| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
|
| |
Bug: 16031597
Change-Id: Ifa547cdfde19f67a3c85d39c1d50ae317bca5932
|
|
|
|
| |
Change-Id: Id1e3730d897e9f7e8f416fdae0d1da106a63097c
|
|
|
|
|
| |
Change-Id: Iadc3979ff2a08d485d92a12ff29472f3cd447743
Auto-generated-cl: translation import
|
|
|
|
|
| |
Change-Id: Ie463803531bd671303f0dbafe61b635c5c447bff
Auto-generated-cl: translation import
|
|
|
|
| |
Change-Id: I27a990d7f46fb95667931327bfc2f48b2c40e5b3
|
|
|
|
|
| |
Change-Id: I503597eb90961f762c3f699015a0e42a0222d6fa
Auto-generated-cl: translation import
|
|
|
|
|
| |
Change-Id: Ibcb3dafb5d326790ca66cc351648104cfba486c4
Auto-generated-cl: translation import
|
|
|
|
|
| |
Bug: 11412863
Change-Id: Ia377c1a61a5e9339173c9877cb78a6d6873c5ab0
|
|
|
|
|
| |
Change-Id: Ia9a5e1d4d928ceff2b3bcac536c5745eca573597
Auto-generated-cl: translation import
|
|
|
|
|
| |
Bug: 11412863
Change-Id: Ia6619d96e4652816f173631a4a084b2ee8b2533a
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Following locales have been replaced by locale+country code
et -> et-rEE
ms -> ms-rMY
km -> km-rKH
lo -> lo-rLA
hy -> hy-rAM
ka -> ka-rGE
mn -> mn-rMN
Bug: 11412863
Change-Id: I3ca4f564ef6dfcad562329171238b58871f280c3
|
|
|
|
|
| |
Change-Id: I505fb183ef41641fe7c6c40f9bb31a3276f8381b
Auto-generated-cl: translation import
|
|
|
|
|
| |
Change-Id: I009be1df92820ccc7471a68720e87ec7aaf6bdb2
Auto-generated-cl: translation import
|
|
|
|
|
| |
Change-Id: I383364c5c4c5945496ddd5fbe3cf376430d50043
Auto-generated-cl: translation import
|
|
|
|
|
| |
Change-Id: Id6811ddabfaf683b62b4a55074421c7c66972d39
Auto-generated-cl: translation import
|
|
|
|
|
| |
Change-Id: I7a090f2436d7dbec310711f6c4a54aff0e7fc965
Auto-generated-cl: translation import
|
|
|
|
|
|
| |
It doesn't build if llvm-rs-cc is not available.
Change-Id: Ib9b970af9b712d25f08c6a13aeae32c98d0a8cf8
|
|
|
|
|
| |
Change-Id: I2e8d65682d5a198cb6e7c199c4f96369bc618ed9
Auto-generated-cl: translation import
|
|
|
|
| |
Change-Id: I6ed4fd0b153689bf163d82c4b30879c2a1fb52c5
|
|
|
|
| |
Change-Id: I43380ab215b529e1f23c76fb58f2ff1fbfeee180
|
|
|
|
| |
Change-Id: I880296bb76b6a1d7226541b7587ae650182dd75d
|
|
|
|
|
|
|
|
| |
Code follows the java implementation as closely as possible to make
comparisons and future bug fixes easy.
Bug: 5613515
Change-Id: I14db2f2d659be919b4b12e5a7fe53af7ec137467
|
|
|
|
| |
Change-Id: I958131674c746b90d62b6e428b65e554983c52f6
|
|
|
|
| |
Change-Id: I85294e37b07eb6f024f86a0ddd9149085a01d9d6
|
|
|
|
| |
Change-Id: I10541172d99e1ede5b27bfcef106afadc0e540d4
|
|
|
|
| |
Change-Id: Iaf1b05b79b8628a1453fbb94e04eee3fbb72fe8b
|
|
|
|
| |
Change-Id: I67796f3c85ece30d3ad06b33eed63d7e85ad8eba
|
|
|
|
| |
Change-Id: I4eded0d405b56903aa46596f8c688d99d8c11abe
|
|
|
|
|
|
|
|
| |
This causes issues with our new header file structure for Renderscript,
since the build system is using the modern llvm-rs-cc with preexisting
header files from before ICS.
Change-Id: Ie20c7ee445451d2dcc12f7082f03d7204d1e8bf3
|
|
|
|
| |
Change-Id: I95527858918366b2fde54a71b0b831dda0551172
|
|
|
|
|
|
|
|
|
|
|
|
| |
The count was wrong for the number of items in the allocation pool since we
now prefetch more data.
Add demo code to show dialogs for events generated by Carousel
(select texture, select detail, long press).
Some other minor code cleanups.
Change-Id: Iebecd1945c115d96bd60a3968985c1b104b586d6
|
|
|
|
| |
Change-Id: Iee58e5260654883beece74a04fc5f3a2eae329cb
|
|
|
|
| |
Change-Id: I2299266b21b4d7ded2501f87abe90e11e8ec22ad
|
|
|
|
| |
Change-Id: Ieb21634a68d8f5b9f1b411c547c18e32349c4d9e
|
|
|
|
|
|
|
|
| |
This changes the behavior to avoid snapping back to the original
size, which was a very jarring effect. Instead, Carousel now has
a scale-down animation that is triggered when the user releases the item.
Change-Id: I81a91e65d0da991a98211f2ecb92fa89c8c81e22
|
|
|
|
|
|
| |
This fixes an issue where book shadows disappeared too quickly.
Change-Id: Idc8ef5af2c334fa4eb3f0514e79a25fd82f5cae8
|
|\ |
|
| |
| |
| |
| |
| |
| |
| |
| | |
Previously, prefetched cards were considered visible,
which is inefficient when the number of cards to prefetch is
large, and can confuse the hit detection logic (see b3468797)
Change-Id: I91e150c52b88608eb8627edc8cce30ecfb44e672
|
|/
|
|
|
|
|
|
|
|
|
| |
Before this change, Carousel sometimes showed chaotic movement
when a long history of motion was captured because of the
infinite history of velocities used to calculate the average.
The code now limits the number of historical velocities to a
small number, which results in much more predictable motion.
Change-Id: I4edadbde5464749646dcce50495e54b3abb73901
|
|\ |
|
| |
| |
| |
| | |
Change-Id: I4717952f5fe6b5d99699e55d93abcd1cb1fc1986
|
|/
|
|
|
|
| |
Remove use of private depricated APIs.
Change-Id: I1f1fd0046bf5c9680b35c428afa40f5e4e19e222
|
|
|
|
|
| |
Bug: 3404464
Change-Id: Ife6b29241e47727bc980753d0233da3a4a945779
|
|\ |
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
When shrinking the cards array while invalidates are still
in flight, it's possible to have an invalidate or texture
set land on a nonexistent card, causing an out-of-bounds
exception to be thrown, crashing the app. Frequently these
invalidates come from the user dragging the carousel to
another position, so it's not feasible to prevent
invalidates from being sent when we're about to clear the
cards array; nor is it feasible to empty any pending
invalidates, since Renderscript does not provide any
meaningful inspection of the event queue. Therefore, the
only real way we can address the problem is by detecting
these out-of-turn invalidates and handling the errors
cleanly. In practice, since these events are typically to
invalidate a nonexistent card, it's OK to just drop them,
since each such card is already invalid because it does
not exist.
This fix factors out card-get and card-set logic into
common methods, similar to card-set-or-create logic that's
been present for a long time.
Bug: 3381300
Change-Id: I037801de3b4c3bff514b1586fa16417e32c118df
|
|/
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
When invalidateTexture() is told to continue drawing the
current texture until a replacement is found (STATE_STALE),
but the texture was not there in the first place since it
hadn't loaded yet, later attempts to draw the texture based
on the assumption of its being there (by virtue of being in
STATE_STALE or STATE_UPDATING) would result in rendering
problems due to trying to bind a nonexistent texture. The
fix is to make invalidateTexture() only set STATE_STALE if
a valid texture had been there in the first place.
As an added precautionary measure, this change also has a
stopgap solution that will draw the placeholder image if
the state machine indicates the texture is valid but it is
not actually valid. This case should never be encountered.
Bug: 3356101
Change-Id: Ia87e9e651f8ac3138faced3cf0f6199c58ee51e0
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Previously we were arranging cards on the screen using the
individual card's matrix. When the card matrices are
different, however, this leads to an inconsistent layout.
Instead, use the default card matrix for layout, while
using the individaul card matrix (which, when not set
separately, is the same as the default card matrix) for
handling drawing the individual card.
Bug: 3365044
Change-Id: Ibb7ad7760131d2c649cc0882b784c9d2e73a737f
|
|
|
|
| |
Change-Id: I4151f20458b72f339bf636f7324b8dbbe33e6635
|
|
|
|
| |
Change-Id: Ib08f4f7997d5e06891d5ce09cc7bff6c0f61f341
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Set a per-card matrix through the newly expanded
TextureParameters mechanism.
This also changes the behavior of the card matrix so that
it is applied locally to the card, rather than being
applied to the whole carousel. And the detail texture is
no longer drawn with the card matrix.
Bug: 3206251
Change-Id: I5119bc97b136d32cf9d4e29e7aa3751334cc68c1
|