You are viewing a plain text version of this content. The canonical link for it is here.

Posted to dev@hive.apache.org by "Zejun Li (JIRA)" <ji...@apache.org> on 2018/06/03 19:24:00 UTC

[jira] [Created] (HIVE-19779) HCatInputFormat cannot pushdown projection when read an HBase table

Zejun Li created HIVE-19779:
-------------------------------

             Summary: HCatInputFormat cannot pushdown projection when read an HBase table
                 Key: HIVE-19779
                 URL: https://issues.apache.org/jira/browse/HIVE-19779
             Project: Hive
          Issue Type: Bug
         Environment: hive-hcatalog-core 1.1.0-cdh5.14.2
hive-hbase-handler 1.1.0-cdh5.14.2
            Reporter: Zejun Li


I have used `HCatBaseInputFormat.setOutputSchema` to set a columns projection, and the result `HCatRecord` only contains the specified cols indeed. But when I check the `Result` returned by scan, I noticed that all cf and cols is selected.

Can HCatInputFormat do something like projection pushdown? Or is there any documents to describe it?

BTW, will HCatInputFormat do this pushdown on columnar file formats like Parquet?



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