yt_dlp/extractor/mgtv.py

   1 import base64
   2 import time
   3 import uuid
   4
   5 from .common import InfoExtractor
   6 from ..networking.exceptions import HTTPError
   7 from ..utils import (
   8     ExtractorError,
   9     int_or_none,
  10     parse_resolution,
  11     traverse_obj,
  12     try_get,
  13     url_or_none,
  14     urljoin,
  15 )
  16
  17
  18 class MGTVIE(InfoExtractor):
  19     _VALID_URL = r'https?://(?:w(?:ww)?\.)?mgtv\.com/[bv]/(?:[^/]+/)*(?P<id>\d+)\.html'
  20     IE_DESC = '芒果TV'
  21     IE_NAME = 'MangoTV'
  22
  23     _TESTS = [{
  24         'url': 'http://www.mgtv.com/v/1/290525/f/3116640.html',
  25         'info_dict': {
  26             'id': '3116640',
  27             'ext': 'mp4',
  28             'title': '我是歌手 第四季',
  29             'description': '我是歌手第四季双年巅峰会',
  30             'duration': 7461,
  31             'thumbnail': r're:^https?://.*\.jpg$',
  32         },
  33         'params': {'skip_download': 'm3u8'},
  34     }, {
  35         'url': 'https://w.mgtv.com/b/427837/15588271.html',
  36         'info_dict': {
  37             'id': '15588271',
  38             'ext': 'mp4',
  39             'title': '春日迟迟再出发 沉浸版第1期：陆莹结婚半年查出肾炎被离婚 吴雅婷把一半票根退给前夫',
  40             'description': 'md5:a7a05a05b1aa87bd50cae619b19bbca6',
  41             'thumbnail': r're:^https?://.+\.jpg',
  42             'duration': 4026,
  43         },
  44         'params': {'skip_download': 'm3u8'},
  45     }, {
  46         'url': 'https://w.mgtv.com/b/333652/7329822.html',
  47         'info_dict': {
  48             'id': '7329822',
  49             'ext': 'mp4',
  50             'title': '拜托，请你爱我',
  51             'description': 'md5:cd81be6499bafe32e4d143abd822bf9c',
  52             'thumbnail': r're:^https?://.+\.jpg',
  53             'duration': 2656,
  54         },
  55         'params': {'skip_download': 'm3u8'},
  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     _RESOLUTIONS = {
  71         '标清': ('480p', '854x480'),
  72         '高清': ('540p', '960x540'),
  73         '超清': ('720p', '1280x720'),
  74         '蓝光': ('1080p', '1920x1080'),
  75     }
  76
  77     def _real_extract(self, url):
  78         video_id = self._match_id(url)
  79         tk2 = base64.urlsafe_b64encode(
  80             f'did={uuid.uuid4()}|pno=1030|ver=0.3.0301|clit={int(time.time())}'.encode())[::-1]
  81         try:
  82             api_data = self._download_json(
  83                 'https://pcweb.api.mgtv.com/player/video', video_id, query={
  84                     'tk2': tk2,
  85                     'video_id': video_id,
  86                     'type': 'pch5',
  87                 }, headers=self.geo_verification_headers())['data']
  88         except ExtractorError as e:
  89             if isinstance(e.cause, HTTPError) and e.cause.status == 401:
  90                 error = self._parse_json(e.cause.response.read().decode(), None)
  91                 if error.get('code') == 40005:
  92                     self.raise_geo_restricted(countries=self._GEO_COUNTRIES)
  93                 raise ExtractorError(error['msg'], expected=True)
  94             raise
  95
  96         stream_data = self._download_json(
  97             'https://pcweb.api.mgtv.com/player/getSource', video_id, query={
  98                 'tk2': tk2,
  99                 'pm2': api_data['atc']['pm2'],
 100                 'video_id': video_id,
 101                 'type': 'pch5',
 102                 'src': 'intelmgtv',
 103             }, headers=self.geo_verification_headers())['data']
 104         stream_domain = traverse_obj(stream_data, ('stream_domain', ..., {url_or_none}), get_all=False)
 105
 106         formats = []
 107         for idx, stream in enumerate(traverse_obj(stream_data, ('stream', lambda _, v: v['url']))):
 108             stream_name = traverse_obj(stream, 'name', 'standardName', 'barName', expected_type=str)
 109             resolution = traverse_obj(
 110                 self._RESOLUTIONS, (stream_name, 1 if stream.get('scale') == '16:9' else 0))
 111             format_url = traverse_obj(self._download_json(
 112                 urljoin(stream_domain, stream['url']), video_id, fatal=False,
 113                 note=f'Downloading video info for format {resolution or stream_name}'),
 114                 ('info', {url_or_none}))
 115             if not format_url:
 116                 continue
 117             tbr = int_or_none(stream.get('filebitrate') or self._search_regex(
 118                 r'_(\d+)_mp4/', format_url, 'tbr', default=None))
 119             formats.append({
 120                 'format_id': str(tbr or idx),
 121                 'url': format_url,
 122                 'ext': 'mp4',
 123                 'tbr': tbr,
 124                 'vcodec': stream.get('videoFormat'),
 125                 'acodec': stream.get('audioFormat'),
 126                 **parse_resolution(resolution),
 127                 'protocol': 'm3u8_native',
 128                 'http_headers': {
 129                     'Referer': url,
 130                 },
 131                 'format_note': stream_name,
 132             })
 133
 134         return {
 135             'id': video_id,
 136             'formats': formats,
 137             **traverse_obj(api_data, ('info', {
 138                 'title': ('title', {str.strip}),
 139                 'description': ('desc', {str}),
 140                 'duration': ('duration', {int_or_none}),
 141                 'thumbnail': ('thumb', {url_or_none}),
 142             })),
 143             'subtitles': self.extract_subtitles(video_id, stream_domain),
 144         }
 145
 146     def _get_subtitles(self, video_id, domain):
 147         info = self._download_json(f'https://pcweb.api.mgtv.com/video/title?videoId={video_id}',
 148                                    video_id, fatal=False) or {}
 149         subtitles = {}
 150         for sub in try_get(info, lambda x: x['data']['title']) or []:
 151             url_sub = sub.get('url')
 152             if not url_sub:
 153                 continue
 154             locale = sub.get('captionSimpleName') or 'en'
 155             sub = self._download_json(f'{domain}{url_sub}', video_id, fatal=False,
 156                                       note=f'Download subtitle for locale {sub.get("name")} ({locale})') or {}
 157             sub_url = url_or_none(sub.get('info'))
 158             if not sub_url:
 159                 continue
 160             subtitles.setdefault(locale.lower(), []).append({
 161                 'url': sub_url,
 162                 'name': sub.get('name'),
 163                 'ext': 'srt',
 164             })
 165         return subtitles