It's all possible, but just a matter of time.
If those sites do provide API for search and download it can be done much more quickly and stable.
For opensubtitles API see here: http://trac.opensubtitles.org/projec...es/wiki/XMLRPC

If they don't it's still theoretically possible to parse HTML and break CAPTCHA, but doesn't worth it

At the moment I'm finishing rewriting this script into python class so it can be reused by developers of other media-related projects.

Post here links to subtitle APIs and we'll see what can be done.

And another nota bene: this python script can still be used even in nautilus on desktop. Just test it and say if I didn't break anything :]