towards EOSC datasource profiles #130

Merged
alessia.bardi merged 7 commits from datasource_model_eosc_beta into beta 2021-08-23 11:58:35 +02:00
2 changed files with 73 additions and 43 deletions
Showing only changes of commit 3e2a2d6e71 - Show all commits

View File

@ -27,14 +27,20 @@ import eu.dnetlib.dhp.oa.provision.model.ProvisionModelSupport;
import eu.dnetlib.dhp.oa.provision.model.RelatedEntity;
import eu.dnetlib.dhp.oa.provision.model.RelatedEntityWrapper;
import eu.dnetlib.dhp.schema.common.EntityType;
import eu.dnetlib.dhp.schema.oaf.*;
import eu.dnetlib.dhp.schema.oaf.Datasource;
import eu.dnetlib.dhp.schema.oaf.Field;
import eu.dnetlib.dhp.schema.oaf.OafEntity;
import eu.dnetlib.dhp.schema.oaf.Organization;
import eu.dnetlib.dhp.schema.oaf.Project;
import eu.dnetlib.dhp.schema.oaf.Relation;
import eu.dnetlib.dhp.schema.oaf.Result;
import eu.dnetlib.dhp.schema.oaf.StructuredProperty;
import eu.dnetlib.dhp.schema.oaf.utils.ModelHardLimits;
import scala.Tuple2;
/**
* CreateRelatedEntitiesJob: (phase 1): prepare tuples [relation - target entity] (R - T): for each entity type
* E_i map E_i as RelatedEntity T_i to simplify the model and extracting only the necessary information join
* (R.target = T_i.id) save the tuples (R_i, T_i)
* CreateRelatedEntitiesJob: (phase 1): prepare tuples [relation - target entity] (R - T): for each entity type E_i map E_i as RelatedEntity
* T_i to simplify the model and extracting only the necessary information join (R.target = T_i.id) save the tuples (R_i, T_i)
*/
public class CreateRelatedEntitiesJob_phase1 {
@ -42,68 +48,65 @@ public class CreateRelatedEntitiesJob_phase1 {
private static final ObjectMapper OBJECT_MAPPER = new ObjectMapper();
public static void main(String[] args) throws Exception {
public static void main(final String[] args) throws Exception {
String jsonConfiguration = IOUtils
final String jsonConfiguration = IOUtils
.toString(
PrepareRelationsJob.class
.getResourceAsStream(
"/eu/dnetlib/dhp/oa/provision/input_params_related_entities_pahase1.json"));
.getResourceAsStream("/eu/dnetlib/dhp/oa/provision/input_params_related_entities_pahase1.json"));
final ArgumentApplicationParser parser = new ArgumentApplicationParser(jsonConfiguration);
parser.parseArgument(args);
Boolean isSparkSessionManaged = Optional
final Boolean isSparkSessionManaged = Optional
.ofNullable(parser.get("isSparkSessionManaged"))
.map(Boolean::valueOf)
.orElse(Boolean.TRUE);
log.info("isSparkSessionManaged: {}", isSparkSessionManaged);
String inputRelationsPath = parser.get("inputRelationsPath");
final String inputRelationsPath = parser.get("inputRelationsPath");
log.info("inputRelationsPath: {}", inputRelationsPath);
String inputEntityPath = parser.get("inputEntityPath");
final String inputEntityPath = parser.get("inputEntityPath");
log.info("inputEntityPath: {}", inputEntityPath);
String outputPath = parser.get("outputPath");
final String outputPath = parser.get("outputPath");
log.info("outputPath: {}", outputPath);
String graphTableClassName = parser.get("graphTableClassName");
final String graphTableClassName = parser.get("graphTableClassName");
log.info("graphTableClassName: {}", graphTableClassName);
Class<? extends OafEntity> entityClazz = (Class<? extends OafEntity>) Class.forName(graphTableClassName);
final Class<? extends OafEntity> entityClazz = (Class<? extends OafEntity>) Class.forName(graphTableClassName);
SparkConf conf = new SparkConf();
final SparkConf conf = new SparkConf();
conf.set("spark.serializer", "org.apache.spark.serializer.KryoSerializer");
conf.registerKryoClasses(ProvisionModelSupport.getModelClasses());
runWithSparkSession(
conf,
isSparkSessionManaged,
spark -> {
removeOutputDir(spark, outputPath);
joinRelationEntity(spark, inputRelationsPath, inputEntityPath, entityClazz, outputPath);
});
runWithSparkSession(conf, isSparkSessionManaged, spark -> {
removeOutputDir(spark, outputPath);
joinRelationEntity(spark, inputRelationsPath, inputEntityPath, entityClazz, outputPath);
});
}
private static <E extends OafEntity> void joinRelationEntity(
SparkSession spark,
String inputRelationsPath,
String inputEntityPath,
Class<E> clazz,
String outputPath) {
final SparkSession spark,
final String inputRelationsPath,
final String inputEntityPath,
final Class<E> clazz,
final String outputPath) {
Dataset<Tuple2<String, Relation>> relsByTarget = readPathRelation(spark, inputRelationsPath)
final Dataset<Tuple2<String, Relation>> relsByTarget = readPathRelation(spark, inputRelationsPath)
.map(
(MapFunction<Relation, Tuple2<String, Relation>>) r -> new Tuple2<>(r.getTarget(),
r),
Encoders.tuple(Encoders.STRING(), Encoders.kryo(Relation.class)))
.cache();
Dataset<Tuple2<String, RelatedEntity>> entities = readPathEntity(spark, inputEntityPath, clazz)
final Dataset<Tuple2<String, RelatedEntity>> entities = readPathEntity(spark, inputEntityPath, clazz)
.filter("dataInfo.invisible == false")
.map(
(MapFunction<E, Tuple2<String, RelatedEntity>>) e -> new Tuple2<>(e.getId(), asRelatedEntity(e, clazz)),
Encoders.tuple(Encoders.STRING(), Encoders.kryo(RelatedEntity.class)))
Encoders
.tuple(Encoders.STRING(), Encoders.kryo(RelatedEntity.class)))
.cache();
relsByTarget
@ -118,7 +121,9 @@ public class CreateRelatedEntitiesJob_phase1 {
}
private static <E extends OafEntity> Dataset<E> readPathEntity(
SparkSession spark, String inputEntityPath, Class<E> entityClazz) {
final SparkSession spark,
final String inputEntityPath,
final Class<E> entityClazz) {
log.info("Reading Graph table from: {}", inputEntityPath);
return spark
@ -129,7 +134,7 @@ public class CreateRelatedEntitiesJob_phase1 {
Encoders.bean(entityClazz));
}
public static <E extends OafEntity> RelatedEntity asRelatedEntity(E entity, Class<E> clazz) {
public static <E extends OafEntity> RelatedEntity asRelatedEntity(final E entity, final Class<E> clazz) {
final RelatedEntity re = new RelatedEntity();
re.setId(entity.getId());
@ -143,7 +148,7 @@ public class CreateRelatedEntitiesJob_phase1 {
case dataset:
case otherresearchproduct:
case software:
Result result = (Result) entity;
final Result result = (Result) entity;
if (result.getTitle() != null && !result.getTitle().isEmpty()) {
final StructuredProperty title = result.getTitle().stream().findFirst().get();
@ -170,16 +175,17 @@ public class CreateRelatedEntitiesJob_phase1 {
break;
case datasource:
Datasource d = (Datasource) entity;
final Datasource d = (Datasource) entity;
re.setOfficialname(getValue(d.getOfficialname()));
re.setWebsiteurl(getValue(d.getWebsiteurl()));
re.setDatasourcetype(d.getDatasourcetype());
re.setDatasourcetypeui(d.getDatasourcetypeui());
re.setOpenairecompatibility(d.getOpenairecompatibility());
break;
case organization:
Organization o = (Organization) entity;
final Organization o = (Organization) entity;
re.setLegalname(getValue(o.getLegalname()));
re.setLegalshortname(getValue(o.getLegalshortname()));
@ -187,14 +193,14 @@ public class CreateRelatedEntitiesJob_phase1 {
re.setWebsiteurl(getValue(o.getWebsiteurl()));
break;
case project:
Project p = (Project) entity;
final Project p = (Project) entity;
re.setProjectTitle(getValue(p.getTitle()));
re.setCode(getValue(p.getCode()));
re.setAcronym(getValue(p.getAcronym()));
re.setContracttype(p.getContracttype());
List<Field<String>> f = p.getFundingtree();
final List<Field<String>> f = p.getFundingtree();
if (!f.isEmpty()) {
re.setFundingtree(f.stream().map(s -> s.getValue()).collect(Collectors.toList()));
}
@ -203,11 +209,11 @@ public class CreateRelatedEntitiesJob_phase1 {
return re;
}
private static String getValue(Field<String> field) {
private static String getValue(final Field<String> field) {
return getFieldValueWithDefault(field, "");
}
private static <T> T getFieldValueWithDefault(Field<T> f, T defaultValue) {
private static <T> T getFieldValueWithDefault(final Field<T> f, final T defaultValue) {
return Optional
.ofNullable(f)
.filter(Objects::nonNull)
@ -216,21 +222,21 @@ public class CreateRelatedEntitiesJob_phase1 {
}
/**
* Reads a Dataset of eu.dnetlib.dhp.oa.provision.model.SortableRelation objects from a newline delimited json text
* file,
* Reads a Dataset of eu.dnetlib.dhp.oa.provision.model.SortableRelation objects from a newline delimited json text file,
*
* @param spark
* @param relationPath
* @return the Dataset<SortableRelation> containing all the relationships
*/
private static Dataset<Relation> readPathRelation(
SparkSession spark, final String relationPath) {
final SparkSession spark,
final String relationPath) {
log.info("Reading relations from: {}", relationPath);
return spark.read().load(relationPath).as(Encoders.bean(Relation.class));
}
private static void removeOutputDir(SparkSession spark, String path) {
private static void removeOutputDir(final SparkSession spark, final String path) {
HdfsSupport.remove(path, spark.sparkContext().hadoopConfiguration());
}
}

View File

@ -735,6 +735,30 @@ public class XmlRecordFactory implements Serializable {
.collect(Collectors.toList()));
}
if (ds.getJurisdiction() != null) {
metadata.add(XmlSerializationUtils.mapQualifier("jurisdiction", ds.getJurisdiction()));
}
if (ds.getThematic() != null) {
metadata.add(XmlSerializationUtils.asXmlElement("thematic", ds.getThematic().toString()));
}
if (ds.getKnowledgegraph() != null) {
metadata
.add(XmlSerializationUtils.asXmlElement("knowledgegraph", ds.getKnowledgegraph().toString()));
}
if (ds.getContentpolicies() != null) {
metadata
.addAll(
ds
.getContentpolicies()
.stream()
.filter(Objects::nonNull)
.map(q -> XmlSerializationUtils.mapQualifier("contentpolicy", q))
.collect(Collectors.toList()));
}
break;
case organization:
final Organization o = (Organization) entity;