[SDG-FOS] alternative way to get input data to avoid OOM error while getting csv
parent
a706ba0c08
commit
92fd69e25d
@ -0,0 +1,91 @@
|
||||
|
||||
package eu.dnetlib.dhp.actionmanager.createunresolvedentities;
|
||||
|
||||
import static eu.dnetlib.dhp.actionmanager.Constants.DEFAULT_DELIMITER;
|
||||
import static eu.dnetlib.dhp.actionmanager.Constants.isSparkSessionManaged;
|
||||
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.conf.Configuration;
|
||||
import org.apache.hadoop.fs.FileSystem;
|
||||
import org.apache.spark.SparkConf;
|
||||
import org.apache.spark.api.java.function.MapFunction;
|
||||
import org.apache.spark.sql.*;
|
||||
import org.slf4j.Logger;
|
||||
import org.slf4j.LoggerFactory;
|
||||
|
||||
import eu.dnetlib.dhp.actionmanager.createunresolvedentities.model.FOSDataModel;
|
||||
import eu.dnetlib.dhp.application.ArgumentApplicationParser;
|
||||
|
||||
public class GetFOSSparkJob implements Serializable {
|
||||
|
||||
private static final Logger log = LoggerFactory.getLogger(GetFOSSparkJob.class);
|
||||
|
||||
public static void main(String[] args) throws Exception {
|
||||
|
||||
String jsonConfiguration = IOUtils
|
||||
.toString(
|
||||
GetFOSSparkJob.class
|
||||
.getResourceAsStream(
|
||||
"/eu/dnetlib/dhp/actionmanager/createunresolvedentities/get_input_parameters.json"));
|
||||
|
||||
final ArgumentApplicationParser parser = new ArgumentApplicationParser(jsonConfiguration);
|
||||
|
||||
parser.parseArgument(args);
|
||||
|
||||
Boolean isSparkSessionManaged = isSparkSessionManaged(parser);
|
||||
log.info("isSparkSessionManaged: {}", isSparkSessionManaged);
|
||||
|
||||
// the path where the original fos csv file is stored
|
||||
final String sourcePath = parser.get("sourcePath");
|
||||
log.info("sourcePath {}", sourcePath);
|
||||
|
||||
// the path where to put the file as json
|
||||
final String outputPath = parser.get("outputPath");
|
||||
log.info("outputPath {}", outputPath);
|
||||
|
||||
final String delimiter = Optional
|
||||
.ofNullable(parser.get("delimiter"))
|
||||
.orElse(DEFAULT_DELIMITER);
|
||||
|
||||
SparkConf sconf = new SparkConf();
|
||||
runWithSparkSession(
|
||||
sconf,
|
||||
isSparkSessionManaged,
|
||||
spark -> {
|
||||
getFOS(
|
||||
spark,
|
||||
sourcePath,
|
||||
outputPath,
|
||||
delimiter);
|
||||
});
|
||||
}
|
||||
|
||||
private static void getFOS(SparkSession spark, String sourcePath, String outputPath, String delimiter) {
|
||||
Dataset<Row> fosData = spark
|
||||
.read()
|
||||
.format("csv")
|
||||
.option("sep", delimiter)
|
||||
.option("inferSchema", "true")
|
||||
.option("header", "true")
|
||||
.option("quotes", "\"")
|
||||
.load(sourcePath);
|
||||
|
||||
fosData.map((MapFunction<Row, FOSDataModel>) r -> {
|
||||
FOSDataModel fosDataModel = new FOSDataModel();
|
||||
fosDataModel.setDoi(r.getString(0).toLowerCase());
|
||||
fosDataModel.setLevel1(r.getString(1));
|
||||
fosDataModel.setLevel2(r.getString(2));
|
||||
fosDataModel.setLevel3(r.getString(3));
|
||||
return fosDataModel;
|
||||
}, Encoders.bean(FOSDataModel.class))
|
||||
.write()
|
||||
.mode(SaveMode.Overwrite)
|
||||
.json(outputPath);
|
||||
|
||||
}
|
||||
|
||||
}
|
@ -1,80 +0,0 @@
|
||||
|
||||
package eu.dnetlib.dhp.actionmanager.createunresolvedentities;
|
||||
|
||||
import java.io.IOException;
|
||||
import java.io.InputStreamReader;
|
||||
import java.io.Serializable;
|
||||
import java.util.Objects;
|
||||
import java.util.Optional;
|
||||
import java.util.zip.GZIPInputStream;
|
||||
|
||||
import org.apache.commons.io.IOUtils;
|
||||
import org.apache.hadoop.conf.Configuration;
|
||||
import org.apache.hadoop.fs.FileSystem;
|
||||
import org.apache.hadoop.fs.Path;
|
||||
import org.slf4j.Logger;
|
||||
import org.slf4j.LoggerFactory;
|
||||
|
||||
import eu.dnetlib.dhp.application.ArgumentApplicationParser;
|
||||
|
||||
import static eu.dnetlib.dhp.actionmanager.Constants.DEFAULT_DELIMITER;
|
||||
|
||||
public class GetInputData implements Serializable {
|
||||
|
||||
private static final Logger log = LoggerFactory.getLogger(GetInputData.class);
|
||||
|
||||
|
||||
|
||||
public static void main(final String[] args) throws Exception {
|
||||
final ArgumentApplicationParser parser = new ArgumentApplicationParser(
|
||||
IOUtils
|
||||
.toString(
|
||||
Objects
|
||||
.requireNonNull(
|
||||
GetInputData.class
|
||||
.getResourceAsStream(
|
||||
"/eu/dnetlib/dhp/actionmanager/createunresolvedentities/get_input_parameters.json"))));
|
||||
|
||||
parser.parseArgument(args);
|
||||
|
||||
// the path where the original fos csv file is stored
|
||||
final String sourcePath = parser.get("sourcePath");
|
||||
log.info("sourcePath {}", sourcePath);
|
||||
|
||||
// the path where to put the file as json
|
||||
final String outputPath = parser.get("outputPath");
|
||||
log.info("outputPath {}", outputPath);
|
||||
|
||||
final String hdfsNameNode = parser.get("hdfsNameNode");
|
||||
log.info("hdfsNameNode {}", hdfsNameNode);
|
||||
|
||||
final String classForName = parser.get("classForName");
|
||||
log.info("classForName {}", classForName);
|
||||
|
||||
final char delimiter = Optional
|
||||
.ofNullable(parser.get("delimiter"))
|
||||
.map(s -> s.charAt(0))
|
||||
.orElse(DEFAULT_DELIMITER);
|
||||
log.info("delimiter {}", delimiter);
|
||||
|
||||
Configuration conf = new Configuration();
|
||||
conf.set("fs.defaultFS", hdfsNameNode);
|
||||
|
||||
FileSystem fileSystem = FileSystem.get(conf);
|
||||
|
||||
new GetInputData().doRewrite(sourcePath, outputPath, classForName, delimiter, fileSystem);
|
||||
|
||||
}
|
||||
|
||||
public void doRewrite(String inputPath, String outputFile, String classForName, char delimiter, FileSystem fs)
|
||||
throws IOException, ClassNotFoundException {
|
||||
|
||||
|
||||
// reads the csv and writes it as its json equivalent
|
||||
try (InputStreamReader reader = new InputStreamReader(new GZIPInputStream(fs.open(new Path(inputPath))))) {
|
||||
eu.dnetlib.dhp.common.collection.GetCSV.getCsv(fs, reader, outputFile, classForName, delimiter);
|
||||
}
|
||||
|
||||
}
|
||||
|
||||
}
|
@ -0,0 +1,91 @@
|
||||
|
||||
package eu.dnetlib.dhp.actionmanager.createunresolvedentities;
|
||||
|
||||
import static eu.dnetlib.dhp.actionmanager.Constants.DEFAULT_DELIMITER;
|
||||
import static eu.dnetlib.dhp.actionmanager.Constants.isSparkSessionManaged;
|
||||
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.spark.SparkConf;
|
||||
import org.apache.spark.api.java.function.FilterFunction;
|
||||
import org.apache.spark.api.java.function.MapFunction;
|
||||
import org.apache.spark.sql.*;
|
||||
import org.slf4j.Logger;
|
||||
import org.slf4j.LoggerFactory;
|
||||
|
||||
import eu.dnetlib.dhp.actionmanager.createunresolvedentities.model.FOSDataModel;
|
||||
import eu.dnetlib.dhp.actionmanager.createunresolvedentities.model.SDGDataModel;
|
||||
import eu.dnetlib.dhp.application.ArgumentApplicationParser;
|
||||
|
||||
public class GetSDGSparkJob implements Serializable {
|
||||
|
||||
private static final Logger log = LoggerFactory.getLogger(GetSDGSparkJob.class);
|
||||
|
||||
public static void main(String[] args) throws Exception {
|
||||
|
||||
String jsonConfiguration = IOUtils
|
||||
.toString(
|
||||
GetSDGSparkJob.class
|
||||
.getResourceAsStream(
|
||||
"/eu/dnetlib/dhp/actionmanager/createunresolvedentities/get_input_parameters.json"));
|
||||
|
||||
final ArgumentApplicationParser parser = new ArgumentApplicationParser(jsonConfiguration);
|
||||
|
||||
parser.parseArgument(args);
|
||||
|
||||
Boolean isSparkSessionManaged = isSparkSessionManaged(parser);
|
||||
log.info("isSparkSessionManaged: {}", isSparkSessionManaged);
|
||||
|
||||
// the path where the original fos csv file is stored
|
||||
final String sourcePath = parser.get("sourcePath");
|
||||
log.info("sourcePath {}", sourcePath);
|
||||
|
||||
// the path where to put the file as json
|
||||
final String outputPath = parser.get("outputPath");
|
||||
log.info("outputPath {}", outputPath);
|
||||
|
||||
final String delimiter = Optional
|
||||
.ofNullable(parser.get("delimiter"))
|
||||
.orElse(DEFAULT_DELIMITER);
|
||||
|
||||
SparkConf sconf = new SparkConf();
|
||||
runWithSparkSession(
|
||||
sconf,
|
||||
isSparkSessionManaged,
|
||||
spark -> {
|
||||
getSDG(
|
||||
spark,
|
||||
sourcePath,
|
||||
outputPath,
|
||||
delimiter);
|
||||
});
|
||||
}
|
||||
|
||||
private static void getSDG(SparkSession spark, String sourcePath, String outputPath, String delimiter) {
|
||||
Dataset<Row> sdgData = spark
|
||||
.read()
|
||||
.format("csv")
|
||||
.option("sep", delimiter)
|
||||
.option("inferSchema", "true")
|
||||
.option("header", "true")
|
||||
.option("quotes", "\"")
|
||||
.load(sourcePath);
|
||||
|
||||
sdgData.map((MapFunction<Row, SDGDataModel>) r -> {
|
||||
SDGDataModel sdgDataModel = new SDGDataModel();
|
||||
sdgDataModel.setDoi(r.getString(0).toLowerCase());
|
||||
sdgDataModel.setSbj(r.getString(1));
|
||||
|
||||
return sdgDataModel;
|
||||
}, Encoders.bean(SDGDataModel.class))
|
||||
.filter((FilterFunction<SDGDataModel>) sdg -> sdg.getSbj() != null)
|
||||
.write()
|
||||
.mode(SaveMode.Overwrite)
|
||||
.json(outputPath);
|
||||
|
||||
}
|
||||
|
||||
}
|
@ -1,48 +1,47 @@
|
||||
package eu.dnetlib.dhp.actionmanager.createunresolvedentities.model;
|
||||
|
||||
import com.opencsv.bean.CsvBindByPosition;
|
||||
package eu.dnetlib.dhp.actionmanager.createunresolvedentities.model;
|
||||
|
||||
import java.io.Serializable;
|
||||
|
||||
public class SDGDataModel implements Serializable{
|
||||
import com.opencsv.bean.CsvBindByPosition;
|
||||
|
||||
@CsvBindByPosition(position = 0)
|
||||
public class SDGDataModel implements Serializable {
|
||||
|
||||
@CsvBindByPosition(position = 0)
|
||||
// @CsvBindByName(column = "doi")
|
||||
private String doi;
|
||||
private String doi;
|
||||
|
||||
@CsvBindByPosition(position = 1)
|
||||
@CsvBindByPosition(position = 1)
|
||||
// @CsvBindByName(column = "sdg")
|
||||
private String sbj;
|
||||
|
||||
|
||||
public SDGDataModel() {
|
||||
private String sbj;
|
||||
|
||||
}
|
||||
public SDGDataModel() {
|
||||
|
||||
public SDGDataModel(String doi, String sbj) {
|
||||
this.doi = doi;
|
||||
this.sbj = sbj;
|
||||
}
|
||||
|
||||
}
|
||||
public SDGDataModel(String doi, String sbj) {
|
||||
this.doi = doi;
|
||||
this.sbj = sbj;
|
||||
|
||||
public static SDGDataModel newInstance(String d, String sbj) {
|
||||
return new SDGDataModel(d, sbj);
|
||||
}
|
||||
}
|
||||
|
||||
public String getDoi() {
|
||||
return doi;
|
||||
}
|
||||
public static SDGDataModel newInstance(String d, String sbj) {
|
||||
return new SDGDataModel(d, sbj);
|
||||
}
|
||||
|
||||
public void setDoi(String doi) {
|
||||
this.doi = doi;
|
||||
}
|
||||
public String getDoi() {
|
||||
return doi;
|
||||
}
|
||||
|
||||
public void setDoi(String doi) {
|
||||
this.doi = doi;
|
||||
}
|
||||
|
||||
public String getSbj() {
|
||||
return sbj;
|
||||
}
|
||||
public String getSbj() {
|
||||
return sbj;
|
||||
}
|
||||
|
||||
public void setSbj(String sbj) {
|
||||
this.sbj = sbj;
|
||||
}
|
||||
public void setSbj(String sbj) {
|
||||
this.sbj = sbj;
|
||||
}
|
||||
}
|
||||
|
Loading…
Reference in New Issue