[GHC] #15885: Enhancing COMPLETE pragma to support pattern synonyms with polymorphic (output) types

GHC ghc-devs at haskell.org
Sat Nov 10 01:17:46 UTC 2018


#15885: Enhancing COMPLETE pragma to support pattern synonyms with polymorphic
(output) types
-------------------------------------+-------------------------------------
        Reporter:  Shayan-Najd       |                Owner:  (none)
            Type:  task              |               Status:  new
        Priority:  normal            |            Milestone:  8.6.3
       Component:  Compiler          |              Version:  8.6.2
      Resolution:                    |             Keywords:
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 Shayan-Najd):

 It is worth mentioning that the combination of the ordered set of pattern
 synonym names in the pragma and the type of the pattern synonyms
 themselves often guarantees abstraction (currently, I don't know of any
 situation that does not):
 the pattern matching totality checker does not need to know the
 implementation details of pattern synonyms (e.g., such details can be
 omitted from the interface files).
 I noticed the required type constraints guarantee that a pattern synonym
 mentioned in a COMPLETE ordered set is always used in a setting where,
 well, the required constraints hold. This fact immediately implies that,
 for example in `{-# COMPLETE LL :: HasSrcSpan a => a #-}`, `a` always
 satisfies `HasSrcSpan`, hence checking the completeness of a matching on
 `LL` does not need to consider types at all; all types match `a` and
 `HasSrcSpan a` is already guaranteed to hold by the pattern synonym type
 checking.
 I might be wrong about this observation though.

-- 
Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/15885#comment:1>
GHC <http://www.haskell.org/ghc/>
The Glasgow Haskell Compiler


More information about the ghc-tickets mailing list