>    and my understanding is that
>    certain kinds of character data cannot be stored in
> attribute values,
>    since they are CDATA not PCDATA. Am I mistaken about this?)
>
> Yes you are mistaken. All character data that is allowed in element
> content is allowed in attributes

OK, thanks... I don't know where I picked up this misapprehension
but I'm glad to be freed from it. Still, I wonder why I thought that
and whether there is something true and significant that I'm missing.

> (although newlines are normalised
> differently: to #10 in element content and #20 (usually) in
> attributes.

Umm... I just looked this up in the XML 1.0 spec (3.3.3 Attribute-Value
Normalization) and I don't understand how it is reconciled with what you said...

"1. All line breaks must have been normalized on input to #xA as described in 2.11 End-of-Line
Handling [which refers to element values], so the rest of this algorithm operates on text
normalized in this way."

Not that it matters much to me.

> What doesn't happen is that the string is _Parsed_ looking for element
> start tags, etc. Of course you have to quote " if using " to surround
> teh value, but that is the only difference.

Thanks,
Lars


 XSL-List info and archive:  http://www.mulberrytech.com/xsl/xsl-list