#14070: Allow ‘unsafe’ deriving strategy, deriving code with ‘unsafeCoerce’ -------------------------------------+------------------------------------- Reporter: Iceland_jack | Owner: (none) Type: feature request | Status: new Priority: normal | Milestone: Component: Compiler | Version: 8.0.1 Resolution: | Keywords: | QuantifiedContexts, deriving Operating System: Unknown/Multiple | Architecture: | Unknown/Multiple Type of failure: None/Unknown | Test Case: Blocked By: | Blocking: Related Tickets: | Differential Rev(s): Wiki Page: | -------------------------------------+------------------------------------- Comment (by RyanGlScott): Replying to [comment:14 Iceland_jack]:
I was wondering if there is anything GNDerivable in theory that would be outside of GHC's reach even given quantified constraints (and your hunch).. Just as `join` is 'derivable' but is not derivable currently.
In theory, //everything// is GNDable - after all, `GeneralizedNewtypeDeriving` is just a glorified `unsafeCoerce` hack. The real question is if something is GNDable //and// type-safe, but beyond GHC's reach. The only cases that I'm aware of where this happens involves roles, so if you find other scenarios where `GeneralizedNewtypeDeriving` emits code that doesn't typecheck, file another ticket. -- Ticket URL: http://ghc.haskell.org/trac/ghc/ticket/14070#comment:15 GHC http://www.haskell.org/ghc/ The Glasgow Haskell Compiler