Path: utzoo!utgpu!news-server.csri.toronto.edu!rutgers!ucsd!ucbvax!ucdavis!csusac!unify!openlook!openlook-request
From: smith%mito@endor.harvard.edu (Steven Smith)
Newsgroups: comp.windows.open-look
Subject: Re: Hosed xnews menus under Open Windows
Message-ID: <8lcz7lv@openlook.Unify.Com>
Date: 10 Nov 90 17:14:20 GMT
Lines: 36

In article <1990Nov9.172649@wind.Eng.Sun.COM>, naughton@wind.eng.sun.com (Patrick Naughton) writes:
>Other followups to this message were referring to unrelated
>problems...  It sounds like you are on a GX and you have run some other
>program which directly accessed the framebuffer.  X11/NeWS is confused
>about the state of the GX hardware registers since it assumes that it
>will be the only one to leave state in the hardware.
...
>In article <1990Nov8.123120.20324@cs.utk.edu>, de5@ornl.gov (Dave Sill) writes:
>|> Has anyone seen xnews/openwin under SunOS 4.1 get into
>|> a funky state wherein menus are no longer drawn correctly?
>|> 
>|> Clicking on the triangle-in-a-square menu do-hicky has no
>|> visible effect until the mouse is moved down, then only
>|> the rounded menu labels are drawn as the pointer reaches
>|> them.


I have seen this problem as well.  It is not dependent on GX equipped
machines, but does seem to be a problem with COLOR!  I have an application
which uses a CMS Canvas, and when the CMS is set, scrollbar menus
demonstrate this problem.  However, when I use a monochrome canvas,
all is well.  The problem also arises in the drawing of the scrollbars, as
they do not use the proper "background" color so that they blend in with the
frame.  If color is disabled, the problem does not occur.  If  setting the
depth of the canvas is delayed until after the scrollbars are created
(a no-no) all is well until a split view is performed, at which time the NEW
scrollbar, and scrollbar menu have the display problem.  Eventually, the
menu problem causes the program to crash (at least my program) after a few
splits, and joins.

I have a piece of code which demonstrates the problem, and I would be happy
to give it to anyone (from SUN?) who is interested.

Steve Smith
Harvard Genome Lab
smith@nucleus.harvard.edu