Polish up r45109 (changes to header id generation)

1) Make sure internal links are using the new encoding scheme and still
work (this also reduces code duplication)

2) Make sure no id can start with a nonletter, not just header id's

No parser test regressions.

diff --git a/RELEASE-NOTES b/RELEASE-NOTES
index 62cd3b6..8f3d130 100644 (file)
--- a/RELEASE-NOTES
+++ b/RELEASE-NOTES
@@ -457,7 +457,7 @@ The following extensions are migrated into MediaWiki 1.14:
 * (bug 13835) Fix rendering of {{filepath:Wiki.png|nowiki}}
 * (bug 16772) Special:Upload now correctly rejects files with spaces in the
   file extension (e.g. Foo. jpg).
-* Ensure that all HTML id's generated for headers begin with a letter.
+* Ensure that all HTML id's begin with a letter as required by XHTML standard
 
 === API changes in 1.14 ===
 

diff --git a/includes/Sanitizer.php b/includes/Sanitizer.php
index 84cc7f8..28ae4cb 100644 (file)
--- a/includes/Sanitizer.php
+++ b/includes/Sanitizer.php
@@ -617,7 +617,7 @@ class Sanitizer {
                        }
 
                        if ( $attribute === 'id' )
-                               $value = Sanitizer::escapeId( $value );
+                               $value = Sanitizer::escapeId( $value, Sanitizer::NONE );
 
                        // If this attribute was previously set, override it.
                        // Output should only have one attribute of each name.

diff --git a/includes/Title.php b/includes/Title.php
index 4f82e95..abfae90 100644 (file)
--- a/includes/Title.php
+++ b/includes/Title.php
@@ -451,13 +451,7 @@ class Title {
         * Escape a text fragment, say from a link, for a URL
         */
        static function escapeFragmentForURL( $fragment ) {
-               $fragment = str_replace( ' ', '_', $fragment );
-               $fragment = urlencode( Sanitizer::decodeCharReferences( $fragment ) );
-               $replaceArray = array(
-                       '%3A' => ':',
-                       '%' => '.'
-               );
-               return strtr( $fragment, $replaceArray );
+               return Sanitizer::escapeId( $fragment, Sanitizer::NONE );
        }
 
 #----------------------------------------------------------------------------

diff --git a/maintenance/parserTests.txt b/maintenance/parserTests.txt
index 8e514e4..a30db63 100644 (file)
--- a/maintenance/parserTests.txt
+++ b/maintenance/parserTests.txt
@@ -1454,7 +1454,7 @@ Example for such a section: == < ==
 !! input
 [[%23%3c]][[%23%3e]]
 !! result
-<p><a href="#.3C" title="">#&lt;</a><a href="#.3E" title="">#&gt;</a>
+<p><a href="#x.3C" title="">#&lt;</a><a href="#x.3E" title="">#&gt;</a>
 </p>
 !! end
 
@@ -4662,7 +4662,7 @@ Sanitizer: Escaping of spaces, multibyte characters, colons & other stuff in id=
 !! input
 <span id="æ: v">byte</span>[[#æ: v|backlink]]
 !! result
-<p><span id=".C3.A6:_v">byte</span><a href="#.C3.A6:_v" title="">backlink</a>
+<p><span id="x.C3.A6:_v">byte</span><a href="#x.C3.A6:_v" title="">backlink</a>
 </p>
 !! end