[GHC] #12162: Concatenation of type level symbols missing

[GHC]

#12162: Concatenation of type level symbols missing
-------------------------------------+-------------------------------------
        Reporter:  augustss          |                Owner:  phadej
            Type:  feature request   |               Status:  new
        Priority:  normal            |            Milestone:
       Component:  Compiler          |              Version:  8.0.1
      Resolution:                    |             Keywords:
Operating System:  Unknown/Multiple  |         Architecture:
                                     |  Unknown/Multiple
 Type of failure:  None/Unknown      |            Test Case:
      Blocked By:                    |             Blocking:
 Related Tickets:                    |  Differential Rev(s):  Phab:D2632
       Wiki Page:                    |
-------------------------------------+-------------------------------------

Comment (by ekmett):

 If we build in

 {{{#!hs
 type family (++) :: Symbol -> Symbol -> Symbol
 }}}

 that doesn't prevent it from later being used in a potential later or even
 third-party polykinded

 {{{#!hs
 type family (<>) :: k -> k -> k
 type instance (<>) = (++)
 }}}

 and it would be consistent with the fact that the other `GHC.TypeLits`
 such as `(+)` are all specific to one kind.

 That said, none of the existing machinery for `KnownNat` knows anything
 about `(+)`, so it'd be a leap to expect `KnownSymbol` to know about
 `(++)`.

 Also note that complicating evidence by giving it more internal hidden
 constructors like is used to deal with `Coercible` instances is actually
 bad for folks who use things like reflection to mimic this behavior or
 construct symbols at runtime today.

 As an aside, you can actually get type level symbol concatenation and
 tools for working with it out of the `constraints` package at least in
 HEAD today, but it doesn't have any compiler assistance. You have to
 explicitly use the various axioms and `appendSymbol` proofs rather
 explicitly to get the resulting `KnownSymbol` instances, but you can do
 it.

 https://github.com/ekmett/constraints/blob/master/src/Data/Constraint/Symbol.hs

--
Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/12162#comment:13>
GHC <http://www.haskell.org/ghc/>
The Glasgow Haskell Compiler