yt_dlp/extractor/wistia.py

   1 import re
   2
   3 from .common import InfoExtractor
   4 from ..utils import (
   5     ExtractorError,
   6     float_or_none,
   7     int_or_none,
   8     try_call,
   9     try_get,
  10 )
  11
  12
  13 class WistiaBaseIE(InfoExtractor):
  14     _VALID_ID_REGEX = r'(?P<id>[a-z0-9]{10})'
  15     _VALID_URL_BASE = r'https?://(?:\w+\.)?wistia\.(?:net|com)/(?:embed/)?'
  16     _EMBED_BASE_URL = 'http://fast.wistia.com/embed/'
  17
  18     def _download_embed_config(self, config_type, config_id, referer):
  19         base_url = self._EMBED_BASE_URL + '%ss/%s' % (config_type, config_id)
  20         embed_config = self._download_json(
  21             base_url + '.json', config_id, headers={
  22                 'Referer': referer if referer.startswith('http') else base_url,  # Some videos require this.
  23             })
  24
  25         if isinstance(embed_config, dict) and embed_config.get('error'):
  26             raise ExtractorError(
  27                 'Error while getting the playlist', expected=True)
  28
  29         return embed_config
  30
  31     def _extract_media(self, embed_config):
  32         data = embed_config['media']
  33         video_id = data['hashedId']
  34         title = data['name']
  35
  36         formats = []
  37         thumbnails = []
  38         for a in data['assets']:
  39             aurl = a.get('url')
  40             if not aurl:
  41                 continue
  42             astatus = a.get('status')
  43             atype = a.get('type')
  44             if (astatus is not None and astatus != 2) or atype in ('preview', 'storyboard'):
  45                 continue
  46             elif atype in ('still', 'still_image'):
  47                 thumbnails.append({
  48                     'url': aurl,
  49                     'width': int_or_none(a.get('width')),
  50                     'height': int_or_none(a.get('height')),
  51                     'filesize': int_or_none(a.get('size')),
  52                 })
  53             else:
  54                 aext = a.get('ext')
  55                 display_name = a.get('display_name')
  56                 format_id = atype
  57                 if atype and atype.endswith('_video') and display_name:
  58                     format_id = '%s-%s' % (atype[:-6], display_name)
  59                 f = {
  60                     'format_id': format_id,
  61                     'url': aurl,
  62                     'tbr': int_or_none(a.get('bitrate')) or None,
  63                     'quality': 1 if atype == 'original' else None,
  64                 }
  65                 if display_name == 'Audio':
  66                     f.update({
  67                         'vcodec': 'none',
  68                     })
  69                 else:
  70                     f.update({
  71                         'width': int_or_none(a.get('width')),
  72                         'height': int_or_none(a.get('height')),
  73                         'vcodec': a.get('codec'),
  74                     })
  75                 if a.get('container') == 'm3u8' or aext == 'm3u8':
  76                     ts_f = f.copy()
  77                     ts_f.update({
  78                         'ext': 'ts',
  79                         'format_id': f['format_id'].replace('hls-', 'ts-'),
  80                         'url': f['url'].replace('.bin', '.ts'),
  81                     })
  82                     formats.append(ts_f)
  83                     f.update({
  84                         'ext': 'mp4',
  85                         'protocol': 'm3u8_native',
  86                     })
  87                 else:
  88                     f.update({
  89                         'container': a.get('container'),
  90                         'ext': aext,
  91                         'filesize': int_or_none(a.get('size')),
  92                     })
  93                 formats.append(f)
  94
  95         self._sort_formats(formats)
  96
  97         subtitles = {}
  98         for caption in data.get('captions', []):
  99             language = caption.get('language')
 100             if not language:
 101                 continue
 102             subtitles[language] = [{
 103                 'url': self._EMBED_BASE_URL + 'captions/' + video_id + '.vtt?language=' + language,
 104             }]
 105
 106         return {
 107             'id': video_id,
 108             'title': title,
 109             'description': data.get('seoDescription'),
 110             'formats': formats,
 111             'thumbnails': thumbnails,
 112             'duration': float_or_none(data.get('duration')),
 113             'timestamp': int_or_none(data.get('createdAt')),
 114             'subtitles': subtitles,
 115         }
 116
 117
 118 class WistiaIE(WistiaBaseIE):
 119     _VALID_URL = r'(?:wistia:|%s(?:iframe|medias)/)%s' % (WistiaBaseIE._VALID_URL_BASE, WistiaBaseIE._VALID_ID_REGEX)
 120     _EMBED_REGEX = [r'<(?:meta[^>]+?content|(?:iframe|script)[^>]+?src)=["\'](?P<url>(?:https?:)?//(?:fast\.)?wistia\.(?:net|com)/embed/(?:iframe|medias)/[a-z0-9]{10})']
 121     _TESTS = [{
 122         # with hls video
 123         'url': 'wistia:807fafadvk',
 124         'md5': 'daff0f3687a41d9a71b40e0e8c2610fe',
 125         'info_dict': {
 126             'id': '807fafadvk',
 127             'ext': 'mp4',
 128             'title': 'Drip Brennan Dunn Workshop',
 129             'description': 'a JV Webinars video',
 130             'upload_date': '20160518',
 131             'timestamp': 1463607249,
 132             'duration': 4987.11,
 133         },
 134     }, {
 135         'url': 'wistia:sh7fpupwlt',
 136         'only_matching': True,
 137     }, {
 138         'url': 'http://fast.wistia.net/embed/iframe/sh7fpupwlt',
 139         'only_matching': True,
 140     }, {
 141         'url': 'http://fast.wistia.com/embed/iframe/sh7fpupwlt',
 142         'only_matching': True,
 143     }, {
 144         'url': 'http://fast.wistia.net/embed/medias/sh7fpupwlt.json',
 145         'only_matching': True,
 146     }]
 147
 148     # https://wistia.com/support/embed-and-share/video-on-your-website
 149     @classmethod
 150     def _extract_embed_urls(cls, url, webpage):
 151         urls = list(super()._extract_embed_urls(url, webpage))
 152
 153         for match in re.finditer(
 154                 r'''(?sx)
 155                     <div[^>]+class=(["'])(?:(?!\1).)*?\bwistia_async_(?P<id>[a-z0-9]{10})\b(?:(?!\1).)*?\1
 156                 ''', webpage):
 157             urls.append('wistia:%s' % match.group('id'))
 158         for match in re.finditer(r'(?:data-wistia-?id=["\']|Wistia\.embed\(["\']|id=["\']wistia_)(?P<id>[a-z0-9]{10})', webpage):
 159             urls.append('wistia:%s' % match.group('id'))
 160         return urls
 161
 162     @classmethod
 163     def _extract_from_webpage(cls, url, webpage):
 164         from .teachable import TeachableIE
 165
 166         if list(TeachableIE._extract_embed_urls(url, webpage)):
 167             return
 168
 169         for entry in super()._extract_from_webpage(url, webpage):
 170             yield {
 171                 **entry,
 172                 '_type': 'url_transparent',
 173                 'uploader': try_call(lambda: re.match(r'(?:https?://)?([^/]+)/', url).group(1)),
 174             }
 175
 176     def _real_extract(self, url):
 177         video_id = self._match_id(url)
 178         embed_config = self._download_embed_config('media', video_id, url)
 179         return self._extract_media(embed_config)
 180
 181
 182 class WistiaPlaylistIE(WistiaBaseIE):
 183     _VALID_URL = r'%splaylists/%s' % (WistiaBaseIE._VALID_URL_BASE, WistiaBaseIE._VALID_ID_REGEX)
 184
 185     _TEST = {
 186         'url': 'https://fast.wistia.net/embed/playlists/aodt9etokc',
 187         'info_dict': {
 188             'id': 'aodt9etokc',
 189         },
 190         'playlist_count': 3,
 191     }
 192
 193     def _real_extract(self, url):
 194         playlist_id = self._match_id(url)
 195         playlist = self._download_embed_config('playlist', playlist_id, url)
 196
 197         entries = []
 198         for media in (try_get(playlist, lambda x: x[0]['medias']) or []):
 199             embed_config = media.get('embed_config')
 200             if not embed_config:
 201                 continue
 202             entries.append(self._extract_media(embed_config))
 203
 204         return self.playlist_result(entries, playlist_id)