RE: Mozilla Support for XSL XSLT? Yes, please!

["Didier PH Martin" <martind@xxxxxxxxxxxxx>]

HI Dan,

Dan said:
Regardless, yeah, I hope to have a fairly solid implementation by May
2000, when I graduate. I'm keeping a site for the progress of the project,
at:

  http://www.cs.brown.edu/people/dr/xml/

There's not much there right now related to the XSLT stuff, since I've
been mostly working on a never-to-be-finished paper, but I'll keep the
list posted when something cool happens.

Didier reply:
Happy to see a fellow C++ programmer (I heard that we are an endangered
specie :-))

So if you want to integrate your XSLT engine in Mozilla then you'll probably
sit the engine on the Mozilla DOM engine, will you?

What kind of algorithm are you using?

In the OpenJade project we came to this algorithm. (by the way, are you
interested to be part of OpenJade tech discussions, there is small bunch of
C++ high flyers there ;-), please let me know if this is interesting to
you).

Algorithm:
a) the original document is parsed and it is transformed into an internal
structure (i.e. a grove) the internal structure is accessed now with a
custom interface, but we are in the process to provide a DOM interface (in
XPCOM) and a C++ Grove interface (we work with others to make it a
standard).
c) Then we parse the style sheet and for each template, we extract the
node-list (i.e. node set) from the document grove. We continue to parse the
template's content and the interpreter tries to match some keywords
pertinent to the style language (in the case of xsl it is any keyword part
of the xsl name space). When we get a match and that the context is ok then
the keyword (which is, in fact, a small procedure) fires an action like,
most of time extract information from the document grove.

this is the easy way to do it. Have you thought of some optimizations that
may improve the process?

Cheers
Didier PH Martin
mailto:martind@xxxxxxxxxxxxx
http://www.netfolder.com


 XSL-List info and archive:  http://www.mulberrytech.com/xsl/xsl-list