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