List imitator.py

from random import choice
from statistics import mean, stdev
filein = open(r'/home/hellerick/Documents/Data/Geonames/Italian populated places.txt', mode='rU', encoding="utf-8")
# The list should be a text file with a name in each line, and with no blank line in the end
necessary_length = 10000
data = filein.read().split('\n')
# for Python 2:
# data = filein.read().decode('utf8').split('\n')
# encoding="utf-8"

#data = [i.lower() for i in data]
data = [i for i in data]
filein.close()
if data[-1] == '':
    data = data[:-1]
lengths = [len(i) for i in data]
mastermean = mean (lengths)
masterdev = stdev (lengths)
print (mastermean, masterdev)
stat = dict([])
for w in data:
    for i in range(-1,len(w)+1):
        if i == -1:
            prv = '\n'
            cur = '\n'
            nxt = w[0]
        elif i == len(w):
            prv = w[-1]
            cur = '\n'
            nxt = '\n'
        else:
            cur = w[i]
            if i == 0:
                prv = '\n'
            else:
                prv = w[i-1]
            if i == len(w)-1:
                nxt = '\n'
            else:
                nxt = w[i+1]
        if prv+cur in stat:
            if nxt in stat[prv+cur]:
                stat[prv+cur][nxt] += 1
            else:
                stat[prv+cur][nxt] = 1
        else:
            stat[prv+cur] = {nxt:1}
#for i in stat:
#    print (i, stat[i])
result = set()
while len(result)<necessary_length:
    s = '\n\n'
    genname = s
    while True:
        nextchars = []
        for i in stat[s]:
            nextchars = nextchars + [i]*stat[s][i]
        chosenchar = choice(nextchars)
        genname = genname + chosenchar
        s = s[1] + chosenchar
        #print ()
        if s == '\n\n':
            break
    #print (',',genname[2:-2],',')
    genname = genname[2:-2]
    if len(result) < 10:
        result= result | {genname}
    else:
        curlen = [len(i) for i in result]
        curdev = stdev(curlen)
        curmean = mean(curlen)
        attres = result | {genname}
        attlen = [len(i) for i in attres]
        attdev = stdev(attlen)
        attmean = mean(attlen)
        if ((abs(curmean-mastermean)-abs(attmean-mastermean)) + (abs(curdev-masterdev)-abs(attdev-masterdev)) > 0 and (abs(attmean-mastermean)<0.5 and abs(attdev-masterdev)<0.1)==False ) or (abs(attmean-mastermean)<0.5 and abs(attdev-masterdev)<0.1):
            result = attres
            print (genname)

curlen = [len(i) for i in result]
curdev = stdev(curlen)
curmean = mean(curlen)
print (curmean, curdev)

with open('/home/hellerick/Documents/Data/Geonames/Italian populated places - Imitation.txt', mode='wt', encoding="utf-8") as f:
    for r in result:
        f.write(r+'\n')