138 translated messages, 9 untranslated messages.

Transifex (https://explore.transifex.com/xfce/).

147 translated messages.

Transifex (https://explore.transifex.com/xfce/).

147 translated messages.

Transifex (https://explore.transifex.com/xfce/).

147 translated messages.

Transifex (https://explore.transifex.com/xfce/).

147 translated messages.

Transifex (https://explore.transifex.com/xfce/).

144 translated messages, 3 untranslated messages.

Transifex (https://explore.transifex.com/xfce/).

120 translated messages, 27 untranslated messages.

Transifex (https://explore.transifex.com/xfce/).

147 translated messages.

Transifex (https://explore.transifex.com/xfce/).

147 translated messages.

Transifex (https://explore.transifex.com/xfce/).

147 translated messages.

Transifex (https://explore.transifex.com/xfce/).

147 translated messages.

Transifex (https://explore.transifex.com/xfce/).

144 translated messages, 3 untranslated messages.

Transifex (https://explore.transifex.com/xfce/).

147 translated messages.

Transifex (https://explore.transifex.com/xfce/).

134 translated messages, 13 untranslated messages.

Transifex (https://explore.transifex.com/xfce/).

Reinstated 75243250, since xfce4-screensaver really can't do anything in
an environment where support has been disabled.

Related: orage#32, xfce/xfce4-dev-tools!101

Following the addition of the Wayland implementation of GSListener, we
now need these signals to be emitted only once, and at the time
indicated by their name. This also implies some corrections to the
values returned by the functions emitting or connected to these signals.

The ext-session-lock protocol imposes nothing on the compositor
concerning the focus of lock surfaces. In particular, there is no
guarantee that the user will be able to click on a surface to give it
focus, even if the protocol suggests this.

More generally, it's not certain that we receive pointer events on lock
surfaces, and even when we do, we don't necessarily have a reliable way
of knowing which surface is under the pointer at any given time.

At first glance, then, it seems that the only thing we can do to ensure
that the unlock dialog has indeed the focus, is to display it on the
lock surface to which the compositor has given the focus when creating
all the surfaces. This is what the Wayland implementation of
find_window_at_pointer() does in this commit.

However, the active surface may have changed by the time the dialog is
displayed, particularly when user activity takes place just after
surfaces have been created. In this case, we cancel the dialog display,
and another user activity will re-trigger its creation (in practice,
this simply adds a small latency).

Next, we need to add a focus recovery timeout at plug level, in case
focus is lost when exiting text entry, for example by pressing the tab
key (for some reason, pressing this key doesn't automatically return you
to text entry once you've reached the last button in the action area, as
on X11).

gs-grab-x11.c is X11 only and gs-window-x11.c mixes X11 and Wayland.
This also aligns file names with class names.

In preparation for the addition of the GSListener abstract class for X11
and Wayland implementations. This matches file naming, incidentally.

This is the right format, previous commits have already changed some of
it.

In preparation for using WleGtkSocket on Wayland.

In preparation for using WleGtkPlug on Wayland.

The precautions taken in this commit to avoid X11 errors are not always
sufficient. For example, it's possible to get one of these errors on
resuming from suspend when a notification appears at the same time (not
just any notification, apparently; in my case it's from nm-applet).
However, I can't reproduce the problem with openbox, only with xfwm4,
and apparently only with /general/unredirect_overlays enabled.

So user can extend or override the flags provided there by defining the
DISTCHECK_CONFIGURE_FLAGS variable.

First step in distinguishing X11 and Wayland, the rest will come in
next commits.

* Remove unused project files
* Remove dead code/checks/flags
* Fix X11 specific headers

The build actually failed in the !HAVE_SIGACTION case, so it seems safe
not to check for the presence of this function.

It's not xfce4-screensaver's job as long as xfce4-power-manager exists.
The DPMS should only be configurable in one place, which simplifies life
for both the user and the developer.

This search for the best visual doesn't seem very useful, and the GDK
functions that would make this code independent of the windowing
environment have been deprecated, so let's just remove it and use the
system default visual instead.