[python] How to urlencode a querystring in Python?

I am trying to urlencode this string before I submit.

queryString = 'eventName=' + evt.fields["eventName"] + '&' + 'eventDescription=' + evt.fields["eventDescription"]; 

Python 2

What you're looking for is urllib.quote_plus:

>>> urllib.quote_plus('string_of_characters_like_these:$#@=?%^Q^$')
'string_of_characters_like_these%3A%24%23%40%3D%3F%25%5EQ%5E%24'

Python 3

In Python 3, the urllib package has been broken into smaller components. You'll use urllib.parse.quote_plus (note the parse child module)

import urllib.parse
urllib.parse.quote_plus(...)

For use in scripts/programs which need to support both python 2 and 3, the six module provides quote and urlencode functions:

>>> from six.moves.urllib.parse import urlencode, quote
>>> data = {'some': 'query', 'for': 'encoding'}
>>> urlencode(data)
'some=query&for=encoding'
>>> url = '/some/url/with spaces and %;!<>&'
>>> quote(url)
'/some/url/with%20spaces%20and%20%25%3B%21%3C%3E%26'

Try this:

urllib.pathname2url(stringToURLEncode)

urlencode won't work because it only works on dictionaries. quote_plus didn't produce the correct output.

For Python 3 urllib3 works properly, you can use as follow as per its official docs :

import urllib3

http = urllib3.PoolManager()
response = http.request(
     'GET',
     'https://api.prylabs.net/eth/v1alpha1/beacon/attestations',
     fields={  # here fields are the query params
          'epoch': 1234,
          'pageSize': pageSize 
      } 
 )
response = attestations.data.decode('UTF-8')

Try requests instead of urllib and you don't need to bother with urlencode!

import requests
requests.get('http://youraddress.com', params=evt.fields)

EDIT:

If you need ordered name-value pairs or multiple values for a name then set params like so:

params=[('name1','value11'), ('name1','value12'), ('name2','value21'), ...]

instead of using a dictionary.

Python 3:

urllib.parse.quote_plus(string, safe='', encoding=None, errors=None)

for future references (ex: for python3)

>>> import urllib.request as req
>>> query = 'eventName=theEvent&eventDescription=testDesc'
>>> req.pathname2url(query)
>>> 'eventName%3DtheEvent%26eventDescription%3DtestDesc'

Another thing that might not have been mentioned already is that urllib.urlencode() will encode empty values in the dictionary as the string None instead of having that parameter as absent. I don't know if this is typically desired or not, but does not fit my use case, hence I have to use quote_plus.

In Python 3, this worked with me

import urllib

urllib.parse.quote(query)

Context

• Python (version 2.7.2 )

Problem

• You want to generate a urlencoded query string.
• You have a dictionary or object containing the name-value pairs.
• You want to be able to control the output ordering of the name-value pairs.

Solution

• urllib.urlencode
• urllib.quote_plus

Pitfalls

• dictionary output arbitrary ordering of name-value pairs
  • (see also: Why is python ordering my dictionary like so?)
  • (see also: Why is the order in dictionaries and sets arbitrary?)
• handling cases when you DO NOT care about the ordering of the name-value pairs
• handling cases when you DO care about the ordering of the name-value pairs
• handling cases where a single name needs to appear more than once in the set of all name-value pairs

Example

The following is a complete solution, including how to deal with some pitfalls.

### ********************
## init python (version 2.7.2 )
import urllib

### ********************
## first setup a dictionary of name-value pairs
dict_name_value_pairs = {
  "bravo"   : "True != False",
  "alpha"   : "http://www.example.com",
  "charlie" : "hello world",
  "delta"   : "1234567 !@#$%^&*",
  "echo"    : "[email protected]",
  }

### ********************
## setup an exact ordering for the name-value pairs
ary_ordered_names = []
ary_ordered_names.append('alpha')
ary_ordered_names.append('bravo')
ary_ordered_names.append('charlie')
ary_ordered_names.append('delta')
ary_ordered_names.append('echo')

### ********************
## show the output results
if('NO we DO NOT care about the ordering of name-value pairs'):
  queryString  = urllib.urlencode(dict_name_value_pairs)
  print queryString 
  """
  echo=user%40example.com&bravo=True+%21%3D+False&delta=1234567+%21%40%23%24%25%5E%26%2A&charlie=hello+world&alpha=http%3A%2F%2Fwww.example.com
  """

if('YES we DO care about the ordering of name-value pairs'):
  queryString  = "&".join( [ item+'='+urllib.quote_plus(dict_name_value_pairs[item]) for item in ary_ordered_names ] )
  print queryString
  """
  alpha=http%3A%2F%2Fwww.example.com&bravo=True+%21%3D+False&charlie=hello+world&delta=1234567+%21%40%23%24%25%5E%26%2A&echo=user%40example.com
  """ 

If the urllib.parse.urlencode( ) is giving you errors , then Try the urllib3 module .

The syntax is as follows :

import urllib3
urllib3.request.urlencode({"user" : "john" }) 

Note that the urllib.urlencode does not always do the trick. The problem is that some services care about the order of arguments, which gets lost when you create the dictionary. For such cases, urllib.quote_plus is better, as Ricky suggested.