Okay, I've pushed biocore 0.2.1 to my darcs repo, now with (an exported) Stringable instance. Minus the "length" function, I don't like it when things must be qualified to avoid clashing with Prelude - and at least, I don't want to introduce things that break previously working compiles.
http://malde.org/~ketil/biohaskell/biocore - please check and see if/how it works! Christian Höner zu Siederdissen <choe...@tbi.univie.ac.at> writes: > how about making everything 'ListLike'? Yes - at least, we should provide this kind of functionality. It's a bit involved though, so I'd like to be more sure it's the right thing. > Btw since we talk about this > stuff. I'm thinking about sequences of type: > > newtype S t s = S (VU.Vector s) > > or maybe even associated data types. I want to tag ('t') things like > what sequence we have (dna/rna/protein/any) and if are indexing > 0-/1-based. Though for indexing I currently prefer importing to an > internal common format with one lazy and one strict implemention. The old Bio.Sequence used to do that, but I notice Dan removed the phantom type in biofasta. Since Bio.Core is fairly agnostic about the specific data, you can easily do this and instantiate BioSeq, I think. -k _______________________________________________ Biohaskell mailing list Biohaskell@biohaskell.org http://malde.org/cgi-bin/mailman/listinfo/biohaskell
