Set your spinnarets to ON

[ataxi]

If I wanted to set up an RSS feed for my journal on a personal site (say, my UCC public-html area*), how hard would it be and what would I use**? What if I wanted to create a parallel site that aggregated everything I like to read in one place? Could that site also access locked LJ posts with appropriate authentication efforts? Is there a pre-

Comments

[theducks]

I'm currently playing with magpierss .. it's pretty easy:

require_once('magpierss/rss_fetch.inc');
$rss=fetch_rss("http://theducks.livejournal.com/data/rss");

foreach( $rss->items as $item) {
$i++;
if ($i < 5) {
if ($i > 1) { echo ""; }
echo "$item[pubdate] - $item[title]
$item[summary]
Comment on this
";
}
}
echo "Archives";
?>

[ataxi]

Neat. Can you get someone's locked posts that way?

[theducks]

No

[ataxi]

That's one of the main issues for me ... although I guess I could get away with not reading locked posts from here on in.

[anonymous]

Davyd uses the code from Planet UCC to power his home page, which is a replication of his LiveJournal. I'm not sure if the feedreader modules used in Planet UCC support the various authentication schemes offered by LiveJournal, but I'm guessing they probably do (or can be made to work).

Unfortunately, it's written in Python, which is a lovely language for quick hacks like this one but you haven't listed it above...

Also, running it at UCC could be difficult because of the traffic filtering - LiveJournal is not free traffic. Things like Planet UCC are actually powered by machines outside the UWA network, and FTPed in automatically.

- David 'Zanchey' Adam

[dannipenguin]

I see that Zanchey has already told you my magic secret already.

Indeed, the Planet UCC engine is written in Python and makes extensive use of the 'feedparser' module written by Mark Pilgrim. The HTML is also statically generated and uploaded to the UCC when required from a machine not inside the UCC.

The feedparser module is by critical acclaim the most flexible and powerful RSS/Atom parser available on the net (it parses something like 9 not-compatible versions of RSS) as well as Atom.

I am fairly sure you can get locked posts if you pass a valid Livejournal login cookie when you make the request (this might be interesting to implement).

. You'll want to look in the XMLParse2 module to see how to use feedparser.

As Zanchey covered, it's all in Python, but that's ok, because Python is better then PHP anyway.

[ataxi]

I've done some Python. It's easy to do easy stuff with it. I will have a look at your shitz if I ever get around to this little project. Thanks Zanchey for the headsup on the limitations of UCC's hosting.

[dannipenguin]

I can expose the Planet UCC 'CacheObject's via some useful protocol, but they only contain the blogs of people on Planet UCC. They also don't contain locked posts.