ShittyScript.py

import bisect 
import csv
import datetime

survivalFile = open('Survival_lang_no_title.csv')
reader = csv.reader(survivalFile)
reader.next() # pop the first row
minutes = []
langDict = {}


#extract minutes then sort

print 'reading csv', 
readStart = datetime.datetime.now()
for row in reader:
	language = row[1]
	try:
		minute = int(row[3])
	except ValueError:
		continue
	
	try:
		langDict[language].append(minute)
	except KeyError:
		langDict[language] = [minute]
	
	bisect.insort(minutes, minute)  # this keeps shit sorted for the minutes.
	bisect.insort(langDict[language], minute)  # this keeps shit sorted within the dict
print str(datetime.datetime.now()-readStart)

minutes = list(set(minutes))

print len(minutes), max(minutes), min(minutes)
print 'calculating frequencies',
resultDict = {}
freqStart = datetime.datetime.now()
for min in minutes[::30]:
	for language in langDict.keys():
		position = bisect.bisect(langDict[language], min)
		count = len(langDict[language][position:])			
		try:
			resultDict[min][language] = count
		except KeyError:
			resultDict[min] = {language:count}
print str(datetime.datetime.now() - freqStart)

fout = open('niceSurvivalFreqJump30.csv','w')
fout.write('min, lang, unanswered\n')
print 'output!',
outStart = datetime.datetime.now()
for min in resultDict.keys():
	for language in resultDict[min].keys():
		s = '%s, %s, %s\n' % (min, language, resultDict[min][language])
		fout.write(s)
fout.close()
print str(datetime.datetime.now() - outStart)