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 */
68 /* not really important, this one; perhaps I could've put it inline with
69 * the rest. */
70 function find_match($curlscore,$curcscore,$curgtlang,$langval,$charval,
71 $gtlang)
72 {
73 if($curlscore < $langval) {
74 $curlscore=$langval;
75 $curcscore=$charval;
76 $curgtlang=$gtlang;
77 } else if ($curlscore == $langval) {
78 if($curcscore < $charval) {
79 $curcscore=$charval;
80 $curgtlang=$gtlang;
81 }
82 }
83 return array($curlscore, $curcscore, $curgtlang);
84 }
87 function al2gt($gettextlangs, $mime)
88 {
89 /* Check if ACCEPT_LANGUAGE isset */
90 if(!isset($_SERVER["HTTP_ACCEPT_LANGUAGE"])){
91 $_SERVER["HTTP_ACCEPT_LANGUAGE"] = "";
92 }
93 if(!isset($_SERVER["HTTP_ACCEPT_CHARSET"])){
94 $_SERVER["HTTP_ACCEPT_CHARSET"] = "";
95 }
97 /* default to "everything is acceptable", as RFC2616 specifies */
98 $acceptLang=(($_SERVER["HTTP_ACCEPT_LANGUAGE"] == '') ? '*' :
99 $_SERVER["HTTP_ACCEPT_LANGUAGE"]);
100 /* Commented out due to problems with IE7, defaulting to the one below... */
101 #$acceptChar=(($_SERVER["HTTP_ACCEPT_CHARSET"] == '') ? '*' :
102 # $_SERVER["HTTP_ACCEPT_CHARSET"]);
103 $acceptChar=(($_SERVER["HTTP_ACCEPT_CHARSET"] == '') ? 'ISO-8859-1,utf-8;q=0.7,*;q=0.7' :
104 $_SERVER["HTTP_ACCEPT_CHARSET"]);
105 $alparts=@preg_split("/,/",$acceptLang);
106 $acparts=@preg_split("/,/",$acceptChar);
108 /* Parse the contents of the Accept-Language header.*/
109 foreach($alparts as $part) {
110 $part=trim($part);
111 if(preg_match("/;/", $part)) {
112 $lang=@preg_split("/;/",$part);
113 $score=@preg_split("/=/",$lang[1]);
114 $alscores[$lang[0]]=$score[1];
115 } else {
116 $alscores[$part]=1;
117 }
118 }
120 /* Do the same for the Accept-Charset header. */
122 /* RFC2616: ``If no "*" is present in an Accept-Charset field, then
123 * all character sets not explicitly mentioned get a quality value of
124 * 0, except for ISO-8859-1, which gets a quality value of 1 if not
125 * explicitly mentioned.''
126 *
127 * Making it 2 for the time being, so that we
128 * can distinguish between "not specified" and "specified as 1" later
129 * on. */
130 $acscores["ISO-8859-1"]=2;
132 foreach($acparts as $part) {
133 $part=trim($part);
134 if(preg_match("/;/", $part)) {
135 $cs=@preg_split("/;/",$part);
136 $score=@preg_split("/=/",$cs[1]);
137 $acscores[strtoupper($cs[0])]=$score[1];
138 } else {
139 $acscores[strtoupper($part)]=1;
140 }
141 }
142 if($acscores["ISO-8859-1"]==2) {
143 $acscores["ISO-8859-1"]=(isset($acscores["*"])?$acscores["*"]:1);
144 }
146 /*
147 * Loop through the available languages/encodings, and pick the one
148 * with the highest score, excluding the ones with a charset the user
149 * did not include.
150 */
151 $curlscore=0;
152 $curcscore=0;
153 $curgtlang=NULL;
154 foreach($gettextlangs as $gtlang) {
156 $tmp1=preg_replace("/\_/","-",$gtlang);
157 $tmp2=@preg_split("/\./",$tmp1);
158 $allang=strtolower($tmp2[0]);
159 $gtcs=strtoupper($tmp2[1]);
160 $noct=@preg_split("/-/",$allang);
162 if(!isset($alscores["*"])){
163 $alscores["*"] = "";
164 }
166 if(!isset($alscores[$allang])){
167 $alscores[$allang] = "";
168 }
170 if(!isset($alscores[$noct[0]])){
171 $alscores[$noct[0]] = "";
172 }
174 if(!isset($acscores[$gtcs])){
175 $acscores[$gtcs] = "";
176 }
177 $testvals=array(
178 array($alscores[$allang], $acscores[$gtcs]),
179 array($alscores[$noct[0]], $acscores[$gtcs]),
180 array($alscores[$allang], $acscores["*"]),
181 array($alscores[$noct[0]], $acscores["*"]),
182 array($alscores["*"], $acscores[$gtcs]),
183 array($alscores["*"], $acscores["*"]));
185 $found=FALSE;
186 foreach($testvals as $tval) {
187 if(!$found && isset($tval[0]) && isset($tval[1])) {
188 $arr=find_match($curlscore, $curcscore, $curgtlang, $tval[0],
189 $tval[1], $gtlang);
190 $curlscore=$arr[0];
191 $curcscore=$arr[1];
192 $curgtlang=$arr[2];
193 $found=TRUE;
194 }
195 }
196 }
198 /* We must re-parse the gettext-string now, since we may have found it
199 * through a "*" qualifier.*/
200 $gtparts=@preg_split("/\./",$curgtlang);
201 $tmp=strtolower($gtparts[0]);
202 $lang=preg_replace("/\_/", "-", $tmp);
203 header("Content-Language: $lang");
204 if(isset($gtparts[1])){
205 $charset=$gtparts[1];
206 header("Content-Type: $mime; charset=$charset");
207 }
208 return $curgtlang;
209 }
211 // vim:tabstop=2:expandtab:shiftwidth=2:filetype=php:syntax:ruler:
212 ?>