I'm not going to clone a git repo to read a file, but I did waste some time clicking around your site trying to find this file that you mentioned, to no avail. Just FYI.
i mean, in theory, you could do this in Python
>>> print(zlib.decompress(urlopen(
'http://canonical.org/~kragen/sw/pavnotes2.git/objects/03/220089fdc6713fc459ac71da35d6c8abb5b110'
).read()).decode('utf-8'))
but it's easier to just clone the reporeading the file is going to take you an hour or two, so if spending an extra thirty seconds to clone the repo is a significant cost in this context, you probably shouldn't start