Stacked on #26570
Previously we restricted Float methods to only being callable while
rendering. This allowed us to make associations between calls and their
position in the DOM tree, for instance hoisting preinitialized styles
into a ShadowRoot or an iframe Document.
When considering how we are going to support Flight support in Float
however it became clear that this restriction would lead to compromises
on the implementation because the Flight client does not execute within
the context of a client render. We want to be able to disaptch Float
directives coming from Flight as soon as possible and this requires
being able to call them outside of render.
this patch modifies Float so that its methods are callable anywhere. The
main consequence of this change is Float will always use the Document
the renderer script is running within as the HoistableRoot. This means
if you preinit as style inside a component render targeting a ShadowRoot
the style will load in the ownerDocument not the ShadowRoot. Practially
speaking it means that preinit is not useful inside ShadowRoots and
iframes.
This tradeoff was deemed acceptable because these methods are
optimistic, not critical. Additionally, the other methods, preconntect,
prefetchDNS, and preload, are not impacted because they already operated
at the level of the ownerDocument and really only interface with the
Network cache layer.
I added a couple additional fixes that were necessary for getting tests
to pass that are worth considering separately.
The first commit improves the diff for `waitForThrow` so it compares
strings if possible.
The second commit makes invokeGuardedCallback not use metaprogramming
pattern and swallows any novel errors produced from trying to run the
guarded callback. Swallowing may not be the best we can do but it at
least protects React against rapid failure when something causes the
dispatchEvent to throw.
In anticipation of making Fiber use the document global for dispatching
Float methods that arrive from Flight I needed to update some tests that
commonly recreated the JSDOM instance after importing react.
This change updates a few tests to only create JSDOM once per test,
before importing react-dom/client.
Additionally the current act implementation for server streaming did not
adequately model streaming semantics so I rewrite the act implementation
in a way that better mirrors how a browser would parse incoming HTML.
The new act implementation does the following
1. the first time it processes meaningful streamed content it figures
out whether it is rendering into the existing document container or if
it needs to reset the document. this is based on whether the streamed
content contains tags `<html>` or `<body>` etc...
2. Once the streaming container is set it will typically continue to
stream into that container for future calls to act. The exception is if
the streaming container is the `<head>` in which case it will switch to
streaming into the body once it receives a `<body>` tag.
This means for tests that render something like a `<div>...</div>` it
will naturally stream into the default `<div id="container">...` and for
tests that render a full document the HTML will parse like a real
browser would (with some very minor edge case differences)
I also refactored the way we move nodes from buffered content into the
document and execute any scripts we find. Previously we were using
window.eval and I switched this to just setting the external script
content as script text. Additionally the nonce logic is reworked to be a
bit simpler.
This puts the change introduced by #26611 behind a flag until Meta is
able to roll it out. Disabling the flag reverts back to the old
behavior, where retries are throttled if there's still data remaining in
the tree, but not if all the data has finished loading.
The new behavior is still enabled in the public builds.
- substr is Annex B
- substring silently flips its arguments if they're in the "wrong order", which is confusing
- slice is better than sliced bread (no pun intended) and also it works the same way on Arrays so there's less to remember
---
> I'd be down to just lint and enforce a single form just for the potential compression savings by using a repeated string.
_Originally posted by @sebmarkbage in https://github.com/facebook/react/pull/26663#discussion_r1170455401_
This lets you pass a function to `<form action={...}>` or `<button
formAction={...}>` or `<input type="submit formAction={...}>`. This will
behave basically like a `javascript:` URL except not quite implemented
that way. This is a convenience for the `onSubmit={e => {
e.preventDefault(); const fromData = new FormData(e.target); ... }`
pattern.
You can still implement a custom `onSubmit` handler and if it calls
`preventDefault`, it won't invoke the action, just like it would if you
used a full page form navigation or javascript urls. It behaves just
like a navigation and we might implement it with the Navigation API in
the future.
Currently this is just a synchronous function but in a follow up this
will accept async functions, handle pending states and handle errors.
This is implemented by setting `javascript:` URLs, but these only exist
to trigger an error message if something goes wrong instead of
navigating away. Like if you called `stopPropagation` to prevent React
from handling it or if you called `form.submit()` instead of
`form.requestSubmit()` which by-passes the `submit` event. If CSP is
used to ban `javascript:` urls, those will trigger errors when these
URLs are invoked which would be a different error message but it's still
there to notify the user that something went wrong in the plumbing.
Next up is improving the SSR state with action replaying and progressive
enhancement.
Implements initial (client-only) support for async actions behind a
flag. This is an experimental feature and the design isn't completely
finalized but we're getting closer. It will be layered alongside other
features we're working on, so it may not feel complete when considered
in isolation.
The basic description is you can pass an async function to
`startTransition` and all the transition updates that are scheduled
inside that async function will be grouped together. The `isPending`
flag will be set to true immediately, and only set back to false once
the async action has completed (as well as all the updates that it
triggers).
The ideal behavior would be that all updates spawned by the async action
are automatically inferred and grouped together; however, doing this
properly requires the upcoming (stage 2) Async Context API, which is not
yet implemented by browsers. In the meantime, we will fake this by
grouping together all transition updates that occur until the async
function has terminated. This can lead to overgrouping between unrelated
actions, which is not wrong per se, just not ideal.
If the `useTransition` hook is removed from the UI before an async
action has completed — for example, if the user navigates to a new page
— subsequent transitions will no longer be grouped with together with
that action.
Another consequence of the lack of Async Context is that if you call
`setState` inside an action but after an `await`, it must be wrapped in
`startTransition` in order to be grouped properly. If we didn't require
this, then there would be no way to distinguish action updates from
urgent updates caused by user input, too. This is an unfortunate footgun
but we can likely detect the most common mistakes using a lint rule.
Once Async Context lands in browsers, we can start warning in dev if we
detect an update that hasn't been wrapped in `startTransition`. Then,
longer term, once the feature is ubiquitous, we can rely on it for real
and allow you to call `setState` without the additional wrapper.
Things that are _not_ yet implemented in this PR, but will be added as
follow ups:
- Support for non-hook form of `startTransition`
- Canceling the async action scope if the `useTransition` hook is
deleted from the UI
- Anything related to server actions
I accidentally made a behavior change in the refactor. It turns out that
when switching off `checked` to an uncontrolled component, we used to
revert to the concept of "initialChecked" which used to be stored on
state.
When there's a diff to this computed prop and the value of props.checked
is null, then we end up in a case where it sets `checked` to
`initialChecked`:
https://github.com/facebook/react/blob/5cbe6258bc436b1683080a6d978c27849f1d9a22/packages/react-dom-bindings/src/client/ReactDOMInput.js#L69
Since we never changed `initialChecked` and it's not relevant if
non-null `checked` changes value, the only way this "change" could
trigger was if we move from having `checked` to having null.
This wasn't really consistent with how `value` works, where we instead
leave the current value in place regardless. So this is a "bug fix" that
changes `checked` to be consistent with `value` and just leave the
current value in place. This case should already have a warning in it
regardless since it's going from controlled to uncontrolled.
Related to that, there was also another issue observed in
https://github.com/facebook/react/pull/26596#discussion_r1162295872 and
https://github.com/facebook/react/pull/26588
We need to atomically apply mutations on radio buttons. I fixed this by
setting the name to empty before doing mutations to value/checked/type
in updateInput, and then set the name to whatever it should be. Setting
the name is what ends up atomically applying the changes.
---------
Co-authored-by: Sophie Alpert <git@sophiebits.com>
## Summary
Removing `enableNamedHooksFeature`, `enableProfilerChangedHookIndices`,
`enableProfilerComponentTree` feature flags, they are the same for all
configurations.
Some minor changes, observed while working on 24.7.5 release:
- Updated numeration of text instructions
- `reactjs.org` -> `react.dev`
- Fixed using `npm view` command for node 16+, `publish-release` script
currently fails if used with node 16+
Builds on top of https://github.com/facebook/react/pull/26661
This lets you pass FormData objects through the Flight Reply
serialization. It does that by prefixing each entry with the ID of the
reference and then the decoding side creates a new FormData object
containing only those fields (without the prefix).
Ideally this should be more generic. E.g. you should be able to pass
Blobs, Streams and Typed Arrays by reference inside plain objects too.
You should also be able to send Blobs and FormData in the regular Flight
serialization too so that they can go both directions. They should be
symmetrical. We'll get around to adding more of those features in the
Flight protocol as we go.
---------
Co-authored-by: Sophie Alpert <git@sophiebits.com>
Since this is an observable behavior and is hard to think about, seems
good to have tests for this.
The expected value included in each test is the behavior that existed
prior to #26546.
In
https://github.com/facebook/react/pull/26573/commits/2019ddc75f448292ffa6429d7625514af192631b,
we changed to set .defaultValue before .value on updates. In some cases,
setting .defaultValue causes .value to change, and since we only set
.value if it has the wrong value, this resulted in us not assigning to
.value, which resulted in inputValueTracking not knowing the right
value. See new test added.
My fix here is to (a) move the value setting back up first and (b)
narrowing the fix in the aforementioned PR to newly remove the value
attribute only if it defaultValue was previously present in props.
The second half is necessary because for types where the value property
and attribute are indelibly linked (hidden checkbox radio submit image
reset button, i.e. spec modes default or default/on from
https://html.spec.whatwg.org/multipage/input.html#dom-input-value-default),
we can't remove the value attribute after setting .value, because that
will undo the assignment we just did! That is, not having (b) makes all
of those types fail to handle updating props.value.
This code is incredibly hard to think about but I think this is right
(or at least, as right as the old code was) because we set .value here
only if the nextProps.value != null, and we now remove defaultValue only
if lastProps.defaultValue != null. These can't happen at the same time
because we have long warned if value and defaultValue are simultaneously
specified, and also if a component switches between controlled and
uncontrolled.
Also, it fixes the test in https://github.com/facebook/react/pull/26626.
In the extension, currently we do the following:
1. check whether there's at least one React renderer on the page
2. if yes, load the backend to the page
3. initialize the backend
To support multiple versions of backends, we are changing it to:
1. check the versions of React renders on the page
2. load corresponding React DevTools backends that are shipped with the
extension; if they are not contained (usually prod builds of
prereleases), show a UI to allow users to load them from UI
3. initialize each of the backends
To enable this workflow, a backend will ignore React renderers that does
not match its version
This PR adds a new file "backendManager" in the extension for this
purpose.
------
I've tested it on Chrome, Edge and Firefox extensions
This lets the client bundle encode Server References without them first
being passed from an RSC payload. Like if you just import `"use server"`
from the client. A bundler could already emit these proxies to be called
on the client but the subtle difference is that those proxies couldn't
be passed back into the server by reference. They have to be registered
with React.
We don't currently implement importing `"use server"` from client
components in the reference implementation. It'd need to expand the
Webpack plugin with a loader that rewrites files with the `"use server"`
in the client bundle.
```
"use server";
export async function action() {
...
}
```
->
```
import {createServerReference} from "react-server-dom-webpack/client";
import {callServer} from "some-router/call-server";
export const action = createServerReference('1234#action', callServer);
```
The technique I use here is that the compiled output has to call
`createServerReference(id, callServer)` with the `$$id` and proxy
implementation. We then return a proxy function that is registered with
a WeakMap to the particular instance of the Flight Client.
This might be hard to implement because it requires emitting module
imports to a specific stateful runtime module in the compiler. A benefit
is that this ensures that this particular reference is locked to a
specific client if there are multiple - e.g. talking to different
servers.
It's fairly arbitrary whether we use a WeakMap technique (like we do on
the client) vs an `$$id` (like we do on the server). Not sure what's
best overall. The WeakMap is nice because it doesn't leak implementation
details that might be abused to consumers. We should probably pick one
and unify.
We currently don't just "require" a module by its module id/path. We
encode the pair of module id/path AND its export name. That's because
with module splitting, a single original module can end up in two or
more separate modules by name. Therefore the manifest files need to
encode how to require the whole module as well as how to require each
export name.
In practice, we don't currently use this because we end up forcing
Webpack to deopt and keep it together as a single module, and we don't
even have the code in the Webpack plugin to write separate values for
each export name.
The problem is with CJS we don't statically know what all the export
names will be. Since these cases will never be module split, we don't
really need to know.
This changes the Flight requires to first look for the specific name
we're loading and then if that name doesn't exist in the manifest we
fallback to looking for the `"*"` name containing the entire module and
look for the name in there at runtime.
We could probably optimize this a bit if we assume that CJS modules on
the server never get built with a name. That way we don't have to do the
failed lookup.
Additionally, since we've recently merged filepath + name into a single
string instead of two values, we now have to split those back out by
parsing the string. This is especially unfortunate for server references
since those should really not reveal what they are but be a hash or
something. The solution might just be to split them back out into two
separate fields again.
cc @shuding
<!--
Thanks for submitting a pull request!
We appreciate you spending the time to work on these changes. Please
provide enough information so that others can review your pull request.
The three fields below are mandatory.
Before submitting a pull request, please make sure the following is
done:
1. Fork [the repository](https://github.com/facebook/react) and create
your branch from `main`.
2. Run `yarn` in the repository root.
3. If you've fixed a bug or added code that should be tested, add tests!
4. Ensure the test suite passes (`yarn test`). Tip: `yarn test --watch
TestName` is helpful in development.
5. Run `yarn test --prod` to test in the production environment. It
supports the same options as `yarn test`.
6. If you need a debugger, run `yarn debug-test --watch TestName`, open
`chrome://inspect`, and press "Inspect".
7. Format your code with
[prettier](https://github.com/prettier/prettier) (`yarn prettier`).
8. Make sure your code lints (`yarn lint`). Tip: `yarn linc` to only
check changed files.
9. Run the [Flow](https://flowtype.org/) type checks (`yarn flow`).
10. If you haven't already, complete the CLA.
Learn more about contributing:
https://reactjs.org/docs/how-to-contribute.html
-->
## Summary
Browsers restore state like forms and scroll position right after the
popstate event. To make sure the page work as expected on back or
forward button, we need to flush transitions scheduled in a popstate
synchronously, and only yields if it suspends.
This PR adds a new HostConfig method to check if `window.event ===
'popstate'`, and `scheduleMicrotask` if a transition is scheduled in a
`PopStateEvent`.
## How did you test this change?
yarn test
If a Suspense fallback is shown, and the data finishes loading really
quickly after that, we throttle the content from appearing for 500ms to
reduce thrash.
This already works for successive fallback states (like if one fallback
is nested inside another) but it wasn't being applied to the final step
in the sequence: if there were no more unresolved Suspense boundaries in
the tree, the content would appear immediately.
This fixes the throttling behavior so that it applies to all renders
that are the result of suspended data being loaded. (Our internal jargon
term for this is a "retry".)
Many of our Suspense-related tests were written before the `act` API was
introduced, and use the lower level `waitFor` helpers instead. So they
are less resilient to changes in implementation details than they could
be.
This converts some of our test suite to use `act` in more places. I
found these while working on a PR to expand our fallback throttling
mechanism to include all renders that result from a promise resolving,
even if there are no more fallbacks in the tree.
I think this covers all the remaining tests that are affected.
Fixes https://github.com/facebook/react/issues/26500
## Summary
- No more using `clipboard-js` from the backend side, now emitting
custom `saveToClipboard` event, also adding corresponding listener in
`store.js`
- Not migrating to `navigator.clipboard` api yet, there were some issues
with using it on Chrome, will add more details to
https://github.com/facebook/react/pull/26539
## How did you test this change?
- Tested on Chrome, Firefox, Edge
- Tested on standalone electron app: seems like context menu is not
expected to work there (cannot right-click on value, the menu is not
appearing), other logic (pressing on copy icon) was not changed
<!--
Thanks for submitting a pull request!
We appreciate you spending the time to work on these changes. Please
provide enough information so that others can review your pull request.
The three fields below are mandatory.
Before submitting a pull request, please make sure the following is
done:
1. Fork [the repository](https://github.com/facebook/react) and create
your branch from `main`.
2. Run `yarn` in the repository root.
3. If you've fixed a bug or added code that should be tested, add tests!
4. Ensure the test suite passes (`yarn test`). Tip: `yarn test --watch
TestName` is helpful in development.
5. Run `yarn test --prod` to test in the production environment. It
supports the same options as `yarn test`.
6. If you need a debugger, run `yarn test --debug --watch TestName`,
open `chrome://inspect`, and press "Inspect".
7. Format your code with
[prettier](https://github.com/prettier/prettier) (`yarn prettier`).
8. Make sure your code lints (`yarn lint`). Tip: `yarn linc` to only
check changed files.
9. Run the [Flow](https://flowtype.org/) type checks (`yarn flow`).
10. If you haven't already, complete the CLA.
Learn more about contributing:
https://reactjs.org/docs/how-to-contribute.html
-->
## Summary
Addresses #26352.
This PR explicitly passes an icon to `chrome.devtools.panels.create()`,
so that edge devtools will display the icon when in [Focus
Mode](https://learn.microsoft.com/en-us/microsoft-edge/devtools-guide-chromium/experimental-features/focus-mode).
<!--
Explain the **motivation** for making this change. What existing problem
does the pull request solve?
-->
## How did you test this change?
Passing test suite (`yarn test` & `yarn test --prod`) ✅
Passing lint (`yarn linc`) ✅
Passing type checks (`yarn flow`) ✅
**Visual Testing**
Before Changes | After Changes
:-------------------------:|:-------------------------:
|
<!--
Demonstrate the code is solid. Example: The exact commands you ran and
their output, screenshots / videos if the pull request changes the user
interface.
How exactly did you verify that your PR solves the issue you wanted to
solve?
If you leave this empty, your PR will very likely be closed.
-->
Many of our Suspense-related tests were written before the `act` API was
introduced, and use the lower level `waitFor` helpers instead. So they
are less resilient to changes in implementation details than they could
be.
This converts some of our test suite to use `act` in more places. I
found these while working on a PR to expand our fallback throttling
mechanism to include all renders that result from a promise resolving,
even if there are no more fallbacks in the tree. This isn't all the
affected tests, just some of them — I'll be sharding the changes across
multiple PRs.
Since `props.x` is a possibly megamorphic access, it can be slow to
access and trigger recompilation.
When we are looping over the props and pattern matching every key,
anyway, we've already done this work. We can just reuse the same value
by stashing it outside the loop in the stack.
This only makes sense for updates in diffInCommitPhase since otherwise
we don't have the full set of props in that loop.
We also have to be careful not to skip over equal values since we need
to extract them anyway.
This traces back to https://github.com/facebook/react/pull/6449 and then
another before that.
I think that back then we favored the property over the attribute, and
setting the property wouldn't be enough. However, the default path for
these are now using attributes if we don't special case it. So we don't
need it.
The only difference is that we currently have a divergence for
symbol/function behavior between controlled values that use the
getToStringValue helpers which treat them as empty string, where as
everywhere else they're treated as null/missing.
Since this comes with a warning and is a weird error case, it's probably
fine to change.
Updates that are marked as part of a transition are allowed to block a
render from committing. Generally, other updates cannot — however,
there's one exception that's leftover from a previous iteration of our
Suspense architecture. If an update is not the result of a known urgent
event type — known as "Default" updates — then we allow it to suspend
briefly, as long as the delay is short enough that the user won't
notice. We refer to this delay as a "Just Noticable Difference" (JND)
delay. To illustrate, if the user has already waited 400ms for an update
to be reflected on the screen, the theory is that they won't notice if
you wait an additional 100ms. So React can suspend for a bit longer in
case more data comes in. The longer the user has already waited, the
longer the JND.
While we still believe this theory is sound from a UX perspective, we no
longer think the implementation complexity is worth it. The main thing
that's changed is how we handle Default updates. We used to render
Default updates concurrently (i.e. they were time sliced, and were
scheduled with postTask), but now they are blocking. Soon, they will
also be scheduled with rAF, too, which means by the end of the next rAF,
they will have either finished rendering or the main thread will be
blocked until they do. There are various motivations for this but part
of the rationale is that anything that can be made non-blocking should
be marked as a Transition, anyway, so it's not worth adding
implementation complexity to Default.
This commit removes the JND delay for Default updates. They will now
commit immediately once the render phase is complete, even if a
component suspends.
This removes the concept of `prepareUpdate()`, behind a flag.
React Native already does everything in the commit phase, but generates
a temporary update payload before applying it.
React Fabric does it both in the render phase. Now it just moves it to a
single host config.
For DOM I forked updateProperties into one that does diffing and
updating in one pass vs just applying a pre-diffed updatePayload.
There are a few downsides of this approach:
- If only "children" has changed, we end up scheduling an update to be
done in the commit phase. Since we traverse through it anyway, it's
probably not much extra.
- It does more work in the commit phase so for a large tree that is
mostly unchanged, it'll stall longer.
- It does some extra work for special cases since that work happens if
anything has changed. We no longer have a deep bailout.
- The special cases now have to each replicate the "clean up old props"
loop, leading to extra code.
The benefit is that this doesn't allocate temporary extra objects
(possibly multiple per element if the array has to resize). It's less
work overall. It also gives us an option to reuse this function for a
sync render optimization.
Another benefit is that if we do the loop in the commit phase I can do
further optimizations by reading all props that I need for special cases
in that loop instead of polymorphic reads from props. This is what I'd
like to do in future refactors that would be stacked on top of this
change.
We have moved away from HostConfig since the name does not fully
describe the configs we customize per runtime like FlightClient,
FlightServer, Fizz, and Fiber. This commit generalizes $$$hostconfig to
$$$config
part of https://github.com/facebook/react/pull/26571
merging separately to improve tracking of files renames in git
Rename HostConfig files to FiberConfig to clarify they are configs for
Fiber and not Fizz/Flight. This better conforms to the naming used in
Flight and now Fizz of `ReactFlightServerConfig` and `ReactFizzConfig`
Part of https://github.com/facebook/react/pull/26571
Implements wiring for Flight to have it's own "HostConfig" from Fizz.
Historically the ServerFormatConfigs were supposed to be generic enough
to be used by Fizz and Flight. However with the addition of features
like Float the configs have evolved to be more specific to the renderer.
We may want to get back to a place where there is a pure FormatConfig
which can be shared but for now we are embracing the fact that these
runtimes need very different things and DCE cannot adequately remove the
unused stuff for Fizz when pulling this dep into Flight so we are going
to fork the configs and just maintain separate ones.
At first the Flight config will be almost empty but once Float support
in Flight lands it will have a more complex implementation
Additionally this commit normalizes the component files which make up
FlightServerConfig and FlightClientConfig. Now each file that
participates starts with ReactFlightServerConfig... and
ReactFlightClientConfig...
## Summary
This pull request aims to improve the maintainability of the codebase by
consolidating types and constants that are shared between the backend
and frontend. This consolidation will allow us to maintain backwards
compatibility in the frontend in the future.
To achieve this, we have moved the shared types and constants to the
following blessed files:
- react-devtools-shared/src/constants
- react-devtools-shared/src/types
- react-devtools-shared/src/backend/types
- react-devtools-shared/src/backend/NativeStyleEditor/types
Please note that the inclusion of NativeStyleEditor in this list is
temporary, and we plan to remove it once we have a better plugin system
in place.
## How did you test this change?
I have tested it by running `yarn flow dom-node`, which reports no
errors.
`act` uses the `didScheduleLegacyUpdate` field to simulate the behavior
of batching in React <17 and below. It's a quirk leftover from a
previous implementation, not intentionally designed.
This sets `didScheduleLegacyUpdate` every time a legacy root receives an
update as opposed to only when the `executionContext` is empty. There's
no real reason to do it this way over some other way except that it's
how it used to work before #26512 and we should try our best to maintain
the existing behavior, quirks and all, since existing tests may have
come to accidentally rely on it.
This should fix some (though not all) of the internal Meta tests that
started failing after #26512 landed.
Will add a regression test before merging.
In #26573 I changed it so that textareas get their defaultValue reset if
you don't specify one.
However, the way that was implemented, it always set it for any update
even if it hasn't changed.
We have a test for that, but that test only works if no properties
update at all so that no update was scheduled. This fixes the test so
that it updates some unrelated prop.
I also found a test for `checked` that needed a similar fix.
Interestingly, we don't do this deduping for `defaultValue` or
`defaultChecked` on inputs and there's no test for that.
This is mainly renaming some stuff. The behavior change is
hasOwnProperty to nullish check.
I had a bigger refactor that was a dead-end but might as well land this
part and see if I can pick it up later.
Currently, `waitForThrow` tries to diff the expected value against the
thrown value if it doesn't match. However if the expectation is a
string, we are not diffing against the thrown message. This commit makes
it so if we are matching against message we also diff against message.
## Summary
- #26234 is reverted and replaced with a better approach
- introduce a new global devtools variable to decouple the global hook's
dependency on backend/console.js, and add it to react-devtools-inline
and react-devtools-standalone
With this PR, I want to introduce a new principle to hook.js: we should
always be alert when editing this file and avoid importing from other
files.
In the past, we try to inline a lot of the implementation because we use
`.toString()` to inject this function from the extension (we still have
some old comments left). Although it is no longer inlined that way, it
has became now more important to keep it clean as it is a de facto
global API people are using (9.9K files contains it on Github search as
of today).
**File size change for extension:**
Before:
379K installHook.js
After:
21K installHook.js
363K renderer.js
This fixes the "double free" bug illustrated by the regression test
added in the previous commit.
The underlying issue is that `effect.destroy` field is a mutable field
but we read it during render. This is a concurrency bug — if we had a
borrow checker, it would not allow this.
It's rare in practice today because the field is updated during the
commit phase, which takes a lock on the fiber tree until all the effects
have fired. But it's still theoretically wrong because you can have
multiple Fiber copies each with their own reference to a single destroy
function, and indeed we discovered in production a scenario where this
happens via our current APIs.
In the future these types of scenarios will be much more common because
we will introduce features where effects may run concurrently with the
render phase — i.e. an imperative `hide` method that synchronously hides
a React tree and unmounts all its effects without entering the render
phase, and without interrupting a render phase that's already in
progress.
A future version of React may also be able to run the entire commit
phase concurrently with a subsequent render phase. We can't do this now
because our data structures are not fully thread safe (see: the Fiber
alternate model) but we should be able to do this in the future.
The fix I've introduced in this commit is to move the `destroy` field to
a separate object. The effect "instance" is a shared object that remains
the same for the entire lifetime of an effect. In Rust terms, a RefCell.
The field is `undefined` if the effect is unmounted, or if the effect
ran but is not stateful. We don't explicitly track whether the effect is
mounted or unmounted because that can be inferred by the hiddenness of
the fiber in the tree, i.e. whether there is a hidden Offscreen fiber
above it.
It's unfortunate that this is stored on a separate object, because it
adds more memory per effect instance, but it's conceptually sound. I
think there's likely a better data structure we could use for effects;
perhaps just one array of effect instances per fiber. But I think this
is OK for now despite the additional memory and we can follow up with
performance optimizations later.
---------
Co-authored-by: Dan Abramov <dan.abramov@gmail.com>
Co-authored-by: Rick Hanlon <rickhanlonii@gmail.com>
Co-authored-by: Jan Kassens <jan@kassens.net>
## Summary
The electron package was recently upgraded from ^11.1.0 to ^23.1.2
(#26337). However, the WebContents `new-window` event – that is used in
the react-devtools project – was deprecated in
[v12.0.0](https://releases.electronjs.org/release/v12.0.0) and removed
in [v22.2.0](https://releases.electronjs.org/release/v22.2.0). The event
was replaced by `webContents.setWindowOpenHandler()`. This PR replaces
the `new-window` event with `webContents.setWindowOpenHandler()`.
## How did you test this change?
I created a simple electron application with similar functionality:
```
const { app, BrowserWindow, shell } = require('electron')
const createWindow = () => {
const mainWindow = new BrowserWindow({
width: 800,
height: 600
})
mainWindow.webContents.setWindowOpenHandler(({ url }) => {
shell.openExternal(url)
return { action: 'deny' }
})
mainWindow.loadFile('index.html')
}
app.whenReady().then(() => {
createWindow()
})
```
---------
Co-authored-by: root <root@DESKTOP-KCGHLB8.localdomain>
Josh Story
Andrew Clark
Ruslan Lesiutin
Sophie Alpert
Sebastian Markbåge
Jan Kassens
Mengdi Chen
Rubén Norte
Tianyu Yao
Samuel Susla
Harry Zumwalt
Willie-Boy
