dimitris.pierrakos
/
dnet-hadoop
forked from D-Net/dnet-hadoop
1
0
Fork 0
Code Pull Requests Activity

code formatting

This commit is contained in:
Claudio Atzori 2021-10-19 16:19:29 +02:00
parent d517c71458
commit 512e7b0170
2 changed files with 38 additions and 36 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

38
dhp-workflows/dhp-graph-mapper/src/main/java/eu/dnetlib/dhp/oa/graph/raw/CopyHdfsOafApplication.java
View File

@ -1,18 +1,10 @@
package eu.dnetlib.dhp.oa.graph.raw; package eu.dnetlib.dhp.oa.graph.raw;
import com.clearspring.analytics.util.Lists; import static eu.dnetlib.dhp.common.SparkSessionSupport.runWithSparkSession;
import com.fasterxml.jackson.databind.ObjectMapper;
import eu.dnetlib.dhp.application.ArgumentApplicationParser; import java.util.*;
import eu.dnetlib.dhp.common.HdfsSupport;
import eu.dnetlib.dhp.common.vocabulary.VocabularyGroup;
import eu.dnetlib.dhp.oa.graph.raw.common.AbstractMigrationApplication;
import eu.dnetlib.dhp.schema.common.EntityType;
import eu.dnetlib.dhp.schema.common.ModelSupport;
import eu.dnetlib.dhp.schema.oaf.Oaf;
import eu.dnetlib.dhp.schema.oaf.Relation;
import eu.dnetlib.dhp.utils.ISLookupClientFactory;
import eu.dnetlib.enabling.is.lookup.rmi.ISLookUpService;
import org.apache.commons.io.IOUtils; import org.apache.commons.io.IOUtils;
import org.apache.spark.SparkConf; import org.apache.spark.SparkConf;
import org.apache.spark.api.java.JavaSparkContext; import org.apache.spark.api.java.JavaSparkContext;
@ -26,9 +18,19 @@ import org.apache.spark.sql.SparkSession;
import org.slf4j.Logger; import org.slf4j.Logger;
import org.slf4j.LoggerFactory; import org.slf4j.LoggerFactory;
import java.util.*; import com.clearspring.analytics.util.Lists;
import com.fasterxml.jackson.databind.ObjectMapper;
import static eu.dnetlib.dhp.common.SparkSessionSupport.runWithSparkSession; import eu.dnetlib.dhp.application.ArgumentApplicationParser;
import eu.dnetlib.dhp.common.HdfsSupport;
import eu.dnetlib.dhp.common.vocabulary.VocabularyGroup;
import eu.dnetlib.dhp.oa.graph.raw.common.AbstractMigrationApplication;
import eu.dnetlib.dhp.schema.common.EntityType;
import eu.dnetlib.dhp.schema.common.ModelSupport;
import eu.dnetlib.dhp.schema.oaf.Oaf;
import eu.dnetlib.dhp.schema.oaf.Relation;
import eu.dnetlib.dhp.utils.ISLookupClientFactory;
import eu.dnetlib.enabling.is.lookup.rmi.ISLookUpService;
public class CopyHdfsOafApplication extends AbstractMigrationApplication { public class CopyHdfsOafApplication extends AbstractMigrationApplication {
@ -96,12 +98,12 @@ public class CopyHdfsOafApplication extends AbstractMigrationApplication {
if (validPaths.length > 0) { if (validPaths.length > 0) {
// load the dataset // load the dataset
Dataset<Oaf> oaf = spark Dataset<Oaf> oaf = spark
.read() .read()
.load(validPaths) .load(validPaths)
.as(Encoders.kryo(Oaf.class)); .as(Encoders.kryo(Oaf.class));
// dispatch each entity type individually in the respective graph subdirectory in append mode // dispatch each entity type individually in the respective graph subdirectory in append mode
for(Map.Entry<String, Class> e : ModelSupport.oafTypes.entrySet()) { for (Map.Entry<String, Class> e : ModelSupport.oafTypes.entrySet()) {
oaf oaf
.filter((FilterFunction<Oaf>) o -> o.getClass().getSimpleName().toLowerCase().equals(e.getKey())) .filter((FilterFunction<Oaf>) o -> o.getClass().getSimpleName().toLowerCase().equals(e.getKey()))
.map((MapFunction<Oaf, String>) OBJECT_MAPPER::writeValueAsString, Encoders.bean(e.getValue())) .map((MapFunction<Oaf, String>) OBJECT_MAPPER::writeValueAsString, Encoders.bean(e.getValue()))

36
dhp-workflows/dhp-graph-mapper/src/main/java/eu/dnetlib/dhp/oa/graph/raw/common/AbstractMigrationApplication.java
View File

@ -8,7 +8,6 @@ import java.util.Set;
import java.util.concurrent.atomic.AtomicInteger; import java.util.concurrent.atomic.AtomicInteger;
import java.util.stream.Collectors; import java.util.stream.Collectors;
import eu.dnetlib.dhp.schema.mdstore.MDStoreWithInfo;
import org.apache.commons.io.IOUtils; import org.apache.commons.io.IOUtils;
import org.apache.commons.lang3.StringUtils; import org.apache.commons.lang3.StringUtils;
import org.apache.commons.logging.Log; import org.apache.commons.logging.Log;
@ -17,16 +16,17 @@ import org.apache.hadoop.conf.Configuration;
import org.apache.hadoop.fs.Path; import org.apache.hadoop.fs.Path;
import org.apache.hadoop.io.SequenceFile; import org.apache.hadoop.io.SequenceFile;
import org.apache.hadoop.io.Text; import org.apache.hadoop.io.Text;
import com.fasterxml.jackson.core.JsonProcessingException;
import com.fasterxml.jackson.databind.ObjectMapper;
import eu.dnetlib.dhp.schema.oaf.Oaf;
import org.apache.http.client.methods.CloseableHttpResponse; import org.apache.http.client.methods.CloseableHttpResponse;
import org.apache.http.client.methods.HttpGet; import org.apache.http.client.methods.HttpGet;
import org.apache.http.impl.client.CloseableHttpClient; import org.apache.http.impl.client.CloseableHttpClient;
import org.apache.http.impl.client.HttpClients; import org.apache.http.impl.client.HttpClients;
import com.fasterxml.jackson.core.JsonProcessingException;
import com.fasterxml.jackson.databind.ObjectMapper;
import eu.dnetlib.dhp.schema.mdstore.MDStoreWithInfo;
import eu.dnetlib.dhp.schema.oaf.Oaf;
public class AbstractMigrationApplication implements Closeable { public class AbstractMigrationApplication implements Closeable {
private final AtomicInteger counter = new AtomicInteger(0); private final AtomicInteger counter = new AtomicInteger(0);
@ -68,9 +68,9 @@ public class AbstractMigrationApplication implements Closeable {
* @throws IOException in case of HTTP communication issues * @throws IOException in case of HTTP communication issues
*/ */
protected static Set<String> mdstorePaths(final String mdstoreManagerUrl, protected static Set<String> mdstorePaths(final String mdstoreManagerUrl,
final String format, final String format,
final String layout, final String layout,
final String interpretation) throws IOException { final String interpretation) throws IOException {
final String url = mdstoreManagerUrl + "/mdstores/"; final String url = mdstoreManagerUrl + "/mdstores/";
final ObjectMapper objectMapper = new ObjectMapper(); final ObjectMapper objectMapper = new ObjectMapper();
@ -81,15 +81,15 @@ public class AbstractMigrationApplication implements Closeable {
final String json = IOUtils.toString(response.getEntity().getContent()); final String json = IOUtils.toString(response.getEntity().getContent());
final MDStoreWithInfo[] mdstores = objectMapper.readValue(json, MDStoreWithInfo[].class); final MDStoreWithInfo[] mdstores = objectMapper.readValue(json, MDStoreWithInfo[].class);
return Arrays return Arrays
.stream(mdstores) .stream(mdstores)
.filter(md -> md.getFormat().equalsIgnoreCase(format)) .filter(md -> md.getFormat().equalsIgnoreCase(format))
.filter(md -> md.getLayout().equalsIgnoreCase(layout)) .filter(md -> md.getLayout().equalsIgnoreCase(layout))
.filter(md -> md.getInterpretation().equalsIgnoreCase(interpretation)) .filter(md -> md.getInterpretation().equalsIgnoreCase(interpretation))
.filter(md -> StringUtils.isNotBlank(md.getHdfsPath())) .filter(md -> StringUtils.isNotBlank(md.getHdfsPath()))
.filter(md -> StringUtils.isNotBlank(md.getCurrentVersion())) .filter(md -> StringUtils.isNotBlank(md.getCurrentVersion()))
.filter(md -> md.getSize() > 0) .filter(md -> md.getSize() > 0)
.map(md -> md.getHdfsPath() + "/" + md.getCurrentVersion() + "/store") .map(md -> md.getHdfsPath() + "/" + md.getCurrentVersion() + "/store")
.collect(Collectors.toSet()); .collect(Collectors.toSet());
} }
} }
} }