_NET_WM_PID is unreliable! It can be faked or pid might be from
different namespace. Ignore _NET_WM_PID and use XResQueryClientIds
to get pid.

#9

Replace filename with basename in template file and also regenerate
enum types if Makefile.am changes.

1) These buttons are broken with default GTK theme. As nobody has
reported bug I assume these buttons are unused. GTK is default
theme type since 3.2.0.

2) These buttons are not supported by client side decorated
windows.

3) Few Metacity themes I checked had no icons for these buttons.
They had empty draw_ops...

4) These buttons has been removed also from Mutter.

Functionality itself is not removed! Still available using window
menu and/or keybindings.

This button has never been used by Metacity nor by Compiz. Remove
everyting related to this button.

Based on mutter commit:
mutter@1d75d5aa

Reduced resources setting was used to disable minimize effect and
to show wireframe when moving or resizing windows.

Minimize effect still can be disabled by using enable-animations
setting from org.gnome.desktop.interface schema. And if you are
really trying to reduce resource usage you probably already have
animations disabled.

Wireframe mode has not been updated/fixed to work with CSD windows
and/or to exclude shadow/resize area from SSD windows. Also quickly
testing with and without compositor wireframe sometimes leaves
parts of frame on screen.

Considering that there has not been any bug report about broken
wireframe mode I think it is time to remove it instead of wasting
time trying to fix something that nobody is using.

Based on mutter commit:
mutter@92388892

It is always TRUE.

ideal_width and ideal_height has always same value, same with
ideal_mini_width and ideal_mini_height.

meta_read_icons is always used with constant size that is larger
then zero.

Preferred way is to use GTK+ theme for window decorations. Metacity
themes are still "supported" to allow users use old themes...

Also metacity-theme.dtd has not been updated for Theme Format 3.x,
remove it as it can not be used to validate new format files.

We refuse to move focus while a grab operation is in place. While this
generally makes sense, there's no reason why the window that owns the
grab shouldn't be given the regular input focus as well - we pretty
much assume that the grab window is also the focus window anyway.

In fact there's a strong reason for allowing the focus change here:
If the grab window isn't the focus window, it probably has a modal
transient that is focused instead, and a likely reason for the focus
request is that the transient is being unmanaged and we must move
the focus elsewhere.

mutter#15

It doesn't make much sense to globally unset focus when mapping
a new window.

Based on mutter commit:
mutter@998d921d

FOCUSED_WINDOW layer does not exist.

This reverts commit 12c4f35c.

#8

Timeouts are not used since de65967b.

Changes to remove frame for fullscreen windows have introduced
regression - Firefox will cancel fullscreen if it receives focus
out event before _NET_WM_STATE property change event.

Use idle to ensure or destroy frame after window features are
recalculated and window state set.

This is attempt to fix/silence coverity overlapping assignment
issues - CID 1491451 and CID 1491452.

In practice this does not seem to change anything, in both cases
clang and gcc generates same assembly output with optimization
level 1 or higher.