Metadata Extracter

As we transition to the new Hyland Community, some post links may be temporarily disrupted as we finalize redirects. We apologize for the inconvenience and are working to resolve this as quickly as possible. Details.

stebans

Hi all,

I'm currently writing a MetadataExtracter from Word Documents. I want to extract other properties than the default ones (author, title, description) from the properties (like keywords) and from the parsing of the document.  I wrote a Java class which extends the org.alfresco.repo.content.metadata.OfficeMetadataExtracter I could find in Alfresco SDK.

But how can I save the extracted metadata from the Java class?  There exists the ContentModel with a few metadata (ContentModel.PROP_AUTHOR, etc.), but how could I write say document keywords.  Where should I define the PROP_KEYWORD metadata?  Could the property be simply defined in a new ContentModel.xml?

Sorry if the answer is obvious, maybe it's simple Java, but I'm lost!

Thanks,
Best regards,
Stéphane

stebans

Hi,

first of all, I hadn't seen there was another post about the same subject and with almost the same title (http://forums.alfresco.com/viewtopic.php?t=5687), but anyway I found the information which is in this thread on myself.

Anyway, I still cannot make the Extracter extract the metadata!  Here's what I did, and, hopefully someone will be able to help me finish all this up…

1. "my custom model" xml file

<model name="my:model" xmlns="http://www.alfresco.org/model/dictionary/1.0">
   <imports>
      <import uri="http://www.alfresco.org/model/dictionary/1.0" prefix="d" />
      <import uri="http://www.alfresco.org/model/content/1.0" prefix="cm" />
   </imports>

   <namespaces>
      <namespace uri="http://***/model/dictionary/1.0" prefix="my" />
   </namespaces>

   <types>
      <type name="my:model">
         <title>My Model for Documents Metadata</title>
         <parent>cm:content</parent>
         <properties>   
            <property name="my:keywords">
               <title>Keywords for my document</title>
               <description>Keywords for my document</description>
               <type>d:text</type>
               <mandatory>true</mandatory>
            </property>
         </properties>
      </type>
   </types>
</model>
‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍

2. web client configuration

   <config evaluator="string-compare" condition="Content Wizards">
      <content-types>
         <type name="my:model"/>
      </content-types>
   </config>
   <config evaluator="node-type" condition="my:model">
      <property-sheet>
         <show-property name="my:keywords" />
      </property-sheet>
   </config>
‍‍‍‍‍‍‍‍‍‍‍‍

3. Java class which I'm really not sure about.

package my.pckg;

import java.io.IOException;
import java.io.InputStream;
import java.io.Serializable;
import java.util.Map;

import org.alfresco.model.ContentModel;
import org.alfresco.repo.content.metadata.OfficeMetadataExtracter;
import org.alfresco.service.cmr.repository.ContentIOException;
import org.alfresco.service.cmr.repository.ContentReader;
import org.alfresco.service.namespace.QName;
import org.apache.poi.hpsf.PropertySet;
import org.apache.poi.hpsf.PropertySetFactory;
import org.apache.poi.hpsf.SummaryInformation;
import org.apache.poi.poifs.eventfilesystem.POIFSReader;
import org.apache.poi.poifs.eventfilesystem.POIFSReaderEvent;
import org.apache.poi.poifs.eventfilesystem.POIFSReaderListener;


public class MyOfficeMetadataExtracter extends OfficeMetadataExtracter {

   static final QName PROP_KEYWORDS = QName.createQName("http://***/model/dictionary/1.0", "keywords");
   
   public MyOfficeMetadataExtracter() {
      new OfficeMetadataExtracter();
   }

   public void extractInternal(ContentReader reader,
         final Map<QName, Serializable> destination) throws Throwable {
         
      POIFSReaderListener readerListener = new POIFSReaderListener() {

         public void processPOIFSReaderEvent(final POIFSReaderEvent event) {
            try {
               PropertySet ps = PropertySetFactory.create(event.getStream());
               if (ps instanceof SummaryInformation){
                  SummaryInformation si = (SummaryInformation) ps;
                  
                  //trimPut(ContentModel.PROP_AUTHOR, si.getAuthor(), destination);
                  trimPut(PROP_KEYWORDS, si.getKeywords(), destination);                                       
                  
               }

            } catch (Exception ex) {
               throw new ContentIOException("Property set stream: "
                     + event.getPath() + event.getName(), ex);
            }
         }
      };
      
      InputStream is = null;
      try {
         is = reader.getContentInputStream();
         POIFSReader poiFSReader = new POIFSReader();
         poiFSReader.registerListener(readerListener, SummaryInformation.DEFAULT_STREAM_NAME);
         poiFSReader.read(is);
      } catch (Exception e) {
         //…
      } finally {
         if (is != null){
            try {
               is.close();
            } catch (IOException e) {
               //…
            }
         }
      
      }
   }
}
‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍

4. Bean registration in "content-services-context-xml"

   <bean class="my.pckg.MyOfficeMetadataExtracter" parent="org.alfresco.repo.content.metadata.OfficeMetadataExtracter" />
‍‍‍

I don't know if the class is supposed to work at all. Is QName PROP_KEYWORDS = QName.create(namespace, property_name) the right way to go? I'm lost!

Even if it worked out, I don't understand the way to make the mapping between Word documents of my custom content type and the extracter. Where should I configure the mapping?

Hope I'm not too far from the solution and my question is clear enough for a tip…

Thanks
Best Regards
Stéphane