[jira] [Commented] (CARBONDATA-1278) Data Mismatch issue when dictionary column filter values doesn't exists in dictionary

["kumar vishal (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/CARBONDATA-1278?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16093579#comment-16093579 ] 

kumar vishal commented on CARBONDATA-1278:
------------------------------------------

*Root cause*
when filter value is not present in dictionary, end key for that column  is updated with 0,and hence btree jump is not selecting all the leaf node. 
*Fix*
To handle this issue end key should be updated with Integer max value, so that it can go till last leafnode of btree

> Data Mismatch issue when dictionary column filter values doesn't exists in dictionary 
> --------------------------------------------------------------------------------------
>
>                 Key: CARBONDATA-1278
>                 URL: https://issues.apache.org/jira/browse/CARBONDATA-1278
>             Project: CarbonData
>          Issue Type: Bug
>            Reporter: kumar vishal
>            Assignee: kumar vishal
>
> Data Mismatch issue when dictionary column filter values doesn't exists in dictionary when number of blocklet is more than 1 
> Select count(*) from table where column2='xxx' -- returning 100 rows.
> Select count(*) from table where column1='yyy' OR column2='xxx' -- returning 7 rows.
> Select count(*) from table where column1='yyy' is returning 0 rows 
> yyy does not exists in dictionary file.
> Query excepted value is 100 as yyy does not exists and its a or condition 



--
This message was sent by Atlassian JIRA
(v6.4.14#64029)