includes/api/ApiQueryExternalLinks.php

   1 <?php
   2 /**
   3  * Copyright © 2006 Yuri Astrakhan "<Firstname><Lastname>@gmail.com"
   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  */
  22
  23 /**
  24  * A query module to list all external URLs found on a given set of pages.
  25  *
  26  * @ingroup API
  27  */
  28 class ApiQueryExternalLinks extends ApiQueryBase {
  29
  30         public function __construct( ApiQuery $query, $moduleName ) {
  31                 parent::__construct( $query, $moduleName, 'el' );
  32         }
  33
  34         public function execute() {
  35                 if ( $this->getPageSet()->getGoodTitleCount() == 0 ) {
  36                         return;
  37                 }
  38
  39                 $params = $this->extractRequestParams();
  40                 $db = $this->getDB();
  41
  42                 $query = $params['query'];
  43                 $protocol = ApiQueryExtLinksUsage::getProtocolPrefix( $params['protocol'] );
  44
  45                 $this->addFields( [
  46                         'el_from',
  47                         'el_to'
  48                 ] );
  49
  50                 $this->addTables( 'externallinks' );
  51                 $this->addWhereFld( 'el_from', array_keys( $this->getPageSet()->getGoodTitles() ) );
  52
  53                 $orderBy = [];
  54
  55                 // Don't order by el_from if it's constant in the WHERE clause
  56                 if ( count( $this->getPageSet()->getGoodTitles() ) != 1 ) {
  57                         $orderBy[] = 'el_from';
  58                 }
  59
  60                 if ( $query !== null && $query !== '' ) {
  61                         if ( $protocol === null ) {
  62                                 $protocol = 'http://';
  63                         }
  64
  65                         // Normalize query to match the normalization applied for the externallinks table
  66                         $query = Parser::normalizeLinkUrl( $protocol . $query );
  67
  68                         $conds = LinkFilter::getQueryConditions( $query, [
  69                                 'protocol' => '',
  70                                 'oneWildcard' => true,
  71                                 'db' => $db
  72                         ] );
  73                         if ( !$conds ) {
  74                                 $this->dieWithError( 'apierror-badquery' );
  75                         }
  76                         $this->addWhere( $conds );
  77                         if ( !isset( $conds['el_index_60'] ) ) {
  78                                 $orderBy[] = 'el_index_60';
  79                         }
  80                 } else {
  81                         $orderBy[] = 'el_index_60';
  82
  83                         if ( $protocol !== null ) {
  84                                 $this->addWhere( 'el_index_60' . $db->buildLike( "$protocol", $db->anyString() ) );
  85                         } else {
  86                                 // We're querying all protocols, filter out duplicate protocol-relative links
  87                                 $this->addWhere( $db->makeList( [
  88                                         'el_to NOT' . $db->buildLike( '//', $db->anyString() ),
  89                                         'el_index_60 ' . $db->buildLike( 'http://', $db->anyString() ),
  90                                 ], LIST_OR ) );
  91                         }
  92                 }
  93
  94                 $orderBy[] = 'el_id';
  95                 $this->addOption( 'ORDER BY', $orderBy );
  96                 $this->addFields( $orderBy ); // Make sure
  97
  98                 $this->addOption( 'LIMIT', $params['limit'] + 1 );
  99
 100                 if ( $params['continue'] !== null ) {
 101                         $cont = explode( '|', $params['continue'] );
 102                         $this->dieContinueUsageIf( count( $cont ) !== count( $orderBy ) );
 103                         $i = count( $cont ) - 1;
 104                         $cond = $orderBy[$i] . ' >= ' . $db->addQuotes( rawurldecode( $cont[$i] ) );
 105                         while ( $i-- > 0 ) {
 106                                 $field = $orderBy[$i];
 107                                 $v = $db->addQuotes( rawurldecode( $cont[$i] ) );
 108                                 $cond = "($field > $v OR ($field = $v AND $cond))";
 109                         }
 110                         $this->addWhere( $cond );
 111                 }
 112
 113                 $res = $this->select( __METHOD__ );
 114
 115                 $count = 0;
 116                 foreach ( $res as $row ) {
 117                         if ( ++$count > $params['limit'] ) {
 118                                 // We've reached the one extra which shows that
 119                                 // there are additional pages to be had. Stop here...
 120                                 $this->setContinue( $orderBy, $row );
 121                                 break;
 122                         }
 123                         $entry = [];
 124                         $to = $row->el_to;
 125                         // expand protocol-relative urls
 126                         if ( $params['expandurl'] ) {
 127                                 $to = wfExpandUrl( $to, PROTO_CANONICAL );
 128                         }
 129                         ApiResult::setContentValue( $entry, 'url', $to );
 130                         $fit = $this->addPageSubItem( $row->el_from, $entry );
 131                         if ( !$fit ) {
 132                                 $this->setContinue( $orderBy, $row );
 133                                 break;
 134                         }
 135                 }
 136         }
 137
 138         private function setContinue( $orderBy, $row ) {
 139                 $fields = [];
 140                 foreach ( $orderBy as $field ) {
 141                         $fields[] = strtr( $row->$field, [ '%' => '%25', '|' => '%7C' ] );
 142                 }
 143                 $this->setContinueEnumParameter( 'continue', implode( '|', $fields ) );
 144         }
 145
 146         public function getCacheMode( $params ) {
 147                 return 'public';
 148         }
 149
 150         public function getAllowedParams() {
 151                 return [
 152                         'limit' => [
 153                                 ApiBase::PARAM_DFLT => 10,
 154                                 ApiBase::PARAM_TYPE => 'limit',
 155                                 ApiBase::PARAM_MIN => 1,
 156                                 ApiBase::PARAM_MAX => ApiBase::LIMIT_BIG1,
 157                                 ApiBase::PARAM_MAX2 => ApiBase::LIMIT_BIG2
 158                         ],
 159                         'continue' => [
 160                                 ApiBase::PARAM_HELP_MSG => 'api-help-param-continue',
 161                         ],
 162                         'protocol' => [
 163                                 ApiBase::PARAM_TYPE => ApiQueryExtLinksUsage::prepareProtocols(),
 164                                 ApiBase::PARAM_DFLT => '',
 165                         ],
 166                         'query' => null,
 167                         'expandurl' => false,
 168                 ];
 169         }
 170
 171         protected function getExamplesMessages() {
 172                 return [
 173                         'action=query&prop=extlinks&titles=Main%20Page'
 174                                 => 'apihelp-query+extlinks-example-simple',
 175                 ];
 176         }
 177
 178         public function getHelpUrls() {
 179                 return 'https://www.mediawiki.org/wiki/Special:MyLanguage/API:Extlinks';
 180         }
 181 }