message-tray: Don't update message count after destruction
When trying to update the message count after a summary icon has
been destroyed, the label to display the count is no longer valid
and trying to set its text results in some Clutter warnings.
Cosimo Cecchi [Thu, 23 Jun 2011 14:08:17 +0000 (10:08 -0400)]
autorun: prefer Safe Removal over eject/unmount if possible
Basically do what NautilusPlacesSidebar does with the drive/volume/mount
eject/unmount/stop priorities.
We follow this pattern:
- always prefer Safely Remove if available (i.e. drive.stop())
- fallback to ejecting the mount/volume/drive if that's not possible
- finally, fallback to unmounting the mount if even eject is not
available
This also means we don't care about the distinction between
Stop/Eject/Unmount at this level. Disk Utility (or Nautilus) are
available for those who want that degree of control, but the common case
here should do the most useful action without presenting the choice.
Cosimo Cecchi [Wed, 22 Jun 2011 20:43:16 +0000 (16:43 -0400)]
mount-operation: add a ShellMountOperation implementation
Ideally, this would be an entirely-JS implementation, but we have a
couple of issues with gjs and gobject-introspection to work around, so
we need a ShellMountOperation class for the time being.
This first commit implements the show-processes dialog, with a system
modal style very similar to the EndSession dialog.
Implementations of ask-question and ask-password will follow shortly.
The sniffer is a simple helper process, activated as a DBus service,
that tries to crawl as many files as possible in the provided target
directory (i.e. the new mount's root), for a maximum amount of time -
which is set here to 1.5 seconds (i.e. it will crawl either all the
files in the directory tree, or as many as it can before the specified
timeout expires).
Crawled files are ordered by their content type, and a generic estimation
of the type of files composing the directory is returned to the caller,
using generic 'x-content/*' mimetypes.
The process will then set an autoquit timeout on itself, which can be
disabled by setting the env variable HOTPLUG_SNIFFER_PERSIST for
debugging purposes. The HOTPLUG_SNIFFER_DEBUG env variable can also be
set to enable debugging output.
autorun: follow per content-type gsettings preferences for autorun
Autorun preferences can be fine-tweaked at the content-type level from
the System Settings 'Removable Media' panel.
Use those settings to figure out the default action for newly-mounted
mounts.
Cosimo Cecchi [Mon, 20 Jun 2011 19:16:40 +0000 (15:16 -0400)]
automount: add an AutomountManager class
The AutomountManager class is the low-level counterpart of the
previously introduced AutorunManager, and takes care of extracting the
list of valid mounts from a GVolume or a GDrive and mounting them,
provided a number of conditions and requirements are met.
AutomountManager also keeps track of the current session availability
(using the ConsoleKit and gnome-screensaver DBus interfaces) and
inhibits mounting if the current session is locked, or another session
is in use instead.
AutorunManager is a class that takes care of displaying and managing
notifications and UI for storage devices.
When a mount appears and a number of conditions are satisified, a
transient notification will be displayed to immediately interact with
the device. AutorunTransientDispatcher is the object that takes care of
showing/hiding the notification sources as devices appear/disappear.
Likewise, current mounts are kept in a list and presented within a
list in a resident notification, handled by AutorunResidentSource.
The DBus interface for the battery status is now provided by the
power plugin of gnome-settings-daemon, so there's real reason to
keep building gnome-power-manager.
On error, we tried to kill and respawn gnome-power-manager, but
as of commit c5676900 the DBus interface provided by g-s-d's power
plugin is used, so the respawning does not have any effect.
The upower version in the moduleset is too old for the power plugin
of gnome-settings-daemon. As at least Fedora and Ubuntu have a
recent enough version in their repositories, use that instead of
bumping the version in the moduleset.
==17386== 1,669 (88 direct, 1,581 indirect) bytes in 1 blocks are definitely lost in loss record 4,090 of 4,151
==17386== at 0x4C24AF4: calloc (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so)
==17386== by 0x691B099: g_malloc0 (in /usr/lib/libglib-2.0.so.0.2800.8)
==17386== by 0x692006A: g_option_context_new (in /usr/lib/libglib-2.0.so.0.2800.8)
==17386== by 0x5124C57: meta_get_option_context (in /usr/lib/libmutter.so.0.0.0)
==17386== by 0x401D4F: main (in /usr/bin/gnome-shell)
The gconf keys used to restore GNOME aren't in a proper GNOME3 environment.
To mimic what GNOME3 gnome-session does would be extremely complicated, so
just launch the system gnome-shell.
NetworkMenu: take out an item from More... when another is destroyed
When one of the networks in the main menu is removed and we have
a More... submenu, we can take the first out from the submenu and
show it in the main menu.
Allow other clients to preempt the channels we are handling
This is needed if we are handling an incoming text channel and then user tries
to open a chat with the same contact using Empathy. In this case, the Shell
should delegate the channel back to Empathy and just continue observing it as
it does for usual outgoing channels.
Depends on telepathy-glib 0.15.3 as
tp_base_client_set_delegated_channels_callback() has been added in this
version.
The cogl path pads the corners out to the maximum corner radius to make the
math and painting logic easier. Unfortunately, when the radius exceeds the
actor's halfsize, the padding ends up interfering with other corners, creating
a big mess of rendering errors.
It'd be extremely complicated to fix this properly in the Cogl code,
so take the Cairo fallback.
St: Implement CSS3 specification for reducing border-radius
Currently, any cases of overlapping corners were just ignored and rendered incorrectly.
Implement the corner overlap algorithm as specified by the W3C to fix this.
NotificationDaemon: remove source if notification sender is removed from DBus
We don't want sources that are no longer associated with a running application
to stick around in the message tray.
Message tray sources were removed when the associated application’s state
changed to Shell.AppState.STOPPED . This caused sources for applications
that were still running, but did not have any open windows to be removed.
Instead, we should use the notification’s sender removal from DBus as an
indicator for when to remove the associated source from the message tray.
If we're typing we want to send composing. If we empty the entry we
want to send active. If we're typing but don't type any more for
COMPOSING_STOP_TIMEOUT seconds, we want to send paused. Simple.
This behaviour was stolen from Empathy where it has won many awards.
https://bugzilla.gnome.org/show_bug.cgi?id=650196
Based on patch from Jonny Lamb <jonnylamb@gnome.org> Signed-off-by: Alban Crequy <alban.crequy@collabora.co.uk>
workspace: Fix layout errors caused by window zoom
Commit 64b2b4a7d4 changed the monitor layout handling, resulting
in some layout errors due to a subtle change in memory handling:
when zooming a window in the overview, the available zoom area is
calculated by subtracting the panel height from the primary monitor
area. This area used to be a copy of the monitor rect, but as now
the rect itself is returned, zooming a window on the primary monitor
repeatedly modifies the monitor rect, leading to layout errors in
various parts of the shell.
Fix by using a copy when calculating the available zoom area.
shell-util: Add a helper method for saving files from HTTP
Unfortunately, gjs cannot handle binary C strings directly from
gobject-introspection. Add a simple workaround method in C to help
us save random files from the web.
Dan Winship [Tue, 5 Jul 2011 21:21:22 +0000 (17:21 -0400)]
altTab: try to avoid showing the switcher when "flipping"
Use a longer fade-in time, but with an inout transition, so that the
dialog starts fading in very slowly and then picks up speed after
150ms or so. That way if the user releases Alt+Tab right away, they'll
never actually see the dialog.
Dan Winship [Mon, 13 Jun 2011 13:54:05 +0000 (09:54 -0400)]
layout: new file handling shell layout
Remove ShellGlobal's monitor-related methods, and have
Main.layoutManager provide that information instead. Move
Main._relayout() to LayoutManager, and have other objects connect to
the layout manager's 'monitors-changed' signal to know when the screen
geometry has changed.
Colin Walters [Thu, 30 Jun 2011 18:19:20 +0000 (14:19 -0400)]
configure: Depend on gnome-bluetooth 3.1.0, fix linking issue
Most importantly, the new variables include an rpath for the gnome-bluetooth
directory which fixes a linking problem; introspetion needs to be able to find
the library while running the introspected binary at build time.
Also, this removes some hardcoding in the configure script.
Florian Müllner [Mon, 27 Jun 2011 02:16:28 +0000 (04:16 +0200)]
shell-util: Adapt to nautilus' switch to GSettings
shell_util_get_file_display_name_if_mount() uses a nautilus setting
to use a name for $HOME which is consistent with the file manager.
But while nautilus was ported to GSettings a while ago, we are still
trying to access the old GConf setting.
Morten Mjelva [Wed, 29 Jun 2011 21:16:32 +0000 (23:16 +0200)]
build: Use libgdata from git master
Previously we got by with using the 0.8 tarball as evolution-data-server
only requires libgdata >= 0.7, but as gnome-online-accounts is now build
by default, we need it from git master.
Adel Gadllah [Sat, 4 Jun 2011 21:35:24 +0000 (23:35 +0200)]
StScrollViewFade: Fix scrollbar size handling
Only skip the areas of the scrollbars when they are invisible
and add take the horizontal scrollbar into account as well
when calculating the faded area.
Dan Winship [Fri, 10 Jun 2011 15:01:19 +0000 (11:01 -0400)]
chrome: drop visibleInOverview
Every place that called chrome.addActor was specifying
visibleInOverview:true, and no existing designs call for chrome that
disappears when you enter the overview, so just drop that as an
option.
Florian Müllner [Fri, 3 Jun 2011 22:10:55 +0000 (00:10 +0200)]
user-status: Hide "Power off" option if shutdown is disabled
Due to lockdown settings or Polkit policy, shutdown may not be
available. If this is the case, the "Power off ..." action should
be hidden from the user status menu.
Florian Müllner [Thu, 16 Jun 2011 12:53:17 +0000 (14:53 +0200)]
workspaces-display: Don't slide out pager on drag-begin
Sliding out the workspaces pager when starting a drag causes a lot
of motion. With the pager only hiding if workspaces are not used,
it is better to require to explicitly hover the workspaces sidebar
for the sliding.
theme: Make the running indicator look as it's supposed to
background-image is intentionally being clipped to an actor's size,
but the running indicator SVG source didn't take this into account.
So the running indicator didn't actually work properly, and was a
lot more subtle than intended.
PopupMenu: enhance PopupSwitchMenuItem to show status labels
Moves and converts NMDeviceTitleMenuItem from network.js into
PopupSwitchMenuItem, so that it can show both a switch and a
greyed-out status label. This will be soon used by the Bluetooth menu.
NetworkMenu: place network items in the right position
_createNetworkItem was always appending access point at the end of
the menu when their position was < 5 (NUM_VISIBLE_NETWORKS), ignoring
the presence or absence of _overflowItem, which thus ended in the
middle of the network list. Fix that.
Florian Müllner [Tue, 21 Jun 2011 22:28:42 +0000 (00:28 +0200)]
workspaces-view: Handle clip rectangle separately
After completing the overview transition, the workspace view is
clipped to avoid overlapping the search entry/view selector titles
while switching workspaces. For clipping, the view's position and
size was used, which works well assuming that the workspace pager
will start hidden or stay zoomed out while the overview is visible.
However, that assumption holds no longer true, as auto-hiding the
pager now depends on the number of workspaces, and thus may change
while in the overview. For instance, when starting with the pager
being visible, the clip area ends up being too small when moving
all windows to the first workspace (and thus triggering auto-hide).
As a fix, handle the clip rectangle separately from the view's
geometry, and set it always to the area the workspaces would
occupy with the pager hidden.
Dan Winship [Fri, 10 Jun 2011 13:51:55 +0000 (09:51 -0400)]
main: move pointer-sync-on-restack code here
We call global.sync_pointer() on MetaScreen::restack as a hack to try
to fix up the hover state after a pointer grab. Previously we were
doing this in chrome.js, since there was already a ::restack handler
there anyway, but this isn't really related to the chrome at all, so
move it to main.js instead.
projects / gnome-shell.git / log