We already have the GtkFontChooser::font property
which serves the same purpose, so get rid of the
duplication.

It is going away.

This is a missing part of the CSS font level 3 module.

The docs were out of sync with the flags we actually have.

This is unused and leaked.

The signatures of row-inserted and row-deleted are not
identical, so we need to be a bit more careful here.

Instead of separate handlers, just keep a single
handler for both row-inserted and row-deleted.
Slightly less work that way.

The only time a style-updated indicates we need
to reload fonts is when it is synthesized by GtkSettings
in response to a fontconfig timestamp change, but
we are listening to those already, anyway.

Otherwise, the NULL parent window causes a crash in the
GDK_WINDOW_IS_DESTROYED call.

This fixes a crash happening later on when something tries to access the
event's display.

Some emoji fonts (such as Emoji One), render Emoji sequences
such as some of the family variations using multiple individual
glyphs. This rendering is too wide and breaks our grid layout.
Therefore, we will just skip any sequence whose rendering is
more than twice as wide as a simple smiley.

Stash the DESKTOP_AUTOSTART_ID env var in a constructor,
before any threads have been created.

https://bugzilla.gnome.org/show_bug.cgi?id=790963

No need to keep complaining about this.

https://bugzilla.gnome.org/show_bug.cgi?id=791650

This is just a bandaid solution to make scale 3 work.
If people seriously want to go for scales larger than
that, we need a better solution.

Might as well allow this, for 'large scale testing'.

No deprecated api used in here, it seems.

That is just misleading - the entire placeholder text is going
to be replaced by the drop.

Instead, pass the source window to gdk_drag_begin().

Also make Wayland use this window instead of the one under the pointer
(though those 2 Windows are most likely the same anyway).

The function was referencing itself in ways that bamboozled gcc:
  static void
  foo (void)
  {
    g_signal_handlers_disconnect_by_func (NULL, foo, NULL);
  }

Well done, function, you get your own commit!

This is in preparation for splitting these. Unfortunately, source_window
== ipc_window, so this isn't very debuggable yet.

Ignore the passed in IPC window.

The code was asserting something that was not always holding
true. We can hit row == NULL here on page-up too. Handle that
case by moving to the first row.

https://bugzilla.gnome.org/show_bug.cgi?id=791549

The signal has been gone since e3e9bf4b

Those are source-side events that are handled by signals of the
Dragontext these days.

Quietly do nothing when there is already an ongoing operation.
This matches the behavior of the ewmh code, and is much nicer
than a crash.

https://bugzilla.gnome.org/show_bug.cgi?id=789054

Juan Pablo Ugarte authored Dec 14, 2017 and Juan Pablo Ugarte committed Dec 15, 2017

https://bugzilla.gnome.org/show_bug.cgi?id=791281

Just another bullet point.

It has been replaced by gdk_display_get_keymap.

Replace all uses of the old function by the new one.
No functional change.

This is just a better name for gdk_keymap_get_for_display.

It is better to use the explicit keymap getter - the
default display may not even use the current gdk backend.

We haven't been generating events of this type
since we dropped offscreen windows.

We never generate events of this type, so no need
to carry it around.

The numeric values of event types are irrelevant,
so we don't need to declare them explicitly.