Bug 37129 – CSS last-child not being properly updated dynamically after innerHTML update

• Status: RESOLVED
• Resolution: DUPLICATE
• Component: CSS

[Stephen Murphy]

Created attachment 52592 [details]
HTML file showing the innerHTML and last-child bug

When updating a node's content via innerHTML, the last-child CSS attribute isn't re-evaluated properly. I've attached an HTML file showing the issue.

[Stephen Murphy]

This not only affects WebKit nightly but Safari 4.05 (5531.22.7)

[Alexey Proskuryakov]

Confirmed with r57153. I tested that last-child works dynamically when adding several items with appendChild(). Perhaps it's more than just last-child that doesn't work well with innerHTML?

[Stephen Murphy]

One workaround for this is to use nth-child(N) where N is the last item. Mobile Safari is affected as well.

[David Richards]

I guess I duped this bug, but I also included some other examples and workarounds because it also affects styling of a set of classes.

bug: 37944

https://bugs.webkit.org/show_bug.cgi?id=37944

[Yuzo Fujishima]

Created attachment 55829 [details]
Fix for Bug 37129 - CSS last-child not being properly updated dynamically after innerHTML update

[Yuzo Fujishima]

Created attachment 55830 [details]
Fix for Bug 37129 - CSS last-child not being properly updated dynamically after innerHTML update

[Yoshiki Hayashi]

I don't know if both are needed or only one is needed but I happened to have sent a different patch to fix the same bug in different bug.  What a coincidence the time I posted is almost the same time as yours. :)

https://bugs.webkit.org/show_bug.cgi?id=37944

[Yuzo Fujishima]

Ping?

[Alexey Proskuryakov]

I don't know if this is correct or not, but it certainly looks suspicious that a function called beginParsingChildren() is called when not parsing.

[Kent Tamura]

I chose the patch in Bug#37944.  Thank you for working on this.

*** This bug has been marked as a duplicate of bug 37944 ***