[BUG] test.afterAll will return error if test.skip at the last test

[xyngfei]

System info

• Playwright Version: v1.27.1
• Operating System: 13.2.1 (22D68)
• Browser: Chromium
• Other info:

Source code

• [x] I provided exact source code that allows reproducing the issue locally.

or

Test file (self-contained)

import { test } from "@playwright/test";

test.describe.serial("demo", () => {
  test.beforeEach(async ({ page }) => {
    await page.goto("https://demo.applitools.com/index_v2.html");
  });

  test.afterAll(async ({ page }, testInfo) => {
    console.log("test: " + testInfo.title.indexOf("@api"));
    await page.close();
  });

  test("111 @demo", async ({}) => {
    console.log("111 test");
  });

  test("222 @demo", async ({}) => {
    console.log("222 test");
  });

  test.skip("333 @demo", async ({}) => {
    console.log("333 test");
  });
});

Steps

• [Run the test]
• [...]

Expected

It should skip the test without any issue

Actual

[pavelfeldman]

As stated in the error message, you error is that the { page } fixture is used in afterAll. Each test has its own page, so you don't need to close it after all. afterAll can't have any pages, because it runs after all tests.

If you are looking for a serial mode where the page is reused (anti-pattern), you should copy the example from https://playwright.dev/docs/next/test-retries#reuse-single-page-between-tests. Notice that page fixture is not used there.

[pavelfeldman]

Closing as per above, please feel free to open a new issue if this does not cover your use case.