sqlmap/lib/utils/google.py

133 lines
4.1 KiB
Python
Raw Normal View History

2008-10-15 19:38:22 +04:00
#!/usr/bin/env python
"""
2008-10-15 19:56:32 +04:00
$Id$
2008-10-15 19:38:22 +04:00
Copyright (c) 2006-2010 sqlmap developers (http://sqlmap.sourceforge.net/)
See the file doc/COPYING for copying permission.
2008-10-15 19:38:22 +04:00
"""
import cookielib
import re
import socket
2008-10-15 19:38:22 +04:00
import urllib2
2010-06-02 16:45:40 +04:00
from lib.core.common import getUnicode
from lib.core.convert import htmlunescape
2008-10-15 19:38:22 +04:00
from lib.core.convert import urlencode
from lib.core.data import conf
from lib.core.data import kb
from lib.core.data import logger
2008-10-15 19:38:22 +04:00
from lib.core.exception import sqlmapConnectionException
from lib.request.basic import decodePage
2008-10-15 19:38:22 +04:00
class Google:
"""
This class defines methods used to perform Google dorking (command
line option '-g <google dork>'
"""
def __init__(self, handlers):
2008-10-15 19:38:22 +04:00
self.__matches = []
self.__cj = cookielib.LWPCookieJar()
handlers.append(urllib2.HTTPCookieProcessor(self.__cj))
self.opener = urllib2.build_opener(*handlers)
2008-10-15 19:38:22 +04:00
self.opener.addheaders = conf.httpHeaders
def __parsePage(self, page):
"""
Parse Google dork search results page to get the list of
HTTP addresses
"""
matches = []
regExpr = "li class=\042?g\042?\076.+?a href=\042(http[s]*://.+?)\042\sclass=\042?l\042?"
2008-10-15 19:38:22 +04:00
matches = re.findall(regExpr, page, re.I | re.M)
return matches
def getTargetUrls(self):
"""
This method returns the list of hosts with parameters out of
your Google dork search results
"""
for match in self.__matches:
if re.search("(.*?)\?(.+)", match, re.I):
kb.targetUrls.add(( htmlunescape(match), None, None, None ))
2008-10-15 19:38:22 +04:00
def getCookie(self):
"""
This method is the first to be called when initializing a
Google dorking object through this library. It is used to
retrieve the Google session cookie needed to perform the
further search
"""
try:
conn = self.opener.open("http://www.google.com/ncr")
_ = conn.info()
2008-10-15 19:38:22 +04:00
except urllib2.HTTPError, e:
_ = e.info()
except urllib2.URLError, _:
2008-10-15 19:38:22 +04:00
errMsg = "unable to connect to Google"
raise sqlmapConnectionException, errMsg
def search(self, googleDork):
"""
This method performs the effective search on Google providing
the google dork and the Google session cookie
"""
gpage = conf.googlePage if conf.googlePage > 1 else 1
2010-03-05 17:59:33 +03:00
logger.info("using Google result page #%d" % gpage)
2010-09-27 17:41:18 +04:00
2008-10-15 19:38:22 +04:00
if not googleDork:
return None
url = "http://www.google.com/search?"
url += "q=%s&" % urlencode(googleDork)
url += "num=100&hl=en&safe=off&filter=0&btnG=Search"
url += "&start=%d" % ((gpage-1) * 100)
2008-10-15 19:38:22 +04:00
try:
conn = self.opener.open(url)
requestMsg = "HTTP request:\nGET %s HTTP/1.1\n" % url
requestMsg += "\n".join(["%s: %s" % (header, value) for header, value in conn.headers.items()])
requestMsg += "\n"
logger.log(9, requestMsg)
page = conn.read()
code = conn.code
status = conn.msg
responseHeaders = conn.info()
page = decodePage(page, responseHeaders.get("Content-Encoding"), responseHeaders.get("Content-Type"))
responseMsg = "HTTP response (%s - %d):\n" % (status, code)
2010-09-27 17:41:18 +04:00
if conf.verbose <= 4:
2010-06-02 16:45:40 +04:00
responseMsg += getUnicode(responseHeaders)
elif conf.verbose > 4:
responseMsg += "%s\n%s\n" % (responseHeaders, page)
2010-09-27 17:41:18 +04:00
logger.log(8, responseMsg)
2008-10-15 19:38:22 +04:00
except urllib2.HTTPError, e:
2010-05-16 00:44:08 +04:00
try:
page = e.read()
except socket.timeout:
warnMsg = "connection timed out while trying "
warnMsg += "to get error page information (%d)" % e.code
2010-09-27 17:41:18 +04:00
logger.critical(warnMsg)
2010-05-16 00:44:08 +04:00
return None
except (urllib2.URLError, socket.error, socket.timeout), _:
2008-10-15 19:38:22 +04:00
errMsg = "unable to connect to Google"
raise sqlmapConnectionException, errMsg
self.__matches = self.__parsePage(page)
return self.__matches