[GHC] #4108: GHC.Integer.hashInteger is a misnomer and confuses people

[GHC]

#4108: GHC.Integer.hashInteger is a misnomer and confuses people
---------------------------+------------------------------------------------
  Reporter:  duncan        |          Owner:                  
      Type:  bug           |         Status:  new             
  Priority:  normal        |      Milestone:                  
 Component:  Compiler      |        Version:  6.12.2          
Resolution:                |       Keywords:                  
Difficulty:                |             Os:  Unknown/Multiple
  Testcase:                |   Architecture:  Unknown/Multiple
   Failure:  None/Unknown  |  
---------------------------+------------------------------------------------
Changes (by simonmar):

  * status:  closed => new
  * resolution:  invalid =>


Comment:

 Replying to [comment:3 duncan]:
 > The new `berp` package uses it (but will stop using it in the next
 release). The author did not realise that `GHC.Integer.hashInteger` is
 basically the identity function and would not have used it if they had
 known.

 Why's that, just out of interest?

 > By comparison, `Data.HashTable.hashInt` is not an identity. I defer to
 Eric that perhaps the existing `hashInteger` could be usable as a hash
 function in some circumstances. It does however seem a bit special-purpose
 to use that name, and it could be named something less misleading.

 It's probably not a great hash function, in the same way that (`mod`
 power-of-2) is not a great hash function either (disclaimer: I'm not an
 expert on hash functions), because it's too easy to construct a collection
 of values that have the same hash.  So at least we ought to add something
 to the documentation to say what the function does.

-- 
Ticket URL: <http://hackage.haskell.org/trac/ghc/ticket/4108#comment:4>
GHC <http://www.haskell.org/ghc/>
The Glasgow Haskell Compiler