Christian Hoener zu Siederdissen <choe...@tbi.univie.ac.at> writes: >> Anyone opposed?
> If seqid returns an (unspecified) substring of seqheader , it sounds ok. > Unspecified as it may not always be something like the 1st word. IMO, it should return an identifier without whitespace, and ideally unique (but I don't think there's any way to enforce that). The header can return any information content, but typically starts with seqid + wsp. > What about a Functor instance working on the payload? Assuming that one > "copy"s the seqheader? In general I am not a fan of hiding > implementation details, I tend to find odd uses. I'm not really after hiding the details, merely making it more convenient to use (i.e. cleaning up the import lists). I'm not sure how to use Functor, SeqData etc aren't parametric types. We can supply a separate set of maps and folds etc, though, similar to Bytestring. If so, I'd like to put them in a separate module, I don't much care for having to qualify everything because it collides with the Prelude. -k _______________________________________________ Biohaskell mailing list Biohaskell@biohaskell.org http://malde.org/cgi-bin/mailman/listinfo/biohaskell
