
I'm not sure why cmptype lists the tuple instances explicitly, since the plugin also handles tuple types. I can delete those cases and it still passes the tests. -Michael On Sun, Dec 29, 2019 at 12:51 AM Justin Paston-Cooper < paston.cooper@gmail.com> wrote:
Looks like cmptype wins on genericity. Out of interest, is there any alternative these days to writing the tuple instances of CmpType explicitly?
On Sat, 28 Dec 2019 at 23:29, Michael Sloan
wrote: Cool approach!
cmptype does something similar, but requires a ghc plugin:
https://github.com/isovector/type-sets/tree/master/cmptype
I suppose one advantage of cmptype is that it can work with types that
cannot derive Generic, such as GADTs.
-Michael
On Sat, Dec 28, 2019 at 3:23 AM Justin Paston-Cooper <
paston.cooper@gmail.com> wrote:
Hello,
I've defined a module which defines a Data.Type.Ordering over types which are instances of Generic:
https://github.com/jpcooper/generic-type-ordering/blob/master/src/Data/Type/... .
I was considering submitting it to Hackage if anyone finds this useful.
I am using it in conjunction with Data.Type.Set (type-level-sets), which requires an ordering over the used types.
The ordering works by first comparing the package, then the module, then the type name. I take it that the ordering as defined is correct over all types which are an instance of Generic.
I wanted to check first whether this is done anywhere else, or in a different, better way. Can anyone advise?
Cheers,
Justin _______________________________________________ Haskell-Cafe mailing list To (un)subscribe, modify options or view archives go to: http://mail.haskell.org/cgi-bin/mailman/listinfo/haskell-cafe Only members subscribed via the mailman list are allowed to post.