#11390: GHC does not warn about redundant patterns -------------------------------------+------------------------------------- Reporter: Iceland_jack | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 8.1 Resolution: | Keywords: warnings 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 rwbarton): I would say that `case x of {}` is just a case expression with no alternatives. It is, itself, still an ordinary expression. So both versions look like `sillyId x = <expr>` and should be equivalent as far as redundancy of the ''outer'' pattern match, the one done by `sillyId`, is concerned. Since you cannot write a definition with 0 equations, this "redundancy" is unavoidable. (Unless you use LambdaCase and EmptyCase, I guess.) Are you suggesting treating `case x of {}` as some kind of special form? Also, I'm not sure a "stuck" type family application is really empty in the same sense as an empty data type. But #10746 makes it hard to tell what GHC thinks about this. -- Ticket URL: http://ghc.haskell.org/trac/ghc/ticket/11390#comment:5 GHC http://www.haskell.org/ghc/ The Glasgow Haskell Compiler