* Fix some command-line propagation issues

[lhc/web/wiklou.git] / maintenance / generateSitemap.php

diff --git a/maintenance/generateSitemap.php b/maintenance/generateSitemap.php
index c0f1189..52dc33a 100644 (file)
--- a/maintenance/generateSitemap.php
+++ b/maintenance/generateSitemap.php
@@ -2,16 +2,16 @@
 define( 'GS_MAIN', -2 );
 define( 'GS_TALK', -1 );
 /**
- * Creates a Google sitemap for the site
+ * Creates a sitemap for the site
  *
- * @addtogroup 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>
  *
- * @see http://www.google.com/webmasters/sitemaps/docs/en/about.html
- * @see 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
  */
@@ -20,7 +20,7 @@ class GenerateSitemap {
        /**
         * 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
         */
@@ -29,7 +29,7 @@ class GenerateSitemap {
        /**
         * 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
         */
@@ -80,8 +80,8 @@ class GenerateSitemap {
                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',
@@ -148,22 +148,47 @@ class GenerateSitemap {
 
                $this->url_limit = 50000;
                $this->size_limit = pow( 2, 20 ) * 10;
-               $this->fspath = isset( $fspath ) ? $fspath : '';
+               $this->fspath = self::init_path( $fspath );
+
                $this->compress = $compress;
 
                $this->stderr = fopen( 'php://stderr', 'wt' );
                $this->dbr = wfGetDB( DB_SLAVE );
                $this->generateNamespaces();
                $this->timestamp = wfTimestamp( TS_ISO_8601, wfTimestampNow() );
+
+
                $this->findex = fopen( "{$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 ;
+       }
+
        /**
         * Generate a one-dimensional array of existing namespaces
         */
        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(),
@@ -253,7 +278,7 @@ class GenerateSitemap {
                                        $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;
                                }
@@ -345,7 +370,7 @@ class GenerateSitemap {
         * @returns string
         */
        function xmlSchema() {
-               return 'http://www.google.com/schemas/sitemap/0.84';
+               return 'http://www.sitemaps.org/schemas/sitemap/0.9';
        }
 
        /**
@@ -399,7 +424,7 @@ class GenerateSitemap {
         *
         * @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
         *
@@ -450,7 +475,7 @@ if ( in_array( '--help', $argv ) ) {
 Usage: php generateSitemap.php [options]
        --help                  show this message
 
-       --fspath=<path>         The file system path to save to, e.g /tmp/sitemap/
+       --fspath=<path>         The file system path to save to, e.g /tmp/sitemap
 
        --server=<server>       The protocol and server name to use in URLs, e.g.
                http://en.wikipedia.org. This is sometimes necessary because