Indexing License Metadata

[Jason Kivlighn <jkivlighn gmail com>]

Hi all,

I'm interested in extending existing filters to extract license
metadata.  The attached patch demonstrates extracting Creative Commons
license metadata from SVGs, XMP embedded in images and PDFs, and OLE files.

However, I don't know what to do with the licenses extracted.  Does a
new property need to be added?  How do I get the license back out after
indexing a file with a license?

Thanks,
Jason

Index: Util/F-Spot/MetadataStore.cs
===================================================================
--- Util/F-Spot/MetadataStore.cs	(revision 3791)
+++ Util/F-Spot/MetadataStore.cs	(working copy)
@@ -198,6 +198,7 @@
 		public const string RdfsNS = "http://www.w3.org/2000/01/rdf-schema#";;
 		public const string IViewNS = "http://ns.iview-multimedia.com/mediapro/1.0/";;
 		public const string XmlNS = "http://iptc.org/std/Iptc4xmpCore/1.0/xmlns/";;
+		public const string CcNS = "http://creativecommons.org/ns#";;
 
 		// FIXME this needs to be parsable by System.Uri
 		public const string FSpotXMPBase = "http://fakebase.f-spot.org/internal/";;
@@ -219,6 +220,7 @@
 			Namespaces.AddNamespace (RdfNS, "rdf");
 			Namespaces.AddNamespace (RdfsNS, "rdfs");
 			Namespaces.AddNamespace (IViewNS, "mediapro");
+			Namespaces.AddNamespace (CcNS, "cc");
 		}
 
 		public static MetadataStore Descriptions {
Index: Filters/FilterPdf.cs
===================================================================
--- Filters/FilterPdf.cs	(revision 3791)
+++ Filters/FilterPdf.cs	(working copy)
@@ -13,7 +13,10 @@
 
 using Beagle.Util;
 using Beagle.Daemon;
+using Beagle.Util.Xmp;
 
+using SemWeb;
+
 namespace Beagle.Filters {
 
 	public class FilterPdf : Beagle.Daemon.Filter {
@@ -39,7 +42,7 @@
 		{
 			// create new external process
 			pc = new SafeProcess ();
-			pc.Arguments = new string [] { "pdfinfo", FileInfo.FullName };
+			pc.Arguments = new string [] { "pdfinfo", "-meta", FileInfo.FullName };
 			pc.RedirectStandardOutput = true;
 			// See FIXME below for why this is false.
 			pc.RedirectStandardError = false;
@@ -86,6 +89,11 @@
 					case "Producer":
 						strMetaTag = "dc:appname";
 						break;
+					case "Metadata":
+						string xmpString = pout.ReadToEnd();
+						XmpFile xmp = new XmpFile (new MemoryStream(System.Text.Encoding.ASCII.GetBytes(xmpString)));
+						AddXmpProperties(xmp);
+						break;
 					}
 					if (strMetaTag != null) {
 						if (bKeyword)
@@ -196,6 +204,49 @@
 #endif
 			pc.Close ();
 		}
+
+		internal void AddXmpProperties (XmpFile xmp)
+		{
+			Resource subject_anon = null;
+			Resource creator_anon = null;
+			Resource rights_anon = null;
+			Resource title_anon = null;
+
+			foreach (Statement stmt in xmp.Store) {
+				if (stmt.Predicate == MetadataStore.Namespaces.Resolve ("dc:subject")) {
+					//Console.WriteLine ("found subject");
+					subject_anon = stmt.Object;
+				} else if (stmt.Predicate == MetadataStore.Namespaces.Resolve ("dc:creator")) {
+					//Console.WriteLine ("found creator");
+					creator_anon = stmt.Object;
+				} else if (stmt.Predicate == MetadataStore.Namespaces.Resolve ("dc:rights")) {
+					rights_anon = stmt.Object;
+				} else if (stmt.Predicate == MetadataStore.Namespaces.Resolve ("dc:title")) {
+					if (stmt.Object is Literal)
+						AddProperty (Beagle.Property.New ("dc:title", ((Literal)stmt.Object).Value));
+					else if (stmt.Object is BNode)
+						title_anon = stmt.Object;
+				} else if (stmt.Predicate == MetadataStore.Namespaces.Resolve ("cc:license")) {
+					AddProperty (Beagle.Property.NewKeyword ("fixme:license", ((Literal)stmt.Object).Value));
+				}
+			}
+			
+			foreach (Statement stmt in xmp.Store) {
+				if (stmt.Subject == subject_anon && 
+				    stmt.Predicate != MetadataStore.Namespaces.Resolve ("rdf:type")) {
+					AddProperty (Beagle.Property.New ("dc:subject", ((Literal)stmt.Object).Value));
+				} else if (stmt.Subject == creator_anon &&  
+					   stmt.Predicate != MetadataStore.Namespaces.Resolve ("rdf:type")) {
+					AddProperty (Beagle.Property.New ("dc:creator", ((Literal)stmt.Object).Value));
+				} else if (stmt.Subject == rights_anon &&  
+					   stmt.Predicate != MetadataStore.Namespaces.Resolve ("rdf:type")) {
+					AddProperty (Beagle.Property.New ("dc:rights", ((Literal)stmt.Object).Value));
+				} else if (stmt.Subject == title_anon &&
+					   stmt.Predicate != MetadataStore.Namespaces.Resolve ("rdf:type")) {
+					AddProperty (Beagle.Property.New ("dc:title", ((Literal)stmt.Object).Value));
+				}
+			}
+		}
 	}
 }
 
Index: Filters/FilterOle.cs
===================================================================
--- Filters/FilterOle.cs	(revision 3791)
+++ Filters/FilterOle.cs	(working copy)
@@ -141,6 +141,10 @@
 				prop = docSumMeta.Lookup ("gsf:category");
 				if (prop != null)
 					AddProperty (Beagle.Property.New ("fixme:category", prop.Val as string));
+
+				prop = docSumMeta.Lookup ("CreativeCommons_LicenseURL");
+				if (prop != null)
+					AddProperty (Beagle.Property.New ("fixme:license", prop.Val as string));
 			}
 
 			ExtractMetaData (sum_stream, doc_stream);
Index: Filters/FilterImage.cs
===================================================================
--- Filters/FilterImage.cs	(revision 3791)
+++ Filters/FilterImage.cs	(working copy)
@@ -157,6 +157,8 @@
 				} else if (stmt.Predicate == MetadataStore.Namespaces.Resolve ("tiff:Model")) {
 					// NOTE: the namespaces for xmp and beagle don't always match up
 					AddProperty (Beagle.Property.New ("exif:Model", ((Literal)stmt.Object).Value));
+				} else if (stmt.Predicate == MetadataStore.Namespaces.Resolve ("cc:license")) {
+					AddProperty (Beagle.Property.NewKeyword ("fixme:license", ((Literal)stmt.Object).Value));
 				}
 			}
 			
Index: Filters/FilterSvg.cs
===================================================================
--- Filters/FilterSvg.cs	(revision 3791)
+++ Filters/FilterSvg.cs	(working copy)
@@ -41,6 +41,7 @@
 		static private string [] ignore_strings = { "format"};
 		
 		static private string dcnamespace = "http://purl.org/dc/elements/1.1/";;
+		static private string ccnamespace = "http://creativecommons.org/ns#";;
 		
 		public FilterSvg ()
 		{
@@ -154,8 +155,14 @@
 							if (grab_text)
 								break;
 							
-							if (reader.IsEmptyElement)
+							if (reader.IsEmptyElement) {
+ 								if (reader.NamespaceURI == ccnamespace) {
+									if ( reader.LocalName == "license" ) {
+										AddProperty (Property.NewKeyword ("fixme:license", reader.GetAttribute("resource","http://www.w3.org/1999/02/22-rdf-syntax-ns#";)));
+									}
+								}
 								break;
+							}
 
 							if (ArrayFu.IndexOfString (ignore_strings, reader.LocalName) != -1)
 								break;