http
http copied to clipboard
hyperium
Rust HTTP types
`http` is using the stable version of `bytes` crate now (#457) so I guess #369 can be reverted, except for the addition of `from_maybe_shared` constructors until `http` v0.3 or v1.0.
Over in non-binary/http-cache-semantics#2, we're using `request::Parts` and `response::Parts` rather than `Request` or `Response` because we never care about the bodies. But the builders are still the easiest way to construct...
I'm currently building a site using actix-web, which relies on the http crate for strongly-typed header constants, and I noticed that there are a couple of headers which don't yet...
Currently, the only way to build an `Authority` object is to parse a string in the form `:`. This can be a bit inconvenient: In the case the host is...
The user requesting `Ord` for `Uri` in #407 suggests that _some_ or even any impl of `PartialOrd` and `Ord` for `Uri` is helpful for composition. Its also necessary for inserting...
I thought about an implementation like this ```rust impl PartialOrd for Uri { fn partial_cmp(&self, other: &Self) -> Option { Some(self.cmp(other)) } } impl Ord for Uri { fn cmp(&self,...
In the current version the lifetime is tied to self which is needlessly restrictive as the strings are in fact static.
What makes this crate and http_types different? They seem to have very similar goals (foundational types for http to interop between different crates) which would - at first glance -...
I was recently reading name.rs, and was surprised to see a lot of hand-coded byte-by-byte comparisons for reading header names: https://github.com/hyperium/http/blob/975dbdd70a882fbf74a77a5b90ad190602bb98ac/src/header/name.rs#L1312-L1320 I'm curious why the code is written this way,...
