D-Net
/
dnet-hadoop
14
5
Fork 11
Code Issues 14 Pull Requests 12 Projects 2 Releases 2 Wiki Activity

master #59

Closed
claudio.atzori wants to merge 3221 commits from master into stable_ids
pull from: master
merge into: D-Net:stable_ids
Conversation 0 Commits 3221 Files Changed 2141 +30 -9
4 changed files with 30 additions and 9 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files
Showing only changes of commit 3426451d3f - Show all commits

22
dhp-workflows/dhp-dedup-openaire/src/main/java/eu/dnetlib/dhp/oa/dedup/AbstractSparkAction.java
View File

@ -139,12 +139,26 @@ abstract class AbstractSparkAction implements Serializable {
protected boolean isOpenorgs(Relation rel) { protected boolean isOpenorgs(Relation rel) {
return Optional return Optional
.ofNullable(rel.getCollectedfrom()) .ofNullable(rel.getCollectedfrom())
.map( .map(c -> isCollectedFromOpenOrgs(c))
c -> c .orElse(false);
}
protected boolean isOpenorgsDedupRel(Relation rel) {
return isOpenorgs(rel) && isOpenOrgsDedupMergeRelation(rel);
}
private boolean isCollectedFromOpenOrgs(List<KeyValue> c) {
return c
.stream() .stream()
.filter(Objects::nonNull) .filter(Objects::nonNull)
.anyMatch(kv -> ModelConstants.OPENORGS_NAME.equals(kv.getValue()))) .anyMatch(kv -> ModelConstants.OPENORGS_NAME.equals(kv.getValue()));
.orElse(false); }
private boolean isOpenOrgsDedupMergeRelation(Relation rel) {
return ModelConstants.ORG_ORG_RELTYPE.equals(rel.getRelType()) &&
ModelConstants.DEDUP.equals(rel.getSubRelType())
&& (ModelConstants.IS_MERGED_IN.equals(rel.getRelClass()) ||
ModelConstants.MERGES.equals(rel.getRelClass()));
} }
protected static Boolean parseECField(Field<String> field) { protected static Boolean parseECField(Field<String> field) {

2
dhp-workflows/dhp-dedup-openaire/src/main/java/eu/dnetlib/dhp/oa/dedup/SparkCopyRelationsNoOpenorgs.java
View File

@ -61,7 +61,7 @@ public class SparkCopyRelationsNoOpenorgs extends AbstractSparkAction {
.textFile(relationPath) .textFile(relationPath)
.map(patchRelFn(), Encoders.bean(Relation.class)) .map(patchRelFn(), Encoders.bean(Relation.class))
.toJavaRDD() .toJavaRDD()
.filter(x -> !isOpenorgs(x)); .filter(x -> !isOpenorgsDedupRel(x));
if (log.isDebugEnabled()) { if (log.isDebugEnabled()) {
log.debug("Number of non-Openorgs relations collected: {}", simRels.count()); log.debug("Number of non-Openorgs relations collected: {}", simRels.count());

11
dhp-workflows/dhp-dedup-openaire/src/test/java/eu/dnetlib/dhp/oa/dedup/SparkOpenorgsProvisionTest.java
View File

@ -11,6 +11,8 @@ import java.io.IOException;
import java.io.Serializable; import java.io.Serializable;
import java.net.URISyntaxException; import java.net.URISyntaxException;
import java.nio.file.Paths; import java.nio.file.Paths;
import java.util.List;
import java.util.stream.Collectors;
import org.apache.commons.io.FileUtils; import org.apache.commons.io.FileUtils;
import org.apache.commons.io.IOUtils; import org.apache.commons.io.IOUtils;
@ -29,6 +31,8 @@ import org.mockito.Mock;
import org.mockito.Mockito; import org.mockito.Mockito;
import org.mockito.junit.jupiter.MockitoExtension; import org.mockito.junit.jupiter.MockitoExtension;
import com.fasterxml.jackson.databind.ObjectMapper;
import eu.dnetlib.dhp.application.ArgumentApplicationParser; import eu.dnetlib.dhp.application.ArgumentApplicationParser;
import eu.dnetlib.dhp.schema.oaf.Relation; import eu.dnetlib.dhp.schema.oaf.Relation;
import eu.dnetlib.enabling.is.lookup.rmi.ISLookUpException; import eu.dnetlib.enabling.is.lookup.rmi.ISLookUpException;
@ -226,9 +230,10 @@ public class SparkOpenorgsProvisionTest implements Serializable {
new SparkCopyRelationsNoOpenorgs(parser, spark).run(isLookUpService); new SparkCopyRelationsNoOpenorgs(parser, spark).run(isLookUpService);
long relations = jsc.textFile(testDedupGraphBasePath + "/relation").count(); final JavaRDD<String> rels = jsc.textFile(testDedupGraphBasePath + "/relation");
assertEquals(2382, rels.count());
assertEquals(2380, relations);
} }
@Test @Test
@ -250,7 +255,7 @@ public class SparkOpenorgsProvisionTest implements Serializable {
long relations = jsc.textFile(testDedupGraphBasePath + "/relation").count(); long relations = jsc.textFile(testDedupGraphBasePath + "/relation").count();
assertEquals(4894, relations); assertEquals(4896, relations);
// check deletedbyinference // check deletedbyinference
final Dataset<Relation> mergeRels = spark final Dataset<Relation> mergeRels = spark

2
dhp-workflows/dhp-dedup-openaire/src/test/resources/eu/dnetlib/dhp/dedup/openorgs/provision/relation/part-00000
View File

@ -2518,3 +2518,5 @@
{"subRelType": "dedup", "relClass": "isMergedIn", "dataInfo": {"provenanceaction": {"classid": "sysimport:crosswalk:entityregistry", "classname": "sysimport:crosswalk:entityregistry", "schemeid": "dnet:provenanceActions", "schemename": "dnet:provenanceActions"}, "deletedbyinference": false, "inferred": false, "inferenceprovenance": "", "invisible": false, "trust": "0.990"}, "target": "20|openorgs____::5c351d85f02db01ca291acd119f0bd78", "lastupdatetimestamp": 1617801137807, "relType": "organizationOrganization", "source": "20|opendoar____::37248e2f6987b18670dd2b8a51d6ef55", "validationDate": null, "collectedfrom": [{"dataInfo": null, "key": "10|openaire____::0362fcdb3076765d9c0041ad331553e8", "value": "OpenOrgs Database"}], "validated": false, "properties": []} {"subRelType": "dedup", "relClass": "isMergedIn", "dataInfo": {"provenanceaction": {"classid": "sysimport:crosswalk:entityregistry", "classname": "sysimport:crosswalk:entityregistry", "schemeid": "dnet:provenanceActions", "schemename": "dnet:provenanceActions"}, "deletedbyinference": false, "inferred": false, "inferenceprovenance": "", "invisible": false, "trust": "0.990"}, "target": "20|openorgs____::5c351d85f02db01ca291acd119f0bd78", "lastupdatetimestamp": 1617801137807, "relType": "organizationOrganization", "source": "20|opendoar____::37248e2f6987b18670dd2b8a51d6ef55", "validationDate": null, "collectedfrom": [{"dataInfo": null, "key": "10|openaire____::0362fcdb3076765d9c0041ad331553e8", "value": "OpenOrgs Database"}], "validated": false, "properties": []}
{"subRelType": "dedup", "relClass": "merges", "dataInfo": {"provenanceaction": {"classid": "sysimport:crosswalk:entityregistry", "classname": "sysimport:crosswalk:entityregistry", "schemeid": "dnet:provenanceActions", "schemename": "dnet:provenanceActions"}, "deletedbyinference": false, "inferred": false, "inferenceprovenance": "", "invisible": false, "trust": "0.990"}, "target": "20|corda_______::6acb33e6ea8c6fcdabc891c80d083c64", "lastupdatetimestamp": 1617801137807, "relType": "organizationOrganization", "source": "20|openorgs____::e38c1a27fcb0f0ab218828e4f5fc7be9", "validationDate": null, "collectedfrom": [{"dataInfo": null, "key": "10|openaire____::0362fcdb3076765d9c0041ad331553e8", "value": "OpenOrgs Database"}], "validated": false, "properties": []} {"subRelType": "dedup", "relClass": "merges", "dataInfo": {"provenanceaction": {"classid": "sysimport:crosswalk:entityregistry", "classname": "sysimport:crosswalk:entityregistry", "schemeid": "dnet:provenanceActions", "schemename": "dnet:provenanceActions"}, "deletedbyinference": false, "inferred": false, "inferenceprovenance": "", "invisible": false, "trust": "0.990"}, "target": "20|corda_______::6acb33e6ea8c6fcdabc891c80d083c64", "lastupdatetimestamp": 1617801137807, "relType": "organizationOrganization", "source": "20|openorgs____::e38c1a27fcb0f0ab218828e4f5fc7be9", "validationDate": null, "collectedfrom": [{"dataInfo": null, "key": "10|openaire____::0362fcdb3076765d9c0041ad331553e8", "value": "OpenOrgs Database"}], "validated": false, "properties": []}
{"subRelType": "dedup", "relClass": "isMergedIn", "dataInfo": {"provenanceaction": {"classid": "sysimport:crosswalk:entityregistry", "classname": "sysimport:crosswalk:entityregistry", "schemeid": "dnet:provenanceActions", "schemename": "dnet:provenanceActions"}, "deletedbyinference": false, "inferred": false, "inferenceprovenance": "", "invisible": false, "trust": "0.990"}, "target": "20|openorgs____::e38c1a27fcb0f0ab218828e4f5fc7be9", "lastupdatetimestamp": 1617801137807, "relType": "organizationOrganization", "source": "20|corda_______::6acb33e6ea8c6fcdabc891c80d083c64", "validationDate": null, "collectedfrom": [{"dataInfo": null, "key": "10|openaire____::0362fcdb3076765d9c0041ad331553e8", "value": "OpenOrgs Database"}], "validated": false, "properties": []} {"subRelType": "dedup", "relClass": "isMergedIn", "dataInfo": {"provenanceaction": {"classid": "sysimport:crosswalk:entityregistry", "classname": "sysimport:crosswalk:entityregistry", "schemeid": "dnet:provenanceActions", "schemename": "dnet:provenanceActions"}, "deletedbyinference": false, "inferred": false, "inferenceprovenance": "", "invisible": false, "trust": "0.990"}, "target": "20|openorgs____::e38c1a27fcb0f0ab218828e4f5fc7be9", "lastupdatetimestamp": 1617801137807, "relType": "organizationOrganization", "source": "20|corda_______::6acb33e6ea8c6fcdabc891c80d083c64", "validationDate": null, "collectedfrom": [{"dataInfo": null, "key": "10|openaire____::0362fcdb3076765d9c0041ad331553e8", "value": "OpenOrgs Database"}], "validated": false, "properties": []}
{"subRelType": "relationship", "relClass": "IsParentOf", "dataInfo": {"provenanceaction": {"classid": "sysimport:crosswalk:entityregistry", "classname": "sysimport:crosswalk:entityregistry", "schemeid": "dnet:provenanceActions", "schemename": "dnet:provenanceActions"}, "deletedbyinference": false, "inferred": false, "inferenceprovenance": "", "invisible": false, "trust": "0.990"}, "target": "20|openorgs____::e38c1a27fcb0f0ab218828e4f5fc7be9", "lastupdatetimestamp": 1617801137807, "relType": "organizationOrganization", "source": "20|corda_______::6acb33e6ea8c6fcdabc891c80d083c64", "validationDate": null, "collectedfrom": [{"dataInfo": null, "key": "10|openaire____::0362fcdb3076765d9c0041ad331553e8", "value": "OpenOrgs Database"}], "validated": false, "properties": []}
{"subRelType": "relationship", "relClass": "IsChildOf", "dataInfo": {"provenanceaction": {"classid": "sysimport:crosswalk:entityregistry", "classname": "sysimport:crosswalk:entityregistry", "schemeid": "dnet:provenanceActions", "schemename": "dnet:provenanceActions"}, "deletedbyinference": false, "inferred": false, "inferenceprovenance": "", "invisible": false, "trust": "0.990"}, "target": "20|corda_______::6acb33e6ea8c6fcdabc891c80d083c64", "lastupdatetimestamp": 1617801137807, "relType": "organizationOrganization", "source": "20|openorgs____::e38c1a27fcb0f0ab218828e4f5fc7be9", "validationDate": null, "collectedfrom": [{"dataInfo": null, "key": "10|openaire____::0362fcdb3076765d9c0041ad331553e8", "value": "OpenOrgs Database"}], "validated": false, "properties": []}