- was a subtle regression introduced when fixing bug #5753006
- as we are now using SkPaint::kGlyphID_TextEncoding (glyph encoding)
instead of SkPaint::kUTF16_TextEncoding (UTF16 encoding), we need to
force the UTF16 encoding in some cases that are NOT going thru
the TextLayoutCache / Harfbuzz shaping
- fix also breakText() the same way
- also clean some old comment
- Warning: depends also on a CL from Skia for having getBaseGlyphCount() "const"
Change-Id: I3d1fc87f070884876c679b33541f810fbfb5df3f
Improved the documentation a little bit.
Fixed a bug in ContentService wherein if a ContentObserver was
passed as an argument and its deliverSelfNotifications() method
returned true, then notifyChange would tell all observers that
the change was a self-change even though it was only a self-change
from the perspective of the provided observer.
Deprecated ContentObservable.notifyChange since it is never
used and in general it shouldn't be because we want the notification
to be posted to the handler.
Change-Id: Idde49eb40777e011a068f2adae8a32f779dfb923
We now have an animation that can start as soon as we detect to
rotate the screen, and have a different phase once the new UI
is ready to complete the transition into it.
Change-Id: I05b9c12a699232acbddc36b4a994a4452db71476
In change Ief74d0e4f4f28dff7a435e9dab1fab1ca1d9bfaf with a sha1 of
e81932e92a773538e1ad1ad1d4bfd8f241860c8d it seemed like a good idea
to throw away all TLV's on errors. In retrospect that was probably
not a good idea. For example on the MTN Ghana SIM the error
occurs because of some 0 pad bytes at the end, all of the actual
TLV's are good, so throwing away everything is unreasonable. Instead
accepting what is valid seems a better approach.
Also, add a couple debug lines on error paths.
Change-Id: I4add0c9cd242f46e0ef8700229d0ec755b9e4c4c
This fixes a complaint from carriers (that we used 8.8.8.8), but also
fixes the case where there is only room for one live radio
connection: the secondary connection (tethering) doesn't have a
default route to prevent on-device traffic from slipping out on the
tethering connection, but tethered dns is proxied through dnsmasq, so
it is appearing as on-device traffic and is unroutable. By switching
to the carrier-indicated dns servers we can use the host-routes
already set for those and kill two bugs with one fix.
bug:5898904
Change-Id: Ida8777687994f353b2d4f2c7db5d6ea4b6ac3882
The VM is allowed to use null to represent the bootstrap class
loader, so attempting to call methods on it is a bad idea. Use
the system class loader instead.
Change-Id: I9190848945f679d546d5fb30aba10fd27c7e5404
This change enables the use of the EGL_KHR_fence_sync extension in
SurfaceTexture on omap4 and s5pc110 platforms.
Change-Id: Icad5245bab445413ffb8a7c823c296b678bf3250
The special shared-storage step was mistakenly writing its data directly
to the USB output pipe rather than to the proper stacked data handling
chain that applies compression and encryption. Fix this by getting rid
of the custom handling of the shared-storage data, instead folding it
into the normal data handling flow [with a small amount of additional
management because e.g. it doesn't need a "manifest" pseudofile in the
archive stream].
Fixes bug 5897791
Change-Id: I3995b07963334d2f8cce49b247c87d3d3ff93bed
