[GHC] #11390: GHC does not warn about redundant patterns

[GHC]

#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,
                                     |  PatternMatchWarnings
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 gkaracha):

 Actually nothing is redundant in the above examples. As we discuss in the
 paper (section about
 laziness), `(sillyId undefined)` will give you the right hand-side so the
 clause:
 {{{#!hs
 sillyId x = x
 }}}
 is not redundant at all. Of course, this is not the case for:
 {{{#!hs
 sillyId2 :: F1 Char -> F1 Char
 sillyId2 (!x) = x
 }}}
 because it is strict in the argument, so no WHNF can have the type `F1
 Char`. But even in this
 case, the clause is not actually redundant, it just has an inaccessible
 RHS (remember this is
 not Agda). Pattern matching is not just used for discrimination in Haskell
 but also for evaluation,
 which I see as a side effect. Hence, the clause may have an inaccessible
 RHS, but by removing it
 the expression is not evaluated and you change the semantics of
 `sillyId2`.

--
Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/11390#comment:10>
GHC <http://www.haskell.org/ghc/>
The Glasgow Haskell Compiler