Re: [w3c/DOM-Parsing] XML null namespaces need to be preserved (#47)

> 1. Is this a bug in the specification text? If so, where?

IMO it's hard to say this is a specification bug.  The current specification is reasonable, and Edge follows it.  However, now three major browsers except for Edge are against the specification.

If we change the specification so that it matches to non-Edge, we should update 3.5.2.1. 
the second bullet in https://w3c.github.io/DOM-Parsing/#serializing-an-element-s-attributes.

>  "the attr's prefix is null and the ignore namespace definition attribute flag is true"

to something like "the attr's prefix is null, the ignore namespace definition attribute flag is true, and the attr's value does not match to the element's namespaceURI"

> 3. Is this tested by wpt? web-platform-tests/wpt#15779 suggests yes, but maybe there's a nuanced scenario that's not.

[domparsing/XMLSerializer-serializeToString.html](https://wpt.fyi/results/domparsing/XMLSerializer-serializeToString.html) contains some test cases related to this issue. They follow the current specification.
Now "Check if start tag serialization drops element prefix if the namespace is same as inherited default namespace."  fails on Firefox and Safari.  I reverted the Chrome's behavior in this week.  As you know web-platform-tests/wpt#15779 moved the test cases.


-- 
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/w3c/DOM-Parsing/issues/47#issuecomment-473249391