Communication with the no-functional world.

Simon L Peyton Jones simonpj@dcs.gla.ac.uk
Wed, 18 Mar 98 08:57:13 +0000


> I have another wish for Clean, and I would like to know how the Haskell=

> community addresses it. Two or three months ago, I needed to

=2E..

> Since there is a call to "size", and the error handler was showing a
> position very close to it, I was convinced that the error was there. Bu=
t
> no!  The error is in the expression [toHex x \\ x <- (s%(i, i+n-1))]. O=
f
> course, I am happy that functional languages find out my mistakes at
> compile  time, not at run time, as C or Java.  However, it would be muc=
h
> better if  functional languages could be as good to show where the erro=
r
> is, as they are good to choose when to tell me that I made a blunder.

This is a compiler issue not a language issue.
I do not know whether GHC would give a better error message
than Clean.  Quite possibly not; but the way to find out
is to try it.  You just need to convert your program to Haskell
and have a go.

Incidentally, GHC will give much much better error messages if
you give type signatures to your functions.

Simon =