Fuse Integration Designer: access to table in relational database

[simonepringle]

I have seen documentation that indicates that it is possible to integrate with relational databases using direct JDBC access, OpenJPA, hibernate or TopLink.

It is not clear from the documentation how to approach access to a table in a relational database using the Fuse Integration Designer.  I tried using the Direct endpoint using a URI to a Derby DB (jdbc:derby://locahost:1527/testdb).

It's not clear what else to do from there - even assuming this is a possible way to access a relational DB...  In particular, it wasn't clear how to provide additional properties such as user name, password and target table(s).

Any help would be greatly appreciated.

Thanks,

Simone

[socallag]

Hi Simone,

It is possible to interact with Relational Databases using the FUSE Mediation Router (Camel) , the following documentation explains how this is done.

http://camel.apache.org/sql-component.html

http://camel.apache.org/jdbc.html

In the context of using the FID a Generic Endpoint can take any URL that is supported by Camel, which would include the jdbc: and sql: documented above, but you may required to do some additional configuration of such endpoints.

If you can give us a little more detail of your usecase we may be able to help some more.

Regards,

Seán

[simonepringle]

Hi, Sean!

Thanks for the quick reply.  I'll definitely take a look at the links you provided.

In the meantime, we're trying to set up an environment that makes it easy(ier) for folks to service-enable (read) access to data.  We are in an experimental phase of our efforts (i.e. 'researching'), so the first thing we were trying to do was to expose data from a single table as a service.

Imagine, for instance, that I have a table that summarizes reporting data (e.g. last month's statistics on web page visits - each page on the site tracking the total number of visitors per month).  I'd like to be able to access that data (read it) using a web service.

It seems that it should have been possible to do something (generically) like:

JDBC-aware end point -> Mapper -> Web-service publisher

Each of the steps above would be configurable:

     - JDBC-aware end point: connection string including login

     - Mapper: select and potentially manipulate row values and map to target format (XML)

     - Web-service publisher: WSDL-specifiable interface

Does that make sense?

Thanks for your help,

Simone

Edited by: simonepringle on Jun 24, 2009 6:30 PM

[abdeslam]

Hi simonepringle !

i have done same things without FID.

I'm very interested by your experience for integrating JDBC DB and mapping.

Can you share it ?

Regards

abdeslam