Wasmtime GitHub notifications bot (Oct 13 2026 at 16:14): Wasmtime GitHub notifications bot (Oct 13 2026 at 16:14):xia-chao opened PR #14702 from xia-chao:fix/wasi-http-field-map-capacity-error to bytecodealliance:main:
A full HeaderMap made Fields::set/delete report invalid-syntax for valid names. http's try_entry was treating MaxSizeReached as InvalidHeaderName. This PR switches to try_insert/try_append.
Wasmtime GitHub notifications bot (Oct 13 2026 at 16:14):xia-chao requested rvolosatovs for a review on PR #14702.
Wasmtime GitHub notifications bot (Oct 13 2026 at 16:14):xia-chao requested wasmtime-wasi-reviewers for a review on PR #14702.
Wasmtime GitHub notifications bot (Oct 13 2026 at 17:28)::repeat: rvolosatovs submitted PR review.
Wasmtime GitHub notifications bot (Oct 13 2026 at 17:28)::speech_balloon: rvolosatovs created PR review comment:
try_appendcould fail here, which could leave the map in inconsistent state and incorrectly report size
Wasmtime GitHub notifications bot (Oct 13 2026 at 17:28)::speech_balloon: rvolosatovs created PR review comment:
could we make this smaller?
Wasmtime GitHub notifications bot (Oct 13 2026 at 17:28)::speech_balloon: rvolosatovs created PR review comment:
this seems to be testing implementation details of
httpcrate, instead, let's assert the actual results we expect (e.g. error with TooManyFields)I'd suggest breaking each statement line-by-line and asserting result of each operation individually without using an array
Wasmtime GitHub notifications bot (Oct 13 2026 at 17:28)::speech_balloon: rvolosatovs created PR review comment:
The
keywas already parsed here, so could we keep using the entry API if we did something like:map.try_entry(key).map_err(|_| FieldMapError::TooManyFields)?
Wasmtime GitHub notifications bot (Oct 13 2026 at 17:31)::memo: rvolosatovs submitted PR review.
Wasmtime GitHub notifications bot (Oct 13 2026 at 17:31)::speech_balloon: rvolosatovs created PR review comment:
// `try_entry` misreports a full map as `InvalidHeaderName`.
Wasmtime GitHub notifications bot (Oct 13 2026 at 17:32)::speech_balloon: rvolosatovs deleted PR review comment.
Wasmtime GitHub notifications bot (Oct 13 2026 at 17:32)::speech_balloon: rvolosatovs edited PR review comment.
Wasmtime GitHub notifications bot (Oct 13 2026 at 18:46):xia-chao updated PR #14702.
Wasmtime GitHub notifications bot (Oct 13 2026 at 18:53):xia-chao commented on PR #14702:
Done! Thanks!
Wasmtime GitHub notifications bot (Oct 13 2026 at 19:10)::thumbs_up: rvolosatovs submitted PR review:
Thanks!
There's still a little issue with
setsetting keys that already exist, but this is already an improvement and it fixes an existing issue, so I'm going to merge this.Would you like to work on a follow-up for
setspecifically?
Wasmtime GitHub notifications bot (Oct 13 2026 at 19:10)::speech_balloon: rvolosatovs created PR review comment:
this actually looks wrong, because
setshould have replaced the values
Wasmtime GitHub notifications bot (Oct 13 2026 at 19:10)::speech_balloon: rvolosatovs created PR review comment:
Unfortunately, this would still prevent
setfor a key, which already exists when the map is full, even though the values should have been replaced.Perhaps we could work around that by first removing the key?
Wasmtime GitHub notifications bot (Oct 13 2026 at 19:11):rvolosatovs added PR #14702 wasi-http: report a full field map as size-exceeded to the merge queue.
Wasmtime GitHub notifications bot (Oct 13 2026 at 19:35):rvolosatovs removed PR #14702 wasi-http: report a full field map as size-exceeded from the merge queue.
Wasmtime GitHub notifications bot (Oct 13 2026 at 19:35)::check: rvolosatovs merged PR #14702.
Last updated: Oct 31 2026 at 23:03 UTC