Symaptic: os/textandloc/fontservices/textshaperplugin/IcuSource/common/unicode/ucnv.h@260cb5ec6c19 (annotated)

sl@0	1	/*
sl@0	2	**********************************************************************
sl@0	3	* Copyright (C) 1999-2005, International Business Machines
sl@0	4	* Corporation and others. All Rights Reserved.
sl@0	5	**********************************************************************
sl@0	6	* ucnv.h:
sl@0	7	* External APIs for the ICU's codeset conversion library
sl@0	8	* Bertrand A. Damiba
sl@0	9	*
sl@0	10	* Modification History:
sl@0	11	*
sl@0	12	* Date Name Description
sl@0	13	* 04/04/99 helena Fixed internal header inclusion.
sl@0	14	* 05/11/00 helena Added setFallback and usesFallback APIs.
sl@0	15	* 06/29/2000 helena Major rewrite of the callback APIs.
sl@0	16	* 12/07/2000 srl Update of documentation
sl@0	17	*/
sl@0	18
sl@0	19	/**
sl@0	20	* \file
sl@0	21	* \brief C API: Character conversion
sl@0	22	*
sl@0	23	* <h2>Character Conversion C API</h2>
sl@0	24	*
sl@0	25	* <p>This API is used to convert codepage or character encoded data to and
sl@0	26	* from UTF-16. You can open a converter with {@link ucnv_open() }. With that
sl@0	27	* converter, you can get its properties, set options, convert your data and
sl@0	28	* close the converter.</p>
sl@0	29	*
sl@0	30	* <p>Since many software programs recogize different converter names for
sl@0	31	* different types of converters, there are other functions in this API to
sl@0	32	* iterate over the converter aliases. The functions {@link ucnv_getAvailableName() },
sl@0	33	* {@link ucnv_getAlias() } and {@link ucnv_getStandardName() } are some of the
sl@0	34	* more frequently used alias functions to get this information.</p>
sl@0	35	*
sl@0	36	* <p>When a converter encounters an illegal, irregular, invalid or unmappable character
sl@0	37	* its default behavior is to use a substitution character to replace the
sl@0	38	* bad byte sequence. This behavior can be changed by using {@link ucnv_setFromUCallBack() }
sl@0	39	* or {@link ucnv_setToUCallBack() } on the converter. The header ucnv_err.h defines
sl@0	40	* many other callback actions that can be used instead of a character substitution.</p>
sl@0	41	*
sl@0	42	* <p>More information about this API can be found in our
sl@0	43	* <a href="http://icu.sourceforge.net/userguide/conversion.html">User's
sl@0	44	* Guide</a>.</p>
sl@0	45	*/
sl@0	46
sl@0	47	#ifndef UCNV_H
sl@0	48	#define UCNV_H
sl@0	49
sl@0	50	#include "unicode/ucnv_err.h"
sl@0	51	#include "unicode/uenum.h"
sl@0	52
sl@0	53	#ifndef __USET_H__
sl@0	54
sl@0	55	/**
sl@0	56	* USet is the C API type for Unicode sets.
sl@0	57	* It is forward-declared here to avoid including the header file if related
sl@0	58	* conversion APIs are not used.
sl@0	59	* See unicode/uset.h
sl@0	60	*
sl@0	61	* @see ucnv_getUnicodeSet
sl@0	62	* @stable ICU 2.6
sl@0	63	*/
sl@0	64	struct USet;
sl@0	65	/** @stable ICU 2.6 */
sl@0	66	typedef struct USet USet;
sl@0	67
sl@0	68	#endif
sl@0	69
sl@0	70	#if !UCONFIG_NO_CONVERSION
sl@0	71
sl@0	72	U_CDECL_BEGIN
sl@0	73
sl@0	74	/** Maximum length of a converter name including the terminating NULL @stable ICU 2.0 */
sl@0	75	#define UCNV_MAX_CONVERTER_NAME_LENGTH 60
sl@0	76	/** Maximum length of a converter name including path and terminating NULL @stable ICU 2.0 */
sl@0	77	#define UCNV_MAX_FULL_FILE_NAME_LENGTH (600+UCNV_MAX_CONVERTER_NAME_LENGTH)
sl@0	78
sl@0	79	/** Shift in for EBDCDIC_STATEFUL and iso2022 states @stable ICU 2.0 */
sl@0	80	#define UCNV_SI 0x0F
sl@0	81	/** Shift out for EBDCDIC_STATEFUL and iso2022 states @stable ICU 2.0 */
sl@0	82	#define UCNV_SO 0x0E
sl@0	83
sl@0	84	/**
sl@0	85	* Enum for specifying basic types of converters
sl@0	86	* @see ucnv_getType
sl@0	87	* @stable ICU 2.0
sl@0	88	*/
sl@0	89	typedef enum {
sl@0	90	UCNV_UNSUPPORTED_CONVERTER = -1,
sl@0	91	UCNV_SBCS = 0,
sl@0	92	UCNV_DBCS = 1,
sl@0	93	UCNV_MBCS = 2,
sl@0	94	UCNV_LATIN_1 = 3,
sl@0	95	UCNV_UTF8 = 4,
sl@0	96	UCNV_UTF16_BigEndian = 5,
sl@0	97	UCNV_UTF16_LittleEndian = 6,
sl@0	98	UCNV_UTF32_BigEndian = 7,
sl@0	99	UCNV_UTF32_LittleEndian = 8,
sl@0	100	UCNV_EBCDIC_STATEFUL = 9,
sl@0	101	UCNV_ISO_2022 = 10,
sl@0	102
sl@0	103	UCNV_LMBCS_1 = 11,
sl@0	104	UCNV_LMBCS_2,
sl@0	105	UCNV_LMBCS_3,
sl@0	106	UCNV_LMBCS_4,
sl@0	107	UCNV_LMBCS_5,
sl@0	108	UCNV_LMBCS_6,
sl@0	109	UCNV_LMBCS_8,
sl@0	110	UCNV_LMBCS_11,
sl@0	111	UCNV_LMBCS_16,
sl@0	112	UCNV_LMBCS_17,
sl@0	113	UCNV_LMBCS_18,
sl@0	114	UCNV_LMBCS_19,
sl@0	115	UCNV_LMBCS_LAST = UCNV_LMBCS_19,
sl@0	116	UCNV_HZ,
sl@0	117	UCNV_SCSU,
sl@0	118	UCNV_ISCII,
sl@0	119	UCNV_US_ASCII,
sl@0	120	UCNV_UTF7,
sl@0	121	UCNV_BOCU1,
sl@0	122	UCNV_UTF16,
sl@0	123	UCNV_UTF32,
sl@0	124	UCNV_CESU8,
sl@0	125	UCNV_IMAP_MAILBOX,
sl@0	126
sl@0	127	/* Number of converter types for which we have conversion routines. */
sl@0	128	UCNV_NUMBER_OF_SUPPORTED_CONVERTER_TYPES
sl@0	129
sl@0	130	} UConverterType;
sl@0	131
sl@0	132	/**
sl@0	133	* Enum for specifying which platform a converter ID refers to.
sl@0	134	* The use of platform/CCSID is not recommended. See ucnv_openCCSID().
sl@0	135	*
sl@0	136	* @see ucnv_getPlatform
sl@0	137	* @see ucnv_openCCSID
sl@0	138	* @see ucnv_getCCSID
sl@0	139	* @stable ICU 2.0
sl@0	140	*/
sl@0	141	typedef enum {
sl@0	142	UCNV_UNKNOWN = -1,
sl@0	143	UCNV_IBM = 0
sl@0	144	} UConverterPlatform;
sl@0	145
sl@0	146	/**
sl@0	147	* Function pointer for error callback in the codepage to unicode direction.
sl@0	148	* Called when an error has occured in conversion to unicode, or on open/close of the callback (see reason).
sl@0	149	* @param context Pointer to the callback's private data
sl@0	150	* @param args Information about the conversion in progress
sl@0	151	* @param codeUnits Points to 'length' bytes of the concerned codepage sequence
sl@0	152	* @param length Size (in bytes) of the concerned codepage sequence
sl@0	153	* @param reason Defines the reason the callback was invoked
sl@0	154	* @param pErrorCode ICU error code in/out parameter.
sl@0	155	* For converter callback functions, set to a conversion error
sl@0	156	* before the call, and the callback may reset it to U_ZERO_ERROR.
sl@0	157	* @see ucnv_setToUCallBack
sl@0	158	* @see UConverterToUnicodeArgs
sl@0	159	* @stable ICU 2.0
sl@0	160	*/
sl@0	161	typedef void (U_EXPORT2 *UConverterToUCallback) (
sl@0	162	const void* context,
sl@0	163	UConverterToUnicodeArgs *args,
sl@0	164	const char *codeUnits,
sl@0	165	int32_t length,
sl@0	166	UConverterCallbackReason reason,
sl@0	167	UErrorCode *pErrorCode);
sl@0	168
sl@0	169	/**
sl@0	170	* Function pointer for error callback in the unicode to codepage direction.
sl@0	171	* Called when an error has occured in conversion from unicode, or on open/close of the callback (see reason).
sl@0	172	* @param context Pointer to the callback's private data
sl@0	173	* @param args Information about the conversion in progress
sl@0	174	* @param codeUnits Points to 'length' UChars of the concerned Unicode sequence
sl@0	175	* @param length Size (in bytes) of the concerned codepage sequence
sl@0	176	* @param codePoint Single UChar32 (UTF-32) containing the concerend Unicode codepoint.
sl@0	177	* @param reason Defines the reason the callback was invoked
sl@0	178	* @param pErrorCode ICU error code in/out parameter.
sl@0	179	* For converter callback functions, set to a conversion error
sl@0	180	* before the call, and the callback may reset it to U_ZERO_ERROR.
sl@0	181	* @see ucnv_setFromUCallBack
sl@0	182	* @stable ICU 2.0
sl@0	183	*/
sl@0	184	typedef void (U_EXPORT2 *UConverterFromUCallback) (
sl@0	185	const void* context,
sl@0	186	UConverterFromUnicodeArgs *args,
sl@0	187	const UChar* codeUnits,
sl@0	188	int32_t length,
sl@0	189	UChar32 codePoint,
sl@0	190	UConverterCallbackReason reason,
sl@0	191	UErrorCode *pErrorCode);
sl@0	192
sl@0	193	U_CDECL_END
sl@0	194
sl@0	195	/**
sl@0	196	* Character that separates converter names from options and options from each other.
sl@0	197	* @see ucnv_open
sl@0	198	* @stable ICU 2.0
sl@0	199	*/
sl@0	200	#define UCNV_OPTION_SEP_CHAR ','
sl@0	201
sl@0	202	/**
sl@0	203	* String version of UCNV_OPTION_SEP_CHAR.
sl@0	204	* @see ucnv_open
sl@0	205	* @stable ICU 2.0
sl@0	206	*/
sl@0	207	#define UCNV_OPTION_SEP_STRING ","
sl@0	208
sl@0	209	/**
sl@0	210	* Character that separates a converter option from its value.
sl@0	211	* @see ucnv_open
sl@0	212	* @stable ICU 2.0
sl@0	213	*/
sl@0	214	#define UCNV_VALUE_SEP_CHAR '='
sl@0	215
sl@0	216	/**
sl@0	217	* String version of UCNV_VALUE_SEP_CHAR.
sl@0	218	* @see ucnv_open
sl@0	219	* @stable ICU 2.0
sl@0	220	*/
sl@0	221	#define UCNV_VALUE_SEP_STRING "="
sl@0	222
sl@0	223	/**
sl@0	224	* Converter option for specifying a locale.
sl@0	225	* For example, ucnv_open("SCSU,locale=ja", &errorCode);
sl@0	226	* See convrtrs.txt.
sl@0	227	*
sl@0	228	* @see ucnv_open
sl@0	229	* @stable ICU 2.0
sl@0	230	*/
sl@0	231	#define UCNV_LOCALE_OPTION_STRING ",locale="
sl@0	232
sl@0	233	/**
sl@0	234	* Converter option for specifying a version selector (0..9) for some converters.
sl@0	235	* For example, ucnv_open("UTF-7,version=1", &errorCode);
sl@0	236	* See convrtrs.txt.
sl@0	237	*
sl@0	238	* @see ucnv_open
sl@0	239	* @stable ICU 2.4
sl@0	240	*/
sl@0	241	#define UCNV_VERSION_OPTION_STRING ",version="
sl@0	242
sl@0	243	/**
sl@0	244	* Converter option for EBCDIC SBCS or mixed-SBCS/DBCS (stateful) codepages.
sl@0	245	* Swaps Unicode mappings for EBCDIC LF and NL codes, as used on
sl@0	246	* S/390 (z/OS) Unix System Services (Open Edition).
sl@0	247	* For example, ucnv_open("ibm-1047,swaplfnl", &errorCode);
sl@0	248	* See convrtrs.txt.
sl@0	249	*
sl@0	250	* @see ucnv_open
sl@0	251	* @stable ICU 2.4
sl@0	252	*/
sl@0	253	#define UCNV_SWAP_LFNL_OPTION_STRING ",swaplfnl"
sl@0	254
sl@0	255	/**
sl@0	256	* Do a fuzzy compare of a two converter/alias names. The comparison
sl@0	257	* is case-insensitive. It also ignores the characters '-', '_', and
sl@0	258	* ' ' (dash, underscore, and space). Thus the strings "UTF-8",
sl@0	259	* "utf_8", and "Utf 8" are exactly equivalent.
sl@0	260	*
sl@0	261	* @param name1 a converter name or alias, zero-terminated
sl@0	262	* @param name2 a converter name or alias, zero-terminated
sl@0	263	* @return 0 if the names match, or a negative value if the name1
sl@0	264	* lexically precedes name2, or a positive value if the name1
sl@0	265	* lexically follows name2.
sl@0	266	* @stable ICU 2.0
sl@0	267	*/
sl@0	268	U_STABLE int U_EXPORT2
sl@0	269	ucnv_compareNames(const char name1, const char name2);
sl@0	270
sl@0	271
sl@0	272	/**
sl@0	273	* Creates a UConverter object with the names specified as a C string.
sl@0	274	* The actual name will be resolved with the alias file
sl@0	275	* using a case-insensitive string comparison that ignores
sl@0	276	* the delimiters '-', '_', and ' ' (dash, underscore, and space).
sl@0	277	* E.g., the names "UTF8", "utf-8", and "Utf 8" are all equivalent.
sl@0	278	* If <code>NULL</code> is passed for the converter name, it will create one with the
sl@0	279	* getDefaultName return value.
sl@0	280	*
sl@0	281	* <p>A converter name for ICU 1.5 and above may contain options
sl@0	282	* like a locale specification to control the specific behavior of
sl@0	283	* the newly instantiated converter.
sl@0	284	* The meaning of the options depends on the particular converter.
sl@0	285	* If an option is not defined for or recognized by a given converter, then it is ignored.</p>
sl@0	286	*
sl@0	287	* <p>Options are appended to the converter name string, with a
sl@0	288	* <code>UCNV_OPTION_SEP_CHAR</code> between the name and the first option and
sl@0	289	* also between adjacent options.</p>
sl@0	290	*
sl@0	291	* <p>If the alias is ambiguous, then the preferred converter is used
sl@0	292	* and the status is set to U_AMBIGUOUS_ALIAS_WARNING.</p>
sl@0	293	*
sl@0	294	* <p>The conversion behavior and names can vary between platforms. ICU may
sl@0	295	* convert some characters differently from other platforms. Details on this topic
sl@0	296	* are in the <a href="http://icu.sourceforge.net/userguide/conversion.html">User's
sl@0	297	* Guide</a>.</p>
sl@0	298	*
sl@0	299	* @param converterName Name of the uconv table, may have options appended
sl@0	300	* @param err outgoing error status <TT>U_MEMORY_ALLOCATION_ERROR, U_FILE_ACCESS_ERROR</TT>
sl@0	301	* @return the created Unicode converter object, or <TT>NULL</TT> if an error occured
sl@0	302	* @see ucnv_openU
sl@0	303	* @see ucnv_openCCSID
sl@0	304	* @see ucnv_close
sl@0	305	* @stable ICU 2.0
sl@0	306	*/
sl@0	307	U_STABLE UConverter* U_EXPORT2
sl@0	308	ucnv_open(const char converterName, UErrorCode err);
sl@0	309
sl@0	310
sl@0	311	/**
sl@0	312	* Creates a Unicode converter with the names specified as unicode string.
sl@0	313	* The name should be limited to the ASCII-7 alphanumerics range.
sl@0	314	* The actual name will be resolved with the alias file
sl@0	315	* using a case-insensitive string comparison that ignores
sl@0	316	* the delimiters '-', '_', and ' ' (dash, underscore, and space).
sl@0	317	* E.g., the names "UTF8", "utf-8", and "Utf 8" are all equivalent.
sl@0	318	* If <TT>NULL</TT> is passed for the converter name, it will create
sl@0	319	* one with the ucnv_getDefaultName() return value.
sl@0	320	* If the alias is ambiguous, then the preferred converter is used
sl@0	321	* and the status is set to U_AMBIGUOUS_ALIAS_WARNING.
sl@0	322	* @param name : name of the uconv table in a zero terminated
sl@0	323	* Unicode string
sl@0	324	* @param err outgoing error status <TT>U_MEMORY_ALLOCATION_ERROR,
sl@0	325	* U_FILE_ACCESS_ERROR</TT>
sl@0	326	* @return the created Unicode converter object, or <TT>NULL</TT> if an
sl@0	327	* error occured
sl@0	328	* @see ucnv_open
sl@0	329	* @see ucnv_openCCSID
sl@0	330	* @see ucnv_close
sl@0	331	* @see ucnv_getDefaultName
sl@0	332	* @stable ICU 2.0
sl@0	333	*/
sl@0	334	U_STABLE UConverter* U_EXPORT2
sl@0	335	ucnv_openU(const UChar *name,
sl@0	336	UErrorCode *err);
sl@0	337
sl@0	338	/**
sl@0	339	* Creates a UConverter object from a CCSID number and platform pair.
sl@0	340	* Note that the usefulness of this function is limited to platforms with numeric
sl@0	341	* encoding IDs. Only IBM and Microsoft platforms use numeric (16-bit) identifiers for
sl@0	342	* encodings.
sl@0	343	*
sl@0	344	* In addition, IBM CCSIDs and Unicode conversion tables are not 1:1 related.
sl@0	345	* For many IBM CCSIDs there are multiple (up to six) Unicode conversion tables, and
sl@0	346	* for some Unicode conversion tables there are multiple CCSIDs.
sl@0	347	* Some "alternate" Unicode conversion tables are provided by the
sl@0	348	* IBM CDRA conversion table registry.
sl@0	349	* The most prominent example of a systematic modification of conversion tables that is
sl@0	350	* not provided in the form of conversion table files in the repository is
sl@0	351	* that S/390 Unix System Services swaps the codes for Line Feed and New Line in all
sl@0	352	* EBCDIC codepages, which requires such a swap in the Unicode conversion tables as well.
sl@0	353	*
sl@0	354	* Only IBM default conversion tables are accessible with ucnv_openCCSID().
sl@0	355	* ucnv_getCCSID() will return the same CCSID for all conversion tables that are associated
sl@0	356	* with that CCSID.
sl@0	357	*
sl@0	358	* Currently, the only "platform" supported in the ICU converter API is UCNV_IBM.
sl@0	359	*
sl@0	360	* In summary, the use of CCSIDs and the associated API functions is not recommended.
sl@0	361	*
sl@0	362	* In order to open a converter with the default IBM CDRA Unicode conversion table,
sl@0	363	* you can use this function or use the prefix "ibm-":
sl@0	364	* \code
sl@0	365	* char name[20];
sl@0	366	* sprintf(name, "ibm-%hu", ccsid);
sl@0	367	* cnv=ucnv_open(name, &errorCode);
sl@0	368	* \endcode
sl@0	369	*
sl@0	370	* In order to open a converter with the IBM S/390 Unix System Services variant
sl@0	371	* of a Unicode/EBCDIC conversion table,
sl@0	372	* you can use the prefix "ibm-" together with the option string UCNV_SWAP_LFNL_OPTION_STRING:
sl@0	373	* \code
sl@0	374	* char name[20];
sl@0	375	* sprintf(name, "ibm-%hu" UCNV_SWAP_LFNL_OPTION_STRING, ccsid);
sl@0	376	* cnv=ucnv_open(name, &errorCode);
sl@0	377	* \endcode
sl@0	378	*
sl@0	379	* In order to open a converter from a Microsoft codepage number, use the prefix "cp":
sl@0	380	* \code
sl@0	381	* char name[20];
sl@0	382	* sprintf(name, "cp%hu", codepageID);
sl@0	383	* cnv=ucnv_open(name, &errorCode);
sl@0	384	* \endcode
sl@0	385	*
sl@0	386	* If the alias is ambiguous, then the preferred converter is used
sl@0	387	* and the status is set to U_AMBIGUOUS_ALIAS_WARNING.
sl@0	388	*
sl@0	389	* @param codepage codepage number to create
sl@0	390	* @param platform the platform in which the codepage number exists
sl@0	391	* @param err error status <TT>U_MEMORY_ALLOCATION_ERROR, U_FILE_ACCESS_ERROR</TT>
sl@0	392	* @return the created Unicode converter object, or <TT>NULL</TT> if an error
sl@0	393	* occured.
sl@0	394	* @see ucnv_open
sl@0	395	* @see ucnv_openU
sl@0	396	* @see ucnv_close
sl@0	397	* @see ucnv_getCCSID
sl@0	398	* @see ucnv_getPlatform
sl@0	399	* @see UConverterPlatform
sl@0	400	* @stable ICU 2.0
sl@0	401	*/
sl@0	402	U_STABLE UConverter* U_EXPORT2
sl@0	403	ucnv_openCCSID(int32_t codepage,
sl@0	404	UConverterPlatform platform,
sl@0	405	UErrorCode * err);
sl@0	406
sl@0	407	/**
sl@0	408	* <p>Creates a UConverter object specified from a packageName and a converterName.</p>
sl@0	409	*
sl@0	410	* <p>The packageName and converterName must point to an ICU udata object, as defined by
sl@0	411	* <code> udata_open( packageName, "cnv", converterName, err) </code> or equivalent.
sl@0	412	* Typically, packageName will refer to a (.dat) file, or to a package registered with
sl@0	413	* udata_setAppData(). Using a full file or directory pathname for packageName is deprecated.</p>
sl@0	414	*
sl@0	415	* <p>The name will NOT be looked up in the alias mechanism, nor will the converter be
sl@0	416	* stored in the converter cache or the alias table. The only way to open further converters
sl@0	417	* is call this function multiple times, or use the ucnv_safeClone() function to clone a
sl@0	418	* 'master' converter.</p>
sl@0	419	*
sl@0	420	* <p>A future version of ICU may add alias table lookups and/or caching
sl@0	421	* to this function.</p>
sl@0	422	*
sl@0	423	* <p>Example Use:
sl@0	424	* <code>cnv = ucnv_openPackage("myapp", "myconverter", &err);</code>
sl@0	425	* </p>
sl@0	426	*
sl@0	427	* @param packageName name of the package (equivalent to 'path' in udata_open() call)
sl@0	428	* @param converterName name of the data item to be used, without suffix.
sl@0	429	* @param err outgoing error status <TT>U_MEMORY_ALLOCATION_ERROR, U_FILE_ACCESS_ERROR</TT>
sl@0	430	* @return the created Unicode converter object, or <TT>NULL</TT> if an error occured
sl@0	431	* @see udata_open
sl@0	432	* @see ucnv_open
sl@0	433	* @see ucnv_safeClone
sl@0	434	* @see ucnv_close
sl@0	435	* @stable ICU 2.2
sl@0	436	*/
sl@0	437	U_STABLE UConverter* U_EXPORT2
sl@0	438	ucnv_openPackage(const char packageName, const char converterName, UErrorCode *err);
sl@0	439
sl@0	440	/**
sl@0	441	* Thread safe converter cloning operation.
sl@0	442	* For most efficient operation, pass in a stackBuffer (and a *pBufferSize)
sl@0	443	* with at least U_CNV_SAFECLONE_BUFFERSIZE bytes of space.
sl@0	444	* If the buffer size is sufficient, then the clone will use the stack buffer;
sl@0	445	* otherwise, it will be allocated, and *pBufferSize will indicate
sl@0	446	* the actual size. (This should not occur with U_CNV_SAFECLONE_BUFFERSIZE.)
sl@0	447	*
sl@0	448	* You must ucnv_close() the clone in any case.
sl@0	449	*
sl@0	450	* If *pBufferSize==0, (regardless of whether stackBuffer==NULL or not)
sl@0	451	* then *pBufferSize will be changed to a sufficient size
sl@0	452	* for cloning this converter,
sl@0	453	* without actually cloning the converter ("pure pre-flighting").
sl@0	454	*
sl@0	455	* If *pBufferSize is greater than zero but not large enough for a stack-based
sl@0	456	* clone, then the converter is cloned using newly allocated memory
sl@0	457	* and *pBufferSize is changed to the necessary size.
sl@0	458	*
sl@0	459	* If the converter clone fits into the stack buffer but the stack buffer is not
sl@0	460	* sufficiently aligned for the clone, then the clone will use an
sl@0	461	* adjusted pointer and use an accordingly smaller buffer size.
sl@0	462	*
sl@0	463	* @param cnv converter to be cloned
sl@0	464	* @param stackBuffer user allocated space for the new clone. If NULL new memory will be allocated.
sl@0	465	* If buffer is not large enough, new memory will be allocated.
sl@0	466	* Clients can use the U_CNV_SAFECLONE_BUFFERSIZE. This will probably be enough to avoid memory allocations.
sl@0	467	* @param pBufferSize pointer to size of allocated space. pBufferSize must not be NULL.
sl@0	468	* @param status to indicate whether the operation went on smoothly or there were errors
sl@0	469	* An informational status value, U_SAFECLONE_ALLOCATED_WARNING,
sl@0	470	* is used if any allocations were necessary.
sl@0	471	* However, it is better to check if *pBufferSize grew for checking for
sl@0	472	* allocations because warning codes can be overridden by subsequent
sl@0	473	* function calls.
sl@0	474	* @return pointer to the new clone
sl@0	475	* @stable ICU 2.0
sl@0	476	*/
sl@0	477	U_STABLE UConverter * U_EXPORT2
sl@0	478	ucnv_safeClone(const UConverter *cnv,
sl@0	479	void *stackBuffer,
sl@0	480	int32_t *pBufferSize,
sl@0	481	UErrorCode *status);
sl@0	482
sl@0	483	/**
sl@0	484	* \def U_CNV_SAFECLONE_BUFFERSIZE
sl@0	485	* Definition of a buffer size that is designed to be large enough for
sl@0	486	* converters to be cloned with ucnv_safeClone().
sl@0	487	* @stable ICU 2.0
sl@0	488	*/
sl@0	489	#define U_CNV_SAFECLONE_BUFFERSIZE 1024
sl@0	490
sl@0	491	/**
sl@0	492	* Deletes the unicode converter and releases resources associated
sl@0	493	* with just this instance.
sl@0	494	* Does not free up shared converter tables.
sl@0	495	*
sl@0	496	* @param converter the converter object to be deleted
sl@0	497	* @see ucnv_open
sl@0	498	* @see ucnv_openU
sl@0	499	* @see ucnv_openCCSID
sl@0	500	* @stable ICU 2.0
sl@0	501	*/
sl@0	502	U_STABLE void U_EXPORT2
sl@0	503	ucnv_close(UConverter * converter);
sl@0	504
sl@0	505	/**
sl@0	506	* Fills in the output parameter, subChars, with the substitution characters
sl@0	507	* as multiple bytes.
sl@0	508	*
sl@0	509	* @param converter the Unicode converter
sl@0	510	* @param subChars the subsitution characters
sl@0	511	* @param len on input the capacity of subChars, on output the number
sl@0	512	* of bytes copied to it
sl@0	513	* @param err the outgoing error status code.
sl@0	514	* If the substitution character array is too small, an
sl@0	515	* <TT>U_INDEX_OUTOFBOUNDS_ERROR</TT> will be returned.
sl@0	516	* @see ucnv_setSubstChars
sl@0	517	* @stable ICU 2.0
sl@0	518	*/
sl@0	519	U_STABLE void U_EXPORT2
sl@0	520	ucnv_getSubstChars(const UConverter *converter,
sl@0	521	char *subChars,
sl@0	522	int8_t *len,
sl@0	523	UErrorCode *err);
sl@0	524
sl@0	525	/**
sl@0	526	* Sets the substitution chars when converting from unicode to a codepage. The
sl@0	527	* substitution is specified as a string of 1-4 bytes, and may contain
sl@0	528	* <TT>NULL</TT> byte.
sl@0	529	* @param converter the Unicode converter
sl@0	530	* @param subChars the substitution character byte sequence we want set
sl@0	531	* @param len the number of bytes in subChars
sl@0	532	* @param err the error status code. <TT>U_INDEX_OUTOFBOUNDS_ERROR </TT> if
sl@0	533	* len is bigger than the maximum number of bytes allowed in subchars
sl@0	534	* @see ucnv_getSubstChars
sl@0	535	* @stable ICU 2.0
sl@0	536	*/
sl@0	537	U_STABLE void U_EXPORT2
sl@0	538	ucnv_setSubstChars(UConverter *converter,
sl@0	539	const char *subChars,
sl@0	540	int8_t len,
sl@0	541	UErrorCode *err);
sl@0	542
sl@0	543	/**
sl@0	544	* Fills in the output parameter, errBytes, with the error characters from the
sl@0	545	* last failing conversion.
sl@0	546	*
sl@0	547	* @param converter the Unicode converter
sl@0	548	* @param errBytes the codepage bytes which were in error
sl@0	549	* @param len on input the capacity of errBytes, on output the number of
sl@0	550	* bytes which were copied to it
sl@0	551	* @param err the error status code.
sl@0	552	* If the substitution character array is too small, an
sl@0	553	* <TT>U_INDEX_OUTOFBOUNDS_ERROR</TT> will be returned.
sl@0	554	* @stable ICU 2.0
sl@0	555	*/
sl@0	556	U_STABLE void U_EXPORT2
sl@0	557	ucnv_getInvalidChars(const UConverter *converter,
sl@0	558	char *errBytes,
sl@0	559	int8_t *len,
sl@0	560	UErrorCode *err);
sl@0	561
sl@0	562	/**
sl@0	563	* Fills in the output parameter, errChars, with the error characters from the
sl@0	564	* last failing conversion.
sl@0	565	*
sl@0	566	* @param converter the Unicode converter
sl@0	567	* @param errUChars the UChars which were in error
sl@0	568	* @param len on input the capacity of errUChars, on output the number of
sl@0	569	* UChars which were copied to it
sl@0	570	* @param err the error status code.
sl@0	571	* If the substitution character array is too small, an
sl@0	572	* <TT>U_INDEX_OUTOFBOUNDS_ERROR</TT> will be returned.
sl@0	573	* @stable ICU 2.0
sl@0	574	*/
sl@0	575	U_STABLE void U_EXPORT2
sl@0	576	ucnv_getInvalidUChars(const UConverter *converter,
sl@0	577	UChar *errUChars,
sl@0	578	int8_t *len,
sl@0	579	UErrorCode *err);
sl@0	580
sl@0	581	/**
sl@0	582	* Resets the state of a converter to the default state. This is used
sl@0	583	* in the case of an error, to restart a conversion from a known default state.
sl@0	584	* It will also empty the internal output buffers.
sl@0	585	* @param converter the Unicode converter
sl@0	586	* @stable ICU 2.0
sl@0	587	*/
sl@0	588	U_STABLE void U_EXPORT2
sl@0	589	ucnv_reset(UConverter *converter);
sl@0	590
sl@0	591	/**
sl@0	592	* Resets the to-Unicode part of a converter state to the default state.
sl@0	593	* This is used in the case of an error to restart a conversion to
sl@0	594	* Unicode to a known default state. It will also empty the internal
sl@0	595	* output buffers used for the conversion to Unicode codepoints.
sl@0	596	* @param converter the Unicode converter
sl@0	597	* @stable ICU 2.0
sl@0	598	*/
sl@0	599	U_STABLE void U_EXPORT2
sl@0	600	ucnv_resetToUnicode(UConverter *converter);
sl@0	601
sl@0	602	/**
sl@0	603	* Resets the from-Unicode part of a converter state to the default state.
sl@0	604	* This is used in the case of an error to restart a conversion from
sl@0	605	* Unicode to a known default state. It will also empty the internal output
sl@0	606	* buffers used for the conversion from Unicode codepoints.
sl@0	607	* @param converter the Unicode converter
sl@0	608	* @stable ICU 2.0
sl@0	609	*/
sl@0	610	U_STABLE void U_EXPORT2
sl@0	611	ucnv_resetFromUnicode(UConverter *converter);
sl@0	612
sl@0	613	/**
sl@0	614	* Returns the maximum number of bytes that are output per UChar in conversion
sl@0	615	* from Unicode using this converter.
sl@0	616	* The returned number can be used with UCNV_GET_MAX_BYTES_FOR_STRING
sl@0	617	* to calculate the size of a target buffer for conversion from Unicode.
sl@0	618	*
sl@0	619	* Note: Before ICU 2.8, this function did not return reliable numbers for
sl@0	620	* some stateful converters (EBCDIC_STATEFUL, ISO-2022) and LMBCS.
sl@0	621	*
sl@0	622	* This number may not be the same as the maximum number of bytes per
sl@0	623	* "conversion unit". In other words, it may not be the intuitively expected
sl@0	624	* number of bytes per character that would be published for a charset,
sl@0	625	* and may not fulfill any other purpose than the allocation of an output
sl@0	626	* buffer of guaranteed sufficient size for a given input length and converter.
sl@0	627	*
sl@0	628	* Examples for special cases that are taken into account:
sl@0	629	* - Supplementary code points may convert to more bytes than BMP code points.
sl@0	630	* This function returns bytes per UChar (UTF-16 code unit), not per
sl@0	631	* Unicode code point, for efficient buffer allocation.
sl@0	632	* - State-shifting output (SI/SO, escapes, etc.) from stateful converters.
sl@0	633	* - When m input UChars are converted to n output bytes, then the maximum m/n
sl@0	634	* is taken into account.
sl@0	635	*
sl@0	636	* The number returned here does not take into account
sl@0	637	* (see UCNV_GET_MAX_BYTES_FOR_STRING):
sl@0	638	* - callbacks which output more than one charset character sequence per call,
sl@0	639	* like escape callbacks
sl@0	640	* - initial and final non-character bytes that are output by some converters
sl@0	641	* (automatic BOMs, initial escape sequence, final SI, etc.)
sl@0	642	*
sl@0	643	* Examples for returned values:
sl@0	644	* - SBCS charsets: 1
sl@0	645	* - Shift-JIS: 2
sl@0	646	* - UTF-16: 2 (2 per BMP, 4 per surrogate _pair_, BOM not counted)
sl@0	647	* - UTF-8: 3 (3 per BMP, 4 per surrogate _pair_)
sl@0	648	* - EBCDIC_STATEFUL (EBCDIC mixed SBCS/DBCS): 3 (SO + DBCS)
sl@0	649	* - ISO-2022: 3 (always outputs UTF-8)
sl@0	650	* - ISO-2022-JP: 6 (4-byte escape sequences + DBCS)
sl@0	651	* - ISO-2022-CN: 8 (4-byte designator sequences + 2-byte SS2/SS3 + DBCS)
sl@0	652	*
sl@0	653	* @param converter The Unicode converter.
sl@0	654	* @return The maximum number of bytes per UChar that are output by ucnv_fromUnicode(),
sl@0	655	* to be used together with UCNV_GET_MAX_BYTES_FOR_STRING for buffer allocation.
sl@0	656	*
sl@0	657	* @see UCNV_GET_MAX_BYTES_FOR_STRING
sl@0	658	* @see ucnv_getMinCharSize
sl@0	659	* @stable ICU 2.0
sl@0	660	*/
sl@0	661	U_STABLE int8_t U_EXPORT2
sl@0	662	ucnv_getMaxCharSize(const UConverter *converter);
sl@0	663
sl@0	664	#ifndef U_HIDE_DRAFT_API
sl@0	665
sl@0	666	/**
sl@0	667	* Calculates the size of a buffer for conversion from Unicode to a charset.
sl@0	668	* The calculated size is guaranteed to be sufficient for this conversion.
sl@0	669	*
sl@0	670	* It takes into account initial and final non-character bytes that are output
sl@0	671	* by some converters.
sl@0	672	* It does not take into account callbacks which output more than one charset
sl@0	673	* character sequence per call, like escape callbacks.
sl@0	674	* The default (substitution) callback only outputs one charset character sequence.
sl@0	675	*
sl@0	676	* @param length Number of UChars to be converted.
sl@0	677	* @param maxCharSize Return value from ucnv_getMaxCharSize() for the converter
sl@0	678	* that will be used.
sl@0	679	* @return Size of a buffer that will be large enough to hold the output bytes of
sl@0	680	* converting length UChars with the converter that returned the maxCharSize.
sl@0	681	*
sl@0	682	* @see ucnv_getMaxCharSize
sl@0	683	* @stable ICU 2.8
sl@0	684	*/
sl@0	685	#define UCNV_GET_MAX_BYTES_FOR_STRING(length, maxCharSize) \
sl@0	686	(((int32_t)(length)+10)*(int32_t)(maxCharSize))
sl@0	687
sl@0	688	#endif /U_HIDE_DRAFT_API/
sl@0	689
sl@0	690	/**
sl@0	691	* Returns the minimum byte length for characters in this codepage.
sl@0	692	* This is usually either 1 or 2.
sl@0	693	* @param converter the Unicode converter
sl@0	694	* @return the minimum number of bytes allowed by this particular converter
sl@0	695	* @see ucnv_getMaxCharSize
sl@0	696	* @stable ICU 2.0
sl@0	697	*/
sl@0	698	U_STABLE int8_t U_EXPORT2
sl@0	699	ucnv_getMinCharSize(const UConverter *converter);
sl@0	700
sl@0	701	/**
sl@0	702	* Returns the display name of the converter passed in based on the Locale
sl@0	703	* passed in. If the locale contains no display name, the internal ASCII
sl@0	704	* name will be filled in.
sl@0	705	*
sl@0	706	* @param converter the Unicode converter.
sl@0	707	* @param displayLocale is the specific Locale we want to localised for
sl@0	708	* @param displayName user provided buffer to be filled in
sl@0	709	* @param displayNameCapacity size of displayName Buffer
sl@0	710	* @param err error status code
sl@0	711	* @return displayNameLength number of UChar needed in displayName
sl@0	712	* @see ucnv_getName
sl@0	713	* @stable ICU 2.0
sl@0	714	*/
sl@0	715	U_STABLE int32_t U_EXPORT2
sl@0	716	ucnv_getDisplayName(const UConverter *converter,
sl@0	717	const char *displayLocale,
sl@0	718	UChar *displayName,
sl@0	719	int32_t displayNameCapacity,
sl@0	720	UErrorCode *err);
sl@0	721
sl@0	722	/**
sl@0	723	* Gets the internal, canonical name of the converter (zero-terminated).
sl@0	724	* The lifetime of the returned string will be that of the converter
sl@0	725	* passed to this function.
sl@0	726	* @param converter the Unicode converter
sl@0	727	* @param err UErrorCode status
sl@0	728	* @return the internal name of the converter
sl@0	729	* @see ucnv_getDisplayName
sl@0	730	* @stable ICU 2.0
sl@0	731	*/
sl@0	732	U_STABLE const char * U_EXPORT2
sl@0	733	ucnv_getName(const UConverter converter, UErrorCode err);
sl@0	734
sl@0	735	/**
sl@0	736	* Gets a codepage number associated with the converter. This is not guaranteed
sl@0	737	* to be the one used to create the converter. Some converters do not represent
sl@0	738	* platform registered codepages and return zero for the codepage number.
sl@0	739	* The error code fill-in parameter indicates if the codepage number
sl@0	740	* is available.
sl@0	741	* Does not check if the converter is <TT>NULL</TT> or if converter's data
sl@0	742	* table is <TT>NULL</TT>.
sl@0	743	*
sl@0	744	* Important: The use of CCSIDs is not recommended because it is limited
sl@0	745	* to only two platforms in principle and only one (UCNV_IBM) in the current
sl@0	746	* ICU converter API.
sl@0	747	* Also, CCSIDs are insufficient to identify IBM Unicode conversion tables precisely.
sl@0	748	* For more details see ucnv_openCCSID().
sl@0	749	*
sl@0	750	* @param converter the Unicode converter
sl@0	751	* @param err the error status code.
sl@0	752	* @return If any error occurrs, -1 will be returned otherwise, the codepage number
sl@0	753	* will be returned
sl@0	754	* @see ucnv_openCCSID
sl@0	755	* @see ucnv_getPlatform
sl@0	756	* @stable ICU 2.0
sl@0	757	*/
sl@0	758	U_STABLE int32_t U_EXPORT2
sl@0	759	ucnv_getCCSID(const UConverter *converter,
sl@0	760	UErrorCode *err);
sl@0	761
sl@0	762	/**
sl@0	763	* Gets a codepage platform associated with the converter. Currently,
sl@0	764	* only <TT>UCNV_IBM</TT> will be returned.
sl@0	765	* Does not test if the converter is <TT>NULL</TT> or if converter's data
sl@0	766	* table is <TT>NULL</TT>.
sl@0	767	* @param converter the Unicode converter
sl@0	768	* @param err the error status code.
sl@0	769	* @return The codepage platform
sl@0	770	* @stable ICU 2.0
sl@0	771	*/
sl@0	772	U_STABLE UConverterPlatform U_EXPORT2
sl@0	773	ucnv_getPlatform(const UConverter *converter,
sl@0	774	UErrorCode *err);
sl@0	775
sl@0	776	/**
sl@0	777	* Gets the type of the converter
sl@0	778	* e.g. SBCS, MBCS, DBCS, UTF8, UTF16_BE, UTF16_LE, ISO_2022,
sl@0	779	* EBCDIC_STATEFUL, LATIN_1
sl@0	780	* @param converter a valid, opened converter
sl@0	781	* @return the type of the converter
sl@0	782	* @stable ICU 2.0
sl@0	783	*/
sl@0	784	U_STABLE UConverterType U_EXPORT2
sl@0	785	ucnv_getType(const UConverter * converter);
sl@0	786
sl@0	787	/**
sl@0	788	* Gets the "starter" (lead) bytes for converters of type MBCS.
sl@0	789	* Will fill in an <TT>U_ILLEGAL_ARGUMENT_ERROR</TT> if converter passed in
sl@0	790	* is not MBCS. Fills in an array of type UBool, with the value of the byte
sl@0	791	* as offset to the array. For example, if (starters[0x20] == TRUE) at return,
sl@0	792	* it means that the byte 0x20 is a starter byte in this converter.
sl@0	793	* Context pointers are always owned by the caller.
sl@0	794	*
sl@0	795	* @param converter a valid, opened converter of type MBCS
sl@0	796	* @param starters an array of size 256 to be filled in
sl@0	797	* @param err error status, <TT>U_ILLEGAL_ARGUMENT_ERROR</TT> if the
sl@0	798	* converter is not a type which can return starters.
sl@0	799	* @see ucnv_getType
sl@0	800	* @stable ICU 2.0
sl@0	801	*/
sl@0	802	U_STABLE void U_EXPORT2
sl@0	803	ucnv_getStarters(const UConverter* converter,
sl@0	804	UBool starters[256],
sl@0	805	UErrorCode* err);
sl@0	806
sl@0	807
sl@0	808	/**
sl@0	809	* Selectors for Unicode sets that can be returned by ucnv_getUnicodeSet().
sl@0	810	* @see ucnv_getUnicodeSet
sl@0	811	* @stable ICU 2.6
sl@0	812	*/
sl@0	813	typedef enum UConverterUnicodeSet {
sl@0	814	/** Select the set of roundtrippable Unicode code points. @stable ICU 2.6 */
sl@0	815	UCNV_ROUNDTRIP_SET,
sl@0	816	/** Number of UConverterUnicodeSet selectors. @stable ICU 2.6 */
sl@0	817	UCNV_SET_COUNT
sl@0	818	} UConverterUnicodeSet;
sl@0	819
sl@0	820
sl@0	821	/**
sl@0	822	* Returns the set of Unicode code points that can be converted by an ICU converter.
sl@0	823	*
sl@0	824	* The current implementation returns only one kind of set (UCNV_ROUNDTRIP_SET):
sl@0	825	* The set of all Unicode code points that can be roundtrip-converted
sl@0	826	* (converted without any data loss) with the converter.
sl@0	827	* This set will not include code points that have fallback mappings
sl@0	828	* or are only the result of reverse fallback mappings.
sl@0	829	* See UTR #22 "Character Mapping Markup Language"
sl@0	830	* at http://www.unicode.org/reports/tr22/
sl@0	831	*
sl@0	832	* This is useful for example for
sl@0	833	* - checking that a string or document can be roundtrip-converted with a converter,
sl@0	834	* without/before actually performing the conversion
sl@0	835	* - testing if a converter can be used for text for typical text for a certain locale,
sl@0	836	* by comparing its roundtrip set with the set of ExemplarCharacters from
sl@0	837	* ICU's locale data or other sources
sl@0	838	*
sl@0	839	* In the future, there may be more UConverterUnicodeSet choices to select
sl@0	840	* sets with different properties.
sl@0	841	*
sl@0	842	* @param cnv The converter for which a set is requested.
sl@0	843	* @param setFillIn A valid USet *. It will be cleared by this function before
sl@0	844	* the converter's specific set is filled into the USet.
sl@0	845	* @param whichSet A UConverterUnicodeSet selector;
sl@0	846	* currently UCNV_ROUNDTRIP_SET is the only supported value.
sl@0	847	* @param pErrorCode ICU error code in/out parameter.
sl@0	848	* Must fulfill U_SUCCESS before the function call.
sl@0	849	*
sl@0	850	* @see UConverterUnicodeSet
sl@0	851	* @see uset_open
sl@0	852	* @see uset_close
sl@0	853	* @stable ICU 2.6
sl@0	854	*/
sl@0	855	U_STABLE void U_EXPORT2
sl@0	856	ucnv_getUnicodeSet(const UConverter *cnv,
sl@0	857	USet *setFillIn,
sl@0	858	UConverterUnicodeSet whichSet,
sl@0	859	UErrorCode *pErrorCode);
sl@0	860
sl@0	861	/**
sl@0	862	* Gets the current calback function used by the converter when an illegal
sl@0	863	* or invalid codepage sequence is found.
sl@0	864	* Context pointers are always owned by the caller.
sl@0	865	*
sl@0	866	* @param converter the unicode converter
sl@0	867	* @param action fillin: returns the callback function pointer
sl@0	868	* @param context fillin: returns the callback's private void* context
sl@0	869	* @see ucnv_setToUCallBack
sl@0	870	* @stable ICU 2.0
sl@0	871	*/
sl@0	872	U_STABLE void U_EXPORT2
sl@0	873	ucnv_getToUCallBack (const UConverter * converter,
sl@0	874	UConverterToUCallback *action,
sl@0	875	const void **context);
sl@0	876
sl@0	877	/**
sl@0	878	* Gets the current callback function used by the converter when illegal
sl@0	879	* or invalid Unicode sequence is found.
sl@0	880	* Context pointers are always owned by the caller.
sl@0	881	*
sl@0	882	* @param converter the unicode converter
sl@0	883	* @param action fillin: returns the callback function pointer
sl@0	884	* @param context fillin: returns the callback's private void* context
sl@0	885	* @see ucnv_setFromUCallBack
sl@0	886	* @stable ICU 2.0
sl@0	887	*/
sl@0	888	U_STABLE void U_EXPORT2
sl@0	889	ucnv_getFromUCallBack (const UConverter * converter,
sl@0	890	UConverterFromUCallback *action,
sl@0	891	const void **context);
sl@0	892
sl@0	893	/**
sl@0	894	* Changes the callback function used by the converter when
sl@0	895	* an illegal or invalid sequence is found.
sl@0	896	* Context pointers are always owned by the caller.
sl@0	897	* Predefined actions and contexts can be found in the ucnv_err.h header.
sl@0	898	*
sl@0	899	* @param converter the unicode converter
sl@0	900	* @param newAction the new callback function
sl@0	901	* @param newContext the new toUnicode callback context pointer. This can be NULL.
sl@0	902	* @param oldAction fillin: returns the old callback function pointer. This can be NULL.
sl@0	903	* @param oldContext fillin: returns the old callback's private void* context. This can be NULL.
sl@0	904	* @param err The error code status
sl@0	905	* @see ucnv_getToUCallBack
sl@0	906	* @stable ICU 2.0
sl@0	907	*/
sl@0	908	U_STABLE void U_EXPORT2
sl@0	909	ucnv_setToUCallBack (UConverter * converter,
sl@0	910	UConverterToUCallback newAction,
sl@0	911	const void* newContext,
sl@0	912	UConverterToUCallback *oldAction,
sl@0	913	const void** oldContext,
sl@0	914	UErrorCode * err);
sl@0	915
sl@0	916	/**
sl@0	917	* Changes the current callback function used by the converter when
sl@0	918	* an illegal or invalid sequence is found.
sl@0	919	* Context pointers are always owned by the caller.
sl@0	920	* Predefined actions and contexts can be found in the ucnv_err.h header.
sl@0	921	*
sl@0	922	* @param converter the unicode converter
sl@0	923	* @param newAction the new callback function
sl@0	924	* @param newContext the new fromUnicode callback context pointer. This can be NULL.
sl@0	925	* @param oldAction fillin: returns the old callback function pointer. This can be NULL.
sl@0	926	* @param oldContext fillin: returns the old callback's private void* context. This can be NULL.
sl@0	927	* @param err The error code status
sl@0	928	* @see ucnv_getFromUCallBack
sl@0	929	* @stable ICU 2.0
sl@0	930	*/
sl@0	931	U_STABLE void U_EXPORT2
sl@0	932	ucnv_setFromUCallBack (UConverter * converter,
sl@0	933	UConverterFromUCallback newAction,
sl@0	934	const void *newContext,
sl@0	935	UConverterFromUCallback *oldAction,
sl@0	936	const void **oldContext,
sl@0	937	UErrorCode * err);
sl@0	938
sl@0	939	/**
sl@0	940	* Converts an array of unicode characters to an array of codepage
sl@0	941	* characters. This function is optimized for converting a continuous
sl@0	942	* stream of data in buffer-sized chunks, where the entire source and
sl@0	943	* target does not fit in available buffers.
sl@0	944	*
sl@0	945	* The source pointer is an in/out parameter. It starts out pointing where the
sl@0	946	* conversion is to begin, and ends up pointing after the last UChar consumed.
sl@0	947	*
sl@0	948	* Target similarly starts out pointer at the first available byte in the output
sl@0	949	* buffer, and ends up pointing after the last byte written to the output.
sl@0	950	*
sl@0	951	* The converter always attempts to consume the entire source buffer, unless
sl@0	952	* (1.) the target buffer is full, or (2.) a failing error is returned from the
sl@0	953	* current callback function. When a successful error status has been
sl@0	954	* returned, it means that all of the source buffer has been
sl@0	955	* consumed. At that point, the caller should reset the source and
sl@0	956	* sourceLimit pointers to point to the next chunk.
sl@0	957	*
sl@0	958	* At the end of the stream (flush==TRUE), the input is completely consumed
sl@0	959	* when *source==sourceLimit and no error code is set.
sl@0	960	* The converter object is then automatically reset by this function.
sl@0	961	* (This means that a converter need not be reset explicitly between data
sl@0	962	* streams if it finishes the previous stream without errors.)
sl@0	963	*
sl@0	964	* This is a <I>stateful</I> conversion. Additionally, even when all source data has
sl@0	965	* been consumed, some data may be in the converters' internal state.
sl@0	966	* Call this function repeatedly, updating the target pointers with
sl@0	967	* the next empty chunk of target in case of a
sl@0	968	* <TT>U_BUFFER_OVERFLOW_ERROR</TT>, and updating the source pointers
sl@0	969	* with the next chunk of source when a successful error status is
sl@0	970	* returned, until there are no more chunks of source data.
sl@0	971	* @param converter the Unicode converter
sl@0	972	* @param target I/O parameter. Input : Points to the beginning of the buffer to copy
sl@0	973	* codepage characters to. Output : points to after the last codepage character copied
sl@0	974	* to <TT>target</TT>.
sl@0	975	* @param targetLimit the pointer just after last of the <TT>target</TT> buffer
sl@0	976	* @param source I/O parameter, pointer to pointer to the source Unicode character buffer.
sl@0	977	* @param sourceLimit the pointer just after the last of the source buffer
sl@0	978	* @param offsets if NULL is passed, nothing will happen to it, otherwise it needs to have the same number
sl@0	979	* of allocated cells as <TT>target</TT>. Will fill in offsets from target to source pointer
sl@0	980	* 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>
sl@0	981	* For output data carried across calls, and other data without a specific source character
sl@0	982	* (such as from escape sequences or callbacks) -1 will be placed for offsets.
sl@0	983	* @param flush set to <TT>TRUE</TT> if the current source buffer is the last available
sl@0	984	* chunk of the source, <TT>FALSE</TT> otherwise. Note that if a failing status is returned,
sl@0	985	* this function may have to be called multiple times with flush set to <TT>TRUE</TT> until
sl@0	986	* the source buffer is consumed.
sl@0	987	* @param err the error status. <TT>U_ILLEGAL_ARGUMENT_ERROR</TT> will be set if the
sl@0	988	* converter is <TT>NULL</TT>.
sl@0	989	* <code>U_BUFFER_OVERFLOW_ERROR</code> will be set if the target is full and there is
sl@0	990	* still data to be written to the target.
sl@0	991	* @see ucnv_fromUChars
sl@0	992	* @see ucnv_convert
sl@0	993	* @see ucnv_getMinCharSize
sl@0	994	* @see ucnv_setToUCallBack
sl@0	995	* @stable ICU 2.0
sl@0	996	*/
sl@0	997	U_STABLE void U_EXPORT2
sl@0	998	ucnv_fromUnicode (UConverter * converter,
sl@0	999	char **target,
sl@0	1000	const char *targetLimit,
sl@0	1001	const UChar ** source,
sl@0	1002	const UChar * sourceLimit,
sl@0	1003	int32_t* offsets,
sl@0	1004	UBool flush,
sl@0	1005	UErrorCode * err);
sl@0	1006
sl@0	1007	/**
sl@0	1008	* Converts a buffer of codepage bytes into an array of unicode UChars
sl@0	1009	* characters. This function is optimized for converting a continuous
sl@0	1010	* stream of data in buffer-sized chunks, where the entire source and
sl@0	1011	* target does not fit in available buffers.
sl@0	1012	*
sl@0	1013	* The source pointer is an in/out parameter. It starts out pointing where the
sl@0	1014	* conversion is to begin, and ends up pointing after the last byte of source consumed.
sl@0	1015	*
sl@0	1016	* Target similarly starts out pointer at the first available UChar in the output
sl@0	1017	* buffer, and ends up pointing after the last UChar written to the output.
sl@0	1018	* It does NOT necessarily keep UChar sequences together.
sl@0	1019	*
sl@0	1020	* The converter always attempts to consume the entire source buffer, unless
sl@0	1021	* (1.) the target buffer is full, or (2.) a failing error is returned from the
sl@0	1022	* current callback function. When a successful error status has been
sl@0	1023	* returned, it means that all of the source buffer has been
sl@0	1024	* consumed. At that point, the caller should reset the source and
sl@0	1025	* sourceLimit pointers to point to the next chunk.
sl@0	1026	*
sl@0	1027	* At the end of the stream (flush==TRUE), the input is completely consumed
sl@0	1028	* when *source==sourceLimit and no error code is set
sl@0	1029	* The converter object is then automatically reset by this function.
sl@0	1030	* (This means that a converter need not be reset explicitly between data
sl@0	1031	* streams if it finishes the previous stream without errors.)
sl@0	1032	*
sl@0	1033	* This is a <I>stateful</I> conversion. Additionally, even when all source data has
sl@0	1034	* been consumed, some data may be in the converters' internal state.
sl@0	1035	* Call this function repeatedly, updating the target pointers with
sl@0	1036	* the next empty chunk of target in case of a
sl@0	1037	* <TT>U_BUFFER_OVERFLOW_ERROR</TT>, and updating the source pointers
sl@0	1038	* with the next chunk of source when a successful error status is
sl@0	1039	* returned, until there are no more chunks of source data.
sl@0	1040	* @param converter the Unicode converter
sl@0	1041	* @param target I/O parameter. Input : Points to the beginning of the buffer to copy
sl@0	1042	* UChars into. Output : points to after the last UChar copied.
sl@0	1043	* @param targetLimit the pointer just after the end of the <TT>target</TT> buffer
sl@0	1044	* @param source I/O parameter, pointer to pointer to the source codepage buffer.
sl@0	1045	* @param sourceLimit the pointer to the byte after the end of the source buffer
sl@0	1046	* @param offsets if NULL is passed, nothing will happen to it, otherwise it needs to have the same number
sl@0	1047	* of allocated cells as <TT>target</TT>. Will fill in offsets from target to source pointer
sl@0	1048	* 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>
sl@0	1049	* For output data carried across calls, and other data without a specific source character
sl@0	1050	* (such as from escape sequences or callbacks) -1 will be placed for offsets.
sl@0	1051	* @param flush set to <TT>TRUE</TT> if the current source buffer is the last available
sl@0	1052	* chunk of the source, <TT>FALSE</TT> otherwise. Note that if a failing status is returned,
sl@0	1053	* this function may have to be called multiple times with flush set to <TT>TRUE</TT> until
sl@0	1054	* the source buffer is consumed.
sl@0	1055	* @param err the error status. <TT>U_ILLEGAL_ARGUMENT_ERROR</TT> will be set if the
sl@0	1056	* converter is <TT>NULL</TT>.
sl@0	1057	* <code>U_BUFFER_OVERFLOW_ERROR</code> will be set if the target is full and there is
sl@0	1058	* still data to be written to the target.
sl@0	1059	* @see ucnv_fromUChars
sl@0	1060	* @see ucnv_convert
sl@0	1061	* @see ucnv_getMinCharSize
sl@0	1062	* @see ucnv_setFromUCallBack
sl@0	1063	* @see ucnv_getNextUChar
sl@0	1064	* @stable ICU 2.0
sl@0	1065	*/
sl@0	1066	U_STABLE void U_EXPORT2
sl@0	1067	ucnv_toUnicode(UConverter *converter,
sl@0	1068	UChar **target,
sl@0	1069	const UChar *targetLimit,
sl@0	1070	const char **source,
sl@0	1071	const char *sourceLimit,
sl@0	1072	int32_t *offsets,
sl@0	1073	UBool flush,
sl@0	1074	UErrorCode *err);
sl@0	1075
sl@0	1076	/**
sl@0	1077	* Convert the Unicode string into a codepage string using an existing UConverter.
sl@0	1078	* The output string is NUL-terminated if possible.
sl@0	1079	*
sl@0	1080	* This function is a more convenient but less powerful version of ucnv_fromUnicode().
sl@0	1081	* It is only useful for whole strings, not for streaming conversion.
sl@0	1082	*
sl@0	1083	* The maximum output buffer capacity required (barring output from callbacks) will be
sl@0	1084	* UCNV_GET_MAX_BYTES_FOR_STRING(srcLength, ucnv_getMaxCharSize(cnv)).
sl@0	1085	*
sl@0	1086	* @param cnv the converter object to be used (ucnv_resetFromUnicode() will be called)
sl@0	1087	* @param src the input Unicode string
sl@0	1088	* @param srcLength the input string length, or -1 if NUL-terminated
sl@0	1089	* @param dest destination string buffer, can be NULL if destCapacity==0
sl@0	1090	* @param destCapacity the number of chars available at dest
sl@0	1091	* @param pErrorCode normal ICU error code;
sl@0	1092	* common error codes that may be set by this function include
sl@0	1093	* U_BUFFER_OVERFLOW_ERROR, U_STRING_NOT_TERMINATED_WARNING,
sl@0	1094	* U_ILLEGAL_ARGUMENT_ERROR, and conversion errors
sl@0	1095	* @return the length of the output string, not counting the terminating NUL;
sl@0	1096	* if the length is greater than destCapacity, then the string will not fit
sl@0	1097	* and a buffer of the indicated length would need to be passed in
sl@0	1098	* @see ucnv_fromUnicode
sl@0	1099	* @see ucnv_convert
sl@0	1100	* @see UCNV_GET_MAX_BYTES_FOR_STRING
sl@0	1101	* @stable ICU 2.0
sl@0	1102	*/
sl@0	1103	U_STABLE int32_t U_EXPORT2
sl@0	1104	ucnv_fromUChars(UConverter *cnv,
sl@0	1105	char *dest, int32_t destCapacity,
sl@0	1106	const UChar *src, int32_t srcLength,
sl@0	1107	UErrorCode *pErrorCode);
sl@0	1108
sl@0	1109	/**
sl@0	1110	* Convert the codepage string into a Unicode string using an existing UConverter.
sl@0	1111	* The output string is NUL-terminated if possible.
sl@0	1112	*
sl@0	1113	* This function is a more convenient but less powerful version of ucnv_toUnicode().
sl@0	1114	* It is only useful for whole strings, not for streaming conversion.
sl@0	1115	*
sl@0	1116	* The maximum output buffer capacity required (barring output from callbacks) will be
sl@0	1117	* 2*srcLength (each char may be converted into a surrogate pair).
sl@0	1118	*
sl@0	1119	* @param cnv the converter object to be used (ucnv_resetToUnicode() will be called)
sl@0	1120	* @param src the input codepage string
sl@0	1121	* @param srcLength the input string length, or -1 if NUL-terminated
sl@0	1122	* @param dest destination string buffer, can be NULL if destCapacity==0
sl@0	1123	* @param destCapacity the number of UChars available at dest
sl@0	1124	* @param pErrorCode normal ICU error code;
sl@0	1125	* common error codes that may be set by this function include
sl@0	1126	* U_BUFFER_OVERFLOW_ERROR, U_STRING_NOT_TERMINATED_WARNING,
sl@0	1127	* U_ILLEGAL_ARGUMENT_ERROR, and conversion errors
sl@0	1128	* @return the length of the output string, not counting the terminating NUL;
sl@0	1129	* if the length is greater than destCapacity, then the string will not fit
sl@0	1130	* and a buffer of the indicated length would need to be passed in
sl@0	1131	* @see ucnv_toUnicode
sl@0	1132	* @see ucnv_convert
sl@0	1133	* @stable ICU 2.0
sl@0	1134	*/
sl@0	1135	U_STABLE int32_t U_EXPORT2
sl@0	1136	ucnv_toUChars(UConverter *cnv,
sl@0	1137	UChar *dest, int32_t destCapacity,
sl@0	1138	const char *src, int32_t srcLength,
sl@0	1139	UErrorCode *pErrorCode);
sl@0	1140
sl@0	1141	/**
sl@0	1142	* Convert a codepage buffer into Unicode one character at a time.
sl@0	1143	* The input is completely consumed when the U_INDEX_OUTOFBOUNDS_ERROR is set.
sl@0	1144	*
sl@0	1145	* Advantage compared to ucnv_toUnicode() or ucnv_toUChars():
sl@0	1146	* - Faster for small amounts of data, for most converters, e.g.,
sl@0	1147	* US-ASCII, ISO-8859-1, UTF-8/16/32, and most "normal" charsets.
sl@0	1148	* (For complex converters, e.g., SCSU, UTF-7 and ISO 2022 variants,
sl@0	1149	* it uses ucnv_toUnicode() internally.)
sl@0	1150	* - Convenient.
sl@0	1151	*
sl@0	1152	* Limitations compared to ucnv_toUnicode():
sl@0	1153	* - Always assumes flush=TRUE.
sl@0	1154	* This makes ucnv_getNextUChar() unsuitable for "streaming" conversion,
sl@0	1155	* that is, for where the input is supplied in multiple buffers,
sl@0	1156	* because ucnv_getNextUChar() will assume the end of the input at the end
sl@0	1157	* of the first buffer.
sl@0	1158	* - Does not provide offset output.
sl@0	1159	*
sl@0	1160	* It is possible to "mix" ucnv_getNextUChar() and ucnv_toUnicode() because
sl@0	1161	* ucnv_getNextUChar() uses the current state of the converter
sl@0	1162	* (unlike ucnv_toUChars() which always resets first).
sl@0	1163	* However, if ucnv_getNextUChar() is called after ucnv_toUnicode()
sl@0	1164	* stopped in the middle of a character sequence (with flush=FALSE),
sl@0	1165	* then ucnv_getNextUChar() will always use the slower ucnv_toUnicode()
sl@0	1166	* internally until the next character boundary.
sl@0	1167	* (This is new in ICU 2.6. In earlier releases, ucnv_getNextUChar() had to
sl@0	1168	* start at a character boundary.)
sl@0	1169	*
sl@0	1170	* Instead of using ucnv_getNextUChar(), it is recommended
sl@0	1171	* to convert using ucnv_toUnicode() or ucnv_toUChars()
sl@0	1172	* and then iterate over the text using U16_NEXT() or a UCharIterator (uiter.h)
sl@0	1173	* or a C++ CharacterIterator or similar.
sl@0	1174	* This allows streaming conversion and offset output, for example.
sl@0	1175	*
sl@0	1176	* <p>Handling of surrogate pairs and supplementary-plane code points:<br>
sl@0	1177	* There are two different kinds of codepages that provide mappings for surrogate characters:
sl@0	1178	* <ul>
sl@0	1179	* <li>Codepages like UTF-8, UTF-32, and GB 18030 provide direct representations for Unicode
sl@0	1180	* code points U+10000-U+10ffff as well as for single surrogates U+d800-U+dfff.
sl@0	1181	* Each valid sequence will result in exactly one returned code point.
sl@0	1182	* If a sequence results in a single surrogate, then that will be returned
sl@0	1183	* by itself, even if a neighboring sequence encodes the matching surrogate.</li>
sl@0	1184	* <li>Codepages like SCSU and LMBCS (and UTF-16) provide direct representations only for BMP code points
sl@0	1185	* including surrogates. Code points in supplementary planes are represented with
sl@0	1186	* two sequences, each encoding a surrogate.
sl@0	1187	* For these codepages, matching pairs of surrogates will be combined into single
sl@0	1188	* code points for returning from this function.
sl@0	1189	* (Note that SCSU is actually a mix of these codepage types.)</li>
sl@0	1190	* </ul></p>
sl@0	1191	*
sl@0	1192	* @param converter an open UConverter
sl@0	1193	* @param source the address of a pointer to the codepage buffer, will be
sl@0	1194	* updated to point after the bytes consumed in the conversion call.
sl@0	1195	* @param sourceLimit points to the end of the input buffer
sl@0	1196	* @param err fills in error status (see ucnv_toUnicode)
sl@0	1197	* <code>U_INDEX_OUTOFBOUNDS_ERROR</code> will be set if the input
sl@0	1198	* is empty or does not convert to any output (e.g.: pure state-change
sl@0	1199	* codes SI/SO, escape sequences for ISO 2022,
sl@0	1200	* or if the callback did not output anything, ...).
sl@0	1201	* This function will not set a <code>U_BUFFER_OVERFLOW_ERROR</code> because
sl@0	1202	* the "buffer" is the return code. However, there might be subsequent output
sl@0	1203	* stored in the converter object
sl@0	1204	* that will be returned in following calls to this function.
sl@0	1205	* @return a UChar32 resulting from the partial conversion of source
sl@0	1206	* @see ucnv_toUnicode
sl@0	1207	* @see ucnv_toUChars
sl@0	1208	* @see ucnv_convert
sl@0	1209	* @stable ICU 2.0
sl@0	1210	*/
sl@0	1211	U_STABLE UChar32 U_EXPORT2
sl@0	1212	ucnv_getNextUChar(UConverter * converter,
sl@0	1213	const char **source,
sl@0	1214	const char * sourceLimit,
sl@0	1215	UErrorCode * err);
sl@0	1216
sl@0	1217	/**
sl@0	1218	* Convert from one external charset to another using two existing UConverters.
sl@0	1219	* Internally, two conversions - ucnv_toUnicode() and ucnv_fromUnicode() -
sl@0	1220	* are used, "pivoting" through 16-bit Unicode.
sl@0	1221	*
sl@0	1222	* There is a similar function, ucnv_convert(),
sl@0	1223	* which has the following limitations:
sl@0	1224	* - it takes charset names, not converter objects, so that
sl@0	1225	* - two converters are opened for each call
sl@0	1226	* - only single-string conversion is possible, not streaming operation
sl@0	1227	* - it does not provide enough information to find out,
sl@0	1228	* in case of failure, whether the toUnicode or
sl@0	1229	* the fromUnicode conversion failed
sl@0	1230	*
sl@0	1231	* By contrast, ucnv_convertEx()
sl@0	1232	* - takes UConverter parameters instead of charset names
sl@0	1233	* - fully exposes the pivot buffer for complete error handling
sl@0	1234	*
sl@0	1235	* ucnv_convertEx() also provides further convenience:
sl@0	1236	* - an option to reset the converters at the beginning
sl@0	1237	* (if reset==TRUE, see parameters;
sl@0	1238	* also sets pivotTarget=pivotSource=pivotStart)
sl@0	1239	* - allow NUL-terminated input
sl@0	1240	* (only a single NUL byte, will not work for charsets with multi-byte NULs)
sl@0	1241	* (if sourceLimit==NULL, see parameters)
sl@0	1242	* - terminate with a NUL on output
sl@0	1243	* (only a single NUL byte, not useful for charsets with multi-byte NULs),
sl@0	1244	* or set U_STRING_NOT_TERMINATED_WARNING if the output exactly fills
sl@0	1245	* the target buffer
sl@0	1246	* - the pivot buffer can be provided internally;
sl@0	1247	* in this case, the caller will not be able to get details about where an
sl@0	1248	* error occurred
sl@0	1249	* (if pivotStart==NULL, see below)
sl@0	1250	*
sl@0	1251	* The function returns when one of the following is true:
sl@0	1252	* - the entire source text has been converted successfully to the target buffer
sl@0	1253	* - a target buffer overflow occurred (U_BUFFER_OVERFLOW_ERROR)
sl@0	1254	* - a conversion error occurred
sl@0	1255	* (other U_FAILURE(), see description of pErrorCode)
sl@0	1256	*
sl@0	1257	* Limitation compared to the direct use of
sl@0	1258	* ucnv_fromUnicode() and ucnv_toUnicode():
sl@0	1259	* ucnv_convertEx() does not provide offset information.
sl@0	1260	*
sl@0	1261	* Limitation compared to ucnv_fromUChars() and ucnv_toUChars():
sl@0	1262	* ucnv_convertEx() does not support preflighting directly.
sl@0	1263	*
sl@0	1264	* Sample code for converting a single string from
sl@0	1265	* one external charset to UTF-8, ignoring the location of errors:
sl@0	1266	*
sl@0	1267	* \code
sl@0	1268	* int32_t
sl@0	1269	* myToUTF8(UConverter *cnv,
sl@0	1270	* const char *s, int32_t length,
sl@0	1271	* char *u8, int32_t capacity,
sl@0	1272	* UErrorCode *pErrorCode) {
sl@0	1273	* UConverter *utf8Cnv;
sl@0	1274	* char *target;
sl@0	1275	*
sl@0	1276	* if(U_FAILURE(*pErrorCode)) {
sl@0	1277	* return 0;
sl@0	1278	* }
sl@0	1279	*
sl@0	1280	* utf8Cnv=myGetCachedUTF8Converter(pErrorCode);
sl@0	1281	* if(U_FAILURE(*pErrorCode)) {
sl@0	1282	* return 0;
sl@0	1283	* }
sl@0	1284	*
sl@0	1285	* target=u8;
sl@0	1286	* ucnv_convertEx(cnv, utf8Cnv,
sl@0	1287	* &target, u8+capacity,
sl@0	1288	* &s, length>=0 ? s+length : NULL,
sl@0	1289	* NULL, NULL, NULL, NULL,
sl@0	1290	* TRUE, TRUE,
sl@0	1291	* pErrorCode);
sl@0	1292	*
sl@0	1293	* myReleaseCachedUTF8Converter(utf8Cnv);
sl@0	1294	*
sl@0	1295	* // return the output string length, but without preflighting
sl@0	1296	* return (int32_t)(target-u8);
sl@0	1297	* }
sl@0	1298	* \endcode
sl@0	1299	*
sl@0	1300	* @param targetCnv Output converter, used to convert from the UTF-16 pivot
sl@0	1301	* to the target using ucnv_fromUnicode().
sl@0	1302	* @param sourceCnv Input converter, used to convert from the source to
sl@0	1303	* the UTF-16 pivot using ucnv_toUnicode().
sl@0	1304	* @param target I/O parameter, same as for ucnv_fromUChars().
sl@0	1305	* Input: *target points to the beginning of the target buffer.
sl@0	1306	* Output: *target points to the first unit after the last char written.
sl@0	1307	* @param targetLimit Pointer to the first unit after the target buffer.
sl@0	1308	* @param source I/O parameter, same as for ucnv_toUChars().
sl@0	1309	* Input: *source points to the beginning of the source buffer.
sl@0	1310	* Output: *source points to the first unit after the last char read.
sl@0	1311	* @param sourceLimit Pointer to the first unit after the source buffer.
sl@0	1312	* @param pivotStart Pointer to the UTF-16 pivot buffer. If pivotStart==NULL,
sl@0	1313	* then an internal buffer is used and the other pivot
sl@0	1314	* arguments are ignored and can be NULL as well.
sl@0	1315	* @param pivotSource I/O parameter, same as source in ucnv_fromUChars() for
sl@0	1316	* conversion from the pivot buffer to the target buffer.
sl@0	1317	* @param pivotTarget I/O parameter, same as target in ucnv_toUChars() for
sl@0	1318	* conversion from the source buffer to the pivot buffer.
sl@0	1319	* It must be pivotStart<=pivotSource<=pivotTarget<=pivotLimit
sl@0	1320	* and pivotStart<pivotLimit (unless pivotStart==NULL).
sl@0	1321	* @param pivotLimit Pointer to the first unit after the pivot buffer.
sl@0	1322	* @param reset If TRUE, then ucnv_resetToUnicode(sourceCnv) and
sl@0	1323	* ucnv_resetFromUnicode(targetCnv) are called, and the
sl@0	1324	* pivot pointers are reset (pivotTarget=pivotSource=pivotStart).
sl@0	1325	* @param flush If true, indicates the end of the input.
sl@0	1326	* Passed directly to ucnv_toUnicode(), and carried over to
sl@0	1327	* ucnv_fromUnicode() when the source is empty as well.
sl@0	1328	* @param pErrorCode ICU error code in/out parameter.
sl@0	1329	* Must fulfill U_SUCCESS before the function call.
sl@0	1330	* U_BUFFER_OVERFLOW_ERROR always refers to the target buffer
sl@0	1331	* because overflows into the pivot buffer are handled internally.
sl@0	1332	* Other conversion errors are from the source-to-pivot
sl@0	1333	* conversion if *pivotSource==pivotStart, otherwise from
sl@0	1334	* the pivot-to-target conversion.
sl@0	1335	*
sl@0	1336	* @see ucnv_convert
sl@0	1337	* @see ucnv_fromAlgorithmic
sl@0	1338	* @see ucnv_toAlgorithmic
sl@0	1339	* @see ucnv_fromUnicode
sl@0	1340	* @see ucnv_toUnicode
sl@0	1341	* @see ucnv_fromUChars
sl@0	1342	* @see ucnv_toUChars
sl@0	1343	* @stable ICU 2.6
sl@0	1344	*/
sl@0	1345	U_STABLE void U_EXPORT2
sl@0	1346	ucnv_convertEx(UConverter targetCnv, UConverter sourceCnv,
sl@0	1347	char *target, const char targetLimit,
sl@0	1348	const char *source, const char sourceLimit,
sl@0	1349	UChar pivotStart, UChar *pivotSource,
sl@0	1350	UChar *pivotTarget, const UChar pivotLimit,
sl@0	1351	UBool reset, UBool flush,
sl@0	1352	UErrorCode *pErrorCode);
sl@0	1353
sl@0	1354	/**
sl@0	1355	* Convert from one external charset to another.
sl@0	1356	* Internally, two converters are opened according to the name arguments,
sl@0	1357	* then the text is converted to and from the 16-bit Unicode "pivot"
sl@0	1358	* using ucnv_convertEx(), then the converters are closed again.
sl@0	1359	*
sl@0	1360	* This is a convenience function, not an efficient way to convert a lot of text:
sl@0	1361	* ucnv_convert()
sl@0	1362	* - takes charset names, not converter objects, so that
sl@0	1363	* - two converters are opened for each call
sl@0	1364	* - only single-string conversion is possible, not streaming operation
sl@0	1365	* - does not provide enough information to find out,
sl@0	1366	* in case of failure, whether the toUnicode or
sl@0	1367	* the fromUnicode conversion failed
sl@0	1368	* - allows NUL-terminated input
sl@0	1369	* (only a single NUL byte, will not work for charsets with multi-byte NULs)
sl@0	1370	* (if sourceLength==-1, see parameters)
sl@0	1371	* - terminate with a NUL on output
sl@0	1372	* (only a single NUL byte, not useful for charsets with multi-byte NULs),
sl@0	1373	* or set U_STRING_NOT_TERMINATED_WARNING if the output exactly fills
sl@0	1374	* the target buffer
sl@0	1375	* - a pivot buffer is provided internally
sl@0	1376	*
sl@0	1377	* The function returns when one of the following is true:
sl@0	1378	* - the entire source text has been converted successfully to the target buffer
sl@0	1379	* and either the target buffer is terminated with a single NUL byte
sl@0	1380	* or the error code is set to U_STRING_NOT_TERMINATED_WARNING
sl@0	1381	* - a target buffer overflow occurred (U_BUFFER_OVERFLOW_ERROR)
sl@0	1382	* and the full output string length is returned ("preflighting")
sl@0	1383	* - a conversion error occurred
sl@0	1384	* (other U_FAILURE(), see description of pErrorCode)
sl@0	1385	*
sl@0	1386	* @param toConverterName The name of the converter that is used to convert
sl@0	1387	* from the UTF-16 pivot buffer to the target.
sl@0	1388	* @param fromConverterName The name of the converter that is used to convert
sl@0	1389	* from the source to the UTF-16 pivot buffer.
sl@0	1390	* @param target Pointer to the output buffer.
sl@0	1391	* @param targetCapacity Capacity of the target, in bytes.
sl@0	1392	* @param source Pointer to the input buffer.
sl@0	1393	* @param sourceLength Length of the input text, in bytes, or -1 for NUL-terminated input.
sl@0	1394	* @param pErrorCode ICU error code in/out parameter.
sl@0	1395	* Must fulfill U_SUCCESS before the function call.
sl@0	1396	* @return Length of the complete output text in bytes, even if it exceeds the targetCapacity
sl@0	1397	* and a U_BUFFER_OVERFLOW_ERROR is set.
sl@0	1398	*
sl@0	1399	* @see ucnv_convertEx
sl@0	1400	* @see ucnv_fromAlgorithmic
sl@0	1401	* @see ucnv_toAlgorithmic
sl@0	1402	* @see ucnv_fromUnicode
sl@0	1403	* @see ucnv_toUnicode
sl@0	1404	* @see ucnv_fromUChars
sl@0	1405	* @see ucnv_toUChars
sl@0	1406	* @see ucnv_getNextUChar
sl@0	1407	* @stable ICU 2.0
sl@0	1408	*/
sl@0	1409	U_STABLE int32_t U_EXPORT2
sl@0	1410	ucnv_convert(const char *toConverterName,
sl@0	1411	const char *fromConverterName,
sl@0	1412	char *target,
sl@0	1413	int32_t targetCapacity,
sl@0	1414	const char *source,
sl@0	1415	int32_t sourceLength,
sl@0	1416	UErrorCode *pErrorCode);
sl@0	1417
sl@0	1418	/**
sl@0	1419	* Convert from one external charset to another.
sl@0	1420	* Internally, the text is converted to and from the 16-bit Unicode "pivot"
sl@0	1421	* using ucnv_convertEx(). ucnv_toAlgorithmic() works exactly like ucnv_convert()
sl@0	1422	* except that the two converters need not be looked up and opened completely.
sl@0	1423	*
sl@0	1424	* The source-to-pivot conversion uses the cnv converter parameter.
sl@0	1425	* The pivot-to-target conversion uses a purely algorithmic converter
sl@0	1426	* according to the specified type, e.g., UCNV_UTF8 for a UTF-8 converter.
sl@0	1427	*
sl@0	1428	* Internally, the algorithmic converter is opened and closed for each
sl@0	1429	* function call, which is more efficient than using the public ucnv_open()
sl@0	1430	* but somewhat less efficient than only resetting an existing converter
sl@0	1431	* and using ucnv_convertEx().
sl@0	1432	*
sl@0	1433	* This function is more convenient than ucnv_convertEx() for single-string
sl@0	1434	* conversions, especially when "preflighting" is desired (returning the length
sl@0	1435	* of the complete output even if it does not fit into the target buffer;
sl@0	1436	* see the User Guide Strings chapter). See ucnv_convert() for details.
sl@0	1437	*
sl@0	1438	* @param algorithmicType UConverterType constant identifying the desired target
sl@0	1439	* charset as a purely algorithmic converter.
sl@0	1440	* Those are converters for Unicode charsets like
sl@0	1441	* UTF-8, BOCU-1, SCSU, UTF-7, IMAP-mailbox-name, etc.,
sl@0	1442	* as well as US-ASCII and ISO-8859-1.
sl@0	1443	* @param cnv The converter that is used to convert
sl@0	1444	* from the source to the UTF-16 pivot buffer.
sl@0	1445	* @param target Pointer to the output buffer.
sl@0	1446	* @param targetCapacity Capacity of the target, in bytes.
sl@0	1447	* @param source Pointer to the input buffer.
sl@0	1448	* @param sourceLength Length of the input text, in bytes
sl@0	1449	* @param pErrorCode ICU error code in/out parameter.
sl@0	1450	* Must fulfill U_SUCCESS before the function call.
sl@0	1451	* @return Length of the complete output text in bytes, even if it exceeds the targetCapacity
sl@0	1452	* and a U_BUFFER_OVERFLOW_ERROR is set.
sl@0	1453	*
sl@0	1454	* @see ucnv_fromAlgorithmic
sl@0	1455	* @see ucnv_convert
sl@0	1456	* @see ucnv_convertEx
sl@0	1457	* @see ucnv_fromUnicode
sl@0	1458	* @see ucnv_toUnicode
sl@0	1459	* @see ucnv_fromUChars
sl@0	1460	* @see ucnv_toUChars
sl@0	1461	* @stable ICU 2.6
sl@0	1462	*/
sl@0	1463	U_STABLE int32_t U_EXPORT2
sl@0	1464	ucnv_toAlgorithmic(UConverterType algorithmicType,
sl@0	1465	UConverter *cnv,
sl@0	1466	char *target, int32_t targetCapacity,
sl@0	1467	const char *source, int32_t sourceLength,
sl@0	1468	UErrorCode *pErrorCode);
sl@0	1469
sl@0	1470	/**
sl@0	1471	* Convert from one external charset to another.
sl@0	1472	* Internally, the text is converted to and from the 16-bit Unicode "pivot"
sl@0	1473	* using ucnv_convertEx(). ucnv_fromAlgorithmic() works exactly like ucnv_convert()
sl@0	1474	* except that the two converters need not be looked up and opened completely.
sl@0	1475	*
sl@0	1476	* The source-to-pivot conversion uses a purely algorithmic converter
sl@0	1477	* according to the specified type, e.g., UCNV_UTF8 for a UTF-8 converter.
sl@0	1478	* The pivot-to-target conversion uses the cnv converter parameter.
sl@0	1479	*
sl@0	1480	* Internally, the algorithmic converter is opened and closed for each
sl@0	1481	* function call, which is more efficient than using the public ucnv_open()
sl@0	1482	* but somewhat less efficient than only resetting an existing converter
sl@0	1483	* and using ucnv_convertEx().
sl@0	1484	*
sl@0	1485	* This function is more convenient than ucnv_convertEx() for single-string
sl@0	1486	* conversions, especially when "preflighting" is desired (returning the length
sl@0	1487	* of the complete output even if it does not fit into the target buffer;
sl@0	1488	* see the User Guide Strings chapter). See ucnv_convert() for details.
sl@0	1489	*
sl@0	1490	* @param cnv The converter that is used to convert
sl@0	1491	* from the UTF-16 pivot buffer to the target.
sl@0	1492	* @param algorithmicType UConverterType constant identifying the desired source
sl@0	1493	* charset as a purely algorithmic converter.
sl@0	1494	* Those are converters for Unicode charsets like
sl@0	1495	* UTF-8, BOCU-1, SCSU, UTF-7, IMAP-mailbox-name, etc.,
sl@0	1496	* as well as US-ASCII and ISO-8859-1.
sl@0	1497	* @param target Pointer to the output buffer.
sl@0	1498	* @param targetCapacity Capacity of the target, in bytes.
sl@0	1499	* @param source Pointer to the input buffer.
sl@0	1500	* @param sourceLength Length of the input text, in bytes
sl@0	1501	* @param pErrorCode ICU error code in/out parameter.
sl@0	1502	* Must fulfill U_SUCCESS before the function call.
sl@0	1503	* @return Length of the complete output text in bytes, even if it exceeds the targetCapacity
sl@0	1504	* and a U_BUFFER_OVERFLOW_ERROR is set.
sl@0	1505	*
sl@0	1506	* @see ucnv_fromAlgorithmic
sl@0	1507	* @see ucnv_convert
sl@0	1508	* @see ucnv_convertEx
sl@0	1509	* @see ucnv_fromUnicode
sl@0	1510	* @see ucnv_toUnicode
sl@0	1511	* @see ucnv_fromUChars
sl@0	1512	* @see ucnv_toUChars
sl@0	1513	* @stable ICU 2.6
sl@0	1514	*/
sl@0	1515	U_STABLE int32_t U_EXPORT2
sl@0	1516	ucnv_fromAlgorithmic(UConverter *cnv,
sl@0	1517	UConverterType algorithmicType,
sl@0	1518	char *target, int32_t targetCapacity,
sl@0	1519	const char *source, int32_t sourceLength,
sl@0	1520	UErrorCode *pErrorCode);
sl@0	1521
sl@0	1522	/**
sl@0	1523	* Frees up memory occupied by unused, cached converter shared data.
sl@0	1524	*
sl@0	1525	* @return the number of cached converters successfully deleted
sl@0	1526	* @see ucnv_close
sl@0	1527	* @stable ICU 2.0
sl@0	1528	*/
sl@0	1529	U_STABLE int32_t U_EXPORT2
sl@0	1530	ucnv_flushCache(void);
sl@0	1531
sl@0	1532	/**
sl@0	1533	* Returns the number of available converters, as per the alias file.
sl@0	1534	*
sl@0	1535	* @return the number of available converters
sl@0	1536	* @see ucnv_getAvailableName
sl@0	1537	* @stable ICU 2.0
sl@0	1538	*/
sl@0	1539	U_STABLE int32_t U_EXPORT2
sl@0	1540	ucnv_countAvailable(void);
sl@0	1541
sl@0	1542	/**
sl@0	1543	* Gets the canonical converter name of the specified converter from a list of
sl@0	1544	* all available converters contaied in the alias file. All converters
sl@0	1545	* in this list can be opened.
sl@0	1546	*
sl@0	1547	* @param n the index to a converter available on the system (in the range <TT>[0..ucnv_countAvaiable()]</TT>)
sl@0	1548	* @return a pointer a string (library owned), or <TT>NULL</TT> if the index is out of bounds.
sl@0	1549	* @see ucnv_countAvailable
sl@0	1550	* @stable ICU 2.0
sl@0	1551	*/
sl@0	1552	U_STABLE const char* U_EXPORT2
sl@0	1553	ucnv_getAvailableName(int32_t n);
sl@0	1554
sl@0	1555	/**
sl@0	1556	* Returns a UEnumeration to enumerate all of the canonical converter
sl@0	1557	* names, as per the alias file, regardless of the ability to open each
sl@0	1558	* converter.
sl@0	1559	*
sl@0	1560	* @return A UEnumeration object for getting all the recognized canonical
sl@0	1561	* converter names.
sl@0	1562	* @see ucnv_getAvailableName
sl@0	1563	* @see uenum_close
sl@0	1564	* @see uenum_next
sl@0	1565	* @stable ICU 2.4
sl@0	1566	*/
sl@0	1567	U_STABLE UEnumeration * U_EXPORT2
sl@0	1568	ucnv_openAllNames(UErrorCode *pErrorCode);
sl@0	1569
sl@0	1570	/**
sl@0	1571	* Gives the number of aliases for a given converter or alias name.
sl@0	1572	* If the alias is ambiguous, then the preferred converter is used
sl@0	1573	* and the status is set to U_AMBIGUOUS_ALIAS_WARNING.
sl@0	1574	* This method only enumerates the listed entries in the alias file.
sl@0	1575	* @param alias alias name
sl@0	1576	* @param pErrorCode error status
sl@0	1577	* @return number of names on alias list for given alias
sl@0	1578	* @stable ICU 2.0
sl@0	1579	*/
sl@0	1580	U_STABLE uint16_t U_EXPORT2
sl@0	1581	ucnv_countAliases(const char alias, UErrorCode pErrorCode);
sl@0	1582
sl@0	1583	/**
sl@0	1584	* Gives the name of the alias at given index of alias list.
sl@0	1585	* This method only enumerates the listed entries in the alias file.
sl@0	1586	* If the alias is ambiguous, then the preferred converter is used
sl@0	1587	* and the status is set to U_AMBIGUOUS_ALIAS_WARNING.
sl@0	1588	* @param alias alias name
sl@0	1589	* @param n index in alias list
sl@0	1590	* @param pErrorCode result of operation
sl@0	1591	* @return returns the name of the alias at given index
sl@0	1592	* @see ucnv_countAliases
sl@0	1593	* @stable ICU 2.0
sl@0	1594	*/
sl@0	1595	U_STABLE const char * U_EXPORT2
sl@0	1596	ucnv_getAlias(const char alias, uint16_t n, UErrorCode pErrorCode);
sl@0	1597
sl@0	1598	/**
sl@0	1599	* Fill-up the list of alias names for the given alias.
sl@0	1600	* This method only enumerates the listed entries in the alias file.
sl@0	1601	* If the alias is ambiguous, then the preferred converter is used
sl@0	1602	* and the status is set to U_AMBIGUOUS_ALIAS_WARNING.
sl@0	1603	* @param alias alias name
sl@0	1604	* @param aliases fill-in list, aliases is a pointer to an array of
sl@0	1605	* <code>ucnv_countAliases()</code> string-pointers
sl@0	1606	* (<code>const char *</code>) that will be filled in.
sl@0	1607	* The strings themselves are owned by the library.
sl@0	1608	* @param pErrorCode result of operation
sl@0	1609	* @stable ICU 2.0
sl@0	1610	*/
sl@0	1611	U_STABLE void U_EXPORT2
sl@0	1612	ucnv_getAliases(const char alias, const char aliases, UErrorCode pErrorCode);
sl@0	1613
sl@0	1614	/**
sl@0	1615	* Return a new UEnumeration object for enumerating all the
sl@0	1616	* alias names for a given converter that are recognized by a standard.
sl@0	1617	* This method only enumerates the listed entries in the alias file.
sl@0	1618	* The convrtrs.txt file can be modified to change the results of
sl@0	1619	* this function.
sl@0	1620	* The first result in this list is the same result given by
sl@0	1621	* <code>ucnv_getStandardName</code>, which is the default alias for
sl@0	1622	* the specified standard name. The returned object must be closed with
sl@0	1623	* <code>uenum_close</code> when you are done with the object.
sl@0	1624	*
sl@0	1625	* @param convName original converter name
sl@0	1626	* @param standard name of the standard governing the names; MIME and IANA
sl@0	1627	* are such standards
sl@0	1628	* @param pErrorCode The error code
sl@0	1629	* @return A UEnumeration object for getting all aliases that are recognized
sl@0	1630	* by a standard. If any of the parameters are invalid, NULL
sl@0	1631	* is returned.
sl@0	1632	* @see ucnv_getStandardName
sl@0	1633	* @see uenum_close
sl@0	1634	* @see uenum_next
sl@0	1635	* @stable ICU 2.2
sl@0	1636	*/
sl@0	1637	U_STABLE UEnumeration * U_EXPORT2
sl@0	1638	ucnv_openStandardNames(const char *convName,
sl@0	1639	const char *standard,
sl@0	1640	UErrorCode *pErrorCode);
sl@0	1641
sl@0	1642	/**
sl@0	1643	* Gives the number of standards associated to converter names.
sl@0	1644	* @return number of standards
sl@0	1645	* @stable ICU 2.0
sl@0	1646	*/
sl@0	1647	U_STABLE uint16_t U_EXPORT2
sl@0	1648	ucnv_countStandards(void);
sl@0	1649
sl@0	1650	/**
sl@0	1651	* Gives the name of the standard at given index of standard list.
sl@0	1652	* @param n index in standard list
sl@0	1653	* @param pErrorCode result of operation
sl@0	1654	* @return returns the name of the standard at given index. Owned by the library.
sl@0	1655	* @stable ICU 2.0
sl@0	1656	*/
sl@0	1657	U_STABLE const char * U_EXPORT2
sl@0	1658	ucnv_getStandard(uint16_t n, UErrorCode *pErrorCode);
sl@0	1659
sl@0	1660	/**
sl@0	1661	* Returns a standard name for a given converter name.
sl@0	1662	* <p>
sl@0	1663	* Example alias table:<br>
sl@0	1664	* conv alias1 { STANDARD1 } alias2 { STANDARD1* }
sl@0	1665	* <p>
sl@0	1666	* Result of ucnv_getStandardName("conv", "STANDARD1") from example
sl@0	1667	* alias table:<br>
sl@0	1668	* <b>"alias2"</b>
sl@0	1669	*
sl@0	1670	* @param name original converter name
sl@0	1671	* @param standard name of the standard governing the names; MIME and IANA
sl@0	1672	* are such standards
sl@0	1673	* @param pErrorCode result of operation
sl@0	1674	* @return returns the standard converter name;
sl@0	1675	* if a standard converter name cannot be determined,
sl@0	1676	* then <code>NULL</code> is returned. Owned by the library.
sl@0	1677	* @stable ICU 2.0
sl@0	1678	*/
sl@0	1679	U_STABLE const char * U_EXPORT2
sl@0	1680	ucnv_getStandardName(const char name, const char standard, UErrorCode *pErrorCode);
sl@0	1681
sl@0	1682	/**
sl@0	1683	* This function will return the internal canonical converter name of the
sl@0	1684	* tagged alias. This is the opposite of ucnv_openStandardNames, which
sl@0	1685	* returns the tagged alias given the canonical name.
sl@0	1686	* <p>
sl@0	1687	* Example alias table:<br>
sl@0	1688	* conv alias1 { STANDARD1 } alias2 { STANDARD1* }
sl@0	1689	* <p>
sl@0	1690	* Result of ucnv_getStandardName("alias1", "STANDARD1") from example
sl@0	1691	* alias table:<br>
sl@0	1692	* <b>"conv"</b>
sl@0	1693	*
sl@0	1694	* @return returns the canonical converter name;
sl@0	1695	* if a standard or alias name cannot be determined,
sl@0	1696	* then <code>NULL</code> is returned. The returned string is
sl@0	1697	* owned by the library.
sl@0	1698	* @see ucnv_getStandardName
sl@0	1699	* @stable ICU 2.4
sl@0	1700	*/
sl@0	1701	U_STABLE const char * U_EXPORT2
sl@0	1702	ucnv_getCanonicalName(const char alias, const char standard, UErrorCode *pErrorCode);
sl@0	1703
sl@0	1704	/**
sl@0	1705	* returns the current default converter name.
sl@0	1706	*
sl@0	1707	* @return returns the current default converter name;
sl@0	1708	* if a default converter name cannot be determined,
sl@0	1709	* then <code>NULL</code> is returned.
sl@0	1710	* Storage owned by the library
sl@0	1711	* @see ucnv_setDefaultName
sl@0	1712	* @stable ICU 2.0
sl@0	1713	*/
sl@0	1714	U_STABLE const char * U_EXPORT2
sl@0	1715	ucnv_getDefaultName(void);
sl@0	1716
sl@0	1717	/**
sl@0	1718	* sets the current default converter name. Caller must own the storage for 'name'
sl@0	1719	* and preserve it indefinitely.
sl@0	1720	* @param name the converter name to be the default (must exist).
sl@0	1721	* @see ucnv_getDefaultName
sl@0	1722	* @system SYSTEM API
sl@0	1723	* @stable ICU 2.0
sl@0	1724	*/
sl@0	1725	U_STABLE void U_EXPORT2
sl@0	1726	ucnv_setDefaultName(const char *name);
sl@0	1727
sl@0	1728	/**
sl@0	1729	* Fixes the backslash character mismapping. For example, in SJIS, the backslash
sl@0	1730	* character in the ASCII portion is also used to represent the yen currency sign.
sl@0	1731	* When mapping from Unicode character 0x005C, it's unclear whether to map the
sl@0	1732	* character back to yen or backslash in SJIS. This function will take the input
sl@0	1733	* buffer and replace all the yen sign characters with backslash. This is necessary
sl@0	1734	* when the user tries to open a file with the input buffer on Windows.
sl@0	1735	* This function will test the converter to see whether such mapping is
sl@0	1736	* required. You can sometimes avoid using this function by using the correct version
sl@0	1737	* of Shift-JIS.
sl@0	1738	*
sl@0	1739	* @param cnv The converter representing the target codepage.
sl@0	1740	* @param source the input buffer to be fixed
sl@0	1741	* @param sourceLen the length of the input buffer
sl@0	1742	* @see ucnv_isAmbiguous
sl@0	1743	* @stable ICU 2.0
sl@0	1744	*/
sl@0	1745	U_STABLE void U_EXPORT2
sl@0	1746	ucnv_fixFileSeparator(const UConverter cnv, UChar source, int32_t sourceLen);
sl@0	1747
sl@0	1748	/**
sl@0	1749	* Determines if the converter contains ambiguous mappings of the same
sl@0	1750	* character or not.
sl@0	1751	* @param cnv the converter to be tested
sl@0	1752	* @return TRUE if the converter contains ambiguous mapping of the same
sl@0	1753	* character, FALSE otherwise.
sl@0	1754	* @stable ICU 2.0
sl@0	1755	*/
sl@0	1756	U_STABLE UBool U_EXPORT2
sl@0	1757	ucnv_isAmbiguous(const UConverter *cnv);
sl@0	1758
sl@0	1759	/**
sl@0	1760	* Sets the converter to use fallback mapping or not.
sl@0	1761	* @param cnv The converter to set the fallback mapping usage on.
sl@0	1762	* @param usesFallback TRUE if the user wants the converter to take advantage of the fallback
sl@0	1763	* mapping, FALSE otherwise.
sl@0	1764	* @stable ICU 2.0
sl@0	1765	*/
sl@0	1766	U_STABLE void U_EXPORT2
sl@0	1767	ucnv_setFallback(UConverter *cnv, UBool usesFallback);
sl@0	1768
sl@0	1769	/**
sl@0	1770	* Determines if the converter uses fallback mappings or not.
sl@0	1771	* @param cnv The converter to be tested
sl@0	1772	* @return TRUE if the converter uses fallback, FALSE otherwise.
sl@0	1773	* @stable ICU 2.0
sl@0	1774	*/
sl@0	1775	U_STABLE UBool U_EXPORT2
sl@0	1776	ucnv_usesFallback(const UConverter *cnv);
sl@0	1777
sl@0	1778	/**
sl@0	1779	* Detects Unicode signature byte sequences at the start of the byte stream
sl@0	1780	* and returns the charset name of the indicated Unicode charset.
sl@0	1781	* NULL is returned when no Unicode signature is recognized.
sl@0	1782	* The number of bytes in the signature is output as well.
sl@0	1783	*
sl@0	1784	* The caller can ucnv_open() a converter using the charset name.
sl@0	1785	* The first code unit (UChar) from the start of the stream will be U+FEFF
sl@0	1786	* (the Unicode BOM/signature character) and can usually be ignored.
sl@0	1787	*
sl@0	1788	* For most Unicode charsets it is also possible to ignore the indicated
sl@0	1789	* number of initial stream bytes and start converting after them.
sl@0	1790	* However, there are stateful Unicode charsets (UTF-7 and BOCU-1) for which
sl@0	1791	* this will not work. Therefore, it is best to ignore the first output UChar
sl@0	1792	* instead of the input signature bytes.
sl@0	1793	* <p>
sl@0	1794	* Usage:
sl@0	1795	* @code
sl@0	1796	* UErrorCode err = U_ZERO_ERROR;
sl@0	1797	* char input[] = { '\xEF','\xBB', '\xBF','\x41','\x42','\x43' };
sl@0	1798	* int32_t signatureLength = 0;
sl@0	1799	* char *encoding = ucnv_detectUnicodeSignatures(input,sizeof(input),&signatureLength,&err);
sl@0	1800	* UConverter *conv = NULL;
sl@0	1801	* UChar output[100];
sl@0	1802	* UChar target = output, out;
sl@0	1803	* char *source = input;
sl@0	1804	* if(encoding!=NULL && U_SUCCESS(err)){
sl@0	1805	* // should signature be discarded ?
sl@0	1806	* conv = ucnv_open(encoding, &err);
sl@0	1807	* // do the conversion
sl@0	1808	* ucnv_toUnicode(conv,
sl@0	1809	* target, output + sizeof(output)/U_SIZEOF_UCHAR,
sl@0	1810	* source, input + sizeof(input),
sl@0	1811	* NULL, TRUE, &err);
sl@0	1812	* out = output;
sl@0	1813	* if (discardSignature){
sl@0	1814	* ++out; // ignore initial U+FEFF
sl@0	1815	* }
sl@0	1816	* while(out != target) {
sl@0	1817	* printf("%04x ", *out++);
sl@0	1818	* }
sl@0	1819	* puts("");
sl@0	1820	* }
sl@0	1821	*
sl@0	1822	* @endcode
sl@0	1823	*
sl@0	1824	* @param source The source string in which the signature should be detected.
sl@0	1825	* @param sourceLength Length of the input string, or -1 if terminated with a NUL byte.
sl@0	1826	* @param signatureLength A pointer to int32_t to receive the number of bytes that make up the signature
sl@0	1827	* of the detected UTF. 0 if not detected.
sl@0	1828	* Can be a NULL pointer.
sl@0	1829	* @param pErrorCode ICU error code in/out parameter.
sl@0	1830	* Must fulfill U_SUCCESS before the function call.
sl@0	1831	* @return The name of the encoding detected. NULL if encoding is not detected.
sl@0	1832	* @stable ICU 2.4
sl@0	1833	*/
sl@0	1834	U_STABLE const char* U_EXPORT2
sl@0	1835	ucnv_detectUnicodeSignature(const char* source,
sl@0	1836	int32_t sourceLength,
sl@0	1837	int32_t *signatureLength,
sl@0	1838	UErrorCode *pErrorCode);
sl@0	1839
sl@0	1840	/**
sl@0	1841	* Returns the number of UChars held in the converter's internal state
sl@0	1842	* because more input is needed for completing the conversion. This function is
sl@0	1843	* useful for mapping semantics of ICU's converter interface to those of iconv,
sl@0	1844	* and this information is not needed for normal conversion.
sl@0	1845	* @param cnv The converter in which the input is held
sl@0	1846	* @param status ICU error code in/out parameter.
sl@0	1847	* Must fulfill U_SUCCESS before the function call.
sl@0	1848	* @return The number of UChars in the state. -1 if an error is encountered.
sl@0	1849	* @draft ICU 3.4
sl@0	1850	*/
sl@0	1851	U_DRAFT int32_t U_EXPORT2
sl@0	1852	ucnv_fromUCountPending(const UConverter* cnv, UErrorCode* status);
sl@0	1853
sl@0	1854	/**
sl@0	1855	* Returns the number of chars held in the converter's internal state
sl@0	1856	* because more input is needed for completing the conversion. This function is
sl@0	1857	* useful for mapping semantics of ICU's converter interface to those of iconv,
sl@0	1858	* and this information is not needed for normal conversion.
sl@0	1859	* @param cnv The converter in which the input is held as internal state
sl@0	1860	* @param status ICU error code in/out parameter.
sl@0	1861	* Must fulfill U_SUCCESS before the function call.
sl@0	1862	* @return The number of chars in the state. -1 if an error is encountered.
sl@0	1863	* @draft ICU 3.4
sl@0	1864	*/
sl@0	1865	U_DRAFT int32_t U_EXPORT2
sl@0	1866	ucnv_toUCountPending(const UConverter* cnv, UErrorCode* status);
sl@0	1867
sl@0	1868	#endif
sl@0	1869
sl@0	1870	#endif
sl@0	1871	/_UCNV/

author	sl
	Tue, 10 Jun 2014 14:32:02 +0200
changeset 1	260cb5ec6c19
permissions	-rw-r--r--