yt-dlp/yt_dlp/extractor/wistia.py

382 lines
15 KiB
Python
Raw Normal View History

import re
import urllib.error
import urllib.parse
from base64 import b64decode
2013-12-06 03:15:04 -05:00
from .common import InfoExtractor
Switch codebase to use sanitized_Request instead of compat_urllib_request.Request [downloader/dash] Use sanitized_Request [downloader/http] Use sanitized_Request [atresplayer] Use sanitized_Request [bambuser] Use sanitized_Request [bliptv] Use sanitized_Request [brightcove] Use sanitized_Request [cbs] Use sanitized_Request [ceskatelevize] Use sanitized_Request [collegerama] Use sanitized_Request [extractor/common] Use sanitized_Request [crunchyroll] Use sanitized_Request [dailymotion] Use sanitized_Request [dcn] Use sanitized_Request [dramafever] Use sanitized_Request [dumpert] Use sanitized_Request [eitb] Use sanitized_Request [escapist] Use sanitized_Request [everyonesmixtape] Use sanitized_Request [extremetube] Use sanitized_Request [facebook] Use sanitized_Request [fc2] Use sanitized_Request [flickr] Use sanitized_Request [4tube] Use sanitized_Request [gdcvault] Use sanitized_Request [extractor/generic] Use sanitized_Request [hearthisat] Use sanitized_Request [hotnewhiphop] Use sanitized_Request [hypem] Use sanitized_Request [iprima] Use sanitized_Request [ivi] Use sanitized_Request [keezmovies] Use sanitized_Request [letv] Use sanitized_Request [lynda] Use sanitized_Request [metacafe] Use sanitized_Request [minhateca] Use sanitized_Request [miomio] Use sanitized_Request [meovideo] Use sanitized_Request [mofosex] Use sanitized_Request [moniker] Use sanitized_Request [mooshare] Use sanitized_Request [movieclips] Use sanitized_Request [mtv] Use sanitized_Request [myvideo] Use sanitized_Request [neteasemusic] Use sanitized_Request [nfb] Use sanitized_Request [niconico] Use sanitized_Request [noco] Use sanitized_Request [nosvideo] Use sanitized_Request [novamov] Use sanitized_Request [nowness] Use sanitized_Request [nuvid] Use sanitized_Request [played] Use sanitized_Request [pluralsight] Use sanitized_Request [pornhub] Use sanitized_Request [pornotube] Use sanitized_Request [primesharetv] Use sanitized_Request [promptfile] Use sanitized_Request [qqmusic] Use sanitized_Request [rtve] Use sanitized_Request [safari] Use sanitized_Request [sandia] Use sanitized_Request [shared] Use sanitized_Request [sharesix] Use sanitized_Request [sina] Use sanitized_Request [smotri] Use sanitized_Request [sohu] Use sanitized_Request [spankwire] Use sanitized_Request [sportdeutschland] Use sanitized_Request [streamcloud] Use sanitized_Request [streamcz] Use sanitized_Request [tapely] Use sanitized_Request [tube8] Use sanitized_Request [tubitv] Use sanitized_Request [twitch] Use sanitized_Request [twitter] Use sanitized_Request [udemy] Use sanitized_Request [vbox7] Use sanitized_Request [veoh] Use sanitized_Request [vessel] Use sanitized_Request [vevo] Use sanitized_Request [viddler] Use sanitized_Request [videomega] Use sanitized_Request [viewvster] Use sanitized_Request [viki] Use sanitized_Request [vk] Use sanitized_Request [vodlocker] Use sanitized_Request [voicerepublic] Use sanitized_Request [wistia] Use sanitized_Request [xfileshare] Use sanitized_Request [xtube] Use sanitized_Request [xvideos] Use sanitized_Request [yandexmusic] Use sanitized_Request [youku] Use sanitized_Request [youporn] Use sanitized_Request [youtube] Use sanitized_Request [patreon] Use sanitized_Request [extractor/common] Remove unused import [nfb] PEP 8
2015-11-21 11:18:17 -05:00
from ..utils import (
ExtractorError,
float_or_none,
2021-01-01 07:26:37 -05:00
int_or_none,
parse_qs,
traverse_obj,
2021-01-01 07:26:37 -05:00
try_get,
update_url_query,
Switch codebase to use sanitized_Request instead of compat_urllib_request.Request [downloader/dash] Use sanitized_Request [downloader/http] Use sanitized_Request [atresplayer] Use sanitized_Request [bambuser] Use sanitized_Request [bliptv] Use sanitized_Request [brightcove] Use sanitized_Request [cbs] Use sanitized_Request [ceskatelevize] Use sanitized_Request [collegerama] Use sanitized_Request [extractor/common] Use sanitized_Request [crunchyroll] Use sanitized_Request [dailymotion] Use sanitized_Request [dcn] Use sanitized_Request [dramafever] Use sanitized_Request [dumpert] Use sanitized_Request [eitb] Use sanitized_Request [escapist] Use sanitized_Request [everyonesmixtape] Use sanitized_Request [extremetube] Use sanitized_Request [facebook] Use sanitized_Request [fc2] Use sanitized_Request [flickr] Use sanitized_Request [4tube] Use sanitized_Request [gdcvault] Use sanitized_Request [extractor/generic] Use sanitized_Request [hearthisat] Use sanitized_Request [hotnewhiphop] Use sanitized_Request [hypem] Use sanitized_Request [iprima] Use sanitized_Request [ivi] Use sanitized_Request [keezmovies] Use sanitized_Request [letv] Use sanitized_Request [lynda] Use sanitized_Request [metacafe] Use sanitized_Request [minhateca] Use sanitized_Request [miomio] Use sanitized_Request [meovideo] Use sanitized_Request [mofosex] Use sanitized_Request [moniker] Use sanitized_Request [mooshare] Use sanitized_Request [movieclips] Use sanitized_Request [mtv] Use sanitized_Request [myvideo] Use sanitized_Request [neteasemusic] Use sanitized_Request [nfb] Use sanitized_Request [niconico] Use sanitized_Request [noco] Use sanitized_Request [nosvideo] Use sanitized_Request [novamov] Use sanitized_Request [nowness] Use sanitized_Request [nuvid] Use sanitized_Request [played] Use sanitized_Request [pluralsight] Use sanitized_Request [pornhub] Use sanitized_Request [pornotube] Use sanitized_Request [primesharetv] Use sanitized_Request [promptfile] Use sanitized_Request [qqmusic] Use sanitized_Request [rtve] Use sanitized_Request [safari] Use sanitized_Request [sandia] Use sanitized_Request [shared] Use sanitized_Request [sharesix] Use sanitized_Request [sina] Use sanitized_Request [smotri] Use sanitized_Request [sohu] Use sanitized_Request [spankwire] Use sanitized_Request [sportdeutschland] Use sanitized_Request [streamcloud] Use sanitized_Request [streamcz] Use sanitized_Request [tapely] Use sanitized_Request [tube8] Use sanitized_Request [tubitv] Use sanitized_Request [twitch] Use sanitized_Request [twitter] Use sanitized_Request [udemy] Use sanitized_Request [vbox7] Use sanitized_Request [veoh] Use sanitized_Request [vessel] Use sanitized_Request [vevo] Use sanitized_Request [viddler] Use sanitized_Request [videomega] Use sanitized_Request [viewvster] Use sanitized_Request [viki] Use sanitized_Request [vk] Use sanitized_Request [vodlocker] Use sanitized_Request [voicerepublic] Use sanitized_Request [wistia] Use sanitized_Request [xfileshare] Use sanitized_Request [xtube] Use sanitized_Request [xvideos] Use sanitized_Request [yandexmusic] Use sanitized_Request [youku] Use sanitized_Request [youporn] Use sanitized_Request [youtube] Use sanitized_Request [patreon] Use sanitized_Request [extractor/common] Remove unused import [nfb] PEP 8
2015-11-21 11:18:17 -05:00
)
2013-12-06 03:15:04 -05:00
2021-01-01 07:26:37 -05:00
class WistiaBaseIE(InfoExtractor):
_VALID_ID_REGEX = r'(?P<id>[a-z0-9]{10})'
_VALID_URL_BASE = r'https?://(?:\w+\.)?wistia\.(?:net|com)/(?:embed/)?'
_EMBED_BASE_URL = 'http://fast.wistia.net/embed/'
2013-12-06 03:15:04 -05:00
2021-01-01 07:26:37 -05:00
def _download_embed_config(self, config_type, config_id, referer):
base_url = self._EMBED_BASE_URL + '%s/%s' % (config_type, config_id)
2021-01-01 07:26:37 -05:00
embed_config = self._download_json(
base_url + '.json', config_id, headers={
'Referer': referer if referer.startswith('http') else base_url, # Some videos require this.
2016-05-20 10:55:10 -04:00
})
error = traverse_obj(embed_config, 'error')
if error:
2016-05-20 10:55:10 -04:00
raise ExtractorError(
f'Error while getting the playlist: {error}', expected=True)
2016-05-20 10:55:10 -04:00
2021-01-01 07:26:37 -05:00
return embed_config
def _extract_media(self, embed_config):
data = embed_config['media']
video_id = data['hashedId']
2016-03-17 12:48:17 -04:00
title = data['name']
2013-12-06 03:15:04 -05:00
formats = []
thumbnails = []
2016-03-01 12:26:53 -05:00
for a in data['assets']:
aurl = a.get('url')
if not aurl:
continue
2016-03-17 12:48:17 -04:00
astatus = a.get('status')
2016-03-01 12:26:53 -05:00
atype = a.get('type')
if (astatus is not None and astatus != 2) or atype in ('preview', 'storyboard'):
2016-03-17 12:48:17 -04:00
continue
elif atype in ('still', 'still_image'):
2013-12-06 03:15:04 -05:00
thumbnails.append({
'url': aurl,
'width': int_or_none(a.get('width')),
'height': int_or_none(a.get('height')),
'filesize': int_or_none(a.get('size')),
2013-12-06 03:15:04 -05:00
})
2016-03-17 12:48:17 -04:00
else:
2016-05-20 11:16:08 -04:00
aext = a.get('ext')
display_name = a.get('display_name')
format_id = atype
if atype and atype.endswith('_video') and display_name:
format_id = '%s-%s' % (atype[:-6], display_name)
f = {
'format_id': format_id,
'url': aurl,
'tbr': int_or_none(a.get('bitrate')) or None,
'quality': 1 if atype == 'original' else None,
}
if display_name == 'Audio':
f.update({
'vcodec': 'none',
})
else:
f.update({
'width': int_or_none(a.get('width')),
'height': int_or_none(a.get('height')),
'vcodec': a.get('codec'),
})
if a.get('container') == 'm3u8' or aext == 'm3u8':
ts_f = f.copy()
ts_f.update({
'ext': 'ts',
'format_id': f['format_id'].replace('hls-', 'ts-'),
'url': f['url'].replace('.bin', '.ts'),
})
formats.append(ts_f)
f.update({
'ext': 'mp4',
'protocol': 'm3u8_native',
})
else:
f.update({
'container': a.get('container'),
'ext': aext,
'filesize': int_or_none(a.get('size')),
})
formats.append(f)
2013-12-25 09:20:14 -05:00
subtitles = {}
for caption in data.get('captions', []):
language = caption.get('language')
if not language:
continue
subtitles[language] = [{
'url': self._EMBED_BASE_URL + 'captions/' + video_id + '.vtt?language=' + language,
}]
2013-12-06 03:15:04 -05:00
return {
'id': video_id,
2016-03-17 12:48:17 -04:00
'title': title,
'description': data.get('seoDescription'),
2013-12-06 03:15:04 -05:00
'formats': formats,
'thumbnails': thumbnails,
'duration': float_or_none(data.get('duration')),
2016-03-17 12:48:17 -04:00
'timestamp': int_or_none(data.get('createdAt')),
'subtitles': subtitles,
2013-12-06 03:15:04 -05:00
}
2021-01-01 07:26:37 -05:00
@classmethod
def _extract_from_webpage(cls, url, webpage):
from .teachable import TeachableIE
if list(TeachableIE._extract_embed_urls(url, webpage)):
return
yield from super()._extract_from_webpage(url, webpage)
@classmethod
def _extract_wistia_async_embed(cls, webpage):
# https://wistia.com/support/embed-and-share/video-on-your-website
# https://wistia.com/support/embed-and-share/channel-embeds
yield from re.finditer(
r'''(?sx)
<(?:div|section)[^>]+class=([\"'])(?:(?!\1).)*?(?P<type>wistia[a-z_0-9]+)\s*\bwistia_async_(?P<id>[a-z0-9]{10})\b(?:(?!\1).)*?\1
''', webpage)
@classmethod
def _extract_url_media_id(cls, url):
mobj = re.search(r'(?:wmediaid|wvideo(?:id)?)]?=(?P<id>[a-z0-9]{10})', urllib.parse.unquote_plus(url))
if mobj:
return mobj.group('id')
2021-01-01 07:26:37 -05:00
class WistiaIE(WistiaBaseIE):
_VALID_URL = r'(?:wistia:|%s(?:iframe|medias)/)%s' % (WistiaBaseIE._VALID_URL_BASE, WistiaBaseIE._VALID_ID_REGEX)
_EMBED_REGEX = [
r'''(?x)
<(?:meta[^>]+?content|(?:iframe|script)[^>]+?src)=["\']
(?P<url>(?:https?:)?//(?:fast\.)?wistia\.(?:net|com)/embed/(?:iframe|medias)/[a-z0-9]{10})
''']
2021-01-01 07:26:37 -05:00
_TESTS = [{
# with hls video
'url': 'wistia:807fafadvk',
'md5': 'daff0f3687a41d9a71b40e0e8c2610fe',
'info_dict': {
'id': '807fafadvk',
'ext': 'mp4',
'title': 'Drip Brennan Dunn Workshop',
'description': 'a JV Webinars video',
'upload_date': '20160518',
'timestamp': 1463607249,
'duration': 4987.11,
},
'skip': 'video unavailable',
}, {
'url': 'wistia:a6ndpko1wg',
'md5': '10c1ce9c4dde638202513ed17a3767bd',
'info_dict': {
'id': 'a6ndpko1wg',
'ext': 'bin',
'title': 'Episode 2: Boxed Water\'s retention is thirsty',
'upload_date': '20210324',
'description': 'md5:da5994c2c2d254833b412469d9666b7a',
'duration': 966.0,
'timestamp': 1616614369,
'thumbnail': 'https://embed-ssl.wistia.com/deliveries/53dc60239348dc9b9fba3755173ea4c2.bin',
}
}, {
'url': 'wistia:5vd7p4bct5',
'md5': 'b9676d24bf30945d97060638fbfe77f0',
'info_dict': {
'id': '5vd7p4bct5',
'ext': 'bin',
'title': 'md5:eaa9f64c4efd7b5f098b9b6118597679',
'description': 'md5:a9bea0315f0616aa5df2dc413ddcdd0f',
'upload_date': '20220915',
'timestamp': 1663258727,
'duration': 623.019,
'thumbnail': r're:https?://embed(?:-ssl)?.wistia.com/.+\.(?:jpg|bin)$',
},
2021-01-01 07:26:37 -05:00
}, {
'url': 'wistia:sh7fpupwlt',
'only_matching': True,
}, {
'url': 'http://fast.wistia.net/embed/iframe/sh7fpupwlt',
'only_matching': True,
}, {
'url': 'http://fast.wistia.com/embed/iframe/sh7fpupwlt',
'only_matching': True,
}, {
'url': 'http://fast.wistia.net/embed/medias/sh7fpupwlt.json',
'only_matching': True,
}]
_WEBPAGE_TESTS = [{
'url': 'https://www.weidert.com/blog/wistia-channels-video-marketing-tool',
'info_dict': {
'id': 'cqwukac3z1',
'ext': 'bin',
'title': 'How Wistia Channels Can Help Capture Inbound Value From Your Video Content',
'duration': 158.125,
'timestamp': 1618974400,
'description': 'md5:27abc99a758573560be72600ef95cece',
'upload_date': '20210421',
'thumbnail': 'https://embed-ssl.wistia.com/deliveries/6c551820ae950cdee2306d6cbe9ef742.bin',
}
}, {
'url': 'https://study.com/academy/lesson/north-american-exploration-failed-colonies-of-spain-france-england.html#lesson',
'md5': 'b9676d24bf30945d97060638fbfe77f0',
'info_dict': {
'id': '5vd7p4bct5',
'ext': 'bin',
'title': 'paywall_north-american-exploration-failed-colonies-of-spain-france-england',
'upload_date': '20220915',
'timestamp': 1663258727,
'duration': 623.019,
'thumbnail': 'https://embed-ssl.wistia.com/deliveries/83e6ec693e2c05a0ce65809cbaead86a.bin',
'description': 'a Paywall Videos video',
},
}]
def _real_extract(self, url):
video_id = self._match_id(url)
embed_config = self._download_embed_config('medias', video_id, url)
return self._extract_media(embed_config)
@classmethod
def _extract_embed_urls(cls, url, webpage):
urls = list(super()._extract_embed_urls(url, webpage))
for match in cls._extract_wistia_async_embed(webpage):
if match.group('type') != 'wistia_channel':
urls.append('wistia:%s' % match.group('id'))
for match in re.finditer(r'(?:data-wistia-?id=["\']|Wistia\.embed\(["\']|id=["\']wistia_)(?P<id>[a-z0-9]{10})',
webpage):
urls.append('wistia:%s' % match.group('id'))
if not WistiaChannelIE._extract_embed_urls(url, webpage): # Fallback
media_id = cls._extract_url_media_id(url)
if media_id:
urls.append('wistia:%s' % match.group('id'))
2021-01-01 07:26:37 -05:00
return urls
class WistiaPlaylistIE(WistiaBaseIE):
_VALID_URL = r'%splaylists/%s' % (WistiaBaseIE._VALID_URL_BASE, WistiaBaseIE._VALID_ID_REGEX)
2021-01-01 07:26:37 -05:00
_TEST = {
'url': 'https://fast.wistia.net/embed/playlists/aodt9etokc',
'info_dict': {
'id': 'aodt9etokc',
},
'playlist_count': 3,
}
def _real_extract(self, url):
playlist_id = self._match_id(url)
playlist = self._download_embed_config('playlists', playlist_id, url)
2021-01-01 07:26:37 -05:00
entries = []
for media in (try_get(playlist, lambda x: x[0]['medias']) or []):
embed_config = media.get('embed_config')
if not embed_config:
continue
entries.append(self._extract_media(embed_config))
return self.playlist_result(entries, playlist_id)
class WistiaChannelIE(WistiaBaseIE):
_VALID_URL = r'(?:wistiachannel:|%schannel/)%s' % (WistiaBaseIE._VALID_URL_BASE, WistiaBaseIE._VALID_ID_REGEX)
_TESTS = [{
# JSON Embed API returns 403, should fall back to webpage
'url': 'https://fast.wistia.net/embed/channel/yvyvu7wjbg?wchannelid=yvyvu7wjbg',
'info_dict': {
'id': 'yvyvu7wjbg',
'title': 'Copysmith Tutorials and Education!',
'description': 'Learn all things Copysmith via short and informative videos!'
},
'playlist_mincount': 7,
'expected_warnings': ['falling back to webpage'],
}, {
'url': 'https://fast.wistia.net/embed/channel/3802iirk0l',
'info_dict': {
'id': '3802iirk0l',
'title': 'The Roof',
},
'playlist_mincount': 20,
}, {
# link to popup video, follow --no-playlist
'url': 'https://fast.wistia.net/embed/channel/3802iirk0l?wchannelid=3802iirk0l&wmediaid=sp5dqjzw3n',
'info_dict': {
'id': 'sp5dqjzw3n',
'ext': 'bin',
'title': 'The Roof S2: The Modern CRO',
'thumbnail': 'https://embed-ssl.wistia.com/deliveries/dadfa9233eaa505d5e0c85c23ff70741.bin',
'duration': 86.487,
'description': 'A sales leader on The Roof? Man, they really must be letting anyone up here this season.\n',
'timestamp': 1619790290,
'upload_date': '20210430',
},
'params': {'noplaylist': True, 'skip_download': True},
}]
_WEBPAGE_TESTS = [{
'url': 'https://www.profitwell.com/recur/boxed-out',
'info_dict': {
'id': '6jyvmqz6zs',
'title': 'Boxed Out',
'description': 'md5:14a8a93a1dbe236718e6a59f8c8c7bae',
},
'playlist_mincount': 30,
}, {
# section instead of div
'url': 'https://360learning.com/studio/onboarding-joei/',
'info_dict': {
'id': 'z874k93n2o',
'title': 'Onboarding Joei.',
'description': 'Coming to you weekly starting Feb 19th.',
},
'playlist_mincount': 20,
}, {
'url': 'https://amplitude.com/amplify-sessions?amp%5Bwmediaid%5D=pz0m0l0if3&amp%5Bwvideo%5D=pz0m0l0if3&wchannelid=emyjmwjf79&wmediaid=i8um783bdt',
'info_dict': {
'id': 'pz0m0l0if3',
'title': 'A Framework for Improving Product Team Performance',
'ext': 'bin',
'timestamp': 1653935275,
'upload_date': '20220530',
'description': 'Learn how to help your company improve and achieve your product related goals.',
'duration': 1854.39,
'thumbnail': 'https://embed-ssl.wistia.com/deliveries/12fd19e56413d9d6f04e2185c16a6f8854e25226.bin',
},
'params': {'noplaylist': True, 'skip_download': True},
}]
def _real_extract(self, url):
channel_id = self._match_id(url)
media_id = self._extract_url_media_id(url)
if not self._yes_playlist(channel_id, media_id, playlist_label='channel'):
return self.url_result(f'wistia:{media_id}', 'Wistia')
try:
data = self._download_embed_config('channel', channel_id, url)
except (ExtractorError, urllib.error.HTTPError):
# Some channels give a 403 from the JSON API
self.report_warning('Failed to download channel data from API, falling back to webpage.')
webpage = self._download_webpage(f'https://fast.wistia.net/embed/channel/{channel_id}', channel_id)
data = self._parse_json(
self._search_regex(r'wchanneljsonp-%s\'\]\s*=[^\"]*\"([A-Za-z0-9=/]*)' % channel_id, webpage, 'jsonp', channel_id),
channel_id, transform_source=lambda x: urllib.parse.unquote_plus(b64decode(x).decode('utf-8')))
# XXX: can there be more than one series?
series = traverse_obj(data, ('series', 0), default={})
entries = [
self.url_result(f'wistia:{video["hashedId"]}', WistiaIE, title=video.get('name'))
for video in traverse_obj(series, ('sections', ..., 'videos', ...)) or []
if video.get('hashedId')
]
return self.playlist_result(
entries, channel_id, playlist_title=series.get('title'), playlist_description=series.get('description'))
@classmethod
def _extract_embed_urls(cls, url, webpage):
yield from super()._extract_embed_urls(url, webpage)
for match in cls._extract_wistia_async_embed(webpage):
if match.group('type') == 'wistia_channel':
# original url may contain wmediaid query param
yield update_url_query(f'wistiachannel:{match.group("id")}', parse_qs(url))