]> git.saurik.com Git - apple/javascriptcore.git/blame_incremental - icu/unicode/ucnv.h
JavaScriptCore-7601.1.46.3.tar.gz
[apple/javascriptcore.git] / icu / unicode / ucnv.h
... / ...
CommitLineData
1/*
2**********************************************************************
3* Copyright (C) 1999-2010, International Business Machines
4* Corporation and others. All Rights Reserved.
5**********************************************************************
6 * ucnv.h:
7 * External APIs for the ICU's codeset conversion library
8 * Bertrand A. Damiba
9 *
10 * Modification History:
11 *
12 * Date Name Description
13 * 04/04/99 helena Fixed internal header inclusion.
14 * 05/11/00 helena Added setFallback and usesFallback APIs.
15 * 06/29/2000 helena Major rewrite of the callback APIs.
16 * 12/07/2000 srl Update of documentation
17 */
18
19/**
20 * \file
21 * \brief C API: Character conversion
22 *
23 * <h2>Character Conversion C API</h2>
24 *
25 * <p>This API is used to convert codepage or character encoded data to and
26 * from UTF-16. You can open a converter with {@link ucnv_open() }. With that
27 * converter, you can get its properties, set options, convert your data and
28 * close the converter.</p>
29 *
30 * <p>Since many software programs recogize different converter names for
31 * different types of converters, there are other functions in this API to
32 * iterate over the converter aliases. The functions {@link ucnv_getAvailableName() },
33 * {@link ucnv_getAlias() } and {@link ucnv_getStandardName() } are some of the
34 * more frequently used alias functions to get this information.</p>
35 *
36 * <p>When a converter encounters an illegal, irregular, invalid or unmappable character
37 * its default behavior is to use a substitution character to replace the
38 * bad byte sequence. This behavior can be changed by using {@link ucnv_setFromUCallBack() }
39 * or {@link ucnv_setToUCallBack() } on the converter. The header ucnv_err.h defines
40 * many other callback actions that can be used instead of a character substitution.</p>
41 *
42 * <p>More information about this API can be found in our
43 * <a href="http://icu-project.org/userguide/conversion.html">User's
44 * Guide</a>.</p>
45 */
46
47#ifndef UCNV_H
48#define UCNV_H
49
50#include "unicode/ucnv_err.h"
51#include "unicode/uenum.h"
52#include "unicode/localpointer.h"
53
54#ifndef __USET_H__
55
56/**
57 * USet is the C API type for Unicode sets.
58 * It is forward-declared here to avoid including the header file if related
59 * conversion APIs are not used.
60 * See unicode/uset.h
61 *
62 * @see ucnv_getUnicodeSet
63 * @stable ICU 2.6
64 */
65struct USet;
66/** @stable ICU 2.6 */
67typedef struct USet USet;
68
69#endif
70
71#if !UCONFIG_NO_CONVERSION
72
73U_CDECL_BEGIN
74
75/** Maximum length of a converter name including the terminating NULL @stable ICU 2.0 */
76#define UCNV_MAX_CONVERTER_NAME_LENGTH 60
77/** Maximum length of a converter name including path and terminating NULL @stable ICU 2.0 */
78#define UCNV_MAX_FULL_FILE_NAME_LENGTH (600+UCNV_MAX_CONVERTER_NAME_LENGTH)
79
80/** Shift in for EBDCDIC_STATEFUL and iso2022 states @stable ICU 2.0 */
81#define UCNV_SI 0x0F
82/** Shift out for EBDCDIC_STATEFUL and iso2022 states @stable ICU 2.0 */
83#define UCNV_SO 0x0E
84
85/**
86 * Enum for specifying basic types of converters
87 * @see ucnv_getType
88 * @stable ICU 2.0
89 */
90typedef enum {
91 UCNV_UNSUPPORTED_CONVERTER = -1,
92 UCNV_SBCS = 0,
93 UCNV_DBCS = 1,
94 UCNV_MBCS = 2,
95 UCNV_LATIN_1 = 3,
96 UCNV_UTF8 = 4,
97 UCNV_UTF16_BigEndian = 5,
98 UCNV_UTF16_LittleEndian = 6,
99 UCNV_UTF32_BigEndian = 7,
100 UCNV_UTF32_LittleEndian = 8,
101 UCNV_EBCDIC_STATEFUL = 9,
102 UCNV_ISO_2022 = 10,
103
104 UCNV_LMBCS_1 = 11,
105 UCNV_LMBCS_2,
106 UCNV_LMBCS_3,
107 UCNV_LMBCS_4,
108 UCNV_LMBCS_5,
109 UCNV_LMBCS_6,
110 UCNV_LMBCS_8,
111 UCNV_LMBCS_11,
112 UCNV_LMBCS_16,
113 UCNV_LMBCS_17,
114 UCNV_LMBCS_18,
115 UCNV_LMBCS_19,
116 UCNV_LMBCS_LAST = UCNV_LMBCS_19,
117 UCNV_HZ,
118 UCNV_SCSU,
119 UCNV_ISCII,
120 UCNV_US_ASCII,
121 UCNV_UTF7,
122 UCNV_BOCU1,
123 UCNV_UTF16,
124 UCNV_UTF32,
125 UCNV_CESU8,
126 UCNV_IMAP_MAILBOX,
127
128 /* Number of converter types for which we have conversion routines. */
129 UCNV_NUMBER_OF_SUPPORTED_CONVERTER_TYPES
130
131} UConverterType;
132
133/**
134 * Enum for specifying which platform a converter ID refers to.
135 * The use of platform/CCSID is not recommended. See ucnv_openCCSID().
136 *
137 * @see ucnv_getPlatform
138 * @see ucnv_openCCSID
139 * @see ucnv_getCCSID
140 * @stable ICU 2.0
141 */
142typedef enum {
143 UCNV_UNKNOWN = -1,
144 UCNV_IBM = 0
145} UConverterPlatform;
146
147/**
148 * Function pointer for error callback in the codepage to unicode direction.
149 * Called when an error has occured in conversion to unicode, or on open/close of the callback (see reason).
150 * @param context Pointer to the callback's private data
151 * @param args Information about the conversion in progress
152 * @param codeUnits Points to 'length' bytes of the concerned codepage sequence
153 * @param length Size (in bytes) of the concerned codepage sequence
154 * @param reason Defines the reason the callback was invoked
155 * @param pErrorCode ICU error code in/out parameter.
156 * For converter callback functions, set to a conversion error
157 * before the call, and the callback may reset it to U_ZERO_ERROR.
158 * @see ucnv_setToUCallBack
159 * @see UConverterToUnicodeArgs
160 * @stable ICU 2.0
161 */
162typedef void (U_EXPORT2 *UConverterToUCallback) (
163 const void* context,
164 UConverterToUnicodeArgs *args,
165 const char *codeUnits,
166 int32_t length,
167 UConverterCallbackReason reason,
168 UErrorCode *pErrorCode);
169
170/**
171 * Function pointer for error callback in the unicode to codepage direction.
172 * Called when an error has occured in conversion from unicode, or on open/close of the callback (see reason).
173 * @param context Pointer to the callback's private data
174 * @param args Information about the conversion in progress
175 * @param codeUnits Points to 'length' UChars of the concerned Unicode sequence
176 * @param length Size (in bytes) of the concerned codepage sequence
177 * @param codePoint Single UChar32 (UTF-32) containing the concerend Unicode codepoint.
178 * @param reason Defines the reason the callback was invoked
179 * @param pErrorCode ICU error code in/out parameter.
180 * For converter callback functions, set to a conversion error
181 * before the call, and the callback may reset it to U_ZERO_ERROR.
182 * @see ucnv_setFromUCallBack
183 * @stable ICU 2.0
184 */
185typedef void (U_EXPORT2 *UConverterFromUCallback) (
186 const void* context,
187 UConverterFromUnicodeArgs *args,
188 const UChar* codeUnits,
189 int32_t length,
190 UChar32 codePoint,
191 UConverterCallbackReason reason,
192 UErrorCode *pErrorCode);
193
194U_CDECL_END
195
196/**
197 * Character that separates converter names from options and options from each other.
198 * @see ucnv_open
199 * @stable ICU 2.0
200 */
201#define UCNV_OPTION_SEP_CHAR ','
202
203/**
204 * String version of UCNV_OPTION_SEP_CHAR.
205 * @see ucnv_open
206 * @stable ICU 2.0
207 */
208#define UCNV_OPTION_SEP_STRING ","
209
210/**
211 * Character that separates a converter option from its value.
212 * @see ucnv_open
213 * @stable ICU 2.0
214 */
215#define UCNV_VALUE_SEP_CHAR '='
216
217/**
218 * String version of UCNV_VALUE_SEP_CHAR.
219 * @see ucnv_open
220 * @stable ICU 2.0
221 */
222#define UCNV_VALUE_SEP_STRING "="
223
224/**
225 * Converter option for specifying a locale.
226 * For example, ucnv_open("SCSU,locale=ja", &errorCode);
227 * See convrtrs.txt.
228 *
229 * @see ucnv_open
230 * @stable ICU 2.0
231 */
232#define UCNV_LOCALE_OPTION_STRING ",locale="
233
234/**
235 * Converter option for specifying a version selector (0..9) for some converters.
236 * For example,
237 * \code
238 * ucnv_open("UTF-7,version=1", &errorCode);
239 * \endcode
240 * See convrtrs.txt.
241 *
242 * @see ucnv_open
243 * @stable ICU 2.4
244 */
245#define UCNV_VERSION_OPTION_STRING ",version="
246
247/**
248 * Converter option for EBCDIC SBCS or mixed-SBCS/DBCS (stateful) codepages.
249 * Swaps Unicode mappings for EBCDIC LF and NL codes, as used on
250 * S/390 (z/OS) Unix System Services (Open Edition).
251 * For example, ucnv_open("ibm-1047,swaplfnl", &errorCode);
252 * See convrtrs.txt.
253 *
254 * @see ucnv_open
255 * @stable ICU 2.4
256 */
257#define UCNV_SWAP_LFNL_OPTION_STRING ",swaplfnl"
258
259/**
260 * Do a fuzzy compare of two converter/alias names.
261 * The comparison is case-insensitive, ignores leading zeroes if they are not
262 * followed by further digits, and ignores all but letters and digits.
263 * Thus the strings "UTF-8", "utf_8", "u*T@f08" and "Utf 8" are exactly equivalent.
264 * See section 1.4, Charset Alias Matching in Unicode Technical Standard #22
265 * at http://www.unicode.org/reports/tr22/
266 *
267 * @param name1 a converter name or alias, zero-terminated
268 * @param name2 a converter name or alias, zero-terminated
269 * @return 0 if the names match, or a negative value if the name1
270 * lexically precedes name2, or a positive value if the name1
271 * lexically follows name2.
272 * @stable ICU 2.0
273 */
274U_STABLE int U_EXPORT2
275ucnv_compareNames(const char *name1, const char *name2);
276
277
278/**
279 * Creates a UConverter object with the name of a coded character set specified as a C string.
280 * The actual name will be resolved with the alias file
281 * using a case-insensitive string comparison that ignores
282 * leading zeroes and all non-alphanumeric characters.
283 * E.g., the names "UTF8", "utf-8", "u*T@f08" and "Utf 8" are all equivalent.
284 * (See also ucnv_compareNames().)
285 * If <code>NULL</code> is passed for the converter name, it will create one with the
286 * getDefaultName return value.
287 *
288 * <p>A converter name for ICU 1.5 and above may contain options
289 * like a locale specification to control the specific behavior of
290 * the newly instantiated converter.
291 * The meaning of the options depends on the particular converter.
292 * If an option is not defined for or recognized by a given converter, then it is ignored.</p>
293 *
294 * <p>Options are appended to the converter name string, with a
295 * <code>UCNV_OPTION_SEP_CHAR</code> between the name and the first option and
296 * also between adjacent options.</p>
297 *
298 * <p>If the alias is ambiguous, then the preferred converter is used
299 * and the status is set to U_AMBIGUOUS_ALIAS_WARNING.</p>
300 *
301 * <p>The conversion behavior and names can vary between platforms. ICU may
302 * convert some characters differently from other platforms. Details on this topic
303 * are in the <a href="http://icu-project.org/userguide/conversion.html">User's
304 * Guide</a>. Aliases starting with a "cp" prefix have no specific meaning
305 * other than its an alias starting with the letters "cp". Please do not
306 * associate any meaning to these aliases.</p>
307 *
308 * @param converterName Name of the coded character set table.
309 * This may have options appended to the string.
310 * IANA alias character set names, IBM CCSIDs starting with "ibm-",
311 * Windows codepage numbers starting with "windows-" are frequently
312 * used for this parameter. See ucnv_getAvailableName and
313 * ucnv_getAlias for a complete list that is available.
314 * If this parameter is NULL, the default converter will be used.
315 * @param err outgoing error status <TT>U_MEMORY_ALLOCATION_ERROR, U_FILE_ACCESS_ERROR</TT>
316 * @return the created Unicode converter object, or <TT>NULL</TT> if an error occured
317 * @see ucnv_openU
318 * @see ucnv_openCCSID
319 * @see ucnv_getAvailableName
320 * @see ucnv_getAlias
321 * @see ucnv_getDefaultName
322 * @see ucnv_close
323 * @see ucnv_compareNames
324 * @stable ICU 2.0
325 */
326U_STABLE UConverter* U_EXPORT2
327ucnv_open(const char *converterName, UErrorCode *err);
328
329
330/**
331 * Creates a Unicode converter with the names specified as unicode string.
332 * The name should be limited to the ASCII-7 alphanumerics range.
333 * The actual name will be resolved with the alias file
334 * using a case-insensitive string comparison that ignores
335 * leading zeroes and all non-alphanumeric characters.
336 * E.g., the names "UTF8", "utf-8", "u*T@f08" and "Utf 8" are all equivalent.
337 * (See also ucnv_compareNames().)
338 * If <TT>NULL</TT> is passed for the converter name, it will create
339 * one with the ucnv_getDefaultName() return value.
340 * If the alias is ambiguous, then the preferred converter is used
341 * and the status is set to U_AMBIGUOUS_ALIAS_WARNING.
342 *
343 * <p>See ucnv_open for the complete details</p>
344 * @param name Name of the UConverter table in a zero terminated
345 * Unicode string
346 * @param err outgoing error status <TT>U_MEMORY_ALLOCATION_ERROR,
347 * U_FILE_ACCESS_ERROR</TT>
348 * @return the created Unicode converter object, or <TT>NULL</TT> if an
349 * error occured
350 * @see ucnv_open
351 * @see ucnv_openCCSID
352 * @see ucnv_close
353 * @see ucnv_compareNames
354 * @stable ICU 2.0
355 */
356U_STABLE UConverter* U_EXPORT2
357ucnv_openU(const UChar *name,
358 UErrorCode *err);
359
360/**
361 * Creates a UConverter object from a CCSID number and platform pair.
362 * Note that the usefulness of this function is limited to platforms with numeric
363 * encoding IDs. Only IBM and Microsoft platforms use numeric (16-bit) identifiers for
364 * encodings.
365 *
366 * In addition, IBM CCSIDs and Unicode conversion tables are not 1:1 related.
367 * For many IBM CCSIDs there are multiple (up to six) Unicode conversion tables, and
368 * for some Unicode conversion tables there are multiple CCSIDs.
369 * Some "alternate" Unicode conversion tables are provided by the
370 * IBM CDRA conversion table registry.
371 * The most prominent example of a systematic modification of conversion tables that is
372 * not provided in the form of conversion table files in the repository is
373 * that S/390 Unix System Services swaps the codes for Line Feed and New Line in all
374 * EBCDIC codepages, which requires such a swap in the Unicode conversion tables as well.
375 *
376 * Only IBM default conversion tables are accessible with ucnv_openCCSID().
377 * ucnv_getCCSID() will return the same CCSID for all conversion tables that are associated
378 * with that CCSID.
379 *
380 * Currently, the only "platform" supported in the ICU converter API is UCNV_IBM.
381 *
382 * In summary, the use of CCSIDs and the associated API functions is not recommended.
383 *
384 * In order to open a converter with the default IBM CDRA Unicode conversion table,
385 * you can use this function or use the prefix "ibm-":
386 * \code
387 * char name[20];
388 * sprintf(name, "ibm-%hu", ccsid);
389 * cnv=ucnv_open(name, &errorCode);
390 * \endcode
391 *
392 * In order to open a converter with the IBM S/390 Unix System Services variant
393 * of a Unicode/EBCDIC conversion table,
394 * you can use the prefix "ibm-" together with the option string UCNV_SWAP_LFNL_OPTION_STRING:
395 * \code
396 * char name[20];
397 * sprintf(name, "ibm-%hu" UCNV_SWAP_LFNL_OPTION_STRING, ccsid);
398 * cnv=ucnv_open(name, &errorCode);
399 * \endcode
400 *
401 * In order to open a converter from a Microsoft codepage number, use the prefix "cp":
402 * \code
403 * char name[20];
404 * sprintf(name, "cp%hu", codepageID);
405 * cnv=ucnv_open(name, &errorCode);
406 * \endcode
407 *
408 * If the alias is ambiguous, then the preferred converter is used
409 * and the status is set to U_AMBIGUOUS_ALIAS_WARNING.
410 *
411 * @param codepage codepage number to create
412 * @param platform the platform in which the codepage number exists
413 * @param err error status <TT>U_MEMORY_ALLOCATION_ERROR, U_FILE_ACCESS_ERROR</TT>
414 * @return the created Unicode converter object, or <TT>NULL</TT> if an error
415 * occured.
416 * @see ucnv_open
417 * @see ucnv_openU
418 * @see ucnv_close
419 * @see ucnv_getCCSID
420 * @see ucnv_getPlatform
421 * @see UConverterPlatform
422 * @stable ICU 2.0
423 */
424U_STABLE UConverter* U_EXPORT2
425ucnv_openCCSID(int32_t codepage,
426 UConverterPlatform platform,
427 UErrorCode * err);
428
429/**
430 * <p>Creates a UConverter object specified from a packageName and a converterName.</p>
431 *
432 * <p>The packageName and converterName must point to an ICU udata object, as defined by
433 * <code> udata_open( packageName, "cnv", converterName, err) </code> or equivalent.
434 * Typically, packageName will refer to a (.dat) file, or to a package registered with
435 * udata_setAppData(). Using a full file or directory pathname for packageName is deprecated.</p>
436 *
437 * <p>The name will NOT be looked up in the alias mechanism, nor will the converter be
438 * stored in the converter cache or the alias table. The only way to open further converters
439 * is call this function multiple times, or use the ucnv_safeClone() function to clone a
440 * 'master' converter.</p>
441 *
442 * <p>A future version of ICU may add alias table lookups and/or caching
443 * to this function.</p>
444 *
445 * <p>Example Use:
446 * <code>cnv = ucnv_openPackage("myapp", "myconverter", &err);</code>
447 * </p>
448 *
449 * @param packageName name of the package (equivalent to 'path' in udata_open() call)
450 * @param converterName name of the data item to be used, without suffix.
451 * @param err outgoing error status <TT>U_MEMORY_ALLOCATION_ERROR, U_FILE_ACCESS_ERROR</TT>
452 * @return the created Unicode converter object, or <TT>NULL</TT> if an error occured
453 * @see udata_open
454 * @see ucnv_open
455 * @see ucnv_safeClone
456 * @see ucnv_close
457 * @stable ICU 2.2
458 */
459U_STABLE UConverter* U_EXPORT2
460ucnv_openPackage(const char *packageName, const char *converterName, UErrorCode *err);
461
462/**
463 * Thread safe converter cloning operation.
464 * For most efficient operation, pass in a stackBuffer (and a *pBufferSize)
465 * with at least U_CNV_SAFECLONE_BUFFERSIZE bytes of space.
466 * If the buffer size is sufficient, then the clone will use the stack buffer;
467 * otherwise, it will be allocated, and *pBufferSize will indicate
468 * the actual size. (This should not occur with U_CNV_SAFECLONE_BUFFERSIZE.)
469 *
470 * You must ucnv_close() the clone in any case.
471 *
472 * If *pBufferSize==0, (regardless of whether stackBuffer==NULL or not)
473 * then *pBufferSize will be changed to a sufficient size
474 * for cloning this converter,
475 * without actually cloning the converter ("pure pre-flighting").
476 *
477 * If *pBufferSize is greater than zero but not large enough for a stack-based
478 * clone, then the converter is cloned using newly allocated memory
479 * and *pBufferSize is changed to the necessary size.
480 *
481 * If the converter clone fits into the stack buffer but the stack buffer is not
482 * sufficiently aligned for the clone, then the clone will use an
483 * adjusted pointer and use an accordingly smaller buffer size.
484 *
485 * @param cnv converter to be cloned
486 * @param stackBuffer user allocated space for the new clone. If NULL new memory will be allocated.
487 * If buffer is not large enough, new memory will be allocated.
488 * Clients can use the U_CNV_SAFECLONE_BUFFERSIZE. This will probably be enough to avoid memory allocations.
489 * @param pBufferSize pointer to size of allocated space. pBufferSize must not be NULL.
490 * @param status to indicate whether the operation went on smoothly or there were errors
491 * An informational status value, U_SAFECLONE_ALLOCATED_WARNING,
492 * is used if any allocations were necessary.
493 * However, it is better to check if *pBufferSize grew for checking for
494 * allocations because warning codes can be overridden by subsequent
495 * function calls.
496 * @return pointer to the new clone
497 * @stable ICU 2.0
498 */
499U_STABLE UConverter * U_EXPORT2
500ucnv_safeClone(const UConverter *cnv,
501 void *stackBuffer,
502 int32_t *pBufferSize,
503 UErrorCode *status);
504
505/**
506 * \def U_CNV_SAFECLONE_BUFFERSIZE
507 * Definition of a buffer size that is designed to be large enough for
508 * converters to be cloned with ucnv_safeClone().
509 * @stable ICU 2.0
510 */
511#define U_CNV_SAFECLONE_BUFFERSIZE 1024
512
513/**
514 * Deletes the unicode converter and releases resources associated
515 * with just this instance.
516 * Does not free up shared converter tables.
517 *
518 * @param converter the converter object to be deleted
519 * @see ucnv_open
520 * @see ucnv_openU
521 * @see ucnv_openCCSID
522 * @stable ICU 2.0
523 */
524U_STABLE void U_EXPORT2
525ucnv_close(UConverter * converter);
526
527#if U_SHOW_CPLUSPLUS_API
528
529U_NAMESPACE_BEGIN
530
531/**
532 * \class LocalUConverterPointer
533 * "Smart pointer" class, closes a UConverter via ucnv_close().
534 * For most methods see the LocalPointerBase base class.
535 *
536 * @see LocalPointerBase
537 * @see LocalPointer
538 * @stable ICU 4.4
539 */
540U_DEFINE_LOCAL_OPEN_POINTER(LocalUConverterPointer, UConverter, ucnv_close);
541
542U_NAMESPACE_END
543
544#endif
545
546/**
547 * Fills in the output parameter, subChars, with the substitution characters
548 * as multiple bytes.
549 * If ucnv_setSubstString() set a Unicode string because the converter is
550 * stateful, then subChars will be an empty string.
551 *
552 * @param converter the Unicode converter
553 * @param subChars the subsitution characters
554 * @param len on input the capacity of subChars, on output the number
555 * of bytes copied to it
556 * @param err the outgoing error status code.
557 * If the substitution character array is too small, an
558 * <TT>U_INDEX_OUTOFBOUNDS_ERROR</TT> will be returned.
559 * @see ucnv_setSubstString
560 * @see ucnv_setSubstChars
561 * @stable ICU 2.0
562 */
563U_STABLE void U_EXPORT2
564ucnv_getSubstChars(const UConverter *converter,
565 char *subChars,
566 int8_t *len,
567 UErrorCode *err);
568
569/**
570 * Sets the substitution chars when converting from unicode to a codepage. The
571 * substitution is specified as a string of 1-4 bytes, and may contain
572 * <TT>NULL</TT> bytes.
573 * The subChars must represent a single character. The caller needs to know the
574 * byte sequence of a valid character in the converter's charset.
575 * For some converters, for example some ISO 2022 variants, only single-byte
576 * substitution characters may be supported.
577 * The newer ucnv_setSubstString() function relaxes these limitations.
578 *
579 * @param converter the Unicode converter
580 * @param subChars the substitution character byte sequence we want set
581 * @param len the number of bytes in subChars
582 * @param err the error status code. <TT>U_INDEX_OUTOFBOUNDS_ERROR </TT> if
583 * len is bigger than the maximum number of bytes allowed in subchars
584 * @see ucnv_setSubstString
585 * @see ucnv_getSubstChars
586 * @stable ICU 2.0
587 */
588U_STABLE void U_EXPORT2
589ucnv_setSubstChars(UConverter *converter,
590 const char *subChars,
591 int8_t len,
592 UErrorCode *err);
593
594/**
595 * Set a substitution string for converting from Unicode to a charset.
596 * The caller need not know the charset byte sequence for each charset.
597 *
598 * Unlike ucnv_setSubstChars() which is designed to set a charset byte sequence
599 * for a single character, this function takes a Unicode string with
600 * zero, one or more characters, and immediately verifies that the string can be
601 * converted to the charset.
602 * If not, or if the result is too long (more than 32 bytes as of ICU 3.6),
603 * then the function returns with an error accordingly.
604 *
605 * Also unlike ucnv_setSubstChars(), this function works for stateful charsets
606 * by converting on the fly at the point of substitution rather than setting
607 * a fixed byte sequence.
608 *
609 * @param cnv The UConverter object.
610 * @param s The Unicode string.
611 * @param length The number of UChars in s, or -1 for a NUL-terminated string.
612 * @param err Pointer to a standard ICU error code. Its input value must
613 * pass the U_SUCCESS() test, or else the function returns
614 * immediately. Check for U_FAILURE() on output or use with
615 * function chaining. (See User Guide for details.)
616 *
617 * @see ucnv_setSubstChars
618 * @see ucnv_getSubstChars
619 * @stable ICU 3.6
620 */
621U_STABLE void U_EXPORT2
622ucnv_setSubstString(UConverter *cnv,
623 const UChar *s,
624 int32_t length,
625 UErrorCode *err);
626
627/**
628 * Fills in the output parameter, errBytes, with the error characters from the
629 * last failing conversion.
630 *
631 * @param converter the Unicode converter
632 * @param errBytes the codepage bytes which were in error
633 * @param len on input the capacity of errBytes, on output the number of
634 * bytes which were copied to it
635 * @param err the error status code.
636 * If the substitution character array is too small, an
637 * <TT>U_INDEX_OUTOFBOUNDS_ERROR</TT> will be returned.
638 * @stable ICU 2.0
639 */
640U_STABLE void U_EXPORT2
641ucnv_getInvalidChars(const UConverter *converter,
642 char *errBytes,
643 int8_t *len,
644 UErrorCode *err);
645
646/**
647 * Fills in the output parameter, errChars, with the error characters from the
648 * last failing conversion.
649 *
650 * @param converter the Unicode converter
651 * @param errUChars the UChars which were in error
652 * @param len on input the capacity of errUChars, on output the number of
653 * UChars which were copied to it
654 * @param err the error status code.
655 * If the substitution character array is too small, an
656 * <TT>U_INDEX_OUTOFBOUNDS_ERROR</TT> will be returned.
657 * @stable ICU 2.0
658 */
659U_STABLE void U_EXPORT2
660ucnv_getInvalidUChars(const UConverter *converter,
661 UChar *errUChars,
662 int8_t *len,
663 UErrorCode *err);
664
665/**
666 * Resets the state of a converter to the default state. This is used
667 * in the case of an error, to restart a conversion from a known default state.
668 * It will also empty the internal output buffers.
669 * @param converter the Unicode converter
670 * @stable ICU 2.0
671 */
672U_STABLE void U_EXPORT2
673ucnv_reset(UConverter *converter);
674
675/**
676 * Resets the to-Unicode part of a converter state to the default state.
677 * This is used in the case of an error to restart a conversion to
678 * Unicode to a known default state. It will also empty the internal
679 * output buffers used for the conversion to Unicode codepoints.
680 * @param converter the Unicode converter
681 * @stable ICU 2.0
682 */
683U_STABLE void U_EXPORT2
684ucnv_resetToUnicode(UConverter *converter);
685
686/**
687 * Resets the from-Unicode part of a converter state to the default state.
688 * This is used in the case of an error to restart a conversion from
689 * Unicode to a known default state. It will also empty the internal output
690 * buffers used for the conversion from Unicode codepoints.
691 * @param converter the Unicode converter
692 * @stable ICU 2.0
693 */
694U_STABLE void U_EXPORT2
695ucnv_resetFromUnicode(UConverter *converter);
696
697/**
698 * Returns the maximum number of bytes that are output per UChar in conversion
699 * from Unicode using this converter.
700 * The returned number can be used with UCNV_GET_MAX_BYTES_FOR_STRING
701 * to calculate the size of a target buffer for conversion from Unicode.
702 *
703 * Note: Before ICU 2.8, this function did not return reliable numbers for
704 * some stateful converters (EBCDIC_STATEFUL, ISO-2022) and LMBCS.
705 *
706 * This number may not be the same as the maximum number of bytes per
707 * "conversion unit". In other words, it may not be the intuitively expected
708 * number of bytes per character that would be published for a charset,
709 * and may not fulfill any other purpose than the allocation of an output
710 * buffer of guaranteed sufficient size for a given input length and converter.
711 *
712 * Examples for special cases that are taken into account:
713 * - Supplementary code points may convert to more bytes than BMP code points.
714 * This function returns bytes per UChar (UTF-16 code unit), not per
715 * Unicode code point, for efficient buffer allocation.
716 * - State-shifting output (SI/SO, escapes, etc.) from stateful converters.
717 * - When m input UChars are converted to n output bytes, then the maximum m/n
718 * is taken into account.
719 *
720 * The number returned here does not take into account
721 * (see UCNV_GET_MAX_BYTES_FOR_STRING):
722 * - callbacks which output more than one charset character sequence per call,
723 * like escape callbacks
724 * - initial and final non-character bytes that are output by some converters
725 * (automatic BOMs, initial escape sequence, final SI, etc.)
726 *
727 * Examples for returned values:
728 * - SBCS charsets: 1
729 * - Shift-JIS: 2
730 * - UTF-16: 2 (2 per BMP, 4 per surrogate _pair_, BOM not counted)
731 * - UTF-8: 3 (3 per BMP, 4 per surrogate _pair_)
732 * - EBCDIC_STATEFUL (EBCDIC mixed SBCS/DBCS): 3 (SO + DBCS)
733 * - ISO-2022: 3 (always outputs UTF-8)
734 * - ISO-2022-JP: 6 (4-byte escape sequences + DBCS)
735 * - ISO-2022-CN: 8 (4-byte designator sequences + 2-byte SS2/SS3 + DBCS)
736 *
737 * @param converter The Unicode converter.
738 * @return The maximum number of bytes per UChar that are output by ucnv_fromUnicode(),
739 * to be used together with UCNV_GET_MAX_BYTES_FOR_STRING for buffer allocation.
740 *
741 * @see UCNV_GET_MAX_BYTES_FOR_STRING
742 * @see ucnv_getMinCharSize
743 * @stable ICU 2.0
744 */
745U_STABLE int8_t U_EXPORT2
746ucnv_getMaxCharSize(const UConverter *converter);
747
748/**
749 * Calculates the size of a buffer for conversion from Unicode to a charset.
750 * The calculated size is guaranteed to be sufficient for this conversion.
751 *
752 * It takes into account initial and final non-character bytes that are output
753 * by some converters.
754 * It does not take into account callbacks which output more than one charset
755 * character sequence per call, like escape callbacks.
756 * The default (substitution) callback only outputs one charset character sequence.
757 *
758 * @param length Number of UChars to be converted.
759 * @param maxCharSize Return value from ucnv_getMaxCharSize() for the converter
760 * that will be used.
761 * @return Size of a buffer that will be large enough to hold the output bytes of
762 * converting length UChars with the converter that returned the maxCharSize.
763 *
764 * @see ucnv_getMaxCharSize
765 * @stable ICU 2.8
766 */
767#define UCNV_GET_MAX_BYTES_FOR_STRING(length, maxCharSize) \
768 (((int32_t)(length)+10)*(int32_t)(maxCharSize))
769
770/**
771 * Returns the minimum byte length for characters in this codepage.
772 * This is usually either 1 or 2.
773 * @param converter the Unicode converter
774 * @return the minimum number of bytes allowed by this particular converter
775 * @see ucnv_getMaxCharSize
776 * @stable ICU 2.0
777 */
778U_STABLE int8_t U_EXPORT2
779ucnv_getMinCharSize(const UConverter *converter);
780
781/**
782 * Returns the display name of the converter passed in based on the Locale
783 * passed in. If the locale contains no display name, the internal ASCII
784 * name will be filled in.
785 *
786 * @param converter the Unicode converter.
787 * @param displayLocale is the specific Locale we want to localised for
788 * @param displayName user provided buffer to be filled in
789 * @param displayNameCapacity size of displayName Buffer
790 * @param err error status code
791 * @return displayNameLength number of UChar needed in displayName
792 * @see ucnv_getName
793 * @stable ICU 2.0
794 */
795U_STABLE int32_t U_EXPORT2
796ucnv_getDisplayName(const UConverter *converter,
797 const char *displayLocale,
798 UChar *displayName,
799 int32_t displayNameCapacity,
800 UErrorCode *err);
801
802/**
803 * Gets the internal, canonical name of the converter (zero-terminated).
804 * The lifetime of the returned string will be that of the converter
805 * passed to this function.
806 * @param converter the Unicode converter
807 * @param err UErrorCode status
808 * @return the internal name of the converter
809 * @see ucnv_getDisplayName
810 * @stable ICU 2.0
811 */
812U_STABLE const char * U_EXPORT2
813ucnv_getName(const UConverter *converter, UErrorCode *err);
814
815/**
816 * Gets a codepage number associated with the converter. This is not guaranteed
817 * to be the one used to create the converter. Some converters do not represent
818 * platform registered codepages and return zero for the codepage number.
819 * The error code fill-in parameter indicates if the codepage number
820 * is available.
821 * Does not check if the converter is <TT>NULL</TT> or if converter's data
822 * table is <TT>NULL</TT>.
823 *
824 * Important: The use of CCSIDs is not recommended because it is limited
825 * to only two platforms in principle and only one (UCNV_IBM) in the current
826 * ICU converter API.
827 * Also, CCSIDs are insufficient to identify IBM Unicode conversion tables precisely.
828 * For more details see ucnv_openCCSID().
829 *
830 * @param converter the Unicode converter
831 * @param err the error status code.
832 * @return If any error occurrs, -1 will be returned otherwise, the codepage number
833 * will be returned
834 * @see ucnv_openCCSID
835 * @see ucnv_getPlatform
836 * @stable ICU 2.0
837 */
838U_STABLE int32_t U_EXPORT2
839ucnv_getCCSID(const UConverter *converter,
840 UErrorCode *err);
841
842/**
843 * Gets a codepage platform associated with the converter. Currently,
844 * only <TT>UCNV_IBM</TT> will be returned.
845 * Does not test if the converter is <TT>NULL</TT> or if converter's data
846 * table is <TT>NULL</TT>.
847 * @param converter the Unicode converter
848 * @param err the error status code.
849 * @return The codepage platform
850 * @stable ICU 2.0
851 */
852U_STABLE UConverterPlatform U_EXPORT2
853ucnv_getPlatform(const UConverter *converter,
854 UErrorCode *err);
855
856/**
857 * Gets the type of the converter
858 * e.g. SBCS, MBCS, DBCS, UTF8, UTF16_BE, UTF16_LE, ISO_2022,
859 * EBCDIC_STATEFUL, LATIN_1
860 * @param converter a valid, opened converter
861 * @return the type of the converter
862 * @stable ICU 2.0
863 */
864U_STABLE UConverterType U_EXPORT2
865ucnv_getType(const UConverter * converter);
866
867/**
868 * Gets the "starter" (lead) bytes for converters of type MBCS.
869 * Will fill in an <TT>U_ILLEGAL_ARGUMENT_ERROR</TT> if converter passed in
870 * is not MBCS. Fills in an array of type UBool, with the value of the byte
871 * as offset to the array. For example, if (starters[0x20] == TRUE) at return,
872 * it means that the byte 0x20 is a starter byte in this converter.
873 * Context pointers are always owned by the caller.
874 *
875 * @param converter a valid, opened converter of type MBCS
876 * @param starters an array of size 256 to be filled in
877 * @param err error status, <TT>U_ILLEGAL_ARGUMENT_ERROR</TT> if the
878 * converter is not a type which can return starters.
879 * @see ucnv_getType
880 * @stable ICU 2.0
881 */
882U_STABLE void U_EXPORT2
883ucnv_getStarters(const UConverter* converter,
884 UBool starters[256],
885 UErrorCode* err);
886
887
888/**
889 * Selectors for Unicode sets that can be returned by ucnv_getUnicodeSet().
890 * @see ucnv_getUnicodeSet
891 * @stable ICU 2.6
892 */
893typedef enum UConverterUnicodeSet {
894 /** Select the set of roundtrippable Unicode code points. @stable ICU 2.6 */
895 UCNV_ROUNDTRIP_SET,
896 /** Select the set of Unicode code points with roundtrip or fallback mappings. @stable ICU 4.0 */
897 UCNV_ROUNDTRIP_AND_FALLBACK_SET,
898 /** Number of UConverterUnicodeSet selectors. @stable ICU 2.6 */
899 UCNV_SET_COUNT
900} UConverterUnicodeSet;
901
902
903/**
904 * Returns the set of Unicode code points that can be converted by an ICU converter.
905 *
906 * Returns one of several kinds of set:
907 *
908 * 1. UCNV_ROUNDTRIP_SET
909 *
910 * The set of all Unicode code points that can be roundtrip-converted
911 * (converted without any data loss) with the converter (ucnv_fromUnicode()).
912 * This set will not include code points that have fallback mappings
913 * or are only the result of reverse fallback mappings.
914 * This set will also not include PUA code points with fallbacks, although
915 * ucnv_fromUnicode() will always uses those mappings despite ucnv_setFallback().
916 * See UTR #22 "Character Mapping Markup Language"
917 * at http://www.unicode.org/reports/tr22/
918 *
919 * This is useful for example for
920 * - checking that a string or document can be roundtrip-converted with a converter,
921 * without/before actually performing the conversion
922 * - testing if a converter can be used for text for typical text for a certain locale,
923 * by comparing its roundtrip set with the set of ExemplarCharacters from
924 * ICU's locale data or other sources
925 *
926 * 2. UCNV_ROUNDTRIP_AND_FALLBACK_SET
927 *
928 * The set of all Unicode code points that can be converted with the converter (ucnv_fromUnicode())
929 * when fallbacks are turned on (see ucnv_setFallback()).
930 * This set includes all code points with roundtrips and fallbacks (but not reverse fallbacks).
931 *
932 * In the future, there may be more UConverterUnicodeSet choices to select
933 * sets with different properties.
934 *
935 * @param cnv The converter for which a set is requested.
936 * @param setFillIn A valid USet *. It will be cleared by this function before
937 * the converter's specific set is filled into the USet.
938 * @param whichSet A UConverterUnicodeSet selector;
939 * currently UCNV_ROUNDTRIP_SET is the only supported value.
940 * @param pErrorCode ICU error code in/out parameter.
941 * Must fulfill U_SUCCESS before the function call.
942 *
943 * @see UConverterUnicodeSet
944 * @see uset_open
945 * @see uset_close
946 * @stable ICU 2.6
947 */
948U_STABLE void U_EXPORT2
949ucnv_getUnicodeSet(const UConverter *cnv,
950 USet *setFillIn,
951 UConverterUnicodeSet whichSet,
952 UErrorCode *pErrorCode);
953
954/**
955 * Gets the current calback function used by the converter when an illegal
956 * or invalid codepage sequence is found.
957 * Context pointers are always owned by the caller.
958 *
959 * @param converter the unicode converter
960 * @param action fillin: returns the callback function pointer
961 * @param context fillin: returns the callback's private void* context
962 * @see ucnv_setToUCallBack
963 * @stable ICU 2.0
964 */
965U_STABLE void U_EXPORT2
966ucnv_getToUCallBack (const UConverter * converter,
967 UConverterToUCallback *action,
968 const void **context);
969
970/**
971 * Gets the current callback function used by the converter when illegal
972 * or invalid Unicode sequence is found.
973 * Context pointers are always owned by the caller.
974 *
975 * @param converter the unicode converter
976 * @param action fillin: returns the callback function pointer
977 * @param context fillin: returns the callback's private void* context
978 * @see ucnv_setFromUCallBack
979 * @stable ICU 2.0
980 */
981U_STABLE void U_EXPORT2
982ucnv_getFromUCallBack (const UConverter * converter,
983 UConverterFromUCallback *action,
984 const void **context);
985
986/**
987 * Changes the callback function used by the converter when
988 * an illegal or invalid sequence is found.
989 * Context pointers are always owned by the caller.
990 * Predefined actions and contexts can be found in the ucnv_err.h header.
991 *
992 * @param converter the unicode converter
993 * @param newAction the new callback function
994 * @param newContext the new toUnicode callback context pointer. This can be NULL.
995 * @param oldAction fillin: returns the old callback function pointer. This can be NULL.
996 * @param oldContext fillin: returns the old callback's private void* context. This can be NULL.
997 * @param err The error code status
998 * @see ucnv_getToUCallBack
999 * @stable ICU 2.0
1000 */
1001U_STABLE void U_EXPORT2
1002ucnv_setToUCallBack (UConverter * converter,
1003 UConverterToUCallback newAction,
1004 const void* newContext,
1005 UConverterToUCallback *oldAction,
1006 const void** oldContext,
1007 UErrorCode * err);
1008
1009/**
1010 * Changes the current callback function used by the converter when
1011 * an illegal or invalid sequence is found.
1012 * Context pointers are always owned by the caller.
1013 * Predefined actions and contexts can be found in the ucnv_err.h header.
1014 *
1015 * @param converter the unicode converter
1016 * @param newAction the new callback function
1017 * @param newContext the new fromUnicode callback context pointer. This can be NULL.
1018 * @param oldAction fillin: returns the old callback function pointer. This can be NULL.
1019 * @param oldContext fillin: returns the old callback's private void* context. This can be NULL.
1020 * @param err The error code status
1021 * @see ucnv_getFromUCallBack
1022 * @stable ICU 2.0
1023 */
1024U_STABLE void U_EXPORT2
1025ucnv_setFromUCallBack (UConverter * converter,
1026 UConverterFromUCallback newAction,
1027 const void *newContext,
1028 UConverterFromUCallback *oldAction,
1029 const void **oldContext,
1030 UErrorCode * err);
1031
1032/**
1033 * Converts an array of unicode characters to an array of codepage
1034 * characters. This function is optimized for converting a continuous
1035 * stream of data in buffer-sized chunks, where the entire source and
1036 * target does not fit in available buffers.
1037 *
1038 * The source pointer is an in/out parameter. It starts out pointing where the
1039 * conversion is to begin, and ends up pointing after the last UChar consumed.
1040 *
1041 * Target similarly starts out pointer at the first available byte in the output
1042 * buffer, and ends up pointing after the last byte written to the output.
1043 *
1044 * The converter always attempts to consume the entire source buffer, unless
1045 * (1.) the target buffer is full, or (2.) a failing error is returned from the
1046 * current callback function. When a successful error status has been
1047 * returned, it means that all of the source buffer has been
1048 * consumed. At that point, the caller should reset the source and
1049 * sourceLimit pointers to point to the next chunk.
1050 *
1051 * At the end of the stream (flush==TRUE), the input is completely consumed
1052 * when *source==sourceLimit and no error code is set.
1053 * The converter object is then automatically reset by this function.
1054 * (This means that a converter need not be reset explicitly between data
1055 * streams if it finishes the previous stream without errors.)
1056 *
1057 * This is a <I>stateful</I> conversion. Additionally, even when all source data has
1058 * been consumed, some data may be in the converters' internal state.
1059 * Call this function repeatedly, updating the target pointers with
1060 * the next empty chunk of target in case of a
1061 * <TT>U_BUFFER_OVERFLOW_ERROR</TT>, and updating the source pointers
1062 * with the next chunk of source when a successful error status is
1063 * returned, until there are no more chunks of source data.
1064 * @param converter the Unicode converter
1065 * @param target I/O parameter. Input : Points to the beginning of the buffer to copy
1066 * codepage characters to. Output : points to after the last codepage character copied
1067 * to <TT>target</TT>.
1068 * @param targetLimit the pointer just after last of the <TT>target</TT> buffer
1069 * @param source I/O parameter, pointer to pointer to the source Unicode character buffer.
1070 * @param sourceLimit the pointer just after the last of the source buffer
1071 * @param offsets if NULL is passed, nothing will happen to it, otherwise it needs to have the same number
1072 * of allocated cells as <TT>target</TT>. Will fill in offsets from target to source pointer
1073 * e.g: <TT>offsets[3]</TT> is equal to 6, it means that the <TT>target[3]</TT> was a result of transcoding <TT>source[6]</TT>
1074 * For output data carried across calls, and other data without a specific source character
1075 * (such as from escape sequences or callbacks) -1 will be placed for offsets.
1076 * @param flush set to <TT>TRUE</TT> if the current source buffer is the last available
1077 * chunk of the source, <TT>FALSE</TT> otherwise. Note that if a failing status is returned,
1078 * this function may have to be called multiple times with flush set to <TT>TRUE</TT> until
1079 * the source buffer is consumed.
1080 * @param err the error status. <TT>U_ILLEGAL_ARGUMENT_ERROR</TT> will be set if the
1081 * converter is <TT>NULL</TT>.
1082 * <code>U_BUFFER_OVERFLOW_ERROR</code> will be set if the target is full and there is
1083 * still data to be written to the target.
1084 * @see ucnv_fromUChars
1085 * @see ucnv_convert
1086 * @see ucnv_getMinCharSize
1087 * @see ucnv_setToUCallBack
1088 * @stable ICU 2.0
1089 */
1090U_STABLE void U_EXPORT2
1091ucnv_fromUnicode (UConverter * converter,
1092 char **target,
1093 const char *targetLimit,
1094 const UChar ** source,
1095 const UChar * sourceLimit,
1096 int32_t* offsets,
1097 UBool flush,
1098 UErrorCode * err);
1099
1100/**
1101 * Converts a buffer of codepage bytes into an array of unicode UChars
1102 * characters. This function is optimized for converting a continuous
1103 * stream of data in buffer-sized chunks, where the entire source and
1104 * target does not fit in available buffers.
1105 *
1106 * The source pointer is an in/out parameter. It starts out pointing where the
1107 * conversion is to begin, and ends up pointing after the last byte of source consumed.
1108 *
1109 * Target similarly starts out pointer at the first available UChar in the output
1110 * buffer, and ends up pointing after the last UChar written to the output.
1111 * It does NOT necessarily keep UChar sequences together.
1112 *
1113 * The converter always attempts to consume the entire source buffer, unless
1114 * (1.) the target buffer is full, or (2.) a failing error is returned from the
1115 * current callback function. When a successful error status has been
1116 * returned, it means that all of the source buffer has been
1117 * consumed. At that point, the caller should reset the source and
1118 * sourceLimit pointers to point to the next chunk.
1119 *
1120 * At the end of the stream (flush==TRUE), the input is completely consumed
1121 * when *source==sourceLimit and no error code is set
1122 * The converter object is then automatically reset by this function.
1123 * (This means that a converter need not be reset explicitly between data
1124 * streams if it finishes the previous stream without errors.)
1125 *
1126 * This is a <I>stateful</I> conversion. Additionally, even when all source data has
1127 * been consumed, some data may be in the converters' internal state.
1128 * Call this function repeatedly, updating the target pointers with
1129 * the next empty chunk of target in case of a
1130 * <TT>U_BUFFER_OVERFLOW_ERROR</TT>, and updating the source pointers
1131 * with the next chunk of source when a successful error status is
1132 * returned, until there are no more chunks of source data.
1133 * @param converter the Unicode converter
1134 * @param target I/O parameter. Input : Points to the beginning of the buffer to copy
1135 * UChars into. Output : points to after the last UChar copied.
1136 * @param targetLimit the pointer just after the end of the <TT>target</TT> buffer
1137 * @param source I/O parameter, pointer to pointer to the source codepage buffer.
1138 * @param sourceLimit the pointer to the byte after the end of the source buffer
1139 * @param offsets if NULL is passed, nothing will happen to it, otherwise it needs to have the same number
1140 * of allocated cells as <TT>target</TT>. Will fill in offsets from target to source pointer
1141 * e.g: <TT>offsets[3]</TT> is equal to 6, it means that the <TT>target[3]</TT> was a result of transcoding <TT>source[6]</TT>
1142 * For output data carried across calls, and other data without a specific source character
1143 * (such as from escape sequences or callbacks) -1 will be placed for offsets.
1144 * @param flush set to <TT>TRUE</TT> if the current source buffer is the last available
1145 * chunk of the source, <TT>FALSE</TT> otherwise. Note that if a failing status is returned,
1146 * this function may have to be called multiple times with flush set to <TT>TRUE</TT> until
1147 * the source buffer is consumed.
1148 * @param err the error status. <TT>U_ILLEGAL_ARGUMENT_ERROR</TT> will be set if the
1149 * converter is <TT>NULL</TT>.
1150 * <code>U_BUFFER_OVERFLOW_ERROR</code> will be set if the target is full and there is
1151 * still data to be written to the target.
1152 * @see ucnv_fromUChars
1153 * @see ucnv_convert
1154 * @see ucnv_getMinCharSize
1155 * @see ucnv_setFromUCallBack
1156 * @see ucnv_getNextUChar
1157 * @stable ICU 2.0
1158 */
1159U_STABLE void U_EXPORT2
1160ucnv_toUnicode(UConverter *converter,
1161 UChar **target,
1162 const UChar *targetLimit,
1163 const char **source,
1164 const char *sourceLimit,
1165 int32_t *offsets,
1166 UBool flush,
1167 UErrorCode *err);
1168
1169/**
1170 * Convert the Unicode string into a codepage string using an existing UConverter.
1171 * The output string is NUL-terminated if possible.
1172 *
1173 * This function is a more convenient but less powerful version of ucnv_fromUnicode().
1174 * It is only useful for whole strings, not for streaming conversion.
1175 *
1176 * The maximum output buffer capacity required (barring output from callbacks) will be
1177 * UCNV_GET_MAX_BYTES_FOR_STRING(srcLength, ucnv_getMaxCharSize(cnv)).
1178 *
1179 * @param cnv the converter object to be used (ucnv_resetFromUnicode() will be called)
1180 * @param src the input Unicode string
1181 * @param srcLength the input string length, or -1 if NUL-terminated
1182 * @param dest destination string buffer, can be NULL if destCapacity==0
1183 * @param destCapacity the number of chars available at dest
1184 * @param pErrorCode normal ICU error code;
1185 * common error codes that may be set by this function include
1186 * U_BUFFER_OVERFLOW_ERROR, U_STRING_NOT_TERMINATED_WARNING,
1187 * U_ILLEGAL_ARGUMENT_ERROR, and conversion errors
1188 * @return the length of the output string, not counting the terminating NUL;
1189 * if the length is greater than destCapacity, then the string will not fit
1190 * and a buffer of the indicated length would need to be passed in
1191 * @see ucnv_fromUnicode
1192 * @see ucnv_convert
1193 * @see UCNV_GET_MAX_BYTES_FOR_STRING
1194 * @stable ICU 2.0
1195 */
1196U_STABLE int32_t U_EXPORT2
1197ucnv_fromUChars(UConverter *cnv,
1198 char *dest, int32_t destCapacity,
1199 const UChar *src, int32_t srcLength,
1200 UErrorCode *pErrorCode);
1201
1202/**
1203 * Convert the codepage string into a Unicode string using an existing UConverter.
1204 * The output string is NUL-terminated if possible.
1205 *
1206 * This function is a more convenient but less powerful version of ucnv_toUnicode().
1207 * It is only useful for whole strings, not for streaming conversion.
1208 *
1209 * The maximum output buffer capacity required (barring output from callbacks) will be
1210 * 2*srcLength (each char may be converted into a surrogate pair).
1211 *
1212 * @param cnv the converter object to be used (ucnv_resetToUnicode() will be called)
1213 * @param src the input codepage string
1214 * @param srcLength the input string length, or -1 if NUL-terminated
1215 * @param dest destination string buffer, can be NULL if destCapacity==0
1216 * @param destCapacity the number of UChars available at dest
1217 * @param pErrorCode normal ICU error code;
1218 * common error codes that may be set by this function include
1219 * U_BUFFER_OVERFLOW_ERROR, U_STRING_NOT_TERMINATED_WARNING,
1220 * U_ILLEGAL_ARGUMENT_ERROR, and conversion errors
1221 * @return the length of the output string, not counting the terminating NUL;
1222 * if the length is greater than destCapacity, then the string will not fit
1223 * and a buffer of the indicated length would need to be passed in
1224 * @see ucnv_toUnicode
1225 * @see ucnv_convert
1226 * @stable ICU 2.0
1227 */
1228U_STABLE int32_t U_EXPORT2
1229ucnv_toUChars(UConverter *cnv,
1230 UChar *dest, int32_t destCapacity,
1231 const char *src, int32_t srcLength,
1232 UErrorCode *pErrorCode);
1233
1234/**
1235 * Convert a codepage buffer into Unicode one character at a time.
1236 * The input is completely consumed when the U_INDEX_OUTOFBOUNDS_ERROR is set.
1237 *
1238 * Advantage compared to ucnv_toUnicode() or ucnv_toUChars():
1239 * - Faster for small amounts of data, for most converters, e.g.,
1240 * US-ASCII, ISO-8859-1, UTF-8/16/32, and most "normal" charsets.
1241 * (For complex converters, e.g., SCSU, UTF-7 and ISO 2022 variants,
1242 * it uses ucnv_toUnicode() internally.)
1243 * - Convenient.
1244 *
1245 * Limitations compared to ucnv_toUnicode():
1246 * - Always assumes flush=TRUE.
1247 * This makes ucnv_getNextUChar() unsuitable for "streaming" conversion,
1248 * that is, for where the input is supplied in multiple buffers,
1249 * because ucnv_getNextUChar() will assume the end of the input at the end
1250 * of the first buffer.
1251 * - Does not provide offset output.
1252 *
1253 * It is possible to "mix" ucnv_getNextUChar() and ucnv_toUnicode() because
1254 * ucnv_getNextUChar() uses the current state of the converter
1255 * (unlike ucnv_toUChars() which always resets first).
1256 * However, if ucnv_getNextUChar() is called after ucnv_toUnicode()
1257 * stopped in the middle of a character sequence (with flush=FALSE),
1258 * then ucnv_getNextUChar() will always use the slower ucnv_toUnicode()
1259 * internally until the next character boundary.
1260 * (This is new in ICU 2.6. In earlier releases, ucnv_getNextUChar() had to
1261 * start at a character boundary.)
1262 *
1263 * Instead of using ucnv_getNextUChar(), it is recommended
1264 * to convert using ucnv_toUnicode() or ucnv_toUChars()
1265 * and then iterate over the text using U16_NEXT() or a UCharIterator (uiter.h)
1266 * or a C++ CharacterIterator or similar.
1267 * This allows streaming conversion and offset output, for example.
1268 *
1269 * <p>Handling of surrogate pairs and supplementary-plane code points:<br>
1270 * There are two different kinds of codepages that provide mappings for surrogate characters:
1271 * <ul>
1272 * <li>Codepages like UTF-8, UTF-32, and GB 18030 provide direct representations for Unicode
1273 * code points U+10000-U+10ffff as well as for single surrogates U+d800-U+dfff.
1274 * Each valid sequence will result in exactly one returned code point.
1275 * If a sequence results in a single surrogate, then that will be returned
1276 * by itself, even if a neighboring sequence encodes the matching surrogate.</li>
1277 * <li>Codepages like SCSU and LMBCS (and UTF-16) provide direct representations only for BMP code points
1278 * including surrogates. Code points in supplementary planes are represented with
1279 * two sequences, each encoding a surrogate.
1280 * For these codepages, matching pairs of surrogates will be combined into single
1281 * code points for returning from this function.
1282 * (Note that SCSU is actually a mix of these codepage types.)</li>
1283 * </ul></p>
1284 *
1285 * @param converter an open UConverter
1286 * @param source the address of a pointer to the codepage buffer, will be
1287 * updated to point after the bytes consumed in the conversion call.
1288 * @param sourceLimit points to the end of the input buffer
1289 * @param err fills in error status (see ucnv_toUnicode)
1290 * <code>U_INDEX_OUTOFBOUNDS_ERROR</code> will be set if the input
1291 * is empty or does not convert to any output (e.g.: pure state-change
1292 * codes SI/SO, escape sequences for ISO 2022,
1293 * or if the callback did not output anything, ...).
1294 * This function will not set a <code>U_BUFFER_OVERFLOW_ERROR</code> because
1295 * the "buffer" is the return code. However, there might be subsequent output
1296 * stored in the converter object
1297 * that will be returned in following calls to this function.
1298 * @return a UChar32 resulting from the partial conversion of source
1299 * @see ucnv_toUnicode
1300 * @see ucnv_toUChars
1301 * @see ucnv_convert
1302 * @stable ICU 2.0
1303 */
1304U_STABLE UChar32 U_EXPORT2
1305ucnv_getNextUChar(UConverter * converter,
1306 const char **source,
1307 const char * sourceLimit,
1308 UErrorCode * err);
1309
1310/**
1311 * Convert from one external charset to another using two existing UConverters.
1312 * Internally, two conversions - ucnv_toUnicode() and ucnv_fromUnicode() -
1313 * are used, "pivoting" through 16-bit Unicode.
1314 *
1315 * Important: For streaming conversion (multiple function calls for successive
1316 * parts of a text stream), the caller must provide a pivot buffer explicitly,
1317 * and must preserve the pivot buffer and associated pointers from one
1318 * call to another. (The buffer may be moved if its contents and the relative
1319 * pointer positions are preserved.)
1320 *
1321 * There is a similar function, ucnv_convert(),
1322 * which has the following limitations:
1323 * - it takes charset names, not converter objects, so that
1324 * - two converters are opened for each call
1325 * - only single-string conversion is possible, not streaming operation
1326 * - it does not provide enough information to find out,
1327 * in case of failure, whether the toUnicode or
1328 * the fromUnicode conversion failed
1329 *
1330 * By contrast, ucnv_convertEx()
1331 * - takes UConverter parameters instead of charset names
1332 * - fully exposes the pivot buffer for streaming conversion and complete error handling
1333 *
1334 * ucnv_convertEx() also provides further convenience:
1335 * - an option to reset the converters at the beginning
1336 * (if reset==TRUE, see parameters;
1337 * also sets *pivotTarget=*pivotSource=pivotStart)
1338 * - allow NUL-terminated input
1339 * (only a single NUL byte, will not work for charsets with multi-byte NULs)
1340 * (if sourceLimit==NULL, see parameters)
1341 * - terminate with a NUL on output
1342 * (only a single NUL byte, not useful for charsets with multi-byte NULs),
1343 * or set U_STRING_NOT_TERMINATED_WARNING if the output exactly fills
1344 * the target buffer
1345 * - the pivot buffer can be provided internally;
1346 * possible only for whole-string conversion, not streaming conversion;
1347 * in this case, the caller will not be able to get details about where an
1348 * error occurred
1349 * (if pivotStart==NULL, see below)
1350 *
1351 * The function returns when one of the following is true:
1352 * - the entire source text has been converted successfully to the target buffer
1353 * - a target buffer overflow occurred (U_BUFFER_OVERFLOW_ERROR)
1354 * - a conversion error occurred
1355 * (other U_FAILURE(), see description of pErrorCode)
1356 *
1357 * Limitation compared to the direct use of
1358 * ucnv_fromUnicode() and ucnv_toUnicode():
1359 * ucnv_convertEx() does not provide offset information.
1360 *
1361 * Limitation compared to ucnv_fromUChars() and ucnv_toUChars():
1362 * ucnv_convertEx() does not support preflighting directly.
1363 *
1364 * Sample code for converting a single string from
1365 * one external charset to UTF-8, ignoring the location of errors:
1366 *
1367 * \code
1368 * int32_t
1369 * myToUTF8(UConverter *cnv,
1370 * const char *s, int32_t length,
1371 * char *u8, int32_t capacity,
1372 * UErrorCode *pErrorCode) {
1373 * UConverter *utf8Cnv;
1374 * char *target;
1375 *
1376 * if(U_FAILURE(*pErrorCode)) {
1377 * return 0;
1378 * }
1379 *
1380 * utf8Cnv=myGetCachedUTF8Converter(pErrorCode);
1381 * if(U_FAILURE(*pErrorCode)) {
1382 * return 0;
1383 * }
1384 *
1385 * if(length<0) {
1386 * length=strlen(s);
1387 * }
1388 * target=u8;
1389 * ucnv_convertEx(utf8Cnv, cnv,
1390 * &target, u8+capacity,
1391 * &s, s+length,
1392 * NULL, NULL, NULL, NULL,
1393 * TRUE, TRUE,
1394 * pErrorCode);
1395 *
1396 * myReleaseCachedUTF8Converter(utf8Cnv);
1397 *
1398 * // return the output string length, but without preflighting
1399 * return (int32_t)(target-u8);
1400 * }
1401 * \endcode
1402 *
1403 * @param targetCnv Output converter, used to convert from the UTF-16 pivot
1404 * to the target using ucnv_fromUnicode().
1405 * @param sourceCnv Input converter, used to convert from the source to
1406 * the UTF-16 pivot using ucnv_toUnicode().
1407 * @param target I/O parameter, same as for ucnv_fromUChars().
1408 * Input: *target points to the beginning of the target buffer.
1409 * Output: *target points to the first unit after the last char written.
1410 * @param targetLimit Pointer to the first unit after the target buffer.
1411 * @param source I/O parameter, same as for ucnv_toUChars().
1412 * Input: *source points to the beginning of the source buffer.
1413 * Output: *source points to the first unit after the last char read.
1414 * @param sourceLimit Pointer to the first unit after the source buffer.
1415 * @param pivotStart Pointer to the UTF-16 pivot buffer. If pivotStart==NULL,
1416 * then an internal buffer is used and the other pivot
1417 * arguments are ignored and can be NULL as well.
1418 * @param pivotSource I/O parameter, same as source in ucnv_fromUChars() for
1419 * conversion from the pivot buffer to the target buffer.
1420 * @param pivotTarget I/O parameter, same as target in ucnv_toUChars() for
1421 * conversion from the source buffer to the pivot buffer.
1422 * It must be pivotStart<=*pivotSource<=*pivotTarget<=pivotLimit
1423 * and pivotStart<pivotLimit (unless pivotStart==NULL).
1424 * @param pivotLimit Pointer to the first unit after the pivot buffer.
1425 * @param reset If TRUE, then ucnv_resetToUnicode(sourceCnv) and
1426 * ucnv_resetFromUnicode(targetCnv) are called, and the
1427 * pivot pointers are reset (*pivotTarget=*pivotSource=pivotStart).
1428 * @param flush If true, indicates the end of the input.
1429 * Passed directly to ucnv_toUnicode(), and carried over to
1430 * ucnv_fromUnicode() when the source is empty as well.
1431 * @param pErrorCode ICU error code in/out parameter.
1432 * Must fulfill U_SUCCESS before the function call.
1433 * U_BUFFER_OVERFLOW_ERROR always refers to the target buffer
1434 * because overflows into the pivot buffer are handled internally.
1435 * Other conversion errors are from the source-to-pivot
1436 * conversion if *pivotSource==pivotStart, otherwise from
1437 * the pivot-to-target conversion.
1438 *
1439 * @see ucnv_convert
1440 * @see ucnv_fromAlgorithmic
1441 * @see ucnv_toAlgorithmic
1442 * @see ucnv_fromUnicode
1443 * @see ucnv_toUnicode
1444 * @see ucnv_fromUChars
1445 * @see ucnv_toUChars
1446 * @stable ICU 2.6
1447 */
1448U_STABLE void U_EXPORT2
1449ucnv_convertEx(UConverter *targetCnv, UConverter *sourceCnv,
1450 char **target, const char *targetLimit,
1451 const char **source, const char *sourceLimit,
1452 UChar *pivotStart, UChar **pivotSource,
1453 UChar **pivotTarget, const UChar *pivotLimit,
1454 UBool reset, UBool flush,
1455 UErrorCode *pErrorCode);
1456
1457/**
1458 * Convert from one external charset to another.
1459 * Internally, two converters are opened according to the name arguments,
1460 * then the text is converted to and from the 16-bit Unicode "pivot"
1461 * using ucnv_convertEx(), then the converters are closed again.
1462 *
1463 * This is a convenience function, not an efficient way to convert a lot of text:
1464 * ucnv_convert()
1465 * - takes charset names, not converter objects, so that
1466 * - two converters are opened for each call
1467 * - only single-string conversion is possible, not streaming operation
1468 * - does not provide enough information to find out,
1469 * in case of failure, whether the toUnicode or
1470 * the fromUnicode conversion failed
1471 * - allows NUL-terminated input
1472 * (only a single NUL byte, will not work for charsets with multi-byte NULs)
1473 * (if sourceLength==-1, see parameters)
1474 * - terminate with a NUL on output
1475 * (only a single NUL byte, not useful for charsets with multi-byte NULs),
1476 * or set U_STRING_NOT_TERMINATED_WARNING if the output exactly fills
1477 * the target buffer
1478 * - a pivot buffer is provided internally
1479 *
1480 * The function returns when one of the following is true:
1481 * - the entire source text has been converted successfully to the target buffer
1482 * and either the target buffer is terminated with a single NUL byte
1483 * or the error code is set to U_STRING_NOT_TERMINATED_WARNING
1484 * - a target buffer overflow occurred (U_BUFFER_OVERFLOW_ERROR)
1485 * and the full output string length is returned ("preflighting")
1486 * - a conversion error occurred
1487 * (other U_FAILURE(), see description of pErrorCode)
1488 *
1489 * @param toConverterName The name of the converter that is used to convert
1490 * from the UTF-16 pivot buffer to the target.
1491 * @param fromConverterName The name of the converter that is used to convert
1492 * from the source to the UTF-16 pivot buffer.
1493 * @param target Pointer to the output buffer.
1494 * @param targetCapacity Capacity of the target, in bytes.
1495 * @param source Pointer to the input buffer.
1496 * @param sourceLength Length of the input text, in bytes, or -1 for NUL-terminated input.
1497 * @param pErrorCode ICU error code in/out parameter.
1498 * Must fulfill U_SUCCESS before the function call.
1499 * @return Length of the complete output text in bytes, even if it exceeds the targetCapacity
1500 * and a U_BUFFER_OVERFLOW_ERROR is set.
1501 *
1502 * @see ucnv_convertEx
1503 * @see ucnv_fromAlgorithmic
1504 * @see ucnv_toAlgorithmic
1505 * @see ucnv_fromUnicode
1506 * @see ucnv_toUnicode
1507 * @see ucnv_fromUChars
1508 * @see ucnv_toUChars
1509 * @see ucnv_getNextUChar
1510 * @stable ICU 2.0
1511 */
1512U_STABLE int32_t U_EXPORT2
1513ucnv_convert(const char *toConverterName,
1514 const char *fromConverterName,
1515 char *target,
1516 int32_t targetCapacity,
1517 const char *source,
1518 int32_t sourceLength,
1519 UErrorCode *pErrorCode);
1520
1521/**
1522 * Convert from one external charset to another.
1523 * Internally, the text is converted to and from the 16-bit Unicode "pivot"
1524 * using ucnv_convertEx(). ucnv_toAlgorithmic() works exactly like ucnv_convert()
1525 * except that the two converters need not be looked up and opened completely.
1526 *
1527 * The source-to-pivot conversion uses the cnv converter parameter.
1528 * The pivot-to-target conversion uses a purely algorithmic converter
1529 * according to the specified type, e.g., UCNV_UTF8 for a UTF-8 converter.
1530 *
1531 * Internally, the algorithmic converter is opened and closed for each
1532 * function call, which is more efficient than using the public ucnv_open()
1533 * but somewhat less efficient than only resetting an existing converter
1534 * and using ucnv_convertEx().
1535 *
1536 * This function is more convenient than ucnv_convertEx() for single-string
1537 * conversions, especially when "preflighting" is desired (returning the length
1538 * of the complete output even if it does not fit into the target buffer;
1539 * see the User Guide Strings chapter). See ucnv_convert() for details.
1540 *
1541 * @param algorithmicType UConverterType constant identifying the desired target
1542 * charset as a purely algorithmic converter.
1543 * Those are converters for Unicode charsets like
1544 * UTF-8, BOCU-1, SCSU, UTF-7, IMAP-mailbox-name, etc.,
1545 * as well as US-ASCII and ISO-8859-1.
1546 * @param cnv The converter that is used to convert
1547 * from the source to the UTF-16 pivot buffer.
1548 * @param target Pointer to the output buffer.
1549 * @param targetCapacity Capacity of the target, in bytes.
1550 * @param source Pointer to the input buffer.
1551 * @param sourceLength Length of the input text, in bytes
1552 * @param pErrorCode ICU error code in/out parameter.
1553 * Must fulfill U_SUCCESS before the function call.
1554 * @return Length of the complete output text in bytes, even if it exceeds the targetCapacity
1555 * and a U_BUFFER_OVERFLOW_ERROR is set.
1556 *
1557 * @see ucnv_fromAlgorithmic
1558 * @see ucnv_convert
1559 * @see ucnv_convertEx
1560 * @see ucnv_fromUnicode
1561 * @see ucnv_toUnicode
1562 * @see ucnv_fromUChars
1563 * @see ucnv_toUChars
1564 * @stable ICU 2.6
1565 */
1566U_STABLE int32_t U_EXPORT2
1567ucnv_toAlgorithmic(UConverterType algorithmicType,
1568 UConverter *cnv,
1569 char *target, int32_t targetCapacity,
1570 const char *source, int32_t sourceLength,
1571 UErrorCode *pErrorCode);
1572
1573/**
1574 * Convert from one external charset to another.
1575 * Internally, the text is converted to and from the 16-bit Unicode "pivot"
1576 * using ucnv_convertEx(). ucnv_fromAlgorithmic() works exactly like ucnv_convert()
1577 * except that the two converters need not be looked up and opened completely.
1578 *
1579 * The source-to-pivot conversion uses a purely algorithmic converter
1580 * according to the specified type, e.g., UCNV_UTF8 for a UTF-8 converter.
1581 * The pivot-to-target conversion uses the cnv converter parameter.
1582 *
1583 * Internally, the algorithmic converter is opened and closed for each
1584 * function call, which is more efficient than using the public ucnv_open()
1585 * but somewhat less efficient than only resetting an existing converter
1586 * and using ucnv_convertEx().
1587 *
1588 * This function is more convenient than ucnv_convertEx() for single-string
1589 * conversions, especially when "preflighting" is desired (returning the length
1590 * of the complete output even if it does not fit into the target buffer;
1591 * see the User Guide Strings chapter). See ucnv_convert() for details.
1592 *
1593 * @param cnv The converter that is used to convert
1594 * from the UTF-16 pivot buffer to the target.
1595 * @param algorithmicType UConverterType constant identifying the desired source
1596 * charset as a purely algorithmic converter.
1597 * Those are converters for Unicode charsets like
1598 * UTF-8, BOCU-1, SCSU, UTF-7, IMAP-mailbox-name, etc.,
1599 * as well as US-ASCII and ISO-8859-1.
1600 * @param target Pointer to the output buffer.
1601 * @param targetCapacity Capacity of the target, in bytes.
1602 * @param source Pointer to the input buffer.
1603 * @param sourceLength Length of the input text, in bytes
1604 * @param pErrorCode ICU error code in/out parameter.
1605 * Must fulfill U_SUCCESS before the function call.
1606 * @return Length of the complete output text in bytes, even if it exceeds the targetCapacity
1607 * and a U_BUFFER_OVERFLOW_ERROR is set.
1608 *
1609 * @see ucnv_fromAlgorithmic
1610 * @see ucnv_convert
1611 * @see ucnv_convertEx
1612 * @see ucnv_fromUnicode
1613 * @see ucnv_toUnicode
1614 * @see ucnv_fromUChars
1615 * @see ucnv_toUChars
1616 * @stable ICU 2.6
1617 */
1618U_STABLE int32_t U_EXPORT2
1619ucnv_fromAlgorithmic(UConverter *cnv,
1620 UConverterType algorithmicType,
1621 char *target, int32_t targetCapacity,
1622 const char *source, int32_t sourceLength,
1623 UErrorCode *pErrorCode);
1624
1625/**
1626 * Frees up memory occupied by unused, cached converter shared data.
1627 *
1628 * @return the number of cached converters successfully deleted
1629 * @see ucnv_close
1630 * @stable ICU 2.0
1631 */
1632U_STABLE int32_t U_EXPORT2
1633ucnv_flushCache(void);
1634
1635/**
1636 * Returns the number of available converters, as per the alias file.
1637 *
1638 * @return the number of available converters
1639 * @see ucnv_getAvailableName
1640 * @stable ICU 2.0
1641 */
1642U_STABLE int32_t U_EXPORT2
1643ucnv_countAvailable(void);
1644
1645/**
1646 * Gets the canonical converter name of the specified converter from a list of
1647 * all available converters contaied in the alias file. All converters
1648 * in this list can be opened.
1649 *
1650 * @param n the index to a converter available on the system (in the range <TT>[0..ucnv_countAvaiable()]</TT>)
1651 * @return a pointer a string (library owned), or <TT>NULL</TT> if the index is out of bounds.
1652 * @see ucnv_countAvailable
1653 * @stable ICU 2.0
1654 */
1655U_STABLE const char* U_EXPORT2
1656ucnv_getAvailableName(int32_t n);
1657
1658/**
1659 * Returns a UEnumeration to enumerate all of the canonical converter
1660 * names, as per the alias file, regardless of the ability to open each
1661 * converter.
1662 *
1663 * @return A UEnumeration object for getting all the recognized canonical
1664 * converter names.
1665 * @see ucnv_getAvailableName
1666 * @see uenum_close
1667 * @see uenum_next
1668 * @stable ICU 2.4
1669 */
1670U_STABLE UEnumeration * U_EXPORT2
1671ucnv_openAllNames(UErrorCode *pErrorCode);
1672
1673/**
1674 * Gives the number of aliases for a given converter or alias name.
1675 * If the alias is ambiguous, then the preferred converter is used
1676 * and the status is set to U_AMBIGUOUS_ALIAS_WARNING.
1677 * This method only enumerates the listed entries in the alias file.
1678 * @param alias alias name
1679 * @param pErrorCode error status
1680 * @return number of names on alias list for given alias
1681 * @stable ICU 2.0
1682 */
1683U_STABLE uint16_t U_EXPORT2
1684ucnv_countAliases(const char *alias, UErrorCode *pErrorCode);
1685
1686/**
1687 * Gives the name of the alias at given index of alias list.
1688 * This method only enumerates the listed entries in the alias file.
1689 * If the alias is ambiguous, then the preferred converter is used
1690 * and the status is set to U_AMBIGUOUS_ALIAS_WARNING.
1691 * @param alias alias name
1692 * @param n index in alias list
1693 * @param pErrorCode result of operation
1694 * @return returns the name of the alias at given index
1695 * @see ucnv_countAliases
1696 * @stable ICU 2.0
1697 */
1698U_STABLE const char * U_EXPORT2
1699ucnv_getAlias(const char *alias, uint16_t n, UErrorCode *pErrorCode);
1700
1701/**
1702 * Fill-up the list of alias names for the given alias.
1703 * This method only enumerates the listed entries in the alias file.
1704 * If the alias is ambiguous, then the preferred converter is used
1705 * and the status is set to U_AMBIGUOUS_ALIAS_WARNING.
1706 * @param alias alias name
1707 * @param aliases fill-in list, aliases is a pointer to an array of
1708 * <code>ucnv_countAliases()</code> string-pointers
1709 * (<code>const char *</code>) that will be filled in.
1710 * The strings themselves are owned by the library.
1711 * @param pErrorCode result of operation
1712 * @stable ICU 2.0
1713 */
1714U_STABLE void U_EXPORT2
1715ucnv_getAliases(const char *alias, const char **aliases, UErrorCode *pErrorCode);
1716
1717/**
1718 * Return a new UEnumeration object for enumerating all the
1719 * alias names for a given converter that are recognized by a standard.
1720 * This method only enumerates the listed entries in the alias file.
1721 * The convrtrs.txt file can be modified to change the results of
1722 * this function.
1723 * The first result in this list is the same result given by
1724 * <code>ucnv_getStandardName</code>, which is the default alias for
1725 * the specified standard name. The returned object must be closed with
1726 * <code>uenum_close</code> when you are done with the object.
1727 *
1728 * @param convName original converter name
1729 * @param standard name of the standard governing the names; MIME and IANA
1730 * are such standards
1731 * @param pErrorCode The error code
1732 * @return A UEnumeration object for getting all aliases that are recognized
1733 * by a standard. If any of the parameters are invalid, NULL
1734 * is returned.
1735 * @see ucnv_getStandardName
1736 * @see uenum_close
1737 * @see uenum_next
1738 * @stable ICU 2.2
1739 */
1740U_STABLE UEnumeration * U_EXPORT2
1741ucnv_openStandardNames(const char *convName,
1742 const char *standard,
1743 UErrorCode *pErrorCode);
1744
1745/**
1746 * Gives the number of standards associated to converter names.
1747 * @return number of standards
1748 * @stable ICU 2.0
1749 */
1750U_STABLE uint16_t U_EXPORT2
1751ucnv_countStandards(void);
1752
1753/**
1754 * Gives the name of the standard at given index of standard list.
1755 * @param n index in standard list
1756 * @param pErrorCode result of operation
1757 * @return returns the name of the standard at given index. Owned by the library.
1758 * @stable ICU 2.0
1759 */
1760U_STABLE const char * U_EXPORT2
1761ucnv_getStandard(uint16_t n, UErrorCode *pErrorCode);
1762
1763/**
1764 * Returns a standard name for a given converter name.
1765 * <p>
1766 * Example alias table:<br>
1767 * conv alias1 { STANDARD1 } alias2 { STANDARD1* }
1768 * <p>
1769 * Result of ucnv_getStandardName("conv", "STANDARD1") from example
1770 * alias table:<br>
1771 * <b>"alias2"</b>
1772 *
1773 * @param name original converter name
1774 * @param standard name of the standard governing the names; MIME and IANA
1775 * are such standards
1776 * @param pErrorCode result of operation
1777 * @return returns the standard converter name;
1778 * if a standard converter name cannot be determined,
1779 * then <code>NULL</code> is returned. Owned by the library.
1780 * @stable ICU 2.0
1781 */
1782U_STABLE const char * U_EXPORT2
1783ucnv_getStandardName(const char *name, const char *standard, UErrorCode *pErrorCode);
1784
1785/**
1786 * This function will return the internal canonical converter name of the
1787 * tagged alias. This is the opposite of ucnv_openStandardNames, which
1788 * returns the tagged alias given the canonical name.
1789 * <p>
1790 * Example alias table:<br>
1791 * conv alias1 { STANDARD1 } alias2 { STANDARD1* }
1792 * <p>
1793 * Result of ucnv_getStandardName("alias1", "STANDARD1") from example
1794 * alias table:<br>
1795 * <b>"conv"</b>
1796 *
1797 * @return returns the canonical converter name;
1798 * if a standard or alias name cannot be determined,
1799 * then <code>NULL</code> is returned. The returned string is
1800 * owned by the library.
1801 * @see ucnv_getStandardName
1802 * @stable ICU 2.4
1803 */
1804U_STABLE const char * U_EXPORT2
1805ucnv_getCanonicalName(const char *alias, const char *standard, UErrorCode *pErrorCode);
1806
1807/**
1808 * Returns the current default converter name. If you want to open
1809 * a default converter, you do not need to use this function.
1810 * It is faster if you pass a NULL argument to ucnv_open the
1811 * default converter.
1812 *
1813 * If U_CHARSET_IS_UTF8 is defined to 1 in utypes.h then this function
1814 * always returns "UTF-8".
1815 *
1816 * @return returns the current default converter name.
1817 * Storage owned by the library
1818 * @see ucnv_setDefaultName
1819 * @stable ICU 2.0
1820 */
1821U_STABLE const char * U_EXPORT2
1822ucnv_getDefaultName(void);
1823
1824/**
1825 * This function is not thread safe. DO NOT call this function when ANY ICU
1826 * function is being used from more than one thread! This function sets the
1827 * current default converter name. If this function needs to be called, it
1828 * should be called during application initialization. Most of the time, the
1829 * results from ucnv_getDefaultName() or ucnv_open with a NULL string argument
1830 * is sufficient for your application.
1831 *
1832 * If U_CHARSET_IS_UTF8 is defined to 1 in utypes.h then this function
1833 * does nothing.
1834 *
1835 * @param name the converter name to be the default (must be known by ICU).
1836 * @see ucnv_getDefaultName
1837 * @system
1838 * @stable ICU 2.0
1839 */
1840U_STABLE void U_EXPORT2
1841ucnv_setDefaultName(const char *name);
1842
1843/**
1844 * Fixes the backslash character mismapping. For example, in SJIS, the backslash
1845 * character in the ASCII portion is also used to represent the yen currency sign.
1846 * When mapping from Unicode character 0x005C, it's unclear whether to map the
1847 * character back to yen or backslash in SJIS. This function will take the input
1848 * buffer and replace all the yen sign characters with backslash. This is necessary
1849 * when the user tries to open a file with the input buffer on Windows.
1850 * This function will test the converter to see whether such mapping is
1851 * required. You can sometimes avoid using this function by using the correct version
1852 * of Shift-JIS.
1853 *
1854 * @param cnv The converter representing the target codepage.
1855 * @param source the input buffer to be fixed
1856 * @param sourceLen the length of the input buffer
1857 * @see ucnv_isAmbiguous
1858 * @stable ICU 2.0
1859 */
1860U_STABLE void U_EXPORT2
1861ucnv_fixFileSeparator(const UConverter *cnv, UChar *source, int32_t sourceLen);
1862
1863/**
1864 * Determines if the converter contains ambiguous mappings of the same
1865 * character or not.
1866 * @param cnv the converter to be tested
1867 * @return TRUE if the converter contains ambiguous mapping of the same
1868 * character, FALSE otherwise.
1869 * @stable ICU 2.0
1870 */
1871U_STABLE UBool U_EXPORT2
1872ucnv_isAmbiguous(const UConverter *cnv);
1873
1874/**
1875 * Sets the converter to use fallback mappings or not.
1876 * Regardless of this flag, the converter will always use
1877 * fallbacks from Unicode Private Use code points, as well as
1878 * reverse fallbacks (to Unicode).
1879 * For details see ".ucm File Format"
1880 * in the Conversion Data chapter of the ICU User Guide:
1881 * http://www.icu-project.org/userguide/conversion-data.html#ucmformat
1882 *
1883 * @param cnv The converter to set the fallback mapping usage on.
1884 * @param usesFallback TRUE if the user wants the converter to take advantage of the fallback
1885 * mapping, FALSE otherwise.
1886 * @stable ICU 2.0
1887 * @see ucnv_usesFallback
1888 */
1889U_STABLE void U_EXPORT2
1890ucnv_setFallback(UConverter *cnv, UBool usesFallback);
1891
1892/**
1893 * Determines if the converter uses fallback mappings or not.
1894 * This flag has restrictions, see ucnv_setFallback().
1895 *
1896 * @param cnv The converter to be tested
1897 * @return TRUE if the converter uses fallback, FALSE otherwise.
1898 * @stable ICU 2.0
1899 * @see ucnv_setFallback
1900 */
1901U_STABLE UBool U_EXPORT2
1902ucnv_usesFallback(const UConverter *cnv);
1903
1904/**
1905 * Detects Unicode signature byte sequences at the start of the byte stream
1906 * and returns the charset name of the indicated Unicode charset.
1907 * NULL is returned when no Unicode signature is recognized.
1908 * The number of bytes in the signature is output as well.
1909 *
1910 * The caller can ucnv_open() a converter using the charset name.
1911 * The first code unit (UChar) from the start of the stream will be U+FEFF
1912 * (the Unicode BOM/signature character) and can usually be ignored.
1913 *
1914 * For most Unicode charsets it is also possible to ignore the indicated
1915 * number of initial stream bytes and start converting after them.
1916 * However, there are stateful Unicode charsets (UTF-7 and BOCU-1) for which
1917 * this will not work. Therefore, it is best to ignore the first output UChar
1918 * instead of the input signature bytes.
1919 * <p>
1920 * Usage:
1921 * @code
1922 * UErrorCode err = U_ZERO_ERROR;
1923 * char input[] = { '\xEF','\xBB', '\xBF','\x41','\x42','\x43' };
1924 * int32_t signatureLength = 0;
1925 * char *encoding = ucnv_detectUnicodeSignature(input,sizeof(input),&signatureLength,&err);
1926 * UConverter *conv = NULL;
1927 * UChar output[100];
1928 * UChar *target = output, *out;
1929 * char *source = input;
1930 * if(encoding!=NULL && U_SUCCESS(err)){
1931 * // should signature be discarded ?
1932 * conv = ucnv_open(encoding, &err);
1933 * // do the conversion
1934 * ucnv_toUnicode(conv,
1935 * target, output + sizeof(output)/U_SIZEOF_UCHAR,
1936 * source, input + sizeof(input),
1937 * NULL, TRUE, &err);
1938 * out = output;
1939 * if (discardSignature){
1940 * ++out; // ignore initial U+FEFF
1941 * }
1942 * while(out != target) {
1943 * printf("%04x ", *out++);
1944 * }
1945 * puts("");
1946 * }
1947 *
1948 * @endcode
1949 *
1950 * @param source The source string in which the signature should be detected.
1951 * @param sourceLength Length of the input string, or -1 if terminated with a NUL byte.
1952 * @param signatureLength A pointer to int32_t to receive the number of bytes that make up the signature
1953 * of the detected UTF. 0 if not detected.
1954 * Can be a NULL pointer.
1955 * @param pErrorCode ICU error code in/out parameter.
1956 * Must fulfill U_SUCCESS before the function call.
1957 * @return The name of the encoding detected. NULL if encoding is not detected.
1958 * @stable ICU 2.4
1959 */
1960U_STABLE const char* U_EXPORT2
1961ucnv_detectUnicodeSignature(const char* source,
1962 int32_t sourceLength,
1963 int32_t *signatureLength,
1964 UErrorCode *pErrorCode);
1965
1966/**
1967 * Returns the number of UChars held in the converter's internal state
1968 * because more input is needed for completing the conversion. This function is
1969 * useful for mapping semantics of ICU's converter interface to those of iconv,
1970 * and this information is not needed for normal conversion.
1971 * @param cnv The converter in which the input is held
1972 * @param status ICU error code in/out parameter.
1973 * Must fulfill U_SUCCESS before the function call.
1974 * @return The number of UChars in the state. -1 if an error is encountered.
1975 * @stable ICU 3.4
1976 */
1977U_STABLE int32_t U_EXPORT2
1978ucnv_fromUCountPending(const UConverter* cnv, UErrorCode* status);
1979
1980/**
1981 * Returns the number of chars held in the converter's internal state
1982 * because more input is needed for completing the conversion. This function is
1983 * useful for mapping semantics of ICU's converter interface to those of iconv,
1984 * and this information is not needed for normal conversion.
1985 * @param cnv The converter in which the input is held as internal state
1986 * @param status ICU error code in/out parameter.
1987 * Must fulfill U_SUCCESS before the function call.
1988 * @return The number of chars in the state. -1 if an error is encountered.
1989 * @stable ICU 3.4
1990 */
1991U_STABLE int32_t U_EXPORT2
1992ucnv_toUCountPending(const UConverter* cnv, UErrorCode* status);
1993
1994#endif
1995
1996#endif
1997/*_UCNV*/