[GHC] #14771: TypeError reported too eagerly

[GHC]

#14771: TypeError reported too eagerly
-------------------------------------+-------------------------------------
        Reporter:  hsyl20            |                Owner:  (none)
            Type:  bug               |               Status:  new
        Priority:  normal            |            Milestone:
       Component:  Compiler (Type    |              Version:  8.2.2
  checker)                           |             Keywords:
      Resolution:                    |  CustomTypeErrors
Operating System:  Unknown/Multiple  |         Architecture:
 Type of failure:  Incorrect         |  Unknown/Multiple
  error/warning at compile-time      |            Test Case:
      Blocked By:                    |             Blocking:
 Related Tickets:                    |  Differential Rev(s):
       Wiki Page:                    |
-------------------------------------+-------------------------------------

Comment (by diatchki):

 For issue 1:

 the high-level question is what to do when we encounter an unevaluated
 type function with an error parameter (e.g., `T x (TypeError ...)`).
 Currently, we always report such errors, which is simple, but probably too
 eager in some cases.

 I guess a better solution, although more complex, would be too look at the
 available equations for `T` and see if evaluation is "stuck" due to the
 type error.  If so, we report the custom error.  Otherwise, we leave the
 expression as is.  Note, however, that if we choose to not report the
 custom error we'd get a rather ugly "normal" error, as GHC would still
 print the description of the error (in combinator form) when it is
 rendering the type.


 For issue 2:
 I am not sure I understand 100% what the request is.  Could you give a
 concrete example?

-- 
Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/14771#comment:3>
GHC <http://www.haskell.org/ghc/>
The Glasgow Haskell Compiler