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

Re: constraints - odd question

• From: Rick Jelliffe <ricko@a...>
• To: xml-dev@l...
• Date: Mon, 09 Jul 2001 16:54:19 +0800

 From: "Simon St.Laurent" <simonstl@s...>

> No, the problem is that you don't know what the element names are in
> advance.  Any elements are permitted, but there can be no duplication of
> names.  I don't think the solution you proposed deals with that, unless
> I've missed something significant in XML Schema.

Oh, I get it.  OK, then what about a variant of Ken's expression?

<pattern>
    <rule context="properties/*">
        <report test="previous-sibling::*[name(.)=name(current())]"
        >There should not be duplicate properties</report>
    </rule>
</pattern>

I made it previous-sibling because then the error is reported at the
subsequent
occurrence rather than at the first one. A different way to do it might be

<pattern>
    <rule context="properties/*">
        <assert test="count(../*[name(.)=name(current())])=1"
        >There should not be duplicate properties</report>
    </rule>
</pattern>

but you would get a repeated message for the original and the
duplicate and involve twice the comparisons as the former.

Cheers
Rick Jelliffe

• References:
  • constraints - odd question
    • From: "Simon St.Laurent" <simonstl@s...>
  • Re: constraints - odd question
    • From: Rick Jelliffe <ricko@a...>
  • Re: constraints - odd question
    • From: "Simon St.Laurent" <simonstl@s...>

• Prev by Date: Re: XML Blueberry (non-ASCII name characters in Japan)
• Next by Date: Re: CDATA vs. EMPTY
• Previous by thread: Re: constraints - odd question
• Next by thread: FYI - XML & DTD/XML & RDF articles
• Index(es):
  • Date
  • Thread