pythonurlliburl-encodingquotereserved

Why does urllib.parse.quote encode non reserved characters in urls?

The parse function in urllib.parse can be used to encode url components. But its behavior is different from the standard javascript encoder.

In python

>>> import urllib
>>> urllib.parse.quote('(a+b)')
... '%28a%2Bb%29'

in Javascript

>>> encodeURIComponent('(a+b)')
... "(a%2Bb)"

Why is the python function more "strict" when encoding the url component?

If I understood it right, brackets are not reserved characters in urls. So I don't understand why they are escaped in the urllib parse function.

Solution

  • As of RFC 3986, brackets are reserved.

    By default, Python will percent-encode every character passed to quote() except for _.-/. However, quote() is tunable. If you want strict RFC 3986 behavior, set safe to '~':

    urllib.parse.quote(string, safe='~')

    If you want to minimally match javascript-on-your-platform's behavior that you showed (you didn't state which parts of which ECMAScript standard it conforms to):

    urllib.parse.quote(string, safe='()')