Summary:
The RootView being managed by Fabric should have an id of View.NO_ID when it is "handed over" to RN. This is true for Fabric and non-Fabric
and setting a custom id on the ReactRootView has never been supported. I'm temporarily (?) adding an additional check earlier and into ReactRootView to hopefully
catch any of these issues early.
Changelog: [Internal]
Reviewed By: mdvacca
Differential Revision: D26077509
fbshipit-source-id: 59e1ec080504e50698acc654c29120f039238a96
Summary:
In some cases, onMeasure/onLayout are called on the RootView before `startSurface` in Fabric has been able to set surfaceId on the RootView.
With the new SurfaceMountingManager, this causes a crash because we need a valid surfaceId to perform an operation. Before the SurfaceMountingManager refactor, a surfaceId of 0 would be passed to `mBinding.setConstraints` in the FabricUIManager, and setConstraints in C++ noops if there's an invalid surfaceId.
For now, FabricUIManager will also fail silently if the surfaceId is invalid when updateRootLayoutSpecs is called, just to be conservative and to be consistent with previous behavior. This will be upgraded to a hard-crash in the future.
Changelog: [Internal]
Reviewed By: mdvacca
Differential Revision: D26051266
fbshipit-source-id: ca2d80f899cdba9b3962af68546bd83b77be0680
Summary:
This diff fixes a race condition in the execution of FabricUIManager.StartSurface method.
The rootcause is that startSurface is executing getViewportOffset from a background thread.
changelog: [internal]
Reviewed By: shergin
Differential Revision: D25617154
fbshipit-source-id: 9351201088164e74bb0b9454e30651e1de0da912
Summary:
The root cause of this bug is a race condition between the onMeasure method and setupReactContext.
ReactInstanceManager.attachRootView() method is responsible of the initialization of ReactRootViews and it is invoked by ReactRootView.onMeasure() method in the UIThread
Important initialization steps:
1. Clear the Id of the ReactRootView
2. Add the ReactRootView to the mAttachedReactRoots
3. Call StartSurface (if the bridge has been initialized)
Sometimes, when this method is invoked for the first time, the bridge is not initialized, in those cases we delay the start of the surface.
Once the bridge is initialized, StartSurface is called by the setupReactContext() running in the NativeModuleThread.
Since onMeasure can be called multiple times, it is possible that we call "StartSurface" twice for the same ReactRootView, causing the bug reported on T78832286.
This diff adds an extra check to prevent calling "StartSurface" twice. The fix is done using an AtomicInteger comparison and it is gated by the flag "enableStartSurfaceRaceConditionFix". Once we verify this works fine in production we will clean up the code, remove the flags and maybe revisit the API of ReactRoot.
changelog: [Android] Fix race-condition on the initialization of ReactRootViews
Reviewed By: JoshuaGross
Differential Revision: D25255877
fbshipit-source-id: ca8fb00f50e86891fb4c5a06240177cc1a0186d9
Summary:
This feature flag (enableStopSurfaceOnRootViewUnmount) was used to gate usage of the "stopSurface" API, which is now fully supported, and has been used
in the FB app for several months. This is safe to ship in code.
Changelog: [Internal]
Reviewed By: mdvacca
Differential Revision: D25275192
fbshipit-source-id: fa22bfd00aa023297bc19c83c138f133e9ff1645
Summary:
Add three markers for ReactRootView perf logging: onMeasure, attachToReactInstanceManager, and updateLayoutSpecs.
It is suspected that one or all of these have regressed under Fabric in some cases.
Changelog: [Internal]
Reviewed By: shergin
Differential Revision: D24909635
fbshipit-source-id: 6b6c0cc792c4b2d72f2360d6cffc02ef00e8a88b
Summary:
Currently the dimensions are created once, and then cached. This change will reload the dimensions when the device orientation changes to insure that dimension update events follow orientation changed events.
this should help address the following issues, that I know of:
https://github.com/facebook/react-native/issues/29105https://github.com/facebook/react-native/issues/29451https://github.com/facebook/react-native/issues/29323
## 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
-->
[Android] [Fixed] - Dimension update events are now properly sent following orientation change
Pull Request resolved: https://github.com/facebook/react-native/pull/30324
Test Plan: Open up RNTester app. Select the Dimensions API list item. Rotate the device and verify that the dimensions are correct based on orientation.
Reviewed By: fkgozali
Differential Revision: D24874733
Pulled By: ejanzer
fbshipit-source-id: 867681ecb009d368a2ae7b67d94d6355e67dea7b
Summary:
Sometimes stopSurface crashes when unmountReactApplication is called under Fabric; knowing the stack trace up to this point might assist in debugging.
Changelog: [Internal]
Reviewed By: sammy-SC
Differential Revision: D24532027
fbshipit-source-id: f350e77fb1a2de52eb146b449f1d2f6e960fa017
Summary:
There are cases under investigation where unmountReactApplication is called before the ReactRootView gets an ID; in some or all of those cases, UIManagerBinding.stopSurface cannot get the ReactFabric JSI module and crashes there.
It's still unclear why `unmountReactApplication` is being called in these circumstances - maybe another crash causing React Native teardown?
Changelog: [Internal]
Reviewed By: sammy-SC
Differential Revision: D24214712
fbshipit-source-id: 796594653f4ff6d87088c4841b89f06cc873b46f
Summary:
In D23640968 (https://github.com/facebook/react-native/commit/78b42d7fb7180102c1e8ec917dcccd2d9d4076db) I introduced a mechanism to update offsetX/offsetY whenever onMeasure/onLayout were called, to ensure that `measureInWindow` had the correct metrics and would work properly.
However, now `uiManager.updateRootLayoutSpecs` gets spammed and is called too often. For example, whenever a TextInput is focused/blurred, `uiManager.updateRootLayoutSpecs` may be called 5+ times even though
the measure specs/offsets may only change once.
Thus, we just compare with previous values before calling into the UIManager. This should give us a very small perf improvement.
Changelog: [Internal]
Reviewed By: shergin
Differential Revision: D24176867
fbshipit-source-id: f0dcc816e651a843607e9e5d40d8f3489894d4ba
Summary:
Changelog: [internal]
# Why is text laid out twice in Fabric?
Layout constraints (min and max size) change during startup of Fabric surface.
1. `Scheduler::startSurface` is called with max size being {inf, inf}.
2. `Scheduler::constraintSurfaceLayout` is called with max size equal to viewport.
These are two operations that don't happen one after the other and on Android, CompleteRoot is called from JS before second operation is called. This triggers layout with max size {inf, inf} and later when second operation is called. Layout happens again now with correct size.
# Fix
Make sure `Scheduler::startSurface` is called with proper values and not {inf, inf}.
Reviewed By: JoshuaGross, yungsters
Differential Revision: D23866735
fbshipit-source-id: b16307543cc75c689d0c1f0a16aa581458f7417d
Summary:
In the past, in Fabric (Android), we never called stopSurface. Ever! This is bad for memory and can cause other issues, so... let's not do that.
Instead, when the ReactRootView is being torn down, we check the View ID to see if it's a Fabric RootView, and if it is, we call Fabric's stopSurface method.
Fabric stopSurface only has impact on (1) Fabric mount instructions being executed after that point and (2) tells JS to stop running the surface, on the JS thread, asynchronously.
Anecdotally it looks like all the MountItems involved in deleting and removing views from the hierarchy are executed before stopSurface is called.
Changelog: [Internal]
Reviewed By: mdvacca
Differential Revision: D21965837
fbshipit-source-id: 5169c5a1e339fd9e016ae1234d8389b2bd30be70
Summary:
Removing logs related to T62192299 that we don't need anymore.
Changelog: [Internal]
Reviewed By: mdvacca
Differential Revision: D21773629
fbshipit-source-id: d16c01f87be3ed7512fe90b6e261b4c7efbd3068
Summary:
This diff removes the variable 'useSurface' which wasn't beeing used in ReactRootView anymore
changelog: [Internal][Android] Internal cleanup of ReactRootView
Reviewed By: JoshuaGross
Differential Revision: D21318109
fbshipit-source-id: f850b27811608d16b22b6a3964455b172705c4c7
Summary:
This is the codemod portion of the parent diff.
I modified all call-sites to `ReactContext.getNativeModule` to do a null check on the returned NativeModule.
Changelog:
[Android][Fixed] - Check if NativeModules returned from CatalystInstanceImpl.getNativeModule are null before using them.
Reviewed By: JoshuaGross
Differential Revision: D21272028
fbshipit-source-id: 6bd16c6bf30605f2dfdf4c481352063712965342
Summary:
Fabric debugging.
I also have an agenda of removing all instances of `ReactConstants.TAG`, because it's overly broad and not helpful during debugging.
Changelog: [Internal]
Reviewed By: mdvacca
Differential Revision: D20016401
fbshipit-source-id: 91e5cd7e8eb662b1a4a887c1e4e66a31ec741b46
Summary:
This diff refactors the usages of UIManagerHelper.getUIManager() to make sure we always consider null objects.
Some of the callsites were throwing a NullPointerExcetpion, now they throw a more explicit exception.
changelog: [internal]
Reviewed By: makovkastar
Differential Revision: D19383064
fbshipit-source-id: 1806a37528e80cab1c8fdff5eb631aaf47bde819
Summary:
ReactRootView currently caches the last seen DisplayMetrics so we can compare them against the current values in DisplayMetricsHolder to determine if the screen dimensions have changed. (If they have changed, we emit an event to notify JS of the new dimensions).
However, ReactRootView initializes these member variables with empty DisplayMetrics, which means that the first time we check against them in onGlobalLayout, we will *always* emit an event to JS.
This seems reasonable if you only have one ReactRootView, but if you create a new RRV for each RN screen in your app, then you're going to get updated dimensions on each navigation event, even though the screen dimensions have probably not changed.
In this diff, I'm no longer storing the DisplayMetrics in ReactRootView at all, but instead am using temporary variables to check the new DisplayMetrics values against the old.
Changelog: [Android][Fixed] Only update dimensions in ReactRootView if they've changed
Reviewed By: JoshuaGross, mdvacca
Differential Revision: D19395326
fbshipit-source-id: c01aee73064764503c9b49208032c790b83a1d29
Summary:
This diff exposes the surfaceID as part of ReactRootView.
Changelog: [internal]
Reviewed By: JoshuaGross
Differential Revision: D18474014
fbshipit-source-id: 98f651b211d3cc4df88c6b1eb257bc12129eff57
Summary:
Fixes T54997838 by preventing any view mutations during `onMeasure` calls.
There might still be places where this is possible, but this is where I'm seeing all the crashes currently.
See comments in ReactRootView for why views were mutated during onMeasure.
Changelog: [Internal]
Reviewed By: mdvacca
Differential Revision: D18518591
fbshipit-source-id: 1406af8a6b0bfcc86f4cc5b451b3967f312dfd85
Summary:
Add UI thread assertions and annotations to ReactRootView. Shouldn't have any immediate effect since these methods already call other methods that assert they're on the UI thread. Doing this to hoist assumptions higher up.
Doing some very simple refactoring to the way cleanup happens to ensure aggressive cleanup in more instances.
Reviewed By: shergin
Differential Revision: D17860291
fbshipit-source-id: c87e0336594fa2327271b648eb8340e250235250
Summary:
This diff migrates the usages Nullable and NonNull annotations to AndroidX instead of javax.
The purpose of this change is to bring consistency in the annotations used by the core of RN
Reviewed By: makovkastar
Differential Revision: D16054504
fbshipit-source-id: 21d888854da088d2a14615a90d4dc058e5286b91
Summary:
This diff formats the Java class files inside xplat/js/react-native-github. Since google-java-format was enabled in D16071401 we want to codemode the existing code so that users don't have to deal with formatter lint noise at diff-time.
```arc f --paths-cmd 'hg files -I "**/*.java"'```
drop-conflicts
Reviewed By: cpojer
Differential Revision: D16071725
fbshipit-source-id: fc6e3852e45742c109f0c5ac4065d64201c74204
Summary:
Every call site is either already using `createReactContextInBackground` correctly or guarding the invocation using `hasStartedCreatingInitialContext`. This is an unnecessary and overly complex dance that can be simplified.
This revision simplifies the use of `createReactContextInBackground` by integrating the check. This is not a breaking change.
Reviewed By: zackargyle, mdvacca
Differential Revision: D15566632
fbshipit-source-id: 7b50285c9ac6776d1297d2c9c53dff208851b722
Summary:
We currently have two different codepaths for actually rendering a surface with Fabric on iOS and Android: on iOS we use Fabric's `UIManagerBinding.startSurface` to call `AppRegistry.runApplication`, but on Android we don't; instead we use the same codepath as paper, calling `ReactRootView.runApplication`.
This diff does a few different things:
1. Unify iOS and Android by removing the `#ifndef` for Android so that we call `startSurface` for both
2. Pass through the JS module name on Android so that this actually works (it currently passes in an empty string)
3. Remove the call to `ReactRootView.runApplication` for Fabric so that we don't end up doing this twice
4. Copy over some logic that we need from `ReactRootView.runApplication` (make sure that root layout specs get updated, and that content appeared gets logged)
Reviewed By: mdvacca
Differential Revision: D15501666
fbshipit-source-id: 5c96c8cf036261cb99729b1dbdff0f7c09a32d76
Summary:
Set duration=0 for android keyboard events. Brings actual implementation closer to existing flowtypes, and duration is set to 0 to minimize impact on existing keyboard event consumers.
Follow up to #24947, upon cpojer's [input](https://github.com/facebook/react-native/pull/24947#issuecomment-494681618) :)
## Changelog
[Android] [Added] - Set duration=0 for android keyboard events
Pull Request resolved: https://github.com/facebook/react-native/pull/24994
Differential Revision: D15449394
Pulled By: cpojer
fbshipit-source-id: d43096238bd38d189fbec54fc2d93f17010d9ddb
Summary:
This pull request enhances the Keyboard API event emitter for Android upon `keyboardDidHide` by returning a `KeyboardEvent` with a meaningful `endCoordinates` property (instead of emitting a null as of current implementation). This change standardizes the `keyboardDidHide` keyboard event emission across both iOS and Android, which makes it easier for developers to use the API.
In particular, the semantics of `endCoordinates` emitted during the `keyboardDidHide` event on Android will match nicely with semantics of the same event emitted on iOS:
- `screenY` will be height of the screen, as that the keyboard has collapsed to the bottom of the screen
- `screenX` will be 0, as the keyboard will always be flush to the sides of the screen
- `height` will be 0, as the keyboard has fully collapsed
- `width` will be the width of the screen, as the keyboard will always extend to the width of the screen
Also, the flowtypes for `KeyboardEvent` have been further improved and are more explicit to better highlight the different object shapes (see `IOSKeyboardEvent` and `AndroidKeyboardEvent`) depending on the platform.
## Changelog
[Android] [Added] - Return endCoordinates for keyboardDidHide keyboard event
Pull Request resolved: https://github.com/facebook/react-native/pull/24947
Differential Revision: D15413441
Pulled By: cpojer
fbshipit-source-id: aa3998542b7068e9852467038f57310355018379
Summary:
This diff ensures the method scheduler.constraintSurfaceLayout is executed before the JS run application start.
This is necessary to properly set the pointScaleFactor for the Root before running JS.
This is a workaround to fix a bug when the pointScaleFactor changes over time for the rootShadowNode. The bug is easily reproducible when rendering the "fabric" indicator on Fabric screens. During the first render of a Fabric screen this method was called before "JS run application" starts, and the Fabric indicator was render correctly.
Beacuse of timing of measure APIS, the second time a Fabric screen is rendered the method is called after the "JS run application process started", as a consecuence the Fabric indicator is not rendered correctlly (the pointScaleFactor is incorrectly assigned into the layout metrics of the Fabric indicator text).
We still need to analyze why the pointScaleFactor is not correctly assigned when it is set after the "JS run application process started", but this will be part of another diff.
Reviewed By: shergin
Differential Revision: D15303554
fbshipit-source-id: 7d985cefee20fd40dbe04166c1a1358b3f3ddc85
Summary: Remove dependency on ReactRootView in ReactInstanceManager by creating a rough interface (ReactRoot) so that either a Surface or ReactRootView can be attached to ReactInstanceManager.
Reviewed By: ejanzer, mdvacca
Differential Revision: D14158890
fbshipit-source-id: b7ab4654b1e0ef8343230a3c15023653a7f23a4b
Summary: Flesh out how surface will be used with a flag in ReactRootView
Reviewed By: mdvacca
Differential Revision: D14112897
fbshipit-source-id: adf6078048dbf83452d3523f0530a4d6dca7b3e8
Summary: This diff changes the onMeasure method of the RootView to optimize the amount of times we call updateRootLayoutSpecs in Fabric
Reviewed By: shergin
Differential Revision: D14198155
fbshipit-source-id: ff2deee04540899c25d4e38b0bd93333f74c6ace
Summary:
Instead of using measure specs to set certain yoga properties on the root node (like max width, auto width, specific width), use yoga's calculateLayout(width, height) instead. The measure specs will be stored in the shadow node. This allows us to remove duplicated code that processes the measure specs and allows us to remove other logic like the enableLayoutCalculation() method.
This diff also removes MeasureSpecProvider. MeasureSpecProvider was originally introduced to pass previously measured view measure specs to the initial creation of the root shadow node, but it turns out that this is unnecessary. We can update the root layout specs from the root view instead.
Reviewed By: mdvacca
Differential Revision: D9729744
fbshipit-source-id: 79b0b27ca879758f5dc3fc7cc8a0d38856a6cc79
Summary: SizeMonitoringFrameLayout was used to set layout contraints on the root shadow node when the native view's size changes. Since then, we introduced ways for the root node to use proper layout contraints using the root view's measure specs, which provides more accurate constraints for the root node, so SizeMonitoringFrameLayout is no longer needed. This ends up making a lot of UIManagerModule's method signatures cleaner
Reviewed By: mdvacca
Differential Revision: D9565720
fbshipit-source-id: c569cd15991a09987cc01e89612dc9193ad99b45
Summary:
React Native's minSdkVersion is 16, or we support Android versions 16 (Jelly Bean) and above. But in the code we have many checks if Android is Jelly Bean or newer, which are unnecessary. This PR removes unnecessary Android version checks, also uses Android version names instead of numbers.
[Android] [Changes] - remove unnecessary Android version checks
Pull Request resolved: https://github.com/facebook/react-native/pull/23277
Differential Revision: D13955909
Pulled By: cpojer
fbshipit-source-id: 6b1caa5ef4fe42273d3c69a6617fff140a697b5c
Summary:
We are writing a native navigation library. When there are more than a dozen `ReactRootView` in the application, the application will be very stuck. It doesn't improve much, even if we remove the invisible ReactRootView from the View hierarchy temporarily.
After debugging, we found that we were not able to remove `OnGlobalLayoutListener` in `onDetachedFromWindow` because this listener was added twice. One is in `onAttachedToWindow` and the other is in `attachToReactInstanceManager`.
We added the following code by inheriting ReactRootView.
```java
Override
protected void onAttachedToWindow() {
try {
Method method = ReactRootView.class.getDeclaredMethod("getCustomGlobalLayoutListener");
method.setAccessible(true);
// remove duplicated GlobalLayoutListener
if (Build.VERSION.SDK_INT >= Build.VERSION_CODES.JELLY_BEAN) {
getViewTreeObserver().removeOnGlobalLayoutListener((ViewTreeObserver.OnGlobalLayoutListener) method.invoke(this));
} else {
getViewTreeObserver().removeGlobalOnLayoutListener((ViewTreeObserver.OnGlobalLayoutListener) method.invoke(this));
}
} catch (NoSuchMethodException e) {
e.printStackTrace();
} catch (IllegalAccessException e) {
e.printStackTrace();
} catch (InvocationTargetException e) {
e.printStackTrace();
}
super.onAttachedToWindow();
}
```
It turns out that there is no pressure even if opening hundreds of pages that contain ReactRootView (remove ReactRootView from the View hierarchy temporarily in the invisible pages).
Changelog:
----------
[Android] [Fixed] - Remove duplicated `OnGlobalLayoutListener` to improve performance.
Pull Request resolved: https://github.com/facebook/react-native/pull/22849
Differential Revision: D13682818
Pulled By: hramos
fbshipit-source-id: 1f1bc8cd5fc4e01a1bc2f2a4784822b4eae92fdc
Summary:
It works great on iOS, and mostly works on Android, and is now OTA'able as part of the screen config! Haven't done template view yet. One remaining issue:
Layout is borked on Android. I'm guessing the issue has to do with the timing of setting the constraints in `updateRootLayoutSpecs` and calling `mBinding.startSurface` which actually builds the shadow tree. If I try to call `updateRootLayoutSpecs` earlier, it just crashes immediately. Here's the layout it spits out, which clearly has -440 for the x of 420006, which is the RCTText component, causing it to get cut off on the left of the screen:
```
updateLayoutMountItem for reactTag: 420006 x: -440, y: -13, width: 931, height: 78
updateLayoutMountItem for reactTag: 420010 x: 26, y: 79, width: 0, height: 1651
updateLayoutMountItem for reactTag: 420012 x: 0, y: 26, width: 0, height: 158
updateLayoutMountItem for reactTag: 420016 x: 0, y: 210, width: 454, height: 454
updateLayoutMountItem for reactTag: 420018 x: 454, y: 210, width: 455, height: 454
updateLayoutMountItem for reactTag: 420022 x: 0, y: 690, width: 454, height: 454
updateLayoutMountItem for reactTag: 420024 x: 454, y: 690, width: 455, height: 454
updateLayoutMountItem for reactTag: 420028 x: 0, y: 1171, width: 454, height: 454
updateLayoutMountItem for reactTag: 420030 x: 454, y: 1171, width: 455, height: 454
updateLayoutMountItem for reactTag: 420032 x: 0, y: 1651, width: 0, height: 0
```
Reviewed By: mdvacca
Differential Revision: D12813192
fbshipit-source-id: 450d646af4883ff25184141721351da67b091b7c
Summary: There hasn't been any reports of overflow issues since the last issue was fixed in v183 (July 30th). Let's remove this flag
Reviewed By: yungsters
Differential Revision: D10020006
fbshipit-source-id: 593c7d2c2bef5f0bd60d5de1941cd58f7fd6ccc7
Summary: This change drops the year from the copyright headers and the LICENSE file.
Reviewed By: yungsters
Differential Revision: D9727774
fbshipit-source-id: df4fc1e4390733fe774b1a160dd41b4a3d83302a
Summary: This diff updates the size of RootShadowNode and re-render RN views when the Size of the Android React View changes
Reviewed By: achen1
Differential Revision: D9173758
fbshipit-source-id: 7cc6bbfb646025c3ec1773ab041eb9207623af71
Summary:
Set clipChildren to false by default in ReactRootView.java so that Overflow: visible/hidden will work for the root view.
Moved sDefaultOverflowHidden from ReactViewGroup.java to ViewProps.java so that it can be used in both ReactViewGroup.java and ReactRootView.java.
Reviewed By: mdvacca
Differential Revision: D8727140
fbshipit-source-id: b593bed63e479cdbd22e4a025b936e6aeb28fc8c
Joshua Gross
David Vacca
ajpaulingalls
Samuel Susla
fabriziobertoglio1987
Ramanpreet Nara
Emily Janzer
Andres Suarez
Oleksandr Melnykov
Tim Yung
Emily Janzer
nossbigg
Luna Wei
Dulmandakh
Listen
Spencer Ahrens
Andrew Chen (Eng)
Héctor Ramos
Jiaqi Wu
