def setUp(self):
"""
Method for instantiating test environment.
"""
self.n_sites = 35
self.c = DistanceMatrix(level="versioned-site")
self.df = self.c.get_dataframe()
class DistanceMatrixTestCase(TestCase):
"""
Tests for the DistanceMatrix() class.
"""
def setUp(self):
"""
Method for instantiating test environment.
"""
self.n_sites = 35
self.c = DistanceMatrix(level="versioned-site")
self.df = self.c.get_dataframe()
def test_some_results(self):
"""
DistanceMatrix() returns a dataframe that contains hand-picked results.
"""
set_df = self.df.set_index("site_id_from")
self.assertAlmostEqual(
round(set_df.loc["8wPojr"]["distance"].values[0]), 789, 2
)
self.assertAlmostEqual(
round(set_df.loc["8wPojr"]["distance"].values[2]), 769, 2
)
self.assertAlmostEqual(
round(set_df.loc["8wPojr"]["distance"].values[4]), 758, 2
)
def test_result_has_correct_length(self):
"""
DistanceMatrix() has the correct length.
"""
self.assertEqual(len(self.df), self.n_sites ** 2)
def __init__(
self,
start: str,
stop: str,
reference_location: BaseLocation,
statistic: str = "avg",
unit: str = "km",
hours: Union[str, Tuple[int, int]] = "all",
table: Union[str, List[str]] = "all",
subscriber_identifier: str = "msisdn",
ignore_nulls: bool = True,
return_subscribers_not_seen: bool = False,
subscriber_subset: Optional[Query] = None,
):
self.return_subscribers_not_seen = return_subscribers_not_seen
self.start = start
self.stop = stop
self.spatial_unit = reference_location.spatial_unit
subscriber_locations = SubscriberLocations(
self.start,
self.stop,
spatial_unit=self.spatial_unit,
hours=hours,
table=table,
subscriber_identifier=subscriber_identifier,
ignore_nulls=ignore_nulls,
subscriber_subset=subscriber_subset,
)
self.statistic = statistic.lower()
if self.statistic not in valid_stats:
raise ValueError(
"{} is not a valid statistic. Use one of {}".format(
self.statistic, valid_stats))
if not isinstance(reference_location, BaseLocation):
raise ValueError(
"Argument 'reference_location' should be an instance of BaseLocation class. "
f"Got: {type(reference_location)}")
else:
self.reference_location = reference_location
self.joined = reference_location.join(
other=subscriber_locations,
on_left=["subscriber"],
left_append="_from",
right_append="_to",
).join(
DistanceMatrix(spatial_unit=self.spatial_unit),
on_left=[
f"{col}_{direction}" for direction in ("from", "to")
for col in self.spatial_unit.location_id_columns
],
right_append="_dist",
how="left outer",
)
self.unit = unit
super().__init__()
def __init__(
self,
*,
subscriber_locations: SubscriberLocations,
reference_location: Union[BaseLocation, Tuple[float, float]] = (0, 0),
statistic: str = "avg",
time_bucket: str = "day",
):
subscriber_locations.spatial_unit.verify_criterion("has_geography")
subscriber_locations.spatial_unit.verify_criterion("has_lon_lat_columns")
self.spatial_unit = subscriber_locations.spatial_unit
if time_bucket.lower() in valid_time_buckets:
self.aggregate_by = time_bucket.lower()
else:
raise ValueError(
f"'{time_bucket}' is not a valid value for time_bucket. Use one of {valid_time_buckets}"
)
if statistic.lower() not in valid_stats:
raise ValueError(
f"'{statistic}' is not a valid statistic. Use one of {valid_stats}"
)
self.statistic = statistic.lower()
self.start = standardise_date(subscriber_locations.start)
self.stop = standardise_date(subscriber_locations.stop)
if isinstance(reference_location, tuple):
self.reference_location = reference_location
self.joined = subscriber_locations
elif isinstance(reference_location, BaseLocation):
if reference_location.spatial_unit != subscriber_locations.spatial_unit:
raise ValueError(
"reference_location must have the same spatial unit as subscriber_locations."
)
self.reference_location = reference_location
self.joined = reference_location.join(
other=subscriber_locations,
on_left=["subscriber"],
left_append="_from",
right_append="_to",
).join(
DistanceMatrix(spatial_unit=self.spatial_unit),
on_left=[
f"{col}_{direction}"
for direction in ("from", "to")
for col in self.spatial_unit.location_id_columns
],
right_append="_dist",
how="left outer",
)
else:
raise ValueError(
"Argument 'reference_location' should be an instance of BaseLocation class or a tuple of two floats. "
f"Got: {type(reference_location).__name__}"
)
super().__init__()
def test_some_results(get_dataframe):
"""
DistanceMatrix() returns a dataframe that contains hand-picked results.
"""
c = DistanceMatrix(level="versioned-site")
df = get_dataframe(c)
set_df = df.set_index("site_id_from")
assert round(set_df.loc["8wPojr"]["distance"].values[0]) == 789
assert round(set_df.loc["8wPojr"]["distance"].values[2]) == 769
assert round(set_df.loc["8wPojr"]["distance"].values[4]) == 758
def test_some_results(get_dataframe):
"""
DistanceMatrix() returns a dataframe that contains hand-picked results.
"""
c = DistanceMatrix(spatial_unit=make_spatial_unit("versioned-site"))
df = get_dataframe(c)
set_df = df.set_index(
["site_id_from", "version_from", "site_id_to", "version_to"])
assert set_df.loc[("8wPojr", 1, "GN2k0G",
0)]["value"] == pytest.approx(789.23239740488)
assert set_df.loc[("8wPojr", 0, "GN2k0G",
0)]["value"] == pytest.approx(769.20155628077)
assert set_df.loc[("8wPojr", 1, "DbWg4K",
0)]["value"] == pytest.approx(757.97771793683)
def test_not_storeable():
with pytest.raises(UnstorableQueryError):
DistanceMatrix(
spatial_unit=make_spatial_unit("versioned-site")).store()
def test_result_has_correct_length(spatial_unit_type, length, get_length):
"""
DistanceMatrix has the correct length.
"""
c = DistanceMatrix(spatial_unit=make_spatial_unit(spatial_unit_type))
assert get_length(c) == length
def test_result_has_correct_length(get_length):
"""
DistanceMatrix() has the correct length.
"""
c = DistanceMatrix(level="versioned-site")
assert get_length(c) == 35 ** 2