Re: [xsl] Fwd: Combing two different documents

["andrew welch" <andrew.j.welch@xxxxxxxxx>]

On 7/17/06, andrew welch <andrew.j.welch@xxxxxxxxx> wrote:
> On 7/17/06, David B <daavidb@xxxxxxxxx> wrote:
> > Hello,
> >
> > I am trying to combine to seperate XML documents that have the same
> > structure but different leaf nodes, e.g:
> > ---File 1.xml---
> > <base>
> >      <foo>
> >           <bar1>123</bar1>
> >      </foo>
> > </base>
> > --File 2.xml----
> > <base>
> >      <foo>
> >          <bar2>abc</bar2>
> >      </foo>
> > </base>
> >
> > I want the output to be:
> > <base>
> >      <foo>
> >          <bar1>123</bar1>
> >          <bar2>abc</bar2>
> >      </foo>
> > </base>
> >
> > Preferably without hardcoding too much of the structure of the file
> > into the .xsl.
>
> How about something like:
>
> <xsl:template match="@*|node()">
>         <xsl:copy>
>                 <xsl:apply-templates select="@*|node()"/>
>         </xsl:copy>
> </xsl:template>
>
> <xsl:template match="*[not(*)]">
>         <xsl:copy-of select="."/>
>         <xsl:copy-of select="$file2//*[not(*)][count(preceding::*) =
> count(current()/preceding::*)]"/>
> </xsl:template>
>
> You would apply this transform to file1.xml and reference file2.xml
> using the document() function.
>
> You could speed it up by using a key for the lookup into $file2 as the
> heavy using of the preceding axis would likely make this very slow.

If there are a different numbers of leaf nodes at the same point in
each file then you would need work on the level above, eg:

<xsl:template match="*[*[not(*)]]">
	<xsl:copy>
		<xsl:copy-of select="*"/>
		<xsl:copy-of select="$file2//*[*[not(*)]][count(preceding::*) =
count(current()/preceding::*)]/*"/>
	</xsl:copy>
</xsl:template>

If there is mixed content at the level above (<foo> level) use copy-of
select="node()", but if all your text nodes are contained by leaf
nodes then this should be fine.

cheers
andrew