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
def test_ttest_delta(self, tmp_clustered_files):
# Arrange
tmpdir, inputs = tmp_clustered_files
input_files = [x["filename"] for x in inputs]
exp_avg_pre = [x["exp_values"]["avg_sum_pre"] for x in inputs]
exp_avg_post = [x["exp_values"]["avg_sum_post"] for x in inputs]
output_file = os.path.join(tmpdir, "test_ttest_delta_output.csv")
# Pre-act Assert
# The output file should not exist before the test runs.
assert os.path.exists(output_file) == False
# Act
TTest().ttest_delta(input_files, output_file)
# Assert
assert os.path.exists(output_file)
output_info = Base.get_publications(output_file)
# Check if files have header
assert BaseTestCase.assert_str_list_equal(list(output_info), TTest.TTEST_HEADER)
# check the value of avg pre citations
assert BaseTestCase.assert_lists_equal(output_info["Average Pre Citations"], exp_avg_pre)
# check the value of avg post citations
assert BaseTestCase.assert_lists_equal(output_info["Average Post Citations"], exp_avg_post)
# Check if the values exist and are not 0.0 or NaN;
# the correctness of their values is asserted via other tests.
for column in output_info:
if column in ["Repository", "Interpretation", "Growth"]:
continue
assert output_info[column].dtype == 'float64'
for cell in output_info[column]:
assert cell > 0.0
assert not math.isnan(cell)
assert not math.isinf(cell)