Any symbols (including class properties) that should be visible
outside the module it's defined in need to be defined as global.
For now gjs still allows the access for 'const', but get rid of
the warnings spill now by changing it.
main: Define global properties with 'var' instead of 'let'
Symbols that are defined with 'let' are no longer visible outside
the module that defines them. To unbreak the code base, define all
non-private properties as global.
The first time that the session is started, it can happen that the
AT SPI hasn't been correctly initialized, and this results in a crash
when attempting to register the caret or focus listeners.
In order to avoid this, these changes check the result of initializing
the AT SPI, to allow further attempts when it has failed.
Carlos Garnacho [Wed, 5 Jul 2017 22:05:22 +0000 (00:05 +0200)]
padOsd: Implement edition of actions for rings/strips
Customization of keycombo actions for strips/rings was lost in the
porting to new incarnation of Wacom support.
The UI here is slightly different, instead of requiring the user to
rotate/swipe in each direction to map each keycombo, the UI will
navigate the user through edition of both options, first one, then
the other.
Commit 39a840e2c3d added an additional parameter to shell_app_launch().
When adjusting callers, the parameter was also added accidentally to
calls of the confusingly similar g_app_info_launch() ...
Remove those to fix some warnings.
Florian Müllner [Wed, 17 May 2017 22:03:28 +0000 (00:03 +0200)]
extensionSystem: Handle all initExtension() extension errors
The method currently catches errors that occur when calling the
extension's init() method, but throws itself an error if the
expected extension.js file is missing. The former is pointless
if we expect all callers to handle errors themselves anyway, and
we should avoid the latter if we don't - opt for the second option
and handle a missing extension.js file gracefully.
Florian Müllner [Wed, 17 May 2017 21:57:18 +0000 (23:57 +0200)]
extensionSystem: Catch import errors
While we catch errors that occur when calling init(), enable() or
disable(), the import itself can throw an exception, for instance
if the extension imports an unavailable typelib or tries to draw
in a conflicting library.
Both reloadExtensions() and enableExtensions() are already expected
to catch extension errors. If they don't, this is the bug that
should be fixed instead of catching unhandled exceptions in the
caller.
Those will go away when we port authentication prompts to the new
MessageDialogContent widget, so pick the style classes from there
and adjust individual properties with more specific rules to re-
produce the existing style.
Carlos Garnacho [Thu, 19 Jan 2017 18:49:52 +0000 (19:49 +0100)]
ui: Refactor modal dialog into separate Dialog class
This is the basic dialog actor implementation, which will allow us to
use the same implementation on the session-global modal dialogs. The
ModalDialog class now uses it underneath, and so do all users of it.
Carlos Garnacho [Fri, 14 Jul 2017 20:23:53 +0000 (22:23 +0200)]
layout: Allow adding already parented actors for chrome tracking
let trackChrome accept actors that are not children of chrome actors.
this will be useful for the MetaCloseDialog in gnome-shell, which
is already included in the MetaWindowGroup, but needs to be tracked
as chrome for the dialog to receive pointer events on X11.
Carlos Garnacho [Fri, 14 Jul 2017 23:24:40 +0000 (01:24 +0200)]
slider: Cache slider width for dragging handlers
The actor allocation will be invalid on CLUTTER_TOUCH_BEGIN, because
it comes together with a CLUTTER_ENTER event that will recalculate
styles, and queue a relayout in result.
The net result is that on CLUTTER_TOUCH_BEGIN, the relayout has been
already queued, so the slider width comes up as 0, and the value ends
up as 1. Later touch events already happen on a validated actor, so
it is corrected. Still, not fun when modifying the volume slider on a
touchscreen.
gnome-shell: Make search result description semitransparent
There is a clear difference between the file name and its associated
description which the mockups clearly highlight. Consequently,
make the description label semitransparent, in order to emphasize
the file name, not the description itself.
Having descriptions with multiple lines will clutter
the view and make it more confusing for the user. Apart
from that, it also makes the search result a lot bigger,
potentially losing general vertical alignment.
Rares Visalom [Tue, 27 Jun 2017 22:01:13 +0000 (01:01 +0300)]
search: Highlight search terms in description
The classes extending the SearchResult can now connect
to the 'terms-changed' signal emitted by the SearchResult
class. This signal enables each object to update its
internal description in order to apply the bold style
onto strings that match the search terms.
Rares Visalom [Tue, 27 Jun 2017 21:58:07 +0000 (00:58 +0300)]
search: Propagate searchResults down the hierarchy
In order to prepare for applying the bold style to the part
of the description that matches the search terms, we need
to listen for the signal that announces the fact that the
search terms have changed. Given the fact that the
SearchResults class is aware of the changes regarding the
search terms, the classes that extent SearchResult need
to have a reference to it in order to listen for the
to-be-implemented 'terms-changed' signal.
Rares Visalom [Fri, 23 Jun 2017 14:24:34 +0000 (17:24 +0300)]
search: Increase maximum number of visible list search results
Since results are now much more concise and take up less
space, we can use the 'saved' space to provide the user
with more search results for each provider.
gnome-shell: Use standard font size for the search-result-title class
The current size of the text is way bigger than what
we want now. The default front size is what we are
looking after, therefore no explicit value is needed
to specify the font size.
search: Replace the gradient separator with a simple line
The current separator uses a gradient effect as a separator
between search results. As the mockups suggest, the gradient
separator is no longer needed, in favor of a more simple one,
which is a thin semitransparent line.
In order to match the current mockups, the providerIcon
class needed to include both the name of the provider
and the label that informs the user about how many more
search results are available for that specific provider.
The latter replaces the plus sign icon that has been
used so far.
Rares Visalom [Tue, 20 Jun 2017 19:13:48 +0000 (22:13 +0300)]
search: Make the results take more horizontal space
In order to make gnome-shell search functionality fit on
smaller screens, like those of devices, search results
need to take advantage of more horizontal space so that
any extra space can be used efficiently.
In order to do so, change the layout of the ListSearchResult
class from a vertical one, to a horizontal one and also
decrease the padding of the list-search-result-content css
class.
Carlos Garnacho [Wed, 12 Jul 2017 20:38:31 +0000 (22:38 +0200)]
padOsd: strengthen ring/strip label creation
If the padOsd is given a nonexistent ring/strip, things would fail
badly later when trying to paint a 0x0 StLabel. Just avoid creating
more ring/strip labels than those known by libwacom.
This is unlikely to happen, but seems better to protect against it.
ui: Reset workspace thumbnail porthole on monitor changes
The porthole will not be destroyed when the scale factor changed.
That makes workspace thumbnail porthole still wrong size in the first
seeing after the scale factor changed.
Jonas Ådahl [Wed, 7 Jun 2017 06:53:25 +0000 (14:53 +0800)]
shell-util: Fix capturing scaled views
libmutter-clutter fixed stage capturing when views are scaled, and for
that to work properly when compositing the screenshot, we need to deal
with the cairo device scale here too.
Florian Müllner [Sat, 17 Jun 2017 16:14:20 +0000 (18:14 +0200)]
dateMenu: Set _delegate on non-menu-item menu child
PopupMenu needs special-treatment of certain types of menu items,
which it determines via children's _delegate property. However as
the calendar drop-down is very unmenu-ish, we use regular actors
rather than PopupMenuItems and the missing _delegate property
triggers a warning. Just add it as the bare minimum to make
PopupMenu happy.
Florian Müllner [Tue, 13 Jun 2017 03:51:31 +0000 (05:51 +0200)]
windowManager: Fix a warning
The destroy effect currently relies on a monkey-patched property
added from the map effect. However on X11 it is possible that we
did no map animation for a window that is destroyed when the shell
was restarted or had taken over from another WM. Just use the real
MetaWindow property to avoid a warning in that case.
layout: Skip strut computation in the no-monitor case
It's possible for updateRegions() to be called before monitors have
been properly initialized. Instead of throwing an error in that case,
just skip the strut computation (that doesn't make sense anyway without
a monitor).
Florian Müllner [Tue, 13 Jun 2017 03:39:06 +0000 (05:39 +0200)]
layout: Range-check index before array lookup
findMonitorForActor() may be called before the layoutManager gets
to initialize monitors, so make sure the monitor index is in range
to avoid a warning.
Florian Müllner [Tue, 13 Jun 2017 03:03:17 +0000 (05:03 +0200)]
ibusManager: Fix a warning
Don't try to access a non-existent engine - it probably makes sense to
use Map() instead of a plain object to track engines in the future, but
for now just add an additional check to shut up a warning.
Florian Müllner [Tue, 13 Jun 2017 02:56:34 +0000 (04:56 +0200)]
history: Fix bogus return value
We only use lastItem() to reset the history index to the end, so
nobody noticed the utter nonsense in the return value until gjs
started to warn about it. As we don't actually use the value
anywhere, we could just remove it, but the function name implies
that an item is returned, so fix it to behave as advertised.
Florian Müllner [Tue, 13 Jun 2017 02:46:37 +0000 (04:46 +0200)]
panel: Fix a JS warning
We need to track the open-status of indicator menus, but don't want
to hook up signals more than once, so we check for the handler ID
we store on the object. As the property is only defined once we did
set up the signal connection, this check now logs a warning. We
can avoid it by checking for the existence of the property rather
than a particular value.
Florian Müllner [Tue, 13 Jun 2017 02:17:39 +0000 (04:17 +0200)]
remoteMenu: Add actionGroup accessor
To avoid recreating the app menu unnecessarily, the panel checks
whether the menu's current actionGroup already matches the target
one. However as the menu's actionGroup property is currently private,
the test always fails, whoops.
Florian Müllner [Tue, 13 Jun 2017 01:41:42 +0000 (03:41 +0200)]
dash: Fix some JS warnings
We currently use "array[index]" to test whether an array has an
element at index before using it. However nowadays gjs warns about
accessing non-existent array elements, so the test itself already
produces a warning. Avoid this by checking the array length before
using an index to access an element.
Florian Müllner [Tue, 13 Jun 2017 02:24:12 +0000 (04:24 +0200)]
Initialize properties in _init()
While we've always considered it good style to initialize JS properties,
some code that relies on uninitialized properties having an implicit
value of 'undefined' has slipped in over time. The updated SpiderMonkey
version used by gjs now warns when accessing those properties, so we
should make sure that they are properly initialized to avoid log spam,
even though all warnings addressed here occur in conditionals that
produce the correct result with 'undefined'.
Florian Müllner [Fri, 9 Jun 2017 15:34:43 +0000 (17:34 +0200)]
loginDialog: Only highlight a single item at any time
The user list uses the same indication for hover and focus, so it
is possible for two items to be highlighted at the same time. Using
different styling would improve the situation, but only to some
extent - the user would still need to figure out which highlight
corresponds to which activation method. So instead, copy the
approach we use in popup menus and use a single property for
highlights that is updated by both focus- and hover changes.
Florian Müllner [Fri, 9 Jun 2017 14:06:39 +0000 (16:06 +0200)]
volume: Show OSD popup when changing volume via scroll
It is possible to use the scroll wheel to adjust the volume without
opening the system menu, but there is no feedback other than the
icon itself in that case. To provide a less coarse indication for
the volume level, display the OSD window when adjusting the volume
while the slider isn't visible.
Florian Müllner [Fri, 16 Jun 2017 15:08:07 +0000 (17:08 +0200)]
system: Emulate click action button release
Since commit 2c070d38, we add a ClickAction to the visible AltSwitcher
button to track long-presses. As a result, we now have two components
that will grab and ungrab the pointer for the button, so to make sure
we don't end up with a stuck grab, we need to release the second's
component grab when the first activates.
Currently we only drop the StButton grab on long-press, we also need
to cancel any initiated long-press on click.
Ray Strode [Wed, 7 Jun 2017 19:00:11 +0000 (15:00 -0400)]
gdmUtil: Fix auth prompt drawing over user list
If the user fails to enter their password then hits escape, we
jump back to the user list, then ask again for a password in a
garbled screen. this commit fixes that by skipping a retry if
the operation is cancelled.
Florian Müllner [Thu, 29 Jun 2017 22:06:53 +0000 (00:06 +0200)]
st: Don't try to create shadow for empty texture
Even though the API documentation doesn't say so, the underlying
Cogl texture of a ClutterTexture may be unset, so check for that
case to avoid a runtime warning.
Florian Müllner [Tue, 20 Jun 2017 08:16:05 +0000 (10:16 +0200)]
panel: Consider visibility changes for solid style
We currently assume that window state changes are accompanied by an
allocation change (triggered for example by the minimize animation).
However this misses the case where a window actor is simply hidden
without any transition, as is the case with the 'show-desktop' action
for instance, so start tracking plain visibility changes as well.
Florian Müllner [Fri, 23 Jun 2017 15:37:13 +0000 (17:37 +0200)]
util: Force text direction of time strings
While the string returned by formatTime() should follow the locale's
text direction as a whole, the actual time part is always expected
to put hours on the left and minutes to the right. It is possible to
enforce that by inserting a left-to-right mark, but so far this is
only done by the Hebrew translation. So in order to not require all
other RTL translations to be fixed individually, just insert the
mark into the returned string ourselves like gnome-desktop's WallClock
code does[0].
workspace: Reset source IDs to zero after removing them
There were some source IDs that were not being reset to zero when
removing the associated sources, resulting on some critical errors
being dumped when _realRecalculateWindowPositions() got called
after that point, via _delayedWindowRepositioning().
Ray Strode [Wed, 7 Jun 2017 19:00:11 +0000 (15:00 -0400)]
gdmUtil: Fix auth prompt drawing over user list
If the user fails to enter their password then hits escape, we
jump back to the user list, then ask again for a password in a
garbled screen. this commit fixes that by guarding against the retry
if the fail counter is reset.
calendar-server: Specify the timezone for the view query
Otherwise UTC is used which results in a different interpretation of
floating time events. This then could lead to a mismatch with events
generated e_cal_client_generate_instances_sync() which always uses the
default timezone. Such a mismatch would then cause constant invalidation
and reloading.
appDisplay: Add a null-check for the scrollview's fade ClutterEffect
This effect will only be created when the StScrollView actor has either
a non-zero vertical or horizontal fade offset defined, so we need to
add a null-check in these two cases before assuming it's there.
Cosimo Cecchi [Tue, 27 Aug 2013 20:47:19 +0000 (13:47 -0700)]
Remove an explicit destroy to the ClutterText actor
This is the same as the previous commit, but for StEntry.
We don't have any need to explicitly destroy this actor in our dispose
implementation, and doing so breaks the assumption that we can access
the clutter_text from within destroy.
projects / gnome-shell.git / log