[Git][ghc/ghc][wip/int-index/t2t-expr] Comments for IsPunnedVarOcc

Vladislav Zavialov (@int-index) gitlab at gitlab.haskell.org
Mon Oct 2 15:38:06 UTC 2023



Vladislav Zavialov pushed to branch wip/int-index/t2t-expr at Glasgow Haskell Compiler / GHC


Commits:
96bbea26 by Vladislav Zavialov at 2023-10-02T18:34:15+03:00
Comments for IsPunnedVarOcc

- - - - -


1 changed file:

- compiler/GHC/Hs/Expr.hs


Changes:

=====================================
compiler/GHC/Hs/Expr.hs
=====================================
@@ -222,9 +222,16 @@ type instance XOverLabel     GhcTc = DataConCantHappen
 
 -- ---------------------------------------------------------------------
 
+-- | Is this variable occurrence punned, i.e. is there a variable with the same
+-- 'occNameFS' but a different 'occNameSpace' in the context?
+--
+-- This is used to reject punned variable occurrences when converting required
+-- type arguments from HsExpr to HsType (see the "T2T-Mapping" section of GHC Proposal #281).
 data IsPunnedVarOcc =
     DistinctVarOcc
-  | PunnedVarOcc Name Name
+  | PunnedVarOcc
+      Name  -- how the variable was actually resolved
+      Name  -- how it could have been resolved if we were to look in a different namespace
   deriving (Eq, Data)
 
 type instance XVar GhcPs = NoExtField



View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/commit/96bbea260945576e201a4931e79ec7db755cc821

-- 
View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/commit/96bbea260945576e201a4931e79ec7db755cc821
You're receiving this email because of your account on gitlab.haskell.org.


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.haskell.org/pipermail/ghc-commits/attachments/20231002/d16e5e31/attachment.html>


More information about the ghc-commits mailing list