This is for https://github.com/flutter/flutter/issues/127803: a text field should unregister from the scribble scope, when it becomes unfocusable.
When a `FocusNode` has listeners and its `_canRequestFocus` flag is set to true, it adds `+1` to `_focusabilityListeningDescendantCount` of all ancestors until it reaches the first ancestor with `descendantsAreFocusable = false`. When the a `FocusNode`'s `descendantsAreFocusable` changes, all listeners that contributed to its `_focusabilityListeningDescendantCount` will be notified.
Reverts flutter/flutter#144207
Initiated by: CaseyHillers
Reason for reverting: b/327301206 - Breaking a customer test
Original PR Author: LongCatIsLooong
Reviewed By: {gspencergoog}
This change reverts the following previous change:
Original Description:
`FocusNode.canRequestFocus` was doing a double traversal if no ancestor disallows focus. The last for loop only has to reach as far as the enclosing scope.
Also this caches the `FocusNode.enclosingScope` since the getter access happens much more frequently than node reparenting.
`FocusNode.canRequestFocus` was doing a double traversal if no ancestor disallows focus. The last for loop only has to reach as far as the enclosing scope.
Also this caches the `FocusNode.enclosingScope` since the getter access happens much more frequently than node reparenting.
The original PR was reverted because the new caret positioning callpath triggered a skparagraph assert. The assert has been removed. Relanding the PR with no changes applied.
This pull request fixes#143803 by taking advantage of Dart's null-aware operators.
And unlike `switch` expressions ([9 PRs](https://github.com/flutter/flutter/pull/143634) and counting), the Flutter codebase is already fantastic when it comes to null-aware coding. After refactoring the entire repo, all the changes involving `?.` and `??` can fit into a single pull request.
Reverts flutter/flutter#143281
Initiated by: LongCatIsLooong
Reason for reverting: https://github.com/flutter/flutter/issues/143797
Original PR Author: LongCatIsLooong
Reviewed By: {justinmc, jason-simmons}
This change reverts the following previous change:
Original Description:
The behavior largely remains the same, except:
1. The EOT cursor `(textLength, downstream)` for text ending in the opposite writing direction as the paragraph is now placed at the visual end of the last line.
For example, in a LTR paragraph, the EOT cursor for `aA` (lowercase for LTR and uppercase for RTL) is placed to the right of the line: `aA|` (it was `a|A` before).
This matches the behavior of most applications that do logical order arrow key navigation instead of visual order navigation.
And it makes the navigation order consistent for `aA\naA`:
```
|aA => aA| => aA| => aA => aA => aA
aA aA aA |aA aA| aA|
(1) (2) (3) (4) (5) (6)
```
This is indeed still pretty confusing as (2) and (3), as well as (5) and (6) are hard to distinguish (when the I beam has a large width they are actually visually distinguishable -- they use the same anchor but one gets painted to the left and the other to the right. I noticed that emacs does the same).
But logical order navigation will always be confusing in bidi text, in one way or another.
Interestingly there are 3 different behaviors I've observed in chrome:
- the chrome download dialog (which I think uses GTK text widgets but not sure which version) gives me 2 cursors when navigating bidi text, and
- its HTML fields only show one, and presumably they place the I beam at the **trailing edge** of the character (which makes more sense for backspacing I guess).
- On the other hand, its (new) omnibar seems to use visual order arrow navigation
Side note: we may need to update the "tap to place the caret here" logic to handle the case where the tap lands outside of the text and the text ends in the opposite writing direction.
2. Removed the logarithmic search. The same could be done using the characters package but when glyphInfo tells you about the baseline location in the future we probably don't need the `getBoxesForRange` call. This should fix https://github.com/flutter/flutter/issues/123424.
## Internal Tests
This is going to change the image output of some internal golden tests. I'm planning to merge https://github.com/flutter/flutter/pull/143281 before this to avoid updating the same golden files twice for invalid selections.
Fixes https://github.com/flutter/flutter/issues/138912
Change `ItemExtentBuilder`'s return value nullable, it should return null if asked to build an item extent with a greater index than exists.
The behavior largely remains the same, except:
1. The EOT cursor `(textLength, downstream)` for text ending in the opposite writing direction as the paragraph is now placed at the visual end of the last line.
For example, in a LTR paragraph, the EOT cursor for `aA` (lowercase for LTR and uppercase for RTL) is placed to the right of the line: `aA|` (it was `a|A` before).
This matches the behavior of most applications that do logical order arrow key navigation instead of visual order navigation.
And it makes the navigation order consistent for `aA\naA`:
```
|aA => aA| => aA| => aA => aA => aA
aA aA aA |aA aA| aA|
(1) (2) (3) (4) (5) (6)
```
This is indeed still pretty confusing as (2) and (3), as well as (5) and (6) are hard to distinguish (when the I beam has a large width they are actually visually distinguishable -- they use the same anchor but one gets painted to the left and the other to the right. I noticed that emacs does the same).
But logical order navigation will always be confusing in bidi text, in one way or another.
Interestingly there are 3 different behaviors I've observed in chrome:
- the chrome download dialog (which I think uses GTK text widgets but not sure which version) gives me 2 cursors when navigating bidi text, and
- its HTML fields only show one, and presumably they place the I beam at the **trailing edge** of the character (which makes more sense for backspacing I guess).
- On the other hand, its (new) omnibar seems to use visual order arrow navigation
Side note: we may need to update the "tap to place the caret here" logic to handle the case where the tap lands outside of the text and the text ends in the opposite writing direction.
2. Removed the logarithmic search. The same could be done using the characters package but when glyphInfo tells you about the baseline location in the future we probably don't need the `getBoxesForRange` call. This should fix https://github.com/flutter/flutter/issues/123424.
## Internal Tests
This is going to change the image output of some internal golden tests. I'm planning to merge https://github.com/flutter/flutter/pull/143281 before this to avoid updating the same golden files twice for invalid selections.
fixes#87061
It doesn't matter whether I'm using Google Chrome, VS Code, Discord, or a Terminal window: any time a text cursor is blinking, it means that the characters I type will show up there.
And this isn't limited to text fields: if I repeatedly press `Tab` to navigate through a website, there's a visual indicator that goes away if I click away from the window, and it comes back if I click or `Alt+Tab` back into it.
<details open>
<summary>Example (Chrome):</summary>
</details>
<details open>
<summary>This PR adds the same functionality to Flutter apps:</summary>
</details>
Fixes https://github.com/flutter/flutter/issues/138610.
When `RenderImage` receives a new `Image` it only needs to fire up the layout machinery when the dimensions of the image have changed compared to the previous image. If the dimensions are the same, a repaint is sufficient.
This is part 3 of a broken down version of the #140101 refactor.
There's some TODOs showing where I think we should change the behaviour, but in this PR the behaviour is unchanged.
A future PR will remove the tests that are redundant with these tests, but I wanted to make sure we had both sets in the codebase at the same time first.
This PR includes a change to the golden control test so that we can verify that these specific values do work on main. It would be extremely surprising if not, but in the interests of rigour...
`precacheImage` was failing to dipose the `ImageInfo` it receives. That's part of the contract of being an image listener.
I'm doing this in the frame callback for the same reason as evicting it from the cache.
This change affects Android and iOS devices using the TextField's context menu. After this change the context menu will fade out when scrolling the text and fade in when the scroll ends.
If the scroll ends and the selection is outside of the view, then the toolbar will be scheduled to show in a future scroll end. This toolbar scheduling can be invalidated if the `TextEditingValue` changed anytime between the scheduling and when the toolbar is ready to be shown.
This change also fixes a regression where the TextField context menu would not fade when the selection handles where not visible.
When using the native browser context menu this behavior is not controlled by Flutter.
https://github.com/flutter/flutter/assets/948037/3f46bcbb-ba6f-456c-8473-e42919b9d572Fixes#52425Fixes#105804Fixes#52426
Fixes https://github.com/flutter/flutter/issues/142480.
This fixes a crash occurring during hot reload when a `ViewAnchor` is used between a `ParentDataWidget` (like `Positioned`) and its closest `RenderObject` descendant. Prior to the fix, the `ParentDataWidget` was accidentally applying its parent data to the render object in the `ViewAnchor.view` slot, which crashed because that render object wasn't (and shouldn't be) setup to accept parent data (after all, it is in a different render tree). Instead, the parent data should only be applied to the render object in the `ViewAnchor.child` slot. Luckily, with `Element.renderObjectAttachingChild` we already have API in place to walk the widget tree such as that only `RenderObjectWidgets` from the same render tree are considered.
Fixes https://github.com/flutter/flutter/issues/142183
This fixes a bug in the SliverGeometry of SliverMainAxisGroup. The cacheExtent represents how many pixels the sliver has consumed in the SliverConstraints.remainingCacheExtent. Since it was not set, slivers that came after a SliverMainAxisGroup that filled the whole screen did not properly lay out their own children, in some cases making lazy sliver more eager than they should be.
Reverts flutter/flutter#142339
In the original change one of the tests included the same view twice which resulted in a different failure than the expected one. The second commit contains the fix for this. I don't understand how this wasn't caught presubmit on CI.
It fixes assertion failure due to unstable state of children list during reordering in `RenderTwoDimensionalViewport.parentDataOf`. This changes the assertion to check debug orphan list and `keepAlive` bucket in addition to children list to determine whether child belongs to this render object or not.
- Fixes#141101
Reverts flutter/flutter#141484
Initiated by: eliasyishak
This change reverts the following previous change:
Original Description:
The existing `runApp` bootstraps the widget tree and renders the provided widget into the default view (which is currently the implicit View from `PlatformDispatcher.implicitView` and - in the future - may be a default-created window). Apps, that want more control over the View they are rendered in, need a new way to bootstrap the widget tree: `runWidget`. It does not make any assumptions about the View the provided widget is rendered into. Instead, it is up to the caller to include a View widget in the provided widget tree that specifies where content should be rendered. In the future, this may enable developers to create a custom window for their app instead of relying on the default-created one.
The existing `runApp` bootstraps the widget tree and renders the provided widget into the default view (which is currently the implicit View from `PlatformDispatcher.implicitView` and - in the future - may be a default-created window). Apps, that want more control over the View they are rendered in, need a new way to bootstrap the widget tree: `runWidget`. It does not make any assumptions about the View the provided widget is rendered into. Instead, it is up to the caller to include a View widget in the provided widget tree that specifies where content should be rendered. In the future, this may enable developers to create a custom window for their app instead of relying on the default-created one.
Fixes#129590
* Consider `AxisDirection` when calculating scroll offset used in determining TextSelection during a drag/long press drag. Previously it seems that we were assuming the direction was always vertical 30cc831985/packages/flutter/lib/src/widgets/text_selection.dart (L2842-L2844) .
* SelectableText now considers RenderEditable offset changes and Scrollable offset changes when calculating the TextSelection during a long press drag.
