And a way to derive "sub"-adjustments from the main one.
The main and private StAdjustment properly represents
workspaces, and has all relevant properties (lower, upper,
and value) set by the workspace manager.
The main adjustment is not bound to any particular actor,
which means we cannot call the 'ease' method directly.
Consumers of this API should create adjustments using
Main.createWorkspacesAdjustment(), and this new adjustment
is bound to the actor that the consumer passed. Consumers
must not change any property of the derived adjustment other
than the 'value' property.
The 'value' property is synchronized between all adjustments
created, which guarantees that all elements that represent
workspaces can have a shared and up-to-date understanding of
where in the workspace layout we are.
Dallas Strouse [Fri, 11 Aug 2023 15:20:53 +0000 (15:20 +0000)]
util: Export trySpawn()
It was private before, and so any users of it (including the
Background Apps menu) would fail to utilize it, making apps fail
to get told to shut down from the Background Apps menu.
Florian Müllner [Thu, 10 Aug 2023 15:50:08 +0000 (17:50 +0200)]
tests/unit: Access shell sources from resource
Unit tests currently load shell sources directly from the
filesystem.
This is currently blocking generated sources - namely config.js - to
ESM, because a relative import from the source dir will fail to
locate the file in the build dir.
Address this by using the same GResource as gnome-shell instead of
direct filesystem access, as the resource will always include all
sources files at the expected location.
Out of the members of the notification daemon(s), these are the most
necessary and flexible to override. The daemons are used as singletons
and notifications are rather temporary, while the sources are more
flexible and make the most sense to override (and revert).
Florian Müllner [Wed, 9 Aug 2023 14:24:07 +0000 (16:24 +0200)]
screenshot: Export showScreen*UI() functions
showScreenshotUI() is needed by the system action, and
showScreenRecordingUI() is conceptually so similar that
it would be odd to export one and not the other.
Florian Müllner [Mon, 7 Aug 2023 01:26:04 +0000 (03:26 +0200)]
ci: Simplify run-eslint script
Now that all code conforms with the new style, we can remove all
the tricky bits that compare errors from regular- and legacy
config or limit checks to changed lines from a git diff.
All that is left over the eslint CLI tool is the ability to output
results both as junit for gitlab and plain text for logs without
duplicating the linting, but that's well worth preserving.
Florian Müllner [Mon, 7 Aug 2023 01:24:35 +0000 (03:24 +0200)]
ci: Remove eslint_mr job
There is no longer a difference between regular- and legacy config.
That means that all code now follows the regular configuration, and
we no longer need a separate check that enforces this for new code.
Florian Müllner [Mon, 7 Aug 2023 01:15:51 +0000 (03:15 +0200)]
cleanup: Fixup wrong JSDoc comments
There are a handful of JSDoc comments that are invalid or don't
conform to the configured format. Fix them to remove the last
remaining bit of legacy configuration.
Florian Müllner [Sun, 6 Aug 2023 22:40:20 +0000 (00:40 +0200)]
cleanup: Remove spaces in object literals
We only adopted this style relatively recently, so there's a bit
more to adjust. Still, it's manageable and another step towards
getting rid of the legacy style.
Carlos Garnacho [Tue, 8 Aug 2023 22:26:52 +0000 (00:26 +0200)]
st: Mass update to ClutterActorClass event vmethod changes
These are now forwarded as ClutterEvent in order to remain accessible
on the Javascript side. This also means all ClutterActor implementations
in ST need to update to this change.
Carlos Garnacho [Tue, 8 Aug 2023 16:22:11 +0000 (18:22 +0200)]
js: Avoid Clutter.get_current_event() in Clutter.Actor event vfuncs
We sometimes obtained the current event through the general machinery
just so we could pass it along as a ClutterEvent instead of a specialized
subtype.
We now get that out of the box, so may avoid getting the current event
which is just a cast of the same current event we already have.
Carlos Garnacho [Tue, 8 Aug 2023 16:14:04 +0000 (18:14 +0200)]
js: Mass move to Clutter.Event getter methods in Clutter.Actor vfuncs
These traditionally got the various ClutterEvent subtype structs as their
argument, so it was not allowed to use ClutterEvent generic getter methods
in these vfuncs. These methods used direct access to struct fields instead.
This got spoiled with the move to make ClutterEvent opaque types, since
these are no longer public structs so GNOME Shell most silently failed to
fetch the expected values from event fields. But since they are not
ClutterEvents either, the getters could not be used on them.
Mutter is changing so that these vmethods all contain an alias to the
one and only Clutter.Event type, thus lifting those barriers, and making
it possible to use the ClutterEvent methods in these vfuncs.
Carlos Garnacho [Wed, 9 Aug 2023 09:47:28 +0000 (11:47 +0200)]
inputMethod: Check that there is an existing cursor location
Commit 17d9ec5788fb made the input method call update() more eagerly,
but also at times that it does not have a cursor position yet. Make
it bail out correctly in that situation.
Fixes: 17d9ec5788fb ("inputMethod: Keep Capabilite.FOCUS before context.focus_in/focus_out")
Part-of: <https://gitlab.gnome.org/GNOME/gnome-shell/-/merge_requests/2876>
Carlos Garnacho [Tue, 8 Aug 2023 14:10:39 +0000 (16:10 +0200)]
ui/slider: Handle emulated scroll events through the Clutter.EventFlags
Emulated scroll events no longer get event.is_pointer_emulated() set to
True, instead this information is propagated through the Clutter.EventFlags
of the event.
Carlos Garnacho [Tue, 8 Aug 2023 14:08:36 +0000 (16:08 +0200)]
st: Handle emulated scroll events through the ClutterEventFlag
Emulated scroll events no longer get clutter_event_is_pointer_emulated()
set to TRUE, instead this information is propagated through the
ClutterEventFlags of the event.
Takao Fujiwara [Wed, 1 Mar 2023 14:42:18 +0000 (23:42 +0900)]
inputMethod: Keep Capabilite.FOCUS before context.focus_in/focus_out
If context.focus_out() is called *after* context.set_capabilities(0),
The FocusOut D-Bus method is ignored because of no FOCUS capability.
If context.focus_out() is called *before* context.set_capabilities(0),
The 0 capability is set to the next focused context and the
FocusIn D-Bus method is ignored because of no FOCUS capability.
So context.set_capabilities(0) should not be called.
Florian Müllner [Mon, 7 Aug 2023 15:13:48 +0000 (17:13 +0200)]
js: Port dbusUtils to ESM
The module is shared between the various D-Bus services and the
main gnome-shell process, so it was originally left out to allow
porting different bits at their own speed.
Now that everything has been ported to ESM, there is no reason
to not move that particular module as well.
Florian Müllner [Sat, 5 Aug 2023 19:12:55 +0000 (21:12 +0200)]
status/backlight: Sync checked on level changes
Since switching to property bindings/notifications, value changes
only happen on actual changes, so we cannot rely on them as the sole
source of updating the checked state.
Daniel van Vugt [Thu, 23 Mar 2023 09:51:46 +0000 (17:51 +0800)]
main: Correct the pointer value of strtab on riscv/mips architectures
Glibc defines `d_ptr`/`d_val` to be a relocated address on most
architectures, except for riscv and mips where it is just an offset from
the binary load address. So we need to convert `strtab` from an offset
into a pointer for riscv and mips.
Internally within glibc there is the `D_PTR` macro for doing this, which
relies on private data hidden in an ABI-unstable part of `link_map`. So
we can't use the same conditional logic as glibc does internally.
Our solution is to detect when `strtab` is unreasonably low (below the
address of our binary) and so it must be an unrelocated offset. In that
case we just do the relocation manually.
Florian Müllner [Wed, 2 Aug 2023 16:09:46 +0000 (18:09 +0200)]
status/backlight: Notify slider value changes
The custom setter used by the slider item isn't emitting change
notifications, so the property binding that uses it as source
never propagates the new value.
We have always defaults to an empty ornament, so that menu items
are always aligned, even when radio items are used.
However radio items are fairly rare, so most of the time we end
up with an extra margin with no purpose. The design team now
prefers radio items to only align with each other, so that regular
items get the expected margin.
We currently consider idle nodes as in use to avoid too fast state
changes and thus flickering of the indicator icon. However, the current
idle timeout is rather long and arguably confusing for users.
Thus switch to only consider running nodes as in use, but delay
disabling of the indicator by 500ms.
Florian Müllner [Tue, 13 Jun 2023 19:36:20 +0000 (21:36 +0200)]
status/backlight: Add new quick toggle
Settings no longer exposes a slider for the keyboard brightness,
leaving keyboard shortcuts as the only way of adjusting it.
This is good enough in most cases, because devices with keyboard
backlight usually include corresponding keys on their keyboard.
However for devices without those keys, it would be good for the
settings to be exposed somewhere again. Quick settings seems like
a more appropriate place than "proper" Settings, since it gives
quick access that doesn't require a major focus change.
extensions: Make ExtensionPreferences more flexible
Extensions must now export a class with a fillPreferencesWindow()
method in their prefs. That is less convenient for extensions
with simple preferences than the old buildPrefsWidget() hook, as
they must wrap their widget in page/group widgets.
Address this by adding a default fillPreferencesWindow() implementation
that calls a getPreferencesWidget() method and wraps it as necessary.
This is flexible enough to support different cases fairly conveniently,
from simple single-widget prefs over tweaking the window to complex
multi-page prefs:
```js
class SimplePreferences extends ExtensionPreferences {
getPreferencesWidget() {
return new SimplePrefsWidget();
}
}
class TinkerPreferences extends ExtensionPreferences {
getPreferencesWidget() {
return new SimplePrefsWidget();
}
If an extension provides the gettext domain in its metadata,
then we can simply set up translations from the constructor,
so do that for a bit more convenience.
All extension lookups now happen through the dedicated
Extension/ExtensionPreferences classes, so the shared
code no longer has to access the extensionManager.
Use the new defineTranslationFunctions() method from the previous
commit to create gettext functions for the module, instead of
re-exporting from the shared module.
It is now up to extension developers to use the more effective
The method can be used to define a set of gettext functions that
call the corresponding method of an extension.
Those functions are very similar to the gettext functions we are
exporting, except that:
- looking up the extension is delegated to the
Extension/Preferences class
- it is possible to avoid examining the stack
when called with `import.meta.url`
projects / gnome-shell.git / log