Summary:
React Native uses JSON to marshal the data across the bridge.
And because of this we have to avoid using NaN and INF values in events and other pieces of data that suppose to be transfered to/from JS side.
(We also don't want to introduce additional wrapping/escaping semantics for perfomance reasons.)
So, we have to gate all particular cases where there is a possibility of NaN or INF values, and replace these value with something meaningful for each particular case.
We are using `0` as NaN substitution here because:
* NaN in touch event is super rare case;
* Conversion to `0` is fast;
* `0` is okay value for product code in most cases;
* In all cases `0` is decent analog to "undefined position on screen" for touch event;
* Nobody will explicitly handle NaN case in product code, just because it is super rare case and actually indicates that something else went wrong.
Reviewed By: javache
Differential Revision: D4918669
fbshipit-source-id: e95fa29e59dcdc40b57519e307b74c1f293da188
Summary:
In the Docs of `NSURLSession` ,
>IMPORTANT
>The session object keeps a strong reference to the delegate until your app exits or explicitly >invalidates the session. If you do not invalidate the session, your app leaks memory until it exits.
The RCTMultipartDataTask will cause memory leaks, it will make `RCTBatchedBridge` and ` RCTMultipartDataTask` will not release.
So call `[session finishTasksAndInvalidate];` at the end of `startTask` function.
Closes https://github.com/facebook/react-native/pull/12673
Reviewed By: shergin
Differential Revision: D4896497
Pulled By: javache
fbshipit-source-id: eb5f8761f67ad33a7de081a68a9a7e1d4329bfc0
Summary:
Issue Fix: #13485
Yet another race condition that was found by XCode's Thread Sanitizer.
Happens because wasBatchActive is read/write from multiple threads at the same time
- opened UIExplorer and see it works
- npm run test pass
- tested on own project as well
Signed CLA
Closes https://github.com/facebook/react-native/pull/13505
Differential Revision: D4906096
Pulled By: javache
fbshipit-source-id: 5d4329aafcfe9491ce0188fa1e2dd71e09b31031
Summary:
Output the reason for the error when failing to load source code. This was a big help when trying to diagnose https://github.com/facebook/react-native/issues/13299.
~~Unfortunately there still seems to be no way to get the offending line number (because `loadError.userInfo[RCTJSStackTraceKey]` is empty), but this is good enough.~~
Before:
```
[warn][tid:com.facebook.react.JavaScript][RCTBatchedBridge.m:510] Failed to execute source code. Unhandled JS Exception: SyntaxError
```
After:
```
[warn][tid:com.facebook.react.JavaScript][RCTJSCErrorHandling.mm:30] Couldn't get stack trace for http://localhost:8081/index.ios.bundle?platform=ios&dev=true&minify=false:81886
[warn][tid:com.facebook.react.JavaScript][RCTBatchedBridge.m:510] Failed to execute source code. Unhandled JS Exception: SyntaxError Unexpected keyword 'var'
```
Closes https://github.com/facebook/react-native/pull/13561
Differential Revision: D4908501
Pulled By: javache
fbshipit-source-id: a316dc70739b917b3cc690309d0ff37a8bb5d412
Summary:
* It complicates Touch Handling mechanism.
* The same functionality can be (and should be) implemented via overriding standard `hitTest:` method.
* It was marked as deprecated a while ago.
Reviewed By: mmmulani
Differential Revision: D4667776
fbshipit-source-id: 2e047c3308563a2849ea351a242270f0800fead2
Summary:
…th RCTSharedApplication()
Thanks for submitting a PR! Please read these instructions carefully:
- [ ] Explain the **motivation** for making this change.
Using React Native latest version with Cocoapods 1.2.0 causes the following error inside iOS app extensions
> /react-native/React/Modules/RCTAccessibilityManager.m:67:70: ‘sharedApplication’ is unavailable: not available on iOS (App Extension) — Use view controller based solutions where appropriate instead.
Moving the use of [UIApplication sharedApplication] to RCTSharedApplication() which is safe on app extension
- [ ] Provide a **test plan** demonstrating that the code is solid.
I am not sure how to test such that all the features which touch the modified code are tested.
- [ ] Match the **code formatting** of the rest of the codebase.
- [ ] Target the `master` branch, NOT a "stable" branch.
What existing problem does the pull request solve?
Using React Native latest v
Closes https://github.com/facebook/react-native/pull/13227
Differential Revision: D4816338
Pulled By: javache
fbshipit-source-id: e3e3c77882990ad1817b0b633521cff52571ecd0
Summary:
Motivation: When logging to RCTRedBox, if any of the stack frames lacks a line number or a column, the application will crash with `[NSNull integerValue]: unrecognized selector sent to instance`.
Closes https://github.com/facebook/react-native/pull/13242
Differential Revision: D4812185
Pulled By: hramos
fbshipit-source-id: 4b1c3c38f67cf59034a383c95d4280d1b6380300
Summary:
Various fixes of xcode projects and cleaning up some warnings
Closes https://github.com/facebook/react-native/pull/13109
Differential Revision: D4762652
Pulled By: lacker
fbshipit-source-id: b452976a58962439de4adecc8e703264af40cb38
Summary: `autoresizingMask` is supposed to be set outside self class, this is UIKit convention.
Reviewed By: mmmulani
Differential Revision: D4697098
fbshipit-source-id: 7e0aa5d3032184de980b3cecafebbc4ce8ef9ada
Summary:
Moving setting `availableSize` for `RCTRootShadowView` on earlier stage allows to prevent situations where `availableSize` is not specified yet, but Yoga layout is already happening.
Because `availableSize` equals {infinity, infinity} by default (in this case), Yoga returns a lot of nodes with infinit metrics, which confises UIKit.
Reviewed By: mmmulani
Differential Revision: D4672170
fbshipit-source-id: f9d8c84799dcbdb6b9230ddef6284d84df268833
Summary:
This prevents someone from only putting `RCT_EXPORT_METHOD(...)` around some of the arguments to a method and generally causing confusion. Yes, before this diff it would still compile correctly if you did something like:
```
RCT_EXPORT_METHOD(lol:(NSNumber *)lol)
blah:(id)blah
{
NSLog(@"%@ %@", lol, blah);
}
```
Reviewed By: fkgozali
Differential Revision: D4660019
fbshipit-source-id: 1829a47e5c8d5e8ce93edca2ac7efd7e2bfdf840
Summary:
This diff adds display:none support to React Native. This enables hiding components which still calling their render method and keeping them within the state of your application. This enables preserving state in a component even though the component is not visible. Previously this was often implemented by rendering a component off screen as a work around. See below playground for usage.
```
class Playground extends React.Component {
render() {
return (
<View style={{width: '100%', height: '100%', flexDirection: 'row', backgroundColor: 'white'}}>
<View style={{width: 100, height: 100, display: 'none', backgroundColor: 'red'}}/>
<View style={{width: 100, height: 100, backgroundColor: 'blue'}}/>
</View>
);
}
}
```
Reviewed By: astreet
Differential Revision: D4611771
fbshipit-source-id: 0dbe0494d989df42994ab9ad5125d47f3233cc5a
Summary:
* The dev support code moved into a `DevSupport` subspec, meaning that only if the subspec is specified in the user’s Podfile will the packager client, dev menu, etc be included. This is mainly done through checks for header availability.
It also improves the weird situation where you had to specify the `RCTWebSocket` subspec if you wanted to be able to use the packager client during development.
* I removed hardcoding the release version in the podspec on release, because the podspec still relies on `package.json` when evaluating, so there’s no real point in not also getting the version number from there. This should remove any requirement to perform maintenance of the OSS release script regarding the podspec.
Closes https://github.com/facebook/react-native/pull/12602
Differential Revision: D4621021
Pulled By: ericvicenti
fbshipit-source-id: 6c208371fc40ea607809a6ab05dd3714ed9980cf
Summary:
Nothing actually changed except the deprecation.
Existed `intrinsicSize` was already implemented as `intrinsicContentSize` and this change only removes redundancy.
Moreover, we do not need `rootViewDidChangeIntrinsicSize` delegate method anymore; this is now mentioned in its description.
Depends on D4577890
Reviewed By: mmmulani
Differential Revision: D4589344
fbshipit-source-id: 16ed62cbf6bf72678bd7f7c11d4812c5aa36c743
Summary:
Now RCTRootView is much more reliable citizen of UIKit, it got:
* Implemented `sizeThatFits:`;
* Implemented `instrinsicContentSize`;
* Notifying superview via `setNeedsLayout` about changed size.
All it make possible painless integration of ReactNative-powered widgets inside existing native apps.
Reviewed By: javache
Differential Revision: D4577890
fbshipit-source-id: 9897cb002c9d658a97fd436240c2ac947ba2084b
Summary:
Since we are reading from a file, we should make sure this struct is packed, just in case we change it down the line and the compiler decides it might want to introduce padding, we're now protected against that.
There was also a discussion about the fact that people might use `ptr += sizeof(BundleHeader)` as an idiom in their code, which would currently be incorrect, if padding was introduced at the end of the file. Actually, it remains incorrect to do that now, because a RAM bundle header is a different size to a BC Bundle header. If people are properly testing their code, they should spot this pretty quickly, because it will always be an incorrect thing to do with a RAM bundle, so this isn't as bad as previously thought: where the code only succeeds when the compiler deigns to not pad the struct at the end.
This diff also cleans up how headers are initialised. `BundleHeader` has a constructor that explicitly zero-initialises it so we can rely on the default initializer to do the right thing now.
Reviewed By: mhorowitz
Differential Revision: D4572032
fbshipit-source-id: 7dc50cfa9438dfdfb9f842dc39d8f15334813c63
Summary:
Now things look much more clear, I hope.
This diff:
* Introduces new property of `RCTRootShadowView` `availableSize` which represents exactly what we transmit to layout engine;
* Illuminates conflict between logical `availableSize` and explicitly specified size of DOM node (current `size`);
* Splits overcomplicated `setSize:forView:` method into two unrelated ones;
* Changes actual values of `RCTRootViewSizeFlexibility` enum constants for simpler usage;
* Completely removes `sizeFlexibility` concept from `RCTRootShadowView` (in favor of special values of `availableSize`);
* Makes the code clearer finally.
This is beginning of big effort to improve `RCTRootView` and co.
Reviewed By: mmmulani
Differential Revision: D4562834
fbshipit-source-id: f5baaf2859ea430d44645a6b5d35f222f15a668e
Summary:
...to reflect the modern world we live in with dynamic DPI platforms :)
Closes https://github.com/facebook/yoga/pull/375
Reviewed By: dshahidehpour
Differential Revision: D4528518
Pulled By: emilsjolander
fbshipit-source-id: e422bd4ae148e02c598a7b484a6adfa8c0e1e0c9
Summary: Now layout direction (LTR or LTR) can be specified not only for whole app but also for view subtree via `direction` style property.
Reviewed By: mmmulani
Differential Revision: D4510206
fbshipit-source-id: 4e56c5886b6e42f2343165eb76be897e681c5ba4
Summary:
Eliminates a build warning related to the use of the deprecated `NSStringArray` typedef.
This fix was more complex than I'd anticipated because `NSStringArray` was also being used as part of a macro-generated selector name, and in two different ways for debug/release. I've added a macro which allows the selector name to be specified explicitly, thus generally allowing for converters which return arrays of templated types.
There's an argument for ditching `RCT_JSON_ARRAY_CONVERTER` in favour of `RCT_JSON_ARRAY_CONVERTER_NAMED` as they're both private, but `RCT_ARRAY_CONVERTER` is in the public API so we'd at least need to retain that. There are also arguments for ditching the use of the macro for the nested array case(s) - since afaik this is the only one at the moment. Feedback appreciated :)
Tested with the `UIExplorer` unit tests and by diffing the preprocessor output of `RCTConvert.m` in both release and debug configs, verifying that they're identical apart from that `NSStringArray` is replaced by
Closes https://github.com/facebook/react-native/pull/11799
Differential Revision: D4441197
fbshipit-source-id: 7535ebe6f8ad4566df742e805b0a64530d1b269f
Summary:
In theory, we should be able to animate any non-layout property, including custom ones. While there is still work to be done on the native side to fully enable this, we should start by dropping the prop whitelist.
Closes https://github.com/facebook/react-native/pull/10658
Differential Revision: D4379031
Pulled By: ericvicenti
fbshipit-source-id: fe9c30ea101e93a8b260d7d09a909fafbb82fee6
Summary: Switch to using IS_TESTING on the Platform module. While IS_TESTING has to be explicitly set in the test harness, this makes it more usable and stops people from relying on brittle variables in the (larger) environment.
Reviewed By: fkgozali
Differential Revision: D4423661
fbshipit-source-id: 27a80867778b9374bcba67b69f9c93d32c0a74b0
Summary:
In the context of an app an image exists in three resolutions on the server: `thumb` (30px) `feed` (300px) `full` (900px). When looking at an individual item a user can come either from the feed, via a permalink or from other parts of the app. This allows a situation where the `feed` image might or might not already be loaded somewhere in the app. In the detail view I want to render `thumb` with a blur (to quickly display something), then the `feed` image if it exists to have something decent to display until `full` loads. However it is quite a waste to load the `feed` image if it isn't already in cache, and will slow down the time until `full` is loaded. It is possible to track the navigation from feed->detail and that the `feed` image has actually completed loading by the feed component however as component hierarchies grow this turns into quite a lot of prop passing and bad separation of concerns.
NSURLRequests accepts a [Cache Policy](https://developer.apple.com/reference/fo
Closes https://github.com/facebook/react-native/pull/10844
Differential Revision: D4425959
Pulled By: lacker
fbshipit-source-id: 679835439c761a2fc894f56eb6d744c036cf0b49
Alex Dvornikov
Valentin Shergin
Pieter De Baets
wenzhao.yin
Andrew Tokarev
Kevin Cooper
Shruti Malugu
Ben Roth
Pieter De Baets
Marc Horowitz
Jeff Thomas
Ashwin Bharambe
Mehdi Mulani
Emil Sjolander
Eloy Durán
Ashok Menon
Alexey Lang
David Hart
rh389
Martin Konicek
Ryan Gomba
Peter Salanki