Re: [Wxruby-dev] OS X 10.2.6 compile error

[Kevin Smith]

On Thu, 2003-07-10 at 14:43, John Platte wrote:
> And as I mentioned in my post a few minutes ago, now the window doesn't 
> come up at all. I think this happened as of the CVS checkin on the 8th 
> abt 20:00.

That's really, really strange. The only change that should affect any
code that's being executed for you is the evt_menu diagnostic attempt.
I've taken that back out (since that problem seems to be fixed). I can't
see anything else that could possibly have an effect. But I suppose
things can be deceiving.

> But samples/minimal/minimal.rb almost completely succeeds. There's no 
> menu bar, if it's supposed to have that, and I can't bring the window 
> in front of other programs' windows. BUT I get no errors in dialogs or 
> STDERR, and I can activate, resize, and close the window.

Minimal is supposed to have a window. And it does some other things, so
it isn't really "minimal". I've added a nothing.rb in that same
directory that is REALLY a minimal wxruby app. Please try it and see if
it is any better (or worse).

> For some reason, samples/artprov still gives me the "no bitmap handler 
> for type 9 defined." dialog. 

artprov.rb is one of the stranger samples, so I wouldn't rely too much
on it. I would recommend controls.rb as a complex but "normal" sample.
Most of the rest are also less unusual.

> None of these have burped out to STDERR.

Can you try running them from a shell command line? We might see better
diagnostics.

> (Just for kicks, can I try running Redwood to see what works and what 
> doesn't?)

As soon as I put together some kind of release, sure. It's a pretty
complex UI, so I don't expect it to be any better than the
samples--maybe toolbar.rb for instance. It also requires the 'nmh' set
of mail tools to get anything done.

> Again, this message is based on the CVS head at 2003-07-08T19:00. The 
> checkin after that seems to've broken my machine worse.

If this is still the case, I'd appreciate if you could do a diff between
your tree version and the sourceforge tip. Maybe that will reveal
something that I'm not seeing.

Thanks again,

Kevin