cocoon-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Stefano Mazzocchi <stef...@apache.org>
Subject Re: R:R:pathargs problem
Date Sat, 04 Mar 2000 03:26:05 GMT
Scott_Boag@lotus.com wrote:
> 
> "Timm, Sean" <STimm@mailgo.com> wrote:
> > I don't see any reason that you *can't* specify xsl:version, but it's
> > not required.
> 
> Review the XSLT recommendation quote again:
> 
> "An element from the XSLT namespace may have any attribute not from the
> XSLT namespace, provided that the expanded-name of the attribute has a
> non-null namespace URI."
> 
> Over the past year I've tried to read James Clark-speak.  Since
> xslt:version *is* in the XSLT namespace, then the above would seem to imply
> that it is not legal.  

maybe it's a problem with my english, but the above says "may" not
"must".

"MAY have any attribute NOT from the XSLT namespace"

doesn't say anything about attributes that already ARE on the XSLT
namespace.

> If you invert the first part: "An element from the
> XSLT namespace may *not* have any attribute *in* the XSLT namespace", then
> you get how I am reading it.  

Sorry, but this is totally wrong.

More formally:

 attribute is valid if
   doesn't belong to the XSLT namespace
 AND
   its expanded-name has a non-null namespace URI

the negation is (following the DeMorgan theorem)

 attribute is NOT valid if
   NOT(doesn't belong to the XSLT namespace)
 OR
   NOT(its expanded-name has a non-null namespace URI)

which becomes

 attribute is invalid if
   belongs to the XSLT namespace
 OR
   its expanded-name has a null namespace URI

which clearly breaks the spec. So there is something wrong in your
negation approach.

> This may be a misinterpretation somehow, or
> an over application of legalism.  Hell if I know.  I just want a stylesheet
> that Xalan let's you produce be interoperable with other processors.  So
> clearly this has to be disambiguated by the WG.  Because we all want to
> live in a world where users have choices about what software to run, these
> legalisms are very important when it comes to standards.

oh sure. but in general, I see no problem in allowing the user of the
XSLT namespace for attributes.

-- 
Stefano Mazzocchi      One must still have chaos in oneself to be
                          able to give birth to a dancing star.
<stefano@apache.org>                             Friedrich Nietzsche
--------------------------------------------------------------------
 Come to the first official Apache Software Foundation Conference!  
------------------------- http://ApacheCon.Com ---------------------


Mime
View raw message