Refactor legacy update queue #17510
Closed
Refactor legacy update queue #17510
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
|
This pull request is automatically built and testable in CodeSandbox. To see build info of the built libraries, click here or the icon next to each commit SHA. Latest deployment of this branch, based on commit 99444a2:
|
Details of bundled changes.Comparing: 3c1efa0...99444a2 react-dom
react-native-renderer
react-test-renderer
react-noop-renderer
react-art
react-reconciler
ReactDOM: size: 0.0%, gzip: 0.0% Size changes (experimental) |
Details of bundled changes.Comparing: 3c1efa0...99444a2 react-reconciler
react-noop-renderer
react-dom
react-art
react-test-renderer
react-native-renderer
ReactDOM: size: -0.4%, gzip: 🔺+0.1% Size changes (stable) |
acdlite
force-pushed
the
queue-refactor
branch
3 times, most recently
from
1cc8f85
to
275cc2f
Compare
acdlite
force-pushed
the
queue-refactor
branch
from
275cc2f
to
f08b8ba
Compare
sebmarkbage
reviewed
Dec 6, 2019
| const first = pending.next; | ||
| if (first !== null) { | ||
| // Still circular. | ||
| update.next = first; |
There was a problem hiding this comment.
What's this about, we need this for it to stay circular, no? Maybe we're missing a test?
There was a problem hiding this comment.
All I did was remove the first !== null check. Which is unnecessary because update.next is never null.
There was a problem hiding this comment.
I think the diff makes it look like I deleted the assignment but really all I deleted was the type check that surrounds it.
There was a problem hiding this comment.
Actually, to rephrase, it's not really a type check that I deleted. In the old implementation, the queue was only sometimes circular. It would eventually "unravel" to linear so that the old nodes could be garbage collected. But in the new implementation, the pending queue is always circular. So you don't need to check for if it's circular or not.
sebmarkbage
reviewed
Dec 9, 2019
| @@ -160,7 +160,7 @@ describe('createSubscription', () => { | |||
|
|
|||
| it('should still work if unsubscription is managed incorrectly', async () => { | |||
| const Subscription = createSubscription({ | |||
| getCurrentValue: source => undefined, | |||
| getCurrentValue: source => source.value, | |||
There was a problem hiding this comment.
Is this a breaking change?
There was a problem hiding this comment.
I looked into this for a while and I think the old version was just wrong and it accidentally worked with the old implementation. I'm going to try to write a failing test for what's in master.
There was a problem hiding this comment.
Never mind, I found the bug. Will add a test.
Adds a failing test case where an update that was committed is later skipped over during a rebase. This should never happen.
sebmarkbage
reviewed
Dec 9, 2019
| lastCapturedEffect: Update<State> | null, | ||
| baseQueue: Update<State> | null, | ||
| shared: SharedQueue<State>, | ||
| effects: Array<Update<State>> | null, |
There was a problem hiding this comment.
Why is this an array as opposed to a linked list like the hooks form? Harder to see how they correlate.
I was considering a refactor to arrays but with a different take (inline fields instead of objects) . However, I wanted to do that as a separate refactor.
There was a problem hiding this comment.
Idk I just chose one :D
This felt like a different trade off from the hooks one because these effects are relatively rare. Only when you pass a callback to setState. It's more like the emitEffect API.
There was a problem hiding this comment.
So it didn't seem worth putting a nextEffect field on the update type
acdlite
force-pushed
the
queue-refactor
branch
2 times, most recently
from
57816d9
to
862287b
Compare
sebmarkbage
approved these changes
Dec 9, 2019
This test passes in the old legacy update queue implementation. I'm adding this before the refactor to prevent a regression.
Refactors legacy update queue to incoporate rebasing fix. Uses nearly the same approach as the hook update queue but has to handle a few other cases.
acdlite
force-pushed
the
queue-refactor
branch
from
862287b
to
99444a2
Compare
acdlite
added a commit
that referenced
this pull request
Dec 9, 2019
Fixes a bug related to rebasing updates. Once an update has committed, it should never un-commit, even if interrupted by a higher priority update. The fix includes a refactor of how update queues work. This commit is a combination of two PRs: - #17483 by @sebmarkbage refactors the hook update queue - #17510 by @acdlite refactors the class and root update queue Landing one without the other would cause state updates to sometimes be inconsistent across components, so I've combined them into a single commit in case they need to be reverted. Co-authored-by: Sebastian Markbåge <sema@fb.com> Co-authored-by: Andrew Clark <git@andrewclark.io>
|
I squashed this with the hooks refactor and landed as b617db3 so it can be atomically reverted. |
acdlite
added a commit
to acdlite/react
that referenced
this pull request
Dec 9, 2019
Fixes a bug related to rebasing updates. Once an update has committed, it should never un-commit, even if interrupted by a higher priority update. The fix includes a refactor of how update queues work. This commit is a combination of two PRs: - facebook#17483 by @sebmarkbage refactors the hook update queue - facebook#17510 by @acdlite refactors the class and root update queue Landing one without the other would cause state updates to sometimes be inconsistent across components, so I've combined them into a single commit in case they need to be reverted. Co-authored-by: Sebastian Markbåge <sema@fb.com> Co-authored-by: Andrew Clark <git@andrewclark.io>
acdlite
added a commit
that referenced
this pull request
Dec 11, 2019
* Refactor Update Queues to Fix Rebasing Bug Fixes a bug related to rebasing updates. Once an update has committed, it should never un-commit, even if interrupted by a higher priority update. The fix includes a refactor of how update queues work. This commit is a combination of two PRs: - #17483 by @sebmarkbage refactors the hook update queue - #17510 by @acdlite refactors the class and root update queue Landing one without the other would cause state updates to sometimes be inconsistent across components, so I've combined them into a single commit in case they need to be reverted. Co-authored-by: Sebastian Markbåge <sema@fb.com> Co-authored-by: Andrew Clark <git@andrewclark.io> * Initialize update queue object on mount Instead of lazily initializing update queue objects on the first update, class and host root queues are created on mount. This simplifies the logic for appending new updates and matches what we do for hooks.
Merged
This was referenced Mar 6, 2020
This was referenced Sep 25, 2024
This was referenced Sep 28, 2024
This was referenced Sep 29, 2024
This was referenced Oct 4, 2024
This was referenced Oct 5, 2024
This was referenced Oct 12, 2024
This was referenced Oct 25, 2024
This was referenced Oct 26, 2024
This was referenced Oct 27, 2024
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Based on #17483
Refactors legacy update queue to incorporate rebasing fix. Uses nearly the same approach as the hook update queue but has to handle a few other cases.
Managed to remove some older, poorly factored code related to error handling.
To-do