forrest-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Ross Gardler <>
Subject Re: Help with sitemap.xmap - Docbook Plugin
Date Sat, 30 Oct 2004 12:58:43 GMT

Sean Wheller wrote:
> On Saturday 30 October 2004 09:42, David Crossley wrote:
>>Sean Wheller wrote:
>>>Ok Ross,
>>>My map:transform is now like this:
>>>        <map:resource name="transform-to-xhtml">
>>>            <map:act type="sourcetype" src="{src}">
>>>                <map:select type="parameter">
>>>                    <map:parameter name="parameter-selector-test"
>>>                    <map:when test="docbook-v4.3">
>>>                        <map:transform
>>>xsl"/> <map:transform
>>>                    </map:when>
>>>                    <map:otherwise/>
>>>                </map:select>
>>>            </map:act>
>>>        </map:resource>
>>>    </map:resources>
>>>I get a title only in the body.
>>I remember answering this one or two weeks ago.
>>The trouble i saw was that the docbook.xsl wraps
>>the whole content in div elements. I thought of inserting
>>a little stylesheet before html2document, to strip those
>>outer divs. Alas, no time to experiment.
>  Yes you're right. However, Do you really want to strip the "div" elements. 
> Doing that breaks the CSS provided with docbook.

I was doing some experimentation last night and could see no reason to 
strip them. Although I did not do extensive testing to see what impact 
it would have. For your use case you need to maintain the divs so give 
it a go and see what happens (Note there is a matcher in html2document 
that intentionally strips divs, just remove that match and the divs will 
be copied across).

> If we are able to go the route of DB -> XHTML1.0 then I don't think it is 
> worth breaking the CSS. Let me know what is required in order to use XHTML1.0 
> instead of Document and I will start hacking it.

We do not want to use XHTML. We used to use a html2htmlbody.xsl 
stylesheet to do this, but the problems with skinning were too great, 
hence we introduced the xhtml2document.xsl. It works, but is not tested 
in large enough documents to be mature yet. Your work will bring it to 
maturity. Please do not give up on it yet, discuss with us and we will 
find you a solution. (i.e. see paragraph above).

To give you a little encouragement I was using this same technique last 
night to create a WYSIWYG editor. So far I have not found a problem that 
cannot be solved. I *may* find one yet, but with us both working on it 
we have a better chance of success.

The end goal, as you know, is the pipeline:

someformat -> XHTML1 -> XHTML2 -> skinned XHTML

If we work towards that you will be enabling anyone to use any format 
that utilises XHTML1 (yesterday for exmple someone wanted Latex support).


View raw message