Merge commit '6d42d80653f2c41f3e72a878a1d9a6f9693b89f7' into eclair-mr2
* commit '6d42d80653f2c41f3e72a878a1d9a6f9693b89f7':
Fix issue 2304669: VoiceIME: starting and canceling voice IME yields persistent "error 8" state on future attempts and breaks voice search.
Merge commit '8445f367070514dce3154be73827107b8a93f234' into eclair-mr2
* commit '8445f367070514dce3154be73827107b8a93f234':
1) Capture the start memory before the test started
* changes:
Change OMXCodec::Create to return an sp<MediaSource> instead of an sp<OMXCodec>, this is more general and does not sacrifice any functionality as the remaining OMXCodec APIs are not meant to be public anyway.
Fixed AudioFlinger::openInput() broken in change ddb78e7753be03937ad57ce7c3c842c52bdad65e
so that an invalid IO handle (0) is returned in case of failure.
Applied the same correction to openOutput().
Modified RecordThread start procedure so that a failure occuring during the first read from audio input stream is detected and causes
the record start to fail.
Modified RecordThread stop procedure to make sure that audio input stream fd is closed before we exit the stop function.
Fixed AudioRecord JAVA and JNI implementation to take status of native AudioRecord::start() into account
and not change mRecordingState to RECORDSTATE_RECORDING if start fails.
Merge commit '0019215fc395ef12c191049b1903eeabf70859cf' into eclair-mr2
* commit '0019215fc395ef12c191049b1903eeabf70859cf':
Revert "When using MDP, we needed to use a texture for diming."
Without this, boot fails with:
12-05 21:06:56.984: ERROR/Zygote(435):
java.lang.IllegalStateException: Missing class(es) for preloading,
update preloaded-classes [android.server.search.SearchableInfo]
This was caused by https://android-git.corp.google.com/g/34797
Change-Id: I195b7e72a411401e2de8c1d6c2e7db0fe1badec6
We can now locate event log tag definitions in individual packages
(and java constants for the tag numbers get auto-generated), so move
all the tags used by the system server into the package.
* changes:
Improve error message thrown when eglCreateContext fails.
Improve error reporting by always returning the public EGL10.EGL_NO_XXX objects for displays, surfaces, and contexts.
When the user lifts her finger up, AbsListView computes the fling velocity
and report a SCROLL_STATE_FLING if the velocity is high enough. However,
when the fling is not initiated because the velocity is too low (case of
a normal scroll), the idle state was not set nor reported.
Merge commit 'ae952b3bcc3eb744cceb5cd0ae65b2c7a83f9de7' into eclair-mr2
* commit 'ae952b3bcc3eb744cceb5cd0ae65b2c7a83f9de7':
If the usage stats file doesn't exist in the first place there is no need to
The StringPool indicates the length of a string with a 16-bit
integer. If the length of the string is greater than 0x7FFF,
it splits it into two 16-bit integers with the first one
having the high bit set.
The length calculation has a small bug that masks off the
19 bits instead of the first 15 bits as intended.
Merge commit 'd3bc1994b0c2c5e9c60fa748de47eebbe1984fda' into eclair-mr2
* commit 'd3bc1994b0c2c5e9c60fa748de47eebbe1984fda':
Don't accidentally signal multiple exceptions in setThreadPriority()
Merge commit '052d9c8b414dc9eb1c70d849660592d9bc3b2644' into eclair-mr2
* commit '052d9c8b414dc9eb1c70d849660592d9bc3b2644':
Images.Media.DATE_TAKEN should be in milliseconds.
After signalling an exception due to set_sched_policy() failing (probably
because the thread in question had exited before this function was called), we
proceeded to call setpriority() anyway, and raised *another* exception when it
also failed. We don't do this any more; we just bail after signalling the first
exception.
Change-Id: I397b6a93c73d3f0870176e26759f2bec33e56205
We can now locate event log tag definitions in individual packages
(and java constants for the tag numbers get auto-generated), so move
all the tags used by the system server into the package.
Cached views would, in some situations, not update if they contained an
animating child. This was caused by clearing the dirty cache flag too
early in View.buildDrawingCache().
Approved by Dr. No mcleron.
Change-Id: I8c5f2fc3e6605657e0da625d60d50b55bb133666
WebCore treats */*+xml as content that it can render. This prevents Activities
from intercepting those mime types. Check the same regular expression as
DOMImplementation and then use the PackageManager to look for a matching
Activity.
Requires a change to webcore.
Bug: 2256744
There's no need to have constants like OP_UNUSED_FF in the API. They
serve no purpose, and it makes life confusing when an opcode is defined
(does the "unused" one go away, breaking the API? do we have the real
definition and the "unused" definition simultaneously?).
Discussed / approved by API council.
We need to expose SearchableInfo in the SDK in order to unbundle
Quick Search Box. Since the android.server.search package is
hidden, I'm moving SearchableInfo to android.app, where
SearchManager lives.
This change doesn't actually expose SearchableInfo. I'll do
that in a separate change to keep the change that
api-council needs to review small.
This is part of the fix for: http://b/issue?id=2270838
Change-Id: I9589f9c2c11d36c958beedff8245fe0c3319c6ba
EGL10.EGL_NO_XXX objects for displays, surfaces, and contexts.
This allows clients to compare the returned object against
the public EGL10.EGL_NO_XXX object using a simple == operation.
This fixes bug 2303947 "Java layer EGL API makes it difficult
to tell when certain methods have failed"
Rename the old MockContentProvider to MockIContentProvider since it is
more appropriate name.
Detail:
Current developers inevitably depend on the backend used by ContentProvider,
which is useful but not ideal nor "testable" from the view of them.
Current MockContentResolver only accepts exact "ContentProvider" class, not
IContentProvider interface, since we want to hide "IContentProvider" while
the old MockContentProvider implements IContentProvider and as a result some
methods we want to hide may be exposed to the public SDK now and probably
for the future.
On the other hand, ContentProvider is not interface but an exact class
heavily depends on the internal logic and not suitable for external
developers to use for tests.
The new MockContentProvider introduces the mock implementation
for ContentProvider.
It extends ContentProvider, so "is" ContentProvider, but tries to avoid
depending on any backend System like IPC in Android, etc.
This should be useful from the view of application developers who do not
want to be confused with ContentProvider/ContentResolver backend
implementation "at all" and want to use MockContentResolver without
any other ContentProvider implementations tightly connected to the
external worlds.
