Atom 1.0 comments feed? Re: [wp-hackers] WP issues

[Geoffrey Sneddon]

On 31 May 2007, at 22:34, Lloyd Budd wrote:

> On 5/31/07, Geoffrey Sneddon <foolistbar at googlemail.com> wrote:
>>
>> On 31 May 2007, at 20:52, Lloyd Budd wrote:
>>
>> > On 5/31/07, Geoffrey Sneddon <foolistbar at googlemail.com> wrote:
>> >> 3. Why was no Atom 1.0 feed offered?
>> >
>> > I am guess that should read Atom 1.0 comments feed?
>>
>> I mean any Atom 1.0 feed at all.
>>
>> >
>> >>         a) "Aggregators I tried with the existing patch didn't
>> >> work."[TICKET1526] — We already offer multiple feeds to make sure
>> >> the UA can parse one (though in the above case, all of Photo  
>> Matt's
>> >> comment feeds are currently broken, and any parser that parses  
>> it is
>> >> broken).
>> >>         b) "There is no satisfactory patch  
>> available."[TICKET1526] —
>> >> Waiting doesn't seem to have made this much better than patches  
>> that
>> >> were around several years ago, see 2.
>> >>         c) "We have no way currently to ensure XHTML
>> >> validity."[TICKET1526]
>> >> — See 1.
>> >
>> >> [TICKET1526]: http://trac.wordpress.org/ticket/1526
>> >
>> > That has a lot of information -- and I don't have the expertise to
>> > follow it. Maybe, a new suscinct ticket would move the issue  
>> forward?
>> > Of course, patches and availability to discuss on #wordpress-dev  
>> helps
>> > even more!
>>
>> How would that help this issue? It was _finally_ fixed in 2.2, when
>> Atom 1.0 support was added at the site level (the shipped comments
>> feed is unusable, so it isn't really added, IMO). This is, along with
>> various other points, an issue with WP's development process.
>
> Sorry, I don't understand. As well as development process concerns,
> does 3. describe any current technical issues?

The third comment relates to using the XHTML content construct in  
Atom, which requires the XHTML to be well-formed XML. We have no  
method of ensuring that (the only way to properly do so is to either  
reject anything at an authoring level that isn't well formed, or use  
an XML serialiser). What was eventually done was using the HTML  
content construct, and relying on HTML error handling (so that empty  
tags aren't treated as NETs).

It's also rather odd that Matt is saying that, as we didn't use the  
the XML construct in Atom 0.3, but rather we used the HTML one. Why  
must we use the XHTML one in Atom 1.0, and not continue doing what  
we're already doing?


- Geoffrey Sneddon