X-Git-Url: https://git.proxmox.com/?p=pve-docs.git;a=blobdiff_plain;f=pve-docs-mediawiki-import.in;h=14dffa09d9e7556f1c13c72145f303f7497b308f;hp=5ba556bec3f5692086fd8bd6e6e37f599000618c;hb=856993e4166495537f42e0b9c3a51c966227feab;hpb=4068809967eae3c4e3ff7d32ef2e7eabc944e162

diff --git a/pve-docs-mediawiki-import.in b/pve-docs-mediawiki-import.in
index 5ba556b..14dffa0 100755
--- a/pve-docs-mediawiki-import.in
+++ b/pve-docs-mediawiki-import.in
@@ -7,6 +7,7 @@ use Data::Dumper;
 use IO::File;
 use File::Basename;
 use MediaWiki::API;
+use HTML::Parser;
 
 use JSON;
 
@@ -17,10 +18,10 @@ my $fileinfo = decode_json($data_str);
 
 my $config_fn = "/root/.pve-docs"; # format 'username:pw'
 
-my $fh = IO::File->new("$config_fn") || 
+my $fh = IO::File->new("$config_fn") ||
     die "Please configure the mediawiki user/passswd in '$config_fn'\n";
 
-my $api_url = "http://localhost/mediawiki/api.php";
+my $api_url = "https://pve.proxmox.com/mediawiki/api.php";
 
 my $config = <$fh>;
 chomp $config;
@@ -35,27 +36,62 @@ $mw->login({ lgname => $username, lgpassword => $passwd })
     || die $mw->{error}->{code} . ': ' . $mw->{error}->{details};
 
 sub update_page {
-    my ($pagename, $include, $category) = @_;
+    my ($pagename, $filename, $category) = @_;
 
     print "update mediawiki page: $pagename\n";
 
     my $ref = $mw->get_page( { title => $pagename } );
     my $page =  $ref->{'*'} || '';
 
-    if ($page !~ m/^\{\{#pvedocs:.*\}\}\s*$/m) {
-	$page = "{{#pvedocs:$include}}\n$page";
-    } else {
-	$page =~ s/^\{\{#pvedocs:.*\}\}\s*$/\{\{#pvedocs:$include\}\}\n/m;
-    }
+    my $pve_content = "<!-- Do not edit - this is autogenerated content -->\n";
+
+    $pve_content .= "{{#pvedocs:$filename}}\n";
+    $pve_content .= "[[Category:$category]]\n" if $category;
+
+    my $starttag = '<!--PVE_IMPORT_START_MARKER-->';
+    my $endtag = '<!--PVE_IMPORT_END_MARKER-->';
+
+    $pve_content .= "<pvehide>\n";
 
-    if ($category) {
-	my $catstr = "Category:$category";
+    my $parser_opts = {
+	api_version => 3,
+	text_h => [ sub { $pve_content .= shift }, "text" ],
+    };
+    my $parser = HTML::Parser->new(%$parser_opts);
+    $parser->ignore_elements(qw(script style));
 
-	if ($page !~ m/^\[\[$catstr\]\]\s*$/m) {
-	    $page .= "\n[[$catstr]]\n";
-	}
+    my $fh = IO::File->new("/usr/share/pve-docs/$filename", "r") or
+	die "unable to open file '$filename' - $!\n";
+    while (defined(my $line = <$fh>)) {
+	$parser->parse($line);
+    }
+    $pve_content .= "</pvehide>\n";
+
+    $pve_content =~ s/\s+$//gm;
+
+    chomp $pve_content;
+
+    if ($page =~ m/^(.*)$starttag\n.*\n$endtag\n?(.*)$/s) {
+	my ($top_content, $bottom_content) = ($1, $2);
+	$page = $top_content;
+	$page .= "$starttag\n";
+	$page .= $pve_content;
+	$page .= "\n$endtag\n";
+	$page .= $bottom_content;
+    } elsif ($page =~ m/(.*)\{\{#pvedocs:.*?\}\}(.*)$/) {
+	# old style
+	my ($top_content, $bottom_content) = ($1, $2);
+	chomp $top_content;
+	chomp $bottom_content;
+	$page = $top_content;
+	$page .= "$starttag\n";
+	$page .= $pve_content;
+	$page .= "\n$endtag\n";
+	$page .= $bottom_content;
+    } else {
+	$page = "$starttag\n$pve_content\n$endtag\n$page";
     }
-    
+
     my $timestamp = $ref->{timestamp};
     my $wcmd = {
 	action => 'edit',
@@ -64,7 +100,7 @@ sub update_page {
 	text => $page,
     };
 
-    $mw->edit($wcmd) || 
+    $mw->edit($wcmd) ||
 	die $mw->{error}->{code} . ': ' . $mw->{error}->{details};
 }
 
@@ -79,8 +115,8 @@ foreach my $source (sort keys %{$fileinfo->{toplevel}->{wiki}}) {
 
     my $path = "/usr/share/pve-docs/$filename";
     die "no such file '$path'" if ! -f $path;
-    
-    update_page($title, $filename, category => $cat_refdoc);
+
+    update_page($title, $filename, $cat_refdoc);
 }
 
 # also update 'Get support' page, because this is used since a long