datahub.spark.model.dataset.JdbcDataset Maven / Gradle / Ivy

Go to download

Show more of this group Show more artifacts with this name
Show all versions of datahub-spark-lineage Show documentation

Library to push data lineage from spark to datahub

There is a newer version: 0.14.1-12rc3

Show newest version

package datahub.spark.model.dataset;

import com.linkedin.common.FabricType;
import io.opentracing.contrib.jdbc.parser.URLParser;
import java.util.HashMap;
import java.util.Map;
import lombok.ToString;

@ToString
public class JdbcDataset extends SparkDataset {
  // TODO: Should map to the central location on datahub for platform names
  private static final Map PLATFORM_NAME_MAPPING = new HashMap<>();

  static {
    PLATFORM_NAME_MAPPING.put("postgresql", "postgres");
  }

  public JdbcDataset(String url, String tbl, String platformInstance, FabricType fabricType) {
    super(platformName(url), platformInstance, dsName(url, tbl), fabricType);
  }

  private static String platformName(String url) {
    String dbType = URLParser.parse(url).getDbType();
    return PLATFORM_NAME_MAPPING.getOrDefault(dbType, dbType);
  }

  private static String dsName(String url, String tbl) {
    return URLParser.parse(url).getDbInstance() + "." + tbl;
  }
}