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