This API is used to convert codepage or character encoded data to and sl@0: * from UTF-16. You can open a converter with {@link ucnv_open() }. With that sl@0: * converter, you can get its properties, set options, convert your data and sl@0: * close the converter.
sl@0: * sl@0: *Since many software programs recogize different converter names for sl@0: * different types of converters, there are other functions in this API to sl@0: * iterate over the converter aliases. The functions {@link ucnv_getAvailableName() }, sl@0: * {@link ucnv_getAlias() } and {@link ucnv_getStandardName() } are some of the sl@0: * more frequently used alias functions to get this information.
sl@0: * sl@0: *When a converter encounters an illegal, irregular, invalid or unmappable character sl@0: * its default behavior is to use a substitution character to replace the sl@0: * bad byte sequence. This behavior can be changed by using {@link ucnv_setFromUCallBack() } sl@0: * or {@link ucnv_setToUCallBack() } on the converter. The header ucnv_err.h defines sl@0: * many other callback actions that can be used instead of a character substitution.
sl@0: * sl@0: *More information about this API can be found in our sl@0: * User's sl@0: * Guide.
sl@0: */ sl@0: sl@0: #ifndef UCNV_H sl@0: #define UCNV_H sl@0: sl@0: #include "unicode/ucnv_err.h" sl@0: #include "unicode/uenum.h" sl@0: sl@0: #ifndef __USET_H__ sl@0: sl@0: /** sl@0: * USet is the C API type for Unicode sets. sl@0: * It is forward-declared here to avoid including the header file if related sl@0: * conversion APIs are not used. sl@0: * See unicode/uset.h sl@0: * sl@0: * @see ucnv_getUnicodeSet sl@0: * @stable ICU 2.6 sl@0: */ sl@0: struct USet; sl@0: /** @stable ICU 2.6 */ sl@0: typedef struct USet USet; sl@0: sl@0: #endif sl@0: sl@0: #if !UCONFIG_NO_CONVERSION sl@0: sl@0: U_CDECL_BEGIN sl@0: sl@0: /** Maximum length of a converter name including the terminating NULL @stable ICU 2.0 */ sl@0: #define UCNV_MAX_CONVERTER_NAME_LENGTH 60 sl@0: /** Maximum length of a converter name including path and terminating NULL @stable ICU 2.0 */ sl@0: #define UCNV_MAX_FULL_FILE_NAME_LENGTH (600+UCNV_MAX_CONVERTER_NAME_LENGTH) sl@0: sl@0: /** Shift in for EBDCDIC_STATEFUL and iso2022 states @stable ICU 2.0 */ sl@0: #define UCNV_SI 0x0F sl@0: /** Shift out for EBDCDIC_STATEFUL and iso2022 states @stable ICU 2.0 */ sl@0: #define UCNV_SO 0x0E sl@0: sl@0: /** sl@0: * Enum for specifying basic types of converters sl@0: * @see ucnv_getType sl@0: * @stable ICU 2.0 sl@0: */ sl@0: typedef enum { sl@0: UCNV_UNSUPPORTED_CONVERTER = -1, sl@0: UCNV_SBCS = 0, sl@0: UCNV_DBCS = 1, sl@0: UCNV_MBCS = 2, sl@0: UCNV_LATIN_1 = 3, sl@0: UCNV_UTF8 = 4, sl@0: UCNV_UTF16_BigEndian = 5, sl@0: UCNV_UTF16_LittleEndian = 6, sl@0: UCNV_UTF32_BigEndian = 7, sl@0: UCNV_UTF32_LittleEndian = 8, sl@0: UCNV_EBCDIC_STATEFUL = 9, sl@0: UCNV_ISO_2022 = 10, sl@0: sl@0: UCNV_LMBCS_1 = 11, sl@0: UCNV_LMBCS_2, sl@0: UCNV_LMBCS_3, sl@0: UCNV_LMBCS_4, sl@0: UCNV_LMBCS_5, sl@0: UCNV_LMBCS_6, sl@0: UCNV_LMBCS_8, sl@0: UCNV_LMBCS_11, sl@0: UCNV_LMBCS_16, sl@0: UCNV_LMBCS_17, sl@0: UCNV_LMBCS_18, sl@0: UCNV_LMBCS_19, sl@0: UCNV_LMBCS_LAST = UCNV_LMBCS_19, sl@0: UCNV_HZ, sl@0: UCNV_SCSU, sl@0: UCNV_ISCII, sl@0: UCNV_US_ASCII, sl@0: UCNV_UTF7, sl@0: UCNV_BOCU1, sl@0: UCNV_UTF16, sl@0: UCNV_UTF32, sl@0: UCNV_CESU8, sl@0: UCNV_IMAP_MAILBOX, sl@0: sl@0: /* Number of converter types for which we have conversion routines. */ sl@0: UCNV_NUMBER_OF_SUPPORTED_CONVERTER_TYPES sl@0: sl@0: } UConverterType; sl@0: sl@0: /** sl@0: * Enum for specifying which platform a converter ID refers to. sl@0: * The use of platform/CCSID is not recommended. See ucnv_openCCSID(). sl@0: * sl@0: * @see ucnv_getPlatform sl@0: * @see ucnv_openCCSID sl@0: * @see ucnv_getCCSID sl@0: * @stable ICU 2.0 sl@0: */ sl@0: typedef enum { sl@0: UCNV_UNKNOWN = -1, sl@0: UCNV_IBM = 0 sl@0: } UConverterPlatform; sl@0: sl@0: /** sl@0: * Function pointer for error callback in the codepage to unicode direction. sl@0: * Called when an error has occured in conversion to unicode, or on open/close of the callback (see reason). sl@0: * @param context Pointer to the callback's private data sl@0: * @param args Information about the conversion in progress sl@0: * @param codeUnits Points to 'length' bytes of the concerned codepage sequence sl@0: * @param length Size (in bytes) of the concerned codepage sequence sl@0: * @param reason Defines the reason the callback was invoked sl@0: * @param pErrorCode ICU error code in/out parameter. sl@0: * For converter callback functions, set to a conversion error sl@0: * before the call, and the callback may reset it to U_ZERO_ERROR. sl@0: * @see ucnv_setToUCallBack sl@0: * @see UConverterToUnicodeArgs sl@0: * @stable ICU 2.0 sl@0: */ sl@0: typedef void (U_EXPORT2 *UConverterToUCallback) ( sl@0: const void* context, sl@0: UConverterToUnicodeArgs *args, sl@0: const char *codeUnits, sl@0: int32_t length, sl@0: UConverterCallbackReason reason, sl@0: UErrorCode *pErrorCode); sl@0: sl@0: /** sl@0: * Function pointer for error callback in the unicode to codepage direction. sl@0: * Called when an error has occured in conversion from unicode, or on open/close of the callback (see reason). sl@0: * @param context Pointer to the callback's private data sl@0: * @param args Information about the conversion in progress sl@0: * @param codeUnits Points to 'length' UChars of the concerned Unicode sequence sl@0: * @param length Size (in bytes) of the concerned codepage sequence sl@0: * @param codePoint Single UChar32 (UTF-32) containing the concerend Unicode codepoint. sl@0: * @param reason Defines the reason the callback was invoked sl@0: * @param pErrorCode ICU error code in/out parameter. sl@0: * For converter callback functions, set to a conversion error sl@0: * before the call, and the callback may reset it to U_ZERO_ERROR. sl@0: * @see ucnv_setFromUCallBack sl@0: * @stable ICU 2.0 sl@0: */ sl@0: typedef void (U_EXPORT2 *UConverterFromUCallback) ( sl@0: const void* context, sl@0: UConverterFromUnicodeArgs *args, sl@0: const UChar* codeUnits, sl@0: int32_t length, sl@0: UChar32 codePoint, sl@0: UConverterCallbackReason reason, sl@0: UErrorCode *pErrorCode); sl@0: sl@0: U_CDECL_END sl@0: sl@0: /** sl@0: * Character that separates converter names from options and options from each other. sl@0: * @see ucnv_open sl@0: * @stable ICU 2.0 sl@0: */ sl@0: #define UCNV_OPTION_SEP_CHAR ',' sl@0: sl@0: /** sl@0: * String version of UCNV_OPTION_SEP_CHAR. sl@0: * @see ucnv_open sl@0: * @stable ICU 2.0 sl@0: */ sl@0: #define UCNV_OPTION_SEP_STRING "," sl@0: sl@0: /** sl@0: * Character that separates a converter option from its value. sl@0: * @see ucnv_open sl@0: * @stable ICU 2.0 sl@0: */ sl@0: #define UCNV_VALUE_SEP_CHAR '=' sl@0: sl@0: /** sl@0: * String version of UCNV_VALUE_SEP_CHAR. sl@0: * @see ucnv_open sl@0: * @stable ICU 2.0 sl@0: */ sl@0: #define UCNV_VALUE_SEP_STRING "=" sl@0: sl@0: /** sl@0: * Converter option for specifying a locale. sl@0: * For example, ucnv_open("SCSU,locale=ja", &errorCode); sl@0: * See convrtrs.txt. sl@0: * sl@0: * @see ucnv_open sl@0: * @stable ICU 2.0 sl@0: */ sl@0: #define UCNV_LOCALE_OPTION_STRING ",locale=" sl@0: sl@0: /** sl@0: * Converter option for specifying a version selector (0..9) for some converters. sl@0: * For example, ucnv_open("UTF-7,version=1", &errorCode); sl@0: * See convrtrs.txt. sl@0: * sl@0: * @see ucnv_open sl@0: * @stable ICU 2.4 sl@0: */ sl@0: #define UCNV_VERSION_OPTION_STRING ",version=" sl@0: sl@0: /** sl@0: * Converter option for EBCDIC SBCS or mixed-SBCS/DBCS (stateful) codepages. sl@0: * Swaps Unicode mappings for EBCDIC LF and NL codes, as used on sl@0: * S/390 (z/OS) Unix System Services (Open Edition). sl@0: * For example, ucnv_open("ibm-1047,swaplfnl", &errorCode); sl@0: * See convrtrs.txt. sl@0: * sl@0: * @see ucnv_open sl@0: * @stable ICU 2.4 sl@0: */ sl@0: #define UCNV_SWAP_LFNL_OPTION_STRING ",swaplfnl" sl@0: sl@0: /** sl@0: * Do a fuzzy compare of a two converter/alias names. The comparison sl@0: * is case-insensitive. It also ignores the characters '-', '_', and sl@0: * ' ' (dash, underscore, and space). Thus the strings "UTF-8", sl@0: * "utf_8", and "Utf 8" are exactly equivalent. sl@0: * sl@0: * @param name1 a converter name or alias, zero-terminated sl@0: * @param name2 a converter name or alias, zero-terminated sl@0: * @return 0 if the names match, or a negative value if the name1 sl@0: * lexically precedes name2, or a positive value if the name1 sl@0: * lexically follows name2. sl@0: * @stable ICU 2.0 sl@0: */ sl@0: U_STABLE int U_EXPORT2 sl@0: ucnv_compareNames(const char *name1, const char *name2); sl@0: sl@0: sl@0: /** sl@0: * Creates a UConverter object with the names specified as a C string. sl@0: * The actual name will be resolved with the alias file sl@0: * using a case-insensitive string comparison that ignores sl@0: * the delimiters '-', '_', and ' ' (dash, underscore, and space). sl@0: * E.g., the names "UTF8", "utf-8", and "Utf 8" are all equivalent. sl@0: * IfNULL is passed for the converter name, it will create one with the
sl@0: * getDefaultName return value.
sl@0: *
sl@0: * A converter name for ICU 1.5 and above may contain options sl@0: * like a locale specification to control the specific behavior of sl@0: * the newly instantiated converter. sl@0: * The meaning of the options depends on the particular converter. sl@0: * If an option is not defined for or recognized by a given converter, then it is ignored.
sl@0: * sl@0: *Options are appended to the converter name string, with a
sl@0: * UCNV_OPTION_SEP_CHAR between the name and the first option and
sl@0: * also between adjacent options.
If the alias is ambiguous, then the preferred converter is used sl@0: * and the status is set to U_AMBIGUOUS_ALIAS_WARNING.
sl@0: * sl@0: *The conversion behavior and names can vary between platforms. ICU may sl@0: * convert some characters differently from other platforms. Details on this topic sl@0: * are in the User's sl@0: * Guide.
sl@0: * sl@0: * @param converterName Name of the uconv table, may have options appended sl@0: * @param err outgoing error status U_MEMORY_ALLOCATION_ERROR, U_FILE_ACCESS_ERROR sl@0: * @return the created Unicode converter object, or NULL if an error occured sl@0: * @see ucnv_openU sl@0: * @see ucnv_openCCSID sl@0: * @see ucnv_close sl@0: * @stable ICU 2.0 sl@0: */ sl@0: U_STABLE UConverter* U_EXPORT2 sl@0: ucnv_open(const char *converterName, UErrorCode *err); sl@0: sl@0: sl@0: /** sl@0: * Creates a Unicode converter with the names specified as unicode string. sl@0: * The name should be limited to the ASCII-7 alphanumerics range. sl@0: * The actual name will be resolved with the alias file sl@0: * using a case-insensitive string comparison that ignores sl@0: * the delimiters '-', '_', and ' ' (dash, underscore, and space). sl@0: * E.g., the names "UTF8", "utf-8", and "Utf 8" are all equivalent. sl@0: * If NULL is passed for the converter name, it will create sl@0: * one with the ucnv_getDefaultName() return value. sl@0: * If the alias is ambiguous, then the preferred converter is used sl@0: * and the status is set to U_AMBIGUOUS_ALIAS_WARNING. sl@0: * @param name : name of the uconv table in a zero terminated sl@0: * Unicode string sl@0: * @param err outgoing error status U_MEMORY_ALLOCATION_ERROR, sl@0: * U_FILE_ACCESS_ERROR sl@0: * @return the created Unicode converter object, or NULL if an sl@0: * error occured sl@0: * @see ucnv_open sl@0: * @see ucnv_openCCSID sl@0: * @see ucnv_close sl@0: * @see ucnv_getDefaultName sl@0: * @stable ICU 2.0 sl@0: */ sl@0: U_STABLE UConverter* U_EXPORT2 sl@0: ucnv_openU(const UChar *name, sl@0: UErrorCode *err); sl@0: sl@0: /** sl@0: * Creates a UConverter object from a CCSID number and platform pair. sl@0: * Note that the usefulness of this function is limited to platforms with numeric sl@0: * encoding IDs. Only IBM and Microsoft platforms use numeric (16-bit) identifiers for sl@0: * encodings. sl@0: * sl@0: * In addition, IBM CCSIDs and Unicode conversion tables are not 1:1 related. sl@0: * For many IBM CCSIDs there are multiple (up to six) Unicode conversion tables, and sl@0: * for some Unicode conversion tables there are multiple CCSIDs. sl@0: * Some "alternate" Unicode conversion tables are provided by the sl@0: * IBM CDRA conversion table registry. sl@0: * The most prominent example of a systematic modification of conversion tables that is sl@0: * not provided in the form of conversion table files in the repository is sl@0: * that S/390 Unix System Services swaps the codes for Line Feed and New Line in all sl@0: * EBCDIC codepages, which requires such a swap in the Unicode conversion tables as well. sl@0: * sl@0: * Only IBM default conversion tables are accessible with ucnv_openCCSID(). sl@0: * ucnv_getCCSID() will return the same CCSID for all conversion tables that are associated sl@0: * with that CCSID. sl@0: * sl@0: * Currently, the only "platform" supported in the ICU converter API is UCNV_IBM. sl@0: * sl@0: * In summary, the use of CCSIDs and the associated API functions is not recommended. sl@0: * sl@0: * In order to open a converter with the default IBM CDRA Unicode conversion table, sl@0: * you can use this function or use the prefix "ibm-": sl@0: * \code sl@0: * char name[20]; sl@0: * sprintf(name, "ibm-%hu", ccsid); sl@0: * cnv=ucnv_open(name, &errorCode); sl@0: * \endcode sl@0: * sl@0: * In order to open a converter with the IBM S/390 Unix System Services variant sl@0: * of a Unicode/EBCDIC conversion table, sl@0: * you can use the prefix "ibm-" together with the option string UCNV_SWAP_LFNL_OPTION_STRING: sl@0: * \code sl@0: * char name[20]; sl@0: * sprintf(name, "ibm-%hu" UCNV_SWAP_LFNL_OPTION_STRING, ccsid); sl@0: * cnv=ucnv_open(name, &errorCode); sl@0: * \endcode sl@0: * sl@0: * In order to open a converter from a Microsoft codepage number, use the prefix "cp": sl@0: * \code sl@0: * char name[20]; sl@0: * sprintf(name, "cp%hu", codepageID); sl@0: * cnv=ucnv_open(name, &errorCode); sl@0: * \endcode sl@0: * sl@0: * If the alias is ambiguous, then the preferred converter is used sl@0: * and the status is set to U_AMBIGUOUS_ALIAS_WARNING. sl@0: * sl@0: * @param codepage codepage number to create sl@0: * @param platform the platform in which the codepage number exists sl@0: * @param err error status U_MEMORY_ALLOCATION_ERROR, U_FILE_ACCESS_ERROR sl@0: * @return the created Unicode converter object, or NULL if an error sl@0: * occured. sl@0: * @see ucnv_open sl@0: * @see ucnv_openU sl@0: * @see ucnv_close sl@0: * @see ucnv_getCCSID sl@0: * @see ucnv_getPlatform sl@0: * @see UConverterPlatform sl@0: * @stable ICU 2.0 sl@0: */ sl@0: U_STABLE UConverter* U_EXPORT2 sl@0: ucnv_openCCSID(int32_t codepage, sl@0: UConverterPlatform platform, sl@0: UErrorCode * err); sl@0: sl@0: /** sl@0: *Creates a UConverter object specified from a packageName and a converterName.
sl@0: * sl@0: *The packageName and converterName must point to an ICU udata object, as defined by
sl@0: * udata_open( packageName, "cnv", converterName, err) or equivalent.
sl@0: * Typically, packageName will refer to a (.dat) file, or to a package registered with
sl@0: * udata_setAppData(). Using a full file or directory pathname for packageName is deprecated.
The name will NOT be looked up in the alias mechanism, nor will the converter be sl@0: * stored in the converter cache or the alias table. The only way to open further converters sl@0: * is call this function multiple times, or use the ucnv_safeClone() function to clone a sl@0: * 'master' converter.
sl@0: * sl@0: *A future version of ICU may add alias table lookups and/or caching sl@0: * to this function.
sl@0: * sl@0: *Example Use:
sl@0: * cnv = ucnv_openPackage("myapp", "myconverter", &err);
sl@0: *
U_BUFFER_OVERFLOW_ERROR will be set if the target is full and there is
sl@0: * still data to be written to the target.
sl@0: * @see ucnv_fromUChars
sl@0: * @see ucnv_convert
sl@0: * @see ucnv_getMinCharSize
sl@0: * @see ucnv_setToUCallBack
sl@0: * @stable ICU 2.0
sl@0: */
sl@0: U_STABLE void U_EXPORT2
sl@0: ucnv_fromUnicode (UConverter * converter,
sl@0: char **target,
sl@0: const char *targetLimit,
sl@0: const UChar ** source,
sl@0: const UChar * sourceLimit,
sl@0: int32_t* offsets,
sl@0: UBool flush,
sl@0: UErrorCode * err);
sl@0:
sl@0: /**
sl@0: * Converts a buffer of codepage bytes into an array of unicode UChars
sl@0: * characters. This function is optimized for converting a continuous
sl@0: * stream of data in buffer-sized chunks, where the entire source and
sl@0: * target does not fit in available buffers.
sl@0: *
sl@0: * The source pointer is an in/out parameter. It starts out pointing where the
sl@0: * conversion is to begin, and ends up pointing after the last byte of source consumed.
sl@0: *
sl@0: * Target similarly starts out pointer at the first available UChar in the output
sl@0: * buffer, and ends up pointing after the last UChar written to the output.
sl@0: * It does NOT necessarily keep UChar sequences together.
sl@0: *
sl@0: * The converter always attempts to consume the entire source buffer, unless
sl@0: * (1.) the target buffer is full, or (2.) a failing error is returned from the
sl@0: * current callback function. When a successful error status has been
sl@0: * returned, it means that all of the source buffer has been
sl@0: * consumed. At that point, the caller should reset the source and
sl@0: * sourceLimit pointers to point to the next chunk.
sl@0: *
sl@0: * At the end of the stream (flush==TRUE), the input is completely consumed
sl@0: * when *source==sourceLimit and no error code is set
sl@0: * The converter object is then automatically reset by this function.
sl@0: * (This means that a converter need not be reset explicitly between data
sl@0: * streams if it finishes the previous stream without errors.)
sl@0: *
sl@0: * This is a stateful conversion. Additionally, even when all source data has
sl@0: * been consumed, some data may be in the converters' internal state.
sl@0: * Call this function repeatedly, updating the target pointers with
sl@0: * the next empty chunk of target in case of a
sl@0: * U_BUFFER_OVERFLOW_ERROR, and updating the source pointers
sl@0: * with the next chunk of source when a successful error status is
sl@0: * returned, until there are no more chunks of source data.
sl@0: * @param converter the Unicode converter
sl@0: * @param target I/O parameter. Input : Points to the beginning of the buffer to copy
sl@0: * UChars into. Output : points to after the last UChar copied.
sl@0: * @param targetLimit the pointer just after the end of the target buffer
sl@0: * @param source I/O parameter, pointer to pointer to the source codepage buffer.
sl@0: * @param sourceLimit the pointer to the byte after the end of the source buffer
sl@0: * @param offsets if NULL is passed, nothing will happen to it, otherwise it needs to have the same number
sl@0: * of allocated cells as target. Will fill in offsets from target to source pointer
sl@0: * e.g: offsets[3] is equal to 6, it means that the target[3] was a result of transcoding source[6]
sl@0: * For output data carried across calls, and other data without a specific source character
sl@0: * (such as from escape sequences or callbacks) -1 will be placed for offsets.
sl@0: * @param flush set to TRUE if the current source buffer is the last available
sl@0: * chunk of the source, FALSE otherwise. Note that if a failing status is returned,
sl@0: * this function may have to be called multiple times with flush set to TRUE until
sl@0: * the source buffer is consumed.
sl@0: * @param err the error status. U_ILLEGAL_ARGUMENT_ERROR will be set if the
sl@0: * converter is NULL.
sl@0: * U_BUFFER_OVERFLOW_ERROR will be set if the target is full and there is
sl@0: * still data to be written to the target.
sl@0: * @see ucnv_fromUChars
sl@0: * @see ucnv_convert
sl@0: * @see ucnv_getMinCharSize
sl@0: * @see ucnv_setFromUCallBack
sl@0: * @see ucnv_getNextUChar
sl@0: * @stable ICU 2.0
sl@0: */
sl@0: U_STABLE void U_EXPORT2
sl@0: ucnv_toUnicode(UConverter *converter,
sl@0: UChar **target,
sl@0: const UChar *targetLimit,
sl@0: const char **source,
sl@0: const char *sourceLimit,
sl@0: int32_t *offsets,
sl@0: UBool flush,
sl@0: UErrorCode *err);
sl@0:
sl@0: /**
sl@0: * Convert the Unicode string into a codepage string using an existing UConverter.
sl@0: * The output string is NUL-terminated if possible.
sl@0: *
sl@0: * This function is a more convenient but less powerful version of ucnv_fromUnicode().
sl@0: * It is only useful for whole strings, not for streaming conversion.
sl@0: *
sl@0: * The maximum output buffer capacity required (barring output from callbacks) will be
sl@0: * UCNV_GET_MAX_BYTES_FOR_STRING(srcLength, ucnv_getMaxCharSize(cnv)).
sl@0: *
sl@0: * @param cnv the converter object to be used (ucnv_resetFromUnicode() will be called)
sl@0: * @param src the input Unicode string
sl@0: * @param srcLength the input string length, or -1 if NUL-terminated
sl@0: * @param dest destination string buffer, can be NULL if destCapacity==0
sl@0: * @param destCapacity the number of chars available at dest
sl@0: * @param pErrorCode normal ICU error code;
sl@0: * common error codes that may be set by this function include
sl@0: * U_BUFFER_OVERFLOW_ERROR, U_STRING_NOT_TERMINATED_WARNING,
sl@0: * U_ILLEGAL_ARGUMENT_ERROR, and conversion errors
sl@0: * @return the length of the output string, not counting the terminating NUL;
sl@0: * if the length is greater than destCapacity, then the string will not fit
sl@0: * and a buffer of the indicated length would need to be passed in
sl@0: * @see ucnv_fromUnicode
sl@0: * @see ucnv_convert
sl@0: * @see UCNV_GET_MAX_BYTES_FOR_STRING
sl@0: * @stable ICU 2.0
sl@0: */
sl@0: U_STABLE int32_t U_EXPORT2
sl@0: ucnv_fromUChars(UConverter *cnv,
sl@0: char *dest, int32_t destCapacity,
sl@0: const UChar *src, int32_t srcLength,
sl@0: UErrorCode *pErrorCode);
sl@0:
sl@0: /**
sl@0: * Convert the codepage string into a Unicode string using an existing UConverter.
sl@0: * The output string is NUL-terminated if possible.
sl@0: *
sl@0: * This function is a more convenient but less powerful version of ucnv_toUnicode().
sl@0: * It is only useful for whole strings, not for streaming conversion.
sl@0: *
sl@0: * The maximum output buffer capacity required (barring output from callbacks) will be
sl@0: * 2*srcLength (each char may be converted into a surrogate pair).
sl@0: *
sl@0: * @param cnv the converter object to be used (ucnv_resetToUnicode() will be called)
sl@0: * @param src the input codepage string
sl@0: * @param srcLength the input string length, or -1 if NUL-terminated
sl@0: * @param dest destination string buffer, can be NULL if destCapacity==0
sl@0: * @param destCapacity the number of UChars available at dest
sl@0: * @param pErrorCode normal ICU error code;
sl@0: * common error codes that may be set by this function include
sl@0: * U_BUFFER_OVERFLOW_ERROR, U_STRING_NOT_TERMINATED_WARNING,
sl@0: * U_ILLEGAL_ARGUMENT_ERROR, and conversion errors
sl@0: * @return the length of the output string, not counting the terminating NUL;
sl@0: * if the length is greater than destCapacity, then the string will not fit
sl@0: * and a buffer of the indicated length would need to be passed in
sl@0: * @see ucnv_toUnicode
sl@0: * @see ucnv_convert
sl@0: * @stable ICU 2.0
sl@0: */
sl@0: U_STABLE int32_t U_EXPORT2
sl@0: ucnv_toUChars(UConverter *cnv,
sl@0: UChar *dest, int32_t destCapacity,
sl@0: const char *src, int32_t srcLength,
sl@0: UErrorCode *pErrorCode);
sl@0:
sl@0: /**
sl@0: * Convert a codepage buffer into Unicode one character at a time.
sl@0: * The input is completely consumed when the U_INDEX_OUTOFBOUNDS_ERROR is set.
sl@0: *
sl@0: * Advantage compared to ucnv_toUnicode() or ucnv_toUChars():
sl@0: * - Faster for small amounts of data, for most converters, e.g.,
sl@0: * US-ASCII, ISO-8859-1, UTF-8/16/32, and most "normal" charsets.
sl@0: * (For complex converters, e.g., SCSU, UTF-7 and ISO 2022 variants,
sl@0: * it uses ucnv_toUnicode() internally.)
sl@0: * - Convenient.
sl@0: *
sl@0: * Limitations compared to ucnv_toUnicode():
sl@0: * - Always assumes flush=TRUE.
sl@0: * This makes ucnv_getNextUChar() unsuitable for "streaming" conversion,
sl@0: * that is, for where the input is supplied in multiple buffers,
sl@0: * because ucnv_getNextUChar() will assume the end of the input at the end
sl@0: * of the first buffer.
sl@0: * - Does not provide offset output.
sl@0: *
sl@0: * It is possible to "mix" ucnv_getNextUChar() and ucnv_toUnicode() because
sl@0: * ucnv_getNextUChar() uses the current state of the converter
sl@0: * (unlike ucnv_toUChars() which always resets first).
sl@0: * However, if ucnv_getNextUChar() is called after ucnv_toUnicode()
sl@0: * stopped in the middle of a character sequence (with flush=FALSE),
sl@0: * then ucnv_getNextUChar() will always use the slower ucnv_toUnicode()
sl@0: * internally until the next character boundary.
sl@0: * (This is new in ICU 2.6. In earlier releases, ucnv_getNextUChar() had to
sl@0: * start at a character boundary.)
sl@0: *
sl@0: * Instead of using ucnv_getNextUChar(), it is recommended
sl@0: * to convert using ucnv_toUnicode() or ucnv_toUChars()
sl@0: * and then iterate over the text using U16_NEXT() or a UCharIterator (uiter.h)
sl@0: * or a C++ CharacterIterator or similar.
sl@0: * This allows streaming conversion and offset output, for example.
sl@0: *
sl@0: * Handling of surrogate pairs and supplementary-plane code points:
sl@0: * There are two different kinds of codepages that provide mappings for surrogate characters:
sl@0: *
U_INDEX_OUTOFBOUNDS_ERROR will be set if the input
sl@0: * is empty or does not convert to any output (e.g.: pure state-change
sl@0: * codes SI/SO, escape sequences for ISO 2022,
sl@0: * or if the callback did not output anything, ...).
sl@0: * This function will not set a U_BUFFER_OVERFLOW_ERROR because
sl@0: * the "buffer" is the return code. However, there might be subsequent output
sl@0: * stored in the converter object
sl@0: * that will be returned in following calls to this function.
sl@0: * @return a UChar32 resulting from the partial conversion of source
sl@0: * @see ucnv_toUnicode
sl@0: * @see ucnv_toUChars
sl@0: * @see ucnv_convert
sl@0: * @stable ICU 2.0
sl@0: */
sl@0: U_STABLE UChar32 U_EXPORT2
sl@0: ucnv_getNextUChar(UConverter * converter,
sl@0: const char **source,
sl@0: const char * sourceLimit,
sl@0: UErrorCode * err);
sl@0:
sl@0: /**
sl@0: * Convert from one external charset to another using two existing UConverters.
sl@0: * Internally, two conversions - ucnv_toUnicode() and ucnv_fromUnicode() -
sl@0: * are used, "pivoting" through 16-bit Unicode.
sl@0: *
sl@0: * There is a similar function, ucnv_convert(),
sl@0: * which has the following limitations:
sl@0: * - it takes charset names, not converter objects, so that
sl@0: * - two converters are opened for each call
sl@0: * - only single-string conversion is possible, not streaming operation
sl@0: * - it does not provide enough information to find out,
sl@0: * in case of failure, whether the toUnicode or
sl@0: * the fromUnicode conversion failed
sl@0: *
sl@0: * By contrast, ucnv_convertEx()
sl@0: * - takes UConverter parameters instead of charset names
sl@0: * - fully exposes the pivot buffer for complete error handling
sl@0: *
sl@0: * ucnv_convertEx() also provides further convenience:
sl@0: * - an option to reset the converters at the beginning
sl@0: * (if reset==TRUE, see parameters;
sl@0: * also sets *pivotTarget=*pivotSource=pivotStart)
sl@0: * - allow NUL-terminated input
sl@0: * (only a single NUL byte, will not work for charsets with multi-byte NULs)
sl@0: * (if sourceLimit==NULL, see parameters)
sl@0: * - terminate with a NUL on output
sl@0: * (only a single NUL byte, not useful for charsets with multi-byte NULs),
sl@0: * or set U_STRING_NOT_TERMINATED_WARNING if the output exactly fills
sl@0: * the target buffer
sl@0: * - the pivot buffer can be provided internally;
sl@0: * in this case, the caller will not be able to get details about where an
sl@0: * error occurred
sl@0: * (if pivotStart==NULL, see below)
sl@0: *
sl@0: * The function returns when one of the following is true:
sl@0: * - the entire source text has been converted successfully to the target buffer
sl@0: * - a target buffer overflow occurred (U_BUFFER_OVERFLOW_ERROR)
sl@0: * - a conversion error occurred
sl@0: * (other U_FAILURE(), see description of pErrorCode)
sl@0: *
sl@0: * Limitation compared to the direct use of
sl@0: * ucnv_fromUnicode() and ucnv_toUnicode():
sl@0: * ucnv_convertEx() does not provide offset information.
sl@0: *
sl@0: * Limitation compared to ucnv_fromUChars() and ucnv_toUChars():
sl@0: * ucnv_convertEx() does not support preflighting directly.
sl@0: *
sl@0: * Sample code for converting a single string from
sl@0: * one external charset to UTF-8, ignoring the location of errors:
sl@0: *
sl@0: * \code
sl@0: * int32_t
sl@0: * myToUTF8(UConverter *cnv,
sl@0: * const char *s, int32_t length,
sl@0: * char *u8, int32_t capacity,
sl@0: * UErrorCode *pErrorCode) {
sl@0: * UConverter *utf8Cnv;
sl@0: * char *target;
sl@0: *
sl@0: * if(U_FAILURE(*pErrorCode)) {
sl@0: * return 0;
sl@0: * }
sl@0: *
sl@0: * utf8Cnv=myGetCachedUTF8Converter(pErrorCode);
sl@0: * if(U_FAILURE(*pErrorCode)) {
sl@0: * return 0;
sl@0: * }
sl@0: *
sl@0: * target=u8;
sl@0: * ucnv_convertEx(cnv, utf8Cnv,
sl@0: * &target, u8+capacity,
sl@0: * &s, length>=0 ? s+length : NULL,
sl@0: * NULL, NULL, NULL, NULL,
sl@0: * TRUE, TRUE,
sl@0: * pErrorCode);
sl@0: *
sl@0: * myReleaseCachedUTF8Converter(utf8Cnv);
sl@0: *
sl@0: * // return the output string length, but without preflighting
sl@0: * return (int32_t)(target-u8);
sl@0: * }
sl@0: * \endcode
sl@0: *
sl@0: * @param targetCnv Output converter, used to convert from the UTF-16 pivot
sl@0: * to the target using ucnv_fromUnicode().
sl@0: * @param sourceCnv Input converter, used to convert from the source to
sl@0: * the UTF-16 pivot using ucnv_toUnicode().
sl@0: * @param target I/O parameter, same as for ucnv_fromUChars().
sl@0: * Input: *target points to the beginning of the target buffer.
sl@0: * Output: *target points to the first unit after the last char written.
sl@0: * @param targetLimit Pointer to the first unit after the target buffer.
sl@0: * @param source I/O parameter, same as for ucnv_toUChars().
sl@0: * Input: *source points to the beginning of the source buffer.
sl@0: * Output: *source points to the first unit after the last char read.
sl@0: * @param sourceLimit Pointer to the first unit after the source buffer.
sl@0: * @param pivotStart Pointer to the UTF-16 pivot buffer. If pivotStart==NULL,
sl@0: * then an internal buffer is used and the other pivot
sl@0: * arguments are ignored and can be NULL as well.
sl@0: * @param pivotSource I/O parameter, same as source in ucnv_fromUChars() for
sl@0: * conversion from the pivot buffer to the target buffer.
sl@0: * @param pivotTarget I/O parameter, same as target in ucnv_toUChars() for
sl@0: * conversion from the source buffer to the pivot buffer.
sl@0: * It must be pivotStart<=*pivotSource<=*pivotTarget<=pivotLimit
sl@0: * and pivotStartucnv_countAliases() string-pointers
sl@0: * (const char *) that will be filled in.
sl@0: * The strings themselves are owned by the library.
sl@0: * @param pErrorCode result of operation
sl@0: * @stable ICU 2.0
sl@0: */
sl@0: U_STABLE void U_EXPORT2
sl@0: ucnv_getAliases(const char *alias, const char **aliases, UErrorCode *pErrorCode);
sl@0:
sl@0: /**
sl@0: * Return a new UEnumeration object for enumerating all the
sl@0: * alias names for a given converter that are recognized by a standard.
sl@0: * This method only enumerates the listed entries in the alias file.
sl@0: * The convrtrs.txt file can be modified to change the results of
sl@0: * this function.
sl@0: * The first result in this list is the same result given by
sl@0: * ucnv_getStandardName, which is the default alias for
sl@0: * the specified standard name. The returned object must be closed with
sl@0: * uenum_close when you are done with the object.
sl@0: *
sl@0: * @param convName original converter name
sl@0: * @param standard name of the standard governing the names; MIME and IANA
sl@0: * are such standards
sl@0: * @param pErrorCode The error code
sl@0: * @return A UEnumeration object for getting all aliases that are recognized
sl@0: * by a standard. If any of the parameters are invalid, NULL
sl@0: * is returned.
sl@0: * @see ucnv_getStandardName
sl@0: * @see uenum_close
sl@0: * @see uenum_next
sl@0: * @stable ICU 2.2
sl@0: */
sl@0: U_STABLE UEnumeration * U_EXPORT2
sl@0: ucnv_openStandardNames(const char *convName,
sl@0: const char *standard,
sl@0: UErrorCode *pErrorCode);
sl@0:
sl@0: /**
sl@0: * Gives the number of standards associated to converter names.
sl@0: * @return number of standards
sl@0: * @stable ICU 2.0
sl@0: */
sl@0: U_STABLE uint16_t U_EXPORT2
sl@0: ucnv_countStandards(void);
sl@0:
sl@0: /**
sl@0: * Gives the name of the standard at given index of standard list.
sl@0: * @param n index in standard list
sl@0: * @param pErrorCode result of operation
sl@0: * @return returns the name of the standard at given index. Owned by the library.
sl@0: * @stable ICU 2.0
sl@0: */
sl@0: U_STABLE const char * U_EXPORT2
sl@0: ucnv_getStandard(uint16_t n, UErrorCode *pErrorCode);
sl@0:
sl@0: /**
sl@0: * Returns a standard name for a given converter name.
sl@0: *
sl@0: * Example alias table:
sl@0: * conv alias1 { STANDARD1 } alias2 { STANDARD1* }
sl@0: *
sl@0: * Result of ucnv_getStandardName("conv", "STANDARD1") from example
sl@0: * alias table:
sl@0: * "alias2"
sl@0: *
sl@0: * @param name original converter name
sl@0: * @param standard name of the standard governing the names; MIME and IANA
sl@0: * are such standards
sl@0: * @param pErrorCode result of operation
sl@0: * @return returns the standard converter name;
sl@0: * if a standard converter name cannot be determined,
sl@0: * then NULL is returned. Owned by the library.
sl@0: * @stable ICU 2.0
sl@0: */
sl@0: U_STABLE const char * U_EXPORT2
sl@0: ucnv_getStandardName(const char *name, const char *standard, UErrorCode *pErrorCode);
sl@0:
sl@0: /**
sl@0: * This function will return the internal canonical converter name of the
sl@0: * tagged alias. This is the opposite of ucnv_openStandardNames, which
sl@0: * returns the tagged alias given the canonical name.
sl@0: *
sl@0: * Example alias table:
sl@0: * conv alias1 { STANDARD1 } alias2 { STANDARD1* }
sl@0: *
sl@0: * Result of ucnv_getStandardName("alias1", "STANDARD1") from example
sl@0: * alias table:
sl@0: * "conv"
sl@0: *
sl@0: * @return returns the canonical converter name;
sl@0: * if a standard or alias name cannot be determined,
sl@0: * then NULL is returned. The returned string is
sl@0: * owned by the library.
sl@0: * @see ucnv_getStandardName
sl@0: * @stable ICU 2.4
sl@0: */
sl@0: U_STABLE const char * U_EXPORT2
sl@0: ucnv_getCanonicalName(const char *alias, const char *standard, UErrorCode *pErrorCode);
sl@0:
sl@0: /**
sl@0: * returns the current default converter name.
sl@0: *
sl@0: * @return returns the current default converter name;
sl@0: * if a default converter name cannot be determined,
sl@0: * then NULL is returned.
sl@0: * Storage owned by the library
sl@0: * @see ucnv_setDefaultName
sl@0: * @stable ICU 2.0
sl@0: */
sl@0: U_STABLE const char * U_EXPORT2
sl@0: ucnv_getDefaultName(void);
sl@0:
sl@0: /**
sl@0: * sets the current default converter name. Caller must own the storage for 'name'
sl@0: * and preserve it indefinitely.
sl@0: * @param name the converter name to be the default (must exist).
sl@0: * @see ucnv_getDefaultName
sl@0: * @system SYSTEM API
sl@0: * @stable ICU 2.0
sl@0: */
sl@0: U_STABLE void U_EXPORT2
sl@0: ucnv_setDefaultName(const char *name);
sl@0:
sl@0: /**
sl@0: * Fixes the backslash character mismapping. For example, in SJIS, the backslash
sl@0: * character in the ASCII portion is also used to represent the yen currency sign.
sl@0: * When mapping from Unicode character 0x005C, it's unclear whether to map the
sl@0: * character back to yen or backslash in SJIS. This function will take the input
sl@0: * buffer and replace all the yen sign characters with backslash. This is necessary
sl@0: * when the user tries to open a file with the input buffer on Windows.
sl@0: * This function will test the converter to see whether such mapping is
sl@0: * required. You can sometimes avoid using this function by using the correct version
sl@0: * of Shift-JIS.
sl@0: *
sl@0: * @param cnv The converter representing the target codepage.
sl@0: * @param source the input buffer to be fixed
sl@0: * @param sourceLen the length of the input buffer
sl@0: * @see ucnv_isAmbiguous
sl@0: * @stable ICU 2.0
sl@0: */
sl@0: U_STABLE void U_EXPORT2
sl@0: ucnv_fixFileSeparator(const UConverter *cnv, UChar *source, int32_t sourceLen);
sl@0:
sl@0: /**
sl@0: * Determines if the converter contains ambiguous mappings of the same
sl@0: * character or not.
sl@0: * @param cnv the converter to be tested
sl@0: * @return TRUE if the converter contains ambiguous mapping of the same
sl@0: * character, FALSE otherwise.
sl@0: * @stable ICU 2.0
sl@0: */
sl@0: U_STABLE UBool U_EXPORT2
sl@0: ucnv_isAmbiguous(const UConverter *cnv);
sl@0:
sl@0: /**
sl@0: * Sets the converter to use fallback mapping or not.
sl@0: * @param cnv The converter to set the fallback mapping usage on.
sl@0: * @param usesFallback TRUE if the user wants the converter to take advantage of the fallback
sl@0: * mapping, FALSE otherwise.
sl@0: * @stable ICU 2.0
sl@0: */
sl@0: U_STABLE void U_EXPORT2
sl@0: ucnv_setFallback(UConverter *cnv, UBool usesFallback);
sl@0:
sl@0: /**
sl@0: * Determines if the converter uses fallback mappings or not.
sl@0: * @param cnv The converter to be tested
sl@0: * @return TRUE if the converter uses fallback, FALSE otherwise.
sl@0: * @stable ICU 2.0
sl@0: */
sl@0: U_STABLE UBool U_EXPORT2
sl@0: ucnv_usesFallback(const UConverter *cnv);
sl@0:
sl@0: /**
sl@0: * Detects Unicode signature byte sequences at the start of the byte stream
sl@0: * and returns the charset name of the indicated Unicode charset.
sl@0: * NULL is returned when no Unicode signature is recognized.
sl@0: * The number of bytes in the signature is output as well.
sl@0: *
sl@0: * The caller can ucnv_open() a converter using the charset name.
sl@0: * The first code unit (UChar) from the start of the stream will be U+FEFF
sl@0: * (the Unicode BOM/signature character) and can usually be ignored.
sl@0: *
sl@0: * For most Unicode charsets it is also possible to ignore the indicated
sl@0: * number of initial stream bytes and start converting after them.
sl@0: * However, there are stateful Unicode charsets (UTF-7 and BOCU-1) for which
sl@0: * this will not work. Therefore, it is best to ignore the first output UChar
sl@0: * instead of the input signature bytes.
sl@0: *
sl@0: * Usage: sl@0: * @code sl@0: * UErrorCode err = U_ZERO_ERROR; sl@0: * char input[] = { '\xEF','\xBB', '\xBF','\x41','\x42','\x43' }; sl@0: * int32_t signatureLength = 0; sl@0: * char *encoding = ucnv_detectUnicodeSignatures(input,sizeof(input),&signatureLength,&err); sl@0: * UConverter *conv = NULL; sl@0: * UChar output[100]; sl@0: * UChar *target = output, *out; sl@0: * char *source = input; sl@0: * if(encoding!=NULL && U_SUCCESS(err)){ sl@0: * // should signature be discarded ? sl@0: * conv = ucnv_open(encoding, &err); sl@0: * // do the conversion sl@0: * ucnv_toUnicode(conv, sl@0: * target, output + sizeof(output)/U_SIZEOF_UCHAR, sl@0: * source, input + sizeof(input), sl@0: * NULL, TRUE, &err); sl@0: * out = output; sl@0: * if (discardSignature){ sl@0: * ++out; // ignore initial U+FEFF sl@0: * } sl@0: * while(out != target) { sl@0: * printf("%04x ", *out++); sl@0: * } sl@0: * puts(""); sl@0: * } sl@0: * sl@0: * @endcode sl@0: * sl@0: * @param source The source string in which the signature should be detected. sl@0: * @param sourceLength Length of the input string, or -1 if terminated with a NUL byte. sl@0: * @param signatureLength A pointer to int32_t to receive the number of bytes that make up the signature sl@0: * of the detected UTF. 0 if not detected. sl@0: * Can be a NULL pointer. sl@0: * @param pErrorCode ICU error code in/out parameter. sl@0: * Must fulfill U_SUCCESS before the function call. sl@0: * @return The name of the encoding detected. NULL if encoding is not detected. sl@0: * @stable ICU 2.4 sl@0: */ sl@0: U_STABLE const char* U_EXPORT2 sl@0: ucnv_detectUnicodeSignature(const char* source, sl@0: int32_t sourceLength, sl@0: int32_t *signatureLength, sl@0: UErrorCode *pErrorCode); sl@0: sl@0: /** sl@0: * Returns the number of UChars held in the converter's internal state sl@0: * because more input is needed for completing the conversion. This function is sl@0: * useful for mapping semantics of ICU's converter interface to those of iconv, sl@0: * and this information is not needed for normal conversion. sl@0: * @param cnv The converter in which the input is held sl@0: * @param status ICU error code in/out parameter. sl@0: * Must fulfill U_SUCCESS before the function call. sl@0: * @return The number of UChars in the state. -1 if an error is encountered. sl@0: * @draft ICU 3.4 sl@0: */ sl@0: U_DRAFT int32_t U_EXPORT2 sl@0: ucnv_fromUCountPending(const UConverter* cnv, UErrorCode* status); sl@0: sl@0: /** sl@0: * Returns the number of chars held in the converter's internal state sl@0: * because more input is needed for completing the conversion. This function is sl@0: * useful for mapping semantics of ICU's converter interface to those of iconv, sl@0: * and this information is not needed for normal conversion. sl@0: * @param cnv The converter in which the input is held as internal state sl@0: * @param status ICU error code in/out parameter. sl@0: * Must fulfill U_SUCCESS before the function call. sl@0: * @return The number of chars in the state. -1 if an error is encountered. sl@0: * @draft ICU 3.4 sl@0: */ sl@0: U_DRAFT int32_t U_EXPORT2 sl@0: ucnv_toUCountPending(const UConverter* cnv, UErrorCode* status); sl@0: sl@0: #endif sl@0: sl@0: #endif sl@0: /*_UCNV*/