[Webkit-unassigned] [Bug 267900] InvalidStateError not thrown in Safari 17 when dialog showModal() called twice

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=267900

sideshowbarker <mike at w3.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Keywords|NeedsReduction              |
                 CC|                            |annevk at annevk.nl

--- Comment #4 from sideshowbarker <mike at w3.org> ---
(In reply to Jeff Johnson from comment #3)
> (In reply to sideshowbarker from comment #2)
> > So it seems that for anybody to be able to investigate this issue further,
> > what would be needed is a minimal reproducible example/reduction that
> > demonstrates Safari not throwing when it should — because the existing test
> > cases in WPT show Safari throwing as expected.
> 
> There was already a minimal reproducible example attached to my bug report:
> https://bug-267900-attachments.webkit.org/attachment.cgi?id=469511

ah OK — sorry for not having noticed that.

So I see that while that example doesn’t cause Safari 17.3 to throw, it also doesn’t cause Chrome or Firefox to throw either. Instead all three browsers log the “This should not be reached” message.

> The difference between our test cases is that yours has an open dialog
> already in the HTML and calls showModal() only once, whereas mine calls
> showModal() twice, hence the title of this bug report.

Since the behavior in Safari, Chrome, and Firefox is the same for your test case, it seems that either the test case doesn’t actually conform to the requirement stated in the spec — or else all three browser engines (not just WebKit) are failing to conform to the requirement.

[I’m not stating one or the other — because I haven’t yet looked at what the relevant requirement is in the spec (which may be different from what that https://developer.mozilla.org/docs/Web/API/HTMLDialogElement/showModal#invalidstateerror MDN page states].

-- 
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.webkit.org/pipermail/webkit-unassigned/attachments/20240130/a0aa56ce/attachment-0001.htm>