Re: About tests 37-41 (headers)

I am afraid you did not understand me. I agree with your example (it is 
perfectly OK), but I did not misunderstand tests 37-41.

Please, take notice that I said that "there should be at least one h1 
somewhere before every h2":

//h2[preceding::h1 = ()] = ()

instead of "the previous header of every h2 should be a h1", which would 
be the following expression (which I also disagree with):

//h2[let $prev := preceding::(h1|h2|h3|h4|h5|h6)[1] return $prev = () or 
node-name($prev) != "h1"] = ()

This is another example of why I consider XQuery expressions useful for 
avoiding misinterpretations. :-)))

The same applies for h3-6. Your sequence validates all my XQuery 
expressions, so it is OK.

Best regards.

On Wed, 15 Jun 2005, Joe Clark wrote:

>
> On Wed, 15 Jun 2005, Vicente Luque Centeno wrote:
>
>> * Those h2 having no preceding h1 are an error:
>
> A common misunderstanding. A sequence like
>
> 	h1 h2 h3 h3 h3 h2 h3 h3 h2 h3 h3 h3
>
> is perfectly imaginable for some documents and is the correct nesting of 
> heading levels for those documents.
>
> You seem to think that the only permissible sequence is h1 h2 h3 h4 h5 h6, 
> with all but h1 occurring in any quantity in a cluster.
>

Vicente Luque Centeno
Dep. Ingeniería Telemática
Universidad Carlos III de Madrid
http://www.it.uc3m.es/vlc