wcsmbs/mbrtoc16.c

   1 /* Copyright (C) 2011-2024 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 <dlfcn.h>
  20 #include <errno.h>
  21 #include <gconv.h>
  22 #include <uchar.h>
  23 #include <wcsmbsload.h>
  24
  25 #include <pointer_guard.h>
  26
  27 #ifndef EILSEQ
  28 # define EILSEQ EINVAL
  29 #endif
  30
  31
  32 /* This is the private state used if PS is NULL.  */
  33 static mbstate_t state;
  34
  35 size_t
  36 mbrtoc16 (char16_t *pc16, const char *s, size_t n, mbstate_t *ps)
  37 {
  38   if (ps == NULL)
  39     ps = &state;
  40
  41   /* The standard text does not say that S being NULL means the state
  42      is reset even if the second half of a surrogate still have to be
  43      returned.  In fact, the error code description indicates
  44      otherwise.  Therefore always first try to return a second
  45      half.  */
  46   if (ps->__count & 0x80000000)
  47     {
  48       /* We have to return the second word for a surrogate.  */
  49       ps->__count &= 0x7fffffff;
  50       *pc16 = ps->__value.__wch;
  51       ps->__value.__wch = L'\0';
  52       return (size_t) -3;
  53     }
  54
  55   wchar_t wc;
  56   struct __gconv_step_data data;
  57   int status;
  58   size_t result;
  59   size_t dummy;
  60   const unsigned char *inbuf, *endbuf;
  61   unsigned char *outbuf = (unsigned char *) &wc;
  62   const struct gconv_fcts *fcts;
  63
  64   /* Set information for this step.  */
  65   data.__invocation_counter = 0;
  66   data.__internal_use = 1;
  67   data.__flags = __GCONV_IS_LAST;
  68   data.__statep = ps;
  69
  70   /* A first special case is if S is NULL.  This means put PS in the
  71      initial state.  */
  72   if (s == NULL)
  73     {
  74       pc16 = NULL;
  75       s = "";
  76       n = 1;
  77     }
  78
  79   if (n == 0)
  80     return (size_t) -2;
  81
  82   /* Tell where we want the result.  */
  83   data.__outbuf = outbuf;
  84   data.__outbufend = outbuf + sizeof (wchar_t);
  85
  86   /* Get the conversion functions.  */
  87   fcts = get_gconv_fcts (_NL_CURRENT_DATA (LC_CTYPE));
  88
  89   /* Do a normal conversion.  */
  90   inbuf = (const unsigned char *) s;
  91   endbuf = inbuf + n;
  92   if (__glibc_unlikely (endbuf < inbuf))
  93     {
  94       endbuf = (const unsigned char *) ~(uintptr_t) 0;
  95       if (endbuf == inbuf)
  96         goto ilseq;
  97     }
  98   __gconv_fct fct = fcts->towc->__fct;
  99   if (fcts->towc->__shlib_handle != NULL)
 100     PTR_DEMANGLE (fct);
 101
 102   status = DL_CALL_FCT (fct, (fcts->towc, &data, &inbuf, endbuf,
 103                               NULL, &dummy, 0, 1));
 104
 105   /* There must not be any problems with the conversion but illegal input
 106      characters.  The output buffer must be large enough, otherwise the
 107      definition of MB_CUR_MAX is not correct.  All the other possible
 108      errors also must not happen.  */
 109   assert (status == __GCONV_OK || status == __GCONV_EMPTY_INPUT
 110           || status == __GCONV_ILLEGAL_INPUT
 111           || status == __GCONV_INCOMPLETE_INPUT
 112           || status == __GCONV_FULL_OUTPUT);
 113
 114   if (status == __GCONV_OK || status == __GCONV_EMPTY_INPUT
 115       || status == __GCONV_FULL_OUTPUT)
 116     {
 117       result = inbuf - (const unsigned char *) s;
 118
 119       if (wc < 0x10000)
 120         {
 121           if (pc16 != NULL)
 122             *pc16 = wc;
 123
 124           if (data.__outbuf != outbuf && wc == L'\0')
 125             {
 126               /* The converted character is the NUL character.  */
 127               assert (__mbsinit (data.__statep));
 128               result = 0;
 129             }
 130         }
 131       else
 132         {
 133           /* This is a surrogate.  */
 134           if (pc16 != NULL)
 135             *pc16 = 0xd7c0 + (wc >> 10);
 136
 137           ps->__count |= 0x80000000;
 138           ps->__value.__wch = 0xdc00 + (wc & 0x3ff);
 139         }
 140     }
 141   else if (status == __GCONV_INCOMPLETE_INPUT)
 142     result = (size_t) -2;
 143   else
 144     {
 145     ilseq:
 146       result = (size_t) -1;
 147       __set_errno (EILSEQ);
 148     }
 149
 150   return result;
 151 }