Potential feed validation issue

[yellowled]

Completely not my expertise, but someone pointed out to me on Twitter that s9y might not be generating valid RSS and/or Atom feeds.

Example: http://feedvalidator.org/check.cgi?url= ... s/atom.xml

Note: Might be fixed already, said blog is running on 1.5.5.

YL

[yellowled]

More info: According to the owner of the blog, this is supposed to be related to the flattr plugin …?

The person who pointed this out to me also said to ask if/why (again, not my expertise) XML is (not) being serialized or something …

YL

[garvinhicking]

Hi!

The validity of a RSS file often relies on the author; if they enter invalid XHTML, this gets passed along to the feed. There is the XHTML fixup plugin, that would fix many of the issues; in case of zeitspuk, the <a> tag contains an "&" instead of "&".

The flattr plugin AFAIK doesn't emit this "var ..." stuff anymore, maybe it's not the most up to date version of the plugin?

s9y passes along the actual input of the blog as the simple content. It's really, really hard to fix up bad XHTML from the user. There is absolutely no way to make this bulletproof, once a user i.e. wraps "<b>...</strong>" wrongly, how would we be able to fix this.

XML is simple a hard format, it's not meant to be loosely validated; that's why the user has his part in this...either by not writing HTML manually and using "simple" markup, or by using WYSIWYG editors....

The flattr issue of course I'd like to fix, but for that first we need to be able to investigate/reproduce it.

Regards,
Garvin

[yellowled]

The flattr issue of course I'd like to fix, but for that first we need to be able to investigate/reproduce it.

I already asked for the blogger to maybe report back here in the forum to help troubleshoot this. Since he's running s9y 1.5.5, it's likely his plugins are not up to date, either. Not really sure how to proceed here.

YL

[hds]

i'll try and update everything. then lets see. lazy me...

[hds]

hi there.
i updated now. can't really tell what's wrong with the feed, though.

[garvinhicking]

Hi!

Great. Seems the only issue left is this:

Code: Select all

http://www.amazon.de/Ekstasis/dp/B0081TBZRO/ref=sr_1_3?ie=UTF8&qid

in your article you have a link to that, you need to change that to:

Code: Select all

http://www.amazon.de/Ekstasis/dp/B0081TBZRO/ref=sr_1_3?ie=UTF8&qid

(note the "&" vs. "&").

You could also install the xhtml cleanup event plugin that should fix an issue like this; however, doing it manually is less performance-expensive like installing a plugin

Regards,
Garvin