reads the blacklist from the blacklist db and writes it as a set of relations on hdfs

This commit is contained in:
Miriam Baglioni 2020-04-29 17:29:49 +02:00
parent f7695e833c
commit b85ad7012a
4 changed files with 189 additions and 0 deletions

View File

@ -0,0 +1,4 @@
package eu.dnetlib.dhp.schema.common;
public class RelationInverse {
}

View File

@ -0,0 +1,15 @@
<?xml version="1.0" encoding="UTF-8"?>
<project xmlns="http://maven.apache.org/POM/4.0.0"
xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
xsi:schemaLocation="http://maven.apache.org/POM/4.0.0 http://maven.apache.org/xsd/maven-4.0.0.xsd">
<parent>
<artifactId>dhp-workflows</artifactId>
<groupId>eu.dnetlib.dhp</groupId>
<version>1.1.7-SNAPSHOT</version>
</parent>
<modelVersion>4.0.0</modelVersion>
<artifactId>dhp-blacklist</artifactId>
</project>

View File

@ -0,0 +1,138 @@
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;
import java.io.BufferedWriter;
import java.io.Closeable;
import java.io.IOException;
import java.io.OutputStreamWriter;
import java.nio.charset.StandardCharsets;
import java.sql.ResultSet;
import java.util.Arrays;
import java.util.List;
import java.util.function.Consumer;
import java.util.function.Function;
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;
public class ReadBlacklistFromDB implements Closeable {
private final DbClient dbClient;
private static final Log log = LogFactory.getLog(ReadBlacklistFromDB.class);
private final Configuration conf;
private final BufferedWriter writer;
private final static String query = "SELECT source_type, unnest(original_source_objects) as source, " +
"target_type, unnest(original_target_objects) as target, " +
"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(
"/eu/dnetlib/dhp/blacklist/blacklist_parameters.json")));
parser.parseArgument(args);
final String dbUrl = parser.get("postgresUrl");
final String dbUser = parser.get("postgresUser");
final String dbPassword = parser.get("postgresPassword");
final String hdfsPath = parser.get("hdfsPath");
final String hdfsNameNode = parser.get("hdfsNameNode");
try (final ReadBlacklistFromDB rbl =
new ReadBlacklistFromDB(hdfsPath, hdfsNameNode, dbUrl, dbUser, dbPassword)) {
log.info("Processing blacklist...");
rbl.execute(query, rbl::processBlacklistEntry);
}
}
public void execute(final String sql, final Function<ResultSet, List<Relation>> producer)
throws Exception {
final Consumer<ResultSet> consumer = rs -> producer.apply(rs).forEach(r -> writeRelation(r));
dbClient.processResults(sql, consumer);
}
public List<Relation> processBlacklistEntry(ResultSet rs){
try {
Relation direct = new Relation();
Relation inverse = new Relation();
String source_prefix = ModelSupport.entityIdPrefix.get(rs.getString("source_type"));
String target_prefix = ModelSupport.entityIdPrefix.get(rs.getString("target_type"));
String source_direct = source_prefix + "|" + rs.getString("source");
direct.setSource(source_direct);
inverse.setTarget(source_direct);
String target_direct = target_prefix + "|" + rs.getString("target");
direct.setTarget(target_direct);
inverse.setSource(target_direct);
String encoding = rs.getString("relationship");
RelationInverse ri = ModelSupport.relationInverseMap.get(encoding);
direct.setRelClass(ri.getRelation());
inverse.setRelClass(ri.getInverse());
direct.setRelType(ri.getRelType());
inverse.setRelType(ri.getRelType());
direct.setSubRelType(ri.getSubReltype());
inverse.setSubRelType(ri.getSubReltype());
return Arrays.asList(direct, inverse);
} catch (final Exception e) {
throw new RuntimeException(e);
}
}
@Override
public void close() throws IOException {
dbClient.close();
writer.close();
}
public ReadBlacklistFromDB(
final String hdfsPath, String hdfsNameNode, final String dbUrl, final String dbUser, final String dbPassword)
throws Exception {
this.dbClient = new DbClient(dbUrl, dbUser, dbPassword);
this.conf = new Configuration();
this.conf.set("fs.defaultFS", hdfsNameNode);
FileSystem fileSystem = FileSystem.get(this.conf);
Path hdfsWritePath = new Path(hdfsPath);
FSDataOutputStream fsDataOutputStream = fileSystem.append(hdfsWritePath);
this.writer = new BufferedWriter(new OutputStreamWriter(fsDataOutputStream, StandardCharsets.UTF_8));
}
protected void writeRelation(final Relation r) {
try {
writer.write(new ObjectMapper().writeValueAsString(r));
writer.newLine();
} catch (final Exception e) {
throw new RuntimeException(e);
}
}
}

View File

@ -0,0 +1,32 @@
[
{
"paramName": "p",
"paramLongName": "hdfsPath",
"paramDescription": "the path where storing the sequential file",
"paramRequired": true
},
{
"paramName": "pgurl",
"paramLongName": "postgresUrl",
"paramDescription": "postgres url, example: jdbc:postgresql://localhost:5432/testdb",
"paramRequired": true
},
{
"paramName": "pguser",
"paramLongName": "postgresUser",
"paramDescription": "postgres user",
"paramRequired": false
},
{
"paramName": "pgpasswd",
"paramLongName": "postgresPassword",
"paramDescription": "postgres password",
"paramRequired": false
},
{
"paramName": "a",
"paramLongName": "action",
"paramDescription": "process claims",
"paramRequired": false
}
]