[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': 'flv',
            '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': 156.967,
            'view_count': int,
            'categories': list,
        },
    }]
    _API_DOMAIN = 'sumo.tv2.no'
    _PROTOCOLS = ('HDS', 'HLS', 'DASH')
    _GEO_COUNTRIES = ['NO']

    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'):
            raise ExtractorError('This video is DRM protected.', expected=True)
        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 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(TV2IE):
    _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']


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',
8d3a3a99	27	'ext': 'flv',
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',
	32	'duration': 156.967,
	33	'view_count': int,
	34	'categories': list,
ed1a3905	35	},
2181983a	36	}]
8d3a3a99 RA	37	_API_DOMAIN = 'sumo.tv2.no'
	38	_PROTOCOLS = ('HDS', 'HLS', 'DASH')
	39	_GEO_COUNTRIES = ['NO']
bc0f937b S	40
	41	def _real_extract(self, url):
	42	video_id = self._match_id(url)
8d3a3a99	43	api_base = 'http://%s/api/web/asset/%s' % (self._API_DOMAIN, video_id)
bc0f937b	44
2181983a	45	asset = self._download_json(
	46	api_base + '.json', video_id,
	47	'Downloading metadata JSON')['asset']
	48	title = asset.get('subtitle') or asset['title']
	49	is_live = asset.get('live') is True
	50
bc0f937b S	51	formats = []
bc0f937b S	52	format_urls = []
8d3a3a99 RA	53	for protocol in self._PROTOCOLS:
	54	try:
	55	data = self._download_json(
	56	api_base + '/play.json?protocol=%s&videoFormat=SMIL+ISMUSP' % protocol,
	57	video_id, 'Downloading play JSON')['playback']
	58	except ExtractorError as e:
	59	if isinstance(e.cause, compat_HTTPError) and e.cause.code == 401:
	60	error = self._parse_json(e.cause.read().decode(), video_id)['error']
	61	error_code = error.get('code')
	62	if error_code == 'ASSET_PLAYBACK_INVALID_GEO_LOCATION':
	63	self.raise_geo_restricted(countries=self._GEO_COUNTRIES)
	64	elif error_code == 'SESSION_NOT_AUTHENTICATED':
	65	self.raise_login_required()
	66	raise ExtractorError(error['description'])
	67	raise
9a621ddc S	68	items = try_get(data, lambda x: x['items']['item'])
	69	if not items:
	70	continue
	71	if not isinstance(items, list):
	72	items = [items]
	73	for item in items:
	74	if not isinstance(item, dict):
	75	continue
bc0f937b S	76	video_url = item.get('url')
	77	if not video_url or video_url in format_urls:
	78	continue
	79	format_id = '%s-%s' % (protocol.lower(), item.get('mediaFormat'))
	80	if not self._is_valid_url(video_url, video_id, format_id):
	81	continue
	82	format_urls.append(video_url)
	83	ext = determine_ext(video_url)
	84	if ext == 'f4m':
	85	formats.extend(self._extract_f4m_formats(
dbc0b39b	86	video_url, video_id, f4m_id=format_id, fatal=False))
bc0f937b	87	elif ext == 'm3u8':
0b25af9b RA	88	if not data.get('drmProtected'):
0b25af9b RA	89	formats.extend(self._extract_m3u8_formats(
2181983a	90	video_url, video_id, 'mp4',
2181983a	91	'm3u8' if is_live else 'm3u8_native',
0b25af9b	92	m3u8_id=format_id, fatal=False))
8d3a3a99 RA	93	elif ext == 'mpd':
	94	formats.extend(self._extract_mpd_formats(
	95	video_url, video_id, format_id, fatal=False))
bc0f937b S	96	elif ext == 'ism' or video_url.endswith('.ism/Manifest'):
	97	pass
	98	else:
	99	formats.append({
	100	'url': video_url,
	101	'format_id': format_id,
	102	'tbr': int_or_none(item.get('bitrate')),
	103	'filesize': int_or_none(item.get('fileSize')),
	104	})
0b25af9b RA	105	if not formats and data.get('drmProtected'):
0b25af9b RA	106	raise ExtractorError('This video is DRM protected.', expected=True)
bc0f937b S	107	self._sort_formats(formats)
bc0f937b S	108
bc0f937b S	109	thumbnails = [{
	110	'id': thumbnail.get('@type'),
	111	'url': thumbnail.get('url'),
0b25af9b	112	} for _, thumbnail in (asset.get('imageVersions') or {}).items()]
bc0f937b S	113
	114	return {
	115	'id': video_id,
	116	'url': video_url,
2181983a	117	'title': self._live_title(title) if is_live else title,
8d3a3a99	118	'description': strip_or_none(asset.get('description')),
bc0f937b	119	'thumbnails': thumbnails,
8d3a3a99 RA	120	'timestamp': parse_iso8601(asset.get('createTime')),
	121	'duration': float_or_none(asset.get('accurateDuration') or asset.get('duration')),
	122	'view_count': int_or_none(asset.get('views')),
	123	'categories': asset.get('keywords', '').split(','),
bc0f937b	124	'formats': formats,
2181983a	125	'is_live': is_live,
bc0f937b	126	}
588b82bb S	127
	128
	129	class TV2ArticleIE(InfoExtractor):
5886b38d	130	_VALID_URL = r'https?://(?:www\.)?tv2\.no/(?:a\|\d{4}/\d{2}/\d{2}(/[^/]+)+)/(?P<id>\d+)'
588b82bb S	131	_TESTS = [{
	132	'url': 'http://www.tv2.no/2015/05/16/nyheter/alesund/krim/pingvin/6930542',
	133	'info_dict': {
	134	'id': '6930542',
481c5c51	135	'title': 'Russen hetses etter pingvintyveri - innrømmer å ha åpnet luken på buret',
8d3a3a99	136	'description': 'De fire siktede nekter fortsatt for å ha stjålet pingvinbabyene, men innrømmer å ha åpnet luken til de små kyllingene.',
588b82bb S	137	},
	138	'playlist_count': 2,
	139	}, {
	140	'url': 'http://www.tv2.no/a/6930542',
	141	'only_matching': True,
	142	}]
	143
	144	def _real_extract(self, url):
	145	playlist_id = self._match_id(url)
	146
	147	webpage = self._download_webpage(url, playlist_id)
	148
481c5c51 S	149	# Old embed pattern (looks unused nowadays)
	150	assets = re.findall(r'data-assetid=["\'](\d+)', webpage)
	151
	152	if not assets:
	153	# New embed pattern
8d3a3a99	154	for v in re.findall(r'(?s)TV2ContentboxVideo\(({.+?})\)', webpage):
481c5c51 S	155	video = self._parse_json(
	156	v, playlist_id, transform_source=js_to_json, fatal=False)
	157	if not video:
	158	continue
	159	asset = video.get('assetId')
	160	if asset:
	161	assets.append(asset)
	162
588b82bb	163	entries = [
481c5c51 S	164	self.url_result('http://www.tv2.no/v/%s' % asset_id, 'TV2')
481c5c51 S	165	for asset_id in assets]
588b82bb S	166
	167	title = remove_end(self._og_search_title(webpage), ' - TV2.no')
	168	description = remove_end(self._og_search_description(webpage), ' - TV2.no')
	169
	170	return self.playlist_result(entries, playlist_id, title, description)
8d3a3a99 RA	171
	172
	173	class KatsomoIE(TV2IE):
2181983a	174	_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	175	_TESTS = [{
8d3a3a99 RA	176	'url': 'https://www.mtv.fi/sarja/mtv-uutiset-live-33001002003/lahden-pelicans-teki-kovan-ratkaisun-ville-nieminen-pihalle-1181321',
	177	'info_dict': {
	178	'id': '1181321',
	179	'ext': 'mp4',
2181983a	180	'title': 'Lahden Pelicans teki kovan ratkaisun – Ville Nieminen pihalle',
8d3a3a99 RA	181	'description': 'Päätöksen teki Pelicansin hallitus.',
	182	'timestamp': 1575116484,
	183	'upload_date': '20191130',
	184	'duration': 37.12,
	185	'view_count': int,
	186	'categories': list,
	187	},
	188	'params': {
	189	# m3u8 download
	190	'skip_download': True,
	191	},
2181983a	192	}, {
	193	'url': 'http://www.katsomo.fi/#!/jakso/33001005/studio55-fi/658521/jukka-kuoppamaki-tekee-yha-lauluja-vaikka-lentokoneessa',
	194	'only_matching': True,
	195	}, {
	196	'url': 'https://www.mtvuutiset.fi/video/prog1311159',
	197	'only_matching': True,
	198	}, {
	199	'url': 'https://www.katsomo.fi/#!/jakso/1311159',
	200	'only_matching': True,
	201	}]
8d3a3a99 RA	202	_API_DOMAIN = 'api.katsomo.fi'
	203	_PROTOCOLS = ('HLS', 'MPD')
	204	_GEO_COUNTRIES = ['FI']
2181983a	205
	206
	207	class MTVUutisetArticleIE(InfoExtractor):
	208	_VALID_URL = r'https?://(?:www\.)mtvuutiset\.fi/artikkeli/[^/]+/(?P<id>\d+)'
	209	_TESTS = [{
	210	'url': 'https://www.mtvuutiset.fi/artikkeli/tallaisia-vaurioita-viking-amorellassa-on-useamman-osaston-alla-vetta/7931384',
	211	'info_dict': {
	212	'id': '1311159',
	213	'ext': 'mp4',
	214	'title': 'Viking Amorellan matkustajien evakuointi on alkanut – tältä operaatio näyttää laivalla',
	215	'description': 'Viking Amorellan matkustajien evakuointi on alkanut – tältä operaatio näyttää laivalla',
	216	'timestamp': 1600608966,
	217	'upload_date': '20200920',
	218	'duration': 153.7886666,
	219	'view_count': int,
	220	'categories': list,
	221	},
	222	'params': {
	223	# m3u8 download
	224	'skip_download': True,
	225	},
	226	}, {
	227	# multiple Youtube embeds
	228	'url': 'https://www.mtvuutiset.fi/artikkeli/50-vuotta-subarun-vastaiskua/6070962',
	229	'only_matching': True,
	230	}]
	231
	232	def _real_extract(self, url):
	233	article_id = self._match_id(url)
	234	article = self._download_json(
	235	'http://api.mtvuutiset.fi/mtvuutiset/api/json/' + article_id,
	236	article_id)
	237
	238	def entries():
	239	for video in (article.get('videos') or []):
	240	video_type = video.get('videotype')
	241	video_url = video.get('url')
	242	if not (video_url and video_type in ('katsomo', 'youtube')):
	243	continue
	244	yield self.url_result(
	245	video_url, video_type.capitalize(), video.get('video_id'))
	246
	247	return self.playlist_result(
	248	entries(), article_id, article.get('title'), article.get('description'))