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

Switch to Plain View
Hive >> mail # user >> partitioned by usage


+
Kishore kumar 2014-01-07, 09:13
+
Nitin Pawar 2014-01-07, 10:21
+
Kishore kumar 2014-01-07, 10:24
Copy link to this message
-
Re: partitioned by usage
its something like this

create table xyz (a int, b string) partitioned by (c string);
LOAD DATA LOCAL INPATH 'abc' INTO TABLE xyx PARTITION(c="abc");

remember if your data has multiple values on partition column and you do
not want to write mapreduce code or pig scripts then you will need a
temporary table and then enable dynamic partitioning while loading into
real table
On Tue, Jan 7, 2014 at 3:54 PM, Kishore kumar <[EMAIL PROTECTED]> wrote:

> How to create partitioned table without creating intermediate table?
> simply..
>
>
> On Tue, Jan 7, 2014 at 10:21 AM, Nitin Pawar <[EMAIL PROTECTED]>wrote:
>
>> can you put your question in with an example?
>>
>>
>> On Tue, Jan 7, 2014 at 2:43 PM, Kishore kumar <[EMAIL PROTECTED]>wrote:
>>
>>> Hi Experts,
>>>
>>> As per this link
>>>
>>>
>>> http://stackoverflow.com/questions/10276584/hive-table-partition-with-column-in-the-middle
>>>
>>> I understood that we can create the partitioned table when we have
>>> already a non partitioned table, if it is correct, when we will use
>>> partitioned by clause to create a new partitioned table.
>>>
>>> --
>>>
>>> *Kishore Kumar*
>>>
>>>
>>
>>
>> --
>> Nitin Pawar
>>
>
>
>
> --
>
> *Kishore Kumar*
> ITIM
>
> Bidstalk - Ingenius Programmatic Platform
>
> Email: [EMAIL PROTECTED]| Tel: +1 415 423 8230  | Cell: +91 741 135
> 8658 | skype: kishore.alajangi | YM: kk_asn2004 | Twitter:
> __kishorealajangi
> [image: Inline image 1]
>

--
Nitin Pawar