Mike Pilgrem
Mike Pilgrem
I have not investigated what the current CI actually tests but, having identified the original GHC issue https://gitlab.haskell.org/ghc/ghc/-/issues/3837, I tested on Windows 11 with the given `A.hsc`: ~~~Haskell -- エ...
I have converted the pull request to draft because it has been suggested that, compared to other approaches, `unsnoc` has performance issues. I am looking into that.
@sol, this does, de facto, only affect the readability of generated Cabal files. However, I think that is not unimportant: an important purpose of Cabal files is to be read...
(A related but distinct question is: could the Hpack format itself avoid a 'staircase'? Maybe something like: ~~~yaml case: - condition: os(windows) source-dirs: windows - condition: os(darwin) source-dirs: macos -...
The use of the conditions directly as YAML keys is interesting, but a YAML dictionary is not ordered - I assume you had in mind a list: ~~~yaml when: -...
@ulysses4ever, I'm happy to raise it on GHC's issue tracker. What I was not sure about is that switching between GHCs also switches between the version of the `Cabal` library...
@ulysses4ever, I am pretty sure Stack builds a `Setup.exe` with the version of `Cabal` that comes with the relevant version of GHC, and then uses that to interact with Cabal...
The GHC issue is here: https://gitlab.haskell.org/ghc/ghc/-/issues/22333
May I ask, why was the release of `servant-client-0.20.2` only a minor version change when the visible data constructor of `ClientEnv` was changed? I read Hackage's PVP as requiring a...
Yes, that is correct #1720.