D-Net
/
dnet-hadoop
14
5
Fork 11
Code Issues 14 Pull Requests 13 Projects 2 Releases 2 Wiki Activity

[Enrichment: Propagation through parent-child relationships] Added counters, and changed constraint to verify if filtering out the relation (from classname = harvested to classid != propagation)

This commit is contained in:
Miriam Baglioni 2021-11-03 13:55:37 +01:00
parent 2aca6bfa0a
commit b9d124bb7c
4 changed files with 8 additions and 36 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

5
dhp-workflows/dhp-enrichment/src/main/java/eu/dnetlib/dhp/resulttoorganizationfromsemrel/PrepareInfo.java
View File

@ -6,9 +6,7 @@ import static eu.dnetlib.dhp.common.SparkSessionSupport.runWithSparkHiveSession;
import java.io.Serializable;
import java.util.*;
import java.util.stream.Collectors;
import org.apache.commons.collections.iterators.ArrayListIterator;
import org.apache.commons.io.IOUtils;
import org.apache.spark.SparkConf;
import org.apache.spark.api.java.function.*;
@ -17,15 +15,12 @@ import org.slf4j.Logger;
import org.slf4j.LoggerFactory;
import com.fasterxml.jackson.databind.ObjectMapper;
import com.sun.tools.internal.ws.processor.model.Model;
import eu.dnetlib.dhp.KeyValueSet;
import eu.dnetlib.dhp.PropagationConstant;
import eu.dnetlib.dhp.application.ArgumentApplicationParser;
import eu.dnetlib.dhp.resulttoorganizationfrominstrepo.SparkResultToOrganizationFromIstRepoJob;
import eu.dnetlib.dhp.schema.common.ModelConstants;
import eu.dnetlib.dhp.schema.oaf.Relation;
import net.sf.saxon.expr.instruct.ForEach;
import scala.Tuple2;
/**

31
dhp-workflows/dhp-enrichment/src/main/java/eu/dnetlib/dhp/resulttoorganizationfromsemrel/SparkResultToOrganizationFromSemRel.java
View File

@ -2,7 +2,7 @@
package eu.dnetlib.dhp.resulttoorganizationfromsemrel;
import static eu.dnetlib.dhp.PropagationConstant.*;
import static eu.dnetlib.dhp.common.Constants.*;
import static eu.dnetlib.dhp.common.SparkSessionSupport.runWithSparkHiveSession;
import java.io.Serializable;
@ -22,12 +22,13 @@ import org.slf4j.Logger;
import org.slf4j.LoggerFactory;
import eu.dnetlib.dhp.KeyValueSet;
import eu.dnetlib.dhp.PropagationConstant;
import eu.dnetlib.dhp.application.ArgumentApplicationParser;
import eu.dnetlib.dhp.resulttoorganizationfrominstrepo.SparkResultToOrganizationFromIstRepoJob;
import eu.dnetlib.dhp.schema.common.ModelConstants;
import eu.dnetlib.dhp.schema.oaf.Relation;
public class SparkResultToOrganizationFromSemRel implements Serializable {
private static final Logger log = LoggerFactory.getLogger(SparkResultToOrganizationFromSemRel.class);
private static final int MAX_ITERATION = 5;
@ -113,7 +114,7 @@ public class SparkResultToOrganizationFromSemRel implements Serializable {
String resultOrganizationPath, String graphPath, String workingPath, String outputPath,
PropagationCounter propagationCounter) {
int iteration = 0;
long leavesCount = 0;
long leavesCount;
do {
iteration++;
@ -199,30 +200,6 @@ public class SparkResultToOrganizationFromSemRel implements Serializable {
.json(outputPath);
}
// per ogni figlio nel file delle organizzazioni
// devo fare una roba iterativa che legge info da un file e le cambia via via
// passo 1: creo l'informazione iniale: organizzazioni che non hanno figli con almeno un padre
// ogni organizzazione punta alla lista di padri
// eseguo la propagazione dall'organizzazione figlio all'organizzazione padre
// ricerco nel dataset delle relazioni se l'organizzazione a cui ho propagato ha, a sua volta, dei padri
// e propago anche a quelli e cosi' via fino a che arrivo ad organizzazione senza padre
// organizationFile:
// f => {p1, p2, ..., pn}
// resultFile
// o => {r1, r2, ... rm}
// supponiamo che f => {r1, r2} e che nessuno dei padri abbia gia' l'associazione con questi result
// quindi
// p1 => {r1, r2}
// p2 => {r1, r2}
// pn => {r1, r2}
// mi serve un file con tutta la gerarchia per organizzazioni
// un file con le organizzazioni foglia da joinare con l'altro file
// un file con le associazioni organizzazione -> result forse meglio result -> organization
// eseguo gli step fino a che ho foglie nel set
// quando non ne ho piu' creo relazioni doppio verso per le nuove propagate che ho introdotto
}

6
dhp-workflows/dhp-enrichment/src/main/java/eu/dnetlib/dhp/resulttoorganizationfromsemrel/StepActions.java
View File

@ -131,11 +131,11 @@ public class StepActions implements Serializable {
if (relationList
.stream()
.filter(
rel -> rel
rel -> !rel
.getDataInfo()
.getProvenanceaction()
.getClassname()
.equals(ModelConstants.HARVESTED))
.getClassid()
.equals(PROPAGATION_RELATION_RESULT_ORGANIZATION_SEM_REL_CLASS_ID))
.count() > 0) {
return null;
}

2
dhp-workflows/dhp-enrichment/src/main/resources/eu/dnetlib/dhp/resulttoorganizationfromsemrel/oozie_app/workflow.xml
View File

@ -176,7 +176,7 @@
<arg>--graphPath</arg><arg>${sourcePath}</arg>
<arg>--outputPath</arg><arg>${outputPath}/relation</arg>
<arg>--leavesPath</arg><arg>${workingDir}/preparedInfo/leavesPath</arg>
<arg>--leavesPath</arg><arg>${workingDir}/preparedInfo/childParentPath</arg>
<arg>--childParentPath</arg><arg>${workingDir}/preparedInfo/childParentPath</arg>
<arg>--resultOrgPath</arg><arg>${workingDir}/preparedInfo/resultOrgPath</arg>
<arg>--hive_metastore_uris</arg><arg>${hive_metastore_uris}</arg>
<arg>--workingDir</arg><arg>${workingDir}/working</arg>