After a day optimizing ASTs, Magnolia now gets approximately 12x better compile-time performance for typeclass derivation than Shapeless!
Show this thread
-
Yes. For starters, I was generating the code for every typeclass twice.
But I also reduced my anonymous class count.
And managed to reuse typeclasses repeated in the same case class. Would be nice to deduplicate across case classes, but I'm not there yet.
-
-
deduplicating might not be trivial & cost a lot... that will be some kind of compromise certainly...
-
I don't think it would be a huge cost, compared to typechecking. My original code wasn't written with this optimization in mind, though...
End of conversation
New conversation -
-
-
Wow! Impressive. Magnolia really has the potential to be one of those tools that makes Scala so much nicer.
-
Wow. I'll scrub my failed attempt with shapeless and try again with magnolia (if it supports the equivalent of LabelledGenerics)
- Show replies
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.