Currently, the emergency calls are put into an readonly property ro.ril.ecclist.
It is possible that ecclist has to be changed before and after SIM pin lock.
And ecclist can also be changed after network broadcasts the local emgergency numbers.
So add r-w property ril.ecclist to make emergency number list changable, while still
check r-o property to make old RIL work.
Fix a bug in DataConnection state machine where the notification
of disconnection completion was sent before we actually transitioned
to the in active state. Also, change conn.reset to send a response
so the user can know when the transition to the in active state
completes for that also.
bug: 2471897
Change-Id: I5776324ac89a607925d07f4a600bc5b34c3f3ed6
This is the framework part, moving classes around so the framework
no longer needs to link to android-common. Makes some APIs public,
others that didn't need to be public are private in the framework,
some small things are copied.
When performing a logical OR on a byte value, make sure to truncate its
value to prevent possible sign extension when it is converted to 32-bit.
Change-Id: I7bdfded32cec3e8a5f6b5c45316b2af5ab9471d1
When a signed byte value is taken from an array, make sure to truncate
it so it doesn't have a problem when it's sign extended to 32-bit.
Change-Id: Ic2b35c317e21c63dff960501145a4f3074b687ad
This solves a race condition between the network type changing and
a signal strength changing. This could cause StatusBarPolicy to see
the -1 values for the evdo strength parameters and hence 0 bars would
be displayed.
Bug: 2419251
Change-Id: Iba0bdbd2023b6a5e49fc1f82c997ec3219158286
events (and in one case, a DropBox entry).
Add a simple intent that triggers master-clear (and toggle EFS), given the
right permissions.
Bug: 2264596
Bug: 2350452
Bug: 2264596
This is a new implementation of TTY support.
Previous implementation in commit aead64def1fe58c95c086a0ca00cf0b13fa32ef1 is reverted.
The new method does not rely any more on the kernel headset driver to send a UEvent containing
current TTY mode.
Most of these are file moves; a couple notable exceptions are the changes due to the move, and fixing up test code:
- database/DatabaseCursorTest.java
- database/DatabaseStatementTest.java
- net/UriTest.java
Merge commit '64c92f75df7415084aa9f0b304f2a99089b6be9f'
* commit '64c92f75df7415084aa9f0b304f2a99089b6be9f':
Handle emergency calls only status in ServiceStateTracker.
FCC raised the issue of not allowing users to configure
channel counts beyond 11. This change enforces the channel
count based on MCC values.
Bug: 2378844
Backport of MR2 change 421c72b6.
Removes entries sooner. A bug.
Doesn't add one for phone process at all. This was intended to be removed long ago.
bug: 2329900
bug: 2377507
Typically the CallerInfo is looked up based on a phone number, in which case the '_id' column
contains the contact id.
However when the lookup is based on a phone row, the '_id' column is the row number we queried.
The contact id is stored somewhere else in a 'contact_id' column.
The current fix, checks the mime type of the contactRef URL to find out which column contains
the contact id.
I think the contact DB should be more consistent and always return the contact id AS contact_id even
in the phone number scenario.
Bug:2269240
Bug 2339655. Should hangup the existing voice call (including onhold ones) before powering off radio to ensure the proper sequence at radio protocol level. Otherwise the carrier will treat it as drop calls.
sequences (which nobody used) and streamline the API, adding
documentation in preparation for inclusion in the SDK.
Gut and deprecate EventLogTags, which unfortunately was put
into the public SDK (an oversight). Include the functionality
in EventLog proper, in a simpler and easier to use manner.
This change doesn't actually un-@hide anything, but it does
change it to @pending.
