Summary:
Pull Request resolved: https://github.com/facebook/react-native/pull/45334
Adds support for the `xml` file extension as a loadable asset, and lets Flow treat the type signature as an image
Changelog: [Internal]
Reviewed By: robhogan
Differential Revision: D58261501
fbshipit-source-id: b29cffee81d0438827529711e86267edd5d2a0f7
Summary:
I'm picking 1630b5c743 in main as it's currently missing (available only on `0.75-stable`).
## Changelog:
[INTERNAL] - Update testing scripts to work with any version of React native
Pull Request resolved: https://github.com/facebook/react-native/pull/45498
Test Plan: Nothing to test as this is a backport
Reviewed By: cipolleschi
Differential Revision: D59861440
Pulled By: cortinico
fbshipit-source-id: 57f642c66c7a6976f5a5cd53debaeb2e461a1f30
Summary:
Pull Request resolved: https://github.com/facebook/react-native/pull/45457
With this change, we are implementing in Android a similar logic that we implemented in iOS.
1. When the user stops dragging a scroll view, it tells native animated modle that a scroll has finished
2. NativeAnimated module asks to the NativeAnimatedNodesModule if there are native node listening to the scroll
3. In case they are, it emits an event to JS
4. JS listen to the events and resync the Shadow Tree and the Native Tree (this implemented in a previous change)
## Changelog
[Android][Fixed] - Sync the Shadow Tree and the Native Tree with Native animation when scroll is driving the animation
Reviewed By: sammy-SC
Differential Revision: D59756577
fbshipit-source-id: e558557b477f4da9da1f89fb31ba86d0ea1390a3
Summary:
Pull Request resolved: https://github.com/facebook/react-native/pull/45383
This is the second step required to fix the onTouchMove event in the new architecture.
In this change, we are retrieving the list of nodes that are connected by the animation, and we are sending an event to the nodes so that we can trigger the commit.
## Changelog
[iOS][Added] - retrieve the tags of the nodes connected by the animation and send them to JS
Reviewed By: sammy-SC
Differential Revision: D59524617
fbshipit-source-id: 584317afa8e4cf0ad9f98f38e4e5d436c5fe3ac5
Summary:
Pull Request resolved: https://github.com/facebook/react-native/pull/45382
This change is the first step to tr and solve the pressability's `onTouchMove` issue with animation driven natively in the New Architecture.
The idea is to trigger a special event from native to let JS know that a scroll event has ended (`scrollViewDidEndDragging` or `scrollViewdidEndDecelerating`).
When this happens, we need to send an event to JS to let him know that it has to sync the Native Tree with the Shadow Tree.
Step 2 is to connect Native with JS
## Changelog:
[iOS][Added] - Send onScrollEnded event to NativeTurboAnimatedModule
Reviewed By: sammy-SC
Differential Revision: D59459989
fbshipit-source-id: cb425cddcdaa9d700ec40accaf4ab3ce1f3c5038
Summary:
Pull Request resolved: https://github.com/facebook/react-native/pull/45495
Unify type definitions to semver@7, to fix a [type-sync](https://www.internalfb.com/intern/test/281475050813096/) test that was broken by D59378011. The test is very simple and doesn't actually understand the typing.
I don't believe there is a significant difference in the typing, esp. with how we're using it. Flow will tell us if this is the case though (🏖️🏰).
Changelog: [Internal]
Reviewed By: huntie
Differential Revision: D59855434
fbshipit-source-id: ae3c6b7aa81b3cde25468d72a7922bcb2b6f652f
Summary:
Pull Request resolved: https://github.com/facebook/react-native/pull/45478
Some code is far too recursive. This consumes buffer space and causes problems for the Perfetto frontend. Let's limit it to 50 frames.
Changelog: [Internal]
Reviewed By: rubennorte, sammy-SC
Differential Revision: D59813638
fbshipit-source-id: 69068f9c2193d706ec0cc00ffc0d5950ae094e05
Summary:
Pull Request resolved: https://github.com/facebook/react-native/pull/45474
Our actions inputs are now a mixture of different casing.
I'm moving everything to be kebab-case
Changelog:
[Internal] [Changed] - Composite actions inputs should be kebab-case
Reviewed By: cipolleschi
Differential Revision: D59809181
fbshipit-source-id: af6d541c2b4f5fa162dcde412fb8808bae1ef2d3
Summary:
Pull Request resolved: https://github.com/facebook/react-native/pull/45480
We currently use the default GITHUB_ACTION which makes a lot of interaction
appear as user "GitHub Actions". Instead we could use the `REACT_NATIVE_BOT_GITHUB_TOKEN`
which we have as secret so the bot will actually perform the actions.
Changelog:
[Internal] [Changed] - Act as react-native-bot on all the actions
Reviewed By: cipolleschi
Differential Revision: D59815201
fbshipit-source-id: 702b121ec07d0db10abf25e23f7ddf5658dd5d62
Summary:
Pull Request resolved: https://github.com/facebook/react-native/pull/45491
The build_android job was missing the sonatype credentials so could not
publish a -SNAPSHOT version. This fixes it.
Changelog:
[Internal] [Changed] - Unbreak nightlies by fixing secrets
Reviewed By: cipolleschi
Differential Revision: D59848810
fbshipit-source-id: 2cc1d03b090d0aeb3886590ec0696f9c3a6556b9
Summary:
Pull Request resolved: https://github.com/facebook/react-native/pull/45485
## Changelog:
the new functions I introduced in https://github.com/facebook/react-native/pull/45365 to read color channel value, like `alphaFromColor`, will return float between 0~255 on iOS and 8bit unsigned ints on other platforms, because of different platform implementation. However that way we cannot assume consistent return value across platforms, which kind of contradicts with RN's cross platform design.
so here I make `alphaFromColor` in Color.h (the cross platform method) to always return uint8_t, while still allow `alphaFromHostPlatformColor` to return different result
[Internal] [Fixed] -
Reviewed By: christophpurrer
Differential Revision: D59826142
fbshipit-source-id: 4401918be29980474bdc8601443ae33155710f22
Summary:
FIXES [45404](https://github.com/facebook/react-native/issues/45404)
sending headers from Image component not working in new arch , implementation was missing
```
<Image
source={{
uri: "http://localhost:3000/image",
headers: {
"test-header": 'test',
"hello":"tested"
}
}}
style={{
width: 300,
height: 300,
}}
/>
```
## Changelog:
[IOS] [ADDED]- sending missing **headers** field with **Image** component in fabric
Pull Request resolved: https://github.com/facebook/react-native/pull/45415
Test Plan:
# Tested
Attaching the below video to show how headers are getting received on server from Image component running in new arch
https://github.com/user-attachments/assets/c816265d-0bb5-4670-bde0-cfec72d7618f
Reviewed By: javache, cipolleschi
Differential Revision: D59807462
Pulled By: blakef
fbshipit-source-id: dffa4d80db58de6a81947ac876aa76ec7e62dd48
Summary:
Pull Request resolved: https://github.com/facebook/react-native/pull/45445
Roll our own base64 encoding and revert D59685218, which fixed the missing Folly dependency implied by D54309633.
I assumed Folly base64 was already elsewhere in RN but given it isn't, and we only need simple, non-perf-sensitive encoding for the debugger (not the SIMD or delegated implementations, or decoding), it might be best to just include our own encoder.
Changelog: [Internal]
Reviewed By: cipolleschi
Differential Revision: D58323859
fbshipit-source-id: 5ce98561e9ced82765e8e7c18e5d2ebfa8148c8c
Summary:
Pull Request resolved: https://github.com/facebook/react-native/pull/45468
This should greatly reduce the time spent on build_npm_package
because we're moving all the publishing logic to build_android.
I need to do a bit more testing with nightlies to make sure that everything is published correctly.
Changelog:
[Internal] [Changed] - Make build_android publish to the stating repositories
Reviewed By: cipolleschi
Differential Revision: D59804015
fbshipit-source-id: be3f0b6e16f5fdbf760ec7a5e16c8e258e06dd28
Summary:
The `LinkingModule.getInitialURL` method is declared as returning a `Proimise<string>` but looking at the implementation it can also return null:
a96272e27e/packages/react-native/Libraries/LinkingIOS/RCTLinkingManager.mm (L166)
which is returned by
a96272e27e/packages/react-native/Libraries/Linking/Linking.js (L96)
which happens to expand the type to the correct type for most external users.
React-native-window's turbomodule codegen is more type strict, and so cannot return null with the current spec.
## Changelog:
[INTERNAL] [FIXED] - Use more accurate type in LinkingManager spec file
Pull Request resolved: https://github.com/facebook/react-native/pull/45450
Test Plan:
Type change only, so should only affect build time change which will be caught by CI.
Have verified that this type change allows react-native-windows to return null.
Reviewed By: robhogan
Differential Revision: D59816089
Pulled By: zeyap
fbshipit-source-id: 9810f150ce84b503883e72b1f29518d2e62258b6
Summary:
Pull Request resolved: https://github.com/facebook/react-native/pull/45426
The initial implementation of `Network.loadNetworkResource` and the accompanying `IO.read` (D54202854) base64-encodes all data as if it is binary. This is the more general case, and we'll continue to base64-encode non-text resources.
In the common case of text resources (particularly JS and JSON), it'd be preferable to do as Chrome does and send UTF-8 over the wire directly. This has a few performance benefits:
- Less CPU and RAM footprint on device (UTF-8 truncation is constant-time, fast, and in-place), similarly less decoding for the frontend.
- 25% less data per chunk (base64 encodes 3 bytes as 4 characters), implies up to 25% fewer network round trips for large resources.
It also has the benefit of being human-readable in the CDP protocol inspector.
## Determining whether data is text
We use exactly Chromium's heuristic for this (code pointers in comments), which is based only on the `Content-Type` header, and assuming any text mime type is UTF-8.
## UTF-8 truncation
The slight implementation complexity here is that `IO.read` requests may specify a maximum number of bytes, and so we must slice a raw buffer up into valid UTF-8 sequences. This turns out to be fairly simple and cheap:
1. Naively truncate the buffer, inspect the last byte
2. If the last byte has topmost bit =0, it's ASCII (single byte) and we're done.
3. Otherwise, look back at most 3 bytes to find the first byte of the code point (topmost bits 11), counting the number of "continuationBytes" at the end of our buffer. If we don't find one within 3 bytes then the string isn't UTF-8 - throw.
4. Read the code point length, which is encoded into the first byte.
5. Resize to remove the last code point fragment, unless it terminates correctly exactly at the end of our buffer.
## Edge cases + divergence from Chrome
Chrome's behaviour here in at least one case is questionable and we intentionally differ:
- If a response has header "content-type: text/plain" but content eg`0x80` (not valid UTF-8), Chrome will respond to an `IO.read` with `{ "data": "", "base64Encoded": false, "eof": false }`, ie an empty string, but will move its internal pointer such that the next or some subsequent `IO.read` will have `"eof": true`. To the client, this is indistinguishable from a successfully received resource, when in fact it is effectively corrupted.
- Instead, we respond with a CDP error to the `IO.read`. We do not immediately cancel the request or discard data, since not all `IO.read` errors are necessarily fatal. I've verified that CDT sends `IO.close` after an error, so we'll clean up that way (this isn't strictly guaranteed by any spec, but nor is `IO.close` after a resource is successfully consumed).
Changelog:
[General][Added] Debugger: Support text responses to CDP `IO.read` requests
Reviewed By: hoxyq
Differential Revision: D58323790
fbshipit-source-id: def8bf8426266f16bb305d836a6efe8927a9dfc4
Summary:
When running the entire unit test suite of `RNTesterPods` with `TSan`, I saw that occasionally a data race was detected on line 843 of `RCTImageLoader`. It seems the completion handler that does contain the lock around `cancelLoad` is called concurrently with the value being assigned on line 843. Here there is no lock in place.
Here is the output of `TSan` when I comment out my fix:
```
WARNING: ThreadSanitizer: data race (pid=72490)
Write of size 8 at 0x000144151ce8 by main thread:
#0 -[RCTImageLoader loadImageWithURLRequest:size:scale:clipped:resizeMode:priority:attribution:progressBlock:partialLoadBlock:completionBlock:] <null> (RNTesterUnitTests:arm64+0x16e8030)
https://github.com/facebook/react-native/issues/1 -[RCTImageLoader loadImageWithURLRequest:size:scale:clipped:resizeMode:priority:progressBlock:partialLoadBlock:completionBlock:] <null> (RNTesterUnitTests:arm64+0x16df8dc)
https://github.com/facebook/react-native/issues/2 -[RCTImageLoader loadImageWithURLRequest:size:scale:clipped:resizeMode:progressBlock:partialLoadBlock:completionBlock:] <null> (RNTesterUnitTests:arm64+0x16df534)
https://github.com/facebook/react-native/issues/3 -[RCTImageLoaderTests testImageLoaderUsesImageURLLoaderWithHighestPriority] <null> (RNTesterUnitTests:arm64+0x7cb8)
https://github.com/facebook/react-native/issues/4 __invoking___ <null> (CoreFoundation:arm64+0x13371c)
Previous write of size 8 at 0x000144151ce8 by thread T4 (mutexes: write M0):
#0 __140-[RCTImageLoader loadImageWithURLRequest:size:scale:clipped:resizeMode:priority:attribution:progressBlock:partialLoadBlock:completionBlock:]_block_invoke_2 <null> (RNTesterUnitTests:arm64+0x16e8894)
https://github.com/facebook/react-native/issues/1 __139-[RCTImageLoader _loadImageOrDataWithURLRequest:size:scale:resizeMode:priority:attribution:progressBlock:partialLoadBlock:completionBlock:]_block_invoke <null> (RNTesterUnitTests:arm64+0x16e3430)
https://github.com/facebook/react-native/issues/2 __139-[RCTImageLoader _loadImageOrDataWithURLRequest:size:scale:resizeMode:priority:attribution:progressBlock:partialLoadBlock:completionBlock:]_block_invoke_3 <null> (RNTesterUnitTests:arm64+0x16e52a8)
https://github.com/facebook/react-native/issues/3 __75-[RCTImageLoaderTests testImageLoaderUsesImageURLLoaderWithHighestPriority]_block_invoke_2 <null> (RNTesterUnitTests:arm64+0x7f24)
https://github.com/facebook/react-native/issues/4 -[RCTConcreteImageURLLoader loadImageForURL:size:scale:resizeMode:progressHandler:partialLoadHandler:completionHandler:] <null> (RNTesterUnitTests:arm64+0x6c470)
https://github.com/facebook/react-native/issues/5 __139-[RCTImageLoader _loadImageOrDataWithURLRequest:size:scale:resizeMode:priority:attribution:progressBlock:partialLoadBlock:completionBlock:]_block_invoke.172 <null> (RNTesterUnitTests:arm64+0x16e4964)
https://github.com/facebook/react-native/issues/6 __tsan::invoke_and_release_block(void*) <null> (libclang_rt.tsan_iossim_dynamic.dylib:arm64+0x77ee0)
https://github.com/facebook/react-native/issues/7 _dispatch_client_callout <null> (libdispatch.dylib:arm64+0x3974)
Location is heap block of size 48 at 0x000144151cc0 allocated by main thread:
#0 malloc <null> (libclang_rt.tsan_iossim_dynamic.dylib:arm64+0x4fa48)
https://github.com/facebook/react-native/issues/1 _malloc_type_malloc_outlined <null> (libsystem_malloc.dylib:arm64+0xf3ec)
https://github.com/facebook/react-native/issues/2 _call_copy_helpers_excp <null> (libsystem_blocks.dylib:arm64+0x10b4)
https://github.com/facebook/react-native/issues/3 -[RCTImageLoader loadImageWithURLRequest:size:scale:clipped:resizeMode:priority:progressBlock:partialLoadBlock:completionBlock:] <null> (RNTesterUnitTests:arm64+0x16df8dc)
https://github.com/facebook/react-native/issues/4 -[RCTImageLoader loadImageWithURLRequest:size:scale:clipped:resizeMode:progressBlock:partialLoadBlock:completionBlock:] <null> (RNTesterUnitTests:arm64+0x16df534)
https://github.com/facebook/react-native/issues/5 -[RCTImageLoaderTests testImageLoaderUsesImageURLLoaderWithHighestPriority] <null> (RNTesterUnitTests:arm64+0x7cb8)
https://github.com/facebook/react-native/issues/6 __invoking___ <null> (CoreFoundation:arm64+0x13371c)
Mutex M0 (0x000108f316e8) created at:
#0 pthread_mutex_init <null> (libclang_rt.tsan_iossim_dynamic.dylib:arm64+0x2cc98)
https://github.com/facebook/react-native/issues/1 -[NSLock init] <null> (Foundation:arm64+0x5ca14c)
https://github.com/facebook/react-native/issues/2 -[RCTImageLoader loadImageWithURLRequest:size:scale:clipped:resizeMode:priority:progressBlock:partialLoadBlock:completionBlock:] <null> (RNTesterUnitTests:arm64+0x16df8dc)
https://github.com/facebook/react-native/issues/3 -[RCTImageLoader loadImageWithURLRequest:size:scale:clipped:resizeMode:progressBlock:partialLoadBlock:completionBlock:] <null> (RNTesterUnitTests:arm64+0x16df534)
https://github.com/facebook/react-native/issues/4 -[RCTImageLoaderTests testImageLoaderUsesImageURLLoaderWithHighestPriority] <null> (RNTesterUnitTests:arm64+0x7cb8)
https://github.com/facebook/react-native/issues/5 __invoking___ <null> (CoreFoundation:arm64+0x13371c)
Thread T4 (tid=10935088, running) is a GCD worker thread
```
## Changelog:
[iOS][Fixed] Data race in `RCTImageLoader` related to assignment of cancellation block.
Pull Request resolved: https://github.com/facebook/react-native/pull/45454
Test Plan: There are already tests in place for `RCTImageLoader`. I hope these will cover the fix.
Reviewed By: realsoelynn
Differential Revision: D59816000
Pulled By: zeyap
fbshipit-source-id: f959d472eb60f83f39ced6711ee395949ab37e7c
Summary:
Pull Request resolved: https://github.com/facebook/react-native/pull/45479
This just prints a summary on PRs if the Gradle task fails so it's easier to jump directly
to the failure.
Changelog:
[Internal] [Changed] - Enable add-job-summary-as-pr-comment for failed jobs
Reviewed By: cipolleschi
Differential Revision: D59812845
fbshipit-source-id: 2069a1c8db7d264ca1af3c1182fa443cb0a69646
Summary:
Pull Request resolved: https://github.com/facebook/react-native/pull/45469
Enables React Native DevTools by default on `main`, ahead of the 0.76 release. We've observed no new bug reports internally over the last two weeks, and are moving forward with our rollout plan.
Changelog: [Internal]
Reviewed By: blakef
Differential Revision: D59804882
fbshipit-source-id: 0cb6302f4d940718786db2e5d8fb652fae6a8c54
Summary:
Pull Request resolved: https://github.com/facebook/react-native/pull/45475
Instead of zipping all the RNTester's APK together, let's
upload them per buildVariant so it's easier to retrieve them later.
Changelog:
[Internal] [Changed] - Split the rntester APK artifacts in 4
Reviewed By: cipolleschi
Differential Revision: D59809721
fbshipit-source-id: 2d375475d5cee71c212f4d1f3a4a9edf3442358f
Summary:
We don't need to specify a minor/patch for actions/upload-artifact.
We also have all sorts of different versions scattered around the codebase.
This aligns them to the latest sable in the 4.x series.
Changelog:
[Internal] [Changed] - actions/upload-artifact to v4.x
Reviewed By: cipolleschi
Differential Revision: D59811525
fbshipit-source-id: 7264db097bcb2ff34b3ace467996e8308c0f2034
Summary:
Pull Request resolved: https://github.com/facebook/react-native/pull/45463
Changelog: [Internal]
State updates will clone shadow nodes for an shadow tree revision that is outdated.
This can lead to accessing deallocated shadow node references because the JS renderer committed a newer revision and deallocated the one used by the pending state update.
By using a weak pointer to hold a reference to the runtime shadow node reference, we can only update references for wrappers that are still valid.
Reviewed By: javache
Differential Revision: D59804999
fbshipit-source-id: 89c9967d139d3cac7d7252994beae419bc591e79
Summary:
Pull Request resolved: https://github.com/facebook/react-native/pull/45413
## Changes
Add an example in RNTester to test that pressability with NativeDrivers works properly.
## Context
The pressability handling is a bit peculiar.
We have to handle 3 main behaviors:
* `PressIn` -> `PressOut` => triggers the `onPress`
* `PressIn` -> move inside the rectangle -> `PressOut` => triggers the `onPress`
* `PressIn` -> move outside the rectangle -> `PressOut` => cancel `onPress`.
For the first case, we detect whether the press happens inside a component in the Native layer only. And everything works.
When a move is involved, we:
1. Detect the initial press in the Native layer
2. We move the coursor and we delegate the detection of whether we are inside of a rect or not to the JS layer
3. The JS layer asks the C++ layer about the layout and decide whether we are in case 2 (move but still inside the rect) or in case 3 (move but outside the rect).
The problem is that with `nativeDriver` and animations, the C++ layer doesn't know about where the receiver view actually is.
This results in issues like the one shown by [#36504](https://github.com/facebook/react-native/issues/36504), where the onMove is not handled correctly.
## Solution
The solution is to keep detecting whether we are in the receiver view or not in the Native layer and pass the receiver view position and size back to JS so that the JS layer don't have to jump to C++ to make this decision.
We decided to pass the frame information because the JS layer is adding some padding and configurations to the final rectangle and we don't want to lose those configurations.
## Changelog
[General][Added] - Add example in RNTester to show that pressability works properly with NativeDrivers
Reviewed By: sammy-SC
Differential Revision: D58182480
fbshipit-source-id: 9ca4fb9a3ca1a8af52ccbe208cbfe8434175f87d
Summary:
CCI on main is broken. We suspect that's due to cache issues which restore a wrong layout for the Folly pod.
This PR is an attempt to fix it
## Changelog:
[Internal] - Fix missing folly base 64
Pull Request resolved: https://github.com/facebook/react-native/pull/45460
Test Plan: CCI and GHA are green
Reviewed By: sammy-SC, huntie
Differential Revision: D59804748
Pulled By: cipolleschi
fbshipit-source-id: 44d6b169cf3319f4d7ee9e0a5833f07bc6ba4bb3
Summary:
Pull Request resolved: https://github.com/facebook/react-native/pull/45458
Checking for packager is an async operation, which may return when we've already destroyed the ReactInstanceManager. Prevent the CatalystInstance from being created if the ReactInstanceManager has been invalidated.
Changelog: [Internal]
Reviewed By: cortinico
Differential Revision: D59753247
fbshipit-source-id: e3ac2b6dd142330e2d4051519b9863584b33f8a6
Summary:
Pull Request resolved: https://github.com/facebook/react-native/pull/45459
Seeing failures on main with GHA for gradle builds, in the Post Setup gradle step:
```
Could not get unknown property 'cleanupTime' for object of type org.gradle.api.internal.cache.DefaultCacheConfigurations.
```
This is a speculative change to get CI back to stable.
Changelog: [Internal]
Reviewed By: cortinico
Differential Revision: D59802517
fbshipit-source-id: c7b5259397fddef9420570043263e92f21718934
Summary:
This change factors out build-android job so we can reuse it
## Changelog:
[Internal] - Factor out the build-android job for code reuse
Pull Request resolved: https://github.com/facebook/react-native/pull/45455
Test Plan: GHA are green
Reviewed By: blakef
Differential Revision: D59802116
Pulled By: cipolleschi
fbshipit-source-id: 12ece8004da3bfd1f275b4af8e9822d4b0ccc0f0
Summary:
Pull Request resolved: https://github.com/facebook/react-native/pull/45448
changelog: [internal]
I would like to experiment with executing insert mount instructions after layout and state update. It has shown small improvement in local testing
Reviewed By: mdvacca
Differential Revision: D59582123
fbshipit-source-id: 3ee6ec12a533a287ed32f7373863175f3a107548
Summary:
Pull Request resolved: https://github.com/facebook/react-native/pull/45410
changelog: [internal]
For better performance, let's test setting up the animation graph from passive effects. This will delay the work, not blocking the paint.
Reviewed By: rubennorte
Differential Revision: D59644374
fbshipit-source-id: ff951ee7c1a1d47e13c55fc7c7f6c0690aa465f7
Summary:
Pull Request resolved: https://github.com/facebook/react-native/pull/45444
## Changelog:
[Internal] -
Even though `ScrollView.scrollIndicatorInsets` isn't supported on the vanilla Android platform, it still may be used on some other variations of it, which means that the changes may not potentially find the way into C++/Fabric, opening a door of all kinds of weird corner case issues.
Reviewed By: christophpurrer
Differential Revision: D59761458
fbshipit-source-id: 4dae5c96791ca924d589a3d803d8fa60fdca1b67
Summary:
Pull Request resolved: https://github.com/facebook/react-native/pull/45304
Add support for most keyword values of mix-blend-mode on iOS and added RNTester Example
Missing compositing operators and global values
Changelog: [Internal]
Reviewed By: NickGerleman
Differential Revision: D59402969
fbshipit-source-id: b7e1aaed01fbf8f80e04ad0fa73d2ef63b5ad933
Summary:
Pull Request resolved: https://github.com/facebook/react-native/pull/45306
Adding missing drop-shadow test to rn-tester.
Added with alpha-hotdog image to show we are creating the shadow with the alpha channel of the view.
Changelog: [Internal]
Reviewed By: cortinico
Differential Revision: D59410148
fbshipit-source-id: 5a03ee84313979f99585b8ca7e07abf9cdbe2396
Summary:
Pull Request resolved: https://github.com/facebook/react-native/pull/45443
We are only building these out for Fabric. This means only one natural iOS impl, but that other Android bits will not work fully correctly on Paper (like setting containing block for filter element). This also means we can remove view configs once we're on Fabric CSS parser. We will do the same for boxShadow once that is ready.
Changelog: [internal]
Reviewed By: mdvacca
Differential Revision: D59762282
fbshipit-source-id: 14ce07f04b822c6aee908894c9081419594fc484
Summary:
Pull Request resolved: https://github.com/facebook/react-native/pull/45421
RNTester contains Android resources that are loaded by name and not resolved by Metro. As a result, these assets are not automatically linked when RNTester JS code is embedded in other projects. This is considered "legacy" loading and is generally discouraged, but is still showcased as an alernative way of loading resources.
I also modified the Image test to ensure that flag status is printed so it's obvious why the vector drawable hasn't loaded.
Changelog: [Internal]
Reviewed By: javache
Differential Revision: D59585555
fbshipit-source-id: d42fb44d8846d8e7c7aa01dca4cec89ae85a9195
Summary:
Pull Request resolved: https://github.com/facebook/react-native/pull/45442
This deprecates also `DeveloperSettings.isStartSamplingProfilerOnInit`
so we can remove it in a future version of React Native.
This field is unused so you should not be using it at all.
Changelog:
[Android] [Changed] - Deprecate DeveloperSettings.isStartSamplingProfilerOnInit
Reviewed By: blakef
Differential Revision: D59757500
fbshipit-source-id: dc879ba46f2f937e5f259a4101646c2f060db548
Summary:
## Summary
Now that HostContext determination for Fabric is a DEV-only behavior, we
can move the HostContext determination to resolve from the ViewConfig
for a given type. Doing this will allow arbitrary types to register
themselves as potential parents of raw text string children. This is the
first of two diffs for react as we'll:
1. Add the new property to the ViewConfig types
2. Update React Native to include the `supportsRawText` property for
`RCTText`, `RCTVirtualText`, `AndroidTextInput`, etc.
3. Switch the behavior of react to read from the ViewConfig rather than
a static list of types.
Changelog: [Internal]
## Test Plan
- yarn test
- yarn test --prod
- Pulled change into react-native, added `supportsRawText` props to
RCTText, RCTVirtualText, etc. ViewConfigs and confirmed everything type
checks.
DiffTrain build for commit a5cc797b88.
bypass-github-export-checks
Reviewed By: poteto
Differential Revision: D59641180
Pulled By: rozele
fbshipit-source-id: a3ddb1bc810a70d5f782e708cb845e3eae136d78
Summary:
Pull Request resolved: https://github.com/facebook/react-native/pull/45365
## Changelog
As a followup to https://github.com/facebook/react-native/pull/45139
There I only implemented `hostPlatformColorFromRGBA` and `alpha/red/green/blueFromHostPlatformColor` on cxx platform, then used cxx platform specific method at some places. but really I should implement and use methods in `Color.h` that are platform agnostic
* cxx/android: platform color format is int32_t, RGBA are 8bit unsigned int ([0,255])
* windows: platform color format is `winrt::Windows::UI::Color` where RGBA props are 8bit unsigned ints ([0,255])
* apple: platform color format is `UIColor` where RGBA props are floats in [0,1]
[Internal]
previous change D58872165
Reviewed By: christophpurrer
Differential Revision: D59593659
fbshipit-source-id: 5d18419039817510e607d4e3f632c207d25c30a3
Summary:
Pull Request resolved: https://github.com/facebook/react-native/pull/45436
changelog: [internal]
In D58355433, delete tree experiment was cleaned up. These are the remainings of the experiment that are no longer used. Let's delete them to clean up dead code
Reviewed By: javache
Differential Revision: D59732137
fbshipit-source-id: a22c0b14eda70e62817e80224f367ccb9006acc9
Summary:
Pull Request resolved: https://github.com/facebook/react-native/pull/45354
Fresco has indicated that they have no plans to support loading vector assets and similar drawable types in Drawee-backed views ([issue](https://github.com/facebook/fresco/issues/329), [issue](https://github.com/facebook/fresco/issues/1463), [issue](https://github.com/facebook/fresco/issues/2463)). Guidance has been to instead load the vector drawable onto the backing image view ourselves. On the React Native side, having the ability to load vector drawables has been requested many times ([issue](https://github.com/facebook/react-native/issues/16651), [issue](https://github.com/facebook/react-native/issues/27502)).
I went this route over using a custom Fresco decoder for XML assets because vector drawables are compiled down into binary XML and I couldn't find a trivial, performant way to parse those files in a context-aware manner. This change only accounts for vector drawables, not any of the other XML-based drawable types (layer lists, level lists, state lists, 9-patch, etc.). Support could be added easily in the future by expanding the `getDrawableIfUnsupported` function.
## Changelog
[Android] [Added] - Added support for rendering XML assets provided to `Image`
Reviewed By: javache
Differential Revision: D59530172
fbshipit-source-id: 3d427c06238287e0a3b7f9570ac20e43d76126c7
Summary:
This is just a minor nit to make it easier to copy-n-paste
## Changelog:
[INTERNAL] - Remove extra dot from close-pr.yml
Pull Request resolved: https://github.com/facebook/react-native/pull/45441
Test Plan: N/A
Reviewed By: cipolleschi
Differential Revision: D59752879
Pulled By: cortinico
fbshipit-source-id: caa398010b64024e2a0259d177762fd76082507f
