Re: [opendocument-users] Implementor notes

[Dave Pawson <dave.pawson@gmail.com>]

2009/6/18 Hanssens Bart <Bart.Hanssens@fedict.be>:
>
> Hi Dennis,
>
> I didn't forget about your hard work on the wiki ;-)
>
> The reason I suggested xhtml and atom for the implementation notes
> is that the wiki is read-only for non OIC members. It sure is a great place
> to summarize the issues found and analyzing the spec, absolutely.
>
> But I'm not sure if it still scales if we have to maintain the links to notes
> (perhaps in several languages) and bug tracks of various products, as well
> as general remarks etc. Or maybe that's just me dreaming of a perfect
> world again :-)

perhaps a bit ambitious without lots of prep work

>
> Anyway, regardless of all this, I think we can all agree that having a spec
> published XHTML opens new opportunities :-)


I guess the only id values you could take over from say 1.2 to 1.3 would
be the section numbers, even then with some level of risk, albeit low.

suggestion, retaining that idea and taking such a risk.
Create the xhtml version. Not difficult.
use xt from James clark to generate id values (simplest form used of
all the XSLT processors)
Attach those values, prefixed by the section numbers?

<p>
<xsl:attribute name='id' select='concat("s4",generate-id())'/>

that would give you such as

<p id='s4h2345'>
which is probably as logical as can be expected  without manual intervention?

I did that and finished up with things like
 <xxx  xreflabel="tests.para11p1s1">
      <d:title>Section 11 para 1 s 1. (Form Content)</d:title>

for section 1, para 1 section 11. possible, just a drag to do.


HTH



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