When a keyboard grab activate/deactives, the input focus jumps to the window
which does the grab. These FocusIn events sometimes managed to confuse awesome.
The symptom was that a newly mapped/started client didn't receive the input
focus:
- You press your key-combo to start a terminal. This activates a passive grab
and the input focus jumps to the root window
- The terminal opens and the "manage" rule does client.focus = c
- This doesn't set the focus yet, but instead causes
globalconf.focus.need_update = true and .client = new_terminal
- Before the focus is updated, the key combo is released and the focus jumps
back to the previously focused client
- The FocusIn with mode == Ungrab causes awesome to think that the previously
focused client got focused again and thus globalconf.client.focus gets re-set
to what it was before
- Finally the focus is refreshed and the previously focused client gets focused
The fix is simple: We don't need the FocusIn events which are generated when a
grab activates or deactivates, so we can just ignore them.
Thanks to Majic for reporting this to me.
Signed-off-by: Uli Schlachter <psychon@znc.in>
xcb_event_get_{request,error}_label() return NULL for non-core requests and
error codes. Obviously, that's not really helpful at all, so also print the
numeric information about what failed.
Signed-off-by: Uli Schlachter <psychon@znc.in>
Thanks to muni who managed to crash his WM with conky. When conky runs in its
own window, clicking on it will sometimes (depending on the conkyrc) use
XSendEvent() to forward the event to the WM.
Signed-off-by: Uli Schlachter <psychon@znc.in>
Returning true from the callback just to signal keygrabber to continue grabbing
felt redundant (and silly :|).
This will break old code that relied on returning false to stop grabbing,
instead of calling keygrabber.stop.
And fix keygrabber docs.
Signed-off-by: Anurag Priyam <anurag08priyam@gmail.com>
Signed-off-by: Uli Schlachter <psychon@znc.in>
This option is no longer valid in modelines, so it has been removed from
all modelines using the following shellscript:
#!/bin/ksh
git ls-tree -r HEAD | cut -f2 | while read f; do
egrep -e '^(//|--) vim: .*encoding=' $f >/dev/null || continue
sed -E -e '/^(\/\/|--) vim:/s/:encoding=utf-8//' $f > /tmp/foo
mv /tmp/foo $f
done
Signed-off-by: Gregor Best <gbe@ring0.de>
Signed-off-by: Uli Schlachter <psychon@znc.in>
X11 only let's you query the state of mouse button 1 to 5. However, it can
generate ButtonPress and ButtonRelease events for up to 256 mouse buttons.
Instead of asking the server which buttons are pressed, we will now remember
the button state from those ButtonPress and ButtonRelease events. Currently
this let's us keep track of up to 32 mouse buttons.
Signed-off-by: Uli Schlachter <psychon@znc.in>
xcb-util is now split into several repositories since 0.3.8. This
release also cleaned up the API a lot, thus update the code
accordingly.
Signed-off-by: Arnaud Fontaine <arnau@debian.org>
Signed-off-by: Uli Schlachter <psychon@znc.in>
Both client_isvisible() and client_maybevisible() where almost exclusively
called with the client's screen as their second argument. Remove this second
argument and let these functions instead always act on the client's screen.
Signed-off-by: Uli Schlachter <psychon@znc.in>
event_button_callback() pops the top element from the stack, but
event_emit_button() doesn't. This got was expecting both to pop the top from the
stack so Bad Things(tm) happen.
Signed-off-by: Uli Schlachter <psychon@znc.in>
When we receive a DestroyNotify, we have to unmanage the client. However, we
were just informed that the window was destroyed, so it doesn't make sense to
try to do "something" with the window.
This commits fixes some errors from the X server (which we ignored, but still
they are gone now).
Signed-off-by: Uli Schlachter <psychon@znc.in>
These signals also contain the x and y coordinate of the event which the
capi.button()-based signals can't provide.
Signed-off-by: Uli Schlachter <psychon@znc.in>
This makes awesome support only a single X11 protocol screen. If you are still
using zaphod mode, you can run multiple instances of awesome on the single
screens, e.g.:
DISPLAY=:1.0 awesome & DISPLAY=:1.1 awesome &
Signed-off-by: Uli Schlachter <psychon@znc.in>
This reverts commit 0682499895.
We have to handle all the UnmapNotifies due to ICCCM. The bug this was trying to
fix is now fixed by 9ab8cd4039 ("Set our event
mask on the root window later") instead.
We got XCB_EVENT_MASK_SUBSTRUCTURE_NOTIFY on the root window, but we also have
XCB_EVENT_MASK_STRUCTURE_NOTIFY on the individual client windows. Since we are
reparenting our windows, we can safely ignore all UnmapNotifies that we get
through the root window.
This fixes a bug if there were any windows already present before awesome
started. Those were obviously already mapped which means that ReparentWindow
would cause an UnmapNotify for them on the root window. This commit makes us
ignore that UnmapNotify and thus not throwing away the window immediately again.
Signed-off-by: Uli Schlachter <psychon@znc.in>
When a window is moved, it should be sent a synthetic ConfigureNotify describing
its new position. This also documents why we send a synthetic event if nothing
was changed in response to a ConfigureRequest.
Signed-off-by: Uli Schlachter <psychon@znc.in>
When a client is unmanaged, we destroy our frame window. But since the client's
own window was still a child of the frame window, it was destroyed, too. This
commit fixes this by reparenting the client's window back to the root window
first.
Signed-off-by: Uli Schlachter <psychon@znc.in>
ICCCM mandates that window managers have to trigger the conversion to withdrawn
state on the synthetic and the real unmap notify, no matter if any is missing.
Previously we couldn't do this, but thanks to the reparenting, we can now do
this properly.
Signed-off-by: Uli Schlachter <psychon@znc.in>
This splits up client_getbywin() into two different functions. One of them finds
a client by its frame window, the other checks against the client's own window.
Signed-off-by: Uli Schlachter <psychon@znc.in>
Each window is now reparented under a window created by awesome. This window is
NOT visible, this commit should not cause any visible differences. Instead, this
is done to work around problems with X11.
The only alternative would be to ban windows by moving them offscreen which is
ugly in itself and breaks pagers. Doing it like this seems to be the better
solution.
Signed-off-by: Uli Schlachter <psychon@znc.in>
We no longer use libxcb-event, so we don't have to follow it's API any more.
This means the pointless arguments and return values can be removed.
Signed-off-by: Uli Schlachter <psychon@znc.in>
Upstream removed most of the code from libxcb-event and there is no
event-related stuff left in this library. We now no longer use any of the parts
that were removed.
Signed-off-by: Uli Schlachter <psychon@znc.in>
This commits makes awesome remove systray windows from the save set when they
are unmapped. This happens for the same reason as with normal client windows.
Signed-off-by: Uli Schlachter <psychon@znc.in>
This will store the widgets references that the wibox have inside their
environment table, and not in the global registry, avoiding memory leaks.
This should fix FS#771.
Signed-off-by: Julien Danjou <julien@danjou.info>
Second try:
Turns out I messed up with XCB_EVENT_SENT() and had a "!" too much. The old code
already tried to cope with this, but forgot to actually unmap the window which
it just set to withdrawn state.
This time I tested the patch *again* and now I found even less bugs than on my
last try.
P.S.: I suck.
Signed-off-by: Uli Schlachter <psychon@znc.in>
Signed-off-by: Julien Danjou <julien@danjou.info>
According to the Inter-Client Communication Conventions Manual, if you want to
switch your window to withdrawn state, you unmap it and send a synthetic
UnmapNotify to the root window.
This synthetic event fixes a race condition. When you map and unmap a window
quickly, the map will generate a MapRequest for the WM but won't actually map
the window. Thus, the unmap will be discarded (-> window not yet mapped) and the
window stays map once the WM handles the MapRequest
Before this patch, awesome just ignored the synthetic unmap notify which caused
the bug to appear again. With this patch it doesn't happen anymore.
Signed-off-by: Uli Schlachter <psychon@znc.in>
Signed-off-by: Julien Danjou <julien@danjou.info>
This changes wibox_t::mouse_over to a proper reference. That way one can't
remove that widget from underneath us which would lead to an unprotected lua error.
Signed-off-by: Uli Schlachter <psychon@znc.in>
Signed-off-by: Julien Danjou <julien@danjou.info>
Wiboxes are lazily updated. This means that we could receive an expose event on
them between making them visible and actually painting their content.
Due to this we were copying undefined content to the wibox, because the pixmap
was only created just now, but it wasn't actually filled with anything yet.
Signed-off-by: Uli Schlachter <psychon@znc.in>
Signed-off-by: Julien Danjou <julien@danjou.info>
This avoid the problem we can had while modifying globalconf.keys in a
callback function from a key or button event. Now we push all matching
objects on the stack, and call each callback function. If something
modify globalconf.keys, we do not care.
Signed-off-by: Julien Danjou <julien@danjou.info>
This variable didn't really have any effect on anything. It was only ever
checked in a function that would never be called if this flag was false.
Signed-off-by: Uli Schlachter <psychon@znc.in>
Signed-off-by: Julien Danjou <julien@danjou.info>
This patch fixes focus hooks calls - for every call to focus hook, there
should be call to unfocus hook.
Focus related info:
In this shape, awesome doesn't support multiple focused clients, that
means it follows the rule "there is only a single focus", which is not
true for MPX. To change this, I think it will need some magic with
FocusOut events handling and changes to some structures
(e.g. globalconf.screen_focus, screen_t.client_focus should be arrays) :p
Now we don't need to handle FocusOut events.
Signed-off-by: Mariusz Ceier <mceier@gmail.com>
Signed-off-by: Julien Danjou <julien@danjou.info>
We stop retrieving useless key, since we have constant.
We also grab correctly all keycode for a keysym.
Signed-off-by: Julien Danjou <julien@danjou.info>
Pretty much every single source file needs this struct, so it makes sense to
define it in a common header instead of in every single .c file.
Signed-off-by: Uli Schlachter <psychon@znc.in>
Signed-off-by: Julien Danjou <julien@danjou.info>
- We are tracking focus, using FocusIn/FocusOut events handlers, so user
should never be confused about which client has focus
- window_setfocus function generates focus change requests to the X server
- client_focus uses window_setfocus to set input focus
- revert_to when setting input focus set to Parent, compliant with
ICCCM convention ([1])
- DEBUG flag for those who want to debug focus handlers
Most of the changes, are compliant with X11 handbook ([0]), but some
where obtained experimentally.
Kudos to Maarten Maathuis who helped a lot with this.
[0]
http://cgit.freedesktop.org/xorg/doc/xorg-docs/plain/hardcopy/X11/xlib.PS.gz
[1] http://tronche.com/gui/x/icccm/sec-4.html#s-4.2.7
Signed-off-by: Maarten Maathuis <madman2003@gmail.com>
Signed-off-by: Mariusz Ceier <mceier@gmail.com>
Signed-off-by: Julien Danjou <julien@danjou.info>
Uli Schlachter
Anurag Priyam
Gregor Best
Arnaud Fontaine
Julien Danjou
Mariusz Ceier
Maarten Maathuis