[Sommaire] [Imprimer]

   1  <?php
   2  //
   3  // Definition of eZMBStringMapper class
   4  //
   5  // Created on: <12-Jul-2002 12:56:48 amos>
   6  //
   7  // SOFTWARE NAME: eZ publish
   8  // SOFTWARE RELEASE: 3.9.0
   9  // BUILD VERSION: 17785
  10  // COPYRIGHT NOTICE: Copyright (C) 1999-2006 eZ systems AS
  11  // SOFTWARE LICENSE: GNU General Public License v2.0
  12  // NOTICE: >
  13  //   This program is free software; you can redistribute it and/or
  14  //   modify it under the terms of version 2.0  of the GNU General
  15  //   Public License as published by the Free Software Foundation.
  16  //
  17  //   This program is distributed in the hope that it will be useful,
  18  //   but WITHOUT ANY WARRANTY; without even the implied warranty of
  19  //   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  20  //   GNU General Public License for more details.
  21  //
  22  //   You should have received a copy of version 2.0 of the GNU General
  23  //   Public License along with this program; if not, write to the Free
  24  //   Software Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston,
  25  //   MA 02110-1301, USA.
  26  //
  27  //
  28  
  29  /*! \file ezmbstringmapper.php
  30  */
  31  
  32  /*!
  33    \class eZMBStringMapper ezmbstringmapper.php
  34    \ingroup eZI18N
  35    \brief The class eZMBStringMapper does
  36  
  37    The mbstring extension supports the following charset:
  38    UCS-4, UCS-4BE, UCS-4LE, UCS-2, UCS-2BE, UCS-2LE, UTF-32, UTF-32BE, UTF-32LE, UCS-2LE, UTF-16,
  39    UTF-16BE, UTF-16LE, UTF-8, UTF-7, ASCII, EUC-JP, SJIS, eucJP-win, SJIS-win, ISO-2022-JP, JIS,
  40    ISO-8859-1, ISO-8859-2, ISO-8859-3, ISO-8859-4, ISO-8859-5, ISO-8859-6, ISO-8859-7, ISO-8859-8,
  41    ISO-8859-9, ISO-8859-10, ISO-8859-13, ISO-8859-14, ISO-8859-15, byte2be, byte2le, byte4be,
  42    byte4le, BASE64, 7bit, 8bit and UTF7-IMAP.
  43  */
  44  
  45  include_once ( "lib/ezi18n/classes/ezcharsetinfo.php" );
  46  
  47  class eZMBStringMapper
  48  {
  49      /*!
  50       Constructor
  51      */
  52      function eZMBStringMapper( $input_charset_code, $output_charset_code )
  53      {
  54          $this->RequestedInputCharsetCode = $input_charset_code;
  55          $this->InputCharsetCode = eZCharsetInfo::realCharsetCode( $input_charset_code );
  56          $this->RequestedOutputCharsetCode = $output_charset_code;
  57          $this->OutputCharsetCode = eZCharsetInfo::realCharsetCode( $output_charset_code );
  58          $this->Valid = false;
  59          if ( !$this->isCharsetSupported( $input_charset_code ) )
  60          {
  61              eZDebug::writeError( "Input charset $input_charset_code not supported", "eZMBStringMapper" );
  62          }
  63          else if ( !$this->isCharsetSupported( $output_charset_code ) )
  64          {
  65              eZDebug::writeError( "Output charset $output_charset_code not supported", "eZMBStringMapper" );
  66          }
  67          else if ( $this->hasMBStringExtension() )
  68              $this->Valid = true;
  69          else
  70              eZDebug::writeError( "No mbstring functions available", "eZMBStringMapper" );
  71      }
  72  
  73      /*!
  74       \static
  75       \note This function is duplicated in eZTextCodec::eZTextCodec(), remember to update both places.
  76      */
  77      function &charsetList()
  78      {
  79          $charsets =& $GLOBALS["eZMBCharsetList"];
  80          if ( !is_array( $charsets ) )
  81          {
  82              $charsetList = array( "ucs-4", "ucs-4be", "ucs-4le", "ucs-2", "ucs-2be", "ucs-2le", "utf-32", "utf-32be", "utf-32le", "utf-16",
  83                                    "utf-16be", "utf-16le", "utf-8", "utf-7", "ascii", "euc-jp", "sjis", "eucjp-win", "sjis-win", "iso-2022-jp", "jis",
  84                                    "iso-8859-1", "iso-8859-2", "iso-8859-3", "iso-8859-4", "iso-8859-5", "iso-8859-6", "iso-8859-7", "iso-8859-8",
  85                                    "iso-8859-9", "iso-8859-10", "iso-8859-13", "iso-8859-14", "iso-8859-15", "byte2be", "byte2le", "byte4be",
  86                                    "byte4le", "base64", "7bit", "8bit", "utf7-imap" );
  87              $charsets = array();
  88              foreach ( $charsetList as $charset )
  89              {
  90                  $charsets[$charset] = $charset;
  91              }
  92          }
  93          return $charsets;
  94      }
  95  
  96      /*!
  97       \static
  98       \return \c true if the mbstring can be used.
  99       \note The following function must be present for the function to return \c true.
 100             mb_convert_encoding
 101             mb_substitute_character
 102             mb_strcut
 103             mb_strlen
 104             mb_strpos
 105             mb_strrpos
 106             mb_strwidth
 107             mb_substr
 108       \note This function is duplicated in eZTextCodec::eZTextCodec(), remember to update both places.
 109      */
 110      function hasMBStringExtension()
 111      {
 112          return ( function_exists( "mb_convert_encoding" ) and
 113                   function_exists( "mb_substitute_character" ) and
 114                   function_exists( "mb_strcut" ) and
 115                   function_exists( "mb_strlen" ) and
 116                   function_exists( "mb_strpos" ) and
 117                   function_exists( "mb_strrpos" ) and
 118                   function_exists( "mb_strwidth" ) and
 119                   function_exists( "mb_substr" ) );
 120      }
 121  
 122      function inputCharsetCode()
 123      {
 124          return $this->InputCharsetCode;
 125      }
 126  
 127      function outputCharsetCode()
 128      {
 129          return $this->OutputCharsetCode;
 130      }
 131  
 132      function requestedInputCharsetCode()
 133      {
 134          return $this->RequestedInputCharsetCode;
 135      }
 136  
 137      function requestedOutputCharsetCode()
 138      {
 139          return $this->RequestedOutputCharsetCode;
 140      }
 141  
 142      function isCharsetSupported( $charset_code )
 143      {
 144          $charset_code = eZCharsetInfo::realCharsetCode( $charset_code );
 145          return in_array( $charset_code, eZMBStringMapper::charsetList() );
 146      }
 147  
 148      function substituteCharacter()
 149      {
 150          if ( !$this->Valid )
 151              return null;
 152          return mb_substitute_character();
 153      }
 154  
 155      function setSubstituteCharacter( $char )
 156      {
 157          if ( $this->Valid )
 158              mb_substitute_character( $char );
 159      }
 160  
 161      function convertString( $str )
 162      {
 163          if ( !$this->Valid )
 164              return $str;
 165          return mb_convert_encoding( $str, $this->OutputCharsetCode, $this->InputCharsetCode );
 166      }
 167  
 168      function strlen( $str )
 169      {
 170          return mb_strlen( $str, $this->InputCharsetCode );
 171      }
 172  
 173      function strpos( $haystack, $needle, $offset = 0 )
 174      {
 175          return mb_strpos( $haystack, $needle, $offset, $this->InputCharsetCode );
 176      }
 177  
 178      function strrpos( $haystack, $needle )
 179      {
 180          return mb_strrpos( $haystack, $needle, $this->InputCharsetCode );
 181      }
 182  
 183      function substr( $str, $start, $length )
 184      {
 185          return mb_substr( $str, $start, $length, $this->InputCharsetCode );
 186      }
 187  
 188      function instance( $input_charset_code, $output_charset_code )
 189      {
 190          $mb =& $GLOBALS["eZMBStringMapper-$input_charset_code-$output_charset_code"];
 191          if ( get_class( $mb ) != "ezmbstringmapper" )
 192          {
 193              $mb = new eZMBStringMapper( $input_charset_code, $output_charset_code );
 194          }
 195          return $mb;
 196      }
 197  }
 198  
 199  ?>