utf8proc

annotate utf8proc.h @ 5:c18366878af9

find changesets by author, revision, files, or words in the commit message
Version 1.0.2

- included a check in Integer#utf8, which raises an exception, if the given code-point is invalid because of being too high (this was missing yet)
- added support for PostgreSQL version 8.2
author jbe
date Tue Dec 26 12:00:00 2006 +0100 (2006-12-26)
parents 4ee0d5f54af1
children fcfd8c836c64
rev   line source
jbe@0 1 /*
jbe@0 2 * Copyright (c) 2006, FlexiGuided GmbH, Berlin, Germany
jbe@0 3 * Author: Jan Behrens <jan.behrens@flexiguided.de>
jbe@0 4 * All rights reserved.
jbe@0 5 *
jbe@0 6 * Redistribution and use in source and binary forms, with or without
jbe@0 7 * modification, are permitted provided that the following conditions are
jbe@0 8 * met:
jbe@0 9 *
jbe@0 10 * 1. Redistributions of source code must retain the above copyright
jbe@0 11 * notice, this list of conditions and the following disclaimer.
jbe@0 12 * 2. Redistributions in binary form must reproduce the above copyright
jbe@0 13 * notice, this list of conditions and the following disclaimer in the
jbe@0 14 * documentation and/or other materials provided with the distribution.
jbe@0 15 * 3. Neither the name of the FlexiGuided GmbH nor the names of its
jbe@0 16 * contributors may be used to endorse or promote products derived from
jbe@0 17 * this software without specific prior written permission.
jbe@0 18 *
jbe@0 19 * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
jbe@0 20 * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
jbe@0 21 * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A
jbe@0 22 * PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER
jbe@0 23 * OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL,
jbe@0 24 * EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO,
jbe@0 25 * PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR
jbe@0 26 * PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF
jbe@0 27 * LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING
jbe@0 28 * NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
jbe@0 29 * SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
jbe@0 30 *
jbe@0 31 *
jbe@0 32 * This library contains derived data from a modified version of the
jbe@0 33 * Unicode data files.
jbe@0 34 *
jbe@0 35 * The original data files are available at
jbe@0 36 * http://www.unicode.org/Public/UNIDATA/
jbe@0 37 *
jbe@0 38 * Please notice the copyright statement in the file "utf8proc_data.c".
jbe@0 39 *
jbe@0 40 */
jbe@0 41
jbe@0 42
jbe@0 43 /*
jbe@0 44 * File name: utf8proc.h
jbe@3 45 * Version: 1.0
jbe@3 46 * Last changed: 2006-09-17
jbe@0 47 *
jbe@0 48 * Description:
jbe@0 49 * Header files for libutf8proc, which is a mapping tool for UTF-8 strings
jbe@0 50 * with following features:
jbe@0 51 * - decomposing and composing of strings
jbe@0 52 * - replacing compatibility characters with their equivalents
jbe@0 53 * - stripping of "default ignorable characters"
jbe@0 54 * like SOFT-HYPHEN or ZERO-WIDTH-SPACE
jbe@3 55 * - folding of certain characters for string comparison
jbe@3 56 * (e.g. HYPHEN U+2010 and MINUS U+2212 to ASCII "-")
jbe@3 57 * (see "LUMP" option)
jbe@0 58 * - optional rejection of strings containing non-assigned code points
jbe@0 59 * - stripping of control characters
jbe@3 60 * - stripping of character marks (accents, etc.)
jbe@0 61 * - transformation of LF, CRLF, CR and NEL to line-feed (LF)
jbe@0 62 * or to the unicode chararacters for paragraph separation (PS)
jbe@0 63 * or line separation (LS).
jbe@0 64 * - unicode case folding (for case insensitive string comparisons)
jbe@0 65 * - rejection of illegal UTF-8 data (i.e. UTF-8 encoded UTF-16 surrogates)
jbe@0 66 * - support for korean hangul characters
jbe@2 67 * Unicode Version 5.0.0 is supported.
jbe@0 68 */
jbe@0 69
jbe@0 70
jbe@0 71 #ifndef UTF8PROC_H
jbe@0 72 #define UTF8PROC_H
jbe@0 73
jbe@0 74
jbe@0 75 #include <stdlib.h>
jbe@0 76 #include <stdbool.h>
jbe@0 77 #include <sys/types.h>
jbe@0 78 #include <inttypes.h>
jbe@0 79 #include <limits.h>
jbe@0 80
jbe@0 81 #ifndef SSIZE_MAX
jbe@0 82 #define SSIZE_MAX (SIZE_MAX/2)
jbe@0 83 #endif
jbe@0 84
jbe@2 85 #define UTF8PROC_NULLTERM (1<<0)
jbe@2 86 #define UTF8PROC_STABLE (1<<1)
jbe@2 87 #define UTF8PROC_COMPAT (1<<2)
jbe@2 88 #define UTF8PROC_COMPOSE (1<<3)
jbe@2 89 #define UTF8PROC_DECOMPOSE (1<<4)
jbe@2 90 #define UTF8PROC_IGNORE (1<<5)
jbe@2 91 #define UTF8PROC_REJECTNA (1<<6)
jbe@2 92 #define UTF8PROC_NLF2LS (1<<7)
jbe@2 93 #define UTF8PROC_NLF2PS (1<<8)
jbe@2 94 #define UTF8PROC_NLF2LF (UTF8PROC_NLF2LS | UTF8PROC_NLF2PS)
jbe@2 95 #define UTF8PROC_STRIPCC (1<<9)
jbe@2 96 #define UTF8PROC_CASEFOLD (1<<10)
jbe@2 97 #define UTF8PROC_CHARBOUND (1<<11)
jbe@3 98 #define UTF8PROC_LUMP (1<<12)
jbe@3 99 #define UTF8PROC_STRIPMARK (1<<13)
jbe@0 100 /*
jbe@0 101 * Flags being regarded by several functions in the library:
jbe@2 102 * NULLTERM: The given UTF-8 input is NULL terminated.
jbe@2 103 * STABLE: Unicode Versioning Stability has to be respected.
jbe@2 104 * COMPAT: Compatiblity decomposition
jbe@2 105 * (i.e. formatting information is lost)
jbe@2 106 * COMPOSE: Return a result with composed characters.
jbe@2 107 * DECOMPOSE: Return a result with decomposed characters.
jbe@2 108 * IGNORE: Strip "default ignorable characters"
jbe@2 109 * REJECTNA: Return an error, if the input contains unassigned code points.
jbe@2 110 * NLF2LS: Indicating that NLF-sequences (LF, CRLF, CR, NEL) are
jbe@2 111 * representing a line break, and should be converted to the
jbe@2 112 * unicode character for line separation (LS).
jbe@2 113 * NLF2PS: Indicating that NLF-sequences are representing a paragraph
jbe@2 114 * break, and should be converted to the unicode character for
jbe@2 115 * paragraph separation (PS).
jbe@2 116 * NLF2LF: Indicating that the meaning of NLF-sequences is unknown.
jbe@2 117 * STRIPCC: Strips and/or convers control characters.
jbe@2 118 * NLF-sequences are transformed into space, except if one of the
jbe@2 119 * NLF2LS/PS/LF options is given.
jbe@2 120 * HorizontalTab (HT) and FormFeed (FF) are treated as a
jbe@2 121 * NLF-sequence in this case.
jbe@2 122 * All other control characters are simply removed.
jbe@2 123 * CASEFOLD: Performs unicode case folding, to be able to do a
jbe@2 124 * case-insensitive string comparison.
jbe@2 125 * CHARBOUND: Inserts 0xFF bytes at the beginning of each sequence which is
jbe@2 126 * representing a single grapheme cluster (a single character).
jbe@3 127 * LUMP: Lumps certain characters together
jbe@3 128 * (e.g. HYPHEN U+2010 and MINUS U+2212 to ASCII "-").
jbe@3 129 * (See lump.txt for details.)
jbe@3 130 * If NLF2LF is set, this includes a transformation of paragraph
jbe@3 131 * and line separators to ASCII line-feed (LF).
jbe@3 132 * STRIPMARK: Strips all character markings
jbe@3 133 * (non-spacing, spacing and enclosing) (i.e. accents)
jbe@3 134 * NOTE: this option works only with COMPOSE or DECOMPOSE
jbe@0 135 */
jbe@0 136
jbe@0 137 #define UTF8PROC_ERROR_NOMEM -1
jbe@0 138 #define UTF8PROC_ERROR_OVERFLOW -2
jbe@0 139 #define UTF8PROC_ERROR_INVALIDUTF8 -3
jbe@0 140 #define UTF8PROC_ERROR_NOTASSIGNED -4
jbe@3 141 #define UTF8PROC_ERROR_INVALIDOPTS -5
jbe@0 142 /*
jbe@0 143 * Error codes being returned by almost all functions:
jbe@0 144 * ERROR_NOMEM: Memory could not be allocated.
jbe@0 145 * ERROR_OVERFLOW: The given string is too long to be processed.
jbe@0 146 * ERROR_INVALIDUTF8: The given string is not a legal UTF-8 string.
jbe@0 147 * ERROR_NOTASSIGNED: The REJECTNA flag was set,
jbe@0 148 * and an unassigned code point was found.
jbe@3 149 * ERROR_INVALIDOPTS: Invalid options have been used.
jbe@0 150 */
jbe@0 151
jbe@3 152 typedef int16_t utf8proc_propval_t;
jbe@0 153 typedef struct utf8proc_property_struct {
jbe@3 154 utf8proc_propval_t category;
jbe@3 155 utf8proc_propval_t combining_class;
jbe@3 156 utf8proc_propval_t bidi_class;
jbe@3 157 utf8proc_propval_t decomp_type;
jbe@0 158 const int32_t *decomp_mapping;
jbe@0 159 const unsigned bidi_mirrored:1;
jbe@0 160 const int32_t uppercase_mapping;
jbe@0 161 const int32_t lowercase_mapping;
jbe@0 162 const int32_t titlecase_mapping;
jbe@0 163 const int32_t comb1st_index;
jbe@0 164 const int32_t comb2nd_index;
jbe@0 165 const unsigned comp_exclusion:1;
jbe@0 166 const unsigned ignorable:1;
jbe@2 167 const unsigned control_boundary:1;
jbe@2 168 const unsigned extend:1;
jbe@0 169 const int32_t *casefold_mapping;
jbe@0 170 } utf8proc_property_t;
jbe@0 171
jbe@2 172 #define UTF8PROC_CATEGORY_LU 1
jbe@2 173 #define UTF8PROC_CATEGORY_LL 2
jbe@2 174 #define UTF8PROC_CATEGORY_LT 3
jbe@2 175 #define UTF8PROC_CATEGORY_LM 4
jbe@2 176 #define UTF8PROC_CATEGORY_LO 5
jbe@2 177 #define UTF8PROC_CATEGORY_MN 6
jbe@2 178 #define UTF8PROC_CATEGORY_MC 7
jbe@2 179 #define UTF8PROC_CATEGORY_ME 8
jbe@2 180 #define UTF8PROC_CATEGORY_ND 9
jbe@2 181 #define UTF8PROC_CATEGORY_NL 10
jbe@2 182 #define UTF8PROC_CATEGORY_NO 11
jbe@2 183 #define UTF8PROC_CATEGORY_PC 12
jbe@2 184 #define UTF8PROC_CATEGORY_PD 13
jbe@2 185 #define UTF8PROC_CATEGORY_PS 14
jbe@2 186 #define UTF8PROC_CATEGORY_PE 15
jbe@2 187 #define UTF8PROC_CATEGORY_PI 16
jbe@2 188 #define UTF8PROC_CATEGORY_PF 17
jbe@2 189 #define UTF8PROC_CATEGORY_PO 18
jbe@2 190 #define UTF8PROC_CATEGORY_SM 19
jbe@2 191 #define UTF8PROC_CATEGORY_SC 20
jbe@2 192 #define UTF8PROC_CATEGORY_SK 21
jbe@2 193 #define UTF8PROC_CATEGORY_SO 22
jbe@2 194 #define UTF8PROC_CATEGORY_ZS 23
jbe@2 195 #define UTF8PROC_CATEGORY_ZL 24
jbe@2 196 #define UTF8PROC_CATEGORY_ZP 25
jbe@2 197 #define UTF8PROC_CATEGORY_CC 26
jbe@2 198 #define UTF8PROC_CATEGORY_CF 27
jbe@2 199 #define UTF8PROC_CATEGORY_CS 28
jbe@2 200 #define UTF8PROC_CATEGORY_CO 29
jbe@2 201 #define UTF8PROC_CATEGORY_CN 30
jbe@2 202 #define UTF8PROC_BIDI_CLASS_L 1
jbe@2 203 #define UTF8PROC_BIDI_CLASS_LRE 2
jbe@2 204 #define UTF8PROC_BIDI_CLASS_LRO 3
jbe@2 205 #define UTF8PROC_BIDI_CLASS_R 4
jbe@2 206 #define UTF8PROC_BIDI_CLASS_AL 5
jbe@2 207 #define UTF8PROC_BIDI_CLASS_RLE 6
jbe@2 208 #define UTF8PROC_BIDI_CLASS_RLO 7
jbe@2 209 #define UTF8PROC_BIDI_CLASS_PDF 8
jbe@2 210 #define UTF8PROC_BIDI_CLASS_EN 9
jbe@2 211 #define UTF8PROC_BIDI_CLASS_ES 10
jbe@2 212 #define UTF8PROC_BIDI_CLASS_ET 11
jbe@2 213 #define UTF8PROC_BIDI_CLASS_AN 12
jbe@2 214 #define UTF8PROC_BIDI_CLASS_CS 13
jbe@2 215 #define UTF8PROC_BIDI_CLASS_NSM 14
jbe@2 216 #define UTF8PROC_BIDI_CLASS_BN 15
jbe@2 217 #define UTF8PROC_BIDI_CLASS_B 16
jbe@2 218 #define UTF8PROC_BIDI_CLASS_S 17
jbe@2 219 #define UTF8PROC_BIDI_CLASS_WS 18
jbe@2 220 #define UTF8PROC_BIDI_CLASS_ON 19
jbe@2 221 #define UTF8PROC_DECOMP_TYPE_FONT 1
jbe@2 222 #define UTF8PROC_DECOMP_TYPE_NOBREAK 2
jbe@2 223 #define UTF8PROC_DECOMP_TYPE_INITIAL 3
jbe@2 224 #define UTF8PROC_DECOMP_TYPE_MEDIAL 4
jbe@2 225 #define UTF8PROC_DECOMP_TYPE_FINAL 5
jbe@2 226 #define UTF8PROC_DECOMP_TYPE_ISOLATED 6
jbe@2 227 #define UTF8PROC_DECOMP_TYPE_CIRCLE 7
jbe@2 228 #define UTF8PROC_DECOMP_TYPE_SUPER 8
jbe@2 229 #define UTF8PROC_DECOMP_TYPE_SUB 9
jbe@2 230 #define UTF8PROC_DECOMP_TYPE_VERTICAL 10
jbe@2 231 #define UTF8PROC_DECOMP_TYPE_WIDE 11
jbe@2 232 #define UTF8PROC_DECOMP_TYPE_NARROW 12
jbe@2 233 #define UTF8PROC_DECOMP_TYPE_SMALL 13
jbe@2 234 #define UTF8PROC_DECOMP_TYPE_SQUARE 14
jbe@2 235 #define UTF8PROC_DECOMP_TYPE_FRACTION 15
jbe@2 236 #define UTF8PROC_DECOMP_TYPE_COMPAT 16
jbe@2 237
jbe@0 238 extern const int8_t utf8proc_utf8class[256];
jbe@0 239
jbe@0 240 const char *utf8proc_errmsg(ssize_t errcode);
jbe@0 241 /*
jbe@0 242 * Returns a static error string for the given error code.
jbe@0 243 */
jbe@0 244
jbe@0 245 ssize_t utf8proc_iterate(uint8_t *str, ssize_t strlen, int32_t *dst);
jbe@0 246 /*
jbe@0 247 * Reads a single char from the UTF-8 sequence being pointed to by 'str'.
jbe@0 248 * The maximum number of bytes read is 'strlen', unless 'strlen' is
jbe@0 249 * negative.
jbe@0 250 * If a valid unicode char could be read, it is stored in the variable
jbe@0 251 * being pointed to by 'dst', otherwise that variable will be set to -1.
jbe@0 252 * In case of success the number of bytes read is returned, otherwise a
jbe@0 253 * negative error code is returned.
jbe@0 254 */
jbe@0 255
jbe@0 256 ssize_t utf8proc_encode_char(int32_t uc, uint8_t *dst);
jbe@0 257 /*
jbe@0 258 * Encodes the unicode char with the code point 'uc' as an UTF-8 string in
jbe@0 259 * the byte array being pointed to by 'dst'. This array has to be at least
jbe@0 260 * 4 bytes long.
jbe@0 261 * In case of success the number of bytes written is returned, otherwise 0.
jbe@0 262 * This function does not check if 'uc' is a valid unicode code point.
jbe@0 263 */
jbe@0 264
jbe@0 265 const utf8proc_property_t *utf8proc_get_property(int32_t uc);
jbe@0 266 /*
jbe@0 267 * Returns a pointer to a (constant) struct containing information about
jbe@0 268 * the unicode char with the given code point 'uc'.
jbe@0 269 * If the character is not existent a pointer to a special struct is
jbe@0 270 * returned, where 'category' is a NULL pointer.
jbe@0 271 * WARNING: The parameter 'uc' has to be in the range of 0x0000 to
jbe@0 272 * 0x10FFFF, otherwise the program might crash!
jbe@0 273 */
jbe@0 274
jbe@0 275 ssize_t utf8proc_decompose_char(int32_t uc, int32_t *dst, ssize_t bufsize,
jbe@2 276 int options, int *last_boundclass);
jbe@0 277 /*
jbe@0 278 * Writes a decomposition of the unicode char 'uc' into the array being
jbe@0 279 * pointed to by 'dst'.
jbe@0 280 * Following flags in the 'options' field are regarded:
jbe@2 281 * REJECTNA: an unassigned unicode code point leads to an error
jbe@2 282 * IGNORE: "default ignorable" chars are stripped
jbe@2 283 * CASEFOLD: unicode casefolding is applied
jbe@2 284 * COMPAT: replace certain characters with their
jbe@2 285 * compatibility decomposition
jbe@2 286 * CHARBOUND: Inserts 0xFF bytes before each grapheme cluster
jbe@3 287 * LUMP: lumps certain different characters together
jbe@3 288 * STRIPMARK: removes all character marks
jbe@2 289 * The pointer 'last_boundclass' has to point to an integer variable which is
jbe@2 290 * storing the last character boundary class, if the CHARBOUND option is
jbe@2 291 * used.
jbe@0 292 * In case of success the number of chars written is returned,
jbe@0 293 * in case of an error, a negative error code is returned.
jbe@0 294 * If the number of written chars would be bigger than 'bufsize',
jbe@0 295 * the buffer (up to 'bufsize') has inpredictable data, and the needed
jbe@0 296 * buffer size is returned.
jbe@0 297 * WARNING: The parameter 'uc' has to be in the range of 0x0000 to
jbe@0 298 * 0x10FFFF, otherwise the program might crash!
jbe@0 299 */
jbe@0 300
jbe@0 301 ssize_t utf8proc_decompose(uint8_t *str, ssize_t strlen,
jbe@0 302 int32_t *buffer, ssize_t bufsize, int options);
jbe@0 303 /*
jbe@0 304 * Does the same as 'utf8proc_decompose_char', but acts on a whole UTF-8
jbe@0 305 * string, and orders the decomposed sequences correctly.
jbe@0 306 * If the NULLTERM flag in 'options' is set, processing will be stopped,
jbe@0 307 * when a NULL byte is encounted, otherwise 'strlen' bytes are processed.
jbe@0 308 * The result in form of unicode code points is written into the buffer
jbe@0 309 * being pointed to by 'buffer', having the length of 'bufsize' entries.
jbe@0 310 * In case of success the number of chars written is returned,
jbe@0 311 * in case of an error, a negative error code is returned.
jbe@0 312 * If the number of written chars would be bigger than 'bufsize',
jbe@0 313 * the buffer (up to 'bufsize') has inpredictable data, and the needed
jbe@0 314 * buffer size is returned.
jbe@0 315 */
jbe@0 316
jbe@0 317 ssize_t utf8proc_reencode(int32_t *buffer, ssize_t length, int options);
jbe@0 318 /*
jbe@0 319 * Reencodes the sequence of unicode characters given by the pointer
jbe@0 320 * 'buffer' and 'length' as UTF-8.
jbe@0 321 * The result is stored in the same memory area where the data is read.
jbe@0 322 * Following flags in the 'options' field are regarded:
jbe@0 323 * NLF2LS: converts LF, CRLF, CR and NEL into LS
jbe@0 324 * NLF2PS: converts LF, CRLF, CR and NEL into PS
jbe@0 325 * NLF2LF: converts LF, CRLF, CR and NEL into LF
jbe@0 326 * STRIPCC: strips or converts all non-affected control characters
jbe@0 327 * COMPOSE: tries to combine decomposed characters into composite characters
jbe@0 328 * STABLE: prohibits combining characters which would violate
jbe@0 329 * the unicode versioning stability
jbe@0 330 * In case of success the length of the resulting UTF-8 string is returned,
jbe@0 331 * otherwise a negative error code is returned.
jbe@0 332 * WARNING: The amount of free space being pointed to by 'buffer', has to
jbe@0 333 * exceed the amount of the input data by one byte, and the
jbe@0 334 * entries of the array pointed to by 'str' have to be in the
jbe@0 335 * range of 0x0000 to 0x10FFFF, otherwise the program might crash!
jbe@0 336 */
jbe@0 337
jbe@0 338 ssize_t utf8proc_map(uint8_t *str, ssize_t strlen, uint8_t **dstptr,
jbe@0 339 int options);
jbe@0 340 /*
jbe@0 341 * Maps the given UTF-8 string being pointed to by 'str' to a new UTF-8
jbe@0 342 * string, which is allocated dynamically, and afterwards pointed to by the
jbe@0 343 * pointer being pointed to by 'dstptr'.
jbe@0 344 * If the NULLTERM flag in the 'options' field is set, the length is
jbe@0 345 * determined by a NULL terminator, otherwise the parameter 'strlen' is
jbe@0 346 * evaluated to determine the string length, but in any case the result
jbe@0 347 * will be NULL terminated (though it might contain NULL characters before).
jbe@0 348 * Other flags in the 'options' field are passed to the functions defined
jbe@0 349 * above, and regarded as described.
jbe@0 350 * In case of success the length of the new string is returned,
jbe@0 351 * otherwise a negative error code is returned.
jbe@0 352 * NOTICE: The memory of the new UTF-8 string will have been allocated with
jbe@0 353 * 'malloc', and has theirfore to be freed with 'free'.
jbe@0 354 */
jbe@0 355
jbe@0 356 uint8_t *utf8proc_NFD(uint8_t *str);
jbe@0 357 uint8_t *utf8proc_NFC(uint8_t *str);
jbe@0 358 uint8_t *utf8proc_NFKD(uint8_t *str);
jbe@0 359 uint8_t *utf8proc_NFKC(uint8_t *str);
jbe@0 360 /*
jbe@0 361 * Returns a pointer to newly allocated memory of a NFD, NFC, NFKD or NFKC
jbe@0 362 * normalized version of the null-terminated string 'str'.
jbe@0 363 */
jbe@0 364
jbe@0 365
jbe@0 366 #endif
jbe@0 367
jbe@0 368

Impressum / About Us