[Webkit-unassigned] [Bug 258282] REGRESSION (STP 172): Broke custom element built-in polyfill by making form properties non deletable

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=258282

Andrea Giammarchi <andrea.giammarchi at gmail.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |andrea.giammarchi at gmail.com

--- Comment #3 from Andrea Giammarchi <andrea.giammarchi at gmail.com> ---
Polyfill author here. I am not sure if it's a regression or an intended behavior but the form gets an own `0` property which I believe is the button within the form (I can't test as I'm on WebKit non TP)

Live example here: https://codepen.io/bakura10/pen/mdQERdX

Before I explain the fix: some class might define accessors and when a node gets upgraded, these accessors won't trigger unless removed and re-attached (to trigger accessors at the prototype level) later on.

The current live demo reveals that the form gets an own, enumerable, property as index `0` and when the upgrade dance kicks in, this can't be deleted or an error is thrown.

I believe the polyfill fix was necessary regardless to avoid this kind of situation in the future, even if this never happened before (builtin don't usually get own enumerable properties out of the blue) but it was surprising to find that index `0` attached, although I am sure I shouldn't delete it and re-attach it, if that's not meant to ever happen.

Hopefully this helps clarifying what you should do or decide if it's a regression or not, yet it would be extremely helpful if you folks could test that polyfill when adding new features as part of your tests ... it's been around for years, there are business depending on it, and beside adding a +1 to the request "please just add builtin extends and let's move forward" the polyfill doesn't change much in time, if ever, as it's rock solid and working very well so it shouldn't add particular burden to your flow.

Thank you in advance for eventually considering my hint.

-- 
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/20230620/341c42da/attachment.htm>