Re: Which file should be the basis for pointing into XSLT?

["G. Ken Holman" <gkholman@CraneSoftwrights.com>]

At 00/09/01 16:29 -0400, Tony Graham wrote:
>At 31 Aug 2000 19:54 -0400, G. Ken Holman wrote:
>  > At 00/08/31 15:19 -0400, Tony Graham wrote:
>....
>  > But I was going to try and base the XPointers on IDs if possible 
> instead of
>  > the root ... hopefully this will be more resilient to change.
>
>They will be more resilient to change but not impervious.

Yes, that was why I chose that word.

I gather the work on submitter test description requirements David is doing 
supports the citation of a number of sections of the Recommendations for a 
given test.  What I've done in special renditions of the two 
recommendations is programmed the display on the status line of the 
original XML document XPath address (relative to the closest hierarchical 
unique identifier) of each text node you hover over, as well as encode that 
address as a unique identifier in the HTML rendition.  Using these 
renditions, a submitter of tests can cite the locations in the XML document 
of the recommendation.

Please see the four new links on the work page:

   http://www.oasis-open.org/committees/xslt/work.html

I created two renditions of each Recommendation: one where the status line 
is blanked when you are no longer hovering over a text node and the other 
where the status line persists until you hit another text node.  I found 
the former one better to distinguish where text nodes start and end, but I 
found the latter one better to cite the location when using another window 
on the desktop (since changing window focus doesn't change the status line).

These renditions have also coded named anchors for each text node where the 
name of the anchor is the XPath address of the text node.

I suggest that for submitters to cite a particular location from a 
specification:

  (1) - size your browser window such that the status line is visible when 
focus changes to your editing window
  (2) - load the persistent status line rendition in your browser
  (3) - hover over the text node desired
  (4) - use Alt-TAB to change focus to your editing window
  (5) - transcribe the address as required in the XML of the test description

>However, if, as I
>recall from the Lotus test files, the identifiers are embedded in
>comments in test files,

I recall that David was tasked with prototyping the structuring of the 
submitter's document model ... I think one of the requirements for a given 
test is that a number of citation be associated with the test.

>I was going to ask how we were going to support identifying that span
>of text, but in a rare moment of lucidity and in the interests of
>keeping things simple (or simpler), I'm going to agree with Ken

:{)}

>and
>ask whether we can get by with associating the test with the <p>
>containing the significant text.

Yes, let's discuss this on Wednesday ... though what I've prototyped today 
goes to the numbered text node.

>  > But I've discovered that it wasn't XSLT used to create the XSLT
>  > Recommendation ... it is a DSSSL script using James' SGML 
> semantics.  I've
>  > written the supplied address for Ben Trafford and it bounced as 
> un-sendable.
>
>I don't know where Ben Trafford comes into this,

Ben is cited in http://www.w3.org/XML/1998/06/xmlspec-report-v21.htm as 
being the source of the "XSLT stylesheet intended to provide 100% coverage 
of the official XMLspec DTD".  The address there bounces, so I cannot reach 
Ben.

>but surely a person
>of your skill and experience could rewrite the DSSSL stylesheet as
>XSLT?

Sure, but in the interest of time and consistency with W3C renditions, I 
just obtained James' collection (also sited in the report above) and 
utilized his DSSSL script in my process.

Unfortunately, I then discovered that James' stylesheets must have been for 
his internal use only and not for final publication because there are a 
number of elements not accommodated in this collection of stylesheets that 
are accommodated in the W3C renditions.

However, what I've done so far should be useful.  I will ask James who 
might have the stylesheets used to create the final renditions.

>  > >The generated HTML can be referred to from other parts of the test
>  > >suite's documentation, plus it should be possible to get from a span
>  > >of text in the HTML to the tests that test for the specific behaviour
>  > >defined by that span of text.
>  >
>  > Hmmmmmmmm ... let me think on that.  I may have to make that a separate
>  > document if I cannot combine both sets of requirements.
>
>I thought that was always something that came up in conversation at
>our meetings: that a user could or should be able to browse the text
>of the Recommendations and see which tests apply to a particular run
>of text in the Recommendation.

Browsers appear to accept arbitrary CDATA in the namespace of unique 
identifiers, so I encoded the XML XPath address as a named anchor in the 
renditions.

(in the URL's below, mailers may corrupt and chunk the addresses; be sure 
to recreate the URL's as required before trying to use them)

So, if I were to look at the first of your discretionary items:

  (1) - the name of the item is "signal-non-xslt-attribute-uri" and the 
absolute location to the normative document is:

    http://www.w3.org/TR/1999/REC-xslt-19991116.xml#xpointer(/spec[1]/body[1 
http://www.w3.org/TR/1999/REC-xslt-19991116.xml#xpointer(/spec[1]/body[1]/di 
v1[2]/div2[1]/p[4])

  (2) - find that highlighted text "an XSLT processor is always free" in 
the rendition of the specification found from OASIS:

    http://www.oasis-open.org/committees/xslt/xslt-19991116-addr-persist.html

  (3) - hovering over that particular text one discovers the relative XPath 
address is:

    id(xslt-namespace)/p[4]/text()[2]

(though I suppose you could use: id(xslt-namespace)/p[4]/text()[1] for the 
start of the paragraph)

  (4) - the test can cite the address and the recommendation (the following 
would be encoded as suggested by David in his DTD for submitters to 
describe their tests):

    Recommendation: xslt
    Location: id(xslt-namespace)/p[4]/text()[2]

  (5) - the rendition we create of the test suite can allow the user of the 
test suite to navigate to a local location of a copy of the modified 
rendition of the Recommendation:

    xslt-19991116-addr.html#id(xslt-namespace)/p[4]/text()[2]

  (6) - if need be (and I don't think so because I would like the test 
suite to be run entirely locally without being connected to the Internet) 
one could cite the rendition on the OASIS site:

    http://www.oasis-open.org/committees/xslt/xslt-19991116-addr.html#id(xsl 
http://www.oasis-open.org/committees/xslt/xslt-19991116-addr.html#id(xslt-na 
mespace)/p[4]/text()[2]

(Also, for some reason, using IE5 I'm having problems citing the OASIS copy 
with the named anchor while not experiencing problems citing a local 
Personal Web Server copy or a local disk copy ... I think it has something 
to do with how quickly the file is loaded (I have very slow access at all 
times))

I didn't realize named anchors could use arbitrary CDATA text, so I don't 
think the above is necessarily a violation of accepted practice ... and it 
seems to work.

Let's discuss this more on Wednesday.

................... Ken

--
G. Ken Holman                    mailto:gkholman@CraneSoftwrights.com
Crane Softwrights Ltd.             http://www.CraneSoftwrights.com/s/
Box 266, Kars, Ontario CANADA K0A-2E0   +1(613)489-0999   (Fax:-0995)
Web site: XSL/XML/DSSSL/SGML services, training, libraries, products.
Book: Practical Transformation Using XSLT and XPath ISBN1-894049-05-5
Article:          What is XSLT? http://www.xml.com/pub/2000/08/holman
Next public instructor-led training:     2000-09-19/20,2000-10-03/05,
-                  2000-10-09/10,2000-10-19,2000-11-06/07,2000-11-12,
-                                            2000-12-03/04,2001-01-27