恢复 Bilibili 收藏夹中失效视频

'''
需要 Python 3.6 + requests
'''
import requests
from math import ceil
from urllib import parse
import time


class favlist():
    apiUrl = 'https://api.bilibili.com/medialist/gateway/base/spaceDetail?media_id={fid}&pn={pn}&ps=20'

    def __init__(self, url: str):
        '''
        初始化类，对 URL 进行解析，并从 API 中获得收藏夹视频数目

        :param url: 收藏夹 URL
        Usage::
            favlist('https://space.bilibili.com/3992364/favlist?fid=45027464')
        '''
        params = self.parse_url(url)
        self.fid = params['fid'][0]
        favList = self.get_content(1)
        self.mediaCount = favList['data']['info']['media_count']

    def parse_url(self, inUrl: str):
        '''
        解析 URL，返回参数列表

        :param inUrl: 被解析 URL
        :rtype: dict
        '''
        paramStr = parse.urlsplit(inUrl)
        params = parse.parse_qs(paramStr[3])
        return params

    def get_content(self, pn: int):
        '''
        通过 API 获得收藏夹内容

        :param pn: 收藏夹页数
        :rtype: dict
        '''
        currUrl = self.apiUrl.format(fid=self.fid, pn=pn)
        try:
            apiResp = requests.get(currUrl)
        except requests.HTTPError as e:
            print(f'get_content() error, HTTPError {e}')
            return {}
        except requests.ConnectionError as e:
            print(f'get_content() error, ConnectionError {e}')
            return {}
        else:
            favListJson = apiResp.json()
            return favListJson

    def get_failure_media(self):
        '''
        根据标题筛选失效视频

        :return: 失效视频列表
        :rtype: list
        '''
        failMedia = []
        currPage = 1
        totalPage = ceil(self.mediaCount/20)
        print('Failure media:')
        while currPage <= totalPage:
            favList = self.get_content(currPage)
            print(f'Page {currPage}:')
            # B站程序员蜜汁英语水平，这个`medias`是怎么回事
            for i in favList['data']['medias']:
                if i['title'] == '已失效视频':
                    print(f'av{i["id"]}')
                    failMedia.append(i)
            currPage += 1
        return failMedia


def get_biliplus_data(aid: int):
    plusUrl = f'https://hd.biliplus.com/api/aidinfo?aid={aid}'
    try:
        plusResp = requests.get(plusUrl, timeout=18)
        plusJson = plusResp.json()
        if plusJson['code'] != 0:
            # biliPlus 文档有毒吧，code 的意义只能实验出来...
            raise requests.HTTPError(f'status code {plusJson["code"]}')
    except requests.HTTPError as e:
        print(f'get_biliplus_data(), HTTPError {e}')
        return {}
    except requests.ConnectionError as e:
        print(f'get_biliplus_data(), ConnectionError {e}')
        return {}
    else:
        mediaData = plusJson["data"][str(aid)]
        print(
            f'title: {mediaData["title"]}, author: {mediaData["author"]}, av{aid}')
        return mediaData


def save_to_file(inStr: str, fileName: str):
    try:
        fp = open(fileName, 'a')
    except IOError as identifier:
        fp = open(fileName, 'w')
    finally:
        fp.write(inStr)
        fp.close()


if __name__ == "__main__":
    hint = '输入收藏夹URL\n例子: https://space.bilibili.com/3992364/favlist?fid=45027464\n> '
    url = input(hint)
    f = favlist(url)
    fail = f.get_failure_media()
    print('Failure videos detail:')
    for i in fail:
        save_to_file(f'av{i["id"]}\n', 'avNum.txt')
        plus = get_biliplus_data(i['id'])
        if len(plus) != 0:
            save_to_file(f'{plus["title"]},{plus["author"]},av{i["id"]}\n', 'result.csv')
        # biliPlus 的土豆服务器一分钟只能请求 5 次
        # 而且还经常爆炸
        time.sleep(20)