Re: Resources for AtomPub parser validation

[Tim Bray <Tim.Bray@xxxxxxx>]

On Jun 30, 2008, at 9:10 AM, Daniel Jalkut wrote:

> In  particular, I continue to have mind-bending problems  
> interpreting the ins and outs of how particular content needs to be  
> treated in the Atom format that is used by AtomPub.  Right now, for  
> instance, I'm trying to learn definitively whether an escaped  
> "&" inside a content type xhtml div should be left alone or  
> converted to an ampersand in the parsed content output.

If it says <content type="xhtml"> then you just let the XML parser  
keep running and take what it gives you.  In the example from RFC4287  
(title not content, but shouldn't make a difference):

<title type="xhtml" xmlns:xhtml="http://www.w3.org/1999/xhtml";>
  <xhtml:div>
    Less: <xhtml:em> &lt; </xhtml:em>
  </xhtml:div>
</title>

The XML parser will tell you that the content of xhtml:em is " < ".   
Obviously, if you're going to spit that out again to html or anything  
xml-based, you're going to have to re-encode it.

I'm a little concerned that this is bothering people who I know are  
smart, experienced implementors.  Maybe you've stumbled on a corner  
case?

 -T


> It would be a great aid if there was some "reference" feed that  
> exercised many of the common mistakes parsers make, and at the same  
> time expressed definitive advice for how these entities should be  
> handled.
>
> Anybody know of such a resource?
>
> Daniel