Code Convention: Eclipse formatter

2018-02-27 Thread Gokulakannan Muralidharan
Hi, Google code style format for Java is not working for Zeppelin project as mentioned in the docs , getting checkstyle errors when trying

Re: Cannot define UDAF in %spark interpreter

2018-02-27 Thread Vannson, Raphael
Hello Paul, Many thanks for your quick answer. This did the trick! Fantastic! Best, Raphael ***PARAGRAPH INPUT:*** val AggregatedChangepointAnalyzer = new UserDefinedAggregateFunction { … } ***PARAGRAPH OUTPUT:*** AggregatedChangepointAnalyzer: org.apache.spark.sql.expressions.UserDefinedAgg

Cannot define UDAF in %spark interpreter

2018-02-27 Thread Paul Brenner
Unfortunately, I don’t know why code that is working for you in spark shell isn’t working in Zeppelin. But if you are looking for a quick fix perhaps this could help? I’ve had luck defining my UDAFs in zeppelin like: val myUDAF = new UserDefinedAggregateFunction {} So for example the following

Cannot define UDAF in %spark interpreter

2018-02-27 Thread Vannson, Raphael
Hello, I am having trouble defining a UDAF, using the same code in spark-shell in :paste mode works fine. Environment: - Amazon EMR - Apache Zeppelin Version 0.7.3 - Spark version 2.2.1 - Using Scala version 2.11.8 (OpenJDK 64-Bit Server VM, Java 1.8.0_161) 1) Is there a way to configure th