Re: [GHC] #11122: Ambiguous inferred type causes a panic

23 Nov 2015

      #11122: Ambiguous inferred type causes a panic
-------------------------------------+-------------------------------------
        Reporter:  tuplanolla        |                Owner:
            Type:  bug               |               Status:  new
        Priority:  normal            |            Milestone:
       Component:  Compiler (Type    |              Version:  7.10.2
  checker)                           |
      Resolution:                    |             Keywords:
Operating System:  Linux             |         Architecture:
                                     |  Unknown/Multiple
 Type of failure:  None/Unknown      |            Test Case:
      Blocked By:                    |             Blocking:
 Related Tickets:                    |  Differential Rev(s):
       Wiki Page:                    |
-------------------------------------+-------------------------------------
Description changed by tuplanolla:

Old description:
...
Consider the following program.
{{{#!hs
{-# LANGUAGE PartialTypeSignatures #-}
module Main where
import Text.Parsec
import Text.Parsec.String
parser :: Parser _
parser = read <$> many digit
data Wrapper = Wrapper Int
wrapperParser = Wrapper <$> parser
}}}
I am not sure whether it is valid or not, but it breaks the type checker
as follows.
{{{#!hs
Program.hs:9:1:
    No instance for (Read w_)
    When checking that ‘parser’ has the specified type
      parser :: forall w_. Parser w_
    Probable cause: the inferred type is ambiguous
Program.hs:13:29:
    Couldn't match type ‘w_’ with ‘Int’
      ‘w_’ is untouchable
        inside the constraints ()
        bound by the inferred type of
                 wrapperParser :: ParsecT
                                    String ()
Data.Functor.Identity.Identity Wrapper
        at Program.hs:13:1-34ghc: panic! (the 'impossible' happened)
  (GHC version 7.10.2 for x86_64-unknown-linux):
        No skolem info: w__avnz[sk]
}}}
Disabling `PartialTypeSignatures` makes no difference.
Switching from `parsec-3.1.9` to `megaparsec-4.1.1` and changing `many
digit` to `some digitChar` does not have an effect either.
Replacing `_` with `Int` seems to be the only obvious way to make the
problem disappear.
Alas that defeats the point of using partial type signatures or typed
holes in the first place.
New description:

 Consider the following program.

 {{{#!hs
 {-# LANGUAGE NoMonomorphismRestriction, PartialTypeSignatures #-}

 module Main where

 import Text.Parsec
 import Text.Parsec.String

 parser :: Parser _
 parser = read <$> many digit

 data Wrapper = Wrapper Int deriving Show

 wrapperParser = Wrapper <$> parser

 main :: IO ()
 main = parseTest wrapperParser "0"
 }}}

 I am not sure whether it is valid or not, but
 it breaks the type checker as follows.

 {{{#!hs
 Main.hs:9:1:
     No instance for (Read w_)
     When checking that ‘parser’ has the specified type
       parser :: forall w_. Parser w_
     Probable cause: the inferred type is ambiguous

 Main.hs:13:29:
     Couldn't match type ‘w_’ with ‘Int’
       ‘w_’ is untouchable
         inside the constraints ()
         bound by the inferred type of
                  wrapperParser :: ParsecT
                                     String ()
 Data.Functor.Identity.Identity Wrapper
         at Main.hs:13:1-34ghc: panic! (the 'impossible' happened)
   (GHC version 7.10.2 for x86_64-unknown-linux):
         No skolem info: w__a2BN[sk]
 }}}

 Disabling `PartialTypeSignatures` makes no difference.
 Switching from `parsec` to `megaparsec` and
 changing `many digit` to `some digitChar` does not have an effect either.
 Removing `NoMonomorphismRestriction` makes the problem disappear,
 as does changing `_` to `Int`.

 The libraries used were

 * `array-0.5.1.0`,
 * `base-4.8.1.0`,
 * `binary-0.7.5.0`,
 * `bytestring-0.10.6.0`,
 * `containers-0.5.6.2`,
 * `deepseq-1.4.1.1`,
 * `ghc-prim-0.4.0.0`,
 * `integer-gmp-1.0.0.0`,
 * `megaparsec-4.1.1`,
 * `mtl-2.2.1`,
 * `parsec-3.1.9`,
 * `text-1.2.1.3` and
 * `transformers-0.4.2.0`.

--

--
Ticket URL: http://ghc.haskell.org/trac/ghc/ticket/11122#comment:1
GHC http://www.haskell.org/ghc/
The Glasgow Haskell Compiler