includes/specials/SpecialLinkSearch.php

   1 <?php
   2 /**
   3  * Implements Special:LinkSearch
   4  *
   5  * This program is free software; you can redistribute it and/or modify
   6  * it under the terms of the GNU General Public License as published by
   7  * the Free Software Foundation; either version 2 of the License, or
   8  * (at your option) any later version.
   9  *
  10  * This program is distributed in the hope that it will be useful,
  11  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  12  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
  13  * GNU General Public License for more details.
  14  *
  15  * You should have received a copy of the GNU General Public License along
  16  * with this program; if not, write to the Free Software Foundation, Inc.,
  17  * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
  18  * http://www.gnu.org/copyleft/gpl.html
  19  *
  20  * @file
  21  * @ingroup SpecialPage
  22  * @author Brion Vibber
  23  */
  24
  25 use Wikimedia\Rdbms\ResultWrapper;
  26 use Wikimedia\Rdbms\IDatabase;
  27
  28 /**
  29  * Special:LinkSearch to search the external-links table.
  30  * @ingroup SpecialPage
  31  */
  32 class LinkSearchPage extends QueryPage {
  33         /** @var array|bool */
  34         private $mungedQuery = false;
  35
  36         function setParams( $params ) {
  37                 $this->mQuery = $params['query'];
  38                 $this->mNs = $params['namespace'];
  39                 $this->mProt = $params['protocol'];
  40         }
  41
  42         function __construct( $name = 'LinkSearch' ) {
  43                 parent::__construct( $name );
  44
  45                 // Since we don't control the constructor parameters, we can't inject services that way.
  46                 // Instead, we initialize services in the execute() method, and allow them to be overridden
  47                 // using the setServices() method.
  48         }
  49
  50         function isCacheable() {
  51                 return false;
  52         }
  53
  54         public function execute( $par ) {
  55                 $this->setHeaders();
  56                 $this->outputHeader();
  57
  58                 $out = $this->getOutput();
  59                 $out->allowClickjacking();
  60
  61                 $request = $this->getRequest();
  62                 $target = $request->getVal( 'target', $par );
  63                 $namespace = $request->getIntOrNull( 'namespace' );
  64
  65                 $protocols_list = [];
  66                 foreach ( $this->getConfig()->get( 'UrlProtocols' ) as $prot ) {
  67                         if ( $prot !== '//' ) {
  68                                 $protocols_list[] = $prot;
  69                         }
  70                 }
  71
  72                 $target2 = $target;
  73                 // Get protocol, default is http://
  74                 $protocol = 'http://';
  75                 $bits = wfParseUrl( $target );
  76                 if ( isset( $bits['scheme'] ) && isset( $bits['delimiter'] ) ) {
  77                         $protocol = $bits['scheme'] . $bits['delimiter'];
  78                         // Make sure wfParseUrl() didn't make some well-intended correction in the
  79                         // protocol
  80                         if ( strcasecmp( $protocol, substr( $target, 0, strlen( $protocol ) ) ) === 0 ) {
  81                                 $target2 = substr( $target, strlen( $protocol ) );
  82                         } else {
  83                                 // If it did, let LinkFilter::makeLikeArray() handle this
  84                                 $protocol = '';
  85                         }
  86                 }
  87
  88                 $out->addWikiMsg(
  89                         'linksearch-text',
  90                         '<nowiki>' . $this->getLanguage()->commaList( $protocols_list ) . '</nowiki>',
  91                         count( $protocols_list )
  92                 );
  93                 $fields = [
  94                         'target' => [
  95                                 'type' => 'text',
  96                                 'name' => 'target',
  97                                 'id' => 'target',
  98                                 'size' => 50,
  99                                 'label-message' => 'linksearch-pat',
 100                                 'default' => $target,
 101                                 'dir' => 'ltr',
 102                         ]
 103                 ];
 104                 if ( !$this->getConfig()->get( 'MiserMode' ) ) {
 105                         $fields += [
 106                                 'namespace' => [
 107                                         'type' => 'namespaceselect',
 108                                         'name' => 'namespace',
 109                                         'label-message' => 'linksearch-ns',
 110                                         'default' => $namespace,
 111                                         'id' => 'namespace',
 112                                         'all' => '',
 113                                         'cssclass' => 'namespaceselector',
 114                                 ],
 115                         ];
 116                 }
 117                 $hiddenFields = [
 118                         'title' => $this->getPageTitle()->getPrefixedDBkey(),
 119                 ];
 120                 $htmlForm = HTMLForm::factory( 'ooui', $fields, $this->getContext() );
 121                 $htmlForm->addHiddenFields( $hiddenFields );
 122                 $htmlForm->setSubmitTextMsg( 'linksearch-ok' );
 123                 $htmlForm->setWrapperLegendMsg( 'linksearch' );
 124                 $htmlForm->setAction( wfScript() );
 125                 $htmlForm->setMethod( 'get' );
 126                 $htmlForm->prepareForm()->displayForm( false );
 127                 $this->addHelpLink( 'Help:Linksearch' );
 128
 129                 if ( $target != '' ) {
 130                         $this->setParams( [
 131                                 'query' => Parser::normalizeLinkUrl( $target2 ),
 132                                 'namespace' => $namespace,
 133                                 'protocol' => $protocol ] );
 134                         parent::execute( $par );
 135                         if ( $this->mungedQuery === false ) {
 136                                 $out->addWikiMsg( 'linksearch-error' );
 137                         }
 138                 }
 139         }
 140
 141         /**
 142          * Disable RSS/Atom feeds
 143          * @return bool
 144          */
 145         function isSyndicated() {
 146                 return false;
 147         }
 148
 149         /**
 150          * Return an appropriately formatted LIKE query and the clause
 151          *
 152          * @param string $query Search pattern to search for
 153          * @param string $prot Protocol, e.g. 'http://'
 154          *
 155          * @return array
 156          */
 157         static function mungeQuery( $query, $prot ) {
 158                 $field = 'el_index';
 159                 $dbr = wfGetDB( DB_REPLICA );
 160
 161                 if ( $query === '*' && $prot !== '' ) {
 162                         // Allow queries like 'ftp://*' to find all ftp links
 163                         $rv = [ $prot, $dbr->anyString() ];
 164                 } else {
 165                         $rv = LinkFilter::makeLikeArray( $query, $prot );
 166                 }
 167
 168                 if ( $rv === false ) {
 169                         // LinkFilter doesn't handle wildcard in IP, so we'll have to munge here.
 170                         $pattern = '/^(:?[0-9]{1,3}\.)+\*\s*$|^(:?[0-9]{1,3}\.){3}[0-9]{1,3}:[0-9]*\*\s*$/';
 171                         if ( preg_match( $pattern, $query ) ) {
 172                                 $rv = [ $prot . rtrim( $query, " \t*" ), $dbr->anyString() ];
 173                                 $field = 'el_to';
 174                         }
 175                 }
 176
 177                 return [ $rv, $field ];
 178         }
 179
 180         function linkParameters() {
 181                 $params = [];
 182                 $params['target'] = $this->mProt . $this->mQuery;
 183                 if ( $this->mNs !== null && !$this->getConfig()->get( 'MiserMode' ) ) {
 184                         $params['namespace'] = $this->mNs;
 185                 }
 186
 187                 return $params;
 188         }
 189
 190         public function getQueryInfo() {
 191                 $dbr = wfGetDB( DB_REPLICA );
 192                 // strip everything past first wildcard, so that
 193                 // index-based-only lookup would be done
 194                 list( $this->mungedQuery, $clause ) = self::mungeQuery( $this->mQuery, $this->mProt );
 195                 if ( $this->mungedQuery === false ) {
 196                         // Invalid query; return no results
 197                         return [ 'tables' => 'page', 'fields' => 'page_id', 'conds' => '0=1' ];
 198                 }
 199
 200                 $stripped = LinkFilter::keepOneWildcard( $this->mungedQuery );
 201                 $like = $dbr->buildLike( $stripped );
 202                 $retval = [
 203                         'tables' => [ 'page', 'externallinks' ],
 204                         'fields' => [
 205                                 'namespace' => 'page_namespace',
 206                                 'title' => 'page_title',
 207                                 'value' => 'el_index',
 208                                 'url' => 'el_to'
 209                         ],
 210                         'conds' => [
 211                                 'page_id = el_from',
 212                                 "$clause $like"
 213                         ],
 214                         'options' => [ 'USE INDEX' => $clause ]
 215                 ];
 216
 217                 if ( $this->mNs !== null && !$this->getConfig()->get( 'MiserMode' ) ) {
 218                         $retval['conds']['page_namespace'] = $this->mNs;
 219                 }
 220
 221                 return $retval;
 222         }
 223
 224         /**
 225          * Pre-fill the link cache
 226          *
 227          * @param IDatabase $db
 228          * @param ResultWrapper $res
 229          */
 230         function preprocessResults( $db, $res ) {
 231                 $this->executeLBFromResultWrapper( $res );
 232         }
 233
 234         /**
 235          * @param Skin $skin
 236          * @param object $result Result row
 237          * @return string
 238          */
 239         function formatResult( $skin, $result ) {
 240                 $title = new TitleValue( (int)$result->namespace, $result->title );
 241                 $pageLink = $this->getLinkRenderer()->makeLink( $title );
 242
 243                 $url = $result->url;
 244                 $urlLink = Linker::makeExternalLink( $url, $url );
 245
 246                 return $this->msg( 'linksearch-line' )->rawParams( $urlLink, $pageLink )->escaped();
 247         }
 248
 249         /**
 250          * Override to squash the ORDER BY.
 251          * We do a truncated index search, so the optimizer won't trust
 252          * it as good enough for optimizing sort. The implicit ordering
 253          * from the scan will usually do well enough for our needs.
 254          * @return array
 255          */
 256         function getOrderFields() {
 257                 return [];
 258         }
 259
 260         protected function getGroupName() {
 261                 return 'redirects';
 262         }
 263
 264         /**
 265          * enwiki complained about low limits on this special page
 266          *
 267          * @see T130058
 268          * @todo FIXME This special page should not use LIMIT for paging
 269          * @return int
 270          */
 271         protected function getMaxResults() {
 272                 return max( parent::getMaxResults(), 60000 );
 273         }
 274 }