mirror of
https://github.com/hanzoai/datastore.git
synced 2026-09-26 13:55:07 +00:00
Nikolai Kochetov
85937b3981
When set in the server config, this option overrides the session setting `allow_experimental_analyzer` for `MutationsInterpreter`. Absent (default) means the session setting is used. The override is parsed once per config reload in `Server.cpp` and stored as a tri-state atomic on the shared context, so each mutation only pays an atomic load instead of re-reading the config.
98 lines
2.7 KiB
Python
98 lines
2.7 KiB
Python
#!/usr/bin/env python3
|
|
|
|
"""
|
|
Verifies the server-side `<use_analyzer_for_mutations>` config option.
|
|
|
|
`MutationsInterpreter` emits `Will use old analyzer to prepare mutation` at
|
|
`test` level only when the old path is taken. The integration test captures
|
|
that signal from `system.text_log`:
|
|
* with the override set to `true`, the message must not appear;
|
|
* with the override set to `false`, the message must appear.
|
|
"""
|
|
|
|
import pytest
|
|
|
|
from helpers.cluster import ClickHouseCluster
|
|
|
|
cluster = ClickHouseCluster(__file__)
|
|
|
|
node = cluster.add_instance(
|
|
"node",
|
|
main_configs=[
|
|
"configs/use_analyzer_for_mutations.xml",
|
|
"configs/text_log.xml",
|
|
],
|
|
stay_alive=True,
|
|
)
|
|
|
|
OLD_ANALYZER_LOG_MESSAGE = "Will use old analyzer to prepare mutation"
|
|
CONFIG_PATH = "/etc/clickhouse-server/config.d/use_analyzer_for_mutations.xml"
|
|
|
|
|
|
@pytest.fixture(scope="module")
|
|
def start_cluster():
|
|
try:
|
|
cluster.start()
|
|
yield cluster
|
|
finally:
|
|
cluster.shutdown()
|
|
|
|
|
|
def run_mutation(table_name: str) -> None:
|
|
node.query(f"DROP TABLE IF EXISTS {table_name} SYNC")
|
|
node.query(
|
|
f"CREATE TABLE {table_name} (id UInt64, v UInt64) ENGINE = MergeTree ORDER BY id"
|
|
)
|
|
node.query(f"INSERT INTO {table_name} VALUES (1, 10), (2, 20), (3, 30)")
|
|
node.query(
|
|
f"ALTER TABLE {table_name} UPDATE v = v + 1 WHERE id = 2 SETTINGS mutations_sync = 2"
|
|
)
|
|
|
|
|
|
def count_old_analyzer_log_lines(table_name: str) -> int:
|
|
node.query("SYSTEM FLUSH LOGS")
|
|
result = node.query(
|
|
f"""
|
|
SELECT count()
|
|
FROM system.text_log
|
|
WHERE logger_name = 'MutationsInterpreter(default.{table_name})'
|
|
AND message = '{OLD_ANALYZER_LOG_MESSAGE}'
|
|
"""
|
|
).strip()
|
|
return int(result)
|
|
|
|
|
|
def set_override(value: str) -> None:
|
|
"""Rewrite the override config to `<use_analyzer_for_mutations>{value}</...>` and reload."""
|
|
node.exec_in_container(
|
|
[
|
|
"bash",
|
|
"-c",
|
|
(
|
|
"cat > {path} <<'EOF'\n"
|
|
"<clickhouse>\n"
|
|
" <use_analyzer_for_mutations>{value}</use_analyzer_for_mutations>\n"
|
|
"</clickhouse>\n"
|
|
"EOF"
|
|
).format(path=CONFIG_PATH, value=value),
|
|
]
|
|
)
|
|
node.query("SYSTEM RELOAD CONFIG")
|
|
|
|
|
|
def test_override_forces_new_analyzer(start_cluster):
|
|
set_override("true")
|
|
table = "t_override_true"
|
|
run_mutation(table)
|
|
assert count_old_analyzer_log_lines(table) == 0
|
|
|
|
|
|
def test_override_forces_old_analyzer(start_cluster):
|
|
set_override("false")
|
|
try:
|
|
table = "t_override_false"
|
|
run_mutation(table)
|
|
assert count_old_analyzer_log_lines(table) >= 1
|
|
finally:
|
|
set_override("true")
|