hive-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Ashutosh Chauhan (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (HIVE-18945) Support "analyze table T"
Date Tue, 02 Oct 2018 23:24:00 GMT

    [ https://issues.apache.org/jira/browse/HIVE-18945?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16636227#comment-16636227
] 

Ashutosh Chauhan commented on HIVE-18945:
-----------------------------------------

I beg to differ. {{analyze table T}} syntax will cause more confusion then necessary. We have
made {{analyze T for cols}} to gather basic stats too so user just need to run one command.
Having 2 syntax which achieves same thing is not useful. Further, in longer run we may do
more than collect stats in analyze statement, e.g., some DBs has syntax like {{analyze table
T validate}} which performs bunch of validation on Table. On Hive, one such validation could
be making sure data files correspond to metadata w.r.t file format. If we ever introduce such
syntax in future, it will be puzzling for user that shortcut of {{analyze T}} does stats collection
or validation. So, IMHO we shall not add this syntax.

> Support "analyze table T"
> -------------------------
>
>                 Key: HIVE-18945
>                 URL: https://issues.apache.org/jira/browse/HIVE-18945
>             Project: Hive
>          Issue Type: Sub-task
>            Reporter: Zoltan Haindrich
>            Assignee: Laszlo Bodor
>            Priority: Major
>
> I think it would be good to have it behave the same as 
> {code}
> analyze table T compute statistics for columns
> {code}
> this could help people who not yet know the different analyze commands ; to run the probably
 most appropriate one



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)

Mime
View raw message