You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@hive.apache.org by "Sergey Shelukhin (JIRA)" <ji...@apache.org> on 2014/11/18 00:17:33 UTC

[jira] [Commented] (HIVE-8848) data loading from text files or text file processing doesn't handle nulls correctly

    [ https://issues.apache.org/jira/browse/HIVE-8848?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14215366#comment-14215366 ] 

Sergey Shelukhin commented on HIVE-8848:
----------------------------------------

[~jpullokkaran] [~ashutoshc] this happens in tests when loading cbo t1,t2,t3 tables. Do you guys know about how nulls are supposed to work, and if they work as designed (which would be pretty ugly) can you please +1 the patch.

> data loading from text files or text file processing doesn't handle nulls correctly
> -----------------------------------------------------------------------------------
>
>                 Key: HIVE-8848
>                 URL: https://issues.apache.org/jira/browse/HIVE-8848
>             Project: Hive
>          Issue Type: Bug
>            Reporter: Sergey Shelukhin
>         Attachments: HIVE-8848.patch
>
>
> I am not sure how nulls are supposed to be stored in text tables, but after loading some data with "null" or "NULL" strings, or x00 characters, we get bunch of annoying logging from LazyPrimitive that data is not in INT format and was converted to null, with data being "null" (string saying "null", I assume from the code).
> Either load should load them as nulls, or there should be some defined way to load nulls.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)