Updated language detection code

git-svn-id: https://oss.gonicus.de/repositories/gosa/branches/2.5@7034 594d385d-05f5-0310-b6e9-bd551577e9d8

diff --git a/AUTHORS b/AUTHORS
index 2ee3e3749795920514c1655c8d0fab04bf5a0f27..8b81b67fbd581b498b4d281435cea971dfe183b3 100644 (file)
--- a/AUTHORS
+++ b/AUTHORS
@@ -74,3 +74,6 @@ documentation and additional help.
 * Vincent Seynhaeve
   Xls export plugin <vincent.seynhaeve@opensides.be>
   
+* Wouter Verhelst <wouter@debian.org>
+  accept-to-gettext code that helps for language conversation
+

diff --git a/Changelog b/Changelog
index 782eee341277bc1c7331599480847b4e9a546c1c..fcd2668e970f8d479579003509ae05674d4d1020 100644 (file)
--- a/Changelog
+++ b/Changelog
@@ -9,7 +9,7 @@ GOsa2 changelog
   - Added copy'n paste for mimetypes
   - Cut'n paste objects are now greyed out
   - Added swedish locale
-  - Unified language code
+  - Improved language detection
   - Added a statistic footer to lists
   - Added the ssh plugin
   - Layout fixes

diff --git a/html/index.php b/html/index.php
index 5a4f2128ace55ec478da35ddf9bf0fb293c0283c..fcd6130849d44c0aceb65e1dc4faa71bcb85afb1 100644 (file)
--- a/html/index.php
+++ b/html/index.php
@@ -152,7 +152,6 @@ if ($config->data['MAIN']['LANG'] == ""){
 } else {
   $lang= $config->data['MAIN']['LANG'];
 }
-$lang.=".UTF-8";
 putenv("LANGUAGE=");
 putenv("LANG=$lang");
 setlocale(LC_ALL, $lang);

diff --git a/html/main.php b/html/main.php
index dc952df2e8348b6cb121647799b9c7cc16e88229..0093a965321d4b746ee549820a870b5c127ca0bf 100644 (file)
--- a/html/main.php
+++ b/html/main.php
@@ -135,8 +135,6 @@ if ($config->data['MAIN']['LANG'] == ""){
   $lang= $config->data['MAIN']['LANG'];
 }
 
-$lang.=".UTF-8";
-
 putenv("LANGUAGE=");
 putenv("LANG=$lang");
 setlocale(LC_ALL, $lang);

diff --git a/html/setup.php b/html/setup.php
index 177b38216d6c6c21ea12a724c0c58e5ff9b9c497..297e073165f041e18282faf9d89b3de35680406c 100644 (file)
--- a/html/setup.php
+++ b/html/setup.php
@@ -86,7 +86,6 @@ if(isset($_POST['lang_selected'])){
 }
 
 $lang = $_SESSION['lang'];
-$lang.=".UTF-8";
 putenv("LANGUAGE=");
 putenv("LANG=$lang");
 setlocale(LC_ALL, $lang);

diff --git a/include/accept-to-gettext.inc b/include/accept-to-gettext.inc
new file mode 100644 (file)
index 0000000..f04bb4f
--- /dev/null
+++ b/include/accept-to-gettext.inc
@@ -0,0 +1,188 @@
+<?php
+/*
+ * accept-to-gettext.inc -- convert information in 'Accept-*' headers to
+ * gettext language identifiers.
+ * Copyright (c) 2003, Wouter Verhelst <wouter@debian.org>
+ * 
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
+ *
+ * Usage:
+ *
+ *  $locale=al2gt(<array of supported languages/charsets in gettext syntax>,
+ *                <MIME type of document>);
+ *  setlocale('LC_ALL', $locale); // or 'LC_MESSAGES', or whatever...
+ *
+ * Example:
+ *
+ *  $langs=array('nl_BE.ISO-8859-15','nl_BE.UTF-8','en_US.UTF-8','en_GB.UTF-8');
+ *  $locale=al2gt($langs, 'text/html');
+ *  setlocale('LC_ALL', $locale);
+ *
+ * Note that this will send out header information (to be
+ * RFC2616-compliant), so it must be called before anything is sent to
+ * the user.
+ * 
+ * Assumptions made:
+ * * Charset encodings are written the same way as the Accept-Charset
+ *   HTTP header specifies them (RFC2616), except that they're parsed
+ *   case-insensitive.
+ * * Country codes and language codes are the same in both gettext and
+ *   the Accept-Language syntax (except for the case differences, which
+ *   are dealt with easily). If not, some input may be ignored.
+ * * The provided gettext-strings are fully qualified; i.e., no "en_US";
+ *   always "en_US.ISO-8859-15" or "en_US.UTF-8", or whichever has been
+ *   used. "en.ISO-8859-15" is OK, though.
+ * * The language is more important than the charset; i.e., if the
+ *   following is given:
+ * 
+ *   Accept-Language: nl-be, nl;q=0.8, en-us;q=0.5, en;q=0.3
+ *   Accept-Charset: ISO-8859-15, utf-8;q=0.5
+ *
+ *   And the supplied parameter contains (amongst others) nl_BE.UTF-8
+ *   and nl.ISO-8859-15, then nl_BE.UTF-8 will be picked.
+ * 
+ * $Log: accept-to-gettext.inc,v $
+ * Revision 1.1.1.1  2003/11/19 19:31:15  wouter
+ * * moved to new CVS repo after death of the old
+ * * Fixed code to apply a default to both Accept-Charset and
+ *   Accept-Language if none of those headers are supplied; patch from
+ *   Dominic Chambers <dominic@encasa.com>
+ *
+ * Revision 1.2  2003/08/14 10:23:59  wouter
+ * Removed little error in Content-Type header syntaxis.
+ *
+ */
+
+/* not really important, this one; perhaps I could've put it inline with
+ * the rest. */
+function find_match($curlscore,$curcscore,$curgtlang,$langval,$charval,
+                    $gtlang)
+{
+  if($curlscore < $langval) {
+    $curlscore=$langval;
+    $curcscore=$charval;
+    $curgtlang=$gtlang;
+  } else if ($curlscore == $langval) {
+    if($curcscore < $charval) {
+      $curcscore=$charval;
+      $curgtlang=$gtlang;
+    }
+  }
+  return array($curlscore, $curcscore, $curgtlang);
+}
+
+
+function al2gt($gettextlangs, $mime) {
+  error_reporting(0);
+
+  /* default to "everything is acceptable", as RFC2616 specifies */
+  $acceptLang=(($_SERVER["HTTP_ACCEPT_LANGUAGE"] == '') ? '*' :
+    $_SERVER["HTTP_ACCEPT_LANGUAGE"]);
+  $acceptChar=(($_SERVER["HTTP_ACCEPT_CHARSET"] == '') ? '*' :
+    $_SERVER["HTTP_ACCEPT_CHARSET"]);
+  $alparts=@preg_split("/,/",$acceptLang);
+  $acparts=@preg_split("/,/",$acceptChar);
+  
+  /* Parse the contents of the Accept-Language header.*/
+  foreach($alparts as $part) {
+    $part=trim($part);
+    if(preg_match("/;/", $part)) {
+      $lang=@preg_split("/;/",$part);
+      $score=@preg_split("/=/",$lang[1]);
+      $alscores[$lang[0]]=$score[1];
+    } else {
+      $alscores[$part]=1;
+    }
+  }
+
+  /* Do the same for the Accept-Charset header. */
+
+  /* RFC2616: ``If no "*" is present in an Accept-Charset field, then
+   * all character sets not explicitly mentioned get a quality value of
+   * 0, except for ISO-8859-1, which gets a quality value of 1 if not
+   * explicitly mentioned.''
+   * 
+   * Making it 2 for the time being, so that we
+   * can distinguish between "not specified" and "specified as 1" later
+   * on. */
+  $acscores["ISO-8859-1"]=2;
+
+  foreach($acparts as $part) {
+    $part=trim($part);
+    if(preg_match("/;/", $part)) {
+      $cs=@preg_split("/;/",$part);
+      $score=@preg_split("/=/",$cs[1]);
+      $acscores[strtoupper($cs[0])]=$score[1];
+    } else {
+      $acscores[strtoupper($part)]=1;
+    }
+  }
+  if($acscores["ISO-8859-1"]==2) {
+    $acscores["ISO-8859-1"]=(isset($acscores["*"])?$acscores["*"]:1);
+  }
+
+  /* 
+   * Loop through the available languages/encodings, and pick the one
+   * with the highest score, excluding the ones with a charset the user
+   * did not include.
+   */
+  $curlscore=0;
+  $curcscore=0;
+  $curgtlang=NULL;
+  foreach($gettextlangs as $gtlang) {
+
+    $tmp1=preg_replace("/\_/","-",$gtlang);
+    $tmp2=@preg_split("/\./",$tmp1);
+    $allang=strtolower($tmp2[0]);
+    $gtcs=strtoupper($tmp2[1]);
+    $noct=@preg_split("/-/",$allang);
+
+    $testvals=array(
+        array($alscores[$allang], $acscores[$gtcs]),
+        array($alscores[$noct[0]], $acscores[$gtcs]),
+        array($alscores[$allang], $acscores["*"]),
+        array($alscores[$noct[0]], $acscores["*"]),
+        array($alscores["*"], $acscores[$gtcs]),
+        array($alscores["*"], $acscores["*"]));
+
+    $found=FALSE;
+    foreach($testvals as $tval) {
+      if(!$found && isset($tval[0]) && isset($tval[1])) {
+        $arr=find_match($curlscore, $curcscore, $curgtlang, $tval[0],
+            $tval[1], $gtlang);
+        $curlscore=$arr[0];
+        $curcscore=$arr[1];
+        $curgtlang=$arr[2];
+        $found=TRUE;
+      }
+    }
+  }
+
+  /* We must re-parse the gettext-string now, since we may have found it
+   * through a "*" qualifier.*/
+  
+  $gtparts=@preg_split("/\./",$curgtlang);
+  $tmp=strtolower($gtparts[0]);
+  $lang=preg_replace("/\_/", "-", $tmp);
+  $charset=$gtparts[1];
+
+  header("Content-Language: $lang");
+  header("Content-Type: $mime; charset=$charset");
+
+  error_reporting(E_ALL);
+  return $curgtlang;
+}
+
+?>

diff --git a/include/functions.inc b/include/functions.inc
index e9e3622ac726f45629a4e2cd2bcef4c907f4784a..be17de81fe9af4635bfeefddf00f4d266d4256bb 100644 (file)
--- a/include/functions.inc
+++ b/include/functions.inc
@@ -45,6 +45,7 @@ require_once ("class_mail-methods.inc");
 require_once("class_password-methods.inc");
 require_once ("functions_debug.inc");
 require_once ("functions_dns.inc");
+require_once ("accept-to-gettext.inc");
 require_once ("class_MultiSelectWindow.inc");
 
 /* Define constants for debugging */
@@ -146,54 +147,19 @@ function DEBUG($level, $line, $function, $file, $data, $info="")
 }
 
 
-/* Simple function to get browser language and convert it to
-   xx_XY needed by locales. Ignores sublanguages and weights. */
 function get_browser_language()
 {
-  global $BASE_DIR;
+  /* Load supported languages */
+  $gosa_languages= get_languages();
 
-  /* Try to use users primary language */
-  $ui= get_userinfo();
-  if ($ui != NULL){
-    if ($ui->language != ""){
-      return ($ui->language);
-    }
-  }
-
-  /* Get list of languages */
-  if (isset($_SERVER['HTTP_ACCEPT_LANGUAGE'])){
-    $lang= preg_replace("/\s+/", "", $_SERVER['HTTP_ACCEPT_LANGUAGE']);
-    $languages= split (',', $lang);
-    $languages[]= "C";
-  } else {
-    $languages= array("C");
-  }
-
-  /* Walk through languages and get first supported */
-  foreach ($languages as $val){
-
-    /* Strip off weight */
-    $lang= preg_replace("/;q=.*$/i", "", $val);
-
-    /* Simplify sub language handling */
-    $lang= preg_replace("/-.*$/", "", $lang);
-
-    /* Cancel loop if available in GOsa, or the last
-       entry has been reached */
-    if (is_dir("$BASE_DIR/locale/$lang")){
-      break;
-    }
-  }
-
-  /* We've just one zh variation. Fix code... */
-  if (preg_match('/zh/', $lang)){
-    return ("zh_CN");
-  }
-  if (preg_match('/sv/', $lang)){
-    return ("sv_SE");
+  /* Move supported languages to flat list */
+  $langs= array();
+  foreach($gosa_languages as $lang => $dummy){
+    $langs[]= $lang.'.UTF-8';
   }
 
-  return (strtolower($lang)."_".strtoupper($lang));
+  /* Return gettext based string */
+  return (al2gt($langs, 'text/html'));
 }