jbe@0: /* jbe@10: * Copyright (c) 2009 Public Software Group e. V., Berlin, Germany jbe@0: * jbe@7: * Permission is hereby granted, free of charge, to any person obtaining a jbe@7: * copy of this software and associated documentation files (the "Software"), jbe@7: * to deal in the Software without restriction, including without limitation jbe@7: * the rights to use, copy, modify, merge, publish, distribute, sublicense, jbe@7: * and/or sell copies of the Software, and to permit persons to whom the jbe@7: * Software is furnished to do so, subject to the following conditions: jbe@0: * jbe@7: * The above copyright notice and this permission notice shall be included in jbe@7: * all copies or substantial portions of the Software. jbe@0: * jbe@7: * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR jbe@7: * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, jbe@7: * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE jbe@7: * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER jbe@7: * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING jbe@7: * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER jbe@7: * DEALINGS IN THE SOFTWARE. jbe@0: */ jbe@7: jbe@0: jbe@0: /* jbe@0: * File name: utf8proc.h jbe@0: * jbe@0: * Description: jbe@0: * Header files for libutf8proc, which is a mapping tool for UTF-8 strings jbe@0: * with following features: jbe@0: * - decomposing and composing of strings jbe@0: * - replacing compatibility characters with their equivalents jbe@0: * - stripping of "default ignorable characters" jbe@0: * like SOFT-HYPHEN or ZERO-WIDTH-SPACE jbe@3: * - folding of certain characters for string comparison jbe@3: * (e.g. HYPHEN U+2010 and MINUS U+2212 to ASCII "-") jbe@3: * (see "LUMP" option) jbe@0: * - optional rejection of strings containing non-assigned code points jbe@0: * - stripping of control characters jbe@3: * - stripping of character marks (accents, etc.) jbe@0: * - transformation of LF, CRLF, CR and NEL to line-feed (LF) jbe@0: * or to the unicode chararacters for paragraph separation (PS) jbe@0: * or line separation (LS). jbe@0: * - unicode case folding (for case insensitive string comparisons) jbe@7: * - rejection of illegal UTF-8 data jbe@7: * (i.e. UTF-8 encoded UTF-16 surrogates) jbe@0: * - support for korean hangul characters Jiahao@15: * Unicode Version 7.0.0 is supported. jbe@0: */ jbe@0: jbe@0: jbe@0: #ifndef UTF8PROC_H jbe@0: #define UTF8PROC_H jbe@0: jbe@0: jbe@0: #include Jan@12: #include Jan@12: #ifdef _MSC_VER Jan@12: typedef signed char int8_t; Jan@12: typedef unsigned char uint8_t; Jan@12: typedef short int16_t; Jan@12: typedef unsigned short uint16_t; Jan@12: typedef int int32_t; Jiahao@15: # ifdef _WIN64 Jiahao@15: # define ssize_t __int64 Jiahao@15: # else Jiahao@15: # define ssize_t int Jiahao@15: # endif Jiahao@15: # ifndef __cplusplus Jan@12: typedef unsigned char bool; Jan@12: enum {false, true}; Jiahao@15: # endif Jan@12: #else Jiahao@15: # include Jiahao@15: # include Jan@12: #endif jbe@0: #include jbe@0: Jiahao@15: #ifdef _WIN32 Jiahao@15: # ifdef UTF8PROC_EXPORTS Jiahao@15: # define DLLEXPORT __declspec(dllexport) Jiahao@15: # else Jiahao@15: # define DLLEXPORT __declspec(dllimport) Jiahao@15: # endif Jiahao@15: #elif __GNUC__ >= 4 Jiahao@15: # define DLLEXPORT __attribute__ ((visibility("default"))) Jiahao@15: #else Jiahao@15: # define DLLEXPORT Jiahao@15: #endif Jiahao@15: Jan@12: #ifdef __cplusplus Jan@12: extern "C" { Jan@12: #endif Jan@12: jbe@0: #ifndef SSIZE_MAX jbe@10: #define SSIZE_MAX ((size_t)SIZE_MAX/2) jbe@0: #endif jbe@0: jbe@2: #define UTF8PROC_NULLTERM (1<<0) jbe@2: #define UTF8PROC_STABLE (1<<1) jbe@2: #define UTF8PROC_COMPAT (1<<2) jbe@2: #define UTF8PROC_COMPOSE (1<<3) jbe@2: #define UTF8PROC_DECOMPOSE (1<<4) jbe@2: #define UTF8PROC_IGNORE (1<<5) jbe@2: #define UTF8PROC_REJECTNA (1<<6) jbe@2: #define UTF8PROC_NLF2LS (1<<7) jbe@2: #define UTF8PROC_NLF2PS (1<<8) jbe@2: #define UTF8PROC_NLF2LF (UTF8PROC_NLF2LS | UTF8PROC_NLF2PS) jbe@2: #define UTF8PROC_STRIPCC (1<<9) jbe@2: #define UTF8PROC_CASEFOLD (1<<10) jbe@2: #define UTF8PROC_CHARBOUND (1<<11) jbe@3: #define UTF8PROC_LUMP (1<<12) jbe@3: #define UTF8PROC_STRIPMARK (1<<13) jbe@0: /* jbe@0: * Flags being regarded by several functions in the library: jbe@2: * NULLTERM: The given UTF-8 input is NULL terminated. jbe@2: * STABLE: Unicode Versioning Stability has to be respected. Jiahao@15: * COMPAT: Compatibility decomposition jbe@2: * (i.e. formatting information is lost) jbe@2: * COMPOSE: Return a result with composed characters. jbe@2: * DECOMPOSE: Return a result with decomposed characters. jbe@2: * IGNORE: Strip "default ignorable characters" jbe@7: * REJECTNA: Return an error, if the input contains unassigned jbe@7: * code points. jbe@2: * NLF2LS: Indicating that NLF-sequences (LF, CRLF, CR, NEL) are jbe@2: * representing a line break, and should be converted to the jbe@2: * unicode character for line separation (LS). jbe@2: * NLF2PS: Indicating that NLF-sequences are representing a paragraph jbe@2: * break, and should be converted to the unicode character for jbe@2: * paragraph separation (PS). jbe@2: * NLF2LF: Indicating that the meaning of NLF-sequences is unknown. jbe@2: * STRIPCC: Strips and/or convers control characters. jbe@7: * NLF-sequences are transformed into space, except if one of jbe@7: * the NLF2LS/PS/LF options is given. jbe@2: * HorizontalTab (HT) and FormFeed (FF) are treated as a jbe@2: * NLF-sequence in this case. jbe@2: * All other control characters are simply removed. jbe@2: * CASEFOLD: Performs unicode case folding, to be able to do a jbe@2: * case-insensitive string comparison. jbe@7: * CHARBOUND: Inserts 0xFF bytes at the beginning of each sequence which jbe@7: * is representing a single grapheme cluster (see UAX#29). jbe@3: * LUMP: Lumps certain characters together jbe@3: * (e.g. HYPHEN U+2010 and MINUS U+2212 to ASCII "-"). jbe@3: * (See lump.txt for details.) jbe@7: * If NLF2LF is set, this includes a transformation of jbe@7: * paragraph and line separators to ASCII line-feed (LF). jbe@3: * STRIPMARK: Strips all character markings jbe@3: * (non-spacing, spacing and enclosing) (i.e. accents) jbe@3: * NOTE: this option works only with COMPOSE or DECOMPOSE jbe@0: */ jbe@0: jbe@0: #define UTF8PROC_ERROR_NOMEM -1 jbe@0: #define UTF8PROC_ERROR_OVERFLOW -2 jbe@0: #define UTF8PROC_ERROR_INVALIDUTF8 -3 jbe@0: #define UTF8PROC_ERROR_NOTASSIGNED -4 jbe@3: #define UTF8PROC_ERROR_INVALIDOPTS -5 jbe@0: /* jbe@0: * Error codes being returned by almost all functions: jbe@0: * ERROR_NOMEM: Memory could not be allocated. jbe@0: * ERROR_OVERFLOW: The given string is too long to be processed. jbe@0: * ERROR_INVALIDUTF8: The given string is not a legal UTF-8 string. jbe@0: * ERROR_NOTASSIGNED: The REJECTNA flag was set, jbe@0: * and an unassigned code point was found. jbe@3: * ERROR_INVALIDOPTS: Invalid options have been used. jbe@0: */ jbe@0: jbe@3: typedef int16_t utf8proc_propval_t; jbe@0: typedef struct utf8proc_property_struct { jbe@3: utf8proc_propval_t category; jbe@3: utf8proc_propval_t combining_class; jbe@3: utf8proc_propval_t bidi_class; jbe@3: utf8proc_propval_t decomp_type; jbe@0: const int32_t *decomp_mapping; jbe@7: unsigned bidi_mirrored:1; jbe@7: int32_t uppercase_mapping; jbe@7: int32_t lowercase_mapping; jbe@7: int32_t titlecase_mapping; jbe@7: int32_t comb1st_index; jbe@7: int32_t comb2nd_index; jbe@7: unsigned comp_exclusion:1; jbe@7: unsigned ignorable:1; jbe@7: unsigned control_boundary:1; jbe@7: unsigned extend:1; jbe@0: const int32_t *casefold_mapping; jbe@0: } utf8proc_property_t; jbe@0: jbe@2: #define UTF8PROC_CATEGORY_LU 1 jbe@2: #define UTF8PROC_CATEGORY_LL 2 jbe@2: #define UTF8PROC_CATEGORY_LT 3 jbe@2: #define UTF8PROC_CATEGORY_LM 4 jbe@2: #define UTF8PROC_CATEGORY_LO 5 jbe@2: #define UTF8PROC_CATEGORY_MN 6 jbe@2: #define UTF8PROC_CATEGORY_MC 7 jbe@2: #define UTF8PROC_CATEGORY_ME 8 jbe@2: #define UTF8PROC_CATEGORY_ND 9 jbe@2: #define UTF8PROC_CATEGORY_NL 10 jbe@2: #define UTF8PROC_CATEGORY_NO 11 jbe@2: #define UTF8PROC_CATEGORY_PC 12 jbe@2: #define UTF8PROC_CATEGORY_PD 13 jbe@2: #define UTF8PROC_CATEGORY_PS 14 jbe@2: #define UTF8PROC_CATEGORY_PE 15 jbe@2: #define UTF8PROC_CATEGORY_PI 16 jbe@2: #define UTF8PROC_CATEGORY_PF 17 jbe@2: #define UTF8PROC_CATEGORY_PO 18 jbe@2: #define UTF8PROC_CATEGORY_SM 19 jbe@2: #define UTF8PROC_CATEGORY_SC 20 jbe@2: #define UTF8PROC_CATEGORY_SK 21 jbe@2: #define UTF8PROC_CATEGORY_SO 22 jbe@2: #define UTF8PROC_CATEGORY_ZS 23 jbe@2: #define UTF8PROC_CATEGORY_ZL 24 jbe@2: #define UTF8PROC_CATEGORY_ZP 25 jbe@2: #define UTF8PROC_CATEGORY_CC 26 jbe@2: #define UTF8PROC_CATEGORY_CF 27 jbe@2: #define UTF8PROC_CATEGORY_CS 28 jbe@2: #define UTF8PROC_CATEGORY_CO 29 jbe@2: #define UTF8PROC_CATEGORY_CN 30 jbe@2: #define UTF8PROC_BIDI_CLASS_L 1 jbe@2: #define UTF8PROC_BIDI_CLASS_LRE 2 jbe@2: #define UTF8PROC_BIDI_CLASS_LRO 3 jbe@2: #define UTF8PROC_BIDI_CLASS_R 4 jbe@2: #define UTF8PROC_BIDI_CLASS_AL 5 jbe@2: #define UTF8PROC_BIDI_CLASS_RLE 6 jbe@2: #define UTF8PROC_BIDI_CLASS_RLO 7 jbe@2: #define UTF8PROC_BIDI_CLASS_PDF 8 jbe@2: #define UTF8PROC_BIDI_CLASS_EN 9 jbe@2: #define UTF8PROC_BIDI_CLASS_ES 10 jbe@2: #define UTF8PROC_BIDI_CLASS_ET 11 jbe@2: #define UTF8PROC_BIDI_CLASS_AN 12 jbe@2: #define UTF8PROC_BIDI_CLASS_CS 13 jbe@2: #define UTF8PROC_BIDI_CLASS_NSM 14 jbe@2: #define UTF8PROC_BIDI_CLASS_BN 15 jbe@2: #define UTF8PROC_BIDI_CLASS_B 16 jbe@2: #define UTF8PROC_BIDI_CLASS_S 17 jbe@2: #define UTF8PROC_BIDI_CLASS_WS 18 jbe@2: #define UTF8PROC_BIDI_CLASS_ON 19 Jiahao@15: #define UTF8PROC_BIDI_CLASS_LRI 20 /* new in Unicode 6.3 */ Jiahao@15: #define UTF8PROC_BIDI_CLASS_RLI 21 /* new in Unicode 6.3 */ Jiahao@15: #define UTF8PROC_BIDI_CLASS_FSI 22 /* new in Unicode 6.3 */ Jiahao@15: #define UTF8PROC_BIDI_CLASS_PDI 23 /* new in Unicode 6.3 */ jbe@2: #define UTF8PROC_DECOMP_TYPE_FONT 1 jbe@2: #define UTF8PROC_DECOMP_TYPE_NOBREAK 2 jbe@2: #define UTF8PROC_DECOMP_TYPE_INITIAL 3 jbe@2: #define UTF8PROC_DECOMP_TYPE_MEDIAL 4 jbe@2: #define UTF8PROC_DECOMP_TYPE_FINAL 5 jbe@2: #define UTF8PROC_DECOMP_TYPE_ISOLATED 6 jbe@2: #define UTF8PROC_DECOMP_TYPE_CIRCLE 7 jbe@2: #define UTF8PROC_DECOMP_TYPE_SUPER 8 jbe@2: #define UTF8PROC_DECOMP_TYPE_SUB 9 jbe@2: #define UTF8PROC_DECOMP_TYPE_VERTICAL 10 jbe@2: #define UTF8PROC_DECOMP_TYPE_WIDE 11 jbe@2: #define UTF8PROC_DECOMP_TYPE_NARROW 12 jbe@2: #define UTF8PROC_DECOMP_TYPE_SMALL 13 jbe@2: #define UTF8PROC_DECOMP_TYPE_SQUARE 14 jbe@2: #define UTF8PROC_DECOMP_TYPE_FRACTION 15 jbe@2: #define UTF8PROC_DECOMP_TYPE_COMPAT 16 jbe@2: Jiahao@15: DLLEXPORT extern const int8_t utf8proc_utf8class[256]; jbe@0: Jiahao@15: DLLEXPORT const char *utf8proc_version(void); jbe@9: Jiahao@15: DLLEXPORT const char *utf8proc_errmsg(ssize_t errcode); jbe@0: /* jbe@0: * Returns a static error string for the given error code. jbe@0: */ jbe@0: Jiahao@15: DLLEXPORT ssize_t utf8proc_iterate(const uint8_t *str, ssize_t strlen, int32_t *dst); jbe@0: /* jbe@0: * Reads a single char from the UTF-8 sequence being pointed to by 'str'. jbe@0: * The maximum number of bytes read is 'strlen', unless 'strlen' is jbe@0: * negative. jbe@0: * If a valid unicode char could be read, it is stored in the variable jbe@0: * being pointed to by 'dst', otherwise that variable will be set to -1. jbe@0: * In case of success the number of bytes read is returned, otherwise a jbe@0: * negative error code is returned. jbe@0: */ jbe@0: Jiahao@15: DLLEXPORT bool utf8proc_codepoint_valid(int32_t uc); jbe@7: /* jbe@7: * Returns 1, if the given unicode code-point is valid, otherwise 0. jbe@7: */ jbe@7: Jiahao@15: DLLEXPORT ssize_t utf8proc_encode_char(int32_t uc, uint8_t *dst); jbe@0: /* jbe@0: * Encodes the unicode char with the code point 'uc' as an UTF-8 string in jbe@0: * the byte array being pointed to by 'dst'. This array has to be at least jbe@0: * 4 bytes long. jbe@7: * In case of success the number of bytes written is returned, jbe@7: * otherwise 0. jbe@0: * This function does not check if 'uc' is a valid unicode code point. jbe@0: */ jbe@0: Jiahao@15: DLLEXPORT const utf8proc_property_t *utf8proc_get_property(int32_t uc); jbe@0: /* jbe@0: * Returns a pointer to a (constant) struct containing information about jbe@0: * the unicode char with the given code point 'uc'. jbe@0: * If the character is not existent a pointer to a special struct is jbe@0: * returned, where 'category' is a NULL pointer. jbe@0: * WARNING: The parameter 'uc' has to be in the range of 0x0000 to jbe@0: * 0x10FFFF, otherwise the program might crash! jbe@0: */ jbe@0: Jiahao@15: DLLEXPORT ssize_t utf8proc_decompose_char( jbe@7: int32_t uc, int32_t *dst, ssize_t bufsize, jbe@7: int options, int *last_boundclass jbe@7: ); jbe@0: /* jbe@0: * Writes a decomposition of the unicode char 'uc' into the array being jbe@0: * pointed to by 'dst'. jbe@0: * Following flags in the 'options' field are regarded: jbe@2: * REJECTNA: an unassigned unicode code point leads to an error jbe@2: * IGNORE: "default ignorable" chars are stripped jbe@2: * CASEFOLD: unicode casefolding is applied jbe@2: * COMPAT: replace certain characters with their jbe@2: * compatibility decomposition jbe@2: * CHARBOUND: Inserts 0xFF bytes before each grapheme cluster jbe@3: * LUMP: lumps certain different characters together jbe@3: * STRIPMARK: removes all character marks jbe@7: * The pointer 'last_boundclass' has to point to an integer variable which jbe@7: * is storing the last character boundary class, if the CHARBOUND option jbe@7: * is used. jbe@0: * In case of success the number of chars written is returned, jbe@0: * in case of an error, a negative error code is returned. jbe@0: * If the number of written chars would be bigger than 'bufsize', jbe@0: * the buffer (up to 'bufsize') has inpredictable data, and the needed jbe@0: * buffer size is returned. jbe@0: * WARNING: The parameter 'uc' has to be in the range of 0x0000 to jbe@0: * 0x10FFFF, otherwise the program might crash! jbe@0: */ jbe@0: Jiahao@15: DLLEXPORT ssize_t utf8proc_decompose( jbe@7: const uint8_t *str, ssize_t strlen, jbe@7: int32_t *buffer, ssize_t bufsize, int options jbe@7: ); jbe@0: /* jbe@0: * Does the same as 'utf8proc_decompose_char', but acts on a whole UTF-8 jbe@0: * string, and orders the decomposed sequences correctly. jbe@0: * If the NULLTERM flag in 'options' is set, processing will be stopped, jbe@0: * when a NULL byte is encounted, otherwise 'strlen' bytes are processed. jbe@0: * The result in form of unicode code points is written into the buffer jbe@0: * being pointed to by 'buffer', having the length of 'bufsize' entries. jbe@0: * In case of success the number of chars written is returned, jbe@0: * in case of an error, a negative error code is returned. jbe@0: * If the number of written chars would be bigger than 'bufsize', jbe@0: * the buffer (up to 'bufsize') has inpredictable data, and the needed jbe@0: * buffer size is returned. jbe@0: */ jbe@0: Jiahao@15: DLLEXPORT ssize_t utf8proc_reencode(int32_t *buffer, ssize_t length, int options); jbe@0: /* jbe@0: * Reencodes the sequence of unicode characters given by the pointer jbe@0: * 'buffer' and 'length' as UTF-8. jbe@0: * The result is stored in the same memory area where the data is read. jbe@0: * Following flags in the 'options' field are regarded: jbe@0: * NLF2LS: converts LF, CRLF, CR and NEL into LS jbe@0: * NLF2PS: converts LF, CRLF, CR and NEL into PS jbe@0: * NLF2LF: converts LF, CRLF, CR and NEL into LF jbe@0: * STRIPCC: strips or converts all non-affected control characters jbe@7: * COMPOSE: tries to combine decomposed characters into composite jbe@7: * characters jbe@0: * STABLE: prohibits combining characters which would violate jbe@0: * the unicode versioning stability jbe@7: * In case of success the length of the resulting UTF-8 string is jbe@7: * returned, otherwise a negative error code is returned. jbe@0: * WARNING: The amount of free space being pointed to by 'buffer', has to jbe@0: * exceed the amount of the input data by one byte, and the jbe@0: * entries of the array pointed to by 'str' have to be in the jbe@7: * range of 0x0000 to 0x10FFFF, otherwise the program might jbe@7: * crash! jbe@0: */ jbe@0: Jiahao@15: DLLEXPORT ssize_t utf8proc_map( jbe@7: const uint8_t *str, ssize_t strlen, uint8_t **dstptr, int options jbe@7: ); jbe@0: /* jbe@0: * Maps the given UTF-8 string being pointed to by 'str' to a new UTF-8 jbe@7: * string, which is allocated dynamically, and afterwards pointed to by jbe@7: * the pointer being pointed to by 'dstptr'. jbe@0: * If the NULLTERM flag in the 'options' field is set, the length is jbe@0: * determined by a NULL terminator, otherwise the parameter 'strlen' is jbe@0: * evaluated to determine the string length, but in any case the result jbe@7: * will be NULL terminated (though it might contain NULL characters jbe@7: * before). Other flags in the 'options' field are passed to the functions jbe@7: * defined above, and regarded as described. jbe@0: * In case of success the length of the new string is returned, jbe@0: * otherwise a negative error code is returned. jbe@0: * NOTICE: The memory of the new UTF-8 string will have been allocated with jbe@0: * 'malloc', and has theirfore to be freed with 'free'. jbe@0: */ jbe@0: Jiahao@15: DLLEXPORT uint8_t *utf8proc_NFD(const uint8_t *str); Jiahao@15: DLLEXPORT uint8_t *utf8proc_NFC(const uint8_t *str); Jiahao@15: DLLEXPORT uint8_t *utf8proc_NFKD(const uint8_t *str); Jiahao@15: DLLEXPORT uint8_t *utf8proc_NFKC(const uint8_t *str); jbe@0: /* jbe@0: * Returns a pointer to newly allocated memory of a NFD, NFC, NFKD or NFKC jbe@0: * normalized version of the null-terminated string 'str'. jbe@0: */ jbe@0: Jan@12: #ifdef __cplusplus Jan@12: } Jan@12: #endif jbe@0: jbe@0: #endif jbe@0: