Path: utzoo!attcan!utgpu!jarvis.csri.toronto.edu!cs.utexas.edu!tut.cis.ohio-state.edu!ucbvax!bloom-beacon!EXPO.LCS.MIT.EDU!kit
From: kit@EXPO.LCS.MIT.EDU (Chris D. Peterson)
Newsgroups: comp.windows.x
Subject: Re: help with athena widgets
Message-ID: <8912132238.AA15157@expo.lcs.mit.edu>
Date: 13 Dec 89 22:38:13 GMT
References: <8912132142.AA23380@cunixf.cc.columbia.edu>
Sender: root@athena.mit.edu (Wizard A. Root)
Organization: The Internet
Lines: 35


> I've gotten my program to do mostly what i want.
> what i ended up doing is setting XtNoverrideRedirect to True for a
> topLevelShell widget

This is okay for a demo, but really EVIL for a real application, since it does
not let the user reposition this window with the window manager.  The rule of
thumb that I use is not to use OverrideRedirect unless there is a server grab
active for the entire time the window is active.  Have you tried using a
transient shell instead?

> but i did have to place the line
> 	{ "select-start", NULL },
> into my action table to trick the action parser into thinking it knew
> where that routine was.

This is probabally because you are using a resource line of:

*Text*tranlations:

This assignes translations to the children of the Text widget, which in this
case is a scrollbar that does not know about the select-start action.

> is it true, then, that window manager like things (geometry, decoration)
> for the toplevel widget (that created with XtInitialize) must be
> modified only from the window manager end?

I am not sure what you are asking, these are things the window manager controls,
and as such only the owner of the window manager can modify them, or be 100%
certain that things act a certain way.  There are certain conventions described
in the ICCCM manual which help application programmers do the "right" thing.



						Chris