generate action sets and saves them in the hdfs path for the actions sets

dhp-workflows/dhp-aggregation/src/main/java/eu/dnetlib/dhp/actionmanager/project/SparkAtomicActionJob.java

@@ -3,11 +3,16 @@ package eu.dnetlib.dhp.actionmanager.project;
 
 import static eu.dnetlib.dhp.common.SparkSessionSupport.runWithSparkSession;
 
+import java.io.IOException;
 import java.util.Arrays;
 import java.util.HashMap;
 import java.util.Optional;
+import java.util.function.Consumer;
 
+import eu.dnetlib.dhp.schema.action.AtomicAction;
 import org.apache.commons.io.IOUtils;
+import org.apache.hadoop.fs.Path;
+import org.apache.hadoop.io.IntWritable;
 import org.apache.spark.SparkConf;
 import org.apache.spark.api.java.function.MapFunction;
 import org.apache.spark.sql.Dataset;
@@ -28,6 +33,15 @@ import eu.dnetlib.dhp.schema.oaf.Programme;
 import eu.dnetlib.dhp.schema.oaf.Project;
 import eu.dnetlib.dhp.utils.DHPUtils;
 
+import org.apache.hadoop.conf.Configuration;
+
+import org.apache.hadoop.io.SequenceFile;
+import org.apache.spark.rdd.SequenceFileRDDFunctions;
+import org.apache.hadoop.io.Text;
+import scala.Function1;
+import scala.Tuple2;
+import scala.runtime.BoxedUnit;
+
 public class SparkAtomicActionJob {
 	private static final Logger log = LoggerFactory.getLogger(SparkAtomicActionJob.class);
 	private static final ObjectMapper OBJECT_MAPPER = new ObjectMapper();
@@ -61,6 +75,8 @@ public class SparkAtomicActionJob {
 		final String programmePath = parser.get("programmePath");
 		log.info("programmePath {}: ", programmePath);
 
+		final String nameNode = parser.get("hdfsNameNode");
+
 		SparkConf conf = new SparkConf();
 
 		runWithSparkSession(
@@ -72,7 +88,8 @@ public class SparkAtomicActionJob {
 					spark,
 					projectPath,
 					programmePath,
-					outputPath);
+					outputPath,
+						nameNode);
 			});
 	}
 
@@ -82,7 +99,8 @@ public class SparkAtomicActionJob {
 
 	private static void getAtomicActions(SparkSession spark, String projectPatH,
 		String programmePath,
-		String outputPath) {
+		String outputPath,
+										 String nameNode) throws Exception{
 
 		Dataset<CSVProject> project = readPath(spark, projectPatH, CSVProject.class);
 		Dataset<CSVProgramme> programme = readPath(spark, programmePath, CSVProgramme.class);
@@ -103,17 +121,16 @@ public class SparkAtomicActionJob {
 						pm.setCode(csvProject.getProgramme());
 						pm.setDescription(csvProgramme.get().getShortTitle());
 						p.setProgramme(Arrays.asList(pm));
-					return p;
+						return new AtomicAction<>(Project.class, p);
 					}
 
 					return null;
-			}, Encoders.bean(Project.class))
-			.filter(p -> !(p == null))
-			// .map(p -> new AtomicAction<>(Project.class, p), Encoders.bean(AtomicAction.class))
-			.write()
-			.option("compression", "gzip")
-			.mode(SaveMode.Overwrite)
-			.json(outputPath);
+				}, Encoders.bean(AtomicAction.class))
+				.filter(aa -> !(aa == null))
+				.toJavaRDD()
+				.mapToPair(aa->new Tuple2<>(aa.getClazz().getCanonicalName(), OBJECT_MAPPER.writeValueAsString(aa)))
+				.saveAsHadoopFile(outputPath, Text.class, Text.class, null);
+
 	}
 
 	public static <R> Dataset<R> readPath(

dhp-workflows/dhp-aggregation/src/main/resources/eu/dnetlib/dhp/actionmanager/project/oozie_app/workflow.xml

@@ -24,7 +24,6 @@
         <fs>
             <delete path='${outputPath}'/>
             <mkdir path='${outputPath}'/>
-            <delete path="/tmp/h2020programme"/>
         </fs>
         <ok to="get_project_file"/>
         <error to="Kill"/>