Summary:
Before invoking TurboModule ObjC methods, we loop through all the arguments and transform them from `jsi::Value` to ObjC data structures. `jsi::Value`s that represent JS Objects get converted to `NSDictionary`s in ObjC. This isn't good enough because `NSDictionary` isn't typed. What we really need is a C/C++ struct that represents the type of the JS Object.
Therefore, for every argument of a TurboModule method that is a JS Object, this diff allows you to specify a method on `RCTCxxConvert` (which you have to declare and implement) that transforms that type of JS Object into something else. Basically, with the changes in this diff, we'll be able to transform JS Objects into C++ struct instances, which gives us type safety for JS Object method argumetns to TurboModule functions.
I modified the codegen to also create a mapping from NativeModule method name => argument num => `RCTCxxModule` conversion selector. This way, the FB codegen that generates the `RCTCxxConversion` function also informs our TurboModule system which conversion function to use before we call the method requiring complex argument. I just had to extend the `ObjCTurboModule` class to accomplish this.
The old system relies on the additional method generated by `RCT_EXPORT_METHOD` macro. It takes the written code, does string processing on it to parse the type of the struct arguments, and then replaces all instances of `::` with `_`. Super hacky. I didn't take this approach because it seemed unnecessarily hacky brittle.
The other approach I considered was to try to use reflection to infer the type of the struct at runtime. We'd have to do a bit of string processing and concat the TurboModule class name with the struct type. The solution would be simpler because it'd only modify the objective C and it wouldn't touch the Codegen system. **Edit:** I implemented it here: D14513078. The one downside of this design is that the name of the conversion function is individually constructed in two different locations in the code (i.e: no source of truth for this name). The other downside is that we have to rely on `RCTBridgeModuleNameForClass`, which we want to eliminate. This also computationally more expensive since it requires string processing/regex parsing and additional reflection. Therefore, we decided to move on with the current solution.
Reviewed By: fkgozali
Differential Revision: D14513429
fbshipit-source-id: 3d1b87e02ee908a19305686ff82b2ed624d8ac67
Summary: [iOS] [Fixed] - The existing logic defaults `undefined` & `null` in JS to be `[NSNull null]` when converting JS object to `NSDictionary`. Let's not insert the prop to the dictionary if it's `undefined`.
Reviewed By: blairvanderhoof
Differential Revision: D14571128
fbshipit-source-id: e03c713b055672b0a001d3305d694912ee36ab36
Summary:
JSBigString was inadvertently changed to a shared mapping. This means
that any changes to the string will be written back to the file. Ensure
we have a private (COW) mapping.
Reviewed By: kodafb
Differential Revision: D14532757
fbshipit-source-id: 6afb9635493496c90904f1432847c2f0da882c58
Summary:
It crashes with IllegalStateException in case you pass a wrong URL.
It crashes if it meets unexpected symbols in the header name and value,
while standard says it is not recommended to use those symbols not that
they are prohibited.
The headers handing is a special use case as a client might have an auth token in the header. In this case, we want to get 401 error response
from the server to find out that token is wrong. In case of the onerror
client will continue to retry with an existing token.
[ANDROID][Fixed] - Networking: Passing invalid URL not crashes the app instead onerror callback of HttpClient is called. Invalid symbols are stripped from the headers to allow HTTP query to fail with 401 error code in case of the broken token.
Pull Request resolved: https://github.com/facebook/react-native/pull/21231
Reviewed By: axe-fb
Differential Revision: D10222129
Pulled By: hramos
fbshipit-source-id: b23340692d0fb059a90e338fa85ad4d9612275f2
Summary:
Fixes#23342.
Since Metro [v0.47](https://github.com/facebook/metro/releases/tag/v0.47.0), some babel plugins such as [babel-plugin-module-resolver](https://github.com/tleunen/babel-plugin-module-resolver) or [babel-plugin-import-graphql](https://github.com/detrohutt/babel-plugin-import-graphql) fail to resolve files if the packager is started through Xcode. They receive wrong filenames from Babel such as `${projectRoot}/node_modules/react-native/src/index.js` instead of `${projectRoot}/src/index.js`.
It happens because the start command of the local-cli is not executed in the projectRoot directory. In this case, the cwd will be `${projectRoot}/node_modules/react-native`.
This issue doesn't occur if you start the packager yourself using `node node_modules/react-native/local-cli/cli.js start` from your project root.
It comes from this [line](b640b6faf7/scripts/packager.sh (L11)). This script changed the working directory to `${projectRoot}/node_modules/react-native` and started Metro from there.
Starting Metro from the project root fixes this issue.
[iOS] [Fixed] - Start Metro packager from project root
Pull Request resolved: https://github.com/facebook/react-native/pull/24070
Differential Revision: D14563996
Pulled By: cpojer
fbshipit-source-id: cdeff34610f1ebb5fb7bc82a96f4ac9eec750d16
Summary: This diff changes `js1 upgrade jest` to add a caret when upgrading so that the open source version of RN is not locked to a concrete version of Jest which doesn't really make much sense. I also added `pretty-format` to the packages that get updated because it is part of Jest's repo.
Reviewed By: rubennorte
Differential Revision: D14503180
fbshipit-source-id: 683c9a6ac768baa578771b9ad677e02df4912baa
Summary:
With recent activity to extract out non-core things into separate repos, I did an audit of the package.json to see what can be removed, and what packages remain that have newer versions that result in less/smaller javascript code. The goal is to:
1. make developer experience better by removing unneeded dependencies, and
2. eliminate things the app bundler needs to process to make the production app bundle smaller.
[internal] [chore] - Remove unneeded packages, bump package versions who now have smaller size
Pull Request resolved: https://github.com/facebook/react-native/pull/23968
Differential Revision: D14519070
Pulled By: cpojer
fbshipit-source-id: 579f035f925668a34d07293810e3d35a7d0ae076
Summary:
Existing glob might not work very well with `spec.js` and `test.js` files unless they're placed in `__tests__` directory. This PR aims to bring back the support of `jest` globals in these files, even if they're outside of `__tests__` directory.
[General] [Fixed] - Fixed globs for `spec.js` and `test.js` files.
Pull Request resolved: https://github.com/facebook/react-native/pull/24088
Differential Revision: D14562085
Pulled By: cpojer
fbshipit-source-id: 543d67e3f8a154256f454b34ccc68bb070197a75
Summary:
The goal of this PR is to enable eslint checks in the projects generated by `react-native init` command. I added `template/_eslintrc` file, that would be replaced in an initialized project with `.eslintrc` file. This PR comes in parallel with https://github.com/react-native-community/react-native-cli/pull/229
[General] [Added] - Added `.eslintrc` file to generated template.
Pull Request resolved: https://github.com/facebook/react-native/pull/23901
Differential Revision: D14561084
Pulled By: cpojer
fbshipit-source-id: 6eb717bf03c45d83ae8a393e6a0abb79e1e2f915
Summary:
@public
Yoga's Java license headers were not in the correct format.
Reviewed By: muraziz
Differential Revision: D14541087
fbshipit-source-id: 5b3cff398875bd59dadeaddbb43020700ef027e2
Summary:
Fix for this issue I rasied: https://github.com/facebook/react-native/issues/24024
When I toggle `Show Perf Monitor` and reload JS `CMD+R` the Perf Monitor will be hidden, but settings in dev menu will persist. So to fix this state and need to `Hide Perf Monitor` and `Show Perf Monitor` again to see it.
[iOS] [Fixed] - Show Perf Monitor, after reloading JS
Pull Request resolved: https://github.com/facebook/react-native/pull/24073
Differential Revision: D14560025
Pulled By: cpojer
fbshipit-source-id: cd5602bd6ee041b8b3e61d163d10bd8bc47237b9
Summary: Conceptually, this assert is correct, however, sometimes a new node got allocated by same address as old parent node (which does not exist already) which makes the assert fires.
Reviewed By: mdvacca
Differential Revision: D14533070
fbshipit-source-id: 3fcc71c25e7d724180dc85aaf2457227d22ddba0
Summary:
Google recommends to use Gravity.START and Gravity.END instead of Gravity.LEFT and Gravity.RIGHT to support RTL better.
[Android] [Changed] - Improve RTL support
Pull Request resolved: https://github.com/facebook/react-native/pull/24069
Differential Revision: D14541569
Pulled By: cpojer
fbshipit-source-id: 5c104d8bd666e1270d5410216c7f2efa6152692a
Summary:
For the use-case where iOSdebug builds prefer to use static bundles over bundles served dynamically by metro. There's currently a bug caused by the packager client repeatedly attempting to connect to metro. This leads to endless log statements in xcode as well as other side effects:
https://github.com/facebook/react-native/issues/21030. Fix is based off of https://github.com/facebook/react-native/issues/21030 which fixes the behavior for the inspector.
Stops endless logging by packager client for debug builds that use static bundles.
[CATEGORY] [TYPE] - Message
Pull Request resolved: https://github.com/facebook/react-native/pull/24050
Differential Revision: D14541574
Pulled By: cpojer
fbshipit-source-id: 09c2c82eb2d4aeb0e7ba322fd81b0dd7bf9e683c
Summary:
Custom containers are only enabled in release mode.
Using custom stuff complicates debugging process because it breaks embedded into IDE introspections mechanisms.
Reviewed By: mdvacca
Differential Revision: D14508299
fbshipit-source-id: d2dbe87764b17d75ccd544c0a4baf03dd9e0cd0b
Summary:
Added a child class of YogaNodeJNIBase which will be used to separate layout outputs transfer logic.
This change set is for adding experiment for layout outputs batching using a float array
Reviewed By: davidaurelio
Differential Revision: D14368098
fbshipit-source-id: e0f10fb61cd09ee47cf9ce41fb400f4cfb3dd795
Summary:
It allows HTTP connections in debug builds, and requires no configuration compared to previous/current solution where you need to edit config file to test on device.
Consulted with Salakar about this.
[Android] [Changed] - Allow HTTP in debug builds
Pull Request resolved: https://github.com/facebook/react-native/pull/24066
Differential Revision: D14540255
Pulled By: cpojer
fbshipit-source-id: f1239154c27c36c21c9b080a826f8b1d0eb0fc7d
Summary:
@public
Contexts of nodes only hold a bit mask after we got rid of weak JNI refs.
We can simply store that data in a pointer-sized unsigned int.
Here, we replace all context heap allocations with usage of the node context (`void *`) as bitmask. We also add a couple of utility operators in order to keep the code comprehensible.
Reviewed By: fabiomassimo
Differential Revision: D14425742
fbshipit-source-id: f32c2184a1f09268c39dbb8cd09ac96517339674
Summary:
when conditional rendering is used then picker breaks when the child is null in iOS
This conditional rendering inside Picker fails when "someBooleanValue" variable is false
```
{
this.state.someBooleanValue && <Picker.Item label="value" value="value" />
}
```
[iOS] [Fixed] - null check on children by using filter
Pull Request resolved: https://github.com/facebook/react-native/pull/24057
Differential Revision: D14538337
Pulled By: cpojer
fbshipit-source-id: d9324671931b5f1dac65d8058d9aa957b650af25
Summary:
Renamed class from YogaNodeJNI to YogaNodeJNIBase.
This change set is for adding experiment for layout outputs batching using a float array where we will have two separate classes which will override how layout outputs are transferred to java YogaNode object.
We needed two separate classes because having everything in one class was causing memory issues as both the individual fields for width, height etc. and float array for batching needs to be present in code.
Reviewed By: davidaurelio
Differential Revision: D14368069
fbshipit-source-id: 0e98e28c8c7a9788345ccb92b2cd0f2cd4a53525
Summary:
This module was merged with `Alert`. Landing the removal now will mean that it is gone from 0.61 which means the deprecation warning will have been there for two full versions.
It will also make it easier to ship D14538298.
Reviewed By: rubennorte
Differential Revision: D14539026
fbshipit-source-id: fb36482324866c277811756cc7461451a1482b23
Summary:
Jest 24 includes [`testMatch` and `moduleFileExtensions`](c5fd7aae93/packages/jest-config/src/Defaults.ts (L70)) that align with the ones that are currently there on master, because it added default handling for TypeScript as well. I think it's time for us to move to Jest 24. Is there a way we can tell users to upgrade Jest to certain version?
Fixes https://github.com/facebook/react-native/issues/24060
[General] [Changed] - update Jest preset to align with Jest 24
Pull Request resolved: https://github.com/facebook/react-native/pull/24062
Differential Revision: D14538988
Pulled By: cpojer
fbshipit-source-id: d8d152b8e8517b34144970f1cc1ed0b49f8b4e54
Summary:
Change the function called when an error is returned from the completionHandler of the `RCTJavaScriptLoader(loadBundleAtURL:onProgress:onComplete:)`.
* AS-IS
- `RCTLogError()`
- This error method does not post any notification for users. So users cannot receive notification when `RCTJavaScriptLoader(loadBundleAtURL:onProgress:onComplete:)` complete with error.(i.e. bundleURL is is invalid but not nil, `http://1`, `http://localho:8081/index.bundle`)
* TO-BE
- `handleError(error:)`
- Call this method will post notification for users when `completionHandler` fails with error.
[iOS] [Fixed] - Change the function called when an error is returned from the completionHandler of the `RCTJavaScriptLoader(loadBundleAtURL:onProgress:onComplete:)`.
Pull Request resolved: https://github.com/facebook/react-native/pull/23837
Differential Revision: D14538790
Pulled By: cpojer
fbshipit-source-id: da3306904c0d8113d204edfaa0f9e2a23793981a
Summary:
This PR fixes the case where the content a VirtualizedList loads with a contentLength of 0, causing a crash in the `renderDebugOverlay` function. The result of that crash is a red screen when turning on debug on FlatList and other VirtualizedList components as described in #24053.
[LIST] [FIX] - Fix VirtualizedList debug mode crash
Pull Request resolved: https://github.com/facebook/react-native/pull/24058
Differential Revision: D14538317
Pulled By: cpojer
fbshipit-source-id: 7b17bf51c388561c517bab1f775a31200abdc5a9
Summary:
Put `prepareForRecycle` to last before enqueue to recycle pool, ensure only call it when count lower than RCTComponentViewRegistryRecyclePoolMaxSize.
cc. shergin .
[General] [Changed] - Reorder prepareForRecycle before adding recycle pool
Pull Request resolved: https://github.com/facebook/react-native/pull/24025
Differential Revision: D14536843
Pulled By: shergin
fbshipit-source-id: 82a816e2c0afb5a6bb72637d7d55d6a4fda708af
Summary:
I was annoyed by warnings from yarn/npm about eslint peer dependencies not being met, so I dived in to try and get rid of some of them. Sometimes it meant bumping a plugin, but then that plugin needed a newer babel-eslint, so it was a dance.
Some we can't easily update to latest (eslint-plugin-prettier) because the rule format has changed a bit. Happy to do that in this PR if folks think its a good idea. eslint-config-fbjs itself needs to be updated and republished to eliminate the last few warnings.
There are a few new warnings (the repo wasn't linting cleanly for me from the start). I can fix those in this PR, or a separate one, based on people's preferences.
[internal] [chore] - Eliminate some peer dependency warnings and bump some eslint packages to latest.
Pull Request resolved: https://github.com/facebook/react-native/pull/23969
Differential Revision: D14519076
Pulled By: cpojer
fbshipit-source-id: aa44cfd05814d7b79069414aa78d089e448c2235
Summary:
The copyright headers in the Wrapper example referenced a license file that doesn't exist anymore. This PR updates the copyright header to match other example files in the repo
Fixes https://github.com/facebook/react-native/issues/23215
[General] [Changed] Updated the license of RCTWrapperExample to match the rest of the Examples
Pull Request resolved: https://github.com/facebook/react-native/pull/24037
Differential Revision: D14521327
Pulled By: cpojer
fbshipit-source-id: 983741d4d93178cd90b532ad5f8ff0717670aff4
Summary:
Fixes#22873. Given that this code is meant to only work on macOS, we should be fine not using a portable hash. In the packages for which this matters at the bottom of this file, it produces the same hashes so it should be fine.
[General] [Fixed] - Don't use `-p` when invoking shasum.
Pull Request resolved: https://github.com/facebook/react-native/pull/24036
Differential Revision: D14521134
Pulled By: cpojer
fbshipit-source-id: 882e762b7817d9d46bdd405c9e2e9e8b4d7cfaf4
Summary:
It was not possible to run RNTester in Android P and connect to the packager.
[Android][RNTester] add react_native_config to RNTester for the packager to be able to connect with Android P devices
Pull Request resolved: https://github.com/facebook/react-native/pull/24033
Differential Revision: D14520803
Pulled By: cpojer
fbshipit-source-id: 567a0461d14d880cd78e9fa551ba25fe60af798d
Summary: Currently we can't access navigationOptions from measureAppRequire and that breaks fetchRelayQuery logs.
Reviewed By: fkgozali
Differential Revision: D14505525
fbshipit-source-id: 97ae4461ebb0e703633f782323b13b4e4f7c5cb7
Summary:
```
let index = params.itemIndex + 1;
```
to
```
let index = Platform.OS === 'ios' ? params.itemIndex : params.itemIndex - 1;
```
to fix an issue on iOS. Note however, how the sign for non iOS changed from `+` to `-` causing a crash on Android when trying to scroll to index 0 as that will be evaluated to -1 instead of 1 and thus be out of bounds.
[Android] [Fixed] - Fixed regression in SectionList caused by #21577 not being able to scroll to top on android
Pull Request resolved: https://github.com/facebook/react-native/pull/24034
Differential Revision: D14520796
Pulled By: cpojer
fbshipit-source-id: bb49619f49752fd3f343ef3b7bf1b86ac48af7f8
Summary: Adds support for the native type Point to the rn codegen
Reviewed By: TheSavior
Differential Revision: D14462164
fbshipit-source-id: 942b5697d616c6aa6289d01bb56382fd7adac203
Summary: Same algorithm using a new API.
Reviewed By: JoshuaGross
Differential Revision: D14423509
fbshipit-source-id: c40f61fdd001f68785512c304941f523585d641c
Summary: The basic idea of the implementation is the same, but it uses the new method for finding parenting chain and it does not use `shared_from_this()` anymore.
Reviewed By: JoshuaGross
Differential Revision: D14416949
fbshipit-source-id: 13ef928505a60280e2a6c30c76ac87d91cee326e
Summary:
The algorithm is quite simply:
1. Get family objects of the two nodes (inner and outer).
2. Traverse the list of family objects starting from the inner one and form a reversed list of them.
3. tarting from the inner node, traverse the tree layer-by-layer choosing a next element of the path by comparing the family object of each node on the level and an object from the list.
Reviewed By: JoshuaGross
Differential Revision: D14416950
fbshipit-source-id: 23c659a9e01690f90174193650a2b0ef09eadb4d
Summary:
One of the core feature of ShadowNodeFamily is having a pointer to a parent family. This diff implements it.
I don't think there is a hard retain cycle there, but for more lean memory usage we use weak_ptr here.
Reviewed By: JoshuaGross
Differential Revision: D14416948
fbshipit-source-id: 05fd2c4833146f007228363b1d958776b4a2a9cf
Summary:
ShadowNodeFamily is a new concept that allows us to implement an efficient way to solve the problem of finding a path to some ancestor node (practically reimplement ShadowNode::constructAncestorPath() in some efficient way).
This diff is the first one in the series. It introduces a new class and moves some data into it.
Reviewed By: JoshuaGross
Differential Revision: D14416947
fbshipit-source-id: c93865a8929a2128498e34d3589487696aac6283
Summary: This is a leftover from the ealy days of Fabric. The mehtod does not have implementation counterpart and any usage.
Reviewed By: JoshuaGross
Differential Revision: D14402636
fbshipit-source-id: aa2e919084ae7382d3a62af761e1f6eac334fc75
Summary: The removed code does nothing because we are replacing the oldChild with newChild several lines above.
Reviewed By: JoshuaGross
Differential Revision: D14402637
fbshipit-source-id: 731a950f373e20f7d5bae3cbf6470335d3694ccc
