[ 
https://issues.apache.org/jira/browse/HIVE-4568?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
 ]

Xuefu Zhang updated HIVE-4568:
------------------------------

    Description: 
Previous Hive CLI allows user to specify hive variables at the command line 
using option "--hivevar". In user's script, reference to a hive variable will 
be substituted with the value of the variable. In such way, user can 
parameterize his/her script and invoke the script with different hive variable 
values. The following script is one usage:
{code}
hive --hivevar
     INPUT=/user/jenkins/oozie.1371538916178/examples/input-data/table
     --hivevar
     OUTPUT=/user/jenkins/oozie.1371538916178/examples/output-data/hive
     -f script.q
{code}

script.q makes use of hive variables:
{code}
CREATE EXTERNAL TABLE test (a INT) STORED AS TEXTFILE LOCATION '${INPUT}';
INSERT OVERWRITE DIRECTORY '${OUTPUT}' SELECT * FROM test;
{code}

However, after upgrade to hiveserver2 and beeline, this functionality is 
missing. Beeline doesn't take --hivevar option, and any hive variable isn't 
passed to server so it cannot be used for substitution.

This JIRA is to address this issue, providing a backward compatible behavior at 
Beeline.

  was:Beeline currently doesn't support variable (system, env, etc) 
substitution as hive client does. Supporting this feature will certainly make 
it more usable.

    
> Beeline needs to support resolving variables
> --------------------------------------------
>
>                 Key: HIVE-4568
>                 URL: https://issues.apache.org/jira/browse/HIVE-4568
>             Project: Hive
>          Issue Type: Improvement
>    Affects Versions: 0.10.0
>            Reporter: Xuefu Zhang
>            Assignee: Xuefu Zhang
>            Priority: Minor
>         Attachments: HIVE-4568.patch
>
>
> Previous Hive CLI allows user to specify hive variables at the command line 
> using option "--hivevar". In user's script, reference to a hive variable will 
> be substituted with the value of the variable. In such way, user can 
> parameterize his/her script and invoke the script with different hive 
> variable values. The following script is one usage:
> {code}
> hive --hivevar
>      INPUT=/user/jenkins/oozie.1371538916178/examples/input-data/table
>      --hivevar
>      OUTPUT=/user/jenkins/oozie.1371538916178/examples/output-data/hive
>      -f script.q
> {code}
> script.q makes use of hive variables:
> {code}
> CREATE EXTERNAL TABLE test (a INT) STORED AS TEXTFILE LOCATION '${INPUT}';
> INSERT OVERWRITE DIRECTORY '${OUTPUT}' SELECT * FROM test;
> {code}
> However, after upgrade to hiveserver2 and beeline, this functionality is 
> missing. Beeline doesn't take --hivevar option, and any hive variable isn't 
> passed to server so it cannot be used for substitution.
> This JIRA is to address this issue, providing a backward compatible behavior 
> at Beeline.

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira

Reply via email to