CODEWITHSUNDEEP
pythonstringurl
user1637000
user1637000

Reputation: 162

Google URL Builder in Python

I'm trying add Google URL Builder's functionality into my application.

https://support.google.com/analytics/answer/1033867?hl=en

Unfortunately, I'm not sure to get the exact results..

My code

        def buildurl(url):

            #take out old url builder
            url = sub('\?utm_source=.*?(&|$)utm_medium=.*?(&|$)|utm_term=.*?(&|$)|utm_content=.*?(&|$)|utm_campaign=.*?(&|$)','',url)

            #build url
            header += '?utm_source=' + self.data['source']
            header += '&utm_medium=' + self.data['medium']
            header += '&utm_campaign=' + self.data['campaign']

            #return long url
            return(url header + urllib.quote(header)

My code returns this: http://iipdigital.usembassy.gov/st/english/article/2014/08/20140813305633.html#axzz3ANwb5XD?utm_source=source&utm_medium=medi&utm_campaign=testu

Google's URL Builder Returns this: http://iipdigital.usembassy.gov/st/english/article/2014/08/20140813305633.html?utm_source=source&utm_medium=medi&utm_campaign=test#axzz3ANwb5XDu

I could push the #axzz3ANwb5XDu to the back, but is there a way to parse and reconstruct the url in a standardized way?

Upvotes: 1

Views: 1370

Answers (3)

chjortlund
chjortlund

Reputation: 4017

I would go for Pythons urllib - it's a build in library.

import urllib.parse

getVars = {'var1': 'some_data', 'var2': 1337}
url = 'http://domain.com/somepage/?'

print(url + urllib.parse.urlencode(getVars))

Output:

http://domain.com/somepage/?var2=1337&var1=some_data

Upvotes: 1

phihag
phihag

Reputation: 287835

There is a way to parse the URL; it's called urlparse:

try:
    from urllib.parse import urlparse, urlunparse
except ImportError:  # Python 2.x
    from urlparse import urlparse, urlunparse


def buildurl(url):
    scheme, netloc, path, params, query, fragment = urlparse(url)

    #take out old url builder
    query = sub('\?utm_source=.*?(&|$)utm_medium=.*?(&|$)|utm_term=.*?(&|$)|utm_content=.*?(&|$)|utm_campaign=.*?(&|$)', '', query)

    #build url
    query += '?utm_source=' + self.data['source']
    query += '&utm_medium=' + self.data['medium']
    query += '&utm_campaign=' + self.data['campaign']
    return urlunparse((scheme, netloc, path, params, query, fragment))

Upvotes: 0

Matt
Matt

Reputation: 5168

You should checkout the urlparse module. I have modified your code such that it removes the existing url builder parts but keeps any other parts of the query.

from urlparse import urlparse, urlunparse

def buildurl(url):

    #take out old url builder.
    url = sub('utm_source=.*?(&|$)utm_medium=.*?(&|$)|utm_term=.*?(&|$)|utm_content=.*?(&|$)|utm_campaign=.*?(&|$)','',url)


    #Parse the url.        
    o = urlparse(url)

    #build url query.
    query = o.query
    query += 'utm_source=' + self.data['source']
    query += '&utm_medium=' + self.data['medium']
    query += '&utm_campaign=' + self.data['campaign']

    #return the url with the corrected query.
    return urlunparse(o.scheme, o.netloc, o.path, o.params, query, o.fragment)

The fragment should be at the end of the url.

Upvotes: 1

Related Questions