stub slovene file; start consolidating duplicate checkTitleEncoding funcs

diff --git a/includes/Utf8Case.php b/includes/Utf8Case.php
index 1f0208e..bcf4eeb 100644 (file)
--- a/includes/Utf8Case.php
+++ b/includes/Utf8Case.php
@@ -6,6 +6,7 @@ $wgOutputEncoding       = "utf-8";
 # Won't get context-sensitive things yet
 
 # Hack for bugs in ucfirst() and company
+# TODO: store this in shared memory or something
 
 $wikiUpperChars = array(
        "a" => "A",
@@ -1496,11 +1497,11 @@ $wikiLowerChars = array (
 # Base stuff useful to all UTF-8 based language files
 class LanguageUtf8 extends Language {
 
-    function ucfirst( $string ) {
+       function ucfirst( $string ) {
                # For most languages, this is a wrapper for ucfirst()
                # But that doesn't work right in a UTF-8 locale
                global $wikiUpperChars, $wikiLowerChars;
-        return preg_replace (
+               return preg_replace (
                "/^([\\x00-\\x7f]|[\\xc0-\\xff][\\x80-\\xbf]*)/e",
                "strtr ( \"\$1\" , \$wikiUpperChars )",
                $string );
@@ -1515,6 +1516,27 @@ class LanguageUtf8 extends Language {
                  "'U8' . bin2hex( strtr( \"\$1\", \$wikiLowerChars ) )",
                  $string );
        }
+
+       function fallback8bitEncoding() {
+               # Windows codepage 1252 is a superset of iso 8859-1
+               # override this to use difference source encoding to
+               # translate incoming 8-bit URLs.
+               return "windows-1252";
+       }
+
+       function checkTitleEncoding( $s ) {
+               global $wgInputEncoding;
+
+               # Check for non-UTF-8 URLs
+               $ishigh = preg_match( '/[\x80-\xff]/', $s);
+               if(!$ishigh) return $s;
+               
+               $isutf8 = preg_match( '/^([\x00-\x7f]|[\xc0-\xdf][\x80-\xbf]|' .
+                '[\xe0-\xef][\x80-\xbf]{2}|[\xf0-\xf7][\x80-\xbf]{3})+$/', $s );
+               if( $isutf8 ) return $s;
+
+               return $this->iconv( $this->fallback8bitEncoding(), "utf-8", $s );
+       }
 }
 
 ?>
\ No newline at end of file

diff --git a/languages/LanguageSl.php b/languages/LanguageSl.php
new file mode 100644 (file)
index 0000000..e66a545
--- /dev/null
+++ b/languages/LanguageSl.php
@@ -0,0 +1,14 @@
+<?
+# See language.doc
+global $IP;
+include_once("$IP/Utf8Case.php");
+
+class LanguageSl extends LanguageUtf8 {
+       # Inherit everything
+
+       function fallback8bitEncoding() {
+               return "windows-1250"; /* or iso 8859-2? */
+       }
+}
+
+?>