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

Switch to Threaded View
Hive, mail # user - Define field delimiters for output files


Copy link to this message
-
Re: Define field delimiters for output files
Bejoy KS 2012-06-19, 08:02
Yes, external Tables are no different that managed tables. The main difference  in case of external tables is, if you drop a managed table the data in hdfs is also deleted where as in case of external tables the data is not deleted from hdfs only the entry from metastore is removed.
Regards
Bejoy KS

Sent from handheld, please excuse typos.

-----Original Message-----
From: tishan pubudu kanishka dahanayakage <[EMAIL PROTECTED]>
Date: Tue, 19 Jun 2012 13:26:49
To: <[EMAIL PROTECTED]>; <[EMAIL PROTECTED]>
Reply-To: [EMAIL PROTECTED]
Subject: Re: Define field delimiters for output files

Hi Bejoy,
Thanks. Will an external table also do the same thing?

On Tue, Jun 19, 2012 at 12:35 PM, Bejoy KS <[EMAIL PROTECTED]> wrote:

> **
> Hi Tishan
>
> AFAIK there is no way to specify the delimiter in INSERT OVERWRITE
> command. The delimiter could be specified on table creation only.
>
> You can get your requirement done my creating an intermediate table .
>
> 1) Define the intermediate table with the required delimiter of your
> choice.
> 2) Load data from original table to this intermediate table.
> 3) Do the Insert Overwrite to local directory from this table
>
>
> Regards
> Bejoy KS
>
> Sent from handheld, please excuse typos.
> ------------------------------
> *From: * tishan pubudu kanishka dahanayakage <[EMAIL PROTECTED]>
> *Date: *Tue, 19 Jun 2012 12:17:42 +0530
> *To: *<[EMAIL PROTECTED]>
> *ReplyTo: * [EMAIL PROTECTED]
> *Subject: *Define field delimiters for output files
>
> Hi all,
> Is there a way to define the field delimiter for output file when using
> the command
> INSERT OVERWRITE LOCAL DIRECTORY 'path'
>
>
> --
> Regards,
> Tishan
>
>
--
Regards,
Tishan