[Haskell-cafe] Haskell grammar ambiguity

[Viktor Dukhovni]

> On 16 Jun 2021, at 10:38 am, Askar Safin via Haskell-Cafe <haskell-cafe at haskell.org> wrote:
> 
> I agree. I always use curly braces with "do" and "case", i. e.:
> do {
>  ...;
>  ...;
> }

I, on the other hand find, e.g., BlockArguments and ApplicativeDo
to be useful decluttering extensions.  These are not without their
"pockets of support".  Perhaps some of the error messages could be
less optimistic in suggesting "BlockArguments", the user who meant
to use them would typically deduce the forgotten pragma even without
an explicit hint when the hint degrades error quality for others.

The specific error message choices are not inherent to the extension.

-- 
	Viktor.