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Outdated suggestions cannot be applied.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.
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
fs, stream: add initial
Symbol.dispose
andSymbol.asyncDispose
support #48518fs, stream: add initial
Symbol.dispose
andSymbol.asyncDispose
support #48518Changes from 8 commits
8a25397
8dc2ee7
223a2d8
40ba972
e8a014f
48602bb
39b6ace
a6d798d
98285ae
5d0d435
File filter
Filter by extension
Conversations
Jump to
There are no files selected for viewing
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Doesn't this mean Symbol.keyFor of these symbols won't properly return undefined?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Would this be a good thing to polyfill them? Wouldn't users assume certain behaviors from the runtime because they exists?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I don't think there is a way for typescript or bable to recognize these symbols without them being global.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
@mcollina this is confirmed (with the TS team) to play nice. Users would be able to use them with the polyfill but not the syntactic sugar until v8 ships
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
That breaks
core-js
.There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
#48699
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
What is the outcome of the following?
Ideally, the second call should be a noop, but it seems like the second call could actually throw the
AbortError
recorded by the first call, since theAbortError
is only ignored in the first call.For reference, here is the guidance from the spec related to how a
Symbol.asyncDispose
method should behave:There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
The second call would not call
this.destroy
since this.destroyed would be true and thus would not cause any side effects and reject/fulfill with the same way as the first one. So:The only difference is if the stream itself was destroyed with an error (e.g. the first call was unable to destroy it correctly) the second call would also reject with the same error as the first one.
The
AbortError
is filtered out but is still needed since someone may be iterating the stream while we are disposing it and from their point of view an error happened.Another thing to consider is two concurrent calls too
readable[Symbol.asyncDispose]();
for example:In that case I think it's possible the error is emitted twice which may be a bug, but I think it's fine since
.destroy
sets.destroyed
synchronously IIRC - what do you think @ronag , should be fine ?There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Thanks for explaining. I wanted to verify because it seemed that, if
eos
could emit theAbortError
from the call tothis.destroy()
, then it would potentially be possible to observe thatAbortError
out of band in either a sequential or concurrent call.It might make sense to have tests for these two cases, at least, to verify these expectations:
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
@rbuckton wanna open a PR? If you do, the easiest would be to add it to test/parallel/test-stream-readable-dispose.js and you don't have to build node for it you can use a nightly since it's (hopefully) a passing test.
It could be aa good opportunity to practice contributing code to Ndoe :)
Otherwise I'll add the test when I tackle the next stream API. Would this sort of test make sense for any disposable? Is there a list of principles we should follow?