Replacing jobs
Is your job running with outdated information? Then just replace it...
Do you need to replace an already enqueued job? Or a job that is already being processed? Thanks to the JobRunr Pro enqueueOrReplace and scheduleOrReplace you can update the existing job easily.
UUID cityId = city.getId();
Observation observation = observationService.getLatestObservation(cityId); // the original observation
BackgroundJob.enqueue(myId, () -> weatherService.predictWeather(cityId, observation));
Observation observation = observationService.getLatestObservation(cityId); // the updated observation after a storm
BackgroundJob.enqueueOrReplace(myId, () -> weatherService.predictWeather(cityId, observation));
By default, JobRunr ignores jobs with an id that already exist in the JobRunr database. The reason for this is that JobRunr allows to create jobs from JMS messages in a load-balanced environment. By ignoring a job that already exists, we’re sure to run the job only once.
Creating a UUID from an arbitrary string
In the above example, your city id might not be a UUID but an arbitrary string that also uniquely identify the city. With JobRunr Pro you can use the method fromIdentifier available on JobId, JobProId and JobProRequestId to convert any string to a UUID.
Going back to the above example, our city id is the same as the city name (e.g., Brussels) and we don’t want to enqueue multiple jobs with the same city name.
UUID jobId = JobProId.fromIdentifier(city.name());
Observation observation = observationService.getLatestObservation(city.name()); // the original observation
BackgroundJob.enqueue(jobId, () -> weatherService.predictWeather(city.name(), observation));
Observation observation = observationService.getLatestObservation(city.name()); // the updated observation after a storm
BackgroundJob.enqueueOrReplace(jobId, () -> weatherService.predictWeather(city.name(), observation));