[commit: ghc] master: Comments only (c64926f)

[Simon Peyton Jones]

Repository : ssh://darcs.haskell.org//srv/darcs/ghc

On branch  : master

http://hackage.haskell.org/trac/ghc/changeset/c64926f3f9c7e8b6abc67b35b548143107f51958

>---------------------------------------------------------------

commit c64926f3f9c7e8b6abc67b35b548143107f51958
Author: Simon Peyton Jones <simonpj at microsoft.com>
Date:   Mon Oct 24 15:39:44 2011 +0100

    Comments only

>---------------------------------------------------------------

 compiler/typecheck/TcTyDecls.lhs |    4 ++--
 1 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/compiler/typecheck/TcTyDecls.lhs b/compiler/typecheck/TcTyDecls.lhs
index d4800ac..a3dd2b8 100644
--- a/compiler/typecheck/TcTyDecls.lhs
+++ b/compiler/typecheck/TcTyDecls.lhs
@@ -166,8 +166,8 @@ Where expand is defined as follows:
 
 Eqn (1) is conservative; when there's a type variable at the head,
 look in all the argument types.  Eqn (2) expands superclasses; the
-third component of the union is like Eqn (1).  Eqn (3) really never
-happens.
+third component of the union is like Eqn (1).  Eqn (3) happens mainly
+when the context is a (constraint) tuple, such as (Eq a, Show a).
 
 Furthermore, expand always looks through type synonyms.