forked from D-Net/dnet-hadoop
[graph cleaning] testing the collectedfron and hostedby patch procedure
This commit is contained in:
parent
58c05731f9
commit
8e3edba318
|
|
@ -25,11 +25,13 @@ 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.common.action.model.MasterDuplicate;
|
||||
import eu.dnetlib.dhp.oa.graph.clean.country.CleanCountrySparkJob;
|
||||
import eu.dnetlib.dhp.schema.oaf.Instance;
|
||||
import eu.dnetlib.dhp.schema.oaf.KeyValue;
|
||||
import eu.dnetlib.dhp.schema.oaf.Result;
|
||||
import eu.dnetlib.dhp.utils.DHPUtils;
|
||||
import scala.Tuple2;
|
||||
|
||||
public class CleanCfHbSparkJob {
|
||||
|
|
@ -76,6 +78,8 @@ public class CleanCfHbSparkJob {
|
|||
conf,
|
||||
isSparkSessionManaged,
|
||||
spark -> {
|
||||
HdfsSupport.remove(outputPath, spark.sparkContext().hadoopConfiguration());
|
||||
HdfsSupport.remove(resolvedPath, spark.sparkContext().hadoopConfiguration());
|
||||
cleanCfHb(
|
||||
spark, inputPath, entityClazz, resolvedPath, dsMasterDuplicatePath, outputPath);
|
||||
});
|
||||
|
|
@ -95,30 +99,12 @@ public class CleanCfHbSparkJob {
|
|||
.read()
|
||||
.textFile(inputPath)
|
||||
.map(as(entityClazz), Encoders.bean(entityClazz))
|
||||
.flatMap(
|
||||
(FlatMapFunction<T, IdCfHbMapping>) r -> {
|
||||
final List<IdCfHbMapping> list = Stream
|
||||
.concat(
|
||||
r.getCollectedfrom().stream().map(KeyValue::getKey),
|
||||
Stream
|
||||
.concat(
|
||||
r.getInstance().stream().map(Instance::getHostedby).map(KeyValue::getKey),
|
||||
r.getInstance().stream().map(Instance::getCollectedfrom).map(KeyValue::getKey)))
|
||||
.distinct()
|
||||
.map(s -> asIdCfHbMapping(r.getId(), s))
|
||||
.collect(Collectors.toList());
|
||||
return list.iterator();
|
||||
},
|
||||
Encoders.bean(IdCfHbMapping.class));
|
||||
.flatMap(flattenCfHbFn(), Encoders.bean(IdCfHbMapping.class));
|
||||
|
||||
// set the EMPTY master ID/NAME and save it
|
||||
resolved
|
||||
.joinWith(md, resolved.col("cfhb").equalTo(md.col("duplicateId")))
|
||||
.map((MapFunction<Tuple2<IdCfHbMapping, MasterDuplicate>, IdCfHbMapping>) t -> {
|
||||
t._1().setMasterId(t._2().getMasterId());
|
||||
t._1().setMasterName(t._2().getMasterName());
|
||||
return t._1();
|
||||
}, Encoders.bean(IdCfHbMapping.class))
|
||||
.map(asIdCfHbMapping(), Encoders.bean(IdCfHbMapping.class))
|
||||
.write()
|
||||
.mode(SaveMode.Overwrite)
|
||||
.json(resolvedPath);
|
||||
|
|
@ -140,18 +126,37 @@ public class CleanCfHbSparkJob {
|
|||
.joinWith(resolvedDS, res.col("id").equalTo(resolvedDS.col("resultId")), "left")
|
||||
.groupByKey((MapFunction<Tuple2<T, IdCfHbMapping>, String>) t -> t._1().getId(), Encoders.STRING())
|
||||
.mapGroups(getMapGroupsFunction(), Encoders.bean(entityClazz))
|
||||
//.agg(new IdCfHbMappingAggregator(entityClazz).toColumn())
|
||||
.write()
|
||||
.mode(SaveMode.Overwrite)
|
||||
.option("compression", "gzip")
|
||||
.json(outputPath);
|
||||
}
|
||||
|
||||
@NotNull
|
||||
private static MapFunction<Tuple2<IdCfHbMapping, MasterDuplicate>, IdCfHbMapping> asIdCfHbMapping() {
|
||||
return t -> {
|
||||
t._1().setMasterId(t._2().getMasterId());
|
||||
t._1().setMasterName(t._2().getMasterName());
|
||||
return t._1();
|
||||
};
|
||||
}
|
||||
|
||||
private static <T extends Result> FlatMapFunction<T, IdCfHbMapping> flattenCfHbFn() {
|
||||
return r -> Stream
|
||||
.concat(
|
||||
r.getCollectedfrom().stream().map(KeyValue::getKey),
|
||||
Stream
|
||||
.concat(
|
||||
r.getInstance().stream().map(Instance::getHostedby).map(KeyValue::getKey),
|
||||
r.getInstance().stream().map(Instance::getCollectedfrom).map(KeyValue::getKey)))
|
||||
.distinct()
|
||||
.map(s -> asIdCfHbMapping(r.getId(), s))
|
||||
.iterator();
|
||||
}
|
||||
|
||||
private static <T extends Result> MapGroupsFunction<String, Tuple2<T, IdCfHbMapping>, T> getMapGroupsFunction() {
|
||||
return new MapGroupsFunction<String, Tuple2<T, IdCfHbMapping>, T>() {
|
||||
@Override
|
||||
public T call(String key, Iterator<Tuple2<T, IdCfHbMapping>> values) throws Exception {
|
||||
public T call(String key, Iterator<Tuple2<T, IdCfHbMapping>> values) {
|
||||
final Tuple2<T, IdCfHbMapping> first = values.next();
|
||||
final T res = first._1();
|
||||
|
||||
|
|
|
|||
|
|
@ -1,8 +1,16 @@
|
|||
|
||||
package eu.dnetlib.dhp.oa.graph.clean.cfhb;
|
||||
|
||||
import com.fasterxml.jackson.databind.ObjectMapper;
|
||||
import eu.dnetlib.dhp.schema.oaf.Dataset;
|
||||
import eu.dnetlib.dhp.schema.oaf.Publication;
|
||||
import static org.junit.jupiter.api.Assertions.assertEquals;
|
||||
import static org.junit.jupiter.api.Assertions.assertTrue;
|
||||
|
||||
import java.io.File;
|
||||
import java.io.IOException;
|
||||
import java.net.URISyntaxException;
|
||||
import java.nio.file.Files;
|
||||
import java.nio.file.Path;
|
||||
import java.nio.file.Paths;
|
||||
|
||||
import org.apache.commons.io.FileUtils;
|
||||
import org.apache.spark.SparkConf;
|
||||
import org.apache.spark.api.java.JavaSparkContext;
|
||||
|
|
@ -13,12 +21,10 @@ import org.junit.jupiter.api.*;
|
|||
import org.slf4j.Logger;
|
||||
import org.slf4j.LoggerFactory;
|
||||
|
||||
import java.io.File;
|
||||
import java.io.IOException;
|
||||
import java.net.URISyntaxException;
|
||||
import java.nio.file.Files;
|
||||
import java.nio.file.Path;
|
||||
import java.nio.file.Paths;
|
||||
import com.fasterxml.jackson.databind.ObjectMapper;
|
||||
|
||||
import eu.dnetlib.dhp.schema.oaf.Dataset;
|
||||
import eu.dnetlib.dhp.schema.oaf.Publication;
|
||||
|
||||
public class CleanCfHbSparkJobTest {
|
||||
|
||||
|
|
@ -56,11 +62,13 @@ public class CleanCfHbSparkJobTest {
|
|||
FileUtils
|
||||
.copyFileToDirectory(
|
||||
Paths
|
||||
.get(CleanCfHbSparkJobTest.class.getResource("/eu/dnetlib/dhp/oa/graph/clean/cfhb/masterduplicate.json").toURI())
|
||||
.get(
|
||||
CleanCfHbSparkJobTest.class
|
||||
.getResource("/eu/dnetlib/dhp/oa/graph/clean/cfhb/masterduplicate.json")
|
||||
.toURI())
|
||||
.toFile(),
|
||||
testBaseTmpPath.resolve("workingDir").resolve("masterduplicate").toFile());
|
||||
|
||||
|
||||
graphInputPath = testBaseTmpPath.resolve("input").resolve("entities").toString();
|
||||
resolvedPath = testBaseTmpPath.resolve("workingDir").resolve("cfHbResolved").toString();
|
||||
graphOutputPath = testBaseTmpPath.resolve("workingDir").resolve("cfHbPatched").toString();
|
||||
|
|
@ -89,27 +97,114 @@ public class CleanCfHbSparkJobTest {
|
|||
@Test
|
||||
void testCleanCfHbSparkJob() throws Exception {
|
||||
final String outputPath = graphOutputPath + "/dataset";
|
||||
final String inputPath = graphInputPath + "/dataset";
|
||||
|
||||
org.apache.spark.sql.Dataset<Dataset> records = read(spark, inputPath, Dataset.class);
|
||||
Dataset d = records
|
||||
.filter("id = '50|doi_________::09821844208a5cd6300b2bfb13bca1b9'")
|
||||
.first();
|
||||
assertEquals("10|re3data_____::4c4416659cb74c2e0e891a883a047cbc", d.getCollectedfrom().get(0).getKey());
|
||||
assertEquals("Bacterial Protein Interaction Database - DUP", d.getCollectedfrom().get(0).getValue());
|
||||
assertEquals(
|
||||
"10|re3data_____::4c4416659cb74c2e0e891a883a047cbc", d.getInstance().get(0).getCollectedfrom().getKey());
|
||||
assertEquals(
|
||||
"Bacterial Protein Interaction Database - DUP", d.getInstance().get(0).getCollectedfrom().getValue());
|
||||
|
||||
d = records
|
||||
.filter("id = '50|DansKnawCris::0dd644304b7116e8e58da3a5e3adc37a'")
|
||||
.first();
|
||||
assertEquals("10|opendoar____::788b4ac1e172d8e520c2b9461c0a3d35", d.getCollectedfrom().get(0).getKey());
|
||||
assertEquals("FILUR DATA - DUP", d.getCollectedfrom().get(0).getValue());
|
||||
assertEquals(
|
||||
"10|opendoar____::788b4ac1e172d8e520c2b9461c0a3d35", d.getInstance().get(0).getCollectedfrom().getKey());
|
||||
assertEquals("FILUR DATA - DUP", d.getInstance().get(0).getCollectedfrom().getValue());
|
||||
assertEquals(
|
||||
"10|re3data_____::6ffd7bc058f762912dc494cd9c175341", d.getInstance().get(0).getHostedby().getKey());
|
||||
assertEquals("depositar - DUP", d.getInstance().get(0).getHostedby().getValue());
|
||||
|
||||
d = records
|
||||
.filter("id = '50|DansKnawCris::203a27996ddc0fd1948258e5b7dec61c'")
|
||||
.first();
|
||||
assertEquals("10|openaire____::c6df70599aa984f16ee52b4b86d2e89f", d.getCollectedfrom().get(0).getKey());
|
||||
assertEquals("DANS (Data Archiving and Networked Services)", d.getCollectedfrom().get(0).getValue());
|
||||
assertEquals(
|
||||
"10|openaire____::c6df70599aa984f16ee52b4b86d2e89f", d.getInstance().get(0).getCollectedfrom().getKey());
|
||||
assertEquals(
|
||||
"DANS (Data Archiving and Networked Services)", d.getInstance().get(0).getCollectedfrom().getValue());
|
||||
assertEquals(
|
||||
"10|openaire____::c6df70599aa984f16ee52b4b86d2e89f", d.getInstance().get(0).getHostedby().getKey());
|
||||
assertEquals("DANS (Data Archiving and Networked Services)", d.getInstance().get(0).getHostedby().getValue());
|
||||
|
||||
CleanCfHbSparkJob
|
||||
.main(
|
||||
new String[] {
|
||||
"--isSparkSessionManaged", Boolean.FALSE.toString(),
|
||||
"--inputPath", graphInputPath + "/dataset",
|
||||
"--inputPath", inputPath,
|
||||
"--outputPath", outputPath,
|
||||
"--resolvedPath", resolvedPath + "/dataset",
|
||||
"--graphTableClassName", Dataset.class.getCanonicalName(),
|
||||
"--datasourceMasterDuplicate", dsMasterDuplicatePath
|
||||
});
|
||||
|
||||
//final JavaSparkContext sc = JavaSparkContext.fromSparkContext(spark.sparkContext());
|
||||
assertTrue(Files.exists(Paths.get(graphOutputPath, "dataset")));
|
||||
|
||||
Assertions.assertTrue(Files.exists(Paths.get(graphOutputPath, "dataset")));
|
||||
records = read(spark, outputPath, Dataset.class);
|
||||
|
||||
final org.apache.spark.sql.Dataset<Dataset> d = spark
|
||||
assertEquals(3, records.count());
|
||||
|
||||
d = records
|
||||
.filter("id = '50|doi_________::09821844208a5cd6300b2bfb13bca1b9'")
|
||||
.first();
|
||||
assertEquals("10|fairsharing_::a29d1598024f9e87beab4b98411d48ce", d.getCollectedfrom().get(0).getKey());
|
||||
assertEquals("Bacterial Protein Interaction Database", d.getCollectedfrom().get(0).getValue());
|
||||
assertEquals(
|
||||
"10|fairsharing_::a29d1598024f9e87beab4b98411d48ce", d.getInstance().get(0).getCollectedfrom().getKey());
|
||||
assertEquals("Bacterial Protein Interaction Database", d.getInstance().get(0).getCollectedfrom().getValue());
|
||||
|
||||
d = records
|
||||
.filter("id = '50|DansKnawCris::0dd644304b7116e8e58da3a5e3adc37a'")
|
||||
.first();
|
||||
assertEquals("10|re3data_____::fc1db64b3964826913b1e9eafe830490", d.getCollectedfrom().get(0).getKey());
|
||||
assertEquals("FULIR Data", d.getCollectedfrom().get(0).getValue());
|
||||
assertEquals(
|
||||
"10|re3data_____::fc1db64b3964826913b1e9eafe830490", d.getInstance().get(0).getCollectedfrom().getKey());
|
||||
assertEquals("FULIR Data", d.getInstance().get(0).getCollectedfrom().getValue());
|
||||
assertEquals(
|
||||
"10|fairsharing_::3f647cadf56541fb9513cb63ec370187", d.getInstance().get(0).getHostedby().getKey());
|
||||
assertEquals("depositar", d.getInstance().get(0).getHostedby().getValue());
|
||||
|
||||
d = records
|
||||
.filter("id = '50|DansKnawCris::203a27996ddc0fd1948258e5b7dec61c'")
|
||||
.first();
|
||||
assertEquals("10|openaire____::c6df70599aa984f16ee52b4b86d2e89f", d.getCollectedfrom().get(0).getKey());
|
||||
assertEquals("DANS (Data Archiving and Networked Services)", d.getCollectedfrom().get(0).getValue());
|
||||
assertEquals(
|
||||
"10|openaire____::c6df70599aa984f16ee52b4b86d2e89f", d.getInstance().get(0).getCollectedfrom().getKey());
|
||||
assertEquals(
|
||||
"DANS (Data Archiving and Networked Services)", d.getInstance().get(0).getCollectedfrom().getValue());
|
||||
assertEquals(
|
||||
"10|openaire____::c6df70599aa984f16ee52b4b86d2e89f", d.getInstance().get(0).getHostedby().getKey());
|
||||
assertEquals("DANS (Data Archiving and Networked Services)", d.getInstance().get(0).getHostedby().getValue());
|
||||
|
||||
d = records
|
||||
.filter("id = '50|DansKnawCris::203a27996ddc0fd1948258e5b7dec61c'")
|
||||
.first();
|
||||
assertEquals("10|openaire____::c6df70599aa984f16ee52b4b86d2e89f", d.getCollectedfrom().get(0).getKey());
|
||||
assertEquals("DANS (Data Archiving and Networked Services)", d.getCollectedfrom().get(0).getValue());
|
||||
assertEquals(
|
||||
"10|openaire____::c6df70599aa984f16ee52b4b86d2e89f", d.getInstance().get(0).getCollectedfrom().getKey());
|
||||
assertEquals(
|
||||
"DANS (Data Archiving and Networked Services)", d.getInstance().get(0).getCollectedfrom().getValue());
|
||||
assertEquals(
|
||||
"10|openaire____::c6df70599aa984f16ee52b4b86d2e89f", d.getInstance().get(0).getHostedby().getKey());
|
||||
assertEquals("DANS (Data Archiving and Networked Services)", d.getInstance().get(0).getHostedby().getValue());
|
||||
}
|
||||
|
||||
private <R> org.apache.spark.sql.Dataset<R> read(SparkSession spark, String path, Class<R> clazz) {
|
||||
return spark
|
||||
.read()
|
||||
.textFile(outputPath)
|
||||
.map(as(Dataset.class), Encoders.bean(Dataset.class));
|
||||
Assertions.assertEquals(3, d.count());
|
||||
|
||||
.textFile(path)
|
||||
.map(as(clazz), Encoders.bean(clazz));
|
||||
}
|
||||
|
||||
private static <R> MapFunction<String, R> as(Class<R> clazz) {
|
||||
|
|
|
|||
Loading…
Reference in New Issue