yt_dlp/extractor/mgtv.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import base64
   5 import time
   6 import uuid
   7
   8 from .common import InfoExtractor
   9 from ..compat import (
  10     compat_HTTPError,
  11     compat_str,
  12 )
  13 from ..utils import (
  14     ExtractorError,
  15     int_or_none,
  16     try_get,
  17     url_or_none,
  18 )
  19
  20
  21 class MGTVIE(InfoExtractor):
  22     _VALID_URL = r'https?://(?:w(?:ww)?\.)?mgtv\.com/(v|b)/(?:[^/]+/)*(?P<id>\d+)\.html'
  23     IE_DESC = '芒果TV'
  24     IE_NAME = 'MangoTV'
  25
  26     _TESTS = [{
  27         'url': 'http://www.mgtv.com/v/1/290525/f/3116640.html',
  28         'info_dict': {
  29             'id': '3116640',
  30             'ext': 'mp4',
  31             'title': '我是歌手 第四季',
  32             'description': '我是歌手第四季双年巅峰会',
  33             'duration': 7461,
  34             'thumbnail': r're:^https?://.*\.jpg$',
  35         },
  36     }, {
  37         'url': 'https://w.mgtv.com/b/427837/15588271.html',
  38         'info_dict': {
  39             'id': '15588271',
  40             'ext': 'mp4',
  41             'title': '春日迟迟再出发 沉浸版',
  42             'description': 'md5:a7a05a05b1aa87bd50cae619b19bbca6',
  43             'thumbnail': r're:^https?://.+\.jpg',
  44             'duration': 4026,
  45         },
  46     }, {
  47         'url': 'https://w.mgtv.com/b/333652/7329822.html',
  48         'info_dict': {
  49             'id': '7329822',
  50             'ext': 'mp4',
  51             'title': '拜托，请你爱我',
  52             'description': 'md5:cd81be6499bafe32e4d143abd822bf9c',
  53             'thumbnail': r're:^https?://.+\.jpg',
  54             'duration': 2656,
  55         },
  56     }, {
  57         'url': 'https://w.mgtv.com/b/427837/15591647.html',
  58         'only_matching': True,
  59     }, {
  60         'url': 'https://w.mgtv.com/b/388252/15634192.html?fpa=33318&fpos=4&lastp=ch_home',
  61         'only_matching': True,
  62     }, {
  63         'url': 'http://www.mgtv.com/b/301817/3826653.html',
  64         'only_matching': True,
  65     }, {
  66         'url': 'https://w.mgtv.com/b/301817/3826653.html',
  67         'only_matching': True,
  68     }]
  69
  70     def _real_extract(self, url):
  71         video_id = self._match_id(url)
  72         tk2 = base64.urlsafe_b64encode(
  73             f'did={compat_str(uuid.uuid4()).encode()}|pno=1030|ver=0.3.0301|clit={int(time.time())}'.encode())[::-1]
  74         try:
  75             api_data = self._download_json(
  76                 'https://pcweb.api.mgtv.com/player/video', video_id, query={
  77                     'tk2': tk2,
  78                     'video_id': video_id,
  79                     'type': 'pch5'
  80                 }, headers=self.geo_verification_headers())['data']
  81         except ExtractorError as e:
  82             if isinstance(e.cause, compat_HTTPError) and e.cause.code == 401:
  83                 error = self._parse_json(e.cause.read().decode(), None)
  84                 if error.get('code') == 40005:
  85                     self.raise_geo_restricted(countries=self._GEO_COUNTRIES)
  86                 raise ExtractorError(error['msg'], expected=True)
  87             raise
  88         info = api_data['info']
  89         title = info['title'].strip()
  90         stream_data = self._download_json(
  91             'https://pcweb.api.mgtv.com/player/getSource', video_id, query={
  92                 'pm2': api_data['atc']['pm2'],
  93                 'tk2': tk2,
  94                 'video_id': video_id,
  95                 'src': 'intelmgtv',
  96             }, headers=self.geo_verification_headers())['data']
  97         stream_domain = stream_data['stream_domain'][0]
  98
  99         formats = []
 100         for idx, stream in enumerate(stream_data['stream']):
 101             stream_path = stream.get('url')
 102             if not stream_path:
 103                 continue
 104             format_data = self._download_json(
 105                 stream_domain + stream_path, video_id,
 106                 note=f'Download video info for format #{idx}')
 107             format_url = format_data.get('info')
 108             if not format_url:
 109                 continue
 110             tbr = int_or_none(stream.get('filebitrate') or self._search_regex(
 111                 r'_(\d+)_mp4/', format_url, 'tbr', default=None))
 112             formats.append({
 113                 'format_id': compat_str(tbr or idx),
 114                 'url': url_or_none(format_url),
 115                 'ext': 'mp4',
 116                 'tbr': tbr,
 117                 'protocol': 'm3u8_native',
 118                 'http_headers': {
 119                     'Referer': url,
 120                 },
 121                 'format_note': stream.get('name'),
 122             })
 123         self._sort_formats(formats)
 124
 125         return {
 126             'id': video_id,
 127             'title': title,
 128             'formats': formats,
 129             'description': info.get('desc'),
 130             'duration': int_or_none(info.get('duration')),
 131             'thumbnail': info.get('thumb'),
 132             'subtitles': self.extract_subtitles(video_id, stream_domain),
 133         }
 134
 135     def _get_subtitles(self, video_id, domain):
 136         info = self._download_json(f'https://pcweb.api.mgtv.com/video/title?videoId={video_id}',
 137                                    video_id, fatal=False) or {}
 138         subtitles = {}
 139         for sub in try_get(info, lambda x: x['data']['title']) or []:
 140             url_sub = sub.get('url')
 141             if not url_sub:
 142                 continue
 143             locale = sub.get('captionCountrySimpleName')
 144             sub = self._download_json(f'{domain}{url_sub}', video_id, fatal=False,
 145                                       note=f'Download subtitle for locale {sub.get("name")} ({locale})') or {}
 146             sub_url = url_or_none(sub.get('info'))
 147             if not sub_url:
 148                 continue
 149             subtitles.setdefault(locale or 'en', []).append({
 150                 'url': sub_url,
 151                 'ext': 'srt'
 152             })
 153         return subtitles