> From: Matt Birkholz <m...@birchwood-abbey.net>
> Date: Sun, 5 Jun 2016 00:06:47 -0700
>
> > From: Patric Jonsson <pat...@kth.se>
> > Date: Fri, 3 Jun 2016 17:47:31 +0000
> >
> > [...]
> > ;;; Bug ported to scheme in new x11/x11base.scm (reading the source,
> > ;;; methinks this can't be right)
>
> Thanks for giving it a whirl. I'm happy to take a closer look knowing
> that someone cares. But Phoenix Comicon is happening and she's turned
> me into an elf for the weekend.
I got better.
And I believe I have fixed this. I applied the change you suggested
to both microcode/x11base.c and x11/x11base.scm.
fd4ff59 microcode/x11base.c (char_ptr_to_prop_data_32): Fix for 64bits.
I also fixed x_get_window_property so that it actually returns
something (blush). Now I get the same list of atoms on 32bit and
64bit machines.
I added tests to x11 and x11-screen so that `make check' (when DISPLAY
is set) will print the list (as well as create a window). I did not
test the old μmodule. And I did not mess with x-change-property
though a quick read of the manpage suggests it has the same problem.
"If the specified format is 32, the property data must be a long array."
If someone wants to volunteer an x-change-property test, I might try
to fix it.
_______________________________________________
MIT-Scheme-devel mailing list
MIT-Scheme-devel@gnu.org
https://lists.gnu.org/mailman/listinfo/mit-scheme-devel
