[yt-dlp.git] / yt_dlp / extractor / redbee.py

import json
import re
import time
import urllib.parse
import uuid

from .common import InfoExtractor
from ..utils import (
    ExtractorError,
    float_or_none,
    int_or_none,
    strip_or_none,
    traverse_obj,
    try_call,
    unified_timestamp,
)


class RedBeeBaseIE(InfoExtractor):
    _DEVICE_ID = str(uuid.uuid4())

    @property
    def _API_URL(self):
        """
        Ref: https://apidocs.emp.ebsd.ericsson.net
        Subclasses must set _REDBEE_CUSTOMER, _REDBEE_BUSINESS_UNIT
        """
        return f'https://exposure.api.redbee.live/v2/customer/{self._REDBEE_CUSTOMER}/businessunit/{self._REDBEE_BUSINESS_UNIT}'

    def _get_bearer_token(self, asset_id, jwt=None):
        request = {
            'deviceId': self._DEVICE_ID,
            'device': {
                'deviceId': self._DEVICE_ID,
                'name': 'Mozilla Firefox 102',
                'type': 'WEB',
            },
        }
        if jwt:
            request['jwt'] = jwt

        return self._download_json(
            f'{self._API_URL}/auth/{"gigyaLogin" if jwt else "anonymous"}',
            asset_id, data=json.dumps(request).encode('utf-8'), headers={
                'Content-Type': 'application/json;charset=utf-8'
            })['sessionToken']

    def _get_formats_and_subtitles(self, asset_id, **kwargs):
        bearer_token = self._get_bearer_token(asset_id, **kwargs)
        api_response = self._download_json(
            f'{self._API_URL}/entitlement/{asset_id}/play',
            asset_id, headers={
                'Authorization': f'Bearer {bearer_token}',
                'Accept': 'application/json, text/plain, */*'
            })

        formats, subtitles = [], {}
        for format in api_response['formats']:
            if not format.get('mediaLocator'):
                continue

            fmts, subs = [], {}
            if format.get('format') == 'DASH':
                fmts, subs = self._extract_mpd_formats_and_subtitles(
                    format['mediaLocator'], asset_id, fatal=False)
            elif format.get('format') == 'SMOOTHSTREAMING':
                fmts, subs = self._extract_ism_formats_and_subtitles(
                    format['mediaLocator'], asset_id, fatal=False)
            elif format.get('format') == 'HLS':
                fmts, subs = self._extract_m3u8_formats_and_subtitles(
                    format['mediaLocator'], asset_id, fatal=False)

            if format.get('drm'):
                for f in fmts:
                    f['has_drm'] = True

            formats.extend(fmts)
            self._merge_subtitles(subs, target=subtitles)

        return formats, subtitles


class ParliamentLiveUKIE(RedBeeBaseIE):
    IE_NAME = 'parliamentlive.tv'
    IE_DESC = 'UK parliament videos'
    _VALID_URL = r'(?i)https?://(?:www\.)?parliamentlive\.tv/Event/Index/(?P<id>[\da-f]{8}-[\da-f]{4}-[\da-f]{4}-[\da-f]{4}-[\da-f]{12})'

    _REDBEE_CUSTOMER = 'UKParliament'
    _REDBEE_BUSINESS_UNIT = 'ParliamentLive'

    _TESTS = [{
        'url': 'http://parliamentlive.tv/Event/Index/c1e9d44d-fd6c-4263-b50f-97ed26cc998b',
        'info_dict': {
            'id': 'c1e9d44d-fd6c-4263-b50f-97ed26cc998b',
            'ext': 'mp4',
            'title': 'Home Affairs Committee',
            'timestamp': 1395153872,
            'upload_date': '20140318',
            'thumbnail': r're:https?://[^?#]+c1e9d44d-fd6c-4263-b50f-97ed26cc998b[^/]*/thumbnail',
        },
    }, {
        'url': 'http://parliamentlive.tv/event/index/3f24936f-130f-40bf-9a5d-b3d6479da6a4',
        'only_matching': True,
    }, {
        'url': 'https://parliamentlive.tv/Event/Index/27cf25e4-e77b-42a3-93c5-c815cd6d7377',
        'info_dict': {
            'id': '27cf25e4-e77b-42a3-93c5-c815cd6d7377',
            'ext': 'mp4',
            'title': 'House of Commons',
            'timestamp': 1658392447,
            'upload_date': '20220721',
            'thumbnail': r're:https?://[^?#]+27cf25e4-e77b-42a3-93c5-c815cd6d7377[^/]*/thumbnail',
        },
    }]

    def _real_extract(self, url):
        video_id = self._match_id(url)

        formats, subtitles = self._get_formats_and_subtitles(video_id)

        video_info = self._download_json(
            f'https://www.parliamentlive.tv/Event/GetShareVideo/{video_id}', video_id, fatal=False)

        return {
            'id': video_id,
            'formats': formats,
            'subtitles': subtitles,
            'title': traverse_obj(video_info, ('event', 'title')),
            'thumbnail': traverse_obj(video_info, 'thumbnailUrl'),
            'timestamp': traverse_obj(
                video_info, ('event', 'publishedStartTime'), expected_type=unified_timestamp),
            '_format_sort_fields': ('res', 'proto'),
        }


class RTBFIE(RedBeeBaseIE):
    _VALID_URL = r'''(?x)
        https?://(?:www\.)?rtbf\.be/
        (?:
            video/[^?]+\?.*\bid=|
            ouftivi/(?:[^/]+/)*[^?]+\?.*\bvideoId=|
            auvio/[^/]+\?.*\b(?P<live>l)?id=
        )(?P<id>\d+)'''
    _NETRC_MACHINE = 'rtbf'

    _REDBEE_CUSTOMER = 'RTBF'
    _REDBEE_BUSINESS_UNIT = 'Auvio'

    _TESTS = [{
        'url': 'https://www.rtbf.be/video/detail_les-diables-au-coeur-episode-2?id=1921274',
        'md5': '8c876a1cceeb6cf31b476461ade72384',
        'info_dict': {
            'id': '1921274',
            'ext': 'mp4',
            'title': 'Les Diables au coeur (épisode 2)',
            'description': '(du 25/04/2014)',
            'duration': 3099.54,
            'upload_date': '20140425',
            'timestamp': 1398456300,
        },
        'skip': 'No longer available',
    }, {
        # geo restricted
        'url': 'http://www.rtbf.be/ouftivi/heros/detail_scooby-doo-mysteres-associes?id=1097&videoId=2057442',
        'only_matching': True,
    }, {
        'url': 'http://www.rtbf.be/ouftivi/niouzz?videoId=2055858',
        'only_matching': True,
    }, {
        'url': 'http://www.rtbf.be/auvio/detail_jeudi-en-prime-siegfried-bracke?id=2102996',
        'only_matching': True,
    }, {
        # Live
        'url': 'https://www.rtbf.be/auvio/direct_pure-fm?lid=134775',
        'only_matching': True,
    }, {
        # Audio
        'url': 'https://www.rtbf.be/auvio/detail_cinq-heures-cinema?id=2360811',
        'only_matching': True,
    }, {
        # With Subtitle
        'url': 'https://www.rtbf.be/auvio/detail_les-carnets-du-bourlingueur?id=2361588',
        'only_matching': True,
    }, {
        'url': 'https://www.rtbf.be/auvio/detail_investigation?id=2921926',
        'md5': 'd5d11bb62169fef38d7ce7ac531e034f',
        'info_dict': {
            'id': '2921926',
            'ext': 'mp4',
            'title': 'Le handicap un confinement perpétuel - Maladie de Lyme',
            'description': 'md5:dcbd5dcf6015488c9069b057c15ccc52',
            'duration': 5258.8,
            'upload_date': '20220727',
            'timestamp': 1658934000,
            'series': '#Investigation',
            'thumbnail': r're:^https?://[^?&]+\.jpg$',
        },
    }, {
        'url': 'https://www.rtbf.be/auvio/detail_la-belgique-criminelle?id=2920492',
        'md5': '054f9f143bc79c89647c35e5a7d35fa8',
        'info_dict': {
            'id': '2920492',
            'ext': 'mp4',
            'title': '04 - Le crime de la rue Royale',
            'description': 'md5:0c3da1efab286df83f2ab3f8f96bd7a6',
            'duration': 1574.6,
            'upload_date': '20220723',
            'timestamp': 1658596887,
            'series': 'La Belgique criminelle - TV',
            'thumbnail': r're:^https?://[^?&]+\.jpg$',
        },
    }]

    _IMAGE_HOST = 'http://ds1.ds.static.rtbf.be'
    _PROVIDERS = {
        'YOUTUBE': 'Youtube',
        'DAILYMOTION': 'Dailymotion',
        'VIMEO': 'Vimeo',
    }
    _QUALITIES = [
        ('mobile', 'SD'),
        ('web', 'MD'),
        ('high', 'HD'),
    ]
    _LOGIN_URL = 'https://login.rtbf.be/accounts.login'
    _GIGYA_API_KEY = '3_kWKuPgcdAybqnqxq_MvHVk0-6PN8Zk8pIIkJM_yXOu-qLPDDsGOtIDFfpGivtbeO'
    _LOGIN_COOKIE_ID = f'glt_{_GIGYA_API_KEY}'

    def _perform_login(self, username, password):
        if self._get_cookies(self._LOGIN_URL).get(self._LOGIN_COOKIE_ID):
            return

        self._set_cookie('.rtbf.be', 'gmid', 'gmid.ver4', secure=True, expire_time=time.time() + 3600)

        login_response = self._download_json(
            self._LOGIN_URL, None, data=urllib.parse.urlencode({
                'loginID': username,
                'password': password,
                'APIKey': self._GIGYA_API_KEY,
                'targetEnv': 'jssdk',
                'sessionExpiration': '-2',
            }).encode('utf-8'), headers={
                'Content-Type': 'application/x-www-form-urlencoded',
            })

        if login_response['statusCode'] != 200:
            raise ExtractorError('Login failed. Server message: %s' % login_response['errorMessage'], expected=True)

        self._set_cookie('.rtbf.be', self._LOGIN_COOKIE_ID, login_response['sessionInfo']['login_token'],
                         secure=True, expire_time=time.time() + 3600)

    def _get_formats_and_subtitles(self, url, media_id):
        login_token = self._get_cookies(url).get(self._LOGIN_COOKIE_ID)
        if not login_token:
            self.raise_login_required()

        session_jwt = try_call(lambda: self._get_cookies(url)['rtbf_jwt'].value) or self._download_json(
            'https://login.rtbf.be/accounts.getJWT', media_id, query={
                'login_token': login_token.value,
                'APIKey': self._GIGYA_API_KEY,
                'sdk': 'js_latest',
                'authMode': 'cookie',
                'pageURL': url,
                'sdkBuild': '13273',
                'format': 'json',
            })['id_token']

        return super()._get_formats_and_subtitles(media_id, jwt=session_jwt)

    def _real_extract(self, url):
        live, media_id = self._match_valid_url(url).groups()
        embed_page = self._download_webpage(
            'https://www.rtbf.be/auvio/embed/' + ('direct' if live else 'media'),
            media_id, query={'id': media_id})

        media_data = self._html_search_regex(r'data-media="([^"]+)"', embed_page, 'media data', fatal=False)
        if not media_data:
            if re.search(r'<div[^>]+id="js-error-expired"[^>]+class="(?![^"]*hidden)', embed_page):
                raise ExtractorError('Livestream has ended.', expected=True)
            if re.search(r'<div[^>]+id="js-sso-connect"[^>]+class="(?![^"]*hidden)', embed_page):
                self.raise_login_required()

            raise ExtractorError('Could not find media data')

        data = self._parse_json(media_data, media_id)

        error = data.get('error')
        if error:
            raise ExtractorError('%s said: %s' % (self.IE_NAME, error), expected=True)

        provider = data.get('provider')
        if provider in self._PROVIDERS:
            return self.url_result(data['url'], self._PROVIDERS[provider])

        title = traverse_obj(data, 'subtitle', 'title')
        is_live = data.get('isLive')
        height_re = r'-(\d+)p\.'
        formats, subtitles = [], {}

        # The old api still returns m3u8 and mpd manifest for livestreams, but these are 'fake'
        # since all they contain is a 20s video that is completely unrelated.
        # https://github.com/yt-dlp/yt-dlp/issues/4656#issuecomment-1214461092
        m3u8_url = None if data.get('isLive') else traverse_obj(data, 'urlHlsAes128', 'urlHls')
        if m3u8_url:
            fmts, subs = self._extract_m3u8_formats_and_subtitles(
                m3u8_url, media_id, 'mp4', m3u8_id='hls', fatal=False)
            formats.extend(fmts)
            self._merge_subtitles(subs, target=subtitles)

        fix_url = lambda x: x.replace('//rtbf-vod.', '//rtbf.') if '/geo/drm/' in x else x
        http_url = data.get('url')
        if formats and http_url and re.search(height_re, http_url):
            http_url = fix_url(http_url)
            for m3u8_f in formats[:]:
                height = m3u8_f.get('height')
                if not height:
                    continue
                f = m3u8_f.copy()
                del f['protocol']
                f.update({
                    'format_id': m3u8_f['format_id'].replace('hls-', 'http-'),
                    'url': re.sub(height_re, '-%dp.' % height, http_url),
                })
                formats.append(f)
        else:
            sources = data.get('sources') or {}
            for key, format_id in self._QUALITIES:
                format_url = sources.get(key)
                if not format_url:
                    continue
                height = int_or_none(self._search_regex(
                    height_re, format_url, 'height', default=None))
                formats.append({
                    'format_id': format_id,
                    'url': fix_url(format_url),
                    'height': height,
                })

        mpd_url = None if data.get('isLive') else data.get('urlDash')
        if mpd_url and (self.get_param('allow_unplayable_formats') or not data.get('drm')):
            fmts, subs = self._extract_mpd_formats_and_subtitles(
                mpd_url, media_id, mpd_id='dash', fatal=False)
            formats.extend(fmts)
            self._merge_subtitles(subs, target=subtitles)

        audio_url = data.get('urlAudio')
        if audio_url:
            formats.append({
                'format_id': 'audio',
                'url': audio_url,
                'vcodec': 'none',
            })

        for track in (data.get('tracks') or {}).values():
            sub_url = track.get('url')
            if not sub_url:
                continue
            subtitles.setdefault(track.get('lang') or 'fr', []).append({
                'url': sub_url,
            })

        if not formats:
            fmts, subs = self._get_formats_and_subtitles(url, f'live_{media_id}' if is_live else media_id)
            formats.extend(fmts)
            self._merge_subtitles(subs, target=subtitles)

        return {
            'id': media_id,
            'formats': formats,
            'title': title,
            'description': strip_or_none(data.get('description')),
            'thumbnail': data.get('thumbnail'),
            'duration': float_or_none(data.get('realDuration')),
            'timestamp': int_or_none(data.get('liveFrom')),
            'series': data.get('programLabel'),
            'subtitles': subtitles,
            'is_live': is_live,
            '_format_sort_fields': ('res', 'proto'),
        }
Commit	Line	Data
2a5e5477 E	1	import json
	2	import re
	3	import time
	4	import urllib.parse
	5	import uuid
	6
	7	from .common import InfoExtractor
	8	from ..utils import (
	9	ExtractorError,
	10	float_or_none,
	11	int_or_none,
	12	strip_or_none,
	13	traverse_obj,
b85703d1	14	try_call,
2a5e5477 E	15	unified_timestamp,
	16	)
	17
	18
	19	class RedBeeBaseIE(InfoExtractor):
	20	_DEVICE_ID = str(uuid.uuid4())
	21
	22	@property
	23	def _API_URL(self):
	24	"""
	25	Ref: https://apidocs.emp.ebsd.ericsson.net
	26	Subclasses must set _REDBEE_CUSTOMER, _REDBEE_BUSINESS_UNIT
	27	"""
	28	return f'https://exposure.api.redbee.live/v2/customer/{self._REDBEE_CUSTOMER}/businessunit/{self._REDBEE_BUSINESS_UNIT}'
	29
	30	def _get_bearer_token(self, asset_id, jwt=None):
	31	request = {
	32	'deviceId': self._DEVICE_ID,
	33	'device': {
	34	'deviceId': self._DEVICE_ID,
	35	'name': 'Mozilla Firefox 102',
	36	'type': 'WEB',
	37	},
	38	}
	39	if jwt:
	40	request['jwt'] = jwt
	41
	42	return self._download_json(
	43	f'{self._API_URL}/auth/{"gigyaLogin" if jwt else "anonymous"}',
	44	asset_id, data=json.dumps(request).encode('utf-8'), headers={
	45	'Content-Type': 'application/json;charset=utf-8'
	46	})['sessionToken']
	47
	48	def _get_formats_and_subtitles(self, asset_id, **kwargs):
	49	bearer_token = self._get_bearer_token(asset_id, **kwargs)
	50	api_response = self._download_json(
	51	f'{self._API_URL}/entitlement/{asset_id}/play',
	52	asset_id, headers={
	53	'Authorization': f'Bearer {bearer_token}',
	54	'Accept': 'application/json, text/plain, /'
	55	})
	56
	57	formats, subtitles = [], {}
	58	for format in api_response['formats']:
	59	if not format.get('mediaLocator'):
	60	continue
	61
	62	fmts, subs = [], {}
	63	if format.get('format') == 'DASH':
	64	fmts, subs = self._extract_mpd_formats_and_subtitles(
	65	format['mediaLocator'], asset_id, fatal=False)
	66	elif format.get('format') == 'SMOOTHSTREAMING':
	67	fmts, subs = self._extract_ism_formats_and_subtitles(
	68	format['mediaLocator'], asset_id, fatal=False)
	69	elif format.get('format') == 'HLS':
	70	fmts, subs = self._extract_m3u8_formats_and_subtitles(
	71	format['mediaLocator'], asset_id, fatal=False)
	72
2b3e43e2 E	73	if format.get('drm'):
	74	for f in fmts:
	75	f['has_drm'] = True
	76
2a5e5477 E	77	formats.extend(fmts)
	78	self._merge_subtitles(subs, target=subtitles)
	79
	80	return formats, subtitles
	81
	82
	83	class ParliamentLiveUKIE(RedBeeBaseIE):
	84	IE_NAME = 'parliamentlive.tv'
	85	IE_DESC = 'UK parliament videos'
	86	_VALID_URL = r'(?i)https?://(?:www\.)?parliamentlive\.tv/Event/Index/(?P<id>[\da-f]{8}-[\da-f]{4}-[\da-f]{4}-[\da-f]{4}-[\da-f]{12})'
	87
	88	_REDBEE_CUSTOMER = 'UKParliament'
	89	_REDBEE_BUSINESS_UNIT = 'ParliamentLive'
	90
	91	_TESTS = [{
	92	'url': 'http://parliamentlive.tv/Event/Index/c1e9d44d-fd6c-4263-b50f-97ed26cc998b',
	93	'info_dict': {
	94	'id': 'c1e9d44d-fd6c-4263-b50f-97ed26cc998b',
	95	'ext': 'mp4',
	96	'title': 'Home Affairs Committee',
	97	'timestamp': 1395153872,
	98	'upload_date': '20140318',
	99	'thumbnail': r're:https?://[^?#]+c1e9d44d-fd6c-4263-b50f-97ed26cc998b[^/]*/thumbnail',
	100	},
	101	}, {
	102	'url': 'http://parliamentlive.tv/event/index/3f24936f-130f-40bf-9a5d-b3d6479da6a4',
	103	'only_matching': True,
	104	}, {
	105	'url': 'https://parliamentlive.tv/Event/Index/27cf25e4-e77b-42a3-93c5-c815cd6d7377',
	106	'info_dict': {
	107	'id': '27cf25e4-e77b-42a3-93c5-c815cd6d7377',
	108	'ext': 'mp4',
	109	'title': 'House of Commons',
	110	'timestamp': 1658392447,
	111	'upload_date': '20220721',
	112	'thumbnail': r're:https?://[^?#]+27cf25e4-e77b-42a3-93c5-c815cd6d7377[^/]*/thumbnail',
	113	},
	114	}]
	115
	116	def _real_extract(self, url):
	117	video_id = self._match_id(url)
	118
	119	formats, subtitles = self._get_formats_and_subtitles(video_id)
2a5e5477 E	120
	121	video_info = self._download_json(
	122	f'https://www.parliamentlive.tv/Event/GetShareVideo/{video_id}', video_id, fatal=False)
	123
2a5e5477 E	124	return {
	125	'id': video_id,
	126	'formats': formats,
	127	'subtitles': subtitles,
	128	'title': traverse_obj(video_info, ('event', 'title')),
	129	'thumbnail': traverse_obj(video_info, 'thumbnailUrl'),
	130	'timestamp': traverse_obj(
	131	video_info, ('event', 'publishedStartTime'), expected_type=unified_timestamp),
9f14daf2	132	'_format_sort_fields': ('res', 'proto'),
2a5e5477 E	133	}
	134
	135
	136	class RTBFIE(RedBeeBaseIE):
	137	_VALID_URL = r'''(?x)
	138	https?://(?:www\.)?rtbf\.be/
	139	(?:
	140	video/[^?]+\?.*\bid=\|
	141	ouftivi/(?:[^/]+/)[^?]+\?.\bvideoId=\|
	142	auvio/[^/]+\?.*\b(?P<live>l)?id=
	143	)(?P<id>\d+)'''
	144	_NETRC_MACHINE = 'rtbf'
	145
	146	_REDBEE_CUSTOMER = 'RTBF'
	147	_REDBEE_BUSINESS_UNIT = 'Auvio'
	148
	149	_TESTS = [{
	150	'url': 'https://www.rtbf.be/video/detail_les-diables-au-coeur-episode-2?id=1921274',
	151	'md5': '8c876a1cceeb6cf31b476461ade72384',
	152	'info_dict': {
	153	'id': '1921274',
	154	'ext': 'mp4',
	155	'title': 'Les Diables au coeur (épisode 2)',
	156	'description': '(du 25/04/2014)',
	157	'duration': 3099.54,
	158	'upload_date': '20140425',
	159	'timestamp': 1398456300,
	160	},
	161	'skip': 'No longer available',
	162	}, {
	163	# geo restricted
	164	'url': 'http://www.rtbf.be/ouftivi/heros/detail_scooby-doo-mysteres-associes?id=1097&videoId=2057442',
	165	'only_matching': True,
	166	}, {
	167	'url': 'http://www.rtbf.be/ouftivi/niouzz?videoId=2055858',
	168	'only_matching': True,
	169	}, {
	170	'url': 'http://www.rtbf.be/auvio/detail_jeudi-en-prime-siegfried-bracke?id=2102996',
	171	'only_matching': True,
	172	}, {
	173	# Live
	174	'url': 'https://www.rtbf.be/auvio/direct_pure-fm?lid=134775',
	175	'only_matching': True,
	176	}, {
	177	# Audio
	178	'url': 'https://www.rtbf.be/auvio/detail_cinq-heures-cinema?id=2360811',
	179	'only_matching': True,
	180	}, {
	181	# With Subtitle
	182	'url': 'https://www.rtbf.be/auvio/detail_les-carnets-du-bourlingueur?id=2361588',
	183	'only_matching': True,
	184	}, {
	185	'url': 'https://www.rtbf.be/auvio/detail_investigation?id=2921926',
	186	'md5': 'd5d11bb62169fef38d7ce7ac531e034f',
	187	'info_dict': {
	188	'id': '2921926',
	189	'ext': 'mp4',
	190	'title': 'Le handicap un confinement perpétuel - Maladie de Lyme',
	191	'description': 'md5:dcbd5dcf6015488c9069b057c15ccc52',
	192	'duration': 5258.8,
	193	'upload_date': '20220727',
	194	'timestamp': 1658934000,
	195	'series': '#Investigation',
	196	'thumbnail': r're:^https?://[^?&]+\.jpg$',
197	},
198	}, {
199	'url': 'https://www.rtbf.be/auvio/detail_la-belgique-criminelle?id=2920492',
200	'md5': '054f9f143bc79c89647c35e5a7d35fa8',
201	'info_dict': {
202	'id': '2920492',
203	'ext': 'mp4',
204	'title': '04 - Le crime de la rue Royale',
205	'description': 'md5:0c3da1efab286df83f2ab3f8f96bd7a6',
206	'duration': 1574.6,
207	'upload_date': '20220723',
208	'timestamp': 1658596887,
209	'series': 'La Belgique criminelle - TV',
210	'thumbnail': r're:^https?://[^?&]+\.jpg$',
211	},
212	}]
213
214	_IMAGE_HOST = 'http://ds1.ds.static.rtbf.be'
215	_PROVIDERS = {
216	'YOUTUBE': 'Youtube',
217	'DAILYMOTION': 'Dailymotion',
218	'VIMEO': 'Vimeo',
219	}
220	_QUALITIES = [
221	('mobile', 'SD'),
222	('web', 'MD'),
223	('high', 'HD'),
224	]
225	_LOGIN_URL = 'https://login.rtbf.be/accounts.login'
226	_GIGYA_API_KEY = '3_kWKuPgcdAybqnqxq_MvHVk0-6PN8Zk8pIIkJM_yXOu-qLPDDsGOtIDFfpGivtbeO'
227	_LOGIN_COOKIE_ID = f'glt_{_GIGYA_API_KEY}'
228
229	def _perform_login(self, username, password):
230	if self._get_cookies(self._LOGIN_URL).get(self._LOGIN_COOKIE_ID):
231	return
232
233	self._set_cookie('.rtbf.be', 'gmid', 'gmid.ver4', secure=True, expire_time=time.time() + 3600)
234
235	login_response = self._download_json(
236	self._LOGIN_URL, None, data=urllib.parse.urlencode({
237	'loginID': username,
238	'password': password,
239	'APIKey': self._GIGYA_API_KEY,
240	'targetEnv': 'jssdk',
241	'sessionExpiration': '-2',
242	}).encode('utf-8'), headers={
243	'Content-Type': 'application/x-www-form-urlencoded',
244	})
245
246	if login_response['statusCode'] != 200:
247	raise ExtractorError('Login failed. Server message: %s' % login_response['errorMessage'], expected=True)
248
249	self._set_cookie('.rtbf.be', self._LOGIN_COOKIE_ID, login_response['sessionInfo']['login_token'],
250	secure=True, expire_time=time.time() + 3600)
251
252	def _get_formats_and_subtitles(self, url, media_id):
253	login_token = self._get_cookies(url).get(self._LOGIN_COOKIE_ID)
254	if not login_token:
255	self.raise_login_required()
256
b85703d1	257	session_jwt = try_call(lambda: self._get_cookies(url)['rtbf_jwt'].value) or self._download_json(
2a5e5477 E	258	'https://login.rtbf.be/accounts.getJWT', media_id, query={
	259	'login_token': login_token.value,
	260	'APIKey': self._GIGYA_API_KEY,
	261	'sdk': 'js_latest',
	262	'authMode': 'cookie',
	263	'pageURL': url,
	264	'sdkBuild': '13273',
	265	'format': 'json',
	266	})['id_token']
	267
	268	return super()._get_formats_and_subtitles(media_id, jwt=session_jwt)
	269
	270	def _real_extract(self, url):
	271	live, media_id = self._match_valid_url(url).groups()
	272	embed_page = self._download_webpage(
	273	'https://www.rtbf.be/auvio/embed/' + ('direct' if live else 'media'),
	274	media_id, query={'id': media_id})
2b3e43e2 E	275
	276	media_data = self._html_search_regex(r'data-media="([^"]+)"', embed_page, 'media data', fatal=False)
	277	if not media_data:
	278	if re.search(r'<div[^>]+id="js-error-expired"[^>]+class="(?![^"]*hidden)', embed_page):
	279	raise ExtractorError('Livestream has ended.', expected=True)
	280	if re.search(r'<div[^>]+id="js-sso-connect"[^>]+class="(?![^"]*hidden)', embed_page):
	281	self.raise_login_required()
	282
	283	raise ExtractorError('Could not find media data')
	284
	285	data = self._parse_json(media_data, media_id)
2a5e5477 E	286
	287	error = data.get('error')
	288	if error:
	289	raise ExtractorError('%s said: %s' % (self.IE_NAME, error), expected=True)
	290
	291	provider = data.get('provider')
	292	if provider in self._PROVIDERS:
	293	return self.url_result(data['url'], self._PROVIDERS[provider])
	294
2b3e43e2	295	title = traverse_obj(data, 'subtitle', 'title')
2a5e5477 E	296	is_live = data.get('isLive')
2a5e5477 E	297	height_re = r'-(\d+)p\.'
2b3e43e2	298	formats, subtitles = [], {}
2a5e5477	299
2b3e43e2 E	300	# The old api still returns m3u8 and mpd manifest for livestreams, but these are 'fake'
	301	# since all they contain is a 20s video that is completely unrelated.
	302	# https://github.com/yt-dlp/yt-dlp/issues/4656#issuecomment-1214461092
	303	m3u8_url = None if data.get('isLive') else traverse_obj(data, 'urlHlsAes128', 'urlHls')
2a5e5477	304	if m3u8_url:
2b3e43e2 E	305	fmts, subs = self._extract_m3u8_formats_and_subtitles(
	306	m3u8_url, media_id, 'mp4', m3u8_id='hls', fatal=False)
	307	formats.extend(fmts)
	308	self._merge_subtitles(subs, target=subtitles)
2a5e5477 E	309
	310	fix_url = lambda x: x.replace('//rtbf-vod.', '//rtbf.') if '/geo/drm/' in x else x
	311	http_url = data.get('url')
	312	if formats and http_url and re.search(height_re, http_url):
	313	http_url = fix_url(http_url)
	314	for m3u8_f in formats[:]:
	315	height = m3u8_f.get('height')
	316	if not height:
	317	continue
	318	f = m3u8_f.copy()
	319	del f['protocol']
	320	f.update({
	321	'format_id': m3u8_f['format_id'].replace('hls-', 'http-'),
	322	'url': re.sub(height_re, '-%dp.' % height, http_url),
	323	})
	324	formats.append(f)
	325	else:
	326	sources = data.get('sources') or {}
	327	for key, format_id in self._QUALITIES:
	328	format_url = sources.get(key)
	329	if not format_url:
	330	continue
	331	height = int_or_none(self._search_regex(
	332	height_re, format_url, 'height', default=None))
	333	formats.append({
	334	'format_id': format_id,
	335	'url': fix_url(format_url),
	336	'height': height,
	337	})
	338
2b3e43e2	339	mpd_url = None if data.get('isLive') else data.get('urlDash')
2a5e5477	340	if mpd_url and (self.get_param('allow_unplayable_formats') or not data.get('drm')):
2b3e43e2 E	341	fmts, subs = self._extract_mpd_formats_and_subtitles(
	342	mpd_url, media_id, mpd_id='dash', fatal=False)
	343	formats.extend(fmts)
	344	self._merge_subtitles(subs, target=subtitles)
2a5e5477 E	345
	346	audio_url = data.get('urlAudio')
	347	if audio_url:
	348	formats.append({
	349	'format_id': 'audio',
	350	'url': audio_url,
	351	'vcodec': 'none',
	352	})
	353
2a5e5477 E	354	for track in (data.get('tracks') or {}).values():
	355	sub_url = track.get('url')
	356	if not sub_url:
	357	continue
	358	subtitles.setdefault(track.get('lang') or 'fr', []).append({
	359	'url': sub_url,
	360	})
	361
	362	if not formats:
2b3e43e2	363	fmts, subs = self._get_formats_and_subtitles(url, f'live_{media_id}' if is_live else media_id)
2a5e5477 E	364	formats.extend(fmts)
	365	self._merge_subtitles(subs, target=subtitles)
	366
2a5e5477 E	367	return {
	368	'id': media_id,
	369	'formats': formats,
	370	'title': title,
	371	'description': strip_or_none(data.get('description')),
	372	'thumbnail': data.get('thumbnail'),
	373	'duration': float_or_none(data.get('realDuration')),
	374	'timestamp': int_or_none(data.get('liveFrom')),
	375	'series': data.get('programLabel'),
	376	'subtitles': subtitles,
	377	'is_live': is_live,
9f14daf2	378	'_format_sort_fields': ('res', 'proto'),
2a5e5477	379	}