Skip to content

test.md

Latest commit

 

History

History
93 lines (69 loc) · 5.04 KB

test.md

File metadata and controls

93 lines (69 loc) · 5.04 KB
Raw

Test Pipeline

This repo uses a small unit-test layer for pure TypeScript modules plus the existing type-check and build verification steps.

Local Commands

npm run test
npm run test:watch
npm run check
npm run build

Recommended local sequence before pushing:

  1. npm run test
  2. npm run check
  3. npm run build

For PWA or speech work, also verify the built app exposes the expected manifest/service-worker assets and that browser-only speech helpers degrade cleanly when APIs are unavailable.

Current Scope

The current unit tests focus on fast, deterministic modules that do not require live provider APIs, a browser, or a database:

  • src/lib/utils/note-structure.test.ts
    • verifies the shared assistant note-section contract
    • checks required section order, approved optional-section placement, deprecated-heading rejection, and heading normalization
  • src/lib/server/ai/prompts.test.ts
    • verifies create/update assistant prompts both inject the same shared note-structure contract
  • src/lib/server/assistant/routing.test.ts
    • verifies inference-first routing keeps plain learning prompts conversational
    • verifies follow-up asks like add it to notes or save this to my notes route into note creation
  • src/lib/client/speech.test.ts
    • verifies canSpeak and canListen return false in SSR (no window) and correctly detect API presence using vi.stubGlobal
    • verifies extractHtmlText strips tags, decodes common entities, and collapses whitespace
    • verifies extractMarkdownText strips headings, bold/italic, links, images, wikilinks, code blocks, and list markers
    • verifies speak and stopSpeaking manage utterance lifecycle, cancel before starting new speech, trigger onEnd on end/error events, and no-op when the API is absent
    • browser speech APIs are mocked via vi.stubGlobal; no real microphone or system voices required
  • src/lib/server/practice/practice.test.ts
    • verifies validateManualImport accepts valid complete and minimal payloads, rejects missing or whitespace-only required fields, trims whitespace, and filters empty tag strings
    • verifies isValidStatus accepts all four documented status values and rejects invalid strings, null, and numeric values
    • verifies validateProgressInput accepts valid minimal and full payloads; rejects missing fields, invalid status, negative or fractional attempts, non-string notes, and non-string non-null codeSnapshot; accepts null codeSnapshot
    • verifies LeetCodeFetchDisabledError and LeetCodeFetchUpstreamError carry their documented error codes and names
    • verifies fetchLeetCodeDaily throws LeetCodeFetchDisabledError when LEETCODE_DAILY_FETCH_ENABLED is absent or not "true", confirming the disabled state documented in docs/API.md
    • verifies validateTutorInput accepts valid minimal and full payloads; rejects missing fields, invalid hint levels, non-string code, and disallowed model identifiers; accepts all documented hint levels
    • verifies isValidHintLevel accepts the four documented levels and rejects out-of-vocabulary strings
    • verifies the tutor module's non-persistence contract: tutor.ts exports no conversation creation or message append functions, confirming that tutor turns are never written to conversations or conversation_messages
    • $env/dynamic/private and the DB module are mocked via vi.mock so tests run without live secrets or a database connection

Remaining planned additions for the PWA and voice phase:

  • route tests for optional speech transcription error states where practical
  • build checks that the PWA manifest and service-worker registration compile without server-only import leaks

Shared Contract

The assistant note-section rules live in src/lib/utils/note-structure.ts.

That module is the source of truth for:

  • required sections
  • approved optional sections
  • deprecated headings
  • the shared prompt guidance text
  • pure note-body validation used by the assistant commit boundary

This keeps prompt-time instructions and commit-time validation aligned.

CI Pipeline

GitHub Actions runs .github/workflows/ci.yml on pushes to main and on pull requests targeting main.

The CI sequence is:

  1. npm ci
  2. npm run test
  3. npm run build
  4. npm run check

Build still uses dummy private env values so the workflow catches server-only import leaks without requiring real secrets.

Adding Tests

Prefer unit tests for pure modules first. They are faster, easier to maintain, and run cleanly in CI without app bootstrapping.

When adding tests:

  • keep them close to the module under test with *.test.ts
  • avoid live network calls and provider SDK calls
  • mock or isolate runtime-heavy boundaries instead of depending on external services
  • add route-level or browser-level tests only when unit coverage is not enough to protect the behavior
  • for speech features, mock browser speech APIs instead of depending on device microphone, installed system voices, or live transcription providers
  • for PWA features, prefer build/output verification and manifest assertions over brittle install-prompt automation