The v1 inventory is a hierarchical structure of feed entities in JSON-LD format. This schema allowed early access-program partners to use the schema to integrate with Ordering End-to-End. The v1 inventory schema is now deprecated.
The latest version of the data feed specification, v2 inventory, has a flat structure and uses a relational schema model. This allows partners to extract and transform entities from their relational database in a more straightforward manner to create the data feeds. The v2 inventory is in NDJSON, which is a newline delimited JSON format.
Why you should migrate from v1 to v2
Although we continue to provide maintenance support to partners with Ordering End-to-End v1 inventory schema integrations, all new partner integrations must use the v2 inventory.
New features in the v2 inventory might not be available in the v1 inventory. For an example, see Deal.
The use of v2 inventory requires a separate Ordering End-to-End project and a new integration. The following process explains how to migrate a v1 inventory to a v2 inventory.
- To create a new v2 project, first complete the integration steps.
- To implement the data feeds, use the relational inventory schema.
- If applicable, reuse the fulfillment endpoint from your v1 project.
- When your v2 project is ready for launch, follow these steps:
- Ensure that your v2 project meets the launch readiness checklist.
If you want to rollback to v1 after deployment of your v2 project in production, you can undeploy the production release in your v2 project and then redeploy your v1 project’s production release.
Impact on partner data and provider ranking
The following are potential impacts to partner data and provider ranking.
- Transaction logs and analytics data from your v1 project don't transfer to your v2 project.
- Performance metrics, such as conversion rates and API metrics from your v1 project, aren't ported to the v2 project. This resets your provider ranking in Ordering End-to-End.