Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!mnetor!uunet!husc6!mit-eddie!ll-xn!ames!sdcsvax!ucbvax!ZERMATT.LCS.MIT.EDU!RWS
From: RWS@ZERMATT.LCS.MIT.EDU (Robert Scheifler)
Newsgroups: comp.windows.x
Subject: re: fixes breaking
Message-ID: <871014101901.9.RWS@KILLINGTON.LCS.MIT.EDU>
Date: Wed, 14-Oct-87 10:19:00 EDT
Article-I.D.: KILLINGT.871014101901.9.RWS
Posted: Wed Oct 14 10:19:00 1987
Date-Received: Fri, 16-Oct-87 06:16:24 EDT
References: <8710131921.AA07321@frozen>
Sender: daemon@ucbvax.BERKELEY.EDU
Organization: The ARPA Internet
Lines: 33

I wouldn't normally blat to the whole list like this,
but since he did ...

    Date: Tue, 13 Oct 87 12:21:31 PDT
    From: franz!frozen!jdi@ucbarpa.Berkeley.EDU (John)

    Well, one of the newer fixes (last week or so) to the X11 server has made
    a couple things strange on my Sun 3/260 (running in mono mode).

    The scrollbar in xterm now works (it used to be always black) (I use
    reverse-video mode).

I find this a bit hard to believe.  I can't think of any of the changes
that would affect this.  I could not reproduce this on a Sun 3/110 here
(the scrollbar is still black with -rv).  Furthermore, on inspection, it
is pretty clear this is bogosity in lib/oldXtk/Scroll.c; the statement

  if (data->foreground == NULL)
	data->foreground = BlackPixel(data->dpy,DefaultScreen(dpy));

is effectively a no-op on VAXes and Apollos, but causes the bug you
mention on RTs and BW Suns (I can't say for sure about -mono mode).
This test for NULL is bogus, since zero is a perfectly valid pixel
value.  I suspect the correct fix is to simply delete this statement (at
least a quick test produced correct results on an Apollo and a Sun), but
this is *not* an official bug fix.

    However, the cursor now looks very strange.  It is like the old cursor
    but surrounded in a white box.  Ie: the whole area contained by the cursor
    that isn't defined in the mask doesn't show through any more -- it's white.

I cannot reproduce this on a Sun 3/110 or on an Apollo (or on a QVSS,
although it would be difficult to prove much about cursors on a QVSS).