Re: Type Attibute for LI element.

On Tue, 3 Feb 1998, Jukka Korpela wrote:
> On Mon, 2 Feb 1998, Russell Steven Shawn O'Connor wrote:
> > <UL>
> > <LI Type="A">foo
> > </UL>
> > 
> > This snippet of HTML I would expect to be illegal.  I can't seem to find
> > any indication in the specs saying it is. 
>
> The specs say that in an LI element which is an immediate constituent
> of the content of a UL element, the only allowed values for the TYPE
> attribute are "disc", "square", "circle" (case insensitively, and the
> quotes being optional).
> See http://www.w3.org/TR/REC-html32.html#ul
> and http://www.w3.org/TR/REC-html40/struct/lists.html#type-values
> (to which the description of the TYPE attribute for LI refers, too).

I still don't see it in the specs, but maybe I'm just blind.  The specs
refert to the type parameter in UL and OL, but doesn't seem to talk about
the LI elements that it contain.
 
> There are technical difficulties in imposing this restriction
> in a DTD. As far as the DTD is concerned, the value of a TYPE
> attribute of an LI element is just a string (CDATA). Strictly
> speaking this means that a validator proper must not flag your
> snippet as illegal, since it formally conforms to the DTD.

Perhaps we have different definitions for validator, but I don't think
that every documents that validates as as SGML document against the DTD is
a legal HTML document.  It is possible to validate against the DTD and
stil be an ilegal HTML documents.  That's why I am working on this
porgram.  To catch most (all) other errors that can be in an HTML
document.

-- 
Russell O'Connor                           roconnor@uwaterloo.ca
    <URL:http://www.undergrad.math.uwaterloo.ca/%7Eroconnor/>
"And truth irreversibly destroys the meaning of its own message"
-- Anindita Dutta, "The Paradox of Truth, the Truth of Entropy"

Received on Tuesday, 3 February 1998 10:53:11 UTC