Skip to content

test-coverage-status.md

Latest commit

 

History

History
84 lines (66 loc) · 4.33 KB

test-coverage-status.md

File metadata and controls

84 lines (66 loc) · 4.33 KB

Test Coverage Status — v3.0

Last updated: 2026-05-24 (post-T094, GA gate set) Audit frequency: weekly during v3.0 implementation; one row per coverage-ratchet event.

Current gate (jest.config.js → coverageThreshold.global)

Metric Threshold Last measured Slack
Lines 40 % 58.22 %* +18 pts
Statements 40 % 57.34 % +17 pts
Functions 35 % 50.45 % +15 pts
Branches 25 % 45.75 % +20 pts

** Last measured numbers are from T046's unit-only run. Integration tests added in Phases 9 (Rollback), 10 (Notifications/Auth/Audit/AutoRecovery), and 13 (Projects/Users/Deployments) expand the denominator significantly; re-run with the full suite + Redis + DB up before tagging v3.0.0.

Client (vitest)

Metric Threshold Notes
Lines 30 % GA gate per NFR-006 (T094)
Statements 30 %
Functions 25 %
Branches 25 %

* Measured only against files actually loaded by the current test suite (unit tests). When integration tests come online (Phase 8+), the denominator expands; current slack will compress accordingly. Plan for the slack to shrink by ≥ 20 pts once notification + workspace + template features add their own untested src/ surface.

Ratchet schedule (research D-10)

Week Date target Gate target Tasks Status
1 2026-05-23 0 % (stub) T008 ✅ done
2 2026-05-30 20 % T046 ✅ done
3 2026-06-06 30 % T077 ✅ done
4 2026-06-13 40 % / 30 % client (GA gate) T094 ✅ done (this audit)

Files / areas currently uncovered (must improve for next ratchet)

Untouched src directories (need tests added in T075 / T077 / T092 / T093)

Area Coverage est. Coming in
Services/DeploymentService.ts < 5 % (only via import) T093 integration
Services/NotificationService.ts + dispatchers n/a (Phase 8 not built yet) T075
Services/AuditLogService.ts 0 % T078
Services/AutoRecovery.ts 0 % T078
Services/PipelineService.ts 0 % T093 (indirect, via deployment run)
Controllers/*.ts low T092 (Projects), T093 (Deployments), T077 (Auth)
Routes/*.ts n/a until integration covers them T092+T093
Migrations/*.ts excluded by config n/a
Models/*.ts low covered transitively via integration

Hot paths with existing tests but partial coverage

File Lines covered (approx) Notes
Utils/EncryptionHelper.ts 100 % T041
Utils/PasswordHelper.ts 100 % T041
Utils/LogFormatter.ts ~90 % T042 — Format* color path partly exercised
Utils/SshKeyGenerator.ts ~75 % T042 (generation tests skip when ssh-keygen probe fails)
Services/QueueService.ts ~80 % T043 (retry test opt-in via RUN_SLOW_QUEUE_TEST=1)
Services/EnvironmentVariableService.ts ~80 % covered via T044 integration

Operator notes / pre-existing tech debt blocking gate accuracy

Issue Impact Owner
Utils/SshKeyGenerator.CheckSshKeygenAvailable uses invalid ssh-keygen -V flag → always returns false → key-gen tests auto-skip even when ssh-keygen IS on PATH Generation tests never execute on most machines; coverage on gen paths stays at ~0 % Pre-existing v2.1 bug. Fix would be 1-line (use -y or which). Out of v3.0 scope unless promoted.
ESLint v9 needs eslint.config.js flat config — npm run lint fails silently with exit 0 today CI lint step is no-op Pre-existing v2.1 bug, separate task.
MigrationRunner.ts imports 008_... twice (for slots 007 + 008) — migration 007 effectively never runs Pre-existing data-shape risk Pre-existing v2.1 bug.

What an audit should check next week (before T077)

  1. Run npx jest --coverage --forceExit against the FULL suite (with Redis + DB up). Capture the global Lines % from coverage/coverage-summary.json.
  2. Confirm new tests written in Phase 8 (T075 + T076) lift the global number ≥ 30 %.
  3. If slack < +5 pts, identify lowest-coverage file via coverage/lcov-report/index.html and write one targeted test per gap.
  4. Bump coverageThreshold.global.lines to 30 (T077) only AFTER step 2 passes.