# E2E Testing Guide for Super Productivity This guide provides comprehensive information for writing and maintaining end-to-end tests for Super Productivity using Playwright. ## Table of Contents - [Overview](#overview) - [Running Tests](#running-tests) - [Test Structure](#test-structure) - [Page Objects](#page-objects) - [Common Patterns](#common-patterns) - [Selectors](#selectors) - [Wait Utilities](#wait-utilities) - [Writing New Tests](#writing-new-tests) - [Best Practices](#best-practices) - [Troubleshooting](#troubleshooting) --- ## Overview Our E2E tests are built with Playwright and follow the Page Object Model (POM) pattern for maintainability and reusability. Tests are organized by feature and use shared fixtures for common setup. ### Key Technologies - **Playwright**: Modern E2E testing framework - **TypeScript**: Type-safe test code - **Page Object Model**: Encapsulates page interactions - **Fixtures**: Shared setup and utilities --- ## Running Tests ### Basic Commands ```bash # Run all tests npm run e2e # Run tests in UI mode (interactive) npm run e2e:ui # Run a single test file with detailed output npm run e2e:file tests/task-basic/task-crud.spec.ts # Run tests in headed mode (see browser) npm run e2e:headed # Run tests in debug mode npm run e2e:debug # Show test report npm run e2e:show-report ``` ### WebDAV Sync Tests ```bash # Run WebDAV tests (starts Docker container) npm run e2e:webdav ``` --- ## Test Structure ### Directory Layout ``` e2e/ ├── constants/ # Shared selectors and constants │ └── selectors.ts # Centralized CSS selectors ├── fixtures/ # Test fixtures and setup │ └── test.fixture.ts # Custom test fixtures with page objects ├── helpers/ # Test helper functions │ └── plugin-test.helpers.ts ├── pages/ # Page Object Models │ ├── base.page.ts # Base page with common methods │ ├── work-view.page.ts │ ├── project.page.ts │ ├── task.page.ts │ ├── settings.page.ts │ ├── dialog.page.ts │ ├── planner.page.ts │ ├── schedule.page.ts │ ├── side-nav.page.ts │ ├── sync.page.ts │ ├── tag.page.ts │ └── note.page.ts ├── tests/ # Test specifications │ ├── task-basic/ │ ├── project/ │ ├── planner/ │ └── ... ├── utils/ # Utility functions │ ├── waits.ts # Wait helpers │ └── sync-helpers.ts ├── playwright.config.ts └── global-setup.ts ``` --- ## Page Objects Page Objects encapsulate interactions with specific pages or components. All page objects extend `BasePage` and receive a `page` and optional `testPrefix`. ### Available Page Objects #### 1. **BasePage** (`base.page.ts`) Base class for all page objects. Provides common functionality: ```typescript class BasePage { async addTask(taskName: string): Promise; // Adds a task with automatic test prefix } ``` **Example:** ```typescript await workViewPage.addTask('My Task'); // Creates task with name "W0-P0-My Task" (prefixed for isolation) ``` #### 2. **WorkViewPage** (`work-view.page.ts`) Interactions with the main work view: ```typescript class WorkViewPage extends BasePage { async waitForTaskList(): Promise; async addSubTask(task: Locator, subTaskName: string): Promise; } ``` **Example:** ```typescript await workViewPage.waitForTaskList(); await workViewPage.addTask('Parent Task'); const task = page.locator('task').first(); await workViewPage.addSubTask(task, 'Child Task'); ``` #### 3. **TaskPage** (`task.page.ts`) Task-specific operations: ```typescript class TaskPage extends BasePage { getTask(index: number): Locator; getTaskByText(text: string): Locator; async markTaskAsDone(task: Locator): Promise; async editTaskTitle(task: Locator, newTitle: string): Promise; async openTaskDetail(task: Locator): Promise; async getTaskCount(): Promise; async isTaskDone(task: Locator): Promise; getDoneTasks(): Locator; getUndoneTasks(): Locator; async waitForTaskWithText(text: string): Promise; async taskHasTag(task: Locator, tagName: string): Promise; } ``` **Example:** ```typescript const task = taskPage.getTask(1); // First task await taskPage.markTaskAsDone(task); await expect(taskPage.getDoneTasks()).toHaveCount(1); ``` #### 4. **ProjectPage** (`project.page.ts`) Project management: ```typescript class ProjectPage extends BasePage { async createProject(projectName: string): Promise; async navigateToProjectByName(projectName: string): Promise; async createAndGoToTestProject(): Promise; async addNote(noteContent: string): Promise; async archiveDoneTasks(): Promise; } ``` **Example:** ```typescript await projectPage.createProject('My Project'); await projectPage.navigateToProjectByName('My Project'); await projectPage.addNote('Project notes here'); ``` #### 5. **SettingsPage** (`settings.page.ts`) Settings and configuration: ```typescript class SettingsPage extends BasePage { async navigateToSettings(): Promise; async expandSection(sectionSelector: string): Promise; async expandPluginSection(): Promise; async navigateToPluginSettings(): Promise; async enablePlugin(pluginName: string): Promise; async disablePlugin(pluginName: string): Promise; async isPluginEnabled(pluginName: string): Promise; async uploadPlugin(pluginPath: string): Promise; } ``` **Example:** ```typescript await settingsPage.navigateToPluginSettings(); await settingsPage.enablePlugin('Test Plugin'); expect(await settingsPage.isPluginEnabled('Test Plugin')).toBeTruthy(); ``` #### 6. **DialogPage** (`dialog.page.ts`) Dialog and modal interactions: ```typescript class DialogPage extends BasePage { async waitForDialog(): Promise; async waitForDialogToClose(): Promise; async clickDialogButton(buttonText: string): Promise; async clickSaveButton(): Promise; async fillDialogInput(selector: string, value: string): Promise; async fillMarkdownDialog(content: string): Promise; async saveMarkdownDialog(): Promise; async editDateTime(dateValue?: string, timeValue?: string): Promise; } ``` **Example:** ```typescript await dialogPage.waitForDialog(); await dialogPage.fillDialogInput('input[name="title"]', 'New Title'); await dialogPage.clickSaveButton(); await dialogPage.waitForDialogToClose(); ``` --- ## Common Patterns ### Pattern 1: Basic Task CRUD ```typescript test('should create and edit task', async ({ page, workViewPage, taskPage }) => { await workViewPage.waitForTaskList(); // Create await workViewPage.addTask('Test Task'); await expect(taskPage.getAllTasks()).toHaveCount(1); // Edit const task = taskPage.getTask(1); await taskPage.editTaskTitle(task, 'Updated Task'); await expect(taskPage.getTaskTitle(task)).toContainText('Updated Task'); // Mark as done await taskPage.markTaskAsDone(task); await expect(taskPage.getDoneTasks()).toHaveCount(1); }); ``` ### Pattern 2: Project Workflow ```typescript test('should create project and add tasks', async ({ projectPage, workViewPage }) => { await projectPage.createAndGoToTestProject(); await workViewPage.addTask('Project Task 1'); await workViewPage.addTask('Project Task 2'); await expect(page.locator('task')).toHaveCount(2); }); ``` ### Pattern 3: Settings Configuration ```typescript test('should enable plugin', async ({ settingsPage, waitForNav }) => { await settingsPage.navigateToPluginSettings(); await settingsPage.enablePlugin('My Plugin'); await waitForNav(); expect(await settingsPage.isPluginEnabled('My Plugin')).toBeTruthy(); }); ``` ### Pattern 4: Dialog Interactions ```typescript test('should edit date in dialog', async ({ taskPage, dialogPage }) => { const task = taskPage.getTask(1); await taskPage.openTaskDetail(task); const dateInfo = dialogPage.getDateInfo('Created'); await dateInfo.click(); await dialogPage.editDateTime('12/25/2025', undefined); await dialogPage.clickSaveButton(); }); ``` --- ## Selectors All selectors are centralized in `constants/selectors.ts`. Always use these constants instead of hardcoding selectors in tests. ### Using Selectors ```typescript import { cssSelectors } from '../constants/selectors'; const { TASK, TASK_TITLE, TASK_DONE_BTN } = cssSelectors; // In test: const task = page.locator(TASK).first(); const title = task.locator(TASK_TITLE); ``` ### Selector Categories - **Navigation**: `SIDENAV`, `NAV_ITEM`, `SETTINGS_BTN` - **Layout**: `ROUTE_WRAPPER`, `BACKDROP`, `PAGE_TITLE` - **Tasks**: `TASK`, `TASK_TITLE`, `TASK_DONE_BTN`, `SUB_TASK` - **Add Task**: `ADD_TASK_INPUT`, `ADD_TASK_SUBMIT` - **Dialogs**: `MAT_DIALOG`, `DIALOG_FULLSCREEN_MARKDOWN` - **Settings**: `PAGE_SETTINGS`, `PLUGIN_SECTION`, `PLUGIN_MANAGEMENT` - **Projects**: `PAGE_PROJECT`, `CREATE_PROJECT_BTN`, `WORK_CONTEXT_MENU` --- ## Wait Utilities Located in `utils/waits.ts`, these utilities help handle Angular's async nature. ### Available Wait Functions #### `waitForAngularStability(page, timeout?)` Waits for Angular to finish all async operations. ```typescript await waitForAngularStability(page); ``` #### `waitForAppReady(page, options?)` Comprehensive wait for app initialization. ```typescript await waitForAppReady(page, { selector: 'task-list', ensureRoute: true, routeRegex: /#\/project\/\w+/, }); ``` #### `waitForStatePersistence(page)` Waits for IndexedDB persistence to complete (important before sync operations). ```typescript await workViewPage.addTask('Task'); await waitForStatePersistence(page); // Ensure saved to IndexedDB // Now safe to trigger sync ``` --- ## Writing New Tests ### Step 1: Create Test File ```typescript // e2e/tests/my-feature/my-feature.spec.ts import { test, expect } from '../../fixtures/test.fixture'; test.describe('My Feature', () => { test('should do something', async ({ page, workViewPage, taskPage }) => { // Test code here }); }); ``` ### Step 2: Use Page Objects ```typescript test('my test', async ({ workViewPage, taskPage, dialogPage }) => { // Wait for page ready await workViewPage.waitForTaskList(); // Use page objects for interactions await workViewPage.addTask('Task 1'); const task = taskPage.getTask(1); await taskPage.markTaskAsDone(task); // Assertions await expect(taskPage.getDoneTasks()).toHaveCount(1); }); ``` ### Step 3: Handle Waits Properly ```typescript // GOOD: Use Angular stability waits await workViewPage.addTask('Task'); await waitForAngularStability(page); await expect(page.locator('task')).toBeVisible(); // BAD: Arbitrary timeouts await page.waitForTimeout(5000); // Avoid unless necessary ``` ### Step 4: Use Selectors from Constants ```typescript import { cssSelectors } from '../../constants/selectors'; const { TASK, TASK_TITLE } = cssSelectors; const title = page.locator(TASK).first().locator(TASK_TITLE); ``` --- ## Best Practices ### ✅ DO 1. **Use page objects** for all interactions 2. **Use centralized selectors** from `constants/selectors.ts` 3. **Wait for Angular stability** after state changes 4. **Use test prefixes** (automatic via fixtures) for isolation 5. **Test one thing per test** - keep tests focused 6. **Use descriptive test names** - "should create task and mark as done" 7. **Clean up state** - tests should be independent 8. **Use role-based selectors** when possible (accessibility) ```typescript // GOOD await page.getByRole('button', { name: 'Save' }).click(); // LESS GOOD await page.locator('.save-btn').click(); ``` ### ❌ DON'T 1. **Don't hardcode selectors** - use `cssSelectors` 2. **Don't use arbitrary waits** - use `waitForAngularStability` 3. **Don't share state between tests** - each test should be independent 4. **Don't access DOM directly** - use page objects 5. **Don't skip error handling** - tests should fail clearly 6. **Don't use `any` types** - maintain type safety ### Test Isolation Each test gets: - Isolated browser context (clean storage) - Unique test prefix (`W0-P0-`, `W1-P0-`, etc.) - Fresh page instance This ensures tests don't interfere with each other. ### Handling Flakiness ```typescript // Use waitFor with explicit conditions await page.waitForFunction(() => document.querySelectorAll('task').length === 3, { timeout: 10000, }); // Use locator assertions (auto-retry) await expect(page.locator('task')).toHaveCount(3); // Avoid fixed timeouts await page.waitForTimeout(1000); // BAD await waitForAngularStability(page); // GOOD ``` --- ## Troubleshooting ### Test Fails with "Element not found" 1. Check if selector is correct in `constants/selectors.ts` 2. Add wait before interaction: `await waitForAngularStability(page)` 3. Use `await element.waitFor({ state: 'visible' })` 4. Check if element is in a different context (iframe, shadow DOM) ### Test Timeout 1. Increase timeout in specific waitFor calls 2. Check if Angular is stuck - look for pending HTTP requests 3. Use `page.pause()` to debug interactively 4. Check network tab for failed requests ### Flaky Tests 1. Add proper waits: `waitForAngularStability`, `waitForAppReady` 2. Avoid `page.waitForTimeout()` - use condition-based waits 3. Check for race conditions - ensure state is persisted 4. Use `waitForStatePersistence` before operations that depend on saved state ### Debugging ```typescript // Pause execution and open Playwright Inspector await page.pause(); // Take screenshot await page.screenshot({ path: 'debug.png' }); // Console log page content console.log(await page.content()); // Get element text for debugging const text = await page.locator('task').first().textContent(); console.log('Task text:', text); ``` ### Running Single Test ```bash # Run specific file npm run e2e:file tests/task-basic/task-crud.spec.ts # Run in debug mode npm run e2e:debug # Run in headed mode to see browser npm run e2e:headed ``` --- ## Examples ### Example 1: Full Task CRUD Test ```typescript import { test, expect } from '../../fixtures/test.fixture'; test.describe('Task CRUD', () => { test('should create, edit, and delete tasks', async ({ page, workViewPage, taskPage, }) => { await workViewPage.waitForTaskList(); // Create await workViewPage.addTask('Task 1'); await workViewPage.addTask('Task 2'); await expect(taskPage.getAllTasks()).toHaveCount(2); // Edit const firstTask = taskPage.getTask(1); await taskPage.editTaskTitle(firstTask, 'Updated Task'); await expect(taskPage.getTaskTitle(firstTask)).toContainText('Updated Task'); // Mark as done await taskPage.markTaskAsDone(firstTask); await expect(taskPage.getDoneTasks()).toHaveCount(1); await expect(taskPage.getUndoneTasks()).toHaveCount(1); }); }); ``` ### Example 2: Project Workflow ```typescript test('should create project with tasks', async ({ projectPage, workViewPage, taskPage, }) => { await projectPage.createAndGoToTestProject(); await workViewPage.addTask('Project Task'); await projectPage.addNote('Important notes'); const task = taskPage.getTask(1); await taskPage.markTaskAsDone(task); await projectPage.archiveDoneTasks(); await expect(taskPage.getUndoneTasks()).toHaveCount(0); }); ``` ### Example 3: Settings Test ```typescript test('should configure plugin', async ({ settingsPage, page }) => { await settingsPage.navigateToPluginSettings(); const pluginExists = await settingsPage.pluginExists('Test Plugin'); expect(pluginExists).toBeTruthy(); await settingsPage.enablePlugin('Test Plugin'); expect(await settingsPage.isPluginEnabled('Test Plugin')).toBeTruthy(); await settingsPage.navigateBackToWorkView(); await expect(page).toHaveURL(/tag\/TODAY/); }); ``` --- ## Getting Help - Check existing tests in `e2e/tests/` for examples - Review page objects in `e2e/pages/` for available methods - Look at `constants/selectors.ts` for available selectors - Use Playwright Inspector (`npm run e2e:debug`) for debugging - Check Playwright docs: https://playwright.dev/ --- ## Summary Checklist When writing a new test: - [ ] Create test file in appropriate `tests/` subdirectory - [ ] Import `test` and `expect` from `fixtures/test.fixture.ts` - [ ] Use page objects for all interactions - [ ] Use selectors from `constants/selectors.ts` - [ ] Add proper waits (`waitForAngularStability`, etc.) - [ ] Use descriptive test names - [ ] Ensure test is isolated (no shared state) - [ ] Run test locally before committing - [ ] Test passes consistently (run 3+ times)