languages/classes/LanguageEo.php

   1 <?php
   2
   3 /** Esperanto (Esperanto)
   4  *
   5  * @ingroup Language
   6  * @author Brion Vibber <brion@pobox.com>
   7  */
   8 class LanguageEo extends Language {
   9         /**
  10          * Wrapper for charset conversions.
  11          *
  12          * In most languages, this calls through to standard system iconv(), but
  13          * for Esperanto we're also adding a special pseudo-charset to convert
  14          * accented characters to/from the ASCII-friendly "X" surrogate coding:
  15          *
  16          *     cx = ĉ     cxx = cx
  17          *     gx = ĝ     gxx = gx
  18          *     hx = ĥ     hxx = hx
  19          *     jx = ĵ     jxx = jx
  20          *     sx = ŝ     sxx = sx
  21          *     ux = ŭ     uxx = ux
  22          *     xx = x
  23          *
  24          *   http://en.wikipedia.org/wiki/Esperanto_orthography#X-system
  25          *   http://eo.wikipedia.org/wiki/X-sistemo
  26          *
  27          * X-conversion is applied, in either direction, between "utf-8" and "x" charsets;
  28          * this comes into effect when input is run through $wgRequest->getText() and the
  29          * $wgEditEncoding is set to 'x'.
  30          *
  31          * In the long run, this should be moved out of here and into the client-side
  32          * editor behavior; the original server-side translation system dates to 2002-2003
  33          * when many browsers with really bad Unicode support were still in use.
  34          *
  35          * @param string $in input character set
  36          * @param string $out output character set
  37          * @param string $string text to be converted
  38          * @return string
  39          */
  40         function iconv( $in, $out, $string ) {
  41                 if ( strcasecmp( $in, 'x' ) == 0 && strcasecmp( $out, 'utf-8' ) == 0 ) {
  42                         return preg_replace_callback (
  43                                 '/([cghjsu]x?)((?:xx)*)(?!x)/i',
  44                                 array( $this, 'strrtxuCallback' ), $string      );
  45                 } else if ( strcasecmp( $in, 'UTF-8' ) == 0 && strcasecmp( $out, 'x' ) == 0 ) {
  46                         # Double Xs only if they follow cxapelutaj literoj.
  47                         return preg_replace_callback(
  48                                 '/((?:[cghjsu]|\xc4[\x88\x89\x9c\x9d\xa4\xa5\xb4\xb5]|\xc5[\x9c\x9d\xac\xad])x*)/i',
  49                                 array( $this, 'strrtuxCallback' ), $string );
  50                 }
  51                 return parent::iconv( $in, $out, $string );
  52         }
  53
  54         function strrtuxCallback( $matches ) {
  55                 static $ux = array (
  56                         'x' => 'xx' , 'X' => 'Xx' ,
  57                         "\xc4\x88" => "Cx" , "\xc4\x89" => "cx" ,
  58                         "\xc4\x9c" => "Gx" , "\xc4\x9d" => "gx" ,
  59                         "\xc4\xa4" => "Hx" , "\xc4\xa5" => "hx" ,
  60                         "\xc4\xb4" => "Jx" , "\xc4\xb5" => "jx" ,
  61                         "\xc5\x9c" => "Sx" , "\xc5\x9d" => "sx" ,
  62                         "\xc5\xac" => "Ux" , "\xc5\xad" => "ux"
  63                 );
  64                 return strtr( $matches[1], $ux );
  65         }
  66
  67         function strrtxuCallback( $matches ) {
  68                 static $xu = array (
  69                         'xx' => 'x' , 'xX' => 'x' ,
  70                         'Xx' => 'X' , 'XX' => 'X' ,
  71                         "Cx" => "\xc4\x88" , "CX" => "\xc4\x88" ,
  72                         "cx" => "\xc4\x89" , "cX" => "\xc4\x89" ,
  73                         "Gx" => "\xc4\x9c" , "GX" => "\xc4\x9c" ,
  74                         "gx" => "\xc4\x9d" , "gX" => "\xc4\x9d" ,
  75                         "Hx" => "\xc4\xa4" , "HX" => "\xc4\xa4" ,
  76                         "hx" => "\xc4\xa5" , "hX" => "\xc4\xa5" ,
  77                         "Jx" => "\xc4\xb4" , "JX" => "\xc4\xb4" ,
  78                         "jx" => "\xc4\xb5" , "jX" => "\xc4\xb5" ,
  79                         "Sx" => "\xc5\x9c" , "SX" => "\xc5\x9c" ,
  80                         "sx" => "\xc5\x9d" , "sX" => "\xc5\x9d" ,
  81                         "Ux" => "\xc5\xac" , "UX" => "\xc5\xac" ,
  82                         "ux" => "\xc5\xad" , "uX" => "\xc5\xad"
  83                 );
  84                 return strtr( $matches[1], $xu ) . strtr( $matches[2], $xu );
  85         }
  86
  87         function checkTitleEncoding( $s ) {
  88                 # Check for X-system backwards-compatibility URLs
  89                 $ishigh = preg_match( '/[\x80-\xff]/', $s );
  90                 $isutf = preg_match( '/^([\x00-\x7f]|[\xc0-\xdf][\x80-\xbf]|' .
  91                         '[\xe0-\xef][\x80-\xbf]{2}|[\xf0-\xf7][\x80-\xbf]{3})+$/', $s );
  92
  93                 if ( $ishigh and !$isutf ) {
  94                         # Assume Latin1
  95                         $s = utf8_encode( $s );
  96                 } else {
  97                         if ( preg_match( '/(\xc4[\x88\x89\x9c\x9d\xa4\xa5\xb4\xb5]' .
  98                                 '|\xc5[\x9c\x9d\xac\xad])/', $s ) )
  99                         return $s;
 100                 }
 101
 102                 // if( preg_match( '/[cghjsu]x/i', $s ) )
 103                 //      return $this->iconv( 'x', 'utf-8', $s );
 104                 return $s;
 105         }
 106
 107         function initEncoding() {
 108                 global $wgEditEncoding;
 109                 $wgEditEncoding = 'x';
 110         }
 111 }