Legoktm has uploaded a new change for review.
https://gerrit.wikimedia.org/r/231748
Change subject: [WIP] Attempt to canonicalize URLs before storing them
......................................................................
[WIP] Attempt to canonicalize URLs before storing them
Convert index.php?title=$1 to /wiki/$1
TODO:
* Write tests
Bug: T108602
Change-Id: Ib47ca7c9e0c75aa295d9edb81b5fd0f671a6fb57
---
M UrlShortener.utils.php
1 file changed, 22 insertions(+), 4 deletions(-)
git pull ssh://gerrit.wikimedia.org:29418/mediawiki/extensions/UrlShortener
refs/changes/48/231748/1
diff --git a/UrlShortener.utils.php b/UrlShortener.utils.php
index 72ad2df..34e6d5e 100755
--- a/UrlShortener.utils.php
+++ b/UrlShortener.utils.php
@@ -28,10 +28,7 @@
* @return Status with value of base36 encoded shortcode that refers to
the $url
*/
public static function maybeCreateShortCode( $url, User $user ) {
- // First, cannonicalize the URL
- // store everything in the db as HTTP, we'll convert it before
- // redirecting users
- $url = self::convertToProtocol( $url, PROTO_HTTP );
+ $url = self::normalizeUrl( $url );
$dbw = self::getDB( DB_MASTER );
$id = $dbw->selectField(
@@ -68,6 +65,27 @@
return Status::newGood( self::encodeId( $id ) );
}
+ public static function normalizeUrl( $url ) {
+ global $wgArticlePath;
+ // First, force the protocol to HTTP, we'll convert
+ // it to a different one when redirecting
+ $url = self::convertToProtocol( $url, PROTO_HTTP );
+
+ // If the wiki is using an article path (e.g. /wiki/$1) try
+ // and convert plain index.php?title=$1 URLs to the canonical
form
+ if ( $wgArticlePath !== false && strpos( $url, '?' ) ) {
+ $parsed = wfParseUrl( $url );
+ $query = wfCgiToArray( $parsed['query'] );
+ if ( count( $query ) === 1 && isset( $query['title'] )
&& $parsed['path'] === wfScript() ) {
+ $parsed['path'] = str_replace( '$1',
$query['title'], $wgArticlePath );
+ unset( $parsed['query'] );
+ }
+ $url = wfAssembleUrl( $parsed );
+ }
+
+ return $url;
+ }
+
/**
* Converts a possibly protocol'd url to the one specified
*
--
To view, visit https://gerrit.wikimedia.org/r/231748
To unsubscribe, visit https://gerrit.wikimedia.org/r/settings
Gerrit-MessageType: newchange
Gerrit-Change-Id: Ib47ca7c9e0c75aa295d9edb81b5fd0f671a6fb57
Gerrit-PatchSet: 1
Gerrit-Project: mediawiki/extensions/UrlShortener
Gerrit-Branch: master
Gerrit-Owner: Legoktm <[email protected]>
_______________________________________________
MediaWiki-commits mailing list
[email protected]
https://lists.wikimedia.org/mailman/listinfo/mediawiki-commits
