dimitris.pierrakos
/
dnet-hadoop
forked from D-Net/dnet-hadoop
1
0
Fork 0
Code Pull Requests Activity

refactoring

This commit is contained in:
Miriam Baglioni 2021-04-23 12:10:30 +02:00
parent 4ae6fba01d
commit 72e5aa3b42
1 changed files with 10 additions and 10 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

20
dhp-workflows/dhp-enrichment/src/main/java/eu/dnetlib/dhp/resulttoorganizationfrominstrepo/PrepareResultInstRepoAssociation.java
View File

@ -4,6 +4,11 @@ package eu.dnetlib.dhp.resulttoorganizationfrominstrepo;
import static eu.dnetlib.dhp.PropagationConstant.*; import static eu.dnetlib.dhp.PropagationConstant.*;
import static eu.dnetlib.dhp.common.SparkSessionSupport.runWithSparkHiveSession; import static eu.dnetlib.dhp.common.SparkSessionSupport.runWithSparkHiveSession;
import java.util.ArrayList;
import java.util.Arrays;
import java.util.List;
import java.util.Optional;
import org.apache.commons.io.IOUtils; import org.apache.commons.io.IOUtils;
import org.apache.hadoop.io.compress.GzipCodec; import org.apache.hadoop.io.compress.GzipCodec;
import org.apache.spark.SparkConf; import org.apache.spark.SparkConf;
@ -22,11 +27,6 @@ import eu.dnetlib.dhp.schema.oaf.Datasource;
import eu.dnetlib.dhp.schema.oaf.Organization; import eu.dnetlib.dhp.schema.oaf.Organization;
import eu.dnetlib.dhp.schema.oaf.Relation; import eu.dnetlib.dhp.schema.oaf.Relation;
import java.util.ArrayList;
import java.util.Arrays;
import java.util.List;
import java.util.Optional;
public class PrepareResultInstRepoAssociation { public class PrepareResultInstRepoAssociation {
private static final Logger log = LoggerFactory.getLogger(PrepareResultInstRepoAssociation.class); private static final Logger log = LoggerFactory.getLogger(PrepareResultInstRepoAssociation.class);
@ -56,9 +56,10 @@ public class PrepareResultInstRepoAssociation {
final String alreadyLinkedPath = parser.get("alreadyLinkedPath"); final String alreadyLinkedPath = parser.get("alreadyLinkedPath");
log.info("alreadyLinkedPath {}: ", alreadyLinkedPath); log.info("alreadyLinkedPath {}: ", alreadyLinkedPath);
List<String> blacklist = Optional.ofNullable(parser.get("blacklist")) List<String> blacklist = Optional
.map(v -> Arrays.asList(v.split(";"))) .ofNullable(parser.get("blacklist"))
.orElse(new ArrayList<>()); .map(v -> Arrays.asList(v.split(";")))
.orElse(new ArrayList<>());
SparkConf conf = new SparkConf(); SparkConf conf = new SparkConf();
conf.set("hive.metastore.uris", parser.get("hive_metastore_uris")); conf.set("hive.metastore.uris", parser.get("hive_metastore_uris"));
@ -91,14 +92,13 @@ public class PrepareResultInstRepoAssociation {
private static void prepareDatasourceOrganization( private static void prepareDatasourceOrganization(
SparkSession spark, String datasourceOrganizationPath, List<String> blacklist) { SparkSession spark, String datasourceOrganizationPath, List<String> blacklist) {
String blacklisted = ""; String blacklisted = "";
if(blacklist.size() > 0 ){ if (blacklist.size() > 0) {
blacklisted = " AND d.id != '" + blacklist.get(0) + "'"; blacklisted = " AND d.id != '" + blacklist.get(0) + "'";
for (int i = 1; i < blacklist.size(); i++) { for (int i = 1; i < blacklist.size(); i++) {
blacklisted += " AND d.id != '" + blacklist.get(i) + "'"; blacklisted += " AND d.id != '" + blacklist.get(i) + "'";
} }
} }
String query = "SELECT source datasourceId, target organizationId " String query = "SELECT source datasourceId, target organizationId "
+ "FROM ( SELECT id " + "FROM ( SELECT id "
+ "FROM datasource " + "FROM datasource "