fix: DRY-up job class/args extraction

[ixti]

Related-PR: #184

[ixti]

cc: @jlledom

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 98.98%. Comparing base (3c1c543) to head (f6a7b0d).

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #185      +/-   ##
==========================================
+ Coverage   98.95%   98.98%   +0.03%     
==========================================
  Files          18       19       +1     
  Lines         383      396      +13     
  Branches       55       56       +1     
==========================================
+ Hits          379      392      +13     
  Misses          4        4              

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[freemanoid]

LCTM

[jlledom]

Works fine for me. 👍

[mnovelo]

Sidekiq::Pro specs pass too!

[jeffbax]

Pardon the crudeness of the example, but I had noticed a similar situation to #184 attempting to adopt this gem over the past week or so (moving from https://github.com/veeqo/activejob-uniqueness) although this is for using the threshold strategy not the concurrency one.

ERROR: Failed to get key suffix: wrong number of arguments (given 1, expected 2+)

basically when I substitute

key_suffix: ->(args) do
  puts args
  raise
end

My output is

 {
   "job_class" => "Events::JobKlass", 
   "job_id" => "abaccc09-9916-4ca4-9765-77968d09bbde", 
   "provider_job_id" => nil, 
   "queue_name" => "data_queue_low", 
   "priority" => nil, 
   "arguments" => ["Events::SavedEvent", { "id" => 307, "status_id_changes" => [1, 2], "_aj_symbol_keys" => [] }], 
   "executions" => 0, 
   "exception_executions" => {}, 
   "locale" => "en", 
   "timezone" => "Eastern Time (US & Canada)", 
   "enqueued_at" => "2024-03-28T18:14:14Z", 
   "first_enqueued_at" => 1711649654,
 }

Whereas I was expecting to get the position arguments from the arguments key:

• "Events::SavedEvent"
• { "id" => 307, "status_id_changes" => [1, 2], "_aj_symbol_keys" => [] }

Is this an incorrect assumption on my part or perhaps a similar snag?

[ixti]

Will finish this PR and cut patch release this week.

[ixti]

Released as v1.4.0

[jeffbax]

Released as v1.4.0

thanks! will give it a shot :)