[osis-core] <div> model question???

[Patrick Durusau]

Guys,

I am slowing but surely working my way backwards (perhaps I should be 
going forwards?) ;-) through the OSIS posts.

I think I may have been misunderstanding Harry's (and Chris's) posts on 
<div> so I will try again (not that I am convinced by the tail recursion 
consideration but if we can have that and a useful content model, so 
much the better).

I understand the model to allow <div> to nest within <div> to any level, 
but that the nested <div> would always occur after any other content of 
the <div>.

Hmmm, the more I think about actual texts that I have encoded (mainly 
periodicals and some monographs), the less likely I think you are to 
encounter what I suggested in my earlier post.

Counter cases anyone?

Assuming I don't  have any by the time I post later today, will be using 
Harry's suggested:

><xs:element name="div">
> <xs:complexType mixed="true">
>  <xs:sequence>
>   <xs:choice minOccurs="0" maxOccurs="unbounded"> [lots of stuff here]
></xs:choice>
>   <xs:element ref="div" minOccurs="0" maxOccurs="unbounded/>
>  </xs:sequence>
>
with the change that I will be dropping <div> from the choice group.

Sorry if I was being thick on this one. Something still nags me about it 
but that is not a good reason to not do it this way.

Patrick


-- 
Patrick Durusau
Director of Research and Development
Society of Biblical Literature
pdurusau@emory.edu