[jira] [Commented] (HIVE-14040) insert overwrite for HBase doesn't overwrite

["Pengcheng Xiong (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/HIVE-14040?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15342537#comment-15342537 ] 

Pengcheng Xiong commented on HIVE-14040:
----------------------------------------

Here is some other finding. It seems that we allow create partition table with HBaseHandler but insert overwrite does not work.
{code}
set hive.exec.dynamic.partition.mode=nonstrict;
set hive.exec.max.dynamic.partitions.pernode=100000;

DROP TABLE users;
CREATE TABLE users(k string, state string, country string) partitioned by(key string)
STORED BY 'org.apache.hadoop.hive.hbase.HBaseStorageHandler'
WITH SERDEPROPERTIES (
"hbase.columns.mapping" = "info:state,info:country"
);

INSERT OVERWRITE TABLE users partition(key) SELECT 'user1', 'IA', 'USA', key FROM src;

desc formatted users;

select * from users;
{code}

the output
{code}
POSTHOOK: query: desc formatted users
POSTHOOK: type: DESCTABLE
POSTHOOK: Input: default@users
# col_name              data_type               comment

k                       string
state                   string
country                 string

# Partition Information
# col_name              data_type               comment

key                     string

# Detailed Table Information
Database:               default
#### A masked pattern was here ####
Retention:              0
#### A masked pattern was here ####
Table Type:             MANAGED_TABLE
Table Parameters:
        storage_handler         org.apache.hadoop.hive.hbase.HBaseStorageHandler
#### A masked pattern was here ####

# Storage Information
SerDe Library:          org.apache.hadoop.hive.hbase.HBaseSerDe
InputFormat:            null
OutputFormat:           null
Compressed:             No
Num Buckets:            -1
Bucket Columns:         []
Sort Columns:           []
Storage Desc Params:
        hbase.columns.mapping   info:state,info:country
        serialization.format    1
PREHOOK: query: select * from users
PREHOOK: type: QUERY
PREHOOK: Input: default@users
#### A masked pattern was here ####
POSTHOOK: query: select * from users
POSTHOOK: type: QUERY
POSTHOOK: Input: default@users
#### A masked pattern was here ####
{code}

> insert overwrite for HBase doesn't overwrite
> --------------------------------------------
>
>                 Key: HIVE-14040
>                 URL: https://issues.apache.org/jira/browse/HIVE-14040
>             Project: Hive
>          Issue Type: Sub-task
>            Reporter: Sergey Shelukhin
>            Assignee: Sergey Shelukhin
>
> Creating a table and doing insert overwrite twice with two different rows (for example) results in the table with both rows, rather than only one as per "overwrite"



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)