yt_dlp/extractor/voxmedia.py

   1 from .common import InfoExtractor
   2 from .once import OnceIE
   3 from ..compat import compat_urllib_parse_unquote
   4 from ..utils import (
   5     ExtractorError,
   6     int_or_none,
   7     try_get,
   8     unified_timestamp,
   9 )
  10
  11
  12 class VoxMediaVolumeIE(OnceIE):
  13     _VALID_URL = r'https?://volume\.vox-cdn\.com/embed/(?P<id>[0-9a-f]{9})'
  14
  15     def _real_extract(self, url):
  16         video_id = self._match_id(url)
  17         webpage = self._download_webpage(url, video_id)
  18
  19         setup = self._parse_json(self._search_regex(
  20             r'setup\s*=\s*({.+});', webpage, 'setup'), video_id)
  21         player_setup = setup.get('player_setup') or setup
  22         video_data = player_setup.get('video') or {}
  23         formatted_metadata = video_data.get('formatted_metadata') or {}
  24         info = {
  25             'id': video_id,
  26             'title': player_setup.get('title') or video_data.get('title_short'),
  27             'description': video_data.get('description_long') or video_data.get('description_short'),
  28             'thumbnail': formatted_metadata.get('thumbnail') or video_data.get('brightcove_thumbnail'),
  29             'timestamp': unified_timestamp(formatted_metadata.get('video_publish_date')),
  30         }
  31         asset = try_get(setup, lambda x: x['embed_assets']['chorus'], dict) or {}
  32
  33         formats = []
  34         hls_url = asset.get('hls_url')
  35         if hls_url:
  36             formats.extend(self._extract_m3u8_formats(
  37                 hls_url, video_id, 'mp4', 'm3u8_native', m3u8_id='hls', fatal=False))
  38         mp4_url = asset.get('mp4_url')
  39         if mp4_url:
  40             tbr = self._search_regex(r'-(\d+)k\.', mp4_url, 'bitrate', default=None)
  41             format_id = 'http'
  42             if tbr:
  43                 format_id += '-' + tbr
  44             formats.append({
  45                 'format_id': format_id,
  46                 'url': mp4_url,
  47                 'tbr': int_or_none(tbr),
  48             })
  49         if formats:
  50             self._sort_formats(formats)
  51             info['formats'] = formats
  52             info['duration'] = int_or_none(asset.get('duration'))
  53             return info
  54
  55         for provider_video_type in ('ooyala', 'youtube', 'brightcove'):
  56             provider_video_id = video_data.get('%s_id' % provider_video_type)
  57             if not provider_video_id:
  58                 continue
  59             if provider_video_type == 'brightcove':
  60                 info['formats'] = self._extract_once_formats(provider_video_id)
  61                 self._sort_formats(info['formats'])
  62             else:
  63                 info.update({
  64                     '_type': 'url_transparent',
  65                     'url': provider_video_id if provider_video_type == 'youtube' else '%s:%s' % (provider_video_type, provider_video_id),
  66                     'ie_key': provider_video_type.capitalize(),
  67                 })
  68             return info
  69         raise ExtractorError('Unable to find provider video id')
  70
  71
  72 class VoxMediaIE(InfoExtractor):
  73     _VALID_URL = r'https?://(?:www\.)?(?:(?:theverge|vox|sbnation|eater|polygon|curbed|racked|funnyordie)\.com|recode\.net)/(?:[^/]+/)*(?P<id>[^/?]+)'
  74     _EMBED_REGEX = [r'<iframe[^>]+?src="(?P<url>https?://(?:www\.)?funnyordie\.com/embed/[^"]+)"']
  75     _TESTS = [{
  76         # Volume embed, Youtube
  77         'url': 'http://www.theverge.com/2014/6/27/5849272/material-world-how-google-discovered-what-software-is-made-of',
  78         'info_dict': {
  79             'id': 'j4mLW6x17VM',
  80             'ext': 'mp4',
  81             'title': 'Material world: how Google discovered what software is made of',
  82             'description': 'md5:dfc17e7715e3b542d66e33a109861382',
  83             'upload_date': '20190710',
  84             'uploader_id': 'TheVerge',
  85             'uploader': 'The Verge',
  86         },
  87         'add_ie': ['Youtube'],
  88     }, {
  89         # Volume embed, Youtube
  90         'url': 'http://www.theverge.com/2014/10/21/7025853/google-nexus-6-hands-on-photos-video-android-phablet',
  91         'md5': 'fd19aa0cf3a0eea515d4fd5c8c0e9d68',
  92         'info_dict': {
  93             'id': 'Gy8Md3Eky38',
  94             'ext': 'mp4',
  95             'title': 'The Nexus 6: hands-on with Google\'s phablet',
  96             'description': 'md5:d9f0216e5fb932dd2033d6db37ac3f1d',
  97             'uploader_id': 'TheVerge',
  98             'upload_date': '20141021',
  99             'uploader': 'The Verge',
 100             'timestamp': 1413907200,
 101         },
 102         'add_ie': ['Youtube'],
 103         'skip': 'similar to the previous test',
 104     }, {
 105         # Volume embed, Youtube
 106         'url': 'http://www.vox.com/2016/3/31/11336640/mississippi-lgbt-religious-freedom-bill',
 107         'info_dict': {
 108             'id': '22986359b',
 109             'ext': 'mp4',
 110             'title': "Mississippi's laws are so bad that its anti-LGBTQ law isn't needed to allow discrimination",
 111             'description': 'md5:fc1317922057de31cd74bce91eb1c66c',
 112             'upload_date': '20150915',
 113             'timestamp': 1442332800,
 114             'duration': 285,
 115         },
 116         'add_ie': ['Youtube'],
 117         'skip': 'similar to the previous test',
 118     }, {
 119         # youtube embed
 120         'url': 'http://www.vox.com/2016/3/24/11291692/robot-dance',
 121         'md5': '83b3080489fb103941e549352d3e0977',
 122         'info_dict': {
 123             'id': 'FcNHTJU1ufM',
 124             'ext': 'mp4',
 125             'title': 'How "the robot" became the greatest novelty dance of all time',
 126             'description': 'md5:b081c0d588b8b2085870cda55e6da176',
 127             'upload_date': '20160324',
 128             'uploader_id': 'voxdotcom',
 129             'uploader': 'Vox',
 130         },
 131         'add_ie': ['Youtube'],
 132         'skip': 'Page no longer contain videos',
 133     }, {
 134         # SBN.VideoLinkset.entryGroup multiple ooyala embeds
 135         'url': 'http://www.sbnation.com/college-football-recruiting/2015/2/3/7970291/national-signing-day-rationalizations-itll-be-ok-itll-be-ok',
 136         'info_dict': {
 137             'id': 'national-signing-day-rationalizations-itll-be-ok-itll-be-ok',
 138             'title': '25 lies you will tell yourself on National Signing Day',
 139             'description': 'It\'s the most self-delusional time of the year, and everyone\'s gonna tell the same lies together!',
 140         },
 141         'playlist': [{
 142             'md5': '721fededf2ab74ae4176c8c8cbfe092e',
 143             'info_dict': {
 144                 'id': 'p3cThlMjE61VDi_SD9JlIteSNPWVDBB9',
 145                 'ext': 'mp4',
 146                 'title': 'Buddy Hield vs Steph Curry (and the world)',
 147                 'description': 'Let’s dissect only the most important Final Four storylines.',
 148             },
 149         }, {
 150             'md5': 'bf0c5cc115636af028be1bab79217ea9',
 151             'info_dict': {
 152                 'id': 'BmbmVjMjE6esPHxdALGubTrouQ0jYLHj',
 153                 'ext': 'mp4',
 154                 'title': 'Chasing Cinderella 2016: Syracuse basketball',
 155                 'description': 'md5:e02d56b026d51aa32c010676765a690d',
 156             },
 157         }],
 158         'skip': 'Page no longer contain videos',
 159     }, {
 160         # volume embed, Brightcove Once
 161         'url': 'https://www.recode.net/2014/6/17/11628066/post-post-pc-ceo-the-full-code-conference-video-of-microsofts-satya',
 162         'md5': '2dbc77b8b0bff1894c2fce16eded637d',
 163         'info_dict': {
 164             'id': '1231c973d',
 165             'ext': 'mp4',
 166             'title': 'Post-Post-PC CEO: The Full Code Conference Video of Microsoft\'s Satya Nadella',
 167             'description': 'The longtime veteran was chosen earlier this year as the software giant\'s third leader in its history.',
 168             'timestamp': 1402938000,
 169             'upload_date': '20140616',
 170             'duration': 4114,
 171         },
 172         'add_ie': ['VoxMediaVolume'],
 173     }]
 174
 175     def _real_extract(self, url):
 176         display_id = self._match_id(url)
 177         webpage = compat_urllib_parse_unquote(self._download_webpage(url, display_id))
 178
 179         def create_entry(provider_video_id, provider_video_type, title=None, description=None):
 180             video_url = {
 181                 'youtube': '%s',
 182                 'ooyala': 'ooyala:%s',
 183                 'volume': 'http://volume.vox-cdn.com/embed/%s',
 184             }[provider_video_type] % provider_video_id
 185             return {
 186                 '_type': 'url_transparent',
 187                 'url': video_url,
 188                 'title': title or self._og_search_title(webpage),
 189                 'description': description or self._og_search_description(webpage),
 190             }
 191
 192         entries = []
 193         entries_data = self._search_regex([
 194             r'Chorus\.VideoContext\.addVideo\((\[{.+}\])\);',
 195             r'var\s+entry\s*=\s*({.+});',
 196             r'SBN\.VideoLinkset\.entryGroup\(\s*(\[.+\])',
 197         ], webpage, 'video data', default=None)
 198         if entries_data:
 199             entries_data = self._parse_json(entries_data, display_id)
 200             if isinstance(entries_data, dict):
 201                 entries_data = [entries_data]
 202             for video_data in entries_data:
 203                 provider_video_id = video_data.get('provider_video_id')
 204                 provider_video_type = video_data.get('provider_video_type')
 205                 if provider_video_id and provider_video_type:
 206                     entries.append(create_entry(
 207                         provider_video_id, provider_video_type,
 208                         video_data.get('title'), video_data.get('description')))
 209
 210         provider_video_id = self._search_regex(
 211             r'data-ooyala-id="([^"]+)"', webpage, 'ooyala id', default=None)
 212         if provider_video_id:
 213             entries.append(create_entry(provider_video_id, 'ooyala'))
 214
 215         volume_uuid = self._search_regex(
 216             r'data-volume-uuid="([^"]+)"', webpage, 'volume uuid', default=None)
 217         if volume_uuid:
 218             entries.append(create_entry(volume_uuid, 'volume'))
 219
 220         if len(entries) == 1:
 221             return entries[0]
 222         else:
 223             return self.playlist_result(entries, display_id, self._og_search_title(webpage), self._og_search_description(webpage))