wcsmbs/mbrtowc.c

   1 /* Copyright (C) 1996 Free Software Foundation, Inc.
   2 This file is part of the GNU C Library.
   3 Contributed by Ulrich Drepper <drepper@gnu.ai.mit.edu>, 1996.
   4
   5 The GNU C Library is free software; you can redistribute it and/or
   6 modify it under the terms of the GNU Library General Public License as
   7 published by the Free Software Foundation; either version 2 of the
   8 License, or (at your option) any later version.
   9
  10 The GNU C Library 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 GNU
  13 Library General Public License for more details.
  14
  15 You should have received a copy of the GNU Library General Public
  16 License along with the GNU C Library; see the file COPYING.LIB.  If
  17 not, write to the Free Software Foundation, Inc., 59 Temple Place - Suite 330,
  18 Boston, MA 02111-1307, USA.  */
  19
  20 #include <errno.h>
  21 #include <wchar.h>
  22
  23 #ifndef EILSEQ
  24 #define EILSEQ EINVAL
  25 #endif
  26
  27
  28 static mbstate_t internal;
  29
  30 size_t
  31 mbrtowc (wchar_t *pwc, const char *s, size_t n, mbstate_t *ps)
  32 {
  33   wchar_t to_wide;
  34   size_t used = 0;
  35
  36   if (ps == NULL)
  37     ps = &internal;
  38
  39   if (s == NULL)
  40     {
  41       /* See first paragraph of description in 7.16.6.3.2.  */
  42       pwc = NULL;
  43       s = "";
  44       n = 1;
  45     }
  46
  47   if (n > 0)
  48     {
  49       if (ps->count == 0)
  50         {
  51           unsigned char byte = (unsigned char) *s++;
  52           ++used;
  53
  54           /* We must look for a possible first byte of a UTF8 sequence.  */
  55           if (byte < 0x80)
  56             {
  57               /* One byte sequence.  */
  58               if (pwc != NULL)
  59                 *pwc = (wchar_t) byte;
  60               return byte ? used : 0;
  61             }
  62
  63           if ((byte & 0xc0) == 0x80 || (byte & 0xfe) == 0xfe)
  64             {
  65               /* Oh, oh.  An encoding error.  */
  66               errno = EILSEQ;
  67               return (size_t) -1;
  68             }
  69
  70           if ((byte & 0xe0) == 0xc0)
  71             {
  72               /* We expect two bytes.  */
  73               ps->count = 1;
  74               ps->value = byte & 0x1f;
  75             }
  76           else if ((byte & 0xf0) == 0xe0)
  77             {
  78               /* We expect three bytes.  */
  79               ps->count = 2;
  80               ps->value = byte & 0x0f;
  81             }
  82           else if ((byte & 0xf8) == 0xf0)
  83             {
  84               /* We expect four bytes.  */
  85               ps->count = 3;
  86               ps->value = byte & 0x07;
  87             }
  88           else if ((byte & 0xfc) == 0xf8)
  89             {
  90               /* We expect five bytes.  */
  91               ps->count = 4;
  92               ps->value = byte & 0x03;
  93             }
  94           else
  95             {
  96               /* We expect six bytes.  */
  97               ps->count = 5;
  98               ps->value = byte & 0x01;
  99             }
 100         }
 101
 102       /* We know we have to handle a multibyte character and there are
 103          some more bytes to read.  */
 104       while (used < n)
 105         {
 106           /* The second to sixths byte must be of the form 10xxxxxx.  */
 107           unsigned char byte = (unsigned char) *s++;
 108           ++used;
 109
 110           if ((byte & 0xc0) != 0x80)
 111             {
 112               /* Oh, oh.  An encoding error.  */
 113               errno = EILSEQ;
 114               return (size_t) -1;
 115             }
 116
 117           ps->value <<= 6;
 118           ps->value |= byte & 0x3f;
 119
 120           if (--ps->count == 0)
 121             {
 122               /* The character is finished.  */
 123               if (pwc != NULL)
 124                 *pwc = (wchar_t) ps->value;
 125               return ps->value ? used : 0;
 126             }
 127         }
 128     }
 129
 130   return (size_t) -2;
 131 }