mdw@git.distorted.org.uk Git - disorder/blob - lib/unicode.h

   1 /*
   2  * This file is part of DisOrde
   3  * Copyright (C) 2007 Richard Kettlewell
   4  *
   5  * This program is free software; you can redistribute it and/or modify
   6  * it under the terms of the GNU General Public License as published by
   7  * the Free Software Foundation; either version 2 of the License, or
   8  * (at your option) any later version.
   9  *
  10  * This program is distributed in the hope that it will be useful, but
  11  * WITHOUT ANY WARRANTY; without even the implied warranty of
  12  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
  13  * General Public License for more details.
  14  *
  15  * You should have received a copy of the GNU General Public License
  16  * along with this program; if not, write to the Free Software
  17  * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307
  18  * USA
  19  */
  20 /** @file lib/unicode.h
  21  * @brief Unicode support functions
  22  */
  23
  24 #ifndef UNICODE_H
  25 #define UNICODE_H
  26
  27 /** @brief Smart pointer into a string
  28  *
  29  * Iterators can be efficiently moved either forwards or back to the start of
  30  * the string.  They cannot (currently) efficiently be moved backwards.  Their
  31  * advantage is that they remember internal state to speed up boundary
  32  * detection.
  33  *
  34  * Iterators can point to any code point of the string, or to a hypothetical
  35  * post-final code point of value 0, but not outside the string.
  36  */
  37 typedef struct utf32_iterator_data *utf32_iterator;
  38
  39 char *utf32_to_utf8(const uint32_t *s, size_t ns, size_t *nd);
  40 uint32_t *utf8_to_utf32(const char *s, size_t ns, size_t *nd);
  41 int utf8_valid(const char *s, size_t ns);
  42
  43 size_t utf32_len(const uint32_t *s);
  44 int utf32_cmp(const uint32_t *a, const uint32_t *b);
  45
  46 uint32_t *utf32_decompose_canon(const uint32_t *s, size_t ns, size_t *ndp);
  47 char *utf8_decompose_canon(const char *s, size_t ns, size_t *ndp);
  48
  49 uint32_t *utf32_decompose_compat(const uint32_t *s, size_t ns, size_t *ndp);
  50 char *utf8_decompose_compat(const char *s, size_t ns, size_t *ndp);
  51
  52 uint32_t *utf32_compose_canon(const uint32_t *s, size_t ns, size_t *ndp);
  53 char *utf8_compose_canon(const char *s, size_t ns, size_t *ndp);
  54
  55 uint32_t *utf32_compose_compat(const uint32_t *s, size_t ns, size_t *ndp);
  56 char *utf8_compose_compat(const char *s, size_t ns, size_t *ndp);
  57
  58 uint32_t *utf32_casefold_canon(const uint32_t *s, size_t ns, size_t *ndp);
  59 char *utf8_casefold_canon(const char *s, size_t ns, size_t *ndp);
  60
  61 uint32_t *utf32_casefold_compat(const uint32_t *s, size_t ns, size_t *ndp);
  62 char *utf8_casefold_compat(const char *s, size_t ns, size_t *ndp);
  63
  64 int utf32_is_grapheme_boundary(const uint32_t *s, size_t ns, size_t n);
  65 int utf32_is_word_boundary(const uint32_t *s, size_t ns, size_t n);
  66
  67 utf32_iterator utf32_iterator_new(const uint32_t *s, size_t ns);
  68 void utf32_iterator_destroy(utf32_iterator it);
  69
  70 size_t utf32_iterator_where(utf32_iterator it);
  71 int utf32_iterator_set(utf32_iterator it, size_t n);
  72 int utf32_iterator_advance(utf32_iterator it, size_t n);
  73 uint32_t utf32_iterator_code(utf32_iterator it);
  74 int utf32_iterator_grapheme_boundary(utf32_iterator it);
  75 int utf32_iterator_word_boundary(utf32_iterator it);
  76
  77 /** @brief Convert 0-terminated UTF-32 to UTF-8
  78  * @param s 0-terminated UTF-32 string
  79  * @return 0-terminated UTF-8 string or 0 on error
  80  *
  81  * See utf32_to_utf8() for possible causes of errors.
  82  */
  83 static inline char *utf32nt_to_utf8(const uint32_t *s) {
  84   return utf32_to_utf8(s, utf32_len(s), 0);
  85 }
  86
  87 /** @brief Convert 0-terminated UTF-8 to UTF-32
  88  * @param s 0-terminated UTF-8 string
  89  * @return 0-terminated UTF-32 string or 0 on error
  90  *
  91  * See utf8_to_utf32() for possible causes of errors.
  92  */
  93 static inline uint32_t *utf8nt_to_utf32(const char *s) {
  94   return utf8_to_utf32(s, strlen(s), 0);
  95 }
  96
  97 #endif /* UNICODE_H */
  98
  99 /*
 100 Local Variables:
 101 c-basic-offset:2
 102 comment-column:40
 103 fill-column:79
 104 indent-tabs-mode:nil
 105 End:
 106 */