lsmyrnaios
/
dnet-hadoop
forked from D-Net/dnet-hadoop
1
0
Fork 0
Code Pull Requests Activity

merging project to result propagation logic to master

This commit is contained in:
Miriam Baglioni 2020-02-19 18:04:59 +01:00
parent e8af7a6b64 8aa3b4d7c0
commit b2bdc9b99b
5 changed files with 225 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

11
dhp-workflows/dhp-propagation/src/main/java/eu/dnetlib/dhp/PropagationConstant.java
View File

@ -23,6 +23,9 @@ public class PropagationConstant {
public final static String PROPAGATION_COUNTRY_INSTREPO_CLASS_NAME = "Propagation of country to result collected from datasources of type institutional repositories"; public final static String PROPAGATION_COUNTRY_INSTREPO_CLASS_NAME = "Propagation of country to result collected from datasources of type institutional repositories";
public final static String PROPAGATION_RELATION_RESULT_ORGANIZATION_INST_REPO_CLASS_ID = "result:organization:instrepo"; public final static String PROPAGATION_RELATION_RESULT_ORGANIZATION_INST_REPO_CLASS_ID = "result:organization:instrepo";
public final static String PROPAGATION_RELATION_RESULT_ORGANIZATION_INST_REPO_CLASS_NAME = "Propagation of affiliation to result collected from datasources of type institutional repository"; public final static String PROPAGATION_RELATION_RESULT_ORGANIZATION_INST_REPO_CLASS_NAME = "Propagation of affiliation to result collected from datasources of type institutional repository";
public final static String PROPAGATION_RELATION_RESULT_PROJECT_SEM_REL_CLASS_ID = "result:project:semrel";
public final static String PROPAGATION_RELATION_RESULT_PROJECT_SEM_REL_CLASS_NAME = "Propagation of result to project through semantic relation";
public final static String RELATION_DATASOURCEORGANIZATION_REL_TYPE = "datasourceOrganization"; public final static String RELATION_DATASOURCEORGANIZATION_REL_TYPE = "datasourceOrganization";
public final static String RELATION_DATASOURCEORGANIZATION_SUBREL_TYPE = "provision"; public final static String RELATION_DATASOURCEORGANIZATION_SUBREL_TYPE = "provision";
@ -34,6 +37,14 @@ public class PropagationConstant {
public final static String RELATION_ORGANIZATION_RESULT_REL_CLASS = "isAuthorInstitutionOf"; public final static String RELATION_ORGANIZATION_RESULT_REL_CLASS = "isAuthorInstitutionOf";
public final static String RELATION_RESULT_ORGANIZATION_REL_CLASS = "hasAuthorInstitution"; public final static String RELATION_RESULT_ORGANIZATION_REL_CLASS = "hasAuthorInstitution";
public static final String RELATION_RESULTRESULT_REL_TYPE = "resultResult";
public static final String RELATION_RESULTRESULT_SUBREL_TYPE = "supplement";
public static final String RELATION_RESULTPROJECT_REL_TYPE = "resultProject";
public static final String RELATION_RESULTPROJECT_SUBREL_TYPE = "outcome";
public static final String RELATION_RESULT_PROJECT_REL_CLASS = "isProducedBy";
public static final String RELATION_PROJECT_RESULT_REL_CLASS = "produces";
public static Country getCountry(String country){ public static Country getCountry(String country){
Country nc = new Country(); Country nc = new Country();

120
dhp-workflows/dhp-propagation/src/main/java/eu/dnetlib/dhp/projecttoresult/SparkResultToProjectThroughSemRelJob.java Normal file
View File

@ -0,0 +1,120 @@
package eu.dnetlib.dhp.projecttoresult;
import com.fasterxml.jackson.databind.ObjectMapper;
import eu.dnetlib.dhp.TypedRow;
import eu.dnetlib.dhp.application.ArgumentApplicationParser;
import eu.dnetlib.dhp.schema.oaf.*;
import org.apache.commons.io.IOUtils;
import org.apache.hadoop.io.Text;
import org.apache.spark.api.java.JavaPairRDD;
import org.apache.spark.api.java.JavaRDD;
import org.apache.spark.api.java.JavaSparkContext;
import org.apache.spark.sql.SparkSession;
import scala.Tuple2;
import java.io.File;
import java.util.ArrayList;
import java.util.Arrays;
import java.util.HashSet;
import java.util.List;
import static eu.dnetlib.dhp.PropagationConstant.*;
import static eu.dnetlib.dhp.PropagationConstant.toPair;
public class SparkResultToProjectThroughSemRelJob {
public static void main(String[] args) throws Exception {
final ArgumentApplicationParser parser = new ArgumentApplicationParser(IOUtils.toString(SparkResultToProjectThroughSemRelJob.class.getResourceAsStream("/eu/dnetlib/dhp/projecttoresult/input_projecttoresult_parameters.json")));
parser.parseArgument(args);
final SparkSession spark = SparkSession
.builder()
.appName(SparkResultToProjectThroughSemRelJob.class.getSimpleName())
.master(parser.get("master"))
.enableHiveSupport()
.getOrCreate();
final JavaSparkContext sc = new JavaSparkContext(spark.sparkContext());
final String inputPath = parser.get("sourcePath");
final String outputPath = "/tmp/provision/propagation/projecttoresult";
final List<String> allowedsemrel = Arrays.asList(parser.get("allowedsemrels").split(";"));
File directory = new File(outputPath);
if (!directory.exists()) {
directory.mkdirs();
}
JavaRDD<Relation> relations = sc.sequenceFile(inputPath + "/relation", Text.class, Text.class)
.map(item -> new ObjectMapper().readValue(item._2().toString(), Relation.class)).cache();
JavaPairRDD<String, TypedRow> result_result = relations
.filter(r -> !r.getDataInfo().getDeletedbyinference())
.filter(r -> allowedsemrel.contains(r.getRelClass()) && RELATION_RESULTRESULT_REL_TYPE.equals(r.getRelType()))
.map(r -> new TypedRow().setSourceId(r.getSource()).setTargetId(r.getTarget()))
.mapToPair(toPair());
JavaPairRDD<String, TypedRow> result_project = relations
.filter(r -> !r.getDataInfo().getDeletedbyinference())
.filter(r -> RELATION_RESULT_PROJECT_REL_CLASS.equals(r.getRelClass()) && RELATION_RESULTPROJECT_REL_TYPE.equals(r.getRelType()))
.map(r -> new TypedRow().setSourceId(r.getSource()).setTargetId(r.getTarget()))
.mapToPair(toPair());
//relationships from project to result. One pair for each relationship for results having allowed semantics relation with another result
JavaPairRDD<String, TypedRow> project_result = result_project.join(result_result)
.map(c -> {
String projectId = c._2()._1().getTargetId();
String resultId = c._2()._2().getTargetId();
return new TypedRow().setSourceId(projectId).setTargetId(resultId);
})
.mapToPair(toPair());
//relationships from project to result. One Pair for each project => project id list of results related to the project
JavaPairRDD<String, TypedRow> project_results = relations
.filter(r -> !r.getDataInfo().getDeletedbyinference())
.filter(r -> RELATION_PROJECT_RESULT_REL_CLASS.equals(r.getRelClass()) && RELATION_RESULTPROJECT_REL_TYPE.equals(r.getRelType()))
.map(r -> new TypedRow().setSourceId(r.getSource()).setTargetId(r.getTarget()))
.mapToPair(toPair())
.reduceByKey((a, b) -> {
if (a == null) {
return b;
}
if (b == null) {
return a;
}
a.addAll(b.getAccumulator());
return a;
});
JavaRDD<Relation> newRels = project_result.join(project_results)
.flatMap(c -> {
String resId = c._2()._1().getTargetId();
if (c._2()._2().getAccumulator().contains(resId)) {
return null;
}
String progId = c._2()._1().getSourceId();
List<Relation> rels = new ArrayList();
rels.add(getRelation(progId, resId, RELATION_PROJECT_RESULT_REL_CLASS,
RELATION_RESULTPROJECT_REL_TYPE, RELATION_RESULTPROJECT_SUBREL_TYPE, PROPAGATION_DATA_INFO_TYPE,
PROPAGATION_RELATION_RESULT_PROJECT_SEM_REL_CLASS_ID, PROPAGATION_RELATION_RESULT_PROJECT_SEM_REL_CLASS_NAME));
rels.add(getRelation(resId, progId, RELATION_RESULT_PROJECT_REL_CLASS,
RELATION_RESULTPROJECT_REL_TYPE, RELATION_RESULTPROJECT_SUBREL_TYPE, PROPAGATION_DATA_INFO_TYPE,
PROPAGATION_RELATION_RESULT_PROJECT_SEM_REL_CLASS_ID, PROPAGATION_RELATION_RESULT_PROJECT_SEM_REL_CLASS_NAME));
return rels.iterator();
})
.cache();
newRels.map(p -> new ObjectMapper().writeValueAsString(p))
.saveAsTextFile(outputPath + "/relation_new");
newRels.union(relations).map(p -> new ObjectMapper().writeValueAsString(p))
.saveAsTextFile(outputPath + "/relation");
}
}

21
dhp-workflows/dhp-propagation/src/main/resources/eu/dnetlib/dhp/projecttoresult/input_projecttoresult_parameters.json Normal file
View File

@ -0,0 +1,21 @@
[
{
"paramName":"mt",
"paramLongName":"master",
"paramDescription": "should be local or yarn",
"paramRequired": true
},
{
"paramName":"s",
"paramLongName":"sourcePath",
"paramDescription": "the path of the sequencial file to read",
"paramRequired": true
},
{
"paramName":"asr",
"paramLongName":"allowedsemrels",
"paramDescription": "the types of the allowed datasources. Split by ;",
"paramRequired": true
}
]

18
dhp-workflows/dhp-propagation/src/main/resources/eu/dnetlib/dhp/projecttoresult/oozie_app/config-default.xml Normal file
View File

@ -0,0 +1,18 @@
<configuration>
<property>
<name>jobTracker</name>
<value>yarnRM</value>
</property>
<property>
<name>nameNode</name>
<value>hdfs://nameservice1</value>
</property>
<property>
<name>oozie.use.system.libpath</name>
<value>true</value>
</property>
<property>
<name>oozie.action.sharelib.for.spark</name>
<value>spark2</value>
</property>
</configuration>

55
dhp-workflows/dhp-propagation/src/main/resources/eu/dnetlib/dhp/projecttoresult/oozie_app/workflow.xml Normal file
View File

@ -0,0 +1,55 @@
<workflow-app name="project_to_result_propagation" xmlns="uri:oozie:workflow:0.5">
<parameters>
<property>
<name>sourcePath</name>
<description>the source path</description>
</property>
<property>
<name>allowedsemrels</name>
<description>the allowed semantics </description>
</property>
<property>
<name>sparkDriverMemory</name>
<description>memory for driver process</description>
</property>
<property>
<name>sparkExecutorMemory</name>
<description>memory for individual executor</description>
</property>
<property>
<name>sparkExecutorCores</name>
<description>number of cores used by single executor</description>
</property>
</parameters>
<start to="ProjectPropagation"/>
<kill name="Kill">
<message>Action failed, error message[${wf:errorMessage(wf:lastErrorNode())}]</message>
</kill>
<action name="ProjectPropagation">
<spark xmlns="uri:oozie:spark-action:0.2">
<job-tracker>${jobTracker}</job-tracker>
<name-node>${nameNode}</name-node>
<master>yarn-cluster</master>
<mode>cluster</mode>
<name>ProjectToResultPropagation</name>
<class>eu.dnetlib.dhp.projecttoresult.SparkResultToProjectThroughSemRelJob</class>
<jar>dhp-propagation-${projectVersion}.jar</jar>
<spark-opts>--executor-memory ${sparkExecutorMemory}
--executor-cores ${sparkExecutorCores}
--driver-memory=${sparkDriverMemory}
--conf spark.extraListeners="com.cloudera.spark.lineage.NavigatorAppListener"
--conf spark.sql.queryExecutionListeners="com.cloudera.spark.lineage.NavigatorQueryListener"
</spark-opts>
<arg>-mt</arg> <arg>yarn-cluster</arg>
<arg>--sourcePath</arg><arg>${sourcePath}</arg>
<arg>--allowedsemrels</arg><arg>${allowedsemrels}</arg>
</spark>
<ok to="End"/>
<error to="Kill"/>
</action>
<end name="End"/>
</workflow-app>