<@UE2FBFZFH> do you have a haskell -> jvm bytec...
# arrowj
jimn
10/26/2019, 1:53 PM@Jannis do you have a haskell -> jvm bytecode emitter?
j
Jannis
10/26/2019, 1:59 PMThere is eta-lang (ghc haskell compiler frontent + jvm bytecode backend), which is usable but discontinued afaik. If you need the jvm then I'd look at scala for equivalents or write a basic parser combinator library yourself (I'd love to help out, but I have some other projects atm that need to be done first). Basic parser combinators are not too hard, but require a bit of work.
Jannis
10/26/2019, 2:02 PMWriting a parser lib is definitly on my todo list as that is something arrow will excel at, especially with implicit typeclass resolution and that is also immensly useful.
Jannis
10/26/2019, 2:02 PMAlso it's fun ^-^
j
jimn
10/26/2019, 2:04 PMkotlin is at a good centroid of many concerns, and imho there's no reason not to add the haskell killer features to the basic kotlin toolkit, whether by arrow or elsewhere.
jimn
10/26/2019, 2:06 PMhotspot, llvm, and js runtimes are at least adequate runtime targets with working sample code. these may need individual optimizations.
jimn
10/26/2019, 2:07 PMarrow has gone ahead with compiler extensions and bucked a few kotlin limitations to date.
j
Jannis
10/26/2019, 2:15 PMhaskell imo is far far superior to kotlin in most language related aspects. Arrow with the compiler plugin will catch up a tiny bit soon tho...
GHC has a fair few backends it supports as well (native, llvm, js and soon web assembly, all with some limitations tho), but a missing jvm backend is quite annoying with some many existing projects being mostly jvm projects.
j
jimn
10/26/2019, 3:17 PMi don't know enough about the fundamental impedences to haskell but im of the opinion that some kotlinc extentions might enable hornclauses that define grammars that enable compile-time polymoprhic syntax changes to match a problem without divorcing the toolkit and runtime or forcing an external library binding both ways
jimn
10/26/2019, 3:19 PMi don't honestly know if Arrow is fundamentally moving toward that particular direction, but it seems to need some progress with similar factors to arrive at the FP features
2 Views