[Webkit-unassigned] [Bug 17403] WebKit Creates Invalid Xhtml Links with Ajax

bugzilla-daemon at webkit.org bugzilla-daemon at webkit.org
Mon Apr 14 03:50:49 PDT 2008


http://bugs.webkit.org/show_bug.cgi?id=17403





------- Comment #19 from julien.chaffraix at gmail.com  2008-04-14 03:50 PDT -------
> The only problem I can see is that is that createContextualFragment is still
> not contextual - it doesn't use any context from the node it is invoked on, see
> this test case. Even in the original test case, the elements would end up in
> null namespace AFAICT. 

IMHO, I think you are wrong here. Check XMLTokenizer::startElementNs: there is
a check for the namespace. If none is provided by libxml then we check the
prefix against m_prefixToNamespaceMap (which is initialized from the Element
provided as a parent). If no prefix was found, we put the node in the default
namespace.
(LayoutTests/fast/dom/set-innnerHTML also check for namespace and I had a hard
time with it)

> Daniel says there are many things that are contextual,
> besides namespaces - are there any others that we should care about?

I would say that others are internal to libxml (like tree structure or parser
state to detect if the tree is well formed).

I have tried to match libxml behaviour to avoid any regression here (preferring
portable API whenever I could).

>  483     // XML_PARSE_NODICT: default dictionnary option.
> 
> An extra "n" in "dictionary" here.
> 
Oups...

There is also a memory leak as we never free m_context.


-- 
Configure bugmail: http://bugs.webkit.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug, or are watching the assignee.



More information about the webkit-unassigned mailing list