Re: Parsing of HTML fragments

> A lot of this is going to depend on the application.  Guess the question is,
> to pick a specific case - should the Atom spec disallow too-new HTML tags?
> Should Atom disallow presentation-related tags (like <font>) that are
> unlikely to play well in the parent document?

Atom Text Constructs are allowed to contain [in the @type="html" case]
any content "suitable for handling as HTML" (referencing HTML 4.01),
and in the @type="xhtml" case, any content "suitable for handling as
XHTML" (referencing  Modularization of XHTML). [1] I think a
reasonable notion of suitability would allow for the use of something
like the WHATWG parsing algorithm -- effectively, this is more or less
what liberal feed parsers are doing already.


-- 
Edward O'Connor

1. http://atompub.org/rfc4287.html#rfc.section.3.1.1

Received on Wednesday, 25 April 2007 00:10:13 UTC