mirror of
https://github.com/ClickHouse/ClickHouse.git
synced 2024-11-27 18:12:02 +00:00
b75963d370
This PR formats all the `*.py` files found under the `tests/integration` folder. It also reorders the imports and cleans up a bunch of unused imports. The formatting also takes care of other things like wrapping lines and fixing spaces and indents such that the tests look more readable.
79 lines
3.1 KiB
Python
79 lines
3.1 KiB
Python
import pytest
|
|
from helpers.client import QueryRuntimeException
|
|
from helpers.cluster import ClickHouseCluster
|
|
|
|
cluster = ClickHouseCluster(__file__)
|
|
ch1 = cluster.add_instance('ch1', main_configs=["configs/config.d/clusters.xml", "configs/config.d/ddl.xml"],
|
|
with_zookeeper=True)
|
|
ch2 = cluster.add_instance('ch2', main_configs=["configs/config.d/clusters.xml", "configs/config.d/ddl.xml"],
|
|
with_zookeeper=True)
|
|
ch3 = cluster.add_instance('ch3', main_configs=["configs/config.d/clusters.xml", "configs/config.d/ddl.xml"],
|
|
with_zookeeper=True)
|
|
ch4 = cluster.add_instance('ch4', main_configs=["configs/config.d/clusters.xml", "configs/config.d/ddl.xml"],
|
|
with_zookeeper=True)
|
|
|
|
|
|
@pytest.fixture(scope="module")
|
|
def started_cluster():
|
|
try:
|
|
cluster.start()
|
|
ch1.query(
|
|
"CREATE TABLE sometbl ON CLUSTER 'cluster' (key UInt64, value String) ENGINE = MergeTree ORDER by key")
|
|
yield cluster
|
|
|
|
finally:
|
|
cluster.shutdown()
|
|
|
|
|
|
def test_dictionary_ddl_on_cluster(started_cluster):
|
|
for node in [ch1, ch2, ch3, ch4]:
|
|
assert node.query("SELECT count() from sometbl") == "0\n"
|
|
|
|
for num, node in enumerate([ch1, ch2, ch3, ch4]):
|
|
node.query("insert into sometbl values ({}, '{}')".format(num, node.name))
|
|
|
|
ch1.query(
|
|
"""
|
|
CREATE DICTIONARY somedict ON CLUSTER 'cluster' (
|
|
key UInt64,
|
|
value String
|
|
)
|
|
PRIMARY KEY key
|
|
LAYOUT(FLAT())
|
|
SOURCE(CLICKHOUSE(HOST 'localhost' PORT 9000 USER 'default' TABLE 'sometbl' DB 'default'))
|
|
LIFETIME(10)
|
|
""")
|
|
|
|
for num, node in enumerate([ch1, ch2, ch3, ch4]):
|
|
assert node.query("SELECT count() from sometbl") == "1\n"
|
|
assert node.query(
|
|
"SELECT dictGetString('default.somedict', 'value', toUInt64({}))".format(num)) == node.name + '\n'
|
|
|
|
ch1.query("DETACH DICTIONARY default.somedict ON CLUSTER 'cluster'")
|
|
|
|
for node in [ch1, ch2, ch3, ch4]:
|
|
with pytest.raises(QueryRuntimeException):
|
|
node.query("SELECT dictGetString('default.somedict', 'value', toUInt64(1))")
|
|
|
|
ch1.query("ATTACH DICTIONARY default.somedict ON CLUSTER 'cluster'")
|
|
|
|
for num, node in enumerate([ch1, ch2, ch3, ch4]):
|
|
assert node.query("SELECT count() from sometbl") == "1\n"
|
|
assert node.query(
|
|
"SELECT dictGetString('default.somedict', 'value', toUInt64({}))".format(num)) == node.name + '\n'
|
|
|
|
for num, node in enumerate([ch1, ch2, ch3, ch4]):
|
|
node.query("ALTER TABLE sometbl UPDATE value = 'new_key' WHERE 1")
|
|
|
|
ch1.query("SYSTEM RELOAD DICTIONARY ON CLUSTER 'cluster' `default.somedict`")
|
|
|
|
for num, node in enumerate([ch1, ch2, ch3, ch4]):
|
|
assert node.query(
|
|
"SELECT dictGetString('default.somedict', 'value', toUInt64({}))".format(num)) == 'new_key' + '\n'
|
|
|
|
ch1.query("DROP DICTIONARY default.somedict ON CLUSTER 'cluster'")
|
|
|
|
for node in [ch1, ch2, ch3, ch4]:
|
|
with pytest.raises(QueryRuntimeException):
|
|
node.query("SELECT dictGetString('default.somedict', 'value', toUInt64(1))")
|