Daniel Aleksandersen wrote:
Hi emailing list, (message contains Unicode characters)
I am curious about the status of the required <div> container when
including xhtml in text constructs. With it be changed to become optional
instead? The below example never made much sense to me.
No. it is required.
<feed … xmlns:xht="http://www.w3.org/1999/xhtml";>
…
<summary type="xhtml">
<xht:div>
<xht:p>This is a <xht:em>summary</xht:em> paragraph.</xht:p>
</xht:div>
</summary>
…
</feed>
Instead I am currently using a ‘hack’ which involves declaring the type as
application/xhtml+xml instead of xhtml. It is not directly invalid,
according to the Atom format specifications. However, it is not very well
supported either. I always contain the XHTML in paragraphs or other XHTML
element; so it stays valid XML all the time.
<feed … xmlns:xht="http://www.w3.org/1999/xhtml";>
…
<summary type="application/xhtml+xml">
<xht:p>This is a <xht:em>summary</xht:em> paragraph.</xht:p>
</summary>
…
</feed>
Don't do this. It's not valid either way. When
type="application/xhtml+xml", the content is expected to be processable
as a complete XHTML document, which obviously is not the case. Hacks
like this are most likely just to lead to interoperability problems later.
Is there a better way of doing this? —with ‘better’, I am ‘better
supported and with a higher adoption rate’. My main concern is that some
feed readers may not be able to support my undocumented
implementation/interpretation. Anyone got any experiences in this field?
The best approach is to stick to the standard and to not try to work
around it.
- James
What about the <article> or <section> elements from the HTML 5 (+XML)
working draft? They would both become much better containers than a
meaningless <div> if it is absolutely required to have one.
