Line data Source code
1 : // © 2016 and later: Unicode, Inc. and others.
2 : // License & terms of use: http://www.unicode.org/copyright.html
3 : /*
4 : *******************************************************************************
5 : * Copyright (C) 2004-2016, International Business Machines
6 : * Corporation and others. All Rights Reserved.
7 : *******************************************************************************
8 : * file name: ucol_sit.cpp
9 : * encoding: UTF-8
10 : * tab size: 8 (not used)
11 : * indentation:4
12 : *
13 : * Modification history
14 : * Date Name Comments
15 : * 03/12/2004 weiv Creation
16 : */
17 :
18 : #include "unicode/ustring.h"
19 : #include "unicode/udata.h"
20 : #include "unicode/utf16.h"
21 : #include "utracimp.h"
22 : #include "ucol_imp.h"
23 : #include "cmemory.h"
24 : #include "cstring.h"
25 : #include "uresimp.h"
26 : #include "unicode/coll.h"
27 :
28 : #ifdef UCOL_TRACE_SIT
29 : # include <stdio.h>
30 : #endif
31 :
32 : #if !UCONFIG_NO_COLLATION
33 :
34 : #include "unicode/tblcoll.h"
35 :
36 : enum OptionsList {
37 : UCOL_SIT_LANGUAGE = 0,
38 : UCOL_SIT_SCRIPT = 1,
39 : UCOL_SIT_REGION = 2,
40 : UCOL_SIT_VARIANT = 3,
41 : UCOL_SIT_KEYWORD = 4,
42 : UCOL_SIT_PROVIDER = 5,
43 : UCOL_SIT_LOCELEMENT_MAX = UCOL_SIT_PROVIDER, /* the last element that's part of LocElements */
44 :
45 : UCOL_SIT_BCP47,
46 : UCOL_SIT_STRENGTH,
47 : UCOL_SIT_CASE_LEVEL,
48 : UCOL_SIT_CASE_FIRST,
49 : UCOL_SIT_NUMERIC_COLLATION,
50 : UCOL_SIT_ALTERNATE_HANDLING,
51 : UCOL_SIT_NORMALIZATION_MODE,
52 : UCOL_SIT_FRENCH_COLLATION,
53 : UCOL_SIT_HIRAGANA_QUATERNARY,
54 : UCOL_SIT_VARIABLE_TOP,
55 : UCOL_SIT_VARIABLE_TOP_VALUE,
56 : UCOL_SIT_ITEMS_COUNT
57 : };
58 :
59 : /* option starters chars. */
60 : static const char alternateHArg = 'A';
61 : static const char variableTopValArg = 'B';
62 : static const char caseFirstArg = 'C';
63 : static const char numericCollArg = 'D';
64 : static const char caseLevelArg = 'E';
65 : static const char frenchCollArg = 'F';
66 : static const char hiraganaQArg = 'H';
67 : static const char keywordArg = 'K';
68 : static const char languageArg = 'L';
69 : static const char normArg = 'N';
70 : static const char providerArg = 'P';
71 : static const char regionArg = 'R';
72 : static const char strengthArg = 'S';
73 : static const char variableTopArg = 'T';
74 : static const char variantArg = 'V';
75 : static const char RFC3066Arg = 'X';
76 : static const char scriptArg = 'Z';
77 :
78 : static const char collationKeyword[] = "@collation=";
79 : static const char providerKeyword[] = "@sp=";
80 :
81 :
82 : static const int32_t locElementCount = UCOL_SIT_LOCELEMENT_MAX+1;
83 : static const int32_t locElementCapacity = 32;
84 : static const int32_t loc3066Capacity = 256;
85 : static const int32_t locProviderCapacity = 10;
86 : static const int32_t internalBufferSize = 512;
87 :
88 : /* structure containing specification of a collator. Initialized
89 : * from a short string. Also used to construct a short string from a
90 : * collator instance
91 : */
92 : struct CollatorSpec {
93 : char locElements[locElementCount][locElementCapacity];
94 : char locale[loc3066Capacity];
95 : char provider[locProviderCapacity];
96 : UColAttributeValue options[UCOL_ATTRIBUTE_COUNT];
97 : uint32_t variableTopValue;
98 : UChar variableTopString[locElementCapacity];
99 : int32_t variableTopStringLen;
100 : UBool variableTopSet;
101 : struct {
102 : const char *start;
103 : int32_t len;
104 : } entries[UCOL_SIT_ITEMS_COUNT];
105 : };
106 :
107 :
108 : /* structure for converting between character attribute
109 : * representation and real collation attribute value.
110 : */
111 : struct AttributeConversion {
112 : char letter;
113 : UColAttributeValue value;
114 : };
115 :
116 : static const AttributeConversion conversions[12] = {
117 : { '1', UCOL_PRIMARY },
118 : { '2', UCOL_SECONDARY },
119 : { '3', UCOL_TERTIARY },
120 : { '4', UCOL_QUATERNARY },
121 : { 'D', UCOL_DEFAULT },
122 : { 'I', UCOL_IDENTICAL },
123 : { 'L', UCOL_LOWER_FIRST },
124 : { 'N', UCOL_NON_IGNORABLE },
125 : { 'O', UCOL_ON },
126 : { 'S', UCOL_SHIFTED },
127 : { 'U', UCOL_UPPER_FIRST },
128 : { 'X', UCOL_OFF }
129 : };
130 :
131 :
132 : static UColAttributeValue
133 0 : ucol_sit_letterToAttributeValue(char letter, UErrorCode *status) {
134 0 : uint32_t i = 0;
135 0 : for(i = 0; i < UPRV_LENGTHOF(conversions); i++) {
136 0 : if(conversions[i].letter == letter) {
137 0 : return conversions[i].value;
138 : }
139 : }
140 0 : *status = U_ILLEGAL_ARGUMENT_ERROR;
141 : #ifdef UCOL_TRACE_SIT
142 : fprintf(stderr, "%s:%d: unknown letter %c: %s\n", __FILE__, __LINE__, letter, u_errorName(*status));
143 : #endif
144 0 : return UCOL_DEFAULT;
145 : }
146 :
147 : /* function prototype for functions used to parse a short string */
148 : U_CDECL_BEGIN
149 : typedef const char* U_CALLCONV
150 : ActionFunction(CollatorSpec *spec, uint32_t value1, const char* string,
151 : UErrorCode *status);
152 : U_CDECL_END
153 :
154 : U_CDECL_BEGIN
155 : static const char* U_CALLCONV
156 0 : _processLocaleElement(CollatorSpec *spec, uint32_t value, const char* string,
157 : UErrorCode *status)
158 : {
159 0 : int32_t len = 0;
160 0 : do {
161 0 : if(value == UCOL_SIT_LANGUAGE || value == UCOL_SIT_KEYWORD || value == UCOL_SIT_PROVIDER) {
162 0 : spec->locElements[value][len++] = uprv_tolower(*string);
163 : } else {
164 0 : spec->locElements[value][len++] = *string;
165 : }
166 0 : } while(*(++string) != '_' && *string && len < locElementCapacity);
167 0 : if(len >= locElementCapacity) {
168 0 : *status = U_BUFFER_OVERFLOW_ERROR;
169 0 : return string;
170 : }
171 : // don't skip the underscore at the end
172 0 : return string;
173 : }
174 : U_CDECL_END
175 :
176 : U_CDECL_BEGIN
177 : static const char* U_CALLCONV
178 0 : _processRFC3066Locale(CollatorSpec *spec, uint32_t, const char* string,
179 : UErrorCode *status)
180 : {
181 0 : char terminator = *string;
182 0 : string++;
183 0 : const char *end = uprv_strchr(string+1, terminator);
184 0 : if(end == NULL || end - string >= loc3066Capacity) {
185 0 : *status = U_BUFFER_OVERFLOW_ERROR;
186 0 : return string;
187 : } else {
188 0 : uprv_strncpy(spec->locale, string, end-string);
189 0 : return end+1;
190 : }
191 : }
192 :
193 : U_CDECL_END
194 :
195 : U_CDECL_BEGIN
196 : static const char* U_CALLCONV
197 0 : _processCollatorOption(CollatorSpec *spec, uint32_t option, const char* string,
198 : UErrorCode *status)
199 : {
200 0 : spec->options[option] = ucol_sit_letterToAttributeValue(*string, status);
201 0 : if((*(++string) != '_' && *string) || U_FAILURE(*status)) {
202 : #ifdef UCOL_TRACE_SIT
203 : fprintf(stderr, "%s:%d: unknown collator option at '%s': %s\n", __FILE__, __LINE__, string, u_errorName(*status));
204 : #endif
205 0 : *status = U_ILLEGAL_ARGUMENT_ERROR;
206 : }
207 0 : return string;
208 : }
209 : U_CDECL_END
210 :
211 :
212 : static UChar
213 0 : readHexCodeUnit(const char **string, UErrorCode *status)
214 : {
215 0 : UChar result = 0;
216 0 : int32_t value = 0;
217 : char c;
218 0 : int32_t noDigits = 0;
219 0 : while((c = **string) != 0 && noDigits < 4) {
220 0 : if( c >= '0' && c <= '9') {
221 0 : value = c - '0';
222 0 : } else if ( c >= 'a' && c <= 'f') {
223 0 : value = c - 'a' + 10;
224 0 : } else if ( c >= 'A' && c <= 'F') {
225 0 : value = c - 'A' + 10;
226 : } else {
227 0 : *status = U_ILLEGAL_ARGUMENT_ERROR;
228 : #ifdef UCOL_TRACE_SIT
229 : fprintf(stderr, "%s:%d: Bad hex char at '%s': %s\n", __FILE__, __LINE__, *string, u_errorName(*status));
230 : #endif
231 0 : return 0;
232 : }
233 0 : result = (result << 4) | (UChar)value;
234 0 : noDigits++;
235 0 : (*string)++;
236 : }
237 : // if the string was terminated before we read 4 digits, set an error
238 0 : if(noDigits < 4) {
239 0 : *status = U_ILLEGAL_ARGUMENT_ERROR;
240 : #ifdef UCOL_TRACE_SIT
241 : fprintf(stderr, "%s:%d: Short (only %d digits, wanted 4) at '%s': %s\n", __FILE__, __LINE__, noDigits,*string, u_errorName(*status));
242 : #endif
243 : }
244 0 : return result;
245 : }
246 :
247 : U_CDECL_BEGIN
248 : static const char* U_CALLCONV
249 0 : _processVariableTop(CollatorSpec *spec, uint32_t value1, const char* string, UErrorCode *status)
250 : {
251 : // get four digits
252 0 : int32_t i = 0;
253 0 : if(!value1) {
254 0 : while(U_SUCCESS(*status) && i < locElementCapacity && *string != 0 && *string != '_') {
255 0 : spec->variableTopString[i++] = readHexCodeUnit(&string, status);
256 : }
257 0 : spec->variableTopStringLen = i;
258 0 : if(i == locElementCapacity && *string != 0 && *string != '_') {
259 0 : *status = U_BUFFER_OVERFLOW_ERROR;
260 : }
261 : } else {
262 0 : spec->variableTopValue = readHexCodeUnit(&string, status);
263 : }
264 0 : if(U_SUCCESS(*status)) {
265 0 : spec->variableTopSet = TRUE;
266 : }
267 0 : return string;
268 : }
269 : U_CDECL_END
270 :
271 :
272 : /* Table for parsing short strings */
273 : struct ShortStringOptions {
274 : char optionStart;
275 : ActionFunction *action;
276 : uint32_t attr;
277 : };
278 :
279 : static const ShortStringOptions options[UCOL_SIT_ITEMS_COUNT] =
280 : {
281 : /* 10 ALTERNATE_HANDLING */ {alternateHArg, _processCollatorOption, UCOL_ALTERNATE_HANDLING }, // alternate N, S, D
282 : /* 15 VARIABLE_TOP_VALUE */ {variableTopValArg, _processVariableTop, 1 },
283 : /* 08 CASE_FIRST */ {caseFirstArg, _processCollatorOption, UCOL_CASE_FIRST }, // case first L, U, X, D
284 : /* 09 NUMERIC_COLLATION */ {numericCollArg, _processCollatorOption, UCOL_NUMERIC_COLLATION }, // codan O, X, D
285 : /* 07 CASE_LEVEL */ {caseLevelArg, _processCollatorOption, UCOL_CASE_LEVEL }, // case level O, X, D
286 : /* 12 FRENCH_COLLATION */ {frenchCollArg, _processCollatorOption, UCOL_FRENCH_COLLATION }, // french O, X, D
287 : /* 13 HIRAGANA_QUATERNARY] */ {hiraganaQArg, _processCollatorOption, UCOL_HIRAGANA_QUATERNARY_MODE }, // hiragana O, X, D
288 : /* 04 KEYWORD */ {keywordArg, _processLocaleElement, UCOL_SIT_KEYWORD }, // keyword
289 : /* 00 LANGUAGE */ {languageArg, _processLocaleElement, UCOL_SIT_LANGUAGE }, // language
290 : /* 11 NORMALIZATION_MODE */ {normArg, _processCollatorOption, UCOL_NORMALIZATION_MODE }, // norm O, X, D
291 : /* 02 REGION */ {regionArg, _processLocaleElement, UCOL_SIT_REGION }, // region
292 : /* 06 STRENGTH */ {strengthArg, _processCollatorOption, UCOL_STRENGTH }, // strength 1, 2, 3, 4, I, D
293 : /* 14 VARIABLE_TOP */ {variableTopArg, _processVariableTop, 0 },
294 : /* 03 VARIANT */ {variantArg, _processLocaleElement, UCOL_SIT_VARIANT }, // variant
295 : /* 05 RFC3066BIS */ {RFC3066Arg, _processRFC3066Locale, 0 }, // rfc3066bis locale name
296 : /* 01 SCRIPT */ {scriptArg, _processLocaleElement, UCOL_SIT_SCRIPT }, // script
297 : /* PROVIDER */ {providerArg, _processLocaleElement, UCOL_SIT_PROVIDER }
298 : };
299 :
300 :
301 : static
302 0 : const char* ucol_sit_readOption(const char *start, CollatorSpec *spec,
303 : UErrorCode *status)
304 : {
305 0 : int32_t i = 0;
306 :
307 0 : for(i = 0; i < UCOL_SIT_ITEMS_COUNT; i++) {
308 0 : if(*start == options[i].optionStart) {
309 0 : spec->entries[i].start = start;
310 0 : const char* end = options[i].action(spec, options[i].attr, start+1, status);
311 0 : spec->entries[i].len = (int32_t)(end - start);
312 0 : return end;
313 : }
314 : }
315 0 : *status = U_ILLEGAL_ARGUMENT_ERROR;
316 : #ifdef UCOL_TRACE_SIT
317 : fprintf(stderr, "%s:%d: Unknown option at '%s': %s\n", __FILE__, __LINE__, start, u_errorName(*status));
318 : #endif
319 0 : return start;
320 : }
321 :
322 : static
323 0 : void ucol_sit_initCollatorSpecs(CollatorSpec *spec)
324 : {
325 : // reset everything
326 0 : uprv_memset(spec, 0, sizeof(CollatorSpec));
327 : // set collation options to default
328 0 : int32_t i = 0;
329 0 : for(i = 0; i < UCOL_ATTRIBUTE_COUNT; i++) {
330 0 : spec->options[i] = UCOL_DEFAULT;
331 : }
332 0 : }
333 :
334 : static const char*
335 0 : ucol_sit_readSpecs(CollatorSpec *s, const char *string,
336 : UParseError *parseError, UErrorCode *status)
337 : {
338 0 : const char *definition = string;
339 0 : while(U_SUCCESS(*status) && *string) {
340 0 : string = ucol_sit_readOption(string, s, status);
341 : // advance over '_'
342 0 : while(*string && *string == '_') {
343 0 : string++;
344 : }
345 : }
346 0 : if(U_FAILURE(*status)) {
347 0 : parseError->offset = (int32_t)(string - definition);
348 : }
349 0 : return string;
350 : }
351 :
352 : static
353 0 : int32_t ucol_sit_dumpSpecs(CollatorSpec *s, char *destination, int32_t capacity, UErrorCode *status)
354 : {
355 0 : int32_t i = 0, j = 0;
356 0 : int32_t len = 0;
357 : char optName;
358 0 : if(U_SUCCESS(*status)) {
359 0 : for(i = 0; i < UCOL_SIT_ITEMS_COUNT; i++) {
360 0 : if(s->entries[i].start) {
361 0 : if(len) {
362 0 : if(len < capacity) {
363 0 : uprv_strcat(destination, "_");
364 : }
365 0 : len++;
366 : }
367 0 : optName = *(s->entries[i].start);
368 0 : if(optName == languageArg || optName == regionArg || optName == variantArg || optName == keywordArg) {
369 0 : for(j = 0; j < s->entries[i].len; j++) {
370 0 : if(len + j < capacity) {
371 0 : destination[len+j] = uprv_toupper(*(s->entries[i].start+j));
372 : }
373 : }
374 0 : len += s->entries[i].len;
375 : } else {
376 0 : len += s->entries[i].len;
377 0 : if(len < capacity) {
378 0 : uprv_strncat(destination,s->entries[i].start, s->entries[i].len);
379 : }
380 : }
381 : }
382 : }
383 0 : return len;
384 : } else {
385 0 : return 0;
386 : }
387 : }
388 :
389 : static void
390 0 : ucol_sit_calculateWholeLocale(CollatorSpec *s) {
391 : // put the locale together, unless we have a done
392 : // locale
393 0 : if(s->locale[0] == 0) {
394 : // first the language
395 0 : uprv_strcat(s->locale, s->locElements[UCOL_SIT_LANGUAGE]);
396 : // then the script, if present
397 0 : if(*(s->locElements[UCOL_SIT_SCRIPT])) {
398 0 : uprv_strcat(s->locale, "_");
399 0 : uprv_strcat(s->locale, s->locElements[UCOL_SIT_SCRIPT]);
400 : }
401 : // then the region, if present
402 0 : if(*(s->locElements[UCOL_SIT_REGION])) {
403 0 : uprv_strcat(s->locale, "_");
404 0 : uprv_strcat(s->locale, s->locElements[UCOL_SIT_REGION]);
405 0 : } else if(*(s->locElements[UCOL_SIT_VARIANT])) { // if there is a variant, we need an underscore
406 0 : uprv_strcat(s->locale, "_");
407 : }
408 : // add variant, if there
409 0 : if(*(s->locElements[UCOL_SIT_VARIANT])) {
410 0 : uprv_strcat(s->locale, "_");
411 0 : uprv_strcat(s->locale, s->locElements[UCOL_SIT_VARIANT]);
412 : }
413 :
414 : // if there is a collation keyword, add that too
415 0 : if(*(s->locElements[UCOL_SIT_KEYWORD])) {
416 0 : uprv_strcat(s->locale, collationKeyword);
417 0 : uprv_strcat(s->locale, s->locElements[UCOL_SIT_KEYWORD]);
418 : }
419 :
420 : // if there is a provider keyword, add that too
421 0 : if(*(s->locElements[UCOL_SIT_PROVIDER])) {
422 0 : uprv_strcat(s->locale, providerKeyword);
423 0 : uprv_strcat(s->locale, s->locElements[UCOL_SIT_PROVIDER]);
424 : }
425 : }
426 0 : }
427 :
428 :
429 : U_CAPI void U_EXPORT2
430 0 : ucol_prepareShortStringOpen( const char *definition,
431 : UBool,
432 : UParseError *parseError,
433 : UErrorCode *status)
434 : {
435 0 : if(U_FAILURE(*status)) return;
436 :
437 : UParseError internalParseError;
438 :
439 0 : if(!parseError) {
440 0 : parseError = &internalParseError;
441 : }
442 0 : parseError->line = 0;
443 0 : parseError->offset = 0;
444 0 : parseError->preContext[0] = 0;
445 0 : parseError->postContext[0] = 0;
446 :
447 :
448 : // first we want to pick stuff out of short string.
449 : // we'll end up with an UCA version, locale and a bunch of
450 : // settings
451 :
452 : // analyse the string in order to get everything we need.
453 : CollatorSpec s;
454 0 : ucol_sit_initCollatorSpecs(&s);
455 0 : ucol_sit_readSpecs(&s, definition, parseError, status);
456 0 : ucol_sit_calculateWholeLocale(&s);
457 :
458 : char buffer[internalBufferSize];
459 0 : uprv_memset(buffer, 0, internalBufferSize);
460 0 : uloc_canonicalize(s.locale, buffer, internalBufferSize, status);
461 :
462 0 : UResourceBundle *b = ures_open(U_ICUDATA_COLL, buffer, status);
463 : /* we try to find stuff from keyword */
464 0 : UResourceBundle *collations = ures_getByKey(b, "collations", NULL, status);
465 0 : UResourceBundle *collElem = NULL;
466 : char keyBuffer[256];
467 : // if there is a keyword, we pick it up and try to get elements
468 0 : if(!uloc_getKeywordValue(buffer, "collation", keyBuffer, 256, status)) {
469 : // no keyword. we try to find the default setting, which will give us the keyword value
470 0 : UResourceBundle *defaultColl = ures_getByKeyWithFallback(collations, "default", NULL, status);
471 0 : if(U_SUCCESS(*status)) {
472 0 : int32_t defaultKeyLen = 0;
473 0 : const UChar *defaultKey = ures_getString(defaultColl, &defaultKeyLen, status);
474 0 : u_UCharsToChars(defaultKey, keyBuffer, defaultKeyLen);
475 0 : keyBuffer[defaultKeyLen] = 0;
476 : } else {
477 0 : *status = U_INTERNAL_PROGRAM_ERROR;
478 0 : return;
479 : }
480 0 : ures_close(defaultColl);
481 : }
482 0 : collElem = ures_getByKeyWithFallback(collations, keyBuffer, collElem, status);
483 0 : ures_close(collElem);
484 0 : ures_close(collations);
485 0 : ures_close(b);
486 : }
487 :
488 :
489 : U_CAPI UCollator* U_EXPORT2
490 0 : ucol_openFromShortString( const char *definition,
491 : UBool forceDefaults,
492 : UParseError *parseError,
493 : UErrorCode *status)
494 : {
495 : UTRACE_ENTRY_OC(UTRACE_UCOL_OPEN_FROM_SHORT_STRING);
496 : UTRACE_DATA1(UTRACE_INFO, "short string = \"%s\"", definition);
497 :
498 0 : if(U_FAILURE(*status)) return 0;
499 :
500 : UParseError internalParseError;
501 :
502 0 : if(!parseError) {
503 0 : parseError = &internalParseError;
504 : }
505 0 : parseError->line = 0;
506 0 : parseError->offset = 0;
507 0 : parseError->preContext[0] = 0;
508 0 : parseError->postContext[0] = 0;
509 :
510 :
511 : // first we want to pick stuff out of short string.
512 : // we'll end up with an UCA version, locale and a bunch of
513 : // settings
514 :
515 : // analyse the string in order to get everything we need.
516 0 : const char *string = definition;
517 : CollatorSpec s;
518 0 : ucol_sit_initCollatorSpecs(&s);
519 0 : string = ucol_sit_readSpecs(&s, definition, parseError, status);
520 0 : ucol_sit_calculateWholeLocale(&s);
521 :
522 : char buffer[internalBufferSize];
523 0 : uprv_memset(buffer, 0, internalBufferSize);
524 0 : uloc_canonicalize(s.locale, buffer, internalBufferSize, status);
525 :
526 0 : UCollator *result = ucol_open(buffer, status);
527 0 : int32_t i = 0;
528 :
529 0 : for(i = 0; i < UCOL_ATTRIBUTE_COUNT; i++) {
530 0 : if(s.options[i] != UCOL_DEFAULT) {
531 0 : if(forceDefaults || ucol_getAttribute(result, (UColAttribute)i, status) != s.options[i]) {
532 0 : ucol_setAttribute(result, (UColAttribute)i, s.options[i], status);
533 : }
534 :
535 0 : if(U_FAILURE(*status)) {
536 0 : parseError->offset = (int32_t)(string - definition);
537 0 : ucol_close(result);
538 0 : return NULL;
539 : }
540 :
541 : }
542 : }
543 0 : if(s.variableTopSet) {
544 0 : if(s.variableTopString[0]) {
545 0 : ucol_setVariableTop(result, s.variableTopString, s.variableTopStringLen, status);
546 : } else { // we set by value, using 'B'
547 0 : ucol_restoreVariableTop(result, s.variableTopValue, status);
548 : }
549 : }
550 :
551 :
552 0 : if(U_FAILURE(*status)) { // here it can only be a bogus value
553 0 : ucol_close(result);
554 0 : result = NULL;
555 : }
556 :
557 : UTRACE_EXIT_PTR_STATUS(result, *status);
558 0 : return result;
559 : }
560 :
561 :
562 : U_CAPI int32_t U_EXPORT2
563 0 : ucol_getShortDefinitionString(const UCollator *coll,
564 : const char *locale,
565 : char *dst,
566 : int32_t capacity,
567 : UErrorCode *status)
568 : {
569 0 : if(U_FAILURE(*status)) return 0;
570 0 : if(coll == NULL) {
571 0 : *status = U_ILLEGAL_ARGUMENT_ERROR;
572 0 : return 0;
573 : }
574 0 : return ((icu::Collator*)coll)->internalGetShortDefinitionString(locale,dst,capacity,*status);
575 : }
576 :
577 : U_CAPI int32_t U_EXPORT2
578 0 : ucol_normalizeShortDefinitionString(const char *definition,
579 : char *destination,
580 : int32_t capacity,
581 : UParseError *parseError,
582 : UErrorCode *status)
583 : {
584 :
585 0 : if(U_FAILURE(*status)) {
586 0 : return 0;
587 : }
588 :
589 0 : if(destination) {
590 0 : uprv_memset(destination, 0, capacity*sizeof(char));
591 : }
592 :
593 : UParseError pe;
594 0 : if(!parseError) {
595 0 : parseError = &pe;
596 : }
597 :
598 : // validate
599 : CollatorSpec s;
600 0 : ucol_sit_initCollatorSpecs(&s);
601 0 : ucol_sit_readSpecs(&s, definition, parseError, status);
602 0 : return ucol_sit_dumpSpecs(&s, destination, capacity, status);
603 : }
604 :
605 : /**
606 : * Get a set containing the contractions defined by the collator. The set includes
607 : * both the UCA contractions and the contractions defined by the collator
608 : * @param coll collator
609 : * @param conts the set to hold the result
610 : * @param status to hold the error code
611 : * @return the size of the contraction set
612 : */
613 : U_CAPI int32_t U_EXPORT2
614 0 : ucol_getContractions( const UCollator *coll,
615 : USet *contractions,
616 : UErrorCode *status)
617 : {
618 0 : ucol_getContractionsAndExpansions(coll, contractions, NULL, FALSE, status);
619 0 : return uset_getItemCount(contractions);
620 : }
621 :
622 : /**
623 : * Get a set containing the expansions defined by the collator. The set includes
624 : * both the UCA expansions and the expansions defined by the tailoring
625 : * @param coll collator
626 : * @param conts the set to hold the result
627 : * @param addPrefixes add the prefix contextual elements to contractions
628 : * @param status to hold the error code
629 : *
630 : * @draft ICU 3.4
631 : */
632 : U_CAPI void U_EXPORT2
633 0 : ucol_getContractionsAndExpansions( const UCollator *coll,
634 : USet *contractions,
635 : USet *expansions,
636 : UBool addPrefixes,
637 : UErrorCode *status)
638 : {
639 0 : if(U_FAILURE(*status)) {
640 0 : return;
641 : }
642 0 : if(coll == NULL) {
643 0 : *status = U_ILLEGAL_ARGUMENT_ERROR;
644 0 : return;
645 : }
646 0 : const icu::RuleBasedCollator *rbc = icu::RuleBasedCollator::rbcFromUCollator(coll);
647 0 : if(rbc == NULL) {
648 0 : *status = U_UNSUPPORTED_ERROR;
649 0 : return;
650 : }
651 0 : rbc->internalGetContractionsAndExpansions(
652 : icu::UnicodeSet::fromUSet(contractions),
653 : icu::UnicodeSet::fromUSet(expansions),
654 0 : addPrefixes, *status);
655 : }
656 : #endif
|
