Last active
December 29, 2015 14:09
-
-
Save egcodes/7682311 to your computer and use it in GitHub Desktop.
get hacked tr(turkish) pages from zone-h.org
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
from BeautifulSoup import BeautifulSoup | |
import urllib2 | |
from datetime import datetime | |
import time,os | |
pageNo = 1 | |
while pageNo < 10: | |
print "PageNo: %d"%pageNo | |
page = urllib2.urlopen("http://www.zone-h.org/archive/page=%d"%pageNo, timeout=10).read() | |
soup = BeautifulSoup(page) | |
today = str(datetime.now()) | |
today = today[:10] | |
table = soup.find("table") | |
rows = table.findAll('tr') | |
for i, tr in enumerate(rows): | |
cols = tr.findAll('td') | |
for j,td in enumerate(cols): | |
try: | |
text = ''.join(td.find(text=True)) | |
except: | |
if str(td).find("tr.png") != -1: | |
ls = [] | |
for item in tr: | |
try: | |
txt = ''.join(item.find(text=True)) | |
if txt == "mirror": | |
ls.append(str(item)[str(item).find('href') + 1:str(item).rfind('mir') - 1]) | |
else: | |
ls.append(txt.strip()) | |
except Exception, error: | |
pass | |
hack = "Time: ",today + " " + ls[0] + " - " + "Hacked By: ",ls[1] | |
hack2 = "" | |
if ls[2].find('.') != -1: | |
hack2 = " - ","Site: ",ls[2] + " - " + "Os: ", ls[3] + "\n" | |
elif ls[3].find('.') != -1: | |
hack2 = " - ","Site: ",ls[3] + " - " + "Os: ", ls[4] + "\n" | |
elif ls[4].find('.') != -1: | |
hack2 = " - ","Site: ",ls[4] + " - " + "Os: ", ls[5] + "\n" | |
else: | |
hack2 = " - ","Site: ",ls[5] + " - " + "Os: ", ls[6] + "\n" | |
print hack+hack2 | |
pageNo += 1 |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment