I haven't used Zeppelin for a while, so excuse me if I'm speaking out of
turn anyone else, but back in the day I would hand my data from Flink to
Spark/Python/R for visualization.
The "Mahout Tutorial" is still in the trunk it seems. There are examples on
how to do this in there. (e.g. to your point there is no "z" context, but
how to call it deeply and hand data off using the ResourcePools)
On Fri, Feb 9, 2018 at 7:53 AM, André Schütz <an...@wegtam.com> wrote:
> the following examples to parametrize the access to databases from
> Spark is really awesome.
> I try to create such an example with a Flink interpreter but it seems
> that the possibility to use the "sqlContext" and "z context" is not
> possible with Apache Flink interpreter.
> Is there a possibility to create a similar construction like develop in
> the example above with the given Apache Flink Interpreter capabilities?
> It is my aim to create parametrized SELECT queries for users that do
> allow the selection of tables, columns and filters via Dynamic Form.
> Thank you for any help.
> Kind regards,
> Andre Schütz
> COO / Founder - Wegtam GmbH
> an...@wegtam.com | P: +49 (0) 381-80 699 041 | M: +49 (0) 176-218 02 604
> www.wegtam.com | www.tensei-data.com | www.wegtam.net