class TestPrepTemplate(TestCase):
"""Tests the PrepTemplate class"""
def setUp(self):
metadata_dict = {
'SKB8.640193': {'center_name': 'ANL',
'center_project_name': 'Test Project',
'ebi_submission_accession': None,
'EMP_status_id': 1,
'data_type_id': 2,
'str_column': 'Value for sample 1'},
'SKD8.640184': {'center_name': 'ANL',
'center_project_name': 'Test Project',
'ebi_submission_accession': None,
'EMP_status_id': 1,
'data_type_id': 2,
'str_column': 'Value for sample 2'},
'SKB7.640196': {'center_name': 'ANL',
'center_project_name': 'Test Project',
'ebi_submission_accession': None,
'EMP_status_id': 1,
'data_type_id': 2,
'str_column': 'Value for sample 3'}
}
self.metadata = pd.DataFrame.from_dict(metadata_dict, orient='index')
self.test_raw_data = RawData(1)
fd, seqs_fp = mkstemp(suffix='_seqs.fastq')
close(fd)
fd, barcodes_fp = mkstemp(suffix='_barcodes.fastq')
close(fd)
filepaths = [(seqs_fp, 1), (barcodes_fp, 2)]
with open(seqs_fp, "w") as f:
f.write("\n")
with open(barcodes_fp, "w") as f:
f.write("\n")
self.new_raw_data = RawData.create(2, filepaths, [Study(1)])
db_test_raw_dir = join(get_db_files_base_dir(), 'raw_data')
db_seqs_fp = join(db_test_raw_dir, "3_%s" % basename(seqs_fp))
db_barcodes_fp = join(db_test_raw_dir, "3_%s" % basename(barcodes_fp))
self._clean_up_files = [db_seqs_fp, db_barcodes_fp]
self.tester = PrepTemplate(1)
self.exp_sample_ids = {'SKB1.640202', 'SKB2.640194', 'SKB3.640195',
'SKB4.640189', 'SKB5.640181', 'SKB6.640176',
'SKB7.640196', 'SKB8.640193', 'SKB9.640200',
'SKD1.640179', 'SKD2.640178', 'SKD3.640198',
'SKD4.640185', 'SKD5.640186', 'SKD6.640190',
'SKD7.640191', 'SKD8.640184', 'SKD9.640182',
'SKM1.640183', 'SKM2.640199', 'SKM3.640197',
'SKM4.640180', 'SKM5.640177', 'SKM6.640187',
'SKM7.640188', 'SKM8.640201', 'SKM9.640192'}
def tearDown(self):
for f in self._clean_up_files:
remove(f)
def test_init_unknown_error(self):
"""Init raises an error if the id is not known"""
with self.assertRaises(QiitaDBUnknownIDError):
PrepTemplate(2)
def test_init(self):
"""Init successfully instantiates the object"""
st = PrepTemplate(1)
self.assertTrue(st.id, 1)
def test_table_name(self):
"""Table name return the correct string"""
obs = PrepTemplate._table_name(self.test_raw_data)
self.assertEqual(obs, "prep_1")
def test_create_duplicate(self):
"""Create raises an error when creating a duplicated PrepTemplate"""
with self.assertRaises(QiitaDBDuplicateError):
PrepTemplate.create(self.metadata, self.test_raw_data)
def test_create_duplicate_header(self):
"""Create raises an error when duplicate headers are present"""
self.metadata['STR_COLUMN'] = pd.Series(['', '', ''],
index=self.metadata.index)
with self.assertRaises(QiitaDBDuplicateHeaderError):
PrepTemplate.create(self.metadata, self.new_raw_data)
def test_create(self):
"""Creates a new PrepTemplate"""
pt = PrepTemplate.create(self.metadata, self.new_raw_data)
# The returned object has the correct id
self.assertEqual(pt.id, 3)
# The relevant rows to common_prep_info have been added.
obs = self.conn_handler.execute_fetchall(
"SELECT * FROM qiita.common_prep_info WHERE raw_data_id=3")
# raw_data_id, sample_id, center_name, center_project_name,
# ebi_submission_accession, ebi_study_accession, emp_status_id,
# data_type_id
exp = [[3, 'SKB8.640193', 'ANL', 'Test Project', None, None, 1, 2],
[3, 'SKD8.640184', 'ANL', 'Test Project', None, None, 1, 2],
[3, 'SKB7.640196', 'ANL', 'Test Project', None, None, 1, 2]]
self.assertEqual(sorted(obs), sorted(exp))
# The relevant rows have been added to the raw_data_prep_columns
obs = self.conn_handler.execute_fetchall(
"SELECT * FROM qiita.raw_data_prep_columns WHERE raw_data_id=3")
# raw_data_id, column_name, column_type
exp = [[3, "str_column", "varchar"]]
self.assertEqual(obs, exp)
# The new table exists
self.assertTrue(exists_table("prep_3", self.conn_handler))
# The new table hosts the correct values
obs = self.conn_handler.execute_fetchall(
"SELECT * FROM qiita.prep_3")
# sample_id, str_column
exp = [['SKB8.640193', "Value for sample 1"],
['SKD8.640184', "Value for sample 2"],
['SKB7.640196', "Value for sample 3"]]
self.assertEqual(sorted(obs), sorted(exp))
def test_exists_true(self):
"""Exists returns true when the PrepTemplate already exists"""
self.assertTrue(PrepTemplate.exists(self.test_raw_data))
def test_exists_false(self):
"""Exists returns false when the PrepTemplate does not exists"""
self.assertFalse(PrepTemplate.exists(self.new_raw_data))
def test_get_sample_ids(self):
"""get_sample_ids returns the correct set of sample ids"""
obs = self.tester._get_sample_ids(self.conn_handler)
self.assertEqual(obs, self.exp_sample_ids)
def test_len(self):
"""Len returns the correct number of sample ids"""
self.assertEqual(len(self.tester), 27)
def test_getitem(self):
"""Get item returns the correct sample object"""
obs = self.tester['SKM7.640188']
exp = PrepSample('SKM7.640188', self.tester)
self.assertEqual(obs, exp)
def test_getitem_error(self):
"""Get item raises an error if key does not exists"""
with self.assertRaises(KeyError):
self.tester['Not_a_Sample']
def test_setitem(self):
"""setitem raises an error (currently not allowed)"""
with self.assertRaises(QiitaDBNotImplementedError):
self.tester['SKM7.640188'] = PrepSample('SKM7.640188', self.tester)
def test_delitem(self):
"""delitem raises an error (currently not allowed)"""
with self.assertRaises(QiitaDBNotImplementedError):
del self.tester['SKM7.640188']
def test_iter(self):
"""iter returns an iterator over the sample ids"""
obs = self.tester.__iter__()
self.assertTrue(isinstance(obs, Iterable))
self.assertEqual(set(obs), self.exp_sample_ids)
def test_contains_true(self):
"""contains returns true if the sample id exists"""
self.assertTrue('SKM7.640188' in self.tester)
def test_contains_false(self):
"""contains returns false if the sample id does not exists"""
self.assertFalse('Not_a_Sample' in self.tester)
def test_keys(self):
"""keys returns an iterator over the sample ids"""
obs = self.tester.keys()
self.assertTrue(isinstance(obs, Iterable))
self.assertEqual(set(obs), self.exp_sample_ids)
def test_values(self):
"""values returns an iterator over the values"""
obs = self.tester.values()
self.assertTrue(isinstance(obs, Iterable))
exp = {PrepSample('SKB1.640202', self.tester),
PrepSample('SKB2.640194', self.tester),
PrepSample('SKB3.640195', self.tester),
PrepSample('SKB4.640189', self.tester),
PrepSample('SKB5.640181', self.tester),
PrepSample('SKB6.640176', self.tester),
PrepSample('SKB7.640196', self.tester),
PrepSample('SKB8.640193', self.tester),
PrepSample('SKB9.640200', self.tester),
PrepSample('SKD1.640179', self.tester),
PrepSample('SKD2.640178', self.tester),
PrepSample('SKD3.640198', self.tester),
PrepSample('SKD4.640185', self.tester),
PrepSample('SKD5.640186', self.tester),
PrepSample('SKD6.640190', self.tester),
PrepSample('SKD7.640191', self.tester),
PrepSample('SKD8.640184', self.tester),
PrepSample('SKD9.640182', self.tester),
PrepSample('SKM1.640183', self.tester),
PrepSample('SKM2.640199', self.tester),
PrepSample('SKM3.640197', self.tester),
PrepSample('SKM4.640180', self.tester),
PrepSample('SKM5.640177', self.tester),
PrepSample('SKM6.640187', self.tester),
PrepSample('SKM7.640188', self.tester),
PrepSample('SKM8.640201', self.tester),
PrepSample('SKM9.640192', self.tester)}
# Creating a list and looping over it since unittest does not call
# the __eq__ function on the objects
for o, e in zip(sorted(list(obs), key=lambda x: x.id),
sorted(exp, key=lambda x: x.id)):
self.assertEqual(o, e)
def test_items(self):
"""items returns an iterator over the (key, value) tuples"""
obs = self.tester.items()
self.assertTrue(isinstance(obs, Iterable))
exp = [('SKB1.640202', PrepSample('SKB1.640202', self.tester)),
('SKB2.640194', PrepSample('SKB2.640194', self.tester)),
('SKB3.640195', PrepSample('SKB3.640195', self.tester)),
('SKB4.640189', PrepSample('SKB4.640189', self.tester)),
('SKB5.640181', PrepSample('SKB5.640181', self.tester)),
('SKB6.640176', PrepSample('SKB6.640176', self.tester)),
('SKB7.640196', PrepSample('SKB7.640196', self.tester)),
('SKB8.640193', PrepSample('SKB8.640193', self.tester)),
('SKB9.640200', PrepSample('SKB9.640200', self.tester)),
('SKD1.640179', PrepSample('SKD1.640179', self.tester)),
('SKD2.640178', PrepSample('SKD2.640178', self.tester)),
('SKD3.640198', PrepSample('SKD3.640198', self.tester)),
('SKD4.640185', PrepSample('SKD4.640185', self.tester)),
('SKD5.640186', PrepSample('SKD5.640186', self.tester)),
('SKD6.640190', PrepSample('SKD6.640190', self.tester)),
('SKD7.640191', PrepSample('SKD7.640191', self.tester)),
('SKD8.640184', PrepSample('SKD8.640184', self.tester)),
('SKD9.640182', PrepSample('SKD9.640182', self.tester)),
('SKM1.640183', PrepSample('SKM1.640183', self.tester)),
('SKM2.640199', PrepSample('SKM2.640199', self.tester)),
('SKM3.640197', PrepSample('SKM3.640197', self.tester)),
('SKM4.640180', PrepSample('SKM4.640180', self.tester)),
('SKM5.640177', PrepSample('SKM5.640177', self.tester)),
('SKM6.640187', PrepSample('SKM6.640187', self.tester)),
('SKM7.640188', PrepSample('SKM7.640188', self.tester)),
('SKM8.640201', PrepSample('SKM8.640201', self.tester)),
('SKM9.640192', PrepSample('SKM9.640192', self.tester))]
# Creating a list and looping over it since unittest does not call
# the __eq__ function on the objects
for o, e in zip(sorted(list(obs)), sorted(exp)):
self.assertEqual(o, e)
def test_get(self):
"""get returns the correct PrepSample object"""
obs = self.tester.get('SKM7.640188')
exp = PrepSample('SKM7.640188', self.tester)
self.assertEqual(obs, exp)
def test_get_none(self):
"""get returns none if the sample id is not present"""
self.assertTrue(self.tester.get('Not_a_Sample') is None)
def test_to_file(self):
"""to file writes a tab delimited file with all the metadata"""
fd, fp = mkstemp()
close(fd)
pt = PrepTemplate.create(self.metadata, self.new_raw_data)
pt.to_file(fp)
self._clean_up_files.append(fp)
with open(fp, 'U') as f:
obs = f.read()
self.assertEqual(obs, EXP_PREP_TEMPLATE)
class TestPrepTemplate(TestCase):
"""Tests the PrepTemplate class"""
def setUp(self):
metadata_dict = {
'SKB8.640193': {
'center_name': 'ANL',
'center_project_name': 'Test Project',
'ebi_submission_accession': None,
'EMP_status_id': 1,
'data_type_id': 2,
'str_column': 'Value for sample 1'
},
'SKD8.640184': {
'center_name': 'ANL',
'center_project_name': 'Test Project',
'ebi_submission_accession': None,
'EMP_status_id': 1,
'data_type_id': 2,
'str_column': 'Value for sample 2'
},
'SKB7.640196': {
'center_name': 'ANL',
'center_project_name': 'Test Project',
'ebi_submission_accession': None,
'EMP_status_id': 1,
'data_type_id': 2,
'str_column': 'Value for sample 3'
}
}
self.metadata = pd.DataFrame.from_dict(metadata_dict, orient='index')
self.test_raw_data = RawData(1)
fd, seqs_fp = mkstemp(suffix='_seqs.fastq')
close(fd)
fd, barcodes_fp = mkstemp(suffix='_barcodes.fastq')
close(fd)
filepaths = [(seqs_fp, 1), (barcodes_fp, 2)]
with open(seqs_fp, "w") as f:
f.write("\n")
with open(barcodes_fp, "w") as f:
f.write("\n")
self.new_raw_data = RawData.create(2, filepaths, [Study(1)])
db_test_raw_dir = join(get_db_files_base_dir(), 'raw_data')
db_seqs_fp = join(db_test_raw_dir, "3_%s" % basename(seqs_fp))
db_barcodes_fp = join(db_test_raw_dir, "3_%s" % basename(barcodes_fp))
self._clean_up_files = [db_seqs_fp, db_barcodes_fp]
self.tester = PrepTemplate(1)
self.exp_sample_ids = {
'SKB1.640202', 'SKB2.640194', 'SKB3.640195', 'SKB4.640189',
'SKB5.640181', 'SKB6.640176', 'SKB7.640196', 'SKB8.640193',
'SKB9.640200', 'SKD1.640179', 'SKD2.640178', 'SKD3.640198',
'SKD4.640185', 'SKD5.640186', 'SKD6.640190', 'SKD7.640191',
'SKD8.640184', 'SKD9.640182', 'SKM1.640183', 'SKM2.640199',
'SKM3.640197', 'SKM4.640180', 'SKM5.640177', 'SKM6.640187',
'SKM7.640188', 'SKM8.640201', 'SKM9.640192'
}
def tearDown(self):
for f in self._clean_up_files:
remove(f)
def test_init_unknown_error(self):
"""Init raises an error if the id is not known"""
with self.assertRaises(QiitaDBUnknownIDError):
PrepTemplate(2)
def test_init(self):
"""Init successfully instantiates the object"""
st = PrepTemplate(1)
self.assertTrue(st.id, 1)
def test_table_name(self):
"""Table name return the correct string"""
obs = PrepTemplate._table_name(self.test_raw_data)
self.assertEqual(obs, "prep_1")
def test_create_duplicate(self):
"""Create raises an error when creating a duplicated PrepTemplate"""
with self.assertRaises(QiitaDBDuplicateError):
PrepTemplate.create(self.metadata, self.test_raw_data)
def test_create_duplicate_header(self):
"""Create raises an error when duplicate headers are present"""
self.metadata['STR_COLUMN'] = pd.Series(['', '', ''],
index=self.metadata.index)
with self.assertRaises(QiitaDBDuplicateHeaderError):
PrepTemplate.create(self.metadata, self.new_raw_data)
def test_create(self):
"""Creates a new PrepTemplate"""
pt = PrepTemplate.create(self.metadata, self.new_raw_data)
# The returned object has the correct id
self.assertEqual(pt.id, 3)
# The relevant rows to common_prep_info have been added.
obs = self.conn_handler.execute_fetchall(
"SELECT * FROM qiita.common_prep_info WHERE raw_data_id=3")
# raw_data_id, sample_id, center_name, center_project_name,
# ebi_submission_accession, ebi_study_accession, emp_status_id,
# data_type_id
exp = [[3, 'SKB8.640193', 'ANL', 'Test Project', 1, 2],
[3, 'SKD8.640184', 'ANL', 'Test Project', 1, 2],
[3, 'SKB7.640196', 'ANL', 'Test Project', 1, 2]]
self.assertEqual(sorted(obs), sorted(exp))
# The relevant rows have been added to the raw_data_prep_columns
obs = self.conn_handler.execute_fetchall(
"SELECT * FROM qiita.raw_data_prep_columns WHERE raw_data_id=3")
# raw_data_id, column_name, column_type
exp = [[3, 'str_column', 'varchar'],
[3, 'ebi_submission_accession', 'varchar']]
self.assertEqual(obs, exp)
# The new table exists
self.assertTrue(exists_table("prep_3", self.conn_handler))
# The new table hosts the correct values
obs = self.conn_handler.execute_fetchall("SELECT * FROM qiita.prep_3")
# sample_id, str_column
exp = [['SKB8.640193', "Value for sample 1", None],
['SKD8.640184', "Value for sample 2", None],
['SKB7.640196', "Value for sample 3", None]]
self.assertEqual(sorted(obs), sorted(exp))
def test_exists_true(self):
"""Exists returns true when the PrepTemplate already exists"""
self.assertTrue(PrepTemplate.exists(self.test_raw_data))
def test_exists_false(self):
"""Exists returns false when the PrepTemplate does not exists"""
self.assertFalse(PrepTemplate.exists(self.new_raw_data))
def test_get_sample_ids(self):
"""get_sample_ids returns the correct set of sample ids"""
obs = self.tester._get_sample_ids(self.conn_handler)
self.assertEqual(obs, self.exp_sample_ids)
def test_len(self):
"""Len returns the correct number of sample ids"""
self.assertEqual(len(self.tester), 27)
def test_getitem(self):
"""Get item returns the correct sample object"""
obs = self.tester['SKM7.640188']
exp = PrepSample('SKM7.640188', self.tester)
self.assertEqual(obs, exp)
def test_getitem_error(self):
"""Get item raises an error if key does not exists"""
with self.assertRaises(KeyError):
self.tester['Not_a_Sample']
def test_setitem(self):
"""setitem raises an error (currently not allowed)"""
with self.assertRaises(QiitaDBNotImplementedError):
self.tester['SKM7.640188'] = PrepSample('SKM7.640188', self.tester)
def test_delitem(self):
"""delitem raises an error (currently not allowed)"""
with self.assertRaises(QiitaDBNotImplementedError):
del self.tester['SKM7.640188']
def test_iter(self):
"""iter returns an iterator over the sample ids"""
obs = self.tester.__iter__()
self.assertTrue(isinstance(obs, Iterable))
self.assertEqual(set(obs), self.exp_sample_ids)
def test_contains_true(self):
"""contains returns true if the sample id exists"""
self.assertTrue('SKM7.640188' in self.tester)
def test_contains_false(self):
"""contains returns false if the sample id does not exists"""
self.assertFalse('Not_a_Sample' in self.tester)
def test_keys(self):
"""keys returns an iterator over the sample ids"""
obs = self.tester.keys()
self.assertTrue(isinstance(obs, Iterable))
self.assertEqual(set(obs), self.exp_sample_ids)
def test_values(self):
"""values returns an iterator over the values"""
obs = self.tester.values()
self.assertTrue(isinstance(obs, Iterable))
exp = {
PrepSample('SKB1.640202', self.tester),
PrepSample('SKB2.640194', self.tester),
PrepSample('SKB3.640195', self.tester),
PrepSample('SKB4.640189', self.tester),
PrepSample('SKB5.640181', self.tester),
PrepSample('SKB6.640176', self.tester),
PrepSample('SKB7.640196', self.tester),
PrepSample('SKB8.640193', self.tester),
PrepSample('SKB9.640200', self.tester),
PrepSample('SKD1.640179', self.tester),
PrepSample('SKD2.640178', self.tester),
PrepSample('SKD3.640198', self.tester),
PrepSample('SKD4.640185', self.tester),
PrepSample('SKD5.640186', self.tester),
PrepSample('SKD6.640190', self.tester),
PrepSample('SKD7.640191', self.tester),
PrepSample('SKD8.640184', self.tester),
PrepSample('SKD9.640182', self.tester),
PrepSample('SKM1.640183', self.tester),
PrepSample('SKM2.640199', self.tester),
PrepSample('SKM3.640197', self.tester),
PrepSample('SKM4.640180', self.tester),
PrepSample('SKM5.640177', self.tester),
PrepSample('SKM6.640187', self.tester),
PrepSample('SKM7.640188', self.tester),
PrepSample('SKM8.640201', self.tester),
PrepSample('SKM9.640192', self.tester)
}
# Creating a list and looping over it since unittest does not call
# the __eq__ function on the objects
for o, e in zip(sorted(list(obs), key=lambda x: x.id),
sorted(exp, key=lambda x: x.id)):
self.assertEqual(o, e)
def test_items(self):
"""items returns an iterator over the (key, value) tuples"""
obs = self.tester.items()
self.assertTrue(isinstance(obs, Iterable))
exp = [('SKB1.640202', PrepSample('SKB1.640202', self.tester)),
('SKB2.640194', PrepSample('SKB2.640194', self.tester)),
('SKB3.640195', PrepSample('SKB3.640195', self.tester)),
('SKB4.640189', PrepSample('SKB4.640189', self.tester)),
('SKB5.640181', PrepSample('SKB5.640181', self.tester)),
('SKB6.640176', PrepSample('SKB6.640176', self.tester)),
('SKB7.640196', PrepSample('SKB7.640196', self.tester)),
('SKB8.640193', PrepSample('SKB8.640193', self.tester)),
('SKB9.640200', PrepSample('SKB9.640200', self.tester)),
('SKD1.640179', PrepSample('SKD1.640179', self.tester)),
('SKD2.640178', PrepSample('SKD2.640178', self.tester)),
('SKD3.640198', PrepSample('SKD3.640198', self.tester)),
('SKD4.640185', PrepSample('SKD4.640185', self.tester)),
('SKD5.640186', PrepSample('SKD5.640186', self.tester)),
('SKD6.640190', PrepSample('SKD6.640190', self.tester)),
('SKD7.640191', PrepSample('SKD7.640191', self.tester)),
('SKD8.640184', PrepSample('SKD8.640184', self.tester)),
('SKD9.640182', PrepSample('SKD9.640182', self.tester)),
('SKM1.640183', PrepSample('SKM1.640183', self.tester)),
('SKM2.640199', PrepSample('SKM2.640199', self.tester)),
('SKM3.640197', PrepSample('SKM3.640197', self.tester)),
('SKM4.640180', PrepSample('SKM4.640180', self.tester)),
('SKM5.640177', PrepSample('SKM5.640177', self.tester)),
('SKM6.640187', PrepSample('SKM6.640187', self.tester)),
('SKM7.640188', PrepSample('SKM7.640188', self.tester)),
('SKM8.640201', PrepSample('SKM8.640201', self.tester)),
('SKM9.640192', PrepSample('SKM9.640192', self.tester))]
# Creating a list and looping over it since unittest does not call
# the __eq__ function on the objects
for o, e in zip(sorted(list(obs)), sorted(exp)):
self.assertEqual(o, e)
def test_get(self):
"""get returns the correct PrepSample object"""
obs = self.tester.get('SKM7.640188')
exp = PrepSample('SKM7.640188', self.tester)
self.assertEqual(obs, exp)
def test_get_none(self):
"""get returns none if the sample id is not present"""
self.assertTrue(self.tester.get('Not_a_Sample') is None)
def test_to_file(self):
"""to file writes a tab delimited file with all the metadata"""
fd, fp = mkstemp()
close(fd)
pt = PrepTemplate.create(self.metadata, self.new_raw_data)
pt.to_file(fp)
self._clean_up_files.append(fp)
with open(fp, 'U') as f:
obs = f.read()
self.assertEqual(obs, EXP_PREP_TEMPLATE)
