Re: [patch][ox-html] Support for level based containers

[Rick Frankel <rick@rickster.com>]

On 2014-03-17 23:31, Rasmus wrote:
> It's a variable that you can set in your project or in your Org file
> or in your init file.  I don't see why div × 3 is better than section
> article div or something else conditional on two variables being
> explicitly set to get fancy HTML5. . .  In any case, I don't have
> strong—if any—preferences on this.

Because using these tags is assigning semantic meaning which may or
may not be valid for the current document. Based on the spec, your use
of =section= seems ok (but could also be used for the other levels),
but your use of =article= is probably wrong in most cases. From
http://www.w3.org/html/wg/drafts/html/master/sections.html#the-article-element:

The article element represents a complete, or self-contained,
composition in a document, page, application, or site and that is,
in principle, independently distributable or reusable, e.g. in
syndication. This could be a forum post, a magazine or newspaper
article, a blog entry, a user-submitted comment, an interactive
widget or gadget, or any other independent item of content.

As to the =section= element, the from the above doc:

The section element represents a generic section of a document or
application. A section, in this context, is a thematic grouping of
content. The theme of each section should be identified, typically
by including a heading (h1-h6 element) as a child of the section
element.

and

A general rule is that the section element is appropriate only if
the element's contents would be listed explicitly in the
document's outline.

So, using this definition, in html5, the wrappers should be =sections=
to the same level as the toc heading level specified for the document,
and =divs= after.[1]

> org-html-text-markup-alist is nice.  What do you want to see in
> addition to the current structure (in patch v2)?
> 
> Somehow I never saw the original thread, only the email cc'ing me
> directly. I went to gmane to find the patch, and obviously grabbed the
> wrong one.
> 
> Could you please send me the (new) patch so that i can review it?
> 
> Here's the Gmane link.  I believe it's different than what you
> reviewed before, but perhaps I'm wrong. . .

No, i got the wrong patch from gmane. This one looks better modulo:

1. The default should stay the same as it is now -- the string "div"
2. Minor typo, but "backward comparability" should be "backwards
compatibility".

But, after reviewing the spec (see above vis. =section= and
=article=), i would submit that a better patch would be to
implement [1] above -- remove the defcustom (i only added to support
using a different default wrapper element in html5), and use =section=
and =div= based on toc level when html5-fancy is true. As far as i can
tell from the spec, =article= would almost never be correct for the
average org doc. Here's a relevant quote from the spec:

Authors are encouraged to use the article element instead of the
section element when it would make sense to syndicate the contents
of the element.

I think the best way to implement this would be letting the user
specify it with the =HTML_CONTAINER= property already implemented. As
this seems very much in keeping with the spec, i will implement this
change when i have some time in the next couple of weeks if i don't
hear any strong arguments against.

As an aside, the complex semantics of the new html5 tags is why we
have been slow in implementing them in ox-html. =div= is by
definition a non-semantic tag meant to be used for grouping and
styling, but the new tags have very specific meanings associated with
them and their mis-use is worse than their non-use.

rick