-
Notifications
You must be signed in to change notification settings - Fork 46.9k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
[React 19] useTransition()'s pending state does not go back to false (revision 94eed63c49-20240425) #28923
Comments
Does this reproduce in dev and prod? |
@eps1lon I made production build via |
Maybe not a good idea to pass an async value directly or either through a function into state as this behaviour is synchronous. this could result to state inconsistency and also concurrency issues. looking at the delayed function passed into state. |
Resolving the useTransition Hook Issue: Following Official Documentation In my opinion, the issue mentioned above is not an actual issue according to the official React documentation. I resolved the problem using the following method: Solution: I moved the useTransition hook to the top level of the component, which resolved the error. (https://react.dev/reference/react/useTransition) Reason: The official documentation states: "Call useTransition at the top level of your component to mark some state updates as Transitions." Conclusion: When using the useTransition hook, it is essential to declare it at the top level of the component. |
@nakjun12 Can you provide a repro or diff illustrating what you did? From what I can tell, the original repro is valid. |
@eps1lon I did try doing so in my repro code. In function SimpleControlledDisplay({
promise,
onChange,
}: {
promise: Promise<number>;
onChange: (value: Promise<number>) => void;
}) {
+ let [pending, startTransition] = useTransition();
let value = use(promise);
- let [pending, startTransition] = useTransition(); And the transition worked as expected, resetting |
@eps1lon |
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>
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)
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.
defining the |
Summary
I am excited to start using React v19 as it has so many features and QoL improvements I've been waiting for!
There is a bug (new bug comparing to v18.2.0) that I found while reproducing #26814. When using
useTransition()
withuse()
,pending
flag of transition correctly becomestrue
in the beginning, but doesn't go back tofalse
after transition is complete, which means any pending state artifacts in the UI remain visible.Repository for reproducing: https://github.com/alexeyraspopov/react-beta-bugs.
Basic code (since the repo contains more than just this bug):
Additionally here's the video reproduction, using the code from the repo I mentioned:
Screen.Recording.2024-04-26.at.08.27.42.mov
The text was updated successfully, but these errors were encountered: