lib/unistr/u16-to-u8.c

   1 /* Convert UTF-16 string to UTF-8 string.
   2    Copyright (C) 2002, 2006-2007, 2009-2021 Free Software Foundation, Inc.
   3    Written by Bruno Haible <bruno@clisp.org>, 2002.
   4
   5    This program is free software: you can redistribute it and/or modify it
   6    under the terms of the GNU Lesser General Public License as published
   7    by the Free Software Foundation; either version 3 of the License, or
   8    (at your option) any later version.
   9
  10    This program 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 License
  16    along with this program.  If not, see <https://www.gnu.org/licenses/>.  */
  17
  18 #include <config.h>
  19
  20 /* Specification.  */
  21 #include "unistr.h"
  22
  23 #define FUNC u16_to_u8
  24 #define SRC_UNIT uint16_t
  25 #define DST_UNIT uint8_t
  26
  27 #include <errno.h>
  28 #include <stdlib.h>
  29 #include <string.h>
  30
  31 DST_UNIT *
  32 FUNC (const SRC_UNIT *s, size_t n, DST_UNIT *resultbuf, size_t *lengthp)
  33 {
  34   const SRC_UNIT *s_end = s + n;
  35   /* Output string accumulator.  */
  36   DST_UNIT *result;
  37   size_t allocated;
  38   size_t length;
  39
  40   if (resultbuf != NULL)
  41     {
  42       result = resultbuf;
  43       allocated = *lengthp;
  44     }
  45   else
  46     {
  47       result = NULL;
  48       allocated = 0;
  49     }
  50   length = 0;
  51   /* Invariants:
  52      result is either == resultbuf or == NULL or malloc-allocated.
  53      If length > 0, then result != NULL.  */
  54
  55   while (s < s_end)
  56     {
  57       ucs4_t uc;
  58       int count;
  59
  60       /* Fetch a Unicode character from the input string.  */
  61       count = u16_mbtoucr (&uc, s, s_end - s);
  62       if (count < 0)
  63         {
  64           if (!(result == resultbuf || result == NULL))
  65             free (result);
  66           errno = EILSEQ;
  67           return NULL;
  68         }
  69       s += count;
  70
  71       /* Store it in the output string.  */
  72       count = u8_uctomb (result + length, uc, allocated - length);
  73       if (count == -1)
  74         {
  75           if (!(result == resultbuf || result == NULL))
  76             free (result);
  77           errno = EILSEQ;
  78           return NULL;
  79         }
  80       if (count == -2)
  81         {
  82           DST_UNIT *memory;
  83
  84           allocated = (allocated > 0 ? 2 * allocated : 12);
  85           if (length + 6 > allocated)
  86             allocated = length + 6;
  87           if (result == resultbuf || result == NULL)
  88             memory = (DST_UNIT *) malloc (allocated * sizeof (DST_UNIT));
  89           else
  90             memory =
  91               (DST_UNIT *) realloc (result, allocated * sizeof (DST_UNIT));
  92
  93           if (memory == NULL)
  94             {
  95               if (!(result == resultbuf || result == NULL))
  96                 free (result);
  97               errno = ENOMEM;
  98               return NULL;
  99             }
 100           if (result == resultbuf && length > 0)
 101             memcpy ((char *) memory, (char *) result,
 102                     length * sizeof (DST_UNIT));
 103           result = memory;
 104           count = u8_uctomb (result + length, uc, allocated - length);
 105           if (count < 0)
 106             abort ();
 107         }
 108       length += count;
 109     }
 110
 111   if (length == 0)
 112     {
 113       if (result == NULL)
 114         {
 115           /* Return a non-NULL value.  NULL means error.  */
 116           result = (DST_UNIT *) malloc (1);
 117           if (result == NULL)
 118             {
 119               errno = ENOMEM;
 120               return NULL;
 121             }
 122         }
 123     }
 124   else if (result != resultbuf && length < allocated)
 125     {
 126       /* Shrink the allocated memory if possible.  */
 127       DST_UNIT *memory;
 128
 129       memory = (DST_UNIT *) realloc (result, length * sizeof (DST_UNIT));
 130       if (memory != NULL)
 131         result = memory;
 132     }
 133
 134   *lengthp = length;
 135   return result;
 136 }