[orcidenrichment] refactoring

This commit is contained in:
Miriam Baglioni 2024-11-06 14:15:34 +01:00
parent 939c84ede6
commit e4f89f9800
2 changed files with 26 additions and 244 deletions

View File

@ -1,21 +1,32 @@
[
{
"paramName":"s",
"paramLongName":"sourcePath",
"paramLongName":"graphPath",
"paramDescription": "the path of the sequencial file to read",
"paramRequired": true
},
{
"paramName": "out",
"paramLongName": "outputPath",
"paramLongName": "targetPath",
"paramDescription": "the path used to store temporary output files",
"paramRequired": true
}, {
"paramName": "o",
"paramLongName": "orcidPath",
"paramDescription": "the path used to store temporary output files",
"paramRequired": true
}, {
"paramName": "w",
"paramLongName": "workingDir",
"paramDescription": "the path used to store temporary output files",
"paramRequired": true
},
{
"paramName": "ssm",
"paramLongName": "isSparkSessionManaged",
"paramDescription": "true if the spark session is managed, false otherwise",
"paramName": "m",
"paramLongName": "matchingSource",
"paramDescription": "the path used to store temporary output files",
"paramRequired": false
}
]

View File

@ -92,21 +92,14 @@
<error to="Kill"/>
</action>
<join name="copy_wait" to="fork_prepare_assoc_step1"/>
<join name="copy_wait" to="exec_propagation"/>
<fork name="fork_prepare_assoc_step1">
<path start="join_prepare_publication"/>
<path start="join_prepare_dataset"/>
<path start="join_prepare_otherresearchproduct"/>
<path start="join_prepare_software"/>
</fork>
<action name="join_prepare_publication">
<action name="exec_propagation">
<spark xmlns="uri:oozie:spark-action:0.2">
<master>yarn</master>
<mode>cluster</mode>
<name>ORCIDPropagation-PreparePhase1-Publications</name>
<class>eu.dnetlib.dhp.orcidtoresultfromsemrel.PrepareResultOrcidAssociationStep1</class>
<class>eu.dnetlib.dhp.orcidtoresultfromsemrel.SparkPropagateOrcidAuthor</class>
<jar>dhp-enrichment-${projectVersion}.jar</jar>
<spark-opts>
--executor-cores=${sparkExecutorCores}
@ -119,239 +112,17 @@
--conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
--conf spark.sql.shuffle.partitions=8000
</spark-opts>
<arg>--sourcePath</arg><arg>${sourcePath}</arg>
<arg>--hive_metastore_uris</arg><arg>${hive_metastore_uris}</arg>
<arg>--resultTableName</arg><arg>eu.dnetlib.dhp.schema.oaf.Publication</arg>
<arg>--outputPath</arg><arg>${workingDir}/orcid/targetOrcidAssoc</arg>
<arg>--allowedsemrels</arg><arg>${allowedsemrels}</arg>
<arg>--graphPath</arg><arg>${sourcePath}/</arg>
<arg>--orcidPath</arg><arg>${sourcePath}/</arg>
<arg>--workingDir</arg><arg>${workingDir}/</arg>
<arg>--targetPath</arg><arg>${outputPath}/</arg>
<arg>--matchingSource</arg><arg>graph</arg>
</spark>
<ok to="wait"/>
<ok to="End"/>
<error to="Kill"/>
</action>
<action name="join_prepare_dataset">
<spark xmlns="uri:oozie:spark-action:0.2">
<master>yarn</master>
<mode>cluster</mode>
<name>ORCIDPropagation-PreparePhase1-Dataset</name>
<class>eu.dnetlib.dhp.orcidtoresultfromsemrel.PrepareResultOrcidAssociationStep1</class>
<jar>dhp-enrichment-${projectVersion}.jar</jar>
<spark-opts>
--executor-cores=${sparkExecutorCores}
--executor-memory=${sparkExecutorMemory}
--driver-memory=${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}
</spark-opts>
<arg>--sourcePath</arg><arg>${sourcePath}</arg>
<arg>--hive_metastore_uris</arg><arg>${hive_metastore_uris}</arg>
<arg>--resultTableName</arg><arg>eu.dnetlib.dhp.schema.oaf.Dataset</arg>
<arg>--outputPath</arg><arg>${workingDir}/orcid/targetOrcidAssoc</arg>
<arg>--allowedsemrels</arg><arg>${allowedsemrels}</arg>
</spark>
<ok to="wait"/>
<error to="Kill"/>
</action>
<action name="join_prepare_otherresearchproduct">
<spark xmlns="uri:oozie:spark-action:0.2">
<master>yarn</master>
<mode>cluster</mode>
<name>ORCIDPropagation-PreparePhase1-ORP</name>
<class>eu.dnetlib.dhp.orcidtoresultfromsemrel.PrepareResultOrcidAssociationStep1</class>
<jar>dhp-enrichment-${projectVersion}.jar</jar>
<spark-opts>
--executor-cores=${sparkExecutorCores}
--executor-memory=${sparkExecutorMemory}
--driver-memory=${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}
</spark-opts>
<arg>--sourcePath</arg><arg>${sourcePath}</arg>
<arg>--hive_metastore_uris</arg><arg>${hive_metastore_uris}</arg>
<arg>--resultTableName</arg><arg>eu.dnetlib.dhp.schema.oaf.OtherResearchProduct</arg>
<arg>--outputPath</arg><arg>${workingDir}/orcid/targetOrcidAssoc</arg>
<arg>--allowedsemrels</arg><arg>${allowedsemrels}</arg>
</spark>
<ok to="wait"/>
<error to="Kill"/>
</action>
<action name="join_prepare_software">
<spark xmlns="uri:oozie:spark-action:0.2">
<master>yarn</master>
<mode>cluster</mode>
<name>ORCIDPropagation-PreparePhase1-Software</name>
<class>eu.dnetlib.dhp.orcidtoresultfromsemrel.PrepareResultOrcidAssociationStep1</class>
<jar>dhp-enrichment-${projectVersion}.jar</jar>
<spark-opts>
--executor-cores=${sparkExecutorCores}
--executor-memory=${sparkExecutorMemory}
--driver-memory=${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}
</spark-opts>
<arg>--sourcePath</arg><arg>${sourcePath}</arg>
<arg>--hive_metastore_uris</arg><arg>${hive_metastore_uris}</arg>
<arg>--resultTableName</arg><arg>eu.dnetlib.dhp.schema.oaf.Software</arg>
<arg>--outputPath</arg><arg>${workingDir}/orcid/targetOrcidAssoc</arg>
<arg>--allowedsemrels</arg><arg>${allowedsemrels}</arg>
</spark>
<ok to="wait"/>
<error to="Kill"/>
</action>
<join name="wait" to="prepare_assoc_step2"/>
<action name="prepare_assoc_step2">
<spark xmlns="uri:oozie:spark-action:0.2">
<master>yarn</master>
<mode>cluster</mode>
<name>ORCIDPropagation-PreparePhase2</name>
<class>eu.dnetlib.dhp.orcidtoresultfromsemrel.PrepareResultOrcidAssociationStep2</class>
<jar>dhp-enrichment-${projectVersion}.jar</jar>
<spark-opts>
--executor-cores=${sparkExecutorCores}
--executor-memory=${sparkExecutorMemory}
--driver-memory=${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}
</spark-opts>
<arg>--sourcePath</arg><arg>${workingDir}/orcid/targetOrcidAssoc</arg>
<arg>--outputPath</arg><arg>${workingDir}/orcid/mergedOrcidAssoc</arg>
</spark>
<ok to="fork-join-exec-propagation"/>
<error to="Kill"/>
</action>
<fork name="fork-join-exec-propagation">
<path start="join_propagate_publication"/>
<path start="join_propagate_dataset"/>
<path start="join_propagate_otherresearchproduct"/>
<path start="join_propagate_software"/>
</fork>
<action name="join_propagate_publication">
<spark xmlns="uri:oozie:spark-action:0.2">
<master>yarn</master>
<mode>cluster</mode>
<name>ORCIDPropagation-Publication</name>
<class>eu.dnetlib.dhp.orcidtoresultfromsemrel.SparkOrcidToResultFromSemRelJob</class>
<jar>dhp-enrichment-${projectVersion}.jar</jar>
<spark-opts>
--executor-cores=${sparkExecutorCores}
--executor-memory=${sparkExecutorMemory}
--driver-memory=${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=15000
</spark-opts>
<arg>--possibleUpdatesPath</arg><arg>${workingDir}/orcid/mergedOrcidAssoc</arg>
<arg>--sourcePath</arg><arg>${sourcePath}/publication</arg>
<arg>--resultTableName</arg><arg>eu.dnetlib.dhp.schema.oaf.Publication</arg>
<arg>--outputPath</arg><arg>${outputPath}/publication</arg>
</spark>
<ok to="wait2"/>
<error to="Kill"/>
</action>
<action name="join_propagate_dataset">
<spark xmlns="uri:oozie:spark-action:0.2">
<master>yarn</master>
<mode>cluster</mode>
<name>ORCIDPropagation-Dataset</name>
<class>eu.dnetlib.dhp.orcidtoresultfromsemrel.SparkOrcidToResultFromSemRelJob</class>
<jar>dhp-enrichment-${projectVersion}.jar</jar>
<spark-opts>
--executor-cores=${sparkExecutorCores}
--executor-memory=${sparkExecutorMemory}
--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}${spark2EventLogDir}
--conf spark.sql.shuffle.partitions=8000
</spark-opts>
<arg>--possibleUpdatesPath</arg><arg>${workingDir}/orcid/mergedOrcidAssoc</arg>
<arg>--sourcePath</arg><arg>${sourcePath}/dataset</arg>
<arg>--resultTableName</arg><arg>eu.dnetlib.dhp.schema.oaf.Dataset</arg>
<arg>--outputPath</arg><arg>${outputPath}/dataset</arg>
</spark>
<ok to="wait2"/>
<error to="Kill"/>
</action>
<action name="join_propagate_otherresearchproduct">
<spark xmlns="uri:oozie:spark-action:0.2">
<master>yarn</master>
<mode>cluster</mode>
<name>ORCIDPropagation-ORP</name>
<class>eu.dnetlib.dhp.orcidtoresultfromsemrel.SparkOrcidToResultFromSemRelJob</class>
<jar>dhp-enrichment-${projectVersion}.jar</jar>
<spark-opts>
--executor-cores=${sparkExecutorCores}
--executor-memory=${sparkExecutorMemory}
--driver-memory=${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=8000
</spark-opts>
<arg>--possibleUpdatesPath</arg><arg>${workingDir}/orcid/mergedOrcidAssoc</arg>
<arg>--sourcePath</arg><arg>${sourcePath}/otherresearchproduct</arg>
<arg>--resultTableName</arg><arg>eu.dnetlib.dhp.schema.oaf.OtherResearchProduct</arg>
<arg>--outputPath</arg><arg>${outputPath}/otherresearchproduct</arg>
</spark>
<ok to="wait2"/>
<error to="Kill"/>
</action>
<action name="join_propagate_software">
<spark xmlns="uri:oozie:spark-action:0.2">
<master>yarn</master>
<mode>cluster</mode>
<name>ORCIDPropagation-Software</name>
<class>eu.dnetlib.dhp.orcidtoresultfromsemrel.SparkOrcidToResultFromSemRelJob</class>
<jar>dhp-enrichment-${projectVersion}.jar</jar>
<spark-opts>
--executor-cores=${sparkExecutorCores}
--executor-memory=${sparkExecutorMemory}
--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}${spark2EventLogDir}
--conf spark.sql.shuffle.partitions=4000
</spark-opts>
<arg>--possibleUpdatesPath</arg><arg>${workingDir}/orcid/mergedOrcidAssoc</arg>
<arg>--sourcePath</arg><arg>${sourcePath}/software</arg>
<arg>--resultTableName</arg><arg>eu.dnetlib.dhp.schema.oaf.Software</arg>
<arg>--outputPath</arg><arg>${outputPath}/software</arg>
</spark>
<ok to="wait2"/>
<error to="Kill"/>
</action>
<join name="wait2" to="End"/>
<end name="End"/>