1.1 --- /dev/null Thu Jan 01 00:00:00 1970 +0000
1.2 +++ b/os/textandloc/fontservices/textshaperplugin/IcuSource/common/unicode/ucnv.h Fri Jun 15 03:10:57 2012 +0200
1.3 @@ -0,0 +1,1871 @@
1.4 +/*
1.5 +**********************************************************************
1.6 +* Copyright (C) 1999-2005, International Business Machines
1.7 +* Corporation and others. All Rights Reserved.
1.8 +**********************************************************************
1.9 + * ucnv.h:
1.10 + * External APIs for the ICU's codeset conversion library
1.11 + * Bertrand A. Damiba
1.12 + *
1.13 + * Modification History:
1.14 + *
1.15 + * Date Name Description
1.16 + * 04/04/99 helena Fixed internal header inclusion.
1.17 + * 05/11/00 helena Added setFallback and usesFallback APIs.
1.18 + * 06/29/2000 helena Major rewrite of the callback APIs.
1.19 + * 12/07/2000 srl Update of documentation
1.20 + */
1.21 +
1.22 +/**
1.23 + * \file
1.24 + * \brief C API: Character conversion
1.25 + *
1.26 + * <h2>Character Conversion C API</h2>
1.27 + *
1.28 + * <p>This API is used to convert codepage or character encoded data to and
1.29 + * from UTF-16. You can open a converter with {@link ucnv_open() }. With that
1.30 + * converter, you can get its properties, set options, convert your data and
1.31 + * close the converter.</p>
1.32 + *
1.33 + * <p>Since many software programs recogize different converter names for
1.34 + * different types of converters, there are other functions in this API to
1.35 + * iterate over the converter aliases. The functions {@link ucnv_getAvailableName() },
1.36 + * {@link ucnv_getAlias() } and {@link ucnv_getStandardName() } are some of the
1.37 + * more frequently used alias functions to get this information.</p>
1.38 + *
1.39 + * <p>When a converter encounters an illegal, irregular, invalid or unmappable character
1.40 + * its default behavior is to use a substitution character to replace the
1.41 + * bad byte sequence. This behavior can be changed by using {@link ucnv_setFromUCallBack() }
1.42 + * or {@link ucnv_setToUCallBack() } on the converter. The header ucnv_err.h defines
1.43 + * many other callback actions that can be used instead of a character substitution.</p>
1.44 + *
1.45 + * <p>More information about this API can be found in our
1.46 + * <a href="http://icu.sourceforge.net/userguide/conversion.html">User's
1.47 + * Guide</a>.</p>
1.48 + */
1.49 +
1.50 +#ifndef UCNV_H
1.51 +#define UCNV_H
1.52 +
1.53 +#include "unicode/ucnv_err.h"
1.54 +#include "unicode/uenum.h"
1.55 +
1.56 +#ifndef __USET_H__
1.57 +
1.58 +/**
1.59 + * USet is the C API type for Unicode sets.
1.60 + * It is forward-declared here to avoid including the header file if related
1.61 + * conversion APIs are not used.
1.62 + * See unicode/uset.h
1.63 + *
1.64 + * @see ucnv_getUnicodeSet
1.65 + * @stable ICU 2.6
1.66 + */
1.67 +struct USet;
1.68 +/** @stable ICU 2.6 */
1.69 +typedef struct USet USet;
1.70 +
1.71 +#endif
1.72 +
1.73 +#if !UCONFIG_NO_CONVERSION
1.74 +
1.75 +U_CDECL_BEGIN
1.76 +
1.77 +/** Maximum length of a converter name including the terminating NULL @stable ICU 2.0 */
1.78 +#define UCNV_MAX_CONVERTER_NAME_LENGTH 60
1.79 +/** Maximum length of a converter name including path and terminating NULL @stable ICU 2.0 */
1.80 +#define UCNV_MAX_FULL_FILE_NAME_LENGTH (600+UCNV_MAX_CONVERTER_NAME_LENGTH)
1.81 +
1.82 +/** Shift in for EBDCDIC_STATEFUL and iso2022 states @stable ICU 2.0 */
1.83 +#define UCNV_SI 0x0F
1.84 +/** Shift out for EBDCDIC_STATEFUL and iso2022 states @stable ICU 2.0 */
1.85 +#define UCNV_SO 0x0E
1.86 +
1.87 +/**
1.88 + * Enum for specifying basic types of converters
1.89 + * @see ucnv_getType
1.90 + * @stable ICU 2.0
1.91 + */
1.92 +typedef enum {
1.93 + UCNV_UNSUPPORTED_CONVERTER = -1,
1.94 + UCNV_SBCS = 0,
1.95 + UCNV_DBCS = 1,
1.96 + UCNV_MBCS = 2,
1.97 + UCNV_LATIN_1 = 3,
1.98 + UCNV_UTF8 = 4,
1.99 + UCNV_UTF16_BigEndian = 5,
1.100 + UCNV_UTF16_LittleEndian = 6,
1.101 + UCNV_UTF32_BigEndian = 7,
1.102 + UCNV_UTF32_LittleEndian = 8,
1.103 + UCNV_EBCDIC_STATEFUL = 9,
1.104 + UCNV_ISO_2022 = 10,
1.105 +
1.106 + UCNV_LMBCS_1 = 11,
1.107 + UCNV_LMBCS_2,
1.108 + UCNV_LMBCS_3,
1.109 + UCNV_LMBCS_4,
1.110 + UCNV_LMBCS_5,
1.111 + UCNV_LMBCS_6,
1.112 + UCNV_LMBCS_8,
1.113 + UCNV_LMBCS_11,
1.114 + UCNV_LMBCS_16,
1.115 + UCNV_LMBCS_17,
1.116 + UCNV_LMBCS_18,
1.117 + UCNV_LMBCS_19,
1.118 + UCNV_LMBCS_LAST = UCNV_LMBCS_19,
1.119 + UCNV_HZ,
1.120 + UCNV_SCSU,
1.121 + UCNV_ISCII,
1.122 + UCNV_US_ASCII,
1.123 + UCNV_UTF7,
1.124 + UCNV_BOCU1,
1.125 + UCNV_UTF16,
1.126 + UCNV_UTF32,
1.127 + UCNV_CESU8,
1.128 + UCNV_IMAP_MAILBOX,
1.129 +
1.130 + /* Number of converter types for which we have conversion routines. */
1.131 + UCNV_NUMBER_OF_SUPPORTED_CONVERTER_TYPES
1.132 +
1.133 +} UConverterType;
1.134 +
1.135 +/**
1.136 + * Enum for specifying which platform a converter ID refers to.
1.137 + * The use of platform/CCSID is not recommended. See ucnv_openCCSID().
1.138 + *
1.139 + * @see ucnv_getPlatform
1.140 + * @see ucnv_openCCSID
1.141 + * @see ucnv_getCCSID
1.142 + * @stable ICU 2.0
1.143 + */
1.144 +typedef enum {
1.145 + UCNV_UNKNOWN = -1,
1.146 + UCNV_IBM = 0
1.147 +} UConverterPlatform;
1.148 +
1.149 +/**
1.150 + * Function pointer for error callback in the codepage to unicode direction.
1.151 + * Called when an error has occured in conversion to unicode, or on open/close of the callback (see reason).
1.152 + * @param context Pointer to the callback's private data
1.153 + * @param args Information about the conversion in progress
1.154 + * @param codeUnits Points to 'length' bytes of the concerned codepage sequence
1.155 + * @param length Size (in bytes) of the concerned codepage sequence
1.156 + * @param reason Defines the reason the callback was invoked
1.157 + * @param pErrorCode ICU error code in/out parameter.
1.158 + * For converter callback functions, set to a conversion error
1.159 + * before the call, and the callback may reset it to U_ZERO_ERROR.
1.160 + * @see ucnv_setToUCallBack
1.161 + * @see UConverterToUnicodeArgs
1.162 + * @stable ICU 2.0
1.163 + */
1.164 +typedef void (U_EXPORT2 *UConverterToUCallback) (
1.165 + const void* context,
1.166 + UConverterToUnicodeArgs *args,
1.167 + const char *codeUnits,
1.168 + int32_t length,
1.169 + UConverterCallbackReason reason,
1.170 + UErrorCode *pErrorCode);
1.171 +
1.172 +/**
1.173 + * Function pointer for error callback in the unicode to codepage direction.
1.174 + * Called when an error has occured in conversion from unicode, or on open/close of the callback (see reason).
1.175 + * @param context Pointer to the callback's private data
1.176 + * @param args Information about the conversion in progress
1.177 + * @param codeUnits Points to 'length' UChars of the concerned Unicode sequence
1.178 + * @param length Size (in bytes) of the concerned codepage sequence
1.179 + * @param codePoint Single UChar32 (UTF-32) containing the concerend Unicode codepoint.
1.180 + * @param reason Defines the reason the callback was invoked
1.181 + * @param pErrorCode ICU error code in/out parameter.
1.182 + * For converter callback functions, set to a conversion error
1.183 + * before the call, and the callback may reset it to U_ZERO_ERROR.
1.184 + * @see ucnv_setFromUCallBack
1.185 + * @stable ICU 2.0
1.186 + */
1.187 +typedef void (U_EXPORT2 *UConverterFromUCallback) (
1.188 + const void* context,
1.189 + UConverterFromUnicodeArgs *args,
1.190 + const UChar* codeUnits,
1.191 + int32_t length,
1.192 + UChar32 codePoint,
1.193 + UConverterCallbackReason reason,
1.194 + UErrorCode *pErrorCode);
1.195 +
1.196 +U_CDECL_END
1.197 +
1.198 +/**
1.199 + * Character that separates converter names from options and options from each other.
1.200 + * @see ucnv_open
1.201 + * @stable ICU 2.0
1.202 + */
1.203 +#define UCNV_OPTION_SEP_CHAR ','
1.204 +
1.205 +/**
1.206 + * String version of UCNV_OPTION_SEP_CHAR.
1.207 + * @see ucnv_open
1.208 + * @stable ICU 2.0
1.209 + */
1.210 +#define UCNV_OPTION_SEP_STRING ","
1.211 +
1.212 +/**
1.213 + * Character that separates a converter option from its value.
1.214 + * @see ucnv_open
1.215 + * @stable ICU 2.0
1.216 + */
1.217 +#define UCNV_VALUE_SEP_CHAR '='
1.218 +
1.219 +/**
1.220 + * String version of UCNV_VALUE_SEP_CHAR.
1.221 + * @see ucnv_open
1.222 + * @stable ICU 2.0
1.223 + */
1.224 +#define UCNV_VALUE_SEP_STRING "="
1.225 +
1.226 +/**
1.227 + * Converter option for specifying a locale.
1.228 + * For example, ucnv_open("SCSU,locale=ja", &errorCode);
1.229 + * See convrtrs.txt.
1.230 + *
1.231 + * @see ucnv_open
1.232 + * @stable ICU 2.0
1.233 + */
1.234 +#define UCNV_LOCALE_OPTION_STRING ",locale="
1.235 +
1.236 +/**
1.237 + * Converter option for specifying a version selector (0..9) for some converters.
1.238 + * For example, ucnv_open("UTF-7,version=1", &errorCode);
1.239 + * See convrtrs.txt.
1.240 + *
1.241 + * @see ucnv_open
1.242 + * @stable ICU 2.4
1.243 + */
1.244 +#define UCNV_VERSION_OPTION_STRING ",version="
1.245 +
1.246 +/**
1.247 + * Converter option for EBCDIC SBCS or mixed-SBCS/DBCS (stateful) codepages.
1.248 + * Swaps Unicode mappings for EBCDIC LF and NL codes, as used on
1.249 + * S/390 (z/OS) Unix System Services (Open Edition).
1.250 + * For example, ucnv_open("ibm-1047,swaplfnl", &errorCode);
1.251 + * See convrtrs.txt.
1.252 + *
1.253 + * @see ucnv_open
1.254 + * @stable ICU 2.4
1.255 + */
1.256 +#define UCNV_SWAP_LFNL_OPTION_STRING ",swaplfnl"
1.257 +
1.258 +/**
1.259 + * Do a fuzzy compare of a two converter/alias names. The comparison
1.260 + * is case-insensitive. It also ignores the characters '-', '_', and
1.261 + * ' ' (dash, underscore, and space). Thus the strings "UTF-8",
1.262 + * "utf_8", and "Utf 8" are exactly equivalent.
1.263 + *
1.264 + * @param name1 a converter name or alias, zero-terminated
1.265 + * @param name2 a converter name or alias, zero-terminated
1.266 + * @return 0 if the names match, or a negative value if the name1
1.267 + * lexically precedes name2, or a positive value if the name1
1.268 + * lexically follows name2.
1.269 + * @stable ICU 2.0
1.270 + */
1.271 +U_STABLE int U_EXPORT2
1.272 +ucnv_compareNames(const char *name1, const char *name2);
1.273 +
1.274 +
1.275 +/**
1.276 + * Creates a UConverter object with the names specified as a C string.
1.277 + * The actual name will be resolved with the alias file
1.278 + * using a case-insensitive string comparison that ignores
1.279 + * the delimiters '-', '_', and ' ' (dash, underscore, and space).
1.280 + * E.g., the names "UTF8", "utf-8", and "Utf 8" are all equivalent.
1.281 + * If <code>NULL</code> is passed for the converter name, it will create one with the
1.282 + * getDefaultName return value.
1.283 + *
1.284 + * <p>A converter name for ICU 1.5 and above may contain options
1.285 + * like a locale specification to control the specific behavior of
1.286 + * the newly instantiated converter.
1.287 + * The meaning of the options depends on the particular converter.
1.288 + * If an option is not defined for or recognized by a given converter, then it is ignored.</p>
1.289 + *
1.290 + * <p>Options are appended to the converter name string, with a
1.291 + * <code>UCNV_OPTION_SEP_CHAR</code> between the name and the first option and
1.292 + * also between adjacent options.</p>
1.293 + *
1.294 + * <p>If the alias is ambiguous, then the preferred converter is used
1.295 + * and the status is set to U_AMBIGUOUS_ALIAS_WARNING.</p>
1.296 + *
1.297 + * <p>The conversion behavior and names can vary between platforms. ICU may
1.298 + * convert some characters differently from other platforms. Details on this topic
1.299 + * are in the <a href="http://icu.sourceforge.net/userguide/conversion.html">User's
1.300 + * Guide</a>.</p>
1.301 + *
1.302 + * @param converterName Name of the uconv table, may have options appended
1.303 + * @param err outgoing error status <TT>U_MEMORY_ALLOCATION_ERROR, U_FILE_ACCESS_ERROR</TT>
1.304 + * @return the created Unicode converter object, or <TT>NULL</TT> if an error occured
1.305 + * @see ucnv_openU
1.306 + * @see ucnv_openCCSID
1.307 + * @see ucnv_close
1.308 + * @stable ICU 2.0
1.309 + */
1.310 +U_STABLE UConverter* U_EXPORT2
1.311 +ucnv_open(const char *converterName, UErrorCode *err);
1.312 +
1.313 +
1.314 +/**
1.315 + * Creates a Unicode converter with the names specified as unicode string.
1.316 + * The name should be limited to the ASCII-7 alphanumerics range.
1.317 + * The actual name will be resolved with the alias file
1.318 + * using a case-insensitive string comparison that ignores
1.319 + * the delimiters '-', '_', and ' ' (dash, underscore, and space).
1.320 + * E.g., the names "UTF8", "utf-8", and "Utf 8" are all equivalent.
1.321 + * If <TT>NULL</TT> is passed for the converter name, it will create
1.322 + * one with the ucnv_getDefaultName() return value.
1.323 + * If the alias is ambiguous, then the preferred converter is used
1.324 + * and the status is set to U_AMBIGUOUS_ALIAS_WARNING.
1.325 + * @param name : name of the uconv table in a zero terminated
1.326 + * Unicode string
1.327 + * @param err outgoing error status <TT>U_MEMORY_ALLOCATION_ERROR,
1.328 + * U_FILE_ACCESS_ERROR</TT>
1.329 + * @return the created Unicode converter object, or <TT>NULL</TT> if an
1.330 + * error occured
1.331 + * @see ucnv_open
1.332 + * @see ucnv_openCCSID
1.333 + * @see ucnv_close
1.334 + * @see ucnv_getDefaultName
1.335 + * @stable ICU 2.0
1.336 + */
1.337 +U_STABLE UConverter* U_EXPORT2
1.338 +ucnv_openU(const UChar *name,
1.339 + UErrorCode *err);
1.340 +
1.341 +/**
1.342 + * Creates a UConverter object from a CCSID number and platform pair.
1.343 + * Note that the usefulness of this function is limited to platforms with numeric
1.344 + * encoding IDs. Only IBM and Microsoft platforms use numeric (16-bit) identifiers for
1.345 + * encodings.
1.346 + *
1.347 + * In addition, IBM CCSIDs and Unicode conversion tables are not 1:1 related.
1.348 + * For many IBM CCSIDs there are multiple (up to six) Unicode conversion tables, and
1.349 + * for some Unicode conversion tables there are multiple CCSIDs.
1.350 + * Some "alternate" Unicode conversion tables are provided by the
1.351 + * IBM CDRA conversion table registry.
1.352 + * The most prominent example of a systematic modification of conversion tables that is
1.353 + * not provided in the form of conversion table files in the repository is
1.354 + * that S/390 Unix System Services swaps the codes for Line Feed and New Line in all
1.355 + * EBCDIC codepages, which requires such a swap in the Unicode conversion tables as well.
1.356 + *
1.357 + * Only IBM default conversion tables are accessible with ucnv_openCCSID().
1.358 + * ucnv_getCCSID() will return the same CCSID for all conversion tables that are associated
1.359 + * with that CCSID.
1.360 + *
1.361 + * Currently, the only "platform" supported in the ICU converter API is UCNV_IBM.
1.362 + *
1.363 + * In summary, the use of CCSIDs and the associated API functions is not recommended.
1.364 + *
1.365 + * In order to open a converter with the default IBM CDRA Unicode conversion table,
1.366 + * you can use this function or use the prefix "ibm-":
1.367 + * \code
1.368 + * char name[20];
1.369 + * sprintf(name, "ibm-%hu", ccsid);
1.370 + * cnv=ucnv_open(name, &errorCode);
1.371 + * \endcode
1.372 + *
1.373 + * In order to open a converter with the IBM S/390 Unix System Services variant
1.374 + * of a Unicode/EBCDIC conversion table,
1.375 + * you can use the prefix "ibm-" together with the option string UCNV_SWAP_LFNL_OPTION_STRING:
1.376 + * \code
1.377 + * char name[20];
1.378 + * sprintf(name, "ibm-%hu" UCNV_SWAP_LFNL_OPTION_STRING, ccsid);
1.379 + * cnv=ucnv_open(name, &errorCode);
1.380 + * \endcode
1.381 + *
1.382 + * In order to open a converter from a Microsoft codepage number, use the prefix "cp":
1.383 + * \code
1.384 + * char name[20];
1.385 + * sprintf(name, "cp%hu", codepageID);
1.386 + * cnv=ucnv_open(name, &errorCode);
1.387 + * \endcode
1.388 + *
1.389 + * If the alias is ambiguous, then the preferred converter is used
1.390 + * and the status is set to U_AMBIGUOUS_ALIAS_WARNING.
1.391 + *
1.392 + * @param codepage codepage number to create
1.393 + * @param platform the platform in which the codepage number exists
1.394 + * @param err error status <TT>U_MEMORY_ALLOCATION_ERROR, U_FILE_ACCESS_ERROR</TT>
1.395 + * @return the created Unicode converter object, or <TT>NULL</TT> if an error
1.396 + * occured.
1.397 + * @see ucnv_open
1.398 + * @see ucnv_openU
1.399 + * @see ucnv_close
1.400 + * @see ucnv_getCCSID
1.401 + * @see ucnv_getPlatform
1.402 + * @see UConverterPlatform
1.403 + * @stable ICU 2.0
1.404 + */
1.405 +U_STABLE UConverter* U_EXPORT2
1.406 +ucnv_openCCSID(int32_t codepage,
1.407 + UConverterPlatform platform,
1.408 + UErrorCode * err);
1.409 +
1.410 +/**
1.411 + * <p>Creates a UConverter object specified from a packageName and a converterName.</p>
1.412 + *
1.413 + * <p>The packageName and converterName must point to an ICU udata object, as defined by
1.414 + * <code> udata_open( packageName, "cnv", converterName, err) </code> or equivalent.
1.415 + * Typically, packageName will refer to a (.dat) file, or to a package registered with
1.416 + * udata_setAppData(). Using a full file or directory pathname for packageName is deprecated.</p>
1.417 + *
1.418 + * <p>The name will NOT be looked up in the alias mechanism, nor will the converter be
1.419 + * stored in the converter cache or the alias table. The only way to open further converters
1.420 + * is call this function multiple times, or use the ucnv_safeClone() function to clone a
1.421 + * 'master' converter.</p>
1.422 + *
1.423 + * <p>A future version of ICU may add alias table lookups and/or caching
1.424 + * to this function.</p>
1.425 + *
1.426 + * <p>Example Use:
1.427 + * <code>cnv = ucnv_openPackage("myapp", "myconverter", &err);</code>
1.428 + * </p>
1.429 + *
1.430 + * @param packageName name of the package (equivalent to 'path' in udata_open() call)
1.431 + * @param converterName name of the data item to be used, without suffix.
1.432 + * @param err outgoing error status <TT>U_MEMORY_ALLOCATION_ERROR, U_FILE_ACCESS_ERROR</TT>
1.433 + * @return the created Unicode converter object, or <TT>NULL</TT> if an error occured
1.434 + * @see udata_open
1.435 + * @see ucnv_open
1.436 + * @see ucnv_safeClone
1.437 + * @see ucnv_close
1.438 + * @stable ICU 2.2
1.439 + */
1.440 +U_STABLE UConverter* U_EXPORT2
1.441 +ucnv_openPackage(const char *packageName, const char *converterName, UErrorCode *err);
1.442 +
1.443 +/**
1.444 + * Thread safe converter cloning operation.
1.445 + * For most efficient operation, pass in a stackBuffer (and a *pBufferSize)
1.446 + * with at least U_CNV_SAFECLONE_BUFFERSIZE bytes of space.
1.447 + * If the buffer size is sufficient, then the clone will use the stack buffer;
1.448 + * otherwise, it will be allocated, and *pBufferSize will indicate
1.449 + * the actual size. (This should not occur with U_CNV_SAFECLONE_BUFFERSIZE.)
1.450 + *
1.451 + * You must ucnv_close() the clone in any case.
1.452 + *
1.453 + * If *pBufferSize==0, (regardless of whether stackBuffer==NULL or not)
1.454 + * then *pBufferSize will be changed to a sufficient size
1.455 + * for cloning this converter,
1.456 + * without actually cloning the converter ("pure pre-flighting").
1.457 + *
1.458 + * If *pBufferSize is greater than zero but not large enough for a stack-based
1.459 + * clone, then the converter is cloned using newly allocated memory
1.460 + * and *pBufferSize is changed to the necessary size.
1.461 + *
1.462 + * If the converter clone fits into the stack buffer but the stack buffer is not
1.463 + * sufficiently aligned for the clone, then the clone will use an
1.464 + * adjusted pointer and use an accordingly smaller buffer size.
1.465 + *
1.466 + * @param cnv converter to be cloned
1.467 + * @param stackBuffer user allocated space for the new clone. If NULL new memory will be allocated.
1.468 + * If buffer is not large enough, new memory will be allocated.
1.469 + * Clients can use the U_CNV_SAFECLONE_BUFFERSIZE. This will probably be enough to avoid memory allocations.
1.470 + * @param pBufferSize pointer to size of allocated space. pBufferSize must not be NULL.
1.471 + * @param status to indicate whether the operation went on smoothly or there were errors
1.472 + * An informational status value, U_SAFECLONE_ALLOCATED_WARNING,
1.473 + * is used if any allocations were necessary.
1.474 + * However, it is better to check if *pBufferSize grew for checking for
1.475 + * allocations because warning codes can be overridden by subsequent
1.476 + * function calls.
1.477 + * @return pointer to the new clone
1.478 + * @stable ICU 2.0
1.479 + */
1.480 +U_STABLE UConverter * U_EXPORT2
1.481 +ucnv_safeClone(const UConverter *cnv,
1.482 + void *stackBuffer,
1.483 + int32_t *pBufferSize,
1.484 + UErrorCode *status);
1.485 +
1.486 +/**
1.487 + * \def U_CNV_SAFECLONE_BUFFERSIZE
1.488 + * Definition of a buffer size that is designed to be large enough for
1.489 + * converters to be cloned with ucnv_safeClone().
1.490 + * @stable ICU 2.0
1.491 + */
1.492 +#define U_CNV_SAFECLONE_BUFFERSIZE 1024
1.493 +
1.494 +/**
1.495 + * Deletes the unicode converter and releases resources associated
1.496 + * with just this instance.
1.497 + * Does not free up shared converter tables.
1.498 + *
1.499 + * @param converter the converter object to be deleted
1.500 + * @see ucnv_open
1.501 + * @see ucnv_openU
1.502 + * @see ucnv_openCCSID
1.503 + * @stable ICU 2.0
1.504 + */
1.505 +U_STABLE void U_EXPORT2
1.506 +ucnv_close(UConverter * converter);
1.507 +
1.508 +/**
1.509 + * Fills in the output parameter, subChars, with the substitution characters
1.510 + * as multiple bytes.
1.511 + *
1.512 + * @param converter the Unicode converter
1.513 + * @param subChars the subsitution characters
1.514 + * @param len on input the capacity of subChars, on output the number
1.515 + * of bytes copied to it
1.516 + * @param err the outgoing error status code.
1.517 + * If the substitution character array is too small, an
1.518 + * <TT>U_INDEX_OUTOFBOUNDS_ERROR</TT> will be returned.
1.519 + * @see ucnv_setSubstChars
1.520 + * @stable ICU 2.0
1.521 + */
1.522 +U_STABLE void U_EXPORT2
1.523 +ucnv_getSubstChars(const UConverter *converter,
1.524 + char *subChars,
1.525 + int8_t *len,
1.526 + UErrorCode *err);
1.527 +
1.528 +/**
1.529 + * Sets the substitution chars when converting from unicode to a codepage. The
1.530 + * substitution is specified as a string of 1-4 bytes, and may contain
1.531 + * <TT>NULL</TT> byte.
1.532 + * @param converter the Unicode converter
1.533 + * @param subChars the substitution character byte sequence we want set
1.534 + * @param len the number of bytes in subChars
1.535 + * @param err the error status code. <TT>U_INDEX_OUTOFBOUNDS_ERROR </TT> if
1.536 + * len is bigger than the maximum number of bytes allowed in subchars
1.537 + * @see ucnv_getSubstChars
1.538 + * @stable ICU 2.0
1.539 + */
1.540 +U_STABLE void U_EXPORT2
1.541 +ucnv_setSubstChars(UConverter *converter,
1.542 + const char *subChars,
1.543 + int8_t len,
1.544 + UErrorCode *err);
1.545 +
1.546 +/**
1.547 + * Fills in the output parameter, errBytes, with the error characters from the
1.548 + * last failing conversion.
1.549 + *
1.550 + * @param converter the Unicode converter
1.551 + * @param errBytes the codepage bytes which were in error
1.552 + * @param len on input the capacity of errBytes, on output the number of
1.553 + * bytes which were copied to it
1.554 + * @param err the error status code.
1.555 + * If the substitution character array is too small, an
1.556 + * <TT>U_INDEX_OUTOFBOUNDS_ERROR</TT> will be returned.
1.557 + * @stable ICU 2.0
1.558 + */
1.559 +U_STABLE void U_EXPORT2
1.560 +ucnv_getInvalidChars(const UConverter *converter,
1.561 + char *errBytes,
1.562 + int8_t *len,
1.563 + UErrorCode *err);
1.564 +
1.565 +/**
1.566 + * Fills in the output parameter, errChars, with the error characters from the
1.567 + * last failing conversion.
1.568 + *
1.569 + * @param converter the Unicode converter
1.570 + * @param errUChars the UChars which were in error
1.571 + * @param len on input the capacity of errUChars, on output the number of
1.572 + * UChars which were copied to it
1.573 + * @param err the error status code.
1.574 + * If the substitution character array is too small, an
1.575 + * <TT>U_INDEX_OUTOFBOUNDS_ERROR</TT> will be returned.
1.576 + * @stable ICU 2.0
1.577 + */
1.578 +U_STABLE void U_EXPORT2
1.579 +ucnv_getInvalidUChars(const UConverter *converter,
1.580 + UChar *errUChars,
1.581 + int8_t *len,
1.582 + UErrorCode *err);
1.583 +
1.584 +/**
1.585 + * Resets the state of a converter to the default state. This is used
1.586 + * in the case of an error, to restart a conversion from a known default state.
1.587 + * It will also empty the internal output buffers.
1.588 + * @param converter the Unicode converter
1.589 + * @stable ICU 2.0
1.590 + */
1.591 +U_STABLE void U_EXPORT2
1.592 +ucnv_reset(UConverter *converter);
1.593 +
1.594 +/**
1.595 + * Resets the to-Unicode part of a converter state to the default state.
1.596 + * This is used in the case of an error to restart a conversion to
1.597 + * Unicode to a known default state. It will also empty the internal
1.598 + * output buffers used for the conversion to Unicode codepoints.
1.599 + * @param converter the Unicode converter
1.600 + * @stable ICU 2.0
1.601 + */
1.602 +U_STABLE void U_EXPORT2
1.603 +ucnv_resetToUnicode(UConverter *converter);
1.604 +
1.605 +/**
1.606 + * Resets the from-Unicode part of a converter state to the default state.
1.607 + * This is used in the case of an error to restart a conversion from
1.608 + * Unicode to a known default state. It will also empty the internal output
1.609 + * buffers used for the conversion from Unicode codepoints.
1.610 + * @param converter the Unicode converter
1.611 + * @stable ICU 2.0
1.612 + */
1.613 +U_STABLE void U_EXPORT2
1.614 +ucnv_resetFromUnicode(UConverter *converter);
1.615 +
1.616 +/**
1.617 + * Returns the maximum number of bytes that are output per UChar in conversion
1.618 + * from Unicode using this converter.
1.619 + * The returned number can be used with UCNV_GET_MAX_BYTES_FOR_STRING
1.620 + * to calculate the size of a target buffer for conversion from Unicode.
1.621 + *
1.622 + * Note: Before ICU 2.8, this function did not return reliable numbers for
1.623 + * some stateful converters (EBCDIC_STATEFUL, ISO-2022) and LMBCS.
1.624 + *
1.625 + * This number may not be the same as the maximum number of bytes per
1.626 + * "conversion unit". In other words, it may not be the intuitively expected
1.627 + * number of bytes per character that would be published for a charset,
1.628 + * and may not fulfill any other purpose than the allocation of an output
1.629 + * buffer of guaranteed sufficient size for a given input length and converter.
1.630 + *
1.631 + * Examples for special cases that are taken into account:
1.632 + * - Supplementary code points may convert to more bytes than BMP code points.
1.633 + * This function returns bytes per UChar (UTF-16 code unit), not per
1.634 + * Unicode code point, for efficient buffer allocation.
1.635 + * - State-shifting output (SI/SO, escapes, etc.) from stateful converters.
1.636 + * - When m input UChars are converted to n output bytes, then the maximum m/n
1.637 + * is taken into account.
1.638 + *
1.639 + * The number returned here does not take into account
1.640 + * (see UCNV_GET_MAX_BYTES_FOR_STRING):
1.641 + * - callbacks which output more than one charset character sequence per call,
1.642 + * like escape callbacks
1.643 + * - initial and final non-character bytes that are output by some converters
1.644 + * (automatic BOMs, initial escape sequence, final SI, etc.)
1.645 + *
1.646 + * Examples for returned values:
1.647 + * - SBCS charsets: 1
1.648 + * - Shift-JIS: 2
1.649 + * - UTF-16: 2 (2 per BMP, 4 per surrogate _pair_, BOM not counted)
1.650 + * - UTF-8: 3 (3 per BMP, 4 per surrogate _pair_)
1.651 + * - EBCDIC_STATEFUL (EBCDIC mixed SBCS/DBCS): 3 (SO + DBCS)
1.652 + * - ISO-2022: 3 (always outputs UTF-8)
1.653 + * - ISO-2022-JP: 6 (4-byte escape sequences + DBCS)
1.654 + * - ISO-2022-CN: 8 (4-byte designator sequences + 2-byte SS2/SS3 + DBCS)
1.655 + *
1.656 + * @param converter The Unicode converter.
1.657 + * @return The maximum number of bytes per UChar that are output by ucnv_fromUnicode(),
1.658 + * to be used together with UCNV_GET_MAX_BYTES_FOR_STRING for buffer allocation.
1.659 + *
1.660 + * @see UCNV_GET_MAX_BYTES_FOR_STRING
1.661 + * @see ucnv_getMinCharSize
1.662 + * @stable ICU 2.0
1.663 + */
1.664 +U_STABLE int8_t U_EXPORT2
1.665 +ucnv_getMaxCharSize(const UConverter *converter);
1.666 +
1.667 +#ifndef U_HIDE_DRAFT_API
1.668 +
1.669 +/**
1.670 + * Calculates the size of a buffer for conversion from Unicode to a charset.
1.671 + * The calculated size is guaranteed to be sufficient for this conversion.
1.672 + *
1.673 + * It takes into account initial and final non-character bytes that are output
1.674 + * by some converters.
1.675 + * It does not take into account callbacks which output more than one charset
1.676 + * character sequence per call, like escape callbacks.
1.677 + * The default (substitution) callback only outputs one charset character sequence.
1.678 + *
1.679 + * @param length Number of UChars to be converted.
1.680 + * @param maxCharSize Return value from ucnv_getMaxCharSize() for the converter
1.681 + * that will be used.
1.682 + * @return Size of a buffer that will be large enough to hold the output bytes of
1.683 + * converting length UChars with the converter that returned the maxCharSize.
1.684 + *
1.685 + * @see ucnv_getMaxCharSize
1.686 + * @stable ICU 2.8
1.687 + */
1.688 +#define UCNV_GET_MAX_BYTES_FOR_STRING(length, maxCharSize) \
1.689 + (((int32_t)(length)+10)*(int32_t)(maxCharSize))
1.690 +
1.691 +#endif /*U_HIDE_DRAFT_API*/
1.692 +
1.693 +/**
1.694 + * Returns the minimum byte length for characters in this codepage.
1.695 + * This is usually either 1 or 2.
1.696 + * @param converter the Unicode converter
1.697 + * @return the minimum number of bytes allowed by this particular converter
1.698 + * @see ucnv_getMaxCharSize
1.699 + * @stable ICU 2.0
1.700 + */
1.701 +U_STABLE int8_t U_EXPORT2
1.702 +ucnv_getMinCharSize(const UConverter *converter);
1.703 +
1.704 +/**
1.705 + * Returns the display name of the converter passed in based on the Locale
1.706 + * passed in. If the locale contains no display name, the internal ASCII
1.707 + * name will be filled in.
1.708 + *
1.709 + * @param converter the Unicode converter.
1.710 + * @param displayLocale is the specific Locale we want to localised for
1.711 + * @param displayName user provided buffer to be filled in
1.712 + * @param displayNameCapacity size of displayName Buffer
1.713 + * @param err error status code
1.714 + * @return displayNameLength number of UChar needed in displayName
1.715 + * @see ucnv_getName
1.716 + * @stable ICU 2.0
1.717 + */
1.718 +U_STABLE int32_t U_EXPORT2
1.719 +ucnv_getDisplayName(const UConverter *converter,
1.720 + const char *displayLocale,
1.721 + UChar *displayName,
1.722 + int32_t displayNameCapacity,
1.723 + UErrorCode *err);
1.724 +
1.725 +/**
1.726 + * Gets the internal, canonical name of the converter (zero-terminated).
1.727 + * The lifetime of the returned string will be that of the converter
1.728 + * passed to this function.
1.729 + * @param converter the Unicode converter
1.730 + * @param err UErrorCode status
1.731 + * @return the internal name of the converter
1.732 + * @see ucnv_getDisplayName
1.733 + * @stable ICU 2.0
1.734 + */
1.735 +U_STABLE const char * U_EXPORT2
1.736 +ucnv_getName(const UConverter *converter, UErrorCode *err);
1.737 +
1.738 +/**
1.739 + * Gets a codepage number associated with the converter. This is not guaranteed
1.740 + * to be the one used to create the converter. Some converters do not represent
1.741 + * platform registered codepages and return zero for the codepage number.
1.742 + * The error code fill-in parameter indicates if the codepage number
1.743 + * is available.
1.744 + * Does not check if the converter is <TT>NULL</TT> or if converter's data
1.745 + * table is <TT>NULL</TT>.
1.746 + *
1.747 + * Important: The use of CCSIDs is not recommended because it is limited
1.748 + * to only two platforms in principle and only one (UCNV_IBM) in the current
1.749 + * ICU converter API.
1.750 + * Also, CCSIDs are insufficient to identify IBM Unicode conversion tables precisely.
1.751 + * For more details see ucnv_openCCSID().
1.752 + *
1.753 + * @param converter the Unicode converter
1.754 + * @param err the error status code.
1.755 + * @return If any error occurrs, -1 will be returned otherwise, the codepage number
1.756 + * will be returned
1.757 + * @see ucnv_openCCSID
1.758 + * @see ucnv_getPlatform
1.759 + * @stable ICU 2.0
1.760 + */
1.761 +U_STABLE int32_t U_EXPORT2
1.762 +ucnv_getCCSID(const UConverter *converter,
1.763 + UErrorCode *err);
1.764 +
1.765 +/**
1.766 + * Gets a codepage platform associated with the converter. Currently,
1.767 + * only <TT>UCNV_IBM</TT> will be returned.
1.768 + * Does not test if the converter is <TT>NULL</TT> or if converter's data
1.769 + * table is <TT>NULL</TT>.
1.770 + * @param converter the Unicode converter
1.771 + * @param err the error status code.
1.772 + * @return The codepage platform
1.773 + * @stable ICU 2.0
1.774 + */
1.775 +U_STABLE UConverterPlatform U_EXPORT2
1.776 +ucnv_getPlatform(const UConverter *converter,
1.777 + UErrorCode *err);
1.778 +
1.779 +/**
1.780 + * Gets the type of the converter
1.781 + * e.g. SBCS, MBCS, DBCS, UTF8, UTF16_BE, UTF16_LE, ISO_2022,
1.782 + * EBCDIC_STATEFUL, LATIN_1
1.783 + * @param converter a valid, opened converter
1.784 + * @return the type of the converter
1.785 + * @stable ICU 2.0
1.786 + */
1.787 +U_STABLE UConverterType U_EXPORT2
1.788 +ucnv_getType(const UConverter * converter);
1.789 +
1.790 +/**
1.791 + * Gets the "starter" (lead) bytes for converters of type MBCS.
1.792 + * Will fill in an <TT>U_ILLEGAL_ARGUMENT_ERROR</TT> if converter passed in
1.793 + * is not MBCS. Fills in an array of type UBool, with the value of the byte
1.794 + * as offset to the array. For example, if (starters[0x20] == TRUE) at return,
1.795 + * it means that the byte 0x20 is a starter byte in this converter.
1.796 + * Context pointers are always owned by the caller.
1.797 + *
1.798 + * @param converter a valid, opened converter of type MBCS
1.799 + * @param starters an array of size 256 to be filled in
1.800 + * @param err error status, <TT>U_ILLEGAL_ARGUMENT_ERROR</TT> if the
1.801 + * converter is not a type which can return starters.
1.802 + * @see ucnv_getType
1.803 + * @stable ICU 2.0
1.804 + */
1.805 +U_STABLE void U_EXPORT2
1.806 +ucnv_getStarters(const UConverter* converter,
1.807 + UBool starters[256],
1.808 + UErrorCode* err);
1.809 +
1.810 +
1.811 +/**
1.812 + * Selectors for Unicode sets that can be returned by ucnv_getUnicodeSet().
1.813 + * @see ucnv_getUnicodeSet
1.814 + * @stable ICU 2.6
1.815 + */
1.816 +typedef enum UConverterUnicodeSet {
1.817 + /** Select the set of roundtrippable Unicode code points. @stable ICU 2.6 */
1.818 + UCNV_ROUNDTRIP_SET,
1.819 + /** Number of UConverterUnicodeSet selectors. @stable ICU 2.6 */
1.820 + UCNV_SET_COUNT
1.821 +} UConverterUnicodeSet;
1.822 +
1.823 +
1.824 +/**
1.825 + * Returns the set of Unicode code points that can be converted by an ICU converter.
1.826 + *
1.827 + * The current implementation returns only one kind of set (UCNV_ROUNDTRIP_SET):
1.828 + * The set of all Unicode code points that can be roundtrip-converted
1.829 + * (converted without any data loss) with the converter.
1.830 + * This set will not include code points that have fallback mappings
1.831 + * or are only the result of reverse fallback mappings.
1.832 + * See UTR #22 "Character Mapping Markup Language"
1.833 + * at http://www.unicode.org/reports/tr22/
1.834 + *
1.835 + * This is useful for example for
1.836 + * - checking that a string or document can be roundtrip-converted with a converter,
1.837 + * without/before actually performing the conversion
1.838 + * - testing if a converter can be used for text for typical text for a certain locale,
1.839 + * by comparing its roundtrip set with the set of ExemplarCharacters from
1.840 + * ICU's locale data or other sources
1.841 + *
1.842 + * In the future, there may be more UConverterUnicodeSet choices to select
1.843 + * sets with different properties.
1.844 + *
1.845 + * @param cnv The converter for which a set is requested.
1.846 + * @param setFillIn A valid USet *. It will be cleared by this function before
1.847 + * the converter's specific set is filled into the USet.
1.848 + * @param whichSet A UConverterUnicodeSet selector;
1.849 + * currently UCNV_ROUNDTRIP_SET is the only supported value.
1.850 + * @param pErrorCode ICU error code in/out parameter.
1.851 + * Must fulfill U_SUCCESS before the function call.
1.852 + *
1.853 + * @see UConverterUnicodeSet
1.854 + * @see uset_open
1.855 + * @see uset_close
1.856 + * @stable ICU 2.6
1.857 + */
1.858 +U_STABLE void U_EXPORT2
1.859 +ucnv_getUnicodeSet(const UConverter *cnv,
1.860 + USet *setFillIn,
1.861 + UConverterUnicodeSet whichSet,
1.862 + UErrorCode *pErrorCode);
1.863 +
1.864 +/**
1.865 + * Gets the current calback function used by the converter when an illegal
1.866 + * or invalid codepage sequence is found.
1.867 + * Context pointers are always owned by the caller.
1.868 + *
1.869 + * @param converter the unicode converter
1.870 + * @param action fillin: returns the callback function pointer
1.871 + * @param context fillin: returns the callback's private void* context
1.872 + * @see ucnv_setToUCallBack
1.873 + * @stable ICU 2.0
1.874 + */
1.875 +U_STABLE void U_EXPORT2
1.876 +ucnv_getToUCallBack (const UConverter * converter,
1.877 + UConverterToUCallback *action,
1.878 + const void **context);
1.879 +
1.880 +/**
1.881 + * Gets the current callback function used by the converter when illegal
1.882 + * or invalid Unicode sequence is found.
1.883 + * Context pointers are always owned by the caller.
1.884 + *
1.885 + * @param converter the unicode converter
1.886 + * @param action fillin: returns the callback function pointer
1.887 + * @param context fillin: returns the callback's private void* context
1.888 + * @see ucnv_setFromUCallBack
1.889 + * @stable ICU 2.0
1.890 + */
1.891 +U_STABLE void U_EXPORT2
1.892 +ucnv_getFromUCallBack (const UConverter * converter,
1.893 + UConverterFromUCallback *action,
1.894 + const void **context);
1.895 +
1.896 +/**
1.897 + * Changes the callback function used by the converter when
1.898 + * an illegal or invalid sequence is found.
1.899 + * Context pointers are always owned by the caller.
1.900 + * Predefined actions and contexts can be found in the ucnv_err.h header.
1.901 + *
1.902 + * @param converter the unicode converter
1.903 + * @param newAction the new callback function
1.904 + * @param newContext the new toUnicode callback context pointer. This can be NULL.
1.905 + * @param oldAction fillin: returns the old callback function pointer. This can be NULL.
1.906 + * @param oldContext fillin: returns the old callback's private void* context. This can be NULL.
1.907 + * @param err The error code status
1.908 + * @see ucnv_getToUCallBack
1.909 + * @stable ICU 2.0
1.910 + */
1.911 +U_STABLE void U_EXPORT2
1.912 +ucnv_setToUCallBack (UConverter * converter,
1.913 + UConverterToUCallback newAction,
1.914 + const void* newContext,
1.915 + UConverterToUCallback *oldAction,
1.916 + const void** oldContext,
1.917 + UErrorCode * err);
1.918 +
1.919 +/**
1.920 + * Changes the current callback function used by the converter when
1.921 + * an illegal or invalid sequence is found.
1.922 + * Context pointers are always owned by the caller.
1.923 + * Predefined actions and contexts can be found in the ucnv_err.h header.
1.924 + *
1.925 + * @param converter the unicode converter
1.926 + * @param newAction the new callback function
1.927 + * @param newContext the new fromUnicode callback context pointer. This can be NULL.
1.928 + * @param oldAction fillin: returns the old callback function pointer. This can be NULL.
1.929 + * @param oldContext fillin: returns the old callback's private void* context. This can be NULL.
1.930 + * @param err The error code status
1.931 + * @see ucnv_getFromUCallBack
1.932 + * @stable ICU 2.0
1.933 + */
1.934 +U_STABLE void U_EXPORT2
1.935 +ucnv_setFromUCallBack (UConverter * converter,
1.936 + UConverterFromUCallback newAction,
1.937 + const void *newContext,
1.938 + UConverterFromUCallback *oldAction,
1.939 + const void **oldContext,
1.940 + UErrorCode * err);
1.941 +
1.942 +/**
1.943 + * Converts an array of unicode characters to an array of codepage
1.944 + * characters. This function is optimized for converting a continuous
1.945 + * stream of data in buffer-sized chunks, where the entire source and
1.946 + * target does not fit in available buffers.
1.947 + *
1.948 + * The source pointer is an in/out parameter. It starts out pointing where the
1.949 + * conversion is to begin, and ends up pointing after the last UChar consumed.
1.950 + *
1.951 + * Target similarly starts out pointer at the first available byte in the output
1.952 + * buffer, and ends up pointing after the last byte written to the output.
1.953 + *
1.954 + * The converter always attempts to consume the entire source buffer, unless
1.955 + * (1.) the target buffer is full, or (2.) a failing error is returned from the
1.956 + * current callback function. When a successful error status has been
1.957 + * returned, it means that all of the source buffer has been
1.958 + * consumed. At that point, the caller should reset the source and
1.959 + * sourceLimit pointers to point to the next chunk.
1.960 + *
1.961 + * At the end of the stream (flush==TRUE), the input is completely consumed
1.962 + * when *source==sourceLimit and no error code is set.
1.963 + * The converter object is then automatically reset by this function.
1.964 + * (This means that a converter need not be reset explicitly between data
1.965 + * streams if it finishes the previous stream without errors.)
1.966 + *
1.967 + * This is a <I>stateful</I> conversion. Additionally, even when all source data has
1.968 + * been consumed, some data may be in the converters' internal state.
1.969 + * Call this function repeatedly, updating the target pointers with
1.970 + * the next empty chunk of target in case of a
1.971 + * <TT>U_BUFFER_OVERFLOW_ERROR</TT>, and updating the source pointers
1.972 + * with the next chunk of source when a successful error status is
1.973 + * returned, until there are no more chunks of source data.
1.974 + * @param converter the Unicode converter
1.975 + * @param target I/O parameter. Input : Points to the beginning of the buffer to copy
1.976 + * codepage characters to. Output : points to after the last codepage character copied
1.977 + * to <TT>target</TT>.
1.978 + * @param targetLimit the pointer just after last of the <TT>target</TT> buffer
1.979 + * @param source I/O parameter, pointer to pointer to the source Unicode character buffer.
1.980 + * @param sourceLimit the pointer just after the last of the source buffer
1.981 + * @param offsets if NULL is passed, nothing will happen to it, otherwise it needs to have the same number
1.982 + * of allocated cells as <TT>target</TT>. Will fill in offsets from target to source pointer
1.983 + * 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>
1.984 + * For output data carried across calls, and other data without a specific source character
1.985 + * (such as from escape sequences or callbacks) -1 will be placed for offsets.
1.986 + * @param flush set to <TT>TRUE</TT> if the current source buffer is the last available
1.987 + * chunk of the source, <TT>FALSE</TT> otherwise. Note that if a failing status is returned,
1.988 + * this function may have to be called multiple times with flush set to <TT>TRUE</TT> until
1.989 + * the source buffer is consumed.
1.990 + * @param err the error status. <TT>U_ILLEGAL_ARGUMENT_ERROR</TT> will be set if the
1.991 + * converter is <TT>NULL</TT>.
1.992 + * <code>U_BUFFER_OVERFLOW_ERROR</code> will be set if the target is full and there is
1.993 + * still data to be written to the target.
1.994 + * @see ucnv_fromUChars
1.995 + * @see ucnv_convert
1.996 + * @see ucnv_getMinCharSize
1.997 + * @see ucnv_setToUCallBack
1.998 + * @stable ICU 2.0
1.999 + */
1.1000 +U_STABLE void U_EXPORT2
1.1001 +ucnv_fromUnicode (UConverter * converter,
1.1002 + char **target,
1.1003 + const char *targetLimit,
1.1004 + const UChar ** source,
1.1005 + const UChar * sourceLimit,
1.1006 + int32_t* offsets,
1.1007 + UBool flush,
1.1008 + UErrorCode * err);
1.1009 +
1.1010 +/**
1.1011 + * Converts a buffer of codepage bytes into an array of unicode UChars
1.1012 + * characters. This function is optimized for converting a continuous
1.1013 + * stream of data in buffer-sized chunks, where the entire source and
1.1014 + * target does not fit in available buffers.
1.1015 + *
1.1016 + * The source pointer is an in/out parameter. It starts out pointing where the
1.1017 + * conversion is to begin, and ends up pointing after the last byte of source consumed.
1.1018 + *
1.1019 + * Target similarly starts out pointer at the first available UChar in the output
1.1020 + * buffer, and ends up pointing after the last UChar written to the output.
1.1021 + * It does NOT necessarily keep UChar sequences together.
1.1022 + *
1.1023 + * The converter always attempts to consume the entire source buffer, unless
1.1024 + * (1.) the target buffer is full, or (2.) a failing error is returned from the
1.1025 + * current callback function. When a successful error status has been
1.1026 + * returned, it means that all of the source buffer has been
1.1027 + * consumed. At that point, the caller should reset the source and
1.1028 + * sourceLimit pointers to point to the next chunk.
1.1029 + *
1.1030 + * At the end of the stream (flush==TRUE), the input is completely consumed
1.1031 + * when *source==sourceLimit and no error code is set
1.1032 + * The converter object is then automatically reset by this function.
1.1033 + * (This means that a converter need not be reset explicitly between data
1.1034 + * streams if it finishes the previous stream without errors.)
1.1035 + *
1.1036 + * This is a <I>stateful</I> conversion. Additionally, even when all source data has
1.1037 + * been consumed, some data may be in the converters' internal state.
1.1038 + * Call this function repeatedly, updating the target pointers with
1.1039 + * the next empty chunk of target in case of a
1.1040 + * <TT>U_BUFFER_OVERFLOW_ERROR</TT>, and updating the source pointers
1.1041 + * with the next chunk of source when a successful error status is
1.1042 + * returned, until there are no more chunks of source data.
1.1043 + * @param converter the Unicode converter
1.1044 + * @param target I/O parameter. Input : Points to the beginning of the buffer to copy
1.1045 + * UChars into. Output : points to after the last UChar copied.
1.1046 + * @param targetLimit the pointer just after the end of the <TT>target</TT> buffer
1.1047 + * @param source I/O parameter, pointer to pointer to the source codepage buffer.
1.1048 + * @param sourceLimit the pointer to the byte after the end of the source buffer
1.1049 + * @param offsets if NULL is passed, nothing will happen to it, otherwise it needs to have the same number
1.1050 + * of allocated cells as <TT>target</TT>. Will fill in offsets from target to source pointer
1.1051 + * 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>
1.1052 + * For output data carried across calls, and other data without a specific source character
1.1053 + * (such as from escape sequences or callbacks) -1 will be placed for offsets.
1.1054 + * @param flush set to <TT>TRUE</TT> if the current source buffer is the last available
1.1055 + * chunk of the source, <TT>FALSE</TT> otherwise. Note that if a failing status is returned,
1.1056 + * this function may have to be called multiple times with flush set to <TT>TRUE</TT> until
1.1057 + * the source buffer is consumed.
1.1058 + * @param err the error status. <TT>U_ILLEGAL_ARGUMENT_ERROR</TT> will be set if the
1.1059 + * converter is <TT>NULL</TT>.
1.1060 + * <code>U_BUFFER_OVERFLOW_ERROR</code> will be set if the target is full and there is
1.1061 + * still data to be written to the target.
1.1062 + * @see ucnv_fromUChars
1.1063 + * @see ucnv_convert
1.1064 + * @see ucnv_getMinCharSize
1.1065 + * @see ucnv_setFromUCallBack
1.1066 + * @see ucnv_getNextUChar
1.1067 + * @stable ICU 2.0
1.1068 + */
1.1069 +U_STABLE void U_EXPORT2
1.1070 +ucnv_toUnicode(UConverter *converter,
1.1071 + UChar **target,
1.1072 + const UChar *targetLimit,
1.1073 + const char **source,
1.1074 + const char *sourceLimit,
1.1075 + int32_t *offsets,
1.1076 + UBool flush,
1.1077 + UErrorCode *err);
1.1078 +
1.1079 +/**
1.1080 + * Convert the Unicode string into a codepage string using an existing UConverter.
1.1081 + * The output string is NUL-terminated if possible.
1.1082 + *
1.1083 + * This function is a more convenient but less powerful version of ucnv_fromUnicode().
1.1084 + * It is only useful for whole strings, not for streaming conversion.
1.1085 + *
1.1086 + * The maximum output buffer capacity required (barring output from callbacks) will be
1.1087 + * UCNV_GET_MAX_BYTES_FOR_STRING(srcLength, ucnv_getMaxCharSize(cnv)).
1.1088 + *
1.1089 + * @param cnv the converter object to be used (ucnv_resetFromUnicode() will be called)
1.1090 + * @param src the input Unicode string
1.1091 + * @param srcLength the input string length, or -1 if NUL-terminated
1.1092 + * @param dest destination string buffer, can be NULL if destCapacity==0
1.1093 + * @param destCapacity the number of chars available at dest
1.1094 + * @param pErrorCode normal ICU error code;
1.1095 + * common error codes that may be set by this function include
1.1096 + * U_BUFFER_OVERFLOW_ERROR, U_STRING_NOT_TERMINATED_WARNING,
1.1097 + * U_ILLEGAL_ARGUMENT_ERROR, and conversion errors
1.1098 + * @return the length of the output string, not counting the terminating NUL;
1.1099 + * if the length is greater than destCapacity, then the string will not fit
1.1100 + * and a buffer of the indicated length would need to be passed in
1.1101 + * @see ucnv_fromUnicode
1.1102 + * @see ucnv_convert
1.1103 + * @see UCNV_GET_MAX_BYTES_FOR_STRING
1.1104 + * @stable ICU 2.0
1.1105 + */
1.1106 +U_STABLE int32_t U_EXPORT2
1.1107 +ucnv_fromUChars(UConverter *cnv,
1.1108 + char *dest, int32_t destCapacity,
1.1109 + const UChar *src, int32_t srcLength,
1.1110 + UErrorCode *pErrorCode);
1.1111 +
1.1112 +/**
1.1113 + * Convert the codepage string into a Unicode string using an existing UConverter.
1.1114 + * The output string is NUL-terminated if possible.
1.1115 + *
1.1116 + * This function is a more convenient but less powerful version of ucnv_toUnicode().
1.1117 + * It is only useful for whole strings, not for streaming conversion.
1.1118 + *
1.1119 + * The maximum output buffer capacity required (barring output from callbacks) will be
1.1120 + * 2*srcLength (each char may be converted into a surrogate pair).
1.1121 + *
1.1122 + * @param cnv the converter object to be used (ucnv_resetToUnicode() will be called)
1.1123 + * @param src the input codepage string
1.1124 + * @param srcLength the input string length, or -1 if NUL-terminated
1.1125 + * @param dest destination string buffer, can be NULL if destCapacity==0
1.1126 + * @param destCapacity the number of UChars available at dest
1.1127 + * @param pErrorCode normal ICU error code;
1.1128 + * common error codes that may be set by this function include
1.1129 + * U_BUFFER_OVERFLOW_ERROR, U_STRING_NOT_TERMINATED_WARNING,
1.1130 + * U_ILLEGAL_ARGUMENT_ERROR, and conversion errors
1.1131 + * @return the length of the output string, not counting the terminating NUL;
1.1132 + * if the length is greater than destCapacity, then the string will not fit
1.1133 + * and a buffer of the indicated length would need to be passed in
1.1134 + * @see ucnv_toUnicode
1.1135 + * @see ucnv_convert
1.1136 + * @stable ICU 2.0
1.1137 + */
1.1138 +U_STABLE int32_t U_EXPORT2
1.1139 +ucnv_toUChars(UConverter *cnv,
1.1140 + UChar *dest, int32_t destCapacity,
1.1141 + const char *src, int32_t srcLength,
1.1142 + UErrorCode *pErrorCode);
1.1143 +
1.1144 +/**
1.1145 + * Convert a codepage buffer into Unicode one character at a time.
1.1146 + * The input is completely consumed when the U_INDEX_OUTOFBOUNDS_ERROR is set.
1.1147 + *
1.1148 + * Advantage compared to ucnv_toUnicode() or ucnv_toUChars():
1.1149 + * - Faster for small amounts of data, for most converters, e.g.,
1.1150 + * US-ASCII, ISO-8859-1, UTF-8/16/32, and most "normal" charsets.
1.1151 + * (For complex converters, e.g., SCSU, UTF-7 and ISO 2022 variants,
1.1152 + * it uses ucnv_toUnicode() internally.)
1.1153 + * - Convenient.
1.1154 + *
1.1155 + * Limitations compared to ucnv_toUnicode():
1.1156 + * - Always assumes flush=TRUE.
1.1157 + * This makes ucnv_getNextUChar() unsuitable for "streaming" conversion,
1.1158 + * that is, for where the input is supplied in multiple buffers,
1.1159 + * because ucnv_getNextUChar() will assume the end of the input at the end
1.1160 + * of the first buffer.
1.1161 + * - Does not provide offset output.
1.1162 + *
1.1163 + * It is possible to "mix" ucnv_getNextUChar() and ucnv_toUnicode() because
1.1164 + * ucnv_getNextUChar() uses the current state of the converter
1.1165 + * (unlike ucnv_toUChars() which always resets first).
1.1166 + * However, if ucnv_getNextUChar() is called after ucnv_toUnicode()
1.1167 + * stopped in the middle of a character sequence (with flush=FALSE),
1.1168 + * then ucnv_getNextUChar() will always use the slower ucnv_toUnicode()
1.1169 + * internally until the next character boundary.
1.1170 + * (This is new in ICU 2.6. In earlier releases, ucnv_getNextUChar() had to
1.1171 + * start at a character boundary.)
1.1172 + *
1.1173 + * Instead of using ucnv_getNextUChar(), it is recommended
1.1174 + * to convert using ucnv_toUnicode() or ucnv_toUChars()
1.1175 + * and then iterate over the text using U16_NEXT() or a UCharIterator (uiter.h)
1.1176 + * or a C++ CharacterIterator or similar.
1.1177 + * This allows streaming conversion and offset output, for example.
1.1178 + *
1.1179 + * <p>Handling of surrogate pairs and supplementary-plane code points:<br>
1.1180 + * There are two different kinds of codepages that provide mappings for surrogate characters:
1.1181 + * <ul>
1.1182 + * <li>Codepages like UTF-8, UTF-32, and GB 18030 provide direct representations for Unicode
1.1183 + * code points U+10000-U+10ffff as well as for single surrogates U+d800-U+dfff.
1.1184 + * Each valid sequence will result in exactly one returned code point.
1.1185 + * If a sequence results in a single surrogate, then that will be returned
1.1186 + * by itself, even if a neighboring sequence encodes the matching surrogate.</li>
1.1187 + * <li>Codepages like SCSU and LMBCS (and UTF-16) provide direct representations only for BMP code points
1.1188 + * including surrogates. Code points in supplementary planes are represented with
1.1189 + * two sequences, each encoding a surrogate.
1.1190 + * For these codepages, matching pairs of surrogates will be combined into single
1.1191 + * code points for returning from this function.
1.1192 + * (Note that SCSU is actually a mix of these codepage types.)</li>
1.1193 + * </ul></p>
1.1194 + *
1.1195 + * @param converter an open UConverter
1.1196 + * @param source the address of a pointer to the codepage buffer, will be
1.1197 + * updated to point after the bytes consumed in the conversion call.
1.1198 + * @param sourceLimit points to the end of the input buffer
1.1199 + * @param err fills in error status (see ucnv_toUnicode)
1.1200 + * <code>U_INDEX_OUTOFBOUNDS_ERROR</code> will be set if the input
1.1201 + * is empty or does not convert to any output (e.g.: pure state-change
1.1202 + * codes SI/SO, escape sequences for ISO 2022,
1.1203 + * or if the callback did not output anything, ...).
1.1204 + * This function will not set a <code>U_BUFFER_OVERFLOW_ERROR</code> because
1.1205 + * the "buffer" is the return code. However, there might be subsequent output
1.1206 + * stored in the converter object
1.1207 + * that will be returned in following calls to this function.
1.1208 + * @return a UChar32 resulting from the partial conversion of source
1.1209 + * @see ucnv_toUnicode
1.1210 + * @see ucnv_toUChars
1.1211 + * @see ucnv_convert
1.1212 + * @stable ICU 2.0
1.1213 + */
1.1214 +U_STABLE UChar32 U_EXPORT2
1.1215 +ucnv_getNextUChar(UConverter * converter,
1.1216 + const char **source,
1.1217 + const char * sourceLimit,
1.1218 + UErrorCode * err);
1.1219 +
1.1220 +/**
1.1221 + * Convert from one external charset to another using two existing UConverters.
1.1222 + * Internally, two conversions - ucnv_toUnicode() and ucnv_fromUnicode() -
1.1223 + * are used, "pivoting" through 16-bit Unicode.
1.1224 + *
1.1225 + * There is a similar function, ucnv_convert(),
1.1226 + * which has the following limitations:
1.1227 + * - it takes charset names, not converter objects, so that
1.1228 + * - two converters are opened for each call
1.1229 + * - only single-string conversion is possible, not streaming operation
1.1230 + * - it does not provide enough information to find out,
1.1231 + * in case of failure, whether the toUnicode or
1.1232 + * the fromUnicode conversion failed
1.1233 + *
1.1234 + * By contrast, ucnv_convertEx()
1.1235 + * - takes UConverter parameters instead of charset names
1.1236 + * - fully exposes the pivot buffer for complete error handling
1.1237 + *
1.1238 + * ucnv_convertEx() also provides further convenience:
1.1239 + * - an option to reset the converters at the beginning
1.1240 + * (if reset==TRUE, see parameters;
1.1241 + * also sets *pivotTarget=*pivotSource=pivotStart)
1.1242 + * - allow NUL-terminated input
1.1243 + * (only a single NUL byte, will not work for charsets with multi-byte NULs)
1.1244 + * (if sourceLimit==NULL, see parameters)
1.1245 + * - terminate with a NUL on output
1.1246 + * (only a single NUL byte, not useful for charsets with multi-byte NULs),
1.1247 + * or set U_STRING_NOT_TERMINATED_WARNING if the output exactly fills
1.1248 + * the target buffer
1.1249 + * - the pivot buffer can be provided internally;
1.1250 + * in this case, the caller will not be able to get details about where an
1.1251 + * error occurred
1.1252 + * (if pivotStart==NULL, see below)
1.1253 + *
1.1254 + * The function returns when one of the following is true:
1.1255 + * - the entire source text has been converted successfully to the target buffer
1.1256 + * - a target buffer overflow occurred (U_BUFFER_OVERFLOW_ERROR)
1.1257 + * - a conversion error occurred
1.1258 + * (other U_FAILURE(), see description of pErrorCode)
1.1259 + *
1.1260 + * Limitation compared to the direct use of
1.1261 + * ucnv_fromUnicode() and ucnv_toUnicode():
1.1262 + * ucnv_convertEx() does not provide offset information.
1.1263 + *
1.1264 + * Limitation compared to ucnv_fromUChars() and ucnv_toUChars():
1.1265 + * ucnv_convertEx() does not support preflighting directly.
1.1266 + *
1.1267 + * Sample code for converting a single string from
1.1268 + * one external charset to UTF-8, ignoring the location of errors:
1.1269 + *
1.1270 + * \code
1.1271 + * int32_t
1.1272 + * myToUTF8(UConverter *cnv,
1.1273 + * const char *s, int32_t length,
1.1274 + * char *u8, int32_t capacity,
1.1275 + * UErrorCode *pErrorCode) {
1.1276 + * UConverter *utf8Cnv;
1.1277 + * char *target;
1.1278 + *
1.1279 + * if(U_FAILURE(*pErrorCode)) {
1.1280 + * return 0;
1.1281 + * }
1.1282 + *
1.1283 + * utf8Cnv=myGetCachedUTF8Converter(pErrorCode);
1.1284 + * if(U_FAILURE(*pErrorCode)) {
1.1285 + * return 0;
1.1286 + * }
1.1287 + *
1.1288 + * target=u8;
1.1289 + * ucnv_convertEx(cnv, utf8Cnv,
1.1290 + * &target, u8+capacity,
1.1291 + * &s, length>=0 ? s+length : NULL,
1.1292 + * NULL, NULL, NULL, NULL,
1.1293 + * TRUE, TRUE,
1.1294 + * pErrorCode);
1.1295 + *
1.1296 + * myReleaseCachedUTF8Converter(utf8Cnv);
1.1297 + *
1.1298 + * // return the output string length, but without preflighting
1.1299 + * return (int32_t)(target-u8);
1.1300 + * }
1.1301 + * \endcode
1.1302 + *
1.1303 + * @param targetCnv Output converter, used to convert from the UTF-16 pivot
1.1304 + * to the target using ucnv_fromUnicode().
1.1305 + * @param sourceCnv Input converter, used to convert from the source to
1.1306 + * the UTF-16 pivot using ucnv_toUnicode().
1.1307 + * @param target I/O parameter, same as for ucnv_fromUChars().
1.1308 + * Input: *target points to the beginning of the target buffer.
1.1309 + * Output: *target points to the first unit after the last char written.
1.1310 + * @param targetLimit Pointer to the first unit after the target buffer.
1.1311 + * @param source I/O parameter, same as for ucnv_toUChars().
1.1312 + * Input: *source points to the beginning of the source buffer.
1.1313 + * Output: *source points to the first unit after the last char read.
1.1314 + * @param sourceLimit Pointer to the first unit after the source buffer.
1.1315 + * @param pivotStart Pointer to the UTF-16 pivot buffer. If pivotStart==NULL,
1.1316 + * then an internal buffer is used and the other pivot
1.1317 + * arguments are ignored and can be NULL as well.
1.1318 + * @param pivotSource I/O parameter, same as source in ucnv_fromUChars() for
1.1319 + * conversion from the pivot buffer to the target buffer.
1.1320 + * @param pivotTarget I/O parameter, same as target in ucnv_toUChars() for
1.1321 + * conversion from the source buffer to the pivot buffer.
1.1322 + * It must be pivotStart<=*pivotSource<=*pivotTarget<=pivotLimit
1.1323 + * and pivotStart<pivotLimit (unless pivotStart==NULL).
1.1324 + * @param pivotLimit Pointer to the first unit after the pivot buffer.
1.1325 + * @param reset If TRUE, then ucnv_resetToUnicode(sourceCnv) and
1.1326 + * ucnv_resetFromUnicode(targetCnv) are called, and the
1.1327 + * pivot pointers are reset (*pivotTarget=*pivotSource=pivotStart).
1.1328 + * @param flush If true, indicates the end of the input.
1.1329 + * Passed directly to ucnv_toUnicode(), and carried over to
1.1330 + * ucnv_fromUnicode() when the source is empty as well.
1.1331 + * @param pErrorCode ICU error code in/out parameter.
1.1332 + * Must fulfill U_SUCCESS before the function call.
1.1333 + * U_BUFFER_OVERFLOW_ERROR always refers to the target buffer
1.1334 + * because overflows into the pivot buffer are handled internally.
1.1335 + * Other conversion errors are from the source-to-pivot
1.1336 + * conversion if *pivotSource==pivotStart, otherwise from
1.1337 + * the pivot-to-target conversion.
1.1338 + *
1.1339 + * @see ucnv_convert
1.1340 + * @see ucnv_fromAlgorithmic
1.1341 + * @see ucnv_toAlgorithmic
1.1342 + * @see ucnv_fromUnicode
1.1343 + * @see ucnv_toUnicode
1.1344 + * @see ucnv_fromUChars
1.1345 + * @see ucnv_toUChars
1.1346 + * @stable ICU 2.6
1.1347 + */
1.1348 +U_STABLE void U_EXPORT2
1.1349 +ucnv_convertEx(UConverter *targetCnv, UConverter *sourceCnv,
1.1350 + char **target, const char *targetLimit,
1.1351 + const char **source, const char *sourceLimit,
1.1352 + UChar *pivotStart, UChar **pivotSource,
1.1353 + UChar **pivotTarget, const UChar *pivotLimit,
1.1354 + UBool reset, UBool flush,
1.1355 + UErrorCode *pErrorCode);
1.1356 +
1.1357 +/**
1.1358 + * Convert from one external charset to another.
1.1359 + * Internally, two converters are opened according to the name arguments,
1.1360 + * then the text is converted to and from the 16-bit Unicode "pivot"
1.1361 + * using ucnv_convertEx(), then the converters are closed again.
1.1362 + *
1.1363 + * This is a convenience function, not an efficient way to convert a lot of text:
1.1364 + * ucnv_convert()
1.1365 + * - takes charset names, not converter objects, so that
1.1366 + * - two converters are opened for each call
1.1367 + * - only single-string conversion is possible, not streaming operation
1.1368 + * - does not provide enough information to find out,
1.1369 + * in case of failure, whether the toUnicode or
1.1370 + * the fromUnicode conversion failed
1.1371 + * - allows NUL-terminated input
1.1372 + * (only a single NUL byte, will not work for charsets with multi-byte NULs)
1.1373 + * (if sourceLength==-1, see parameters)
1.1374 + * - terminate with a NUL on output
1.1375 + * (only a single NUL byte, not useful for charsets with multi-byte NULs),
1.1376 + * or set U_STRING_NOT_TERMINATED_WARNING if the output exactly fills
1.1377 + * the target buffer
1.1378 + * - a pivot buffer is provided internally
1.1379 + *
1.1380 + * The function returns when one of the following is true:
1.1381 + * - the entire source text has been converted successfully to the target buffer
1.1382 + * and either the target buffer is terminated with a single NUL byte
1.1383 + * or the error code is set to U_STRING_NOT_TERMINATED_WARNING
1.1384 + * - a target buffer overflow occurred (U_BUFFER_OVERFLOW_ERROR)
1.1385 + * and the full output string length is returned ("preflighting")
1.1386 + * - a conversion error occurred
1.1387 + * (other U_FAILURE(), see description of pErrorCode)
1.1388 + *
1.1389 + * @param toConverterName The name of the converter that is used to convert
1.1390 + * from the UTF-16 pivot buffer to the target.
1.1391 + * @param fromConverterName The name of the converter that is used to convert
1.1392 + * from the source to the UTF-16 pivot buffer.
1.1393 + * @param target Pointer to the output buffer.
1.1394 + * @param targetCapacity Capacity of the target, in bytes.
1.1395 + * @param source Pointer to the input buffer.
1.1396 + * @param sourceLength Length of the input text, in bytes, or -1 for NUL-terminated input.
1.1397 + * @param pErrorCode ICU error code in/out parameter.
1.1398 + * Must fulfill U_SUCCESS before the function call.
1.1399 + * @return Length of the complete output text in bytes, even if it exceeds the targetCapacity
1.1400 + * and a U_BUFFER_OVERFLOW_ERROR is set.
1.1401 + *
1.1402 + * @see ucnv_convertEx
1.1403 + * @see ucnv_fromAlgorithmic
1.1404 + * @see ucnv_toAlgorithmic
1.1405 + * @see ucnv_fromUnicode
1.1406 + * @see ucnv_toUnicode
1.1407 + * @see ucnv_fromUChars
1.1408 + * @see ucnv_toUChars
1.1409 + * @see ucnv_getNextUChar
1.1410 + * @stable ICU 2.0
1.1411 + */
1.1412 +U_STABLE int32_t U_EXPORT2
1.1413 +ucnv_convert(const char *toConverterName,
1.1414 + const char *fromConverterName,
1.1415 + char *target,
1.1416 + int32_t targetCapacity,
1.1417 + const char *source,
1.1418 + int32_t sourceLength,
1.1419 + UErrorCode *pErrorCode);
1.1420 +
1.1421 +/**
1.1422 + * Convert from one external charset to another.
1.1423 + * Internally, the text is converted to and from the 16-bit Unicode "pivot"
1.1424 + * using ucnv_convertEx(). ucnv_toAlgorithmic() works exactly like ucnv_convert()
1.1425 + * except that the two converters need not be looked up and opened completely.
1.1426 + *
1.1427 + * The source-to-pivot conversion uses the cnv converter parameter.
1.1428 + * The pivot-to-target conversion uses a purely algorithmic converter
1.1429 + * according to the specified type, e.g., UCNV_UTF8 for a UTF-8 converter.
1.1430 + *
1.1431 + * Internally, the algorithmic converter is opened and closed for each
1.1432 + * function call, which is more efficient than using the public ucnv_open()
1.1433 + * but somewhat less efficient than only resetting an existing converter
1.1434 + * and using ucnv_convertEx().
1.1435 + *
1.1436 + * This function is more convenient than ucnv_convertEx() for single-string
1.1437 + * conversions, especially when "preflighting" is desired (returning the length
1.1438 + * of the complete output even if it does not fit into the target buffer;
1.1439 + * see the User Guide Strings chapter). See ucnv_convert() for details.
1.1440 + *
1.1441 + * @param algorithmicType UConverterType constant identifying the desired target
1.1442 + * charset as a purely algorithmic converter.
1.1443 + * Those are converters for Unicode charsets like
1.1444 + * UTF-8, BOCU-1, SCSU, UTF-7, IMAP-mailbox-name, etc.,
1.1445 + * as well as US-ASCII and ISO-8859-1.
1.1446 + * @param cnv The converter that is used to convert
1.1447 + * from the source to the UTF-16 pivot buffer.
1.1448 + * @param target Pointer to the output buffer.
1.1449 + * @param targetCapacity Capacity of the target, in bytes.
1.1450 + * @param source Pointer to the input buffer.
1.1451 + * @param sourceLength Length of the input text, in bytes
1.1452 + * @param pErrorCode ICU error code in/out parameter.
1.1453 + * Must fulfill U_SUCCESS before the function call.
1.1454 + * @return Length of the complete output text in bytes, even if it exceeds the targetCapacity
1.1455 + * and a U_BUFFER_OVERFLOW_ERROR is set.
1.1456 + *
1.1457 + * @see ucnv_fromAlgorithmic
1.1458 + * @see ucnv_convert
1.1459 + * @see ucnv_convertEx
1.1460 + * @see ucnv_fromUnicode
1.1461 + * @see ucnv_toUnicode
1.1462 + * @see ucnv_fromUChars
1.1463 + * @see ucnv_toUChars
1.1464 + * @stable ICU 2.6
1.1465 + */
1.1466 +U_STABLE int32_t U_EXPORT2
1.1467 +ucnv_toAlgorithmic(UConverterType algorithmicType,
1.1468 + UConverter *cnv,
1.1469 + char *target, int32_t targetCapacity,
1.1470 + const char *source, int32_t sourceLength,
1.1471 + UErrorCode *pErrorCode);
1.1472 +
1.1473 +/**
1.1474 + * Convert from one external charset to another.
1.1475 + * Internally, the text is converted to and from the 16-bit Unicode "pivot"
1.1476 + * using ucnv_convertEx(). ucnv_fromAlgorithmic() works exactly like ucnv_convert()
1.1477 + * except that the two converters need not be looked up and opened completely.
1.1478 + *
1.1479 + * The source-to-pivot conversion uses a purely algorithmic converter
1.1480 + * according to the specified type, e.g., UCNV_UTF8 for a UTF-8 converter.
1.1481 + * The pivot-to-target conversion uses the cnv converter parameter.
1.1482 + *
1.1483 + * Internally, the algorithmic converter is opened and closed for each
1.1484 + * function call, which is more efficient than using the public ucnv_open()
1.1485 + * but somewhat less efficient than only resetting an existing converter
1.1486 + * and using ucnv_convertEx().
1.1487 + *
1.1488 + * This function is more convenient than ucnv_convertEx() for single-string
1.1489 + * conversions, especially when "preflighting" is desired (returning the length
1.1490 + * of the complete output even if it does not fit into the target buffer;
1.1491 + * see the User Guide Strings chapter). See ucnv_convert() for details.
1.1492 + *
1.1493 + * @param cnv The converter that is used to convert
1.1494 + * from the UTF-16 pivot buffer to the target.
1.1495 + * @param algorithmicType UConverterType constant identifying the desired source
1.1496 + * charset as a purely algorithmic converter.
1.1497 + * Those are converters for Unicode charsets like
1.1498 + * UTF-8, BOCU-1, SCSU, UTF-7, IMAP-mailbox-name, etc.,
1.1499 + * as well as US-ASCII and ISO-8859-1.
1.1500 + * @param target Pointer to the output buffer.
1.1501 + * @param targetCapacity Capacity of the target, in bytes.
1.1502 + * @param source Pointer to the input buffer.
1.1503 + * @param sourceLength Length of the input text, in bytes
1.1504 + * @param pErrorCode ICU error code in/out parameter.
1.1505 + * Must fulfill U_SUCCESS before the function call.
1.1506 + * @return Length of the complete output text in bytes, even if it exceeds the targetCapacity
1.1507 + * and a U_BUFFER_OVERFLOW_ERROR is set.
1.1508 + *
1.1509 + * @see ucnv_fromAlgorithmic
1.1510 + * @see ucnv_convert
1.1511 + * @see ucnv_convertEx
1.1512 + * @see ucnv_fromUnicode
1.1513 + * @see ucnv_toUnicode
1.1514 + * @see ucnv_fromUChars
1.1515 + * @see ucnv_toUChars
1.1516 + * @stable ICU 2.6
1.1517 + */
1.1518 +U_STABLE int32_t U_EXPORT2
1.1519 +ucnv_fromAlgorithmic(UConverter *cnv,
1.1520 + UConverterType algorithmicType,
1.1521 + char *target, int32_t targetCapacity,
1.1522 + const char *source, int32_t sourceLength,
1.1523 + UErrorCode *pErrorCode);
1.1524 +
1.1525 +/**
1.1526 + * Frees up memory occupied by unused, cached converter shared data.
1.1527 + *
1.1528 + * @return the number of cached converters successfully deleted
1.1529 + * @see ucnv_close
1.1530 + * @stable ICU 2.0
1.1531 + */
1.1532 +U_STABLE int32_t U_EXPORT2
1.1533 +ucnv_flushCache(void);
1.1534 +
1.1535 +/**
1.1536 + * Returns the number of available converters, as per the alias file.
1.1537 + *
1.1538 + * @return the number of available converters
1.1539 + * @see ucnv_getAvailableName
1.1540 + * @stable ICU 2.0
1.1541 + */
1.1542 +U_STABLE int32_t U_EXPORT2
1.1543 +ucnv_countAvailable(void);
1.1544 +
1.1545 +/**
1.1546 + * Gets the canonical converter name of the specified converter from a list of
1.1547 + * all available converters contaied in the alias file. All converters
1.1548 + * in this list can be opened.
1.1549 + *
1.1550 + * @param n the index to a converter available on the system (in the range <TT>[0..ucnv_countAvaiable()]</TT>)
1.1551 + * @return a pointer a string (library owned), or <TT>NULL</TT> if the index is out of bounds.
1.1552 + * @see ucnv_countAvailable
1.1553 + * @stable ICU 2.0
1.1554 + */
1.1555 +U_STABLE const char* U_EXPORT2
1.1556 +ucnv_getAvailableName(int32_t n);
1.1557 +
1.1558 +/**
1.1559 + * Returns a UEnumeration to enumerate all of the canonical converter
1.1560 + * names, as per the alias file, regardless of the ability to open each
1.1561 + * converter.
1.1562 + *
1.1563 + * @return A UEnumeration object for getting all the recognized canonical
1.1564 + * converter names.
1.1565 + * @see ucnv_getAvailableName
1.1566 + * @see uenum_close
1.1567 + * @see uenum_next
1.1568 + * @stable ICU 2.4
1.1569 + */
1.1570 +U_STABLE UEnumeration * U_EXPORT2
1.1571 +ucnv_openAllNames(UErrorCode *pErrorCode);
1.1572 +
1.1573 +/**
1.1574 + * Gives the number of aliases for a given converter or alias name.
1.1575 + * If the alias is ambiguous, then the preferred converter is used
1.1576 + * and the status is set to U_AMBIGUOUS_ALIAS_WARNING.
1.1577 + * This method only enumerates the listed entries in the alias file.
1.1578 + * @param alias alias name
1.1579 + * @param pErrorCode error status
1.1580 + * @return number of names on alias list for given alias
1.1581 + * @stable ICU 2.0
1.1582 + */
1.1583 +U_STABLE uint16_t U_EXPORT2
1.1584 +ucnv_countAliases(const char *alias, UErrorCode *pErrorCode);
1.1585 +
1.1586 +/**
1.1587 + * Gives the name of the alias at given index of alias list.
1.1588 + * This method only enumerates the listed entries in the alias file.
1.1589 + * If the alias is ambiguous, then the preferred converter is used
1.1590 + * and the status is set to U_AMBIGUOUS_ALIAS_WARNING.
1.1591 + * @param alias alias name
1.1592 + * @param n index in alias list
1.1593 + * @param pErrorCode result of operation
1.1594 + * @return returns the name of the alias at given index
1.1595 + * @see ucnv_countAliases
1.1596 + * @stable ICU 2.0
1.1597 + */
1.1598 +U_STABLE const char * U_EXPORT2
1.1599 +ucnv_getAlias(const char *alias, uint16_t n, UErrorCode *pErrorCode);
1.1600 +
1.1601 +/**
1.1602 + * Fill-up the list of alias names for the given alias.
1.1603 + * This method only enumerates the listed entries in the alias file.
1.1604 + * If the alias is ambiguous, then the preferred converter is used
1.1605 + * and the status is set to U_AMBIGUOUS_ALIAS_WARNING.
1.1606 + * @param alias alias name
1.1607 + * @param aliases fill-in list, aliases is a pointer to an array of
1.1608 + * <code>ucnv_countAliases()</code> string-pointers
1.1609 + * (<code>const char *</code>) that will be filled in.
1.1610 + * The strings themselves are owned by the library.
1.1611 + * @param pErrorCode result of operation
1.1612 + * @stable ICU 2.0
1.1613 + */
1.1614 +U_STABLE void U_EXPORT2
1.1615 +ucnv_getAliases(const char *alias, const char **aliases, UErrorCode *pErrorCode);
1.1616 +
1.1617 +/**
1.1618 + * Return a new UEnumeration object for enumerating all the
1.1619 + * alias names for a given converter that are recognized by a standard.
1.1620 + * This method only enumerates the listed entries in the alias file.
1.1621 + * The convrtrs.txt file can be modified to change the results of
1.1622 + * this function.
1.1623 + * The first result in this list is the same result given by
1.1624 + * <code>ucnv_getStandardName</code>, which is the default alias for
1.1625 + * the specified standard name. The returned object must be closed with
1.1626 + * <code>uenum_close</code> when you are done with the object.
1.1627 + *
1.1628 + * @param convName original converter name
1.1629 + * @param standard name of the standard governing the names; MIME and IANA
1.1630 + * are such standards
1.1631 + * @param pErrorCode The error code
1.1632 + * @return A UEnumeration object for getting all aliases that are recognized
1.1633 + * by a standard. If any of the parameters are invalid, NULL
1.1634 + * is returned.
1.1635 + * @see ucnv_getStandardName
1.1636 + * @see uenum_close
1.1637 + * @see uenum_next
1.1638 + * @stable ICU 2.2
1.1639 + */
1.1640 +U_STABLE UEnumeration * U_EXPORT2
1.1641 +ucnv_openStandardNames(const char *convName,
1.1642 + const char *standard,
1.1643 + UErrorCode *pErrorCode);
1.1644 +
1.1645 +/**
1.1646 + * Gives the number of standards associated to converter names.
1.1647 + * @return number of standards
1.1648 + * @stable ICU 2.0
1.1649 + */
1.1650 +U_STABLE uint16_t U_EXPORT2
1.1651 +ucnv_countStandards(void);
1.1652 +
1.1653 +/**
1.1654 + * Gives the name of the standard at given index of standard list.
1.1655 + * @param n index in standard list
1.1656 + * @param pErrorCode result of operation
1.1657 + * @return returns the name of the standard at given index. Owned by the library.
1.1658 + * @stable ICU 2.0
1.1659 + */
1.1660 +U_STABLE const char * U_EXPORT2
1.1661 +ucnv_getStandard(uint16_t n, UErrorCode *pErrorCode);
1.1662 +
1.1663 +/**
1.1664 + * Returns a standard name for a given converter name.
1.1665 + * <p>
1.1666 + * Example alias table:<br>
1.1667 + * conv alias1 { STANDARD1 } alias2 { STANDARD1* }
1.1668 + * <p>
1.1669 + * Result of ucnv_getStandardName("conv", "STANDARD1") from example
1.1670 + * alias table:<br>
1.1671 + * <b>"alias2"</b>
1.1672 + *
1.1673 + * @param name original converter name
1.1674 + * @param standard name of the standard governing the names; MIME and IANA
1.1675 + * are such standards
1.1676 + * @param pErrorCode result of operation
1.1677 + * @return returns the standard converter name;
1.1678 + * if a standard converter name cannot be determined,
1.1679 + * then <code>NULL</code> is returned. Owned by the library.
1.1680 + * @stable ICU 2.0
1.1681 + */
1.1682 +U_STABLE const char * U_EXPORT2
1.1683 +ucnv_getStandardName(const char *name, const char *standard, UErrorCode *pErrorCode);
1.1684 +
1.1685 +/**
1.1686 + * This function will return the internal canonical converter name of the
1.1687 + * tagged alias. This is the opposite of ucnv_openStandardNames, which
1.1688 + * returns the tagged alias given the canonical name.
1.1689 + * <p>
1.1690 + * Example alias table:<br>
1.1691 + * conv alias1 { STANDARD1 } alias2 { STANDARD1* }
1.1692 + * <p>
1.1693 + * Result of ucnv_getStandardName("alias1", "STANDARD1") from example
1.1694 + * alias table:<br>
1.1695 + * <b>"conv"</b>
1.1696 + *
1.1697 + * @return returns the canonical converter name;
1.1698 + * if a standard or alias name cannot be determined,
1.1699 + * then <code>NULL</code> is returned. The returned string is
1.1700 + * owned by the library.
1.1701 + * @see ucnv_getStandardName
1.1702 + * @stable ICU 2.4
1.1703 + */
1.1704 +U_STABLE const char * U_EXPORT2
1.1705 +ucnv_getCanonicalName(const char *alias, const char *standard, UErrorCode *pErrorCode);
1.1706 +
1.1707 +/**
1.1708 + * returns the current default converter name.
1.1709 + *
1.1710 + * @return returns the current default converter name;
1.1711 + * if a default converter name cannot be determined,
1.1712 + * then <code>NULL</code> is returned.
1.1713 + * Storage owned by the library
1.1714 + * @see ucnv_setDefaultName
1.1715 + * @stable ICU 2.0
1.1716 + */
1.1717 +U_STABLE const char * U_EXPORT2
1.1718 +ucnv_getDefaultName(void);
1.1719 +
1.1720 +/**
1.1721 + * sets the current default converter name. Caller must own the storage for 'name'
1.1722 + * and preserve it indefinitely.
1.1723 + * @param name the converter name to be the default (must exist).
1.1724 + * @see ucnv_getDefaultName
1.1725 + * @system SYSTEM API
1.1726 + * @stable ICU 2.0
1.1727 + */
1.1728 +U_STABLE void U_EXPORT2
1.1729 +ucnv_setDefaultName(const char *name);
1.1730 +
1.1731 +/**
1.1732 + * Fixes the backslash character mismapping. For example, in SJIS, the backslash
1.1733 + * character in the ASCII portion is also used to represent the yen currency sign.
1.1734 + * When mapping from Unicode character 0x005C, it's unclear whether to map the
1.1735 + * character back to yen or backslash in SJIS. This function will take the input
1.1736 + * buffer and replace all the yen sign characters with backslash. This is necessary
1.1737 + * when the user tries to open a file with the input buffer on Windows.
1.1738 + * This function will test the converter to see whether such mapping is
1.1739 + * required. You can sometimes avoid using this function by using the correct version
1.1740 + * of Shift-JIS.
1.1741 + *
1.1742 + * @param cnv The converter representing the target codepage.
1.1743 + * @param source the input buffer to be fixed
1.1744 + * @param sourceLen the length of the input buffer
1.1745 + * @see ucnv_isAmbiguous
1.1746 + * @stable ICU 2.0
1.1747 + */
1.1748 +U_STABLE void U_EXPORT2
1.1749 +ucnv_fixFileSeparator(const UConverter *cnv, UChar *source, int32_t sourceLen);
1.1750 +
1.1751 +/**
1.1752 + * Determines if the converter contains ambiguous mappings of the same
1.1753 + * character or not.
1.1754 + * @param cnv the converter to be tested
1.1755 + * @return TRUE if the converter contains ambiguous mapping of the same
1.1756 + * character, FALSE otherwise.
1.1757 + * @stable ICU 2.0
1.1758 + */
1.1759 +U_STABLE UBool U_EXPORT2
1.1760 +ucnv_isAmbiguous(const UConverter *cnv);
1.1761 +
1.1762 +/**
1.1763 + * Sets the converter to use fallback mapping or not.
1.1764 + * @param cnv The converter to set the fallback mapping usage on.
1.1765 + * @param usesFallback TRUE if the user wants the converter to take advantage of the fallback
1.1766 + * mapping, FALSE otherwise.
1.1767 + * @stable ICU 2.0
1.1768 + */
1.1769 +U_STABLE void U_EXPORT2
1.1770 +ucnv_setFallback(UConverter *cnv, UBool usesFallback);
1.1771 +
1.1772 +/**
1.1773 + * Determines if the converter uses fallback mappings or not.
1.1774 + * @param cnv The converter to be tested
1.1775 + * @return TRUE if the converter uses fallback, FALSE otherwise.
1.1776 + * @stable ICU 2.0
1.1777 + */
1.1778 +U_STABLE UBool U_EXPORT2
1.1779 +ucnv_usesFallback(const UConverter *cnv);
1.1780 +
1.1781 +/**
1.1782 + * Detects Unicode signature byte sequences at the start of the byte stream
1.1783 + * and returns the charset name of the indicated Unicode charset.
1.1784 + * NULL is returned when no Unicode signature is recognized.
1.1785 + * The number of bytes in the signature is output as well.
1.1786 + *
1.1787 + * The caller can ucnv_open() a converter using the charset name.
1.1788 + * The first code unit (UChar) from the start of the stream will be U+FEFF
1.1789 + * (the Unicode BOM/signature character) and can usually be ignored.
1.1790 + *
1.1791 + * For most Unicode charsets it is also possible to ignore the indicated
1.1792 + * number of initial stream bytes and start converting after them.
1.1793 + * However, there are stateful Unicode charsets (UTF-7 and BOCU-1) for which
1.1794 + * this will not work. Therefore, it is best to ignore the first output UChar
1.1795 + * instead of the input signature bytes.
1.1796 + * <p>
1.1797 + * Usage:
1.1798 + * @code
1.1799 + * UErrorCode err = U_ZERO_ERROR;
1.1800 + * char input[] = { '\xEF','\xBB', '\xBF','\x41','\x42','\x43' };
1.1801 + * int32_t signatureLength = 0;
1.1802 + * char *encoding = ucnv_detectUnicodeSignatures(input,sizeof(input),&signatureLength,&err);
1.1803 + * UConverter *conv = NULL;
1.1804 + * UChar output[100];
1.1805 + * UChar *target = output, *out;
1.1806 + * char *source = input;
1.1807 + * if(encoding!=NULL && U_SUCCESS(err)){
1.1808 + * // should signature be discarded ?
1.1809 + * conv = ucnv_open(encoding, &err);
1.1810 + * // do the conversion
1.1811 + * ucnv_toUnicode(conv,
1.1812 + * target, output + sizeof(output)/U_SIZEOF_UCHAR,
1.1813 + * source, input + sizeof(input),
1.1814 + * NULL, TRUE, &err);
1.1815 + * out = output;
1.1816 + * if (discardSignature){
1.1817 + * ++out; // ignore initial U+FEFF
1.1818 + * }
1.1819 + * while(out != target) {
1.1820 + * printf("%04x ", *out++);
1.1821 + * }
1.1822 + * puts("");
1.1823 + * }
1.1824 + *
1.1825 + * @endcode
1.1826 + *
1.1827 + * @param source The source string in which the signature should be detected.
1.1828 + * @param sourceLength Length of the input string, or -1 if terminated with a NUL byte.
1.1829 + * @param signatureLength A pointer to int32_t to receive the number of bytes that make up the signature
1.1830 + * of the detected UTF. 0 if not detected.
1.1831 + * Can be a NULL pointer.
1.1832 + * @param pErrorCode ICU error code in/out parameter.
1.1833 + * Must fulfill U_SUCCESS before the function call.
1.1834 + * @return The name of the encoding detected. NULL if encoding is not detected.
1.1835 + * @stable ICU 2.4
1.1836 + */
1.1837 +U_STABLE const char* U_EXPORT2
1.1838 +ucnv_detectUnicodeSignature(const char* source,
1.1839 + int32_t sourceLength,
1.1840 + int32_t *signatureLength,
1.1841 + UErrorCode *pErrorCode);
1.1842 +
1.1843 +/**
1.1844 + * Returns the number of UChars held in the converter's internal state
1.1845 + * because more input is needed for completing the conversion. This function is
1.1846 + * useful for mapping semantics of ICU's converter interface to those of iconv,
1.1847 + * and this information is not needed for normal conversion.
1.1848 + * @param cnv The converter in which the input is held
1.1849 + * @param status ICU error code in/out parameter.
1.1850 + * Must fulfill U_SUCCESS before the function call.
1.1851 + * @return The number of UChars in the state. -1 if an error is encountered.
1.1852 + * @draft ICU 3.4
1.1853 + */
1.1854 +U_DRAFT int32_t U_EXPORT2
1.1855 +ucnv_fromUCountPending(const UConverter* cnv, UErrorCode* status);
1.1856 +
1.1857 +/**
1.1858 + * Returns the number of chars held in the converter's internal state
1.1859 + * because more input is needed for completing the conversion. This function is
1.1860 + * useful for mapping semantics of ICU's converter interface to those of iconv,
1.1861 + * and this information is not needed for normal conversion.
1.1862 + * @param cnv The converter in which the input is held as internal state
1.1863 + * @param status ICU error code in/out parameter.
1.1864 + * Must fulfill U_SUCCESS before the function call.
1.1865 + * @return The number of chars in the state. -1 if an error is encountered.
1.1866 + * @draft ICU 3.4
1.1867 + */
1.1868 +U_DRAFT int32_t U_EXPORT2
1.1869 +ucnv_toUCountPending(const UConverter* cnv, UErrorCode* status);
1.1870 +
1.1871 +#endif
1.1872 +
1.1873 +#endif
1.1874 +/*_UCNV*/