Question

A google search gives me the following first result on HTML:

<h3 class="r"><a href="https://rads.stackoverflow.com/amzn/click/com/0470284889" rel="nofollow noreferrer" class="l vst" onmousedown="return rwt(this,'','','','1','AFQjCNEv1W9YC2jcSKYdEo2kNqBMJ-Utmg','k89K9hF4cVNpxQYHtEKiUQ','0CCoQFjAA',null,event)"><em>Quantitative Trading</em>: <em>How to Build Your Own Algorithmic</em> <b>...</b> - Amazon</a></h3>

I would like to extract the link http://www.amazon.com/Quantitative-Trading-Build-Algorithmic-Business/dp/0470284889 from this, but when I use beautiful soup to extract the information, I obtain

soup.find("h3").find("a").get("href")

I obtain the following string instead:

/url?q=http://www.amazon.com/Quantitative-Trading-Build-Algorithmic-Business/dp/0470284889&sa=U&ei=P2ycT6OoNuasiAL2ncV5&ved=0CBIQFjAA&usg=AFQjCNEo_ujANAKnjheWDRlBKnJ1BGeA7A

I know that the link is in there and I could parse it by deleting the /url?q= and everything after the & symbol, but I was wondering if there was a cleaner solution.

Thanks!

Answer 1

You can use a combination of urlparse.urlparse and urlparse.parse_qs, e.g

>>> import urlparse
>>> url = '/url?q=http://www.amazon.com/Quantitative-Trading-Build-Algorithmic-Business/dp/0470284889&sa=U&ei=P2ycT6OoNuasiAL2ncV5&ved=0CBIQFjAA&usg=AFQjCNEo_ujANAKnjheWDRlBKnJ1BGe'
>>> data = urlparse.parse_qs(
...     urlparse.urlparse(url).query
... )
>>> data
{'ei': ['P2ycT6OoNuasiAL2ncV5'],
 'q': ['http://www.amazon.com/Quantitative-Trading-Build-Algorithmic-Business/dp/0470284889'],
 'sa': ['U'],
 'usg': ['AFQjCNEo_ujANAKnjheWDRlBKnJ1BGe'],
 'ved': ['0CBIQFjAA']}
>>> data['q'][0]
'http://www.amazon.com/Quantitative-Trading-Build-Algorithmic-Business/dp/0470284889'