[Haskell-cafe] data vs newtype (was: OOP'er with (hopefully) trivial questions)

[Jonathan Cast]

On 17 Dec 2007, at 7:39 AM, David Menendez wrote:

> On Dec 17, 2007 8:51 AM, Bayley, Alistair  
> <Alistair_Bayley at invescoperpetual.co.uk> wrote:
> As an aside, I was wondering exactly what the differences are between
> newtype and data i.e. between
>
> > newtype A a = A a
>
> and
>
> > data A a = A a
>
> According to:
>   http://www.haskell.org/onlinereport/decls.html#sect4.2.3
> newtype is, umm, stricter than data i.e. newtype A undefined =
> undefined, but data A undefined = A undefined. Other than that,  
> newtype
> just seems to be an optimization hint. Is that a more-or-less correct
> interpretation?
>
> Pretty much. Newtype is nice to have, but I don't think there's any  
> program you can write that couldn't be rewritten to use data (with  
> a possible loss of efficiency).
>
> The difference that surprised me is the difference between
>
>     newtype A a = A a
>
> and
>
>     data A a = A !a
>
> If we define a function like this,
>
>     seqA (A a) = ()
>
> Under the first definition of A,
>
>     seqA undefined = ()
>
> Under the second,
>
>    seqA undefined = undefined
>
> The difference is that pattern-matching a newtype doesn't do any  
> evaluation.

So there is a program (or, rather, type) you can write with newtype  
that can't be written with data:

newtype T = T T

jcc

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.haskell.org/pipermail/haskell-cafe/attachments/20071217/aec238fe/attachment.htm