[ https://issues.apache.org/jira/browse/HIVE-19919?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16515797#comment-16515797 ] 

Sergey Shelukhin commented on HIVE-19919:
-----------------------------------------

Hive already has this, it's called sysdb. See sysdb.q test file on master. Unfortunately I think currently it needs to be initialized explicitly cc [~hagleitn]

> HiveServer2 - expose queryable data dictionary (ie Oracles' ALL_TAB_COLUMNS)
> ----------------------------------------------------------------------------
>
>                 Key: HIVE-19919
>                 URL: https://issues.apache.org/jira/browse/HIVE-19919
>             Project: Hive
>          Issue Type: New Feature
>          Components: SQL
>    Affects Versions: 3.0.0, 2.3.2
>            Reporter: t oo
>            Priority: Major
>
> All major db vendors have a table like information_schema.columns, all_tab_columns or syscolumns containing table_name,column_name, data_type, col_order. Adding this feature to HiveServer2 would be very convenient for users.
> This information is currently only available in the mysql metastore ie TBLS, COLS but should be exposed up into the HiveServer2 10000 port connection. Thus saving users from having 2 connections (1 to see data, 1 to see metadata). For security reason too, mysql can be firewalled from end-users.



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