Re: Some XSL Questions

Subject: Re: Some XSL Questions
From: Norman Walsh <norm@xxxxxxxxxxxxx>
Date: Mon, 29 Dec 1997 09:30:06 -0500
Sean Mc Grath <digitome@xxxxxx> writes:

> Threee questions on XSL that I would appreciate some help on:-

It is important to remember that XSL does not yet exist. It is
only a submission to the W3C at this point. Since the
submission, informal discussions have revealed a number of major
areas (never mind the details ;-) where the final spec will
almost certainly differ from the submission.

> 1) How many tree walks?
> At the end of section 5 there is an example of using a named mode to achieve
> a table of contents built from TITLE elements followed by the content
> proper. Does the invocation of a named mode initiate a new walk of the
> document tree? Every TITLE element will need to be traversed twice, right?

A named mode does not initiate a new walk of the tree, but
<select-elements> does.

> 2) Inline Style
> Section 4.2 says that a document instance may contain XSL rules "at the
> beginning". Where
> is that? Under the root element? Presumably XSL will then need either
> namespace or
> XLL style #FIXED attributes to allow documents to use element type names
> other than
> the defaults.

The submission is incomplete in this regard an no further
substantive discussion of this issue has taken place (that I'm
aware of).

Personally, I encourage you to think in terms of providing a
wrapper (perhaps a multi-part MIME document?) that includes both
the stylesheet and the document instance.  In the long run,
encoding bits of stylesheet or even pointers to a stylesheet in
a document instance is "a bad thing", IMHO.

> 3) Making Stylesheets valid w.r.t. a fixed XSL dtd:
> Section 3.3 illustrates how the use of the invoke element for macros. e.g.
> <define-macro name = "foo">
> ...
> <invoke macro="foo">
> Rather than:
> <define-macro name = "foo">
> ...
> <foo>
> It says that doing so can make style-sheets valid XML w.r.t. an XSL DTD. But
> the DTD
> in appendix C "for exegesis only" uses exclusion:

That's true. I'm confident that the final spec will include a
proper XML DTD which can be used to validate stylesheets that
avoid the use of macros as element type names.

Norman Walsh <nwalsh@xxxxxxxxxxxxx> | "Bother", said Pooh, as he
Senior Application Analyst          | deleted his root directory.
ArborText, Inc. ( | 
413.549.3868 Voice/FAX              | 

 DSSSList info and archive:

Current Thread
  • Some XSL Questions
    • Sean Mc Grath - from mail1.ability.netby (8.8.5/8.6.12) with ESMTP id LAA11732Sun, 28 Dec 1997 11:42:51 -0500 (EST)
      • Norman Walsh - from mail1.ability.netby (8.8.5/8.6.12) with ESMTP id JAA23164Mon, 29 Dec 1997 09:32:24 -0500 (EST) <=
      • DSSSList Owner - from mail1.ability.netby (8.8.5/8.6.12) with ESMTP id OAA25275Mon, 29 Dec 1997 14:45:55 -0500 (EST)
        • Norman Walsh - from mail1.ability.netby (8.8.5/8.6.12) with ESMTP id PAA25558Mon, 29 Dec 1997 15:38:26 -0500 (EST)
          • DSSSList Owner - from mail1.ability.netby (8.8.5/8.6.12) with ESMTP id WAA28000Mon, 29 Dec 1997 22:53:05 -0500 (EST)
      • <Possible follow-ups>
      • Sean Mc Grath - from mail1.ability.netby (8.8.5/8.6.12) with ESMTP id GAA05090Tue, 30 Dec 1997 06:00:48 -0500 (EST)