Felix Paulusma

I've also made a PoC of a type families based Auth handling -> #1560 If anyone has any feedback, that'd be great. EDIT: Seems I misappropriated the speed up to...

I think I found the problem. It's the `authCheck` in the `HasServer` instance for `Auth`. The `runAuth` will happily return something other than `Authenticated`, but this doesn't actually throw a...

Just ran the same benchmark as @bergey in the first post on my machine (laptop from like 2014-2015), and then changed all benchmark implementations to using `NamedRoutes`. I built everything...

Here examples of how the modules look and the complete output of the bench: Example of the anonymous routes ```haskell {-# LANGUAGE TypeOperators #-} {-# LANGUAGE DataKinds #-} {-# LANGUAGE...

Is it maybe because the flat API isn't actually flat with the anonymous routes, but nested with every extra `:`? ``` Route1 : (Route2 : (Route3 : ...) ) )...

@tfausak I'm not completely sure, but it looks like [you're building with `-O1`](https://gist.github.com/tfausak/8019733fb5c703994d1665143c60ad0f#file-brittany-benchmark-hs-L137), where we're all talking about compiling with `-O2` (at least that's what the title of this issue...

Ok, rebuilt again with just `lts-18.27` and then with `extra-deps: [servant-0.19, servant-server-0.19]` (added the NamedRoutes times again just for reference, did not rebench those) So yeah... if anyone knows the...

> Just tried with GHC 8.10.7 in a slightly modified servant repository — still can't reproduce the huge gains witnessed by some, so obviously I am not able to `git...

> Just to clarify my understanding here (I am not overly familiar with `servant-auth`). With the current `Auth` combinator, if two modes of authentication are defined (say, `Cookie` and `JWT`,...

Note to self: * make stack build every Servant package with `lts-18.27` * make two instances of `HasServer (NewAuth ...` * `TypeError ... => NewAuth '[]` * `NewAuth (auth ':...