parsing script for proxy users

'''
creates json and csv of top 10 urls visited by user, country
from sergiomitm.com proxy users.
input: csv of freedom house rankings, geocoded ips from squidanalyzer
'''

import csv, re, json

def main():
    # get the freedom hous3 rankings out of a csv
    with open("fh_rankings.csv", "r") as f:
        reader = csv.reader(f)
        rankings = []
        for row in reader:
            country = row[0]
            ranking = row[1]
            rankings.append([country,ranking])

    # this big loop counts users and requests per country and url
    # end with a list like [num_users, num_requests, country, url, freedom_house_rank]
    with open("geoip_users.csv","r") as f:
        reader = csv.reader(f)
        alldata = []
        for row in reader:
            if row[0]=='Users':continue
            ip = row[0]
            try:
                requests = int(re.search('(^\d+)',row[1]).group(1))
            except:
                requests = 0
            try:
                country = row[7]
            except:
                continue
            url = row[6]
            found = False
            for x in alldata:
                if x['country'] == country and x['top_url']== url:
                    x['users']+=1
                    x['requests']+=requests
                    found = True
                    break
            if not found:
                freedom_rank='Unknown'
                for r in rankings:
                    if r[0]==country: freedom_rank = r[1]
                d = {}
                d['country']=country
                d['requests']=requests
                d['users']=1
                d['freedom_rank']=freedom_rank
                d['top_url']=url
                alldata.append(d)

    # trim to only top 10 urls per country
    sorted_data = sorted(alldata, key=lambda x: x['users'], reverse=True)
    outdata = []
    for s in sorted_data:
        country = s['country']
        count = 0
        for o in outdata:
            if o['country']==country: count +=1
        if count >9: continue
        outdata.append(s)

    with open("country_summary.json", "w") as f:
        json.dump(outdata,f)
    with open("country_summary.csv", "w") as f:
        fn = ['users','requests','country','freedom_rank','top_url']
        wr = csv.DictWriter(f, fn)
        wr.writeheader()
        wr.writerows(outdata)

if __name__ == "__main__":
    main()