> Things most likely changed and for a good reason. However if this does
> not sound like the result of a modification to Jaybird.

It is. We have now JDBC-specs conformant behavior in auto-commit mode. See
release notes and JDBC specs for information when result set is closed.

> Then a bug? Or improper usage? I have little to know control over how OO
> accesses the db, transaction wise and etc. (At least to my knowledge)
>
> Any thoughts, comments, feedback, etc is all greatly appreciated.

We need somehow to switch the auto-commit mode off. On the other hand, I
think we can add in next release a parameter that would construct holdable
result sets by default (in that case result set won't be closed but cached
completely in memory).

Roman