Turns out just supporting the right value for `kDebugMode` was a lot simpler than I thought. Debug builds used to never go through the build system code path when using `flutter run`, but now that we have wasm this can occur with the run command.
This should address https://github.com/flutter/flutter/issues/148850
## Description
This PR attemps to clarify the `InputDecoration.collapsed` documentation by explaining that it is not meant to be use with helper, label, counter, icons, prefixes and suffixes.
It also adds some parameters that make sense for a collapsed decoration (`hintMaxLines`, `hintFadeDuration`, `constraints`).
Removing parameters that should not have been added (`floatingLabelBehavior` and `floatingLabelAlignment`) will be part of another PR as it will require deprecations.
## Related Issue
Fixes https://github.com/flutter/flutter/issues/61331
## Tests
Adds 3 tests.
The current example:
> for example, when unlimited width is available and
/// you would like a child that would otherwise attempt to expand infinitely to
/// instead size itself to a more reasonable width.
doesn't seem to be the most useful example because most leaf widgets will just size themselves to their intrinsic size if the incoming constraints are not tight, so when
I was reading the doc I wasn't able to come up with any concrete widget combination that would require the use of `IntrinsicHeight`.
Updates the expected steps in the async function defined within `stepping_project.dart`.
The Dart web team is updating the async semantics of DDC to bring them in line with the other backends. Currently, the DDC async semantics don't adhere to the Dart spec and this can lead to inconsistent and surprising results.
However, the step-over operation doesn't work well yet with the new DDC async semantics. In the long run we intend to improve this but until then the debug stepper will have sporadic results that we can't model well with this test. When we are able to fix the stepper functionality, we will return this test to cover more of the async function being stepped over.
Upcoming changes to DDC change the async semantics of code produced by the compiler. The changes will bring the semantics more in line with those of dart2js and fix several bugs in the old semantics. However in landing those changes I experienced a test failure in [obscured_animated_image_test](https://github.com/flutter/flutter/blob/master/packages/flutter/test/widgets/obscured_animated_image_test.dart).
Some debugging uncovered that this is due to the use of `SynchronousFuture` in this `FakeCodec`. The old DDC async semantics forced an async gap when that future was [awaited](https://github.com/flutter/flutter/blob/master/packages/flutter/lib/src/painting/image_stream.dart#L1064). The new async semantics do not create an async gap here. This changes the render ordering of the widget tree created in the test leading to the test failure.
`Future.value` should be a reasonable substitute here and should achieve what the test was trying to achieve while also preserving the correct render ordering given the new DDC semantics.
This PR rewrites `CupertinoAlertDialog` in a cleaner logic, mostly its layout.
There are two major difficulties to lay out `CupertinoAlertDialog`:
* Laying out the actions section, which switches between a horizontal mode (two buttons in a row) and a vertical mode (several buttons in a column). This PR rewrites it in a special layout widget, `_CupertinoAlertActionSection`.
* Allocating vertical space between the content section and the actions section. This reuses `_PriorityColumn`, which was created for the action sheet.
In a similar fashion to the action sheet, the management and configuration for dividers and background (overscrolls) were rewritten as well.
This PR tries to keep as much original code and behavior as possible since this PR is already very large. As a result, almost no tests are broken. Further improvement will be done in future PRs.
* The test that verifies painting overscrolls is replaced by a golden test, since the original test assumes that the background is painted in one rectangle.
Fixes#150268
The issue was related to the check for selection geometry here: 22a5c6cb0a/packages/flutter/lib/src/widgets/selectable_region.dart (L2469-L2476) . Since `otherList == myList` is a reference check this would fail even if the selection rects inside the list contained in SelectionGeometry where the same causing the selectables inside the selection but outside the selectable containing the tapped position to have their selection cleared, use `listEquals` instead.
Changes:
- Introduced `alignmentOffset` property to `DropdownMenu` to allow adjustment of menu alignment.
- Updated the `MenuAnchor` widget to utilize the new `alignmentOffset` property.
- Default value for `alignmentOffset` is `Offset.zero`, meaning no additional spacing by default.
Motivation:
- This PR closes#151524
- @nate-thegrate, please let me know if I need to make changes in order for this PR to be merged
The motivation behind this change is to provide developers with more control over the spacing between the `MenuAnchor` and `dropdownMenuEntries` in the `DropdownMenu` widget. This enhancement allows for better alignment and customization of the dropdown menu appearance.
| before | after |
| :---: | :---: |
| <img width="372" alt="Screenshot 2024-07-14 at 8 03 14â¯PM" src="https://github.com/user-attachments/assets/4a45b843-7fa4-44fd-843c-c7209c6f57ae"> | <img width="364" alt="Screenshot 2024-07-14 at 8 03 27â¯PM" src="https://github.com/user-attachments/assets/12e8b857-aefb-4aaf-a310-4a002abcbc2f"> |
Initially, it was suggested to use a `SizedBox` to introduce spacing. However, upon further examination of the source code, it was discovered that the `DropdownMenuEntries` are rendered on the screen via an `OverlayPortal`. This necessitated leveraging the existing `alignmentOffset` property within the `MenuAnchor` for a more seamless and effective alignment adjustment.
https://github.com/flutter/flutter/issues/150800
> 2172 out-of-scope references in 138 files, 27 unique symbols were left unresolved.
Only did the most straightforward ones. There will be a part 2.
Reverts: flutter/flutter#152049
Initiated by: cbracken
Reason for reverting: iOS builds failing in post-submit
Original PR Author: loic-sharma
Reviewed By: {jmagman}
This change reverts the following previous change:
Changes:
1. Enables Swift Package Manager by default on the main/master channel
2. Fixes tests that fail if Swift Package Manager is enabled
Corresponding docs change: https://github.com/flutter/website/pull/10938
Addresses https://github.com/flutter/flutter/issues/151567
OverlayPortal attaches its overlaychild's renderobject to overlay directly while keeps its semantics tree under overlayportal.
This become a problem when the `overlaychild` markNeedsSemanticsUpdate that it propagate upward the rendering tree.
This means instead of marking dirty upward to the OverlayPortal, it directly mark Overlay dirty instead and skip `OverlayPortal`.
Currently this does not pose an issue other than unnecessary rebuilds, but it become a problem when I try to optimize the semantics tree compilation https://github.com/flutter/flutter/pull/150394.
After the optimization it won't rebuild semantics node unless it is marked dirty. Since the OverlayPortal widget does not get marked dirty, it won't update the subtree.
This changes the return type from `List<Decoration>` to `List<Decoration?>`, matching the corresponding setter.
As such it is a breaking change.
I believe the current type is a *bug*, and this is the correct fix.
The underlying value has a nullable element type, and the setter accepts a list which can contain `null`, but the getter tries to create a list with non-nullable elements from the stored value.
Calling this getter while the list contains `null` will throw.
(If this fix is too simplistic, I'll file a bug for the issue instead.)
https://github.com/flutter/flutter/issues/150800.
After applying the patch:
```
Can not fully resolve ../../flutter/packages/flutter/lib/src/material/material_state.dart: (MaterialState.pressed, MaterialState.focused, MaterialState.hovered, ..., value, update)
Can not fully resolve ../../flutter/packages/flutter/lib/src/material/material_state_mixin.dart: (MaterialStateProperty.resolve, MaterialState.disabled, MaterialState.dragged, ..., MaterialState.scrolledUnder, MaterialState.selected)
Can not fully resolve ../../flutter/packages/flutter/lib/src/material/reorderable_list.dart: (TargetPlatformVariant.desktop, TargetPlatformVariant.mobile)
Can not fully resolve ../../flutter/packages/flutter/lib/src/material/app.dart: (GlobalMaterialLocalizations)
Can not fully resolve ../../flutter/packages/flutter/lib/src/material/material_localizations.dart: (GlobalMaterialLocalizations)
57 out-of-scope references in 5 files, 22 unique symbols were left unresolved.
```
I'm going to leave `material_state` stuff alone because of https://github.com/flutter/flutter/issues/150800#issuecomment-2243923869,
`TargetPlatformVariant.desktop` requires `flutter_test` and `GlobalMaterialLocalizations` requires `package:flutter_localizations/flutter_localizations.dart`.
Adds an `includeSemantics` property to the Shortcuts widget.
@gspencergoog and I discussed this change to make introducing Shortcuts easier on the MenuAnchor widget. This change should also make testing semantics trees slightly less onerous (assuming this doesn't harm accessibility, that is).
On the point of accessibility, I'm not sure what useful semantic information is exposed by the Shortcuts widget's semantic node, since it's basically just an unfocusable keyboard listener. However, I kept the `includeSemantics` defaulting to true so as to not break semantics tests in the Flutter repo.
Fixes https://github.com/flutter/flutter/issues/152076
When the `"${native_assets_path}"*.framework` glob doesn't resolve anything, the bash will run the loop once with the original unglobbed value: `/path/to/native/assets/*.framework`. Skip this case to prevent the build from failing when there are no frameworks to sign.
To reproduce this build failure:
1. Enable native assets in the Flutter tool: `flutter config --enable-native-assets`
2. Create a Flutter project with the default template: `flutter create test_native_assets`
3. Add a build hook that does nothing (`hook/build.dart`).
4. Try to build/run the app: `flutter run --debug -d macos`
1. Instead of getting the `FULL_PRODUCT_NAME` Xcode build setting (`Runner.app`) instead use `PRODUCT_NAME` since most places really want the product name, and the extension stripping wasn't correct when the name contained periods.
2. Don't instruct the user to open the `xcarchive` in Xcode if it doesn't exist.
Fixes https://github.com/flutter/flutter/issues/140212
For the necessary background knowledge, see the flutter.dev content on [Resolution-aware image assets](https://docs.flutter.dev/ui/assets/assets-and-images#resolution-aware) and [Conditional bundling of assets based on app flavor](https://docs.flutter.dev/ui/assets/assets-and-images#conditional-bundling-of-assets-based-on-app-flavor) if you don't have a basic understanding of these features.
Fixes https://github.com/flutter/flutter/issues/151813 by using unique temporary directories, per asset file, for transformations. Currently, only a single directory is used and the name of the temporary files was based only on the basename of files. This means that `assets/image.png` and `assets/2x/image.png` would share an output path (`<temp dir path>/image.png`), causing a race. If this quick and rough explanation is a bit confusing, the original issueâ#151813âprovides a full repro and correct identification of the exact cause of the failure that can occur in the asset transformation process.
https://github.com/flutter/flutter/issues/150800. I skipped files that contain generated code because the script isn't smart enough, and I did not fix some dubious comment refs. Those will be addressed in a different PR.
Fixes a bug where WillPopScope no longer worked on the home route.
With this PR, Android's predictive back feature will be explicitly disabled when a WillPopScope widget is in the widget tree. To get the same behavior and still support predictive back, use PopScope.
This PR properly resets the drag state when losing the gesture arena or when the recognizer stops tracking the current pointer. The _dragState enum was reset properly, but I had forgotten to also reset the `_start`, this caused an issue when the recognizer won the gesture arena the next time, as it tries to detect a drag given the old `_start` in `acceptGesture`, but the `_dragState` has been reset causing an assertion to trigger.
This pull request implements [enhanced enum](https://dart.dev/language/enums#declaring-enhanced-enums) features for the new `WidgetState` enum, in order to improve the developer experience when creating and using `WidgetStateProperty` objects.
`WidgetState` now has a `.matchesSet()` method:
```dart
// identical to "states.contains(WidgetState.error)"
final bool hasError = WidgetState.error.isSatisfiedBy(states);
```
This addition allows for wide variety of `WidgetStateProperty` objects to be constructed in a simple manner.
<br><br>
```dart
// before
final style = MaterialStateTextStyle.resolveWith((states) {
if (states.contains(MaterialState.error)) {
return TextStyle(color: Colors.red);
} else if (states.contains(MaterialState.focused)) {
return TextStyle(color: Colors.blue);
}
return TextStyle(color: Colors.black);
});
// after
final style = WidgetStateTextStyle.fromMap({
WidgetState.error: TextStyle(color: Colors.red),
WidgetState.focused: TextStyle(color: Colors.blue),
WidgetState.any: TextStyle(color: Colors.black), // "any" is a static const member, not an enum value
});
```
```dart
// before
final color = MaterialStateProperty.resolveWith((states) {
if (states.contains(MaterialState.focused)) {
return Colors.blue;
} else if (!states.contains(MaterialState.disabled)) {
return Colors.black;
}
return null;
});
// after
final color = WidgetStateProperty<Color?>.fromMap({
WidgetState.focused: Colors.blue,
~WidgetState.disabled: Colors.black,
});
```
```dart
// before
const activeStates = [MaterialState.selected, MaterialState.focused, MaterialState.scrolledUnder];
final color = MaterialStateColor.resolveWith((states) {
if (activeStates.any(states.contains)) {
if (states.contains(MaterialState.hovered) {
return Colors.blueAccent;
}
return Colors.blue;
}
return Colors.black;
});
// after
final active = WidgetState.selected | WidgetState.focused | WidgetState.scrolledUnder;
final color = WidgetStateColor.fromMap({
active & WidgetState.hovered: Colors.blueAccent,
active: Colors.blue,
~active: Colors.black,
});
```
<br>
(fixes#146042, and also fixes#143488)
In particular, without this cross-reference it's easy for a reader
looking at [Image.frameBuilder] to think that the meanings of these
parameters just aren't properly documented. (I had that thought for
a minute, before thinking *surely* they're documented, and then
trying the link to the typedef.)
fixes https://github.com/flutter/flutter/issues/146414
---
I saw @abikko submitted a PR https://github.com/flutter/flutter/pull/146630, but it was not completed due to CLA and lack of test cases.
I am willing to complete this PR in combination with @abikko's code (I don't know if this is allowed)
This PR adds contorlAffinity to ListTileTheme so that [CheckboxListTile], [RadioListTile], [SwitchListTile], and [ExpansionTile] can read and use it
For example: If ListTileTheme in Theme sets contorlAffinity, then [CheckboxListTile] can directly use contorlAffinity in ListTileTheme. Of course, if contorlAffinity is also set in [CheckboxListTile], the property in [CheckboxListTile] will be used first.