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

import json

from .common import InfoExtractor
from .brightcove import BrightcoveNewIE

from ..compat import compat_str
from ..utils import (
    base_url,
    clean_html,
    determine_ext,
    extract_attributes,
    ExtractorError,
    get_element_by_class,
    JSON_LD_RE,
    merge_dicts,
    parse_duration,
    smuggle_url,
    try_get,
    url_or_none,
    url_basename,
    urljoin,
)


class ITVIE(InfoExtractor):
    _VALID_URL = r'https?://(?:www\.)?itv\.com/hub/[^/]+/(?P<id>[0-9a-zA-Z]+)'
    _GEO_COUNTRIES = ['GB']
    _TESTS = [{
        'url': 'https://www.itv.com/hub/plebs/2a1873a0002',
        'info_dict': {
            'id': '2a1873a0002',
            'ext': 'mp4',
            'title': 'Plebs - The Orgy',
            'description': 'md5:4d7159af53ebd5b36e8b3ec82a41fdb4',
            'series': 'Plebs',
            'season_number': 1,
            'episode_number': 1,
            'thumbnail': r're:https?://hubimages\.itv\.com/episode/2_1873_0002'
        },
        'params': {
            # m3u8 download
            'skip_download': True,
        },
    }, {
        'url': 'https://www.itv.com/hub/the-jonathan-ross-show/2a1166a0209',
        'info_dict': {
            'id': '2a1166a0209',
            'ext': 'mp4',
            'title': 'The Jonathan Ross Show - Series 17 - Episode 8',
            'description': 'md5:3023dcdd375db1bc9967186cdb3f1399',
            'series': 'The Jonathan Ross Show',
            'episode_number': 8,
            'season_number': 17,
            'thumbnail': r're:https?://hubimages\.itv\.com/episode/2_1873_0002'
        },
        'params': {
            # m3u8 download
            'skip_download': True,
        },
    }, {
        # unavailable via data-playlist-url
        'url': 'https://www.itv.com/hub/through-the-keyhole/2a2271a0033',
        'only_matching': True,
    }, {
        # InvalidVodcrid
        'url': 'https://www.itv.com/hub/james-martins-saturday-morning/2a5159a0034',
        'only_matching': True,
    }, {
        # ContentUnavailable
        'url': 'https://www.itv.com/hub/whos-doing-the-dishes/2a2898a0024',
        'only_matching': True,
    }]

    def _generate_api_headers(self, hmac):
        return merge_dicts({
            'Accept': 'application/vnd.itv.vod.playlist.v2+json',
            'Content-Type': 'application/json',
            'hmac': hmac.upper(),
        }, self.geo_verification_headers())

    def _call_api(self, video_id, playlist_url, headers, platform_tag, featureset, fatal=True):
        return self._download_json(
            playlist_url, video_id, data=json.dumps({
                'user': {
                    'itvUserId': '',
                    'entitlements': [],
                    'token': ''
                },
                'device': {
                    'manufacturer': 'Safari',
                    'model': '5',
                    'os': {
                        'name': 'Windows NT',
                        'version': '6.1',
                        'type': 'desktop'
                    }
                },
                'client': {
                    'version': '4.1',
                    'id': 'browser'
                },
                'variantAvailability': {
                    'featureset': {
                        'min': featureset,
                        'max': featureset
                    },
                    'platformTag': platform_tag
                }
            }).encode(), headers=headers, fatal=fatal)

    def _get_subtitles(self, video_id, variants, ios_playlist_url, headers, *args, **kwargs):
        subtitles = {}
        # Prefer last matching featureset
        # See: https://github.com/yt-dlp/yt-dlp/issues/986
        platform_tag_subs, featureset_subs = next(
            ((platform_tag, featureset)
             for platform_tag, featuresets in reversed(list(variants.items())) for featureset in featuresets
             if try_get(featureset, lambda x: x[2]) == 'outband-webvtt'),
            (None, None))

        if platform_tag_subs and featureset_subs:
            subs_playlist = self._call_api(
                video_id, ios_playlist_url, headers, platform_tag_subs, featureset_subs, fatal=False)
            subs = try_get(subs_playlist, lambda x: x['Playlist']['Video']['Subtitles'], list) or []
            for sub in subs:
                if not isinstance(sub, dict):
                    continue
                href = url_or_none(sub.get('Href'))
                if not href:
                    continue
                subtitles.setdefault('en', []).append({'url': href})
        return subtitles

    def _real_extract(self, url):
        video_id = self._match_id(url)
        webpage = self._download_webpage(url, video_id)
        params = extract_attributes(self._search_regex(
            r'(?s)(<[^>]+id="video"[^>]*>)', webpage, 'params'))
        variants = self._parse_json(
            try_get(params, lambda x: x['data-video-variants'], compat_str) or '{}',
            video_id, fatal=False)
        # Prefer last matching featureset
        # See: https://github.com/yt-dlp/yt-dlp/issues/986
        platform_tag_video, featureset_video = next(
            ((platform_tag, featureset)
             for platform_tag, featuresets in reversed(list(variants.items())) for featureset in featuresets
             if set(try_get(featureset, lambda x: x[:2]) or []) == {'aes', 'hls'}),
            (None, None))
        if not platform_tag_video or not featureset_video:
            raise ExtractorError('No downloads available', expected=True, video_id=video_id)

        ios_playlist_url = params.get('data-video-playlist') or params['data-video-id']
        headers = self._generate_api_headers(params['data-video-hmac'])
        ios_playlist = self._call_api(
            video_id, ios_playlist_url, headers, platform_tag_video, featureset_video)

        video_data = try_get(ios_playlist, lambda x: x['Playlist']['Video'], dict) or {}
        ios_base_url = video_data.get('Base')
        formats = []
        for media_file in (video_data.get('MediaFiles') or []):
            href = media_file.get('Href')
            if not href:
                continue
            if ios_base_url:
                href = ios_base_url + href
            ext = determine_ext(href)
            if ext == 'm3u8':
                formats.extend(self._extract_m3u8_formats(
                    href, video_id, 'mp4', entry_protocol='m3u8_native',
                    m3u8_id='hls', fatal=False))
            else:
                formats.append({
                    'url': href,
                })
        info = self._search_json_ld(webpage, video_id, default={})
        if not info:
            json_ld = self._parse_json(self._search_regex(
                JSON_LD_RE, webpage, 'JSON-LD', '{}',
                group='json_ld'), video_id, fatal=False)
            if json_ld and json_ld.get('@type') == 'BreadcrumbList':
                for ile in (json_ld.get('itemListElement:') or []):
                    item = ile.get('item:') or {}
                    if item.get('@type') == 'TVEpisode':
                        item['@context'] = 'http://schema.org'
                        info = self._json_ld(item, video_id, fatal=False) or {}
                        break

        thumbnails = []
        thumbnail_url = try_get(params, lambda x: x['data-video-posterframe'], compat_str)
        if thumbnail_url:
            thumbnails.extend([{
                'url': thumbnail_url.format(width=1920, height=1080, quality=100, blur=0, bg='false'),
                'width': 1920,
                'height': 1080,
            }, {
                'url': urljoin(base_url(thumbnail_url), url_basename(thumbnail_url)),
                'preference': -2
            }])

        thumbnail_url = self._html_search_meta(['og:image', 'twitter:image'], webpage, default=None)
        if thumbnail_url:
            thumbnails.append({
                'url': thumbnail_url,
            })
        self._remove_duplicate_formats(thumbnails)

        return merge_dicts({
            'id': video_id,
            'title': self._html_search_meta(['og:title', 'twitter:title'], webpage),
            'formats': formats,
            'subtitles': self.extract_subtitles(video_id, variants, ios_playlist_url, headers),
            'duration': parse_duration(video_data.get('Duration')),
            'description': clean_html(get_element_by_class('episode-info__synopsis', webpage)),
            'thumbnails': thumbnails
        }, info)


class ITVBTCCIE(InfoExtractor):
    _VALID_URL = r'https?://(?:www\.)?itv\.com/(?:news|btcc)/(?:[^/]+/)*(?P<id>[^/?#&]+)'
    _TESTS = [{
        'url': 'https://www.itv.com/btcc/articles/btcc-2019-brands-hatch-gp-race-action',
        'info_dict': {
            'id': 'btcc-2019-brands-hatch-gp-race-action',
            'title': 'BTCC 2019: Brands Hatch GP race action',
        },
        'playlist_count': 12,
    }, {
        'url': 'https://www.itv.com/news/2021-10-27/i-have-to-protect-the-country-says-rishi-sunak-as-uk-faces-interest-rate-hike',
        'info_dict': {
            'id': 'i-have-to-protect-the-country-says-rishi-sunak-as-uk-faces-interest-rate-hike',
            'title': 'md5:6ef054dd9f069330db3dcc66cb772d32'
        },
        'playlist_count': 4
    }]
    BRIGHTCOVE_URL_TEMPLATE = 'http://players.brightcove.net/%s/%s_default/index.html?videoId=%s'

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

        webpage = self._download_webpage(url, playlist_id)

        json_map = try_get(
            self._search_nextjs_data(webpage, playlist_id),
            lambda x: x['props']['pageProps']['article']['body']['content']) or []

        entries = []
        for video in json_map:
            if not any(video['data'].get(attr) == 'Brightcove' for attr in ('name', 'type')):
                continue
            video_id = video['data']['id']
            account_id = video['data']['accountId']
            player_id = video['data']['playerId']
            entries.append(self.url_result(
                smuggle_url(self.BRIGHTCOVE_URL_TEMPLATE % (account_id, player_id, video_id), {
                    # ITV does not like some GB IP ranges, so here are some
                    # IP blocks it accepts
                    'geo_ip_blocks': [
                        '193.113.0.0/16', '54.36.162.0/23', '159.65.16.0/21'
                    ],
                    'referrer': url,
                }),
                ie=BrightcoveNewIE.ie_key(), video_id=video_id))

        title = self._og_search_title(webpage, fatal=False)

        return self.playlist_result(entries, playlist_id, title)
Commit	Line	Data
a71b8d3b RA	1	import json
	2
	3	from .common import InfoExtractor
ea1f5e5d	4	from .brightcove import BrightcoveNewIE
ea706726	5
ea706726	6	from ..compat import compat_str
a71b8d3b	7	from ..utils import (
ea706726	8	base_url,
29f7c58a	9	clean_html,
a4ec4517	10	determine_ext,
a71b8d3b	11	extract_attributes,
ea706726	12	ExtractorError,
29f7c58a	13	get_element_by_class,
29f7c58a	14	JSON_LD_RE,
30374f4d	15	merge_dicts,
a71b8d3b	16	parse_duration,
ea1f5e5d	17	smuggle_url,
6857df60	18	try_get,
3052a30d	19	url_or_none,
ea706726	20	url_basename,
ea706726	21	urljoin,
a71b8d3b RA	22	)
	23
	24
	25	class ITVIE(InfoExtractor):
f592ff98	26	_VALID_URL = r'https?://(?:www\.)?itv\.com/hub/[^/]+/(?P<id>[0-9a-zA-Z]+)'
4248dad9	27	_GEO_COUNTRIES = ['GB']
3fae11ac	28	_TESTS = [{
ea706726	29	'url': 'https://www.itv.com/hub/plebs/2a1873a0002',
	30	'info_dict': {
	31	'id': '2a1873a0002',
	32	'ext': 'mp4',
	33	'title': 'Plebs - The Orgy',
	34	'description': 'md5:4d7159af53ebd5b36e8b3ec82a41fdb4',
	35	'series': 'Plebs',
	36	'season_number': 1,
	37	'episode_number': 1,
	38	'thumbnail': r're:https?://hubimages\.itv\.com/episode/2_1873_0002'
	39	},
	40	'params': {
	41	# m3u8 download
	42	'skip_download': True,
	43	},
	44	}, {
	45	'url': 'https://www.itv.com/hub/the-jonathan-ross-show/2a1166a0209',
a71b8d3b	46	'info_dict': {
ea706726	47	'id': '2a1166a0209',
29f7c58a	48	'ext': 'mp4',
ea706726	49	'title': 'The Jonathan Ross Show - Series 17 - Episode 8',
	50	'description': 'md5:3023dcdd375db1bc9967186cdb3f1399',
	51	'series': 'The Jonathan Ross Show',
	52	'episode_number': 8,
	53	'season_number': 17,
	54	'thumbnail': r're:https?://hubimages\.itv\.com/episode/2_1873_0002'
a71b8d3b RA	55	},
a71b8d3b RA	56	'params': {
29f7c58a	57	# m3u8 download
a71b8d3b RA	58	'skip_download': True,
a71b8d3b RA	59	},
3fae11ac S	60	}, {
	61	# unavailable via data-playlist-url
	62	'url': 'https://www.itv.com/hub/through-the-keyhole/2a2271a0033',
	63	'only_matching': True,
c18142da S	64	}, {
	65	# InvalidVodcrid
	66	'url': 'https://www.itv.com/hub/james-martins-saturday-morning/2a5159a0034',
	67	'only_matching': True,
	68	}, {
	69	# ContentUnavailable
	70	'url': 'https://www.itv.com/hub/whos-doing-the-dishes/2a2898a0024',
	71	'only_matching': True,
3fae11ac	72	}]
a71b8d3b	73
ea706726	74	def _generate_api_headers(self, hmac):
ea706726	75	return merge_dicts({
29f7c58a	76	'Accept': 'application/vnd.itv.vod.playlist.v2+json',
	77	'Content-Type': 'application/json',
	78	'hmac': hmac.upper(),
ea706726	79	}, self.geo_verification_headers())
	80
	81	def _call_api(self, video_id, playlist_url, headers, platform_tag, featureset, fatal=True):
	82	return self._download_json(
	83	playlist_url, video_id, data=json.dumps({
29f7c58a	84	'user': {
	85	'itvUserId': '',
	86	'entitlements': [],
	87	'token': ''
	88	},
	89	'device': {
	90	'manufacturer': 'Safari',
	91	'model': '5',
	92	'os': {
	93	'name': 'Windows NT',
	94	'version': '6.1',
	95	'type': 'desktop'
	96	}
	97	},
	98	'client': {
	99	'version': '4.1',
	100	'id': 'browser'
	101	},
	102	'variantAvailability': {
	103	'featureset': {
ea706726	104	'min': featureset,
ea706726	105	'max': featureset
29f7c58a	106	},
ea706726	107	'platformTag': platform_tag
29f7c58a	108	}
ea706726	109	}).encode(), headers=headers, fatal=fatal)
	110
	111	def _get_subtitles(self, video_id, variants, ios_playlist_url, headers, args, *kwargs):
	112	subtitles = {}
298bf1d2	113	# Prefer last matching featureset
298bf1d2	114	# See: https://github.com/yt-dlp/yt-dlp/issues/986
ea706726	115	platform_tag_subs, featureset_subs = next(
ea706726	116	((platform_tag, featureset)
bc8ab44e	117	for platform_tag, featuresets in reversed(list(variants.items())) for featureset in featuresets
ea706726	118	if try_get(featureset, lambda x: x[2]) == 'outband-webvtt'),
ea706726	119	(None, None))
298bf1d2	120
298bf1d2	121	if platform_tag_subs and featureset_subs:
ea706726	122	subs_playlist = self._call_api(
	123	video_id, ios_playlist_url, headers, platform_tag_subs, featureset_subs, fatal=False)
	124	subs = try_get(subs_playlist, lambda x: x['Playlist']['Video']['Subtitles'], list) or []
	125	for sub in subs:
	126	if not isinstance(sub, dict):
	127	continue
	128	href = url_or_none(sub.get('Href'))
	129	if not href:
	130	continue
	131	subtitles.setdefault('en', []).append({'url': href})
	132	return subtitles
	133
	134	def _real_extract(self, url):
	135	video_id = self._match_id(url)
	136	webpage = self._download_webpage(url, video_id)
	137	params = extract_attributes(self._search_regex(
	138	r'(?s)(<[^>]+id="video"[^>]*>)', webpage, 'params'))
	139	variants = self._parse_json(
	140	try_get(params, lambda x: x['data-video-variants'], compat_str) or '{}',
	141	video_id, fatal=False)
298bf1d2	142	# Prefer last matching featureset
298bf1d2	143	# See: https://github.com/yt-dlp/yt-dlp/issues/986
ea706726	144	platform_tag_video, featureset_video = next(
ea706726	145	((platform_tag, featureset)
bc8ab44e	146	for platform_tag, featuresets in reversed(list(variants.items())) for featureset in featuresets
da27aeea	147	if set(try_get(featureset, lambda x: x[:2]) or []) == {'aes', 'hls'}),
ea706726	148	(None, None))
	149	if not platform_tag_video or not featureset_video:
	150	raise ExtractorError('No downloads available', expected=True, video_id=video_id)
3fae11ac	151
ea706726	152	ios_playlist_url = params.get('data-video-playlist') or params['data-video-id']
	153	headers = self._generate_api_headers(params['data-video-hmac'])
	154	ios_playlist = self._call_api(
	155	video_id, ios_playlist_url, headers, platform_tag_video, featureset_video)
	156
	157	video_data = try_get(ios_playlist, lambda x: x['Playlist']['Video'], dict) or {}
	158	ios_base_url = video_data.get('Base')
3fae11ac	159	formats = []
29f7c58a	160	for media_file in (video_data.get('MediaFiles') or []):
	161	href = media_file.get('Href')
	162	if not href:
	163	continue
	164	if ios_base_url:
	165	href = ios_base_url + href
	166	ext = determine_ext(href)
	167	if ext == 'm3u8':
	168	formats.extend(self._extract_m3u8_formats(
	169	href, video_id, 'mp4', entry_protocol='m3u8_native',
	170	m3u8_id='hls', fatal=False))
30374f4d	171	else:
29f7c58a	172	formats.append({
29f7c58a	173	'url': href,
30374f4d	174	})
29f7c58a	175	info = self._search_json_ld(webpage, video_id, default={})
	176	if not info:
	177	json_ld = self._parse_json(self._search_regex(
	178	JSON_LD_RE, webpage, 'JSON-LD', '{}',
	179	group='json_ld'), video_id, fatal=False)
	180	if json_ld and json_ld.get('@type') == 'BreadcrumbList':
	181	for ile in (json_ld.get('itemListElement:') or []):
	182	item = ile.get('item:') or {}
	183	if item.get('@type') == 'TVEpisode':
	184	item['@context'] = 'http://schema.org'
	185	info = self._json_ld(item, video_id, fatal=False) or {}
	186	break
	187
ea706726	188	thumbnails = []
	189	thumbnail_url = try_get(params, lambda x: x['data-video-posterframe'], compat_str)
	190	if thumbnail_url:
	191	thumbnails.extend([{
	192	'url': thumbnail_url.format(width=1920, height=1080, quality=100, blur=0, bg='false'),
	193	'width': 1920,
	194	'height': 1080,
	195	}, {
	196	'url': urljoin(base_url(thumbnail_url), url_basename(thumbnail_url)),
	197	'preference': -2
	198	}])
	199
	200	thumbnail_url = self._html_search_meta(['og:image', 'twitter:image'], webpage, default=None)
	201	if thumbnail_url:
	202	thumbnails.append({
	203	'url': thumbnail_url,
	204	})
	205	self._remove_duplicate_formats(thumbnails)
	206
29f7c58a	207	return merge_dicts({
a71b8d3b	208	'id': video_id,
29f7c58a	209	'title': self._html_search_meta(['og:title', 'twitter:title'], webpage),
a71b8d3b	210	'formats': formats,
ea706726	211	'subtitles': self.extract_subtitles(video_id, variants, ios_playlist_url, headers),
29f7c58a	212	'duration': parse_duration(video_data.get('Duration')),
29f7c58a	213	'description': clean_html(get_element_by_class('episode-info__synopsis', webpage)),
ea706726	214	'thumbnails': thumbnails
29f7c58a	215	}, info)
ea1f5e5d S	216
	217
	218	class ITVBTCCIE(InfoExtractor):
3783b5f1	219	_VALID_URL = r'https?://(?:www\.)?itv\.com/(?:news\|btcc)/(?:[^/]+/)*(?P<id>[^/?#&]+)'
3783b5f1	220	_TESTS = [{
6857df60	221	'url': 'https://www.itv.com/btcc/articles/btcc-2019-brands-hatch-gp-race-action',
ea1f5e5d	222	'info_dict': {
6857df60 W	223	'id': 'btcc-2019-brands-hatch-gp-race-action',
6857df60 W	224	'title': 'BTCC 2019: Brands Hatch GP race action',
ea1f5e5d	225	},
85da4055	226	'playlist_count': 12,
3783b5f1	227	}, {
	228	'url': 'https://www.itv.com/news/2021-10-27/i-have-to-protect-the-country-says-rishi-sunak-as-uk-faces-interest-rate-hike',
	229	'info_dict': {
	230	'id': 'i-have-to-protect-the-country-says-rishi-sunak-as-uk-faces-interest-rate-hike',
	231	'title': 'md5:6ef054dd9f069330db3dcc66cb772d32'
	232	},
	233	'playlist_count': 4
	234	}]
	235	BRIGHTCOVE_URL_TEMPLATE = 'http://players.brightcove.net/%s/%s_default/index.html?videoId=%s'
ea1f5e5d S	236
	237	def _real_extract(self, url):
	238	playlist_id = self._match_id(url)
	239
	240	webpage = self._download_webpage(url, playlist_id)
	241
135dfa2c	242	json_map = try_get(
135dfa2c	243	self._search_nextjs_data(webpage, playlist_id),
6857df60 W	244	lambda x: x['props']['pageProps']['article']['body']['content']) or []
6857df60 W	245
3783b5f1	246	entries = []
6857df60	247	for video in json_map:
3783b5f1	248	if not any(video['data'].get(attr) == 'Brightcove' for attr in ('name', 'type')):
	249	continue
	250	video_id = video['data']['id']
	251	account_id = video['data']['accountId']
	252	player_id = video['data']['playerId']
	253	entries.append(self.url_result(
	254	smuggle_url(self.BRIGHTCOVE_URL_TEMPLATE % (account_id, player_id, video_id), {
ea1f5e5d S	255	# ITV does not like some GB IP ranges, so here are some
	256	# IP blocks it accepts
	257	'geo_ip_blocks': [
	258	'193.113.0.0/16', '54.36.162.0/23', '159.65.16.0/21'
	259	],
	260	'referrer': url,
	261	}),
3783b5f1	262	ie=BrightcoveNewIE.ie_key(), video_id=video_id))
ea1f5e5d S	263
	264	title = self._og_search_title(webpage, fatal=False)
	265
	266	return self.playlist_result(entries, playlist_id, title)