sabeel
/
dnet-hadoop
forked from D-Net/dnet-hadoop
1
0
Fork 0
Code Pull Requests Activity

updated the transformation Baseline workflow to include mdstore rollback/commit action

This commit is contained in:
Sandro La Bruzzo 2023-12-18 11:47:00 +01:00
parent 242d647146
commit 1f1a6a5f5f
2 changed files with 80 additions and 11 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

73
dhp-workflows/dhp-aggregation/src/main/resources/eu/dnetlib/dhp/sx/bio/pubmed/oozie_app/workflow.xml
View File

@ -9,8 +9,12 @@
<description>The IS lookUp service endopoint</description>
</property>
<property>
<name>targetPath</name>
<description>The target path</description>
<name>mdStoreOutputId</name>
<description>the identifier of the cleaned MDStore</description>
</property>
<property>
<name>mdStoreManagerURI</name>
<description>the path of the cleaned mdstore</description>
</property>
<property>
<name>skipUpdate</name>
@ -19,12 +23,31 @@
</property>
</parameters>
<start to="ConvertDataset"/>
<start to="StartTransaction"/>
<kill name="Kill">
<message>Action failed, error message[${wf:errorMessage(wf:lastErrorNode())}]</message>
</kill>
<action name="StartTransaction">
<java>
<configuration>
<property>
<name>oozie.launcher.mapreduce.user.classpath.first</name>
<value>true</value>
</property>
</configuration>
<main-class>eu.dnetlib.dhp.aggregation.mdstore.MDStoreActionNode</main-class>
<arg>--action</arg><arg>NEW_VERSION</arg>
<arg>--mdStoreID</arg><arg>${mdStoreOutputId}</arg>
<arg>--mdStoreManagerURI</arg><arg>${mdStoreManagerURI}</arg>
<capture-output/>
</java>
<ok to="ConvertDataset"/>
<error to="RollBack"/>
</action>
<action name="ConvertDataset">
<spark xmlns="uri:oozie:spark-action:0.2">
<master>yarn</master>
@ -43,16 +66,52 @@
--conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
</spark-opts>
<arg>--workingPath</arg><arg>${baselineWorkingPath}</arg>
<arg>--targetPath</arg><arg>${targetPath}</arg>
<arg>--mdstoreOutputVersion</arg><arg>${wf:actionData('StartTransaction')['mdStoreVersion']}</arg>
<arg>--master</arg><arg>yarn</arg>
<arg>--isLookupUrl</arg><arg>${isLookupUrl}</arg>
<arg>--hdfsServerUri</arg><arg>${nameNode}</arg>
<arg>--skipUpdate</arg><arg>${skipUpdate}</arg>
</spark>
<ok to="End"/>
<error to="Kill"/>
<ok to="CommitVersion"/>
<error to="RollBack"/>
</action>
<end name="End"/>
<action name="CommitVersion">
<java>
<configuration>
<property>
<name>oozie.launcher.mapreduce.user.classpath.first</name>
<value>true</value>
</property>
</configuration>
<main-class>eu.dnetlib.dhp.aggregation.mdstore.MDStoreActionNode</main-class>
<arg>--action</arg><arg>COMMIT</arg>
<arg>--namenode</arg><arg>${nameNode}</arg>
<arg>--mdStoreVersion</arg><arg>${wf:actionData('StartTransaction')['mdStoreVersion']}</arg>
<arg>--mdStoreManagerURI</arg><arg>${mdStoreManagerURI}</arg>
</java>
<ok to="End"/>
<error to="Kill"/>
</action>
<action name="RollBack">
<java>
<configuration>
<property>
<name>oozie.launcher.mapreduce.user.classpath.first</name>
<value>true</value>
</property>
</configuration>
<main-class>eu.dnetlib.dhp.aggregation.mdstore.MDStoreActionNode</main-class>
<arg>--action</arg><arg>ROLLBACK</arg>
<arg>--mdStoreVersion</arg><arg>${wf:actionData('StartTransaction')['mdStoreVersion']}</arg>
<arg>--mdStoreManagerURI</arg><arg>${mdStoreManagerURI}</arg>
</java>
<ok to="Kill"/>
<error to="Kill"/>
</action>
<end name="End"/>
</workflow-app>

18
dhp-workflows/dhp-aggregation/src/main/scala/eu/dnetlib/dhp/sx/bio/ebi/SparkCreateBaselineDataFrame.scala
View File

@ -2,9 +2,12 @@ package eu.dnetlib.dhp.sx.bio.ebi
import eu.dnetlib.dhp.application.ArgumentApplicationParser
import eu.dnetlib.dhp.collection.CollectionUtils
import eu.dnetlib.dhp.common.Constants.{MDSTORE_DATA_PATH, MDSTORE_SIZE_PATH}
import eu.dnetlib.dhp.common.vocabulary.VocabularyGroup
import eu.dnetlib.dhp.schema.mdstore.MDStoreVersion
import eu.dnetlib.dhp.schema.oaf.{Oaf, Result}
import eu.dnetlib.dhp.sx.bio.pubmed._
import eu.dnetlib.dhp.utils.DHPUtils.{MAPPER, writeHdfsFile}
import eu.dnetlib.dhp.utils.ISLookupClientFactory
import org.apache.commons.io.IOUtils
import org.apache.hadoop.conf.Configuration
@ -164,11 +167,15 @@ object SparkCreateBaselineDataFrame {
val workingPath = parser.get("workingPath")
log.info("workingPath: {}", workingPath)
val targetPath = parser.get("targetPath")
log.info("targetPath: {}", targetPath)
val mdstoreOutputVersion = parser.get("mdstoreOutputVersion")
log.info("mdstoreOutputVersion: {}", mdstoreOutputVersion)
val cleanedMdStoreVersion = MAPPER.readValue(mdstoreOutputVersion, classOf[MDStoreVersion])
val outputBasePath = cleanedMdStoreVersion.getHdfsPath
log.info("outputBasePath: {}", outputBasePath)
val hdfsServerUri = parser.get("hdfsServerUri")
log.info("hdfsServerUri: {}", targetPath)
log.info("hdfsServerUri: {}", hdfsServerUri)
val skipUpdate = parser.get("skipUpdate")
log.info("skipUpdate: {}", skipUpdate)
@ -216,8 +223,11 @@ object SparkCreateBaselineDataFrame {
.map(a => PubMedToOaf.convert(a, vocabularies))
.as[Oaf]
.filter(p => p != null),
targetPath
s"$outputBasePath/$MDSTORE_DATA_PATH"
)
val df = spark.read.text(s"$outputBasePath/$MDSTORE_DATA_PATH")
val mdStoreSize = df.count
writeHdfsFile(spark.sparkContext.hadoopConfiguration, s"$mdStoreSize", s"$outputBasePath/$MDSTORE_SIZE_PATH")
}
}