Re: page-sequence-wrapper

On 3 December 2010 15:46, Tony Graham
<Tony.Graham@menteithconsulting.com> wrote:

>>> On Thu, 2010-12-02 at 17:06 +0000, Dave Pawson wrote:
>>>> 6.4.6 fo:page-sequence-wrapper
>>
>> A little judicous choice and I see
>>
>> Possible completions are:
>> azimuth       border-collapse
>
> That is not accurate and not what Paul said.  For starters,
> "allowed-height-scale", "allowed-width-scale", and "auto-restore" are
> all inherited properties that would list before "azimuth".

Where are properties prioritised Tony?
Pauls comment was 'where sensible/appropriate' or similar.
That seems rather more apposite and useful to an author?


>
>> OK Paul, I agree, just about *any*  :-)
>
> Any.

Which is plainly daft and of little use to a stylesheet author?



>> I wonder how many weeks work it would be to produce a first
>> cut Schematron file.
>
> Just allow any property anywhere except on empty FOs other than
> fo:table-column, which gets its applicable properties and all
> inheritable properties.

Which surely is back in the domain of daft again?
there must be some way to apply common sense to
this aspect of xsl-fo.

>
> The challenge for a schema or validator is probably more in handling
> what FO can't contain itself or what FO can't be a descendant of what.

I'm almost sure relax ng could handle that for the Pareto.



>
> The challenge for an authoring environment is presenting the plethora of
> properties in a way that's useful, e.g., directly applicable first, then
> other inheritable, then the rest, but getting the list is all too easy.

It may be for the entire list. I've pulled apart Appendix A
to produce 'a' list.
Picking a sensible list likely to
implemented for an fo seems less easy.

Prioritizing them might be interesting.
Expressing a common sense list per element would be a
sensible goal.

regards

-- 
Dave Pawson
XSLT XSL-FO FAQ.
Docbook FAQ.
http://www.dpawson.co.uk

Received on Friday, 3 December 2010 16:10:00 UTC