def test_non_standard_params(self):
"""
Non-standard parameters don't trigger reading the cache.
"""
dl = daily_location("2016-01-01", level="cell")
table_name = dl.table_name
dl.store().result()
dl2 = daily_location("2016-01-01", level="cell", method="most-common")
self.assertFalse(table_name in dl2.get_query())
dl.invalidate_db_cache()
def setUp(self):
self.dls = [
daily_location("2016-01-01 18:00:00",
stop="2016-01-02 06:00:00",
method="most-common"),
daily_location("2016-01-02 18:00:00",
stop="2016-01-03 06:00:00",
method="most-common"),
daily_location("2016-01-03 18:00:00",
stop="2016-01-04 06:00:00",
method="most-common"),
]
def test_hours(self):
"""
Test that daily locations handles the hours parameter
"""
# Lower level test test that subsetdates handles this correctly
# we're just testing that it is passed on in this case.
dl1 = daily_location("2016-01-01", level="cell")
dl2 = daily_location("2016-01-01", level="cell", hours=(19, 23))
dl3 = daily_location("2016-01-01", level="cell", hours=(19, 20))
self.assertTrue(len(dl1) > len(dl2) > len(dl3))
def test_can_be_stored(self):
"""
We can store a daily_location() in the right place.
"""
schema = "cache"
dl = daily_location("2016-01-01", level="cell")
table_name = dl.table_name.split(".")[1]
dl.store().result()
self.assertTrue(dl.connection.has_table(table_name, schema=schema))
dl = daily_location("2016-01-01", level="cell")
self.assertTrue(table_name in dl.get_query())
dl.invalidate_db_cache()
def test_daily_locs_errors(self):
"""
daily_location() errors when we ask for a date that does not exist.
"""
with self.assertRaises(MissingDateError):
dl = daily_location("2016-01-31")
def test_can_return_df(self):
"""
daily_location() can be initialised and return a dataframe.
"""
dl = daily_location("2016-01-01")
self.assertIs(type(dl.get_dataframe()), pd.DataFrame)
def setUp(self):
self.hl = HomeLocation(*[
daily_location(d)
for d in list_of_dates("2016-01-01", "2016-01-03")
])
self.hdf = self.hl.get_dataframe()
self.hdf = self.hl.get_dataframe().set_index("subscriber")
def test_works_with_pcods(self):
"""
We can get daily locations with p-codes rather than the standard names.
"""
dl = daily_location("2016-01-05",
level="admin3",
column_name="admin3pcod")
df = dl.get_dataframe()
self.assertTrue(df.admin3pcod[0].startswith("524"))
def test_storing_updates(self):
"""
Storing updates the query after storing.
"""
dl = daily_location("2016-01-01", level="cell")
dl.store().result()
sql = dl.get_query()
self.assertTrue(sql.startswith("SELECT * FROM cache"))
dl.invalidate_db_cache()
def test_equivalent_to_locate_subscribers(self):
"""
daily_location() is equivalent to the MostFrequentLocation().
"""
mfl = MostFrequentLocation("2016-01-01", "2016-01-02")
mfl_df = mfl.get_dataframe()
dl = daily_location("2016-01-01", method="most-common")
dl_df = dl.get_dataframe()
self.assertTrue((dl_df == mfl_df).all().all())
def test_can_be_aggregated_latlong(self):
"""
Query can be aggregated to a spatial level with lat-lon data.
"""
hl = HomeLocation(*[
daily_location(d, level="lat-lon", method="last")
for d in list_of_dates("2016-01-01", "2016-01-03")
])
agg = hl.aggregate()
df = agg.get_dataframe()
self.assertIs(type(df), pd.DataFrame)
self.assertEqual(list(df.columns), ["lat", "lon", "total"])
def test_force(self):
"""
We can force an overwrite of the table.
"""
dl = daily_location("2016-01-01", level="cell")
dl.store().result()
sql = "DELETE FROM {}".format(dl.table_name)
dl.connection.engine.execute(sql)
# This should not raise an error
dl.store().result()
self.assertTrue(len(dl) == 0)
dl.store(force=True)
self.assertTrue(len(dl.head()) != 0)
dl.invalidate_db_cache()
def test_equivalent_to_locate_subscribers_with_time(self):
"""
daily_location() is equivalent to the MostFrequentLocation() with timestamps.
"""
mfl = MostFrequentLocation("2016-01-01 18:00:00",
"2016-01-02 06:00:00")
mfl_df = mfl.get_dataframe()
dl = daily_location("2016-01-01 18:00:00",
stop="2016-01-02 06:00:00",
method="most-common")
dl_df = dl.get_dataframe()
self.assertTrue((dl_df == mfl_df).all().all())
