yt-dlp/yt_dlp/extractor/canvas.py

368 lines
15 KiB
Python
Raw Normal View History

import json
2015-10-12 07:07:18 +00:00
2015-10-12 07:07:18 +00:00
from .common import InfoExtractor
from .gigya import GigyaBaseIE
from ..compat import compat_HTTPError
from ..utils import (
ExtractorError,
clean_html,
float_or_none,
get_element_by_class,
str_or_none,
strip_or_none,
url_or_none,
urlencode_postdata
)
2015-10-12 07:07:18 +00:00
class CanvasIE(InfoExtractor):
2022-09-11 17:39:17 +00:00
_VALID_URL = r'https://media-services-public\.vrt\.be/media-aggregator/v2/media-items/(?P<video_id>.+)'
_TESTS = [{
'url': 'https://mediazone.vrt.be/api/v1/ketnet/assets/md-ast-4ac54990-ce66-4d00-a8ca-9eac86f4c475',
'md5': '37b2b7bb9b3dcaa05b67058dc3a714a9',
'info_dict': {
'id': 'md-ast-4ac54990-ce66-4d00-a8ca-9eac86f4c475',
'display_id': 'md-ast-4ac54990-ce66-4d00-a8ca-9eac86f4c475',
'ext': 'mp4',
'title': 'Nachtwacht: De Greystook',
'description': 'Nachtwacht: De Greystook',
'thumbnail': r're:^https?://.*\.jpg$',
'duration': 1468.02,
},
'expected_warnings': ['is not a supported codec'],
}, {
'url': 'https://mediazone.vrt.be/api/v1/canvas/assets/mz-ast-5e5f90b6-2d72-4c40-82c2-e134f884e93e',
'only_matching': True,
}]
2021-01-08 16:14:50 +00:00
_GEO_BYPASS = False
_HLS_ENTRY_PROTOCOLS_MAP = {
'HLS': 'm3u8_native',
'HLS_AES': 'm3u8_native',
}
2022-09-11 17:39:17 +00:00
_REST_API_BASE_TOKEN = 'https://media-services-public.vrt.be/vualto-video-aggregator-web/rest/external/v2'
_REST_API_BASE_VIDEO = 'https://media-services-public.vrt.be/media-aggregator/v2'
def _real_extract(self, url):
mobj = self._match_valid_url(url)
2022-09-11 17:39:17 +00:00
video_id = mobj.group('video_id')
2021-01-08 16:14:50 +00:00
data = None
2022-09-11 17:39:17 +00:00
vrtnutoken = self._download_json('https://token.vrt.be/refreshtoken',
2022-09-13 20:14:13 +00:00
video_id, note='refreshtoken: Retrieve vrtnutoken',
errnote='refreshtoken failed')['vrtnutoken']
2022-09-11 17:39:17 +00:00
headers = self.geo_verification_headers()
headers.update({'Content-Type': 'application/json; charset=utf-8'})
vrtPlayerToken = self._download_json(
'%s/tokens' % self._REST_API_BASE_TOKEN, video_id,
'Downloading token', headers=headers, data=json.dumps({
'identityToken': vrtnutoken
}).encode('utf-8'))['vrtPlayerToken']
data = self._download_json(
'%s/media-items/%s' % (self._REST_API_BASE_VIDEO, video_id),
video_id, 'Downloading video JSON', query={
'vrtPlayerToken': vrtPlayerToken,
'client': 'vrtnu-web@PROD',
}, expected_status=400)
if 'title' not in data:
code = data.get('code')
if code == 'AUTHENTICATION_REQUIRED':
self.raise_login_required()
elif code == 'INVALID_LOCATION':
self.raise_geo_restricted(countries=['BE'])
raise ExtractorError(data.get('message') or code, expected=True)
# Note: The title may be an empty string
2022-09-11 17:39:17 +00:00
title = data['title'] or f'{video_id}'
description = data.get('description')
formats = []
subtitles = {}
for target in data['targetUrls']:
format_url, format_type = url_or_none(target.get('url')), str_or_none(target.get('type'))
if not format_url or not format_type:
continue
format_type = format_type.upper()
if format_type in self._HLS_ENTRY_PROTOCOLS_MAP:
fmts, subs = self._extract_m3u8_formats_and_subtitles(
format_url, video_id, 'mp4', self._HLS_ENTRY_PROTOCOLS_MAP[format_type],
m3u8_id=format_type, fatal=False)
formats.extend(fmts)
subtitles = self._merge_subtitles(subtitles, subs)
elif format_type == 'HDS':
formats.extend(self._extract_f4m_formats(
format_url, video_id, f4m_id=format_type, fatal=False))
elif format_type == 'MPEG_DASH':
fmts, subs = self._extract_mpd_formats_and_subtitles(
format_url, video_id, mpd_id=format_type, fatal=False)
formats.extend(fmts)
subtitles = self._merge_subtitles(subtitles, subs)
elif format_type == 'HSS':
fmts, subs = self._extract_ism_formats_and_subtitles(
format_url, video_id, ism_id='mss', fatal=False)
formats.extend(fmts)
subtitles = self._merge_subtitles(subtitles, subs)
else:
formats.append({
'format_id': format_type,
'url': format_url,
})
self._sort_formats(formats)
subtitle_urls = data.get('subtitleUrls')
if isinstance(subtitle_urls, list):
for subtitle in subtitle_urls:
subtitle_url = subtitle.get('url')
if subtitle_url and subtitle.get('type') == 'CLOSED':
subtitles.setdefault('nl', []).append({'url': subtitle_url})
return {
'id': video_id,
'display_id': video_id,
'title': title,
'description': description,
'formats': formats,
'duration': float_or_none(data.get('duration'), 1000),
'thumbnail': data.get('posterImageUrl'),
'subtitles': subtitles,
}
class CanvasEenIE(InfoExtractor):
2017-04-08 15:45:45 +00:00
IE_DESC = 'canvas.be and een.be'
_VALID_URL = r'https?://(?:www\.)?(?P<site_id>canvas|een)\.be/(?:[^/]+/)*(?P<id>[^/?#&]+)'
2016-02-15 14:59:16 +00:00
_TESTS = [{
2015-10-12 07:07:18 +00:00
'url': 'http://www.canvas.be/video/de-afspraak/najaar-2015/de-afspraak-veilt-voor-de-warmste-week',
'md5': 'ed66976748d12350b118455979cca293',
2015-10-12 07:07:18 +00:00
'info_dict': {
2016-01-14 17:15:12 +00:00
'id': 'mz-ast-5e5f90b6-2d72-4c40-82c2-e134f884e93e',
'display_id': 'de-afspraak-veilt-voor-de-warmste-week',
'ext': 'flv',
2016-01-14 17:15:12 +00:00
'title': 'De afspraak veilt voor de Warmste Week',
'description': 'md5:24cb860c320dc2be7358e0e5aa317ba6',
'thumbnail': r're:^https?://.*\.jpg$',
2016-01-14 17:15:12 +00:00
'duration': 49.02,
},
'expected_warnings': ['is not a supported codec'],
2016-02-15 14:59:16 +00:00
}, {
# with subtitles
'url': 'http://www.canvas.be/video/panorama/2016/pieter-0167',
'info_dict': {
'id': 'mz-ast-5240ff21-2d30-4101-bba6-92b5ec67c625',
'display_id': 'pieter-0167',
'ext': 'mp4',
'title': 'Pieter 0167',
'description': 'md5:943cd30f48a5d29ba02c3a104dc4ec4e',
'thumbnail': r're:^https?://.*\.jpg$',
2016-02-15 14:59:16 +00:00
'duration': 2553.08,
'subtitles': {
'nl': [{
'ext': 'vtt',
}],
},
},
'params': {
'skip_download': True,
},
'skip': 'Pagina niet gevonden',
}, {
'url': 'https://www.een.be/thuis/emma-pakt-thilly-aan',
'info_dict': {
'id': 'md-ast-3a24ced2-64d7-44fb-b4ed-ed1aafbf90b8',
'display_id': 'emma-pakt-thilly-aan',
'ext': 'mp4',
'title': 'Emma pakt Thilly aan',
'description': 'md5:c5c9b572388a99b2690030afa3f3bad7',
'thumbnail': r're:^https?://.*\.jpg$',
'duration': 118.24,
},
'params': {
'skip_download': True,
},
'expected_warnings': ['is not a supported codec'],
}, {
'url': 'https://www.canvas.be/check-point/najaar-2016/de-politie-uw-vriend',
'only_matching': True,
2016-02-15 14:59:16 +00:00
}]
2015-10-12 07:07:18 +00:00
def _real_extract(self, url):
mobj = self._match_valid_url(url)
site_id, display_id = mobj.group('site_id'), mobj.group('id')
2015-10-12 07:07:18 +00:00
2016-01-14 17:15:12 +00:00
webpage = self._download_webpage(url, display_id)
2015-10-12 07:07:18 +00:00
title = strip_or_none(self._search_regex(
2016-01-14 17:15:12 +00:00
r'<h1[^>]+class="video__body__header__title"[^>]*>(.+?)</h1>',
webpage, 'title', default=None) or self._og_search_title(
webpage, default=None))
2016-01-14 17:15:12 +00:00
video_id = self._html_search_regex(
r'data-video=(["\'])(?P<id>(?:(?!\1).)+)\1', webpage, 'video id',
group='id')
2016-01-14 17:15:12 +00:00
2015-10-12 07:07:18 +00:00
return {
'_type': 'url_transparent',
'url': 'https://mediazone.vrt.be/api/v1/%s/assets/%s' % (site_id, video_id),
'ie_key': CanvasIE.ie_key(),
2015-10-12 07:07:18 +00:00
'id': video_id,
2016-01-14 17:15:12 +00:00
'display_id': display_id,
2015-10-12 07:07:18 +00:00
'title': title,
2016-01-14 17:15:12 +00:00
'description': self._og_search_description(webpage),
2015-10-12 07:07:18 +00:00
}
class VrtNUIE(GigyaBaseIE):
IE_DESC = 'VrtNU.be'
2022-09-11 17:39:17 +00:00
_VALID_URL = r'https?://(?:www\.)?vrt\.be/vrtnu/a-z/(?P<show>.+)/(?P<year>[0-9]{4})/(?P<id>.+)/?'
_TESTS = [{
# Available via old API endpoint
2021-01-08 16:14:50 +00:00
'url': 'https://www.vrt.be/vrtnu/a-z/postbus-x/1989/postbus-x-s1989a1/',
'info_dict': {
2021-01-08 16:14:50 +00:00
'id': 'pbs-pub-e8713dac-899e-41de-9313-81269f4c04ac$vid-90c932b1-e21d-4fb8-99b1-db7b49cf74de',
'ext': 'mp4',
2021-01-08 16:14:50 +00:00
'title': 'Postbus X - Aflevering 1 (Seizoen 1989)',
'description': 'md5:b704f669eb9262da4c55b33d7c6ed4b7',
'duration': 1457.04,
'thumbnail': r're:^https?://.*\.jpg$',
2021-01-08 16:14:50 +00:00
'series': 'Postbus X',
'season': 'Seizoen 1989',
'season_number': 1989,
'episode': 'De zwarte weduwe',
'episode_number': 1,
2021-01-08 16:14:50 +00:00
'timestamp': 1595822400,
'upload_date': '20200727',
},
'skip': 'This video is only available for registered users',
'expected_warnings': ['is not a supported codec'],
}, {
# Only available via new API endpoint
'url': 'https://www.vrt.be/vrtnu/a-z/kamp-waes/1/kamp-waes-s1a5/',
'info_dict': {
'id': 'pbs-pub-0763b56c-64fb-4d38-b95b-af60bf433c71$vid-ad36a73c-4735-4f1f-b2c0-a38e6e6aa7e1',
'ext': 'mp4',
'title': 'Aflevering 5',
'description': 'Wie valt door de mand tijdens een missie?',
'duration': 2967.06,
'season': 'Season 1',
'season_number': 1,
'episode_number': 5,
},
'skip': 'This video is only available for registered users',
'expected_warnings': ['Unable to download asset JSON', 'is not a supported codec', 'Unknown MIME type'],
}]
_NETRC_MACHINE = 'vrtnu'
_APIKEY = '3_0Z2HujMtiWq_pkAjgnS2Md2E11a1AwZjYiBETtwNE-EoEHDINgtnvcAOpNgmrVGy'
_CONTEXT_ID = 'R3595707040'
def _perform_login(self, username, password):
auth_info = self._gigya_login({
'APIKey': self._APIKEY,
'targetEnv': 'jssdk',
'loginID': username,
'password': password,
'authMode': 'cookie',
})
if auth_info.get('errorDetails'):
raise ExtractorError('Unable to login: VrtNU said: ' + auth_info.get('errorDetails'), expected=True)
# Sometimes authentication fails for no good reason, retry
login_attempt = 1
while login_attempt <= 3:
try:
self._request_webpage('https://token.vrt.be/vrtnuinitlogin',
None, note='Requesting XSRF Token', errnote='Could not get XSRF Token',
query={'provider': 'site', 'destination': 'https://www.vrt.be/vrtnu/'})
post_data = {
'UID': auth_info['UID'],
'UIDSignature': auth_info['UIDSignature'],
'signatureTimestamp': auth_info['signatureTimestamp'],
'_csrf': self._get_cookies('https://login.vrt.be').get('OIDCXSRF').value,
}
self._request_webpage(
'https://login.vrt.be/perform_login',
None, note='Performing login', errnote='perform login failed',
headers={}, query={
'client_id': 'vrtnu-site'
}, data=urlencode_postdata(post_data))
except ExtractorError as e:
if isinstance(e.cause, compat_HTTPError) and e.cause.code == 401:
login_attempt += 1
self.report_warning('Authentication failed')
self._sleep(1, None, msg_template='Waiting for %(timeout)s seconds before trying again')
else:
raise e
else:
break
def _real_extract(self, url):
display_id = self._match_id(url)
2022-09-11 17:39:17 +00:00
episode_data = self._download_json(f'{url.strip("/")}.model.json', display_id,
2022-09-13 20:14:13 +00:00
'Downloading asset JSON', 'Unable to download asset JSON')
2022-09-11 17:39:17 +00:00
details = episode_data.get('details')
actions = details.get('actions')
episode_publication_id = actions[2].get('episodePublicationId')
episode_video_id = actions[2].get('episodeVideoId')
video_id = f'{episode_publication_id}${episode_video_id}'
2022-09-11 17:39:17 +00:00
return {
'_type': 'url_transparent',
2022-09-11 17:39:17 +00:00
'url': 'https://media-services-public.vrt.be/media-aggregator/v2/media-items/%s' % video_id,
'ie_key': 'Canvas',
'id': video_id,
2022-09-11 17:39:17 +00:00
'display_id': display_id
}
class DagelijkseKostIE(InfoExtractor):
IE_DESC = 'dagelijksekost.een.be'
_VALID_URL = r'https?://dagelijksekost\.een\.be/gerechten/(?P<id>[^/?#&]+)'
_TEST = {
'url': 'https://dagelijksekost.een.be/gerechten/hachis-parmentier-met-witloof',
'md5': '30bfffc323009a3e5f689bef6efa2365',
'info_dict': {
'id': 'md-ast-27a4d1ff-7d7b-425e-b84f-a4d227f592fa',
'display_id': 'hachis-parmentier-met-witloof',
'ext': 'mp4',
'title': 'Hachis parmentier met witloof',
'description': 'md5:9960478392d87f63567b5b117688cdc5',
'thumbnail': r're:^https?://.*\.jpg$',
'duration': 283.02,
},
'expected_warnings': ['is not a supported codec'],
}
def _real_extract(self, url):
display_id = self._match_id(url)
webpage = self._download_webpage(url, display_id)
title = strip_or_none(get_element_by_class(
'dish-metadata__title', webpage
) or self._html_search_meta(
'twitter:title', webpage))
description = clean_html(get_element_by_class(
'dish-description', webpage)
) or self._html_search_meta(
('description', 'twitter:description', 'og:description'),
webpage)
video_id = self._html_search_regex(
r'data-url=(["\'])(?P<id>(?:(?!\1).)+)\1', webpage, 'video id',
group='id')
return {
'_type': 'url_transparent',
'url': 'https://mediazone.vrt.be/api/v1/dako/assets/%s' % video_id,
'ie_key': CanvasIE.ie_key(),
'id': video_id,
'display_id': display_id,
'title': title,
'description': description,
}