All, We generate Java objects for use in Spark RDDs from an .avdl schema using the Avro Maven Plugin
https://github.com/bigdatagenomics/bdg-formats <https://github.com/bigdatagenomics/bdg-formats> We also separately generate Scala products and projection field enums for use in Spark SQL Datasets, a bit more hacked together, declared in https://github.com/bigdatagenomics/adam/tree/master/adam-codegen <https://github.com/bigdatagenomics/adam/tree/master/adam-codegen> and used via exec-maven-plugin in https://github.com/bigdatagenomics/adam/blob/master/adam-core/pom.xml#L91 <https://github.com/bigdatagenomics/adam/blob/master/adam-core/pom.xml#L91> If of interest to the Avro project, could someone familiar with codegen and the Avro Maven Plugin help me refactor or otherwise adapt this for contribution? Thank you in advance, michael
