Skip to content

test(openapi): add executable coverage checks between client handlers and bundled spec #67

@joshrotenberg

Description

@joshrotenberg

Summary

The repo vendors an OpenAPI fixture and has some validation tests, but the current checks are mostly structural. They verify that the fixture loads, has enough endpoints, and contains expected categories/schemas. They do not currently prove that the client covers the advertised routes or that uncovered API paths are explicitly tracked.

Why this matters

The README claims comprehensive REST API coverage, but CI does not have an executable check that connects handler methods to the bundled spec. That makes it easy for coverage claims and implementation reality to drift apart.

Current validation

  • tests/openapi_validation.rs
  • tests/fixtures/cloud_openapi.json

Gaps

Suggested direction

  • define a route inventory for the Rust client (manually maintained table or generated metadata)
  • compare that inventory against the bundled spec in CI
  • keep an explicit allowlist for intentionally deferred endpoints
  • use the resulting report to drive follow-up coverage issues instead of relying on README claims alone

Acceptance criteria

  • CI fails when a spec path/method becomes uncovered without an explicit allowlist entry
  • the repo contains a machine-readable inventory or mapping of supported routes
  • intentionally unsupported or deferred endpoints are tracked explicitly
  • README/docs coverage claims align with the executable check

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type
    No fields configured for issues without a type.

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions