enrico.ottonello
/
dnet-hadoop
forked from D-Net/dnet-hadoop
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity

changes in the blacklist and workflow definition

This commit is contained in:
Miriam Baglioni 2020-04-30 10:26:50 +02:00
parent 564e5d6279
commit 354f0162be
2 changed files with 109 additions and 99 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

43
dhp-workflows/dhp-blacklist/src/main/java/eu/dnetlib/dhp/blacklist/ReadBlacklistFromDB.java
View File

@ -1,13 +1,5 @@
package eu.dnetlib.dhp.blacklist;
import eu.dnetlib.dhp.application.ArgumentApplicationParser;
import eu.dnetlib.dhp.oa.graph.raw.common.DbClient;
import eu.dnetlib.dhp.schema.common.ModelSupport;
import eu.dnetlib.dhp.schema.common.RelationInverse;
import eu.dnetlib.dhp.schema.oaf.Relation;
import org.apache.commons.io.IOUtils;
import org.apache.commons.logging.Log;
import org.apache.commons.logging.LogFactory;
package eu.dnetlib.dhp.blacklist;
import java.io.BufferedWriter;
import java.io.Closeable;
@ -19,12 +11,22 @@ import java.util.Arrays;
import java.util.List;
import java.util.function.Consumer;
import java.util.function.Function;
import org.apache.commons.io.IOUtils;
import org.apache.commons.logging.Log;
import org.apache.commons.logging.LogFactory;
import org.apache.hadoop.conf.Configuration;
import com.fasterxml.jackson.databind.ObjectMapper;
import org.apache.hadoop.fs.FSDataOutputStream;
import org.apache.hadoop.fs.FileSystem;
import org.apache.hadoop.fs.Path;
import com.fasterxml.jackson.databind.ObjectMapper;
import eu.dnetlib.dhp.application.ArgumentApplicationParser;
import eu.dnetlib.dhp.oa.graph.raw.common.DbClient;
import eu.dnetlib.dhp.schema.common.ModelSupport;
import eu.dnetlib.dhp.schema.common.RelationInverse;
import eu.dnetlib.dhp.schema.oaf.Relation;
public class ReadBlacklistFromDB implements Closeable {
@ -38,10 +40,11 @@ public class ReadBlacklistFromDB implements Closeable {
"relationship FROM blacklist WHERE status = 'ACCEPTED'";
public static void main(final String[] args) throws Exception {
final ArgumentApplicationParser parser =
new ArgumentApplicationParser(
IOUtils.toString(
ReadBlacklistFromDB.class.getResourceAsStream(
final ArgumentApplicationParser parser = new ArgumentApplicationParser(
IOUtils
.toString(
ReadBlacklistFromDB.class
.getResourceAsStream(
"/eu/dnetlib/dhp/blacklist/blacklist_parameters.json")));
parser.parseArgument(args);
@ -52,10 +55,8 @@ public class ReadBlacklistFromDB implements Closeable {
final String hdfsPath = parser.get("hdfsPath");
final String hdfsNameNode = parser.get("hdfsNameNode");
try (final ReadBlacklistFromDB rbl =
new ReadBlacklistFromDB(hdfsPath, hdfsNameNode, dbUrl, dbUser, dbPassword)) {
try (final ReadBlacklistFromDB rbl = new ReadBlacklistFromDB(hdfsPath, hdfsNameNode, dbUrl, dbUser,
dbPassword)) {
log.info("Processing blacklist...");
rbl.execute(query, rbl::processBlacklistEntry);
@ -63,7 +64,6 @@ public class ReadBlacklistFromDB implements Closeable {
}
}
public void execute(final String sql, final Function<ResultSet, List<Relation>> producer)
throws Exception {
@ -72,7 +72,7 @@ public class ReadBlacklistFromDB implements Closeable {
dbClient.processResults(sql, consumer);
}
public List<Relation> processBlacklistEntry(ResultSet rs){
public List<Relation> processBlacklistEntry(ResultSet rs) {
try {
Relation direct = new Relation();
Relation inverse = new Relation();
@ -132,7 +132,4 @@ public class ReadBlacklistFromDB implements Closeable {
}
}
}

21
dhp-workflows/dhp-blacklist/src/main/resources/eu/dnetlib/dhp/blacklist/oozie_app/workflow.xml
View File

@ -1,5 +1,18 @@
<workflow-app name="blacklsting" xmlns="uri:oozie:workflow:0.5">
<workflow-app name="blacklisting" xmlns="uri:oozie:workflow:0.5">
<parameters>
<property>
<name>postgresURL</name>
<description>the url of the postgress server to query</description>
</property>
<property>
<name>postgresUser</name>
<description>the username to access the postgres db</description>
</property>
<property>
<name>postgresPassword</name>
<description>the postgres password</description>
</property>
</parameters>
<start to="reset-outputpath"/>
<kill name="Kill">
@ -8,7 +21,7 @@
<action name="reset-outputpath">
<fs>
<delete path='${hdfsPath}'/>
<delete path='${workingDir}/blacklist'/>
</fs>
<ok to="read_blacklist"/>
<error to="Kill"/>
@ -21,7 +34,7 @@
<main-class>eu.dnetlib.dhp.blacklist.ReadBlacklistFromDB</main-class>
<arg>--hdfsPath</arg><arg>${workingDir}/blacklist</arg>
<arg>--hdfsNameNode</arg><arg>${nameNode}</arg>
<arg>--postgresUrl</arg><arg>${postgresUrl}</arg>
<arg>--postgresUrl</arg><arg>${postgresURL}</arg>
<arg>--postgresUser</arg><arg>${postgresUser}</arg>
<arg>--postgresPassword</arg><arg>${postgresPassword}</arg>
</java>