gistfile1.py

import csv, codecs
import simplejson as json
import operator

arquivo = codecs.open('raw/20130731_Remuneracao.csv', 'rU')
reader = csv.DictReader((line.replace('\0','') for line in arquivo), delimiter='\t')
first_names = { 'TOTAL' : 0}
lista = []
second_loop = []
print 'First loop...'
for i in reader:
    if 1==1:
        nome = i['NOME'].split()
        if not first_names.has_key(nome[0]):
            first_names[nome[0]] = 1
        else:
            first_names[nome[0]] += 1
        first_names['TOTAL'] += 1
        second_loop.append({'nome' : i['NOME'].split(), 'salario' : i['REMUNERA\xc7\xc3O B\xc1SICA BRUTA (R$)']})

names = { 'TOTAL' : { 'count' : 0, 'salario' : 0.0}}
print 'Second loop...'
for i in second_loop:
    i['salario'] = float(i['salario'].replace(',','.'))
    if first_names.has_key(i['nome'][1]) and first_names[i['nome'][1]] > 100:
        if not names.has_key(i['nome'][0] + ' ' + i['nome'][1]):
            names[i['nome'][0] + ' ' + i['nome'][1]] = { 'salario' : i['salario'], 'count' : 1 }
        else:
            names[i['nome'][0] + ' ' + i['nome'][1]]['count'] += 1
            names[i['nome'][0] + ' ' + i['nome'][1]]['salario'] += i['salario']
    elif not names.has_key(i['nome'][0]):
        names[i['nome'][0]] = { 'salario' : i['salario'], 'count' : 1 }
    else:
        names[i['nome'][0]]['count'] += 1
        names[i['nome'][0]]['salario'] += i['salario']
    names['TOTAL']['count'] += 1
    names['TOTAL']['salario'] += i['salario']

print 'Sorting and printing for TODOS...'
fifteen_names = sorted(names.iteritems(), key=lambda x:x[1])
fifteen_names.reverse()

for i, f in enumerate(fifteen_names[1:51]):
    media = "{0:.2f}".format(f[1]['salario']/f[1]['count'])
    print f[0] + ',' + media


for p in names:
    names[p]['media'] = names[p]['salario']/names[p]['count']

riches = sorted(names.iteritems(), key=lambda x:x[1])