2 *******************************************************************************
3 * Copyright (C) 1997-2011,2014 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"
22 #include "unicode/uset.h"
24 /********************************************************************
26 ********************************************************************
30 * \brief C API: Transliterator
32 * <h2> Transliteration </h2>
33 * The data structures and functions described in this header provide
34 * transliteration services. Transliteration services are implemented
35 * as C++ classes. The comments and documentation in this header
36 * assume the reader is familiar with the C++ headers translit.h and
37 * associated documentation.
39 * A significant but incomplete subset of the C++ transliteration
40 * services are available to C code through this header. In order to
41 * access more complex transliteration services, refer to the C++
42 * headers and documentation.
44 * There are two sets of functions for working with transliterator IDs:
46 * An old, deprecated set uses char * IDs, which works for true and pure
47 * identifiers that these APIs were designed for,
48 * for example "Cyrillic-Latin".
49 * It does not work when the ID contains filters ("[:Script=Cyrl:]")
50 * or even a complete set of rules because then the ID string contains more
51 * than just "invariant" characters (see utypes.h).
53 * A new set of functions replaces the old ones and uses UChar * IDs,
54 * paralleling the UnicodeString IDs in the C++ API. (New in ICU 2.8.)
57 /********************************************************************
59 ********************************************************************/
62 * An opaque transliterator for use in C. Open with utrans_openxxx()
63 * and close with utrans_close() when done. Equivalent to the C++ class
64 * Transliterator and its subclasses.
68 typedef void* UTransliterator
;
71 * Direction constant indicating the direction in a transliterator,
72 * e.g., the forward or reverse rules of a RuleBasedTransliterator.
73 * Specified when a transliterator is opened. An "A-B" transliterator
74 * transliterates A to B when operating in the forward direction, and
75 * B to A when operating in the reverse direction.
78 typedef enum UTransDirection
{
81 * UTRANS_FORWARD means from <source> to <target> for a
82 * transliterator with ID <source>-<target>. For a transliterator
83 * opened using a rule, it means forward direction rules, e.g.,
89 * UTRANS_REVERSE means from <target> to <source> for a
90 * transliterator with ID <source>-<target>. For a transliterator
91 * opened using a rule, it means reverse direction rules, e.g.,
99 * Position structure for utrans_transIncremental() incremental
100 * transliteration. This structure defines two substrings of the text
101 * being transliterated. The first region, [contextStart,
102 * contextLimit), defines what characters the transliterator will read
103 * as context. The second region, [start, limit), defines what
104 * characters will actually be transliterated. The second region
105 * should be a subset of the first.
107 * <p>After a transliteration operation, some of the indices in this
108 * structure will be modified. See the field descriptions for
111 * <p>contextStart <= start <= limit <= contextLimit
113 * <p>Note: All index values in this structure must be at code point
114 * boundaries. That is, none of them may occur between two code units
115 * of a surrogate pair. If any index does split a surrogate pair,
116 * results are unspecified.
120 typedef struct UTransPosition
{
123 * Beginning index, inclusive, of the context to be considered for
124 * a transliteration operation. The transliterator will ignore
125 * anything before this index. INPUT/OUTPUT parameter: This parameter
126 * is updated by a transliteration operation to reflect the maximum
127 * amount of antecontext needed by a transliterator.
130 int32_t contextStart
;
133 * Ending index, exclusive, of the context to be considered for a
134 * transliteration operation. The transliterator will ignore
135 * anything at or after this index. INPUT/OUTPUT parameter: This
136 * parameter is updated to reflect changes in the length of the
137 * text, but points to the same logical position in the text.
140 int32_t contextLimit
;
143 * Beginning index, inclusive, of the text to be transliteratd.
144 * INPUT/OUTPUT parameter: This parameter is advanced past
145 * characters that have already been transliterated by a
146 * transliteration operation.
152 * Ending index, exclusive, of the text to be transliteratd.
153 * INPUT/OUTPUT parameter: This parameter is updated to reflect
154 * changes in the length of the text, but points to the same
155 * logical position in the text.
162 /********************************************************************
164 ********************************************************************/
167 * Open a custom transliterator, given a custom rules string
169 * a system transliterator, given its ID.
170 * Any non-NULL result from this function should later be closed with
173 * @param id a valid transliterator ID
174 * @param idLength the length of the ID string, or -1 if NUL-terminated
175 * @param dir the desired direction
176 * @param rules the transliterator rules. See the C++ header rbt.h for
177 * rules syntax. If NULL then a system transliterator matching
178 * the ID is returned.
179 * @param rulesLength the length of the rules, or -1 if the rules
180 * are NUL-terminated.
181 * @param parseError a pointer to a UParseError struct to receive the details
182 * of any parsing errors. This parameter may be NULL if no
183 * parsing error details are desired.
184 * @param pErrorCode a pointer to the UErrorCode
185 * @return a transliterator pointer that may be passed to other
186 * utrans_xxx() functions, or NULL if the open call fails.
189 U_STABLE UTransliterator
* U_EXPORT2
190 utrans_openU(const UChar
*id
,
195 UParseError
*parseError
,
196 UErrorCode
*pErrorCode
);
199 * Open an inverse of an existing transliterator. For this to work,
200 * the inverse must be registered with the system. For example, if
201 * the Transliterator "A-B" is opened, and then its inverse is opened,
202 * the result is the Transliterator "B-A", if such a transliterator is
203 * registered with the system. Otherwise the result is NULL and a
204 * failing UErrorCode is set. Any non-NULL result from this function
205 * should later be closed with utrans_close().
207 * @param trans the transliterator to open the inverse of.
208 * @param status a pointer to the UErrorCode
209 * @return a pointer to a newly-opened transliterator that is the
210 * inverse of trans, or NULL if the open call fails.
213 U_STABLE UTransliterator
* U_EXPORT2
214 utrans_openInverse(const UTransliterator
* trans
,
218 * Create a copy of a transliterator. Any non-NULL result from this
219 * function should later be closed with utrans_close().
221 * @param trans the transliterator to be copied.
222 * @param status a pointer to the UErrorCode
223 * @return a transliterator pointer that may be passed to other
224 * utrans_xxx() functions, or NULL if the clone call fails.
227 U_STABLE UTransliterator
* U_EXPORT2
228 utrans_clone(const UTransliterator
* trans
,
232 * Close a transliterator. Any non-NULL pointer returned by
233 * utrans_openXxx() or utrans_clone() should eventually be closed.
234 * @param trans the transliterator to be closed.
237 U_STABLE
void U_EXPORT2
238 utrans_close(UTransliterator
* trans
);
240 #if U_SHOW_CPLUSPLUS_API
245 * \class LocalUTransliteratorPointer
246 * "Smart pointer" class, closes a UTransliterator via utrans_close().
247 * For most methods see the LocalPointerBase base class.
249 * @see LocalPointerBase
253 U_DEFINE_LOCAL_OPEN_POINTER(LocalUTransliteratorPointer
, UTransliterator
, utrans_close
);
260 * Return the programmatic identifier for this transliterator.
261 * If this identifier is passed to utrans_openU(), it will open
262 * a transliterator equivalent to this one, if the ID has been
265 * @param trans the transliterator to return the ID of.
266 * @param resultLength pointer to an output variable receiving the length
267 * of the ID string; can be NULL
268 * @return the NUL-terminated ID string. This pointer remains
269 * valid until utrans_close() is called on this transliterator.
273 U_STABLE
const UChar
* U_EXPORT2
274 utrans_getUnicodeID(const UTransliterator
*trans
,
275 int32_t *resultLength
);
278 * Register an open transliterator with the system. When
279 * utrans_open() is called with an ID string that is equal to that
280 * returned by utrans_getID(adoptedTrans,...), then
281 * utrans_clone(adoptedTrans,...) is returned.
283 * <p>NOTE: After this call the system owns the adoptedTrans and will
284 * close it. The user must not call utrans_close() on adoptedTrans.
286 * @param adoptedTrans a transliterator, typically the result of
287 * utrans_openRules(), to be registered with the system.
288 * @param status a pointer to the UErrorCode
291 U_STABLE
void U_EXPORT2
292 utrans_register(UTransliterator
* adoptedTrans
,
296 * Unregister a transliterator from the system. After this call the
297 * system will no longer recognize the given ID when passed to
298 * utrans_open(). If the ID is invalid then nothing is done.
300 * @param id an ID to unregister
301 * @param idLength the length of id, or -1 if id is zero-terminated
304 U_STABLE
void U_EXPORT2
305 utrans_unregisterID(const UChar
* id
, int32_t idLength
);
308 * Set the filter used by a transliterator. A filter can be used to
309 * make the transliterator pass certain characters through untouched.
310 * The filter is expressed using a UnicodeSet pattern. If the
311 * filterPattern is NULL or the empty string, then the transliterator
312 * will be reset to use no filter.
314 * @param trans the transliterator
315 * @param filterPattern a pattern string, in the form accepted by
316 * UnicodeSet, specifying which characters to apply the
317 * transliteration to. May be NULL or the empty string to indicate no
319 * @param filterPatternLen the length of filterPattern, or -1 if
320 * filterPattern is zero-terminated
321 * @param status a pointer to the UErrorCode
325 U_STABLE
void U_EXPORT2
326 utrans_setFilter(UTransliterator
* trans
,
327 const UChar
* filterPattern
,
328 int32_t filterPatternLen
,
332 * Return the number of system transliterators.
333 * It is recommended to use utrans_openIDs() instead.
335 * @return the number of system transliterators.
338 U_STABLE
int32_t U_EXPORT2
339 utrans_countAvailableIDs(void);
342 * Return a UEnumeration for the available transliterators.
344 * @param pErrorCode Pointer to the UErrorCode in/out parameter.
345 * @return UEnumeration for the available transliterators.
346 * Close with uenum_close().
350 U_STABLE UEnumeration
* U_EXPORT2
351 utrans_openIDs(UErrorCode
*pErrorCode
);
353 /********************************************************************
354 * Transliteration API
355 ********************************************************************/
358 * Transliterate a segment of a UReplaceable string. The string is
359 * passed in as a UReplaceable pointer rep and a UReplaceableCallbacks
360 * function pointer struct repFunc. Functions in the repFunc struct
361 * will be called in order to modify the rep string.
363 * @param trans the transliterator
364 * @param rep a pointer to the string. This will be passed to the
366 * @param repFunc a set of function pointers that will be used to
367 * modify the string pointed to by rep.
368 * @param start the beginning index, inclusive; <code>0 <= start <=
370 * @param limit pointer to the ending index, exclusive; <code>start <=
371 * limit <= repFunc->length(rep)</code>. Upon return, *limit will
372 * contain the new limit index. The text previously occupying
373 * <code>[start, limit)</code> has been transliterated, possibly to a
374 * string of a different length, at <code>[start,
375 * </code><em>new-limit</em><code>)</code>, where <em>new-limit</em>
376 * is the return value.
377 * @param status a pointer to the UErrorCode
380 U_STABLE
void U_EXPORT2
381 utrans_trans(const UTransliterator
* trans
,
383 UReplaceableCallbacks
* repFunc
,
389 * Transliterate the portion of the UReplaceable text buffer that can
390 * be transliterated unambiguosly. This method is typically called
391 * after new text has been inserted, e.g. as a result of a keyboard
392 * event. The transliterator will try to transliterate characters of
393 * <code>rep</code> between <code>index.cursor</code> and
394 * <code>index.limit</code>. Characters before
395 * <code>index.cursor</code> will not be changed.
397 * <p>Upon return, values in <code>index</code> will be updated.
398 * <code>index.start</code> will be advanced to the first
399 * character that future calls to this method will read.
400 * <code>index.cursor</code> and <code>index.limit</code> will
401 * be adjusted to delimit the range of text that future calls to
402 * this method may change.
404 * <p>Typical usage of this method begins with an initial call
405 * with <code>index.start</code> and <code>index.limit</code>
406 * set to indicate the portion of <code>text</code> to be
407 * transliterated, and <code>index.cursor == index.start</code>.
408 * Thereafter, <code>index</code> can be used without
409 * modification in future calls, provided that all changes to
410 * <code>text</code> are made via this method.
412 * <p>This method assumes that future calls may be made that will
413 * insert new text into the buffer. As a result, it only performs
414 * unambiguous transliterations. After the last call to this method,
415 * there may be untransliterated text that is waiting for more input
416 * to resolve an ambiguity. In order to perform these pending
417 * transliterations, clients should call utrans_trans() with a start
418 * of index.start and a limit of index.end after the last call to this
419 * method has been made.
421 * @param trans the transliterator
422 * @param rep a pointer to the string. This will be passed to the
424 * @param repFunc a set of function pointers that will be used to
425 * modify the string pointed to by rep.
426 * @param pos a struct containing the start and limit indices of the
427 * text to be read and the text to be transliterated
428 * @param status a pointer to the UErrorCode
431 U_STABLE
void U_EXPORT2
432 utrans_transIncremental(const UTransliterator
* trans
,
434 UReplaceableCallbacks
* repFunc
,
439 * Transliterate a segment of a UChar* string. The string is passed
440 * in in a UChar* buffer. The string is modified in place. If the
441 * result is longer than textCapacity, it is truncated. The actual
442 * length of the result is returned in *textLength, if textLength is
443 * non-NULL. *textLength may be greater than textCapacity, but only
444 * textCapacity UChars will be written to *text, including the zero
447 * @param trans the transliterator
448 * @param text a pointer to a buffer containing the text to be
449 * transliterated on input and the result text on output.
450 * @param textLength a pointer to the length of the string in text.
451 * If the length is -1 then the string is assumed to be
452 * zero-terminated. Upon return, the new length is stored in
453 * *textLength. If textLength is NULL then the string is assumed to
454 * be zero-terminated.
455 * @param textCapacity a pointer to the length of the text buffer.
457 * @param start the beginning index, inclusive; <code>0 <= start <=
459 * @param limit pointer to the ending index, exclusive; <code>start <=
460 * limit <= repFunc->length(rep)</code>. Upon return, *limit will
461 * contain the new limit index. The text previously occupying
462 * <code>[start, limit)</code> has been transliterated, possibly to a
463 * string of a different length, at <code>[start,
464 * </code><em>new-limit</em><code>)</code>, where <em>new-limit</em>
465 * is the return value.
466 * @param status a pointer to the UErrorCode
469 U_STABLE
void U_EXPORT2
470 utrans_transUChars(const UTransliterator
* trans
,
473 int32_t textCapacity
,
479 * Transliterate the portion of the UChar* text buffer that can be
480 * transliterated unambiguosly. See utrans_transIncremental(). The
481 * string is passed in in a UChar* buffer. The string is modified in
482 * place. If the result is longer than textCapacity, it is truncated.
483 * The actual length of the result is returned in *textLength, if
484 * textLength is non-NULL. *textLength may be greater than
485 * textCapacity, but only textCapacity UChars will be written to
486 * *text, including the zero terminator. See utrans_transIncremental()
489 * @param trans the transliterator
490 * @param text a pointer to a buffer containing the text to be
491 * transliterated on input and the result text on output.
492 * @param textLength a pointer to the length of the string in text.
493 * If the length is -1 then the string is assumed to be
494 * zero-terminated. Upon return, the new length is stored in
495 * *textLength. If textLength is NULL then the string is assumed to
496 * be zero-terminated.
497 * @param textCapacity the length of the text buffer
498 * @param pos a struct containing the start and limit indices of the
499 * text to be read and the text to be transliterated
500 * @param status a pointer to the UErrorCode
501 * @see utrans_transIncremental
504 U_STABLE
void U_EXPORT2
505 utrans_transIncrementalUChars(const UTransliterator
* trans
,
508 int32_t textCapacity
,
512 #ifndef U_HIDE_DRAFT_API
514 * Create a rule string that can be passed to utrans_openU to recreate this
517 * @param trans The transliterator
518 * @param escapeUnprintable if TRUE then convert unprintable characters to their
519 * hex escape representations, \\uxxxx or \\Uxxxxxxxx.
520 * Unprintable characters are those other than
521 * U+000A, U+0020..U+007E.
522 * @param result A pointer to a buffer to receive the rules.
523 * @param resultLength The maximum size of result.
524 * @param status A pointer to the UErrorCode. In case of error status, the
525 * contents of result are undefined.
526 * @return int32_t The length of the rule string (may be greater than resultLength,
527 * in which case an error is returned).
530 U_DRAFT
int32_t U_EXPORT2
531 utrans_toRules( const UTransliterator
* trans
,
532 UBool escapeUnprintable
,
533 UChar
* result
, int32_t resultLength
,
537 * Returns the set of all characters that may be modified in the input text by
538 * this UTransliterator, optionally ignoring the transliterator's current filter.
539 * @param trans The transliterator.
540 * @param ignoreFilter If FALSE, the returned set incorporates the
541 * UTransliterator's current filter; if the filter is changed,
542 * the return value of this function will change. If TRUE, the
543 * returned set ignores the effect of the UTransliterator's
545 * @param fillIn Pointer to a USet object to receive the modifiable characters
546 * set. Previous contents of fillIn are lost. <em>If fillIn is
547 * NULL, then a new USet is created and returned. The caller
548 * owns the result and must dispose of it by calling uset_close.</em>
549 * @param status A pointer to the UErrorCode.
550 * @return USet* Either fillIn, or if fillIn is NULL, a pointer to a
551 * newly-allocated USet that the user must close. In case of
552 * error, NULL is returned.
555 U_DRAFT USet
* U_EXPORT2
556 utrans_getSourceSet(const UTransliterator
* trans
,
560 #endif /* U_HIDE_DRAFT_API */
562 /* deprecated API ----------------------------------------------------------- */
564 #ifndef U_HIDE_DEPRECATED_API
566 /* see utrans.h documentation for why these functions are deprecated */
569 * Deprecated, use utrans_openU() instead.
570 * Open a custom transliterator, given a custom rules string
572 * a system transliterator, given its ID.
573 * Any non-NULL result from this function should later be closed with
576 * @param id a valid ID, as returned by utrans_getAvailableID()
577 * @param dir the desired direction
578 * @param rules the transliterator rules. See the C++ header rbt.h
579 * for rules syntax. If NULL then a system transliterator matching
580 * the ID is returned.
581 * @param rulesLength the length of the rules, or -1 if the rules
582 * are zero-terminated.
583 * @param parseError a pointer to a UParseError struct to receive the
584 * details of any parsing errors. This parameter may be NULL if no
585 * parsing error details are desired.
586 * @param status a pointer to the UErrorCode
587 * @return a transliterator pointer that may be passed to other
588 * utrans_xxx() functions, or NULL if the open call fails.
589 * @deprecated ICU 2.8 Use utrans_openU() instead, see utrans.h
591 U_DEPRECATED UTransliterator
* U_EXPORT2
592 utrans_open(const char* id
,
594 const UChar
* rules
, /* may be Null */
595 int32_t rulesLength
, /* -1 if null-terminated */
596 UParseError
* parseError
, /* may be Null */
600 * Deprecated, use utrans_getUnicodeID() instead.
601 * Return the programmatic identifier for this transliterator.
602 * If this identifier is passed to utrans_open(), it will open
603 * a transliterator equivalent to this one, if the ID has been
605 * @param trans the transliterator to return the ID of.
606 * @param buf the buffer in which to receive the ID. This may be
607 * NULL, in which case no characters are copied.
608 * @param bufCapacity the capacity of the buffer. Ignored if buf is
610 * @return the actual length of the ID, not including
611 * zero-termination. This may be greater than bufCapacity.
612 * @deprecated ICU 2.8 Use utrans_getUnicodeID() instead, see utrans.h
614 U_DEPRECATED
int32_t U_EXPORT2
615 utrans_getID(const UTransliterator
* trans
,
617 int32_t bufCapacity
);
620 * Deprecated, use utrans_unregisterID() instead.
621 * Unregister a transliterator from the system. After this call the
622 * system will no longer recognize the given ID when passed to
623 * utrans_open(). If the id is invalid then nothing is done.
625 * @param id a zero-terminated ID
626 * @deprecated ICU 2.8 Use utrans_unregisterID() instead, see utrans.h
628 U_DEPRECATED
void U_EXPORT2
629 utrans_unregister(const char* id
);
632 * Deprecated, use utrans_openIDs() instead.
633 * Return the ID of the index-th system transliterator. The result
634 * is placed in the given buffer. If the given buffer is too small,
635 * the initial substring is copied to buf. The result in buf is
636 * always zero-terminated.
638 * @param index the number of the transliterator to return. Must
639 * satisfy 0 <= index < utrans_countAvailableIDs(). If index is out
640 * of range then it is treated as if it were 0.
641 * @param buf the buffer in which to receive the ID. This may be
642 * NULL, in which case no characters are copied.
643 * @param bufCapacity the capacity of the buffer. Ignored if buf is
645 * @return the actual length of the index-th ID, not including
646 * zero-termination. This may be greater than bufCapacity.
647 * @deprecated ICU 2.8 Use utrans_openIDs() instead, see utrans.h
649 U_DEPRECATED
int32_t U_EXPORT2
650 utrans_getAvailableID(int32_t index
,
652 int32_t bufCapacity
);
654 #endif /* U_HIDE_DEPRECATED_API */
656 #endif /* #if !UCONFIG_NO_TRANSLITERATION */