forked from D-Net/dnet-hadoop
93 lines
2.7 KiB
Java
93 lines
2.7 KiB
Java
|
|
package eu.dnetlib.dhp.ircdl_extention;
|
|
|
|
import static eu.dnetlib.dhp.common.SparkSessionSupport.runWithSparkHiveSession;
|
|
|
|
import java.util.Optional;
|
|
|
|
import org.apache.commons.io.IOUtils;
|
|
import org.apache.spark.SparkConf;
|
|
import org.apache.spark.api.java.function.FilterFunction;
|
|
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.Result;
|
|
import eu.dnetlib.dhp.schema.oaf.Datasource;
|
|
|
|
public class PrepareResultFromInstRepo {
|
|
|
|
public static void main(String[] args) throws Exception {
|
|
|
|
String jsonConfiguration = IOUtils
|
|
.toString(
|
|
PrepareResultFromInstRepo.class
|
|
.getResourceAsStream(
|
|
"/eu/dnetlib/dhp/ircdl_extention/prepare_instrepo_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");
|
|
|
|
final String datasourcePath = parser.get("datasourcePath");
|
|
|
|
SparkConf conf = new SparkConf();
|
|
conf.set("hive.metastore.uris", "thrift://iis-cdh5-test-m3.ocean.icm.edu.pl:9083");
|
|
|
|
runWithSparkHiveSession(
|
|
conf,
|
|
isSparkSessionManaged,
|
|
spark -> {
|
|
Utils.removeOutputDir(spark, outputPath);
|
|
selectResultFromInstRepo(spark, inputPath, outputPath, datasourcePath);
|
|
});
|
|
}
|
|
|
|
private static void selectResultFromInstRepo(SparkSession spark, String inputPath, String output_path,
|
|
String datasourcePath) {
|
|
Dataset<Datasource> datasource = Utils.readPath(spark, datasourcePath, Datasource.class);
|
|
Dataset<Result> res = Utils
|
|
.readPath(
|
|
spark, inputPath, Result.class)
|
|
.filter(
|
|
(FilterFunction<Result>) r -> !r.getId().startsWith("50|doiboost")
|
|
&& !r.getId().startsWith("50|scholix")
|
|
&& !r.getId().startsWith("50|datacite")
|
|
&& !r.getId().startsWith("50|dedup"));
|
|
|
|
datasource.createOrReplaceTempView("datasource");
|
|
res.createOrReplaceTempView("result");
|
|
|
|
spark
|
|
.sql(
|
|
"SELECT t.id, t.deletedbyinference, t.name, t.surname, t.cf, t.fullname, t.pid, t.oid " +
|
|
"FROM " +
|
|
"(Select * " +
|
|
"from result " +
|
|
"LATERAL VIEW explode(cf.key) c as cfromkey) as t " +
|
|
"join " +
|
|
"datasource d " +
|
|
"on " +
|
|
"d.id = t.cfromkey " +
|
|
"and d.datasourcetype.classid = 'pubsrepository::institutional'")
|
|
.as(Encoders.bean(Result.class))
|
|
.write()
|
|
.option("compressio", "gzip")
|
|
.mode(SaveMode.Overwrite)
|
|
.json(output_path);
|
|
|
|
}
|
|
|
|
}
|