[XML-DEV Mailing List Archive Home] [By Thread] [By Date] [Recent Entries] [Reply To This Message]

RE: PSVI and XPath 2.0 data model

• To: <michael.h.kay@n...>
• Subject: RE: PSVI and XPath 2.0 data model
• From: Elliotte Rusty Harold <elharo@m...>
• Date: Wed, 8 May 2002 09:25:25 -0400
• Cc: <public-qt-comments@w...>, <xml-dev@l...>
• In-reply-to: <001e01c1f61b$7cb0b570$6501a8c0@pcukmka>
• References: <001e01c1f61b$7cb0b570$6501a8c0@pcukmka>

At 12:04 AM +0100 5/8/02, Michael Kay wrote:

>It is not acceptable for an XSLT2 processor (or an XSLT1 processor) to
>modify the tree that is provided as its input. It is acceptable for
>arbitrary processing to take place during the construction of that tree.
>This processing is not part of what an XSLT processor does, but it can be
>done by other software that you choose to run before running the XSLT
>processor.

Not good enough. If it's acceptable for "arbitrary processing to take 
place during the construction of that tree" then absolutely anything 
can happen during the construction of the tree. It is a loophole in 
the XSLT2 spec big enough to drive a  $268 billion company from 
Redmond through. If it is impossible to pass a genuine XML document 
in its only normative form (a sequence of characters) to an XSLT 
processor and not expect that the processor refrain from changing the 
meaning of the document during source tree construction by randomly 
replacing and reordering elements, then conformance is meaningless.

>I think the closest we could get to that might be to require that an
>implementor documents a method of constructing the input tree from an XML
>document using the "standard mapping" of XML documents to trees.

That's a start. However, this must be the default behavior. Anything 
beyond that should be at user option only.

>We would also need to consider whether to put in a similar provision about
>serialization

Yes.
-- 

+-----------------------+------------------------+-------------------+
| Elliotte Rusty Harold | elharo@m... | Writer/Programmer |
+-----------------------+------------------------+-------------------+
|          The XML Bible, 2nd Edition (Hungry Minds, 2001)           |
|             http://www.cafeconleche.org/books/bible2/              |
|   http://www.amazon.com/exec/obidos/ISBN=0764547607/cafeaulaitA/   |
+----------------------------------+---------------------------------+
|  Read Cafe au Lait for Java News:  http://www.cafeaulait.org/      |
|  Read Cafe con Leche for XML News: http://www.cafeconleche.org/    |
+----------------------------------+---------------------------------+

• References:
  • RE: PSVI and XPath 2.0 data model
    • From: "Michael Kay" <michael.h.kay@n...>

• Prev by Date: Re: XPath 1.5? (was RE: typing and markup)
• Next by Date: Re: XPath 1.5? (was RE: typing and markup)
• Previous by thread: Re: RE: PSVI and XPath 2.0 data model
• Next by thread: Re: PSVI and XPath 2.0 data model
• Index(es):
  • Date
  • Thread