can't get my RSS to validate - Help!!

[lawrence]

I'm running this page:

http://www.krubner.com/rss/page938.xml


through this validator:

http://rss.scripting.com/?url=http://www.krubner.com/rss/page938.xml


I can't get this page to validate. I'm getting two errors:

1. pubDate must be an RFC-822 date 9 9
2. Undefined rss element: item 12 7

The second one makes no sense to me. "item" is used in all the
examples they give for .91 RSS, so why is the validator giving me
grief?

 

[Rolf Magnus]

I'm running this page:

http://www.krubner.com/rss/page938.xml


through this validator:

http://rss.scripting.com/?url=http://www.krubner.com/rss 2Fpage938.xml


I can't get this page to validate. I'm getting two errors:

1. pubDate must be an RFC-822 date 9 9
2. Undefined rss element: item 12 7

The second one makes no sense to me. "item" is used in all the
examples they give for .91 RSS, so why is the validator giving me
grief?

I don't know rss, but your items are direct children of rss, and
according to the dtd, the only valid child element for rss is channel
and the only valid parent element for item is also channel.

 

[Andy Dingley]

I'm running this page:

http://www.krubner.com/rss/page938.xml

through this validator:

http://rss.scripting.com/?url=http://www.krubner.com/rss/page938.xml

I'd believe George Bush before I trusted the accuracy of a validator
from scripting.com -- and for similar reasons; both are always "right"
(i.e. not obvious enough to clearly catch them at it) but they'll
cheerfully redefine black as white to push their own agendas.

I can't get this page to validate. I'm getting two errors:

1. pubDate must be an RFC-822 date 9 9

2. Undefined rss element: item 12 7

You need to have <item> as a child of <channel>. This 0.91 feed has it
as a child of <rss> (which is where RSS 1.0 places it, so that might
have caused some confusion)

I wouldn't suggest 0.91 either, as a format for new work. 1.0 is best
(a hobby horse of mine), but go to 0.92 as a minimum.

 

[lawrence]

I'm running this page:

http://www.krubner.com/rss/page938.xml


through this validator:

http://rss.scripting.com/?url=http://www.krubner.com/rss/page938.xml


I can't get this page to validate. I'm getting two errors:

1. pubDate must be an RFC-822 date 9 9
2. Undefined rss element: item 12 7

The second one makes no sense to me. "item" is used in all the
examples they give for .91 RSS, so why is the validator giving me
grief?

Thanks very much. I rearranged things and got it to validate, using
the validator from scripting.com.

However, it still fails validation at
http://feeds.archive.org/validator/

The problem is "bad characters". I'm not sure how to start to debug
that. What is the first thing I should look for?

I wrote the 0.91 script (a PHP script) first because I assumed it
would be easy. I figured once I worked out the troubles at 0.91, I
could move up and do scripts for 1.0 and 2.0. The fact that I'm having
trouble even at the 0.91 level makes me happy that I didn't jump
straight to the 1.0 or 2.0 level.

 

[Johannes Koch]

However, it still fails validation at
http://feeds.archive.org/validator/

The problem is "bad characters". I'm not sure how to start to debug
that. What is the first thing I should look for?

Did you change anything? I get "This is a valid RSS feed." with <>
http://feeds.archive.org/validator/>.

 

[Andy Dingley]

However, it still fails validation at

http://feeds.archive.org/validator/

The version this morning seems OK. I suggest you keep an archive of
these bad feeds under static URLs, so that we can more easily see the
problems.

The problem is "bad characters". I'm not sure how to start to debug
that. What is the first thing I should look for?

Most likely thing is the use of HTML entities (like &eacute; ) that
aren't valid because RSS is an XML protocol and doesn't recognise
these HTML-defined entities. A _very_ common error in RSS feeds.

I wrote the 0.91 script (a PHP script) first because I assumed it
would be easy.

I have to ask why anyone needs to write RSS scripts these days ?
(Although I spent yesterday doing it myself). There are very many
already out there, and it's a rare situation that really needs
something written from scratch.

I suggest that you learn a bit more detailed XML (entities for one
thing, namespaces for another) and learn how to read a formal DTD.
It's one thing to make a feed work once during testing, but quite
another to make a reliable feed that handles all the data it will meet
over its lifetime. Sadly RSS tools suffer badly from this - they run
for a week, then crash when they meet their first accented European
character. To get a reliable feed, you really do need to know how to
_understand_ the specification, not just match up one example.

As to the versions, then I'd support 0.92 and 1.0 1.0 is best, 0.92
solves compatibility issues for some older or simpler aggregators. 2.0
is pointless.

You don't need to understand Dublin Core to use RSS 1.0, but you will
do if you want to really use it. Worth the effort of studying it
anyway (Learning Dublin Core is vastly more useful than learning RSS
anyway)

Hang in there - RSS versions don't really vary that much between
themselves.

 

[lawrence]

The version this morning seems OK. I suggest you keep an archive of
these bad feeds under static URLs, so that we can more easily see the
problems.


Most likely thing is the use of HTML entities (like &eacute; ) that
aren't valid because RSS is an XML protocol and doesn't recognise
these HTML-defined entities. A _very_ common error in RSS feeds.

I'm stumped when I think of all the possible user inputs that can mess
up an RSS feed. Myself and a friend have been working on some weblog
software for a year now - the weblog entries show up in the RSS feed.
Lots and lots of problems come up - people might copy and paste
something from Word in their weblog and then get a Windows character
that is not in the defined char set for the RSS feed. How to protect
against that. Or people can, as you say, add in some HTML entities, or
sometimes the software does that automatically.

It's true that I could write filters for each possible problem as it
comes up, but it seems like it would take a mountain of code and there
would never be an end to it - after I figure out how to convert
unexpected European accented characters, then I'd have all the Asian
alphabets to work on.

I guess I'm asking if there is an easy, elegant way to convert a set
of characters to some particular character set. I don't know of any
myself. What strategies have other programmers used when tackling this
problem?

 

[lawrence]

(e-mail address removed) (Andy Dingley) wrote in message >

I have to ask why anyone needs to write RSS scripts these days ?
(Although I spent yesterday doing it myself). There are very many
already out there, and it's a rare situation that really needs
something written from scratch.

I think that is a fair question. Part of the answer would be, I guess,
we want to be able to assert that our weblog software is "complete" in
some sense. Without RSS, it is not complete. The other aspect of it,
call it an institutional mission, is that we want to donate all our
code to the public domain. That means we often have to do things from
scratch, because other we might use is copyrighted. There is a lot of
good code under the GPL, but the GPL is too restrictive for our
purposes.

I suggest that you learn a bit more detailed XML (entities for one
thing, namespaces for another) and learn how to read a formal DTD.
It's one thing to make a feed work once during testing, but quite
another to make a reliable feed that handles all the data it will meet
over its lifetime.

I really would love to learn more about XML and I've bought a bunch
books from O'Reilly. Sadly, I'm stretched a little thin. Me and some
friends have been working on a content management system and I've had
to learn a lot about a lot of different stuff, and I haven't had the
chance to become very good at anything. I'd like to know more about
XML, XHTML, Apache, CSS 2.0, RSS, Linux, Windows XP, mime types,
character encodings, SOAP, Java, web services, Flash, PostGreSql,
Swing, MySql, encryption and a whole lot more. It's all just too much.
My hope, at this point, is they'll come a point when the basic
abilities of our CMS are stable enough that I can go over it again and
go deeper into certain areas when my knowledge is better. That, or
maybe our project can pick up some programmers who are more talented
than I.

We're giving away our (still quite buggy) software here:

http://www.publicdomainsoftware.org/

 

[Andy Dingley]

I'm stumped when I think of all the possible user inputs that can mess
up an RSS feed.

Can't be done. Never was possible - users are just too inventive.

Instead, look at it the other way. Find the set of all things that are
_valid_ RSS and exclude anything else. This is a smaller set than the
invalid stuff, and it's better documented. If you _only_ allow the
valid sequences, then you have implicitly forbidden all the invalid
stuff.

Definitely read this
http://diveintomark.org/archives/2004/02/04/incompatible-rss

It's true that I could write filters for each possible problem as it
comes up,

No, you can't. You might possibly manage to do it for one day, but
there will always be something new coming along tomorrow. Take it the
other way.

 

[lawrence]

Can't be done. Never was possible - users are just too inventive.

Instead, look at it the other way. Find the set of all things that are
_valid_ RSS and exclude anything else. This is a smaller set than the
invalid stuff, and it's better documented. If you _only_ allow the
valid sequences, then you have implicitly forbidden all the invalid
stuff.

Definitely read this
http://diveintomark.org/archives/2004/02/04/incompatible-rss

Thanks for the link to the Mark Pilgrim article. Despite the obvious
anti-Dave Winer bias in the article, I thought it was very good and
very informative.

I like your suggested style for moving forward but don't understand
how to do it. Should I take an input (when a user posts) and take
every character and put it into an array and then test it against an
array full of those character's that are allowed? That seems
cumbersome, though it only happens when the user inputs stuff, so
being cumbersome at that point isn't lethal. But is there a more
graceful way to validate the input?