selenium模拟百度搜索

selenium_demo.py

+from selenium import webdriver
+from selenium.webdriver.common.by import By
+import time
+def open_driver():
+    option = webdriver.EdgeOptions()
+    option.add_experimental_option("detach",True)
+    option.add_experimental_option('excludeSwitches', ['enable-automation'])  # 此步骤很重要，设置为开发者模式，防止被各大网站识别出来使用了Selenium
+    edge = webdriver.Edge(options=option)
+    edge.maximize_window()
+    edge.get("https://www.baidu.com")
+    # 判断“百度一下”在网页的标题上
+    assert "百度一下" in edge.title
+    # ele = edge.find_element(By.ID,"kw")
+
+    # edge.find_element_by_id('kw')
+    return edge
+
+def search(driver):
+    input_ele = driver.find_element(By.ID,"kw")
+    input_ele.clear()
+    input_ele.send_keys("程序猿小董")
+    button = driver.find_element(By.ID,"su")
+    button.click()
+    driver.implicitly_wait(10)
+    # driver.find_elements(By.CLASS_NAME,"")
+    results = driver.find_elements(By.CSS_SELECTOR,".t a , em , .c-title-text")
+    for result in results:
+        if result.get_attribute("href"):
+            if "-" not in result.get_attribute("text"):
+                # 搜索结果的标题
+                title = result.get_attribute("text").strip()
+                # 搜索结果的网址
+                link = result.get_attribute("href")
+                print(driver.title)
+                print(driver.current_url)
+                print(driver.window_handles)
+                result.click()
+                print(driver.window_handles)
+                driver.switch_to.window(driver.window_handles[-1])
+                print(driver.title)
+                print(driver.current_url)
+                #滚动到目的元素
+                # ele = driver.find_element(By.CLASS_NAME,"tw0jb")
+                # driver.execute_script("arguments[0].scrollIntoView();",ele)
+                #向下滚动1000像素
+                driver.execute_script("window.scrollBy(0,1000)")
+                time.sleep(60)
+                driver.switch_to.window(driver.window_handles[0])
+                print(driver.title)
+                print(driver.current_url)
+
+driver = open_driver()
+search(driver)