def crack_sougou(self, url):
log('------开始处理未成功的URL：{}'.format(url))
if re.search('weixin\.sogou\.com', url):
log('------开始处理搜狗验证码------')
self.driver.get(url)
time.sleep(2)
if '搜公众号' in self.driver.page_source:
for i in range(30):
self.driver.get(url)
log('浏览器页面正常')
if '搜公众号' not in self.driver.page_source:
break
try:
img = self.wait.until(EC.presence_of_element_located((By.ID, 'seccodeImage')))
log('------出现验证码页面------')
location = img.location
size = img.size
left = location['x']
top = location['y']
right = location['x'] + size['width']
bottom = location['y'] + size['height']
screenshot = self.driver.get_screenshot_as_png()
screenshot = Image.open(BytesIO(screenshot))
captcha = screenshot.crop((left, top, right, bottom))
captcha_path = os.path.join(IMAGE_DIR, CAPTCHA_NAME)
captcha.save(captcha_path)
with open(captcha_path, "rb") as f:
filebytes = f.read()
captch_input = captch_upload_image(filebytes)
log('------验证码：{}------'.format(captch_input))
if captch_input:
input_text = self.wait.until(EC.presence_of_element_located((By.ID, 'seccodeInput')))
input_text.clear()
input_text.send_keys(captch_input)
submit = self.wait.until(EC.element_to_be_clickable((By.ID, 'submit')))
submit.click()
time.sleep(2)
try:
if '搜公众号' not in self.driver.page_source:
log('验证失败')
return
log('------验证码正确------')
except:
log('--22222222----验证码输入错误------')
except Exception as e:
log('------未跳转到验证码页面，跳转到首页，忽略------')
elif re.search('mp\.weixin\.qq\.com', url):
log('------开始处理微信验证码------')
cert = random.random()
image_url = 'https://mp.weixin.qq.com/mp/verifycode?cert={}'.format(cert)
respones = self.s.get(image_url, cookies=self.cookies)
captch_input = captch_upload_image(respones.content)
log('------验证码：{}------'.format(captch_input))
data = {
'cert': cert,
'input': captch_input
}
self.s.post(image_url, cookies=self.cookies, data=data)
log('------cookies已更新------')
def crack_sougou(self, url):
print('------开始处理未成功的URL：{}'.format(url))
# if 'weixin\.sogou\.com' in url:
print('------开始处理搜狗验证码------')
chrome_options = webdriver.ChromeOptions()
# chrome_options.add_argument('--headless')
browser = webdriver.Chrome(chrome_options=chrome_options)
browser.get(url)
time.sleep(2)
try:
wait = WebDriverWait(browser, 10)
img = wait.until(EC.presence_of_element_located((By.ID, 'seccodeImage')))
print('------出现验证码页面------')
location = img.location
size = img.size
left = location['x']
top = location['y']
right = location['x'] + size['width']
bottom = location['y'] + size['height']
screenshot = browser.get_screenshot_as_png()
screenshot = Image.open(BytesIO(screenshot))
captcha = screenshot.crop((left, top, right, bottom))
captcha_path = os.path.join(IMAGE_DIR, CAPTCHA_NAME)
captcha.save(captcha_path)
with open(captcha_path, "rb") as f:
filebytes = f.read()
captch_input = captch_upload_image(filebytes)
print('------验证码：{}------'.format(captch_input))
if captch_input:
input_text = wait.until(EC.presence_of_element_located((By.ID, 'seccodeInput')))
input_text.clear()
input_text.send_keys(captch_input)
submit = wait.until(EC.element_to_be_clickable((By.ID, 'submit')))
submit.click()
try:
print('------输入验证码------')
error_tips = wait.until(EC.presence_of_element_located((By.ID, 'error-tips'))).text
if len(error_tips):
print('------验证码输入错误------')
return
wait.until(EC.presence_of_element_located((By.CLASS_NAME, 'login-info')))
print('------验证码正确------')
cookies = browser.get_cookies()
new_cookie = {}
for items in cookies:
new_cookie[items.get('name')] = items.get('value')
self.cookies = new_cookie
print('------cookies已更新------')
return new_cookie
except:
print('------验证码输入错误------')
except:
print('------未跳转到验证码页面，跳转到首页，忽略------')
def account_homepage(self):
# 搜索并进入公众号主页
search_url = self.url.format(self.name)
resp_search = self.s.get(search_url, headers=self.headers, cookies=self.cookies)
if 'class="b404-box" id="noresult_part1_container"' in resp_search.text:
log("找不到该公众号: {}".format(self.name))
return
e = pq(resp_search.text)
if e(".tit").eq(0).text() == self.name:
account_link = e(".tit").find('a').attr('href')
elif len(e(".tit").eq(0).text()) > 1:
log("不能匹配正确的公众号: {}".format(self.name))
return
else:
# 处理验证码
self.crack_sougou(search_url)
print("验证完毕")
# 被跳过的公众号要不要抓取 大概 4次
return
account_match = re.search(r'微信号：\w*', e.text())
account_search = account_match.group().replace('微信号：', '') if account_match else ''
homepage = self.s.get(account_link, cookies=self.cookies)
if '<title>请输入验证码 </title>' in homepage.text:
print("出现验码")
from verification_code import captch_upload_image
print('------开始处理微信验证码------')
cert = random.random()
image_url = 'https://mp.weixin.qq.com/mp/verifycode?cert={}'.format(cert)
respones = self.s.get(image_url, )
captch_input = captch_upload_image(respones.content)
print('------验证码：{}------'.format(captch_input))
data = {
'cert': cert,
'input': captch_input
}
respones = self.s.post(image_url, data=data, cookies=self.cookies)
cookies = requests.utils.dict_from_cookiejar(respones.cookies)
print('adffa', cookies)
homepage = self.s.get(account_link, cookies=self.cookies)
print('破解验证码之后')
account = pq(homepage.text)('.profile_account').text().replace('微信号: ', '')
# 搜索页面有account，公众号主页有account，确保找到account
return homepage.text, account or account_search
def crack_sougou(self, url):
log.info('------开始处理未成功的URL：{}'.format(url))
if re.search('weixin\.sogou\.com', url):
log.info('------开始处理搜狗验证码------')
self.driver.get(url)
time.sleep(2)
if '搜公众号' in self.driver.page_source:
log.info('浏览器页面正常' + '直接返回')
return
try:
img = self.wait.until(
EC.presence_of_element_located((By.ID, 'seccodeImage')))
log.info('------出现验证码页面------')
location = img.location
size = img.size
left = location['x']
top = location['y']
right = location['x'] + size['width']
bottom = location['y'] + size['height']
screenshot = self.driver.get_screenshot_as_png()
screenshot = Image.open(BytesIO(screenshot))
captcha = screenshot.crop((left, top, right, bottom))
captcha_path = get_captcha_path()
captcha.save(captcha_path)
captcha_name = os.path.basename(captcha_path)
try:
captch_input = ''
files = {
'img': (captcha_name, open(captcha_path,
'rb'), 'image/png', {})
}
res = requests.post(url=GETCAPTCHA_URL, files=files)
res = res.json()
if res.get('Success'):
captch_input = res.get('Captcha')
except Exception as e:
log.info('搜狗验证码获取失败'.format(e))
with open(captcha_path, "rb") as f:
filebytes = f.read()
captch_input = captch_upload_image(filebytes)
# log.info('------验证码：{}------'.format(captch_input))
log.info('------验证码：{}------'.format(captch_input))
if captch_input:
input_text = self.wait.until(
EC.presence_of_element_located(
(By.ID, 'seccodeInput')))
input_text.clear()
input_text.send_keys(captch_input)
submit = self.wait.until(
EC.element_to_be_clickable((By.ID, 'submit')))
submit.click()
time.sleep(2)
try:
if '搜公众号' not in self.driver.page_source:
log.info('验证失败')
return
log.info('------验证码正确------')
except Exception as e:
log.info('--22222222----验证码输入错误------ {}'.format(e))
except Exception as e:
log.info('------未跳转到验证码页面，跳转到首页，忽略------ {}'.format(e))
elif re.search('mp\.weixin\.qq\.com', url):
log.info('------开始处理微信验证码------')
cert = random.random()
image_url = 'https://mp.weixin.qq.com/mp/verifycode?cert={}'.format(
cert)
respones = self.s.get(image_url, cookies=self.cookies)
captch_input = captch_upload_image(respones.content)
log.info('------验证码：{}------'.format(captch_input))
data = {'cert': cert, 'input': captch_input}
r = self.s.post(image_url, cookies=self.cookies, data=data)
log.info('------cookies已更新------{}'.format(r.status_code))
def account_homepage(self):
# 搜索并进入公众号主页
count = 0
while True:
search_url = self.url.format(self.name)
resp_search = self.s.get(search_url, headers=self.headers, cookies=self.cookies)
if '相关的官方认证订阅号' in resp_search.text:
log("找不到该公众号: {}".format(self.name))
return
e = pq(resp_search.text)
if self.name in e(".tit").eq(0).text():
account_link = e(".tit").find('a').attr('href')
elif len(e(".tit").eq(0).text()) > 1:
log("不能匹配正确的公众号: {}".format(self.name))
return
else:
log(search_url)
# log(resp_search.text)
log('验证之前的cookie', self.cookies)
try_count = 0
while True:
try_count += 1
self.crack_sougou(search_url)
if '搜公众号' in self.driver.page_source:
log('------cookies更新------')
cookies = self.driver.get_cookies()
new_cookie = {}
for items in cookies:
new_cookie[items.get('name')] = items.get('value')
self.cookies = new_cookie
log('------cookies已更新------', self.cookies)
break
elif try_count > 6:
log("浏览器验证失败")
break
log("验证完毕")
time.sleep(2)
# 被跳过的公众号要不要抓取 大概 4次
continue
account_match = re.search(r'微信号：\w*', e.text())
account_search = account_match.group().replace('微信号：', '') if account_match else ''
homepage = self.s.get(account_link, cookies=self.cookies)
if '<title>请输入验证码 </title>' in homepage.text:
print("出现验码")
print('------开始处理微信验证码------')
cert = random.random()
image_url = 'https://mp.weixin.qq.com/mp/verifycode?cert={}'.format(cert)
respones = self.s.get(image_url, )
captch_input = captch_upload_image(respones.content)
print('------验证码：{}------'.format(captch_input))
data = {
'cert': cert,
'input': captch_input
}
respones = self.s.post(image_url, data=data, cookies=self.cookies)
cookies = requests.utils.dict_from_cookiejar(respones.cookies)
print('adffa', cookies)
homepage = self.s.get(account_link, cookies=self.cookies)
print('破解验证码之后')
account = pq(homepage.text)('.profile_account').text().replace('微信号: ', '')
# 搜索页面有account，公众号主页有account，确保找到account
return homepage.text, account or account_search
bottom = location['y'] + size['height']
screenshot = browser.get_screenshot_as_png()
screenshot = Image.open(BytesIO(screenshot))
captcha = screenshot.crop((left, top, right, bottom))
BASE_DIR = os.path.abspath(os.path.dirname(__file__))
IMAGE_DIR = os.path.join(BASE_DIR, 'images')
CAPTCHA_NAME = 'captcha.png'
captcha_path = os.path.join(IMAGE_DIR, CAPTCHA_NAME)
captcha.save(captcha_path)
with open(captcha_path, "rb") as f:
filebytes = f.read()
from verification_code import captch_upload_image
captch_input = captch_upload_image(filebytes)
print('------验证码：{}------'.format(captch_input))
if captch_input:
input_text = wait.until(
EC.presence_of_element_located((By.ID, 'seccodeInput')))
input_text.clear()
input_text.send_keys(captch_input)
submit = wait.until(EC.element_to_be_clickable((By.ID, 'submit')))
submit.click()
try:
print('------输入验证码------')
error_tips = wait.until(
EC.presence_of_element_located((By.ID, 'error-tips'))).text
if len(error_tips):
print('------验证码输入错误------')
