哔哩哔哩视频爬取源码分享

背景：

　　无意间发现B站有个老师的课程特别好（python教学的视频），单位的网络限制了视频网站访问，所以尝试着去把视频下载保存起来，经过一段时间的研究终于完成代码的开发，大家有需要的可以搞下来，后续我会进行一个延伸优化，争取做到通过前端页面的视频名称输入作为爬取条件进行下载。

 

第一版；2021年02月23日

第一版；2021年02月24日

解决间歇不触发点击保存按钮的问题

#_author_='Lucky';
#date:  2021/2/18

import time

from pywinauto import application
from selenium import webdriver
from selenium.webdriver.common.action_chains import ActionChains


def chrome_options_setting(web_driver):
    """
    设置 Chrome Browser 的下载前询问每个文件的保存位置选项为打开（true）
    :param web_driver: 浏览器驱动
    :return: None
    """
    web_driver.get("chrome://settings/downloads")
    time.sleep(2)
    web_element = web_driver.find_element_by_xpath("//settings-ui")
    shadowRoot = web_driver.execute_script("return arguments[0].shadowRoot", web_element)
    # shadowRoot 节点下不能使用 xpath 选择器
    web_element = shadowRoot.find_element_by_id("container").find_element_by_id("main")
    shadowRoot = web_driver.execute_script("return arguments[0].shadowRoot", web_element)
    web_element = shadowRoot.find_element_by_css_selector("settings-basic-page[role='main']")
    shadowRoot = web_driver.execute_script("return arguments[0].shadowRoot", web_element)
    web_element = shadowRoot.find_element_by_css_selector("settings-downloads-page")
    shadowRoot = web_driver.execute_script("return arguments[0].shadowRoot", web_element)
    web_element = shadowRoot.find_element_by_css_selector("settings-toggle-button")
    shadowRoot = web_driver.execute_script("return arguments[0].shadowRoot", web_element)
    result = shadowRoot.find_element_by_css_selector(
        "#outerRow > cr-toggle[aria-describedby='sub-label-text']").get_attribute("aria-pressed")
    if result == "false":
        shadowRoot.find_element_by_css_selector("#outerRow > cr-toggle[aria-describedby='sub-label-text']").click()


def file_name_save(url, xpath):
    bili_browser.get(url)
    time.sleep(3)
    file_name = bili_browser.find_element_by_xpath(xpath).text
    print(file_name)
    return file_name


def save_as_window(file_name):
    app = application.Application().connect(title_re=u"另存为", class_name="#32770")
    save_as_spec = app.window(title=u"另存为", class_name="#32770")
    # print(save_as_spec.print_control_identifiers())

    edit = save_as_spec["Edit"]
    edit.set_text(file_name)  # 第一种方法是直接设置edit的text
    # edit.type_keys(file_name_save(), with_spaces=True)  # 第二种是在里面模拟键盘输入(如果字符串中没有空格，可以省略后面的参数)，殊途同归

    while True:
        time.sleep(1)
        app['另存为']['保存(&S)'].click()
        if app.window(title=u"另存为", class_name="#32770").exists() is False: break


def main_download(url):
    bili_browser.get("https://xbeibeix.com/api/bilibili")
    time.sleep(3)
    bili_browser.find_element_by_xpath("//*[@placeholder='输入地址']").clear()
    time.sleep(1)
    bili_browser.find_element_by_xpath("//*[@placeholder='输入地址']").send_keys(url)
    time.sleep(1)
    bili_browser.find_element_by_id('button-1').click()
    time.sleep(2)
    element2 = bili_browser.find_element_by_xpath("//a[contains(text(),'MP4地址')]")
    ActionChains(bili_browser).key_down(u'\ue00a').click(element2).perform()
    ActionChains(bili_browser).key_up(u'\ue00a')


if __name__ == "__main__":
    bili_browser = webdriver.Chrome()
    bili_browser.maximize_window()

    # 设置Chrome浏览器下载前询问每个文件的保存位置选项
    chrome_options_setting(web_driver=bili_browser)
    time.sleep(2)

    for index in range(456, 634):
        video_url = 'https://www.bilibili.com/video/BV197411G75w?p=' + str(index)
        file_name_xpath = '//*[@id="multi_page"]/div[2]/ul/li[%d]/a/div/div[1]/span[2]' % index

        video_name = file_name_save(url=video_url, xpath=file_name_xpath)
        print(file_name_xpath)
        # 处理Windows10 系统“另存为”窗口，并修改保存文件名称
        main_download(video_url)
        time.sleep(1)

        save_as_window(file_name=video_name)
        time.sleep(3)

    time.sleep(5)
    bili_browser.quit()