lib/accept-to-gettext.php

   1 <?php
   2 /*
   3  * accept-to-gettext.inc -- convert information in 'Accept-*' headers to
   4  * gettext language identifiers.
   5  * Copyright (c) 2003, Wouter Verhelst <wouter@debian.org>
   6  *
   7  * This program is free software; you can redistribute it and/or modify
   8  * it under the terms of the GNU General Public License as published by
   9  * the Free Software Foundation; either version 2 of the License, or
  10  * (at your option) any later version.
  11  *
  12  * This program is distributed in the hope that it will be useful,
  13  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  14  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  15  * GNU General Public License for more details.
  16  *
  17  * You should have received a copy of the GNU General Public License
  18  * along with this program; if not, write to the Free Software
  19  * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
  20  *
  21  * Usage:
  22  *
  23  *  $locale=al2gt(<array of supported languages/charsets in gettext syntax>,
  24  *                <MIME type of document>);
  25  *  setlocale('LC_ALL', $locale); // or 'LC_MESSAGES', or whatever...
  26  *
  27  * Example:
  28  *
  29  *  $langs=array('nl_BE.ISO-8859-15','nl_BE.UTF-8','en_US.UTF-8','en_GB.UTF-8');
  30  *  $locale=al2gt($langs, 'text/html');
  31  *  setlocale('LC_ALL', $locale);
  32  *
  33  * Note that this will send out header information (to be
  34  * RFC2616-compliant), so it must be called before anything is sent to
  35  * the user.
  36  *
  37  * Assumptions made:
  38  * * Charset encodings are written the same way as the Accept-Charset
  39  *   HTTP header specifies them (RFC2616), except that they're parsed
  40  *   case-insensitive.
  41  * * Country codes and language codes are the same in both gettext and
  42  *   the Accept-Language syntax (except for the case differences, which
  43  *   are dealt with easily). If not, some input may be ignored.
  44  * * The provided gettext-strings are fully qualified; i.e., no "en_US";
  45  *   always "en_US.ISO-8859-15" or "en_US.UTF-8", or whichever has been
  46  *   used. "en.ISO-8859-15" is OK, though.
  47  * * The language is more important than the charset; i.e., if the
  48  *   following is given:
  49  *
  50  *   Accept-Language: nl-be, nl;q=0.8, en-us;q=0.5, en;q=0.3
  51  *   Accept-Charset: ISO-8859-15, utf-8;q=0.5
  52  *
  53  *   And the supplied parameter contains (amongst others) nl_BE.UTF-8
  54  *   and nl.ISO-8859-15, then nl_BE.UTF-8 will be picked.
  55  *
  56  * $Log: accept-to-gettext.inc,v $
  57  * Revision 1.1.1.1  2003/11/19 19:31:15  wouter
  58  * * moved to new CVS repo after death of the old
  59  * * Fixed code to apply a default to both Accept-Charset and
  60  *   Accept-Language if none of those headers are supplied; patch from
  61  *   Dominic Chambers <dominic@encasa.com>
  62  *
  63  * Revision 1.2  2003/08/14 10:23:59  wouter
  64  * Removed little error in Content-Type header syntaxis.
  65  *
  66  * 2007-04-01
  67  * add '@' before use of arrays, to avoid PHP warnings.
  68  */
  69
  70 /* not really important, this one; perhaps I could've put it inline with
  71  * the rest. */
  72 function find_match($curlscore,$curcscore,$curgtlang,$langval,$charval,
  73                     $gtlang)
  74 {
  75   if($curlscore < $langval) {
  76     $curlscore=$langval;
  77     $curcscore=$charval;
  78     $curgtlang=$gtlang;
  79   } else if ($curlscore == $langval) {
  80     if($curcscore < $charval) {
  81       $curcscore=$charval;
  82       $curgtlang=$gtlang;
  83     }
  84   }
  85   return array($curlscore, $curcscore, $curgtlang);
  86 }
  87
  88 function al2gt($gettextlangs, $mime) {
  89   /* default to "everything is acceptable", as RFC2616 specifies */
  90   $acceptLang=(($_SERVER["HTTP_ACCEPT_LANGUAGE"] == '') ? '*' :
  91       $_SERVER["HTTP_ACCEPT_LANGUAGE"]);
  92   $acceptChar=(($_SERVER["HTTP_ACCEPT_CHARSET"] == '') ? '*' :
  93       $_SERVER["HTTP_ACCEPT_CHARSET"]);
  94   $alparts=@preg_split("/,/",$acceptLang);
  95   $acparts=@preg_split("/,/",$acceptChar);
  96
  97   /* Parse the contents of the Accept-Language header.*/
  98   foreach($alparts as $part) {
  99     $part=trim($part);
 100     if(preg_match("/;/", $part)) {
 101       $lang=@preg_split("/;/",$part);
 102       $score=@preg_split("/=/",$lang[1]);
 103       $alscores[$lang[0]]=$score[1];
 104     } else {
 105       $alscores[$part]=1;
 106     }
 107   }
 108
 109   /* Do the same for the Accept-Charset header. */
 110
 111   /* RFC2616: ``If no "*" is present in an Accept-Charset field, then
 112    * all character sets not explicitly mentioned get a quality value of
 113    * 0, except for ISO-8859-1, which gets a quality value of 1 if not
 114    * explicitly mentioned.''
 115    *
 116    * Making it 2 for the time being, so that we
 117    * can distinguish between "not specified" and "specified as 1" later
 118    * on. */
 119   $acscores["ISO-8859-1"]=2;
 120
 121   foreach($acparts as $part) {
 122     $part=trim($part);
 123     if(preg_match("/;/", $part)) {
 124       $cs=@preg_split("/;/",$part);
 125       $score=@preg_split("/=/",$cs[1]);
 126       $acscores[strtoupper($cs[0])]=$score[1];
 127     } else {
 128       $acscores[strtoupper($part)]=1;
 129     }
 130   }
 131   if($acscores["ISO-8859-1"]==2) {
 132     $acscores["ISO-8859-1"]=(isset($acscores["*"])?$acscores["*"]:1);
 133   }
 134
 135   /*
 136    * Loop through the available languages/encodings, and pick the one
 137    * with the highest score, excluding the ones with a charset the user
 138    * did not include.
 139    */
 140   $curlscore=0;
 141   $curcscore=0;
 142   $curgtlang=NULL;
 143   foreach($gettextlangs as $gtlang) {
 144
 145     $tmp1=preg_replace("/\_/","-",$gtlang);
 146     $tmp2=@preg_split("/\./",$tmp1);
 147     $allang=strtolower($tmp2[0]);
 148     $gtcs=strtoupper($tmp2[1]);
 149     $noct=@preg_split("/-/",$allang);
 150
 151     $testvals=array(
 152          array(@$alscores[$allang], @$acscores[$gtcs]),
 153      array(@$alscores[$noct[0]], @$acscores[$gtcs]),
 154      array(@$alscores[$allang], @$acscores["*"]),
 155      array(@$alscores[$noct[0]], @$acscores["*"]),
 156      array(@$alscores["*"], @$acscores[$gtcs]),
 157      array(@$alscores["*"], @$acscores["*"]));
 158
 159     $found=FALSE;
 160     foreach($testvals as $tval) {
 161       if(!$found && isset($tval[0]) && isset($tval[1])) {
 162         $arr=find_match($curlscore, $curcscore, $curgtlang, $tval[0],
 163               $tval[1], $gtlang);
 164         $curlscore=$arr[0];
 165         $curcscore=$arr[1];
 166         $curgtlang=$arr[2];
 167     $found=TRUE;
 168       }
 169     }
 170   }
 171
 172   /* We must re-parse the gettext-string now, since we may have found it
 173    * through a "*" qualifier.*/
 174
 175   $gtparts=@preg_split("/\./",$curgtlang);
 176   $tmp=strtolower($gtparts[0]);
 177   $lang=preg_replace("/\_/", "-", $tmp);
 178   $charset=$gtparts[1];
 179
 180   header("Content-Language: $lang");
 181   header("Content-Type: $mime; charset=$charset");
 182
 183   return $curgtlang;
 184 }
 185
 186 ?>