wcsmbs/mbrtoc8.c

   1 /* Multibyte to UTF-8 conversion.
   2    Copyright (C) 2022-2024 Free Software Foundation, Inc.
   3    This file is part of the GNU C Library.
   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, see
  17    <https://www.gnu.org/licenses/>.  */
  18
  19 #include <assert.h>
  20 #include <dlfcn.h>
  21 #include <errno.h>
  22 #include <gconv.h>
  23 #include <uchar.h>
  24 #include <wcsmbsload.h>
  25
  26 #ifndef EILSEQ
  27 # define EILSEQ EINVAL
  28 #endif
  29
  30
  31 /* This is the private state used if PS is NULL.  */
  32 static mbstate_t state;
  33
  34 size_t
  35 mbrtoc8 (char8_t *pc8, const char *s, size_t n, mbstate_t *ps)
  36 {
  37   /* This implementation depends on the converter invoked by mbrtowc not
  38      needing to retain state in either the top most bit of ps->__count or
  39      in ps->__value between invocations.  This implementation uses the
  40      top most bit of ps->__count to indicate that trailing code units are
  41      yet to be written and uses ps->__value to store those code units.  */
  42
  43   if (ps == NULL)
  44     ps = &state;
  45
  46   /* If state indicates that trailing code units are yet to be written, write
  47      those first regardless of whether 's' is a null pointer.  */
  48   if (ps->__count & 0x80000000)
  49     {
  50       /* ps->__value.__wchb[3] stores the index of the next code unit to
  51          write.  Code units are stored in reverse order.  */
  52       size_t i = ps->__value.__wchb[3];
  53       if (pc8 != NULL)
  54         {
  55           *pc8 = ps->__value.__wchb[i];
  56         }
  57       if (i == 0)
  58         {
  59           ps->__count &= 0x7fffffff;
  60           ps->__value.__wch = 0;
  61         }
  62       else
  63         --ps->__value.__wchb[3];
  64       return -3;
  65     }
  66
  67   if (s == NULL)
  68     {
  69       /* if 's' is a null pointer, behave as if a null pointer was passed for
  70          'pc8', an empty string was passed for 's', and 1 passed for 'n'.  */
  71       pc8 = NULL;
  72       s = "";
  73       n = 1;
  74     }
  75
  76   wchar_t wc;
  77   size_t result;
  78
  79   result = mbrtowc (&wc, s, n, ps);
  80   if (result <= n)
  81     {
  82       if (wc <= 0x7F)
  83         {
  84           if (pc8 != NULL)
  85             *pc8 = wc;
  86         }
  87       else if (wc <= 0x7FF)
  88         {
  89           if (pc8 != NULL)
  90             *pc8 = 0xC0 + ((wc >> 6) & 0x1F);
  91           ps->__value.__wchb[0] = 0x80 + (wc & 0x3F);
  92           ps->__value.__wchb[3] = 0;
  93           ps->__count |= 0x80000000;
  94         }
  95       else if (wc <= 0xFFFF)
  96         {
  97           if (pc8 != NULL)
  98             *pc8 = 0xE0 + ((wc >> 12) & 0x0F);
  99           ps->__value.__wchb[1] = 0x80 + ((wc >> 6) & 0x3F);
 100           ps->__value.__wchb[0] = 0x80 + (wc & 0x3F);
 101           ps->__value.__wchb[3] = 1;
 102           ps->__count |= 0x80000000;
 103         }
 104       else if (wc <= 0x10FFFF)
 105         {
 106           if (pc8 != NULL)
 107             *pc8 = 0xF0 + ((wc >> 18) & 0x07);
 108           ps->__value.__wchb[2] = 0x80 + ((wc >> 12) & 0x3F);
 109           ps->__value.__wchb[1] = 0x80 + ((wc >> 6) & 0x3F);
 110           ps->__value.__wchb[0] = 0x80 + (wc & 0x3F);
 111           ps->__value.__wchb[3] = 2;
 112           ps->__count |= 0x80000000;
 113         }
 114     }
 115   if (result == 0 && wc != 0)
 116     {
 117       /* mbrtowc() never returns -3.  When a MB sequence converts to multiple
 118          WCs, no input is consumed when writing the subsequent WCs resulting
 119          in a result of 0 even if a null character wasn't written.  */
 120       result = -3;
 121     }
 122
 123   return result;
 124 }