Home | About | Sematext search-lucene.com search-hadoop.com
 Search Hadoop and all its subprojects:

Switch to Threaded View
Sqoop, mail # user - Possible bug (export) --direct and --input-fields-terminated-by '\t' not working correctly?


Copy link to this message
-
Re: Possible bug (export) --direct and --input-fields-terminated-by '\t' not working correctly?
Jarek Jarcec Cecho 2012-12-15, 22:18
Hi sir,
Would you mind sharing with us entire sqoop command that you've used for export job that worked?

Jarcec

On Thu, Dec 13, 2012 at 04:07:37PM +0800, wzc wrote:
> I run into this too, I try add "--fields-terminated-by " and get the right
> result.
>
> refer:
> --fields-terminated-by <char>Sets the field separator character
> 2012/7/27 Connell, Chuck <[EMAIL PROTECTED]>
>
> > Thanks Kathleen,
> >
> > I read that report and I think mine is different. I do not get an error
> > message, but the export yields the wrong result. I will file a jira.
> >
> > Chuck
> >
> >
> > ________________________________________
> > From: Kathleen Ting [[EMAIL PROTECTED]]
> > Sent: Thursday, July 26, 2012 6:01 PM
> > To: [EMAIL PROTECTED]
> > Subject: Re: Possible bug (export) --direct and
> > --input-fields-terminated-by '\t' not working correctly?
> >
> > Chuck,
> >
> > I think you might be hitting
> > https://issues.apache.org/jira/browse/SQOOP-450, fixed in upcoming
> > release Sqoop 1.4.2
> >
> > Regards, Kathleen
> >
> > On Thu, Jul 26, 2012 at 6:46 AM, Connell, Chuck
> > <[EMAIL PROTECTED]> wrote:
> > > I have a Hive table that has tab-separated fields. I am exporting it to
> > MySQL. The export works fine (using --input-fields-terminated-by '\t') and
> > I get correct results in MySQL.
> > >
> > > BUT, when I add the --direct option, the export fails. The source file
> > (in Hive warehouse) is not parsed correctly and the tabs are ignored. The
> > fields are not separated correctly when placed into MySQL.
> > >
> > > Anyone else run into this?
> > >
> > > Chuck Connell
> > > Nuance R&D Data Team
> > > Burlington, MA
> > >
> >