[jira] Commented: (HIVE-1558) introducing the "dual" table

["John Sichi (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/HIVE-1558?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12899991#action_12899991 ] 

John Sichi commented on HIVE-1558:
----------------------------------

SQL standard has the VALUES clause, so you can do

INSERT INTO t VALUES(3, 'hi');  -- inserts one row

INSERT INTO t VALUES (3, 'hi'), (4, 'bye');  -- inserts two rows

and

SELECT * FROM (VALUES(3, 'hi'), (4, 'bye'))  -- inline table results

If we add dual, it would also be nice to support at least the standard INSERT syntax since that has been around forever.


> introducing the "dual" table
> ----------------------------
>
>                 Key: HIVE-1558
>                 URL: https://issues.apache.org/jira/browse/HIVE-1558
>             Project: Hadoop Hive
>          Issue Type: New Feature
>            Reporter: Ning Zhang
>
> The "dual" table in MySQL and Oracle is very convenient in testing UDFs or constructing rows without reading any other tables. 
> If dual is the only data source we could leverage the local mode execution. 

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