Version
v16.13.2
Platform
Darwin Seans-MacBook-Pro.local 21.2.0 Darwin Kernel Version 21.2.0: Sun Nov 28 20:28:41 PST 2021; root:xnu-8019.61.5~1/RELEASE_ARM64_T6000 arm64
Subsystem
No response
What steps will reproduce the bug?
const stream = new ReadableStream({
pull: c => {
const v = new Uint8Array(c.byobRequest.view.buffer, 0, 3);
v.set([20, 21, 22]);
c.byobRequest.respondWithNewView(v);
},
autoAllocateChunkSize: 10,
type: 'bytes'
});
const reader = stream.getReader();
const view = reader.read();How often does it reproduce? Is there a required condition?
Reproduces everytime, it is dependent on using the autoAllocateChunkSize and NOT using a BYOB reader on the stream.
What is the expected behavior?
I expect no error. By the spec (https://streams.spec.whatwg.org/#rs-byob-request-prototype) it should allowed respond with a new view as long as the underlying buffer is the same as the original view. I don't see a mention of this method being specifically only for BYOB readers and not autoAllocated buffers.
Not that this works without error on Chrome (only browser that currently supports byte streams): https://jsfiddle.net/w1jb7rq2/1/
What do you see instead?
RangeError [ERR_INVALID_ARG_VALUE]: The argument 'view' is invalid. Received Uint8Array(3) [ 20, 21, 22 ]
at new NodeError (node:internal/errors:371:5)
at readableByteStreamControllerRespondWithNewView (node:internal/webstreams/readablestream:2523:11)
at ReadableStreamBYOBRequest.respondWithNewView (node:internal/webstreams/readablestream:682:5)
at Object.pull (/Users/squinlan/Documents/node/test/parallel/test-whatwg-readablebytestream.js:243:21)
at ensureIsPromise (node:internal/webstreams/util:172:19)
at readableByteStreamControllerCallPullIfNeeded (node:internal/webstreams/readablestream:2545:5)
at node:internal/webstreams/readablestream:2662:7 {
code: 'ERR_INVALID_ARG_VALUE'
}Additional information
This is due to readableByteStreamControllerPullSteps missing bufferByteLength in the desc added to pendingPullIntos
Version
v16.13.2
Platform
Darwin Seans-MacBook-Pro.local 21.2.0 Darwin Kernel Version 21.2.0: Sun Nov 28 20:28:41 PST 2021; root:xnu-8019.61.5~1/RELEASE_ARM64_T6000 arm64
Subsystem
No response
What steps will reproduce the bug?
How often does it reproduce? Is there a required condition?
Reproduces everytime, it is dependent on using the autoAllocateChunkSize and NOT using a BYOB reader on the stream.
What is the expected behavior?
I expect no error. By the spec (https://streams.spec.whatwg.org/#rs-byob-request-prototype) it should allowed respond with a new view as long as the underlying buffer is the same as the original view. I don't see a mention of this method being specifically only for BYOB readers and not autoAllocated buffers.
Not that this works without error on Chrome (only browser that currently supports byte streams): https://jsfiddle.net/w1jb7rq2/1/
What do you see instead?
Additional information
This is due to readableByteStreamControllerPullSteps missing bufferByteLength in the desc added to pendingPullIntos