However I don't have the same runtime performance as with shapeless. Optim in progress !
-
Show this thread
-
Replying to @ahoy_jon
I believe
@propensive would be much interested to know the example and impact3 replies 0 retweets 1 like -
-
I'm planning to take a quick look at this with
@fommil today. He's doing some profiling now. I won't have time until the evening, but go ahead and share ideas if you have any! I have some changes planned to avoid constructing various intermediate objects.1 reply 0 retweets 1 like -
This Tweet is unavailable.
-
It's exactly that, issue is in the code, not in magnolia, but the code is structured by magnolia. Maybe that https://github.com/propensive/magnolia/issues/107 … (or https://github.com/propensive/magnolia/issues/115 …) would help. Using foldleft-like HoF definitions instead of seq...
0 replies 0 retweets 0 likes -
This Tweet is unavailable.
-
This Tweet is unavailable.
-
So i got back some perf back by defining the Typeclass as a sealed trait for this function and returning different sub Typeclasses for combine and dispatch.pic.twitter.com/ACe6AVdgrH
1 reply 0 retweets 1 like -
I did some experiments with
@fommil this evening, and we gained some more performance internally, too! Now looking faster than shapeless at runtime. I've just pushed a 0.9.1 release including these enhancements, if you want to try this, too!1 reply 1 retweet 6 likes
Your approach looks quite interesting, though. I'm sure @fommil will want to try that. We should take a look tomorrow!
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.