/**
*
*/
function wfSpecialExport($page = '')
{
global $wgOut, $wgLang, $wgRequest;
if ($wgRequest->getVal('action') == 'submit') {
$page = $wgRequest->getText('pages');
$curonly = $wgRequest->getCheck('curonly');
} else {
# Pre-check the 'current version only' box in the UI
$curonly = true;
}
if ($page != '') {
$wgOut->disable();
header("Content-type: application/xml; charset=utf-8");
$pages = explode("\n", $page);
$db =& wfGetDB(DB_SLAVE);
$history = $curonly ? MW_EXPORT_CURRENT : MW_EXPORT_FULL;
$exporter = new WikiExporter($db, $history);
$exporter->openStream();
$exporter->pagesByName($pages);
$exporter->closeStream();
return;
}
$wgOut->addWikiText(wfMsg("exporttext"));
$titleObj = Title::makeTitle(NS_SPECIAL, "Export");
$action = $titleObj->escapeLocalURL('action=submit');
$wgOut->addHTML("\n<form method='post' action=\"{$action}\">\n<input type='hidden' name='action' value='submit' />\n<textarea name='pages' cols='40' rows='10'></textarea><br />\n<label><input type='checkbox' name='curonly' value='true' checked='checked' />\n" . wfMsg("exportcuronly") . "</label><br />\n<input type='submit' />\n</form>\n");
}
/**
* @covers WikiExporter::pageByTitle
*/
public function testPageByTitle()
{
global $wgContLang;
$pageTitle = 'UTPage';
$exporter = new WikiExporter($this->db, WikiExporter::FULL);
$title = Title::newFromText($pageTitle);
$sink = new DumpStringOutput();
$exporter->setOutputSink($sink);
$exporter->openStream();
$exporter->pageByTitle($title);
$exporter->closeStream();
// This throws error if invalid xml output
$xmlObject = simplexml_load_string($sink);
/**
* Check namespaces match xml
*/
$xmlNamespaces = (array) $xmlObject->siteinfo->namespaces->namespace;
$xmlNamespaces = str_replace(' ', '_', $xmlNamespaces);
unset($xmlNamespaces['@attributes']);
foreach ($xmlNamespaces as &$namespaceObject) {
if (is_object($namespaceObject)) {
$namespaceObject = '';
}
}
$actualNamespaces = (array) $wgContLang->getNamespaces();
$actualNamespaces = array_values($actualNamespaces);
$this->assertEquals($actualNamespaces, $xmlNamespaces);
// Check xml page title correct
$xmlTitle = (array) $xmlObject->page->title;
$this->assertEquals($pageTitle, $xmlTitle[0]);
// Check xml page text is not empty
$text = (array) $xmlObject->page->revision->text;
$this->assertNotEquals('', $text[0]);
}
public function testExportDemo()
{
$fname = "../../docs/export-demo.xml";
$version = WikiExporter::schemaVersion();
$dom = new DomDocument();
$dom->load($fname);
// Ensure, the demo is for the current version
$this->assertEquals($dom->documentElement->getAttribute('version'), $version, 'export-demo.xml should have the current version');
$this->assertTrue($dom->schemaValidate("../../docs/export-" . $version . ".xsd"), "schemaValidate has found an error");
}
/**
* Validates a xml file against the xsd.
*
* The validation is slow, because php has to read the xsd on each call.
*
* @param $fname string: name of file to validate
*/
protected function validateXmlFileAgainstXsd($fname)
{
$version = WikiExporter::schemaVersion();
$dom = new DomDocument();
$dom->load($fname);
try {
$this->assertTrue($dom->schemaValidate("../../docs/export-" . $version . ".xsd"), "schemaValidate has found an error");
} catch (Exception $e) {
$this->fail("xml not valid against xsd: " . $e->getMessage());
}
}
/**
* Validates a xml file against the xsd.
*
* The validation is slow, because php has to read the xsd on each call.
*
* @param $fname string: name of file to validate
*/
protected function validateXmlFileAgainstXsd($fname)
{
$version = WikiExporter::schemaVersion();
$dom = new DomDocument();
$dom->load($fname);
// Ensure, the demo is for the current version
$this->assertEquals($dom->documentElement->getAttribute('version'), $version, 'export-demo.xml should have the current version');
try {
$this->assertTrue($dom->schemaValidate("../../docs/export-" . $version . ".xsd"), "schemaValidate has found an error");
} catch (Exception $e) {
$this->fail("xml not valid against xsd: " . $e->getMessage());
}
}
/**
* @covers WikiExporter::pageByTitle
*/
public function testPageByTitle()
{
global $wgContLang;
$pageTitle = 'UTPage';
$exporter = new WikiExporter($this->db, WikiExporter::FULL);
$title = Title::newFromText($pageTitle);
ob_start();
$exporter->openStream();
$exporter->pageByTitle($title);
$exporter->closeStream();
$xmlString = ob_get_clean();
// This throws error if invalid xml output
$xmlObject = simplexml_load_string($xmlString);
/**
* Check namespaces match xml
* FIXME: PHP 5.3 support. When we don't support PHP 5.3,
* add ->namespace to object and remove from array
*/
$xmlNamespaces = (array) $xmlObject->siteinfo->namespaces;
$xmlNamespaces = str_replace(' ', '_', $xmlNamespaces['namespace']);
unset($xmlNamespaces['@attributes']);
foreach ($xmlNamespaces as &$namespaceObject) {
if (is_object($namespaceObject)) {
$namespaceObject = '';
}
}
$actualNamespaces = (array) $wgContLang->getNamespaces();
$actualNamespaces = array_values($actualNamespaces);
$this->assertEquals($actualNamespaces, $xmlNamespaces);
// Check xml page title correct
$xmlTitle = (array) $xmlObject->page->title;
$this->assertEquals($pageTitle, $xmlTitle[0]);
// Check xml page text is not empty
$text = (array) $xmlObject->page->revision->text;
$this->assertNotEquals('', $text[0]);
}
function dump($history)
{
# This shouldn't happen if on console... ;)
header('Content-type: text/html; charset=UTF-8');
# Notice messages will foul up your XML output even if they're
# relatively harmless.
ini_set('display_errors', false);
$this->startTime = wfTime();
$dbr =& wfGetDB(DB_SLAVE);
$this->maxCount = $dbr->selectField('page', 'MAX(page_id)', '', 'BackupDumper::dump');
$this->startTime = wfTime();
$db =& $this->backupDb();
$exporter = new WikiExporter($db, $history, MW_EXPORT_STREAM);
$exporter->setPageCallback(array(&$this, 'reportPage'));
$exporter->setRevisionCallback(array(&$this, 'revCount'));
$exporter->openStream();
if (is_null($this->pages)) {
$exporter->allPages();
} else {
$exporter->pagesByName($this->pages);
}
$exporter->closeStream();
$this->report(true);
}
function __construct( &$db ) {
parent::__construct( $db, WikiExporter::FULL );
}
/**
*
*/
function wfSpecialExport($page = '')
{
global $wgOut, $wgRequest, $wgExportAllowListContributors;
global $wgExportAllowHistory, $wgExportMaxHistory;
$curonly = true;
if ($wgRequest->getVal('action') == 'submit') {
$page = $wgRequest->getText('pages');
$curonly = $wgRequest->getCheck('curonly');
}
if ($wgRequest->getCheck('history')) {
$curonly = false;
}
if (!$wgExportAllowHistory) {
// Override
$curonly = true;
}
$list_authors = $wgRequest->getCheck('listauthors');
if (!$curonly || !$wgExportAllowListContributors) {
$list_authors = false;
}
if ($page != '') {
$wgOut->disable();
// Cancel output buffering and gzipping if set
// This should provide safer streaming for pages with history
while ($status = ob_get_status()) {
ob_end_clean();
if ($status['name'] == 'ob_gzhandler') {
header('Content-Encoding:');
}
}
header("Content-type: application/xml; charset=utf-8");
$pages = explode("\n", $page);
$db =& wfGetDB(DB_SLAVE);
$history = $curonly ? MW_EXPORT_CURRENT : MW_EXPORT_FULL;
$exporter = new WikiExporter($db, $history);
$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;
}
}
}
$exporter->pageByName($page);
}
$exporter->closeStream();
return;
}
$wgOut->addWikiText(wfMsg("exporttext"));
$titleObj = Title::makeTitle(NS_SPECIAL, "Export");
$form = wfOpenElement('form', array('method' => 'post', 'action' => $titleObj->getLocalUrl()));
$form .= wfOpenElement('textarea', array('name' => 'pages', 'cols' => 40, 'rows' => 10)) . '</textarea><br />';
if ($wgExportAllowHistory) {
$form .= wfCheck('curonly', true, array('value' => 'true', 'id' => 'curonly'));
$form .= wfLabel(wfMsg('exportcuronly'), 'curonly') . '<br />';
} else {
$wgOut->addWikiText(wfMsg('exportnohistory'));
}
$form .= wfHidden('action', 'submit');
$form .= wfSubmitButton(wfMsg('export-submit')) . '</form>';
$wgOut->addHtml($form);
}
/**
*
*/
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) {
/**
* @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.
*/
$catpages = wfExportGetPagesFromCategory($t);
if ($catpages) {
$page .= "\n" . implode("\n", $catpages);
}
}
}
} else {
if ($wgRequest->wasPosted() && $page == '') {
$page = $wgRequest->getText('pages');
$curonly = $wgRequest->getCheck('curonly');
$rawOffset = $wgRequest->getVal('offset');
if ($rawOffset) {
$offset = wfTimestamp(TS_MW, $rawOffset);
} else {
$offset = null;
}
$limit = $wgRequest->getInt('limit');
$dir = $wgRequest->getVal('dir');
$history = array('dir' => 'asc', 'offset' => false, 'limit' => $wgExportMaxHistory);
$historyCheck = $wgRequest->getCheck('history');
if ($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 != '') {
$doexport = true;
}
} else {
// Default to current-only for GET requests
$page = $wgRequest->getText('pages', $page);
$historyCheck = $wgRequest->getCheck('history');
if ($historyCheck) {
$history = WikiExporter::FULL;
} else {
$history = WikiExporter::CURRENT;
}
if ($page != '') {
$doexport = true;
}
}
}
if (!$wgExportAllowHistory) {
// Override
$history = WikiExporter::CURRENT;
}
$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}");
}
/* Split up the input and look up linked pages */
$inputPages = array_filter(explode("\n", $page), 'wfFilterPage');
$pageSet = array_flip($inputPages);
if ($wgRequest->getCheck('templates')) {
$pageSet = wfExportGetTemplates($inputPages, $pageSet);
}
/*
// Enable this when we can do something useful exporting/importing image information. :)
if( $wgRequest->getCheck( 'images' ) ) {
$pageSet = wfExportGetImages( $inputPages, $pageSet );
}
*/
$pages = array_keys($pageSet);
/* Ok, let's get to it... */
if ($history == WikiExporter::CURRENT) {
$lb = false;
$db = wfGetDB(DB_SLAVE);
$buffer = WikiExporter::BUFFER;
} else {
// Use an unbuffered query; histories may be very long!
$lb = wfGetLBFactory()->newMainLB();
$db = $lb->getConnection(DB_SLAVE);
$buffer = WikiExporter::STREAM;
// This might take a while... :D
wfSuppressWarnings();
set_time_limit(0);
wfRestoreWarnings();
}
$exporter = new WikiExporter($db, $history, $buffer);
$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;
}
}
}*/
#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'));
}
$form .= Xml::checkLabel(wfMsg('export-templates'), 'templates', 'wpExportTemplates', false) . '<br />';
// Enable this when we can do something useful exporting/importing image information. :)
//$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);
}
/**
*
*/
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_CATEGORY, $catname);
if ($t) {
$catpages = wfExportGetPagesFromCategory($t);
if ($catpages) {
$page .= "\n" . implode("\n", $catpages);
}
}
}
} else {
if ($wgRequest->wasPosted() && $page == '') {
$page = $wgRequest->getText('pages');
$curonly = $wgRequest->getCheck('curonly');
$rawOffset = $wgRequest->getVal('offset');
if ($rawOffset) {
$offset = wfTimestamp(TS_MW, $rawOffset);
} else {
$offset = null;
}
$limit = $wgRequest->getInt('limit');
$dir = $wgRequest->getVal('dir');
$history = array('dir' => 'asc', 'offset' => false, 'limit' => $wgExportMaxHistory);
$historyCheck = $wgRequest->getCheck('history');
if ($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 != '') {
$doexport = true;
}
} else {
// Default to current-only for GET requests
$page = $wgRequest->getText('pages', $page);
$historyCheck = $wgRequest->getCheck('history');
if ($historyCheck) {
$history = WikiExporter::FULL;
} else {
$history = WikiExporter::CURRENT;
}
if ($page != '') {
$doexport = true;
}
}
}
if (!$wgExportAllowHistory) {
// Override
$history = WikiExporter::CURRENT;
}
$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}");
}
$pages = explode("\n", $page);
$db = wfGetDB(DB_SLAVE);
$exporter = new WikiExporter($db, $history);
$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;
}
}
}*/
#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->userCan('read')) {
continue;
}
#TODO: perhaps output an <error> tag or something.
$exporter->pageByTitle($title);
}
$exporter->closeStream();
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'));
}
$form .= Xml::checkLabel(wfMsg('export-download'), 'wpDownload', 'wpDownload', true) . '<br />';
$form .= Xml::submitButton(wfMsg('export-submit'));
$form .= Xml::closeElement('form');
$wgOut->addHtml($form);
}
/**
* Do the actual page exporting
*
* @param $page String: user input on what page(s) to export
* @param $history Mixed: one of the WikiExporter history export constants
* @param $list_authors Boolean: Whether to add distinct author list (when
* not returning full history)
* @param $exportall Boolean: Whether to export everything
*/
private function doExport($page, $history, $list_authors, $exportall)
{
// If we are grabbing everything, enable full history and ignore the rest
if ($exportall) {
$history = WikiExporter::FULL;
} else {
$pageSet = array();
// Inverted index of all pages to look up
// Split up and normalize input
foreach (explode("\n", $page) as $pageName) {
$pageName = trim($pageName);
$title = Title::newFromText($pageName);
if ($title && $title->getInterwiki() == '' && $title->getText() !== '') {
// Only record each page once!
$pageSet[$title->getPrefixedText()] = true;
}
}
// Set of original pages to pass on to further manipulation...
$inputPages = array_keys($pageSet);
// Look up any linked pages if asked...
if ($this->templates) {
$pageSet = $this->getTemplates($inputPages, $pageSet);
}
$linkDepth = $this->pageLinkDepth;
if ($linkDepth) {
$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);
// Normalize titles to the same format and remove dupes, see bug 17374
foreach ($pages as $k => $v) {
$pages[$k] = str_replace(" ", "_", $v);
}
$pages = array_unique($pages);
}
/* Ok, let's get to it... */
if ($history == WikiExporter::CURRENT) {
$lb = false;
$db = wfGetDB(DB_SLAVE);
$buffer = WikiExporter::BUFFER;
} else {
// Use an unbuffered query; histories may be very long!
$lb = wfGetLBFactory()->newMainLB();
$db = $lb->getConnection(DB_SLAVE);
$buffer = WikiExporter::STREAM;
// This might take a while... :D
wfSuppressWarnings();
set_time_limit(0);
wfRestoreWarnings();
}
$exporter = new WikiExporter($db, $history, $buffer);
$exporter->list_authors = $list_authors;
$exporter->openStream();
if ($exportall) {
$exporter->allPages();
} else {
foreach ($pages as $page) {
/*
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->userCan('read', $this->getUser())) {
continue;
#TODO: perhaps output an <error> tag or something.
}
$exporter->pageByTitle($title);
}
}
$exporter->closeStream();
if ($lb) {
$lb->closeAll();
}
}
/**
* Appends an element for each page in the current pageSet with the
* most general information (id, title), plus any title normalizations
* and missing or invalid title/pageids/revids.
*/
private function outputGeneralPageInfo()
{
$pageSet = $this->getPageSet();
$result = $this->getResult();
# We don't check for a full result set here because we can't be adding
# more than 380K. The maximum revision size is in the megabyte range,
# and the maximum result size must be even higher than that.
// Title normalizations
$normValues = array();
foreach ($pageSet->getNormalizedTitles() as $rawTitleStr => $titleStr) {
$normValues[] = array('from' => $rawTitleStr, 'to' => $titleStr);
}
if (count($normValues)) {
$result->setIndexedTagName($normValues, 'n');
$result->addValue('query', 'normalized', $normValues);
}
// Interwiki titles
$intrwValues = array();
foreach ($pageSet->getInterwikiTitles() as $rawTitleStr => $interwikiStr) {
$intrwValues[] = array('title' => $rawTitleStr, 'iw' => $interwikiStr);
}
if (count($intrwValues)) {
$result->setIndexedTagName($intrwValues, 'i');
$result->addValue('query', 'interwiki', $intrwValues);
}
// Show redirect information
$redirValues = array();
foreach ($pageSet->getRedirectTitles() as $titleStrFrom => $titleStrTo) {
$redirValues[] = array('from' => strval($titleStrFrom), 'to' => $titleStrTo);
}
if (count($redirValues)) {
$result->setIndexedTagName($redirValues, 'r');
$result->addValue('query', 'redirects', $redirValues);
}
//
// Missing revision elements
//
$missingRevIDs = $pageSet->getMissingRevisionIDs();
if (count($missingRevIDs)) {
$revids = array();
foreach ($missingRevIDs as $revid) {
$revids[$revid] = array('revid' => $revid);
}
$result->setIndexedTagName($revids, 'rev');
$result->addValue('query', 'badrevids', $revids);
}
//
// Page elements
//
$pages = array();
// Report any missing titles
foreach ($pageSet->getMissingTitles() as $fakeId => $title) {
$vals = array();
ApiQueryBase::addTitleInfo($vals, $title);
$vals['missing'] = '';
$pages[$fakeId] = $vals;
}
// Report any invalid titles
foreach ($pageSet->getInvalidTitles() as $fakeId => $title) {
$pages[$fakeId] = array('title' => $title, 'invalid' => '');
}
// Report any missing page ids
foreach ($pageSet->getMissingPageIDs() as $pageid) {
$pages[$pageid] = array('pageid' => $pageid, 'missing' => '');
}
// Output general page information for found titles
foreach ($pageSet->getGoodTitles() as $pageid => $title) {
$vals = array();
$vals['pageid'] = $pageid;
ApiQueryBase::addTitleInfo($vals, $title);
$pages[$pageid] = $vals;
}
if (count($pages)) {
if ($this->params['indexpageids']) {
$pageIDs = array_keys($pages);
// json treats all map keys as strings - converting to match
$pageIDs = array_map('strval', $pageIDs);
$result->setIndexedTagName($pageIDs, 'id');
$result->addValue('query', 'pageids', $pageIDs);
}
$result->setIndexedTagName($pages, 'page');
$result->addValue('query', 'pages', $pages);
}
if ($this->params['export']) {
$exporter = new WikiExporter($this->getDB());
// WikiExporter writes to stdout, so catch its
// output with an ob
ob_start();
$exporter->openStream();
foreach (@$pageSet->getGoodTitles() as $title) {
if ($title->userCanRead()) {
$exporter->pageByTitle($title);
}
}
$exporter->closeStream();
$exportxml = ob_get_contents();
ob_end_clean();
// Don't check the size of exported stuff
// It's not continuable, so it would cause more
// problems than it'd solve
$result->disableSizeCheck();
if ($this->params['exportnowrap']) {
$result->reset();
// Raw formatter will handle this
$result->addValue(null, 'text', $exportxml);
$result->addValue(null, 'mime', 'text/xml');
} else {
$r = array();
ApiResult::setContent($r, $exportxml);
$result->addValue('query', 'export', $r);
}
$result->enableSizeCheck();
}
}
/**
* @param ApiPageSet $pageSet Pages to be exported
* @param ApiResult $result Result to output to
*/
private function doExport($pageSet, $result)
{
$exportTitles = [];
$titles = $pageSet->getGoodTitles();
if (count($titles)) {
$user = $this->getUser();
/** @var $title Title */
foreach ($titles as $title) {
if ($title->userCan('read', $user)) {
$exportTitles[] = $title;
}
}
}
$exporter = new WikiExporter($this->getDB());
$sink = new DumpStringOutput();
$exporter->setOutputSink($sink);
$exporter->openStream();
foreach ($exportTitles as $title) {
$exporter->pageByTitle($title);
}
$exporter->closeStream();
// Don't check the size of exported stuff
// It's not continuable, so it would cause more
// problems than it'd solve
if ($this->mParams['exportnowrap']) {
$result->reset();
// Raw formatter will handle this
$result->addValue(null, 'text', $sink, ApiResult::NO_SIZE_CHECK);
$result->addValue(null, 'mime', 'text/xml', ApiResult::NO_SIZE_CHECK);
} else {
$result->addValue('query', 'export', $sink, ApiResult::NO_SIZE_CHECK);
$result->addValue('query', ApiResult::META_BC_SUBELEMENTS, ['export']);
}
}
/**
* Opens the XML output stream's root "<mediawiki>" element.
* This does not include an xml directive, so is safe to include
* as a subelement in a larger XML stream. Namespace and XML Schema
* references are included.
*
* Output will be encoded in UTF-8.
*
* @return string
*/
function openStream()
{
global $wgLanguageCode;
$ver = WikiExporter::schemaVersion();
return Xml::element('mediawiki', array('xmlns' => "http://www.mediawiki.org/xml/export-{$ver}/", 'xmlns:xsi' => "http://www.w3.org/2001/XMLSchema-instance", 'xsi:schemaLocation' => "http://www.mediawiki.org/xml/export-{$ver}/ " . "http://www.mediawiki.org/xml/export-{$ver}.xsd", 'version' => $ver, 'xml:lang' => $wgLanguageCode), null) . "\n" . $this->siteInfo();
}
/**
* Do the actual page exporting
*
* @param string $page User input on what page(s) to export
* @param int $history One of the WikiExporter history export constants
* @param bool $list_authors Whether to add distinct author list (when
* not returning full history)
* @param bool $exportall Whether to export everything
*/
private function doExport($page, $history, $list_authors, $exportall)
{
// If we are grabbing everything, enable full history and ignore the rest
if ($exportall) {
$history = WikiExporter::FULL;
} else {
$pageSet = [];
// Inverted index of all pages to look up
// Split up and normalize input
foreach (explode("\n", $page) as $pageName) {
$pageName = trim($pageName);
$title = Title::newFromText($pageName);
if ($title && !$title->isExternal() && $title->getText() !== '') {
// Only record each page once!
$pageSet[$title->getPrefixedText()] = true;
}
}
// Set of original pages to pass on to further manipulation...
$inputPages = array_keys($pageSet);
// Look up any linked pages if asked...
if ($this->templates) {
$pageSet = $this->getTemplates($inputPages, $pageSet);
}
$linkDepth = $this->pageLinkDepth;
if ($linkDepth) {
$pageSet = $this->getPageLinks($inputPages, $pageSet, $linkDepth);
}
$pages = array_keys($pageSet);
// Normalize titles to the same format and remove dupes, see bug 17374
foreach ($pages as $k => $v) {
$pages[$k] = str_replace(" ", "_", $v);
}
$pages = array_unique($pages);
}
/* Ok, let's get to it... */
if ($history == WikiExporter::CURRENT) {
$lb = false;
$db = wfGetDB(DB_REPLICA);
$buffer = WikiExporter::BUFFER;
} else {
// Use an unbuffered query; histories may be very long!
$lb = wfGetLBFactory()->newMainLB();
$db = $lb->getConnection(DB_REPLICA);
$buffer = WikiExporter::STREAM;
// This might take a while... :D
MediaWiki\suppressWarnings();
set_time_limit(0);
MediaWiki\restoreWarnings();
}
$exporter = new WikiExporter($db, $history, $buffer);
$exporter->list_authors = $list_authors;
$exporter->openStream();
if ($exportall) {
$exporter->allPages();
} else {
foreach ($pages as $page) {
# Bug 8824: Only export pages the user can read
$title = Title::newFromText($page);
if (is_null($title)) {
// @todo Perhaps output an <error> tag or something.
continue;
}
if (!$title->userCan('read', $this->getUser())) {
// @todo Perhaps output an <error> tag or something.
continue;
}
$exporter->pageByTitle($title);
}
}
$exporter->closeStream();
if ($lb) {
$lb->closeAll();
}
}
/**
* 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", $page), array($this, 'filterPage'));
$pageSet = array_flip($inputPages);
if ($this->templates) {
$pageSet = $this->getTemplates($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);
$buffer = WikiExporter::BUFFER;
} else {
// Use an unbuffered query; histories may be very long!
$lb = wfGetLBFactory()->newMainLB();
$db = $lb->getConnection(DB_SLAVE);
$buffer = WikiExporter::STREAM;
// This might take a while... :D
wfSuppressWarnings();
set_time_limit(0);
wfRestoreWarnings();
}
$exporter = new WikiExporter($db, $history, $buffer);
$exporter->list_authors = $list_authors;
$exporter->openStream();
foreach ($pages as $page) {
/*
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();
}
}
/**
* @param $pageSet ApiPageSet Pages to be exported
* @param $result ApiResult Result to output to
*/
private function doExport($pageSet, $result)
{
$exportTitles = array();
$titles = $pageSet->getGoodTitles();
if (count($titles)) {
$user = $this->getUser();
/** @var $title Title */
foreach ($titles as $title) {
if ($title->userCan('read', $user)) {
$exportTitles[] = $title;
}
}
}
$exporter = new WikiExporter($this->getDB());
// WikiExporter writes to stdout, so catch its
// output with an ob
ob_start();
$exporter->openStream();
foreach ($exportTitles as $title) {
$exporter->pageByTitle($title);
}
$exporter->closeStream();
$exportxml = ob_get_contents();
ob_end_clean();
// Don't check the size of exported stuff
// It's not continuable, so it would cause more
// problems than it'd solve
$result->disableSizeCheck();
if ($this->mParams['exportnowrap']) {
$result->reset();
// Raw formatter will handle this
$result->addValue(null, 'text', $exportxml);
$result->addValue(null, 'mime', 'text/xml');
} else {
$r = array();
ApiResult::setContent($r, $exportxml);
$result->addValue('query', 'export', $r);
}
$result->enableSizeCheck();
}
/**
*
*/
function wfSpecialExport($page = '')
{
global $wgOut, $wgRequest, $wgExportAllowListContributors;
global $wgExportAllowHistory, $wgExportMaxHistory;
$curonly = true;
$fullHistory = array('dir' => 'asc', 'offset' => false, 'limit' => $wgExportMaxHistory);
if ($wgRequest->wasPosted()) {
$page = $wgRequest->getText('pages');
$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 = $wgRequest->getCheck('history');
if ($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';
}
}
} else {
// Default to current-only for GET requests
$page = $wgRequest->getText('pages', $page);
$historyCheck = $wgRequest->getCheck('history');
if ($historyCheck) {
$history = WikiExporter::FULL;
} else {
$history = WikiExporter::CURRENT;
}
}
if (!$wgExportAllowHistory) {
// Override
$history = WikiExporter::CURRENT;
}
$list_authors = $wgRequest->getCheck('listauthors');
if (!$curonly || !$wgExportAllowListContributors) {
$list_authors = false;
}
if ($page != '') {
$wgOut->disable();
// Cancel output buffering and gzipping if set
// This should provide safer streaming for pages with history
while ($status = ob_get_status()) {
ob_end_clean();
if ($status['name'] == 'ob_gzhandler') {
header('Content-Encoding:');
}
}
header("Content-type: application/xml; charset=utf-8");
$pages = explode("\n", $page);
$db =& wfGetDB(DB_SLAVE);
$exporter = new WikiExporter($db, $history);
$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;
}
}
}*/
$exporter->pageByName($page);
}
$exporter->closeStream();
return;
}
$wgOut->addWikiText(wfMsg("exporttext"));
$titleObj = Title::makeTitle(NS_SPECIAL, "Export");
$form = wfOpenElement('form', array('method' => 'post', 'action' => $titleObj->getLocalUrl()));
$form .= wfOpenElement('textarea', array('name' => 'pages', 'cols' => 40, 'rows' => 10)) . '</textarea><br />';
if ($wgExportAllowHistory) {
$form .= wfCheck('curonly', true, array('value' => 'true', 'id' => 'curonly'));
$form .= wfLabel(wfMsg('exportcuronly'), 'curonly') . '<br />';
} else {
$wgOut->addWikiText(wfMsg('exportnohistory'));
}
$form .= wfHidden('action', 'submit');
$form .= wfSubmitButton(wfMsg('export-submit')) . '</form>';
$wgOut->addHtml($form);
}