通过网上爬虫获取了全国所有企业，然后就需要补充企业信息，首先想到的就是企查查，启信宝等专业网站，最终选择了企查查，尝试了多种方法：
1、selenium爬虫，绕过企查查的登录验证，但账号和IP限制太大，最终放弃
2、通过requests直接请求+cookies，遇到了cookie有效期和限制问题
不断的尝试和修改参数，最终发现一种有效方式selenium + wep
只需要IP代理，不需要账号，没有限制，因为是没有登录，拿到的信息有限，能展示的都能获取。
image
一、初始化selenium
sysstr = platform.system()
if(sysstr =="Windows"):
chromedriver_path = os.getcwd() + "\\utools\\chromedriver.exe"
else: #mac
chromedriver_path = os.getcwd() + "/mac_chromedriver"
logger.info("chromedriver_path: %s" %(chromedriver_path,))
default_agent = '--user-agent="Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/65.0.3325.146 Safari/537.36"'
class wap_QiChaCha(object):
def __init__(self, user_agent_header=default_agent, chromedriver_path=chromedriver_path):
self.options = webdriver.ChromeOptions() # 创建一个配置对象
self.options.add_argument('--no-sandbox')
self.options.add_argument('--disable-dev-shm-usage')
self.options.add_argument('--proxy-server=http://47.98.154.206:3008')
self.options.add_argument("--headless") # 开启无界面模式cd
self.options.add_argument("--disable-gpu") # 可选项：禁用gpu，可以解决一些莫名的问题
self.options.add_argument(user_agent_header)
mobileEmulation = {'deviceName': 'iPhone X'}
self.options.add_experimental_option('mobileEmulation', mobileEmulation)
def init(self):
self.driver = webdriver.Chrome(executable_path=chromedriver_path, chrome_options=self.options)
# 打开登录页面
self.driver.get('https://m.qichacha.com/')
self.error_encounter = 0
二、判断公司存在
def search_company(self, company_name):
#time.sleep(0.3)
try:
result = {}
result[COMPANY.NAME] = utils.normalizeCompanyName(company_name)
logger.info("search for: %s" %(company_name,))
'''
self.driver.get('https://m.qichacha.com/')
self.driver.find_element_by_id('searchkey').send_keys(company_name)
# 单击搜索按钮
srh_btn = self.driver.find_element_by_xpath('//*[@id="V3_Index_S"]//span')
srh_btn.click()
'''
self.driver.get('https://m.qcc.com/search?key=%s' %(company_name))
utils.alertWait(WebDriverWait(self.driver, 3).until,
expected_conditions.presence_of_element_located(
(By.XPATH, '//*[contains(@class,"text-danger") or contains(@class,"nodata")]')), 5, 0,
"not found text-danger or nodata")
# 检测企业是不是存在
inc_full = self.driver.find_element_by_xpath('//*[@class="text-danger"]').text
self.error_encounter = 0
if inc_full == "0":
logger.error("company %s not found" %(company_name,))
return None
# 获取首个企业文本
cname = self.driver.find_element_by_xpath('//div[@class="list-item-name"]').text
href = self.driver.find_element_by_xpath('//a[@class="a-decoration"]').get_attribute("href")
# 曾用名
cym = None
try:
stock_or_others = self.driver.find_element_by_xpath('//div[@class="list-item-bottom"]').text
# print(stock_or_others)
# 称呼不同：曾用名 或 历史股东等
if utils.normalizeCompanyName(company_name) in stock_or_others:
company_bottom = stock_or_others.replace("：", ":")
cym = company_bottom.split(":")[1]
except:
# 获取下面显示失败
pass
if utils.normalizeCompanyName(cname) == utils.normalizeCompanyName(company_name) \
or utils.normalizeCompanyName(cym) == utils.normalizeCompanyName(company_name):
result[COMPANY.URL] = href
# time.sleep(0.2)
return self.company_detail(href, result)
except Exception as err:
# self.driver.delete_all_cookies()
logger.error(err)
self.error_encounter = self.error_encounter + 1
if self.error_encounter >= 3:
self.driver.quit()
self.init()
return None
finally:
pass
image
三、获取公司信息
def company_detail(self, href, result):
self.driver.get(href)
utils.alertWait(WebDriverWait(self.driver, 3).until,
expected_conditions.presence_of_element_located((By.XPATH, '//*[@class="company-name"]')), 5, 0,
"not found text-danger")
try:
phone = self.driver.find_element_by_xpath('/html/body/div[1]/div[2]/div[1]/div[3]/a[1]').text
if phone and len(phone) > 0:
result[COMPANY.TEL] = phone.strip()
except Exception as e:
pass
# logger.info("没有手机号")
try:
email = self.driver.find_element_by_xpath('/html/body/div[1]/div[2]/div[1]/div[3]/a[2]').text
if email and len(email) > 0:
result[COMPANY.EMAIL] = email.strip()
except Exception as e:
pass
# logger.info("没有邮箱")
try:
address = self.driver.find_element_by_xpath('/html/body/div[1]/div[2]/div[1]/div[4]').text
if address and len(address) > 0:
result[COMPANY.ADDRESS] = address.strip()
except Exception as e:
pass
# logger.info("没有地址")
try:
infos = self.driver.find_element_by_xpath('//div[@class="basic-wrap"]/table')
# infos = self.driver.find_element_by_xpath('//*[@id="Cominfo"]/table')
except:
return result
result[COMPANY.TAX_LEVEL] = "税务等级&&"
try:
taxcreditlist = self.driver.find_element_by_xpath('//div[@id="taxcreditlist"]').text
info = str(taxcreditlist).replace("\n", "&").strip()
result[COMPANY.TAX_LEVEL] = result[COMPANY.TAX_LEVEL] + info
except:
return result
# 转为etree
data = etree.HTML(infos.get_property("innerHTML"))
data_info = data.xpath('.//tr')
result[COMPANY.BUSINESS] = "工商信息"
for info in data_info:
info_list = info.xpath(".//td//text()")
new_info_list = []
for info in list(info_list):
new_info = str(info).replace("\n", "").strip()
new_info_list.append(new_info)
new_info_list = [i for i in new_info_list if i != '']
self.retrieveInfo(new_info_list, result)
result[COMPANY.BUSINESS] = result[COMPANY.BUSINESS] + " && " + " && ".join(
map(str, new_info_list)) # 以 && 分割 连接 list 内容
return result
image
网友评论