lib/mbrtowc.c

   1 /* Convert multibyte character to wide character.
   2    Copyright (C) 1999-2002, 2005-2021 Free Software Foundation, Inc.
   3    Written by Bruno Haible <bruno@clisp.org>, 2008.
   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 3 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,
  11    but WITHOUT ANY WARRANTY; without even the implied warranty of
  12    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  13    GNU 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, see <https://www.gnu.org/licenses/>.  */
  17
  18 #include <config.h>
  19
  20 /* Specification.  */
  21 #include <wchar.h>
  22
  23 #if GNULIB_defined_mbstate_t
  24 /* Implement mbrtowc() on top of mbtowc() for the non-UTF-8 locales
  25    and directly for the UTF-8 locales.  */
  26
  27 # include <errno.h>
  28 # include <stdint.h>
  29 # include <stdlib.h>
  30
  31 # if defined _WIN32 && !defined __CYGWIN__
  32
  33 #  define WIN32_LEAN_AND_MEAN  /* avoid including junk */
  34 #  include <windows.h>
  35
  36 # elif HAVE_PTHREAD_API
  37
  38 #  include <pthread.h>
  39 #  if HAVE_THREADS_H && HAVE_WEAK_SYMBOLS
  40 #   include <threads.h>
  41 #   pragma weak thrd_exit
  42 #   define c11_threads_in_use() (thrd_exit != NULL)
  43 #  else
  44 #   define c11_threads_in_use() 0
  45 #  endif
  46
  47 # elif HAVE_THREADS_H
  48
  49 #  include <threads.h>
  50
  51 # endif
  52
  53 # include "attribute.h"
  54 # include "verify.h"
  55 # include "lc-charset-dispatch.h"
  56 # include "mbtowc-lock.h"
  57
  58 verify (sizeof (mbstate_t) >= 4);
  59 static char internal_state[4];
  60
  61 size_t
  62 mbrtowc (wchar_t *pwc, const char *s, size_t n, mbstate_t *ps)
  63 {
  64 # define FITS_IN_CHAR_TYPE(wc)  ((wc) <= WCHAR_MAX)
  65 # include "mbrtowc-impl.h"
  66 }
  67
  68 #else
  69 /* Override the system's mbrtowc() function.  */
  70
  71 # if MBRTOWC_IN_C_LOCALE_MAYBE_EILSEQ
  72 #  include "hard-locale.h"
  73 #  include <locale.h>
  74 # endif
  75
  76 # undef mbrtowc
  77
  78 size_t
  79 rpl_mbrtowc (wchar_t *pwc, const char *s, size_t n, mbstate_t *ps)
  80 {
  81   size_t ret;
  82   wchar_t wc;
  83
  84 # if MBRTOWC_NULL_ARG2_BUG || MBRTOWC_RETVAL_BUG || MBRTOWC_EMPTY_INPUT_BUG
  85   if (s == NULL)
  86     {
  87       pwc = NULL;
  88       s = "";
  89       n = 1;
  90     }
  91 # endif
  92
  93 # if MBRTOWC_EMPTY_INPUT_BUG
  94   if (n == 0)
  95     return (size_t) -2;
  96 # endif
  97
  98   if (! pwc)
  99     pwc = &wc;
 100
 101 # if MBRTOWC_RETVAL_BUG
 102   {
 103     static mbstate_t internal_state;
 104
 105     /* Override mbrtowc's internal state.  We cannot call mbsinit() on the
 106        hidden internal state, but we can call it on our variable.  */
 107     if (ps == NULL)
 108       ps = &internal_state;
 109
 110     if (!mbsinit (ps))
 111       {
 112         /* Parse the rest of the multibyte character byte for byte.  */
 113         size_t count = 0;
 114         for (; n > 0; s++, n--)
 115           {
 116             ret = mbrtowc (&wc, s, 1, ps);
 117
 118             if (ret == (size_t)(-1))
 119               return (size_t)(-1);
 120             count++;
 121             if (ret != (size_t)(-2))
 122               {
 123                 /* The multibyte character has been completed.  */
 124                 *pwc = wc;
 125                 return (wc == 0 ? 0 : count);
 126               }
 127           }
 128         return (size_t)(-2);
 129       }
 130   }
 131 # endif
 132
 133 # if MBRTOWC_STORES_INCOMPLETE_BUG
 134   ret = mbrtowc (&wc, s, n, ps);
 135   if (ret < (size_t) -2 && pwc != NULL)
 136     *pwc = wc;
 137 # else
 138   ret = mbrtowc (pwc, s, n, ps);
 139 # endif
 140
 141 # if MBRTOWC_NUL_RETVAL_BUG
 142   if (ret < (size_t) -2 && !*pwc)
 143     return 0;
 144 # endif
 145
 146 # if MBRTOWC_IN_C_LOCALE_MAYBE_EILSEQ
 147   if ((size_t) -2 <= ret && n != 0 && ! hard_locale (LC_CTYPE))
 148     {
 149       unsigned char uc = *s;
 150       *pwc = uc;
 151       return 1;
 152     }
 153 # endif
 154
 155   return ret;
 156 }
 157
 158 #endif