Age | Commit message (Collapse) | Author |
|
Author: LuK1337 <priv.luk@gmail.com>
Date: 2018-10-31 16:31:19 +0100
SystemUI: Show bluetooth battery level when available
* Somewhat inspired by change committed by Gavin Ni <gisngy@gmail.com>
back in cm-13.0 days (see commit 88e7a6c). Since then completely
rewritten using BluetoothDevice API introduced in Oreo, with new
drawables meant to be used with 0-9 battery level range.
Change-Id: I6179bfd41e033591534e8cf3c6adc98ce715a13d
Author: Luca Stefani <luca.stefani.ge1@gmail.com>
Date: 2021-08-23 10:11:54 +0200
fixup! SystemUI: Show bluetooth battery level when available
The current update method is triggered in 2 cases:
the SystemUI bluetooth controller changes device state
the battery level changed intent is sent
Sadly the current SysUI controller doesn't handle battery level
changes, and is instead triggered in a few other cases ( ex audio
state changed ).
And since we use the data from SysUI controller for battery level
from changes in the global bt controller we may have 2 different
set of devices to work on. This could ( and does ) trigger
race conditions.
Instead add support battery level state changes within the SystemUI
controller to work on the same data set.
Change-Id: Ib57096fc2ebe151ac3860277a4db2319f230b514
Change-Id: I34344907711b7b9fc4c1d15d4516bd53bf30ebb2
|
|
Author: ezio84 <brabus84@gmail.com>
Date: Tue Dec 19 14:37:35 2017 +0100
Screenshot: append app name to filename
Kang Samsung's idea.
Change-Id: Ibc6a52b5e2597e6014a2da6a4211febe17ec02c7
Author: Wang Han <wanghan1995315@gmail.com>
Date: Fri May 11 12:47:40 2018 +0200
SystemUI: Don't append app name to file on lockscreen
When device is locked, current implementation will append the current
app name to file when taking screenshot. This causes information leaks.
To prevent this, we check for isKeyGuardLocked() and fall back not to
append the app name when taking screenshot on lockscreen.
Change-Id: I04498faf51986e1a3723a32befd97d29e1f3fe58
Author: DroidFreak32 <rushabshah32@gmail.com>
Date: Thu Oct 4 11:33:07 2018 +0530
Screenshot: Append app name after screenshot date.
Change-Id: I2ee5d65006ff22dfa381aae50e5757b8e7596e91
Author: ezio84 <brabus84@gmail.com>
Date: Sat Jun 16 16:57:19 2018 +0200
GlobalScreenshot: Fix screenshot not saved with some languages
like Virgin Islands English when taking a screenshot of the Settings app
Change-Id: Ic04f66f5813b9597c96835d15c74509c93530a5c
Author: Han Wang <416810799@qq.com>
Date: Mon May 25 11:18:01 2020 +0200
SystemUI: Do not convert application name to ISO-8859-15
* ISO-8859-15 does not handle CJK chars, so all of them are
'unknown' and empty eventually.
Change-Id: I27c689f5d737190d381146e49c1989e998b4f623
Author: LuK1337 <priv.luk@gmail.com>
Date: Mon Aug 24 17:52:39 2020 +0200
SystemUI: Use TaskStackChangeListener API for contextual screenshot package name
* This should hopefully fix mislabeled screenshots.
Change-Id: Ifde106c0e306a6700081e4d724c1cf0c06dd126c
Change-Id: Ic2a42536bf763a0c6f7fb9ac4bbbe84f73723f91
|
|
Bug: 220074017
Change-Id: Idfdd94e902f656ac65a2a75dfdd199f6f85ba472
|
|
Change-Id: I61ca87692541f793479961246a296db67593d73d
|
|
Change-Id: Ie250e3592a53519f58f691675a61adf7fddc335c
|
|
Before, when the device was rotated we would get the new orientation value
from the event but then use the value from getResources() to decide
which layout to show. This discrepancy is a potential cause for an issue
where QS uses the landscape layout when the device was actually portrait
Fixes: 159737544
Test: atest QSPanelControllerBaseTest QSPanelControllerTest
Change-Id: Icd15712cd4821b7833a3373978a92f8b5959ac12
Merged-In: Icd15712cd4821b7833a3373978a92f8b5959ac12
(cherry picked from commit 917a11a54a41f79e01b176fb5d98d346e0f76ff5)
(cherry picked from commit a8e1bbf7c0195875f4f6692a8108e2e48db2a255)
Merged-In:Icd15712cd4821b7833a3373978a92f8b5959ac12
|
|
A background app that changes the wallpaper, was able to change the
device theme.
This change will apply the theme if the app is on the foreground, like
WallpaperPicker would be, but would defer the event until the next power
button cycles, like we do with Live Wallpapers, to avoid the same type
of DoS.
Test: manual
Test: atest ThemeOverlayControllerTest
Fixes: 205140487
Change-Id: I2086b29ef9bf3bb6fb7d9ebba6e7b8db9392e459
Merged-In: I2086b29ef9bf3bb6fb7d9ebba6e7b8db9392e459
(cherry picked from commit efaa814a99bc86992c0823f07b015b9a6f80ad75)
Merged-In:I2086b29ef9bf3bb6fb7d9ebba6e7b8db9392e459
|
|
Because on Pixel 4 and other devices, the light reveal scrim doesn't
have full alpha, the notification panel was marked opaque which removed
the wallpaper. This could lead to the wallpaper not being visible in
a couple of situations where it should have been, leaving AOD exposed.
Because of a separate bug (b/196306312), whenever the wallpaper is
hidden, we'd see a color layer, which would lead to the screen becoming
flickery and awake on AOD.
Fixes: 194760238
Test: atest SystemUITests
Merged-In: I5a7e80e2ced8f15f1b215549d0c051f1d61d2953
Change-Id: I5a7e80e2ced8f15f1b215549d0c051f1d61d2953
(cherry picked from commit d61e4732c78c56182dff98508c41f61871a3d939)
(cherry picked from commit 925211481c04e5861ce3b85e44f64bde09a67676)
Merged-In:I5a7e80e2ced8f15f1b215549d0c051f1d61d2953
|
|
Test: manual
Bug: 197586676
Bug: 202500642
Merged-In: If531a23f5e7b242d3d637966b74b67acfd303199
Change-Id: If531a23f5e7b242d3d637966b74b67acfd303199
(cherry picked from commit 022081595bf4a082f8354b852eae6d5a015be3fe)
Merged-In:If531a23f5e7b242d3d637966b74b67acfd303199
|
|
Change-Id: Ic2889f5eb531008340529eadc36ec8efc62b1984
|
|
Change-Id: If7c68a17857eee4194f28413b938d5647820162c
|
|
Test: manual
Bug: 197586676
Bug: 202500642
Merged-In: If531a23f5e7b242d3d637966b74b67acfd303199
Change-Id: If531a23f5e7b242d3d637966b74b67acfd303199
(cherry picked from commit 022081595bf4a082f8354b852eae6d5a015be3fe)
Merged-In:If531a23f5e7b242d3d637966b74b67acfd303199
|
|
(QS Internet Dialog)
- When the location is disabled, the WiFi scanning service will also
stop.
- Bypass the WifiManager#isScanAlwaysAvailable() result.
Bug: 203471614
Test: manual test on the device.
atest -c InternetDialogControllerTest \
InternetDialogTest
Change-Id: I2f159820d577f226e640b3ea735f6b18c52a16f7
(cherry picked from commit 194a61951688b8ed48b6c0e254398f66c5b1d79a)
Merged-In:I2f159820d577f226e640b3ea735f6b18c52a16f7
|
|
1. The log should be controlled by the DEBUG, otherwise it will be printed infinitely and the performance will be very slow.
The log can be removed after checking, so I removing it instead of adding the DEBUG tag.
2. Do too much get icon in the main thread, move it into bg thread
3. It is unnecessary that update the mobile data layout from the callback, onAccessPointsChanged, because it is used for Wi-Fi.
Bug: 201200079
Test: manual and Q verified pass
Change-Id: I5f4623f9ffb68fae6ca377625aaa039f0f33dd38
(cherry picked from commit 5f685c98cc877f61e0297a9507412cb83f83d84b)
Merged-In:I5f4623f9ffb68fae6ca377625aaa039f0f33dd38
|
|
(QS Internet Dialog)
- When the location is disabled, the WiFi scanning service will also
stop.
- Bypass the WifiManager#isScanAlwaysAvailable() result.
Bug: 203471614
Test: manual test on the device.
atest -c InternetDialogControllerTest \
InternetDialogTest
Change-Id: I2f159820d577f226e640b3ea735f6b18c52a16f7
(cherry picked from commit 194a61951688b8ed48b6c0e254398f66c5b1d79a)
|
|
Because we were only checking for the bouncer to be shown and not
the alternative auth interceptor, the unlock mode always ended up
being unlock_collapse instead of dismiss_bouncer.
The isBouncerShowing() was already taking into account the alternative
auth path, but the bouncerIsOrWillBeShowing() didn't
Fixes: 204049350
Test: atest packages/SystemUI/tests/src/com/android/systemui/statusbar/phone/StatusBarKeyguardViewManagerTest.java
Merged-In: Id8bf8460bff26ec5694c82466ea63e3cdf61216e
Change-Id: Id8bf8460bff26ec5694c82466ea63e3cdf61216e
(cherry picked from commit 571fdde909a333a3486d93ac94c677da787b4413)
|
|
1. The log should be controlled by the DEBUG, otherwise it will be printed infinitely and the performance will be very slow.
The log can be removed after checking, so I removing it instead of adding the DEBUG tag.
2. Do too much get icon in the main thread, move it into bg thread
3. It is unnecessary that update the mobile data layout from the callback, onAccessPointsChanged, because it is used for Wi-Fi.
Bug: 201200079
Test: manual and Q verified pass
Change-Id: I5f4623f9ffb68fae6ca377625aaa039f0f33dd38
(cherry picked from commit 5a1bb06a0a8d6d730714f02826ab957a4cc220c7)
|
|
- Show message when Wi-Fi is OFF and Wi-Fi Scanning is ON
- Don't show message when device is locked
- Show a "Change" link to launch Wi-Fi Scanning Settings
Bug: 202323013
Test: manual test to verify on the device.
atest -c InternetDialogControllerTest \
InternetDialogTest
Change-Id: I5eb4563a69e85050532c8b4fe3748e01cc7d9ca5
(cherry picked from commit 8c0847b0fe8df54c35fe747220d2c76ffe8a59c7)
|
|
This was broken by ag/15974634, but the presubmits did not catch that for some reason. That CL changed the brightness logic to be <= the dim value, so this failure (and fix) make sense.
Test: atest SystemUITests, since apparently treehugger won't do that and will +2 this anyway
Fixes: 202212576
Change-Id: Ibe18322fbe1e1fbae86c5274615ad0b6f727c8e2
(cherry picked from commit 20d85a91d4d2ac22760111483b4bdeaecace6c5f)
(cherry picked from commit fc534e4f63a4ac9953394cb69d2f8f25ca508464)
|
|
- Align doze brightness logic with DisplayPowerController, so that the 'minimum dim amount' is taken into account by DozeScreenBrightness.
- Clamp brightness to the now-correct dim value during the entire screen off animation, rather than sometimes transitioning to the sensor value. Apply the sensor brightness value on changing to the DOZE Display state, which happens after the screen off animation.
Bug: 194972547
Test: at 0%, 45%, and 100% brightness, allow the screen to time out in bright and dark rooms, observe minimal brightness changes
Change-Id: I528a944e04f8906cd8c7740eea97dd8fba7c3d3e
(cherry picked from commit 2c26ea64bc74c7c75e3f65bd9e8cebd6494571d5)
|
|
- Show message when Wi-Fi is OFF and Wi-Fi Scanning is ON
- Don't show message when device is locked
- Show a "Change" link to launch Wi-Fi Scanning Settings
Bug: 202323013
Test: manual test to verify on the device.
atest -c InternetDialogControllerTest \
InternetDialogTest
Change-Id: I5eb4563a69e85050532c8b4fe3748e01cc7d9ca5
|
|
Original change: https://googleplex-android-review.googlesource.com/c/platform/frameworks/base/+/15994907
Change-Id: I3e7e18f6c023842eb6de84845246c0a66ff012f0
|
|
timeout." into sc-qpr1-dev
|
|
This was broken by ag/15974634, but the presubmits did not catch that for some reason. That CL changed the brightness logic to be <= the dim value, so this failure (and fix) make sense.
Test: atest SystemUITests, since apparently treehugger won't do that and will +2 this anyway
Fixes: 202212576
Change-Id: Ibe18322fbe1e1fbae86c5274615ad0b6f727c8e2
|
|
Else, the clock may animate in at the wrong font weight.
Test: atest AnimatableClockControllerTest
Test: manually swipe away all notifications on LS,
see the large clock animate in at full font weight
Fixes: 201615075
Change-Id: I139e5a5c88551f23f99decff3e7ffc8bdaa5bba4
|
|
- Align doze brightness logic with DisplayPowerController, so that the 'minimum dim amount' is taken into account by DozeScreenBrightness.
- Clamp brightness to the now-correct dim value during the entire screen off animation, rather than sometimes transitioning to the sensor value. Apply the sensor brightness value on changing to the DOZE Display state, which happens after the screen off animation.
Bug: 194972547
Test: at 0%, 45%, and 100% brightness, allow the screen to time out in bright and dark rooms, observe minimal brightness changes
Change-Id: I528a944e04f8906cd8c7740eea97dd8fba7c3d3e
(cherry picked from commit bb8219d84b2dcf91176674dabc856dc98fc03f7a)
|
|
timeout." into sc-dev
|
|
sc-dev
|
|
- Align doze brightness logic with DisplayPowerController, so that the 'minimum dim amount' is taken into account by DozeScreenBrightness.
- Clamp brightness to the now-correct dim value during the entire screen off animation, rather than sometimes transitioning to the sensor value. Apply the sensor brightness value on changing to the DOZE Display state, which happens after the screen off animation.
Bug: 194972547
Test: at 0%, 45%, and 100% brightness, allow the screen to time out in bright and dark rooms, observe minimal brightness changes
Change-Id: I528a944e04f8906cd8c7740eea97dd8fba7c3d3e
|
|
- Align doze brightness logic with DisplayPowerController, so that the 'minimum dim amount' is taken into account by DozeScreenBrightness.
- Clamp brightness to the now-correct dim value during the entire screen off animation, rather than sometimes transitioning to the sensor value. Apply the sensor brightness value on changing to the DOZE Display state, which happens after the screen off animation.
Bug: 194972547
Test: at 0%, 45%, and 100% brightness, allow the screen to time out in bright and dark rooms, observe minimal brightness changes
Change-Id: I528a944e04f8906cd8c7740eea97dd8fba7c3d3e
|
|
Change-Id: Id27ea30be4f20be6f5d7d6336c8f53d2644c8e25
|
|
sc-qpr1-dev
|
|
Even if the shade is expanding. This ensures that when the user
expands a notification that is hiding sensitive content,
that we still show the udfps bouncer even though the
transition to shade is also occuring (which normally hides the
udfps affordance).
Test: manual, atest UdfpsKeyguardViewControllerTest
Fixes: 197119617
Change-Id: Iac398a60ecf5137d0c3a8c77d700146e12fad0b7
(cherry picked from commit 5a5cd54366f2d64a0bdff6455f381e7746eb741a)
Merged-In: Iac398a60ecf5137d0c3a8c77d700146e12fad0b7
(cherry picked from commit 1c79bbde1586666cbdd5a1127cb9c49f4ff881c7)
|
|
|
|
|
|
This is due to BiometricUnlockController#isBiometricUnlock() returning
true after finishing unlocking.
In order to have a more resilient logic, we can also verify if they
device is already unlocked and the keyguard is gone.
Test: manual
Test: atest StatusBarTest
Fixes: 200882274
Change-Id: I4e0aeeec407db1a9dcbcad2960315cb68d7ac3b3
|
|
- Instead, just animate the current nav bar view (which doesn't cause
insets changes). This should be safe to do because:
1) Nothing else animates the alpha of this particular part of the
view hierarchy
2) Confirmed with SysUI folks that we always get onStartedWakingUp
after onStartedGoingToSleep
3) This otherwise reverts to the same nav bar visibility logic in R
Bug: 196789451
Bug: 190368572
Test: atest StatusBarKeyguardViewManagerTest
Test: In three button nav w/ AOD, turn off the screen
- also try above with interrupting the screen off animation
- also try above with power button instantly locks disabled
Change-Id: I1be44a9a8dbb8cea17bbf354a2e03dd4bad5a4a2
|
|
|
|
There was discrepancy between how null values and empty strings were
treated in KeyguardIndication and KeyguardIndicationController. This
could lead to a crash in systemui. Modify the
KeyguardIndicationController to match the logic in KeyguardIndication.
Bug: 201339576
Test: KeyguardIndicationControllerTest
Change-Id: I8b7b4d7b33b1ee809df17e64347c816f65a50727
|
|
See bug comment #9 for an in-depth explanation.
Test: New unit tests.
Test: I created a test app that sends an ongoing call notification
then removes it 200ms later, and I confirmed that the bug is triggered
without this CL, and is fixed with this CL.
Bug: 199600334
Change-Id: I0f681ccfe55597490bee0d4b190f9fb2c38e28f2
(cherry picked from commit cd755525c4f172862d27466f06f5d4230b57ade5)
Merged-In: I0f681ccfe55597490bee0d4b190f9fb2c38e28f2
|
|
sc-qpr1-dev
|
|
After the device is setup, if the smartspace component is not
available/initialized, don't attempt to start a session.
Fixes: 199986217
Test: atest LockscreenSmartspaceControllerTest
Change-Id: I36d6f75b37cc5f5274b1cea3a35338f6da424a20
|
|
|
|
Even if the shade is expanding. This ensures that when the user
expands a notification that is hiding sensitive content,
that we still show the udfps bouncer even though the
transition to shade is also occuring (which normally hides the
udfps affordance).
Test: manual, atest UdfpsKeyguardViewControllerTest
Fixes: 197119617
Change-Id: Iac398a60ecf5137d0c3a8c77d700146e12fad0b7
(cherry picked from commit 5a5cd54366f2d64a0bdff6455f381e7746eb741a)
Merged-In: Iac398a60ecf5137d0c3a8c77d700146e12fad0b7
|
|
* update the UDFPS icon burn-in parameters to be in pixels
because we don't want display density to increase the
position movement
* when the clock is centered aligned, we want to make sure
it doesn't overlap with the udfps icon
Test: atest SystemUITests, manually update clock times with
with DozeUi.BURN_IN_TESTING_ENABLED enabled and see clock
doesn't interfere with the UDFPS icon.
Fixes: 192423460
Change-Id: I6f95fe3ad763971b804129b517f4c98192ab3674
Merged-In: I6f95fe3ad763971b804129b517f4c98192ab3674
|
|
|
|
Before, when the device was rotated we would get the new orientation value
from the event but then use the value from getResources() to decide
which layout to show. This discrepancy is a potential cause for an issue
where QS uses the landscape layout when the device was actually portrait
Fixes: 159737544
Test: atest QSPanelControllerBaseTest QSPanelControllerTest
Change-Id: Icd15712cd4821b7833a3373978a92f8b5959ac12
Merged-In: Icd15712cd4821b7833a3373978a92f8b5959ac12
|
|
Bug: 193149550
Test: follow repro steps
Merged-In: I49e2b8bcec7b2ce0a9776ff30a64c07f24949da7
Change-Id: I49e2b8bcec7b2ce0a9776ff30a64c07f24949da7
|
|
Bug: 193149550
Test: follow repro steps
Change-Id: I49e2b8bcec7b2ce0a9776ff30a64c07f24949da7
|
|
sc-qpr1-dev
|