oodt-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Gerard, Ryan S. (GSFC-586.0)[COLUMBUS TECHNOLOGIES AND SERVICES INC]" <ryan.s.ger...@nasa.gov>
Subject RE: postgres DB
Date Fri, 09 Mar 2012 18:23:54 GMT
Thanks very much Chris. To answer your question, yes, I am attempting to hook up DataSourceRepositoryManager
and DataSourceCatalogManager to Postgres. I actually like the XMLRepositoryManager so I would
be comfortable defining some configuration here.  Thanks very much for the patching instructions.
I'll let you know how we plan to proceed. 

Thanks,
Ryan

---
Hi Ryan,

On Mar 8, 2012, at 9:54 AM, Gerard, Ryan S. (GSFC-586.0)[COLUMBUS TECHNOLOGIES AND SERVICES
INC] wrote:

> I apologize if this should have gone to the user list.

No worries! This is perfect, read on below.

> 
> I have configured my filemanager to use the Postgres for the database.

Cool. Just to be sure, are you referring to using Postgres as the DataSourceCatalog (Catalog
extension point) and as the DataSourceRepositoryManager (RepositoryManager extension point)?
Both, or one or the other?

> I am trying to run the example(blah.txt) but I noticed the product-types.xml file defines
some extractors of metadata and not all of the metadata tables are getting filled out.

Ah, yeah I think that this is because the XMLRepositoryManager has evolved over time and has
been updated, but the DataSourceRepositoryManager hasn't necessarily been updated to reflect
the new data structures for Product Types over time.

> Using DataSourceRepositoryManager instead of XMLRepositoryManager, I see no way to define
such extractors (via the DB or any other way).

Definitely possible that that's the case. I don't think a lot of folks use the DataSourceRepositoryManager.
Would you be comfortable using the XML files for now?
Another option is to help work on patching the DataSourceRepositoryManager to read and insert
that extra information into the DB. You game? The process would
be:

1. File a JIRA issue [1].
2. Present a patch 
   - to create the patch, svn co http://svn.apache.org/repos/asf/oodt/trunk oodt
   - cd oodt/filemgr/...
   - edit/add code
   - svn status (make sure your changes were picked up)
      - you might have to svn add any new files here
   - svn diff > OODT-xxx.rgerard.<yyMMdd>.patch.txt (where xxx is the JIRA issue
ID from #1) 3. Attach patch from #2 onto #1.

> Do I need to install the another component to define the extractors or can it be done
within the filemanager component as it was within product-types.xml file?

At the moment I think it's only supported with the XMLRepositoryManager. It would be great,
if you are committed to using the DataSourceRepositoryManager, to add the extractor functionality
to it. You game?

At the moment, I believe it still works, you are using missing on the the ability to do File
Manager server side extraction and in turn, missing out on the automatically populated met
keys provided by the CoreMetExtractor. If you are game to work on a patch I'll help you shepherd
it in.

Thanks!

Cheers,
Chris

[1] http//issues.apache.org/jira/browse/OODT

> 
> Thanks very much.

Anytime!

++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
Chris Mattmann, Ph.D.
Senior Computer Scientist
NASA Jet Propulsion Laboratory Pasadena, CA 91109 USA
Office: 171-266B, Mailstop: 171-246
Email: chris.a.mattmann@nasa.gov
WWW:   http://sunset.usc.edu/~mattmann/
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
Adjunct Assistant Professor, Computer Science Department University of Southern California,
Los Angeles, CA 90089 USA
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++


Mime
View raw message