[
https://issues.apache.org/jira/browse/HIVE-19770?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16551271#comment-16551271
]
Vineet Garg commented on HIVE-19770:
------------------------------------
Earlier patch removed the same column restriction from CBO path and added
ambiguity check (same as non-cbo path). But it turns out that ambiguity check
is too restrictive e.g. queries like {code:sql} select count(*) from (select
key, key from src) subq {code} is not allowed with this check.
There are also queries which were successfully planned by CBO but now with this
check are not allowed e.g. {code:sql} select count(*) from (select t1.key,
t2.key from src t1 inner join src t2 on t1.key=t2.key)subq {code}
Therefore I have removed the ambiguity check for now and opened HIVE-20215 for
fix this. Fixing this will require checking ambiguity during column reference
instead of generating column list for select.
> Support for CBO for queries with multiple same columns in select
> ----------------------------------------------------------------
>
> Key: HIVE-19770
> URL: https://issues.apache.org/jira/browse/HIVE-19770
> Project: Hive
> Issue Type: Improvement
> Reporter: Vineet Garg
> Assignee: Vineet Garg
> Priority: Major
> Attachments: HIVE-19770.1.patch, HIVE-19770.2.patch,
> HIVE-19770.3.patch, HIVE-19770.4.patch, HIVE-19770.5.patch, HIVE-19770.6.patch
>
>
> Currently queries such as {code:sql} select a,a from t1 where b > 10 {code}
> are not supported for CBO.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
