def test_events_table_subset_column_names(columns):
"""Test that EventTableSubset column_names property is accurate."""
etu = EventTableSubset(start="2016-01-01",
stop="2016-01-02",
columns=columns,
table="events.calls")
assert etu.head(0).columns.tolist() == etu.column_names
def test_turn_on_caching():
"""
*.get_dataframe() dataframe is retained when we turning on caching.
"""
sd = EventTableSubset(start="2016-01-01", stop="2016-01-02")
sd.get_dataframe()
sd.turn_off_caching()
sd.turn_on_caching()
sd.get_dataframe()
assert isinstance(sd._df, pd.DataFrame)
def test_events_table_subscriber_ident_substitutions(ident):
"""Test that EventTableSubset replaces the subscriber ident column name with subscriber."""
etu = EventTableSubset(
start="2016-01-01",
stop="2016-01-02",
columns=[ident],
table="events.calls",
subscriber_identifier=ident,
)
assert "subscriber" == etu.head(0).columns[0]
assert ["subscriber"] == etu.column_names
def test_graph():
"""Test that dependency graph util runs and has some correct entries."""
g = daily_location("2016-01-01").dependency_graph()
sd = EventTableSubset("2016-01-01",
"2016-01-02",
columns=["msisdn", "datetime", "location_id"])
assert "x{}".format(sd.md5) in g.nodes()
def test_cache_is_returned():
"""
Cache property is returned when called.
"""
sd = EventTableSubset(start="2016-01-01", stop="2016-01-02")
sd.get_dataframe()
sd.turn_on_caching()
assert sd.cache
sd.turn_off_caching()
assert not sd.cache
def test_subset_correct(subscriber_list, get_dataframe):
"""Test that pushed in subsetting matches .subset result"""
su = EventTableSubset(start="2016-01-01",
stop="2016-01-03",
subscriber_subset=subscriber_list)
subsu = EventTableSubset(start="2016-01-01",
stop="2016-01-03").subset("subscriber",
subscriber_list)
assert all(get_dataframe(su) == get_dataframe(subsu))
su = ModalLocation(*[
daily_location(d, subscriber_subset=subscriber_list)
for d in list_of_dates("2016-01-01", "2016-01-07")
])
subsu = ModalLocation(
*
[daily_location(d)
for d in list_of_dates("2016-01-01", "2016-01-03")]).subset(
"subscriber", subscriber_list)
assert all(get_dataframe(su) == get_dataframe(subsu))
def test_get_df_without_caching():
"""
*.get_dataframe() can still retrieve the dataframe without caching.
"""
sd = EventTableSubset(start="2016-01-01", stop="2016-01-02")
sd.get_dataframe()
sd.turn_off_caching()
assert isinstance(sd.get_dataframe(), pd.DataFrame)
assert isinstance(sd.get_dataframe(), pd.DataFrame)
def test_omitted_subscriber_column(get_dataframe, subscriber_list):
"""Test that a result is returned and warning is raised when omitting a subscriber column."""
with pytest.warns(UserWarning):
su_omit_col = get_dataframe(
EventTableSubset(
start="2016-01-01",
stop="2016-01-03",
subscriber_subset=subscriber_list,
columns=["duration"],
))
su_all_cols = get_dataframe(
EventTableSubset(
start="2016-01-01",
stop="2016-01-03",
subscriber_subset=subscriber_list,
columns=["msisdn", "duration"],
))
assert su_omit_col.duration.values.tolist(
) == su_all_cols.duration.values.tolist()
assert su_omit_col.columns.tolist() == ["duration"]
def test_calculate_dependency_graph():
"""
Test that calculate_dependency_graph() runs and the returned graph has some correct entries.
"""
query = daily_location("2016-01-01")
G = calculate_dependency_graph(query, analyse=True)
sd = EventTableSubset(
start="2016-01-01",
stop="2016-01-02",
columns=["msisdn", "datetime", "location_id"],
)
assert f"x{sd.query_id}" in G.nodes()
assert G.nodes[f"x{sd.query_id}"]["query_object"].query_id == sd.query_id
def test_can_subset_by_sampler(get_dataframe):
"""Test that we can use the output of another query to subset by."""
unique_subs_sample = UniqueSubscribers(
"2016-01-01", "2016-01-07").random_sample(size=10,
sampling_method="system",
seed=0.1)
su = EventTableSubset(start="2016-01-01",
stop="2016-01-03",
subscriber_subset=unique_subs_sample)
su_set = set(get_dataframe(su).subscriber)
uu_set = set(get_dataframe(unique_subs_sample).subscriber)
assert su_set == uu_set
assert len(su_set) == 10
def test_turn_off_caching():
"""
*.turn_off_caching() 'forgets' generated dataframe.
"""
sd = EventTableSubset(start="2016-01-01", stop="2016-01-02")
sd.get_dataframe()
sd.turn_off_caching()
with pytest.raises(AttributeError):
sd._df
def test_cdrs_can_be_subset_by_list(get_dataframe, subscriber_list):
"""
We can subset CDRs with a list.
"""
su = EventTableSubset(start="2016-01-01",
stop="2016-01-03",
subscriber_subset=subscriber_list)
df = get_dataframe(su)
# Get the set of subscribers present in the dataframe, we need to handle the logic
# of msisdn_from/msisdn_to
calculated_subscriber_set = set(df.subscriber)
assert calculated_subscriber_set == set(subscriber_list)
def test_turn_off_caching_handles_error():
"""
*.turn_off_caching() works even if ._df attribute is not present.
"""
sd = EventTableSubset(start="2016-01-01", stop="2016-01-02")
sd.get_dataframe()
sd.turn_off_caching()
sd.turn_on_caching()
sd.get_dataframe()
del sd._df
sd.turn_off_caching()
