Finally got around to updating py_shoutcast to the new Shoutcast XML format/layout. This was actually reported to me back at the end of May by Mats Joensson (along with a patch), just haven’t had the time to look at it before now. I would have much preferred to use his patch, but the upstream changes pretty much required a complete rewrite in the parsing code.
If anyone is actually using the python module for anything other than the Freevo plugin, you’ll notice calling ShoutcastFeed.parse_stations() no longer does it all. This is because Shoutcast now requires a fetch of the genre list, then fetching a list of of the stations in that genre. Now you have to use GenreFeed.parse_genres() to grab the list genres, then call ShoutFeed.parse_stations() for each genre.
The parameters have also changed for parse_stations(), most notably the no_compress and feed_max have been removed. They simply don’t exist anymore upstream. One nicety is that I was able to remove the genre regex kludge, and the genre/station list split had resulted in things feeling much faster in the Freevo plugin.
I’ve setup a Trac install, so you’ll be able to find the newest releases on the download page there. Please submit any bugs, patches (my favorite) or feature requests on the ticket tracker.