Skip to content

Conversation

@dicej
Copy link
Collaborator

@dicej dicej commented Nov 10, 2025

I realized that some of my componentize-py tests weren't actually testing what they were supposed to, and that was due to the {stream|future}.{read|write} intrinsics being lowered synchronously 🤦. This fixes that.

Note that we're still lowering the {stream|future}.cancel-{read|write} intrinsics synchronously, same as the Rust wit-bindgen does. We might need to change that (or make it configurable) for toolchains that actually have a good async cancellation paradigm.

I realized that some of my `componentize-py` tests weren't actually testing what
they were supposed to, and that was due to the `{stream|future}.{read|write}`
intrinsics being lowered synchronously 🤦.  This fixes that.

Note that we're still lowering the `{stream|future}.cancel-{read|write}`
intrinsics synchronously, same as the Rust `wit-bindgen` does.  We might need to
change that (or make it configurable) for toolchains that actually have a good
async cancellation paradigm.

Signed-off-by: Joel Dice <[email protected]>
@dicej dicej requested a review from alexcrichton November 10, 2025 20:08
@dicej dicej requested a review from a team as a code owner November 10, 2025 20:08
dicej added a commit to dicej/componentize-py that referenced this pull request Nov 10, 2025
Due to the `wit-dylib` issue that
bytecodealliance/wasm-tools#2371 addresses, these tests
weren't actually testing what I thought they were.  Fixing that revealed a
couple of bugs involving stream writes, which this fixes.

Also, bump version to 0.19.2

Signed-off-by: Joel Dice <[email protected]>
dicej added a commit to dicej/componentize-py that referenced this pull request Nov 10, 2025
Due to the `wit-dylib` issue that
bytecodealliance/wasm-tools#2371 addresses, these tests
weren't actually testing what I thought they were.  Fixing that revealed a
couple of bugs involving stream writes, which this fixes.

Also, bump version to 0.19.2

Signed-off-by: Joel Dice <[email protected]>
@dicej dicej added this pull request to the merge queue Nov 10, 2025
Merged via the queue into bytecodealliance:main with commit 37f05db Nov 10, 2025
34 checks passed
@dicej dicej deleted the async-read-write branch November 10, 2025 21:21
dicej added a commit to dicej/componentize-py that referenced this pull request Nov 10, 2025
Due to the `wit-dylib` issue that
bytecodealliance/wasm-tools#2371 addresses, these tests
weren't actually testing what I thought they were.  Fixing that revealed a
couple of bugs involving stream writes, which this fixes.

Also, bump version to 0.19.2

Signed-off-by: Joel Dice <[email protected]>
dicej added a commit to bytecodealliance/componentize-py that referenced this pull request Nov 10, 2025
Due to the `wit-dylib` issue that
bytecodealliance/wasm-tools#2371 addresses, these tests
weren't actually testing what I thought they were.  Fixing that revealed a
couple of bugs involving stream writes, which this fixes.

Also, bump version to 0.19.2

Signed-off-by: Joel Dice <[email protected]>
mxgordon pushed a commit to YellowPapayas/wasm-tools that referenced this pull request Nov 11, 2025
…iance#2371)

I realized that some of my `componentize-py` tests weren't actually testing what
they were supposed to, and that was due to the `{stream|future}.{read|write}`
intrinsics being lowered synchronously 🤦.  This fixes that.

Note that we're still lowering the `{stream|future}.cancel-{read|write}`
intrinsics synchronously, same as the Rust `wit-bindgen` does.  We might need to
change that (or make it configurable) for toolchains that actually have a good
async cancellation paradigm.

Signed-off-by: Joel Dice <[email protected]>
mxgordon added a commit to YellowPapayas/wasm-tools that referenced this pull request Nov 11, 2025
mxgordon added a commit to YellowPapayas/wasm-tools that referenced this pull request Nov 11, 2025
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants