forcedotcom
diff --git a/‎.cursor/plans/e2e-plan.md‎
Lines changed: 36 additions & 0 deletions b/‎.cursor/plans/e2e-plan.md‎
Lines changed: 36 additions & 0 deletions
diff --git a/‎.cursor/plans/e2e_tests_for_playwright_vscode_extension_a7c36853.plan.md‎
Lines changed: 283 additions & 0 deletions b/‎.cursor/plans/e2e_tests_for_playwright_vscode_extension_a7c36853.plan.md‎
Lines changed: 283 additions & 0 deletions
@@ -0,0 +1,36 @@
+# end to end testing improvements
+
+we want e2e playwright tests for the commands in metadata extension
+keep in mind that test run in parallel but are "backed" by the same org, so make sure tests don't interfere with each other in flaky ways
+
+Code changes should only happen in the playwright tests AND/OR the playwright extension.
+
+It's possible that some code is really broken on windows and that e2e windows test failures are "true positive". If so, we let's report that but don't attempt to fix.
+
+## Commands in metadata extension
+
+1. `sf.project.deploy.start` - Project deploy start
+2. `sf.project.deploy.start.ignore.conflicts` - Project deploy start (ignore conflicts)
+3. `sf.project.retrieve.start` - Project retrieve start
+4. `sf.project.retrieve.start.ignore.conflicts` - Project retrieve start (ignore conflicts)
+5. `sf.view.all.changes` - View all changes
+6. `sf.view.local.changes` - View local changes
+7. `sf.view.remote.changes` - View remote changes
+8. `sf.source.tracking.reset.remote` - Reset remote tracking
+9. `sf.apex.generate.class` - Generate Apex class
+10. `sf.delete.source` - Delete source
+11. `sf.delete.source.current.file` - Delete source (current file)
+12. `sf.deploy.source.path` - Deploy source path
+13. `sf.deploy.active.editor` - Deploy active editor
+14. `sf.deploy.in.manifest` - Deploy in manifest
+15. `sf.retrieve.source.path` - Retrieve source path
+16. `sf.retrieve.current.source.file` - Retrieve current source file
+17. `sf.retrieve.in.manifest` - Retrieve in manifest
+
+## Things that don't matter and should be ignored
+
+- message about all extensions being deactivated
+
+# Mac right-click
+
+there is no way to run right click commands on mac. they should be skipped
@@ -0,0 +1,283 @@
+---
+name: E2E Tests for Playwright VSCode Extension
+overview: Create comprehensive end-to-end tests for the playwright-vscode-ext package that verify all foundational actions work correctly on Mac, Windows, and Web platforms without requiring an org or specific project setup.
+todos: []
+---
+
+# E2E Tests for Playwright VSCode Extension
+
+## Overview
+
+Create foundational e2e tests for `packages/playwright-vscode-ext` to verify core utilities work correctly across Mac, Windows, and Web platforms. These tests should run without an org or specific project/workspace setup.
+
+## Foundational Actions to Test
+
+Based on the exported functions from `packages/playwright-vscode-ext/src/index.ts`, we need to test:
+
+1. **Command Palette** (`executeCommandWithCommandPalette`)
+   - Opening command palette with F1
+   - Typing and executing commands
+   - Windows fallback behavior (Ctrl+Shift+P)
+
+2. **File Operations** (`createFileWithContents`, `openFileByName`, `editOpenFile`)
+   - Creating new files with content
+   - Opening files via Quick Open (Ctrl+P)
+   - Editing open files
+   - Saving files (Ctrl+S)
+
+3. **Settings** (`openSettingsUI`, `upsertSettings`)
+   - Opening settings UI
+   - Searching for settings
+   - Modifying textbox settings
+   - Modifying checkbox settings
+
+4. **Output Channel** (`ensureOutputPanelOpen`, `selectOutputChannel`, `waitForOutputChannelText`, `outputChannelContains`, `clearOutputChannel`)
+   - Opening output panel
+   - Selecting output channels
+   - Waiting for text to appear
+   - Checking if text exists
+   - Clearing output
+
+5. **Context Menus** (`executeEditorContextMenuCommand`, `executeExplorerContextMenuCommand`)
+   - Editor context menu (skip on Mac desktop)
+   - Explorer context menu (skip on Mac desktop)
+
+6. **Helper Functions** (`closeWelcomeTabs`, `waitForWorkspaceReady`, `waitForVSCodeWorkbench`)
+   - Closing welcome tabs
+   - Waiting for workspace to be ready
+   - Waiting for workbench to load
+
+## Test Structure
+
+### Directory Structure
+
+```
+packages/playwright-vscode-ext/
+├── test/
+│   ├── playwright/
+│   │   ├── fixtures/
+│   │   │   ├── index.ts          # Export test fixture (web/desktop)
+│   │   │   └── desktopFixtures.ts # Desktop-specific fixtures
+│   │   ├── pages/                # Page objects (if needed)
+│   │   ├── specs/
+│   │   │   ├── commandPalette.headless.spec.ts
+│   │   │   ├── fileOperations.headless.spec.ts
+│   │   │   ├── settings.headless.spec.ts
+│   │   │   ├── outputChannel.headless.spec.ts
+│   │   │   ├── contextMenu.headless.spec.ts
+│   │   │   └── helpers.headless.spec.ts
+│   │   └── web/
+│   │       └── headlessServer.ts # Web server setup
+│   ├── playwright.config.web.ts
+│   └── playwright.config.desktop.ts
+```
+
+### Test Files
+
+#### 1. `commandPalette.headless.spec.ts`
+
+- Test opening command palette with F1
+- Test executing a simple command (e.g., "View: Close All Editors")
+- Test Windows fallback to Ctrl+Shift+P
+- Verify command execution completes
+
+#### 2. `fileOperations.headless.spec.ts`
+
+- Test creating a new file with content
+- Test opening file via Quick Open (Ctrl+P)
+- Test editing an open file
+- Test saving file (Ctrl+S)
+- Verify file appears in editor tabs
+
+#### 3. `settings.headless.spec.ts`
+
+- Test opening settings UI
+- Test searching for a setting
+- Test modifying a textbox setting
+- Test modifying a checkbox setting
+- Verify settings are saved
+
+#### 4. `outputChannel.headless.spec.ts`
+
+- Test opening output panel
+- Test selecting an output channel
+- Test waiting for text in output
+- Test checking if text exists in output
+- Test clearing output channel
+
+#### 5. `contextMenu.headless.spec.ts`
+
+- Test editor context menu (skip on Mac desktop)
+- Test explorer context menu (skip on Mac desktop)
+- Verify menu items are clickable
+
+#### 6. `helpers.headless.spec.ts`
+
+- Test closing welcome tabs
+- Test waiting for workspace ready
+- Test waiting for workbench to load
+
+## Implementation Details
+
+### Fixtures
+
+Create `test/playwright/fixtures/index.ts` similar to org-browser:
+
+- Export `test` based on `VSCODE_DESKTOP` env var
+- Support both web and desktop fixtures
+- Add debug mode support
+
+Create `test/playwright/fixtures/desktopFixtures.ts`:
+
+- Use `createDesktopTest` from playwright-vscode-ext
+- Pass `fixturesDir: __dirname`
+
+### Web Server
+
+Create `test/playwright/web/headlessServer.ts`:
+
+- Use `createHeadlessServer` from playwright-vscode-ext
+- Extension name: "Playwright VSCode Ext"
+- Caller dirname: `__dirname`
+
+### Playwright Configs
+
+Create `playwright.config.web.ts`:
+
+- Use `createWebConfig()` from playwright-vscode-ext
+
+Create `playwright.config.desktop.ts`:
+
+- Use `createDesktopConfig()` from playwright-vscode-ext
+
+### Package.json Updates
+
+Add scripts:
+
+- `"test:web": "wireit"`
+- `"test:web:ui": "DEBUG_MODE=1 npm run test:web -- --headed"`
+- `"test:web:debug": "npm run test:web -- --debug"`
+- `"test:desktop": "wireit"`
+- `"test:desktop:debug": "npm run test:desktop -- --debug"`
+
+Add wireit configs for `test:web` and `test:desktop`:
+
+- Depend on `compile` and `../playwright-vscode-ext:compile`
+- Command: `playwright test --config=playwright.config.{web|desktop}.ts`
+- Files: `test/playwright/**/*.ts`, `playwright.config.{web|desktop}.ts`
+
+## GitHub Actions Workflow
+
+Create `.github/workflows/playwrightVscodeExtE2E.yml`:
+
+```yaml
+name: Playwright VSCode Ext E2E
+
+on:
+  workflow_dispatch:
+  pull_request:
+    paths:
+      - 'packages/playwright-vscode-ext/**'
+      - '.github/workflows/playwrightVscodeExtE2E.yml'
+
+jobs:
+  e2e-web:
+    runs-on: ubuntu-latest
+    timeout-minutes: 30
+    steps:
+      - Checkout
+      - Setup Node.js
+      - Install dependencies (npm ci)
+      - Install Playwright browsers
+      - Run web tests (npm run test:web -w playwright-vscode-ext)
+      - Upload Playwright reports
+
+  e2e-desktop-mac:
+    runs-on: macos-latest
+    timeout-minutes: 30
+    steps:
+      - Checkout
+      - Setup Node.js
+      - Install dependencies
+      - Install Playwright browsers
+      - Run desktop tests (VSCODE_DESKTOP=1 npm run test:desktop -w playwright-vscode-ext)
+      - Upload Playwright reports
+
+  e2e-desktop-windows:
+    runs-on: windows-latest
+    timeout-minutes: 30
+    steps:
+      - Checkout
+      - Setup Node.js
+      - Install dependencies
+      - Install Playwright browsers
+      - Run desktop tests (VSCODE_DESKTOP=1 npm run test:desktop -w playwright-vscode-ext)
+      - Upload Playwright reports
+```
+
+## Test Requirements
+
+### Platform Support
+
+- All tests must run on Mac, Windows, and Web
+- Context menu tests skip on Mac desktop (use `isMacDesktop()` check)
+- Use `test.skip()` for platform-specific skips
+
+### Test Isolation
+
+- Each test should be independent
+- No shared state between tests
+- Use minimal workspace (empty or minimal project)
+
+### Assertions
+
+- Use clear error messages in `expect()` assertions
+- Import `expect` from `@playwright/test`
+- Fail early, avoid fallbacks/retries
+
+### Code Style
+
+- Use `test.step()` to organize sequential tests
+- Prefer `getByRole` over CSS selectors
+- Use `Control` for all shortcuts (not `ControlOrMeta`)
+- Never use `waitForTimeout` - wait for specific elements
+- Never use Node.js fs/path or VS Code API - use UI interactions only
+
+## Files to Create/Modify
+
+### New Files
+
+- `packages/playwright-vscode-ext/test/playwright/fixtures/index.ts`
+- `packages/playwright-vscode-ext/test/playwright/fixtures/desktopFixtures.ts`
+- `packages/playwright-vscode-ext/test/playwright/web/headlessServer.ts`
+- `packages/playwright-vscode-ext/test/playwright/specs/commandPalette.headless.spec.ts`
+- `packages/playwright-vscode-ext/test/playwright/specs/fileOperations.headless.spec.ts`
+- `packages/playwright-vscode-ext/test/playwright/specs/settings.headless.spec.ts`
+- `packages/playwright-vscode-ext/test/playwright/specs/outputChannel.headless.spec.ts`
+- `packages/playwright-vscode-ext/test/playwright/specs/contextMenu.headless.spec.ts`
+- `packages/playwright-vscode-ext/test/playwright/specs/helpers.headless.spec.ts`
+- `packages/playwright-vscode-ext/playwright.config.web.ts`
+- `packages/playwright-vscode-ext/playwright.config.desktop.ts`
+- `.github/workflows/playwrightVscodeExtE2E.yml`
+
+### Modified Files
+
+- `packages/playwright-vscode-ext/package.json` (add test scripts and wireit configs)
+- `packages/playwright-vscode-ext/tsconfig.json` (ensure test files are included)
+
+## Testing Strategy
+
+1. Start with web tests locally
+2. Verify desktop tests locally (Mac first, then Windows)
+3. Create GitHub Actions workflow
+4. Run in CI and fix platform-specific issues
+5. Iterate: remove fallbacks, consolidate locators, increase DRY
+
+## Success Criteria
+
+- All tests pass on Mac, Windows, and Web
+- Tests run without requiring an org or specific project
+- Tests follow coding-playwright-tests.mdc rules
+- Tests follow iterating-playwright-tests.mdc sequence
+- GitHub Actions workflow runs successfully
+- No flaky tests or platform-specific failures