Mercurial > repos > rliterman > csp2

annotate CSP2/CSP2_env/env-d9b9114564458d9d-741b3de822f2aaca6c6caa4325c4afce/include/unicode/ucnv.h @ 69:33d812a61356

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
planemo upload commit 2e9511a184a1ca667c7be0c6321a36dc4e3d116d
author jpayne
date Tue, 18 Mar 2025 17:55:14 -0400
parents
children
rev   line source
jpayne@69 1 // © 2016 and later: Unicode, Inc. and others.
jpayne@69 2 // License & terms of use: http://www.unicode.org/copyright.html
jpayne@69 3 /*
jpayne@69 4 **********************************************************************
jpayne@69 5 * Copyright (C) 1999-2014, International Business Machines
jpayne@69 6 * Corporation and others. All Rights Reserved.
jpayne@69 7 **********************************************************************
jpayne@69 8 * ucnv.h:
jpayne@69 9 * External APIs for the ICU's codeset conversion library
jpayne@69 10 * Bertrand A. Damiba
jpayne@69 11 *
jpayne@69 12 * Modification History:
jpayne@69 13 *
jpayne@69 14 * Date Name Description
jpayne@69 15 * 04/04/99 helena Fixed internal header inclusion.
jpayne@69 16 * 05/11/00 helena Added setFallback and usesFallback APIs.
jpayne@69 17 * 06/29/2000 helena Major rewrite of the callback APIs.
jpayne@69 18 * 12/07/2000 srl Update of documentation
jpayne@69 19 */
jpayne@69 20
jpayne@69 21 /**
jpayne@69 22 * \file
jpayne@69 23 * \brief C API: Character conversion
jpayne@69 24 *
jpayne@69 25 * <h2>Character Conversion C API</h2>
jpayne@69 26 *
jpayne@69 27 * <p>This API is used to convert codepage or character encoded data to and
jpayne@69 28 * from UTF-16. You can open a converter with {@link ucnv_open() }. With that
jpayne@69 29 * converter, you can get its properties, set options, convert your data and
jpayne@69 30 * close the converter.</p>
jpayne@69 31 *
jpayne@69 32 * <p>Since many software programs recognize different converter names for
jpayne@69 33 * different types of converters, there are other functions in this API to
jpayne@69 34 * iterate over the converter aliases. The functions {@link ucnv_getAvailableName() },
jpayne@69 35 * {@link ucnv_getAlias() } and {@link ucnv_getStandardName() } are some of the
jpayne@69 36 * more frequently used alias functions to get this information.</p>
jpayne@69 37 *
jpayne@69 38 * <p>When a converter encounters an illegal, irregular, invalid or unmappable character
jpayne@69 39 * its default behavior is to use a substitution character to replace the
jpayne@69 40 * bad byte sequence. This behavior can be changed by using {@link ucnv_setFromUCallBack() }
jpayne@69 41 * or {@link ucnv_setToUCallBack() } on the converter. The header ucnv_err.h defines
jpayne@69 42 * many other callback actions that can be used instead of a character substitution.</p>
jpayne@69 43 *
jpayne@69 44 * <p>More information about this API can be found in our
jpayne@69 45 * <a href="http://icu-project.org/userguide/conversion.html">User's
jpayne@69 46 * Guide</a>.</p>
jpayne@69 47 */
jpayne@69 48
jpayne@69 49 #ifndef UCNV_H
jpayne@69 50 #define UCNV_H
jpayne@69 51
jpayne@69 52 #include "unicode/ucnv_err.h"
jpayne@69 53 #include "unicode/uenum.h"
jpayne@69 54 #include "unicode/localpointer.h"
jpayne@69 55
jpayne@69 56 #if !defined(USET_DEFINED) && !defined(U_IN_DOXYGEN)
jpayne@69 57
jpayne@69 58 #define USET_DEFINED
jpayne@69 59
jpayne@69 60 /**
jpayne@69 61 * USet is the C API type corresponding to C++ class UnicodeSet.
jpayne@69 62 * It is forward-declared here to avoid including unicode/uset.h file if related
jpayne@69 63 * conversion APIs are not used.
jpayne@69 64 *
jpayne@69 65 * @see ucnv_getUnicodeSet
jpayne@69 66 * @stable ICU 2.4
jpayne@69 67 */
jpayne@69 68 typedef struct USet USet;
jpayne@69 69
jpayne@69 70 #endif
jpayne@69 71
jpayne@69 72 #if !UCONFIG_NO_CONVERSION
jpayne@69 73
jpayne@69 74 U_CDECL_BEGIN
jpayne@69 75
jpayne@69 76 /** Maximum length of a converter name including the terminating NULL @stable ICU 2.0 */
jpayne@69 77 #define UCNV_MAX_CONVERTER_NAME_LENGTH 60
jpayne@69 78 /** Maximum length of a converter name including path and terminating NULL @stable ICU 2.0 */
jpayne@69 79 #define UCNV_MAX_FULL_FILE_NAME_LENGTH (600+UCNV_MAX_CONVERTER_NAME_LENGTH)
jpayne@69 80
jpayne@69 81 /** Shift in for EBDCDIC_STATEFUL and iso2022 states @stable ICU 2.0 */
jpayne@69 82 #define UCNV_SI 0x0F
jpayne@69 83 /** Shift out for EBDCDIC_STATEFUL and iso2022 states @stable ICU 2.0 */
jpayne@69 84 #define UCNV_SO 0x0E
jpayne@69 85
jpayne@69 86 /**
jpayne@69 87 * Enum for specifying basic types of converters
jpayne@69 88 * @see ucnv_getType
jpayne@69 89 * @stable ICU 2.0
jpayne@69 90 */
jpayne@69 91 typedef enum {
jpayne@69 92 /** @stable ICU 2.0 */
jpayne@69 93 UCNV_UNSUPPORTED_CONVERTER = -1,
jpayne@69 94 /** @stable ICU 2.0 */
jpayne@69 95 UCNV_SBCS = 0,
jpayne@69 96 /** @stable ICU 2.0 */
jpayne@69 97 UCNV_DBCS = 1,
jpayne@69 98 /** @stable ICU 2.0 */
jpayne@69 99 UCNV_MBCS = 2,
jpayne@69 100 /** @stable ICU 2.0 */
jpayne@69 101 UCNV_LATIN_1 = 3,
jpayne@69 102 /** @stable ICU 2.0 */
jpayne@69 103 UCNV_UTF8 = 4,
jpayne@69 104 /** @stable ICU 2.0 */
jpayne@69 105 UCNV_UTF16_BigEndian = 5,
jpayne@69 106 /** @stable ICU 2.0 */
jpayne@69 107 UCNV_UTF16_LittleEndian = 6,
jpayne@69 108 /** @stable ICU 2.0 */
jpayne@69 109 UCNV_UTF32_BigEndian = 7,
jpayne@69 110 /** @stable ICU 2.0 */
jpayne@69 111 UCNV_UTF32_LittleEndian = 8,
jpayne@69 112 /** @stable ICU 2.0 */
jpayne@69 113 UCNV_EBCDIC_STATEFUL = 9,
jpayne@69 114 /** @stable ICU 2.0 */
jpayne@69 115 UCNV_ISO_2022 = 10,
jpayne@69 116
jpayne@69 117 /** @stable ICU 2.0 */
jpayne@69 118 UCNV_LMBCS_1 = 11,
jpayne@69 119 /** @stable ICU 2.0 */
jpayne@69 120 UCNV_LMBCS_2,
jpayne@69 121 /** @stable ICU 2.0 */
jpayne@69 122 UCNV_LMBCS_3,
jpayne@69 123 /** @stable ICU 2.0 */
jpayne@69 124 UCNV_LMBCS_4,
jpayne@69 125 /** @stable ICU 2.0 */
jpayne@69 126 UCNV_LMBCS_5,
jpayne@69 127 /** @stable ICU 2.0 */
jpayne@69 128 UCNV_LMBCS_6,
jpayne@69 129 /** @stable ICU 2.0 */
jpayne@69 130 UCNV_LMBCS_8,
jpayne@69 131 /** @stable ICU 2.0 */
jpayne@69 132 UCNV_LMBCS_11,
jpayne@69 133 /** @stable ICU 2.0 */
jpayne@69 134 UCNV_LMBCS_16,
jpayne@69 135 /** @stable ICU 2.0 */
jpayne@69 136 UCNV_LMBCS_17,
jpayne@69 137 /** @stable ICU 2.0 */
jpayne@69 138 UCNV_LMBCS_18,
jpayne@69 139 /** @stable ICU 2.0 */
jpayne@69 140 UCNV_LMBCS_19,
jpayne@69 141 /** @stable ICU 2.0 */
jpayne@69 142 UCNV_LMBCS_LAST = UCNV_LMBCS_19,
jpayne@69 143 /** @stable ICU 2.0 */
jpayne@69 144 UCNV_HZ,
jpayne@69 145 /** @stable ICU 2.0 */
jpayne@69 146 UCNV_SCSU,
jpayne@69 147 /** @stable ICU 2.0 */
jpayne@69 148 UCNV_ISCII,
jpayne@69 149 /** @stable ICU 2.0 */
jpayne@69 150 UCNV_US_ASCII,
jpayne@69 151 /** @stable ICU 2.0 */
jpayne@69 152 UCNV_UTF7,
jpayne@69 153 /** @stable ICU 2.2 */
jpayne@69 154 UCNV_BOCU1,
jpayne@69 155 /** @stable ICU 2.2 */
jpayne@69 156 UCNV_UTF16,
jpayne@69 157 /** @stable ICU 2.2 */
jpayne@69 158 UCNV_UTF32,
jpayne@69 159 /** @stable ICU 2.2 */
jpayne@69 160 UCNV_CESU8,
jpayne@69 161 /** @stable ICU 2.4 */
jpayne@69 162 UCNV_IMAP_MAILBOX,
jpayne@69 163 /** @stable ICU 4.8 */
jpayne@69 164 UCNV_COMPOUND_TEXT,
jpayne@69 165
jpayne@69 166 /* Number of converter types for which we have conversion routines. */
jpayne@69 167 UCNV_NUMBER_OF_SUPPORTED_CONVERTER_TYPES
jpayne@69 168 } UConverterType;
jpayne@69 169
jpayne@69 170 /**
jpayne@69 171 * Enum for specifying which platform a converter ID refers to.
jpayne@69 172 * The use of platform/CCSID is not recommended. See ucnv_openCCSID().
jpayne@69 173 *
jpayne@69 174 * @see ucnv_getPlatform
jpayne@69 175 * @see ucnv_openCCSID
jpayne@69 176 * @see ucnv_getCCSID
jpayne@69 177 * @stable ICU 2.0
jpayne@69 178 */
jpayne@69 179 typedef enum {
jpayne@69 180 UCNV_UNKNOWN = -1,
jpayne@69 181 UCNV_IBM = 0
jpayne@69 182 } UConverterPlatform;
jpayne@69 183
jpayne@69 184 /**
jpayne@69 185 * Function pointer for error callback in the codepage to unicode direction.
jpayne@69 186 * Called when an error has occurred in conversion to unicode, or on open/close of the callback (see reason).
jpayne@69 187 * @param context Pointer to the callback's private data
jpayne@69 188 * @param args Information about the conversion in progress
jpayne@69 189 * @param codeUnits Points to 'length' bytes of the concerned codepage sequence
jpayne@69 190 * @param length Size (in bytes) of the concerned codepage sequence
jpayne@69 191 * @param reason Defines the reason the callback was invoked
jpayne@69 192 * @param pErrorCode ICU error code in/out parameter.
jpayne@69 193 * For converter callback functions, set to a conversion error
jpayne@69 194 * before the call, and the callback may reset it to U_ZERO_ERROR.
jpayne@69 195 * @see ucnv_setToUCallBack
jpayne@69 196 * @see UConverterToUnicodeArgs
jpayne@69 197 * @stable ICU 2.0
jpayne@69 198 */
jpayne@69 199 typedef void (U_EXPORT2 *UConverterToUCallback) (
jpayne@69 200 const void* context,
jpayne@69 201 UConverterToUnicodeArgs *args,
jpayne@69 202 const char *codeUnits,
jpayne@69 203 int32_t length,
jpayne@69 204 UConverterCallbackReason reason,
jpayne@69 205 UErrorCode *pErrorCode);
jpayne@69 206
jpayne@69 207 /**
jpayne@69 208 * Function pointer for error callback in the unicode to codepage direction.
jpayne@69 209 * Called when an error has occurred in conversion from unicode, or on open/close of the callback (see reason).
jpayne@69 210 * @param context Pointer to the callback's private data
jpayne@69 211 * @param args Information about the conversion in progress
jpayne@69 212 * @param codeUnits Points to 'length' UChars of the concerned Unicode sequence
jpayne@69 213 * @param length Size (in bytes) of the concerned codepage sequence
jpayne@69 214 * @param codePoint Single UChar32 (UTF-32) containing the concerend Unicode codepoint.
jpayne@69 215 * @param reason Defines the reason the callback was invoked
jpayne@69 216 * @param pErrorCode ICU error code in/out parameter.
jpayne@69 217 * For converter callback functions, set to a conversion error
jpayne@69 218 * before the call, and the callback may reset it to U_ZERO_ERROR.
jpayne@69 219 * @see ucnv_setFromUCallBack
jpayne@69 220 * @stable ICU 2.0
jpayne@69 221 */
jpayne@69 222 typedef void (U_EXPORT2 *UConverterFromUCallback) (
jpayne@69 223 const void* context,
jpayne@69 224 UConverterFromUnicodeArgs *args,
jpayne@69 225 const UChar* codeUnits,
jpayne@69 226 int32_t length,
jpayne@69 227 UChar32 codePoint,
jpayne@69 228 UConverterCallbackReason reason,
jpayne@69 229 UErrorCode *pErrorCode);
jpayne@69 230
jpayne@69 231 U_CDECL_END
jpayne@69 232
jpayne@69 233 /**
jpayne@69 234 * Character that separates converter names from options and options from each other.
jpayne@69 235 * @see ucnv_open
jpayne@69 236 * @stable ICU 2.0
jpayne@69 237 */
jpayne@69 238 #define UCNV_OPTION_SEP_CHAR ','
jpayne@69 239
jpayne@69 240 /**
jpayne@69 241 * String version of UCNV_OPTION_SEP_CHAR.
jpayne@69 242 * @see ucnv_open
jpayne@69 243 * @stable ICU 2.0
jpayne@69 244 */
jpayne@69 245 #define UCNV_OPTION_SEP_STRING ","
jpayne@69 246
jpayne@69 247 /**
jpayne@69 248 * Character that separates a converter option from its value.
jpayne@69 249 * @see ucnv_open
jpayne@69 250 * @stable ICU 2.0
jpayne@69 251 */
jpayne@69 252 #define UCNV_VALUE_SEP_CHAR '='
jpayne@69 253
jpayne@69 254 /**
jpayne@69 255 * String version of UCNV_VALUE_SEP_CHAR.
jpayne@69 256 * @see ucnv_open
jpayne@69 257 * @stable ICU 2.0
jpayne@69 258 */
jpayne@69 259 #define UCNV_VALUE_SEP_STRING "="
jpayne@69 260
jpayne@69 261 /**
jpayne@69 262 * Converter option for specifying a locale.
jpayne@69 263 * For example, ucnv_open("SCSU,locale=ja", &errorCode);
jpayne@69 264 * See convrtrs.txt.
jpayne@69 265 *
jpayne@69 266 * @see ucnv_open
jpayne@69 267 * @stable ICU 2.0
jpayne@69 268 */
jpayne@69 269 #define UCNV_LOCALE_OPTION_STRING ",locale="
jpayne@69 270
jpayne@69 271 /**
jpayne@69 272 * Converter option for specifying a version selector (0..9) for some converters.
jpayne@69 273 * For example,
jpayne@69 274 * \code
jpayne@69 275 * ucnv_open("UTF-7,version=1", &errorCode);
jpayne@69 276 * \endcode
jpayne@69 277 * See convrtrs.txt.
jpayne@69 278 *
jpayne@69 279 * @see ucnv_open
jpayne@69 280 * @stable ICU 2.4
jpayne@69 281 */
jpayne@69 282 #define UCNV_VERSION_OPTION_STRING ",version="
jpayne@69 283
jpayne@69 284 /**
jpayne@69 285 * Converter option for EBCDIC SBCS or mixed-SBCS/DBCS (stateful) codepages.
jpayne@69 286 * Swaps Unicode mappings for EBCDIC LF and NL codes, as used on
jpayne@69 287 * S/390 (z/OS) Unix System Services (Open Edition).
jpayne@69 288 * For example, ucnv_open("ibm-1047,swaplfnl", &errorCode);
jpayne@69 289 * See convrtrs.txt.
jpayne@69 290 *
jpayne@69 291 * @see ucnv_open
jpayne@69 292 * @stable ICU 2.4
jpayne@69 293 */
jpayne@69 294 #define UCNV_SWAP_LFNL_OPTION_STRING ",swaplfnl"
jpayne@69 295
jpayne@69 296 /**
jpayne@69 297 * Do a fuzzy compare of two converter/alias names.
jpayne@69 298 * The comparison is case-insensitive, ignores leading zeroes if they are not
jpayne@69 299 * followed by further digits, and ignores all but letters and digits.
jpayne@69 300 * Thus the strings "UTF-8", "utf_8", "u*T@f08" and "Utf 8" are exactly equivalent.
jpayne@69 301 * See section 1.4, Charset Alias Matching in Unicode Technical Standard #22
jpayne@69 302 * at http://www.unicode.org/reports/tr22/
jpayne@69 303 *
jpayne@69 304 * @param name1 a converter name or alias, zero-terminated
jpayne@69 305 * @param name2 a converter name or alias, zero-terminated
jpayne@69 306 * @return 0 if the names match, or a negative value if the name1
jpayne@69 307 * lexically precedes name2, or a positive value if the name1
jpayne@69 308 * lexically follows name2.
jpayne@69 309 * @stable ICU 2.0
jpayne@69 310 */
jpayne@69 311 U_STABLE int U_EXPORT2
jpayne@69 312 ucnv_compareNames(const char *name1, const char *name2);
jpayne@69 313
jpayne@69 314
jpayne@69 315 /**
jpayne@69 316 * Creates a UConverter object with the name of a coded character set specified as a C string.
jpayne@69 317 * The actual name will be resolved with the alias file
jpayne@69 318 * using a case-insensitive string comparison that ignores
jpayne@69 319 * leading zeroes and all non-alphanumeric characters.
jpayne@69 320 * E.g., the names "UTF8", "utf-8", "u*T@f08" and "Utf 8" are all equivalent.
jpayne@69 321 * (See also ucnv_compareNames().)
jpayne@69 322 * If <code>NULL</code> is passed for the converter name, it will create one with the
jpayne@69 323 * getDefaultName return value.
jpayne@69 324 *
jpayne@69 325 * <p>A converter name for ICU 1.5 and above may contain options
jpayne@69 326 * like a locale specification to control the specific behavior of
jpayne@69 327 * the newly instantiated converter.
jpayne@69 328 * The meaning of the options depends on the particular converter.
jpayne@69 329 * If an option is not defined for or recognized by a given converter, then it is ignored.</p>
jpayne@69 330 *
jpayne@69 331 * <p>Options are appended to the converter name string, with a
jpayne@69 332 * <code>UCNV_OPTION_SEP_CHAR</code> between the name and the first option and
jpayne@69 333 * also between adjacent options.</p>
jpayne@69 334 *
jpayne@69 335 * <p>If the alias is ambiguous, then the preferred converter is used
jpayne@69 336 * and the status is set to U_AMBIGUOUS_ALIAS_WARNING.</p>
jpayne@69 337 *
jpayne@69 338 * <p>The conversion behavior and names can vary between platforms. ICU may
jpayne@69 339 * convert some characters differently from other platforms. Details on this topic
jpayne@69 340 * are in the <a href="http://icu-project.org/userguide/conversion.html">User's
jpayne@69 341 * Guide</a>. Aliases starting with a "cp" prefix have no specific meaning
jpayne@69 342 * other than its an alias starting with the letters "cp". Please do not
jpayne@69 343 * associate any meaning to these aliases.</p>
jpayne@69 344 *
jpayne@69 345 * \snippet samples/ucnv/convsamp.cpp ucnv_open
jpayne@69 346 *
jpayne@69 347 * @param converterName Name of the coded character set table.
jpayne@69 348 * This may have options appended to the string.
jpayne@69 349 * IANA alias character set names, IBM CCSIDs starting with "ibm-",
jpayne@69 350 * Windows codepage numbers starting with "windows-" are frequently
jpayne@69 351 * used for this parameter. See ucnv_getAvailableName and
jpayne@69 352 * ucnv_getAlias for a complete list that is available.
jpayne@69 353 * If this parameter is NULL, the default converter will be used.
jpayne@69 354 * @param err outgoing error status <TT>U_MEMORY_ALLOCATION_ERROR, U_FILE_ACCESS_ERROR</TT>
jpayne@69 355 * @return the created Unicode converter object, or <TT>NULL</TT> if an error occurred
jpayne@69 356 * @see ucnv_openU
jpayne@69 357 * @see ucnv_openCCSID
jpayne@69 358 * @see ucnv_getAvailableName
jpayne@69 359 * @see ucnv_getAlias
jpayne@69 360 * @see ucnv_getDefaultName
jpayne@69 361 * @see ucnv_close
jpayne@69 362 * @see ucnv_compareNames
jpayne@69 363 * @stable ICU 2.0
jpayne@69 364 */
jpayne@69 365 U_STABLE UConverter* U_EXPORT2
jpayne@69 366 ucnv_open(const char *converterName, UErrorCode *err);
jpayne@69 367
jpayne@69 368
jpayne@69 369 /**
jpayne@69 370 * Creates a Unicode converter with the names specified as unicode string.
jpayne@69 371 * The name should be limited to the ASCII-7 alphanumerics range.
jpayne@69 372 * The actual name will be resolved with the alias file
jpayne@69 373 * using a case-insensitive string comparison that ignores
jpayne@69 374 * leading zeroes and all non-alphanumeric characters.
jpayne@69 375 * E.g., the names "UTF8", "utf-8", "u*T@f08" and "Utf 8" are all equivalent.
jpayne@69 376 * (See also ucnv_compareNames().)
jpayne@69 377 * If <TT>NULL</TT> is passed for the converter name, it will create
jpayne@69 378 * one with the ucnv_getDefaultName() return value.
jpayne@69 379 * If the alias is ambiguous, then the preferred converter is used
jpayne@69 380 * and the status is set to U_AMBIGUOUS_ALIAS_WARNING.
jpayne@69 381 *
jpayne@69 382 * <p>See ucnv_open for the complete details</p>
jpayne@69 383 * @param name Name of the UConverter table in a zero terminated
jpayne@69 384 * Unicode string
jpayne@69 385 * @param err outgoing error status <TT>U_MEMORY_ALLOCATION_ERROR,
jpayne@69 386 * U_FILE_ACCESS_ERROR</TT>
jpayne@69 387 * @return the created Unicode converter object, or <TT>NULL</TT> if an
jpayne@69 388 * error occurred
jpayne@69 389 * @see ucnv_open
jpayne@69 390 * @see ucnv_openCCSID
jpayne@69 391 * @see ucnv_close
jpayne@69 392 * @see ucnv_compareNames
jpayne@69 393 * @stable ICU 2.0
jpayne@69 394 */
jpayne@69 395 U_STABLE UConverter* U_EXPORT2
jpayne@69 396 ucnv_openU(const UChar *name,
jpayne@69 397 UErrorCode *err);
jpayne@69 398
jpayne@69 399 /**
jpayne@69 400 * Creates a UConverter object from a CCSID number and platform pair.
jpayne@69 401 * Note that the usefulness of this function is limited to platforms with numeric
jpayne@69 402 * encoding IDs. Only IBM and Microsoft platforms use numeric (16-bit) identifiers for
jpayne@69 403 * encodings.
jpayne@69 404 *
jpayne@69 405 * In addition, IBM CCSIDs and Unicode conversion tables are not 1:1 related.
jpayne@69 406 * For many IBM CCSIDs there are multiple (up to six) Unicode conversion tables, and
jpayne@69 407 * for some Unicode conversion tables there are multiple CCSIDs.
jpayne@69 408 * Some "alternate" Unicode conversion tables are provided by the
jpayne@69 409 * IBM CDRA conversion table registry.
jpayne@69 410 * The most prominent example of a systematic modification of conversion tables that is
jpayne@69 411 * not provided in the form of conversion table files in the repository is
jpayne@69 412 * that S/390 Unix System Services swaps the codes for Line Feed and New Line in all
jpayne@69 413 * EBCDIC codepages, which requires such a swap in the Unicode conversion tables as well.
jpayne@69 414 *
jpayne@69 415 * Only IBM default conversion tables are accessible with ucnv_openCCSID().
jpayne@69 416 * ucnv_getCCSID() will return the same CCSID for all conversion tables that are associated
jpayne@69 417 * with that CCSID.
jpayne@69 418 *
jpayne@69 419 * Currently, the only "platform" supported in the ICU converter API is UCNV_IBM.
jpayne@69 420 *
jpayne@69 421 * In summary, the use of CCSIDs and the associated API functions is not recommended.
jpayne@69 422 *
jpayne@69 423 * In order to open a converter with the default IBM CDRA Unicode conversion table,
jpayne@69 424 * you can use this function or use the prefix "ibm-":
jpayne@69 425 * \code
jpayne@69 426 * char name[20];
jpayne@69 427 * sprintf(name, "ibm-%hu", ccsid);
jpayne@69 428 * cnv=ucnv_open(name, &errorCode);
jpayne@69 429 * \endcode
jpayne@69 430 *
jpayne@69 431 * In order to open a converter with the IBM S/390 Unix System Services variant
jpayne@69 432 * of a Unicode/EBCDIC conversion table,
jpayne@69 433 * you can use the prefix "ibm-" together with the option string UCNV_SWAP_LFNL_OPTION_STRING:
jpayne@69 434 * \code
jpayne@69 435 * char name[20];
jpayne@69 436 * sprintf(name, "ibm-%hu" UCNV_SWAP_LFNL_OPTION_STRING, ccsid);
jpayne@69 437 * cnv=ucnv_open(name, &errorCode);
jpayne@69 438 * \endcode
jpayne@69 439 *
jpayne@69 440 * In order to open a converter from a Microsoft codepage number, use the prefix "cp":
jpayne@69 441 * \code
jpayne@69 442 * char name[20];
jpayne@69 443 * sprintf(name, "cp%hu", codepageID);
jpayne@69 444 * cnv=ucnv_open(name, &errorCode);
jpayne@69 445 * \endcode
jpayne@69 446 *
jpayne@69 447 * If the alias is ambiguous, then the preferred converter is used
jpayne@69 448 * and the status is set to U_AMBIGUOUS_ALIAS_WARNING.
jpayne@69 449 *
jpayne@69 450 * @param codepage codepage number to create
jpayne@69 451 * @param platform the platform in which the codepage number exists
jpayne@69 452 * @param err error status <TT>U_MEMORY_ALLOCATION_ERROR, U_FILE_ACCESS_ERROR</TT>
jpayne@69 453 * @return the created Unicode converter object, or <TT>NULL</TT> if an error
jpayne@69 454 * occurred.
jpayne@69 455 * @see ucnv_open
jpayne@69 456 * @see ucnv_openU
jpayne@69 457 * @see ucnv_close
jpayne@69 458 * @see ucnv_getCCSID
jpayne@69 459 * @see ucnv_getPlatform
jpayne@69 460 * @see UConverterPlatform
jpayne@69 461 * @stable ICU 2.0
jpayne@69 462 */
jpayne@69 463 U_STABLE UConverter* U_EXPORT2
jpayne@69 464 ucnv_openCCSID(int32_t codepage,
jpayne@69 465 UConverterPlatform platform,
jpayne@69 466 UErrorCode * err);
jpayne@69 467
jpayne@69 468 /**
jpayne@69 469 * <p>Creates a UConverter object specified from a packageName and a converterName.</p>
jpayne@69 470 *
jpayne@69 471 * <p>The packageName and converterName must point to an ICU udata object, as defined by
jpayne@69 472 * <code> udata_open( packageName, "cnv", converterName, err) </code> or equivalent.
jpayne@69 473 * Typically, packageName will refer to a (.dat) file, or to a package registered with
jpayne@69 474 * udata_setAppData(). Using a full file or directory pathname for packageName is deprecated.</p>
jpayne@69 475 *
jpayne@69 476 * <p>The name will NOT be looked up in the alias mechanism, nor will the converter be
jpayne@69 477 * stored in the converter cache or the alias table. The only way to open further converters
jpayne@69 478 * is call this function multiple times, or use the ucnv_safeClone() function to clone a
jpayne@69 479 * 'master' converter.</p>
jpayne@69 480 *
jpayne@69 481 * <p>A future version of ICU may add alias table lookups and/or caching
jpayne@69 482 * to this function.</p>
jpayne@69 483 *
jpayne@69 484 * <p>Example Use:
jpayne@69 485 * <code>cnv = ucnv_openPackage("myapp", "myconverter", &err);</code>
jpayne@69 486 * </p>
jpayne@69 487 *
jpayne@69 488 * @param packageName name of the package (equivalent to 'path' in udata_open() call)
jpayne@69 489 * @param converterName name of the data item to be used, without suffix.
jpayne@69 490 * @param err outgoing error status <TT>U_MEMORY_ALLOCATION_ERROR, U_FILE_ACCESS_ERROR</TT>
jpayne@69 491 * @return the created Unicode converter object, or <TT>NULL</TT> if an error occurred
jpayne@69 492 * @see udata_open
jpayne@69 493 * @see ucnv_open
jpayne@69 494 * @see ucnv_safeClone
jpayne@69 495 * @see ucnv_close
jpayne@69 496 * @stable ICU 2.2
jpayne@69 497 */
jpayne@69 498 U_STABLE UConverter* U_EXPORT2
jpayne@69 499 ucnv_openPackage(const char *packageName, const char *converterName, UErrorCode *err);
jpayne@69 500
jpayne@69 501 /**
jpayne@69 502 * Thread safe converter cloning operation.
jpayne@69 503 * For most efficient operation, pass in a stackBuffer (and a *pBufferSize)
jpayne@69 504 * with at least U_CNV_SAFECLONE_BUFFERSIZE bytes of space.
jpayne@69 505 * If the buffer size is sufficient, then the clone will use the stack buffer;
jpayne@69 506 * otherwise, it will be allocated, and *pBufferSize will indicate
jpayne@69 507 * the actual size. (This should not occur with U_CNV_SAFECLONE_BUFFERSIZE.)
jpayne@69 508 *
jpayne@69 509 * You must ucnv_close() the clone in any case.
jpayne@69 510 *
jpayne@69 511 * If *pBufferSize==0, (regardless of whether stackBuffer==NULL or not)
jpayne@69 512 * then *pBufferSize will be changed to a sufficient size
jpayne@69 513 * for cloning this converter,
jpayne@69 514 * without actually cloning the converter ("pure pre-flighting").
jpayne@69 515 *
jpayne@69 516 * If *pBufferSize is greater than zero but not large enough for a stack-based
jpayne@69 517 * clone, then the converter is cloned using newly allocated memory
jpayne@69 518 * and *pBufferSize is changed to the necessary size.
jpayne@69 519 *
jpayne@69 520 * If the converter clone fits into the stack buffer but the stack buffer is not
jpayne@69 521 * sufficiently aligned for the clone, then the clone will use an
jpayne@69 522 * adjusted pointer and use an accordingly smaller buffer size.
jpayne@69 523 *
jpayne@69 524 * @param cnv converter to be cloned
jpayne@69 525 * @param stackBuffer <em>Deprecated functionality as of ICU 52, use NULL.</em><br>
jpayne@69 526 * user allocated space for the new clone. If NULL new memory will be allocated.
jpayne@69 527 * If buffer is not large enough, new memory will be allocated.
jpayne@69 528 * Clients can use the U_CNV_SAFECLONE_BUFFERSIZE. This will probably be enough to avoid memory allocations.
jpayne@69 529 * @param pBufferSize <em>Deprecated functionality as of ICU 52, use NULL or 1.</em><br>
jpayne@69 530 * pointer to size of allocated space.
jpayne@69 531 * @param status to indicate whether the operation went on smoothly or there were errors
jpayne@69 532 * An informational status value, U_SAFECLONE_ALLOCATED_WARNING,
jpayne@69 533 * is used if any allocations were necessary.
jpayne@69 534 * However, it is better to check if *pBufferSize grew for checking for
jpayne@69 535 * allocations because warning codes can be overridden by subsequent
jpayne@69 536 * function calls.
jpayne@69 537 * @return pointer to the new clone
jpayne@69 538 * @stable ICU 2.0
jpayne@69 539 */
jpayne@69 540 U_STABLE UConverter * U_EXPORT2
jpayne@69 541 ucnv_safeClone(const UConverter *cnv,
jpayne@69 542 void *stackBuffer,
jpayne@69 543 int32_t *pBufferSize,
jpayne@69 544 UErrorCode *status);
jpayne@69 545
jpayne@69 546 #ifndef U_HIDE_DEPRECATED_API
jpayne@69 547
jpayne@69 548 /**
jpayne@69 549 * \def U_CNV_SAFECLONE_BUFFERSIZE
jpayne@69 550 * Definition of a buffer size that is designed to be large enough for
jpayne@69 551 * converters to be cloned with ucnv_safeClone().
jpayne@69 552 * @deprecated ICU 52. Do not rely on ucnv_safeClone() cloning into any provided buffer.
jpayne@69 553 */
jpayne@69 554 #define U_CNV_SAFECLONE_BUFFERSIZE 1024
jpayne@69 555
jpayne@69 556 #endif /* U_HIDE_DEPRECATED_API */
jpayne@69 557
jpayne@69 558 /**
jpayne@69 559 * Deletes the unicode converter and releases resources associated
jpayne@69 560 * with just this instance.
jpayne@69 561 * Does not free up shared converter tables.
jpayne@69 562 *
jpayne@69 563 * @param converter the converter object to be deleted
jpayne@69 564 * @see ucnv_open
jpayne@69 565 * @see ucnv_openU
jpayne@69 566 * @see ucnv_openCCSID
jpayne@69 567 * @stable ICU 2.0
jpayne@69 568 */
jpayne@69 569 U_STABLE void U_EXPORT2
jpayne@69 570 ucnv_close(UConverter * converter);
jpayne@69 571
jpayne@69 572 #if U_SHOW_CPLUSPLUS_API
jpayne@69 573
jpayne@69 574 U_NAMESPACE_BEGIN
jpayne@69 575
jpayne@69 576 /**
jpayne@69 577 * \class LocalUConverterPointer
jpayne@69 578 * "Smart pointer" class, closes a UConverter via ucnv_close().
jpayne@69 579 * For most methods see the LocalPointerBase base class.
jpayne@69 580 *
jpayne@69 581 * @see LocalPointerBase
jpayne@69 582 * @see LocalPointer
jpayne@69 583 * @stable ICU 4.4
jpayne@69 584 */
jpayne@69 585 U_DEFINE_LOCAL_OPEN_POINTER(LocalUConverterPointer, UConverter, ucnv_close);
jpayne@69 586
jpayne@69 587 U_NAMESPACE_END
jpayne@69 588
jpayne@69 589 #endif
jpayne@69 590
jpayne@69 591 /**
jpayne@69 592 * Fills in the output parameter, subChars, with the substitution characters
jpayne@69 593 * as multiple bytes.
jpayne@69 594 * If ucnv_setSubstString() set a Unicode string because the converter is
jpayne@69 595 * stateful, then subChars will be an empty string.
jpayne@69 596 *
jpayne@69 597 * @param converter the Unicode converter
jpayne@69 598 * @param subChars the substitution characters
jpayne@69 599 * @param len on input the capacity of subChars, on output the number
jpayne@69 600 * of bytes copied to it
jpayne@69 601 * @param err the outgoing error status code.
jpayne@69 602 * If the substitution character array is too small, an
jpayne@69 603 * <TT>U_INDEX_OUTOFBOUNDS_ERROR</TT> will be returned.
jpayne@69 604 * @see ucnv_setSubstString
jpayne@69 605 * @see ucnv_setSubstChars
jpayne@69 606 * @stable ICU 2.0
jpayne@69 607 */
jpayne@69 608 U_STABLE void U_EXPORT2
jpayne@69 609 ucnv_getSubstChars(const UConverter *converter,
jpayne@69 610 char *subChars,
jpayne@69 611 int8_t *len,
jpayne@69 612 UErrorCode *err);
jpayne@69 613
jpayne@69 614 /**
jpayne@69 615 * Sets the substitution chars when converting from unicode to a codepage. The
jpayne@69 616 * substitution is specified as a string of 1-4 bytes, and may contain
jpayne@69 617 * <TT>NULL</TT> bytes.
jpayne@69 618 * The subChars must represent a single character. The caller needs to know the
jpayne@69 619 * byte sequence of a valid character in the converter's charset.
jpayne@69 620 * For some converters, for example some ISO 2022 variants, only single-byte
jpayne@69 621 * substitution characters may be supported.
jpayne@69 622 * The newer ucnv_setSubstString() function relaxes these limitations.
jpayne@69 623 *
jpayne@69 624 * @param converter the Unicode converter
jpayne@69 625 * @param subChars the substitution character byte sequence we want set
jpayne@69 626 * @param len the number of bytes in subChars
jpayne@69 627 * @param err the error status code. <TT>U_INDEX_OUTOFBOUNDS_ERROR </TT> if
jpayne@69 628 * len is bigger than the maximum number of bytes allowed in subchars
jpayne@69 629 * @see ucnv_setSubstString
jpayne@69 630 * @see ucnv_getSubstChars
jpayne@69 631 * @stable ICU 2.0
jpayne@69 632 */
jpayne@69 633 U_STABLE void U_EXPORT2
jpayne@69 634 ucnv_setSubstChars(UConverter *converter,
jpayne@69 635 const char *subChars,
jpayne@69 636 int8_t len,
jpayne@69 637 UErrorCode *err);
jpayne@69 638
jpayne@69 639 /**
jpayne@69 640 * Set a substitution string for converting from Unicode to a charset.
jpayne@69 641 * The caller need not know the charset byte sequence for each charset.
jpayne@69 642 *
jpayne@69 643 * Unlike ucnv_setSubstChars() which is designed to set a charset byte sequence
jpayne@69 644 * for a single character, this function takes a Unicode string with
jpayne@69 645 * zero, one or more characters, and immediately verifies that the string can be
jpayne@69 646 * converted to the charset.
jpayne@69 647 * If not, or if the result is too long (more than 32 bytes as of ICU 3.6),
jpayne@69 648 * then the function returns with an error accordingly.
jpayne@69 649 *
jpayne@69 650 * Also unlike ucnv_setSubstChars(), this function works for stateful charsets
jpayne@69 651 * by converting on the fly at the point of substitution rather than setting
jpayne@69 652 * a fixed byte sequence.
jpayne@69 653 *
jpayne@69 654 * @param cnv The UConverter object.
jpayne@69 655 * @param s The Unicode string.
jpayne@69 656 * @param length The number of UChars in s, or -1 for a NUL-terminated string.
jpayne@69 657 * @param err Pointer to a standard ICU error code. Its input value must
jpayne@69 658 * pass the U_SUCCESS() test, or else the function returns
jpayne@69 659 * immediately. Check for U_FAILURE() on output or use with
jpayne@69 660 * function chaining. (See User Guide for details.)
jpayne@69 661 *
jpayne@69 662 * @see ucnv_setSubstChars
jpayne@69 663 * @see ucnv_getSubstChars
jpayne@69 664 * @stable ICU 3.6
jpayne@69 665 */
jpayne@69 666 U_STABLE void U_EXPORT2
jpayne@69 667 ucnv_setSubstString(UConverter *cnv,
jpayne@69 668 const UChar *s,
jpayne@69 669 int32_t length,
jpayne@69 670 UErrorCode *err);
jpayne@69 671
jpayne@69 672 /**
jpayne@69 673 * Fills in the output parameter, errBytes, with the error characters from the
jpayne@69 674 * last failing conversion.
jpayne@69 675 *
jpayne@69 676 * @param converter the Unicode converter
jpayne@69 677 * @param errBytes the codepage bytes which were in error
jpayne@69 678 * @param len on input the capacity of errBytes, on output the number of
jpayne@69 679 * bytes which were copied to it
jpayne@69 680 * @param err the error status code.
jpayne@69 681 * If the substitution character array is too small, an
jpayne@69 682 * <TT>U_INDEX_OUTOFBOUNDS_ERROR</TT> will be returned.
jpayne@69 683 * @stable ICU 2.0
jpayne@69 684 */
jpayne@69 685 U_STABLE void U_EXPORT2
jpayne@69 686 ucnv_getInvalidChars(const UConverter *converter,
jpayne@69 687 char *errBytes,
jpayne@69 688 int8_t *len,
jpayne@69 689 UErrorCode *err);
jpayne@69 690
jpayne@69 691 /**
jpayne@69 692 * Fills in the output parameter, errChars, with the error characters from the
jpayne@69 693 * last failing conversion.
jpayne@69 694 *
jpayne@69 695 * @param converter the Unicode converter
jpayne@69 696 * @param errUChars the UChars which were in error
jpayne@69 697 * @param len on input the capacity of errUChars, on output the number of
jpayne@69 698 * UChars which were copied to it
jpayne@69 699 * @param err the error status code.
jpayne@69 700 * If the substitution character array is too small, an
jpayne@69 701 * <TT>U_INDEX_OUTOFBOUNDS_ERROR</TT> will be returned.
jpayne@69 702 * @stable ICU 2.0
jpayne@69 703 */
jpayne@69 704 U_STABLE void U_EXPORT2
jpayne@69 705 ucnv_getInvalidUChars(const UConverter *converter,
jpayne@69 706 UChar *errUChars,
jpayne@69 707 int8_t *len,
jpayne@69 708 UErrorCode *err);
jpayne@69 709
jpayne@69 710 /**
jpayne@69 711 * Resets the state of a converter to the default state. This is used
jpayne@69 712 * in the case of an error, to restart a conversion from a known default state.
jpayne@69 713 * It will also empty the internal output buffers.
jpayne@69 714 * @param converter the Unicode converter
jpayne@69 715 * @stable ICU 2.0
jpayne@69 716 */
jpayne@69 717 U_STABLE void U_EXPORT2
jpayne@69 718 ucnv_reset(UConverter *converter);
jpayne@69 719
jpayne@69 720 /**
jpayne@69 721 * Resets the to-Unicode part of a converter state to the default state.
jpayne@69 722 * This is used in the case of an error to restart a conversion to
jpayne@69 723 * Unicode to a known default state. It will also empty the internal
jpayne@69 724 * output buffers used for the conversion to Unicode codepoints.
jpayne@69 725 * @param converter the Unicode converter
jpayne@69 726 * @stable ICU 2.0
jpayne@69 727 */
jpayne@69 728 U_STABLE void U_EXPORT2
jpayne@69 729 ucnv_resetToUnicode(UConverter *converter);
jpayne@69 730
jpayne@69 731 /**
jpayne@69 732 * Resets the from-Unicode part of a converter state to the default state.
jpayne@69 733 * This is used in the case of an error to restart a conversion from
jpayne@69 734 * Unicode to a known default state. It will also empty the internal output
jpayne@69 735 * buffers used for the conversion from Unicode codepoints.
jpayne@69 736 * @param converter the Unicode converter
jpayne@69 737 * @stable ICU 2.0
jpayne@69 738 */
jpayne@69 739 U_STABLE void U_EXPORT2
jpayne@69 740 ucnv_resetFromUnicode(UConverter *converter);
jpayne@69 741
jpayne@69 742 /**
jpayne@69 743 * Returns the maximum number of bytes that are output per UChar in conversion
jpayne@69 744 * from Unicode using this converter.
jpayne@69 745 * The returned number can be used with UCNV_GET_MAX_BYTES_FOR_STRING
jpayne@69 746 * to calculate the size of a target buffer for conversion from Unicode.
jpayne@69 747 *
jpayne@69 748 * Note: Before ICU 2.8, this function did not return reliable numbers for
jpayne@69 749 * some stateful converters (EBCDIC_STATEFUL, ISO-2022) and LMBCS.
jpayne@69 750 *
jpayne@69 751 * This number may not be the same as the maximum number of bytes per
jpayne@69 752 * "conversion unit". In other words, it may not be the intuitively expected
jpayne@69 753 * number of bytes per character that would be published for a charset,
jpayne@69 754 * and may not fulfill any other purpose than the allocation of an output
jpayne@69 755 * buffer of guaranteed sufficient size for a given input length and converter.
jpayne@69 756 *
jpayne@69 757 * Examples for special cases that are taken into account:
jpayne@69 758 * - Supplementary code points may convert to more bytes than BMP code points.
jpayne@69 759 * This function returns bytes per UChar (UTF-16 code unit), not per
jpayne@69 760 * Unicode code point, for efficient buffer allocation.
jpayne@69 761 * - State-shifting output (SI/SO, escapes, etc.) from stateful converters.
jpayne@69 762 * - When m input UChars are converted to n output bytes, then the maximum m/n
jpayne@69 763 * is taken into account.
jpayne@69 764 *
jpayne@69 765 * The number returned here does not take into account
jpayne@69 766 * (see UCNV_GET_MAX_BYTES_FOR_STRING):
jpayne@69 767 * - callbacks which output more than one charset character sequence per call,
jpayne@69 768 * like escape callbacks
jpayne@69 769 * - initial and final non-character bytes that are output by some converters
jpayne@69 770 * (automatic BOMs, initial escape sequence, final SI, etc.)
jpayne@69 771 *
jpayne@69 772 * Examples for returned values:
jpayne@69 773 * - SBCS charsets: 1
jpayne@69 774 * - Shift-JIS: 2
jpayne@69 775 * - UTF-16: 2 (2 per BMP, 4 per surrogate _pair_, BOM not counted)
jpayne@69 776 * - UTF-8: 3 (3 per BMP, 4 per surrogate _pair_)
jpayne@69 777 * - EBCDIC_STATEFUL (EBCDIC mixed SBCS/DBCS): 3 (SO + DBCS)
jpayne@69 778 * - ISO-2022: 3 (always outputs UTF-8)
jpayne@69 779 * - ISO-2022-JP: 6 (4-byte escape sequences + DBCS)
jpayne@69 780 * - ISO-2022-CN: 8 (4-byte designator sequences + 2-byte SS2/SS3 + DBCS)
jpayne@69 781 *
jpayne@69 782 * @param converter The Unicode converter.
jpayne@69 783 * @return The maximum number of bytes per UChar (16 bit code unit)
jpayne@69 784 * that are output by ucnv_fromUnicode(),
jpayne@69 785 * to be used together with UCNV_GET_MAX_BYTES_FOR_STRING
jpayne@69 786 * for buffer allocation.
jpayne@69 787 *
jpayne@69 788 * @see UCNV_GET_MAX_BYTES_FOR_STRING
jpayne@69 789 * @see ucnv_getMinCharSize
jpayne@69 790 * @stable ICU 2.0
jpayne@69 791 */
jpayne@69 792 U_STABLE int8_t U_EXPORT2
jpayne@69 793 ucnv_getMaxCharSize(const UConverter *converter);
jpayne@69 794
jpayne@69 795 /**
jpayne@69 796 * Calculates the size of a buffer for conversion from Unicode to a charset.
jpayne@69 797 * The calculated size is guaranteed to be sufficient for this conversion.
jpayne@69 798 *
jpayne@69 799 * It takes into account initial and final non-character bytes that are output
jpayne@69 800 * by some converters.
jpayne@69 801 * It does not take into account callbacks which output more than one charset
jpayne@69 802 * character sequence per call, like escape callbacks.
jpayne@69 803 * The default (substitution) callback only outputs one charset character sequence.
jpayne@69 804 *
jpayne@69 805 * @param length Number of UChars to be converted.
jpayne@69 806 * @param maxCharSize Return value from ucnv_getMaxCharSize() for the converter
jpayne@69 807 * that will be used.
jpayne@69 808 * @return Size of a buffer that will be large enough to hold the output bytes of
jpayne@69 809 * converting length UChars with the converter that returned the maxCharSize.
jpayne@69 810 *
jpayne@69 811 * @see ucnv_getMaxCharSize
jpayne@69 812 * @stable ICU 2.8
jpayne@69 813 */
jpayne@69 814 #define UCNV_GET_MAX_BYTES_FOR_STRING(length, maxCharSize) \
jpayne@69 815 (((int32_t)(length)+10)*(int32_t)(maxCharSize))
jpayne@69 816
jpayne@69 817 /**
jpayne@69 818 * Returns the minimum byte length (per codepoint) for characters in this codepage.
jpayne@69 819 * This is usually either 1 or 2.
jpayne@69 820 * @param converter the Unicode converter
jpayne@69 821 * @return the minimum number of bytes per codepoint allowed by this particular converter
jpayne@69 822 * @see ucnv_getMaxCharSize
jpayne@69 823 * @stable ICU 2.0
jpayne@69 824 */
jpayne@69 825 U_STABLE int8_t U_EXPORT2
jpayne@69 826 ucnv_getMinCharSize(const UConverter *converter);
jpayne@69 827
jpayne@69 828 /**
jpayne@69 829 * Returns the display name of the converter passed in based on the Locale
jpayne@69 830 * passed in. If the locale contains no display name, the internal ASCII
jpayne@69 831 * name will be filled in.
jpayne@69 832 *
jpayne@69 833 * @param converter the Unicode converter.
jpayne@69 834 * @param displayLocale is the specific Locale we want to localized for
jpayne@69 835 * @param displayName user provided buffer to be filled in
jpayne@69 836 * @param displayNameCapacity size of displayName Buffer
jpayne@69 837 * @param err error status code
jpayne@69 838 * @return displayNameLength number of UChar needed in displayName
jpayne@69 839 * @see ucnv_getName
jpayne@69 840 * @stable ICU 2.0
jpayne@69 841 */
jpayne@69 842 U_STABLE int32_t U_EXPORT2
jpayne@69 843 ucnv_getDisplayName(const UConverter *converter,
jpayne@69 844 const char *displayLocale,
jpayne@69 845 UChar *displayName,
jpayne@69 846 int32_t displayNameCapacity,
jpayne@69 847 UErrorCode *err);
jpayne@69 848
jpayne@69 849 /**
jpayne@69 850 * Gets the internal, canonical name of the converter (zero-terminated).
jpayne@69 851 * The lifetime of the returned string will be that of the converter
jpayne@69 852 * passed to this function.
jpayne@69 853 * @param converter the Unicode converter
jpayne@69 854 * @param err UErrorCode status
jpayne@69 855 * @return the internal name of the converter
jpayne@69 856 * @see ucnv_getDisplayName
jpayne@69 857 * @stable ICU 2.0
jpayne@69 858 */
jpayne@69 859 U_STABLE const char * U_EXPORT2
jpayne@69 860 ucnv_getName(const UConverter *converter, UErrorCode *err);
jpayne@69 861
jpayne@69 862 /**
jpayne@69 863 * Gets a codepage number associated with the converter. This is not guaranteed
jpayne@69 864 * to be the one used to create the converter. Some converters do not represent
jpayne@69 865 * platform registered codepages and return zero for the codepage number.
jpayne@69 866 * The error code fill-in parameter indicates if the codepage number
jpayne@69 867 * is available.
jpayne@69 868 * Does not check if the converter is <TT>NULL</TT> or if converter's data
jpayne@69 869 * table is <TT>NULL</TT>.
jpayne@69 870 *
jpayne@69 871 * Important: The use of CCSIDs is not recommended because it is limited
jpayne@69 872 * to only two platforms in principle and only one (UCNV_IBM) in the current
jpayne@69 873 * ICU converter API.
jpayne@69 874 * Also, CCSIDs are insufficient to identify IBM Unicode conversion tables precisely.
jpayne@69 875 * For more details see ucnv_openCCSID().
jpayne@69 876 *
jpayne@69 877 * @param converter the Unicode converter
jpayne@69 878 * @param err the error status code.
jpayne@69 879 * @return If any error occurs, -1 will be returned otherwise, the codepage number
jpayne@69 880 * will be returned
jpayne@69 881 * @see ucnv_openCCSID
jpayne@69 882 * @see ucnv_getPlatform
jpayne@69 883 * @stable ICU 2.0
jpayne@69 884 */
jpayne@69 885 U_STABLE int32_t U_EXPORT2
jpayne@69 886 ucnv_getCCSID(const UConverter *converter,
jpayne@69 887 UErrorCode *err);
jpayne@69 888
jpayne@69 889 /**
jpayne@69 890 * Gets a codepage platform associated with the converter. Currently,
jpayne@69 891 * only <TT>UCNV_IBM</TT> will be returned.
jpayne@69 892 * Does not test if the converter is <TT>NULL</TT> or if converter's data
jpayne@69 893 * table is <TT>NULL</TT>.
jpayne@69 894 * @param converter the Unicode converter
jpayne@69 895 * @param err the error status code.
jpayne@69 896 * @return The codepage platform
jpayne@69 897 * @stable ICU 2.0
jpayne@69 898 */
jpayne@69 899 U_STABLE UConverterPlatform U_EXPORT2
jpayne@69 900 ucnv_getPlatform(const UConverter *converter,
jpayne@69 901 UErrorCode *err);
jpayne@69 902
jpayne@69 903 /**
jpayne@69 904 * Gets the type of the converter
jpayne@69 905 * e.g. SBCS, MBCS, DBCS, UTF8, UTF16_BE, UTF16_LE, ISO_2022,
jpayne@69 906 * EBCDIC_STATEFUL, LATIN_1
jpayne@69 907 * @param converter a valid, opened converter
jpayne@69 908 * @return the type of the converter
jpayne@69 909 * @stable ICU 2.0
jpayne@69 910 */
jpayne@69 911 U_STABLE UConverterType U_EXPORT2
jpayne@69 912 ucnv_getType(const UConverter * converter);
jpayne@69 913
jpayne@69 914 /**
jpayne@69 915 * Gets the "starter" (lead) bytes for converters of type MBCS.
jpayne@69 916 * Will fill in an <TT>U_ILLEGAL_ARGUMENT_ERROR</TT> if converter passed in
jpayne@69 917 * is not MBCS. Fills in an array of type UBool, with the value of the byte
jpayne@69 918 * as offset to the array. For example, if (starters[0x20] == TRUE) at return,
jpayne@69 919 * it means that the byte 0x20 is a starter byte in this converter.
jpayne@69 920 * Context pointers are always owned by the caller.
jpayne@69 921 *
jpayne@69 922 * @param converter a valid, opened converter of type MBCS
jpayne@69 923 * @param starters an array of size 256 to be filled in
jpayne@69 924 * @param err error status, <TT>U_ILLEGAL_ARGUMENT_ERROR</TT> if the
jpayne@69 925 * converter is not a type which can return starters.
jpayne@69 926 * @see ucnv_getType
jpayne@69 927 * @stable ICU 2.0
jpayne@69 928 */
jpayne@69 929 U_STABLE void U_EXPORT2
jpayne@69 930 ucnv_getStarters(const UConverter* converter,
jpayne@69 931 UBool starters[256],
jpayne@69 932 UErrorCode* err);
jpayne@69 933
jpayne@69 934
jpayne@69 935 /**
jpayne@69 936 * Selectors for Unicode sets that can be returned by ucnv_getUnicodeSet().
jpayne@69 937 * @see ucnv_getUnicodeSet
jpayne@69 938 * @stable ICU 2.6
jpayne@69 939 */
jpayne@69 940 typedef enum UConverterUnicodeSet {
jpayne@69 941 /** Select the set of roundtrippable Unicode code points. @stable ICU 2.6 */
jpayne@69 942 UCNV_ROUNDTRIP_SET,
jpayne@69 943 /** Select the set of Unicode code points with roundtrip or fallback mappings. @stable ICU 4.0 */
jpayne@69 944 UCNV_ROUNDTRIP_AND_FALLBACK_SET,
jpayne@69 945 #ifndef U_HIDE_DEPRECATED_API
jpayne@69 946 /**
jpayne@69 947 * Number of UConverterUnicodeSet selectors.
jpayne@69 948 * @deprecated ICU 58 The numeric value may change over time, see ICU ticket #12420.
jpayne@69 949 */
jpayne@69 950 UCNV_SET_COUNT
jpayne@69 951 #endif // U_HIDE_DEPRECATED_API
jpayne@69 952 } UConverterUnicodeSet;
jpayne@69 953
jpayne@69 954
jpayne@69 955 /**
jpayne@69 956 * Returns the set of Unicode code points that can be converted by an ICU converter.
jpayne@69 957 *
jpayne@69 958 * Returns one of several kinds of set:
jpayne@69 959 *
jpayne@69 960 * 1. UCNV_ROUNDTRIP_SET
jpayne@69 961 *
jpayne@69 962 * The set of all Unicode code points that can be roundtrip-converted
jpayne@69 963 * (converted without any data loss) with the converter (ucnv_fromUnicode()).
jpayne@69 964 * This set will not include code points that have fallback mappings
jpayne@69 965 * or are only the result of reverse fallback mappings.
jpayne@69 966 * This set will also not include PUA code points with fallbacks, although
jpayne@69 967 * ucnv_fromUnicode() will always uses those mappings despite ucnv_setFallback().
jpayne@69 968 * See UTR #22 "Character Mapping Markup Language"
jpayne@69 969 * at http://www.unicode.org/reports/tr22/
jpayne@69 970 *
jpayne@69 971 * This is useful for example for
jpayne@69 972 * - checking that a string or document can be roundtrip-converted with a converter,
jpayne@69 973 * without/before actually performing the conversion
jpayne@69 974 * - testing if a converter can be used for text for typical text for a certain locale,
jpayne@69 975 * by comparing its roundtrip set with the set of ExemplarCharacters from
jpayne@69 976 * ICU's locale data or other sources
jpayne@69 977 *
jpayne@69 978 * 2. UCNV_ROUNDTRIP_AND_FALLBACK_SET
jpayne@69 979 *
jpayne@69 980 * The set of all Unicode code points that can be converted with the converter (ucnv_fromUnicode())
jpayne@69 981 * when fallbacks are turned on (see ucnv_setFallback()).
jpayne@69 982 * This set includes all code points with roundtrips and fallbacks (but not reverse fallbacks).
jpayne@69 983 *
jpayne@69 984 * In the future, there may be more UConverterUnicodeSet choices to select
jpayne@69 985 * sets with different properties.
jpayne@69 986 *
jpayne@69 987 * @param cnv The converter for which a set is requested.
jpayne@69 988 * @param setFillIn A valid USet *. It will be cleared by this function before
jpayne@69 989 * the converter's specific set is filled into the USet.
jpayne@69 990 * @param whichSet A UConverterUnicodeSet selector;
jpayne@69 991 * currently UCNV_ROUNDTRIP_SET is the only supported value.
jpayne@69 992 * @param pErrorCode ICU error code in/out parameter.
jpayne@69 993 * Must fulfill U_SUCCESS before the function call.
jpayne@69 994 *
jpayne@69 995 * @see UConverterUnicodeSet
jpayne@69 996 * @see uset_open
jpayne@69 997 * @see uset_close
jpayne@69 998 * @stable ICU 2.6
jpayne@69 999 */
jpayne@69 1000 U_STABLE void U_EXPORT2
jpayne@69 1001 ucnv_getUnicodeSet(const UConverter *cnv,
jpayne@69 1002 USet *setFillIn,
jpayne@69 1003 UConverterUnicodeSet whichSet,
jpayne@69 1004 UErrorCode *pErrorCode);
jpayne@69 1005
jpayne@69 1006 /**
jpayne@69 1007 * Gets the current calback function used by the converter when an illegal
jpayne@69 1008 * or invalid codepage sequence is found.
jpayne@69 1009 * Context pointers are always owned by the caller.
jpayne@69 1010 *
jpayne@69 1011 * @param converter the unicode converter
jpayne@69 1012 * @param action fillin: returns the callback function pointer
jpayne@69 1013 * @param context fillin: returns the callback's private void* context
jpayne@69 1014 * @see ucnv_setToUCallBack
jpayne@69 1015 * @stable ICU 2.0
jpayne@69 1016 */
jpayne@69 1017 U_STABLE void U_EXPORT2
jpayne@69 1018 ucnv_getToUCallBack (const UConverter * converter,
jpayne@69 1019 UConverterToUCallback *action,
jpayne@69 1020 const void **context);
jpayne@69 1021
jpayne@69 1022 /**
jpayne@69 1023 * Gets the current callback function used by the converter when illegal
jpayne@69 1024 * or invalid Unicode sequence is found.
jpayne@69 1025 * Context pointers are always owned by the caller.
jpayne@69 1026 *
jpayne@69 1027 * @param converter the unicode converter
jpayne@69 1028 * @param action fillin: returns the callback function pointer
jpayne@69 1029 * @param context fillin: returns the callback's private void* context
jpayne@69 1030 * @see ucnv_setFromUCallBack
jpayne@69 1031 * @stable ICU 2.0
jpayne@69 1032 */
jpayne@69 1033 U_STABLE void U_EXPORT2
jpayne@69 1034 ucnv_getFromUCallBack (const UConverter * converter,
jpayne@69 1035 UConverterFromUCallback *action,
jpayne@69 1036 const void **context);
jpayne@69 1037
jpayne@69 1038 /**
jpayne@69 1039 * Changes the callback function used by the converter when
jpayne@69 1040 * an illegal or invalid sequence is found.
jpayne@69 1041 * Context pointers are always owned by the caller.
jpayne@69 1042 * Predefined actions and contexts can be found in the ucnv_err.h header.
jpayne@69 1043 *
jpayne@69 1044 * @param converter the unicode converter
jpayne@69 1045 * @param newAction the new callback function
jpayne@69 1046 * @param newContext the new toUnicode callback context pointer. This can be NULL.
jpayne@69 1047 * @param oldAction fillin: returns the old callback function pointer. This can be NULL.
jpayne@69 1048 * @param oldContext fillin: returns the old callback's private void* context. This can be NULL.
jpayne@69 1049 * @param err The error code status
jpayne@69 1050 * @see ucnv_getToUCallBack
jpayne@69 1051 * @stable ICU 2.0
jpayne@69 1052 */
jpayne@69 1053 U_STABLE void U_EXPORT2
jpayne@69 1054 ucnv_setToUCallBack (UConverter * converter,
jpayne@69 1055 UConverterToUCallback newAction,
jpayne@69 1056 const void* newContext,
jpayne@69 1057 UConverterToUCallback *oldAction,
jpayne@69 1058 const void** oldContext,
jpayne@69 1059 UErrorCode * err);
jpayne@69 1060
jpayne@69 1061 /**
jpayne@69 1062 * Changes the current callback function used by the converter when
jpayne@69 1063 * an illegal or invalid sequence is found.
jpayne@69 1064 * Context pointers are always owned by the caller.
jpayne@69 1065 * Predefined actions and contexts can be found in the ucnv_err.h header.
jpayne@69 1066 *
jpayne@69 1067 * @param converter the unicode converter
jpayne@69 1068 * @param newAction the new callback function
jpayne@69 1069 * @param newContext the new fromUnicode callback context pointer. This can be NULL.
jpayne@69 1070 * @param oldAction fillin: returns the old callback function pointer. This can be NULL.
jpayne@69 1071 * @param oldContext fillin: returns the old callback's private void* context. This can be NULL.
jpayne@69 1072 * @param err The error code status
jpayne@69 1073 * @see ucnv_getFromUCallBack
jpayne@69 1074 * @stable ICU 2.0
jpayne@69 1075 */
jpayne@69 1076 U_STABLE void U_EXPORT2
jpayne@69 1077 ucnv_setFromUCallBack (UConverter * converter,
jpayne@69 1078 UConverterFromUCallback newAction,
jpayne@69 1079 const void *newContext,
jpayne@69 1080 UConverterFromUCallback *oldAction,
jpayne@69 1081 const void **oldContext,
jpayne@69 1082 UErrorCode * err);
jpayne@69 1083
jpayne@69 1084 /**
jpayne@69 1085 * Converts an array of unicode characters to an array of codepage
jpayne@69 1086 * characters. This function is optimized for converting a continuous
jpayne@69 1087 * stream of data in buffer-sized chunks, where the entire source and
jpayne@69 1088 * target does not fit in available buffers.
jpayne@69 1089 *
jpayne@69 1090 * The source pointer is an in/out parameter. It starts out pointing where the
jpayne@69 1091 * conversion is to begin, and ends up pointing after the last UChar consumed.
jpayne@69 1092 *
jpayne@69 1093 * Target similarly starts out pointer at the first available byte in the output
jpayne@69 1094 * buffer, and ends up pointing after the last byte written to the output.
jpayne@69 1095 *
jpayne@69 1096 * The converter always attempts to consume the entire source buffer, unless
jpayne@69 1097 * (1.) the target buffer is full, or (2.) a failing error is returned from the
jpayne@69 1098 * current callback function. When a successful error status has been
jpayne@69 1099 * returned, it means that all of the source buffer has been
jpayne@69 1100 * consumed. At that point, the caller should reset the source and
jpayne@69 1101 * sourceLimit pointers to point to the next chunk.
jpayne@69 1102 *
jpayne@69 1103 * At the end of the stream (flush==TRUE), the input is completely consumed
jpayne@69 1104 * when *source==sourceLimit and no error code is set.
jpayne@69 1105 * The converter object is then automatically reset by this function.
jpayne@69 1106 * (This means that a converter need not be reset explicitly between data
jpayne@69 1107 * streams if it finishes the previous stream without errors.)
jpayne@69 1108 *
jpayne@69 1109 * This is a <I>stateful</I> conversion. Additionally, even when all source data has
jpayne@69 1110 * been consumed, some data may be in the converters' internal state.
jpayne@69 1111 * Call this function repeatedly, updating the target pointers with
jpayne@69 1112 * the next empty chunk of target in case of a
jpayne@69 1113 * <TT>U_BUFFER_OVERFLOW_ERROR</TT>, and updating the source pointers
jpayne@69 1114 * with the next chunk of source when a successful error status is
jpayne@69 1115 * returned, until there are no more chunks of source data.
jpayne@69 1116 * @param converter the Unicode converter
jpayne@69 1117 * @param target I/O parameter. Input : Points to the beginning of the buffer to copy
jpayne@69 1118 * codepage characters to. Output : points to after the last codepage character copied
jpayne@69 1119 * to <TT>target</TT>.
jpayne@69 1120 * @param targetLimit the pointer just after last of the <TT>target</TT> buffer
jpayne@69 1121 * @param source I/O parameter, pointer to pointer to the source Unicode character buffer.
jpayne@69 1122 * @param sourceLimit the pointer just after the last of the source buffer
jpayne@69 1123 * @param offsets if NULL is passed, nothing will happen to it, otherwise it needs to have the same number
jpayne@69 1124 * of allocated cells as <TT>target</TT>. Will fill in offsets from target to source pointer
jpayne@69 1125 * 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>
jpayne@69 1126 * For output data carried across calls, and other data without a specific source character
jpayne@69 1127 * (such as from escape sequences or callbacks) -1 will be placed for offsets.
jpayne@69 1128 * @param flush set to <TT>TRUE</TT> if the current source buffer is the last available
jpayne@69 1129 * chunk of the source, <TT>FALSE</TT> otherwise. Note that if a failing status is returned,
jpayne@69 1130 * this function may have to be called multiple times with flush set to <TT>TRUE</TT> until
jpayne@69 1131 * the source buffer is consumed.
jpayne@69 1132 * @param err the error status. <TT>U_ILLEGAL_ARGUMENT_ERROR</TT> will be set if the
jpayne@69 1133 * converter is <TT>NULL</TT>.
jpayne@69 1134 * <code>U_BUFFER_OVERFLOW_ERROR</code> will be set if the target is full and there is
jpayne@69 1135 * still data to be written to the target.
jpayne@69 1136 * @see ucnv_fromUChars
jpayne@69 1137 * @see ucnv_convert
jpayne@69 1138 * @see ucnv_getMinCharSize
jpayne@69 1139 * @see ucnv_setToUCallBack
jpayne@69 1140 * @stable ICU 2.0
jpayne@69 1141 */
jpayne@69 1142 U_STABLE void U_EXPORT2
jpayne@69 1143 ucnv_fromUnicode (UConverter * converter,
jpayne@69 1144 char **target,
jpayne@69 1145 const char *targetLimit,
jpayne@69 1146 const UChar ** source,
jpayne@69 1147 const UChar * sourceLimit,
jpayne@69 1148 int32_t* offsets,
jpayne@69 1149 UBool flush,
jpayne@69 1150 UErrorCode * err);
jpayne@69 1151
jpayne@69 1152 /**
jpayne@69 1153 * Converts a buffer of codepage bytes into an array of unicode UChars
jpayne@69 1154 * characters. This function is optimized for converting a continuous
jpayne@69 1155 * stream of data in buffer-sized chunks, where the entire source and
jpayne@69 1156 * target does not fit in available buffers.
jpayne@69 1157 *
jpayne@69 1158 * The source pointer is an in/out parameter. It starts out pointing where the
jpayne@69 1159 * conversion is to begin, and ends up pointing after the last byte of source consumed.
jpayne@69 1160 *
jpayne@69 1161 * Target similarly starts out pointer at the first available UChar in the output
jpayne@69 1162 * buffer, and ends up pointing after the last UChar written to the output.
jpayne@69 1163 * It does NOT necessarily keep UChar sequences together.
jpayne@69 1164 *
jpayne@69 1165 * The converter always attempts to consume the entire source buffer, unless
jpayne@69 1166 * (1.) the target buffer is full, or (2.) a failing error is returned from the
jpayne@69 1167 * current callback function. When a successful error status has been
jpayne@69 1168 * returned, it means that all of the source buffer has been
jpayne@69 1169 * consumed. At that point, the caller should reset the source and
jpayne@69 1170 * sourceLimit pointers to point to the next chunk.
jpayne@69 1171 *
jpayne@69 1172 * At the end of the stream (flush==TRUE), the input is completely consumed
jpayne@69 1173 * when *source==sourceLimit and no error code is set
jpayne@69 1174 * The converter object is then automatically reset by this function.
jpayne@69 1175 * (This means that a converter need not be reset explicitly between data
jpayne@69 1176 * streams if it finishes the previous stream without errors.)
jpayne@69 1177 *
jpayne@69 1178 * This is a <I>stateful</I> conversion. Additionally, even when all source data has
jpayne@69 1179 * been consumed, some data may be in the converters' internal state.
jpayne@69 1180 * Call this function repeatedly, updating the target pointers with
jpayne@69 1181 * the next empty chunk of target in case of a
jpayne@69 1182 * <TT>U_BUFFER_OVERFLOW_ERROR</TT>, and updating the source pointers
jpayne@69 1183 * with the next chunk of source when a successful error status is
jpayne@69 1184 * returned, until there are no more chunks of source data.
jpayne@69 1185 * @param converter the Unicode converter
jpayne@69 1186 * @param target I/O parameter. Input : Points to the beginning of the buffer to copy
jpayne@69 1187 * UChars into. Output : points to after the last UChar copied.
jpayne@69 1188 * @param targetLimit the pointer just after the end of the <TT>target</TT> buffer
jpayne@69 1189 * @param source I/O parameter, pointer to pointer to the source codepage buffer.
jpayne@69 1190 * @param sourceLimit the pointer to the byte after the end of the source buffer
jpayne@69 1191 * @param offsets if NULL is passed, nothing will happen to it, otherwise it needs to have the same number
jpayne@69 1192 * of allocated cells as <TT>target</TT>. Will fill in offsets from target to source pointer
jpayne@69 1193 * 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>
jpayne@69 1194 * For output data carried across calls, and other data without a specific source character
jpayne@69 1195 * (such as from escape sequences or callbacks) -1 will be placed for offsets.
jpayne@69 1196 * @param flush set to <TT>TRUE</TT> if the current source buffer is the last available
jpayne@69 1197 * chunk of the source, <TT>FALSE</TT> otherwise. Note that if a failing status is returned,
jpayne@69 1198 * this function may have to be called multiple times with flush set to <TT>TRUE</TT> until
jpayne@69 1199 * the source buffer is consumed.
jpayne@69 1200 * @param err the error status. <TT>U_ILLEGAL_ARGUMENT_ERROR</TT> will be set if the
jpayne@69 1201 * converter is <TT>NULL</TT>.
jpayne@69 1202 * <code>U_BUFFER_OVERFLOW_ERROR</code> will be set if the target is full and there is
jpayne@69 1203 * still data to be written to the target.
jpayne@69 1204 * @see ucnv_fromUChars
jpayne@69 1205 * @see ucnv_convert
jpayne@69 1206 * @see ucnv_getMinCharSize
jpayne@69 1207 * @see ucnv_setFromUCallBack
jpayne@69 1208 * @see ucnv_getNextUChar
jpayne@69 1209 * @stable ICU 2.0
jpayne@69 1210 */
jpayne@69 1211 U_STABLE void U_EXPORT2
jpayne@69 1212 ucnv_toUnicode(UConverter *converter,
jpayne@69 1213 UChar **target,
jpayne@69 1214 const UChar *targetLimit,
jpayne@69 1215 const char **source,
jpayne@69 1216 const char *sourceLimit,
jpayne@69 1217 int32_t *offsets,
jpayne@69 1218 UBool flush,
jpayne@69 1219 UErrorCode *err);
jpayne@69 1220
jpayne@69 1221 /**
jpayne@69 1222 * Convert the Unicode string into a codepage string using an existing UConverter.
jpayne@69 1223 * The output string is NUL-terminated if possible.
jpayne@69 1224 *
jpayne@69 1225 * This function is a more convenient but less powerful version of ucnv_fromUnicode().
jpayne@69 1226 * It is only useful for whole strings, not for streaming conversion.
jpayne@69 1227 *
jpayne@69 1228 * The maximum output buffer capacity required (barring output from callbacks) will be
jpayne@69 1229 * UCNV_GET_MAX_BYTES_FOR_STRING(srcLength, ucnv_getMaxCharSize(cnv)).
jpayne@69 1230 *
jpayne@69 1231 * @param cnv the converter object to be used (ucnv_resetFromUnicode() will be called)
jpayne@69 1232 * @param src the input Unicode string
jpayne@69 1233 * @param srcLength the input string length, or -1 if NUL-terminated
jpayne@69 1234 * @param dest destination string buffer, can be NULL if destCapacity==0
jpayne@69 1235 * @param destCapacity the number of chars available at dest
jpayne@69 1236 * @param pErrorCode normal ICU error code;
jpayne@69 1237 * common error codes that may be set by this function include
jpayne@69 1238 * U_BUFFER_OVERFLOW_ERROR, U_STRING_NOT_TERMINATED_WARNING,
jpayne@69 1239 * U_ILLEGAL_ARGUMENT_ERROR, and conversion errors
jpayne@69 1240 * @return the length of the output string, not counting the terminating NUL;
jpayne@69 1241 * if the length is greater than destCapacity, then the string will not fit
jpayne@69 1242 * and a buffer of the indicated length would need to be passed in
jpayne@69 1243 * @see ucnv_fromUnicode
jpayne@69 1244 * @see ucnv_convert
jpayne@69 1245 * @see UCNV_GET_MAX_BYTES_FOR_STRING
jpayne@69 1246 * @stable ICU 2.0
jpayne@69 1247 */
jpayne@69 1248 U_STABLE int32_t U_EXPORT2
jpayne@69 1249 ucnv_fromUChars(UConverter *cnv,
jpayne@69 1250 char *dest, int32_t destCapacity,
jpayne@69 1251 const UChar *src, int32_t srcLength,
jpayne@69 1252 UErrorCode *pErrorCode);
jpayne@69 1253
jpayne@69 1254 /**
jpayne@69 1255 * Convert the codepage string into a Unicode string using an existing UConverter.
jpayne@69 1256 * The output string is NUL-terminated if possible.
jpayne@69 1257 *
jpayne@69 1258 * This function is a more convenient but less powerful version of ucnv_toUnicode().
jpayne@69 1259 * It is only useful for whole strings, not for streaming conversion.
jpayne@69 1260 *
jpayne@69 1261 * The maximum output buffer capacity required (barring output from callbacks) will be
jpayne@69 1262 * 2*srcLength (each char may be converted into a surrogate pair).
jpayne@69 1263 *
jpayne@69 1264 * @param cnv the converter object to be used (ucnv_resetToUnicode() will be called)
jpayne@69 1265 * @param src the input codepage string
jpayne@69 1266 * @param srcLength the input string length, or -1 if NUL-terminated
jpayne@69 1267 * @param dest destination string buffer, can be NULL if destCapacity==0
jpayne@69 1268 * @param destCapacity the number of UChars available at dest
jpayne@69 1269 * @param pErrorCode normal ICU error code;
jpayne@69 1270 * common error codes that may be set by this function include
jpayne@69 1271 * U_BUFFER_OVERFLOW_ERROR, U_STRING_NOT_TERMINATED_WARNING,
jpayne@69 1272 * U_ILLEGAL_ARGUMENT_ERROR, and conversion errors
jpayne@69 1273 * @return the length of the output string, not counting the terminating NUL;
jpayne@69 1274 * if the length is greater than destCapacity, then the string will not fit
jpayne@69 1275 * and a buffer of the indicated length would need to be passed in
jpayne@69 1276 * @see ucnv_toUnicode
jpayne@69 1277 * @see ucnv_convert
jpayne@69 1278 * @stable ICU 2.0
jpayne@69 1279 */
jpayne@69 1280 U_STABLE int32_t U_EXPORT2
jpayne@69 1281 ucnv_toUChars(UConverter *cnv,
jpayne@69 1282 UChar *dest, int32_t destCapacity,
jpayne@69 1283 const char *src, int32_t srcLength,
jpayne@69 1284 UErrorCode *pErrorCode);
jpayne@69 1285
jpayne@69 1286 /**
jpayne@69 1287 * Convert a codepage buffer into Unicode one character at a time.
jpayne@69 1288 * The input is completely consumed when the U_INDEX_OUTOFBOUNDS_ERROR is set.
jpayne@69 1289 *
jpayne@69 1290 * Advantage compared to ucnv_toUnicode() or ucnv_toUChars():
jpayne@69 1291 * - Faster for small amounts of data, for most converters, e.g.,
jpayne@69 1292 * US-ASCII, ISO-8859-1, UTF-8/16/32, and most "normal" charsets.
jpayne@69 1293 * (For complex converters, e.g., SCSU, UTF-7 and ISO 2022 variants,
jpayne@69 1294 * it uses ucnv_toUnicode() internally.)
jpayne@69 1295 * - Convenient.
jpayne@69 1296 *
jpayne@69 1297 * Limitations compared to ucnv_toUnicode():
jpayne@69 1298 * - Always assumes flush=TRUE.
jpayne@69 1299 * This makes ucnv_getNextUChar() unsuitable for "streaming" conversion,
jpayne@69 1300 * that is, for where the input is supplied in multiple buffers,
jpayne@69 1301 * because ucnv_getNextUChar() will assume the end of the input at the end
jpayne@69 1302 * of the first buffer.
jpayne@69 1303 * - Does not provide offset output.
jpayne@69 1304 *
jpayne@69 1305 * It is possible to "mix" ucnv_getNextUChar() and ucnv_toUnicode() because
jpayne@69 1306 * ucnv_getNextUChar() uses the current state of the converter
jpayne@69 1307 * (unlike ucnv_toUChars() which always resets first).
jpayne@69 1308 * However, if ucnv_getNextUChar() is called after ucnv_toUnicode()
jpayne@69 1309 * stopped in the middle of a character sequence (with flush=FALSE),
jpayne@69 1310 * then ucnv_getNextUChar() will always use the slower ucnv_toUnicode()
jpayne@69 1311 * internally until the next character boundary.
jpayne@69 1312 * (This is new in ICU 2.6. In earlier releases, ucnv_getNextUChar() had to
jpayne@69 1313 * start at a character boundary.)
jpayne@69 1314 *
jpayne@69 1315 * Instead of using ucnv_getNextUChar(), it is recommended
jpayne@69 1316 * to convert using ucnv_toUnicode() or ucnv_toUChars()
jpayne@69 1317 * and then iterate over the text using U16_NEXT() or a UCharIterator (uiter.h)
jpayne@69 1318 * or a C++ CharacterIterator or similar.
jpayne@69 1319 * This allows streaming conversion and offset output, for example.
jpayne@69 1320 *
jpayne@69 1321 * <p>Handling of surrogate pairs and supplementary-plane code points:<br>
jpayne@69 1322 * There are two different kinds of codepages that provide mappings for surrogate characters:
jpayne@69 1323 * <ul>
jpayne@69 1324 * <li>Codepages like UTF-8, UTF-32, and GB 18030 provide direct representations for Unicode
jpayne@69 1325 * code points U+10000-U+10ffff as well as for single surrogates U+d800-U+dfff.
jpayne@69 1326 * Each valid sequence will result in exactly one returned code point.
jpayne@69 1327 * If a sequence results in a single surrogate, then that will be returned
jpayne@69 1328 * by itself, even if a neighboring sequence encodes the matching surrogate.</li>
jpayne@69 1329 * <li>Codepages like SCSU and LMBCS (and UTF-16) provide direct representations only for BMP code points
jpayne@69 1330 * including surrogates. Code points in supplementary planes are represented with
jpayne@69 1331 * two sequences, each encoding a surrogate.
jpayne@69 1332 * For these codepages, matching pairs of surrogates will be combined into single
jpayne@69 1333 * code points for returning from this function.
jpayne@69 1334 * (Note that SCSU is actually a mix of these codepage types.)</li>
jpayne@69 1335 * </ul></p>
jpayne@69 1336 *
jpayne@69 1337 * @param converter an open UConverter
jpayne@69 1338 * @param source the address of a pointer to the codepage buffer, will be
jpayne@69 1339 * updated to point after the bytes consumed in the conversion call.
jpayne@69 1340 * @param sourceLimit points to the end of the input buffer
jpayne@69 1341 * @param err fills in error status (see ucnv_toUnicode)
jpayne@69 1342 * <code>U_INDEX_OUTOFBOUNDS_ERROR</code> will be set if the input
jpayne@69 1343 * is empty or does not convert to any output (e.g.: pure state-change
jpayne@69 1344 * codes SI/SO, escape sequences for ISO 2022,
jpayne@69 1345 * or if the callback did not output anything, ...).
jpayne@69 1346 * This function will not set a <code>U_BUFFER_OVERFLOW_ERROR</code> because
jpayne@69 1347 * the "buffer" is the return code. However, there might be subsequent output
jpayne@69 1348 * stored in the converter object
jpayne@69 1349 * that will be returned in following calls to this function.
jpayne@69 1350 * @return a UChar32 resulting from the partial conversion of source
jpayne@69 1351 * @see ucnv_toUnicode
jpayne@69 1352 * @see ucnv_toUChars
jpayne@69 1353 * @see ucnv_convert
jpayne@69 1354 * @stable ICU 2.0
jpayne@69 1355 */
jpayne@69 1356 U_STABLE UChar32 U_EXPORT2
jpayne@69 1357 ucnv_getNextUChar(UConverter * converter,
jpayne@69 1358 const char **source,
jpayne@69 1359 const char * sourceLimit,
jpayne@69 1360 UErrorCode * err);
jpayne@69 1361
jpayne@69 1362 /**
jpayne@69 1363 * Convert from one external charset to another using two existing UConverters.
jpayne@69 1364 * Internally, two conversions - ucnv_toUnicode() and ucnv_fromUnicode() -
jpayne@69 1365 * are used, "pivoting" through 16-bit Unicode.
jpayne@69 1366 *
jpayne@69 1367 * Important: For streaming conversion (multiple function calls for successive
jpayne@69 1368 * parts of a text stream), the caller must provide a pivot buffer explicitly,
jpayne@69 1369 * and must preserve the pivot buffer and associated pointers from one
jpayne@69 1370 * call to another. (The buffer may be moved if its contents and the relative
jpayne@69 1371 * pointer positions are preserved.)
jpayne@69 1372 *
jpayne@69 1373 * There is a similar function, ucnv_convert(),
jpayne@69 1374 * which has the following limitations:
jpayne@69 1375 * - it takes charset names, not converter objects, so that
jpayne@69 1376 * - two converters are opened for each call
jpayne@69 1377 * - only single-string conversion is possible, not streaming operation
jpayne@69 1378 * - it does not provide enough information to find out,
jpayne@69 1379 * in case of failure, whether the toUnicode or
jpayne@69 1380 * the fromUnicode conversion failed
jpayne@69 1381 *
jpayne@69 1382 * By contrast, ucnv_convertEx()
jpayne@69 1383 * - takes UConverter parameters instead of charset names
jpayne@69 1384 * - fully exposes the pivot buffer for streaming conversion and complete error handling
jpayne@69 1385 *
jpayne@69 1386 * ucnv_convertEx() also provides further convenience:
jpayne@69 1387 * - an option to reset the converters at the beginning
jpayne@69 1388 * (if reset==TRUE, see parameters;
jpayne@69 1389 * also sets *pivotTarget=*pivotSource=pivotStart)
jpayne@69 1390 * - allow NUL-terminated input
jpayne@69 1391 * (only a single NUL byte, will not work for charsets with multi-byte NULs)
jpayne@69 1392 * (if sourceLimit==NULL, see parameters)
jpayne@69 1393 * - terminate with a NUL on output
jpayne@69 1394 * (only a single NUL byte, not useful for charsets with multi-byte NULs),
jpayne@69 1395 * or set U_STRING_NOT_TERMINATED_WARNING if the output exactly fills
jpayne@69 1396 * the target buffer
jpayne@69 1397 * - the pivot buffer can be provided internally;
jpayne@69 1398 * possible only for whole-string conversion, not streaming conversion;
jpayne@69 1399 * in this case, the caller will not be able to get details about where an
jpayne@69 1400 * error occurred
jpayne@69 1401 * (if pivotStart==NULL, see below)
jpayne@69 1402 *
jpayne@69 1403 * The function returns when one of the following is true:
jpayne@69 1404 * - the entire source text has been converted successfully to the target buffer
jpayne@69 1405 * - a target buffer overflow occurred (U_BUFFER_OVERFLOW_ERROR)
jpayne@69 1406 * - a conversion error occurred
jpayne@69 1407 * (other U_FAILURE(), see description of pErrorCode)
jpayne@69 1408 *
jpayne@69 1409 * Limitation compared to the direct use of
jpayne@69 1410 * ucnv_fromUnicode() and ucnv_toUnicode():
jpayne@69 1411 * ucnv_convertEx() does not provide offset information.
jpayne@69 1412 *
jpayne@69 1413 * Limitation compared to ucnv_fromUChars() and ucnv_toUChars():
jpayne@69 1414 * ucnv_convertEx() does not support preflighting directly.
jpayne@69 1415 *
jpayne@69 1416 * Sample code for converting a single string from
jpayne@69 1417 * one external charset to UTF-8, ignoring the location of errors:
jpayne@69 1418 *
jpayne@69 1419 * \code
jpayne@69 1420 * int32_t
jpayne@69 1421 * myToUTF8(UConverter *cnv,
jpayne@69 1422 * const char *s, int32_t length,
jpayne@69 1423 * char *u8, int32_t capacity,
jpayne@69 1424 * UErrorCode *pErrorCode) {
jpayne@69 1425 * UConverter *utf8Cnv;
jpayne@69 1426 * char *target;
jpayne@69 1427 *
jpayne@69 1428 * if(U_FAILURE(*pErrorCode)) {
jpayne@69 1429 * return 0;
jpayne@69 1430 * }
jpayne@69 1431 *
jpayne@69 1432 * utf8Cnv=myGetCachedUTF8Converter(pErrorCode);
jpayne@69 1433 * if(U_FAILURE(*pErrorCode)) {
jpayne@69 1434 * return 0;
jpayne@69 1435 * }
jpayne@69 1436 *
jpayne@69 1437 * if(length<0) {
jpayne@69 1438 * length=strlen(s);
jpayne@69 1439 * }
jpayne@69 1440 * target=u8;
jpayne@69 1441 * ucnv_convertEx(utf8Cnv, cnv,
jpayne@69 1442 * &target, u8+capacity,
jpayne@69 1443 * &s, s+length,
jpayne@69 1444 * NULL, NULL, NULL, NULL,
jpayne@69 1445 * TRUE, TRUE,
jpayne@69 1446 * pErrorCode);
jpayne@69 1447 *
jpayne@69 1448 * myReleaseCachedUTF8Converter(utf8Cnv);
jpayne@69 1449 *
jpayne@69 1450 * // return the output string length, but without preflighting
jpayne@69 1451 * return (int32_t)(target-u8);
jpayne@69 1452 * }
jpayne@69 1453 * \endcode
jpayne@69 1454 *
jpayne@69 1455 * @param targetCnv Output converter, used to convert from the UTF-16 pivot
jpayne@69 1456 * to the target using ucnv_fromUnicode().
jpayne@69 1457 * @param sourceCnv Input converter, used to convert from the source to
jpayne@69 1458 * the UTF-16 pivot using ucnv_toUnicode().
jpayne@69 1459 * @param target I/O parameter, same as for ucnv_fromUChars().
jpayne@69 1460 * Input: *target points to the beginning of the target buffer.
jpayne@69 1461 * Output: *target points to the first unit after the last char written.
jpayne@69 1462 * @param targetLimit Pointer to the first unit after the target buffer.
jpayne@69 1463 * @param source I/O parameter, same as for ucnv_toUChars().
jpayne@69 1464 * Input: *source points to the beginning of the source buffer.
jpayne@69 1465 * Output: *source points to the first unit after the last char read.
jpayne@69 1466 * @param sourceLimit Pointer to the first unit after the source buffer.
jpayne@69 1467 * @param pivotStart Pointer to the UTF-16 pivot buffer. If pivotStart==NULL,
jpayne@69 1468 * then an internal buffer is used and the other pivot
jpayne@69 1469 * arguments are ignored and can be NULL as well.
jpayne@69 1470 * @param pivotSource I/O parameter, same as source in ucnv_fromUChars() for
jpayne@69 1471 * conversion from the pivot buffer to the target buffer.
jpayne@69 1472 * @param pivotTarget I/O parameter, same as target in ucnv_toUChars() for
jpayne@69 1473 * conversion from the source buffer to the pivot buffer.
jpayne@69 1474 * It must be pivotStart<=*pivotSource<=*pivotTarget<=pivotLimit
jpayne@69 1475 * and pivotStart<pivotLimit (unless pivotStart==NULL).
jpayne@69 1476 * @param pivotLimit Pointer to the first unit after the pivot buffer.
jpayne@69 1477 * @param reset If TRUE, then ucnv_resetToUnicode(sourceCnv) and
jpayne@69 1478 * ucnv_resetFromUnicode(targetCnv) are called, and the
jpayne@69 1479 * pivot pointers are reset (*pivotTarget=*pivotSource=pivotStart).
jpayne@69 1480 * @param flush If true, indicates the end of the input.
jpayne@69 1481 * Passed directly to ucnv_toUnicode(), and carried over to
jpayne@69 1482 * ucnv_fromUnicode() when the source is empty as well.
jpayne@69 1483 * @param pErrorCode ICU error code in/out parameter.
jpayne@69 1484 * Must fulfill U_SUCCESS before the function call.
jpayne@69 1485 * U_BUFFER_OVERFLOW_ERROR always refers to the target buffer
jpayne@69 1486 * because overflows into the pivot buffer are handled internally.
jpayne@69 1487 * Other conversion errors are from the source-to-pivot
jpayne@69 1488 * conversion if *pivotSource==pivotStart, otherwise from
jpayne@69 1489 * the pivot-to-target conversion.
jpayne@69 1490 *
jpayne@69 1491 * @see ucnv_convert
jpayne@69 1492 * @see ucnv_fromAlgorithmic
jpayne@69 1493 * @see ucnv_toAlgorithmic
jpayne@69 1494 * @see ucnv_fromUnicode
jpayne@69 1495 * @see ucnv_toUnicode
jpayne@69 1496 * @see ucnv_fromUChars
jpayne@69 1497 * @see ucnv_toUChars
jpayne@69 1498 * @stable ICU 2.6
jpayne@69 1499 */
jpayne@69 1500 U_STABLE void U_EXPORT2
jpayne@69 1501 ucnv_convertEx(UConverter *targetCnv, UConverter *sourceCnv,
jpayne@69 1502 char **target, const char *targetLimit,
jpayne@69 1503 const char **source, const char *sourceLimit,
jpayne@69 1504 UChar *pivotStart, UChar **pivotSource,
jpayne@69 1505 UChar **pivotTarget, const UChar *pivotLimit,
jpayne@69 1506 UBool reset, UBool flush,
jpayne@69 1507 UErrorCode *pErrorCode);
jpayne@69 1508
jpayne@69 1509 /**
jpayne@69 1510 * Convert from one external charset to another.
jpayne@69 1511 * Internally, two converters are opened according to the name arguments,
jpayne@69 1512 * then the text is converted to and from the 16-bit Unicode "pivot"
jpayne@69 1513 * using ucnv_convertEx(), then the converters are closed again.
jpayne@69 1514 *
jpayne@69 1515 * This is a convenience function, not an efficient way to convert a lot of text:
jpayne@69 1516 * ucnv_convert()
jpayne@69 1517 * - takes charset names, not converter objects, so that
jpayne@69 1518 * - two converters are opened for each call
jpayne@69 1519 * - only single-string conversion is possible, not streaming operation
jpayne@69 1520 * - does not provide enough information to find out,
jpayne@69 1521 * in case of failure, whether the toUnicode or
jpayne@69 1522 * the fromUnicode conversion failed
jpayne@69 1523 * - allows NUL-terminated input
jpayne@69 1524 * (only a single NUL byte, will not work for charsets with multi-byte NULs)
jpayne@69 1525 * (if sourceLength==-1, see parameters)
jpayne@69 1526 * - terminate with a NUL on output
jpayne@69 1527 * (only a single NUL byte, not useful for charsets with multi-byte NULs),
jpayne@69 1528 * or set U_STRING_NOT_TERMINATED_WARNING if the output exactly fills
jpayne@69 1529 * the target buffer
jpayne@69 1530 * - a pivot buffer is provided internally
jpayne@69 1531 *
jpayne@69 1532 * The function returns when one of the following is true:
jpayne@69 1533 * - the entire source text has been converted successfully to the target buffer
jpayne@69 1534 * and either the target buffer is terminated with a single NUL byte
jpayne@69 1535 * or the error code is set to U_STRING_NOT_TERMINATED_WARNING
jpayne@69 1536 * - a target buffer overflow occurred (U_BUFFER_OVERFLOW_ERROR)
jpayne@69 1537 * and the full output string length is returned ("preflighting")
jpayne@69 1538 * - a conversion error occurred
jpayne@69 1539 * (other U_FAILURE(), see description of pErrorCode)
jpayne@69 1540 *
jpayne@69 1541 * @param toConverterName The name of the converter that is used to convert
jpayne@69 1542 * from the UTF-16 pivot buffer to the target.
jpayne@69 1543 * @param fromConverterName The name of the converter that is used to convert
jpayne@69 1544 * from the source to the UTF-16 pivot buffer.
jpayne@69 1545 * @param target Pointer to the output buffer.
jpayne@69 1546 * @param targetCapacity Capacity of the target, in bytes.
jpayne@69 1547 * @param source Pointer to the input buffer.
jpayne@69 1548 * @param sourceLength Length of the input text, in bytes, or -1 for NUL-terminated input.
jpayne@69 1549 * @param pErrorCode ICU error code in/out parameter.
jpayne@69 1550 * Must fulfill U_SUCCESS before the function call.
jpayne@69 1551 * @return Length of the complete output text in bytes, even if it exceeds the targetCapacity
jpayne@69 1552 * and a U_BUFFER_OVERFLOW_ERROR is set.
jpayne@69 1553 *
jpayne@69 1554 * @see ucnv_convertEx
jpayne@69 1555 * @see ucnv_fromAlgorithmic
jpayne@69 1556 * @see ucnv_toAlgorithmic
jpayne@69 1557 * @see ucnv_fromUnicode
jpayne@69 1558 * @see ucnv_toUnicode
jpayne@69 1559 * @see ucnv_fromUChars
jpayne@69 1560 * @see ucnv_toUChars
jpayne@69 1561 * @see ucnv_getNextUChar
jpayne@69 1562 * @stable ICU 2.0
jpayne@69 1563 */
jpayne@69 1564 U_STABLE int32_t U_EXPORT2
jpayne@69 1565 ucnv_convert(const char *toConverterName,
jpayne@69 1566 const char *fromConverterName,
jpayne@69 1567 char *target,
jpayne@69 1568 int32_t targetCapacity,
jpayne@69 1569 const char *source,
jpayne@69 1570 int32_t sourceLength,
jpayne@69 1571 UErrorCode *pErrorCode);
jpayne@69 1572
jpayne@69 1573 /**
jpayne@69 1574 * Convert from one external charset to another.
jpayne@69 1575 * Internally, the text is converted to and from the 16-bit Unicode "pivot"
jpayne@69 1576 * using ucnv_convertEx(). ucnv_toAlgorithmic() works exactly like ucnv_convert()
jpayne@69 1577 * except that the two converters need not be looked up and opened completely.
jpayne@69 1578 *
jpayne@69 1579 * The source-to-pivot conversion uses the cnv converter parameter.
jpayne@69 1580 * The pivot-to-target conversion uses a purely algorithmic converter
jpayne@69 1581 * according to the specified type, e.g., UCNV_UTF8 for a UTF-8 converter.
jpayne@69 1582 *
jpayne@69 1583 * Internally, the algorithmic converter is opened and closed for each
jpayne@69 1584 * function call, which is more efficient than using the public ucnv_open()
jpayne@69 1585 * but somewhat less efficient than only resetting an existing converter
jpayne@69 1586 * and using ucnv_convertEx().
jpayne@69 1587 *
jpayne@69 1588 * This function is more convenient than ucnv_convertEx() for single-string
jpayne@69 1589 * conversions, especially when "preflighting" is desired (returning the length
jpayne@69 1590 * of the complete output even if it does not fit into the target buffer;
jpayne@69 1591 * see the User Guide Strings chapter). See ucnv_convert() for details.
jpayne@69 1592 *
jpayne@69 1593 * @param algorithmicType UConverterType constant identifying the desired target
jpayne@69 1594 * charset as a purely algorithmic converter.
jpayne@69 1595 * Those are converters for Unicode charsets like
jpayne@69 1596 * UTF-8, BOCU-1, SCSU, UTF-7, IMAP-mailbox-name, etc.,
jpayne@69 1597 * as well as US-ASCII and ISO-8859-1.
jpayne@69 1598 * @param cnv The converter that is used to convert
jpayne@69 1599 * from the source to the UTF-16 pivot buffer.
jpayne@69 1600 * @param target Pointer to the output buffer.
jpayne@69 1601 * @param targetCapacity Capacity of the target, in bytes.
jpayne@69 1602 * @param source Pointer to the input buffer.
jpayne@69 1603 * @param sourceLength Length of the input text, in bytes
jpayne@69 1604 * @param pErrorCode ICU error code in/out parameter.
jpayne@69 1605 * Must fulfill U_SUCCESS before the function call.
jpayne@69 1606 * @return Length of the complete output text in bytes, even if it exceeds the targetCapacity
jpayne@69 1607 * and a U_BUFFER_OVERFLOW_ERROR is set.
jpayne@69 1608 *
jpayne@69 1609 * @see ucnv_fromAlgorithmic
jpayne@69 1610 * @see ucnv_convert
jpayne@69 1611 * @see ucnv_convertEx
jpayne@69 1612 * @see ucnv_fromUnicode
jpayne@69 1613 * @see ucnv_toUnicode
jpayne@69 1614 * @see ucnv_fromUChars
jpayne@69 1615 * @see ucnv_toUChars
jpayne@69 1616 * @stable ICU 2.6
jpayne@69 1617 */
jpayne@69 1618 U_STABLE int32_t U_EXPORT2
jpayne@69 1619 ucnv_toAlgorithmic(UConverterType algorithmicType,
jpayne@69 1620 UConverter *cnv,
jpayne@69 1621 char *target, int32_t targetCapacity,
jpayne@69 1622 const char *source, int32_t sourceLength,
jpayne@69 1623 UErrorCode *pErrorCode);
jpayne@69 1624
jpayne@69 1625 /**
jpayne@69 1626 * Convert from one external charset to another.
jpayne@69 1627 * Internally, the text is converted to and from the 16-bit Unicode "pivot"
jpayne@69 1628 * using ucnv_convertEx(). ucnv_fromAlgorithmic() works exactly like ucnv_convert()
jpayne@69 1629 * except that the two converters need not be looked up and opened completely.
jpayne@69 1630 *
jpayne@69 1631 * The source-to-pivot conversion uses a purely algorithmic converter
jpayne@69 1632 * according to the specified type, e.g., UCNV_UTF8 for a UTF-8 converter.
jpayne@69 1633 * The pivot-to-target conversion uses the cnv converter parameter.
jpayne@69 1634 *
jpayne@69 1635 * Internally, the algorithmic converter is opened and closed for each
jpayne@69 1636 * function call, which is more efficient than using the public ucnv_open()
jpayne@69 1637 * but somewhat less efficient than only resetting an existing converter
jpayne@69 1638 * and using ucnv_convertEx().
jpayne@69 1639 *
jpayne@69 1640 * This function is more convenient than ucnv_convertEx() for single-string
jpayne@69 1641 * conversions, especially when "preflighting" is desired (returning the length
jpayne@69 1642 * of the complete output even if it does not fit into the target buffer;
jpayne@69 1643 * see the User Guide Strings chapter). See ucnv_convert() for details.
jpayne@69 1644 *
jpayne@69 1645 * @param cnv The converter that is used to convert
jpayne@69 1646 * from the UTF-16 pivot buffer to the target.
jpayne@69 1647 * @param algorithmicType UConverterType constant identifying the desired source
jpayne@69 1648 * charset as a purely algorithmic converter.
jpayne@69 1649 * Those are converters for Unicode charsets like
jpayne@69 1650 * UTF-8, BOCU-1, SCSU, UTF-7, IMAP-mailbox-name, etc.,
jpayne@69 1651 * as well as US-ASCII and ISO-8859-1.
jpayne@69 1652 * @param target Pointer to the output buffer.
jpayne@69 1653 * @param targetCapacity Capacity of the target, in bytes.
jpayne@69 1654 * @param source Pointer to the input buffer.
jpayne@69 1655 * @param sourceLength Length of the input text, in bytes
jpayne@69 1656 * @param pErrorCode ICU error code in/out parameter.
jpayne@69 1657 * Must fulfill U_SUCCESS before the function call.
jpayne@69 1658 * @return Length of the complete output text in bytes, even if it exceeds the targetCapacity
jpayne@69 1659 * and a U_BUFFER_OVERFLOW_ERROR is set.
jpayne@69 1660 *
jpayne@69 1661 * @see ucnv_fromAlgorithmic
jpayne@69 1662 * @see ucnv_convert
jpayne@69 1663 * @see ucnv_convertEx
jpayne@69 1664 * @see ucnv_fromUnicode
jpayne@69 1665 * @see ucnv_toUnicode
jpayne@69 1666 * @see ucnv_fromUChars
jpayne@69 1667 * @see ucnv_toUChars
jpayne@69 1668 * @stable ICU 2.6
jpayne@69 1669 */
jpayne@69 1670 U_STABLE int32_t U_EXPORT2
jpayne@69 1671 ucnv_fromAlgorithmic(UConverter *cnv,
jpayne@69 1672 UConverterType algorithmicType,
jpayne@69 1673 char *target, int32_t targetCapacity,
jpayne@69 1674 const char *source, int32_t sourceLength,
jpayne@69 1675 UErrorCode *pErrorCode);
jpayne@69 1676
jpayne@69 1677 /**
jpayne@69 1678 * Frees up memory occupied by unused, cached converter shared data.
jpayne@69 1679 *
jpayne@69 1680 * @return the number of cached converters successfully deleted
jpayne@69 1681 * @see ucnv_close
jpayne@69 1682 * @stable ICU 2.0
jpayne@69 1683 */
jpayne@69 1684 U_STABLE int32_t U_EXPORT2
jpayne@69 1685 ucnv_flushCache(void);
jpayne@69 1686
jpayne@69 1687 /**
jpayne@69 1688 * Returns the number of available converters, as per the alias file.
jpayne@69 1689 *
jpayne@69 1690 * @return the number of available converters
jpayne@69 1691 * @see ucnv_getAvailableName
jpayne@69 1692 * @stable ICU 2.0
jpayne@69 1693 */
jpayne@69 1694 U_STABLE int32_t U_EXPORT2
jpayne@69 1695 ucnv_countAvailable(void);
jpayne@69 1696
jpayne@69 1697 /**
jpayne@69 1698 * Gets the canonical converter name of the specified converter from a list of
jpayne@69 1699 * all available converters contaied in the alias file. All converters
jpayne@69 1700 * in this list can be opened.
jpayne@69 1701 *
jpayne@69 1702 * @param n the index to a converter available on the system (in the range <TT>[0..ucnv_countAvaiable()]</TT>)
jpayne@69 1703 * @return a pointer a string (library owned), or <TT>NULL</TT> if the index is out of bounds.
jpayne@69 1704 * @see ucnv_countAvailable
jpayne@69 1705 * @stable ICU 2.0
jpayne@69 1706 */
jpayne@69 1707 U_STABLE const char* U_EXPORT2
jpayne@69 1708 ucnv_getAvailableName(int32_t n);
jpayne@69 1709
jpayne@69 1710 /**
jpayne@69 1711 * Returns a UEnumeration to enumerate all of the canonical converter
jpayne@69 1712 * names, as per the alias file, regardless of the ability to open each
jpayne@69 1713 * converter.
jpayne@69 1714 *
jpayne@69 1715 * @return A UEnumeration object for getting all the recognized canonical
jpayne@69 1716 * converter names.
jpayne@69 1717 * @see ucnv_getAvailableName
jpayne@69 1718 * @see uenum_close
jpayne@69 1719 * @see uenum_next
jpayne@69 1720 * @stable ICU 2.4
jpayne@69 1721 */
jpayne@69 1722 U_STABLE UEnumeration * U_EXPORT2
jpayne@69 1723 ucnv_openAllNames(UErrorCode *pErrorCode);
jpayne@69 1724
jpayne@69 1725 /**
jpayne@69 1726 * Gives the number of aliases for a given converter or alias name.
jpayne@69 1727 * If the alias is ambiguous, then the preferred converter is used
jpayne@69 1728 * and the status is set to U_AMBIGUOUS_ALIAS_WARNING.
jpayne@69 1729 * This method only enumerates the listed entries in the alias file.
jpayne@69 1730 * @param alias alias name
jpayne@69 1731 * @param pErrorCode error status
jpayne@69 1732 * @return number of names on alias list for given alias
jpayne@69 1733 * @stable ICU 2.0
jpayne@69 1734 */
jpayne@69 1735 U_STABLE uint16_t U_EXPORT2
jpayne@69 1736 ucnv_countAliases(const char *alias, UErrorCode *pErrorCode);
jpayne@69 1737
jpayne@69 1738 /**
jpayne@69 1739 * Gives the name of the alias at given index of alias list.
jpayne@69 1740 * This method only enumerates the listed entries in the alias file.
jpayne@69 1741 * If the alias is ambiguous, then the preferred converter is used
jpayne@69 1742 * and the status is set to U_AMBIGUOUS_ALIAS_WARNING.
jpayne@69 1743 * @param alias alias name
jpayne@69 1744 * @param n index in alias list
jpayne@69 1745 * @param pErrorCode result of operation
jpayne@69 1746 * @return returns the name of the alias at given index
jpayne@69 1747 * @see ucnv_countAliases
jpayne@69 1748 * @stable ICU 2.0
jpayne@69 1749 */
jpayne@69 1750 U_STABLE const char * U_EXPORT2
jpayne@69 1751 ucnv_getAlias(const char *alias, uint16_t n, UErrorCode *pErrorCode);
jpayne@69 1752
jpayne@69 1753 /**
jpayne@69 1754 * Fill-up the list of alias names for the given alias.
jpayne@69 1755 * This method only enumerates the listed entries in the alias file.
jpayne@69 1756 * If the alias is ambiguous, then the preferred converter is used
jpayne@69 1757 * and the status is set to U_AMBIGUOUS_ALIAS_WARNING.
jpayne@69 1758 * @param alias alias name
jpayne@69 1759 * @param aliases fill-in list, aliases is a pointer to an array of
jpayne@69 1760 * <code>ucnv_countAliases()</code> string-pointers
jpayne@69 1761 * (<code>const char *</code>) that will be filled in.
jpayne@69 1762 * The strings themselves are owned by the library.
jpayne@69 1763 * @param pErrorCode result of operation
jpayne@69 1764 * @stable ICU 2.0
jpayne@69 1765 */
jpayne@69 1766 U_STABLE void U_EXPORT2
jpayne@69 1767 ucnv_getAliases(const char *alias, const char **aliases, UErrorCode *pErrorCode);
jpayne@69 1768
jpayne@69 1769 /**
jpayne@69 1770 * Return a new UEnumeration object for enumerating all the
jpayne@69 1771 * alias names for a given converter that are recognized by a standard.
jpayne@69 1772 * This method only enumerates the listed entries in the alias file.
jpayne@69 1773 * The convrtrs.txt file can be modified to change the results of
jpayne@69 1774 * this function.
jpayne@69 1775 * The first result in this list is the same result given by
jpayne@69 1776 * <code>ucnv_getStandardName</code>, which is the default alias for
jpayne@69 1777 * the specified standard name. The returned object must be closed with
jpayne@69 1778 * <code>uenum_close</code> when you are done with the object.
jpayne@69 1779 *
jpayne@69 1780 * @param convName original converter name
jpayne@69 1781 * @param standard name of the standard governing the names; MIME and IANA
jpayne@69 1782 * are such standards
jpayne@69 1783 * @param pErrorCode The error code
jpayne@69 1784 * @return A UEnumeration object for getting all aliases that are recognized
jpayne@69 1785 * by a standard. If any of the parameters are invalid, NULL
jpayne@69 1786 * is returned.
jpayne@69 1787 * @see ucnv_getStandardName
jpayne@69 1788 * @see uenum_close
jpayne@69 1789 * @see uenum_next
jpayne@69 1790 * @stable ICU 2.2
jpayne@69 1791 */
jpayne@69 1792 U_STABLE UEnumeration * U_EXPORT2
jpayne@69 1793 ucnv_openStandardNames(const char *convName,
jpayne@69 1794 const char *standard,
jpayne@69 1795 UErrorCode *pErrorCode);
jpayne@69 1796
jpayne@69 1797 /**
jpayne@69 1798 * Gives the number of standards associated to converter names.
jpayne@69 1799 * @return number of standards
jpayne@69 1800 * @stable ICU 2.0
jpayne@69 1801 */
jpayne@69 1802 U_STABLE uint16_t U_EXPORT2
jpayne@69 1803 ucnv_countStandards(void);
jpayne@69 1804
jpayne@69 1805 /**
jpayne@69 1806 * Gives the name of the standard at given index of standard list.
jpayne@69 1807 * @param n index in standard list
jpayne@69 1808 * @param pErrorCode result of operation
jpayne@69 1809 * @return returns the name of the standard at given index. Owned by the library.
jpayne@69 1810 * @stable ICU 2.0
jpayne@69 1811 */
jpayne@69 1812 U_STABLE const char * U_EXPORT2
jpayne@69 1813 ucnv_getStandard(uint16_t n, UErrorCode *pErrorCode);
jpayne@69 1814
jpayne@69 1815 /**
jpayne@69 1816 * Returns a standard name for a given converter name.
jpayne@69 1817 * <p>
jpayne@69 1818 * Example alias table:<br>
jpayne@69 1819 * conv alias1 { STANDARD1 } alias2 { STANDARD1* }
jpayne@69 1820 * <p>
jpayne@69 1821 * Result of ucnv_getStandardName("conv", "STANDARD1") from example
jpayne@69 1822 * alias table:<br>
jpayne@69 1823 * <b>"alias2"</b>
jpayne@69 1824 *
jpayne@69 1825 * @param name original converter name
jpayne@69 1826 * @param standard name of the standard governing the names; MIME and IANA
jpayne@69 1827 * are such standards
jpayne@69 1828 * @param pErrorCode result of operation
jpayne@69 1829 * @return returns the standard converter name;
jpayne@69 1830 * if a standard converter name cannot be determined,
jpayne@69 1831 * then <code>NULL</code> is returned. Owned by the library.
jpayne@69 1832 * @stable ICU 2.0
jpayne@69 1833 */
jpayne@69 1834 U_STABLE const char * U_EXPORT2
jpayne@69 1835 ucnv_getStandardName(const char *name, const char *standard, UErrorCode *pErrorCode);
jpayne@69 1836
jpayne@69 1837 /**
jpayne@69 1838 * This function will return the internal canonical converter name of the
jpayne@69 1839 * tagged alias. This is the opposite of ucnv_openStandardNames, which
jpayne@69 1840 * returns the tagged alias given the canonical name.
jpayne@69 1841 * <p>
jpayne@69 1842 * Example alias table:<br>
jpayne@69 1843 * conv alias1 { STANDARD1 } alias2 { STANDARD1* }
jpayne@69 1844 * <p>
jpayne@69 1845 * Result of ucnv_getStandardName("alias1", "STANDARD1") from example
jpayne@69 1846 * alias table:<br>
jpayne@69 1847 * <b>"conv"</b>
jpayne@69 1848 *
jpayne@69 1849 * @return returns the canonical converter name;
jpayne@69 1850 * if a standard or alias name cannot be determined,
jpayne@69 1851 * then <code>NULL</code> is returned. The returned string is
jpayne@69 1852 * owned by the library.
jpayne@69 1853 * @see ucnv_getStandardName
jpayne@69 1854 * @stable ICU 2.4
jpayne@69 1855 */
jpayne@69 1856 U_STABLE const char * U_EXPORT2
jpayne@69 1857 ucnv_getCanonicalName(const char *alias, const char *standard, UErrorCode *pErrorCode);
jpayne@69 1858
jpayne@69 1859 /**
jpayne@69 1860 * Returns the current default converter name. If you want to open
jpayne@69 1861 * a default converter, you do not need to use this function.
jpayne@69 1862 * It is faster if you pass a NULL argument to ucnv_open the
jpayne@69 1863 * default converter.
jpayne@69 1864 *
jpayne@69 1865 * If U_CHARSET_IS_UTF8 is defined to 1 in utypes.h then this function
jpayne@69 1866 * always returns "UTF-8".
jpayne@69 1867 *
jpayne@69 1868 * @return returns the current default converter name.
jpayne@69 1869 * Storage owned by the library
jpayne@69 1870 * @see ucnv_setDefaultName
jpayne@69 1871 * @stable ICU 2.0
jpayne@69 1872 */
jpayne@69 1873 U_STABLE const char * U_EXPORT2
jpayne@69 1874 ucnv_getDefaultName(void);
jpayne@69 1875
jpayne@69 1876 #ifndef U_HIDE_SYSTEM_API
jpayne@69 1877 /**
jpayne@69 1878 * This function is not thread safe. DO NOT call this function when ANY ICU
jpayne@69 1879 * function is being used from more than one thread! This function sets the
jpayne@69 1880 * current default converter name. If this function needs to be called, it
jpayne@69 1881 * should be called during application initialization. Most of the time, the
jpayne@69 1882 * results from ucnv_getDefaultName() or ucnv_open with a NULL string argument
jpayne@69 1883 * is sufficient for your application.
jpayne@69 1884 *
jpayne@69 1885 * If U_CHARSET_IS_UTF8 is defined to 1 in utypes.h then this function
jpayne@69 1886 * does nothing.
jpayne@69 1887 *
jpayne@69 1888 * @param name the converter name to be the default (must be known by ICU).
jpayne@69 1889 * @see ucnv_getDefaultName
jpayne@69 1890 * @system
jpayne@69 1891 * @stable ICU 2.0
jpayne@69 1892 */
jpayne@69 1893 U_STABLE void U_EXPORT2
jpayne@69 1894 ucnv_setDefaultName(const char *name);
jpayne@69 1895 #endif /* U_HIDE_SYSTEM_API */
jpayne@69 1896
jpayne@69 1897 /**
jpayne@69 1898 * Fixes the backslash character mismapping. For example, in SJIS, the backslash
jpayne@69 1899 * character in the ASCII portion is also used to represent the yen currency sign.
jpayne@69 1900 * When mapping from Unicode character 0x005C, it's unclear whether to map the
jpayne@69 1901 * character back to yen or backslash in SJIS. This function will take the input
jpayne@69 1902 * buffer and replace all the yen sign characters with backslash. This is necessary
jpayne@69 1903 * when the user tries to open a file with the input buffer on Windows.
jpayne@69 1904 * This function will test the converter to see whether such mapping is
jpayne@69 1905 * required. You can sometimes avoid using this function by using the correct version
jpayne@69 1906 * of Shift-JIS.
jpayne@69 1907 *
jpayne@69 1908 * @param cnv The converter representing the target codepage.
jpayne@69 1909 * @param source the input buffer to be fixed
jpayne@69 1910 * @param sourceLen the length of the input buffer
jpayne@69 1911 * @see ucnv_isAmbiguous
jpayne@69 1912 * @stable ICU 2.0
jpayne@69 1913 */
jpayne@69 1914 U_STABLE void U_EXPORT2
jpayne@69 1915 ucnv_fixFileSeparator(const UConverter *cnv, UChar *source, int32_t sourceLen);
jpayne@69 1916
jpayne@69 1917 /**
jpayne@69 1918 * Determines if the converter contains ambiguous mappings of the same
jpayne@69 1919 * character or not.
jpayne@69 1920 * @param cnv the converter to be tested
jpayne@69 1921 * @return TRUE if the converter contains ambiguous mapping of the same
jpayne@69 1922 * character, FALSE otherwise.
jpayne@69 1923 * @stable ICU 2.0
jpayne@69 1924 */
jpayne@69 1925 U_STABLE UBool U_EXPORT2
jpayne@69 1926 ucnv_isAmbiguous(const UConverter *cnv);
jpayne@69 1927
jpayne@69 1928 /**
jpayne@69 1929 * Sets the converter to use fallback mappings or not.
jpayne@69 1930 * Regardless of this flag, the converter will always use
jpayne@69 1931 * fallbacks from Unicode Private Use code points, as well as
jpayne@69 1932 * reverse fallbacks (to Unicode).
jpayne@69 1933 * For details see ".ucm File Format"
jpayne@69 1934 * in the Conversion Data chapter of the ICU User Guide:
jpayne@69 1935 * http://www.icu-project.org/userguide/conversion-data.html#ucmformat
jpayne@69 1936 *
jpayne@69 1937 * @param cnv The converter to set the fallback mapping usage on.
jpayne@69 1938 * @param usesFallback TRUE if the user wants the converter to take advantage of the fallback
jpayne@69 1939 * mapping, FALSE otherwise.
jpayne@69 1940 * @stable ICU 2.0
jpayne@69 1941 * @see ucnv_usesFallback
jpayne@69 1942 */
jpayne@69 1943 U_STABLE void U_EXPORT2
jpayne@69 1944 ucnv_setFallback(UConverter *cnv, UBool usesFallback);
jpayne@69 1945
jpayne@69 1946 /**
jpayne@69 1947 * Determines if the converter uses fallback mappings or not.
jpayne@69 1948 * This flag has restrictions, see ucnv_setFallback().
jpayne@69 1949 *
jpayne@69 1950 * @param cnv The converter to be tested
jpayne@69 1951 * @return TRUE if the converter uses fallback, FALSE otherwise.
jpayne@69 1952 * @stable ICU 2.0
jpayne@69 1953 * @see ucnv_setFallback
jpayne@69 1954 */
jpayne@69 1955 U_STABLE UBool U_EXPORT2
jpayne@69 1956 ucnv_usesFallback(const UConverter *cnv);
jpayne@69 1957
jpayne@69 1958 /**
jpayne@69 1959 * Detects Unicode signature byte sequences at the start of the byte stream
jpayne@69 1960 * and returns the charset name of the indicated Unicode charset.
jpayne@69 1961 * NULL is returned when no Unicode signature is recognized.
jpayne@69 1962 * The number of bytes in the signature is output as well.
jpayne@69 1963 *
jpayne@69 1964 * The caller can ucnv_open() a converter using the charset name.
jpayne@69 1965 * The first code unit (UChar) from the start of the stream will be U+FEFF
jpayne@69 1966 * (the Unicode BOM/signature character) and can usually be ignored.
jpayne@69 1967 *
jpayne@69 1968 * For most Unicode charsets it is also possible to ignore the indicated
jpayne@69 1969 * number of initial stream bytes and start converting after them.
jpayne@69 1970 * However, there are stateful Unicode charsets (UTF-7 and BOCU-1) for which
jpayne@69 1971 * this will not work. Therefore, it is best to ignore the first output UChar
jpayne@69 1972 * instead of the input signature bytes.
jpayne@69 1973 * <p>
jpayne@69 1974 * Usage:
jpayne@69 1975 * \snippet samples/ucnv/convsamp.cpp ucnv_detectUnicodeSignature
jpayne@69 1976 *
jpayne@69 1977 * @param source The source string in which the signature should be detected.
jpayne@69 1978 * @param sourceLength Length of the input string, or -1 if terminated with a NUL byte.
jpayne@69 1979 * @param signatureLength A pointer to int32_t to receive the number of bytes that make up the signature
jpayne@69 1980 * of the detected UTF. 0 if not detected.
jpayne@69 1981 * Can be a NULL pointer.
jpayne@69 1982 * @param pErrorCode ICU error code in/out parameter.
jpayne@69 1983 * Must fulfill U_SUCCESS before the function call.
jpayne@69 1984 * @return The name of the encoding detected. NULL if encoding is not detected.
jpayne@69 1985 * @stable ICU 2.4
jpayne@69 1986 */
jpayne@69 1987 U_STABLE const char* U_EXPORT2
jpayne@69 1988 ucnv_detectUnicodeSignature(const char* source,
jpayne@69 1989 int32_t sourceLength,
jpayne@69 1990 int32_t *signatureLength,
jpayne@69 1991 UErrorCode *pErrorCode);
jpayne@69 1992
jpayne@69 1993 /**
jpayne@69 1994 * Returns the number of UChars held in the converter's internal state
jpayne@69 1995 * because more input is needed for completing the conversion. This function is
jpayne@69 1996 * useful for mapping semantics of ICU's converter interface to those of iconv,
jpayne@69 1997 * and this information is not needed for normal conversion.
jpayne@69 1998 * @param cnv The converter in which the input is held
jpayne@69 1999 * @param status ICU error code in/out parameter.
jpayne@69 2000 * Must fulfill U_SUCCESS before the function call.
jpayne@69 2001 * @return The number of UChars in the state. -1 if an error is encountered.
jpayne@69 2002 * @stable ICU 3.4
jpayne@69 2003 */
jpayne@69 2004 U_STABLE int32_t U_EXPORT2
jpayne@69 2005 ucnv_fromUCountPending(const UConverter* cnv, UErrorCode* status);
jpayne@69 2006
jpayne@69 2007 /**
jpayne@69 2008 * Returns the number of chars held in the converter's internal state
jpayne@69 2009 * because more input is needed for completing the conversion. This function is
jpayne@69 2010 * useful for mapping semantics of ICU's converter interface to those of iconv,
jpayne@69 2011 * and this information is not needed for normal conversion.
jpayne@69 2012 * @param cnv The converter in which the input is held as internal state
jpayne@69 2013 * @param status ICU error code in/out parameter.
jpayne@69 2014 * Must fulfill U_SUCCESS before the function call.
jpayne@69 2015 * @return The number of chars in the state. -1 if an error is encountered.
jpayne@69 2016 * @stable ICU 3.4
jpayne@69 2017 */
jpayne@69 2018 U_STABLE int32_t U_EXPORT2
jpayne@69 2019 ucnv_toUCountPending(const UConverter* cnv, UErrorCode* status);
jpayne@69 2020
jpayne@69 2021 /**
jpayne@69 2022 * Returns whether or not the charset of the converter has a fixed number of bytes
jpayne@69 2023 * per charset character.
jpayne@69 2024 * An example of this are converters that are of the type UCNV_SBCS or UCNV_DBCS.
jpayne@69 2025 * Another example is UTF-32 which is always 4 bytes per character.
jpayne@69 2026 * A Unicode code point may be represented by more than one UTF-8 or UTF-16 code unit
jpayne@69 2027 * but a UTF-32 converter encodes each code point with 4 bytes.
jpayne@69 2028 * Note: This method is not intended to be used to determine whether the charset has a
jpayne@69 2029 * fixed ratio of bytes to Unicode codes <i>units</i> for any particular Unicode encoding form.
jpayne@69 2030 * FALSE is returned with the UErrorCode if error occurs or cnv is NULL.
jpayne@69 2031 * @param cnv The converter to be tested
jpayne@69 2032 * @param status ICU error code in/out paramter
jpayne@69 2033 * @return TRUE if the converter is fixed-width
jpayne@69 2034 * @stable ICU 4.8
jpayne@69 2035 */
jpayne@69 2036 U_STABLE UBool U_EXPORT2
jpayne@69 2037 ucnv_isFixedWidth(UConverter *cnv, UErrorCode *status);
jpayne@69 2038
jpayne@69 2039 #endif
jpayne@69 2040
jpayne@69 2041 #endif
jpayne@69 2042 /*_UCNV*/