1 /* 2 ********************************************************************** 3 * Copyright (C) 1999-2011, International Business Machines 4 * Corporation and others. All Rights Reserved. 5 ********************************************************************** 6 * ucnv.h: 7 * External APIs for the ICU's codeset conversion library 8 * Bertrand A. Damiba 9 * 10 * Modification History: 11 * 12 * Date Name Description 13 * 04/04/99 helena Fixed internal header inclusion. 14 * 05/11/00 helena Added setFallback and usesFallback APIs. 15 * 06/29/2000 helena Major rewrite of the callback APIs. 16 * 12/07/2000 srl Update of documentation 17 */ 18 19 /** 20 * \file 21 * \brief C API: Character conversion 22 * 23 * <h2>Character Conversion C API</h2> 24 * 25 * <p>This API is used to convert codepage or character encoded data to and 26 * from UTF-16. You can open a converter with {@link ucnv_open() }. With that 27 * converter, you can get its properties, set options, convert your data and 28 * close the converter.</p> 29 * 30 * <p>Since many software programs recogize different converter names for 31 * different types of converters, there are other functions in this API to 32 * iterate over the converter aliases. The functions {@link ucnv_getAvailableName() }, 33 * {@link ucnv_getAlias() } and {@link ucnv_getStandardName() } are some of the 34 * more frequently used alias functions to get this information.</p> 35 * 36 * <p>When a converter encounters an illegal, irregular, invalid or unmappable character 37 * its default behavior is to use a substitution character to replace the 38 * bad byte sequence. This behavior can be changed by using {@link ucnv_setFromUCallBack() } 39 * or {@link ucnv_setToUCallBack() } on the converter. The header ucnv_err.h defines 40 * many other callback actions that can be used instead of a character substitution.</p> 41 * 42 * <p>More information about this API can be found in our 43 * <a href="http://icu-project.org/userguide/conversion.html">User's 44 * Guide</a>.</p> 45 */ 46 47 #ifndef UCNV_H 48 #define UCNV_H 49 50 #include "unicode/ucnv_err.h" 51 #include "unicode/uenum.h" 52 #include "unicode/localpointer.h" 53 54 #ifndef __USET_H__ 55 56 /** 57 * USet is the C API type for Unicode sets. 58 * It is forward-declared here to avoid including the header file if related 59 * conversion APIs are not used. 60 * See unicode/uset.h 61 * 62 * @see ucnv_getUnicodeSet 63 * @stable ICU 2.6 64 */ 65 struct USet; 66 /** @stable ICU 2.6 */ 67 typedef struct USet USet; 68 69 #endif 70 71 #if !UCONFIG_NO_CONVERSION 72 73 U_CDECL_BEGIN 74 75 /** Maximum length of a converter name including the terminating NULL @stable ICU 2.0 */ 76 #define UCNV_MAX_CONVERTER_NAME_LENGTH 60 77 /** Maximum length of a converter name including path and terminating NULL @stable ICU 2.0 */ 78 #define UCNV_MAX_FULL_FILE_NAME_LENGTH (600+UCNV_MAX_CONVERTER_NAME_LENGTH) 79 80 /** Shift in for EBDCDIC_STATEFUL and iso2022 states @stable ICU 2.0 */ 81 #define UCNV_SI 0x0F 82 /** Shift out for EBDCDIC_STATEFUL and iso2022 states @stable ICU 2.0 */ 83 #define UCNV_SO 0x0E 84 85 /** 86 * Enum for specifying basic types of converters 87 * @see ucnv_getType 88 * @stable ICU 2.0 89 */ 90 typedef enum { 91 /** @stable ICU 2.0 */ 92 UCNV_UNSUPPORTED_CONVERTER = -1, 93 /** @stable ICU 2.0 */ 94 UCNV_SBCS = 0, 95 /** @stable ICU 2.0 */ 96 UCNV_DBCS = 1, 97 /** @stable ICU 2.0 */ 98 UCNV_MBCS = 2, 99 /** @stable ICU 2.0 */ 100 UCNV_LATIN_1 = 3, 101 /** @stable ICU 2.0 */ 102 UCNV_UTF8 = 4, 103 /** @stable ICU 2.0 */ 104 UCNV_UTF16_BigEndian = 5, 105 /** @stable ICU 2.0 */ 106 UCNV_UTF16_LittleEndian = 6, 107 /** @stable ICU 2.0 */ 108 UCNV_UTF32_BigEndian = 7, 109 /** @stable ICU 2.0 */ 110 UCNV_UTF32_LittleEndian = 8, 111 /** @stable ICU 2.0 */ 112 UCNV_EBCDIC_STATEFUL = 9, 113 /** @stable ICU 2.0 */ 114 UCNV_ISO_2022 = 10, 115 116 /** @stable ICU 2.0 */ 117 UCNV_LMBCS_1 = 11, 118 /** @stable ICU 2.0 */ 119 UCNV_LMBCS_2, 120 /** @stable ICU 2.0 */ 121 UCNV_LMBCS_3, 122 /** @stable ICU 2.0 */ 123 UCNV_LMBCS_4, 124 /** @stable ICU 2.0 */ 125 UCNV_LMBCS_5, 126 /** @stable ICU 2.0 */ 127 UCNV_LMBCS_6, 128 /** @stable ICU 2.0 */ 129 UCNV_LMBCS_8, 130 /** @stable ICU 2.0 */ 131 UCNV_LMBCS_11, 132 /** @stable ICU 2.0 */ 133 UCNV_LMBCS_16, 134 /** @stable ICU 2.0 */ 135 UCNV_LMBCS_17, 136 /** @stable ICU 2.0 */ 137 UCNV_LMBCS_18, 138 /** @stable ICU 2.0 */ 139 UCNV_LMBCS_19, 140 /** @stable ICU 2.0 */ 141 UCNV_LMBCS_LAST = UCNV_LMBCS_19, 142 /** @stable ICU 2.0 */ 143 UCNV_HZ, 144 /** @stable ICU 2.0 */ 145 UCNV_SCSU, 146 /** @stable ICU 2.0 */ 147 UCNV_ISCII, 148 /** @stable ICU 2.0 */ 149 UCNV_US_ASCII, 150 /** @stable ICU 2.0 */ 151 UCNV_UTF7, 152 /** @stable ICU 2.2 */ 153 UCNV_BOCU1, 154 /** @stable ICU 2.2 */ 155 UCNV_UTF16, 156 /** @stable ICU 2.2 */ 157 UCNV_UTF32, 158 /** @stable ICU 2.2 */ 159 UCNV_CESU8, 160 /** @stable ICU 2.4 */ 161 UCNV_IMAP_MAILBOX, 162 /** @draft ICU 4.8 */ 163 UCNV_COMPOUND_TEXT, 164 165 /* Number of converter types for which we have conversion routines. */ 166 UCNV_NUMBER_OF_SUPPORTED_CONVERTER_TYPES 167 } UConverterType; 168 169 /** 170 * Enum for specifying which platform a converter ID refers to. 171 * The use of platform/CCSID is not recommended. See ucnv_openCCSID(). 172 * 173 * @see ucnv_getPlatform 174 * @see ucnv_openCCSID 175 * @see ucnv_getCCSID 176 * @stable ICU 2.0 177 */ 178 typedef enum { 179 UCNV_UNKNOWN = -1, 180 UCNV_IBM = 0 181 } UConverterPlatform; 182 183 /** 184 * Function pointer for error callback in the codepage to unicode direction. 185 * Called when an error has occured in conversion to unicode, or on open/close of the callback (see reason). 186 * @param context Pointer to the callback's private data 187 * @param args Information about the conversion in progress 188 * @param codeUnits Points to 'length' bytes of the concerned codepage sequence 189 * @param length Size (in bytes) of the concerned codepage sequence 190 * @param reason Defines the reason the callback was invoked 191 * @param pErrorCode ICU error code in/out parameter. 192 * For converter callback functions, set to a conversion error 193 * before the call, and the callback may reset it to U_ZERO_ERROR. 194 * @see ucnv_setToUCallBack 195 * @see UConverterToUnicodeArgs 196 * @stable ICU 2.0 197 */ 198 typedef void (U_EXPORT2 *UConverterToUCallback) ( 199 const void* context, 200 UConverterToUnicodeArgs *args, 201 const char *codeUnits, 202 int32_t length, 203 UConverterCallbackReason reason, 204 UErrorCode *pErrorCode); 205 206 /** 207 * Function pointer for error callback in the unicode to codepage direction. 208 * Called when an error has occured in conversion from unicode, or on open/close of the callback (see reason). 209 * @param context Pointer to the callback's private data 210 * @param args Information about the conversion in progress 211 * @param codeUnits Points to 'length' UChars of the concerned Unicode sequence 212 * @param length Size (in bytes) of the concerned codepage sequence 213 * @param codePoint Single UChar32 (UTF-32) containing the concerend Unicode codepoint. 214 * @param reason Defines the reason the callback was invoked 215 * @param pErrorCode ICU error code in/out parameter. 216 * For converter callback functions, set to a conversion error 217 * before the call, and the callback may reset it to U_ZERO_ERROR. 218 * @see ucnv_setFromUCallBack 219 * @stable ICU 2.0 220 */ 221 typedef void (U_EXPORT2 *UConverterFromUCallback) ( 222 const void* context, 223 UConverterFromUnicodeArgs *args, 224 const UChar* codeUnits, 225 int32_t length, 226 UChar32 codePoint, 227 UConverterCallbackReason reason, 228 UErrorCode *pErrorCode); 229 230 U_CDECL_END 231 232 /** 233 * Character that separates converter names from options and options from each other. 234 * @see ucnv_open 235 * @stable ICU 2.0 236 */ 237 #define UCNV_OPTION_SEP_CHAR ',' 238 239 /** 240 * String version of UCNV_OPTION_SEP_CHAR. 241 * @see ucnv_open 242 * @stable ICU 2.0 243 */ 244 #define UCNV_OPTION_SEP_STRING "," 245 246 /** 247 * Character that separates a converter option from its value. 248 * @see ucnv_open 249 * @stable ICU 2.0 250 */ 251 #define UCNV_VALUE_SEP_CHAR '=' 252 253 /** 254 * String version of UCNV_VALUE_SEP_CHAR. 255 * @see ucnv_open 256 * @stable ICU 2.0 257 */ 258 #define UCNV_VALUE_SEP_STRING "=" 259 260 /** 261 * Converter option for specifying a locale. 262 * For example, ucnv_open("SCSU,locale=ja", &errorCode); 263 * See convrtrs.txt. 264 * 265 * @see ucnv_open 266 * @stable ICU 2.0 267 */ 268 #define UCNV_LOCALE_OPTION_STRING ",locale=" 269 270 /** 271 * Converter option for specifying a version selector (0..9) for some converters. 272 * For example, 273 * \code 274 * ucnv_open("UTF-7,version=1", &errorCode); 275 * \endcode 276 * See convrtrs.txt. 277 * 278 * @see ucnv_open 279 * @stable ICU 2.4 280 */ 281 #define UCNV_VERSION_OPTION_STRING ",version=" 282 283 /** 284 * Converter option for EBCDIC SBCS or mixed-SBCS/DBCS (stateful) codepages. 285 * Swaps Unicode mappings for EBCDIC LF and NL codes, as used on 286 * S/390 (z/OS) Unix System Services (Open Edition). 287 * For example, ucnv_open("ibm-1047,swaplfnl", &errorCode); 288 * See convrtrs.txt. 289 * 290 * @see ucnv_open 291 * @stable ICU 2.4 292 */ 293 #define UCNV_SWAP_LFNL_OPTION_STRING ",swaplfnl" 294 295 /** 296 * Do a fuzzy compare of two converter/alias names. 297 * The comparison is case-insensitive, ignores leading zeroes if they are not 298 * followed by further digits, and ignores all but letters and digits. 299 * Thus the strings "UTF-8", "utf_8", "u*T@f08" and "Utf 8" are exactly equivalent. 300 * See section 1.4, Charset Alias Matching in Unicode Technical Standard #22 301 * at http://www.unicode.org/reports/tr22/ 302 * 303 * @param name1 a converter name or alias, zero-terminated 304 * @param name2 a converter name or alias, zero-terminated 305 * @return 0 if the names match, or a negative value if the name1 306 * lexically precedes name2, or a positive value if the name1 307 * lexically follows name2. 308 * @stable ICU 2.0 309 */ 310 U_STABLE int U_EXPORT2 311 ucnv_compareNames(const char *name1, const char *name2); 312 313 314 /** 315 * Creates a UConverter object with the name of a coded character set specified as a C string. 316 * The actual name will be resolved with the alias file 317 * using a case-insensitive string comparison that ignores 318 * leading zeroes and all non-alphanumeric characters. 319 * E.g., the names "UTF8", "utf-8", "u*T@f08" and "Utf 8" are all equivalent. 320 * (See also ucnv_compareNames().) 321 * If <code>NULL</code> is passed for the converter name, it will create one with the 322 * getDefaultName return value. 323 * 324 * <p>A converter name for ICU 1.5 and above may contain options 325 * like a locale specification to control the specific behavior of 326 * the newly instantiated converter. 327 * The meaning of the options depends on the particular converter. 328 * If an option is not defined for or recognized by a given converter, then it is ignored.</p> 329 * 330 * <p>Options are appended to the converter name string, with a 331 * <code>UCNV_OPTION_SEP_CHAR</code> between the name and the first option and 332 * also between adjacent options.</p> 333 * 334 * <p>If the alias is ambiguous, then the preferred converter is used 335 * and the status is set to U_AMBIGUOUS_ALIAS_WARNING.</p> 336 * 337 * <p>The conversion behavior and names can vary between platforms. ICU may 338 * convert some characters differently from other platforms. Details on this topic 339 * are in the <a href="http://icu-project.org/userguide/conversion.html">User's 340 * Guide</a>. Aliases starting with a "cp" prefix have no specific meaning 341 * other than its an alias starting with the letters "cp". Please do not 342 * associate any meaning to these aliases.</p> 343 * 344 * @param converterName Name of the coded character set table. 345 * This may have options appended to the string. 346 * IANA alias character set names, IBM CCSIDs starting with "ibm-", 347 * Windows codepage numbers starting with "windows-" are frequently 348 * used for this parameter. See ucnv_getAvailableName and 349 * ucnv_getAlias for a complete list that is available. 350 * If this parameter is NULL, the default converter will be used. 351 * @param err outgoing error status <TT>U_MEMORY_ALLOCATION_ERROR, U_FILE_ACCESS_ERROR</TT> 352 * @return the created Unicode converter object, or <TT>NULL</TT> if an error occured 353 * @see ucnv_openU 354 * @see ucnv_openCCSID 355 * @see ucnv_getAvailableName 356 * @see ucnv_getAlias 357 * @see ucnv_getDefaultName 358 * @see ucnv_close 359 * @see ucnv_compareNames 360 * @stable ICU 2.0 361 */ 362 U_STABLE UConverter* U_EXPORT2 363 ucnv_open(const char *converterName, UErrorCode *err); 364 365 366 /** 367 * Creates a Unicode converter with the names specified as unicode string. 368 * The name should be limited to the ASCII-7 alphanumerics range. 369 * The actual name will be resolved with the alias file 370 * using a case-insensitive string comparison that ignores 371 * leading zeroes and all non-alphanumeric characters. 372 * E.g., the names "UTF8", "utf-8", "u*T@f08" and "Utf 8" are all equivalent. 373 * (See also ucnv_compareNames().) 374 * If <TT>NULL</TT> is passed for the converter name, it will create 375 * one with the ucnv_getDefaultName() return value. 376 * If the alias is ambiguous, then the preferred converter is used 377 * and the status is set to U_AMBIGUOUS_ALIAS_WARNING. 378 * 379 * <p>See ucnv_open for the complete details</p> 380 * @param name Name of the UConverter table in a zero terminated 381 * Unicode string 382 * @param err outgoing error status <TT>U_MEMORY_ALLOCATION_ERROR, 383 * U_FILE_ACCESS_ERROR</TT> 384 * @return the created Unicode converter object, or <TT>NULL</TT> if an 385 * error occured 386 * @see ucnv_open 387 * @see ucnv_openCCSID 388 * @see ucnv_close 389 * @see ucnv_compareNames 390 * @stable ICU 2.0 391 */ 392 U_STABLE UConverter* U_EXPORT2 393 ucnv_openU(const UChar *name, 394 UErrorCode *err); 395 396 /** 397 * Creates a UConverter object from a CCSID number and platform pair. 398 * Note that the usefulness of this function is limited to platforms with numeric 399 * encoding IDs. Only IBM and Microsoft platforms use numeric (16-bit) identifiers for 400 * encodings. 401 * 402 * In addition, IBM CCSIDs and Unicode conversion tables are not 1:1 related. 403 * For many IBM CCSIDs there are multiple (up to six) Unicode conversion tables, and 404 * for some Unicode conversion tables there are multiple CCSIDs. 405 * Some "alternate" Unicode conversion tables are provided by the 406 * IBM CDRA conversion table registry. 407 * The most prominent example of a systematic modification of conversion tables that is 408 * not provided in the form of conversion table files in the repository is 409 * that S/390 Unix System Services swaps the codes for Line Feed and New Line in all 410 * EBCDIC codepages, which requires such a swap in the Unicode conversion tables as well. 411 * 412 * Only IBM default conversion tables are accessible with ucnv_openCCSID(). 413 * ucnv_getCCSID() will return the same CCSID for all conversion tables that are associated 414 * with that CCSID. 415 * 416 * Currently, the only "platform" supported in the ICU converter API is UCNV_IBM. 417 * 418 * In summary, the use of CCSIDs and the associated API functions is not recommended. 419 * 420 * In order to open a converter with the default IBM CDRA Unicode conversion table, 421 * you can use this function or use the prefix "ibm-": 422 * \code 423 * char name[20]; 424 * sprintf(name, "ibm-%hu", ccsid); 425 * cnv=ucnv_open(name, &errorCode); 426 * \endcode 427 * 428 * In order to open a converter with the IBM S/390 Unix System Services variant 429 * of a Unicode/EBCDIC conversion table, 430 * you can use the prefix "ibm-" together with the option string UCNV_SWAP_LFNL_OPTION_STRING: 431 * \code 432 * char name[20]; 433 * sprintf(name, "ibm-%hu" UCNV_SWAP_LFNL_OPTION_STRING, ccsid); 434 * cnv=ucnv_open(name, &errorCode); 435 * \endcode 436 * 437 * In order to open a converter from a Microsoft codepage number, use the prefix "cp": 438 * \code 439 * char name[20]; 440 * sprintf(name, "cp%hu", codepageID); 441 * cnv=ucnv_open(name, &errorCode); 442 * \endcode 443 * 444 * If the alias is ambiguous, then the preferred converter is used 445 * and the status is set to U_AMBIGUOUS_ALIAS_WARNING. 446 * 447 * @param codepage codepage number to create 448 * @param platform the platform in which the codepage number exists 449 * @param err error status <TT>U_MEMORY_ALLOCATION_ERROR, U_FILE_ACCESS_ERROR</TT> 450 * @return the created Unicode converter object, or <TT>NULL</TT> if an error 451 * occured. 452 * @see ucnv_open 453 * @see ucnv_openU 454 * @see ucnv_close 455 * @see ucnv_getCCSID 456 * @see ucnv_getPlatform 457 * @see UConverterPlatform 458 * @stable ICU 2.0 459 */ 460 U_STABLE UConverter* U_EXPORT2 461 ucnv_openCCSID(int32_t codepage, 462 UConverterPlatform platform, 463 UErrorCode * err); 464 465 /** 466 * <p>Creates a UConverter object specified from a packageName and a converterName.</p> 467 * 468 * <p>The packageName and converterName must point to an ICU udata object, as defined by 469 * <code> udata_open( packageName, "cnv", converterName, err) </code> or equivalent. 470 * Typically, packageName will refer to a (.dat) file, or to a package registered with 471 * udata_setAppData(). Using a full file or directory pathname for packageName is deprecated.</p> 472 * 473 * <p>The name will NOT be looked up in the alias mechanism, nor will the converter be 474 * stored in the converter cache or the alias table. The only way to open further converters 475 * is call this function multiple times, or use the ucnv_safeClone() function to clone a 476 * 'master' converter.</p> 477 * 478 * <p>A future version of ICU may add alias table lookups and/or caching 479 * to this function.</p> 480 * 481 * <p>Example Use: 482 * <code>cnv = ucnv_openPackage("myapp", "myconverter", &err);</code> 483 * </p> 484 * 485 * @param packageName name of the package (equivalent to 'path' in udata_open() call) 486 * @param converterName name of the data item to be used, without suffix. 487 * @param err outgoing error status <TT>U_MEMORY_ALLOCATION_ERROR, U_FILE_ACCESS_ERROR</TT> 488 * @return the created Unicode converter object, or <TT>NULL</TT> if an error occured 489 * @see udata_open 490 * @see ucnv_open 491 * @see ucnv_safeClone 492 * @see ucnv_close 493 * @stable ICU 2.2 494 */ 495 U_STABLE UConverter* U_EXPORT2 496 ucnv_openPackage(const char *packageName, const char *converterName, UErrorCode *err); 497 498 /** 499 * Thread safe converter cloning operation. 500 * For most efficient operation, pass in a stackBuffer (and a *pBufferSize) 501 * with at least U_CNV_SAFECLONE_BUFFERSIZE bytes of space. 502 * If the buffer size is sufficient, then the clone will use the stack buffer; 503 * otherwise, it will be allocated, and *pBufferSize will indicate 504 * the actual size. (This should not occur with U_CNV_SAFECLONE_BUFFERSIZE.) 505 * 506 * You must ucnv_close() the clone in any case. 507 * 508 * If *pBufferSize==0, (regardless of whether stackBuffer==NULL or not) 509 * then *pBufferSize will be changed to a sufficient size 510 * for cloning this converter, 511 * without actually cloning the converter ("pure pre-flighting"). 512 * 513 * If *pBufferSize is greater than zero but not large enough for a stack-based 514 * clone, then the converter is cloned using newly allocated memory 515 * and *pBufferSize is changed to the necessary size. 516 * 517 * If the converter clone fits into the stack buffer but the stack buffer is not 518 * sufficiently aligned for the clone, then the clone will use an 519 * adjusted pointer and use an accordingly smaller buffer size. 520 * 521 * @param cnv converter to be cloned 522 * @param stackBuffer user allocated space for the new clone. If NULL new memory will be allocated. 523 * If buffer is not large enough, new memory will be allocated. 524 * Clients can use the U_CNV_SAFECLONE_BUFFERSIZE. This will probably be enough to avoid memory allocations. 525 * @param pBufferSize pointer to size of allocated space. pBufferSize must not be NULL. 526 * @param status to indicate whether the operation went on smoothly or there were errors 527 * An informational status value, U_SAFECLONE_ALLOCATED_WARNING, 528 * is used if any allocations were necessary. 529 * However, it is better to check if *pBufferSize grew for checking for 530 * allocations because warning codes can be overridden by subsequent 531 * function calls. 532 * @return pointer to the new clone 533 * @stable ICU 2.0 534 */ 535 U_STABLE UConverter * U_EXPORT2 536 ucnv_safeClone(const UConverter *cnv, 537 void *stackBuffer, 538 int32_t *pBufferSize, 539 UErrorCode *status); 540 541 /** 542 * \def U_CNV_SAFECLONE_BUFFERSIZE 543 * Definition of a buffer size that is designed to be large enough for 544 * converters to be cloned with ucnv_safeClone(). 545 * @stable ICU 2.0 546 */ 547 #define U_CNV_SAFECLONE_BUFFERSIZE 1024 548 549 /** 550 * Deletes the unicode converter and releases resources associated 551 * with just this instance. 552 * Does not free up shared converter tables. 553 * 554 * @param converter the converter object to be deleted 555 * @see ucnv_open 556 * @see ucnv_openU 557 * @see ucnv_openCCSID 558 * @stable ICU 2.0 559 */ 560 U_STABLE void U_EXPORT2 561 ucnv_close(UConverter * converter); 562 563 #if U_SHOW_CPLUSPLUS_API 564 565 U_NAMESPACE_BEGIN 566 567 /** 568 * \class LocalUConverterPointer 569 * "Smart pointer" class, closes a UConverter via ucnv_close(). 570 * For most methods see the LocalPointerBase base class. 571 * 572 * @see LocalPointerBase 573 * @see LocalPointer 574 * @stable ICU 4.4 575 */ 576 U_DEFINE_LOCAL_OPEN_POINTER(LocalUConverterPointer, UConverter, ucnv_close); 577 578 U_NAMESPACE_END 579 580 #endif 581 582 /** 583 * Fills in the output parameter, subChars, with the substitution characters 584 * as multiple bytes. 585 * If ucnv_setSubstString() set a Unicode string because the converter is 586 * stateful, then subChars will be an empty string. 587 * 588 * @param converter the Unicode converter 589 * @param subChars the subsitution characters 590 * @param len on input the capacity of subChars, on output the number 591 * of bytes copied to it 592 * @param err the outgoing error status code. 593 * If the substitution character array is too small, an 594 * <TT>U_INDEX_OUTOFBOUNDS_ERROR</TT> will be returned. 595 * @see ucnv_setSubstString 596 * @see ucnv_setSubstChars 597 * @stable ICU 2.0 598 */ 599 U_STABLE void U_EXPORT2 600 ucnv_getSubstChars(const UConverter *converter, 601 char *subChars, 602 int8_t *len, 603 UErrorCode *err); 604 605 /** 606 * Sets the substitution chars when converting from unicode to a codepage. The 607 * substitution is specified as a string of 1-4 bytes, and may contain 608 * <TT>NULL</TT> bytes. 609 * The subChars must represent a single character. The caller needs to know the 610 * byte sequence of a valid character in the converter's charset. 611 * For some converters, for example some ISO 2022 variants, only single-byte 612 * substitution characters may be supported. 613 * The newer ucnv_setSubstString() function relaxes these limitations. 614 * 615 * @param converter the Unicode converter 616 * @param subChars the substitution character byte sequence we want set 617 * @param len the number of bytes in subChars 618 * @param err the error status code. <TT>U_INDEX_OUTOFBOUNDS_ERROR </TT> if 619 * len is bigger than the maximum number of bytes allowed in subchars 620 * @see ucnv_setSubstString 621 * @see ucnv_getSubstChars 622 * @stable ICU 2.0 623 */ 624 U_STABLE void U_EXPORT2 625 ucnv_setSubstChars(UConverter *converter, 626 const char *subChars, 627 int8_t len, 628 UErrorCode *err); 629 630 /** 631 * Set a substitution string for converting from Unicode to a charset. 632 * The caller need not know the charset byte sequence for each charset. 633 * 634 * Unlike ucnv_setSubstChars() which is designed to set a charset byte sequence 635 * for a single character, this function takes a Unicode string with 636 * zero, one or more characters, and immediately verifies that the string can be 637 * converted to the charset. 638 * If not, or if the result is too long (more than 32 bytes as of ICU 3.6), 639 * then the function returns with an error accordingly. 640 * 641 * Also unlike ucnv_setSubstChars(), this function works for stateful charsets 642 * by converting on the fly at the point of substitution rather than setting 643 * a fixed byte sequence. 644 * 645 * @param cnv The UConverter object. 646 * @param s The Unicode string. 647 * @param length The number of UChars in s, or -1 for a NUL-terminated string. 648 * @param err Pointer to a standard ICU error code. Its input value must 649 * pass the U_SUCCESS() test, or else the function returns 650 * immediately. Check for U_FAILURE() on output or use with 651 * function chaining. (See User Guide for details.) 652 * 653 * @see ucnv_setSubstChars 654 * @see ucnv_getSubstChars 655 * @stable ICU 3.6 656 */ 657 U_STABLE void U_EXPORT2 658 ucnv_setSubstString(UConverter *cnv, 659 const UChar *s, 660 int32_t length, 661 UErrorCode *err); 662 663 /** 664 * Fills in the output parameter, errBytes, with the error characters from the 665 * last failing conversion. 666 * 667 * @param converter the Unicode converter 668 * @param errBytes the codepage bytes which were in error 669 * @param len on input the capacity of errBytes, on output the number of 670 * bytes which were copied to it 671 * @param err the error status code. 672 * If the substitution character array is too small, an 673 * <TT>U_INDEX_OUTOFBOUNDS_ERROR</TT> will be returned. 674 * @stable ICU 2.0 675 */ 676 U_STABLE void U_EXPORT2 677 ucnv_getInvalidChars(const UConverter *converter, 678 char *errBytes, 679 int8_t *len, 680 UErrorCode *err); 681 682 /** 683 * Fills in the output parameter, errChars, with the error characters from the 684 * last failing conversion. 685 * 686 * @param converter the Unicode converter 687 * @param errUChars the UChars which were in error 688 * @param len on input the capacity of errUChars, on output the number of 689 * UChars which were copied to it 690 * @param err the error status code. 691 * If the substitution character array is too small, an 692 * <TT>U_INDEX_OUTOFBOUNDS_ERROR</TT> will be returned. 693 * @stable ICU 2.0 694 */ 695 U_STABLE void U_EXPORT2 696 ucnv_getInvalidUChars(const UConverter *converter, 697 UChar *errUChars, 698 int8_t *len, 699 UErrorCode *err); 700 701 /** 702 * Resets the state of a converter to the default state. This is used 703 * in the case of an error, to restart a conversion from a known default state. 704 * It will also empty the internal output buffers. 705 * @param converter the Unicode converter 706 * @stable ICU 2.0 707 */ 708 U_STABLE void U_EXPORT2 709 ucnv_reset(UConverter *converter); 710 711 /** 712 * Resets the to-Unicode part of a converter state to the default state. 713 * This is used in the case of an error to restart a conversion to 714 * Unicode to a known default state. It will also empty the internal 715 * output buffers used for the conversion to Unicode codepoints. 716 * @param converter the Unicode converter 717 * @stable ICU 2.0 718 */ 719 U_STABLE void U_EXPORT2 720 ucnv_resetToUnicode(UConverter *converter); 721 722 /** 723 * Resets the from-Unicode part of a converter state to the default state. 724 * This is used in the case of an error to restart a conversion from 725 * Unicode to a known default state. It will also empty the internal output 726 * buffers used for the conversion from Unicode codepoints. 727 * @param converter the Unicode converter 728 * @stable ICU 2.0 729 */ 730 U_STABLE void U_EXPORT2 731 ucnv_resetFromUnicode(UConverter *converter); 732 733 /** 734 * Returns the maximum number of bytes that are output per UChar in conversion 735 * from Unicode using this converter. 736 * The returned number can be used with UCNV_GET_MAX_BYTES_FOR_STRING 737 * to calculate the size of a target buffer for conversion from Unicode. 738 * 739 * Note: Before ICU 2.8, this function did not return reliable numbers for 740 * some stateful converters (EBCDIC_STATEFUL, ISO-2022) and LMBCS. 741 * 742 * This number may not be the same as the maximum number of bytes per 743 * "conversion unit". In other words, it may not be the intuitively expected 744 * number of bytes per character that would be published for a charset, 745 * and may not fulfill any other purpose than the allocation of an output 746 * buffer of guaranteed sufficient size for a given input length and converter. 747 * 748 * Examples for special cases that are taken into account: 749 * - Supplementary code points may convert to more bytes than BMP code points. 750 * This function returns bytes per UChar (UTF-16 code unit), not per 751 * Unicode code point, for efficient buffer allocation. 752 * - State-shifting output (SI/SO, escapes, etc.) from stateful converters. 753 * - When m input UChars are converted to n output bytes, then the maximum m/n 754 * is taken into account. 755 * 756 * The number returned here does not take into account 757 * (see UCNV_GET_MAX_BYTES_FOR_STRING): 758 * - callbacks which output more than one charset character sequence per call, 759 * like escape callbacks 760 * - initial and final non-character bytes that are output by some converters 761 * (automatic BOMs, initial escape sequence, final SI, etc.) 762 * 763 * Examples for returned values: 764 * - SBCS charsets: 1 765 * - Shift-JIS: 2 766 * - UTF-16: 2 (2 per BMP, 4 per surrogate _pair_, BOM not counted) 767 * - UTF-8: 3 (3 per BMP, 4 per surrogate _pair_) 768 * - EBCDIC_STATEFUL (EBCDIC mixed SBCS/DBCS): 3 (SO + DBCS) 769 * - ISO-2022: 3 (always outputs UTF-8) 770 * - ISO-2022-JP: 6 (4-byte escape sequences + DBCS) 771 * - ISO-2022-CN: 8 (4-byte designator sequences + 2-byte SS2/SS3 + DBCS) 772 * 773 * @param converter The Unicode converter. 774 * @return The maximum number of bytes per UChar that are output by ucnv_fromUnicode(), 775 * to be used together with UCNV_GET_MAX_BYTES_FOR_STRING for buffer allocation. 776 * 777 * @see UCNV_GET_MAX_BYTES_FOR_STRING 778 * @see ucnv_getMinCharSize 779 * @stable ICU 2.0 780 */ 781 U_STABLE int8_t U_EXPORT2 782 ucnv_getMaxCharSize(const UConverter *converter); 783 784 /** 785 * Calculates the size of a buffer for conversion from Unicode to a charset. 786 * The calculated size is guaranteed to be sufficient for this conversion. 787 * 788 * It takes into account initial and final non-character bytes that are output 789 * by some converters. 790 * It does not take into account callbacks which output more than one charset 791 * character sequence per call, like escape callbacks. 792 * The default (substitution) callback only outputs one charset character sequence. 793 * 794 * @param length Number of UChars to be converted. 795 * @param maxCharSize Return value from ucnv_getMaxCharSize() for the converter 796 * that will be used. 797 * @return Size of a buffer that will be large enough to hold the output bytes of 798 * converting length UChars with the converter that returned the maxCharSize. 799 * 800 * @see ucnv_getMaxCharSize 801 * @stable ICU 2.8 802 */ 803 #define UCNV_GET_MAX_BYTES_FOR_STRING(length, maxCharSize) \ 804 (((int32_t)(length)+10)*(int32_t)(maxCharSize)) 805 806 /** 807 * Returns the minimum byte length for characters in this codepage. 808 * This is usually either 1 or 2. 809 * @param converter the Unicode converter 810 * @return the minimum number of bytes allowed by this particular converter 811 * @see ucnv_getMaxCharSize 812 * @stable ICU 2.0 813 */ 814 U_STABLE int8_t U_EXPORT2 815 ucnv_getMinCharSize(const UConverter *converter); 816 817 /** 818 * Returns the display name of the converter passed in based on the Locale 819 * passed in. If the locale contains no display name, the internal ASCII 820 * name will be filled in. 821 * 822 * @param converter the Unicode converter. 823 * @param displayLocale is the specific Locale we want to localised for 824 * @param displayName user provided buffer to be filled in 825 * @param displayNameCapacity size of displayName Buffer 826 * @param err error status code 827 * @return displayNameLength number of UChar needed in displayName 828 * @see ucnv_getName 829 * @stable ICU 2.0 830 */ 831 U_STABLE int32_t U_EXPORT2 832 ucnv_getDisplayName(const UConverter *converter, 833 const char *displayLocale, 834 UChar *displayName, 835 int32_t displayNameCapacity, 836 UErrorCode *err); 837 838 /** 839 * Gets the internal, canonical name of the converter (zero-terminated). 840 * The lifetime of the returned string will be that of the converter 841 * passed to this function. 842 * @param converter the Unicode converter 843 * @param err UErrorCode status 844 * @return the internal name of the converter 845 * @see ucnv_getDisplayName 846 * @stable ICU 2.0 847 */ 848 U_STABLE const char * U_EXPORT2 849 ucnv_getName(const UConverter *converter, UErrorCode *err); 850 851 /** 852 * Gets a codepage number associated with the converter. This is not guaranteed 853 * to be the one used to create the converter. Some converters do not represent 854 * platform registered codepages and return zero for the codepage number. 855 * The error code fill-in parameter indicates if the codepage number 856 * is available. 857 * Does not check if the converter is <TT>NULL</TT> or if converter's data 858 * table is <TT>NULL</TT>. 859 * 860 * Important: The use of CCSIDs is not recommended because it is limited 861 * to only two platforms in principle and only one (UCNV_IBM) in the current 862 * ICU converter API. 863 * Also, CCSIDs are insufficient to identify IBM Unicode conversion tables precisely. 864 * For more details see ucnv_openCCSID(). 865 * 866 * @param converter the Unicode converter 867 * @param err the error status code. 868 * @return If any error occurrs, -1 will be returned otherwise, the codepage number 869 * will be returned 870 * @see ucnv_openCCSID 871 * @see ucnv_getPlatform 872 * @stable ICU 2.0 873 */ 874 U_STABLE int32_t U_EXPORT2 875 ucnv_getCCSID(const UConverter *converter, 876 UErrorCode *err); 877 878 /** 879 * Gets a codepage platform associated with the converter. Currently, 880 * only <TT>UCNV_IBM</TT> will be returned. 881 * Does not test if the converter is <TT>NULL</TT> or if converter's data 882 * table is <TT>NULL</TT>. 883 * @param converter the Unicode converter 884 * @param err the error status code. 885 * @return The codepage platform 886 * @stable ICU 2.0 887 */ 888 U_STABLE UConverterPlatform U_EXPORT2 889 ucnv_getPlatform(const UConverter *converter, 890 UErrorCode *err); 891 892 /** 893 * Gets the type of the converter 894 * e.g. SBCS, MBCS, DBCS, UTF8, UTF16_BE, UTF16_LE, ISO_2022, 895 * EBCDIC_STATEFUL, LATIN_1 896 * @param converter a valid, opened converter 897 * @return the type of the converter 898 * @stable ICU 2.0 899 */ 900 U_STABLE UConverterType U_EXPORT2 901 ucnv_getType(const UConverter * converter); 902 903 /** 904 * Gets the "starter" (lead) bytes for converters of type MBCS. 905 * Will fill in an <TT>U_ILLEGAL_ARGUMENT_ERROR</TT> if converter passed in 906 * is not MBCS. Fills in an array of type UBool, with the value of the byte 907 * as offset to the array. For example, if (starters[0x20] == TRUE) at return, 908 * it means that the byte 0x20 is a starter byte in this converter. 909 * Context pointers are always owned by the caller. 910 * 911 * @param converter a valid, opened converter of type MBCS 912 * @param starters an array of size 256 to be filled in 913 * @param err error status, <TT>U_ILLEGAL_ARGUMENT_ERROR</TT> if the 914 * converter is not a type which can return starters. 915 * @see ucnv_getType 916 * @stable ICU 2.0 917 */ 918 U_STABLE void U_EXPORT2 919 ucnv_getStarters(const UConverter* converter, 920 UBool starters[256], 921 UErrorCode* err); 922 923 924 /** 925 * Selectors for Unicode sets that can be returned by ucnv_getUnicodeSet(). 926 * @see ucnv_getUnicodeSet 927 * @stable ICU 2.6 928 */ 929 typedef enum UConverterUnicodeSet { 930 /** Select the set of roundtrippable Unicode code points. @stable ICU 2.6 */ 931 UCNV_ROUNDTRIP_SET, 932 /** Select the set of Unicode code points with roundtrip or fallback mappings. @stable ICU 4.0 */ 933 UCNV_ROUNDTRIP_AND_FALLBACK_SET, 934 /** Number of UConverterUnicodeSet selectors. @stable ICU 2.6 */ 935 UCNV_SET_COUNT 936 } UConverterUnicodeSet; 937 938 939 /** 940 * Returns the set of Unicode code points that can be converted by an ICU converter. 941 * 942 * Returns one of several kinds of set: 943 * 944 * 1. UCNV_ROUNDTRIP_SET 945 * 946 * The set of all Unicode code points that can be roundtrip-converted 947 * (converted without any data loss) with the converter (ucnv_fromUnicode()). 948 * This set will not include code points that have fallback mappings 949 * or are only the result of reverse fallback mappings. 950 * This set will also not include PUA code points with fallbacks, although 951 * ucnv_fromUnicode() will always uses those mappings despite ucnv_setFallback(). 952 * See UTR #22 "Character Mapping Markup Language" 953 * at http://www.unicode.org/reports/tr22/ 954 * 955 * This is useful for example for 956 * - checking that a string or document can be roundtrip-converted with a converter, 957 * without/before actually performing the conversion 958 * - testing if a converter can be used for text for typical text for a certain locale, 959 * by comparing its roundtrip set with the set of ExemplarCharacters from 960 * ICU's locale data or other sources 961 * 962 * 2. UCNV_ROUNDTRIP_AND_FALLBACK_SET 963 * 964 * The set of all Unicode code points that can be converted with the converter (ucnv_fromUnicode()) 965 * when fallbacks are turned on (see ucnv_setFallback()). 966 * This set includes all code points with roundtrips and fallbacks (but not reverse fallbacks). 967 * 968 * In the future, there may be more UConverterUnicodeSet choices to select 969 * sets with different properties. 970 * 971 * @param cnv The converter for which a set is requested. 972 * @param setFillIn A valid USet *. It will be cleared by this function before 973 * the converter's specific set is filled into the USet. 974 * @param whichSet A UConverterUnicodeSet selector; 975 * currently UCNV_ROUNDTRIP_SET is the only supported value. 976 * @param pErrorCode ICU error code in/out parameter. 977 * Must fulfill U_SUCCESS before the function call. 978 * 979 * @see UConverterUnicodeSet 980 * @see uset_open 981 * @see uset_close 982 * @stable ICU 2.6 983 */ 984 U_STABLE void U_EXPORT2 985 ucnv_getUnicodeSet(const UConverter *cnv, 986 USet *setFillIn, 987 UConverterUnicodeSet whichSet, 988 UErrorCode *pErrorCode); 989 990 /** 991 * Gets the current calback function used by the converter when an illegal 992 * or invalid codepage sequence is found. 993 * Context pointers are always owned by the caller. 994 * 995 * @param converter the unicode converter 996 * @param action fillin: returns the callback function pointer 997 * @param context fillin: returns the callback's private void* context 998 * @see ucnv_setToUCallBack 999 * @stable ICU 2.0 1000 */ 1001 U_STABLE void U_EXPORT2 1002 ucnv_getToUCallBack (const UConverter * converter, 1003 UConverterToUCallback *action, 1004 const void **context); 1005 1006 /** 1007 * Gets the current callback function used by the converter when illegal 1008 * or invalid Unicode sequence is found. 1009 * Context pointers are always owned by the caller. 1010 * 1011 * @param converter the unicode converter 1012 * @param action fillin: returns the callback function pointer 1013 * @param context fillin: returns the callback's private void* context 1014 * @see ucnv_setFromUCallBack 1015 * @stable ICU 2.0 1016 */ 1017 U_STABLE void U_EXPORT2 1018 ucnv_getFromUCallBack (const UConverter * converter, 1019 UConverterFromUCallback *action, 1020 const void **context); 1021 1022 /** 1023 * Changes the callback function used by the converter when 1024 * an illegal or invalid sequence is found. 1025 * Context pointers are always owned by the caller. 1026 * Predefined actions and contexts can be found in the ucnv_err.h header. 1027 * 1028 * @param converter the unicode converter 1029 * @param newAction the new callback function 1030 * @param newContext the new toUnicode callback context pointer. This can be NULL. 1031 * @param oldAction fillin: returns the old callback function pointer. This can be NULL. 1032 * @param oldContext fillin: returns the old callback's private void* context. This can be NULL. 1033 * @param err The error code status 1034 * @see ucnv_getToUCallBack 1035 * @stable ICU 2.0 1036 */ 1037 U_STABLE void U_EXPORT2 1038 ucnv_setToUCallBack (UConverter * converter, 1039 UConverterToUCallback newAction, 1040 const void* newContext, 1041 UConverterToUCallback *oldAction, 1042 const void** oldContext, 1043 UErrorCode * err); 1044 1045 /** 1046 * Changes the current callback function used by the converter when 1047 * an illegal or invalid sequence is found. 1048 * Context pointers are always owned by the caller. 1049 * Predefined actions and contexts can be found in the ucnv_err.h header. 1050 * 1051 * @param converter the unicode converter 1052 * @param newAction the new callback function 1053 * @param newContext the new fromUnicode callback context pointer. This can be NULL. 1054 * @param oldAction fillin: returns the old callback function pointer. This can be NULL. 1055 * @param oldContext fillin: returns the old callback's private void* context. This can be NULL. 1056 * @param err The error code status 1057 * @see ucnv_getFromUCallBack 1058 * @stable ICU 2.0 1059 */ 1060 U_STABLE void U_EXPORT2 1061 ucnv_setFromUCallBack (UConverter * converter, 1062 UConverterFromUCallback newAction, 1063 const void *newContext, 1064 UConverterFromUCallback *oldAction, 1065 const void **oldContext, 1066 UErrorCode * err); 1067 1068 /** 1069 * Converts an array of unicode characters to an array of codepage 1070 * characters. This function is optimized for converting a continuous 1071 * stream of data in buffer-sized chunks, where the entire source and 1072 * target does not fit in available buffers. 1073 * 1074 * The source pointer is an in/out parameter. It starts out pointing where the 1075 * conversion is to begin, and ends up pointing after the last UChar consumed. 1076 * 1077 * Target similarly starts out pointer at the first available byte in the output 1078 * buffer, and ends up pointing after the last byte written to the output. 1079 * 1080 * The converter always attempts to consume the entire source buffer, unless 1081 * (1.) the target buffer is full, or (2.) a failing error is returned from the 1082 * current callback function. When a successful error status has been 1083 * returned, it means that all of the source buffer has been 1084 * consumed. At that point, the caller should reset the source and 1085 * sourceLimit pointers to point to the next chunk. 1086 * 1087 * At the end of the stream (flush==TRUE), the input is completely consumed 1088 * when *source==sourceLimit and no error code is set. 1089 * The converter object is then automatically reset by this function. 1090 * (This means that a converter need not be reset explicitly between data 1091 * streams if it finishes the previous stream without errors.) 1092 * 1093 * This is a <I>stateful</I> conversion. Additionally, even when all source data has 1094 * been consumed, some data may be in the converters' internal state. 1095 * Call this function repeatedly, updating the target pointers with 1096 * the next empty chunk of target in case of a 1097 * <TT>U_BUFFER_OVERFLOW_ERROR</TT>, and updating the source pointers 1098 * with the next chunk of source when a successful error status is 1099 * returned, until there are no more chunks of source data. 1100 * @param converter the Unicode converter 1101 * @param target I/O parameter. Input : Points to the beginning of the buffer to copy 1102 * codepage characters to. Output : points to after the last codepage character copied 1103 * to <TT>target</TT>. 1104 * @param targetLimit the pointer just after last of the <TT>target</TT> buffer 1105 * @param source I/O parameter, pointer to pointer to the source Unicode character buffer. 1106 * @param sourceLimit the pointer just after the last of the source buffer 1107 * @param offsets if NULL is passed, nothing will happen to it, otherwise it needs to have the same number 1108 * of allocated cells as <TT>target</TT>. Will fill in offsets from target to source pointer 1109 * 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> 1110 * For output data carried across calls, and other data without a specific source character 1111 * (such as from escape sequences or callbacks) -1 will be placed for offsets. 1112 * @param flush set to <TT>TRUE</TT> if the current source buffer is the last available 1113 * chunk of the source, <TT>FALSE</TT> otherwise. Note that if a failing status is returned, 1114 * this function may have to be called multiple times with flush set to <TT>TRUE</TT> until 1115 * the source buffer is consumed. 1116 * @param err the error status. <TT>U_ILLEGAL_ARGUMENT_ERROR</TT> will be set if the 1117 * converter is <TT>NULL</TT>. 1118 * <code>U_BUFFER_OVERFLOW_ERROR</code> will be set if the target is full and there is 1119 * still data to be written to the target. 1120 * @see ucnv_fromUChars 1121 * @see ucnv_convert 1122 * @see ucnv_getMinCharSize 1123 * @see ucnv_setToUCallBack 1124 * @stable ICU 2.0 1125 */ 1126 U_STABLE void U_EXPORT2 1127 ucnv_fromUnicode (UConverter * converter, 1128 char **target, 1129 const char *targetLimit, 1130 const UChar ** source, 1131 const UChar * sourceLimit, 1132 int32_t* offsets, 1133 UBool flush, 1134 UErrorCode * err); 1135 1136 /** 1137 * Converts a buffer of codepage bytes into an array of unicode UChars 1138 * characters. This function is optimized for converting a continuous 1139 * stream of data in buffer-sized chunks, where the entire source and 1140 * target does not fit in available buffers. 1141 * 1142 * The source pointer is an in/out parameter. It starts out pointing where the 1143 * conversion is to begin, and ends up pointing after the last byte of source consumed. 1144 * 1145 * Target similarly starts out pointer at the first available UChar in the output 1146 * buffer, and ends up pointing after the last UChar written to the output. 1147 * It does NOT necessarily keep UChar sequences together. 1148 * 1149 * The converter always attempts to consume the entire source buffer, unless 1150 * (1.) the target buffer is full, or (2.) a failing error is returned from the 1151 * current callback function. When a successful error status has been 1152 * returned, it means that all of the source buffer has been 1153 * consumed. At that point, the caller should reset the source and 1154 * sourceLimit pointers to point to the next chunk. 1155 * 1156 * At the end of the stream (flush==TRUE), the input is completely consumed 1157 * when *source==sourceLimit and no error code is set 1158 * The converter object is then automatically reset by this function. 1159 * (This means that a converter need not be reset explicitly between data 1160 * streams if it finishes the previous stream without errors.) 1161 * 1162 * This is a <I>stateful</I> conversion. Additionally, even when all source data has 1163 * been consumed, some data may be in the converters' internal state. 1164 * Call this function repeatedly, updating the target pointers with 1165 * the next empty chunk of target in case of a 1166 * <TT>U_BUFFER_OVERFLOW_ERROR</TT>, and updating the source pointers 1167 * with the next chunk of source when a successful error status is 1168 * returned, until there are no more chunks of source data. 1169 * @param converter the Unicode converter 1170 * @param target I/O parameter. Input : Points to the beginning of the buffer to copy 1171 * UChars into. Output : points to after the last UChar copied. 1172 * @param targetLimit the pointer just after the end of the <TT>target</TT> buffer 1173 * @param source I/O parameter, pointer to pointer to the source codepage buffer. 1174 * @param sourceLimit the pointer to the byte after the end of the source buffer 1175 * @param offsets if NULL is passed, nothing will happen to it, otherwise it needs to have the same number 1176 * of allocated cells as <TT>target</TT>. Will fill in offsets from target to source pointer 1177 * 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> 1178 * For output data carried across calls, and other data without a specific source character 1179 * (such as from escape sequences or callbacks) -1 will be placed for offsets. 1180 * @param flush set to <TT>TRUE</TT> if the current source buffer is the last available 1181 * chunk of the source, <TT>FALSE</TT> otherwise. Note that if a failing status is returned, 1182 * this function may have to be called multiple times with flush set to <TT>TRUE</TT> until 1183 * the source buffer is consumed. 1184 * @param err the error status. <TT>U_ILLEGAL_ARGUMENT_ERROR</TT> will be set if the 1185 * converter is <TT>NULL</TT>. 1186 * <code>U_BUFFER_OVERFLOW_ERROR</code> will be set if the target is full and there is 1187 * still data to be written to the target. 1188 * @see ucnv_fromUChars 1189 * @see ucnv_convert 1190 * @see ucnv_getMinCharSize 1191 * @see ucnv_setFromUCallBack 1192 * @see ucnv_getNextUChar 1193 * @stable ICU 2.0 1194 */ 1195 U_STABLE void U_EXPORT2 1196 ucnv_toUnicode(UConverter *converter, 1197 UChar **target, 1198 const UChar *targetLimit, 1199 const char **source, 1200 const char *sourceLimit, 1201 int32_t *offsets, 1202 UBool flush, 1203 UErrorCode *err); 1204 1205 /** 1206 * Convert the Unicode string into a codepage string using an existing UConverter. 1207 * The output string is NUL-terminated if possible. 1208 * 1209 * This function is a more convenient but less powerful version of ucnv_fromUnicode(). 1210 * It is only useful for whole strings, not for streaming conversion. 1211 * 1212 * The maximum output buffer capacity required (barring output from callbacks) will be 1213 * UCNV_GET_MAX_BYTES_FOR_STRING(srcLength, ucnv_getMaxCharSize(cnv)). 1214 * 1215 * @param cnv the converter object to be used (ucnv_resetFromUnicode() will be called) 1216 * @param src the input Unicode string 1217 * @param srcLength the input string length, or -1 if NUL-terminated 1218 * @param dest destination string buffer, can be NULL if destCapacity==0 1219 * @param destCapacity the number of chars available at dest 1220 * @param pErrorCode normal ICU error code; 1221 * common error codes that may be set by this function include 1222 * U_BUFFER_OVERFLOW_ERROR, U_STRING_NOT_TERMINATED_WARNING, 1223 * U_ILLEGAL_ARGUMENT_ERROR, and conversion errors 1224 * @return the length of the output string, not counting the terminating NUL; 1225 * if the length is greater than destCapacity, then the string will not fit 1226 * and a buffer of the indicated length would need to be passed in 1227 * @see ucnv_fromUnicode 1228 * @see ucnv_convert 1229 * @see UCNV_GET_MAX_BYTES_FOR_STRING 1230 * @stable ICU 2.0 1231 */ 1232 U_STABLE int32_t U_EXPORT2 1233 ucnv_fromUChars(UConverter *cnv, 1234 char *dest, int32_t destCapacity, 1235 const UChar *src, int32_t srcLength, 1236 UErrorCode *pErrorCode); 1237 1238 /** 1239 * Convert the codepage string into a Unicode string using an existing UConverter. 1240 * The output string is NUL-terminated if possible. 1241 * 1242 * This function is a more convenient but less powerful version of ucnv_toUnicode(). 1243 * It is only useful for whole strings, not for streaming conversion. 1244 * 1245 * The maximum output buffer capacity required (barring output from callbacks) will be 1246 * 2*srcLength (each char may be converted into a surrogate pair). 1247 * 1248 * @param cnv the converter object to be used (ucnv_resetToUnicode() will be called) 1249 * @param src the input codepage string 1250 * @param srcLength the input string length, or -1 if NUL-terminated 1251 * @param dest destination string buffer, can be NULL if destCapacity==0 1252 * @param destCapacity the number of UChars available at dest 1253 * @param pErrorCode normal ICU error code; 1254 * common error codes that may be set by this function include 1255 * U_BUFFER_OVERFLOW_ERROR, U_STRING_NOT_TERMINATED_WARNING, 1256 * U_ILLEGAL_ARGUMENT_ERROR, and conversion errors 1257 * @return the length of the output string, not counting the terminating NUL; 1258 * if the length is greater than destCapacity, then the string will not fit 1259 * and a buffer of the indicated length would need to be passed in 1260 * @see ucnv_toUnicode 1261 * @see ucnv_convert 1262 * @stable ICU 2.0 1263 */ 1264 U_STABLE int32_t U_EXPORT2 1265 ucnv_toUChars(UConverter *cnv, 1266 UChar *dest, int32_t destCapacity, 1267 const char *src, int32_t srcLength, 1268 UErrorCode *pErrorCode); 1269 1270 /** 1271 * Convert a codepage buffer into Unicode one character at a time. 1272 * The input is completely consumed when the U_INDEX_OUTOFBOUNDS_ERROR is set. 1273 * 1274 * Advantage compared to ucnv_toUnicode() or ucnv_toUChars(): 1275 * - Faster for small amounts of data, for most converters, e.g., 1276 * US-ASCII, ISO-8859-1, UTF-8/16/32, and most "normal" charsets. 1277 * (For complex converters, e.g., SCSU, UTF-7 and ISO 2022 variants, 1278 * it uses ucnv_toUnicode() internally.) 1279 * - Convenient. 1280 * 1281 * Limitations compared to ucnv_toUnicode(): 1282 * - Always assumes flush=TRUE. 1283 * This makes ucnv_getNextUChar() unsuitable for "streaming" conversion, 1284 * that is, for where the input is supplied in multiple buffers, 1285 * because ucnv_getNextUChar() will assume the end of the input at the end 1286 * of the first buffer. 1287 * - Does not provide offset output. 1288 * 1289 * It is possible to "mix" ucnv_getNextUChar() and ucnv_toUnicode() because 1290 * ucnv_getNextUChar() uses the current state of the converter 1291 * (unlike ucnv_toUChars() which always resets first). 1292 * However, if ucnv_getNextUChar() is called after ucnv_toUnicode() 1293 * stopped in the middle of a character sequence (with flush=FALSE), 1294 * then ucnv_getNextUChar() will always use the slower ucnv_toUnicode() 1295 * internally until the next character boundary. 1296 * (This is new in ICU 2.6. In earlier releases, ucnv_getNextUChar() had to 1297 * start at a character boundary.) 1298 * 1299 * Instead of using ucnv_getNextUChar(), it is recommended 1300 * to convert using ucnv_toUnicode() or ucnv_toUChars() 1301 * and then iterate over the text using U16_NEXT() or a UCharIterator (uiter.h) 1302 * or a C++ CharacterIterator or similar. 1303 * This allows streaming conversion and offset output, for example. 1304 * 1305 * <p>Handling of surrogate pairs and supplementary-plane code points:<br> 1306 * There are two different kinds of codepages that provide mappings for surrogate characters: 1307 * <ul> 1308 * <li>Codepages like UTF-8, UTF-32, and GB 18030 provide direct representations for Unicode 1309 * code points U+10000-U+10ffff as well as for single surrogates U+d800-U+dfff. 1310 * Each valid sequence will result in exactly one returned code point. 1311 * If a sequence results in a single surrogate, then that will be returned 1312 * by itself, even if a neighboring sequence encodes the matching surrogate.</li> 1313 * <li>Codepages like SCSU and LMBCS (and UTF-16) provide direct representations only for BMP code points 1314 * including surrogates. Code points in supplementary planes are represented with 1315 * two sequences, each encoding a surrogate. 1316 * For these codepages, matching pairs of surrogates will be combined into single 1317 * code points for returning from this function. 1318 * (Note that SCSU is actually a mix of these codepage types.)</li> 1319 * </ul></p> 1320 * 1321 * @param converter an open UConverter 1322 * @param source the address of a pointer to the codepage buffer, will be 1323 * updated to point after the bytes consumed in the conversion call. 1324 * @param sourceLimit points to the end of the input buffer 1325 * @param err fills in error status (see ucnv_toUnicode) 1326 * <code>U_INDEX_OUTOFBOUNDS_ERROR</code> will be set if the input 1327 * is empty or does not convert to any output (e.g.: pure state-change 1328 * codes SI/SO, escape sequences for ISO 2022, 1329 * or if the callback did not output anything, ...). 1330 * This function will not set a <code>U_BUFFER_OVERFLOW_ERROR</code> because 1331 * the "buffer" is the return code. However, there might be subsequent output 1332 * stored in the converter object 1333 * that will be returned in following calls to this function. 1334 * @return a UChar32 resulting from the partial conversion of source 1335 * @see ucnv_toUnicode 1336 * @see ucnv_toUChars 1337 * @see ucnv_convert 1338 * @stable ICU 2.0 1339 */ 1340 U_STABLE UChar32 U_EXPORT2 1341 ucnv_getNextUChar(UConverter * converter, 1342 const char **source, 1343 const char * sourceLimit, 1344 UErrorCode * err); 1345 1346 /** 1347 * Convert from one external charset to another using two existing UConverters. 1348 * Internally, two conversions - ucnv_toUnicode() and ucnv_fromUnicode() - 1349 * are used, "pivoting" through 16-bit Unicode. 1350 * 1351 * Important: For streaming conversion (multiple function calls for successive 1352 * parts of a text stream), the caller must provide a pivot buffer explicitly, 1353 * and must preserve the pivot buffer and associated pointers from one 1354 * call to another. (The buffer may be moved if its contents and the relative 1355 * pointer positions are preserved.) 1356 * 1357 * There is a similar function, ucnv_convert(), 1358 * which has the following limitations: 1359 * - it takes charset names, not converter objects, so that 1360 * - two converters are opened for each call 1361 * - only single-string conversion is possible, not streaming operation 1362 * - it does not provide enough information to find out, 1363 * in case of failure, whether the toUnicode or 1364 * the fromUnicode conversion failed 1365 * 1366 * By contrast, ucnv_convertEx() 1367 * - takes UConverter parameters instead of charset names 1368 * - fully exposes the pivot buffer for streaming conversion and complete error handling 1369 * 1370 * ucnv_convertEx() also provides further convenience: 1371 * - an option to reset the converters at the beginning 1372 * (if reset==TRUE, see parameters; 1373 * also sets *pivotTarget=*pivotSource=pivotStart) 1374 * - allow NUL-terminated input 1375 * (only a single NUL byte, will not work for charsets with multi-byte NULs) 1376 * (if sourceLimit==NULL, see parameters) 1377 * - terminate with a NUL on output 1378 * (only a single NUL byte, not useful for charsets with multi-byte NULs), 1379 * or set U_STRING_NOT_TERMINATED_WARNING if the output exactly fills 1380 * the target buffer 1381 * - the pivot buffer can be provided internally; 1382 * possible only for whole-string conversion, not streaming conversion; 1383 * in this case, the caller will not be able to get details about where an 1384 * error occurred 1385 * (if pivotStart==NULL, see below) 1386 * 1387 * The function returns when one of the following is true: 1388 * - the entire source text has been converted successfully to the target buffer 1389 * - a target buffer overflow occurred (U_BUFFER_OVERFLOW_ERROR) 1390 * - a conversion error occurred 1391 * (other U_FAILURE(), see description of pErrorCode) 1392 * 1393 * Limitation compared to the direct use of 1394 * ucnv_fromUnicode() and ucnv_toUnicode(): 1395 * ucnv_convertEx() does not provide offset information. 1396 * 1397 * Limitation compared to ucnv_fromUChars() and ucnv_toUChars(): 1398 * ucnv_convertEx() does not support preflighting directly. 1399 * 1400 * Sample code for converting a single string from 1401 * one external charset to UTF-8, ignoring the location of errors: 1402 * 1403 * \code 1404 * int32_t 1405 * myToUTF8(UConverter *cnv, 1406 * const char *s, int32_t length, 1407 * char *u8, int32_t capacity, 1408 * UErrorCode *pErrorCode) { 1409 * UConverter *utf8Cnv; 1410 * char *target; 1411 * 1412 * if(U_FAILURE(*pErrorCode)) { 1413 * return 0; 1414 * } 1415 * 1416 * utf8Cnv=myGetCachedUTF8Converter(pErrorCode); 1417 * if(U_FAILURE(*pErrorCode)) { 1418 * return 0; 1419 * } 1420 * 1421 * if(length<0) { 1422 * length=strlen(s); 1423 * } 1424 * target=u8; 1425 * ucnv_convertEx(utf8Cnv, cnv, 1426 * &target, u8+capacity, 1427 * &s, s+length, 1428 * NULL, NULL, NULL, NULL, 1429 * TRUE, TRUE, 1430 * pErrorCode); 1431 * 1432 * myReleaseCachedUTF8Converter(utf8Cnv); 1433 * 1434 * // return the output string length, but without preflighting 1435 * return (int32_t)(target-u8); 1436 * } 1437 * \endcode 1438 * 1439 * @param targetCnv Output converter, used to convert from the UTF-16 pivot 1440 * to the target using ucnv_fromUnicode(). 1441 * @param sourceCnv Input converter, used to convert from the source to 1442 * the UTF-16 pivot using ucnv_toUnicode(). 1443 * @param target I/O parameter, same as for ucnv_fromUChars(). 1444 * Input: *target points to the beginning of the target buffer. 1445 * Output: *target points to the first unit after the last char written. 1446 * @param targetLimit Pointer to the first unit after the target buffer. 1447 * @param source I/O parameter, same as for ucnv_toUChars(). 1448 * Input: *source points to the beginning of the source buffer. 1449 * Output: *source points to the first unit after the last char read. 1450 * @param sourceLimit Pointer to the first unit after the source buffer. 1451 * @param pivotStart Pointer to the UTF-16 pivot buffer. If pivotStart==NULL, 1452 * then an internal buffer is used and the other pivot 1453 * arguments are ignored and can be NULL as well. 1454 * @param pivotSource I/O parameter, same as source in ucnv_fromUChars() for 1455 * conversion from the pivot buffer to the target buffer. 1456 * @param pivotTarget I/O parameter, same as target in ucnv_toUChars() for 1457 * conversion from the source buffer to the pivot buffer. 1458 * It must be pivotStart<=*pivotSource<=*pivotTarget<=pivotLimit 1459 * and pivotStart<pivotLimit (unless pivotStart==NULL). 1460 * @param pivotLimit Pointer to the first unit after the pivot buffer. 1461 * @param reset If TRUE, then ucnv_resetToUnicode(sourceCnv) and 1462 * ucnv_resetFromUnicode(targetCnv) are called, and the 1463 * pivot pointers are reset (*pivotTarget=*pivotSource=pivotStart). 1464 * @param flush If true, indicates the end of the input. 1465 * Passed directly to ucnv_toUnicode(), and carried over to 1466 * ucnv_fromUnicode() when the source is empty as well. 1467 * @param pErrorCode ICU error code in/out parameter. 1468 * Must fulfill U_SUCCESS before the function call. 1469 * U_BUFFER_OVERFLOW_ERROR always refers to the target buffer 1470 * because overflows into the pivot buffer are handled internally. 1471 * Other conversion errors are from the source-to-pivot 1472 * conversion if *pivotSource==pivotStart, otherwise from 1473 * the pivot-to-target conversion. 1474 * 1475 * @see ucnv_convert 1476 * @see ucnv_fromAlgorithmic 1477 * @see ucnv_toAlgorithmic 1478 * @see ucnv_fromUnicode 1479 * @see ucnv_toUnicode 1480 * @see ucnv_fromUChars 1481 * @see ucnv_toUChars 1482 * @stable ICU 2.6 1483 */ 1484 U_STABLE void U_EXPORT2 1485 ucnv_convertEx(UConverter *targetCnv, UConverter *sourceCnv, 1486 char **target, const char *targetLimit, 1487 const char **source, const char *sourceLimit, 1488 UChar *pivotStart, UChar **pivotSource, 1489 UChar **pivotTarget, const UChar *pivotLimit, 1490 UBool reset, UBool flush, 1491 UErrorCode *pErrorCode); 1492 1493 /** 1494 * Convert from one external charset to another. 1495 * Internally, two converters are opened according to the name arguments, 1496 * then the text is converted to and from the 16-bit Unicode "pivot" 1497 * using ucnv_convertEx(), then the converters are closed again. 1498 * 1499 * This is a convenience function, not an efficient way to convert a lot of text: 1500 * ucnv_convert() 1501 * - takes charset names, not converter objects, so that 1502 * - two converters are opened for each call 1503 * - only single-string conversion is possible, not streaming operation 1504 * - does not provide enough information to find out, 1505 * in case of failure, whether the toUnicode or 1506 * the fromUnicode conversion failed 1507 * - allows NUL-terminated input 1508 * (only a single NUL byte, will not work for charsets with multi-byte NULs) 1509 * (if sourceLength==-1, see parameters) 1510 * - terminate with a NUL on output 1511 * (only a single NUL byte, not useful for charsets with multi-byte NULs), 1512 * or set U_STRING_NOT_TERMINATED_WARNING if the output exactly fills 1513 * the target buffer 1514 * - a pivot buffer is provided internally 1515 * 1516 * The function returns when one of the following is true: 1517 * - the entire source text has been converted successfully to the target buffer 1518 * and either the target buffer is terminated with a single NUL byte 1519 * or the error code is set to U_STRING_NOT_TERMINATED_WARNING 1520 * - a target buffer overflow occurred (U_BUFFER_OVERFLOW_ERROR) 1521 * and the full output string length is returned ("preflighting") 1522 * - a conversion error occurred 1523 * (other U_FAILURE(), see description of pErrorCode) 1524 * 1525 * @param toConverterName The name of the converter that is used to convert 1526 * from the UTF-16 pivot buffer to the target. 1527 * @param fromConverterName The name of the converter that is used to convert 1528 * from the source to the UTF-16 pivot buffer. 1529 * @param target Pointer to the output buffer. 1530 * @param targetCapacity Capacity of the target, in bytes. 1531 * @param source Pointer to the input buffer. 1532 * @param sourceLength Length of the input text, in bytes, or -1 for NUL-terminated input. 1533 * @param pErrorCode ICU error code in/out parameter. 1534 * Must fulfill U_SUCCESS before the function call. 1535 * @return Length of the complete output text in bytes, even if it exceeds the targetCapacity 1536 * and a U_BUFFER_OVERFLOW_ERROR is set. 1537 * 1538 * @see ucnv_convertEx 1539 * @see ucnv_fromAlgorithmic 1540 * @see ucnv_toAlgorithmic 1541 * @see ucnv_fromUnicode 1542 * @see ucnv_toUnicode 1543 * @see ucnv_fromUChars 1544 * @see ucnv_toUChars 1545 * @see ucnv_getNextUChar 1546 * @stable ICU 2.0 1547 */ 1548 U_STABLE int32_t U_EXPORT2 1549 ucnv_convert(const char *toConverterName, 1550 const char *fromConverterName, 1551 char *target, 1552 int32_t targetCapacity, 1553 const char *source, 1554 int32_t sourceLength, 1555 UErrorCode *pErrorCode); 1556 1557 /** 1558 * Convert from one external charset to another. 1559 * Internally, the text is converted to and from the 16-bit Unicode "pivot" 1560 * using ucnv_convertEx(). ucnv_toAlgorithmic() works exactly like ucnv_convert() 1561 * except that the two converters need not be looked up and opened completely. 1562 * 1563 * The source-to-pivot conversion uses the cnv converter parameter. 1564 * The pivot-to-target conversion uses a purely algorithmic converter 1565 * according to the specified type, e.g., UCNV_UTF8 for a UTF-8 converter. 1566 * 1567 * Internally, the algorithmic converter is opened and closed for each 1568 * function call, which is more efficient than using the public ucnv_open() 1569 * but somewhat less efficient than only resetting an existing converter 1570 * and using ucnv_convertEx(). 1571 * 1572 * This function is more convenient than ucnv_convertEx() for single-string 1573 * conversions, especially when "preflighting" is desired (returning the length 1574 * of the complete output even if it does not fit into the target buffer; 1575 * see the User Guide Strings chapter). See ucnv_convert() for details. 1576 * 1577 * @param algorithmicType UConverterType constant identifying the desired target 1578 * charset as a purely algorithmic converter. 1579 * Those are converters for Unicode charsets like 1580 * UTF-8, BOCU-1, SCSU, UTF-7, IMAP-mailbox-name, etc., 1581 * as well as US-ASCII and ISO-8859-1. 1582 * @param cnv The converter that is used to convert 1583 * from the source to the UTF-16 pivot buffer. 1584 * @param target Pointer to the output buffer. 1585 * @param targetCapacity Capacity of the target, in bytes. 1586 * @param source Pointer to the input buffer. 1587 * @param sourceLength Length of the input text, in bytes 1588 * @param pErrorCode ICU error code in/out parameter. 1589 * Must fulfill U_SUCCESS before the function call. 1590 * @return Length of the complete output text in bytes, even if it exceeds the targetCapacity 1591 * and a U_BUFFER_OVERFLOW_ERROR is set. 1592 * 1593 * @see ucnv_fromAlgorithmic 1594 * @see ucnv_convert 1595 * @see ucnv_convertEx 1596 * @see ucnv_fromUnicode 1597 * @see ucnv_toUnicode 1598 * @see ucnv_fromUChars 1599 * @see ucnv_toUChars 1600 * @stable ICU 2.6 1601 */ 1602 U_STABLE int32_t U_EXPORT2 1603 ucnv_toAlgorithmic(UConverterType algorithmicType, 1604 UConverter *cnv, 1605 char *target, int32_t targetCapacity, 1606 const char *source, int32_t sourceLength, 1607 UErrorCode *pErrorCode); 1608 1609 /** 1610 * Convert from one external charset to another. 1611 * Internally, the text is converted to and from the 16-bit Unicode "pivot" 1612 * using ucnv_convertEx(). ucnv_fromAlgorithmic() works exactly like ucnv_convert() 1613 * except that the two converters need not be looked up and opened completely. 1614 * 1615 * The source-to-pivot conversion uses a purely algorithmic converter 1616 * according to the specified type, e.g., UCNV_UTF8 for a UTF-8 converter. 1617 * The pivot-to-target conversion uses the cnv converter parameter. 1618 * 1619 * Internally, the algorithmic converter is opened and closed for each 1620 * function call, which is more efficient than using the public ucnv_open() 1621 * but somewhat less efficient than only resetting an existing converter 1622 * and using ucnv_convertEx(). 1623 * 1624 * This function is more convenient than ucnv_convertEx() for single-string 1625 * conversions, especially when "preflighting" is desired (returning the length 1626 * of the complete output even if it does not fit into the target buffer; 1627 * see the User Guide Strings chapter). See ucnv_convert() for details. 1628 * 1629 * @param cnv The converter that is used to convert 1630 * from the UTF-16 pivot buffer to the target. 1631 * @param algorithmicType UConverterType constant identifying the desired source 1632 * charset as a purely algorithmic converter. 1633 * Those are converters for Unicode charsets like 1634 * UTF-8, BOCU-1, SCSU, UTF-7, IMAP-mailbox-name, etc., 1635 * as well as US-ASCII and ISO-8859-1. 1636 * @param target Pointer to the output buffer. 1637 * @param targetCapacity Capacity of the target, in bytes. 1638 * @param source Pointer to the input buffer. 1639 * @param sourceLength Length of the input text, in bytes 1640 * @param pErrorCode ICU error code in/out parameter. 1641 * Must fulfill U_SUCCESS before the function call. 1642 * @return Length of the complete output text in bytes, even if it exceeds the targetCapacity 1643 * and a U_BUFFER_OVERFLOW_ERROR is set. 1644 * 1645 * @see ucnv_fromAlgorithmic 1646 * @see ucnv_convert 1647 * @see ucnv_convertEx 1648 * @see ucnv_fromUnicode 1649 * @see ucnv_toUnicode 1650 * @see ucnv_fromUChars 1651 * @see ucnv_toUChars 1652 * @stable ICU 2.6 1653 */ 1654 U_STABLE int32_t U_EXPORT2 1655 ucnv_fromAlgorithmic(UConverter *cnv, 1656 UConverterType algorithmicType, 1657 char *target, int32_t targetCapacity, 1658 const char *source, int32_t sourceLength, 1659 UErrorCode *pErrorCode); 1660 1661 /** 1662 * Frees up memory occupied by unused, cached converter shared data. 1663 * 1664 * @return the number of cached converters successfully deleted 1665 * @see ucnv_close 1666 * @stable ICU 2.0 1667 */ 1668 U_STABLE int32_t U_EXPORT2 1669 ucnv_flushCache(void); 1670 1671 /** 1672 * Returns the number of available converters, as per the alias file. 1673 * 1674 * @return the number of available converters 1675 * @see ucnv_getAvailableName 1676 * @stable ICU 2.0 1677 */ 1678 U_STABLE int32_t U_EXPORT2 1679 ucnv_countAvailable(void); 1680 1681 /** 1682 * Gets the canonical converter name of the specified converter from a list of 1683 * all available converters contaied in the alias file. All converters 1684 * in this list can be opened. 1685 * 1686 * @param n the index to a converter available on the system (in the range <TT>[0..ucnv_countAvaiable()]</TT>) 1687 * @return a pointer a string (library owned), or <TT>NULL</TT> if the index is out of bounds. 1688 * @see ucnv_countAvailable 1689 * @stable ICU 2.0 1690 */ 1691 U_STABLE const char* U_EXPORT2 1692 ucnv_getAvailableName(int32_t n); 1693 1694 /** 1695 * Returns a UEnumeration to enumerate all of the canonical converter 1696 * names, as per the alias file, regardless of the ability to open each 1697 * converter. 1698 * 1699 * @return A UEnumeration object for getting all the recognized canonical 1700 * converter names. 1701 * @see ucnv_getAvailableName 1702 * @see uenum_close 1703 * @see uenum_next 1704 * @stable ICU 2.4 1705 */ 1706 U_STABLE UEnumeration * U_EXPORT2 1707 ucnv_openAllNames(UErrorCode *pErrorCode); 1708 1709 /** 1710 * Gives the number of aliases for a given converter or alias name. 1711 * If the alias is ambiguous, then the preferred converter is used 1712 * and the status is set to U_AMBIGUOUS_ALIAS_WARNING. 1713 * This method only enumerates the listed entries in the alias file. 1714 * @param alias alias name 1715 * @param pErrorCode error status 1716 * @return number of names on alias list for given alias 1717 * @stable ICU 2.0 1718 */ 1719 U_STABLE uint16_t U_EXPORT2 1720 ucnv_countAliases(const char *alias, UErrorCode *pErrorCode); 1721 1722 /** 1723 * Gives the name of the alias at given index of alias list. 1724 * This method only enumerates the listed entries in the alias file. 1725 * If the alias is ambiguous, then the preferred converter is used 1726 * and the status is set to U_AMBIGUOUS_ALIAS_WARNING. 1727 * @param alias alias name 1728 * @param n index in alias list 1729 * @param pErrorCode result of operation 1730 * @return returns the name of the alias at given index 1731 * @see ucnv_countAliases 1732 * @stable ICU 2.0 1733 */ 1734 U_STABLE const char * U_EXPORT2 1735 ucnv_getAlias(const char *alias, uint16_t n, UErrorCode *pErrorCode); 1736 1737 /** 1738 * Fill-up the list of alias names for the given alias. 1739 * This method only enumerates the listed entries in the alias file. 1740 * If the alias is ambiguous, then the preferred converter is used 1741 * and the status is set to U_AMBIGUOUS_ALIAS_WARNING. 1742 * @param alias alias name 1743 * @param aliases fill-in list, aliases is a pointer to an array of 1744 * <code>ucnv_countAliases()</code> string-pointers 1745 * (<code>const char *</code>) that will be filled in. 1746 * The strings themselves are owned by the library. 1747 * @param pErrorCode result of operation 1748 * @stable ICU 2.0 1749 */ 1750 U_STABLE void U_EXPORT2 1751 ucnv_getAliases(const char *alias, const char **aliases, UErrorCode *pErrorCode); 1752 1753 /** 1754 * Return a new UEnumeration object for enumerating all the 1755 * alias names for a given converter that are recognized by a standard. 1756 * This method only enumerates the listed entries in the alias file. 1757 * The convrtrs.txt file can be modified to change the results of 1758 * this function. 1759 * The first result in this list is the same result given by 1760 * <code>ucnv_getStandardName</code>, which is the default alias for 1761 * the specified standard name. The returned object must be closed with 1762 * <code>uenum_close</code> when you are done with the object. 1763 * 1764 * @param convName original converter name 1765 * @param standard name of the standard governing the names; MIME and IANA 1766 * are such standards 1767 * @param pErrorCode The error code 1768 * @return A UEnumeration object for getting all aliases that are recognized 1769 * by a standard. If any of the parameters are invalid, NULL 1770 * is returned. 1771 * @see ucnv_getStandardName 1772 * @see uenum_close 1773 * @see uenum_next 1774 * @stable ICU 2.2 1775 */ 1776 U_STABLE UEnumeration * U_EXPORT2 1777 ucnv_openStandardNames(const char *convName, 1778 const char *standard, 1779 UErrorCode *pErrorCode); 1780 1781 /** 1782 * Gives the number of standards associated to converter names. 1783 * @return number of standards 1784 * @stable ICU 2.0 1785 */ 1786 U_STABLE uint16_t U_EXPORT2 1787 ucnv_countStandards(void); 1788 1789 /** 1790 * Gives the name of the standard at given index of standard list. 1791 * @param n index in standard list 1792 * @param pErrorCode result of operation 1793 * @return returns the name of the standard at given index. Owned by the library. 1794 * @stable ICU 2.0 1795 */ 1796 U_STABLE const char * U_EXPORT2 1797 ucnv_getStandard(uint16_t n, UErrorCode *pErrorCode); 1798 1799 /** 1800 * Returns a standard name for a given converter name. 1801 * <p> 1802 * Example alias table:<br> 1803 * conv alias1 { STANDARD1 } alias2 { STANDARD1* } 1804 * <p> 1805 * Result of ucnv_getStandardName("conv", "STANDARD1") from example 1806 * alias table:<br> 1807 * <b>"alias2"</b> 1808 * 1809 * @param name original converter name 1810 * @param standard name of the standard governing the names; MIME and IANA 1811 * are such standards 1812 * @param pErrorCode result of operation 1813 * @return returns the standard converter name; 1814 * if a standard converter name cannot be determined, 1815 * then <code>NULL</code> is returned. Owned by the library. 1816 * @stable ICU 2.0 1817 */ 1818 U_STABLE const char * U_EXPORT2 1819 ucnv_getStandardName(const char *name, const char *standard, UErrorCode *pErrorCode); 1820 1821 /** 1822 * This function will return the internal canonical converter name of the 1823 * tagged alias. This is the opposite of ucnv_openStandardNames, which 1824 * returns the tagged alias given the canonical name. 1825 * <p> 1826 * Example alias table:<br> 1827 * conv alias1 { STANDARD1 } alias2 { STANDARD1* } 1828 * <p> 1829 * Result of ucnv_getStandardName("alias1", "STANDARD1") from example 1830 * alias table:<br> 1831 * <b>"conv"</b> 1832 * 1833 * @return returns the canonical converter name; 1834 * if a standard or alias name cannot be determined, 1835 * then <code>NULL</code> is returned. The returned string is 1836 * owned by the library. 1837 * @see ucnv_getStandardName 1838 * @stable ICU 2.4 1839 */ 1840 U_STABLE const char * U_EXPORT2 1841 ucnv_getCanonicalName(const char *alias, const char *standard, UErrorCode *pErrorCode); 1842 1843 /** 1844 * Returns the current default converter name. If you want to open 1845 * a default converter, you do not need to use this function. 1846 * It is faster if you pass a NULL argument to ucnv_open the 1847 * default converter. 1848 * 1849 * If U_CHARSET_IS_UTF8 is defined to 1 in utypes.h then this function 1850 * always returns "UTF-8". 1851 * 1852 * @return returns the current default converter name. 1853 * Storage owned by the library 1854 * @see ucnv_setDefaultName 1855 * @stable ICU 2.0 1856 */ 1857 U_STABLE const char * U_EXPORT2 1858 ucnv_getDefaultName(void); 1859 1860 /** 1861 * This function is not thread safe. DO NOT call this function when ANY ICU 1862 * function is being used from more than one thread! This function sets the 1863 * current default converter name. If this function needs to be called, it 1864 * should be called during application initialization. Most of the time, the 1865 * results from ucnv_getDefaultName() or ucnv_open with a NULL string argument 1866 * is sufficient for your application. 1867 * 1868 * If U_CHARSET_IS_UTF8 is defined to 1 in utypes.h then this function 1869 * does nothing. 1870 * 1871 * @param name the converter name to be the default (must be known by ICU). 1872 * @see ucnv_getDefaultName 1873 * @system 1874 * @stable ICU 2.0 1875 */ 1876 U_STABLE void U_EXPORT2 1877 ucnv_setDefaultName(const char *name); 1878 1879 /** 1880 * Fixes the backslash character mismapping. For example, in SJIS, the backslash 1881 * character in the ASCII portion is also used to represent the yen currency sign. 1882 * When mapping from Unicode character 0x005C, it's unclear whether to map the 1883 * character back to yen or backslash in SJIS. This function will take the input 1884 * buffer and replace all the yen sign characters with backslash. This is necessary 1885 * when the user tries to open a file with the input buffer on Windows. 1886 * This function will test the converter to see whether such mapping is 1887 * required. You can sometimes avoid using this function by using the correct version 1888 * of Shift-JIS. 1889 * 1890 * @param cnv The converter representing the target codepage. 1891 * @param source the input buffer to be fixed 1892 * @param sourceLen the length of the input buffer 1893 * @see ucnv_isAmbiguous 1894 * @stable ICU 2.0 1895 */ 1896 U_STABLE void U_EXPORT2 1897 ucnv_fixFileSeparator(const UConverter *cnv, UChar *source, int32_t sourceLen); 1898 1899 /** 1900 * Determines if the converter contains ambiguous mappings of the same 1901 * character or not. 1902 * @param cnv the converter to be tested 1903 * @return TRUE if the converter contains ambiguous mapping of the same 1904 * character, FALSE otherwise. 1905 * @stable ICU 2.0 1906 */ 1907 U_STABLE UBool U_EXPORT2 1908 ucnv_isAmbiguous(const UConverter *cnv); 1909 1910 /** 1911 * Sets the converter to use fallback mappings or not. 1912 * Regardless of this flag, the converter will always use 1913 * fallbacks from Unicode Private Use code points, as well as 1914 * reverse fallbacks (to Unicode). 1915 * For details see ".ucm File Format" 1916 * in the Conversion Data chapter of the ICU User Guide: 1917 * http://www.icu-project.org/userguide/conversion-data.html#ucmformat 1918 * 1919 * @param cnv The converter to set the fallback mapping usage on. 1920 * @param usesFallback TRUE if the user wants the converter to take advantage of the fallback 1921 * mapping, FALSE otherwise. 1922 * @stable ICU 2.0 1923 * @see ucnv_usesFallback 1924 */ 1925 U_STABLE void U_EXPORT2 1926 ucnv_setFallback(UConverter *cnv, UBool usesFallback); 1927 1928 /** 1929 * Determines if the converter uses fallback mappings or not. 1930 * This flag has restrictions, see ucnv_setFallback(). 1931 * 1932 * @param cnv The converter to be tested 1933 * @return TRUE if the converter uses fallback, FALSE otherwise. 1934 * @stable ICU 2.0 1935 * @see ucnv_setFallback 1936 */ 1937 U_STABLE UBool U_EXPORT2 1938 ucnv_usesFallback(const UConverter *cnv); 1939 1940 /** 1941 * Detects Unicode signature byte sequences at the start of the byte stream 1942 * and returns the charset name of the indicated Unicode charset. 1943 * NULL is returned when no Unicode signature is recognized. 1944 * The number of bytes in the signature is output as well. 1945 * 1946 * The caller can ucnv_open() a converter using the charset name. 1947 * The first code unit (UChar) from the start of the stream will be U+FEFF 1948 * (the Unicode BOM/signature character) and can usually be ignored. 1949 * 1950 * For most Unicode charsets it is also possible to ignore the indicated 1951 * number of initial stream bytes and start converting after them. 1952 * However, there are stateful Unicode charsets (UTF-7 and BOCU-1) for which 1953 * this will not work. Therefore, it is best to ignore the first output UChar 1954 * instead of the input signature bytes. 1955 * <p> 1956 * Usage: 1957 * @code 1958 * UErrorCode err = U_ZERO_ERROR; 1959 * char input[] = { '\xEF','\xBB', '\xBF','\x41','\x42','\x43' }; 1960 * int32_t signatureLength = 0; 1961 * char *encoding = ucnv_detectUnicodeSignature(input,sizeof(input),&signatureLength,&err); 1962 * UConverter *conv = NULL; 1963 * UChar output[100]; 1964 * UChar *target = output, *out; 1965 * char *source = input; 1966 * if(encoding!=NULL && U_SUCCESS(err)){ 1967 * // should signature be discarded ? 1968 * conv = ucnv_open(encoding, &err); 1969 * // do the conversion 1970 * ucnv_toUnicode(conv, 1971 * target, output + sizeof(output)/U_SIZEOF_UCHAR, 1972 * source, input + sizeof(input), 1973 * NULL, TRUE, &err); 1974 * out = output; 1975 * if (discardSignature){ 1976 * ++out; // ignore initial U+FEFF 1977 * } 1978 * while(out != target) { 1979 * printf("%04x ", *out++); 1980 * } 1981 * puts(""); 1982 * } 1983 * 1984 * @endcode 1985 * 1986 * @param source The source string in which the signature should be detected. 1987 * @param sourceLength Length of the input string, or -1 if terminated with a NUL byte. 1988 * @param signatureLength A pointer to int32_t to receive the number of bytes that make up the signature 1989 * of the detected UTF. 0 if not detected. 1990 * Can be a NULL pointer. 1991 * @param pErrorCode ICU error code in/out parameter. 1992 * Must fulfill U_SUCCESS before the function call. 1993 * @return The name of the encoding detected. NULL if encoding is not detected. 1994 * @stable ICU 2.4 1995 */ 1996 U_STABLE const char* U_EXPORT2 1997 ucnv_detectUnicodeSignature(const char* source, 1998 int32_t sourceLength, 1999 int32_t *signatureLength, 2000 UErrorCode *pErrorCode); 2001 2002 /** 2003 * Returns the number of UChars held in the converter's internal state 2004 * because more input is needed for completing the conversion. This function is 2005 * useful for mapping semantics of ICU's converter interface to those of iconv, 2006 * and this information is not needed for normal conversion. 2007 * @param cnv The converter in which the input is held 2008 * @param status ICU error code in/out parameter. 2009 * Must fulfill U_SUCCESS before the function call. 2010 * @return The number of UChars in the state. -1 if an error is encountered. 2011 * @stable ICU 3.4 2012 */ 2013 U_STABLE int32_t U_EXPORT2 2014 ucnv_fromUCountPending(const UConverter* cnv, UErrorCode* status); 2015 2016 /** 2017 * Returns the number of chars held in the converter's internal state 2018 * because more input is needed for completing the conversion. This function is 2019 * useful for mapping semantics of ICU's converter interface to those of iconv, 2020 * and this information is not needed for normal conversion. 2021 * @param cnv The converter in which the input is held as internal state 2022 * @param status ICU error code in/out parameter. 2023 * Must fulfill U_SUCCESS before the function call. 2024 * @return The number of chars in the state. -1 if an error is encountered. 2025 * @stable ICU 3.4 2026 */ 2027 U_STABLE int32_t U_EXPORT2 2028 ucnv_toUCountPending(const UConverter* cnv, UErrorCode* status); 2029 2030 /** 2031 * Returns whether or not the charset of the converter has a fixed number of bytes 2032 * per charset character. 2033 * An example of this are converters that are of the type UCNV_SBCS or UCNV_DBCS. 2034 * Another example is UTF-32 which is always 4 bytes per character. 2035 * A Unicode code point may be represented by more than one UTF-8 or UTF-16 code unit 2036 * but a UTF-32 converter encodes each code point with 4 bytes. 2037 * Note: This method is not intended to be used to determine whether the charset has a 2038 * fixed ratio of bytes to Unicode codes <i>units</i> for any particular Unicode encoding form. 2039 * FALSE is returned with the UErrorCode if error occurs or cnv is NULL. 2040 * @param cnv The converter to be tested 2041 * @param status ICU error code in/out paramter 2042 * @return TRUE if the converter is fixed-width 2043 * @draft ICU 4.8 2044 */ 2045 U_DRAFT UBool U_EXPORT2 2046 ucnv_isFixedWidth(UConverter *cnv, UErrorCode *status); 2047 2048 #endif 2049 2050 #endif 2051 /*_UCNV*/ 2052