ducdetronquito

> >Has an extra never_index field (link) > > This should continue to exist Could you explain why ? It seems only related to how `Headers` internal storage works, but...

Oh sorry, I had only the HTTP/1.1 in mind: thanks for the explanation ! I will make a PR this weekend to make `HeaderEntry` public and provide an allocation free...

Hi @daurnimator, After re-reading the HTTP/2 specs you linked about [never-indexed literals](https://httpwg.org/specs/rfc7541.html#rfc.section.7.1.3), it seems this concept only apply for the HPACK compression step. Moreover, the specs refers to it as...

Hi everyone, I think it is a good move to remove the HTTP headers implementation from the standard library, as it can allow us more freedom to experiment things at...

Thanks !

First rework step made with the PR https://github.com/ducdetronquito/http/pull/7. The idea is to avoid the complexity of a proper multimap implementation by providing a simple list of headers. **Advantages**: - Very...

Headers in the standard library `asynchttpserver` and `asynchttpcore` do not follow the [RFC 6265](https://tools.ietf.org/html/rfc6265). When I have a better understanding of the RFC I may submit PRs to Nim, in...

Hi @elerch ! Sorry for the very, very late answer: I hadn't time to work on requestz after summer... Thanks for your PR, it was a good catch ! I...

Hi @helinwang ! Could you give me a snippet of code to reproduce your problem ? Have a nice day :)

After very light analysis, it seems that `api.forismatic.com` used TLS 1.3 but [iguanaTLS](https://github.com/alexnask/iguanaTLS), the library *requestz* uses to enable TLS, only support TLS 1.2. I am not very confortable with...