Adding upstream version 20.9.0.upstream/20.9.0

Signed-off-by: Daniel Baumann <daniel.baumann@progress-linux.org>

1 files changed, 113 insertions, 27 deletions

diff --git a/tests/dialects/test_duckdb.py b/tests/dialects/test_duckdb.py
index f915168..e5f7e0c 100644
--- a/tests/dialects/test_duckdb.py
+++ b/tests/dialects/test_duckdb.py
@@ -7,6 +7,41 @@ class TestDuckDB(Validator):
     dialect = "duckdb"
 
     def test_duckdb(self):
+        struct_pack = parse_one('STRUCT_PACK("a b" := 1)', read="duckdb")
+        self.assertIsInstance(struct_pack.expressions[0].this, exp.Identifier)
+        self.assertEqual(struct_pack.sql(dialect="duckdb"), "{'a b': 1}")
+
+        self.validate_all(
+            "SELECT SUM(X) OVER (ORDER BY x)",
+            write={
+                "bigquery": "SELECT SUM(X) OVER (ORDER BY x NULLS LAST)",
+                "duckdb": "SELECT SUM(X) OVER (ORDER BY x)",
+                "mysql": "SELECT SUM(X) OVER (ORDER BY CASE WHEN x IS NULL THEN 1 ELSE 0 END, x)",
+            },
+        )
+        self.validate_all(
+            "SELECT SUM(X) OVER (ORDER BY x RANGE BETWEEN 1 PRECEDING AND CURRENT ROW)",
+            write={
+                "bigquery": "SELECT SUM(X) OVER (ORDER BY x RANGE BETWEEN 1 PRECEDING AND CURRENT ROW)",
+                "duckdb": "SELECT SUM(X) OVER (ORDER BY x RANGE BETWEEN 1 PRECEDING AND CURRENT ROW)",
+                "mysql": "SELECT SUM(X) OVER (ORDER BY x RANGE BETWEEN 1 PRECEDING AND CURRENT ROW)",
+            },
+        )
+        self.validate_all(
+            "SELECT * FROM x ORDER BY 1 NULLS LAST",
+            write={
+                "duckdb": "SELECT * FROM x ORDER BY 1",
+                "mysql": "SELECT * FROM x ORDER BY 1",
+            },
+        )
+
+        self.validate_all(
+            "CREATE TEMPORARY FUNCTION f1(a, b) AS (a + b)",
+            read={"bigquery": "CREATE TEMP FUNCTION f1(a INT64, b INT64) AS (a + b)"},
+        )
+        self.validate_identity("SELECT 1 WHERE x > $1")
+        self.validate_identity("SELECT 1 WHERE x > $name")
+
         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"),
@@ -55,11 +90,35 @@ class TestDuckDB(Validator):
             )
 
         self.validate_all(
+            """SELECT JSON('{"fruit":"banana"}') -> 'fruit'""",
+            write={
+                "duckdb": """SELECT JSON('{"fruit":"banana"}') -> 'fruit'""",
+                "snowflake": """SELECT PARSE_JSON('{"fruit":"banana"}')['fruit']""",
+            },
+        )
+        self.validate_all(
+            """SELECT JSON('{"fruit": {"foo": "banana"}}') -> 'fruit' -> 'foo'""",
+            write={
+                "duckdb": """SELECT JSON('{"fruit": {"foo": "banana"}}') -> 'fruit' -> 'foo'""",
+                "snowflake": """SELECT PARSE_JSON('{"fruit": {"foo": "banana"}}')['fruit']['foo']""",
+            },
+        )
+        self.validate_all(
+            "WITH _data AS (SELECT [{'a': 1, 'b': 2}, {'a': 2, 'b': 3}] AS col) SELECT (SELECT col['b'] FROM UNNEST(col) AS t(col) WHERE col['a'] = 1) FROM _data",
+            write={
+                "bigquery": "WITH _data AS (SELECT [STRUCT(1 AS a, 2 AS b), STRUCT(2 AS a, 3 AS b)] AS col) SELECT (SELECT col.b FROM UNNEST(col) AS col WHERE col.a = 1) FROM _data",
+                "duckdb": "WITH _data AS (SELECT [{'a': 1, 'b': 2}, {'a': 2, 'b': 3}] AS col) SELECT (SELECT col['b'] FROM UNNEST(col) AS t(col) WHERE col['a'] = 1) FROM _data",
+            },
+        )
+        self.validate_all(
             "SELECT {'bla': column1, 'foo': column2, 'bar': column3} AS data FROM source_table",
             read={
                 "bigquery": "SELECT STRUCT(column1 AS bla, column2 AS foo, column3 AS bar) AS data FROM source_table",
                 "duckdb": "SELECT {'bla': column1, 'foo': column2, 'bar': column3} AS data FROM source_table",
             },
+            write={
+                "bigquery": "SELECT STRUCT(column1 AS bla, column2 AS foo, column3 AS bar) AS data FROM source_table",
+            },
         )
         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",
@@ -86,8 +145,8 @@ class TestDuckDB(Validator):
         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(_u.pos = _u_2.pos_2, _u_2.col) AS col, IF(_u.pos = _u_3.pos_3, _u_3.col_2) AS col_2, IF(_u.pos = _u_4.pos_4, _u_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 ((_u.pos = _u_2.pos_2 OR (_u.pos > CARDINALITY(ARRAY[1, 2, 3]) AND _u_2.pos_2 = CARDINALITY(ARRAY[1, 2, 3]))) AND (_u.pos = _u_3.pos_3 OR (_u.pos > CARDINALITY(ARRAY[4, 5]) AND _u_3.pos_3 = CARDINALITY(ARRAY[4, 5])))) AND (_u.pos = _u_4.pos_4 OR (_u.pos > CARDINALITY(ARRAY[6]) AND _u_4.pos_4 = CARDINALITY(ARRAY[6])))",
+                "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 CROSS JOIN 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(_u.pos = _u_2.pos_2, _u_2.col) AS col, IF(_u.pos = _u_3.pos_3, _u_3.col_2) AS col_2, IF(_u.pos = _u_4.pos_4, _u_4.col_3) AS col_3 FROM x CROSS JOIN 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 ((_u.pos = _u_2.pos_2 OR (_u.pos > CARDINALITY(ARRAY[1, 2, 3]) AND _u_2.pos_2 = CARDINALITY(ARRAY[1, 2, 3]))) AND (_u.pos = _u_3.pos_3 OR (_u.pos > CARDINALITY(ARRAY[4, 5]) AND _u_3.pos_3 = CARDINALITY(ARRAY[4, 5])))) AND (_u.pos = _u_4.pos_4 OR (_u.pos > CARDINALITY(ARRAY[6]) AND _u_4.pos_4 = CARDINALITY(ARRAY[6])))",
             },
         )
         self.validate_all(
@@ -165,6 +224,13 @@ class TestDuckDB(Validator):
         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 STRFTIME(CAST('2020-01-01' AS TIMESTAMP), CONCAT('%Y', '%m'))",
+            write={
+                "duckdb": "SELECT STRFTIME(CAST('2020-01-01' AS TIMESTAMP), CONCAT('%Y', '%m'))",
+                "tsql": "SELECT FORMAT(CAST('2020-01-01' AS DATETIME2), CONCAT('yyyy', 'MM'))",
+            },
+        )
+        self.validate_all(
             "SELECT * FROM produce PIVOT(SUM(sales) FOR quarter IN ('Q1', 'Q2'))",
             read={
                 "duckdb": "SELECT * FROM produce PIVOT(SUM(sales) FOR quarter IN ('Q1', 'Q2'))",
@@ -188,7 +254,7 @@ class TestDuckDB(Validator):
             },
         )
         self.validate_all(
-            "DATE_DIFF('day', CAST(b AS DATE), CAST(a AS DATE))",
+            "DATE_DIFF('DAY', CAST(b AS DATE), CAST(a AS DATE))",
             read={
                 "duckdb": "DATE_DIFF('day', CAST(b AS DATE), CAST(a AS DATE))",
                 "hive": "DATEDIFF(a, b)",
@@ -226,15 +292,15 @@ class TestDuckDB(Validator):
         self.validate_all(
             """SELECT DATEDIFF('day', t1."A", t1."B") FROM "table" AS t1""",
             write={
-                "duckdb": """SELECT DATE_DIFF('day', t1."A", t1."B") FROM "table" AS t1""",
-                "trino": """SELECT DATE_DIFF('day', t1."A", t1."B") FROM "table" AS t1""",
+                "duckdb": """SELECT DATE_DIFF('DAY', t1."A", t1."B") FROM "table" AS t1""",
+                "trino": """SELECT DATE_DIFF('DAY', t1."A", t1."B") FROM "table" AS t1""",
             },
         )
         self.validate_all(
             "SELECT DATE_DIFF('day', DATE '2020-01-01', DATE '2020-01-05')",
             write={
-                "duckdb": "SELECT DATE_DIFF('day', CAST('2020-01-01' AS DATE), CAST('2020-01-05' AS DATE))",
-                "trino": "SELECT DATE_DIFF('day', CAST('2020-01-01' AS DATE), CAST('2020-01-05' AS DATE))",
+                "duckdb": "SELECT DATE_DIFF('DAY', CAST('2020-01-01' AS DATE), CAST('2020-01-05' AS DATE))",
+                "trino": "SELECT DATE_DIFF('DAY', CAST('2020-01-01' AS DATE), CAST('2020-01-05' AS DATE))",
             },
         )
         self.validate_all(
@@ -366,7 +432,7 @@ class TestDuckDB(Validator):
             },
         )
         self.validate_all(
-            "IF(y <> 0, x / y, NULL)",
+            "IF((y) <> 0, (x) / (y), NULL)",
             read={
                 "bigquery": "SAFE_DIVIDE(x, y)",
             },
@@ -374,6 +440,7 @@ class TestDuckDB(Validator):
         self.validate_all(
             "STRUCT_PACK(x := 1, y := '2')",
             write={
+                "bigquery": "STRUCT(1 AS x, '2' AS y)",
                 "duckdb": "{'x': 1, 'y': '2'}",
                 "spark": "STRUCT(1 AS x, '2' AS y)",
             },
@@ -381,6 +448,7 @@ class TestDuckDB(Validator):
         self.validate_all(
             "STRUCT_PACK(key1 := 'value1', key2 := 42)",
             write={
+                "bigquery": "STRUCT('value1' AS key1, 42 AS key2)",
                 "duckdb": "{'key1': 'value1', 'key2': 42}",
                 "spark": "STRUCT('value1' AS key1, 42 AS key2)",
             },
@@ -478,6 +546,8 @@ class TestDuckDB(Validator):
             "SELECT CAST('2020-05-06' AS DATE) + INTERVAL 5 DAY",
             read={"bigquery": "SELECT DATE_ADD(CAST('2020-05-06' AS DATE), INTERVAL 5 DAY)"},
         )
+        self.validate_identity("SELECT PERCENTILE_CONT(0.25) WITHIN GROUP (ORDER BY y DESC) FROM t")
+        self.validate_identity("SELECT PERCENTILE_DISC(0.25) WITHIN GROUP (ORDER BY y DESC) FROM t")
         self.validate_all(
             "SELECT QUANTILE_CONT(x, q) FROM t",
             write={
@@ -513,6 +583,14 @@ class TestDuckDB(Validator):
 
         self.validate_identity("SELECT ISNAN(x)")
 
+        self.validate_all(
+            "SELECT COUNT_IF(x)",
+            write={
+                "duckdb": "SELECT COUNT_IF(x)",
+                "bigquery": "SELECT COUNTIF(x)",
+            },
+        )
+
     def test_array_index(self):
         with self.assertLogs(helper_logger) as cm:
             self.validate_all(
@@ -569,10 +647,10 @@ class TestDuckDB(Validator):
         )
         self.validate_all(
             "SELECT INTERVAL '1 quarter'",
-            write={"duckdb": "SELECT (90 * INTERVAL '1' day)"},
+            write={"duckdb": "SELECT (90 * INTERVAL '1' DAY)"},
         )
         self.validate_all(
-            "SELECT ((DATE_TRUNC('DAY', CAST(CAST(DATE_TRUNC('DAY', CURRENT_TIMESTAMP) AS DATE) AS TIMESTAMP) + INTERVAL (0 - MOD((DAYOFWEEK(CAST(CAST(DATE_TRUNC('DAY', CURRENT_TIMESTAMP) AS DATE) AS TIMESTAMP)) % 7) - 1 + 7, 7)) day) + (7 * INTERVAL (-5) day))) AS t1",
+            "SELECT ((DATE_TRUNC('DAY', CAST(CAST(DATE_TRUNC('DAY', CURRENT_TIMESTAMP) AS DATE) AS TIMESTAMP) + INTERVAL (0 - MOD((DAYOFWEEK(CAST(CAST(DATE_TRUNC('DAY', CURRENT_TIMESTAMP) AS DATE) AS TIMESTAMP)) % 7) - 1 + 7, 7)) DAY) + (7 * INTERVAL (-5) DAY))) AS t1",
             read={
                 "presto": "SELECT ((DATE_ADD('week', -5, DATE_TRUNC('DAY', DATE_ADD('day', (0 - MOD((DAY_OF_WEEK(CAST(CAST(DATE_TRUNC('DAY', NOW()) AS DATE) AS TIMESTAMP)) % 7) - 1 + 7, 7)), CAST(CAST(DATE_TRUNC('DAY', NOW()) AS DATE) AS TIMESTAMP)))))) AS t1",
             },
@@ -594,7 +672,7 @@ class TestDuckDB(Validator):
             write={
                 "bigquery": "TIMESTAMP_MILLIS(x)",
                 "duckdb": "EPOCH_MS(x)",
-                "presto": "FROM_UNIXTIME(CAST(x AS DOUBLE) / 1000)",
+                "presto": "FROM_UNIXTIME(CAST(x AS DOUBLE) / POW(10, 3))",
                 "spark": "TIMESTAMP_MILLIS(x)",
             },
         )
@@ -659,36 +737,44 @@ class TestDuckDB(Validator):
         )
 
     def test_sample(self):
-        self.validate_all(
+        self.validate_identity(
             "SELECT * FROM tbl USING SAMPLE 5",
-            write={"duckdb": "SELECT * FROM tbl USING SAMPLE (5)"},
+            "SELECT * FROM tbl USING SAMPLE (5 ROWS)",
         )
-        self.validate_all(
+        self.validate_identity(
             "SELECT * FROM tbl USING SAMPLE 10%",
-            write={"duckdb": "SELECT * FROM tbl USING SAMPLE (10 PERCENT)"},
+            "SELECT * FROM tbl USING SAMPLE (10 PERCENT)",
         )
-        self.validate_all(
+        self.validate_identity(
             "SELECT * FROM tbl USING SAMPLE 10 PERCENT (bernoulli)",
-            write={"duckdb": "SELECT * FROM tbl USING SAMPLE BERNOULLI (10 PERCENT)"},
+            "SELECT * FROM tbl USING SAMPLE BERNOULLI (10 PERCENT)",
         )
-        self.validate_all(
+        self.validate_identity(
             "SELECT * FROM tbl USING SAMPLE reservoir(50 ROWS) REPEATABLE (100)",
-            write={"duckdb": "SELECT * FROM tbl USING SAMPLE RESERVOIR (50 ROWS) REPEATABLE (100)"},
+            "SELECT * FROM tbl USING SAMPLE RESERVOIR (50 ROWS) REPEATABLE (100)",
         )
-        self.validate_all(
+        self.validate_identity(
             "SELECT * FROM tbl USING SAMPLE 10% (system, 377)",
-            write={"duckdb": "SELECT * FROM tbl USING SAMPLE SYSTEM (10 PERCENT) REPEATABLE (377)"},
+            "SELECT * FROM tbl USING SAMPLE SYSTEM (10 PERCENT) REPEATABLE (377)",
         )
-        self.validate_all(
+        self.validate_identity(
             "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"
-            },
+            "SELECT * FROM tbl TABLESAMPLE RESERVOIR (20 PERCENT), tbl2 WHERE tbl.i = tbl2.i",
         )
-        self.validate_all(
+        self.validate_identity(
             "SELECT * FROM tbl, tbl2 WHERE tbl.i=tbl2.i USING SAMPLE RESERVOIR(20%)",
+            "SELECT * FROM tbl, tbl2 WHERE tbl.i = tbl2.i USING SAMPLE RESERVOIR (20 PERCENT)",
+        )
+
+        self.validate_all(
+            "SELECT * FROM example TABLESAMPLE (3 ROWS) REPEATABLE (82)",
+            read={
+                "duckdb": "SELECT * FROM example TABLESAMPLE (3) REPEATABLE (82)",
+                "snowflake": "SELECT * FROM example SAMPLE (3 ROWS) SEED (82)",
+            },
             write={
-                "duckdb": "SELECT * FROM tbl, tbl2 WHERE tbl.i = tbl2.i USING SAMPLE RESERVOIR (20 PERCENT)"
+                "duckdb": "SELECT * FROM example TABLESAMPLE (3 ROWS) REPEATABLE (82)",
+                "snowflake": "SELECT * FROM example TABLESAMPLE (3 ROWS) SEED (82)",
             },
         )