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

Switch to Threaded View
Pig, mail # user - Many to One UDF Problem


Copy link to this message
-
Re: Many to One UDF Problem
Prashant Kommireddi 2012-05-10, 05:29
I messed up, your original UDF does not need to be changed.

Just pass in all fields (*) as I suggested in my previous email, and access
them the way you were doing it before:
String query = (String)input.get(0);
String query1 = (String)input.get(1);

That should work.

-Prashant
On Wed, May 9, 2012 at 7:00 PM, DIPESH KUMAR SINGH <[EMAIL PROTECTED]>wrote:

> MapReduce job runs now, but string output of UDF is not coming. It shows
> something
> like this:
>
> (Jeff,13,)
> (John,12,)
>
> May be something needs to be changed in output schema, i was passing
> earlier:
>
> @Override
> public Schema outputSchema(Schema input) {
> return new Schema(new
> Schema.FieldSchema(getSchemaName(this.getClass().getName().toLowerCase(),
> input), DataType.CHARARRAY));
> }
>
> Thanks,
> Dipesh
>
> On Thu, May 10, 2012 at 7:09 AM, Prashant Kommireddi <[EMAIL PROTECTED]
> >wrote:
>
> > public List<FuncSpec> getArgToFuncMapping() throws FrontendException
> > needs to be modified accordingly, since you are now passing your UDF
> > the entire tuple. You don't really need to implement it if there is no
> > overloaded function.
> >
> > Sent from my iPhone
> >
> > On May 9, 2012, at 6:19 PM, DIPESH KUMAR SINGH <[EMAIL PROTECTED]>
> > wrote:
> >
> > > public List<FuncSpec> getArgToFuncMapping() throws FrontendException
> >
>
>
>
> --
> Dipesh Kr. Singh
>