Schema Registry¶

PatternOps provides two specialised schema registry sub-capabilities for managing schema definitions, versions, and compatibility.

Two Registries¶

Registry	Package	Use Case
Messaging	`capabilities.schema`	Kafka-style subject/version schemas
General	`capabilities.schema`	Datasets, APIs, configs, any structured data

Messaging Schema Registry¶

Kafka-style schema management with subject-based versioning:

public interface MessagingSchemaRegistryContract extends BaseProviderContract {
    RegistrationResult registerSchema(String subject, Object schema, String format);
    Object getSchema(String subject, int version);
    Object getLatestSchema(String subject);
    CompatibilityVerdict checkCompatibility(String subject, Object proposedSchema);
    List<String> listSubjects();
    List<Integer> listVersions(String subject);
    void deleteSubject(String subject);
}

Schema Formats¶

Format	Description
`avro`	Apache Avro schemas
`json_schema`	JSON Schema
`protobuf`	Protocol Buffers
`custom`	Organisation-specific format

Compatibility Modes¶

Mode	Rule
`BACKWARD`	New schema can read data written by old schema
`FORWARD`	Old schema can read data written by new schema
`FULL`	Both backward and forward compatible
`NONE`	No compatibility checks

General Schema Registry¶

For datasets, APIs, and configuration schemas:

public interface GeneralSchemaRegistryContract extends BaseProviderContract {
    RegistrationResult registerSchema(String namespace, String name, Object schema, String format);
    Object getSchema(String namespace, String name, int version);
    Object getLatestSchema(String namespace, String name);
    EvolutionResult evolveSchema(String namespace, String name, Object proposed, String strategy);
    List<SchemaDescriptor> listSchemas(String namespace);
    SchemaDiff compareVersions(String namespace, String name, int versionA, int versionB);
    void deprecateSchema(String namespace, String name, int version);
}

Schema Evolution¶

EvolutionResult result = registry.evolveSchema(
    "production", "customer-orders",
    newSchema, "BACKWARD"
);
// result indicates: compatible, backward-compatible, or breaking

Version Comparison¶

SchemaDiff diff = registry.compareVersions(
    "production", "customer-orders", 1, 2
);
// diff.added() → new fields
// diff.removed() → removed fields
// diff.modified() → changed fields

Example Providers¶

Provider	Type
Confluent Schema Registry	Messaging
AWS Glue Schema Registry	Messaging + General
Custom Database-backed	General

Integration with Data Contracts¶

Schema Registry works alongside Data Contracts:

Data Contract defines the agreement (schema + quality + SLA)
Schema Registry manages the schema versions and compatibility
Echo Service tracks contract history and publishes changes