[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Bug#239510: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=239510



On Thu, Apr 08, 2004 at 12:54:41AM -0500, Branden Robinson wrote:
> On Tue, Apr 06, 2004 at 06:24:10AM -0400, Thomas Dickey wrote:
> > I see some indication in the code (will have to test/see) that it might have
> > changed the way the widget is initialized.  I can verify that by making a
> > debug-version on a "good" system.
> 
> Okay, consider me interested in what you learn.

I ran a quick test on a Solaris 8 box, and see that the underlying cause
does show up in my debug trace.  See in charproc.c

	TRACE(("Color resource initialization:\n"));
	TRACE(("   Default foreground %#lx\n", wnew->dft_foreground));
	TRACE(("   Default background %#lx\n", wnew->dft_background));
	TRACE(("   Screen foreground  %#lx\n", request->screen.foreground));
	TRACE(("   Screen background  %#lx\n", request->core.background_pixel));

For Solaris 8, the latter two values change when -rv is used, for XFree86 they
do not.  (The first two change as expected for both).

Checking an old XFree86 (3.3.1), I see that it still does not in that case.  So
it's something that I simply didn't notice (or have forgotten) since the
instances where I would do "xterm -rv" were mostly on the X11R5 or X11R6
systems.  I seem to recall some discussion of this (regarding the X server),
but that would be quite a while ago.

I have this in my resource file, so it's not simply that I don't use the -rv
option, but had formed a habit:

XTerm*VT100*translations:       #override \n\
        Shift<Key>KP_Multiply:  set-reverse-video(toggle)\n\

I'll probably make a workaround for this as I do for the cursor color (check if
default foreground is the same as screen background, etc.), unless I can find a
more reliable way (will have to read some code, but my reading of it the other
night led me to suspect that I can't do it in a straightforward manner).

To explain:  I'm testing the default foreground/background colors to ensure the
cursor is visible, but using the screen colors to setup the graphics.  The fix
isn't as simple as using different variables (since they do have different
values if the -fg, -bg options are used), but I can probably get it to work.

In any case, it's not a new bug.

-- 
Thomas E. Dickey
http://invisible-island.net
ftp://invisible-island.net

Attachment: pgpRUlM3F1Eyn.pgp
Description: PGP signature


Reply to: