Python Selenium自动化10fastfingers脚本输入初始单词后崩溃的排查与解决

我编写了一个自动化脚本用于完成10fastfingers的英文打字测试，但脚本在输入页面初始显示的单词集后就崩溃，无法继续输入后续内容。

from selenium import webdriver
from selenium.webdriver.common.keys import Keys
import time

try:
    driver = webdriver.Firefox()
    driver.get("https://10fastfingers.com/typing-test/english")

    # Wait for the page to fully load
    time.sleep(6)

    # Find the text input element on the page
    text_input = driver.find_element("id", "inputfield")

    current_word = driver.find_element("id", "row1")
    print(type(driver))  # <-- Add this line
    # Type the words into the input field
    while True:
        text_to_type = current_word.text

        for character in text_to_type:
            text_input.send_keys(character)
            time.sleep(0.1)

        # Check if the next sibling element exists
        next_word = current_word.find_element_by_xpath("following-sibling::span[1]")
        if not next_word:
            break

        current_word = next_word

    # Submit the form to complete the typing test
    text_input.send_keys(Keys.RETURN)

except Exception as e:
    print("An error occurred:", e)

finally:
    # Close the web browser
  driver.quit()

• 模拟人类输入延迟：误以为浏览器拦截自动输入，给每个字符和单词添加随机延迟，但问题依旧：

for word in text_to_type:
        for character in word:
            text_input.send_keys(character)
            time.sleep(random.uniform(0.05, 0.1))
        text_input.send_keys(Keys.SPACE)
        time.sleep(random.uniform(0.5, 1.0))

• 重新获取单词块：猜测页面输入后会动态更新内容，尝试循环获取#row1下的span元素，但没有效果：

while True:
        # Get the next word to type
        word_elements = driver.find_elements("css selector", "#row1 span")
        if not word_elements:
            break
        word = word_elements[0].text

问题出在元素定位逻辑上，最初通过row1和兄弟元素遍历的方式无法适配页面动态更新的结构。改用//span[@wordnr]这个XPath可以直接获取所有带wordnr属性的单词元素，这些元素是页面预加载的全部测试单词，无需处理后续的动态更新。

from selenium import webdriver
from selenium.webdriver.common.keys import Keys
import time

try:
    driver = webdriver.Firefox()
    driver.get("https://10fastfingers.com/typing-test/english")

    # 等待页面完全加载
    time.sleep(6)

    # 定位输入框
    text_input = driver.find_element("id", "inputfield")

    # 获取所有带wordnr属性的单词元素（即所有需要输入的单词）
    words = driver.find_elements("xpath", "//span[@wordnr]")

    # 逐个输入单词
    for word in words:
        text_to_type = word.text
        for character in text_to_type:
            text_input.send_keys(character)
            time.sleep(0)
        text_input.send_keys(" ")
        time.sleep(0)

    # 提交完成测试
    text_input.send_keys(Keys.RETURN)

except Exception as e:
    print("发生错误:", e)

# finally:
    # 关闭浏览器
    # driver.quit()

• 放弃通过row1和兄弟元素遍历的方式，直接定位所有带wordnr属性的span元素，一次性获取全部测试单词，避开页面动态更新带来的元素定位问题。
• 移除不必要的延迟（或设置为0），确保脚本快速执行的同时，避免因延迟导致元素状态变化引发的异常。

内容的提问来源于stack exchange，提问作者Rashad Aliyev