wcsmbs/mbrtoc8.c

   1 /* Multibyte to UTF-8 conversion.
   2    Copyright (C) 2022 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 #include <sysdep.h>
  27
  28 #ifndef EILSEQ
  29 # define EILSEQ EINVAL
  30 #endif
  31
  32
  33 /* This is the private state used if PS is NULL.  */
  34 static mbstate_t state;
  35
  36 size_t
  37 mbrtoc8 (char8_t *pc8, const char *s, size_t n, mbstate_t *ps)
  38 {
  39   /* This implementation depends on the converter invoked by mbrtowc not
  40      needing to retain state in either the top most bit of ps->__count or
  41      in ps->__value between invocations.  This implementation uses the
  42      top most bit of ps->__count to indicate that trailing code units are
  43      yet to be written and uses ps->__value to store those code units.  */
  44
  45   if (ps == NULL)
  46     ps = &state;
  47
  48   /* If state indicates that trailing code units are yet to be written, write
  49      those first regardless of whether 's' is a null pointer.  */
  50   if (ps->__count & 0x80000000)
  51     {
  52       /* ps->__value.__wchb[3] stores the index of the next code unit to
  53          write.  Code units are stored in reverse order.  */
  54       size_t i = ps->__value.__wchb[3];
  55       if (pc8 != NULL)
  56         {
  57           *pc8 = ps->__value.__wchb[i];
  58         }
  59       if (i == 0)
  60         {
  61           ps->__count &= 0x7fffffff;
  62           ps->__value.__wch = 0;
  63         }
  64       else
  65         --ps->__value.__wchb[3];
  66       return -3;
  67     }
  68
  69   if (s == NULL)
  70     {
  71       /* if 's' is a null pointer, behave as if a null pointer was passed for
  72          'pc8', an empty string was passed for 's', and 1 passed for 'n'.  */
  73       pc8 = NULL;
  74       s = "";
  75       n = 1;
  76     }
  77
  78   wchar_t wc;
  79   size_t result;
  80
  81   result = mbrtowc (&wc, s, n, ps);
  82   if (result <= n)
  83     {
  84       if (wc <= 0x7F)
  85         {
  86           if (pc8 != NULL)
  87             *pc8 = wc;
  88         }
  89       else if (wc <= 0x7FF)
  90         {
  91           if (pc8 != NULL)
  92             *pc8 = 0xC0 + ((wc >> 6) & 0x1F);
  93           ps->__value.__wchb[0] = 0x80 + (wc & 0x3F);
  94           ps->__value.__wchb[3] = 0;
  95           ps->__count |= 0x80000000;
  96         }
  97       else if (wc <= 0xFFFF)
  98         {
  99           if (pc8 != NULL)
 100             *pc8 = 0xE0 + ((wc >> 12) & 0x0F);
 101           ps->__value.__wchb[1] = 0x80 + ((wc >> 6) & 0x3F);
 102           ps->__value.__wchb[0] = 0x80 + (wc & 0x3F);
 103           ps->__value.__wchb[3] = 1;
 104           ps->__count |= 0x80000000;
 105         }
 106       else if (wc <= 0x10FFFF)
 107         {
 108           if (pc8 != NULL)
 109             *pc8 = 0xF0 + ((wc >> 18) & 0x07);
 110           ps->__value.__wchb[2] = 0x80 + ((wc >> 12) & 0x3F);
 111           ps->__value.__wchb[1] = 0x80 + ((wc >> 6) & 0x3F);
 112           ps->__value.__wchb[0] = 0x80 + (wc & 0x3F);
 113           ps->__value.__wchb[3] = 2;
 114           ps->__count |= 0x80000000;
 115         }
 116     }
 117   if (result == 0 && wc != 0)
 118     {
 119       /* mbrtowc() never returns -3.  When a MB sequence converts to multiple
 120          WCs, no input is consumed when writing the subsequent WCs resulting
 121          in a result of 0 even if a null character wasn't written.  */
 122       result = -3;
 123     }
 124
 125   return result;
 126 }