Ruben Verborgh
Ruben Verborgh
Might be fixed with #662.
@megoth It was a priority for @timbl.
If you ask me, we can just have the exact same logic, except we don't write a body (so catch all `res.end/write` calls or `pipe`s to it). That should do...
> This should be part of the HTTP stack of the server environment Yes, but we're playing server on a very low level here. I think it is good that...
> either WHATWG would need to change fetch FYI, we were routinely hitting this issue in practice with our RDF agents. This prompted us to propose such a change in...
The problem is that @kristophor will have put `krischen` as his WebID, and that is indeed not an absolute URL.
And browser? Maybe good to use it conditionally for now.
Sweet, I like the idea 👍
RE: degradation, know that the benchmark might have a flawed design. It's an atypical insertion pattern. Retrieval is typically much more common than insertion. Could be an option to have...
Working as intended (writer always expects valid input), but we can indeed accept a PR along the lines of #165 to perform optional input sanitization.