forked from D-Net/dnet-hadoop
103 lines
3.3 KiB
Java
103 lines
3.3 KiB
Java
|
|
package eu.dnetlib.dhp.sx.provision;
|
|
|
|
import static eu.dnetlib.dhp.common.SparkSessionSupport.runWithSparkSession;
|
|
|
|
import java.io.IOException;
|
|
import java.util.Objects;
|
|
import java.util.Optional;
|
|
|
|
import org.apache.commons.cli.ParseException;
|
|
import org.apache.commons.io.IOUtils;
|
|
import org.apache.solr.common.SolrInputDocument;
|
|
import org.apache.spark.SparkConf;
|
|
import org.apache.spark.api.java.JavaRDD;
|
|
import org.apache.spark.api.java.function.MapFunction;
|
|
import org.apache.spark.sql.Encoder;
|
|
import org.apache.spark.sql.Encoders;
|
|
import org.apache.spark.sql.SparkSession;
|
|
import org.slf4j.Logger;
|
|
import org.slf4j.LoggerFactory;
|
|
|
|
import com.lucidworks.spark.util.SolrSupport;
|
|
|
|
import eu.dnetlib.dhp.application.ArgumentApplicationParser;
|
|
import eu.dnetlib.dhp.oa.provision.ProvisionConstants;
|
|
import eu.dnetlib.dhp.oa.provision.model.SerializableSolrInputDocument;
|
|
import eu.dnetlib.dhp.oa.provision.scholix.ScholixToSolr;
|
|
import eu.dnetlib.dhp.oa.provision.utils.ISLookupClient;
|
|
import eu.dnetlib.dhp.utils.ISLookupClientFactory;
|
|
|
|
public class SparkIndexCollectionOnSOLR {
|
|
|
|
private static final Integer DEFAULT_BATCH_SIZE = 1000;
|
|
|
|
// LOGGER initialized
|
|
private static final Logger log = LoggerFactory.getLogger(SparkIndexCollectionOnSOLR.class);
|
|
|
|
public static void main(String[] args) throws IOException, ParseException {
|
|
final ArgumentApplicationParser parser = new ArgumentApplicationParser(
|
|
IOUtils
|
|
.toString(
|
|
Objects
|
|
.requireNonNull(
|
|
SparkIndexCollectionOnSOLR.class
|
|
.getResourceAsStream("/eu/dnetlib/dhp/sx/provision/index_solr_parameters.json"))));
|
|
|
|
parser.parseArgument(args);
|
|
|
|
final String cluster = parser.get("cluster");
|
|
log.info("Cluster is {}", cluster);
|
|
|
|
final String format = parser.get("format");
|
|
log.info("Index format name is {}", format);
|
|
|
|
final String isLookupUrl = parser.get("isURL");
|
|
log.info("isURL is {}", isLookupUrl);
|
|
|
|
final String inputPath = parser.get("inputPath");
|
|
log.info("inputPath: {}", inputPath);
|
|
|
|
Boolean isSparkSessionManaged = Optional
|
|
.ofNullable(parser.get("isSparkSessionManaged"))
|
|
.map(Boolean::valueOf)
|
|
.orElse(Boolean.TRUE);
|
|
log.info("isSparkSessionManaged: {}", isSparkSessionManaged);
|
|
|
|
final Integer batchSize = Optional
|
|
.ofNullable(parser.get("batchSize"))
|
|
.map(Integer::valueOf)
|
|
.orElse(DEFAULT_BATCH_SIZE);
|
|
log.info("batchSize: {}", batchSize);
|
|
|
|
final SparkConf conf = new SparkConf();
|
|
conf.registerKryoClasses(new Class[] {
|
|
SerializableSolrInputDocument.class
|
|
});
|
|
|
|
runWithSparkSession(
|
|
conf,
|
|
isSparkSessionManaged,
|
|
spark -> {
|
|
final ISLookupClient isLookup = new ISLookupClient(ISLookupClientFactory.getLookUpService(isLookupUrl));
|
|
final String zkHost = isLookup.getZkHost();
|
|
log.info("zkHost: {}", zkHost);
|
|
final String collection = ProvisionConstants.getCollectionName(format);
|
|
log.info("collection: {}", collection);
|
|
feedScholixToSOLRIndex(spark, inputPath, collection, batchSize, zkHost);
|
|
});
|
|
}
|
|
|
|
public static void feedScholixToSOLRIndex(final SparkSession spark, final String inputPath, final String collection,
|
|
Integer batchSize, final String zkHost) {
|
|
final JavaRDD<SolrInputDocument> docs = spark
|
|
.read()
|
|
.text(inputPath)
|
|
.as(Encoders.STRING())
|
|
.map(new ScholixToSolr(), Encoders.kryo(SolrInputDocument.class))
|
|
.toJavaRDD();
|
|
SolrSupport.indexDocs(zkHost, collection, batchSize, docs.rdd());
|
|
|
|
}
|
|
}
|