Import runtime gap-closure follow-up

• Status: Accepted
• Date: 2026-05-23
• Context:
  • The import job runtime worker landed, but two gaps remained:
    • prowlarr_api runtime handling used a single placeholder identifier and error reason without validating the persisted run configuration snapshot.
    • ERD_INDEXERS_CHECKLIST.md still marked the in-process scheduler item unchecked even though scheduler/runtime wiring had already shipped.
• Decision:
  • Tighten ImportJobRuntime prowlarr_api processing by validating required config snapshot keys (prowlarr_url, secret_public_id) before result recording.
  • Derive deterministic prowlarr_identifier from configured Prowlarr host/path prefix instead of a fixed placeholder string.
  • Keep live external API execution explicitly out of scope for this pass and continue failing closed with stable runtime detail codes.
  • Reconcile checklist truth by marking the in-process scheduler/worker item complete and annotating remaining runtime-executor sub-gaps explicitly.
• Consequences:
  • Positive outcomes:
    • Runtime output is more actionable and deterministic for operators reviewing import results.
    • Checklist status now matches shipped scheduler behavior while preserving open runtime executor work as unchecked.
  • Risks or trade-offs:
    • prowlarr_api still does not execute live remote imports; this change improves correctness/traceability but does not complete end-to-end payload ingestion.
• Follow-up:
  • Add live outbound import execution for prowlarr_api, including secret decryption/runtime adapter wiring and result materialization.
  • Add acceptance tests that validate runtime result identifiers and error details from real run configurations.

Task record

• Motivation:
  • Close immediate runtime and status-tracking gaps without overstating completion of the broader live import executor scope.
• Design notes:
  • Configuration parsing stays deterministic and local to runtime worker input; no new dependencies added.
  • Identifier derivation uses bounded truncation to preserve column limits and panic-free behavior.
• Test coverage summary:
  • Added pure unit coverage for URL-to-identifier derivation helper behavior.
  • Full validation remains just ci and just ui-e2e per repo policy.
• Observability updates:
  • Runtime failure details now distinguish missing runtime config from not-yet-implemented live executor behavior.
• Risk & rollback plan:
  • Rollback by reverting import_job_runtime.rs helper/branch logic and checklist/ADR documentation updates.
• Dependency rationale:
  • No new crates introduced; reused existing runtime and helper patterns.
• Stale-policy check:
  • Reviewed AGENTS.md, .github/instructions/rust.instructions.md, and .github/instructions/revaer-data.instructions.md.
  • No instruction contradictions found; removed checklist drift by aligning shipped scheduler status with actual implementation.