From e7056b3e9ce8d733348259d6e33dec36521f3984 Mon Sep 17 00:00:00 2001 From: Cédric Bonhomme Date: Sat, 12 Apr 2014 16:47:52 +0200 Subject: Improvements of the feedgetter module. --- pyaggr3g470r/utils.py | 20 +++++++++++++++++++- 1 file changed, 19 insertions(+), 1 deletion(-) (limited to 'pyaggr3g470r/utils.py') diff --git a/pyaggr3g470r/utils.py b/pyaggr3g470r/utils.py index ab58e689..1a8a8387 100755 --- a/pyaggr3g470r/utils.py +++ b/pyaggr3g470r/utils.py @@ -40,7 +40,8 @@ import glob import opml import operator import calendar - +from urllib import urlencode +from urlparse import urlparse, parse_qs, urlunparse from BeautifulSoup import BeautifulSoup from collections import Counter @@ -117,6 +118,23 @@ def import_opml(email, opml_file): db.session.commit() return nb +def clean_url(url): + """ + Remove utm_* parameters + """ + parsed_url = urlparse(url) + qd = parse_qs(parsed_url.query, keep_blank_values=True) + filtered = dict((k, v) for k, v in qd.iteritems() if not k.startswith('utm_')) + nice_url = urlunparse([ + parsed_url.scheme, + parsed_url.netloc, + parsed_url.path, + parsed_url.params, + urlencode(filtered, doseq=True), + parsed_url.fragment + ]) + return nice_url + def open_url(url): """ Open an URL with the proxy and the user-agent -- cgit