Skip to content
Branch: master
Find file Copy path
Fetching contributors…
Cannot retrieve contributors at this time
48 lines (42 sloc) 1.58 KB
import os
import csv
import subprocess
import time
import rxe
from collections import defaultdict
def execute(cmd):
p = subprocess.Popen(cmd, shell=True, stdout=subprocess.PIPE)
out, _ = p.communicate()
return out.strip()
def get_post_ids(filename):
with open(filename) as csv_file:
csv_reader = csv.reader(csv_file, delimiter=',')
return [row[0] for row in csv_reader]
def get_post(post_id):
path = "cache/%s.json" % post_id
if not os.path.isfile(path):
print('Downloading post data into cache...')
url = """""" % post_id
contents = execute("""curl -s %s > %s""" % (url, path))
with open(path, 'r') as f:
def get_commenters(post_id):
post = get_post(post_id)
r ='"author":"').one_or_more(rxe.set_except(['"'])).one('"')
return set([fragment[10:-1] for fragment in r.findall(post)])
post_ids = get_post_ids('top_10k_posts.csv')
user_comments = defaultdict(lambda: [])
for post_id in post_ids:
print('Getting comments for post %s' % post_id)
for user_id in get_commenters(post_id):
with open("post_comments_10k.csv", "w") as f:
num_prolific = 0
for user_id in user_comments:
if len(user_comments[user_id]) >= 3 and len(user_comments[user_id]) <= 50:
num_prolific += 1
for post_id in user_comments[user_id]:
f.write('%s,%s\n' % (post_id, user_id))
print('Qualifying commenters = %d' % num_prolific)
You can’t perform that action at this time.
You signed in with another tab or window. Reload to refresh your session. You signed out in another tab or window. Reload to refresh your session.