I have tried to read/write Parquet files from/to Azurite using Spark like this:
import com.holdenkarau.spark.testing.DatasetSuiteBase import org.apache.spark.