Adds a test that verifies that the view of the local time is the same in
the Dart isolate and the process that is running the test.
Specifically, this test is useful to verify that the code paths for
timezone retrieval do not break while the underlying FIDL protocols are
being refactored.
However, since the check is generally useful, the test is written as a
general flutter test.
Running this on Fuchsia required adding `fuchsia.intl.ProfileProvider`
to the CMX file that is used for all build Fuchsia packages.
Testing is a bit involved on Fuchsia. You must build the Fuchsia
package `fluter/shell/common:shell_tests` and publish it to the dev
repository for your Fuchsia device. It seems that a way to do so is to
modify the script `flutter/tools/fuchsia/build_fuchsia_artifacts.py` and
modify its function `GetTargetsToBuild` like so:
```
def GetTargetsToBuild(product=False):
targets_to_build = [
'flutter/shell/platform/fuchsia:fuchsia',
'flutter/shell/common:shell_tests',
]
return targets_to_build
```
Next, the Fuchsia packages need to be compiled and published.
Once done, the following `fx` invocation will run the test, assuming
that you have your Fuchsia setup:
```
fx shell run \
fuchsia-pkg://fuchsia.com/shell_tests#meta/shell_tests.cmx \
-- --gtest_filter=ShellTest.LocaltimesMatch
```
Adds an `allowUpscale` parameter to the Dart interfaces for instantiateImageCodec and decodeImageFromPIxels. If set to false, the image bounds will be clamped at their intrinsic dimensions.
A later commit will change the default to true.
* Hook up PersistentCache to ShellTestPlatformViewVulkan
* [vulkan][fuchsia] enable vulkan without swiftshader in fuchsia tests on arm64
* [vulkan][fuchsia] Disable second half of ShellTest.CacheSkSLWorks on
vulkan backend
GrContext::precompileShader() is not implemented for vulkan contexts, so
dont run the portion of this test that depends on that behavior on
Vulkan.
Co-authored-by: George Wright <gw280@google.com>
SkImage references get held by both our Image and Picture objects. The GC has no idea about this, and so sometimes pictures look very small when they're actually holding a reference to a large chunk of memory. This patch helps make sure that the GC can more adequately catch the real size impact of Picture objects, and combined with an upstream patch in Dart allows for much more aggressive collection of unused image related memory.
Added a new `FlutterEngineAOTData` argument to `FlutterProjectArgs`. Embedders can instantiate and destroy this object via the new `FlutterEngineCreateAOTData` and `FlutterEngineCollectAOTData` methods provided.
If an embedder provides more than one source of AOT data to `FlutterEngineInitialize` or `FlutterEngineRun` (e.g. snapshots as well as `FlutterEngineAOTData`), the engine will error out.
Resolves: https://github.com/flutter/flutter/issues/50778
This reverts commit 74823c212d418597775d332d8c272673c83f6f63 and relands our reland https://github.com/flutter/engine/pull/17915.
Additionally, we fixed the cull rect logic in `OpacityLayer::Preroll` which is the root cause of https://github.com/flutter/flutter/issues/56298. We've always had that root problem before but it did not trigger performance issues because we were using the OpacityLayer's `paint_bounds`, instead of its child's `paint_bounds` for preparing the layer raster cache. A correct handling of the cull rect should allow us to cull at any level.
It also turns out that our ios32 (iPhone4s) performacne can regress a lot
without snapping. My theory is that although the picture has a
fractional top left corner, many drawing operations inside the picture
have integral coordinations. In older hardwares, keeping those
coordinates integral seems to be performance critical.
To avoid flutter/flutter#41654, the snapping
will still be disabled if the matrix has non-scale-translation
transformations.
This reverts commit c72ff4a and relands #17712.
Fixesflutter/flutter#53288 and flutter/flutter#41654.
Together with #17791, this reland addresses some of Jim's concerns in the original PR #17712.
The major part of this PR is still the same as the original PR, and the performance / golden image impacts should be the same.
* Bumped up the timeout for testAccessibilityFocusOnTextSemanticsProducesCorrectIosViews
* ran formatter
* made an alternative version of the assertion
* removed log statements, seems to mess with the buildbot's xcpretty
