Summary:
^
No need for a constructor here.
Reviewed By: fkgozali
Differential Revision: D15056802
fbshipit-source-id: 32a2c541e5c1fa6336b7f6f2131772198740626c
Summary:
This is a quick change to avoid having "Collect Test Results" fail on `test_android` when unit tests have not run. If the raw Buck results are not available, avoid running the buck-to-junit conversion tool.
You can find an example of the situation we want to avoid here: https://circleci.com/gh/facebook/react-native/101397?utm_campaign=vcs-integration-link&utm_medium=referral&utm_source=github-build-link
## Changelog
[Internal] [Changed] - Do not fail test collection step when unit tests fail
Pull Request resolved: https://github.com/facebook/react-native/pull/25621
Test Plan:
Verified on macOS + bash, the conditional will fail if all-results-raw.xml does not exist.
Circle CI:
The `test_android` job as a whole failed (as expected, master is broken), but Collect Test Results remained green even in the absence of a Buck results file.
https://pxl.cl/Fbjl
Reviewed By: fkgozali
Differential Revision: D16230885
Pulled By: hramos
fbshipit-source-id: 73a8290bc4679f2e0dfa88d1e7f71fed279c2104
Summary: For better compatibility re: https://github.com/facebook/react-native/pull/25393, this target should just use `RCTTypeSafety`
Reviewed By: PeteTheHeat
Differential Revision: D16210888
fbshipit-source-id: 6a55d631453cc420909247a7d5a64379587225b7
Summary:
**Final Goal**
Currently unit test by default unmock everything and dependencies have to be explicitely mocked. The goal is to mock everything by default and whitelist (unmock) explicitely.
**This Diff**
Fixes the typing in the mock and add unit tests
Differential Revision: D16198318
fbshipit-source-id: d2bb71e9c940f4aeb48f8ac92bf3c5444ebc2025
Summary:
I believe there's a mismatch between the type definitions and the expected prop in Android for `TextInput`'s `autoComplete` prop.
* Android is expecting `autoComplete`.
* JS types are expecting `autoCompleteType`.
* Latest documentation documents `autoCompleteType`.
Prop added here: https://github.com/facebook/react-native/commit/179d490607620a988a53aacb01031ed300d4ac66
This change updates the JS types to match what Android is expecting (`autoComplete`). Can update documentation if this is the approach we'd prefer (rather than updating Android to expect `autoCompleteType`).
## Changelog
[Javascript] [Fixed] - Update types for `TextInput`'s `autoComplete` prop.
Pull Request resolved: https://github.com/facebook/react-native/pull/25549
Test Plan:
Before:
* Pass invalid value to `TextInput`'s `autoComplete` prop, see no type errors on JS side, and Android blows up with:
```sh
Invalid autocomplete option: foobar
updateViewProp
ViewManagersPropertyCache.java:95
setProperty
ViewManagerPropertyUpdater.java:132
updateProps
ViewManagerPropertyUpdater.java:51
updateProperties
ViewManager.java:37
```
After:
* Pass invalid value to `TextInput`'s `autoComplete` prop, see PropType warning for `autoComplete` prop.
Differential Revision: D16220809
Pulled By: mdvacca
fbshipit-source-id: e25e198cbcbe721c8d71f069bba293856bf5f36d
Summary:
Currently users have to use the global `react-native` command or run `yarn react-native run-x` which isn't very nice.
This PR adds `android` and `ios` scripts to `package.json` so users can run `yarn android` or `yarn ios` directly.
Pull Request resolved: https://github.com/facebook/react-native/pull/25618
Differential Revision: D16223229
Pulled By: cpojer
fbshipit-source-id: 69b082760ff2ee31ab0406251f9a50339fd227bf
Summary:
Following out internal communication we decided to change logic of current parser to consider types which extend 'TurboModule' instead of looking at default exports.
It also cassed for minor changes in testing logic and updating snapshots.
Reviewed By: rickhanlonii
Differential Revision: D16205707
fbshipit-source-id: c75cd4febf773ba5101e6b60ed1a921773246009
Summary:
In order to extend / use `DestructorThread.Destructor` outside of `com.facebook.jni`, we need access modifiers to be less strict:
- `Destructor#Destructor()`: package protected -> public
- `Destructor#destruct()`: package protected -> protected
This will enable Yoga to move from finalizers to `DestructorThread.Destructor` without having to buy into `HybridData` completely.
Reviewed By: cjhopman
Differential Revision: D16182362
fbshipit-source-id: ad616c403df8e7c1e3d751131cfb7a9cfe62cf24
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: Code generators for modules and components don't have a lot of in common so there's no point to keepo them in the same directory and mix their files.
Reviewed By: rickhanlonii
Differential Revision: D16183434
fbshipit-source-id: b6ee32c6b223e8d6e4bc843b2e111598bee94dc5
Summary:
Add generating cpp files following NativeModuleSpec.
We still need to figure out how to deal with module names, so I currently made in workable by extracting `Sample` from `SampleNativeModule` and filing into proper places, but it's a temporary solution.
Reviewed By: rickhanlonii
Differential Revision: D16181292
fbshipit-source-id: 0f50352d7610f99f0228045ae78309383d3e8bfa
Summary:
Add basic Native Module header generator
We still need to figure out how to deal with module names, so I currently made in workable by extracting `Sample` from `SampleNativeModule` and filing into proper places, but it's a temporary solution.
Reviewed By: rickhanlonii
Differential Revision: D16180561
fbshipit-source-id: bb27592d565ae439707d781fb2650f2bdb140146
Summary: I this diff I include basic codegen skeleton. It does not work yet, but I'd like to consider it as kind of playgroud for further changes. I have splited tests for component generation from test for module generation.
Reviewed By: rickhanlonii
Differential Revision: D16161909
fbshipit-source-id: a0ecb81898810fe6aa09f76238c433894c731b73
Summary:
Pull Request resolved: https://github.com/facebook/react-native/pull/25583
We now use CocoaPods for better maintainability.
Reviewed By: hramos
Differential Revision: D16193719
fbshipit-source-id: 26382f2da4eaba14a71771540b587fdc80b41108
Summary: For now, suppress this warning - they are harmless.
Reviewed By: mdvacca
Differential Revision: D16198994
fbshipit-source-id: b167d0e98bbc4abcd0461d50f01f364d8d560aec
Summary:
Following out internal communication we decided to change logic of current parser to consider types which extend 'TurboModule' instead of looking at default exports.
It also cassed for minor changes i n testing logic and updating snapshots.
Reviewed By: rickhanlonii
Differential Revision: D16200939
fbshipit-source-id: a21cbfc461647df2b9210c0eca4c2c95c285dfe1
Summary: Int32 and Float can be represented on native site in a different after cpp code generation. Inspired by component codegen.
Reviewed By: rickhanlonii
Differential Revision: D16201358
fbshipit-source-id: a5c6c449d69f162ad9cd2a998d57e9c65bc17706
Summary:
This diff fixes a crash caused by an IllegalStateException thrown from the `TextView.onEditorAction()`. This could happen if we don't return false from the `OnEditorActionListener.onEditorAction()` and Android will fallback to the default behaviour, which will try to search and focus the next/previous view in case of `EditorInfo.IME_ACTION_NEXT` or `EditorInfo.IME_ACTION_PREVIOUS` accordingly. Because ReactEditText prevents requesting focus from Android (`ReactEditText.requestFocus()` returns false), the following piece of code from `TextView.onEditorAction()` will crash the app:
```
} else if (actionCode == EditorInfo.IME_ACTION_PREVIOUS) {
View v = focusSearch(FOCUS_BACKWARD);
if (v != null) {
if (!v.requestFocus(FOCUS_BACKWARD)) {
throw new IllegalStateException("focus search returned a view "
+ "that wasn't able to take focus!");
}
}
return;
} else if (actionCode == EditorInfo.IME_ACTION_DONE) {
InputMethodManager imm = InputMethodManager.peekInstance();
if (imm != null && imm.isActive(this)) {
imm.hideSoftInputFromWindow(getWindowToken(), 0);
}
return;
}
```
To prevent this we have to catch `EditorInfo.IME_ACTION_NEXT` and `EditorInfo.IME_ACTION_PREVIOUS` inside `OnEditorActionListener.onEditorAction()` and prevent the default Android behaviour.
Reviewed By: mdvacca
Differential Revision: D16180306
fbshipit-source-id: 6118257c16a7a4a205ae05da671cd76d3a18d565
Summary:
We added a test to make sure button and accessibility actions would not have unwanted behavior. Additionally we added support for accessibility actions for all touchables. However we discovered that RCTTextView and possibly anything else that does not derive from RCTView does not support accessibility actions and need to be children off a component that does support it for it to not crash in ios. This became noticeable when TouchableWithoutFeedback only worked if text is a child of view on ios.
In a local branch we where able to modify RCTTextView to support accessibility actions and text no longer needed to be a child of view for it to work.
## Changelog
[General] [Added] - Button test with accessibility actions
[General] [Added] - Support for accessibility actions to all Touchables. With TouchableWithoutFeedback being a special case where text must be a child of view. (See AccessibilityExample.js for an example)
Pull Request resolved: https://github.com/facebook/react-native/pull/25582
Test Plan:
Test plan is testing in RNTester making sure the examples work
## Open Question
What would you say is the best practice for adding accessibility action support for all the components that do not extended from RCTView?
Reviewed By: cpojer
Differential Revision: D16192919
Pulled By: osdnk
fbshipit-source-id: 7d4e186ba1f30393f2b4d08a0e227b960f83586c
Summary: This fixes the following warning from appearing when you have a FlatList render a Modal where the content of the Modal also contains a FlatList: https://fburl.com/p953k985. Spencer addressed an issue similar to this in D7863625, but we still get a yellow box due to the fact that `scrollContext` still exists, but `this.context.virtualizedList` is null from this line in Modal.js https://fburl.com/nqc261a1.
Reviewed By: cpojer
Differential Revision: D16160666
fbshipit-source-id: ba222d3eef234f4c8c4c2bddbc71bec27df81e0a
Summary: TurboModule is now included by default, so no need to have a toggle in autolink script anymore.
Reviewed By: mdvacca
Differential Revision: D16173820
fbshipit-source-id: 215ce7d188ce526b5a71df881bdc992c1ef34b34
Summary:
This is the first step towards fixing https://github.com/facebook/react-native/issues/25349. These are the changes to the podspec to correctly update dependencies and build config that will cause any breaking change for users or libraries.
I am breaking these changes out from https://github.com/facebook/react-native/pull/25393 as suggested by fkgozali in https://github.com/facebook/react-native/pull/25393#issuecomment-508322884.
These are the changes:
- Made C++ headers in `React-Core` private by default so that ObjC files can import the module without failures.
- Reduced the number of `yoga` headers that are exposed for the same reason as above. As far as I can see this doesn't cause issues but we can find another solution if it does.
- Adding some missing dependencies to fix undefined symbols errors.
- Added `DoubleConversion` to `HEADER_SEARCH_PATHS` where it was missing.
## Changelog
[iOS] [Fixed] - Updated podspecs for improved compatibility with different install types.
Pull Request resolved: https://github.com/facebook/react-native/pull/25496
Test Plan:
Everything should work exactly as before. I have a branch on my [sample project](https://github.com/jtreanor/react-native-cocoapods-frameworks) here which points at this branch to show that it is still working `Podfile` to demonstrate this is fixed.
You can see that it works with these steps:
1. `git clone git@github.com:jtreanor/react-native-cocoapods-frameworks.git`
2. `git checkout podspec-updates`
3. `cd ios && pod install`
4. `cd .. && react-native run-ios`
The sample app will build and run successfully.
Reviewed By: mmmulani
Differential Revision: D16167346
Pulled By: fkgozali
fbshipit-source-id: 1917b2f8779cb172362a457fb3fce686c55056d3
Summary: That should help to fail early in situations when we lose critical class data members.
Reviewed By: sammy-SC
Differential Revision: D16179539
fbshipit-source-id: da73b81568c2f3657b9bc2bd1cc7ee6624e75626
Summary: This is a very similar mechanism that we use in UIManager and it should be eventually unified.
Reviewed By: sammy-SC
Differential Revision: D16179524
fbshipit-source-id: 7c8c45b7581ac4a1db3a773d62004ff368f18321
Summary: Adds Baseline start and end events to be handled later for instrumentation
Reviewed By: davidaurelio
Differential Revision: D16048790
fbshipit-source-id: 8409dbb633168753a7bf8fab20bc6551d113ddd6
Summary: Using layoutPassStart and LayoutPassEnd events instead of YGMarkerLayout for instrumentation
Reviewed By: davidaurelio
Differential Revision: D16048789
fbshipit-source-id: 041a35bc2cb1b7281ca83cf9d35041b4011cfeb9
Summary: We have exactly same code in classic RCTScrollView component. We need that to not interfere with iOS automagical behaviours.
Reviewed By: sammy-SC
Differential Revision: D16160846
fbshipit-source-id: 5b5affcc64abe5e3204f619216412f9da5b03b78
Summary:
We are running into some problems with the spread operator. The solution is to ask for a newer minimum babel version.
See https://github.com/babel/babel/issues/10179
Note: this doesn't actually change behavior. When creating a new app a new babel version will already be used. This simply ensures that the minimum version fulfills the requirements.
Reviewed By: axe-fb
Differential Revision: D16181111
fbshipit-source-id: f5207318b3a7bd9f092c0e64a8065d0f713012da
Summary:
This diff includes minor codestyle changes.
All these fixes was discussed with Rick.
Reviewed By: rickhanlonii
Differential Revision: D16169332
fbshipit-source-id: e561e2f2b116b6fdf8434c3dfc20c3e610d7ecad
Summary:
returning type of Bubbling and Direct Event should be always void of Promise (if async). Other situations shouldn't be permitted.
Reformated all cases when it the function wasn't void.
Reviewed By: rickhanlonii
Differential Revision: D16165962
fbshipit-source-id: 7c1377c3ed4bd54a431a13e5bcda4f7ec0adf4dc
Summary: This diff fixes a potential memory leak which can occur if an exception is thrown inside the try block and `outputStream.close()` is not called. By wrapping `outputStream.write(data)` inside try-with-resource we guarantee that outputStream will be closed regardless of whether the try statement completes normally or abruptly.
Reviewed By: sammy-SC
Differential Revision: D16148850
fbshipit-source-id: c5c0a78b36375857f6e717bb581e8686a4a94bb9
Summary:
I'd like to use `Modal`’s flow types in my application to make a reusable component.
## Changelog
[JavaScript] [Added] - Exported `Modal`’s types
Pull Request resolved: https://github.com/facebook/react-native/pull/25554
Test Plan: n/a
Differential Revision: D16180231
Pulled By: cpojer
fbshipit-source-id: 9cfd5163a187954783102bfe4d9b4d1dbc8c6e6d
Summary: It's pointless to handle non optional key if value has withDefaul so I'm adding a rule into parser preventing from such cases
Reviewed By: lunaleaps
Differential Revision: D16166709
fbshipit-source-id: 38cef522b217917a3a4886d857720932f2ebb475
Summary:
This removes the `--projectRoot` flag in `launchPackager.bat` as it was removed, fixing [this bundler not opening issue on cli repo](https://github.com/react-native-community/cli/issues/484) when `yarn react-native run-android` is run.
This is Windows related only so should be only tested on Windows.
## Changelog
[Internal] [Fixed] - Fixed Metro Bundler not opening when running `yarn react-native run-android`
Pull Request resolved: https://github.com/facebook/react-native/pull/25517
Test Plan:
1. Create a new repo using `react-native init`
2. Edit the `launchPackager.bat` file
3. Run `yarn react-native run-android`
Differential Revision: D16162108
Pulled By: cpojer
fbshipit-source-id: c12d9853ad49f00d56b9a8254a5b2c40a358cb2e
Summary: This diff changes a few things around so that a diff coming on top of this stack will be smaller. The aim of this change is to add a method `registerEntryPoint` which will allow a client to subscribe to updates for multiple bundles.
Reviewed By: gaearon
Differential Revision: D16131963
fbshipit-source-id: d460d6647b15a711021c7a3a51f52486a1aea535
Summary:
[Android] [Added] - Release underlying resources when JS instance is GC'ed on Android
D15826082 was reverted because it introduced a crash in Ads Manager for Android (see P67222724).
This diff fixes the crash and re-applies D15826082. The problem was that `jni::findClassStatic` in the destructor of BlobCollector.cpp couldn't find the Java class `com/facebook/react/modules/blob/BlobModule` and crashed the app.
JNI didn't seem to have access to the Java class loader probably because the destructor was called from a non-Java thread (https://our.intern.facebook.com/intern/wiki/Fbjni/environment-and-thread-management/?vitals_event=wiki_click_navigation_link#threads). The fix is to wrap the code in the destructor inside `ThreadScope::WithClassLoader `, which will allow to run code that has full access to Java as though you were running in a Java thread.
Reviewed By: shergin
Differential Revision: D16122059
fbshipit-source-id: 12f14fa4a58218242a482c2c3e2149bb6770e8ec
Summary: Accessibility is essential but seems we don't have any use of AccessibleShadowNode in Fabric.
Reviewed By: sammy-SC
Differential Revision: D16139595
fbshipit-source-id: a6aec6d22c4a6050d7ee5e6b21ef4ad04d80ffce
Summary:
The instrumentation header only needs the forward declarations for ostream, so
we can use just include `iosfwd`, as suggested by Riley in an earlier diff.
Reviewed By: kodafb
Differential Revision: D16152451
fbshipit-source-id: 2afbc40e623b180dfc5917fc8093ab15bf647968
David Vacca
Jeff Thomas
Héctor Ramos
Kevin Gozali
Florian Schoellhammer
Dan Gilbert
satyajit.happy
Michał Osadnik
David Aurelio
Ram N
Ben Cheng
Oleksandr Melnykov
Kevin Gonzales
Valentin Shergin
Blair Vanderhoof
James Treanor
Sidharth Guglani
Christoph Nakazawa
Cory Reed
thecodrr
Michał Pierzchała
Ashok Menon