fix:ptt执行效率

pc_twitter.py

@@ -22,6 +22,7 @@ from selenium.webdriver.support import expected_conditions as EC
 # ---------------   selenium 依赖 end ----------------
 
 import platform
+
 '''
 打开指定网页，并使用 Selenium 模拟点击 "GDPR-accept" 按钮，然后循环点击 "search-show-more-button" 按钮来加载更多数据，直到按钮不再可点击为止。最后，获取完整的分页数据并关闭浏览器驱动。
 '''
@@ -35,15 +36,19 @@ def reptile(browser=None, search_word=""):
     """
     print(f"搜索词:{search_word}")
     base_url = "https://twitter.com/"
-    if platform.system() == "Windows":
-        browser = browser or create(no_headless=True, using_user_data=True)
-    else:
-        browser = browser or create(no_headless=True, using_user_data=True)
+    browser = browser or create(no_headless=False, using_user_data=True)
     # print(browser)
     # 打开网页
     browser.get(base_url)
     time.sleep(2)
     try:
+        try:
+            login_button = browser.find_element('xpath', "//a[@href='/login']")
+            login_button.click()
+            time.sleep(2)
+        except:
+            error = ""
+
         # wait = WebDriverWait(browser, 20)
         # wait.until(EC.presence_of_element_located((By.XPATH, "//input[@autocomplete='username']")))
         # 检测是否要登录
@@ -59,9 +64,11 @@ def reptile(browser=None, search_word=""):
         # # 获取登录按钮
         button_login = browser.find_element('xpath', "//div[@data-testid='LoginForm_Login_Button']")
         button_login.click()
-        time.sleep(1)
+        time.sleep(2)
     except:
-        print("------")
+        # print("------")
+        error = ""
+
     url = 'https://twitter.com/search?q=' + search_word + '&src=typed_query'
     browser.get(url)
     wait = WebDriverWait(browser, 10)
@@ -86,6 +93,15 @@ def reptile(browser=None, search_word=""):
         except:
             link_str = ""
             timestamp = time.time()
+
+        # 删除多余div
+        parent_div = soup.find("div")
+        # 找到所有的 <div> 子元素
+        div_elements = parent_div.find_all('div', recursive=False)
+        for key, item in enumerate(div_elements):
+            if key == 0 or key == len(div_elements)-1:
+                item.extract()
+
         author = element_authors_list[index].text
         # 标题取：作者+日期
         title = f"{author}-{datetime.fromtimestamp(int(timestamp))}"
@@ -113,8 +129,8 @@ def reptile(browser=None, search_word=""):
                 custom_video["src"] = ""
                 parent_div.append(custom_video)
         else:
-            print("")
-
+            # print("")
+            error = ""
         picture_url = []
         if len(image_list) > 0:
             for key, element in enumerate(image_list):
@@ -130,7 +146,9 @@ def reptile(browser=None, search_word=""):
                     element['src'] = access_address
                     picture_url.append(download_dir)
         else:
-            print("")
+            # print("")
+            error = ""
+
         content = soup.prettify()
 
         # ---------------- 判断类型 start ----------