2 *******************************************************************************
3 * Copyright (C) 1997-2011, International Business Machines
4 * Corporation and others. All Rights Reserved.
5 *******************************************************************************
6 * Date Name Description
7 * 06/21/00 aliu Creation.
8 *******************************************************************************
14 #include "unicode/utypes.h"
16 #if !UCONFIG_NO_TRANSLITERATION
18 #include "unicode/localpointer.h"
19 #include "unicode/urep.h"
20 #include "unicode/parseerr.h"
21 #include "unicode/uenum.h"
23 /********************************************************************
25 ********************************************************************
29 * \brief C API: Transliterator
31 * <h2> Transliteration </h2>
32 * The data structures and functions described in this header provide
33 * transliteration services. Transliteration services are implemented
34 * as C++ classes. The comments and documentation in this header
35 * assume the reader is familiar with the C++ headers translit.h and
36 * associated documentation.
38 * A significant but incomplete subset of the C++ transliteration
39 * services are available to C code through this header. In order to
40 * access more complex transliteration services, refer to the C++
41 * headers and documentation.
43 * There are two sets of functions for working with transliterator IDs:
45 * An old, deprecated set uses char * IDs, which works for true and pure
46 * identifiers that these APIs were designed for,
47 * for example "Cyrillic-Latin".
48 * It does not work when the ID contains filters ("[:Script=Cyrl:]")
49 * or even a complete set of rules because then the ID string contains more
50 * than just "invariant" characters (see utypes.h).
52 * A new set of functions replaces the old ones and uses UChar * IDs,
53 * paralleling the UnicodeString IDs in the C++ API. (New in ICU 2.8.)
56 /********************************************************************
58 ********************************************************************/
61 * An opaque transliterator for use in C. Open with utrans_openxxx()
62 * and close with utrans_close() when done. Equivalent to the C++ class
63 * Transliterator and its subclasses.
67 typedef void* UTransliterator
;
70 * Direction constant indicating the direction in a transliterator,
71 * e.g., the forward or reverse rules of a RuleBasedTransliterator.
72 * Specified when a transliterator is opened. An "A-B" transliterator
73 * transliterates A to B when operating in the forward direction, and
74 * B to A when operating in the reverse direction.
77 typedef enum UTransDirection
{
80 * UTRANS_FORWARD means from <source> to <target> for a
81 * transliterator with ID <source>-<target>. For a transliterator
82 * opened using a rule, it means forward direction rules, e.g.,
88 * UTRANS_REVERSE means from <target> to <source> for a
89 * transliterator with ID <source>-<target>. For a transliterator
90 * opened using a rule, it means reverse direction rules, e.g.,
98 * Position structure for utrans_transIncremental() incremental
99 * transliteration. This structure defines two substrings of the text
100 * being transliterated. The first region, [contextStart,
101 * contextLimit), defines what characters the transliterator will read
102 * as context. The second region, [start, limit), defines what
103 * characters will actually be transliterated. The second region
104 * should be a subset of the first.
106 * <p>After a transliteration operation, some of the indices in this
107 * structure will be modified. See the field descriptions for
110 * <p>contextStart <= start <= limit <= contextLimit
112 * <p>Note: All index values in this structure must be at code point
113 * boundaries. That is, none of them may occur between two code units
114 * of a surrogate pair. If any index does split a surrogate pair,
115 * results are unspecified.
119 typedef struct UTransPosition
{
122 * Beginning index, inclusive, of the context to be considered for
123 * a transliteration operation. The transliterator will ignore
124 * anything before this index. INPUT/OUTPUT parameter: This parameter
125 * is updated by a transliteration operation to reflect the maximum
126 * amount of antecontext needed by a transliterator.
129 int32_t contextStart
;
132 * Ending index, exclusive, of the context to be considered for a
133 * transliteration operation. The transliterator will ignore
134 * anything at or after this index. INPUT/OUTPUT parameter: This
135 * parameter is updated to reflect changes in the length of the
136 * text, but points to the same logical position in the text.
139 int32_t contextLimit
;
142 * Beginning index, inclusive, of the text to be transliteratd.
143 * INPUT/OUTPUT parameter: This parameter is advanced past
144 * characters that have already been transliterated by a
145 * transliteration operation.
151 * Ending index, exclusive, of the text to be transliteratd.
152 * INPUT/OUTPUT parameter: This parameter is updated to reflect
153 * changes in the length of the text, but points to the same
154 * logical position in the text.
161 /********************************************************************
163 ********************************************************************/
166 * Open a custom transliterator, given a custom rules string
168 * a system transliterator, given its ID.
169 * Any non-NULL result from this function should later be closed with
172 * @param id a valid transliterator ID
173 * @param idLength the length of the ID string, or -1 if NUL-terminated
174 * @param dir the desired direction
175 * @param rules the transliterator rules. See the C++ header rbt.h for
176 * rules syntax. If NULL then a system transliterator matching
177 * the ID is returned.
178 * @param rulesLength the length of the rules, or -1 if the rules
179 * are NUL-terminated.
180 * @param parseError a pointer to a UParseError struct to receive the details
181 * of any parsing errors. This parameter may be NULL if no
182 * parsing error details are desired.
183 * @param pErrorCode a pointer to the UErrorCode
184 * @return a transliterator pointer that may be passed to other
185 * utrans_xxx() functions, or NULL if the open call fails.
188 U_STABLE UTransliterator
* U_EXPORT2
189 utrans_openU(const UChar
*id
,
194 UParseError
*parseError
,
195 UErrorCode
*pErrorCode
);
198 * Open an inverse of an existing transliterator. For this to work,
199 * the inverse must be registered with the system. For example, if
200 * the Transliterator "A-B" is opened, and then its inverse is opened,
201 * the result is the Transliterator "B-A", if such a transliterator is
202 * registered with the system. Otherwise the result is NULL and a
203 * failing UErrorCode is set. Any non-NULL result from this function
204 * should later be closed with utrans_close().
206 * @param trans the transliterator to open the inverse of.
207 * @param status a pointer to the UErrorCode
208 * @return a pointer to a newly-opened transliterator that is the
209 * inverse of trans, or NULL if the open call fails.
212 U_STABLE UTransliterator
* U_EXPORT2
213 utrans_openInverse(const UTransliterator
* trans
,
217 * Create a copy of a transliterator. Any non-NULL result from this
218 * function should later be closed with utrans_close().
220 * @param trans the transliterator to be copied.
221 * @param status a pointer to the UErrorCode
222 * @return a transliterator pointer that may be passed to other
223 * utrans_xxx() functions, or NULL if the clone call fails.
226 U_STABLE UTransliterator
* U_EXPORT2
227 utrans_clone(const UTransliterator
* trans
,
231 * Close a transliterator. Any non-NULL pointer returned by
232 * utrans_openXxx() or utrans_clone() should eventually be closed.
233 * @param trans the transliterator to be closed.
236 U_STABLE
void U_EXPORT2
237 utrans_close(UTransliterator
* trans
);
239 #if U_SHOW_CPLUSPLUS_API
244 * \class LocalUTransliteratorPointer
245 * "Smart pointer" class, closes a UTransliterator via utrans_close().
246 * For most methods see the LocalPointerBase base class.
248 * @see LocalPointerBase
252 U_DEFINE_LOCAL_OPEN_POINTER(LocalUTransliteratorPointer
, UTransliterator
, utrans_close
);
259 * Return the programmatic identifier for this transliterator.
260 * If this identifier is passed to utrans_openU(), it will open
261 * a transliterator equivalent to this one, if the ID has been
264 * @param trans the transliterator to return the ID of.
265 * @param resultLength pointer to an output variable receiving the length
266 * of the ID string; can be NULL
267 * @return the NUL-terminated ID string. This pointer remains
268 * valid until utrans_close() is called on this transliterator.
272 U_STABLE
const UChar
* U_EXPORT2
273 utrans_getUnicodeID(const UTransliterator
*trans
,
274 int32_t *resultLength
);
277 * Register an open transliterator with the system. When
278 * utrans_open() is called with an ID string that is equal to that
279 * returned by utrans_getID(adoptedTrans,...), then
280 * utrans_clone(adoptedTrans,...) is returned.
282 * <p>NOTE: After this call the system owns the adoptedTrans and will
283 * close it. The user must not call utrans_close() on adoptedTrans.
285 * @param adoptedTrans a transliterator, typically the result of
286 * utrans_openRules(), to be registered with the system.
287 * @param status a pointer to the UErrorCode
290 U_STABLE
void U_EXPORT2
291 utrans_register(UTransliterator
* adoptedTrans
,
295 * Unregister a transliterator from the system. After this call the
296 * system will no longer recognize the given ID when passed to
297 * utrans_open(). If the ID is invalid then nothing is done.
299 * @param id an ID to unregister
300 * @param idLength the length of id, or -1 if id is zero-terminated
303 U_STABLE
void U_EXPORT2
304 utrans_unregisterID(const UChar
* id
, int32_t idLength
);
307 * Set the filter used by a transliterator. A filter can be used to
308 * make the transliterator pass certain characters through untouched.
309 * The filter is expressed using a UnicodeSet pattern. If the
310 * filterPattern is NULL or the empty string, then the transliterator
311 * will be reset to use no filter.
313 * @param trans the transliterator
314 * @param filterPattern a pattern string, in the form accepted by
315 * UnicodeSet, specifying which characters to apply the
316 * transliteration to. May be NULL or the empty string to indicate no
318 * @param filterPatternLen the length of filterPattern, or -1 if
319 * filterPattern is zero-terminated
320 * @param status a pointer to the UErrorCode
324 U_STABLE
void U_EXPORT2
325 utrans_setFilter(UTransliterator
* trans
,
326 const UChar
* filterPattern
,
327 int32_t filterPatternLen
,
331 * Return the number of system transliterators.
332 * It is recommended to use utrans_openIDs() instead.
334 * @return the number of system transliterators.
337 U_STABLE
int32_t U_EXPORT2
338 utrans_countAvailableIDs(void);
341 * Return a UEnumeration for the available transliterators.
343 * @param pErrorCode Pointer to the UErrorCode in/out parameter.
344 * @return UEnumeration for the available transliterators.
345 * Close with uenum_close().
349 U_STABLE UEnumeration
* U_EXPORT2
350 utrans_openIDs(UErrorCode
*pErrorCode
);
352 /********************************************************************
353 * Transliteration API
354 ********************************************************************/
357 * Transliterate a segment of a UReplaceable string. The string is
358 * passed in as a UReplaceable pointer rep and a UReplaceableCallbacks
359 * function pointer struct repFunc. Functions in the repFunc struct
360 * will be called in order to modify the rep string.
362 * @param trans the transliterator
363 * @param rep a pointer to the string. This will be passed to the
365 * @param repFunc a set of function pointers that will be used to
366 * modify the string pointed to by rep.
367 * @param start the beginning index, inclusive; <code>0 <= start <=
369 * @param limit pointer to the ending index, exclusive; <code>start <=
370 * limit <= repFunc->length(rep)</code>. Upon return, *limit will
371 * contain the new limit index. The text previously occupying
372 * <code>[start, limit)</code> has been transliterated, possibly to a
373 * string of a different length, at <code>[start,
374 * </code><em>new-limit</em><code>)</code>, where <em>new-limit</em>
375 * is the return value.
376 * @param status a pointer to the UErrorCode
379 U_STABLE
void U_EXPORT2
380 utrans_trans(const UTransliterator
* trans
,
382 UReplaceableCallbacks
* repFunc
,
388 * Transliterate the portion of the UReplaceable text buffer that can
389 * be transliterated unambiguosly. This method is typically called
390 * after new text has been inserted, e.g. as a result of a keyboard
391 * event. The transliterator will try to transliterate characters of
392 * <code>rep</code> between <code>index.cursor</code> and
393 * <code>index.limit</code>. Characters before
394 * <code>index.cursor</code> will not be changed.
396 * <p>Upon return, values in <code>index</code> will be updated.
397 * <code>index.start</code> will be advanced to the first
398 * character that future calls to this method will read.
399 * <code>index.cursor</code> and <code>index.limit</code> will
400 * be adjusted to delimit the range of text that future calls to
401 * this method may change.
403 * <p>Typical usage of this method begins with an initial call
404 * with <code>index.start</code> and <code>index.limit</code>
405 * set to indicate the portion of <code>text</code> to be
406 * transliterated, and <code>index.cursor == index.start</code>.
407 * Thereafter, <code>index</code> can be used without
408 * modification in future calls, provided that all changes to
409 * <code>text</code> are made via this method.
411 * <p>This method assumes that future calls may be made that will
412 * insert new text into the buffer. As a result, it only performs
413 * unambiguous transliterations. After the last call to this method,
414 * there may be untransliterated text that is waiting for more input
415 * to resolve an ambiguity. In order to perform these pending
416 * transliterations, clients should call utrans_trans() with a start
417 * of index.start and a limit of index.end after the last call to this
418 * method has been made.
420 * @param trans the transliterator
421 * @param rep a pointer to the string. This will be passed to the
423 * @param repFunc a set of function pointers that will be used to
424 * modify the string pointed to by rep.
425 * @param pos a struct containing the start and limit indices of the
426 * text to be read and the text to be transliterated
427 * @param status a pointer to the UErrorCode
430 U_STABLE
void U_EXPORT2
431 utrans_transIncremental(const UTransliterator
* trans
,
433 UReplaceableCallbacks
* repFunc
,
438 * Transliterate a segment of a UChar* string. The string is passed
439 * in in a UChar* buffer. The string is modified in place. If the
440 * result is longer than textCapacity, it is truncated. The actual
441 * length of the result is returned in *textLength, if textLength is
442 * non-NULL. *textLength may be greater than textCapacity, but only
443 * textCapacity UChars will be written to *text, including the zero
446 * @param trans the transliterator
447 * @param text a pointer to a buffer containing the text to be
448 * transliterated on input and the result text on output.
449 * @param textLength a pointer to the length of the string in text.
450 * If the length is -1 then the string is assumed to be
451 * zero-terminated. Upon return, the new length is stored in
452 * *textLength. If textLength is NULL then the string is assumed to
453 * be zero-terminated.
454 * @param textCapacity a pointer to the length of the text buffer.
456 * @param start the beginning index, inclusive; <code>0 <= start <=
458 * @param limit pointer to the ending index, exclusive; <code>start <=
459 * limit <= repFunc->length(rep)</code>. Upon return, *limit will
460 * contain the new limit index. The text previously occupying
461 * <code>[start, limit)</code> has been transliterated, possibly to a
462 * string of a different length, at <code>[start,
463 * </code><em>new-limit</em><code>)</code>, where <em>new-limit</em>
464 * is the return value.
465 * @param status a pointer to the UErrorCode
468 U_STABLE
void U_EXPORT2
469 utrans_transUChars(const UTransliterator
* trans
,
472 int32_t textCapacity
,
478 * Transliterate the portion of the UChar* text buffer that can be
479 * transliterated unambiguosly. See utrans_transIncremental(). The
480 * string is passed in in a UChar* buffer. The string is modified in
481 * place. If the result is longer than textCapacity, it is truncated.
482 * The actual length of the result is returned in *textLength, if
483 * textLength is non-NULL. *textLength may be greater than
484 * textCapacity, but only textCapacity UChars will be written to
485 * *text, including the zero terminator. See utrans_transIncremental()
488 * @param trans the transliterator
489 * @param text a pointer to a buffer containing the text to be
490 * transliterated on input and the result text on output.
491 * @param textLength a pointer to the length of the string in text.
492 * If the length is -1 then the string is assumed to be
493 * zero-terminated. Upon return, the new length is stored in
494 * *textLength. If textLength is NULL then the string is assumed to
495 * be zero-terminated.
496 * @param textCapacity the length of the text buffer
497 * @param pos a struct containing the start and limit indices of the
498 * text to be read and the text to be transliterated
499 * @param status a pointer to the UErrorCode
500 * @see utrans_transIncremental
503 U_STABLE
void U_EXPORT2
504 utrans_transIncrementalUChars(const UTransliterator
* trans
,
507 int32_t textCapacity
,
511 /* deprecated API ----------------------------------------------------------- */
513 #ifndef U_HIDE_DEPRECATED_API
515 /* see utrans.h documentation for why these functions are deprecated */
518 * Deprecated, use utrans_openU() instead.
519 * Open a custom transliterator, given a custom rules string
521 * a system transliterator, given its ID.
522 * Any non-NULL result from this function should later be closed with
525 * @param id a valid ID, as returned by utrans_getAvailableID()
526 * @param dir the desired direction
527 * @param rules the transliterator rules. See the C++ header rbt.h
528 * for rules syntax. If NULL then a system transliterator matching
529 * the ID is returned.
530 * @param rulesLength the length of the rules, or -1 if the rules
531 * are zero-terminated.
532 * @param parseError a pointer to a UParseError struct to receive the
533 * details of any parsing errors. This parameter may be NULL if no
534 * parsing error details are desired.
535 * @param status a pointer to the UErrorCode
536 * @return a transliterator pointer that may be passed to other
537 * utrans_xxx() functions, or NULL if the open call fails.
538 * @deprecated ICU 2.8 Use utrans_openU() instead, see utrans.h
540 U_DEPRECATED UTransliterator
* U_EXPORT2
541 utrans_open(const char* id
,
543 const UChar
* rules
, /* may be Null */
544 int32_t rulesLength
, /* -1 if null-terminated */
545 UParseError
* parseError
, /* may be Null */
549 * Deprecated, use utrans_getUnicodeID() instead.
550 * Return the programmatic identifier for this transliterator.
551 * If this identifier is passed to utrans_open(), it will open
552 * a transliterator equivalent to this one, if the ID has been
554 * @param trans the transliterator to return the ID of.
555 * @param buf the buffer in which to receive the ID. This may be
556 * NULL, in which case no characters are copied.
557 * @param bufCapacity the capacity of the buffer. Ignored if buf is
559 * @return the actual length of the ID, not including
560 * zero-termination. This may be greater than bufCapacity.
561 * @deprecated ICU 2.8 Use utrans_getUnicodeID() instead, see utrans.h
563 U_DEPRECATED
int32_t U_EXPORT2
564 utrans_getID(const UTransliterator
* trans
,
566 int32_t bufCapacity
);
569 * Deprecated, use utrans_unregisterID() instead.
570 * Unregister a transliterator from the system. After this call the
571 * system will no longer recognize the given ID when passed to
572 * utrans_open(). If the id is invalid then nothing is done.
574 * @param id a zero-terminated ID
575 * @deprecated ICU 2.8 Use utrans_unregisterID() instead, see utrans.h
577 U_DEPRECATED
void U_EXPORT2
578 utrans_unregister(const char* id
);
581 * Deprecated, use utrans_openIDs() instead.
582 * Return the ID of the index-th system transliterator. The result
583 * is placed in the given buffer. If the given buffer is too small,
584 * the initial substring is copied to buf. The result in buf is
585 * always zero-terminated.
587 * @param index the number of the transliterator to return. Must
588 * satisfy 0 <= index < utrans_countAvailableIDs(). If index is out
589 * of range then it is treated as if it were 0.
590 * @param buf the buffer in which to receive the ID. This may be
591 * NULL, in which case no characters are copied.
592 * @param bufCapacity the capacity of the buffer. Ignored if buf is
594 * @return the actual length of the index-th ID, not including
595 * zero-termination. This may be greater than bufCapacity.
596 * @deprecated ICU 2.8 Use utrans_openIDs() instead, see utrans.h
598 U_DEPRECATED
int32_t U_EXPORT2
599 utrans_getAvailableID(int32_t index
,
601 int32_t bufCapacity
);
603 #endif /* U_HIDE_DEPRECATED_API */
605 #endif /* #if !UCONFIG_NO_TRANSLITERATION */