def test_persistent_tables(self):
with Session() as session1:
session1 = Session()
session1.run_script('t = None')
t = session1.empty_table(10)
session1.bind_table('t', t)
with Session(sync_fields=SYNC_ONCE) as session2:
self.assertIn('t', session2.tables)
def test_never_timeout(self):
session = Session()
for _ in range(2):
token1 = session.session_token
sleep(300)
token2 = session.session_token
self.assertNotEqual(token1, token2)
session.close()
def test_shared_tables(self):
session1 = Session(sync_fields=SYNC_REPEATED)
session1.run_script('t = None')
session2 = Session()
t = session2.empty_table(10)
session2.bind_table('t', t)
@timeout_decorator.timeout(seconds=1)
def wait_for_table():
while 't' not in session1.tables:
pass
try:
wait_for_table()
except timeout_decorator.TimeoutError:
self.fail('table did not get synced to session1')
def main():
with Session(host="localhost", port=10000) as dh_session:
taxi_data_table = import_taxi_records(dh_session)
bottom_5_fares_table = demo_chained_table_ops(taxi_data_table)
# download the table to the client in the form of pyarrow table and convert it into a Pandas DataFrame
snapshot_data = bottom_5_fares_table.snapshot()
df = snapshot_data.to_pandas()
pd.set_option("max_columns", 20)
print(df)
def main():
with Session(host="localhost", port=10000) as dh_session:
taxi_data_table = import_taxi_records(dh_session)
variable_name = "t"
dh_session.bind_table(taxi_data_table, variable_name)
bottom_5_fares_table = run_script(dh_session=dh_session)
snapshot_data = bottom_5_fares_table.snapshot()
df = snapshot_data.to_pandas()
pd.set_option("max_columns", 20)
print(df)
def test_merge_tables(self):
session = Session()
pa_table = csv.read_csv(self.csv_file)
table1 = session.import_table(pa_table)
table2 = table1.group_by(by=["a", "c"]).ungroup(cols=["b", "d", "e"])
table3 = table1.where(["a % 2 > 0 && b % 3 == 1"])
result_table = session.merge_tables(tables=[table1, table2, table3],
order_by="a")
self.assertTrue(result_table.size > table1.size)
self.assertTrue(result_table.size > table2.size)
self.assertTrue(result_table.size > table3.size)
def main():
with Session(host="localhost", port=10000) as dh_session:
taxi_data_table = import_taxi_records(dh_session)
top_5_fares_table = demo_query(dh_session=dh_session, taxi_data_table=taxi_data_table)
bottom_5_fares_table = demo_chained_table_ops(taxi_data_table)
combined_fares_table = dh_session.merge_tables(tables=[top_5_fares_table, bottom_5_fares_table])
snapshot_data = combined_fares_table.snapshot()
df = snapshot_data.to_pandas()
pd.set_option("max_columns", 20)
print(df)
def test_multiple_sessions(self):
sessions = []
for i in range(100):
sessions.append(Session())
tables = []
for session in sessions:
pa_table = csv.read_csv(self.csv_file)
table1 = session.import_table(pa_table)
table2 = table1.group_by()
self.assertEqual(table2.size, 1)
tables.append(table1)
for i, table in enumerate(tables[:-1]):
j_table = table.natural_join(tables[i + 1],
on=["a", "b", "c", "d", "e"])
self.assertEqual(table.size, j_table.size)
for session in sessions:
session.close()
def setUp(self) -> None:
self.session = Session()
def main():
with Session(host="localhost", port=10000) as dh_session:
joined_table = demo_asof_join(dh_session)
df = joined_table.snapshot().to_pandas()
print(df)
def test_time_table(self):
session = Session()
t = session.time_table(period=100000)
self.assertFalse(t.is_static)
session.close()
def test_empty_table(self):
session = Session()
t = session.empty_table(1000)
self.assertEqual(t.size, 1000)
session.close()
def test_connect_failure(self):
with self.assertRaises(DHError):
session = Session(port=80)
def test_close(self):
session = Session()
session.close()
self.assertEqual(False, session.is_connected)
self.assertEqual(False, session.is_alive)
