[Haskell] "Classic" vs. "Cunning Newtype" Derivation
simonpj at microsoft.com
Tue Mar 8 03:47:52 EST 2005
[redirecting to GHC users list]
John is right. I've added something to the user manual to say so.
| -----Original Message-----
| From: haskell-bounces at haskell.org [mailto:haskell-bounces at haskell.org] On Behalf Of John Meacham
| Sent: 08 March 2005 00:20
| To: haskell at haskell.org
| Subject: Re: [Haskell] "Classic" vs. "Cunning Newtype" Derivation
| On Mon, Mar 07, 2005 at 04:06:44PM -0800, Ashley Yakeley wrote:
| > I make quite a lot of use of "cunning newtype" deriviation (in GHC),
| > because it's easy to understand.
| > newtype Mytype = MkMytype T deriving C
| > However, if I use one of the standard derivable classes (Eq, Ord, Enum,
| > Bounded, Show, and Read), I believe that form of deriving takes priority.
| > AFAICT from the H98 Report ch. 10, for Eq, Ord and Bounded this works
| > out as the same thing, Enum doesn't apply, and Read and Show do
| > something different. Is this correct, or are there subtle differences?
| Read, Show, Typeable, and Data are the only differences AFAIK.
| John Meacham - ⑆repetae.net⑆john⑈
| Haskell mailing list
| Haskell at haskell.org
More information about the Glasgow-haskell-users