def test_single_sample(
self, tmpdir, path_proj_conf_file, which_sample_index):
""" Single Sample is perfectly valid for Project and sheet. """
# Pull out the values for the current sample.
values = DATA[which_sample_index]
# Write the annotations.
anns_path = os.path.join(tmpdir.strpath, NAME_ANNOTATIONS_FILE)
with open(anns_path, 'w') as anns_file:
anns_file.write("{}\n".format(",".join(COLUMNS)))
anns_file.write("{}\n".format(",".join([str(v) for v in values])))
# Build the sheet.
p = Project(path_proj_conf_file)
sheet = p.build_sheet()
# It should be a single-row DataFrame.
assert isinstance(sheet, pd.DataFrame)
assert 1 == len(sheet)
assert 1 == p.num_samples
# There will be additional values added from the Project,
# but the core data values will have remained the same.
sample = list(p.samples)[0]
for attr, exp_val in zip(COLUMNS, values):
obs_val = getattr(sample, attr)
try:
assert exp_val == obs_val
except AssertionError as e:
try:
assert exp_val == int(obs_val)
except AssertionError:
raise e
def test_multiple_samples(self, protocols, path_anns_file,
path_proj_conf_file):
""" Project also processes multiple Sample fine. """
p = Project(path_proj_conf_file)
# Total sample count is constant.
assert len(SAMPLE_NAMES) == sum(1 for _ in p.samples)
# But the sheet permits filtering to specific protocol(s).
exp_num_samples = len(SAMPLE_NAMES) if not protocols else \
sum(sum(1 for p2 in PROTOCOLS if p2 == p1) for p1 in protocols)
sheet = p.build_sheet(*protocols)
assert exp_num_samples == len(sheet)
if protocols:
def as_expected(sd):
return sd.protocol in set(protocols)
else:
def as_expected(sd):
return sd.protocol not in set(protocols)
for _, sample_data in sheet.iterrows():
assert as_expected(sample_data)
def test_single_sample(self, tmpdir, path_proj_conf_file,
which_sample_index):
""" Single Sample is perfectly valid for Project and sheet. """
# Pull out the values for the current sample.
values = DATA[which_sample_index]
# Write the annotations.
anns_path = os.path.join(tmpdir.strpath, NAME_ANNOTATIONS_FILE)
with open(anns_path, 'w') as anns_file:
anns_file.write("{}\n".format(",".join(COLUMNS)))
anns_file.write("{}\n".format(",".join([str(v) for v in values])))
# Build the sheet.
p = Project(path_proj_conf_file)
sheet = p.build_sheet()
# It should be a single-row DataFrame.
assert isinstance(sheet, pd.DataFrame)
assert 1 == len(sheet)
assert 1 == p.num_samples
# There will be additional values added from the Project,
# but the core data values will have remained the same.
sample = list(p.samples)[0]
for attr, exp_val in zip(COLUMNS, values):
obs_val = getattr(sample, attr)
try:
assert exp_val == obs_val
except AssertionError as e:
try:
assert exp_val == int(obs_val)
except AssertionError:
raise e
def test_no_samples(self, protocols, delimiter, path_empty_project):
""" Lack of Samples is unproblematic for the sheet build. """
# Regardless of protocol(s), the sheet should be empty.
print("Test config file: {}".format(path_empty_project))
p = Project(path_empty_project)
sheet = p.build_sheet(*protocols)
assert sheet.empty
def test_multiple_samples(
self, protocols, path_anns_file, path_proj_conf_file):
""" Project also processes multiple Sample fine. """
p = Project(path_proj_conf_file)
# Total sample count is constant.
assert len(SAMPLE_NAMES) == sum(1 for _ in p.samples)
# But the sheet permits filtering to specific protocol(s).
exp_num_samples = len(SAMPLE_NAMES) if not protocols else \
sum(sum(1 for p2 in PROTOCOLS if p2 == p1) for p1 in protocols)
sheet = p.build_sheet(*protocols)
assert exp_num_samples == len(sheet)
if protocols:
def as_expected(sd):
return sd.protocol in set(protocols)
else:
def as_expected(sd):
return sd.protocol not in set(protocols)
for _, sample_data in sheet.iterrows():
assert as_expected(sample_data)
def test_no_samples(self, protocols, delimiter, path_empty_project):
""" Lack of Samples is unproblematic for the sheet build. """
# Regardless of protocol(s), the sheet should be empty.
p = Project(path_empty_project)
sheet = p.build_sheet(*protocols)
assert sheet.empty