Merge pull request 'Clean Country' (#241) from clean_country into beta
Reviewed-on: #241transformation_wf
commit
6ad38ade74
@ -0,0 +1,168 @@
|
||||
|
||||
package eu.dnetlib.dhp.oa.graph.clean.country;
|
||||
|
||||
import static eu.dnetlib.dhp.common.SparkSessionSupport.runWithSparkSession;
|
||||
|
||||
import java.io.Serializable;
|
||||
import java.util.List;
|
||||
import java.util.Optional;
|
||||
import java.util.stream.Collectors;
|
||||
|
||||
import javax.swing.text.html.Option;
|
||||
|
||||
import eu.dnetlib.dhp.schema.oaf.utils.PidType;
|
||||
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.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;
|
||||
|
||||
/**
|
||||
* @author miriam.baglioni
|
||||
* @Date 20/07/22
|
||||
*/
|
||||
import com.fasterxml.jackson.databind.ObjectMapper;
|
||||
|
||||
import eu.dnetlib.dhp.application.ArgumentApplicationParser;
|
||||
import eu.dnetlib.dhp.oa.graph.clean.CleanContextSparkJob;
|
||||
import eu.dnetlib.dhp.schema.oaf.Country;
|
||||
import eu.dnetlib.dhp.schema.oaf.Result;
|
||||
import eu.dnetlib.dhp.schema.oaf.StructuredProperty;
|
||||
|
||||
public class CleanCountrySparkJob implements Serializable {
|
||||
private static final Logger log = LoggerFactory.getLogger(CleanCountrySparkJob.class);
|
||||
|
||||
private static final ObjectMapper OBJECT_MAPPER = new ObjectMapper();
|
||||
|
||||
public static void main(String[] args) throws Exception {
|
||||
|
||||
String jsonConfiguration = IOUtils
|
||||
.toString(
|
||||
CleanContextSparkJob.class
|
||||
.getResourceAsStream(
|
||||
"/eu/dnetlib/dhp/oa/graph/input_clean_country_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);
|
||||
|
||||
String inputPath = parser.get("inputPath");
|
||||
log.info("inputPath: {}", inputPath);
|
||||
|
||||
String workingPath = parser.get("workingPath");
|
||||
log.info("workingPath: {}", workingPath);
|
||||
|
||||
String datasourcePath = parser.get("hostedBy");
|
||||
log.info("datasourcePath: {}", datasourcePath);
|
||||
|
||||
String country = parser.get("country");
|
||||
log.info("country: {}", country);
|
||||
|
||||
String[] verifyParam = parser.get("verifyParam").split(";");
|
||||
log.info("verifyParam: {}", verifyParam);
|
||||
|
||||
String collectedfrom = parser.get("collectedfrom");
|
||||
log.info("collectedfrom: {}", collectedfrom);
|
||||
|
||||
String graphTableClassName = parser.get("graphTableClassName");
|
||||
log.info("graphTableClassName: {}", graphTableClassName);
|
||||
|
||||
Class<? extends Result> entityClazz = (Class<? extends Result>) Class.forName(graphTableClassName);
|
||||
|
||||
SparkConf conf = new SparkConf();
|
||||
runWithSparkSession(
|
||||
conf,
|
||||
isSparkSessionManaged,
|
||||
spark -> {
|
||||
|
||||
cleanCountry(
|
||||
spark, country, verifyParam, inputPath, entityClazz, workingPath, collectedfrom, datasourcePath);
|
||||
});
|
||||
}
|
||||
|
||||
private static <T extends Result> void cleanCountry(SparkSession spark, String country, String[] verifyParam,
|
||||
String inputPath, Class<T> entityClazz, String workingPath, String collectedfrom, String datasourcePath) {
|
||||
|
||||
List<String> hostedBy = spark
|
||||
.read()
|
||||
.textFile(datasourcePath)
|
||||
.collectAsList();
|
||||
|
||||
Dataset<T> res = spark
|
||||
.read()
|
||||
.textFile(inputPath)
|
||||
.map(
|
||||
(MapFunction<String, T>) value -> OBJECT_MAPPER.readValue(value, entityClazz),
|
||||
Encoders.bean(entityClazz));
|
||||
|
||||
res.map((MapFunction<T, T>) r -> {
|
||||
if (r.getInstance().stream().anyMatch(i -> hostedBy.contains(i.getHostedby().getKey())) ||
|
||||
!r.getCollectedfrom().stream().anyMatch(cf -> cf.getValue().equals(collectedfrom))) {
|
||||
return r;
|
||||
}
|
||||
|
||||
if (r
|
||||
.getPid()
|
||||
.stream()
|
||||
.anyMatch(p -> p.getQualifier().getClassid()
|
||||
.equals(PidType.doi) && pidInParam(p.getValue(), verifyParam))) {
|
||||
r
|
||||
.setCountry(
|
||||
r
|
||||
.getCountry()
|
||||
.stream()
|
||||
.filter(
|
||||
c -> toTakeCountry(c, country))
|
||||
.collect(Collectors.toList()));
|
||||
|
||||
}
|
||||
|
||||
return r;
|
||||
}, Encoders.bean(entityClazz))
|
||||
.write()
|
||||
.mode(SaveMode.Overwrite)
|
||||
.option("compression", "gzip")
|
||||
.json(workingPath);
|
||||
|
||||
spark
|
||||
.read()
|
||||
.textFile(workingPath)
|
||||
.map(
|
||||
(MapFunction<String, T>) value -> OBJECT_MAPPER.readValue(value, entityClazz),
|
||||
Encoders.bean(entityClazz))
|
||||
.write()
|
||||
.mode(SaveMode.Overwrite)
|
||||
.option("compression", "gzip")
|
||||
.json(inputPath);
|
||||
}
|
||||
|
||||
private static boolean pidInParam(String value, String[] verifyParam) {
|
||||
for (String s : verifyParam)
|
||||
if (value.startsWith(s))
|
||||
return true;
|
||||
return false;
|
||||
}
|
||||
|
||||
private static boolean toTakeCountry(Country c, String country) {
|
||||
// If dataInfo is not set, or dataInfo.inferenceprovenance is not set or not present then it cannot be
|
||||
// inserted via propagation
|
||||
if (!Optional.ofNullable(c.getDataInfo()).isPresent())
|
||||
return true;
|
||||
if (!Optional.ofNullable(c.getDataInfo().getInferenceprovenance()).isPresent())
|
||||
return true;
|
||||
return !(c
|
||||
.getClassid()
|
||||
.equalsIgnoreCase(country) &&
|
||||
c.getDataInfo().getInferenceprovenance().equals("propagation"));
|
||||
}
|
||||
|
||||
}
|
||||
@ -0,0 +1,108 @@
|
||||
|
||||
package eu.dnetlib.dhp.oa.graph.clean.country;
|
||||
|
||||
import static eu.dnetlib.dhp.common.SparkSessionSupport.runWithSparkSession;
|
||||
|
||||
import java.io.Serializable;
|
||||
import java.util.List;
|
||||
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.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.oa.graph.clean.CleanContextSparkJob;
|
||||
import eu.dnetlib.dhp.schema.common.ModelConstants;
|
||||
import eu.dnetlib.dhp.schema.oaf.*;
|
||||
import scala.Tuple2;
|
||||
|
||||
/**
|
||||
* @author miriam.baglioni
|
||||
* @Date 22/07/22
|
||||
*/
|
||||
public class GetDatasourceFromCountry implements Serializable {
|
||||
private static final Logger log = LoggerFactory.getLogger(GetDatasourceFromCountry.class);
|
||||
|
||||
private static final ObjectMapper OBJECT_MAPPER = new ObjectMapper();
|
||||
|
||||
public static void main(String[] args) throws Exception {
|
||||
|
||||
String jsonConfiguration = IOUtils
|
||||
.toString(
|
||||
GetDatasourceFromCountry.class
|
||||
.getResourceAsStream(
|
||||
"/eu/dnetlib/dhp/oa/graph/input_datasource_country_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);
|
||||
|
||||
String inputPath = parser.get("inputPath");
|
||||
log.info("inputPath: {}", inputPath);
|
||||
|
||||
String workingPath = parser.get("workingPath");
|
||||
log.info("workingPath: {}", workingPath);
|
||||
|
||||
String country = parser.get("country");
|
||||
log.info("country: {}", country);
|
||||
|
||||
SparkConf conf = new SparkConf();
|
||||
runWithSparkSession(
|
||||
conf,
|
||||
isSparkSessionManaged,
|
||||
spark -> {
|
||||
|
||||
getDatasourceFromCountry(spark, country, inputPath, workingPath);
|
||||
});
|
||||
}
|
||||
|
||||
private static void getDatasourceFromCountry(SparkSession spark, String country, String inputPath,
|
||||
String workingPath) {
|
||||
|
||||
Dataset<Organization> organization = spark
|
||||
.read()
|
||||
.textFile(inputPath + "/organization")
|
||||
.map(
|
||||
(MapFunction<String, Organization>) value -> OBJECT_MAPPER.readValue(value, Organization.class),
|
||||
Encoders.bean(Organization.class))
|
||||
.filter(
|
||||
(FilterFunction<Organization>) o -> !o.getDataInfo().getDeletedbyinference() &&
|
||||
o.getCountry().getClassid().length() > 0 &&
|
||||
o.getCountry().getClassid().equals(country));
|
||||
;
|
||||
|
||||
// filtering of the relations taking the non deleted by inference and those with IsProvidedBy as relclass
|
||||
Dataset<Relation> relation = spark
|
||||
.read()
|
||||
.textFile(inputPath + "/relation")
|
||||
.map(
|
||||
(MapFunction<String, Relation>) value -> OBJECT_MAPPER.readValue(value, Relation.class),
|
||||
Encoders.bean(Relation.class))
|
||||
.filter(
|
||||
(FilterFunction<Relation>) rel -> rel.getRelClass().equalsIgnoreCase(ModelConstants.IS_PROVIDED_BY) &&
|
||||
!rel.getDataInfo().getDeletedbyinference());
|
||||
|
||||
organization
|
||||
.joinWith(relation, organization.col("id").equalTo(relation.col("target")), "left")
|
||||
.map((MapFunction<Tuple2<Organization, Relation>, String>) t2 -> t2._2().getSource(), Encoders.STRING())
|
||||
.write()
|
||||
.mode(SaveMode.Overwrite)
|
||||
.option("compression", "gzip")
|
||||
.json(workingPath);
|
||||
|
||||
}
|
||||
}
|
||||
@ -0,0 +1,49 @@
|
||||
[
|
||||
{
|
||||
"paramName": "issm",
|
||||
"paramLongName": "isSparkSessionManaged",
|
||||
"paramDescription": "when true will stop SparkSession after job execution",
|
||||
"paramRequired": false
|
||||
},
|
||||
{
|
||||
"paramName": "in",
|
||||
"paramLongName": "inputPath",
|
||||
"paramDescription": "the path to the graph data dump to read",
|
||||
"paramRequired": true
|
||||
},
|
||||
{
|
||||
"paramName": "wp",
|
||||
"paramLongName": "workingPath",
|
||||
"paramDescription": "the path to store the output graph",
|
||||
"paramRequired": true
|
||||
},
|
||||
{
|
||||
"paramName": "c",
|
||||
"paramLongName": "country",
|
||||
"paramDescription": "the id of the context to be removed",
|
||||
"paramRequired": true
|
||||
},
|
||||
{
|
||||
"paramName": "class",
|
||||
"paramLongName": "graphTableClassName",
|
||||
"paramDescription": "class name moelling the graph table",
|
||||
"paramRequired": true
|
||||
},{
|
||||
"paramName": "vf",
|
||||
"paramLongName": "verifyParam",
|
||||
"paramDescription": "the parameter to be verified to remove the country",
|
||||
"paramRequired": true
|
||||
},
|
||||
{
|
||||
"paramName": "cf",
|
||||
"paramLongName": "collectedfrom",
|
||||
"paramDescription": "the collectedfrom value for which we should apply the cleaning",
|
||||
"paramRequired": true
|
||||
},
|
||||
{
|
||||
"paramName": "hb",
|
||||
"paramLongName": "hostedBy",
|
||||
"paramDescription": "the set of datasources having the specified country in the graph searched for in the hostedby of the results",
|
||||
"paramRequired": true
|
||||
}
|
||||
]
|
||||
@ -0,0 +1,26 @@
|
||||
[
|
||||
{
|
||||
"paramName": "issm",
|
||||
"paramLongName": "isSparkSessionManaged",
|
||||
"paramDescription": "when true will stop SparkSession after job execution",
|
||||
"paramRequired": false
|
||||
},
|
||||
{
|
||||
"paramName": "in",
|
||||
"paramLongName": "inputPath",
|
||||
"paramDescription": "the path to the graph data dump to read",
|
||||
"paramRequired": true
|
||||
},
|
||||
{
|
||||
"paramName": "wp",
|
||||
"paramLongName": "workingPath",
|
||||
"paramDescription": "the path to store the output graph",
|
||||
"paramRequired": true
|
||||
},
|
||||
{
|
||||
"paramName": "c",
|
||||
"paramLongName": "country",
|
||||
"paramDescription": "the id of the context to be removed",
|
||||
"paramRequired": true
|
||||
}
|
||||
]
|
||||
@ -0,0 +1,150 @@
|
||||
|
||||
package eu.dnetlib.dhp.oa.graph.clean;
|
||||
|
||||
import java.io.IOException;
|
||||
import java.nio.file.Files;
|
||||
import java.nio.file.Path;
|
||||
|
||||
import org.apache.commons.io.FileUtils;
|
||||
import org.apache.spark.SparkConf;
|
||||
import org.apache.spark.api.java.JavaRDD;
|
||||
import org.apache.spark.api.java.JavaSparkContext;
|
||||
import org.apache.spark.api.java.function.MapFunction;
|
||||
import org.apache.spark.sql.Encoders;
|
||||
import org.apache.spark.sql.SparkSession;
|
||||
import org.junit.jupiter.api.AfterAll;
|
||||
import org.junit.jupiter.api.Assertions;
|
||||
import org.junit.jupiter.api.BeforeAll;
|
||||
import org.junit.jupiter.api.Test;
|
||||
import org.slf4j.Logger;
|
||||
import org.slf4j.LoggerFactory;
|
||||
|
||||
/**
|
||||
* @author miriam.baglioni
|
||||
* @Date 20/07/22
|
||||
*/
|
||||
import com.fasterxml.jackson.databind.ObjectMapper;
|
||||
|
||||
import eu.dnetlib.dhp.oa.graph.clean.country.CleanCountrySparkJob;
|
||||
import eu.dnetlib.dhp.schema.oaf.Publication;
|
||||
|
||||
public class CleanCountryTest {
|
||||
private static final ObjectMapper OBJECT_MAPPER = new ObjectMapper();
|
||||
|
||||
private static SparkSession spark;
|
||||
|
||||
private static Path workingDir;
|
||||
|
||||
private static final Logger log = LoggerFactory.getLogger(CleanContextTest.class);
|
||||
|
||||
@BeforeAll
|
||||
public static void beforeAll() throws IOException {
|
||||
workingDir = Files.createTempDirectory(CleanCountryTest.class.getSimpleName());
|
||||
log.info("using work dir {}", workingDir);
|
||||
|
||||
SparkConf conf = new SparkConf();
|
||||
conf.setAppName(CleanCountryTest.class.getSimpleName());
|
||||
|
||||
conf.setMaster("local[*]");
|
||||
conf.set("spark.driver.host", "localhost");
|
||||
conf.set("hive.metastore.local", "true");
|
||||
conf.set("spark.ui.enabled", "false");
|
||||
conf.set("spark.sql.warehouse.dir", workingDir.toString());
|
||||
conf.set("hive.metastore.warehouse.dir", workingDir.resolve("warehouse").toString());
|
||||
|
||||
spark = SparkSession
|
||||
.builder()
|
||||
.appName(CleanCountryTest.class.getSimpleName())
|
||||
.config(conf)
|
||||
.getOrCreate();
|
||||
}
|
||||
|
||||
@AfterAll
|
||||
public static void afterAll() throws IOException {
|
||||
FileUtils.deleteDirectory(workingDir.toFile());
|
||||
spark.stop();
|
||||
}
|
||||
|
||||
@Test
|
||||
public void testResultClean() throws Exception {
|
||||
final String sourcePath = getClass()
|
||||
.getResource("/eu/dnetlib/dhp/oa/graph/clean/publication_clean_country.json")
|
||||
.getPath();
|
||||
|
||||
spark
|
||||
.read()
|
||||
.textFile(sourcePath)
|
||||
.map(
|
||||
(MapFunction<String, Publication>) r -> OBJECT_MAPPER.readValue(r, Publication.class),
|
||||
Encoders.bean(Publication.class))
|
||||
.write()
|
||||
.json(workingDir.toString() + "/publication");
|
||||
|
||||
CleanCountrySparkJob.main(new String[] {
|
||||
"--isSparkSessionManaged", Boolean.FALSE.toString(),
|
||||
"--inputPath", workingDir.toString() + "/publication",
|
||||
"-graphTableClassName", Publication.class.getCanonicalName(),
|
||||
"-workingPath", workingDir.toString() + "/working",
|
||||
"-country", "NL",
|
||||
"-verifyParam", "10.17632",
|
||||
"-collectedfrom", "NARCIS",
|
||||
"-hostedBy", getClass()
|
||||
.getResource("/eu/dnetlib/dhp/oa/graph/clean/hostedBy")
|
||||
.getPath()
|
||||
});
|
||||
|
||||
final JavaSparkContext sc = JavaSparkContext.fromSparkContext(spark.sparkContext());
|
||||
JavaRDD<Publication> tmp = sc
|
||||
.textFile(workingDir.toString() + "/publication")
|
||||
.map(item -> OBJECT_MAPPER.readValue(item, Publication.class));
|
||||
|
||||
Assertions.assertEquals(8, tmp.count());
|
||||
|
||||
// original result with NL country and doi starting with Mendely prefix, but not collectedfrom NARCIS
|
||||
Assertions
|
||||
.assertEquals(
|
||||
1,
|
||||
tmp
|
||||
.filter(p -> p.getId().equals("50|DansKnawCris::0224aae28af558f21768dbc6439c7a95"))
|
||||
.collect()
|
||||
.get(0)
|
||||
.getCountry()
|
||||
.size());
|
||||
|
||||
// original result with NL country and pid not starting with Mendely prefix
|
||||
Assertions
|
||||
.assertEquals(
|
||||
1,
|
||||
tmp
|
||||
.filter(p -> p.getId().equals("50|DansKnawCris::20c414a3b1c742d5dd3851f1b67df2d9"))
|
||||
.collect()
|
||||
.get(0)
|
||||
.getCountry()
|
||||
.size());
|
||||
|
||||
// original result with NL country and doi starting with Mendely prefix and collectedfrom NARCIS but not
|
||||
// inserted with propagation
|
||||
Assertions
|
||||
.assertEquals(
|
||||
1,
|
||||
tmp
|
||||
.filter(p -> p.getId().equals("50|DansKnawCris::3c81248c335f0aa07e06817ece6fa6af"))
|
||||
.collect()
|
||||
.get(0)
|
||||
.getCountry()
|
||||
.size());
|
||||
|
||||
// original result with NL country and doi starting with Mendely prefix and collectedfrom NARCIS inserted with
|
||||
// propagation
|
||||
Assertions
|
||||
.assertEquals(
|
||||
0,
|
||||
tmp
|
||||
.filter(p -> p.getId().equals("50|DansKnawCris::3c81248c335f0aa07e06817ece6fa6ag"))
|
||||
.collect()
|
||||
.get(0)
|
||||
.getCountry()
|
||||
.size());
|
||||
}
|
||||
|
||||
}
|
||||
File diff suppressed because one or more lines are too long
Loading…
Reference in New Issue