Summary:
* When I build the latest bishop (using buck), I get a failure and a suggestion:
```
Rule //fbobjc/Apps/Bishop:BishopPackage (//fbobjc/Apps/Bishop:BishopPlatform) FAILED because Command failed with exit code 1.
command: [/bin/bash, -e, /data/sandcastle/boxes/fbsource/buck-out/tmp/genrule-15268472800256890202.sh]
stderr: The target: fbsource//fbobjc/Apps/Bishop:BishopShareExtension is an App Extension, and must only use APIs available in App Extensions. However, it contains dependencies that do not set 'extension_api_only = True' in their BUCK files:
- //xplat/js/react-native-github:RCTSettingsApple
This command may be able automatically to fix this issue:
arc buildozer 'set extension_api_only True' //xplat/js/react-native-github:RCTSettingsApple
```
* This diff is the result of running that command
Reviewed By: christophpurrer
Differential Revision: D41121880
fbshipit-source-id: b1ecd077270d7f3909739364237af32dbc617f52
Summary:
This PR backport two fixes we did in 0.71 to unblock the release process:
* the change in `publish-npm` is needed because of the introduction of .strict() from https://github.com/facebook/react-native/commit/4f3ca8facfc41f27169c49ad8b3a28d3022d5f0a
* the removal of the other script (added originally here https://github.com/facebook/react-native/commit/e4b5d3eec9bae44c46795c4be097b31cb48593ae) is because:
1) that step is not needed anymore (we don't publish/upload hermes artifacts to the GH release)
2) by the time this job gets run the release crew has already setup the GH release
3) the logic for the versioning was broken and even on the 0.71-rc pipeline it was tagging stuff as 1000.0.0
## Changelog
<!-- Help reviewers and the release process by writing your own changelog entry. For an example, see:
https://reactnative.dev/contributing/changelogs-in-pull-requests
-->
[Internal] [Fixed] - Fix release scripts for "release" pipeline scenario
Pull Request resolved: https://github.com/facebook/react-native/pull/35258
Test Plan: The fact that 0.71-rc0 was released is the ✅ for this.
Reviewed By: jacdebug
Differential Revision: D41120888
Pulled By: cipolleschi
fbshipit-source-id: 06d108f0659ad1db53c6324fe1d735f52c34a3c5
Summary:
Pull Request resolved: https://github.com/facebook/react-native/pull/35226
Changelog: [Internal]
This adds scaffolding for the C++ side of NativePerformanceObserver module.
Thanks to christophpurrer for helping set this up, as this is the first one of this kind inside core/OSS.
Reviewed By: rubennorte
Differential Revision: D41028555
fbshipit-source-id: 4acf0e71a254a42044cbbe5f94f40938342c6aa2
Summary:
Pull Request resolved: https://github.com/facebook/react-native/pull/35106
This Diff remove the assert on the initializer in the EventEmitter in place of a more idiomatic check when the method is invoked.
It aims to solve an internal error and promotes best practices for the iOS platform.
## Changelog:
[iOS][Changed] Refactor RCTEventEmitter initialization
Reviewed By: sammy-SC
Differential Revision: D40762253
fbshipit-source-id: 83d26616ce147914948e536e9e4b1010758fb690
Summary:
This PR updates he cache strategy for the `checkout_with_cache command`.
The previous strategy was using three keys in descending priority order to restore from the cache:
* `<< parameters.checkout_base_cache_key >>-{{ arch }}-{{ .Branch }}-{{ .Revision }}`
* `<< parameters.checkout_base_cache_key >>-{{ arch }}-{{ .Branch }}`
*`<< parameters.checkout_base_cache_key >>-{{ arch }}`
When it saves, it always saves using the first key.
The restore works as it follows:
1. It tries to restore the cache using the first key
2. If it fails, it checks whether there is a cache hit for a key that matches the second key as a pattern
3. If it fails, it checks whether there is a cache hit for a key that matches the third pattern
4. Otherwise, it is a cache miss.
This does not work well. Imagine that you submit some code in commit `xxxx` for branch `abc`.
The CI run the first time, it misses all the three keys and checks out the code normally.
Then, it stores the checked out code in the `checkout_key-abc-xxxx` key.
Then, you submit a commit `yyyy` in the same branch.
The CI starts, it tries with the key `checkout_key-abc-yyyy` but it misses
It tries with the key `checkout_key-abc` and it finds the cache for `checkout_key-abc-xxxx` and it restores it, forgetting about your changes.
While doing the release, we created a tag in a commit X. Then we manually had to remove it, but the CI had a cached version of .git with the tag for
the `0.71-stable` branch. And the release failed because the tag was already existing.
### Why this should work
With this solution, we are going to cache using the commit. If there is no cache for a specific commit, it will be a miss. It won't try to be smart and
retrieve the code from previous caches.
This should prevent stale caches and if we manually remove a tag, and then we do a new commit, it should work.
This is a good trade-off that allows to pay the checkout cost only for the first batch of jobs of the pipeline.
**NOTE:** This still won't work if we don't do a new commit.
## Changelog
[General] [Fixed] - Change Cache strategy to avoid cache bumps in Release
Pull Request resolved: https://github.com/facebook/react-native/pull/35259
Test Plan: 1. CircleCI must be green
Reviewed By: jacdebug
Differential Revision: D41120895
Pulled By: cipolleschi
fbshipit-source-id: 2b45da01803197dbe4a25a313a9dfc53a976d096
Summary:
Changelog:
[Internal][Changed] - In order to make Dev Loading View cross platform, refactoring the accessary show methods.
Reviewed By: cortinico
Differential Revision: D41029102
fbshipit-source-id: 475949548fe98217e61d6cf64accbbdc0fb0f1c5
Summary:
Pull Request resolved: https://github.com/facebook/react-native/pull/35163
# What
This diff contains all the changes from D40333083 (https://github.com/facebook/react-native/commit/0fac9817df403e31d8256befe52409c948614706) (aka https://github.com/facebook/react-native/pull/34964), **except** the change to `setUpReactDevTools.js`, which actually uses the new files.
# Why
* We want to ship the Buck, C++, etc. changes before the JavaScript changes that depend on those files.
* Otherwise, apps can fail at startup with the message:
```
`TurboModuleRegistry.getEnforcing(...): '${name}' could not be found. ` +
'Verify that a module by this name is registered in the native binary.',
```
* Note that this only occurs if you are using a previously-built version of the C++, Obj C, etc. files in RN, but a more recent version of the JavaScript files. If you are building from matching sources, this does not occur.
* After a few days, we can land the JS files.
## Changelog
Changelog
[General][Added] Add, but don't use, DevTools Settings Manager.
Reviewed By: NickGerleman
Differential Revision: D40873390
fbshipit-source-id: c7bac6ae65f85666b8616443db278ebb175b691b
Summary:
Pull Request resolved: https://github.com/facebook/react-native/pull/35212
A previous change - https://github.com/facebook/react-native/pull/34011 - already fixed basic usage of <react/bridging/.../ imports.
However that change was only tailored towards the usage of: <react/bridging/CallbackWrapper.h>
Any other header besides <react/bridging/CallbackWrapper.h> from <react/bridging/... can't be imported at this time in Xcode ... ... which is bad.
For C++ TurboModules we need to be able to access *any* <react/bridging/...> header via the React-Codegen CocoaPod.
Hence adding bridging now as a sub-spec to the ReactCommon CocoaPod
Changelog: [Internal]
Reviewed By: cipolleschi
Differential Revision: D41057878
fbshipit-source-id: 83c117bc5252d84dd419cdb72f145f65547d23b2
Summary:
## Changelog
<!-- Help reviewers and the release process by writing your own changelog entry. For an example, see:
https://reactnative.dev/contributing/changelogs-in-pull-requests
-->
[Android] [Fixed] - fixed typo in template build.gradle
Pull Request resolved: https://github.com/facebook/react-native/pull/35209
Reviewed By: cipolleschi
Differential Revision: D41080347
Pulled By: cortinico
fbshipit-source-id: d2e36f232f798a636f98843edbc5651730125fc4
Summary:
Noticed these types could be improved based on the tests added in D40979066 (https://github.com/facebook/react-native/commit/e81c98c842380d8b72c1dc8d4a6e64f760e2a58c).
Changelog: [Android][Fixed] Corrected Nullable annotations for parameters and return values in TurboModules codegen
Reviewed By: mdvacca, cipolleschi
Differential Revision: D40979940
fbshipit-source-id: cfc352a9e7eb9f59e2cce3d7da110a9a8d32db4b
Summary:
This is the 2nd iteration of D39468561 (https://github.com/facebook/react-native/commit/4d1a56813c7975c848631de331e424587054fd57). We first check if the `BridgeDevSupportManager` can be used before we return the `PerfTestDevSupportManager`. This is to avoid a breakage of Quantum that happened on the previous diff.
Add a `DevSupportManager` that can be used for performance testing. This `DevSupportManager` allows the inspector connection to be established, but leaves everything else disabled.
Previously, if Developer Support was enabled on a release build, the application would present an error as it unsuccessfully attempted to use the bridge dev support manager.
This is now conceptually the new flow for deciding what DevSupportManager to choose.
```
if (developerSupportEnabled) {
if (full support available) {
use full support (i.e. bridge)
} else {
use profiling-only support (i.e. perftest)
}
} else {
disable dev support
}
```
The first attempt at this diff erroneously used this logic:
```
if (developerSupportEnabled) {
if (debug build) {
use full support (i.e. bridge)
} else {
use profiling-only support (i.e. perftest)
}
} else {
disable dev support
}
```
So now we are always checking to see if the `BridgeDevSupportManager` is available, and if it is, we use it.
(`enableOnCrease` indicates the development mode setting: https://www.internalfb.com/code/fbsource/[6b8a941fdf2a0fd58d9db36f5a59fa5fb53ad2df]/xplat/js/react-native-github/ReactAndroid/src/main/java/com/facebook/react/ReactInstanceManager.java?lines=259)
Changelog: [internal]
Reviewed By: makovkastar
Differential Revision: D40948243
fbshipit-source-id: 50c6b6b905f5b9c5b5ecc090b36edbd6090ea774
Summary:
[Changelog][Internal]
This adds module specs for the native part of PerformanceObserver, as well as the interaction logic vs the NativePerformanceObserver API.
See https://fb.quip.com/MdqgAk1Eb2dV for more detail.
Reviewed By: rubennorte
Differential Revision: D40897006
fbshipit-source-id: 77475f21dad9ee9dbe15df5a989eb08d314e6db2
Summary:
The method that would restore these files was removed in https://github.com/facebook/react-native/pull/34846. The action performed by `saveFilesToRestore` is no longer necessary (thanks kelset for pointing this out).
Changelog: [internal]
Reviewed By: cortinico
Differential Revision: D41003911
fbshipit-source-id: bbc057ac450e7f134c4664173291ca56c18f1b17
Summary:
The project.android block is unnecessary and contains a wrong path. Let's use the default path which is `./android`.
## Changelog
[Internal] - Remove project.android block from rn-tester's CLI config
Pull Request resolved: https://github.com/facebook/react-native/pull/35197
Test Plan: Tested locally + will wait for a CI run
Reviewed By: robhogan
Differential Revision: D41013193
Pulled By: cortinico
fbshipit-source-id: 5c76c9b571b23ad71a23a8f3f05a9acb4d6e20b9
Summary:
We accidentally added `-DCMAKE_BUILD_TYPE=Debug` for Hermes which is causing
a really slow engine execution. This fixes it and keep the -DCMAKE_BUILD_TYPE
aligned with the one used on Android.
Changelog:
[iOS] [Fixed] - Correctly set -DCMAKE_BUILD_TYPE for Hermes on iOS
Reviewed By: hramos, cipolleschi
Differential Revision: D40980934
fbshipit-source-id: 30f0455990a911b51eb109bebca8272507302650
Summary:
[Changelog][Internal]
This is to prevent an accidental use of PerformanceObserver while it's still a work in progress.
PerformanceObserver is also moved into a dedicated folder, `Libraries/WebPerformance`.
Reviewed By: rubennorte
Differential Revision: D40978749
fbshipit-source-id: 09645a95bed72902870ebc00c1b4a3d81ab4c829
Summary:
## Changelog:
[Android] [Fixed] - Fix a bug that returns a random number from callback argument `timeRemaining` of `idleCallbacks` registered by `requestIdleCallbacks`.
`global.performance.now()` returns a time from an unknown origin, while `frameTime` is epoch time. Thus, `global.performance.now() - frameTime` is just a random number rather than an elapsed time from the frame start time.
As a simple solution, I would suggest to use `Date.now()`.
Its resolution is lower than `frameTime` due to the security issue in JavaScript VM. So, it loses some precision, but, at least, it is not wrong.
Reviewed By: javache
Differential Revision: D40941461
fbshipit-source-id: b0094e181b97a844d133a9268fe48cd8c8fadfda
Antoine Doubovetzky
generatedunixname89002005287564
Mike Vitousek
Robert Balicki
Lorenzo Sciandra
Ruslan Shestopalyuk
Riccardo Cipolleschi
Riccardo Cipolleschi
Christoph Purrer
Stanley Shi
fabriziobertoglio1987
Ramanpreet Nara
Gabriel Donadel Dall'Agnol
Arushi Kesarwani
Dmitry Rykun
Samuel Susla
Thibault Malbranche
Pieter De Baets
Michael Anthony Leon
Mitch Powell
Ruslan Lesiutin
Sam Zhou
Pranav Yadav
Abraham Romero
Héctor Ramos
Nicola Corti
Nicola Corti
Yoonseok Ko
