The way to establish an offset in Phoenix is using our row value constructor mechanism: https://phoenix.apache.org/paged.html. This can be part of a predicate and Phoenix will use it to form a startRow on the scan (which is essentially an offset).
On Tue, Aug 4, 2015 at 10:15 AM, Josh Mahonin <[email protected]> wrote: > Hi, > > The phoenix-spark module allows passing in a custom 'predicate', as long > as no aggregation functions are used. You can see examples here: > > https://github.com/apache/phoenix/blob/master/phoenix-spark/src/it/scala/org/apache/phoenix/spark/PhoenixSparkIT.scala#L215-L234 > > Re: offsetting, once you have the RDD (or DataFrame), you can skip rows > there, but there's no way to pass in an offset to the original query. > > Josh > > On Tue, Aug 4, 2015 at 11:13 AM, Hafiz Mujadid <[email protected]> > wrote: > >> >> Hi all! >> >> How can we pass custom query to spark phoenix load method? do we have >> some way to skip first n rows in result just like in sql we have offset >> keyword to skip first n records. >> >> Thanks >> > >
