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

from .common import InfoExtractor
from ..utils import (
    ExtractorError,
    float_or_none,
    int_or_none,
    join_nonempty,
    parse_iso8601,
    qualities,
    try_get,
)


class SRGSSRIE(InfoExtractor):
    _VALID_URL = r'''(?x)
                    (?:
                        https?://tp\.srgssr\.ch/p(?:/[^/]+)+\?urn=urn|
                        srgssr
                    ):
                    (?P<bu>
                        srf|rts|rsi|rtr|swi
                    ):(?:[^:]+:)?
                    (?P<type>
                        video|audio
                    ):
                    (?P<id>
                        [0-9a-f\-]{36}|\d+
                    )
                    '''
    _GEO_BYPASS = False
    _GEO_COUNTRIES = ['CH']

    _ERRORS = {
        'AGERATING12': 'To protect children under the age of 12, this video is only available between 8 p.m. and 6 a.m.',
        'AGERATING18': 'To protect children under the age of 18, this video is only available between 11 p.m. and 5 a.m.',
        # 'ENDDATE': 'For legal reasons, this video was only available for a specified period of time.',
        'GEOBLOCK': 'For legal reasons, this video is only available in Switzerland.',
        'LEGAL': 'The video cannot be transmitted for legal reasons.',
        'STARTDATE': 'This video is not yet available. Please try again later.',
    }
    _DEFAULT_LANGUAGE_CODES = {
        'srf': 'de',
        'rts': 'fr',
        'rsi': 'it',
        'rtr': 'rm',
        'swi': 'en',
    }

    def _get_tokenized_src(self, url, video_id, format_id):
        token = self._download_json(
            'http://tp.srgssr.ch/akahd/token?acl=*',
            video_id, f'Downloading {format_id} token', fatal=False) or {}
        auth_params = try_get(token, lambda x: x['token']['authparams'])
        if auth_params:
            url += ('?' if '?' not in url else '&') + auth_params
        return url

    def _get_media_data(self, bu, media_type, media_id):
        query = {'onlyChapters': True} if media_type == 'video' else {}
        full_media_data = self._download_json(
            f'https://il.srgssr.ch/integrationlayer/2.0/{bu}/mediaComposition/{media_type}/{media_id}.json',
            media_id, query=query)['chapterList']
        try:
            media_data = next(
                x for x in full_media_data if x.get('id') == media_id)
        except StopIteration:
            raise ExtractorError('No media information found')

        block_reason = media_data.get('blockReason')
        if block_reason and block_reason in self._ERRORS:
            message = self._ERRORS[block_reason]
            if block_reason == 'GEOBLOCK':
                self.raise_geo_restricted(
                    msg=message, countries=self._GEO_COUNTRIES)
            raise ExtractorError(
                f'{self.IE_NAME} said: {message}', expected=True)

        return media_data

    def _real_extract(self, url):
        bu, media_type, media_id = self._match_valid_url(url).groups()
        media_data = self._get_media_data(bu, media_type, media_id)
        title = media_data['title']

        formats = []
        subtitles = {}
        q = qualities(['SD', 'HD'])
        for source in (media_data.get('resourceList') or []):
            format_url = source.get('url')
            if not format_url:
                continue
            protocol = source.get('protocol')
            quality = source.get('quality')
            format_id = join_nonempty(protocol, source.get('encoding'), quality)

            if protocol in ('HDS', 'HLS'):
                if source.get('tokenType') == 'AKAMAI':
                    format_url = self._get_tokenized_src(
                        format_url, media_id, format_id)
                    fmts, subs = self._extract_akamai_formats_and_subtitles(
                        format_url, media_id)
                    formats.extend(fmts)
                    subtitles = self._merge_subtitles(subtitles, subs)
                elif protocol == 'HLS':
                    m3u8_fmts, m3u8_subs = self._extract_m3u8_formats_and_subtitles(
                        format_url, media_id, 'mp4', 'm3u8_native',
                        m3u8_id=format_id, fatal=False)
                    formats.extend(m3u8_fmts)
                    subtitles = self._merge_subtitles(subtitles, m3u8_subs)
            elif protocol in ('HTTP', 'HTTPS'):
                formats.append({
                    'format_id': format_id,
                    'url': format_url,
                    'quality': q(quality),
                })

        # This is needed because for audio medias the podcast url is usually
        # always included, even if is only an audio segment and not the
        # whole episode.
        if int_or_none(media_data.get('position')) == 0:
            for p in ('S', 'H'):
                podcast_url = media_data.get(f'podcast{p}dUrl')
                if not podcast_url:
                    continue
                quality = p + 'D'
                formats.append({
                    'format_id': 'PODCAST-' + quality,
                    'url': podcast_url,
                    'quality': q(quality),
                })

        if media_type == 'video':
            for sub in (media_data.get('subtitleList') or []):
                sub_url = sub.get('url')
                if not sub_url:
                    continue
                lang = sub.get('locale') or self._DEFAULT_LANGUAGE_CODES[bu]
                subtitles.setdefault(lang, []).append({
                    'url': sub_url,
                })

        return {
            'id': media_id,
            'title': title,
            'description': media_data.get('description'),
            'timestamp': parse_iso8601(media_data.get('date')),
            'thumbnail': media_data.get('imageUrl'),
            'duration': float_or_none(media_data.get('duration'), 1000),
            'subtitles': subtitles,
            'formats': formats,
        }


class SRGSSRPlayIE(InfoExtractor):
    IE_DESC = 'srf.ch, rts.ch, rsi.ch, rtr.ch and swissinfo.ch play sites'
    _VALID_URL = r'''(?x)
                    https?://
                        (?:(?:www|play)\.)?
                        (?P<bu>srf|rts|rsi|rtr|swissinfo)\.ch/play/(?:tv|radio)/
                        (?:
                            [^/]+/(?P<type>video|audio)/[^?]+|
                            popup(?P<type_2>video|audio)player
                        )
                        \?.*?\b(?:id=|urn=urn:[^:]+:video:)(?P<id>[0-9a-f\-]{36}|\d+)
                    '''

    _TESTS = [{
        'url': 'http://www.srf.ch/play/tv/10vor10/video/snowden-beantragt-asyl-in-russland?id=28e1a57d-5b76-4399-8ab3-9097f071e6c5',
        'md5': '6db2226ba97f62ad42ce09783680046c',
        'info_dict': {
            'id': '28e1a57d-5b76-4399-8ab3-9097f071e6c5',
            'ext': 'mp4',
            'upload_date': '20130701',
            'title': 'Snowden beantragt Asyl in Russland',
            'timestamp': 1372708215,
            'duration': 113.827,
            'thumbnail': r're:^https?://.*1383719781\.png$',
        },
        'expected_warnings': ['Unable to download f4m manifest'],
    }, {
        'url': 'http://www.rtr.ch/play/radio/actualitad/audio/saira-tujetsch-tuttina-cuntinuar-cun-sedrun-muster-turissem?id=63cb0778-27f8-49af-9284-8c7a8c6d15fc',
        'info_dict': {
            'id': '63cb0778-27f8-49af-9284-8c7a8c6d15fc',
            'ext': 'mp3',
            'upload_date': '20151013',
            'title': 'Saira: Tujetsch - tuttina cuntinuar cun Sedrun Mustér Turissem',
            'timestamp': 1444709160,
            'duration': 336.816,
        },
        'params': {
            # rtmp download
            'skip_download': True,
        },
    }, {
        'url': 'http://www.rts.ch/play/tv/-/video/le-19h30?id=6348260',
        'md5': '67a2a9ae4e8e62a68d0e9820cc9782df',
        'info_dict': {
            'id': '6348260',
            'display_id': '6348260',
            'ext': 'mp4',
            'duration': 1796.76,
            'title': 'Le 19h30',
            'upload_date': '20141201',
            'timestamp': 1417458600,
            'thumbnail': r're:^https?://.*\.image',
        },
        'params': {
            # m3u8 download
            'skip_download': True,
        },
    }, {
        'url': 'http://play.swissinfo.ch/play/tv/business/video/why-people-were-against-tax-reforms?id=42960270',
        'info_dict': {
            'id': '42960270',
            'ext': 'mp4',
            'title': 'Why people were against tax reforms',
            'description': 'md5:7ac442c558e9630e947427469c4b824d',
            'duration': 94.0,
            'upload_date': '20170215',
            'timestamp': 1487173560,
            'thumbnail': r're:https?://www\.swissinfo\.ch/srgscalableimage/42961964',
            'subtitles': 'count:9',
        },
        'params': {
            'skip_download': True,
        },
    }, {
        'url': 'https://www.srf.ch/play/tv/popupvideoplayer?id=c4dba0ca-e75b-43b2-a34f-f708a4932e01',
        'only_matching': True,
    }, {
        'url': 'https://www.srf.ch/play/tv/10vor10/video/snowden-beantragt-asyl-in-russland?urn=urn:srf:video:28e1a57d-5b76-4399-8ab3-9097f071e6c5',
        'only_matching': True,
    }, {
        'url': 'https://www.rts.ch/play/tv/19h30/video/le-19h30?urn=urn:rts:video:6348260',
        'only_matching': True,
    }, {
        # audio segment, has podcastSdUrl of the full episode
        'url': 'https://www.srf.ch/play/radio/popupaudioplayer?id=50b20dc8-f05b-4972-bf03-e438ff2833eb',
        'only_matching': True,
    }]

    def _real_extract(self, url):
        mobj = self._match_valid_url(url)
        bu = mobj.group('bu')
        media_type = mobj.group('type') or mobj.group('type_2')
        media_id = mobj.group('id')
        return self.url_result(f'srgssr:{bu[:3]}:{media_type}:{media_id}', 'SRGSSR')
Commit	Line	Data
1ef15636	1	from .common import InfoExtractor
	2	from ..utils import (
	3	ExtractorError,
ec5e77c5	4	float_or_none,
ec5e77c5	5	int_or_none,
34921b43	6	join_nonempty,
1ef15636	7	parse_iso8601,
1ef15636	8	qualities,
ec5e77c5	9	try_get,
1ef15636	10	)
	11
	12
	13	class SRGSSRIE(InfoExtractor):
ec5e77c5	14	_VALID_URL = r'''(?x)
	15	(?:
	16	https?://tp\.srgssr\.ch/p(?:/[^/]+)+\?urn=urn\|
	17	srgssr
	18	):
	19	(?P<bu>
	20	srf\|rts\|rsi\|rtr\|swi
	21	):(?:[^:]+:)?
	22	(?P<type>
	23	video\|audio
	24	):
	25	(?P<id>
	26	[0-9a-f\-]{36}\|\d+
	27	)
	28	'''
4248dad9 S	29	_GEO_BYPASS = False
4248dad9 S	30	_GEO_COUNTRIES = ['CH']
1ef15636	31
	32	_ERRORS = {
	33	'AGERATING12': 'To protect children under the age of 12, this video is only available between 8 p.m. and 6 a.m.',
	34	'AGERATING18': 'To protect children under the age of 18, this video is only available between 11 p.m. and 5 a.m.',
7db2897d	35	# 'ENDDATE': 'For legal reasons, this video was only available for a specified period of time.',
1ef15636	36	'GEOBLOCK': 'For legal reasons, this video is only available in Switzerland.',
	37	'LEGAL': 'The video cannot be transmitted for legal reasons.',
	38	'STARTDATE': 'This video is not yet available. Please try again later.',
	39	}
ec5e77c5	40	_DEFAULT_LANGUAGE_CODES = {
	41	'srf': 'de',
	42	'rts': 'fr',
	43	'rsi': 'it',
	44	'rtr': 'rm',
	45	'swi': 'en',
	46	}
1ef15636	47
68601ef3	48	def _get_tokenized_src(self, url, video_id, format_id):
68601ef3	49	token = self._download_json(
ec5e77c5	50	'http://tp.srgssr.ch/akahd/token?acl=*',
add96eb9	51	video_id, f'Downloading {format_id} token', fatal=False) or {}
ec5e77c5	52	auth_params = try_get(token, lambda x: x['token']['authparams'])
68601ef3	53	if auth_params:
ec5e77c5	54	url += ('?' if '?' not in url else '&') + auth_params
68601ef3 RA	55	return url
68601ef3 RA	56
ec5e77c5	57	def _get_media_data(self, bu, media_type, media_id):
	58	query = {'onlyChapters': True} if media_type == 'video' else {}
	59	full_media_data = self._download_json(
add96eb9	60	f'https://il.srgssr.ch/integrationlayer/2.0/{bu}/mediaComposition/{media_type}/{media_id}.json',
ec5e77c5	61	media_id, query=query)['chapterList']
	62	try:
	63	media_data = next(
	64	x for x in full_media_data if x.get('id') == media_id)
	65	except StopIteration:
	66	raise ExtractorError('No media information found')
	67
	68	block_reason = media_data.get('blockReason')
	69	if block_reason and block_reason in self._ERRORS:
	70	message = self._ERRORS[block_reason]
	71	if block_reason == 'GEOBLOCK':
4248dad9 S	72	self.raise_geo_restricted(
4248dad9 S	73	msg=message, countries=self._GEO_COUNTRIES)
71631862	74	raise ExtractorError(
add96eb9	75	f'{self.IE_NAME} said: {message}', expected=True)
7db2897d	76
	77	return media_data
	78
	79	def _real_extract(self, url):
5ad28e7f	80	bu, media_type, media_id = self._match_valid_url(url).groups()
ec5e77c5	81	media_data = self._get_media_data(bu, media_type, media_id)
ec5e77c5	82	title = media_data['title']
7db2897d	83
1ef15636	84	formats = []
efe9dba5	85	subtitles = {}
ec5e77c5	86	q = qualities(['SD', 'HD'])
	87	for source in (media_data.get('resourceList') or []):
	88	format_url = source.get('url')
	89	if not format_url:
	90	continue
	91	protocol = source.get('protocol')
	92	quality = source.get('quality')
34921b43	93	format_id = join_nonempty(protocol, source.get('encoding'), quality)
ec5e77c5	94
	95	if protocol in ('HDS', 'HLS'):
	96	if source.get('tokenType') == 'AKAMAI':
	97	format_url = self._get_tokenized_src(
	98	format_url, media_id, format_id)
efe9dba5 F	99	fmts, subs = self._extract_akamai_formats_and_subtitles(
	100	format_url, media_id)
	101	formats.extend(fmts)
	102	subtitles = self._merge_subtitles(subtitles, subs)
ec5e77c5	103	elif protocol == 'HLS':
efe9dba5	104	m3u8_fmts, m3u8_subs = self._extract_m3u8_formats_and_subtitles(
ec5e77c5	105	format_url, media_id, 'mp4', 'm3u8_native',
efe9dba5 F	106	m3u8_id=format_id, fatal=False)
	107	formats.extend(m3u8_fmts)
	108	subtitles = self._merge_subtitles(subtitles, m3u8_subs)
ec5e77c5	109	elif protocol in ('HTTP', 'HTTPS'):
	110	formats.append({
	111	'format_id': format_id,
	112	'url': format_url,
	113	'quality': q(quality),
	114	})
	115
	116	# This is needed because for audio medias the podcast url is usually
	117	# always included, even if is only an audio segment and not the
	118	# whole episode.
	119	if int_or_none(media_data.get('position')) == 0:
	120	for p in ('S', 'H'):
add96eb9	121	podcast_url = media_data.get(f'podcast{p}dUrl')
ec5e77c5	122	if not podcast_url:
	123	continue
	124	quality = p + 'D'
	125	formats.append({
	126	'format_id': 'PODCAST-' + quality,
	127	'url': podcast_url,
	128	'quality': q(quality),
	129	})
1ef15636	130
ec5e77c5	131	if media_type == 'video':
	132	for sub in (media_data.get('subtitleList') or []):
	133	sub_url = sub.get('url')
	134	if not sub_url:
	135	continue
	136	lang = sub.get('locale') or self._DEFAULT_LANGUAGE_CODES[bu]
	137	subtitles.setdefault(lang, []).append({
	138	'url': sub_url,
	139	})
	140
1ef15636	141	return {
	142	'id': media_id,
	143	'title': title,
ec5e77c5	144	'description': media_data.get('description'),
	145	'timestamp': parse_iso8601(media_data.get('date')),
	146	'thumbnail': media_data.get('imageUrl'),
	147	'duration': float_or_none(media_data.get('duration'), 1000),
	148	'subtitles': subtitles,
1ef15636	149	'formats': formats,
	150	}
	151
	152
	153	class SRGSSRPlayIE(InfoExtractor):
ce7d243c	154	IE_DESC = 'srf.ch, rts.ch, rsi.ch, rtr.ch and swissinfo.ch play sites'
25b83c2a S	155	_VALID_URL = r'''(?x)
	156	https?://
	157	(?:(?:www\|play)\.)?
	158	(?P<bu>srf\|rts\|rsi\|rtr\|swissinfo)\.ch/play/(?:tv\|radio)/
	159	(?:
	160	[^/]+/(?P<type>video\|audio)/[^?]+\|
	161	popup(?P<type_2>video\|audio)player
	162	)
06cd4cdb	163	\?.*?\b(?:id=\|urn=urn:[^:]+:video:)(?P<id>[0-9a-f\-]{36}\|\d+)
25b83c2a	164	'''
1ef15636	165
	166	_TESTS = [{
	167	'url': 'http://www.srf.ch/play/tv/10vor10/video/snowden-beantragt-asyl-in-russland?id=28e1a57d-5b76-4399-8ab3-9097f071e6c5',
ec5e77c5	168	'md5': '6db2226ba97f62ad42ce09783680046c',
1ef15636	169	'info_dict': {
1ef15636	170	'id': '28e1a57d-5b76-4399-8ab3-9097f071e6c5',
68601ef3	171	'ext': 'mp4',
1ef15636	172	'upload_date': '20130701',
1ef15636	173	'title': 'Snowden beantragt Asyl in Russland',
ec5e77c5	174	'timestamp': 1372708215,
	175	'duration': 113.827,
	176	'thumbnail': r're:^https?://.*1383719781\.png$',
1ef15636	177	},
ec5e77c5	178	'expected_warnings': ['Unable to download f4m manifest'],
7db2897d	179	}, {
e09f58b3	180	'url': 'http://www.rtr.ch/play/radio/actualitad/audio/saira-tujetsch-tuttina-cuntinuar-cun-sedrun-muster-turissem?id=63cb0778-27f8-49af-9284-8c7a8c6d15fc',
	181	'info_dict': {
	182	'id': '63cb0778-27f8-49af-9284-8c7a8c6d15fc',
	183	'ext': 'mp3',
	184	'upload_date': '20151013',
	185	'title': 'Saira: Tujetsch - tuttina cuntinuar cun Sedrun Mustér Turissem',
ec5e77c5	186	'timestamp': 1444709160,
ec5e77c5	187	'duration': 336.816,
e09f58b3	188	},
	189	'params': {
	190	# rtmp download
	191	'skip_download': True,
	192	},
7db2897d	193	}, {
	194	'url': 'http://www.rts.ch/play/tv/-/video/le-19h30?id=6348260',
	195	'md5': '67a2a9ae4e8e62a68d0e9820cc9782df',
	196	'info_dict': {
	197	'id': '6348260',
	198	'display_id': '6348260',
422f7c11	199	'ext': 'mp4',
ec5e77c5	200	'duration': 1796.76,
7db2897d	201	'title': 'Le 19h30',
7db2897d	202	'upload_date': '20141201',
7db2897d	203	'timestamp': 1417458600,
ec85ded8	204	'thumbnail': r're:^https?://.*\.image',
7db2897d	205	},
422f7c11	206	'params': {
	207	# m3u8 download
	208	'skip_download': True,
add96eb9	209	},
ec5e77c5	210	}, {
	211	'url': 'http://play.swissinfo.ch/play/tv/business/video/why-people-were-against-tax-reforms?id=42960270',
	212	'info_dict': {
	213	'id': '42960270',
	214	'ext': 'mp4',
	215	'title': 'Why people were against tax reforms',
	216	'description': 'md5:7ac442c558e9630e947427469c4b824d',
	217	'duration': 94.0,
	218	'upload_date': '20170215',
	219	'timestamp': 1487173560,
	220	'thumbnail': r're:https?://www\.swissinfo\.ch/srgscalableimage/42961964',
	221	'subtitles': 'count:9',
	222	},
	223	'params': {
	224	'skip_download': True,
add96eb9	225	},
3fe77472 SP	226	}, {
3fe77472 SP	227	'url': 'https://www.srf.ch/play/tv/popupvideoplayer?id=c4dba0ca-e75b-43b2-a34f-f708a4932e01',
25b83c2a	228	'only_matching': True,
06cd4cdb S	229	}, {
	230	'url': 'https://www.srf.ch/play/tv/10vor10/video/snowden-beantragt-asyl-in-russland?urn=urn:srf:video:28e1a57d-5b76-4399-8ab3-9097f071e6c5',
	231	'only_matching': True,
	232	}, {
	233	'url': 'https://www.rts.ch/play/tv/19h30/video/le-19h30?urn=urn:rts:video:6348260',
	234	'only_matching': True,
ec5e77c5	235	}, {
	236	# audio segment, has podcastSdUrl of the full episode
	237	'url': 'https://www.srf.ch/play/radio/popupaudioplayer?id=50b20dc8-f05b-4972-bf03-e438ff2833eb',
	238	'only_matching': True,
1ef15636	239	}]
	240
	241	def _real_extract(self, url):
5ad28e7f	242	mobj = self._match_valid_url(url)
25b83c2a S	243	bu = mobj.group('bu')
	244	media_type = mobj.group('type') or mobj.group('type_2')
	245	media_id = mobj.group('id')
add96eb9	246	return self.url_result(f'srgssr:{bu[:3]}:{media_type}:{media_id}', 'SRGSSR')