wcsmbs/mbsnrtowcs.c

   1 /* Copyright (C) 1996, 1997 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 not,
  17    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 /* We don't need the state really because we don't have shift states
  29    to maintain between calls to this function.  */
  30 static mbstate_t internal;
  31
  32 /* This is a non-standard function but it is very useful in the
  33    implementation of stdio because we have to deal with unterminated
  34    buffers.  At most NMC bytes will be converted.  */
  35 size_t
  36 __mbsnrtowcs (dst, src, nmc, len, ps)
  37      wchar_t *dst;
  38      const char **src;
  39      size_t nmc;
  40      size_t len;
  41      mbstate_t *ps;
  42 {
  43   size_t written = 0;
  44   const char *run = *src;
  45   const char *last = run + nmc;
  46   wchar_t value;
  47   size_t count;
  48
  49   if (ps == NULL)
  50     ps = &internal;
  51
  52   /* Get information from last use of this state.  */
  53   count = ps->count;
  54   value = ps->value;
  55
  56   if (dst == NULL)
  57     /* The LEN parameter has to be ignored if we don't actually write
  58        anything.  */
  59     len = ~0;
  60
  61   /* Copy all words.  */
  62   while (written < len && run < last)
  63     {
  64       unsigned char byte;
  65
  66       /* Store address of next byte to process.  */
  67       *src = run;
  68
  69       /* Start reading a new character only if we are in the initial
  70          state.  */
  71       if (count == 0)
  72         {
  73           byte = *run++;
  74
  75           /* We expect a start of a new multibyte character.  */
  76           if (byte < 0x80)
  77             {
  78               /* One byte sequence.  */
  79               count = 0;
  80               value = byte;
  81             }
  82           else if ((byte & 0xe0) == 0xc0)
  83             {
  84               count = 1;
  85               value = byte & 0x1f;
  86             }
  87           else if ((byte & 0xf0) == 0xe0)
  88             {
  89               /* We expect three bytes.  */
  90               count = 2;
  91               value = byte & 0x0f;
  92             }
  93           else if ((byte & 0xf8) == 0xf0)
  94             {
  95               /* We expect four bytes.  */
  96               count = 3;
  97               value = byte & 0x07;
  98             }
  99           else if ((byte & 0xfc) == 0xf8)
 100             {
 101               /* We expect five bytes.  */
 102               count = 4;
 103               value = byte & 0x03;
 104             }
 105           else if ((byte & 0xfe) == 0xfc)
 106             {
 107               /* We expect six bytes.  */
 108               count = 5;
 109               value = byte & 0x01;
 110             }
 111           else
 112             {
 113               /* This is an illegal encoding.  */
 114               __set_errno (EILSEQ);
 115               return (size_t) -1;
 116             }
 117         }
 118
 119       /* Read the possible remaining bytes.  */
 120       while (run < last && count > 0)
 121         {
 122           byte = *run++;
 123           --count;
 124
 125           if ((byte & 0xc0) != 0x80)
 126             {
 127               /* This is an illegal encoding.  */
 128               __set_errno (EILSEQ);
 129               return (size_t) -1;
 130             }
 131
 132           value <<= 6;
 133           value |= byte & 0x3f;
 134         }
 135
 136       /* If this character is only partially available remember this.  */
 137       if (run == last && count != 0)
 138         {
 139           ps->count = count;
 140           ps->value = value;
 141           break;
 142         }
 143
 144       /* Store value is required.  */
 145       if (dst != NULL)
 146         *dst++ = value;
 147
 148       /* The whole sequence is read.  Check whether end of string is
 149          reached.  */
 150       if (value == L'\0')
 151         {
 152           /* Found the end of the string.  */
 153           *src = NULL;
 154           ps->count = 0;
 155           return written;
 156         }
 157
 158       /* Increment counter of produced words.  */
 159       ++written;
 160     }
 161
 162   /* Store address of next byte to process.  */
 163   *src = run;
 164
 165   return written;
 166 }
 167 weak_alias (__mbsnrtowcs, mbsnrtowcs)