UDF

udf.java

import static org.apache.spark.sql.functions.*;
import org.apache.spark.sql.expressions.UserDefinedFunction;


StructField [] sf1 = new StructField[] {
				 DataTypes.createStructField("uid",DataTypes.IntegerType, true),
				 DataTypes.createStructField("mid",DataTypes.IntegerType,true),
				 DataTypes.createStructField("rating",DataTypes.IntegerType, true),
				 DataTypes.createStructField("time",DataTypes.IntegerType, true),
		 };
		 
		 StructType st1 = DataTypes.createStructType(sf1);
		 
		
		Dataset<Row> mv =  spark
				.read()
				 .schema(st1)
				.format("com.databricks.spark.csv")
				.option("delimiter", "\t")
				
				.csv("/home/hasura/Desktop/SparkData/u.data");
		


	UserDefinedFunction increaserating = udf(
		 		(Integer s) -> s+1,DataTypes.IntegerType
				);
		
		mv.
		 	withColumn("rating",increaserating.apply(mv.col("rating")))
		.show();