[comp.windows.x] R4 twm problems

grady@fxgrp.fx.com (Steven Grady) (01/06/90)

I'm running X11R4 on a color Sparcstation-1, compiled with cc.
I'm having some problems with the new twm.
(Should I have sent this to xbugs as well?)

1] With the following binding/function, grid lines are left around on
obscuring windows:
    Button1= m      : all   : f.function "lower-or-resize"

    NoRaiseOnResize
    Function "lower-or-resize"
    {
	f.resize
	f.deltastop
	f.lower
    }
Try putting a window in front of another one, then hit meta-button1
to lower it.  The grid lines were left around.  I tried this both
with XorValue unspecified (ie the default) and set to 1.

2] In general twm doesn't handle deltas quite right.  The behavior
I'd like to have is for meta-button1-delta to resize, and
meta-button1-nodelta to lower (which is what I had from uwm,
which was similar to xwm).  Instead the window is always lowered.
This is because f.resize is asynchronous.  Since it continues immediately,
the user hasn't moved the pointer yet, so there is no delta, so
f.deltastop doesn't get invoked, so f.lower gets invoked.  Instead
it should act like f.move, which operates synchronously.  Since resizes
grab the server (see the beginning of StartResize()), it would probably
just as simple to make f.resize synchronous as well.

<These are requests rather than bugs>
3] There should be two different functions (eg f.iconify and f.toggleicon),
to allow something like:
    "F1" =	: "XTerm": f.iconify
which would make F1 iconify all xterms.  With the current form of f.iconify,
this command iconifies uniconified xterms, and uniconifies iconified
ones.

4] It would be nice if there were an f.warpback function, which warped
the pointer back to where is was before the last f.warp* function.

Other than these things, way cool, guys.
-- 
	Steven
	...!ucbvax!grady
	grady@postgres.berkeley.edu

"There's another weird lizard farm coming up fast at eleven o'clock."
"I feel the warmth of its presence, Sam."