src/charset.cpp

   1 /***************************************************************************
   2  *   Copyright (C) 2008-2009 by Andrzej Rybczak                            *
   3  *   electricityispower@gmail.com                                          *
   4  *                                                                         *
   5  *   This program is free software; you can redistribute it and/or modify  *
   6  *   it under the terms of the GNU General Public License as published by  *
   7  *   the Free Software Foundation; either version 2 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         *
  13  *   GNU General Public License for more details.                          *
  14  *                                                                         *
  15  *   You should have received a copy of the GNU General Public License     *
  16  *   along with this program; if not, write to the                         *
  17  *   Free Software Foundation, Inc.,                                       *
  18  *   51 Franklin St, Fifth Floor, Boston, MA 02110-1301, USA.              *
  19  ***************************************************************************/
  20
  21 #include "charset.h"
  22
  23 #ifdef HAVE_ICONV_H
  24
  25 #include <iconv.h>
  26 #include <cstdlib>
  27 #include <cstring>
  28 #include <fstream>
  29 #include <iostream>
  30 #include <stdexcept>
  31
  32 #include "settings.h"
  33
  34 namespace
  35 {
  36         inline bool char_non_ascii(char ch)
  37         {
  38                 return (ch & 0x80) != 0;
  39         }
  40
  41         bool has_non_ascii_chars(const char *s)
  42         {
  43                 for (; *s; ++s)
  44                         if (char_non_ascii(*s))
  45                                 return true;
  46                 return false;
  47         }
  48
  49         bool has_non_ascii_chars(const std::string &s)
  50         {
  51                 for (std::string::const_iterator it = s.begin(); it != s.end(); ++it)
  52                         if (char_non_ascii(*it))
  53                                 return true;
  54                 return false;
  55         }
  56
  57         void charset_convert(const char *from, const char *to, const char *&inbuf, bool delete_old, size_t len = 0)
  58         {
  59                 if (!inbuf || !from || !to)
  60                         return;
  61
  62                 iconv_t cd = iconv_open(to, from);
  63
  64                 if (cd == iconv_t(-1))
  65                         return;
  66
  67                 if (!len)
  68                         len = strlen(inbuf);
  69                 size_t buflen = len*6+1;
  70                 char *outbuf = new char[buflen];
  71                 char *outstart = outbuf;
  72                 const char *instart = inbuf;
  73
  74                 if (iconv(cd, const_cast<ICONV_CONST char **>(&inbuf), &len, &outbuf, &buflen) == size_t(-1))
  75                 {
  76                         inbuf = instart;
  77                         delete [] outstart;
  78                         iconv_close(cd);
  79                         return;
  80                 }
  81                 iconv_close(cd);
  82                 *outbuf = 0;
  83                 if (delete_old)
  84                         delete [] instart;
  85                 inbuf = outstart;
  86         }
  87 }
  88
  89 void utf_to_locale(std::string &s)
  90 {
  91         if (s.empty() || Config.system_encoding.empty() || !has_non_ascii_chars(s))
  92                 return;
  93         const char *tmp = strdup(s.c_str());
  94         charset_convert("utf-8", Config.system_encoding.c_str(), tmp, 1, s.length());
  95         s = tmp;
  96         delete [] tmp;
  97 }
  98
  99 std::string utf_to_locale_cpy(const std::string &s)
 100 {
 101         std::string result = s;
 102         utf_to_locale(result);
 103         return result;
 104 }
 105
 106 void locale_to_utf(std::string &s)
 107 {
 108         if (s.empty() || Config.system_encoding.empty() || !has_non_ascii_chars(s))
 109                 return;
 110         const char *tmp = strdup(s.c_str());
 111         charset_convert(Config.system_encoding.c_str(), "utf-8", tmp, 1, s.length());
 112         s = tmp;
 113         delete [] tmp;
 114 }
 115
 116 std::string locale_to_utf_cpy(const std::string &s)
 117 {
 118         std::string result = s;
 119         locale_to_utf(result);
 120         return result;
 121 }
 122
 123 void utf_to_locale(const char *&s, bool delete_old)
 124 {
 125         if (!s || Config.system_encoding.empty() || !has_non_ascii_chars(s))
 126                 return;
 127         charset_convert("utf-8", Config.system_encoding.c_str(), s, delete_old);
 128 }
 129
 130 void locale_to_utf(const char *&s, bool delete_old)
 131 {
 132         if (!s || Config.system_encoding.empty() || !has_non_ascii_chars(s))
 133                 return;
 134         charset_convert(Config.system_encoding.c_str(), "utf-8", s, delete_old);
 135 }
 136
 137 #endif // HAVE_ICONV_H
 138