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