Summary:
Pull Request resolved: https://github.com/facebook/react-native/pull/32443
This diff removes all the custom Gradle machinery to build the native code and delegates to AGP
the triggering of the `ndk-build` command. This means that the native build will be now invoked
with the `:ReactAndroid:externalNativeBuild<Variant>` task.
An important thing to notice is that that task will always run, and will delegate to Make the
compilation avoidance. If you invoke the task twice, the second time it will be significantly faster.
On my machine this takes ~6/7 mins the first time, and 30 seconds the second time.
There are some gotchas that are worth noting:
* The native build will run on every build now. Given the complexity of our native build graph,
even with an up-to-date build, Make will still take ~30 seconds on my machine to analyse all the
targets and mention that there is no work to be done. I believe this could be impactful for local
development experience. The mitigation I found was to apply an `abiFilter` to build only the ABI
of the target device (e.g. arm64 for a real device and so on).
This reduces the native build to ~10 seconds.
* All the change to the `react-native-gradle-plugin` source will cause the Gradle tasks to be
considered invalid. Therefore they will re-extract the header files inside the folders that are
used by Make to compile, triggering a near-full rebuild. This can be a bit painful when building
locally, if you plan to edit react-native-gradle-plugin and relaunch
rn-tester (seems to be like an edge case scenario but worth pointing out). The mitigation here
would be to invoke the tasks like
```
gw :packages:rn-tester:android:app:installHermesDebug -x prepareBoost -x prepareLibevent -x prepareGlog \
-x prepareJSC -x extractNativeDependencies -x generateCodegenArtifactsFromSchema \
-x generateCodegenSchemaFromJavaScript
```
Changelog:
[Internal] [Changed] - Refactor Extract Headers and JNI from AARs to an internal task
Reviewed By: ShikaSD
Differential Revision: D31683721
fbshipit-source-id: fa85793c567796f4e04751e10503717a88cb0620
Summary:
Change usages of `exports.foo` and `module.exports.foo` to just `foo`.
This will fix future errors when Flow no longer allows reads from `exports`/`module.exports`.
Changelog: [Internal]
Reviewed By: bradzacher
Differential Revision: D32009490
fbshipit-source-id: bb609b37ba948b44c3877f8dbbfa9137b963586b
Summary:
The debug message stays `Building RNTester with Fabric enabled.` even if we set `fabric_enabled = false`
This pull request changes the message like below when `fabric_enabled = false`
`Building RNTester with Fabric disabled.`
## 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] - Fix the debug message for Fabric in rn-tester
Pull Request resolved: https://github.com/facebook/react-native/pull/32502
Reviewed By: sota000
Differential Revision: D32027858
Pulled By: lunaleaps
fbshipit-source-id: baf5301581e354f588830acd12a5e8171799b1ca
Summary:
Make `generate-specs-cli.js` use named arguments.
Updated all `generate-specs-cli.js` callsites to make use of named arguments.
Changelog: [Internal]
Reviewed By: sota000
Differential Revision: D31908041
fbshipit-source-id: f2cb5967db3c3b847e1095e35e8d5d21585be27b
Summary:
When calculating the offset range, we assume the first item is always at offset zero position and skipped that (as the smallerOffset is zero). However, this may not be the case in some situations. This diff changes the range measuring loop to always start from the first item.
Changelog:
[Android][Fixed] - Do NOT skip the first child view in the scroll view group when measuring the lower and upper bounds for snapping.
Reviewed By: mdvacca
Differential Revision: D31887086
fbshipit-source-id: af7221a621b2719d057afa6b64aa91c94ac01295
Summary:
Publishing a new version of the Gradle plugin as we need to distribute some
changes before releasing the playbook to the public.
Changelog:
[Internal] [Changed] - Bump react-native-gradle-plugin to 0.0.2
Reviewed By: ShikaSD
Differential Revision: D31957686
fbshipit-source-id: ef6742a205f2b568714c349dcb6481af33254b9b
Summary:
I'm removing the `src/test` folder from the npm bundle as it's not really
needed for the sake of distributing the gradle plugin.
Changelog:
[Internal] [Changed] - Do not publish tests inside the react-native-gradle-plugin npm package
Reviewed By: ShikaSD
Differential Revision: D31956449
fbshipit-source-id: b84138d24a7c51c44b92ebfc84f35afb3668dee5
Summary:
D31809012 (f7e4c07c84) introduced a condition where codegen files weren't generated in a correct order so the build fails in `yarn test-ios` if it was a first time to run the command. So it broke ci/circleci: test_ios_unit_hermes.
In this diff
Pull Request resolved: https://github.com/facebook/react-native/pull/32480
Changelog: [intermal]
Reviewed By: cortinico
Differential Revision: D31953580
fbshipit-source-id: db854d6cfed8167dc4aae2667d379738bc261cfe
Summary:
While working updating the codeden Makefile template for Android, I've removed
the `libreact_debug` and `libreact_render_debug` dependencies as they were unused in a
simple turbomodule. Turns out that `:ReactAndroid` is depending on having those dependencies.
Moving the build file to use AGP APIs, triggers this scenario and is making `ReactAndroid`
failing to build (see the build status for https://github.com/facebook/react-native/pull/32443).
I'm updating the codegen Makefile template to reintrodce the two libraries + I've update the prebuilt
makefile (used in the playbook) to include `react_debug` prebuilts as they were missing.
Changelog:
[Internal] [Changed] - Re-add libreact_debug to codegen makefile and add prebuilts for them
Reviewed By: ShikaSD
Differential Revision: D31900675
fbshipit-source-id: ff188c0498a0dca4a951a548a580ca8dd0674782
Summary:
ReactSwitch component is crashing on Android when it is initialised with both a backgroundColor and thumbColor, `style={{ backgroundColor: "anyColor" }} thumbColor="anyColor"`, due to IllegalCastException.
When setting a background color, BaseViewManagerDelegate is calling `setBackgroundColor` which replaces the background drawable with a ColorDrawale, hence [this line](72ea0e111f/ReactAndroid/src/main/java/com/facebook/react/views/switchview/ReactSwitch.java (L68)) fails.
Instead, given the ripple effect needs to be preserved, one should initialise a RippleDrawable using the current background drawable and set it as the background of the switch.
Given the RippleDrawable should be preserved, overriding the `setBackgroundColor` seemed the sensible thing to do.
## Changelog
[Android] [Fixed] - Fix crash when a Switch is initialised with both backgroundColor and thumbColor.
Pull Request resolved: https://github.com/facebook/react-native/pull/32468
Test Plan:
### Setup:
Initialise an empty React Native project. Add a switch component:
`<Switch
style={{backgroundColor: 'red'}}
thumbColor={'https://github.com/facebook/react-native/issues/356'}
/>`
Run the project `yarn android`
### Current state (RN 65+):
Red screen will show highlighting an IllegalCastException.
<img src="https://user-images.githubusercontent.com/4354327/138616661-3ba1370c-6a2b-48c2-ba70-b99415a4256f.png" width="200"/>
### With fix:
- The component is expected to have a red background.
- When pressed a ripple effect shows inside the backgrounds bounding box.
- Business as usual otherwise.
`backgroundColor` with `thumbColor`:
![backgroundColor + thumbColor](https://user-images.githubusercontent.com/4354327/138615603-141660d2-a5cd-49d7-aa5e-9c93ebc6d680.gif)
Just `thumbColor`:
![Screen Recording 2021-10-25 at 00 23 57](https://user-images.githubusercontent.com/4354327/138615658-baa380dd-2cbb-4d0f-a25e-a003ef67c977.gif)
Reviewed By: ShikaSD
Differential Revision: D31895690
Pulled By: cortinico
fbshipit-source-id: 60af16de7db61440ccfbf11d67a3d945dd90b562
Summary:
Some of the prerendered surfaces rely on Android context being present to have correct theming (e.g. for platform colors) and measurements of platform components. This change uses context provided to initialize the surface as themed context before view is attached.
This way it is possible to configure theming with `ContextThemeWrapper` the same way as Litho does it for prerendering. The assumption is that any kind of customization done through Android theme will be applied from prerendering entry point as well.
Changelog: [Internal] - Use context from surface for prerendering
Reviewed By: mdvacca
Differential Revision: D31906091
fbshipit-source-id: 344fc96eb2f85ba5b762bee64d1a29443b3fd1d3
Summary:
I missed to push this change in the previous diff D31809012 (f7e4c07c84).
This diff adds a check so that only when fabric is enabled it include the React-graphic dependency.
Changelog: [internal]
Reviewed By: fkgozali
Differential Revision: D31919354
fbshipit-source-id: 0b4e7f489155f868cdf58bec3f61f309470ca0c6
Summary:
In this diff, it moves the codegen output location out of node_modules and to build/generated/ios folder.
A temp pod spec will be created so that those files will be included in the Xcode project.
Changelog: [Internal]
Reviewed By: hramos, cortinico
Differential Revision: D31809012
fbshipit-source-id: ba1c884c8024306ba0fd2102837b7dbebc6e18ac
Summary:
Several desktop forks (`react-native-macos`, `react-native-windows`, `react-native-web`) support mouse events, and while the stock Pressable component has the ability to support mouse events, it seems we aren't forwarding some props properly from Pressable -> Pressability.
Pressability will calculate onMouseEnter / onMouseLeave event handlers based on the `onHoverIn/onHoverOut` callbacks passed into PressabilityConfig.
ad0d4534a7/Libraries/Pressability/Pressability.js (L552)
However, Pressable does not pass take in onHoverIn/onHoverOut props to pass to PressabilityConfig, so we can't take advantage of this functionality. This change should simply address that by passing the props through.
## 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
-->
[General] [Fixed] - Pressabel not passing hover props and event handlers to PressabilityConfig
Pull Request resolved: https://github.com/facebook/react-native/pull/32405
Test Plan: I fixed a similar issue in `react-native-macos` that I am now trying to contribute back upstream. https://github.com/microsoft/react-native-macos/pull/855
Reviewed By: yungsters
Differential Revision: D31667737
Pulled By: sota000
fbshipit-source-id: f0bbe48302703bb2c45280d2afeec8d7a4586b6a
Summary:
As per the XMLHttpRequest specification [1], getAllResponseHeaders() should return a string of headers with lowercased names and sorted by their uppercase representation, with each header ending with '\r\n'.
[1] https://xhr.spec.whatwg.org/#the-getallresponseheaders()-method
## Changelog
[General] [Changed] XMLHttpRequest.getAllResponseHeaders() now returns headers with names lowercased and sorted in ascending order, as per specification
Pull Request resolved: https://github.com/facebook/react-native/pull/32363
Test Plan:
Test derived from Web Platform Test repository:
https://github.com/web-platform-tests/wpt/tree/master/xhr
Reviewed By: yungsters
Differential Revision: D31626217
Pulled By: sota000
fbshipit-source-id: 299d005facbe1c15b8cda5eed6750db75addca80
Summary:
Adding an error check to make debugging easier when codegen fails when invalid library_type option is passed.
Changelog: [internal]
Reviewed By: hramos
Differential Revision: D31907880
fbshipit-source-id: c1ffa6bbd7b3e4faede88da2ee8d3378fa086780
Summary:
Currently, some filtering of generated files are done in multiple places (e.g. generate-specs-cli.js, react_native_pods.rb, etc). I am introducing this arguments so that only needed files are generated for components and modules.
Changelog: [internal]
Reviewed By: hramos, cortinico
Differential Revision: D31878098
fbshipit-source-id: d2dc8f51ea14a5d0ba1548bd481814220c9ae3a2
Summary: Changelog: [Internal] Remove github action to close upgrade-related issues. We will be moving these issue to react-native repo and using this tag
Reviewed By: cortinico
Differential Revision: D31837195
fbshipit-source-id: 51fdd6efd88c0ebd5c504730406b217c8dbb3a88
Summary:
Fixes https://github.com/facebook/react-native/issues/27052
Since react-native 0.62, the `devDisabledIn${buildType}` syntax has stopped working for apps with multiple `productFlavors`. This PR adds the `disableDevForVariant` lambda to allow dev mode to be disabled for different variants.
## 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] - Fix devDisabledIn not working with multiple productFlavors
Pull Request resolved: https://github.com/facebook/react-native/pull/30606
Test Plan:
I added the following log into `react.gradle` and ran the Android build for my app:
```
println("devEnabled: ${targetName}, ${devEnabled}")
```
```
# build.gradle
project.ext.react = [
entryFile: "index.android.js",
enableHermes: true, // clean and rebuild if changing
bundleInLive: true,
disableDevForVariant: {
def variant -> variant.name.toLowerCase().contains('release') || variant.name.toLowerCase().contains('live')
},
]
...
flavorDimensions 'branding'
productFlavors {
cve {
dimension 'branding'
}
whce {
dimension 'branding'
}
}
```
Console output:
```
Reading env from: env/cve/live
devEnabled: CveDebug, true
devEnabled: CveRelease, false
devEnabled: CveLive, false
devEnabled: WhceDebug, true
devEnabled: WhceRelease, false
devEnabled: WhceLive, false
```
Reviewed By: cortinico
Differential Revision: D31649977
Pulled By: ShikaSD
fbshipit-source-id: 520734314f4bca7608b8dca67c7c5ce0be6d31a5
Summary:
Fixes https://github.com/facebook/react-native/issues/32434: RCTLogBox window is orphaned, covering entire screen.
After this change, the logbox window once again is removed from the screen.
## Changelog
Some third-party SDKs may hold references to created UIWindow, UIViewController, or UIView objects. Doing so means that the current code's `hide` method that releases the reference to the UIWindow in LogBox will not cause the window to be dealloc'd, and thus instead it will remain on the screen. This change explicitly hides the LogBox window when the reference is released, so that even if some other SDK holds onto the window it will still be taken off the screen.
<!-- 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] - 32434
Pull Request resolved: https://github.com/facebook/react-native/pull/32435
Test Plan:
1. Use console.warn to generate a yellow warning message in log box. Also install a third-party SDK that holds onto a reference to UIWindow -- for example the Facebook SDK, the Data Dog SDK, or any number of other SDKs that use `swizzling` to intercept calls like `viewDidAppear:`.
2. click the log
3. tap "dismiss"
4. try to tap anywhere
5. Use Xcode view debugger to inspect the UI state
## Expected
The app still responds to the touch.
In Xcode, there is not an extra UIWindow covering the screen
Reviewed By: philIip
Differential Revision: D31794242
Pulled By: sshic
fbshipit-source-id: 28aa247b3ed3fd60b8e7c2ed7d0606cbf5c42408
Summary: Changelog: [Internal] Fix up basic styling of dark mode for examples. Individual examples still may be broken and will be fixed up individually
Reviewed By: yungsters
Differential Revision: D31710790
fbshipit-source-id: 6ca4fb8a6238f38ff484ec91518057b243ba1d7b
Summary:
Single and composite animations were handled inconsistently in AnimatedMock. Also added a guard to prevent callbacks from triggering additional animations, since we had a test-scenario that did exactly that.
Changelog:
[General][Fixed] - Composite animations will now be ran immediately when the app is in testing mode
Reviewed By: yungsters
Differential Revision: D31826967
fbshipit-source-id: a6416b42e227fe79f5c3a55a9c51beb8451874f8
Summary:
Bumping the codegen to a new version to ship a change to the Android template.
Changelog:
[General] [Changed] - Bump react-native-codegen to 0.0.9
Reviewed By: hramos, mdvacca
Differential Revision: D31762139
fbshipit-source-id: 71a96210a3577c12ff3c9f9013c6e72adf4a0ecb
Summary:
Deprecates `prop-types` from React Native.
Existing use cases will be presented with a warning to migrate to the newly published `deprecated-react-native-prop-types` module.
In a subsequent release, these will be removed from React Native.
Changelog:
[General][Changed] - Accessing `Image.propTypes`, `Text.propTypes`, `TextInput.propTypes`, `ColorPropType`, `EdgeInsetsPropType`, `PointPropType`, or `ViewPropTypes` now emits a deprecation warning.
Reviewed By: kacieb
Differential Revision: D29019309
fbshipit-source-id: 21e518e588fa05c498cc75ba81f69cfa8a9d0613
Summary:
Fix the `scripts/update-ruby.sh` so it always use the correct [bundle config](https://bundler.io/man/bundle-config.1.html#DESCRIPTION). In the current version it wasn't using the correct configuration inside the `template/` directory, resulting in incorrect platform for `template/Gemfile.lock`.
While at that, update the gems to their latest version:
- ethon 0.14.0 -> 0.15.0
- json 0.5.1 -> 0.6.0
- zeitwerk 2.4.2 -> 2.5.1
- bundler 2.2.28 -> 2.2.29
## Changelog
No changelog
Pull Request resolved: https://github.com/facebook/react-native/pull/32456
Test Plan:
Run `bump-oss-version.js` and see `template/Gemfile.lock` lists `ruby` as the `PLATFORM` (no diff in that line).
## References
- e18cf90d71 (r58230816)
Reviewed By: yungsters
Differential Revision: D31841524
Pulled By: charlesbdudley
fbshipit-source-id: 695c245fcb344c866afed45f747e04233e5c91e4
Summary:
Adds utility script which crawls through a React Native app's Node dependencies and, for each compatible library, generates the relevant native code artifacts.
This script is for development purposes, and is not hooked into the existing codegen integration by design.
Changelog: [Internal]
Reviewed By: sota000
Differential Revision: D28915433
fbshipit-source-id: de36d3e1dc0e11aad3ca55cea5e6731db09c5377
Summary:
The `ReactScrollView.java` constructor is using `ReactContext` instead of `Context`, which is inconsistent to the horizontal scroll view. This is the result from D3863966 (2cf2fdbc04) when an OSS issue needs to be addressed. That issue and all call sites to use the `ReactContext` are deprecated now.
Revert this back to use `Context` to be less restrictive.
Changelog:
[Android][Fixed] - Revert `ReactScrollView` to use `Context` instead of `ReactContext` in the constructor to be less restrictive.
Reviewed By: javache
Differential Revision: D31819799
fbshipit-source-id: 3f00d64850aebd2e20615033b2e1f1c721fed37e
Summary:
Changelog: [internal]
This change had to be reverted because it caused some issues internally at Facebook. After we solved those we re-applied the changes but this also has issues in OSS apps. Specifically, the option in Metro to inject modules to execute before the EntryPoint (in this case `InitializeCore`) doesn't work if those modules aren't part of the bundle in the first place.
This reverts again so we can think about a long-term solution for this.
Original commit changeset: ba145a30ead5
Reviewed By: ShikaSD
Differential Revision: D31825736
fbshipit-source-id: 393b3d5d2a726951a9ac386dc2b1c19ef4a916a5
Summary:
Colocate the surface ID generate with the code that consumes it. This allows us to re-use this method in other event emitter locations.
Changelog: [Android][Changed] Add helper to get surfaceId for event dispatching
Reviewed By: philIip
Differential Revision: D31651881
fbshipit-source-id: 109e189f90261d3ba0077ffa519c3d12a9111439
Summary:
Changelog:
[Android][Changed] RootView's onChildStartedNativeGesture now takes the child view as its first argument
Reviewed By: philIip
Differential Revision: D31399515
fbshipit-source-id: b9438f6118e604a04799ef67d0b46303a06d6434
Summary:
i saw this a lot in the codebase, it's not optimal bc we're using two selectors when we only need one.
fastmod --extensions m,mm '\[\[(.*) alloc] init]' '[${1} new]' --dir xplat/js/react-native-github/*
i manually updated the callsites that this codemod couldn't handle (e.g., where there were more than one of these instances in a single line)
Changelog: [Internal]
Reviewed By: RSNara
Differential Revision: D31776561
fbshipit-source-id: 1b16da240e8a79b54da67383d548921b82b05a9f
Summary:
Adds the `glob` type definition via `flow-typed` so that the `combine-js-to-schema-cli.js` script does not need a Flow suppression.
Changelog:
[Internal]
Reviewed By: kacieb
Differential Revision: D31800809
fbshipit-source-id: 3e01a096d2153639a619a7152bd06332df57f06b
Summary:
Many have reported about the misguiding error `Fatal Exception: java.lang.UnsatisfiedLinkError: couldn't find DSO to load: libhermes.so` even though they don't use Hermes (for example issues https://github.com/facebook/react-native/issues/26075#25923).
**The current code does not handle errors correctly when loading JSC or Hermes in `ReactInstanceManagerBuilder`**.
**ReactAndroid/src/main/java/com/facebook/react/ReactInstanceManagerBuilder.java:**
```java
try {
return new HermesExecutorFactory();
} catch (UnsatisfiedLinkError hermesE) {
// We never get here because "new HermesExecutorFactory()" does not throw an exception!
hermesE.printStackTrace();
throw jscE;
}
```
In Java, when an exception is thrown in static block, it will be RuntimeException and it can't be caught. For example the exception from `SoLoader.loadLibrary` can't be caught and it will crash the app.
**ReactAndroid/src/main/java/com/facebook/hermes/reactexecutor/HermesExecutor.java:**
```java
static {
// Exception from this code block will be RuntimeException and it can't be caught!
SoLoader.loadLibrary("hermes");
try {
SoLoader.loadLibrary("hermes-executor-debug");
mode_ = "Debug";
} catch (UnsatisfiedLinkError e) {
SoLoader.loadLibrary("hermes-executor-release");
mode_ = "Release";
}
}
```
This PR fixes the code so that the original exception from failed JSC loading is not swallowed. It does not fix the original issue why JSC loading is failing with some devices, but it can be really helpful to know what the real error is. For example Firebase Crashlytics shows wrong stack trace with current code.
I'm sure that this fix could have been written better. It feels wrong to import `JSCExecutor` and `HermesExecutor` in `ReactInstanceManagerBuilder.java`. However, the main point of this PR is to give the idea what is wrong with the current code.
## 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] - Fix error handling when loading JSC or Hermes
Pull Request resolved: https://github.com/facebook/react-native/pull/30749
Test Plan:
* from this PR, modify `ReactAndroid/src/main/java/com/facebook/react/jscexecutor/JSCExecutor.java` so that JSC loading will fail:
```java
// original
SoLoader.loadLibrary("jscexecutor");
// changed
SoLoader.loadLibrary("jscexecutor-does-not-exist");
```
* Run `rn-tester` app
* Check from Logcat that the app crashed with correct exception and stacktrace. It should **not** be `java.lang.UnsatisfiedLinkError: couldn't find DSO to load: libhermes.so`
Tested with Hermes
```
SoLoader.loadLibrary("hermes-executor-test");
```
Got this one in logcat
```
09-24 20:12:39.552 6412 6455 E AndroidRuntime: java.lang.UnsatisfiedLinkError: couldn't find DSO to load: libhermes-executor-test.so
```
Reviewed By: cortinico
Differential Revision: D30346032
Pulled By: sota000
fbshipit-source-id: 09b032a9e471af233b7ac90b571c311952ab6342
Summary:
Changelog: [Internal]
these are returning bools for some reason even though no one is using the returned value. changing them to return void
Reviewed By: RSNara
Differential Revision: D31594241
fbshipit-source-id: 04c115b573b74996eaf2fef631eedb12c6734ea8
Summary:
Changelog: [Internal]
`start` and `stop` are already part of `RCTSurfaceProtocol` which is a public protocol conformance, we don't need to add these to this header, it's just extra work for the compiler
Reviewed By: RSNara
Differential Revision: D31776005
fbshipit-source-id: d89ad4dbe35e1b67cfa750c6414c40f9b4fc7f24