mirror of
https://github.com/laramies/theHarvester.git
synced 2024-09-23 08:46:32 +08:00
89 lines
3.1 KiB
Python
89 lines
3.1 KiB
Python
import string
|
|
import http.client
|
|
import sys
|
|
import myparser
|
|
import re
|
|
import time
|
|
import requests
|
|
|
|
class search_bing:
|
|
|
|
def __init__(self, word, limit, start):
|
|
self.word = word.replace(' ', '%20')
|
|
self.results = ""
|
|
self.totalresults = ""
|
|
self.server = "www.bing.com"
|
|
self.apiserver = "api.search.live.net"
|
|
self.hostname = "www.bing.com"
|
|
self.userAgent = "(Mozilla/5.0 (Windows; U; Windows NT 6.0;en-US; rv:1.9.2) Gecko/20100115 Firefox/3.6"
|
|
self.quantity = "50"
|
|
self.limit = int(limit)
|
|
self.bingApi = ""
|
|
self.counter = start
|
|
|
|
def do_search(self):
|
|
headers = {
|
|
'Host': self.hostname,
|
|
'Cookie':'SRCHHPGUSR=ADLT=DEMOTE&NRSLT=50',
|
|
'Accept-Language': 'en-us,en',
|
|
'User-agent': self.userAgent
|
|
}
|
|
h = requests.get(url=('http://'+self.server + "/search?q=%40" + self.word + "&count=50&first=" + str(self.counter)),headers=headers)
|
|
self.results = str(h.content)
|
|
self.totalresults += self.results
|
|
|
|
def do_search_api(self):
|
|
h = http.client.HTTPConnection(self.apiserver)
|
|
h.putrequest('GET', "/xml.aspx?Appid=" + self.bingApi + "&query=%40" +
|
|
self.word + "&sources=web&web.count=40&web.offset=" + str(self.counter))
|
|
h.putheader('Host', "api.search.live.net")
|
|
h.putheader('User-agent', self.userAgent)
|
|
h.endheaders()
|
|
self.results = str(h.getresponse().read())
|
|
self.totalresults += self.results
|
|
|
|
def do_search_vhost(self):
|
|
headers = {
|
|
'Host': self.hostname,
|
|
'Cookie':'mkt=en-US;ui=en-US;SRCHHPGUSR=NEWWND=0&ADLT=DEMOTE&NRSLT=50',
|
|
'Accept-Language': 'en-us,en',
|
|
'User-agent':self.userAgent
|
|
}
|
|
url = 'http://' + self.server + "/search?q=ip:" + self.word + "&go=&count=50&FORM=QBHL&qs=n&first=" + str(self.counter)
|
|
h = requests.get(url=url,headers=headers)
|
|
self.results = str(h.content)
|
|
self.totalresults += self.results
|
|
|
|
def get_emails(self):
|
|
rawres = myparser.parser(self.totalresults, self.word)
|
|
return rawres.emails()
|
|
|
|
def get_hostnames(self):
|
|
rawres = myparser.parser(self.totalresults, self.word)
|
|
return rawres.hostnames()
|
|
|
|
def get_allhostnames(self):
|
|
rawres = myparser.parser(self.totalresults, self.word)
|
|
return rawres.hostnames_all()
|
|
|
|
def process(self, api):
|
|
if api == "yes":
|
|
if self.bingApi == "":
|
|
print("Please insert your API key in the discovery/bingsearch.py")
|
|
sys.exit()
|
|
while (self.counter < self.limit):
|
|
if api == "yes":
|
|
self.do_search_api()
|
|
time.sleep(0.3)
|
|
else:
|
|
self.do_search()
|
|
time.sleep(1)
|
|
self.counter += 50
|
|
print("\tSearching " + str(self.counter) + " results...")
|
|
|
|
def process_vhost(self):
|
|
# Maybe it is good to use other limit for this.
|
|
while (self.counter < self.limit):
|
|
self.do_search_vhost()
|
|
self.counter += 50
|