This is one of the properties that need to be explicitly set.

Initialize the `nearest` variable.

When filechooer is re-used (instead of being created and destroyed
every time it's needed), it might happen that the directory the
application wishes to initially show is the same directory that
was shown last time the filechooser was used. In that case, make sure
filechooser knows that it does not need to do its default "select $pwd
or $home" routine.

https://bugzilla.gnome.org/show_bug.cgi?id=766122

This will let us expand the information e.g. with code examples,
and can serve as starting material for a future 3->4 migration
guide.

Resize grips are no more.

Make sure to keep parity with the number of times we froze the window
when we unmap it.
Otherwise it will permanently stay frozen after being remapped.

https://bugzilla.gnome.org/show_bug.cgi?id=766643
https://bugzilla.mozilla.org/show_bug.cgi?id=1225044

adds a paragraph to gtk_text_view_get_pixels_below_lines(),
describing how it can be used together with
gtk_text_view_get_pixels_above_lines() to get the total
amount of line spacing between two paragraphs.

https://bugzilla.gnome.org/show_bug.cgi?id=747206

Mark GtkWidget:style property as deprecated in the docs.

Among other things, explain the relationship of background-color
and background-image.

Remove explicit reference to color from the
gtk_info_bar_set_message_type() docs.

https://bugzilla.gnome.org/show_bug.cgi?id=759037

If the backend does not override GdkWindowClass.create_gl_context()
then we should provide a fallback that gives an error, instead of
crashing.

https://bugzilla.gnome.org/show_bug.cgi?id=766782

Calling gdk_gl_context_realize() should always result in a valid result,
so we need to provide a default implementation, to avoid a call to a
NULL function pointer.

If the popover's relative-to widget is unparented/reparented, we end
up unparenting/reparenting the popover as well. In that case, at the
moment of reparenting, the widget might have been visible (and is
thus mapped again), but priv->window hasn't been set yet.

We must first set priv->window, and then call gtk_window_add_popover(),
that way gtk_popover_map() has its prerequisites straight.

https://bugzilla.gnome.org/show_bug.cgi?id=766323

gtk_widget_set_parent() will map the widget if the parent is mapped
and the widget is both visible and child-visible. As we currently
only set the child visibility after adding the child, we immediately
map all children that are added to a mapped stack, even when they
are not actually shown. Avoid this by setting the child visibility
before adding the child, so widgets are only mapped when shown.

https://bugzilla.gnome.org/show_bug.cgi?id=766737

See https://bugzilla.gnome.org/show_bug.cgi?id=766642

make the switch height and width 1px smaller.

Use g_drive_is_removable for external drives detection. Current heuristic
fails in some cases (e.g. when removable drive is attached before login),
see Bug 765457.

Bump GLib version accordingly.

https://bugzilla.gnome.org/show_bug.cgi?id=765924

This was causing problems with the gtk# scanner

When starting a rubberband selection from an empty area, we could run
into crashes if the selection moves over the rows and then back out
to unpopulated area. Handle this case without crashing.

https://bugzilla.gnome.org/show_bug.cgi?id=766336

The style classes are reflected in the output here.

We should use the same style classes here, to avoid theme
confusion.

sync the alpha scale styleclass too.

rely on toplevel styleclass for scale with marks.

See https://bugzilla.gnome.org/show_bug.cgi?id=766440

This seems to make some issue with the gtk# scanner.

Use .marks-before/after to indicate the presence of marks.
As Lapo points out, compatibility with the previous names
is not really that important, since everything else changed
around it.

https://bugzilla.gnome.org/show_bug.cgi?id=766440

Warn about the situation when we've found a resource or file path,
but gdk-pixbuf fails to give us a pixbuf. This generally means that
either pixbuf loaders are not found or the shared-mime database
is missing.

This can be made to happen eg by setting XDG_DATA_DIRS and
XDG_DATA_HOME to /. Not a useful value, but not a good reason
to crash either.

There is no need to make it a public API, move it to the private header
instead.

https://bugzilla.gnome.org/show_bug.cgi?id=766566