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

import re
import time

from .common import InfoExtractor
from ..utils import (
    ExtractorError,
    determine_ext,
    js_to_json,
    parse_qs,
    traverse_obj,
    urlencode_postdata,
)


class IPrimaIE(InfoExtractor):
    _VALID_URL = r'https?://(?!cnn)(?:[^/]+)\.iprima\.cz/(?:[^/]+/)*(?P<id>[^/?#&]+)'
    _GEO_BYPASS = False
    _NETRC_MACHINE = 'iprima'
    _AUTH_ROOT = 'https://auth.iprima.cz'
    access_token = None

    _TESTS = [{
        'url': 'https://prima.iprima.cz/particka/92-epizoda',
        'info_dict': {
            'id': 'p51388',
            'ext': 'mp4',
            'title': 'Partička (92)',
            'description': 'md5:859d53beae4609e6dd7796413f1b6cac',
            'upload_date': '20201103',
            'timestamp': 1604437480,
        },
        'params': {
            'skip_download': True,  # m3u8 download
        },
    }, {
        'url': 'http://play.iprima.cz/particka/particka-92',
        'only_matching': True,
    }, {
        # geo restricted
        'url': 'http://play.iprima.cz/closer-nove-pripady/closer-nove-pripady-iv-1',
        'only_matching': True,
    }, {
        'url': 'https://prima.iprima.cz/my-little-pony/mapa-znameni-2-2',
        'only_matching': True,
    }, {
        'url': 'https://prima.iprima.cz/porady/jak-se-stavi-sen/rodina-rathousova-praha',
        'only_matching': True,
    }, {
        'url': 'http://www.iprima.cz/filmy/desne-rande',
        'only_matching': True,
    }, {
        'url': 'https://zoom.iprima.cz/10-nejvetsich-tajemstvi-zahad/posvatna-mista-a-stavby',
        'only_matching': True,
    }, {
        'url': 'https://krimi.iprima.cz/mraz-0/sebevrazdy',
        'only_matching': True,
    }, {
        'url': 'https://cool.iprima.cz/derava-silnice-nevadi',
        'only_matching': True,
    }, {
        'url': 'https://love.iprima.cz/laska-az-za-hrob/slib-dany-bratrovi',
        'only_matching': True,
    }]

    def _perform_login(self, username, password):
        if self.access_token:
            return

        login_page = self._download_webpage(
            f'{self._AUTH_ROOT}/oauth2/login', None, note='Downloading login page',
            errnote='Downloading login page failed')

        login_form = self._hidden_inputs(login_page)

        login_form.update({
            '_email': username,
            '_password': password})

        profile_select_html, login_handle = self._download_webpage_handle(
            f'{self._AUTH_ROOT}/oauth2/login', None, data=urlencode_postdata(login_form),
            note='Logging in')

        # a profile may need to be selected first, even when there is only a single one
        if '/profile-select' in login_handle.url:
            profile_id = self._search_regex(
                r'data-identifier\s*=\s*["\']?(\w+)', profile_select_html, 'profile id')

            login_handle = self._request_webpage(
                f'{self._AUTH_ROOT}/user/profile-select-perform/{profile_id}', None,
                query={'continueUrl': '/user/login?redirect_uri=/user/'}, note='Selecting profile')

        code = traverse_obj(login_handle.url, ({parse_qs}, 'code', 0))
        if not code:
            raise ExtractorError('Login failed', expected=True)

        token_request_data = {
            'scope': 'openid+email+profile+phone+address+offline_access',
            'client_id': 'prima_sso',
            'grant_type': 'authorization_code',
            'code': code,
            'redirect_uri': f'{self._AUTH_ROOT}/sso/auth-check'}

        token_data = self._download_json(
            f'{self._AUTH_ROOT}/oauth2/token', None,
            note='Downloading token', errnote='Downloading token failed',
            data=urlencode_postdata(token_request_data))

        self.access_token = token_data.get('access_token')
        if self.access_token is None:
            raise ExtractorError('Getting token failed', expected=True)

    def _real_initialize(self):
        if not self.access_token:
            self.raise_login_required('Login is required to access any iPrima content', method='password')

    def _raise_access_error(self, error_code):
        if error_code == 'PLAY_GEOIP_DENIED':
            self.raise_geo_restricted(countries=['CZ'], metadata_available=True)
        elif error_code is not None:
            self.raise_no_formats('Access to stream infos forbidden', expected=True)

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

        webpage = self._download_webpage(url, video_id)

        title = self._html_extract_title(webpage) or self._html_search_meta(
            ['og:title', 'twitter:title'],
            webpage, 'title', default=None)

        video_id = self._search_regex((
            r'productId\s*=\s*([\'"])(?P<id>p\d+)\1',
            r'pproduct_id\s*=\s*([\'"])(?P<id>p\d+)\1',
        ), webpage, 'real id', group='id', default=None)

        if not video_id:
            nuxt_data = self._search_nuxt_data(webpage, video_id, traverse='data', fatal=False)
            video_id = traverse_obj(
                nuxt_data, (..., 'content', 'additionals', 'videoPlayId', {str}), get_all=False)

        if not video_id:
            nuxt_data = self._search_json(
                r'<script[^>]+\bid=["\']__NUXT_DATA__["\'][^>]*>',
                webpage, 'nuxt data', None, end_pattern=r'</script>', contains_pattern=r'\[(?s:.+)\]')

            video_id = traverse_obj(nuxt_data, lambda _, v: re.fullmatch(r'p\d+', v), get_all=False)

        if not video_id:
            self.raise_no_formats('Unable to extract video ID from webpage')

        metadata = self._download_json(
            f'https://api.play-backend.iprima.cz/api/v1//products/id-{video_id}/play',
            video_id, note='Getting manifest URLs', errnote='Failed to get manifest URLs',
            headers={'X-OTT-Access-Token': self.access_token},
            expected_status=403)

        self._raise_access_error(metadata.get('errorCode'))

        stream_infos = metadata.get('streamInfos')
        formats = []
        if stream_infos is None:
            self.raise_no_formats('Reading stream infos failed', expected=True)
        else:
            for manifest in stream_infos:
                manifest_type = manifest.get('type')
                manifest_url = manifest.get('url')
                ext = determine_ext(manifest_url)
                if manifest_type == 'HLS' or ext == 'm3u8':
                    formats += self._extract_m3u8_formats(
                        manifest_url, video_id, 'mp4', entry_protocol='m3u8_native',
                        m3u8_id='hls', fatal=False)
                elif manifest_type == 'DASH' or ext == 'mpd':
                    formats += self._extract_mpd_formats(
                        manifest_url, video_id, mpd_id='dash', fatal=False)

        final_result = self._search_json_ld(webpage, video_id, default={})
        final_result.update({
            'id': video_id,
            'title': title,
            'thumbnail': self._html_search_meta(
                ['thumbnail', 'og:image', 'twitter:image'],
                webpage, 'thumbnail', default=None),
            'formats': formats,
            'description': self._html_search_meta(
                ['description', 'og:description', 'twitter:description'],
                webpage, 'description', default=None)})

        return final_result


class IPrimaCNNIE(InfoExtractor):
    _VALID_URL = r'https?://cnn\.iprima\.cz/(?:[^/]+/)*(?P<id>[^/?#&]+)'
    _GEO_BYPASS = False

    _TESTS = [{
        'url': 'https://cnn.iprima.cz/porady/strunc/24072020-koronaviru-mam-plne-zuby-strasit-druhou-vlnou-je-absurdni-rika-senatorka-dernerova',
        'info_dict': {
            'id': 'p716177',
            'ext': 'mp4',
            'title': 'md5:277c6b1ed0577e51b40ddd35602ff43e',
        },
        'params': {
            'skip_download': 'm3u8'
        }
    }]

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

        self._set_cookie('play.iprima.cz', 'ott_adult_confirmed', '1')

        webpage = self._download_webpage(url, video_id)

        title = self._og_search_title(
            webpage, default=None) or self._search_regex(
            r'<h1>([^<]+)', webpage, 'title')

        video_id = self._search_regex(
            (r'<iframe[^>]+\bsrc=["\'](?:https?:)?//(?:api\.play-backend\.iprima\.cz/prehravac/embedded|prima\.iprima\.cz/[^/]+/[^/]+)\?.*?\bid=(p\d+)',
             r'data-product="([^"]+)">',
             r'id=["\']player-(p\d+)"',
             r'playerId\s*:\s*["\']player-(p\d+)',
             r'\bvideos\s*=\s*["\'](p\d+)'),
            webpage, 'real id')

        playerpage = self._download_webpage(
            'http://play.iprima.cz/prehravac/init',
            video_id, note='Downloading player', query={
                '_infuse': 1,
                '_ts': round(time.time()),
                'productId': video_id,
            }, headers={'Referer': url})

        formats = []

        def extract_formats(format_url, format_key=None, lang=None):
            ext = determine_ext(format_url)
            new_formats = []
            if format_key == 'hls' or ext == 'm3u8':
                new_formats = self._extract_m3u8_formats(
                    format_url, video_id, 'mp4', entry_protocol='m3u8_native',
                    m3u8_id='hls', fatal=False)
            elif format_key == 'dash' or ext == 'mpd':
                return
                new_formats = self._extract_mpd_formats(
                    format_url, video_id, mpd_id='dash', fatal=False)
            if lang:
                for f in new_formats:
                    if not f.get('language'):
                        f['language'] = lang
            formats.extend(new_formats)

        options = self._parse_json(
            self._search_regex(
                r'(?s)(?:TDIPlayerOptions|playerOptions)\s*=\s*({.+?});\s*\]\]',
                playerpage, 'player options', default='{}'),
            video_id, transform_source=js_to_json, fatal=False)
        if options:
            for key, tracks in options.get('tracks', {}).items():
                if not isinstance(tracks, list):
                    continue
                for track in tracks:
                    src = track.get('src')
                    if src:
                        extract_formats(src, key.lower(), track.get('lang'))

        if not formats:
            for _, src in re.findall(r'src["\']\s*:\s*(["\'])(.+?)\1', playerpage):
                extract_formats(src)

        if not formats and '>GEO_IP_NOT_ALLOWED<' in playerpage:
            self.raise_geo_restricted(countries=['CZ'], metadata_available=True)

        return {
            'id': video_id,
            'title': title,
            'thumbnail': self._og_search_thumbnail(webpage, default=None),
            'formats': formats,
            'description': self._og_search_description(webpage, default=None),
        }
Commit	Line	Data
7881a644	1	import re
f406c787	2	import time
7881a644	3
7881a644	4	from .common import InfoExtractor
1cc79574	5	from ..utils import (
e897bd82	6	ExtractorError,
369e7e3f S	7	determine_ext,
369e7e3f S	8	js_to_json,
9fddc12a	9	parse_qs,
e897bd82 SS	10	traverse_obj,
e897bd82 SS	11	urlencode_postdata,
82642235	12	)
7881a644	13
	14
	15	class IPrimaIE(InfoExtractor):
e1b7c54d	16	_VALID_URL = r'https?://(?!cnn)(?:[^/]+)\.iprima\.cz/(?:[^/]+/)*(?P<id>[^/?#&]+)'
da42ff06	17	_GEO_BYPASS = False
e1b7c54d	18	_NETRC_MACHINE = 'iprima'
9fddc12a	19	_AUTH_ROOT = 'https://auth.iprima.cz'
e1b7c54d	20	access_token = None
7881a644	21
7881a644	22	_TESTS = [{
30fa5c60	23	'url': 'https://prima.iprima.cz/particka/92-epizoda',
7881a644	24	'info_dict': {
30fa5c60	25	'id': 'p51388',
f406c787	26	'ext': 'mp4',
30fa5c60 S	27	'title': 'Partička (92)',
30fa5c60 S	28	'description': 'md5:859d53beae4609e6dd7796413f1b6cac',
e1b7c54d	29	'upload_date': '20201103',
e1b7c54d	30	'timestamp': 1604437480,
7881a644	31	},
7881a644	32	'params': {
f406c787	33	'skip_download': True, # m3u8 download
7881a644	34	},
973f2532	35	}, {
f406c787	36	'url': 'http://play.iprima.cz/particka/particka-92',
bc03e585	37	'only_matching': True,
da42ff06 S	38	}, {
	39	# geo restricted
	40	'url': 'http://play.iprima.cz/closer-nove-pripady/closer-nove-pripady-iv-1',
	41	'only_matching': True,
a2637a2d	42	}, {
a2637a2d S	43	'url': 'https://prima.iprima.cz/my-little-pony/mapa-znameni-2-2',
	44	'only_matching': True,
	45	}, {
a2637a2d S	46	'url': 'https://prima.iprima.cz/porady/jak-se-stavi-sen/rodina-rathousova-praha',
a2637a2d S	47	'only_matching': True,
9235b509 S	48	}, {
	49	'url': 'http://www.iprima.cz/filmy/desne-rande',
	50	'only_matching': True,
90046d77	51	}, {
	52	'url': 'https://zoom.iprima.cz/10-nejvetsich-tajemstvi-zahad/posvatna-mista-a-stavby',
	53	'only_matching': True,
	54	}, {
	55	'url': 'https://krimi.iprima.cz/mraz-0/sebevrazdy',
	56	'only_matching': True,
	57	}, {
	58	'url': 'https://cool.iprima.cz/derava-silnice-nevadi',
	59	'only_matching': True,
	60	}, {
	61	'url': 'https://love.iprima.cz/laska-az-za-hrob/slib-dany-bratrovi',
	62	'only_matching': True,
e1b7c54d	63	}]
e1b7c54d	64
52efa4b3	65	def _perform_login(self, username, password):
	66	if self.access_token:
	67	return
e1b7c54d	68
e1b7c54d	69	login_page = self._download_webpage(
9fddc12a	70	f'{self._AUTH_ROOT}/oauth2/login', None, note='Downloading login page',
e1b7c54d	71	errnote='Downloading login page failed')
	72
	73	login_form = self._hidden_inputs(login_page)
	74
	75	login_form.update({
	76	'_email': username,
	77	'_password': password})
	78
9fddc12a	79	profile_select_html, login_handle = self._download_webpage_handle(
9fddc12a	80	f'{self._AUTH_ROOT}/oauth2/login', None, data=urlencode_postdata(login_form),
e1b7c54d	81	note='Logging in')
e1b7c54d	82
9fddc12a	83	# a profile may need to be selected first, even when there is only a single one
3d2623a8	84	if '/profile-select' in login_handle.url:
9fddc12a	85	profile_id = self._search_regex(
	86	r'data-identifier\s=\s["\']?(\w+)', profile_select_html, 'profile id')
	87
	88	login_handle = self._request_webpage(
	89	f'{self._AUTH_ROOT}/user/profile-select-perform/{profile_id}', None,
	90	query={'continueUrl': '/user/login?redirect_uri=/user/'}, note='Selecting profile')
	91
3d2623a8	92	code = traverse_obj(login_handle.url, ({parse_qs}, 'code', 0))
e1b7c54d	93	if not code:
	94	raise ExtractorError('Login failed', expected=True)
	95
	96	token_request_data = {
	97	'scope': 'openid+email+profile+phone+address+offline_access',
	98	'client_id': 'prima_sso',
	99	'grant_type': 'authorization_code',
	100	'code': code,
9fddc12a	101	'redirect_uri': f'{self._AUTH_ROOT}/sso/auth-check'}
e1b7c54d	102
e1b7c54d	103	token_data = self._download_json(
9fddc12a	104	f'{self._AUTH_ROOT}/oauth2/token', None,
e1b7c54d	105	note='Downloading token', errnote='Downloading token failed',
	106	data=urlencode_postdata(token_request_data))
	107
	108	self.access_token = token_data.get('access_token')
	109	if self.access_token is None:
	110	raise ExtractorError('Getting token failed', expected=True)
	111
52efa4b3	112	def _real_initialize(self):
	113	if not self.access_token:
	114	self.raise_login_required('Login is required to access any iPrima content', method='password')
	115
e1b7c54d	116	def _raise_access_error(self, error_code):
	117	if error_code == 'PLAY_GEOIP_DENIED':
	118	self.raise_geo_restricted(countries=['CZ'], metadata_available=True)
	119	elif error_code is not None:
	120	self.raise_no_formats('Access to stream infos forbidden', expected=True)
	121
e1b7c54d	122	def _real_extract(self, url):
	123	video_id = self._match_id(url)
	124
	125	webpage = self._download_webpage(url, video_id)
	126
9fddc12a	127	title = self._html_extract_title(webpage) or self._html_search_meta(
e1b7c54d	128	['og:title', 'twitter:title'],
	129	webpage, 'title', default=None)
	130
	131	video_id = self._search_regex((
	132	r'productId\s=\s([\'"])(?P<id>p\d+)\1',
9fddc12a	133	r'pproduct_id\s=\s([\'"])(?P<id>p\d+)\1',
	134	), webpage, 'real id', group='id', default=None)
	135
	136	if not video_id:
568f0805	137	nuxt_data = self._search_nuxt_data(webpage, video_id, traverse='data', fatal=False)
9fddc12a	138	video_id = traverse_obj(
	139	nuxt_data, (..., 'content', 'additionals', 'videoPlayId', {str}), get_all=False)
	140
568f0805	141	if not video_id:
	142	nuxt_data = self._search_json(
	143	r'<script[^>]+\bid=["\']__NUXT_DATA__["\'][^>]*>',
	144	webpage, 'nuxt data', None, end_pattern=r'</script>', contains_pattern=r'\[(?s:.+)\]')
	145
	146	video_id = traverse_obj(nuxt_data, lambda _, v: re.fullmatch(r'p\d+', v), get_all=False)
	147
9fddc12a	148	if not video_id:
9fddc12a	149	self.raise_no_formats('Unable to extract video ID from webpage')
e1b7c54d	150
	151	metadata = self._download_json(
	152	f'https://api.play-backend.iprima.cz/api/v1//products/id-{video_id}/play',
	153	video_id, note='Getting manifest URLs', errnote='Failed to get manifest URLs',
	154	headers={'X-OTT-Access-Token': self.access_token},
	155	expected_status=403)
	156
	157	self._raise_access_error(metadata.get('errorCode'))
	158
	159	stream_infos = metadata.get('streamInfos')
	160	formats = []
	161	if stream_infos is None:
	162	self.raise_no_formats('Reading stream infos failed', expected=True)
	163	else:
	164	for manifest in stream_infos:
	165	manifest_type = manifest.get('type')
	166	manifest_url = manifest.get('url')
	167	ext = determine_ext(manifest_url)
	168	if manifest_type == 'HLS' or ext == 'm3u8':
	169	formats += self._extract_m3u8_formats(
	170	manifest_url, video_id, 'mp4', entry_protocol='m3u8_native',
	171	m3u8_id='hls', fatal=False)
	172	elif manifest_type == 'DASH' or ext == 'mpd':
	173	formats += self._extract_mpd_formats(
	174	manifest_url, video_id, mpd_id='dash', fatal=False)
e1b7c54d	175
2530b68d	176	final_result = self._search_json_ld(webpage, video_id, default={})
e1b7c54d	177	final_result.update({
	178	'id': video_id,
	179	'title': title,
	180	'thumbnail': self._html_search_meta(
	181	['thumbnail', 'og:image', 'twitter:image'],
	182	webpage, 'thumbnail', default=None),
	183	'formats': formats,
	184	'description': self._html_search_meta(
	185	['description', 'og:description', 'twitter:description'],
	186	webpage, 'description', default=None)})
	187
	188	return final_result
	189
	190
	191	class IPrimaCNNIE(InfoExtractor):
	192	_VALID_URL = r'https?://cnn\.iprima\.cz/(?:[^/]+/)*(?P<id>[^/?#&]+)'
	193	_GEO_BYPASS = False
	194
	195	_TESTS = [{
	196	'url': 'https://cnn.iprima.cz/porady/strunc/24072020-koronaviru-mam-plne-zuby-strasit-druhou-vlnou-je-absurdni-rika-senatorka-dernerova',
	197	'info_dict': {
	198	'id': 'p716177',
	199	'ext': 'mp4',
	200	'title': 'md5:277c6b1ed0577e51b40ddd35602ff43e',
	201	},
	202	'params': {
	203	'skip_download': 'm3u8'
	204	}
973f2532	205	}]
7881a644	206
7881a644	207	def _real_extract(self, url):
369e7e3f	208	video_id = self._match_id(url)
7881a644	209
09322ccc S	210	self._set_cookie('play.iprima.cz', 'ott_adult_confirmed', '1')
09322ccc S	211
7881a644	212	webpage = self._download_webpage(url, video_id)
7881a644	213
30fa5c60 S	214	title = self._og_search_title(
	215	webpage, default=None) or self._search_regex(
	216	r'<h1>([^<]+)', webpage, 'title')
	217
a2637a2d S	218	video_id = self._search_regex(
a2637a2d S	219	(r'<iframe[^>]+\bsrc=["\'](?:https?:)?//(?:api\.play-backend\.iprima\.cz/prehravac/embedded\|prima\.iprima\.cz/[^/]+/[^/]+)\?.*?\bid=(p\d+)',
30fa5c60 S	220	r'data-product="([^"]+)">',
30fa5c60 S	221	r'id=["\']player-(p\d+)"',
b7770046 U	222	r'playerId\s:\s["\']player-(p\d+)',
b7770046 U	223	r'\bvideos\s=\s["\'](p\d+)'),
a2637a2d	224	webpage, 'real id')
82642235	225
82f66218 S	226	playerpage = self._download_webpage(
	227	'http://play.iprima.cz/prehravac/init',
	228	video_id, note='Downloading player', query={
	229	'_infuse': 1,
	230	'_ts': round(time.time()),
	231	'productId': video_id,
	232	}, headers={'Referer': url})
7881a644	233
369e7e3f	234	formats = []
7881a644	235
369e7e3f S	236	def extract_formats(format_url, format_key=None, lang=None):
	237	ext = determine_ext(format_url)
	238	new_formats = []
	239	if format_key == 'hls' or ext == 'm3u8':
	240	new_formats = self._extract_m3u8_formats(
	241	format_url, video_id, 'mp4', entry_protocol='m3u8_native',
	242	m3u8_id='hls', fatal=False)
	243	elif format_key == 'dash' or ext == 'mpd':
	244	return
	245	new_formats = self._extract_mpd_formats(
	246	format_url, video_id, mpd_id='dash', fatal=False)
	247	if lang:
	248	for f in new_formats:
	249	if not f.get('language'):
	250	f['language'] = lang
	251	formats.extend(new_formats)
	252
	253	options = self._parse_json(
	254	self._search_regex(
0bbcc8a1	255	r'(?s)(?:TDIPlayerOptions\|playerOptions)\s=\s({.+?});\s*\]\]',
369e7e3f S	256	playerpage, 'player options', default='{}'),
	257	video_id, transform_source=js_to_json, fatal=False)
	258	if options:
	259	for key, tracks in options.get('tracks', {}).items():
	260	if not isinstance(tracks, list):
	261	continue
	262	for track in tracks:
	263	src = track.get('src')
	264	if src:
	265	extract_formats(src, key.lower(), track.get('lang'))
	266
	267	if not formats:
	268	for _, src in re.findall(r'src["\']\s:\s(["\'])(.+?)\1', playerpage):
	269	extract_formats(src)
91264ce5	270
3c6b3bf2	271	if not formats and '>GEO_IP_NOT_ALLOWED<' in playerpage:
b7da73eb	272	self.raise_geo_restricted(countries=['CZ'], metadata_available=True)
3c6b3bf2	273
7881a644	274	return {
f406c787	275	'id': video_id,
30fa5c60 S	276	'title': title,
30fa5c60 S	277	'thumbnail': self._og_search_thumbnail(webpage, default=None),
7881a644	278	'formats': formats,
30fa5c60	279	'description': self._og_search_description(webpage, default=None),
91264ce5	280	}