--- "A. Pagaltzis" <[EMAIL PROTECTED]> wrote: > * John Napiorkowski <[EMAIL PROTECTED]> [2006-09-25 > 19:00]: > > Perhaps I don't fully understand how this works, > but my current > > understanding is that ATOM allows you to specify > xml inside the > > content tags as long as you set a type attribute > of something > > with 'xml' in it, like "application/xml" for > example. > > The rules for XML MIME types are more restrictive > than that, but > otherwise this is correct. > > > Now the atom spec says I can do XML content and > have it look > > like (same example as above): > > > > <content type="application/xml"> > > <member xmlns="http://mysite.com/ns#member"; > > > <name>blah blah</name> > > <title>nobody</title> > > </member> > > </content> > > Correct. > > > However nothing I do seems to get the XML::Atom > plugin to work > > this way. Now, I realize this is not a tradional > use for this > > kind of feed, but the specification seems to > indicate it's a > > valid one. > > Itâs not just valid, itâs explicitly catered > for. This was one of > the use cases that was considered when âweâ > first drafted Atom > (quote marks because I was not on the WG at the > time) and > remained a central concern throughout the process. > Atom is not > just for weblog delivery; think of it more as a > webservice > messaging wire format. It can do a lot of things > besides weblogs. > > Check this out: > http://www.codezoo.com/about/doap_over_atom.csp > > > Another approach I could take is how Google is > doing this by > > adding a new namespace and placing the data > elements inside the > > entry tag and not the content tag. Maybe this is a > better way? > > This would look (very roughly like) > > > > <entry member:xmlsn="http://mysite.com/ns#member";> > > <member:name>blah blah</name> > > <member:title>nobody</title> > > </entry> > > Please donât. Atom is not RSS2. Most of the > interesting things > one might want to do can be done with the provisions > of the core > format, without any need for lots of underspecified > and badly > interacting extensions that contribute to the mess > in RSS2-land, > each requiring a special-case implementation. > > See also http://www.snellspace.com/wp/?p=314 for a > critique of > GData. > > > It just seems neater to me to place all my stuff > inside the > > content tag, > > It *is* neater. It is the intended purpose of > `content`. > > PS.: if you are putting non-human-readable stuff > there, then see > if you can provide a human-readable (possibly short) > description > in `summary`. (Think of `summary` as being to > `content` what > HTMLâs `img`âs `alt` attribute is to `src`.) > > Regards, > -- > Aristotle Pagaltzis // <http://plasmasturm.org/>
Thanks for the links, these are all great examples for me to show my group as examples of exactly what we are trying to do. After looking at the XML::Atom code I really don't think I'm going to get this behavior without patches. That's an option for me since I have developer time paid for already in this project. As a temporary workaround in development the Atom spec says I can use a 'src' attribute instead of inline content. In some ways I like that better, makes the entry cleaner. However I'm not see too many people do that. Anyone know if there is a good reason to stay away from <content src="...' /> Thanks for all the suggestions, appreciated, John __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com _______________________________________________ List: [email protected] Listinfo: http://lists.rawmode.org/mailman/listinfo/catalyst Searchable archive: http://www.mail-archive.com/[email protected]/ Dev site: http://dev.catalyst.perl.org/
