wcsmbs/mbsrtowcs_l.c

   1 /* Copyright (C) 2002, 2004, 2005 Free Software Foundation, Inc.
   2    This file is part of the GNU C Library.
   3    Contributed by Ulrich Drepper <drepper@gnu.org>, 2002.
   4
   5    The GNU C Library is free software; you can redistribute it and/or
   6    modify it under the terms of the GNU Lesser General Public
   7    License as published by the Free Software Foundation; either
   8    version 2.1 of the 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    Lesser General Public License for more details.
  14
  15    You should have received a copy of the GNU Lesser General Public
  16    License along with the GNU C Library; if not, write to the Free
  17    Software Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA
  18    02111-1307 USA.  */
  19
  20 #include <ctype.h>
  21 #include <string.h>
  22 #include "wcsmbsload.h"
  23 #include <dlfcn.h>
  24 #include <errno.h>
  25 #include <gconv.h>
  26 #include <stdlib.h>
  27 #include <string.h>
  28 #include <wchar.h>
  29 #include <wcsmbsload.h>
  30
  31 #include <assert.h>
  32
  33 #ifndef EILSEQ
  34 # define EILSEQ EINVAL
  35 #endif
  36
  37
  38 size_t
  39 attribute_hidden
  40 __mbsrtowcs_l (dst, src, len, ps, l)
  41      wchar_t *dst;
  42      const char **src;
  43      size_t len;
  44      mbstate_t *ps;
  45      __locale_t l;
  46 {
  47   struct __gconv_step_data data;
  48   size_t result;
  49   int status;
  50   struct __gconv_step *towc;
  51   size_t non_reversible;
  52   const struct gconv_fcts *fcts;
  53
  54   /* Tell where we want the result.  */
  55   data.__invocation_counter = 0;
  56   data.__internal_use = 1;
  57   data.__flags = __GCONV_IS_LAST;
  58   data.__statep = ps;
  59   data.__trans = NULL;
  60
  61   /* Get the conversion functions.  */
  62   fcts = get_gconv_fcts (l->__locales[LC_CTYPE]);
  63
  64   /* Get the structure with the function pointers.  */
  65   towc = fcts->towc;
  66
  67   /* We have to handle DST == NULL special.  */
  68   if (dst == NULL)
  69     {
  70       mbstate_t temp_state;
  71       wchar_t buf[64];          /* Just an arbitrary size.  */
  72       const unsigned char *inbuf = (const unsigned char *) *src;
  73       const unsigned char *srcend = inbuf + strlen (*src) + 1;
  74
  75       temp_state = *data.__statep;
  76       data.__statep = &temp_state;
  77
  78       result = 0;
  79       data.__outbufend = (unsigned char *) buf + sizeof (buf);
  80       do
  81         {
  82           data.__outbuf = (unsigned char *) buf;
  83
  84           status = DL_CALL_FCT (towc->__fct,
  85                                 (towc, &data, &inbuf, srcend, NULL,
  86                                  &non_reversible, 0, 1));
  87
  88           result += (wchar_t *) data.__outbuf - buf;
  89         }
  90       while (status == __GCONV_FULL_OUTPUT);
  91
  92       if (status == __GCONV_OK || status == __GCONV_EMPTY_INPUT)
  93         {
  94           /* There better should be a NUL wide char at the end.  */
  95           assert (((wchar_t *) data.__outbuf)[-1] == L'\0');
  96           /* Don't count the NUL character in.  */
  97           --result;
  98         }
  99     }
 100   else
 101     {
 102       /* This code is based on the safe assumption that all internal
 103          multi-byte encodings use the NUL byte only to mark the end
 104          of the string.  */
 105       const unsigned char *srcp = (const unsigned char *) *src;
 106       const unsigned char *srcend;
 107
 108       data.__outbuf = (unsigned char *) dst;
 109       data.__outbufend = data.__outbuf + len * sizeof (wchar_t);
 110
 111       status = __GCONV_FULL_OUTPUT;
 112
 113       while (len > 0)
 114         {
 115           /* Pessimistic guess as to how much input we can use.  In the
 116              worst case we need one input byte for one output wchar_t.  */
 117           srcend = srcp + __strnlen ((const char *) srcp, len) + 1;
 118
 119           status = DL_CALL_FCT (towc->__fct,
 120                                 (towc, &data, &srcp, srcend, NULL,
 121                                  &non_reversible, 0, 1));
 122           if ((status != __GCONV_EMPTY_INPUT
 123                && status != __GCONV_INCOMPLETE_INPUT)
 124               /* Not all input read.  */
 125               || srcp != srcend
 126               /* Reached the end of the input.  */
 127               || srcend[-1] == '\0')
 128             break;
 129
 130           len = (wchar_t *) data.__outbufend - (wchar_t *) data.__outbuf;
 131         }
 132
 133       /* Make the end if the input known to the caller.  */
 134       *src = (const char *) srcp;
 135
 136       result = (wchar_t *) data.__outbuf - dst;
 137
 138       /* We have to determine whether the last character converted
 139          is the NUL character.  */
 140       if ((status == __GCONV_OK || status == __GCONV_EMPTY_INPUT)
 141           && ((wchar_t *) dst)[result - 1] == L'\0')
 142         {
 143           assert (result > 0);
 144           assert (__mbsinit (data.__statep));
 145           *src = NULL;
 146           --result;
 147         }
 148     }
 149
 150   /* There must not be any problems with the conversion but illegal input
 151      characters.  */
 152   assert (status == __GCONV_OK || status == __GCONV_EMPTY_INPUT
 153           || status == __GCONV_ILLEGAL_INPUT
 154           || status == __GCONV_INCOMPLETE_INPUT
 155           || status == __GCONV_FULL_OUTPUT);
 156
 157   if (status != __GCONV_OK && status != __GCONV_FULL_OUTPUT
 158       && status != __GCONV_EMPTY_INPUT && status != __GCONV_INCOMPLETE_INPUT)
 159     {
 160       result = (size_t) -1;
 161       __set_errno (EILSEQ);
 162     }
 163
 164   return result;
 165 }