lib/uniconv/u-conv-to-enc.h

   1 /* Conversion from UTF-16/UTF-32 to legacy encodings.
   2    Copyright (C) 2002, 2006-2020 Free Software Foundation, Inc.
   3
   4    This program is free software: you can redistribute it and/or modify it
   5    under the terms of the GNU Lesser General Public License as published
   6    by the Free Software Foundation; either version 3 of the License, or
   7    (at your option) any later version.
   8
   9    This program 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 License
  15    along with this program.  If not, see <https://www.gnu.org/licenses/>.  */
  16
  17 char *
  18 FUNC (const char *tocode,
  19       enum iconv_ilseq_handler handler,
  20       const UNIT *src, size_t srclen,
  21       size_t *offsets,
  22       char *resultbuf, size_t *lengthp)
  23 {
  24 #if HAVE_UTF_NAME
  25   size_t *scaled_offsets;
  26   char *result;
  27   size_t length;
  28
  29   if (offsets != NULL && srclen > 0)
  30     {
  31       scaled_offsets =
  32         (size_t *) malloc (srclen * sizeof (UNIT) * sizeof (size_t));
  33       if (scaled_offsets == NULL)
  34         {
  35           errno = ENOMEM;
  36           return NULL;
  37         }
  38     }
  39   else
  40     scaled_offsets = NULL;
  41
  42   result = resultbuf;
  43   length = *lengthp;
  44   if (mem_iconveha ((const char *) src, srclen * sizeof (UNIT),
  45                     UTF_NAME, tocode,
  46                     handler == iconveh_question_mark, handler,
  47                     scaled_offsets, &result, &length) < 0)
  48     {
  49       int saved_errno = errno;
  50       free (scaled_offsets);
  51       errno = saved_errno;
  52       return NULL;
  53     }
  54
  55   if (offsets != NULL)
  56     {
  57       /* Convert scaled_offsets[srclen * sizeof (UNIT)] to
  58          offsets[srclen].  */
  59       size_t i;
  60
  61       for (i = 0; i < srclen; i++)
  62         offsets[i] = scaled_offsets[i * sizeof (UNIT)];
  63       free (scaled_offsets);
  64     }
  65
  66   if (result == NULL) /* when (resultbuf == NULL && length == 0)  */
  67     {
  68       result = (char *) malloc (1);
  69       if (result == NULL)
  70         {
  71           errno = ENOMEM;
  72           return NULL;
  73         }
  74     }
  75   *lengthp = length;
  76   return result;
  77 #else
  78   uint8_t tmpbuf[4096];
  79   size_t tmpbufsize = SIZEOF (tmpbuf);
  80   uint8_t *utf8_src;
  81   size_t utf8_srclen;
  82   size_t *scaled_offsets;
  83   char *result;
  84
  85   utf8_src = U_TO_U8 (src, srclen, tmpbuf, &tmpbufsize);
  86   if (utf8_src == NULL)
  87     return NULL;
  88   utf8_srclen = tmpbufsize;
  89
  90   if (offsets != NULL && utf8_srclen > 0)
  91     {
  92       scaled_offsets = (size_t *) malloc (utf8_srclen * sizeof (size_t));
  93       if (scaled_offsets == NULL)
  94         {
  95           if (utf8_src != tmpbuf)
  96             free (utf8_src);
  97           errno = ENOMEM;
  98           return NULL;
  99         }
 100     }
 101   else
 102     scaled_offsets = NULL;
 103
 104   result = u8_conv_to_encoding (tocode, handler, utf8_src, utf8_srclen,
 105                                 scaled_offsets, resultbuf, lengthp);
 106   if (result == NULL)
 107     {
 108       int saved_errno = errno;
 109       free (scaled_offsets);
 110       if (utf8_src != tmpbuf)
 111         free (utf8_src);
 112       errno = saved_errno;
 113       return NULL;
 114     }
 115   if (offsets != NULL)
 116     {
 117       size_t iunit;     /* offset into src */
 118       size_t i8;        /* offset into utf8_src */
 119
 120       for (iunit = 0; iunit < srclen; iunit++)
 121         offsets[iunit] = (size_t)(-1);
 122
 123       iunit = 0;
 124       i8 = 0;
 125       while (iunit < srclen && i8 < utf8_srclen)
 126         {
 127           int countunit;
 128           int count8;
 129
 130           offsets[iunit] = scaled_offsets[i8];
 131
 132           countunit = U_MBLEN (src + iunit, srclen - iunit);
 133           count8 = u8_mblen (utf8_src + i8, utf8_srclen - i8);
 134           if (countunit < 0 || count8 < 0)
 135             abort ();
 136           iunit += countunit;
 137           i8 += count8;
 138         }
 139       /* Check that utf8_src has been traversed entirely.  */
 140       if (i8 < utf8_srclen)
 141         abort ();
 142       /* Check that src has been traversed entirely, except possibly for an
 143          incomplete sequence of units at the end.  */
 144       if (iunit < srclen)
 145         {
 146           offsets[iunit] = *lengthp;
 147           if (!(U_MBLEN (src + iunit, srclen - iunit) < 0))
 148             abort ();
 149         }
 150       free (scaled_offsets);
 151     }
 152   if (utf8_src != tmpbuf)
 153     free (utf8_src);
 154   return result;
 155 #endif
 156 }