:i doesn't work for ~

Prelude> :set -XTypeOperators
Prelude> :i (~)

<interactive>:1:2: parse error on input ‘~’
Prelude> :k (~)
(~) :: k -> k -> Constraint

Trac metadata

Trac field	Value
Version	7.11
Type	Bug
TypeOfFailure	OtherFailure
Priority	normal
Resolution	Unresolved
Component	GHCi
Test case
Differential revisions
BlockedBy
Related
Blocking
CC	hvr
Operating system
Architecture

mentioned in issue #9194 (closed)

mentioned in issue #10056 (closed)

changed weight to 5

added GHCi Tbug Trac import labels

See ticket:10056#comment:95839. The last point there explains, essentially, why this is happening. Both #10056 (closed) and this ticket have the same resolution.

I'm not quite labeling this a duplicate of #10056 (closed), because this is worth a separate regression test.

Trac metadata

Trac field	Value
Related	→ #10056 (closed)

mentioned in issue #12023 (closed)

mentioned in issue #14316 (closed)

Trac metadata

Trac field	Value
Related	#10056 (closed) → #10056 (closed), #12023 (closed)

After some digging, it turns out that the reason this doesn't work is because the identifier parser production doesn't catch (~). It turns out that this patch:

diff --git a/compiler/parser/Parser.y b/compiler/parser/Parser.y
index c6face8..06a5722 100644
--- a/compiler/parser/Parser.y
+++ b/compiler/parser/Parser.y
@@ -629,6 +629,8 @@ identifier :: { Located RdrName }
         | qconop                        { $1 }
     | '(' '->' ')'      {% ams (sLL $1 $> $ getRdrName funTyCon)
                                [mj AnnOpenP $1,mu AnnRarrow $2,mj AnnCloseP $3] }
+    | '(' '~' ')'       {% ams (sLL $1 $> $ eqTyCon_RDR)
+                               [mop $1,mj AnnTilde $2,mcp $3] }
 
 -----------------------------------------------------------------------------
 -- Backpack stuff

Is enough to make :info (~) go through:

λ> :info (~)
class (a ~ b) => (~) (a :: k) (b :: k)
        -- Defined in ‘Data.Type.Equality’
instance [incoherent] forall k (a :: k) (b :: k). (a ~ b) => a ~ b
  -- Defined in ‘Data.Type.Equality’

(The pretty-printing of the class definition of (~) is gnarly, though—I'm still figuring out how to work around that.)

Then again, it turns out that the :info output for (~~) is similarly broken:

GHCi, version 8.4.3: http://www.haskell.org/ghc/  :? for help
Loaded GHCi configuration from /home/rgscott/.ghci
λ> import Data.Type.Equality
λ> :i (~~)
class ((a :: k0) ~~ (b :: k1)) => (~~) (a :: k0) (b :: k1)
        -- Defined in ‘GHC.Types’

Both :infos look correct to me. Turn on -fprint-equality-relations to get better output. I don't know how to do better, given -fprint-equality-relations.

In that case, let's do this!

Trac metadata

Trac field	Value
Differential revisions	- → D4877

mentioned in commit f4dce6cf

changed milestone to %8.6.1

This could be merged to GHC 8.6 if desired.

Trac metadata

Trac field	Value
Test case	- → ghci/scripts/T10059

added 1 deleted label

closed

This is in 8.6.

Trac metadata

Trac field	Value
Resolution	Unresolved → ResolvedFixed

removed 1 deleted label

added Pnormal label