class ThreeFourSixFour(object):
def __init__(self):
"""3464网站的爬虫"""
self.getter = GETTER(rtimes=10)
self.cm = ConnMysql()
self.bf = BloomFilter(key='allip')
self.url = "http://www.3464.com/data/Proxy/http/"
def parser(self):
html = self.getter.rget_data(self.url)
html_ele = etree.HTML(html)
tr_list = html_ele.xpath(
'//div[@class="CommonBody"]/table[6]//table//tr')[1:]
sql_list = list()
for tr in tr_list:
try:
ip = tr.xpath('./td[1]/text()')[0]
port = tr.xpath('./td[2]/text()')[0]
except Exception:
continue
# 校验是否已有
if not self.bf.isContains(ip):
sql_list.append(
"""INSERT INTO allip (`ip`, `port`, `type`) VALUES ('{}', '{}', '{}')"""
.format(ip, port, "http"))
self.bf.insert(ip)
else:
pass
for sql in sql_list: # 一次性操作数据库
self.cm.exe(sql)
class YouDaiLi(object):
def __init__(self):
self.getter = GETTER(rtimes=10)
self.cm = ConnMysql()
self.bf = BloomFilter(key='allip')
self.url = "http://www.youdaili.net/Daili/http/"
def parser(self):
url = etree.HTML(self.getter.rget_data(
self.url)).xpath('//div[@class="chunlist"]/ul/li[1]/p/a/@href')[0]
time.sleep(2)
html = self.getter.rget_data(url)
soup = BeautifulSoup(html, 'lxml')
p_tag = soup.find_all('p')
sql_list = list()
for p in p_tag:
ip_list = re.findall(
'(.*?) ———— (.*?) ———— (.*?) ———— (.*?) ',
p.get_text())
if ip_list:
# [('61.130.226.39', '20753', '浙江湖州', 'HTTPS')]
ip = ip_list[0][0]
port = ip_list[0][1]
lx = ip_list[0][3]
if not self.bf.isContains(ip):
sql_list.append(
"""INSERT INTO allip (`ip`, `port`, `type`) VALUES ('{}', '{}', '{}')"""
.format(ip, port, lx))
self.bf.insert(ip)
else:
pass
for sql in sql_list: # 一次性操作数据库
# print(sql)
self.cm.exe(sql)
self.cm.close()
class ThreeThreeSixSix(object):
def __init__(self):
"""3366代理网站的爬虫"""
self.getter = GETTER(rtimes=10)
self.cm = ConnMysql()
self.bf = BloomFilter(key='allip')
self.url = "http://www.ip3366.net/?stype={}&page={}"
def parser(self):
for stype in range(1, 6):
for page in range(1, 11):
url = self.url.format(stype, page)
time.sleep(2)
try:
html = self.getter.rget_data(url)
except Exception:
continue
html_ele = etree.HTML(html)
tr_list = html_ele.xpath('//table/tbody/tr')
sql_list = list()
for tr in tr_list:
ip = tr.xpath('./td[1]/text()')[0]
port = tr.xpath('./td[2]/text()')[0]
lx = tr.xpath('./td[4]/text()')[0]
if not self.bf.isContains(ip):
sql_list.append(
"""INSERT INTO allip (`ip`, `port`, `type`) VALUES ('{}', '{}', '{}')"""
.format(ip, port, lx))
self.bf.insert(ip)
else:
pass
for sql in sql_list: # 一次性操作数据库
self.cm.exe(sql)
class TXTIPPage(object):
def __init__(self):
self.getter = GETTER(rtimes=10)
self.cm = ConnMysql()
self.bf = BloomFilter(key='allip')
self.url_list = [
'https://www.rmccurdy.com/scripts/proxy/output/http/ALL',
'https://www.rmccurdy.com/scripts/proxy/output/socks/ALL',
'https://www.rmccurdy.com/scripts/proxy/proxylist.txt',
'http://www.proxylists.net/http_highanon.txt',
'http://ab57.ru/downloads/proxyold.txt'
]
def run(self):
for url in self.url_list:
data = self.getter.rget_data(url)
ip_list = re.findall('\d+\.\d+\.\d+\.\d+:\d+', data)
temp_l = [[ipport.split(":")[0],
ipport.split(":")[1]] for ipport in ip_list]
sql_list = list()
for temp in temp_l:
ip = temp[0]
port = temp[1]
if not self.bf.isContains(ip):
sql_list.append(
"""INSERT INTO allip (`ip`, `port`, `type`) VALUES ('{}', '{}', '{}')"""
.format(ip, port, 'http'))
self.bf.insert(ip)
else:
pass
for sql in sql_list: # 一次性操作数据库
self.cm.exe(sql)
self.cm.close()
class CoderBusy(object):
def __init__(self):
"""码农很忙代理爬虫"""
self.getter = GETTER(rtimes=10)
self.cm = ConnMysql()
self.bf = BloomFilter(key='allip')
self.url = [
'https://proxy.coderbusy.com/', # 首页
'https://proxy.coderbusy.com/classical/https-ready.aspx?page={}', # HTTPS代理
'https://proxy.coderbusy.com/classical/post-ready.aspx?page={}', # 支持POST的代理
'https://proxy.coderbusy.com/classical/anonymous-type/transparent.aspx?page={}', # 透明代理
'https://proxy.coderbusy.com/classical/anonymous-type/anonymous.aspx?page={}', # 匿名代理
'https://proxy.coderbusy.com/classical/anonymous-type/highanonymous.aspx?page={}', # 高匿代理
]
def parser(self, page_lx):
page = 1
while True:
try:
html = self.getter.rget_data(page_lx.format(page))
except Exception as e:
print("出先错误为{}".format(e))
continue
time.sleep(2) # 睡两秒,防止被干掉
next_page = etree.HTML(html).xpath('//nav[@class="text-center"]/ul/li[@title="下一页"]/a/@href')
soup = BeautifulSoup(html, 'lxml')
proxies_list = soup.find('table', 'table').find_all('tr')
sql_list = list()
for proxy in proxies_list:
temp = proxy.find_all('td')
if temp:
# 获取ip
ip = temp[0].get_text().strip()
# 获取端口
port = int(temp[2].get("data-i"))
for num in ip.split('.'):
port -= int(num)
# 获取类型
if temp[8].find('i'):
lx = 'https'
else:
lx = 'http'
# 校验是否已有
if not self.bf.isContains(ip):
sql_list.append("""INSERT INTO allip (`ip`, `port`, `type`) VALUES ('{}', '{}', '{}')""".format(ip, port, lx))
self.bf.insert(ip)
else:
pass
for sql in sql_list: # 一次性操作数据库
self.cm.exe(sql)
if next_page:
page += 1
else:
break
def run(self):
for page_lx in self.url:
time.sleep(2)
self.parser(page_lx)
class Horocn(object):
def __init__(self):
"""
蜻蜓代理的爬虫
https://proxy.horocn.com/free-proxy.html?page={}
"""
self.getter = GETTER(rtimes=10)
self.cm = ConnMysql()
self.bf = BloomFilter(key='allip')
self.img = ImagePort()
self.port = "12345"
self.url = "https://proxy.horocn.com/free-proxy.html?page={}"
def parser(self):
# page = 1
page = 3000
while True:
try:
html = self.getter.rget_data(self.url.format(page))
except Exception as e:
print("出现错误为{}".format(e))
continue
time.sleep(2) # 睡两秒,防止被干掉
html_ele = etree.HTML(html)
next_page = html_ele.xpath('//ul[@class="pager"]//a[text()="下一页 →"]/@href')[0]
tr_list = html_ele.xpath('//table/tbody/tr')
sql_list = list()
path_list = list()
for tr in tr_list:
ip = tr.xpath('./th[1]/text()')[0]
# 开始保存图片
base_port_image = tr.xpath('./th[2]/img/@src')[0]
photo = base64.b64decode(re.search(r"data:image/jpeg;base64,(.*)", base_port_image).group(1))
path = "./{}.jpg".format(tr_list.index(tr))
path_list.append(path) # 将其放入列表一次性操作
with open(path, "wb") as f:
f.write(photo)
for times in range(10):
try:
self.port = int(self.img.run(path))
except Exception:
continue
else:
break
# 校验是否已有
if not self.bf.isContains(ip):
sql_list.append("""INSERT INTO allip (`ip`, `port`, `type`) VALUES ('{}', '{}', '{}')""".format(ip, self.port, "http"))
self.bf.insert(ip)
else:
pass
for path in path_list:
os.remove(path)
for sql in sql_list: # 一次性操作数据库
self.cm.exe(sql)
if next_page != "javascript:;":
page += 1
else:
break
class MiMi(object):
def __init__(self):
"""秘密代理的IP抓取"""
self.getter = GETTER(rtimes=10)
self.cm = ConnMysql()
self.bf = BloomFilter(key='allip')
self.url = [
"http://www.mimiip.com/gngao/{}", # 高匿代理IP
"http://www.mimiip.com/gnpu/{}", # 普匿代理IP
"http://www.mimiip.com/gntou/{}", # 透明代理IP
"http://www.mimiip.com/hw/{}" # 国外代理IP
]
def parser(self, page_lx):
page = 1
while True:
try:
html = self.getter.rget_data(page_lx.format(page))
except Exception as e:
print("出先错误为{}".format(e))
continue
time.sleep(2) # 睡两秒,防止被干掉
next_page = etree.HTML(html).xpath('//div[@class="pagination"]//*[text()="下一页 ›"]/@href')
soup = BeautifulSoup(html, 'lxml')
proxies_list = soup.find('table', 'list').find_all('tr')
sql_list = list()
for proxy in proxies_list:
temp = proxy.find_all('td')
if temp:
# 获取ip
ip = temp[0].get_text()
# 获取端口
port = temp[1].get_text()
# 获取类型
lx = temp[4].get_text().lower()
# 校验是否已有
if not self.bf.isContains(ip):
sql_list.append("""INSERT INTO allip (`ip`, `port`, `type`) VALUES ('{}', '{}', '{}')""".format(ip, port, lx))
self.bf.insert(ip)
else:
pass
for sql in sql_list: # 一次性操作数据库
self.cm.exe(sql)
if next_page:
page += 1
else:
break
def run(self):
for page_lx in self.url:
time.sleep(2)
self.parser(page_lx)
class KuaiDaiLi(object):
def __init__(self):
"""快代理的爬虫"""
self.getter = GETTER(rtimes=10)
self.cm = ConnMysql()
self.bf = BloomFilter(key='allip')
def parser(self, total_url, xpath_str, format_url):
total = int(
etree.HTML(self.getter.rget_data(total_url)).xpath(xpath_str)[0])
time.sleep(2)
for pageNum in range(1, total):
url = format_url.format(pageNum) # 拼接url
try:
html = self.getter.rget_data(url) # 访问页面
except Exception as e:
print("出先错误为{}".format(e))
continue
time.sleep(2) # 睡两秒,防止被干掉
soup = BeautifulSoup(html, 'lxml')
proxy_list = soup.find(
'table', {
'class': 'table table-bordered table-striped'
}).find('tbody')
sql_list = list()
for proxy in proxy_list.find_all('tr'):
tmp = proxy.find_all('td')
ip = tmp[0].get_text()
port = tmp[1].get_text()
lx = tmp[3].get_text().lower()
if not self.bf.isContains(ip):
sql_list.append(
"""INSERT INTO allip (`ip`, `port`, `type`) VALUES ('{}', '{}', '{}')"""
.format(ip, port, lx))
self.bf.insert(ip)
else:
pass
for sql in sql_list: # 一次性操作数据库
self.cm.exe(sql)
def run(self):
# 获取国内高匿部分
self.parser("https://www.kuaidaili.com/free/inha/1/",
'//div[@id="listnav"]/ul/li[last()-1]/a/text()',
"https://www.kuaidaili.com/free/inha/{}/")
# 获取国内普通部分
time.sleep(3)
self.parser("https://www.kuaidaili.com/free/intr/1/",
'//div[@id="listnav"]/ul/li[last()-1]/a/text()',
"https://www.kuaidaili.com/free/intr/{}/")
self.cm.close() # 关闭数据库连接
class HinkyDink(object):
def __init__(self):
"""Hinky Dink's代理的爬虫"""
self.getter = GETTER(rtimes=10)
self.cm = ConnMysql()
self.bf = BloomFilter(key='allip')
def parser(self, total_url, xpath_str, format_url):
total = int(
etree.HTML(self.getter.rget_data(total_url)).xpath(xpath_str)
[0].strip("[").strip("]"))
time.sleep(2)
for pageNum in range(1, total):
if pageNum == 1:
url = total_url
else:
url = format_url.format(pageNum) # 拼接url
try:
html = self.getter.rget_data(url) # 访问页面
except Exception as e:
print("出先错误为{}".format(e))
continue
time.sleep(2) # 睡两秒,防止被干掉
html_ele = etree.HTML(html)
tr_list = html_ele.xpath(
'//table[2]//tr[2]/td[3]/table//tr/td//table//tr[@class="text"]'
)
sql_list = list()
for tr in tr_list:
ip = tr.xpath('./td[1]/text()')[0]
port = tr.xpath('./td[2]/text()')[0]
if not self.bf.isContains(ip):
sql_list.append(
"""INSERT INTO allip (`ip`, `port`, `type`) VALUES ('{}', '{}', '{}')"""
.format(ip, port, 'http'))
self.bf.insert(ip)
else:
pass
for sql in sql_list: # 一次性操作数据库
self.cm.exe(sql)
def run(self):
self.parser(
"http://www.mrhinkydink.com/proxies.htm", # 第一页url
'//table[2]//tr[2]/td[3]/table//tr/td//table//tr[last()]/td/a[last()]/text()',
"http://www.mrhinkydink.com/proxies{}.htm" # 第二页,开始格式化的url
)
class ThreeSixZero(object):
def __init__(self):
"""360代理的爬虫"""
self.getter = GETTER(rtimes=10)
self.cm = ConnMysql()
self.bf = BloomFilter(key='allip')
self.url = [
"http://www.swei360.com/free/?page={}", # 国内高匿代理
"http://www.swei360.com/free/?stype=2&page={}", # 国内普通代理
"http://www.swei360.com/free/?stype=3&page={}", # 国外高匿代理
"http://www.swei360.com/free/?stype=4&page={}" # 国外普通代理
]
def parser(self, format_url):
for pageNum in range(1, 8): # 他只有7页
url = format_url.format(pageNum) # 拼接url
try:
html = self.getter.rget_data(url) # 访问页面
except Exception as e:
print("出现错误为{}".format(e))
continue
time.sleep(2) # 睡两秒,防止被干掉
html_ele = etree.HTML(html)
tr_list = html_ele.xpath('//table/tbody/tr')
sql_list = list()
for tr in tr_list:
ip = tr.xpath('./td[1]/text()')[0]
port = tr.xpath('./td[2]/text()')[0]
lx = tr.xpath('./td[4]/text()')[0]
if not self.bf.isContains(ip):
sql_list.append("""INSERT INTO allip (`ip`, `port`, `type`) VALUES ('{}', '{}', '{}')""".format(ip, port, lx))
self.bf.insert(ip)
else:
pass
for sql in sql_list: # 一次性操作数据库
self.cm.exe(sql)
def run(self):
for format_url in self.url:
time.sleep(2)
self.parser(format_url)
class SixSixIP(object):
def __init__(self):
"""66ip代理的爬虫"""
self.getter = GETTER(rtimes=10)
self.cm = ConnMysql()
self.bf = BloomFilter(key='allip')
self.url = "http://www.66ip.cn/{}.html"
def run(self):
total = int(
etree.HTML(
self.getter.rget_data("http://www.66ip.cn/1.html")).xpath(
'//div[@id="PageList"]/a[last()-1]/text()')[0])
time.sleep(3)
# for pageNum in range(1, total):
# for pageNum in range(1176, total):
for pageNum in range(1200, total):
url = self.url.format(pageNum) # 拼接url
try:
html = self.getter.rget_data(url) # 访问页面
except Exception as e:
print("出先错误为{}".format(e))
continue
time.sleep(3) # 睡两秒,防止被干掉
soup = BeautifulSoup(html, 'lxml')
proxy_list = soup.find('table', {"border": "2px"})
sql_list = list()
for proxy in proxy_list.find_all('tr')[1:]:
ip = proxy.find_all('td')[0].get_text() # 获取ip
port = proxy.find_all('td')[1].get_text() # 获取端口
if not self.bf.isContains(ip):
sql_list.append(
"""INSERT INTO allip (`ip`, `port`, `type`) VALUES ('{}', '{}', '{}')"""
.format(ip, port, 'http'))
self.bf.insert(ip)
else:
pass
for sql in sql_list: # 一次性操作数据库)
self.cm.exe(sql)
self.cm.close()
class ListProxy(object):
def __init__(self):
self.getter = GETTER(rtimes=10)
self.cm = ConnMysql()
self.bf = BloomFilter(key='allip')
self.url = "https://list.proxylistplus.com/Fresh-HTTP-Proxy-List-{}"
def parser(self):
total = int(etree.HTML(self.getter.rget_data(self.url.format(1))).xpath('//div[@id="page"]/table[3]/tr/td[1]/a[last()]/text()')[0].strip('[').strip(']'))
time.sleep(3)
for pageNum in range(1, total):
url = self.url.format(pageNum) # 拼接url
try:
html = self.getter.rget_data(url) # 访问页面
except Exception as e:
print("出先错误为{}".format(e))
continue
time.sleep(3) # 睡两秒,防止被干掉
soup = BeautifulSoup(html, 'lxml')
proxy_list = soup.find("table", 'bg').find_all('tr')
sql_list = list()
for proxy in proxy_list:
tmp = proxy.find_all("td")
if tmp:
ip = tmp[1].get_text()
port = tmp[2].get_text()
lx = tmp[6].get_text()
if lx == "yes":
lx = 'https'
else:
lx = 'http'
if not self.bf.isContains(ip):
sql_list.append("""INSERT INTO allip (`ip`, `port`, `type`) VALUES ('{}', '{}', '{}')""".format(ip, port, lx))
self.bf.insert(ip)
else:
pass
for sql in sql_list: # 一次性操作数据库
self.cm.exe(sql)
self.cm.close()
class Data5U(object):
def __init__(self):
"""data5u代理的爬虫"""
self.getter = GETTER(rtimes=10)
self.cm = ConnMysql()
self.bf = BloomFilter(key='allip')
self.url = "http://www.data5u.com/free/index.shtml"
def parser(self):
html = self.getter.rget_data(self.url)
soup = BeautifulSoup(html, "lxml")
proxy_list = soup.find_all('ul', {'class': "l2"})
sql_list = list() # 一次性操作数据库
for proxy in proxy_list:
tmp = proxy.find_all('li')
ip = tmp[0].get_text()
port_zimu = list(tmp[1].attrs.values())[0][1]
lx = tmp[3].get_text()
port = self.mk_port(port_zimu)
# 对ip进行布隆去重
if not self.bf.isContains(ip):
sql_list.append(
"""INSERT INTO allip (`ip`, `port`, `type`) VALUES ('{}', '{}', '{}')"""
.format(ip, port, lx))
self.bf.insert(ip)
else:
pass
for sql in sql_list: # 一次性操作数据库
self.cm.exe(sql)
self.cm.close() # 关闭数据库连接
def mk_port(self, port_word):
word = list(port_word)
num_list = []
for item in word:
num = 'ABCDEFGHIZ'.find(item)
num_list.append(str(num))
port = int("".join(num_list)) >> 0x3
return port
class IP181(object):
def __init__(self):
self.getter = GETTER(rtimes=10)
self.cm = ConnMysql()
self.bf = BloomFilter(key='allip')
self.url = 'http://www.ip181.com/'
def parser(self):
js_data = self.getter.rget_data(self.url)
sql_list = list() # 一次性操作数据库
for proxy in json.loads(js_data).get("RESULT"):
ip = proxy.get('ip')
port = proxy.get('port')
if not self.bf.isContains(ip):
sql_list.append(
"""INSERT INTO allip (`ip`, `port`, `type`) VALUES ('{}', '{}', '{}')"""
.format(ip, port, 'http'))
self.bf.insert(ip)
else:
pass
for sql in sql_list: # 一次性操作数据库
self.cm.exe(sql)
self.cm.close() # 关闭数据库连接
class ProxyDB(object):
def __init__(self):
"""ProxyDB代理的爬虫"""
self.getter = GETTER(rtimes=10)
self.cm = ConnMysql()
self.bf = BloomFilter(key='allip')
self.url = "http://proxydb.net/?offset={}"
def parser(self, page_lx):
# page = 0
page = 150
while True:
try:
html = self.getter.rget_data(page_lx.format(page * 15))
except Exception as e:
print("出先错误为{}".format(e))
continue
time.sleep(3) # 睡两秒,防止被干掉
html_ele = etree.HTML(html)
next_page = html_ele.xpath('//nav/ul/li[2]/a/@href')
add_num = html_ele.xpath('//div[@style="display:none"]/@*')[1]
td_list = html_ele.xpath(
'//table[contains(@class, "table")]/tbody/tr/td[1]/script/text()'
)
lx_list = html_ele.xpath(
'//table[contains(@class, "table")]/tbody/tr/td[5]/text()')
sql_list = list()
for td in td_list:
ip_h_reve = re.search(r"'(.*?)'.split", td).group(1) # 提取ip头部
ip_t_b64 = re.search(r"atob\('(.*?)'.replace",
td).group(1) # 提取base64编码部分
p = re.search(r"pp = \((\d+) - \(", td).group(1) # 提取待相加的port
ip, port = self.mk_ip_port(ip_h_reve, ip_t_b64, p, add_num)
lx = lx_list[td_list.index(td)].strip().lower()
if "socket" in lx:
lx = "http"
# 校验是否已有
if not self.bf.isContains(ip):
sql_list.append(
"""INSERT INTO allip (`ip`, `port`, `type`) VALUES ('{}', '{}', '{}')"""
.format(ip, port, lx))
self.bf.insert(ip)
else:
pass
for sql in sql_list: # 一次性操作数据库
self.cm.exe(sql)
if next_page:
page += 1
else:
break
def mk_ip_port(self, ip_h_reve, ip_t_b64, p, add_n):
"""
将网页上抓取下来的参数,直接组织成ip和port
:param ip_h_reve: 待翻转的ip前一部分
:param ip_t_b64: base64加密部分的字母
:param p: 直接抓取到的端口,需要相加
:param add_n: 要相加的值
:return:
"""
l_ip_head = list(ip_h_reve)
l_ip_head.reverse()
ip_head = ""
for char in l_ip_head:
ip_head += char
# 下面这句codecs.getdecoder("unicode_escape")(ip_t_b64)[0]超级重要.取消了转义而使用的
ip_tail = base64.b64decode(
codecs.getdecoder("unicode_escape")(ip_t_b64)[0]).decode()
ip = ip_head + ip_tail
port = int(p) + int(add_n)
return ip, port
def run(self):
self.parser(self.url)
class CoolProxy(object):
def __init__(self):
"""cool-proxy.net代理的爬虫"""
self.getter = GETTER(rtimes=10)
self.cm = ConnMysql()
self.bf = BloomFilter(key='allip')
self.url = "https://www.cool-proxy.net/proxies/http_proxy_list/sort:score/direction:desc/page:{}"
def parser(self):
page = 1
while True:
try:
html = self.getter.rget_data(self.url.format(page))
except Exception as e:
print("出现错误为{}".format(e))
continue
time.sleep(2) # 睡两秒,防止被干掉
next_page = etree.HTML(html).xpath(
'//table//tr[last()]//span[last()]/a')
soup = BeautifulSoup(html, 'lxml')
tr_list = soup.find('table').find_all('tr')
sql_list = list()
for tr in tr_list:
temp = tr.find_all('td')
if temp:
try:
ip = self.mk_ip(
re.search(
r"str_rot13\(\"(.*?)\"\)",
temp[0].find('script').get_text()).group(1))
except Exception:
continue # 里面有混淆的tr
port = temp[1].get_text()
# 校验是否已有
if not self.bf.isContains(ip):
sql_list.append(
"""INSERT INTO allip (`ip`, `port`, `type`) VALUES ('{}', '{}', '{}')"""
.format(ip, port, "http"))
self.bf.insert(ip)
else:
pass
for sql in sql_list: # 一次性操作数据库
self.cm.exe(sql)
if next_page:
page += 1
else:
break
def mk_ip(self, en_ip):
"""
将拿到的-->`ZGH5Ywt5YwVlBF42At==`这种ip解码成可用ip-->159.89.229.66
:param en_ip:爬取到的加密ip
:return:解密后的ip
"""
letter_str = ""
for char in en_ip:
if char in "0123456789==": # 数字和等号用来混淆,直接拼接
letter_str += char
else:
head = ord(char[0]) # 获得该字母的Unicode的编码
tail = 13 if char.lower() < 'n' else -13 # 盐
letter_str += chr(head + tail) # 讲加密后的值解析成字母并拼接
return base64.b64decode(letter_str).decode() # base64解码拼接后的字符串
class XiCiDaiLi(object):
def __init__(self):
"""西刺代理爬虫"""
self.getter = GETTER(rtimes=10)
self.cm = ConnMysql()
self.bf = BloomFilter(key='allip')
def parser(self, total_url, xpath_str, format_url):
total = int(
etree.HTML(self.getter.rget_data(total_url)).xpath(xpath_str)[0])
time.sleep(2)
for pageNum in range(1, total):
url = format_url.format(pageNum) # 拼接url
try:
html = self.getter.rget_data(url) # 访问页面
except Exception as e:
print("出先错误为{}".format(e))
continue
time.sleep(3) # 睡两秒,防止被干掉
soup = BeautifulSoup(html, 'lxml')
proxy_list = soup.find('table', {
'id': 'ip_list'
}).find_all('tr')[1:]
sql_list = list()
for proxy in proxy_list:
tmp = proxy.find_all('td')
ip = tmp[1].get_text()
port = tmp[2].get_text()
lx = tmp[5].get_text().lower()
if "socks" in lx:
lx = "http"
if not self.bf.isContains(ip):
sql_list.append(
"""INSERT INTO allip (`ip`, `port`, `type`) VALUES ('{}', '{}', '{}')"""
.format(ip, port, lx))
self.bf.insert(ip)
else:
pass
for sql in sql_list: # 一次性操作数据库
self.cm.exe(sql)
def run(self):
# 国内高匿代理IP
self.parser("http://www.xicidaili.com/nn/1",
'//div[@id="body"]/div[2]/a[last()-1]/text()',
"http://www.xicidaili.com/nn/{}")
time.sleep(2)
# 国内透明代理IP
self.parser("http://www.xicidaili.com/nt/1",
'//div[@id="body"]/div[2]/a[last()-1]/text()',
"http://www.xicidaili.com/nt/{}")
time.sleep(2)
# HTTPS代理IP
self.parser("http://www.xicidaili.com/wn/1",
'//div[@id="body"]/div[2]/a[last()-1]/text()',
"http://www.xicidaili.com/wn/{}")
time.sleep(2)
# HTTP代理IP
self.parser("http://www.xicidaili.com/wt/1",
'//div[@id="body"]/div[2]/a[last()-1]/text()',
"http://www.xicidaili.com/wt/{}")
self.cm.close()
class GouBanJia(object):
def __init__(self):
"""全网代理IP的爬虫"""
self.getter = GETTER(rtimes=10)
self.cm = ConnMysql()
self.bf = BloomFilter(key='allip')
self.url = "http://www.goubanjia.com/"
def parser(self):
"""
↓此函数借鉴于↓
https://blog.csdn.net/weixin_37586648/article/details/78868015
"""
html = self.getter.rget_data(self.url)
# 解析html
soup = BeautifulSoup(html, "lxml")
# ---------↓自己添加获取类型↓---------
lx_list = list()
ip_port_list = list()
for tr in soup.find_all('tr'):
temp = tr.find_all('td')
if temp:
lx = temp[2].get_text()
lx_list.append(lx)
# ---------↑自己添加获取类型↑---------
# 获取所有的ip的td
td_list = soup.select('td[class="ip"]')
for td in td_list:
# 获取当前td所以的子标签
child_list = td.find_all()
ip_port = ""
for child in child_list:
if 'style' in child.attrs.keys():
if child.attrs['style'].replace(
' ', '') == "display:inline-block;":
if child.string is not None:
ip_port = ip_port + child.string
# 过滤出端口号
elif 'class' in child.attrs.keys():
class_list = child.attrs['class']
if 'port' in class_list:
port = self.mk_port(class_list[1])
# 拼接端口
ip_port = ip_port + ":" + str(port)
else:
if child.string is not None:
ip_port = ip_port + child.string
# 接下来是我自己的
ip_port_list.append(ip_port)
return lx_list, ip_port_list
def run(self):
lx_list, ip_port_list = self.parser()
sql_list = list()
for ip_port in ip_port_list:
lx = lx_list[ip_port_list.index(ip_port)]
ip = ip_port.split(":")[0]
port = ip_port.split(":")[1]
# 对ip进行布隆去重
if not self.bf.isContains(ip):
sql_list.append(
"""INSERT INTO allip (`ip`, `port`, `type`) VALUES ('{}', '{}', '{}')"""
.format(ip, port, lx))
self.bf.insert(ip)
else:
pass
for sql in sql_list: # 一次性操作数据库
self.cm.exe(sql)
self.cm.close() # 关闭数据库连接
def mk_port(self, port_word):
word = list(port_word)
num_list = []
for item in word:
num = 'ABCDEFGHIZ'.find(item)
num_list.append(str(num))
port = int("".join(num_list)) >> 0x3
return port