Environment (Required on creation):

All distributions, all databases

Description (Required on creation; please attach any relevant screenshots, stacktraces, log files, etc. to the ticket):

When using the Java API to create a deployment with the name being null, filtering duplicates does not work. There is no warning about this anywhere and as a user, I am left wondering why filtering does not work.

Steps to reproduce (Required on creation):

processEngine.getRepositoryService()
  .createDeployment()
  .addClasspathResource("workflows/example-workflow.bpmn")
  .enableDuplicateFiltering(true)
  .deploy();

Observed Behavior (Required on creation):

Each time a new version of the process is deployed, regardless if it has changed or not. There is no warning about this anywhere.

Expected behavior (Required on creation):

I can see a warning that duplicate filtering will not work due to no name being provided.

Root Cause (Required on prioritization):

The query that is used to select resources for filtering is always considering them by name. It does not consider null-name deployments specifically.

Solution Ideas (Optional):

Log a warning in the application log when such a deployment is created.

Hints (Optional):

• Fixing this for good, i.e. enable filtering for NULL name deployments can lead to confusing situations where completely unrelated deployments are compared because they both have no name but don't share any similarities otherwise.
• Throwing an exception in such a case might be a bit strict since technically the deployment is fine but rather one feature will not work as expected.