Conversation

Notices

1. Embed this notice
   tech? no! man, see... (technomancy@icosahedron.website)'s status on Wednesday, 14-Jun-2023 01:19:36 JST tech? no! man, see...

   • Devine Lu Linvega
   • golly josh darn

   @neauoire @klardotsh well, there are a bunch of very obvious rules like break after every top-level form, break after every function arglist, break after every k/v pair in a let, between pattern match clauses, etc

   the main one I'm not happy with is that if the current form would push it over 80 columns, we break there, but it gives you weird results sometimes because a human would have noticed earlier that you were coming up on the limit and broken the line earlier

   kinda figure I need to add some logic to be able to go back in time a bit?

   • Embed this notice
     tech? no! man, see... (technomancy@icosahedron.website)'s status on Wednesday, 14-Jun-2023 02:40:49 JST tech? no! man, see...

     in reply to

     • Devine Lu Linvega
     • Reid D. M.
     • golly josh darn

     @arrdem @neauoire @klardotsh I appreciate the sentiment but when we're talking about fixing a bug in a 300-line program and the first step is "read a thesis written by a Haskell user" I'm thinking that's probably not happening =)

   • Embed this notice
     Reid D. M. (arrdem@macaw.social)'s status on Wednesday, 14-Jun-2023 02:40:50 JST Reid D. M.

     in reply to

     • Devine Lu Linvega
     • golly josh darn

     @technomancy @neauoire @klardotsh There's a very fun Haskell community thesis I remember reading that deals explicitly with building backtracking/elastic formatting engines for this sort of thing. I think it's referenced by one of the clojure pretty-printers but it's been a while.