fix(optimizer): Avoid extra queries for prefetches with existing pref…

…etch hints

Fix #559

strawberry_django/optimizer.py

@@ -763,6 +763,13 @@ def _get_hints_from_django_relation(
         store.prefetch_related.append(model_fieldname)
         return store
 
+    field_store = getattr(field, "store", None)
+    if field_store and field_store.prefetch_related:
+        # When we have a prefetch_related in the field, skip the optimization as
+        # it will probably be filtering/annotating the queryset differently, and doing
+        # it here would result in an extra unused query
+        return store
+
     remote_field = model_field.remote_field
     remote_model = remote_field.model
     field_store = _get_model_hints(

tests/test_optimizer.py

@@ -1365,3 +1365,63 @@ def test_nested_prefetch_with_get_queryset(
         },
     }
     mock_get_queryset.assert_called_once()
+
+
+@pytest.mark.django_db(transaction=True)
+def test_ensure_prefetch_hint_with_field_as_same_name_as_the_model_doesnt_cause_extra_db_queries(
+    db,
+):
+    @strawberry_django.type(Issue)
+    class IssueType:
+        pk: strawberry.ID
+
+    @strawberry_django.type(Milestone)
+    class MilestoneType:
+        pk: strawberry.ID
+
+        @strawberry_django.field(
+            prefetch_related=[
+                lambda info: Prefetch(
+                    "issues",
+                    queryset=Issue.objects.filter(name__startswith="Foo"),
+                    to_attr="_my_issues",
+                ),
+            ],
+        )
+        def issues(self) -> List[IssueType]:
+            return self._my_issues  # type: ignore
+
+    @strawberry.type
+    class Query:
+        milestone: MilestoneType = strawberry_django.field()
+
+    schema = strawberry.Schema(query=Query, extensions=[DjangoOptimizerExtension])
+
+    milestone1 = MilestoneFactory.create()
+    milestone2 = MilestoneFactory.create()
+
+    issue1 = IssueFactory.create(name="Foo", milestone=milestone1)
+    IssueFactory.create(name="Bar", milestone=milestone1)
+    IssueFactory.create(name="Foo", milestone=milestone2)
+
+    query = """\
+      query TestQuery ($pk: ID!) {
+        milestone(pk: $pk) {
+          pk
+          issues {
+            pk
+          }
+        }
+      }
+    """
+
+    with assert_num_queries(2):
+        res = schema.execute_sync(query, {"pk": milestone1.pk})
+
+    assert res.errors is None
+    assert res.data == {
+        "milestone": {
+            "pk": str(milestone1.pk),
+            "issues": [{"pk": str(issue1.pk)}],
+        },
+    }

tests/utils.py

@@ -97,8 +97,9 @@ def assert_num_queries(n: int, *, using=DEFAULT_DB_ALIAS):
     # FIXME: Async will not have access to the correct number of queries without
     # execing CaptureQueriesContext.(__enter__|__exit__) wrapped in sync_to_async
     # How can we fix this?
-    if _client.get().is_async and executed == 0:
-        return
+    with contextlib.suppress(LookupError):
+        if _client.get().is_async and executed == 0:
+            return
 
     assert (
         executed == n