btc.py

# Standard modules, wget, Selenium
import random
import os
import sys
import requests
import re
from bs4 import BeautifulSoup
from time import sleep

global goods
goods = 0
global wallets
wallets = 1

def getBlockChain(t1):
    global wallets
    global goods
    # From directory.io, query blockchain
    target = t1[0]
    header = {"User-Agent":"Mozilla/5.0 (Macintosh; Intel Mac OS X 10_11_6) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/53.0.2785.143 Safari/537.36", "Cache-Control":"no-cache"}
    response = requests.get(target,headers=header)
    html = response.content
    soup = BeautifulSoup(html,'lxml')
    transactions = soup.findAll('td', {'id':re.compile('final_balance')})
    c = str(transactions[0])[57:]
    c = c[:-23]
    coins = c.split('">')
    satoshis = int(coins[0])
    btcs = float(coins[1])
    if satoshis == 0:
        print "Wallet %s is empty. Checked %s wallets in %s pages (%s good)" % (t1[1], wallets, t1[3], goods)
    elif satoshis > 0:
        loot = "THERE'S LOOT HERE!\nPage: %s\nWallet: %s\nBTC(Satoshis): %s(%s)\nChecked %s wallets, %s good\n" % (t1[2], t1[1], btcs, satoshis, wallets, goods)
        print loot
        print "Writing to loot.txt..."
        lootxt = open("lootxt", 'a')
        lootxt.write(loot)
        lootxt.close()
    wallets = wallets + 1

def getpagedata(index,iters):
    print "Getting page %s! (%s pages so far!)" % (index,iters)
    target = "http://directory.io/"
    target = target + str(index)
    header = {"User-Agent":"Mozilla/5.0 (Macintosh; Intel Mac OS X 10_11_6) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/53.0.2785.143 Safari/537.36"}
    response = requests.get(target,headers=header)
    html = response.content
    soup = BeautifulSoup(html, "lxml")
    links = soup.findAll('a', attrs={'href': re.compile("^https://blockchain.info/address/")})

    array = []

    for i in links:
        array.append(i)

    while array:
        string = str(array[0])
        temp = string.split('<a href="')
        temp.pop(0)
        temp = temp[0].split('">')
        temp.append(index)
        temp.append(iters)
        temp[1] = temp[1][:-4]
        getBlockChain(temp)
        array.pop(0)
        wait = random.uniform(0.0,1.0)
        sleep(wait)

    getpagedata(int(index)+1,int(iters)+1)

index = raw_input("What page do we start at?\n>")
getpagedata(index,1)