ghci> :set -XBlockArguments
ghci> let be = id in do be do be do ()
()

On Wed, Feb 21, 2024 at 3:18 AM Johannes Waldmann <johannes.waldmann@htwk-leipzig.de> wrote:
Dear Cafe,

I was surprised to see  ghc  accept this code

ghci> do let {x = False}; not x
True

because I initially thought: there's a `do`, so there must be some monad,
but which is it? some implicit Identity monad perhaps?

But the type is indeed plain `Bool`, and the explanation is (I think):
there is a `do` but there is no monad, since the translation according to
( https://www.haskell.org/onlinereport/haskell2010/haskellch3.html#x8-470003.14 )
does never produce `(>>=)`.

So, all is fine, nothing to see here.
Perhaps keep in mind as an edge case, useful to confuse students.
Actually, to motivate them to read the language standard ...

- J.W.
_______________________________________________
Haskell-Cafe mailing list
To (un)subscribe, modify options or view archives go to:
http://mail.haskell.org/cgi-bin/mailman/listinfo/haskell-cafe
Only members subscribed via the mailman list are allowed to post.