Cool 😄 Could you please update CONTRIBUTORS.md too?

👍 for this 😄

Very cool, can't wait for this to get merged. But what about unicode lambda?

But what about unicode lambda?

Oh yeah, that'd be nice to have too.

I will take a look at it later. I tried a similar quick fix but λ isn't a symbol but a identifier so I have to do some more work.

I don't think we need a new lexeme. We can just use reserved to parse a lambda, surely?

No reserved doesn't work because λa is a valid identifier. Lexing λa → a will result in Right ["λa",->,"a"] instead ofRight [,"a",->,"a"]`.

Oh, that's a shame. I'm not sure what the best option is, in that case.

Perhaps we merge without λ, as you had it before, in 0.8, and revisit λ in 0.8.*. That would be my vote.

Another option is to use a space between the λ and the binder, but that's not going to look very good.

In either case, though, I think we can solve this without adding \ as a lexeme. Maybe a better solution is for char 'λ' *> notFollowedBy letter to be a lexeme.

Related discussion for the same thing in GHC: https://ghc.haskell.org/trac/ghc/ticket/1102

@paf31 While I still think lambda would fit in to be a lexeme without cluttering it, I also agree that it would be best to revert to the previous commit and discuss later how to solve the lambda problem

Could you please rebase, and I'll get this merged. Thanks.

I'm sorry, by rebase I meant "and squash commits". Could you please do that so we can keep the commit log as simple as possible?

Looks great, thanks! I'll wait for Travis and then get this merged in.

@DavidLindbom Would you be interested in showing this off in 5-10 minutes at the next online meetup? Are you on Twitter? Otherwise I can email.

Maybe a better solution is for char 'λ' *> notFollowedBy letter to be a lexeme.

I had the same idea. Were there any downsides?