CôaCôaCôa

Il est beau mon RSS

This morning, thanks to .net Weblog Archives, I see that Scot tW bookmark a tip to translate RSS with XSLT. And it's reminds me another beautifull way to syndicate an rss feed with some XSLT and JavaScript:

• RSS + XSLT
• More details : s l a m (by Marc Barrot)
• My updated XLST file

And after reading Scobleizer, following some links ... I found WebOutliner (by the same Marc Barrot). And still with this link, I came to Pepys, an hypertext notebook for Windows.

Ma Peugeot est fantastique

MS SpeedRegulator.NET, MS SpeedLimiter.NET, MS RainSensor.NET, MS LightSensors.NET, MS GPSlocalisation.NET... Soon ?

{via Scobleizer}

Agglutinage

I'm trying to build an application which would combine several rss feeds into only one feed. I had this brilliant idea (hum!) and my first tests are visible here.

For my own use, even if I assimilated that C# didn't like « if (bCheck = true) », I'm stiill wrong with « If (bCheck == true) ». In a more general way, I found some nice things.

Rss requests

Instead of requesting each feed regularly, I scan weblogues.com (a weblogs.com like for a French-speaking weblogs) every hour. So I know which weblogs were updated, wheter a weblog has a feed or not,  and then I request only those updated with a feed.

That appears to me as a good alternative to ETag and If-Modified-Since checkings.

Rss url

For the moment, I parse weblogues.com source page to determine wheter a weblog has an rss feed (because when I started changes.xml wasn't reliable).

Then I first "discover" rss url with a trick. For a weblog hosted at Joueb.com, rss feed is always http://joueb.com/xxxxxxxx/rss.shtml, for a weblog hosted at U-Blog.net, rss feed is always http://www.u-blog.net/xxxxxxxx/rss and so on (this works with weblogs.net too)...

If this easy way can't work, I do a really work:

• RSS auto-discovery.
• When RSS auto-discovery isn't supported, I saw Aggie try some standards rss feed names, so do I.
• If I still don't find rss url, then I give up (let's say 1%).

Now that changes.xml is quite ok, I consider to use it. But it doesn't indicate if a weblog has a feed or not. Consequently, I will have to do the checking, including on weblogs which don't have rss feed. Thus I need to improve my RSS locator (source RSS-Bandit).

And now, with BlogStreet it must be possible to have rather good results. More than 10000 rss feeds in its directory,a web service to read this information and even a rss generator.

I'll put that to my todo list. But first I will send a mail to the webmaster to see whether he can add a "rssUrl" attribute to the changes.xml file.

Déclinaisons

• Content © 2003 Microsoft Corporation (Don Box)
• This weblog does not represent the thoughts, intentions, plans or strategies of my employer. It is solely my opinion (DevHawk)
• If you think these posts reflect the opinion of my company, you're a f#%*ing moron (InkBlog)