From f9e35d418164c849a5468acb747bc5dee40f2838 Mon Sep 17 00:00:00 2001 From: vaggelisd Date: Tue, 8 Oct 2024 13:47:58 +0300 Subject: [PATCH 1/3] feat: Make exp.Update be a DML node --- sqlglot/expressions.py | 2 +- sqlglot/optimizer/scope.py | 4 ++-- tests/test_optimizer.py | 6 ++++++ 3 files changed, 9 insertions(+), 3 deletions(-) diff --git a/sqlglot/expressions.py b/sqlglot/expressions.py index 77a067404b..8bfbfdd874 100644 --- a/sqlglot/expressions.py +++ b/sqlglot/expressions.py @@ -3278,7 +3278,7 @@ class Intersect(SetOperation): pass -class Update(Expression): +class Update(DML): arg_types = { "with": False, "this": False, diff --git a/sqlglot/optimizer/scope.py b/sqlglot/optimizer/scope.py index fafd61fe15..0bd7ab26f0 100644 --- a/sqlglot/optimizer/scope.py +++ b/sqlglot/optimizer/scope.py @@ -562,8 +562,8 @@ def _traverse_scope(scope): elif isinstance(expression, exp.DML): yield from _traverse_ctes(scope) for query in find_all_in_scope(expression, exp.Query): - # This check ensures we don't yield the CTE queries twice - if not isinstance(query.parent, exp.CTE): + # This check ensures we don't yield the CTE/nested queries twice + if not isinstance(query.parent, (exp.CTE, exp.Subquery)): yield from _traverse_scope(Scope(query, cte_sources=scope.cte_sources)) return else: diff --git a/tests/test_optimizer.py b/tests/test_optimizer.py index 2c2015b3d9..b1c2b7bbae 100644 --- a/tests/test_optimizer.py +++ b/tests/test_optimizer.py @@ -762,6 +762,12 @@ def test_scope(self): ) self.assertEqual(set(scopes[3].sources), {""}) + sql = ( + "UPDATE customers SET total_spent = (SELECT 1 FROM t1) WHERE EXISTS (SELECT 1 FROM t2)" + ) + expression = parse_one(sql) + self.assertEqual(len(traverse_scope(expression)), 3) + inner_query = "SELECT bar FROM baz" for udtf in (f"UNNEST(({inner_query}))", f"LATERAL ({inner_query})"): sql = f"SELECT a FROM foo CROSS JOIN {udtf}" From 8a9bb9ca9f01b646013b83e67eb1c0a1172ca75f Mon Sep 17 00:00:00 2001 From: vaggelisd Date: Tue, 8 Oct 2024 14:46:23 +0300 Subject: [PATCH 2/3] Add more tests --- tests/test_optimizer.py | 18 ++++++++++++------ 1 file changed, 12 insertions(+), 6 deletions(-) diff --git a/tests/test_optimizer.py b/tests/test_optimizer.py index b1c2b7bbae..2a51d84cac 100644 --- a/tests/test_optimizer.py +++ b/tests/test_optimizer.py @@ -762,12 +762,6 @@ def test_scope(self): ) self.assertEqual(set(scopes[3].sources), {""}) - sql = ( - "UPDATE customers SET total_spent = (SELECT 1 FROM t1) WHERE EXISTS (SELECT 1 FROM t2)" - ) - expression = parse_one(sql) - self.assertEqual(len(traverse_scope(expression)), 3) - inner_query = "SELECT bar FROM baz" for udtf in (f"UNNEST(({inner_query}))", f"LATERAL ({inner_query})"): sql = f"SELECT a FROM foo CROSS JOIN {udtf}" @@ -785,6 +779,18 @@ def test_scope(self): self.assertEqual(scopes[2].expression.sql(), f"SELECT a FROM foo CROSS JOIN {udtf}") self.assertEqual(set(scopes[2].sources), {"", "foo"}) + # Check DML statement scopes + sql = ( + "UPDATE customers SET total_spent = (SELECT 1 FROM t1) WHERE EXISTS (SELECT 1 FROM t2)" + ) + self.assertEqual(len(traverse_scope(parse_one(sql))), 3) + + sql = "UPDATE tbl1 SET col = 1 WHERE EXISTS (SELECT 1 FROM tbl2 WHERE tbl1.id = tbl2.id)" + self.assertEqual(len(traverse_scope(parse_one(sql))), 1) + + sql = "UPDATE tbl1 SET col = 0" + self.assertEqual(len(traverse_scope(parse_one(sql))), 0) + @patch("sqlglot.optimizer.scope.logger") def test_scope_warning(self, logger): self.assertEqual(len(traverse_scope(parse_one("WITH q AS (@y) SELECT * FROM q"))), 1) From 87d18fb0a2b4c74503018d529dde4192434064d2 Mon Sep 17 00:00:00 2001 From: vaggelisd Date: Tue, 8 Oct 2024 15:17:26 +0300 Subject: [PATCH 3/3] Add qualify test --- tests/fixtures/optimizer/qualify_tables.sql | 3 +++ 1 file changed, 3 insertions(+) diff --git a/tests/fixtures/optimizer/qualify_tables.sql b/tests/fixtures/optimizer/qualify_tables.sql index 61d0b9677f..49e07faf9b 100644 --- a/tests/fixtures/optimizer/qualify_tables.sql +++ b/tests/fixtures/optimizer/qualify_tables.sql @@ -184,3 +184,6 @@ COPY INTO (SELECT * FROM c.db.x AS x) TO 'data' WITH (FORMAT 'CSV'); # title: tablesample SELECT 1 FROM x TABLESAMPLE SYSTEM (10 PERCENT) CROSS JOIN y TABLESAMPLE SYSTEM (10 PERCENT); SELECT 1 FROM c.db.x AS x TABLESAMPLE SYSTEM (10 PERCENT) CROSS JOIN c.db.y AS y TABLESAMPLE SYSTEM (10 PERCENT); + +WITH cte_tbl AS (SELECT 1 AS col2) UPDATE y SET col1 = (SELECT * FROM x) WHERE EXISTS(SELECT 1 FROM cte_tbl); +WITH cte_tbl AS (SELECT 1 AS col2) UPDATE c.db.y SET col1 = (SELECT * FROM c.db.x AS x) WHERE EXISTS(SELECT 1 FROM cte_tbl AS cte_tbl);