That subject line makes almost no sense... Sorry. Anyway; we've been generating our persistence layer from within eclipse using the hibernate-tools, but we're moving it out to an ant task so we can control the environment a little better.

We have a USR table that is mapped like this to get the object to be 'User'.



This works great, however...

When we generate in eclipse, we get methods like this:
User getUsrByUsrId()

Outside of eclipse we get this:
User getUserByUsrId()

Note that the second one is using the class we've defined (getUser) in the method name, and the first isn't (getUsr). I like the second one better actually, but all of our code was written using the first style.

So, my question is why might they be different? I tried to use exactly the same jars (and versions) the eclipse reverse engineering uses, but I may have missed something. Or, is there a setting I'm missing?

Any help would be greatly appreciated.

Thanks
M

this is controlled by the reverse engineering strategy and I don't think you are using the same version in eclipse and ant (are you actually using the jar from inside eclipse plugins ?)

Yep. Turns out the eclipse platform was indeed using beta7 instead of beta9a, as I had been led to believe. Running the ant version with the beta7 libraries resulted in consistent results. Thanks much.