I am working on a small project, which focuses on processing some data, and I'm building it based on rihanna. One part of the process is a background job, that spawns multiple followup jobs, and, since Rihanna jobs are based on postgres, I am doing that by introducing multiple rows in one query. It's awesome, and it's enabling the project to work in the first place, since this particular operation allows it to work on a cheap host. The thing is, I got it to work by copying rihanna schemas from https://github.com/samsondav/rihanna_ui, and doing
now = DateTime.utc_now() |> DateTime.truncate(:second)
jobs_data =
for entity_id <- entity_ids do
term = {MyProject.JobModule, [entity_id]}
[term: term, enqueued_at: now, priority: 30]
end
Homepage.Repo.insert_all(RihannaJob, jobs_data)
Do you think this could be supported by Rihanna natively? It really is a killer feature.
I tried to make it work through INSERT INTO SELECT statement, that would really be amazing, but I don't think it's possible in current architecture, since each job data is serialized with :erlang.term_to_binary
I am working on a small project, which focuses on processing some data, and I'm building it based on rihanna. One part of the process is a background job, that spawns multiple followup jobs, and, since Rihanna jobs are based on postgres, I am doing that by introducing multiple rows in one query. It's awesome, and it's enabling the project to work in the first place, since this particular operation allows it to work on a cheap host. The thing is, I got it to work by copying rihanna schemas from https://github.com/samsondav/rihanna_ui, and doing
Do you think this could be supported by Rihanna natively? It really is a killer feature. I tried to make it work through
INSERT INTO SELECT
statement, that would really be amazing, but I don't think it's possible in current architecture, since each job data is serialized with:erlang.term_to_binary
WDYT?