D-Net
/
dnet-hadoop
14
5
Fork 11
Code Issues 14 Pull Requests 13 Projects 2 Releases 2 Wiki Activity

close the SparkSession at the end

This commit is contained in:
Claudio Atzori 2020-03-04 10:53:31 +01:00
parent 25ceec29ab
commit 9af3e904be
1 changed files with 24 additions and 19 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

43
dhp-workflows/dhp-graph-mapper/src/main/java/eu/dnetlib/dhp/graph/SparkGraphImporterJob.java
View File

@ -15,30 +15,35 @@ public class SparkGraphImporterJob {
final ArgumentApplicationParser parser = new ArgumentApplicationParser(IOUtils.toString(SparkGraphImporterJob.class.getResourceAsStream("/eu/dnetlib/dhp/graph/input_graph_parameters.json")));
parser.parseArgument(args);
final SparkSession spark = SparkSession
try(SparkSession spark = getSparkSession(parser)) {
final JavaSparkContext sc = new JavaSparkContext(spark.sparkContext());
final String inputPath = parser.get("sourcePath");
final String hiveDbName = parser.get("hive_db_name");
spark.sql(String.format("DROP DATABASE IF EXISTS %s CASCADE", hiveDbName));
spark.sql(String.format("CREATE DATABASE IF NOT EXISTS %s", hiveDbName));
// Read the input file and convert it into RDD of serializable object
GraphMappingUtils.types.forEach((name, clazz) -> {
spark.createDataset(sc.sequenceFile(inputPath + "/" + name, Text.class, Text.class)
.map(s -> new ObjectMapper().readValue(s._2().toString(), clazz))
.rdd(), Encoders.bean(clazz))
.write()
.mode(SaveMode.Overwrite)
.saveAsTable(hiveDbName + "." + name);
});
}
}
private static SparkSession getSparkSession(ArgumentApplicationParser parser) {
return SparkSession
.builder()
.appName(SparkGraphImporterJob.class.getSimpleName())
.master(parser.get("master"))
.config("hive.metastore.uris", parser.get("hive_metastore_uris"))
.enableHiveSupport()
.getOrCreate();
final JavaSparkContext sc = new JavaSparkContext(spark.sparkContext());
final String inputPath = parser.get("sourcePath");
final String hiveDbName = parser.get("hive_db_name");
spark.sql(String.format("DROP DATABASE IF EXISTS %s CASCADE", hiveDbName));
spark.sql(String.format("CREATE DATABASE IF NOT EXISTS %s", hiveDbName));
// Read the input file and convert it into RDD of serializable object
GraphMappingUtils.types.forEach((name, clazz) -> {
spark.createDataset(sc.sequenceFile(inputPath + "/" + name, Text.class, Text.class)
.map(s -> new ObjectMapper().readValue(s._2().toString(), clazz))
.rdd(), Encoders.bean(clazz))
.write()
.mode(SaveMode.Overwrite)
.saveAsTable(hiveDbName + "." + name);
});
}
}