Multiple results tabs opening when executing script under Apache Spark connector

Description

When executing a script on an Apache Spark connection, a results window gets opened for each statement whether results are returned or not. Consider the following query:

CREATE TABLE MyTable(PersonName VARCHAR(50));
INSERT INTO MyTable(PersonName) SELECT 'John Doe';
SELECT * FROM MyTable;
DROP TABLE MyTable;

On other connections (tested on MS SQL Server), one result window is opened, which is expected since there is one SELECT statement. On Apache Spark, however, four tabs are opened, one for each statement.

MS SQL Server:

Apache Spark:

DBeaver Version

22.2.4.202211061524

Operating System

Windows 11 22H2

Database and driver

Spark 3.3.1 Apache Spark driver 1.2.1.spark2

Steps to reproduce

1. Create an Apache Spark connection
2. Create a script consisting of several SQL DML statements that do not produce results
3. Run all the statements as a single script by pressing Alt-X or the Execute script button
4. Multiple result windows are opened

Additional context

No response

[ShadelessFox]

Wow, thank you! I believe this would be enough.

[chipmunkofdoom2]

@ShadelessFox Are there any settings I can change or play around with on the driver? Or in DBeaver?

If you need a Spark dev environment, it’s fairly straightforward to stand up a single node Spark cluster in standalone mode. Just tested these commands in an LXC container running Ubuntu 22.04:

# Optional: if you don't have Java installed or this is new VM/container, install JDK and set Java home
sudo apt-get update; sudo apt-get install openjdk-11-jdk -y
echo 'export JAVA_HOME=/usr/lib/jvm/java-11-openjdk-amd64' >> ~/.bashrc
source ~/.bashrc

# Download latest Spark release and unpack
wget https://dlcdn.apache.org/spark/spark-3.3.1/spark-3.3.1-bin-hadoop3.tgz
tar -xf spark-3.3.1-bin-hadoop3.tgz
mv spark-3.3.1-bin-hadoop3 spark

# Start master node. Explicitly set listen address to 127.0.0.1
spark/sbin/start-master.sh --host 127.0.0.1

# Start worker node pointing at the master node we just created
spark/sbin/start-worker.sh spark://127.0.0.1:7077

#Start thriftserver pointing at our master node. This is how DBeaver will connect
spark/sbin/start-thriftserver.sh --master spark://127.0.0.1:7077

Next, add a Spark connection in DBeaver. For the host, use the IP/name of the machine hosting Spark. Leave port as 10000. I put default as the database, but I don’t think this matters. Leave username/password blank:

You should then be up and running.

I can also help with testing. I don’t have Java experience, but if there are any guides for compiling/running tests, I can follow them.