#15681: Take {-# COMPLETE #-} pragma into consideration when using MonadFailDesugaring -------------------------------------+------------------------------------- Reporter: chshersh | Owner: (none) Type: bug | Status: new Priority: normal | Milestone: 8.6.1 Component: Compiler | Version: 8.6.1 Resolution: | Keywords: pattern- | matching,monadfail,desugaring 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 simonpj):
That's because it's determining if a pattern synonym is irrefutable by consulting the plain old data constructors that correspond to the type constructor that heads its return type
Yes, that is pretty dodgy! Eg I think it would give the wrong answer for {{{ pattern P a = (Just a, True) }}} The outer constructor is `(,)`, but that doesn't mean that `P x` is irrefutable. Naively, one might think that the simple thing do to is to behave as if the type synonym was expanded at the use site. But that breaks the abstraction that is part of the purpose of having a pattern synonym. And in fact, in separate compilation, GHC does not record the original definition directly; it just exports teh builder and matcher functions for the pattern synonym. The right thing must surely be to use the `COMPLETE` sets, somehow. -- Ticket URL: http://ghc.haskell.org/trac/ghc/ticket/15681#comment:2 GHC http://www.haskell.org/ghc/ The Glasgow Haskell Compiler