Defining a Job

A Job represents both the object to be serialized to a Connection to eventually work as well as the code to execute when working a Job.

Extending from AbstractJob

The first step to define a Job is to extend from AbstractJob. This provides the necessary helper methods to run a Job through the cbq pipeline.

component
    name="SendWelcomeEmailJob"
    extends="cbq.models.Jobs.AbstractJob"
{

}

Properties

Defining properties on your job allow you to see at a glance what data your Job expects when constructing it. These properties and their values will be serialized to a Connection when dispatching a Job.

Defining the properties is not strictly necessary, but your future self will thank you when you try to remember what properties your job is using.

component
    name="SendWelcomeEmailJob"
    extends="cbq.models.Jobs.AbstractJob"
{

    property name="email";
    property name="greeting";

}

The handle method

The handle method is called when a Job is worked. Before being called, a Job will be reconstructed with the serialized data from the Connection. This method is the only required method when defining a Job.

Prefer using provider: injections or inline getInstance calls for logic in your handle method. The Job component is created both when dispatching and when working your job, so utilizing these tools will reduce unnecessary processing time when dispatching your job.

Job Execution Properties

A job can define several execution properties on the job itself. If defined, these values override the module, connection, or worker defaults. It can still be overridden using the job methods when creating a job.

Lifecycle Methods

A Job can define a before or after method that will be called as part of the Job lifecycle.

Manual Control

release

A job can be manually released from inside a Job. Releasing a Job sends it back to the queue and increments the attempt count. An optional delay (in seconds) can be passed to release.

cancel

A job can be cancelled from inside a Job. Cancelling a Job fails the Job and skips any future retries. It accepts an optional error message as its argument.

ProviderContext

One field on a job is provided exclusively for Providers — the providerContext field. Providers can use this field to store any data necessary on the Job instance itself. This can be no data, a struct of metadata, or an underlying Java class. This makes it easier for the provider to get that data back in the Provider lifecycle methods later.

Last updated