From bd2d949d1f2fb728cf4c429dd3ae9a1510e10182 Mon Sep 17 00:00:00 2001 From: Daniel Baumann Date: Mon, 25 Sep 2023 10:20:06 +0200 Subject: Adding upstream version 18.7.0. Signed-off-by: Daniel Baumann --- tests/dialects/test_duckdb.py | 97 ++++++++++++++++++++++++++++++++++++++++++- 1 file changed, 95 insertions(+), 2 deletions(-) (limited to 'tests/dialects/test_duckdb.py') diff --git a/tests/dialects/test_duckdb.py b/tests/dialects/test_duckdb.py index 36fca7c..dbf0a87 100644 --- a/tests/dialects/test_duckdb.py +++ b/tests/dialects/test_duckdb.py @@ -6,6 +6,92 @@ class TestDuckDB(Validator): dialect = "duckdb" def test_duckdb(self): + self.assertEqual( + parse_one("select * from t limit (select 5)").sql(dialect="duckdb"), + exp.select("*").from_("t").limit(exp.select("5").subquery()).sql(dialect="duckdb"), + ) + + for struct_value in ("{'a': 1}", "struct_pack(a := 1)"): + self.validate_all(struct_value, write={"presto": UnsupportedError}) + + for join_type in ("SEMI", "ANTI"): + exists = "EXISTS" if join_type == "SEMI" else "NOT EXISTS" + + self.validate_all( + f"SELECT * FROM t1 {join_type} JOIN t2 ON t1.x = t2.x", + write={ + "bigquery": f"SELECT * FROM t1 WHERE {exists}(SELECT 1 FROM t2 WHERE t1.x = t2.x)", + "clickhouse": f"SELECT * FROM t1 {join_type} JOIN t2 ON t1.x = t2.x", + "databricks": f"SELECT * FROM t1 {join_type} JOIN t2 ON t1.x = t2.x", + "doris": f"SELECT * FROM t1 WHERE {exists}(SELECT 1 FROM t2 WHERE t1.x = t2.x)", + "drill": f"SELECT * FROM t1 WHERE {exists}(SELECT 1 FROM t2 WHERE t1.x = t2.x)", + "duckdb": f"SELECT * FROM t1 {join_type} JOIN t2 ON t1.x = t2.x", + "hive": f"SELECT * FROM t1 {join_type} JOIN t2 ON t1.x = t2.x", + "mysql": f"SELECT * FROM t1 WHERE {exists}(SELECT 1 FROM t2 WHERE t1.x = t2.x)", + "oracle": f"SELECT * FROM t1 {join_type} JOIN t2 ON t1.x = t2.x", + "postgres": f"SELECT * FROM t1 WHERE {exists}(SELECT 1 FROM t2 WHERE t1.x = t2.x)", + "presto": f"SELECT * FROM t1 WHERE {exists}(SELECT 1 FROM t2 WHERE t1.x = t2.x)", + "redshift": f"SELECT * FROM t1 WHERE {exists}(SELECT 1 FROM t2 WHERE t1.x = t2.x)", + "snowflake": f"SELECT * FROM t1 WHERE {exists}(SELECT 1 FROM t2 WHERE t1.x = t2.x)", + "spark": f"SELECT * FROM t1 {join_type} JOIN t2 ON t1.x = t2.x", + "sqlite": f"SELECT * FROM t1 WHERE {exists}(SELECT 1 FROM t2 WHERE t1.x = t2.x)", + "starrocks": f"SELECT * FROM t1 WHERE {exists}(SELECT 1 FROM t2 WHERE t1.x = t2.x)", + "teradata": f"SELECT * FROM t1 WHERE {exists}(SELECT 1 FROM t2 WHERE t1.x = t2.x)", + "trino": f"SELECT * FROM t1 WHERE {exists}(SELECT 1 FROM t2 WHERE t1.x = t2.x)", + "tsql": f"SELECT * FROM t1 WHERE {exists}(SELECT 1 FROM t2 WHERE t1.x = t2.x)", + }, + ) + self.validate_all( + f"SELECT * FROM t1 {join_type} JOIN t2 ON t1.x = t2.x", + read={ + "duckdb": f"SELECT * FROM t1 {join_type} JOIN t2 ON t1.x = t2.x", + "spark": f"SELECT * FROM t1 LEFT {join_type} JOIN t2 ON t1.x = t2.x", + }, + ) + + self.validate_all( + "WITH cte(x) AS (SELECT 1 UNION ALL SELECT 2 UNION ALL SELECT 3) SELECT AVG(x) FILTER (WHERE x > 1) FROM cte", + write={ + "duckdb": "WITH cte(x) AS (SELECT 1 UNION ALL SELECT 2 UNION ALL SELECT 3) SELECT AVG(x) FILTER(WHERE x > 1) FROM cte", + "snowflake": "WITH cte(x) AS (SELECT 1 UNION ALL SELECT 2 UNION ALL SELECT 3) SELECT AVG(IFF(x > 1, x, NULL)) FROM cte", + }, + ) + self.validate_all( + "SELECT AVG(x) FILTER (WHERE TRUE) FROM t", + write={ + "duckdb": "SELECT AVG(x) FILTER(WHERE TRUE) FROM t", + "snowflake": "SELECT AVG(IFF(TRUE, x, NULL)) FROM t", + }, + ) + self.validate_all( + "SELECT UNNEST(ARRAY[1, 2, 3]), UNNEST(ARRAY[4, 5]), UNNEST(ARRAY[6])", + write={ + "bigquery": "SELECT IF(pos = pos_2, col, NULL) AS col, IF(pos = pos_3, col_2, NULL) AS col_2, IF(pos = pos_4, col_3, NULL) AS col_3 FROM UNNEST(GENERATE_ARRAY(0, GREATEST(ARRAY_LENGTH([1, 2, 3]), ARRAY_LENGTH([4, 5]), ARRAY_LENGTH([6])) - 1)) AS pos CROSS JOIN UNNEST([1, 2, 3]) AS col WITH OFFSET AS pos_2 CROSS JOIN UNNEST([4, 5]) AS col_2 WITH OFFSET AS pos_3 CROSS JOIN UNNEST([6]) AS col_3 WITH OFFSET AS pos_4 WHERE ((pos = pos_2 OR (pos > (ARRAY_LENGTH([1, 2, 3]) - 1) AND pos_2 = (ARRAY_LENGTH([1, 2, 3]) - 1))) AND (pos = pos_3 OR (pos > (ARRAY_LENGTH([4, 5]) - 1) AND pos_3 = (ARRAY_LENGTH([4, 5]) - 1)))) AND (pos = pos_4 OR (pos > (ARRAY_LENGTH([6]) - 1) AND pos_4 = (ARRAY_LENGTH([6]) - 1)))", + "presto": "SELECT IF(pos = pos_2, col) AS col, IF(pos = pos_3, col_2) AS col_2, IF(pos = pos_4, col_3) AS col_3 FROM UNNEST(SEQUENCE(1, GREATEST(CARDINALITY(ARRAY[1, 2, 3]), CARDINALITY(ARRAY[4, 5]), CARDINALITY(ARRAY[6])))) AS _u(pos) CROSS JOIN UNNEST(ARRAY[1, 2, 3]) WITH ORDINALITY AS _u_2(col, pos_2) CROSS JOIN UNNEST(ARRAY[4, 5]) WITH ORDINALITY AS _u_3(col_2, pos_3) CROSS JOIN UNNEST(ARRAY[6]) WITH ORDINALITY AS _u_4(col_3, pos_4) WHERE ((pos = pos_2 OR (pos > CARDINALITY(ARRAY[1, 2, 3]) AND pos_2 = CARDINALITY(ARRAY[1, 2, 3]))) AND (pos = pos_3 OR (pos > CARDINALITY(ARRAY[4, 5]) AND pos_3 = CARDINALITY(ARRAY[4, 5])))) AND (pos = pos_4 OR (pos > CARDINALITY(ARRAY[6]) AND pos_4 = CARDINALITY(ARRAY[6])))", + }, + ) + + self.validate_all( + "SELECT UNNEST(ARRAY[1, 2, 3]), UNNEST(ARRAY[4, 5]), UNNEST(ARRAY[6]) FROM x", + write={ + "bigquery": "SELECT IF(pos = pos_2, col, NULL) AS col, IF(pos = pos_3, col_2, NULL) AS col_2, IF(pos = pos_4, col_3, NULL) AS col_3 FROM x, UNNEST(GENERATE_ARRAY(0, GREATEST(ARRAY_LENGTH([1, 2, 3]), ARRAY_LENGTH([4, 5]), ARRAY_LENGTH([6])) - 1)) AS pos CROSS JOIN UNNEST([1, 2, 3]) AS col WITH OFFSET AS pos_2 CROSS JOIN UNNEST([4, 5]) AS col_2 WITH OFFSET AS pos_3 CROSS JOIN UNNEST([6]) AS col_3 WITH OFFSET AS pos_4 WHERE ((pos = pos_2 OR (pos > (ARRAY_LENGTH([1, 2, 3]) - 1) AND pos_2 = (ARRAY_LENGTH([1, 2, 3]) - 1))) AND (pos = pos_3 OR (pos > (ARRAY_LENGTH([4, 5]) - 1) AND pos_3 = (ARRAY_LENGTH([4, 5]) - 1)))) AND (pos = pos_4 OR (pos > (ARRAY_LENGTH([6]) - 1) AND pos_4 = (ARRAY_LENGTH([6]) - 1)))", + "presto": "SELECT IF(pos = pos_2, col) AS col, IF(pos = pos_3, col_2) AS col_2, IF(pos = pos_4, col_3) AS col_3 FROM x, UNNEST(SEQUENCE(1, GREATEST(CARDINALITY(ARRAY[1, 2, 3]), CARDINALITY(ARRAY[4, 5]), CARDINALITY(ARRAY[6])))) AS _u(pos) CROSS JOIN UNNEST(ARRAY[1, 2, 3]) WITH ORDINALITY AS _u_2(col, pos_2) CROSS JOIN UNNEST(ARRAY[4, 5]) WITH ORDINALITY AS _u_3(col_2, pos_3) CROSS JOIN UNNEST(ARRAY[6]) WITH ORDINALITY AS _u_4(col_3, pos_4) WHERE ((pos = pos_2 OR (pos > CARDINALITY(ARRAY[1, 2, 3]) AND pos_2 = CARDINALITY(ARRAY[1, 2, 3]))) AND (pos = pos_3 OR (pos > CARDINALITY(ARRAY[4, 5]) AND pos_3 = CARDINALITY(ARRAY[4, 5])))) AND (pos = pos_4 OR (pos > CARDINALITY(ARRAY[6]) AND pos_4 = CARDINALITY(ARRAY[6])))", + }, + ) + self.validate_all( + "SELECT UNNEST(x) + 1", + write={ + "bigquery": "SELECT IF(pos = pos_2, col, NULL) + 1 AS col FROM UNNEST(GENERATE_ARRAY(0, GREATEST(ARRAY_LENGTH(x)) - 1)) AS pos CROSS JOIN UNNEST(x) AS col WITH OFFSET AS pos_2 WHERE pos = pos_2 OR (pos > (ARRAY_LENGTH(x) - 1) AND pos_2 = (ARRAY_LENGTH(x) - 1))", + }, + ) + self.validate_all( + "SELECT UNNEST(x) + 1 AS y", + write={ + "bigquery": "SELECT IF(pos = pos_2, y, NULL) + 1 AS y FROM UNNEST(GENERATE_ARRAY(0, GREATEST(ARRAY_LENGTH(x)) - 1)) AS pos CROSS JOIN UNNEST(x) AS y WITH OFFSET AS pos_2 WHERE pos = pos_2 OR (pos > (ARRAY_LENGTH(x) - 1) AND pos_2 = (ARRAY_LENGTH(x) - 1))", + }, + ) + + self.validate_identity("SELECT i FROM RANGE(5) AS _(i) ORDER BY i ASC") self.validate_identity("[x.STRING_SPLIT(' ')[1] FOR x IN ['1', '2', 3] IF x.CONTAINS('1')]") self.validate_identity("INSERT INTO x BY NAME SELECT 1 AS y") self.validate_identity("SELECT 1 AS x UNION ALL BY NAME SELECT 2 AS x") @@ -61,6 +147,13 @@ class TestDuckDB(Validator): self.validate_all("0x1010", write={"": "0 AS x1010"}) self.validate_all("x ~ y", write={"duckdb": "REGEXP_MATCHES(x, y)"}) self.validate_all("SELECT * FROM 'x.y'", write={"duckdb": 'SELECT * FROM "x.y"'}) + self.validate_all( + "SELECT UNNEST([1, 2, 3])", + write={ + "duckdb": "SELECT UNNEST([1, 2, 3])", + "snowflake": "SELECT IFF(pos = pos_2, col, NULL) AS col FROM (SELECT value FROM TABLE(FLATTEN(INPUT => ARRAY_GENERATE_RANGE(0, (GREATEST(ARRAY_SIZE([1, 2, 3])) - 1) + 1)))) AS _u(pos) CROSS JOIN (SELECT value, index FROM TABLE(FLATTEN(INPUT => [1, 2, 3]))) AS _u_2(col, pos_2) WHERE pos = pos_2 OR (pos > (ARRAY_SIZE([1, 2, 3]) - 1) AND pos_2 = (ARRAY_SIZE([1, 2, 3]) - 1))", + }, + ) self.validate_all( "VAR_POP(x)", read={ @@ -304,8 +397,8 @@ class TestDuckDB(Validator): self.validate_all( "SELECT fname, lname, age FROM person ORDER BY age DESC NULLS FIRST, fname ASC NULLS LAST, lname", write={ - "": "SELECT fname, lname, age FROM person ORDER BY age DESC NULLS FIRST, fname NULLS LAST, lname NULLS LAST", - "duckdb": "SELECT fname, lname, age FROM person ORDER BY age DESC NULLS FIRST, fname, lname", + "": "SELECT fname, lname, age FROM person ORDER BY age DESC NULLS FIRST, fname ASC NULLS LAST, lname NULLS LAST", + "duckdb": "SELECT fname, lname, age FROM person ORDER BY age DESC NULLS FIRST, fname ASC, lname", }, ) self.validate_all( -- cgit v1.2.3