Lens shading correction should be full correction that
corrects both color shading and vignetting.
Bug: 18175853
Change-Id: Ia552345b2bd78b1b2eed2472096ebae80680d70b
- system image download links for DP2
- OTA download links for DP2
- documentation download links for DP2
Change-Id: Ibe0127fd7d5efec90478de07eb2a7b3d342bd67c
Previously if we were in multi-window mode and some task called
startLockTask() or kiosk mode was started - other tasks remained
on screen. User could still interact with other tasks, locked or
pinned task could be removed.
This CL makes tasks move to fullscreen when they are locked.
Bug: 27876860
Change-Id: I547a94be50276c7f418343600877e51b6b40c7a7
The goal of this CL is to enable
InputMethodSubtype#getLocaleDisplayName() to return more natural locale
display name in terms of capitalization rules.
The key idea here is to use LocaleDisplayNames#localeDisplayName() with
an appropriate DisplayContext.CAPITALIZATION_FOR_* parameter rather than
relying on Locale#getDisplayName().
Bug: 22845728
Change-Id: If105082ce703db7a86738455db7e9fb37f3c6fe8
* Check null result for all MTP APIs that can return null.
* Don't regard valid parent handles that are more than 0x80000000 as
error.
* Check boolean result from importFile.
Change-Id: I262511ac3c935ebb54e52c03eaefd3535a3e2f29
Fix: 28155538
This includes following animation.
- Fix focus change animation duration among the PIP control buttons
- Apply ripple animation for PIP control buttons
- PIP menu enter/exit animation.
Bug: 27677096
Change-Id: I2f992416efc8a0994eeb56d5dec00201fd08a7aa
If the user's wallpaper isn't direct-boot aware, wait around for
the user to be unlocked, instead of clearing the wallpaper.
Also switch a few classes to using SystemService lifecycle, since
events are dispatched faster than through broadcasts. Fix bug where
ContentService.systemReady() was never called, and make sure
EntropyMixer doesn't risk being GC'ed.
Bug: 26280055
Change-Id: I9fff468a439b868baa68cf11bb6ee9f7d52b7b5a
When calculating all the frames and insets, we need to use the actual
frame instead of layoutContainingFrame. To do this, we layout mFrame
with layoutContaining/layoutDisplayFrame, calculate all the content
frames and insets, and then offset everything by the constant offset.
Bug: 28075359
Change-Id: I78f0a54ca2a0cc6c7c8be21153c2b2c8f1d5c0a9
Windows of TYPE_APPLICATION (as opposed to TYPE_BASE_APPLICATION)
are not child windows in the sense of SurfaceView, etc, as they are
independent windows like Modal Dialogs rather than embedded parts
of other windows. Still though, we expect them to reappear following
activity relaunch, and they won't be covered by window preservation,
so we need to mark them for replacement.
Bug: 26668339
Change-Id: I652b4137085f6ef4d6c9d54de609727f966ef4d6
am: cd40b61
* commit 'cd40b619e8e955e3005fec15c074f7cf0ed705d7':
docs: Continuing work on the N DP2 release notes.
Change-Id: I9e3872e8b53ea83b23a83065d5b2d975e7658d35
Because of a min/max issue, while processing each paragraph StaticLayout
would trim the left/right indent arrays incorrectly.
Bug: 28090810
Change-Id: Ib2b9b48963861e0952bd45a079179e3cca86ffcf
am: f5266e5
* commit 'f5266e563341f9c4f07c1306e983f32a24317f09':
docs: Continuing work on the N DP2 API Overview.
Change-Id: If5736642ea6809e0ace1a043069bb176a8692599
Also make them unfocusable, and don't focus the docked
stack when tapping into it in that state.
Bug: 27972642
Change-Id: Ic24ff9a5f39f596fe4a2f50567566d4400f9c125
DHCP receive thread was spinning on a boolean variable to know when to
stop. That variable had no volatile qualifier, potentially preventing
the thread from stopping at all. Without a volatile qualifier, the reads
and writes in halt() could be reordered in such a way that the running
thread is interrupted before mStopped is set t true. Also, the optimizer
could decide to hoist mStopped in a register inside the running thread,
preventing the loop from exiting.
Change-Id: I5b30c1247808114f0e5b46e230978ee7ea18ab8e
