[Haskell-cafe] There is no null; Maybe/Option types

[Evan Laforge]

On Sat, Apr 23, 2011 at 5:57 PM, Ketil Malde <ketil at malde.org> wrote:
> Evan Laforge <qdunkan at gmail.com> writes:
>
>>> Most of the (non-IO) runtime errors I get using Haskell software is
>>> due to head or fromJust, it's actually quite annoying.
>
>> Just singling this one out because I've heard it before.  I've never
>> gotten a runtime error from these.  It seems quite easy to just not
>> use these functions.
>
> Unfortunately, it is quite easy to just use them, too :-)  Sometimes I
> just *know* that this list will never be empty, or that that Maybe will
> never be nothing - but often I'm wrong.

That's the thing I was commenting on... I do a lot of things wrong,
but maybe somewhere I'm doing something right, because for whatever
reason this never comes up for me.  The only times I can think of are
'maximum' that I know will complete because I wrote 'maximum (0 :
stuff)' and "split" functions, which are guaranteed to return at least
one element.  As Henning points out, (a, [a]) is an explicit type for
those, but usually I just trust the postcondition of the function and
match right away on its output.

I have lots of 'maybe [] id' or whatever, but it's always cases where
I expect a Nothing so I have an escape route planned.