alessia.bardi
/
dnet-hadoop
forked from D-Net/dnet-hadoop
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

[Graph Dump] changed to mirror the changes in the model

This commit is contained in:
Miriam Baglioni 2021-11-19 15:56:25 +01:00
parent 9fae872181
commit 0506fa2654
2 changed files with 5 additions and 12 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

10
dhp-workflows/dhp-graph-mapper/src/main/java/eu/dnetlib/dhp/oa/graph/dump/ResultMapper.java
View File

@ -12,14 +12,12 @@ import eu.dnetlib.dhp.schema.common.ModelConstants;
import eu.dnetlib.dhp.schema.dump.oaf.*; import eu.dnetlib.dhp.schema.dump.oaf.*;
import eu.dnetlib.dhp.schema.dump.oaf.AccessRight; import eu.dnetlib.dhp.schema.dump.oaf.AccessRight;
import eu.dnetlib.dhp.schema.dump.oaf.Author; import eu.dnetlib.dhp.schema.dump.oaf.Author;
import eu.dnetlib.dhp.schema.dump.oaf.Country;
import eu.dnetlib.dhp.schema.dump.oaf.GeoLocation; import eu.dnetlib.dhp.schema.dump.oaf.GeoLocation;
import eu.dnetlib.dhp.schema.dump.oaf.Instance; import eu.dnetlib.dhp.schema.dump.oaf.Instance;
import eu.dnetlib.dhp.schema.dump.oaf.KeyValue;
import eu.dnetlib.dhp.schema.dump.oaf.Measure; import eu.dnetlib.dhp.schema.dump.oaf.Measure;
import eu.dnetlib.dhp.schema.dump.oaf.OpenAccessRoute; import eu.dnetlib.dhp.schema.dump.oaf.OpenAccessRoute;
import eu.dnetlib.dhp.schema.dump.oaf.Qualifier;
import eu.dnetlib.dhp.schema.dump.oaf.Result; import eu.dnetlib.dhp.schema.dump.oaf.Result;
import eu.dnetlib.dhp.schema.dump.oaf.community.CfHbKeyValue;
import eu.dnetlib.dhp.schema.dump.oaf.community.CommunityInstance; import eu.dnetlib.dhp.schema.dump.oaf.community.CommunityInstance;
import eu.dnetlib.dhp.schema.dump.oaf.community.CommunityResult; import eu.dnetlib.dhp.schema.dump.oaf.community.CommunityResult;
import eu.dnetlib.dhp.schema.dump.oaf.community.Context; import eu.dnetlib.dhp.schema.dump.oaf.community.Context;
@ -231,7 +229,7 @@ public class ResultMapper implements Serializable {
input input
.getCollectedfrom() .getCollectedfrom()
.stream() .stream()
.map(cf -> KeyValue.newInstance(cf.getKey(), cf.getValue())) .map(cf -> CfHbKeyValue.newInstance(cf.getKey(), cf.getValue()))
.collect(Collectors.toList())); .collect(Collectors.toList()));
Set<String> communities = communityMap.keySet(); Set<String> communities = communityMap.keySet();
@ -424,12 +422,12 @@ public class ResultMapper implements Serializable {
instance instance
.setCollectedfrom( .setCollectedfrom(
KeyValue CfHbKeyValue
.newInstance(i.getCollectedfrom().getKey(), i.getCollectedfrom().getValue())); .newInstance(i.getCollectedfrom().getKey(), i.getCollectedfrom().getValue()));
instance instance
.setHostedby( .setHostedby(
KeyValue.newInstance(i.getHostedby().getKey(), i.getHostedby().getValue())); CfHbKeyValue.newInstance(i.getHostedby().getKey(), i.getHostedby().getValue()));
return instance; return instance;

7
dhp-workflows/dhp-graph-mapper/src/main/java/eu/dnetlib/dhp/oa/graph/dump/complete/DumpGraphEntities.java
View File

@ -1,7 +1,6 @@
package eu.dnetlib.dhp.oa.graph.dump.complete; package eu.dnetlib.dhp.oa.graph.dump.complete;
import static com.jayway.jsonpath.Filter.filter;
import static eu.dnetlib.dhp.common.SparkSessionSupport.runWithSparkSession; import static eu.dnetlib.dhp.common.SparkSessionSupport.runWithSparkSession;
import java.io.Serializable; import java.io.Serializable;
@ -11,9 +10,7 @@ import java.util.stream.Collectors;
import org.apache.spark.SparkConf; import org.apache.spark.SparkConf;
import org.apache.spark.api.java.function.FilterFunction; import org.apache.spark.api.java.function.FilterFunction;
import org.apache.spark.api.java.function.ForeachFunction;
import org.apache.spark.api.java.function.MapFunction; 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.Encoders;
import org.apache.spark.sql.SaveMode; import org.apache.spark.sql.SaveMode;
import org.apache.spark.sql.SparkSession; import org.apache.spark.sql.SparkSession;
@ -22,8 +19,6 @@ import org.dom4j.DocumentException;
import org.dom4j.Node; import org.dom4j.Node;
import org.dom4j.io.SAXReader; import org.dom4j.io.SAXReader;
import com.fasterxml.jackson.databind.ObjectMapper;
import eu.dnetlib.dhp.oa.graph.dump.DumpProducts; import eu.dnetlib.dhp.oa.graph.dump.DumpProducts;
import eu.dnetlib.dhp.oa.graph.dump.Utils; import eu.dnetlib.dhp.oa.graph.dump.Utils;
import eu.dnetlib.dhp.schema.common.ModelSupport; import eu.dnetlib.dhp.schema.common.ModelSupport;
@ -133,7 +128,7 @@ public class DumpGraphEntities implements Serializable {
.ifPresent( .ifPresent(
pids -> pids pids -> pids
.stream() .stream()
.map(p -> ControlledField.newInstance(p.getQualifier().getClassid(), p.getValue())) .map(p -> DatasourcePid.newInstance(p.getQualifier().getClassid(), p.getValue()))
.collect(Collectors.toList())); .collect(Collectors.toList()));
Optional Optional