Change the showing of hidden pages by the SiteIndex to be configurable. Fixes RFE #11044.

Also, change the SiteIndex to use an iterator.

docs/changelog/7.x.x.txt

@@ -5,6 +5,7 @@
  - fixed #11372: All Search Forms should use GET
  - fixed #11373: Problem creating FilePump bundles
  - fixed #11374: USPS shipping drivers take 3 minutes to timeout
+ - fixed #11044: Optionally include hidden pages in sitemap.xml
 
 7.8.10
  - fixed #11332: Pagination in webgui.org forum urls

docs/gotcha.txt

@@ -11,6 +11,11 @@ save you many hours of grief.
 --------------------------------------------------------------------
      * WebGUI now requires DateTime::Format::HTTP, to handle parsing HTTP dates.
 
+     * The Google sitemap generator in WebGUI was including hidden pages in the list of
+       pages.  This is now configurable, with the default being to NOT show hidden pages.
+       If you wish hidden pages to be shown, then in the config file set showHiddenPages
+       to 1.
+
 7.8.6
 --------------------------------------------------------------------
     * WebGUI now requires LWP 5.833 or higher, to fix a bug in that module.

docs/upgrades/upgrade_7.8.10-7.8.11.pl

@@ -31,6 +31,7 @@ my $quiet; # this line required
 my $session = start(); # this line required
 
 # upgrade functions go here
+addSiteIndexToConfig($session);
 
 finish($session); # this line required
 
@@ -44,6 +45,16 @@ finish($session); # this line required
 #    print "DONE!\n" unless $quiet;
 #}
 
+#----------------------------------------------------------------------------
+# Describe what our function does
+sub addSiteIndexToConfig {
+    my $session = shift;
+    print "\tAdd Site Config indexing section to config file... " unless $quiet;
+    $session->config->set('siteIndex', { showHiddenPages => 0 });
+    # and here's our code
+    print "DONE!\n" unless $quiet;
+}
+
 
 # -------------- DO NOT EDIT BELOW THIS LINE --------------------------------
 

etc/WebGUI.conf.original

@@ -1071,4 +1071,10 @@
       "Opera Mini"
    ],
 
+#A list of UserAgents of recognized mobile platforms.  If useMobileStyle is set in the
+#Admin settings, then the mobile style will be used for these browsers.
+   "siteIndex" : {
+      "showHiddenPages" : 0
+   }
+
 }

lib/WebGUI/Content/SiteIndex.pm

@@ -16,6 +16,7 @@ package WebGUI::Content::SiteIndex;
 
 use strict;
 use WebGUI::Asset;
+use WebGUI::Exception;
 use XML::Simple;
 
 =head1 NAME
@@ -53,17 +54,28 @@ sub handler {
     unless ($p =~ m/sitemap\.xml$/i) {
         return undef;
     }
+
+    my $whereClause = "assetData.groupIdView = 7";
+    if (! $session->config->get("siteIndex")->{showHiddenPages}) {
+        $whereClause .= ' AND assetData.isHidden=0';
+    }
     
-    my $pages  = WebGUI::Asset->getRoot($session)->getLineage(["self","descendants"],{
+    my $pages  = WebGUI::Asset->getRoot($session)->getLineageIterator(["self","descendants"],{
         returnObjects      => 1,
         includeOnlyClasses => ["WebGUI::Asset::Wobject::Layout"],
-        whereClause        => "assetData.groupIdView = 7",
+        whereClause        => $whereClause,
         limit              => 20000
     });
     
     
     my $url          = [];
-    foreach my $page (@{$pages}) {
+    ASSET: while (1) {
+        my $page = eval { $pages->() };
+        if (my $e = Exception::Class->caught()) {
+            $session->log->error($@);
+            next ASSET;
+        }
+        last ASSET unless $page;
         push(@{$url},{
             loc     => $session->url->getSiteURL().formatXML($page->getUrl),
             lastmod => $session->datetime->epochToSet($page->get("revisionDate")),
@@ -112,4 +124,3 @@ sub formatXML {
 
 
 1;
-