Home | About | Sematext search-lucene.com search-hadoop.com
NEW: Monitor These Apps!
elasticsearch, apache solr, apache hbase, hadoop, redis, casssandra, amazon cloudwatch, mysql, memcached, apache kafka, apache zookeeper, apache storm, ubuntu, centOS, red hat, debian, puppet labs, java, senseiDB
 Search Hadoop and all its subprojects:

Switch to Threaded View
Hive >> mail # dev >> Extending Explode


Copy link to this message
-
Re: Extending Explode
FYI,
  Brickhouse provides a numeric_range UDTF, which explodes integer values,
and an array_index UDF, so you could solve your problem by exploding on a
numeric range of the size of the array. ie.

select n, array_index(arr, n )
  from mytable
lateral view numeric_range(0, size(arr) -1 ) n1 as n
;

-- jerome
On Fri, Jul 26, 2013 at 6:40 AM, Edward Capriolo <[EMAIL PROTECTED]>wrote:

> So the explode output should be in the order of the array or the map you
> are exploding.
>
> I think you could use rank or row-sequence to give the exploded attay a
> number. If that does not wok adding a new udf might make more sence then
> extwnding in this case.:
>
> On Friday, July 26, 2013,  <[EMAIL PROTECTED]> wrote:
> > Hi,
> >
> > I'd like to make a patch that extends the functionality of "explode" to
> include an output column with the position of each item in the original
> array.
> >
> > I imagine this could be useful to the greater community and am wondering
> if I should extend the current explode function or if I should write a
> completely new function. Any thoughts on what will be more useful and more
> likely to be added to the hive-trunk would be greatly appreciated.
> >
> > Thanks,
> > Niko
> >
> >
>
NEW: Monitor These Apps!
elasticsearch, apache solr, apache hbase, hadoop, redis, casssandra, amazon cloudwatch, mysql, memcached, apache kafka, apache zookeeper, apache storm, ubuntu, centOS, red hat, debian, puppet labs, java, senseiDB