Ensure log_clusters array is big enough.
Bug: 5714171
Explicitly handle the cases where the entire string or a single
run might have a length of 0.
Harfbuzz assumes the length of the item is at least 1.
If the length is zero, then it will clobber memory at index -1
into the log_clusters array.
Bug: 5705479
Change-Id: If28a9866221081f69973c1d12d7fe0cf8db2edd0
We were not passing the length of the UTF-16 string to
String16::setTo. As a result, it was copying the contents of
the text up to the first null it found.
First problem, these strings are not typically null terminated!
Second problem, if the string contained a null character, then
we might truncate it. However, we only truncated the string
when the copy constructor was invoked (say, when we called
get() on the cache) but not in internalTextCopy() (before
adding the key to the cache).
As a result of the second problem, we would first search
the cache for a key that matched a partially copied truncated
string (potentially reading uninitialized memory that followed it).
Finding none, we would add the entry to the cache using
the correct key.
If the cache already had a value associated with the correct key,
then the put would fail, returning false. Charging ever onwards,
we would add the size of the entry to the cache size.
Proceeding in this manner, it was possible for the cache to
believe it had less remaining space than it really did. At that
point, it was possible for the cache to evict all entries and
yet still not think it had room to add a new one, so it would
continue trying to make space indefinitely.
Bug: 5576812
Change-Id: I05251594f6b2da0a5dc09f7200f04fe9100ec766
* commit 'bcf05a69090f342d328f1537d1d83406b883290b':
Fix bug #5553401 TextLayoutCache is too verbose: "computeValuesWithHarfbuzz -- need to force to single run"
- welcome back start / count
- goodbye log clusters
- clean Paint code
- make private some functions as they should be
- improve memory allocation (create only one Shaper and reuse it for for shaping the runs in
the same input text)
Change-Id: I89a320c7f041319851308c8c9a919fbeafa82cdd
- the gain is about 5% and the timing is more stable
- use compare_type() and strictly_order_type()
Change-Id: Iab81869a8ba461ce786a468b6c59b8f34e8db838
- makes TextLayoutCache not carring about start/count. Basically he will cache the result for
the full string and gives back the "chunk" corresponding to start/count
- changed the TextLayoutCacheValue API to take start/count parameters
- add Harfbuzz LogClusters in TextLayoutCacheValue as it is needed for extracting the start/count "chunk"
Change-Id: I4b38a4442428606de9a093303bbbe98181e1f89c
- use vector (instead of array) for advances and glyphs
- reverse glyphs directly in computeRunValuesWithHarfbuzz() (instead of reversing them after)
Change-Id: I716a8f914fd043818d7cb80cca76ee5fb0effb96
- need a copy constructor for the key as the GenerationCache we are using
is actually a KeyedVector<K, sp<Entry<K, V> > >
- use the getText() API to access the text in the cache key
Change-Id: I5b60ebc062b62308ed7ac1284cfe2a9f28e2b8b1
- makes TextLayoutCache not carring about start/count. Basically he will cache the result for
the full string and gives back the "chunk" corresponding to start/count
- changed the TextLayoutCacheValue API to take start/count parameters
- added the Harfbuzz LogClusters in TextLayoutCacheValue as it is needed for extracting the start/count "chunk"
- fix potential issue of cache key leaking
Change-Id: I9276f9bec744e8de36349acfba8429f7c6f83394
- revert back to old working code where the script was setup for Harfbuzz\
depending of the direction of the run
Change-Id: I7ec740732b51ccf05b6744b7f9d2fcb35555478d
This removes the ParcelSurfaceTexture class since that functionality has been
folded into Surface.java. The change also updates the MediaPlayer to get rid
of setParcelSurfaceTexture() and modifies setTexture() to use the new Surface
functionality in order to simplify the code.
Change-Id: Iafa75ea3188263928128325d8a726786971b4de4
- better check parameters consistency
- return calls as soon as possible (when null or empty text)
Change-Id: I46744e517b04e3fba0ec37132d7de400177f214b
The Harfbuzz script setting was wired to HB_Script_Common for left-
to-right text and HB_Script_Arabic for right-to-left. This change
selects from additional scripts using some utility APIs from
Harfbuzz.
Change-Id: I34a6f1e0407c8d122bc968443948e7863d1f91ed
If the target is smaller on a given axis than the source then we
just draw a downscaled version of the NinePatch. If we use the
current path and the source has transparency then areas of the
NinePatch overlap and are blended together resulting in visual
inconsistancy.
bug: 5041053
Change-Id: I0fcc6fb5c214b188a164acf0651aa4ab2f35946d
- optimization for single run case was broken
- pass isRTL boolean along the call stack instead of the dirFlags integer
(which was only used as a "isRTL" in the shaper)
- update unit tests
Change-Id: I33110b76a433633a0b92fbd1db03785204e0c3e6
This change alters the conditions under which the onFrameAvailable
callback gets called by the C++ SurfaceTexture class. The new behavior
is to call the callback whenever a frame gets queued that will be
visible to the buffer consumer. This means that buffers queued in
synchronous mode always trigger the callback, as those buffers will
remain pending until they are consumed. Buffers queued in asynchronous
mode will only trigger the callback if there was not previously an
unconsumed buffer pending.
The new behavior means that a consumer should perform a draw operation
exactly once for every onFrameAvailable call that it recieves. This
change also modifies SurfaceFlinger and the SurfaceTexture JNI to
support of the new behavior.
Change-Id: I8b2c6e00961d3d58b11c6af50b555b6e4c5f5b40
