biofasta 0.0.2 → 0.0.3

raw patch · 3 files changed

+36/−35 lines, 3 filesdep ~biocorePVP ok

version bump matches the API change (PVP)

Dependency ranges changed: biocore

API changes (from Hackage documentation)

Files

• biofasta.cabal +4/−3
• src/Bio/Sequence/Fasta.hs +16/−32
• src/Test.hs +16/−0

@@ -1,19 +1,20 @@ Name:                biofasta-Version:             0.0.2+Version:             0.0.3 Synopsis:            Library for reading fasta sequence files Description:         Library for reading fasta sequence files Homepage:	     https://patch-tag.com/r/dfornika/biofasta/home License:             GPL License-file:        LICENSE Cabal-Version:       >=1.6-Author:              Dan Fornika <dfornika@gmail.com>+Author:              Ketil Malde <ketil@malde.org> Maintainer:          dfornika@gmail.com Stability:	     Provisional Category:	     Bioinformatics Build-Type:          Simple+Extra-Source-Files:  src/Test.hs  Library-  Build-depends:     base >= 2 && < 5, biocore, bytestring+  Build-depends:     base >= 2 && < 5, biocore >= 0.3.1, bytestring   Exposed-modules:   Bio.Sequence.Fasta   Hs-source-dirs:    src 

@@ -4,11 +4,7 @@    This module incorporates functionality for reading and writing    sequence data in the Fasta format.    Each sequence consists of a header (with a '>' prefix)-   and a set of lines containing the sequence data.--   As Fasta is used for both amino acids and nucleotides, the-   resulting 'Sequence's are type-tagged with 'Unknown'.  If you know the -   type of sequence you are reading, use 'castToAmino' or 'castToNuc'.+   and a set of lines containing the sequence data.. -}  module Bio.Sequence.Fasta@@ -19,6 +15,8 @@     , countSeqs     -- * Helper function for reading your own sequences     , mkSeqs+    -- * Other+    , toStr, seqid, seqheader, seqdata, seqlength ) where  @@ -36,32 +34,15 @@ splitsAt n s = let (s1,s2) = B.splitAt (unOff n) s                in if B.null s2 then [s1] else s1 : splitsAt n s2 -{---- ugly?-class SeqType sd where-    toSeq :: sd -> sd -> Sequence-    fromSeq :: Sequence -> (sd,sd)--instance SeqType B.ByteString where-    toSeq = Seq-    fromSeq (Seq x y) = (x,y)--instance SeqType BS.ByteString where-    toSeq h s = Seq (B.fromChunks [h]) (B.fromChunks [s])-    fromSeq (Seq x y) = (c x, c y) where c = BS.concat . B.toChunks--}- data Sequence = Seq SeqLabel SeqData (Maybe QualData)-                deriving Eq+                deriving (Show, Eq)  instance BioSeq Sequence where-  seqlabel  (Seq lab seq mqual) = lab+  seqid     (Seq lab seq mqual) = SeqLabel {unSL = B.takeWhile (/= ' ') $ unSL lab}+  seqheader (Seq lab seq mqual) = lab   seqdata   (Seq lab seq mqual) = seq   seqlength (Seq lab seq mqual) = Offset {unOff = B.length $ unSD seq} -instance Show Sequence where-  show (Seq lab seq qual) = ">" ++ (B.unpack $ unSL lab) ++ "\n" ++  (B.unpack $ unSD seq)- toStr :: SeqData -> String toStr  = B.unpack . unSD @@ -106,17 +87,20 @@ mkSeqs = map mkSeq . blocks  mkSeq :: [ByteString] -> Sequence-mkSeq (l:ls) -  -- maybe check this?  | B.length l < 2 || isSpace (B.head $ B.tail l) -  --  = error "Trying to read sequence without a name...and failing."-  | otherwise = Seq (SeqLabel (B.drop 1 l)) (SeqData (B.filter (not . isSpace) $ B.concat $ takeWhile isSeq ls)) Nothing-    where isSeq s = (not . B.null) s && ((flip elem) (['A'..'Z']++['a'..'z']) . B.head) s-mkSeq [] = error "empty input to mkSeq"+mkSeq (l:ls) = Seq (SeqLabel (B.drop 1 l))+               (SeqData (B.filter (not . isSpace) $ B.concat $ takeWhile isSeq ls))+               Nothing+               where isSeq s = (not . B.null) s &&+                               (flip elem (['A'..'Z']++['a'..'z']) . B.head) s+mkSeq []     = error "empty input to mkSeq"  -- | Split lines into blocks starting with '>' characters --   Filter out # comments (but not semicolons?) blocks :: [ByteString] -> [[ByteString]]-blocks = groupBy (const (('>' /=) . B.head)) . filter ((/='#') . B.head) . dropWhile (('>' /=) . B.head) . filter (not . B.null)+blocks = groupBy (const (('>' /=) . B.head))+         . filter ((/='#') . B.head)+         . dropWhile (('>' /=) . B.head)+         . filter (not . B.null)  countSeqs :: FilePath -> IO Int countSeqs f = do

@@ -0,0 +1,16 @@+import Test.QuickCheck+import Text.Printf (printf)++import qualified Data.ByteString.Lazy.Char8 as B++main = mapM_ (\(s,a) -> printf "%-25s: " s >> a) tests++prop_test_00 s = s == s+  where _ = s :: Int++prop_block seq = block seq+  where seq = B.pack '>' B.cons (s :: Bytestring)++tests = [("test_00", quickCheck prop_test_00)+        ,("test_01", quickCheck prop_block  )+        ]