python/sqlglot
1
0
Fork 0
Code Issues Activity

Merging upstream version 26.24.0.

Browse source 
Signed-off-by: Daniel Baumann <daniel@debian.org>
This commit is contained in:
Daniel Baumann 2025-06-02 05:53:55 +02:00
parent c78999c8c9
commit 2b9f8478b0
Signed by: daniel
GPG key ID: FBB4F0E80A80222F
53 changed files with 3642 additions and 3447 deletions
Download patch file Download diff file Expand all files Collapse all files

56
tests/dialects/test_bigquery.py
View file

@ -55,6 +55,10 @@ class TestBigQuery(Validator):
select_with_quoted_udf = self.validate_identity("SELECT `p.d.UdF`(data) FROM `p.d.t`")
self.assertEqual(select_with_quoted_udf.selects[0].name, "p.d.UdF")
self.validate_identity("ARRAY_CONCAT_AGG(x ORDER BY ARRAY_LENGTH(x) LIMIT 2)")
self.validate_identity("ARRAY_CONCAT_AGG(x LIMIT 2)")
self.validate_identity("ARRAY_CONCAT_AGG(x ORDER BY ARRAY_LENGTH(x))")
self.validate_identity("ARRAY_CONCAT_AGG(x)")
self.validate_identity("PARSE_TIMESTAMP('%Y-%m-%dT%H:%M:%E*S%z', x)")
self.validate_identity("SELECT ARRAY_CONCAT([1])")
self.validate_identity("SELECT * FROM READ_CSV('bla.csv')")
@ -2560,3 +2564,55 @@ OPTIONS (
self.assertEqual(qualified.sql("bigquery"), "SELECT * FROM `P`.`D`.`T` AS `T`")
finally:
BigQuery.NORMALIZATION_STRATEGY = NormalizationStrategy.CASE_INSENSITIVE
def test_array_agg(self):
for distinct in ("", "DISTINCT "):
self.validate_all(
f"SELECT ARRAY_AGG({distinct}x ORDER BY x)",
write={
"bigquery": f"SELECT ARRAY_AGG({distinct}x ORDER BY x)",
"snowflake": f"SELECT ARRAY_AGG({distinct}x) WITHIN GROUP (ORDER BY x NULLS FIRST)",
},
)
for nulls in ("", " IGNORE NULLS", " RESPECT NULLS"):
self.validate_all(
f"SELECT ARRAY_AGG(x{nulls} ORDER BY col1 ASC, col2 DESC)",
write={
"bigquery": f"SELECT ARRAY_AGG(x{nulls} ORDER BY col1 ASC, col2 DESC)",
"snowflake": "SELECT ARRAY_AGG(x) WITHIN GROUP (ORDER BY col1 ASC NULLS FIRST, col2 DESC NULLS LAST)",
},
)
def test_select_as_struct(self):
self.validate_all(
"SELECT ARRAY(SELECT AS STRUCT x1 AS x1, x2 AS x2 FROM t) AS array_col",
write={
"bigquery": "SELECT ARRAY(SELECT AS STRUCT x1 AS x1, x2 AS x2 FROM t) AS array_col",
"snowflake": "SELECT (SELECT ARRAY_AGG(OBJECT_CONSTRUCT('x1', x1, 'x2', x2)) FROM t) AS array_col",
},
)
self.validate_all(
"WITH t1 AS (SELECT ARRAY(SELECT AS STRUCT x1 AS alias_x1, x2 /* test */ FROM t2) AS array_col) SELECT array_col[0].alias_x1, array_col[0].x2 FROM t1",
write={
"bigquery": "WITH t1 AS (SELECT ARRAY(SELECT AS STRUCT x1 AS alias_x1, x2 /* test */ FROM t2) AS array_col) SELECT array_col[0].alias_x1, array_col[0].x2 FROM t1",
"snowflake": "WITH t1 AS (SELECT (SELECT ARRAY_AGG(OBJECT_CONSTRUCT('alias_x1', x1, 'x2', x2 /* test */)) FROM t2) AS array_col) SELECT array_col[0].alias_x1, array_col[0].x2 FROM t1",
},
)
self.validate_all(
"WITH t1 AS (SELECT ARRAY(SELECT AS STRUCT 1 AS a, 2 AS b) AS array_col) SELECT array_col[0].a, array_col[0].b FROM t1",
write={
"bigquery": "WITH t1 AS (SELECT ARRAY(SELECT AS STRUCT 1 AS a, 2 AS b) AS array_col) SELECT array_col[0].a, array_col[0].b FROM t1",
"snowflake": "WITH t1 AS (SELECT (SELECT ARRAY_AGG(OBJECT_CONSTRUCT('a', 1, 'b', 2))) AS array_col) SELECT array_col[0].a, array_col[0].b FROM t1",
},
)
self.validate_all(
"WITH t1 AS (SELECT ARRAY(SELECT AS STRUCT x1 AS alias_x1, x2 /* test */ FROM t2 WHERE x2 = 4) AS array_col) SELECT array_col[0].alias_x1, array_col[0].x2 FROM t1",
write={
"bigquery": "WITH t1 AS (SELECT ARRAY(SELECT AS STRUCT x1 AS alias_x1, x2 /* test */ FROM t2 WHERE x2 = 4) AS array_col) SELECT array_col[0].alias_x1, array_col[0].x2 FROM t1",
"snowflake": "WITH t1 AS (SELECT (SELECT ARRAY_AGG(OBJECT_CONSTRUCT('alias_x1', x1, 'x2', x2 /* test */)) FROM t2 WHERE x2 = 4) AS array_col) SELECT array_col[0].alias_x1, array_col[0].x2 FROM t1",
},
)

2
tests/dialects/test_hive.py
View file

@ -161,7 +161,7 @@ class TestHive(Validator):
"CREATE TABLE test STORED AS parquet TBLPROPERTIES ('x'='1', 'Z'='2') AS SELECT 1",
write={
"duckdb": "CREATE TABLE test AS SELECT 1",
"presto": "CREATE TABLE test WITH (FORMAT='PARQUET', x='1', Z='2') AS SELECT 1",
"presto": "CREATE TABLE test WITH (format='parquet', x='1', Z='2') AS SELECT 1",
"hive": "CREATE TABLE test STORED AS PARQUET TBLPROPERTIES ('x'='1', 'Z'='2') AS SELECT 1",
"spark": "CREATE TABLE test USING PARQUET TBLPROPERTIES ('x'='1', 'Z'='2') AS SELECT 1",
},

6
tests/dialects/test_presto.py
View file

@ -465,7 +465,7 @@ class TestPresto(Validator):
"CREATE TABLE test WITH (FORMAT = 'PARQUET') AS SELECT 1",
write={
"duckdb": "CREATE TABLE test AS SELECT 1",
"presto": "CREATE TABLE test WITH (FORMAT='PARQUET') AS SELECT 1",
"presto": "CREATE TABLE test WITH (format='PARQUET') AS SELECT 1",
"hive": "CREATE TABLE test STORED AS PARQUET AS SELECT 1",
"spark": "CREATE TABLE test USING PARQUET AS SELECT 1",
},
@ -474,7 +474,7 @@ class TestPresto(Validator):
"CREATE TABLE test STORED AS 'PARQUET' AS SELECT 1",
write={
"duckdb": "CREATE TABLE test AS SELECT 1",
"presto": "CREATE TABLE test WITH (FORMAT='PARQUET') AS SELECT 1",
"presto": "CREATE TABLE test WITH (format='PARQUET') AS SELECT 1",
"hive": "CREATE TABLE test STORED AS PARQUET AS SELECT 1",
"spark": "CREATE TABLE test USING PARQUET AS SELECT 1",
},
@ -483,7 +483,7 @@ class TestPresto(Validator):
"CREATE TABLE test WITH (FORMAT = 'PARQUET', X = '1', Z = '2') AS SELECT 1",
write={
"duckdb": "CREATE TABLE test AS SELECT 1",
"presto": "CREATE TABLE test WITH (FORMAT='PARQUET', X='1', Z='2') AS SELECT 1",
"presto": "CREATE TABLE test WITH (format='PARQUET', X='1', Z='2') AS SELECT 1",
"hive": "CREATE TABLE test STORED AS PARQUET TBLPROPERTIES ('X'='1', 'Z'='2') AS SELECT 1",
"spark": "CREATE TABLE test USING PARQUET TBLPROPERTIES ('X'='1', 'Z'='2') AS SELECT 1",
},

17
tests/dialects/test_snowflake.py
View file

@ -1071,6 +1071,14 @@ class TestSnowflake(Validator):
},
)
self.validate_all(
"SELECT DATE_PART(WEEKISO, CAST('2013-12-25' AS DATE))",
read={
"bigquery": "SELECT EXTRACT(ISOWEEK FROM CAST('2013-12-25' AS DATE))",
"snowflake": "SELECT DATE_PART(WEEKISO, CAST('2013-12-25' AS DATE))",
},
)
def test_null_treatment(self):
self.validate_all(
r"SELECT FIRST_VALUE(TABLE1.COLUMN1) OVER (PARTITION BY RANDOM_COLUMN1, RANDOM_COLUMN2 ROWS BETWEEN UNBOUNDED PRECEDING AND UNBOUNDED FOLLOWING) AS MY_ALIAS FROM TABLE1",
@ -1102,6 +1110,15 @@ class TestSnowflake(Validator):
"snowflake": r"SELECT FIRST_VALUE(TABLE1.COLUMN1) IGNORE NULLS OVER (PARTITION BY RANDOM_COLUMN1, RANDOM_COLUMN2 ROWS BETWEEN UNBOUNDED PRECEDING AND UNBOUNDED FOLLOWING) AS MY_ALIAS FROM TABLE1"
},
)
self.validate_all(
"SELECT * FROM foo WHERE 'str' IN (SELECT value FROM TABLE(FLATTEN(INPUT => vals)) AS _u(seq, key, path, index, value, this))",
read={
"bigquery": "SELECT * FROM foo WHERE 'str' IN UNNEST(vals)",
},
write={
"snowflake": "SELECT * FROM foo WHERE 'str' IN (SELECT value FROM TABLE(FLATTEN(INPUT => vals)) AS _u(seq, key, path, index, value, this))",
},
)
def test_staged_files(self):
# Ensure we don't treat staged file paths as identifiers (i.e. they're not normalized)

8
tests/dialects/test_spark.py
View file

@ -56,7 +56,7 @@ class TestSpark(Validator):
"CREATE TABLE x USING ICEBERG PARTITIONED BY (MONTHS(y)) LOCATION 's3://z'",
write={
"duckdb": "CREATE TABLE x",
"presto": "CREATE TABLE x WITH (FORMAT='ICEBERG', PARTITIONED_BY=ARRAY['MONTHS(y)'])",
"presto": "CREATE TABLE x WITH (format='ICEBERG', PARTITIONED_BY=ARRAY['MONTHS(y)'])",
"hive": "CREATE TABLE x STORED AS ICEBERG PARTITIONED BY (MONTHS(y)) LOCATION 's3://z'",
"spark": "CREATE TABLE x USING ICEBERG PARTITIONED BY (MONTHS(y)) LOCATION 's3://z'",
},
@ -65,7 +65,9 @@ class TestSpark(Validator):
"CREATE TABLE test STORED AS PARQUET AS SELECT 1",
write={
"duckdb": "CREATE TABLE test AS SELECT 1",
"presto": "CREATE TABLE test WITH (FORMAT='PARQUET') AS SELECT 1",
"presto": "CREATE TABLE test WITH (format='PARQUET') AS SELECT 1",
"trino": "CREATE TABLE test WITH (format='PARQUET') AS SELECT 1",
"athena": "CREATE TABLE test WITH (format='PARQUET') AS SELECT 1", # note: lowercase format property is important for Athena
"hive": "CREATE TABLE test STORED AS PARQUET AS SELECT 1",
"spark": "CREATE TABLE test USING PARQUET AS SELECT 1",
},
@ -83,7 +85,7 @@ class TestSpark(Validator):
COMMENT 'Test comment: blah'
WITH (
PARTITIONED_BY=ARRAY['date'],
FORMAT='ICEBERG',
format='ICEBERG',
x='1'
)""",
"hive": """CREATE TABLE blah (