summaryrefslogtreecommitdiffstats
path: root/tests
diff options
context:
space:
mode:
authorDaniel Baumann <daniel.baumann@progress-linux.org>2023-03-08 07:22:12 +0000
committerDaniel Baumann <daniel.baumann@progress-linux.org>2023-03-08 07:22:12 +0000
commit243d9a15edd910136313a7a9fa7d7f24b494e169 (patch)
tree22d768ca708dc5caa8230705a9155ba1a343502b /tests
parentAdding upstream version 11.3.0. (diff)
downloadsqlglot-243d9a15edd910136313a7a9fa7d7f24b494e169.tar.xz
sqlglot-243d9a15edd910136313a7a9fa7d7f24b494e169.zip
Adding upstream version 11.3.3.upstream/11.3.3
Signed-off-by: Daniel Baumann <daniel.baumann@progress-linux.org>
Diffstat (limited to '')
-rw-r--r--tests/dialects/test_bigquery.py17
-rw-r--r--tests/dialects/test_dialect.py1
-rw-r--r--tests/dialects/test_duckdb.py34
-rw-r--r--tests/dialects/test_hive.py26
-rw-r--r--tests/dialects/test_mysql.py1
-rw-r--r--tests/dialects/test_postgres.py8
-rw-r--r--tests/dialects/test_redshift.py6
-rw-r--r--tests/dialects/test_snowflake.py88
-rw-r--r--tests/dialects/test_sqlite.py7
-rw-r--r--tests/fixtures/identity.sql21
-rw-r--r--tests/test_build.py6
-rw-r--r--tests/test_optimizer.py5
-rw-r--r--tests/test_transpile.py8
13 files changed, 197 insertions, 31 deletions
diff --git a/tests/dialects/test_bigquery.py b/tests/dialects/test_bigquery.py
index 7b18a6a..22387da 100644
--- a/tests/dialects/test_bigquery.py
+++ b/tests/dialects/test_bigquery.py
@@ -8,6 +8,8 @@ class TestBigQuery(Validator):
def test_bigquery(self):
self.validate_identity("SELECT STRUCT<ARRAY<STRING>>(['2023-01-17'])")
self.validate_identity("SELECT * FROM q UNPIVOT(values FOR quarter IN (b, c))")
+
+ self.validate_all("LEAST(x, y)", read={"sqlite": "MIN(x, y)"})
self.validate_all(
"REGEXP_CONTAINS('foo', '.*')",
read={"bigquery": "REGEXP_CONTAINS('foo', '.*')"},
@@ -390,3 +392,18 @@ class TestBigQuery(Validator):
"bigquery": "INSERT INTO test (cola, colb) VALUES (CAST(7 AS STRING), CAST(14 AS STRING))",
},
)
+
+ def test_merge(self):
+ self.validate_all(
+ """
+ MERGE dataset.Inventory T
+ USING dataset.NewArrivals S ON FALSE
+ WHEN NOT MATCHED BY TARGET AND product LIKE '%a%'
+ THEN DELETE
+ WHEN NOT MATCHED BY SOURCE AND product LIKE '%b%'
+ THEN DELETE""",
+ write={
+ "bigquery": "MERGE INTO dataset.Inventory AS T USING dataset.NewArrivals AS S ON FALSE WHEN NOT MATCHED AND product LIKE '%a%' THEN DELETE WHEN NOT MATCHED BY SOURCE AND product LIKE '%b%' THEN DELETE",
+ "snowflake": "MERGE INTO dataset.Inventory AS T USING dataset.NewArrivals AS S ON FALSE WHEN NOT MATCHED AND product LIKE '%a%' THEN DELETE WHEN NOT MATCHED AND product LIKE '%b%' THEN DELETE",
+ },
+ )
diff --git a/tests/dialects/test_dialect.py b/tests/dialects/test_dialect.py
index 5054d94..69563cb 100644
--- a/tests/dialects/test_dialect.py
+++ b/tests/dialects/test_dialect.py
@@ -510,6 +510,7 @@ class TestDialect(Validator):
"DATE_ADD(x, 1, 'day')",
read={
"mysql": "DATE_ADD(x, INTERVAL 1 DAY)",
+ "snowflake": "DATEADD('day', 1, x)",
"starrocks": "DATE_ADD(x, INTERVAL 1 DAY)",
},
write={
diff --git a/tests/dialects/test_duckdb.py b/tests/dialects/test_duckdb.py
index 0efb7e7..a1a0090 100644
--- a/tests/dialects/test_duckdb.py
+++ b/tests/dialects/test_duckdb.py
@@ -75,6 +75,40 @@ class TestDuckDB(Validator):
},
)
+ def test_sample(self):
+ self.validate_all(
+ "SELECT * FROM tbl USING SAMPLE 5",
+ write={"duckdb": "SELECT * FROM tbl USING SAMPLE (5)"},
+ )
+ self.validate_all(
+ "SELECT * FROM tbl USING SAMPLE 10%",
+ write={"duckdb": "SELECT * FROM tbl USING SAMPLE (10 PERCENT)"},
+ )
+ self.validate_all(
+ "SELECT * FROM tbl USING SAMPLE 10 PERCENT (bernoulli)",
+ write={"duckdb": "SELECT * FROM tbl USING SAMPLE BERNOULLI (10 PERCENT)"},
+ )
+ self.validate_all(
+ "SELECT * FROM tbl USING SAMPLE reservoir(50 ROWS) REPEATABLE (100)",
+ write={"duckdb": "SELECT * FROM tbl USING SAMPLE RESERVOIR (50 ROWS) REPEATABLE (100)"},
+ )
+ self.validate_all(
+ "SELECT * FROM tbl USING SAMPLE 10% (system, 377)",
+ write={"duckdb": "SELECT * FROM tbl USING SAMPLE SYSTEM (10 PERCENT) REPEATABLE (377)"},
+ )
+ self.validate_all(
+ "SELECT * FROM tbl TABLESAMPLE RESERVOIR(20%), tbl2 WHERE tbl.i=tbl2.i",
+ write={
+ "duckdb": "SELECT * FROM tbl TABLESAMPLE RESERVOIR (20 PERCENT), tbl2 WHERE tbl.i = tbl2.i"
+ },
+ )
+ self.validate_all(
+ "SELECT * FROM tbl, tbl2 WHERE tbl.i=tbl2.i USING SAMPLE RESERVOIR(20%)",
+ write={
+ "duckdb": "SELECT * FROM tbl, tbl2 WHERE tbl.i = tbl2.i USING SAMPLE RESERVOIR (20 PERCENT)"
+ },
+ )
+
def test_duckdb(self):
self.validate_identity("SELECT {'a': 1} AS x")
self.validate_identity("SELECT {'a': {'b': {'c': 1}}, 'd': {'e': 2}} AS x")
diff --git a/tests/dialects/test_hive.py b/tests/dialects/test_hive.py
index a067764..8484805 100644
--- a/tests/dialects/test_hive.py
+++ b/tests/dialects/test_hive.py
@@ -482,9 +482,9 @@ class TestHive(Validator):
self.validate_all(
"SELECT * FROM x TABLESAMPLE(10) y",
write={
- "presto": "SELECT * FROM x AS y TABLESAMPLE(10)",
- "hive": "SELECT * FROM x TABLESAMPLE(10) AS y",
- "spark": "SELECT * FROM x TABLESAMPLE(10) AS y",
+ "presto": "SELECT * FROM x AS y TABLESAMPLE (10)",
+ "hive": "SELECT * FROM x TABLESAMPLE (10) AS y",
+ "spark": "SELECT * FROM x TABLESAMPLE (10) AS y",
},
)
self.validate_all(
@@ -626,25 +626,25 @@ class TestHive(Validator):
},
)
self.validate_all(
- "SELECT * FROM x TABLESAMPLE(1) AS foo",
+ "SELECT * FROM x TABLESAMPLE (1) AS foo",
read={
- "presto": "SELECT * FROM x AS foo TABLESAMPLE(1)",
+ "presto": "SELECT * FROM x AS foo TABLESAMPLE (1)",
},
write={
- "presto": "SELECT * FROM x AS foo TABLESAMPLE(1)",
- "hive": "SELECT * FROM x TABLESAMPLE(1) AS foo",
- "spark": "SELECT * FROM x TABLESAMPLE(1) AS foo",
+ "presto": "SELECT * FROM x AS foo TABLESAMPLE (1)",
+ "hive": "SELECT * FROM x TABLESAMPLE (1) AS foo",
+ "spark": "SELECT * FROM x TABLESAMPLE (1) AS foo",
},
)
self.validate_all(
- "SELECT * FROM x TABLESAMPLE(1) AS foo",
+ "SELECT * FROM x TABLESAMPLE (1) AS foo",
read={
- "presto": "SELECT * FROM x AS foo TABLESAMPLE(1)",
+ "presto": "SELECT * FROM x AS foo TABLESAMPLE (1)",
},
write={
- "presto": "SELECT * FROM x AS foo TABLESAMPLE(1)",
- "hive": "SELECT * FROM x TABLESAMPLE(1) AS foo",
- "spark": "SELECT * FROM x TABLESAMPLE(1) AS foo",
+ "presto": "SELECT * FROM x AS foo TABLESAMPLE (1)",
+ "hive": "SELECT * FROM x TABLESAMPLE (1) AS foo",
+ "spark": "SELECT * FROM x TABLESAMPLE (1) AS foo",
},
)
self.validate_all(
diff --git a/tests/dialects/test_mysql.py b/tests/dialects/test_mysql.py
index 192f9fc..5f8560a 100644
--- a/tests/dialects/test_mysql.py
+++ b/tests/dialects/test_mysql.py
@@ -64,6 +64,7 @@ class TestMySQL(Validator):
self.validate_identity("SET TRANSACTION READ ONLY")
self.validate_identity("SET GLOBAL TRANSACTION ISOLATION LEVEL REPEATABLE READ, READ WRITE")
self.validate_identity("SELECT SCHEMA()")
+ self.validate_identity("SELECT DATABASE()")
def test_types(self):
self.validate_all(
diff --git a/tests/dialects/test_postgres.py b/tests/dialects/test_postgres.py
index 5c4a23e..0881a89 100644
--- a/tests/dialects/test_postgres.py
+++ b/tests/dialects/test_postgres.py
@@ -11,6 +11,14 @@ class TestPostgres(Validator):
self.validate_identity("CREATE TABLE test (foo JSONB)")
self.validate_identity("CREATE TABLE test (foo VARCHAR(64)[])")
+ self.validate_identity("INSERT INTO x VALUES (1, 'a', 2.0) RETURNING a")
+ self.validate_identity("INSERT INTO x VALUES (1, 'a', 2.0) RETURNING a, b")
+ self.validate_identity("INSERT INTO x VALUES (1, 'a', 2.0) RETURNING *")
+ self.validate_identity(
+ "DELETE FROM event USING sales AS s WHERE event.eventid = s.eventid RETURNING a"
+ )
+ self.validate_identity("UPDATE tbl_name SET foo = 123 RETURNING a")
+
self.validate_all(
"CREATE OR REPLACE FUNCTION function_name (input_a character varying DEFAULT NULL::character varying)",
write={
diff --git a/tests/dialects/test_redshift.py b/tests/dialects/test_redshift.py
index 640706a..ad83b99 100644
--- a/tests/dialects/test_redshift.py
+++ b/tests/dialects/test_redshift.py
@@ -85,6 +85,12 @@ class TestRedshift(Validator):
"presto": "DATE_DIFF(d, a, b)",
},
)
+ self.validate_all(
+ "SELECT TOP 1 x FROM y",
+ write={
+ "redshift": "SELECT x FROM y LIMIT 1",
+ },
+ )
def test_identity(self):
self.validate_identity(
diff --git a/tests/dialects/test_snowflake.py b/tests/dialects/test_snowflake.py
index 3358227..c28c58d 100644
--- a/tests/dialects/test_snowflake.py
+++ b/tests/dialects/test_snowflake.py
@@ -63,6 +63,28 @@ class TestSnowflake(Validator):
},
)
self.validate_all(
+ "ZEROIFNULL(foo)",
+ write={
+ "snowflake": "IFF(foo IS NULL, 0, foo)",
+ "sqlite": "CASE WHEN foo IS NULL THEN 0 ELSE foo END",
+ "presto": "IF(foo IS NULL, 0, foo)",
+ "spark": "IF(foo IS NULL, 0, foo)",
+ "hive": "IF(foo IS NULL, 0, foo)",
+ "duckdb": "CASE WHEN foo IS NULL THEN 0 ELSE foo END",
+ },
+ )
+ self.validate_all(
+ "NULLIFZERO(foo)",
+ write={
+ "snowflake": "IFF(foo = 0, NULL, foo)",
+ "sqlite": "CASE WHEN foo = 0 THEN NULL ELSE foo END",
+ "presto": "IF(foo = 0, NULL, foo)",
+ "spark": "IF(foo = 0, NULL, foo)",
+ "hive": "IF(foo = 0, NULL, foo)",
+ "duckdb": "CASE WHEN foo = 0 THEN NULL ELSE foo END",
+ },
+ )
+ self.validate_all(
"CREATE OR REPLACE TEMPORARY TABLE x (y NUMBER IDENTITY(0, 1))",
write={
"snowflake": "CREATE OR REPLACE TEMPORARY TABLE x (y DECIMAL AUTOINCREMENT START 0 INCREMENT 1)",
@@ -280,12 +302,6 @@ class TestSnowflake(Validator):
},
)
self.validate_all(
- "SELECT a FROM test SAMPLE BLOCK (0.5) SEED (42)",
- write={
- "snowflake": "SELECT a FROM test TABLESAMPLE BLOCK (0.5) SEED (42)",
- },
- )
- self.validate_all(
"SELECT a FROM test pivot",
write={
"snowflake": "SELECT a FROM test AS pivot",
@@ -356,6 +372,51 @@ class TestSnowflake(Validator):
},
)
+ def test_sample(self):
+ self.validate_identity("SELECT * FROM testtable TABLESAMPLE BERNOULLI (20.3)")
+ self.validate_identity("SELECT * FROM testtable TABLESAMPLE (100)")
+ self.validate_identity(
+ "SELECT i, j FROM table1 AS t1 INNER JOIN table2 AS t2 TABLESAMPLE (50) WHERE t2.j = t1.i"
+ )
+ self.validate_identity(
+ "SELECT * FROM (SELECT * FROM t1 JOIN t2 ON t1.a = t2.c) TABLESAMPLE (1)"
+ )
+ self.validate_identity("SELECT * FROM testtable TABLESAMPLE SYSTEM (3) SEED (82)")
+ self.validate_identity("SELECT * FROM testtable TABLESAMPLE (10 ROWS)")
+
+ self.validate_all(
+ "SELECT * FROM testtable SAMPLE (10)",
+ write={"snowflake": "SELECT * FROM testtable TABLESAMPLE (10)"},
+ )
+ self.validate_all(
+ "SELECT * FROM testtable SAMPLE ROW (0)",
+ write={"snowflake": "SELECT * FROM testtable TABLESAMPLE ROW (0)"},
+ )
+ self.validate_all(
+ "SELECT a FROM test SAMPLE BLOCK (0.5) SEED (42)",
+ write={
+ "snowflake": "SELECT a FROM test TABLESAMPLE BLOCK (0.5) SEED (42)",
+ },
+ )
+ self.validate_all(
+ """
+ SELECT i, j
+ FROM
+ table1 AS t1 SAMPLE (25) -- 25% of rows in table1
+ INNER JOIN
+ table2 AS t2 SAMPLE (50) -- 50% of rows in table2
+ WHERE t2.j = t1.i""",
+ write={
+ "snowflake": "SELECT i, j FROM table1 AS t1 TABLESAMPLE (25) /* 25% of rows in table1 */ INNER JOIN table2 AS t2 TABLESAMPLE (50) /* 50% of rows in table2 */ WHERE t2.j = t1.i",
+ },
+ )
+ self.validate_all(
+ "SELECT * FROM testtable SAMPLE BLOCK (0.012) REPEATABLE (99992)",
+ write={
+ "snowflake": "SELECT * FROM testtable TABLESAMPLE BLOCK (0.012) SEED (99992)",
+ },
+ )
+
def test_timestamps(self):
self.validate_identity("SELECT EXTRACT(month FROM a)")
@@ -415,6 +476,13 @@ class TestSnowflake(Validator):
"presto": "SELECT TO_UNIXTIME(CAST(foo AS TIMESTAMP)) * 1000 AS ddate FROM table_name",
},
)
+ self.validate_all(
+ "DATEADD(DAY, 5, CAST('2008-12-25' AS DATE))",
+ write={
+ "bigquery": "DATE_ADD(CAST('2008-12-25' AS DATE), INTERVAL 5 DAY)",
+ "snowflake": "DATEADD(DAY, 5, CAST('2008-12-25' AS DATE))",
+ },
+ )
def test_semi_structured_types(self):
self.validate_identity("SELECT CAST(a AS VARIANT)")
@@ -655,6 +723,14 @@ FROM persons AS p, LATERAL FLATTEN(input => p.c, path => 'contact') AS f, LATERA
},
)
+ self.validate_all(
+ """SELECT $1 AS "_1" FROM VALUES ('a'), ('b')""",
+ write={
+ "snowflake": """SELECT $1 AS "_1" FROM (VALUES ('a'), ('b'))""",
+ "spark": """SELECT @1 AS `_1` FROM VALUES ('a'), ('b')""",
+ },
+ )
+
def test_describe_table(self):
self.validate_all(
"DESCRIBE TABLE db.table",
diff --git a/tests/dialects/test_sqlite.py b/tests/dialects/test_sqlite.py
index 19a88f3..a3e4cc9 100644
--- a/tests/dialects/test_sqlite.py
+++ b/tests/dialects/test_sqlite.py
@@ -81,6 +81,13 @@ class TestSQLite(Validator):
"sqlite": "SELECT fname, lname, age FROM person ORDER BY age DESC NULLS FIRST, fname NULLS LAST, lname",
},
)
+ self.validate_all("x", read={"snowflake": "LEAST(x)"})
+ self.validate_all("MIN(x)", read={"snowflake": "MIN(x)"}, write={"snowflake": "MIN(x)"})
+ self.validate_all(
+ "MIN(x, y, z)",
+ read={"snowflake": "LEAST(x, y, z)"},
+ write={"snowflake": "LEAST(x, y, z)"},
+ )
def test_hexadecimal_literal(self):
self.validate_all(
diff --git a/tests/fixtures/identity.sql b/tests/fixtures/identity.sql
index 0677a05..085880c 100644
--- a/tests/fixtures/identity.sql
+++ b/tests/fixtures/identity.sql
@@ -153,6 +153,7 @@ SUM(ROW() OVER (PARTITION BY x AND y))
CASE WHEN (x > 1) THEN 1 ELSE 0 END
CASE (1) WHEN 1 THEN 1 ELSE 0 END
CASE 1 WHEN 1 THEN 1 ELSE 0 END
+CASE 1 WHEN 1 THEN timestamp ELSE date END
x AT TIME ZONE 'UTC'
CAST('2025-11-20 00:00:00+00' AS TIMESTAMP) AT TIME ZONE 'Africa/Cairo'
SET x = 1
@@ -191,6 +192,7 @@ SELECT DISTINCT TIMESTAMP_TRUNC(time_field, MONTH) AS time_value FROM "table"
SELECT DISTINCT ON (x) x, y FROM z
SELECT DISTINCT ON (x, y + 1) * FROM z
SELECT DISTINCT ON (x.y) * FROM z
+SELECT DISTINCT FROM_SOMETHING
SELECT top.x
SELECT TIMESTAMP(DATE_TRUNC(DATE(time_field), MONTH)) AS time_value FROM "table"
SELECT GREATEST((3 + 1), LEAST(3, 4))
@@ -295,13 +297,13 @@ SELECT CASE CASE x > 1 WHEN TRUE THEN 1 END WHEN 1 THEN 1 ELSE 2 END
SELECT a FROM (SELECT a FROM test) AS x
SELECT a FROM (SELECT a FROM (SELECT a FROM test) AS y) AS x
SELECT a FROM test WHERE a IN (1, 2, 3) OR b BETWEEN 1 AND 4
-SELECT a FROM test AS x TABLESAMPLE(BUCKET 1 OUT OF 5)
-SELECT a FROM test TABLESAMPLE(BUCKET 1 OUT OF 5)
-SELECT a FROM test TABLESAMPLE(BUCKET 1 OUT OF 5 ON x)
-SELECT a FROM test TABLESAMPLE(BUCKET 1 OUT OF 5 ON RAND())
-SELECT a FROM test TABLESAMPLE(0.1 PERCENT)
-SELECT a FROM test TABLESAMPLE(100)
-SELECT a FROM test TABLESAMPLE(100 ROWS)
+SELECT a FROM test AS x TABLESAMPLE (BUCKET 1 OUT OF 5)
+SELECT a FROM test TABLESAMPLE (BUCKET 1 OUT OF 5)
+SELECT a FROM test TABLESAMPLE (BUCKET 1 OUT OF 5 ON x)
+SELECT a FROM test TABLESAMPLE (BUCKET 1 OUT OF 5 ON RAND())
+SELECT a FROM test TABLESAMPLE (0.1 PERCENT)
+SELECT a FROM test TABLESAMPLE (100)
+SELECT a FROM test TABLESAMPLE (100 ROWS)
SELECT a FROM test TABLESAMPLE BERNOULLI (50)
SELECT a FROM test TABLESAMPLE SYSTEM (75)
SELECT a FROM test PIVOT(SUM(x) FOR y IN ('z', 'q'))
@@ -310,7 +312,7 @@ SELECT a FROM test PIVOT(SOMEAGG(x, y, z) FOR q IN (1))
SELECT a FROM test PIVOT(SUM(x) FOR y IN ('z', 'q')) PIVOT(MAX(b) FOR c IN ('d'))
SELECT a FROM (SELECT a, b FROM test) PIVOT(SUM(x) FOR y IN ('z', 'q'))
SELECT a FROM test UNPIVOT(x FOR y IN (z, q)) AS x
-SELECT a FROM test PIVOT(SUM(x) FOR y IN ('z', 'q')) AS x TABLESAMPLE(0.1)
+SELECT a FROM test PIVOT(SUM(x) FOR y IN ('z', 'q')) AS x TABLESAMPLE (0.1)
SELECT a FROM test PIVOT(SUM(x) FOR y IN ('z', 'q')) UNPIVOT(x FOR y IN (z, q)) AS x
SELECT ABS(a) FROM test
SELECT AVG(a) FROM test
@@ -466,6 +468,7 @@ SELECT SUM(x) OVER (PARTITION BY a RANGE BETWEEN 1 AND 3)
SELECT SUM(x) OVER (PARTITION BY a RANGE BETWEEN 1 FOLLOWING AND 3)
SELECT SUM(x) OVER (PARTITION BY a RANGE BETWEEN 1 FOLLOWING AND UNBOUNDED FOLLOWING)
SELECT AVG(x) OVER (ROWS BETWEEN UNBOUNDED PRECEDING AND UNBOUNDED FOLLOWING) FROM t
+SELECT SUM(x) OVER (PARTITION BY a ORDER BY date ROWS BETWEEN 1 PRECEDING AND 1 FOLLOWING)
SELECT LISTAGG(x) WITHIN GROUP (ORDER BY x) AS y
SELECT LISTAGG(x) WITHIN GROUP (ORDER BY x DESC)
SELECT PERCENTILE_CONT(0.5) WITHIN GROUP (ORDER BY x)
@@ -708,6 +711,7 @@ SELECT 1 /* c1 */ + 2 /* c2 */, 3 /* c3 */
SELECT x FROM a.b.c /* x */, e.f.g /* x */
SELECT FOO(x /* c */) /* FOO */, b /* b */
SELECT FOO(x /* c1 */ + y /* c2 */ + BLA(5 /* c3 */)) FROM (VALUES (1 /* c4 */, "test" /* c5 */)) /* c6 */
+INSERT INTO foo SELECT * FROM bar /* comment */
SELECT a FROM x WHERE a COLLATE 'utf8_general_ci' = 'b'
SELECT x AS INTO FROM bla
SELECT * INTO newevent FROM event
@@ -731,6 +735,7 @@ ALTER TABLE orders DROP PARTITION(dt = '2014-05-14', country = 'IN')
ALTER TABLE orders DROP IF EXISTS PARTITION(dt = '2014-05-14', country = 'IN')
ALTER TABLE orders DROP PARTITION(dt = '2014-05-14', country = 'IN'), PARTITION(dt = '2014-05-15', country = 'IN')
ALTER TABLE mydataset.mytable DELETE WHERE x = 1
+ALTER TABLE table1 MODIFY COLUMN name1 SET TAG foo='bar'
SELECT div.a FROM test_table AS div
WITH view AS (SELECT 1 AS x) SELECT * FROM view
ARRAY<STRUCT<INT, DOUBLE, ARRAY<INT>>>
diff --git a/tests/test_build.py b/tests/test_build.py
index fbfbb62..718e471 100644
--- a/tests/test_build.py
+++ b/tests/test_build.py
@@ -504,6 +504,12 @@ class TestBuild(unittest.TestCase):
.window("d AS (PARTITION BY g ORDER BY h)"),
"SELECT AVG(a) OVER b, MIN(c) OVER d FROM table WINDOW b AS (PARTITION BY e ORDER BY f), d AS (PARTITION BY g ORDER BY h)",
),
+ (
+ lambda: select("*")
+ .from_("table")
+ .qualify("row_number() OVER (PARTITION BY a ORDER BY b) = 1"),
+ "SELECT * FROM table QUALIFY ROW_NUMBER() OVER (PARTITION BY a ORDER BY b) = 1",
+ ),
]:
with self.subTest(sql):
self.assertEqual(expression().sql(dialect[0] if dialect else None), sql)
diff --git a/tests/test_optimizer.py b/tests/test_optimizer.py
index 8ddd95f..e10d05e 100644
--- a/tests/test_optimizer.py
+++ b/tests/test_optimizer.py
@@ -364,6 +364,11 @@ FROM READ_CSV('tests/fixtures/optimizer/tpc-h/nation.csv.gz', 'delimiter', '|')
expression = annotate_types(parse_one("ARRAY(1)::ARRAY<INT>"))
self.assertEqual(expression.type, parse_one("ARRAY<INT>", into=exp.DataType))
+ expression = annotate_types(parse_one("CAST(x AS INTERVAL)"))
+ self.assertEqual(expression.type.this, exp.DataType.Type.INTERVAL)
+ self.assertEqual(expression.this.type.this, exp.DataType.Type.UNKNOWN)
+ self.assertEqual(expression.args["to"].type.this, exp.DataType.Type.INTERVAL)
+
def test_cache_annotation(self):
expression = annotate_types(
parse_one("CACHE LAZY TABLE x OPTIONS('storageLevel' = 'value') AS SELECT 1")
diff --git a/tests/test_transpile.py b/tests/test_transpile.py
index 0463aed..6355400 100644
--- a/tests/test_transpile.py
+++ b/tests/test_transpile.py
@@ -202,9 +202,9 @@ FROM bar /* comment 5 */, tbl /* comment 6 */""",
)
def test_types(self):
- self.validate("INT x", "CAST(x AS INT)")
- self.validate("VARCHAR x y", "CAST(x AS VARCHAR) AS y")
- self.validate("STRING x y", "CAST(x AS TEXT) AS y")
+ self.validate("INT 1", "CAST(1 AS INT)")
+ self.validate("VARCHAR 'x' y", "CAST('x' AS VARCHAR) AS y")
+ self.validate("STRING 'x' y", "CAST('x' AS TEXT) AS y")
self.validate("x::INT", "CAST(x AS INT)")
self.validate("x::INTEGER", "CAST(x AS INT)")
self.validate("x::INT y", "CAST(x AS INT) AS y")
@@ -221,7 +221,7 @@ FROM bar /* comment 5 */, tbl /* comment 6 */""",
self.validate("a NOT BETWEEN b AND c", "NOT a BETWEEN b AND c")
self.validate("a NOT IN (1, 2)", "NOT a IN (1, 2)")
self.validate("a IS NOT NULL", "NOT a IS NULL")
- self.validate("a LIKE TEXT y", "a LIKE CAST(y AS TEXT)")
+ self.validate("a LIKE TEXT 'y'", "a LIKE CAST('y' AS TEXT)")
def test_extract(self):
self.validate(