> Removing 'Launcher' as parameter from state methods called by StateManager
> Converting state properties to methods for easier abstraction
> Moving state handling drom state definition to activity class
Change-Id: I997627df606a7e0bb3bf32688d045a942a47fc94
> Fixing scale up calculator for swipe-down
> Offsetting pivot so that the preview is aligned to bottom-right
> Allowing insets to be available in multi-window mode as well
> Offsetting taskViewSimulator appropriately in multi-window mode
Change-Id: I7da4c145efca72ef219a5ffcaf23d726812df270
Was able to repo the bug reliably when calling cancel() and
end() immediately after starting the StaggeredWorkspaceAnim.
Bug: 153584267
Change-Id: I954d38ad69e72a83535f4dd1ec21dba0f61a3cfb
The rect we passed in was in screen space to match the window, but needs to be
be local to launcher.
Also updated code/comments to make it more readable.
Note that the bounds of the targets being off is a separate issue b/153581126
Bug: 135195169
Change-Id: I4a8c0aa2f504bdf69342a927725cd440fb324a36
We don't clear state on EdgeBackGestureHandler
when we go to Recents or Home.
A more robust solution is in the works with
WM team and will probably be built on top of this.
Needs ag/11273364 to work.
Bug: 154580671
Test: Launch app directly in landscape,
should be able to swipe home on first try.
Change-Id: Ia4c685e16645d0237f278a611f6c61f252f3b79c
This still shows up a bit strange with
notched phones because we are using
RecentsView padding to give ClearAll the
right amount of spacing. However, we're only
adding padding to the left and right and not
the top/bottom.
That's captured here: b/154436527
Fixes: 150389833
Test: Tested landscape/portrait/seascape
launcher, w/ RTL as well
Change-Id: I49defd58048e609a1f7a821abb97374bcbb94b9f
Prevent extra IPC calls when checking for
multiWindowMode by using existing DeviceProfile.
This might be made irrelevant by b/142282126
Fixes: 149571513
Change-Id: Ia41cbfd56c67dd78f4d9c10bf1469d0d2a9e581d
Disable overview rotation in 2-Button
until we can figure out the correct
transition/animation.
TODO: b/154665738
Fixes: 154181816
Test: Tested no button and 2 button,
overview actions worked as before.
Change-Id: Ib67906b11e24def870a39f57b5156e29db258fd7
Fixes: 154955192, 150958400
Test: Switched from gesture nav to 3 button,
was no longer able to swipe up
Change-Id: Icc7b27eb7bd1eb5ddf25d7e7c3464a77a3a88f24
Test:
- start sharesheet from overview action, swipe up
- start sharesheet after taking screenshot on home, swipe up
- start sharesheet from an app, e.g. Chrome
The first two use OWFIC, last one uses OtherActivityInputConsumer
as normal.
Bug: 153673244
Change-Id: I7d7887e1f6818346ada75776ec63b43aaf3708f1
Home gesture also exits Back and Home tutorials on the
completion screens (as does back gesture).
Test: Manually verified Back and Home tutorial flows.
Bug: 148542211
Change-Id: I8026745e875e9ccb8cfd60bc438213b2d896254d
Now with hierarachical animation, the layer of
RecentsAnimation / RemoteAnimation are controlled by system.
Now Launcher side don't need to have other way to set layer for ZBoost
case, so remove setLayer & withLayer related logics for cleaning up.
Fix: 151024899
Test: manual, launch google map app -> enter PiP mode,
swipe up to overview screen or launch activity from shortcut,
make sure the PiP window is not covered by other apps.
Change-Id: I8028cd50d66f7d77e5d1ec6514e953c8be9935d6
(cherry picked from commit 99f7da7dd2)
Strictmode leak detector is still a goal, but we might not be able to
achieve it in R. Strictmode has several framework-side bugs that perhaps
hide Launcher-side strictmode violations, while the time to fix
everything is limited, and new leaks get introduced all the time.
For now, implementing a check that is slightly more relaxed than
Strictmode, but still ensures the absence of leaks. I’ll keep
eliminating Strictmode violations as well as keep strengthening the
makeshift checker conditions until we’ll be able to enable Strictmode in
continuous testing.
I’m disabling Strictmode checks for now so that they don’t generate
unnecessary hprof dumps, but leaving the code dealing with strictmode.
Bug: 139137636
Change-Id: Ib10136b0d4e9892f70a19cd052ae5a54cf0a4efb
After reaching overview, we let the user translate RecentsView
if they continue dragging around. But when they let go, we need
to return RecentsView to translation 0 since that's no longer
part of the state machine.
Change-Id: I30b51485339a3b6c3dd52bda113b1a05b6e885fa
