雪球网年报脚本维护

comData/annualReport/雪球网-年报.py

-# -*- coding: utf-8 -*-
+# -*- coding: utf-8 -*-
@@ -21,19 +21,20 @@ from selenium import webdriver
 
 # chromedriver = "D:/chrome/chromedriver.exe"
 # browser = webdriver.Chrome(chromedriver)
+def createDriver():
+    opt = webdriver.ChromeOptions()
+    opt.add_argument(
+            'user-agent=Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/86.0.4240.198 Safari/537.36')
 
-opt = webdriver.ChromeOptions()
-opt.add_argument(
-        'user-agent=Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/86.0.4240.198 Safari/537.36')
-
-opt.add_argument("--ignore-certificate-errors")
-opt.add_argument("--ignore-ssl-errors")
-opt.add_experimental_option("excludeSwitches", ["enable-automation"])
-opt.add_experimental_option('excludeSwitches', ['enable-logging'])
-opt.add_experimental_option('useAutomationExtension', False)
-opt.binary_location = r'D:/Google/Chrome/Application/chrome.exe'
-chromedriver = r'D:/cmd100/chromedriver.exe'
-browser = webdriver.Chrome(chrome_options=opt, executable_path=chromedriver)
+    opt.add_argument("--ignore-certificate-errors")
+    opt.add_argument("--ignore-ssl-errors")
+    opt.add_experimental_option("excludeSwitches", ["enable-automation"])
+    opt.add_experimental_option('excludeSwitches', ['enable-logging'])
+    opt.add_experimental_option('useAutomationExtension', False)
+    opt.binary_location = r'D:/Google/Chrome/Application/chrome.exe'
+    chromedriver = r'D:/cmd100/chromedriver.exe'
+    browser = webdriver.Chrome(chrome_options=opt, executable_path=chromedriver)
+    return browser
 log = baseCore.getLogger()
 requests.adapters.DEFAULT_RETRIES = 3
 
@@ -67,7 +68,7 @@ def spider_annual_report(dict_info,num):
     com_name = dict_info['com_name']
     code = dict_info['code']
     url_1 = f'https://vip.stock.finance.sina.com.cn/corp/go.php/vCB_Bulletin/stockid/{code}/page_type/ndbg.phtml'
-
+    browser = createDriver()
     browser.get(url_1)
     time.sleep(3)
     page_source = browser.page_source