BrBETA_dnet-hadoop/dhp-workflows/dhp-workflows-common/src/main/java/eu/dnetlib/dhp/common/GraphSupport.java

package eu.dnetlib.dhp.common;

import org.apache.spark.sql.DataFrameWriter;
import org.apache.spark.sql.Dataset;
import org.apache.spark.sql.SaveMode;
import org.slf4j.Logger;
import org.slf4j.LoggerFactory;

import eu.dnetlib.dhp.schema.common.ModelSupport;
import eu.dnetlib.dhp.schema.oaf.Oaf;

public class GraphSupport {

	private static final Logger log = LoggerFactory.getLogger(GraphSupport.class);

	private static <T extends Oaf> void saveGraphTable(Dataset<T> dataset, Class<T> clazz, String outputGraph,
		eu.dnetlib.dhp.common.SaveMode saveMode) {

		log.info("saving graph in {} mode to {}", outputGraph, saveMode.toString());

		final DataFrameWriter<T> writer = dataset.write().mode(SaveMode.Overwrite);
		switch (saveMode) {
			case JSON:
				writer.option("compression", "gzip").json(outputGraph);
				break;
			case PARQUET:
				final String db_table = ModelSupport.tableIdentifier(outputGraph, clazz);
				writer.saveAsTable(db_table);
				break;
		}

	}

}