[
https://issues.apache.org/jira/browse/SQOOP-399?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Jarek Jarcec Cecho resolved SQOOP-399.
--------------------------------------
Resolution: Won't Fix
Assignee: Jarek Jarcec Cecho
Hi Miki,
thank you very much for reporting this issue. I agree that it's really strange
when default mapping differes for different file formats. However there might
be users out there that are already using sqoop and are fine with this strange
mapping and we definitely don't want to break it for them.
We recently started working on sqoop 2 where we will rewrite most sqoop parts
and where we will be able to do backward incompatible changes. We will
definitely make sure that default mapping will be the same across all file
formats in this version.
For the time being, you can overcame it by using explicit mapping that was
added into sqoop recently and is covered by SQOOP-342. This change was included
in sqoop version 1.4.0.
Jarcec
> Sqoop import TIMESTAMP and DATE differ between Avro and CSV imports
> -------------------------------------------------------------------
>
> Key: SQOOP-399
> URL: https://issues.apache.org/jira/browse/SQOOP-399
> Project: Sqoop
> Issue Type: Bug
> Reporter: Miki Tebeka
> Assignee: Jarek Jarcec Cecho
>
> When importing from PostgreSQL, TIMESTAMP and DATE columns are imported into
> long fields in Avro but into some kind of string representation in CSV import.
> Sqoop should be consistent and either use both epoch or string representation
> in both.
> (Also [here|https://issues.cloudera.org/browse/DISTRO-340]).
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators:
https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
