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

Switch to Threaded View
Hive, mail # user - Regex and serde with hive


Copy link to this message
-
Re: Regex and serde with hive
Vijay 2011-12-23, 07:30
If the format is simply delimited like this, you don't need to use the
RegexSerde. Hive's default format with the right "FIELDS TERMINATED
BY" setting will work great.

-Vijay

On Thu, Dec 22, 2011 at 8:49 PM, Raghunath, Ranjith
<[EMAIL PROTECTED]> wrote:
> I have been struggling with this for a while so I would appreciate any
> advice that you any of you may have.
>
>
>
> I have a file of the format
>
>
>
> “Xyz”~”qsd”~”1234”
>
>
>
> I created the following table definition to get the data loaded
>
>
>
> CREATE TABLE dummy
>
> (f1   string,
>
>   f2    string,
>
>   f3     string)
>
> ROW FORMAT SERDE 'org.apache.hadoop.hive.contrib.serde2.RegexSerDe'
>
> WITH SERDEPROPERTIES  (
>
> "input.regex" = "([^\"~]*) ([^\"~]*) ([^\"~]*)?",
>
> "output.format.string" = "%1$s %2$s %3$s");
>
>
>
> When I load the data in and try to perform a select get NULL values. Thanks
> again.
>
> Thank you,
>
> Ranjith
>
>