Summary:
Android documentation was not updated accordingly following a native Android API surface modification which ended up renaming a couple of methods (introduced in 19a1c4c229)
This PR is fixing the documentation while at the same time updating the documentation in the same section adding another method which was introduced in the same commit and which needs to be called as well (please see comment in [ReactInstanceManager.java](5b871ad9d7/ReactAndroid/src/main/java/com/facebook/react/ReactInstanceManager.java (L44)) : `It's required to pass owning activity's lifecycle events to the instance manager [...]`)
Closes https://github.com/facebook/react-native/pull/8032
Differential Revision: D3424674
fbshipit-source-id: 867f98ef210d7e9402afc278fb7b8e08890de3f0
Summary:
The default priority for AsyncTask is THREAD_PRIORITY_BACKGROUND which puts the thread in a limited cgroup that can utilize ~10% CPU time. Because the whole TTI depends on this task we should run it on a normal priority.
The priority will be reset for the nest task and this has been true since donut: https://github.com/android/platform_frameworks_base/blob/donut-release/core/java/android/os/AsyncTask.java
Reviewed By: astreet
Differential Revision: D3417232
fbshipit-source-id: e1e35f82b35b31ff7ebf4fc59509ca2df21e2bdd
Summary: Expose the ability to provide extra information to Systrace events from JS. This is specifically useful for instrumenting generic recursive method that are defined completely through their params.
Reviewed By: tadeuzagallo
Differential Revision: D3423595
fbshipit-source-id: 7f2dedf02bf901401a6b391b85b1a0bb6782349c
Summary:
**motivation**
Lots of people does not sure what to do after a PR.
common questions like:
* should I keep rebasing this PR to the latest master branch ?
* should I keep squashing all commits in a PR into single one?
This may clear it a bit and makes people less nervous 😱
**Test plan (required)**
Closes https://github.com/facebook/react-native/pull/8081
Differential Revision: D3424076
fbshipit-source-id: 6c5f1b0b2f0bee03c647c2f8cad00b788130a5fc
Summary:
Submitting PR for adding textAlignment to the placeholder of textView so that it matches the alignment of the actual value text
Closes https://github.com/facebook/react-native/pull/7429
Differential Revision: D3422007
fbshipit-source-id: e255c2d86dbaa0e197328a56ec81cbab2735c53d
Summary: Slower bounce during mount and right swipe, so it feels less jarring to users.
Reviewed By: hedgerwang
Differential Revision: D3420169
fbshipit-source-id: d5da87c4a64eecf478edd626c96f5541ab94b56e
Summary:
Re: javache 's suggestions from https://github.com/facebook/react-native/pull/7878. Didn't want to deal with the merge conflict so I'm opening a separate PR. Here's the original justification:
If I want to set NODE_ENV to "baconator", I should be allowed to. Mutating global state that most devs assume to be immutable is just abysmal dev practice, especially since this mutation only happens when you're building for prod, not running on the simulator.
To test this, run env NODE_ENV=baconator ./gradlew assembleRelease with babel-plugin-transform-inline-environment-variables in your app/.babelrc. You'll see that the final app has NODE_ENV=production.
As a side note, running with babel-plugin-transform-inline-environment-variables in the top-level .babelrc crashes horribly with a compiler error.
For anybody who runs into this bug and doesn't feel like waiting for this to get merged, I wrote a quick babel plugin to remove assignments to process.env, which is sufficient to fix this issue.
Closes https://github.com/facebook/react-native/pull/8057
Differential Revision: D3419950
Pulled By: javache
fbshipit-source-id: dc541cad0a99906433e5c14bbc93ce66b4ed325e
Summary:
When remote debugging is enabled, stack traces start at `/debuggerWorker.js`. Since this is not a valid bundle URL, the packager fails to decipher it to find its sourcemap.
This changes the packager to skip the `/debuggerWorker.js` stack frame if one exists.
Reviewed By: frantic
Differential Revision: D3418341
fbshipit-source-id: 7434aa45dea7d120d9d77c060101dd9403989d0c
Summary:
Fix a bug that happens when views are added and removed in the same manageChildren block with a delete animation. What happens is that the inserted view is not inserted at the proper index if the deleted view index is smaller than the inserted one. This is because the view is not immediately removed from the subviews array so we need to offset the insert index for each view that is going to be deleted with an animation and is before the inserted view.
To do this I separated `_removeChildren` into 2 different functions, one for animated delete and one for normal delete. The animated one returns an array of `RCTComponent` that are going to be animated. We can then use this array to offset the insert index.
**Test plan (required)**
Tested that this fixed the bug in an app where I noticed it, also tested the UIExplorer example to make sure LayoutAnimations still worked properly.
Closes https://github.com/facebook/react-native/pull/7942
Differential Revision: D3417194
Pulled By: bestander
fbshipit-source-id: 9145a783e520c6718dd023a1e006646acb09cb7c
Summary:
Currently if an RN app is started in response to a remote notification, that notification's data is available on startup via `PushNotificationIOS.popInitialNotification()`. However, if the app is started in response to a "local" notification, that information is never passed in. This PR modifies the `popInitialNotification` behavior so it will return the notification used to launch the app, no matter if it was local or remote.
I've tested this change in my app and ensured that when the app is woken up with a `localNotification` it's passed in to `PushNotificationIOS.popInitialNotification`. I've also tested that the `localNotification` event continues working as before.
Closes https://github.com/facebook/react-native/pull/7765
Differential Revision: D3417267
Pulled By: nicklockwood
fbshipit-source-id: 0b5b432e9a75dda7d3c50289a3bf0f1c1ffcf061
Summary:
Fix a bug that happens when views are added and removed in the same manageChildren block with a delete animation. What happens is that the inserted view is not inserted at the proper index if the deleted view index is smaller than the inserted one. This is because the view is not immediately removed from the subviews array so we need to offset the insert index for each view that is going to be deleted with an animation and is before the inserted view.
To do this I separated `_removeChildren` into 2 different functions, one for animated delete and one for normal delete. The animated one returns an array of `RCTComponent` that are going to be animated. We can then use this array to offset the insert index.
**Test plan (required)**
Tested that this fixed the bug in an app where I noticed it, also tested the UIExplorer example to make sure LayoutAnimations still worked properly.
Closes https://github.com/facebook/react-native/pull/7942
Differential Revision: D3417194
Pulled By: nicklockwood
fbshipit-source-id: 790f4ac15a8552323b359e6466cecfa80418c63c
Summary: ImageCropper is broken on iOS. Can not scroll up and down
Differential Revision: D3413397
fbshipit-source-id: 75096fc1d5dd14764c0ddd4fd3888a9576c1d1ce
Summary:
Remove prop `onNavigate` from these views.
- NavigationAnimatedView
- NavigationCardStack
- NavigationCard
Also, the `sceneProps` onject that is passed to the `renderScene` function
no longer contains `onNavigate`.
The contract that `onNavigate` expects has been vague. Different data flow
system may expect complete different params for such function
For instance,
* onNavigate({type: 'back'});
* onNavigate({type: 'BACK'});
* onNavigate('back'});
We have no intention to unify such generic API since it's more likely to be
constrained by the data flow frameworks such as redux or flux.
Also, passing the prop `onNavigate` all the way down to the component that
invokes the navigation action can be really tedious. We'd expect developer
to either pass such callback (onNavigate) via context or just set up some
kind of static actions that any component can call directly.
`onNavigate` was previously added as a part of (redux-like) reducers-friendly
feature but that's no longer the case.
This new prop `onNavigateBack` is used to explicitly handle the case when the back button or back gesture
is performed.
Reviewed By: ericvicenti
Differential Revision: D3410873
fbshipit-source-id: a703cf0debd474cff33d6610e858b9c4bb3ecbf5
Summary:
The documentation currently list a few but not all of the steps required. This changes completes that list of required project changes.
This addresses #5612
Closes https://github.com/facebook/react-native/pull/7985
Differential Revision: D3414576
Pulled By: nicklockwood
fbshipit-source-id: 2c7ea6598ee4cd8b6945d9bb229bed5b592a68f6
Summary:
Currently on iOS animations are being performed on the JS thread. This ports animations over to the native thread and performs them natively. A lot of this work has already been done on Android, but this PR enables a few animation nodes that Android doesn't yet support such as Transform, Multiplication, and Addition nodes.
Also there is a demo of the native animations added to the UIExplorer app.
Closes https://github.com/facebook/react-native/pull/7884
Reviewed By: javache
Differential Revision: D3409179
Pulled By: nicklockwood
fbshipit-source-id: ef2d8840032e0c32f49e4a16ba86d448662e1751
Summary:
Under rare and as-yet-to-be determined circumstances, images can sometimes fail to load/download and get "stuck", without producing an error.
Because the `RCTNetworkTask` for these images is stuck in the "in progress" state, they clog up the RCTImageLoader task queue, which has a limit of 4 concurrent in-progress tasks.
This was previously masked by the fact that we automatically cancelled image requests when the RCTImageView moved offscreen, but we no longer do that.
This diff adds logic to detect some types of stuck task and remove them, thereby unblocking the queue. I've also restored the functionality of cancelling downloads for offscreen images (but not unloading the image itself) so that stuck images will be cancelled when you move to another screen, instead of using up space in the queue forever.
Reviewed By: fkgozali
Differential Revision: D3398105
fbshipit-source-id: 75ee40d06a872ae8e1cb57f02f9cad57c459143c
Summary: Converted the zIndex property on iOS to NSInteger instead of double. This is consistent with the CSS spec, and helps to simplify the Android implementation.
Reviewed By: javache
Differential Revision: D3411491
fbshipit-source-id: 902ebc29aac39a65f7e8707a28607655f9f5052c
Summary:
When we hit the back button and unmount the ReactRootView, we tell JS to unmount the application root node, which causes JS to asynchronously come back and tell the UIManager to drop the corresponding root view.
This issue was that by the time JS gets back to us, we likely will have already paused the UIManager frame callback which means the view unmounting never actually happens: it just gets stuck in the queue.
The solution is to immediately execute batches when they are enqueued when the frame callback isn't running.
Reviewed By: lexs
Differential Revision: D3398958
fbshipit-source-id: 0de81061a97a119be4cb0b12d6f01c1cec8e8171
Summary:
Returns a promise-like object with a new cancel function that will dig through the queue
and remove relevant tasks before they are executed. Handy when tasks are scheduled in react
components but should be cleaned up in unmount.
Reviewed By: devknoll
Differential Revision: D3406953
fbshipit-source-id: edf1157d831d5d6b63f13ee64cfd1c46843e79fa
Summary:
Clears the debug console whenever the React Native JS is reloaded.
NOTE: This respects "Preserve log" in Chrome by default.
Reviewed By: jingc
Differential Revision: D3409713
fbshipit-source-id: ce215e3125cf43ab3ea5811c707fab9dfa4bcbb3
Summary:
== API Breaking Change ==
- Add unit tests to ensure that NavigationStateUtils does the right thing.
- Remove the logics that lets NavigationStateUtils accept empty value as input
and return a new state.
- Remove the method `NavigationStateUtils.getParent`, `NavigationStateUtils.set`. These methods are rarely used and they can be replaced by other methods.
Reviewed By: ericvicenti
Differential Revision: D3374934
fbshipit-source-id: 0fdf538d014d7c5b4aa1f15a0ee8db9dc91e33cd
Summary:
`rootTag` is a lie, it's passed around but never actually used. IIRC
`findInstanceByNativeTag` needed it but seems like not anymore.
Reviewed By: spicyj
Differential Revision: D3382144
fbshipit-source-id: eb96870a3848333e66bf045e78e95c7763812cc4
Summary:
`renderApplication` is cluttered with unrelated Dev-only stuff
with duplicate code across iOS and Android. This is the first
diff in series of refactorings, with the end goal of making element
inspector work on Modals.
Reviewed By: spicyj
Differential Revision: D3381956
fbshipit-source-id: 4ac6525633e7482628d2b064eb894da2806daf8c
Summary:
The Getting Started guide is one of the first documents a new user will encounter. This pull request aims to minimize the amount of time it takes to get a React Native app up and running.
* The original section title, "Required prerequisites", is redundant. "Installing React Native" is a better description of what this section is about.
* Detailed installation instructions for each of the required tools are delegated to the first party instructions where available.
* If the installation instructions already take care of installing the latest version, there's no need to warn the user about the minimum required version.
* Assume the user is familiar with Homebrew or Chocolatey, and defer installation instructions to the package manager's website. The installation and explanation of what a package manager is is out of scope within this document.
* Link to Node.js package manager instructions and let savvy Linux users use the pack
Closes https://github.com/facebook/react-native/pull/8010
Differential Revision: D3407029
Pulled By: JoelMarcey
fbshipit-source-id: c8b25d5b176c40eb58e5d7d3c6f13d43cde65166
Summary:
For navigation actions at high level, reducers from NavigationReducers does not
know anything about the app-specific state thus people won't use these reducers.
Instead, people should build their own reducers.
There are a lot of good libraries available that help people to reducing things if that's
what they really need.
At the low level, for navigation state changes that don't involve app-specific state,
`NavigationStateUtils` should server that kind of need.
`NavigationReducers` serves little benefit cause it does not know the app state, it does
not know how to traverse the navigation states which can be a tree, a list or a map.
That said, we hold no interest in owning in the core navigation library.
Reviewed By: ericvicenti
Differential Revision: D3372910
fbshipit-source-id: 797382b46e7d64b7ad578b51dd37e2b941faa83d
Summary:
Add more flexibility to the iOS UIExplorer test scripts:
- support environments without `xcpretty` installed
- support custom xcode destination argument
Initial PR is based on the last git commit which passed on Travis today (2a92b52) in order to test Travis build.
I'll rebase to latest, in order to fix merge conflicts, before merging.
**Test plan (required)**
Make sure tests pass on Travis.
Closes https://github.com/facebook/react-native/pull/7982
Differential Revision: D3404574
Pulled By: avaly
fbshipit-source-id: 48aabd81fba67d482af46728a9c3975842f03060
Benoit Lemaire
Alexander Blom
Pieter De Baets
zxcpoiu
Mark Wang
Guillermo Orellana
Dave
Fred Liu
Valeri Karpov
Yinan Na
Keri Clowes
Jing Chen
Tim Yung
Konstantin Raev
Tucker Connelly
Janic Duplessis
Kyle Corbitt
Sokovikov
Fabian Eichinger
Paul Shen
Wenjing Wang
Hedger Wang
Kevin Smith
Kasim Tan
Dave Miller
MacKentoch
Ovidiu Viorel Iepure
Brandon Withrow
Nick Lockwood
Andy Street
Nathan Azaria
Spencer Ahrens
Alex Kotliarskyi
Héctor Ramos
Olivier Notteghem
Yann Pringault
Valentin Agachi
