def _presort_by_abundance(self, seq_path):
""" Preform pre-sorting of input by abundance """
# Turn off uclust's sorting
self.Params["suppress_sort"] = True
# Get a temp file name for the sorted fasta file
sorted_input_seqs_filepath = get_tmp_filename(prefix=self.Name, suffix=".fasta")
# Sort input seqs by abundance, and write to the temp
# file
sort_fasta_by_abundance(open(seq_path, "U"), open(sorted_input_seqs_filepath, "w"))
# Return the sorted sequences filepath
return sorted_input_seqs_filepath
def test_sort_fasta_by_abundance(self):
"""sort_fasta_by_abundance functions as expected"""
class FakeOutF(object):
def __init__(self):
self.s = ""
def write(self,s):
self.s += s
actual = FakeOutF()
expected = ""
sort_fasta_by_abundance([],actual)
self.assertEqual(actual.s,expected)
# no sorting necessary
actual = FakeOutF()
expected1 = "\n".join(['>s1','ACCGT','>s2 comment','ATTGC',''])
expected2 = "\n".join(['>s2 comment','ATTGC','>s1','ACCGT',''])
sort_fasta_by_abundance(['>s1','ACCGT','>s2 comment','ATTGC'],actual)
# order is unimportant here
self.assertTrue(actual.s == expected1 or actual.s == expected2)
# sorting necessary
actual = FakeOutF()
inseqs = ['>s1','ACCGT',
'>s2 comment','ATTGC',
'>s3 blah','ATTGC']
expected = "\n".join(['>s2 comment','ATTGC',
'>s3 blah','ATTGC',
'>s1','ACCGT',''])
sort_fasta_by_abundance(inseqs,actual)
self.assertEqual(actual.s,expected)
