| Index: trunk/phase3/includes/specials/SpecialExport.php |
| — | — | @@ -22,378 +22,378 @@ |
| 23 | 23 | */ |
| 24 | 24 | |
| 25 | 25 | class SpecialExport extends SpecialPage { |
| 26 | | - |
| 27 | | - private $curonly, $doExport, $pageLinkDepth, $templates; |
| 28 | | - private $images; |
| 29 | | - |
| 30 | | - public function __construct() { |
| 31 | | - parent::__construct( 'Export' ); |
| 32 | | - } |
| 33 | | - |
| 34 | | - public function execute( $par ) { |
| 35 | | - global $wgOut, $wgRequest, $wgSitename, $wgExportAllowListContributors; |
| 36 | | - global $wgExportAllowHistory, $wgExportMaxHistory, $wgExportMaxLinkDepth; |
| 37 | | - |
| 38 | | - $this->setHeaders(); |
| 39 | | - $this->outputHeader(); |
| 40 | | - |
| 41 | | - // Set some variables |
| 42 | | - $this->curonly = true; |
| 43 | | - $this->doExport = false; |
| 44 | | - $this->templates = $wgRequest->getCheck( 'templates' ); |
| 45 | | - $this->images = $wgRequest->getCheck( 'images' ); // Doesn't do anything yet |
| 46 | | - $this->pageLinkDepth = $this->validateLinkDepth( |
| 47 | | - $wgRequest->getIntOrNull( 'pagelink-depth' ) ); |
| 48 | | - |
| 49 | | - if ( $wgRequest->getCheck( 'addcat' ) ) { |
| 50 | | - $page = $wgRequest->getText( 'pages' ); |
| 51 | | - $catname = $wgRequest->getText( 'catname' ); |
| 52 | | - |
| 53 | | - if ( $catname !== '' && $catname !== NULL && $catname !== false ) { |
| 54 | | - $t = Title::makeTitleSafe( NS_MAIN, $catname ); |
| 55 | | - if ( $t ) { |
| 56 | | - /** |
| 57 | | - * @fixme This can lead to hitting memory limit for very large |
| 58 | | - * categories. Ideally we would do the lookup synchronously |
| 59 | | - * during the export in a single query. |
| 60 | | - */ |
| 61 | | - $catpages = $this->getPagesFromCategory( $t ); |
| 62 | | - if ( $catpages ) $page .= "\n" . implode( "\n", $catpages ); |
| 63 | | - } |
| 64 | | - } |
| 65 | | - } |
| 66 | | - else if( $wgRequest->getCheck( 'addns' ) ) { |
| 67 | | - $page = $wgRequest->getText( 'pages' ); |
| 68 | | - $nsindex = $wgRequest->getText( 'nsindex' ); |
| 69 | | - |
| 70 | | - if ( $nsindex !== '' && $nsindex !== NULL && $nsindex !== false ) { |
| 71 | | - /** |
| 72 | | - * Same implementation as above, so same @fixme |
| 73 | | - */ |
| 74 | | - $nspages = $this->getPagesFromNamespace( $nsindex ); |
| 75 | | - if ( $nspages ) $page .= "\n" . implode( "\n", $nspages ); |
| 76 | | - } |
| 77 | | - } |
| 78 | | - else if( $wgRequest->wasPosted() && $par == '' ) { |
| 79 | | - $page = $wgRequest->getText( 'pages' ); |
| 80 | | - $this->curonly = $wgRequest->getCheck( 'curonly' ); |
| 81 | | - $rawOffset = $wgRequest->getVal( 'offset' ); |
| 82 | | - if( $rawOffset ) { |
| 83 | | - $offset = wfTimestamp( TS_MW, $rawOffset ); |
| 84 | | - } else { |
| 85 | | - $offset = null; |
| 86 | | - } |
| 87 | | - $limit = $wgRequest->getInt( 'limit' ); |
| 88 | | - $dir = $wgRequest->getVal( 'dir' ); |
| 89 | | - $history = array( |
| 90 | | - 'dir' => 'asc', |
| 91 | | - 'offset' => false, |
| 92 | | - 'limit' => $wgExportMaxHistory, |
| 93 | | - ); |
| 94 | | - $historyCheck = $wgRequest->getCheck( 'history' ); |
| 95 | | - if ( $this->curonly ) { |
| 96 | | - $history = WikiExporter::CURRENT; |
| 97 | | - } elseif ( !$historyCheck ) { |
| 98 | | - if ( $limit > 0 && $limit < $wgExportMaxHistory ) { |
| 99 | | - $history['limit'] = $limit; |
| 100 | | - } |
| 101 | | - if ( !is_null( $offset ) ) { |
| 102 | | - $history['offset'] = $offset; |
| 103 | | - } |
| 104 | | - if ( strtolower( $dir ) == 'desc' ) { |
| 105 | | - $history['dir'] = 'desc'; |
| 106 | | - } |
| 107 | | - } |
| 108 | | - |
| 109 | | - if( $page != '' ) $this->doExport = true; |
| 110 | | - } else { |
| 111 | | - // Default to current-only for GET requests |
| 112 | | - $page = $wgRequest->getText( 'pages', $par ); |
| 113 | | - $historyCheck = $wgRequest->getCheck( 'history' ); |
| 114 | | - if( $historyCheck ) { |
| 115 | | - $history = WikiExporter::FULL; |
| 116 | | - } else { |
| 117 | | - $history = WikiExporter::CURRENT; |
| 118 | | - } |
| 119 | | - |
| 120 | | - if( $page != '' ) $this->doExport = true; |
| 121 | | - } |
| 122 | | - |
| 123 | | - if( !$wgExportAllowHistory ) { |
| 124 | | - // Override |
| 125 | | - $history = WikiExporter::CURRENT; |
| 126 | | - } |
| 127 | | - |
| 128 | | - $list_authors = $wgRequest->getCheck( 'listauthors' ); |
| 129 | | - if ( !$this->curonly || !$wgExportAllowListContributors ) $list_authors = false ; |
| 130 | | - |
| 131 | | - if ( $this->doExport ) { |
| 132 | | - $wgOut->disable(); |
| 133 | | - // Cancel output buffering and gzipping if set |
| 134 | | - // This should provide safer streaming for pages with history |
| 135 | | - wfResetOutputBuffers(); |
| 136 | | - header( "Content-type: application/xml; charset=utf-8" ); |
| 137 | | - if( $wgRequest->getCheck( 'wpDownload' ) ) { |
| 138 | | - // Provide a sane filename suggestion |
| 139 | | - $filename = urlencode( $wgSitename . '-' . wfTimestampNow() . '.xml' ); |
| 140 | | - $wgRequest->response()->header( "Content-disposition: attachment;filename={$filename}" ); |
| 141 | | - } |
| 142 | | - $this->doExport( $page, $history, $list_authors ); |
| 143 | | - return; |
| 144 | | - } |
| 145 | | - |
| 146 | | - $wgOut->addWikiMsg( 'exporttext' ); |
| 147 | | - |
| 148 | | - $form = Xml::openElement( 'form', array( 'method' => 'post', |
| 149 | | - 'action' => $this->getTitle()->getLocalUrl( 'action=submit' ) ) ); |
| 150 | | - $form .= Xml::inputLabel( wfMsg( 'export-addcattext' ) , 'catname', 'catname', 40 ) . ' '; |
| 151 | | - $form .= Xml::submitButton( wfMsg( 'export-addcat' ), array( 'name' => 'addcat' ) ) . '<br />'; |
| 152 | | - |
| 153 | | - $form .= Xml::namespaceSelector( '', null, 'nsindex', wfMsg( 'export-addnstext' ) ) . ' '; |
| 154 | | - $form .= Xml::submitButton( wfMsg( 'export-addns' ), array( 'name' => 'addns' ) ) . '<br />'; |
| 155 | | - |
| 156 | | - $form .= Xml::element( 'textarea', array( 'name' => 'pages', 'cols' => 40, 'rows' => 10 ), $page, false ); |
| 157 | | - $form .= '<br />'; |
| 158 | | - |
| 159 | | - if( $wgExportAllowHistory ) { |
| 160 | | - $form .= Xml::checkLabel( wfMsg( 'exportcuronly' ), 'curonly', 'curonly', true ) . '<br />'; |
| 161 | | - } else { |
| 162 | | - $wgOut->addHTML( wfMsgExt( 'exportnohistory', 'parse' ) ); |
| 163 | | - } |
| 164 | | - $form .= Xml::checkLabel( wfMsg( 'export-templates' ), 'templates', 'wpExportTemplates', false ) . '<br />'; |
| 165 | | - if( $wgExportMaxLinkDepth || $this->userCanOverrideExportDepth() ) { |
| 166 | | - $form .= Xml::inputLabel( wfMsg( 'export-pagelinks' ), 'pagelink-depth', 'pagelink-depth', 20, 0 ) . '<br />'; |
| 167 | | - } |
| 168 | | - // Enable this when we can do something useful exporting/importing image information. :) |
| 169 | | - //$form .= Xml::checkLabel( wfMsg( 'export-images' ), 'images', 'wpExportImages', false ) . '<br />'; |
| 170 | | - $form .= Xml::checkLabel( wfMsg( 'export-download' ), 'wpDownload', 'wpDownload', true ) . '<br />'; |
| 171 | | - |
| 172 | | - $form .= Xml::submitButton( wfMsg( 'export-submit' ), array( 'accesskey' => 's' ) ); |
| 173 | | - $form .= Xml::closeElement( 'form' ); |
| 174 | | - $wgOut->addHTML( $form ); |
| 175 | | - } |
| 176 | | - |
| 177 | | - private function userCanOverrideExportDepth() { |
| 178 | | - global $wgUser; |
| 179 | | - |
| 180 | | - return in_array( 'override-export-depth', $wgUser->getRights()); |
| 181 | | - } |
| 182 | | - |
| 183 | | - /** |
| 184 | | - * Do the actual page exporting |
| 185 | | - * @param string $page User input on what page(s) to export |
| 186 | | - * @param mixed $history one of the WikiExporter history export constants |
| 187 | | - */ |
| 188 | | - private function doExport( $page, $history, $list_authors ) { |
| 189 | | - global $wgExportMaxHistory; |
| 190 | | - |
| 191 | | - /* Split up the input and look up linked pages */ |
| 192 | | - $inputPages = array_filter( explode( "\n", $page ), array( $this, 'filterPage' ) ); |
| 193 | | - $pageSet = array_flip( $inputPages ); |
| 194 | | - |
| 195 | | - if( $this->templates ) { |
| 196 | | - $pageSet = $this->getTemplates( $inputPages, $pageSet ); |
| 197 | | - } |
| 198 | | - |
| 199 | | - if( $linkDepth = $this->pageLinkDepth ) { |
| 200 | | - $pageSet = $this->getPageLinks( $inputPages, $pageSet, $linkDepth ); |
| 201 | | - } |
| 202 | | - |
| 203 | | - /* |
| 204 | | - // Enable this when we can do something useful exporting/importing image information. :) |
| 205 | | - if( $this->images ) ) { |
| 206 | | - $pageSet = $this->getImages( $inputPages, $pageSet ); |
| 207 | | - } |
| 208 | | - */ |
| 209 | | - |
| 210 | | - $pages = array_keys( $pageSet ); |
| 211 | | - |
| 212 | | - /* Ok, let's get to it... */ |
| 213 | | - if( $history == WikiExporter::CURRENT ) { |
| 214 | | - $lb = false; |
| 215 | | - $db = wfGetDB( DB_SLAVE ); |
| 216 | | - $buffer = WikiExporter::BUFFER; |
| 217 | | - } else { |
| 218 | | - // Use an unbuffered query; histories may be very long! |
| 219 | | - $lb = wfGetLBFactory()->newMainLB(); |
| 220 | | - $db = $lb->getConnection( DB_SLAVE ); |
| 221 | | - $buffer = WikiExporter::STREAM; |
| 222 | | - |
| 223 | | - // This might take a while... :D |
| 224 | | - wfSuppressWarnings(); |
| 225 | | - set_time_limit(0); |
| 226 | | - wfRestoreWarnings(); |
| 227 | | - } |
| 228 | | - $exporter = new WikiExporter( $db, $history, $buffer ); |
| 229 | | - $exporter->list_authors = $list_authors; |
| 230 | | - $exporter->openStream(); |
| 231 | | - foreach( $pages as $page ) { |
| 232 | | - /* |
| 233 | | - if( $wgExportMaxHistory && !$this->curonly ) { |
| 234 | | - $title = Title::newFromText( $page ); |
| 235 | | - if( $title ) { |
| 236 | | - $count = Revision::countByTitle( $db, $title ); |
| 237 | | - if( $count > $wgExportMaxHistory ) { |
| 238 | | - wfDebug( __FUNCTION__ . |
| 239 | | - ": Skipped $page, $count revisions too big\n" ); |
| 240 | | - continue; |
| 241 | | - } |
| 242 | | - } |
| 243 | | - }*/ |
| 244 | | - #Bug 8824: Only export pages the user can read |
| 245 | | - $title = Title::newFromText( $page ); |
| 246 | | - if( is_null( $title ) ) continue; #TODO: perhaps output an <error> tag or something. |
| 247 | | - if( !$title->userCanRead() ) continue; #TODO: perhaps output an <error> tag or something. |
| 248 | | - |
| 249 | | - $exporter->pageByTitle( $title ); |
| 250 | | - } |
| 251 | | - |
| 252 | | - $exporter->closeStream(); |
| 253 | | - if( $lb ) { |
| 254 | | - $lb->closeAll(); |
| 255 | | - } |
| 256 | | - } |
| 257 | | - |
| 258 | | - |
| 259 | | - private function getPagesFromCategory( $title ) { |
| 260 | | - global $wgContLang; |
| 261 | | - |
| 262 | | - $name = $title->getDBkey(); |
| 263 | | - |
| 264 | | - $dbr = wfGetDB( DB_SLAVE ); |
| 265 | | - $res = $dbr->select( array('page', 'categorylinks' ), |
| 266 | | - array( 'page_namespace', 'page_title' ), |
| 267 | | - array('cl_from=page_id', 'cl_to' => $name ), |
| 268 | | - __METHOD__, array('LIMIT' => '5000')); |
| 269 | | - |
| 270 | | - $pages = array(); |
| 271 | | - while ( $row = $dbr->fetchObject( $res ) ) { |
| 272 | | - $n = $row->page_title; |
| 273 | | - if ($row->page_namespace) { |
| 274 | | - $ns = $wgContLang->getNsText( $row->page_namespace ); |
| 275 | | - $n = $ns . ':' . $n; |
| 276 | | - } |
| 277 | | - |
| 278 | | - $pages[] = $n; |
| 279 | | - } |
| 280 | | - $dbr->freeResult($res); |
| 281 | | - |
| 282 | | - return $pages; |
| 283 | | - } |
| 284 | | - |
| 285 | | - private function getPagesFromNamespace( $nsindex ) { |
| 286 | | - global $wgContLang; |
| 287 | | - |
| 288 | | - $dbr = wfGetDB( DB_SLAVE ); |
| 289 | | - $res = $dbr->select( 'page', array('page_namespace', 'page_title'), |
| 290 | | - array('page_namespace' => $nsindex), |
| 291 | | - __METHOD__, array('LIMIT' => '5000') ); |
| 292 | | - |
| 293 | | - $pages = array(); |
| 294 | | - while ( $row = $dbr->fetchObject( $res ) ) { |
| 295 | | - $n = $row->page_title; |
| 296 | | - if ($row->page_namespace) { |
| 297 | | - $ns = $wgContLang->getNsText( $row->page_namespace ); |
| 298 | | - $n = $ns . ':' . $n; |
| 299 | | - } |
| 300 | | - |
| 301 | | - $pages[] = $n; |
| 302 | | - } |
| 303 | | - $dbr->freeResult($res); |
| 304 | | - |
| 305 | | - return $pages; |
| 306 | | - } |
| 307 | | - /** |
| 308 | | - * Expand a list of pages to include templates used in those pages. |
| 309 | | - * @param $inputPages array, list of titles to look up |
| 310 | | - * @param $pageSet array, associative array indexed by titles for output |
| 311 | | - * @return array associative array index by titles |
| 312 | | - */ |
| 313 | | - private function getTemplates( $inputPages, $pageSet ) { |
| 314 | | - return $this->getLinks( $inputPages, $pageSet, |
| 315 | | - 'templatelinks', |
| 316 | | - array( 'tl_namespace AS namespace', 'tl_title AS title' ), |
| 317 | | - array( 'page_id=tl_from' ) ); |
| 318 | | - } |
| 319 | | - |
| 320 | | - /** |
| 321 | | - * Validate link depth setting, if available. |
| 322 | | - */ |
| 323 | | - private function validateLinkDepth( $depth ) { |
| 324 | | - global $wgExportMaxLinkDepth, $wgExportMaxLinkDepthLimit; |
| 325 | | - if( $depth < 0 ) { |
| 326 | | - return 0; |
| 327 | | - } |
| | 26 | + |
| | 27 | + private $curonly, $doExport, $pageLinkDepth, $templates; |
| | 28 | + private $images; |
| | 29 | + |
| | 30 | + public function __construct() { |
| | 31 | + parent::__construct( 'Export' ); |
| | 32 | + } |
| | 33 | + |
| | 34 | + public function execute( $par ) { |
| | 35 | + global $wgOut, $wgRequest, $wgSitename, $wgExportAllowListContributors; |
| | 36 | + global $wgExportAllowHistory, $wgExportMaxHistory, $wgExportMaxLinkDepth; |
| | 37 | + |
| | 38 | + $this->setHeaders(); |
| | 39 | + $this->outputHeader(); |
| | 40 | + |
| | 41 | + // Set some variables |
| | 42 | + $this->curonly = true; |
| | 43 | + $this->doExport = false; |
| | 44 | + $this->templates = $wgRequest->getCheck( 'templates' ); |
| | 45 | + $this->images = $wgRequest->getCheck( 'images' ); // Doesn't do anything yet |
| | 46 | + $this->pageLinkDepth = $this->validateLinkDepth( |
| | 47 | + $wgRequest->getIntOrNull( 'pagelink-depth' ) ); |
| | 48 | + |
| | 49 | + if ( $wgRequest->getCheck( 'addcat' ) ) { |
| | 50 | + $page = $wgRequest->getText( 'pages' ); |
| | 51 | + $catname = $wgRequest->getText( 'catname' ); |
| | 52 | + |
| | 53 | + if ( $catname !== '' && $catname !== NULL && $catname !== false ) { |
| | 54 | + $t = Title::makeTitleSafe( NS_MAIN, $catname ); |
| | 55 | + if ( $t ) { |
| | 56 | + /** |
| | 57 | + * @fixme This can lead to hitting memory limit for very large |
| | 58 | + * categories. Ideally we would do the lookup synchronously |
| | 59 | + * during the export in a single query. |
| | 60 | + */ |
| | 61 | + $catpages = $this->getPagesFromCategory( $t ); |
| | 62 | + if ( $catpages ) $page .= "\n" . implode( "\n", $catpages ); |
| | 63 | + } |
| | 64 | + } |
| | 65 | + } |
| | 66 | + else if( $wgRequest->getCheck( 'addns' ) ) { |
| | 67 | + $page = $wgRequest->getText( 'pages' ); |
| | 68 | + $nsindex = $wgRequest->getText( 'nsindex' ); |
| | 69 | + |
| | 70 | + if ( $nsindex !== '' && $nsindex !== NULL && $nsindex !== false ) { |
| | 71 | + /** |
| | 72 | + * Same implementation as above, so same @fixme |
| | 73 | + */ |
| | 74 | + $nspages = $this->getPagesFromNamespace( $nsindex ); |
| | 75 | + if ( $nspages ) $page .= "\n" . implode( "\n", $nspages ); |
| | 76 | + } |
| | 77 | + } |
| | 78 | + else if( $wgRequest->wasPosted() && $par == '' ) { |
| | 79 | + $page = $wgRequest->getText( 'pages' ); |
| | 80 | + $this->curonly = $wgRequest->getCheck( 'curonly' ); |
| | 81 | + $rawOffset = $wgRequest->getVal( 'offset' ); |
| | 82 | + if( $rawOffset ) { |
| | 83 | + $offset = wfTimestamp( TS_MW, $rawOffset ); |
| | 84 | + } else { |
| | 85 | + $offset = null; |
| | 86 | + } |
| | 87 | + $limit = $wgRequest->getInt( 'limit' ); |
| | 88 | + $dir = $wgRequest->getVal( 'dir' ); |
| | 89 | + $history = array( |
| | 90 | + 'dir' => 'asc', |
| | 91 | + 'offset' => false, |
| | 92 | + 'limit' => $wgExportMaxHistory, |
| | 93 | + ); |
| | 94 | + $historyCheck = $wgRequest->getCheck( 'history' ); |
| | 95 | + if ( $this->curonly ) { |
| | 96 | + $history = WikiExporter::CURRENT; |
| | 97 | + } elseif ( !$historyCheck ) { |
| | 98 | + if ( $limit > 0 && $limit < $wgExportMaxHistory ) { |
| | 99 | + $history['limit'] = $limit; |
| | 100 | + } |
| | 101 | + if ( !is_null( $offset ) ) { |
| | 102 | + $history['offset'] = $offset; |
| | 103 | + } |
| | 104 | + if ( strtolower( $dir ) == 'desc' ) { |
| | 105 | + $history['dir'] = 'desc'; |
| | 106 | + } |
| | 107 | + } |
| | 108 | + |
| | 109 | + if( $page != '' ) $this->doExport = true; |
| | 110 | + } else { |
| | 111 | + // Default to current-only for GET requests |
| | 112 | + $page = $wgRequest->getText( 'pages', $par ); |
| | 113 | + $historyCheck = $wgRequest->getCheck( 'history' ); |
| | 114 | + if( $historyCheck ) { |
| | 115 | + $history = WikiExporter::FULL; |
| | 116 | + } else { |
| | 117 | + $history = WikiExporter::CURRENT; |
| | 118 | + } |
| | 119 | + |
| | 120 | + if( $page != '' ) $this->doExport = true; |
| | 121 | + } |
| | 122 | + |
| | 123 | + if( !$wgExportAllowHistory ) { |
| | 124 | + // Override |
| | 125 | + $history = WikiExporter::CURRENT; |
| | 126 | + } |
| | 127 | + |
| | 128 | + $list_authors = $wgRequest->getCheck( 'listauthors' ); |
| | 129 | + if ( !$this->curonly || !$wgExportAllowListContributors ) $list_authors = false ; |
| | 130 | + |
| | 131 | + if ( $this->doExport ) { |
| | 132 | + $wgOut->disable(); |
| | 133 | + // Cancel output buffering and gzipping if set |
| | 134 | + // This should provide safer streaming for pages with history |
| | 135 | + wfResetOutputBuffers(); |
| | 136 | + header( "Content-type: application/xml; charset=utf-8" ); |
| | 137 | + if( $wgRequest->getCheck( 'wpDownload' ) ) { |
| | 138 | + // Provide a sane filename suggestion |
| | 139 | + $filename = urlencode( $wgSitename . '-' . wfTimestampNow() . '.xml' ); |
| | 140 | + $wgRequest->response()->header( "Content-disposition: attachment;filename={$filename}" ); |
| | 141 | + } |
| | 142 | + $this->doExport( $page, $history, $list_authors ); |
| | 143 | + return; |
| | 144 | + } |
| | 145 | + |
| | 146 | + $wgOut->addWikiMsg( 'exporttext' ); |
| | 147 | + |
| | 148 | + $form = Xml::openElement( 'form', array( 'method' => 'post', |
| | 149 | + 'action' => $this->getTitle()->getLocalUrl( 'action=submit' ) ) ); |
| | 150 | + $form .= Xml::inputLabel( wfMsg( 'export-addcattext' ) , 'catname', 'catname', 40 ) . ' '; |
| | 151 | + $form .= Xml::submitButton( wfMsg( 'export-addcat' ), array( 'name' => 'addcat' ) ) . '<br />'; |
| | 152 | + |
| | 153 | + $form .= Xml::namespaceSelector( '', null, 'nsindex', wfMsg( 'export-addnstext' ) ) . ' '; |
| | 154 | + $form .= Xml::submitButton( wfMsg( 'export-addns' ), array( 'name' => 'addns' ) ) . '<br />'; |
| | 155 | + |
| | 156 | + $form .= Xml::element( 'textarea', array( 'name' => 'pages', 'cols' => 40, 'rows' => 10 ), $page, false ); |
| | 157 | + $form .= '<br />'; |
| | 158 | + |
| | 159 | + if( $wgExportAllowHistory ) { |
| | 160 | + $form .= Xml::checkLabel( wfMsg( 'exportcuronly' ), 'curonly', 'curonly', true ) . '<br />'; |
| | 161 | + } else { |
| | 162 | + $wgOut->addHTML( wfMsgExt( 'exportnohistory', 'parse' ) ); |
| | 163 | + } |
| | 164 | + $form .= Xml::checkLabel( wfMsg( 'export-templates' ), 'templates', 'wpExportTemplates', false ) . '<br />'; |
| | 165 | + if( $wgExportMaxLinkDepth || $this->userCanOverrideExportDepth() ) { |
| | 166 | + $form .= Xml::inputLabel( wfMsg( 'export-pagelinks' ), 'pagelink-depth', 'pagelink-depth', 20, 0 ) . '<br />'; |
| | 167 | + } |
| | 168 | + // Enable this when we can do something useful exporting/importing image information. :) |
| | 169 | + //$form .= Xml::checkLabel( wfMsg( 'export-images' ), 'images', 'wpExportImages', false ) . '<br />'; |
| | 170 | + $form .= Xml::checkLabel( wfMsg( 'export-download' ), 'wpDownload', 'wpDownload', true ) . '<br />'; |
| | 171 | + |
| | 172 | + $form .= Xml::submitButton( wfMsg( 'export-submit' ), array( 'accesskey' => 's' ) ); |
| | 173 | + $form .= Xml::closeElement( 'form' ); |
| | 174 | + $wgOut->addHTML( $form ); |
| | 175 | + } |
| | 176 | + |
| | 177 | + private function userCanOverrideExportDepth() { |
| | 178 | + global $wgUser; |
| | 179 | + |
| | 180 | + return in_array( 'override-export-depth', $wgUser->getRights()); |
| | 181 | + } |
| | 182 | + |
| | 183 | + /** |
| | 184 | + * Do the actual page exporting |
| | 185 | + * @param string $page User input on what page(s) to export |
| | 186 | + * @param mixed $history one of the WikiExporter history export constants |
| | 187 | + */ |
| | 188 | + private function doExport( $page, $history, $list_authors ) { |
| | 189 | + global $wgExportMaxHistory; |
| | 190 | + |
| | 191 | + /* Split up the input and look up linked pages */ |
| | 192 | + $inputPages = array_filter( explode( "\n", $page ), array( $this, 'filterPage' ) ); |
| | 193 | + $pageSet = array_flip( $inputPages ); |
| | 194 | + |
| | 195 | + if( $this->templates ) { |
| | 196 | + $pageSet = $this->getTemplates( $inputPages, $pageSet ); |
| | 197 | + } |
| | 198 | + |
| | 199 | + if( $linkDepth = $this->pageLinkDepth ) { |
| | 200 | + $pageSet = $this->getPageLinks( $inputPages, $pageSet, $linkDepth ); |
| | 201 | + } |
| | 202 | + |
| | 203 | + /* |
| | 204 | + // Enable this when we can do something useful exporting/importing image information. :) |
| | 205 | + if( $this->images ) ) { |
| | 206 | + $pageSet = $this->getImages( $inputPages, $pageSet ); |
| | 207 | + } |
| | 208 | + */ |
| | 209 | + |
| | 210 | + $pages = array_keys( $pageSet ); |
| | 211 | + |
| | 212 | + /* Ok, let's get to it... */ |
| | 213 | + if( $history == WikiExporter::CURRENT ) { |
| | 214 | + $lb = false; |
| | 215 | + $db = wfGetDB( DB_SLAVE ); |
| | 216 | + $buffer = WikiExporter::BUFFER; |
| | 217 | + } else { |
| | 218 | + // Use an unbuffered query; histories may be very long! |
| | 219 | + $lb = wfGetLBFactory()->newMainLB(); |
| | 220 | + $db = $lb->getConnection( DB_SLAVE ); |
| | 221 | + $buffer = WikiExporter::STREAM; |
| | 222 | + |
| | 223 | + // This might take a while... :D |
| | 224 | + wfSuppressWarnings(); |
| | 225 | + set_time_limit(0); |
| | 226 | + wfRestoreWarnings(); |
| | 227 | + } |
| | 228 | + $exporter = new WikiExporter( $db, $history, $buffer ); |
| | 229 | + $exporter->list_authors = $list_authors; |
| | 230 | + $exporter->openStream(); |
| | 231 | + foreach( $pages as $page ) { |
| | 232 | + /* |
| | 233 | + if( $wgExportMaxHistory && !$this->curonly ) { |
| | 234 | + $title = Title::newFromText( $page ); |
| | 235 | + if( $title ) { |
| | 236 | + $count = Revision::countByTitle( $db, $title ); |
| | 237 | + if( $count > $wgExportMaxHistory ) { |
| | 238 | + wfDebug( __FUNCTION__ . |
| | 239 | + ": Skipped $page, $count revisions too big\n" ); |
| | 240 | + continue; |
| | 241 | + } |
| | 242 | + } |
| | 243 | + }*/ |
| | 244 | + #Bug 8824: Only export pages the user can read |
| | 245 | + $title = Title::newFromText( $page ); |
| | 246 | + if( is_null( $title ) ) continue; #TODO: perhaps output an <error> tag or something. |
| | 247 | + if( !$title->userCanRead() ) continue; #TODO: perhaps output an <error> tag or something. |
| | 248 | + |
| | 249 | + $exporter->pageByTitle( $title ); |
| | 250 | + } |
| | 251 | + |
| | 252 | + $exporter->closeStream(); |
| | 253 | + if( $lb ) { |
| | 254 | + $lb->closeAll(); |
| | 255 | + } |
| | 256 | + } |
| | 257 | + |
| | 258 | + |
| | 259 | + private function getPagesFromCategory( $title ) { |
| | 260 | + global $wgContLang; |
| | 261 | + |
| | 262 | + $name = $title->getDBkey(); |
| | 263 | + |
| | 264 | + $dbr = wfGetDB( DB_SLAVE ); |
| | 265 | + $res = $dbr->select( array('page', 'categorylinks' ), |
| | 266 | + array( 'page_namespace', 'page_title' ), |
| | 267 | + array('cl_from=page_id', 'cl_to' => $name ), |
| | 268 | + __METHOD__, array('LIMIT' => '5000')); |
| | 269 | + |
| | 270 | + $pages = array(); |
| | 271 | + while ( $row = $dbr->fetchObject( $res ) ) { |
| | 272 | + $n = $row->page_title; |
| | 273 | + if ($row->page_namespace) { |
| | 274 | + $ns = $wgContLang->getNsText( $row->page_namespace ); |
| | 275 | + $n = $ns . ':' . $n; |
| | 276 | + } |
| | 277 | + |
| | 278 | + $pages[] = $n; |
| | 279 | + } |
| | 280 | + $dbr->freeResult($res); |
| | 281 | + |
| | 282 | + return $pages; |
| | 283 | + } |
| | 284 | + |
| | 285 | + private function getPagesFromNamespace( $nsindex ) { |
| | 286 | + global $wgContLang; |
| | 287 | + |
| | 288 | + $dbr = wfGetDB( DB_SLAVE ); |
| | 289 | + $res = $dbr->select( 'page', array('page_namespace', 'page_title'), |
| | 290 | + array('page_namespace' => $nsindex), |
| | 291 | + __METHOD__, array('LIMIT' => '5000') ); |
| | 292 | + |
| | 293 | + $pages = array(); |
| | 294 | + while ( $row = $dbr->fetchObject( $res ) ) { |
| | 295 | + $n = $row->page_title; |
| | 296 | + if ($row->page_namespace) { |
| | 297 | + $ns = $wgContLang->getNsText( $row->page_namespace ); |
| | 298 | + $n = $ns . ':' . $n; |
| | 299 | + } |
| | 300 | + |
| | 301 | + $pages[] = $n; |
| | 302 | + } |
| | 303 | + $dbr->freeResult($res); |
| | 304 | + |
| | 305 | + return $pages; |
| | 306 | + } |
| | 307 | + /** |
| | 308 | + * Expand a list of pages to include templates used in those pages. |
| | 309 | + * @param $inputPages array, list of titles to look up |
| | 310 | + * @param $pageSet array, associative array indexed by titles for output |
| | 311 | + * @return array associative array index by titles |
| | 312 | + */ |
| | 313 | + private function getTemplates( $inputPages, $pageSet ) { |
| | 314 | + return $this->getLinks( $inputPages, $pageSet, |
| | 315 | + 'templatelinks', |
| | 316 | + array( 'tl_namespace AS namespace', 'tl_title AS title' ), |
| | 317 | + array( 'page_id=tl_from' ) ); |
| | 318 | + } |
| | 319 | + |
| | 320 | + /** |
| | 321 | + * Validate link depth setting, if available. |
| | 322 | + */ |
| | 323 | + private function validateLinkDepth( $depth ) { |
| | 324 | + global $wgExportMaxLinkDepth, $wgExportMaxLinkDepthLimit; |
| | 325 | + if( $depth < 0 ) { |
| | 326 | + return 0; |
| | 327 | + } |
| 328 | 328 | if ( !$this->userCanOverrideExportDepth() ) { |
| 329 | | - if( $depth > $wgExportMaxLinkDepth ) { |
| 330 | | - return $wgExportMaxLinkDepth; |
| 331 | | - } |
| | 329 | + if( $depth > $wgExportMaxLinkDepth ) { |
| | 330 | + return $wgExportMaxLinkDepth; |
| | 331 | + } |
| 332 | 332 | } |
| 333 | 333 | /* |
| 334 | 334 | * There's a HARD CODED limit of 5 levels of recursion here to prevent a |
| 335 | 335 | * crazy-big export from being done by someone setting the depth |
| 336 | 336 | * number too high. In other words, last resort safety net. |
| 337 | 337 | */ |
| 338 | | - return intval( min( $depth, 5 ) ); |
| 339 | | - } |
| 340 | | - |
| 341 | | - /** Expand a list of pages to include pages linked to from that page. */ |
| 342 | | - private function getPageLinks( $inputPages, $pageSet, $depth ) { |
| 343 | | - for( $depth=$depth; $depth>0; --$depth ) { |
| 344 | | - $pageSet = $this->getLinks( $inputPages, $pageSet, 'pagelinks', |
| 345 | | - array( 'pl_namespace AS namespace', 'pl_title AS title' ), |
| 346 | | - array( 'page_id=pl_from' ) ); |
| 347 | | - } |
| 348 | | - return $pageSet; |
| 349 | | - } |
| 350 | | - |
| 351 | | - /** |
| 352 | | - * Expand a list of pages to include images used in those pages. |
| 353 | | - * @param $inputPages array, list of titles to look up |
| 354 | | - * @param $pageSet array, associative array indexed by titles for output |
| 355 | | - * @return array associative array index by titles |
| 356 | | - */ |
| 357 | | - private function getImages( $inputPages, $pageSet ) { |
| 358 | | - return $this->getLinks( $inputPages, $pageSet, |
| 359 | | - 'imagelinks', |
| 360 | | - array( NS_FILE . ' AS namespace', 'il_to AS title' ), |
| 361 | | - array( 'page_id=il_from' ) ); |
| 362 | | - } |
| 363 | | - |
| 364 | | - /** |
| 365 | | - * Expand a list of pages to include items used in those pages. |
| 366 | | - * @private |
| 367 | | - */ |
| 368 | | - private function getLinks( $inputPages, $pageSet, $table, $fields, $join ) { |
| 369 | | - $dbr = wfGetDB( DB_SLAVE ); |
| 370 | | - foreach( $inputPages as $page ) { |
| 371 | | - $title = Title::newFromText( $page ); |
| 372 | | - if( $title ) { |
| 373 | | - $pageSet[$title->getPrefixedText()] = true; |
| 374 | | - /// @fixme May or may not be more efficient to batch these |
| 375 | | - /// by namespace when given multiple input pages. |
| 376 | | - $result = $dbr->select( |
| 377 | | - array( 'page', $table ), |
| 378 | | - $fields, |
| 379 | | - array_merge( $join, |
| 380 | | - array( |
| 381 | | - 'page_namespace' => $title->getNamespace(), |
| 382 | | - 'page_title' => $title->getDBKey() ) ), |
| 383 | | - __METHOD__ ); |
| 384 | | - foreach( $result as $row ) { |
| 385 | | - $template = Title::makeTitle( $row->namespace, $row->title ); |
| 386 | | - $pageSet[$template->getPrefixedText()] = true; |
| 387 | | - } |
| 388 | | - } |
| 389 | | - } |
| 390 | | - return $pageSet; |
| 391 | | - } |
| 392 | | - |
| 393 | | - /** |
| 394 | | - * Callback function to remove empty strings from the pages array. |
| 395 | | - */ |
| 396 | | - private function filterPage( $page ) { |
| 397 | | - return $page !== '' && $page !== null; |
| 398 | | - } |
| | 338 | + return intval( min( $depth, 5 ) ); |
| | 339 | + } |
| | 340 | + |
| | 341 | + /** Expand a list of pages to include pages linked to from that page. */ |
| | 342 | + private function getPageLinks( $inputPages, $pageSet, $depth ) { |
| | 343 | + for( $depth=$depth; $depth>0; --$depth ) { |
| | 344 | + $pageSet = $this->getLinks( $inputPages, $pageSet, 'pagelinks', |
| | 345 | + array( 'pl_namespace AS namespace', 'pl_title AS title' ), |
| | 346 | + array( 'page_id=pl_from' ) ); |
| | 347 | + } |
| | 348 | + return $pageSet; |
| | 349 | + } |
| | 350 | + |
| | 351 | + /** |
| | 352 | + * Expand a list of pages to include images used in those pages. |
| | 353 | + * @param $inputPages array, list of titles to look up |
| | 354 | + * @param $pageSet array, associative array indexed by titles for output |
| | 355 | + * @return array associative array index by titles |
| | 356 | + */ |
| | 357 | + private function getImages( $inputPages, $pageSet ) { |
| | 358 | + return $this->getLinks( $inputPages, $pageSet, |
| | 359 | + 'imagelinks', |
| | 360 | + array( NS_FILE . ' AS namespace', 'il_to AS title' ), |
| | 361 | + array( 'page_id=il_from' ) ); |
| | 362 | + } |
| | 363 | + |
| | 364 | + /** |
| | 365 | + * Expand a list of pages to include items used in those pages. |
| | 366 | + * @private |
| | 367 | + */ |
| | 368 | + private function getLinks( $inputPages, $pageSet, $table, $fields, $join ) { |
| | 369 | + $dbr = wfGetDB( DB_SLAVE ); |
| | 370 | + foreach( $inputPages as $page ) { |
| | 371 | + $title = Title::newFromText( $page ); |
| | 372 | + if( $title ) { |
| | 373 | + $pageSet[$title->getPrefixedText()] = true; |
| | 374 | + /// @fixme May or may not be more efficient to batch these |
| | 375 | + /// by namespace when given multiple input pages. |
| | 376 | + $result = $dbr->select( |
| | 377 | + array( 'page', $table ), |
| | 378 | + $fields, |
| | 379 | + array_merge( $join, |
| | 380 | + array( |
| | 381 | + 'page_namespace' => $title->getNamespace(), |
| | 382 | + 'page_title' => $title->getDBKey() ) ), |
| | 383 | + __METHOD__ ); |
| | 384 | + foreach( $result as $row ) { |
| | 385 | + $template = Title::makeTitle( $row->namespace, $row->title ); |
| | 386 | + $pageSet[$template->getPrefixedText()] = true; |
| | 387 | + } |
| | 388 | + } |
| | 389 | + } |
| | 390 | + return $pageSet; |
| | 391 | + } |
| | 392 | + |
| | 393 | + /** |
| | 394 | + * Callback function to remove empty strings from the pages array. |
| | 395 | + */ |
| | 396 | + private function filterPage( $page ) { |
| | 397 | + return $page !== '' && $page !== null; |
| | 398 | + } |
| 399 | 399 | } |
| 400 | 400 | |