inteligr8/alfresco-community-repo
1
0
Fork 0
mirror of https://github.com/Alfresco/alfresco-community-repo.git synced 2025-07-24 17:32:48 +00:00
Code Issues Packages Projects Releases Wiki Activity

. Outlook email format meta-data extractor

Browse Source 
- expects .msg files in native Outlook format
  - uses POI library for the parsing of the horrid OLE2 compound document format
  - extracts addressee(s), sent date and originator email address
  ...for the future - could be modified and used as a transformer to allow full-text indexing of Outlook format emails

. Add new aspect "emailed" to the contentmodel to support properties for above extractor

git-svn-id: https://svn.alfresco.com/repos/alfresco-enterprise/alfresco/HEAD/root@3387 c4b6b30b-aa2e-2d43-bbcb-ca4b014f7261
This commit is contained in:
Kevin Roast
2006-07-24 15:05:48 +00:00
parent e02ee2fd09
commit e31e027039
15 changed files with 229 additions and 33 deletions
Download Patch File Download Diff File Expand all files Collapse all files

20
source/java/org/alfresco/repo/content/metadata/OfficeMetadataExtracter.java
View File

@@ -1,5 +1,5 @@
/*
* Copyright (C) 2005 Jesper Steen M<>ller
* Copyright (C) 2005 Jesper Steen M<>ller
*
* Licensed under the Mozilla Public License version 1.1
* with a permitted attribution clause. You may obtain a
@@ -28,7 +28,6 @@ import org.alfresco.repo.content.MimetypeMap;
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.DocumentSummaryInformation;
import org.apache.poi.hpsf.PropertySet;
import org.apache.poi.hpsf.PropertySetFactory;
import org.apache.poi.hpsf.SummaryInformation;
@@ -37,15 +36,16 @@ import org.apache.poi.poifs.eventfilesystem.POIFSReaderEvent;
import org.apache.poi.poifs.eventfilesystem.POIFSReaderListener;
/**
* Office file format Metadata Extracter
*
* @author Jesper Steen M<EFBFBD>ller
* @author Jesper Steen Møller
*/
public class OfficeMetadataExtracter extends AbstractMetadataExtracter
{
public static String[] SUPPORTED_MIMETYPES = new String[] {
MimetypeMap.MIMETYPE_WORD,
MimetypeMap.MIMETYPE_EXCEL,
MimetypeMap.MIMETYPE_PPT };
MimetypeMap.MIMETYPE_PPT};
public OfficeMetadataExtracter()
{
@@ -64,6 +64,7 @@ public class OfficeMetadataExtracter extends AbstractMetadataExtracter
if (ps instanceof SummaryInformation)
{
SummaryInformation si = (SummaryInformation) ps;
// Titled aspect
trimPut(ContentModel.PROP_TITLE, si.getTitle(), destination);
trimPut(ContentModel.PROP_DESCRIPTION, si.getSubject(), destination);
@@ -73,16 +74,6 @@ public class OfficeMetadataExtracter extends AbstractMetadataExtracter
trimPut(ContentModel.PROP_MODIFIED, si.getLastSaveDateTime(), destination);
trimPut(ContentModel.PROP_AUTHOR, si.getAuthor(), destination);
}
else if (ps instanceof DocumentSummaryInformation)
{
// DocumentSummaryInformation dsi = (DocumentSummaryInformation) ps;
// These are not really interesting to any aspect:
// trimPut(ContentModel.PROP_xxx, dsi.getCompany(),
// destination);
// trimPut(ContentModel.PROP_yyy, dsi.getManager(),
// destination);
}
}
catch (Exception ex)
{
@@ -90,6 +81,7 @@ public class OfficeMetadataExtracter extends AbstractMetadataExtracter
}
}
};
InputStream is = null;
try
{