D-Net
/
dnet-hadoop
14
5
Fork 11
Code Issues 14 Pull Requests 13 Projects 2 Releases 2 Wiki Activity

OpenAIRE graph dump #51

Merged
claudio.atzori merged 5 commits from miriam.baglioni/dnet-hadoop:dump into master 2020-11-05 17:09:53 +01:00
Conversation 0 Commits 5 Files Changed 3 +11 -12
4 changed files with 11 additions and 12 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files
Showing only changes of commit be5ed8f554 - Show all commits

13
dhp-workflows/dhp-graph-mapper/src/main/java/eu/dnetlib/dhp/oa/graph/dump/MakeTar.java
View File

@ -4,7 +4,6 @@ package eu.dnetlib.dhp.oa.graph.dump;
import java.io.*;
import java.util.Optional;
import eu.dnetlib.dhp.common.MakeTarArchive;
import org.apache.commons.compress.archivers.ar.ArArchiveEntry;
import org.apache.commons.compress.archivers.ar.ArArchiveOutputStream;
import org.apache.commons.compress.archivers.tar.TarArchiveEntry;
@ -16,6 +15,7 @@ import org.slf4j.Logger;
import org.slf4j.LoggerFactory;
import eu.dnetlib.dhp.application.ArgumentApplicationParser;
import eu.dnetlib.dhp.common.MakeTarArchive;
import eu.dnetlib.dhp.oa.graph.dump.community.CommunityMap;
public class MakeTar implements Serializable {
@ -41,9 +41,10 @@ public class MakeTar implements Serializable {
final String inputPath = parser.get("sourcePath");
log.info("input path : {}", inputPath);
final int gBperSplit = Optional.ofNullable(parser.get("splitSize"))
.map(Integer::valueOf)
.orElse(10);
final int gBperSplit = Optional
.ofNullable(parser.get("splitSize"))
.map(Integer::valueOf)
.orElse(10);
Configuration conf = new Configuration();
conf.set("fs.defaultFS", hdfsNameNode);
@ -54,7 +55,8 @@ public class MakeTar implements Serializable {
}
public static void makeTArArchive(FileSystem fileSystem, String inputPath, String outputPath, int gBperSplit) throws IOException {
public static void makeTArArchive(FileSystem fileSystem, String inputPath, String outputPath, int gBperSplit)
throws IOException {
RemoteIterator<LocatedFileStatus> dir_iterator = fileSystem.listLocatedStatus(new Path(inputPath));
@ -70,5 +72,4 @@ public class MakeTar implements Serializable {
}
}

4
dhp-workflows/dhp-graph-mapper/src/main/java/eu/dnetlib/dhp/oa/graph/dump/SendToZenodoHDFS.java
View File

@ -98,8 +98,10 @@ public class SendToZenodoHDFS implements Serializable {
}
}
if (!metadata.equals("")) {
zenodoApiClient.sendMretadata(metadata);
}
zenodoApiClient.sendMretadata(metadata);
if (publish)
zenodoApiClient.publish();

4
dhp-workflows/dhp-graph-mapper/src/main/java/eu/dnetlib/dhp/oa/graph/dump/Utils.java
View File

@ -5,7 +5,6 @@ import java.io.BufferedReader;
import java.io.IOException;
import java.io.InputStreamReader;
import org.apache.hadoop.fs.FileSystem;
import org.apache.hadoop.fs.Path;
import org.apache.spark.api.java.function.MapFunction;
@ -13,14 +12,12 @@ import org.apache.spark.sql.Dataset;
import org.apache.spark.sql.Encoders;
import org.apache.spark.sql.SparkSession;
import com.fasterxml.jackson.databind.ObjectMapper;
import com.google.gson.Gson;
import eu.dnetlib.dhp.common.HdfsSupport;
import eu.dnetlib.dhp.oa.graph.dump.community.CommunityMap;
import eu.dnetlib.dhp.oa.graph.dump.graph.Constants;
import eu.dnetlib.dhp.utils.DHPUtils;
import eu.dnetlib.dhp.utils.ISLookupClientFactory;
import eu.dnetlib.enabling.is.lookup.rmi.ISLookUpService;
@ -74,5 +71,4 @@ public class Utils {
return new Gson().fromJson(sb.toString(), CommunityMap.class);
}
}

2
dhp-workflows/dhp-graph-mapper/src/main/resources/eu/dnetlib/dhp/oa/graph/dump/oozie_app/workflow.xml
View File

@ -418,7 +418,7 @@
<arg>--metadata</arg><arg>${metadata}</arg>
<arg>--communityMapPath</arg><arg>${workingDir}/communityMap</arg>
<arg>--conceptRecordId</arg><arg>${conceptRecordId}</arg>
<arg>--newDeposition</arg><arg>${newDeposition}</arg>
<arg>--depositionType</arg><arg>${depositionType}</arg>
</java>
<ok to="End"/>
<error to="Kill"/>