Subject: Re: [xsl] Shallow copy and base URI From: David Carlisle <davidc@xxxxxxxxx> Date: Fri, 27 Jan 2006 16:19:12 GMT |
Actually, I guess using a function as a launcher for the identity copy could be better, because I could use @select and avoid create a document node. The document node is created (for xslt 1 compatibility as much as anything else) if you don't use any funk new attributes but if you go (for example) <xsl:variable name="tmp.tree" as="element()"> <tree> then the variable will hold a (parentless) element node rather than a document node. this is useful sometimes although parentless eleemnts are a bit scary, especially if you make two of them <xsl:variable name="tmp.tree" as="element()"> <tree/> <tree/> </xsl:variable> makes a variable that holds a sequence of two tree elements, but they are not siblings of each other they are not even siblings of themselves, if the current node is one of those tree elements ../* will not select anything not even the current element as .. is empty. Of course an alternative to making this extra <tree> element to hold xml:base would be to modify mode="the.mode" so it added xml:base attributes to some or all of the elements it generates. Which is easier depends on what the mode="the.mode" templates look like. David ________________________________________________________________________ This e-mail has been scanned for all viruses by Star. The service is powered by MessageLabs. For more information on a proactive anti-virus service working around the clock, around the globe, visit: http://www.star.net.uk ________________________________________________________________________
Current Thread |
---|
|
<- Previous | Index | Next -> |
---|---|---|
RE: [xsl] Shallow copy and base URI, Michael Kay | Thread | Re: [xsl] Shallow copy and base URI, Florent Georges |
RE: [xsl] Shallow copy and base URI, Michael Kay | Date | Re: [xsl] Shallow copy and base URI, David Carlisle |
Month |