root/maint/gnulib/lib/unistr/u8-to-u32.c

/* */

DEFINITIONS

1. FUNC

   1 /* Convert UTF-8 string to UTF-32 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 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 "unistr.h"
  22 
  23 #define FUNC u8_to_u32
  24 #define SRC_UNIT uint8_t
  25 #define DST_UNIT uint32_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 = u8_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       if (length + 1 > allocated)
  73         {
  74           DST_UNIT *memory;
  75 
  76           allocated = (allocated > 0 ? 2 * allocated : 12);
  77           if (length + 1 > allocated)
  78             allocated = length + 1;
  79           if (result == resultbuf || result == NULL)
  80             memory = (DST_UNIT *) malloc (allocated * sizeof (DST_UNIT));
  81           else
  82             memory =
  83               (DST_UNIT *) realloc (result, allocated * sizeof (DST_UNIT));
  84 
  85           if (memory == NULL)
  86             {
  87               if (!(result == resultbuf || result == NULL))
  88                 free (result);
  89               errno = ENOMEM;
  90               return NULL;
  91             }
  92           if (result == resultbuf && length > 0)
  93             memcpy ((char *) memory, (char *) result,
  94                     length * sizeof (DST_UNIT));
  95           result = memory;
  96         }
  97       result[length++] = uc;
  98     }
  99 
 100   if (length == 0)
 101     {
 102       if (result == NULL)
 103         {
 104           /* Return a non-NULL value.  NULL means error.  */
 105           result = (DST_UNIT *) malloc (1);
 106           if (result == NULL)
 107             {
 108               errno = ENOMEM;
 109               return NULL;
 110             }
 111         }
 112     }
 113   else if (result != resultbuf && length < allocated)
 114     {
 115       /* Shrink the allocated memory if possible.  */
 116       DST_UNIT *memory;
 117 
 118       memory = (DST_UNIT *) realloc (result, length * sizeof (DST_UNIT));
 119       if (memory != NULL)
 120         result = memory;
 121     }
 122 
 123   *lengthp = length;
 124   return result;
 125 }

/* */