I’m just starting to explore Couchbase, as it’s been prescribed for a project I’m working on.
I’m considering which of the three JDBC driver implementations I should consider, or whether I should consider JDBC at all. I’m working on porting an existing app using Oracle JDBC, so a JDBC solution would be convenient.
The two JDBC drivers recommended on the Couchbase website are commercial, with trial periods.
There is also a JDBC driver on Github, at: https://github.com/jdbc-json/jdbc-cb . Does anyone have experience using this driver who can say it’s worth using?
While there is no hard-and-fast rule, the JDBC drivers are primarily targeted at existing platforms which expect a JDBC interface. Examples would be things like reporting tools that are built on JDBC’s ability to get schema, iterate over large data sets, use accessors.
For many application cases though, say a Java app, directly using Couchbase’s SDKs at a DAO layer or whatever is appropriate in your app or framework may be lower friction and more flexible. A good example of this is Spring Data’s repository pattern and exposing the different services from Couchbase through Spring Framework.
So, it’s hard to give a blanket answer, but hopefully that helps as you evaluate things.