includes/specials/SpecialExport.php

   1 <?php
   2 # Copyright (C) 2003-2008 Brion Vibber <brion@pobox.com>
   3 # http://www.mediawiki.org/
   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  */
  23
  24 class SpecialExport extends SpecialPage {
  25
  26         private $curonly, $doExport, $pageLinkDepth, $templates;
  27         private $images;
  28
  29         public function __construct() {
  30                 parent::__construct( 'Export' );
  31         }
  32
  33         public function execute( $par ) {
  34                 global $wgOut, $wgRequest, $wgSitename, $wgExportAllowListContributors;
  35                 global $wgExportAllowHistory, $wgExportMaxHistory, $wgExportMaxLinkDepth;
  36
  37                 $this->setHeaders();
  38                 $this->outputHeader();
  39
  40                 // Set some variables
  41                 $this->curonly = true;
  42                 $this->doExport = false;
  43                 $this->templates = $wgRequest->getCheck( 'templates' );
  44                 $this->images = $wgRequest->getCheck( 'images' ); // Doesn't do anything yet
  45                 $this->pageLinkDepth = $this->validateLinkDepth(
  46                         $wgRequest->getIntOrNull( 'pagelink-depth' ) );
  47
  48                 if ( $wgRequest->getCheck( 'addcat' ) ) {
  49                         $page = $wgRequest->getText( 'pages' );
  50                         $catname = $wgRequest->getText( 'catname' );
  51
  52                         if ( $catname !== '' && $catname !== NULL && $catname !== false ) {
  53                                 $t = Title::makeTitleSafe( NS_MAIN, $catname );
  54                                 if ( $t ) {
  55                                         /**
  56                                          * @fixme This can lead to hitting memory limit for very large
  57                                          * categories. Ideally we would do the lookup synchronously
  58                                          * during the export in a single query.
  59                                          */
  60                                         $catpages = $this->getPagesFromCategory( $t );
  61                                         if ( $catpages ) $page .= "\n" . implode( "\n", $catpages );
  62                                 }
  63                         }
  64                 }
  65                 else if( $wgRequest->getCheck( 'addns' ) ) {
  66                         $page = $wgRequest->getText( 'pages' );
  67                         $nsindex = $wgRequest->getText( 'nsindex' );
  68
  69                         if ( $nsindex !== '' && $nsindex !== NULL && $nsindex !== false ) {
  70                                 /**
  71                                  * Same implementation as above, so same @fixme
  72                                  */
  73                                 $nspages = $this->getPagesFromNamespace( $nsindex );
  74                                 if ( $nspages ) $page .= "\n" . implode( "\n", $nspages );
  75                         }
  76                 }
  77                 else if( $wgRequest->wasPosted() && $par == '' ) {
  78                         $page = $wgRequest->getText( 'pages' );
  79                         $this->curonly = $wgRequest->getCheck( 'curonly' );
  80                         $rawOffset = $wgRequest->getVal( 'offset' );
  81                         if( $rawOffset ) {
  82                                 $offset = wfTimestamp( TS_MW, $rawOffset );
  83                         } else {
  84                                 $offset = null;
  85                         }
  86                         $limit = $wgRequest->getInt( 'limit' );
  87                         $dir = $wgRequest->getVal( 'dir' );
  88                         $history = array(
  89                                 'dir' => 'asc',
  90                                 'offset' => false,
  91                                 'limit' => $wgExportMaxHistory,
  92                         );
  93                         $historyCheck = $wgRequest->getCheck( 'history' );
  94                         if ( $this->curonly ) {
  95                                 $history = WikiExporter::CURRENT;
  96                         } elseif ( !$historyCheck ) {
  97                                 if ( $limit > 0 && $limit < $wgExportMaxHistory ) {
  98                                         $history['limit'] = $limit;
  99                                 }
 100                                 if ( !is_null( $offset ) ) {
 101                                         $history['offset'] = $offset;
 102                                 }
 103                                 if ( strtolower( $dir ) == 'desc' ) {
 104                                         $history['dir'] = 'desc';
 105                                 }
 106                         }
 107
 108                         if( $page != '' ) $this->doExport = true;
 109                 } else {
 110                         // Default to current-only for GET requests
 111                         $page = $wgRequest->getText( 'pages', $par );
 112                         $historyCheck = $wgRequest->getCheck( 'history' );
 113                         if( $historyCheck ) {
 114                                 $history = WikiExporter::FULL;
 115                         } else {
 116                                 $history = WikiExporter::CURRENT;
 117                         }
 118
 119                         if( $page != '' ) $this->doExport = true;
 120                 }
 121
 122                 if( !$wgExportAllowHistory ) {
 123                         // Override
 124                         $history = WikiExporter::CURRENT;
 125                 }
 126
 127                 $list_authors = $wgRequest->getCheck( 'listauthors' );
 128                 if ( !$this->curonly || !$wgExportAllowListContributors ) $list_authors = false ;
 129
 130                 if ( $this->doExport ) {
 131                         $wgOut->disable();
 132                         // Cancel output buffering and gzipping if set
 133                         // This should provide safer streaming for pages with history
 134                         wfResetOutputBuffers();
 135                         header( "Content-type: application/xml; charset=utf-8" );
 136                         if( $wgRequest->getCheck( 'wpDownload' ) ) {
 137                                 // Provide a sane filename suggestion
 138                                 $filename = urlencode( $wgSitename . '-' . wfTimestampNow() . '.xml' );
 139                                 $wgRequest->response()->header( "Content-disposition: attachment;filename={$filename}" );
 140                         }
 141                         $this->doExport( $page, $history, $list_authors );
 142                         return;
 143                 }
 144
 145                 $wgOut->addWikiMsg( 'exporttext' );
 146
 147                 $form = Xml::openElement( 'form', array( 'method' => 'post',
 148                         'action' => $this->getTitle()->getLocalUrl( 'action=submit' ) ) );
 149                 $form .= Xml::inputLabel( wfMsg( 'export-addcattext' )  , 'catname', 'catname', 40 ) . '&nbsp;';
 150                 $form .= Xml::submitButton( wfMsg( 'export-addcat' ), array( 'name' => 'addcat' ) ) . '<br />';
 151
 152                 $form .= Xml::namespaceSelector( '', null, 'nsindex', wfMsg( 'export-addnstext' ) ) . '&nbsp;';
 153                 $form .= Xml::submitButton( wfMsg( 'export-addns' ), array( 'name' => 'addns' ) ) . '<br />';
 154
 155                 $form .= Xml::element( 'textarea', array( 'name' => 'pages', 'cols' => 40, 'rows' => 10 ), $page, false );
 156                 $form .= '<br />';
 157
 158                 if( $wgExportAllowHistory ) {
 159                         $form .= Xml::checkLabel( wfMsg( 'exportcuronly' ), 'curonly', 'curonly', true ) . '<br />';
 160                 } else {
 161                         $wgOut->addHTML( wfMsgExt( 'exportnohistory', 'parse' ) );
 162                 }
 163                 $form .= Xml::checkLabel( wfMsg( 'export-templates' ), 'templates', 'wpExportTemplates', false ) . '<br />';
 164                 if( $wgExportMaxLinkDepth ) {
 165                         $form .= Xml::inputLabel( wfMsg( 'export-pagelinks' ), 'pagelink-depth', 'pagelink-depth', 20, 0 ) . '<br />';
 166                 }
 167                 // Enable this when we can do something useful exporting/importing image information. :)
 168                 //$form .= Xml::checkLabel( wfMsg( 'export-images' ), 'images', 'wpExportImages', false ) . '<br />';
 169                 $form .= Xml::checkLabel( wfMsg( 'export-download' ), 'wpDownload', 'wpDownload', true ) . '<br />';
 170
 171                 $form .= Xml::submitButton( wfMsg( 'export-submit' ), array( 'accesskey' => 's' ) );
 172                 $form .= Xml::closeElement( 'form' );
 173                 $wgOut->addHTML( $form );
 174         }
 175
 176         /**
 177          * Do the actual page exporting
 178          * @param string $page User input on what page(s) to export
 179          * @param mixed  $history one of the WikiExporter history export constants
 180          */
 181         private function doExport( $page, $history, $list_authors ) {
 182                 global $wgExportMaxHistory;
 183
 184                 /* Split up the input and look up linked pages */
 185                 $inputPages = array_filter( explode( "\n", $page ), array( $this, 'filterPage' ) );
 186                 $pageSet = array_flip( $inputPages );
 187
 188                 if( $this->templates ) {
 189                         $pageSet = $this->getTemplates( $inputPages, $pageSet );
 190                 }
 191
 192                 if( $linkDepth = $this->pageLinkDepth ) {
 193                         $pageSet = $this->getPageLinks( $inputPages, $pageSet, $linkDepth );
 194                 }
 195
 196                 /*
 197                 // Enable this when we can do something useful exporting/importing image information. :)
 198                 if( $this->images ) ) {
 199                         $pageSet = $this->getImages( $inputPages, $pageSet );
 200                 }
 201                 */
 202
 203                 $pages = array_keys( $pageSet );
 204
 205                 /* Ok, let's get to it... */
 206                 if( $history == WikiExporter::CURRENT ) {
 207                         $lb = false;
 208                         $db = wfGetDB( DB_SLAVE );
 209                         $buffer = WikiExporter::BUFFER;
 210                 } else {
 211                         // Use an unbuffered query; histories may be very long!
 212                         $lb = wfGetLBFactory()->newMainLB();
 213                         $db = $lb->getConnection( DB_SLAVE );
 214                         $buffer = WikiExporter::STREAM;
 215
 216                         // This might take a while... :D
 217                         wfSuppressWarnings();
 218                         set_time_limit(0);
 219                         wfRestoreWarnings();
 220                 }
 221                 $exporter = new WikiExporter( $db, $history, $buffer );
 222                 $exporter->list_authors = $list_authors;
 223                 $exporter->openStream();
 224                 foreach( $pages as $page ) {
 225                         /*
 226                         if( $wgExportMaxHistory && !$this->curonly ) {
 227                                 $title = Title::newFromText( $page );
 228                                 if( $title ) {
 229                                         $count = Revision::countByTitle( $db, $title );
 230                                         if( $count > $wgExportMaxHistory ) {
 231                                                 wfDebug( __FUNCTION__ .
 232                                                         ": Skipped $page, $count revisions too big\n" );
 233                                                 continue;
 234                                         }
 235                                 }
 236                         }*/
 237                         #Bug 8824: Only export pages the user can read
 238                         $title = Title::newFromText( $page );
 239                         if( is_null( $title ) ) continue; #TODO: perhaps output an <error> tag or something.
 240                                 if( !$title->userCanRead() ) continue; #TODO: perhaps output an <error> tag or something.
 241
 242                                 $exporter->pageByTitle( $title );
 243                 }
 244
 245                 $exporter->closeStream();
 246                 if( $lb ) {
 247                         $lb->closeAll();
 248                 }
 249         }
 250
 251
 252         private function getPagesFromCategory( $title ) {
 253                 global $wgContLang;
 254
 255                 $name = $title->getDBkey();
 256
 257                 $dbr = wfGetDB( DB_SLAVE );
 258                 $res = $dbr->select( array('page', 'categorylinks' ),
 259                                         array( 'page_namespace', 'page_title' ),
 260                                         array('cl_from=page_id', 'cl_to' => $name ),
 261                                         __METHOD__, array('LIMIT' => '5000'));
 262
 263                 $pages = array();
 264                 while ( $row = $dbr->fetchObject( $res ) ) {
 265                         $n = $row->page_title;
 266                         if ($row->page_namespace) {
 267                                 $ns = $wgContLang->getNsText( $row->page_namespace );
 268                                 $n = $ns . ':' . $n;
 269                         }
 270
 271                         $pages[] = $n;
 272                 }
 273                 $dbr->freeResult($res);
 274
 275                 return $pages;
 276         }
 277
 278         private function getPagesFromNamespace( $nsindex ) {
 279                 global $wgContLang;
 280
 281                 $dbr = wfGetDB( DB_SLAVE );
 282                 $res = $dbr->select( 'page', array('page_namespace', 'page_title'),
 283                                         array('page_namespace' => $nsindex),
 284                                         __METHOD__, array('LIMIT' => '5000') );
 285
 286                 $pages = array();
 287                 while ( $row = $dbr->fetchObject( $res ) ) {
 288                         $n = $row->page_title;
 289                         if ($row->page_namespace) {
 290                                 $ns = $wgContLang->getNsText( $row->page_namespace );
 291                                 $n = $ns . ':' . $n;
 292                         }
 293
 294                         $pages[] = $n;
 295                 }
 296                 $dbr->freeResult($res);
 297
 298                 return $pages;
 299         }
 300         /**
 301          * Expand a list of pages to include templates used in those pages.
 302          * @param $inputPages array, list of titles to look up
 303          * @param $pageSet array, associative array indexed by titles for output
 304          * @return array associative array index by titles
 305          */
 306         private function getTemplates( $inputPages, $pageSet ) {
 307                 return $this->getLinks( $inputPages, $pageSet,
 308                                         'templatelinks',
 309                                         array( 'tl_namespace AS namespace', 'tl_title AS title' ),
 310                                         array( 'page_id=tl_from' ) );
 311         }
 312
 313         /**
 314          * Validate link depth setting, if available.
 315          */
 316         private function validateLinkDepth( $depth ) {
 317                 global $wgExportMaxLinkDepth;
 318                 if( $depth < 0 ) {
 319                         return 0;
 320                 }
 321                 if( $depth > $wgExportMaxLinkDepth ) {
 322                         return $wgExportMaxLinkDepth;
 323                 }
 324                 return intval( $depth );
 325         }
 326
 327         /** Expand a list of pages to include pages linked to from that page. */
 328         private function getPageLinks( $inputPages, $pageSet, $depth ) {
 329                 for( $depth=$depth; $depth>0; --$depth ) {
 330                 $pageSet = $this->getLinks( $inputPages, $pageSet, 'pagelinks',
 331                         array( 'pl_namespace AS namespace', 'pl_title AS title' ),
 332                         array( 'page_id=pl_from' ) );
 333                 }
 334                 return $pageSet;
 335         }
 336
 337         /**
 338          * Expand a list of pages to include images used in those pages.
 339          * @param $inputPages array, list of titles to look up
 340          * @param $pageSet array, associative array indexed by titles for output
 341          * @return array associative array index by titles
 342          */
 343         private function getImages( $inputPages, $pageSet ) {
 344                 return $this->getLinks( $inputPages, $pageSet,
 345                                         'imagelinks',
 346                                         array( NS_FILE . ' AS namespace', 'il_to AS title' ),
 347                                         array( 'page_id=il_from' ) );
 348         }
 349
 350         /**
 351          * Expand a list of pages to include items used in those pages.
 352          * @private
 353          */
 354         private function getLinks( $inputPages, $pageSet, $table, $fields, $join ) {
 355                 $dbr = wfGetDB( DB_SLAVE );
 356                 foreach( $inputPages as $page ) {
 357                         $title = Title::newFromText( $page );
 358                         if( $title ) {
 359                                 $pageSet[$title->getPrefixedText()] = true;
 360                                 /// @fixme May or may not be more efficient to batch these
 361                                 ///        by namespace when given multiple input pages.
 362                                 $result = $dbr->select(
 363                                         array( 'page', $table ),
 364                                         $fields,
 365                                         array_merge( $join,
 366                                                 array(
 367                                                         'page_namespace' => $title->getNamespace(),
 368                                                         'page_title' => $title->getDBKey() ) ),
 369                                         __METHOD__ );
 370                                 foreach( $result as $row ) {
 371                                         $template = Title::makeTitle( $row->namespace, $row->title );
 372                                         $pageSet[$template->getPrefixedText()] = true;
 373                                 }
 374                         }
 375                 }
 376                 return $pageSet;
 377         }
 378
 379         /**
 380          * Callback function to remove empty strings from the pages array.
 381          */
 382         private function filterPage( $page ) {
 383                 return $page !== '' && $page !== null;
 384         }
 385 }