Batch Syndication Jobs

Developers can initiate batch syndication jobs by sending a batch event message to the ep.batch.catalogsyndication queue using Java Messaging Service (JMS).

Create Batch Syndication JMS Message

When the catalog syndication batch job in the Elastic Path batch server gets an event message from the ep.batch.catalogsyndication queue, the batch projection builder starts the corresponding batch jobs. For Elastic Path batch server to initiate the batch jobs, developers must create a message and send it to the ep.batch.catalogsyndication queue in JMS in the following format:

{
    "eventType":{
        "@class":"CatalogBatchEventType",
        "name":"START_JOB"
      },
      "guid":"BUILD_ALL_OFFERS",
      "data": {
          "cleanUpDatabase":"true"
      }
}

note

Ensure that the cleanUpDatabase flag is set to true to clean the entire database before building the batch projection. If the flag is set to false, only projections for entities that have changed are cleaned. Cleaning up the database ensures that data from the previous build projection is removed.

Build Projection

The batch projection builder builds each projection in the following sequence:

1. Runs the cleanUpDatabase job. This step is optional
2. Fetches all entities of the corresponding job type, such as brand, SKU option, or product
3. Builds multiple projections from the single domain entity
4. Writes the projections to the database

The batch projection builder can build all entity projections together or as separate entities. The entities are built in the following order if built all together:

1. BUILD_ALL_OPTIONS
2. BUILD_ALL_ATTRIBUTES
3. BUILD_ALL_BRANDS
4. BUILD_ALL_FIELDMETADATA
5. BUILD_ALL_OFFERS
6. BUILD_ALL_PROJECTIONS

Expired Projections

You can set a disable date and time on products, SKUs and categories in the catalog. These values are syndicated to the TCATALOGPROJECTIONS database table. When you run a batch job, the job checks the TCATALOGPROJECTIONS table. It deletes the product from the catalog if the DISABLE_DATE_TIME is set to the current date and time or is in the past. The job schedules its next run to be on the closest DISABLE_DATE_TIME or within an hour, whichever comes first.