Fix three match cartesian sequential scanning (#1555)

src/query/plan/rule_based_planner.hpp

@@ -511,10 +511,6 @@ class RuleBasedPlanner {
 
     std::set<ExpansionGroupId> visited_expansion_groups;
 
-    last_op =
-        GenerateExpansionOnAlreadySeenSymbols(std::move(last_op), matching, visited_expansion_groups, symbol_table,
-                                              storage, bound_symbols, new_symbols, named_paths, filters, view);
-
     // We want to create separate branches of scan operators for each expansion group group of patterns
     // Whenever there are 2 scan branches, they will be joined with a Cartesian operator
 
@@ -528,6 +524,14 @@ class RuleBasedPlanner {
         continue;
       }
 
+      last_op =
+          GenerateExpansionOnAlreadySeenSymbols(std::move(last_op), matching, visited_expansion_groups, symbol_table,
+                                                storage, bound_symbols, new_symbols, named_paths, filters, view);
+
+      if (visited_expansion_groups.contains(expansion.expansion_group_id)) {
+        continue;
+      }
+
       std::unique_ptr<LogicalOperator> starting_expansion_operator = nullptr;
       if (!initial_expansion_done) {
         starting_expansion_operator = std::move(last_op);

tests/e2e/CMakeLists.txt

@@ -73,6 +73,7 @@ add_subdirectory(inspect_query)
 add_subdirectory(filter_info)
 add_subdirectory(queries)
 add_subdirectory(garbage_collection)
+add_subdirectory(query_planning)
 
 copy_e2e_python_files(pytest_runner pytest_runner.sh "")
 copy_e2e_python_files(x x.sh "")

tests/e2e/query_planning/CMakeLists.txt

@@ -0,0 +1,6 @@
+function(copy_query_planning_e2e_python_files FILE_NAME)
+    copy_e2e_python_files(query_planning ${FILE_NAME})
+endfunction()
+
+copy_query_planning_e2e_python_files(common.py)
+copy_query_planning_e2e_python_files(query_planning_cartesian.py)

tests/e2e/query_planning/common.py

@@ -0,0 +1,24 @@
+# Copyright 2023 Memgraph Ltd.
+#
+# Use of this software is governed by the Business Source License
+# included in the file licenses/BSL.txt; by using this file, you agree to be bound by the terms of the Business Source
+# License, and you may not use this file except in compliance with the Business Source License.
+#
+# As of the Change Date specified in that file, in accordance with
+# the Business Source License, use of this software will be governed
+# by the Apache License, Version 2.0, included in the file
+# licenses/APL.txt.
+
+import pytest
+from gqlalchemy import Memgraph
+
+
+@pytest.fixture
+def memgraph(**kwargs) -> Memgraph:
+    memgraph = Memgraph()
+
+    yield memgraph
+
+    memgraph.drop_indexes()
+    memgraph.ensure_constraints([])
+    memgraph.drop_database()

tests/e2e/query_planning/query_planning_cartesian.py

@@ -0,0 +1,42 @@
+# Copyright 2023 Memgraph Ltd.
+#
+# Use of this software is governed by the Business Source License
+# included in the file licenses/BSL.txt; by using this file, you agree to be bound by the terms of the Business Source
+# License, and you may not use this file except in compliance with the Business Source License.
+#
+# As of the Change Date specified in that file, in accordance with
+# the Business Source License, use of this software will be governed
+# by the Apache License, Version 2.0, included in the file
+# licenses/APL.txt.
+
+import sys
+
+import pytest
+from common import memgraph
+
+QUERY_PLAN = "QUERY PLAN"
+
+
+def test_indexed_join_with_indices(memgraph):
+    memgraph.execute("CREATE INDEX ON :Node(id);")
+
+    expected_explain = [
+        f" * Produce {{a, b, r}}",
+        f" * Filter (a :Node), {{a.id}}",
+        f" * Expand (b)-[r:EDGE]-(a)",
+        f" * ScanAllByLabelPropertyValue (b :Node {{id}})",
+        f" * Once",
+    ]
+
+    results = list(
+        memgraph.execute_and_fetch(
+            "EXPLAIN MATCH (a:Node {id: 1}) MATCH (b:Node {id: 2}) MATCH (a)-[r:EDGE]-(b) return a,b,r;"
+        )
+    )
+    actual_explain = [x[QUERY_PLAN] for x in results]
+
+    assert expected_explain == actual_explain
+
+
+if __name__ == "__main__":
+    sys.exit(pytest.main([__file__, "-rA"]))

tests/e2e/query_planning/workloads.yaml

@@ -0,0 +1,14 @@
+queries_cluster: &queries_cluster
+  cluster:
+    main:
+      args: ["--bolt-port", "7687", "--log-level=TRACE"]
+      log_file: "query_planning.log"
+      setup_queries: []
+      validation_queries: []
+
+
+workloads:
+  - name: "Query planning cartesian"
+    binary: "tests/e2e/pytest_runner.sh"
+    args: ["query_planning/query_planning_cartesian.py"]
+    <<: *queries_cluster