Thanks....That was exactly what I was looking for.

'A' wont be the document elements, and will be be a
normal kind of element. And it wont have nested
descendants :).

Jehanzeb

--- Colin Paul Adams <colin@xxxxxxxxxxxxxxxxxx> wrote:
> >>>>> "Jehanzeb" == Jehanzeb Hameed
> <jehanzebhameed@xxxxxxxxx> writes:
> 
>     Jehanzeb> Hello I am trying to use xslt to copy
> a modified version
>     Jehanzeb> of a xml file. Given a node A, I want
> to generate a new
>     Jehanzeb> XML file, which is exactly the same as
> old XML file, but
> 
> OK. So it's a variation on the identity transform.
> 
>     Jehanzeb> with node A, now also the sibling of
> its parent.
> 
> So in addition to the standard identity transform,
> you want a template
> that matches all nodes that have an A child.
> 
> Note that if A is the document element, then the
> result will not be a
> well-formed XML document. 
> 
> e.g. the tree:
> 
>            /
>            |
>            A
>           / \
>           B  C
> 
> Cannot be transformed into anything sensible - the
> document node cannot
> have a sibling. I presume you want to limit A to
> being a child of an
> element. 
> 
> In which case you want a template something like:
> 
> <xsl:template match="*[child::A]">
>   <xsl:copy-of select="."/>
>   <xsl:copy-of select="A"/>
> </xsl:template>
> 
> Is A also to be only an element, or can it be some
> other kind of node?
> 
> If you can have A's nested as descendants of A's,
> then it will be more complicated.
> -- 
> Colin Paul Adams
> Preston Lancashire
> 
> 

__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com