Or, more trivially, just call it directly (_XmVendorExtRealize()).
Neither closure nor call_data is used. A workaround would be a
private unsafe version of XtAdd/Remove/Call/Callback; it would be
thread safe if you knew the callers were. You might want more than
one function to be called at realize time.
Mitch
> -----Original Message-----
> From: Martin Simmons [mailto:[EMAIL PROTECTED]]
> Sent: Wednesday, July 19, 2000 3:35 PM
> To: [EMAIL PROTECTED]
> Cc: [EMAIL PROTECTED]; [EMAIL PROTECTED]
> Subject: Re: BaseClass (was Broken pipe with Grace)
>
>
> >>>>> On Wed, 19 Jul 2000 11:23:57 -0400, "Mitch Miers" <[EMAIL PROTECTED]>
>said:
>
> Mitch> Uh, oh. I think there's another problem here. People who use
> Mitch> the threadsafe version of libXt may see a segfault when the
> Mitch> callback is called - XtCallCallbackList in this case needs
> Mitch> XtWidgetToApplicationContext() to work, which it doesn't for
> Mitch> ExtObjs. It works ok as long as you don't use the threadsafe
> Mitch> version.
>
> Does the realize_callback (XmNrealizeCallback) really need to be a resourced
> XtCallCallbackList? I can't find it anywhere in the Motif man pages and
> _XmVendorExtRealize is the only function added to it in Lesstif, so perhaps it
> can be done with a function slot (c.f. wrap->realize)?
>
> __Martin
>
