def test_silva_taxonomy_format_validate_negative_nonnumeric_values(self):
filepath = self.get_data_path('silva_taxa.tsv')
badtax = pd.read_csv(filepath, sep='\t', index_col=0, header=None)
# turn the taxids to a non-numeric column
badtax[1] = 'blarg'
junkpath = os.path.join(self.temp_dir.name, 'junk.tsv')
badtax.to_csv(junkpath, sep='\t', header=False)
format = SILVATaxonomyFormat(junkpath, mode='r')
with self.assertRaisesRegex(ValidationError, 'non-numeric value'):
format.validate()
def test_silva_taxonomy_format_validate_negative_invalid_taxonomy(self):
filepath = self.get_data_path('silva_taxa.tsv')
# index_col=1 sets the wrong index, creating an invalid taxonomy
badtax = pd.read_csv(filepath, sep='\t', index_col=1, header=None)
junkpath = os.path.join(self.temp_dir.name, 'junk.tsv')
badtax.to_csv(junkpath, sep='\t', header=False)
format = SILVATaxonomyFormat(junkpath, mode='r')
with self.assertRaisesRegex(ValidationError,
'not in SILVA taxonomy format'):
format.validate()
def test_silva_taxonomy_format_validate_negative_empty_file(self):
filepath = self.get_data_path('empty.tsv')
format = SILVATaxonomyFormat(filepath, mode='r')
with self.assertRaisesRegex(ValidationError,
'must be at least one.*record'):
format.validate()
def test_silva_taxonomy_format_validate_negative_incorrect_row_count(self):
filepath = self.get_data_path('silva_taxamap.tsv')
format = SILVATaxonomyFormat(filepath, mode='r')
with self.assertRaisesRegex(ValidationError, 'Expected.*5 fields'):
format.validate()
def test_silva_taxonomy_format_validate_positive(self):
filepath = self.get_data_path('silva_taxa.tsv')
format = SILVATaxonomyFormat(filepath, mode='r')
# These should both just succeed
format.validate('min')
format.validate('max')