Long is used in PropertyValuesHolder class to store native pointers
as they can be 64-bit. Note that jmethodID, a pointer to structures,
is also carried in long rather than int to support 64-bit system.
Change-Id: Ifb514fc8473d7c41c0d6194fc6eb85d4816b2cd9
Signed-off-by: Marcus Oakland <marcus.oakland@arm.com>
Signed-off-by: Ashok Bhat <ashok.bhat@arm.com>
All pointers are now 64bits wide, so should be
represented as java longs and not ints.
Also changed DelegateManager and SparseWeakArray to
reflect the new world order.
(cherry picked from commit 88a8364c386c694f7ad56662ef89713dbf7c9d63)
Change-Id: Ic2f55dd6235751169c5014f9d2ccf3f544259a87
All pointers are now 64bits wide, so should be
represented as java longs and not ints.
Also changed DelegateManager and SparseWeakArray to
reflect the new world order.
Change-Id: Ic32b6b53818dbae9b949f03004c4fb6dae26cdbe
This change ensures that assets that are not needed for the
preferred density are stripped from additional directories
specified by the -S flag
For simplicity, the logic for deciding which assets are not
needed is still applied per directory, rather than globally,
which may still result in some unneeded assets being included
in the final output.
Bug: 12604267
Change-Id: Ic1378570c2d9cd1c854c507e784233818c19efca
This lets us share zip archive processing code with both
the runtime (Art, dalvik) and critical java code
(StrictJarFile).
This change also moves several utility methods to ZipUtils
and dedups code across several zip inflation methods.
One of the side effects of this change is that several
processing loops are now O(n) instead of O(n^2).
bug: 10193060
(cherry picked from commit e9d380f9303692950166472b1dfab083e1af8133)
Change-Id: I53e7d0c0af266cb1fe4e293e1177b2590cc44224
It forces aapt to return an error if aapt fails to find an resource
entry for a configuration.
Bug: 11259444
Change-Id: Ie5674a29dff5d4455e7d7c94f6b25560fb1305b7
If the manifest specifies a maxSdkVersion for which to
stop granting a certain permission, output that value
when dumping badging.
bug:11630700
Change-Id: I922a3186340383828e1af3ce0815efb407d9d535
Also, changed getComponentName() to output a String8 object
instead of a const char * because the const char * is an internal
buffer of a String8 object which gets immediately destroyed
after returning from getComponentName().
Bug: 11329761
Change-Id: Ic459dec0ad3b20162c36de0ee492bcc022863b12
In frameworks/base commit 1125235da15a5655e1fdf3b5f24df48dafccee90,
the IPowerManager interface was updated to add updateWakeLockUids,
but this file was not updated.
Fixed.
Change-Id: I8d5def02b7cadf5a154634bdff18cacb84b718f4
When a preferred density is specified, prune all but the
closest matching density when the exact density resource does not
exist.
If a preferred density resource does not exist and a higher density
resource exists, prune all but the resource with the next highest
density relative to the preferred density.
If a preferred density resource does not exist and there is no
higher density resource, prune all but the resource with the next
lowest density relative to the preferred density.
Manual merge of I99b3950fe3d8116a5625968ed383442315ae8526 from
frameworks/base/tools to frameworks/tools
bug:11224465
Change-Id: Idddf55dad27f114ffc429b61317c3a5b1435d808
This fixes outputting string as well as not crashing when the
<meta-data> element has an "android:resource" attribute instead
of an "android:value" attribute.
Bug: 11255844
Change-Id: Iadb62b5dcb18ea3db8dbd2ba3241f489606d535d
Rather than ignoring resources that do not match the specified
product, we keep track of the ignored ones and make sure that
some variant of the resource that matches the product was processed.
bug:10860838
Change-Id: I7a35f37fda2c9561634f75323bd4cb2b9047e29b
Originally from I4d6773a88ea2932982278127a3c96d38be54ddf5
- file has move to a different git project in master, so manually
re-applied the diff that landed in klp-dev
Change-Id: I813210bd6160ab71c7a4dff9c221db31d8cc305c
Bug: 10447005
Call png_set_interlace_handling explicitly instead of relying on
implicit handling that logs warnings
Include filename when printing warnings
Change-Id: Ia343427f5522dc8ab1010f8d7017e86f389caf99
There are still some errors
1. Little vertical clippping for extra tall glyphs.
2. Breaking into scripts isn't perfect which results in incorrect layout
of text.
Change-Id: Ib7b008b4ab5e689038678825289199aeb4bd815b
This changeset adds the framework resources for RTL locales and mirrors
the layout if the application is RTL aware.
Use ICU to check the character orientation of the locale - right to left
or left to right. Set the layout direction on the top level layout
accordingly. Also, load the RTL resources for Nav Bar when the locale is
RTL.
Change-Id: Icbb57ee2ac7c6d8dfc34c2f04dce34d820b9b1ed
Part of the Transition API changes, except that the tools/ directory
got moved between klp and master, so re-making the same changes here.
Issue #10460684 KLP API Review: android.view.transition and android.animation
Change-Id: Icedca1dc689285b4b213f9376656b26f18c7f23e
Cookies are really indices into vectors and arrays, so
they don't need to be void*. We choose int32_t instead
of size_t to allow their width to be well specified.
(cherry picked from commit ebfdd0f467e39c3af8d92cade78263935340acb7)
(cherry picked from commit a7fa2e592e2e579e5acdb903dba83fc074ebc215)
(cherry picked from commit a9d5701b034ed2d9771b3f0943e1add00741d7cd)
Change-Id: I2aed3db568b6fdc487bf99e2c5dd123206736fda
The height of a layout should be zero if it is assigned a layout_weight.
This way, the layout is measured only once and prevents spurious errors.
Bug: https://code.google.com/p/android/issues/detail?id=58398
Change-Id: If49a7480e5eb82cb86780e00f2f5b65ee053fc2a
