introduced post processing step following the hive DB creation/population

This commit is contained in:
Claudio Atzori 2020-03-04 10:56:50 +01:00
parent 1e563bc15e
commit 0233987603
2 changed files with 26 additions and 1 deletions

View File

@ -0,0 +1,8 @@
CREATE view result as
select id, dateofcollection, bestaccessright, datainfo, collectedfrom, pid, author, resulttype, language, country, subject, description, dateofacceptance, embargoenddate, resourcetype, context, instance from ${hive_db_name}.publication p
union all
select id, dateofcollection, bestaccessright, datainfo, collectedfrom, pid, author, resulttype, language, country, subject, description, dateofacceptance, embargoenddate, resourcetype, context, instance from ${hive_db_name}.dataset d
union all
select id, dateofcollection, bestaccessright, datainfo, collectedfrom, pid, author, resulttype, language, country, subject, description, dateofacceptance, embargoenddate, resourcetype, context, instance from ${hive_db_name}.software s
union all
select id, dateofcollection, bestaccessright, datainfo, collectedfrom, pid, author, resulttype, language, country, subject, description, dateofacceptance, embargoenddate, resourcetype, context, instance from ${hive_db_name}.otherresearchproduct o;

View File

@ -37,12 +37,29 @@
<name>MapGraphIntoDataFrame</name> <name>MapGraphIntoDataFrame</name>
<class>eu.dnetlib.dhp.graph.SparkGraphImporterJob</class> <class>eu.dnetlib.dhp.graph.SparkGraphImporterJob</class>
<jar>dhp-graph-mapper-${projectVersion}.jar</jar> <jar>dhp-graph-mapper-${projectVersion}.jar</jar>
<spark-opts>--executor-memory ${sparkExecutorMemory} --executor-cores ${sparkExecutorCores} --driver-memory=${sparkDriverMemory} --conf spark.extraListeners="com.cloudera.spark.lineage.NavigatorAppListener" --conf spark.sql.queryExecutionListeners="com.cloudera.spark.lineage.NavigatorQueryListener" --conf spark.sql.warehouse.dir="/user/hive/warehouse"</spark-opts> <spark-opts>
--executor-memory ${sparkExecutorMemory}
--executor-cores ${sparkExecutorCores}
--driver-memory=${sparkDriverMemory}
--conf spark.extraListeners="com.cloudera.spark.lineage.NavigatorAppListener"
--conf spark.sql.queryExecutionListeners="com.cloudera.spark.lineage.NavigatorQueryListener"
--conf spark.sql.warehouse.dir="/user/hive/warehouse"</spark-opts>
<arg>-mt</arg> <arg>yarn-cluster</arg> <arg>-mt</arg> <arg>yarn-cluster</arg>
<arg>--sourcePath</arg><arg>${sourcePath}</arg> <arg>--sourcePath</arg><arg>${sourcePath}</arg>
<arg>--hive_db_name</arg><arg>${hive_db_name}</arg> <arg>--hive_db_name</arg><arg>${hive_db_name}</arg>
<arg>--hive_metastore_uris</arg><arg>${hive_metastore_uris}</arg> <arg>--hive_metastore_uris</arg><arg>${hive_metastore_uris}</arg>
</spark> </spark>
<ok to="PostProcessing"/>
<error to="Kill"/>
</action>
<action name="PostProcessing">
<hive xmlns="uri:oozie:hive-action:0.2">
<job-tracker>${jobTracker}</job-tracker>
<name-node>${nameNode}</name-node>
<script>/eu/dnetlib/dhp/graph/hive/postprocessing.sql</script>
<param>hive_db_name=${hive_db_name}</param>
</hive>
<ok to="End"/> <ok to="End"/>
<error to="Kill"/> <error to="Kill"/>
</action> </action>