json-transformation

Json Transformation

An integration layer receives JSON payloads from a partner API in a deeply nested schema that doesn't match the internal data model. Fields need renaming, nested objects need flattening, arrays need filtering, and the output must conform to the internal schema before downstream services can consume it.

Pipeline

[jt_parse_input]
     |
     v
[jt_map_fields]
     |
     v
[jt_restructure_nested]
     |
     v
[jt_validate_schema]
     |
     v
[jt_emit_output]

Workflow inputs: sourceJson, mappingRules

Workers

EmitOutputWorker (task: jt_emit_output)

Emits the final transformed record.

• Reads finalRecord, isValid. Writes record, emitted

MapFieldsWorker (task: jt_map_fields)

Maps source fields to target fields with renaming and value transforms.

• Lowercases strings, uppercases strings, trims whitespace
• Reads data. Writes mapped, mappedFieldCount

ParseInputWorker (task: jt_parse_input)

Parses the incoming source JSON and counts its top-level fields.

• Reads sourceJson. Writes parsed, fieldCount

RestructureNestedWorker (task: jt_restructure_nested)

Restructures a flat mapped record into nested groups.

• Reads mapped. Writes restructured

ValidateSchemaWorker (task: jt_validate_schema)

Validates that the restructured record conforms to the expected schema. Checks that identity.id and contact.email exist and are non-empty.

• Reads record. Writes validated, isValid

---

43 tests | Workflow: json_transformation | Timeout: 60s

See RUNNING.md for setup and usage.