def test_get_clustered_files(self, tmpdir):
"""
Asserts if the `get_files` method reads only **clustered** files
when its flag is set (i.e., those with the cluster postfix).
For instance, from a directory as the following, it should read
only `file_1_clustered.csv`:
├─── file_1.csv
├─── file_2.csv
├─── file_1_clustered.csv
└─── file_3.txt
:type tmpdir: string
:param tmpdir: The ‘tmpdir’ fixture is a py.path.local object
which will provide a temporary directory unique
to the test invocation.
"""
x = "content"
for i in range(CSV_FILES_COUNT):
tmpdir.join(f"file_{i}{CLUSTERED_FILENAME_POSFIX}.csv").write(x)
tmpdir.join(f"file_{i}.csv").write(x)
tmpdir.join(f"file_n.txt").write(x)
# Act
files = Base.get_files(tmpdir, include_clustered_files=True)
# Assert
assert len(files) == CSV_FILES_COUNT
# checks if files are returned with their absolute path.
for file in files:
assert os.path.isabs(file) == True
def test_cluster_numbers(self, tmp_clustered_files):
# Arrange
tmpdir = tmp_clustered_files[0]
repos = tmp_clustered_files[1]
test_pubs = BaseTestCase.get_test_publications()
# Act
Cluster().run(tmpdir)
files = Base.get_files(tmpdir, include_clustered_files=True)
# Assert
# TODO: This assertion is anti-pattern; must be re-implemented in a much better way.
for file in files:
publications = Base.get_publications(file)
checked = False
for idx, row in publications.iterrows():
for test_pub in test_pubs:
for idx2, row2 in test_pub[0].iterrows():
if row.get("Tools") == row2.get("Tools"):
assert row.get(
CLUSTER_NAME_COLUMN_LABEL) == row2.get(
CLUSTER_NAME_COLUMN_LABEL)
checked = True
assert checked == True