Homestar Runner Wiki Forum

It would be really cool if the forum was in RSS/XML.
But that may be to hard to do....

Your wish is.. well, it's a good way for me to kill some time. Turns out this was really simple, since there's just such a mod in the phpBB verified mods database. The syndication URL is:

http://forum.hrwiki.org/rss.php

There's some extras, too. You can specify an argument for how many posts you want it to display:

Show 15 posts (c stands for count):
http://forum.hrwiki.org/rss.php?c=15

And you can specify which forum you want posts from (skip this and it'll show all posts from all forums):

Show only posts from the Toons forum (f stands for forum):
http://forum.hrwiki.org/rss.php?f=4

By the way, the number that goes with each forum is the same number that's at the end of the URL when you view that forum normally (e.g. Emails = 14, Off-subject = 12, etc.).

Oh, and you can combine them:

http://forum.hrwiki.org/rss.php?c=15&f=4

There's also a 't' option that's supposed to show only topics (instead of full posts), but, er, I can't get it to work.

Enjoy.

_________________

Cool! Now you just need a link on the main page.

All in good time. Eventually there'll be an RSS feed for the Wiki as well.

_________________

You guys are awesome.

I don't know about the other mods... but I think bored is probably a better term. (no offense jones, I do know how much time you all put in here, and certainly appreciate it... a lot)

_________________

How's the status?

Also, if not all, you could just RSS the H*R updates page at first, which I check the most often (if it's easier to just set it up for one page, that is).

Don't slap me with a herring for being ignorant...but what is RSS?

These aren't the droids you're looking for..

Please look here.

_________________

Really Simple Syntax (not sure of the last one)
It's a news delivering service that uses readers to receive it. It's like a news article, but online.

_________________
Evidence of the ol' glassies! Nothing up our sleeves, no magic little Alex! A job for two who are now of job age! The police!

If you want a better (and more accurate) explanation, please visit my link above.

_________________

Nice. Will there also be one for the H*R Updates page?

No, because we'd have to build it by hand every time there was an update. The Wiki won't generate something like that automagically (though it would be an interesting project.. hmmm..).

If you want a feed of all of the updates on H*R.com, there are two guerilla feeds out there that I've found:

http://interglacial.com/rss/homestar.html
http://www.livejournal.com/~homestar_rss/

For the latter, click on "syndication information" to get the URI for your aggregator. Neither of them seem perfect.

_________________

It looks quite doable, although I have very little experience with PHP and such.

The way I see it, you'd just have to make a script that would convert http://www.hrwiki.org/index.php/Templat ... action=raw into something like this (made that manually, but it seems to work in Opera's RSS reader).

I'm looking into it myself, but if it's easy to do and not very time-consuming, some of you more experienced members might help me out.

Whoa.. I didn't know there was an action=raw option.

But actually, from a screen-scraping perspective, it would be just as easy to use the regular HTML version of page the page to generate a feed as it would to use the "raw" version. The purist in me says the raw version would be better because it's so bare -- it has a very high signal-to-noise ratio. However, the HTML that MediaWiki generates is very clean and structured. Every day on the page is an <h2/> tag followed by an unordered list. And as a bonus, the page validates as XHTML Transitional. This is a cinch for a DOM parser. Another advantage over the "raw" version is that all of the links are already built, so it would take minimal parsing to just add "http://www.hrwiki.org" in front of each href, and any text formatting is already taken care of. So using the HTML version of the page you end up throwing a lot more away, but in the end it's easier because there are already libraries in place to parse the data.

I think I'm gonna have to play with this a bit later. You've played me like a fiddle.

_________________

I'm using Firefox, and usually I can see if there is an RSS feed in the lower right corner of the browser. For example, it appears on the Recent Changes page of the Wiki. However, that RSS icon doesn't appear on the forum. I see that you link to an RSS feed above, so what does it take to get Firefox to notice it? Does it need to be linked to in some tag?

Yes, it needs to be in a <link rel="alternate"/> tag, and that would require editing the template.

If you want to use the feed as a live bookmark in Firefox, just copy the feed URI (it'll end in .xml or .rss) and then in the Bookmarks Manager (Bookmarks > Manage Bookmarks...) and go to File > New Live Bookmark... and past the URI in the "Feed Location:" box.

_________________

Justas, you were right: Using the action=raw output is a lot easier.

I made a lot of progress using a SAX parser (XML_HTMLSax3), since a DOM-based solution wasn't going to work, not without some serious over-my-head XPath mojo (since a section's content is not a child of the section heading; boo), but I rant into more problems with the SAX approach (I couldn't treat everything in a <li/> as one chunk, I had to deal with every part (e.g. data, <a>, data, </a>, data) as a separate chunk. What a nightmare. But it was fun while it lasted.

So next time 'round I'm just going to parse the action=raw output line by line and hope that whoever edited the page last didn't leave a mess. Had I started out this way I'd be asleep by now; but oh well, it was a learning experience.

_________________

Thumbs up!

Also, to prevent you from the dangers of malnutrition for yet another sleepless night, I've brought you cookies.

Awesome. I love crumbs.

_________________

How's the status?

Er, busy.

_________________

It seems that a bit of HRWiki magic is happening, what with the emoticon keep your pants on and the temporary wiki failure. Would it be possible to add the <link rel="alternate"/> for the RSS feed while all these changes/upgrades are happening?

Done.

[annoying kid on the backseat]

Any progress on the RSS front?

[/annoying kid on the backseat]

[irate parent in the front seat]

Don't make me turn this car around!

[/irate parent in the front seat]

Seriously, I've been busy, and tired, and, um, lazy.

_________________

Ah, yes. The three words with which I am unfortunately also well familiar. Nonetheless, good luck for yon future and may ye be blesseth with sufficient stamina to carry forth this undertaking till its end!

And some nachos, too.

...and then I made one myself:

http://justas.ggcmedia.com/hrw.php

It doesn't do link conversion or fully support all the wiki tags somebody might decide to put there, but I suppose the fact that it works at all is promising, considering my near non-existant PHP skills (most of which I picked up today, during the making of this).

I can also provide the source file, if anyone is interested in having a look.

Justas,

Great work! It's always nice when somebody else does what I was too lazy to get done myself. I wouldn't mind seeing your source.

A couple comments:

Firstly, it's really slow. I don't know if this is the script or your server's bandwidth or my connection, but it takes a long time to retrieve.

Secondly, I personally would have made each individual update (i.e. each bullet item rather than each day) as its own feed item, as it would make new updates more at-a-glance. But that's just me.

Lastly, concerning links, I found this bit of code that converts Wikipedia links into HTML links. It's in the form of a Wordpress plugin, but I think it would be really easy to adapt to any PHP script.

Again, great work.

_________________

That was probably our server. At first, I also thought that all those loops and newbie code probably ate up the server's CPU speed, but the refreshing/loading time varied from 2 seconds to half a minute, so I'm assuming something was temporarily wrong over there. But the script could probably use some optimisation as well, so:


Here you go: http://justas.ggcmedia.com/hrwc.txt

Looks rather clumsy, compared to that other converter you found there.
I added a lot of comments here, to make it more legible to the average human being. But anyway, you're free to improve upon this, implementing the links feature and/or more items per update, etc. As long as I get partial credit for any of my crap that survives the editions