(CC: the biohaskell mailing list, in case somebody else is interested) Sönke Hahn <s...@zerobuzz.net> writes:
>> Of course...we're trying to deprecate [the old biolib] in favor of a >> set of smaller packages. :-) > That's interesting. Are they on hackage already? They should be, but I'm not sure how current or updated those versions are. I tend to scratch my own itches first, please report any inconsistencies! See also: http://biohaskell.org/Libraries If there's specific functionality not yet covered, feel free to prod me, or just factor it out from the old biolib. > http://open-projects.net/~shahn/darcs/biolib/ Great - pulling it now. > I just tested compilation with ghc-7.4.2. > > One of the patches changes the cabal constraints for binary. ghc-7.6.2 > ships with binary-0.5.1.1 so it would be good to allow at least 0.5.*. > My patch also allows 0.6.* and 0.7.*, which we already have on hackage. > I tested compilation with binary-0.6.4.0 and binary-0.7.0.1 on > ghc-7.6.2. So I guess allowing these newer binary versions is ok. About that: there is a small problem: strictness of binary changed in 0.5. This means that for SFF files (454 sequences), the old code pulls in the whole file (which can be gigabytes) in one go. If you use biosff, the code there has been rewritten to work lazily with newer versions of binary, but it is unfortunately quite a bit slower. -k -- If I haven't seen further, it is by standing in the footprints of giants _______________________________________________ Biohaskell mailing list Biohaskell@biohaskell.org http://malde.org/cgi-bin/mailman/listinfo/biohaskell
