Merge remote-tracking branch 'origin/beta_to_master_may2024' into beta_to_master_may2024

dhp-shade-package/dependency-reduced-pom.xml

@@ -0,0 +1,113 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<project xmlns="http://maven.apache.org/POM/4.0.0" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://maven.apache.org/POM/4.0.0 http://maven.apache.org/maven-v4_0_0.xsd">
+  <parent>
+    <artifactId>dhp</artifactId>
+    <groupId>eu.dnetlib.dhp</groupId>
+    <version>1.2.5-SNAPSHOT</version>
+  </parent>
+  <modelVersion>4.0.0</modelVersion>
+  <artifactId>dhp-shade-package</artifactId>
+  <description>This module create a jar of all module dependencies</description>
+  <build>
+    <plugins>
+      <plugin>
+        <artifactId>maven-shade-plugin</artifactId>
+        <executions>
+          <execution>
+            <phase>package</phase>
+            <goals>
+              <goal>shade</goal>
+            </goals>
+            <configuration>
+              <transformers>
+                <transformer>
+                  <mainClass>eu.dnetlib.dhp.oa.dedup.SparkCreateSimRels</mainClass>
+                </transformer>
+                <transformer />
+                <transformer>
+                  <resource>META-INF/cxf/bus-extensions.txt</resource>
+                </transformer>
+              </transformers>
+              <filters>
+                <filter>
+                  <artifact>*:*</artifact>
+                  <excludes>
+                    <exclude>META-INF/maven/**</exclude>
+                    <exclude>META-INF/*.SF</exclude>
+                    <exclude>META-INF/*.DSA</exclude>
+                    <exclude>META-INF/*.RSA</exclude>
+                  </excludes>
+                </filter>
+              </filters>
+              <relocations>
+                <relocation>
+                  <pattern>com</pattern>
+                  <shadedPattern>repackaged.com.google.common</shadedPattern>
+                  <includes>
+                    <include>com.google.common.**</include>
+                  </includes>
+                </relocation>
+              </relocations>
+            </configuration>
+          </execution>
+        </executions>
+      </plugin>
+    </plugins>
+  </build>
+  <dependencies>
+    <dependency>
+      <groupId>org.projectlombok</groupId>
+      <artifactId>lombok</artifactId>
+      <version>1.18.28</version>
+      <scope>provided</scope>
+    </dependency>
+    <dependency>
+      <groupId>org.junit.jupiter</groupId>
+      <artifactId>junit-jupiter</artifactId>
+      <version>5.6.1</version>
+      <scope>test</scope>
+      <exclusions>
+        <exclusion>
+          <artifactId>junit-jupiter-api</artifactId>
+          <groupId>org.junit.jupiter</groupId>
+        </exclusion>
+        <exclusion>
+          <artifactId>junit-jupiter-params</artifactId>
+          <groupId>org.junit.jupiter</groupId>
+        </exclusion>
+        <exclusion>
+          <artifactId>junit-jupiter-engine</artifactId>
+          <groupId>org.junit.jupiter</groupId>
+        </exclusion>
+      </exclusions>
+    </dependency>
+    <dependency>
+      <groupId>org.mockito</groupId>
+      <artifactId>mockito-core</artifactId>
+      <version>3.3.3</version>
+      <scope>test</scope>
+      <exclusions>
+        <exclusion>
+          <artifactId>byte-buddy</artifactId>
+          <groupId>net.bytebuddy</groupId>
+        </exclusion>
+        <exclusion>
+          <artifactId>byte-buddy-agent</artifactId>
+          <groupId>net.bytebuddy</groupId>
+        </exclusion>
+      </exclusions>
+    </dependency>
+    <dependency>
+      <groupId>org.mockito</groupId>
+      <artifactId>mockito-junit-jupiter</artifactId>
+      <version>3.3.3</version>
+      <scope>test</scope>
+    </dependency>
+  </dependencies>
+  <distributionManagement>
+    <site>
+      <id>DHPSite</id>
+      <url>${dhp.site.stage.path}/dhp-common</url>
+    </site>
+  </distributionManagement>
+</project>

dhp-workflows/dhp-actionmanager/src/main/resources/eu/dnetlib/dhp/actionmanager/wf/dataset/oozie_app/workflow.xml

@@ -103,6 +103,7 @@
                 --executor-memory=${sparkExecutorMemory}
                 --executor-cores=${sparkExecutorCores}
                 --driver-memory=${sparkDriverMemory}
+                --conf spark.executor.memoryOverhead=${sparkExecutorMemory}
                 --conf spark.extraListeners=${spark2ExtraListeners}
                 --conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
                 --conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
@@ -156,6 +157,7 @@
                 --executor-memory=${sparkExecutorMemory}
                 --executor-cores=${sparkExecutorCores}
                 --driver-memory=${sparkDriverMemory}
+                --conf spark.executor.memoryOverhead=${sparkExecutorMemory}
                 --conf spark.extraListeners=${spark2ExtraListeners}
                 --conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
                 --conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}

dhp-workflows/dhp-actionmanager/src/main/resources/eu/dnetlib/dhp/actionmanager/wf/datasource/oozie_app/workflow.xml

@@ -95,6 +95,7 @@
                 --executor-memory=${sparkExecutorMemory}
                 --executor-cores=${sparkExecutorCores}
                 --driver-memory=${sparkDriverMemory}
+                --conf spark.executor.memoryOverhead=${sparkExecutorMemory}
                 --conf spark.extraListeners=${spark2ExtraListeners}
                 --conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
                 --conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}

dhp-workflows/dhp-actionmanager/src/main/resources/eu/dnetlib/dhp/actionmanager/wf/main/oozie_app/workflow.xml

@@ -125,6 +125,7 @@
                 --executor-memory=${sparkExecutorMemory}
                 --executor-cores=${sparkExecutorCores}
                 --driver-memory=${sparkDriverMemory}
+                --conf spark.executor.memoryOverhead=${sparkExecutorMemory}
                 --conf spark.extraListeners=${spark2ExtraListeners}
                 --conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
                 --conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}

dhp-workflows/dhp-actionmanager/src/main/resources/eu/dnetlib/dhp/actionmanager/wf/organization/oozie_app/workflow.xml

@@ -95,6 +95,7 @@
                 --executor-memory=${sparkExecutorMemory}
                 --executor-cores=${sparkExecutorCores}
                 --driver-memory=${sparkDriverMemory}
+                --conf spark.executor.memoryOverhead=${sparkExecutorMemory}
                 --conf spark.extraListeners=${spark2ExtraListeners}
                 --conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
                 --conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}

dhp-workflows/dhp-actionmanager/src/main/resources/eu/dnetlib/dhp/actionmanager/wf/otherresearchproduct/oozie_app/workflow.xml

@@ -103,6 +103,7 @@
                 --executor-memory=${sparkExecutorMemory}
                 --executor-cores=${sparkExecutorCores}
                 --driver-memory=${sparkDriverMemory}
+                --conf spark.executor.memoryOverhead=${sparkExecutorMemory}
                 --conf spark.extraListeners=${spark2ExtraListeners}
                 --conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
                 --conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
@@ -155,11 +156,12 @@
                 --executor-memory=${sparkExecutorMemory}
                 --executor-cores=${sparkExecutorCores}
                 --driver-memory=${sparkDriverMemory}
+                --conf spark.executor.memoryOverhead=${sparkExecutorMemory}
                 --conf spark.extraListeners=${spark2ExtraListeners}
                 --conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
                 --conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
                 --conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
-                --conf spark.sql.shuffle.partitions=2560
+                --conf spark.sql.shuffle.partitions=8000
             </spark-opts>
             <arg>--inputGraphTablePath</arg><arg>${workingDir}/otherresearchproduct</arg>
             <arg>--graphTableClassName</arg><arg>eu.dnetlib.dhp.schema.oaf.OtherResearchProduct</arg>

dhp-workflows/dhp-actionmanager/src/main/resources/eu/dnetlib/dhp/actionmanager/wf/project/oozie_app/workflow.xml

@@ -95,6 +95,7 @@
                 --executor-memory=${sparkExecutorMemory}
                 --executor-cores=${sparkExecutorCores}
                 --driver-memory=${sparkDriverMemory}
+                --conf spark.executor.memoryOverhead=${sparkExecutorMemory}
                 --conf spark.extraListeners=${spark2ExtraListeners}
                 --conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
                 --conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}

dhp-workflows/dhp-actionmanager/src/main/resources/eu/dnetlib/dhp/actionmanager/wf/publication/oozie_app/workflow.xml

@@ -103,11 +103,12 @@
                 --executor-memory=${sparkExecutorMemory}
                 --executor-cores=${sparkExecutorCores}
                 --driver-memory=${sparkDriverMemory}
+                --conf spark.executor.memoryOverhead=${sparkExecutorMemory}
                 --conf spark.extraListeners=${spark2ExtraListeners}
                 --conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
                 --conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
                 --conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
-                --conf spark.sql.shuffle.partitions=7000
+                --conf spark.sql.shuffle.partitions=15000
             </spark-opts>
             <arg>--inputGraphTablePath</arg><arg>${inputGraphRootPath}/publication</arg>
             <arg>--graphTableClassName</arg><arg>eu.dnetlib.dhp.schema.oaf.Publication</arg>
@@ -156,11 +157,12 @@
                 --executor-memory=${sparkExecutorMemory}
                 --executor-cores=${sparkExecutorCores}
                 --driver-memory=${sparkDriverMemory}
+                --conf spark.executor.memoryOverhead=${sparkExecutorMemory}
                 --conf spark.extraListeners=${spark2ExtraListeners}
                 --conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
                 --conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
                 --conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
-                --conf spark.sql.shuffle.partitions=7000
+                --conf spark.sql.shuffle.partitions=15000
             </spark-opts>
             <arg>--inputGraphTablePath</arg><arg>${workingDir}/publication</arg>
             <arg>--graphTableClassName</arg><arg>eu.dnetlib.dhp.schema.oaf.Publication</arg>

dhp-workflows/dhp-actionmanager/src/main/resources/eu/dnetlib/dhp/actionmanager/wf/relation/oozie_app/workflow.xml

@@ -95,11 +95,12 @@
                 --executor-memory=${sparkExecutorMemory}
                 --executor-cores=${sparkExecutorCores}
                 --driver-memory=${sparkDriverMemory}
+                --conf spark.executor.memoryOverhead=${sparkExecutorMemory}
                 --conf spark.extraListeners=${spark2ExtraListeners}
                 --conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
                 --conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
                 --conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
-                --conf spark.sql.shuffle.partitions=10000
+                --conf spark.sql.shuffle.partitions=15000
             </spark-opts>
             <arg>--inputGraphTablePath</arg><arg>${inputGraphRootPath}/relation</arg>
             <arg>--graphTableClassName</arg><arg>eu.dnetlib.dhp.schema.oaf.Relation</arg>

dhp-workflows/dhp-actionmanager/src/main/resources/eu/dnetlib/dhp/actionmanager/wf/software/oozie_app/workflow.xml

@@ -103,6 +103,7 @@
                 --executor-memory=${sparkExecutorMemory}
                 --executor-cores=${sparkExecutorCores}
                 --driver-memory=${sparkDriverMemory}
+                --conf spark.executor.memoryOverhead=${sparkExecutorMemory}
                 --conf spark.extraListeners=${spark2ExtraListeners}
                 --conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
                 --conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
@@ -155,11 +156,12 @@
                 --executor-memory=${sparkExecutorMemory}
                 --executor-cores=${sparkExecutorCores}
                 --driver-memory=${sparkDriverMemory}
+                --conf spark.executor.memoryOverhead=${sparkExecutorMemory}
                 --conf spark.extraListeners=${spark2ExtraListeners}
                 --conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
                 --conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
                 --conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
-                --conf spark.sql.shuffle.partitions=2560
+                --conf spark.sql.shuffle.partitions=4000
             </spark-opts>
             <arg>--inputGraphTablePath</arg><arg>${workingDir}/software</arg>
             <arg>--graphTableClassName</arg><arg>eu.dnetlib.dhp.schema.oaf.Software</arg>

dhp-workflows/dhp-graph-mapper/src/main/scala/eu/dnetlib/dhp/oa/graph/hostedbymap/SparkApplyHostedByMapToResult.scala

@@ -27,10 +27,12 @@ object SparkApplyHostedByMapToResult {
             val inst: Instance = i.head
             patchInstance(p, ei, inst)
 
-          } else if (i.size == 2) {
+          } else {
-            if (i.map(ii => ii.getCollectedfrom.getValue).contains("UnpayWall")) {
+            val cf = i.map(ii => ii.getCollectedfrom.getValue)
-              val inst: Instance = i.filter(ii => "Crossref".equals(ii.getCollectedfrom.getValue)).head
+            if (cf.contains("Crossref")) {
-              patchInstance(p, ei, inst)
+              i.foreach(ii => {
+                patchInstance(p, ei, ii)
+              })
             }
           }
         }

dhp-workflows/dhp-graph-provision/src/main/java/eu/dnetlib/dhp/oa/oaipmh/IrishOaiExporterJob.java

@@ -31,7 +31,6 @@ import org.slf4j.Logger;
 import org.slf4j.LoggerFactory;
 
 import eu.dnetlib.dhp.application.ArgumentApplicationParser;
-
 import eu.dnetlib.dhp.oa.provision.model.SerializableSolrInputDocument;
 import eu.dnetlib.dhp.oa.provision.model.TupleWrapper;
 
@@ -48,7 +47,7 @@ public class IrishOaiExporterJob {
 		final ArgumentApplicationParser parser = new ArgumentApplicationParser(
 			IOUtils
 				.toString(
-						IrishOaiExporterJob.class
+					IrishOaiExporterJob.class
 						.getResourceAsStream("/eu/dnetlib/dhp/oa/oaipmh/input_params_irish_oai_exporter.json")));
 		parser.parseArgument(args);
 

dhp-workflows/dhp-graph-provision/src/main/java/eu/dnetlib/dhp/oa/provision/utils/XmlSerializationUtils.java

@@ -176,13 +176,13 @@ public class XmlSerializationUtils {
 		StringBuilder sb = new StringBuilder();
 		for (KeyValue kv : measure.getUnit()) {
 			sb
-					.append("<")
+				.append("<")
-					.append(name)
+				.append(name)
-					.append(" ")
+				.append(" ")
-					.append(attr("id", measure.getId()))
+				.append(attr("id", measure.getId()))
-					.append(attr("count", kv.getValue()))
+				.append(attr("count", kv.getValue()))
-					.append(attr("datasource", kv.getKey()))
+				.append(attr("datasource", kv.getKey()))
-					.append(" />");
+				.append(" />");
 		}
 		return sb.toString();
 	}

dhp-workflows/dhp-graph-provision/src/test/java/eu/dnetlib/dhp/oa/provision/XmlRecordFactoryTest.java

@@ -43,7 +43,6 @@ public class XmlRecordFactoryTest {
 		final Publication p = OBJECT_MAPPER
 			.readValue(IOUtils.toString(getClass().getResourceAsStream("publication.json")), Publication.class);
 
-
 		final String xml = xmlRecordFactory.build(new JoinedEntity(p));
 
 		assertNotNull(xml);