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

import json
import re

from .common import InfoExtractor
from ..dependencies import Cryptodome
from ..utils import ExtractorError, int_or_none, qualities


class IviIE(InfoExtractor):
    IE_DESC = 'ivi.ru'
    IE_NAME = 'ivi'
    _VALID_URL = r'https?://(?:www\.)?ivi\.(?:ru|tv)/(?:watch/(?:[^/]+/)?|video/player\?.*?videoId=)(?P<id>\d+)'
    _EMBED_REGEX = [r'<embed[^>]+?src=(["\'])(?P<url>https?://(?:www\.)?ivi\.ru/video/player.+?)\1']
    _GEO_BYPASS = False
    _GEO_COUNTRIES = ['RU']
    _LIGHT_KEY = b'\xf1\x02\x32\xb7\xbc\x5c\x7a\xe8\xf7\x96\xc1\x33\x2b\x27\xa1\x8c'
    _LIGHT_URL = 'https://api.ivi.ru/light/'

    _TESTS = [
        # Single movie
        {
            'url': 'http://www.ivi.ru/watch/53141',
            'md5': '6ff5be2254e796ed346251d117196cf4',
            'info_dict': {
                'id': '53141',
                'ext': 'mp4',
                'title': 'Иван Васильевич меняет профессию',
                'description': 'md5:b924063ea1677c8fe343d8a72ac2195f',
                'duration': 5498,
                'thumbnail': r're:^https?://.*\.jpg$',
            },
            'skip': 'Only works from Russia',
        },
        # Serial's series
        {
            'url': 'http://www.ivi.ru/watch/dvoe_iz_lartsa/9549',
            'md5': '221f56b35e3ed815fde2df71032f4b3e',
            'info_dict': {
                'id': '9549',
                'ext': 'mp4',
                'title': 'Двое из ларца - Дело Гольдберга (1 часть)',
                'series': 'Двое из ларца',
                'season': 'Сезон 1',
                'season_number': 1,
                'episode': 'Дело Гольдберга (1 часть)',
                'episode_number': 1,
                'duration': 2655,
                'thumbnail': r're:^https?://.*\.jpg$',
            },
            'skip': 'Only works from Russia',
        },
        {
            # with MP4-HD720 format
            'url': 'http://www.ivi.ru/watch/146500',
            'md5': 'd63d35cdbfa1ea61a5eafec7cc523e1e',
            'info_dict': {
                'id': '146500',
                'ext': 'mp4',
                'title': 'Кукла',
                'description': 'md5:ffca9372399976a2d260a407cc74cce6',
                'duration': 5599,
                'thumbnail': r're:^https?://.*\.jpg$',
            },
            'skip': 'Only works from Russia',
        },
        {
            'url': 'https://www.ivi.tv/watch/33560/',
            'only_matching': True,
        },
    ]

    # Sorted by quality
    _KNOWN_FORMATS = (
        'MP4-low-mobile', 'MP4-mobile', 'FLV-lo', 'MP4-lo', 'FLV-hi', 'MP4-hi',
        'MP4-SHQ', 'MP4-HD720', 'MP4-HD1080')

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

        data = json.dumps({
            'method': 'da.content.get',
            'params': [
                video_id, {
                    'site': 's%d',
                    'referrer': f'http://www.ivi.ru/watch/{video_id}',
                    'contentid': video_id,
                },
            ],
        })

        for site in (353, 183):
            content_data = (data % site).encode()
            if site == 353:
                if not Cryptodome.CMAC:
                    continue

                timestamp = (self._download_json(
                    self._LIGHT_URL, video_id,
                    'Downloading timestamp JSON', data=json.dumps({
                        'method': 'da.timestamp.get',
                        'params': [],
                    }).encode(), fatal=False) or {}).get('result')
                if not timestamp:
                    continue

                query = {
                    'ts': timestamp,
                    'sign': Cryptodome.CMAC.new(self._LIGHT_KEY, timestamp.encode() + content_data,
                                                Cryptodome.Blowfish).hexdigest(),
                }
            else:
                query = {}

            video_json = self._download_json(
                self._LIGHT_URL, video_id,
                'Downloading video JSON', data=content_data, query=query)

            error = video_json.get('error')
            if error:
                origin = error.get('origin')
                message = error.get('message') or error.get('user_message')
                extractor_msg = 'Unable to download video %s'
                if origin == 'NotAllowedForLocation':
                    self.raise_geo_restricted(message, self._GEO_COUNTRIES)
                elif origin == 'NoRedisValidData':
                    extractor_msg = 'Video %s does not exist'
                elif site == 353:
                    continue
                elif not Cryptodome.CMAC:
                    raise ExtractorError('pycryptodomex not found. Please install', expected=True)
                elif message:
                    extractor_msg += ': ' + message
                raise ExtractorError(extractor_msg % video_id, expected=True)
            else:
                break

        result = video_json['result']
        title = result['title']

        quality = qualities(self._KNOWN_FORMATS)

        formats = []
        for f in result.get('files', []):
            f_url = f.get('url')
            content_format = f.get('content_format')
            if not f_url:
                continue
            if (not self.get_param('allow_unplayable_formats')
                    and ('-MDRM-' in content_format or '-FPS-' in content_format)):
                continue
            formats.append({
                'url': f_url,
                'format_id': content_format,
                'quality': quality(content_format),
                'filesize': int_or_none(f.get('size_in_bytes')),
            })

        compilation = result.get('compilation')
        episode = title if compilation else None

        title = f'{compilation} - {title}' if compilation is not None else title

        thumbnails = [{
            'url': preview['url'],
            'id': preview.get('content_format'),
        } for preview in result.get('preview', []) if preview.get('url')]

        webpage = self._download_webpage(url, video_id)

        season = self._search_regex(
            r'<li[^>]+class="season active"[^>]*><a[^>]+>([^<]+)',
            webpage, 'season', default=None)
        season_number = int_or_none(self._search_regex(
            r'<li[^>]+class="season active"[^>]*><a[^>]+data-season(?:-index)?="(\d+)"',
            webpage, 'season number', default=None))

        episode_number = int_or_none(self._search_regex(
            r'[^>]+itemprop="episode"[^>]*>\s*<meta[^>]+itemprop="episodeNumber"[^>]+content="(\d+)',
            webpage, 'episode number', default=None))

        description = self._og_search_description(webpage, default=None) or self._html_search_meta(
            'description', webpage, 'description', default=None)

        return {
            'id': video_id,
            'title': title,
            'series': compilation,
            'season': season,
            'season_number': season_number,
            'episode': episode,
            'episode_number': episode_number,
            'thumbnails': thumbnails,
            'description': description,
            'duration': int_or_none(result.get('duration')),
            'formats': formats,
        }


class IviCompilationIE(InfoExtractor):
    IE_DESC = 'ivi.ru compilations'
    IE_NAME = 'ivi:compilation'
    _VALID_URL = r'https?://(?:www\.)?ivi\.ru/watch/(?!\d+)(?P<compilationid>[a-z\d_-]+)(?:/season(?P<seasonid>\d+))?$'
    _TESTS = [{
        'url': 'http://www.ivi.ru/watch/dvoe_iz_lartsa',
        'info_dict': {
            'id': 'dvoe_iz_lartsa',
            'title': 'Двое из ларца (2006 - 2008)',
        },
        'playlist_mincount': 24,
    }, {
        'url': 'http://www.ivi.ru/watch/dvoe_iz_lartsa/season1',
        'info_dict': {
            'id': 'dvoe_iz_lartsa/season1',
            'title': 'Двое из ларца (2006 - 2008) 1 сезон',
        },
        'playlist_mincount': 12,
    }]

    def _extract_entries(self, html, compilation_id):
        return [
            self.url_result(
                f'http://www.ivi.ru/watch/{compilation_id}/{serie}', IviIE.ie_key())
            for serie in re.findall(
                rf'<a\b[^>]+\bhref=["\']/watch/{compilation_id}/(\d+)["\']', html)]

    def _real_extract(self, url):
        mobj = self._match_valid_url(url)
        compilation_id = mobj.group('compilationid')
        season_id = mobj.group('seasonid')

        if season_id is not None:  # Season link
            season_page = self._download_webpage(
                url, compilation_id, f'Downloading season {season_id} web page')
            playlist_id = f'{compilation_id}/season{season_id}'
            playlist_title = self._html_search_meta('title', season_page, 'title')
            entries = self._extract_entries(season_page, compilation_id)
        else:  # Compilation link
            compilation_page = self._download_webpage(url, compilation_id, 'Downloading compilation web page')
            playlist_id = compilation_id
            playlist_title = self._html_search_meta('title', compilation_page, 'title')
            seasons = re.findall(
                rf'<a href="/watch/{compilation_id}/season(\d+)', compilation_page)
            if not seasons:  # No seasons in this compilation
                entries = self._extract_entries(compilation_page, compilation_id)
            else:
                entries = []
                for season_id in seasons:
                    season_page = self._download_webpage(
                        f'http://www.ivi.ru/watch/{compilation_id}/season{season_id}',
                        compilation_id, f'Downloading season {season_id} web page')
                    entries.extend(self._extract_entries(season_page, compilation_id))

        return self.playlist_result(entries, playlist_id, playlist_title)
Commit	Line	Data
77aa6b32	1	import json
80a51fc2	2	import re
77aa6b32	3
77aa6b32	4	from .common import InfoExtractor
f6a765ce	5	from ..dependencies import Cryptodome
f6a765ce	6	from ..utils import ExtractorError, int_or_none, qualities
77aa6b32	7
	8
	9	class IviIE(InfoExtractor):
ceb2b7d2	10	IE_DESC = 'ivi.ru'
ceb2b7d2	11	IE_NAME = 'ivi'
022218f2	12	_VALID_URL = r'https?://(?:www\.)?ivi\.(?:ru\|tv)/(?:watch/(?:[^/]+/)?\|video/player\?.*?videoId=)(?P<id>\d+)'
bfd973ec	13	_EMBED_REGEX = [r'<embed[^>]+?src=(["\'])(?P<url>https?://(?:www\.)?ivi\.ru/video/player.+?)\1']
42dcdbe1 S	14	_GEO_BYPASS = False
42dcdbe1 S	15	_GEO_COUNTRIES = ['RU']
656c2001 RA	16	_LIGHT_KEY = b'\xf1\x02\x32\xb7\xbc\x5c\x7a\xe8\xf7\x96\xc1\x33\x2b\x27\xa1\x8c'
656c2001 RA	17	_LIGHT_URL = 'https://api.ivi.ru/light/'
77aa6b32	18
	19	_TESTS = [
	20	# Single movie
	21	{
ceb2b7d2	22	'url': 'http://www.ivi.ru/watch/53141',
ceb2b7d2	23	'md5': '6ff5be2254e796ed346251d117196cf4',
ceb2b7d2	24	'info_dict': {
84dd7031 S	25	'id': '53141',
84dd7031 S	26	'ext': 'mp4',
ceb2b7d2	27	'title': 'Иван Васильевич меняет профессию',
	28	'description': 'md5:b924063ea1677c8fe343d8a72ac2195f',
	29	'duration': 5498,
ec85ded8	30	'thumbnail': r're:^https?://.*\.jpg$',
77aa6b32	31	},
ceb2b7d2	32	'skip': 'Only works from Russia',
77aa6b32	33	},
dfb1b146	34	# Serial's series
77aa6b32	35	{
6ebb46c1 S	36	'url': 'http://www.ivi.ru/watch/dvoe_iz_lartsa/9549',
6ebb46c1 S	37	'md5': '221f56b35e3ed815fde2df71032f4b3e',
ceb2b7d2	38	'info_dict': {
6ebb46c1	39	'id': '9549',
84dd7031	40	'ext': 'mp4',
ab3176af S	41	'title': 'Двое из ларца - Дело Гольдберга (1 часть)',
ab3176af S	42	'series': 'Двое из ларца',
1463c5b9 S	43	'season': 'Сезон 1',
1463c5b9 S	44	'season_number': 1,
ab3176af S	45	'episode': 'Дело Гольдберга (1 часть)',
ab3176af S	46	'episode_number': 1,
6ebb46c1	47	'duration': 2655,
ec85ded8	48	'thumbnail': r're:^https?://.*\.jpg$',
77aa6b32	49	},
ceb2b7d2	50	'skip': 'Only works from Russia',
cf143c4d S	51	},
	52	{
	53	# with MP4-HD720 format
	54	'url': 'http://www.ivi.ru/watch/146500',
	55	'md5': 'd63d35cdbfa1ea61a5eafec7cc523e1e',
	56	'info_dict': {
	57	'id': '146500',
	58	'ext': 'mp4',
	59	'title': 'Кукла',
	60	'description': 'md5:ffca9372399976a2d260a407cc74cce6',
	61	'duration': 5599,
ec85ded8	62	'thumbnail': r're:^https?://.*\.jpg$',
cf143c4d S	63	},
cf143c4d S	64	'skip': 'Only works from Russia',
022218f2 S	65	},
	66	{
	67	'url': 'https://www.ivi.tv/watch/33560/',
	68	'only_matching': True,
	69	},
77aa6b32	70	]
ceb2b7d2	71
77aa6b32	72	# Sorted by quality
cf143c4d S	73	_KNOWN_FORMATS = (
	74	'MP4-low-mobile', 'MP4-mobile', 'FLV-lo', 'MP4-lo', 'FLV-hi', 'MP4-hi',
	75	'MP4-SHQ', 'MP4-HD720', 'MP4-HD1080')
77aa6b32	76
77aa6b32	77	def _real_extract(self, url):
63be3b89	78	video_id = self._match_id(url)
77aa6b32	79
656c2001	80	data = json.dumps({
63be3b89 S	81	'method': 'da.content.get',
	82	'params': [
	83	video_id, {
1bba88ef	84	'site': 's%d',
add96eb9	85	'referrer': f'http://www.ivi.ru/watch/{video_id}',
	86	'contentid': video_id,
	87	},
	88	],
f8015c15	89	})
77aa6b32	90
76d9eca4	91	for site in (353, 183):
f8015c15	92	content_data = (data % site).encode()
76d9eca4	93	if site == 353:
65f6e807	94	if not Cryptodome.CMAC:
f6a765ce	95	continue
76d9eca4 RA	96
	97	timestamp = (self._download_json(
	98	self._LIGHT_URL, video_id,
	99	'Downloading timestamp JSON', data=json.dumps({
	100	'method': 'da.timestamp.get',
add96eb9	101	'params': [],
76d9eca4 RA	102	}).encode(), fatal=False) or {}).get('result')
	103	if not timestamp:
	104	continue
	105
	106	query = {
	107	'ts': timestamp,
65f6e807	108	'sign': Cryptodome.CMAC.new(self._LIGHT_KEY, timestamp.encode() + content_data,
65f6e807	109	Cryptodome.Blowfish).hexdigest(),
76d9eca4 RA	110	}
	111	else:
	112	query = {}
1bba88ef	113
76d9eca4	114	video_json = self._download_json(
1bba88ef	115	self._LIGHT_URL, video_id,
76d9eca4 RA	116	'Downloading video JSON', data=content_data, query=query)
	117
	118	error = video_json.get('error')
	119	if error:
	120	origin = error.get('origin')
	121	message = error.get('message') or error.get('user_message')
	122	extractor_msg = 'Unable to download video %s'
	123	if origin == 'NotAllowedForLocation':
	124	self.raise_geo_restricted(message, self._GEO_COUNTRIES)
	125	elif origin == 'NoRedisValidData':
	126	extractor_msg = 'Video %s does not exist'
	127	elif site == 353:
	128	continue
65f6e807	129	elif not Cryptodome.CMAC:
49e7e9c3	130	raise ExtractorError('pycryptodomex not found. Please install', expected=True)
76d9eca4 RA	131	elif message:
	132	extractor_msg += ': ' + message
	133	raise ExtractorError(extractor_msg % video_id, expected=True)
	134	else:
	135	break
77aa6b32	136
ceb2b7d2	137	result = video_json['result']
656c2001	138	title = result['title']
77aa6b32	139
cf143c4d S	140	quality = qualities(self._KNOWN_FORMATS)
cf143c4d S	141
656c2001 RA	142	formats = []
	143	for f in result.get('files', []):
	144	f_url = f.get('url')
	145	content_format = f.get('content_format')
06869367	146	if not f_url:
06869367	147	continue
a06916d9	148	if (not self.get_param('allow_unplayable_formats')
06869367	149	and ('-MDRM-' in content_format or '-FPS-' in content_format)):
656c2001 RA	150	continue
	151	formats.append({
	152	'url': f_url,
	153	'format_id': content_format,
	154	'quality': quality(content_format),
	155	'filesize': int_or_none(f.get('size_in_bytes')),
	156	})
bf5b0a1b	157
ab3176af S	158	compilation = result.get('compilation')
	159	episode = title if compilation else None
	160
add96eb9	161	title = f'{compilation} - {title}' if compilation is not None else title
77aa6b32	162
ab3176af S	163	thumbnails = [{
	164	'url': preview['url'],
	165	'id': preview.get('content_format'),
	166	} for preview in result.get('preview', []) if preview.get('url')]
	167
	168	webpage = self._download_webpage(url, video_id)
	169
1463c5b9 S	170	season = self._search_regex(
	171	r'<li[^>]+class="season active"[^>]*><a[^>]+>([^<]+)',
	172	webpage, 'season', default=None)
	173	season_number = int_or_none(self._search_regex(
	174	r'<li[^>]+class="season active"[^>]*><a[^>]+data-season(?:-index)?="(\d+)"',
	175	webpage, 'season number', default=None))
	176
ab3176af	177	episode_number = int_or_none(self._search_regex(
3d897cc7	178	r'[^>]+itemprop="episode"[^>]>\s<meta[^>]+itemprop="episodeNumber"[^>]+content="(\d+)',
ab3176af	179	webpage, 'episode number', default=None))
77aa6b32	180
ab3176af S	181	description = self._og_search_description(webpage, default=None) or self._html_search_meta(
ab3176af S	182	'description', webpage, 'description', default=None)
77aa6b32	183
	184	return {
	185	'id': video_id,
	186	'title': title,
ab3176af	187	'series': compilation,
1463c5b9 S	188	'season': season,
1463c5b9 S	189	'season_number': season_number,
ab3176af S	190	'episode': episode,
	191	'episode_number': episode_number,
	192	'thumbnails': thumbnails,
77aa6b32	193	'description': description,
656c2001	194	'duration': int_or_none(result.get('duration')),
77aa6b32	195	'formats': formats,
	196	}
	197
	198
	199	class IviCompilationIE(InfoExtractor):
ceb2b7d2	200	IE_DESC = 'ivi.ru compilations'
ceb2b7d2	201	IE_NAME = 'ivi:compilation'
84dd7031	202	_VALID_URL = r'https?://(?:www\.)?ivi\.ru/watch/(?!\d+)(?P<compilationid>[a-z\d_-]+)(?:/season(?P<seasonid>\d+))?$'
22a6f150 PH	203	_TESTS = [{
	204	'url': 'http://www.ivi.ru/watch/dvoe_iz_lartsa',
	205	'info_dict': {
	206	'id': 'dvoe_iz_lartsa',
	207	'title': 'Двое из ларца (2006 - 2008)',
	208	},
	209	'playlist_mincount': 24,
	210	}, {
	211	'url': 'http://www.ivi.ru/watch/dvoe_iz_lartsa/season1',
	212	'info_dict': {
	213	'id': 'dvoe_iz_lartsa/season1',
	214	'title': 'Двое из ларца (2006 - 2008) 1 сезон',
	215	},
	216	'playlist_mincount': 12,
	217	}]
77aa6b32	218
77aa6b32	219	def _extract_entries(self, html, compilation_id):
c6270b2e S	220	return [
c6270b2e S	221	self.url_result(
add96eb9	222	f'http://www.ivi.ru/watch/{compilation_id}/{serie}', IviIE.ie_key())
c6270b2e	223	for serie in re.findall(
add96eb9	224	rf'<a\b[^>]+\bhref=["\']/watch/{compilation_id}/(\d+)["\']', html)]
77aa6b32	225
77aa6b32	226	def _real_extract(self, url):
5ad28e7f	227	mobj = self._match_valid_url(url)
77aa6b32	228	compilation_id = mobj.group('compilationid')
	229	season_id = mobj.group('seasonid')
	230
5f6a1245	231	if season_id is not None: # Season link
c6270b2e	232	season_page = self._download_webpage(
add96eb9	233	url, compilation_id, f'Downloading season {season_id} web page')
add96eb9	234	playlist_id = f'{compilation_id}/season{season_id}'
ceb2b7d2	235	playlist_title = self._html_search_meta('title', season_page, 'title')
77aa6b32	236	entries = self._extract_entries(season_page, compilation_id)
5f6a1245	237	else: # Compilation link
ceb2b7d2	238	compilation_page = self._download_webpage(url, compilation_id, 'Downloading compilation web page')
77aa6b32	239	playlist_id = compilation_id
ceb2b7d2	240	playlist_title = self._html_search_meta('title', compilation_page, 'title')
c6270b2e	241	seasons = re.findall(
add96eb9	242	rf'<a href="/watch/{compilation_id}/season(\d+)', compilation_page)
c6270b2e	243	if not seasons: # No seasons in this compilation
77aa6b32	244	entries = self._extract_entries(compilation_page, compilation_id)
	245	else:
	246	entries = []
	247	for season_id in seasons:
ceb2b7d2	248	season_page = self._download_webpage(
add96eb9	249	f'http://www.ivi.ru/watch/{compilation_id}/season{season_id}',
add96eb9	250	compilation_id, f'Downloading season {season_id} web page')
77aa6b32	251	entries.extend(self._extract_entries(season_page, compilation_id))
77aa6b32	252
5f6a1245	253	return self.playlist_result(entries, playlist_id, playlist_title)