D-Net
/
dnet-hadoop
14
5
Fork 11
Code Issues 14 Pull Requests 13 Projects 2 Releases 2 Wiki Activity

added ODF invisible stores in raw_all workflow

This commit is contained in:
Michele Artini 2024-07-23 09:56:27 +02:00
parent 4f4c73d65b
commit d27e9ea50f
1 changed files with 28 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

30
dhp-workflows/dhp-graph-mapper/src/main/resources/eu/dnetlib/dhp/oa/graph/raw_all/oozie_app/workflow.xml
View File

@ -368,6 +368,32 @@
<arg>--mdLayout</arg><arg>store</arg> <arg>--mdLayout</arg><arg>store</arg>
<arg>--mdInterpretation</arg><arg>cleaned</arg> <arg>--mdInterpretation</arg><arg>cleaned</arg>
</spark> </spark>
<ok to="ImportODF_hdfs_invisible"/>
<error to="Kill"/>
</action>
<action name="ImportODF_hdfs_invisible">
<spark xmlns="uri:oozie:spark-action:0.2">
<master>yarn</master>
<mode>cluster</mode>
<name>ImportODF_hdfs_invisible</name>
<class>eu.dnetlib.dhp.oa.graph.raw.MigrateHdfsMdstoresApplication</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}${spark2EventLogDir}
</spark-opts>
<arg>--hdfsPath</arg><arg>${contentPath}/odf_mdstore_hdfs_invisible</arg>
<arg>--mdstoreManagerUrl</arg><arg>${mdstoreManagerUrl}</arg>
<arg>--mdFormat</arg><arg>ODF</arg>
<arg>--mdLayout</arg><arg>store</arg>
<arg>--mdInterpretation</arg><arg>intersection</arg>
</spark>
<ok to="reuse_oaf_hdfs"/> <ok to="reuse_oaf_hdfs"/>
<error to="Kill"/> <error to="Kill"/>
</action> </action>
@ -529,7 +555,7 @@
--conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress} --conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
--conf spark.eventLog.dir=${nameNode}${spark2EventLogDir} --conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
</spark-opts> </spark-opts>
<arg>--sourcePaths</arg><arg>${contentPath}/db_openaire,${contentPath}/db_openorgs,${contentPath}/oaf_mdstore_hdfs,${contentPath}/odf_mdstore_hdfs,${contentPath}/mdstore/*/*</arg> <arg>--sourcePaths</arg><arg>${contentPath}/db_openaire,${contentPath}/db_openorgs,${contentPath}/oaf_mdstore_hdfs,${contentPath}/odf_mdstore_hdfs,${contentPath}/odf_mdstore_hdfs_invisible,${contentPath}/mdstore/*/*</arg>
<arg>--invalidPath</arg><arg>${workingDir}/invalid_records</arg> <arg>--invalidPath</arg><arg>${workingDir}/invalid_records</arg>
<arg>--isLookupUrl</arg><arg>${isLookupUrl}</arg> <arg>--isLookupUrl</arg><arg>${isLookupUrl}</arg>
</spark> </spark>
@ -553,7 +579,7 @@
--conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress} --conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
--conf spark.eventLog.dir=${nameNode}${spark2EventLogDir} --conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
</spark-opts> </spark-opts>
<arg>--sourcePaths</arg><arg>${contentPath}/db_openaire,${contentPath}/db_openorgs,${contentPath}/oaf_mdstore_hdfs,${contentPath}/odf_mdstore_hdfs,${contentPath}/mdstore/*/*</arg> <arg>--sourcePaths</arg><arg>${contentPath}/db_openaire,${contentPath}/db_openorgs,${contentPath}/oaf_mdstore_hdfs,${contentPath}/odf_mdstore_hdfs,${contentPath}/odf_mdstore_hdfs_invisible,${contentPath}/mdstore/*/*</arg>
<arg>--targetPath</arg><arg>${workingDir}/entities</arg> <arg>--targetPath</arg><arg>${workingDir}/entities</arg>
<arg>--isLookupUrl</arg><arg>${isLookupUrl}</arg> <arg>--isLookupUrl</arg><arg>${isLookupUrl}</arg>
<arg>--shouldHashId</arg><arg>${shouldHashId}</arg> <arg>--shouldHashId</arg><arg>${shouldHashId}</arg>