lsmyrnaios
/
dnet-hadoop
forked from D-Net/dnet-hadoop
1
0
Fork 0
Code Pull Requests Activity

cleanup

This commit is contained in:
Claudio Atzori 2020-04-18 12:24:22 +02:00
parent 9374ff03ea
commit a2938dd059
3 changed files with 4 additions and 12 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

6
dhp-workflows/dhp-dedup-openaire/src/main/java/eu/dnetlib/dhp/oa/dedup/DedupRecordFactory.java
View File

@ -27,15 +27,15 @@ public class DedupRecordFactory {
.configure(DeserializationFeature.FAIL_ON_UNKNOWN_PROPERTIES, false); .configure(DeserializationFeature.FAIL_ON_UNKNOWN_PROPERTIES, false);
public static <T extends OafEntity> Dataset<T> createDedupRecord( public static <T extends OafEntity> Dataset<T> createDedupRecord(
final SparkSession spark, final String mergeRelsInputPath, final String entitiesInputPath, final Class<T> clazz, final DedupConfig dedupConf) { final SparkSession spark, final String mergeRelsInputPath, final String entitiesInputPath, final Class<T> clazz) {
long ts = System.currentTimeMillis(); long ts = System.currentTimeMillis();
//<id, json_entity> //<id, json_entity>
Dataset<Tuple2<String, T>> entities = spark.read() Dataset<Tuple2<String, T>> entities = spark.read()
.textFile(entitiesInputPath) .textFile(entitiesInputPath)
.map((MapFunction<String, Tuple2<String, T>>) it -> { .map((MapFunction<String, Tuple2<String, T>>) s -> {
T entity = OBJECT_MAPPER.readValue(it, clazz); T entity = OBJECT_MAPPER.readValue(s, clazz);
return new Tuple2<>(entity.getId(), entity); return new Tuple2<>(entity.getId(), entity);
}, Encoders.tuple(Encoders.STRING(), Encoders.kryo(clazz))); }, Encoders.tuple(Encoders.STRING(), Encoders.kryo(clazz)));

5
dhp-workflows/dhp-dedup-openaire/src/main/java/eu/dnetlib/dhp/oa/dedup/SparkCreateDedupRecord.java
View File

@ -10,10 +10,7 @@ import eu.dnetlib.enabling.is.lookup.rmi.ISLookUpService;
import eu.dnetlib.pace.config.DedupConfig; import eu.dnetlib.pace.config.DedupConfig;
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.JavaRDD;
import org.apache.spark.api.java.JavaSparkContext;
import org.apache.spark.api.java.function.MapFunction; import org.apache.spark.api.java.function.MapFunction;
import org.apache.spark.sql.Dataset;
import org.apache.spark.sql.Encoders; import org.apache.spark.sql.Encoders;
import org.apache.spark.sql.SaveMode; import org.apache.spark.sql.SaveMode;
import org.apache.spark.sql.SparkSession; import org.apache.spark.sql.SparkSession;
@ -70,7 +67,7 @@ public class SparkCreateDedupRecord extends AbstractSparkAction {
Class<OafEntity> clazz = ModelSupport.entityTypes.get(EntityType.valueOf(subEntity)); Class<OafEntity> clazz = ModelSupport.entityTypes.get(EntityType.valueOf(subEntity));
DedupRecordFactory.createDedupRecord(spark, mergeRelPath, entityPath, clazz, dedupConf) DedupRecordFactory.createDedupRecord(spark, mergeRelPath, entityPath, clazz)
.map((MapFunction<OafEntity, String>) value -> OBJECT_MAPPER.writeValueAsString(value), Encoders.STRING()) .map((MapFunction<OafEntity, String>) value -> OBJECT_MAPPER.writeValueAsString(value), Encoders.STRING())
.write() .write()
.mode(SaveMode.Overwrite) .mode(SaveMode.Overwrite)

5
dhp-workflows/dhp-dedup-openaire/src/test/java/eu/dnetlib/dhp/oa/dedup/SparkDedupTest.java
View File

@ -95,7 +95,6 @@ public class SparkDedupTest implements Serializable {
IOUtils.toString( IOUtils.toString(
SparkCreateSimRels.class.getResourceAsStream("/eu/dnetlib/dhp/oa/dedup/createSimRels_parameters.json"))); SparkCreateSimRels.class.getResourceAsStream("/eu/dnetlib/dhp/oa/dedup/createSimRels_parameters.json")));
parser.parseArgument(new String[]{ parser.parseArgument(new String[]{
"-mt", "local[*]",
"-i", testGraphBasePath, "-i", testGraphBasePath,
"-asi", testActionSetId, "-asi", testActionSetId,
"-la", "lookupurl", "-la", "lookupurl",
@ -120,7 +119,6 @@ public class SparkDedupTest implements Serializable {
IOUtils.toString( IOUtils.toString(
SparkCreateMergeRels.class.getResourceAsStream("/eu/dnetlib/dhp/oa/dedup/createCC_parameters.json"))); SparkCreateMergeRels.class.getResourceAsStream("/eu/dnetlib/dhp/oa/dedup/createCC_parameters.json")));
parser.parseArgument(new String[]{ parser.parseArgument(new String[]{
"-mt", "local[*]",
"-i", testGraphBasePath, "-i", testGraphBasePath,
"-asi", testActionSetId, "-asi", testActionSetId,
"-la", "lookupurl", "-la", "lookupurl",
@ -145,7 +143,6 @@ public class SparkDedupTest implements Serializable {
IOUtils.toString( IOUtils.toString(
SparkCreateDedupRecord.class.getResourceAsStream("/eu/dnetlib/dhp/oa/dedup/createDedupRecord_parameters.json"))); SparkCreateDedupRecord.class.getResourceAsStream("/eu/dnetlib/dhp/oa/dedup/createDedupRecord_parameters.json")));
parser.parseArgument(new String[]{ parser.parseArgument(new String[]{
"-mt", "local[*]",
"-i", testGraphBasePath, "-i", testGraphBasePath,
"-asi", testActionSetId, "-asi", testActionSetId,
"-la", "lookupurl", "-la", "lookupurl",
@ -170,7 +167,6 @@ public class SparkDedupTest implements Serializable {
IOUtils.toString( IOUtils.toString(
SparkUpdateEntity.class.getResourceAsStream("/eu/dnetlib/dhp/oa/dedup/updateEntity_parameters.json"))); SparkUpdateEntity.class.getResourceAsStream("/eu/dnetlib/dhp/oa/dedup/updateEntity_parameters.json")));
parser.parseArgument(new String[]{ parser.parseArgument(new String[]{
"-mt", "local[*]",
"-i", testGraphBasePath, "-i", testGraphBasePath,
"-w", testOutputBasePath, "-w", testOutputBasePath,
"-o", testDedupGraphBasePath "-o", testDedupGraphBasePath
@ -221,7 +217,6 @@ public class SparkDedupTest implements Serializable {
IOUtils.toString( IOUtils.toString(
SparkPropagateRelation.class.getResourceAsStream("/eu/dnetlib/dhp/oa/dedup/propagateRelation_parameters.json"))); SparkPropagateRelation.class.getResourceAsStream("/eu/dnetlib/dhp/oa/dedup/propagateRelation_parameters.json")));
parser.parseArgument(new String[]{ parser.parseArgument(new String[]{
"-mt", "local[*]",
"-i", testGraphBasePath, "-i", testGraphBasePath,
"-w", testOutputBasePath, "-w", testOutputBasePath,
"-o", testDedupGraphBasePath "-o", testDedupGraphBasePath