yt_dlp/extractor/gopro.py

   1 from .common import InfoExtractor
   2 from ..utils import (
   3     int_or_none,
   4     remove_end,
   5     str_or_none,
   6     try_get,
   7     unified_timestamp,
   8     url_or_none,
   9 )
  10
  11
  12 class GoProIE(InfoExtractor):
  13     _VALID_URL = r'https?://(www\.)?gopro\.com/v/(?P<id>[A-Za-z0-9]+)'
  14
  15     _TESTS = [{
  16         'url': 'https://gopro.com/v/ZNVvED8QDzR5V',
  17         'info_dict': {
  18             'id': 'ZNVvED8QDzR5V',
  19             'title': 'My GoPro Adventure - 9/19/21',
  20             'thumbnail': r're:https?://.+',
  21             'ext': 'mp4',
  22             'timestamp': 1632072947,
  23             'upload_date': '20210919',
  24             'uploader_id': 'fireydive30018',
  25             'duration': 396062,
  26         },
  27     }, {
  28         'url': 'https://gopro.com/v/KRm6Vgp2peg4e',
  29         'info_dict': {
  30             'id': 'KRm6Vgp2peg4e',
  31             'title': 'じゃがいも カリカリ オーブン焼き',
  32             'thumbnail': r're:https?://.+',
  33             'ext': 'mp4',
  34             'timestamp': 1607231125,
  35             'upload_date': '20201206',
  36             'uploader_id': 'dc9bcb8b-47d2-47c6-afbc-4c48f9a3769e',
  37             'duration': 45187,
  38             'track': 'The Sky Machine',
  39         },
  40     }, {
  41         'url': 'https://gopro.com/v/kVrK9wlJvBMwn',
  42         'info_dict': {
  43             'id': 'kVrK9wlJvBMwn',
  44             'title': 'DARKNESS',
  45             'thumbnail': r're:https?://.+',
  46             'ext': 'mp4',
  47             'timestamp': 1594183735,
  48             'upload_date': '20200708',
  49             'uploader_id': '闇夜乃皇帝',
  50             'duration': 313075,
  51             'track': 'Battery (Live)',
  52             'artist': 'Metallica',
  53         },
  54     }]
  55
  56     def _real_extract(self, url):
  57         video_id = self._match_id(url)
  58         webpage = self._download_webpage(url, video_id)
  59
  60         metadata = self._search_json(
  61             r'window\.__reflectData\s*=', webpage, 'metadata', video_id)
  62
  63         video_info = metadata['collectionMedia'][0]
  64         media_data = self._download_json(
  65             'https://api.gopro.com/media/{}/download'.format(video_info['id']), video_id)
  66
  67         formats = []
  68         for fmt in try_get(media_data, lambda x: x['_embedded']['variations']) or []:
  69             format_url = url_or_none(fmt.get('url'))
  70             if not format_url:
  71                 continue
  72             formats.append({
  73                 'url': format_url,
  74                 'format_id': str_or_none(fmt.get('quality')),
  75                 'format_note': str_or_none(fmt.get('label')),
  76                 'ext': str_or_none(fmt.get('type')),
  77                 'width': int_or_none(fmt.get('width')),
  78                 'height': int_or_none(fmt.get('height')),
  79             })
  80
  81         title = str_or_none(
  82             try_get(metadata, lambda x: x['collection']['title'])
  83             or self._html_search_meta(['og:title', 'twitter:title'], webpage)
  84             or remove_end(self._html_search_regex(
  85                 r'<title[^>]*>([^<]+)</title>', webpage, 'title', fatal=False), ' | GoPro'))
  86         if title:
  87             title = title.replace('\n', ' ')
  88
  89         return {
  90             'id': video_id,
  91             'title': title,
  92             'formats': formats,
  93             'thumbnail': url_or_none(
  94                 self._html_search_meta(['og:image', 'twitter:image'], webpage)),
  95             'timestamp': unified_timestamp(
  96                 try_get(metadata, lambda x: x['collection']['created_at'])),
  97             'uploader_id': str_or_none(
  98                 try_get(metadata, lambda x: x['account']['nickname'])),
  99             'duration': int_or_none(
 100                 video_info.get('source_duration')),
 101             'artist': str_or_none(
 102                 video_info.get('music_track_artist')) or None,
 103             'track': str_or_none(
 104                 video_info.get('music_track_name')) or None,
 105         }