migrate configuration to ocean, fix publication dataset creation

This commit is contained in:
Enrico Ottonello 2020-07-28 15:23:52 +02:00
parent a6acb37689
commit c82b15b5f4
4 changed files with 68 additions and 50 deletions

View File

@ -91,19 +91,18 @@ public class SparkGenEnrichedOrcidWorks {
Encoders.tuple(Encoders.STRING(), Encoders.STRING()))
.filter(Objects::nonNull)
.toJavaRDD();
logger.info("Works enriched data created: " + enrichedWorksRDD.count());
enrichedWorksRDD.saveAsTextFile(workingPath + outputEnrichedWorksPath);
logger.info("Works enriched data saved");
JavaRDD<Tuple2<String, Publication>> oafPublicationRDD = enrichedWorksRDD.map(e -> {
JavaRDD<Publication> oafPublicationRDD = enrichedWorksRDD.map(e -> {
JsonElement j = new JsonParser().parse(e._2());
return new Tuple2<>(e._1(), (Publication) PublicationToOaf
.generatePublicationActionsFromDump(j.getAsJsonObject()));
});
return (Publication) PublicationToOaf
.generatePublicationActionsFromDump(j.getAsJsonObject());
}).filter(p -> p != null);
Dataset<Tuple2<String, Publication>> publicationDataset = spark
Dataset<Publication> publicationDataset = spark
.createDataset(
oafPublicationRDD.repartition(1).rdd(),
Encoders.tuple(Encoders.STRING(), Encoders.bean(Publication.class)));
Encoders.bean(Publication.class));
publicationDataset.write().mode(SaveMode.Overwrite).save(workingPath + "no_doi_dataset/output");
});
}

View File

@ -1,31 +0,0 @@
<configuration>
<property>
<name>oozie.action.sharelib.for.java</name>
<value>spark2</value>
</property>
<property>
<name>oozie.launcher.mapreduce.user.classpath.first</name>
<value>true</value>
</property>
<property>
<name>oozie.launcher.mapreduce.map.java.opts</name>
<value>-Xmx4g</value>
</property>
<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>

View File

@ -1,11 +1,56 @@
<workflow-app name="Gen Enriched Orcid Works" xmlns="uri:oozie:workflow:0.5">
<workflow-app name="gen_orcid_no_doi_dataset" xmlns="uri:oozie:workflow:0.5">
<parameters>
<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>
<property>
<name>oozieActionShareLibForSpark2</name>
<description>oozie action sharelib for spark 2.*</description>
</property>
<property>
<name>spark2ExtraListeners</name>
<value>com.cloudera.spark.lineage.NavigatorAppListener</value>
<description>spark 2.* extra listeners classname</description>
</property>
<property>
<name>spark2SqlQueryExecutionListeners</name>
<value>com.cloudera.spark.lineage.NavigatorQueryListener</value>
<description>spark 2.* sql query execution listeners classname</description>
</property>
<property>
<name>spark2YarnHistoryServerAddress</name>
<description>spark 2.* yarn history server address</description>
</property>
<property>
<name>spark2EventLogDir</name>
<description>spark 2.* event log dir location</description>
</property>
<property>
<name>workingPath</name>
<description>the working dir base path</description>
</property>
</parameters>
<global>
<job-tracker>${jobTracker}</job-tracker>
<name-node>${nameNode}</name-node>
<configuration>
<property>
<name>oozie.action.sharelib.for.spark</name>
<value>${oozieActionShareLibForSpark2}</value>
</property>
</configuration>
</global>
<start to="ResetWorkingPath"/>
<kill name="Kill">
@ -16,20 +61,25 @@
<fs>
<delete path='${workingPath}/no_doi_enriched_works/output'/>
</fs>
<ok to="Gen_Enriched_Orcid_Works"/>
<ok to="GenOrcidNoDoiDataset"/>
<error to="Kill"/>
</action>
<action name="Gen_Enriched_Orcid_Works">
<action name="GenOrcidNoDoiDataset">
<spark xmlns="uri:oozie:spark-action:0.2">
<job-tracker>${jobTracker}</job-tracker>
<name-node>${nameNode}</name-node>
<master>yarn</master>
<master>yarn-cluster</master>
<mode>cluster</mode>
<name>Gen_Enriched_Orcid_Works</name>
<name>GenOrcidNoDoiDataset</name>
<class>eu.dnetlib.doiboost.orcidnodoi.SparkGenEnrichedOrcidWorks</class>
<jar>dhp-doiboost-1.2.4-SNAPSHOT.jar</jar>
<spark-opts>--num-executors 10 --conf spark.yarn.jars=&quot;hdfs://hadoop-rm1.garr-pa1.d4science.org:8020/user/oozie/share/lib/lib_20180405103059/spark2&quot; --executor-memory=${sparkExecutorMemory} --executor-cores=${sparkExecutorCores} --driver-memory=${sparkDriverMemory}
<jar>dhp-doiboost-${projectVersion}.jar</jar>
<spark-opts>
--executor-memory=${sparkExecutorMemory}
--executor-cores=${sparkExecutorCores}
--driver-memory=${sparkDriverMemory}
--conf spark.extraListeners=${spark2ExtraListeners}
--conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
--conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
--conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
</spark-opts>
<arg>-w</arg><arg>${workingPath}/</arg>
<arg>-n</arg><arg>${nameNode}</arg>

View File

@ -12,7 +12,7 @@ import eu.dnetlib.doiboost.orcid.model.WorkData;
public class XMLRecordParserTest {
@Test
public void testOrcidAuthorDataXMLParser() throws Exception {
private void testOrcidAuthorDataXMLParser() throws Exception {
String xml = IOUtils.toString(this.getClass().getResourceAsStream("summary_0000-0001-6828-479X.xml"));
@ -27,7 +27,7 @@ public class XMLRecordParserTest {
}
@Test
public void testOrcidXMLErrorRecordParser() throws Exception {
private void testOrcidXMLErrorRecordParser() throws Exception {
String xml = IOUtils.toString(this.getClass().getResourceAsStream("summary_error.xml"));
@ -40,7 +40,7 @@ public class XMLRecordParserTest {
}
@Test
public void testOrcidWorkDataXMLParser() throws Exception {
private void testOrcidWorkDataXMLParser() throws Exception {
String xml = IOUtils
.toString(