lib/mbsinit.c

   1 /* Test for initial conversion state.
   2    Copyright (C) 2008-2024 Free Software Foundation, Inc.
   3    Written by Bruno Haible <bruno@clisp.org>, 2008.
   4
   5    This file is free software: you can redistribute it and/or modify
   6    it under the terms of the GNU Lesser General Public License as
   7    published by the Free Software Foundation; either version 2.1 of the
   8    License, or (at your option) any later version.
   9
  10    This file 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 Lesser General Public License for more details.
  14
  15    You should have received a copy of the GNU Lesser 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
  24 #if GNULIB_defined_mbstate_t
  25
  26 /* Platforms that lack mbsinit() also lack mbrlen(), mbrtowc(), mbsrtowcs()
  27    and wcrtomb(), wcsrtombs().
  28    We assume that
  29      - sizeof (mbstate_t) >= 4,
  30      - only stateless encodings are supported (such as UTF-8 and EUC-JP, but
  31        not ISO-2022 variants),
  32      - for each encoding, the number of bytes for a wide character is <= 4.
  33        (This maximum is attained for UTF-8, GB18030, EUC-TW.)
  34    We define the meaning of mbstate_t as follows:
  35      - In mb -> wc direction, mbstate_t's first byte contains the number of
  36        buffered bytes (in the range 0..3), followed by up to 3 buffered bytes.
  37        See mbrtowc.c.
  38      - In wc -> mb direction, mbstate_t contains no information. In other
  39        words, it is always in an initial state.  */
  40
  41 static_assert (sizeof (mbstate_t) >= 4);
  42
  43 int
  44 mbsinit (const mbstate_t *ps)
  45 {
  46   const char *pstate = (const char *)ps;
  47
  48   return pstate == NULL || pstate[0] == 0;
  49 }
  50
  51 #else
  52
  53 int
  54 mbsinit (const mbstate_t *ps)
  55 {
  56 # if defined _WIN32 && !defined __CYGWIN__
  57   /* Native Windows.  */
  58   /* MSVC defines 'mbstate_t' as an 8-byte struct; the first 4 bytes matter.
  59      On mingw, 'mbstate_t' is sometimes defined as 'int', sometimes defined as
  60      an 8-byte struct, of which the first 4 bytes matter.  */
  61   return ps == NULL || *(const unsigned int *)ps == 0;
  62 # else
  63   /* Minix, HP-UX 11.00, Solaris 2.6, Interix, ...  */
  64   /* Maybe this definition works, maybe not...  */
  65   return ps == NULL || *(const char *)ps == 0;
  66 # endif
  67 }
  68
  69 #endif