Fixes #10

[erfaan]

There is no need to manually join the urls when we have urljoin

[peterbe]

Do all tests work with this?

I'm wondering if I didn't use urlparse.urljoin for a reason. I can't remember.

[peterbe]

I can't merge this without tests but I'll take a look if you don't have time to write some.

[peterbe]

I did find a flaw in the proxy/app.py script to do with script tags. Look at it now.
http://cl.ly/Mqcb

[erfaan]

Yes all tests ran smoothly. Just tested it. The issue is not related to proxy/app.py but the processor.py
What exactly did you change in proxy/app.py?

I am not sure how you managed to make it working. I tried the following code and it failed:

>>> from mincss.processor import Processor
>>> p = Processor()
>>> p.process_url('http://www.i.com.pk')
>>> p.process()

Regarding urljoin, I guess that is the proper way to join urls.

[peterbe]

Ok. Great I can reproduce that now.

I still can't merge your pull request if it doesn't have tests.

And lastly, please don't teach me about how urljoin works.

[peterbe]

You were right. It works with plain urljoin. I can no longer find any reason why it wouldn't work. And I added some unit tests too.
2786c69