yt_dlp/extractor/wistia.py

   1 import base64
   2 import re
   3 import urllib.parse
   4
   5 from .common import InfoExtractor
   6 from ..networking import HEADRequest
   7 from ..networking.exceptions import HTTPError
   8 from ..utils import (
   9     ExtractorError,
  10     determine_ext,
  11     filter_dict,
  12     float_or_none,
  13     int_or_none,
  14     parse_qs,
  15     traverse_obj,
  16     try_get,
  17     update_url_query,
  18     urlhandle_detect_ext,
  19 )
  20
  21
  22 class WistiaBaseIE(InfoExtractor):
  23     _VALID_ID_REGEX = r'(?P<id>[a-z0-9]{10})'
  24     _VALID_URL_BASE = r'https?://(?:\w+\.)?wistia\.(?:net|com)/(?:embed/)?'
  25     _EMBED_BASE_URL = 'http://fast.wistia.net/embed/'
  26
  27     def _download_embed_config(self, config_type, config_id, referer):
  28         base_url = self._EMBED_BASE_URL + f'{config_type}/{config_id}'
  29         video_password = self.get_param('videopassword')
  30         embed_config = self._download_json(
  31             base_url + '.json', config_id, headers={
  32                 'Referer': referer if referer.startswith('http') else base_url,  # Some videos require this.
  33             }, query=filter_dict({'password': video_password}))
  34
  35         error = traverse_obj(embed_config, 'error')
  36         if error:
  37             raise ExtractorError(
  38                 f'Error while getting the playlist: {error}', expected=True)
  39
  40         if traverse_obj(embed_config, (
  41                 'media', ('embed_options', 'embedOptions'), 'plugin',
  42                 'passwordProtectedVideo', 'on', any)) == 'true':
  43             if video_password:
  44                 raise ExtractorError('Invalid video password', expected=True)
  45             raise ExtractorError(
  46                 'This content is password-protected. Use the --video-password option', expected=True)
  47
  48         return embed_config
  49
  50     def _get_real_ext(self, url):
  51         ext = determine_ext(url, default_ext='bin')
  52         if ext == 'bin':
  53             urlh = self._request_webpage(
  54                 HEADRequest(url), None, note='Checking media extension',
  55                 errnote='HEAD request returned error', fatal=False)
  56             if urlh:
  57                 ext = urlhandle_detect_ext(urlh, default='bin')
  58         return 'mp4' if ext == 'mov' else ext
  59
  60     def _extract_media(self, embed_config):
  61         data = embed_config['media']
  62         video_id = data['hashedId']
  63         title = data['name']
  64
  65         formats = []
  66         thumbnails = []
  67         for a in data['assets']:
  68             aurl = a.get('url')
  69             if not aurl:
  70                 continue
  71             astatus = a.get('status')
  72             atype = a.get('type')
  73             if (astatus is not None and astatus != 2) or atype in ('preview', 'storyboard'):
  74                 continue
  75             elif atype in ('still', 'still_image'):
  76                 thumbnails.append({
  77                     'url': aurl.replace('.bin', f'.{self._get_real_ext(aurl)}'),
  78                     'width': int_or_none(a.get('width')),
  79                     'height': int_or_none(a.get('height')),
  80                     'filesize': int_or_none(a.get('size')),
  81                 })
  82             else:
  83                 aext = a.get('ext') or self._get_real_ext(aurl)
  84                 display_name = a.get('display_name')
  85                 format_id = atype
  86                 if atype and atype.endswith('_video') and display_name:
  87                     format_id = f'{atype[:-6]}-{display_name}'
  88                 f = {
  89                     'format_id': format_id,
  90                     'url': aurl,
  91                     'tbr': int_or_none(a.get('bitrate')) or None,
  92                     'quality': 1 if atype == 'original' else None,
  93                 }
  94                 if display_name == 'Audio':
  95                     f.update({
  96                         'vcodec': 'none',
  97                     })
  98                 else:
  99                     f.update({
 100                         'width': int_or_none(a.get('width')),
 101                         'height': int_or_none(a.get('height')),
 102                         'vcodec': a.get('codec'),
 103                     })
 104                 if a.get('container') == 'm3u8' or aext == 'm3u8':
 105                     ts_f = f.copy()
 106                     ts_f.update({
 107                         'ext': 'ts',
 108                         'format_id': f['format_id'].replace('hls-', 'ts-'),
 109                         'url': f['url'].replace('.bin', '.ts'),
 110                     })
 111                     formats.append(ts_f)
 112                     f.update({
 113                         'ext': 'mp4',
 114                         'protocol': 'm3u8_native',
 115                     })
 116                 else:
 117                     f.update({
 118                         'container': a.get('container'),
 119                         'ext': aext,
 120                         'filesize': int_or_none(a.get('size')),
 121                     })
 122                 formats.append(f)
 123
 124         subtitles = {}
 125         for caption in data.get('captions', []):
 126             language = caption.get('language')
 127             if not language:
 128                 continue
 129             subtitles[language] = [{
 130                 'url': self._EMBED_BASE_URL + 'captions/' + video_id + '.vtt?language=' + language,
 131             }]
 132
 133         return {
 134             'id': video_id,
 135             'title': title,
 136             'description': data.get('seoDescription'),
 137             'formats': formats,
 138             'thumbnails': thumbnails,
 139             'duration': float_or_none(data.get('duration')),
 140             'timestamp': int_or_none(data.get('createdAt')),
 141             'subtitles': subtitles,
 142         }
 143
 144     @classmethod
 145     def _extract_from_webpage(cls, url, webpage):
 146         from .teachable import TeachableIE
 147
 148         if list(TeachableIE._extract_embed_urls(url, webpage)):
 149             return
 150
 151         yield from super()._extract_from_webpage(url, webpage)
 152
 153     @classmethod
 154     def _extract_wistia_async_embed(cls, webpage):
 155         # https://wistia.com/support/embed-and-share/video-on-your-website
 156         # https://wistia.com/support/embed-and-share/channel-embeds
 157         yield from re.finditer(
 158             r'''(?sx)
 159                 <(?:div|section)[^>]+class=([\"'])(?:(?!\1).)*?(?P<type>wistia[a-z_0-9]+)\s*\bwistia_async_(?P<id>[a-z0-9]{10})\b(?:(?!\1).)*?\1
 160             ''', webpage)
 161
 162     @classmethod
 163     def _extract_url_media_id(cls, url):
 164         mobj = re.search(r'(?:wmediaid|wvideo(?:id)?)]?=(?P<id>[a-z0-9]{10})', urllib.parse.unquote_plus(url))
 165         if mobj:
 166             return mobj.group('id')
 167
 168
 169 class WistiaIE(WistiaBaseIE):
 170     _VALID_URL = rf'(?:wistia:|{WistiaBaseIE._VALID_URL_BASE}(?:iframe|medias)/){WistiaBaseIE._VALID_ID_REGEX}'
 171     _EMBED_REGEX = [
 172         r'''(?x)
 173             <(?:meta[^>]+?content|(?:iframe|script)[^>]+?src)=["\']
 174             (?P<url>(?:https?:)?//(?:fast\.)?wistia\.(?:net|com)/embed/(?:iframe|medias)/[a-z0-9]{10})
 175             ''']
 176     _TESTS = [{
 177         # with hls video
 178         'url': 'wistia:807fafadvk',
 179         'md5': 'daff0f3687a41d9a71b40e0e8c2610fe',
 180         'info_dict': {
 181             'id': '807fafadvk',
 182             'ext': 'mp4',
 183             'title': 'Drip Brennan Dunn Workshop',
 184             'description': 'a JV Webinars video',
 185             'upload_date': '20160518',
 186             'timestamp': 1463607249,
 187             'duration': 4987.11,
 188         },
 189         'skip': 'video unavailable',
 190     }, {
 191         'url': 'wistia:a6ndpko1wg',
 192         'md5': '10c1ce9c4dde638202513ed17a3767bd',
 193         'info_dict': {
 194             'id': 'a6ndpko1wg',
 195             'ext': 'mp4',
 196             'title': 'Episode 2: Boxed Water\'s retention is thirsty',
 197             'upload_date': '20210324',
 198             'description': 'md5:da5994c2c2d254833b412469d9666b7a',
 199             'duration': 966.0,
 200             'timestamp': 1616614369,
 201             'thumbnail': 'https://embed-ssl.wistia.com/deliveries/53dc60239348dc9b9fba3755173ea4c2.png',
 202         },
 203     }, {
 204         'url': 'wistia:5vd7p4bct5',
 205         'md5': 'b9676d24bf30945d97060638fbfe77f0',
 206         'info_dict': {
 207             'id': '5vd7p4bct5',
 208             'ext': 'mp4',
 209             'title': 'md5:eaa9f64c4efd7b5f098b9b6118597679',
 210             'description': 'md5:a9bea0315f0616aa5df2dc413ddcdd0f',
 211             'upload_date': '20220915',
 212             'timestamp': 1663258727,
 213             'duration': 623.019,
 214             'thumbnail': r're:https?://embed(?:-ssl)?.wistia.com/.+\.jpg$',
 215         },
 216     }, {
 217         'url': 'wistia:sh7fpupwlt',
 218         'only_matching': True,
 219     }, {
 220         'url': 'http://fast.wistia.net/embed/iframe/sh7fpupwlt',
 221         'only_matching': True,
 222     }, {
 223         'url': 'http://fast.wistia.com/embed/iframe/sh7fpupwlt',
 224         'only_matching': True,
 225     }, {
 226         'url': 'http://fast.wistia.net/embed/medias/sh7fpupwlt.json',
 227         'only_matching': True,
 228     }]
 229
 230     _WEBPAGE_TESTS = [{
 231         'url': 'https://www.weidert.com/blog/wistia-channels-video-marketing-tool',
 232         'info_dict': {
 233             'id': 'cqwukac3z1',
 234             'ext': 'mp4',
 235             'title': 'How Wistia Channels Can Help Capture Inbound Value From Your Video Content',
 236             'duration': 158.125,
 237             'timestamp': 1618974400,
 238             'description': 'md5:27abc99a758573560be72600ef95cece',
 239             'upload_date': '20210421',
 240             'thumbnail': 'https://embed-ssl.wistia.com/deliveries/6c551820ae950cdee2306d6cbe9ef742.jpg',
 241         },
 242     }, {
 243         'url': 'https://study.com/academy/lesson/north-american-exploration-failed-colonies-of-spain-france-england.html#lesson',
 244         'md5': 'b9676d24bf30945d97060638fbfe77f0',
 245         'info_dict': {
 246             'id': '5vd7p4bct5',
 247             'ext': 'mp4',
 248             'title': 'paywall_north-american-exploration-failed-colonies-of-spain-france-england',
 249             'upload_date': '20220915',
 250             'timestamp': 1663258727,
 251             'duration': 623.019,
 252             'thumbnail': 'https://embed-ssl.wistia.com/deliveries/83e6ec693e2c05a0ce65809cbaead86a.jpg',
 253             'description': 'a Paywall Videos video',
 254         },
 255     }]
 256
 257     def _real_extract(self, url):
 258         video_id = self._match_id(url)
 259         embed_config = self._download_embed_config('medias', video_id, url)
 260         return self._extract_media(embed_config)
 261
 262     @classmethod
 263     def _extract_embed_urls(cls, url, webpage):
 264         urls = list(super()._extract_embed_urls(url, webpage))
 265         for match in cls._extract_wistia_async_embed(webpage):
 266             if match.group('type') != 'wistia_channel':
 267                 urls.append('wistia:{}'.format(match.group('id')))
 268         for match in re.finditer(r'(?:data-wistia-?id=["\']|Wistia\.embed\(["\']|id=["\']wistia_)(?P<id>[a-z0-9]{10})',
 269                                  webpage):
 270             urls.append('wistia:{}'.format(match.group('id')))
 271         if not WistiaChannelIE._extract_embed_urls(url, webpage):  # Fallback
 272             media_id = cls._extract_url_media_id(url)
 273             if media_id:
 274                 urls.append('wistia:{}'.format(match.group('id')))
 275         return urls
 276
 277
 278 class WistiaPlaylistIE(WistiaBaseIE):
 279     _VALID_URL = rf'{WistiaBaseIE._VALID_URL_BASE}playlists/{WistiaBaseIE._VALID_ID_REGEX}'
 280
 281     _TEST = {
 282         'url': 'https://fast.wistia.net/embed/playlists/aodt9etokc',
 283         'info_dict': {
 284             'id': 'aodt9etokc',
 285         },
 286         'playlist_count': 3,
 287     }
 288
 289     def _real_extract(self, url):
 290         playlist_id = self._match_id(url)
 291         playlist = self._download_embed_config('playlists', playlist_id, url)
 292
 293         entries = []
 294         for media in (try_get(playlist, lambda x: x[0]['medias']) or []):
 295             embed_config = media.get('embed_config')
 296             if not embed_config:
 297                 continue
 298             entries.append(self._extract_media(embed_config))
 299
 300         return self.playlist_result(entries, playlist_id)
 301
 302
 303 class WistiaChannelIE(WistiaBaseIE):
 304     _VALID_URL = rf'(?:wistiachannel:|{WistiaBaseIE._VALID_URL_BASE}channel/){WistiaBaseIE._VALID_ID_REGEX}'
 305
 306     _TESTS = [{
 307         # JSON Embed API returns 403, should fall back to webpage
 308         'url': 'https://fast.wistia.net/embed/channel/yvyvu7wjbg?wchannelid=yvyvu7wjbg',
 309         'info_dict': {
 310             'id': 'yvyvu7wjbg',
 311             'title': 'Copysmith Tutorials and Education!',
 312             'description': 'Learn all things Copysmith via short and informative videos!',
 313         },
 314         'playlist_mincount': 7,
 315         'expected_warnings': ['falling back to webpage'],
 316     }, {
 317         'url': 'https://fast.wistia.net/embed/channel/3802iirk0l',
 318         'info_dict': {
 319             'id': '3802iirk0l',
 320             'title': 'The Roof',
 321         },
 322         'playlist_mincount': 20,
 323     }, {
 324         # link to popup video, follow --no-playlist
 325         'url': 'https://fast.wistia.net/embed/channel/3802iirk0l?wchannelid=3802iirk0l&wmediaid=sp5dqjzw3n',
 326         'info_dict': {
 327             'id': 'sp5dqjzw3n',
 328             'ext': 'mp4',
 329             'title': 'The Roof S2: The Modern CRO',
 330             'thumbnail': 'https://embed-ssl.wistia.com/deliveries/dadfa9233eaa505d5e0c85c23ff70741.png',
 331             'duration': 86.487,
 332             'description': 'A sales leader on The Roof? Man, they really must be letting anyone up here this season.\n',
 333             'timestamp': 1619790290,
 334             'upload_date': '20210430',
 335         },
 336         'params': {'noplaylist': True, 'skip_download': True},
 337     }]
 338     _WEBPAGE_TESTS = [{
 339         'url': 'https://www.profitwell.com/recur/boxed-out',
 340         'info_dict': {
 341             'id': '6jyvmqz6zs',
 342             'title': 'Boxed Out',
 343             'description': 'md5:14a8a93a1dbe236718e6a59f8c8c7bae',
 344         },
 345         'playlist_mincount': 30,
 346     }, {
 347         # section instead of div
 348         'url': 'https://360learning.com/studio/onboarding-joei/',
 349         'info_dict': {
 350             'id': 'z874k93n2o',
 351             'title': 'Onboarding Joei.',
 352             'description': 'Coming to you weekly starting Feb 19th.',
 353         },
 354         'playlist_mincount': 20,
 355     }, {
 356         'url': 'https://amplitude.com/amplify-sessions?amp%5Bwmediaid%5D=pz0m0l0if3&amp%5Bwvideo%5D=pz0m0l0if3&wchannelid=emyjmwjf79&wmediaid=i8um783bdt',
 357         'info_dict': {
 358             'id': 'pz0m0l0if3',
 359             'title': 'A Framework for Improving Product Team Performance',
 360             'ext': 'mp4',
 361             'timestamp': 1653935275,
 362             'upload_date': '20220530',
 363             'description': 'Learn how to help your company improve and achieve your product related goals.',
 364             'duration': 1854.39,
 365             'thumbnail': 'https://embed-ssl.wistia.com/deliveries/12fd19e56413d9d6f04e2185c16a6f8854e25226.png',
 366         },
 367         'params': {'noplaylist': True, 'skip_download': True},
 368     }]
 369
 370     def _real_extract(self, url):
 371         channel_id = self._match_id(url)
 372         media_id = self._extract_url_media_id(url)
 373         if not self._yes_playlist(channel_id, media_id, playlist_label='channel'):
 374             return self.url_result(f'wistia:{media_id}', 'Wistia')
 375
 376         try:
 377             data = self._download_embed_config('channel', channel_id, url)
 378         except (ExtractorError, HTTPError):
 379             # Some channels give a 403 from the JSON API
 380             self.report_warning('Failed to download channel data from API, falling back to webpage.')
 381             webpage = self._download_webpage(f'https://fast.wistia.net/embed/channel/{channel_id}', channel_id)
 382             data = self._parse_json(
 383                 self._search_regex(rf'wchanneljsonp-{channel_id}\'\]\s*=[^\"]*\"([A-Za-z0-9=/]*)', webpage, 'jsonp', channel_id),
 384                 channel_id, transform_source=lambda x: urllib.parse.unquote_plus(base64.b64decode(x).decode('utf-8')))
 385
 386         # XXX: can there be more than one series?
 387         series = traverse_obj(data, ('series', 0), default={})
 388
 389         entries = [
 390             self.url_result(f'wistia:{video["hashedId"]}', WistiaIE, title=video.get('name'))
 391             for video in traverse_obj(series, ('sections', ..., 'videos', ...)) or []
 392             if video.get('hashedId')
 393         ]
 394
 395         return self.playlist_result(
 396             entries, channel_id, playlist_title=series.get('title'), playlist_description=series.get('description'))
 397
 398     @classmethod
 399     def _extract_embed_urls(cls, url, webpage):
 400         yield from super()._extract_embed_urls(url, webpage)
 401         for match in cls._extract_wistia_async_embed(webpage):
 402             if match.group('type') == 'wistia_channel':
 403                 # original url may contain wmediaid query param
 404                 yield update_url_query(f'wistiachannel:{match.group("id")}', parse_qs(url))