Khafra

> Fields ([Section 5](https://www.rfc-editor.org/rfc/rfc9110.html#fields)) that are sent or received before the content are referred to as "header fields" (or just "headers", colloquially). > field-value = *field-content field-content = field-vchar [...

I wouldn't expect a malformed formdata to parse successfully, this change seems correct.

Same here, but I'm happily surprised undici is that close in both benchmarks (other than very large messages).

Interestingly, at a certain threshold the fragmentation tests start to pass, such as 9.3.9 & 9.4.9. Sending out smaller chunks = tests fail for some reason.

It's literally a single bug and the rest of the tests will start passing. That's super annoying.

non-strict is passing, no reason to touch them until there's nothing else to do

Do you have an example?

> and still stay within spec. https://fetch.spec.whatwg.org/#dom-request > 3. If parsedURL [includes credentials](https://url.spec.whatwg.org/#include-credentials), then [throw](https://webidl.spec.whatwg.org/#dfn-throw) a [TypeError](https://webidl.spec.whatwg.org/#exceptiondef-typeerror).

There are related failures, but I don't necessarily think this should land as-is even if those failures were fixed.

I don't see the benefit in having 100% code coverage.