Luka Zdravic created SPARK-60000:
------------------------------------
Summary: MATCH_CONDITION rejects a column hidden by a nested USING
join. ON accepts it.
Key: SPARK-60000
URL: https://issues.apache.org/jira/browse/SPARK-60000
Project: Spark
Issue Type: Bug
Components: SQL
Affects Versions: 4.4.0
Reporter: Luka Zdravic
{code:sql}
-- fails: [ASOF_JOIN_MATCH_CONDITION_TABLE_REFERENCE] ... Got: "lts" and "k".
SELECT l.lts, r.rts
FROM (VALUES (1, 10), (2, 20)) AS l(k, lts)
ASOF JOIN ((VALUES (1, 5), (2, 15)) AS r(k, rts) JOIN (VALUES (1), (2)) AS
d(k) USING (k))
MATCH_CONDITION (l.lts >= d.k);
-- works:
SELECT l.lts, r.rts
FROM (VALUES (1, 10), (2, 20)) AS l(k, lts)
JOIN ((VALUES (1, 5), (2, 15)) AS r(k, rts) JOIN (VALUES (1), (2)) AS d(k)
USING (k))
ON l.lts >= d.k;
{code}
Cause: the side check looks only at visible columns (left.outputSet /
right.outputSet in ResolveAsOfJoin), so d.k has no side.
Expected: the ASOF query returns (10, 15) and (20, 15).
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
