& XSLT On Jul 14, 2006, at 6:51 AM, Thomas, Mark - BLS CTR wrote:

>  You can do much better than that! All you need is one loop.
> 
> 
> > my @DataLists = $xp->findnodes("/CompanyData/Data/Datalist);
> >
> 
>  This is a typical newbie mistake. You've stopped too early. You're not
>  really doing anything with the DataList element itself, right? It's  
>  the
>  children you're interested in. Go straight to them!
> 
>  What you really want are the DataListElementX nodes, or put another  
>  way,
>  the children of DataList following the Id. So that's what we'll loop
>  over:
> 
>  foreach my $elem ($x->findnodes('//DataList/*[preceding- 
>  sibling::Id]')){
> 
>    # Get the Id which is the key to the hash
>    my $Id = $elem->findvalue('preceding-sibling::Id');
> 
>    # Now push it on to the hash as an array element
>    push @{$DataList{$Id}}, $elem->string_value;
>  }
> 

I'm a bit lost on the [preceding-sibling::Id] notation.  The w3 site  
defines "preceding-sibling" as "selects all siblings before the  
current node", what gets (or more specifically how) selected as the  
"current node" with the notation that you've used?


>  That's all you need to do. This is the result:
> 
>  %DataList = (
>    'first' => [' first_element_1 ',' ... ',' ... '],
>    'second' => [' second_element_1 ',' ... ',' ... '],
>    'third' => [' ... ',' ... ',' ... ']
>  );
> 

Yep, looks to work well - my structures are in reality a bit more  
complex, but this is a great starting point.  Thanks for your help!

-lev


_______________________________________________
Perl-XML mailing list
Perl-XML@[...].com
To unsubscribe: http://listserv.ActiveState.com/mailman/mysubs