Bug 22086521
Views matched by instance, or appearing/disappearing, or
named views were not limited.
Change-Id: I70fd1497c58a3cde771f4eafe11d8ebe0889ab6c
Cherry pick of b561f39d01c211425bfefaaa7b31ebe097e7ba79 from AOSP master.
Due to precision loss of float math, we sometimes get 1.000001f for
dotProduct. This causes NaN result from acosf() and floor() funcs.
At the moment, this does not cause any problems on ARM, as casting
NaN to int results in 0. On mips however (possibly on x86), such cast
gives INT_MAX, so crash occurs when trying to use the resulting value.
Change-Id: I8e0285a0306a65b8469d9f4885c19665066fc4c8
Bug 22063111
When transition.end() is run, it removes itself from
the list of running transitions and perturbs the list.
Change-Id: I4feb7ebe19717a0e2302844d4e4e0d19a55ec57c
When transitioning from ON to OFF with LE Advertisers, advertiser do not
get a chance to unregister themselves as the stopAdvertising checks the
state of the stack and throws before unregistering the object.
It will then never remove the callback objects causing a leak.
b/22092678 | Remote service crash after switching to restricted profile
Change-Id: I04817026a524d10d60abdd8b533554a71a0112e2
In practice, if we ever see an apparent transaction-too-large result
with a modest-sized payload, it means that the remote process died at
just the right time (with the binder transaction already in flight
so it wasn't detected as a DEAD_OBJECT up front). Don't throw
TransactionTooLargeException in this case, because we really do need
to distinguish that from dead-remote more accurately.
In particular, certain common execution patterns on existing hardware
trigger this circumstance, and they wind up crashing the system.
This is bad, so now we avoid it unless we're pretty sure that is
really what happened.
Bug 21801759
Change-Id: Id05f1eecc0d23dc8d0505c402e2cb68396782135
Bug: 22090564
These tests were not touched for at least 3 years, and with webview
moving to chromium, they are tested at the chromium side.
Change-Id: I4424871dd97c5dc23bb3dbdcee0915d5ab893c7e
Selection start can be bigger than selection end and this kind
of selection is generated by mouse selection at least from L MR1.
To fix and support SHARE button for this kind of selection,
getSelectedText need to invert selection start and selection end
when the selection start is bigger than selection end.
Bug: 22065979
Change-Id: Ide4ac41e3f59d256033aca004450712ca0bb0f03
OnPreDraw is called even if the View is not visible.
So need to check isShown() and hasWindowFocus() before calling
starting selection action mode.
This hack is originally introduced for keeping selection on device rotation.
I manually verified this issue does not revive with this CL.
Bug: 22036870
Change-Id: I814db6165e2345fcacedcbd399c1a3985501c8fd
Some apps such as OpenVPN Connect for some reason generate
PKCS#1-padded RSA signatures using Cipher (initialized in
Cipher.ENCRYPT_MODE with private key!) instead of using Signature.
Unfortunately, RI supports this strange behavior and previous releases
of Android Keystore did the same. So, we have to continue supporting
this craziness.
Bug: 22083023
Change-Id: Ife3950d0d4ceb4e44e08014635312c1252878b69
The text selection handles should be hidden / shown when the window
loses / regains focus.
Additionally renames method to make more sense.
Bug: 22062480
Change-Id: I6e160234cf112ee285367637e2f1c14defd82e89
