What I’ve been up to

Work is far too busy, and it doesn’t give me anything to talk about – my job all day is to put out fires in Windows Vista, and there’s definitely no shortage of work in that department (though we’re definitely making progress!)

My Yahoo! Pipe

However, since I haven’t got any juicy Microsoft secrets to spill, instead I will share my clever Yahoo! Pipes rig that works like TiVo only free’er. Here’s a screenshot of its PVC goodness, click it to be taken to the page:

Yahoo! Pipes + Awesome = Very Yes

But what does this do?

What does this actually do for us though? Well, the output of this pipe is an RSS feed of Bittorrent files – put these into your favorite RSS-enabled torrent program, and it will watch the output of this pipe, and add all of the torrents to be downloaded! My favorite one is KTorrent:

Auto-downloaded goodness

How it all works

Here’s what some of these boxes do – keep in mind, our goal is to end up with a list of torrent files to download of our favorite new shows, with no duplicates or files we already have. I didn’t really arrange the pipes diagram very well – the flow of data is from right to left, and ending up in the center. So, starting at the right, we see that we have a lot of Fetch Feed blocks. These blocks serve as our data source, and fetch the results of searches from tvrss.net. All of these go into a Union operator, which just mixes all of our sources into one list. So after we do that, we end up with data that looks like:

As you can see, we’ve got some problems: first, we don’t really want to treat the 0TV and the REPACK versions as separate, so let’s filter out the release group from the title with a Regex block; it’s like ‘search and replace’ but with a vague ‘search’ part. We replace ‘[anything]‘ with nothing – in code it looks like “\[.*?\]“.

After that, I only want certain seasons of shows, so we only permit items if they contain certain text, like “Scrubs 7x”. Once we’ve done that, we’re almost finished; our only problem is that we’ve got a lot of duplicate entries, and they’re in random order. To fix that, we’ll use the Unique and Sort blocks to put them in chronological order.

The results