quoteall

__author__ = 'rorke'
#simple program which allows for optimal shitposting via
#a print out of all post numbers itt

import urllib.request
import re # normies out

def read_by_tokens(fileobj):
    for line in fileobj:
        for token in line.split():
            yield token

filename = input("enter text file name: ")
url = input("enter url:")

page = urllib.request.urlopen(url)

text_page = str(page.read(), encoding="utf-8")#convert bytes to string

text_file = open(filename, 'r+')
text_file.write(text_page)

replies_list = []

with open(filename) as f:
    for token in read_by_tokens(f):
        if re.search("javascript:quote", token):
            token = token[24:]# delete first 24 characters
            token = token[:8]# deletes all bar first 8 characters
            token = ">>" + token + "\n"
            replies_list.append(token)
            print(token)

text_file = open(filename, 'w')

replies = list(set(replies_list))#convert to set then back to list to remove duplicates


for you in replies:
    text_file.write(you)
text_file.close()