X-Git-Url: https://scripts.mit.edu/gitweb/autoinstallsdev/mediawiki.git/blobdiff_plain/19e297c21b10b1b8a3acad5e73fc71dcb35db44a..6932310fd58ebef145fa01eb76edf7150284d8ea:/maintenance/generateSitemap.php diff --git a/maintenance/generateSitemap.php b/maintenance/generateSitemap.php index e483f7c9..26a9c399 100644 --- a/maintenance/generateSitemap.php +++ b/maintenance/generateSitemap.php @@ -1,8 +1,6 @@ and * Brion Vibber @@ -28,9 +26,17 @@ define( 'GS_TALK', -1 ); * @see http://www.sitemaps.org/schemas/sitemap/0.9/sitemap.xsd */ -require_once( dirname( __FILE__ ) . '/Maintenance.php' ); +require_once __DIR__ . '/Maintenance.php'; +/** + * Maintenance script that generates a sitemap for the site. + * + * @ingroup Maintenance + */ class GenerateSitemap extends Maintenance { + const GS_MAIN = -2; + const GS_TALK = -1; + /** * The maximum amount of urls in a sitemap file * @@ -38,7 +44,7 @@ class GenerateSitemap extends Maintenance { * * @var int */ - var $url_limit; + public $url_limit; /** * The maximum size of a sitemap file @@ -47,88 +53,122 @@ class GenerateSitemap extends Maintenance { * * @var int */ - var $size_limit; + public $size_limit; /** * The path to prepend to the filename * * @var string */ - var $fspath; + public $fspath; /** - * The URL path to prepend to filenames in the index; should resolve to the same directory as $fspath + * The URL path to prepend to filenames in the index; + * should resolve to the same directory as $fspath. * * @var string */ - var $urlpath; + public $urlpath; /** * Whether or not to use compression * * @var bool */ - var $compress; + public $compress; + + /** + * Whether or not to include redirection pages + * + * @var bool + */ + public $skipRedirects; /** * The number of entries to save in each sitemap file * * @var array */ - var $limit = array(); + public $limit = []; /** * Key => value entries of namespaces and their priorities * * @var array */ - var $priorities = array(); + public $priorities = []; /** * A one-dimensional array of namespaces in the wiki * * @var array */ - var $namespaces = array(); + public $namespaces = []; /** * When this sitemap batch was generated * * @var string */ - var $timestamp; + public $timestamp; /** - * A database slave object + * A database replica DB object * * @var object */ - var $dbr; + public $dbr; /** * A resource pointing to the sitemap index file * * @var resource */ - var $findex; - + public $findex; /** * A resource pointing to a sitemap file * * @var resource */ - var $file; + public $file; /** - * Constructor + * Identifier to use in filenames, default $wgDBname + * + * @var string */ + private $identifier; + public function __construct() { parent::__construct(); - $this->mDescription = "Creates a sitemap for the site"; - $this->addOption( 'fspath', 'The file system path to save to, e.g. /tmp/sitemap; defaults to current directory', false, true ); - $this->addOption( 'urlpath', 'The URL path corresponding to --fspath, prepended to filenames in the index; defaults to an empty string', false, true ); - $this->addOption( 'compress', 'Compress the sitemap files, can take value yes|no, default yes', false, true ); + $this->addDescription( 'Creates a sitemap for the site' ); + $this->addOption( + 'fspath', + 'The file system path to save to, e.g. /tmp/sitemap; defaults to current directory', + false, + true + ); + $this->addOption( + 'urlpath', + 'The URL path corresponding to --fspath, prepended to filenames in the index; ' + . 'defaults to an empty string', + false, + true + ); + $this->addOption( + 'compress', + 'Compress the sitemap files, can take value yes|no, default yes', + false, + true + ); + $this->addOption( 'skip-redirects', 'Do not include redirecting articles in the sitemap' ); + $this->addOption( + 'identifier', + 'What site identifier to use for the wiki, defaults to $wgDBname', + false, + true + ); } /** @@ -138,24 +178,35 @@ class GenerateSitemap extends Maintenance { $this->setNamespacePriorities(); $this->url_limit = 50000; $this->size_limit = pow( 2, 20 ) * 10; - $this->fspath = self::init_path( $this->getOption( 'fspath', getcwd() ) ); + + # Create directory if needed + $fspath = $this->getOption( 'fspath', getcwd() ); + if ( !wfMkdirParents( $fspath, null, __METHOD__ ) ) { + $this->error( "Can not create directory $fspath.", 1 ); + } + + $this->fspath = realpath( $fspath ) . DIRECTORY_SEPARATOR; $this->urlpath = $this->getOption( 'urlpath', "" ); if ( $this->urlpath !== "" && substr( $this->urlpath, -1 ) !== '/' ) { $this->urlpath .= '/'; } + $this->identifier = $this->getOption( 'identifier', wfWikiID() ); $this->compress = $this->getOption( 'compress', 'yes' ) !== 'no'; - $this->dbr = wfGetDB( DB_SLAVE ); + $this->skipRedirects = $this->hasOption( 'skip-redirects' ); + $this->dbr = $this->getDB( DB_REPLICA ); $this->generateNamespaces(); $this->timestamp = wfTimestamp( TS_ISO_8601, wfTimestampNow() ); - $this->findex = fopen( "{$this->fspath}sitemap-index-" . wfWikiID() . ".xml", 'wb' ); + $this->findex = fopen( "{$this->fspath}sitemap-index-{$this->identifier}.xml", 'wb' ); $this->main(); } private function setNamespacePriorities() { + global $wgSitemapNamespacesPriorities; + // Custom main namespaces - $this->priorities[GS_MAIN] = '0.5'; + $this->priorities[self::GS_MAIN] = '0.5'; // Custom talk namesspaces - $this->priorities[GS_TALK] = '0.1'; + $this->priorities[self::GS_TALK] = '0.1'; // MediaWiki standard namespaces $this->priorities[NS_MAIN] = '1.0'; $this->priorities[NS_TALK] = '0.1'; @@ -173,21 +224,22 @@ class GenerateSitemap extends Maintenance { $this->priorities[NS_HELP_TALK] = '0.1'; $this->priorities[NS_CATEGORY] = '0.5'; $this->priorities[NS_CATEGORY_TALK] = '0.1'; - } - /** - * Create directory if it does not exist and return pathname with a trailing slash - */ - private static function init_path( $fspath ) { - if ( !isset( $fspath ) ) { - return null; - } - # Create directory if needed - if ( $fspath && !is_dir( $fspath ) ) { - wfMkdirParents( $fspath ) or die( "Can not create directory $fspath.\n" ); + // Custom priorities + if ( $wgSitemapNamespacesPriorities !== false ) { + /** + * @var $wgSitemapNamespacesPriorities array + */ + foreach ( $wgSitemapNamespacesPriorities as $namespace => $priority ) { + $float = floatval( $priority ); + if ( $float > 1.0 ) { + $priority = '1.0'; + } elseif ( $float < 0.0 ) { + $priority = '0.0'; + } + $this->priorities[$namespace] = $priority; + } } - - return realpath( $fspath ) . DIRECTORY_SEPARATOR ; } /** @@ -198,31 +250,35 @@ class GenerateSitemap extends Maintenance { global $wgSitemapNamespaces; if ( is_array( $wgSitemapNamespaces ) ) { $this->namespaces = $wgSitemapNamespaces; + return; } $res = $this->dbr->select( 'page', - array( 'page_namespace' ), - array(), + [ 'page_namespace' ], + [], __METHOD__, - array( + [ 'GROUP BY' => 'page_namespace', 'ORDER BY' => 'page_namespace', - ) + ] ); - foreach ( $res as $row ) + foreach ( $res as $row ) { $this->namespaces[] = $row->page_namespace; + } } /** * Get the priority of a given namespace * - * @param $namespace Integer: the namespace to get the priority for - * @return String + * @param int $namespace The namespace to get the priority for + * @return string */ function priority( $namespace ) { - return isset( $this->priorities[$namespace] ) ? $this->priorities[$namespace] : $this->guessPriority( $namespace ); + return isset( $this->priorities[$namespace] ) + ? $this->priorities[$namespace] + : $this->guessPriority( $namespace ); } /** @@ -230,27 +286,30 @@ class GenerateSitemap extends Maintenance { * default priority for the namespace, varies depending on whether it's * a talkpage or not. * - * @param $namespace Integer: the namespace to get the priority for - * @return String + * @param int $namespace The namespace to get the priority for + * @return string */ function guessPriority( $namespace ) { - return MWNamespace::isMain( $namespace ) ? $this->priorities[GS_MAIN] : $this->priorities[GS_TALK]; + return MWNamespace::isSubject( $namespace ) + ? $this->priorities[self::GS_MAIN] + : $this->priorities[self::GS_TALK]; } /** * Return a database resolution of all the pages in a given namespace * - * @param $namespace Integer: limit the query to this namespace + * @param int $namespace Limit the query to this namespace * @return Resource */ function getPageRes( $namespace ) { return $this->dbr->select( 'page', - array( + [ 'page_namespace', 'page_title', 'page_touched', - ), - array( 'page_namespace' => $namespace ), + 'page_is_redirect' + ], + [ 'page_namespace' => $namespace ], __METHOD__ ); } @@ -271,9 +330,18 @@ class GenerateSitemap extends Maintenance { $i = $smcount = 0; $fns = $wgContLang->getFormattedNsText( $namespace ); - $this->output( "$namespace ($fns)" ); + $this->output( "$namespace ($fns)\n" ); + $skippedRedirects = 0; // Number of redirects skipped for that namespace foreach ( $res as $row ) { - if ( $i++ === 0 || $i === $this->url_limit + 1 || $length + $this->limit[1] + $this->limit[2] > $this->size_limit ) { + if ( $this->skipRedirects && $row->page_is_redirect ) { + $skippedRedirects++; + continue; + } + + if ( $i++ === 0 + || $i === $this->url_limit + 1 + || $length + $this->limit[1] + $this->limit[2] > $this->size_limit + ) { if ( $this->file !== false ) { $this->write( $this->file, $this->closeFile() ); $this->close( $this->file ); @@ -288,20 +356,31 @@ class GenerateSitemap extends Maintenance { } $title = Title::makeTitle( $row->page_namespace, $row->page_title ); $date = wfTimestamp( TS_ISO_8601, $row->page_touched ); - $entry = $this->fileEntry( $title->getFullURL(), $date, $this->priority( $namespace ) ); + $entry = $this->fileEntry( $title->getCanonicalURL(), $date, $this->priority( $namespace ) ); $length += strlen( $entry ); $this->write( $this->file, $entry ); // generate pages for language variants if ( $wgContLang->hasVariants() ) { $variants = $wgContLang->getVariants(); foreach ( $variants as $vCode ) { - if ( $vCode == $wgContLang->getCode() ) continue; // we don't want default variant - $entry = $this->fileEntry( $title->getFullURL( '', $vCode ), $date, $this->priority( $namespace ) ); + if ( $vCode == $wgContLang->getCode() ) { + continue; // we don't want default variant + } + $entry = $this->fileEntry( + $title->getCanonicalURL( '', $vCode ), + $date, + $this->priority( $namespace ) + ); $length += strlen( $entry ); $this->write( $this->file, $entry ); } } } + + if ( $this->skipRedirects && $skippedRedirects > 0 ) { + $this->output( " skipped $skippedRedirects redirect(s)\n" ); + } + if ( $this->file ) { $this->write( $this->file, $this->closeFile() ); $this->close( $this->file ); @@ -314,42 +393,61 @@ class GenerateSitemap extends Maintenance { /** * gzopen() / fopen() wrapper * - * @return Resource + * @param string $file + * @param string $flags + * @return resource */ function open( $file, $flags ) { - return $this->compress ? gzopen( $file, $flags ) : fopen( $file, $flags ); + $resource = $this->compress ? gzopen( $file, $flags ) : fopen( $file, $flags ); + if ( $resource === false ) { + throw new MWException( __METHOD__ + . " error opening file $file with flags $flags. Check permissions?" ); + } + + return $resource; } /** * gzwrite() / fwrite() wrapper + * + * @param resource &$handle + * @param string $str */ function write( &$handle, $str ) { - if ( $this->compress ) + if ( $handle === true || $handle === false ) { + throw new MWException( __METHOD__ . " was passed a boolean as a file handle.\n" ); + } + if ( $this->compress ) { gzwrite( $handle, $str ); - else + } else { fwrite( $handle, $str ); + } } /** * gzclose() / fclose() wrapper + * + * @param resource &$handle */ function close( &$handle ) { - if ( $this->compress ) + if ( $this->compress ) { gzclose( $handle ); - else + } else { fclose( $handle ); + } } /** * Get a sitemap filename * - * @param $namespace Integer: the namespace - * @param $count Integer: the count - * @return String + * @param int $namespace The namespace + * @param int $count The count + * @return string */ function sitemapFilename( $namespace, $count ) { $ext = $this->compress ? '.gz' : ''; - return "sitemap-" . wfWikiID() . "-NS_$namespace-$count.xml$ext"; + + return "sitemap-{$this->identifier}-NS_$namespace-$count.xml$ext"; } /** @@ -364,7 +462,7 @@ class GenerateSitemap extends Maintenance { /** * Return the XML schema being used * - * @return String + * @return string */ function xmlSchema() { return 'http://www.sitemaps.org/schemas/sitemap/0.9'; @@ -373,7 +471,7 @@ class GenerateSitemap extends Maintenance { /** * Return the XML required to open a sitemap index file * - * @return String + * @return string */ function openIndex() { return $this->xmlHead() . '' . "\n"; @@ -382,8 +480,8 @@ class GenerateSitemap extends Maintenance { /** * Return the XML for a single sitemap indexfile entry * - * @param $filename String: the filename of the sitemap file - * @return String + * @param string $filename The filename of the sitemap file + * @return string */ function indexEntry( $filename ) { return @@ -396,7 +494,7 @@ class GenerateSitemap extends Maintenance { /** * Return the XML required to close a sitemap index file * - * @return String + * @return string */ function closeIndex() { return "\n"; @@ -405,7 +503,7 @@ class GenerateSitemap extends Maintenance { /** * Return the XML required to open a sitemap file * - * @return String + * @return string */ function openFile() { return $this->xmlHead() . '' . "\n"; @@ -414,15 +512,16 @@ class GenerateSitemap extends Maintenance { /** * Return the XML for a single sitemap entry * - * @param $url String: an RFC 2396 compliant URL - * @param $date String: a ISO 8601 date - * @param $priority String: a priority indicator, 0.0 - 1.0 inclusive with a 0.1 stepsize - * @return String + * @param string $url An RFC 2396 compliant URL + * @param string $date A ISO 8601 date + * @param string $priority A priority indicator, 0.0 - 1.0 inclusive with a 0.1 stepsize + * @return string */ function fileEntry( $url, $date, $priority ) { return "\t\n" . - "\t\t$url\n" . + // T36666: $url may contain bad characters such as ampersands. + "\t\t" . htmlspecialchars( $url ) . "\n" . "\t\t$date\n" . "\t\t$priority\n" . "\t\n"; @@ -431,7 +530,7 @@ class GenerateSitemap extends Maintenance { /** * Return the XML required to close sitemap file * - * @return String + * @return string */ function closeFile() { return "\n"; @@ -439,17 +538,24 @@ class GenerateSitemap extends Maintenance { /** * Populate $this->limit + * + * @param int $namespace */ function generateLimit( $namespace ) { + // T19961: make a title with the longest possible URL in this namespace $title = Title::makeTitle( $namespace, str_repeat( "\xf0\xa8\xae\x81", 63 ) . "\xe5\x96\x83" ); - $this->limit = array( + $this->limit = [ strlen( $this->openFile() ), - strlen( $this->fileEntry( $title->getFullUrl(), wfTimestamp( TS_ISO_8601, wfTimestamp() ), $this->priority( $namespace ) ) ), + strlen( $this->fileEntry( + $title->getCanonicalURL(), + wfTimestamp( TS_ISO_8601, wfTimestamp() ), + $this->priority( $namespace ) + ) ), strlen( $this->closeFile() ) - ); + ]; } } $maintClass = "GenerateSitemap"; -require_once( RUN_MAINTENANCE_IF_MAIN ); +require_once RUN_MAINTENANCE_IF_MAIN;