The switch for the current network was only updated to TRUE after a
network change. However, if a new network was not enabled, the switch
should be set to FALSE. This is now properly handled.
Furthermore, the Off/Enabled/Active indicator was also inconsistent
after network changes due to the is_visible check failing sometimes,
even after changing to an already enabled network. Because the actual
current network state is now better tracked, we can just check the state
of the switch instead.
In `cc_sharing_networks_new_row`, the `network_name` is not being set to
the new row, but to `self->current_row`. This is incorrect behavior,
as it leaves the new row without any title, and overwrites the title of
the current row incorrectly to the last network entry in the enabled
networks list. This fix correctly sets the name to the new row instead.
This problem is at least discussed in #2318 (point 3) and causes some
confusion in #2299 as well.
Network names have many allowed characters, among which `&`, `<`, and
`>`. These are problematic for Pango markup. By default, AdwActionRow
has markup enabled for its title. We could escape strings at every
`adw_preferences_row_set_title` call, but this might be easy to forget
in the future. Instead, let's disable markup altogether for these rows.
The ability to tweak the state property was mainly introduced in
order to implement delayed state change via the state-set signal.
Since GTK 4.9.3[1] the active and state properties are no more
interchangeable.
[1] https://gitlab.gnome.org/GNOME/gtk/-/merge_requests/5442
GtkStyleContext will be deprecated in gtk 4.10.
https://docs.gtk.org/gtk4/class.StyleContext.html
This preserves code blocks where additional GtkStyleContext operations
were used, such as gtk_style_context_save/restore.
Currently if a network connection's name is long enough, there will
be no space between the name label widget and the switch button widget
next to it.
Fix that by setting the end margin of label widget as 6 to keep some
space between those two widgets.
We need to block ourselves around the gtk_switch_set_active() call in
the error path, when we can't speak to the sharing dbus service, or we
end up in an endless loop since we're the ::state-set handler.
https://bugzilla.gnome.org/show_bug.cgi?id=739886
