lib/iconv_open.c

   1 /* Character set conversion.
   2    Copyright (C) 2007, 2009-2018 Free Software Foundation, Inc.
   3
   4    This program is free software; you can redistribute it and/or modify
   5    it under the terms of the GNU General Public License as published by
   6    the Free Software Foundation; either version 2, or (at your option)
   7    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
  12    GNU General Public License for more details.
  13
  14    You should have received a copy of the GNU General Public License along
  15    with this program; if not, see <https://www.gnu.org/licenses/>.  */
  16
  17 #include <config.h>
  18
  19 /* Specification.  */
  20 #include <iconv.h>
  21
  22 #include <errno.h>
  23 #include <string.h>
  24 #include "c-ctype.h"
  25 #include "c-strcase.h"
  26
  27 #define SIZEOF(a) (sizeof(a) / sizeof(a[0]))
  28
  29 /* Namespace cleanliness.  */
  30 #define mapping_lookup rpl_iconv_open_mapping_lookup
  31
  32 /* The macro ICONV_FLAVOR is defined to one of these or undefined.  */
  33
  34 #define ICONV_FLAVOR_AIX "iconv_open-aix.h"
  35 #define ICONV_FLAVOR_HPUX "iconv_open-hpux.h"
  36 #define ICONV_FLAVOR_IRIX "iconv_open-irix.h"
  37 #define ICONV_FLAVOR_OSF "iconv_open-osf.h"
  38 #define ICONV_FLAVOR_SOLARIS "iconv_open-solaris.h"
  39
  40 #ifdef ICONV_FLAVOR
  41 # include ICONV_FLAVOR
  42 #endif
  43
  44 iconv_t
  45 rpl_iconv_open (const char *tocode, const char *fromcode)
  46 #undef iconv_open
  47 {
  48   char fromcode_upper[32];
  49   char tocode_upper[32];
  50   char *fromcode_upper_end;
  51   char *tocode_upper_end;
  52
  53 #if REPLACE_ICONV_UTF
  54   /* Special handling of conversion between UTF-8 and UTF-{16,32}{BE,LE}.
  55      Do this here, before calling the real iconv_open(), because  OSF/1 5.1
  56      iconv() to these encoding inserts a BOM, which is wrong.
  57      We do not need to handle conversion between arbitrary encodings and
  58      UTF-{16,32}{BE,LE}, because the 'striconveh' module implements two-step
  59      conversion through UTF-8.
  60      The _ICONV_* constants are chosen to be disjoint from any iconv_t
  61      returned by the system's iconv_open() functions.  Recall that iconv_t
  62      is a scalar type.  */
  63   if (c_toupper (fromcode[0]) == 'U'
  64       && c_toupper (fromcode[1]) == 'T'
  65       && c_toupper (fromcode[2]) == 'F'
  66       && fromcode[3] == '-')
  67     {
  68       if (c_toupper (tocode[0]) == 'U'
  69           && c_toupper (tocode[1]) == 'T'
  70           && c_toupper (tocode[2]) == 'F'
  71           && tocode[3] == '-')
  72         {
  73           if (strcmp (fromcode + 4, "8") == 0)
  74             {
  75               if (c_strcasecmp (tocode + 4, "16BE") == 0)
  76                 return _ICONV_UTF8_UTF16BE;
  77               if (c_strcasecmp (tocode + 4, "16LE") == 0)
  78                 return _ICONV_UTF8_UTF16LE;
  79               if (c_strcasecmp (tocode + 4, "32BE") == 0)
  80                 return _ICONV_UTF8_UTF32BE;
  81               if (c_strcasecmp (tocode + 4, "32LE") == 0)
  82                 return _ICONV_UTF8_UTF32LE;
  83             }
  84           else if (strcmp (tocode + 4, "8") == 0)
  85             {
  86               if (c_strcasecmp (fromcode + 4, "16BE") == 0)
  87                 return _ICONV_UTF16BE_UTF8;
  88               if (c_strcasecmp (fromcode + 4, "16LE") == 0)
  89                 return _ICONV_UTF16LE_UTF8;
  90               if (c_strcasecmp (fromcode + 4, "32BE") == 0)
  91                 return _ICONV_UTF32BE_UTF8;
  92               if (c_strcasecmp (fromcode + 4, "32LE") == 0)
  93                 return _ICONV_UTF32LE_UTF8;
  94             }
  95         }
  96     }
  97 #endif
  98
  99   /* Do *not* add special support for 8-bit encodings like ASCII or ISO-8859-1
 100      here.  This would lead to programs that work in some locales (such as the
 101      "C" or "en_US" locales) but do not work in East Asian locales.  It is
 102      better if programmers make their programs depend on GNU libiconv (except
 103      on glibc systems), e.g. by using the AM_ICONV macro and documenting the
 104      dependency in an INSTALL or DEPENDENCIES file.  */
 105
 106   /* Try with the original names first.
 107      This covers the case when fromcode or tocode is a lowercase encoding name
 108      that is understood by the system's iconv_open but not listed in our
 109      mappings table.  */
 110   {
 111     iconv_t cd = iconv_open (tocode, fromcode);
 112     if (cd != (iconv_t)(-1))
 113       return cd;
 114   }
 115
 116   /* Convert the encodings to upper case, because
 117        1. in the arguments of iconv_open() on AIX, HP-UX, and OSF/1 the case
 118           matters,
 119        2. it makes searching in the table faster.  */
 120   {
 121     const char *p = fromcode;
 122     char *q = fromcode_upper;
 123     while ((*q = c_toupper (*p)) != '\0')
 124       {
 125         p++;
 126         q++;
 127         if (q == &fromcode_upper[SIZEOF (fromcode_upper)])
 128           {
 129             errno = EINVAL;
 130             return (iconv_t)(-1);
 131           }
 132       }
 133     fromcode_upper_end = q;
 134   }
 135
 136   {
 137     const char *p = tocode;
 138     char *q = tocode_upper;
 139     while ((*q = c_toupper (*p)) != '\0')
 140       {
 141         p++;
 142         q++;
 143         if (q == &tocode_upper[SIZEOF (tocode_upper)])
 144           {
 145             errno = EINVAL;
 146             return (iconv_t)(-1);
 147           }
 148       }
 149     tocode_upper_end = q;
 150   }
 151
 152 #ifdef ICONV_FLAVOR
 153   /* Apply the mappings.  */
 154   {
 155     const struct mapping *m =
 156       mapping_lookup (fromcode_upper, fromcode_upper_end - fromcode_upper);
 157
 158     fromcode = (m != NULL ? m->vendor_name : fromcode_upper);
 159   }
 160   {
 161     const struct mapping *m =
 162       mapping_lookup (tocode_upper, tocode_upper_end - tocode_upper);
 163
 164     tocode = (m != NULL ? m->vendor_name : tocode_upper);
 165   }
 166 #else
 167   fromcode = fromcode_upper;
 168   tocode = tocode_upper;
 169 #endif
 170
 171   return iconv_open (tocode, fromcode);
 172 }