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