• Home
  • Line#
  • Scopes#
  • Navigate#
  • Raw
  • Download
1 /********************************************************************
2  * COPYRIGHT:
3  * Copyright (c) 1997-2007, International Business Machines Corporation and
4  * others. All Rights Reserved.
5  ********************************************************************/
6 /********************************************************************************
7 *
8 * File CAPITEST.C
9 *
10 * Modification History:
11 *        Name                     Description
12 *     Madhu Katragadda             Ported for C API
13 *********************************************************************************
14 *//* C API TEST For COLLATOR */
15 
16 #include "unicode/utypes.h"
17 
18 #if !UCONFIG_NO_COLLATION
19 
20 #include <stdio.h>
21 #include <stdlib.h>
22 #include <string.h>
23 #include "unicode/uloc.h"
24 #include "unicode/ustring.h"
25 #include "unicode/ures.h"
26 #include "unicode/ucoleitr.h"
27 #include "cintltst.h"
28 #include "capitst.h"
29 #include "ccolltst.h"
30 #include "putilimp.h"
31 
32 static void TestAttribute(void);
33 static void TestDefault(void);
34         int TestBufferSize();    /* defined in "colutil.c" */
35 
36 
37 
38 
39 /* next two function is modified from "i18n/ucol.cpp" to avoid include "ucol_imp.h" */
uprv_appendByteToHexString(char * dst,uint8_t val)40 static void uprv_appendByteToHexString(char *dst, uint8_t val) {
41   uint32_t len = (uint32_t)strlen(dst);
42   sprintf(dst+len, "%02X", val);
43 }
44 
ucol_sortKeyToString(const UCollator * coll,const uint8_t * sortkey,char * buffer,uint32_t * len)45 static char* U_EXPORT2 ucol_sortKeyToString(const UCollator *coll, const uint8_t *sortkey, char *buffer, uint32_t *len) {
46   int32_t strength = UCOL_PRIMARY;
47   uint32_t res_size = 0;
48   UBool doneCase = FALSE;
49 
50   char *current = buffer;
51   const uint8_t *currentSk = sortkey;
52 
53   UErrorCode error_code = U_ZERO_ERROR;
54 
55   strcpy(current, "[");
56 
57   while(strength <= UCOL_QUATERNARY && strength <= ucol_getAttribute(coll,UCOL_STRENGTH, &error_code)) {
58     if(U_FAILURE(error_code)) {
59       log_err("ucol_getAttribute returned error: %s\n", u_errorName(error_code));
60     }
61     if(strength > UCOL_PRIMARY) {
62       strcat(current, " . ");
63     }
64     while(*currentSk != 0x01 && *currentSk != 0x00) { /* print a level */
65       uprv_appendByteToHexString(current, *currentSk++);
66       strcat(current, " ");
67     }
68     if(ucol_getAttribute(coll,UCOL_CASE_LEVEL, &error_code) == UCOL_ON && strength == UCOL_SECONDARY && doneCase == FALSE) {
69         doneCase = TRUE;
70     } else if(ucol_getAttribute(coll,UCOL_CASE_LEVEL, &error_code) == UCOL_OFF || doneCase == TRUE || strength != UCOL_SECONDARY) {
71       strength ++;
72     }
73     if(U_FAILURE(error_code)) {
74       log_err("ucol_getAttribute returned error: %s\n", u_errorName(error_code));
75     }
76     uprv_appendByteToHexString(current, *currentSk++); /* This should print '01' */
77     if(strength == UCOL_QUATERNARY && ucol_getAttribute(coll,UCOL_ALTERNATE_HANDLING, &error_code) == UCOL_NON_IGNORABLE) {
78       break;
79     }
80   }
81 
82   if(ucol_getAttribute(coll,UCOL_STRENGTH, &error_code) == UCOL_IDENTICAL) {
83     strcat(current, " . ");
84     while(*currentSk != 0) {
85       uprv_appendByteToHexString(current, *currentSk++);
86       strcat(current, " ");
87     }
88 
89     uprv_appendByteToHexString(current, *currentSk++);
90   }
91   if(U_FAILURE(error_code)) {
92     log_err("ucol_getAttribute returned error: %s\n", u_errorName(error_code));
93   }
94   strcat(current, "]");
95 
96   if(res_size > *len) {
97     return NULL;
98   }
99 
100   return buffer;
101 }
102 /* end of  avoid include "ucol_imp.h" */
103 
104 
addCollAPITest(TestNode ** root)105 void addCollAPITest(TestNode** root)
106 {
107     /* WEIVTODO: return tests here */
108     addTest(root, &TestProperty,      "tscoll/capitst/TestProperty");
109     addTest(root, &TestRuleBasedColl, "tscoll/capitst/TestRuleBasedColl");
110     addTest(root, &TestCompare,       "tscoll/capitst/TestCompare");
111     addTest(root, &TestSortKey,       "tscoll/capitst/TestSortKey");
112     addTest(root, &TestHashCode,      "tscoll/capitst/TestHashCode");
113     addTest(root, &TestElemIter,      "tscoll/capitst/TestElemIter");
114     addTest(root, &TestGetAll,        "tscoll/capitst/TestGetAll");
115     /*addTest(root, &TestGetDefaultRules, "tscoll/capitst/TestGetDefaultRules");*/
116     addTest(root, &TestDecomposition, "tscoll/capitst/TestDecomposition");
117     addTest(root, &TestSafeClone, "tscoll/capitst/TestSafeClone");
118     addTest(root, &TestCloneBinary, "tscoll/capitst/TestCloneBinary");
119     addTest(root, &TestGetSetAttr, "tscoll/capitst/TestGetSetAttr");
120     addTest(root, &TestBounds, "tscoll/capitst/TestBounds");
121     addTest(root, &TestGetLocale, "tscoll/capitst/TestGetLocale");
122     addTest(root, &TestSortKeyBufferOverrun, "tscoll/capitst/TestSortKeyBufferOverrun");
123     addTest(root, &TestAttribute, "tscoll/capitst/TestAttribute");
124     addTest(root, &TestGetTailoredSet, "tscoll/capitst/TestGetTailoredSet");
125     addTest(root, &TestMergeSortKeys, "tscoll/capitst/TestMergeSortKeys");
126     addTest(root, &TestShortString, "tscoll/capitst/TestShortString");
127     addTest(root, &TestGetContractionsAndUnsafes, "tscoll/capitst/TestGetContractionsAndUnsafes");
128     addTest(root, &TestOpenBinary, "tscoll/capitst/TestOpenBinary");
129     addTest(root, &TestDefault, "tscoll/capitst/TestDefault");
130 }
131 
TestGetSetAttr(void)132 void TestGetSetAttr(void) {
133   UErrorCode status = U_ZERO_ERROR;
134   UCollator *coll = ucol_open(NULL, &status);
135   struct attrTest {
136     UColAttribute att;
137     UColAttributeValue val[5];
138     uint32_t valueSize;
139     UColAttributeValue nonValue;
140   } attrs[] = {
141     {UCOL_FRENCH_COLLATION, {UCOL_ON, UCOL_OFF}, 2, UCOL_SHIFTED},
142     {UCOL_ALTERNATE_HANDLING, {UCOL_NON_IGNORABLE, UCOL_SHIFTED}, 2, UCOL_OFF},/* attribute for handling variable elements*/
143     {UCOL_CASE_FIRST, {UCOL_OFF, UCOL_LOWER_FIRST, UCOL_UPPER_FIRST}, 3, UCOL_SHIFTED},/* who goes first, lower case or uppercase */
144     {UCOL_CASE_LEVEL, {UCOL_ON, UCOL_OFF}, 2, UCOL_SHIFTED},/* do we have an extra case level */
145     {UCOL_NORMALIZATION_MODE, {UCOL_ON, UCOL_OFF}, 2, UCOL_SHIFTED},/* attribute for normalization */
146     {UCOL_DECOMPOSITION_MODE, {UCOL_ON, UCOL_OFF}, 2, UCOL_SHIFTED},
147     {UCOL_STRENGTH,         {UCOL_PRIMARY, UCOL_SECONDARY, UCOL_TERTIARY, UCOL_QUATERNARY, UCOL_IDENTICAL}, 5, UCOL_SHIFTED},/* attribute for strength */
148     {UCOL_HIRAGANA_QUATERNARY_MODE, {UCOL_ON, UCOL_OFF}, 2, UCOL_SHIFTED},/* when turned on, this attribute */
149   };
150   UColAttribute currAttr;
151   UColAttributeValue value;
152   uint32_t i = 0, j = 0;
153 
154   for(i = 0; i<sizeof(attrs)/sizeof(attrs[0]); i++) {
155     currAttr = attrs[i].att;
156     ucol_setAttribute(coll, currAttr, UCOL_DEFAULT, &status);
157     if(U_FAILURE(status)) {
158       log_err("ucol_setAttribute with the default value returned error: %s\n", u_errorName(status));
159       break;
160     }
161     value = ucol_getAttribute(coll, currAttr, &status);
162     if(U_FAILURE(status)) {
163       log_err("ucol_getAttribute returned error: %s\n", u_errorName(status));
164       break;
165     }
166     for(j = 0; j<attrs[i].valueSize; j++) {
167       ucol_setAttribute(coll, currAttr, attrs[i].val[j], &status);
168       if(U_FAILURE(status)) {
169         log_err("ucol_setAttribute with the value %i returned error: %s\n", attrs[i].val[j], u_errorName(status));
170         break;
171       }
172     }
173     status = U_ZERO_ERROR;
174     ucol_setAttribute(coll, currAttr, attrs[i].nonValue, &status);
175     if(U_SUCCESS(status)) {
176       log_err("ucol_setAttribute with the bad value didn't return an error\n");
177       break;
178     }
179     status = U_ZERO_ERROR;
180 
181     ucol_setAttribute(coll, currAttr, value, &status);
182     if(U_FAILURE(status)) {
183       log_err("ucol_setAttribute with the default valuereturned error: %s\n", u_errorName(status));
184       break;
185     }
186 
187   }
188   status = U_ZERO_ERROR;
189   value = ucol_getAttribute(coll, UCOL_ATTRIBUTE_COUNT, &status);
190   if(U_SUCCESS(status)) {
191     log_err("ucol_getAttribute for UCOL_ATTRIBUTE_COUNT didn't return an error\n");
192   }
193   status = U_ZERO_ERROR;
194   ucol_setAttribute(coll, UCOL_ATTRIBUTE_COUNT, UCOL_DEFAULT, &status);
195   if(U_SUCCESS(status)) {
196     log_err("ucol_setAttribute for UCOL_ATTRIBUTE_COUNT didn't return an error\n");
197   }
198   status = U_ZERO_ERROR;
199   ucol_close(coll);
200 }
201 
202 
doAssert(int condition,const char * message)203 static void doAssert(int condition, const char *message)
204 {
205     if (condition==0) {
206         log_err("ERROR :  %s\n", message);
207     }
208 }
209 
210 #if 0
211 /* We don't have default rules, at least not in the previous sense */
212 void TestGetDefaultRules(){
213     uint32_t size=0;
214     UErrorCode status=U_ZERO_ERROR;
215     UCollator *coll=NULL;
216     int32_t len1 = 0, len2=0;
217     uint8_t *binColData = NULL;
218 
219     UResourceBundle *res = NULL;
220     UResourceBundle *binColl = NULL;
221     uint8_t *binResult = NULL;
222 
223 
224     const UChar * defaultRulesArray=ucol_getDefaultRulesArray(&size);
225     log_verbose("Test the function ucol_getDefaultRulesArray()\n");
226 
227     coll = ucol_openRules(defaultRulesArray, size, UCOL_ON, UCOL_PRIMARY, &status);
228     if(U_SUCCESS(status) && coll !=NULL) {
229         binColData = (uint8_t*)ucol_cloneRuleData(coll, &len1, &status);
230 
231     }
232 
233 
234     status=U_ZERO_ERROR;
235     res=ures_open(NULL, "root", &status);
236     if(U_FAILURE(status)){
237         log_err("ERROR: Failed to get resource for \"root Locale\" with %s", myErrorName(status));
238         return;
239     }
240     binColl=ures_getByKey(res, "%%Collation", binColl, &status);
241     if(U_SUCCESS(status)){
242         binResult=(uint8_t*)ures_getBinary(binColl,  &len2, &status);
243         if(U_FAILURE(status)){
244             log_err("ERROR: ures_getBinary() failed\n");
245         }
246     }else{
247         log_err("ERROR: ures_getByKey(locale(default), %%Collation) failed");
248     }
249 
250 
251     if(len1 != len2){
252         log_err("Error: ucol_getDefaultRulesArray() failed to return the correct length.\n");
253     }
254     if(memcmp(binColData, binResult, len1) != 0){
255         log_err("Error: ucol_getDefaultRulesArray() failed\n");
256     }
257 
258     free(binColData);
259     ures_close(binColl);
260     ures_close(res);
261     ucol_close(coll);
262 
263 }
264 #endif
265 
266 /* Collator Properties
267  ucol_open, ucol_strcoll,  getStrength/setStrength
268  getDecomposition/setDecomposition, getDisplayName*/
TestProperty()269 void TestProperty()
270 {
271     UCollator *col, *ruled;
272     UChar *disName;
273     int32_t len = 0, i = 0;
274     UChar *source, *target;
275     int32_t tempLength;
276     UErrorCode status = U_ZERO_ERROR;
277     /*
278       All the collations have the same version in an ICU
279       version.
280       ICU 2.0 currVersionArray = {0x18, 0xC0, 0x02, 0x02};
281       ICU 2.1 currVersionArray = {0x19, 0x00, 0x03, 0x03};
282       ICU 2.2 currVersionArray = {0x21, 0x40, 0x04, 0x04};
283       ICU 2.4 currVersionArray = {0x21, 0x40, 0x04, 0x04};
284       ICU 2.6 currVersionArray = {0x21, 0x40, 0x03, 0x03};
285       ICU 2.8 currVersionArray = {0x29, 0x80, 0x00, 0x04};
286       ICU 3.4 currVersionArray = {0x31, 0xC0, 0x00, 0x04};
287     */
288     UVersionInfo currVersionArray = {0x31, 0xC0, 0x00, 0x05};
289     /* ICU 3.4 had UCA 4.1 */
290     /*UVersionInfo currUCAVersionArray = {4, 1, 0, 0};*/
291     UVersionInfo currUCAVersionArray = {5, 0, 0, 0};
292     UVersionInfo versionArray = {0, 0, 0, 0};
293     UVersionInfo versionUCAArray = {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("Default Collator creation failed.: %s\n", myErrorName(status));
300         return;
301     }
302 
303     ucol_getVersion(col, versionArray);
304     for (i=0; i<4; ++i) {
305       if (versionArray[i] != currVersionArray[i]) {
306         log_err("Testing ucol_getVersion() - unexpected result: %hu.%hu.%hu.%hu\n",
307             versionArray[0], versionArray[1], versionArray[2], versionArray[3]);
308         break;
309       }
310     }
311 
312     ucol_getUCAVersion(col, versionUCAArray);
313     for (i=0; i<4; ++i) {
314       if (versionUCAArray[i] != currUCAVersionArray[i]) {
315         log_err("Testing ucol_getUCAVersion() - unexpected result: %hu.%hu.%hu.%hu\n",
316             versionUCAArray[0], versionUCAArray[1], versionUCAArray[2], versionUCAArray[3]);
317         break;
318       }
319     }
320 
321     source=(UChar*)malloc(sizeof(UChar) * 12);
322     target=(UChar*)malloc(sizeof(UChar) * 12);
323 
324 
325     u_uastrcpy(source, "ab");
326     u_uastrcpy(target, "abc");
327 
328     doAssert((ucol_strcoll(col, source, u_strlen(source), target, u_strlen(target)) == UCOL_LESS), "ab < abc comparison failed");
329 
330     u_uastrcpy(source, "ab");
331     u_uastrcpy(target, "AB");
332 
333     doAssert((ucol_strcoll(col, source, u_strlen(source), target, u_strlen(target)) == UCOL_LESS), "ab < AB comparison failed");
334 /*    u_uastrcpy(source, "black-bird");
335     u_uastrcpy(target, "blackbird"); */
336     u_uastrcpy(target, "black-bird");
337     u_uastrcpy(source, "blackbird");
338 
339     doAssert((ucol_strcoll(col, source, u_strlen(source), target, u_strlen(target)) == UCOL_GREATER),
340         "black-bird > blackbird comparison failed");
341     u_uastrcpy(source, "black bird");
342     u_uastrcpy(target, "black-bird");
343     doAssert((ucol_strcoll(col, source, u_strlen(source), target, u_strlen(target)) == UCOL_LESS),
344         "black bird < black-bird comparison failed");
345     u_uastrcpy(source, "Hello");
346     u_uastrcpy(target, "hello");
347 
348     doAssert((ucol_strcoll(col, source, u_strlen(source), target, u_strlen(target)) == UCOL_GREATER),
349         "Hello > hello comparison failed");
350     free(source);
351     free(target);
352     log_verbose("Test ucol_strcoll ends.\n");
353 
354     log_verbose("testing ucol_getStrength() method ...\n");
355     doAssert( (ucol_getStrength(col) == UCOL_TERTIARY), "collation object has the wrong strength");
356     doAssert( (ucol_getStrength(col) != UCOL_PRIMARY), "collation object's strength is primary difference");
357 
358     log_verbose("testing ucol_setStrength() method ...\n");
359     ucol_setStrength(col, UCOL_SECONDARY);
360     doAssert( (ucol_getStrength(col) != UCOL_TERTIARY), "collation object's strength is secondary difference");
361     doAssert( (ucol_getStrength(col) != UCOL_PRIMARY), "collation object's strength is primary difference");
362     doAssert( (ucol_getStrength(col) == UCOL_SECONDARY), "collation object has the wrong strength");
363 
364 
365     log_verbose("Get display name for the default collation in German : \n");
366 
367     len=ucol_getDisplayName("en_US", "de_DE", NULL, 0,  &status);
368     if(status==U_BUFFER_OVERFLOW_ERROR){
369         status=U_ZERO_ERROR;
370         disName=(UChar*)malloc(sizeof(UChar) * (len+1));
371         ucol_getDisplayName("en_US", "de_DE", disName, len+1,  &status);
372         log_verbose("the display name for default collation in german: %s\n", austrdup(disName) );
373         free(disName);
374     }
375     if(U_FAILURE(status)){
376         log_err("ERROR: in getDisplayName: %s\n", myErrorName(status));
377         return;
378     }
379     log_verbose("Default collation getDisplayName ended.\n");
380 
381     ruled = ucol_open("da_DK", &status);
382     log_verbose("ucol_getRules() testing ...\n");
383     ucol_getRules(ruled, &tempLength);
384     doAssert( tempLength != 0, "getRules() result incorrect" );
385     log_verbose("getRules tests end.\n");
386     {
387         UChar *buffer = (UChar *)malloc(200000*sizeof(UChar));
388         int32_t bufLen = 200000;
389         buffer[0] = '\0';
390         log_verbose("ucol_getRulesEx() testing ...\n");
391         tempLength = ucol_getRulesEx(col,UCOL_TAILORING_ONLY,buffer,bufLen );
392         doAssert( tempLength == 0x0a, "getRulesEx() result incorrect" );
393         log_verbose("getRules tests end.\n");
394 
395         log_verbose("ucol_getRulesEx() testing ...\n");
396         tempLength=ucol_getRulesEx(col,UCOL_FULL_RULES,buffer,bufLen );
397         doAssert( tempLength != 0, "getRulesEx() result incorrect" );
398         log_verbose("getRules tests end.\n");
399         free(buffer);
400     }
401     ucol_close(ruled);
402     ucol_close(col);
403 
404     log_verbose("open an collator for french locale");
405     col = ucol_open("fr_FR", &status);
406     if (U_FAILURE(status)) {
407        log_err("ERROR: Creating French collation failed.: %s\n", myErrorName(status));
408         return;
409     }
410     ucol_setStrength(col, UCOL_PRIMARY);
411     log_verbose("testing ucol_getStrength() method again ...\n");
412     doAssert( (ucol_getStrength(col) != UCOL_TERTIARY), "collation object has the wrong strength");
413     doAssert( (ucol_getStrength(col) == UCOL_PRIMARY), "collation object's strength is not primary difference");
414 
415     log_verbose("testing French ucol_setStrength() method ...\n");
416     ucol_setStrength(col, UCOL_TERTIARY);
417     doAssert( (ucol_getStrength(col) == UCOL_TERTIARY), "collation object's strength is not tertiary difference");
418     doAssert( (ucol_getStrength(col) != UCOL_PRIMARY), "collation object's strength is primary difference");
419     doAssert( (ucol_getStrength(col) != UCOL_SECONDARY), "collation object's strength is secondary difference");
420     ucol_close(col);
421 
422     log_verbose("Get display name for the french collation in english : \n");
423     len=ucol_getDisplayName("fr_FR", "en_US", NULL, 0,  &status);
424     if(status==U_BUFFER_OVERFLOW_ERROR){
425         status=U_ZERO_ERROR;
426         disName=(UChar*)malloc(sizeof(UChar) * (len+1));
427         ucol_getDisplayName("fr_FR", "en_US", disName, len+1,  &status);
428         log_verbose("the display name for french collation in english: %s\n", austrdup(disName) );
429         free(disName);
430     }
431     if(U_FAILURE(status)){
432         log_err("ERROR: in getDisplayName: %s\n", myErrorName(status));
433         return;
434     }
435     log_verbose("Default collation getDisplayName ended.\n");
436 
437 }
438 
439 /* Test RuleBasedCollator and getRules*/
TestRuleBasedColl()440 void TestRuleBasedColl()
441 {
442     UCollator *col1, *col2, *col3, *col4;
443     UCollationElements *iter1, *iter2;
444     UChar ruleset1[60];
445     UChar ruleset2[50];
446     UChar teststr[10];
447     UChar teststr2[10];
448     const UChar *rule1, *rule2, *rule3, *rule4;
449     int32_t tempLength;
450     UErrorCode status = U_ZERO_ERROR;
451     u_uastrcpy(ruleset1, "&9 < a, A < b, B < c, C; ch, cH, Ch, CH < d, D, e, E");
452     u_uastrcpy(ruleset2, "&9 < a, A < b, B < c, C < d, D, e, E");
453 
454 
455     col1 = ucol_openRules(ruleset1, u_strlen(ruleset1), UCOL_DEFAULT, UCOL_DEFAULT_STRENGTH, NULL,&status);
456     if (U_FAILURE(status)) {
457         log_err("RuleBased Collator creation failed.: %s\n", myErrorName(status));
458         return;
459     }
460     else
461         log_verbose("PASS: RuleBased Collator creation passed\n");
462 
463     status = U_ZERO_ERROR;
464     col2 = ucol_openRules(ruleset2, u_strlen(ruleset2),  UCOL_DEFAULT, UCOL_DEFAULT_STRENGTH, NULL, &status);
465     if (U_FAILURE(status)) {
466         log_err("RuleBased Collator creation failed.: %s\n", myErrorName(status));
467         return;
468     }
469     else
470         log_verbose("PASS: RuleBased Collator creation passed\n");
471 
472 
473     status = U_ZERO_ERROR;
474     col3= ucol_open(NULL, &status);
475     if (U_FAILURE(status)) {
476         log_err("Default Collator creation failed.: %s\n", myErrorName(status));
477         return;
478     }
479     else
480         log_verbose("PASS: Default Collator creation passed\n");
481 
482     rule1 = ucol_getRules(col1, &tempLength);
483     rule2 = ucol_getRules(col2, &tempLength);
484     rule3 = ucol_getRules(col3, &tempLength);
485 
486     doAssert((u_strcmp(rule1, rule2) != 0), "Default collator getRules failed");
487     doAssert((u_strcmp(rule2, rule3) != 0), "Default collator getRules failed");
488     doAssert((u_strcmp(rule1, rule3) != 0), "Default collator getRules failed");
489 
490     col4=ucol_openRules(rule2, u_strlen(rule2), UCOL_DEFAULT, UCOL_DEFAULT_STRENGTH, NULL, &status);
491     if (U_FAILURE(status)) {
492         log_err("RuleBased Collator creation failed.: %s\n", myErrorName(status));
493         return;
494     }
495     rule4= ucol_getRules(col4, &tempLength);
496     doAssert((u_strcmp(rule2, rule4) == 0), "Default collator getRules failed");
497 
498     ucol_close(col1);
499     ucol_close(col2);
500     ucol_close(col3);
501     ucol_close(col4);
502 
503     /* tests that modifier ! is always ignored */
504     u_uastrcpy(ruleset1, "!&a<b");
505     teststr[0] = 0x0e40;
506     teststr[1] = 0x0e01;
507     teststr[2] = 0x0e2d;
508     col1 = ucol_openRules(ruleset1, u_strlen(ruleset1), UCOL_DEFAULT, UCOL_DEFAULT_STRENGTH, NULL, &status);
509     if (U_FAILURE(status)) {
510         log_err("RuleBased Collator creation failed.: %s\n", myErrorName(status));
511         return;
512     }
513     col2 = ucol_open("en_US", &status);
514     if (U_FAILURE(status)) {
515         log_err("en_US Collator creation failed.: %s\n", myErrorName(status));
516         return;
517     }
518     iter1 = ucol_openElements(col1, teststr, 3, &status);
519     iter2 = ucol_openElements(col2, teststr, 3, &status);
520     if(U_FAILURE(status)) {
521         log_err("ERROR: CollationElement iterator creation failed.: %s\n", myErrorName(status));
522         return;
523     }
524     while (TRUE) {
525         /* testing with en since thai has its own tailoring */
526         uint32_t ce = ucol_next(iter1, &status);
527         uint32_t ce2 = ucol_next(iter2, &status);
528         if(U_FAILURE(status)) {
529             log_err("ERROR: CollationElement iterator creation failed.: %s\n", myErrorName(status));
530             return;
531         }
532         if (ce2 != ce) {
533              log_err("! modifier test failed");
534         }
535         if (ce == UCOL_NULLORDER) {
536             break;
537         }
538     }
539     ucol_closeElements(iter1);
540     ucol_closeElements(iter2);
541     ucol_close(col1);
542     ucol_close(col2);
543     /* test that we can start a rule without a & or < */
544     u_uastrcpy(ruleset1, "< z < a");
545     col1 = ucol_openRules(ruleset1, u_strlen(ruleset1), UCOL_DEFAULT, UCOL_DEFAULT_STRENGTH, NULL, &status);
546     if (U_FAILURE(status)) {
547         log_err("RuleBased Collator creation failed.: %s\n", myErrorName(status));
548         return;
549     }
550     u_uastrcpy(teststr, "z");
551     u_uastrcpy(teststr2, "a");
552     if (ucol_greaterOrEqual(col1, teststr, 1, teststr2, 1)) {
553         log_err("Rule \"z < a\" fails");
554     }
555     ucol_close(col1);
556 }
557 
TestCompare()558 void TestCompare()
559 {
560     UErrorCode status = U_ZERO_ERROR;
561     UCollator *col;
562     UChar* test1;
563     UChar* test2;
564 
565     log_verbose("The compare tests begin : \n");
566     status=U_ZERO_ERROR;
567     col = ucol_open("en_US", &status);
568     if(U_FAILURE(status)) {
569         log_err("ucal_open() collation creation failed.: %s\n", myErrorName(status));
570         return;
571     }
572     test1=(UChar*)malloc(sizeof(UChar) * 6);
573     test2=(UChar*)malloc(sizeof(UChar) * 6);
574     u_uastrcpy(test1, "Abcda");
575     u_uastrcpy(test2, "abcda");
576 
577     log_verbose("Use tertiary comparison level testing ....\n");
578 
579     doAssert( (!ucol_equal(col, test1, u_strlen(test1), test2, u_strlen(test2))), "Result should be \"Abcda\" != \"abcda\" ");
580     doAssert( (ucol_greater(col, test1, u_strlen(test1), test2, u_strlen(test2))), "Result should be \"Abcda\" >>> \"abcda\" ");
581     doAssert( (ucol_greaterOrEqual(col, test1, u_strlen(test1), test2, u_strlen(test2))), "Result should be \"Abcda\" >>> \"abcda\"");
582 
583     ucol_setStrength(col, UCOL_SECONDARY);
584     log_verbose("Use secondary comparison level testing ....\n");
585 
586     doAssert( (ucol_equal(col, test1, u_strlen(test1), test2, u_strlen(test2) )), "Result should be \"Abcda\" == \"abcda\"");
587     doAssert( (!ucol_greater(col, test1, u_strlen(test1), test2, u_strlen(test2))), "Result should be \"Abcda\" == \"abcda\"");
588     doAssert( (ucol_greaterOrEqual(col, test1, u_strlen(test1), test2, u_strlen(test2) )), "Result should be \"Abcda\" == \"abcda\"");
589 
590     ucol_setStrength(col, UCOL_PRIMARY);
591     log_verbose("Use primary comparison level testing ....\n");
592 
593     doAssert( (ucol_equal(col, test1, u_strlen(test1), test2, u_strlen(test2))), "Result should be \"Abcda\" == \"abcda\"");
594     doAssert( (!ucol_greater(col, test1, u_strlen(test1), test2, u_strlen(test2))), "Result should be \"Abcda\" == \"abcda\"");
595     doAssert( (ucol_greaterOrEqual(col, test1, u_strlen(test1), test2, u_strlen(test2))), "Result should be \"Abcda\" == \"abcda\"");
596 
597 
598     log_verbose("The compare tests end.\n");
599     ucol_close(col);
600     free(test1);
601     free(test2);
602 
603 }
604 /*
605 ---------------------------------------------
606  tests decomposition setting
607 */
TestDecomposition()608 void TestDecomposition() {
609     UErrorCode status = U_ZERO_ERROR;
610     UCollator *en_US, *el_GR, *vi_VN;
611     en_US = ucol_open("en_US", &status);
612     el_GR = ucol_open("el_GR", &status);
613     vi_VN = ucol_open("vi_VN", &status);
614 
615     if (U_FAILURE(status)) {
616         log_err("ERROR: collation creation failed.: %s\n", myErrorName(status));
617         return;
618     }
619 
620     if (ucol_getAttribute(vi_VN, UCOL_NORMALIZATION_MODE, &status) != UCOL_ON ||
621         U_FAILURE(status))
622     {
623         log_err("ERROR: vi_VN collation did not have cannonical decomposition for normalization!\n");
624     }
625 
626     status = U_ZERO_ERROR;
627     if (ucol_getAttribute(el_GR, UCOL_NORMALIZATION_MODE, &status) != UCOL_ON ||
628         U_FAILURE(status))
629     {
630         log_err("ERROR: el_GR collation did not have cannonical decomposition for normalization!\n");
631     }
632 
633     status = U_ZERO_ERROR;
634     if (ucol_getAttribute(en_US, UCOL_NORMALIZATION_MODE, &status) != UCOL_OFF ||
635         U_FAILURE(status))
636     {
637         log_err("ERROR: en_US collation had cannonical decomposition for normalization!\n");
638     }
639 
640     ucol_close(en_US);
641     ucol_close(el_GR);
642     ucol_close(vi_VN);
643 }
644 
645 #define CLONETEST_COLLATOR_COUNT 4
646 
TestSafeClone()647 void TestSafeClone() {
648     UChar* test1;
649     UChar* test2;
650     static const UChar umlautUStr[] = {0x00DC, 0};
651     static const UChar oeStr[] = {0x0055, 0x0045, 0};
652     UCollator * someCollators [CLONETEST_COLLATOR_COUNT];
653     UCollator * someClonedCollators [CLONETEST_COLLATOR_COUNT];
654     UCollator * col;
655     UErrorCode err = U_ZERO_ERROR;
656     int8_t testSize = 6;    /* Leave this here to test buffer alingment in memory*/
657     uint8_t buffer [CLONETEST_COLLATOR_COUNT] [U_COL_SAFECLONE_BUFFERSIZE];
658     int32_t bufferSize = U_COL_SAFECLONE_BUFFERSIZE;
659     const char sampleRuleChars[] = "&Z < CH";
660     UChar sampleRule[sizeof(sampleRuleChars)];
661     int index;
662 
663     if (TestBufferSize()) {
664         log_err("U_COL_SAFECLONE_BUFFERSIZE should be larger than sizeof(UCollator)\n");
665         return;
666     }
667 
668     test1=(UChar*)malloc(sizeof(UChar) * testSize);
669     test2=(UChar*)malloc(sizeof(UChar) * testSize);
670     u_uastrcpy(test1, "abCda");
671     u_uastrcpy(test2, "abcda");
672     u_uastrcpy(sampleRule, sampleRuleChars);
673 
674     /* one default collator & two complex ones */
675     someCollators[0] = ucol_open("en_US", &err);
676     someCollators[1] = ucol_open("ko", &err);
677     someCollators[2] = ucol_open("ja_JP", &err);
678     someCollators[3] = ucol_openRules(sampleRule, -1, UCOL_ON, UCOL_TERTIARY, NULL, &err);
679     if(U_FAILURE(err)) {
680         log_data_err("Couldn't open one or more collators\n");
681         return;
682     }
683 
684     /* Check the various error & informational states: */
685 
686     /* Null status - just returns NULL */
687     if (0 != ucol_safeClone(someCollators[0], buffer[0], &bufferSize, 0))
688     {
689         log_err("FAIL: Cloned Collator failed to deal correctly with null status\n");
690     }
691     /* error status - should return 0 & keep error the same */
692     err = U_MEMORY_ALLOCATION_ERROR;
693     if (0 != ucol_safeClone(someCollators[0], buffer[0], &bufferSize, &err) || err != U_MEMORY_ALLOCATION_ERROR)
694     {
695         log_err("FAIL: Cloned Collator failed to deal correctly with incoming error status\n");
696     }
697     err = U_ZERO_ERROR;
698 
699     /* Null buffer size pointer - just returns NULL & set error to U_ILLEGAL_ARGUMENT_ERROR*/
700     if (0 != ucol_safeClone(someCollators[0], buffer[0], 0, &err) || err != U_ILLEGAL_ARGUMENT_ERROR)
701     {
702         log_err("FAIL: Cloned Collator failed to deal correctly with null bufferSize pointer\n");
703     }
704     err = U_ZERO_ERROR;
705 
706     /* buffer size pointer is 0 - fill in pbufferSize with a size */
707     bufferSize = 0;
708     if (0 != ucol_safeClone(someCollators[0], buffer[0], &bufferSize, &err) || U_FAILURE(err) || bufferSize <= 0)
709     {
710         log_err("FAIL: Cloned Collator failed a sizing request ('preflighting')\n");
711     }
712     /* Verify our define is large enough  */
713     if (U_COL_SAFECLONE_BUFFERSIZE < bufferSize)
714     {
715         log_err("FAIL: Pre-calculated buffer size is too small\n");
716     }
717     /* Verify we can use this run-time calculated size */
718     if (0 == (col = ucol_safeClone(someCollators[0], buffer[0], &bufferSize, &err)) || U_FAILURE(err))
719     {
720         log_err("FAIL: Collator can't be cloned with run-time size\n");
721     }
722     if (col) ucol_close(col);
723     /* size one byte too small - should allocate & let us know */
724     --bufferSize;
725     if (0 == (col = ucol_safeClone(someCollators[0], 0, &bufferSize, &err)) || err != U_SAFECLONE_ALLOCATED_WARNING)
726     {
727         log_err("FAIL: Cloned Collator failed to deal correctly with too-small buffer size\n");
728     }
729     if (col) ucol_close(col);
730     err = U_ZERO_ERROR;
731     bufferSize = U_COL_SAFECLONE_BUFFERSIZE;
732 
733 
734     /* Null buffer pointer - return Collator & set error to U_SAFECLONE_ALLOCATED_ERROR */
735     if (0 == (col = ucol_safeClone(someCollators[0], 0, &bufferSize, &err)) || err != U_SAFECLONE_ALLOCATED_WARNING)
736     {
737         log_err("FAIL: Cloned Collator failed to deal correctly with null buffer pointer\n");
738     }
739     if (col) ucol_close(col);
740     err = U_ZERO_ERROR;
741 
742     /* Null Collator - return NULL & set U_ILLEGAL_ARGUMENT_ERROR */
743     if (0 != ucol_safeClone(0, buffer[0], &bufferSize, &err) || err != U_ILLEGAL_ARGUMENT_ERROR)
744     {
745         log_err("FAIL: Cloned Collator failed to deal correctly with null Collator pointer\n");
746     }
747 
748     err = U_ZERO_ERROR;
749 
750     /* Test that a cloned collator doesn't accidentally use UCA. */
751     col=ucol_open("de@collation=phonebook", &err);
752     bufferSize = U_COL_SAFECLONE_BUFFERSIZE;
753     someClonedCollators[0] = ucol_safeClone(col, buffer[0], &bufferSize, &err);
754     doAssert( (ucol_greater(col, umlautUStr, u_strlen(umlautUStr), oeStr, u_strlen(oeStr))), "Original German phonebook collation sorts differently than expected");
755     doAssert( (ucol_greater(someClonedCollators[0], umlautUStr, u_strlen(umlautUStr), oeStr, u_strlen(oeStr))), "Cloned German phonebook collation sorts differently than expected");
756     if (!ucol_equals(someClonedCollators[0], col)) {
757         log_err("FAIL: Cloned German phonebook collator is not equal to original.\n");
758     }
759     ucol_close(col);
760     ucol_close(someClonedCollators[0]);
761 
762     err = U_ZERO_ERROR;
763 
764     /* change orig & clone & make sure they are independent */
765 
766     for (index = 0; index < CLONETEST_COLLATOR_COUNT; index++)
767     {
768         ucol_setStrength(someCollators[index], UCOL_IDENTICAL);
769         bufferSize = 1;
770         err = U_ZERO_ERROR;
771         ucol_close(ucol_safeClone(someCollators[index], buffer[index], &bufferSize, &err));
772         if (err != U_SAFECLONE_ALLOCATED_WARNING) {
773             log_err("FAIL: collator number %d was not allocated.\n", index);
774         }
775 
776         bufferSize = U_COL_SAFECLONE_BUFFERSIZE;
777         err = U_ZERO_ERROR;
778         someClonedCollators[index] = ucol_safeClone(someCollators[index], buffer[index], &bufferSize, &err);
779         if (someClonedCollators[index] == NULL
780             || someClonedCollators[index] < (UCollator *)buffer[index]
781             || someClonedCollators[index] > (UCollator *)(buffer[index]+(U_COL_SAFECLONE_BUFFERSIZE-1)))
782         {
783             log_err("FAIL: Cloned collator didn't use provided buffer.\n");
784             return;
785         }
786         if (!ucol_equals(someClonedCollators[index], someCollators[index])) {
787             log_err("FAIL: Cloned collator is not equal to original at index = %d.\n", index);
788         }
789 
790         /* Check the usability */
791         ucol_setStrength(someCollators[index], UCOL_PRIMARY);
792         ucol_setAttribute(someCollators[index], UCOL_CASE_LEVEL, UCOL_OFF, &err);
793 
794         doAssert( (ucol_equal(someCollators[index], test1, u_strlen(test1), test2, u_strlen(test2))), "Result should be \"abcda\" == \"abCda\"");
795 
796         /* Close the original to make sure that the clone is usable. */
797         ucol_close(someCollators[index]);
798 
799         ucol_setStrength(someClonedCollators[index], UCOL_TERTIARY);
800         ucol_setAttribute(someClonedCollators[index], UCOL_CASE_LEVEL, UCOL_OFF, &err);
801         doAssert( (ucol_greater(someClonedCollators[index], test1, u_strlen(test1), test2, u_strlen(test2))), "Result should be \"abCda\" >>> \"abcda\" ");
802 
803         ucol_close(someClonedCollators[index]);
804     }
805     free(test1);
806     free(test2);
807 }
808 
TestCloneBinary()809 void TestCloneBinary(){
810     UErrorCode err = U_ZERO_ERROR;
811     UCollator * col = ucol_open("en_US", &err);
812     UCollator * c;
813     int32_t size;
814     uint8_t * buffer;
815 
816     if (U_FAILURE(err)) {
817         log_data_err("Couldn't open collator. Error: %s\n", u_errorName(err));
818         return;
819     }
820 
821     size = ucol_cloneBinary(col, NULL, 0, &err);
822     if(size==0 || err!=U_BUFFER_OVERFLOW_ERROR) {
823         log_err("ucol_cloneBinary - couldn't check size. Error: %s\n", u_errorName(err));
824         return;
825     }
826     err = U_ZERO_ERROR;
827 
828     buffer = (uint8_t *) malloc(size);
829     ucol_cloneBinary(col, buffer, size, &err);
830     if(U_FAILURE(err)) {
831         log_err("ucol_cloneBinary - couldn't clone.. Error: %s\n", u_errorName(err));
832         free(buffer);
833         return;
834     }
835 
836     /* how to check binary result ? */
837 
838     c = ucol_openBinary(buffer, size, col, &err);
839     if(U_FAILURE(err)) {
840         log_err("ucol_openBinary failed. Error: %s\n", u_errorName(err));
841     } else {
842         UChar t[] = {0x41, 0x42, 0x43, 0};  /* ABC */
843         uint8_t  *k1, *k2;
844         int l1, l2;
845         l1 = ucol_getSortKey(col, t, -1, NULL,0);
846         l2 = ucol_getSortKey(c, t, -1, NULL,0);
847         k1 = (uint8_t *) malloc(sizeof(uint8_t) * l1);
848         k2 = (uint8_t *) malloc(sizeof(uint8_t) * l2);
849         ucol_getSortKey(col, t, -1, k1, l1);
850         ucol_getSortKey(col, t, -1, k2, l2);
851         if (strcmp((char *)k1,(char *)k2) != 0){
852             log_err("ucol_openBinary - new collator should equal to old one\n");
853         };
854         free(k1);
855         free(k2);
856     }
857     free(buffer);
858     ucol_close(c);
859     ucol_close(col);
860 }
861 /*
862 ----------------------------------------------------------------------------
863  ctor -- Tests the getSortKey
864 */
TestSortKey()865 void TestSortKey()
866 {
867     uint8_t *sortk1 = NULL, *sortk2 = NULL, *sortk3 = NULL, *sortkEmpty = NULL;
868     uint8_t sortk2_compat[] = {
869         /* 3.6 key, from UCA 5.0 */
870         0x29, 0x2b, 0x2d, 0x2f, 0x29, 0x01,
871         0x09, 0x01, 0x09, 0x01, 0x28, 0x01,
872         0x92, 0x93, 0x94, 0x95, 0x92, 0x00
873         /* 3.4 key, from UCA 4.1 */
874         /* 0x28, 0x2a, 0x2c, 0x2e, 0x28, 0x01, 0x09, 0x01, 0x09, 0x01, 0x27, 0x01, 0x92, 0x93, 0x94, 0x95, 0x92, 0x00 */
875         /* 2.6.1 key */
876         /* 0x26, 0x28, 0x2A, 0x2C, 0x26, 0x01, 0x09, 0x01, 0x09, 0x01, 0x25, 0x01, 0x92, 0x93, 0x94, 0x95, 0x92, 0x00 */
877         /* 2.2 key */
878         /*0x1D, 0x1F, 0x21, 0x23, 0x1D, 0x01, 0x09, 0x01, 0x09, 0x01, 0x1C, 0x01, 0x92, 0x93, 0x94, 0x95, 0x92, 0x00*/
879         /* 2.0 key */
880         /*0x19, 0x1B, 0x1D, 0x1F, 0x19, 0x01, 0x09, 0x01, 0x09, 0x01, 0x18, 0x01, 0x92, 0x93, 0x94, 0x95, 0x92, 0x00*/
881         /* 1.8.1 key.*/
882         /*0x19, 0x1B, 0x1D, 0x1F, 0x19, 0x01, 0x0A, 0x01, 0x0A, 0x01, 0x92, 0x93, 0x94, 0x95, 0x92, 0x00*/
883         /*this is a 1.8 sortkey */
884         /*0x17, 0x19, 0x1B, 0x1D, 0x17, 0x01, 0x08, 0x01, 0x08, 0x00*/
885         /*this is a 1.7 sortkey */
886         /*0x02, 0x54, 0x02, 0x55, 0x02, 0x56, 0x02, 0x57, 0x02, 0x54, 0x01, 0x02, 0x02, 0x02, 0x02, 0x02, 0x01, 0x02, 0x02, 0x02, 0x02, 0x02, 0x00*/
887         /* this is a 1.6 sortkey */
888       /*0x00, 0x53, 0x00, 0x54, 0x00, 0x55, 0x00, 0x56, 0x00, 0x53, 0x00, 0x00, 0x00, 0x01, 0x00, 0x01, 0x00, 0x01, 0x00, 0x01, 0x00, 0x01, 0x00, 0x00, 0x00, 0x01, 0x00, 0x01, 0x00, 0x01, 0x00, 0x01, 0x00, 0x01, 0x00, 0x00*/
889     };
890 
891     int32_t sortklen, osortklen;
892     uint32_t toStringLen=0;
893     UCollator *col;
894     UChar *test1, *test2, *test3;
895     UErrorCode status = U_ZERO_ERROR;
896     char toStringBuffer[256], *resultP;
897 
898 
899     uint8_t s1[] = { 0x9f, 0x00 };
900     uint8_t s2[] = { 0x61, 0x00 };
901     int  strcmpResult;
902 
903     strcmpResult = strcmp((const char *)s1, (const char *)s2);
904     log_verbose("strcmp(0x9f..., 0x61...) = %d\n", strcmpResult);
905 
906     if(strcmpResult <= 0) {
907       log_err("ERR: expected strcmp(\"9f 00\", \"61 00\") to be >=0 (GREATER).. got %d. Calling strcmp() for sortkeys may not work! \n",
908               strcmpResult);
909     }
910 
911 
912     log_verbose("testing SortKey begins...\n");
913     /* this is supposed to open default date format, but later on it treats it like it is "en_US"
914        - very bad if you try to run the tests on machine where default locale is NOT "en_US" */
915     /* col = ucol_open(NULL, &status); */
916     col = ucol_open("en_US", &status);
917     if (U_FAILURE(status)) {
918         log_err("ERROR: Default collation creation failed.: %s\n", myErrorName(status));
919         return;
920     }
921 
922 
923     if(ucol_getStrength(col) != UCOL_DEFAULT_STRENGTH)
924     {
925         log_err("ERROR: default collation did not have UCOL_DEFAULT_STRENGTH !\n");
926     }
927     /* Need to use identical strength */
928     ucol_setAttribute(col, UCOL_STRENGTH, UCOL_IDENTICAL, &status);
929 
930     test1=(UChar*)malloc(sizeof(UChar) * 6);
931     test2=(UChar*)malloc(sizeof(UChar) * 6);
932     test3=(UChar*)malloc(sizeof(UChar) * 6);
933 
934     memset(test1,0xFE, sizeof(UChar)*6);
935     memset(test2,0xFE, sizeof(UChar)*6);
936     memset(test3,0xFE, sizeof(UChar)*6);
937 
938 
939     u_uastrcpy(test1, "Abcda");
940     u_uastrcpy(test2, "abcda");
941     u_uastrcpy(test3, "abcda");
942 
943     log_verbose("Use tertiary comparison level testing ....\n");
944 
945     sortklen=ucol_getSortKey(col, test1, u_strlen(test1),  NULL, 0);
946     sortk1=(uint8_t*)malloc(sizeof(uint8_t) * (sortklen+1));
947     memset(sortk1,0xFE, sortklen);
948     ucol_getSortKey(col, test1, u_strlen(test1), sortk1, sortklen+1);
949 
950     sortklen=ucol_getSortKey(col, test2, u_strlen(test2),  NULL, 0);
951     sortk2=(uint8_t*)malloc(sizeof(uint8_t) * (sortklen+1));
952     memset(sortk2,0xFE, sortklen);
953     ucol_getSortKey(col, test2, u_strlen(test2), sortk2, sortklen+1);
954 
955     osortklen = sortklen;
956     sortklen=ucol_getSortKey(col, test2, u_strlen(test3),  NULL, 0);
957     sortk3=(uint8_t*)malloc(sizeof(uint8_t) * (sortklen+1));
958     memset(sortk3,0xFE, sortklen);
959     ucol_getSortKey(col, test2, u_strlen(test2), sortk3, sortklen+1);
960 
961     doAssert( (sortklen == osortklen), "Sortkey length should be the same (abcda, abcda)");
962 
963     doAssert( (memcmp(sortk1, sortk2, sortklen) > 0), "Result should be \"Abcda\" > \"abcda\"");
964     doAssert( (memcmp(sortk2, sortk1, sortklen) < 0), "Result should be \"abcda\" < \"Abcda\"");
965     doAssert( (memcmp(sortk2, sortk3, sortklen) == 0), "Result should be \"abcda\" ==  \"abcda\"");
966 
967     doAssert( (memcmp(sortk2, sortk2_compat, sortklen) == 0), "Binary format for 'abcda' sortkey different!");
968 
969     resultP = ucol_sortKeyToString(col, sortk2_compat, toStringBuffer, &toStringLen);
970     doAssert( (resultP != 0), "sortKeyToString failed!");
971 
972 #if 1 /* verobse log of sortkeys */
973     {
974       char junk2[1000];
975       char junk3[1000];
976       int i;
977 
978       strcpy(junk2, "abcda[2] ");
979       strcpy(junk3, " abcda[3] ");
980 
981       for(i=0;i<sortklen;i++)
982         {
983           sprintf(junk2+strlen(junk2), "%02X ",(int)( 0xFF & sortk2[i]));
984           sprintf(junk3+strlen(junk3), "%02X ",(int)( 0xFF & sortk3[i]));
985         }
986 
987       log_verbose("%s\n", junk2);
988       log_verbose("%s\n", junk3);
989     }
990 #endif
991 
992     free(sortk1);
993     free(sortk2);
994     free(sortk3);
995 
996     log_verbose("Use secondary comparision level testing ...\n");
997     ucol_setStrength(col, UCOL_SECONDARY);
998     sortklen=ucol_getSortKey(col, test1, u_strlen(test1),  NULL, 0);
999     sortk1=(uint8_t*)malloc(sizeof(uint8_t) * (sortklen+1));
1000     ucol_getSortKey(col, test1, u_strlen(test1), sortk1, sortklen+1);
1001     sortklen=ucol_getSortKey(col, test2, u_strlen(test2),  NULL, 0);
1002     sortk2=(uint8_t*)malloc(sizeof(uint8_t) * (sortklen+1));
1003     ucol_getSortKey(col, test2, u_strlen(test2), sortk2, sortklen+1);
1004 
1005     doAssert( !(memcmp(sortk1, sortk2, sortklen) > 0), "Result should be \"Abcda\" == \"abcda\"");
1006     doAssert( !(memcmp(sortk2, sortk1, sortklen) < 0), "Result should be \"abcda\" == \"Abcda\"");
1007     doAssert( (memcmp(sortk1, sortk2, sortklen) == 0), "Result should be \"abcda\" ==  \"abcda\"");
1008 
1009     log_verbose("getting sortkey for an empty string\n");
1010     ucol_setAttribute(col, UCOL_STRENGTH, UCOL_TERTIARY, &status);
1011     sortklen = ucol_getSortKey(col, test1, 0, NULL, 0);
1012     sortkEmpty = (uint8_t*)malloc(sizeof(uint8_t) * sortklen+1);
1013     sortklen = ucol_getSortKey(col, test1, 0, sortkEmpty, sortklen+1);
1014     if(sortklen != 3 || sortkEmpty[0] != 1 || sortkEmpty[0] != 1 || sortkEmpty[2] != 0) {
1015       log_err("Empty string generated wrong sortkey!\n");
1016     }
1017     free(sortkEmpty);
1018 
1019     log_verbose("testing passing invalid string\n");
1020     sortklen = ucol_getSortKey(col, NULL, 0, NULL, 0);
1021     if(sortklen != 0) {
1022       log_err("Invalid string didn't return sortkey size of 0\n");
1023     }
1024 
1025 
1026     log_verbose("testing sortkey ends...\n");
1027     ucol_close(col);
1028     free(test1);
1029     free(test2);
1030     free(test3);
1031     free(sortk1);
1032     free(sortk2);
1033 
1034 }
TestHashCode()1035 void TestHashCode()
1036 {
1037     uint8_t *sortk1, *sortk2, *sortk3;
1038     int32_t sortk1len, sortk2len, sortk3len;
1039     UCollator *col;
1040     UChar *test1, *test2, *test3;
1041     UErrorCode status = U_ZERO_ERROR;
1042     log_verbose("testing getHashCode begins...\n");
1043     col = ucol_open("en_US", &status);
1044     if (U_FAILURE(status)) {
1045         log_err("ERROR: Default collation creation failed.: %s\n", myErrorName(status));
1046         return;
1047     }
1048     test1=(UChar*)malloc(sizeof(UChar) * 6);
1049     test2=(UChar*)malloc(sizeof(UChar) * 6);
1050     test3=(UChar*)malloc(sizeof(UChar) * 6);
1051     u_uastrcpy(test1, "Abcda");
1052     u_uastrcpy(test2, "abcda");
1053     u_uastrcpy(test3, "abcda");
1054 
1055     log_verbose("Use tertiary comparison level testing ....\n");
1056     sortk1len=ucol_getSortKey(col, test1, u_strlen(test1),  NULL, 0);
1057     sortk1=(uint8_t*)malloc(sizeof(uint8_t) * (sortk1len+1));
1058     ucol_getSortKey(col, test1, u_strlen(test1), sortk1, sortk1len+1);
1059     sortk2len=ucol_getSortKey(col, test2, u_strlen(test2),  NULL, 0);
1060     sortk2=(uint8_t*)malloc(sizeof(uint8_t) * (sortk2len+1));
1061     ucol_getSortKey(col, test2, u_strlen(test2), sortk2, sortk2len+1);
1062     sortk3len=ucol_getSortKey(col, test2, u_strlen(test3),  NULL, 0);
1063     sortk3=(uint8_t*)malloc(sizeof(uint8_t) * (sortk3len+1));
1064     ucol_getSortKey(col, test2, u_strlen(test2), sortk3, sortk3len+1);
1065 
1066 
1067     log_verbose("ucol_hashCode() testing ...\n");
1068 
1069     doAssert( ucol_keyHashCode(sortk1, sortk1len) != ucol_keyHashCode(sortk2, sortk2len), "Hash test1 result incorrect" );
1070     doAssert( !(ucol_keyHashCode(sortk1, sortk1len) == ucol_keyHashCode(sortk2, sortk2len)), "Hash test2 result incorrect" );
1071     doAssert( ucol_keyHashCode(sortk2, sortk2len) == ucol_keyHashCode(sortk3, sortk3len), "Hash result not equal" );
1072 
1073     log_verbose("hashCode tests end.\n");
1074     ucol_close(col);
1075     free(sortk1);
1076     free(sortk2);
1077     free(sortk3);
1078     free(test1);
1079     free(test2);
1080     free(test3);
1081 
1082 
1083 }
1084 /*
1085  *----------------------------------------------------------------------------
1086  * Tests the UCollatorElements API.
1087  *
1088  */
TestElemIter()1089 void TestElemIter()
1090 {
1091     int32_t offset;
1092     int32_t order1, order2, order3;
1093     UChar *testString1, *testString2;
1094     UCollator *col;
1095     UCollationElements *iterator1, *iterator2, *iterator3;
1096     UErrorCode status = U_ZERO_ERROR;
1097     log_verbose("testing UCollatorElements begins...\n");
1098     col = ucol_open("en_US", &status);
1099     ucol_setAttribute(col, UCOL_NORMALIZATION_MODE, UCOL_OFF, &status);
1100     if (U_FAILURE(status)) {
1101         log_err("ERROR: Default collation creation failed.: %s\n", myErrorName(status));
1102         return;
1103     }
1104 
1105     testString1=(UChar*)malloc(sizeof(UChar) * 150);
1106     testString2=(UChar*)malloc(sizeof(UChar) * 150);
1107     u_uastrcpy(testString1, "XFILE What subset of all possible test cases has the highest probability of detecting the most errors?");
1108     u_uastrcpy(testString2, "Xf_ile What subset of all possible test cases has the lowest probability of detecting the least errors?");
1109 
1110     log_verbose("Constructors and comparison testing....\n");
1111 
1112     iterator1 = ucol_openElements(col, testString1, u_strlen(testString1), &status);
1113     if(U_FAILURE(status)) {
1114         log_err("ERROR: Default collationElement iterator creation failed.: %s\n", myErrorName(status));
1115         ucol_close(col);
1116         return;
1117     }
1118     else{ log_verbose("PASS: Default collationElement iterator1 creation passed\n");}
1119 
1120     iterator2 = ucol_openElements(col, testString1, u_strlen(testString1), &status);
1121     if(U_FAILURE(status)) {
1122         log_err("ERROR: Default collationElement iterator creation failed.: %s\n", myErrorName(status));
1123         ucol_close(col);
1124         return;
1125     }
1126     else{ log_verbose("PASS: Default collationElement iterator2 creation passed\n");}
1127 
1128     iterator3 = ucol_openElements(col, testString2, u_strlen(testString2), &status);
1129     if(U_FAILURE(status)) {
1130         log_err("ERROR: Default collationElement iterator creation failed.: %s\n", myErrorName(status));
1131         ucol_close(col);
1132         return;
1133     }
1134     else{ log_verbose("PASS: Default collationElement iterator3 creation passed\n");}
1135 
1136     offset=ucol_getOffset(iterator1);
1137     ucol_setOffset(iterator1, 6, &status);
1138     if (U_FAILURE(status)) {
1139         log_err("Error in setOffset for UCollatorElements iterator.: %s\n", myErrorName(status));
1140         return;
1141     }
1142     if(ucol_getOffset(iterator1)==6)
1143         log_verbose("setOffset and getOffset working fine\n");
1144     else{
1145         log_err("error in set and get Offset got %d instead of 6\n", ucol_getOffset(iterator1));
1146     }
1147 
1148     ucol_setOffset(iterator1, 0, &status);
1149     order1 = ucol_next(iterator1, &status);
1150     if (U_FAILURE(status)) {
1151         log_err("Somehow ran out of memory stepping through the iterator1.: %s\n", myErrorName(status));
1152         return;
1153     }
1154     order2=ucol_getOffset(iterator2);
1155     doAssert((order1 != order2), "The first iterator advance failed");
1156     order2 = ucol_next(iterator2, &status);
1157     if (U_FAILURE(status)) {
1158         log_err("Somehow ran out of memory stepping through the iterator2.: %s\n", myErrorName(status));
1159         return;
1160     }
1161     order3 = ucol_next(iterator3, &status);
1162     if (U_FAILURE(status)) {
1163         log_err("Somehow ran out of memory stepping through the iterator3.: %s\n", myErrorName(status));
1164         return;
1165     }
1166 
1167     doAssert((order1 == order2), "The second iterator advance failed should be the same as first one");
1168 
1169 doAssert( (ucol_primaryOrder(order1) == ucol_primaryOrder(order3)), "The primary orders should be identical");
1170 doAssert( (ucol_secondaryOrder(order1) == ucol_secondaryOrder(order3)), "The secondary orders should be identical");
1171 doAssert( (ucol_tertiaryOrder(order1) == ucol_tertiaryOrder(order3)), "The tertiary orders should be identical");
1172 
1173     order1=ucol_next(iterator1, &status);
1174     if (U_FAILURE(status)) {
1175         log_err("Somehow ran out of memory stepping through the iterator2.: %s\n", myErrorName(status));
1176         return;
1177     }
1178     order3=ucol_next(iterator3, &status);
1179     if (U_FAILURE(status)) {
1180         log_err("Somehow ran out of memory stepping through the iterator2.: %s\n", myErrorName(status));
1181         return;
1182     }
1183 doAssert( (ucol_primaryOrder(order1) == ucol_primaryOrder(order3)), "The primary orders should be identical");
1184 doAssert( (ucol_tertiaryOrder(order1) != ucol_tertiaryOrder(order3)), "The tertiary orders should be different");
1185 
1186     order1=ucol_next(iterator1, &status);
1187     if (U_FAILURE(status)) {
1188         log_err("Somehow ran out of memory stepping through the iterator2.: %s\n", myErrorName(status));
1189         return;
1190     }
1191     order3=ucol_next(iterator3, &status);
1192     if (U_FAILURE(status)) {
1193         log_err("Somehow ran out of memory stepping through the iterator2.: %s\n", myErrorName(status));
1194         return;
1195     }
1196     /* this here, my friends, is either pure lunacy or something so obsolete that even it's mother
1197      * doesn't care about it. Essentialy, this test complains if secondary values for 'I' and '_'
1198      * are the same. According to the UCA, this is not true. Therefore, remove the test.
1199      * Besides, if primary strengths for two code points are different, it doesn't matter one bit
1200      * what is the relation between secondary or any other strengths.
1201      * killed by weiv 06/11/2002.
1202      */
1203     /*
1204     doAssert( ((order1 & UCOL_SECONDARYMASK) != (order3 & UCOL_SECONDARYMASK)), "The secondary orders should be different");
1205     */
1206     doAssert( (order1 != UCOL_NULLORDER), "Unexpected end of iterator reached");
1207 
1208     free(testString1);
1209     free(testString2);
1210     ucol_closeElements(iterator1);
1211     ucol_closeElements(iterator2);
1212     ucol_closeElements(iterator3);
1213     ucol_close(col);
1214 
1215     log_verbose("testing CollationElementIterator ends...\n");
1216 }
1217 
TestGetLocale()1218 void TestGetLocale() {
1219   UErrorCode status = U_ZERO_ERROR;
1220   const char *rules = "&a<x<y<z";
1221   UChar rlz[256] = {0};
1222   uint32_t rlzLen = u_unescape(rules, rlz, 256);
1223 
1224   UCollator *coll = NULL;
1225   const char *locale = NULL;
1226 
1227   int32_t i = 0;
1228 
1229   /* Now that the collation tree is separate, actual==valid at all times. [alan] */
1230   static const struct {
1231     const char* requestedLocale;
1232     const char* validLocale;
1233     const char* actualLocale;
1234   } testStruct[] = {
1235     { "sr_RS", "sr_Cyrl_RS", "ru" },
1236     { "sh_YU", "sr_Latn_RS", "hr" }, /* this used to be sh, but now sh collation aliases hr */
1237     { "en_BE_FOO", "en_BE", "en_BE" },
1238     { "fr_FR_NONEXISTANT", "fr_FR", "fr" }
1239   };
1240 
1241   /* test opening collators for different locales */
1242   for(i = 0; i<sizeof(testStruct)/sizeof(testStruct[0]); i++) {
1243     status = U_ZERO_ERROR;
1244     coll = ucol_open(testStruct[i].requestedLocale, &status);
1245     if(U_FAILURE(status)) {
1246       log_err("Failed to open collator for %s with %s\n", testStruct[i].requestedLocale, u_errorName(status));
1247       ucol_close(coll);
1248       continue;
1249     }
1250    locale = ucol_getLocale(coll, ULOC_REQUESTED_LOCALE, &status);
1251     if(strcmp(locale, testStruct[i].requestedLocale) != 0) {
1252       log_err("[Coll %s]: Error in requested locale, expected %s, got %s\n", testStruct[i].requestedLocale, testStruct[i].requestedLocale, locale);
1253     }
1254     locale = ucol_getLocale(coll, ULOC_VALID_LOCALE, &status);
1255     if(strcmp(locale, testStruct[i].validLocale) != 0) {
1256       log_err("[Coll %s]: Error in valid locale, expected %s, got %s\n", testStruct[i].requestedLocale, testStruct[i].validLocale, locale);
1257     }
1258     locale = ucol_getLocale(coll, ULOC_ACTUAL_LOCALE, &status);
1259     if(strcmp(locale, testStruct[i].actualLocale) != 0) {
1260       log_err("[Coll %s]: Error in actual locale, expected %s, got %s\n", testStruct[i].requestedLocale, testStruct[i].actualLocale, locale);
1261     }
1262     ucol_close(coll);
1263   }
1264 
1265   /* completely non-existant locale for collator should get a default collator */
1266   {
1267     UCollator *defaultColl = ucol_open(NULL, &status);
1268     coll = ucol_open("blahaha", &status);
1269     if(U_SUCCESS(status)) {
1270       if(strcmp(ucol_getLocale(coll, ULOC_REQUESTED_LOCALE, &status), "blahaha")) {
1271         log_err("Nonexisting locale didn't preserve the requested locale\n");
1272       }
1273       if(strcmp(ucol_getLocale(coll, ULOC_VALID_LOCALE, &status),
1274         ucol_getLocale(defaultColl, ULOC_VALID_LOCALE, &status))) {
1275         log_err("Valid locale for nonexisting locale locale collator differs "
1276           "from valid locale for default collator\n");
1277       }
1278       if(strcmp(ucol_getLocale(coll, ULOC_ACTUAL_LOCALE, &status),
1279         ucol_getLocale(defaultColl, ULOC_ACTUAL_LOCALE, &status))) {
1280         log_err("Actual locale for nonexisting locale locale collator differs "
1281           "from actual locale for default collator\n");
1282       }
1283       ucol_close(coll);
1284       ucol_close(defaultColl);
1285     } else {
1286       log_data_err("Couldn't open collators\n");
1287     }
1288   }
1289 
1290 
1291 
1292   /* collator instantiated from rules should have all three locales NULL */
1293   coll = ucol_openRules(rlz, rlzLen, UCOL_DEFAULT, UCOL_DEFAULT, NULL, &status);
1294   locale = ucol_getLocale(coll, ULOC_REQUESTED_LOCALE, &status);
1295   if(locale != NULL) {
1296     log_err("For collator instantiated from rules, requested locale returned %s instead of NULL\n", locale);
1297   }
1298   locale = ucol_getLocale(coll, ULOC_VALID_LOCALE, &status);
1299   if(locale != NULL) {
1300     log_err("For collator instantiated from rules,  valid locale returned %s instead of NULL\n", locale);
1301   }
1302   locale = ucol_getLocale(coll, ULOC_ACTUAL_LOCALE, &status);
1303   if(locale != NULL) {
1304     log_err("For collator instantiated from rules, actual locale returned %s instead of NULL\n", locale);
1305   }
1306   ucol_close(coll);
1307 
1308 }
1309 
1310 
TestGetAll()1311 void TestGetAll()
1312 {
1313     int32_t i, count;
1314     count=ucol_countAvailable();
1315     /* use something sensible w/o hardcoding the count */
1316     if(count < 0){
1317         log_err("Error in countAvailable(), it returned %d\n", count);
1318     }
1319     else{
1320         log_verbose("PASS: countAvailable() successful, it returned %d\n", count);
1321     }
1322     for(i=0;i<count;i++)
1323         log_verbose("%s\n", ucol_getAvailable(i));
1324 
1325 
1326 }
1327 
1328 
1329 struct teststruct {
1330     const char *original;
1331     uint8_t key[256];
1332 } ;
1333 
compare_teststruct(const void * string1,const void * string2)1334 static int compare_teststruct(const void *string1, const void *string2) {
1335     return(strcmp((const char *)((struct teststruct *)string1)->key, (const char *)((struct teststruct *)string2)->key));
1336 }
1337 
TestBounds()1338 void TestBounds() {
1339     UErrorCode status = U_ZERO_ERROR;
1340 
1341     UCollator *coll = ucol_open("sh", &status);
1342 
1343     uint8_t sortkey[512], lower[512], upper[512];
1344     UChar buffer[512];
1345 
1346     static const char * const test[] = {
1347         "John Smith",
1348         "JOHN SMITH",
1349         "john SMITH",
1350         "j\\u00F6hn sm\\u00EFth",
1351         "J\\u00F6hn Sm\\u00EFth",
1352         "J\\u00D6HN SM\\u00CFTH",
1353         "john smithsonian",
1354         "John Smithsonian",
1355     };
1356 
1357     struct teststruct tests[] = {
1358         {"\\u010CAKI MIHALJ" } ,
1359         {"\\u010CAKI MIHALJ" } ,
1360         {"\\u010CAKI PIRO\\u0160KA" },
1361         {"\\u010CABAI ANDRIJA" } ,
1362         {"\\u010CABAI LAJO\\u0160" } ,
1363         {"\\u010CABAI MARIJA" } ,
1364         {"\\u010CABAI STEVAN" } ,
1365         {"\\u010CABAI STEVAN" } ,
1366         {"\\u010CABARKAPA BRANKO" } ,
1367         {"\\u010CABARKAPA MILENKO" } ,
1368         {"\\u010CABARKAPA MIROSLAV" } ,
1369         {"\\u010CABARKAPA SIMO" } ,
1370         {"\\u010CABARKAPA STANKO" } ,
1371         {"\\u010CABARKAPA TAMARA" } ,
1372         {"\\u010CABARKAPA TOMA\\u0160" } ,
1373         {"\\u010CABDARI\\u0106 NIKOLA" } ,
1374         {"\\u010CABDARI\\u0106 ZORICA" } ,
1375         {"\\u010CABI NANDOR" } ,
1376         {"\\u010CABOVI\\u0106 MILAN" } ,
1377         {"\\u010CABRADI AGNEZIJA" } ,
1378         {"\\u010CABRADI IVAN" } ,
1379         {"\\u010CABRADI JELENA" } ,
1380         {"\\u010CABRADI LJUBICA" } ,
1381         {"\\u010CABRADI STEVAN" } ,
1382         {"\\u010CABRDA MARTIN" } ,
1383         {"\\u010CABRILO BOGDAN" } ,
1384         {"\\u010CABRILO BRANISLAV" } ,
1385         {"\\u010CABRILO LAZAR" } ,
1386         {"\\u010CABRILO LJUBICA" } ,
1387         {"\\u010CABRILO SPASOJA" } ,
1388         {"\\u010CADE\\u0160 ZDENKA" } ,
1389         {"\\u010CADESKI BLAGOJE" } ,
1390         {"\\u010CADOVSKI VLADIMIR" } ,
1391         {"\\u010CAGLJEVI\\u0106 TOMA" } ,
1392         {"\\u010CAGOROVI\\u0106 VLADIMIR" } ,
1393         {"\\u010CAJA VANKA" } ,
1394         {"\\u010CAJI\\u0106 BOGOLJUB" } ,
1395         {"\\u010CAJI\\u0106 BORISLAV" } ,
1396         {"\\u010CAJI\\u0106 RADOSLAV" } ,
1397         {"\\u010CAK\\u0160IRAN MILADIN" } ,
1398         {"\\u010CAKAN EUGEN" } ,
1399         {"\\u010CAKAN EVGENIJE" } ,
1400         {"\\u010CAKAN IVAN" } ,
1401         {"\\u010CAKAN JULIJAN" } ,
1402         {"\\u010CAKAN MIHAJLO" } ,
1403         {"\\u010CAKAN STEVAN" } ,
1404         {"\\u010CAKAN VLADIMIR" } ,
1405         {"\\u010CAKAN VLADIMIR" } ,
1406         {"\\u010CAKAN VLADIMIR" } ,
1407         {"\\u010CAKARA ANA" } ,
1408         {"\\u010CAKAREVI\\u0106 MOMIR" } ,
1409         {"\\u010CAKAREVI\\u0106 NEDELJKO" } ,
1410         {"\\u010CAKI \\u0160ANDOR" } ,
1411         {"\\u010CAKI AMALIJA" } ,
1412         {"\\u010CAKI ANDRA\\u0160" } ,
1413         {"\\u010CAKI LADISLAV" } ,
1414         {"\\u010CAKI LAJO\\u0160" } ,
1415         {"\\u010CAKI LASLO" } ,
1416     };
1417 
1418 
1419 
1420     int32_t i = 0, j = 0, k = 0, buffSize = 0, skSize = 0, lowerSize = 0, upperSize = 0;
1421     int32_t arraySize = sizeof(tests)/sizeof(tests[0]);
1422 
1423     if(U_SUCCESS(status) && coll) {
1424         for(i = 0; i<arraySize; i++) {
1425             buffSize = u_unescape(tests[i].original, buffer, 512);
1426             skSize = ucol_getSortKey(coll, buffer, buffSize, tests[i].key, 512);
1427         }
1428 
1429         qsort(tests, arraySize, sizeof(struct teststruct), compare_teststruct);
1430 
1431         for(i = 0; i < arraySize-1; i++) {
1432             for(j = i+1; j < arraySize; j++) {
1433                 lowerSize = ucol_getBound(tests[i].key, -1, UCOL_BOUND_LOWER, 1, lower, 512, &status);
1434                 upperSize = ucol_getBound(tests[j].key, -1, UCOL_BOUND_UPPER, 1, upper, 512, &status);
1435                 for(k = i; k <= j; k++) {
1436                     if(strcmp((const char *)lower, (const char *)tests[k].key) > 0) {
1437                         log_err("Problem with lower! j = %i (%s vs %s)\n", k, tests[k].original, tests[i].original);
1438                     }
1439                     if(strcmp((const char *)upper, (const char *)tests[k].key) <= 0) {
1440                         log_err("Problem with upper! j = %i (%s vs %s)\n", k, tests[k].original, tests[j].original);
1441                     }
1442                 }
1443             }
1444         }
1445 
1446 
1447 #if 0
1448         for(i = 0; i < 1000; i++) {
1449             lowerRND = (rand()/(RAND_MAX/arraySize));
1450             upperRND = lowerRND + (rand()/(RAND_MAX/(arraySize-lowerRND)));
1451 
1452             lowerSize = ucol_getBound(tests[lowerRND].key, -1, UCOL_BOUND_LOWER, 1, lower, 512, &status);
1453             upperSize = ucol_getBound(tests[upperRND].key, -1, UCOL_BOUND_UPPER_LONG, 1, upper, 512, &status);
1454 
1455             for(j = lowerRND; j<=upperRND; j++) {
1456                 if(strcmp(lower, tests[j].key) > 0) {
1457                     log_err("Problem with lower! j = %i (%s vs %s)\n", j, tests[j].original, tests[lowerRND].original);
1458                 }
1459                 if(strcmp(upper, tests[j].key) <= 0) {
1460                     log_err("Problem with upper! j = %i (%s vs %s)\n", j, tests[j].original, tests[upperRND].original);
1461                 }
1462             }
1463         }
1464 #endif
1465 
1466 
1467 
1468 
1469 
1470         for(i = 0; i<sizeof(test)/sizeof(test[0]); i++) {
1471             buffSize = u_unescape(test[i], buffer, 512);
1472             skSize = ucol_getSortKey(coll, buffer, buffSize, sortkey, 512);
1473             lowerSize = ucol_getBound(sortkey, skSize, UCOL_BOUND_LOWER, 1, lower, 512, &status);
1474             upperSize = ucol_getBound(sortkey, skSize, UCOL_BOUND_UPPER_LONG, 1, upper, 512, &status);
1475             for(j = i+1; j<sizeof(test)/sizeof(test[0]); j++) {
1476                 buffSize = u_unescape(test[j], buffer, 512);
1477                 skSize = ucol_getSortKey(coll, buffer, buffSize, sortkey, 512);
1478                 if(strcmp((const char *)lower, (const char *)sortkey) > 0) {
1479                     log_err("Problem with lower! i = %i, j = %i (%s vs %s)\n", i, j, test[i], test[j]);
1480                 }
1481                 if(strcmp((const char *)upper, (const char *)sortkey) <= 0) {
1482                     log_err("Problem with upper! i = %i, j = %i (%s vs %s)\n", i, j, test[i], test[j]);
1483                 }
1484             }
1485         }
1486         ucol_close(coll);
1487     } else {
1488         log_data_err("Couldn't open collator\n");
1489     }
1490 
1491 }
1492 
doOverrunTest(UCollator * coll,const UChar * uString,int32_t strLen)1493 static void doOverrunTest(UCollator *coll, const UChar *uString, int32_t strLen) {
1494     int32_t skLen = 0, skLen2 = 0;
1495     uint8_t sortKey[256];
1496     int32_t i, j;
1497     uint8_t filler = 0xFF;
1498 
1499     skLen = ucol_getSortKey(coll, uString, strLen, NULL, 0);
1500 
1501     for(i = 0; i < skLen; i++) {
1502         memset(sortKey, filler, 256);
1503         skLen2 = ucol_getSortKey(coll, uString, strLen, sortKey, i);
1504         if(skLen != skLen2) {
1505             log_err("For buffer size %i, got different sortkey length. Expected %i got %i\n", i, skLen, skLen2);
1506         }
1507         for(j = i; j < 256; j++) {
1508             if(sortKey[j] != filler) {
1509                 log_err("Something run over index %i\n", j);
1510                 break;
1511             }
1512         }
1513     }
1514 }
1515 
1516 /* j1865 reports that if a shorter buffer is passed to
1517 * to get sort key, a buffer overrun happens in some
1518 * cases. This test tries to check this.
1519 */
TestSortKeyBufferOverrun(void)1520 void TestSortKeyBufferOverrun(void) {
1521     UErrorCode status = U_ZERO_ERROR;
1522     const char* cString = "A very Merry liTTle-lamB..";
1523     UChar uString[256];
1524     int32_t strLen = 0;
1525     UCollator *coll = ucol_open("root", &status);
1526     strLen = u_unescape(cString, uString, 256);
1527 
1528     if(U_SUCCESS(status)) {
1529         log_verbose("testing non ignorable\n");
1530         ucol_setAttribute(coll, UCOL_ALTERNATE_HANDLING, UCOL_NON_IGNORABLE, &status);
1531         doOverrunTest(coll, uString, strLen);
1532 
1533         log_verbose("testing shifted\n");
1534         ucol_setAttribute(coll, UCOL_ALTERNATE_HANDLING, UCOL_SHIFTED, &status);
1535         doOverrunTest(coll, uString, strLen);
1536 
1537         log_verbose("testing shifted quaternary\n");
1538         ucol_setAttribute(coll, UCOL_STRENGTH, UCOL_QUATERNARY, &status);
1539         doOverrunTest(coll, uString, strLen);
1540 
1541         log_verbose("testing with french secondaries\n");
1542         ucol_setAttribute(coll, UCOL_FRENCH_COLLATION, UCOL_ON, &status);
1543         ucol_setAttribute(coll, UCOL_STRENGTH, UCOL_TERTIARY, &status);
1544         ucol_setAttribute(coll, UCOL_ALTERNATE_HANDLING, UCOL_NON_IGNORABLE, &status);
1545         doOverrunTest(coll, uString, strLen);
1546 
1547     }
1548     ucol_close(coll);
1549 }
1550 
TestAttribute()1551 static void TestAttribute()
1552 {
1553     UErrorCode error = U_ZERO_ERROR;
1554     UCollator *coll = ucol_open(NULL, &error);
1555 
1556     if (U_FAILURE(error)) {
1557         log_err("Creation of default collator failed");
1558         return;
1559     }
1560 
1561     ucol_setAttribute(coll, UCOL_FRENCH_COLLATION, UCOL_OFF, &error);
1562     if (ucol_getAttribute(coll, UCOL_FRENCH_COLLATION, &error) != UCOL_OFF ||
1563         U_FAILURE(error)) {
1564         log_err("Setting and retrieving of the french collation failed");
1565     }
1566 
1567     ucol_setAttribute(coll, UCOL_FRENCH_COLLATION, UCOL_ON, &error);
1568     if (ucol_getAttribute(coll, UCOL_FRENCH_COLLATION, &error) != UCOL_ON ||
1569         U_FAILURE(error)) {
1570         log_err("Setting and retrieving of the french collation failed");
1571     }
1572 
1573     ucol_setAttribute(coll, UCOL_ALTERNATE_HANDLING, UCOL_SHIFTED, &error);
1574     if (ucol_getAttribute(coll, UCOL_ALTERNATE_HANDLING, &error) != UCOL_SHIFTED ||
1575         U_FAILURE(error)) {
1576         log_err("Setting and retrieving of the alternate handling failed");
1577     }
1578 
1579     ucol_setAttribute(coll, UCOL_ALTERNATE_HANDLING, UCOL_NON_IGNORABLE, &error);
1580     if (ucol_getAttribute(coll, UCOL_ALTERNATE_HANDLING, &error) != UCOL_NON_IGNORABLE ||
1581         U_FAILURE(error)) {
1582         log_err("Setting and retrieving of the alternate handling failed");
1583     }
1584 
1585     ucol_setAttribute(coll, UCOL_CASE_FIRST, UCOL_LOWER_FIRST, &error);
1586     if (ucol_getAttribute(coll, UCOL_CASE_FIRST, &error) != UCOL_LOWER_FIRST ||
1587         U_FAILURE(error)) {
1588         log_err("Setting and retrieving of the case first attribute failed");
1589     }
1590 
1591     ucol_setAttribute(coll, UCOL_CASE_FIRST, UCOL_UPPER_FIRST, &error);
1592     if (ucol_getAttribute(coll, UCOL_CASE_FIRST, &error) != UCOL_UPPER_FIRST ||
1593         U_FAILURE(error)) {
1594         log_err("Setting and retrieving of the case first attribute failed");
1595     }
1596 
1597     ucol_setAttribute(coll, UCOL_CASE_LEVEL, UCOL_ON, &error);
1598     if (ucol_getAttribute(coll, UCOL_CASE_LEVEL, &error) != UCOL_ON ||
1599         U_FAILURE(error)) {
1600         log_err("Setting and retrieving of the case level attribute failed");
1601     }
1602 
1603     ucol_setAttribute(coll, UCOL_CASE_LEVEL, UCOL_OFF, &error);
1604     if (ucol_getAttribute(coll, UCOL_CASE_LEVEL, &error) != UCOL_OFF ||
1605         U_FAILURE(error)) {
1606         log_err("Setting and retrieving of the case level attribute failed");
1607     }
1608 
1609     ucol_setAttribute(coll, UCOL_NORMALIZATION_MODE, UCOL_ON, &error);
1610     if (ucol_getAttribute(coll, UCOL_NORMALIZATION_MODE, &error) != UCOL_ON ||
1611         U_FAILURE(error)) {
1612         log_err("Setting and retrieving of the normalization on/off attribute failed");
1613     }
1614 
1615     ucol_setAttribute(coll, UCOL_NORMALIZATION_MODE, UCOL_OFF, &error);
1616     if (ucol_getAttribute(coll, UCOL_NORMALIZATION_MODE, &error) != UCOL_OFF ||
1617         U_FAILURE(error)) {
1618         log_err("Setting and retrieving of the normalization on/off attribute failed");
1619     }
1620 
1621     ucol_setAttribute(coll, UCOL_STRENGTH, UCOL_PRIMARY, &error);
1622     if (ucol_getAttribute(coll, UCOL_STRENGTH, &error) != UCOL_PRIMARY ||
1623         U_FAILURE(error)) {
1624         log_err("Setting and retrieving of the collation strength failed");
1625     }
1626 
1627     ucol_setAttribute(coll, UCOL_STRENGTH, UCOL_SECONDARY, &error);
1628     if (ucol_getAttribute(coll, UCOL_STRENGTH, &error) != UCOL_SECONDARY ||
1629         U_FAILURE(error)) {
1630         log_err("Setting and retrieving of the collation strength failed");
1631     }
1632 
1633     ucol_setAttribute(coll, UCOL_STRENGTH, UCOL_TERTIARY, &error);
1634     if (ucol_getAttribute(coll, UCOL_STRENGTH, &error) != UCOL_TERTIARY ||
1635         U_FAILURE(error)) {
1636         log_err("Setting and retrieving of the collation strength failed");
1637     }
1638 
1639     ucol_setAttribute(coll, UCOL_STRENGTH, UCOL_QUATERNARY, &error);
1640     if (ucol_getAttribute(coll, UCOL_STRENGTH, &error) != UCOL_QUATERNARY ||
1641         U_FAILURE(error)) {
1642         log_err("Setting and retrieving of the collation strength failed");
1643     }
1644 
1645     ucol_setAttribute(coll, UCOL_STRENGTH, UCOL_IDENTICAL, &error);
1646     if (ucol_getAttribute(coll, UCOL_STRENGTH, &error) != UCOL_IDENTICAL ||
1647         U_FAILURE(error)) {
1648         log_err("Setting and retrieving of the collation strength failed");
1649     }
1650 
1651     ucol_close(coll);
1652 }
1653 
TestGetTailoredSet()1654 void TestGetTailoredSet() {
1655   struct {
1656     const char *rules;
1657     const char *tests[20];
1658     int32_t testsize;
1659   } setTest[] = {
1660     { "&a < \\u212b", { "\\u212b", "A\\u030a", "\\u00c5" }, 3},
1661     { "& S < \\u0161 <<< \\u0160", { "\\u0161", "s\\u030C", "\\u0160", "S\\u030C" }, 4}
1662   };
1663 
1664   int32_t i = 0, j = 0;
1665   UErrorCode status = U_ZERO_ERROR;
1666   UParseError pError;
1667 
1668   UCollator *coll = NULL;
1669   UChar buff[1024];
1670   int32_t buffLen = 0;
1671   USet *set = NULL;
1672 
1673   for(i = 0; i < sizeof(setTest)/sizeof(setTest[0]); i++) {
1674     buffLen = u_unescape(setTest[i].rules, buff, 1024);
1675     coll = ucol_openRules(buff, buffLen, UCOL_DEFAULT, UCOL_DEFAULT, &pError, &status);
1676     if(U_SUCCESS(status)) {
1677       set = ucol_getTailoredSet(coll, &status);
1678       if(uset_size(set) != setTest[i].testsize) {
1679         log_err("Tailored set size different (%d) than expected (%d)\n", uset_size(set), setTest[i].testsize);
1680       }
1681       for(j = 0; j < setTest[i].testsize; j++) {
1682         buffLen = u_unescape(setTest[i].tests[j], buff, 1024);
1683         if(!uset_containsString(set, buff, buffLen)) {
1684           log_err("Tailored set doesn't contain %s... It should\n", setTest[i].tests[j]);
1685         }
1686       }
1687       uset_close(set);
1688     } else {
1689       log_err("Couldn't open collator with rules %s\n", setTest[i].rules);
1690     }
1691     ucol_close(coll);
1692   }
1693 }
1694 
tMemCmp(const uint8_t * first,const uint8_t * second)1695 static int tMemCmp(const uint8_t *first, const uint8_t *second) {
1696    int32_t firstLen = (int32_t)strlen((const char *)first);
1697    int32_t secondLen = (int32_t)strlen((const char *)second);
1698    return memcmp(first, second, uprv_min(firstLen, secondLen));
1699 }
1700 static const char * strengthsC[] = {
1701      "UCOL_PRIMARY",
1702      "UCOL_SECONDARY",
1703      "UCOL_TERTIARY",
1704      "UCOL_QUATERNARY",
1705      "UCOL_IDENTICAL"
1706 };
1707 
TestMergeSortKeys(void)1708 void TestMergeSortKeys(void) {
1709    UErrorCode status = U_ZERO_ERROR;
1710    UCollator *coll = ucol_open("en", &status);
1711    if(U_SUCCESS(status)) {
1712 
1713      const char* cases[] = {
1714        "abc",
1715          "abcd",
1716          "abcde"
1717      };
1718      uint32_t casesSize = sizeof(cases)/sizeof(cases[0]);
1719      const char* prefix = "foo";
1720      const char* suffix = "egg";
1721      char outBuff1[256], outBuff2[256];
1722 
1723      uint8_t **sortkeys = (uint8_t **)malloc(casesSize*sizeof(uint8_t *));
1724      uint8_t **mergedPrefixkeys = (uint8_t **)malloc(casesSize*sizeof(uint8_t *));
1725      uint8_t **mergedSuffixkeys = (uint8_t **)malloc(casesSize*sizeof(uint8_t *));
1726      uint32_t *sortKeysLen = (uint32_t *)malloc(casesSize*sizeof(uint32_t));
1727      uint8_t prefixKey[256], suffixKey[256];
1728      uint32_t prefixKeyLen = 0, suffixKeyLen = 0, i = 0;
1729      UChar buffer[256];
1730      uint32_t unescapedLen = 0, l1 = 0, l2 = 0;
1731      UColAttributeValue strength;
1732 
1733      log_verbose("ucol_mergeSortkeys test\n");
1734      log_verbose("Testing order of the test cases\n");
1735      genericLocaleStarter("en", cases, casesSize);
1736 
1737      for(i = 0; i<casesSize; i++) {
1738        sortkeys[i] = (uint8_t *)malloc(256*sizeof(uint8_t));
1739        mergedPrefixkeys[i] = (uint8_t *)malloc(256*sizeof(uint8_t));
1740        mergedSuffixkeys[i] = (uint8_t *)malloc(256*sizeof(uint8_t));
1741      }
1742 
1743      unescapedLen = u_unescape(prefix, buffer, 256);
1744      prefixKeyLen = ucol_getSortKey(coll, buffer, unescapedLen, prefixKey, 256);
1745 
1746      unescapedLen = u_unescape(suffix, buffer, 256);
1747      suffixKeyLen = ucol_getSortKey(coll, buffer, unescapedLen, suffixKey, 256);
1748 
1749      log_verbose("Massaging data with prefixes and different strengths\n");
1750      strength = UCOL_PRIMARY;
1751      while(strength <= UCOL_IDENTICAL) {
1752        log_verbose("Strength %s\n", strengthsC[strength<=UCOL_QUATERNARY?strength:4]);
1753        ucol_setAttribute(coll, UCOL_STRENGTH, strength, &status);
1754        for(i = 0; i<casesSize; i++) {
1755          unescapedLen = u_unescape(cases[i], buffer, 256);
1756          sortKeysLen[i] = ucol_getSortKey(coll, buffer, unescapedLen, sortkeys[i], 256);
1757          ucol_mergeSortkeys(prefixKey, prefixKeyLen, sortkeys[i], sortKeysLen[i], mergedPrefixkeys[i], 256);
1758          ucol_mergeSortkeys(sortkeys[i], sortKeysLen[i], suffixKey, suffixKeyLen, mergedSuffixkeys[i], 256);
1759          if(i>0) {
1760            if(tMemCmp(mergedPrefixkeys[i-1], mergedPrefixkeys[i]) >= 0) {
1761              log_err("Error while comparing prefixed keys @ strength %s:\n", strengthsC[strength<=UCOL_QUATERNARY?strength:4]);
1762              log_err("%s\n%s\n",
1763                          ucol_sortKeyToString(coll, mergedPrefixkeys[i-1], outBuff1, &l1),
1764                          ucol_sortKeyToString(coll, mergedPrefixkeys[i], outBuff2, &l2));
1765            }
1766            if(tMemCmp(mergedSuffixkeys[i-1], mergedSuffixkeys[i]) >= 0) {
1767              log_err("Error while comparing suffixed keys @ strength %s:\n", strengthsC[strength<=UCOL_QUATERNARY?strength:4]);
1768              log_err("%s\n%s\n",
1769                          ucol_sortKeyToString(coll, mergedSuffixkeys[i-1], outBuff1, &l1),
1770                          ucol_sortKeyToString(coll, mergedSuffixkeys[i], outBuff2, &l2));
1771            }
1772          }
1773        }
1774        if(strength == UCOL_QUATERNARY) {
1775          strength = UCOL_IDENTICAL;
1776        } else {
1777          strength++;
1778        }
1779      }
1780 
1781      {
1782        uint8_t smallBuf[3];
1783        uint32_t reqLen = 0;
1784        log_verbose("testing buffer overflow\n");
1785        reqLen = ucol_mergeSortkeys(prefixKey, prefixKeyLen, suffixKey, suffixKeyLen, smallBuf, 3);
1786        if(reqLen != (prefixKeyLen+suffixKeyLen-1)) {
1787          log_err("Wrong preflight size for merged sortkey\n");
1788        }
1789      }
1790 
1791      {
1792        UChar empty = 0;
1793        uint8_t emptyKey[20], abcKey[50], mergedKey[100];
1794        int32_t emptyKeyLen = 0, abcKeyLen = 0, mergedKeyLen = 0;
1795 
1796        log_verbose("testing merging with sortkeys generated for empty strings\n");
1797        emptyKeyLen = ucol_getSortKey(coll, &empty, 0, emptyKey, 20);
1798        unescapedLen = u_unescape(cases[0], buffer, 256);
1799        abcKeyLen = ucol_getSortKey(coll, buffer, unescapedLen, abcKey, 50);
1800        mergedKeyLen = ucol_mergeSortkeys(emptyKey, emptyKeyLen, abcKey, abcKeyLen, mergedKey, 100);
1801        if(mergedKey[0] != 2) {
1802          log_err("Empty sortkey didn't produce a level separator\n");
1803        }
1804        /* try with zeros */
1805        mergedKeyLen = ucol_mergeSortkeys(emptyKey, 0, abcKey, abcKeyLen, mergedKey, 100);
1806        if(mergedKeyLen != 0 || mergedKey[0] != 0) {
1807          log_err("Empty key didn't produce null mergedKey\n");
1808        }
1809        mergedKeyLen = ucol_mergeSortkeys(abcKey, abcKeyLen, emptyKey, 0, mergedKey, 100);
1810        if(mergedKeyLen != 0 || mergedKey[0] != 0) {
1811          log_err("Empty key didn't produce null mergedKey\n");
1812        }
1813 
1814      }
1815 
1816      for(i = 0; i<casesSize; i++) {
1817        free(sortkeys[i]);
1818        free(mergedPrefixkeys[i]);
1819        free(mergedSuffixkeys[i]);
1820      }
1821      free(sortkeys);
1822      free(mergedPrefixkeys);
1823      free(mergedSuffixkeys);
1824      free(sortKeysLen);
1825      ucol_close(coll);
1826      /* need to finish this up */
1827    } else {
1828      log_data_err("Couldn't open collator");
1829    }
1830 }
TestShortString(void)1831 static void TestShortString(void)
1832 {
1833     struct {
1834         const char *input;
1835         const char *expectedOutput;
1836         const char *locale;
1837         UErrorCode expectedStatus;
1838         int32_t    expectedOffset;
1839         uint32_t   expectedIdentifier;
1840     } testCases[] = {
1841         {"LDE_RDE_KPHONEBOOK_T0041_ZLATN","B2900_KPHONEBOOK_LDE", "de@collation=phonebook", U_USING_FALLBACK_WARNING, 0, 0 },
1842         {"LEN_RUS_NO_AS_S4","AS_LEN_NO_S4", NULL, U_USING_FALLBACK_WARNING, 0, 0 },
1843         {"LDE_VPHONEBOOK_EO_SI","EO_KPHONEBOOK_LDE_SI", "de@collation=phonebook", U_ZERO_ERROR, 0, 0 },
1844         {"LDE_Kphonebook","KPHONEBOOK_LDE", "de@collation=phonebook", U_ZERO_ERROR, 0, 0 },
1845         {"Xqde_DE@collation=phonebookq_S3_EX","KPHONEBOOK_LDE", "de@collation=phonebook", U_USING_FALLBACK_WARNING, 0, 0 },
1846         {"LFR_FO", "LFR", NULL, U_ZERO_ERROR, 0, 0 },
1847         {"SO_LX_AS", "", NULL, U_ILLEGAL_ARGUMENT_ERROR, 8, 0 },
1848         {"S3_ASS_MMM", "", NULL, U_ILLEGAL_ARGUMENT_ERROR, 5, 0 }
1849     };
1850 
1851     int32_t i = 0, j = 0;
1852     UCollator *coll = NULL, *fromID = NULL, *fromNormalized = NULL;
1853     UParseError parseError;
1854     UErrorCode status = U_ZERO_ERROR;
1855     char fromShortBuffer[256], fromIDBuffer[256], fromIDRoundtrip[256], normalizedBuffer[256], fromNormalizedBuffer[256];
1856     uint32_t identifier = 0, idFromSS = 0;
1857     const char* locale = NULL;
1858 
1859 
1860     for(i = 0; i < sizeof(testCases)/sizeof(testCases[0]); i++) {
1861         status = U_ZERO_ERROR;
1862         if(testCases[i].locale) {
1863             locale = testCases[i].locale;
1864         } else {
1865             locale = NULL;
1866         }
1867 
1868         coll = ucol_openFromShortString(testCases[i].input, FALSE, &parseError, &status);
1869         if(status != testCases[i].expectedStatus) {
1870             log_err("Got status '%s' that is different from expected '%s' for '%s'\n",
1871                 u_errorName(status), u_errorName(testCases[i].expectedStatus), testCases[i].input);
1872         }
1873 
1874         if(U_SUCCESS(status)) {
1875             ucol_getShortDefinitionString(coll, locale, fromShortBuffer, 256, &status);
1876 
1877             if(strcmp(fromShortBuffer, testCases[i].expectedOutput)) {
1878                 log_err("Got short string '%s' from the collator. Expected '%s' for input '%s'\n",
1879                     fromShortBuffer, testCases[i].expectedOutput, testCases[i].input);
1880             }
1881 
1882             ucol_normalizeShortDefinitionString(testCases[i].input, normalizedBuffer, 256, &parseError, &status);
1883             fromNormalized = ucol_openFromShortString(normalizedBuffer, FALSE, &parseError, &status);
1884             ucol_getShortDefinitionString(fromNormalized, locale, fromNormalizedBuffer, 256, &status);
1885 
1886             if(strcmp(fromShortBuffer, fromNormalizedBuffer)) {
1887                 log_err("Strings obtained from collators instantiated by short string ('%s') and from normalized string ('%s') differ\n",
1888                     fromShortBuffer, fromNormalizedBuffer);
1889             }
1890 
1891 
1892             if(!ucol_equals(coll, fromNormalized)) {
1893                 log_err("Collator from short string ('%s') differs from one obtained through a normalized version ('%s')\n",
1894                     testCases[i].input, normalizedBuffer);
1895             }
1896 
1897             /* test identifiers */
1898             identifier = ucol_collatorToIdentifier(coll, locale, &status);
1899             if(identifier < UCOL_SIT_COLLATOR_NOT_ENCODABLE) {
1900                 ucol_identifierToShortString(identifier, fromIDBuffer, 256, FALSE, &status);
1901                 fromID = ucol_openFromIdentifier(identifier, FALSE, &status);
1902                 if(!ucol_equals(coll, fromID)) {
1903                     log_err("Collator from short string ('%s') differs from one obtained through an identifier ('%s')\n",
1904                         testCases[i].input, fromIDBuffer);
1905                 }
1906                 ucol_close(fromID);
1907             }
1908 
1909             /* round-trip short string - identifier */
1910             for(j = 1; j < 2; j++) {
1911                 idFromSS = ucol_shortStringToIdentifier(testCases[i].input, (UBool)j, &status);
1912                 ucol_identifierToShortString(idFromSS, fromIDBuffer, 256, (UBool)j, &status);
1913                 identifier = ucol_shortStringToIdentifier(fromIDBuffer, (UBool)j, &status);
1914                 ucol_identifierToShortString(identifier, fromIDRoundtrip, 256, (UBool)j, &status);
1915 
1916                 if(idFromSS != identifier) {
1917                     log_err("FD = %i, id didn't round trip. %08X vs %08X (%s)\n",
1918                         j, idFromSS, identifier, testCases[i].input);
1919                 }
1920                 if(strcmp(fromIDBuffer, fromIDRoundtrip)) {
1921                     log_err("FD = %i, SS didn't round trip. %s vs %s (%s)\n",
1922                         j, fromIDBuffer, fromIDRoundtrip, testCases[i].input);
1923                 }
1924             }
1925 
1926             ucol_close(fromNormalized);
1927             ucol_close(coll);
1928 
1929         } else {
1930             if(parseError.offset != testCases[i].expectedOffset) {
1931                 log_err("Got parse error offset %i, but expected %i instead for '%s'\n",
1932                     parseError.offset, testCases[i].expectedOffset, testCases[i].input);
1933             }
1934         }
1935     }
1936 
1937 }
1938 
1939 static void
doSetsTest(const char * locale,const USet * ref,USet * set,const char * inSet,const char * outSet,UErrorCode * status)1940 doSetsTest(const char *locale, const USet *ref, USet *set, const char* inSet, const char* outSet, UErrorCode *status) {
1941     UChar buffer[512];
1942     int32_t bufLen;
1943 
1944     uset_clear(set);
1945     bufLen = u_unescape(inSet, buffer, 512);
1946     uset_applyPattern(set, buffer, bufLen, 0, status);
1947     if(U_FAILURE(*status)) {
1948         log_err("%s: Failure setting pattern %s\n", locale, u_errorName(*status));
1949     }
1950 
1951     if(!uset_containsAll(ref, set)) {
1952         log_err("%s: Some stuff from %s is not present in the set\n", locale, inSet);
1953     }
1954 
1955     uset_clear(set);
1956     bufLen = u_unescape(outSet, buffer, 512);
1957     uset_applyPattern(set, buffer, bufLen, 0, status);
1958     if(U_FAILURE(*status)) {
1959         log_err("%s: Failure setting pattern %s\n", locale, u_errorName(*status));
1960     }
1961 
1962     if(!uset_containsNone(ref, set)) {
1963         log_err("%s: Some stuff from %s is present in the set\n", locale, outSet);
1964     }
1965 }
1966 
1967 
1968 
1969 
1970 static void
TestGetContractionsAndUnsafes(void)1971 TestGetContractionsAndUnsafes(void)
1972 {
1973     static struct {
1974         const char* locale;
1975         const char* inConts;
1976         const char* outConts;
1977         const char* inExp;
1978         const char* outExp;
1979         const char* unsafeCodeUnits;
1980         const char* safeCodeUnits;
1981     } tests[] = {
1982         { "ru",
1983             "[{\\u0474\\u030F}{\\u0475\\u030F}{\\u04D8\\u0308}{\\u04D9\\u0308}{\\u04E8\\u0308}{\\u04E9\\u0308}]",
1984             "[{\\u0430\\u0306}{\\u0410\\u0306}{\\u0430\\u0308}{\\u0410\\u0306}{\\u0433\\u0301}{\\u0413\\u0301}]",
1985             "[\\u00e6]",
1986             "[a]",
1987             "[\\u0474\\u0475\\u04d8\\u04d9\\u04e8\\u04e9]",
1988             "[aAbB\\u0430\\u0410\\u0433\\u0413]"
1989         },
1990         { "uk",
1991             "[{\\u0474\\u030F}{\\u0475\\u030F}{\\u04D8\\u0308}{\\u04D9\\u0308}{\\u04E8\\u0308}{\\u04E9\\u0308}"
1992             "{\\u0430\\u0306}{\\u0410\\u0306}{\\u0430\\u0308}{\\u0410\\u0306}{\\u0433\\u0301}{\\u0413\\u0301}]",
1993             "[]",
1994             "[\\u00e6]",
1995             "[a]",
1996             "[\\u0474\\u0475\\u04D8\\u04D9\\u04E8\\u04E9\\u0430\\u0410\\u0433\\u0413]",
1997             "[aAbBxv]",
1998         },
1999         { "sh",
2000             "[{C\\u0301}{C\\u030C}{C\\u0341}{DZ\\u030C}{Dz\\u030C}{D\\u017D}{D\\u017E}{lj}{nj}]",
2001             "[{\\u309d\\u3099}{\\u30fd\\u3099}]",
2002             "[\\u00e6]",
2003             "[a]",
2004             "[nlcdzNLCDZ]",
2005             "[jabv]"
2006         },
2007         { "ja",
2008           "[{\\u3053\\u3099\\u309D}{\\u3053\\u3099\\u309D\\u3099}{\\u3053\\u3099\\u309E}{\\u3053\\u3099\\u30FC}{\\u3053\\u309D}{\\u3053\\u309D\\u3099}{\\u3053\\u309E}{\\u3053\\u30FC}{\\u30B3\\u3099\\u30FC}{\\u30B3\\u3099\\u30FD}{\\u30B3\\u3099\\u30FD\\u3099}{\\u30B3\\u3099\\u30FE}{\\u30B3\\u30FC}{\\u30B3\\u30FD}{\\u30B3\\u30FD\\u3099}{\\u30B3\\u30FE}]",
2009           "[{\\u30FD\\u3099}{\\u309D\\u3099}{\\u3053\\u3099}{\\u30B3\\u3099}{lj}{nj}]",
2010             "[\\u30FE\\u00e6]",
2011             "[a]",
2012             "[\\u3099]",
2013             "[]"
2014         }
2015     };
2016 
2017 
2018 
2019 
2020     UErrorCode status = U_ZERO_ERROR;
2021     UCollator *coll = NULL;
2022     int32_t i = 0;
2023     int32_t noConts = 0;
2024     USet *conts = uset_open(0,0);
2025     USet *exp = uset_open(0, 0);
2026     USet *set  = uset_open(0,0);
2027     int32_t setBufferLen = 65536;
2028     UChar buffer[65536];
2029     int32_t setLen = 0;
2030 
2031     for(i = 0; i < sizeof(tests)/sizeof(tests[0]); i++) {
2032         log_verbose("Testing locale: %s\n", tests[i].locale);
2033         coll = ucol_open(tests[i].locale, &status);
2034         ucol_getContractionsAndExpansions(coll, conts, exp, TRUE, &status);
2035         doSetsTest(tests[i].locale, conts, set, tests[i].inConts, tests[i].outConts, &status);
2036         setLen = uset_toPattern(conts, buffer, setBufferLen, TRUE, &status);
2037         if(U_SUCCESS(status)) {
2038             /*log_verbose("Contractions %i: %s\n", uset_getItemCount(conts), aescstrdup(buffer, setLen));*/
2039         } else {
2040             log_err("error %s. %i\n", u_errorName(status), setLen);
2041             status = U_ZERO_ERROR;
2042         }
2043         doSetsTest(tests[i].locale, exp, set, tests[i].inExp, tests[i].outExp, &status);
2044         setLen = uset_toPattern(exp, buffer, setBufferLen, TRUE, &status);
2045         if(U_SUCCESS(status)) {
2046             /*log_verbose("Expansions %i: %s\n", uset_getItemCount(exp), aescstrdup(buffer, setLen));*/
2047         } else {
2048             log_err("error %s. %i\n", u_errorName(status), setLen);
2049             status = U_ZERO_ERROR;
2050         }
2051 
2052         noConts = ucol_getUnsafeSet(coll, conts, &status);
2053         doSetsTest(tests[i].locale, conts, set, tests[i].unsafeCodeUnits, tests[i].safeCodeUnits, &status);
2054         setLen = uset_toPattern(conts, buffer, setBufferLen, TRUE, &status);
2055         if(U_SUCCESS(status)) {
2056             log_verbose("Unsafe %i: %s\n", uset_getItemCount(exp), aescstrdup(buffer, setLen));
2057         } else {
2058             log_err("error %s. %i\n", u_errorName(status), setLen);
2059             status = U_ZERO_ERROR;
2060         }
2061 
2062         ucol_close(coll);
2063     }
2064 
2065 
2066     uset_close(conts);
2067     uset_close(exp);
2068     uset_close(set);
2069 }
2070 
2071 static void
TestOpenBinary(void)2072 TestOpenBinary(void)
2073 {
2074     UErrorCode status = U_ZERO_ERROR;
2075     /*
2076     char rule[] = "&h < d < c < b";
2077     char *wUCA[] = { "a", "h", "d", "c", "b", "i" };
2078     char *noUCA[] = {"d", "c", "b", "a", "h", "i" };
2079     */
2080     /* we have to use Cyrillic letters because latin-1 always gets copied */
2081     const char rule[] = "&\\u0452 < \\u0434 < \\u0433 < \\u0432"; /* &dje < d < g < v */
2082     const char *wUCA[] = { "\\u0430", "\\u0452", "\\u0434", "\\u0433", "\\u0432", "\\u0435" }; /* a, dje, d, g, v, e */
2083     const char *noUCA[] = {"\\u0434", "\\u0433", "\\u0432", "\\u0430", "\\u0435", "\\u0452" }; /* d, g, v, a, e, dje */
2084 
2085     UChar uRules[256];
2086     int32_t uRulesLen = u_unescape(rule, uRules, 256);
2087 
2088     UCollator *coll = ucol_openRules(uRules, uRulesLen, UCOL_DEFAULT, UCOL_DEFAULT, NULL, &status);
2089     UCollator *UCA = ucol_open("root", &status);
2090     UCollator *cloneNOUCA = NULL, *cloneWUCA = NULL;
2091 
2092     uint8_t imageBuffer[32768];
2093     uint8_t *image = imageBuffer;
2094     int32_t imageBufferCapacity = 32768;
2095 
2096     int32_t imageSize;
2097 
2098     if((coll==NULL)||(UCA==NULL)||(U_FAILURE(status))) {
2099         log_data_err("could not load collators or error occured: %s\n",
2100             u_errorName(status));
2101         return;
2102     }
2103     imageSize = ucol_cloneBinary(coll, image, imageBufferCapacity, &status);
2104     if(U_FAILURE(status)) {
2105         image = (uint8_t *)malloc(imageSize*sizeof(uint8_t));
2106         status = U_ZERO_ERROR;
2107         imageSize = ucol_cloneBinary(coll, imageBuffer, imageSize, &status);
2108     }
2109 
2110 
2111     cloneWUCA = ucol_openBinary(image, imageSize, UCA, &status);
2112     cloneNOUCA = ucol_openBinary(image, imageSize, NULL, &status);
2113 
2114     genericOrderingTest(coll, wUCA, sizeof(wUCA)/sizeof(wUCA[0]));
2115 
2116     genericOrderingTest(cloneWUCA, wUCA, sizeof(wUCA)/sizeof(wUCA[0]));
2117     genericOrderingTest(cloneNOUCA, noUCA, sizeof(noUCA)/sizeof(noUCA[0]));
2118 
2119     if(image != imageBuffer) {
2120         free(image);
2121     }
2122     ucol_close(coll);
2123     ucol_close(cloneNOUCA);
2124     ucol_close(cloneWUCA);
2125     ucol_close(UCA);
2126 }
2127 
TestDefault(void)2128 static void TestDefault(void) {
2129     /* Tests for code coverage. */
2130     UErrorCode status = U_ZERO_ERROR;
2131     UCollator *coll = ucol_open("es@collation=pinyin", &status);
2132     if (status != U_USING_DEFAULT_WARNING) {
2133         /* What do you mean that you know about using pinyin collation in Spanish!? This should be in the zh locale. */
2134         log_err("es@collation=pinyin should return U_USING_DEFAULT_WARNING, but returned %s\n", u_errorName(status));
2135     }
2136     ucol_close(coll);
2137     if (ucol_getKeywordValues("funky", &status) != NULL) {
2138         log_err("Collators should not know about the funky keyword.\n");
2139     }
2140     if (status != U_ILLEGAL_ARGUMENT_ERROR) {
2141         log_err("funky keyword didn't fail as expected %s\n", u_errorName(status));
2142     }
2143     if (ucol_getKeywordValues("collation", &status) != NULL) {
2144         log_err("ucol_getKeywordValues should not work when given a bad status.\n");
2145     }
2146 }
2147 
2148 #endif /* #if !UCONFIG_NO_COLLATION */
2149