Every so often I get the strangest stuff in my web server log. I've scratched my head over it many a time. Looks like somebody has downloaded my weblog software and decided to go through every line of the source code and turn it into a web request. These requests always fail. What kind of clueless idiot would do such a thing?

Finally figured out the answer. It isn't a human at all. It also isn't turning every line of the source file into a web request - only those that contain the string 'href='. It's a very poorly written web crawler. It grabbed my source code file and ignored the fact that it's a file of type 'application/x-tar'. It thinks it's a normal web page (which would be type 'text/html'). And it is extracting everything on the page that looks remotely like a URL and fetching it. The only problem is that these aren't URL's. They are computer code which constructs URL's. They are nonsense unless you run it through a processor to execute the code -  which then creates the URL.