Re: [Haskell-cafe] Reading open data types

2007-06-13 Thread Isaac Dupree
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Benja Fallenstein wrote: Hi all, We've had a discussion on #haskell about how we can make a function that reads in serialized values of an open data type, such as [...] However, this is still kind of boring. Is there a better way? If not,

Re: [Haskell-cafe] Reading open data types

2007-06-13 Thread Benja Fallenstein
Hi Isaac, 2007/6/13, Isaac Dupree [EMAIL PROTECTED]: Since Show instances can overlap (e.g. (show (1::Int)) == (show (1::Integer))), we need to tag with the type. Indeed. But that's the easy part :-) Reminds me of Typeable. Since GHC lets us derive Typeable with a guarantee of different

Re: [Haskell-cafe] Reading open data types

2007-06-13 Thread Bulat Ziganshin
Hello Benja, Wednesday, June 13, 2007, 6:12:25 PM, you wrote: We've had a discussion on #haskell about how we can make a function that reads in serialized values of an open data type, such as look at Data.Generics.Text which may be implements exactly what you need -- Best regards, Bulat

Re: [Haskell-cafe] Reading open data types

2007-06-13 Thread Benja Fallenstein
Hi Bulat, 2007/6/13, Bulat Ziganshin [EMAIL PROTECTED]: We've had a discussion on #haskell about how we can make a function that reads in serialized values of an open data type, such as look at Data.Generics.Text which may be implements exactly what you need Unfortunately not.

Re: [Haskell-cafe] Reading open data types

2007-06-13 Thread Stefan O'Rear
On Wed, Jun 13, 2007 at 05:12:25PM +0300, Benja Fallenstein wrote: However, this is still kind of boring. Is there a better way? If not, would it be a good idea to have compiler support for building this kind of type table? The compiler does build exactly such a table - it's called a symbol

Re: [Haskell-cafe] Reading open data types

2007-06-13 Thread Benja Fallenstein
2007/6/14, Stefan O'Rear [EMAIL PROTECTED]: On Wed, Jun 13, 2007 at 05:12:25PM +0300, Benja Fallenstein wrote: However, this is still kind of boring. Is there a better way? If not, would it be a good idea to have compiler support for building this kind of type table? The compiler does build