Subject: Re: [xsl] What is actually a "fragment" ? From: Michael Kay <mike@xxxxxxxxxxxx> Date: Sat, 26 Apr 2014 23:35:06 +0100 |
On 26 Apr 2014, at 20:03, Dimitre Novatchev <dnovatchev@xxxxxxxxx> wrote: > According to the XDM (both 2.0 and 3.0), > > "[Definition: A tree whose root node is not a Document Node is > referred to as a fragment.]" > > So a fragment is a tree. > > However, I have been taught (by the books of Dr. Michael Kay) that an > fragment is a node-set, that by itself may not be a well-formed > document, but wrapping this node-set in a single element parent will > make this a well-formed document. > > There is an obvious contradiction in these two definitions -- in the > former a fragment must be a tree (have a root node), while in the > latter this isn't required. > Interesting. I'm not aware of any normative use of the XDM-defined term anywhere in our specs, so I don't think it's a big issue. But I'm more familiar with the use in the sense of the DOM DocumentFragment object, which is essentially a Document without the constraint of having exactly one element node and no text node children. Either that, or the URI "fragment identifier" which means something quite different. Michael Kay Saxonica
Current Thread |
---|
|
<- Previous | Index | Next -> |
---|---|---|
[xsl] What is actually a "fragment", Dimitre Novatchev | Thread | [xsl] Grouping By Column Heading (b, G. T. Stresen-Reuter |
[xsl] What is actually a "fragment", Dimitre Novatchev | Date | Re: [xsl] dangling attribute creati, David Carlisle |
Month |