Re: Keys and select distinct - is that the solution ?

Christian Rasmussen <christiankrasmussen@xxxxxxxxx> writes:

> I still don't see why it has to do with grouping. you should be able
> to pick out elements by their content, and check to see if that same
> content is present in other siblings? Is it really necessary to do
> grouping to do just that?

I had completely forgotten about the "check for identical preceeding
elements" method, in fact.  yes, you're right that you could use that
approach.  but as David points out, this becomes unworkable with
moderately large documents.  better, IMO, is to start off with
something likely to work reasonably well in all cases.

> All the examples I've found - including the lesson at jenitenissons
> website, only concentrate on the element names, not the content.

you'll need to generalize a bit.  there's nothing fundamentally
different between element names and content, in this context.

-- 

joe