Changed public constants from integers to strings. Internally
everything is still integers, since we want that more efficient
representation for most things.
Changed the Callback interface to OnOpChangedListener. We also
have a private versin that again takes an int, and tricks to
make both work.
Reworked the class documentation to be appropriate to the SDK
(as much as it can be); most of the existing documentation is
moved to the private implementation. Also added documentation
of the MODE constants.
Change-Id: I4f7e73cc99fe66beff9194e960e072e2aa9458f8
When a file already exists on disk, try adding a counter suffix to
make a unique name. Move services near top of roots list, just below
recents. Remove "Documents" root.
Increase number of recents allowed from single provider, and add more
logging to diagnose wedged loaders.
When launching GET_CONTENT apps, wait for successful result before
relaying result; canceled requests now return to DocumentsUI.
Add CloseGuard to ContentProviderClients, since leaked instances can
keep the remote process alive.
Fix UI bug around trailing breadcrumbs. Fix bug that dropped Recents
from roots list. Add up action to Settings activity. Give our
activity a default icon while waiting for async roots to load.
Bug: 10818683, 10819461, 10819461, 10819196, 10860199
Change-Id: I7b9e26b1cf8353dd3175458b23da2b4bda6c5831
Previously I was calling setIsConnectedToProvisioningNetwork(false) always,
but all MDST's receive every broadcast. Thus we could over write an MDST's
mNetworkInfo.mIsConnectedToProvisioningNetwork to false, unless the MDST
that was set to true was last, i.e the code was order dependent.
If the provisioning networks value was false instead of true
when handleMobileProvisioningAction was called we wouldn't invoke
mdst.enableMobileProvisioning because network info would be null.
Thus the provisioning network would never transition to CONNECTED and
a default route wouldn't get setup and the browser couldn't access the
website.
Now setIsConnectedToProvisioningNetwork is only set to false when the
apnType matches and we won't indiscriminately change it and are not
order dependent.
Bug: 10853805
Change-Id: I68a4f9bdf5dc18d90f4cdef7a60811f57be67261
- CaptureResult#getRequest is used to tie a result to a request (for convenience)
- Add new CaptureFailure class to describe capture failure
- Results/frame numbers also return frame numbers, sequence ids
- Captures now all return the sequence id
- A sequence id onComplete is available in the CaptureListener
Bug: 10360518
Change-Id: I9ebaa45698c718a1185b5ae920b7975925fe2f60
...device to another mode (portrait or landscape) on Main page.
So, it turns out that Bundle claimed to have an invariant that either
mParcelledData or mMap would hold its data, never both. The new
implementation on top of ArrayMap assumed this was the case. However,
there is one situation where it is not true: an application can take
an existing Bundle that contains data, and call readFromParcel() on it.
The implementation of readFromParcel() would just pull out the
parceled data and stuff it in to mParcelledData for later unparceling,
even if that Bundle already had a non-empty mMap.
To fix this, we just look for this case in readFromParcel() and
immediately unparcel at that point into the existing map, using a
new unparcelling method that doesn't rely on the target map being
empty.
Change-Id: Ib816b6876a6cd2760b7a3372c7a79ca2f12dfeba
The user has to explicitly enable a print service from the settings UI
before using it. Usually, users very rarely if at all interact with print
services, therefore all print service management task are performed from
the print settings. We also have to get user consent warning that the
user's data is about to be given to a third-party app. We now post a
notification allowing the user to go directly to the settings screen to
turn the service on.
bug:10447510
Change-Id: Iea56c0825f0bf38328ad94912f0ea5576e9339b3
* commit '47d620ac282870f1f7142bd86ab32d14ad9ee91f':
Ensure access to the map of geofence callbacks is synchronized in all cases Fix Java reference leak in the JNI layer
This patch fixes behavior where attempting to place the cursor at
the end of a string where the last character is a combining accent
actually placed it before that accent. This was especially annoying
for editing Thai text, because it made it difficult to delete a
trailing tone mark.
Fixes bug 8947569 "Can't place cursor after combining accent" and bug
10398332 "[Thailand] Thai Language(IME) -- Can't delete Thai tone mark
while writing message"
Change-Id: Ida1933c8f0ab6cdb0200db39891e9389e4bdba86
Apps without sdcard_r or sdcard_rw are still able to write to
their package-specific directory, but someone needs to first make
that directory on their behalf. This change will delegate the
mkdirs() call through to vold when an app fails to create directly.
MountService validates that the path belongs to the calling user, and
that it's actually on external storage, before passing to vold.
Update Environment to make app-vs-vold paths clearer.
Bug: 10577808
Change-Id: I43b4a77fd6d2b9af2a0d899790da8d9d89386776
