yt_dlp/extractor/amadeustv.py

   1 from .common import InfoExtractor
   2 from ..utils import (
   3     ExtractorError,
   4     float_or_none,
   5     int_or_none,
   6     parse_iso8601,
   7     url_or_none,
   8 )
   9 from ..utils.traversal import traverse_obj
  10
  11
  12 class AmadeusTVIE(InfoExtractor):
  13     _VALID_URL = r'https?://(?:www\.)?amadeus\.tv/library/(?P<id>[\da-f]+)'
  14     _TESTS = [{
  15         'url': 'http://www.amadeus.tv/library/65091a87ff85af59d9fc54c3',
  16         'info_dict': {
  17             'id': '5576678021301411311',
  18             'ext': 'mp4',
  19             'title': 'Jieon Park - 第五届珠海莫扎特国际青少年音乐周小提琴C组第三轮',
  20             'thumbnail': 'http://1253584441.vod2.myqcloud.com/a0046a27vodtransbj1253584441/7db4af535576678021301411311/coverBySnapshot_10_0.jpg',
  21             'duration': 1264.8,
  22             'upload_date': '20230918',
  23             'timestamp': 1695034800,
  24             'display_id': '65091a87ff85af59d9fc54c3',
  25             'view_count': int,
  26             'description': 'md5:a0357b9c215489e2067cbae0b777bb95',
  27         },
  28     }]
  29
  30     def _real_extract(self, url):
  31         display_id = self._match_id(url)
  32         webpage = self._download_webpage(url, display_id)
  33
  34         nuxt_data = self._search_nuxt_data(webpage, display_id, traverse=('fetch', '0'))
  35         video_id = traverse_obj(nuxt_data, ('item', 'video', {str}))
  36
  37         if not video_id:
  38             raise ExtractorError('Unable to extract actual video ID')
  39
  40         video_data = self._download_json(
  41             f'http://playvideo.qcloud.com/getplayinfo/v2/1253584441/{video_id}',
  42             video_id, headers={'Referer': 'http://www.amadeus.tv/'})
  43
  44         formats = []
  45         for video in traverse_obj(video_data, ('videoInfo', ('sourceVideo', ('transcodeList', ...)), {dict})):
  46             if not url_or_none(video.get('url')):
  47                 continue
  48             formats.append({
  49                 **traverse_obj(video, {
  50                     'url': 'url',
  51                     'format_id': ('definition', {lambda x: f'http-{x or "0"}'}),
  52                     'width': ('width', {int_or_none}),
  53                     'height': ('height', {int_or_none}),
  54                     'filesize': (('totalSize', 'size'), {int_or_none}),
  55                     'vcodec': ('videoStreamList', 0, 'codec'),
  56                     'acodec': ('audioStreamList', 0, 'codec'),
  57                     'fps': ('videoStreamList', 0, 'fps', {float_or_none}),
  58                 }, get_all=False),
  59                 'http_headers': {'Referer': 'http://www.amadeus.tv/'},
  60             })
  61
  62         return {
  63             'id': video_id,
  64             'display_id': display_id,
  65             'formats': formats,
  66             **traverse_obj(video_data, {
  67                 'title': ('videoInfo', 'basicInfo', 'name', {str}),
  68                 'thumbnail': ('coverInfo', 'coverUrl', {url_or_none}),
  69                 'duration': ('videoInfo', 'sourceVideo', ('floatDuration', 'duration'), {float_or_none}),
  70             }, get_all=False),
  71             **traverse_obj(nuxt_data, ('item', {
  72                 'title': (('title', 'title_en', 'title_cn'), {str}),
  73                 'description': (('description', 'description_en', 'description_cn'), {str}),
  74                 'timestamp': ('date', {parse_iso8601}),
  75                 'view_count': ('view', {int_or_none}),
  76             }), get_all=False),
  77         }