Paul,

I still have this as a potentially open issue. Can we start from scratch on
it?

Thanks,
Jason Wharton
CPS - Mesa AZ
http://www.ibobjects.com

----- Original Message -----
From: "Paul Schmidt" <paul@...>
To: "Jason Wharton" <jwharton@...>; <IBObjects@yahoogroups.com>
Sent: Wednesday, January 31, 2001 8:43 AM
Subject: Re: [IBO] Exploding Application


>
>
> On 30 Jan 2001, at 13:01, Jason Wharton wrote:
>
> To: <IBObjects@yahoogroups.com>
> From: "Jason Wharton" <jwharton@...>
> Date sent: Tue, 30 Jan 2001 13:01:13 -0700
> Send reply to: IBObjects@yahoogroups.com
> Subject: Re: [IBO] Exploding Application
>
> > Paul,
> >
> > > The application is an MDI application. the child form in question is
> > > created and released from the stack, then after it's released,
> > > (using the release statement), and the main child form becomes
> > > active, it explodes. But so far it's after I modify a record. For
> > > the two variables involved, neither looks like a sane value.
> >
> > Unfortunately I don't use MDI ever so I don't know if there is
> > something about this that could be tripping you up.
> >
>
> I sent you another message with some more info on this....
>
> > > Now there is another MDI child form that is set up to work the same,
> > > and it doesn't give me an error. I am going to look at the
> > > properties, and make sure that they match up. If it's a property
> > > difference, I will let you know which one works, and which one
> > > explodes.
> >
> > Thanks, that will help.
> >
> > > > This suggests to me that there is either an uninitialized variable
> > > > or that something has performed an invalid write to memory and
> > > > trashed some of it. I am in favor of the thought that something is
> > > > getting trashed somehow.
> > >
> > > Gee I didn't think Delphi did that, then again anything is possible.
> >
> > This would be a bug in IBO most likely.
> >
>
> When I figured out it was CachedUpdates, I was thinking that it could
> be a pointer into some resource on the form (like the TIB_Query
> control), and CachedUpdates didn't realise that the form and
> therefore the TIB_Query didn't exist anymore, and that is what caused
> it to access memory that didn't belong to the application anymore.
>
> > > > What I need is a reproducible test case that duplicates this
> > > > problem and then I can look at it.
> > >
> > > > I have been somewhat buried in things lately so I am not certain
> > > > when I could have the problem resolved. I suspect in your efforts
> > > > to isolate this you will hopefully solve it while you are at it.
> > > >
> > > > Keep us posted on your progress and in the meantime I will keep in
> > > > the back of my mind what could be causing this.
> > >
> > > Will do, not sure how much I will do today, my wife spent the last
> > > month dying of a cold, now I have it, and had to cancel two
> > > important appointments today..... I may spend the afternoon
> > > watching Star Trek re-runs on the Space channel. I hear the remote
> > > calling me....
>
> Still suffering, my father used to swaer that the best solution was a
> good stiff shot of brandy. However as a Born Again Christian, I
> don't believe in solving problems with alcohol, although even the
> Puritans believed in "medicinal purposes".
>
> Oh well, I need a couple of days off anyway, (cough, hack, wheeze,
> sniff).
>
>
> Paul Schmidt,
> Tricat Technologies
> Email: paul@...
> Website: www.tricattechnologies.com
>
>
>
>