[webkit-dev] XPath Issues?

[Alex Milowski]

On Mon, Nov 8, 2010 at 7:09 PM, Frans Englich <fenglich at fastmail.fm> wrote:
>
> As far as I know, no, but I'm busy with studies these days. I guess one of the first questions is rather how the two stacks, Qt and Webkit, should relate to each other on this particular field, what the aim of the integration is and the nature of the integration, as well as dependencies. But it seems interesting to have these technologies in Webkit. What's your thoughts on how it would be carried be done?
>

I think the simplest answer is for one to be able to "turn on Qt XML
support" without having to use the whole Qt port.  In my research into
how all this is put together, the XPath and XSLT support within the
browser is separable.  Any library could be integrated by simply
adding a build option to do so.

I've been looking at XQuilla in this same way.  They have an XPath
2.0, XQuery, and XSLT 2.0 implementation that looks to be very easy to
integrate.  The only is that they rely on some internals from Xerces-C
that seem rather superfluous.  That is, you don't have to use the
Xerces-C parser to use their system.

Any implementation of XSLT or XPath needs to be able to use the XML
DOM directly in some way.  This would avoid any kind of strange
serialization (which is what happens when XSLT is called from
Javascript with libxslt) that currently causes bugs.  In addition, the
XPath implementation can't really serialize and expect to get the
context node correct.

> (I wrote the XPath, XQuery and XSL-T code and mentored the Schema code. However, I'm no longer employed by Nokia.)
>

That's good to know!

Have the compliance tests for XSLT 2.0 and XQuery been run?

-- 
--Alex Milowski
"The excellence of grammar as a guide is proportional to the paucity of the
inflexions, i.e. to the degree of analysis effected by the language
considered."

Bertrand Russell in a footnote of Principles of Mathematics