Question

我正试图从TripAdvisor搜索酒店的评论并将其写在文本文件中。到目前为止，代码表现良好，除了事实上它偶尔会在我编写文本文件的行上抛出StaleElementException。这是我的代码：

for num in range(page_count):
    try:
        if num != 0:
            try:
                nxt = WebDriverWait(driver, 10).until(EC.presence_of_element_located((By.CSS_SELECTOR, "a.nav.next.rndBtn.ui_button.primary.taLnk")))
                #nxt = driver.find_element_by_css_selector("a.nav.next.rndBtn.ui_button.primary.taLnk")
                nxt.click()
                driver.implicitly_wait(5)
            except NoSuchElementException:
                driver.refresh()
                #driver.implicitly_wait(5)
                nxt = WebDriverWait(driver, 10).until(EC.presence_of_element_located((By.CSS_SELECTOR, "a.nav.next.rndBtn.ui_button.primary.taLnk")))
                nxt.click()
                driver.implicitly_wait(5)
        try:
            more = WebDriverWait(driver, 10).until(EC.element_to_be_clickable((By.CSS_SELECTOR, "span.taLnk.ulBlueLinks")))
            #more = driver.find_element_by_css_selector("span.taLnk.ulBlueLinks")
            more.click()
            time.sleep(1)
        except TimeoutException:
            print("There is no 'more' button on page %d" % (num+1))
        except WebDriverException:
            nxt = WebDriverWait(driver, 10).until(EC.presence_of_element_located((By.CSS_SELECTOR, "span.taLnk.ulBlueLinks")))
            nxt.click()
            driver.implicitly_wait(5)

        review_result = WebDriverWait(driver, 10).until(EC.presence_of_all_elements_located((By.CLASS_NAME, 'entry')))

        with open('New_Review.txt', 'a') as fid:
            for review in review_result:
                    fid.write(unidecode(review.text))
                    fid.write(sep)
                    fid.write(line_break)
        print ("processing done till page number %d" % (num+1))
    except StaleElementReferenceException:
        driver.refresh()
        driver.implicitly_wait(5)

        try:
            more = WebDriverWait(driver, 10).until(EC.element_to_be_clickable((By.CSS_SELECTOR, "span.taLnk.ulBlueLinks")))
            #more = driver.find_element_by_css_selector("span.taLnk.ulBlueLinks")
            more.click()
        except TimeoutException:
            print("There is no 'more' button on page %d" % (num+1))
        except WebDriverException:
            nxt = WebDriverWait(driver, 10).until(EC.presence_of_element_located((By.CSS_SELECTOR, "span.taLnk.ulBlueLinks")))
            nxt.click()
            driver.implicitly_wait(5)

        review_result = WebDriverWait(driver, 10).until(EC.presence_of_all_elements_located((By.CLASS_NAME, 'entry')))

        #print (review_result[1].text)
        with open('New_Review.csv', 'a') as fid:
            writer = csv.writer(fid, delimiter = ',', line_break = '\n')
            for review in review_result:
                    fid.write(unidecode(review.text))
                    fid.write(sep)
                    fid.write(line_break)
        print ("processing done till page number %d" % (num+1))

这是错误：

StaleElementReferenceException：陈旧元素引用：元素是未附加到页面文档

追溯给出了这一行：

fid.write(unidecode(review.text))

我已经尝试处理异常，但它不适合我，我很难弄清楚到底哪里错了。任何帮助表示赞赏！

Answer 1

尝试创建辅助方法，例如

def get_text(locator):
   staled = True
   while staled:
     try:
        return WebDriverWait(driver, 10).until(EC.presence_of_element(locator)).text
    except StaleElementReferenceException:
       `log something or limit retry to certain times`

然后改变你获得文本的方式

review_result = WebDriverWait(driver, 10).until(EC.presence_of_all_elements_located())
num_of_review - review_result.length
        with open('New_Review.txt', 'a') as fid:
            for index in range(1, num_of_review):
                    review_text = get_text((By.XPATH, "//*[@class='entry'][index]"))
                    fid.write(unidecode(review_text))
                    fid.write(sep)
                    fid.write(line_break)

使用selenium webdriver

1 个答案: