[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 (
    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 = [{
        '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'},
    }, {
        '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 = []
        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)

            if not dinfo:
                continue

            video = traverse_obj(dinfo, ('video', {dict}))
            if video:
                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')

            meta = traverse_obj(dinfo, ('meta', {dict}))
            if meta:
                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'))

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