Analyze the activity of Fedora Hosted projects over a period of time.

fedorahosted-active.py

import operator
import feedparser
import multiprocessing.pool

DAYS_BACK = 498 # 2014-1-1 -> 2015-05-14 https://daycalc.appspot.com/01/01/2015
RSS = 'https://fedorahosted.org/{}/timeline?ticket=on&changeset=on&milestone=on&wiki=on&max=50&format=rss&daysback={}&authors='

active_projects = {}

def handle(project):
    project = project.strip()
    feed = feedparser.parse(RSS.format(project, DAYS_BACK))
    entries = feed['entries']
    if entries:
        active_projects[project] = len(entries)

with file('fedorahosted-projects.txt') as fh:
    projects = fh.readlines()

total = len(projects)

pool = multiprocessing.pool.ThreadPool(20)
pool.map(handle, projects)

num_active = len(active_projects)
print('{} out of {} ({:.2f}%) active Fedora Hosted projects since 2014-01-01'.format(num_active, total, num_active / float(total) * 100))
#for project, entries in sorted(active_projects.items(), key=operator.itemgetter(1)):
#    print('{} {}'.format(entries, project))