Support Python3

[ClemPi]

Code bytes http response. Handle urllib refactor.

Address #2

[valerytschopp]

Hi ClemPi, For the urllib refactor, wouldn't it be better to do something like this:

try:
    from urllib.parse import urlparse
except ImportError:
     from urlparse import urlparse

instead of explicitly checking the python version?

[ClemPi]

I did this because the method is not the same and we have to explicitly check the version when calling urlencode : https://github.com/valerytschopp/python-radosgw-admin/pull/4/commits/566adf64a0bae09c967109703153a11a2c13dc38#diff-fe3c0a5df953368eb825f1815aa2b8c9R99. But we can do the other way if you prefer

[valerytschopp]

Yes, I understand. Sorry my example was bad, I really would prefer:

try:
    from urllib.parse import urlencode
except ImportError:
    from urllib import urlencode

...
      urlencode(query_params)
...

[ClemPi]

Yes, better alternative. Just updated the PR