Page 1 of 1

PHP.NET's RSS Feed and Firefox

Posted: Wed Mar 15, 2006 8:48 pm
by Ambush Commander
Is it just me, or did PHP.NET's RSS feed stop parsing correctly in Firefox 1.5? (actually, I think this has been going on for a while)

Posted: Wed Mar 15, 2006 9:19 pm
by Benjamin
Doesn't work in Firefox 1.0.7 either.

Posted: Wed Mar 15, 2006 10:23 pm
by AKA Panama Jack
Works just fine in Opera. :D

Re: PHP.NET's RSS Feed and Firefox

Posted: Thu Mar 16, 2006 1:06 am
by Roja
Ambush Commander wrote:Is it just me, or did PHP.NET's RSS feed stop parsing correctly in Firefox 1.5? (actually, I think this has been going on for a while)
Its not Firefox. The feed is no longer valid:

http://feedvalidator.org/check.cgi?url= ... 2Fnews.rss

The XML spec encourages authors to be conservative in what they accept, so Firefox is just following the rules.

Contact the PHP group and let them know.

Posted: Sat Mar 18, 2006 11:54 am
by AKA Panama Jack
Actually if you really check out the syntax the feed validator site is misinterpreting the feed.

The RDF:About flag is wrong in this case as it is interpreting the <item rdf:about= ...> improperly in the context it is being used. The other error is another mistake by the feed validator. It is interpreting &mdash; as it it were something like &nbsp; because it is bracketed by an & and ; without spaces. If it was & mdash; then it would pass. It's a programming fault on the validators side.

You will find stand alone RSS viewers along with Opera interpret the feed properly.

Online validators are not 100% perfect. :)

Posted: Sat Mar 18, 2006 12:46 pm
by Roja
AKA Panama Jack wrote:The RDF:About flag is wrong in this case as it is interpreting the <item rdf:about= ...> improperly in the context it is being used. The other error is another mistake by the feed validator. It is interpreting &mdash; as it it were something like &nbsp; because it is bracketed by an & and ; without spaces. If it was & mdash; then it would pass. It's a programming fault on the validators side.
You may be right on it being a fault, but not on the reasoning.

It wouldn't pass if it was & mdash; - that wouldnt be valid, as that is not a valid character entity (Check the XML entity list - there can be no spaces inside). However, it does seem to be flagging the entity incorrectly. I can only guess, based on the output, that it is because there is no *leading* space before the entity begins, and after it ends. That seems dumb. :)

Submit some feedback on the issue on their mailing list to help them improve: http://lists.sourceforge.net/lists/list ... ator-users

Regardless, thats only the second error. The first error (duplicating the about field) is accurate, and should cause an error in any compliant viewer.
AKA Panama Jack wrote:Online validators are not 100% perfect. :)
Neither are browsers including Opera.

Posted: Mon Mar 20, 2006 12:45 am
by AKA Panama Jack
Roja wrote:It wouldn't pass if it was & mdash; - that wouldnt be valid, as that is not a valid character entity (Check the XML entity list - there can be no spaces inside). However, it does seem to be flagging the entity incorrectly. I can only guess, based on the output, that it is because there is no *leading* space before the entity begins, and after it ends. That seems dumb. :)
Actually, they are treating it like an entity and they shouldn't.
Roja wrote:Regardless, thats only the second error. The first error (duplicating the about field) is accurate, and should cause an error in any compliant viewer.
And the second one is NOT accurate because each rdb:about is in a different item and it should NOT be flagged as a duplicate.

Posted: Mon Mar 20, 2006 5:08 am
by Roja
AKA Panama Jack wrote:Actually, they are treating it like an entity and they shouldn't.
Why shouldn't they? (&)mdash (; ) is a defined xml entity.
AKA Panama Jack wrote:And the second one is NOT accurate because each rdb:about is in a different item and it should NOT be flagged as a duplicate.
No, even in a different item, the spec says you cannot have duplicate about's.