[yt-dlp.git] / youtube_dl / extractor / canvas.py

from __future__ import unicode_literals

import re
import json

from .common import InfoExtractor
from .gigya import GigyaBaseIE
from ..compat import compat_HTTPError
from ..utils import (
    ExtractorError,
    strip_or_none,
    float_or_none,
    int_or_none,
    merge_dicts,
    parse_iso8601,
)


class CanvasIE(InfoExtractor):
    _VALID_URL = r'https?://mediazone\.vrt\.be/api/v1/(?P<site_id>canvas|een|ketnet|vrt(?:video|nieuws)|sporza)/assets/(?P<id>[^/?#&]+)'
    _TESTS = [{
        'url': 'https://mediazone.vrt.be/api/v1/ketnet/assets/md-ast-4ac54990-ce66-4d00-a8ca-9eac86f4c475',
        'md5': '90139b746a0a9bd7bb631283f6e2a64e',
        'info_dict': {
            'id': 'md-ast-4ac54990-ce66-4d00-a8ca-9eac86f4c475',
            'display_id': 'md-ast-4ac54990-ce66-4d00-a8ca-9eac86f4c475',
            'ext': 'flv',
            'title': 'Nachtwacht: De Greystook',
            'description': 'md5:1db3f5dc4c7109c821261e7512975be7',
            'thumbnail': r're:^https?://.*\.jpg$',
            'duration': 1468.03,
        },
        'expected_warnings': ['is not a supported codec', 'Unknown MIME type'],
    }, {
        'url': 'https://mediazone.vrt.be/api/v1/canvas/assets/mz-ast-5e5f90b6-2d72-4c40-82c2-e134f884e93e',
        'only_matching': True,
    }]
    _HLS_ENTRY_PROTOCOLS_MAP = {
        'HLS': 'm3u8_native',
        'HLS_AES': 'm3u8',
    }

    def _real_extract(self, url):
        mobj = re.match(self._VALID_URL, url)
        site_id, video_id = mobj.group('site_id'), mobj.group('id')

        data = self._download_json(
            'https://mediazone.vrt.be/api/v1/%s/assets/%s'
            % (site_id, video_id), video_id)

        title = data['title']
        description = data.get('description')

        formats = []
        for target in data['targetUrls']:
            format_url, format_type = target.get('url'), target.get('type')
            if not format_url or not format_type:
                continue
            if format_type in self._HLS_ENTRY_PROTOCOLS_MAP:
                formats.extend(self._extract_m3u8_formats(
                    format_url, video_id, 'mp4', self._HLS_ENTRY_PROTOCOLS_MAP[format_type],
                    m3u8_id=format_type, fatal=False))
            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':
                formats.extend(self._extract_mpd_formats(
                    format_url, video_id, mpd_id=format_type, fatal=False))
            elif format_type == 'HSS':
                formats.extend(self._extract_ism_formats(
                    format_url, video_id, ism_id='mss', fatal=False))
            else:
                formats.append({
                    'format_id': format_type,
                    'url': format_url,
                })
        self._sort_formats(formats)

        subtitles = {}
        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):
    IE_DESC = 'canvas.be and een.be'
    _VALID_URL = r'https?://(?:www\.)?(?P<site_id>canvas|een)\.be/(?:[^/]+/)*(?P<id>[^/?#&]+)'
    _TESTS = [{
        'url': 'http://www.canvas.be/video/de-afspraak/najaar-2015/de-afspraak-veilt-voor-de-warmste-week',
        'md5': 'ed66976748d12350b118455979cca293',
        'info_dict': {
            'id': 'mz-ast-5e5f90b6-2d72-4c40-82c2-e134f884e93e',
            'display_id': 'de-afspraak-veilt-voor-de-warmste-week',
            'ext': 'flv',
            'title': 'De afspraak veilt voor de Warmste Week',
            'description': 'md5:24cb860c320dc2be7358e0e5aa317ba6',
            'thumbnail': r're:^https?://.*\.jpg$',
            'duration': 49.02,
        },
        'expected_warnings': ['is not a supported codec'],
    }, {
        # with subtitles
        'url': 'http://www.canvas.be/video/panorama/2016/pieter-0167',
        'info_dict': {
            'id': 'mz-ast-5240ff21-2d30-4101-bba6-92b5ec67c625',
            'display_id': 'pieter-0167',
            'ext': 'mp4',
            'title': 'Pieter 0167',
            'description': 'md5:943cd30f48a5d29ba02c3a104dc4ec4e',
            'thumbnail': r're:^https?://.*\.jpg$',
            'duration': 2553.08,
            'subtitles': {
                'nl': [{
                    'ext': 'vtt',
                }],
            },
        },
        'params': {
            'skip_download': True,
        },
        'skip': 'Pagina niet gevonden',
    }, {
        'url': 'https://www.een.be/sorry-voor-alles/herbekijk-sorry-voor-alles',
        'info_dict': {
            'id': 'mz-ast-11a587f8-b921-4266-82e2-0bce3e80d07f',
            'display_id': 'herbekijk-sorry-voor-alles',
            'ext': 'mp4',
            'title': 'Herbekijk Sorry voor alles',
            'description': 'md5:8bb2805df8164e5eb95d6a7a29dc0dd3',
            'thumbnail': r're:^https?://.*\.jpg$',
            'duration': 3788.06,
        },
        'params': {
            'skip_download': True,
        },
        'skip': 'Episode no longer available',
    }, {
        'url': 'https://www.canvas.be/check-point/najaar-2016/de-politie-uw-vriend',
        'only_matching': True,
    }]

    def _real_extract(self, url):
        mobj = re.match(self._VALID_URL, url)
        site_id, display_id = mobj.group('site_id'), mobj.group('id')

        webpage = self._download_webpage(url, display_id)

        title = strip_or_none(self._search_regex(
            r'<h1[^>]+class="video__body__header__title"[^>]*>(.+?)</h1>',
            webpage, 'title', default=None) or self._og_search_title(
            webpage, default=None))

        video_id = self._html_search_regex(
            r'data-video=(["\'])(?P<id>(?:(?!\1).)+)\1', webpage, 'video id',
            group='id')

        return {
            '_type': 'url_transparent',
            'url': 'https://mediazone.vrt.be/api/v1/%s/assets/%s' % (site_id, video_id),
            'ie_key': CanvasIE.ie_key(),
            'id': video_id,
            'display_id': display_id,
            'title': title,
            'description': self._og_search_description(webpage),
        }


class VrtNUIE(GigyaBaseIE):
    IE_DESC = 'VrtNU.be'
    _VALID_URL = r'https?://(?:www\.)?vrt\.be/(?P<site_id>vrtnu)/(?:[^/]+/)*(?P<id>[^/?#&]+)'
    _TESTS = [{
        'url': 'https://www.vrt.be/vrtnu/a-z/postbus-x/1/postbus-x-s1a1/',
        'info_dict': {
            'id': 'pbs-pub-2e2d8c27-df26-45c9-9dc6-90c78153044d$vid-90c932b1-e21d-4fb8-99b1-db7b49cf74de',
            'ext': 'flv',
            'title': 'De zwarte weduwe',
            'description': 'md5:d90c21dced7db869a85db89a623998d4',
            'duration': 1457.04,
            'thumbnail': r're:^https?://.*\.jpg$',
            'season': '1',
            'season_number': 1,
            'episode_number': 1,
        },
        'skip': 'This video is only available for registered users'
    }]
    _NETRC_MACHINE = 'vrtnu'
    _APIKEY = '3_0Z2HujMtiWq_pkAjgnS2Md2E11a1AwZjYiBETtwNE-EoEHDINgtnvcAOpNgmrVGy'
    _CONTEXT_ID = 'R3595707040'

    def _real_initialize(self):
        self._login()

    def _login(self):
        username, password = self._get_login_info()
        if username is None:
            return

        auth_data = {
            'APIKey': self._APIKEY,
            'targetEnv': 'jssdk',
            'loginID': username,
            'password': password,
            'authMode': 'cookie',
        }

        auth_info = self._gigya_login(auth_data)

        # Sometimes authentication fails for no good reason, retry
        login_attempt = 1
        while login_attempt <= 3:
            try:
                # When requesting a token, no actual token is returned, but the
                # necessary cookies are set.
                self._request_webpage(
                    'https://token.vrt.be',
                    None, note='Requesting a token', errnote='Could not get a token',
                    headers={
                        'Content-Type': 'application/json',
                        'Referer': 'https://www.vrt.be/vrtnu/',
                    },
                    data=json.dumps({
                        'uid': auth_info['UID'],
                        'uidsig': auth_info['UIDSignature'],
                        'ts': auth_info['signatureTimestamp'],
                        'email': auth_info['profile']['email'],
                    }).encode('utf-8'))
            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)

        webpage, urlh = self._download_webpage_handle(url, display_id)

        info = self._search_json_ld(webpage, display_id, default={})

        # title is optional here since it may be extracted by extractor
        # that is delegated from here
        title = strip_or_none(self._html_search_regex(
            r'(?ms)<h1 class="content__heading">(.+?)</h1>',
            webpage, 'title', default=None))

        description = self._html_search_regex(
            r'(?ms)<div class="content__description">(.+?)</div>',
            webpage, 'description', default=None)

        season = self._html_search_regex(
            [r'''(?xms)<div\ class="tabs__tab\ tabs__tab--active">\s*
                    <span>seizoen\ (.+?)</span>\s*
                </div>''',
             r'<option value="seizoen (\d{1,3})" data-href="[^"]+?" selected>'],
            webpage, 'season', default=None)

        season_number = int_or_none(season)

        episode_number = int_or_none(self._html_search_regex(
            r'''(?xms)<div\ class="content__episode">\s*
                    <abbr\ title="aflevering">afl</abbr>\s*<span>(\d+)</span>
                </div>''',
            webpage, 'episode_number', default=None))

        release_date = parse_iso8601(self._html_search_regex(
            r'(?ms)<div class="content__broadcastdate">\s*<time\ datetime="(.+?)"',
            webpage, 'release_date', default=None))

        # If there's a ? or a # in the URL, remove them and everything after
        clean_url = urlh.geturl().split('?')[0].split('#')[0].strip('/')
        securevideo_url = clean_url + '.mssecurevideo.json'

        try:
            video = self._download_json(securevideo_url, display_id)
        except ExtractorError as e:
            if isinstance(e.cause, compat_HTTPError) and e.cause.code == 401:
                self.raise_login_required()
            raise

        # We are dealing with a '../<show>.relevant' URL
        redirect_url = video.get('url')
        if redirect_url:
            return self.url_result(self._proto_relative_url(redirect_url, 'https:'))

        # There is only one entry, but with an unknown key, so just get
        # the first one
        video_id = list(video.values())[0].get('videoid')

        return merge_dicts(info, {
            '_type': 'url_transparent',
            'url': 'https://mediazone.vrt.be/api/v1/vrtvideo/assets/%s' % video_id,
            'ie_key': CanvasIE.ie_key(),
            'id': video_id,
            'display_id': display_id,
            'title': title,
            'description': description,
            'season': season,
            'season_number': season_number,
            'episode_number': episode_number,
            'release_date': release_date,
        })
Commit	Line	Data
be2d40a5 TG	1	from __future__ import unicode_literals
be2d40a5 TG	2
bb5ebd44	3	import re
7913e0fc	4	import json
bb5ebd44	5
be2d40a5	6	from .common import InfoExtractor
7913e0fc	7	from .gigya import GigyaBaseIE
7913e0fc	8	from ..compat import compat_HTTPError
117589df	9	from ..utils import (
7913e0fc	10	ExtractorError,
117589df	11	strip_or_none,
7913e0fc	12	float_or_none,
7913e0fc	13	int_or_none,
fd62b366	14	merge_dicts,
7913e0fc	15	parse_iso8601,
117589df	16	)
be2d40a5 TG	17
	18
	19	class CanvasIE(InfoExtractor):
170d6444	20	_VALID_URL = r'https?://mediazone\.vrt\.be/api/v1/(?P<site_id>canvas\|een\|ketnet\|vrt(?:video\|nieuws)\|sporza)/assets/(?P<id>[^/?#&]+)'
117589df S	21	_TESTS = [{
117589df S	22	'url': 'https://mediazone.vrt.be/api/v1/ketnet/assets/md-ast-4ac54990-ce66-4d00-a8ca-9eac86f4c475',
19a107f2	23	'md5': '90139b746a0a9bd7bb631283f6e2a64e',
117589df S	24	'info_dict': {
	25	'id': 'md-ast-4ac54990-ce66-4d00-a8ca-9eac86f4c475',
	26	'display_id': 'md-ast-4ac54990-ce66-4d00-a8ca-9eac86f4c475',
19a107f2	27	'ext': 'flv',
117589df	28	'title': 'Nachtwacht: De Greystook',
19a107f2	29	'description': 'md5:1db3f5dc4c7109c821261e7512975be7',
117589df	30	'thumbnail': r're:^https?://.*\.jpg$',
19a107f2	31	'duration': 1468.03,
117589df S	32	},
	33	'expected_warnings': ['is not a supported codec', 'Unknown MIME type'],
	34	}, {
	35	'url': 'https://mediazone.vrt.be/api/v1/canvas/assets/mz-ast-5e5f90b6-2d72-4c40-82c2-e134f884e93e',
	36	'only_matching': True,
	37	}]
170d6444 RA	38	_HLS_ENTRY_PROTOCOLS_MAP = {
	39	'HLS': 'm3u8_native',
	40	'HLS_AES': 'm3u8',
	41	}
117589df S	42
	43	def _real_extract(self, url):
	44	mobj = re.match(self._VALID_URL, url)
	45	site_id, video_id = mobj.group('site_id'), mobj.group('id')
	46
	47	data = self._download_json(
	48	'https://mediazone.vrt.be/api/v1/%s/assets/%s'
19a107f2	49	% (site_id, video_id), video_id)
117589df S	50
	51	title = data['title']
	52	description = data.get('description')
	53
	54	formats = []
	55	for target in data['targetUrls']:
19a107f2	56	format_url, format_type = target.get('url'), target.get('type')
117589df S	57	if not format_url or not format_type:
117589df S	58	continue
170d6444	59	if format_type in self._HLS_ENTRY_PROTOCOLS_MAP:
117589df	60	formats.extend(self._extract_m3u8_formats(
170d6444	61	format_url, video_id, 'mp4', self._HLS_ENTRY_PROTOCOLS_MAP[format_type],
117589df S	62	m3u8_id=format_type, fatal=False))
	63	elif format_type == 'HDS':
	64	formats.extend(self._extract_f4m_formats(
	65	format_url, video_id, f4m_id=format_type, fatal=False))
	66	elif format_type == 'MPEG_DASH':
	67	formats.extend(self._extract_mpd_formats(
	68	format_url, video_id, mpd_id=format_type, fatal=False))
	69	elif format_type == 'HSS':
	70	formats.extend(self._extract_ism_formats(
	71	format_url, video_id, ism_id='mss', fatal=False))
	72	else:
	73	formats.append({
	74	'format_id': format_type,
	75	'url': format_url,
	76	})
	77	self._sort_formats(formats)
	78
	79	subtitles = {}
	80	subtitle_urls = data.get('subtitleUrls')
	81	if isinstance(subtitle_urls, list):
	82	for subtitle in subtitle_urls:
	83	subtitle_url = subtitle.get('url')
	84	if subtitle_url and subtitle.get('type') == 'CLOSED':
	85	subtitles.setdefault('nl', []).append({'url': subtitle_url})
	86
	87	return {
	88	'id': video_id,
	89	'display_id': video_id,
	90	'title': title,
	91	'description': description,
	92	'formats': formats,
	93	'duration': float_or_none(data.get('duration'), 1000),
	94	'thumbnail': data.get('posterImageUrl'),
	95	'subtitles': subtitles,
	96	}
	97
	98
	99	class CanvasEenIE(InfoExtractor):
41b263ac	100	IE_DESC = 'canvas.be and een.be'
bb5ebd44	101	_VALID_URL = r'https?://(?:www\.)?(?P<site_id>canvas\|een)\.be/(?:[^/]+/)*(?P<id>[^/?#&]+)'
6eff2605	102	_TESTS = [{
be2d40a5	103	'url': 'http://www.canvas.be/video/de-afspraak/najaar-2015/de-afspraak-veilt-voor-de-warmste-week',
117589df	104	'md5': 'ed66976748d12350b118455979cca293',
be2d40a5	105	'info_dict': {
4e2743ab S	106	'id': 'mz-ast-5e5f90b6-2d72-4c40-82c2-e134f884e93e',
4e2743ab S	107	'display_id': 'de-afspraak-veilt-voor-de-warmste-week',
117589df	108	'ext': 'flv',
4e2743ab S	109	'title': 'De afspraak veilt voor de Warmste Week',
4e2743ab S	110	'description': 'md5:24cb860c320dc2be7358e0e5aa317ba6',
ec85ded8	111	'thumbnail': r're:^https?://.*\.jpg$',
4e2743ab	112	'duration': 49.02,
117589df S	113	},
117589df S	114	'expected_warnings': ['is not a supported codec'],
6eff2605 S	115	}, {
	116	# with subtitles
	117	'url': 'http://www.canvas.be/video/panorama/2016/pieter-0167',
	118	'info_dict': {
	119	'id': 'mz-ast-5240ff21-2d30-4101-bba6-92b5ec67c625',
	120	'display_id': 'pieter-0167',
	121	'ext': 'mp4',
	122	'title': 'Pieter 0167',
	123	'description': 'md5:943cd30f48a5d29ba02c3a104dc4ec4e',
ec85ded8	124	'thumbnail': r're:^https?://.*\.jpg$',
6eff2605 S	125	'duration': 2553.08,
	126	'subtitles': {
	127	'nl': [{
	128	'ext': 'vtt',
	129	}],
	130	},
	131	},
	132	'params': {
	133	'skip_download': True,
117589df S	134	},
117589df S	135	'skip': 'Pagina niet gevonden',
bb5ebd44	136	}, {
19a107f2	137	'url': 'https://www.een.be/sorry-voor-alles/herbekijk-sorry-voor-alles',
bb5ebd44	138	'info_dict': {
19a107f2 AG	139	'id': 'mz-ast-11a587f8-b921-4266-82e2-0bce3e80d07f',
19a107f2 AG	140	'display_id': 'herbekijk-sorry-voor-alles',
bb5ebd44	141	'ext': 'mp4',
19a107f2 AG	142	'title': 'Herbekijk Sorry voor alles',
19a107f2 AG	143	'description': 'md5:8bb2805df8164e5eb95d6a7a29dc0dd3',
ec85ded8	144	'thumbnail': r're:^https?://.*\.jpg$',
19a107f2	145	'duration': 3788.06,
bb5ebd44 S	146	},
	147	'params': {
	148	'skip_download': True,
117589df	149	},
19a107f2	150	'skip': 'Episode no longer available',
bb5ebd44 S	151	}, {
	152	'url': 'https://www.canvas.be/check-point/najaar-2016/de-politie-uw-vriend',
	153	'only_matching': True,
6eff2605	154	}]
be2d40a5 TG	155
be2d40a5 TG	156	def _real_extract(self, url):
bb5ebd44 S	157	mobj = re.match(self._VALID_URL, url)
bb5ebd44 S	158	site_id, display_id = mobj.group('site_id'), mobj.group('id')
be2d40a5	159
4e2743ab	160	webpage = self._download_webpage(url, display_id)
be2d40a5	161
117589df	162	title = strip_or_none(self._search_regex(
4e2743ab	163	r'<h1[^>]+class="video__body__header__title"[^>]*>(.+?)</h1>',
bb5ebd44	164	webpage, 'title', default=None) or self._og_search_title(
117589df	165	webpage, default=None))
4e2743ab S	166
4e2743ab S	167	video_id = self._html_search_regex(
117589df S	168	r'data-video=(["\'])(?P<id>(?:(?!\1).)+)\1', webpage, 'video id',
117589df S	169	group='id')
4e2743ab	170
be2d40a5	171	return {
117589df S	172	'_type': 'url_transparent',
	173	'url': 'https://mediazone.vrt.be/api/v1/%s/assets/%s' % (site_id, video_id),
	174	'ie_key': CanvasIE.ie_key(),
be2d40a5	175	'id': video_id,
4e2743ab	176	'display_id': display_id,
be2d40a5	177	'title': title,
4e2743ab	178	'description': self._og_search_description(webpage),
be2d40a5	179	}
7913e0fc	180
	181
	182	class VrtNUIE(GigyaBaseIE):
	183	IE_DESC = 'VrtNU.be'
	184	_VALID_URL = r'https?://(?:www\.)?vrt\.be/(?P<site_id>vrtnu)/(?:[^/]+/)*(?P<id>[^/?#&]+)'
	185	_TESTS = [{
	186	'url': 'https://www.vrt.be/vrtnu/a-z/postbus-x/1/postbus-x-s1a1/',
	187	'info_dict': {
	188	'id': 'pbs-pub-2e2d8c27-df26-45c9-9dc6-90c78153044d$vid-90c932b1-e21d-4fb8-99b1-db7b49cf74de',
19a107f2	189	'ext': 'flv',
7913e0fc	190	'title': 'De zwarte weduwe',
19a107f2	191	'description': 'md5:d90c21dced7db869a85db89a623998d4',
7913e0fc	192	'duration': 1457.04,
7913e0fc	193	'thumbnail': r're:^https?://.*\.jpg$',
19a107f2	194	'season': '1',
7913e0fc	195	'season_number': 1,
	196	'episode_number': 1,
	197	},
19a107f2	198	'skip': 'This video is only available for registered users'
7913e0fc	199	}]
	200	_NETRC_MACHINE = 'vrtnu'
	201	_APIKEY = '3_0Z2HujMtiWq_pkAjgnS2Md2E11a1AwZjYiBETtwNE-EoEHDINgtnvcAOpNgmrVGy'
	202	_CONTEXT_ID = 'R3595707040'
	203
	204	def _real_initialize(self):
	205	self._login()
	206
	207	def _login(self):
	208	username, password = self._get_login_info()
	209	if username is None:
b8c6ffc5	210	return
7913e0fc	211
	212	auth_data = {
	213	'APIKey': self._APIKEY,
	214	'targetEnv': 'jssdk',
	215	'loginID': username,
	216	'password': password,
	217	'authMode': 'cookie',
	218	}
	219
	220	auth_info = self._gigya_login(auth_data)
	221
	222	# Sometimes authentication fails for no good reason, retry
	223	login_attempt = 1
	224	while login_attempt <= 3:
	225	try:
	226	# When requesting a token, no actual token is returned, but the
	227	# necessary cookies are set.
	228	self._request_webpage(
	229	'https://token.vrt.be',
	230	None, note='Requesting a token', errnote='Could not get a token',
	231	headers={
	232	'Content-Type': 'application/json',
	233	'Referer': 'https://www.vrt.be/vrtnu/',
	234	},
	235	data=json.dumps({
	236	'uid': auth_info['UID'],
	237	'uidsig': auth_info['UIDSignature'],
	238	'ts': auth_info['signatureTimestamp'],
	239	'email': auth_info['profile']['email'],
	240	}).encode('utf-8'))
	241	except ExtractorError as e:
	242	if isinstance(e.cause, compat_HTTPError) and e.cause.code == 401:
	243	login_attempt += 1
	244	self.report_warning('Authentication failed')
	245	self._sleep(1, None, msg_template='Waiting for %(timeout)s seconds before trying again')
	246	else:
	247	raise e
	248	else:
	249	break
	250
	251	def _real_extract(self, url):
	252	display_id = self._match_id(url)
	253
64f34528	254	webpage, urlh = self._download_webpage_handle(url, display_id)
7913e0fc	255
fd62b366 S	256	info = self._search_json_ld(webpage, display_id, default={})
	257
	258	# title is optional here since it may be extracted by extractor
	259	# that is delegated from here
	260	title = strip_or_none(self._html_search_regex(
7913e0fc	261	r'(?ms)<h1 class="content__heading">(.+?)</h1>',
fd62b366	262	webpage, 'title', default=None))
7913e0fc	263
	264	description = self._html_search_regex(
	265	r'(?ms)<div class="content__description">(.+?)</div>',
	266	webpage, 'description', default=None)
	267
	268	season = self._html_search_regex(
	269	[r'''(?xms)<div\ class="tabs__tab\ tabs__tab--active">\s*
	270	<span>seizoen\ (.+?)</span>\s*
	271	</div>''',
	272	r'<option value="seizoen (\d{1,3})" data-href="[^"]+?" selected>'],
	273	webpage, 'season', default=None)
	274
	275	season_number = int_or_none(season)
	276
	277	episode_number = int_or_none(self._html_search_regex(
	278	r'''(?xms)<div\ class="content__episode">\s*
	279	<abbr\ title="aflevering">afl</abbr>\s*<span>(\d+)</span>
	280	</div>''',
	281	webpage, 'episode_number', default=None))
	282
	283	release_date = parse_iso8601(self._html_search_regex(
	284	r'(?ms)<div class="content__broadcastdate">\s*<time\ datetime="(.+?)"',
	285	webpage, 'release_date', default=None))
	286
	287	# If there's a ? or a # in the URL, remove them and everything after
64f34528	288	clean_url = urlh.geturl().split('?')[0].split('#')[0].strip('/')
7913e0fc	289	securevideo_url = clean_url + '.mssecurevideo.json'
7913e0fc	290
b8c6ffc5 S	291	try:
	292	video = self._download_json(securevideo_url, display_id)
	293	except ExtractorError as e:
	294	if isinstance(e.cause, compat_HTTPError) and e.cause.code == 401:
	295	self.raise_login_required()
	296	raise
	297
7913e0fc	298	# We are dealing with a '../<show>.relevant' URL
b8c6ffc5	299	redirect_url = video.get('url')
7913e0fc	300	if redirect_url:
b8c6ffc5 S	301	return self.url_result(self._proto_relative_url(redirect_url, 'https:'))
	302
	303	# There is only one entry, but with an unknown key, so just get
	304	# the first one
	305	video_id = list(video.values())[0].get('videoid')
7913e0fc	306
fd62b366	307	return merge_dicts(info, {
7913e0fc	308	'_type': 'url_transparent',
	309	'url': 'https://mediazone.vrt.be/api/v1/vrtvideo/assets/%s' % video_id,
	310	'ie_key': CanvasIE.ie_key(),
	311	'id': video_id,
	312	'display_id': display_id,
	313	'title': title,
	314	'description': description,
	315	'season': season,
	316	'season_number': season_number,
	317	'episode_number': episode_number,
	318	'release_date': release_date,
fd62b366	319	})