提交 c442ebb4 作者: 薛凌堃

天眼查核心人员更新

上级 0fe52c2b
...@@ -45,6 +45,9 @@ def get_html(tycid, s, headers): ...@@ -45,6 +45,9 @@ def get_html(tycid, s, headers):
# div_part.find('div', class_='dimHeader_root__XTCLe') # div_part.find('div', class_='dimHeader_root__XTCLe')
except: except:
return -1 return -1
if div_part is None:
return -2
else:
try: try:
tmp_field = div_part.find('div', class_='dim-tab-root').find('span').text tmp_field = div_part.find('div', class_='dim-tab-root').find('span').text
if '最新公示' in tmp_field: if '最新公示' in tmp_field:
...@@ -64,7 +67,10 @@ def get_page(url, s, headers): ...@@ -64,7 +67,10 @@ def get_page(url, s, headers):
if res.status_code != 200: if res.status_code != 200:
raise raise
data_page = res.json() data_page = res.json()
try:
total_page_ = data_page['data']['total'] total_page_ = data_page['data']['total']
except:
raise
return total_page_ return total_page_
...@@ -77,7 +83,7 @@ def doJob(): ...@@ -77,7 +83,7 @@ def doJob():
'Accept-Encoding': 'gzip, deflate, br', 'Accept-Encoding': 'gzip, deflate, br',
'Accept-Language': 'zh-CN,zh;q=0.9', 'Accept-Language': 'zh-CN,zh;q=0.9',
'Cache-Control': 'max-age=0', 'Cache-Control': 'max-age=0',
'Connection': 'keep-alive', # 'Connection': 'keep-alive',
'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/120.0.0.0 Safari/537.36', 'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/120.0.0.0 Safari/537.36',
'version': 'TYC-Web' 'version': 'TYC-Web'
} }
...@@ -90,7 +96,7 @@ def doJob(): ...@@ -90,7 +96,7 @@ def doJob():
# 根据从Redis中拿到的社会信用代码,在数据库中获取对应基本信息 # 根据从Redis中拿到的社会信用代码,在数据库中获取对应基本信息
# social_code = baseCore.redicPullData('CorPersonEnterprise:gnqy_socialCode') # social_code = baseCore.redicPullData('CorPersonEnterprise:gnqy_socialCode')
# 判断 如果Redis中已经没有数据,则等待 # 判断 如果Redis中已经没有数据,则等待
social_code = '91440300MA5EU1QM0T' social_code = '91110108780992804C'
if social_code == None: if social_code == None:
time.sleep(20) time.sleep(20)
continue continue
...@@ -163,6 +169,11 @@ def doJob(): ...@@ -163,6 +169,11 @@ def doJob():
log.info(f"{id}---{xydm}----{tycid}----请求失败----重新放入redis") log.info(f"{id}---{xydm}----{tycid}----请求失败----重新放入redis")
time.sleep(2) time.sleep(2)
continue continue
elif charge == -2:
# 该企业没有人员信息
log.info(f"{id}---{xydm}----{tycid}----没有核心人员")
continue
elif charge == 0: elif charge == 0:
log.info(f"{id}---{xydm}----{tycid}----没有最新公示") log.info(f"{id}---{xydm}----{tycid}----没有最新公示")
url1 = f'https://capi.tianyancha.com/cloud-company-background/company/dim/staff?_={t}&gid={tycid}&pageSize=20&pageNum=1' url1 = f'https://capi.tianyancha.com/cloud-company-background/company/dim/staff?_={t}&gid={tycid}&pageSize=20&pageNum=1'
...@@ -240,6 +251,8 @@ def doJob(): ...@@ -240,6 +251,8 @@ def doJob():
pass pass
else: else:
log.info(f'{id}---{xydm}----{tycid}----没有高管信息') log.info(f'{id}---{xydm}----{tycid}----没有高管信息')
# todo: 关闭连接
res.close()
if flag == 1: if flag == 1:
for one_info in list_all: for one_info in list_all:
name = one_info['name'] name = one_info['name']
......
Markdown 格式
0%
您添加了 0 到此讨论。请谨慎行事。
请先完成此评论的编辑!
注册 或者 后发表评论