114 lines
3.8 KiB
Java
114 lines
3.8 KiB
Java
|
|
package eu.dnetlib.dhp.oa.graph.dump.eosc;
|
|
|
|
import static eu.dnetlib.dhp.common.SparkSessionSupport.runWithSparkSession;
|
|
|
|
import java.io.Serializable;
|
|
import java.util.Optional;
|
|
|
|
import org.apache.commons.io.IOUtils;
|
|
import org.apache.hadoop.io.Text;
|
|
import org.apache.spark.SparkConf;
|
|
import org.apache.spark.api.java.JavaRDD;
|
|
import org.apache.spark.api.java.JavaSparkContext;
|
|
import org.apache.spark.api.java.function.MapFunction;
|
|
import org.apache.spark.sql.Dataset;
|
|
import org.apache.spark.sql.Encoders;
|
|
import org.apache.spark.sql.SaveMode;
|
|
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.eosc.model.Indicator;
|
|
import eu.dnetlib.dhp.eosc.model.Result;
|
|
import eu.dnetlib.dhp.eosc.model.UsageCounts;
|
|
import eu.dnetlib.dhp.schema.action.AtomicAction;
|
|
import scala.Tuple2;
|
|
|
|
/**
|
|
* @author miriam.baglioni
|
|
* @Date 07/11/22
|
|
*/
|
|
public class ExtendWithUsageCounts implements Serializable {
|
|
|
|
private static final Logger log = LoggerFactory.getLogger(ExtendEoscResultWithOrganizationStep2.class);
|
|
|
|
public static void main(String[] args) throws Exception {
|
|
String jsonConfiguration = IOUtils
|
|
.toString(
|
|
ExtendEoscResultWithOrganizationStep2.class
|
|
.getResourceAsStream(
|
|
"/eu/dnetlib/dhp/oa/graph/dump/eosc_extend_result_with_indicators_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);
|
|
|
|
final String actionSetPath = parser.get("actionSetPath");
|
|
log.info("actionSetPath: {}", actionSetPath);
|
|
|
|
final String resultPath = parser.get("resultPath");
|
|
log.info("resultPath: {}", resultPath);
|
|
|
|
final String outputPath = parser.get("outputPath");
|
|
log.info("outputPath: {}", outputPath);
|
|
|
|
SparkConf conf = new SparkConf();
|
|
|
|
runWithSparkSession(
|
|
conf,
|
|
isSparkSessionManaged,
|
|
spark -> {
|
|
Utils.removeOutputDir(spark, outputPath);
|
|
addIndicators(spark, actionSetPath, outputPath, resultPath);
|
|
});
|
|
}
|
|
|
|
private static void addIndicators(SparkSession spark, String actionSetPath, String outputPath, String resultPath) {
|
|
final JavaSparkContext sc = new JavaSparkContext(spark.sparkContext());
|
|
JavaRDD<eu.dnetlib.dhp.schema.oaf.Result> resultsWithIndicatorsRdd = sc
|
|
.sequenceFile(actionSetPath, Text.class, Text.class)
|
|
.map(value -> new ObjectMapper().readValue(value._2().toString(), AtomicAction.class))
|
|
.map(aa -> ((eu.dnetlib.dhp.schema.oaf.Result) aa.getPayload()));
|
|
|
|
Dataset<eu.dnetlib.dhp.schema.oaf.Result> resultWithIndicators = spark
|
|
.createDataset(resultsWithIndicatorsRdd.rdd(), Encoders.bean(eu.dnetlib.dhp.schema.oaf.Result.class));
|
|
|
|
Dataset<Result> result = Utils
|
|
.readPath(spark, resultPath, Result.class);
|
|
|
|
result
|
|
.joinWith(resultWithIndicators, result.col("id").equalTo(resultWithIndicators.col("id")), "left")
|
|
.map((MapFunction<Tuple2<Result, eu.dnetlib.dhp.schema.oaf.Result>, Result>) t2 -> {
|
|
if (Optional.ofNullable(t2._2()).isPresent()) {
|
|
Indicator indicator = new Indicator();
|
|
UsageCounts uc = new UsageCounts();
|
|
t2._2().getMeasures().stream().forEach(m -> {
|
|
if (m.getId().equals("downloads")) {
|
|
uc.setDownloads(m.getUnit().get(0).getValue());
|
|
} else {
|
|
uc.setViews(m.getUnit().get(0).getValue());
|
|
}
|
|
indicator.setUsageCounts(uc);
|
|
t2._1().setIndicator(indicator);
|
|
});
|
|
|
|
}
|
|
return t2._1();
|
|
}, Encoders.bean(Result.class))
|
|
.write()
|
|
.mode(SaveMode.Overwrite)
|
|
.option("compression", "gzip")
|
|
.json(outputPath);
|
|
|
|
}
|
|
}
|