article.movies missing 'http:'

[adamlwgriffiths]

I've noticed results from Article.movies are missing the protocol prefix.

>>> import newspaper
>>> url = 'http://www.rockpapershotgun.com/2014/07/24/top-down-tracy-third-eye-crime/'
>>> a = newspaper.Article(url)
>>> a.download()
>>> a.parse()
>>> a.movies
['//www.youtube.com/embed/LgNLRT6QyQE', '//www.youtube.com/embed/jsqVLa1yy1M']

$ pip show newspaper

---
Name: newspaper
Version: 0.0.7
Location: /home/adamgriffiths/.anaconda/envs/collected-redux/lib/python2.7/site-packages
Requires: lxml, requests, nltk, Pillow, cssselect, BeautifulSoup

[adamlwgriffiths]

I also just noticed that those urls aren't unicode, but in the README they are

>>> article.movies
[u'http://youtube.com/path/to/link.com', ...]

[adamlwgriffiths]

Ok, the missing http:// is from the html actually having the link as '//youtube.com/...'
How thats even valid is beyond me.

I'm not sure if you consider this an issue or not. Feel free to close if not.
Preferably newspaper would validate the links and ensure they are absolute links.

[codelucas]

Refer to: http://stackoverflow.com/questions/11881054/is-a-url-starting-with-valid

Those URLs are "scheme relative" and are totally valid.
A great example of these URLs in action: https://developers.google.com/speed/libraries/devguide
Look at the URLs in the <script> snippets.

[adamlwgriffiths]

Interesting, I've never seen that before.
Cheers for the clarification =)