162441 - Content Model for this testcase isnot correct

Status: RESOLVED INVALID

(Core :: DOM: HTML Parser, defect)

Description

[Jerry]

Testcase: <p/>abc

when access using DOM inspector, it seems like '<p>abc'.
I think mozilla should treat it as '<p></p>abc'

Comment 1

[Jerry]

I think it is a parser's bug.
Fix on htmlparser can fix this bug.

so set component as parser.

Comment 2

[Boris Zbarsky [:bzbarsky]]

Is this in HTML or in XML?

Comment 3

[Jerry]

Testcase is html

Comment 4

[Boris Zbarsky [:bzbarsky]]

You may want to post the testcase in such cases, you know.... ;)

For HTML, we should not be closing out the <p> when we see <p/>.  We used to do
this, and it breaks many sites.  If a non-XHTML document is doing this, there is
absolutely no reason to parse it per the XML rules (and in fact, we could make a
case for parsing per the SGML rules then, so <p/> would be the same as <p>&gt;).
 If it's XHTML1, it being served as text/html is a violation of the XHTML1
specification, which says that only XHTML1 documents that follow Appendix C of
the specification should be served as text/html.  See section C.3 of Appendix C,
which covers this case.