forked from D-Net/dnet-hadoop
113 lines
3.4 KiB
Java
113 lines
3.4 KiB
Java
|
|
package eu.dnetlib.dhp.export.zenodo;
|
|
|
|
import java.io.*;
|
|
|
|
import org.apache.commons.compress.archivers.tar.TarArchiveEntry;
|
|
import org.apache.commons.compress.archivers.tar.TarArchiveOutputStream;
|
|
import org.apache.commons.io.IOUtils;
|
|
import org.apache.hadoop.conf.Configuration;
|
|
import org.apache.hadoop.fs.*;
|
|
import org.slf4j.Logger;
|
|
import org.slf4j.LoggerFactory;
|
|
|
|
import eu.dnetlib.dhp.application.ArgumentApplicationParser;
|
|
import eu.dnetlib.dhp.common.MakeTarArchive;
|
|
|
|
public class MakeTar implements Serializable {
|
|
|
|
private static final Logger log = LoggerFactory.getLogger(MakeTar.class);
|
|
|
|
public static void main(String[] args) throws Exception {
|
|
String jsonConfiguration = IOUtils
|
|
.toString(
|
|
MakeTar.class
|
|
.getResourceAsStream(
|
|
"/eu/dnetlib/dhp/export/input_maketar_parameters.json"));
|
|
|
|
final ArgumentApplicationParser parser = new ArgumentApplicationParser(jsonConfiguration);
|
|
parser.parseArgument(args);
|
|
|
|
final String outputPath = parser.get("targetPath");
|
|
log.info("hdfsPath: {}", outputPath);
|
|
|
|
final String hdfsNameNode = parser.get("nameNode");
|
|
log.info("nameNode: {}", hdfsNameNode);
|
|
|
|
final String inputPath = parser.get("sourcePath");
|
|
log.info("input path : {}", inputPath);
|
|
|
|
Configuration conf = new Configuration();
|
|
conf.set("fs.defaultFS", hdfsNameNode);
|
|
|
|
FileSystem fileSystem = FileSystem.get(conf);
|
|
|
|
MakeTarArchive.tarMaxSize(fileSystem, inputPath, outputPath, "scholix_dump", 25);
|
|
|
|
}
|
|
|
|
// public static void makeTArArchive(FileSystem fileSystem, String inputPath, String outputPath) throws IOException {
|
|
//
|
|
// RemoteIterator<LocatedFileStatus> dir_iterator = fileSystem.listLocatedStatus(new Path(inputPath));
|
|
//
|
|
// while (dir_iterator.hasNext()) {
|
|
// LocatedFileStatus fileStatus = dir_iterator.next();
|
|
//
|
|
// Path p = fileStatus.getPath();
|
|
// String p_string = p.toString();
|
|
// String entity = p_string.substring(p_string.lastIndexOf("/") + 1);
|
|
//
|
|
// write(fileSystem, p_string, outputPath + "/" + entity + ".tar", entity);
|
|
// }
|
|
//
|
|
// }
|
|
//
|
|
// private static void write(FileSystem fileSystem, String inputPath, String outputPath, String dir_name)
|
|
// throws IOException {
|
|
//
|
|
// Path hdfsWritePath = new Path(outputPath);
|
|
// FSDataOutputStream fsDataOutputStream = null;
|
|
// if (fileSystem.exists(hdfsWritePath)) {
|
|
// fileSystem.delete(hdfsWritePath, true);
|
|
//
|
|
// }
|
|
// fsDataOutputStream = fileSystem.create(hdfsWritePath);
|
|
//
|
|
// TarArchiveOutputStream ar = new TarArchiveOutputStream(fsDataOutputStream.getWrappedStream());
|
|
//
|
|
// RemoteIterator<LocatedFileStatus> fileStatusListIterator = fileSystem
|
|
// .listFiles(
|
|
// new Path(inputPath), true);
|
|
//
|
|
// while (fileStatusListIterator.hasNext()) {
|
|
// LocatedFileStatus fileStatus = fileStatusListIterator.next();
|
|
//
|
|
// Path p = fileStatus.getPath();
|
|
// String p_string = p.toString();
|
|
// if (!p_string.endsWith("_SUCCESS")) {
|
|
// String name = p_string.substring(p_string.lastIndexOf("/") + 1);
|
|
// TarArchiveEntry entry = new TarArchiveEntry(dir_name + "/" + name + ".json.gz");
|
|
// entry.setSize(fileStatus.getLen());
|
|
// ar.putArchiveEntry(entry);
|
|
//
|
|
// InputStream is = fileSystem.open(fileStatus.getPath());
|
|
//
|
|
// BufferedInputStream bis = new BufferedInputStream(is);
|
|
//
|
|
// int count;
|
|
// byte data[] = new byte[1024];
|
|
// while ((count = bis.read(data, 0, data.length)) != -1) {
|
|
// ar.write(data, 0, count);
|
|
// }
|
|
// bis.close();
|
|
// ar.closeArchiveEntry();
|
|
//
|
|
// }
|
|
//
|
|
// }
|
|
//
|
|
// ar.close();
|
|
// }
|
|
|
|
}
|