includes/api/ApiQueryExternalLinks.php

   1 <?php
   2 /**
   3  *
   4  *
   5  * Created on May 13, 2007
   6  *
   7  * Copyright © 2006 Yuri Astrakhan "<Firstname><Lastname>@gmail.com"
   8  *
   9  * This program is free software; you can redistribute it and/or modify
  10  * it under the terms of the GNU General Public License as published by
  11  * the Free Software Foundation; either version 2 of the License, or
  12  * (at your option) any later version.
  13  *
  14  * This program is distributed in the hope that it will be useful,
  15  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  16  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
  17  * GNU General Public License for more details.
  18  *
  19  * You should have received a copy of the GNU General Public License along
  20  * with this program; if not, write to the Free Software Foundation, Inc.,
  21  * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
  22  * http://www.gnu.org/copyleft/gpl.html
  23  *
  24  * @file
  25  */
  26
  27 /**
  28  * A query module to list all external URLs found on a given set of pages.
  29  *
  30  * @ingroup API
  31  */
  32 class ApiQueryExternalLinks extends ApiQueryBase {
  33
  34         public function __construct( $query, $moduleName ) {
  35                 parent::__construct( $query, $moduleName, 'el' );
  36         }
  37
  38         public function execute() {
  39                 if ( $this->getPageSet()->getGoodTitleCount() == 0 ) {
  40                         return;
  41                 }
  42
  43                 $params = $this->extractRequestParams();
  44
  45                 $query = $params['query'];
  46                 $protocol = ApiQueryExtLinksUsage::getProtocolPrefix( $params['protocol'] );
  47
  48                 $this->addFields( array(
  49                         'el_from',
  50                         'el_to'
  51                 ) );
  52
  53                 $this->addTables( 'externallinks' );
  54                 $this->addWhereFld( 'el_from', array_keys( $this->getPageSet()->getGoodTitles() ) );
  55
  56                 $whereQuery = $this->prepareUrlQuerySearchString( $query, $protocol );
  57
  58                 if ( $whereQuery !== null ) {
  59                         $this->addWhere( $whereQuery );
  60                 }
  61
  62                 // Don't order by el_from if it's constant in the WHERE clause
  63                 if ( count( $this->getPageSet()->getGoodTitles() ) != 1 ) {
  64                         $this->addOption( 'ORDER BY', 'el_from' );
  65                 }
  66
  67                 // If we're querying all protocols, use DISTINCT to avoid repeating protocol-relative links twice
  68                 if ( $protocol === null ) {
  69                         $this->addOption( 'DISTINCT' );
  70                 }
  71
  72                 $this->addOption( 'LIMIT', $params['limit'] + 1 );
  73                 $offset = isset( $params['offset'] ) ? $params['offset'] : 0;
  74                 if ( $offset ) {
  75                         $this->addOption( 'OFFSET', $params['offset'] );
  76                 }
  77
  78                 $res = $this->select( __METHOD__ );
  79
  80                 $count = 0;
  81                 foreach ( $res as $row ) {
  82                         if ( ++$count > $params['limit'] ) {
  83                                 // We've reached the one extra which shows that
  84                                 // there are additional pages to be had. Stop here...
  85                                 $this->setContinueEnumParameter( 'offset', $offset + $params['limit'] );
  86                                 break;
  87                         }
  88                         $entry = array();
  89                         $to = $row->el_to;
  90                         // expand protocol-relative urls
  91                         if ( $params['expandurl'] ) {
  92                                 $to = wfExpandUrl( $to, PROTO_CANONICAL );
  93                         }
  94                         ApiResult::setContent( $entry, $to );
  95                         $fit = $this->addPageSubItem( $row->el_from, $entry );
  96                         if ( !$fit ) {
  97                                 $this->setContinueEnumParameter( 'offset', $offset + $count - 1 );
  98                                 break;
  99                         }
 100                 }
 101         }
 102
 103         public function getCacheMode( $params ) {
 104                 return 'public';
 105         }
 106
 107         public function getAllowedParams() {
 108                 return array(
 109                         'limit' => array(
 110                                 ApiBase::PARAM_DFLT => 10,
 111                                 ApiBase::PARAM_TYPE => 'limit',
 112                                 ApiBase::PARAM_MIN => 1,
 113                                 ApiBase::PARAM_MAX => ApiBase::LIMIT_BIG1,
 114                                 ApiBase::PARAM_MAX2 => ApiBase::LIMIT_BIG2
 115                         ),
 116                         'offset' => array(
 117                                 ApiBase::PARAM_TYPE => 'integer'
 118                         ),
 119                         'protocol' => array(
 120                                 ApiBase::PARAM_TYPE => ApiQueryExtLinksUsage::prepareProtocols(),
 121                                 ApiBase::PARAM_DFLT => '',
 122                         ),
 123                         'query' => null,
 124                         'expandurl' => false,
 125                 );
 126         }
 127
 128         public function getParamDescription() {
 129                 $p = $this->getModulePrefix();
 130
 131                 return array(
 132                         'limit' => 'How many links to return',
 133                         'offset' => 'When more results are available, use this to continue',
 134                         'protocol' => array(
 135                                 "Protocol of the URL. If empty and {$p}query set, the protocol is http.",
 136                                 "Leave both this and {$p}query empty to list all external links"
 137                         ),
 138                         'query' => 'Search string without protocol. Useful for checking ' .
 139                                 'whether a certain page contains a certain external url',
 140                         'expandurl' => 'Expand protocol-relative URLs with the canonical protocol',
 141                 );
 142         }
 143
 144         public function getResultProperties() {
 145                 return array(
 146                         '' => array(
 147                                 '*' => 'string'
 148                         )
 149                 );
 150         }
 151
 152         public function getDescription() {
 153                 return 'Returns all external URLs (not interwikis) from the given page(s).';
 154         }
 155
 156         public function getPossibleErrors() {
 157                 return array_merge( parent::getPossibleErrors(), array(
 158                         array( 'code' => 'bad_query', 'info' => 'Invalid query' ),
 159                 ) );
 160         }
 161
 162         public function getExamples() {
 163                 return array(
 164                         'api.php?action=query&prop=extlinks&titles=Main%20Page'
 165                                 => 'Get a list of external links on the [[Main Page]]',
 166                 );
 167         }
 168
 169         public function getHelpUrls() {
 170                 return 'https://www.mediawiki.org/wiki/API:Properties#extlinks_.2F_el';
 171         }
 172 }