2020-07-13 19:26:46 +02:00
|
|
|
|
2020-07-13 18:22:34 +02:00
|
|
|
package eu.dnetlib.dhp.oa.dedup;
|
|
|
|
|
|
|
|
import java.io.IOException;
|
|
|
|
|
|
|
|
import org.apache.commons.io.IOUtils;
|
|
|
|
import org.apache.spark.SparkConf;
|
|
|
|
import org.apache.spark.api.java.JavaPairRDD;
|
|
|
|
import org.apache.spark.api.java.JavaRDD;
|
|
|
|
import org.apache.spark.api.java.JavaSparkContext;
|
|
|
|
import org.apache.spark.api.java.function.PairFunction;
|
|
|
|
import org.apache.spark.sql.Encoders;
|
|
|
|
import org.apache.spark.sql.SaveMode;
|
|
|
|
import org.apache.spark.sql.SparkSession;
|
|
|
|
import org.dom4j.DocumentException;
|
|
|
|
import org.slf4j.Logger;
|
|
|
|
import org.slf4j.LoggerFactory;
|
2020-07-13 19:26:46 +02:00
|
|
|
|
2020-07-13 18:22:34 +02:00
|
|
|
import eu.dnetlib.dhp.application.ArgumentApplicationParser;
|
|
|
|
import eu.dnetlib.dhp.oa.dedup.model.Block;
|
2020-07-13 19:26:46 +02:00
|
|
|
import eu.dnetlib.dhp.oa.dedup.model.BlockStats;
|
2020-07-13 18:22:34 +02:00
|
|
|
import eu.dnetlib.dhp.utils.ISLookupClientFactory;
|
|
|
|
import eu.dnetlib.enabling.is.lookup.rmi.ISLookUpException;
|
|
|
|
import eu.dnetlib.enabling.is.lookup.rmi.ISLookUpService;
|
|
|
|
import eu.dnetlib.pace.config.DedupConfig;
|
|
|
|
import eu.dnetlib.pace.model.FieldListImpl;
|
|
|
|
import eu.dnetlib.pace.model.FieldValueImpl;
|
|
|
|
import eu.dnetlib.pace.model.MapDocument;
|
|
|
|
import eu.dnetlib.pace.util.MapDocumentUtil;
|
|
|
|
import scala.Tuple2;
|
|
|
|
|
|
|
|
public class SparkBlockStats extends AbstractSparkAction {
|
|
|
|
|
2020-07-13 19:53:18 +02:00
|
|
|
private static final Logger log = LoggerFactory.getLogger(SparkBlockStats.class);
|
2020-07-13 18:22:34 +02:00
|
|
|
|
|
|
|
public SparkBlockStats(ArgumentApplicationParser parser, SparkSession spark) {
|
|
|
|
super(parser, spark);
|
|
|
|
}
|
|
|
|
|
|
|
|
public static void main(String[] args) throws Exception {
|
|
|
|
ArgumentApplicationParser parser = new ArgumentApplicationParser(
|
|
|
|
IOUtils
|
|
|
|
.toString(
|
2020-07-13 19:53:18 +02:00
|
|
|
SparkBlockStats.class
|
2020-07-13 18:22:34 +02:00
|
|
|
.getResourceAsStream(
|
|
|
|
"/eu/dnetlib/dhp/oa/dedup/createBlockStats_parameters.json")));
|
|
|
|
parser.parseArgument(args);
|
|
|
|
|
|
|
|
SparkConf conf = new SparkConf();
|
|
|
|
conf.set("spark.serializer", "org.apache.spark.serializer.KryoSerializer");
|
|
|
|
conf
|
|
|
|
.registerKryoClasses(
|
|
|
|
new Class[] {
|
|
|
|
MapDocument.class, FieldListImpl.class, FieldValueImpl.class, Block.class
|
|
|
|
});
|
|
|
|
|
2020-07-13 19:53:18 +02:00
|
|
|
new SparkBlockStats(parser, getSparkSession(conf))
|
2020-07-13 18:22:34 +02:00
|
|
|
.run(ISLookupClientFactory.getLookUpService(parser.get("isLookUpUrl")));
|
|
|
|
}
|
|
|
|
|
|
|
|
public Long computeComparisons(Long blockSize, Long slidingWindowSize){
|
|
|
|
|
|
|
|
if (slidingWindowSize >= blockSize)
|
|
|
|
return (slidingWindowSize * (slidingWindowSize - 1)) / 2;
|
|
|
|
else {
|
|
|
|
return (blockSize - slidingWindowSize + 1) * (slidingWindowSize * (slidingWindowSize - 1)) / 2;
|
|
|
|
}
|
|
|
|
}
|
2020-07-13 19:26:46 +02:00
|
|
|
|
|
|
|
@Override
|
|
|
|
public void run(ISLookUpService isLookUpService)
|
|
|
|
throws DocumentException, IOException, ISLookUpException {
|
|
|
|
|
|
|
|
// read oozie parameters
|
|
|
|
final String graphBasePath = parser.get("graphBasePath");
|
|
|
|
final String isLookUpUrl = parser.get("isLookUpUrl");
|
|
|
|
final String actionSetId = parser.get("actionSetId");
|
|
|
|
final String workingPath = parser.get("workingPath");
|
|
|
|
|
|
|
|
log.info("graphBasePath: '{}'", graphBasePath);
|
|
|
|
log.info("isLookUpUrl: '{}'", isLookUpUrl);
|
|
|
|
log.info("actionSetId: '{}'", actionSetId);
|
|
|
|
log.info("workingPath: '{}'", workingPath);
|
|
|
|
|
|
|
|
// for each dedup configuration
|
|
|
|
for (DedupConfig dedupConf : getConfigurations(isLookUpService, actionSetId)) {
|
|
|
|
|
|
|
|
final String subEntity = dedupConf.getWf().getSubEntityValue();
|
|
|
|
log.info("Creating blockstats for: '{}'", subEntity);
|
|
|
|
|
|
|
|
final String outputPath = DedupUtility.createBlockStatsPath(workingPath, actionSetId, subEntity);
|
|
|
|
removeOutputDir(spark, outputPath);
|
|
|
|
|
|
|
|
JavaSparkContext sc = JavaSparkContext.fromSparkContext(spark.sparkContext());
|
|
|
|
|
|
|
|
JavaPairRDD<String, MapDocument> mapDocuments = sc
|
|
|
|
.textFile(DedupUtility.createEntityPath(graphBasePath, subEntity))
|
|
|
|
.mapToPair(
|
|
|
|
(PairFunction<String, String, MapDocument>) s -> {
|
|
|
|
MapDocument d = MapDocumentUtil.asMapDocumentWithJPath(dedupConf, s);
|
|
|
|
return new Tuple2<>(d.getIdentifier(), d);
|
|
|
|
});
|
|
|
|
|
|
|
|
// create blocks for deduplication
|
|
|
|
JavaPairRDD<String, Block> blocks = Deduper.createSortedBlocks(mapDocuments, dedupConf);
|
|
|
|
|
|
|
|
JavaRDD<BlockStats> blockStats = blocks
|
|
|
|
.map(
|
|
|
|
b -> new BlockStats(
|
|
|
|
b._1(),
|
|
|
|
(long) b._2().getDocuments().size(),
|
|
|
|
computeComparisons(
|
|
|
|
(long) b._2().getDocuments().size(), (long) dedupConf.getWf().getSlidingWindowSize())));
|
|
|
|
|
|
|
|
// save the blockstats in the workingdir
|
|
|
|
spark
|
|
|
|
.createDataset(blockStats.rdd(), Encoders.bean(BlockStats.class))
|
|
|
|
.write()
|
|
|
|
.mode(SaveMode.Overwrite)
|
|
|
|
.save(outputPath);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2020-07-13 18:22:34 +02:00
|
|
|
}
|