forked from D-Net/dnet-hadoop
76 lines
2.1 KiB
Java
76 lines
2.1 KiB
Java
|
|
package eu.dnetlib.dhp.ircdl_extention;
|
|
|
|
import static eu.dnetlib.dhp.common.SparkSessionSupport.runWithSparkSession;
|
|
|
|
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 eu.dnetlib.dhp.application.ArgumentApplicationParser;
|
|
import eu.dnetlib.dhp.ircdl_extention.model.Orcid;
|
|
|
|
public class PrepareNormalizedOrcid {
|
|
|
|
public static void main(String[] args) throws Exception {
|
|
|
|
String jsonConfiguration = IOUtils
|
|
.toString(
|
|
PrepareNormalizedOrcid.class
|
|
.getResourceAsStream(
|
|
"/eu/dnetlib/dhp/ircdl_extention/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);
|
|
|
|
final String inputPath = parser.get("inputPath");
|
|
|
|
final String outputPath = parser.get("outputPath");
|
|
|
|
SparkConf conf = new SparkConf();
|
|
|
|
runWithSparkSession(
|
|
conf,
|
|
isSparkSessionManaged,
|
|
spark -> {
|
|
Utils.removeOutputDir(spark, outputPath);
|
|
execNormalize(spark, outputPath, inputPath);
|
|
});
|
|
}
|
|
|
|
private static void execNormalize(SparkSession spark, String outputPath, String inputPath) {
|
|
Dataset<Orcid> orcid = Utils.readPath(spark, inputPath, Orcid.class);
|
|
orcid.map((MapFunction<Orcid, Orcid>) o -> {
|
|
o.setName(Utils.normalizeString(o.getName()));
|
|
o.setSurname(Utils.normalizeString(o.getSurname()));
|
|
o.setCreditname(Utils.normalizeString(o.getCreditname()));
|
|
o
|
|
.setOtherNames(
|
|
o
|
|
.getOtherNames()
|
|
.stream()
|
|
.map(on -> Utils.normalizeString(on))
|
|
.collect(Collectors.toList()));
|
|
return o;
|
|
}, Encoders.bean(Orcid.class))
|
|
.write()
|
|
.option("compression", "gzip")
|
|
.mode(SaveMode.Overwrite)
|
|
.json(outputPath);
|
|
}
|
|
|
|
}
|