Selenium 自动化测试：识别并提取页面下拉框和输入框信息

selects_attributes = {}
selects_attributes_ele = {}
selects = driver.find_elements(By.TAG_NAME, 'dt')
print('下拉框数量：', len(selects))
for element in selects:
    if not element.is_displayed():  # text == ''
        continue
    text = element.text
    location = element.location
    describe = Input_describe(location, text_attributes)

    if text == '':
        text = describe.split('。')[0]
    if text in selects_attributes:
        # if location not in [t['coordinate'] for t in selects_attributes[text]]:
        # if abs(location['x'] - selects_attributes[text][-1]['coordinate']['x']) < 15 or abs(
        #         location['y'] - selects_attributes[text][-1]['coordinate']['y']) < 15:
        #     continue
        selects_attributes[text].append(
            {'coordinate': (location['x'], location['y']), 'describe': Input_describe(location, text_attributes)})
        # selects_attributes[text].append(
        #     {'coordinate': (location['x'], location['y']), 'id': element.id, 'class': element.get_attribute('class'),
        #      'describe': Input_describe(location, text_attributes)})
        selects_attributes_ele[text].append(
            {'coordinate': (location['x'], location['y']), 'id': element.id,
             'class': element.get_attribute('class'),
             'describe': Input_describe(location, text_attributes), 'element': element})
    else:
        selects_attributes[text] = [
            {'coordinate': (location['x'], location['y']), 'describe': Input_describe(location, text_attributes)}]
        # driver.execute_script('arguments[0].click();', element)
        # selects_attributes[text] = [
        #     {'coordinate': (location['x'], location['y']), 'id': element.id, 'class': element.get_attribute('class'),
        #      'describe': Input_describe(location, text_attributes)}]
        selects_attributes_ele[text] = [
            {'coordinate': (location['x'], location['y']), 'id': element.id,
             'class': element.get_attribute('class'),
             'describe': Input_describe(location, text_attributes), 'element': element}]
print(selects_attributes)

# inputs = driver.find_elements(By.TAG_NAME, 'input')
inputs = driver.find_elements(By.XPATH, '//input | //textarea')
print('输入框数量：', len(inputs))
inputs_attributes = {}
inputs_attributes_ele = {}
for element in inputs:
    if element.is_displayed() and element.get_attribute('type') != 'hidden':
        key = element.get_attribute('value')
        location = element.location
        describe = Input_describe(location, text_attributes)
        if key == '':
            key = describe.split('。')[0]
        skip = False
        for select_text, select_attributes in selects_attributes.items():
            for select_attribute in select_attributes:
                select_coordinate = select_attribute['coordinate']
                distance = ((location['x'] - select_coordinate[0]) ** 2 + (location['y'] - select_coordinate[1]) ** 2) ** 0.5
                if distance < 10:
                    skip = True
                    break
            if skip:
                break
        if skip:
            continue
        if key in inputs_attributes:
            if abs(location['x'] - inputs_attributes[key][-1]['coordinate']['x']) < 15 and abs(
                    location['y'] - inputs_attributes[key][-1]['coordinate']['y']) < 15:
                continue
            inputs_attributes[key].append({'coordinate': (location['x'], location['y']),
                                           'data-placeholder': element.get_attribute('data-placeholder'),
                                           'type': element.get_attribute('type'),
                                           'accessible_name': element.accessible_name,
                                           'describe': describe + '。' + element.accessible_name
                                           })
            inputs_attributes_ele[key].append({'id': element.get_attribute('id'),
                                               'coordinate': (location['x'], location['y']),
                                               'data-placeholder': element.get_attribute(
                                                   'data-placeholder'),
                                               'type': element.get_attribute('type'),
                                               'accessible_name': element.accessible_name,
                                               'class': element.get_attribute('class'),
                                               'describe': describe + '。' + element.accessible_name,
                                               'element': element
                                               })
        else:
            inputs_attributes[key] = [{'coordinate': (location['x'], location['y']),
                                       'data-placeholder': element.get_attribute('data-placeholder'),
                                       'type': element.get_attribute('type'),
                                       'accessible_name': element.accessible_name,
                                       'describe': describe + '。' + element.accessible_name
                                       }]
            inputs_attributes_ele[key] = [{'id': element.get_attribute('id'),
                                           'coordinate': (location['x'], location['y']),
                                           'data-placeholder': element.get_attribute('data-placeholder'),
                                           'type': element.get_attribute('type'),
                                           'accessible_name': element.accessible_name,
                                           'class': element.get_attribute('class'),
                                           'describe': describe + '。' + element.accessible_name,
                                           'element': element
                                           }]
print(inputs_attributes)

代码中，首先通过 driver.find_elements(By.TAG_NAME, 'dt') 获取页面中的所有 dt 标签元素，这些元素通常代表下拉框。然后遍历每个 dt 元素，提取其文本内容、坐标位置、描述文本等信息，并存储到 selects_attributes 和 selects_attributes_ele 字典中。

接着，通过 driver.find_elements(By.XPATH, '//input | //textarea') 获取页面中的所有输入框元素，遍历每个输入框元素，提取其属性信息，例如 value、data-placeholder、type、accessible_name 等。在遍历输入框元素时，代码中还添加了判断条件，如果输入框元素的坐标位置与 selects_attributes 字典中存储的下拉框元素坐标位置的欧式距离小于 10，则跳过该输入框元素，避免重复提取。

最后，代码将提取到的下拉框和输入框信息分别存储到 selects_attributes 和 inputs_attributes 字典中，方便后续使用。

该代码可以帮助您快速识别网页中的下拉框和输入框元素，并提取其属性信息，从而方便您进行自动化测试或其他网页分析工作。