Added a test for suppressed mouse events on canceled pointerdown

[mustaqahmed]

Closes #3189

[wpt-pr-bot]

Reviewers for this pull request are: @EvgenyAgafonchikov, @Steditor, @bethge, @jacobrossi, @plehegar, @scottgonzalez, and @staktrace.

[hoppipolla-critic-bot]

Critic review: https://critic.hoppipolla.co.uk/r/6644

This is an external review system which you may optionally use for the code review of your pull request.

In order to help critic track your changes, please do not make in-place history rewrites (e.g. via git rebase -i or git commit --amend) when updating this pull request.

[mustaqahmed]

ptal, resolved the sync problem in my branch.

@scottgonzalez any comments?

[RByers]

Looks reasonable to me. Is it worth also adding a "drag" case so we can verify that no mousemove is generated? Obviously it would have to be robust to the number of move events - so maybe append to the log at most once for move events?

[mustaqahmed]

I tried exactly that in my first commit. The problem is that neither Edge nor Chrome fires any mouse events on touch drag, so the no-canceled pointerdown case fails.

Does it make sense to split the test into touch and non-touch case?

[scottgonzalez]

These lines are indented one too many times.

[mustaqahmed]

Done, ptal.

Also split the test into two cases: click/tap (for any pointer type) and drag (only for mouse).

[RByers]

Splitting into two tests makes sense to me. We should be designing our tests for an environment (like Edge, blink and jQuery's harnesses) where they're all automated anyway (since manual tests are of very little value in practice).

There's a lot of boilerplate / duplicated code between theses tests though, but I guess that's a maintenance burden issue for many of the pointer events tests (not necessarily something to be solved in this commit).

[mustaqahmed]

FYI: both tests are passing on Chrome & Edge.

[RByers]

LGTM, thanks