天眼查-获取天眼查id

enterprise_tyc/getTycId.py

+# 根据信用代码获取天眼查id
+import json
+import random
+import sys
+import time
+
+import pymysql
+import requests
+from retry import retry
+
+from base import BaseCore
+import urllib3
+
+urllib3.disable_warnings(urllib3.exceptions.InsecureRequestWarning)
+requests.adapters.DEFAULT_RETRIES = 5
+baseCore = BaseCore.BaseCore()
+log = baseCore.getLogger()
+# headers = {
+#         'Accept': 'application/json, text/plain, */*',
+#         'Accept-Language': 'zh-CN,zh;q=0.9',
+#         'Connection': 'keep-alive',
+#         'Content-Length': '32',
+#         'Content-Type': 'application/json',
+#         'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/119.0.0.0 Safari/537.36',
+#         'version': 'TYC-Web'
+# }
+
+headers = {
+    'Accept-Language': 'zh-CN,zh;q=0.9',
+    'Content-Type': 'application/json',
+    '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',
+    'version': 'TYC-Web'
+}
+
+header = {
+    'Accept': 'application/json, text/plain, */*',
+    'Accept-Encoding': 'gzip, deflate, br, zstd',
+    'Accept-Language': 'zh-CN,zh;q=0.9',
+    'Connection': 'keep-alive',
+    'Content-Length': '93',
+    'Content-Type': 'application/json',
+    'Host': 'capi.tianyancha.com',
+    'Origin': 'https://www.tianyancha.com',
+    'Referer': 'https://www.tianyancha.com/',
+    'Sec-Fetch-Dest': 'empty',
+    'Sec-Fetch-Mode': 'cors',
+    'Sec-Fetch-Site': 'same-site',
+    'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/124.0.0.0 Safari/537.36',
+    'X-AUTH-TOKEN': 'eyJhbGciOiJIUzUxMiJ9.eyJzdWIiOiIxMzYzNjcxMTc0NiIsImlhdCI6MTcxNDk1Njg3MywiZXhwIjoxNzE3NTQ4ODczfQ.qMEvtETT7RS3Rhwq9idu5H2AKMxc2cjtr5bDDW6C6yOFKR-ErgDwT4SOBX9PB2LWDexAG2hNaeAvn6swr-n6VA',
+    'X-TYCID': 'dad485900fcc11ee8c0de34479b5b939',
+    'sec-ch-ua': '"Chromium";v="124", "Google Chrome";v="124", "Not-A.Brand";v="99"',
+    'sec-ch-ua-mobile': '?0',
+    'sec-ch-ua-platform': '"Windows"',
+    'version': 'TYC-Web'
+}
+
+# cnx = pymysql.connect(host='114.116.44.11', user='caiji', password='f7s0&7qqtK', db='dbScore', charset='utf8mb4')
+# cursor= cnx.cursor()
+cnx_ = baseCore.cnx
+cursor_ = baseCore.cursor
+taskType = '天眼查企业id/天眼查'
+
+
+# 根据信用代码获取天眼查id 企业名字等信息
+
+@retry(tries=5, delay=3)
+def getTycIdByXYDM(com_name, s):
+    retData = {'state': False, 'tycData': None, 'reput': True}
+    url = f"https://capi.tianyancha.com/cloud-tempest/search/suggest/v3?_={baseCore.getNowTime(3)}"
+    # url=f"https://capi.tianyancha.com/cloud-tempest/search/suggest/v3"
+    # ip = baseCore.get_proxy()
+    paramJsonData = {'keyword': com_name}
+
+    # response = requests.post(url=url, json=paramJsonData, headers=header, verify=False, proxies=ip)
+    response = s.post(url, json=paramJsonData, headers=headers)
+    time.sleep(random.randint(3, 5))
+    retJsonData = json.loads(response.content.decode('utf-8'))
+    if retJsonData['data'] and retJsonData['state'] == 'ok':
+        pass
+    else:
+        log.error(f"---{com_name}-未查询到该企业---")
+        retData['reput'] = False
+        return retData
+    matchType = retJsonData['data'][0]['matchType']
+    if matchType == '信用代码匹配' or matchType == '公司名称匹配':
+        retData['state'] = True
+        retData['tycData'] = retJsonData['data'][0]
+        response.close()
+        return retData
+    else:
+        log.error(f"{com_name}------{retJsonData}")
+        response.close()
+        return retData
+
+
+# 更新天眼查企业基本信息
+def updateTycInfo():
+    while True:
+        # 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('NewsEnterprise:gnqy_socialCode')
+        # social_code = '9111000066990444XF'
+        # 判断 如果Redis中已经没有数据，则等待
+        if social_code == None:
+            time.sleep(20)
+            continue
+        start = time.time()
+
+        data = baseCore.getInfomation(social_code)
+        if len(data) != 0:
+            pass
+        else:
+            # 数据重新塞入redis
+            baseCore.rePutIntoR('NewsEnterprise:gnqy_socialCode', social_code)
+            continue
+
+        xydm = data[2]
+        tycid = data[11]
+        if tycid == None or tycid == '':
+            try:
+                retData = getTycIdByXYDM(xydm)
+                if retData['tycData'] and retData['reput']:
+                    tycid = retData['id']
+                    # todo:写入数据库
+                    updateSql = f"update EnterpriseInfo set TYCID = '{tycid}' where SocialCode = '{xydm}'"
+                    cursor_.execute(updateSql)
+                    cnx_.commit()
+                elif not retData['tycData'] and retData['reput']:
+                    state = 0
+                    takeTime = baseCore.getTimeCost(start, time.time())
+                    baseCore.recordLog(social_code, taskType, state, takeTime, '', '获取天眼查id失败')
+                    log.info(f'======={social_code}====重新放入redis====')
+                    baseCore.rePutIntoR('NewsEnterprise:gnqy_socialCode', social_code)
+                    continue
+                elif not retData['reput'] and not retData['tycData']:
+                    continue
+            except Exception as e:
+                log.error(e)
+                state = 0
+                takeTime = baseCore.getTimeCost(start, time.time())
+                baseCore.recordLog(social_code, taskType, state, takeTime, '', '获取天眼查id失败')
+                baseCore.rePutIntoR('NewsEnterprise:gnqy_socialCode', social_code)
+                continue
+
+
+# 从数据里获取天眼查id
+def getTycIdByDB(social_code, cursor, start, item, s):
+
+    data = baseCore.getInfomation(social_code)
+    if len(data) != 0:
+        id = data[0]
+        com_name = data[1]
+        xydm = data[2]
+        tycid = data[11]
+        count = data[17]
+    else:
+        # 数据重新塞入redis
+        # log.info(f'数据库中无该企业{social_code}')
+        sql = f"SELECT * FROM sys_base_enterprise WHERE social_credit_code = '{social_code}'"
+        cursor.execute(sql)
+        data = cursor.fetchone()
+        if data:
+            pass
+        else:
+            # 数据库中并没有该企业 需要新增
+            pass
+        id = data[0]
+        com_name = data[3]
+        xydm = data[1]
+        conut = 0
+        # 写入数据库
+        insert = "INSERT INTO EnterpriseInfo(CompanyName, SocialCode) VALUES (%s, %s)"
+        cursor_.execute(insert, (com_name, xydm))
+        cnx_.commit()
+        tycid = ''
+    if tycid == None or tycid == '':
+        try:
+            retData = getTycIdByXYDM(com_name, s)
+            if retData['state']:
+                tycid = retData['tycData']['id']
+                # # todo:写入数据库
+                updateSql = f"update EnterpriseInfo set TYCID = '{tycid}' where SocialCode = '{xydm}'"
+                cursor_.execute(updateSql)
+                cnx_.commit()
+            else:
+                state = 0
+                takeTime = baseCore.getTimeCost(start, time.time())
+                baseCore.recordLog(social_code, taskType, state, takeTime, '', '获取天眼查id失败')
+                log.info(f'======={social_code}====重新放入redis====')
+                baseCore.rePutIntoR('CorPersonEnterpriseError:gnqy_socialCode', item)
+                return None
+        except:
+            state = 0
+            takeTime = baseCore.getTimeCost(start, time.time())
+            baseCore.recordLog(social_code, taskType, state, takeTime, '', '获取天眼查id失败')
+            baseCore.rePutIntoR('CorPersonEnterpriseError:gnqy_socialCode', item)
+            return None
+
+    return tycid
+
+
+if __name__ == '__main__':
+    updateTycInfo()