Data Models

The ROAR backend architecture uses data models to represent the data entities and their relationships. These models are designed to be flexible, scalable, and maintainable, allowing for easy integration with various data sources and services.

Data Model Design Principles

Each data model is an interface that defines the structure and behavior of a specific data entity. It provides a clear and consistent way to interact with the data, ensuring data integrity and consistency across the system. Data models should be designed with extensibility in mind, allowing for easy addition of new fields or relationships without breaking existing code. Data models are agnostic to the underlying data storage mechanism, allowing for easy integration with various data sources and services.

Core Data Models

The ROAR platform is built around several key data models:

Administration Model

The Administration model represents a collection of assessments administered to a specific set of organizations. It includes:

  • Assessment configurations
  • Organization settings
  • Access control information
  • Status tracking

User Model

The User model represents individuals who interact with the platform and includes:

  • Basic profile information
  • Authentication details
  • System-wide unique identifier
  • Creation and update timestamps

Organization Model

The OrgBase model represents educational institutions and other organizations using the platform:

  • Hierarchical organization structure
  • Organizational metadata
  • Identification information

Identity Provider Model

The IdentityProvider model manages integration with external authentication systems:

  • Provider type and configuration
  • Synchronization status and metadata
  • Mapping between external and internal identifiers

Model Relationships

Data models in ROAR maintain relationships through references rather than direct embedding:

  • Administration to Organization: Many-to-many relationship tracked through organization lists
  • User to Organization: Many-to-many relationship managed through claims
  • Administration to Assessment: One-to-many relationship with assessments as properties of an administration
  • User to Identity Provider: Many-to-one relationship establishing authentication source

Base Model

All data models extend the BaseModel interface which provides common properties:

  • id: A unique identifier within the collection
  • createdAt: Timestamp indicating when the entity was created
  • updatedAt: Timestamp indicating the last modification
  • name: Human-readable identifier (where applicable)

Type Safety and Validation

The ROAR platform uses TypeScript interfaces to ensure type safety across all data models. This provides:

  • Compile-time validation
  • Better developer experience with IDE autocompletion
  • Clear documentation of expected data structures
  • Runtime type checking when interfacing with external systems

Data Model Evolution

As the ROAR platform evolves, data models may need to change. The architecture supports:

  • Backward compatibility through optional fields
  • Version tracking for data migrations
  • Transformation utilities for converting between different versions of models
  • Schema validation to ensure data integrity during transitions

Data Model Implementation

Each data model has an associated repository class that implements the interface. The class provides methods for creating, reading, updating, and deleting data entities. It also provides methods for querying and filtering data entities based on specific criteria. The attributes of each model can also be defined as an interface that extends the base model interface. This allows for increased type safety and better code organization.

You can find the complete definition of all data models in the Interfaces section of the API documentation.