reading graph dump as text files, encoded as newline-delimited JSON records, as indicated in the wiki

This commit is contained in:
Claudio Atzori 2020-03-10 17:19:17 +01:00
parent 60aedb1110
commit 7b6f0c8756
1 changed files with 14 additions and 11 deletions

View File

@ -3,7 +3,7 @@ package eu.dnetlib.dhp.graph;
import com.fasterxml.jackson.databind.ObjectMapper; import com.fasterxml.jackson.databind.ObjectMapper;
import eu.dnetlib.dhp.application.ArgumentApplicationParser; import eu.dnetlib.dhp.application.ArgumentApplicationParser;
import org.apache.commons.io.IOUtils; import org.apache.commons.io.IOUtils;
import org.apache.hadoop.io.Text; import org.apache.spark.SparkConf;
import org.apache.spark.api.java.JavaSparkContext; import org.apache.spark.api.java.JavaSparkContext;
import org.apache.spark.sql.Encoders; import org.apache.spark.sql.Encoders;
import org.apache.spark.sql.SaveMode; import org.apache.spark.sql.SaveMode;
@ -13,7 +13,9 @@ public class SparkGraphImporterJob {
public static void main(String[] args) throws Exception { public static void main(String[] args) throws Exception {
final ArgumentApplicationParser parser = new ArgumentApplicationParser(IOUtils.toString(SparkGraphImporterJob.class.getResourceAsStream("/eu/dnetlib/dhp/graph/input_graph_parameters.json"))); final ArgumentApplicationParser parser = new ArgumentApplicationParser(
IOUtils.toString(SparkGraphImporterJob.class.getResourceAsStream(
"/eu/dnetlib/dhp/graph/input_graph_parameters.json")));
parser.parseArgument(args); parser.parseArgument(args);
try(SparkSession spark = getSparkSession(parser)) { try(SparkSession spark = getSparkSession(parser)) {
@ -26,23 +28,24 @@ public class SparkGraphImporterJob {
spark.sql(String.format("CREATE DATABASE IF NOT EXISTS %s", hiveDbName)); spark.sql(String.format("CREATE DATABASE IF NOT EXISTS %s", hiveDbName));
// Read the input file and convert it into RDD of serializable object // Read the input file and convert it into RDD of serializable object
GraphMappingUtils.types.forEach((name, clazz) -> { GraphMappingUtils.types.forEach((name, clazz) -> spark.createDataset(sc.textFile(inputPath + "/" + name)
spark.createDataset(sc.sequenceFile(inputPath + "/" + name, Text.class, Text.class) .map(s -> new ObjectMapper().readValue(s, clazz))
.map(s -> new ObjectMapper().readValue(s._2().toString(), clazz))
.rdd(), Encoders.bean(clazz)) .rdd(), Encoders.bean(clazz))
.write() .write()
.mode(SaveMode.Overwrite) .mode(SaveMode.Overwrite)
.saveAsTable(hiveDbName + "." + name); .saveAsTable(hiveDbName + "." + name));
});
} }
} }
private static SparkSession getSparkSession(ArgumentApplicationParser parser) { private static SparkSession getSparkSession(ArgumentApplicationParser parser) {
SparkConf conf = new SparkConf();
conf.set("hive.metastore.uris", parser.get("hive_metastore_uris"));
return SparkSession return SparkSession
.builder() .builder()
.appName(SparkGraphImporterJob.class.getSimpleName()) .appName(SparkGraphImporterJob.class.getSimpleName())
.master(parser.get("master")) .master(parser.get("master"))
.config("hive.metastore.uris", parser.get("hive_metastore_uris")) .config(conf)
.enableHiveSupport() .enableHiveSupport()
.getOrCreate(); .getOrCreate();
} }