3 * Implements Special:Export
5 * Copyright © 2003-2008 Brion Vibber <brion@pobox.com>
7 * This program is free software; you can redistribute it and/or modify
8 * it under the terms of the GNU General Public License as published by
9 * the Free Software Foundation; either version 2 of the License, or
10 * (at your option) any later version.
12 * This program is distributed in the hope that it will be useful,
13 * but WITHOUT ANY WARRANTY; without even the implied warranty of
14 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
15 * GNU General Public License for more details.
17 * You should have received a copy of the GNU General Public License along
18 * with this program; if not, write to the Free Software Foundation, Inc.,
19 * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
20 * http://www.gnu.org/copyleft/gpl.html
23 * @ingroup SpecialPage
27 * A special page that allows users to export pages in a XML file
29 * @ingroup SpecialPage
31 class SpecialExport
extends SpecialPage
{
32 private $curonly, $doExport, $pageLinkDepth, $templates;
35 public function __construct() {
36 parent
::__construct( 'Export' );
39 public function execute( $par ) {
41 $this->outputHeader();
42 $config = $this->getConfig();
45 $this->curonly
= true;
46 $this->doExport
= false;
47 $request = $this->getRequest();
48 $this->templates
= $request->getCheck( 'templates' );
49 $this->images
= $request->getCheck( 'images' ); // Doesn't do anything yet
50 $this->pageLinkDepth
= $this->validateLinkDepth(
51 $request->getIntOrNull( 'pagelink-depth' )
56 if ( $request->getCheck( 'addcat' ) ) {
57 $page = $request->getText( 'pages' );
58 $catname = $request->getText( 'catname' );
60 if ( $catname !== '' && $catname !== null && $catname !== false ) {
61 $t = Title
::makeTitleSafe( NS_MAIN
, $catname );
64 * @todo FIXME: This can lead to hitting memory limit for very large
65 * categories. Ideally we would do the lookup synchronously
66 * during the export in a single query.
68 $catpages = $this->getPagesFromCategory( $t );
70 $page .= "\n" . implode( "\n", $catpages );
74 } elseif ( $request->getCheck( 'addns' ) && $config->get( 'ExportFromNamespaces' ) ) {
75 $page = $request->getText( 'pages' );
76 $nsindex = $request->getText( 'nsindex', '' );
78 if ( strval( $nsindex ) !== '' ) {
80 * Same implementation as above, so same @todo
82 $nspages = $this->getPagesFromNamespace( $nsindex );
84 $page .= "\n" . implode( "\n", $nspages );
87 } elseif ( $request->getCheck( 'exportall' ) && $config->get( 'ExportAllowAll' ) ) {
88 $this->doExport
= true;
91 /* Although $page and $history are not used later on, we
92 nevertheless set them to avoid that PHP notices about using
93 undefined variables foul up our XML output (see call to
94 doExport(...) further down) */
97 } elseif ( $request->wasPosted() && $par == '' ) {
98 $page = $request->getText( 'pages' );
99 $this->curonly
= $request->getCheck( 'curonly' );
100 $rawOffset = $request->getVal( 'offset' );
103 $offset = wfTimestamp( TS_MW
, $rawOffset );
108 $maxHistory = $config->get( 'ExportMaxHistory' );
109 $limit = $request->getInt( 'limit' );
110 $dir = $request->getVal( 'dir' );
114 'limit' => $maxHistory,
116 $historyCheck = $request->getCheck( 'history' );
118 if ( $this->curonly
) {
119 $history = WikiExporter
::CURRENT
;
120 } elseif ( !$historyCheck ) {
121 if ( $limit > 0 && ( $maxHistory == 0 ||
$limit < $maxHistory ) ) {
122 $history['limit'] = $limit;
125 if ( !is_null( $offset ) ) {
126 $history['offset'] = $offset;
129 if ( strtolower( $dir ) == 'desc' ) {
130 $history['dir'] = 'desc';
135 $this->doExport
= true;
138 // Default to current-only for GET requests.
139 $page = $request->getText( 'pages', $par );
140 $historyCheck = $request->getCheck( 'history' );
142 if ( $historyCheck ) {
143 $history = WikiExporter
::FULL
;
145 $history = WikiExporter
::CURRENT
;
149 $this->doExport
= true;
153 if ( !$config->get( 'ExportAllowHistory' ) ) {
155 $history = WikiExporter
::CURRENT
;
158 $list_authors = $request->getCheck( 'listauthors' );
159 if ( !$this->curonly ||
!$config->get( 'ExportAllowListContributors' ) ) {
160 $list_authors = false;
163 if ( $this->doExport
) {
164 $this->getOutput()->disable();
166 // Cancel output buffering and gzipping if set
167 // This should provide safer streaming for pages with history
168 wfResetOutputBuffers();
169 $request->response()->header( "Content-type: application/xml; charset=utf-8" );
171 if ( $request->getCheck( 'wpDownload' ) ) {
172 // Provide a sane filename suggestion
173 $filename = urlencode( $config->get( 'Sitename' ) . '-' . wfTimestampNow() . '.xml' );
174 $request->response()->header( "Content-disposition: attachment;filename={$filename}" );
177 $this->doExport( $page, $history, $list_authors, $exportall );
182 $out = $this->getOutput();
183 $out->addWikiMsg( 'exporttext' );
185 $form = Xml
::openElement( 'form', array( 'method' => 'post',
186 'action' => $this->getPageTitle()->getLocalURL( 'action=submit' ) ) );
187 $form .= Xml
::inputLabel(
188 $this->msg( 'export-addcattext' )->text(),
193 $form .= Xml
::submitButton(
194 $this->msg( 'export-addcat' )->text(),
195 array( 'name' => 'addcat' )
198 if ( $config->get( 'ExportFromNamespaces' ) ) {
199 $form .= Html
::namespaceSelector(
201 'selected' => $nsindex,
202 'label' => $this->msg( 'export-addnstext' )->text()
206 'class' => 'namespaceselector',
209 $form .= Xml
::submitButton(
210 $this->msg( 'export-addns' )->text(),
211 array( 'name' => 'addns' )
215 if ( $config->get( 'ExportAllowAll' ) ) {
216 $form .= Xml
::checkLabel(
217 $this->msg( 'exportall' )->text(),
220 $request->wasPosted() ?
$request->getCheck( 'exportall' ) : false
224 $form .= Xml
::element(
226 array( 'name' => 'pages', 'cols' => 40, 'rows' => 10 ),
232 if ( $config->get( 'ExportAllowHistory' ) ) {
233 $form .= Xml
::checkLabel(
234 $this->msg( 'exportcuronly' )->text(),
237 $request->wasPosted() ?
$request->getCheck( 'curonly' ) : true
240 $out->addWikiMsg( 'exportnohistory' );
243 $form .= Xml
::checkLabel(
244 $this->msg( 'export-templates' )->text(),
247 $request->wasPosted() ?
$request->getCheck( 'templates' ) : false
250 if ( $config->get( 'ExportMaxLinkDepth' ) ||
$this->userCanOverrideExportDepth() ) {
251 $form .= Xml
::inputLabel(
252 $this->msg( 'export-pagelinks' )->text(),
260 /* Enable this when we can do something useful exporting/importing image information.
261 $form .= Xml::checkLabel(
262 $this->msg( 'export-images' )->text(),
268 $form .= Xml
::checkLabel(
269 $this->msg( 'export-download' )->text(),
272 $request->wasPosted() ?
$request->getCheck( 'wpDownload' ) : true
275 if ( $config->get( 'ExportAllowListContributors' ) ) {
276 $form .= Xml
::checkLabel(
277 $this->msg( 'exportlistauthors' )->text(),
280 $request->wasPosted() ?
$request->getCheck( 'listauthors' ) : false
284 $form .= Xml
::submitButton(
285 $this->msg( 'export-submit' )->text(),
286 Linker
::tooltipAndAccesskeyAttribs( 'export' )
288 $form .= Xml
::closeElement( 'form' );
290 $out->addHTML( $form );
296 private function userCanOverrideExportDepth() {
297 return $this->getUser()->isAllowed( 'override-export-depth' );
301 * Do the actual page exporting
303 * @param string $page User input on what page(s) to export
304 * @param int $history One of the WikiExporter history export constants
305 * @param bool $list_authors Whether to add distinct author list (when
306 * not returning full history)
307 * @param bool $exportall Whether to export everything
309 private function doExport( $page, $history, $list_authors, $exportall ) {
311 // If we are grabbing everything, enable full history and ignore the rest
313 $history = WikiExporter
::FULL
;
316 $pageSet = array(); // Inverted index of all pages to look up
318 // Split up and normalize input
319 foreach ( explode( "\n", $page ) as $pageName ) {
320 $pageName = trim( $pageName );
321 $title = Title
::newFromText( $pageName );
322 if ( $title && !$title->isExternal() && $title->getText() !== '' ) {
323 // Only record each page once!
324 $pageSet[$title->getPrefixedText()] = true;
328 // Set of original pages to pass on to further manipulation...
329 $inputPages = array_keys( $pageSet );
331 // Look up any linked pages if asked...
332 if ( $this->templates
) {
333 $pageSet = $this->getTemplates( $inputPages, $pageSet );
335 $linkDepth = $this->pageLinkDepth
;
337 $pageSet = $this->getPageLinks( $inputPages, $pageSet, $linkDepth );
340 // Enable this when we can do something useful exporting/importing image information.
341 // if( $this->images ) ) {
342 // $pageSet = $this->getImages( $inputPages, $pageSet );
345 $pages = array_keys( $pageSet );
347 // Normalize titles to the same format and remove dupes, see bug 17374
348 foreach ( $pages as $k => $v ) {
349 $pages[$k] = str_replace( " ", "_", $v );
352 $pages = array_unique( $pages );
355 /* Ok, let's get to it... */
356 if ( $history == WikiExporter
::CURRENT
) {
358 $db = wfGetDB( DB_SLAVE
);
359 $buffer = WikiExporter
::BUFFER
;
361 // Use an unbuffered query; histories may be very long!
362 $lb = wfGetLBFactory()->newMainLB();
363 $db = $lb->getConnection( DB_SLAVE
);
364 $buffer = WikiExporter
::STREAM
;
366 // This might take a while... :D
367 wfSuppressWarnings();
372 $exporter = new WikiExporter( $db, $history, $buffer );
373 $exporter->list_authors
= $list_authors;
374 $exporter->openStream();
377 $exporter->allPages();
379 foreach ( $pages as $page ) {
380 #Bug 8824: Only export pages the user can read
381 $title = Title
::newFromText( $page );
382 if ( is_null( $title ) ) {
383 // @todo Perhaps output an <error> tag or something.
387 if ( !$title->userCan( 'read', $this->getUser() ) ) {
388 // @todo Perhaps output an <error> tag or something.
392 $exporter->pageByTitle( $title );
396 $exporter->closeStream();
404 * @param Title $title
407 private function getPagesFromCategory( $title ) {
410 $name = $title->getDBkey();
412 $dbr = wfGetDB( DB_SLAVE
);
414 array( 'page', 'categorylinks' ),
415 array( 'page_namespace', 'page_title' ),
416 array( 'cl_from=page_id', 'cl_to' => $name ),
418 array( 'LIMIT' => '5000' )
423 foreach ( $res as $row ) {
424 $n = $row->page_title
;
425 if ( $row->page_namespace
) {
426 $ns = $wgContLang->getNsText( $row->page_namespace
);
437 * @param int $nsindex
440 private function getPagesFromNamespace( $nsindex ) {
443 $dbr = wfGetDB( DB_SLAVE
);
446 array( 'page_namespace', 'page_title' ),
447 array( 'page_namespace' => $nsindex ),
449 array( 'LIMIT' => '5000' )
454 foreach ( $res as $row ) {
455 $n = $row->page_title
;
457 if ( $row->page_namespace
) {
458 $ns = $wgContLang->getNsText( $row->page_namespace
);
469 * Expand a list of pages to include templates used in those pages.
470 * @param array $inputPages List of titles to look up
471 * @param array $pageSet Associative array indexed by titles for output
472 * @return array Associative array index by titles
474 private function getTemplates( $inputPages, $pageSet ) {
475 return $this->getLinks( $inputPages, $pageSet,
477 array( 'namespace' => 'tl_namespace', 'title' => 'tl_title' ),
478 array( 'page_id=tl_from' )
483 * Validate link depth setting, if available.
487 private function validateLinkDepth( $depth ) {
492 if ( !$this->userCanOverrideExportDepth() ) {
493 $maxLinkDepth = $this->getConfig()->get( 'ExportMaxLinkDepth' );
494 if ( $depth > $maxLinkDepth ) {
495 return $maxLinkDepth;
500 * There's a HARD CODED limit of 5 levels of recursion here to prevent a
501 * crazy-big export from being done by someone setting the depth
502 * number too high. In other words, last resort safety net.
505 return intval( min( $depth, 5 ) );
509 * Expand a list of pages to include pages linked to from that page.
510 * @param array $inputPages
511 * @param array $pageSet
515 private function getPageLinks( $inputPages, $pageSet, $depth ) {
516 // @codingStandardsIgnoreStart Squiz.WhiteSpace.SemicolonSpacing.Incorrect
517 for ( ; $depth > 0; --$depth ) {
518 // @codingStandardsIgnoreEnd
519 $pageSet = $this->getLinks(
520 $inputPages, $pageSet, 'pagelinks',
521 array( 'namespace' => 'pl_namespace', 'title' => 'pl_title' ),
522 array( 'page_id=pl_from' )
524 $inputPages = array_keys( $pageSet );
531 * Expand a list of pages to include images used in those pages.
533 * @param array $inputPages List of titles to look up
534 * @param array $pageSet Associative array indexed by titles for output
536 * @return array Associative array index by titles
538 private function getImages( $inputPages, $pageSet ) {
539 return $this->getLinks(
543 array( 'namespace' => NS_FILE
, 'title' => 'il_to' ),
544 array( 'page_id=il_from' )
549 * Expand a list of pages to include items used in those pages.
550 * @param array $inputPages Array of page titles
551 * @param array $pageSet
552 * @param string $table
553 * @param array $fields Array of field names
557 private function getLinks( $inputPages, $pageSet, $table, $fields, $join ) {
558 $dbr = wfGetDB( DB_SLAVE
);
560 foreach ( $inputPages as $page ) {
561 $title = Title
::newFromText( $page );
564 $pageSet[$title->getPrefixedText()] = true;
565 /// @todo FIXME: May or may not be more efficient to batch these
566 /// by namespace when given multiple input pages.
567 $result = $dbr->select(
568 array( 'page', $table ),
573 'page_namespace' => $title->getNamespace(),
574 'page_title' => $title->getDBkey()
580 foreach ( $result as $row ) {
581 $template = Title
::makeTitle( $row->namespace, $row->title
);
582 $pageSet[$template->getPrefixedText()] = true;
590 protected function getGroupName() {