RE: XProc and iXML

Hi Norm, Martin and group,

If this has turned into a Saxon or XML Calabash question feel free to send me off. Thanks for the hints so far!

On the Saxon documentation page for XSLT configuration https://www.saxonica.com/documentation12/index.html#!configuration/configuration-file/config-xslt some features are indeed marked as "Applies to HE" - so I guess there is hope.

But apologies for being dense, while I see an "xslt/extensionElement" element for binding extension elements (instructions), there is nothing evident for binding extension functions. (Maybe I need an example - I've also looked at the 'global` element.) And in any case those are PE/EE features and won't work under my 'unencumbered use of standard features' requirement....

Put that way I know this might be a bit of a tall order. Nevertheless I'd be grateful for any more advice or guidance. (This does work fine under Saxon-HE when I can call the initializer.) I'm a little hesitant to ask the Saxon list for a how-do-you-avoid-buying-Saxon question 😊 -- even if my users are arguably next year's Saxon customers....

Thanks, Wendell

-----Original Message-----
From: Norm Tovey-Walsh <ndw@nwalsh.com> 
Sent: Monday, February 5, 2024 9:44 AM
To: Piez, Wendell A. (Fed) <wendell.piez@nist.gov>
Cc: Stein, A.J. Mr. (Fed) <alexander.stein@nist.gov>; xproc-dev@w3.org
Subject: Re: XProc and iXML

> There is one thing I can’t get working yet, however, under XML 
> Calabash using Saxon-HE, namely configure Saxon’s extension 
> functionality to use Norm’s CoffeeSacks XSLT extension. (Saxon-HE is extensible only by means of an initialization called either programmatically or via an -init flag).

I thought there was a flag for that, or an XML Calabash configuration option (I’d have *sworn*…), but I went looking the other day on behalf of someone else and I couldn’t find it.

I’ll try to sort that out in the near future.

In the meantime, I think Martin’s proposal to do it in the *Saxon* configuration file will work. Just put the names of the classes that implement the functions in there:

  org.nineml.coffeesacks.LoadGrammar
  org.nineml.coffeesacks.MakeParser
  org.nineml.coffeesacks.HygieneReportFunction

                                        Be seeing you,
                                          norm

--
Norm Tovey-Walsh <ndw@nwalsh.com>
https://norm.tovey-walsh.com/


> One must not confuse scientific fact with scientific misconception.
> They're more or less opposites, of course. But a scientific 
> misconception, if widely believed, creates a kind of truth of its own.
> This doesn't seem to make sense, but it's true.--Adam Felber

Received on Monday, 5 February 2024 15:32:41 UTC