Yuming Wang created SPARK-37226:
-----------------------------------
Summary: Filter push down through window
Key: SPARK-37226
URL: https://issues.apache.org/jira/browse/SPARK-37226
Project: Spark
Issue Type: Improvement
Components: SQL
Affects Versions: 3.3.0
Reporter: Yuming Wang
Assignee: Yuming Wang
{code:scala}
spark.sql("CREATE TABLE t1 using parquet as select id as a, id as b from
range(1000)")
spark.sql("select * from (SELECT a, count(*) cnt, row_number() over (order by a
desc) as rn from t1 group by a) where rn <= 10").explain(true)
{code}
We can optimize this query:
{noformat}
== Optimized Logical Plan ==
Filter (rn#4 <= 10)
+- Window [row_number() windowspecdefinition(a#7L DESC NULLS LAST,
specifiedwindowframe(RowFrame, unboundedpreceding$(), currentrow$())) AS rn#4],
[a#7L DESC NULLS LAST]
+- GlobalLimit 10
+- LocalLimit 10
+- Sort [a#7L DESC NULLS LAST], true
+- Aggregate [a#7L], [a#7L, count(1) AS cnt#3L]
+- Project [a#7L]
+- Relation default.t1[a#7L,b#8L] parquet
{noformat}
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org
For additional commands, e-mail: issues-h...@spark.apache.org
