Re: [Springframework-developer] Code changes for 0.8?

[Rod J. <rod...@in...>]

Sounds good. I didn't realise this about JDO. I doubt there are any specific
threading issues: a race condition would just initialize (correctly) twice,
I guess.

Rod

----- Original Message -----
From: <tri...@tr...>
To: "Rod Johnson" <rod...@in...>
Cc: <spr...@li...>
Sent: Friday, May 23, 2003 6:49 PM
Subject: Re: [Springframework-developer] Code changes for 0.8?


> Rod,
>
> A JDO Query supports compile(), but if the Query is not compiled when
execute()
> is called, then the Query is compiled automatically.  I think we should
provide
> the same functionality.  I have to look closer at the code to determine if
there
> are any threading issues we have to address.
>
> --Tomas
>
> > > While we are on the subject of JDBC.  Do we really have to require an
> > explicit
> > > compile for the RdbmsOperation and its subclasses.  Currently we throw
> > > InvalidDataAccessApiUsageException when we try to use an operation
that
> > has
> > > not been comiled.  We could just as easily execute the compile() at
that
> > > point, since we know what the problem is.
> >
> > Good question. I guess I was concerned about thread safety implications
of
> > compiling lazily. What do you think?
> >
> > Certainly there'd be a bonus in making the API slightly easier to use.
> >
> > Regards,
> > Rod
> >
> >
> >
>
>
>
>