/**
* Some functions to help implement an external link filter for spam control.
*
- * TODO: implement the filter. Currently these are just some functions to help
+ * @todo implement the filter. Currently these are just some functions to help
* maintenance/cleanupSpam.php remove links to a single specified domain. The
* next thing is to implement functions for checking a given page against a big
* list of domains.
* Another cool thing to do would be a web interface for fast spam removal.
*/
class LinkFilter {
+
/**
- * @static
+ * Check whether $text contains a link to $filterEntry
+ *
+ * @param $text String: text to check
+ * @param $filterEntry String: domainparts, see makeRegex() for more details
+ * @return Integer: 0 if no match or 1 if there's at least one match
*/
static function matchEntry( $text, $filterEntry ) {
$regex = LinkFilter::makeRegex( $filterEntry );
}
/**
- * @static
+ * Builds a regex pattern for $filterEntry.
+ *
+ * @param $filterEntry String: URL, if it begins with "*.", it'll be
+ * replaced to match any subdomain
+ * @return String: regex pattern, for preg_match()
*/
private static function makeRegex( $filterEntry ) {
$regex = '!http://';
*
* Asterisks in any other location are considered invalid.
*
- * @static
* @param $filterEntry String: domainparts
* @param $prot String: protocol
+ * @return String
* @deprecated Use makeLikeArray() and pass result to Database::buildLike() instead
*/
public static function makeLike( $filterEntry , $prot = 'http://' ) {
wfDeprecated( __METHOD__ );
- $db = wfGetDB( DB_MASTER );
- if ( substr( $filterEntry, 0, 2 ) == '*.' ) {
- $subdomains = true;
- $filterEntry = substr( $filterEntry, 2 );
- if ( $filterEntry == '' ) {
- // We don't want to make a clause that will match everything,
- // that could be dangerous
- return false;
- }
- } else {
- $subdomains = false;
- }
- // No stray asterisks, that could cause confusion
- // It's not simple or efficient to handle it properly so we don't
- // handle it at all.
- if ( strpos( $filterEntry, '*' ) !== false ) {
+
+ $like = self::makeLikeArray( $filterEntry , $prot );
+ if ( !$like ) {
return false;
}
- $slash = strpos( $filterEntry, '/' );
- if ( $slash !== false ) {
- $path = substr( $filterEntry, $slash );
- $host = substr( $filterEntry, 0, $slash );
- } else {
- $path = '/';
- $host = $filterEntry;
- }
- // Reverse the labels in the hostname, convert to lower case
- // For emails reverse domainpart only
- if ( $prot == 'mailto:' && strpos($host, '@') ) {
- // complete email adress
- $mailparts = explode( '@', $host );
- $domainpart = strtolower( implode( '.', array_reverse( explode( '.', $mailparts[1] ) ) ) );
- $host = $domainpart . '@' . $mailparts[0];
- $like = $db->escapeLike( "$prot$host" ) . "%";
- } elseif ( $prot == 'mailto:' ) {
- // domainpart of email adress only. do not add '.'
- $host = strtolower( implode( '.', array_reverse( explode( '.', $host ) ) ) );
- $like = $db->escapeLike( "$prot$host" ) . "%";
+ $dbw = wfGetDB( DB_MASTER );
+ $s = $dbw->buildLike( $like );
+ $m = false;
+ if ( preg_match( "/^ *LIKE '(.*)' *$/", $s, $m ) ) {
+ return $m[1];
} else {
- $host = strtolower( implode( '.', array_reverse( explode( '.', $host ) ) ) );
- if ( substr( $host, -1, 1 ) !== '.' ) {
- $host .= '.';
- }
- $like = $db->escapeLike( "$prot$host" );
-
- if ( $subdomains ) {
- $like .= '%';
- }
- if ( !$subdomains || $path !== '/' ) {
- $like .= $db->escapeLike( $path ) . '%';
- }
+ throw new MWException( __METHOD__.': this DBMS is not supported by this function.' );
}
- return $like;
}
/**
- * Make an array to be used for calls to Database::like(), which will match the specified
- * string. There are several kinds of filter entry:
+ * Make an array to be used for calls to DatabaseBase::buildLike(), which
+ * will match the specified string. There are several kinds of filter entry:
* *.domain.com - Produces http://com.domain.%, matches domain.com
* and www.domain.com
* domain.com - Produces http://com.domain./%, matches domain.com
* domain.com/xy but not www.domain.com/xy
*
* Asterisks in any other location are considered invalid.
- *
- * @static
+ *
* @param $filterEntry String: domainparts
* @param $prot String: protocol
+ * @return Array to be passed to DatabaseBase::buildLike() or false on error
*/
public static function makeLikeArray( $filterEntry , $prot = 'http://' ) {
$db = wfGetDB( DB_MASTER );
/**
* Filters an array returned by makeLikeArray(), removing everything past first pattern placeholder.
- * @static
+ *
* @param $arr array: array to filter
* @return filtered array
*/