Re: [Haskell-cafe] Haskell's type inference considered harmful [Re: [Haskell] A riddle...]

You will be warned about the top-level definitions not including a type-signature if you use the -Wall flag. This isn't really a complete solution to your gripes, but it does address the change in behaviour that you saw when adding/removing the commented code, and would draw your attention to the logical error of trying to squeeze numbers that were too large into a Word16. I've been caught by unwarned truncation of numeric literals before too, so it would be great if there were warnings for this. On Tue, Jul 17, 2012 at 3:40 PM, Christopher Done wrote:

On 17 July 2012 09:27, Andreas Abel wrote:

...

1. Haskell's type inference is NON-COMPOSITIONAL!

In the riddle below, I am defining two things f ("rgbliste") and g ("farbliste"). Even though they are not strongly connected, but g comes after f in the definition order, the code of g influences the type of f. THAT'S WRONG! :-(

Bindings at the same level in Haskell are mutually recursive. Order of declaration does not matter. These two terms are unified by the type system. So I'm not sure what you expect to happen here.

_______________________________________________ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe