coverage-run

coverage-run

Execute tests with coverage collection across multiple stacks.

Commands

/coverage-run              # Run all stacks
/coverage-run unit         # Vitest only
/coverage-run e2e          # Playwright only
/coverage-run python       # pytest only
/coverage-run api          # Vitest for packages/api only
/coverage-run web          # Vitest for apps/web only

Execution Flow

1. Read Configuration

cat coverage-config.json

Extract stack commands and report paths from the config.

2. Auto-Detect Stack

Input	Stack	Command
unit, vitest	Vitest	pnpm test -- --coverage (turbo passthrough)
e2e, playwright	Playwright	pnpm -C apps/web test:e2e
python, pytest	pytest	cd apps/ana-service && pytest --cov
api, database, etc	Vitest (scoped)	pnpm test --filter="@repo/{name}" -- --coverage
(none)	All	Run sequentially

Important: Use -- --coverage to pass args through Turborepo to Vitest.

3. Pre-flight Checks

Before running tests:

# Check if DB is accessible (for integration tests)
docker ps | grep postgres || echo "Warning: PostgreSQL not running"

# Check if Python venv exists for pytest
test -d apps/ana-service/.venv || echo "Warning: Python venv not found"

4. Execute Coverage

Vitest (all packages):

# Use -- to pass args through Turborepo
pnpm test -- --coverage --coverage.reporter=json-summary --coverage.reporter=json

Vitest (filtered - recommended):

# Filter to packages with working test scripts
pnpm test --filter="@repo/api" --filter="@repo/database" --filter="@repo/validators" --filter="@repo/logger" -- --coverage --coverage.reporter=json-summary --coverage.reporter=json

Note: Some packages (e.g., @repo/test-utils) have test scripts that don't accept coverage args. Use --filter to exclude them if needed.

Playwright (route coverage - count specs):

# Count specs and calculate route coverage
SPECS=$(find apps/web/e2e/flows -name "*.spec.ts" | wc -l)
TOTAL_ROUTES=216  # from e2e-coverage-plan.md
echo "E2E Coverage: $SPECS specs covering routes"