Summary:
Fixes#13691.
Prior to this change, VoiceOver generally didn't say anything when selecting a multiline TextInput. The only exception was VoiceOver would announce the TextInput's placeholder if it had one set.
After this change, VoiceOver announces the following for multiline TextInputs:
- The TextInput's content.
- The TextInput's placeholder if the TextInput doesn't have any content in it.
- The fact that the TextInput is a textfield.
- The cursor position if the TextInput is being edited.
This is similar to the behavior of single line TextInputs.
This change achieves this by disabling `RCTTextView` as an accessibility element. `RCTTextView` is a subclass of `RCTView` so VoiceOver doesn't recognize this as a textfield. Instead, VoiceOver now sees the child `RCTUITextView` which is a subclass of `UITextView` so VoiceOver does recognize it as a textfield. Additionally, an `accessibilityLabel` implementation was added to `RCTUITextView` in order to take the value of the placeholder into account.
Verified the announcements of TextInputs with various props:
- No placeholder and no content
- Just a placeholder
- Just content
- Both a placeholder and content
Did this for both singe line inputs and multiline inputs. For setting content in multiline inputs, I tested both using the `value` prop and passing children. All other props being equal, these configurations resulted in similar announcements.
I verified that the following accessibility props work the same on singleline and multiline TextInputs:
- `accessible`
- `accessibilityLabel`
- `accessibilityTraits`
- `accessibilityViewIsModal`
- `onAccessibilityTap`
- `onMagicTap`
Additionally, my team has been using this change in our app.
Adam Comella
Microsoft Corp.
Closes https://github.com/facebook/react-native/pull/14200
Differential Revision: D5185727
Pulled By: shergin
fbshipit-source-id: 94271e6c8b089eb82006b52fe7917649d69e74af
Summary:
Motivation:
* We maintain two different implementation of <TextInput> (multilined and singlelined), this change makes the implementations much similar which will help us to support and improve both of them in the (near) future;
* We have to have separated RCTView-based container view for (TextField) to support sofisticated bordering and so on;
* It opens to us possibility to unify UITextView and UITextField subclasses and remove code duplication across RCTTextView and RCTTextField;
* Making things decoupled in general will allow us to fix existing bugs with events.
Reviewed By: mmmulani
Differential Revision: D5083010
fbshipit-source-id: 2f2d42c2244d2b39256c51480c1f16f4e3947c01
Summary: Now padding, border and intinsic sizes are computed same way as for singlelined text input.
Reviewed By: mmmulani
Differential Revision: D5075880
fbshipit-source-id: 1bc2fd479c13a003c717b1fc3d9c69f4639d4444
Summary:
Pair `reactWillMakeFirstResponder` and `reactDidMakeFirstResponder` was replaced with just `reactFocus` method
which is supposed to incapsulate all "focus" and "focus-later-if-needed" functionality.
Reviewed By: mmmulani
Differential Revision: D4664626
fbshipit-source-id: 8d3b7935ca26d32ba1d1826a585cce0396fcc885
Adam Comella
Valentin Shergin