1 /* 2 * Copyright (c) 2015 Steven G. Johnson, Jiahao Chen, Peter Colberg, Tony Kelman, Scott P. Jones, and other contributors. 3 * Copyright (c) 2009 Public Software Group e. V., Berlin, Germany 4 * 5 * Permission is hereby granted, free of charge, to any person obtaining a 6 * copy of this software and associated documentation files (the "Software"), 7 * to deal in the Software without restriction, including without limitation 8 * the rights to use, copy, modify, merge, publish, distribute, sublicense, 9 * and/or sell copies of the Software, and to permit persons to whom the 10 * Software is furnished to do so, subject to the following conditions: 11 * 12 * The above copyright notice and this permission notice shall be included in 13 * all copies or substantial portions of the Software. 14 * 15 * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR 16 * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, 17 * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE 18 * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER 19 * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING 20 * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER 21 * DEALINGS IN THE SOFTWARE. 22 */ 23 24 25 /** 26 * @mainpage 27 * 28 * utf8proc is a free/open-source (MIT/expat licensed) C library 29 * providing Unicode normalization, case-folding, and other operations 30 * for strings in the UTF-8 encoding, supporting Unicode version 31 * 8.0.0. See the utf8proc home page (http://julialang.org/utf8proc/) 32 * for downloads and other information, or the source code on github 33 * (https://github.com/JuliaLang/utf8proc). 34 * 35 * For the utf8proc API documentation, see: @ref utf8proc.h 36 * 37 * The features of utf8proc include: 38 * 39 * - Transformation of strings (@ref utf8proc_map) to: 40 * - decompose (@ref UTF8PROC_DECOMPOSE) or compose (@ref UTF8PROC_COMPOSE) Unicode combining characters (http://en.wikipedia.org/wiki/Combining_character) 41 * - canonicalize Unicode compatibility characters (@ref UTF8PROC_COMPAT) 42 * - strip "ignorable" (@ref UTF8PROC_IGNORE) characters, control characters (@ref UTF8PROC_STRIPCC), or combining characters such as accents (@ref UTF8PROC_STRIPMARK) 43 * - case-folding (@ref UTF8PROC_CASEFOLD) 44 * - Unicode normalization: @ref utf8proc_NFD, @ref utf8proc_NFC, @ref utf8proc_NFKD, @ref utf8proc_NFKC 45 * - Detecting grapheme boundaries (@ref utf8proc_grapheme_break and @ref UTF8PROC_CHARBOUND) 46 * - Character-width computation: @ref utf8proc_charwidth 47 * - Classification of characters by Unicode category: @ref utf8proc_category and @ref utf8proc_category_string 48 * - Encode (@ref utf8proc_encode_char) and decode (@ref utf8proc_iterate) Unicode codepoints to/from UTF-8. 49 */ 50 51 /** @file */ 52 53 #ifndef UTF8PROC_H 54 #define UTF8PROC_H 55 56 /** @name API version 57 * 58 * The utf8proc API version MAJOR.MINOR.PATCH, following 59 * semantic-versioning rules (http://semver.org) based on API 60 * compatibility. 61 * 62 * This is also returned at runtime by @ref utf8proc_version; however, the 63 * runtime version may append a string like "-dev" to the version number 64 * for prerelease versions. 65 * 66 * @note The shared-library version number in the Makefile 67 * (and CMakeLists.txt, and MANIFEST) may be different, 68 * being based on ABI compatibility rather than API compatibility. 69 */ 70 /** @{ */ 71 /** The MAJOR version number (increased when backwards API compatibility is broken). */ 72 #define UTF8PROC_VERSION_MAJOR 2 73 /** The MINOR version number (increased when new functionality is added in a backwards-compatible manner). */ 74 #define UTF8PROC_VERSION_MINOR 1 75 /** The PATCH version (increased for fixes that do not change the API). */ 76 #define UTF8PROC_VERSION_PATCH 0 77 /** @} */ 78 79 #include <stdlib.h> 80 #include <sys/types.h> 81 #include <apr.h> 82 typedef char utf8proc_int8_t; 83 typedef unsigned char utf8proc_uint8_t; 84 typedef apr_int16_t utf8proc_int16_t; 85 typedef apr_uint16_t utf8proc_uint16_t; 86 typedef apr_int32_t utf8proc_int32_t; 87 typedef apr_uint32_t utf8proc_uint32_t; 88 #ifdef _MSC_VER 89 # ifdef _WIN64 90 typedef apr_int64_t utf8proc_ssize_t; 91 typedef apr_uint64_t utf8proc_size_t; 92 # else 93 typedef apr_int32_t utf8proc_ssize_t; 94 typedef apr_uint32_t utf8proc_size_t; 95 # endif 96 #else 97 typedef apr_size_t utf8proc_size_t; 98 typedef apr_ssize_t utf8proc_ssize_t; 99 #endif 100 #ifdef __cplusplus 101 typedef bool utf8proc_bool; 102 #elif defined HAVE_STDBOOL_H 103 # include <stdbool.h> 104 typedef bool utf8proc_bool; 105 #else 106 /* emulate C99 bool */ 107 # ifndef __bool_true_false_are_defined 108 # define false 0 109 # define true 1 110 # define __bool_true_false_are_defined 1 111 # endif 112 typedef apr_byte_t utf8proc_bool; 113 #endif 114 #include <limits.h> 115 116 117 /* 118 * Define UTF8PROC_INLINE and include utf8proc.c to embed a static 119 * version of utf8proc in your program or library without exporting 120 * any of its symbols. 121 */ 122 #ifdef UTF8PROC_INLINE 123 # define UTF8PROC_DATA static 124 # undef UTF8PROC_DATA_EXPORT 125 # define UTF8PROC_DLLEXPORT static 126 #else 127 # define UTF8PROC_DATA 128 # define UTF8PROC_DATA_EXPORT 129 # ifdef _WIN32 130 # ifdef UTF8PROC_EXPORTS 131 # define UTF8PROC_DLLEXPORT __declspec(dllexport) 132 # else 133 # define UTF8PROC_DLLEXPORT __declspec(dllimport) 134 # endif 135 # elif __GNUC__ >= 4 136 # define UTF8PROC_DLLEXPORT __attribute__ ((visibility("default"))) 137 # else 138 # define UTF8PROC_DLLEXPORT 139 # endif 140 #endif 141 142 #ifdef __cplusplus 143 extern "C" { 144 #endif 145 146 #ifndef SSIZE_MAX 147 #define SSIZE_MAX ((size_t)SIZE_MAX/2) 148 #endif 149 150 #ifndef UINT16_MAX 151 # define UINT16_MAX 65535U 152 #endif 153 154 /** 155 * Option flags used by several functions in the library. 156 */ 157 typedef enum { 158 /** The given UTF-8 input is NULL terminated. */ 159 UTF8PROC_NULLTERM = (1<<0), 160 /** Unicode Versioning Stability has to be respected. */ 161 UTF8PROC_STABLE = (1<<1), 162 /** Compatibility decomposition (i.e. formatting information is lost). */ 163 UTF8PROC_COMPAT = (1<<2), 164 /** Return a result with decomposed characters. */ 165 UTF8PROC_COMPOSE = (1<<3), 166 /** Return a result with decomposed characters. */ 167 UTF8PROC_DECOMPOSE = (1<<4), 168 /** Strip "default ignorable characters" such as SOFT-HYPHEN or ZERO-WIDTH-SPACE. */ 169 UTF8PROC_IGNORE = (1<<5), 170 /** Return an error, if the input contains unassigned codepoints. */ 171 UTF8PROC_REJECTNA = (1<<6), 172 /** 173 * Indicating that NLF-sequences (LF, CRLF, CR, NEL) are representing a 174 * line break, and should be converted to the codepoint for line 175 * separation (LS). 176 */ 177 UTF8PROC_NLF2LS = (1<<7), 178 /** 179 * Indicating that NLF-sequences are representing a paragraph break, and 180 * should be converted to the codepoint for paragraph separation 181 * (PS). 182 */ 183 UTF8PROC_NLF2PS = (1<<8), 184 /** Indicating that the meaning of NLF-sequences is unknown. */ 185 UTF8PROC_NLF2LF = (UTF8PROC_NLF2LS | UTF8PROC_NLF2PS), 186 /** Strips and/or convers control characters. 187 * 188 * NLF-sequences are transformed into space, except if one of the 189 * NLF2LS/PS/LF options is given. HorizontalTab (HT) and FormFeed (FF) 190 * are treated as a NLF-sequence in this case. All other control 191 * characters are simply removed. 192 */ 193 UTF8PROC_STRIPCC = (1<<9), 194 /** 195 * Performs unicode case folding, to be able to do a case-insensitive 196 * string comparison. 197 */ 198 UTF8PROC_CASEFOLD = (1<<10), 199 /** 200 * Inserts 0xFF bytes at the beginning of each sequence which is 201 * representing a single grapheme cluster (see UAX#29). 202 */ 203 UTF8PROC_CHARBOUND = (1<<11), 204 /** Lumps certain characters together. 205 * 206 * E.g. HYPHEN U+2010 and MINUS U+2212 to ASCII "-". See lump.md for details. 207 * 208 * If NLF2LF is set, this includes a transformation of paragraph and 209 * line separators to ASCII line-feed (LF). 210 */ 211 UTF8PROC_LUMP = (1<<12), 212 /** Strips all character markings. 213 * 214 * This includes non-spacing, spacing and enclosing (i.e. accents). 215 * @note This option works only with @ref UTF8PROC_COMPOSE or 216 * @ref UTF8PROC_DECOMPOSE 217 */ 218 UTF8PROC_STRIPMARK = (1<<13), 219 } utf8proc_option_t; 220 221 /** @name Error codes 222 * Error codes being returned by almost all functions. 223 */ 224 /** @{ */ 225 /** Memory could not be allocated. */ 226 #define UTF8PROC_ERROR_NOMEM -1 227 /** The given string is too long to be processed. */ 228 #define UTF8PROC_ERROR_OVERFLOW -2 229 /** The given string is not a legal UTF-8 string. */ 230 #define UTF8PROC_ERROR_INVALIDUTF8 -3 231 /** The @ref UTF8PROC_REJECTNA flag was set and an unassigned codepoint was found. */ 232 #define UTF8PROC_ERROR_NOTASSIGNED -4 233 /** Invalid options have been used. */ 234 #define UTF8PROC_ERROR_INVALIDOPTS -5 235 /** @} */ 236 237 /* @name Types */ 238 239 /** Holds the value of a property. */ 240 typedef utf8proc_int16_t utf8proc_propval_t; 241 242 /** Struct containing information about a codepoint. */ 243 typedef struct utf8proc_property_struct { 244 /** 245 * Unicode category. 246 * @see utf8proc_category_t. 247 */ 248 utf8proc_propval_t category; 249 utf8proc_propval_t combining_class; 250 /** 251 * Bidirectional class. 252 * @see utf8proc_bidi_class_t. 253 */ 254 utf8proc_propval_t bidi_class; 255 /** 256 * @anchor Decomposition type. 257 * @see utf8proc_decomp_type_t. 258 */ 259 utf8proc_propval_t decomp_type; 260 utf8proc_uint16_t decomp_seqindex; 261 utf8proc_uint16_t casefold_seqindex; 262 utf8proc_uint16_t uppercase_seqindex; 263 utf8proc_uint16_t lowercase_seqindex; 264 utf8proc_uint16_t titlecase_seqindex; 265 utf8proc_uint16_t comb_index; 266 unsigned bidi_mirrored:1; 267 unsigned comp_exclusion:1; 268 /** 269 * Can this codepoint be ignored? 270 * 271 * Used by @ref utf8proc_decompose_char when @ref UTF8PROC_IGNORE is 272 * passed as an option. 273 */ 274 unsigned ignorable:1; 275 unsigned control_boundary:1; 276 /** The width of the codepoint. */ 277 unsigned charwidth:2; 278 unsigned pad:2; 279 /** 280 * Boundclass. 281 * @see utf8proc_boundclass_t. 282 */ 283 unsigned boundclass:8; 284 } utf8proc_property_t; 285 286 /** Unicode categories. */ 287 typedef enum { 288 UTF8PROC_CATEGORY_CN = 0, /**< Other, not assigned */ 289 UTF8PROC_CATEGORY_LU = 1, /**< Letter, uppercase */ 290 UTF8PROC_CATEGORY_LL = 2, /**< Letter, lowercase */ 291 UTF8PROC_CATEGORY_LT = 3, /**< Letter, titlecase */ 292 UTF8PROC_CATEGORY_LM = 4, /**< Letter, modifier */ 293 UTF8PROC_CATEGORY_LO = 5, /**< Letter, other */ 294 UTF8PROC_CATEGORY_MN = 6, /**< Mark, nonspacing */ 295 UTF8PROC_CATEGORY_MC = 7, /**< Mark, spacing combining */ 296 UTF8PROC_CATEGORY_ME = 8, /**< Mark, enclosing */ 297 UTF8PROC_CATEGORY_ND = 9, /**< Number, decimal digit */ 298 UTF8PROC_CATEGORY_NL = 10, /**< Number, letter */ 299 UTF8PROC_CATEGORY_NO = 11, /**< Number, other */ 300 UTF8PROC_CATEGORY_PC = 12, /**< Punctuation, connector */ 301 UTF8PROC_CATEGORY_PD = 13, /**< Punctuation, dash */ 302 UTF8PROC_CATEGORY_PS = 14, /**< Punctuation, open */ 303 UTF8PROC_CATEGORY_PE = 15, /**< Punctuation, close */ 304 UTF8PROC_CATEGORY_PI = 16, /**< Punctuation, initial quote */ 305 UTF8PROC_CATEGORY_PF = 17, /**< Punctuation, final quote */ 306 UTF8PROC_CATEGORY_PO = 18, /**< Punctuation, other */ 307 UTF8PROC_CATEGORY_SM = 19, /**< Symbol, math */ 308 UTF8PROC_CATEGORY_SC = 20, /**< Symbol, currency */ 309 UTF8PROC_CATEGORY_SK = 21, /**< Symbol, modifier */ 310 UTF8PROC_CATEGORY_SO = 22, /**< Symbol, other */ 311 UTF8PROC_CATEGORY_ZS = 23, /**< Separator, space */ 312 UTF8PROC_CATEGORY_ZL = 24, /**< Separator, line */ 313 UTF8PROC_CATEGORY_ZP = 25, /**< Separator, paragraph */ 314 UTF8PROC_CATEGORY_CC = 26, /**< Other, control */ 315 UTF8PROC_CATEGORY_CF = 27, /**< Other, format */ 316 UTF8PROC_CATEGORY_CS = 28, /**< Other, surrogate */ 317 UTF8PROC_CATEGORY_CO = 29, /**< Other, private use */ 318 } utf8proc_category_t; 319 320 /** Bidirectional character classes. */ 321 typedef enum { 322 UTF8PROC_BIDI_CLASS_L = 1, /**< Left-to-Right */ 323 UTF8PROC_BIDI_CLASS_LRE = 2, /**< Left-to-Right Embedding */ 324 UTF8PROC_BIDI_CLASS_LRO = 3, /**< Left-to-Right Override */ 325 UTF8PROC_BIDI_CLASS_R = 4, /**< Right-to-Left */ 326 UTF8PROC_BIDI_CLASS_AL = 5, /**< Right-to-Left Arabic */ 327 UTF8PROC_BIDI_CLASS_RLE = 6, /**< Right-to-Left Embedding */ 328 UTF8PROC_BIDI_CLASS_RLO = 7, /**< Right-to-Left Override */ 329 UTF8PROC_BIDI_CLASS_PDF = 8, /**< Pop Directional Format */ 330 UTF8PROC_BIDI_CLASS_EN = 9, /**< European Number */ 331 UTF8PROC_BIDI_CLASS_ES = 10, /**< European Separator */ 332 UTF8PROC_BIDI_CLASS_ET = 11, /**< European Number Terminator */ 333 UTF8PROC_BIDI_CLASS_AN = 12, /**< Arabic Number */ 334 UTF8PROC_BIDI_CLASS_CS = 13, /**< Common Number Separator */ 335 UTF8PROC_BIDI_CLASS_NSM = 14, /**< Nonspacing Mark */ 336 UTF8PROC_BIDI_CLASS_BN = 15, /**< Boundary Neutral */ 337 UTF8PROC_BIDI_CLASS_B = 16, /**< Paragraph Separator */ 338 UTF8PROC_BIDI_CLASS_S = 17, /**< Segment Separator */ 339 UTF8PROC_BIDI_CLASS_WS = 18, /**< Whitespace */ 340 UTF8PROC_BIDI_CLASS_ON = 19, /**< Other Neutrals */ 341 UTF8PROC_BIDI_CLASS_LRI = 20, /**< Left-to-Right Isolate */ 342 UTF8PROC_BIDI_CLASS_RLI = 21, /**< Right-to-Left Isolate */ 343 UTF8PROC_BIDI_CLASS_FSI = 22, /**< First Strong Isolate */ 344 UTF8PROC_BIDI_CLASS_PDI = 23, /**< Pop Directional Isolate */ 345 } utf8proc_bidi_class_t; 346 347 /** Decomposition type. */ 348 typedef enum { 349 UTF8PROC_DECOMP_TYPE_FONT = 1, /**< Font */ 350 UTF8PROC_DECOMP_TYPE_NOBREAK = 2, /**< Nobreak */ 351 UTF8PROC_DECOMP_TYPE_INITIAL = 3, /**< Initial */ 352 UTF8PROC_DECOMP_TYPE_MEDIAL = 4, /**< Medial */ 353 UTF8PROC_DECOMP_TYPE_FINAL = 5, /**< Final */ 354 UTF8PROC_DECOMP_TYPE_ISOLATED = 6, /**< Isolated */ 355 UTF8PROC_DECOMP_TYPE_CIRCLE = 7, /**< Circle */ 356 UTF8PROC_DECOMP_TYPE_SUPER = 8, /**< Super */ 357 UTF8PROC_DECOMP_TYPE_SUB = 9, /**< Sub */ 358 UTF8PROC_DECOMP_TYPE_VERTICAL = 10, /**< Vertical */ 359 UTF8PROC_DECOMP_TYPE_WIDE = 11, /**< Wide */ 360 UTF8PROC_DECOMP_TYPE_NARROW = 12, /**< Narrow */ 361 UTF8PROC_DECOMP_TYPE_SMALL = 13, /**< Small */ 362 UTF8PROC_DECOMP_TYPE_SQUARE = 14, /**< Square */ 363 UTF8PROC_DECOMP_TYPE_FRACTION = 15, /**< Fraction */ 364 UTF8PROC_DECOMP_TYPE_COMPAT = 16, /**< Compat */ 365 } utf8proc_decomp_type_t; 366 367 /** Boundclass property. (TR29) */ 368 typedef enum { 369 UTF8PROC_BOUNDCLASS_START = 0, /**< Start */ 370 UTF8PROC_BOUNDCLASS_OTHER = 1, /**< Other */ 371 UTF8PROC_BOUNDCLASS_CR = 2, /**< Cr */ 372 UTF8PROC_BOUNDCLASS_LF = 3, /**< Lf */ 373 UTF8PROC_BOUNDCLASS_CONTROL = 4, /**< Control */ 374 UTF8PROC_BOUNDCLASS_EXTEND = 5, /**< Extend */ 375 UTF8PROC_BOUNDCLASS_L = 6, /**< L */ 376 UTF8PROC_BOUNDCLASS_V = 7, /**< V */ 377 UTF8PROC_BOUNDCLASS_T = 8, /**< T */ 378 UTF8PROC_BOUNDCLASS_LV = 9, /**< Lv */ 379 UTF8PROC_BOUNDCLASS_LVT = 10, /**< Lvt */ 380 UTF8PROC_BOUNDCLASS_REGIONAL_INDICATOR = 11, /**< Regional indicator */ 381 UTF8PROC_BOUNDCLASS_SPACINGMARK = 12, /**< Spacingmark */ 382 UTF8PROC_BOUNDCLASS_PREPEND = 13, /**< Prepend */ 383 UTF8PROC_BOUNDCLASS_ZWJ = 14, /**< Zero Width Joiner */ 384 UTF8PROC_BOUNDCLASS_E_BASE = 15, /**< Emoji Base */ 385 UTF8PROC_BOUNDCLASS_E_MODIFIER = 16, /**< Emoji Modifier */ 386 UTF8PROC_BOUNDCLASS_GLUE_AFTER_ZWJ = 17, /**< Glue_After_ZWJ */ 387 UTF8PROC_BOUNDCLASS_E_BASE_GAZ = 18, /**< E_BASE + GLUE_AFTER_ZJW */ 388 } utf8proc_boundclass_t; 389 390 /** 391 * Function pointer type passed to @ref utf8proc_map_custom and 392 * @ref utf8proc_decompose_custom, which is used to specify a user-defined 393 * mapping of codepoints to be applied in conjunction with other mappings. 394 */ 395 typedef utf8proc_int32_t (*utf8proc_custom_func)(utf8proc_int32_t codepoint, void *data); 396 397 #if defined(DOXYGEN) || defined(UTF8PROC_DATA_EXPORT) 398 /** 399 * Array containing the byte lengths of a UTF-8 encoded codepoint based 400 * on the first byte. 401 */ 402 UTF8PROC_DLLEXPORT extern const utf8proc_int8_t utf8proc_utf8class[256]; 403 #endif 404 405 /** 406 * Returns the utf8proc API version as a string MAJOR.MINOR.PATCH 407 * (http://semver.org format), possibly with a "-dev" suffix for 408 * development versions. 409 */ 410 UTF8PROC_DLLEXPORT const char *utf8proc_version(void); 411 412 /** 413 * Returns an informative error string for the given utf8proc error code 414 * (e.g. the error codes returned by @ref utf8proc_map). 415 */ 416 UTF8PROC_DLLEXPORT const char *utf8proc_errmsg(utf8proc_ssize_t errcode); 417 418 /** 419 * Reads a single codepoint from the UTF-8 sequence being pointed to by `str`. 420 * The maximum number of bytes read is `strlen`, unless `strlen` is 421 * negative (in which case up to 4 bytes are read). 422 * 423 * If a valid codepoint could be read, it is stored in the variable 424 * pointed to by `codepoint_ref`, otherwise that variable will be set to -1. 425 * In case of success, the number of bytes read is returned; otherwise, a 426 * negative error code is returned. 427 */ 428 UTF8PROC_DLLEXPORT utf8proc_ssize_t utf8proc_iterate(const utf8proc_uint8_t *str, utf8proc_ssize_t strlen, utf8proc_int32_t *codepoint_ref); 429 430 /** 431 * Check if a codepoint is valid (regardless of whether it has been 432 * assigned a value by the current Unicode standard). 433 * 434 * @return 1 if the given `codepoint` is valid and otherwise return 0. 435 */ 436 UTF8PROC_DLLEXPORT utf8proc_bool utf8proc_codepoint_valid(utf8proc_int32_t codepoint); 437 438 /** 439 * Encodes the codepoint as an UTF-8 string in the byte array pointed 440 * to by `dst`. This array must be at least 4 bytes long. 441 * 442 * In case of success the number of bytes written is returned, and 443 * otherwise 0 is returned. 444 * 445 * This function does not check whether `codepoint` is valid Unicode. 446 */ 447 UTF8PROC_DLLEXPORT utf8proc_ssize_t utf8proc_encode_char(utf8proc_int32_t codepoint, utf8proc_uint8_t *dst); 448 449 /** 450 * Look up the properties for a given codepoint. 451 * 452 * @param codepoint The Unicode codepoint. 453 * 454 * @returns 455 * A pointer to a (constant) struct containing information about 456 * the codepoint. 457 * @par 458 * If the codepoint is unassigned or invalid, a pointer to a special struct is 459 * returned in which `category` is 0 (@ref UTF8PROC_CATEGORY_CN). 460 */ 461 UTF8PROC_DLLEXPORT const utf8proc_property_t *utf8proc_get_property(utf8proc_int32_t codepoint); 462 463 /** Decompose a codepoint into an array of codepoints. 464 * 465 * @param codepoint the codepoint. 466 * @param dst the destination buffer. 467 * @param bufsize the size of the destination buffer. 468 * @param options one or more of the following flags: 469 * - @ref UTF8PROC_REJECTNA - return an error `codepoint` is unassigned 470 * - @ref UTF8PROC_IGNORE - strip "default ignorable" codepoints 471 * - @ref UTF8PROC_CASEFOLD - apply Unicode casefolding 472 * - @ref UTF8PROC_COMPAT - replace certain codepoints with their 473 * compatibility decomposition 474 * - @ref UTF8PROC_CHARBOUND - insert 0xFF bytes before each grapheme cluster 475 * - @ref UTF8PROC_LUMP - lump certain different codepoints together 476 * - @ref UTF8PROC_STRIPMARK - remove all character marks 477 * @param last_boundclass 478 * Pointer to an integer variable containing 479 * the previous codepoint's boundary class if the @ref UTF8PROC_CHARBOUND 480 * option is used. Otherwise, this parameter is ignored. 481 * 482 * @return 483 * In case of success, the number of codepoints written is returned; in case 484 * of an error, a negative error code is returned (@ref utf8proc_errmsg). 485 * @par 486 * If the number of written codepoints would be bigger than `bufsize`, the 487 * required buffer size is returned, while the buffer will be overwritten with 488 * undefined data. 489 */ 490 UTF8PROC_DLLEXPORT utf8proc_ssize_t utf8proc_decompose_char( 491 utf8proc_int32_t codepoint, utf8proc_int32_t *dst, utf8proc_ssize_t bufsize, 492 utf8proc_option_t options, int *last_boundclass 493 ); 494 495 /** 496 * The same as @ref utf8proc_decompose_char, but acts on a whole UTF-8 497 * string and orders the decomposed sequences correctly. 498 * 499 * If the @ref UTF8PROC_NULLTERM flag in `options` is set, processing 500 * will be stopped, when a NULL byte is encounted, otherwise `strlen` 501 * bytes are processed. The result (in the form of 32-bit unicode 502 * codepoints) is written into the buffer being pointed to by 503 * `buffer` (which must contain at least `bufsize` entries). In case of 504 * success, the number of codepoints written is returned; in case of an 505 * error, a negative error code is returned (@ref utf8proc_errmsg). 506 * See @ref utf8proc_decompose_custom to supply additional transformations. 507 * 508 * If the number of written codepoints would be bigger than `bufsize`, the 509 * required buffer size is returned, while the buffer will be overwritten with 510 * undefined data. 511 */ 512 UTF8PROC_DLLEXPORT utf8proc_ssize_t utf8proc_decompose( 513 const utf8proc_uint8_t *str, utf8proc_ssize_t strlen, 514 utf8proc_int32_t *buffer, utf8proc_ssize_t bufsize, utf8proc_option_t options 515 ); 516 517 /** 518 * The same as @ref utf8proc_decompose, but also takes a `custom_func` mapping function 519 * that is called on each codepoint in `str` before any other transformations 520 * (along with a `custom_data` pointer that is passed through to `custom_func`). 521 * The `custom_func` argument is ignored if it is `NULL`. See also @ref utf8proc_map_custom. 522 */ 523 UTF8PROC_DLLEXPORT utf8proc_ssize_t utf8proc_decompose_custom( 524 const utf8proc_uint8_t *str, utf8proc_ssize_t strlen, 525 utf8proc_int32_t *buffer, utf8proc_ssize_t bufsize, utf8proc_option_t options, 526 utf8proc_custom_func custom_func, void *custom_data 527 ); 528 529 /** 530 * Normalizes the sequence of `length` codepoints pointed to by `buffer` 531 * in-place (i.e., the result is also stored in `buffer`). 532 * 533 * @param buffer the (native-endian UTF-32) unicode codepoints to re-encode. 534 * @param length the length (in codepoints) of the buffer. 535 * @param options a bitwise or (`|`) of one or more of the following flags: 536 * - @ref UTF8PROC_NLF2LS - convert LF, CRLF, CR and NEL into LS 537 * - @ref UTF8PROC_NLF2PS - convert LF, CRLF, CR and NEL into PS 538 * - @ref UTF8PROC_NLF2LF - convert LF, CRLF, CR and NEL into LF 539 * - @ref UTF8PROC_STRIPCC - strip or convert all non-affected control characters 540 * - @ref UTF8PROC_COMPOSE - try to combine decomposed codepoints into composite 541 * codepoints 542 * - @ref UTF8PROC_STABLE - prohibit combining characters that would violate 543 * the unicode versioning stability 544 * 545 * @return 546 * In case of success, the length (in codepoints) of the normalized UTF-32 string is 547 * returned; otherwise, a negative error code is returned (@ref utf8proc_errmsg). 548 * 549 * @warning The entries of the array pointed to by `str` have to be in the 550 * range `0x0000` to `0x10FFFF`. Otherwise, the program might crash! 551 */ 552 UTF8PROC_DLLEXPORT utf8proc_ssize_t utf8proc_normalize_utf32(utf8proc_int32_t *buffer, utf8proc_ssize_t length, utf8proc_option_t options); 553 554 /** 555 * Reencodes the sequence of `length` codepoints pointed to by `buffer` 556 * UTF-8 data in-place (i.e., the result is also stored in `buffer`). 557 * Can optionally normalize the UTF-32 sequence prior to UTF-8 conversion. 558 * 559 * @param buffer the (native-endian UTF-32) unicode codepoints to re-encode. 560 * @param length the length (in codepoints) of the buffer. 561 * @param options a bitwise or (`|`) of one or more of the following flags: 562 * - @ref UTF8PROC_NLF2LS - convert LF, CRLF, CR and NEL into LS 563 * - @ref UTF8PROC_NLF2PS - convert LF, CRLF, CR and NEL into PS 564 * - @ref UTF8PROC_NLF2LF - convert LF, CRLF, CR and NEL into LF 565 * - @ref UTF8PROC_STRIPCC - strip or convert all non-affected control characters 566 * - @ref UTF8PROC_COMPOSE - try to combine decomposed codepoints into composite 567 * codepoints 568 * - @ref UTF8PROC_STABLE - prohibit combining characters that would violate 569 * the unicode versioning stability 570 * - @ref UTF8PROC_CHARBOUND - insert 0xFF bytes before each grapheme cluster 571 * 572 * @return 573 * In case of success, the length (in bytes) of the resulting nul-terminated 574 * UTF-8 string is returned; otherwise, a negative error code is returned 575 * (@ref utf8proc_errmsg). 576 * 577 * @warning The amount of free space pointed to by `buffer` must 578 * exceed the amount of the input data by one byte, and the 579 * entries of the array pointed to by `str` have to be in the 580 * range `0x0000` to `0x10FFFF`. Otherwise, the program might crash! 581 */ 582 UTF8PROC_DLLEXPORT utf8proc_ssize_t utf8proc_reencode(utf8proc_int32_t *buffer, utf8proc_ssize_t length, utf8proc_option_t options); 583 584 /** 585 * Given a pair of consecutive codepoints, return whether a grapheme break is 586 * permitted between them (as defined by the extended grapheme clusters in UAX#29). 587 * 588 * @param state Beginning with Version 29 (Unicode 9.0.0), this algorithm requires 589 * state to break graphemes. This state can be passed in as a pointer 590 * in the `state` argument and should initially be set to 0. If the 591 * state is not passed in (i.e. a null pointer is passed), UAX#29 rules 592 * GB10/12/13 which require this state will not be applied, essentially 593 * matching the rules in Unicode 8.0.0. 594 * 595 * @warning If the state parameter is used, `utf8proc_grapheme_break_stateful` must 596 * be called IN ORDER on ALL potential breaks in a string. 597 */ 598 UTF8PROC_DLLEXPORT utf8proc_bool utf8proc_grapheme_break_stateful( 599 utf8proc_int32_t codepoint1, utf8proc_int32_t codepoint2, utf8proc_int32_t *state); 600 601 /** 602 * Same as @ref utf8proc_grapheme_break_stateful, except without support for the 603 * Unicode 9 additions to the algorithm. Supported for legacy reasons. 604 */ 605 UTF8PROC_DLLEXPORT utf8proc_bool utf8proc_grapheme_break( 606 utf8proc_int32_t codepoint1, utf8proc_int32_t codepoint2); 607 608 609 /** 610 * Given a codepoint `c`, return the codepoint of the corresponding 611 * lower-case character, if any; otherwise (if there is no lower-case 612 * variant, or if `c` is not a valid codepoint) return `c`. 613 */ 614 UTF8PROC_DLLEXPORT utf8proc_int32_t utf8proc_tolower(utf8proc_int32_t c); 615 616 /** 617 * Given a codepoint `c`, return the codepoint of the corresponding 618 * upper-case character, if any; otherwise (if there is no upper-case 619 * variant, or if `c` is not a valid codepoint) return `c`. 620 */ 621 UTF8PROC_DLLEXPORT utf8proc_int32_t utf8proc_toupper(utf8proc_int32_t c); 622 623 /** 624 * Given a codepoint `c`, return the codepoint of the corresponding 625 * title-case character, if any; otherwise (if there is no title-case 626 * variant, or if `c` is not a valid codepoint) return `c`. 627 */ 628 UTF8PROC_DLLEXPORT utf8proc_int32_t utf8proc_totitle(utf8proc_int32_t c); 629 630 /** 631 * Given a codepoint, return a character width analogous to `wcwidth(codepoint)`, 632 * except that a width of 0 is returned for non-printable codepoints 633 * instead of -1 as in `wcwidth`. 634 * 635 * @note 636 * If you want to check for particular types of non-printable characters, 637 * (analogous to `isprint` or `iscntrl`), use @ref utf8proc_category. */ 638 UTF8PROC_DLLEXPORT int utf8proc_charwidth(utf8proc_int32_t codepoint); 639 640 /** 641 * Return the Unicode category for the codepoint (one of the 642 * @ref utf8proc_category_t constants.) 643 */ 644 UTF8PROC_DLLEXPORT utf8proc_category_t utf8proc_category(utf8proc_int32_t codepoint); 645 646 /** 647 * Return the two-letter (nul-terminated) Unicode category string for 648 * the codepoint (e.g. `"Lu"` or `"Co"`). 649 */ 650 UTF8PROC_DLLEXPORT const char *utf8proc_category_string(utf8proc_int32_t codepoint); 651 652 /** 653 * Maps the given UTF-8 string pointed to by `str` to a new UTF-8 654 * string, allocated dynamically by `malloc` and returned via `dstptr`. 655 * 656 * If the @ref UTF8PROC_NULLTERM flag in the `options` field is set, 657 * the length is determined by a NULL terminator, otherwise the 658 * parameter `strlen` is evaluated to determine the string length, but 659 * in any case the result will be NULL terminated (though it might 660 * contain NULL characters with the string if `str` contained NULL 661 * characters). Other flags in the `options` field are passed to the 662 * functions defined above, and regarded as described. See also 663 * @ref utfproc_map_custom to supply a custom codepoint transformation. 664 * 665 * In case of success the length of the new string is returned, 666 * otherwise a negative error code is returned. 667 * 668 * @note The memory of the new UTF-8 string will have been allocated 669 * with `malloc`, and should therefore be deallocated with `free`. 670 */ 671 UTF8PROC_DLLEXPORT utf8proc_ssize_t utf8proc_map( 672 const utf8proc_uint8_t *str, utf8proc_ssize_t strlen, utf8proc_uint8_t **dstptr, utf8proc_option_t options 673 ); 674 675 /** 676 * Like @ref utf8proc_map, but also takes a `custom_func` mapping function 677 * that is called on each codepoint in `str` before any other transformations 678 * (along with a `custom_data` pointer that is passed through to `custom_func`). 679 * The `custom_func` argument is ignored if it is `NULL`. 680 */ 681 UTF8PROC_DLLEXPORT utf8proc_ssize_t utf8proc_map_custom( 682 const utf8proc_uint8_t *str, utf8proc_ssize_t strlen, utf8proc_uint8_t **dstptr, utf8proc_option_t options, 683 utf8proc_custom_func custom_func, void *custom_data 684 ); 685 686 /** @name Unicode normalization 687 * 688 * Returns a pointer to newly allocated memory of a NFD, NFC, NFKD or NFKC 689 * normalized version of the null-terminated string `str`. These 690 * are shortcuts to calling @ref utf8proc_map with @ref UTF8PROC_NULLTERM 691 * combined with @ref UTF8PROC_STABLE and flags indicating the normalization. 692 */ 693 /** @{ */ 694 /** NFD normalization (@ref UTF8PROC_DECOMPOSE). */ 695 UTF8PROC_DLLEXPORT utf8proc_uint8_t *utf8proc_NFD(const utf8proc_uint8_t *str); 696 /** NFC normalization (@ref UTF8PROC_COMPOSE). */ 697 UTF8PROC_DLLEXPORT utf8proc_uint8_t *utf8proc_NFC(const utf8proc_uint8_t *str); 698 /** NFKD normalization (@ref UTF8PROC_DECOMPOSE and @ref UTF8PROC_COMPAT). */ 699 UTF8PROC_DLLEXPORT utf8proc_uint8_t *utf8proc_NFKD(const utf8proc_uint8_t *str); 700 /** NFKC normalization (@ref UTF8PROC_COMPOSE and @ref UTF8PROC_COMPAT). */ 701 UTF8PROC_DLLEXPORT utf8proc_uint8_t *utf8proc_NFKC(const utf8proc_uint8_t *str); 702 /** @} */ 703 704 #ifdef __cplusplus 705 } 706 #endif 707 708 #endif 709