wcsmbs/mbsrtowcs_l.c

   1 /* Copyright (C) 2002-2022 Free Software Foundation, Inc.
   2    This file is part of the GNU C Library.
   3
   4    The GNU C Library is free software; you can redistribute it and/or
   5    modify it under the terms of the GNU Lesser General Public
   6    License as published by the Free Software Foundation; either
   7    version 2.1 of the License, or (at your option) any later version.
   8
   9    The GNU C Library is distributed in the hope that it will be useful,
  10    but WITHOUT ANY WARRANTY; without even the implied warranty of
  11    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
  12    Lesser General Public License for more details.
  13
  14    You should have received a copy of the GNU Lesser General Public
  15    License along with the GNU C Library; if not, see
  16    <https://www.gnu.org/licenses/>.  */
  17
  18 #include <assert.h>
  19 #include <ctype.h>
  20 #include <string.h>
  21 #include "wcsmbsload.h"
  22 #include <dlfcn.h>
  23 #include <errno.h>
  24 #include <gconv.h>
  25 #include <stdlib.h>
  26 #include <string.h>
  27 #include <wchar.h>
  28 #include <wcsmbsload.h>
  29
  30 #include <sysdep.h>
  31
  32 #ifndef EILSEQ
  33 # define EILSEQ EINVAL
  34 #endif
  35
  36
  37 size_t
  38 attribute_hidden
  39 __mbsrtowcs_l (wchar_t *dst, const char **src, size_t len, mbstate_t *ps,
  40                locale_t l)
  41 {
  42   struct __gconv_step_data data;
  43   size_t result;
  44   int status;
  45   struct __gconv_step *towc;
  46   size_t non_reversible;
  47   const struct gconv_fcts *fcts;
  48
  49   /* Tell where we want the result.  */
  50   data.__invocation_counter = 0;
  51   data.__internal_use = 1;
  52   data.__flags = __GCONV_IS_LAST;
  53   data.__statep = ps;
  54
  55   /* Get the conversion functions.  */
  56   fcts = get_gconv_fcts (l->__locales[LC_CTYPE]);
  57
  58   /* Get the structure with the function pointers.  */
  59   towc = fcts->towc;
  60   __gconv_fct fct = towc->__fct;
  61 #ifdef PTR_DEMANGLE
  62   if (towc->__shlib_handle != NULL)
  63     PTR_DEMANGLE (fct);
  64 #endif
  65
  66   /* We have to handle DST == NULL special.  */
  67   if (dst == NULL)
  68     {
  69       mbstate_t temp_state;
  70       wchar_t buf[64];          /* Just an arbitrary size.  */
  71       const unsigned char *inbuf = (const unsigned char *) *src;
  72       const unsigned char *srcend = inbuf + strlen (*src) + 1;
  73
  74       temp_state = *data.__statep;
  75       data.__statep = &temp_state;
  76
  77       result = 0;
  78       data.__outbufend = (unsigned char *) buf + sizeof (buf);
  79       do
  80         {
  81           data.__outbuf = (unsigned char *) buf;
  82
  83           status = DL_CALL_FCT (fct, (towc, &data, &inbuf, srcend, NULL,
  84                                       &non_reversible, 0, 1));
  85
  86           result += (wchar_t *) data.__outbuf - buf;
  87         }
  88       while (status == __GCONV_FULL_OUTPUT);
  89
  90       if (status == __GCONV_OK || status == __GCONV_EMPTY_INPUT)
  91         {
  92           /* There better should be a NUL wide char at the end.  */
  93           assert (((wchar_t *) data.__outbuf)[-1] == L'\0');
  94           /* Don't count the NUL character in.  */
  95           --result;
  96         }
  97     }
  98   else
  99     {
 100       /* This code is based on the safe assumption that all internal
 101          multi-byte encodings use the NUL byte only to mark the end
 102          of the string.  */
 103       const unsigned char *srcp = (const unsigned char *) *src;
 104       const unsigned char *srcend;
 105
 106       data.__outbuf = (unsigned char *) dst;
 107       data.__outbufend = data.__outbuf + len * sizeof (wchar_t);
 108
 109       status = __GCONV_FULL_OUTPUT;
 110
 111       while (len > 0)
 112         {
 113           /* Pessimistic guess as to how much input we can use.  In the
 114              worst case we need one input byte for one output wchar_t.  */
 115           srcend = srcp + __strnlen ((const char *) srcp, len) + 1;
 116
 117           status = DL_CALL_FCT (fct, (towc, &data, &srcp, srcend, NULL,
 118                                       &non_reversible, 0, 1));
 119           if ((status != __GCONV_EMPTY_INPUT
 120                && status != __GCONV_INCOMPLETE_INPUT)
 121               /* Not all input read.  */
 122               || srcp != srcend
 123               /* Reached the end of the input.  */
 124               || srcend[-1] == '\0')
 125             break;
 126
 127           len = (wchar_t *) data.__outbufend - (wchar_t *) data.__outbuf;
 128         }
 129
 130       /* Make the end if the input known to the caller.  */
 131       *src = (const char *) srcp;
 132
 133       result = (wchar_t *) data.__outbuf - dst;
 134
 135       /* We have to determine whether the last character converted
 136          is the NUL character.  */
 137       if ((status == __GCONV_OK || status == __GCONV_EMPTY_INPUT)
 138           && ((wchar_t *) dst)[result - 1] == L'\0')
 139         {
 140           assert (result > 0);
 141           assert (__mbsinit (data.__statep));
 142           *src = NULL;
 143           --result;
 144         }
 145     }
 146
 147   /* There must not be any problems with the conversion but illegal input
 148      characters.  */
 149   assert (status == __GCONV_OK || status == __GCONV_EMPTY_INPUT
 150           || status == __GCONV_ILLEGAL_INPUT
 151           || status == __GCONV_INCOMPLETE_INPUT
 152           || status == __GCONV_FULL_OUTPUT);
 153
 154   if (status != __GCONV_OK && status != __GCONV_FULL_OUTPUT
 155       && status != __GCONV_EMPTY_INPUT && status != __GCONV_INCOMPLETE_INPUT)
 156     {
 157       result = (size_t) -1;
 158       __set_errno (EILSEQ);
 159     }
 160
 161   return result;
 162 }