On Sat, Dec 20, 2008 at 9:10 PM, Jeff Wheeler <j...@nokrev.com> wrote:
> On Sat, 2008-12-20 at 21:00 +0100, nicolas.pouillard wrote:
>> Excerpts from Gwern Branwen's message of Sat Dec 20 20:54:11 +0100 2008:
>> > -----BEGIN PGP SIGNED MESSAGE-----
>> > Hash: SHA512
>> >
>> > I doubt everyone has installed or even heard of hlint here. hlint is
>> > Neil Mitchell's latest project, which was released today. It is as it
>> > sounds like: it finds suboptimal or wasteful Haskell code and suggests
>> > cleaner ways of writing it.
>>
>> Of course I've heard of it :) That's a great tool!
>
> Indeed. I saw discussion of it on haskell-cafe (I think), and it seems
> very useful. Anybody mind if I collect a bunch of these and commit them
> together? I'd very much like to see us minimize the number of
> suggestions.
>
I am a bit unsure about some things:
1. How confident are we that the suggestions are semantics-preserving?
I'd recommend thinking twice before applying non-trivial things.
2. I find conversion from point-wise to point-free... pointless :)
Indeed, as I have said before this can be detrimental to code clarity:
the name of the point can add useful documentation, etc.
3. I could not take a look at the report (somehow gmail screws it up),
so I have no idea (yet) what hlint proposed.
I think I'd be quite in favour of running this and applying the suggestions, as
long as it could be configured. (In particular I'd drop the systematic
transformation to point-free)
Cheers,
JP.
--~--~---------~--~----~------------~-------~--~----~
Yi development mailing list
yi-devel@googlegroups.com
http://groups.google.com/group/yi-devel
-~----------~----~----~----~------~----~------~--~---
