forked from D-Net/dnet-hadoop
75 lines
2.0 KiB
Java
75 lines
2.0 KiB
Java
|
|
package eu.dnetlib.dhp.actionmanager.project;
|
|
|
|
import static eu.dnetlib.dhp.common.SparkSessionSupport.runWithSparkSession;
|
|
|
|
import java.util.Optional;
|
|
|
|
import org.apache.commons.io.IOUtils;
|
|
import org.apache.spark.SparkConf;
|
|
import org.apache.spark.sql.SparkSession;
|
|
import org.slf4j.Logger;
|
|
import org.slf4j.LoggerFactory;
|
|
|
|
import com.fasterxml.jackson.databind.ObjectMapper;
|
|
|
|
import eu.dnetlib.dhp.application.ArgumentApplicationParser;
|
|
import eu.dnetlib.dhp.common.HdfsSupport;
|
|
|
|
public class SparkAtomicActionJob {
|
|
private static final Logger log = LoggerFactory.getLogger(SparkAtomicActionJob.class);
|
|
private static final ObjectMapper OBJECT_MAPPER = new ObjectMapper();
|
|
|
|
public static void main(String[] args) throws Exception {
|
|
|
|
String jsonConfiguration = IOUtils
|
|
.toString(
|
|
SparkAtomicActionJob.class
|
|
.getResourceAsStream(
|
|
"/eu/dnetlib/dhp/actionmanager/project/action_set_parameters.json"));
|
|
|
|
final ArgumentApplicationParser parser = new ArgumentApplicationParser(jsonConfiguration);
|
|
|
|
parser.parseArgument(args);
|
|
|
|
Boolean isSparkSessionManaged = Optional
|
|
.ofNullable(parser.get("isSparkSessionManaged"))
|
|
.map(Boolean::valueOf)
|
|
.orElse(Boolean.TRUE);
|
|
|
|
log.info("isSparkSessionManaged: {}", isSparkSessionManaged);
|
|
|
|
String projectPath = parser.get("projectPath");
|
|
log.info("projectPath: {}", projectPath);
|
|
|
|
final String outputPath = parser.get("outputPath");
|
|
log.info("outputPath {}: ", outputPath);
|
|
|
|
final String programmePath = parser.get("programmePath");
|
|
log.info("programmePath {}: ", programmePath);
|
|
|
|
SparkConf conf = new SparkConf();
|
|
|
|
runWithSparkSession(
|
|
conf,
|
|
isSparkSessionManaged,
|
|
spark -> {
|
|
removeOutputDir(spark, outputPath);
|
|
getAtomicActions(
|
|
spark,
|
|
projectPath,
|
|
programmePath,
|
|
outputPath);
|
|
});
|
|
}
|
|
|
|
private static void removeOutputDir(SparkSession spark, String path) {
|
|
HdfsSupport.remove(path, spark.sparkContext().hadoopConfiguration());
|
|
}
|
|
|
|
private static void getAtomicActions(SparkSession spark, String projectPatj, String programmePath,
|
|
String outputPath) {
|
|
|
|
}
|
|
}
|