[
https://issues.apache.org/jira/browse/DRILL-6193?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16381032#comment-16381032
]
Julian Hyde commented on DRILL-6193:
------------------------------------
By design, you cannot tell where the relational expression came from.
One case where a cartesian join is "safe" is where one or both sides have at
zero or one rows. Then the join has no multiplying effect. Looks like that may
hold in this case, if you have a PK on Orders. There is a statistic
RelMdMaxRowCount.
If you are going to disable cartesian joins, why not do it at the SQL level
rather than the algebra level? There are legitimate patterns where cartesian
joins are the best plan.
> Latest Calcite optimized out join condition and cause "This query cannot be
> planned possibly due to either a cartesian join or an inequality join"
> --------------------------------------------------------------------------------------------------------------------------------------------------
>
> Key: DRILL-6193
> URL: https://issues.apache.org/jira/browse/DRILL-6193
> Project: Apache Drill
> Issue Type: Bug
> Components: Query Planning & Optimization
> Affects Versions: 1.13.0
> Reporter: Chunhui Shi
> Assignee: Hanumath Rao Maduri
> Priority: Blocker
> Fix For: 1.13.0
>
>
> I got the same error on apache master's MapR profile on the tip(before Hive
> upgrade) and on changeset 9e944c97ee6f6c0d1705f09d531af35deed2e310, the last
> commit of Calcite upgrade with the failed query reported in functional test
> but now it is on parquet file:
>
> {quote}SELECT L.L_QUANTITY, L.L_DISCOUNT, L.L_EXTENDEDPRICE, L.L_TAX
>
> FROM cp.`tpch/lineitem.parquet` L, cp.`tpch/orders.parquet` O
> WHERE cast(L.L_ORDERKEY as int) = cast(O.O_ORDERKEY as int) AND
> cast(L.L_LINENUMBER as int) = 7 AND cast(L.L_ORDERKEY as int) = 10208 AND
> cast(O.O_ORDERKEY as int) = 10208;
> {quote}
> However, built Drill on commit ef0fafea214e866556fa39c902685d48a56001e1, the
> commit right before Calcite upgrade commits, the same query worked.
> This was caused by latest Calcite simplified the predicates and during this
> process, "cast(L.L_ORDERKEY as int) = cast(O.O_ORDERKEY as int) " was
> considered redundant and was removed, so the logical plan of this query is
> getting an always true condition for Join:
> {quote}DrillJoinRel(condition=[true], joinType=[inner])
> {quote}
> While in previous version we have
> {quote}DrillJoinRel(condition=[=($5, $0)], joinType=[inner])
> {quote}
>
>
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
