Hakob,

yes, I think we ought to take a similar approach; surely one won't have
to recompile the code of the shared library in order to add an extension
function. I gather that in Xalan, you overload the method used for
extensions when you instantiate the processor-like class. In Sablotron,
it would be more consistent to let you register a callback (an
"extension handler") to handle calls to extension functions. This
remains to be done though. Sorry if my reply was misleading.

Tom 

Hakob Zarobyan wrote:
> 
> Hi Tom,
> 
> Thanks for giving me a hint. BTW, are there any plans to make the
> implantation of extension function similar to xalan
> (http://xml.apache.org/xalan-c/index.html).  In this case there will be  no
> need to modify code base each time you need extension.
> 
> with regards,
> Hakob
> 
> -----Original Message-----
> From: Kaiserovi [mailto:[EMAIL PROTECTED]]
> Sent: September 6, 2000 7:30 AM
> To: Sablotron Mailing List
> Subject: Re: [Sab] extension functions support
> 
> Hakob,
> 
> no extension functions are currently available; however, it shouldn't be
> that hard to add some. It will amount to writing the code in C++, same
> as when adding the currently unimplemented XSLT functions. Look at
> Expression::callFunc() in engine/expr.cpp to get an idea.
> 
> Tom
> 
> Hakob Zarobyan wrote:
> >
> >         Hi All,
> > As far as i understand from previous discussions in the mailing list, this
> > implementation supports extension function
> > (http://www.w3.org/TR/1999/REC-xslt-19991116,     14.2 Extension
> Functions).
> > However, i did not find any information/instruction which provide me with
> > the tips, how to add new extension functions. I will appreciate, if you
> > feedback me on this point.
> > Thanks.
> > regards,
> > Hakob



Reply via email to