I’m working on v0.3 of the SharePoint RSS Feed Generator (from now on I’m calling in WSS RSS). The big feature of v0.3 will be OPML support. However, I’ve also made a structural change as well. Instead of using HTTP Handlers, which requires me either modify a web.config file or create a vDir off the main WSS web server, I’ll be using empty WebForms. These files have nothin but the <%@ Page header in the .aspx file, and I’m manually writing out the RSS/OPML from inside the Page_Load event. It seems sloppy, but since WSS remaps the HTTP Handlers to do away with everything but .aspx/System.Web.UI.PageHandlerFactory mapping, it’s the best approach. Now I can drop rss.aspx/opml.aspx into the _layouts vDir and the compiled DLL into the bin subdirectory and I’m ready to go.

Other than import/export, what kind of OPML support is available in news aggregators? With WSS RSS, you’ll be able to pull down an OPML directory of a WSS site. It would be nice to have an interface to pick and choose which feeds you care about, rather than importing all of them and then deleting the ones you don’t want.

Also, is there any list of standard OPML outline attributes? I’m using: title, description, type, xmlUrl, htmlUrl and SharePointID. The last one is my own creation, and I put it in its own namespace. The others were the attributes of the OPML export of my subscribed news feeds in SharpReader.

Back from vacation, caught up on email and weblogs, and ready to code. Scott sent me some of his weblog code to look at on the plane to Phoenix. How is it that I always seem to work on my blog in Phoenix? Looking forward to also checking out BlogX 1.1 or BlogY or whatever they call it. Also have a few updates to the SharePoint RSS feed generator to implement. There’s been interest inside Microsoft, but I haven’t heard much interest on the outside.

I’m on vacation, but still checking my email. Many thanks to Mickey Williams who pointed out to me that my referral log had been hacked. I’m not sure hacked was quite the right word for it – I was breaking the first rule of Secure Code – Trust User Input at Your Own Peril. The referral HTTP header is a form of user input, and I was happily echoing it back out to the site without any sort of check whatsoever. I guess I should consider myself lucky that I ended up with a page full of porn links rather than something more serious. Obviously, I’ve taken the page down. When I get back from vacation, I’ll check the server log to see when this started happening. Anyone else blindly storing and echoing referrals should keep an eye on their log.

I asked a while ago about canonical weblog names. At the time, I wanted unify the entries in my referral log that pointed back to the same weblog. Now, I want to also eliminate bogus entries as well. Is pingback/trackback the answer?

With the launch of VS.NET 2003, there’s a bunch of new architecture information from Microsoft. A whole new Pattern Repository site is up, featuring patterns for web presentation, deployment, distributed systems and performance and reliability. All of the patterns have communities – check out the Community Resources for Architecture and Design on GotDotNet, including areas on Enterprise Architecture, Application Architecture, Systems Architecture and Patterns and Templates. Tools like a TDL Policy Editor and new Enterprise Templates are also available.

I’ve updated my SharePoint RSS Feed Generator to version 0.2. Primary new feature: RSS feeds for individual lists. As I mentioned in my last post, I’m on vacation next week, so v0.3 will be a while. Current top features being considered for v0.3 are an OPML feed and modifying the WSS UI to include links to RSS feeds (both web feeds and list feeds).

I’ve also been working on a tool to help me explore the Windows SharePoint database. Any interest in that?