dnet-hadoop/dhp-workflows/dhp-graph-mapper/src/main/resources/eu/dnetlib/dhp/oa/graph/dump/oozie_app/workflow.xml

430 lines
19 KiB
XML

<workflow-app name="dump_community_products" xmlns="uri:oozie:workflow:0.5">
<parameters>
<property>
<name>sourcePath</name>
<description>the source path</description>
</property>
<property>
<name>isLookUpUrl</name>
<description>the isLookup service endpoint</description>
</property>
<property>
<name>outputPath</name>
<description>the output path</description>
</property>
<property>
<name>accessToken</name>
<description>the access token used for the deposition in Zenodo</description>
</property>
<property>
<name>connectionUrl</name>
<description>the connection url for Zenodo</description>
</property>
<property>
<name>metadata</name>
<description> the metadata associated to the deposition</description>
</property>
<property>
<name>newDeposition</name>
<description>true if it is a brand new depositon. false for new version of an old deposition</description>
</property>
<property>
<name>conceptRecordId</name>
<description>for new version, the id of the record for the old deposition</description>
</property>
<property>
<name>hiveDbName</name>
<description>the target hive database name</description>
</property>
<property>
<name>hiveJdbcUrl</name>
<description>hive server jdbc url</description>
</property>
<property>
<name>hiveMetastoreUris</name>
<description>hive server metastore URIs</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>
<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>
</parameters>
<global>
<job-tracker>${jobTracker}</job-tracker>
<name-node>${nameNode}</name-node>
<configuration>
<property>
<name>mapreduce.job.queuename</name>
<value>${queueName}</value>
</property>
<property>
<name>oozie.launcher.mapred.job.queue.name</name>
<value>${oozieLauncherQueueName}</value>
</property>
<property>
<name>oozie.action.sharelib.for.spark</name>
<value>${oozieActionShareLibForSpark2}</value>
</property>
</configuration>
</global>
<start to="reset_outputpath"/>
<kill name="Kill">
<message>Action failed, error message[${wf:errorMessage(wf:lastErrorNode())}]</message>
</kill>
<action name="reset_outputpath">
<fs>
<delete path="${outputPath}"/>
<mkdir path="${outputPath}"/>
</fs>
<ok to="save_community_map"/>
<error to="Kill"/>
</action>
<action name="save_community_map">
<java>
<main-class>eu.dnetlib.dhp.oa.graph.dump.SaveCommunityMap</main-class>
<arg>--outputPath</arg><arg>${workingDir}/communityMap</arg>
<arg>--nameNode</arg><arg>${nameNode}</arg>
<arg>--isLookUpUrl</arg><arg>${isLookUpUrl}</arg>
</java>
<ok to="fork_dump"/>
<error to="Kill"/>
</action>
<fork name="fork_dump">
<path start="dump_publication"/>
<path start="dump_dataset"/>
<path start="dump_orp"/>
<path start="dump_software"/>
</fork>
<action name="dump_publication">
<spark xmlns="uri:oozie:spark-action:0.2">
<master>yarn</master>
<mode>cluster</mode>
<name>Dump table publication for community related products</name>
<class>eu.dnetlib.dhp.oa.graph.dump.community.SparkDumpCommunityProducts</class>
<jar>dhp-graph-mapper-${projectVersion}.jar</jar>
<spark-opts>
--executor-memory=${sparkExecutorMemory}
--executor-cores=${sparkExecutorCores}
--driver-memory=${sparkDriverMemory}
--conf spark.extraListeners=${spark2ExtraListeners}
--conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
--conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
--conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
--conf spark.sql.warehouse.dir=${sparkSqlWarehouseDir}
</spark-opts>
<arg>--sourcePath</arg><arg>${sourcePath}/publication</arg>
<arg>--resultTableName</arg><arg>eu.dnetlib.dhp.schema.oaf.Publication</arg>
<arg>--outputPath</arg><arg>${workingDir}/publication</arg>
<arg>--communityMapPath</arg><arg>${workingDir}/communityMap</arg>
</spark>
<ok to="join_dump"/>
<error to="Kill"/>
</action>
<action name="dump_dataset">
<spark xmlns="uri:oozie:spark-action:0.2">
<master>yarn</master>
<mode>cluster</mode>
<name>Dump table dataset for community related products</name>
<class>eu.dnetlib.dhp.oa.graph.dump.community.SparkDumpCommunityProducts</class>
<jar>dhp-graph-mapper-${projectVersion}.jar</jar>
<spark-opts>
--executor-memory=${sparkExecutorMemory}
--executor-cores=${sparkExecutorCores}
--driver-memory=${sparkDriverMemory}
--conf spark.extraListeners=${spark2ExtraListeners}
--conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
--conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
--conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
--conf spark.sql.warehouse.dir=${sparkSqlWarehouseDir}
</spark-opts>
<arg>--sourcePath</arg><arg>${sourcePath}/dataset</arg>
<arg>--resultTableName</arg><arg>eu.dnetlib.dhp.schema.oaf.Dataset</arg>
<arg>--outputPath</arg><arg>${workingDir}/dataset</arg>
<arg>--communityMapPath</arg><arg>${workingDir}/communityMap</arg>
</spark>
<ok to="join_dump"/>
<error to="Kill"/>
</action>
<action name="dump_orp">
<spark xmlns="uri:oozie:spark-action:0.2">
<master>yarn</master>
<mode>cluster</mode>
<name>Dump table ORP for community related products</name>
<class>eu.dnetlib.dhp.oa.graph.dump.community.SparkDumpCommunityProducts</class>
<jar>dhp-graph-mapper-${projectVersion}.jar</jar>
<spark-opts>
--executor-memory=${sparkExecutorMemory}
--executor-cores=${sparkExecutorCores}
--driver-memory=${sparkDriverMemory}
--conf spark.extraListeners=${spark2ExtraListeners}
--conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
--conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
--conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
--conf spark.sql.warehouse.dir=${sparkSqlWarehouseDir}
</spark-opts>
<arg>--sourcePath</arg><arg>${sourcePath}/otherresearchproduct</arg>
<arg>--resultTableName</arg><arg>eu.dnetlib.dhp.schema.oaf.OtherResearchProduct</arg>
<arg>--outputPath</arg><arg>${workingDir}/otherresearchproduct</arg>
<arg>--communityMapPath</arg><arg>${workingDir}/communityMap</arg>
</spark>
<ok to="join_dump"/>
<error to="Kill"/>
</action>
<action name="dump_software">
<spark xmlns="uri:oozie:spark-action:0.2">
<master>yarn</master>
<mode>cluster</mode>
<name>Dump table software for community related products</name>
<class>eu.dnetlib.dhp.oa.graph.dump.community.SparkDumpCommunityProducts</class>
<jar>dhp-graph-mapper-${projectVersion}.jar</jar>
<spark-opts>
--executor-memory=${sparkExecutorMemory}
--executor-cores=${sparkExecutorCores}
--driver-memory=${sparkDriverMemory}
--conf spark.extraListeners=${spark2ExtraListeners}
--conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
--conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
--conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
--conf spark.sql.warehouse.dir=${sparkSqlWarehouseDir}
</spark-opts>
<arg>--sourcePath</arg><arg>${sourcePath}/software</arg>
<arg>--resultTableName</arg><arg>eu.dnetlib.dhp.schema.oaf.Software</arg>
<arg>--outputPath</arg><arg>${workingDir}/software</arg>
<arg>--communityMapPath</arg><arg>${workingDir}/communityMap</arg>
</spark>
<ok to="join_dump"/>
<error to="Kill"/>
</action>
<join name="join_dump" to="prepareResultProject"/>
<action name="prepareResultProject">
<spark xmlns="uri:oozie:spark-action:0.2">
<master>yarn</master>
<mode>cluster</mode>
<name>Prepare association result subset of project info</name>
<class>eu.dnetlib.dhp.oa.graph.dump.community.SparkPrepareResultProject</class>
<jar>dhp-graph-mapper-${projectVersion}.jar</jar>
<spark-opts>
--executor-memory=${sparkExecutorMemory}
--executor-cores=${sparkExecutorCores}
--driver-memory=${sparkDriverMemory}
--conf spark.extraListeners=${spark2ExtraListeners}
--conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
--conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
--conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
--conf spark.sql.warehouse.dir=${sparkSqlWarehouseDir}
</spark-opts>
<arg>--sourcePath</arg><arg>${sourcePath}</arg>
<arg>--outputPath</arg><arg>${workingDir}/preparedInfo</arg>
</spark>
<ok to="fork_extendWithProject"/>
<error to="Kill"/>
</action>
<fork name="fork_extendWithProject">
<path start="extend_publication"/>
<path start="extend_dataset"/>
<path start="extend_orp"/>
<path start="extend_software"/>
</fork>
<action name="extend_publication">
<spark xmlns="uri:oozie:spark-action:0.2">
<master>yarn</master>
<mode>cluster</mode>
<name>Extend dumped publications with information about project</name>
<class>eu.dnetlib.dhp.oa.graph.dump.community.SparkUpdateProjectInfo</class>
<jar>dhp-graph-mapper-${projectVersion}.jar</jar>
<spark-opts>
--executor-memory=${sparkExecutorMemory}
--executor-cores=${sparkExecutorCores}
--driver-memory=${sparkDriverMemory}
--conf spark.extraListeners=${spark2ExtraListeners}
--conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
--conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
--conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
--conf spark.sql.warehouse.dir=${sparkSqlWarehouseDir}
</spark-opts>
<arg>--sourcePath</arg><arg>${workingDir}/publication</arg>
<arg>--outputPath</arg><arg>${workingDir}/ext/publication</arg>
<arg>--preparedInfoPath</arg><arg>${workingDir}/preparedInfo</arg>
</spark>
<ok to="join_extend"/>
<error to="Kill"/>
</action>
<action name="extend_dataset">
<spark xmlns="uri:oozie:spark-action:0.2">
<master>yarn</master>
<mode>cluster</mode>
<name>Extend dumped dataset with information about project</name>
<class>eu.dnetlib.dhp.oa.graph.dump.community.SparkUpdateProjectInfo</class>
<jar>dhp-graph-mapper-${projectVersion}.jar</jar>
<spark-opts>
--executor-memory=${sparkExecutorMemory}
--executor-cores=${sparkExecutorCores}
--driver-memory=${sparkDriverMemory}
--conf spark.extraListeners=${spark2ExtraListeners}
--conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
--conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
--conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
--conf spark.sql.warehouse.dir=${sparkSqlWarehouseDir}
</spark-opts>
<arg>--sourcePath</arg><arg>${workingDir}/dataset</arg>
<arg>--outputPath</arg><arg>${workingDir}/ext/dataset</arg>
<arg>--preparedInfoPath</arg><arg>${workingDir}/preparedInfo</arg>
</spark>
<ok to="join_extend"/>
<error to="Kill"/>
</action>
<action name="extend_orp">
<spark xmlns="uri:oozie:spark-action:0.2">
<master>yarn</master>
<mode>cluster</mode>
<name>Extend dumped ORP with information about project</name>
<class>eu.dnetlib.dhp.oa.graph.dump.community.SparkUpdateProjectInfo</class>
<jar>dhp-graph-mapper-${projectVersion}.jar</jar>
<spark-opts>
--executor-memory=${sparkExecutorMemory}
--executor-cores=${sparkExecutorCores}
--driver-memory=${sparkDriverMemory}
--conf spark.extraListeners=${spark2ExtraListeners}
--conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
--conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
--conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
--conf spark.sql.warehouse.dir=${sparkSqlWarehouseDir}
</spark-opts>
<arg>--sourcePath</arg><arg>${workingDir}/otherresearchproduct</arg>
<arg>--outputPath</arg><arg>${workingDir}/ext/orp</arg>
<arg>--preparedInfoPath</arg><arg>${workingDir}/preparedInfo</arg>
</spark>
<ok to="join_extend"/>
<error to="Kill"/>
</action>
<action name="extend_software">
<spark xmlns="uri:oozie:spark-action:0.2">
<master>yarn</master>
<mode>cluster</mode>
<name>Extend dumped software with information about project</name>
<class>eu.dnetlib.dhp.oa.graph.dump.community.SparkUpdateProjectInfo</class>
<jar>dhp-graph-mapper-${projectVersion}.jar</jar>
<spark-opts>
--executor-memory=${sparkExecutorMemory}
--executor-cores=${sparkExecutorCores}
--driver-memory=${sparkDriverMemory}
--conf spark.extraListeners=${spark2ExtraListeners}
--conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
--conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
--conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
--conf spark.sql.warehouse.dir=${sparkSqlWarehouseDir}
</spark-opts>
<arg>--sourcePath</arg><arg>${workingDir}/software</arg>
<arg>--outputPath</arg><arg>${workingDir}/ext/software</arg>
<arg>--preparedInfoPath</arg><arg>${workingDir}/preparedInfo</arg>
</spark>
<ok to="join_extend"/>
<error to="Kill"/>
</action>
<join name="join_extend" to="splitForCommunities"/>
<action name="splitForCommunities">
<spark xmlns="uri:oozie:spark-action:0.2">
<master>yarn</master>
<mode>cluster</mode>
<name>Split dumped result for community</name>
<class>eu.dnetlib.dhp.oa.graph.dump.community.SparkSplitForCommunity</class>
<jar>dhp-graph-mapper-${projectVersion}.jar</jar>
<spark-opts>
--executor-memory=${sparkExecutorMemory}
--executor-cores=${sparkExecutorCores}
--driver-memory=${sparkDriverMemory}
--conf spark.extraListeners=${spark2ExtraListeners}
--conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
--conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
--conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
--conf spark.sql.warehouse.dir=${sparkSqlWarehouseDir}
</spark-opts>
<arg>--sourcePath</arg><arg>${workingDir}/ext</arg>
<arg>--outputPath</arg><arg>${workingDir}/split</arg>
<arg>--communityMapPath</arg><arg>${workingDir}/communityMap</arg>
</spark>
<ok to="make_archive"/>
<error to="Kill"/>
</action>
<action name="make_archive">
<java>
<main-class>eu.dnetlib.dhp.oa.graph.dump.MakeTar</main-class>
<arg>--hdfsPath</arg><arg>${outputPath}</arg>
<arg>--nameNode</arg><arg>${nameNode}</arg>
<arg>--sourcePath</arg><arg>${workingDir}/split</arg>
</java>
<ok to="send_zenodo"/>
<error to="Kill"/>
</action>
<action name="send_zenodo">
<java>
<main-class>eu.dnetlib.dhp.oa.graph.dump.SendToZenodoHDFS</main-class>
<arg>--hdfsPath</arg><arg>${outputPath}</arg>
<arg>--nameNode</arg><arg>${nameNode}</arg>
<arg>--accessToken</arg><arg>${accessToken}</arg>
<arg>--connectionUrl</arg><arg>${connectionUrl}</arg>
<arg>--metadata</arg><arg>${metadata}</arg>
<arg>--communityMapPath</arg><arg>${workingDir}/communityMap</arg>
<arg>--conceptRecordId</arg><arg>${conceptRecordId}</arg>
<arg>--depositionType</arg><arg>${depositionType}</arg>
</java>
<ok to="End"/>
<error to="Kill"/>
</action>
<end name="End"/>
</workflow-app>