dbcli/sqlglot
1
0
Fork 0
Code Issues Activity

Merging upstream version 26.14.0.

Browse source 
Signed-off-by: Daniel Baumann <daniel@debian.org>
This commit is contained in:
Daniel Baumann 2025-04-16 09:04:43 +02:00
parent 68f1150572
commit e9f53ab285
Signed by: daniel
GPG key ID: FBB4F0E80A80222F
84 changed files with 63872 additions and 61909 deletions
Download patch file Download diff file Expand all files Collapse all files

6
tests/dialects/test_athena.py
View file

@ -154,6 +154,12 @@ class TestAthena(Validator):
write_sql='CREATE TABLE "foo" AS WITH "foo" AS (SELECT "a", "b" FROM "bar") SELECT * FROM "foo"',
)
# CTAS with Union should still hit the Trino engine and not Hive
self.validate_identity(
'CREATE TABLE `foo` AS WITH `foo` AS (SELECT "a", `b` FROM "bar") SELECT * FROM "foo" UNION SELECT * FROM "foo"',
write_sql='CREATE TABLE "foo" AS WITH "foo" AS (SELECT "a", "b" FROM "bar") SELECT * FROM "foo" UNION SELECT * FROM "foo"',
)
self.validate_identity("DESCRIBE foo.bar", write_sql="DESCRIBE `foo`.`bar`", identify=True)
def test_dml_quoting(self):

48
tests/dialects/test_bigquery.py
View file

@ -1404,11 +1404,13 @@ LANGUAGE js AS
)
self.validate_all(
"TO_JSON_STRING(x)",
read={"bigquery": "TO_JSON_STRING(x)"},
read={
"bigquery": "TO_JSON_STRING(x)",
},
write={
"bigquery": "TO_JSON_STRING(x)",
"duckdb": "CAST(TO_JSON(x) AS TEXT)",
"presto": "JSON_FORMAT(x)",
"presto": "JSON_FORMAT(CAST(x AS JSON))",
"spark": "TO_JSON(x)",
},
)
@ -1486,7 +1488,7 @@ WHERE
"snowflake": "SELECT LENGTH(foo)",
},
write={
"duckdb": "SELECT CASE TYPEOF(foo) WHEN 'VARCHAR' THEN LENGTH(CAST(foo AS TEXT)) WHEN 'BLOB' THEN OCTET_LENGTH(CAST(foo AS BLOB)) END",
"duckdb": "SELECT CASE TYPEOF(foo) WHEN 'BLOB' THEN OCTET_LENGTH(CAST(foo AS BLOB)) ELSE LENGTH(CAST(foo AS TEXT)) END",
"snowflake": "SELECT LENGTH(foo)",
"": "SELECT LENGTH(foo)",
},
@ -2382,3 +2384,43 @@ OPTIONS (
for select in annotated.selects:
self.assertEqual(select.type.sql("bigquery"), "TIMESTAMP")
def test_set_operations(self):
self.validate_identity("SELECT 1 AS foo INNER UNION ALL SELECT 3 AS foo, 4 AS bar")
for side in ("", " LEFT", " FULL"):
for kind in ("", " OUTER"):
for name in (
"",
" BY NAME",
" BY NAME ON (foo, bar)",
):
with self.subTest(f"Testing {side} {kind} {name} in test_set_operations"):
self.validate_identity(
f"SELECT 1 AS foo{side}{kind} UNION ALL{name} SELECT 3 AS foo, 4 AS bar",
)
self.validate_identity(
"SELECT 1 AS x UNION ALL CORRESPONDING SELECT 2 AS x",
"SELECT 1 AS x INNER UNION ALL BY NAME SELECT 2 AS x",
)
self.validate_identity(
"SELECT 1 AS x UNION ALL CORRESPONDING BY (foo, bar) SELECT 2 AS x",
"SELECT 1 AS x INNER UNION ALL BY NAME ON (foo, bar) SELECT 2 AS x",
)
self.validate_identity(
"SELECT 1 AS x LEFT UNION ALL CORRESPONDING SELECT 2 AS x",
"SELECT 1 AS x LEFT UNION ALL BY NAME SELECT 2 AS x",
)
self.validate_identity(
"SELECT 1 AS x UNION ALL STRICT CORRESPONDING SELECT 2 AS x",
"SELECT 1 AS x UNION ALL BY NAME SELECT 2 AS x",
)
self.validate_identity(
"SELECT 1 AS x UNION ALL STRICT CORRESPONDING BY (foo, bar) SELECT 2 AS x",
"SELECT 1 AS x UNION ALL BY NAME ON (foo, bar) SELECT 2 AS x",
)

50
tests/dialects/test_clickhouse.py
View file

@ -33,6 +33,7 @@ class TestClickhouse(Validator):
self.assertEqual(expr.sql(dialect="clickhouse"), "COUNT(x)")
self.assertIsNone(expr._meta)
self.validate_identity('SELECT DISTINCT ON ("id") * FROM t')
self.validate_identity("SELECT 1 OR (1 = 2)")
self.validate_identity("SELECT 1 AND (1 = 2)")
self.validate_identity("SELECT json.a.:Int64")
@ -105,8 +106,10 @@ class TestClickhouse(Validator):
self.validate_identity("SELECT * FROM table LIMIT 1 BY a, b")
self.validate_identity("SELECT * FROM table LIMIT 2 OFFSET 1 BY a, b")
self.validate_identity("TRUNCATE TABLE t1 ON CLUSTER test_cluster")
self.validate_identity("TRUNCATE TABLE t1 ON CLUSTER '{cluster}'")
self.validate_identity("TRUNCATE DATABASE db")
self.validate_identity("TRUNCATE DATABASE db ON CLUSTER test_cluster")
self.validate_identity("TRUNCATE DATABASE db ON CLUSTER '{cluster}'")
self.validate_identity(
"SELECT DATE_BIN(toDateTime('2023-01-01 14:45:00'), INTERVAL '1' MINUTE, toDateTime('2023-01-01 14:35:30'), 'UTC')",
)
@ -155,12 +158,21 @@ class TestClickhouse(Validator):
self.validate_identity(
"CREATE TABLE test ON CLUSTER default (id UInt8) ENGINE=AggregatingMergeTree() ORDER BY tuple()"
)
self.validate_identity(
"CREATE TABLE test ON CLUSTER '{cluster}' (id UInt8) ENGINE=AggregatingMergeTree() ORDER BY tuple()"
)
self.validate_identity(
"CREATE MATERIALIZED VIEW test_view ON CLUSTER cl1 (id UInt8) ENGINE=AggregatingMergeTree() ORDER BY tuple() AS SELECT * FROM test_data"
)
self.validate_identity(
"CREATE MATERIALIZED VIEW test_view ON CLUSTER '{cluster}' (id UInt8) ENGINE=AggregatingMergeTree() ORDER BY tuple() AS SELECT * FROM test_data"
)
self.validate_identity(
"CREATE MATERIALIZED VIEW test_view ON CLUSTER cl1 TO table1 AS SELECT * FROM test_data"
)
self.validate_identity(
"CREATE MATERIALIZED VIEW test_view ON CLUSTER '{cluster}' TO table1 AS SELECT * FROM test_data"
)
self.validate_identity(
"CREATE MATERIALIZED VIEW test_view TO db.table1 (id UInt8) AS SELECT * FROM test_data"
)
@ -184,7 +196,7 @@ class TestClickhouse(Validator):
)
self.validate_identity(
"INSERT INTO tab VALUES ({'key1': 1, 'key2': 10}), ({'key1': 2, 'key2': 20}), ({'key1': 3, 'key2': 30})",
"INSERT INTO tab VALUES (map('key1', 1, 'key2', 10)), (map('key1', 2, 'key2', 20)), (map('key1', 3, 'key2', 30))",
"INSERT INTO tab VALUES ((map('key1', 1, 'key2', 10))), ((map('key1', 2, 'key2', 20))), ((map('key1', 3, 'key2', 30)))",
)
self.validate_identity(
"SELECT (toUInt8('1') + toUInt8('2')) IS NOT NULL",
@ -507,11 +519,12 @@ class TestClickhouse(Validator):
"INSERT INTO FUNCTION s3('url', 'CSV', 'name String, value UInt32', 'gzip') SELECT name, value FROM existing_table"
)
self.validate_identity(
"INSERT INTO FUNCTION remote('localhost', default.simple_table) VALUES (100, 'inserted via remote()')"
"INSERT INTO FUNCTION remote('localhost', default.simple_table) VALUES (100, 'inserted via remote()')",
"INSERT INTO FUNCTION remote('localhost', default.simple_table) VALUES ((100), ('inserted via remote()'))",
)
self.validate_identity(
"""INSERT INTO TABLE FUNCTION hdfs('hdfs://hdfs1:9000/test', 'TSV', 'name String, column2 UInt32, column3 UInt32') VALUES ('test', 1, 2)""",
"""INSERT INTO FUNCTION hdfs('hdfs://hdfs1:9000/test', 'TSV', 'name String, column2 UInt32, column3 UInt32') VALUES ('test', 1, 2)""",
"""INSERT INTO FUNCTION hdfs('hdfs://hdfs1:9000/test', 'TSV', 'name String, column2 UInt32, column3 UInt32') VALUES (('test'), (1), (2))""",
)
self.validate_identity("SELECT 1 FORMAT TabSeparated")
@ -546,22 +559,23 @@ class TestClickhouse(Validator):
)
self.validate_identity("ALTER TABLE visits REPLACE PARTITION ID '201901' FROM visits_tmp")
self.validate_identity("ALTER TABLE visits ON CLUSTER test_cluster DROP COLUMN col1")
self.validate_identity("ALTER TABLE visits ON CLUSTER '{cluster}' DROP COLUMN col1")
self.validate_identity("DELETE FROM tbl ON CLUSTER test_cluster WHERE date = '2019-01-01'")
self.validate_identity("DELETE FROM tbl ON CLUSTER '{cluster}' WHERE date = '2019-01-01'")
self.assertIsInstance(
parse_one("Tuple(select Int64)", into=exp.DataType, read="clickhouse"), exp.DataType
)
self.validate_identity("INSERT INTO t (col1, col2) VALUES ('abcd', 1234)")
self.validate_identity(
"INSERT INTO t (col1, col2) VALUES ('abcd', 1234)",
"INSERT INTO t (col1, col2) VALUES (('abcd'), (1234))",
)
self.validate_all(
"INSERT INTO t (col1, col2) VALUES ('abcd', 1234)",
read={
# looks like values table function, but should be parsed as VALUES block
"clickhouse": "INSERT INTO t (col1, col2) values('abcd', 1234)"
},
write={
"clickhouse": "INSERT INTO t (col1, col2) VALUES ('abcd', 1234)",
"postgres": "INSERT INTO t (col1, col2) VALUES ('abcd', 1234)",
"clickhouse": "INSERT INTO t (col1, col2) VALUES (('abcd'), (1234))",
"postgres": "INSERT INTO t (col1, col2) VALUES (('abcd'), (1234))",
},
)
self.validate_identity("SELECT TRIM(TRAILING ')' FROM '( Hello, world! )')")
@ -591,6 +605,9 @@ class TestClickhouse(Validator):
self.validate_identity("SELECT arrayConcat([1, 2], [3, 4])")
def test_clickhouse_values(self):
ast = self.parse_one("SELECT * FROM VALUES (1, 2, 3)")
self.assertEqual(len(list(ast.find_all(exp.Tuple))), 4)
values = exp.select("*").from_(
exp.values([exp.tuple_(1, 2, 3)], alias="subq", columns=["a", "b", "c"])
)
@ -599,10 +616,18 @@ class TestClickhouse(Validator):
"SELECT * FROM (SELECT 1 AS a, 2 AS b, 3 AS c) AS subq",
)
self.validate_identity("INSERT INTO t (col1, col2) VALUES ('abcd', 1234)")
self.validate_identity("SELECT * FROM VALUES ((1, 1), (2, 1), (3, 1), (4, 1))")
self.validate_identity(
"SELECT type, id FROM VALUES ('id Int, type Int', (1, 1), (2, 1), (3, 1), (4, 1))"
)
self.validate_identity(
"INSERT INTO t (col1, col2) VALUES ('abcd', 1234)",
"INSERT INTO t (col1, col2) VALUES (('abcd'), (1234))",
)
self.validate_identity(
"INSERT INTO t (col1, col2) FORMAT Values('abcd', 1234)",
"INSERT INTO t (col1, col2) VALUES ('abcd', 1234)",
"INSERT INTO t (col1, col2) VALUES (('abcd'), (1234))",
)
self.validate_all(
@ -1171,6 +1196,7 @@ LIFETIME(MIN 0 MAX 0)""",
for creatable in ("DATABASE", "TABLE", "VIEW", "DICTIONARY", "FUNCTION"):
with self.subTest(f"Test DROP {creatable} ON CLUSTER"):
self.validate_identity(f"DROP {creatable} test ON CLUSTER test_cluster")
self.validate_identity(f"DROP {creatable} test ON CLUSTER '{{cluster}}'")
def test_datetime_funcs(self):
# Each datetime func has an alias that is roundtripped to the original name e.g. (DATE_SUB, DATESUB) -> DATE_SUB

18
tests/dialects/test_databricks.py
View file

@ -68,6 +68,24 @@ class TestDatabricks(Validator):
"FROM_UTC_TIMESTAMP(CAST(x AS TIMESTAMP), tz)",
)
self.validate_all(
"SELECT c1:item[1].price",
read={
"spark": "SELECT GET_JSON_OBJECT(c1, '$.item[1].price')",
},
write={
"databricks": "SELECT c1:item[1].price",
"spark": "SELECT GET_JSON_OBJECT(c1, '$.item[1].price')",
},
)
self.validate_all(
"SELECT GET_JSON_OBJECT(c1, '$.item[1].price')",
write={
"databricks": "SELECT c1:item[1].price",
"spark": "SELECT GET_JSON_OBJECT(c1, '$.item[1].price')",
},
)
self.validate_all(
"CREATE TABLE foo (x INT GENERATED ALWAYS AS (YEAR(y)))",
write={

7
tests/dialects/test_duckdb.py
View file

@ -400,6 +400,9 @@ class TestDuckDB(Validator):
self.validate_identity(
"SELECT * FROM (PIVOT Cities ON Year USING SUM(Population) GROUP BY Country) AS pivot_alias"
)
self.validate_identity(
"SELECT * FROM cities PIVOT(SUM(population) FOR year IN (2000, 2010, 2020) GROUP BY country)"
)
self.validate_identity(
# QUALIFY comes after WINDOW
"SELECT schema_name, function_name, ROW_NUMBER() OVER my_window AS function_rank FROM DUCKDB_FUNCTIONS() WINDOW my_window AS (PARTITION BY schema_name ORDER BY function_name) QUALIFY ROW_NUMBER() OVER my_window < 3"
@ -1595,3 +1598,7 @@ class TestDuckDB(Validator):
"SELECT (@-1) + 1",
"SELECT (ABS(-1)) + 1",
)
def test_show_tables(self):
self.validate_identity("SHOW TABLES").assert_is(exp.Show)
self.validate_identity("SHOW ALL TABLES").assert_is(exp.Show)

3
tests/dialects/test_postgres.py
View file

@ -76,6 +76,9 @@ class TestPostgres(Validator):
self.validate_identity("SELECT CURRENT_SCHEMA")
self.validate_identity("SELECT CURRENT_USER")
self.validate_identity("SELECT * FROM ONLY t1")
self.validate_identity(
"SELECT * FROM test_data, LATERAL JSONB_ARRAY_ELEMENTS(data) WITH ORDINALITY AS elem(value, ordinality)"
)
self.validate_identity(
"SELECT id, name FROM xml_data AS t, XMLTABLE('/root/user' PASSING t.xml COLUMNS id INT PATH '@id', name TEXT PATH 'name/text()') AS x"
)

3
tests/dialects/test_presto.py
View file

@ -1021,9 +1021,6 @@ class TestPresto(Validator):
)
self.validate_all(
"JSON_FORMAT(x)",
read={
"spark": "TO_JSON(x)",
},
write={
"bigquery": "TO_JSON_STRING(x)",
"duckdb": "CAST(TO_JSON(x) AS TEXT)",

48
tests/dialects/test_snowflake.py
View file

@ -1,6 +1,6 @@
from unittest import mock
from sqlglot import UnsupportedError, exp, parse_one
from sqlglot import UnsupportedError, exp, parse_one, ParseError
from sqlglot.optimizer.normalize_identifiers import normalize_identifiers
from sqlglot.optimizer.qualify_columns import quote_identifiers
from tests.dialects.test_dialect import Validator
@ -1481,9 +1481,30 @@ class TestSnowflake(Validator):
self.validate_identity("CREATE TAG cost_center ALLOWED_VALUES 'a', 'b'")
self.validate_identity("CREATE WAREHOUSE x").this.assert_is(exp.Identifier)
self.validate_identity("CREATE STREAMLIT x").this.assert_is(exp.Identifier)
self.validate_identity(
"CREATE TEMPORARY STAGE stage1 FILE_FORMAT=(TYPE=PARQUET)"
).this.assert_is(exp.Table)
self.validate_identity(
"CREATE STAGE stage1 FILE_FORMAT='format1'",
"CREATE STAGE stage1 FILE_FORMAT=(FORMAT_NAME='format1')",
)
self.validate_identity("CREATE STAGE stage1 FILE_FORMAT=(FORMAT_NAME=stage1.format1)")
self.validate_identity("CREATE STAGE stage1 FILE_FORMAT=(FORMAT_NAME='stage1.format1')")
self.validate_identity(
"CREATE STAGE stage1 FILE_FORMAT=schema1.format1",
"CREATE STAGE stage1 FILE_FORMAT=(FORMAT_NAME=schema1.format1)",
)
with self.assertRaises(ParseError):
self.parse_one("CREATE STAGE stage1 FILE_FORMAT=123", dialect="snowflake")
self.validate_identity(
"CREATE STAGE s1 URL='s3://bucket-123' FILE_FORMAT=(TYPE='JSON') CREDENTIALS=(aws_key_id='test' aws_secret_key='test')"
)
self.validate_identity(
"CREATE OR REPLACE TAG IF NOT EXISTS cost_center COMMENT='cost_center tag'"
).this.assert_is(exp.Identifier)
self.validate_identity(
"CREATE TEMPORARY FILE FORMAT fileformat1 TYPE=PARQUET COMPRESSION=auto"
).this.assert_is(exp.Table)
self.validate_identity(
"CREATE DYNAMIC TABLE product (pre_tax_profit, taxes, after_tax_profit) TARGET_LAG='20 minutes' WAREHOUSE=mywh AS SELECT revenue - cost, (revenue - cost) * tax_rate, (revenue - cost) * (1.0 - tax_rate) FROM staging_table"
)
@ -1499,9 +1520,6 @@ class TestSnowflake(Validator):
self.validate_identity(
"CREATE TABLE orders_clone_restore CLONE orders BEFORE (STATEMENT => '8e5d0ca9-005e-44e6-b858-a8f5b37c5726')"
)
self.validate_identity(
"CREATE TABLE a (x DATE, y BIGINT) PARTITION BY (x) integration='q' auto_refresh=TRUE file_format=(type = parquet)"
)
self.validate_identity(
"CREATE SCHEMA mytestschema_clone_restore CLONE testschema BEFORE (TIMESTAMP => TO_TIMESTAMP(40 * 365 * 86400))"
)
@ -1544,8 +1562,8 @@ class TestSnowflake(Validator):
partition by (col1,col2,col3)
location=@s2/logs/
partition_type = user_specified
file_format = (type = parquet)""",
"CREATE EXTERNAL TABLE et2 (col1 DATE AS (CAST(GET_PATH(PARSE_JSON(metadata$external_table_partition), 'COL1') AS DATE)), col2 VARCHAR AS (CAST(GET_PATH(PARSE_JSON(metadata$external_table_partition), 'COL2') AS VARCHAR)), col3 DECIMAL(38, 0) AS (CAST(GET_PATH(PARSE_JSON(metadata$external_table_partition), 'COL3') AS DECIMAL(38, 0)))) LOCATION @s2/logs/ PARTITION BY (col1, col2, col3) partition_type=user_specified file_format=(type = parquet)",
file_format = (type = parquet compression = gzip binary_as_text = false)""",
"CREATE EXTERNAL TABLE et2 (col1 DATE AS (CAST(GET_PATH(PARSE_JSON(metadata$external_table_partition), 'COL1') AS DATE)), col2 VARCHAR AS (CAST(GET_PATH(PARSE_JSON(metadata$external_table_partition), 'COL2') AS VARCHAR)), col3 DECIMAL(38, 0) AS (CAST(GET_PATH(PARSE_JSON(metadata$external_table_partition), 'COL3') AS DECIMAL(38, 0)))) PARTITION BY (col1, col2, col3) LOCATION=@s2/logs/ partition_type=user_specified FILE_FORMAT=(type=parquet compression=gzip binary_as_text=FALSE)",
)
self.validate_all(
@ -2092,6 +2110,15 @@ MATCH_RECOGNIZE (
self.assertEqual(ast.this, "DATABASES")
self.assertEqual(ast.args.get("scope_kind"), "ACCOUNT")
def test_show_file_formats(self):
self.validate_identity("SHOW FILE FORMATS")
self.validate_identity("SHOW FILE FORMATS LIKE 'foo' IN DATABASE db1")
self.validate_identity("SHOW FILE FORMATS LIKE 'foo' IN SCHEMA db1.schema1")
ast = parse_one("SHOW FILE FORMATS IN ACCOUNT", read="snowflake")
self.assertEqual(ast.this, "FILE FORMATS")
self.assertEqual(ast.args.get("scope_kind"), "ACCOUNT")
def test_show_functions(self):
self.validate_identity("SHOW FUNCTIONS")
self.validate_identity("SHOW FUNCTIONS LIKE 'foo' IN CLASS bla")
@ -2109,6 +2136,15 @@ MATCH_RECOGNIZE (
self.assertEqual(ast.this, "PROCEDURES")
self.assertEqual(ast.args.get("scope_kind"), "ACCOUNT")
def test_show_stages(self):
self.validate_identity("SHOW STAGES")
self.validate_identity("SHOW STAGES LIKE 'foo' IN DATABASE db1")
self.validate_identity("SHOW STAGES LIKE 'foo' IN SCHEMA db1.schema1")
ast = parse_one("SHOW STAGES IN ACCOUNT", read="snowflake")
self.assertEqual(ast.this, "STAGES")
self.assertEqual(ast.args.get("scope_kind"), "ACCOUNT")
def test_show_warehouses(self):
self.validate_identity("SHOW WAREHOUSES")
self.validate_identity("SHOW WAREHOUSES LIKE 'foo' WITH PRIVILEGES USAGE, MODIFY")

10
tests/dialects/test_spark.py
View file

@ -245,6 +245,7 @@ TBLPROPERTIES (
"REFRESH TABLE t",
)
self.validate_identity("IF(cond, foo AS bar, bla AS baz)")
self.validate_identity("any_value(col, true)", "ANY_VALUE(col) IGNORE NULLS")
self.validate_identity("first(col, true)", "FIRST(col) IGNORE NULLS")
self.validate_identity("first_value(col, true)", "FIRST_VALUE(col) IGNORE NULLS")
@ -307,6 +308,15 @@ TBLPROPERTIES (
"SELECT STR_TO_MAP('a:1,b:2,c:3')",
"SELECT STR_TO_MAP('a:1,b:2,c:3', ',', ':')",
)
self.validate_all(
"SELECT TO_JSON(STRUCT('blah' AS x)) AS y",
write={
"presto": "SELECT JSON_FORMAT(CAST(CAST(ROW('blah') AS ROW(x VARCHAR)) AS JSON)) AS y",
"spark": "SELECT TO_JSON(STRUCT('blah' AS x)) AS y",
"trino": "SELECT JSON_FORMAT(CAST(CAST(ROW('blah') AS ROW(x VARCHAR)) AS JSON)) AS y",
},
)
self.validate_all(
"SELECT TRY_ELEMENT_AT(ARRAY(1, 2, 3), 2)",
read={

4
tests/dialects/test_sqlite.py
View file

@ -109,6 +109,10 @@ class TestSQLite(Validator):
"SELECT * FROM station WHERE NOT city IS ''",
)
self.validate_identity("SELECT JSON_OBJECT('col1', 1, 'col2', '1')")
self.validate_identity(
'CREATE TABLE "foo t" ("foo t id" TEXT NOT NULL, PRIMARY KEY ("foo t id"))',
'CREATE TABLE "foo t" ("foo t id" TEXT NOT NULL PRIMARY KEY)',
)
def test_strftime(self):
self.validate_identity("SELECT STRFTIME('%Y/%m/%d', 'now')")

10
tests/dialects/test_tsql.py
View file

@ -17,6 +17,8 @@ class TestTSQL(Validator):
# tsql allows .. which means use the default schema
self.validate_identity("SELECT * FROM a..b")
self.validate_identity("GO").assert_is(exp.Command)
self.validate_identity("SELECT go").selects[0].assert_is(exp.Column)
self.validate_identity("CREATE view a.b.c", "CREATE VIEW b.c")
self.validate_identity("DROP view a.b.c", "DROP VIEW b.c")
self.validate_identity("ROUND(x, 1, 0)")
@ -559,6 +561,14 @@ class TestTSQL(Validator):
with self.assertRaises(ParseError, msg=f"When running '{query}'"):
self.parse_one(query)
self.validate_all(
"SELECT col FROM t OPTION(LABEL = 'foo')",
write={
"tsql": "SELECT col FROM t OPTION(LABEL = 'foo')",
"databricks": UnsupportedError,
},
)
def test_types(self):
self.validate_identity("CAST(x AS XML)")
self.validate_identity("CAST(x AS UNIQUEIDENTIFIER)")

18
tests/fixtures/optimizer/merge_subqueries.sql vendored
View file

@ -481,3 +481,21 @@ FROM (
LEFT OUTER JOIN tbl AS ITBL ON OTBL.id = ITBL.id
) AS ITBL;
WITH tbl AS (SELECT 1 AS id) SELECT OTBL.id AS id FROM tbl AS OTBL LEFT OUTER JOIN tbl AS ITBL_2 ON OTBL.id = ITBL_2.id LEFT OUTER JOIN tbl AS ITBL_3 ON OTBL.id = ITBL_3.id LEFT OUTER JOIN tbl AS ITBL ON OTBL.id = ITBL.id;
# title: Inner query contains subquery with an alias that conflicts with outer query
WITH i AS (
SELECT
a
FROM (
SELECT 1 a
) AS conflict
), j AS (
SELECT 1 AS a
)
SELECT
i.a,
conflict.a
FROM i
LEFT JOIN j AS conflict
ON i.a = conflict.a;
WITH j AS (SELECT 1 AS a) SELECT conflict_2.a AS a, conflict.a AS a FROM (SELECT 1 AS a) AS conflict_2 LEFT JOIN j AS conflict ON conflict_2.a = conflict.a;

10
tests/fixtures/optimizer/optimizer.sql vendored
View file

@ -760,10 +760,12 @@ SELECT
`_q_0`.`first_half_sales` AS `first_half_sales`,
`_q_0`.`second_half_sales` AS `second_half_sales`
FROM `produce` AS `produce`
UNPIVOT((`first_half_sales`, `second_half_sales`) FOR `semesters` IN (
(`produce`.`q1`, `produce`.`q2`) AS 'semester_1',
(`produce`.`q3`, `produce`.`q4`) AS 'semester_2'
)) AS `_q_0`;
UNPIVOT((`first_half_sales`, `second_half_sales`) FOR
`semesters` IN (
(`produce`.`q1`, `produce`.`q2`) AS 'semester_1',
(`produce`.`q3`, `produce`.`q4`) AS 'semester_2'
)
) AS `_q_0`;
# title: quoting is preserved
# dialect: snowflake

82
tests/fixtures/optimizer/qualify_columns.sql vendored
View file

@ -325,6 +325,65 @@ SELECT _q_0.a AS a FROM (SELECT x.a AS a FROM x AS x UNION SELECT x.a AS a FROM
((select a from x where a < 1)) UNION ((select a from x where a > 2));
((SELECT x.a AS a FROM x AS x WHERE x.a < 1)) UNION ((SELECT x.a AS a FROM x AS x WHERE x.a > 2));
# dialect: bigquery
# execute: false
SELECT * FROM (SELECT 1 AS foo, 2 AS bar INNER UNION ALL BY NAME SELECT 3 AS bar, 4 AS baz);
SELECT _q_0.bar AS bar FROM (SELECT 1 AS foo, 2 AS bar INNER UNION ALL BY NAME SELECT 3 AS bar, 4 AS baz) AS _q_0;
# dialect: bigquery
# execute: false
SELECT * FROM (SELECT 1 AS foo, 2 AS bar UNION ALL CORRESPONDING SELECT 3 AS bar, 4 AS baz);
SELECT _q_0.bar AS bar FROM (SELECT 1 AS foo, 2 AS bar INNER UNION ALL BY NAME SELECT 3 AS bar, 4 AS baz) AS _q_0;
# dialect: bigquery
# execute: false
SELECT * FROM (SELECT 1 AS foo, 2 AS bar LEFT UNION ALL BY NAME SELECT 3 AS bar, 4 AS baz);
SELECT _q_0.foo AS foo, _q_0.bar AS bar FROM (SELECT 1 AS foo, 2 AS bar LEFT UNION ALL BY NAME SELECT 3 AS bar, 4 AS baz) AS _q_0;
# dialect: bigquery
# execute: false
SELECT * FROM (SELECT 1 AS foo, 2 AS bar FULL UNION ALL BY NAME SELECT 3 AS bar, 4 AS baz);
SELECT _q_0.foo AS foo, _q_0.bar AS bar, _q_0.baz AS baz FROM (SELECT 1 AS foo, 2 AS bar FULL UNION ALL BY NAME SELECT 3 AS bar, 4 AS baz) AS _q_0;
# dialect: bigquery
# execute: false
SELECT * FROM (SELECT 1 AS foo, 2 AS bar LEFT UNION ALL CORRESPONDING SELECT 3 AS bar, 4 AS baz);
SELECT _q_0.foo AS foo, _q_0.bar AS bar FROM (SELECT 1 AS foo, 2 AS bar LEFT UNION ALL BY NAME SELECT 3 AS bar, 4 AS baz) AS _q_0;
# dialect: bigquery
# execute: false
SELECT * FROM (SELECT 1 AS foo, 2 AS bar FULL UNION ALL CORRESPONDING SELECT 3 AS bar, 4 AS baz);
SELECT _q_0.foo AS foo, _q_0.bar AS bar, _q_0.baz AS baz FROM (SELECT 1 AS foo, 2 AS bar FULL UNION ALL BY NAME SELECT 3 AS bar, 4 AS baz) AS _q_0;
# dialect: bigquery
# execute: false
SELECT * FROM (SELECT 1 AS foo, 2 AS bar FULL UNION ALL CORRESPONDING BY (foo, bar) SELECT 3 AS bar, 4 AS baz);
SELECT _q_0.foo AS foo, _q_0.bar AS bar FROM (SELECT 1 AS foo, 2 AS bar FULL UNION ALL BY NAME ON (foo, bar) SELECT 3 AS bar, 4 AS baz) AS _q_0;
# dialect: bigquery
# execute: false
SELECT * FROM (SELECT 1 AS foo, 2 AS bar FULL UNION ALL BY NAME ON (foo, bar) SELECT 3 AS bar, 4 AS baz);
SELECT _q_0.foo AS foo, _q_0.bar AS bar FROM (SELECT 1 AS foo, 2 AS bar FULL UNION ALL BY NAME ON (foo, bar) SELECT 3 AS bar, 4 AS baz) AS _q_0;
# dialect: bigquery
# execute: false
SELECT * FROM ((SELECT 1 AS foo, 2 AS bar LEFT UNION ALL BY NAME SELECT 3 AS bar, 4 AS baz) LEFT UNION ALL BY NAME ON (bar) SELECT 3 AS foo, 4 AS bar);
SELECT _q_0.bar AS bar FROM ((SELECT 1 AS foo, 2 AS bar LEFT UNION ALL BY NAME SELECT 3 AS bar, 4 AS baz) LEFT UNION ALL BY NAME ON (bar) SELECT 3 AS foo, 4 AS bar) AS _q_0;
# dialect: bigquery
# execute: false
SELECT * FROM ((SELECT 1 AS foo, 2 AS bar LEFT UNION ALL BY NAME SELECT 3 AS bar, 4 AS baz) FULL UNION ALL BY NAME ON (foo, qux) SELECT 3 AS qux, 4 AS bar);
SELECT _q_0.foo AS foo, _q_0.qux AS qux FROM ((SELECT 1 AS foo, 2 AS bar LEFT UNION ALL BY NAME SELECT 3 AS bar, 4 AS baz) FULL UNION ALL BY NAME ON (foo, qux) SELECT 3 AS qux, 4 AS bar) AS _q_0;
# dialect: bigquery
# execute: false
SELECT * FROM (((SELECT 1 AS foo, 2 AS bar LEFT UNION ALL BY NAME SELECT 3 AS bar, 4 AS baz) FULL UNION ALL BY NAME ON (foo, qux) SELECT 3 AS qux, 4 AS bar) INNER UNION ALL BY NAME ON (foo) SELECT 6 AS foo);
SELECT _q_0.foo AS foo FROM (((SELECT 1 AS foo, 2 AS bar LEFT UNION ALL BY NAME SELECT 3 AS bar, 4 AS baz) FULL UNION ALL BY NAME ON (foo, qux) SELECT 3 AS qux, 4 AS bar) INNER UNION ALL BY NAME ON (foo) SELECT 6 AS foo) AS _q_0;
--------------------------------------
-- Subqueries
--------------------------------------
@ -435,6 +494,29 @@ WITH tbl1 AS (SELECT STRUCT(1 AS col1, 2 AS col1) AS col) SELECT tbl1.col.* FROM
SELECT * FROM READ_CSV('file.csv');
SELECT * FROM READ_CSV('file.csv') AS _q_0;
# dialect: clickhouse
# Title: Expand tuples in VALUES using the structure provided
# execute: false
SELECT * FROM VALUES ('person String, place String', ('Noah', 'Paris'));
SELECT _q_0.person AS person, _q_0.place AS place FROM VALUES ('person String, place String', ('Noah', 'Paris')) AS _q_0(person, place);
# dialect: clickhouse
# Title: Expand tuples in VALUES using the default naming scheme in CH
# execute: false
SELECT * FROM VALUES ((1, 1), (2, 2));
SELECT _q_0.c1 AS c1, _q_0.c2 AS c2 FROM VALUES ((1, 1), (2, 2)) AS _q_0(c1, c2);
# dialect: clickhouse
# Title: Expand fields in VALUES using the default naming scheme in CH
# execute: false
SELECT * FROM VALUES (1, 2, 3);
SELECT _q_0.c1 AS c1 FROM VALUES ((1), (2), (3)) AS _q_0(c1);
# title: Expand PIVOT column combinations
# dialect: duckdb
WITH cities AS (SELECT * FROM (VALUES ('nl', 'amsterdam', 2000, 1005)) AS t(country, name, year, population)) SELECT * FROM cities PIVOT(SUM(population) AS total, COUNT(population) AS count FOR country IN ('nl', 'us') year IN (2000, 2010) name IN ('amsterdam', 'seattle'));
WITH cities AS (SELECT t.country AS country, t.name AS name, t.year AS year, t.population AS population FROM (VALUES ('nl', 'amsterdam', 2000, 1005)) AS t(country, name, year, population)) SELECT _q_0.nl_2000_amsterdam_total AS nl_2000_amsterdam_total, _q_0.nl_2000_amsterdam_count AS nl_2000_amsterdam_count, _q_0.nl_2000_seattle_total AS nl_2000_seattle_total, _q_0.nl_2000_seattle_count AS nl_2000_seattle_count, _q_0.nl_2010_amsterdam_total AS nl_2010_amsterdam_total, _q_0.nl_2010_amsterdam_count AS nl_2010_amsterdam_count, _q_0.nl_2010_seattle_total AS nl_2010_seattle_total, _q_0.nl_2010_seattle_count AS nl_2010_seattle_count, _q_0.us_2000_amsterdam_total AS us_2000_amsterdam_total, _q_0.us_2000_amsterdam_count AS us_2000_amsterdam_count, _q_0.us_2000_seattle_total AS us_2000_seattle_total, _q_0.us_2000_seattle_count AS us_2000_seattle_count, _q_0.us_2010_amsterdam_total AS us_2010_amsterdam_total, _q_0.us_2010_amsterdam_count AS us_2010_amsterdam_count, _q_0.us_2010_seattle_total AS us_2010_seattle_total, _q_0.us_2010_seattle_count AS us_2010_seattle_count FROM cities AS cities PIVOT(SUM(population) AS total, COUNT(population) AS count FOR country IN ('nl', 'us') year IN (2000, 2010) name IN ('amsterdam', 'seattle')) AS _q_0;
--------------------------------------
-- CTEs
--------------------------------------

5
tests/test_optimizer.py
View file

@ -55,13 +55,10 @@ def simplify(expression, **kwargs):
def annotate_functions(expression, **kwargs):
from sqlglot.dialects import Dialect
dialect = kwargs.get("dialect")
schema = kwargs.get("schema")
annotators = Dialect.get_or_raise(dialect).ANNOTATORS
annotated = annotate_types(expression, annotators=annotators, schema=schema)
annotated = annotate_types(expression, dialect=dialect, schema=schema)
return annotated.expressions[0]

62
tests/test_parser.py
View file

@ -646,6 +646,27 @@ class TestParser(unittest.TestCase):
) PIVOT (AVG("PrIcE"), MAX(quality) FOR partname IN ('prop' AS prop1, 'rudder'))
"""
two_in_clauses_duckdb = """
SELECT * FROM cities PIVOT (
sum(population) AS total,
count(population) AS count
FOR
year IN (2000, 2010)
country IN ('NL', 'US')
)
"""
three_in_clauses_duckdb = """
SELECT * FROM cities PIVOT (
sum(population) AS total,
count(population) AS count
FOR
year IN (2000, 2010)
country IN ('NL', 'US')
name IN ('Amsterdam', 'Seattle')
)
"""
query_to_column_names = {
nothing_aliased: {
"bigquery": ["prop", "rudder"],
@ -707,13 +728,48 @@ class TestParser(unittest.TestCase):
'"rudder_max(quality)"',
],
},
two_in_clauses_duckdb: {
"duckdb": [
'"2000_NL_total"',
'"2000_NL_count"',
'"2000_US_total"',
'"2000_US_count"',
'"2010_NL_total"',
'"2010_NL_count"',
'"2010_US_total"',
'"2010_US_count"',
],
},
three_in_clauses_duckdb: {
"duckdb": [
'"2000_NL_Amsterdam_total"',
'"2000_NL_Amsterdam_count"',
'"2000_NL_Seattle_total"',
'"2000_NL_Seattle_count"',
'"2000_US_Amsterdam_total"',
'"2000_US_Amsterdam_count"',
'"2000_US_Seattle_total"',
'"2000_US_Seattle_count"',
'"2010_NL_Amsterdam_total"',
'"2010_NL_Amsterdam_count"',
'"2010_NL_Seattle_total"',
'"2010_NL_Seattle_count"',
'"2010_US_Amsterdam_total"',
'"2010_US_Amsterdam_count"',
'"2010_US_Seattle_total"',
'"2010_US_Seattle_count"',
],
},
}
for query, dialect_columns in query_to_column_names.items():
for dialect, expected_columns in dialect_columns.items():
expr = parse_one(query, read=dialect)
columns = expr.args["from"].this.args["pivots"][0].args["columns"]
self.assertEqual(expected_columns, [col.sql(dialect=dialect) for col in columns])
with self.subTest(f"Testing query '{query}' for dialect {dialect}"):
expr = parse_one(query, read=dialect)
columns = expr.args["from"].this.args["pivots"][0].args["columns"]
self.assertEqual(
expected_columns, [col.sql(dialect=dialect) for col in columns]
)
def test_parse_nested(self):
def warn_over_threshold(query: str, max_threshold: float = 0.2):