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

import re
import urllib.parse

from .common import InfoExtractor
from .dailymotion import DailymotionIE
from ..networking import HEADRequest
from ..utils import (
    clean_html,
    determine_ext,
    filter_dict,
    format_field,
    int_or_none,
    join_nonempty,
    parse_iso8601,
    smuggle_url,
    unsmuggle_url,
    url_or_none,
)
from ..utils.traversal import traverse_obj


class FranceTVBaseInfoExtractor(InfoExtractor):
    def _make_url_result(self, video_id, url=None):
        video_id = video_id.split('@')[0]  # for compat with old @catalog IDs
        full_id = f'francetv:{video_id}'
        if url:
            full_id = smuggle_url(full_id, {'hostname': urllib.parse.urlparse(url).hostname})
        return self.url_result(full_id, FranceTVIE, video_id)


class FranceTVIE(InfoExtractor):
    _VALID_URL = r'francetv:(?P<id>[^@#]+)'
    _GEO_COUNTRIES = ['FR']
    _GEO_BYPASS = False

    _TESTS = [{
        # tokenized url is in dinfo['video']['token']
        'url': 'francetv:ec217ecc-0733-48cf-ac06-af1347b849d1',
        'info_dict': {
            'id': 'ec217ecc-0733-48cf-ac06-af1347b849d1',
            'ext': 'mp4',
            'title': '13h15, le dimanche... - Les mystères de Jésus',
            'timestamp': 1502623500,
            'duration': 2580,
            'thumbnail': r're:^https?://.*\.jpg$',
            'upload_date': '20170813',
        },
        'params': {'skip_download': 'm3u8'},
    }, {
        # tokenized url is in dinfo['video']['token']['akamai']
        'url': 'francetv:c5bda21d-2c6f-4470-8849-3d8327adb2ba',
        'info_dict': {
            'id': 'c5bda21d-2c6f-4470-8849-3d8327adb2ba',
            'ext': 'mp4',
            'title': '13h15, le dimanche... - Les mystères de Jésus',
            'timestamp': 1514118300,
            'duration': 2880,
            'thumbnail': r're:^https?://.*\.jpg$',
            'upload_date': '20171224',
        },
        'params': {'skip_download': 'm3u8'},
    }, {
        'url': 'francetv:162311093',
        'only_matching': True,
    }, {
        'url': 'francetv:NI_1004933@Zouzous',
        'only_matching': True,
    }, {
        'url': 'francetv:NI_983319@Info-web',
        'only_matching': True,
    }, {
        'url': 'francetv:NI_983319',
        'only_matching': True,
    }, {
        'url': 'francetv:NI_657393@Regions',
        'only_matching': True,
    }, {
        # france-3 live
        'url': 'francetv:SIM_France3',
        'only_matching': True,
    }]

    def _extract_video(self, video_id, hostname=None):
        is_live = None
        videos = []
        drm_formats = False
        title = None
        subtitle = None
        episode_number = None
        season_number = None
        image = None
        duration = None
        timestamp = None
        spritesheets = None

        # desktop+chrome returns dash; mobile+safari returns hls
        for device_type, browser in [('desktop', 'chrome'), ('mobile', 'safari')]:
            dinfo = self._download_json(
                f'https://k7.ftven.fr/videos/{video_id}', video_id,
                f'Downloading {device_type} {browser} video JSON', query=filter_dict({
                    'device_type': device_type,
                    'browser': browser,
                    'domain': hostname,
                }), fatal=False, expected_status=422)  # 422 json gives detailed error code/message

            if not dinfo:
                continue

            if video := traverse_obj(dinfo, ('video', {dict})):
                videos.append(video)
                if duration is None:
                    duration = video.get('duration')
                if is_live is None:
                    is_live = video.get('is_live')
                if spritesheets is None:
                    spritesheets = video.get('spritesheets')
            elif code := traverse_obj(dinfo, ('code', {int})):
                if code == 2009:
                    self.raise_geo_restricted(countries=self._GEO_COUNTRIES)
                elif code in (2015, 2017):
                    # 2015: L'accès à cette vidéo est impossible. (DRM-only)
                    # 2017: Cette vidéo n'est pas disponible depuis le site web mobile (b/c DRM)
                    drm_formats = True
                    continue
                self.report_warning(
                    f'{self.IE_NAME} said: {code} "{clean_html(dinfo.get("message"))}"')
                continue

            if meta := traverse_obj(dinfo, ('meta', {dict})):
                if title is None:
                    title = meta.get('title')
                # meta['pre_title'] contains season and episode number for series in format "S<ID> E<ID>"
                season_number, episode_number = self._search_regex(
                    r'S(\d+)\s*E(\d+)', meta.get('pre_title'), 'episode info', group=(1, 2), default=(None, None))
                if subtitle is None:
                    subtitle = meta.get('additional_title')
                if image is None:
                    image = meta.get('image_url')
                if timestamp is None:
                    timestamp = parse_iso8601(meta.get('broadcasted_at'))

        if not videos and drm_formats:
            self.report_drm(video_id)

        formats, subtitles, video_url = [], {}, None
        for video in traverse_obj(videos, lambda _, v: url_or_none(v['url'])):
            video_url = video['url']
            format_id = video.get('format')

            if token_url := traverse_obj(video, ('token', (None, 'akamai'), {url_or_none}, any)):
                tokenized_url = traverse_obj(self._download_json(
                    token_url, video_id, f'Downloading signed {format_id} manifest URL',
                    fatal=False, query={
                        'format': 'json',
                        'url': video_url,
                    }), ('url', {url_or_none}))
                if tokenized_url:
                    video_url = tokenized_url

            ext = determine_ext(video_url)
            if ext == 'f4m':
                formats.extend(self._extract_f4m_formats(
                    video_url, video_id, f4m_id=format_id or ext, fatal=False))
            elif ext == 'm3u8':
                format_id = format_id or 'hls'
                fmts, subs = self._extract_m3u8_formats_and_subtitles(
                    video_url, video_id, 'mp4', m3u8_id=format_id, fatal=False)
                for f in traverse_obj(fmts, lambda _, v: v['vcodec'] == 'none' and v.get('tbr') is None):
                    if mobj := re.match(rf'{format_id}-[Aa]udio-\w+-(?P<bitrate>\d+)', f['format_id']):
                        f.update({
                            'tbr': int_or_none(mobj.group('bitrate')),
                            'acodec': 'mp4a',
                        })
                formats.extend(fmts)
                self._merge_subtitles(subs, target=subtitles)
            elif ext == 'mpd':
                fmts, subs = self._extract_mpd_formats_and_subtitles(
                    video_url, video_id, mpd_id=format_id or 'dash', fatal=False)
                formats.extend(fmts)
                self._merge_subtitles(subs, target=subtitles)
            elif video_url.startswith('rtmp'):
                formats.append({
                    'url': video_url,
                    'format_id': join_nonempty('rtmp', format_id),
                    'ext': 'flv',
                })
            else:
                if self._is_valid_url(video_url, video_id, format_id):
                    formats.append({
                        'url': video_url,
                        'format_id': format_id,
                    })

            # XXX: what is video['captions']?

        if not formats and video_url:
            urlh = self._request_webpage(
                HEADRequest(video_url), video_id, 'Checking for geo-restriction',
                fatal=False, expected_status=403)
            if urlh and urlh.headers.get('x-errortype') == 'geo':
                self.raise_geo_restricted(countries=self._GEO_COUNTRIES, metadata_available=True)

        for f in formats:
            if f.get('acodec') != 'none' and f.get('language') in ('qtz', 'qad'):
                f['language_preference'] = -10
                f['format_note'] = 'audio description{}'.format(format_field(f, 'format_note', ', %s'))

        if spritesheets:
            formats.append({
                'format_id': 'spritesheets',
                'format_note': 'storyboard',
                'acodec': 'none',
                'vcodec': 'none',
                'ext': 'mhtml',
                'protocol': 'mhtml',
                'url': 'about:invalid',
                'fragments': [{
                    'url': sheet,
                    # XXX: not entirely accurate; each spritesheet seems to be
                    # a 10x10 grid of thumbnails corresponding to approximately
                    # 2 seconds of the video; the last spritesheet may be shorter
                    'duration': 200,
                } for sheet in traverse_obj(spritesheets, (..., {url_or_none}))],
            })

        return {
            'id': video_id,
            'title': join_nonempty(title, subtitle, delim=' - ').strip(),
            'thumbnail': image,
            'duration': duration,
            'timestamp': timestamp,
            'is_live': is_live,
            'formats': formats,
            'subtitles': subtitles,
            'episode': subtitle if episode_number else None,
            'series': title if episode_number else None,
            'episode_number': int_or_none(episode_number),
            'season_number': int_or_none(season_number),
            '_format_sort_fields': ('res', 'tbr', 'proto'),  # prioritize m3u8 over dash
        }

    def _real_extract(self, url):
        url, smuggled_data = unsmuggle_url(url, {})
        video_id = self._match_id(url)
        hostname = smuggled_data.get('hostname') or 'www.france.tv'

        return self._extract_video(video_id, hostname=hostname)


class FranceTVSiteIE(FranceTVBaseInfoExtractor):
    _VALID_URL = r'https?://(?:(?:www\.)?france\.tv|mobile\.france\.tv)/(?:[^/]+/)*(?P<id>[^/]+)\.html'

    _TESTS = [{
        'url': 'https://www.france.tv/france-2/13h15-le-dimanche/140921-les-mysteres-de-jesus.html',
        'info_dict': {
            'id': 'c5bda21d-2c6f-4470-8849-3d8327adb2ba',
            'ext': 'mp4',
            'title': '13h15, le dimanche... - Les mystères de Jésus',
            'timestamp': 1514118300,
            'duration': 2880,
            'thumbnail': r're:^https?://.*\.jpg$',
            'upload_date': '20171224',
        },
        'params': {
            'skip_download': True,
        },
        'add_ie': [FranceTVIE.ie_key()],
    }, {
        # geo-restricted
        'url': 'https://www.france.tv/enfants/six-huit-ans/foot2rue/saison-1/3066387-duel-au-vieux-port.html',
        'info_dict': {
            'id': 'a9050959-eedd-4b4a-9b0d-de6eeaa73e44',
            'ext': 'mp4',
            'title': 'Foot2Rue - Duel au vieux port',
            'episode': 'Duel au vieux port',
            'series': 'Foot2Rue',
            'episode_number': 1,
            'season_number': 1,
            'timestamp': 1642761360,
            'upload_date': '20220121',
            'season': 'Season 1',
            'thumbnail': r're:^https?://.*\.jpg$',
            'duration': 1441,
        },
    }, {
        # geo-restricted livestream (workflow == 'token-akamai')
        'url': 'https://www.france.tv/france-4/direct.html',
        'info_dict': {
            'id': '9a6a7670-dde9-4264-adbc-55b89558594b',
            'ext': 'mp4',
            'title': r're:France 4 en direct .+',
            'live_status': 'is_live',
        },
        'skip': 'geo-restricted livestream',
    }, {
        # livestream (workflow == 'dai')
        'url': 'https://www.france.tv/france-2/direct.html',
        'info_dict': {
            'id': '006194ea-117d-4bcf-94a9-153d999c59ae',
            'ext': 'mp4',
            'title': r're:France 2 en direct .+',
            'live_status': 'is_live',
        },
        'params': {'skip_download': 'livestream'},
    }, {
        # france3
        'url': 'https://www.france.tv/france-3/des-chiffres-et-des-lettres/139063-emission-du-mardi-9-mai-2017.html',
        'only_matching': True,
    }, {
        # france4
        'url': 'https://www.france.tv/france-4/hero-corp/saison-1/134151-apres-le-calme.html',
        'only_matching': True,
    }, {
        # france5
        'url': 'https://www.france.tv/france-5/c-a-dire/saison-10/137013-c-a-dire.html',
        'only_matching': True,
    }, {
        # franceo
        'url': 'https://www.france.tv/france-o/archipels/132249-mon-ancetre-l-esclave.html',
        'only_matching': True,
    }, {
        'url': 'https://www.france.tv/documentaires/histoire/136517-argentine-les-500-bebes-voles-de-la-dictature.html',
        'only_matching': True,
    }, {
        'url': 'https://www.france.tv/jeux-et-divertissements/divertissements/133965-le-web-contre-attaque.html',
        'only_matching': True,
    }, {
        'url': 'https://mobile.france.tv/france-5/c-dans-l-air/137347-emission-du-vendredi-12-mai-2017.html',
        'only_matching': True,
    }, {
        'url': 'https://www.france.tv/142749-rouge-sang.html',
        'only_matching': True,
    }, {
        # france-3 live
        'url': 'https://www.france.tv/france-3/direct.html',
        'only_matching': True,
    }]

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

        webpage = self._download_webpage(url, display_id)

        video_id = self._search_regex(
            r'(?:data-main-video\s*=|videoId["\']?\s*[:=])\s*(["\'])(?P<id>(?:(?!\1).)+)\1',
            webpage, 'video id', default=None, group='id')

        if not video_id:
            video_id = self._html_search_regex(
                r'(?:href=|player\.setVideo\(\s*)"http://videos?\.francetv\.fr/video/([^@"]+@[^"]+)"',
                webpage, 'video ID')

        return self._make_url_result(video_id, url=url)


class FranceTVInfoIE(FranceTVBaseInfoExtractor):
    IE_NAME = 'francetvinfo.fr'
    _VALID_URL = r'https?://(?:www|mobile|france3-regions)\.francetvinfo\.fr/(?:[^/]+/)*(?P<id>[^/?#&.]+)'

    _TESTS = [{
        'url': 'https://www.francetvinfo.fr/replay-jt/france-3/soir-3/jt-grand-soir-3-jeudi-22-aout-2019_3561461.html',
        'info_dict': {
            'id': 'd12458ee-5062-48fe-bfdd-a30d6a01b793',
            'ext': 'mp4',
            'title': 'Soir 3',
            'upload_date': '20190822',
            'timestamp': 1566510730,
            'thumbnail': r're:^https?://.*\.jpe?g$',
            'duration': 1637,
            'subtitles': {
                'fr': 'mincount:2',
            },
        },
        'params': {
            'skip_download': True,
        },
        'add_ie': [FranceTVIE.ie_key()],
    }, {
        'note': 'Only an image exists in initial webpage instead of the video',
        'url': 'https://www.francetvinfo.fr/sante/maladie/coronavirus/covid-19-en-inde-une-situation-catastrophique-a-new-dehli_4381095.html',
        'info_dict': {
            'id': '7d204c9e-a2d3-11eb-9e4c-000d3a23d482',
            'ext': 'mp4',
            'title': 'Covid-19 : une situation catastrophique à New Dehli - Édition du mercredi 21 avril 2021',
            'thumbnail': r're:^https?://.*\.jpe?g$',
            'duration': 76,
            'timestamp': 1619028518,
            'upload_date': '20210421',
        },
        'params': {
            'skip_download': True,
        },
        'add_ie': [FranceTVIE.ie_key()],
    }, {
        'url': 'http://www.francetvinfo.fr/elections/europeennes/direct-europeennes-regardez-le-debat-entre-les-candidats-a-la-presidence-de-la-commission_600639.html',
        'only_matching': True,
    }, {
        'url': 'http://www.francetvinfo.fr/economie/entreprises/les-entreprises-familiales-le-secret-de-la-reussite_933271.html',
        'only_matching': True,
    }, {
        'url': 'http://france3-regions.francetvinfo.fr/bretagne/cotes-d-armor/thalassa-echappee-breizh-ce-venredi-dans-les-cotes-d-armor-954961.html',
        'only_matching': True,
    }, {
        # Dailymotion embed
        'url': 'http://www.francetvinfo.fr/politique/notre-dame-des-landes/video-sur-france-inter-cecile-duflot-denonce-le-regard-meprisant-de-patrick-cohen_1520091.html',
        'md5': 'ee7f1828f25a648addc90cb2687b1f12',
        'info_dict': {
            'id': 'x4iiko0',
            'ext': 'mp4',
            'title': 'NDDL, référendum, Brexit : Cécile Duflot répond à Patrick Cohen',
            'description': 'md5:fdcb582c370756293a65cdfbc6ecd90e',
            'timestamp': 1467011958,
            'uploader': 'France Inter',
            'uploader_id': 'x2q2ez',
            'upload_date': '20160627',
            'view_count': int,
            'tags': ['Politique', 'France Inter', '27 juin 2016', 'Linvité de 8h20', 'Cécile Duflot', 'Patrick Cohen'],
            'age_limit': 0,
            'duration': 640,
            'like_count': int,
            'thumbnail': r're:https://[^/?#]+/v/[^/?#]+/x1080',
        },
        'add_ie': ['Dailymotion'],
    }, {
        'url': 'http://france3-regions.francetvinfo.fr/limousin/emissions/jt-1213-limousin',
        'only_matching': True,
    }, {
        # "<figure id=" pattern (#28792)
        'url': 'https://www.francetvinfo.fr/culture/patrimoine/incendie-de-notre-dame-de-paris/notre-dame-de-paris-de-l-incendie-de-la-cathedrale-a-sa-reconstruction_4372291.html',
        'only_matching': True,
    }]

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

        webpage = self._download_webpage(url, display_id)

        dailymotion_urls = tuple(DailymotionIE._extract_embed_urls(url, webpage))
        if dailymotion_urls:
            return self.playlist_result([
                self.url_result(dailymotion_url, DailymotionIE.ie_key())
                for dailymotion_url in dailymotion_urls])

        video_id = self._search_regex(
            (r'player\.load[^;]+src:\s*["\']([^"\']+)',
             r'id-video=([^@]+@[^"]+)',
             r'<a[^>]+href="(?:https?:)?//videos\.francetv\.fr/video/([^@]+@[^"]+)"',
             r'(?:data-id|<figure[^<]+\bid)=["\']([\da-f]{8}-[\da-f]{4}-[\da-f]{4}-[\da-f]{4}-[\da-f]{12})'),
            webpage, 'video id')

        return self._make_url_result(video_id, url=url)
Commit	Line	Data
ede624d1	1	import re
9749ac7f	2	import urllib.parse
9749ac7f	3
5d8afe69	4	from .common import InfoExtractor
71f28097	5	from .dailymotion import DailymotionIE
9749ac7f	6	from ..networking import HEADRequest
1cc79574	7	from ..utils import (
3690c2f5	8	clean_html,
71f28097	9	determine_ext,
9749ac7f	10	filter_dict,
28fe35b4	11	format_field,
71f28097 PG	12	int_or_none,
71f28097 PG	13	join_nonempty,
28fe35b4	14	parse_iso8601,
9749ac7f	15	smuggle_url,
	16	unsmuggle_url,
	17	url_or_none,
5d8afe69	18	)
9749ac7f	19	from ..utils.traversal import traverse_obj
5d8afe69 PR	20
5d8afe69 PR	21
648d25d4	22	class FranceTVBaseInfoExtractor(InfoExtractor):
ede624d1	23	def _make_url_result(self, video_id, url=None):
	24	video_id = video_id.split('@')[0] # for compat with old @catalog IDs
	25	full_id = f'francetv:{video_id}'
9749ac7f	26	if url:
9749ac7f	27	full_id = smuggle_url(full_id, {'hostname': urllib.parse.urlparse(url).hostname})
ede624d1	28	return self.url_result(full_id, FranceTVIE, video_id)
99892e99 S	29
	30
	31	class FranceTVIE(InfoExtractor):
ede624d1	32	_VALID_URL = r'francetv:(?P<id>[^@#]+)'
9749ac7f	33	_GEO_COUNTRIES = ['FR']
9749ac7f	34	_GEO_BYPASS = False
99892e99 S	35
99892e99 S	36	_TESTS = [{
081708d6	37	# tokenized url is in dinfo['video']['token']
ede624d1	38	'url': 'francetv:ec217ecc-0733-48cf-ac06-af1347b849d1',
99892e99	39	'info_dict': {
ede624d1	40	'id': 'ec217ecc-0733-48cf-ac06-af1347b849d1',
99892e99 S	41	'ext': 'mp4',
99892e99 S	42	'title': '13h15, le dimanche... - Les mystères de Jésus',
99892e99	43	'timestamp': 1502623500,
ede624d1	44	'duration': 2580,
ede624d1	45	'thumbnail': r're:^https?://.*\.jpg$',
99892e99 S	46	'upload_date': '20170813',
99892e99 S	47	},
ede624d1	48	'params': {'skip_download': 'm3u8'},
081708d6	49	}, {
	50	# tokenized url is in dinfo['video']['token']['akamai']
	51	'url': 'francetv:c5bda21d-2c6f-4470-8849-3d8327adb2ba',
	52	'info_dict': {
	53	'id': 'c5bda21d-2c6f-4470-8849-3d8327adb2ba',
	54	'ext': 'mp4',
	55	'title': '13h15, le dimanche... - Les mystères de Jésus',
	56	'timestamp': 1514118300,
	57	'duration': 2880,
	58	'thumbnail': r're:^https?://.*\.jpg$',
	59	'upload_date': '20171224',
	60	},
	61	'params': {'skip_download': 'm3u8'},
99892e99 S	62	}, {
	63	'url': 'francetv:162311093',
	64	'only_matching': True,
	65	}, {
	66	'url': 'francetv:NI_1004933@Zouzous',
	67	'only_matching': True,
	68	}, {
	69	'url': 'francetv:NI_983319@Info-web',
	70	'only_matching': True,
	71	}, {
	72	'url': 'francetv:NI_983319',
	73	'only_matching': True,
	74	}, {
	75	'url': 'francetv:NI_657393@Regions',
	76	'only_matching': True,
760f8121 S	77	}, {
	78	# france-3 live
	79	'url': 'francetv:SIM_France3',
	80	'only_matching': True,
99892e99 S	81	}]
99892e99 S	82
ede624d1	83	def _extract_video(self, video_id, hostname=None):
760f8121	84	is_live = None
8bdd16b4	85	videos = []
3690c2f5	86	drm_formats = False
28fe35b4 F	87	title = None
28fe35b4 F	88	subtitle = None
71f28097 PG	89	episode_number = None
71f28097 PG	90	season_number = None
28fe35b4 F	91	image = None
	92	duration = None
	93	timestamp = None
	94	spritesheets = None
	95
ede624d1	96	# desktop+chrome returns dash; mobile+safari returns hls
ede624d1	97	for device_type, browser in [('desktop', 'chrome'), ('mobile', 'safari')]:
28fe35b4	98	dinfo = self._download_json(
ede624d1	99	f'https://k7.ftven.fr/videos/{video_id}', video_id,
ede624d1	100	f'Downloading {device_type} {browser} video JSON', query=filter_dict({
28fe35b4	101	'device_type': device_type,
ede624d1	102	'browser': browser,
9749ac7f	103	'domain': hostname,
3690c2f5	104	}), fatal=False, expected_status=422) # 422 json gives detailed error code/message
28fe35b4 F	105
28fe35b4 F	106	if not dinfo:
8bdd16b4	107	continue
8bdd16b4	108
3690c2f5	109	if video := traverse_obj(dinfo, ('video', {dict})):
28fe35b4 F	110	videos.append(video)
	111	if duration is None:
	112	duration = video.get('duration')
	113	if is_live is None:
	114	is_live = video.get('is_live')
	115	if spritesheets is None:
	116	spritesheets = video.get('spritesheets')
3690c2f5	117	elif code := traverse_obj(dinfo, ('code', {int})):
	118	if code == 2009:
	119	self.raise_geo_restricted(countries=self._GEO_COUNTRIES)
	120	elif code in (2015, 2017):
	121	# 2015: L'accès à cette vidéo est impossible. (DRM-only)
	122	# 2017: Cette vidéo n'est pas disponible depuis le site web mobile (b/c DRM)
	123	drm_formats = True
	124	continue
	125	self.report_warning(
	126	f'{self.IE_NAME} said: {code} "{clean_html(dinfo.get("message"))}"')
	127	continue
28fe35b4	128
3690c2f5	129	if meta := traverse_obj(dinfo, ('meta', {dict})):
28fe35b4 F	130	if title is None:
28fe35b4 F	131	title = meta.get('title')
71f28097 PG	132	# meta['pre_title'] contains season and episode number for series in format "S<ID> E<ID>"
	133	season_number, episode_number = self._search_regex(
	134	r'S(\d+)\s*E(\d+)', meta.get('pre_title'), 'episode info', group=(1, 2), default=(None, None))
28fe35b4 F	135	if subtitle is None:
	136	subtitle = meta.get('additional_title')
	137	if image is None:
	138	image = meta.get('image_url')
	139	if timestamp is None:
	140	timestamp = parse_iso8601(meta.get('broadcasted_at'))
8bdd16b4	141
3690c2f5	142	if not videos and drm_formats:
	143	self.report_drm(video_id)
	144
9749ac7f	145	formats, subtitles, video_url = [], {}, None
	146	for video in traverse_obj(videos, lambda _, v: url_or_none(v['url'])):
	147	video_url = video['url']
8bdd16b4	148	format_id = video.get('format')
28fe35b4	149
081708d6	150	if token_url := traverse_obj(video, ('token', (None, 'akamai'), {url_or_none}, any)):
9749ac7f	151	tokenized_url = traverse_obj(self._download_json(
	152	token_url, video_id, f'Downloading signed {format_id} manifest URL',
	153	fatal=False, query={
	154	'format': 'json',
	155	'url': video_url,
	156	}), ('url', {url_or_none}))
	157	if tokenized_url:
	158	video_url = tokenized_url
28fe35b4	159
bc03228a S	160	ext = determine_ext(video_url)
bc03228a S	161	if ext == 'f4m':
8faa338f	162	formats.extend(self._extract_f4m_formats(
ede624d1	163	video_url, video_id, f4m_id=format_id or ext, fatal=False))
bc03228a	164	elif ext == 'm3u8':
ede624d1	165	format_id = format_id or 'hls'
28fe35b4	166	fmts, subs = self._extract_m3u8_formats_and_subtitles(
ede624d1	167	video_url, video_id, 'mp4', m3u8_id=format_id, fatal=False)
	168	for f in traverse_obj(fmts, lambda _, v: v['vcodec'] == 'none' and v.get('tbr') is None):
	169	if mobj := re.match(rf'{format_id}-[Aa]udio-\w+-(?P<bitrate>\d+)', f['format_id']):
	170	f.update({
	171	'tbr': int_or_none(mobj.group('bitrate')),
	172	'acodec': 'mp4a',
	173	})
28fe35b4 F	174	formats.extend(fmts)
28fe35b4 F	175	self._merge_subtitles(subs, target=subtitles)
8bdd16b4	176	elif ext == 'mpd':
28fe35b4	177	fmts, subs = self._extract_mpd_formats_and_subtitles(
ede624d1	178	video_url, video_id, mpd_id=format_id or 'dash', fatal=False)
28fe35b4 F	179	formats.extend(fmts)
28fe35b4 F	180	self._merge_subtitles(subs, target=subtitles)
64892c0b S	181	elif video_url.startswith('rtmp'):
	182	formats.append({
	183	'url': video_url,
ede624d1	184	'format_id': join_nonempty('rtmp', format_id),
64892c0b	185	'ext': 'flv',
64892c0b S	186	})
64892c0b S	187	else:
3c20208e S	188	if self._is_valid_url(video_url, video_id, format_id):
	189	formats.append({
	190	'url': video_url,
	191	'format_id': format_id,
	192	})
8bdd16b4	193
28fe35b4 F	194	# XXX: what is video['captions']?
28fe35b4 F	195
9749ac7f	196	if not formats and video_url:
	197	urlh = self._request_webpage(
	198	HEADRequest(video_url), video_id, 'Checking for geo-restriction',
	199	fatal=False, expected_status=403)
	200	if urlh and urlh.headers.get('x-errortype') == 'geo':
	201	self.raise_geo_restricted(countries=self._GEO_COUNTRIES, metadata_available=True)
	202
28fe35b4 F	203	for f in formats:
	204	if f.get('acodec') != 'none' and f.get('language') in ('qtz', 'qad'):
	205	f['language_preference'] = -10
add96eb9	206	f['format_note'] = 'audio description{}'.format(format_field(f, 'format_note', ', %s'))
28fe35b4 F	207
	208	if spritesheets:
	209	formats.append({
	210	'format_id': 'spritesheets',
	211	'format_note': 'storyboard',
	212	'acodec': 'none',
	213	'vcodec': 'none',
	214	'ext': 'mhtml',
	215	'protocol': 'mhtml',
9222c381	216	'url': 'about:invalid',
28fe35b4	217	'fragments': [{
b3edc806	218	'url': sheet,
28fe35b4	219	# XXX: not entirely accurate; each spritesheet seems to be
add96eb9	220	# a 10x10 grid of thumbnails corresponding to approximately
28fe35b4 F	221	# 2 seconds of the video; the last spritesheet may be shorter
28fe35b4 F	222	'duration': 200,
add96eb9	223	} for sheet in traverse_obj(spritesheets, (..., {url_or_none}))],
28fe35b4 F	224	})
28fe35b4 F	225
7e8d73c1 JMF	226	return {
7e8d73c1 JMF	227	'id': video_id,
71f28097	228	'title': join_nonempty(title, subtitle, delim=' - ').strip(),
28fe35b4 F	229	'thumbnail': image,
	230	'duration': duration,
	231	'timestamp': timestamp,
760f8121	232	'is_live': is_live,
7e8d73c1	233	'formats': formats,
5dadae07	234	'subtitles': subtitles,
71f28097 PG	235	'episode': subtitle if episode_number else None,
	236	'series': title if episode_number else None,
	237	'episode_number': int_or_none(episode_number),
	238	'season_number': int_or_none(season_number),
ede624d1	239	'_format_sort_fields': ('res', 'tbr', 'proto'), # prioritize m3u8 over dash
7e8d73c1	240	}
648d25d4	241
99892e99	242	def _real_extract(self, url):
9749ac7f	243	url, smuggled_data = unsmuggle_url(url, {})
ede624d1	244	video_id = self._match_id(url)
ede624d1	245	hostname = smuggled_data.get('hostname') or 'www.france.tv'
99892e99	246
ede624d1	247	return self._extract_video(video_id, hostname=hostname)
99892e99	248
648d25d4	249
99892e99	250	class FranceTVSiteIE(FranceTVBaseInfoExtractor):
4489d418	251	_VALID_URL = r'https?://(?:(?:www\.)?france\.tv\|mobile\.france\.tv)/(?:[^/]+/)*(?P<id>[^/]+)\.html'
5d8afe69	252
6d1ded75 S	253	_TESTS = [{
	254	'url': 'https://www.france.tv/france-2/13h15-le-dimanche/140921-les-mysteres-de-jesus.html',
	255	'info_dict': {
081708d6	256	'id': 'c5bda21d-2c6f-4470-8849-3d8327adb2ba',
6d1ded75 S	257	'ext': 'mp4',
6d1ded75 S	258	'title': '13h15, le dimanche... - Les mystères de Jésus',
081708d6	259	'timestamp': 1514118300,
081708d6	260	'duration': 2880,
71f28097	261	'thumbnail': r're:^https?://.*\.jpg$',
081708d6	262	'upload_date': '20171224',
6d1ded75 S	263	},
6d1ded75 S	264	'params': {
6d1ded75 S	265	'skip_download': True,
6d1ded75 S	266	},
99892e99	267	'add_ie': [FranceTVIE.ie_key()],
71f28097	268	}, {
ede624d1	269	# geo-restricted
71f28097 PG	270	'url': 'https://www.france.tv/enfants/six-huit-ans/foot2rue/saison-1/3066387-duel-au-vieux-port.html',
	271	'info_dict': {
	272	'id': 'a9050959-eedd-4b4a-9b0d-de6eeaa73e44',
	273	'ext': 'mp4',
	274	'title': 'Foot2Rue - Duel au vieux port',
	275	'episode': 'Duel au vieux port',
	276	'series': 'Foot2Rue',
	277	'episode_number': 1,
	278	'season_number': 1,
	279	'timestamp': 1642761360,
	280	'upload_date': '20220121',
	281	'season': 'Season 1',
	282	'thumbnail': r're:^https?://.*\.jpg$',
	283	'duration': 1441,
	284	},
e4fbe5f8	285	}, {
	286	# geo-restricted livestream (workflow == 'token-akamai')
	287	'url': 'https://www.france.tv/france-4/direct.html',
	288	'info_dict': {
	289	'id': '9a6a7670-dde9-4264-adbc-55b89558594b',
	290	'ext': 'mp4',
	291	'title': r're:France 4 en direct .+',
	292	'live_status': 'is_live',
	293	},
	294	'skip': 'geo-restricted livestream',
	295	}, {
	296	# livestream (workflow == 'dai')
	297	'url': 'https://www.france.tv/france-2/direct.html',
	298	'info_dict': {
	299	'id': '006194ea-117d-4bcf-94a9-153d999c59ae',
	300	'ext': 'mp4',
	301	'title': r're:France 2 en direct .+',
	302	'live_status': 'is_live',
	303	},
	304	'params': {'skip_download': 'livestream'},
6d1ded75 S	305	}, {
	306	# france3
	307	'url': 'https://www.france.tv/france-3/des-chiffres-et-des-lettres/139063-emission-du-mardi-9-mai-2017.html',
	308	'only_matching': True,
	309	}, {
	310	# france4
	311	'url': 'https://www.france.tv/france-4/hero-corp/saison-1/134151-apres-le-calme.html',
	312	'only_matching': True,
	313	}, {
	314	# france5
	315	'url': 'https://www.france.tv/france-5/c-a-dire/saison-10/137013-c-a-dire.html',
	316	'only_matching': True,
	317	}, {
	318	# franceo
	319	'url': 'https://www.france.tv/france-o/archipels/132249-mon-ancetre-l-esclave.html',
	320	'only_matching': True,
6d1ded75 S	321	}, {
	322	'url': 'https://www.france.tv/documentaires/histoire/136517-argentine-les-500-bebes-voles-de-la-dictature.html',
	323	'only_matching': True,
	324	}, {
	325	'url': 'https://www.france.tv/jeux-et-divertissements/divertissements/133965-le-web-contre-attaque.html',
	326	'only_matching': True,
12f01118 S	327	}, {
	328	'url': 'https://mobile.france.tv/france-5/c-dans-l-air/137347-emission-du-vendredi-12-mai-2017.html',
	329	'only_matching': True,
4489d418 S	330	}, {
	331	'url': 'https://www.france.tv/142749-rouge-sang.html',
	332	'only_matching': True,
49702e36 S	333	}, {
	334	# france-3 live
	335	'url': 'https://www.france.tv/france-3/direct.html',
	336	'only_matching': True,
6d1ded75	337	}]
5d8afe69 PR	338
5d8afe69 PR	339	def _real_extract(self, url):
0a192fbe S	340	display_id = self._match_id(url)
	341
	342	webpage = self._download_webpage(url, display_id)
	343
6d1ded75	344	video_id = self._search_regex(
9d74ea6d	345	r'(?:data-main-video\s=\|videoId["\']?\s[:=])\s*(["\'])(?P<id>(?:(?!\1).)+)\1',
6d1ded75 S	346	webpage, 'video id', default=None, group='id')
6d1ded75 S	347
0a192fbe	348	if not video_id:
ede624d1	349	video_id = self._html_search_regex(
	350	r'(?:href=\|player\.setVideo\(\s*)"http://videos?\.francetv\.fr/video/([^@"]+@[^"]+)"',
	351	webpage, 'video ID')
99892e99	352
ede624d1	353	return self._make_url_result(video_id, url=url)
6d1ded75 S	354
6d1ded75 S	355
6d1ded75	356	class FranceTVInfoIE(FranceTVBaseInfoExtractor):
3c1b4669	357	IE_NAME = 'francetvinfo.fr'
99892e99	358	_VALID_URL = r'https?://(?:www\|mobile\|france3-regions)\.francetvinfo\.fr/(?:[^/]+/)*(?P<id>[^/?#&.]+)'
5d8afe69	359
5c30b268	360	_TESTS = [{
13c30d1d	361	'url': 'https://www.francetvinfo.fr/replay-jt/france-3/soir-3/jt-grand-soir-3-jeudi-22-aout-2019_3561461.html',
3c1b4669	362	'info_dict': {
13c30d1d	363	'id': 'd12458ee-5062-48fe-bfdd-a30d6a01b793',
3c20208e	364	'ext': 'mp4',
3c1b4669	365	'title': 'Soir 3',
13c30d1d	366	'upload_date': '20190822',
ede624d1	367	'timestamp': 1566510730,
	368	'thumbnail': r're:^https?://.*\.jpe?g$',
	369	'duration': 1637,
c137cc0d S	370	'subtitles': {
	371	'fr': 'mincount:2',
	372	},
648d25d4	373	},
3c20208e	374	'params': {
3c20208e S	375	'skip_download': True,
3c20208e S	376	},
99892e99	377	'add_ie': [FranceTVIE.ie_key()],
bbed5763	378	}, {
	379	'note': 'Only an image exists in initial webpage instead of the video',
	380	'url': 'https://www.francetvinfo.fr/sante/maladie/coronavirus/covid-19-en-inde-une-situation-catastrophique-a-new-dehli_4381095.html',
	381	'info_dict': {
	382	'id': '7d204c9e-a2d3-11eb-9e4c-000d3a23d482',
	383	'ext': 'mp4',
ede624d1	384	'title': 'Covid-19 : une situation catastrophique à New Dehli - Édition du mercredi 21 avril 2021',
ede624d1	385	'thumbnail': r're:^https?://.*\.jpe?g$',
bbed5763	386	'duration': 76,
	387	'timestamp': 1619028518,
	388	'upload_date': '20210421',
	389	},
	390	'params': {
	391	'skip_download': True,
	392	},
	393	'add_ie': [FranceTVIE.ie_key()],
5c30b268 PH	394	}, {
5c30b268 PH	395	'url': 'http://www.francetvinfo.fr/elections/europeennes/direct-europeennes-regardez-le-debat-entre-les-candidats-a-la-presidence-de-la-commission_600639.html',
99892e99	396	'only_matching': True,
6f96e308 YCH	397	}, {
6f96e308 YCH	398	'url': 'http://www.francetvinfo.fr/economie/entreprises/les-entreprises-familiales-le-secret-de-la-reussite_933271.html',
99892e99	399	'only_matching': True,
db264e3c S	400	}, {
db264e3c S	401	'url': 'http://france3-regions.francetvinfo.fr/bretagne/cotes-d-armor/thalassa-echappee-breizh-ce-venredi-dans-les-cotes-d-armor-954961.html',
99892e99	402	'only_matching': True,
ad213a1d YCH	403	}, {
	404	# Dailymotion embed
	405	'url': 'http://www.francetvinfo.fr/politique/notre-dame-des-landes/video-sur-france-inter-cecile-duflot-denonce-le-regard-meprisant-de-patrick-cohen_1520091.html',
	406	'md5': 'ee7f1828f25a648addc90cb2687b1f12',
	407	'info_dict': {
	408	'id': 'x4iiko0',
	409	'ext': 'mp4',
	410	'title': 'NDDL, référendum, Brexit : Cécile Duflot répond à Patrick Cohen',
ede624d1	411	'description': 'md5:fdcb582c370756293a65cdfbc6ecd90e',
ad213a1d	412	'timestamp': 1467011958,
ad213a1d YCH	413	'uploader': 'France Inter',
ad213a1d YCH	414	'uploader_id': 'x2q2ez',
ede624d1	415	'upload_date': '20160627',
	416	'view_count': int,
	417	'tags': ['Politique', 'France Inter', '27 juin 2016', 'Linvité de 8h20', 'Cécile Duflot', 'Patrick Cohen'],
	418	'age_limit': 0,
	419	'duration': 640,
	420	'like_count': int,
	421	'thumbnail': r're:https://[^/?#]+/v/[^/?#]+/x1080',
ad213a1d YCH	422	},
ad213a1d YCH	423	'add_ie': ['Dailymotion'],
30b25d38 S	424	}, {
	425	'url': 'http://france3-regions.francetvinfo.fr/limousin/emissions/jt-1213-limousin',
	426	'only_matching': True,
41d1cca3	427	}, {
	428	# "<figure id=" pattern (#28792)
	429	'url': 'https://www.francetvinfo.fr/culture/patrimoine/incendie-de-notre-dame-de-paris/notre-dame-de-paris-de-l-incendie-de-la-cathedrale-a-sa-reconstruction_4372291.html',
	430	'only_matching': True,
5c30b268	431	}]
648d25d4 JMF	432
648d25d4 JMF	433	def _real_extract(self, url):
99892e99 S	434	display_id = self._match_id(url)
	435
	436	webpage = self._download_webpage(url, display_id)
6f96e308	437
43aebb7d	438	dailymotion_urls = tuple(DailymotionIE._extract_embed_urls(url, webpage))
ad213a1d YCH	439	if dailymotion_urls:
	440	return self.playlist_result([
	441	self.url_result(dailymotion_url, DailymotionIE.ie_key())
	442	for dailymotion_url in dailymotion_urls])
6f96e308	443
876fed6b	444	video_id = self._search_regex(
	445	(r'player\.load[^;]+src:\s*["\']([^"\']+)',
	446	r'id-video=([^@]+@[^"]+)',
13c30d1d	447	r'<a[^>]+href="(?:https?:)?//videos\.francetv\.fr/video/([^@]+@[^"]+)"',
41d1cca3	448	r'(?:data-id\|<figure[^<]+\bid)=["\']([\da-f]{8}-[\da-f]{4}-[\da-f]{4}-[\da-f]{4}-[\da-f]{12})'),
876fed6b	449	webpage, 'video id')
99892e99	450
9749ac7f	451	return self._make_url_result(video_id, url=url)