def test_wrong_format():
"""
Test the behavior of providing as input_file an invalid file. In this test
case the input_file will be a .csv. The avg_delivery_time function should
raise an InvalidFileException given that the mandatory format is jsonlines
(one valid JSON per line).
"""
with pytest.raises(InvalidFileException):
avg_delivery_time('data/csv_format.json', None, 10)
def test_json_format():
"""
Test the behavior of providing as input_file a valid JSON format, with an
array of lists. The avg_delivery_time function should raise an
InvalidFileException given that the mandatory format is jsonlines (one valid
JSON per line).
"""
with pytest.raises(InvalidFileException):
avg_delivery_time('data/json_format.json', None, 10)
def test_empty_file():
"""
Test behavior when the input is an empty file. The application should
generate an empty output file (first assertion) and the file should be
empty, i.e., with no lines (second assertion).
"""
avg_delivery_time('data/empty.jsonl', 'data/empty_output.jsonl', 10)
assert os.path.isfile('data/empty_output.jsonl')
with open('data/empty_output.jsonl', 'r') as fp:
lines = fp.readlines()
assert len(lines) == 0
fp.close()
def test_sample_file():
"""
Test if the avg_delivery_time continues to generate the same output expected
output for the 'data/events.jsonl' file. This is the example provided in
the challenge README.md.
The test is based on comparing the contents of the
'data/events_exp_out.jsonl' (expected output) and 'data/events.out.json'
(regenerated output).
"""
avg_delivery_time('data/events.jsonl', 'data/events_out.jsonl', 10)
df_expected_output = pd.read_json('data/events_exp_out.jsonl',
orient='records', lines=True)
df_output = pd.read_json('data/events_out.jsonl', orient='records',
lines=True)
# usage of the pandas.DataFrame.equals() to test whether the to DataFrames
# are the same
assert df_output.equals(df_expected_output), 'different expected output ' \
'in data/events_out.jsonl'