Loading…
Loading…
Onboarding is the no-code process that takes a SOAP (or REST) contract and turns it into a live, converted endpoint with a generated machine-readable contract,…
Onboarding is the no-code process that takes a SOAP (or REST) contract and turns it into a live, converted endpoint with a generated machine-readable contract, deterministic full-field mapping, and observability — without writing a mediation script.
Since v1.0 onboarding runs in two directions on the same runtime:
See Introduction → Publish / Consume for the conceptual model.
Discover → Import WSDL → Configure backend → Review OpenAPI → Publish & version
Every generated operation gets a deterministic, full-field mapping (how mapping works). Manual work is for exceptions only.
You can collapse Discover → Import → Generate into a single step with zero-touch autopilot: the platform imports the WSDL, generates the service, operations, and mappings, and reports a readiness score per operation, leaving you to review and publish.
The Consume (SOAP-facade) direction modernizes the backend while keeping legacy SOAP clients working. It requires four grounding inputs, supplied in order:
Service identity → Old WSDL → REST OpenAPI → REST backend config (incl. auth)
→ Example payloads → Mapping review → Publish (WSDL serve-back)
The runtime accepts SOAP 1.1 or 1.2 inbound, converts to a REST call against the migrated backend (with outbound REST auth), and wraps the REST response back into a SOAP envelope. The consumer-facing contract is the WSDL served back at the consume endpoint, not OpenAPI.
See Consume direction (SOAP facade) for the full flow.
Every runtime invocation uses the same direction-prefixed path:
/{publish|consume}/{environment}/{serviceKey}/{operationPath...}
The direction segment is required; a path without it is rejected with HTTP 400. The
published OpenAPI servers[].url (Publish) and the emitted WSDL <soap:address> (Consume)
both carry the direction segment so generated clients hit the canonical route.