Re: get the current xml file name

The scenario I was interested in is where you follow a link to an XML file
that contains a stylesheet-pi and the transform performed in the browser
depends on the URL so that you can have more than one view of the same
content. The URL of the XML is well defined here, but as far as I can tell,
inaccessible to the transform the browser is running.

I think (and I may get flamed for this) that what you describe was the
intention, but it never really worked in practice.  In reality you
would more likely follow a link to an HTML page that contains
javascript that performs the transform.

Take that a stage further and instead of following the link to an HTML
page which runs the transform on the client, an app server would
intercept the call and run the transform serverside.

The reason for all this is the inflexability of running transforms on
the client, which is what you're finding now.