Field Definition

The foundational constraint that dictates the value space and permissible operations on a field. This is not merely a formatting suggestion but a strict contract enforced during schema validation.

• Primitive Types: string, integer, boolean, float for basic scalar values.
• Temporal Types: date, datetime, duration for time-series and scheduling data.
• Complex Types: object for nested structures, array for ordered lists.
• Semantic Formats: Sub-typing like email, uri, or uuid adds validation logic without changing the underlying primitive.

Defines the multiplicity of a field's occurrence within a parent content type. This is critical for data integrity and query optimization.

• One-to-One (1:1): A single, required instance. Example: A Product must have exactly one sku.
• One-to-Many (1:N): An array of values. Example: A Blog Post can have multiple tags.
• Optionality: A minimum cardinality of 0 makes a field nullable. A maximum of unbounded allows arbitrary list growth.
• Uniqueness: A boolean constraint ensuring no two records share the same value for this field, often used for primary identifiers.

Specifies the fallback value automatically assigned when no explicit data is provided during content creation. This ensures data completeness and reduces authoring friction.

• Static Defaults: A hardcoded value like false for a boolean or "USD" for a currency code.
• Dynamic Defaults: System-generated values such as now() for a creation timestamp or a uuid() for a unique identifier.
• Inheritance Logic: Defaults can be configured to cascade from a parent configuration or a global data dictionary.

A set of logical predicates that restrict the lexical or numerical space of a value beyond its base type. Validation is executed atomically during write operations.

• String Patterns: Regular expressions (regex) enforce formats like phone numbers or product codes.
• Numerical Bounds: minimum, maximum, exclusiveMinimum constraints for integers and floats.
• Enumeration: A closed set of allowed values (enum), transforming a free-text field into a controlled vocabulary.
• Custom Keywords: Domain-specific assertions (e.g., isMutuallyExclusive) defined in a JSON Schema dialect.

Metadata that bridges the gap between raw data structure and business meaning. These annotations power automated metadata tagging and UI rendering.

• Display Labels: Human-readable title and description for auto-generating forms.
• Taxonomic Binding: Linking a field to a concept in an external ontology or taxonomy (e.g., linking a category field to a product hierarchy).
• Unit of Measure: Explicitly tagging a numeric field with unit: "meters" or unit: "USD" to prevent calculation errors in multi-source data fusion.

Security and display logic embedded directly into the field definition to govern data governance and editorial workflow.

• Read-Only: Fields like createdAt or computedScore that cannot be mutated by standard clients.
• Write-Only: Sensitive fields like password or apiKey that are accepted on input but never returned in API responses.
• Deprecation Status: A boolean flag indicating a field is scheduled for removal, guiding schema evolution and client migration.
• Conditional Visibility: Rules that show or hide a field in a UI based on the value of another field (e.g., showing a shippingAddress field only if isPhysicalProduct is true).