Fixes scroll event bubbling when Flutter web is embedded in an
iframe(#156985).
When a scrollable handles a wheel event, it now calls
respond(allowPlatformDefault: false) to signal the engine.
In iframe mode, the engine always calls preventDefault() to block native
scroll chaining, then uses postMessage to explicitly scroll the parent
page only when all scrollables are at boundary
Before change:
https://issue-156985-before.web.app/
After change:
https://issue-156985-after.web.app/
This rewrites many of the include paths in Fuchsia engine code to use
the full path of the file. For example:
```diff
-#include "runtime/dart/utils/inlines.h"
+#include "flutter/shell/platform/fuchsia/runtime/dart/utils/inlines.h"
```
I also cleaned up a few unused includes and convered the includes for
the Fuchsia SDK to using angle backets.
## Pre-launch Checklist
- [X] I read the [Contributor Guide] and followed the process outlined
there for submitting PRs.
- [X] I read the [Tree Hygiene] wiki page, which explains my
responsibilities.
- [X] I read and followed the [Flutter Style Guide], including [Features
we expect every widget to implement].
- [X] I signed the [CLA].
- [ ] I listed at least one issue that this PR fixes in the description
above.
- [X] I updated/added relevant documentation (doc comments with `///`).
- [X] I added new tests to check the change I am making, or this PR is
[test-exempt].
- [X] I followed the [breaking change policy] and added [Data Driven
Fixes] where supported.
- [ ] All existing and new tests are passing.
Work towards https://github.com/dart-lang/sdk/issues/58729
The lint rule will suggest using a super-parameter for a named parameter
even if the positional parameters cannot be super-parameters.
## Description
Fixes https://github.com/flutter/flutter/issues/181967
`WrappedTextureSourceVK` in the embedder delegate path creates a
`VkImageView` every frame via `createImageView()` but never destroys it.
The handle is stored as a raw non-owning `vk::ImageView` and the
destructor is empty, leaking one `VkImageView` per frame (~60/sec on a
typical app).
This switches to `vk::UniqueImageView` for RAII ownership so the image
view is destroyed when the `WrappedTextureSourceVK` is released.
## Why this is safe
The resource lifetime is correct because Impeller's `TrackedObjectsVK`
holds a `shared_ptr` to the `TextureSourceVK` until the GPU fence
signals completion:
1. `RenderPassVK` calls `Track(attachment.texture)` on all render target
textures
2. `CommandBufferVK::Track()` stores `shared_ptr<TextureSourceVK>` in
`TrackedObjectsVK`
3. `TrackedObjectsVK` is cleared only in the fence completion callback,
after the GPU finishes
The `VkImageView` cannot be destroyed while the GPU is still using it.
## Scope
This only affects the **embedder delegate code path** (when `delegate_
!= nullptr` in `GPUSurfaceVulkanImpeller::AcquireFrame`). Android passes
`nullptr` as the delegate and uses `AcquireNextSurface()` instead, so
this change has **no effect on Android, iOS, or standard swapchain
rendering**.
## Pre-launch Checklist
- [x] I read the [Contributor Guide] and followed the process outlined
there for submitting PRs.
- [x] I read the [Tree Hygiene] wiki page, which explains my
responsibilities.
- [x] I read and followed the [Flutter Style Guide].
- [x] I listed at least one issue that this PR fixes in the description
above.
- [ ] I added a test for the change if applicable. *(See note below)*
**Note on testing:** This is a resource leak fix in a Vulkan code path
that requires a running Vulkan device and embedder delegate. I was
unable to find existing test infrastructure for the embedder delegate
path in `GPUSurfaceVulkanImpeller`. Happy to add a test if reviewers can
point me to the right test harness.
Found while working on Impeller Vulkan support for Linux/Wayland with a
custom embedder.
[Contributor Guide]:
https://github.com/flutter/flutter/blob/master/docs/contributing/Tree-hygiene.md#overview
[Tree Hygiene]:
https://github.com/flutter/flutter/blob/master/docs/contributing/Tree-hygiene.md
[Flutter Style Guide]:
https://github.com/flutter/flutter/blob/master/docs/contributing/Style-guide-for-Flutter-repo.md
Adds a bunch of tests for uniform setting functionality for custom
fragment shaders on the web.
Deletes redundant tests.
Fixes a discovered issue in the uniform offset calculation. We were
previously using the `location`, which is the integer offset of the
uniform, not the offset in floats.
Make sure the resize listener is not added if content sizing is not
enabled. This should ensure that the resize listener is not causing the
flake. While this won't fix the flake if content sizing is enabled -
this test was passing before content sizing was checked in therefore
this should remove any side effect content sizing is having on the test.
Fixes: #179673
## Pre-launch Checklist
- [X] I read the [Contributor Guide] and followed the process outlined
there for submitting PRs.
- [X] I read the [Tree Hygiene] wiki page, which explains my
responsibilities.
- [X] I read and followed the [Flutter Style Guide], including [Features
we expect every widget to implement].
- [X] I signed the [CLA].
- [X] I listed at least one issue that this PR fixes in the description
above.
- [X] I updated/added relevant documentation (doc comments with `///`).
- [X] I added new tests to check the change I am making, or this PR is
[test-exempt].
- [X] I followed the [breaking change policy] and added [Data Driven
Fixes] where supported.
- [X] All existing and new tests are passing.
#161546 changed the stripped option with the intent to be true for
android, and false for all other targets, but the check never worked as
intended because parser store_true action default value is False so the
final else was never used.
This fixes the build size increase that was noticed on flutter 3.32 for
linux targets
Fixes#181983
## Pre-launch Checklist
- [x] I read the [Contributor Guide] and followed the process outlined
there for submitting PRs.
- [ ] I read the [Tree Hygiene] wiki page, which explains my
responsibilities.
- [ ] I read and followed the [Flutter Style Guide], including [Features
we expect every widget to implement].
- [x] I signed the [CLA].
- [x] I listed at least one issue that this PR fixes in the description
above.
- [x] I updated/added relevant documentation (doc comments with `///`).
- [ ] I added new tests to check the change I am making, or this PR is
[test-exempt].
- [ ] I followed the [breaking change policy] and added [Data Driven
Fixes] where supported.
- [ ] All existing and new tests are passing.
<!--
Thanks for filing a pull request!
Reviewers are typically assigned within a week of filing a request.
To learn more about code review, see our documentation on Tree Hygiene:
https://github.com/flutter/flutter/blob/main/docs/contributing/Tree-hygiene.md
-->
Optimize the loadFontFromList method of SkiaFontCollection. Reduce the
generation of unnecessary FontMgr.
issue: https://github.com/flutter/flutter/issues/181439
## Pre-launch Checklist
- [x] I read the [Contributor Guide] and followed the process outlined
there for submitting PRs.
- [x] I read the [Tree Hygiene] wiki page, which explains my
responsibilities.
- [x] I read and followed the [Flutter Style Guide], including [Features
we expect every widget to implement].
- [x] I signed the [CLA].
- [x] I listed at least one issue that this PR fixes in the description
above.
- [ ] I updated/added relevant documentation (doc comments with `///`).
- [ ] I added new tests to check the change I am making, or this PR is
[test-exempt].
- [ ] I followed the [breaking change policy] and added [Data Driven
Fixes] where supported.
- [x] All existing and new tests are passing.
Co-authored-by: Mouad Debbar <mdebbar@google.com>
Adds wide gamut color support to macOS (matching iOS), upgrades the
surface pixel format from 10-bit BGRA10_XR to 16-bit float RGBA16Float
on both iOS and macOS when enabled, and fixes Impeller's blur filter P3
clamping on macOS.
**macOS Wide Gamut Support**
- Added DoesHardwareSupportWideGamut() hardware capability check
(MTLGPUFamilyApple2 or MTLGPUFamilyMac2)
- Wide gamut enabled when both hardware supports it and
FLTEnableWideGamut plist flag is YES
- Dynamic wide gamut switching when windows move between P3 and sRGB
displays
- Added flutter/screenshot method channel on macOS for integration
testing
**RGBA16Float Surface Format (iOS + macOS)**
- macOS IOSurface: kCVPixelFormatType_64RGBAHalf +
MTLPixelFormatRGBA16Float
- iOS CAMetalLayer: MTLPixelFormatRGBA16Float
- Image decoder: always uses kRGBA_F16_SkColorType for all wide gamut
images (previously only transparent images used 16-bit)
**Fix Blur P3 Clamping on macOS**
macOS uses the compositor/embedder path, not GPUSurfaceMetalImpeller, so
UpdateOffscreenLayerPixelFormat was never called. Added the call in
embedder.cc after wrapping the Metal resolve texture.
**Why RGBA16Float over BGRA10_XR?**
BGRA10_XR has only 10 bits per channel — values outside sRGB gamut get
clamped in intermediate render targets (e.g. blur filters). RGBA16Float
has 16 bits per channel with full floating-point range, preventing P3
color clamping in multi-pass rendering.
**Tests**
- 9 new iOS FlutterView unit tests verifying RGBA16Float pixel format
and extended sRGB color space
- Updated macOS FlutterSurfaceManagerTest for RGBA16Float, dynamic
switching, color space, and pixel format verification
- Updated image decoder and Impeller display list tests for
kR16G16B16A16Float
- 11 macOS/iOS integration tests: image, saveLayer, codecImage, none,
blur, drawnImage, container, linearGradient, radialGradient,
conicalGradient, sweepGradient
### Issues
https://github.com/flutter/flutter/issues/164557
## Pre-launch Checklist
- [x] I read the [Contributor Guide] and followed the process outlined
there for submitting PRs.
- [x] I read the [Tree Hygiene] wiki page, which explains my
responsibilities.
- [x] I read and followed the [Flutter Style Guide], including [Features
we expect every widget to implement].
- [x] I signed the [CLA].
- [x] I listed at least one issue that this PR fixes in the description
above.
- [x] I updated/added relevant documentation (doc comments with `///`).
- [x] I added new tests to check the change I am making, or this PR is
[test-exempt].
- [ ] I followed the [breaking change policy] and added [Data Driven
Fixes] where supported.
- [ ] All existing and new tests are passing.
ASan had been disabled due to issues that were seen when the LUCI bots
were upgraded to Ubuntu 24. That upgrade was reverted, and ASan can be
run on the current Ubuntu 22 bots.
Fixes https://github.com/flutter/flutter/issues/181639
No bug
Gemini code review suggestion on a cp that should instead be made on
master.
https://github.com/flutter/flutter/pull/181732/files#r2747693178
## Pre-launch Checklist
- [x] I read the [Contributor Guide] and followed the process outlined
there for submitting PRs.
- [x] I read the [Tree Hygiene] wiki page, which explains my
responsibilities.
- [x] I read and followed the [Flutter Style Guide], including [Features
we expect every widget to implement].
- [x] I signed the [CLA].
- [ ] I listed at least one issue that this PR fixes in the description
above.
- [ ] I updated/added relevant documentation (doc comments with `///`).
- [ ] I added new tests to check the change I am making, or this PR is
[test-exempt].
- [x] I followed the [breaking change policy] and added [Data Driven
Fixes] where supported.
- [x] All existing and new tests are passing.
https://github.com/flutter/flutter/pull/181157 introduced a call to the
glDebugMessageCallback API using a lambda as the argument. On Windows
builds this produces a warning about a nonstandard implicit conversion
when the Impeller GL API wrappers try to log the callback argument.
<!--
Thanks for filing a pull request!
Reviewers are typically assigned within a week of filing a request.
To learn more about code review, see our documentation on Tree Hygiene:
https://github.com/flutter/flutter/blob/main/docs/contributing/Tree-hygiene.md
-->
### Problem
The P3-to-sRGB color conversion applies an affine matrix directly to
gamma-encoded values in both the C++ engine (`dl_color.cc`) and Dart
(`painting.dart`). This is mathematically incorrect — color space
conversion matrices must operate in linear light. The result is P3
colors that appear nearly identical to sRGB instead of showing the full
gamut difference.
For example, P3 `#1ECAD3` (30/255, 202/255, 211/255) converts to:
- **Old (wrong):** extendedSRGB (-0.12, 0.86, 0.87)
- **New (correct):** extendedSRGB (-0.38, 0.81, 0.84)
The red channel error of 0.26 is clearly visible — colors appear washed
out instead of vivid.
### Fix
Replace the single affine matrix multiply with the correct 3-step
pipeline:
1. **Linearize** — decode gamma via sRGB EOTF
2. **Transform** — apply 3x3 P3-to-sRGB matrix in linear space
3. **Encode** — re-apply gamma via sRGB OETF
Extended range (negative values from out-of-gamut colors) is handled by
mirroring the transfer function.
**C++ (`dl_color.cc`):** Replace affine matrix with `p3ToExtendedSrgb()`
using `double` precision.
**Dart (`painting.dart`):** Replace `_MatrixColorTransform` with
`_P3ToSrgbTransform` and `_SrgbToP3Transform` classes. Also fixes the
sRGB-to-P3 direction.
### Performance
Negligible. The C++ conversion runs once per paint setup in `ReadColor`,
not per frame or per pixel. The Dart conversion runs once per
`Color.withValues()` call.
### Tests
- **C++:** Added `ColorSpaceP3ToExtendedSRGBLinearLight` test in
`dl_color_unittests.cc`
- **Dart:** Added 3 tests in `colors_test.dart`: mid-range
P3→extendedSRGB, P3 green→extendedSRGB, sRGB→P3 round-trip
All new tests fail with the old code and pass with the fix. Existing
tests continue to pass.
## Issue:
https://github.com/flutter/flutter/issues/181717
## Pre-launch Checklist
- [x] I read the [Contributor Guide] and followed the process outlined
there for submitting PRs.
- [x] I read the [Tree Hygiene] wiki page, which explains my
responsibilities.
- [x] I read and followed the [Flutter Style Guide], including [Features
we expect every widget to implement].
- [x] I signed the [CLA].
- [x] I listed at least one issue that this PR fixes in the description
above.
- [x] I updated/added relevant documentation (doc comments with `///`).
- [x] I added new tests to check the change I am making, or this PR is
[test-exempt].
- [x] I followed the [breaking change policy] and added [Data Driven
Fixes] where supported.
- [ ] All existing and new tests are passing. **New tests pass, can't
run golden tests!**
---------
Co-authored-by: gaaclarke <30870216+gaaclarke@users.noreply.github.com>
While testing with a different add-to-app
https://github.com/flutter/samples/pull/2787, an exception is sometimes
thrown when the resize is attempted because the callback comes from a
different thread. It is expected that the raster thread calls this but
with previous
[testing](https://github.com/mboetger/test-add-to-app/tree/content-sizing)
never caused this exception.
This PR ensures the resize happens on the UI thread. It also a flag to
ensure content sizing is disabled.
Fixes: https://github.com/flutter/flutter/issues/181573
## Pre-launch Checklist
- [X] I read the [Contributor Guide] and followed the process outlined
there for submitting PRs.
- [X] I read the [Tree Hygiene] wiki page, which explains my
responsibilities.
- [X] I read and followed the [Flutter Style Guide], including [Features
we expect every widget to implement].
- [X] I signed the [CLA].
- [X] I listed at least one issue that this PR fixes in the description
above.
- [X] I updated/added relevant documentation (doc comments with `///`).
- [X] I added new tests to check the change I am making, or this PR is
[test-exempt].
- [X] I followed the [breaking change policy] and added [Data Driven
Fixes] where supported.
- [X] All existing and new tests are passing.
Add ccache support when building the engine using a custom toolchain.
Fixes#180736
## Pre-launch Checklist
- [x] I read the [Contributor Guide] and followed the process outlined
there for submitting PRs.
- [x] I read the [Tree Hygiene] wiki page, which explains my
responsibilities.
- [x] I read and followed the [Flutter Style Guide], including [Features
we expect every widget to implement].
- [x] I signed the [CLA].
- [x] I listed at least one issue that this PR fixes in the description
above.
Updates the GLFW dependency to master and uses the EGL context creation
API on Linux. This allows Linux users to select an OpenGL implementation
when running the impeller playground tests by defining
`__EGL_VENDOR_LIBRARY_FILENAMES`.
For example, running
```
$ impeller_unittests --gtest_filter=Play/AiksTest.ToImageFromImage/OpenGLES
```
gives, in `description_gles.cc` ,
```
gl_version_string_ // OpenGL ES 3.2 NVIDIA 590.48.01
```
Then running
```
$ __EGL_VENDOR_LIBRARY_FILENAMES=/usr/share/glvnd/egl_vendor.d/50_mesa.json impeller_unittests --gtest_filter=Play/AiksTest.ToImageFromImage/OpenGLES
```
```
gl_version_string_ // OpenGL ES 3.2 Mesa 25.3.3-arch1.3
```
Fixes#181258
## Pre-launch Checklist
- [x] I read the [Contributor Guide] and followed the process outlined
there for submitting PRs.
- [x] I read the [Tree Hygiene] wiki page, which explains my
responsibilities.
- [x] I read and followed the [Flutter Style Guide], including [Features
we expect every widget to implement].
- [x] I signed the [CLA].
- [x] I listed at least one issue that this PR fixes in the description
above.
- [x] I updated/added relevant documentation (doc comments with `///`).
- [x] I added new tests to check the change I am making, or this PR is
[test-exempt].
- [x] I followed the [breaking change policy] and added [Data Driven
Fixes] where supported.
- [x] All existing and new tests are passing.
Adds support for backtraces when requested in host builds
### Before:
```
Note: Google Test filter = Play/RendererTest.BabysFirstTriangle/OpenGLES
[==========] Running 1 test from 1 test suite.
[----------] Global test environment set-up.
[----------] 1 test from Play/RendererTest
[ RUN ] Play/RendererTest.BabysFirstTriangle/OpenGLES
...
Segmentation fault (core dumped) __EGL_VENDOR_LIBRARY_FILENAMES=/usr/share/glvnd/egl_vendor.d/50_mesa.json ./out/host_debug_unopt/exe.unstripped/impeller_unittests --enable_playground --gtest_filter='Play/RendererTest.BabysFirstTriangle/OpenGLES'
```
### After:
```
Note: Google Test filter = Play/RendererTest.BabysFirstTriangle/OpenGLES
[==========] Running 1 test from 1 test suite.
[----------] Global test environment set-up.
[----------] 1 test from Play/RendererTest
[ RUN ] Play/RendererTest.BabysFirstTriangle/OpenGLES
...
[symbolize_elf.inc : 379] RAW: Unable to get high fd: rc=0, limit=1024
[ERROR:flutter/fml/backtrace.cc(108)] Caught signal SIGSEGV during program execution.
Frame 0: 0x55831f19fbc4 impeller::DeviceBufferGLES::GetBufferData()
Frame 1: 0x55831f186cd3 impeller::BufferBindingsGLES::BindUniformBufferV2()
Frame 2: 0x55831f186b66 impeller::BufferBindingsGLES::BindUniformBufferV3()
Frame 3: 0x55831f1861da impeller::BufferBindingsGLES::BindUniformBuffer()
Frame 4: 0x55831f185fea impeller::BufferBindingsGLES::BindUniformData()
Frame 5: 0x55831f1cb112 impeller::EncodeCommandsInReactor()
...
Segmentation fault (core dumped) __EGL_VENDOR_LIBRARY_FILENAMES=/usr/share/glvnd/egl_vendor.d/50_mesa.json ./out/host_debug_unopt/exe.unstripped/impeller_unittests --enable_playground --gtest_filter='Play/RendererTest.BabysFirstTriangle/OpenGLES'
```
Fixes#181156
## Pre-launch Checklist
- [x] I read the [Contributor Guide] and followed the process outlined
there for submitting PRs.
- [x] I read the [Tree Hygiene] wiki page, which explains my
responsibilities.
- [x] I read and followed the [Flutter Style Guide], including [Features
we expect every widget to implement].
- [x] I signed the [CLA].
- [x] I listed at least one issue that this PR fixes in the description
above.
- [x] I updated/added relevant documentation (doc comments with `///`).
- [x] I added new tests to check the change I am making, or this PR is
[test-exempt].
- [x] I followed the [breaking change policy] and added [Data Driven
Fixes] where supported.
- [x] All existing and new tests are passing.
<!--
Thanks for filing a pull request!
Reviewers are typically assigned within a week of filing a request.
To learn more about code review, see our documentation on Tree Hygiene:
https://github.com/flutter/flutter/blob/main/docs/contributing/Tree-hygiene.md
-->
Enables improved OpenGLES debug support in the playground tests if
[KHR_DEBUG](https://registry.khronos.org/OpenGL/extensions/KHR/KHR_debug.txt)
is available. Only activated on unoptimized builds (where `NDEBUG` is
not defined)
Example:
Before:
```
[FATAL:flutter/impeller/renderer/backend/gles/proc_table_gles.h(44)] Fatal GL Error GL_INVALID_ENUM(1280) encountered on call to glBindBuffer
```
After:
```
[ERROR:flutter/impeller/playground/backend/gles/playground_impl_gles.cc(146)] GL Error: GL_INVALID_ENUM error generated. Invalid buffer target enum.
[FATAL:flutter/impeller/renderer/backend/gles/proc_table_gles.h(44)] Fatal GL Error GL_INVALID_ENUM(1280) encountered on call to glBindBuffer
```
Closes#179329
## Pre-launch Checklist
- [x] I read the [Contributor Guide] and followed the process outlined
there for submitting PRs.
- [x] I read the [Tree Hygiene] wiki page, which explains my
responsibilities.
- [x] I read and followed the [Flutter Style Guide], including [Features
we expect every widget to implement].
- [x] I signed the [CLA].
- [x] I listed at least one issue that this PR fixes in the description
above.
- [x] I updated/added relevant documentation (doc comments with `///`).
- [x] I added new tests to check the change I am making, or this PR is
[test-exempt].
- [x] I followed the [breaking change policy] and added [Data Driven
Fixes] where supported.
- [x] All existing and new tests are passing.
RenderPass holds a list of BufferViews that may hold raw pointers to
DeviceBuffers. HostBuffers and the underlying DeviceBuffers must not be
deleted until the RenderPass is no longer using them.
See https://github.com/flutter/flutter/issues/181287
Fixes#178619
Uses defensive null check (?.) instead of null assertion (!) when
calling focusWithoutScroll() on focusedFormElement in
GloballyPositionedTextEditingStrategy.placeElement().
This prevents a crash that can occur due to a race condition where
hasAutofillGroup returns true but focusedFormElement is null by the time
it's accessed.
Does what it says on the tin!
This PR adds struct member information to the runtime flatbuffer format.
This allows dart code to introspect structs at runtime.
Also modifies the runtime_stage tests to verify formats for all
supported uniform types.
Bubbles up struct member information to dart, and uses that information
to grab struct members in `getUniformX` related functions. This is
necessary on Vulkan because all uniforms are packaged into a single
struct. Also re-enables tests
## Pre-launch Checklist
- [ ] I read the [Contributor Guide] and followed the process outlined
there for submitting PRs.
- [ ] I read the [Tree Hygiene] wiki page, which explains my
responsibilities.
- [ ] I read and followed the [Flutter Style Guide], including [Features
we expect every widget to implement].
- [ ] I signed the [CLA].
- [ ] I listed at least one issue that this PR fixes in the description
above.
- [ ] I updated/added relevant documentation (doc comments with `///`).
- [ ] I added new tests to check the change I am making, or this PR is
[test-exempt].
- [ ] I followed the [breaking change policy] and added [Data Driven
Fixes] where supported.
- [ ] All existing and new tests are passing.
Instead of using fake touch events.
Before this patch `FlutterViewController` sends two fake touch events
(down and up), at `(0, 0)` to the framework to signal that the status
bar is tapped on iOS. The scaffold widget and the cupertino page
scaffold widget set up gesture detectors to listen for these fake taps,
and scroll the "primary" scrollable container to the top in response.
This messaging mechanism is sometimes ambiguous, as the framework may
interpret that as a pair of regular pointer tap events (for instance in
#177992 the modal barrier claims the tap gesture and as a result the
modal barrier is dismissed by the fake touch events). This PR changes
that to communicate the status bar tap event via a new system channel,
and dispatch the events via `WidgetsBindingObserver`s in the framework.
See also
https://github.com/flutter/flutter/issues/177992#issuecomment-3499913265Fixes#177992, fixes#175606
It appears that UIKit also has access to the coordinates of the touch
events to determine which scrollable view(s?) to dispatch the scroll to
top event to.
```objc
* frame #0: 0x00000001032f6520 UIPlayground.debug.dylib`MyScrollViewController.scrollViewShouldScrollToTop(scrollView=0x0000000106014800) at UIScrollView.swift:13:3
frame #2: 0x00000001867c9300 UIKitCore`-[UIScrollView _scrollToTopIfPossible:] + 316
frame #3: 0x00000001867c9604 UIKitCore`-[UIScrollView _scrollToTopFromTouchAtScreenLocation:resultHandler:] + 40
frame #4: 0x0000000186299bbc UIKitCore`__71-[UIWindow _scrollToTopViewsUnderScreenPointIfNecessary:resultHandler:]_block_invoke.358 + 168
frame #5: 0x000000018629981c UIKitCore`-[UIWindow _scrollToTopViewsUnderScreenPointIfNecessary:resultHandler:] + 1212
frame #6: 0x000000018581ed8c UIKitCore`-[UIStatusBarManager _handleScrollToTopAtXPosition:] + 192
frame #7: 0x000000018581eb60 UIKitCore`-[UIStatusBarManager handleTapAction:] + 60
```
Unfortunately that information is not available to user application. The
iOS accessibility bridge currently does create dummy UIScrollViews for
each scrollable in the accessibility tree so may be we can take
advantage of that in the future.
## Pre-launch Checklist
- [ ] I read the [Contributor Guide] and followed the process outlined
there for submitting PRs.
- [ ] I read the [Tree Hygiene] wiki page, which explains my
responsibilities.
- [ ] I read and followed the [Flutter Style Guide], including [Features
we expect every widget to implement].
- [ ] I signed the [CLA].
- [ ] I listed at least one issue that this PR fixes in the description
above.
- [ ] I updated/added relevant documentation (doc comments with `///`).
- [ ] I added new tests to check the change I am making, or this PR is
[test-exempt].
- [ ] I followed the [breaking change policy] and added [Data Driven
Fixes] where supported.
- [ ] All existing and new tests are passing.
If you need help, consider asking for advice on the #hackers-new channel
on [Discord].
**Note**: The Flutter team is currently trialing the use of [Gemini Code
Assist for
GitHub](https://developers.google.com/gemini-code-assist/docs/review-github-code).
Comments from the `gemini-code-assist` bot should not be taken as
authoritative feedback from the Flutter team. If you find its comments
useful you can update your code accordingly, but if you are unsure or
disagree with the feedback, please feel free to wait for a Flutter team
member's review for guidance on which automated comments should be
addressed.
<!-- Links -->
[Contributor Guide]:
https://github.com/flutter/flutter/blob/main/docs/contributing/Tree-hygiene.md#overview
[Tree Hygiene]:
https://github.com/flutter/flutter/blob/main/docs/contributing/Tree-hygiene.md
[test-exempt]:
https://github.com/flutter/flutter/blob/main/docs/contributing/Tree-hygiene.md#tests
[Flutter Style Guide]:
https://github.com/flutter/flutter/blob/main/docs/contributing/Style-guide-for-Flutter-repo.md
[Features we expect every widget to implement]:
https://github.com/flutter/flutter/blob/main/docs/contributing/Style-guide-for-Flutter-repo.md#features-we-expect-every-widget-to-implement
[CLA]: https://cla.developers.google.com/
[flutter/tests]: https://github.com/flutter/tests
[breaking change policy]:
https://github.com/flutter/flutter/blob/main/docs/contributing/Tree-hygiene.md#handling-breaking-changes
[Discord]:
https://github.com/flutter/flutter/blob/main/docs/contributing/Chat.md
[Data Driven Fixes]:
https://github.com/flutter/flutter/blob/main/docs/contributing/Data-driven-Fixes.md
