Make ImageFilter objects comparable and printable.
This will help in areas in the Widget and RenderObject trees which try to avoid marking objects for updates if a setter is called with the same value (previously all ImageFilter objects would compare as not equal and appear to be new values).
This will improve font matching in SkParagraph for fonts that are bundled as
assets within the app. Libtxt was using Minikin's FontFamily class to select
the closest matching font, but SkParagraph will rely on the matchStyle
implementation in the asset font manager.
This exposes the line boundary information a line by adding getLineBoundary to return the indices corresponding to a line around a TextPosition. The information is already calculated when calculating line metrics, so that we can enable moving the selection/cursor to the beginning/end of a line in a text field.
Converting the argument to Paragraph.wordBoundary to dynamic temporarily until the framework code is converted to send a TextPosition instead of an int.
I'll submit this, then update the framework side to send a TextPosition, and expect a TextRange
or a List<int>, and then submit that, then I'll change this code to send a TextRange and take a TextPostion only, removing the dynamic here. Once that's done, I'll remove the code in the framework that expects a TextRange or a List<int>, and have it just expect a TextRange.
This is so that we can change the API without breaking the builds.
Landing on red to kick the engine builds.
This removes TextRange from the framework and moves it to the engine, in preparation for using it to return text ranges from the text extent APIs, like Paragraph.getWordBoundary instead of a List<int>.
Also added new tests for TextRange.
Obtaining the SkiaUnrefQueue through the IOManager is unsafe because
UIDartState has a weak pointer to the IOManager that can not be dereferenced
on the UI thread.
This re-enables thread safety checks for WeakPtr. WeakPtrs can't be used on a thread other than the one the WeakPtrFactory was created on.
This fixes the unit tests and adds a getUnsafe() method to WeakPtr to work around the remaining unresolved locations where we are using WeakPtr unsafely.
Since this is currently only meant to be used by the embedding internally, the setter in Objective-C is only exposed via the FlutterDartProject private class extension. Unit tests have been added to the shell_unittests harness.
Fixes https://github.com/flutter/flutter/issues/37641
This adds an isFocusable to SemanticsFlag so that the framework can tell the engine what semantics nodes are allowed to be focused, which will affect what platform flags are applied to the semantics information.
This flag is not yet in use by the frame
This reverts commit fcc4ab32301396986dd5103d6d444bff35fe0f63.
Fixes https://github.com/flutter/flutter/issues/41394 and other
related correctness issues.
TBR: @arbreng @jason-simmons @mehmetf
On Fuchsia, add a build flag for compositing OpacityLayers using the system
compositor vs Skia, which exposes a fastpath for opacity via Scenic.
This will only work under certain circumstances, in particular nested
OpacityLayers will not render correctly!
On Fuchsia, add a build flag for compositing PhysicalShapeLayers using
the system compositor vs Skia. Set to off by default, which restores
performant shadows on Fuchsia.
Remove the opacity exposed from ChildView, as that was added mistakenly.
Finally, we centralize the logic for switching between the
system-composited and in-process-composited paths inside of
ContainerLayer. We also centralize the logic for computing elevation
there. This allows the removal of many OS_FUCHSIA-specific code-paths.
Test: Ran workstation on Fuchsia; benchmarked before and after
Bug: 23711
Bug: 24163
* Fix broken tests
The earlier design speculated that embedders could affect the same
transformations on the layers post engine compositor presentation but before
final composition.
However, the linked issue points out that this design is not suitable for use
with hardware overlay planes. When rendering to the same, to affect the
transformation before composition, embedders would have to render to an
off-screen render target and then apply the transformation before presentation.
This patch negates the need for that off-screen render pass.
To be clear, the previous architecture is still fully viable. Embedders still
have full control over layer transformations before composition. This is an
optimization for the hardware overlay planes use-case.
Fixes b/139758641
This was required by the C++ spec pre-C++11 and still gets picked up in
the Fuchsia tree due to -Wnewline-eof. It fixes:
../../third_party/flutter/lib/ui/text/line_metrics.h:75:47: error: no newline at end of file [-Werror,-Wnewline-eof]
#endif // FLUTTER_LIB_UI_TEXT_LINE_METRICS_H_
Without this, developers have to override `onReportTimings` to listen for `FrameTiming`.
That can potentially break previous `onReportTimings` listeners if they forget to call
the old listener in their new callback.
This PR replaces the similar RP in the framework: https://github.com/flutter/flutter/pull/38574
Once this PR landed, we'll have to create another framework PR to use the stream to replace
`onReportTimings` usages.
Once that's done, we can then propose the breaking change of removing the deprecated
`onReportTimings`.
