A Learning Record Store (LRS) is a persistent store for xAPI statements and associated attachments and documents. The full LRS specification can be found in Part 3 of the xAPI specification. SQL LRS is distinct from other LRSs developed at Yet Analytics for being SQL-based and supporting multiple SQL database management systems (DBMSs) like SQLite, and Postgres.
SQL LRS Accounts can be created using the user interface (see Getting Started) or API calls (see Endpoints). Accounts can be used to create or access SQL LRS credentials. These credentials, which consist of an API key, a secret API key, and their scopes (described in the xAPI spec), are then used as headers for LRS-specific methods to authenticate and authorize the request sender.
If you previously used Yet's Cloud LRS products, it is important to be aware of certain differences:
- Tenancy is not supported in SQL LRS; the entire database can be considered to be a single default tenant.
- All operations in SQL LRS are synchronous; async operations are not supported.
storedtimestamps are not strictly monotonic in SQL LRS; two or more Statements may be assigned the same timestamp if stored in quick succession.- If a Statement voids a target Statement that is itself voiding, SQL LRS will accept it upon insertion, though it will not update the state of the target Statement as per the xAPI spec. (The Cloud LRS, on the other hand, will simply reject the voiding Statement.)