In Pentaho a similar possibility like the oda driver in birt?

[ManInBlack]

Hi,

in BIRT I can develop an "Open Data Access" (ODA) - Driver. That is a fine thing.
Possibly it could be programmed a oda driver, which can access a heterogen system (where therefore isn´t provide no driver, like jdbc, etc.)

But is there a similar feature in pentaho, too?

Thank you for your answer.

Bye
ManInBlack

P.S.
I haven´t found a solution by using the forum- or google search.

[ingo.klose]

I don't know if that is the answer to your question, but starting 3.5 you can use Pentaho Data Integration to feed data into your report and thus have all the flexibility that Pentaho Data Integration has to offer.

In addition you can roll your own data source using Java code.

[Taqua]

if you have a Java-class that can generate a Swing-Tablemodel with the data you like, then you can use the "Java Method Call" datasource, or you can use the scripting datasource, or you can use the Kettle-datasource. For 99% of all cases, these options should be sufficient.

If you want to go the whole way, you can finally implement your own datasources (along with an UI to configure it - this route is definitely not a quick and dirty one). Rolling your own datasources along with the UI work aims at integrators, where the effort to implement the datasources is dwarfed against the gains in everyday productivity.


The engine itself really does not care where the data comes from, as long as you ship the data in a Swing-TableModel. We also do not care whether you use a database, carrier pigeon or messages to a deity of your choice to get the data - as long as the result is a TableModel that is suitably sorted for the report.