEVAN", {0}},
1556 {"\\u010CABARKAPA BRANKO", {0}},
1557 {"\\u010CABARKAPA MILENKO", {0}},
1558 {"\\u010CABARKAPA MIROSLAV", {0}},
1559 {"\\u010CABARKAPA SIMO", {0}},
1560 {"\\u010CABARKAPA STANKO", {0}},
1561 {"\\u010CABARKAPA TAMARA", {0}},
1562 {"\\u010CABARKAPA TOMA\\u0160", {0}},
1563 {"\\u010CABDARI\\u0106 NIKOLA", {0}},
1564 {"\\u010CABDARI\\u0106 ZORICA", {0}},
1565 {"\\u010CABI NANDOR", {0}},
1566 {"\\u010CABOVI\\u0106 MILAN", {0}},
1567 {"\\u010CABRADI AGNEZIJA", {0}},
1568 {"\\u010CABRADI IVAN", {0}},
1569 {"\\u010CABRADI JELENA", {0}},
1570 {"\\u010CABRADI LJUBICA", {0}},
1571 {"\\u010CABRADI STEVAN", {0}},
1572 {"\\u010CABRDA MARTIN", {0}},
1573 {"\\u010CABRILO BOGDAN", {0}},
1574 {"\\u010CABRILO BRANISLAV", {0}},
1575 {"\\u010CABRILO LAZAR", {0}},
1576 {"\\u010CABRILO LJUBICA", {0}},
1577 {"\\u010CABRILO SPASOJA", {0}},
1578 {"\\u010CADE\\u0160 ZDENKA", {0}},
1579 {"\\u010CADESKI BLAGOJE", {0}},
1580 {"\\u010CADOVSKI VLADIMIR", {0}},
1581 {"\\u010CAGLJEVI\\u0106 TOMA", {0}},
1582 {"\\u010CAGOROVI\\u0106 VLADIMIR", {0}},
1583 {"\\u010CAJA VANKA", {0}},
1584 {"\\u010CAJI\\u0106 BOGOLJUB", {0}},
1585 {"\\u010CAJI\\u0106 BORISLAV", {0}},
1586 {"\\u010CAJI\\u0106 RADOSLAV", {0}},
1587 {"\\u010CAK\\u0160IRAN MILADIN", {0}},
1588 {"\\u010CAKAN EUGEN", {0}},
1589 {"\\u010CAKAN EVGENIJE", {0}},
1590 {"\\u010CAKAN IVAN", {0}},
1591 {"\\u010CAKAN JULIJAN", {0}},
1592 {"\\u010CAKAN MIHAJLO", {0}},
1593 {"\\u010CAKAN STEVAN", {0}},
1594 {"\\u010CAKAN VLADIMIR", {0}},
1595 {"\\u010CAKAN VLADIMIR", {0}},
1596 {"\\u010CAKAN VLADIMIR", {0}},
1597 {"\\u010CAKARA ANA", {0}},
1598 {"\\u010CAKAREVI\\u0106 MOMIR", {0}},
1599 {"\\u010CAKAREVI\\u0106 NEDELJKO", {0}},
1600 {"\\u010CAKI \\u0160ANDOR", {0}},
1601 {"\\u010CAKI AMALIJA", {0}},
1602 {"\\u010CAKI ANDRA\\u0160", {0}},
1603 {"\\u010CAKI LADISLAV", {0}},
1604 {"\\u010CAKI LAJO\\u0160", {0}},
1605 {"\\u010CAKI LASLO", {0}},
1606 };
1607
1608
1609
1610 int32_t i = 0, j = 0, k = 0, buffSize = 0, skSize = 0, lowerSize = 0, upperSize = 0;
1611 int32_t arraySize = UPRV_LENGTHOF(tests);
1612
1613 if(U_SUCCESS(status) && coll) {
1614 for(i = 0; i<arraySize; i++) {
1615 buffSize = u_unescape(tests[i].original, buffer, 512);
1616 skSize = ucol_getSortKey(coll, buffer, buffSize, tests[i].key, 512);
1617 }
1618
1619 qsort(tests, arraySize, sizeof(struct teststruct), compare_teststruct);
1620
1621 for(i = 0; i < arraySize-1; i++) {
1622 for(j = i+1; j < arraySize; j++) {
1623 lowerSize = ucol_getBound(tests[i].key, -1, UCOL_BOUND_LOWER, 1, lower, 512, &status);
1624 upperSize = ucol_getBound(tests[j].key, -1, UCOL_BOUND_UPPER, 1, upper, 512, &status);
1625 (void)lowerSize; /* Suppress set but not used warning. */
1626 (void)upperSize;
1627 for(k = i; k <= j; k++) {
1628 if(strcmp((const char *)lower, (const char *)tests[k].key) > 0) {
1629 log_err("Problem with lower! j = %i (%s vs %s)\n", k, tests[k].original, tests[i].original);
1630 }
1631 if(strcmp((const char *)upper, (const char *)tests[k].key) <= 0) {
1632 log_err("Problem with upper! j = %i (%s vs %s)\n", k, tests[k].original, tests[j].original);
1633 }
1634 }
1635 }
1636 }
1637
1638
1639 #if 0
1640 for(i = 0; i < 1000; i++) {
1641 lowerRND = (rand()/(RAND_MAX/arraySize));
1642 upperRND = lowerRND + (rand()/(RAND_MAX/(arraySize-lowerRND)));
1643
1644 lowerSize = ucol_getBound(tests[lowerRND].key, -1, UCOL_BOUND_LOWER, 1, lower, 512, &status);
1645 upperSize = ucol_getBound(tests[upperRND].key, -1, UCOL_BOUND_UPPER_LONG, 1, upper, 512, &status);
1646
1647 for(j = lowerRND; j<=upperRND; j++) {
1648 if(strcmp(lower, tests[j].key) > 0) {
1649 log_err("Problem with lower! j = %i (%s vs %s)\n", j, tests[j].original, tests[lowerRND].original);
1650 }
1651 if(strcmp(upper, tests[j].key) <= 0) {
1652 log_err("Problem with upper! j = %i (%s vs %s)\n", j, tests[j].original, tests[upperRND].original);
1653 }
1654 }
1655 }
1656 #endif
1657
1658
1659
1660
1661
1662 for(i = 0; i<UPRV_LENGTHOF(test); i++) {
1663 buffSize = u_unescape(test[i], buffer, 512);
1664 skSize = ucol_getSortKey(coll, buffer, buffSize, sortkey, 512);
1665 lowerSize = ucol_getBound(sortkey, skSize, UCOL_BOUND_LOWER, 1, lower, 512, &status);
1666 upperSize = ucol_getBound(sortkey, skSize, UCOL_BOUND_UPPER_LONG, 1, upper, 512, &status);
1667 for(j = i+1; j<UPRV_LENGTHOF(test); j++) {
1668 buffSize = u_unescape(test[j], buffer, 512);
1669 skSize = ucol_getSortKey(coll, buffer, buffSize, sortkey, 512);
1670 if(strcmp((const char *)lower, (const char *)sortkey) > 0) {
1671 log_err("Problem with lower! i = %i, j = %i (%s vs %s)\n", i, j, test[i], test[j]);
1672 }
1673 if(strcmp((const char *)upper, (const char *)sortkey) <= 0) {
1674 log_err("Problem with upper! i = %i, j = %i (%s vs %s)\n", i, j, test[i], test[j]);
1675 }
1676 }
1677 }
1678 ucol_close(coll);
1679 } else {
1680 log_data_err("Couldn't open collator\n");
1681 }
1682
1683 }
1684
doOverrunTest(UCollator * coll,const UChar * uString,int32_t strLen)1685 static void doOverrunTest(UCollator *coll, const UChar *uString, int32_t strLen) {
1686 int32_t skLen = 0, skLen2 = 0;
1687 uint8_t sortKey[256];
1688 int32_t i, j;
1689 uint8_t filler = 0xFF;
1690
1691 skLen = ucol_getSortKey(coll, uString, strLen, NULL, 0);
1692
1693 for(i = 0; i < skLen; i++) {
1694 memset(sortKey, filler, 256);
1695 skLen2 = ucol_getSortKey(coll, uString, strLen, sortKey, i);
1696 if(skLen != skLen2) {
1697 log_err("For buffer size %i, got different sortkey length. Expected %i got %i\n", i, skLen, skLen2);
1698 }
1699 for(j = i; j < 256; j++) {
1700 if(sortKey[j] != filler) {
1701 log_err("Something run over index %i\n", j);
1702 break;
1703 }
1704 }
1705 }
1706 }
1707
1708 /* j1865 reports that if a shorter buffer is passed to
1709 * to get sort key, a buffer overrun happens in some
1710 * cases. This test tries to check this.
1711 */
TestSortKeyBufferOverrun(void)1712 void TestSortKeyBufferOverrun(void) {
1713 UErrorCode status = U_ZERO_ERROR;
1714 const char* cString = "A very Merry liTTle-lamB..";
1715 UChar uString[256];
1716 int32_t strLen = 0;
1717 UCollator *coll = ucol_open("root", &status);
1718 strLen = u_unescape(cString, uString, 256);
1719
1720 if(U_SUCCESS(status)) {
1721 log_verbose("testing non ignorable\n");
1722 ucol_setAttribute(coll, UCOL_ALTERNATE_HANDLING, UCOL_NON_IGNORABLE, &status);
1723 doOverrunTest(coll, uString, strLen);
1724
1725 log_verbose("testing shifted\n");
1726 ucol_setAttribute(coll, UCOL_ALTERNATE_HANDLING, UCOL_SHIFTED, &status);
1727 doOverrunTest(coll, uString, strLen);
1728
1729 log_verbose("testing shifted quaternary\n");
1730 ucol_setAttribute(coll, UCOL_STRENGTH, UCOL_QUATERNARY, &status);
1731 doOverrunTest(coll, uString, strLen);
1732
1733 log_verbose("testing with french secondaries\n");
1734 ucol_setAttribute(coll, UCOL_FRENCH_COLLATION, UCOL_ON, &status);
1735 ucol_setAttribute(coll, UCOL_STRENGTH, UCOL_TERTIARY, &status);
1736 ucol_setAttribute(coll, UCOL_ALTERNATE_HANDLING, UCOL_NON_IGNORABLE, &status);
1737 doOverrunTest(coll, uString, strLen);
1738
1739 }
1740 ucol_close(coll);
1741 }
1742
TestAttribute()1743 static void TestAttribute()
1744 {
1745 UErrorCode error = U_ZERO_ERROR;
1746 UCollator *coll = ucol_open(NULL, &error);
1747
1748 if (U_FAILURE(error)) {
1749 log_err_status(error, "Creation of default collator failed\n");
1750 return;
1751 }
1752
1753 ucol_setAttribute(coll, UCOL_FRENCH_COLLATION, UCOL_OFF, &error);
1754 if (ucol_getAttribute(coll, UCOL_FRENCH_COLLATION, &error) != UCOL_OFF ||
1755 U_FAILURE(error)) {
1756 log_err_status(error, "Setting and retrieving of the french collation failed\n");
1757 }
1758
1759 ucol_setAttribute(coll, UCOL_FRENCH_COLLATION, UCOL_ON, &error);
1760 if (ucol_getAttribute(coll, UCOL_FRENCH_COLLATION, &error) != UCOL_ON ||
1761 U_FAILURE(error)) {
1762 log_err_status(error, "Setting and retrieving of the french collation failed\n");
1763 }
1764
1765 ucol_setAttribute(coll, UCOL_ALTERNATE_HANDLING, UCOL_SHIFTED, &error);
1766 if (ucol_getAttribute(coll, UCOL_ALTERNATE_HANDLING, &error) != UCOL_SHIFTED ||
1767 U_FAILURE(error)) {
1768 log_err_status(error, "Setting and retrieving of the alternate handling failed\n");
1769 }
1770
1771 ucol_setAttribute(coll, UCOL_ALTERNATE_HANDLING, UCOL_NON_IGNORABLE, &error);
1772 if (ucol_getAttribute(coll, UCOL_ALTERNATE_HANDLING, &error) != UCOL_NON_IGNORABLE ||
1773 U_FAILURE(error)) {
1774 log_err_status(error, "Setting and retrieving of the alternate handling failed\n");
1775 }
1776
1777 ucol_setAttribute(coll, UCOL_CASE_FIRST, UCOL_LOWER_FIRST, &error);
1778 if (ucol_getAttribute(coll, UCOL_CASE_FIRST, &error) != UCOL_LOWER_FIRST ||
1779 U_FAILURE(error)) {
1780 log_err_status(error, "Setting and retrieving of the case first attribute failed\n");
1781 }
1782
1783 ucol_setAttribute(coll, UCOL_CASE_FIRST, UCOL_UPPER_FIRST, &error);
1784 if (ucol_getAttribute(coll, UCOL_CASE_FIRST, &error) != UCOL_UPPER_FIRST ||
1785 U_FAILURE(error)) {
1786 log_err_status(error, "Setting and retrieving of the case first attribute failed\n");
1787 }
1788
1789 ucol_setAttribute(coll, UCOL_CASE_LEVEL, UCOL_ON, &error);
1790 if (ucol_getAttribute(coll, UCOL_CASE_LEVEL, &error) != UCOL_ON ||
1791 U_FAILURE(error)) {
1792 log_err_status(error, "Setting and retrieving of the case level attribute failed\n");
1793 }
1794
1795 ucol_setAttribute(coll, UCOL_CASE_LEVEL, UCOL_OFF, &error);
1796 if (ucol_getAttribute(coll, UCOL_CASE_LEVEL, &error) != UCOL_OFF ||
1797 U_FAILURE(error)) {
1798 log_err_status(error, "Setting and retrieving of the case level attribute failed\n");
1799 }
1800
1801 ucol_setAttribute(coll, UCOL_NORMALIZATION_MODE, UCOL_ON, &error);
1802 if (ucol_getAttribute(coll, UCOL_NORMALIZATION_MODE, &error) != UCOL_ON ||
1803 U_FAILURE(error)) {
1804 log_err_status(error, "Setting and retrieving of the normalization on/off attribute failed\n");
1805 }
1806
1807 ucol_setAttribute(coll, UCOL_NORMALIZATION_MODE, UCOL_OFF, &error);
1808 if (ucol_getAttribute(coll, UCOL_NORMALIZATION_MODE, &error) != UCOL_OFF ||
1809 U_FAILURE(error)) {
1810 log_err_status(error, "Setting and retrieving of the normalization on/off attribute failed\n");
1811 }
1812
1813 ucol_setAttribute(coll, UCOL_STRENGTH, UCOL_PRIMARY, &error);
1814 if (ucol_getAttribute(coll, UCOL_STRENGTH, &error) != UCOL_PRIMARY ||
1815 U_FAILURE(error)) {
1816 log_err_status(error, "Setting and retrieving of the collation strength failed\n");
1817 }
1818
1819 ucol_setAttribute(coll, UCOL_STRENGTH, UCOL_SECONDARY, &error);
1820 if (ucol_getAttribute(coll, UCOL_STRENGTH, &error) != UCOL_SECONDARY ||
1821 U_FAILURE(error)) {
1822 log_err_status(error, "Setting and retrieving of the collation strength failed\n");
1823 }
1824
1825 ucol_setAttribute(coll, UCOL_STRENGTH, UCOL_TERTIARY, &error);
1826 if (ucol_getAttribute(coll, UCOL_STRENGTH, &error) != UCOL_TERTIARY ||
1827 U_FAILURE(error)) {
1828 log_err_status(error, "Setting and retrieving of the collation strength failed\n");
1829 }
1830
1831 ucol_setAttribute(coll, UCOL_STRENGTH, UCOL_QUATERNARY, &error);
1832 if (ucol_getAttribute(coll, UCOL_STRENGTH, &error) != UCOL_QUATERNARY ||
1833 U_FAILURE(error)) {
1834 log_err_status(error, "Setting and retrieving of the collation strength failed\n");
1835 }
1836
1837 ucol_setAttribute(coll, UCOL_STRENGTH, UCOL_IDENTICAL, &error);
1838 if (ucol_getAttribute(coll, UCOL_STRENGTH, &error) != UCOL_IDENTICAL ||
1839 U_FAILURE(error)) {
1840 log_err_status(error, "Setting and retrieving of the collation strength failed\n");
1841 }
1842
1843 ucol_close(coll);
1844 }
1845
TestGetTailoredSet()1846 void TestGetTailoredSet() {
1847 struct {
1848 const char *rules;
1849 const char *tests[20];
1850 int32_t testsize;
1851 } setTest[] = {
1852 { "&a < \\u212b", { "\\u212b", "A\\u030a", "\\u00c5" }, 3},
1853 { "& S < \\u0161 <<< \\u0160", { "\\u0161", "s\\u030C", "\\u0160", "S\\u030C" }, 4}
1854 };
1855
1856 int32_t i = 0, j = 0;
1857 UErrorCode status = U_ZERO_ERROR;
1858 UParseError pError;
1859
1860 UCollator *coll = NULL;
1861 UChar buff[1024];
1862 int32_t buffLen = 0;
1863 USet *set = NULL;
1864
1865 for(i = 0; i < UPRV_LENGTHOF(setTest); i++) {
1866 buffLen = u_unescape(setTest[i].rules, buff, 1024);
1867 coll = ucol_openRules(buff, buffLen, UCOL_DEFAULT, UCOL_DEFAULT, &pError, &status);
1868 if(U_SUCCESS(status)) {
1869 set = ucol_getTailoredSet(coll, &status);
1870 if(uset_size(set) < setTest[i].testsize) {
1871 log_err("Tailored set size smaller (%d) than expected (%d)\n", uset_size(set), setTest[i].testsize);
1872 }
1873 for(j = 0; j < setTest[i].testsize; j++) {
1874 buffLen = u_unescape(setTest[i].tests[j], buff, 1024);
1875 if(!uset_containsString(set, buff, buffLen)) {
1876 log_err("Tailored set doesn't contain %s... It should\n", setTest[i].tests[j]);
1877 }
1878 }
1879 uset_close(set);
1880 } else {
1881 log_err_status(status, "Couldn't open collator with rules %s\n", setTest[i].rules);
1882 }
1883 ucol_close(coll);
1884 }
1885 }
1886
tMemCmp(const uint8_t * first,const uint8_t * second)1887 static int tMemCmp(const uint8_t *first, const uint8_t *second) {
1888 int32_t firstLen = (int32_t)strlen((const char *)first);
1889 int32_t secondLen = (int32_t)strlen((const char *)second);
1890 return memcmp(first, second, uprv_min(firstLen, secondLen));
1891 }
1892 static const char * strengthsC[] = {
1893 "UCOL_PRIMARY",
1894 "UCOL_SECONDARY",
1895 "UCOL_TERTIARY",
1896 "UCOL_QUATERNARY",
1897 "UCOL_IDENTICAL"
1898 };
1899
TestMergeSortKeys(void)1900 void TestMergeSortKeys(void) {
1901 UErrorCode status = U_ZERO_ERROR;
1902 UCollator *coll = ucol_open("en", &status);
1903 if(U_SUCCESS(status)) {
1904
1905 const char* cases[] = {
1906 "abc",
1907 "abcd",
1908 "abcde"
1909 };
1910 uint32_t casesSize = UPRV_LENGTHOF(cases);
1911 const char* prefix = "foo";
1912 const char* suffix = "egg";
1913 char outBuff1[256], outBuff2[256];
1914
1915 uint8_t **sortkeys = (uint8_t **)malloc(casesSize*sizeof(uint8_t *));
1916 uint8_t **mergedPrefixkeys = (uint8_t **)malloc(casesSize*sizeof(uint8_t *));
1917 uint8_t **mergedSuffixkeys = (uint8_t **)malloc(casesSize*sizeof(uint8_t *));
1918 uint32_t *sortKeysLen = (uint32_t *)malloc(casesSize*sizeof(uint32_t));
1919 uint8_t prefixKey[256], suffixKey[256];
1920 uint32_t prefixKeyLen = 0, suffixKeyLen = 0, i = 0;
1921 UChar buffer[256];
1922 uint32_t unescapedLen = 0, l1 = 0, l2 = 0;
1923 UColAttributeValue strength;
1924
1925 log_verbose("ucol_mergeSortkeys test\n");
1926 log_verbose("Testing order of the test cases\n");
1927 genericLocaleStarter("en", cases, casesSize);
1928
1929 for(i = 0; i<casesSize; i++) {
1930 sortkeys[i] = (uint8_t *)malloc(256*sizeof(uint8_t));
1931 mergedPrefixkeys[i] = (uint8_t *)malloc(256*sizeof(uint8_t));
1932 mergedSuffixkeys[i] = (uint8_t *)malloc(256*sizeof(uint8_t));
1933 }
1934
1935 unescapedLen = u_unescape(prefix, buffer, 256);
1936 prefixKeyLen = ucol_getSortKey(coll, buffer, unescapedLen, prefixKey, 256);
1937
1938 unescapedLen = u_unescape(suffix, buffer, 256);
1939 suffixKeyLen = ucol_getSortKey(coll, buffer, unescapedLen, suffixKey, 256);
1940
1941 log_verbose("Massaging data with prefixes and different strengths\n");
1942 strength = UCOL_PRIMARY;
1943 while(strength <= UCOL_IDENTICAL) {
1944 log_verbose("Strength %s\n", strengthsC[strength<=UCOL_QUATERNARY?strength:4]);
1945 ucol_setAttribute(coll, UCOL_STRENGTH, strength, &status);
1946 for(i = 0; i<casesSize; i++) {
1947 unescapedLen = u_unescape(cases[i], buffer, 256);
1948 sortKeysLen[i] = ucol_getSortKey(coll, buffer, unescapedLen, sortkeys[i], 256);
1949 ucol_mergeSortkeys(prefixKey, prefixKeyLen, sortkeys[i], sortKeysLen[i], mergedPrefixkeys[i], 256);
1950 ucol_mergeSortkeys(sortkeys[i], sortKeysLen[i], suffixKey, suffixKeyLen, mergedSuffixkeys[i], 256);
1951 if(i>0) {
1952 if(tMemCmp(mergedPrefixkeys[i-1], mergedPrefixkeys[i]) >= 0) {
1953 log_err("Error while comparing prefixed keys @ strength %s:\n", strengthsC[strength<=UCOL_QUATERNARY?strength:4]);
1954 log_err("%s\n%s\n",
1955 ucol_sortKeyToString(coll, mergedPrefixkeys[i-1], outBuff1, l1),
1956 ucol_sortKeyToString(coll, mergedPrefixkeys[i], outBuff2, l2));
1957 }
1958 if(tMemCmp(mergedSuffixkeys[i-1], mergedSuffixkeys[i]) >= 0) {
1959 log_err("Error while comparing suffixed keys @ strength %s:\n", strengthsC[strength<=UCOL_QUATERNARY?strength:4]);
1960 log_err("%s\n%s\n",
1961 ucol_sortKeyToString(coll, mergedSuffixkeys[i-1], outBuff1, l1),
1962 ucol_sortKeyToString(coll, mergedSuffixkeys[i], outBuff2, l2));
1963 }
1964 }
1965 }
1966 if(strength == UCOL_QUATERNARY) {
1967 strength = UCOL_IDENTICAL;
1968 } else {
1969 strength++;
1970 }
1971 }
1972
1973 {
1974 uint8_t smallBuf[3];
1975 uint32_t reqLen = 0;
1976 log_verbose("testing buffer overflow\n");
1977 reqLen = ucol_mergeSortkeys(prefixKey, prefixKeyLen, suffixKey, suffixKeyLen, smallBuf, 3);
1978 if(reqLen != (prefixKeyLen+suffixKeyLen)) {
1979 log_err("Wrong preflight size for merged sortkey\n");
1980 }
1981 }
1982
1983 {
1984 UChar empty = 0;
1985 uint8_t emptyKey[20], abcKey[50], mergedKey[100];
1986 int32_t emptyKeyLen = 0, abcKeyLen = 0, mergedKeyLen = 0;
1987
1988 log_verbose("testing merging with sortkeys generated for empty strings\n");
1989 emptyKeyLen = ucol_getSortKey(coll, &empty, 0, emptyKey, 20);
1990 unescapedLen = u_unescape(cases[0], buffer, 256);
1991 abcKeyLen = ucol_getSortKey(coll, buffer, unescapedLen, abcKey, 50);
1992 mergedKeyLen = ucol_mergeSortkeys(emptyKey, emptyKeyLen, abcKey, abcKeyLen, mergedKey, 100);
1993 if(mergedKey[0] != 2) {
1994 log_err("Empty sortkey didn't produce a level separator\n");
1995 }
1996 /* try with zeros */
1997 mergedKeyLen = ucol_mergeSortkeys(emptyKey, 0, abcKey, abcKeyLen, mergedKey, 100);
1998 if(mergedKeyLen != 0 || mergedKey[0] != 0) {
1999 log_err("Empty key didn't produce null mergedKey\n");
2000 }
2001 mergedKeyLen = ucol_mergeSortkeys(abcKey, abcKeyLen, emptyKey, 0, mergedKey, 100);
2002 if(mergedKeyLen != 0 || mergedKey[0] != 0) {
2003 log_err("Empty key didn't produce null mergedKey\n");
2004 }
2005
2006 }
2007
2008 for(i = 0; i<casesSize; i++) {
2009 free(sortkeys[i]);
2010 free(mergedPrefixkeys[i]);
2011 free(mergedSuffixkeys[i]);
2012 }
2013 free(sortkeys);
2014 free(mergedPrefixkeys);
2015 free(mergedSuffixkeys);
2016 free(sortKeysLen);
2017 ucol_close(coll);
2018 /* need to finish this up */
2019 } else {
2020 log_data_err("Couldn't open collator");
2021 }
2022 }
TestShortString(void)2023 static void TestShortString(void)
2024 {
2025 struct {
2026 const char *input;
2027 const char *expectedOutput;
2028 const char *locale;
2029 UErrorCode expectedStatus;
2030 int32_t expectedOffset;
2031 uint32_t expectedIdentifier;
2032 } testCases[] = {
2033 /*
2034 * Note: The first test case sets variableTop to the dollar sign '$'.
2035 * We have agreed to drop support for variableTop in ucol_getShortDefinitionString(),
2036 * related to ticket #10372 "deprecate collation APIs for short definition strings",
2037 * and because it did not work for most spaces/punctuation/symbols,
2038 * as documented in ticket #10386 "collation short definition strings issues":
2039 * The old code wrote only 3 hex digits for primary weights below 0x0FFF,
2040 * which is a syntax error, and then failed to normalize the result.
2041 *
2042 * The "B2700" was removed from the expected result ("B2700_KPHONEBOOK_LDE").
2043 *
2044 * Previously, this test had to be adjusted for root collator changes because the
2045 * primary weight of the variable top character naturally changed
2046 * but was baked into the expected result.
2047 */
2048 {"LDE_RDE_KPHONEBOOK_T0024_ZLATN","KPHONEBOOK_LDE", "de@collation=phonebook", U_USING_FALLBACK_WARNING, 0, 0 },
2049
2050 {"LEN_RUS_NO_AS_S4","AS_LROOT_NO_S4", NULL, U_USING_DEFAULT_WARNING, 0, 0 },
2051 // uloc_canonicalize("de__PHONEBOOK") used to return "de@collation=phonebook"
2052 // and we got U_ZERO_ERROR.
2053 // Since ICU-20187 "drop support for long-obsolete locale ID variants..."
2054 // we actually load the "de__PHONEBOOK" bundle and fall back to "de".
2055 {"LDE_VPHONEBOOK_EO_SI","EO_KPHONEBOOK_LDE_SI", "de@collation=phonebook", U_USING_FALLBACK_WARNING, 0, 0 },
2056 {"LDE_Kphonebook","KPHONEBOOK_LDE", "de@collation=phonebook", U_ZERO_ERROR, 0, 0 },
2057 {"Xqde_DE@collation=phonebookq_S3_EX","KPHONEBOOK_LDE", "de@collation=phonebook", U_USING_FALLBACK_WARNING, 0, 0 },
2058 {"LFR_FO", "FO_LROOT", NULL, U_USING_DEFAULT_WARNING, 0, 0 },
2059 {"SO_LX_AS", "", NULL, U_ILLEGAL_ARGUMENT_ERROR, 8, 0 },
2060 {"S3_ASS_MMM", "", NULL, U_ILLEGAL_ARGUMENT_ERROR, 5, 0 }
2061 };
2062
2063 int32_t i = 0;
2064 UCollator *coll = NULL, *fromNormalized = NULL;
2065 UParseError parseError;
2066 UErrorCode status = U_ZERO_ERROR;
2067 char fromShortBuffer[256], normalizedBuffer[256], fromNormalizedBuffer[256];
2068 const char* locale = NULL;
2069
2070
2071 for(i = 0; i < UPRV_LENGTHOF(testCases); i++) {
2072 status = U_ZERO_ERROR;
2073 if(testCases[i].locale) {
2074 locale = testCases[i].locale;
2075 } else {
2076 locale = NULL;
2077 }
2078
2079 coll = ucol_openFromShortString(testCases[i].input, FALSE, &parseError, &status);
2080 if(status != testCases[i].expectedStatus) {
2081 log_err_status(status, "Got status '%s' that is different from expected '%s' for '%s'\n",
2082 u_errorName(status), u_errorName(testCases[i].expectedStatus), testCases[i].input);
2083 continue;
2084 }
2085
2086 if(U_SUCCESS(status)) {
2087 ucol_getShortDefinitionString(coll, locale, fromShortBuffer, 256, &status);
2088
2089 if(strcmp(fromShortBuffer, testCases[i].expectedOutput)) {
2090 log_err("Got short string '%s' from the collator. Expected '%s' for input '%s'\n",
2091 fromShortBuffer, testCases[i].expectedOutput, testCases[i].input);
2092 }
2093
2094 ucol_normalizeShortDefinitionString(testCases[i].input, normalizedBuffer, 256, &parseError, &status);
2095 fromNormalized = ucol_openFromShortString(normalizedBuffer, FALSE, &parseError, &status);
2096 ucol_getShortDefinitionString(fromNormalized, locale, fromNormalizedBuffer, 256, &status);
2097
2098 if(strcmp(fromShortBuffer, fromNormalizedBuffer)) {
2099 log_err("Strings obtained from collators instantiated by short string ('%s') and from normalized string ('%s') differ\n",
2100 fromShortBuffer, fromNormalizedBuffer);
2101 }
2102
2103
2104 if(!ucol_equals(coll, fromNormalized)) {
2105 log_err("Collator from short string ('%s') differs from one obtained through a normalized version ('%s')\n",
2106 testCases[i].input, normalizedBuffer);
2107 }
2108
2109 ucol_close(fromNormalized);
2110 ucol_close(coll);
2111
2112 } else {
2113 if(parseError.offset != testCases[i].expectedOffset) {
2114 log_err("Got parse error offset %i, but expected %i instead for '%s'\n",
2115 parseError.offset, testCases[i].expectedOffset, testCases[i].input);
2116 }
2117 }
2118 }
2119
2120 }
2121
2122 static void
doSetsTest(const char * locale,const USet * ref,USet * set,const char * inSet,const char * outSet,UErrorCode * status)2123 doSetsTest(const char *locale, const USet *ref, USet *set, const char* inSet, const char* outSet, UErrorCode *status) {
2124 UChar buffer[65536];
2125 int32_t bufLen;
2126
2127 uset_clear(set);
2128 bufLen = u_unescape(inSet, buffer, 512);
2129 uset_applyPattern(set, buffer, bufLen, 0, status);
2130 if(U_FAILURE(*status)) {
2131 log_err("%s: Failure setting pattern %s\n", locale, u_errorName(*status));
2132 }
2133
2134 if(!uset_containsAll(ref, set)) {
2135 log_err("%s: Some stuff from %s is not present in the set\n", locale, inSet);
2136 uset_removeAll(set, ref);
2137 bufLen = uset_toPattern(set, buffer, UPRV_LENGTHOF(buffer), TRUE, status);
2138 log_info(" missing: %s\n", aescstrdup(buffer, bufLen));
2139 bufLen = uset_toPattern(ref, buffer, UPRV_LENGTHOF(buffer), TRUE, status);
2140 log_info(" total: size=%i %s\n", uset_getItemCount(ref), aescstrdup(buffer, bufLen));
2141 }
2142
2143 uset_clear(set);
2144 bufLen = u_unescape(outSet, buffer, 512);
2145 uset_applyPattern(set, buffer, bufLen, 0, status);
2146 if(U_FAILURE(*status)) {
2147 log_err("%s: Failure setting pattern %s\n", locale, u_errorName(*status));
2148 }
2149
2150 if(!uset_containsNone(ref, set)) {
2151 log_err("%s: Some stuff from %s is present in the set\n", locale, outSet);
2152 }
2153 }
2154
2155
2156
2157
2158 static void
TestGetContractionsAndUnsafes(void)2159 TestGetContractionsAndUnsafes(void)
2160 {
2161 static struct {
2162 const char* locale;
2163 const char* inConts;
2164 const char* outConts;
2165 const char* inExp;
2166 const char* outExp;
2167 const char* unsafeCodeUnits;
2168 const char* safeCodeUnits;
2169 } tests[] = {
2170 { "ru",
2171 "[{\\u0418\\u0306}{\\u0438\\u0306}]",
2172 "[\\u0439\\u0457]",
2173 "[\\u00e6]",
2174 "[ae]",
2175 "[\\u0418\\u0438]",
2176 "[aAbB\\u0430\\u0410\\u0433\\u0413]"
2177 },
2178 { "uk",
2179 "[{\\u0406\\u0308}{\\u0456\\u0308}{\\u0418\\u0306}{\\u0438\\u0306}]",
2180 "[\\u0407\\u0419\\u0439\\u0457]",
2181 "[\\u00e6]",
2182 "[ae]",
2183 "[\\u0406\\u0456\\u0418\\u0438]",
2184 "[aAbBxv]",
2185 },
2186 { "sh",
2187 "[{C\\u0301}{C\\u030C}{C\\u0341}{DZ\\u030C}{Dz\\u030C}{D\\u017D}{D\\u017E}{lj}{nj}]",
2188 "[{\\u309d\\u3099}{\\u30fd\\u3099}]",
2189 "[\\u00e6]",
2190 "[a]",
2191 "[nlcdzNLCDZ]",
2192 "[jabv]"
2193 },
2194 { "ja",
2195 /*
2196 * The "collv2" builder omits mappings if the collator maps their
2197 * character sequences to the same CEs.
2198 * For example, it omits Japanese contractions for NFD forms
2199 * of the voiced iteration mark (U+309E = U+309D + U+3099), such as
2200 * {\\u3053\\u3099\\u309D\\u3099}{\\u3053\\u309D\\u3099}
2201 * {\\u30B3\\u3099\\u30FD\\u3099}{\\u30B3\\u30FD\\u3099}.
2202 * It does add mappings for the precomposed forms.
2203 */
2204 "[{\\u3053\\u3099\\u309D}{\\u3053\\u3099\\u309E}{\\u3053\\u3099\\u30FC}"
2205 "{\\u3053\\u309D}{\\u3053\\u309E}{\\u3053\\u30FC}"
2206 "{\\u30B3\\u3099\\u30FC}{\\u30B3\\u3099\\u30FD}{\\u30B3\\u3099\\u30FE}"
2207 "{\\u30B3\\u30FC}{\\u30B3\\u30FD}{\\u30B3\\u30FE}]",
2208 "[{\\u30FD\\u3099}{\\u309D\\u3099}{\\u3053\\u3099}{\\u30B3\\u3099}{lj}{nj}]",
2209 "[\\u30FE\\u00e6]",
2210 "[a]",
2211 "[\\u3099]",
2212 "[]"
2213 }
2214 };
2215
2216 UErrorCode status = U_ZERO_ERROR;
2217 UCollator *coll = NULL;
2218 int32_t i = 0;
2219 int32_t noConts = 0;
2220 USet *conts = uset_open(0,0);
2221 USet *exp = uset_open(0, 0);
2222 USet *set = uset_open(0,0);
2223 int32_t setBufferLen = 65536;
2224 UChar buffer[65536];
2225 int32_t setLen = 0;
2226
2227 for(i = 0; i < UPRV_LENGTHOF(tests); i++) {
2228 log_verbose("Testing locale: %s\n", tests[i].locale);
2229 coll = ucol_open(tests[i].locale, &status);
2230 if (coll == NULL || U_FAILURE(status)) {
2231 log_err_status(status, "Unable to open collator for locale %s ==> %s\n", tests[i].locale, u_errorName(status));
2232 continue;
2233 }
2234 ucol_getContractionsAndExpansions(coll, conts, exp, TRUE, &status);
2235 doSetsTest(tests[i].locale, conts, set, tests[i].inConts, tests[i].outConts, &status);
2236 setLen = uset_toPattern(conts, buffer, setBufferLen, TRUE, &status);
2237 if(U_SUCCESS(status)) {
2238 /*log_verbose("Contractions %i: %s\n", uset_getItemCount(conts), aescstrdup(buffer, setLen));*/
2239 } else {
2240 log_err("error %s. %i\n", u_errorName(status), setLen);
2241 status = U_ZERO_ERROR;
2242 }
2243 doSetsTest(tests[i].locale, exp, set, tests[i].inExp, tests[i].outExp, &status);
2244 setLen = uset_toPattern(exp, buffer, setBufferLen, TRUE, &status);
2245 if(U_SUCCESS(status)) {
2246 /*log_verbose("Expansions %i: %s\n", uset_getItemCount(exp), aescstrdup(buffer, setLen));*/
2247 } else {
2248 log_err("error %s. %i\n", u_errorName(status), setLen);
2249 status = U_ZERO_ERROR;
2250 }
2251
2252 noConts = ucol_getUnsafeSet(coll, conts, &status);
2253 (void)noConts; /* Suppress set but not used warning */
2254 doSetsTest(tests[i].locale, conts, set, tests[i].unsafeCodeUnits, tests[i].safeCodeUnits, &status);
2255 setLen = uset_toPattern(conts, buffer, setBufferLen, TRUE, &status);
2256 if(U_SUCCESS(status)) {
2257 log_verbose("Unsafe %i: %s\n", uset_getItemCount(exp), aescstrdup(buffer, setLen));
2258 } else {
2259 log_err("error %s. %i\n", u_errorName(status), setLen);
2260 status = U_ZERO_ERROR;
2261 }
2262
2263 ucol_close(coll);
2264 }
2265
2266
2267 uset_close(conts);
2268 uset_close(exp);
2269 uset_close(set);
2270 }
2271
2272 static void
TestOpenBinary(void)2273 TestOpenBinary(void)
2274 {
2275 /*
2276 * ucol_openBinary() documents:
2277 * "The API also takes a base collator which usually should be UCA."
2278 * and
2279 * "Currently it cannot be NULL."
2280 *
2281 * However, the check for NULL was commented out in ICU 3.4 (r18149).
2282 * Ticket #4355 requested "Make collation work with minimal data.
2283 * Optionally without UCA, with relevant parts of UCA copied into the tailoring table."
2284 *
2285 * The ICU team agreed with ticket #10517 "require base collator in ucol_openBinary() etc."
2286 * to require base!=NULL again.
2287 */
2288 #define OPEN_BINARY_ACCEPTS_NULL_BASE 0
2289 UErrorCode status = U_ZERO_ERROR;
2290 /*
2291 char rule[] = "&h < d < c < b";
2292 char *wUCA[] = { "a", "h", "d", "c", "b", "i" };
2293 char *noUCA[] = {"d", "c", "b", "a", "h", "i" };
2294 */
2295 /* we have to use Cyrillic letters because latin-1 always gets copied */
2296 const char rule[] = "&\\u0452 < \\u0434 < \\u0433 < \\u0432"; /* &dje < d < g < v */
2297 const char *wUCA[] = { "\\u0430", "\\u0452", "\\u0434", "\\u0433", "\\u0432", "\\u0435" }; /* a, dje, d, g, v, e */
2298 #if OPEN_BINARY_ACCEPTS_NULL_BASE
2299 const char *noUCA[] = {"\\u0434", "\\u0433", "\\u0432", "\\u0430", "\\u0435", "\\u0452" }; /* d, g, v, a, e, dje */
2300 #endif
2301
2302 UChar uRules[256];
2303 int32_t uRulesLen = u_unescape(rule, uRules, 256);
2304
2305 UCollator *coll = ucol_openRules(uRules, uRulesLen, UCOL_DEFAULT, UCOL_DEFAULT, NULL, &status);
2306 UCollator *UCA = NULL;
2307 UCollator *cloneNOUCA = NULL, *cloneWUCA = NULL;
2308
2309 uint8_t imageBuffer[32768];
2310 uint8_t *image = imageBuffer;
2311 int32_t imageBufferCapacity = 32768;
2312
2313 int32_t imageSize;
2314
2315 if((coll==NULL)||(U_FAILURE(status))) {
2316 log_data_err("could not load collators or error occured: %s\n",
2317 u_errorName(status));
2318 return;
2319 }
2320 UCA = ucol_open("root", &status);
2321 if((UCA==NULL)||(U_FAILURE(status))) {
2322 log_data_err("could not load UCA collator or error occured: %s\n",
2323 u_errorName(status));
2324 return;
2325 }
2326 imageSize = ucol_cloneBinary(coll, image, imageBufferCapacity, &status);
2327 if(U_FAILURE(status)) {
2328 image = (uint8_t *)malloc(imageSize*sizeof(uint8_t));
2329 status = U_ZERO_ERROR;
2330 imageSize = ucol_cloneBinary(coll, imageBuffer, imageSize, &status);
2331 }
2332
2333
2334 cloneWUCA = ucol_openBinary(image, imageSize, UCA, &status);
2335 cloneNOUCA = ucol_openBinary(image, imageSize, NULL, &status);
2336 #if !OPEN_BINARY_ACCEPTS_NULL_BASE
2337 if(status != U_ILLEGAL_ARGUMENT_ERROR) {
2338 log_err("ucol_openBinary(base=NULL) unexpectedly did not fail - %s\n", u_errorName(status));
2339 }
2340 #endif
2341
2342 genericOrderingTest(coll, wUCA, UPRV_LENGTHOF(wUCA));
2343
2344 genericOrderingTest(cloneWUCA, wUCA, UPRV_LENGTHOF(wUCA));
2345 #if OPEN_BINARY_ACCEPTS_NULL_BASE
2346 genericOrderingTest(cloneNOUCA, noUCA, UPRV_LENGTHOF(noUCA));
2347 #endif
2348
2349 if(image != imageBuffer) {
2350 free(image);
2351 }
2352 ucol_close(coll);
2353 ucol_close(cloneNOUCA);
2354 ucol_close(cloneWUCA);
2355 ucol_close(UCA);
2356 }
2357
TestDefault(void)2358 static void TestDefault(void) {
2359 /* Tests for code coverage. */
2360 UErrorCode status = U_ZERO_ERROR;
2361 UCollator *coll = ucol_open("es@collation=pinyin", &status);
2362 if (coll == NULL || status == U_FILE_ACCESS_ERROR) {
2363 log_data_err("Unable to open collator es@collation=pinyin\n");
2364 return;
2365 }
2366 if (status != U_USING_DEFAULT_WARNING) {
2367 /* What do you mean that you know about using pinyin collation in Spanish!? This should be in the zh locale. */
2368 log_err("es@collation=pinyin should return U_USING_DEFAULT_WARNING, but returned %s\n", u_errorName(status));
2369 }
2370 ucol_close(coll);
2371 if (ucol_getKeywordValues("funky", &status) != NULL) {
2372 log_err("Collators should not know about the funky keyword.\n");
2373 }
2374 if (status != U_ILLEGAL_ARGUMENT_ERROR) {
2375 log_err("funky keyword didn't fail as expected %s\n", u_errorName(status));
2376 }
2377 if (ucol_getKeywordValues("collation", &status) != NULL) {
2378 log_err("ucol_getKeywordValues should not work when given a bad status.\n");
2379 }
2380 }
2381
TestDefaultKeyword(void)2382 static void TestDefaultKeyword(void) {
2383 /* Tests for code coverage. */
2384 UErrorCode status = U_ZERO_ERROR;
2385 const char *loc = "zh_TW@collation=default";
2386 UCollator *coll = ucol_open(loc, &status);
2387 if(U_FAILURE(status)) {
2388 log_info("Warning: ucol_open(%s, ...) returned %s, at least it didn't crash.\n", loc, u_errorName(status));
2389 } else if (status != U_USING_FALLBACK_WARNING) {
2390 /* Hmm, skip the following test for CLDR 1.9 data and/or ICU 4.6, no longer seems to apply */
2391 #if 0
2392 log_err("ucol_open(%s, ...) should return an error or some sort of U_USING_FALLBACK_WARNING, but returned %s\n", loc, u_errorName(status));
2393 #endif
2394 }
2395 ucol_close(coll);
2396 }
2397
uenum_contains(UEnumeration * e,const char * s,UErrorCode * status)2398 static UBool uenum_contains(UEnumeration *e, const char *s, UErrorCode *status) {
2399 const char *t;
2400 uenum_reset(e, status);
2401 while(((t = uenum_next(e, NULL, status)) != NULL) && U_SUCCESS(*status)) {
2402 if(uprv_strcmp(s, t) == 0) {
2403 return TRUE;
2404 }
2405 }
2406 return FALSE;
2407 }
2408
TestGetKeywordValuesForLocale(void)2409 static void TestGetKeywordValuesForLocale(void) {
2410 #define MAX_NUMBER_OF_KEYWORDS 9
2411 const char *PREFERRED[][MAX_NUMBER_OF_KEYWORDS+1] = {
2412 { "und", "standard", "eor", "search", NULL, NULL, NULL, NULL, NULL, NULL },
2413 { "en_US", "standard", "eor", "search", NULL, NULL, NULL, NULL, NULL, NULL },
2414 { "en_029", "standard", "eor", "search", NULL, NULL, NULL, NULL, NULL, NULL },
2415 { "de_DE", "standard", "phonebook", "search", "eor", NULL, NULL, NULL, NULL, NULL },
2416 { "de_Latn_DE", "standard", "phonebook", "search", "eor", NULL, NULL, NULL, NULL, NULL },
2417 { "zh", "pinyin", "stroke", "eor", "search", "standard", NULL },
2418 { "zh_Hans", "pinyin", "stroke", "eor", "search", "standard", NULL },
2419 { "zh_CN", "pinyin", "stroke", "eor", "search", "standard", NULL },
2420 { "zh_Hant", "stroke", "pinyin", "eor", "search", "standard", NULL },
2421 { "zh_TW", "stroke", "pinyin", "eor", "search", "standard", NULL },
2422 { "zh__PINYIN", "pinyin", "stroke", "eor", "search", "standard", NULL },
2423 { "es_ES", "standard", "search", "traditional", "eor", NULL, NULL, NULL, NULL, NULL },
2424 { "es__TRADITIONAL","traditional", "search", "standard", "eor", NULL, NULL, NULL, NULL, NULL },
2425 { "und@collation=phonebook", "standard", "eor", "search", NULL, NULL, NULL, NULL, NULL, NULL },
2426 { "de_DE@collation=pinyin", "standard", "phonebook", "search", "eor", NULL, NULL, NULL, NULL, NULL },
2427 { "zzz@collation=xxx", "standard", "eor", "search", NULL, NULL, NULL, NULL, NULL, NULL }
2428 };
2429
2430 UErrorCode status = U_ZERO_ERROR;
2431 UEnumeration *keywordValues = NULL;
2432 int32_t i, n, size;
2433 const char *locale = NULL, *value = NULL;
2434 UBool errorOccurred = FALSE;
2435
2436 for (i = 0; i < UPRV_LENGTHOF(PREFERRED) && !errorOccurred; i++) {
2437 locale = PREFERRED[i][0];
2438 value = NULL;
2439 size = 0;
2440
2441 keywordValues = ucol_getKeywordValuesForLocale("collation", locale, TRUE, &status);
2442 if (keywordValues == NULL || U_FAILURE(status)) {
2443 log_err_status(status, "Error getting keyword values: %s\n", u_errorName(status));
2444 break;
2445 }
2446 size = uenum_count(keywordValues, &status);
2447 (void)size;
2448
2449 for (n = 0; (value = PREFERRED[i][n+1]) != NULL; n++) {
2450 if (!uenum_contains(keywordValues, value, &status)) {
2451 if (U_SUCCESS(status)) {
2452 log_err("Keyword value \"%s\" missing for locale: %s\n", value, locale);
2453 } else {
2454 log_err("While getting keyword value from locale: %s got this error: %s\n", locale, u_errorName(status));
2455 errorOccurred = TRUE;
2456 break;
2457 }
2458 }
2459 }
2460 uenum_close(keywordValues);
2461 keywordValues = NULL;
2462 }
2463 uenum_close(keywordValues);
2464 }
2465
TestStrcollNull(void)2466 static void TestStrcollNull(void) {
2467 UErrorCode status = U_ZERO_ERROR;
2468 UCollator *coll;
2469
2470 const UChar u16asc[] = {0x0049, 0x0042, 0x004D, 0};
2471 const int32_t u16ascLen = 3;
2472
2473 const UChar u16han[] = {0x5c71, 0x5ddd, 0};
2474 const int32_t u16hanLen = 2;
2475
2476 const char *u8asc = "\x49\x42\x4D";
2477 const int32_t u8ascLen = 3;
2478
2479 const char *u8han = "\xE5\xB1\xB1\xE5\xB7\x9D";
2480 const int32_t u8hanLen = 6;
2481
2482 coll = ucol_open(NULL, &status);
2483 if (U_FAILURE(status)) {
2484 log_err_status(status, "Default Collator creation failed.: %s\n", myErrorName(status));
2485 return;
2486 }
2487
2488 /* UChar API */
2489 if (ucol_strcoll(coll, NULL, 0, NULL, 0) != 0) {
2490 log_err("ERROR : ucol_strcoll NULL/0 and NULL/0");
2491 }
2492
2493 if (ucol_strcoll(coll, NULL, -1, NULL, 0) != 0) {
2494 /* No error arg, should return equal without crash */
2495 log_err("ERROR : ucol_strcoll NULL/-1 and NULL/0");
2496 }
2497
2498 if (ucol_strcoll(coll, u16asc, -1, NULL, 10) != 0) {
2499 /* No error arg, should return equal without crash */
2500 log_err("ERROR : ucol_strcoll u16asc/u16ascLen and NULL/10");
2501 }
2502
2503 if (ucol_strcoll(coll, u16asc, -1, NULL, 0) <= 0) {
2504 log_err("ERROR : ucol_strcoll u16asc/-1 and NULL/0");
2505 }
2506 if (ucol_strcoll(coll, NULL, 0, u16asc, -1) >= 0) {
2507 log_err("ERROR : ucol_strcoll NULL/0 and u16asc/-1");
2508 }
2509 if (ucol_strcoll(coll, u16asc, u16ascLen, NULL, 0) <= 0) {
2510 log_err("ERROR : ucol_strcoll u16asc/u16ascLen and NULL/0");
2511 }
2512
2513 if (ucol_strcoll(coll, u16han, -1, NULL, 0) <= 0) {
2514 log_err("ERROR : ucol_strcoll u16han/-1 and NULL/0");
2515 }
2516 if (ucol_strcoll(coll, NULL, 0, u16han, -1) >= 0) {
2517 log_err("ERROR : ucol_strcoll NULL/0 and u16han/-1");
2518 }
2519 if (ucol_strcoll(coll, NULL, 0, u16han, u16hanLen) >= 0) {
2520 log_err("ERROR : ucol_strcoll NULL/0 and u16han/u16hanLen");
2521 }
2522
2523 /* UTF-8 API */
2524 status = U_ZERO_ERROR;
2525 if (ucol_strcollUTF8(coll, NULL, 0, NULL, 0, &status) != 0 || U_FAILURE(status)) {
2526 log_err("ERROR : ucol_strcollUTF8 NULL/0 and NULL/0");
2527 }
2528 status = U_ZERO_ERROR;
2529 ucol_strcollUTF8(coll, NULL, -1, NULL, 0, &status);
2530 if (status != U_ILLEGAL_ARGUMENT_ERROR) {
2531 log_err("ERROR: ucol_strcollUTF8 NULL/-1 and NULL/0, should return U_ILLEGAL_ARGUMENT_ERROR");
2532 }
2533 status = U_ZERO_ERROR;
2534 ucol_strcollUTF8(coll, u8asc, u8ascLen, NULL, 10, &status);
2535 if (status != U_ILLEGAL_ARGUMENT_ERROR) {
2536 log_err("ERROR: ucol_strcollUTF8 u8asc/u8ascLen and NULL/10, should return U_ILLEGAL_ARGUMENT_ERROR");
2537 }
2538
2539 status = U_ZERO_ERROR;
2540 if (ucol_strcollUTF8(coll, u8asc, -1, NULL, 0, &status) <= 0 || U_FAILURE(status)) {
2541 log_err("ERROR : ucol_strcollUTF8 u8asc/-1 and NULL/0");
2542 }
2543 status = U_ZERO_ERROR;
2544 if (ucol_strcollUTF8(coll, NULL, 0, u8asc, -1, &status) >= 0 || U_FAILURE(status)) {
2545 log_err("ERROR : ucol_strcollUTF8 NULL/0 and u8asc/-1");
2546 }
2547 status = U_ZERO_ERROR;
2548 if (ucol_strcollUTF8(coll, u8asc, u8ascLen, NULL, 0, &status) <= 0 || U_FAILURE(status)) {
2549 log_err("ERROR : ucol_strcollUTF8 u8asc/u8ascLen and NULL/0");
2550 }
2551
2552 status = U_ZERO_ERROR;
2553 if (ucol_strcollUTF8(coll, u8han, -1, NULL, 0, &status) <= 0 || U_FAILURE(status)) {
2554 log_err("ERROR : ucol_strcollUTF8 u8han/-1 and NULL/0");
2555 }
2556 status = U_ZERO_ERROR;
2557 if (ucol_strcollUTF8(coll, NULL, 0, u8han, -1, &status) >= 0 || U_FAILURE(status)) {
2558 log_err("ERROR : ucol_strcollUTF8 NULL/0 and u8han/-1");
2559 }
2560 status = U_ZERO_ERROR;
2561 if (ucol_strcollUTF8(coll, NULL, 0, u8han, u8hanLen, &status) >= 0 || U_FAILURE(status)) {
2562 log_err("ERROR : ucol_strcollUTF8 NULL/0 and u8han/u8hanLen");
2563 }
2564
2565 ucol_close(coll);
2566 }
2567
2568 #endif /* #if !UCONFIG_NO_COLLATION */
2569