Kontinuation commented on code in PR #611:
URL: https://github.com/apache/sedona-db/pull/611#discussion_r2815281009
##########
rust/sedona-spatial-join/tests/spatial_join_integration.rs:
##########
@@ -1368,3 +1388,93 @@ async fn test_knn_join_include_tie_breakers(
Ok(())
}
+
+/// Verify that a filter on the *object* (build / right) side of a KNN join is
NOT pushed down
+/// into the build side subtree.
+///
+/// If `PushDownFilter` incorrectly pushes `R.id > 5` below the spatial join,
the set of objects
+/// considered for KNN changes, yielding wrong nearest-neighbor results.
+#[tokio::test]
+async fn test_knn_join_object_side_filter_not_pushed_down() -> Result<()> {
+ let sql = "SELECT L.id, R.id \
+ FROM L JOIN R ON ST_KNN(ST_Point(L.x, 0), ST_Point(R.x, 1), 3,
false) \
+ WHERE R.id > 5";
+ let plan = plan_for_filter_pushdown_test(sql).await?;
+
+ let spatial_joins = collect_spatial_join_exec(&plan)?;
+ assert_eq!(
+ spatial_joins.len(),
+ 1,
+ "expected exactly one SpatialJoinExec"
+ );
+ let sj = spatial_joins[0];
+
+ // The build (right / object) side must NOT have a FilterExec pushed into
it.
+ assert!(
+ !subtree_contains_filter_exec(&sj.right),
+ "FilterExec should NOT be pushed into the object (right/build) side of
a KNN join"
+ );
+
+ Ok(())
+}
+
+/// Verify that for a *non-KNN* spatial join, a filter on the build side IS
pushed down
+/// (the normal, desirable behaviour).
+#[tokio::test]
+async fn test_non_knn_join_object_side_filter_is_pushed_down() -> Result<()> {
+ let sql = "SELECT L.id, R.id \
+ FROM L JOIN R ON ST_Intersects(ST_Buffer(ST_Point(L.x, 0),
1.5), ST_Point(R.x, 1)) \
+ WHERE R.id > 5";
+ let plan = plan_for_filter_pushdown_test(sql).await?;
+
+ let spatial_joins = collect_spatial_join_exec(&plan)?;
+ assert_eq!(
+ spatial_joins.len(),
+ 1,
+ "expected exactly one SpatialJoinExec"
+ );
+ let sj = spatial_joins[0];
+
+ // For non-KNN joins, the filter SHOULD be pushed down to the build side.
+ assert!(
+ subtree_contains_filter_exec(&sj.right),
+ "FilterExec should be pushed into the object (right/build) side of a
non-KNN spatial join"
+ );
+
+ Ok(())
+}
+
+/// Recursively check whether any node in the physical plan tree is a
`FilterExec`.
+fn subtree_contains_filter_exec(plan: &Arc<dyn ExecutionPlan>) -> bool {
+ let mut found = false;
+ plan.apply(|node| {
+ if node.as_any().downcast_ref::<FilterExec>().is_some() {
+ found = true;
+ return Ok(TreeNodeRecursion::Stop);
+ }
+ Ok(TreeNodeRecursion::Continue)
+ })
+ .expect("failed to walk plan");
+ found
+}
+
+/// Create a session context with two small tables for filter-pushdown tests.
+///
+/// L(id INT, x DOUBLE) and R(id INT, x DOUBLE) each with 10 rows.
+/// Geometry is constructed in SQL via ST_Point so no geometry column exists
on the table itself.
+async fn plan_for_filter_pushdown_test(sql: &str) -> Result<Arc<dyn
ExecutionPlan>> {
+ let schema = Arc::new(Schema::new(vec![
+ Field::new("id", DataType::Int32, false),
+ Field::new("x", DataType::Float64, false),
+ ]));
+
+ let options = SpatialJoinOptions::default();
+ let ctx = setup_context(Some(options), 100)?;
+ let empty_l: Arc<dyn TableProvider> =
Arc::new(EmptyTable::new(schema.clone()));
+ let empty_r: Arc<dyn TableProvider> =
Arc::new(EmptyTable::new(schema.clone()));
+ ctx.register_table("L", empty_l)?;
+ ctx.register_table("R", empty_r)?;
Review Comment:
Fixed the comment
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
