languages.py

import csv
from collections import defaultdict

with open('/Users/adam/Downloads/languages-2014.csv', 'r') as f:
    results_2014 = [row for row in csv.reader(f)][1:]

with open('/Users/adam/Downloads/languages-2013.csv', 'r') as f:
    results_2013 = [row for row in csv.reader(f)][1:]

with open('/Users/adam/Downloads/languages-2012.csv', 'r') as f:
    results_2012 = [row for row in csv.reader(f)][1:]

ranks_by_language = defaultdict(list)
installs_by_language = defaultdict(list)

for ii, row in enumerate(results_2014):
    ranks_by_language[row[0]].append(ii + 1)
    installs_by_language[row[0]].append(int(row[1]))

for ii, row in enumerate(results_2013):
    ranks_by_language[row[0]].append(ii + 1)
    installs_by_language[row[0]].append(int(row[1]))

for ii, row in enumerate(results_2012):
    ranks_by_language[row[0]].append(ii + 1)
    installs_by_language[row[0]].append(int(row[1]))

ranks_by_language['Clojure']
installs_by_language['Clojure']
installs_by_language['Java']

ranking_list = [(key, ranks[0]) for key, ranks in ranks_by_language.items()]

out = []

for lang, _ in sorted(ranking_list, key=lambda r: r[1]):
    try:
        rank2014 = ranks_by_language[lang][0]
        rank2013 = ranks_by_language[lang][1]
        rank2012 = ranks_by_language[lang][2]

        if rank2014 <= 20 or rank2013 <= 20 or rank2012 <= 20:
            out.append((
                lang,
                ranks_by_language[lang][0],
                ranks_by_language[lang][1],
                ranks_by_language[lang][2],
                installs_by_language[lang][0],
                installs_by_language[lang][1],
                installs_by_language[lang][2],
            ))
    except IndexError:
        continue


with open('languages-compared.csv', 'w') as f:
    w = csv.writer(f)
    w.writerow(['Language', '2014 Rank', '2013 Rank', '2012 Rank', '2014 # Created', '2013 #', '2012 #'])
    for row in out:
        w.writerow(row)