Re: Creating custom Spark-Native catalyst/codegen functions

2019-08-22 Thread Georg Heiler
unctions in my user application without having to fork or modify Spark > itself. > > Thanks, > > Arwin > > From: Georg Heiler > Sent: Wednesday, August 21, 11:18 PM > Subject: Re: Creating custom Spark-Native catalyst/codegen functions > To: Arwin Tio > Cc: user@spark.ap

Re: Creating custom Spark-Native catalyst/codegen functions

2019-08-22 Thread Arwin Tio
Spark-Native catalyst/codegen functions To: Arwin Tio Cc: user@spark.apache.org Look at https://github.com/DataSystemsLab/GeoSpark/tree/master/sql/src/main/scala/org/apache/spark/sql/geospark sql for an example. Using custom function registration and functions residing inside sparks private

Re: Creating custom Spark-Native catalyst/codegen functions

2019-08-22 Thread Georg Heiler
Look at https://github.com/DataSystemsLab/GeoSpark/tree/master/sql/src/main/scala/org/apache/spark/sql/geospark sql for an example. Using custom function registration and functions residing inside sparks private namespace should work. But I am not aware of a public user facing API. Is there any

Creating custom Spark-Native catalyst/codegen functions

2019-08-21 Thread Arwin Tio
Hi friends, I am looking into converting some UDFs/UDAFs to Spark-Native functions to leverage Catalyst and codegen. Looking through some examples (for example: https://github.com/apache/spark/pull/7214/files for Levenshtein) it seems like we need to add these functions to the Spark framework