[yt-dlp.git] / yt_dlp / extractor / tv2.py

# coding: utf-8
from __future__ import unicode_literals

import re

from .common import InfoExtractor
from ..compat import compat_HTTPError
from ..utils import (
    determine_ext,
    ExtractorError,
    int_or_none,
    float_or_none,
    js_to_json,
    parse_iso8601,
    remove_end,
    strip_or_none,
    try_get,
)


class TV2IE(InfoExtractor):
    _VALID_URL = r'https?://(?:www\.)?tv2\.no/v/(?P<id>\d+)'
    _TESTS = [{
        'url': 'http://www.tv2.no/v/916509/',
        'info_dict': {
            'id': '916509',
            'ext': 'mp4',
            'title': 'Se Frode Gryttens hyllest av Steven Gerrard',
            'description': 'TV 2 Sportens huspoet tar avskjed med Liverpools kaptein Steven Gerrard.',
            'timestamp': 1431715610,
            'upload_date': '20150515',
            'duration': 157,
            'view_count': int,
            'categories': list,
        },
    }]
    _PROTOCOLS = ('HLS', 'DASH')
    _GEO_COUNTRIES = ['NO']

    def _real_extract(self, url):
        video_id = self._match_id(url)
        asset = self._download_json('https://sumo.tv2.no/rest/assets/' + video_id, video_id,
                                    'Downloading metadata JSON')
        title = asset['title']
        is_live = asset.get('live') is True

        formats = []
        format_urls = []
        for protocol in self._PROTOCOLS:
            try:
                data = self._download_json('https://api.sumo.tv2.no/play/%s?stream=%s' % (video_id, protocol),
                                           video_id, 'Downloading playabck JSON',
                                           headers={'content-type': 'application/json'},
                                           data='{"device":{"id":"1-1-1","name":"Nettleser (HTML)"}}'.encode())['playback']
            except ExtractorError as e:
                if isinstance(e.cause, compat_HTTPError) and e.cause.code == 401:
                    error = self._parse_json(e.cause.read().decode(), video_id)['error']
                    error_code = error.get('code')
                    if error_code == 'ASSET_PLAYBACK_INVALID_GEO_LOCATION':
                        self.raise_geo_restricted(countries=self._GEO_COUNTRIES)
                    elif error_code == 'SESSION_NOT_AUTHENTICATED':
                        self.raise_login_required()
                    raise ExtractorError(error['description'])
                raise
            items = data.get('streams', [])
            for item in items:
                video_url = item.get('url')
                if not video_url or video_url in format_urls:
                    continue
                format_id = '%s-%s' % (protocol.lower(), item.get('type'))
                if not self._is_valid_url(video_url, video_id, format_id):
                    continue
                format_urls.append(video_url)
                ext = determine_ext(video_url)
                if ext == 'f4m':
                    formats.extend(self._extract_f4m_formats(
                        video_url, video_id, f4m_id=format_id, fatal=False))
                elif ext == 'm3u8':
                    if not data.get('drmProtected'):
                        formats.extend(self._extract_m3u8_formats(
                            video_url, video_id, 'mp4',
                            'm3u8' if is_live else 'm3u8_native',
                            m3u8_id=format_id, fatal=False))
                elif ext == 'mpd':
                    formats.extend(self._extract_mpd_formats(
                        video_url, video_id, format_id, fatal=False))
                elif ext == 'ism' or video_url.endswith('.ism/Manifest'):
                    pass
                else:
                    formats.append({
                        'url': video_url,
                        'format_id': format_id,
                    })
        if not formats and data.get('drmProtected'):
            self.report_drm(video_id)
        self._sort_formats(formats)

        thumbnails = [{
            'id': type,
            'url': thumb_url,
        } for type, thumb_url in (asset.get('images') or {}).items()]

        return {
            'id': video_id,
            'url': video_url,
            'title': self._live_title(title) if is_live else title,
            'description': strip_or_none(asset.get('description')),
            'thumbnails': thumbnails,
            'timestamp': parse_iso8601(asset.get('live_broadcast_time') or asset.get('update_time')),
            'duration': float_or_none(asset.get('accurateDuration') or asset.get('duration')),
            'view_count': int_or_none(asset.get('views')),
            'categories': asset.get('tags', '').split(','),
            'formats': formats,
            'is_live': is_live,
        }


class TV2ArticleIE(InfoExtractor):
    _VALID_URL = r'https?://(?:www\.)?tv2\.no/(?:a|\d{4}/\d{2}/\d{2}(/[^/]+)+)/(?P<id>\d+)'
    _TESTS = [{
        'url': 'http://www.tv2.no/2015/05/16/nyheter/alesund/krim/pingvin/6930542',
        'info_dict': {
            'id': '6930542',
            'title': 'Russen hetses etter pingvintyveri - innrømmer å ha åpnet luken på buret',
            'description': 'De fire siktede nekter fortsatt for å ha stjålet pingvinbabyene, men innrømmer å ha åpnet luken til de små kyllingene.',
        },
        'playlist_count': 2,
    }, {
        'url': 'http://www.tv2.no/a/6930542',
        'only_matching': True,
    }]

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

        webpage = self._download_webpage(url, playlist_id)

        # Old embed pattern (looks unused nowadays)
        assets = re.findall(r'data-assetid=["\'](\d+)', webpage)

        if not assets:
            # New embed pattern
            for v in re.findall(r'(?s)TV2ContentboxVideo\(({.+?})\)', webpage):
                video = self._parse_json(
                    v, playlist_id, transform_source=js_to_json, fatal=False)
                if not video:
                    continue
                asset = video.get('assetId')
                if asset:
                    assets.append(asset)

        entries = [
            self.url_result('http://www.tv2.no/v/%s' % asset_id, 'TV2')
            for asset_id in assets]

        title = remove_end(self._og_search_title(webpage), ' - TV2.no')
        description = remove_end(self._og_search_description(webpage), ' - TV2.no')

        return self.playlist_result(entries, playlist_id, title, description)


class KatsomoIE(InfoExtractor):
    _VALID_URL = r'https?://(?:www\.)?(?:katsomo|mtv(uutiset)?)\.fi/(?:sarja/[0-9a-z-]+-\d+/[0-9a-z-]+-|(?:#!/)?jakso/(?:\d+/[^/]+/)?|video/prog)(?P<id>\d+)'
    _TESTS = [{
        'url': 'https://www.mtv.fi/sarja/mtv-uutiset-live-33001002003/lahden-pelicans-teki-kovan-ratkaisun-ville-nieminen-pihalle-1181321',
        'info_dict': {
            'id': '1181321',
            'ext': 'mp4',
            'title': 'Lahden Pelicans teki kovan ratkaisun – Ville Nieminen pihalle',
            'description': 'Päätöksen teki Pelicansin hallitus.',
            'timestamp': 1575116484,
            'upload_date': '20191130',
            'duration': 37.12,
            'view_count': int,
            'categories': list,
        },
        'params': {
            # m3u8 download
            'skip_download': True,
        },
    }, {
        'url': 'http://www.katsomo.fi/#!/jakso/33001005/studio55-fi/658521/jukka-kuoppamaki-tekee-yha-lauluja-vaikka-lentokoneessa',
        'only_matching': True,
    }, {
        'url': 'https://www.mtvuutiset.fi/video/prog1311159',
        'only_matching': True,
    }, {
        'url': 'https://www.katsomo.fi/#!/jakso/1311159',
        'only_matching': True,
    }]
    _API_DOMAIN = 'api.katsomo.fi'
    _PROTOCOLS = ('HLS', 'MPD')
    _GEO_COUNTRIES = ['FI']

    def _real_extract(self, url):
        video_id = self._match_id(url)
        api_base = 'http://%s/api/web/asset/%s' % (self._API_DOMAIN, video_id)

        asset = self._download_json(
            api_base + '.json', video_id,
            'Downloading metadata JSON')['asset']
        title = asset.get('subtitle') or asset['title']
        is_live = asset.get('live') is True

        formats = []
        format_urls = []
        for protocol in self._PROTOCOLS:
            try:
                data = self._download_json(
                    api_base + '/play.json?protocol=%s&videoFormat=SMIL+ISMUSP' % protocol,
                    video_id, 'Downloading play JSON')['playback']
            except ExtractorError as e:
                if isinstance(e.cause, compat_HTTPError) and e.cause.code == 401:
                    error = self._parse_json(e.cause.read().decode(), video_id)['error']
                    error_code = error.get('code')
                    if error_code == 'ASSET_PLAYBACK_INVALID_GEO_LOCATION':
                        self.raise_geo_restricted(countries=self._GEO_COUNTRIES)
                    elif error_code == 'SESSION_NOT_AUTHENTICATED':
                        self.raise_login_required()
                    raise ExtractorError(error['description'])
                raise
            items = try_get(data, lambda x: x['items']['item'])
            if not items:
                continue
            if not isinstance(items, list):
                items = [items]
            for item in items:
                if not isinstance(item, dict):
                    continue
                video_url = item.get('url')
                if not video_url or video_url in format_urls:
                    continue
                format_id = '%s-%s' % (protocol.lower(), item.get('mediaFormat'))
                if not self._is_valid_url(video_url, video_id, format_id):
                    continue
                format_urls.append(video_url)
                ext = determine_ext(video_url)
                if ext == 'f4m':
                    formats.extend(self._extract_f4m_formats(
                        video_url, video_id, f4m_id=format_id, fatal=False))
                elif ext == 'm3u8':
                    if not data.get('drmProtected'):
                        formats.extend(self._extract_m3u8_formats(
                            video_url, video_id, 'mp4',
                            'm3u8' if is_live else 'm3u8_native',
                            m3u8_id=format_id, fatal=False))
                elif ext == 'mpd':
                    formats.extend(self._extract_mpd_formats(
                        video_url, video_id, format_id, fatal=False))
                elif ext == 'ism' or video_url.endswith('.ism/Manifest'):
                    pass
                else:
                    formats.append({
                        'url': video_url,
                        'format_id': format_id,
                        'tbr': int_or_none(item.get('bitrate')),
                        'filesize': int_or_none(item.get('fileSize')),
                    })
        if not formats and data.get('drmProtected'):
            self.report_drm(video_id)
        self._sort_formats(formats)

        thumbnails = [{
            'id': thumbnail.get('@type'),
            'url': thumbnail.get('url'),
        } for _, thumbnail in (asset.get('imageVersions') or {}).items()]

        return {
            'id': video_id,
            'url': video_url,
            'title': self._live_title(title) if is_live else title,
            'description': strip_or_none(asset.get('description')),
            'thumbnails': thumbnails,
            'timestamp': parse_iso8601(asset.get('createTime')),
            'duration': float_or_none(asset.get('accurateDuration') or asset.get('duration')),
            'view_count': int_or_none(asset.get('views')),
            'categories': asset.get('keywords', '').split(','),
            'formats': formats,
            'is_live': is_live,
        }


class MTVUutisetArticleIE(InfoExtractor):
    _VALID_URL = r'https?://(?:www\.)mtvuutiset\.fi/artikkeli/[^/]+/(?P<id>\d+)'
    _TESTS = [{
        'url': 'https://www.mtvuutiset.fi/artikkeli/tallaisia-vaurioita-viking-amorellassa-on-useamman-osaston-alla-vetta/7931384',
        'info_dict': {
            'id': '1311159',
            'ext': 'mp4',
            'title': 'Viking Amorellan matkustajien evakuointi on alkanut – tältä operaatio näyttää laivalla',
            'description': 'Viking Amorellan matkustajien evakuointi on alkanut – tältä operaatio näyttää laivalla',
            'timestamp': 1600608966,
            'upload_date': '20200920',
            'duration': 153.7886666,
            'view_count': int,
            'categories': list,
        },
        'params': {
            # m3u8 download
            'skip_download': True,
        },
    }, {
        # multiple Youtube embeds
        'url': 'https://www.mtvuutiset.fi/artikkeli/50-vuotta-subarun-vastaiskua/6070962',
        'only_matching': True,
    }]

    def _real_extract(self, url):
        article_id = self._match_id(url)
        article = self._download_json(
            'http://api.mtvuutiset.fi/mtvuutiset/api/json/' + article_id,
            article_id)

        def entries():
            for video in (article.get('videos') or []):
                video_type = video.get('videotype')
                video_url = video.get('url')
                if not (video_url and video_type in ('katsomo', 'youtube')):
                    continue
                yield self.url_result(
                    video_url, video_type.capitalize(), video.get('video_id'))

        return self.playlist_result(
            entries(), article_id, article.get('title'), article.get('description'))
Commit	Line	Data
dcdb292f	1	# coding: utf-8
bc0f937b S	2	from __future__ import unicode_literals
bc0f937b S	3
588b82bb S	4	import re
588b82bb S	5
bc0f937b	6	from .common import InfoExtractor
8d3a3a99	7	from ..compat import compat_HTTPError
bc0f937b S	8	from ..utils import (
bc0f937b S	9	determine_ext,
8d3a3a99	10	ExtractorError,
bc0f937b S	11	int_or_none,
bc0f937b S	12	float_or_none,
481c5c51	13	js_to_json,
bc0f937b	14	parse_iso8601,
588b82bb	15	remove_end,
8d3a3a99	16	strip_or_none,
9a621ddc	17	try_get,
bc0f937b S	18	)
	19
	20
	21	class TV2IE(InfoExtractor):
5886b38d	22	_VALID_URL = r'https?://(?:www\.)?tv2\.no/v/(?P<id>\d+)'
2181983a	23	_TESTS = [{
bc0f937b	24	'url': 'http://www.tv2.no/v/916509/',
bc0f937b S	25	'info_dict': {
bc0f937b S	26	'id': '916509',
ea81966e	27	'ext': 'mp4',
ed1a3905	28	'title': 'Se Frode Gryttens hyllest av Steven Gerrard',
bc0f937b S	29	'description': 'TV 2 Sportens huspoet tar avskjed med Liverpools kaptein Steven Gerrard.',
	30	'timestamp': 1431715610,
	31	'upload_date': '20150515',
ea81966e	32	'duration': 157,
bc0f937b S	33	'view_count': int,
bc0f937b S	34	'categories': list,
ed1a3905	35	},
2181983a	36	}]
ea81966e	37	_PROTOCOLS = ('HLS', 'DASH')
8d3a3a99	38	_GEO_COUNTRIES = ['NO']
bc0f937b S	39
	40	def _real_extract(self, url):
	41	video_id = self._match_id(url)
ea81966e A	42	asset = self._download_json('https://sumo.tv2.no/rest/assets/' + video_id, video_id,
	43	'Downloading metadata JSON')
	44	title = asset['title']
2181983a	45	is_live = asset.get('live') is True
2181983a	46
bc0f937b S	47	formats = []
bc0f937b S	48	format_urls = []
8d3a3a99 RA	49	for protocol in self._PROTOCOLS:
8d3a3a99 RA	50	try:
ea81966e A	51	data = self._download_json('https://api.sumo.tv2.no/play/%s?stream=%s' % (video_id, protocol),
	52	video_id, 'Downloading playabck JSON',
	53	headers={'content-type': 'application/json'},
	54	data='{"device":{"id":"1-1-1","name":"Nettleser (HTML)"}}'.encode())['playback']
8d3a3a99 RA	55	except ExtractorError as e:
	56	if isinstance(e.cause, compat_HTTPError) and e.cause.code == 401:
	57	error = self._parse_json(e.cause.read().decode(), video_id)['error']
	58	error_code = error.get('code')
	59	if error_code == 'ASSET_PLAYBACK_INVALID_GEO_LOCATION':
	60	self.raise_geo_restricted(countries=self._GEO_COUNTRIES)
	61	elif error_code == 'SESSION_NOT_AUTHENTICATED':
	62	self.raise_login_required()
	63	raise ExtractorError(error['description'])
	64	raise
ea81966e	65	items = data.get('streams', [])
9a621ddc	66	for item in items:
bc0f937b S	67	video_url = item.get('url')
	68	if not video_url or video_url in format_urls:
	69	continue
ea81966e	70	format_id = '%s-%s' % (protocol.lower(), item.get('type'))
bc0f937b S	71	if not self._is_valid_url(video_url, video_id, format_id):
	72	continue
	73	format_urls.append(video_url)
	74	ext = determine_ext(video_url)
	75	if ext == 'f4m':
	76	formats.extend(self._extract_f4m_formats(
dbc0b39b	77	video_url, video_id, f4m_id=format_id, fatal=False))
bc0f937b	78	elif ext == 'm3u8':
0b25af9b RA	79	if not data.get('drmProtected'):
0b25af9b RA	80	formats.extend(self._extract_m3u8_formats(
2181983a	81	video_url, video_id, 'mp4',
2181983a	82	'm3u8' if is_live else 'm3u8_native',
0b25af9b	83	m3u8_id=format_id, fatal=False))
8d3a3a99 RA	84	elif ext == 'mpd':
	85	formats.extend(self._extract_mpd_formats(
	86	video_url, video_id, format_id, fatal=False))
bc0f937b S	87	elif ext == 'ism' or video_url.endswith('.ism/Manifest'):
	88	pass
	89	else:
	90	formats.append({
	91	'url': video_url,
	92	'format_id': format_id,
bc0f937b	93	})
0b25af9b	94	if not formats and data.get('drmProtected'):
88acdbc2	95	self.report_drm(video_id)
bc0f937b S	96	self._sort_formats(formats)
bc0f937b S	97
bc0f937b	98	thumbnails = [{
ea81966e A	99	'id': type,
	100	'url': thumb_url,
	101	} for type, thumb_url in (asset.get('images') or {}).items()]
bc0f937b S	102
	103	return {
	104	'id': video_id,
	105	'url': video_url,
2181983a	106	'title': self._live_title(title) if is_live else title,
8d3a3a99	107	'description': strip_or_none(asset.get('description')),
bc0f937b	108	'thumbnails': thumbnails,
ea81966e	109	'timestamp': parse_iso8601(asset.get('live_broadcast_time') or asset.get('update_time')),
8d3a3a99 RA	110	'duration': float_or_none(asset.get('accurateDuration') or asset.get('duration')),
8d3a3a99 RA	111	'view_count': int_or_none(asset.get('views')),
ea81966e	112	'categories': asset.get('tags', '').split(','),
bc0f937b	113	'formats': formats,
2181983a	114	'is_live': is_live,
bc0f937b	115	}
588b82bb S	116
	117
	118	class TV2ArticleIE(InfoExtractor):
5886b38d	119	_VALID_URL = r'https?://(?:www\.)?tv2\.no/(?:a\|\d{4}/\d{2}/\d{2}(/[^/]+)+)/(?P<id>\d+)'
588b82bb S	120	_TESTS = [{
	121	'url': 'http://www.tv2.no/2015/05/16/nyheter/alesund/krim/pingvin/6930542',
	122	'info_dict': {
	123	'id': '6930542',
481c5c51	124	'title': 'Russen hetses etter pingvintyveri - innrømmer å ha åpnet luken på buret',
8d3a3a99	125	'description': 'De fire siktede nekter fortsatt for å ha stjålet pingvinbabyene, men innrømmer å ha åpnet luken til de små kyllingene.',
588b82bb S	126	},
	127	'playlist_count': 2,
	128	}, {
	129	'url': 'http://www.tv2.no/a/6930542',
	130	'only_matching': True,
	131	}]
	132
	133	def _real_extract(self, url):
	134	playlist_id = self._match_id(url)
	135
	136	webpage = self._download_webpage(url, playlist_id)
	137
481c5c51 S	138	# Old embed pattern (looks unused nowadays)
	139	assets = re.findall(r'data-assetid=["\'](\d+)', webpage)
	140
	141	if not assets:
	142	# New embed pattern
8d3a3a99	143	for v in re.findall(r'(?s)TV2ContentboxVideo\(({.+?})\)', webpage):
481c5c51 S	144	video = self._parse_json(
	145	v, playlist_id, transform_source=js_to_json, fatal=False)
	146	if not video:
	147	continue
	148	asset = video.get('assetId')
	149	if asset:
	150	assets.append(asset)
	151
588b82bb	152	entries = [
481c5c51 S	153	self.url_result('http://www.tv2.no/v/%s' % asset_id, 'TV2')
481c5c51 S	154	for asset_id in assets]
588b82bb S	155
	156	title = remove_end(self._og_search_title(webpage), ' - TV2.no')
	157	description = remove_end(self._og_search_description(webpage), ' - TV2.no')
	158
	159	return self.playlist_result(entries, playlist_id, title, description)
8d3a3a99 RA	160
8d3a3a99 RA	161
ea81966e	162	class KatsomoIE(InfoExtractor):
2181983a	163	_VALID_URL = r'https?://(?:www\.)?(?:katsomo\|mtv(uutiset)?)\.fi/(?:sarja/[0-9a-z-]+-\d+/[0-9a-z-]+-\|(?:#!/)?jakso/(?:\d+/[^/]+/)?\|video/prog)(?P<id>\d+)'
2181983a	164	_TESTS = [{
8d3a3a99 RA	165	'url': 'https://www.mtv.fi/sarja/mtv-uutiset-live-33001002003/lahden-pelicans-teki-kovan-ratkaisun-ville-nieminen-pihalle-1181321',
	166	'info_dict': {
	167	'id': '1181321',
	168	'ext': 'mp4',
2181983a	169	'title': 'Lahden Pelicans teki kovan ratkaisun – Ville Nieminen pihalle',
8d3a3a99 RA	170	'description': 'Päätöksen teki Pelicansin hallitus.',
	171	'timestamp': 1575116484,
	172	'upload_date': '20191130',
	173	'duration': 37.12,
	174	'view_count': int,
	175	'categories': list,
	176	},
	177	'params': {
	178	# m3u8 download
	179	'skip_download': True,
	180	},
2181983a	181	}, {
	182	'url': 'http://www.katsomo.fi/#!/jakso/33001005/studio55-fi/658521/jukka-kuoppamaki-tekee-yha-lauluja-vaikka-lentokoneessa',
	183	'only_matching': True,
	184	}, {
	185	'url': 'https://www.mtvuutiset.fi/video/prog1311159',
	186	'only_matching': True,
	187	}, {
	188	'url': 'https://www.katsomo.fi/#!/jakso/1311159',
	189	'only_matching': True,
	190	}]
8d3a3a99 RA	191	_API_DOMAIN = 'api.katsomo.fi'
	192	_PROTOCOLS = ('HLS', 'MPD')
	193	_GEO_COUNTRIES = ['FI']
2181983a	194
ea81966e A	195	def _real_extract(self, url):
	196	video_id = self._match_id(url)
	197	api_base = 'http://%s/api/web/asset/%s' % (self._API_DOMAIN, video_id)
	198
	199	asset = self._download_json(
	200	api_base + '.json', video_id,
	201	'Downloading metadata JSON')['asset']
	202	title = asset.get('subtitle') or asset['title']
	203	is_live = asset.get('live') is True
	204
	205	formats = []
	206	format_urls = []
	207	for protocol in self._PROTOCOLS:
	208	try:
	209	data = self._download_json(
	210	api_base + '/play.json?protocol=%s&videoFormat=SMIL+ISMUSP' % protocol,
	211	video_id, 'Downloading play JSON')['playback']
	212	except ExtractorError as e:
	213	if isinstance(e.cause, compat_HTTPError) and e.cause.code == 401:
	214	error = self._parse_json(e.cause.read().decode(), video_id)['error']
	215	error_code = error.get('code')
	216	if error_code == 'ASSET_PLAYBACK_INVALID_GEO_LOCATION':
	217	self.raise_geo_restricted(countries=self._GEO_COUNTRIES)
	218	elif error_code == 'SESSION_NOT_AUTHENTICATED':
	219	self.raise_login_required()
	220	raise ExtractorError(error['description'])
	221	raise
	222	items = try_get(data, lambda x: x['items']['item'])
	223	if not items:
	224	continue
	225	if not isinstance(items, list):
	226	items = [items]
	227	for item in items:
	228	if not isinstance(item, dict):
	229	continue
	230	video_url = item.get('url')
	231	if not video_url or video_url in format_urls:
	232	continue
	233	format_id = '%s-%s' % (protocol.lower(), item.get('mediaFormat'))
	234	if not self._is_valid_url(video_url, video_id, format_id):
	235	continue
	236	format_urls.append(video_url)
	237	ext = determine_ext(video_url)
	238	if ext == 'f4m':
	239	formats.extend(self._extract_f4m_formats(
	240	video_url, video_id, f4m_id=format_id, fatal=False))
	241	elif ext == 'm3u8':
	242	if not data.get('drmProtected'):
	243	formats.extend(self._extract_m3u8_formats(
	244	video_url, video_id, 'mp4',
	245	'm3u8' if is_live else 'm3u8_native',
	246	m3u8_id=format_id, fatal=False))
	247	elif ext == 'mpd':
	248	formats.extend(self._extract_mpd_formats(
	249	video_url, video_id, format_id, fatal=False))
	250	elif ext == 'ism' or video_url.endswith('.ism/Manifest'):
	251	pass
	252	else:
	253	formats.append({
	254	'url': video_url,
	255	'format_id': format_id,
	256	'tbr': int_or_none(item.get('bitrate')),
	257	'filesize': int_or_none(item.get('fileSize')),
	258	})
259	if not formats and data.get('drmProtected'):
260	self.report_drm(video_id)
261	self._sort_formats(formats)
262
263	thumbnails = [{
264	'id': thumbnail.get('@type'),
265	'url': thumbnail.get('url'),
266	} for _, thumbnail in (asset.get('imageVersions') or {}).items()]
267
268	return {
269	'id': video_id,
270	'url': video_url,
271	'title': self._live_title(title) if is_live else title,
272	'description': strip_or_none(asset.get('description')),
273	'thumbnails': thumbnails,
274	'timestamp': parse_iso8601(asset.get('createTime')),
275	'duration': float_or_none(asset.get('accurateDuration') or asset.get('duration')),
276	'view_count': int_or_none(asset.get('views')),
277	'categories': asset.get('keywords', '').split(','),
278	'formats': formats,
279	'is_live': is_live,
280	}
281
2181983a	282
	283	class MTVUutisetArticleIE(InfoExtractor):
	284	_VALID_URL = r'https?://(?:www\.)mtvuutiset\.fi/artikkeli/[^/]+/(?P<id>\d+)'
	285	_TESTS = [{
	286	'url': 'https://www.mtvuutiset.fi/artikkeli/tallaisia-vaurioita-viking-amorellassa-on-useamman-osaston-alla-vetta/7931384',
	287	'info_dict': {
	288	'id': '1311159',
	289	'ext': 'mp4',
	290	'title': 'Viking Amorellan matkustajien evakuointi on alkanut – tältä operaatio näyttää laivalla',
	291	'description': 'Viking Amorellan matkustajien evakuointi on alkanut – tältä operaatio näyttää laivalla',
	292	'timestamp': 1600608966,
	293	'upload_date': '20200920',
	294	'duration': 153.7886666,
	295	'view_count': int,
	296	'categories': list,
	297	},
	298	'params': {
	299	# m3u8 download
	300	'skip_download': True,
	301	},
	302	}, {
	303	# multiple Youtube embeds
	304	'url': 'https://www.mtvuutiset.fi/artikkeli/50-vuotta-subarun-vastaiskua/6070962',
	305	'only_matching': True,
	306	}]
	307
	308	def _real_extract(self, url):
	309	article_id = self._match_id(url)
	310	article = self._download_json(
	311	'http://api.mtvuutiset.fi/mtvuutiset/api/json/' + article_id,
	312	article_id)
	313
	314	def entries():
	315	for video in (article.get('videos') or []):
	316	video_type = video.get('videotype')
	317	video_url = video.get('url')
	318	if not (video_url and video_type in ('katsomo', 'youtube')):
	319	continue
	320	yield self.url_result(
	321	video_url, video_type.capitalize(), video.get('video_id'))
	322
	323	return self.playlist_result(
	324	entries(), article_id, article.get('title'), article.get('description'))