Add lazy reasons to browser() - #37241
Merged
Merged
Conversation
github-actions
Bot
added
the
React Core Team
gnoff
force-pushed
the
jstory/browser-reason
branch
2 times, most recently
from
August 7, 2026 22:16
42a4856 to
d199178
Compare
acdlite
approved these changes
Aug 8, 2026
gnoff
force-pushed
the
jstory/browser-reason
branch
4 times, most recently
from
August 8, 2026 06:03
7b8c048 to
d56e1cc
Compare
Changes `ReactDOM.browser()` to return a cheap branded recoverable token instead of eagerly constructing an `Error`. It accepts an optional reason string or initializer that runs only when a server renderer consumes the token and may return any value; the client renderer ignores the reason without invoking the initializer, so browser-only rendering does not pay for an unused stack. When Fizz consumes the token through `use()` or `abort()`, it creates a consistent browser-bailout error at the consumption point so its stack identifies the relevant operation. The initialized reason is preserved unchanged as the optional `cause`, allowing strings, errors, and structured framework metadata without runtime validation. If an initializer throws, Fizz substitutes a stable diagnostic fallback so reason generation cannot change rendering control flow. Successful recoveries report the error through `onBrowserBailout`. When no Suspense boundary can recover the render, Fizz clones the branded recoverable error into an unbranded fatal diagnostic while preserving its cause and consumption frames. During an abort, the request retains the original branded error so every remaining task observes the same reason; fatal clones are created only when reporting a fatal root or closing the stream. Centralized recoverable logging uses the brand to route successful bailouts through `onBrowserBailout` and fatal clones through `onError`. The empty recoverable digest and client hydration suppression behavior remain unchanged. Tests cover omitted and direct reasons, lazy string, error, structured, and primitive reasons, repeated use sites, throwing initializers, lazy client behavior, consumption stacks, flattened fatal errors, recoverable and fatal use and abort paths, nested aborts, direct throws, debug tools, and development and production rendering.
gnoff
force-pushed
the
jstory/browser-reason
branch
from
August 8, 2026 15:07
d56e1cc to
db48e6e
Compare
acdlite
approved these changes
Aug 10, 2026
github-actions Bot
pushed a commit
that referenced
this pull request
Aug 10, 2026
Changes `ReactDOM.browser()` to return a cheap branded recoverable token instead of eagerly constructing an `Error`. It accepts an optional reason string or initializer that runs only when a server renderer consumes the token and may return any value; the client renderer ignores the reason without invoking the initializer, so browser-only rendering does not pay for an unused stack. When Fizz consumes the token through `use()` or `abort()`, it creates a consistent browser-bailout error at the consumption point so its stack identifies the relevant operation. The initialized reason is preserved unchanged as the optional `cause`, allowing strings, errors, and structured framework metadata without runtime validation. If an initializer throws, Fizz substitutes a stable diagnostic fallback so reason generation cannot change rendering control flow. Successful recoveries report the error through `onBrowserBailout`. When no Suspense boundary can recover the render, Fizz clones the branded recoverable error into an unbranded fatal diagnostic while preserving its cause and consumption frames. During an abort, the request retains the original branded error so every remaining task observes the same reason; fatal clones are created only when reporting a fatal root or closing the stream. Centralized recoverable logging uses the brand to route successful bailouts through `onBrowserBailout` and fatal clones through `onError`. The empty recoverable digest and client hydration suppression behavior remain unchanged. Tests cover omitted and direct reasons, lazy string, error, structured, and primitive reasons, repeated use sites, throwing initializers, lazy client behavior, consumption stacks, flattened fatal errors, recoverable and fatal use and abort paths, nested aborts, direct throws, debug tools, and development and production rendering. DiffTrain build for [807d21f](807d21f)
github-actions Bot
pushed a commit
that referenced
this pull request
Aug 10, 2026
Changes `ReactDOM.browser()` to return a cheap branded recoverable token instead of eagerly constructing an `Error`. It accepts an optional reason string or initializer that runs only when a server renderer consumes the token and may return any value; the client renderer ignores the reason without invoking the initializer, so browser-only rendering does not pay for an unused stack. When Fizz consumes the token through `use()` or `abort()`, it creates a consistent browser-bailout error at the consumption point so its stack identifies the relevant operation. The initialized reason is preserved unchanged as the optional `cause`, allowing strings, errors, and structured framework metadata without runtime validation. If an initializer throws, Fizz substitutes a stable diagnostic fallback so reason generation cannot change rendering control flow. Successful recoveries report the error through `onBrowserBailout`. When no Suspense boundary can recover the render, Fizz clones the branded recoverable error into an unbranded fatal diagnostic while preserving its cause and consumption frames. During an abort, the request retains the original branded error so every remaining task observes the same reason; fatal clones are created only when reporting a fatal root or closing the stream. Centralized recoverable logging uses the brand to route successful bailouts through `onBrowserBailout` and fatal clones through `onError`. The empty recoverable digest and client hydration suppression behavior remain unchanged. Tests cover omitted and direct reasons, lazy string, error, structured, and primitive reasons, repeated use sites, throwing initializers, lazy client behavior, consumption stacks, flattened fatal errors, recoverable and fatal use and abort paths, nested aborts, direct throws, debug tools, and development and production rendering. DiffTrain build for [807d21f](807d21f)
5 tasks
This file contains hidden or 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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
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.
Changes
ReactDOM.browser()to return a cheap branded recoverable token instead of eagerly constructing anError. It accepts an optional reason string or initializer that runs only when a server renderer consumes the token and may return any value; the client renderer ignores the reason without invoking the initializer, so browser-only rendering does not pay for an unused stack.When Fizz consumes the token through
use()orabort(), it creates a consistent browser-bailout error at the consumption point so its stack identifies the relevant operation. The initialized reason is preserved unchanged as the optionalcause, allowing strings, errors, and structured framework metadata without runtime validation. If an initializer throws, Fizz substitutes a stable diagnostic fallback so reason generation cannot change rendering control flow. Successful recoveries report the error throughonBrowserBailout.When no Suspense boundary can recover the render, Fizz clones the branded recoverable error into an unbranded fatal diagnostic while preserving its cause and consumption frames. During an abort, the request retains the original branded error so every remaining task observes the same reason; fatal clones are created only when reporting a fatal root or closing the stream. Centralized recoverable logging uses the brand to route successful bailouts through
onBrowserBailoutand fatal clones throughonError. The empty recoverable digest and client hydration suppression behavior remain unchanged.Tests cover omitted and direct reasons, lazy string, error, structured, and primitive reasons, repeated use sites, throwing initializers, lazy client behavior, consumption stacks, flattened fatal errors, recoverable and fatal use and abort paths, nested aborts, direct throws, debug tools, and development and production rendering.