Tika for metadata extraction

First pass of converting a few extractors to use Tika rather than 3rd party libraries directly, or use the new style tika structure


git-svn-id: https://svn.alfresco.com/repos/alfresco-enterprise/alfresco/HEAD/root@20640 c4b6b30b-aa2e-2d43-bbcb-ca4b014f7261
This commit is contained in:
Nick Burch
2010-06-14 19:02:37 +00:00
parent 7aeff72605
commit 63b2f5983a
9 changed files with 329 additions and 226 deletions

View File

@@ -18,20 +18,14 @@
*/
package org.alfresco.repo.content.metadata;
import java.io.IOException;
import java.io.InputStream;
import java.io.Serializable;
import java.util.Arrays;
import java.util.HashSet;
import java.util.ArrayList;
import java.util.Map;
import org.alfresco.repo.content.MimetypeMap;
import org.alfresco.service.cmr.repository.ContentReader;
import org.apache.tika.metadata.Metadata;
import org.apache.tika.parser.ParseContext;
import org.apache.tika.parser.Parser;
import org.apache.tika.parser.dwg.DWGParser;
import org.apache.tika.sax.BodyContentHandler;
import org.xml.sax.ContentHandler;
/**
@@ -47,64 +41,40 @@ import org.xml.sax.ContentHandler;
* <b>lastauthor:</b>
* </pre>
*
* TIKA Note - this has been converted to deep-call into Tika.
* This will be replaced with proper calls to Tika at a later date.
* Uses Apache Tika
*
* @author Nick Burch
*/
public class DWGMetadataExtracter extends AbstractMappingMetadataExtracter
public class DWGMetadataExtracter extends TikaPoweredMetadataExtracter
{
private static final String KEY_AUTHOR = "author";
private static final String KEY_COMMENT = "comment";
private static final String KEY_DESCRIPTION = "description";
private static final String KEY_KEYWORD = "keyword";
private static final String KEY_LAST_AUTHOR = "lastAuthor";
private static final String KEY_TITLE = "title";
public static String[] SUPPORTED_MIMETYPES = new String[] {
MimetypeMap.MIMETYPE_APP_DWG,
MimetypeMap.MIMETYPE_IMG_DWG,
"image/x-dwg", // Was used before IANA registration
};
public static ArrayList<String> SUPPORTED_MIMETYPES = buildSupportedMimetypes(
new String[] {
MimetypeMap.MIMETYPE_APP_DWG,
MimetypeMap.MIMETYPE_IMG_DWG,
"image/x-dwg", // Was used before IANA registration
},
new DWGParser()
);
public DWGMetadataExtracter()
{
super(new HashSet<String>(Arrays.asList(SUPPORTED_MIMETYPES)));
super(SUPPORTED_MIMETYPES);
}
@Override
protected Map<String, Serializable> extractSpecific(Metadata metadata,
Map<String, Serializable> properties) {
putRawValue(KEY_KEYWORD, metadata.get(Metadata.KEYWORDS), properties);
putRawValue(KEY_LAST_AUTHOR, metadata.get(Metadata.LAST_AUTHOR), properties);
System.err.println(properties);
return properties;
}
@Override
public Map<String, Serializable> extractRaw(ContentReader reader) throws Throwable
{
Map<String, Serializable> rawProperties = newRawMap();
InputStream is = null;
try
{
is = reader.getContentInputStream();
DWGParser dwgParser = new DWGParser();
ContentHandler handler = new BodyContentHandler() ;
Metadata metadata = new Metadata();
ParseContext context = new ParseContext();
dwgParser.parse(is, handler, metadata, context);
putRawValue(KEY_AUTHOR, metadata.get(Metadata.AUTHOR), rawProperties);
putRawValue(KEY_COMMENT, metadata.get(Metadata.COMMENTS), rawProperties);
putRawValue(KEY_DESCRIPTION, metadata.get(Metadata.DESCRIPTION), rawProperties);
putRawValue(KEY_KEYWORD, metadata.get(Metadata.KEYWORDS), rawProperties);
putRawValue(KEY_LAST_AUTHOR, metadata.get(Metadata.LAST_AUTHOR), rawProperties);
putRawValue(KEY_DESCRIPTION, metadata.get(Metadata.SUBJECT), rawProperties);
putRawValue(KEY_TITLE, metadata.get(Metadata.TITLE), rawProperties);
}
finally
{
if (is != null)
{
try { is.close(); } catch (IOException e) {}
}
}
// Done
return rawProperties;
protected Parser getParser() {
return new DWGParser();
}
}