test(browser): flaky browser specs (WebKit drag, Firefox import waitForEvent)

Open
#416 1 comment 0 reactions 0 assignees View on GitHub

Maintainers usually reply within 1 day

Nobody has claimed this yet.

Assessment

Difficulty
3/5
Estimated time
1-2 days
Newbie friendliness
75/100
Issue type
Bug
Clarity
Clearly specified
Activity status
Active
Tech stack
javascript, playwright, react
Domain
frontend, testing

Research direction

Start with the drag test near line 256 and the press() helper near line 48 in tests/browser/menu-drag.spec.js. Reproduce with the provided Playwright command using WebKit and repeated runs; inspect other press() callers for one-time reads after 150 ms. Determine whether waiting for the chip result or confirming the drop target before release addresses the failure, then verify the flaky case no longer reproduces.

Written by the indexing model from the issue text.

Description

area:tests

The WebKit test "dragging still works › the Month chip by its handle, after its menu was used, moves to Rows" failed once on #415's CI and passed on the retry. The run was https://github.com/eizedev/pivotkit/actions/runs/43685334063, job "browser (webkit 1/2, React 19)". #415 touches neither the test nor the drag code.

Where: tests/browser/menu-drag.spec.js, in the test at line 256 and the press() helper at line 48.

What failed:

  • The assertion before it passed: the dragged chip followed the pointer, and there was one .pk-overlay after the last move.
  • chips(page, "rows") was still ["City"], not ["City", "Date by month of year"]. The check at line 263 reads the chips once, 150 ms after mouse.up().

The cause is not confirmed yet. There are two candidates:

  1. On a busy runner the drop's state update is drawn more than 150 ms after the release, so the single read comes too early.
  2. press() releases right after its last move. If WebKit fires mouseup before dnd-kit has recorded the Rows zone as the target, the drop lands nowhere.

Reproduce: npx playwright test tests/browser/menu-drag.spec.js -g "by its handle" --project=webkit --repeat-each=30 --retries=0, ideally under CPU load.

Fix:

  • For candidate 1: wait for the result instead of reading it once, with await expect.poll(() => chips(page, "rows")).toEqual(["City", MONTH]). Do the same for cols.
  • For candidate 2: in press(), before the release, wait until the target zone shows that it will take the drop.
  • Check the other callers of press() for the same one-time read after 150 ms.
Dominant language
JavaScript
Stars
1
Forks
1
Avg merge
3h 21m
Merged PRs (30d)
250

Getting set up

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

More from eizedev/pivotkit

All issues in eizedev/pivotkit

Similar issues

More JavaScript issues

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.