[jira] [Commented] (HBASE-5339) Add support for compound keys to importtsv

["Harsh J (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/HBASE-5339?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13736500#comment-13736500 ] 

Harsh J commented on HBASE-5339:
--------------------------------

While this is easy enough to implement, I think we're over-complicating the tool. The next demand could be MD5()-ing a part, etc..

We should hand this off to Pig/Hive by adding bulkload support to them. They already do CSV well enough. For Pig I'd once filed https://issues.apache.org/jira/browse/PIG-2921. We could file a similar one for Hive.
                
> Add support for compound keys to importtsv
> ------------------------------------------
>
>                 Key: HBASE-5339
>                 URL: https://issues.apache.org/jira/browse/HBASE-5339
>             Project: HBase
>          Issue Type: Improvement
>          Components: mapreduce
>            Reporter: Lars George
>            Priority: Trivial
>
> Add support that you can combine some columns from the TSV with either a given separator, no separator, or with a custom row key generator class. Syntax could be:
> {code}
> -Dimporttsv.columns=HBASE_ROW_KEY_1,HBASE_ROW_KEY_2,cf1:col1,cf2:col3,HBASE_ROW_KEY_3
> -Dimporttsv.rowkey.separator="-"
> {code}
> Another option of course is using the custom mapper class and handle this there, but this also seems like a nice to have option, probably often covering the 80% this sort of thing is needed.

--
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