forked from D-Net/dnet-hadoop
61 lines
1.8 KiB
Java
61 lines
1.8 KiB
Java
|
|
package eu.dnetlib.dhp.common;
|
|
|
|
import static eu.dnetlib.dhp.common.SparkSessionSupport.runWithSparkHiveSession;
|
|
|
|
import java.util.Optional;
|
|
|
|
import org.apache.commons.io.IOUtils;
|
|
import org.apache.spark.SparkConf;
|
|
import org.apache.spark.sql.SparkSession;
|
|
import org.slf4j.Logger;
|
|
import org.slf4j.LoggerFactory;
|
|
|
|
import eu.dnetlib.dhp.application.ArgumentApplicationParser;
|
|
|
|
public class ResetHiveDbApplication {
|
|
|
|
private static final Logger log = LoggerFactory.getLogger(ResetHiveDbApplication.class);
|
|
|
|
public static void main(final String[] args) throws Exception {
|
|
final ArgumentApplicationParser parser = new ArgumentApplicationParser(
|
|
IOUtils
|
|
.toString(
|
|
ResetHiveDbApplication.class
|
|
.getResourceAsStream(
|
|
"/eu/dnetlib/dhp/common/reset_hive_db_parameters.json")));
|
|
parser.parseArgument(args);
|
|
|
|
Boolean isSparkSessionManaged = Optional
|
|
.ofNullable(parser.get("isSparkSessionManaged"))
|
|
.map(Boolean::valueOf)
|
|
.orElse(Boolean.TRUE);
|
|
log.info("isSparkSessionManaged: {}", isSparkSessionManaged);
|
|
|
|
final String dbName = Optional
|
|
.ofNullable(parser.get("dbName"))
|
|
.orElseThrow(() -> new IllegalArgumentException("missing DB name"));
|
|
log.info("dbName: {}", dbName);
|
|
|
|
String hiveMetastoreUris = parser.get("hiveMetastoreUris");
|
|
log.info("hiveMetastoreUris: {}", hiveMetastoreUris);
|
|
|
|
SparkConf conf = new SparkConf();
|
|
conf.set("hive.metastore.uris", hiveMetastoreUris);
|
|
|
|
runWithSparkHiveSession(
|
|
conf,
|
|
isSparkSessionManaged,
|
|
spark -> {
|
|
runSQL(spark, String.format("DROP DATABASE IF EXISTS %s CASCADE", dbName));
|
|
runSQL(spark, String.format("CREATE DATABASE %s", dbName));
|
|
});
|
|
}
|
|
|
|
protected static void runSQL(SparkSession spark, String sql) {
|
|
log.info("running SQL '{}'", sql);
|
|
spark.sqlContext().sql(sql);
|
|
}
|
|
|
|
}
|