Summary:
## Rationale
For every 1 call to RCTNetworking.sendRequest, we execute 6 calls to RCTNetworking.addListener. This is followed by at least one call to RCTNetworking.removeListeners. Aside from incrementing and decrementing the `_listeners` integer, these two methods accomplish nothing else: RCTNetworking doesn't implement the `startObserving` and `stopObserving` methods.
This diff makes RCTEventEmitter dispatch events without looking at the listeners integer. In the future, this will allow us to stop making these ~8 unnecessary NativeModule calls for every Network request we send.
Changelog: [Internal]
Reviewed By: fkgozali
Differential Revision: D24272560
fbshipit-source-id: 7996eba5abfa4669a89c43a3ffa536c0faa214a8
Summary:
With Xcode 12 being the latest, Detox 15.x has issues - in particular, it means that if you try to `yarn install` the dependencies for the repo, you'll be greeted by this error:
```bash
error /<stuff>/react-native/node_modules/detox: Command failed.
Exit code: 1
Command: node scripts/postinstall.js
Arguments:
Directory: /<stuff>/react-native/node_modules/detox
Output:
/<stuff>/Library/Detox/ios/5824c837515589f21c08f09b716a6eda088aa31f was found, but could not find Detox.framework inside it. This means that the Detox framework build process was interrupted.
deleting /<stuff>/Library/Detox/ios/5824c837515589f21c08f09b716a6eda088aa31f and trying to rebuild.
Extracting Detox sources...
Building Detox.framework from /<stuff>/Developer/OSS/react-native/node_modules/detox/ios_src...
child_process.js:637
throw err;
^
```
With the {emoji:1f44d} of hramos & alloy I've prep'd up a small defensive PR that can be quickly merged before cutting 0.64, that bumps the version of Detox from 15.4.4 to the highest version available within the reach of "no breaking changes" in changelog.
The main reason why with 16.x this error doesn't happen is that from [16.0.0](https://github.com/wix/Detox/releases/tag/16.0.0):
> Detox now comes as a prebuilt framework on iOS, thus lowering npm install times and saving some build issues that happen due to unexpected Xcode setups.
It would have been better to update directly to latest (at the time of writing 17.7.1) but there are at least two versions that had changelogs that seem to involve bigger changes:
* https://github.com/wix/Detox/releases/tag/17.4.7
* https://github.com/wix/Detox/releases/tag/16.8.0
Hopefully CI will will show that the bump doesn't break any test �
## 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] [Changed] - Bumped Detox in the repo to 16.7.2 for Xcode 12 compatibility
Pull Request resolved: https://github.com/facebook/react-native/pull/30084
Test Plan: Running yarn in the main repo with Node 14 & Xcode 12, without this change, will cause the error copy-pasted above. After upgrading to this version, the error disappear.
Reviewed By: cpojer
Differential Revision: D24293226
Pulled By: hramos
fbshipit-source-id: 2b4d23b033be621274966262ec19200bee44df58
Summary: Changelog: [RN][iOS] Allow gate to be set for Fabric logging from the React Module
Reviewed By: fkgozali
Differential Revision: D24256546
fbshipit-source-id: 7b290efb9abd3035559f743e6e5b6701e02053e1
Summary:
This diff ended up being a bit more complicated than I anticipated, since the source files in `ReactInternal` were depending on `RCTEventDispatcher`. I made the following changes:
1. Make `RCTEventDispatcher` a `protocol`, keep it in `ReactInternal`.
2. Rename the `RCTEventDispatcher` NativeModule to `RCTEventDispatcherModule`, make it conform to the `RCTEventEmitter` `protocol`, and move it to `CoreModules`.
3. Where necessary, replace categories of `RCTEventDispatcher` with functions.
Changelog:
[iOS][Added] - Make RCTEventDispatcher TurboModule-comaptible
Reviewed By: fkgozali
Differential Revision: D18439488
fbshipit-source-id: b3da15c29459fddf884519f33b0c3b8c036b5539
Summary:
There are a few places where we have JNI table ref leaks, and more places where we can aggressively clean up smart pointers immediately instead of waiting for them to be cleaned up at some later point.
In theory these smart pointers should be cleaned up immediately, but in cases where many components are being measured at once, the JNI table could grow until all measure calls are done. In extreme cases this
could cause a crash, which I want to avoid. At the very least, freeing memory more aggressively in this case can't hurt.
Changelog: [Internal]
Reviewed By: shergin
Differential Revision: D24293775
fbshipit-source-id: 159741ba955e5a6fe02caf6e65d1e4d6d4afadee
Summary:
This should fix a race condition uncovered by ASAN build that results in `m_shutdown` being referenced after destruction.
This mirrors a similar fix made in react-native-macos: https://github.com/microsoft/react-native-macos/commit/ea7767a211e5d1e5b18171c79404553b712539c7
Changelog: [Internal]
Reviewed By: shergin
Differential Revision: D24263156
fbshipit-source-id: 5fb1e7053c107dfb7050242fce8fcf0435e5592c
Summary: This completes the Picker stack. Use a handwritten view config to avoid calling `requireNativeComponent` in Bridgeless mode.
Differential Revision: D23663596
fbshipit-source-id: 5d0811014fd6f66956803a1db5fee8fd1119d5bc
Summary:
This builds on the last diff to remove type a type union from Picker. This diff focuses on Picker internals.
Changelog: [JS] Remove type union in PickeriOS/PickerNativeComponent
Reviewed By: sammy-SC
Differential Revision: D24254615
fbshipit-source-id: f788a2e123135c1e8b9909870c40f53b2dea0227
Summary:
Flow type unions don't play well with Fabric components. This diff removes a union in `Picker.js` and fixes all the flow errors.
Before this diff, all these surfaces would crash with the new Fabric Picker impl, because the impl asserts that this field is a string.
Reviewed By: sammy-SC
Differential Revision: D24236317
fbshipit-source-id: 6e646c84fcd16658aaabe5e93507f5f33b346a65
Summary:
This adds support for a controlled `<Picker/>` component .
Changelog: [iOS][Fabric] Fabric Picker support
Reviewed By: sammy-SC
Differential Revision: D24005475
fbshipit-source-id: c50e9918f74f6ef5cdfbfe67cb6c132c12d64916
Summary:
This adds support for `<Picker style={}/>` prop for text styling. It reuses most of conversion logic from BaseText. This means that it actually supports more styles than Paper Picker supported. (Paper picker only supported ~4 styles, this supports everything that Text supports, so 10+ styles).
The only tricky thing is that Picker supports multiple ways of setting text color. Both
<Picker
itemStyle={{color: '#008BD0'}} >
<Picker.Item label="Java" value="java" />
<Picker.Item label="JavaScript" value="js" />
</Picker>
and
<Picker>
<Picker.Item label="Java" value="java" color={'#008BD0'} />
<Picker.Item label="JavaScript" value="js" />
</Picker>
technically work in Paper. I've decided to maintain this behaviour (since there's lots of product code callsites to both options).
Changelog: [iOS][Fabric] Fabric Picker support
Reviewed By: sammy-SC
Differential Revision: D23980319
fbshipit-source-id: e469a837e28af0ad97cf0e171df26ee19adff3ab
Summary:
This builds on previous diff to properly parse `<Picker.Item/>` into a cpp struct.
This fixes the dummy text and text color TODOs.
Changelog: [iOS][Fabric] Fabric Picker support
Reviewed By: sammy-SC
Differential Revision: D23964557
fbshipit-source-id: f42c6c9cf410bfc5e66ff078645b6378548481de
Summary:
This is a starting point for the handwritten Fabric Picker component. It is incomplete, and needs to be landed with the rest of the stack above it.
In general, this creates a new `ComponentView`, `ComponentDescriptor`, `ShadowNode`, `Props` and a few other boilerplate classes for Picker. A bunch of the logic in `ComponentView` was copied over from the Paper `RCTPicker` and `RCTPickerManager`.
What works in this diff:
- A `<Picker>` with items can be created in JS, and a corresponding `UIPicker` is created in native with placeholder text, default styling and the correct amount of items
What doesn't work yet (implemented in later diffs):
- Parsing items to use correct text and styling in native
- Events/commands
Reviewed By: sammy-SC
Differential Revision: D23941821
fbshipit-source-id: e049ca6004757fbd1361985644d5dbb8f53e1ce6
Summary:
When switching between non-Fabric and Fabric screens, I believe that `initializeEventListenerForUIManagerType` is not always being called on the NativeAnimatedNodesManager if
`NativeAnimatedModule.initializeLifecycleEventListenersForViewTag` is being called before the NativeAnimatedNodesManager ivar has been set. This should occur very infrequently, but logs
indicate that it /does/ happen in some marginal cases. Protecting against these cases should be trivial, by using the `getNodesManager` method which is responsible for returning a nodes manager or creating a new one.
The existing uses of the `NativeAnimatedNodesManager` ivar also occur on different threads and we were not protecting against this, so I'm changing it to an atomic. It's very likely that
the inconsistency issues in the past were caused not by ordering errors, but thread races.
Changelog: [Internal]
Reviewed By: fkgozali
Differential Revision: D24267118
fbshipit-source-id: 68abbff7ef3d0b2ecc9aa9977165663ad9447ab8
Summary:
RNTester has some imports left over that operate on directory traversal, assuming it is contained within react-native sources. This change makes these imports relative to react-native, enabling usage outside of the RN repo.
Relates to https://github.com/microsoft/react-native-windows/issues/6210
## Changelog
[Internal] [Fixed] - Avoid File Structure Assumptions in RNTester Imports
Pull Request resolved: https://github.com/facebook/react-native/pull/30141
Test Plan: Validated we can bundle and flow-check both iOS + Android
Reviewed By: cpojer
Differential Revision: D24259628
Pulled By: appden
fbshipit-source-id: 0c21b5d354b01785d8402599da3b0a5be81b4c6d
Summary:
shergin added a new assert to dimensions calculations in D24038911 (https://github.com/facebook/react-native/commit/e85372298109abf258d5154e2a28bc6496fb9529). This crashes in bridgeless mode b/c no bridge.
This diff uses `turboModuleRegistry` as a bridge workaround. The registry is a [weak property](https://www.fburl.com/diffusion/sunv3bx9) so retain cycles shouldn't be an issue here, let me know if that's incorrect.
As a nice bonus, this fixes dynamic font sizing in bridgeless mode. It was broken before since it relied on the bridge.
Changelog: [Internal]
Reviewed By: ejanzer
Differential Revision: D24256338
fbshipit-source-id: 141d36239ac6a6e9e87ca96eea7aeec56729095d
Summary:
Changelog: [Internal]
If `uiTouch.view` isn't managed by Fabric, go up the view hierarchy to find first view that is.
This is important for host views wrapped by <Touchable>, otherwise touches are not delivered to the component.
Reviewed By: yungsters
Differential Revision: D24219223
fbshipit-source-id: 17b4e3460735371553ee0d30b41776a977f8eafb
Summary:
This PR attempts to fix issue https://github.com/facebook/react-native/issues/28278 and https://github.com/facebook/react-native/issues/29525
On Crashlytics, the following error occurs in file `RCTWebSocketModule.m` at method `-[RCTWebSocketModule webSocket:didFailWithError:]` when a nil value is inserted into a dictionary as a value.
```
Fatal Exception: NSInvalidArgumentException
*** -[__NSPlaceholderDictionary initWithObjects:forKeys:count:]: attempt to insert nil object from objects[0]
```
This PR is following the suggestion of this: https://github.com/facebook/react-native/issues/28278#issuecomment-597461650 and it replaces the values of any property if it is nil. In detail:
- it converts `error` to empty NSString if the original value is nil
- it converts `socketID` to a NSNumber object, which stores `-1` if the original value is nil
## Changelog
[iOS] [Fixed] - A crash in WebSocket module
Pull Request resolved: https://github.com/facebook/react-native/pull/30066
Test Plan: We were not able to reproduce the crash, but the report itself provided enough information to find a solution for this issue.
Reviewed By: shergin
Differential Revision: D24241147
Pulled By: sammy-SC
fbshipit-source-id: d5d632b49ca77b5d8be8b9c32358bef68f17d30a
Summary:
This makes a few change to the template xcodeproj
- Update to recommended settings for Xcode 12
- Remove the main.jsbundle file that doesn't exists, this file is included automatically in the app bundle and doesn't need to be in xcode (it won't even be at that path).
- Add -e to the bundle images script. It seems like Xcode doesn't fail the build on script errors anymore so this makes it so the build stop. If it doesn't stop the app will launch and crash because the bundle doesn't exist (note this is only if running the app in release mode since in debug we don't generate the bundle).
## Changelog
[iOS] [Added] - Update template to xcode 12
Pull Request resolved: https://github.com/facebook/react-native/pull/30150
Test Plan: Made the same changes in an app and made sure it ran well.
Reviewed By: cpojer
Differential Revision: D24237481
Pulled By: sammy-SC
fbshipit-source-id: 2b06a24c510c423eb45a1a840ea365b64506321b
Summary:
When taking a heap timeline, Hermes wasn't showing any data until the timeline
was written to disk and then reloaded.
Turns out we were missing support for two events:
* `HeapProfiler.lastSeenObjectId`: This event reports the most recently
allocated object ID. Used to know when objects were allocated in the
timeline.
* `HeapProfiler.heapStatsUpdate`: Report how many objects and bytes
exist for a "time fragment", represented by a fragment index. Later updates
to the same index can decrease the amount of live memory, which show up
as grey spikes instead of blue spikes
Previously, we only supported these by writing out to a file, and they didn't work
with a "live" profiling view. To fix this, I changed the periodic sampling thread
to instead be a periodic flush of a sample every few allocations. The performance
impact is tucked away only when profiling is turned on, and it's very non-invasive to
the rest of the GC. The flush calls a callback with the relevant information if the
inspector is on, and the inspector sends a message back to the browser.
Changelog: [Internal] Fix for Hermes heap timeline profiling
Reviewed By: neildhar
Differential Revision: D23993363
fbshipit-source-id: 8e0b571130cbb7e839dfb009b04f584f5179085d
Summary:
## Previously
- When TurboModules system was on, we'd only log events from the TurboModules system.
- When TurboModules system was off, we'd only log events from the NativeModule system.
This ultimately gives us less data to analyze both systems in production.
## Changes in This Diff
When perf. logging is on, we'll log events from both systems. Each QPL event now include an annotation of which system the event is coming from. Concretely, this will allow us to see how much of the NativeModule system is being exercised in the TurboModule test group.
Changelog: [Internal]
Reviewed By: hramos
Differential Revision: D24232594
fbshipit-source-id: 7dff57bd74fc7ef744d3e06ff174304f25790456
Summary:
Index adjustment is tricky. Seems more reliable to adjust each immediate mutation, and then immediately adjust delayed mutations based on it, rinse and repeat.
Previously it was possible to construct examples where the UI would get into a weird state because index adjustment caused items to be inserted in the wrong location.
Changelog: [Internal]
Reviewed By: kacieb
Differential Revision: D24232926
fbshipit-source-id: f8c445213528c2d2aedacf3e0c73c5bbeb62bc3d
Summary:
Changelog: [internal]
RCTViewComponent view should have multipleTouchEnabled set to YES.
Paper has it enabled by default as well.
Reviewed By: JoshuaGross, yungsters
Differential Revision: D24219076
fbshipit-source-id: 74c18632457147b944a8abbacdbecb57e57f62ef
Summary:
We're planning a fix in Flow (D24112595) that uncovers some existing errors that were
previously suppressed. Adding these annotations prevents them from surfacing during the
deploy of the fix.
Changelog: [Internal]
Reviewed By: dsainati1
Differential Revision: D24147994
fbshipit-source-id: fef59a9427da6db79d4824e39768dd5ad0a8d1a3
Summary:
Types First and the various esproposal flags are on by default as of 0.135
Clean up flowconfigs that have to be in sync with the main one
Changelog: [Internal]
Reviewed By: jimmy623
Differential Revision: D24221995
fbshipit-source-id: 5da2867e2f013ad729a436ada192ecab690cb413
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:
Use the ObjC++ native modules files generated by `react-native-codegen` for the FBReactNativeSpec library.
People working with the core native modules may now re-generate these ObjC++ files using `USE_CODEGEN=1 pod install` or `scripts/generate-native-modules-specs.sh`.
Changelog: [Internal]
Reviewed By: RSNara
Differential Revision: D22249978
fbshipit-source-id: a12db23defacdba36099a24ce0b8475c8973bd6d
Summary:
This PR removes unused `qs` package (query string parser) from the main `package.json` file.
I was not able to find any explicit usage of this package in the repository. Looking at the commit which introduced this package it is not clear why it has been added, query string is not processed in any way.
I have also run ` yarn update-lock` to clean up lock after latest dependencies bumps. There are no `qs` package entry removal in yarn lock, because it is already required by `request` package.
Refs: https://github.com/facebook/react-native/commit/b4785e514430dc3ba45ed6d136ec63574be88e26
## 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] [Removed] - remove unused `qs` dependency
Pull Request resolved: https://github.com/facebook/react-native/pull/30125
Test Plan: This change should not affect the code base.
Reviewed By: appden
Differential Revision: D24184372
Pulled By: cpojer
fbshipit-source-id: 4f159f3c554c48ed21839370c8370848df181046
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:
dispatchDraw, dispatchGetDisplayList, updateDisplayListIfDirty, recreateChildDisplayList, etc, can all crash internally for a variety of reasons and it can be very tricky to track down the root cause. This isn't a fix, this just adds extra logging to hopefully make debugging easier.
Changelog: [Internal]
Reviewed By: shergin
Differential Revision: D24166149
fbshipit-source-id: 1bbaf34a92a9bcac5a594a25522c66e6e0cc80ca
Summary:
According to Node's documentation: https://nodejs.org/api/util.html#util_util_inspect_object_showhidden_depth_colors
> The util.inspect() method returns a string representation of object that is intended for debugging. **The output of util.inspect may change at any time and should not be depended upon programmatically.**
Therefore, this diff switches over our RN Codegen snapshot tests to use a `JSON.stringify` call, followed by a replace of `"` with `'`. This gets the job done without compromising readability.
**Question:** Why do we not use `"`? A: Jest escapes all `"` in the snapshots, which makes reading the snapshots/using them in the console harder.
Changelog: [Internal]
Reviewed By: hramos
Differential Revision: D24157056
fbshipit-source-id: 2f1aa2df28ac3ed4aa17bcdbcd23846ddbf804cf
Summary:
This PR changes ReadableNativeMap.getNullableValue to return null if key not found, instead of throwing exception. This matches method signature and nullable annotation.
## Changelog
[Android] [Changed] - fix ReadableNativeMap.getNullableValue to match signature
Pull Request resolved: https://github.com/facebook/react-native/pull/30121
Test Plan: RNTester app builds and runs as expected, and getNullableValue will return null instead of throwing exception.
Reviewed By: JoshuaGross
Differential Revision: D24164302
Pulled By: fkgozali
fbshipit-source-id: 572c1d4ae5fd493aa0018c2df1dfc7fc91cb4b6b
Summary:
Fix ReadableArray annotations, because these methods throw ArrayIndexOutOfBoundsException instead of null if index is not found.
## Changelog
[Android] [Changed] - fix ReadableArray null annotations. Possibly breaking change for Kotlin apps.
Pull Request resolved: https://github.com/facebook/react-native/pull/30122
Test Plan: RNTester app builds and runs as expected, and show correct type in when used with Kotlin code.
Reviewed By: JoshuaGross
Differential Revision: D24164326
Pulled By: fkgozali
fbshipit-source-id: 0c3f8fa9accbd32cc71c50befe9330e5201643f6
Summary:
The captured `instance_` will result in a call to JVM code to release the reference when the lambda is destroyed. However, Hermes may destroy the lambda in a non-JVM background thread.
Since `instance_` is a member of the class, and `this` is already captured, there is no need to capture it.
Changelog: [Internal]
Reviewed By: RSNara
Differential Revision: D24141179
fbshipit-source-id: e3b268485ed229ee933281a80e5dbe2feba55719
Summary:
# Changelog:
[Internal][Changed] - Close the global perf logger when we create the scoped performance logger and absorb the global perf logger into the scoped logger
Reviewed By: rubennorte
Differential Revision: D23898621
fbshipit-source-id: 0d9bfd95175e660c0bd0bc74974126f92fe409b2
Summary:
Our Babel plugin that strips irrelevant platform-specific code does not currently know how to deal with early return statements.
This minor change to a couple call sites enables the output bundle for Android to omit code specific to iOS.
Changelog:
[Internal]
Reviewed By: ejanzer
Differential Revision: D24139922
fbshipit-source-id: 467c0c38dd45679b889a8a94d85f73fd969e5b36
Summary:
Remove extraneous newlines before and after structs, before copyright headers, and other locations.
Changelog: [Internal]
Reviewed By: fkgozali
Differential Revision: D24117137
fbshipit-source-id: 194996019b4cadef9239a78334f31c0bc89e3901
Summary:
react-native has moved more code to be flow-strict. Part of this move involved adding custom flow-typed definitions for some third-party dependencies where not previously present. These definitions are not currently published, which prevents type-checking some parts of react-native externally.
This change publishes these type definitions as part of the react-native NPM package. Impact to package size is pretty negligible.
## Changelog
[Internal] [Fixed] - Publish Custom Flow Library Definitions
Pull Request resolved: https://github.com/facebook/react-native/pull/30102
Reviewed By: TheSavior
Differential Revision: D24099502
Pulled By: cpojer
fbshipit-source-id: bf3942aa973df4cfc2843ec3e7dbec82623f0447
Summary:
The UserFlow API takes in a FlowId as a parameter on all of its calls. However, this type is currently not exported so consumers cannot store a generated FlowId without creating their own matching type.
Changelog: [Internal]
Reviewed By: dmitry-voronkevich
Differential Revision: D24115983
fbshipit-source-id: 63afbca01343a31aa60042db3693a96a6b854939
Ramanpreet Nara
Lorenzo Sciandra
Paige Sun
Lulu Wu
Joshua Gross
Scott Kyle
Peter Argany
Nick Gerleman
simek
Samuel Susla
Mark Sinkovics
Janic Duplessis
Riley Dulin
Luna Wei
Panagiotis Vekris
George Zahariev
Héctor Ramos
Valentin Shergin
Dulmandakh
Christoph Nakazawa
Neil Dhar
Tim Yung
Chris Dinh
