hadoop-hive-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Namit Jain (JIRA)" <j...@apache.org>
Subject [jira] Commented: (HIVE-79) Print number of raws inserted to table(s) when the query is finished.
Date Fri, 30 Jan 2009 18:22:59 GMT

    [ https://issues.apache.org/jira/browse/HIVE-79?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12668965#action_12668965
] 

Namit Jain commented on HIVE-79:
--------------------------------

The id's in genFilesinkPlan() are always incremented. fileSinkDesc is created twice for evey
table (in the common case of column pruning).

So, although, the FileSinkOperator has counters till 15, most of the tables will be doube-counted
in the common case and logging wont happen when there are more than 8 tables.
Moreoever, this can get worse with more optimizer features.

clear the (id->name) map in reset()  and set the counter back to 0

> Print number of raws inserted to table(s) when  the query is finished.
> ----------------------------------------------------------------------
>
>                 Key: HIVE-79
>                 URL: https://issues.apache.org/jira/browse/HIVE-79
>             Project: Hadoop Hive
>          Issue Type: New Feature
>          Components: Logging
>            Reporter: Suresh Antony
>            Assignee: Suresh Antony
>            Priority: Minor
>             Fix For: 0.2.0
>
>         Attachments: patch_79_1.txt
>
>
> It is good to print the number of rows inserted into each table at end of query. 
> insert overwrite table tab1 select a.* from tab2 a where a.col1 = 10;
> This query can print something like:
> tab1 rows=100

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.


Mime
View raw message