On Thu, Feb 23, 2017 at 04:16:49PM +0100 I heard the voice of
Alexander Klein, and lo! it spake thus:
>
> So,
>
> > - tmp_win->wmhints->input = True;
> > + //tmp_win->wmhints->input = True;
>
> seems to be the right way to go!
Alrighty then. It does feel like a moderately bogus thing to do.
Other places we force (now more strongly than before) focus handling
to be enabled for windows that don't tell us anything, but that forced
it on for windows that explicitly told us not to as well.
OTOH, it's been there for a while; it was added in 3.1p3, and that was
a long time ago. So it's presumably there for a reason, though I'm
not sure what. That's long before we have any detailed VCS history,
so the nearest I can see is
I have considerably reworked the focus handling. So I have
probably introduced some problems.
in the 3.2 CHANGES. The question is, what does it fix, and how do we
do both at once? vtwm doesn't seem to globally override a negative
request from _WM_HINTS. fvwm is harder to tell, and seems at least to
most explicitly deal with ICCCM's 4-ple of states, but it doesn't
appear that it does any global overrides either.
So, it may be that we don't actually need it at all, at least outside
of potentially very specific cases. I propose taking it out as an
unconditional, and adding a config var to enable it (possibly with a
win-list, so you can just set it on specific windows that turn out to
need it).
--
Matthew Fuller (MF4839) | [email protected]
Systems/Network Administrator | http://www.over-yonder.net/~fullermd/
On the Internet, nobody can hear you scream.
