1 // © 2016 and later: Unicode, Inc. and others. 2 // License & terms of use: http://www.unicode.org/copyright.html 3 /* 4 ********************************************************************** 5 * Copyright (C) 1999-2014, International Business Machines 6 * Corporation and others. All Rights Reserved. 7 ********************************************************************** 8 * ucnv.h: 9 * External APIs for the ICU's codeset conversion library 10 * Bertrand A. Damiba 11 * 12 * Modification History: 13 * 14 * Date Name Description 15 * 04/04/99 helena Fixed internal header inclusion. 16 * 05/11/00 helena Added setFallback and usesFallback APIs. 17 * 06/29/2000 helena Major rewrite of the callback APIs. 18 * 12/07/2000 srl Update of documentation 19 */ 20 21 /** 22 * \file 23 * \brief C API: Character conversion 24 * 25 * <h2>Character Conversion C API</h2> 26 * 27 * <p>This API is used to convert codepage or character encoded data to and 28 * from UTF-16. You can open a converter with {@link ucnv_open() }. With that 29 * converter, you can get its properties, set options, convert your data and 30 * close the converter.</p> 31 * 32 * <p>Since many software programs recognize different converter names for 33 * different types of converters, there are other functions in this API to 34 * iterate over the converter aliases. The functions {@link ucnv_getAvailableName() }, 35 * {@link ucnv_getAlias() } and {@link ucnv_getStandardName() } are some of the 36 * more frequently used alias functions to get this information.</p> 37 * 38 * <p>When a converter encounters an illegal, irregular, invalid or unmappable character 39 * its default behavior is to use a substitution character to replace the 40 * bad byte sequence. This behavior can be changed by using {@link ucnv_setFromUCallBack() } 41 * or {@link ucnv_setToUCallBack() } on the converter. The header ucnv_err.h defines 42 * many other callback actions that can be used instead of a character substitution.</p> 43 * 44 * <p>More information about this API can be found in our 45 * <a href="http://icu-project.org/userguide/conversion.html">User's 46 * Guide</a>.</p> 47 */ 48 49 #ifndef UCNV_H 50 #define UCNV_H 51 52 #include "unicode/ucnv_err.h" 53 #include "unicode/uenum.h" 54 55 #if U_SHOW_CPLUSPLUS_API 56 #include "unicode/localpointer.h" 57 #endif // U_SHOW_CPLUSPLUS_API 58 59 #if !defined(USET_DEFINED) && !defined(U_IN_DOXYGEN) 60 61 #define USET_DEFINED 62 63 /** 64 * USet is the C API type corresponding to C++ class UnicodeSet. 65 * It is forward-declared here to avoid including unicode/uset.h file if related 66 * conversion APIs are not used. 67 * 68 * @see ucnv_getUnicodeSet 69 * @stable ICU 2.4 70 */ 71 typedef struct USet USet; 72 73 #endif 74 75 #if !UCONFIG_NO_CONVERSION 76 77 U_CDECL_BEGIN 78 79 /** Maximum length of a converter name including the terminating NULL @stable ICU 2.0 */ 80 #define UCNV_MAX_CONVERTER_NAME_LENGTH 60 81 /** Maximum length of a converter name including path and terminating NULL @stable ICU 2.0 */ 82 #define UCNV_MAX_FULL_FILE_NAME_LENGTH (600+UCNV_MAX_CONVERTER_NAME_LENGTH) 83 84 /** Shift in for EBDCDIC_STATEFUL and iso2022 states @stable ICU 2.0 */ 85 #define UCNV_SI 0x0F 86 /** Shift out for EBDCDIC_STATEFUL and iso2022 states @stable ICU 2.0 */ 87 #define UCNV_SO 0x0E 88 89 /** 90 * Enum for specifying basic types of converters 91 * @see ucnv_getType 92 * @stable ICU 2.0 93 */ 94 typedef enum { 95 /** @stable ICU 2.0 */ 96 UCNV_UNSUPPORTED_CONVERTER = -1, 97 /** @stable ICU 2.0 */ 98 UCNV_SBCS = 0, 99 /** @stable ICU 2.0 */ 100 UCNV_DBCS = 1, 101 /** @stable ICU 2.0 */ 102 UCNV_MBCS = 2, 103 /** @stable ICU 2.0 */ 104 UCNV_LATIN_1 = 3, 105 /** @stable ICU 2.0 */ 106 UCNV_UTF8 = 4, 107 /** @stable ICU 2.0 */ 108 UCNV_UTF16_BigEndian = 5, 109 /** @stable ICU 2.0 */ 110 UCNV_UTF16_LittleEndian = 6, 111 /** @stable ICU 2.0 */ 112 UCNV_UTF32_BigEndian = 7, 113 /** @stable ICU 2.0 */ 114 UCNV_UTF32_LittleEndian = 8, 115 /** @stable ICU 2.0 */ 116 UCNV_EBCDIC_STATEFUL = 9, 117 /** @stable ICU 2.0 */ 118 UCNV_ISO_2022 = 10, 119 120 /** @stable ICU 2.0 */ 121 UCNV_LMBCS_1 = 11, 122 /** @stable ICU 2.0 */ 123 UCNV_LMBCS_2, 124 /** @stable ICU 2.0 */ 125 UCNV_LMBCS_3, 126 /** @stable ICU 2.0 */ 127 UCNV_LMBCS_4, 128 /** @stable ICU 2.0 */ 129 UCNV_LMBCS_5, 130 /** @stable ICU 2.0 */ 131 UCNV_LMBCS_6, 132 /** @stable ICU 2.0 */ 133 UCNV_LMBCS_8, 134 /** @stable ICU 2.0 */ 135 UCNV_LMBCS_11, 136 /** @stable ICU 2.0 */ 137 UCNV_LMBCS_16, 138 /** @stable ICU 2.0 */ 139 UCNV_LMBCS_17, 140 /** @stable ICU 2.0 */ 141 UCNV_LMBCS_18, 142 /** @stable ICU 2.0 */ 143 UCNV_LMBCS_19, 144 /** @stable ICU 2.0 */ 145 UCNV_LMBCS_LAST = UCNV_LMBCS_19, 146 /** @stable ICU 2.0 */ 147 UCNV_HZ, 148 /** @stable ICU 2.0 */ 149 UCNV_SCSU, 150 /** @stable ICU 2.0 */ 151 UCNV_ISCII, 152 /** @stable ICU 2.0 */ 153 UCNV_US_ASCII, 154 /** @stable ICU 2.0 */ 155 UCNV_UTF7, 156 /** @stable ICU 2.2 */ 157 UCNV_BOCU1, 158 /** @stable ICU 2.2 */ 159 UCNV_UTF16, 160 /** @stable ICU 2.2 */ 161 UCNV_UTF32, 162 /** @stable ICU 2.2 */ 163 UCNV_CESU8, 164 /** @stable ICU 2.4 */ 165 UCNV_IMAP_MAILBOX, 166 /** @stable ICU 4.8 */ 167 UCNV_COMPOUND_TEXT, 168 169 /* Number of converter types for which we have conversion routines. */ 170 UCNV_NUMBER_OF_SUPPORTED_CONVERTER_TYPES 171 } UConverterType; 172 173 /** 174 * Enum for specifying which platform a converter ID refers to. 175 * The use of platform/CCSID is not recommended. See ucnv_openCCSID(). 176 * 177 * @see ucnv_getPlatform 178 * @see ucnv_openCCSID 179 * @see ucnv_getCCSID 180 * @stable ICU 2.0 181 */ 182 typedef enum { 183 UCNV_UNKNOWN = -1, 184 UCNV_IBM = 0 185 } UConverterPlatform; 186 187 /** 188 * Function pointer for error callback in the codepage to unicode direction. 189 * Called when an error has occurred in conversion to unicode, or on open/close of the callback (see reason). 190 * @param context Pointer to the callback's private data 191 * @param args Information about the conversion in progress 192 * @param codeUnits Points to 'length' bytes of the concerned codepage sequence 193 * @param length Size (in bytes) of the concerned codepage sequence 194 * @param reason Defines the reason the callback was invoked 195 * @param pErrorCode ICU error code in/out parameter. 196 * For converter callback functions, set to a conversion error 197 * before the call, and the callback may reset it to U_ZERO_ERROR. 198 * @see ucnv_setToUCallBack 199 * @see UConverterToUnicodeArgs 200 * @stable ICU 2.0 201 */ 202 typedef void (U_EXPORT2 *UConverterToUCallback) ( 203 const void* context, 204 UConverterToUnicodeArgs *args, 205 const char *codeUnits, 206 int32_t length, 207 UConverterCallbackReason reason, 208 UErrorCode *pErrorCode); 209 210 /** 211 * Function pointer for error callback in the unicode to codepage direction. 212 * Called when an error has occurred in conversion from unicode, or on open/close of the callback (see reason). 213 * @param context Pointer to the callback's private data 214 * @param args Information about the conversion in progress 215 * @param codeUnits Points to 'length' UChars of the concerned Unicode sequence 216 * @param length Size (in bytes) of the concerned codepage sequence 217 * @param codePoint Single UChar32 (UTF-32) containing the concerend Unicode codepoint. 218 * @param reason Defines the reason the callback was invoked 219 * @param pErrorCode ICU error code in/out parameter. 220 * For converter callback functions, set to a conversion error 221 * before the call, and the callback may reset it to U_ZERO_ERROR. 222 * @see ucnv_setFromUCallBack 223 * @stable ICU 2.0 224 */ 225 typedef void (U_EXPORT2 *UConverterFromUCallback) ( 226 const void* context, 227 UConverterFromUnicodeArgs *args, 228 const UChar* codeUnits, 229 int32_t length, 230 UChar32 codePoint, 231 UConverterCallbackReason reason, 232 UErrorCode *pErrorCode); 233 234 U_CDECL_END 235 236 /** 237 * Character that separates converter names from options and options from each other. 238 * @see ucnv_open 239 * @stable ICU 2.0 240 */ 241 #define UCNV_OPTION_SEP_CHAR ',' 242 243 /** 244 * String version of UCNV_OPTION_SEP_CHAR. 245 * @see ucnv_open 246 * @stable ICU 2.0 247 */ 248 #define UCNV_OPTION_SEP_STRING "," 249 250 /** 251 * Character that separates a converter option from its value. 252 * @see ucnv_open 253 * @stable ICU 2.0 254 */ 255 #define UCNV_VALUE_SEP_CHAR '=' 256 257 /** 258 * String version of UCNV_VALUE_SEP_CHAR. 259 * @see ucnv_open 260 * @stable ICU 2.0 261 */ 262 #define UCNV_VALUE_SEP_STRING "=" 263 264 /** 265 * Converter option for specifying a locale. 266 * For example, ucnv_open("SCSU,locale=ja", &errorCode); 267 * See convrtrs.txt. 268 * 269 * @see ucnv_open 270 * @stable ICU 2.0 271 */ 272 #define UCNV_LOCALE_OPTION_STRING ",locale=" 273 274 /** 275 * Converter option for specifying a version selector (0..9) for some converters. 276 * For example, 277 * \code 278 * ucnv_open("UTF-7,version=1", &errorCode); 279 * \endcode 280 * See convrtrs.txt. 281 * 282 * @see ucnv_open 283 * @stable ICU 2.4 284 */ 285 #define UCNV_VERSION_OPTION_STRING ",version=" 286 287 /** 288 * Converter option for EBCDIC SBCS or mixed-SBCS/DBCS (stateful) codepages. 289 * Swaps Unicode mappings for EBCDIC LF and NL codes, as used on 290 * S/390 (z/OS) Unix System Services (Open Edition). 291 * For example, ucnv_open("ibm-1047,swaplfnl", &errorCode); 292 * See convrtrs.txt. 293 * 294 * @see ucnv_open 295 * @stable ICU 2.4 296 */ 297 #define UCNV_SWAP_LFNL_OPTION_STRING ",swaplfnl" 298 299 /** 300 * Do a fuzzy compare of two converter/alias names. 301 * The comparison is case-insensitive, ignores leading zeroes if they are not 302 * followed by further digits, and ignores all but letters and digits. 303 * Thus the strings "UTF-8", "utf_8", "u*T@f08" and "Utf 8" are exactly equivalent. 304 * See section 1.4, Charset Alias Matching in Unicode Technical Standard #22 305 * at http://www.unicode.org/reports/tr22/ 306 * 307 * @param name1 a converter name or alias, zero-terminated 308 * @param name2 a converter name or alias, zero-terminated 309 * @return 0 if the names match, or a negative value if the name1 310 * lexically precedes name2, or a positive value if the name1 311 * lexically follows name2. 312 * @stable ICU 2.0 313 */ 314 U_CAPI int U_EXPORT2 315 ucnv_compareNames(const char *name1, const char *name2); 316 317 318 /** 319 * Creates a UConverter object with the name of a coded character set specified as a C string. 320 * The actual name will be resolved with the alias file 321 * using a case-insensitive string comparison that ignores 322 * leading zeroes and all non-alphanumeric characters. 323 * E.g., the names "UTF8", "utf-8", "u*T@f08" and "Utf 8" are all equivalent. 324 * (See also ucnv_compareNames().) 325 * If <code>NULL</code> is passed for the converter name, it will create one with the 326 * getDefaultName return value. 327 * 328 * <p>A converter name for ICU 1.5 and above may contain options 329 * like a locale specification to control the specific behavior of 330 * the newly instantiated converter. 331 * The meaning of the options depends on the particular converter. 332 * If an option is not defined for or recognized by a given converter, then it is ignored.</p> 333 * 334 * <p>Options are appended to the converter name string, with a 335 * <code>UCNV_OPTION_SEP_CHAR</code> between the name and the first option and 336 * also between adjacent options.</p> 337 * 338 * <p>If the alias is ambiguous, then the preferred converter is used 339 * and the status is set to U_AMBIGUOUS_ALIAS_WARNING.</p> 340 * 341 * <p>The conversion behavior and names can vary between platforms. ICU may 342 * convert some characters differently from other platforms. Details on this topic 343 * are in the <a href="http://icu-project.org/userguide/conversion.html">User's 344 * Guide</a>. Aliases starting with a "cp" prefix have no specific meaning 345 * other than its an alias starting with the letters "cp". Please do not 346 * associate any meaning to these aliases.</p> 347 * 348 * \snippet samples/ucnv/convsamp.cpp ucnv_open 349 * 350 * @param converterName Name of the coded character set table. 351 * This may have options appended to the string. 352 * IANA alias character set names, IBM CCSIDs starting with "ibm-", 353 * Windows codepage numbers starting with "windows-" are frequently 354 * used for this parameter. See ucnv_getAvailableName and 355 * ucnv_getAlias for a complete list that is available. 356 * If this parameter is NULL, the default converter will be used. 357 * @param err outgoing error status <TT>U_MEMORY_ALLOCATION_ERROR, U_FILE_ACCESS_ERROR</TT> 358 * @return the created Unicode converter object, or <TT>NULL</TT> if an error occurred 359 * @see ucnv_openU 360 * @see ucnv_openCCSID 361 * @see ucnv_getAvailableName 362 * @see ucnv_getAlias 363 * @see ucnv_getDefaultName 364 * @see ucnv_close 365 * @see ucnv_compareNames 366 * @stable ICU 2.0 367 */ 368 U_CAPI UConverter* U_EXPORT2 369 ucnv_open(const char *converterName, UErrorCode *err); 370 371 372 /** 373 * Creates a Unicode converter with the names specified as unicode string. 374 * The name should be limited to the ASCII-7 alphanumerics range. 375 * The actual name will be resolved with the alias file 376 * using a case-insensitive string comparison that ignores 377 * leading zeroes and all non-alphanumeric characters. 378 * E.g., the names "UTF8", "utf-8", "u*T@f08" and "Utf 8" are all equivalent. 379 * (See also ucnv_compareNames().) 380 * If <TT>NULL</TT> is passed for the converter name, it will create 381 * one with the ucnv_getDefaultName() return value. 382 * If the alias is ambiguous, then the preferred converter is used 383 * and the status is set to U_AMBIGUOUS_ALIAS_WARNING. 384 * 385 * <p>See ucnv_open for the complete details</p> 386 * @param name Name of the UConverter table in a zero terminated 387 * Unicode string 388 * @param err outgoing error status <TT>U_MEMORY_ALLOCATION_ERROR, 389 * U_FILE_ACCESS_ERROR</TT> 390 * @return the created Unicode converter object, or <TT>NULL</TT> if an 391 * error occurred 392 * @see ucnv_open 393 * @see ucnv_openCCSID 394 * @see ucnv_close 395 * @see ucnv_compareNames 396 * @stable ICU 2.0 397 */ 398 U_CAPI UConverter* U_EXPORT2 399 ucnv_openU(const UChar *name, 400 UErrorCode *err); 401 402 /** 403 * Creates a UConverter object from a CCSID number and platform pair. 404 * Note that the usefulness of this function is limited to platforms with numeric 405 * encoding IDs. Only IBM and Microsoft platforms use numeric (16-bit) identifiers for 406 * encodings. 407 * 408 * In addition, IBM CCSIDs and Unicode conversion tables are not 1:1 related. 409 * For many IBM CCSIDs there are multiple (up to six) Unicode conversion tables, and 410 * for some Unicode conversion tables there are multiple CCSIDs. 411 * Some "alternate" Unicode conversion tables are provided by the 412 * IBM CDRA conversion table registry. 413 * The most prominent example of a systematic modification of conversion tables that is 414 * not provided in the form of conversion table files in the repository is 415 * that S/390 Unix System Services swaps the codes for Line Feed and New Line in all 416 * EBCDIC codepages, which requires such a swap in the Unicode conversion tables as well. 417 * 418 * Only IBM default conversion tables are accessible with ucnv_openCCSID(). 419 * ucnv_getCCSID() will return the same CCSID for all conversion tables that are associated 420 * with that CCSID. 421 * 422 * Currently, the only "platform" supported in the ICU converter API is UCNV_IBM. 423 * 424 * In summary, the use of CCSIDs and the associated API functions is not recommended. 425 * 426 * In order to open a converter with the default IBM CDRA Unicode conversion table, 427 * you can use this function or use the prefix "ibm-": 428 * \code 429 * char name[20]; 430 * sprintf(name, "ibm-%hu", ccsid); 431 * cnv=ucnv_open(name, &errorCode); 432 * \endcode 433 * 434 * In order to open a converter with the IBM S/390 Unix System Services variant 435 * of a Unicode/EBCDIC conversion table, 436 * you can use the prefix "ibm-" together with the option string UCNV_SWAP_LFNL_OPTION_STRING: 437 * \code 438 * char name[20]; 439 * sprintf(name, "ibm-%hu" UCNV_SWAP_LFNL_OPTION_STRING, ccsid); 440 * cnv=ucnv_open(name, &errorCode); 441 * \endcode 442 * 443 * In order to open a converter from a Microsoft codepage number, use the prefix "cp": 444 * \code 445 * char name[20]; 446 * sprintf(name, "cp%hu", codepageID); 447 * cnv=ucnv_open(name, &errorCode); 448 * \endcode 449 * 450 * If the alias is ambiguous, then the preferred converter is used 451 * and the status is set to U_AMBIGUOUS_ALIAS_WARNING. 452 * 453 * @param codepage codepage number to create 454 * @param platform the platform in which the codepage number exists 455 * @param err error status <TT>U_MEMORY_ALLOCATION_ERROR, U_FILE_ACCESS_ERROR</TT> 456 * @return the created Unicode converter object, or <TT>NULL</TT> if an error 457 * occurred. 458 * @see ucnv_open 459 * @see ucnv_openU 460 * @see ucnv_close 461 * @see ucnv_getCCSID 462 * @see ucnv_getPlatform 463 * @see UConverterPlatform 464 * @stable ICU 2.0 465 */ 466 U_CAPI UConverter* U_EXPORT2 467 ucnv_openCCSID(int32_t codepage, 468 UConverterPlatform platform, 469 UErrorCode * err); 470 471 /** 472 * <p>Creates a UConverter object specified from a packageName and a converterName.</p> 473 * 474 * <p>The packageName and converterName must point to an ICU udata object, as defined by 475 * <code> udata_open( packageName, "cnv", converterName, err) </code> or equivalent. 476 * Typically, packageName will refer to a (.dat) file, or to a package registered with 477 * udata_setAppData(). Using a full file or directory pathname for packageName is deprecated.</p> 478 * 479 * <p>The name will NOT be looked up in the alias mechanism, nor will the converter be 480 * stored in the converter cache or the alias table. The only way to open further converters 481 * is call this function multiple times, or use the ucnv_safeClone() function to clone a 482 * 'primary' converter.</p> 483 * 484 * <p>A future version of ICU may add alias table lookups and/or caching 485 * to this function.</p> 486 * 487 * <p>Example Use: 488 * <code>cnv = ucnv_openPackage("myapp", "myconverter", &err);</code> 489 * </p> 490 * 491 * @param packageName name of the package (equivalent to 'path' in udata_open() call) 492 * @param converterName name of the data item to be used, without suffix. 493 * @param err outgoing error status <TT>U_MEMORY_ALLOCATION_ERROR, U_FILE_ACCESS_ERROR</TT> 494 * @return the created Unicode converter object, or <TT>NULL</TT> if an error occurred 495 * @see udata_open 496 * @see ucnv_open 497 * @see ucnv_safeClone 498 * @see ucnv_close 499 * @stable ICU 2.2 500 */ 501 U_CAPI UConverter* U_EXPORT2 502 ucnv_openPackage(const char *packageName, const char *converterName, UErrorCode *err); 503 504 /** 505 * Thread safe converter cloning operation. 506 * For most efficient operation, pass in a stackBuffer (and a *pBufferSize) 507 * with at least U_CNV_SAFECLONE_BUFFERSIZE bytes of space. 508 * If the buffer size is sufficient, then the clone will use the stack buffer; 509 * otherwise, it will be allocated, and *pBufferSize will indicate 510 * the actual size. (This should not occur with U_CNV_SAFECLONE_BUFFERSIZE.) 511 * 512 * You must ucnv_close() the clone in any case. 513 * 514 * If *pBufferSize==0, (regardless of whether stackBuffer==NULL or not) 515 * then *pBufferSize will be changed to a sufficient size 516 * for cloning this converter, 517 * without actually cloning the converter ("pure pre-flighting"). 518 * 519 * If *pBufferSize is greater than zero but not large enough for a stack-based 520 * clone, then the converter is cloned using newly allocated memory 521 * and *pBufferSize is changed to the necessary size. 522 * 523 * If the converter clone fits into the stack buffer but the stack buffer is not 524 * sufficiently aligned for the clone, then the clone will use an 525 * adjusted pointer and use an accordingly smaller buffer size. 526 * 527 * @param cnv converter to be cloned 528 * @param stackBuffer <em>Deprecated functionality as of ICU 52, use NULL.</em><br> 529 * user allocated space for the new clone. If NULL new memory will be allocated. 530 * If buffer is not large enough, new memory will be allocated. 531 * Clients can use the U_CNV_SAFECLONE_BUFFERSIZE. This will probably be enough to avoid memory allocations. 532 * @param pBufferSize <em>Deprecated functionality as of ICU 52, use NULL or 1.</em><br> 533 * pointer to size of allocated space. 534 * @param status to indicate whether the operation went on smoothly or there were errors 535 * An informational status value, U_SAFECLONE_ALLOCATED_WARNING, 536 * is used if any allocations were necessary. 537 * However, it is better to check if *pBufferSize grew for checking for 538 * allocations because warning codes can be overridden by subsequent 539 * function calls. 540 * @return pointer to the new clone 541 * @stable ICU 2.0 542 */ 543 U_CAPI UConverter * U_EXPORT2 544 ucnv_safeClone(const UConverter *cnv, 545 void *stackBuffer, 546 int32_t *pBufferSize, 547 UErrorCode *status); 548 549 #ifndef U_HIDE_DEPRECATED_API 550 551 /** 552 * \def U_CNV_SAFECLONE_BUFFERSIZE 553 * Definition of a buffer size that is designed to be large enough for 554 * converters to be cloned with ucnv_safeClone(). 555 * @deprecated ICU 52. Do not rely on ucnv_safeClone() cloning into any provided buffer. 556 */ 557 #define U_CNV_SAFECLONE_BUFFERSIZE 1024 558 559 #endif /* U_HIDE_DEPRECATED_API */ 560 561 /** 562 * Deletes the unicode converter and releases resources associated 563 * with just this instance. 564 * Does not free up shared converter tables. 565 * 566 * @param converter the converter object to be deleted 567 * @see ucnv_open 568 * @see ucnv_openU 569 * @see ucnv_openCCSID 570 * @stable ICU 2.0 571 */ 572 U_CAPI void U_EXPORT2 573 ucnv_close(UConverter * converter); 574 575 #if U_SHOW_CPLUSPLUS_API 576 577 U_NAMESPACE_BEGIN 578 579 /** 580 * \class LocalUConverterPointer 581 * "Smart pointer" class, closes a UConverter via ucnv_close(). 582 * For most methods see the LocalPointerBase base class. 583 * 584 * @see LocalPointerBase 585 * @see LocalPointer 586 * @stable ICU 4.4 587 */ 588 U_DEFINE_LOCAL_OPEN_POINTER(LocalUConverterPointer, UConverter, ucnv_close); 589 590 U_NAMESPACE_END 591 592 #endif 593 594 /** 595 * Fills in the output parameter, subChars, with the substitution characters 596 * as multiple bytes. 597 * If ucnv_setSubstString() set a Unicode string because the converter is 598 * stateful, then subChars will be an empty string. 599 * 600 * @param converter the Unicode converter 601 * @param subChars the substitution characters 602 * @param len on input the capacity of subChars, on output the number 603 * of bytes copied to it 604 * @param err the outgoing error status code. 605 * If the substitution character array is too small, an 606 * <TT>U_INDEX_OUTOFBOUNDS_ERROR</TT> will be returned. 607 * @see ucnv_setSubstString 608 * @see ucnv_setSubstChars 609 * @stable ICU 2.0 610 */ 611 U_CAPI void U_EXPORT2 612 ucnv_getSubstChars(const UConverter *converter, 613 char *subChars, 614 int8_t *len, 615 UErrorCode *err); 616 617 /** 618 * Sets the substitution chars when converting from unicode to a codepage. The 619 * substitution is specified as a string of 1-4 bytes, and may contain 620 * <TT>NULL</TT> bytes. 621 * The subChars must represent a single character. The caller needs to know the 622 * byte sequence of a valid character in the converter's charset. 623 * For some converters, for example some ISO 2022 variants, only single-byte 624 * substitution characters may be supported. 625 * The newer ucnv_setSubstString() function relaxes these limitations. 626 * 627 * @param converter the Unicode converter 628 * @param subChars the substitution character byte sequence we want set 629 * @param len the number of bytes in subChars 630 * @param err the error status code. <TT>U_INDEX_OUTOFBOUNDS_ERROR </TT> if 631 * len is bigger than the maximum number of bytes allowed in subchars 632 * @see ucnv_setSubstString 633 * @see ucnv_getSubstChars 634 * @stable ICU 2.0 635 */ 636 U_CAPI void U_EXPORT2 637 ucnv_setSubstChars(UConverter *converter, 638 const char *subChars, 639 int8_t len, 640 UErrorCode *err); 641 642 /** 643 * Set a substitution string for converting from Unicode to a charset. 644 * The caller need not know the charset byte sequence for each charset. 645 * 646 * Unlike ucnv_setSubstChars() which is designed to set a charset byte sequence 647 * for a single character, this function takes a Unicode string with 648 * zero, one or more characters, and immediately verifies that the string can be 649 * converted to the charset. 650 * If not, or if the result is too long (more than 32 bytes as of ICU 3.6), 651 * then the function returns with an error accordingly. 652 * 653 * Also unlike ucnv_setSubstChars(), this function works for stateful charsets 654 * by converting on the fly at the point of substitution rather than setting 655 * a fixed byte sequence. 656 * 657 * @param cnv The UConverter object. 658 * @param s The Unicode string. 659 * @param length The number of UChars in s, or -1 for a NUL-terminated string. 660 * @param err Pointer to a standard ICU error code. Its input value must 661 * pass the U_SUCCESS() test, or else the function returns 662 * immediately. Check for U_FAILURE() on output or use with 663 * function chaining. (See User Guide for details.) 664 * 665 * @see ucnv_setSubstChars 666 * @see ucnv_getSubstChars 667 * @stable ICU 3.6 668 */ 669 U_CAPI void U_EXPORT2 670 ucnv_setSubstString(UConverter *cnv, 671 const UChar *s, 672 int32_t length, 673 UErrorCode *err); 674 675 /** 676 * Fills in the output parameter, errBytes, with the error characters from the 677 * last failing conversion. 678 * 679 * @param converter the Unicode converter 680 * @param errBytes the codepage bytes which were in error 681 * @param len on input the capacity of errBytes, on output the number of 682 * bytes which were copied to it 683 * @param err the error status code. 684 * If the substitution character array is too small, an 685 * <TT>U_INDEX_OUTOFBOUNDS_ERROR</TT> will be returned. 686 * @stable ICU 2.0 687 */ 688 U_CAPI void U_EXPORT2 689 ucnv_getInvalidChars(const UConverter *converter, 690 char *errBytes, 691 int8_t *len, 692 UErrorCode *err); 693 694 /** 695 * Fills in the output parameter, errChars, with the error characters from the 696 * last failing conversion. 697 * 698 * @param converter the Unicode converter 699 * @param errUChars the UChars which were in error 700 * @param len on input the capacity of errUChars, on output the number of 701 * UChars which were copied to it 702 * @param err the error status code. 703 * If the substitution character array is too small, an 704 * <TT>U_INDEX_OUTOFBOUNDS_ERROR</TT> will be returned. 705 * @stable ICU 2.0 706 */ 707 U_CAPI void U_EXPORT2 708 ucnv_getInvalidUChars(const UConverter *converter, 709 UChar *errUChars, 710 int8_t *len, 711 UErrorCode *err); 712 713 /** 714 * Resets the state of a converter to the default state. This is used 715 * in the case of an error, to restart a conversion from a known default state. 716 * It will also empty the internal output buffers. 717 * @param converter the Unicode converter 718 * @stable ICU 2.0 719 */ 720 U_CAPI void U_EXPORT2 721 ucnv_reset(UConverter *converter); 722 723 /** 724 * Resets the to-Unicode part of a converter state to the default state. 725 * This is used in the case of an error to restart a conversion to 726 * Unicode to a known default state. It will also empty the internal 727 * output buffers used for the conversion to Unicode codepoints. 728 * @param converter the Unicode converter 729 * @stable ICU 2.0 730 */ 731 U_CAPI void U_EXPORT2 732 ucnv_resetToUnicode(UConverter *converter); 733 734 /** 735 * Resets the from-Unicode part of a converter state to the default state. 736 * This is used in the case of an error to restart a conversion from 737 * Unicode to a known default state. It will also empty the internal output 738 * buffers used for the conversion from Unicode codepoints. 739 * @param converter the Unicode converter 740 * @stable ICU 2.0 741 */ 742 U_CAPI void U_EXPORT2 743 ucnv_resetFromUnicode(UConverter *converter); 744 745 /** 746 * Returns the maximum number of bytes that are output per UChar in conversion 747 * from Unicode using this converter. 748 * The returned number can be used with UCNV_GET_MAX_BYTES_FOR_STRING 749 * to calculate the size of a target buffer for conversion from Unicode. 750 * 751 * Note: Before ICU 2.8, this function did not return reliable numbers for 752 * some stateful converters (EBCDIC_STATEFUL, ISO-2022) and LMBCS. 753 * 754 * This number may not be the same as the maximum number of bytes per 755 * "conversion unit". In other words, it may not be the intuitively expected 756 * number of bytes per character that would be published for a charset, 757 * and may not fulfill any other purpose than the allocation of an output 758 * buffer of guaranteed sufficient size for a given input length and converter. 759 * 760 * Examples for special cases that are taken into account: 761 * - Supplementary code points may convert to more bytes than BMP code points. 762 * This function returns bytes per UChar (UTF-16 code unit), not per 763 * Unicode code point, for efficient buffer allocation. 764 * - State-shifting output (SI/SO, escapes, etc.) from stateful converters. 765 * - When m input UChars are converted to n output bytes, then the maximum m/n 766 * is taken into account. 767 * 768 * The number returned here does not take into account 769 * (see UCNV_GET_MAX_BYTES_FOR_STRING): 770 * - callbacks which output more than one charset character sequence per call, 771 * like escape callbacks 772 * - initial and final non-character bytes that are output by some converters 773 * (automatic BOMs, initial escape sequence, final SI, etc.) 774 * 775 * Examples for returned values: 776 * - SBCS charsets: 1 777 * - Shift-JIS: 2 778 * - UTF-16: 2 (2 per BMP, 4 per surrogate _pair_, BOM not counted) 779 * - UTF-8: 3 (3 per BMP, 4 per surrogate _pair_) 780 * - EBCDIC_STATEFUL (EBCDIC mixed SBCS/DBCS): 3 (SO + DBCS) 781 * - ISO-2022: 3 (always outputs UTF-8) 782 * - ISO-2022-JP: 6 (4-byte escape sequences + DBCS) 783 * - ISO-2022-CN: 8 (4-byte designator sequences + 2-byte SS2/SS3 + DBCS) 784 * 785 * @param converter The Unicode converter. 786 * @return The maximum number of bytes per UChar (16 bit code unit) 787 * that are output by ucnv_fromUnicode(), 788 * to be used together with UCNV_GET_MAX_BYTES_FOR_STRING 789 * for buffer allocation. 790 * 791 * @see UCNV_GET_MAX_BYTES_FOR_STRING 792 * @see ucnv_getMinCharSize 793 * @stable ICU 2.0 794 */ 795 U_CAPI int8_t U_EXPORT2 796 ucnv_getMaxCharSize(const UConverter *converter); 797 798 /** 799 * Calculates the size of a buffer for conversion from Unicode to a charset. 800 * The calculated size is guaranteed to be sufficient for this conversion. 801 * 802 * It takes into account initial and final non-character bytes that are output 803 * by some converters. 804 * It does not take into account callbacks which output more than one charset 805 * character sequence per call, like escape callbacks. 806 * The default (substitution) callback only outputs one charset character sequence. 807 * 808 * @param length Number of UChars to be converted. 809 * @param maxCharSize Return value from ucnv_getMaxCharSize() for the converter 810 * that will be used. 811 * @return Size of a buffer that will be large enough to hold the output bytes of 812 * converting length UChars with the converter that returned the maxCharSize. 813 * 814 * @see ucnv_getMaxCharSize 815 * @stable ICU 2.8 816 */ 817 #define UCNV_GET_MAX_BYTES_FOR_STRING(length, maxCharSize) \ 818 (((int32_t)(length)+10)*(int32_t)(maxCharSize)) 819 820 /** 821 * Returns the minimum byte length (per codepoint) for characters in this codepage. 822 * This is usually either 1 or 2. 823 * @param converter the Unicode converter 824 * @return the minimum number of bytes per codepoint allowed by this particular converter 825 * @see ucnv_getMaxCharSize 826 * @stable ICU 2.0 827 */ 828 U_CAPI int8_t U_EXPORT2 829 ucnv_getMinCharSize(const UConverter *converter); 830 831 /** 832 * Returns the display name of the converter passed in based on the Locale 833 * passed in. If the locale contains no display name, the internal ASCII 834 * name will be filled in. 835 * 836 * @param converter the Unicode converter. 837 * @param displayLocale is the specific Locale we want to localized for 838 * @param displayName user provided buffer to be filled in 839 * @param displayNameCapacity size of displayName Buffer 840 * @param err error status code 841 * @return displayNameLength number of UChar needed in displayName 842 * @see ucnv_getName 843 * @stable ICU 2.0 844 */ 845 U_CAPI int32_t U_EXPORT2 846 ucnv_getDisplayName(const UConverter *converter, 847 const char *displayLocale, 848 UChar *displayName, 849 int32_t displayNameCapacity, 850 UErrorCode *err); 851 852 /** 853 * Gets the internal, canonical name of the converter (zero-terminated). 854 * The lifetime of the returned string will be that of the converter 855 * passed to this function. 856 * @param converter the Unicode converter 857 * @param err UErrorCode status 858 * @return the internal name of the converter 859 * @see ucnv_getDisplayName 860 * @stable ICU 2.0 861 */ 862 U_CAPI const char * U_EXPORT2 863 ucnv_getName(const UConverter *converter, UErrorCode *err); 864 865 /** 866 * Gets a codepage number associated with the converter. This is not guaranteed 867 * to be the one used to create the converter. Some converters do not represent 868 * platform registered codepages and return zero for the codepage number. 869 * The error code fill-in parameter indicates if the codepage number 870 * is available. 871 * Does not check if the converter is <TT>NULL</TT> or if converter's data 872 * table is <TT>NULL</TT>. 873 * 874 * Important: The use of CCSIDs is not recommended because it is limited 875 * to only two platforms in principle and only one (UCNV_IBM) in the current 876 * ICU converter API. 877 * Also, CCSIDs are insufficient to identify IBM Unicode conversion tables precisely. 878 * For more details see ucnv_openCCSID(). 879 * 880 * @param converter the Unicode converter 881 * @param err the error status code. 882 * @return If any error occurs, -1 will be returned otherwise, the codepage number 883 * will be returned 884 * @see ucnv_openCCSID 885 * @see ucnv_getPlatform 886 * @stable ICU 2.0 887 */ 888 U_CAPI int32_t U_EXPORT2 889 ucnv_getCCSID(const UConverter *converter, 890 UErrorCode *err); 891 892 /** 893 * Gets a codepage platform associated with the converter. Currently, 894 * only <TT>UCNV_IBM</TT> will be returned. 895 * Does not test if the converter is <TT>NULL</TT> or if converter's data 896 * table is <TT>NULL</TT>. 897 * @param converter the Unicode converter 898 * @param err the error status code. 899 * @return The codepage platform 900 * @stable ICU 2.0 901 */ 902 U_CAPI UConverterPlatform U_EXPORT2 903 ucnv_getPlatform(const UConverter *converter, 904 UErrorCode *err); 905 906 /** 907 * Gets the type of the converter 908 * e.g. SBCS, MBCS, DBCS, UTF8, UTF16_BE, UTF16_LE, ISO_2022, 909 * EBCDIC_STATEFUL, LATIN_1 910 * @param converter a valid, opened converter 911 * @return the type of the converter 912 * @stable ICU 2.0 913 */ 914 U_CAPI UConverterType U_EXPORT2 915 ucnv_getType(const UConverter * converter); 916 917 /** 918 * Gets the "starter" (lead) bytes for converters of type MBCS. 919 * Will fill in an <TT>U_ILLEGAL_ARGUMENT_ERROR</TT> if converter passed in 920 * is not MBCS. Fills in an array of type UBool, with the value of the byte 921 * as offset to the array. For example, if (starters[0x20] == true) at return, 922 * it means that the byte 0x20 is a starter byte in this converter. 923 * Context pointers are always owned by the caller. 924 * 925 * @param converter a valid, opened converter of type MBCS 926 * @param starters an array of size 256 to be filled in 927 * @param err error status, <TT>U_ILLEGAL_ARGUMENT_ERROR</TT> if the 928 * converter is not a type which can return starters. 929 * @see ucnv_getType 930 * @stable ICU 2.0 931 */ 932 U_CAPI void U_EXPORT2 933 ucnv_getStarters(const UConverter* converter, 934 UBool starters[256], 935 UErrorCode* err); 936 937 938 /** 939 * Selectors for Unicode sets that can be returned by ucnv_getUnicodeSet(). 940 * @see ucnv_getUnicodeSet 941 * @stable ICU 2.6 942 */ 943 typedef enum UConverterUnicodeSet { 944 /** Select the set of roundtrippable Unicode code points. @stable ICU 2.6 */ 945 UCNV_ROUNDTRIP_SET, 946 /** Select the set of Unicode code points with roundtrip or fallback mappings. @stable ICU 4.0 */ 947 UCNV_ROUNDTRIP_AND_FALLBACK_SET, 948 #ifndef U_HIDE_DEPRECATED_API 949 /** 950 * Number of UConverterUnicodeSet selectors. 951 * @deprecated ICU 58 The numeric value may change over time, see ICU ticket #12420. 952 */ 953 UCNV_SET_COUNT 954 #endif // U_HIDE_DEPRECATED_API 955 } UConverterUnicodeSet; 956 957 958 /** 959 * Returns the set of Unicode code points that can be converted by an ICU converter. 960 * 961 * Returns one of several kinds of set: 962 * 963 * 1. UCNV_ROUNDTRIP_SET 964 * 965 * The set of all Unicode code points that can be roundtrip-converted 966 * (converted without any data loss) with the converter (ucnv_fromUnicode()). 967 * This set will not include code points that have fallback mappings 968 * or are only the result of reverse fallback mappings. 969 * This set will also not include PUA code points with fallbacks, although 970 * ucnv_fromUnicode() will always uses those mappings despite ucnv_setFallback(). 971 * See UTR #22 "Character Mapping Markup Language" 972 * at http://www.unicode.org/reports/tr22/ 973 * 974 * This is useful for example for 975 * - checking that a string or document can be roundtrip-converted with a converter, 976 * without/before actually performing the conversion 977 * - testing if a converter can be used for text for typical text for a certain locale, 978 * by comparing its roundtrip set with the set of ExemplarCharacters from 979 * ICU's locale data or other sources 980 * 981 * 2. UCNV_ROUNDTRIP_AND_FALLBACK_SET 982 * 983 * The set of all Unicode code points that can be converted with the converter (ucnv_fromUnicode()) 984 * when fallbacks are turned on (see ucnv_setFallback()). 985 * This set includes all code points with roundtrips and fallbacks (but not reverse fallbacks). 986 * 987 * In the future, there may be more UConverterUnicodeSet choices to select 988 * sets with different properties. 989 * 990 * @param cnv The converter for which a set is requested. 991 * @param setFillIn A valid USet *. It will be cleared by this function before 992 * the converter's specific set is filled into the USet. 993 * @param whichSet A UConverterUnicodeSet selector; 994 * currently UCNV_ROUNDTRIP_SET is the only supported value. 995 * @param pErrorCode ICU error code in/out parameter. 996 * Must fulfill U_SUCCESS before the function call. 997 * 998 * @see UConverterUnicodeSet 999 * @see uset_open 1000 * @see uset_close 1001 * @stable ICU 2.6 1002 */ 1003 U_CAPI void U_EXPORT2 1004 ucnv_getUnicodeSet(const UConverter *cnv, 1005 USet *setFillIn, 1006 UConverterUnicodeSet whichSet, 1007 UErrorCode *pErrorCode); 1008 1009 /** 1010 * Gets the current calback function used by the converter when an illegal 1011 * or invalid codepage sequence is found. 1012 * Context pointers are always owned by the caller. 1013 * 1014 * @param converter the unicode converter 1015 * @param action fillin: returns the callback function pointer 1016 * @param context fillin: returns the callback's private void* context 1017 * @see ucnv_setToUCallBack 1018 * @stable ICU 2.0 1019 */ 1020 U_CAPI void U_EXPORT2 1021 ucnv_getToUCallBack (const UConverter * converter, 1022 UConverterToUCallback *action, 1023 const void **context); 1024 1025 /** 1026 * Gets the current callback function used by the converter when illegal 1027 * or invalid Unicode sequence is found. 1028 * Context pointers are always owned by the caller. 1029 * 1030 * @param converter the unicode converter 1031 * @param action fillin: returns the callback function pointer 1032 * @param context fillin: returns the callback's private void* context 1033 * @see ucnv_setFromUCallBack 1034 * @stable ICU 2.0 1035 */ 1036 U_CAPI void U_EXPORT2 1037 ucnv_getFromUCallBack (const UConverter * converter, 1038 UConverterFromUCallback *action, 1039 const void **context); 1040 1041 /** 1042 * Changes the callback function used by the converter when 1043 * an illegal or invalid sequence is found. 1044 * Context pointers are always owned by the caller. 1045 * Predefined actions and contexts can be found in the ucnv_err.h header. 1046 * 1047 * @param converter the unicode converter 1048 * @param newAction the new callback function 1049 * @param newContext the new toUnicode callback context pointer. This can be NULL. 1050 * @param oldAction fillin: returns the old callback function pointer. This can be NULL. 1051 * @param oldContext fillin: returns the old callback's private void* context. This can be NULL. 1052 * @param err The error code status 1053 * @see ucnv_getToUCallBack 1054 * @stable ICU 2.0 1055 */ 1056 U_CAPI void U_EXPORT2 1057 ucnv_setToUCallBack (UConverter * converter, 1058 UConverterToUCallback newAction, 1059 const void* newContext, 1060 UConverterToUCallback *oldAction, 1061 const void** oldContext, 1062 UErrorCode * err); 1063 1064 /** 1065 * Changes the current callback function used by the converter when 1066 * an illegal or invalid sequence is found. 1067 * Context pointers are always owned by the caller. 1068 * Predefined actions and contexts can be found in the ucnv_err.h header. 1069 * 1070 * @param converter the unicode converter 1071 * @param newAction the new callback function 1072 * @param newContext the new fromUnicode callback context pointer. This can be NULL. 1073 * @param oldAction fillin: returns the old callback function pointer. This can be NULL. 1074 * @param oldContext fillin: returns the old callback's private void* context. This can be NULL. 1075 * @param err The error code status 1076 * @see ucnv_getFromUCallBack 1077 * @stable ICU 2.0 1078 */ 1079 U_CAPI void U_EXPORT2 1080 ucnv_setFromUCallBack (UConverter * converter, 1081 UConverterFromUCallback newAction, 1082 const void *newContext, 1083 UConverterFromUCallback *oldAction, 1084 const void **oldContext, 1085 UErrorCode * err); 1086 1087 /** 1088 * Converts an array of unicode characters to an array of codepage 1089 * characters. This function is optimized for converting a continuous 1090 * stream of data in buffer-sized chunks, where the entire source and 1091 * target does not fit in available buffers. 1092 * 1093 * The source pointer is an in/out parameter. It starts out pointing where the 1094 * conversion is to begin, and ends up pointing after the last UChar consumed. 1095 * 1096 * Target similarly starts out pointer at the first available byte in the output 1097 * buffer, and ends up pointing after the last byte written to the output. 1098 * 1099 * The converter always attempts to consume the entire source buffer, unless 1100 * (1.) the target buffer is full, or (2.) a failing error is returned from the 1101 * current callback function. When a successful error status has been 1102 * returned, it means that all of the source buffer has been 1103 * consumed. At that point, the caller should reset the source and 1104 * sourceLimit pointers to point to the next chunk. 1105 * 1106 * At the end of the stream (flush==true), the input is completely consumed 1107 * when *source==sourceLimit and no error code is set. 1108 * The converter object is then automatically reset by this function. 1109 * (This means that a converter need not be reset explicitly between data 1110 * streams if it finishes the previous stream without errors.) 1111 * 1112 * This is a <I>stateful</I> conversion. Additionally, even when all source data has 1113 * been consumed, some data may be in the converters' internal state. 1114 * Call this function repeatedly, updating the target pointers with 1115 * the next empty chunk of target in case of a 1116 * <TT>U_BUFFER_OVERFLOW_ERROR</TT>, and updating the source pointers 1117 * with the next chunk of source when a successful error status is 1118 * returned, until there are no more chunks of source data. 1119 * @param converter the Unicode converter 1120 * @param target I/O parameter. Input : Points to the beginning of the buffer to copy 1121 * codepage characters to. Output : points to after the last codepage character copied 1122 * to <TT>target</TT>. 1123 * @param targetLimit the pointer just after last of the <TT>target</TT> buffer 1124 * @param source I/O parameter, pointer to pointer to the source Unicode character buffer. 1125 * @param sourceLimit the pointer just after the last of the source buffer 1126 * @param offsets if NULL is passed, nothing will happen to it, otherwise it needs to have the same number 1127 * of allocated cells as <TT>target</TT>. Will fill in offsets from target to source pointer 1128 * 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> 1129 * For output data carried across calls, and other data without a specific source character 1130 * (such as from escape sequences or callbacks) -1 will be placed for offsets. 1131 * @param flush set to <TT>true</TT> if the current source buffer is the last available 1132 * chunk of the source, <TT>false</TT> otherwise. Note that if a failing status is returned, 1133 * this function may have to be called multiple times with flush set to <TT>true</TT> until 1134 * the source buffer is consumed. 1135 * @param err the error status. <TT>U_ILLEGAL_ARGUMENT_ERROR</TT> will be set if the 1136 * converter is <TT>NULL</TT>. 1137 * <code>U_BUFFER_OVERFLOW_ERROR</code> will be set if the target is full and there is 1138 * still data to be written to the target. 1139 * @see ucnv_fromUChars 1140 * @see ucnv_convert 1141 * @see ucnv_getMinCharSize 1142 * @see ucnv_setToUCallBack 1143 * @stable ICU 2.0 1144 */ 1145 U_CAPI void U_EXPORT2 1146 ucnv_fromUnicode (UConverter * converter, 1147 char **target, 1148 const char *targetLimit, 1149 const UChar ** source, 1150 const UChar * sourceLimit, 1151 int32_t* offsets, 1152 UBool flush, 1153 UErrorCode * err); 1154 1155 /** 1156 * Converts a buffer of codepage bytes into an array of unicode UChars 1157 * characters. This function is optimized for converting a continuous 1158 * stream of data in buffer-sized chunks, where the entire source and 1159 * target does not fit in available buffers. 1160 * 1161 * The source pointer is an in/out parameter. It starts out pointing where the 1162 * conversion is to begin, and ends up pointing after the last byte of source consumed. 1163 * 1164 * Target similarly starts out pointer at the first available UChar in the output 1165 * buffer, and ends up pointing after the last UChar written to the output. 1166 * It does NOT necessarily keep UChar sequences together. 1167 * 1168 * The converter always attempts to consume the entire source buffer, unless 1169 * (1.) the target buffer is full, or (2.) a failing error is returned from the 1170 * current callback function. When a successful error status has been 1171 * returned, it means that all of the source buffer has been 1172 * consumed. At that point, the caller should reset the source and 1173 * sourceLimit pointers to point to the next chunk. 1174 * 1175 * At the end of the stream (flush==true), the input is completely consumed 1176 * when *source==sourceLimit and no error code is set 1177 * The converter object is then automatically reset by this function. 1178 * (This means that a converter need not be reset explicitly between data 1179 * streams if it finishes the previous stream without errors.) 1180 * 1181 * This is a <I>stateful</I> conversion. Additionally, even when all source data has 1182 * been consumed, some data may be in the converters' internal state. 1183 * Call this function repeatedly, updating the target pointers with 1184 * the next empty chunk of target in case of a 1185 * <TT>U_BUFFER_OVERFLOW_ERROR</TT>, and updating the source pointers 1186 * with the next chunk of source when a successful error status is 1187 * returned, until there are no more chunks of source data. 1188 * @param converter the Unicode converter 1189 * @param target I/O parameter. Input : Points to the beginning of the buffer to copy 1190 * UChars into. Output : points to after the last UChar copied. 1191 * @param targetLimit the pointer just after the end of the <TT>target</TT> buffer 1192 * @param source I/O parameter, pointer to pointer to the source codepage buffer. 1193 * @param sourceLimit the pointer to the byte after the end of the source buffer 1194 * @param offsets if NULL is passed, nothing will happen to it, otherwise it needs to have the same number 1195 * of allocated cells as <TT>target</TT>. Will fill in offsets from target to source pointer 1196 * 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> 1197 * For output data carried across calls, and other data without a specific source character 1198 * (such as from escape sequences or callbacks) -1 will be placed for offsets. 1199 * @param flush set to <TT>true</TT> if the current source buffer is the last available 1200 * chunk of the source, <TT>false</TT> otherwise. Note that if a failing status is returned, 1201 * this function may have to be called multiple times with flush set to <TT>true</TT> until 1202 * the source buffer is consumed. 1203 * @param err the error status. <TT>U_ILLEGAL_ARGUMENT_ERROR</TT> will be set if the 1204 * converter is <TT>NULL</TT>. 1205 * <code>U_BUFFER_OVERFLOW_ERROR</code> will be set if the target is full and there is 1206 * still data to be written to the target. 1207 * @see ucnv_fromUChars 1208 * @see ucnv_convert 1209 * @see ucnv_getMinCharSize 1210 * @see ucnv_setFromUCallBack 1211 * @see ucnv_getNextUChar 1212 * @stable ICU 2.0 1213 */ 1214 U_CAPI void U_EXPORT2 1215 ucnv_toUnicode(UConverter *converter, 1216 UChar **target, 1217 const UChar *targetLimit, 1218 const char **source, 1219 const char *sourceLimit, 1220 int32_t *offsets, 1221 UBool flush, 1222 UErrorCode *err); 1223 1224 /** 1225 * Convert the Unicode string into a codepage string using an existing UConverter. 1226 * The output string is NUL-terminated if possible. 1227 * 1228 * This function is a more convenient but less powerful version of ucnv_fromUnicode(). 1229 * It is only useful for whole strings, not for streaming conversion. 1230 * 1231 * The maximum output buffer capacity required (barring output from callbacks) will be 1232 * UCNV_GET_MAX_BYTES_FOR_STRING(srcLength, ucnv_getMaxCharSize(cnv)). 1233 * 1234 * @param cnv the converter object to be used (ucnv_resetFromUnicode() will be called) 1235 * @param src the input Unicode string 1236 * @param srcLength the input string length, or -1 if NUL-terminated 1237 * @param dest destination string buffer, can be NULL if destCapacity==0 1238 * @param destCapacity the number of chars available at dest 1239 * @param pErrorCode normal ICU error code; 1240 * common error codes that may be set by this function include 1241 * U_BUFFER_OVERFLOW_ERROR, U_STRING_NOT_TERMINATED_WARNING, 1242 * U_ILLEGAL_ARGUMENT_ERROR, and conversion errors 1243 * @return the length of the output string, not counting the terminating NUL; 1244 * if the length is greater than destCapacity, then the string will not fit 1245 * and a buffer of the indicated length would need to be passed in 1246 * @see ucnv_fromUnicode 1247 * @see ucnv_convert 1248 * @see UCNV_GET_MAX_BYTES_FOR_STRING 1249 * @stable ICU 2.0 1250 */ 1251 U_CAPI int32_t U_EXPORT2 1252 ucnv_fromUChars(UConverter *cnv, 1253 char *dest, int32_t destCapacity, 1254 const UChar *src, int32_t srcLength, 1255 UErrorCode *pErrorCode); 1256 1257 /** 1258 * Convert the codepage string into a Unicode string using an existing UConverter. 1259 * The output string is NUL-terminated if possible. 1260 * 1261 * This function is a more convenient but less powerful version of ucnv_toUnicode(). 1262 * It is only useful for whole strings, not for streaming conversion. 1263 * 1264 * The maximum output buffer capacity required (barring output from callbacks) will be 1265 * 2*srcLength (each char may be converted into a surrogate pair). 1266 * 1267 * @param cnv the converter object to be used (ucnv_resetToUnicode() will be called) 1268 * @param src the input codepage string 1269 * @param srcLength the input string length, or -1 if NUL-terminated 1270 * @param dest destination string buffer, can be NULL if destCapacity==0 1271 * @param destCapacity the number of UChars available at dest 1272 * @param pErrorCode normal ICU error code; 1273 * common error codes that may be set by this function include 1274 * U_BUFFER_OVERFLOW_ERROR, U_STRING_NOT_TERMINATED_WARNING, 1275 * U_ILLEGAL_ARGUMENT_ERROR, and conversion errors 1276 * @return the length of the output string, not counting the terminating NUL; 1277 * if the length is greater than destCapacity, then the string will not fit 1278 * and a buffer of the indicated length would need to be passed in 1279 * @see ucnv_toUnicode 1280 * @see ucnv_convert 1281 * @stable ICU 2.0 1282 */ 1283 U_CAPI int32_t U_EXPORT2 1284 ucnv_toUChars(UConverter *cnv, 1285 UChar *dest, int32_t destCapacity, 1286 const char *src, int32_t srcLength, 1287 UErrorCode *pErrorCode); 1288 1289 /** 1290 * Convert a codepage buffer into Unicode one character at a time. 1291 * The input is completely consumed when the U_INDEX_OUTOFBOUNDS_ERROR is set. 1292 * 1293 * Advantage compared to ucnv_toUnicode() or ucnv_toUChars(): 1294 * - Faster for small amounts of data, for most converters, e.g., 1295 * US-ASCII, ISO-8859-1, UTF-8/16/32, and most "normal" charsets. 1296 * (For complex converters, e.g., SCSU, UTF-7 and ISO 2022 variants, 1297 * it uses ucnv_toUnicode() internally.) 1298 * - Convenient. 1299 * 1300 * Limitations compared to ucnv_toUnicode(): 1301 * - Always assumes flush=true. 1302 * This makes ucnv_getNextUChar() unsuitable for "streaming" conversion, 1303 * that is, for where the input is supplied in multiple buffers, 1304 * because ucnv_getNextUChar() will assume the end of the input at the end 1305 * of the first buffer. 1306 * - Does not provide offset output. 1307 * 1308 * It is possible to "mix" ucnv_getNextUChar() and ucnv_toUnicode() because 1309 * ucnv_getNextUChar() uses the current state of the converter 1310 * (unlike ucnv_toUChars() which always resets first). 1311 * However, if ucnv_getNextUChar() is called after ucnv_toUnicode() 1312 * stopped in the middle of a character sequence (with flush=false), 1313 * then ucnv_getNextUChar() will always use the slower ucnv_toUnicode() 1314 * internally until the next character boundary. 1315 * (This is new in ICU 2.6. In earlier releases, ucnv_getNextUChar() had to 1316 * start at a character boundary.) 1317 * 1318 * Instead of using ucnv_getNextUChar(), it is recommended 1319 * to convert using ucnv_toUnicode() or ucnv_toUChars() 1320 * and then iterate over the text using U16_NEXT() or a UCharIterator (uiter.h) 1321 * or a C++ CharacterIterator or similar. 1322 * This allows streaming conversion and offset output, for example. 1323 * 1324 * <p>Handling of surrogate pairs and supplementary-plane code points:<br> 1325 * There are two different kinds of codepages that provide mappings for surrogate characters: 1326 * <ul> 1327 * <li>Codepages like UTF-8, UTF-32, and GB 18030 provide direct representations for Unicode 1328 * code points U+10000-U+10ffff as well as for single surrogates U+d800-U+dfff. 1329 * Each valid sequence will result in exactly one returned code point. 1330 * If a sequence results in a single surrogate, then that will be returned 1331 * by itself, even if a neighboring sequence encodes the matching surrogate.</li> 1332 * <li>Codepages like SCSU and LMBCS (and UTF-16) provide direct representations only for BMP code points 1333 * including surrogates. Code points in supplementary planes are represented with 1334 * two sequences, each encoding a surrogate. 1335 * For these codepages, matching pairs of surrogates will be combined into single 1336 * code points for returning from this function. 1337 * (Note that SCSU is actually a mix of these codepage types.)</li> 1338 * </ul></p> 1339 * 1340 * @param converter an open UConverter 1341 * @param source the address of a pointer to the codepage buffer, will be 1342 * updated to point after the bytes consumed in the conversion call. 1343 * @param sourceLimit points to the end of the input buffer 1344 * @param err fills in error status (see ucnv_toUnicode) 1345 * <code>U_INDEX_OUTOFBOUNDS_ERROR</code> will be set if the input 1346 * is empty or does not convert to any output (e.g.: pure state-change 1347 * codes SI/SO, escape sequences for ISO 2022, 1348 * or if the callback did not output anything, ...). 1349 * This function will not set a <code>U_BUFFER_OVERFLOW_ERROR</code> because 1350 * the "buffer" is the return code. However, there might be subsequent output 1351 * stored in the converter object 1352 * that will be returned in following calls to this function. 1353 * @return a UChar32 resulting from the partial conversion of source 1354 * @see ucnv_toUnicode 1355 * @see ucnv_toUChars 1356 * @see ucnv_convert 1357 * @stable ICU 2.0 1358 */ 1359 U_CAPI UChar32 U_EXPORT2 1360 ucnv_getNextUChar(UConverter * converter, 1361 const char **source, 1362 const char * sourceLimit, 1363 UErrorCode * err); 1364 1365 /** 1366 * Convert from one external charset to another using two existing UConverters. 1367 * Internally, two conversions - ucnv_toUnicode() and ucnv_fromUnicode() - 1368 * are used, "pivoting" through 16-bit Unicode. 1369 * 1370 * Important: For streaming conversion (multiple function calls for successive 1371 * parts of a text stream), the caller must provide a pivot buffer explicitly, 1372 * and must preserve the pivot buffer and associated pointers from one 1373 * call to another. (The buffer may be moved if its contents and the relative 1374 * pointer positions are preserved.) 1375 * 1376 * There is a similar function, ucnv_convert(), 1377 * which has the following limitations: 1378 * - it takes charset names, not converter objects, so that 1379 * - two converters are opened for each call 1380 * - only single-string conversion is possible, not streaming operation 1381 * - it does not provide enough information to find out, 1382 * in case of failure, whether the toUnicode or 1383 * the fromUnicode conversion failed 1384 * 1385 * By contrast, ucnv_convertEx() 1386 * - takes UConverter parameters instead of charset names 1387 * - fully exposes the pivot buffer for streaming conversion and complete error handling 1388 * 1389 * ucnv_convertEx() also provides further convenience: 1390 * - an option to reset the converters at the beginning 1391 * (if reset==true, see parameters; 1392 * also sets *pivotTarget=*pivotSource=pivotStart) 1393 * - allow NUL-terminated input 1394 * (only a single NUL byte, will not work for charsets with multi-byte NULs) 1395 * (if sourceLimit==NULL, see parameters) 1396 * - terminate with a NUL on output 1397 * (only a single NUL byte, not useful for charsets with multi-byte NULs), 1398 * or set U_STRING_NOT_TERMINATED_WARNING if the output exactly fills 1399 * the target buffer 1400 * - the pivot buffer can be provided internally; 1401 * possible only for whole-string conversion, not streaming conversion; 1402 * in this case, the caller will not be able to get details about where an 1403 * error occurred 1404 * (if pivotStart==NULL, see below) 1405 * 1406 * The function returns when one of the following is true: 1407 * - the entire source text has been converted successfully to the target buffer 1408 * - a target buffer overflow occurred (U_BUFFER_OVERFLOW_ERROR) 1409 * - a conversion error occurred 1410 * (other U_FAILURE(), see description of pErrorCode) 1411 * 1412 * Limitation compared to the direct use of 1413 * ucnv_fromUnicode() and ucnv_toUnicode(): 1414 * ucnv_convertEx() does not provide offset information. 1415 * 1416 * Limitation compared to ucnv_fromUChars() and ucnv_toUChars(): 1417 * ucnv_convertEx() does not support preflighting directly. 1418 * 1419 * Sample code for converting a single string from 1420 * one external charset to UTF-8, ignoring the location of errors: 1421 * 1422 * \code 1423 * int32_t 1424 * myToUTF8(UConverter *cnv, 1425 * const char *s, int32_t length, 1426 * char *u8, int32_t capacity, 1427 * UErrorCode *pErrorCode) { 1428 * UConverter *utf8Cnv; 1429 * char *target; 1430 * 1431 * if(U_FAILURE(*pErrorCode)) { 1432 * return 0; 1433 * } 1434 * 1435 * utf8Cnv=myGetCachedUTF8Converter(pErrorCode); 1436 * if(U_FAILURE(*pErrorCode)) { 1437 * return 0; 1438 * } 1439 * 1440 * if(length<0) { 1441 * length=strlen(s); 1442 * } 1443 * target=u8; 1444 * ucnv_convertEx(utf8Cnv, cnv, 1445 * &target, u8+capacity, 1446 * &s, s+length, 1447 * NULL, NULL, NULL, NULL, 1448 * true, true, 1449 * pErrorCode); 1450 * 1451 * myReleaseCachedUTF8Converter(utf8Cnv); 1452 * 1453 * // return the output string length, but without preflighting 1454 * return (int32_t)(target-u8); 1455 * } 1456 * \endcode 1457 * 1458 * @param targetCnv Output converter, used to convert from the UTF-16 pivot 1459 * to the target using ucnv_fromUnicode(). 1460 * @param sourceCnv Input converter, used to convert from the source to 1461 * the UTF-16 pivot using ucnv_toUnicode(). 1462 * @param target I/O parameter, same as for ucnv_fromUChars(). 1463 * Input: *target points to the beginning of the target buffer. 1464 * Output: *target points to the first unit after the last char written. 1465 * @param targetLimit Pointer to the first unit after the target buffer. 1466 * @param source I/O parameter, same as for ucnv_toUChars(). 1467 * Input: *source points to the beginning of the source buffer. 1468 * Output: *source points to the first unit after the last char read. 1469 * @param sourceLimit Pointer to the first unit after the source buffer. 1470 * @param pivotStart Pointer to the UTF-16 pivot buffer. If pivotStart==NULL, 1471 * then an internal buffer is used and the other pivot 1472 * arguments are ignored and can be NULL as well. 1473 * @param pivotSource I/O parameter, same as source in ucnv_fromUChars() for 1474 * conversion from the pivot buffer to the target buffer. 1475 * @param pivotTarget I/O parameter, same as target in ucnv_toUChars() for 1476 * conversion from the source buffer to the pivot buffer. 1477 * It must be pivotStart<=*pivotSource<=*pivotTarget<=pivotLimit 1478 * and pivotStart<pivotLimit (unless pivotStart==NULL). 1479 * @param pivotLimit Pointer to the first unit after the pivot buffer. 1480 * @param reset If true, then ucnv_resetToUnicode(sourceCnv) and 1481 * ucnv_resetFromUnicode(targetCnv) are called, and the 1482 * pivot pointers are reset (*pivotTarget=*pivotSource=pivotStart). 1483 * @param flush If true, indicates the end of the input. 1484 * Passed directly to ucnv_toUnicode(), and carried over to 1485 * ucnv_fromUnicode() when the source is empty as well. 1486 * @param pErrorCode ICU error code in/out parameter. 1487 * Must fulfill U_SUCCESS before the function call. 1488 * U_BUFFER_OVERFLOW_ERROR always refers to the target buffer 1489 * because overflows into the pivot buffer are handled internally. 1490 * Other conversion errors are from the source-to-pivot 1491 * conversion if *pivotSource==pivotStart, otherwise from 1492 * the pivot-to-target conversion. 1493 * 1494 * @see ucnv_convert 1495 * @see ucnv_fromAlgorithmic 1496 * @see ucnv_toAlgorithmic 1497 * @see ucnv_fromUnicode 1498 * @see ucnv_toUnicode 1499 * @see ucnv_fromUChars 1500 * @see ucnv_toUChars 1501 * @stable ICU 2.6 1502 */ 1503 U_CAPI void U_EXPORT2 1504 ucnv_convertEx(UConverter *targetCnv, UConverter *sourceCnv, 1505 char **target, const char *targetLimit, 1506 const char **source, const char *sourceLimit, 1507 UChar *pivotStart, UChar **pivotSource, 1508 UChar **pivotTarget, const UChar *pivotLimit, 1509 UBool reset, UBool flush, 1510 UErrorCode *pErrorCode); 1511 1512 /** 1513 * Convert from one external charset to another. 1514 * Internally, two converters are opened according to the name arguments, 1515 * then the text is converted to and from the 16-bit Unicode "pivot" 1516 * using ucnv_convertEx(), then the converters are closed again. 1517 * 1518 * This is a convenience function, not an efficient way to convert a lot of text: 1519 * ucnv_convert() 1520 * - takes charset names, not converter objects, so that 1521 * - two converters are opened for each call 1522 * - only single-string conversion is possible, not streaming operation 1523 * - does not provide enough information to find out, 1524 * in case of failure, whether the toUnicode or 1525 * the fromUnicode conversion failed 1526 * - allows NUL-terminated input 1527 * (only a single NUL byte, will not work for charsets with multi-byte NULs) 1528 * (if sourceLength==-1, see parameters) 1529 * - terminate with a NUL on output 1530 * (only a single NUL byte, not useful for charsets with multi-byte NULs), 1531 * or set U_STRING_NOT_TERMINATED_WARNING if the output exactly fills 1532 * the target buffer 1533 * - a pivot buffer is provided internally 1534 * 1535 * The function returns when one of the following is true: 1536 * - the entire source text has been converted successfully to the target buffer 1537 * and either the target buffer is terminated with a single NUL byte 1538 * or the error code is set to U_STRING_NOT_TERMINATED_WARNING 1539 * - a target buffer overflow occurred (U_BUFFER_OVERFLOW_ERROR) 1540 * and the full output string length is returned ("preflighting") 1541 * - a conversion error occurred 1542 * (other U_FAILURE(), see description of pErrorCode) 1543 * 1544 * @param toConverterName The name of the converter that is used to convert 1545 * from the UTF-16 pivot buffer to the target. 1546 * @param fromConverterName The name of the converter that is used to convert 1547 * from the source to the UTF-16 pivot buffer. 1548 * @param target Pointer to the output buffer. 1549 * @param targetCapacity Capacity of the target, in bytes. 1550 * @param source Pointer to the input buffer. 1551 * @param sourceLength Length of the input text, in bytes, or -1 for NUL-terminated input. 1552 * @param pErrorCode ICU error code in/out parameter. 1553 * Must fulfill U_SUCCESS before the function call. 1554 * @return Length of the complete output text in bytes, even if it exceeds the targetCapacity 1555 * and a U_BUFFER_OVERFLOW_ERROR is set. 1556 * 1557 * @see ucnv_convertEx 1558 * @see ucnv_fromAlgorithmic 1559 * @see ucnv_toAlgorithmic 1560 * @see ucnv_fromUnicode 1561 * @see ucnv_toUnicode 1562 * @see ucnv_fromUChars 1563 * @see ucnv_toUChars 1564 * @see ucnv_getNextUChar 1565 * @stable ICU 2.0 1566 */ 1567 U_CAPI int32_t U_EXPORT2 1568 ucnv_convert(const char *toConverterName, 1569 const char *fromConverterName, 1570 char *target, 1571 int32_t targetCapacity, 1572 const char *source, 1573 int32_t sourceLength, 1574 UErrorCode *pErrorCode); 1575 1576 /** 1577 * Convert from one external charset to another. 1578 * Internally, the text is converted to and from the 16-bit Unicode "pivot" 1579 * using ucnv_convertEx(). ucnv_toAlgorithmic() works exactly like ucnv_convert() 1580 * except that the two converters need not be looked up and opened completely. 1581 * 1582 * The source-to-pivot conversion uses the cnv converter parameter. 1583 * The pivot-to-target conversion uses a purely algorithmic converter 1584 * according to the specified type, e.g., UCNV_UTF8 for a UTF-8 converter. 1585 * 1586 * Internally, the algorithmic converter is opened and closed for each 1587 * function call, which is more efficient than using the public ucnv_open() 1588 * but somewhat less efficient than only resetting an existing converter 1589 * and using ucnv_convertEx(). 1590 * 1591 * This function is more convenient than ucnv_convertEx() for single-string 1592 * conversions, especially when "preflighting" is desired (returning the length 1593 * of the complete output even if it does not fit into the target buffer; 1594 * see the User Guide Strings chapter). See ucnv_convert() for details. 1595 * 1596 * @param algorithmicType UConverterType constant identifying the desired target 1597 * charset as a purely algorithmic converter. 1598 * Those are converters for Unicode charsets like 1599 * UTF-8, BOCU-1, SCSU, UTF-7, IMAP-mailbox-name, etc., 1600 * as well as US-ASCII and ISO-8859-1. 1601 * @param cnv The converter that is used to convert 1602 * from the source to the UTF-16 pivot buffer. 1603 * @param target Pointer to the output buffer. 1604 * @param targetCapacity Capacity of the target, in bytes. 1605 * @param source Pointer to the input buffer. 1606 * @param sourceLength Length of the input text, in bytes 1607 * @param pErrorCode ICU error code in/out parameter. 1608 * Must fulfill U_SUCCESS before the function call. 1609 * @return Length of the complete output text in bytes, even if it exceeds the targetCapacity 1610 * and a U_BUFFER_OVERFLOW_ERROR is set. 1611 * 1612 * @see ucnv_fromAlgorithmic 1613 * @see ucnv_convert 1614 * @see ucnv_convertEx 1615 * @see ucnv_fromUnicode 1616 * @see ucnv_toUnicode 1617 * @see ucnv_fromUChars 1618 * @see ucnv_toUChars 1619 * @stable ICU 2.6 1620 */ 1621 U_CAPI int32_t U_EXPORT2 1622 ucnv_toAlgorithmic(UConverterType algorithmicType, 1623 UConverter *cnv, 1624 char *target, int32_t targetCapacity, 1625 const char *source, int32_t sourceLength, 1626 UErrorCode *pErrorCode); 1627 1628 /** 1629 * Convert from one external charset to another. 1630 * Internally, the text is converted to and from the 16-bit Unicode "pivot" 1631 * using ucnv_convertEx(). ucnv_fromAlgorithmic() works exactly like ucnv_convert() 1632 * except that the two converters need not be looked up and opened completely. 1633 * 1634 * The source-to-pivot conversion uses a purely algorithmic converter 1635 * according to the specified type, e.g., UCNV_UTF8 for a UTF-8 converter. 1636 * The pivot-to-target conversion uses the cnv converter parameter. 1637 * 1638 * Internally, the algorithmic converter is opened and closed for each 1639 * function call, which is more efficient than using the public ucnv_open() 1640 * but somewhat less efficient than only resetting an existing converter 1641 * and using ucnv_convertEx(). 1642 * 1643 * This function is more convenient than ucnv_convertEx() for single-string 1644 * conversions, especially when "preflighting" is desired (returning the length 1645 * of the complete output even if it does not fit into the target buffer; 1646 * see the User Guide Strings chapter). See ucnv_convert() for details. 1647 * 1648 * @param cnv The converter that is used to convert 1649 * from the UTF-16 pivot buffer to the target. 1650 * @param algorithmicType UConverterType constant identifying the desired source 1651 * charset as a purely algorithmic converter. 1652 * Those are converters for Unicode charsets like 1653 * UTF-8, BOCU-1, SCSU, UTF-7, IMAP-mailbox-name, etc., 1654 * as well as US-ASCII and ISO-8859-1. 1655 * @param target Pointer to the output buffer. 1656 * @param targetCapacity Capacity of the target, in bytes. 1657 * @param source Pointer to the input buffer. 1658 * @param sourceLength Length of the input text, in bytes 1659 * @param pErrorCode ICU error code in/out parameter. 1660 * Must fulfill U_SUCCESS before the function call. 1661 * @return Length of the complete output text in bytes, even if it exceeds the targetCapacity 1662 * and a U_BUFFER_OVERFLOW_ERROR is set. 1663 * 1664 * @see ucnv_fromAlgorithmic 1665 * @see ucnv_convert 1666 * @see ucnv_convertEx 1667 * @see ucnv_fromUnicode 1668 * @see ucnv_toUnicode 1669 * @see ucnv_fromUChars 1670 * @see ucnv_toUChars 1671 * @stable ICU 2.6 1672 */ 1673 U_CAPI int32_t U_EXPORT2 1674 ucnv_fromAlgorithmic(UConverter *cnv, 1675 UConverterType algorithmicType, 1676 char *target, int32_t targetCapacity, 1677 const char *source, int32_t sourceLength, 1678 UErrorCode *pErrorCode); 1679 1680 /** 1681 * Frees up memory occupied by unused, cached converter shared data. 1682 * 1683 * @return the number of cached converters successfully deleted 1684 * @see ucnv_close 1685 * @stable ICU 2.0 1686 */ 1687 U_CAPI int32_t U_EXPORT2 1688 ucnv_flushCache(void); 1689 1690 /** 1691 * Returns the number of available converters, as per the alias file. 1692 * 1693 * @return the number of available converters 1694 * @see ucnv_getAvailableName 1695 * @stable ICU 2.0 1696 */ 1697 U_CAPI int32_t U_EXPORT2 1698 ucnv_countAvailable(void); 1699 1700 /** 1701 * Gets the canonical converter name of the specified converter from a list of 1702 * all available converters contaied in the alias file. All converters 1703 * in this list can be opened. 1704 * 1705 * @param n the index to a converter available on the system (in the range <TT>[0..ucnv_countAvaiable()]</TT>) 1706 * @return a pointer a string (library owned), or <TT>NULL</TT> if the index is out of bounds. 1707 * @see ucnv_countAvailable 1708 * @stable ICU 2.0 1709 */ 1710 U_CAPI const char* U_EXPORT2 1711 ucnv_getAvailableName(int32_t n); 1712 1713 /** 1714 * Returns a UEnumeration to enumerate all of the canonical converter 1715 * names, as per the alias file, regardless of the ability to open each 1716 * converter. 1717 * 1718 * @return A UEnumeration object for getting all the recognized canonical 1719 * converter names. 1720 * @see ucnv_getAvailableName 1721 * @see uenum_close 1722 * @see uenum_next 1723 * @stable ICU 2.4 1724 */ 1725 U_CAPI UEnumeration * U_EXPORT2 1726 ucnv_openAllNames(UErrorCode *pErrorCode); 1727 1728 /** 1729 * Gives the number of aliases for a given converter or alias name. 1730 * If the alias is ambiguous, then the preferred converter is used 1731 * and the status is set to U_AMBIGUOUS_ALIAS_WARNING. 1732 * This method only enumerates the listed entries in the alias file. 1733 * @param alias alias name 1734 * @param pErrorCode error status 1735 * @return number of names on alias list for given alias 1736 * @stable ICU 2.0 1737 */ 1738 U_CAPI uint16_t U_EXPORT2 1739 ucnv_countAliases(const char *alias, UErrorCode *pErrorCode); 1740 1741 /** 1742 * Gives the name of the alias at given index of alias list. 1743 * This method only enumerates the listed entries in the alias file. 1744 * If the alias is ambiguous, then the preferred converter is used 1745 * and the status is set to U_AMBIGUOUS_ALIAS_WARNING. 1746 * @param alias alias name 1747 * @param n index in alias list 1748 * @param pErrorCode result of operation 1749 * @return returns the name of the alias at given index 1750 * @see ucnv_countAliases 1751 * @stable ICU 2.0 1752 */ 1753 U_CAPI const char * U_EXPORT2 1754 ucnv_getAlias(const char *alias, uint16_t n, UErrorCode *pErrorCode); 1755 1756 /** 1757 * Fill-up the list of alias names for the given alias. 1758 * This method only enumerates the listed entries in the alias file. 1759 * If the alias is ambiguous, then the preferred converter is used 1760 * and the status is set to U_AMBIGUOUS_ALIAS_WARNING. 1761 * @param alias alias name 1762 * @param aliases fill-in list, aliases is a pointer to an array of 1763 * <code>ucnv_countAliases()</code> string-pointers 1764 * (<code>const char *</code>) that will be filled in. 1765 * The strings themselves are owned by the library. 1766 * @param pErrorCode result of operation 1767 * @stable ICU 2.0 1768 */ 1769 U_CAPI void U_EXPORT2 1770 ucnv_getAliases(const char *alias, const char **aliases, UErrorCode *pErrorCode); 1771 1772 /** 1773 * Return a new UEnumeration object for enumerating all the 1774 * alias names for a given converter that are recognized by a standard. 1775 * This method only enumerates the listed entries in the alias file. 1776 * The convrtrs.txt file can be modified to change the results of 1777 * this function. 1778 * The first result in this list is the same result given by 1779 * <code>ucnv_getStandardName</code>, which is the default alias for 1780 * the specified standard name. The returned object must be closed with 1781 * <code>uenum_close</code> when you are done with the object. 1782 * 1783 * @param convName original converter name 1784 * @param standard name of the standard governing the names; MIME and IANA 1785 * are such standards 1786 * @param pErrorCode The error code 1787 * @return A UEnumeration object for getting all aliases that are recognized 1788 * by a standard. If any of the parameters are invalid, NULL 1789 * is returned. 1790 * @see ucnv_getStandardName 1791 * @see uenum_close 1792 * @see uenum_next 1793 * @stable ICU 2.2 1794 */ 1795 U_CAPI UEnumeration * U_EXPORT2 1796 ucnv_openStandardNames(const char *convName, 1797 const char *standard, 1798 UErrorCode *pErrorCode); 1799 1800 /** 1801 * Gives the number of standards associated to converter names. 1802 * @return number of standards 1803 * @stable ICU 2.0 1804 */ 1805 U_CAPI uint16_t U_EXPORT2 1806 ucnv_countStandards(void); 1807 1808 /** 1809 * Gives the name of the standard at given index of standard list. 1810 * @param n index in standard list 1811 * @param pErrorCode result of operation 1812 * @return returns the name of the standard at given index. Owned by the library. 1813 * @stable ICU 2.0 1814 */ 1815 U_CAPI const char * U_EXPORT2 1816 ucnv_getStandard(uint16_t n, UErrorCode *pErrorCode); 1817 1818 /** 1819 * Returns a standard name for a given converter name. 1820 * <p> 1821 * Example alias table:<br> 1822 * conv alias1 { STANDARD1 } alias2 { STANDARD1* } 1823 * <p> 1824 * Result of ucnv_getStandardName("conv", "STANDARD1") from example 1825 * alias table:<br> 1826 * <b>"alias2"</b> 1827 * 1828 * @param name original converter name 1829 * @param standard name of the standard governing the names; MIME and IANA 1830 * are such standards 1831 * @param pErrorCode result of operation 1832 * @return returns the standard converter name; 1833 * if a standard converter name cannot be determined, 1834 * then <code>NULL</code> is returned. Owned by the library. 1835 * @stable ICU 2.0 1836 */ 1837 U_CAPI const char * U_EXPORT2 1838 ucnv_getStandardName(const char *name, const char *standard, UErrorCode *pErrorCode); 1839 1840 /** 1841 * This function will return the internal canonical converter name of the 1842 * tagged alias. This is the opposite of ucnv_openStandardNames, which 1843 * returns the tagged alias given the canonical name. 1844 * <p> 1845 * Example alias table:<br> 1846 * conv alias1 { STANDARD1 } alias2 { STANDARD1* } 1847 * <p> 1848 * Result of ucnv_getStandardName("alias1", "STANDARD1") from example 1849 * alias table:<br> 1850 * <b>"conv"</b> 1851 * 1852 * @return returns the canonical converter name; 1853 * if a standard or alias name cannot be determined, 1854 * then <code>NULL</code> is returned. The returned string is 1855 * owned by the library. 1856 * @see ucnv_getStandardName 1857 * @stable ICU 2.4 1858 */ 1859 U_CAPI const char * U_EXPORT2 1860 ucnv_getCanonicalName(const char *alias, const char *standard, UErrorCode *pErrorCode); 1861 1862 /** 1863 * Returns the current default converter name. If you want to open 1864 * a default converter, you do not need to use this function. 1865 * It is faster if you pass a NULL argument to ucnv_open the 1866 * default converter. 1867 * 1868 * If U_CHARSET_IS_UTF8 is defined to 1 in utypes.h then this function 1869 * always returns "UTF-8". 1870 * 1871 * @return returns the current default converter name. 1872 * Storage owned by the library 1873 * @see ucnv_setDefaultName 1874 * @stable ICU 2.0 1875 */ 1876 U_CAPI const char * U_EXPORT2 1877 ucnv_getDefaultName(void); 1878 1879 #ifndef U_HIDE_SYSTEM_API 1880 /** 1881 * This function is not thread safe. DO NOT call this function when ANY ICU 1882 * function is being used from more than one thread! This function sets the 1883 * current default converter name. If this function needs to be called, it 1884 * should be called during application initialization. Most of the time, the 1885 * results from ucnv_getDefaultName() or ucnv_open with a NULL string argument 1886 * is sufficient for your application. 1887 * 1888 * If U_CHARSET_IS_UTF8 is defined to 1 in utypes.h then this function 1889 * does nothing. 1890 * 1891 * @param name the converter name to be the default (must be known by ICU). 1892 * @see ucnv_getDefaultName 1893 * @system 1894 * @stable ICU 2.0 1895 */ 1896 U_CAPI void U_EXPORT2 1897 ucnv_setDefaultName(const char *name); 1898 #endif /* U_HIDE_SYSTEM_API */ 1899 1900 /** 1901 * Fixes the backslash character mismapping. For example, in SJIS, the backslash 1902 * character in the ASCII portion is also used to represent the yen currency sign. 1903 * When mapping from Unicode character 0x005C, it's unclear whether to map the 1904 * character back to yen or backslash in SJIS. This function will take the input 1905 * buffer and replace all the yen sign characters with backslash. This is necessary 1906 * when the user tries to open a file with the input buffer on Windows. 1907 * This function will test the converter to see whether such mapping is 1908 * required. You can sometimes avoid using this function by using the correct version 1909 * of Shift-JIS. 1910 * 1911 * @param cnv The converter representing the target codepage. 1912 * @param source the input buffer to be fixed 1913 * @param sourceLen the length of the input buffer 1914 * @see ucnv_isAmbiguous 1915 * @stable ICU 2.0 1916 */ 1917 U_CAPI void U_EXPORT2 1918 ucnv_fixFileSeparator(const UConverter *cnv, UChar *source, int32_t sourceLen); 1919 1920 /** 1921 * Determines if the converter contains ambiguous mappings of the same 1922 * character or not. 1923 * @param cnv the converter to be tested 1924 * @return true if the converter contains ambiguous mapping of the same 1925 * character, false otherwise. 1926 * @stable ICU 2.0 1927 */ 1928 U_CAPI UBool U_EXPORT2 1929 ucnv_isAmbiguous(const UConverter *cnv); 1930 1931 /** 1932 * Sets the converter to use fallback mappings or not. 1933 * Regardless of this flag, the converter will always use 1934 * fallbacks from Unicode Private Use code points, as well as 1935 * reverse fallbacks (to Unicode). 1936 * For details see ".ucm File Format" 1937 * in the Conversion Data chapter of the ICU User Guide: 1938 * http://www.icu-project.org/userguide/conversion-data.html#ucmformat 1939 * 1940 * @param cnv The converter to set the fallback mapping usage on. 1941 * @param usesFallback true if the user wants the converter to take advantage of the fallback 1942 * mapping, false otherwise. 1943 * @stable ICU 2.0 1944 * @see ucnv_usesFallback 1945 */ 1946 U_CAPI void U_EXPORT2 1947 ucnv_setFallback(UConverter *cnv, UBool usesFallback); 1948 1949 /** 1950 * Determines if the converter uses fallback mappings or not. 1951 * This flag has restrictions, see ucnv_setFallback(). 1952 * 1953 * @param cnv The converter to be tested 1954 * @return true if the converter uses fallback, false otherwise. 1955 * @stable ICU 2.0 1956 * @see ucnv_setFallback 1957 */ 1958 U_CAPI UBool U_EXPORT2 1959 ucnv_usesFallback(const UConverter *cnv); 1960 1961 /** 1962 * Detects Unicode signature byte sequences at the start of the byte stream 1963 * and returns the charset name of the indicated Unicode charset. 1964 * NULL is returned when no Unicode signature is recognized. 1965 * The number of bytes in the signature is output as well. 1966 * 1967 * The caller can ucnv_open() a converter using the charset name. 1968 * The first code unit (UChar) from the start of the stream will be U+FEFF 1969 * (the Unicode BOM/signature character) and can usually be ignored. 1970 * 1971 * For most Unicode charsets it is also possible to ignore the indicated 1972 * number of initial stream bytes and start converting after them. 1973 * However, there are stateful Unicode charsets (UTF-7 and BOCU-1) for which 1974 * this will not work. Therefore, it is best to ignore the first output UChar 1975 * instead of the input signature bytes. 1976 * <p> 1977 * Usage: 1978 * \snippet samples/ucnv/convsamp.cpp ucnv_detectUnicodeSignature 1979 * 1980 * @param source The source string in which the signature should be detected. 1981 * @param sourceLength Length of the input string, or -1 if terminated with a NUL byte. 1982 * @param signatureLength A pointer to int32_t to receive the number of bytes that make up the signature 1983 * of the detected UTF. 0 if not detected. 1984 * Can be a NULL pointer. 1985 * @param pErrorCode ICU error code in/out parameter. 1986 * Must fulfill U_SUCCESS before the function call. 1987 * @return The name of the encoding detected. NULL if encoding is not detected. 1988 * @stable ICU 2.4 1989 */ 1990 U_CAPI const char* U_EXPORT2 1991 ucnv_detectUnicodeSignature(const char* source, 1992 int32_t sourceLength, 1993 int32_t *signatureLength, 1994 UErrorCode *pErrorCode); 1995 1996 /** 1997 * Returns the number of UChars held in the converter's internal state 1998 * because more input is needed for completing the conversion. This function is 1999 * useful for mapping semantics of ICU's converter interface to those of iconv, 2000 * and this information is not needed for normal conversion. 2001 * @param cnv The converter in which the input is held 2002 * @param status ICU error code in/out parameter. 2003 * Must fulfill U_SUCCESS before the function call. 2004 * @return The number of UChars in the state. -1 if an error is encountered. 2005 * @stable ICU 3.4 2006 */ 2007 U_CAPI int32_t U_EXPORT2 2008 ucnv_fromUCountPending(const UConverter* cnv, UErrorCode* status); 2009 2010 /** 2011 * Returns the number of chars held in the converter's internal state 2012 * because more input is needed for completing the conversion. This function is 2013 * useful for mapping semantics of ICU's converter interface to those of iconv, 2014 * and this information is not needed for normal conversion. 2015 * @param cnv The converter in which the input is held as internal state 2016 * @param status ICU error code in/out parameter. 2017 * Must fulfill U_SUCCESS before the function call. 2018 * @return The number of chars in the state. -1 if an error is encountered. 2019 * @stable ICU 3.4 2020 */ 2021 U_CAPI int32_t U_EXPORT2 2022 ucnv_toUCountPending(const UConverter* cnv, UErrorCode* status); 2023 2024 /** 2025 * Returns whether or not the charset of the converter has a fixed number of bytes 2026 * per charset character. 2027 * An example of this are converters that are of the type UCNV_SBCS or UCNV_DBCS. 2028 * Another example is UTF-32 which is always 4 bytes per character. 2029 * A Unicode code point may be represented by more than one UTF-8 or UTF-16 code unit 2030 * but a UTF-32 converter encodes each code point with 4 bytes. 2031 * Note: This method is not intended to be used to determine whether the charset has a 2032 * fixed ratio of bytes to Unicode codes <i>units</i> for any particular Unicode encoding form. 2033 * false is returned with the UErrorCode if error occurs or cnv is NULL. 2034 * @param cnv The converter to be tested 2035 * @param status ICU error code in/out paramter 2036 * @return true if the converter is fixed-width 2037 * @stable ICU 4.8 2038 */ 2039 U_CAPI UBool U_EXPORT2 2040 ucnv_isFixedWidth(UConverter *cnv, UErrorCode *status); 2041 2042 #endif 2043 2044 #endif 2045 /*_UCNV*/ 2046