languages/LanguageUtf8.php

   1 <?php
   2 #$Id$
   3 if( defined( "MEDIAWIKI" ) ) {
   4
   5 $wgInputEncoding    = "UTF-8";
   6 $wgOutputEncoding       = "UTF-8";
   7
   8 if (function_exists('mb_internal_encoding')) {
   9         mb_internal_encoding('UTF-8');
  10 } else {
  11         # Hack our own case conversion routines
  12
  13         # Loading serialized arrays is faster than parsing code :P
  14         $wikiUpperChars = $wgMemc->get( $key1 = "$wgDBname:utf8:upper" );
  15         $wikiLowerChars = $wgMemc->get( $key2 = "$wgDBname:utf8:lower" );
  16
  17         if(empty( $wikiUpperChars) || empty($wikiLowerChars )) {
  18                 require_once( "includes/Utf8Case.php" );
  19                 $wgMemc->set( $key1, $wikiUpperChars );
  20                 $wgMemc->set( $key2, $wikiLowerChars );
  21         }
  22 }
  23
  24 # Base stuff useful to all UTF-8 based language files
  25 class LanguageUtf8 extends Language {
  26
  27         # These two functions use mbstring library, if it is loaded
  28         # or compiled and character mapping arrays otherwise.
  29         # In case of language-specific character mismatch
  30         # it should be dealt with in Language classes.
  31
  32         function ucfirst( $string ) {
  33                 if (function_exists('mb_strtoupper')) {
  34                         return mb_strtoupper(mb_substr($string,0,1)).mb_substr($string,1);
  35                 } else {
  36                     global $wikiUpperChars;
  37                     return preg_replace (
  38                     "/^([a-z]|[\\xc0-\\xff][\\x80-\\xbf]*)/e",
  39                     "strtr ( \"\$1\" , \$wikiUpperChars )",
  40                     $string );
  41                 }
  42         }
  43
  44         function lcfirst( $string ) {
  45                 if (function_exists('mb_strtolower')) {
  46                         return mb_strtolower(mb_substr($string,0,1)).mb_substr($string,1);
  47                 } else {
  48                     global $wikiLowerChars;
  49                     return preg_replace (
  50                     "/^([A-Z]|[\\xc0-\\xff][\\x80-\\xbf]*)/e",
  51                     "strtr ( \"\$1\" , \$wikiLowerChars )",
  52                     $string );
  53                 }
  54         }
  55
  56         function stripForSearch( $string ) {
  57                 # MySQL fulltext index doesn't grok utf-8, so we
  58                 # need to fold cases and convert to hex
  59
  60                 # In Language:: it just returns lowercase, maybe
  61                 # all strtolower on stripped output or argument
  62                 # should be removed and all stripForSearch
  63                 # methods adjusted to that.
  64
  65                 wfProfileIn( "LanguageUtf8::stripForSearch" );
  66                 if( function_exists( 'mb_strtolower' ) ) {
  67                         $out = preg_replace(
  68                                 "/([\\xc0-\\xff][\\x80-\\xbf]*)/e",
  69                                 "'U8' . bin2hex( \"$1\" )",
  70                                 mb_strtolower( $string ) );
  71                 } else {
  72                         global $wikiLowerChars;
  73                         $out = preg_replace(
  74                                 "/([\\xc0-\\xff][\\x80-\\xbf]*)/e",
  75                                 "'U8' . bin2hex( strtr( \"\$1\", \$wikiLowerChars ) )",
  76                                 $string );
  77                 }
  78                 wfProfileOut( "LanguageUtf8::stripForSearch" );
  79                 return $out;
  80         }
  81
  82         function fallback8bitEncoding() {
  83                 # Windows codepage 1252 is a superset of iso 8859-1
  84                 # override this to use difference source encoding to
  85                 # translate incoming 8-bit URLs.
  86                 return "windows-1252";
  87         }
  88
  89         function checkTitleEncoding( $s ) {
  90                 global $wgInputEncoding;
  91
  92                 # Check for non-UTF-8 URLs
  93                 $ishigh = preg_match( '/[\x80-\xff]/', $s);
  94                 if(!$ishigh) return $s;
  95
  96                 $isutf8 = preg_match( '/^([\x00-\x7f]|[\xc0-\xdf][\x80-\xbf]|' .
  97                 '[\xe0-\xef][\x80-\xbf]{2}|[\xf0-\xf7][\x80-\xbf]{3})+$/', $s );
  98                 if( $isutf8 ) return $s;
  99
 100                 return $this->iconv( $this->fallback8bitEncoding(), "utf-8", $s );
 101         }
 102
 103         function firstChar( $s ) {
 104                 preg_match( '/^([\x00-\x7f]|[\xc0-\xdf][\x80-\xbf]|' .
 105                 '[\xe0-\xef][\x80-\xbf]{2}|[\xf0-\xf7][\x80-\xbf]{3})/', $s, $matches);
 106
 107                 return isset( $matches[1] ) ? $matches[1] : "";
 108         }
 109
 110         # Crop a string from the beginning or end to a certain number of bytes.
 111         # (Bytes are used because our storage has limited byte lengths for some
 112         # columns in the database.) Multibyte charsets will need to make sure that
 113         # only whole characters are included!
 114         #
 115         # $length does not include the optional ellipsis.
 116         # If $length is negative, snip from the beginning
 117         function truncate( $string, $length, $ellipsis = "" ) {
 118                 if( $length == 0 ) {
 119                         return $ellipsis;
 120                 }
 121                 if ( strlen( $string ) <= abs( $length ) ) {
 122                         return $string;
 123                 }
 124                 if( $length > 0 ) {
 125                         $string = substr( $string, 0, $length );
 126                         $char = ord( $string[strlen( $string ) - 1] );
 127                         if ($char >= 0xc0) {
 128                                 # We got the first byte only of a multibyte char; remove it.
 129                                 $string = substr( $string, 0, -1 );
 130                         } elseif( $char >= 0x80 &&
 131                                   preg_match( '/^(.*)(?:[\xe0-\xef][\x80-\xbf]|' .
 132                                               '[\xf0-\xf7][\x80-\xbf]{1,2})$/', $string, $m ) ) {
 133                             # We chopped in the middle of a character; remove it
 134                                 $string = $m[1];
 135                         }
 136                         return $string . $ellipsis;
 137                 } else {
 138                         $string = substr( $string, $length );
 139                         $char = ord( $string[0] );
 140                         if( $char >= 0x80 && $char < 0xc0 ) {
 141                                 # We chopped in the middle of a character; remove the whole thing
 142                                 $string = preg_replace( '/^[\x80-\xbf]+/', '', $string );
 143                         }
 144                         return $ellipsis . $string;
 145                 }
 146         }
 147 }
 148
 149 } # ifdef MEDIAWIKI
 150
 151 ?>