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

import json
import re

from .common import InfoExtractor
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': 'http://www.ivi.ru/watch/%s' % video_id,
                    'contentid': video_id
                }
            ]
        })

        for site in (353, 183):
            content_data = (data % site).encode()
            if site == 353:
                try:
                    from Cryptodome.Cipher import Blowfish
                    from Cryptodome.Hash import CMAC
                    pycryptodome_found = True
                except ImportError:
                    try:
                        from Crypto.Cipher import Blowfish
                        from Crypto.Hash import CMAC
                        pycryptodome_found = True
                    except ImportError:
                        pycryptodome_found = False
                        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': CMAC.new(self._LIGHT_KEY, timestamp.encode() + content_data, 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 pycryptodome_found:
                    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 = '%s - %s' % (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(
                'http://www.ivi.ru/watch/%s/%s' % (compilation_id, serie), IviIE.ie_key())
            for serie in re.findall(
                r'<a\b[^>]+\bhref=["\']/watch/%s/(\d+)["\']' % compilation_id, 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, 'Downloading season %s web page' % season_id)
            playlist_id = '%s/season%s' % (compilation_id, 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(
                r'<a href="/watch/%s/season(\d+)' % compilation_id, 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(
                        'http://www.ivi.ru/watch/%s/season%s' % (compilation_id, season_id),
                        compilation_id, 'Downloading season %s web page' % season_id)
                    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
1cc79574	5	from ..utils import (
77aa6b32	6	ExtractorError,
ab3176af	7	int_or_none,
cf143c4d	8	qualities,
77aa6b32	9	)
	10
	11
	12	class IviIE(InfoExtractor):
ceb2b7d2	13	IE_DESC = 'ivi.ru'
ceb2b7d2	14	IE_NAME = 'ivi'
022218f2	15	_VALID_URL = r'https?://(?:www\.)?ivi\.(?:ru\|tv)/(?:watch/(?:[^/]+/)?\|video/player\?.*?videoId=)(?P<id>\d+)'
bfd973ec	16	_EMBED_REGEX = [r'<embed[^>]+?src=(["\'])(?P<url>https?://(?:www\.)?ivi\.ru/video/player.+?)\1']
42dcdbe1 S	17	_GEO_BYPASS = False
42dcdbe1 S	18	_GEO_COUNTRIES = ['RU']
656c2001 RA	19	_LIGHT_KEY = b'\xf1\x02\x32\xb7\xbc\x5c\x7a\xe8\xf7\x96\xc1\x33\x2b\x27\xa1\x8c'
656c2001 RA	20	_LIGHT_URL = 'https://api.ivi.ru/light/'
77aa6b32	21
	22	_TESTS = [
	23	# Single movie
	24	{
ceb2b7d2	25	'url': 'http://www.ivi.ru/watch/53141',
ceb2b7d2	26	'md5': '6ff5be2254e796ed346251d117196cf4',
ceb2b7d2	27	'info_dict': {
84dd7031 S	28	'id': '53141',
84dd7031 S	29	'ext': 'mp4',
ceb2b7d2	30	'title': 'Иван Васильевич меняет профессию',
	31	'description': 'md5:b924063ea1677c8fe343d8a72ac2195f',
	32	'duration': 5498,
ec85ded8	33	'thumbnail': r're:^https?://.*\.jpg$',
77aa6b32	34	},
ceb2b7d2	35	'skip': 'Only works from Russia',
77aa6b32	36	},
dfb1b146	37	# Serial's series
77aa6b32	38	{
6ebb46c1 S	39	'url': 'http://www.ivi.ru/watch/dvoe_iz_lartsa/9549',
6ebb46c1 S	40	'md5': '221f56b35e3ed815fde2df71032f4b3e',
ceb2b7d2	41	'info_dict': {
6ebb46c1	42	'id': '9549',
84dd7031	43	'ext': 'mp4',
ab3176af S	44	'title': 'Двое из ларца - Дело Гольдберга (1 часть)',
ab3176af S	45	'series': 'Двое из ларца',
1463c5b9 S	46	'season': 'Сезон 1',
1463c5b9 S	47	'season_number': 1,
ab3176af S	48	'episode': 'Дело Гольдберга (1 часть)',
ab3176af S	49	'episode_number': 1,
6ebb46c1	50	'duration': 2655,
ec85ded8	51	'thumbnail': r're:^https?://.*\.jpg$',
77aa6b32	52	},
ceb2b7d2	53	'skip': 'Only works from Russia',
cf143c4d S	54	},
	55	{
	56	# with MP4-HD720 format
	57	'url': 'http://www.ivi.ru/watch/146500',
	58	'md5': 'd63d35cdbfa1ea61a5eafec7cc523e1e',
	59	'info_dict': {
	60	'id': '146500',
	61	'ext': 'mp4',
	62	'title': 'Кукла',
	63	'description': 'md5:ffca9372399976a2d260a407cc74cce6',
	64	'duration': 5599,
ec85ded8	65	'thumbnail': r're:^https?://.*\.jpg$',
cf143c4d S	66	},
cf143c4d S	67	'skip': 'Only works from Russia',
022218f2 S	68	},
	69	{
	70	'url': 'https://www.ivi.tv/watch/33560/',
	71	'only_matching': True,
	72	},
77aa6b32	73	]
ceb2b7d2	74
77aa6b32	75	# Sorted by quality
cf143c4d S	76	_KNOWN_FORMATS = (
	77	'MP4-low-mobile', 'MP4-mobile', 'FLV-lo', 'MP4-lo', 'FLV-hi', 'MP4-hi',
	78	'MP4-SHQ', 'MP4-HD720', 'MP4-HD1080')
77aa6b32	79
77aa6b32	80	def _real_extract(self, url):
63be3b89	81	video_id = self._match_id(url)
77aa6b32	82
656c2001	83	data = json.dumps({
63be3b89 S	84	'method': 'da.content.get',
	85	'params': [
	86	video_id, {
1bba88ef	87	'site': 's%d',
63be3b89 S	88	'referrer': 'http://www.ivi.ru/watch/%s' % video_id,
63be3b89 S	89	'contentid': video_id
77aa6b32	90	}
63be3b89	91	]
f8015c15	92	})
77aa6b32	93
76d9eca4	94	for site in (353, 183):
f8015c15	95	content_data = (data % site).encode()
76d9eca4 RA	96	if site == 353:
	97	try:
	98	from Cryptodome.Cipher import Blowfish
	99	from Cryptodome.Hash import CMAC
edf65256	100	pycryptodome_found = True
76d9eca4	101	except ImportError:
edf65256	102	try:
	103	from Crypto.Cipher import Blowfish
	104	from Crypto.Hash import CMAC
	105	pycryptodome_found = True
	106	except ImportError:
	107	pycryptodome_found = False
	108	continue
76d9eca4 RA	109
	110	timestamp = (self._download_json(
	111	self._LIGHT_URL, video_id,
	112	'Downloading timestamp JSON', data=json.dumps({
	113	'method': 'da.timestamp.get',
	114	'params': []
	115	}).encode(), fatal=False) or {}).get('result')
	116	if not timestamp:
	117	continue
	118
	119	query = {
	120	'ts': timestamp,
	121	'sign': CMAC.new(self._LIGHT_KEY, timestamp.encode() + content_data, Blowfish).hexdigest(),
	122	}
	123	else:
	124	query = {}
1bba88ef	125
76d9eca4	126	video_json = self._download_json(
1bba88ef	127	self._LIGHT_URL, video_id,
76d9eca4 RA	128	'Downloading video JSON', data=content_data, query=query)
	129
	130	error = video_json.get('error')
	131	if error:
	132	origin = error.get('origin')
	133	message = error.get('message') or error.get('user_message')
	134	extractor_msg = 'Unable to download video %s'
	135	if origin == 'NotAllowedForLocation':
	136	self.raise_geo_restricted(message, self._GEO_COUNTRIES)
	137	elif origin == 'NoRedisValidData':
	138	extractor_msg = 'Video %s does not exist'
	139	elif site == 353:
	140	continue
edf65256	141	elif not pycryptodome_found:
49e7e9c3	142	raise ExtractorError('pycryptodomex not found. Please install', expected=True)
76d9eca4 RA	143	elif message:
	144	extractor_msg += ': ' + message
	145	raise ExtractorError(extractor_msg % video_id, expected=True)
	146	else:
	147	break
77aa6b32	148
ceb2b7d2	149	result = video_json['result']
656c2001	150	title = result['title']
77aa6b32	151
cf143c4d S	152	quality = qualities(self._KNOWN_FORMATS)
cf143c4d S	153
656c2001 RA	154	formats = []
	155	for f in result.get('files', []):
	156	f_url = f.get('url')
	157	content_format = f.get('content_format')
06869367	158	if not f_url:
06869367	159	continue
a06916d9	160	if (not self.get_param('allow_unplayable_formats')
06869367	161	and ('-MDRM-' in content_format or '-FPS-' in content_format)):
656c2001 RA	162	continue
	163	formats.append({
	164	'url': f_url,
	165	'format_id': content_format,
	166	'quality': quality(content_format),
	167	'filesize': int_or_none(f.get('size_in_bytes')),
	168	})
bf5b0a1b	169
ab3176af S	170	compilation = result.get('compilation')
	171	episode = title if compilation else None
	172
5f6a1245	173	title = '%s - %s' % (compilation, title) if compilation is not None else title
77aa6b32	174
ab3176af S	175	thumbnails = [{
	176	'url': preview['url'],
	177	'id': preview.get('content_format'),
	178	} for preview in result.get('preview', []) if preview.get('url')]
	179
	180	webpage = self._download_webpage(url, video_id)
	181
1463c5b9 S	182	season = self._search_regex(
	183	r'<li[^>]+class="season active"[^>]*><a[^>]+>([^<]+)',
	184	webpage, 'season', default=None)
	185	season_number = int_or_none(self._search_regex(
	186	r'<li[^>]+class="season active"[^>]*><a[^>]+data-season(?:-index)?="(\d+)"',
	187	webpage, 'season number', default=None))
	188
ab3176af	189	episode_number = int_or_none(self._search_regex(
3d897cc7	190	r'[^>]+itemprop="episode"[^>]>\s<meta[^>]+itemprop="episodeNumber"[^>]+content="(\d+)',
ab3176af	191	webpage, 'episode number', default=None))
77aa6b32	192
ab3176af S	193	description = self._og_search_description(webpage, default=None) or self._html_search_meta(
ab3176af S	194	'description', webpage, 'description', default=None)
77aa6b32	195
	196	return {
	197	'id': video_id,
	198	'title': title,
ab3176af	199	'series': compilation,
1463c5b9 S	200	'season': season,
1463c5b9 S	201	'season_number': season_number,
ab3176af S	202	'episode': episode,
	203	'episode_number': episode_number,
	204	'thumbnails': thumbnails,
77aa6b32	205	'description': description,
656c2001	206	'duration': int_or_none(result.get('duration')),
77aa6b32	207	'formats': formats,
	208	}
	209
	210
	211	class IviCompilationIE(InfoExtractor):
ceb2b7d2	212	IE_DESC = 'ivi.ru compilations'
ceb2b7d2	213	IE_NAME = 'ivi:compilation'
84dd7031	214	_VALID_URL = r'https?://(?:www\.)?ivi\.ru/watch/(?!\d+)(?P<compilationid>[a-z\d_-]+)(?:/season(?P<seasonid>\d+))?$'
22a6f150 PH	215	_TESTS = [{
	216	'url': 'http://www.ivi.ru/watch/dvoe_iz_lartsa',
	217	'info_dict': {
	218	'id': 'dvoe_iz_lartsa',
	219	'title': 'Двое из ларца (2006 - 2008)',
	220	},
	221	'playlist_mincount': 24,
	222	}, {
	223	'url': 'http://www.ivi.ru/watch/dvoe_iz_lartsa/season1',
	224	'info_dict': {
	225	'id': 'dvoe_iz_lartsa/season1',
	226	'title': 'Двое из ларца (2006 - 2008) 1 сезон',
	227	},
	228	'playlist_mincount': 12,
	229	}]
77aa6b32	230
77aa6b32	231	def _extract_entries(self, html, compilation_id):
c6270b2e S	232	return [
	233	self.url_result(
	234	'http://www.ivi.ru/watch/%s/%s' % (compilation_id, serie), IviIE.ie_key())
	235	for serie in re.findall(
d9a2f867	236	r'<a\b[^>]+\bhref=["\']/watch/%s/(\d+)["\']' % compilation_id, html)]
77aa6b32	237
77aa6b32	238	def _real_extract(self, url):
5ad28e7f	239	mobj = self._match_valid_url(url)
77aa6b32	240	compilation_id = mobj.group('compilationid')
	241	season_id = mobj.group('seasonid')
	242
5f6a1245	243	if season_id is not None: # Season link
c6270b2e S	244	season_page = self._download_webpage(
c6270b2e S	245	url, compilation_id, 'Downloading season %s web page' % season_id)
77aa6b32	246	playlist_id = '%s/season%s' % (compilation_id, season_id)
ceb2b7d2	247	playlist_title = self._html_search_meta('title', season_page, 'title')
77aa6b32	248	entries = self._extract_entries(season_page, compilation_id)
5f6a1245	249	else: # Compilation link
ceb2b7d2	250	compilation_page = self._download_webpage(url, compilation_id, 'Downloading compilation web page')
77aa6b32	251	playlist_id = compilation_id
ceb2b7d2	252	playlist_title = self._html_search_meta('title', compilation_page, 'title')
c6270b2e S	253	seasons = re.findall(
	254	r'<a href="/watch/%s/season(\d+)' % compilation_id, compilation_page)
	255	if not seasons: # No seasons in this compilation
77aa6b32	256	entries = self._extract_entries(compilation_page, compilation_id)
	257	else:
	258	entries = []
	259	for season_id in seasons:
ceb2b7d2	260	season_page = self._download_webpage(
	261	'http://www.ivi.ru/watch/%s/season%s' % (compilation_id, season_id),
	262	compilation_id, 'Downloading season %s web page' % season_id)
77aa6b32	263	entries.extend(self._extract_entries(season_page, compilation_id))
77aa6b32	264
5f6a1245	265	return self.playlist_result(entries, playlist_id, playlist_title)