miriam.baglioni
/
dnet-hadoop
forked from D-Net/dnet-hadoop
2
0
Fork 0
Code Issues 2 Pull Requests Releases Wiki Activity

start to implement the business logic

This commit is contained in:
Miriam Baglioni 2020-02-19 17:59:24 +01:00
parent 5f63ab1416
commit d0279af630
1 changed files with 3 additions and 7 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

10
dhp-workflows/dhp-propagation/src/main/java/eu/dnetlib/dhp/communitytoresultthroughsemrel/SparkResultToCommunityThroughSemRelJob.java
View File

@ -1,6 +1,7 @@
package eu.dnetlib.dhp.communitytoresultthroughsemrel;
import com.fasterxml.jackson.databind.ObjectMapper;
import eu.dnetlib.dhp.QueryInformationSystem;
import eu.dnetlib.dhp.TypedRow;
import eu.dnetlib.dhp.application.ArgumentApplicationParser;
import eu.dnetlib.dhp.resulttoorganizationfrominstrepo.SparkResultToOrganizationFromIstRepoJob;
@ -39,7 +40,7 @@ public class SparkResultToCommunityThroughSemRelJob {
final String outputPath = "/tmp/provision/propagation/communitytoresultthroughsemrel";
final List<String> allowedsemrel = Arrays.asList(parser.get("allowedsemrel").split(";"));
final List<String> communityIdList = Arrays.asList(parser.get("communityidlist").split(";"));
final List<String> communityIdList = QueryInformationSystem.getCommunityList(parser.get("isLookupUrl"));
File directory = new File(outputPath);
@ -47,12 +48,7 @@ public class SparkResultToCommunityThroughSemRelJob {
directory.mkdirs();
}
/*
//get the institutional repositories
JavaPairRDD<String, TypedRow> datasources = sc.sequenceFile(inputPath + "/datasource", Text.class, Text.class)
.map(item -> new ObjectMapper().readValue(item._2().toString(), Datasource.class))
.filter(ds -> INSTITUTIONAL_REPO_TYPE.equals(ds.getDatasourcetype().getClassid()))
.map(ds -> new TypedRow().setSourceId(ds.getId()))
.mapToPair(toPair());
JavaPairRDD<String, TypedRow> rel_datasource_organization = sc.sequenceFile(inputPath + "/relation", Text.class, Text.class)