The formats (or 'region') setting might be unset meaning that it's the
same as the language since that's what LC_TIME et al default to when
unset.
We already handle this for the user setting but for the system setting
we're losing that semantic when getting it from localed since we don't
allow priv->system_region to remain unset. This means that when users
change the system language, the system formats will be explicitly set
to the previous value of the system language instead of remaining
unset and thus follow the new value for the system language.
This isn't that bad on multi user systems where we display system
settings separately from user settings, but on single user systems we
change the system values to match the user values which, due to the
above mentioned semantic difference regarding the region setting,
means that when changing the language, the (unset) region will be
displayed as the same as the language but the system region will be
explicitly set to the previous language.
Fix this by making the system region have the same semantic as the
user's, i.e. allow it to remain unset until explicitly set by the
user.
https://bugzilla.gnome.org/show_bug.cgi?id=755648
Excessively long metadata fields cause the the control-center panel to
blow out to wider than screen width. Probably the right thing is to
limit the allocation to the device detail pane from somewhere up the
stack, but for now, enable ellipsizing and set an max-width to
constrain the width of things which are there just for information
anyway; true values are available in the edit dialog.
Closes#759766.
Signed-off-by: Andrew Cowie <andrew@operationaldynamics.com>
When transitioning from an enabled platform rfkill to disabled, the
Bluetooth adapter's power will lag behind. Make sure to not change the
switch's status when we get into this temporary state.
https://bugzilla.gnome.org/show_bug.cgi?id=760168
When enabling Bluetooth, and with older versions of the rfkill
gnome-settings-daemon D-Bus service, we could get into a state where the
Bluetooth rfkill was off, but the hciX rfkill associated with the
adapter is blocked.
Avoid showing filler text in that case.
https://bugzilla.gnome.org/show_bug.cgi?id=760168
The Tracker plugin is enabled by default, and will replace the
MediaExport plugin is enabled. This means that instead of sharing files
listed in the paths set in the UI, we'd be sharing the user's files as
configured in the Search panel. Not what we want to do if this contains
private files.
The two-finger scroll method will be enabled in all cases
(see bgo#759304). When a touchpad without two-finger scrolling
support but with edge scrolling support is available, show the
switch.
Also bump the required gsettings-desktop-schemas version for
the new key.
https://bugzilla.gnome.org/show_bug.cgi?id=759307
This is a pre-requisite for using a header bar.
GtkDialog:use-header-bar is a construct-only property, and has no
effect when set from the XML. Therefore, the only option is to turn
the widget into a template and set use-header-bar from C.
There should be no behavioral changes due to this patch.
Bump required GLib version for the G_DECLARE_FINAL_TYPE macro.
https://bugzilla.gnome.org/show_bug.cgi?id=757516
When running on something other than a VM or a tablet, which have
hard-coded actions when the power button is pressed, offer to configure
the power button to either suspend (default), hibernate, or do nothing.
https://bugzilla.gnome.org/show_bug.cgi?id=757880
Since the first entry in the Touchpad section causes the sensitive
of the items below to change, those items are in a diferent ListBox.
To make it look like it is all the same ListBox, we arbitrarily add
a separator between the first item and the items below.
https://bugzilla.gnome.org/show_bug.cgi?id=756863
Fixes the situation where you unplug a touchpad device which
supports two-finger-scrolling and plug in another one with
edge-scrolling only, we would have two switches. These entries
are mutually exclusive.
https://bugzilla.gnome.org/show_bug.cgi?id=756863
If you don't have any printers, the panel used to look rather bare.
There was also a "No printers available" label, which was confusing,
and the insensitive list might be hard to understand.
https://bugzilla.gnome.org/show_bug.cgi?id=690351
This fixes:
==5944== 64,392 bytes in 4,223 blocks are definitely lost in loss record 16,020 of 16,045
==5944== at 0x4C28C50: malloc (in /usr/lib64/valgrind/vgpreload_memcheck-amd64-linux.so)
==5944== by 0x1445EFCC: g_malloc (gmem.c:94)
==5944== by 0x1445F2AE: g_malloc_n (gmem.c:330)
==5944== by 0x144981DC: g_variant_get_strv (gvariant.c:1572)
==5944== by 0x48FA45: settings_get_binding (cc-keyboard-item.c:369)
==5944== by 0x48FA9D: binding_changed (cc-keyboard-item.c:384)
==5944== by 0x141C3E2F: g_cclosure_marshal_VOID__STRINGv (gmarshal.c:1794)
==5944== by 0x141BFBE3: _g_closure_invoke_va (gclosure.c:864)
==5944== by 0x141DA3E7: g_signal_emit_valist (gsignal.c:3292)
==5944== by 0x141DB55F: g_signal_emit (gsignal.c:3439)
==5944== by 0x13EDC81D: g_settings_real_change_event (gsettings.c:386)
https://bugzilla.gnome.org/show_bug.cgi?id=756762
