Migrating from Bull when using sandboxed and named processors together

[plus-]

I'm looking for some advice or best practices to migrate a project using Bull which was implemented using named and sandboxed processors at the same time.

This issue points to using a function with a switch case on the job name to reproduce the named processor switching, but if we put that function in a file we would end up with a single sandboxed processor instead of multiple for each named processors (I'd like to keep the behavior as close as it was if possible for now).

Otherwise is there a simple way to spawn a sandboxed processor myself (for example with a worker option)?

I guess a solution would be to split the queue in multiple ones but the migration would be more difficult to perform (existing jobs)