JDBC DatabaseMeta Implementation for Mapr Drill DB

[Kousick]

We have launched a 6.1.0.20180926230239.GA version MapR cluster on one of our GCP compute engines. We are using this cluster for multiple Hadoop tasks and also DrillDB.

We are building a JDBC plugin for one of our softwares, for which we are internally using the jdbc-all mapr jar provided by the MapR repository. 

The JDBC driver works fine for raw SQL queries, but fails to behave as expected for internal methods. For example the getTables DatabaseMeta implementation should return the list of tables in a given connection. However, this returns an empty list. I know why it behaves this way., internally it fetches rows from the information_schema table. However MapRDB does not keep table information directly in the information_schema table. We will have to write a different implementation for fetching files and tables from MapRDB.

Has anyone come across this ? Am I doing something wrong here ? Is there any other JDBC implementation that can work as expected ?

Let me know if you need more details.

Thanks!

[Harshkohli]

Can you let me know below:

1) What is the version of JAR being used for drill-jdbc-all?

Are you using 1.16 version.

Also are you using documentation listed below:

https://docs.oracle.com/javase/8/docs/api/java/sql/DatabaseMetaData.html#getTables-java.lang.String-java.lang.String-java.lang.String-java.lang.String:A-

If yes, can you tell me what values you are passing for below:

(String catalog,
                    String schemaPattern,
                    String tableNamePattern,
                    String[] types)

[Harshkohli]

Can you let us know if your problem got solved or it is still existing?

Can you answer my questions above?