8831478021
* agentHost: run E2E suites in parallel Run the deterministic conformance and provider entrypoints concurrently while preserving serial server reuse within each suite. Integrate the parallel runner into coverage and full integration runs.\n\n(Written by Copilot)\n\nCo-authored-by: Copilot <223556219+Copilot@users.noreply.github.com> * agentHost: harden parallel E2E runner Preserve Windows argument boundaries through a PowerShell wrapper and fail coverage runs when a worker does not emit protocol-surface observations. (Written by Copilot) Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com> * agentHost: stabilize terminal clear E2E test Wait for a marker that appears only in command output, not in the shell's echoed input, before clearing terminal state. (Written by Copilot) Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com> --------- Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
Unit Tests
Run (inside Electron)
./scripts/test.[sh|bat]
All unit tests are run inside a Electron renderer environment which access to DOM and Nodejs api. This is the closest to the environment in which VS Code itself ships. Notes:
- use the
--debugto see an electron window with dev tools which allows for debugging - to run only a subset of tests use the
--runor--globoptions - use
npm run watchto automatically compile changes
For instance, ./scripts/test.sh --debug --glob **/extHost*.test.js runs all tests from extHost-files and enables you to debug them.
Run (inside browser)
npm run test-browser -- --browser webkit --browser chromium
Unit tests from layers common and browser are run inside chromium, webkit, and (soon'ish) firefox (using playwright). This complements our electron-based unit test runner and adds more coverage of supported platforms. Notes:
- these tests are part of the continuous build, that means you might have test failures that only happen with webkit on windows or chromium on linux
- you can run these tests locally via
npm run test-browser -- --browser chromium --browser webkit - to debug, open
<vscode>/test/unit/browser/renderer.htmlinside a browser and use the?m=<amd_module>-query to specify what AMD module to load, e.gfile:///Users/jrieken/Code/vscode/test/unit/browser/renderer.html?m=vs/base/test/common/strings.testruns all tests fromstrings.test.ts - to run only a subset of tests use the
--runor--globoptions
Note: you can enable verbose logging of playwright library by setting a DEBUG environment variable before running the tests (https://playwright.dev/docs/debug#verbose-api-logs)
Run (with node)
npm run test-node -- --run src/vs/editor/test/browser/controller/cursor.test.ts
Coverage
The following command will create a coverage folder in the .build folder at the root of the workspace:
OS X and Linux
./scripts/test.sh --coverage
Windows
scripts\test --coverage