Firstyear

I'd support this change :)

I've just run into this via https://github.com/http-rs/surf/issues/295, is there any action to follow this up?

As an extra comment, would it be worth extending this to other areas like headers to limit the size of headers that are received too? Another design consideration is that...

It's hard to see a way to unlike "correctness" at the protocol level and the "practical" elements of networking in this case. We have a part of the protocol indicating...

@danieleades since http-types is already checking this (partially) would it be worth extending this to allow configuration of the maximum allowed request size, and this middleware just exposes that functionality?...

Could this be that h1 or other http layers lower in the stack are actually blocking the request once you hit the limit set in length? If you do the...

Well, if the lower layers are already doing this work, there seems to be little sense in double accounting - it seems like the lower levels should be configurable to...

> Seems that Warp's approach is to not handle this at the lower layers in favour of handling it explicitly at the top layer. In a way I think this...

> Technically, Google OAuth is already OIDC, is it not? Google Oauth is an OIDC authentication provider. This is a request that Snipe supports being a relying party who consumes...

https://github.com/tokio-rs/tokio/blob/master/tokio/src/runtime/driver.rs#L10 When enable_io is used, process, signal and io handlers are all installed. Unless I'm mistaken this is the what I'm asking, if this could be enabled in a more...