[tei-council] <postscript>

[Martin Holmes]

Odd indeed. You could get round it by including the <lb/> inside one of 
the other components, presumably:

<postscript>
<p>some stuff<lb/></p>
<closer><lb/><signed>MM</signed></closer>
</postscript>

but it's a hack.

What would you be wanting as a solution? Should model.global interleave 
with the others in the main group? Is this actually a case that would be 
a good illustration of the need for interleave?

Cheers,
Martin

On 12-05-06 04:16 PM, Sebastian Rahtz wrote:
> A gentleman of my acquaintance notes that this is illegal:
>
> <postscript>
> <p>some stuff</p>
> <lb/>
> <closer><signed>MM</signed></closer>
> </postscript>
>
> which seems harsh.
>
> Would anyone care to know why we defined the content model of postscript
> as
>
> <zeroOrMore>
>       <choice>
>         <group>
> <zeroOrMore xmlns="http://relaxng.org/ns/structure/1.0">
>    <ref name="model.divTopPart"/>
> </zeroOrMore>
> <oneOrMore xmlns="http://relaxng.org/ns/structure/1.0">
>    <ref name="model.common"/>
> </oneOrMore>
> <zeroOrMore xmlns="http://relaxng.org/ns/structure/1.0">
>    <ref name="model.divBottomPart"/>
> </zeroOrMore>
>         </group>
>         <group>
> <ref name="model.global"/>
>         </group>
>       </choice>
>   </zeroOrMore>
>
> ? i mean, why did we make model.global a choice all its own? the effect is that once you've met the<lb/>,
> you go back to the start of the rule, and so<closer>  fails cos there is nothing from model.common.
>
> I know this is head-bending stuff, just vaguely hoping someone can see a rhyme or reason.
>
> Sebastian
>
>