Thomas,

Thanks for your response.

> Well, why do you want autocommit to be disabled by default?
- For my usage pattern I don't want autocommit - I need to specify my
transaction boundaries and it spans execute statements.

> I don't think other databases support disabling autocommit in the connection 
> URL (do you know any?)
- No.

> the connection pool implementation (violation of the spec). I don't see a way 
> to implement your request, sorry.
- As you mentioned, my request would be a violation of the connection
pool spec.  Fair enough.  I was simply expecting that the connections
as I defined them for the pool would continue to behave as I defined
them.  The fact that I could put 'AUTOCOMMIT=false' in my URL led me
to believe that those connections would always work that way.  I don't
have a problem adding 'connection.setAutoCommit(false)' whenever I
pull a connection.  I just had not included it because I figured the
URL definition took care of it for me (and it fact it does for the
first connection...).  Based on my incorrect expectation of behavior I
was getting some rather odd bugs like deadlocks which shouldn't have
been possible in the usage pattern I "thought" I was getting.

Now, having said any of this - H2 is awesome and I appreciate your
time.  I think if anything simply documenting this behavior with the
references to AUTOCOMMIT would be really helpful for others to avoid
my mistake.

Thanks
Joe


On Apr 16, 1:40 pm, Thomas Mueller <[email protected]>
wrote:
> Hi,
>
> Well, why do you want autocommit to be disabled by default? That you
> can disable autocommit in the database URL is more an 'accidental
> feature', I don't see it as something that needs to be supported for
> all cases. I don't think other databases support disabling autocommit
> in the connection URL (do you know any?), so your application would
> not be portable. Instead, I suggest to call conn.setAutoCommit(false).
>
> I think that any connection pool (not just the built-in pool of H2)
> must reset autocommit at that point. Not doing so would be a bug in
> the connection pool implementation (violation of the spec). I don't
> see a way to implement your request, sorry. Or do you (or anybody
> else) know a way?
>
> Regards,
> Thomas
>
> --
> You received this message because you are subscribed to the Google Groups "H2 
> Database" group.
> To post to this group, send email to [email protected].
> To unsubscribe from this group, send email to 
> [email protected].
> For more options, visit this group 
> athttp://groups.google.com/group/h2-database?hl=en.

-- 
You received this message because you are subscribed to the Google Groups "H2 
Database" group.
To post to this group, send email to [email protected].
To unsubscribe from this group, send email to 
[email protected].
For more options, visit this group at 
http://groups.google.com/group/h2-database?hl=en.

Reply via email to