Just omit the setting.  If hibernate doesn't see
hibernate.hbm2ddl.auto, it does nothing.


On 2/24/07, Kalle Korhonen <[EMAIL PROTECTED]> wrote:
In practice, no. I would have to mark a whole bunch of beans as lazy inited,
give up on auto wiring etc. But it does help to post your thoughts. I was
close before.. I don't see hibernate.hbm2ddl.auto=false documented anywhere
(I see one(!) dodgy google result for it), but that works, and then I can
just call validation myself at a later point. This way I don't need to
fiddle with anything else, but only need to make sure that I don't do
database operations before I do the validation, which is simple enough.

Kalle

On 2/24/07, James Carman <[EMAIL PROTECTED]> wrote:
>
> Can you just mark your session factory bean as lazy init in your
> spring application context file?  Or, mark the whole xml file as lazy
> by default?
>
> On 2/24/07, Kalle Korhonen <[EMAIL PROTECTED]> wrote:
> > Using Tapestry 4.0.2 and this is much more a hivemind question really,
> but I
> > wonder if anybody has tried to lazy initialize Tapestry-Spring.
> Currently,
> > if I don't initialize Spring context at the start-up, Tapestry-Spring
> fails
> > on ClassCastException. The use case I have is that I'm using Hibernate
> and
> > I'd like to deploy a new version of the web application using
> > hibernate.hbm2ddl.auto=validate. If the validation fails I'd just show a
> > status message like "Sorry, we are down for maintenance" until the
> schema is
> > updated, which I can in principle do if I delay creating the Spring
> context
> > and (re-)initialize it at some point later. However, ApplicationServlet
> > initialization fails when it tries to construct Hivemind registry. The
> > service point SpringApplicationInitializer has been marked private so I
> > can't override it externally. What I'm trying to do is to avoid granting
> > rights to modify the schema for the webapp's normal database access
> account,
> > and only update the scema externally with a different user credentials
> (more
> > detailed post about Hibernate validation from Spring's perspective at
> > http://forum.springframework.org/showthread.php?t=35274).
> >
> > Suppose I could implement a wrapper/extension for ApplicationServlet to
> > catch the exception and make it able to re-initialize, but then I'm
> already
> > implementing a wrapper for Spring context initializer and doing some
> other
> > tricks to get Acegi filters to initialize lazily, so overall I'm
> wondering
> > if I'm just complicating the design for nothing. Maybe somebody already
> has
> > a solution with schema validation or can see some completely different,
> but
> > simpler way to get to the same end result?
> >
> > Kalle
> >
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: [EMAIL PROTECTED]
> For additional commands, e-mail: [EMAIL PROTECTED]
>
>


---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Reply via email to