|
Re: ReadS with Maybe: msg#00038lang.haskell.cafe
On Mon, Jan 03, 2005 at 08:48:26PM +0100, Henning Thielemann wrote: > > What is the reason for the definition > > ReadS a = [(a, String)] > > not being > > ReadS a = Maybe (a, String) > > ? The latter one reflects that either one or no value is read, whereas the > first definition allows an arbitrary number of read values which is > confusing and unsafe in my opinion. I always wondered why we have readIO :: Read a => String -> IO a instead of the absurdly more useful and strictly superior readM :: (Monad m,Read a) => String -> m a then we can just readM as a list, a maybe, in IO, for pattern matching in arbitrary monads... sigh. I think I am just a real fan of type classes when it comes to code reuse :) John -- John Meacham - ⑆repetae.net⑆john⑈ |
|
| <Prev in Thread] | Current Thread | [Next in Thread> |
|---|---|---|
| Previous by Date: | Re: Re: Typed Lambda-Expressions withOUT GADTs: 00038, Conor McBride |
|---|---|
| Next by Date: | RE: GHC for .NET?: 00038, Don Syme |
| Previous by Thread: | Re: ReadS with Maybei: 00038, Robert Dockins |
| Next by Thread: | Re: ReadS with Maybe: 00038, David Roundy |
| Indexes: | [Date] [Thread] [Top] [All Lists] |
| News | FAQ | advertise |