define( 'GS_MAIN', -2 );
define( 'GS_TALK', -1 );
/**
- * Creates a Google sitemap for the site
+ * Creates a sitemap for the site
*
- * @package MediaWiki
- * @subpackage Maintenance
+ * @ingroup Maintenance
*
* @copyright Copyright © 2005, Ævar Arnfjörð Bjarmason
* @copyright Copyright © 2005, Jens Frank <jeluf@gmx.de>
* @copyright Copyright © 2005, Brion Vibber <brion@pobox.com>
*
- * @link http://www.google.com/webmasters/sitemaps/docs/en/about.html
- * @link http://www.google.com/schemas/sitemap/0.84/sitemap.xsd
+ * @see http://www.sitemaps.org/
+ * @see http://www.sitemaps.org/schemas/sitemap/0.9/sitemap.xsd
*
* @license http://www.gnu.org/copyleft/gpl.html GNU General Public License 2.0 or later
*/
/**
* The maximum amount of urls in a sitemap file
*
- * @link http://www.google.com/schemas/sitemap/0.84/sitemap.xsd
+ * @link http://www.sitemaps.org/schemas/sitemap/0.9/sitemap.xsd
*
* @var int
*/
/**
* The maximum size of a sitemap file
*
- * @link http://www.google.com/webmasters/sitemaps/docs/en/protocol.html#faq_sitemap_size
+ * @link http://www.sitemaps.org/faq.php#faq_sitemap_size
*
* @var int
*/
var $size_limit;
-
+
/**
* The path to prepend to the filename
*
* @var string
*/
var $path;
-
+
/**
* Whether or not to use compression
*
* @var bool
*/
var $compress;
-
+
/**
* The number of entries to save in each sitemap file
*
NS_USER_TALK => '0.1',
NS_PROJECT => '0.5',
NS_PROJECT_TALK => '0.1',
- NS_IMAGE => '0.5',
- NS_IMAGE_TALK => '0.1',
+ NS_FILE => '0.5',
+ NS_FILE_TALK => '0.1',
NS_MEDIAWIKI => '0.0',
NS_MEDIAWIKI_TALK => '0.1',
NS_TEMPLATE => '0.0',
* @var object
*/
var $dbr;
-
+
/**
* A resource pointing to the sitemap index file
*
* @var resource
*/
var $findex;
-
+
/**
* A resource pointing to a sitemap file
* @param string $path The path to append to the domain name
* @param bool $compress Whether to compress the sitemap files
*/
- function GenerateSitemap( $fspath, $path, $compress ) {
- global $wgDBname, $wgScriptPath;
-
+ function GenerateSitemap( $fspath, $compress ) {
+ global $wgScriptPath;
+
$this->url_limit = 50000;
$this->size_limit = pow( 2, 20 ) * 10;
- $this->fspath = isset( $fspath ) ? $fspath : '';
- $this->path = isset( $path ) ? $path : $wgScriptPath;
+ $this->fspath = self::init_path( $fspath );
+
$this->compress = $compress;
$this->stderr = fopen( 'php://stderr', 'wt' );
- $this->dbr =& wfGetDB( DB_SLAVE );
+ $this->dbr = wfGetDB( DB_SLAVE );
$this->generateNamespaces();
$this->timestamp = wfTimestamp( TS_ISO_8601, wfTimestampNow() );
- $this->findex = fopen( "{$this->fspath}sitemap-index-$wgDBname.xml", 'wb' );
+
+
+ $this->findex = fopen( "{$this->fspath}sitemap-index-" . wfWikiID() . ".xml", 'wb' );
+ }
+
+ /**
+ * 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 ) ) {
+ mkdir( $fspath, 0755 ) or die("Can not create directory $fspath.\n");
+ }
+
+ return realpath( $fspath ). DIRECTORY_SEPARATOR ;
}
/**
*/
function generateNamespaces() {
$fname = 'GenerateSitemap::generateNamespaces';
-
+
+ // Only generate for specific namespaces if $wgSitemapNamespaces is an array.
+ global $wgSitemapNamespaces;
+ if( is_array( $wgSitemapNamespaces ) ) {
+ $this->namespaces = $wgSitemapNamespaces;
+ return;
+ }
+
$res = $this->dbr->select( 'page',
array( 'page_namespace' ),
array(),
* @return string
*/
function guessPriority( $namespace ) {
- return Namespace::isMain( $namespace ) ? $this->priorities[GS_MAIN] : $this->priorities[GS_TALK];
+ return MWNamespace::isMain( $namespace ) ? $this->priorities[GS_MAIN] : $this->priorities[GS_TALK];
}
/**
* @access public
*/
function main() {
- global $wgDBname, $wgContLang;
+ global $wgContLang;
fwrite( $this->findex, $this->openIndex() );
-
+
foreach ( $this->namespaces as $namespace ) {
$res = $this->getPageRes( $namespace );
$this->file = false;
$this->generateLimit( $namespace );
$length = $this->limit[0];
$i = $smcount = 0;
-
+
$fns = $wgContLang->getFormattedNsText( $namespace );
$this->debug( "$namespace ($fns)" );
while ( $row = $this->dbr->fetchObject( $res ) ) {
$this->file = $this->open( $this->fspath . $filename, 'wb' );
$this->write( $this->file, $this->openFile() );
fwrite( $this->findex, $this->indexEntry( $filename ) );
- $this->debug( "\t$filename" );
+ $this->debug( "\t$this->fspath$filename" );
$length = $this->limit[0];
$i = 1;
}
$entry = $this->fileEntry( $title->getFullURL(), $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 ) );
+ $length += strlen( $entry );
+ $this->write( $this->file, $entry );
+ }
+ }
}
if ( $this->file ) {
$this->write( $this->file, $this->closeFile() );
function open( $file, $flags ) {
return $this->compress ? gzopen( $file, $flags ) : fopen( $file, $flags );
}
-
+
/**
* gzwrite() / fwrite() wrapper
*/
* @return string
*/
function sitemapFilename( $namespace, $count ) {
- global $wgDBname;
-
$ext = $this->compress ? '.gz' : '';
-
- return "sitemap-$wgDBname-NS_$namespace-$count.xml$ext";
+ return "sitemap-".wfWikiID()."-NS_$namespace-$count.xml$ext";
}
/**
* @returns string
*/
function xmlSchema() {
- return 'http://www.google.com/schemas/sitemap/0.84';
+ return 'http://www.sitemaps.org/schemas/sitemap/0.9';
}
/**
*
* @static
*
- * @param string $url An RFC 2396 compilant URL
+ * @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
*
*/
function generateLimit( $namespace ) {
$title = Title::makeTitle( $namespace, str_repeat( "\xf0\xa8\xae\x81", 63 ) . "\xe5\x96\x83" );
-
+
$this->limit = array(
strlen( $this->openFile() ),
strlen( $this->fileEntry( $title->getFullUrl(), wfTimestamp( TS_ISO_8601, wfTimestamp() ), $this->priority( $namespace ) ) ),
}
}
-if ( in_array( '--help', $argv ) )
- die(
- "Usage: php generateSitemap.php [host] [options]\n" .
- "\thost = hostname\n" .
- "\toptions:\n" .
- "\t\t--help\tshow this message\n" .
- "\t\t--fspath\tThe file system path to save to, e.g /tmp/sitemap/\n" .
- "\t\t--path\tThe http path to use, e.g. /wiki\n" .
- "\t\t--compress=[yes|no]\tcompress the sitemap files, default yes\n"
- );
+if ( in_array( '--help', $argv ) ) {
+ echo <<<EOT
+Usage: php generateSitemap.php [options]
+ --help show this message
+
+ --fspath=<path> The file system path to save to, e.g /tmp/sitemap
+ Saves to current directory if not given.
+
+ --server=<server> The protocol and server name to use in URLs, e.g.
+ http://en.wikipedia.org. This is sometimes necessary because
+ server name detection may fail in command line scripts.
+
+ --compress=[yes|no] compress the sitemap files, default yes
-if ( isset( $argv[1] ) && strpos( $argv[1], '--' ) !== 0 )
- $_SERVER['SERVER_NAME'] = $argv[1];
+EOT;
+ die( -1 );
+}
+
+$optionsWithArgs = array( 'fspath', 'server', 'compress' );
+require_once( dirname( __FILE__ ) . '/commandLine.inc' );
-$optionsWithArgs = array( 'fspath', 'path', 'compress' );
-require_once 'commandLine.inc';
+if ( isset( $options['server'] ) ) {
+ $wgServer = $options['server'];
+}
-$gs = new GenerateSitemap( @$options['fspath'], @$options['path'], @$options['compress'] !== 'no' );
+$gs = new GenerateSitemap( @$options['fspath'], @$options['compress'] !== 'no' );
$gs->main();
-?>
+