Fix: useTransition after use gets stuck in pending state
#29670
Merged
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
|
The latest updates on your projects. Learn more about Vercel for Git ↗︎
|
facebook-github-bot
added
CLA Signed
React Core Team
|
Comparing: ec6fe57...09f46e4 Critical size changesIncludes critical production bundles, as well as any change greater than 2%:
Significant size changesIncludes any change greater than 0.2%: Expand to show
|
eps1lon
approved these changes
May 31, 2024
Comment on lines
1922
to
1924
| // This is a regression test. The root cause was an issue where we failed to | ||
| // switch from the "re-render" dispatcher back to the "update" dispatcher | ||
| // after a `use` suspends and triggers a replay. |
There was a problem hiding this comment.
Now it also makes sense why it didn't get stuck if we called useTransition before use. Thanks for the explainer.
| // after a `use` suspends and triggers a replay. | ||
| let update; | ||
| function App({promise}) { | ||
| useState(false); |
There was a problem hiding this comment.
Is this testing that everything before use is still processed appropriately? May be worth a comment because right now it looks like we could remove it.
There was a problem hiding this comment.
I was testing both and forgot to delete, I'll add another test for this one
Introduces a regression test for a bug where the pending state of a useTransition hook is not set back to `false` if it comes after a `use` that suspended. Co-authored-by: Andrew Clark <git@andrewclark.io>
When a component suspends with `use`, we switch to the "re-render" dispatcher during the subsequent render attempt, so that we can reuse the work from the initial attempt. However, once we run out of hooks from the previous attempt, we should switch back to the regular "update" dispatcher. This is conceptually the same fix as the one introduced in facebook#26232. That fix only accounted for initial mount, but the useTransition regression test added in the previous commit illustrates that we need to handle updates, too. The issue affects more than just useTransition but because most of the behavior between the "re-render" and "update" dispatchers is the same it's hard to contrive other scenarios in a test, which is probably why it took so long for someone to notice.
acdlite
force-pushed
the
fix-stuck-ispending
branch
from
30bfb86
to
09f46e4
Compare
github-actions bot
pushed a commit
that referenced
this pull request
May 31, 2024
When a component suspends with `use`, we switch to the "re-render" dispatcher during the subsequent render attempt, so that we can reuse the work from the initial attempt. However, once we run out of hooks from the previous attempt, we should switch back to the regular "update" dispatcher. This is conceptually the same fix as the one introduced in #26232. That fix only accounted for initial mount, but the useTransition regression test added in f829733 illustrates that we need to handle updates, too. The issue affects more than just useTransition but because most of the behavior between the "re-render" and "update" dispatchers is the same it's hard to contrive other scenarios in a test, which is probably why it took so long for someone to notice. Closes #28923 and #29209 --------- Co-authored-by: eps1lon <sebastian.silbermann@vercel.com> DiffTrain build for [adbec0c](adbec0c)
github-actions bot
pushed a commit
that referenced
this pull request
May 31, 2024
When a component suspends with `use`, we switch to the "re-render" dispatcher during the subsequent render attempt, so that we can reuse the work from the initial attempt. However, once we run out of hooks from the previous attempt, we should switch back to the regular "update" dispatcher. This is conceptually the same fix as the one introduced in #26232. That fix only accounted for initial mount, but the useTransition regression test added in f829733 illustrates that we need to handle updates, too. The issue affects more than just useTransition but because most of the behavior between the "re-render" and "update" dispatchers is the same it's hard to contrive other scenarios in a test, which is probably why it took so long for someone to notice. Closes #28923 and #29209 --------- Co-authored-by: eps1lon <sebastian.silbermann@vercel.com> DiffTrain build for commit adbec0c.
This was referenced Jun 2, 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.
When a component suspends with
use, we switch to the "re-render" dispatcher during the subsequent render attempt, so that we can reuse the work from the initial attempt. However, once we run out of hooks from the previous attempt, we should switch back to the regular "update" dispatcher.This is conceptually the same fix as the one introduced in #26232. That fix only accounted for initial mount, but the useTransition regression test added in f829733 illustrates that we need to handle updates, too.
The issue affects more than just useTransition but because most of the behavior between the "re-render" and "update" dispatchers is the same it's hard to contrive other scenarios in a test, which is probably why it took so long for someone to notice.
Closes #28923 and #29209