Bug 63538 – Range.insertNode doesn't work per DOM 2 Range

• Component: DOM
• Status: RESOLVED CONFIGURATION CHANGED

[Olli Pettay (:smaug)]

Per DOM 2 Range, and still per the new draft version of Range spec
insertNode should not have special handling when comparing to normal
DOM mutations.
If range is collapsed and insertNode is called,
the range object should not be modified.

ACID3 accepts both this invalid behavior and the behavior defined
in DOM 2 Range.

[Olli Pettay (:smaug)]

Created attachment 98924 [details]
testcase

[Alexey Proskuryakov]

According to discussion in <https://bugzilla.mozilla.org/show_bug.cgi?id=433662>, WebKit behavior matches Hixie's reading of the spec.

See also: bug 16867.

[Glen Huang]

The spec has changed https://dom.spec.whatwg.org/#concept-range-insert

10. If start and end are the same, set end to (parent, new offset).

And lastest gecko and blink all follow through.

I think webkit should do the same.

[Lucas Forschler]

Mass moving XML DOM bugs to the "DOM" Component.

[Ahmad Saleem]

I am unable to reproduce this bug in Safari 15.6 on macOS 12.6 to get expected results. All browsers show two dialog matching with each other, where first dialog is 0,1 and then next is 0,0. It is same with Chrome Canary 106 and Firefox Nightly 105. It might be due to change in web-spec or other browsers aligned with Webkit or DOM3 superseded DOM2 behavior but since all browsers are same, I think we can mark it as "RESOLVED CONFIGURATION CHANGED". Thanks!