## Summary
This PR bumps Flow all the way to the latest 0.245.2.
Most of the suppressions comes from Flow v0.239.0's change to include
undefined in the return of `Array.pop`.
I also enabled `react.custom_jsx_typing=true` and added custom jsx
typing to match the old behavior that `React.createElement` is
effectively any typed. This is necessary since various builtin
components like `React.Fragment` is actually symbol in the React repo
instead of `React.AbstractComponent<...>`. It can be made more accurate
by customizing the `React$CustomJSXFactory` type, but I will leave it to
the React team to decide.
## How did you test this change?
`yarn flow` for all the renderers
In https://github.com/facebook/react/pull/29491 I updated the work
scheduler for Flight to use microtasks to perform work when something
pings. This is useful but it does have some downsides in terms of our
ability to do task prioritization. Additionally the initial work is not
instantiated using a microtask which is inconsistent with how pings
work.
In this change I update the scheduling logic to use microtasks
consistently for prerenders and use regular tasks for renders both for
the initial work and pings.
React transpiles some of its own `console.error` calls into a helper
that appends component stacks to those calls. However, this doesn't
cover user space `console.error` calls - which includes React helpers
that React has moved into third parties like createClass and prop-types.
The idea is that any user space component can add a warning just like
React can which is why React DevTools adds them too if they don't
already exist. Having them appended in both places is tricky because now
you have to know whether to remove them from React's logs.
Similarly it's often common for server-side frameworks to forget to
cover the `console.error` logs from other sources since React DevTools
isn't active there. However, it's also annoying to get component stacks
clogging the terminal - depending on where the log came from.
In the future `console.createTask()` will cover this use case natively
and when available we don't append them at all.
The new strategy relies on either:
- React DevTools existing to add them to React logs as well as third
parties.
- `console.createTask` being supported and surfaced.
- A third party framework showing the component stack either in an Error
Dialog or appended to terminal output.
For a third party to be able to implement this they need to be able to
get the component stack. To get the component stack from within a
`console.error` call you need to use the `React.captureOwnerStack()`
helper which is only available in `enableOwnerStacks` flag. However,
it's possible to polyfill with parent stacks using internals as a stop
gap. There's a question of whether React 19 should just go out with
`enableOwnerStacks` to expose this but regardless I think it's best it
doesn't include component stacks from the runtime for consistency.
In practice it's not really a regression though because typically either
of the other options exists and error dialogs don't implement
`console.error` overrides anyway yet. SSR terminals might miss them but
they'd only have them in DEV warnings to begin with an a subset of React
warnings. Typically those are either going to happen on the client
anyway or replayed.
Our tests are written to assert that component stacks work in various
scenarios all over the place. To ensure that this keeps working I
implement a "polyfill" that is similar to that expected a server
framework might do - in `assertConsoleErrorDev` and `toErrorDev`.
This PR doesn't yet change www or RN since they have their own forks of
consoleWithStackDev for now.
Use the same normalizeCodeLocInfo that we use everywhere else.
We should actually test the component stack itself. Not just that it
exists. This was causing false passes.
However, the logic was also wrong before because it wouldn't always
strip out the last line so wouldn't accurately normalize it. Leading to
false failures as well.
When we replay logs we badge them with e.g. `[Server]`. That way it's
easy to identify that the source of the log actually happened on the
Server (RSC). However, when we threw an error we didn't have any such
thing. The error was rethrown on the client and then handled just like
any other client error.
This transfers the `environmentName` in DEV to our restored Error
"sub-class" (conceptually) along with `digest`. That way you can read
`error.environmentName` to print this in your own UI.
I also updated our default for `onCaughtError` (and `onError` in Fizz)
to use the `printToConsole` helper that the Flight Client uses to log it
with the badge format. So by default you get the same experience as
console.error for caught errors:
<img width="810" alt="Screenshot 2024-06-10 at 9 25 12 PM"
src="https://github.com/facebook/react/assets/63648/8490fedc-09f6-4286-9332-fbe6b0faa2d3">
<img width="815" alt="Screenshot 2024-06-10 at 9 39 30 PM"
src="https://github.com/facebook/react/assets/63648/bdcfc554-504a-4b1d-82bf-b717e74975ac">
Unfortunately I can't do the same thing for `onUncaughtError` nor
`onRecoverableError` because they use `reportError` which doesn't have
custom formatting (unless we also prevented default on window.onerror).
However maybe that's ok because 1) you should always have an error
boundary 2) it's not likely that an RSC error can actually recover
because it's not going to be rendered again so shouldn't really happen
outside some parent conditionally rendering maybe.
The other problem with this approach is that the default is no longer
trivial - so reimplementing the default in user space is trickier and
ideally we shouldn't expose our default to be called.
Basically make `console.error` and `console.warn` behave like normal -
when a component stack isn't appended. I need this because I need to be
able to print rich logs with the component stack option and to be able
to disable instrumentation completely in `console.createTask`
environments that don't need it.
Currently we can't print logs with richer objects because they're
toString:ed first. In practice, pretty much all arguments we log are
already toString:ed so it's not necessary anyway. Some might be like a
number. So it would only be a problem if some environment can't handle
proper consoles but then it's up to that environment to toString it
before logging.
The `Warning: ` prefix is historic and is both noisy and confusing. It's
mostly unnecessary since the UI surrounding `console.error` and
`console.warn` tend to have visual treatment around it anyway. However,
it's actively misleading when `console.error` gets prefixed with a
Warning that we consider an error level. There's an argument to be made
that some of our `console.error` don't make the bar for an error but
then the argument is to downgrade each of those to `console.warn` - not
to brand all our actual error logging with `Warning: `.
Apparently something needs to change in React Native before landing this
because it depends on the prefix somehow which probably doesn't make
sense already.
## Overview
Updates `eslint-plugin-jest` and enables the recommended rules with some
turned off that are unhelpful.
The main motivations is:
a) we have a few duplicated tests, which this found an I deleted
b) making sure we don't accidentally commit skipped tests
## Summary
```js
assertConsoleErrorDev([
['Hello', {withoutStack: true}]
])
```
now errors with a helpful diff message if the message mismatched. See
first commit for previous behavior.
## How did you test this change?
- `yarn test --watch
packages/internal-test-utils/__tests__/ReactInternalTestUtils-test.js`
Fixes issue where if the first letter of the expected string appeared
anywhere in actual message, the assertion would pass, leading to false
negatives. We should check the entire expected string.
---------
Co-authored-by: Ricky <rickhanlonii@gmail.com>
## Overview
**Internal React repo tests only**
Depends on https://github.com/facebook/react/pull/28710
Adds three new assertions:
- `assertConsoleLogDev`
- `assertConsoleWarnDev`
- `assertConsoleErrorDev`
These will replace this pattern:
```js
await expect(async () => {
await expect(async () => {
await act(() => {
root.render(<Fail />)
});
}).toThrow();
}).toWarnDev('Warning');
```
With this:
```js
await expect(async () => {
await act(() => {
root.render(<Fail />)
});
}).toThrow();
assertConsoleWarnDev('Warning');
```
It works similar to our other `assertLog` matchers which clear the log
and assert on it, failing the tests if the log is not asserted before
the test ends.
## Diffs
There are a few improvements I also added including better log diffs and
more logging.
When there's a failure, the output will look something like:
<img width="655" alt="Screenshot 2024-04-03 at 11 50 08 AM"
src="https://github.com/facebook/react/assets/2440089/0c4bf1b2-5f63-4204-8af3-09e0c2d752ad">
Check out the test suite for snapshots of all the failures we may log.
Moving this to `internal-test-utils` so I can add helpers in the next PR
for:
- assertLogDev
- assertWarnDev
- assertErrorDev
Which will be exported from `internal-test-utils`. This isn't strictly
necessary, but it makes the factoring nicer, so internal-test-until
doesn't need to depend on `scripts/jest`.
Stacked on top of #28498 for test fixes.
### Don't Rethrow
When we started React it was 1:1 setState calls a series of renders and
if they error, it errors where the setState was called. Simple. However,
then batching came and the error actually got thrown somewhere else.
With concurrent mode, it's not even possible to get setState itself to
throw anymore.
In fact, all APIs that can rethrow out of React are executed either at
the root of the scheduler or inside a DOM event handler.
If you throw inside a React.startTransition callback that's sync, then
that will bubble out of the startTransition but if you throw inside an
async callback or a useTransition we now need to handle it at the hook
site. So in 19 we need to make all React.startTransition swallow the
error (and report them to reportError).
The only one remaining that can throw is flushSync but it doesn't really
make sense for it to throw at the callsite neither because batching.
Just because something rendered in this flush doesn't mean it was
rendered due to what was just scheduled and doesn't mean that it should
abort any of the remaining code afterwards. setState is fire and forget.
It's send an instruction elsewhere, it's not part of the current
imperative code.
Error boundaries never rethrow. Since you should really always have
error boundaries, most of the time, it wouldn't rethrow anyway.
Rethrowing also actually currently drops errors on the floor since we
can only rethrow the first error, so to avoid that we'd need to call
reportError anyway. This happens in RN events.
The other issue with rethrowing is that it logs an extra console.error.
Since we're not sure that user code will actually log it anywhere we
still log it too just like we do with errors inside error boundaries
which leads all of these to log twice.
The goal of this PR is to never rethrow out of React instead, errors
outside of error boundaries get logged to reportError. Event system
errors too.
### Breaking Changes
The main thing this affects is testing where you want to inspect the
errors thrown. To make it easier to port, if you're inside `act` we
track the error into act in an aggregate error and then rethrow it at
the root of `act`. Unlike before though, if you flush synchronously
inside of act it'll still continue until the end of act before
rethrowing.
I expect most user code breakages would be to migrate from `flushSync`
to `act` if you assert on throwing.
However, in the React repo we also have `internalAct` and the
`waitForThrow` helpers. Since these have to use public production
implementations we track these using the global onerror or process
uncaughtException. Unlike regular act, includes both event handler
errors and onRecoverableError by default too. Not just render/commit
errors. So I had to account for that in our tests.
We restore logging an extra log for uncaught errors after the main log
with the component stack in it. We use `console.warn`. This is not yet
ignorable if you preventDefault to the main error event. To avoid
confusion if you don't end up logging the error to console I just added
`An error occurred`.
### Polyfill
All browsers we support really supports `reportError` but not all test
and server environments do, so I implemented a polyfill for browser and
node in `shared/reportGlobalError`. I don't love that this is included
in all builds and gets duplicated into isomorphic even though it's not
actually needed in production. Maybe in the future we can require a
polyfill for this.
### Follow Ups
In a follow up, I'll make caught vs uncaught error handling be
configurable too.
---------
Co-authored-by: Ricky Hanlon <rickhanlonii@gmail.com>
## Overview
For events, the browser will yield to microtasks between calling event
handers, allowing time to flush work inbetween. For example, in the
browser, this code will log the flushes between events:
```js
<body onclick="console.log('body'); Promise.resolve().then(() => console.log('flush body'));">
<div onclick="console.log('div'); Promise.resolve().then(() => console.log('flush div'));">
hi
</div>
</body>
// Logs
div
flush div
body
flush body
```
[Sandbox](https://codesandbox.io/s/eloquent-noether-mw2cjg?file=/index.html)
The problem is, `dispatchEvent` (either in the browser, or JSDOM) does
not yield to microtasks. Which means, this code will log the flushes
after the events:
```js
const target = document.getElementsByTagName("div")[0];
const nativeEvent = document.createEvent("Event");
nativeEvent.initEvent("click", true, true);
target.dispatchEvent(nativeEvent);
// Logs
div
body
flush div
flush body
```
## The problem
This mostly isn't a problem because React attaches event handler at the
root, and calls the event handlers on components via the synthetic event
system. We handle flushing between calling event handlers as needed.
However, if you're mixing capture and bubbling events, or using multiple
roots, then the problem of not flushing microtasks between events can
come into play. This was found when converting a test to `createRoot` in
https://github.com/facebook/react/pull/28050#discussion_r1462118422, and
that test is an example of where this is an issue with nested roots.
Here's a sandox for
[discrete](https://codesandbox.io/p/sandbox/red-http-2wg8k5) and
[continuous](https://codesandbox.io/p/sandbox/gracious-voice-6r7tsc?file=%2Fsrc%2Findex.js%3A25%2C28)
events, showing how the test should behave. The existing test, when
switched to `createRoot` matches the browser behavior for continuous
events, but not discrete. Continuous events should be batched, and
discrete should flush individually.
## The fix
This PR implements the fix suggested by @sebmarkbage, to manually
traverse the path up from the element and dispatch events, yielding
between each call.
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.
This updates the Suspense fuzz tester to use `act` to recursively flush
timers instead of doing it manually.
This still isn't great because ideally the fuzz tester wouldn't fake
timers at all. It should resolve promises using a custom queue instead
of Jest's fake timer queue, like we've started doing in our other
Suspense tests (i.e. the `resolveText` pattern). That's because our
internal `act` API (not the public one, the one we use in our tests)
uses Jest's fake timer queue as a way to force Suspense fallbacks to
appear.
However I'm not interested in upgrading this test suite to a better
strategy right now because if I were writing a Suspense fuzzer today I
would probably use an entirely different approach. So this is just an
incremental improvement to make it slightly less decoupled to React
implementation details.
I rewrote some of our tests that deal with microtasks with the aim of
making them less coupled to implementation details. This is related to
an upcoming change to move update processing into a microtask.
Continuing my journey to migrate all the Scheduler flush* methods to
async versions of the same helpers.
`unstable_flushExpired` is a rarely used helper that is only meant to be
used to test a very particular implementation detail (update starvation
prevention, or what we sometimes refer to as "expiration").
I've prefixed the new helper with `unstable_`, too, to indicate that our
tests should almost always prefer one of the other patterns instead.
Added an explicit type to all $FlowFixMe suppressions to reduce
over-suppressions of new errors that might be caused on the same lines.
Also removes suppressions that aren't used (e.g. in a `@noflow` file as
they're purely misleading)
Test Plan:
yarn flow-ci
Today if something suspends, React will continue rendering the siblings
of that component.
Our original rationale for prerendering the siblings of a suspended
component was to initiate any lazy fetches that they might contain. This
was when we were more bullish about lazy fetching being a good idea some
of the time (when combined with prefetching), as opposed to our latest
thinking, which is that it's almost always a bad idea.
Another rationale for the original behavior was that the render was I/O
bound, anyway, so we might as do some extra work in the meantime. But
this was before we had the concept of instant loading states: when
navigating to a new screen, it's better to show a loading state as soon
as you can (often a skeleton UI), rather than delay the transition.
(There are still cases where we block the render, when a suitable
loading state is not available; it's just not _all_ cases where
something suspends.) So the biggest issue with our existing
implementation is that the prerendering of the siblings happens within
the same render pass as the one that suspended — _before_ the loading
state appears.
What we should do instead is immediately unwind the stack as soon as
something suspends, to unblock the loading state.
If we want to preserve the ability to prerender the siblings, what we
could do is schedule special render pass immediately after the fallback
is displayed. This is likely what we'll do in the future. However, in
the new implementation of `use`, there's another reason we don't
prerender siblings: so we can preserve the state of the stack when
something suspends, and resume where we left of when the promise
resolves without replaying the parents. The only way to do this
currently is to suspend the entire work loop. Fiber does not currently
support rendering multiple siblings in "parallel". Once you move onto
the next sibling, the stack of the previous sibling is discarded and
cannot be restored. We do plan to implement this feature, but it will
require a not-insignificant refactor.
Given that lazy data fetching is already bad for performance, the best
trade off for now seems to be to disable prerendering of siblings. This
gives us the best performance characteristics when you're following best
practices (i.e. hoist data fetches to Server Components or route
loaders), at the expense of making an already bad pattern a bit worse.
Later, when we implement resumable context stacks, we can reenable
sibling prerendering. Though even then the use case will mostly be to
prerender the CPU-bound work, not lazy fetches.
I updated some of the Float tests that intentionally trigger an error to
assert on the specific error message, rather than swallow any errors
that may or may not happen.
Node's MessageChannel implementation will leak if you don't explicitly
close the port. This updates the enqueueTask function we use in our
internal testing helpers to close the port once it's no longer needed.
Prior to #26347, our internal `act` API (not the public API) behaved
differently depending on whether the scope function returned a promise
(i.e. was an async function), for historical reasons that no longer
apply. Now that this is fixed, I've codemodded all async act scopes that
don't contain an await to be sync.
No pressing motivation other than it looks nicer and the codemod was
easy. Might help avoid confusion for new contributors who see async act
scopes with nothing async inside and infer it must be like that for a
reason.
This adds an async gap to our internal implementation of `act` (the one
used by our repo, not the public API). Rather than call the provided
scope function synchronously when `act` is called, we call it in a
separate async task. This is an extra precaution to ensure that our
tests do not accidentally rely on work being queued synchronously,
because that is an implementation detail that we should be allowed to
change. We don't do this in the public version of `act`, though we maybe
should in the future, for the same rationale. That might be tricky,
though, because it could break existing tests.
This also fixes the issue where our internal `act` requires an async
function. You can pass it a regular function, too.
To wait for the microtask queue to empty, our internal test helpers
schedule an arbitrary task using `setImmediate`. It doesn't matter what
kind of task it is, only that it's a separate task from the current one,
because by the time it fires, the microtasks for the current event will
have already been processed.
The issue with `setImmediate` is that Jest mocks it. Which can lead to
weird behavior.
I've changed it to instead use a message event, via the MessageChannel
implementation exposed by the `node:worker_threads` module.
We should consider doing this in the public implementation of `act`,
too.
This is not a public API. We only use it for our internal tests, the
ones in this repo. Let's move it to this private package. Practically
speaking this will also let us use async/await in the implementation.
(This only affects our own internal repo; it's not a public API.)
I think most of us agree this is a less confusing name. It's possible
someone will confuse it with `console.log`. If that becomes a problem we
can warn in dev or something.
I'm in the process of codemodding our test suite to the waitFor pattern.
See #26285 for full context.
This module required a lot of manual changes so I'm doing it as its own
PR. The reason is that most of the tests involved simulating an async
import by wrapping them in `Promise.resolve()`, which means they would
immediately resolve the next time the microtask queue was flushed. I
rewrote the tests to resolve the simulated import explicitly.
While converting these tests, I also realized that the `waitFor` helpers
weren't properly waiting for the entire microtask queue to recursively
finish — if a microtask schedules another microtask, the subsequent one
wouldn't fire until after `waitFor` had resolved. To fix this, I used
the same strategy as `act` — wait for a real task to finish before
proceeding, such as a message event.
Over the years, we've gradually aligned on a set of best practices for
for testing concurrent React features in this repo. The default in most
cases is to use `act`, the same as you would do when testing a real
React app. However, because we're testing React itself, as opposed to an
app that uses React, our internal tests sometimes need to make
assertions on intermediate states that `act` intentionally disallows.
For those cases, we built a custom set of Jest assertion matchers that
provide greater control over the concurrent work queue. It works by
mocking the Scheduler package. (When we eventually migrate to using
native postTask, it would probably work by stubbing that instead.)
A problem with these helpers that we recently discovered is, because
they are synchronous function calls, they aren't sufficient if the work
you need to flush is scheduled in a microtask — we don't control the
microtask queue, and can't mock it.
`act` addresses this problem by encouraging you to await the result of
the `act` call. (It's not currently required to await, but in future
versions of React it likely will be.) It will then continue flushing
work until both the microtask queue and the Scheduler queue is
exhausted.
We can follow a similar strategy for our custom test helpers, by
replacing the current set of synchronous helpers with a corresponding
set of async ones:
- `expect(Scheduler).toFlushAndYield(log)` -> `await waitForAll(log)`
- `expect(Scheduler).toFlushAndYieldThrough(log)` -> `await
waitFor(log)`
- `expect(Scheduler).toFlushUntilNextPaint(log)` -> `await
waitForPaint(log)`
These APIs are inspired by the existing best practice for writing e2e
React tests. Rather than mock all task queues, in an e2e test you set up
a timer loop and wait for the UI to match an expecte condition. Although
we are mocking _some_ of the task queues in our tests, the general
principle still holds: it makes it less likely that our tests will
diverge from real world behavior in an actual browser.
In this commit, I've implemented the new testing helpers and converted
one of the Suspense tests to use them. In subsequent steps, I'll codemod
the rest of our test suite.
