forked from D-Net/dnet-hadoop
428 lines
19 KiB
XML
428 lines
19 KiB
XML
<workflow-app name="orcid_status" xmlns="uri:oozie:workflow:0.5">
|
|
|
|
<parameters>
|
|
<property>
|
|
<name>inputPath</name>
|
|
<description>the input path to read graph content</description>
|
|
</property>
|
|
<property>
|
|
<name>outputPath</name>
|
|
<description>the target path to store cleaned graph</description>
|
|
</property>
|
|
<property>
|
|
<name>orcidInputPath</name>
|
|
<description>the input path where to find the orcid authoritative information</description>
|
|
</property>
|
|
<property>
|
|
<name>orcidReportPath</name>
|
|
</property>
|
|
<property>
|
|
<name>clean</name>
|
|
<value>false</value>
|
|
<description>determines if the orcid should be cleaned in the graph (true) or the report should be produced (false)
|
|
</description>
|
|
</property>
|
|
<property>
|
|
<name>whitelist</name>
|
|
<description>the withelist</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>
|
|
|
|
<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="${outputPath}"/>
|
|
</fs>
|
|
<ok to="perpare_result"/>
|
|
<error to="Kill"/>
|
|
</action>
|
|
|
|
<fork name="perpare_result">
|
|
<path start="prepare_publication"/>
|
|
<path start="prepare_dataset"/>
|
|
<path start="prepare_software"/>
|
|
<path start="prepare_orp"/>
|
|
</fork>
|
|
|
|
<action name="prepare_publication">
|
|
<spark xmlns="uri:oozie:spark-action:0.2">
|
|
<master>yarn</master>
|
|
<mode>cluster</mode>
|
|
<name>orcid prepare publication</name>
|
|
<class>eu.dnetlib.dhp.oa.graph.clean.authorpids.PrepareResultsSparkJob</class>
|
|
<jar>dhp-graph-mapper-${projectVersion}.jar</jar>
|
|
<spark-opts>
|
|
--executor-cores=${sparkExecutorCores}
|
|
--executor-memory=${sparkExecutorMemory}
|
|
--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.shuffle.partitions=7680
|
|
</spark-opts>
|
|
<arg>--inputPath</arg><arg>${inputPath}/publication</arg>
|
|
<arg>--outputPath</arg><arg>${workingDir}/publication</arg>
|
|
<arg>--graphTableClassName</arg><arg>eu.dnetlib.dhp.schema.oaf.Publication</arg>
|
|
|
|
</spark>
|
|
<ok to="wait_prepare"/>
|
|
<error to="Kill"/>
|
|
</action>
|
|
|
|
<action name="prepare_dataset">
|
|
<spark xmlns="uri:oozie:spark-action:0.2">
|
|
<master>yarn</master>
|
|
<mode>cluster</mode>
|
|
<name>orcid prepare dataset</name>
|
|
<class>eu.dnetlib.dhp.oa.graph.clean.authorpids.PrepareResultsSparkJob</class>
|
|
<jar>dhp-graph-mapper-${projectVersion}.jar</jar>
|
|
<spark-opts>
|
|
--executor-cores=${sparkExecutorCores}
|
|
--executor-memory=${sparkExecutorMemory}
|
|
--driver-memory=${sparkDriverMemory}
|
|
--conf spark.extraListeners=${spark2ExtraListeners}
|
|
--conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
|
|
--conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
|
|
--conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
|
|
--conf spark.sql.shuffle.partitions=7680
|
|
</spark-opts>
|
|
<arg>--inputPath</arg><arg>${inputPath}/dataset</arg>
|
|
<arg>--outputPath</arg><arg>${workingDir}/dataset</arg>
|
|
<arg>--graphTableClassName</arg><arg>eu.dnetlib.dhp.schema.oaf.Dataset</arg>
|
|
|
|
</spark>
|
|
<ok to="wait_prepare"/>
|
|
<error to="Kill"/>
|
|
</action>
|
|
|
|
<action name="prepare_software">
|
|
<spark xmlns="uri:oozie:spark-action:0.2">
|
|
<master>yarn</master>
|
|
<mode>cluster</mode>
|
|
<name>orcid prepare software</name>
|
|
<class>eu.dnetlib.dhp.oa.graph.clean.authorpids.PrepareResultsSparkJob</class>
|
|
<jar>dhp-graph-mapper-${projectVersion}.jar</jar>
|
|
<spark-opts>
|
|
--executor-cores=${sparkExecutorCores}
|
|
--executor-memory=${sparkExecutorMemory}
|
|
--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.shuffle.partitions=7680
|
|
</spark-opts>
|
|
<arg>--inputPath</arg><arg>${inputPath}/software</arg>
|
|
<arg>--outputPath</arg><arg>${workingDir}/software</arg>
|
|
<arg>--graphTableClassName</arg><arg>eu.dnetlib.dhp.schema.oaf.Software</arg>
|
|
|
|
</spark>
|
|
<ok to="wait_prepare"/>
|
|
<error to="Kill"/>
|
|
</action>
|
|
|
|
<action name="prepare_orp">
|
|
<spark xmlns="uri:oozie:spark-action:0.2">
|
|
<master>yarn</master>
|
|
<mode>cluster</mode>
|
|
<name>orcid prepare orp</name>
|
|
<class>eu.dnetlib.dhp.oa.graph.clean.authorpids.PrepareResultsSparkJob</class>
|
|
<jar>dhp-graph-mapper-${projectVersion}.jar</jar>
|
|
<spark-opts>
|
|
--executor-cores=${sparkExecutorCores}
|
|
--executor-memory=${sparkExecutorMemory}
|
|
--driver-memory=${sparkDriverMemory}
|
|
--conf spark.extraListeners=${spark2ExtraListeners}
|
|
--conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
|
|
--conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
|
|
--conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
|
|
--conf spark.sql.shuffle.partitions=7680
|
|
</spark-opts>
|
|
<arg>--inputPath</arg><arg>${inputPath}/otherresearchproduct</arg>
|
|
<arg>--outputPath</arg><arg>${workingDir}/otherresearchproduct</arg>
|
|
<arg>--graphTableClassName</arg><arg>eu.dnetlib.dhp.schema.oaf.OtherResearchProduct</arg>
|
|
|
|
</spark>
|
|
<ok to="wait_prepare"/>
|
|
<error to="Kill"/>
|
|
</action>
|
|
|
|
<join name="wait_prepare" to="make_report"/>
|
|
|
|
<decision name="cleanorreport">
|
|
<switch>
|
|
<case to="make_report">${wf:conf('clean') eq false}</case>
|
|
<case to="clean_orcid_copy">${wf:conf('clean') eq true}</case>
|
|
<default to="make_report"/>
|
|
</switch>
|
|
</decision>
|
|
|
|
<fork name="make_report">
|
|
<path start="report_publication"/>
|
|
<path start="report_dataset"/>
|
|
<path start="report_software"/>
|
|
<path start="report_orp"/>
|
|
</fork>
|
|
|
|
<action name="report_publication">
|
|
<spark xmlns="uri:oozie:spark-action:0.2">
|
|
<master>yarn</master>
|
|
<mode>cluster</mode>
|
|
<name>Report ORCID on Publication</name>
|
|
<class>eu.dnetlib.dhp.oa.graph.clean.authorpids.MakeReportSparkJob</class>
|
|
<jar>dhp-graph-mapper-${projectVersion}.jar</jar>
|
|
<spark-opts>
|
|
--executor-cores=${sparkExecutorCores}
|
|
--executor-memory=${sparkExecutorMemory}
|
|
--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.shuffle.partitions=7680
|
|
</spark-opts>
|
|
<arg>--outputPath</arg><arg>${workingDir}/report/publication</arg>
|
|
<arg>--preparedInfoPath</arg><arg>${workingDir}/publication</arg>
|
|
<arg>--orcidInputPath</arg><arg>${orcidInputPath}</arg>
|
|
<arg>--whitelist</arg><arg>${whitelist}</arg>
|
|
</spark>
|
|
<ok to="wait_report"/>
|
|
<error to="Kill"/>
|
|
</action>
|
|
|
|
<action name="report_dataset">
|
|
<spark xmlns="uri:oozie:spark-action:0.2">
|
|
<master>yarn</master>
|
|
<mode>cluster</mode>
|
|
<name>Report ORCID on Dataset</name>
|
|
<class>eu.dnetlib.dhp.oa.graph.clean.authorpids.MakeReportSparkJob</class>
|
|
<jar>dhp-graph-mapper-${projectVersion}.jar</jar>
|
|
<spark-opts>
|
|
--executor-cores=${sparkExecutorCores}
|
|
--executor-memory=${sparkExecutorMemory}
|
|
--driver-memory=${sparkDriverMemory}
|
|
--conf spark.extraListeners=${spark2ExtraListeners}
|
|
--conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
|
|
--conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
|
|
--conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
|
|
--conf spark.sql.shuffle.partitions=7680
|
|
</spark-opts>
|
|
<arg>--preparedInfoPath</arg><arg>${workingDir}/dataset</arg>
|
|
<arg>--outputPath</arg><arg>${workingDir}/report/dataset</arg>
|
|
<arg>--orcidInputPath</arg><arg>${orcidInputPath}</arg>
|
|
<arg>--whitelist</arg><arg>${whitelist}</arg>
|
|
</spark>
|
|
<ok to="wait_report"/>
|
|
<error to="Kill"/>
|
|
</action>
|
|
|
|
<action name="report_orp">
|
|
<spark xmlns="uri:oozie:spark-action:0.2">
|
|
<master>yarn</master>
|
|
<mode>cluster</mode>
|
|
<name>Report ORCID on ORP</name>
|
|
<class>eu.dnetlib.dhp.oa.graph.clean.authorpids.MakeReportSparkJob</class>
|
|
<jar>dhp-graph-mapper-${projectVersion}.jar</jar>
|
|
<spark-opts>
|
|
--executor-cores=${sparkExecutorCores}
|
|
--executor-memory=${sparkExecutorMemory}
|
|
--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.shuffle.partitions=7680
|
|
</spark-opts>
|
|
<arg>--preparedInfoPath</arg><arg>${workingDir}/otherresearchproduct</arg>
|
|
<arg>--outputPath</arg><arg>${workingDir}/report/otherresearchproduct</arg>
|
|
<arg>--orcidInputPath</arg><arg>${orcidInputPath}</arg>
|
|
<arg>--whitelist</arg><arg>${whitelist}</arg>
|
|
</spark>
|
|
<ok to="wait_report"/>
|
|
<error to="Kill"/>
|
|
</action>
|
|
|
|
<action name="report_software">
|
|
<spark xmlns="uri:oozie:spark-action:0.2">
|
|
<master>yarn</master>
|
|
<mode>cluster</mode>
|
|
<name>Report ORCID on Softwar</name>
|
|
<class>eu.dnetlib.dhp.oa.graph.clean.authorpids.MakeReportSparkJob</class>
|
|
<jar>dhp-graph-mapper-${projectVersion}.jar</jar>
|
|
<spark-opts>
|
|
--executor-cores=${sparkExecutorCores}
|
|
--executor-memory=${sparkExecutorMemory}
|
|
--driver-memory=${sparkDriverMemory}
|
|
--conf spark.extraListeners=${spark2ExtraListeners}
|
|
--conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
|
|
--conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
|
|
--conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
|
|
--conf spark.sql.shuffle.partitions=7680
|
|
</spark-opts>
|
|
<arg>--preparedInfoPath</arg><arg>${workingDir}/software</arg>
|
|
<arg>--outputPath</arg><arg>${workingDir}/report/software</arg>
|
|
<arg>--orcidInputPath</arg><arg>${orcidInputPath}</arg>
|
|
<arg>--whitelist</arg><arg>${whitelist}</arg>
|
|
</spark>
|
|
<ok to="wait_report"/>
|
|
<error to="Kill"/>
|
|
</action>
|
|
|
|
<join name="wait_report" to="clean_orcid"/>
|
|
|
|
<fork name="clean_orcid">
|
|
<path start="clean_publication_orcid"/>
|
|
<path start="clean_dataset_orcid"/>
|
|
<path start="clean_orp_orcid"/>
|
|
<path start="clean_software_orcid"/>
|
|
</fork>
|
|
|
|
<action name="clean_publication_orcid">
|
|
<spark xmlns="uri:oozie:spark-action:0.2">
|
|
<master>yarn</master>
|
|
<mode>cluster</mode>
|
|
<name>Clean ORCID for Publications</name>
|
|
<class>eu.dnetlib.dhp.oa.graph.clean.authorpids.CleanAuthorPidsSparkJob</class>
|
|
<jar>dhp-graph-mapper-${projectVersion}.jar</jar>
|
|
<spark-opts>
|
|
--executor-cores=${sparkExecutorCores}
|
|
--executor-memory=${sparkExecutorMemory}
|
|
--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.shuffle.partitions=7680
|
|
</spark-opts>
|
|
<arg>--inputPath</arg><arg>${inputPath}/publication</arg>
|
|
<arg>--outputPath</arg><arg>${outputPath}/publication</arg>
|
|
<arg>--graphTableClassName</arg><arg>eu.dnetlib.dhp.schema.oaf.Publication</arg>
|
|
<arg>--reportPath</arg><arg>${workingDir}/report/publication</arg>
|
|
</spark>
|
|
<ok to="wait_clean"/>
|
|
<error to="Kill"/>
|
|
</action>
|
|
|
|
<action name="clean_dataset_orcid">
|
|
<spark xmlns="uri:oozie:spark-action:0.2">
|
|
<master>yarn</master>
|
|
<mode>cluster</mode>
|
|
<name>Clean ORCID for Datasets</name>
|
|
<class>eu.dnetlib.dhp.oa.graph.clean.authorpids.CleanAuthorPidsSparkJob</class>
|
|
<jar>dhp-graph-mapper-${projectVersion}.jar</jar>
|
|
<spark-opts>
|
|
--executor-cores=${sparkExecutorCores}
|
|
--executor-memory=${sparkExecutorMemory}
|
|
--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.shuffle.partitions=7680
|
|
</spark-opts>
|
|
<arg>--reportPath</arg><arg>${workingDir}/report/dataset</arg>
|
|
<arg>--outputPath</arg><arg>${outputPath}/dataset</arg>
|
|
<arg>--graphTableClassName</arg><arg>eu.dnetlib.dhp.schema.oaf.Dataset</arg>
|
|
<arg>--inputPath</arg><arg>${inputPath}/dataset</arg>
|
|
</spark>
|
|
<ok to="wait_clean"/>
|
|
<error to="Kill"/>
|
|
</action>
|
|
|
|
<action name="clean_orp_orcid">
|
|
<spark xmlns="uri:oozie:spark-action:0.2">
|
|
<master>yarn</master>
|
|
<mode>cluster</mode>
|
|
<name>Clean ORCID for ORP</name>
|
|
<class>eu.dnetlib.dhp.oa.graph.clean.authorpids.CleanAuthorPidsSparkJob</class>
|
|
<jar>dhp-graph-mapper-${projectVersion}.jar</jar>
|
|
<spark-opts>
|
|
--executor-cores=${sparkExecutorCores}
|
|
--executor-memory=${sparkExecutorMemory}
|
|
--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.shuffle.partitions=7680
|
|
</spark-opts>
|
|
<arg>--reportPath</arg><arg>${workingDir}/report/otherresearchproduct</arg>
|
|
<arg>--outputPath</arg><arg>${outputPath}/otherresearchproduct</arg>
|
|
<arg>--graphTableClassName</arg><arg>eu.dnetlib.dhp.schema.oaf.OtherResearchProduct</arg>
|
|
<arg>--inputPath</arg><arg>${inputPath}/otherresearchproduct</arg>
|
|
</spark>
|
|
<ok to="wait_clean"/>
|
|
<error to="Kill"/>
|
|
</action>
|
|
|
|
<action name="clean_software_orcid">
|
|
<spark xmlns="uri:oozie:spark-action:0.2">
|
|
<master>yarn</master>
|
|
<mode>cluster</mode>
|
|
<name>Clean ORCID for Software</name>
|
|
<class>eu.dnetlib.dhp.oa.graph.clean.authorpids.CleanAuthorPidsSparkJob</class>
|
|
<jar>dhp-graph-mapper-${projectVersion}.jar</jar>
|
|
<spark-opts>
|
|
--executor-cores=${sparkExecutorCores}
|
|
--executor-memory=${sparkExecutorMemory}
|
|
--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.shuffle.partitions=7680
|
|
</spark-opts>
|
|
<arg>--reportPath</arg><arg>${workingDir}/report/software</arg>
|
|
<arg>--outputPath</arg><arg>${outputPath}/software</arg>
|
|
<arg>--graphTableClassName</arg><arg>eu.dnetlib.dhp.schema.oaf.Software</arg>
|
|
<arg>--inputPath</arg><arg>${inputPath}/software</arg>
|
|
</spark>
|
|
<ok to="wait_clean"/>
|
|
<error to="Kill"/>
|
|
</action>
|
|
|
|
<join name="wait_clean" to="End"/>
|
|
|
|
<end name="End"/>
|
|
</workflow-app> |