1 /********************************************************************
3 * Copyright (c) 1997-2010, 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
;
282 uiter_setString(&sIter
, source
, sLen
);
283 uiter_setString(&tIter
, target
, tLen
);
284 compareResultIter
= ucol_strcollIter(myCollation
, &sIter
, &tIter
, &status
);
285 if(compareResultIter
!= result
) {
286 log_err("different results in iterative comparison for UTF-16 encoded strings. %s, %s\n", aescstrdup(source
,-1), aescstrdup(target
,-1));
289 /* convert the strings to UTF-8 and do try comparing with char iterator */
290 if(getTestOption(QUICK_OPTION
) <= 0) { /*!QUICK*/
291 char utf8Source
[256], utf8Target
[256];
292 int32_t utf8SourceLen
= 0, utf8TargetLen
= 0;
293 u_strToUTF8(utf8Source
, 256, &utf8SourceLen
, source
, sLen
, &status
);
294 if(U_FAILURE(status
)) { /* probably buffer is not big enough */
295 log_verbose("Src UTF-8 buffer too small! Will not compare!\n");
297 u_strToUTF8(utf8Target
, 256, &utf8TargetLen
, target
, tLen
, &status
);
298 if(U_SUCCESS(status
)) { /* probably buffer is not big enough */
299 UCollationResult compareResultUTF8
= result
, compareResultUTF8Norm
= result
;
300 /*UCharIterator sIter, tIter;*/
301 /*log_verbose("Strings converted to UTF-8:%s, %s\n", aescstrdup(source,-1), aescstrdup(target,-1));*/
302 uiter_setUTF8(&sIter
, utf8Source
, utf8SourceLen
);
303 uiter_setUTF8(&tIter
, utf8Target
, utf8TargetLen
);
304 /*uiter_setString(&sIter, source, sLen);
305 uiter_setString(&tIter, target, tLen);*/
306 compareResultUTF8
= ucol_strcollIter(myCollation
, &sIter
, &tIter
, &status
);
307 ucol_setAttribute(myCollation
, UCOL_NORMALIZATION_MODE
, UCOL_ON
, &status
);
308 sIter
.move(&sIter
, 0, UITER_START
);
309 tIter
.move(&tIter
, 0, UITER_START
);
310 compareResultUTF8Norm
= ucol_strcollIter(myCollation
, &sIter
, &tIter
, &status
);
311 ucol_setAttribute(myCollation
, UCOL_NORMALIZATION_MODE
, norm
, &status
);
312 if(compareResultUTF8
!= compareResultIter
) {
313 log_err("different results in iterative comparison for UTF-16 and UTF-8 encoded strings. %s, %s\n", aescstrdup(source
,-1), aescstrdup(target
,-1));
315 if(compareResultUTF8
!= compareResultUTF8Norm
) {
316 log_err("different results in iterative when normalization is turned on with UTF-8 strings. %s, %s\n", aescstrdup(source
,-1), aescstrdup(target
,-1));
319 log_verbose("Target UTF-8 buffer too small! Did not compare!\n");
321 if(U_FAILURE(status
)) {
322 log_verbose("UTF-8 strcoll failed! Ignoring result\n");
327 /* testing the partial sortkeys */
330 int32_t partialSizes
[] = { 3, 1, 2, 4, 8, 20, 80 }; /* just size 3 in the quick mode */
331 int32_t partialSizesSize
= 1;
332 if(getTestOption(QUICK_OPTION
) <= 0) {
333 partialSizesSize
= 7;
335 /*log_verbose("partial sortkey test piecesize=");*/
336 for(i
= 0; i
< partialSizesSize
; i
++) {
337 UCollationResult partialSKResult
= result
, partialNormalizedSKResult
= result
;
338 /*log_verbose("%i ", partialSizes[i]);*/
340 partialSKResult
= compareUsingPartials(myCollation
, source
, sLen
, target
, tLen
, partialSizes
[i
], &status
);
341 if(partialSKResult
!= result
) {
342 log_err("Partial sortkey comparison returned wrong result (%i exp. %i): %s, %s (size %i)\n",
343 partialSKResult
, result
,
344 aescstrdup(source
,-1), aescstrdup(target
,-1), partialSizes
[i
]);
347 if(getTestOption(QUICK_OPTION
) <= 0 && norm
!= UCOL_ON
) {
348 /*log_verbose("N ");*/
349 ucol_setAttribute(myCollation
, UCOL_NORMALIZATION_MODE
, UCOL_ON
, &status
);
350 partialNormalizedSKResult
= compareUsingPartials(myCollation
, source
, sLen
, target
, tLen
, partialSizes
[i
], &status
);
351 ucol_setAttribute(myCollation
, UCOL_NORMALIZATION_MODE
, norm
, &status
);
352 if(partialSKResult
!= partialNormalizedSKResult
) {
353 log_err("Partial sortkey comparison gets different result when normalization is on: %s, %s (size %i)\n",
354 aescstrdup(source
,-1), aescstrdup(target
,-1), partialSizes
[i
]);
358 /*log_verbose("\n");*/
362 compareResult
= ucol_strcoll(myCollation
, source
, sLen
, target
, tLen
);
363 compareResulta
= ucol_strcoll(myCollation
, source
, -1, target
, -1);
364 if (compareResult
!= compareResulta
) {
365 log_err("ucol_strcoll result from null terminated and explicit length strings differs.\n");
368 sortklen1
=ucol_getSortKey(myCollation
, source
, sLen
, NULL
, 0);
369 sortklen2
=ucol_getSortKey(myCollation
, target
, tLen
, NULL
, 0);
371 sortklenmax
= (sortklen1
>sortklen2
?sortklen1
:sortklen2
);
372 sortklenmin
= (sortklen1
<sortklen2
?sortklen1
:sortklen2
);
374 sortKey1
=(uint8_t*)malloc(sizeof(uint8_t) * (sortklenmax
+1));
375 sortKey1a
=(uint8_t*)malloc(sizeof(uint8_t) * (sortklenmax
+1));
376 ucol_getSortKey(myCollation
, source
, sLen
, sortKey1
, sortklen1
+1);
377 ucol_getSortKey(myCollation
, source
, -1, sortKey1a
, sortklen1
+1);
379 sortKey2
=(uint8_t*)malloc(sizeof(uint8_t) * (sortklenmax
+1));
380 sortKey2a
=(uint8_t*)malloc(sizeof(uint8_t) * (sortklenmax
+1));
381 ucol_getSortKey(myCollation
, target
, tLen
, sortKey2
, sortklen2
+1);
382 ucol_getSortKey(myCollation
, target
, -1, sortKey2a
, sortklen2
+1);
384 /* Check that sort key generated with null terminated string is identical */
385 /* to that generted with a length specified. */
386 if (uprv_strcmp((const char *)sortKey1
, (const char *)sortKey1a
) != 0 ||
387 uprv_strcmp((const char *)sortKey2
, (const char *)sortKey2a
) != 0 ) {
388 log_err("Sort Keys from null terminated and explicit length strings differ.\n");
391 /*memcmp(sortKey1, sortKey2,sortklenmax);*/
392 temp
= uprv_strcmp((const char *)sortKey1
, (const char *)sortKey2
);
393 gSortklen1
= uprv_strlen((const char *)sortKey1
)+1;
394 gSortklen2
= uprv_strlen((const char *)sortKey2
)+1;
395 if(sortklen1
!= gSortklen1
){
396 log_err("SortKey length does not match Expected: %i Got: %i\n",sortklen1
, gSortklen1
);
397 log_verbose("Generated sortkey: %s\n", sortKeyToString(myCollation
, sortKey1
, buffer
, &len
));
399 if(sortklen2
!= gSortklen2
){
400 log_err("SortKey length does not match Expected: %i Got: %i\n", sortklen2
, gSortklen2
);
401 log_verbose("Generated sortkey: %s\n", sortKeyToString(myCollation
, sortKey2
, buffer
, &len
));
408 keyResult
= UCOL_GREATER
;
411 keyResult
= UCOL_EQUAL
;
413 reportCResult( source
, target
, sortKey1
, sortKey2
, compareResult
, keyResult
, compareResultIter
, result
);
421 void doTest(UCollator
* myCollation
, const UChar source
[], const UChar target
[], UCollationResult result
)
424 doTestVariant(myCollation
, source
, target
, result
);
425 if(result
== UCOL_LESS
) {
426 doTestVariant(myCollation
, target
, source
, UCOL_GREATER
);
427 } else if(result
== UCOL_GREATER
) {
428 doTestVariant(myCollation
, target
, source
, UCOL_LESS
);
430 doTestVariant(myCollation
, target
, source
, UCOL_EQUAL
);
433 log_data_err("No collator! Any data around?\n");
439 * Return an integer array containing all of the collation orders
440 * returned by calls to next on the specified iterator
442 OrderAndOffset
* getOrders(UCollationElements
*iter
, int32_t *orderLength
)
446 int32_t maxSize
= 100;
448 int32_t offset
= ucol_getOffset(iter
);
449 OrderAndOffset
*temp
;
450 OrderAndOffset
*orders
=(OrderAndOffset
*)malloc(sizeof(OrderAndOffset
) * maxSize
);
451 status
= U_ZERO_ERROR
;
454 while ((order
=ucol_next(iter
, &status
)) != UCOL_NULLORDER
)
459 temp
= (OrderAndOffset
*)malloc(sizeof(OrderAndOffset
) * maxSize
);
461 memcpy(temp
, orders
, size
* sizeof(OrderAndOffset
));
467 orders
[size
].order
= order
;
468 orders
[size
].offset
= offset
;
470 offset
= ucol_getOffset(iter
);
474 if (maxSize
> size
&& size
> 0)
476 temp
= (OrderAndOffset
*)malloc(sizeof(OrderAndOffset
) * size
);
478 memcpy(temp
, orders
, size
* sizeof(OrderAndOffset
));
491 backAndForth(UCollationElements
*iter
)
493 /* Run through the iterator forwards and stick it into an array */
495 UErrorCode status
= U_ZERO_ERROR
;
496 int32_t orderLength
= 0;
497 OrderAndOffset
*orders
= getOrders(iter
, &orderLength
);
500 /* Now go through it backwards and make sure we get the same values */
504 /* synwee : changed */
505 while ((o
= ucol_previous(iter
, &status
)) != UCOL_NULLORDER
) {
509 ucol_getOffset(iter
);
512 if (o
!= orders
[index
].order
) {
516 while (index
> 0 && orders
[-- index
].order
== 0) {
520 if (o
!= orders
[index
].order
) {
521 log_err("Mismatched order at index %d: 0x%8.8X vs. 0x%8.8X\n", index
,
522 orders
[index
].order
, o
);
529 if (offset
!= orders
[index
].offset
) {
530 log_err("Mismatched offset at index %d: %d vs. %d\n", index
,
531 orders
[index
].offset
, offset
);
538 while (index
!= 0 && orders
[index
- 1].order
== 0) {
543 log_err("Didn't get back to beginning - index is %d\n", index
);
548 if ((o
= ucol_next(iter
, &status
)) != UCOL_NULLORDER
) {
549 log_err("Error at %x\n", o
);
554 if ((o
= ucol_previous(iter
, &status
)) != UCOL_NULLORDER
) {
555 log_err("Error at %x\n", o
);
565 void genericOrderingTestWithResult(UCollator
*coll
, const char * const s
[], uint32_t size
, UCollationResult result
) {
566 UChar t1
[2048] = {0};
567 UChar t2
[2048] = {0};
568 UCollationElements
*iter
;
569 UErrorCode status
= U_ZERO_ERROR
;
571 uint32_t i
= 0, j
= 0;
572 log_verbose("testing sequence:\n");
573 for(i
= 0; i
< size
; i
++) {
574 log_verbose("%s\n", s
[i
]);
577 iter
= ucol_openElements(coll
, t1
, u_strlen(t1
), &status
);
578 if (U_FAILURE(status
)) {
579 log_err("Creation of iterator failed\n");
581 for(i
= 0; i
< size
-1; i
++) {
582 for(j
= i
+1; j
< size
; j
++) {
583 u_unescape(s
[i
], t1
, 2048);
584 u_unescape(s
[j
], t2
, 2048);
585 doTest(coll
, t1
, t2
, result
);
586 /* synwee : added collation element iterator test */
587 ucol_setText(iter
, t1
, u_strlen(t1
), &status
);
589 ucol_setText(iter
, t2
, u_strlen(t2
), &status
);
593 ucol_closeElements(iter
);
596 void genericOrderingTest(UCollator
*coll
, const char * const s
[], uint32_t size
) {
597 genericOrderingTestWithResult(coll
, s
, size
, UCOL_LESS
);
600 void genericLocaleStarter(const char *locale
, const char * const s
[], uint32_t size
) {
601 UErrorCode status
= U_ZERO_ERROR
;
602 UCollator
*coll
= ucol_open(locale
, &status
);
604 log_verbose("Locale starter for %s\n", locale
);
606 if(U_SUCCESS(status
)) {
607 genericOrderingTest(coll
, s
, size
);
608 } else if(status
== U_FILE_ACCESS_ERROR
) {
609 log_data_err("Is your data around?\n");
612 log_err("Unable to open collator for locale %s\n", locale
);
617 void genericLocaleStarterWithResult(const char *locale
, const char * const s
[], uint32_t size
, UCollationResult result
) {
618 UErrorCode status
= U_ZERO_ERROR
;
619 UCollator
*coll
= ucol_open(locale
, &status
);
621 log_verbose("Locale starter for %s\n", locale
);
623 if(U_SUCCESS(status
)) {
624 genericOrderingTestWithResult(coll
, s
, size
, result
);
625 } else if(status
== U_FILE_ACCESS_ERROR
) {
626 log_data_err("Is your data around?\n");
629 log_err("Unable to open collator for locale %s\n", locale
);
634 /* currently not used with options */
635 void genericRulesStarterWithOptionsAndResult(const char *rules
, const char * const s
[], uint32_t size
, const UColAttribute
*attrs
, const UColAttributeValue
*values
, uint32_t attsize
, UCollationResult result
) {
636 UErrorCode status
= U_ZERO_ERROR
;
637 UChar rlz
[RULE_BUFFER_LEN
] = { 0 };
638 uint32_t rlen
= u_unescape(rules
, rlz
, RULE_BUFFER_LEN
);
641 UCollator
*coll
= ucol_openRules(rlz
, rlen
, UCOL_DEFAULT
, UCOL_DEFAULT
,NULL
, &status
);
643 log_verbose("Rules starter for %s\n", rules
);
645 if(U_SUCCESS(status
)) {
646 log_verbose("Setting attributes\n");
647 for(i
= 0; i
< attsize
; i
++) {
648 ucol_setAttribute(coll
, attrs
[i
], values
[i
], &status
);
651 genericOrderingTestWithResult(coll
, s
, size
, result
);
653 log_err_status(status
, "Unable to open collator with rules %s\n", rules
);
658 void genericLocaleStarterWithOptionsAndResult(const char *locale
, const char * const s
[], uint32_t size
, const UColAttribute
*attrs
, const UColAttributeValue
*values
, uint32_t attsize
, UCollationResult result
) {
659 UErrorCode status
= U_ZERO_ERROR
;
662 UCollator
*coll
= ucol_open(locale
, &status
);
664 log_verbose("Locale starter for %s\n", locale
);
666 if(U_SUCCESS(status
)) {
668 log_verbose("Setting attributes\n");
669 for(i
= 0; i
< attsize
; i
++) {
670 ucol_setAttribute(coll
, attrs
[i
], values
[i
], &status
);
673 genericOrderingTestWithResult(coll
, s
, size
, result
);
675 log_err_status(status
, "Unable to open collator for locale %s\n", locale
);
680 void genericLocaleStarterWithOptions(const char *locale
, const char * const s
[], uint32_t size
, const UColAttribute
*attrs
, const UColAttributeValue
*values
, uint32_t attsize
) {
681 genericLocaleStarterWithOptionsAndResult(locale
, s
, size
, attrs
, values
, attsize
, UCOL_LESS
);
684 void genericRulesStarterWithResult(const char *rules
, const char * const s
[], uint32_t size
, UCollationResult result
) {
685 UErrorCode status
= U_ZERO_ERROR
;
686 UChar rlz
[RULE_BUFFER_LEN
] = { 0 };
687 uint32_t rlen
= u_unescape(rules
, rlz
, RULE_BUFFER_LEN
);
689 UCollator
*coll
= NULL
;
690 coll
= ucol_openRules(rlz
, rlen
, UCOL_DEFAULT
, UCOL_DEFAULT
,NULL
, &status
);
691 log_verbose("Rules starter for %s\n", rules
);
693 if(U_SUCCESS(status
)) {
694 genericOrderingTestWithResult(coll
, s
, size
, result
);
696 } else if(status
== U_FILE_ACCESS_ERROR
) {
697 log_data_err("Is your data around?\n");
699 log_err("Unable to open collator with rules %s\n", rules
);
703 void genericRulesStarter(const char *rules
, const char * const s
[], uint32_t size
) {
704 genericRulesStarterWithResult(rules
, s
, size
, UCOL_LESS
);
707 static void TestTertiary()
710 UCollator
*myCollation
;
711 UErrorCode status
=U_ZERO_ERROR
;
712 static const char str
[]="& C < ch, cH, Ch, CH & Five, 5 & Four, 4 & one, 1 & Ampersand; '&' & Two, 2 ";
713 UChar rules
[sizeof(str
)];
715 u_uastrcpy(rules
, str
);
717 myCollation
=ucol_openRules(rules
, len
, UCOL_OFF
, UCOL_DEFAULT_STRENGTH
, NULL
, &status
);
718 if(U_FAILURE(status
)){
719 log_err_status(status
, "ERROR: in creation of rule based collator :%s\n", myErrorName(status
));
723 ucol_setStrength(myCollation
, UCOL_TERTIARY
);
724 for (i
= 0; i
< 17 ; i
++)
726 doTest(myCollation
, testSourceCases
[i
], testTargetCases
[i
], results
[i
]);
728 ucol_close(myCollation
);
732 static void TestPrimary( )
735 UCollator
*myCollation
;
736 UErrorCode status
=U_ZERO_ERROR
;
737 static const char str
[]="& C < ch, cH, Ch, CH & Five, 5 & Four, 4 & one, 1 & Ampersand; '&' & Two, 2 ";
738 UChar rules
[sizeof(str
)];
740 u_uastrcpy(rules
, str
);
742 myCollation
=ucol_openRules(rules
, len
, UCOL_OFF
, UCOL_DEFAULT_STRENGTH
,NULL
, &status
);
743 if(U_FAILURE(status
)){
744 log_err_status(status
, "ERROR: in creation of rule based collator :%s\n", myErrorName(status
));
747 ucol_setStrength(myCollation
, UCOL_PRIMARY
);
749 for (i
= 17; i
< 26 ; i
++)
752 doTest(myCollation
, testSourceCases
[i
], testTargetCases
[i
], results
[i
]);
754 ucol_close(myCollation
);
758 static void TestSecondary()
762 UCollator
*myCollation
;
763 UErrorCode status
=U_ZERO_ERROR
;
764 static const char str
[]="& C < ch, cH, Ch, CH & Five, 5 & Four, 4 & one, 1 & Ampersand; '&' & Two, 2 ";
765 UChar rules
[sizeof(str
)];
767 u_uastrcpy(rules
, str
);
769 myCollation
=ucol_openRules(rules
, len
, UCOL_OFF
, UCOL_DEFAULT_STRENGTH
,NULL
, &status
);
770 if(U_FAILURE(status
)){
771 log_err_status(status
, "ERROR: in creation of rule based collator :%s\n", myErrorName(status
));
774 ucol_setStrength(myCollation
, UCOL_SECONDARY
);
775 for (i
= 26; i
< 34 ; i
++)
777 doTest(myCollation
, testSourceCases
[i
], testTargetCases
[i
], results
[i
]);
779 ucol_close(myCollation
);
783 static void TestIdentical()
787 UCollator
*myCollation
;
788 UErrorCode status
=U_ZERO_ERROR
;
789 static const char str
[]="& C < ch, cH, Ch, CH & Five, 5 & Four, 4 & one, 1 & Ampersand; '&' & Two, 2 ";
790 UChar rules
[sizeof(str
)];
792 u_uastrcpy(rules
, str
);
794 myCollation
=ucol_openRules(rules
, len
, UCOL_OFF
, UCOL_IDENTICAL
, NULL
,&status
);
795 if(U_FAILURE(status
)){
796 log_err_status(status
, "ERROR: in creation of rule based collator :%s\n", myErrorName(status
));
799 for(i
= 34; i
<37; i
++)
801 doTest(myCollation
, testSourceCases
[i
], testTargetCases
[i
], results
[i
]);
803 ucol_close(myCollation
);
807 static void TestExtra()
811 UCollator
*myCollation
;
812 UErrorCode status
= U_ZERO_ERROR
;
813 static const char str
[]="& C < ch, cH, Ch, CH & Five, 5 & Four, 4 & one, 1 & Ampersand; '&' & Two, 2 ";
814 UChar rules
[sizeof(str
)];
816 u_uastrcpy(rules
, str
);
818 myCollation
=ucol_openRules(rules
, len
, UCOL_OFF
, UCOL_DEFAULT_STRENGTH
,NULL
, &status
);
819 if(U_FAILURE(status
)){
820 log_err_status(status
, "ERROR: in creation of rule based collator :%s\n", myErrorName(status
));
823 ucol_setStrength(myCollation
, UCOL_TERTIARY
);
824 for (i
= 0; i
< COUNT_TEST_CASES
-1 ; i
++)
826 for (j
= i
+ 1; j
< COUNT_TEST_CASES
; j
+= 1)
829 doTest(myCollation
, testCases
[i
], testCases
[j
], UCOL_LESS
);
832 ucol_close(myCollation
);
836 static void TestJB581(void)
838 int32_t bufferLen
= 0;
841 UCollationResult result
= UCOL_EQUAL
;
842 uint8_t sourceKeyArray
[100];
843 uint8_t targetKeyArray
[100];
844 int32_t sourceKeyOut
= 0,
846 UCollator
*myCollator
= 0;
847 UErrorCode status
= U_ZERO_ERROR
;
849 /*u_uastrcpy(source, "This is a test.");*/
850 /*u_uastrcpy(target, "THISISATEST.");*/
851 u_uastrcpy(source
, "THISISATEST.");
852 u_uastrcpy(target
, "Thisisatest.");
854 myCollator
= ucol_open("en_US", &status
);
855 if (U_FAILURE(status
)){
856 log_err_status(status
, "ERROR: Failed to create the collator : %s\n", u_errorName(status
));
859 result
= ucol_strcoll(myCollator
, source
, -1, target
, -1);
860 /* result is 1, secondary differences only for ignorable space characters*/
863 log_err("Comparing two strings with only secondary differences in C failed.\n");
865 /* To compare them with just primary differences */
866 ucol_setStrength(myCollator
, UCOL_PRIMARY
);
867 result
= ucol_strcoll(myCollator
, source
, -1, target
, -1);
871 log_err("Comparing two strings with no differences in C failed.\n");
873 /* Now, do the same comparison with keys */
874 sourceKeyOut
= ucol_getSortKey(myCollator
, source
, -1, sourceKeyArray
, 100);
875 targetKeyOut
= ucol_getSortKey(myCollator
, target
, -1, targetKeyArray
, 100);
876 bufferLen
= ((targetKeyOut
> 100) ? 100 : targetKeyOut
);
877 if (memcmp(sourceKeyArray
, targetKeyArray
, bufferLen
) != 0)
879 log_err("Comparing two strings with sort keys in C failed.\n");
881 ucol_close(myCollator
);
884 static void TestJB1401(void)
886 UCollator
*myCollator
= 0;
887 UErrorCode status
= U_ZERO_ERROR
;
888 static UChar NFD_UnsafeStartChars
[] = {
889 0x0f73, /* Tibetan Vowel Sign II */
890 0x0f75, /* Tibetan Vowel Sign UU */
891 0x0f81, /* Tibetan Vowel Sign Reversed II */
897 myCollator
= ucol_open("en_US", &status
);
898 if (U_FAILURE(status
)){
899 log_err_status(status
, "ERROR: Failed to create the collator : %s\n", u_errorName(status
));
902 ucol_setAttribute(myCollator
, UCOL_NORMALIZATION_MODE
, UCOL_ON
, &status
);
903 if (U_FAILURE(status
)){
904 log_err("ERROR: Failed to set normalization mode ON for collator.\n");
914 /* Get the next funny character to be tested, and set up the
915 * three test strings X, Y, Z, consisting of an A-grave + test char,
916 * in original form, NFD, and then NFC form.
918 c
= NFD_UnsafeStartChars
[i
];
921 X
[0]=0xC0; X
[1]=c
; X
[2]=0; /* \u00C0 is A Grave*/
923 unorm_normalize(X
, -1, UNORM_NFD
, 0, Y
, 20, &status
);
924 unorm_normalize(Y
, -1, UNORM_NFC
, 0, Z
, 20, &status
);
925 if (U_FAILURE(status
)){
926 log_err("ERROR: Failed to normalize test of character %x\n", c
);
930 /* Collation test. All three strings should be equal.
931 * doTest does both strcoll and sort keys, with params in both orders.
933 doTest(myCollator
, X
, Y
, UCOL_EQUAL
);
934 doTest(myCollator
, X
, Z
, UCOL_EQUAL
);
935 doTest(myCollator
, Y
, Z
, UCOL_EQUAL
);
937 /* Run collation element iterators over the three strings. Results should be same for each.
940 UCollationElements
*ceiX
, *ceiY
, *ceiZ
;
941 int32_t ceX
, ceY
, ceZ
;
944 ceiX
= ucol_openElements(myCollator
, X
, -1, &status
);
945 ceiY
= ucol_openElements(myCollator
, Y
, -1, &status
);
946 ceiZ
= ucol_openElements(myCollator
, Z
, -1, &status
);
947 if (U_FAILURE(status
)) {
948 log_err("ERROR: uucol_openElements failed.\n");
953 ceX
= ucol_next(ceiX
, &status
);
954 ceY
= ucol_next(ceiY
, &status
);
955 ceZ
= ucol_next(ceiZ
, &status
);
956 if (U_FAILURE(status
)) {
957 log_err("ERROR: ucol_next failed for iteration #%d.\n", j
);
960 if (ceX
!= ceY
|| ceY
!= ceZ
) {
961 log_err("ERROR: ucol_next failed for iteration #%d.\n", j
);
964 if (ceX
== UCOL_NULLORDER
) {
968 ucol_closeElements(ceiX
);
969 ucol_closeElements(ceiY
);
970 ucol_closeElements(ceiZ
);
973 ucol_close(myCollator
);
979 * Tests the [variable top] tag in rule syntax. Since the default [alternate]
980 * tag has the value shifted, any codepoints before [variable top] should give
983 static void TestVariableTop(void)
985 static const char str
[] = "&z = [variable top]";
986 int len
= strlen(str
);
987 UChar rules
[sizeof(str
)];
988 UCollator
*myCollation
;
989 UCollator
*enCollation
;
990 UErrorCode status
= U_ZERO_ERROR
;
994 uint8_t expected
[20];
996 u_uastrcpy(rules
, str
);
998 enCollation
= ucol_open("en_US", &status
);
999 if (U_FAILURE(status
)) {
1000 log_err_status(status
, "ERROR: in creation of collator :%s\n",
1001 myErrorName(status
));
1004 myCollation
= ucol_openRules(rules
, len
, UCOL_OFF
,
1005 UCOL_PRIMARY
,NULL
, &status
);
1006 if (U_FAILURE(status
)) {
1007 ucol_close(enCollation
);
1008 log_err("ERROR: in creation of rule based collator :%s\n",
1009 myErrorName(status
));
1013 ucol_setStrength(enCollation
, UCOL_PRIMARY
);
1014 ucol_setAttribute(enCollation
, UCOL_ALTERNATE_HANDLING
, UCOL_SHIFTED
,
1016 ucol_setAttribute(myCollation
, UCOL_ALTERNATE_HANDLING
, UCOL_SHIFTED
,
1019 if (ucol_getAttribute(myCollation
, UCOL_ALTERNATE_HANDLING
, &status
) !=
1020 UCOL_SHIFTED
|| U_FAILURE(status
)) {
1021 log_err("ERROR: ALTERNATE_HANDLING value can not be set to SHIFTED\n");
1024 uprv_memset(expected
, 0, 20);
1026 /* space is supposed to be a variable */
1028 len
= ucol_getSortKey(enCollation
, source
, 1, result
,
1031 if (uprv_memcmp(expected
, result
, len
) != 0) {
1032 log_err("ERROR: SHIFTED alternate does not return 0 for primary of space\n");
1038 len
= ucol_getSortKey(myCollation
, source
, 1, result
,
1040 if (uprv_memcmp(expected
, result
, len
) != 0) {
1041 log_err("ERROR: SHIFTED alternate does not return 0 for primary of %c\n",
1047 ucol_close(enCollation
);
1048 ucol_close(myCollation
);
1054 * Tests surrogate support.
1055 * NOTE: This test used \\uD801\\uDC01 pair, which is now assigned to Desseret
1056 * Therefore, another (unassigned) code point was used for this test.
1058 static void TestSurrogates(void)
1060 static const char str
[] =
1061 "&z<'\\uD800\\uDC00'<'\\uD800\\uDC0A\\u0308'<A";
1062 int len
= strlen(str
);
1064 UChar rules
[sizeof(str
)];
1065 UCollator
*myCollation
;
1066 UCollator
*enCollation
;
1067 UErrorCode status
= U_ZERO_ERROR
;
1069 {{'z', 0, 0}, {0xD800, 0xDC00, 0}, {0xD800, 0xDC0A, 0x0308, 0}, {0xD800, 0xDC02}};
1071 {{0xD800, 0xDC00, 0}, {0xD800, 0xDC0A, 0x0308, 0}, {'A', 0, 0}, {0xD800, 0xDC03}};
1073 uint8_t enresult
[20], myresult
[20];
1076 /* tests for open rules with surrogate rules */
1077 rlen
= u_unescape(str
, rules
, len
);
1079 enCollation
= ucol_open("en_US", &status
);
1080 if (U_FAILURE(status
)) {
1081 log_err_status(status
, "ERROR: in creation of collator :%s\n",
1082 myErrorName(status
));
1085 myCollation
= ucol_openRules(rules
, rlen
, UCOL_OFF
,
1086 UCOL_TERTIARY
,NULL
, &status
);
1087 if (U_FAILURE(status
)) {
1088 ucol_close(enCollation
);
1089 log_err("ERROR: in creation of rule based collator :%s\n",
1090 myErrorName(status
));
1095 this test is to verify the supplementary sort key order in the english
1098 log_verbose("start of english collation supplementary characters test\n");
1100 doTest(enCollation
, source
[count
], target
[count
], UCOL_LESS
);
1103 doTest(enCollation
, source
[count
], target
[count
], UCOL_GREATER
);
1105 log_verbose("start of tailored collation supplementary characters test\n");
1107 /* tests getting collation elements for surrogates for tailored rules */
1109 doTest(myCollation
, source
[count
], target
[count
], UCOL_LESS
);
1113 /* tests that \uD800\uDC02 still has the same value, not changed */
1114 enlen
= ucol_getSortKey(enCollation
, source
[3], 2, enresult
, 20);
1115 mylen
= ucol_getSortKey(myCollation
, source
[3], 2, myresult
, 20);
1116 if (enlen
!= mylen
||
1117 uprv_memcmp(enresult
, myresult
, enlen
) != 0) {
1118 log_verbose("Failed : non-tailored supplementary characters should have the same value\n");
1121 ucol_close(enCollation
);
1122 ucol_close(myCollation
);
1128 *### TODO: Add more invalid rules to test all different scenarios.
1133 #define MAX_ERROR_STATES 2
1135 static const char* rulesArr
[MAX_ERROR_STATES
] = {
1136 "& C < ch, cH, Ch[this should fail]<d",
1137 "& C < ch, cH, & Ch[variable top]"
1139 static const char* preContextArr
[MAX_ERROR_STATES
] = {
1144 static const char* postContextArr
[MAX_ERROR_STATES
] = {
1150 for(i
= 0;i
<MAX_ERROR_STATES
;i
++){
1151 UChar rules
[1000] = { '\0' };
1152 UChar preContextExp
[1000] = { '\0' };
1153 UChar postContextExp
[1000] = { '\0' };
1154 UParseError parseError
;
1155 UErrorCode status
= U_ZERO_ERROR
;
1157 u_charsToUChars(rulesArr
[i
],rules
,uprv_strlen(rulesArr
[i
])+1);
1158 u_charsToUChars(preContextArr
[i
],preContextExp
,uprv_strlen(preContextArr
[i
])+1);
1159 u_charsToUChars(postContextArr
[i
],postContextExp
,uprv_strlen(postContextArr
[i
])+1);
1160 /* clean up stuff in parseError */
1161 u_memset(parseError
.preContext
,0x0000,U_PARSE_CONTEXT_LEN
);
1162 u_memset(parseError
.postContext
,0x0000,U_PARSE_CONTEXT_LEN
);
1163 /* open the rules and test */
1164 coll
= ucol_openRules(rules
,u_strlen(rules
),UCOL_OFF
,UCOL_DEFAULT_STRENGTH
,&parseError
,&status
);
1165 if(u_strcmp(parseError
.preContext
,preContextExp
)!=0){
1166 log_err_status(status
, "preContext in UParseError for ucol_openRules does not match\n");
1168 if(u_strcmp(parseError
.postContext
,postContextExp
)!=0){
1169 log_err_status(status
, "postContext in UParseError for ucol_openRules does not match\n");
1175 TestJitterbug1098(){
1177 UCollator
* c1
= NULL
;
1178 UErrorCode status
= U_ZERO_ERROR
;
1179 UParseError parseError
;
1180 char preContext
[200]={0};
1181 char postContext
[200]={0};
1183 const char* rules
[] = {
1191 const UCollationResult results1098
[] = {
1197 const UChar input
[][2]= {
1205 u_memset(parseError
.preContext
,0x0000,U_PARSE_CONTEXT_LEN
);
1206 u_memset(parseError
.postContext
,0x0000,U_PARSE_CONTEXT_LEN
);
1207 for(;rules
[i
]!=0;i
++){
1208 u_uastrcpy(rule
, rules
[i
]);
1209 c1
= ucol_openRules(rule
, u_strlen(rule
), UCOL_OFF
, UCOL_DEFAULT_STRENGTH
, &parseError
, &status
);
1210 if(U_FAILURE(status
)){
1211 log_err_status(status
, "Could not parse the rules syntax. Error: %s\n", u_errorName(status
));
1213 if (status
== U_PARSE_ERROR
) {
1214 u_UCharsToChars(parseError
.preContext
,preContext
,20);
1215 u_UCharsToChars(parseError
.postContext
,postContext
,20);
1216 log_verbose("\n\tPre-Context: %s \n\tPost-Context:%s \n",preContext
,postContext
);
1223 doTest(c1
,X
,Y
,results1098
[i
]);
1229 TestFCDCrash(void) {
1230 static const char *test
[] = {
1231 "Gr\\u00F6\\u00DFe",
1235 UErrorCode status
= U_ZERO_ERROR
;
1236 UCollator
*coll
= ucol_open("es", &status
);
1237 if(U_FAILURE(status
)) {
1238 log_err_status(status
, "Couldn't open collator -> %s\n", u_errorName(status
));
1244 coll
= ucol_open("de_DE", &status
);
1245 if(U_FAILURE(status
)) {
1246 log_err_status(status
, "Couldn't open collator -> %s\n", u_errorName(status
));
1249 ucol_setAttribute(coll
, UCOL_NORMALIZATION_MODE
, UCOL_ON
, &status
);
1250 genericOrderingTest(coll
, test
, 2);
1255 find(UEnumeration* list, const char* str, UErrorCode* status){
1256 const char* value = NULL;
1258 if(U_FAILURE(*status)){
1261 uenum_reset(list, status);
1262 while( (value= uenum_next(list, &length, status))!=NULL){
1263 if(strcmp(value, str)==0){
1270 static void TestJ5298(void)
1272 UErrorCode status
= U_ZERO_ERROR
;
1273 char input
[256], output
[256];
1276 UEnumeration
* values
= NULL
;
1277 const char *keywordValue
= NULL
;
1278 log_verbose("Number of collator locales returned : %i \n", ucol_countAvailable());
1279 values
= ucol_getKeywordValues("collation", &status
);
1280 for (i
= 0; i
< ucol_countAvailable(); i
++) {
1281 uenum_reset(values
, &status
);
1282 while ((keywordValue
= uenum_next(values
, NULL
, &status
)) != NULL
) {
1283 strcpy(input
, ucol_getAvailable(i
));
1284 if (strcmp(keywordValue
, "standard") != 0) {
1285 strcat(input
, "@collation=");
1286 strcat(input
, keywordValue
);
1289 ucol_getFunctionalEquivalent(output
, 256, "collation", input
, &isAvailable
, &status
);
1290 if (strcmp(input
, output
) == 0) { /* Unique locale, print it out */
1291 log_verbose("%s, \n", output
);
1295 uenum_close(values
);
1298 #endif /* #if !UCONFIG_NO_COLLATION */