It would have a much more natural vibe to say qualified import A. Let's allow that as well!

It would have a much more natural vibe to say qualified import A. Let's allow that as well!

Yes, technically it's possible, but that would steal syntax.

qualified :: a -> a
qualified x = x

is valid Haskell today. Adding what you suggest would require to steal syntax by turning qualified into true keyword. I will add that remark to alternatives.

since qualified would have to be followed by a true import keyword eventually, maybe Happy's new permutation rules could compensate for that automatically

since qualified would have to be followed by a true import keyword eventually, maybe Happy's new permutation rules could compensate for that automatically

Fair point. It might work indeed. If we write

ghci> qualified import = import

<interactive>:2:11: error: [GHC-58481]
    parse error on input ‘import’

we get confusing parsing error anyway, so in this case we'll probably get error like with

ghci> import qualified Foo Bar

<interactive>:3:22: error: [GHC-58481] parse error on input ‘Bar’

i.e. something like

ghci> qualified import = import

<interactive>:2:17: error: [GHC-58481]
    parse error on input ‘=’

as = will be the first token which doesn't make sense.

rendered PR link is a 404

It is a April Fool Joke Proposal and this day is gone, so maybe it is better to change status to Draft and Closed!
And everyone could still reads it and laughs, but to be not confused with real Proposals.