#10653: PatternSynonyms should be imported/exported as part of the wildcard notation -------------------------------------+------------------------------------- Reporter: gridaphobe | Owner: Type: feature request | Status: new Priority: normal | Milestone: 7.12.1 Component: Compiler | Version: 7.11 Resolution: | Keywords: pattern | synonyms Operating System: Unknown/Multiple | Architecture: | Unknown/Multiple Type of failure: None/Unknown | Test Case: Blocked By: | Blocking: Related Tickets: | Differential Revisions: -------------------------------------+------------------------------------- Comment (by simonpj): Yes, you do have to import-chase to figure out which ''subset'' of data constructors are imported by `T(..)`. But there's an upper bound: it can import no more than all the constructors. With the new proposal, ''any'' old new pattern synonym could be brought into scope by `T(..)`. That is new. Maybe it's not terrible, but it's new. Maybe one would want to associate more things with `T`? Such as a family of functions over `T`, whether or not they are pattern synonyms? Anyway I don't feel terribly strongly about all this ''provided'' it can all be resolved by the renamer (ie not involving type inference). By all means write a wiki page, seek feedback etc. -- Ticket URL: http://ghc.haskell.org/trac/ghc/ticket/10653#comment:15 GHC http://www.haskell.org/ghc/ The Glasgow Haskell Compiler