view sra.py @ 13:45031bbf6b27 draft

better sra_fetch code, compliant (?) datatype, updated dependencies in tool wrappers

"""
Sra class
"""

import binascii
from galaxy.datatypes.binary import Binary
from galaxy.datatypes.data import Data
from galaxy.datatypes.metadata import MetadataElement
from galaxy.datatypes.sniff import *

class Sra( Binary ):
    """ Sequence Read Archive (SRA) """
    file_ext = 'sra'

    def __init__( self, **kwd ):
        Binary.__init__( self, **kwd )
    def sniff( self, filename ):
        """ The first 8 bytes of any NCBI sra file is 'NCIB.sra', and the file is binary. Not sure if EBI and DDBJ files may differ. 
        For details about the format, see http://www.ncbi.nlm.nih.gov/books/n/helpsra/SRA_Overview_BK/#SRA_Overview_BK.4_SRA_Data_Structure 
        """
        try:
            header = open(filename).read(8)
            if binascii.b2a_hex(header) == binascii.hexlify('NCBI.sra'):
                return True
            else:
                return False
        except:
            return False
    def set_peek(self, dataset, is_multi_byte=False):
        if not dataset.dataset.purged:
            dataset.peek  = 'Binary SRA file'
            dataset.blurb = data.nice_size(dataset.get_size())
        else:
            dataset.peek = 'file does not exist'
            dataset.blurb = 'file purged from disk'
    def display_peek(self, dataset):
        try:
            return dataset.peek
        except:
            return 'Binary sra file (%s)' % ( data.nice_size(dataset.get_size()))