RE: [xsl] document() function and error-handling

Subject: RE: [xsl] document() function and error-handling
From: Wendell Piez <wapiez@xxxxxxxxxxxxxxxx>
Date: Mon, 07 Jan 2008 16:08:06 -0500
Hey Scott,

At 05:50 PM 1/4/2008, you wrote:
Thanks for your reply. Well explained, and good practical points on top
of it. The result document in this case would only be used internally by
our system, so <?oati-xml-result?> would work just fine in this case.

Understood. Of course, any name at all would work as long as (1) it wasn't already taken for something else, and (2) only machines, and never users, had anything to do with it.

My reasoning behind <?xml-result?> is this:
I'm writing the behind-the-scenes valid-URL testing to be as generic as
possible in the proxy script, perhaps even resorting to a try-catch XSLT
run with a stylesheet that simply attempts to access the URL via
document() to see if it fails or not. I might port it to other systems
if it proves generic enough, and <?oati-xml-result?> would include the
name of a company that has nothing to do with the process. (Replace
"oati" with "microsoft" here and it might become more poignant. ~_^)


Granted, <?xml-result?> isn't a W3C-endorsed PI at all, but it does
contain information regarding the XML-validity of an attempted reference
to an external document, and nothing more specific than that. Wouldn't
"xml-" be the most appropriate prefix here?

But as David suggested, you might decide to respect both letter and spirit of the W3C rules and say <?result-xml?> or some such.

If "OATI" doesn't scope properly for the long term, maybe the system itself needs a name.

(For the record, I'm really bad at naming things when it comes to
computer stuff. :/ )

Naming things is always hard. Paying attention and learning from mistakes (both those of others and one's own) is about the only way to improve at it.


P.S. Where exactly would you draw the line when it comes to XSLT issues
vs. XML issues? XSLT is, after all, XML. :)

Actually there are those who would argue with that. You're getting away with this one (assuming the list-owner doesn't bark at me) since we're talking about a system architecture for XSLT, but generally even that's out of scope. Nonetheless the line is still clearer than that between, say, dance and ballet. See the rule above for how to improve at naming things.

Wendell Piez                            mailto:wapiez@xxxxxxxxxxxxxxxx
Mulberry Technologies, Inc.      
17 West Jefferson Street                    Direct Phone: 301/315-9635
Suite 207                                          Phone: 301/315-9631
Rockville, MD  20850                                 Fax: 301/315-8285
  Mulberry Technologies: A Consultancy Specializing in SGML and XML

Current Thread