Should exhaustiveness testing be on by default?

[Peter Hercek]

Claus Reinke wrote:
> Given how long http://hackage.haskell.org/trac/ghc/wiki/ExplicitCallStack
> has been under discussion, it is probably time to provide a short-term
> workaround in GHC, just a token to be replaced by the current source
> location.

This would be the best solution.

Although -fno-ignore-asserts is acceptable till I do not need asserts for
 what they are actually supposed to be used for.

The second solution requires QuasiQuotes, so I do not know. If I would want
 to compile with a different compiler it would break. If srcloc can be 
defined
 as a simple token (not requiring special extensions at places where it 
is used)
 then I could define it to an empty string in some low level module if 
trying to
 compile with a different haskell compiler which does not know srcloc.

Thanks for the tips,
Peter.