Summary:
This is a complete rewrite of RCTText, the part of React Native which manages Text and TextInput components.
Key points:
* It's understandable now. It follows a simple architectural pattern, and it's easy to debug and iterate. Text flow layout is a first-class citizen in React Native layout system now, not just a wired special case. It also brings entirely new possibilities such as nested interleaving <Text> and <View> components.
* All <Text>-specific APIs were removed from UIManager and co (it's about ~16 public methods which were used exclusively only by <Text>).
* It relies on new Yoga measurement/cloning API and on-dirty handler. So, it removes built-in dirty propagation subsystem from RN completely.
* It caches string fragments properly and granularly on a per-node basis which makes updating text-containing components more performant.
* It does not instantiate UIView for virtual components which reduces memory utilization.
* It drastically improves <TextInput> capabilities (e.g. rich text inside single line <TextInput> is now supported).
Screenshots:
https://cl.ly/2j3r1V0L0324https://cl.ly/3N2V3C3d3q3R
Reviewed By: mmmulani
Differential Revision: D6617326
fbshipit-source-id: 35d4d81b35c9870e9557d0211c0e934e6072a41e
Summary: Because setting `intrinsicContentSize` for `RCTSurfaceRootView` doesn't have much sense.
Reviewed By: mmmulani
Differential Revision: D6701107
fbshipit-source-id: 259cdd27339bba3e8c9f98b6ca34affeb87f298c
Summary: Now we do not add newly created view to the registry at the preluminary step.
Reviewed By: mmmulani
Differential Revision: D6641403
fbshipit-source-id: c69077aaba871f3cdb3500c75e1efe07546e1b7f
Summary: This is a leftover from recent changes in D6595780 where a prop application step was moved out to separare mount block.
Differential Revision: D6640736
fbshipit-source-id: 70de0f55f992a7912e222ec4bf9ade1c9bad99f2
Summary:
It's always bothered that we have this protocol and I figured it could just be merged with RCTWrapperViewController.
Closes https://github.com/facebook/react-native/pull/17290
Reviewed By: mmmulani
Differential Revision: D6611544
Pulled By: javache
fbshipit-source-id: a50b9d5adbeb2c48dbadbbfc1c77ccf6d1aae144
Summary:
Nobody uses it.
If the reference to the bridge is available (which is should be case for testing purposes at least), it is easy to get same information.
Reviewed By: mmmulani
Differential Revision: D6596376
fbshipit-source-id: 066eeb1e9465b4e0cc9d9b5b6bf41722450870e4
Summary: The `-[RCTShadowView viewName]` prop must exist, we don't need special handling for this case.
Reviewed By: mmmulani
Differential Revision: D6596375
fbshipit-source-id: 3e99a62bd6296e0285156f03dc2ac93db7f630e5
Summary:
Motivation:
* Current implementation of `didUpdateReactSubviews` relies on `processUpdatedProperties:parentProperties:` method of RCTShadowView, which we plan to remove.
* The existing implementation does not call handlers on unmounted nodes (because they are not part of traversing tree), which is not correct.
* The current implementation is tight with RCTComponentData, which is conceptually wrong, it should be a UIManager thing.
* The new implementation must be much more performant because of simplicity. (We can measure it only after removing `processUpdatedProperties`.)
Reviewed By: mmmulani
Differential Revision: D6595780
fbshipit-source-id: a517207c17b5d5db839c9ce99a37136292acf78c
Summary: As it was mentioned in previous diffs, we are removing this because it overcomplicates rendering layer and provides (almost) no benefits (and cannot be implemented 100% accurate way).
Reviewed By: mmmulani
Differential Revision: D6582560
fbshipit-source-id: 0778db96a45dd8e2520268d5d00792677cb01a20
Summary:
We are removing `reactBridgeDidFinishTransaction`.
Why?
* It is a performance drain. Supporting this requires dispatching main-thread block on every single transaction complete;
* It has "too broad" non-conceptual semantic which encouraged using this as a "band-aid solution" for poorly designed components;
* It is conceptually incompatible with new approaches that we are trying to implement to optimize the render layer;
* It was deprecated for very long time.
Reviewed By: mmmulani
Differential Revision: D6549729
fbshipit-source-id: 58094aab982c67cec3d7fa3b616c637cb84d697f
Summary:
Because it is not simply flushing, it (in the future) is more complex process. And the names should represent logical meaning of the process, not particular implementation details.
It also nice to have unified terminology across our reactive UI frameworks.
See the next diffs.
Reviewed By: rsnara
Differential Revision: D6436770
fbshipit-source-id: 0a0b686e8ace89e30f6787a37c0a7965c5af757b
Summary:
New super simple abstraction in RCTUIManager.
Nothing really changed and RCTUIManager became shorter.
Reviewed By: rsnara
Differential Revision: D5990342
fbshipit-source-id: b38397b789a999168ac14625585065eda73d328f
Summary:
Fix the following warning:
> Module RCTTVNavigationEventEmitter requires main queue setup since it overrides `init` but doesn't implement `requiresMainQueueSetup`. In a future release React Native will default to initializing all native modules on a background thread unless explicitly opted-out of.
Trivial change.
[IOS] [MINOR] [RCTTVNavigationEventEmitter] - Implement `requiresMainQueueSetup` in `RCTTVNavigationEventEmitter`
Closes https://github.com/facebook/react-native/pull/16915
Differential Revision: D6394636
Pulled By: shergin
fbshipit-source-id: 3981655832715e73e93ef917d987e25097029b84
Summary:
RCTSurface instance represents React Native-powered piece of a user interface
which can be a full-screen app, separate modal view controller,
or even small widget. It is called "Surface".
The RCTSurface instance is completely thread-safe by design;
it can be created on any thread, and any its method can be called from
any thread (if the opposite is not mentioned explicitly).
The primary goals of the RCTSurface are:
- ability to measure and layout the surface in a thread-safe and synchronous manner;
- ability to create a UIView instance on demand (later);
- ability to communicate the current stage of the surface granularly.
Differential Revision: D6202576
fbshipit-source-id: 8e644c87fcaad2b6a9c9304b58384d7192747556
Summary: This logic was decoupled from RCTRootView to make it reusable.
Reviewed By: javache
Differential Revision: D6214785
fbshipit-source-id: e7419be03ba0e20d95b47c11e41789636aa6e916
Summary:
Queues Problem Intro:
UIManager queue is special queue because it has special relationship with
the Main queue.
This particular relationship comes from two key factors:
1. UIManager initiates execution of many blocks on the Main queue;
2. In some cases, we want to initiate (and wait for) some UIManager's work *synchronously* from
the Main queue.
So, how can we meet these criteria?
"Pseudo UIManager queue" comes to rescue!
"Pseudo UIManager queue" means safe execution of typical UIManager's work
on the Main queue while the UIManager queue is explicitly blocked for preventing
simultaneous/concurrent memory access.
So, how can we technically do this?
1. `RCTAssertUIManagerQueue` is okay with execution on both actual UIManager and
Pseudo UIManager queues.
2. Both `RCTExecuteOnUIManagerQueue` and `RCTUnsafeExecuteOnUIManagerQueueSync`
execute given block *synchronously* if they were called on actual UIManager
or Pseudo UIManager queues.
3. `RCTExecuteOnMainQueue` executes given block *synchronously* if we already on
the Main queue.
4. `RCTUnsafeExecuteOnUIManagerQueueSync` is smart enough to do the trick:
It detects calling on the Main queue and in this case, instead of doing
trivial *synchronous* dispatch, it does:
- Block the Main queue;
- Dispatch the special block on UIManager queue to block the queue and
concurrent memory access;
- Execute the given block on the Main queue;
- Unblock the UIManager queue.
Imagine the analogy: We have two queues: the Main one and UIManager one.
And these queues are two lanes of railway go in parallel. Then,
at some point, we merge UIManager lane with the Main lane, and all cars use
the unified the Main lane.
And then we split lanes again.
This solution assumes that the code running on UIManager queue will never
*explicitly* block the Main queue via calling `RCTUnsafeExecuteOnMainQueueSync`.
Otherwise, it can cause a deadlock.
Reviewed By: mmmulani
Differential Revision: D5935464
fbshipit-source-id: 6a60ff236280d825b4e2b101f06222266097b97f
Summary:
This is pretty normal and harmless case, we should not crash here.
I plan to refactor similar places in this file soon.
Reviewed By: AaaChiuuu
Differential Revision: D5983443
fbshipit-source-id: 922fea8ed12ebef45d249f16739aa81fe3254f19
Summary:
If you are a product developer and you need to fix your app's issues related to iPhone X limitations asap,
you can temporary use `DeviceInfo.isIPhoneX_deprecated`.
You can, but you should not. Please consider use new <SafeAreaView>.
This prop was initially named so ugly because we are trying to discourage the community to use it.
However, we understand that sometimes we need a "band-aid" to prevent our apps bleeding.
Note: This prop (DeviceInfo.isIPhoneX_deprecated) will be removed completely after 06/18.
Reviewed By: fkgozali
Differential Revision: D5946329
fbshipit-source-id: 5d6dcaf0e2d175327d59cde4b5ec2e01cd77ec70
Summary:
Reasons:
* It is more clear and readable;
* It is more semantical;
* It allows us to add some magic (and complexity, to be fair) to the RCTExecuteOnUIManagerQueue. (See next diffs in the stack.)
Reviewed By: javache
Differential Revision: D5935466
fbshipit-source-id: aeb18d6e11c047eb19182a656da581b6ce073602
Summary:
This fixes [#15801](https://github.com/facebook/react-native/issues/15801)
We ran into a strange crash on iOS (debug only). After removing the clutter I was able to reproduce it in a tiny app. You can check it out [here.](https://github.com/simonracz/textinput_stress)
The UI in JS and native are not always in sync (which is okay). Due to this, a native view might call back into JS, which is no longer present in the shadow view hierarchy there. I think this should be also okay.
TextInput in some cases calls into [setIntrinsicContentView](6d67e2dbbc/React/Modules/RCTUIManager.m (L382)), where it triggers an overly enthusiastic `NSAssert` and crashes the app.
Check out [textinput_stress](https://github.com/simonracz/textinput_stress)
Rotate the simulator a few times to see the crash or the lack of crash.
Closes https://github.com/facebook/react-native/pull/16170
Differential Revision: D5959776
Pulled By: shergin
fbshipit-source-id: f39f5a3f1d86b330ecf7cbccd90871bc01fd69d9
Summary:
This is required for D5874536, wherein I'll be introducing direction-aware props for borders.
When a view's border changes due to a direction update, only the frames of its children update. Therefore, only the children `UIView`s get a chance to be re-rendered. This is incorrect because the view that's had its borders changed also needs to re-render. So, I keep a track of the layout direction in a property on all shadow views. Then, when I update that prop within `applyLayoutNode`, I push shadow views into the `viewsWithNewFrames` set.
Reviewed By: mmmulani
Differential Revision: D5944488
fbshipit-source-id: 3f23e9973f3555612920703cdb6cec38e6360d2d
Dmitry Zakharov
Valentin Shergin
Semen Zhydenko
Pieter De Baets
Ben Nham
Alex Dvornikov
Ayush Sood
Paco Estevez Garcia
Rahul Ramachandran
Yujie Liu
Nicolas Charpentier
Adam Ernst
Ramanpreet Nara
Frank Manns
Simon Racz
Jason Carreiro
