diff options
Diffstat (limited to 'includes/specials/SpecialExport.php')
-rw-r--r-- | includes/specials/SpecialExport.php | 551 |
1 files changed, 326 insertions, 225 deletions
diff --git a/includes/specials/SpecialExport.php b/includes/specials/SpecialExport.php index 898b5a78..8bf16a71 100644 --- a/includes/specials/SpecialExport.php +++ b/includes/specials/SpecialExport.php @@ -21,208 +21,197 @@ * @ingroup SpecialPage */ -function wfExportGetPagesFromCategory( $title ) { - global $wgContLang; - - $name = $title->getDBkey(); - - $dbr = wfGetDB( DB_SLAVE ); - - list( $page, $categorylinks ) = $dbr->tableNamesN( 'page', 'categorylinks' ); - $sql = "SELECT page_namespace, page_title FROM $page " . - "JOIN $categorylinks ON cl_from = page_id " . - "WHERE cl_to = " . $dbr->addQuotes( $name ); - - $pages = array(); - $res = $dbr->query( $sql, 'wfExportGetPagesFromCategory' ); - while ( $row = $dbr->fetchObject( $res ) ) { - $n = $row->page_title; - if ($row->page_namespace) { - $ns = $wgContLang->getNsText( $row->page_namespace ); - $n = $ns . ':' . $n; - } - - $pages[] = $n; +class SpecialExport extends SpecialPage { + + private $curonly, $doExport, $pageLinkDepth, $templates; + private $images; + + public function __construct() { + parent::__construct( 'Export' ); } - $dbr->freeResult($res); - - return $pages; -} - -/** - * Expand a list of pages to include templates used in those pages. - * @param $inputPages array, list of titles to look up - * @param $pageSet array, associative array indexed by titles for output - * @return array associative array index by titles - */ -function wfExportGetTemplates( $inputPages, $pageSet ) { - return wfExportGetLinks( $inputPages, $pageSet, - 'templatelinks', - array( 'tl_namespace AS namespace', 'tl_title AS title' ), - array( 'page_id=tl_from' ) ); -} - -/** - * Expand a list of pages to include images used in those pages. - * @param $inputPages array, list of titles to look up - * @param $pageSet array, associative array indexed by titles for output - * @return array associative array index by titles - */ -function wfExportGetImages( $inputPages, $pageSet ) { - return wfExportGetLinks( $inputPages, $pageSet, - 'imagelinks', - array( NS_FILE . ' AS namespace', 'il_to AS title' ), - array( 'page_id=il_from' ) ); -} - -/** - * Expand a list of pages to include items used in those pages. - * @private - */ -function wfExportGetLinks( $inputPages, $pageSet, $table, $fields, $join ) { - $dbr = wfGetDB( DB_SLAVE ); - foreach( $inputPages as $page ) { - $title = Title::newFromText( $page ); - if( $title ) { - $pageSet[$title->getPrefixedText()] = true; - /// @fixme May or may not be more efficient to batch these - /// by namespace when given multiple input pages. - $result = $dbr->select( - array( 'page', $table ), - $fields, - array_merge( $join, - array( - 'page_namespace' => $title->getNamespace(), - 'page_title' => $title->getDBKey() ) ), - __METHOD__ ); - foreach( $result as $row ) { - $template = Title::makeTitle( $row->namespace, $row->title ); - $pageSet[$template->getPrefixedText()] = true; + + public function execute( $par ) { + global $wgOut, $wgRequest, $wgSitename, $wgExportAllowListContributors; + global $wgExportAllowHistory, $wgExportMaxHistory, $wgExportMaxLinkDepth; + global $wgExportFromNamespaces; + + $this->setHeaders(); + $this->outputHeader(); + + // Set some variables + $this->curonly = true; + $this->doExport = false; + $this->templates = $wgRequest->getCheck( 'templates' ); + $this->images = $wgRequest->getCheck( 'images' ); // Doesn't do anything yet + $this->pageLinkDepth = $this->validateLinkDepth( + $wgRequest->getIntOrNull( 'pagelink-depth' ) ); + + if ( $wgRequest->getCheck( 'addcat' ) ) { + $page = $wgRequest->getText( 'pages' ); + $catname = $wgRequest->getText( 'catname' ); + + if ( $catname !== '' && $catname !== NULL && $catname !== false ) { + $t = Title::makeTitleSafe( NS_MAIN, $catname ); + if ( $t ) { + /** + * @fixme This can lead to hitting memory limit for very large + * categories. Ideally we would do the lookup synchronously + * during the export in a single query. + */ + $catpages = $this->getPagesFromCategory( $t ); + if ( $catpages ) $page .= "\n" . implode( "\n", $catpages ); + } } } - } - return $pageSet; -} - -/** - * Callback function to remove empty strings from the pages array. - */ -function wfFilterPage( $page ) { - return $page !== '' && $page !== null; -} - -/** - * - */ -function wfSpecialExport( $page = '' ) { - global $wgOut, $wgRequest, $wgSitename, $wgExportAllowListContributors; - global $wgExportAllowHistory, $wgExportMaxHistory; - - $curonly = true; - $doexport = false; - - if ( $wgRequest->getCheck( 'addcat' ) ) { - $page = $wgRequest->getText( 'pages' ); - $catname = $wgRequest->getText( 'catname' ); - - if ( $catname !== '' && $catname !== NULL && $catname !== false ) { - $t = Title::makeTitleSafe( NS_MAIN, $catname ); - if ( $t ) { + else if( $wgRequest->getCheck( 'addns' ) && $wgExportFromNamespaces ) { + $page = $wgRequest->getText( 'pages' ); + $nsindex = $wgRequest->getText( 'nsindex' ); + + if ( $nsindex !== '' && $nsindex !== NULL && $nsindex !== false ) { /** - * @fixme This can lead to hitting memory limit for very large - * categories. Ideally we would do the lookup synchronously - * during the export in a single query. + * Same implementation as above, so same @fixme */ - $catpages = wfExportGetPagesFromCategory( $t ); - if ( $catpages ) $page .= "\n" . implode( "\n", $catpages ); - } + $nspages = $this->getPagesFromNamespace( $nsindex ); + if ( $nspages ) $page .= "\n" . implode( "\n", $nspages ); + } } - } - else if( $wgRequest->wasPosted() && $page == '' ) { - $page = $wgRequest->getText( 'pages' ); - $curonly = $wgRequest->getCheck( 'curonly' ); - $rawOffset = $wgRequest->getVal( 'offset' ); - if( $rawOffset ) { - $offset = wfTimestamp( TS_MW, $rawOffset ); + else if( $wgRequest->wasPosted() && $par == '' ) { + $page = $wgRequest->getText( 'pages' ); + $this->curonly = $wgRequest->getCheck( 'curonly' ); + $rawOffset = $wgRequest->getVal( 'offset' ); + if( $rawOffset ) { + $offset = wfTimestamp( TS_MW, $rawOffset ); + } else { + $offset = null; + } + $limit = $wgRequest->getInt( 'limit' ); + $dir = $wgRequest->getVal( 'dir' ); + $history = array( + 'dir' => 'asc', + 'offset' => false, + 'limit' => $wgExportMaxHistory, + ); + $historyCheck = $wgRequest->getCheck( 'history' ); + if ( $this->curonly ) { + $history = WikiExporter::CURRENT; + } elseif ( !$historyCheck ) { + if ( $limit > 0 && $limit < $wgExportMaxHistory ) { + $history['limit'] = $limit; + } + if ( !is_null( $offset ) ) { + $history['offset'] = $offset; + } + if ( strtolower( $dir ) == 'desc' ) { + $history['dir'] = 'desc'; + } + } + + if( $page != '' ) $this->doExport = true; } else { - $offset = null; + // Default to current-only for GET requests + $page = $wgRequest->getText( 'pages', $par ); + $historyCheck = $wgRequest->getCheck( 'history' ); + if( $historyCheck ) { + $history = WikiExporter::FULL; + } else { + $history = WikiExporter::CURRENT; + } + + if( $page != '' ) $this->doExport = true; } - $limit = $wgRequest->getInt( 'limit' ); - $dir = $wgRequest->getVal( 'dir' ); - $history = array( - 'dir' => 'asc', - 'offset' => false, - 'limit' => $wgExportMaxHistory, - ); - $historyCheck = $wgRequest->getCheck( 'history' ); - if ( $curonly ) { + + if( !$wgExportAllowHistory ) { + // Override $history = WikiExporter::CURRENT; - } elseif ( !$historyCheck ) { - if ( $limit > 0 && $limit < $wgExportMaxHistory ) { - $history['limit'] = $limit; - } - if ( !is_null( $offset ) ) { - $history['offset'] = $offset; - } - if ( strtolower( $dir ) == 'desc' ) { - $history['dir'] = 'desc'; + } + + $list_authors = $wgRequest->getCheck( 'listauthors' ); + if ( !$this->curonly || !$wgExportAllowListContributors ) $list_authors = false ; + + if ( $this->doExport ) { + $wgOut->disable(); + // Cancel output buffering and gzipping if set + // This should provide safer streaming for pages with history + wfResetOutputBuffers(); + header( "Content-type: application/xml; charset=utf-8" ); + if( $wgRequest->getCheck( 'wpDownload' ) ) { + // Provide a sane filename suggestion + $filename = urlencode( $wgSitename . '-' . wfTimestampNow() . '.xml' ); + $wgRequest->response()->header( "Content-disposition: attachment;filename={$filename}" ); } + $this->doExport( $page, $history, $list_authors ); + return; } - - if( $page != '' ) $doexport = true; - } else { - // Default to current-only for GET requests - $page = $wgRequest->getText( 'pages', $page ); - $historyCheck = $wgRequest->getCheck( 'history' ); - if( $historyCheck ) { - $history = WikiExporter::FULL; + + $wgOut->addWikiMsg( 'exporttext' ); + + $form = Xml::openElement( 'form', array( 'method' => 'post', + 'action' => $this->getTitle()->getLocalUrl( 'action=submit' ) ) ); + $form .= Xml::inputLabel( wfMsg( 'export-addcattext' ) , 'catname', 'catname', 40 ) . ' '; + $form .= Xml::submitButton( wfMsg( 'export-addcat' ), array( 'name' => 'addcat' ) ) . '<br />'; + + if ( $wgExportFromNamespaces ) { + $form .= Xml::namespaceSelector( '', null, 'nsindex', wfMsg( 'export-addnstext' ) ) . ' '; + $form .= Xml::submitButton( wfMsg( 'export-addns' ), array( 'name' => 'addns' ) ) . '<br />'; + } + + $form .= Xml::element( 'textarea', array( 'name' => 'pages', 'cols' => 40, 'rows' => 10 ), $page, false ); + $form .= '<br />'; + + if( $wgExportAllowHistory ) { + $form .= Xml::checkLabel( wfMsg( 'exportcuronly' ), 'curonly', 'curonly', true ) . '<br />'; } else { - $history = WikiExporter::CURRENT; + $wgOut->addHTML( wfMsgExt( 'exportnohistory', 'parse' ) ); } - - if( $page != '' ) $doexport = true; + $form .= Xml::checkLabel( wfMsg( 'export-templates' ), 'templates', 'wpExportTemplates', false ) . '<br />'; + if( $wgExportMaxLinkDepth || $this->userCanOverrideExportDepth() ) { + $form .= Xml::inputLabel( wfMsg( 'export-pagelinks' ), 'pagelink-depth', 'pagelink-depth', 20, 0 ) . '<br />'; + } + // Enable this when we can do something useful exporting/importing image information. :) + //$form .= Xml::checkLabel( wfMsg( 'export-images' ), 'images', 'wpExportImages', false ) . '<br />'; + $form .= Xml::checkLabel( wfMsg( 'export-download' ), 'wpDownload', 'wpDownload', true ) . '<br />'; + + $form .= Xml::submitButton( wfMsg( 'export-submit' ), array( 'accesskey' => 's' ) ); + $form .= Xml::closeElement( 'form' ); + $wgOut->addHTML( $form ); } + + private function userCanOverrideExportDepth() { + global $wgUser; - if( !$wgExportAllowHistory ) { - // Override - $history = WikiExporter::CURRENT; + return $wgUser->isAllowed( 'override-export-depth' ); } - - $list_authors = $wgRequest->getCheck( 'listauthors' ); - if ( !$curonly || !$wgExportAllowListContributors ) $list_authors = false ; - - if ( $doexport ) { - $wgOut->disable(); - - // Cancel output buffering and gzipping if set - // This should provide safer streaming for pages with history - wfResetOutputBuffers(); - header( "Content-type: application/xml; charset=utf-8" ); - if( $wgRequest->getCheck( 'wpDownload' ) ) { - // Provide a sane filename suggestion - $filename = urlencode( $wgSitename . '-' . wfTimestampNow() . '.xml' ); - $wgRequest->response()->header( "Content-disposition: attachment;filename={$filename}" ); - } - + + /** + * Do the actual page exporting + * @param string $page User input on what page(s) to export + * @param mixed $history one of the WikiExporter history export constants + */ + private function doExport( $page, $history, $list_authors ) { + global $wgExportMaxHistory; + /* Split up the input and look up linked pages */ - $inputPages = array_filter( explode( "\n", $page ), 'wfFilterPage' ); + $inputPages = array_filter( explode( "\n", $page ), array( $this, 'filterPage' ) ); $pageSet = array_flip( $inputPages ); - - if( $wgRequest->getCheck( 'templates' ) ) { - $pageSet = wfExportGetTemplates( $inputPages, $pageSet ); + + if( $this->templates ) { + $pageSet = $this->getTemplates( $inputPages, $pageSet ); } - - /* - // Enable this when we can do something useful exporting/importing image information. :) - if( $wgRequest->getCheck( 'images' ) ) { - $pageSet = wfExportGetImages( $inputPages, $pageSet ); + + if( $linkDepth = $this->pageLinkDepth ) { + $pageSet = $this->getPageLinks( $inputPages, $pageSet, $linkDepth ); } - */ - + + /* + // Enable this when we can do something useful exporting/importing image information. :) + if( $this->images ) ) { + $pageSet = $this->getImages( $inputPages, $pageSet ); + } + */ + $pages = array_keys( $pageSet ); - + /* Ok, let's get to it... */ - if( $history == WikiExporter::CURRENT ) { $lb = false; $db = wfGetDB( DB_SLAVE ); @@ -238,65 +227,177 @@ function wfSpecialExport( $page = '' ) { set_time_limit(0); wfRestoreWarnings(); } - $exporter = new WikiExporter( $db, $history, $buffer ); - $exporter->list_authors = $list_authors ; + $exporter->list_authors = $list_authors; $exporter->openStream(); - foreach( $pages as $page ) { /* - if( $wgExportMaxHistory && !$curonly ) { - $title = Title::newFromText( $page ); - if( $title ) { - $count = Revision::countByTitle( $db, $title ); - if( $count > $wgExportMaxHistory ) { - wfDebug( __FUNCTION__ . - ": Skipped $page, $count revisions too big\n" ); - continue; - } - } - }*/ - + if( $wgExportMaxHistory && !$this->curonly ) { + $title = Title::newFromText( $page ); + if( $title ) { + $count = Revision::countByTitle( $db, $title ); + if( $count > $wgExportMaxHistory ) { + wfDebug( __FUNCTION__ . + ": Skipped $page, $count revisions too big\n" ); + continue; + } + } + }*/ #Bug 8824: Only export pages the user can read $title = Title::newFromText( $page ); if( is_null( $title ) ) continue; #TODO: perhaps output an <error> tag or something. if( !$title->userCanRead() ) continue; #TODO: perhaps output an <error> tag or something. - + $exporter->pageByTitle( $title ); } - + $exporter->closeStream(); if( $lb ) { $lb->closeAll(); } - return; } - - $self = SpecialPage::getTitleFor( 'Export' ); - $wgOut->addHTML( wfMsgExt( 'exporttext', 'parse' ) ); - - $form = Xml::openElement( 'form', array( 'method' => 'post', - 'action' => $self->getLocalUrl( 'action=submit' ) ) ); - - $form .= Xml::inputLabel( wfMsg( 'export-addcattext' ) , 'catname', 'catname', 40 ) . ' '; - $form .= Xml::submitButton( wfMsg( 'export-addcat' ), array( 'name' => 'addcat' ) ) . '<br />'; - - $form .= Xml::openElement( 'textarea', array( 'name' => 'pages', 'cols' => 40, 'rows' => 10 ) ); - $form .= htmlspecialchars( $page ); - $form .= Xml::closeElement( 'textarea' ); - $form .= '<br />'; - - if( $wgExportAllowHistory ) { - $form .= Xml::checkLabel( wfMsg( 'exportcuronly' ), 'curonly', 'curonly', true ) . '<br />'; - } else { - $wgOut->addHTML( wfMsgExt( 'exportnohistory', 'parse' ) ); + + + private function getPagesFromCategory( $title ) { + global $wgContLang; + + $name = $title->getDBkey(); + + $dbr = wfGetDB( DB_SLAVE ); + $res = $dbr->select( array('page', 'categorylinks' ), + array( 'page_namespace', 'page_title' ), + array('cl_from=page_id', 'cl_to' => $name ), + __METHOD__, array('LIMIT' => '5000')); + + $pages = array(); + while ( $row = $dbr->fetchObject( $res ) ) { + $n = $row->page_title; + if ($row->page_namespace) { + $ns = $wgContLang->getNsText( $row->page_namespace ); + $n = $ns . ':' . $n; + } + + $pages[] = $n; + } + $dbr->freeResult($res); + + return $pages; + } + + private function getPagesFromNamespace( $nsindex ) { + global $wgContLang; + + $dbr = wfGetDB( DB_SLAVE ); + $res = $dbr->select( 'page', array('page_namespace', 'page_title'), + array('page_namespace' => $nsindex), + __METHOD__, array('LIMIT' => '5000') ); + + $pages = array(); + while ( $row = $dbr->fetchObject( $res ) ) { + $n = $row->page_title; + if ($row->page_namespace) { + $ns = $wgContLang->getNsText( $row->page_namespace ); + $n = $ns . ':' . $n; + } + + $pages[] = $n; + } + $dbr->freeResult($res); + + return $pages; + } + /** + * Expand a list of pages to include templates used in those pages. + * @param $inputPages array, list of titles to look up + * @param $pageSet array, associative array indexed by titles for output + * @return array associative array index by titles + */ + private function getTemplates( $inputPages, $pageSet ) { + return $this->getLinks( $inputPages, $pageSet, + 'templatelinks', + array( 'tl_namespace AS namespace', 'tl_title AS title' ), + array( 'page_id=tl_from' ) ); + } + + /** + * Validate link depth setting, if available. + */ + private function validateLinkDepth( $depth ) { + global $wgExportMaxLinkDepth, $wgExportMaxLinkDepthLimit; + if( $depth < 0 ) { + return 0; + } + if ( !$this->userCanOverrideExportDepth() ) { + if( $depth > $wgExportMaxLinkDepth ) { + return $wgExportMaxLinkDepth; + } + } + /* + * There's a HARD CODED limit of 5 levels of recursion here to prevent a + * crazy-big export from being done by someone setting the depth + * number too high. In other words, last resort safety net. + */ + return intval( min( $depth, 5 ) ); + } + + /** Expand a list of pages to include pages linked to from that page. */ + private function getPageLinks( $inputPages, $pageSet, $depth ) { + for( $depth=$depth; $depth>0; --$depth ) { + $pageSet = $this->getLinks( $inputPages, $pageSet, 'pagelinks', + array( 'pl_namespace AS namespace', 'pl_title AS title' ), + array( 'page_id=pl_from' ) ); + $inputPages = array_keys( $pageSet ); + } + return $pageSet; + } + + /** + * Expand a list of pages to include images used in those pages. + * @param $inputPages array, list of titles to look up + * @param $pageSet array, associative array indexed by titles for output + * @return array associative array index by titles + */ + private function getImages( $inputPages, $pageSet ) { + return $this->getLinks( $inputPages, $pageSet, + 'imagelinks', + array( NS_FILE . ' AS namespace', 'il_to AS title' ), + array( 'page_id=il_from' ) ); + } + + /** + * Expand a list of pages to include items used in those pages. + * @private + */ + private function getLinks( $inputPages, $pageSet, $table, $fields, $join ) { + $dbr = wfGetDB( DB_SLAVE ); + foreach( $inputPages as $page ) { + $title = Title::newFromText( $page ); + if( $title ) { + $pageSet[$title->getPrefixedText()] = true; + /// @fixme May or may not be more efficient to batch these + /// by namespace when given multiple input pages. + $result = $dbr->select( + array( 'page', $table ), + $fields, + array_merge( $join, + array( + 'page_namespace' => $title->getNamespace(), + 'page_title' => $title->getDBKey() ) ), + __METHOD__ ); + foreach( $result as $row ) { + $template = Title::makeTitle( $row->namespace, $row->title ); + $pageSet[$template->getPrefixedText()] = true; + } + } + } + return $pageSet; + } + + /** + * Callback function to remove empty strings from the pages array. + */ + private function filterPage( $page ) { + return $page !== '' && $page !== null; } - $form .= Xml::checkLabel( wfMsg( 'export-templates' ), 'templates', 'wpExportTemplates', false ) . '<br />'; - // Enable this when we can do something useful exporting/importing image information. :) - //$form .= Xml::checkLabel( wfMsg( 'export-images' ), 'images', 'wpExportImages', false ) . '<br />'; - $form .= Xml::checkLabel( wfMsg( 'export-download' ), 'wpDownload', 'wpDownload', true ) . '<br />'; - - $form .= Xml::submitButton( wfMsg( 'export-submit' ), array( 'accesskey' => 's' ) ); - $form .= Xml::closeElement( 'form' ); - $wgOut->addHTML( $form ); } + |