Skip to content

test: assert ClickHouse overwrite SQL #6149

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
mday-io wants to merge 1 commit into
base: main
Choose a base branch
from
Open

test: assert ClickHouse overwrite SQL #6149

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
106 changes: 81 additions & 25 deletions tests/core/engine_adapter/test_clickhouse.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1118,7 +1118,7 @@ def test_insert_overwrite_by_condition_replace(
columns_to_types,
)

to_sql_calls(adapter) == [
assert to_sql_calls(adapter) == [
'CREATE TABLE "__temp_target_abcd" AS "__temp_existing_records_abcd"',
'INSERT INTO "__temp_target_abcd" ("id", "ds") SELECT "id", "ds" FROM (SELECT * FROM "__temp_new_records_abcd") AS "_subquery"',
'EXCHANGE TABLES "__temp_existing_records_abcd" AND "__temp_target_abcd"',
Expand All @@ -1133,15 +1133,18 @@ def test_insert_overwrite_by_condition_where_partitioned(

temp_table_mock = mocker.patch("sqlmesh.core.engine_adapter.EngineAdapter._get_temp_table")
table_name = "target"
temp_table_mock.return_value = make_temp_table_name(table_name, "abcd")
temp_table_mock.side_effect = lambda table, *args, **kwargs: make_temp_table_name(
"affected_partitions" if exp.to_table(table).name == "_affected_partitions" else table_name,
"abcd",
)

fetchone_mock = mocker.patch("sqlmesh.core.engine_adapter.ClickhouseEngineAdapter.fetchone")
fetchone_mock.return_value = "dateTrunc('WEEK', ds)"

fetchall_mock = mocker.patch("sqlmesh.core.engine_adapter.ClickhouseEngineAdapter.fetchall")
fetchall_mock.side_effect = [
[("1",), ("2",), ("3",), ("4",)],
["1", "2", "4"],
[("1",), ("2",), ("4",)],
]

insert_table_name = make_temp_table_name("new_records", "abcd")
Expand All @@ -1167,15 +1170,29 @@ def test_insert_overwrite_by_condition_where_partitioned(
),
)

to_sql_calls(adapter) == [
sql_calls = to_sql_calls(adapter)

assert len(sql_calls) == 7
assert sql_calls[:5] == [
'CREATE TABLE "__temp_target_abcd" AS "__temp_existing_records_abcd"',
"""INSERT INTO "__temp_target_abcd" ("id", "ds") SELECT "id", "ds" FROM (SELECT * FROM "__temp_new_records_abcd") AS "_subquery" WHERE "ds" BETWEEN '2024-02-15' AND '2024-04-30'""",
"""CREATE TABLE IF NOT EXISTS "__temp_target_abcd" ENGINE=MergeTree ORDER BY () AS SELECT DISTINCT "partition_id" FROM (SELECT "_partition_id" AS "partition_id" FROM "__temp_existing_records_abcd" WHERE "ds" BETWEEN '2024-02-15' AND '2024-04-30' UNION DISTINCT SELECT "_partition_id" AS "partition_id" FROM "__temp_target_abcd") AS "_affected_partitions\"""",
"""INSERT INTO "__temp_target_abcd" SELECT "id", "ds" FROM "__temp_existing_records_abcd" WHERE NOT ("ds" BETWEEN '2024-02-15' AND '2024-04-30') AND "_partition_id" IN (SELECT "partition_id" FROM "__temp_target_abcd")""",
"""ALTER TABLE "__temp_existing_records_abcd" REPLACE PARTITION ID '1' FROM "__temp_target_abcd", REPLACE PARTITION ID '2' FROM "__temp_target_abcd", REPLACE PARTITION ID '4' FROM "__temp_target_abcd", DROP PARTITION ID '3'""",
'DROP TABLE IF EXISTS "__temp_target_abcd"',
'''CREATE TABLE IF NOT EXISTS "__temp_affected_partitions_abcd" ENGINE=MergeTree ORDER BY () AS SELECT DISTINCT "partition_id" FROM (SELECT "_partition_id" AS "partition_id" FROM "__temp_existing_records_abcd" WHERE "ds" BETWEEN '2024-02-15' AND '2024-04-30' UNION DISTINCT SELECT "_partition_id" AS "partition_id" FROM "__temp_target_abcd") AS "_affected_partitions"''',
"""INSERT INTO "__temp_target_abcd" SELECT "id", "ds" FROM "__temp_existing_records_abcd" WHERE NOT ("ds" BETWEEN '2024-02-15' AND '2024-04-30') AND "_partition_id" IN (SELECT "partition_id" FROM "__temp_affected_partitions_abcd")""",
'DROP TABLE IF EXISTS "__temp_affected_partitions_abcd"',
]

alter_prefix = 'ALTER TABLE "__temp_existing_records_abcd" '
assert sql_calls[5].startswith(alter_prefix)
assert sorted(sql_calls[5].removeprefix(alter_prefix).split(", ")) == sorted(
[
"REPLACE PARTITION ID '1' FROM \"__temp_target_abcd\"",
"REPLACE PARTITION ID '2' FROM \"__temp_target_abcd\"",
"REPLACE PARTITION ID '4' FROM \"__temp_target_abcd\"",
"DROP PARTITION ID '3'",
]
)
assert sql_calls[6] == 'DROP TABLE IF EXISTS "__temp_target_abcd"'


def test_insert_overwrite_by_condition_by_key(
make_mocked_engine_adapter: t.Callable, mocker: MockerFixture, make_temp_table_name: t.Callable
Expand Down Expand Up @@ -1218,15 +1235,15 @@ def test_insert_overwrite_by_condition_by_key(
dynamic_key_unique=False,
)

to_sql_calls(adapter) == [
assert to_sql_calls(adapter) == [
'CREATE TABLE "__temp_target_abcd" AS "__temp_existing_records_abcd"',
'INSERT INTO "__temp_target_abcd" ("id", "ds") SELECT "id", "ds" FROM (SELECT DISTINCT ON ("id") * FROM "__temp_new_records_abcd") AS "_subquery"',
'INSERT INTO "__temp_target_abcd" SELECT "id", "ds" FROM "__temp_existing_records_abcd" WHERE NOT ("id" IN (SELECT "id" FROM "__temp_target_abcd"))',
'EXCHANGE TABLES "__temp_existing_records_abcd" AND "__temp_target_abcd"',
'DROP TABLE IF EXISTS "__temp_target_abcd"',
'CREATE TABLE "__temp_target_abcd" AS "__temp_existing_records_abcd"',
'INSERT INTO "__temp_target_abcd" ("id", "ds") SELECT "id", "ds" FROM (SELECT * FROM "__temp_new_records_abcd") AS "_subquery"',
'INSERT INTO "__temp_target_abcd" SELECT "id", "ds" FROM "__temp_existing_records_abcd" WHERE NOT ("id" IN (SELECT "id" FROM "__temp_target_abcd"))',
'INSERT INTO "__temp_target_abcd" SELECT "id", "ds" FROM "__temp_existing_records_abcd" WHERE NOT ("id" IN (SELECT DISTINCT "id" FROM "__temp_target_abcd"))',
'EXCHANGE TABLES "__temp_existing_records_abcd" AND "__temp_target_abcd"',
'DROP TABLE IF EXISTS "__temp_target_abcd"',
]
Expand All @@ -1239,17 +1256,20 @@ def test_insert_overwrite_by_condition_by_key_partitioned(

temp_table_mock = mocker.patch("sqlmesh.core.engine_adapter.EngineAdapter._get_temp_table")
table_name = "target"
temp_table_mock.return_value = make_temp_table_name(table_name, "abcd")
temp_table_mock.side_effect = lambda table, *args, **kwargs: make_temp_table_name(
"affected_partitions" if exp.to_table(table).name == "_affected_partitions" else table_name,
"abcd",
)

fetchone_mock = mocker.patch("sqlmesh.core.engine_adapter.ClickhouseEngineAdapter.fetchone")
fetchone_mock.side_effect = ["dateTrunc('WEEK', ds)", "dateTrunc('WEEK', ds)"]

fetchall_mock = mocker.patch("sqlmesh.core.engine_adapter.ClickhouseEngineAdapter.fetchall")
fetchall_mock.side_effect = [
[("1",), ("2",), ("3",), ("4",)],
["1", "2", "4"],
[("1",), ("2",), ("4",)],
[("1",), ("2",), ("3",), ("4",)],
["1", "2", "4"],
[("1",), ("2",), ("4",)],
]

insert_table_name = make_temp_table_name("new_records", "abcd")
Expand Down Expand Up @@ -1281,20 +1301,45 @@ def test_insert_overwrite_by_condition_by_key_partitioned(
dynamic_key_unique=False,
)

to_sql_calls(adapter) == [
sql_calls = to_sql_calls(adapter)

assert len(sql_calls) == 14
assert sql_calls[:4] == [
'CREATE TABLE "__temp_target_abcd" AS "__temp_existing_records_abcd"',
'INSERT INTO "__temp_target_abcd" ("id", "ds") SELECT "id", "ds" FROM (SELECT DISTINCT ON ("id") * FROM "__temp_new_records_abcd") AS "_subquery"',
'CREATE TABLE IF NOT EXISTS "__temp_target_abcd" ENGINE=MergeTree ORDER BY () AS SELECT DISTINCT "partition_id" FROM (SELECT "_partition_id" AS "partition_id" FROM "__temp_existing_records_abcd" WHERE "id" IN (SELECT "id" FROM "__temp_target_abcd") UNION DISTINCT SELECT "_partition_id" AS "partition_id" FROM "__temp_target_abcd") AS "_affected_partitions"',
'INSERT INTO "__temp_target_abcd" SELECT "id", "ds" FROM "__temp_existing_records_abcd" WHERE NOT ("id" IN (SELECT "id" FROM "__temp_target_abcd")) AND "_partition_id" IN (SELECT "partition_id" FROM "__temp_target_abcd")',
"""ALTER TABLE "__temp_existing_records_abcd" REPLACE PARTITION ID '2' FROM "__temp_target_abcd", REPLACE PARTITION ID '1' FROM "__temp_target_abcd", REPLACE PARTITION ID '4' FROM "__temp_target_abcd", DROP PARTITION ID '3'""",
'DROP TABLE IF EXISTS "__temp_target_abcd"',
'CREATE TABLE IF NOT EXISTS "__temp_affected_partitions_abcd" ENGINE=MergeTree ORDER BY () AS SELECT DISTINCT "partition_id" FROM (SELECT "_partition_id" AS "partition_id" FROM "__temp_existing_records_abcd" WHERE "id" IN (SELECT "id" FROM "__temp_target_abcd") UNION DISTINCT SELECT "_partition_id" AS "partition_id" FROM "__temp_target_abcd") AS "_affected_partitions"',
'INSERT INTO "__temp_target_abcd" SELECT "id", "ds" FROM "__temp_existing_records_abcd" WHERE NOT ("id" IN (SELECT "id" FROM "__temp_target_abcd")) AND "_partition_id" IN (SELECT "partition_id" FROM "__temp_affected_partitions_abcd")',
]

assert sql_calls[4] == 'DROP TABLE IF EXISTS "__temp_affected_partitions_abcd"'
alter_prefix = 'ALTER TABLE "__temp_existing_records_abcd" '
assert sql_calls[5].startswith(alter_prefix)
assert sorted(sql_calls[5].removeprefix(alter_prefix).split(", ")) == sorted(
[
"REPLACE PARTITION ID '1' FROM \"__temp_target_abcd\"",
"REPLACE PARTITION ID '2' FROM \"__temp_target_abcd\"",
"REPLACE PARTITION ID '4' FROM \"__temp_target_abcd\"",
"DROP PARTITION ID '3'",
]
)
assert sql_calls[6] == 'DROP TABLE IF EXISTS "__temp_target_abcd"'
assert sql_calls[7:11] == [
'CREATE TABLE "__temp_target_abcd" AS "__temp_existing_records_abcd"',
'INSERT INTO "__temp_target_abcd" ("id", "ds") SELECT "id", "ds" FROM (SELECT * FROM "__temp_new_records_abcd") AS "_subquery"',
'CREATE TABLE IF NOT EXISTS "__temp_target_abcd" ENGINE=MergeTree ORDER BY () AS SELECT DISTINCT "partition_id" FROM (SELECT "_partition_id" AS "partition_id" FROM "__temp_existing_records_abcd" WHERE "id" IN (SELECT "id" FROM "__temp_target_abcd") UNION DISTINCT SELECT "_partition_id" AS "partition_id" FROM "__temp_target_abcd") AS "_affected_partitions"',
'INSERT INTO "__temp_target_abcd" SELECT "id", "ds" FROM "__temp_existing_records_abcd" WHERE NOT ("id" IN (SELECT "id" FROM "__temp_target_abcd")) AND "_partition_id" IN (SELECT "partition_id" FROM "__temp_target_abcd")',
"""ALTER TABLE "__temp_existing_records_abcd" REPLACE PARTITION ID '2' FROM "__temp_target_abcd", REPLACE PARTITION ID '1' FROM "__temp_target_abcd", REPLACE PARTITION ID '4' FROM "__temp_target_abcd", DROP PARTITION ID '3'""",
'DROP TABLE IF EXISTS "__temp_target_abcd"',
'CREATE TABLE IF NOT EXISTS "__temp_affected_partitions_abcd" ENGINE=MergeTree ORDER BY () AS SELECT DISTINCT "partition_id" FROM (SELECT "_partition_id" AS "partition_id" FROM "__temp_existing_records_abcd" WHERE "id" IN (SELECT DISTINCT "id" FROM "__temp_target_abcd") UNION DISTINCT SELECT "_partition_id" AS "partition_id" FROM "__temp_target_abcd") AS "_affected_partitions"',
'INSERT INTO "__temp_target_abcd" SELECT "id", "ds" FROM "__temp_existing_records_abcd" WHERE NOT ("id" IN (SELECT DISTINCT "id" FROM "__temp_target_abcd")) AND "_partition_id" IN (SELECT "partition_id" FROM "__temp_affected_partitions_abcd")',
]
assert sql_calls[11] == 'DROP TABLE IF EXISTS "__temp_affected_partitions_abcd"'
assert sql_calls[12].startswith(alter_prefix)
assert sorted(sql_calls[12].removeprefix(alter_prefix).split(", ")) == sorted(
[
"REPLACE PARTITION ID '1' FROM \"__temp_target_abcd\"",
"REPLACE PARTITION ID '2' FROM \"__temp_target_abcd\"",
"REPLACE PARTITION ID '4' FROM \"__temp_target_abcd\"",
"DROP PARTITION ID '3'",
]
)
assert sql_calls[13] == 'DROP TABLE IF EXISTS "__temp_target_abcd"'


def test_insert_overwrite_by_condition_inc_by_partition(
Expand Down Expand Up @@ -1330,12 +1375,23 @@ def test_insert_overwrite_by_condition_inc_by_partition(
keep_existing_partition_rows=False,
)

to_sql_calls(adapter) == [
sql_calls = to_sql_calls(adapter)

assert len(sql_calls) == 4
assert sql_calls[:2] == [
'CREATE TABLE "__temp_target_abcd" AS "__temp_existing_records_abcd"',
'INSERT INTO "__temp_target_abcd" ("id", "ds") SELECT "id", "ds" FROM (SELECT * FROM "__temp_new_records_abcd") AS "_subquery"',
"""ALTER TABLE "__temp_existing_records_abcd" REPLACE PARTITION ID '1' FROM "__temp_target_abcd", REPLACE PARTITION ID '2' FROM "__temp_target_abcd", REPLACE PARTITION ID '4' FROM "__temp_target_abcd\"""",
'DROP TABLE IF EXISTS "__temp_target_abcd"',
]
alter_prefix = 'ALTER TABLE "__temp_existing_records_abcd" '
assert sql_calls[2].startswith(alter_prefix)
assert sorted(sql_calls[2].removeprefix(alter_prefix).split(", ")) == sorted(
[
"REPLACE PARTITION ID '1' FROM \"__temp_target_abcd\"",
"REPLACE PARTITION ID '2' FROM \"__temp_target_abcd\"",
"REPLACE PARTITION ID '4' FROM \"__temp_target_abcd\"",
]
)
assert sql_calls[3] == 'DROP TABLE IF EXISTS "__temp_target_abcd"'


def test_to_time_column():
Expand Down
Loading