Consider adding a method to comfortably delay failing jobs

[27182818284590]

In my apps, I want to release failing jobs back to the tube with lower priority and with increasing delay.
A job already has all the necessary metadata, but I would love to avoid repeating the same calculations over and over.

Compact-ish version:

prio := uint32(job.Stats.Releases) * job.Stats.Priority
delay := time.Duration(job.Stats.Releases) * job.Stats.Delay
if err := job.ReleaseWithParams(ctx, prio, delay); err != nil {
 logger.Error("Failed to release a failing job", slog.Any("err", err))
}
return

Extended version:

delay := time.Duration(job.Stats.Releases) * job.Stats.Delay
if delay > maxDelay { 
 delay = maxDelay
}
prio := uint32(job.Stats.Releases) * job.Stats.Priority
if prio > maxPrio { 
 prio = maxPrio
}
if err := job.ReleaseWithParams(ctx, prio, delay); err != nil {
 logger.Error("Failed to release a failing job", 
 slog.Any("job_stats", job.Stats), 
 slog.Any("err", err),
 )
}
return

I suggest adding a function that helps to delay a job with proper back-off settings:

It may look like this:

if err := job.ReleaseWithParams(ctx, job.BackoffParams(maxDelay)); err != nil {
 logger.Error("Failed to release a failing job", slog.Any("err", err))
}
return

I am curious to hear your thoughts. Maybe there is already something that I am missing.