Path: utzoo!utgpu!jarvis.csri.toronto.edu!mailrus!iuvax!rutgers!apple!sun-barr!newstop!sun!quintus!pds
From: pds@quintus.UUCP (Peter Schachte)
Newsgroups: comp.windows.x
Subject: XtConvert()'s widget argument
Message-ID: <1210@quintus.UUCP>
Date: 20 Jul 89 17:12:50 GMT
Reply-To: pds@quintus.UUCP (Peter Schachte)
Organization: Quintus Computer Systems, Inc.
Lines: 35

I asked a similar question before, but I didn't get any replies, so I'm
trying again:  what should one pass for the widget argument to
XtConvert() when one doesn't have a widget yet (I need to convert
arguments in order to create a widget)?  I see two basic possibilities:

    1.  Pass the parent widget.  In intrinsic converters and all the
        widget sets I have source for, the converters that need extra
	arguments only need to look at the core part of the widget.  Is
	that guaranteed?

	The problem with this approach comes in converting arguments for
	the shell widget.  In this case, I don't have a parent widget to
	pass.  The only thing I can think of to do here is create a
	shell widget as a dummy, use it to do my conversions, and
	destroy it when I'm done.

    2.  Create the widget first, use it to convert the arguments,
	and use XtSetValues() to install them.  But am I guaranteed that
	creating a widget with an emtpy argument list and calling
	XtSetValues() to specify the initial arguments is equivalent to
	specifying the the arguments when creating the widget?  This
	certainly seems like the simplest and most obvious solution to
	the problem, if the two are equivalent.  But I seem to recall
	somewhere seeing resources that could not be set by
	XtSetValues(), but only at the time of widget creation.

I've read TFM, and can't find either guarantee.  Anyone have any
answers?  Has anyone already faced this question?  Any words of wisdom
would be greatly appreciated.

TIA,
-- 
-Peter Schachte
pds@quintus.uucp
...!sun!quintus!pds