utf8proc

annotate utf8proc.h @ 7:fcfd8c836c64

find changesets by author, revision, files, or words in the commit message
Version 1.1.1

- Added a new PostgreSQL function 'unistrip', which behaves like 'unifold', but also removes all character marks (e.g. accents).
- Changed license from BSD to MIT style.
- Added a new function 'utf8proc_codepoint_valid' to the C library.
- Changed compiler flags in Makefile from -g -O0 to -O2
- The ruby script, which was used to build the utf8proc_data.c file, is now included in the distribution.
author jbe
date Sun Jul 22 12:00:00 2007 +0200 (2007-07-22)
parents 4ee0d5f54af1
children 951e73a98021
rev   line source
jbe@0 1 /*
jbe@7 2 * Copyright (c) 2006-2007 Jan Behrens, FlexiGuided GmbH, Berlin
jbe@0 3 *
jbe@7 4 * Permission is hereby granted, free of charge, to any person obtaining a
jbe@7 5 * copy of this software and associated documentation files (the "Software"),
jbe@7 6 * to deal in the Software without restriction, including without limitation
jbe@7 7 * the rights to use, copy, modify, merge, publish, distribute, sublicense,
jbe@7 8 * and/or sell copies of the Software, and to permit persons to whom the
jbe@7 9 * Software is furnished to do so, subject to the following conditions:
jbe@0 10 *
jbe@7 11 * The above copyright notice and this permission notice shall be included in
jbe@7 12 * all copies or substantial portions of the Software.
jbe@0 13 *
jbe@7 14 * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
jbe@7 15 * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
jbe@7 16 * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
jbe@7 17 * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
jbe@7 18 * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
jbe@7 19 * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
jbe@7 20 * DEALINGS IN THE SOFTWARE.
jbe@0 21 */
jbe@7 22
jbe@0 23
jbe@0 24 /*
jbe@0 25 * File name: utf8proc.h
jbe@7 26 * Version: 1.1.1
jbe@7 27 * Last changed: 2007-07-22
jbe@0 28 *
jbe@0 29 * Description:
jbe@0 30 * Header files for libutf8proc, which is a mapping tool for UTF-8 strings
jbe@0 31 * with following features:
jbe@0 32 * - decomposing and composing of strings
jbe@0 33 * - replacing compatibility characters with their equivalents
jbe@0 34 * - stripping of "default ignorable characters"
jbe@0 35 * like SOFT-HYPHEN or ZERO-WIDTH-SPACE
jbe@3 36 * - folding of certain characters for string comparison
jbe@3 37 * (e.g. HYPHEN U+2010 and MINUS U+2212 to ASCII "-")
jbe@3 38 * (see "LUMP" option)
jbe@0 39 * - optional rejection of strings containing non-assigned code points
jbe@0 40 * - stripping of control characters
jbe@3 41 * - stripping of character marks (accents, etc.)
jbe@0 42 * - transformation of LF, CRLF, CR and NEL to line-feed (LF)
jbe@0 43 * or to the unicode chararacters for paragraph separation (PS)
jbe@0 44 * or line separation (LS).
jbe@0 45 * - unicode case folding (for case insensitive string comparisons)
jbe@7 46 * - rejection of illegal UTF-8 data
jbe@7 47 * (i.e. UTF-8 encoded UTF-16 surrogates)
jbe@0 48 * - support for korean hangul characters
jbe@2 49 * Unicode Version 5.0.0 is supported.
jbe@0 50 */
jbe@0 51
jbe@0 52
jbe@0 53 #ifndef UTF8PROC_H
jbe@0 54 #define UTF8PROC_H
jbe@0 55
jbe@0 56
jbe@0 57 #include <stdlib.h>
jbe@0 58 #include <stdbool.h>
jbe@0 59 #include <sys/types.h>
jbe@0 60 #include <inttypes.h>
jbe@0 61 #include <limits.h>
jbe@0 62
jbe@0 63 #ifndef SSIZE_MAX
jbe@0 64 #define SSIZE_MAX (SIZE_MAX/2)
jbe@0 65 #endif
jbe@0 66
jbe@2 67 #define UTF8PROC_NULLTERM (1<<0)
jbe@2 68 #define UTF8PROC_STABLE (1<<1)
jbe@2 69 #define UTF8PROC_COMPAT (1<<2)
jbe@2 70 #define UTF8PROC_COMPOSE (1<<3)
jbe@2 71 #define UTF8PROC_DECOMPOSE (1<<4)
jbe@2 72 #define UTF8PROC_IGNORE (1<<5)
jbe@2 73 #define UTF8PROC_REJECTNA (1<<6)
jbe@2 74 #define UTF8PROC_NLF2LS (1<<7)
jbe@2 75 #define UTF8PROC_NLF2PS (1<<8)
jbe@2 76 #define UTF8PROC_NLF2LF (UTF8PROC_NLF2LS | UTF8PROC_NLF2PS)
jbe@2 77 #define UTF8PROC_STRIPCC (1<<9)
jbe@2 78 #define UTF8PROC_CASEFOLD (1<<10)
jbe@2 79 #define UTF8PROC_CHARBOUND (1<<11)
jbe@3 80 #define UTF8PROC_LUMP (1<<12)
jbe@3 81 #define UTF8PROC_STRIPMARK (1<<13)
jbe@0 82 /*
jbe@0 83 * Flags being regarded by several functions in the library:
jbe@2 84 * NULLTERM: The given UTF-8 input is NULL terminated.
jbe@2 85 * STABLE: Unicode Versioning Stability has to be respected.
jbe@2 86 * COMPAT: Compatiblity decomposition
jbe@2 87 * (i.e. formatting information is lost)
jbe@2 88 * COMPOSE: Return a result with composed characters.
jbe@2 89 * DECOMPOSE: Return a result with decomposed characters.
jbe@2 90 * IGNORE: Strip "default ignorable characters"
jbe@7 91 * REJECTNA: Return an error, if the input contains unassigned
jbe@7 92 * code points.
jbe@2 93 * NLF2LS: Indicating that NLF-sequences (LF, CRLF, CR, NEL) are
jbe@2 94 * representing a line break, and should be converted to the
jbe@2 95 * unicode character for line separation (LS).
jbe@2 96 * NLF2PS: Indicating that NLF-sequences are representing a paragraph
jbe@2 97 * break, and should be converted to the unicode character for
jbe@2 98 * paragraph separation (PS).
jbe@2 99 * NLF2LF: Indicating that the meaning of NLF-sequences is unknown.
jbe@2 100 * STRIPCC: Strips and/or convers control characters.
jbe@7 101 * NLF-sequences are transformed into space, except if one of
jbe@7 102 * the NLF2LS/PS/LF options is given.
jbe@2 103 * HorizontalTab (HT) and FormFeed (FF) are treated as a
jbe@2 104 * NLF-sequence in this case.
jbe@2 105 * All other control characters are simply removed.
jbe@2 106 * CASEFOLD: Performs unicode case folding, to be able to do a
jbe@2 107 * case-insensitive string comparison.
jbe@7 108 * CHARBOUND: Inserts 0xFF bytes at the beginning of each sequence which
jbe@7 109 * is representing a single grapheme cluster (see UAX#29).
jbe@3 110 * LUMP: Lumps certain characters together
jbe@3 111 * (e.g. HYPHEN U+2010 and MINUS U+2212 to ASCII "-").
jbe@3 112 * (See lump.txt for details.)
jbe@7 113 * If NLF2LF is set, this includes a transformation of
jbe@7 114 * paragraph and line separators to ASCII line-feed (LF).
jbe@3 115 * STRIPMARK: Strips all character markings
jbe@3 116 * (non-spacing, spacing and enclosing) (i.e. accents)
jbe@3 117 * NOTE: this option works only with COMPOSE or DECOMPOSE
jbe@0 118 */
jbe@0 119
jbe@0 120 #define UTF8PROC_ERROR_NOMEM -1
jbe@0 121 #define UTF8PROC_ERROR_OVERFLOW -2
jbe@0 122 #define UTF8PROC_ERROR_INVALIDUTF8 -3
jbe@0 123 #define UTF8PROC_ERROR_NOTASSIGNED -4
jbe@3 124 #define UTF8PROC_ERROR_INVALIDOPTS -5
jbe@0 125 /*
jbe@0 126 * Error codes being returned by almost all functions:
jbe@0 127 * ERROR_NOMEM: Memory could not be allocated.
jbe@0 128 * ERROR_OVERFLOW: The given string is too long to be processed.
jbe@0 129 * ERROR_INVALIDUTF8: The given string is not a legal UTF-8 string.
jbe@0 130 * ERROR_NOTASSIGNED: The REJECTNA flag was set,
jbe@0 131 * and an unassigned code point was found.
jbe@3 132 * ERROR_INVALIDOPTS: Invalid options have been used.
jbe@0 133 */
jbe@0 134
jbe@3 135 typedef int16_t utf8proc_propval_t;
jbe@0 136 typedef struct utf8proc_property_struct {
jbe@3 137 utf8proc_propval_t category;
jbe@3 138 utf8proc_propval_t combining_class;
jbe@3 139 utf8proc_propval_t bidi_class;
jbe@3 140 utf8proc_propval_t decomp_type;
jbe@0 141 const int32_t *decomp_mapping;
jbe@7 142 unsigned bidi_mirrored:1;
jbe@7 143 int32_t uppercase_mapping;
jbe@7 144 int32_t lowercase_mapping;
jbe@7 145 int32_t titlecase_mapping;
jbe@7 146 int32_t comb1st_index;
jbe@7 147 int32_t comb2nd_index;
jbe@7 148 unsigned comp_exclusion:1;
jbe@7 149 unsigned ignorable:1;
jbe@7 150 unsigned control_boundary:1;
jbe@7 151 unsigned extend:1;
jbe@0 152 const int32_t *casefold_mapping;
jbe@0 153 } utf8proc_property_t;
jbe@0 154
jbe@2 155 #define UTF8PROC_CATEGORY_LU 1
jbe@2 156 #define UTF8PROC_CATEGORY_LL 2
jbe@2 157 #define UTF8PROC_CATEGORY_LT 3
jbe@2 158 #define UTF8PROC_CATEGORY_LM 4
jbe@2 159 #define UTF8PROC_CATEGORY_LO 5
jbe@2 160 #define UTF8PROC_CATEGORY_MN 6
jbe@2 161 #define UTF8PROC_CATEGORY_MC 7
jbe@2 162 #define UTF8PROC_CATEGORY_ME 8
jbe@2 163 #define UTF8PROC_CATEGORY_ND 9
jbe@2 164 #define UTF8PROC_CATEGORY_NL 10
jbe@2 165 #define UTF8PROC_CATEGORY_NO 11
jbe@2 166 #define UTF8PROC_CATEGORY_PC 12
jbe@2 167 #define UTF8PROC_CATEGORY_PD 13
jbe@2 168 #define UTF8PROC_CATEGORY_PS 14
jbe@2 169 #define UTF8PROC_CATEGORY_PE 15
jbe@2 170 #define UTF8PROC_CATEGORY_PI 16
jbe@2 171 #define UTF8PROC_CATEGORY_PF 17
jbe@2 172 #define UTF8PROC_CATEGORY_PO 18
jbe@2 173 #define UTF8PROC_CATEGORY_SM 19
jbe@2 174 #define UTF8PROC_CATEGORY_SC 20
jbe@2 175 #define UTF8PROC_CATEGORY_SK 21
jbe@2 176 #define UTF8PROC_CATEGORY_SO 22
jbe@2 177 #define UTF8PROC_CATEGORY_ZS 23
jbe@2 178 #define UTF8PROC_CATEGORY_ZL 24
jbe@2 179 #define UTF8PROC_CATEGORY_ZP 25
jbe@2 180 #define UTF8PROC_CATEGORY_CC 26
jbe@2 181 #define UTF8PROC_CATEGORY_CF 27
jbe@2 182 #define UTF8PROC_CATEGORY_CS 28
jbe@2 183 #define UTF8PROC_CATEGORY_CO 29
jbe@2 184 #define UTF8PROC_CATEGORY_CN 30
jbe@2 185 #define UTF8PROC_BIDI_CLASS_L 1
jbe@2 186 #define UTF8PROC_BIDI_CLASS_LRE 2
jbe@2 187 #define UTF8PROC_BIDI_CLASS_LRO 3
jbe@2 188 #define UTF8PROC_BIDI_CLASS_R 4
jbe@2 189 #define UTF8PROC_BIDI_CLASS_AL 5
jbe@2 190 #define UTF8PROC_BIDI_CLASS_RLE 6
jbe@2 191 #define UTF8PROC_BIDI_CLASS_RLO 7
jbe@2 192 #define UTF8PROC_BIDI_CLASS_PDF 8
jbe@2 193 #define UTF8PROC_BIDI_CLASS_EN 9
jbe@2 194 #define UTF8PROC_BIDI_CLASS_ES 10
jbe@2 195 #define UTF8PROC_BIDI_CLASS_ET 11
jbe@2 196 #define UTF8PROC_BIDI_CLASS_AN 12
jbe@2 197 #define UTF8PROC_BIDI_CLASS_CS 13
jbe@2 198 #define UTF8PROC_BIDI_CLASS_NSM 14
jbe@2 199 #define UTF8PROC_BIDI_CLASS_BN 15
jbe@2 200 #define UTF8PROC_BIDI_CLASS_B 16
jbe@2 201 #define UTF8PROC_BIDI_CLASS_S 17
jbe@2 202 #define UTF8PROC_BIDI_CLASS_WS 18
jbe@2 203 #define UTF8PROC_BIDI_CLASS_ON 19
jbe@2 204 #define UTF8PROC_DECOMP_TYPE_FONT 1
jbe@2 205 #define UTF8PROC_DECOMP_TYPE_NOBREAK 2
jbe@2 206 #define UTF8PROC_DECOMP_TYPE_INITIAL 3
jbe@2 207 #define UTF8PROC_DECOMP_TYPE_MEDIAL 4
jbe@2 208 #define UTF8PROC_DECOMP_TYPE_FINAL 5
jbe@2 209 #define UTF8PROC_DECOMP_TYPE_ISOLATED 6
jbe@2 210 #define UTF8PROC_DECOMP_TYPE_CIRCLE 7
jbe@2 211 #define UTF8PROC_DECOMP_TYPE_SUPER 8
jbe@2 212 #define UTF8PROC_DECOMP_TYPE_SUB 9
jbe@2 213 #define UTF8PROC_DECOMP_TYPE_VERTICAL 10
jbe@2 214 #define UTF8PROC_DECOMP_TYPE_WIDE 11
jbe@2 215 #define UTF8PROC_DECOMP_TYPE_NARROW 12
jbe@2 216 #define UTF8PROC_DECOMP_TYPE_SMALL 13
jbe@2 217 #define UTF8PROC_DECOMP_TYPE_SQUARE 14
jbe@2 218 #define UTF8PROC_DECOMP_TYPE_FRACTION 15
jbe@2 219 #define UTF8PROC_DECOMP_TYPE_COMPAT 16
jbe@2 220
jbe@0 221 extern const int8_t utf8proc_utf8class[256];
jbe@0 222
jbe@0 223 const char *utf8proc_errmsg(ssize_t errcode);
jbe@0 224 /*
jbe@0 225 * Returns a static error string for the given error code.
jbe@0 226 */
jbe@0 227
jbe@7 228 ssize_t utf8proc_iterate(const uint8_t *str, ssize_t strlen, int32_t *dst);
jbe@0 229 /*
jbe@0 230 * Reads a single char from the UTF-8 sequence being pointed to by 'str'.
jbe@0 231 * The maximum number of bytes read is 'strlen', unless 'strlen' is
jbe@0 232 * negative.
jbe@0 233 * If a valid unicode char could be read, it is stored in the variable
jbe@0 234 * being pointed to by 'dst', otherwise that variable will be set to -1.
jbe@0 235 * In case of success the number of bytes read is returned, otherwise a
jbe@0 236 * negative error code is returned.
jbe@0 237 */
jbe@0 238
jbe@7 239 bool utf8proc_codepoint_valid(int32_t uc);
jbe@7 240 /*
jbe@7 241 * Returns 1, if the given unicode code-point is valid, otherwise 0.
jbe@7 242 */
jbe@7 243
jbe@0 244 ssize_t utf8proc_encode_char(int32_t uc, uint8_t *dst);
jbe@0 245 /*
jbe@0 246 * Encodes the unicode char with the code point 'uc' as an UTF-8 string in
jbe@0 247 * the byte array being pointed to by 'dst'. This array has to be at least
jbe@0 248 * 4 bytes long.
jbe@7 249 * In case of success the number of bytes written is returned,
jbe@7 250 * otherwise 0.
jbe@0 251 * This function does not check if 'uc' is a valid unicode code point.
jbe@0 252 */
jbe@0 253
jbe@0 254 const utf8proc_property_t *utf8proc_get_property(int32_t uc);
jbe@0 255 /*
jbe@0 256 * Returns a pointer to a (constant) struct containing information about
jbe@0 257 * the unicode char with the given code point 'uc'.
jbe@0 258 * If the character is not existent a pointer to a special struct is
jbe@0 259 * returned, where 'category' is a NULL pointer.
jbe@0 260 * WARNING: The parameter 'uc' has to be in the range of 0x0000 to
jbe@0 261 * 0x10FFFF, otherwise the program might crash!
jbe@0 262 */
jbe@0 263
jbe@7 264 ssize_t utf8proc_decompose_char(
jbe@7 265 int32_t uc, int32_t *dst, ssize_t bufsize,
jbe@7 266 int options, int *last_boundclass
jbe@7 267 );
jbe@0 268 /*
jbe@0 269 * Writes a decomposition of the unicode char 'uc' into the array being
jbe@0 270 * pointed to by 'dst'.
jbe@0 271 * Following flags in the 'options' field are regarded:
jbe@2 272 * REJECTNA: an unassigned unicode code point leads to an error
jbe@2 273 * IGNORE: "default ignorable" chars are stripped
jbe@2 274 * CASEFOLD: unicode casefolding is applied
jbe@2 275 * COMPAT: replace certain characters with their
jbe@2 276 * compatibility decomposition
jbe@2 277 * CHARBOUND: Inserts 0xFF bytes before each grapheme cluster
jbe@3 278 * LUMP: lumps certain different characters together
jbe@3 279 * STRIPMARK: removes all character marks
jbe@7 280 * The pointer 'last_boundclass' has to point to an integer variable which
jbe@7 281 * is storing the last character boundary class, if the CHARBOUND option
jbe@7 282 * is used.
jbe@0 283 * In case of success the number of chars written is returned,
jbe@0 284 * in case of an error, a negative error code is returned.
jbe@0 285 * If the number of written chars would be bigger than 'bufsize',
jbe@0 286 * the buffer (up to 'bufsize') has inpredictable data, and the needed
jbe@0 287 * buffer size is returned.
jbe@0 288 * WARNING: The parameter 'uc' has to be in the range of 0x0000 to
jbe@0 289 * 0x10FFFF, otherwise the program might crash!
jbe@0 290 */
jbe@0 291
jbe@7 292 ssize_t utf8proc_decompose(
jbe@7 293 const uint8_t *str, ssize_t strlen,
jbe@7 294 int32_t *buffer, ssize_t bufsize, int options
jbe@7 295 );
jbe@0 296 /*
jbe@0 297 * Does the same as 'utf8proc_decompose_char', but acts on a whole UTF-8
jbe@0 298 * string, and orders the decomposed sequences correctly.
jbe@0 299 * If the NULLTERM flag in 'options' is set, processing will be stopped,
jbe@0 300 * when a NULL byte is encounted, otherwise 'strlen' bytes are processed.
jbe@0 301 * The result in form of unicode code points is written into the buffer
jbe@0 302 * being pointed to by 'buffer', having the length of 'bufsize' entries.
jbe@0 303 * In case of success the number of chars written is returned,
jbe@0 304 * in case of an error, a negative error code is returned.
jbe@0 305 * If the number of written chars would be bigger than 'bufsize',
jbe@0 306 * the buffer (up to 'bufsize') has inpredictable data, and the needed
jbe@0 307 * buffer size is returned.
jbe@0 308 */
jbe@0 309
jbe@0 310 ssize_t utf8proc_reencode(int32_t *buffer, ssize_t length, int options);
jbe@0 311 /*
jbe@0 312 * Reencodes the sequence of unicode characters given by the pointer
jbe@0 313 * 'buffer' and 'length' as UTF-8.
jbe@0 314 * The result is stored in the same memory area where the data is read.
jbe@0 315 * Following flags in the 'options' field are regarded:
jbe@0 316 * NLF2LS: converts LF, CRLF, CR and NEL into LS
jbe@0 317 * NLF2PS: converts LF, CRLF, CR and NEL into PS
jbe@0 318 * NLF2LF: converts LF, CRLF, CR and NEL into LF
jbe@0 319 * STRIPCC: strips or converts all non-affected control characters
jbe@7 320 * COMPOSE: tries to combine decomposed characters into composite
jbe@7 321 * characters
jbe@0 322 * STABLE: prohibits combining characters which would violate
jbe@0 323 * the unicode versioning stability
jbe@7 324 * In case of success the length of the resulting UTF-8 string is
jbe@7 325 * returned, otherwise a negative error code is returned.
jbe@0 326 * WARNING: The amount of free space being pointed to by 'buffer', has to
jbe@0 327 * exceed the amount of the input data by one byte, and the
jbe@0 328 * entries of the array pointed to by 'str' have to be in the
jbe@7 329 * range of 0x0000 to 0x10FFFF, otherwise the program might
jbe@7 330 * crash!
jbe@0 331 */
jbe@0 332
jbe@7 333 ssize_t utf8proc_map(
jbe@7 334 const uint8_t *str, ssize_t strlen, uint8_t **dstptr, int options
jbe@7 335 );
jbe@0 336 /*
jbe@0 337 * Maps the given UTF-8 string being pointed to by 'str' to a new UTF-8
jbe@7 338 * string, which is allocated dynamically, and afterwards pointed to by
jbe@7 339 * the pointer being pointed to by 'dstptr'.
jbe@0 340 * If the NULLTERM flag in the 'options' field is set, the length is
jbe@0 341 * determined by a NULL terminator, otherwise the parameter 'strlen' is
jbe@0 342 * evaluated to determine the string length, but in any case the result
jbe@7 343 * will be NULL terminated (though it might contain NULL characters
jbe@7 344 * before). Other flags in the 'options' field are passed to the functions
jbe@7 345 * defined above, and regarded as described.
jbe@0 346 * In case of success the length of the new string is returned,
jbe@0 347 * otherwise a negative error code is returned.
jbe@0 348 * NOTICE: The memory of the new UTF-8 string will have been allocated with
jbe@0 349 * 'malloc', and has theirfore to be freed with 'free'.
jbe@0 350 */
jbe@0 351
jbe@7 352 uint8_t *utf8proc_NFD(const uint8_t *str);
jbe@7 353 uint8_t *utf8proc_NFC(const uint8_t *str);
jbe@7 354 uint8_t *utf8proc_NFKD(const uint8_t *str);
jbe@7 355 uint8_t *utf8proc_NFKC(const uint8_t *str);
jbe@0 356 /*
jbe@0 357 * Returns a pointer to newly allocated memory of a NFD, NFC, NFKD or NFKC
jbe@0 358 * normalized version of the null-terminated string 'str'.
jbe@0 359 */
jbe@0 360
jbe@0 361
jbe@0 362 #endif
jbe@0 363

Impressum / About Us