Summary: No new errors in this version. Just removed a bunch of unused suppressions
Reviewed By: TheSavior
Differential Revision: D8754160
fbshipit-source-id: 2f02240b6d65edecba5d9ed603c7703462547a7f
Summary:
Original commit changeset: 0b0b3a2d7b80
This constant is still in use at Facebook. Its removal has been pushed to sometime in the future.
Reviewed By: mdvacca
Differential Revision: D8721213
fbshipit-source-id: d1197c96804e4d2dc96be27421e5248a2394cdac
Summary: There was an inverted expression leading to logspew. Fix this.
Reviewed By: TheSavior
Differential Revision: D8758023
fbshipit-source-id: 7a83c68db6c95f2b5db6dcc7d7780fc66321b49e
Summary:
@public
... and it's as efficient as it was before.
The previous version of the algorithm used `sourceNode` reference to know the previous state of the node to call the algorithm recursively.
That wasn't so good because of several reasons:
- It was fragile because we had two different sources of the truth of the "previous state of the tree": committed tree and source node pointer;
- We had to store weak pointers to source nodes inside cloned nodes. That is not free in terms of performance;
- The old approach introduced a constraint that all previously used and now reinserted nodes must be cloned to update source node (otherwise, the algorithm would regenerate instructions recreating already existing subtrees);
- That cloning required access to `isSealed` flag which is supposed to be a debug-only thing (that actually affects performance and must be compile-out for release builds).
The new approach compares nodes with same react tag and naturally cloning-artifacts resilient.
Yes, the new approach uses a map of inserted nodes, but the previous one already had it (otherwise there is no way to tell which nodes should be "deleted"). And anyway, this is a very little map that exists for a very little period of time.
Reviewed By: mdvacca
Differential Revision: D8709953
fbshipit-source-id: 027abb326cf45f00f7bb0bbd7c4e612578268c66
Summary: Image source null which is in RC D8628053 has a bug which has a fix but didn't make to RC. Reverting so it can be cleaned up before going in RC.
Reviewed By: achen1
Differential Revision: D8751687
fbshipit-source-id: e08b23a031455be23047880871813bdc840542dd
Summary: Revert the order of "remove mount items", to ensure views are removed from high index to low index.
Reviewed By: shergin
Differential Revision: D8742796
fbshipit-source-id: 6e04c39386d290bf3958ee83256d4fbe23e2c4ca
Summary: This makes it possible to specify an optional Request.Builder when calling `downloadBundleFromURL` (the old method still works as usual).
Reviewed By: davidaurelio
Differential Revision: D8691303
fbshipit-source-id: 2fb2aecd3506355c6b3a72457a7bb9acfd03b18d
Summary:
Original commit changeset: 0b0b3a2d7b80
Reverting D8714400 which removed the `isIPhoneX_deprecated` flag, which is still widely used across the RN codebase https://fburl.com/biggrep/16jg5bzn
Reviewed By: hramos
Differential Revision: D8743401
fbshipit-source-id: cfc44bdd8019eda41e67ca573b20be417d121d12
Summary:
We're unifying on flow types, and propTypes require non-trivial resources to initialize in aggregate.
Some open source code might depend on extracting ScrollView.propTypes which will now fail. To fix, simplly remove these dependencies and use flow or typescript for verifying correct prop usage instead.
Reviewed By: TheSavior
Differential Revision: D8726371
fbshipit-source-id: 19e9540794db97a9e356615b544759a0753fd68c
Summary:
It looks like for some reason when ReactViewGroup uses setClipChildren(false), the ReactNativeARTSurface no longer redraws when its width is expanded.
Disable this new overflow behavior until the underlying issue is fixed.
Reviewed By: achen1
Differential Revision: D8739003
fbshipit-source-id: ffae0e3eb0cd8ce385eae33a87b5ba0325cae3c4
Summary:
In many projects we have different environments that we connect to. For these environments (for example: Test, Staging, Production) we have custom configurations (Debug and Release). While this is not a problem on Android, it is a problem on iOS. With the current implementation of the react-native-xcode.sh script, when using a custom Debug configuration, the app started on iOS device, can't contact the Packager. This pull request solves this issue.
Connect a real device, start the app with a custom debug configuration in Xcode. Shake and Reload.
[IOS][BUGFIX][./scripts/react-native-xcode.sh] Add support for connecting to the Packager when running the iOS app on device when using custom Debug configuration.
Closes https://github.com/facebook/react-native/pull/16451
Differential Revision: D8730537
Pulled By: hramos
fbshipit-source-id: a36007776e8fe9e401c38015040abd2c2bbd7c58
Summary:
Bump gradle to 4.4, version used by android studio 3.x and gradle plugin 3.x. This will help make migration easier and smoother.
CI: https://circleci.com/gh/dulmandakh/react-native/591
FYI: I found that gradle 4.8 works fine with gradle plugin 2.3, but gradle plugin 3.x displays warnings and asks to disable on demand configuration.
Closes https://github.com/facebook/react-native/pull/19899
Reviewed By: fkgozali
Differential Revision: D8697929
Pulled By: hramos
fbshipit-source-id: 21eb625ee1e4a1fa02aa22de4c36a07269ca467b
Summary: When image source doesn't have uri and is neither an array, it should return null.
Reviewed By: yungsters
Differential Revision: D8728688
fbshipit-source-id: 915c4f3f450907ee3435ac99b1fe9849738766da
Summary: `morgan` produces non-json output, which affects `js1 run` when ran with the `--json` flag.
Reviewed By: Kureev
Differential Revision: D8724830
fbshipit-source-id: 11377f6ef39341c658a7f905383398423a721630
Summary:
@public
Upgrades eslint to v5.0.1
Updated `eslint-plugin-eslint-comments`, which was necessary for eslint 5
Disabled the `prefer-const` rule while we wait for https://github.com/eslint/eslint/issues/10520 to be fixed/published.
Reviewed By: zertosh
Differential Revision: D8692838
fbshipit-source-id: fa0cae3e299af2350c8c30ceb94d70740ee84eab
Summary:
Adds a kill switch that reverts the default behavior of `overflow` to be hidden again. The intent of this kill switch is to give applications more time to migrate if necessary (e.g. if they are depending on third party packages with native components that are not compatible with `overflow` being visible by default).
To use the flag, simply set:
import com.facebook.react.views.view.ReactViewGroup;
// Somewhere in the initialization of your application.
ReactViewGroup.sDefaultOverflowHidden = true;
Reviewed By: achen1
Differential Revision: D8718963
fbshipit-source-id: 0eb9aee45dfe04e9ae34d86e3bedcd30a185ef82
Summary: We were supposed to pass in proper eventEmitter, but passed in one with null eventTarget instead, causing assertion failures when dispatching event.
Reviewed By: sebmarkbage, shergin
Differential Revision: D8720793
fbshipit-source-id: 891f3b2a2c76a6dd3e40039623c6e86991aad50b
Summary:
make prettier script works on windows
pass all current ci.
none
[GENERAL] [INTERNAL] [Script] - make prettier script works on windows.
Closes https://github.com/facebook/react-native/pull/19987
Differential Revision: D8714381
Pulled By: hramos
fbshipit-source-id: 0e5c5a6a5817094edd47918eb57d7f5a281560d6
Summary:
Cleanup the `isIPhoneX_deprecated` constant which was said to be removed by June 1st 2018.
Closes https://github.com/facebook/react-native/pull/19920
Differential Revision: D8714400
Pulled By: hramos
fbshipit-source-id: 0b0b3a2d7b8098baf0474afea230780c79b2fe14
Summary: Replace the union with a spread
Reviewed By: blairvanderhoof
Differential Revision: D8715694
fbshipit-source-id: 3b1ce893a065bfc7395b576e485a79f09bacc999
Summary:
Closes https://github.com/facebook/react-native/pull/20007
We removed support for providesModule annotations and maintained support for Haste names in installed modules via `providesModuleNodeModules`, but our default `hasteImpl` doesn't take them into account. We need to find a better way to override core components from plugins but meanwhile this adds an exception for react-native-windows in the default `hasteImpl` to unblock their upgrade to the latest RC.
Fixes https://github.com/facebook/metro/issues/188
Reviewed By: mjesun
Differential Revision: D8695207
fbshipit-source-id: 2ad6cb1e93e600880a148776ac45f6ebd7d205d3
Summary:
Scope of the diff:
1. Middleware
`react-native-github/local-cli` and `react-native-internal-cli` uses a very similar set of middlewares (internal cli extends github version), so I decided to move it to a standalone file (middleware manager) in order to remove duplications and increase readability.
2. Types
Seems that after Flow upgrade to version 0.68 there were many type issues to resolve, so all of them were auto-mocked. This is fine, but I'd like to see Flow assists me with `Metro.createServer` -> `Metro.runServer` migration. Hence, I decided to resolve flow mocks, related to runServer.
3. `runServer` signature
In `react-native-github` repo I cleaned up `runServer` signature by removing `startCallback` and `readyCallback` from the function parameters and moved them to `runServer` instead.
4. Replace `createServer` by `runServer`
In `react-native-github` repo, `createServer` has been replaced by `runServer`. __Some of arguments are not mapped__.
Note that this diff will partially break argument mapping. This is intentional. @[100000044482482:ives] will fix it with a new config package.
Reviewed By: mjesun
Differential Revision: D8711717
fbshipit-source-id: a843ab576360ff7242099910d8f25a9cb0a388c0
Summary:
@public
This compiles, but it works only on iOS for now.
Reviewed By: mdvacca
Differential Revision: D8655540
fbshipit-source-id: 7e9a73fadb317dd62298af6f347344ac4229a8a5
Summary:
@public
This compiles but this does not work.
To make it actually work we have to implement all missing functions in `Color.cpp` and co.
Reviewed By: fkgozali
Differential Revision: D8655537
fbshipit-source-id: 564fb7131445af81cf05407239dc6ba870cf6b83
Summary:
Fixes `ReactScrollView` so that it respects the drawing rect (i.e. the bounding box of the element).
In JavaScript, this is the backing view for `ScrollView` (vertical) on Android.
Reviewed By: fadinghorse
Differential Revision: D8710256
fbshipit-source-id: f3bd96e39b8569cfcb21e486944b70fdb57c12b6
Summary:
@public
We have this feature in the current version of RN, so it would be nice to support that in Fabric as well. This should save us tens of ms of views creation during mounting.
And that's quite easy to do!
Reviewed By: fkgozali
Differential Revision: D8701992
fbshipit-source-id: 4e3049df009ffd65bb43043de388e81795e5e559
Summary:
Removes the concept of instance handle. Instead we pass the event target
to createNode and don't pass it to subsequent clones.
The life time of the event target is managed by native (the event emitter).
It has to be released manually.
Reviewed By: shergin
Differential Revision: D8688330
fbshipit-source-id: e11b61f147ea9ca4dfb453fe07063ed06f24b7ac
Summary:
@public
Most of them are legit issues which should not be compilable anyways (but Clang tolerates thems).
Reviewed By: mdvacca
Differential Revision: D8655539
fbshipit-source-id: 645729fb9d6a120ce1ab2b07542abcdacd72320d
Summary:
@public
Suddenly, it is not supported on Android.
Luckelly `folly:to<std::string>()` is as good as `std::to_string()`.
Reviewed By: mdvacca
Differential Revision: D8655538
fbshipit-source-id: 2b3b970f6a261253aaa6b22dba8338dc66b7195d
Summary:
Adds support for the `overflow` style property on React Native for Android.
This is the second attempt to do this. See 6110a4cc75 (D8666509) for the first attempt.
Similar to the first attempt, this sets `setClipChildren(false)` by default on all `ViewGroup` instances. However, this differs in how it implements `overflow: hidden`. Instead of conditionally setting `setClipChildren`, this manually clips children to the `ViewGroup`'s bounds (which was incidentally what we were doing for background + border radius already).
Reviewed By: achen1
Differential Revision: D8690805
fbshipit-source-id: 58757825cd9d138c18c8758918d85b4ca1915f87
Summary:
Tidies up the hardcoded strings for referencing the `overflow` style values.
Also, the `OVERFLOW` case in the optimized view flattening code path is unnecessary because `OVERFLOW` is already in the `LAYOUT_ONLY_PROPS` set.
Reviewed By: achen1
Differential Revision: D8690804
fbshipit-source-id: 3befbe93ed761e57e45f9b50e59bffc8a29a407f
Summary:
Reverts D8666509. Unfortunately, I misunderstood `setClipChildren` on Android.
When set on a `ViewGroup`, `setClipChildren` configures whether its //children// — not itself — are clipped to their bounds. This is unlike `overflow` (as it behaves on iOS) which configures whether the view itself is clipped to its bounds.
But they are definitely related. In theory, I think we could implement `overflow` using `setClipChildren` by:
- Setting `setClipChildren(false)` by default. (This part, I got right.)
- When `overflow` is set to `hidden` on a `View`, we create an extra `ViewGroup` (child) within the normal `ViewGroup` (parent). Then, we can set `setClipChildren(true)` on the parent `ViewGroup` which will cause the child `ViewGroup` to be clipped to its bounds.
However, I think the tricky thing will be to create the child `ViewGroup` without incurring unintentional side effects.
I need to decide whether or not this is worth trying. The alternative is to add a new `clipChildren` boolean prop that is Android-only, but I really hate further bifurcating the platform. But for now, I am reverting my mistake.
Reviewed By: achen1
Differential Revision: D8690551
fbshipit-source-id: 1ba3bbcc5458ffbd5c475430ea0382b3fd0916b2
Summary:
Improves the examples in `ViewExample.js` that tests overflow behavior. Notable:
- Test view flattening behavior by setting `overflow` on views that only have other layout-only styles.
- Test the default behavior when `overflow` is not set at all.
Reviewed By: achen1
Differential Revision: D8690560
fbshipit-source-id: 6320ef51305952d13bf5724b369651fdfd32ff21
Summary: Minor cleanup of ViewExample.js in the RNTester.
Reviewed By: sahrens
Differential Revision: D8690133
fbshipit-source-id: d034f6d215679dac7f19fab90729bb7e7ef39edd
Summary:
This diff refactors the cloning mechanism for YogaNode used from Fabric UI renderer and RN iOS graphs.
Previously, we were cleaning the owner of the child's cloned node inside the C++ implementation of YogaNode. This was a mistake because this modified the last commited YogaTree, causing side effect in RN iOS graphs.
Reviewed By: shergin
Differential Revision: D8672627
fbshipit-source-id: c9902d00690e0361fd58aed84b506c42258bd995
Summary:
@public
From now, `babel-preset-react-native` is going to be called `metro-react-native-babel-preset` and it's going to use the same versioning system that other metro packages.
This solves a few problems:
* Automated publishing of this package: Currently we have to publish this package manually from the RN repository, so it's hard to make changes to `babel-preset-react-native` and have them available on `metro` (the plugin version has to be manually updated, the package has to be manually published to npm and metro has to be change to depend on the new version).
* Transforming package code before publishing: By being in Metro repository, we're going to transform its source code before publishing it to npm automatically using the same infra as other metro packages, which is going to prevent issues like the current one with Node v6 due to trailing commas added by prettier: https://circleci.com/gh/facebook/metro/2125
* Workspace usage: Now, this package is not going to be installed via npm internally, but instead be used directly from the yarn workspace.
Reviewed By: mjesun
Differential Revision: D8677254
fbshipit-source-id: 14abdd218af64056625c5a997458bfe51d2101bc
Summary:
… in VirtualizedList - fixes#16612
Issue is detailed in #16612
<!--
Thank you for sending the PR! We appreciate you spending the time to work on these changes.
Help us understand your motivation by explaining why you decided to make this change.
You can learn more about contributing to React Native here: http://facebook.github.io/react-native/docs/contributing.html
Happy contributing!
-->
I need `onViewableItemsChanged` to account for the length of any ListHeaderComponents
I couldn't find any tests that currently cover the use-case of a VirtualizedList + ListComponent + scroll with onViewableItemsChanged, so I have not added any tests - all previous tests pass however.
<!--
Help reviewers and the release process by writing your own release notes
**INTERNAL and MINOR tagged notes will not be included in the next version's final release notes.**
CATEGORY
[----------] TYPE
[ CLI ] [-------------] LOCATION
[ DOCS ] [ BREAKING ] [-------------]
[ GENERAL ] [ BUGFIX ] [-{Component}-]
[ INTERNAL ] [ ENHANCEMENT ] [ {File} ]
[ IOS ] [ FEATURE ] [ {Directory} ] |-----------|
[ ANDROID ] [ MINOR ] [ {Framework} ] - | {Message} |
[----------] [-------------] [-------------] |-----------|
[CATEGORY] [TYPE] [LOCATION] - MESSAGE
EXAMPLES:
[IOS] [BREAKING] [FlatList] - Change a thing that breaks other things
[ANDROID] [BUGFIX] [TextInput] - Did a thing to TextInput
[CLI] [FEATURE] [local-cli/info/info.js] - CLI easier to do things with
[DOCS] [BUGFIX] [GettingStarted.md] - Accidentally a thing/word
[GENERAL] [ENHANCEMENT] [Yoga] - Added new yoga thing/position
[INTERNAL] [FEATURE] [./scripts] - Added thing to script that nobody will see
-->
[GENERAL][BUGFIX][VirtualizedList] - account for `ListHeaderComponent` length
Closes https://github.com/facebook/react-native/pull/17415
Differential Revision: D8683555
Pulled By: hramos
fbshipit-source-id: 05df7b79c16e3c07c12468e782f3c4b0bdce7403
Gabe Levi
Yan Soares Couto
Wenting Hu
Wen-Chien Chen
Konstantin Raev
Valentin Shergin
Andrew Chen (Eng)
Himabindu Gadupudi
David Vacca
Florian Schoellhammer
Spencer Ahrens
Kevin Gozali
Alexey Kureev
Zdravko Nikolovski
Dulmandakh
Miguel Jimenez Esun
Peter van der Zee
Tim Yung
Darío Hereñú
gengjiawen
Janic Duplessis
Yunyu Lin
Rubén Norte
Sebastian Markbage
Rafael Oleza
Shane Osbourne