1345301 - elt.firstChild.remove() doesn't trigger a DOMSubtreeModified event

Status: RESOLVED FIXED

(Core :: DOM: Events, enhancement)

Description

[Florian Quèze [:florian]]

Replacing
  subtree.firstChild.removeChild(subtree.firstChild.firstChild);
with
  subtree.firstChild.firstChild.remove();

at http://searchfox.org/mozilla-central/rev/d4adaabd6d2f88be0d0b151e022c06f6554909da/dom/events/test/test_bug288392.html#61 makes that test fail, as seen in the try push from bug 1345253 comment 2.

Comment 1

[Hsin-Yi Tsai (she/her) [:hsinyi]]

As DOMSubtreeModified is deprecated in DOM level 3, I guess we might not consider to fix this?

Comment 2

[Olli Pettay [:smaug][[e-mail]]]

Maybe not, but mutation events working inconsistently hint that there might be some issue with remove() implementation.

Comment 3

[Olli Pettay [:smaug][[e-mail]]]

Yes, nsINode::Remove() is wrong. It calls parent->RemoveChildAt(uint32_t(index), true);
when it most probably should call 
nsINode::RemoveChild(oldChild, errorResult)

Comment 4

[Olli Pettay [:smaug][[e-mail]]]

Attachment: patch

This makes remove() work consistently with removeChild(), including the a bit weird NodeRemoved dispatching when trying to remove anonymous nodes.
But anonymous nodes aren't exposed to the web so this should be fine - at least consistent.

Comment 5

[:Ms2ger (he/him; ⌚ UTC+1/+2)]

Comment on attachment 8844875 [details] [diff] [review]
patch

Review of attachment 8844875 [details] [diff] [review]:
-----------------------------------------------------------------

This seems fine to me.

Comment 6

[Pulsebot]

Pushed by [e-mail]:
https://hg.mozilla.org/integration/mozilla-inbound/rev/ec3adc50935d
elt.firstChild.remove() doesn't trigger a DOMSubtreeModified event, r=ms2ger

Comment 7

[Wes Kocher (:KWierso) (Not reading bugmail; email directly if needed)]

https://hg.mozilla.org/mozilla-central/rev/ec3adc50935d