Salesforce unit test runner, built over @salesforce/apex-node and jsforce. The runner provides a number of reliability features aimed at ensuring the greatest chance of obtaining a good test run.

The main abstraction is a Testall command. You can use this to configure and execute a test run. The runner is assisted by three plugins:

• TestMethodCollector - Locates all test methods that should be executed, this is used to validate there are no missing methods in a run and if necessary arrange for additional test to be executed to fill the gaps.
• TestRunner - This manages the initial test run and subsequent runs that may be needed due to missing tests. It monitors the test runs for progress and can abort/restart runs when progress is not being made.
• OutputGenerator - This manages the generation of test reports, by default JUnit XML and json style outputs are created.

The Testall command also handles re-running tests that failed due to locking after the main run(s) have completed.

You can start a Testall run by providing a set of test classes. If you don't provide any the run will execute against all the "local" (non-packaged) tests in the org.

This project uses the pnpm package manager.

  pnpm install
  pnpm build

To run unit tests:

  pnpm test

To test bundling using webpack:

  pnpm test:pack
  node test-bundle/bundle.js

This should execute without error.

See ./script for some basic scripts to aid org testing. Execute them with ts-node using the run script:

  # Run Testall with default settings
  pnpm run:script -- ./src/scripts/Testall.ts <username> <namespace | unmanaged>

If you encounter difficulties with the installed git hooks, they can be bypassed with --no-verify/-n flag on commit. To disable them completely (after install) use npx husky uninstall or the direct command git config --unset core.hooksPath.

All the source code included uses a 3-clause BSD license, see LICENSE for details.