# Towerops E2E Tests

End-to-end tests using Playwright for testing the Towerops application against local development or staging environments.

## Setup

### 1. Install Dependencies

```bash
cd e2e
npm install
npx playwright install
```

### 2. Create Test User

You need a test user with TOTP enabled:

**Option A: Using Phoenix Console (Recommended)**

```elixir
# Start console
cd ..
iex -S mix

# Create test user
alias Towerops.{Accounts, Organizations, Repo}

{:ok, user} = Accounts.register_user(%{
  email: "test@example.com",
  password: "TestPassword123!",
  password_confirmation: "TestPassword123!"
})

# Enable TOTP (this will print the secret)
{:ok, credential} = Accounts.create_user_credential(user, %{
  type: :totp,
  totp_secret: "JBSWY3DPEHPK3PXP"  # Use this secret in .env
})

# Create an organization for the test user
{:ok, org} = Organizations.create_organization(%{name: "Test Org"}, user.id)
```

**Option B: Using the Web UI**

1. Register a new user at http://localhost:4000/users/register
2. Log in and enable TOTP in settings
3. Get the TOTP secret from the database:
   ```sql
   SELECT totp_secret FROM user_credentials WHERE user_id = '<user-id>';
   ```

### 3. Configure Environment

```bash
cp .env.example .env
```

Edit `.env` and set:
- `TEST_USER_EMAIL` - Your test user email
- `TEST_USER_PASSWORD` - Your test user password
- `TEST_USER_TOTP_SECRET` - TOTP secret from step 2

## Running Tests

### Local Development

Make sure Phoenix is running first:
```bash
cd ..
mix phx.server
```

Then run tests:
```bash
cd e2e
npm test
```

### Against Staging

```bash
npm run test:staging
```

### Other Commands

```bash
# Run tests with UI (interactive mode)
npm run test:ui

# Run tests with browser visible
npm run test:headed

# Debug a specific test
npm run test:debug -- tests/devices.spec.ts

# Generate tests by recording actions
npm run codegen

# View last test report
npm run report
```

## Test Organization

### `tests/auth.setup.ts`

Authentication setup that runs before all other tests. Logs in once and saves the session state so other tests don't need to authenticate.

### Test Files

- `organizations.spec.ts` - Organization listing and switching
- `devices.spec.ts` - Device listing, details, search/filter
- `alerts.spec.ts` - Alert viewing, filtering, resolution

### Adding New Tests

1. Create a new `.spec.ts` file in `tests/`
2. Import test utilities:
   ```typescript
   import { test, expect } from '@playwright/test';
   ```
3. Write tests using Playwright's API

Example:
```typescript
import { test, expect } from '@playwright/test';

test.describe('My Feature', () => {
  test('can do something', async ({ page }) => {
    await page.goto('/my-page');
    await expect(page.getByRole('heading')).toBeVisible();
  });
});
```

## Configuration

### Environment Variables

- `BASE_URL` - Base URL of the application (default: http://localhost:4000)
- `TEST_USER_EMAIL` - Test user email
- `TEST_USER_PASSWORD` - Test user password
- `TEST_USER_TOTP_SECRET` - TOTP secret for test user

### Playwright Config

Edit `playwright.config.ts` to:
- Add/remove browsers
- Change timeout settings
- Configure reporters
- Add mobile viewports
- Enable video recording

## Best Practices

### 1. Use Data Test Attributes

Add `data-test` attributes to important elements in your templates:

```heex
<div data-test="device-card">
  <h3 data-test="device-name"><%= @device.name %></h3>
</div>
```

Then in tests:
```typescript
await page.locator('[data-test="device-card"]').first().click();
```

### 2. Wait for Network Idle

For LiveView pages with dynamic content:
```typescript
await page.waitForLoadState('networkidle');
```

### 3. Use Semantic Selectors

Prefer role-based selectors when possible:
```typescript
await page.getByRole('button', { name: 'Submit' }).click();
await page.getByRole('heading', { name: 'Devices' }).toBeVisible();
```

### 4. Handle Optional Elements

Check if elements exist before interacting:
```typescript
if (await page.getByText('No results').count() > 0) {
  // Handle empty state
}
```

### 5. Avoid Hard Timeouts

Use `waitFor` methods instead of `waitForTimeout`:
```typescript
// Good
await page.getByText('Success').waitFor();

// Bad
await page.waitForTimeout(3000);
```

## CI/CD Integration

### GitHub Actions / GitLab CI

```yaml
e2e-tests:
  runs-on: ubuntu-latest
  steps:
    - uses: actions/checkout@v3
    - uses: actions/setup-node@v3
      with:
        node-version: 18
    - name: Install dependencies
      run: |
        cd e2e
        npm ci
        npx playwright install --with-deps
    - name: Run tests
      run: |
        cd e2e
        BASE_URL=${{ secrets.STAGING_URL }} npm test
      env:
        TEST_USER_EMAIL: ${{ secrets.TEST_USER_EMAIL }}
        TEST_USER_PASSWORD: ${{ secrets.TEST_USER_PASSWORD }}
        TEST_USER_TOTP_SECRET: ${{ secrets.TEST_USER_TOTP_SECRET }}
    - name: Upload test results
      if: always()
      uses: actions/upload-artifact@v3
      with:
        name: playwright-report
        path: e2e/playwright-report/
```

## Debugging

### Visual Debugging

```bash
# Open test in UI mode
npm run test:ui

# Run specific test with browser visible
npm run test:headed -- tests/devices.spec.ts
```

### Debug Mode

```bash
# Opens Playwright Inspector
npm run test:debug -- tests/devices.spec.ts
```

### Screenshots and Videos

Failed tests automatically capture:
- Screenshots (in `test-results/`)
- Videos (in `test-results/`)
- Traces (view with `npx playwright show-trace <trace-file>`)

### Console Logs

Add console logging in tests:
```typescript
test('debug test', async ({ page }) => {
  page.on('console', msg => console.log('Browser:', msg.text()));
  await page.goto('/devices');
});
```

## Troubleshooting

### "Authentication failed" or TOTP errors

- Verify TOTP secret is correct in `.env`
- Check test user exists and has TOTP enabled
- Ensure test user has access to at least one organization

### Tests timing out

- Increase timeout in `playwright.config.ts`
- Check Phoenix server is running
- Verify BASE_URL is correct
- Check network connectivity to staging

### "Element not found" errors

- Add `data-test` attributes to elements
- Use `page.waitForLoadState('networkidle')` before assertions
- Check if LiveView is fully mounted before interacting

### Browser not launching

```bash
# Reinstall browsers
npx playwright install --with-deps
```

## Resources

- [Playwright Documentation](https://playwright.dev/)
- [Playwright Test API](https://playwright.dev/docs/api/class-test)
- [Best Practices](https://playwright.dev/docs/best-practices)
- [Debugging Guide](https://playwright.dev/docs/debug)