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

changed message in log

This commit is contained in:
Miriam Baglioni 2020-05-29 10:41:39 +02:00
parent 8b6e886fb6
commit 6f1eea28b6
1 changed files with 81 additions and 79 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

34
dhp-workflows/dhp-aggregation/src/main/java/eu/dnetlib/dhp/actionmanager/project/ReadProjectsFromDB.java
View File

@ -1,15 +1,5 @@
package eu.dnetlib.dhp.actionmanager.project;
import com.fasterxml.jackson.databind.ObjectMapper;
import eu.dnetlib.dhp.application.ArgumentApplicationParser;
import eu.dnetlib.dhp.common.DbClient;
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 org.apache.hadoop.fs.FSDataOutputStream;
import org.apache.hadoop.fs.FileSystem;
import org.apache.hadoop.fs.Path;
package eu.dnetlib.dhp.actionmanager.project;
import java.io.BufferedWriter;
import java.io.Closeable;
@ -22,6 +12,19 @@ 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 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.common.DbClient;
public class ReadProjectsFromDB implements Closeable {
private final DbClient dbClient;
@ -31,7 +34,7 @@ public class ReadProjectsFromDB implements Closeable {
private final ObjectMapper OBJECT_MAPPER = new ObjectMapper();
private final static String query = "SELECT code " +
"from projects where id like 'corda__h2020%' " ;
"from projects where id like 'corda__h2020%' ";
public static void main(final String[] args) throws Exception {
final ArgumentApplicationParser parser = new ArgumentApplicationParser(
@ -46,17 +49,18 @@ public class ReadProjectsFromDB implements Closeable {
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 hdfsPath = parser.get("hdfsPath");
final String hdfsNameNode = parser.get("hdfsNameNode");
try (final ReadProjectsFromDB rbl = new ReadProjectsFromDB(hdfsPath, hdfsNameNode, dbUrl, dbUser,
dbPassword)) {
log.info("Processing blacklist...");
log.info("Processing projects...");
rbl.execute(query, rbl::processProjectsEntry);
}
}
public void execute(final String sql, final Function<ResultSet, List<ProjectSubset>> producer) throws Exception {
final Consumer<ResultSet> consumer = rs -> producer.apply(rs).forEach(r -> writeProject(r));
@ -100,7 +104,6 @@ public class ReadProjectsFromDB implements Closeable {
}
fsDataOutputStream = fileSystem.create(hdfsWritePath);
this.writer = new BufferedWriter(new OutputStreamWriter(fsDataOutputStream, StandardCharsets.UTF_8));
}
@ -110,4 +113,3 @@ public class ReadProjectsFromDB implements Closeable {
writer.close();
}
}