Re: [Haskell-cafe] [ANN] acme-dont-1.2 - a "don't" construct

On Apr 2, 2020, at 10:03 AM, Joachim Breitner wrote:

a crucial design flaw of do notation (and which could be fixed as part of #216): The inability to specify an operation to apply to the do-block as a whole.

Even having seen the dawn of a new day, this is an interesting point. Built-in `do` notation does not need to do any desugaring to a single statement, but maybe variants do. And even within GHC, the last statement of a do-block has special status -- it's just that do-notation treats it like any other statement. So maybe it's worth extending #216 with a new feature that applies an operation to the last statement. Seems to fit very much with the spirit of #216, with little downside. Just posting here to make sure I'm not crazy before posting to the altogether-more-serious #216 itself. Richard