Merge branch 'master' into Implement-constant-time-label-and-edge-type-retrieval

tests/e2e/CMakeLists.txt

@@ -71,8 +71,10 @@ add_subdirectory(index_hints)
 add_subdirectory(query_modules)
 add_subdirectory(constraints)
 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/analyze_graph/optimize_indexes.py

@@ -62,7 +62,7 @@ def test_analyze_graph_delete_statistics(delete_query, multi_db):
     # After deleting statistics, id2 should be chosen because it has less vertices
     expected_explain_after_delete_analysis = [
         (f" * Produce {{n}}",),
-        (f" * Filter (n :Label), {{n.id1}}, {{n.id2}}",),
+        (f" * Filter {{n.id1}}",),
         (f" * ScanAllByLabelPropertyValue (n :Label {{id2}})",),
         (f" * Once",),
     ]
@@ -96,7 +96,7 @@ def test_analyze_full_graph(analyze_query, multi_db):
     # Choose id2 before tha analysis because it has less vertices
     expected_explain_before_analysis = [
         (f" * Produce {{n}}",),
-        (f" * Filter (n :Label), {{n.id1}}, {{n.id2}}",),
+        (f" * Filter {{n.id1}}",),
         (f" * ScanAllByLabelPropertyValue (n :Label {{id2}})",),
         (f" * Once",),
     ]
@@ -117,7 +117,7 @@ def test_analyze_full_graph(analyze_query, multi_db):
     # After analyzing graph, id1 index should be chosen because it has smaller average group size
     expected_explain_after_analysis = [
         (f" * Produce {{n}}",),
-        (f" * Filter (n :Label), {{n.id1}}, {{n.id2}}",),
+        (f" * Filter {{n.id2}}",),
         (f" * ScanAllByLabelPropertyValue (n :Label {{id1}})",),
         (f" * Once",),
     ]
@@ -152,7 +152,7 @@ def test_cardinality_different_avg_group_size_uniform_dist(multi_db):
     assert analyze_graph_results[1 - first_index] == ("Label", "id2", 100, 20, 5, 0, 0)
     expected_explain_after_analysis = [
         (f" * Produce {{n}}",),
-        (f" * Filter (n :Label), {{n.id1}}, {{n.id2}}",),
+        (f" * Filter {{n.id2}}",),
         (f" * ScanAllByLabelPropertyValue (n :Label {{id1}})",),
         (f" * Once",),
     ]
@@ -183,7 +183,7 @@ def test_cardinality_same_avg_group_size_uniform_dist_diff_vertex_count(multi_db
     assert analyze_graph_results[1 - first_index] == ("Label", "id2", 50, 50, 1, 0, 0)
     expected_explain_after_analysis = [
         (f" * Produce {{n}}",),
-        (f" * Filter (n :Label), {{n.id1}}, {{n.id2}}",),
+        (f" * Filter {{n.id1}}",),
         (f" * ScanAllByLabelPropertyValue (n :Label {{id2}})",),
         (f" * Once",),
     ]
@@ -214,7 +214,7 @@ def test_large_diff_in_num_vertices_v1(multi_db):
     assert analyze_graph_results[1 - first_index] == ("Label", "id2", 99, 1, 99, 0, 0)
     expected_explain_after_analysis = [
         (f" * Produce {{n}}",),
-        (f" * Filter (n :Label), {{n.id1}}, {{n.id2}}",),
+        (f" * Filter {{n.id1}}",),
         (f" * ScanAllByLabelPropertyValue (n :Label {{id2}})",),
         (f" * Once",),
     ]
@@ -245,7 +245,7 @@ def test_large_diff_in_num_vertices_v2(multi_db):
     assert analyze_graph_results[1 - first_index] == ("Label", "id2", 1000, 1000, 1, 0, 0)
     expected_explain_after_analysis = [
         (f" * Produce {{n}}",),
-        (f" * Filter (n :Label), {{n.id1}}, {{n.id2}}",),
+        (f" * Filter {{n.id2}}",),
         (f" * ScanAllByLabelPropertyValue (n :Label {{id1}})",),
         (f" * Once",),
     ]
@@ -286,7 +286,7 @@ def test_same_avg_group_size_diff_distribution(multi_db):
     assert analyze_graph_results[1 - first_index] == ("Label", "id2", 100, 5, 20, 0, 0)
     expected_explain_after_analysis = [
         (f" * Produce {{n}}",),
-        (f" * Filter (n :Label), {{n.id1}}, {{n.id2}}",),
+        (f" * Filter {{n.id1}}",),
         (f" * ScanAllByLabelPropertyValue (n :Label {{id2}})",),
         (f" * Once",),
     ]

tests/e2e/filter_info/CMakeLists.txt

@@ -0,0 +1,6 @@
+function(copy_filter_info_e2e_python_files FILE_NAME)
+    copy_e2e_python_files(filter_info ${FILE_NAME})
+endfunction()
+
+copy_filter_info_e2e_python_files(common.py)
+copy_filter_info_e2e_python_files(filter_info.py)

tests/e2e/filter_info/common.py

@@ -0,0 +1,23 @@
+# 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_database()
+    memgraph.drop_indexes()

tests/e2e/filter_info/filter_info.py

@@ -0,0 +1,39 @@
+# 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
+
+
+def test_label_index_hint(memgraph):
+    memgraph.execute("CREATE (n:Label1:Label2 {prop: 1});")
+    memgraph.execute("CREATE INDEX ON :Label1;")
+
+    # TODO: Fix this test since it should only filter on :Label2 and prop
+    expected_explain = [
+        " * Produce {n}",
+        " * Filter (n :Label1:Label2), {n.prop}",
+        " * ScanAllByLabel (n :Label1)",
+        " * Once",
+    ]
+
+    actual_explain = [
+        row["QUERY PLAN"]
+        for row in memgraph.execute_and_fetch("EXPLAIN MATCH (n:Label1:Label2) WHERE n.prop = 1 return n;")
+    ]
+
+    assert expected_explain == actual_explain
+
+
+if __name__ == "__main__":
+    sys.exit(pytest.main([__file__, "-rA"]))

tests/e2e/filter_info/workloads.yaml

@@ -0,0 +1,13 @@
+filter_info_cluster: &filter_info_cluster
+  cluster:
+    main:
+      args: ["--bolt-port", "7687", "--log-level=TRACE"]
+      log_file: "filter_info.log"
+      setup_queries: []
+      validation_queries: []
+
+workloads:
+  - name: "Filter info information"
+    binary: "tests/e2e/pytest_runner.sh"
+    args: ["filter_info/filter_info.py"]
+    <<: *filter_info_cluster

tests/e2e/index_hints/index_hints.py

@@ -162,12 +162,13 @@ def test_label_property_index_hint(memgraph):
 
     expected_explain_no_hint = [
         " * Produce {n}",
-        " * Filter (n :Label), {n.id1}, {n.id2}",
+        " * Filter {n.id1}",
         " * ScanAllByLabelPropertyValue (n :Label {id2})",
         " * Once",
     ]
     expected_explain_with_hint = [
-        row.replace("(n :Label {id2})", "(n :Label {id1})") for row in expected_explain_no_hint
+        row.replace("(n :Label {id2})", "(n :Label {id1})").replace(" * Filter {n.id1}", " * Filter {n.id2}")
+        for row in expected_explain_no_hint
     ]
 
     explain_no_hint = [
@@ -192,7 +193,7 @@ def test_label_property_index_hint_alternative_orderings(memgraph):
 
     expected_explain_with_hint = [
         " * Produce {n}",
-        " * Filter (n :Label), {n.id1}, {n.id2}",
+        " * Filter {n.id2}",
         " * ScanAllByLabelPropertyValue (n :Label {id1})",
         " * Once",
     ]
@@ -221,7 +222,7 @@ def test_multiple_label_property_index_hints(memgraph):
 
     expected_explain_with_hint = [
         " * Produce {n}",
-        " * Filter (n :Label), {n.id1}, {n.id2}",
+        " * Filter {n.id2}",
         " * ScanAllByLabelPropertyValue (n :Label {id1})",
         " * Once",
     ]
@@ -251,7 +252,7 @@ def test_multiple_applicable_label_property_index_hints(memgraph):
 
     expected_explain_with_hint = [
         " * Produce {n}",
-        " * Filter (n :Label), {n.id1}, {n.id2}",
+        " * Filter {n.id2}",
         " * ScanAllByLabelPropertyValue (n :Label {id1})",
         " * Once",
     ]
@@ -275,12 +276,13 @@ def test_multiple_applicable_label_property_index_hints_alternative_orderings(me
 
     expected_explain_with_hint_1 = [
         " * Produce {n}",
-        " * Filter (n :Label), {n.id1}, {n.id2}",
+        " * Filter {n.id2}",
         " * ScanAllByLabelPropertyValue (n :Label {id1})",
         " * Once",
     ]
     expected_explain_with_hint_2 = [
-        row.replace("(n :Label {id1})", "(n :Label {id2})") for row in expected_explain_with_hint_1
+        row.replace("(n :Label {id1})", "(n :Label {id2})").replace(" * Filter {n.id2}", " * Filter {n.id1}")
+        for row in expected_explain_with_hint_1
     ]
 
     explain_with_hint_ordering_1a = [
@@ -407,6 +409,7 @@ def test_multiple_match_query(memgraph):
     memgraph.execute("CREATE INDEX ON :Label2;")
     memgraph.execute("CREATE INDEX ON :Label3;")
 
+    # TODO: Fix this test since it has the filtering info wrong (filtering by label that's already indexed)
     expected_explain_with_hint = [
         " * Produce {n, m}",
         " * Cartesian {m : n}",
@@ -414,7 +417,7 @@ def test_multiple_match_query(memgraph):
         " | * Filter (n :Label1:Label2), {n.id}",
         " | * ScanAllByLabel (n :Label1)",
         " | * Once",
-        " * Filter (m :Label2:Label3), (n :Label1:Label2), {n.id}",
+        " * Filter (m :Label2:Label3)",
         " * ScanAllByLabel (m :Label2)",
         " * Once",
     ]

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