forked from D-Net/dnet-hadoop
137 lines
3.9 KiB
Java
137 lines
3.9 KiB
Java
|
|
package eu.dnetlib.dhp.patchrefereed;
|
|
|
|
import static eu.dnetlib.dhp.common.SparkSessionSupport.runWithSparkSession;
|
|
|
|
import java.io.Serializable;
|
|
import java.util.List;
|
|
import java.util.Objects;
|
|
import java.util.Optional;
|
|
import java.util.stream.Collectors;
|
|
|
|
import org.apache.commons.io.IOUtils;
|
|
import org.apache.spark.SparkConf;
|
|
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.common.HdfsSupport;
|
|
import eu.dnetlib.dhp.schema.oaf.*;
|
|
|
|
public class SparkPrepareResultInstanceList implements Serializable {
|
|
|
|
private static final Logger log = LoggerFactory.getLogger(SparkPrepareResultInstanceList.class);
|
|
private static final ObjectMapper OBJECT_MAPPER = new ObjectMapper();
|
|
|
|
public static void main(String[] args) throws Exception {
|
|
|
|
String jsonConfiguration = IOUtils
|
|
.toString(
|
|
SparkPrepareResultInstanceList.class
|
|
.getResourceAsStream(
|
|
"/eu/dnetlib/dhp/patchrefereed/prepare_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 outputPath = parser.get("outputPath");
|
|
log.info("outputPath {}: ", outputPath);
|
|
|
|
final String inputPath = parser.get("inputPath");
|
|
log.info("inputPath {}: ", inputPath);
|
|
|
|
final String resultClassName = parser.get("resultTableName");
|
|
log.info("resultTableName: {}", resultClassName);
|
|
|
|
Class<? extends Result> resultClazz = (Class<? extends Result>) Class.forName(resultClassName);
|
|
|
|
SparkConf conf = new SparkConf();
|
|
|
|
runWithSparkSession(
|
|
conf,
|
|
isSparkSessionManaged,
|
|
spark -> {
|
|
removeOutputDir(spark, outputPath);
|
|
getAtomicActions(
|
|
spark,
|
|
inputPath,
|
|
outputPath,
|
|
resultClazz);
|
|
});
|
|
}
|
|
|
|
private static void removeOutputDir(SparkSession spark, String path) {
|
|
HdfsSupport.remove(path, spark.sparkContext().hadoopConfiguration());
|
|
}
|
|
|
|
private static <R extends Result> void getAtomicActions(SparkSession spark,
|
|
String inputPath,
|
|
String outputPath,
|
|
Class<R> resultClazz) {
|
|
|
|
Dataset<R> result = readPath(spark, inputPath, resultClazz);
|
|
|
|
// log.info("number of results: {}", result.count());
|
|
result.map(r -> {
|
|
ResultInstance ri = null;
|
|
List<Instance> instanceList = Optional
|
|
.ofNullable(r.getInstance())
|
|
.map(
|
|
instances -> instances
|
|
.stream()
|
|
.map(instance -> {
|
|
Optional<Qualifier> oRefereed = Optional.ofNullable(instance.getRefereed());
|
|
if (oRefereed.isPresent()) {
|
|
if (!(oRefereed.get().getClassid().equals(Constants.DEFAULT_REFEREED.getClassid()) ||
|
|
oRefereed.get().getClassid().equals("UNKNOWN") ||
|
|
oRefereed.get().getClassid().equals(""))) {
|
|
return instance;
|
|
}
|
|
}
|
|
return null;
|
|
})
|
|
.filter(Objects::nonNull)
|
|
.collect(Collectors.toList()))
|
|
.orElse(null);
|
|
|
|
if ((instanceList != null) && instanceList.size() > 0) {
|
|
ri = new ResultInstance();
|
|
ri.setResultId(r.getId());
|
|
ri.setInstanceList(instanceList);
|
|
}
|
|
|
|
return ri;
|
|
|
|
}, Encoders.bean(ResultInstance.class))
|
|
.filter(Objects::nonNull)
|
|
.write()
|
|
.mode(SaveMode.Overwrite)
|
|
.option("compression", "gzip")
|
|
.json(outputPath);
|
|
}
|
|
|
|
public static <R> Dataset<R> readPath(
|
|
SparkSession spark, String inputPath, Class<R> clazz) {
|
|
return spark
|
|
.read()
|
|
.textFile(inputPath)
|
|
.map((MapFunction<String, R>) value -> OBJECT_MAPPER.readValue(value, clazz), Encoders.bean(clazz));
|
|
}
|
|
|
|
}
|