Automatic download of Crossref #138

Merged
claudio.atzori merged 7 commits from crossref_dw_wf into beta 2021-08-13 17:10:11 +02:00
4 changed files with 73 additions and 82 deletions

View File

@ -29,16 +29,16 @@ public class ExtractCrossrefRecords {
"/eu/dnetlib/dhp/doiboost/crossref_dump_reader.json")));
parser.parseArgument(args);
final String hdfsServerUri = parser.get("hdfsServerUri");
final String workingPath = parser.get("workingPath");
final String workingPath = hdfsServerUri.concat(parser.get("workingPath"));
final String outputPath = parser.get("outputPath");
final String crossrefFileNameTarGz = parser.get("crossrefFileNameTarGz");
Path hdfsreadpath = new Path(hdfsServerUri.concat(crossrefFileNameTarGz));
Path hdfsreadpath = new Path(workingPath.concat("/").concat(crossrefFileNameTarGz));
Configuration conf = new Configuration();
conf.set("fs.defaultFS", hdfsServerUri.concat(workingPath));
conf.set("fs.defaultFS", workingPath);
conf.set("fs.hdfs.impl", org.apache.hadoop.hdfs.DistributedFileSystem.class.getName());
conf.set("fs.file.impl", org.apache.hadoop.fs.LocalFileSystem.class.getName());
FileSystem fs = FileSystem.get(URI.create(hdfsServerUri.concat(workingPath)), conf);
FileSystem fs = FileSystem.get(URI.create(workingPath), conf);
FSDataInputStream crossrefFileStream = fs.open(hdfsreadpath);
try (TarArchiveInputStream tais = new TarArchiveInputStream(
new GzipCompressorInputStream(crossrefFileStream))) {

View File

@ -1,21 +0,0 @@
[
{
"paramName": "s",
"paramLongName": "sourcePath",
"paramDescription": "the source mdstore path",
"paramRequired": true
},
{
"paramName": "t",
"paramLongName": "targetPath",
"paramDescription": "the target mdstore path",
"paramRequired": true
},
{
"paramName": "m",
"paramLongName": "master",
"paramDescription": "the master name",
"paramRequired": true
}
]

View File

@ -0,0 +1,2 @@
#!/bin/bash
curl -LSs -H "Crossref-Plus-API-Token: Bearer $4" $1 | hdfs dfs -put - $2/$3

View File

@ -1,13 +1,5 @@
<workflow-app name="read Crossref dump from HDFS" xmlns="uri:oozie:workflow:0.5">
<workflow-app name="CROSSREF DUMP - DOWNLOAD" xmlns="uri:oozie:workflow:0.5">
<parameters>
<property>
<name>crossrefDumpPath</name>
<description>the working dir base path</description>
</property>
<property>
<name>inputPathCrossref</name>
<description>the working dir base path</description>
</property>
<property>
<name>sparkDriverMemory</name>
<description>memory for driver process</description>
@ -18,27 +10,87 @@
</property>
<property>
<name>sparkExecutorCores</name>
<value>2</value>
<description>number of cores used by single executor</description>
</property>
<!-- Crossref Parameters -->
<property>
<name>crossrefdumpfilename</name>
<description>the Crossref input path</description>
</property>
<property>
<name>crossrefDumpPath</name>
<description>the Crossref dump path</description>
</property>
<property>
<name>crossrefdumptoken</name>
<description>the token for the API dump path</description>
</property>
</parameters>
<start to="GenerateCrossrefDataset"/>
<global>
<job-tracker>${jobTracker}</job-tracker>
<name-node>${nameNode}</name-node>
<configuration>
<property>
<name>oozie.action.sharelib.for.spark</name>
<value>${oozieActionShareLibForSpark2}</value>
</property>
</configuration>
</global>
<start to="removeFiles"/>
<kill name="Kill">
<message>Action failed, error message[${wf:errorMessage(wf:lastErrorNode())}]</message>
</kill>
<action name="removeFiles">
<fs>
<delete path="${crossrefDumpPath}/${crossrefdumpfilename}"/>
<delete path="${crossrefDumpPath}/files"/>
<delete path="${crossrefDumpPath}/crossref_unpack"/>
</fs>
<ok to="DownloadDump"/>
<error to="Kill"/>
</action>
<action name="DownloadDump">
<shell xmlns="uri:oozie:shell-action:0.2">
<job-tracker>${jobTracker}</job-tracker>
<name-node>${nameNode}</name-node>
<configuration>
<property>
<name>mapred.job.queue.name</name>
<value>${queueName}</value>
</property>
</configuration>
<exec>download.sh</exec>
<argument>${url}</argument>
<argument>${crossrefDumpPath}</argument>
<argument>${crossrefdumpfilename}</argument>
<argument>${crossrefdumptoken}</argument>
<env-var>HADOOP_USER_NAME=${wf:user()}</env-var>
<file>download.sh</file>
<capture-output/>
</shell>
<ok to="ImportCrossRef"/>
<error to="Kill"/>
</action>
<action name="ImportCrossRef">
<java>
<job-tracker>${jobTracker}</job-tracker>
<name-node>${nameNode}</name-node>
<main-class>eu.dnetlib.doiboost.crossref.ExtractCrossrefRecords</main-class>
<arg>--hdfsServerUri</arg><arg>${nameNode}</arg>
<arg>--crossrefFileNameTarGz</arg><arg>${crossrefDumpPath}/crossref.tar.gz</arg>
<arg>--crossrefFileNameTarGz</arg><arg>${crossrefdumpfilename}</arg>
<arg>--workingPath</arg><arg>${crossrefDumpPath}</arg>
<arg>--outputPath</arg><arg>${workingDir}/files/</arg>
<arg>--outputPath</arg><arg>${crossrefDumpPath}/files/</arg>
</java>
<ok to="UnpackCrossrefEntries"/>
<error to="Kill"/>
@ -48,7 +100,7 @@
<spark xmlns="uri:oozie:spark-action:0.2">
<master>yarn-cluster</master>
<mode>cluster</mode>
<name>SparkGenerateCrossrefDataset</name>
<name>SparkUnpackCrossrefEntries</name>
<class>eu.dnetlib.doiboost.crossref.UnpackCrtossrefEntries</class>
<jar>dhp-doiboost-${projectVersion}.jar</jar>
<spark-opts>
@ -63,56 +115,14 @@
</spark-opts>
<arg>--master</arg><arg>yarn-cluster</arg>
<arg>--sourcePath</arg><arg>${crossrefDumpPath}/files</arg>
<arg>--targetPath</arg><arg>${inputPathCrossref}/crossref_ds</arg>
</spark>
<ok to="GenerateCrossrefDataset"/>
<error to="Kill"/>
</action>
<action name="GenerateCrossrefDataset">
<spark xmlns="uri:oozie:spark-action:0.2">
<master>yarn-cluster</master>
<mode>cluster</mode>
<name>SparkGenerateCrossrefDataset</name>
<class>eu.dnetlib.doiboost.crossref.GenerateCrossrefDataset</class>
<jar>dhp-doiboost-${projectVersion}.jar</jar>
<spark-opts>
--executor-memory=${sparkExecutorMemory}
--executor-cores=${sparkExecutorCores}
--driver-memory=${sparkDriverMemory}
--conf spark.sql.shuffle.partitions=3840
--conf spark.extraListeners=${spark2ExtraListeners}
--conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
--conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
--conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
</spark-opts>
<arg>--master</arg><arg>yarn-cluster</arg>
<arg>--sourcePath</arg><arg>${inputPathCrossref}/crossref_ds</arg>
<arg>--targetPath</arg><arg>${inputPathCrossref}/crossref_ds_updates</arg>
<arg>--targetPath</arg><arg>${crossrefDumpPath}/crossref_unpack/</arg>
</spark>
<ok to="End"/>
<error to="Kill"/>
</action>
<!-- <action name="removeFiles">-->
<!-- <fs>-->
<!-- <delete path="${workingDir}/files"/>-->
<!-- </fs>-->
<!-- <ok to="renameDataset"/>-->
<!-- <error to="Kill"/>-->
<!-- </action>-->
<!-- <action name="renameDataset">-->
<!-- <fs>-->
<!-- <delete path="${inputPathCrossref}/crossref_ds"/>-->
<!-- <move source="${inputPathCrossref}/crossref_ds_updated"-->
<!-- target="${inputPathCrossref}/crossref_ds"/>-->
<!-- </fs>-->
<!-- <ok to="End"/>-->
<!-- <error to="Kill"/>-->
<!-- </action>-->
<end name="End"/>
</workflow-app>