Re: [xsl] Highlight.js available as a XPath extension function

["Norman Tovey-Walsh ndw@xxxxxxxxxx" <xsl-list-service@xxxxxxxxxxxxxxxxxxxxxx>]

Christophe Marchand cmarchand@xxxxxxxxxx
<xsl-list-service@xxxxxxxxxxxxxxxxxxxxxx> writes:
> which is not yet open-sourced - I wanted to use exactly the same
> highlighter...

Yes. I faced exactly the same problem. I had been using PrismJS for
syntax highlighting in the browser, but I needed to be able to produce
PDFs. It was annoying that the PDFs werenbt syntax highlighted.

Ibve tinkered with calling JavaScript from the JVM (I have a
cx:javascript step), but at a glance, it didnbt seem practical to run
Prism in that context (does highlightJS really not use any of the DOM
APIs? Or does GraalsJS now support them?)

I cooked up an XProc pipeline to use Pygments in the PDF toolchain, but
then the PDFs had different syntax highlighting than the browser :-(

For the DocBook xslTNG stylesheets, I just went with Pygments in both
places. And since Ibm (temporarily!) getting by without an XProc
implementation in that context, I coded it up as an extension function.

Ibve also now coded up a proper XInclude implementation as an extension
function and a RELAX NG validator. [Expletive] pipelines are useful and
being without them is damned inconvenient!

> You complain about forking processes, but embedding a GraalJS engine

Right. I had seen your note about GraalJS being kind of heavyweight and
I didnbt want to give the impression that Pygments was going to be
faster!

Ibll try to get your extension working too.

                                        Be seeing you,
                                          norm

--
Norman Tovey-Walsh <ndw@xxxxxxxxxx>
https://nwalsh.com/

> She was mostly immensely relieved to think that virtually everything
> that anybody had ever told her was wrong. (Mrs. E. Kapelsen)--Douglas
> Adams

[demime 1.01d removed an attachment of type application/pgp-signature which had a name of signature.asc]