Subject | Re: [Firebird-Java] Re: Serious issue or bug with transactions |
---|---|
Author | David Jencks |
Post date | 2002-04-11T21:50:35Z |
Autocommit does work sometimes, right? So there must be commit happening
sometimes with autocommit -- besides I coded it in ;-) It's definitely not
in a call to connection.commit(), though.
david jencks
sometimes with autocommit -- besides I coded it in ;-) It's definitely not
in a call to connection.commit(), though.
david jencks
On 2002.04.11 17:37:26 -0400 rrokytskyy wrote:
> Hi,
>
> I just looked through the code and I didn't find the con.commit().
> There might be a bug in driver (or more correctly, non-implemented
> specs) when driver should commit in setAutoCommit(...) and
> setTransactionIsolation(...) methods. I have to check the specs. If
> this is true, then it explains the behaviour: you do not see the
> change made in one transaction from another transaction until you
> commit that transaction. You might try to set transaction isolation
> to "dirty read" and example should work.
>
> I will check the specs on this issue. Also, it might take some time
> to add this feature because we have to check if there's no side
> effects in case of JCA. But you might add this to your source code if
> you have non-managed case: in methods FBConnection.setAutoCommit
> (boolean) and in FBConnection.setTransactionIsolation(int) add the
> following code on the begining of the method:
>
> if (inTransaction())
> commit();
>
> But again, I'm not sure if this is the reason, and if the code above
> is compliant with specs.
>
> Best regards,
> Roman Rokytskyy
>
> P.S. BTW, are encodings working in your application?
>
>
>
> To unsubscribe from this group, send an email to:
> Firebird-Java-unsubscribe@yahoogroups.com
>
>
>
> Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/
>
>
>
>
>