1 /********************************************************************
3 * Copyright (c) 1997-2014, International Business Machines Corporation and
4 * others. All Rights Reserved.
5 ********************************************************************/
6 /*******************************************************************************
10 * Modification History:
12 * Madhu Katragadda Ported for C API
13 ********************************************************************************
17 * Important: This file is included into intltest/allcoll.cpp so that the
18 * test data is shared. This makes it easier to maintain the test data,
19 * especially since the Unicode data must be portable and quoted character
20 * literals will not work.
21 * If it is included, then there will be a #define INCLUDE_CALLCOLL_C
22 * that must prevent the actual code in here from being part of the
23 * allcoll.cpp compilation.
27 * CollationDummyTest is a third level test class. This tests creation of
28 * a customized collator object. For example, number 1 to be sorted
29 * equlivalent to word 'one'.
35 #include "unicode/utypes.h"
37 #if !UCONFIG_NO_COLLATION
39 #include "unicode/ucol.h"
40 #include "unicode/uloc.h"
41 #include "unicode/ures.h"
42 #include "unicode/udata.h"
43 #include "unicode/ucoleitr.h"
44 #include "unicode/ustring.h"
45 #include "unicode/uclean.h"
46 #include "unicode/putil.h"
47 #include "unicode/uenum.h"
56 /* set to 1 to test offsets in backAndForth() */
57 #define TEST_OFFSETS 0
59 /* perform test with strength PRIMARY */
60 static void TestPrimary(void);
62 /* perform test with strength SECONDARY */
63 static void TestSecondary(void);
65 /* perform test with strength tertiary */
66 static void TestTertiary(void);
68 /*perform tests with strength Identical */
69 static void TestIdentical(void);
71 /* perform extra tests */
72 static void TestExtra(void);
74 /* Test jitterbug 581 */
75 static void TestJB581(void);
77 /* Test jitterbug 1401 */
78 static void TestJB1401(void);
80 /* Test [variable top] in the rule syntax */
81 static void TestVariableTop(void);
84 static void TestSurrogates(void);
86 static void TestInvalidRules(void);
88 static void TestJitterbug1098(void);
90 static void TestFCDCrash(void);
92 static void TestJ5298(void);
94 const UCollationResult results
[] = {
96 UCOL_LESS
, /*UCOL_GREATER,*/
112 /* test primary > 17 */
122 /* test secondary > 26 */
138 void uprv_appendByteToHexString(char *dst
, uint8_t val
) {
139 uint32_t len
= (uint32_t)uprv_strlen(dst
);
140 *(dst
+len
) = T_CString_itosOffset((val
>> 4));
141 *(dst
+len
+1) = T_CString_itosOffset((val
& 0xF));
145 /* this function makes a string with representation of a sortkey */
146 static char* U_EXPORT2
sortKeyToString(const UCollator
*coll
, const uint8_t *sortkey
, char *buffer
, uint32_t *len
) {
147 int32_t strength
= UCOL_PRIMARY
;
148 uint32_t res_size
= 0;
149 UBool doneCase
= FALSE
;
150 UErrorCode errorCode
= U_ZERO_ERROR
;
152 char *current
= buffer
;
153 const uint8_t *currentSk
= sortkey
;
155 uprv_strcpy(current
, "[");
157 while(strength
<= UCOL_QUATERNARY
&& strength
<= ucol_getStrength(coll
)) {
158 if(strength
> UCOL_PRIMARY
) {
159 uprv_strcat(current
, " . ");
161 while(*currentSk
!= 0x01 && *currentSk
!= 0x00) { /* print a level */
162 uprv_appendByteToHexString(current
, *currentSk
++);
163 uprv_strcat(current
, " ");
165 if(ucol_getAttribute(coll
, UCOL_CASE_LEVEL
, &errorCode
) == UCOL_ON
&& strength
== UCOL_SECONDARY
&& doneCase
== FALSE
) {
167 } else if(ucol_getAttribute(coll
, UCOL_CASE_LEVEL
, &errorCode
) == UCOL_OFF
|| doneCase
== TRUE
|| strength
!= UCOL_SECONDARY
) {
171 uprv_appendByteToHexString(current
, *currentSk
++); /* This should print '01' */
173 if(strength
== UCOL_QUATERNARY
&& ucol_getAttribute(coll
, UCOL_ALTERNATE_HANDLING
, &errorCode
) == UCOL_NON_IGNORABLE
) {
178 if(ucol_getStrength(coll
) == UCOL_IDENTICAL
) {
179 uprv_strcat(current
, " . ");
180 while(*currentSk
!= 0) {
181 uprv_appendByteToHexString(current
, *currentSk
++);
182 uprv_strcat(current
, " ");
185 uprv_appendByteToHexString(current
, *currentSk
++);
187 uprv_strcat(current
, "]");
189 if(res_size
> *len
) {
196 void addAllCollTest(TestNode
** root
)
198 addTest(root
, &TestPrimary
, "tscoll/callcoll/TestPrimary");
199 addTest(root
, &TestSecondary
, "tscoll/callcoll/TestSecondary");
200 addTest(root
, &TestTertiary
, "tscoll/callcoll/TestTertiary");
201 addTest(root
, &TestIdentical
, "tscoll/callcoll/TestIdentical");
202 addTest(root
, &TestExtra
, "tscoll/callcoll/TestExtra");
203 addTest(root
, &TestJB581
, "tscoll/callcoll/TestJB581");
204 addTest(root
, &TestVariableTop
, "tscoll/callcoll/TestVariableTop");
205 addTest(root
, &TestSurrogates
, "tscoll/callcoll/TestSurrogates");
206 addTest(root
, &TestInvalidRules
, "tscoll/callcoll/TestInvalidRules");
207 addTest(root
, &TestJB1401
, "tscoll/callcoll/TestJB1401");
208 addTest(root
, &TestJitterbug1098
, "tscoll/callcoll/TestJitterbug1098");
209 addTest(root
, &TestFCDCrash
, "tscoll/callcoll/TestFCDCrash");
210 addTest(root
, &TestJ5298
, "tscoll/callcoll/TestJ5298");
213 UBool
hasCollationElements(const char *locName
) {
215 UErrorCode status
= U_ZERO_ERROR
;
217 UResourceBundle
*loc
= ures_open(U_ICUDATA_NAME U_TREE_SEPARATOR_STRING
"coll", locName
, &status
);;
219 if(U_SUCCESS(status
)) {
220 status
= U_ZERO_ERROR
;
221 loc
= ures_getByKey(loc
, "collations", loc
, &status
);
223 if(status
== U_ZERO_ERROR
) { /* do the test - there are real elements */
230 static UCollationResult
compareUsingPartials(UCollator
*coll
, const UChar source
[], int32_t sLen
, const UChar target
[], int32_t tLen
, int32_t pieceSize
, UErrorCode
*status
) {
231 int32_t partialSKResult
= 0;
232 UCharIterator sIter
, tIter
;
233 uint32_t sState
[2], tState
[2];
234 int32_t sSize
= pieceSize
, tSize
= pieceSize
;
236 uint8_t sBuf
[16384], tBuf
[16384];
237 if(pieceSize
> 16384) {
238 log_err("Partial sortkey size buffer too small. Please consider increasing the buffer!\n");
239 *status
= U_BUFFER_OVERFLOW_ERROR
;
242 *status
= U_ZERO_ERROR
;
243 sState
[0] = 0; sState
[1] = 0;
244 tState
[0] = 0; tState
[1] = 0;
245 while(sSize
== pieceSize
&& tSize
== pieceSize
&& partialSKResult
== 0) {
246 uiter_setString(&sIter
, source
, sLen
);
247 uiter_setString(&tIter
, target
, tLen
);
248 sSize
= ucol_nextSortKeyPart(coll
, &sIter
, sState
, sBuf
, pieceSize
, status
);
249 tSize
= ucol_nextSortKeyPart(coll
, &tIter
, tState
, tBuf
, pieceSize
, status
);
251 if(sState
[0] != 0 || tState
[0] != 0) {
252 /*log_verbose("State != 0 : %08X %08X\n", sState[0], tState[0]);*/
254 /*log_verbose("%i ", i++);*/
256 partialSKResult
= memcmp(sBuf
, tBuf
, pieceSize
);
259 if(partialSKResult
< 0) {
261 } else if(partialSKResult
> 0) {
268 static void doTestVariant(UCollator
* myCollation
, const UChar source
[], const UChar target
[], UCollationResult result
)
270 int32_t sortklen1
, sortklen2
, sortklenmax
, sortklenmin
;
271 int temp
=0, gSortklen1
=0,gSortklen2
=0;
272 UCollationResult compareResult
, compareResulta
, keyResult
, compareResultIter
= result
;
273 uint8_t *sortKey1
, *sortKey2
, *sortKey1a
, *sortKey2a
;
274 uint32_t sLen
= u_strlen(source
);
275 uint32_t tLen
= u_strlen(target
);
278 UErrorCode status
= U_ZERO_ERROR
;
279 UColAttributeValue norm
= ucol_getAttribute(myCollation
, UCOL_NORMALIZATION_MODE
, &status
);
281 UCharIterator sIter
, tIter
;
283 compareResult
= ucol_strcoll(myCollation
, source
, sLen
, target
, tLen
);
284 if (compareResult
!= result
) {
285 log_err("ucol_strcoll with explicit length returned wrong result (%i exp. %i): %s, %s\n",
286 compareResult
, result
, aescstrdup(source
,-1), aescstrdup(target
,-1));
288 compareResulta
= ucol_strcoll(myCollation
, source
, -1, target
, -1);
289 if (compareResulta
!= result
) {
290 log_err("ucol_strcoll with null terminated strings returned wrong result (%i exp. %i): %s, %s\n",
291 compareResult
, result
, aescstrdup(source
,-1), aescstrdup(target
,-1));
294 uiter_setString(&sIter
, source
, sLen
);
295 uiter_setString(&tIter
, target
, tLen
);
296 compareResultIter
= ucol_strcollIter(myCollation
, &sIter
, &tIter
, &status
);
297 if(compareResultIter
!= result
) {
298 log_err("different results in iterative comparison for UTF-16 encoded strings. %s, %s\n", aescstrdup(source
,-1), aescstrdup(target
,-1));
301 /* convert the strings to UTF-8 and do try comparing with char iterator and ucol_strcollUTF8 */
303 char utf8Source
[256], utf8Target
[256];
304 int32_t utf8SourceLen
= 0, utf8TargetLen
= 0;
306 u_strToUTF8(utf8Source
, 256, &utf8SourceLen
, source
, sLen
, &status
);
307 if(U_FAILURE(status
)) { /* probably buffer is not big enough */
308 log_verbose("Src UTF-8 buffer too small! Will not compare!\n");
310 u_strToUTF8(utf8Target
, 256, &utf8TargetLen
, target
, tLen
, &status
);
311 if(U_SUCCESS(status
)) {
313 /* ucol_strcollUTF8 */
314 compareResulta
= ucol_strcollUTF8(myCollation
, utf8Source
, utf8SourceLen
, utf8Target
, utf8TargetLen
, &status
);
315 if (U_FAILURE(status
)) {
316 log_err("Error in ucol_strcollUTF8 with explicit length\n");
317 status
= U_ZERO_ERROR
;
318 } else if (compareResulta
!= result
) {
319 log_err("ucol_strcollUTF8 with explicit length returned wrong result (%i exp. %i): %s, %s\n",
320 compareResulta
, result
, aescstrdup(source
,-1), aescstrdup(target
,-1));
322 compareResulta
= ucol_strcollUTF8(myCollation
, utf8Source
, -1, utf8Target
, -1, &status
);
323 if (U_FAILURE(status
)) {
324 log_err("Error in ucol_strcollUTF8 with null terminated strings\n");
325 status
= U_ZERO_ERROR
;
326 } else if (compareResulta
!= result
) {
327 log_err("ucol_strcollUTF8 with null terminated strings returned wrong result (%i exp. %i): %s, %s\n",
328 compareResulta
, result
, aescstrdup(source
,-1), aescstrdup(target
,-1));
333 /* char iterator over UTF8 */
334 UCollationResult compareResultUTF8Iter
= result
, compareResultUTF8IterNorm
= result
;
336 uiter_setUTF8(&sIter
, utf8Source
, utf8SourceLen
);
337 uiter_setUTF8(&tIter
, utf8Target
, utf8TargetLen
);
338 compareResultUTF8Iter
= ucol_strcollIter(myCollation
, &sIter
, &tIter
, &status
);
340 ucol_setAttribute(myCollation
, UCOL_NORMALIZATION_MODE
, UCOL_ON
, &status
);
341 sIter
.move(&sIter
, 0, UITER_START
);
342 tIter
.move(&tIter
, 0, UITER_START
);
343 compareResultUTF8IterNorm
= ucol_strcollIter(myCollation
, &sIter
, &tIter
, &status
);
345 ucol_setAttribute(myCollation
, UCOL_NORMALIZATION_MODE
, norm
, &status
);
346 if(compareResultUTF8Iter
!= compareResultIter
) {
347 log_err("different results in iterative comparison for UTF-16 and UTF-8 encoded strings. %s, %s\n", aescstrdup(source
,-1), aescstrdup(target
,-1));
349 if(compareResultUTF8Iter
!= compareResultUTF8IterNorm
) {
350 log_err("different results in iterative when normalization is turned on with UTF-8 strings. %s, %s\n", aescstrdup(source
,-1), aescstrdup(target
,-1));
354 log_verbose("Target UTF-8 buffer too small! Did not compare!\n");
356 if(U_FAILURE(status
)) {
357 log_verbose("UTF-8 strcoll failed! Ignoring result\n");
362 /* testing the partial sortkeys */
365 int32_t partialSizes
[] = { 3, 1, 2, 4, 8, 20, 80 }; /* just size 3 in the quick mode */
366 int32_t partialSizesSize
= 1;
367 if(getTestOption(QUICK_OPTION
) <= 0) {
368 partialSizesSize
= 7;
370 /*log_verbose("partial sortkey test piecesize=");*/
371 for(i
= 0; i
< partialSizesSize
; i
++) {
372 UCollationResult partialSKResult
= result
, partialNormalizedSKResult
= result
;
373 /*log_verbose("%i ", partialSizes[i]);*/
375 partialSKResult
= compareUsingPartials(myCollation
, source
, sLen
, target
, tLen
, partialSizes
[i
], &status
);
376 if(partialSKResult
!= result
) {
377 log_err("Partial sortkey comparison returned wrong result (%i exp. %i): %s, %s (size %i)\n",
378 partialSKResult
, result
,
379 aescstrdup(source
,-1), aescstrdup(target
,-1), partialSizes
[i
]);
382 if(getTestOption(QUICK_OPTION
) <= 0 && norm
!= UCOL_ON
) {
383 /*log_verbose("N ");*/
384 ucol_setAttribute(myCollation
, UCOL_NORMALIZATION_MODE
, UCOL_ON
, &status
);
385 partialNormalizedSKResult
= compareUsingPartials(myCollation
, source
, sLen
, target
, tLen
, partialSizes
[i
], &status
);
386 ucol_setAttribute(myCollation
, UCOL_NORMALIZATION_MODE
, norm
, &status
);
387 if(partialSKResult
!= partialNormalizedSKResult
) {
388 log_err("Partial sortkey comparison gets different result when normalization is on: %s, %s (size %i)\n",
389 aescstrdup(source
,-1), aescstrdup(target
,-1), partialSizes
[i
]);
393 /*log_verbose("\n");*/
396 sortklen1
=ucol_getSortKey(myCollation
, source
, sLen
, NULL
, 0);
397 sortklen2
=ucol_getSortKey(myCollation
, target
, tLen
, NULL
, 0);
399 sortklenmax
= (sortklen1
>sortklen2
?sortklen1
:sortklen2
);
400 sortklenmin
= (sortklen1
<sortklen2
?sortklen1
:sortklen2
);
401 (void)sortklenmin
; /* Suppress set but not used warning. */
403 sortKey1
=(uint8_t*)malloc(sizeof(uint8_t) * (sortklenmax
+1));
404 sortKey1a
=(uint8_t*)malloc(sizeof(uint8_t) * (sortklenmax
+1));
405 ucol_getSortKey(myCollation
, source
, sLen
, sortKey1
, sortklen1
+1);
406 ucol_getSortKey(myCollation
, source
, -1, sortKey1a
, sortklen1
+1);
408 sortKey2
=(uint8_t*)malloc(sizeof(uint8_t) * (sortklenmax
+1));
409 sortKey2a
=(uint8_t*)malloc(sizeof(uint8_t) * (sortklenmax
+1));
410 ucol_getSortKey(myCollation
, target
, tLen
, sortKey2
, sortklen2
+1);
411 ucol_getSortKey(myCollation
, target
, -1, sortKey2a
, sortklen2
+1);
413 /* Check that sort key generated with null terminated string is identical */
414 /* to that generted with a length specified. */
415 if (uprv_strcmp((const char *)sortKey1
, (const char *)sortKey1a
) != 0 ||
416 uprv_strcmp((const char *)sortKey2
, (const char *)sortKey2a
) != 0 ) {
417 log_err("Sort Keys from null terminated and explicit length strings differ.\n");
420 /*memcmp(sortKey1, sortKey2,sortklenmax);*/
421 temp
= uprv_strcmp((const char *)sortKey1
, (const char *)sortKey2
);
422 gSortklen1
= uprv_strlen((const char *)sortKey1
)+1;
423 gSortklen2
= uprv_strlen((const char *)sortKey2
)+1;
424 if(sortklen1
!= gSortklen1
){
425 log_err("SortKey length does not match Expected: %i Got: %i\n",sortklen1
, gSortklen1
);
426 log_verbose("Generated sortkey: %s\n", sortKeyToString(myCollation
, sortKey1
, buffer
, &len
));
428 if(sortklen2
!= gSortklen2
){
429 log_err("SortKey length does not match Expected: %i Got: %i\n", sortklen2
, gSortklen2
);
430 log_verbose("Generated sortkey: %s\n", sortKeyToString(myCollation
, sortKey2
, buffer
, &len
));
437 keyResult
= UCOL_GREATER
;
440 keyResult
= UCOL_EQUAL
;
442 reportCResult( source
, target
, sortKey1
, sortKey2
, compareResult
, keyResult
, compareResultIter
, result
);
450 void doTest(UCollator
* myCollation
, const UChar source
[], const UChar target
[], UCollationResult result
)
453 doTestVariant(myCollation
, source
, target
, result
);
454 if(result
== UCOL_LESS
) {
455 doTestVariant(myCollation
, target
, source
, UCOL_GREATER
);
456 } else if(result
== UCOL_GREATER
) {
457 doTestVariant(myCollation
, target
, source
, UCOL_LESS
);
459 doTestVariant(myCollation
, target
, source
, UCOL_EQUAL
);
462 log_data_err("No collator! Any data around?\n");
468 * Return an integer array containing all of the collation orders
469 * returned by calls to next on the specified iterator
471 OrderAndOffset
* getOrders(UCollationElements
*iter
, int32_t *orderLength
)
475 int32_t maxSize
= 100;
477 int32_t offset
= ucol_getOffset(iter
);
478 OrderAndOffset
*temp
;
479 OrderAndOffset
*orders
=(OrderAndOffset
*)malloc(sizeof(OrderAndOffset
) * maxSize
);
480 status
= U_ZERO_ERROR
;
483 while ((order
=ucol_next(iter
, &status
)) != UCOL_NULLORDER
)
488 temp
= (OrderAndOffset
*)malloc(sizeof(OrderAndOffset
) * maxSize
);
490 memcpy(temp
, orders
, size
* sizeof(OrderAndOffset
));
496 orders
[size
].order
= order
;
497 orders
[size
].offset
= offset
;
499 offset
= ucol_getOffset(iter
);
503 if (maxSize
> size
&& size
> 0)
505 temp
= (OrderAndOffset
*)malloc(sizeof(OrderAndOffset
) * size
);
507 memcpy(temp
, orders
, size
* sizeof(OrderAndOffset
));
520 backAndForth(UCollationElements
*iter
)
522 /* Run through the iterator forwards and stick it into an array */
524 UErrorCode status
= U_ZERO_ERROR
;
525 int32_t orderLength
= 0;
526 OrderAndOffset
*orders
= getOrders(iter
, &orderLength
);
529 /* Now go through it backwards and make sure we get the same values */
533 /* synwee : changed */
534 while ((o
= ucol_previous(iter
, &status
)) != UCOL_NULLORDER
) {
538 ucol_getOffset(iter
);
541 if (o
!= orders
[idx
].order
) {
545 while (idx
> 0 && orders
[-- idx
].order
== 0) {
549 if (o
!= orders
[idx
].order
) {
550 log_err("Mismatched order at index %d: 0x%8.8X vs. 0x%8.8X\n", idx
,
551 orders
[idx
].order
, o
);
558 if (offset
!= orders
[idx
].offset
) {
559 log_err("Mismatched offset at index %d: %d vs. %d\n", idx
,
560 orders
[idx
].offset
, offset
);
567 while (idx
!= 0 && orders
[idx
- 1].order
== 0) {
572 log_err("Didn't get back to beginning - index is %d\n", idx
);
577 if ((o
= ucol_next(iter
, &status
)) != UCOL_NULLORDER
) {
578 log_err("Error at %x\n", o
);
583 if ((o
= ucol_previous(iter
, &status
)) != UCOL_NULLORDER
) {
584 log_err("Error at %x\n", o
);
594 void genericOrderingTestWithResult(UCollator
*coll
, const char * const s
[], uint32_t size
, UCollationResult result
) {
595 UChar t1
[2048] = {0};
596 UChar t2
[2048] = {0};
597 UCollationElements
*iter
;
598 UErrorCode status
= U_ZERO_ERROR
;
600 uint32_t i
= 0, j
= 0;
601 log_verbose("testing sequence:\n");
602 for(i
= 0; i
< size
; i
++) {
603 log_verbose("%s\n", s
[i
]);
606 iter
= ucol_openElements(coll
, t1
, u_strlen(t1
), &status
);
607 if (U_FAILURE(status
)) {
608 log_err("Creation of iterator failed\n");
610 for(i
= 0; i
< size
-1; i
++) {
611 for(j
= i
+1; j
< size
; j
++) {
612 u_unescape(s
[i
], t1
, 2048);
613 u_unescape(s
[j
], t2
, 2048);
614 doTest(coll
, t1
, t2
, result
);
615 /* synwee : added collation element iterator test */
616 ucol_setText(iter
, t1
, u_strlen(t1
), &status
);
618 ucol_setText(iter
, t2
, u_strlen(t2
), &status
);
622 ucol_closeElements(iter
);
625 void genericOrderingTest(UCollator
*coll
, const char * const s
[], uint32_t size
) {
626 genericOrderingTestWithResult(coll
, s
, size
, UCOL_LESS
);
629 void genericLocaleStarter(const char *locale
, const char * const s
[], uint32_t size
) {
630 UErrorCode status
= U_ZERO_ERROR
;
631 UCollator
*coll
= ucol_open(locale
, &status
);
633 log_verbose("Locale starter for %s\n", locale
);
635 if(U_SUCCESS(status
)) {
636 genericOrderingTest(coll
, s
, size
);
637 } else if(status
== U_FILE_ACCESS_ERROR
) {
638 log_data_err("Is your data around?\n");
641 log_err("Unable to open collator for locale %s\n", locale
);
646 void genericLocaleStarterWithResult(const char *locale
, const char * const s
[], uint32_t size
, UCollationResult result
) {
647 UErrorCode status
= U_ZERO_ERROR
;
648 UCollator
*coll
= ucol_open(locale
, &status
);
650 log_verbose("Locale starter for %s\n", locale
);
652 if(U_SUCCESS(status
)) {
653 genericOrderingTestWithResult(coll
, s
, size
, result
);
654 } else if(status
== U_FILE_ACCESS_ERROR
) {
655 log_data_err("Is your data around?\n");
658 log_err("Unable to open collator for locale %s\n", locale
);
663 /* currently not used with options */
664 void genericRulesStarterWithOptionsAndResult(const char *rules
, const char * const s
[], uint32_t size
, const UColAttribute
*attrs
, const UColAttributeValue
*values
, uint32_t attsize
, UCollationResult result
) {
665 UErrorCode status
= U_ZERO_ERROR
;
666 UChar rlz
[RULE_BUFFER_LEN
] = { 0 };
667 uint32_t rlen
= u_unescape(rules
, rlz
, RULE_BUFFER_LEN
);
670 UCollator
*coll
= ucol_openRules(rlz
, rlen
, UCOL_DEFAULT
, UCOL_DEFAULT
,NULL
, &status
);
672 log_verbose("Rules starter for %s\n", rules
);
674 if(U_SUCCESS(status
)) {
675 log_verbose("Setting attributes\n");
676 for(i
= 0; i
< attsize
; i
++) {
677 ucol_setAttribute(coll
, attrs
[i
], values
[i
], &status
);
680 genericOrderingTestWithResult(coll
, s
, size
, result
);
682 log_err_status(status
, "Unable to open collator with rules %s\n", rules
);
687 void genericLocaleStarterWithOptionsAndResult(const char *locale
, const char * const s
[], uint32_t size
, const UColAttribute
*attrs
, const UColAttributeValue
*values
, uint32_t attsize
, UCollationResult result
) {
688 UErrorCode status
= U_ZERO_ERROR
;
691 UCollator
*coll
= ucol_open(locale
, &status
);
693 log_verbose("Locale starter for %s\n", locale
);
695 if(U_SUCCESS(status
)) {
697 log_verbose("Setting attributes\n");
698 for(i
= 0; i
< attsize
; i
++) {
699 ucol_setAttribute(coll
, attrs
[i
], values
[i
], &status
);
702 genericOrderingTestWithResult(coll
, s
, size
, result
);
704 log_err_status(status
, "Unable to open collator for locale %s\n", locale
);
709 void genericLocaleStarterWithOptions(const char *locale
, const char * const s
[], uint32_t size
, const UColAttribute
*attrs
, const UColAttributeValue
*values
, uint32_t attsize
) {
710 genericLocaleStarterWithOptionsAndResult(locale
, s
, size
, attrs
, values
, attsize
, UCOL_LESS
);
713 void genericRulesStarterWithResult(const char *rules
, const char * const s
[], uint32_t size
, UCollationResult result
) {
714 UErrorCode status
= U_ZERO_ERROR
;
715 UChar rlz
[RULE_BUFFER_LEN
] = { 0 };
716 uint32_t rlen
= u_unescape(rules
, rlz
, RULE_BUFFER_LEN
);
718 UCollator
*coll
= NULL
;
719 coll
= ucol_openRules(rlz
, rlen
, UCOL_DEFAULT
, UCOL_DEFAULT
,NULL
, &status
);
720 log_verbose("Rules starter for %s\n", rules
);
722 if(U_SUCCESS(status
)) {
723 genericOrderingTestWithResult(coll
, s
, size
, result
);
725 } else if(status
== U_FILE_ACCESS_ERROR
) {
726 log_data_err("Is your data around?\n");
728 log_err("Unable to open collator with rules %s\n", rules
);
732 void genericRulesStarter(const char *rules
, const char * const s
[], uint32_t size
) {
733 genericRulesStarterWithResult(rules
, s
, size
, UCOL_LESS
);
736 static void TestTertiary()
739 UCollator
*myCollation
;
740 UErrorCode status
=U_ZERO_ERROR
;
741 static const char str
[]="& C < ch, cH, Ch, CH & Five, 5 & Four, 4 & one, 1 & Ampersand; '&' & Two, 2 ";
742 UChar rules
[sizeof(str
)];
744 u_uastrcpy(rules
, str
);
746 myCollation
=ucol_openRules(rules
, len
, UCOL_OFF
, UCOL_DEFAULT_STRENGTH
, NULL
, &status
);
747 if(U_FAILURE(status
)){
748 log_err_status(status
, "ERROR: in creation of rule based collator :%s\n", myErrorName(status
));
752 ucol_setStrength(myCollation
, UCOL_TERTIARY
);
753 for (i
= 0; i
< 17 ; i
++)
755 doTest(myCollation
, testSourceCases
[i
], testTargetCases
[i
], results
[i
]);
757 ucol_close(myCollation
);
761 static void TestPrimary( )
764 UCollator
*myCollation
;
765 UErrorCode status
=U_ZERO_ERROR
;
766 static const char str
[]="& C < ch, cH, Ch, CH & Five, 5 & Four, 4 & one, 1 & Ampersand; '&' & Two, 2 ";
767 UChar rules
[sizeof(str
)];
769 u_uastrcpy(rules
, str
);
771 myCollation
=ucol_openRules(rules
, len
, UCOL_OFF
, UCOL_DEFAULT_STRENGTH
,NULL
, &status
);
772 if(U_FAILURE(status
)){
773 log_err_status(status
, "ERROR: in creation of rule based collator :%s\n", myErrorName(status
));
776 ucol_setStrength(myCollation
, UCOL_PRIMARY
);
778 for (i
= 17; i
< 26 ; i
++)
781 doTest(myCollation
, testSourceCases
[i
], testTargetCases
[i
], results
[i
]);
783 ucol_close(myCollation
);
787 static void TestSecondary()
791 UCollator
*myCollation
;
792 UErrorCode status
=U_ZERO_ERROR
;
793 static const char str
[]="& C < ch, cH, Ch, CH & Five, 5 & Four, 4 & one, 1 & Ampersand; '&' & Two, 2 ";
794 UChar rules
[sizeof(str
)];
796 u_uastrcpy(rules
, str
);
798 myCollation
=ucol_openRules(rules
, len
, UCOL_OFF
, UCOL_DEFAULT_STRENGTH
,NULL
, &status
);
799 if(U_FAILURE(status
)){
800 log_err_status(status
, "ERROR: in creation of rule based collator :%s\n", myErrorName(status
));
803 ucol_setStrength(myCollation
, UCOL_SECONDARY
);
804 for (i
= 26; i
< 34 ; i
++)
806 doTest(myCollation
, testSourceCases
[i
], testTargetCases
[i
], results
[i
]);
808 ucol_close(myCollation
);
812 static void TestIdentical()
816 UCollator
*myCollation
;
817 UErrorCode status
=U_ZERO_ERROR
;
818 static const char str
[]="& C < ch, cH, Ch, CH & Five, 5 & Four, 4 & one, 1 & Ampersand; '&' & Two, 2 ";
819 UChar rules
[sizeof(str
)];
821 u_uastrcpy(rules
, str
);
823 myCollation
=ucol_openRules(rules
, len
, UCOL_OFF
, UCOL_IDENTICAL
, NULL
,&status
);
824 if(U_FAILURE(status
)){
825 log_err_status(status
, "ERROR: in creation of rule based collator :%s\n", myErrorName(status
));
828 for(i
= 34; i
<37; i
++)
830 doTest(myCollation
, testSourceCases
[i
], testTargetCases
[i
], results
[i
]);
832 ucol_close(myCollation
);
836 static void TestExtra()
840 UCollator
*myCollation
;
841 UErrorCode status
= U_ZERO_ERROR
;
842 static const char str
[]="& C < ch, cH, Ch, CH & Five, 5 & Four, 4 & one, 1 & Ampersand; '&' & Two, 2 ";
843 UChar rules
[sizeof(str
)];
845 u_uastrcpy(rules
, str
);
847 myCollation
=ucol_openRules(rules
, len
, UCOL_OFF
, UCOL_DEFAULT_STRENGTH
,NULL
, &status
);
848 if(U_FAILURE(status
)){
849 log_err_status(status
, "ERROR: in creation of rule based collator :%s\n", myErrorName(status
));
852 ucol_setStrength(myCollation
, UCOL_TERTIARY
);
853 for (i
= 0; i
< COUNT_TEST_CASES
-1 ; i
++)
855 for (j
= i
+ 1; j
< COUNT_TEST_CASES
; j
+= 1)
858 doTest(myCollation
, testCases
[i
], testCases
[j
], UCOL_LESS
);
861 ucol_close(myCollation
);
865 static void TestJB581(void)
867 int32_t bufferLen
= 0;
870 UCollationResult result
= UCOL_EQUAL
;
871 uint8_t sourceKeyArray
[100];
872 uint8_t targetKeyArray
[100];
873 int32_t sourceKeyOut
= 0,
875 UCollator
*myCollator
= 0;
876 UErrorCode status
= U_ZERO_ERROR
;
878 /*u_uastrcpy(source, "This is a test.");*/
879 /*u_uastrcpy(target, "THISISATEST.");*/
880 u_uastrcpy(source
, "THISISATEST.");
881 u_uastrcpy(target
, "Thisisatest.");
883 myCollator
= ucol_open("en_US", &status
);
884 if (U_FAILURE(status
)){
885 log_err_status(status
, "ERROR: Failed to create the collator : %s\n", u_errorName(status
));
888 result
= ucol_strcoll(myCollator
, source
, -1, target
, -1);
889 /* result is 1, secondary differences only for ignorable space characters*/
892 log_err("Comparing two strings with only secondary differences in C failed.\n");
894 /* To compare them with just primary differences */
895 ucol_setStrength(myCollator
, UCOL_PRIMARY
);
896 result
= ucol_strcoll(myCollator
, source
, -1, target
, -1);
900 log_err("Comparing two strings with no differences in C failed.\n");
902 /* Now, do the same comparison with keys */
903 sourceKeyOut
= ucol_getSortKey(myCollator
, source
, -1, sourceKeyArray
, 100);
904 (void)sourceKeyOut
; /* Suppress set but not used warning. */
905 targetKeyOut
= ucol_getSortKey(myCollator
, target
, -1, targetKeyArray
, 100);
906 bufferLen
= ((targetKeyOut
> 100) ? 100 : targetKeyOut
);
907 if (memcmp(sourceKeyArray
, targetKeyArray
, bufferLen
) != 0)
909 log_err("Comparing two strings with sort keys in C failed.\n");
911 ucol_close(myCollator
);
914 static void TestJB1401(void)
916 UCollator
*myCollator
= 0;
917 UErrorCode status
= U_ZERO_ERROR
;
918 static UChar NFD_UnsafeStartChars
[] = {
919 0x0f73, /* Tibetan Vowel Sign II */
920 0x0f75, /* Tibetan Vowel Sign UU */
921 0x0f81, /* Tibetan Vowel Sign Reversed II */
927 myCollator
= ucol_open("en_US", &status
);
928 if (U_FAILURE(status
)){
929 log_err_status(status
, "ERROR: Failed to create the collator : %s\n", u_errorName(status
));
932 ucol_setAttribute(myCollator
, UCOL_NORMALIZATION_MODE
, UCOL_ON
, &status
);
933 if (U_FAILURE(status
)){
934 log_err("ERROR: Failed to set normalization mode ON for collator.\n");
944 /* Get the next funny character to be tested, and set up the
945 * three test strings X, Y, Z, consisting of an A-grave + test char,
946 * in original form, NFD, and then NFC form.
948 c
= NFD_UnsafeStartChars
[i
];
951 X
[0]=0xC0; X
[1]=c
; X
[2]=0; /* \u00C0 is A Grave*/
953 unorm_normalize(X
, -1, UNORM_NFD
, 0, Y
, 20, &status
);
954 unorm_normalize(Y
, -1, UNORM_NFC
, 0, Z
, 20, &status
);
955 if (U_FAILURE(status
)){
956 log_err("ERROR: Failed to normalize test of character %x\n", c
);
960 /* Collation test. All three strings should be equal.
961 * doTest does both strcoll and sort keys, with params in both orders.
963 doTest(myCollator
, X
, Y
, UCOL_EQUAL
);
964 doTest(myCollator
, X
, Z
, UCOL_EQUAL
);
965 doTest(myCollator
, Y
, Z
, UCOL_EQUAL
);
967 /* Run collation element iterators over the three strings. Results should be same for each.
970 UCollationElements
*ceiX
, *ceiY
, *ceiZ
;
971 int32_t ceX
, ceY
, ceZ
;
974 ceiX
= ucol_openElements(myCollator
, X
, -1, &status
);
975 ceiY
= ucol_openElements(myCollator
, Y
, -1, &status
);
976 ceiZ
= ucol_openElements(myCollator
, Z
, -1, &status
);
977 if (U_FAILURE(status
)) {
978 log_err("ERROR: uucol_openElements failed.\n");
983 ceX
= ucol_next(ceiX
, &status
);
984 ceY
= ucol_next(ceiY
, &status
);
985 ceZ
= ucol_next(ceiZ
, &status
);
986 if (U_FAILURE(status
)) {
987 log_err("ERROR: ucol_next failed for iteration #%d.\n", j
);
990 if (ceX
!= ceY
|| ceY
!= ceZ
) {
991 log_err("ERROR: ucol_next failed for iteration #%d.\n", j
);
994 if (ceX
== UCOL_NULLORDER
) {
998 ucol_closeElements(ceiX
);
999 ucol_closeElements(ceiY
);
1000 ucol_closeElements(ceiZ
);
1003 ucol_close(myCollator
);
1009 * Tests the [variable top] tag in rule syntax. Since the default [alternate]
1010 * tag has the value shifted, any codepoints before [variable top] should give
1011 * a primary ce of 0.
1013 static void TestVariableTop(void)
1017 * Starting with ICU 53, setting the variable top via a pseudo relation string
1018 * is not supported any more.
1019 * It was replaced by the [maxVariable symbol] setting.
1020 * See ICU tickets #9958 and #8032.
1022 static const char str
[] = "&z = [variable top]";
1023 int len
= strlen(str
);
1024 UChar rules
[sizeof(str
)];
1025 UCollator
*myCollation
;
1026 UCollator
*enCollation
;
1027 UErrorCode status
= U_ZERO_ERROR
;
1031 uint8_t expected
[20];
1033 u_uastrcpy(rules
, str
);
1035 enCollation
= ucol_open("en_US", &status
);
1036 if (U_FAILURE(status
)) {
1037 log_err_status(status
, "ERROR: in creation of collator :%s\n",
1038 myErrorName(status
));
1041 myCollation
= ucol_openRules(rules
, len
, UCOL_OFF
,
1042 UCOL_PRIMARY
,NULL
, &status
);
1043 if (U_FAILURE(status
)) {
1044 ucol_close(enCollation
);
1045 log_err("ERROR: in creation of rule based collator :%s\n",
1046 myErrorName(status
));
1050 ucol_setStrength(enCollation
, UCOL_PRIMARY
);
1051 ucol_setAttribute(enCollation
, UCOL_ALTERNATE_HANDLING
, UCOL_SHIFTED
,
1053 ucol_setAttribute(myCollation
, UCOL_ALTERNATE_HANDLING
, UCOL_SHIFTED
,
1056 if (ucol_getAttribute(myCollation
, UCOL_ALTERNATE_HANDLING
, &status
) !=
1057 UCOL_SHIFTED
|| U_FAILURE(status
)) {
1058 log_err("ERROR: ALTERNATE_HANDLING value can not be set to SHIFTED\n");
1061 uprv_memset(expected
, 0, 20);
1063 /* space is supposed to be a variable */
1065 len
= ucol_getSortKey(enCollation
, source
, 1, result
,
1068 if (uprv_memcmp(expected
, result
, len
) != 0) {
1069 log_err("ERROR: SHIFTED alternate does not return 0 for primary of space\n");
1075 len
= ucol_getSortKey(myCollation
, source
, 1, result
,
1077 if (uprv_memcmp(expected
, result
, len
) != 0) {
1078 log_err("ERROR: SHIFTED alternate does not return 0 for primary of %c\n",
1084 ucol_close(enCollation
);
1085 ucol_close(myCollation
);
1092 * Tests surrogate support.
1093 * NOTE: This test used \\uD801\\uDC01 pair, which is now assigned to Desseret
1094 * Therefore, another (unassigned) code point was used for this test.
1096 static void TestSurrogates(void)
1098 static const char str
[] =
1099 "&z<'\\uD800\\uDC00'<'\\uD800\\uDC0A\\u0308'<A";
1100 int len
= strlen(str
);
1102 UChar rules
[sizeof(str
)];
1103 UCollator
*myCollation
;
1104 UCollator
*enCollation
;
1105 UErrorCode status
= U_ZERO_ERROR
;
1107 {{'z', 0, 0}, {0xD800, 0xDC00, 0}, {0xD800, 0xDC0A, 0x0308, 0}, {0xD800, 0xDC02}};
1109 {{0xD800, 0xDC00, 0}, {0xD800, 0xDC0A, 0x0308, 0}, {'A', 0, 0}, {0xD800, 0xDC03}};
1111 uint8_t enresult
[20], myresult
[20];
1114 /* tests for open rules with surrogate rules */
1115 rlen
= u_unescape(str
, rules
, len
);
1117 enCollation
= ucol_open("en_US", &status
);
1118 if (U_FAILURE(status
)) {
1119 log_err_status(status
, "ERROR: in creation of collator :%s\n",
1120 myErrorName(status
));
1123 myCollation
= ucol_openRules(rules
, rlen
, UCOL_OFF
,
1124 UCOL_TERTIARY
,NULL
, &status
);
1125 if (U_FAILURE(status
)) {
1126 ucol_close(enCollation
);
1127 log_err("ERROR: in creation of rule based collator :%s\n",
1128 myErrorName(status
));
1133 this test is to verify the supplementary sort key order in the english
1136 log_verbose("start of english collation supplementary characters test\n");
1138 doTest(enCollation
, source
[count
], target
[count
], UCOL_LESS
);
1141 doTest(enCollation
, source
[count
], target
[count
], UCOL_GREATER
);
1143 log_verbose("start of tailored collation supplementary characters test\n");
1145 /* tests getting collation elements for surrogates for tailored rules */
1147 doTest(myCollation
, source
[count
], target
[count
], UCOL_LESS
);
1151 /* tests that \uD800\uDC02 still has the same value, not changed */
1152 enlen
= ucol_getSortKey(enCollation
, source
[3], 2, enresult
, 20);
1153 mylen
= ucol_getSortKey(myCollation
, source
[3], 2, myresult
, 20);
1154 if (enlen
!= mylen
||
1155 uprv_memcmp(enresult
, myresult
, enlen
) != 0) {
1156 log_verbose("Failed : non-tailored supplementary characters should have the same value\n");
1159 ucol_close(enCollation
);
1160 ucol_close(myCollation
);
1166 *### TODO: Add more invalid rules to test all different scenarios.
1171 #define MAX_ERROR_STATES 2
1173 static const char* rulesArr
[MAX_ERROR_STATES
] = {
1174 "& C < ch, cH, Ch[this should fail]<d",
1175 "& C < ch, cH, & Ch[variable top]"
1177 static const char* preContextArr
[MAX_ERROR_STATES
] = {
1182 static const char* postContextArr
[MAX_ERROR_STATES
] = {
1188 for(i
= 0;i
<MAX_ERROR_STATES
;i
++){
1189 UChar rules
[1000] = { '\0' };
1190 UChar preContextExp
[1000] = { '\0' };
1191 UChar postContextExp
[1000] = { '\0' };
1192 UParseError parseError
;
1193 UErrorCode status
= U_ZERO_ERROR
;
1195 u_charsToUChars(rulesArr
[i
],rules
,uprv_strlen(rulesArr
[i
])+1);
1196 u_charsToUChars(preContextArr
[i
],preContextExp
,uprv_strlen(preContextArr
[i
])+1);
1197 u_charsToUChars(postContextArr
[i
],postContextExp
,uprv_strlen(postContextArr
[i
])+1);
1198 /* clean up stuff in parseError */
1199 u_memset(parseError
.preContext
,0x0000,U_PARSE_CONTEXT_LEN
);
1200 u_memset(parseError
.postContext
,0x0000,U_PARSE_CONTEXT_LEN
);
1201 /* open the rules and test */
1202 coll
= ucol_openRules(rules
,u_strlen(rules
),UCOL_OFF
,UCOL_DEFAULT_STRENGTH
,&parseError
,&status
);
1203 (void)coll
; /* Suppress set but not used warning. */
1204 if(u_strcmp(parseError
.preContext
,preContextExp
)!=0){
1205 log_err_status(status
, "preContext in UParseError for ucol_openRules does not match: \"%s\"\n",
1206 aescstrdup(parseError
.preContext
, -1));
1208 if(u_strcmp(parseError
.postContext
,postContextExp
)!=0){
1209 log_err_status(status
, "postContext in UParseError for ucol_openRules does not match: \"%s\"\n",
1210 aescstrdup(parseError
.postContext
, -1));
1216 TestJitterbug1098(){
1218 UCollator
* c1
= NULL
;
1219 UErrorCode status
= U_ZERO_ERROR
;
1220 UParseError parseError
;
1221 char preContext
[200]={0};
1222 char postContext
[200]={0};
1224 const char* rules
[] = {
1232 const UCollationResult results1098
[] = {
1238 const UChar input
[][2]= {
1246 u_memset(parseError
.preContext
,0x0000,U_PARSE_CONTEXT_LEN
);
1247 u_memset(parseError
.postContext
,0x0000,U_PARSE_CONTEXT_LEN
);
1248 for(;rules
[i
]!=0;i
++){
1249 u_uastrcpy(rule
, rules
[i
]);
1250 c1
= ucol_openRules(rule
, u_strlen(rule
), UCOL_OFF
, UCOL_DEFAULT_STRENGTH
, &parseError
, &status
);
1251 if(U_FAILURE(status
)){
1252 log_err_status(status
, "Could not parse the rules syntax. Error: %s\n", u_errorName(status
));
1254 if (status
== U_PARSE_ERROR
) {
1255 u_UCharsToChars(parseError
.preContext
,preContext
,20);
1256 u_UCharsToChars(parseError
.postContext
,postContext
,20);
1257 log_verbose("\n\tPre-Context: %s \n\tPost-Context:%s \n",preContext
,postContext
);
1264 doTest(c1
,X
,Y
,results1098
[i
]);
1270 TestFCDCrash(void) {
1271 static const char *test
[] = {
1272 "Gr\\u00F6\\u00DFe",
1276 UErrorCode status
= U_ZERO_ERROR
;
1277 UCollator
*coll
= ucol_open("es", &status
);
1278 if(U_FAILURE(status
)) {
1279 log_err_status(status
, "Couldn't open collator -> %s\n", u_errorName(status
));
1285 coll
= ucol_open("de_DE", &status
);
1286 if(U_FAILURE(status
)) {
1287 log_err_status(status
, "Couldn't open collator -> %s\n", u_errorName(status
));
1290 ucol_setAttribute(coll
, UCOL_NORMALIZATION_MODE
, UCOL_ON
, &status
);
1291 genericOrderingTest(coll
, test
, 2);
1296 find(UEnumeration* list, const char* str, UErrorCode* status){
1297 const char* value = NULL;
1299 if(U_FAILURE(*status)){
1302 uenum_reset(list, status);
1303 while( (value= uenum_next(list, &length, status))!=NULL){
1304 if(strcmp(value, str)==0){
1311 static void TestJ5298(void)
1313 UErrorCode status
= U_ZERO_ERROR
;
1314 char input
[256], output
[256];
1317 UEnumeration
* values
= NULL
;
1318 const char *keywordValue
= NULL
;
1319 log_verbose("Number of collator locales returned : %i \n", ucol_countAvailable());
1320 values
= ucol_getKeywordValues("collation", &status
);
1321 while ((keywordValue
= uenum_next(values
, NULL
, &status
)) != NULL
) {
1322 if (strncmp(keywordValue
, "private-", 8) == 0) {
1323 log_err("ucol_getKeywordValues() returns private collation keyword: %s\n", keywordValue
);
1326 for (i
= 0; i
< ucol_countAvailable(); i
++) {
1327 uenum_reset(values
, &status
);
1328 while ((keywordValue
= uenum_next(values
, NULL
, &status
)) != NULL
) {
1329 strcpy(input
, ucol_getAvailable(i
));
1330 if (strcmp(keywordValue
, "standard") != 0) {
1331 strcat(input
, "@collation=");
1332 strcat(input
, keywordValue
);
1335 ucol_getFunctionalEquivalent(output
, 256, "collation", input
, &isAvailable
, &status
);
1336 if (strcmp(input
, output
) == 0) { /* Unique locale, print it out */
1337 log_verbose("%s, \n", output
);
1341 uenum_close(values
);
1344 #endif /* #if !UCONFIG_NO_COLLATION */