On Jul 22, 2008, at 10:19 PM, Vincent Massol wrote:

>
> On Jul 21, 2008, at 10:51 PM, Vincent Massol wrote:
>
>>
>> On Jul 21, 2008, at 9:53 PM, Sergiu Dumitriu wrote:
>>
>>> Vincent Massol wrote:
>>>> Hi,
>>>>
>>>> I'd like to modify our rendering for section titles and use
>>>> matching H
>>>> tags. For example for "1.1" I'd like to use h2 and not h3 and h1  
>>>> for
>>>> "1".
>>>> The main reason for this need (beyond it being more logical) is in
>>>> the
>>>> new rendering. The XHTML parser I use (the one in wikimodel) will
>>>> automatically transform h2 into a level 2 section title.
>>>>
>>>> In addition I'd like the HTML class element to be heading-X (e.g.
>>>> heading-1, heading-2, etc) instead of heading-1, heading-1-1,
>>>> heading-1-1-1, etc.
>>>>
>>>> I have 2 options here:
>>>> 1) modify the current code to generate the <h> tag and the matching
>>>> heading-X class attribute
>>>> 2) only add new css definitions for heading-X class attributes (and
>>>> thus keep the old heading-1-1-1 ones too)
>>>>
>>>> Option 2) is much simpler since option 1 involves modifying our
>>>> wysiwyg editor code too (and I don't know it at all).
>>>>
>>>
>>> +1 for changing the heading levels.
>>>
>>> -0 for keeping a class element at all. H1 already means heading,  
>>> so I
>>> see no (semantic) need for that class (see also
>>> http://markmail.org/message/5vobm4oxm3c235xd)
>>
>> I agree. The only case I can see would be to differentiate a h1
>> generated by the user in a {xhtml} macro vs a wiki syntax level 1
>> heading but then I don't see why we should differentiate them.
>
> ok I've implemented it locally (51 files modified...) and I'm ready to
> commit...
>
> However I have one remaining problem... Whereas before the WYSIWYG
> editor was only converting HTML H elements having a class of
> heading-1* into section titles, it's now converting any HTML H element
> into a section title.
>
> Thus if you have in your page: <h2>hello</h2>
> You get: 1.1 Hello
> Which in XHTML is rendered as: <h2 id="..."><span>hello</span></h2>
>
> If you have <h2><span>hello</h2>
> You get: 1.1 {style:type=span}hello{style}
> Which in XHTML is rendered as: <h2 id="..."><span><span style=""
>> hello</span></span></h2>
>
> Said differently if you edit in WYSIWYG pages like the Dashboard one
> or the WatchList one and you save it you'll get a different result
> than before (these pages use H elements in HTML).
>
> The problem is also in the new rendering. Imagine you have an input of
> "<h2>hello</h2>".
> You run it throught he XHTML parser and you get a SectionBlock with a
> level 2 section.
> Then if you execute the XHTML Renderer you get: "<h2
> id="..."><span>hello</span></h2>"
>
> So I thought about removing the <span> element but I'm told by Laurent
> Lunati that we need to keep it as it serves as a CSS hook when we need
> to set some styles for IE6 or even with Firefox that cannot be done
> using the H element. I didn't fully understand so if anyone knows why
> we need to keep it in more details please let me know.
>
> So basically we have no more ways of detecting a H element that is a
> wiki syntax heading from one entered by the user (unless we use the
> nested <span> as a differentiator).
>
> So, what do we do? Do we leave it like this with the known limitation?
> Do we drop the <span> element? Something else?

Ok I've modified the regex in the wysiwyg editor so that only H  
elements followed by a span are recognized as headings...

I'm still not sure I like the <span> element and I'd like more details  
on why we need it.

Thanks
-Vincent


_______________________________________________
devs mailing list
[email protected]
http://lists.xwiki.org/mailman/listinfo/devs

Reply via email to