Summary:
In Fabric, there are two subtle differences in Animated:
1) On the native side, queued NativeAnimated operations are executed more frequently
2) On the JS side there seems to be a bigger gap between AnimatedNodes being set up, and the final `componentDidMount`/`componentDidUpdate` that results in `_attachNativeEvents` being called.
This results in some visual glitching for certain features like StickyScrollViewHeader and other animations that are frequently rerendered. Rerendering an animated component relies on sync-like behavior
(new animations must be set up in the same frame that old animations are torn down, and the UI must flush any updates in the same frame). This sync-like behavior is trickier to achieve in Fabric, so
we introduce this mechanism, in Fabric only, to queued operations more aggressively on the JS side and only flush them when `componentDidMount`/`componentDidUpdate` have been called on all relevant nodes.
This seems to resolve longstanding issues with animations in Fabric on Android.
There are unrelated issues on iOS that make this difficult to test currently, so I'll keep the new path Android-only for now.
Changelog: [Internal]
Reviewed By: sammy-SC
Differential Revision: D22801173
fbshipit-source-id: 3db2235483636c9074df941cfac0a30f3f97545a
Summary:
Discovered when debugging a React Native app in Chrome. The function `performanceNow` was recently changed, so that in Chrome it now refers to `performance.now()` method. However, `now()` cannot be called standalone without a context.
Reproduced by a synthetic example in Chrome:
```
const performanceNow = window.performance.now;
performanceNow();
Uncaught TypeError: Illegal invocation
at <anonymous>:1:1
```
Changelog:
[General] [Fixed] - Fix failure when debugging code in a browser; was caused by `performanceNow()` function.
Reviewed By: cpojer
Differential Revision: D22739176
fbshipit-source-id: f89b8a215b7b4c430ffd72a1d23539c4f1b31d24
Summary:
In the next react sync we removed the `displayName` from forwardRef and useMemo components ([PR here](https://github.com/facebook/react/pull/18495 )).
This means we need to manually add the displayName in the mock.
Changelog: [General] [Fixed] Fix test renderer mocks to use the displayName more often.
Reviewed By: TheSavior
Differential Revision: D22775470
fbshipit-source-id: 1390dc325e34f7ccea32bbdf1c6a8f6efea3a080
Summary:
This diff adds some integration tests that throw read React errors and asserts on the LogBox data.
Changelog: [Internal]
Reviewed By: motiz88
Differential Revision: D22752147
fbshipit-source-id: eb2e1542524d67558f2450130015164acd4d6093
Summary:
In the next version of React they've changed the format of component stacks from a custom format to the same format as call stacks. This diff adds handling (and test) for the new format.
Changelog: [Internal]
Reviewed By: motiz88
Differential Revision: D22752034
fbshipit-source-id: 38d74bffd9ed623cbdd71e70103d66d8f0765fd9
Summary:
Changelog: Fix [TypeError: Network request failed] on file upload
# Problem
In https://github.com/facebook/react-native/commit/31980094107ed37f8de70972dbcc319cc9a26339 I made method `loadImageForURL` blocking and nil returning since it was no longer cancellable.
However inside `[RCTNetworkTask validateRequestToken]` was a logic counting on request token being non nil.
This diff removes this check and adds `nullable` to `[RCTImageURLLoader loadImageForURL]` making it explicit.
Reviewed By: PeteTheHeat
Differential Revision: D22767174
fbshipit-source-id: 04d5562e381912233b9c14e8156cbf145288f063
Summary:
Only add Sec-WebSocket-Protocol header if has _requestedProtocols.
## Changelog
<!-- Help reviewers and the release process by writing your own changelog entry. For an example, see:
https://github.com/facebook/react-native/wiki/Changelog
-->
[iOS] [Fixed] - when Sec-WebSocket-Protocol header is empty vaulue, IIS server will return error 502.
Pull Request resolved: https://github.com/facebook/react-native/pull/29419
Reviewed By: cpojer
Differential Revision: D22782867
Pulled By: PeteTheHeat
fbshipit-source-id: d588997a345929b0c11c554d4452e612a336901a
Summary:
The Animated native module cannot be converted to a TM because it has perf regressions when used in conjunction with Paper renderer. Instead of fixing these complicated perf issues (which will disappear when Fabric ships this half), temporarily fork the native module. The new fork is converted to a TM, and only used to unblock bridgeless mode testing.
Changelog:[Internal]
Reviewed By: RSNara
Differential Revision: D22651010
fbshipit-source-id: 912123ef38ac8c66025b7bba34a65ec6d98f330d
Summary:
Changelog:
[Internal][Added] - Change signature of `addTimestamp` for IPerformanceLogger and rename old implementation to addTimeAnnotation
Adding this because there is no current ability to add a timespan to our perf logging with pre-defined start and end timestamps. The naming is a bit confusing because the current implementation of `addTimespan` in FB doesn't add a timespan, it adds an annotation of duration. We rename the old implementation to `addTimeAnnotation/s` and update `addTimespan` to suit our needs.
Reviewed By: rubennorte
Differential Revision: D22633202
fbshipit-source-id: 0e32099241f1f3835257cbb4ad108a4f437869e6
Summary:
I want to be able parse error stacks in contexts where I have a call stack string but no error object.
This diff changes parseErrorStack to only accept the stack, instead of a whole error object.
Changelog: [Internal]
Reviewed By: motiz88
Differential Revision: D22752048
fbshipit-source-id: b4b1cd58802eefe736130d48a82bc091241a11ee
Summary:
Changelog: Fix possible race condition inside `KeyboardAvoidingView`
Fabric has different order of events
In Fabric, `keyboardWillChangeFrame` event is fired before `onLayout`, but in Paper it is the other way around.
`KeyboardAvoidingView` depends on the order of events to function properly. Inside `_relativeKeyboardHeight` 0 is returned if `this._frame` is null.
To fix this, `bottom` margin is recalculated whenever `keyboardWillChangeFrame` or `onLayout` is triggered.
Reviewed By: shergin
Differential Revision: D22764192
fbshipit-source-id: 591ac59af4395a4d43c4e78e4fbc0ff118b292f8
Summary:
Cleaning up a bunch of legacy JS (var) and removing some variable names for stuff that is just used once.
Changelog: [Internal]
Reviewed By: GijsWeterings
Differential Revision: D22627397
fbshipit-source-id: 783af84b17cbf37df3b8e2ae10bb39d50dd8180d
Summary:
This PR is to fix https://github.com/facebook/react-native/issues/29279, which image cannot show in iOS 14
As https://github.com/facebook/react-native/issues/29279#issuecomment-658244428 mention, this issue can be fixed by calling ` [super displayLayer:layer];` it it is still image, to let `UIImageView` handle still image rendering
## Changelog
[iOS] [Fixed] - Fix image cannot show in iOS 14
Pull Request resolved: https://github.com/facebook/react-native/pull/29420
Test Plan:
Image can be shown in iOS 14 build with Xcode 12 beta, using
```js
<Image source={require('./images/some_local_image.jpg')}/>
```
It may also need to test gif image is render correctly
```js
<Image source={{uri: 'https://some_remote_gif_image.gif'}}/>
```
Reviewed By: p-sun
Differential Revision: D22619448
Pulled By: shergin
fbshipit-source-id: f4d0ad83af945a6b8099d4eaea5a5f1933c7bfd2
Summary:
As a part of the "Lean Core" effort we are moving many React Native open source modules into React Native. Through this task, we reduced the app size of all React Native applications!
For this task, I moved the ART iOS files to Facebook internal.
Changelog:
[iOS][Changed] - Moved ART iOS files from React Native to Facebook internal
Reviewed By: PeteTheHeat
Differential Revision: D22560856
fbshipit-source-id: 4ed982b0361fef3e875374b6c3df0204037c2bfa
Summary:
This appears to be the only usage of whitelist/blacklist in react-native-github, let's use some more inclusive language.
Changelog: [JS][Internal]
Reviewed By: kacieb
Differential Revision: D22539431
fbshipit-source-id: 21d4cd54a5a2a676996ccec7b02ef15c421efee1
Summary:
In Fabric if you use a Pressable or PressableBackground and only provide an `onPress` but don't provide background color or anything else, the View might be flattened away and with it the event handlers.
We assume that if the product engineer is using a Pressable, they want it to do something in the View hierarchy, so force it to never be flattened.
Changelog: [Internal]
Reviewed By: shergin
Differential Revision: D22557026
fbshipit-source-id: 171c5b51e8d828fbb2816685d188286179a10af9
Summary:
Refs https://github.com/facebook/react-native/issues/28711
This PR moves Hermes badge component from the template to the `NewAppScreen` library. The main motivation behind this change was to simplify a bit template code.
I assumed that it is not important to expose `global.HermesInternal` to the template users:
* If this assumption is true, I think that there are no other reason to leave this component inside `App` in comparison to other `NewAppScreen` components,
* If this assumption is false, I can adjust this PR and move `HermesInternal` check from the badge component to the `App`.
I was trying to avoid calling `useColorScheme` when Hermes is disabled, but placing hook inside the conditional branch causes ESLint warning (react-hooks/rules-of-hooks).
This PR includes also small style tweaks for the badge - since there are no background padding can be omitted and spacing can be added adjusted tweaking `top` and `left` properties and `fontSize` has been adjusted just for the readability.
In the last commit, I have gone a bit further and moved `HermesBadge` to the `Header` component and I have also changed slightly the `Header` title (React -> React Native) and it's styling.
> I'm not sure if after this change `HermesBadge` export in `NewAppScreen` components list is still required, but maybe this badge will be useful for someone. If it's a mistake I can update the PR and remove this export.
## Changelog
[Internal][Changed] move Hermes badge from the template to the NewAppScreen library
Pull Request resolved: https://github.com/facebook/react-native/pull/28783
Test Plan:
Template app do not redbox on Android emulator with and without Hermes enabled.
## Preview
Android with Hermes enabled and adjusted header:
iOS with adjusted header:
Reviewed By: GijsWeterings
Differential Revision: D22493822
Pulled By: cpojer
fbshipit-source-id: 3440e10f2d59f268ca8851a6e002f0ff23fa839c
Summary:
Fixed tests related to changing Jest's fake timer behavior.
And some of the test code in LogBoxData-test.js itself was incorrect so I am fixing it.
LogBoxData-test.js fails when I try to build a new local development environment and fork the React Native source to run a JavaScript test.
The error message is:
runAllImmediates is not available when using modern timers
After checking, runAllImmediates could not be used with the following modification.
https://github.com/facebook/jest/commit/71631f6bf9ccf8937f02a5ecfb64b62712b86c19
## Changelog
[Internal] [Fixed] - Test fixes related to changes in Jest's Fake Timer behavior
Pull Request resolved: https://github.com/facebook/react-native/pull/29011
Test Plan: Ran JavaScript Tests locally.
Reviewed By: cpojer
Differential Revision: D22494991
Pulled By: rickhanlonii
fbshipit-source-id: 4deeaf82b5092ff8b60c4606eb45549beb452a5f
Summary:
Replace react-test-renderer/shallow with react-shallow-renderer.
We should follow up with teams to remove these tests because they will no longer be supported. I would have just removed them but test like `BillingReauthorizeCreditCardContainer-test` seem important.
Changelog: [Internal]
Reviewed By: cpojer
Differential Revision: D22225642
fbshipit-source-id: a6bd559311422cb14576a574165ed7dc0279919d
Summary:
BubblingEventTypes are inconsistent between AndroidTextInputViewConfig.js and ReactTextInputManager.java, this diff fixes this inconsistency
changelog: [internal]
Reviewed By: JoshuaGross
Differential Revision: D22470096
fbshipit-source-id: 3940dcc0ae67a42ac070c06ec2d54bc365eab6b7
Summary:
Replaces `fbjs/performanceNow` call sites in React Native with `performance.now`.
We did not originally polyfill this in `InitializeCore`, but now we do (and back it with a proper `nativePerformanceNow` implementation). Also, added the missing polyfill to our Jest setup.
Changelog:
[Internal]
Reviewed By: cpojer
Differential Revision: D22445948
fbshipit-source-id: dcfd9867c050617f6e2a3d0a1eb6f48a44771dda
Summary:
When you call `LogBox.ignoreAllLogs()` it should ignore logs.
This fixes a bug that made this equivalent to `LogBox.ignoreAllLogs(false)`
## Changelog
[General] [Fixed] - LogBox.ignoreAllLogs() should ignore logs
Pull Request resolved: https://github.com/facebook/react-native/pull/29310
Test Plan: Added tests
Reviewed By: TheSavior
Differential Revision: D22448436
Pulled By: rickhanlonii
fbshipit-source-id: 6ba12b9d9c1f29cf3ac503946ac5ca0097425a7a
Summary:
This PR adds support for the `shadowColor` style on Android.
This is possible as of Android P using the `setOutlineAmbientShadowColor` and `setOutlineSpotShadowColor` View methods. The actual rendered color is a multiplication of the color-alpha, shadow-effect and elevation-value.
## Changelog
`[Android] [Added] - Add support for shadowColor on API level >= 28`
Pull Request resolved: https://github.com/facebook/react-native/pull/28650
Test Plan:
- Only execute code on Android P
- Added Android `BoxShadow` tests to RNTester app
Reviewed By: mdvacca
Differential Revision: D21125479
Pulled By: shergin
fbshipit-source-id: 14dcc023977d7a9d304fabcd3c90bcf34482f137
Summary:
## Why?
1. RCTTurboModuleLookupDelegate sounds a bit nebulous.
2. In JS and Java, we use a `TurboModuleRegistry` interface to require TurboModules. So, this diff will make JS, Java, and ObjC consistent.
Changelog:
[Internal]
Reviewed By: PeteTheHeat
Differential Revision: D22405754
fbshipit-source-id: 30c85c246b39d198c5b8c6ca4432a3196ca0ebfd
Summary:
This moves ios CameraRoll files from React Native open source into FB internal as part of the Lean Core effort.
Changelog: [Breaking][ios] Remove CameraRoll from React Native
Reviewed By: cpojer
Differential Revision: D22208352
fbshipit-source-id: 894d6aff34ece94648dad68060c13b44974c93bb
Summary:
Pet Peeve: Metro is a brand name. You don't say "the Metro server" just like you don't say "the iPhone phone". This is a leftover from when it used to be called "the packager server".
Note: It makes sense to refer to "the Metro server" when talking about it in the context of Metro's features, like if you are discussing "Metro's bundling" and "Metro's server". However, when talking about the tool itself, just Metro is enough.
Changelog: [Internal]
Reviewed By: motiz88
Differential Revision: D22330966
fbshipit-source-id: 667618363c641884df543d88cac65d1e44956ad3
Summary:
The Hermes inspector always logged the message for `console.assert`.
It instead should check the first argument, and only if that argument
is false should it log.
Also remove the polyfill code that tried to avoid this situation.
Changelog: [Internal]
Reviewed By: mhorowitz
Differential Revision: D22299186
fbshipit-source-id: cdf4f8957d4db3d171d6673a82c7fc32b7152af3
Summary:
onAssetDidLoad is defined as bubbling event in iOS, change it to bubbling event and adde it to bubblingEventTypes to fix "missing: topAssetDidLoad" error in Viewpoints iOS.
Changelog: [Internal]
Reviewed By: JoshuaGross, mdvacca
Differential Revision: D22209406
fbshipit-source-id: 5da779d9d1d62c70d85d84ad80807ff688e29e2f
Summary:
Changelog: [Internal]
Introducing InputAccessoryView.
There is one big difference between Fabric's implementation and Paper's implementation.
Fabric searches for text input from InputAccessoryView, unlike Paper where it is the other way around.
Reviewed By: shergin
Differential Revision: D22160445
fbshipit-source-id: 55313fe50afeced7aead5b57137d711dd1cfd3ae
Summary:
After animation has been finished using Native driver there is no final value passed from the native to JS side. This causes a bug from https://github.com/facebook/react-native/issues/28114.
This PR solves this problem in the same way as `react-native-reanimated` library. When detaching it is calling native side to get the last value from Animated node and stores it on the JS side.
Preserving animated value even if animation was using `useNativeDriver: true`
Fixes https://github.com/facebook/react-native/issues/28114
## Changelog
<!-- Help reviewers and the release process by writing your own changelog entry. For an example, see:
https://github.com/facebook/react-native/wiki/Changelog
-->
[Internal] [Fixed] - Save native Animated node value on JS side in detach phase
Pull Request resolved: https://github.com/facebook/react-native/pull/28841
Test Plan: Unit tests for added getValue method passed. Green CI
Reviewed By: mdvacca
Differential Revision: D22211499
Pulled By: JoshuaGross
fbshipit-source-id: 9a3a98a9f9a8536fe2c8764f667cdabe1f6ba82a
Summary:
The docs suggest that TouchableNativeFeedback [inherits `TouchableWithoutFeedback` props](https://reactnative.dev/docs/touchablewithoutfeedback#props) but `accessibilityHint` was missing.
## Changelog
[Android] [Added] - Add accessibilityHint to TouchableNativeFeedback
Pull Request resolved: https://github.com/facebook/react-native/pull/29154
Test Plan: Not sure how this should be tested, but I'm happy to implement what others may suggest
Reviewed By: kacieb
Differential Revision: D22109459
Pulled By: TheSavior
fbshipit-source-id: 573267a26414a97ba23a1a7995bff1608f9ba34f
Summary:
In an effort to simplify and clean up the `EventEmitter` abstractions in React Native, this removes `once()` and `removeCurrentListener()`. Across the Facebook codebase, there were only two callers of `once()` and no callers of `removeCurrentListener()`.
The same behavior can be achieved using:
```
const subscription = emitter.addListener('event', () => {
subscription.remove();
// ...
});
```
Changelog:
[General][Removed] - Removed `once()` and `removeCurrentListener()` fom `DeviceEventEmitter` and `NativeEventEmitter`.
Reviewed By: cpojer
Differential Revision: D22196474
fbshipit-source-id: 06ced186fd812e91d5c57f6580e647c100505807
Summary:
https://github.com/facebook/react-native/issues/27264 changed stylesheet validation to avoid enumerating properties on the prototype of a style. It introduces a secondary behavior change, where null/undefined styles used to be tolerated but now lead to an exception. This is because `for in undefined` will noop where `for of Object.keys(undefined)` will throw.
This scenario of undefined/null styles seems to actually show up in practice and was previously well tolerated. E.g. `Button.js` has code that looks like this:
```jsx
const styles = StyleSheet.create({
button: Platform.select({
ios: {},
android: {
elevation: 4,
// Material design blue from https://material.google.com/style/color.html#color-color-palette
backgroundColor: '#2196F3',
borderRadius: 2,
},
}),
```
For non ios/Android platforms, that creates a style object which looks like:
```js
{
button: undefined,
...
}
```
This previously meant that the component would be unstyled if created, but now means out-of-tree platforms throw if the builtin Button component is required.
This change restores the previous `for in` loop but adds a `hasOwnProperty` check to avoid properties on prototypes.
## Changelog
[General] [Fixed] - Restore Previous Behavior for StyleSheet Validation of Null/Undefined Styles
Pull Request resolved: https://github.com/facebook/react-native/pull/29171
Test Plan: Validated that importing Buttons will no longer cause an exception, and that invalid properties are still caught.
Reviewed By: JoshuaGross
Differential Revision: D22118379
Pulled By: TheSavior
fbshipit-source-id: 650c64b934ccd12a3dc1b75e95debc359925ad73
Summary:
Renames `EventSubscription.js`, `EmitterSubscription.js`, and `EventSubscriptionVendor.js` to `_EventSubscription.js`, `_EmitterSubscription.js`, and `_EventSubscriptionVendor.js`, respectively.
This is to indicate that those files are implementation details and should not be directly referenced. Instead, the `EventSubscription` type exported from `EventEmitter.js` should be used.
The remaining stragglers that are importing `_EmitterSubscription.js` and `_EventSubscriptionVendor.js` after this commit will be cleaned up when `RCTDeviceEventEmitter` is refactored.
Changelog:
[Internal]
(Note: this ignores all push blocking failures!)
Reviewed By: cpojer
Differential Revision: D22182310
fbshipit-source-id: 82be685f231395bd7b8e9986141b5df1367bec71
Summary:
Changes dependents of `EmitterSubscription` to instead import the `EventSubscription` type from `EventEmitter.js`.
Changelog:
[Internal]
(Note: this ignores all push blocking failures!)
Reviewed By: cpojer
Differential Revision: D22182309
fbshipit-source-id: 575f4c59248ef50182ddb33911f1e6f3ba88ec07
Summary:
Changes dependents to import the `EventSubscription` interface type from `EventEmitter` instead of `EventSubscription.js`.
Changelog:
[Internal]
(Note: this ignores all push blocking failures!)
Reviewed By: cpojer
Differential Revision: D22182313
fbshipit-source-id: 31448ea5ee3038f806a417f23597da1edd4e4e9b
Summary:
Changes `EventEmitter.js` to be Flow-typed so that it can export an interface for `EventSubscription`. In order to retain the untyped-ness of `EventEmitter.js`, I moved the entire definition into an untyped `_EventEmitter.js` file (which I hope no one will try importing from).
This new interface will be used to replace all the current type imports of `EventSubscription` and `EmitterSubscription`.
Changelog:
[Internal]
(Note: this ignores all push blocking failures!)
Reviewed By: cpojer
Differential Revision: D22182311
fbshipit-source-id: c604a371a91963116621e89ca63f5a82167090b9
Joshua Gross
Andrey Tserkus
Paito Anderson
Rick Hanlon
Samuel Susla
bill2004158
Peter Argany
Luna Wei
Christoph Nakazawa
Rubén Norte
Tom Cheung
Jayesh Kawli
Tommy Nguyen
David Vacca
simek
makitake2013
Tim Yung
Rick Hanlon
Hein Rutjes
Ramanpreet Nara
Sean Wang
Riley Dulin
Lulu Wu
TMaszko
fabriziobertoglio1987
Chris Dadabo
Nick Gerleman
