1 /* 2 * Copyright (c) 2014-2019 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 up-to-date Unicode versions. 31 * 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 6 75 /** The PATCH version (increased for fixes that do not change the API). */ 76 #define UTF8PROC_VERSION_PATCH 1 77 /** @} */ 78 79 #include <stdlib.h> 80 81 #if defined(_MSC_VER) && _MSC_VER < 1800 82 // MSVC prior to 2013 lacked stdbool.h and inttypes.h 83 typedef signed char utf8proc_int8_t; 84 typedef unsigned char utf8proc_uint8_t; 85 typedef short utf8proc_int16_t; 86 typedef unsigned short utf8proc_uint16_t; 87 typedef int utf8proc_int32_t; 88 typedef unsigned int utf8proc_uint32_t; 89 # ifdef _WIN64 90 typedef __int64 utf8proc_ssize_t; 91 typedef unsigned __int64 utf8proc_size_t; 92 # else 93 typedef int utf8proc_ssize_t; 94 typedef unsigned int utf8proc_size_t; 95 # endif 96 # ifndef __cplusplus 97 // emulate C99 bool 98 typedef unsigned char utf8proc_bool; 99 # ifndef __bool_true_false_are_defined 100 # define false 0 101 # define true 1 102 # define __bool_true_false_are_defined 1 103 # endif 104 # else 105 typedef bool utf8proc_bool; 106 # endif 107 #else 108 # include <stddef.h> 109 # include <stdbool.h> 110 # include <inttypes.h> 111 typedef int8_t utf8proc_int8_t; 112 typedef uint8_t utf8proc_uint8_t; 113 typedef int16_t utf8proc_int16_t; 114 typedef uint16_t utf8proc_uint16_t; 115 typedef int32_t utf8proc_int32_t; 116 typedef uint32_t utf8proc_uint32_t; 117 typedef size_t utf8proc_size_t; 118 typedef ptrdiff_t utf8proc_ssize_t; 119 typedef bool utf8proc_bool; 120 #endif 121 #include <limits.h> 122 123 #ifdef UTF8PROC_STATIC 124 # define UTF8PROC_DLLEXPORT 125 #else 126 # ifdef _WIN32 127 # ifdef UTF8PROC_EXPORTS 128 # define UTF8PROC_DLLEXPORT __declspec(dllexport) 129 # else 130 # define UTF8PROC_DLLEXPORT __declspec(dllimport) 131 # endif 132 # elif __GNUC__ >= 4 133 # define UTF8PROC_DLLEXPORT __attribute__ ((visibility("default"))) 134 # else 135 # define UTF8PROC_DLLEXPORT 136 # endif 137 #endif 138 139 #ifdef __cplusplus 140 extern "C" { 141 #endif 142 143 /** 144 * Option flags used by several functions in the library. 145 */ 146 typedef enum { 147 /** The given UTF-8 input is NULL terminated. */ 148 UTF8PROC_NULLTERM = (1<<0), 149 /** Unicode Versioning Stability has to be respected. */ 150 UTF8PROC_STABLE = (1<<1), 151 /** Compatibility decomposition (i.e. formatting information is lost). */ 152 UTF8PROC_COMPAT = (1<<2), 153 /** Return a result with decomposed characters. */ 154 UTF8PROC_COMPOSE = (1<<3), 155 /** Return a result with decomposed characters. */ 156 UTF8PROC_DECOMPOSE = (1<<4), 157 /** Strip "default ignorable characters" such as SOFT-HYPHEN or ZERO-WIDTH-SPACE. */ 158 UTF8PROC_IGNORE = (1<<5), 159 /** Return an error, if the input contains unassigned codepoints. */ 160 UTF8PROC_REJECTNA = (1<<6), 161 /** 162 * Indicating that NLF-sequences (LF, CRLF, CR, NEL) are representing a 163 * line break, and should be converted to the codepoint for line 164 * separation (LS). 165 */ 166 UTF8PROC_NLF2LS = (1<<7), 167 /** 168 * Indicating that NLF-sequences are representing a paragraph break, and 169 * should be converted to the codepoint for paragraph separation 170 * (PS). 171 */ 172 UTF8PROC_NLF2PS = (1<<8), 173 /** Indicating that the meaning of NLF-sequences is unknown. */ 174 UTF8PROC_NLF2LF = (UTF8PROC_NLF2LS | UTF8PROC_NLF2PS), 175 /** Strips and/or convers control characters. 176 * 177 * NLF-sequences are transformed into space, except if one of the 178 * NLF2LS/PS/LF options is given. HorizontalTab (HT) and FormFeed (FF) 179 * are treated as a NLF-sequence in this case. All other control 180 * characters are simply removed. 181 */ 182 UTF8PROC_STRIPCC = (1<<9), 183 /** 184 * Performs unicode case folding, to be able to do a case-insensitive 185 * string comparison. 186 */ 187 UTF8PROC_CASEFOLD = (1<<10), 188 /** 189 * Inserts 0xFF bytes at the beginning of each sequence which is 190 * representing a single grapheme cluster (see UAX#29). 191 */ 192 UTF8PROC_CHARBOUND = (1<<11), 193 /** Lumps certain characters together. 194 * 195 * E.g. HYPHEN U+2010 and MINUS U+2212 to ASCII "-". See lump.md for details. 196 * 197 * If NLF2LF is set, this includes a transformation of paragraph and 198 * line separators to ASCII line-feed (LF). 199 */ 200 UTF8PROC_LUMP = (1<<12), 201 /** Strips all character markings. 202 * 203 * This includes non-spacing, spacing and enclosing (i.e. accents). 204 * @note This option works only with @ref UTF8PROC_COMPOSE or 205 * @ref UTF8PROC_DECOMPOSE 206 */ 207 UTF8PROC_STRIPMARK = (1<<13), 208 /** 209 * Strip unassigned codepoints. 210 */ 211 UTF8PROC_STRIPNA = (1<<14), 212 } utf8proc_option_t; 213 214 /** @name Error codes 215 * Error codes being returned by almost all functions. 216 */ 217 /** @{ */ 218 /** Memory could not be allocated. */ 219 #define UTF8PROC_ERROR_NOMEM -1 220 /** The given string is too long to be processed. */ 221 #define UTF8PROC_ERROR_OVERFLOW -2 222 /** The given string is not a legal UTF-8 string. */ 223 #define UTF8PROC_ERROR_INVALIDUTF8 -3 224 /** The @ref UTF8PROC_REJECTNA flag was set and an unassigned codepoint was found. */ 225 #define UTF8PROC_ERROR_NOTASSIGNED -4 226 /** Invalid options have been used. */ 227 #define UTF8PROC_ERROR_INVALIDOPTS -5 228 /** @} */ 229 230 /* @name Types */ 231 232 /** Holds the value of a property. */ 233 typedef utf8proc_int16_t utf8proc_propval_t; 234 235 /** Struct containing information about a codepoint. */ 236 typedef struct utf8proc_property_struct { 237 /** 238 * Unicode category. 239 * @see utf8proc_category_t. 240 */ 241 utf8proc_propval_t category; 242 utf8proc_propval_t combining_class; 243 /** 244 * Bidirectional class. 245 * @see utf8proc_bidi_class_t. 246 */ 247 utf8proc_propval_t bidi_class; 248 /** 249 * @anchor Decomposition type. 250 * @see utf8proc_decomp_type_t. 251 */ 252 utf8proc_propval_t decomp_type; 253 utf8proc_uint16_t decomp_seqindex; 254 utf8proc_uint16_t casefold_seqindex; 255 utf8proc_uint16_t uppercase_seqindex; 256 utf8proc_uint16_t lowercase_seqindex; 257 utf8proc_uint16_t titlecase_seqindex; 258 utf8proc_uint16_t comb_index; 259 unsigned bidi_mirrored:1; 260 unsigned comp_exclusion:1; 261 /** 262 * Can this codepoint be ignored? 263 * 264 * Used by @ref utf8proc_decompose_char when @ref UTF8PROC_IGNORE is 265 * passed as an option. 266 */ 267 unsigned ignorable:1; 268 unsigned control_boundary:1; 269 /** The width of the codepoint. */ 270 unsigned charwidth:2; 271 unsigned pad:2; 272 /** 273 * Boundclass. 274 * @see utf8proc_boundclass_t. 275 */ 276 unsigned boundclass:8; 277 } utf8proc_property_t; 278 279 /** Unicode categories. */ 280 typedef enum { 281 UTF8PROC_CATEGORY_CN = 0, /**< Other, not assigned */ 282 UTF8PROC_CATEGORY_LU = 1, /**< Letter, uppercase */ 283 UTF8PROC_CATEGORY_LL = 2, /**< Letter, lowercase */ 284 UTF8PROC_CATEGORY_LT = 3, /**< Letter, titlecase */ 285 UTF8PROC_CATEGORY_LM = 4, /**< Letter, modifier */ 286 UTF8PROC_CATEGORY_LO = 5, /**< Letter, other */ 287 UTF8PROC_CATEGORY_MN = 6, /**< Mark, nonspacing */ 288 UTF8PROC_CATEGORY_MC = 7, /**< Mark, spacing combining */ 289 UTF8PROC_CATEGORY_ME = 8, /**< Mark, enclosing */ 290 UTF8PROC_CATEGORY_ND = 9, /**< Number, decimal digit */ 291 UTF8PROC_CATEGORY_NL = 10, /**< Number, letter */ 292 UTF8PROC_CATEGORY_NO = 11, /**< Number, other */ 293 UTF8PROC_CATEGORY_PC = 12, /**< Punctuation, connector */ 294 UTF8PROC_CATEGORY_PD = 13, /**< Punctuation, dash */ 295 UTF8PROC_CATEGORY_PS = 14, /**< Punctuation, open */ 296 UTF8PROC_CATEGORY_PE = 15, /**< Punctuation, close */ 297 UTF8PROC_CATEGORY_PI = 16, /**< Punctuation, initial quote */ 298 UTF8PROC_CATEGORY_PF = 17, /**< Punctuation, final quote */ 299 UTF8PROC_CATEGORY_PO = 18, /**< Punctuation, other */ 300 UTF8PROC_CATEGORY_SM = 19, /**< Symbol, math */ 301 UTF8PROC_CATEGORY_SC = 20, /**< Symbol, currency */ 302 UTF8PROC_CATEGORY_SK = 21, /**< Symbol, modifier */ 303 UTF8PROC_CATEGORY_SO = 22, /**< Symbol, other */ 304 UTF8PROC_CATEGORY_ZS = 23, /**< Separator, space */ 305 UTF8PROC_CATEGORY_ZL = 24, /**< Separator, line */ 306 UTF8PROC_CATEGORY_ZP = 25, /**< Separator, paragraph */ 307 UTF8PROC_CATEGORY_CC = 26, /**< Other, control */ 308 UTF8PROC_CATEGORY_CF = 27, /**< Other, format */ 309 UTF8PROC_CATEGORY_CS = 28, /**< Other, surrogate */ 310 UTF8PROC_CATEGORY_CO = 29, /**< Other, private use */ 311 } utf8proc_category_t; 312 313 /** Bidirectional character classes. */ 314 typedef enum { 315 UTF8PROC_BIDI_CLASS_L = 1, /**< Left-to-Right */ 316 UTF8PROC_BIDI_CLASS_LRE = 2, /**< Left-to-Right Embedding */ 317 UTF8PROC_BIDI_CLASS_LRO = 3, /**< Left-to-Right Override */ 318 UTF8PROC_BIDI_CLASS_R = 4, /**< Right-to-Left */ 319 UTF8PROC_BIDI_CLASS_AL = 5, /**< Right-to-Left Arabic */ 320 UTF8PROC_BIDI_CLASS_RLE = 6, /**< Right-to-Left Embedding */ 321 UTF8PROC_BIDI_CLASS_RLO = 7, /**< Right-to-Left Override */ 322 UTF8PROC_BIDI_CLASS_PDF = 8, /**< Pop Directional Format */ 323 UTF8PROC_BIDI_CLASS_EN = 9, /**< European Number */ 324 UTF8PROC_BIDI_CLASS_ES = 10, /**< European Separator */ 325 UTF8PROC_BIDI_CLASS_ET = 11, /**< European Number Terminator */ 326 UTF8PROC_BIDI_CLASS_AN = 12, /**< Arabic Number */ 327 UTF8PROC_BIDI_CLASS_CS = 13, /**< Common Number Separator */ 328 UTF8PROC_BIDI_CLASS_NSM = 14, /**< Nonspacing Mark */ 329 UTF8PROC_BIDI_CLASS_BN = 15, /**< Boundary Neutral */ 330 UTF8PROC_BIDI_CLASS_B = 16, /**< Paragraph Separator */ 331 UTF8PROC_BIDI_CLASS_S = 17, /**< Segment Separator */ 332 UTF8PROC_BIDI_CLASS_WS = 18, /**< Whitespace */ 333 UTF8PROC_BIDI_CLASS_ON = 19, /**< Other Neutrals */ 334 UTF8PROC_BIDI_CLASS_LRI = 20, /**< Left-to-Right Isolate */ 335 UTF8PROC_BIDI_CLASS_RLI = 21, /**< Right-to-Left Isolate */ 336 UTF8PROC_BIDI_CLASS_FSI = 22, /**< First Strong Isolate */ 337 UTF8PROC_BIDI_CLASS_PDI = 23, /**< Pop Directional Isolate */ 338 } utf8proc_bidi_class_t; 339 340 /** Decomposition type. */ 341 typedef enum { 342 UTF8PROC_DECOMP_TYPE_FONT = 1, /**< Font */ 343 UTF8PROC_DECOMP_TYPE_NOBREAK = 2, /**< Nobreak */ 344 UTF8PROC_DECOMP_TYPE_INITIAL = 3, /**< Initial */ 345 UTF8PROC_DECOMP_TYPE_MEDIAL = 4, /**< Medial */ 346 UTF8PROC_DECOMP_TYPE_FINAL = 5, /**< Final */ 347 UTF8PROC_DECOMP_TYPE_ISOLATED = 6, /**< Isolated */ 348 UTF8PROC_DECOMP_TYPE_CIRCLE = 7, /**< Circle */ 349 UTF8PROC_DECOMP_TYPE_SUPER = 8, /**< Super */ 350 UTF8PROC_DECOMP_TYPE_SUB = 9, /**< Sub */ 351 UTF8PROC_DECOMP_TYPE_VERTICAL = 10, /**< Vertical */ 352 UTF8PROC_DECOMP_TYPE_WIDE = 11, /**< Wide */ 353 UTF8PROC_DECOMP_TYPE_NARROW = 12, /**< Narrow */ 354 UTF8PROC_DECOMP_TYPE_SMALL = 13, /**< Small */ 355 UTF8PROC_DECOMP_TYPE_SQUARE = 14, /**< Square */ 356 UTF8PROC_DECOMP_TYPE_FRACTION = 15, /**< Fraction */ 357 UTF8PROC_DECOMP_TYPE_COMPAT = 16, /**< Compat */ 358 } utf8proc_decomp_type_t; 359 360 /** Boundclass property. (TR29) */ 361 typedef enum { 362 UTF8PROC_BOUNDCLASS_START = 0, /**< Start */ 363 UTF8PROC_BOUNDCLASS_OTHER = 1, /**< Other */ 364 UTF8PROC_BOUNDCLASS_CR = 2, /**< Cr */ 365 UTF8PROC_BOUNDCLASS_LF = 3, /**< Lf */ 366 UTF8PROC_BOUNDCLASS_CONTROL = 4, /**< Control */ 367 UTF8PROC_BOUNDCLASS_EXTEND = 5, /**< Extend */ 368 UTF8PROC_BOUNDCLASS_L = 6, /**< L */ 369 UTF8PROC_BOUNDCLASS_V = 7, /**< V */ 370 UTF8PROC_BOUNDCLASS_T = 8, /**< T */ 371 UTF8PROC_BOUNDCLASS_LV = 9, /**< Lv */ 372 UTF8PROC_BOUNDCLASS_LVT = 10, /**< Lvt */ 373 UTF8PROC_BOUNDCLASS_REGIONAL_INDICATOR = 11, /**< Regional indicator */ 374 UTF8PROC_BOUNDCLASS_SPACINGMARK = 12, /**< Spacingmark */ 375 UTF8PROC_BOUNDCLASS_PREPEND = 13, /**< Prepend */ 376 UTF8PROC_BOUNDCLASS_ZWJ = 14, /**< Zero Width Joiner */ 377 378 /* the following are no longer used in Unicode 11, but we keep 379 the constants here for backward compatibility */ 380 UTF8PROC_BOUNDCLASS_E_BASE = 15, /**< Emoji Base */ 381 UTF8PROC_BOUNDCLASS_E_MODIFIER = 16, /**< Emoji Modifier */ 382 UTF8PROC_BOUNDCLASS_GLUE_AFTER_ZWJ = 17, /**< Glue_After_ZWJ */ 383 UTF8PROC_BOUNDCLASS_E_BASE_GAZ = 18, /**< E_BASE + GLUE_AFTER_ZJW */ 384 385 /* the Extended_Pictographic property is used in the Unicode 11 386 grapheme-boundary rules, so we store it in the boundclass field */ 387 UTF8PROC_BOUNDCLASS_EXTENDED_PICTOGRAPHIC = 19, 388 UTF8PROC_BOUNDCLASS_E_ZWG = 20, /* UTF8PROC_BOUNDCLASS_EXTENDED_PICTOGRAPHIC + ZWJ */ 389 } utf8proc_boundclass_t; 390 391 /** 392 * Function pointer type passed to @ref utf8proc_map_custom and 393 * @ref utf8proc_decompose_custom, which is used to specify a user-defined 394 * mapping of codepoints to be applied in conjunction with other mappings. 395 */ 396 typedef utf8proc_int32_t (*utf8proc_custom_func)(utf8proc_int32_t codepoint, void *data); 397 398 /** 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]; 403 404 /** 405 * Returns the utf8proc API version as a string MAJOR.MINOR.PATCH 406 * (http://semver.org format), possibly with a "-dev" suffix for 407 * development versions. 408 */ 409 UTF8PROC_DLLEXPORT const char *utf8proc_version(void); 410 411 /** 412 * Returns the utf8proc supported Unicode version as a string MAJOR.MINOR.PATCH. 413 */ 414 UTF8PROC_DLLEXPORT const char *utf8proc_unicode_version(void); 415 416 /** 417 * Returns an informative error string for the given utf8proc error code 418 * (e.g. the error codes returned by @ref utf8proc_map). 419 */ 420 UTF8PROC_DLLEXPORT const char *utf8proc_errmsg(utf8proc_ssize_t errcode); 421 422 /** 423 * Reads a single codepoint from the UTF-8 sequence being pointed to by `str`. 424 * The maximum number of bytes read is `strlen`, unless `strlen` is 425 * negative (in which case up to 4 bytes are read). 426 * 427 * If a valid codepoint could be read, it is stored in the variable 428 * pointed to by `codepoint_ref`, otherwise that variable will be set to -1. 429 * In case of success, the number of bytes read is returned; otherwise, a 430 * negative error code is returned. 431 */ 432 UTF8PROC_DLLEXPORT utf8proc_ssize_t utf8proc_iterate(const utf8proc_uint8_t *str, utf8proc_ssize_t strlen, utf8proc_int32_t *codepoint_ref); 433 434 /** 435 * Check if a codepoint is valid (regardless of whether it has been 436 * assigned a value by the current Unicode standard). 437 * 438 * @return 1 if the given `codepoint` is valid and otherwise return 0. 439 */ 440 UTF8PROC_DLLEXPORT utf8proc_bool utf8proc_codepoint_valid(utf8proc_int32_t codepoint); 441 442 /** 443 * Encodes the codepoint as an UTF-8 string in the byte array pointed 444 * to by `dst`. This array must be at least 4 bytes long. 445 * 446 * In case of success the number of bytes written is returned, and 447 * otherwise 0 is returned. 448 * 449 * This function does not check whether `codepoint` is valid Unicode. 450 */ 451 UTF8PROC_DLLEXPORT utf8proc_ssize_t utf8proc_encode_char(utf8proc_int32_t codepoint, utf8proc_uint8_t *dst); 452 453 /** 454 * Look up the properties for a given codepoint. 455 * 456 * @param codepoint The Unicode codepoint. 457 * 458 * @returns 459 * A pointer to a (constant) struct containing information about 460 * the codepoint. 461 * @par 462 * If the codepoint is unassigned or invalid, a pointer to a special struct is 463 * returned in which `category` is 0 (@ref UTF8PROC_CATEGORY_CN). 464 */ 465 UTF8PROC_DLLEXPORT const utf8proc_property_t *utf8proc_get_property(utf8proc_int32_t codepoint); 466 467 /** Decompose a codepoint into an array of codepoints. 468 * 469 * @param codepoint the codepoint. 470 * @param dst the destination buffer. 471 * @param bufsize the size of the destination buffer. 472 * @param options one or more of the following flags: 473 * - @ref UTF8PROC_REJECTNA - return an error `codepoint` is unassigned 474 * - @ref UTF8PROC_IGNORE - strip "default ignorable" codepoints 475 * - @ref UTF8PROC_CASEFOLD - apply Unicode casefolding 476 * - @ref UTF8PROC_COMPAT - replace certain codepoints with their 477 * compatibility decomposition 478 * - @ref UTF8PROC_CHARBOUND - insert 0xFF bytes before each grapheme cluster 479 * - @ref UTF8PROC_LUMP - lump certain different codepoints together 480 * - @ref UTF8PROC_STRIPMARK - remove all character marks 481 * - @ref UTF8PROC_STRIPNA - remove unassigned codepoints 482 * @param last_boundclass 483 * Pointer to an integer variable containing 484 * the previous codepoint's boundary class if the @ref UTF8PROC_CHARBOUND 485 * option is used. Otherwise, this parameter is ignored. 486 * 487 * @return 488 * In case of success, the number of codepoints written is returned; in case 489 * of an error, a negative error code is returned (@ref utf8proc_errmsg). 490 * @par 491 * If the number of written codepoints would be bigger than `bufsize`, the 492 * required buffer size is returned, while the buffer will be overwritten with 493 * undefined data. 494 */ 495 UTF8PROC_DLLEXPORT utf8proc_ssize_t utf8proc_decompose_char( 496 utf8proc_int32_t codepoint, utf8proc_int32_t *dst, utf8proc_ssize_t bufsize, 497 utf8proc_option_t options, int *last_boundclass 498 ); 499 500 /** 501 * The same as @ref utf8proc_decompose_char, but acts on a whole UTF-8 502 * string and orders the decomposed sequences correctly. 503 * 504 * If the @ref UTF8PROC_NULLTERM flag in `options` is set, processing 505 * will be stopped, when a NULL byte is encountered, otherwise `strlen` 506 * bytes are processed. The result (in the form of 32-bit unicode 507 * codepoints) is written into the buffer being pointed to by 508 * `buffer` (which must contain at least `bufsize` entries). In case of 509 * success, the number of codepoints written is returned; in case of an 510 * error, a negative error code is returned (@ref utf8proc_errmsg). 511 * See @ref utf8proc_decompose_custom to supply additional transformations. 512 * 513 * If the number of written codepoints would be bigger than `bufsize`, the 514 * required buffer size is returned, while the buffer will be overwritten with 515 * undefined data. 516 */ 517 UTF8PROC_DLLEXPORT utf8proc_ssize_t utf8proc_decompose( 518 const utf8proc_uint8_t *str, utf8proc_ssize_t strlen, 519 utf8proc_int32_t *buffer, utf8proc_ssize_t bufsize, utf8proc_option_t options 520 ); 521 522 /** 523 * The same as @ref utf8proc_decompose, but also takes a `custom_func` mapping function 524 * that is called on each codepoint in `str` before any other transformations 525 * (along with a `custom_data` pointer that is passed through to `custom_func`). 526 * The `custom_func` argument is ignored if it is `NULL`. See also @ref utf8proc_map_custom. 527 */ 528 UTF8PROC_DLLEXPORT utf8proc_ssize_t utf8proc_decompose_custom( 529 const utf8proc_uint8_t *str, utf8proc_ssize_t strlen, 530 utf8proc_int32_t *buffer, utf8proc_ssize_t bufsize, utf8proc_option_t options, 531 utf8proc_custom_func custom_func, void *custom_data 532 ); 533 534 /** 535 * Normalizes the sequence of `length` codepoints pointed to by `buffer` 536 * in-place (i.e., the result is also stored in `buffer`). 537 * 538 * @param buffer the (native-endian UTF-32) unicode codepoints to re-encode. 539 * @param length the length (in codepoints) of the buffer. 540 * @param options a bitwise or (`|`) of one or more of the following flags: 541 * - @ref UTF8PROC_NLF2LS - convert LF, CRLF, CR and NEL into LS 542 * - @ref UTF8PROC_NLF2PS - convert LF, CRLF, CR and NEL into PS 543 * - @ref UTF8PROC_NLF2LF - convert LF, CRLF, CR and NEL into LF 544 * - @ref UTF8PROC_STRIPCC - strip or convert all non-affected control characters 545 * - @ref UTF8PROC_COMPOSE - try to combine decomposed codepoints into composite 546 * codepoints 547 * - @ref UTF8PROC_STABLE - prohibit combining characters that would violate 548 * the unicode versioning stability 549 * 550 * @return 551 * In case of success, the length (in codepoints) of the normalized UTF-32 string is 552 * returned; otherwise, a negative error code is returned (@ref utf8proc_errmsg). 553 * 554 * @warning The entries of the array pointed to by `str` have to be in the 555 * range `0x0000` to `0x10FFFF`. Otherwise, the program might crash! 556 */ 557 UTF8PROC_DLLEXPORT utf8proc_ssize_t utf8proc_normalize_utf32(utf8proc_int32_t *buffer, utf8proc_ssize_t length, utf8proc_option_t options); 558 559 /** 560 * Reencodes the sequence of `length` codepoints pointed to by `buffer` 561 * UTF-8 data in-place (i.e., the result is also stored in `buffer`). 562 * Can optionally normalize the UTF-32 sequence prior to UTF-8 conversion. 563 * 564 * @param buffer the (native-endian UTF-32) unicode codepoints to re-encode. 565 * @param length the length (in codepoints) of the buffer. 566 * @param options a bitwise or (`|`) of one or more of the following flags: 567 * - @ref UTF8PROC_NLF2LS - convert LF, CRLF, CR and NEL into LS 568 * - @ref UTF8PROC_NLF2PS - convert LF, CRLF, CR and NEL into PS 569 * - @ref UTF8PROC_NLF2LF - convert LF, CRLF, CR and NEL into LF 570 * - @ref UTF8PROC_STRIPCC - strip or convert all non-affected control characters 571 * - @ref UTF8PROC_COMPOSE - try to combine decomposed codepoints into composite 572 * codepoints 573 * - @ref UTF8PROC_STABLE - prohibit combining characters that would violate 574 * the unicode versioning stability 575 * - @ref UTF8PROC_CHARBOUND - insert 0xFF bytes before each grapheme cluster 576 * 577 * @return 578 * In case of success, the length (in bytes) of the resulting nul-terminated 579 * UTF-8 string is returned; otherwise, a negative error code is returned 580 * (@ref utf8proc_errmsg). 581 * 582 * @warning The amount of free space pointed to by `buffer` must 583 * exceed the amount of the input data by one byte, and the 584 * entries of the array pointed to by `str` have to be in the 585 * range `0x0000` to `0x10FFFF`. Otherwise, the program might crash! 586 */ 587 UTF8PROC_DLLEXPORT utf8proc_ssize_t utf8proc_reencode(utf8proc_int32_t *buffer, utf8proc_ssize_t length, utf8proc_option_t options); 588 589 /** 590 * Given a pair of consecutive codepoints, return whether a grapheme break is 591 * permitted between them (as defined by the extended grapheme clusters in UAX#29). 592 * 593 * @param codepoint1 The first codepoint. 594 * @param codepoint2 The second codepoint, occurring consecutively after `codepoint1`. 595 * @param state Beginning with Version 29 (Unicode 9.0.0), this algorithm requires 596 * state to break graphemes. This state can be passed in as a pointer 597 * in the `state` argument and should initially be set to 0. If the 598 * state is not passed in (i.e. a null pointer is passed), UAX#29 rules 599 * GB10/12/13 which require this state will not be applied, essentially 600 * matching the rules in Unicode 8.0.0. 601 * 602 * @warning If the state parameter is used, `utf8proc_grapheme_break_stateful` must 603 * be called IN ORDER on ALL potential breaks in a string. However, it 604 * is safe to reset the state to zero after a grapheme break. 605 */ 606 UTF8PROC_DLLEXPORT utf8proc_bool utf8proc_grapheme_break_stateful( 607 utf8proc_int32_t codepoint1, utf8proc_int32_t codepoint2, utf8proc_int32_t *state); 608 609 /** 610 * Same as @ref utf8proc_grapheme_break_stateful, except without support for the 611 * Unicode 9 additions to the algorithm. Supported for legacy reasons. 612 */ 613 UTF8PROC_DLLEXPORT utf8proc_bool utf8proc_grapheme_break( 614 utf8proc_int32_t codepoint1, utf8proc_int32_t codepoint2); 615 616 617 /** 618 * Given a codepoint `c`, return the codepoint of the corresponding 619 * lower-case character, if any; otherwise (if there is no lower-case 620 * variant, or if `c` is not a valid codepoint) return `c`. 621 */ 622 UTF8PROC_DLLEXPORT utf8proc_int32_t utf8proc_tolower(utf8proc_int32_t c); 623 624 /** 625 * Given a codepoint `c`, return the codepoint of the corresponding 626 * upper-case character, if any; otherwise (if there is no upper-case 627 * variant, or if `c` is not a valid codepoint) return `c`. 628 */ 629 UTF8PROC_DLLEXPORT utf8proc_int32_t utf8proc_toupper(utf8proc_int32_t c); 630 631 /** 632 * Given a codepoint `c`, return the codepoint of the corresponding 633 * title-case character, if any; otherwise (if there is no title-case 634 * variant, or if `c` is not a valid codepoint) return `c`. 635 */ 636 UTF8PROC_DLLEXPORT utf8proc_int32_t utf8proc_totitle(utf8proc_int32_t c); 637 638 /** 639 * Given a codepoint `c`, return `1` if the codepoint corresponds to a lower-case character 640 * and `0` otherwise. 641 */ 642 UTF8PROC_DLLEXPORT int utf8proc_islower(utf8proc_int32_t c); 643 644 /** 645 * Given a codepoint `c`, return `1` if the codepoint corresponds to an upper-case character 646 * and `0` otherwise. 647 */ 648 UTF8PROC_DLLEXPORT int utf8proc_isupper(utf8proc_int32_t c); 649 650 /** 651 * Given a codepoint, return a character width analogous to `wcwidth(codepoint)`, 652 * except that a width of 0 is returned for non-printable codepoints 653 * instead of -1 as in `wcwidth`. 654 * 655 * @note 656 * If you want to check for particular types of non-printable characters, 657 * (analogous to `isprint` or `iscntrl`), use @ref utf8proc_category. */ 658 UTF8PROC_DLLEXPORT int utf8proc_charwidth(utf8proc_int32_t codepoint); 659 660 /** 661 * Return the Unicode category for the codepoint (one of the 662 * @ref utf8proc_category_t constants.) 663 */ 664 UTF8PROC_DLLEXPORT utf8proc_category_t utf8proc_category(utf8proc_int32_t codepoint); 665 666 /** 667 * Return the two-letter (nul-terminated) Unicode category string for 668 * the codepoint (e.g. `"Lu"` or `"Co"`). 669 */ 670 UTF8PROC_DLLEXPORT const char *utf8proc_category_string(utf8proc_int32_t codepoint); 671 672 /** 673 * Maps the given UTF-8 string pointed to by `str` to a new UTF-8 674 * string, allocated dynamically by `malloc` and returned via `dstptr`. 675 * 676 * If the @ref UTF8PROC_NULLTERM flag in the `options` field is set, 677 * the length is determined by a NULL terminator, otherwise the 678 * parameter `strlen` is evaluated to determine the string length, but 679 * in any case the result will be NULL terminated (though it might 680 * contain NULL characters with the string if `str` contained NULL 681 * characters). Other flags in the `options` field are passed to the 682 * functions defined above, and regarded as described. See also 683 * @ref utf8proc_map_custom to supply a custom codepoint transformation. 684 * 685 * In case of success the length of the new string is returned, 686 * otherwise a negative error code is returned. 687 * 688 * @note The memory of the new UTF-8 string will have been allocated 689 * with `malloc`, and should therefore be deallocated with `free`. 690 */ 691 UTF8PROC_DLLEXPORT utf8proc_ssize_t utf8proc_map( 692 const utf8proc_uint8_t *str, utf8proc_ssize_t strlen, utf8proc_uint8_t **dstptr, utf8proc_option_t options 693 ); 694 695 /** 696 * Like @ref utf8proc_map, but also takes a `custom_func` mapping function 697 * that is called on each codepoint in `str` before any other transformations 698 * (along with a `custom_data` pointer that is passed through to `custom_func`). 699 * The `custom_func` argument is ignored if it is `NULL`. 700 */ 701 UTF8PROC_DLLEXPORT utf8proc_ssize_t utf8proc_map_custom( 702 const utf8proc_uint8_t *str, utf8proc_ssize_t strlen, utf8proc_uint8_t **dstptr, utf8proc_option_t options, 703 utf8proc_custom_func custom_func, void *custom_data 704 ); 705 706 /** @name Unicode normalization 707 * 708 * Returns a pointer to newly allocated memory of a NFD, NFC, NFKD, NFKC or 709 * NFKC_Casefold normalized version of the null-terminated string `str`. These 710 * are shortcuts to calling @ref utf8proc_map with @ref UTF8PROC_NULLTERM 711 * combined with @ref UTF8PROC_STABLE and flags indicating the normalization. 712 */ 713 /** @{ */ 714 /** NFD normalization (@ref UTF8PROC_DECOMPOSE). */ 715 UTF8PROC_DLLEXPORT utf8proc_uint8_t *utf8proc_NFD(const utf8proc_uint8_t *str); 716 /** NFC normalization (@ref UTF8PROC_COMPOSE). */ 717 UTF8PROC_DLLEXPORT utf8proc_uint8_t *utf8proc_NFC(const utf8proc_uint8_t *str); 718 /** NFKD normalization (@ref UTF8PROC_DECOMPOSE and @ref UTF8PROC_COMPAT). */ 719 UTF8PROC_DLLEXPORT utf8proc_uint8_t *utf8proc_NFKD(const utf8proc_uint8_t *str); 720 /** NFKC normalization (@ref UTF8PROC_COMPOSE and @ref UTF8PROC_COMPAT). */ 721 UTF8PROC_DLLEXPORT utf8proc_uint8_t *utf8proc_NFKC(const utf8proc_uint8_t *str); 722 /** 723 * NFKC_Casefold normalization (@ref UTF8PROC_COMPOSE and @ref UTF8PROC_COMPAT 724 * and @ref UTF8PROC_CASEFOLD and @ref UTF8PROC_IGNORE). 725 **/ 726 UTF8PROC_DLLEXPORT utf8proc_uint8_t *utf8proc_NFKC_Casefold(const utf8proc_uint8_t *str); 727 /** @} */ 728 729 #ifdef __cplusplus 730 } 731 #endif 732 733 #endif 734