How to programmatically change the order in which tests are executed?

and one of the common causes of flaky tests is interdependency of tests. Isn't that mostly solved by Playwright Test encouraging, by design, tests to not be interdependent?

The way you could achieve this is by removing the serialisation of tests that'd be required for them to be interdependent

Although not sure about the value here given that users would most likely know their tests are interdependent and therefore that any change to their order or removing serialisation will break them

Isn't that mostly solved by Playwright Test encouraging, by design, tests to not be interdependent?

There is nothing stopping tests from being dependent on each other at the moment.

I was just debugging a case the other day where another test created an authenticated session and it was leaking to the next test

There is nothing stopping tests from being dependent on each other at the moment.

That's pretty much the point of browser contexts, each test gets it's own context & page, unless you purposely re-use a page object between tests: https://playwright.dev/docs/test-retries#reuse-single-page-between-tests https://playwright.dev/docs/browser-contexts

I was just debugging a case the other day where another test created an authenticated session and it was leaking to the next test That's not a flaky test though, that issue would be the same every test run.