views:

276

answers:

3

I am trying to get info from an embedded db called NexusDB using java.

Alternative 1:

I've read in NexusDB website that there is an ODBC driver so I might use it with unixODBC. Then I need to do a JDBC-ODBC Bridge as stated here.

Alternative 2:

Get some sort of application to migrate NexusDB db to another db. Would like to know one.

I would like to know if anyone ever this this, what's the best solution?

Thanks for reading.

+1  A: 

Alternative 1:

Not possible.

unixODBC needs linux drivers and there aren't for nexusDB.

Alternative 2:

Didn't find any.

Solution so far

Writting a small webservice with delphi or get odbc and use that in a small proxy. In other words, instead of connecting to the nexusdb server you connect to a dedicated application or webservice pass on the information and that app does connect to nexusd and writes the data.

Macarse
Do you need to use this db outside of java? You might consider moving to a non-embedded db engine which has proper jdbc support.
kd304
Information I need to query comes from an output of an application which is a zipFile containing nexusdb embbeded db. I can't move to a non-embedded db. All the same, nexusdb isn't only embedded, it's just this particular case.
Macarse
A: 

Hi,

If you connect to Nexus via .NET maybe you can use IKVM to run your Java app and connect through the .NET api?

Otherwise I think your own "solution so far" sounds ok.

Best Anders

anders.norgaard
A: 

Since it is embedded (and you probably have full access to the machine), what about copying database files to a Windows system, setting up NexusDB and its ODBC driver, then using an ODBC-JDBC bridge on that machine? Once the bridge is running, you can set up a new JDBC-compatible DBMS of your choice, connect, and use a little code to SELECT from NexusDB and insert to your new database.

If you're looking to migrate to another embedded DBMS, I'm using H2 Database for Java, and I've got to say it's really sweet. Both embedded and client/server modes, cross-platform, and really fast for anything under 1M rows. Supports pretty much any feature you'd have in an embedded DBMS and then some.

BobMcGee