[yt-dlp.git] / youtube_dl / extractor / vrv.py

# coding: utf-8
from __future__ import unicode_literals

import base64
import json
import hashlib
import hmac
import random
import string
import time

from .common import InfoExtractor
from ..compat import (
    compat_HTTPError,
    compat_urllib_parse_urlencode,
    compat_urllib_parse,
)
from ..utils import (
    ExtractorError,
    float_or_none,
    int_or_none,
)


class VRVBaseIE(InfoExtractor):
    _API_DOMAIN = None
    _API_PARAMS = {}
    _CMS_SIGNING = {}
    _TOKEN = None
    _TOKEN_SECRET = ''

    def _call_api(self, path, video_id, note, data=None):
        # https://tools.ietf.org/html/rfc5849#section-3
        base_url = self._API_DOMAIN + '/core/' + path
        query = [
            ('oauth_consumer_key', self._API_PARAMS['oAuthKey']),
            ('oauth_nonce', ''.join([random.choice(string.ascii_letters) for _ in range(32)])),
            ('oauth_signature_method', 'HMAC-SHA1'),
            ('oauth_timestamp', int(time.time())),
        ]
        if self._TOKEN:
            query.append(('oauth_token', self._TOKEN))
        encoded_query = compat_urllib_parse_urlencode(query)
        headers = self.geo_verification_headers()
        if data:
            data = json.dumps(data).encode()
            headers['Content-Type'] = 'application/json'
        base_string = '&'.join([
            'POST' if data else 'GET',
            compat_urllib_parse.quote(base_url, ''),
            compat_urllib_parse.quote(encoded_query, '')])
        oauth_signature = base64.b64encode(hmac.new(
            (self._API_PARAMS['oAuthSecret'] + '&' + self._TOKEN_SECRET).encode('ascii'),
            base_string.encode(), hashlib.sha1).digest()).decode()
        encoded_query += '&oauth_signature=' + compat_urllib_parse.quote(oauth_signature, '')
        try:
            return self._download_json(
                '?'.join([base_url, encoded_query]), video_id,
                note='Downloading %s JSON metadata' % note, headers=headers, data=data)
        except ExtractorError as e:
            if isinstance(e.cause, compat_HTTPError) and e.cause.code == 401:
                raise ExtractorError(json.loads(e.cause.read().decode())['message'], expected=True)
            raise

    def _call_cms(self, path, video_id, note):
        if not self._CMS_SIGNING:
            self._CMS_SIGNING = self._call_api('index', video_id, 'CMS Signing')['cms_signing']
        return self._download_json(
            self._API_DOMAIN + path, video_id, query=self._CMS_SIGNING,
            note='Downloading %s JSON metadata' % note, headers=self.geo_verification_headers())

    def _get_cms_resource(self, resource_key, video_id):
        return self._call_api(
            'cms_resource', video_id, 'resource path', data={
                'resource_key': resource_key,
            })['__links__']['cms_resource']['href']

    def _real_initialize(self):
        webpage = self._download_webpage(
            'https://vrv.co/', None, headers=self.geo_verification_headers())
        self._API_PARAMS = self._parse_json(self._search_regex(
            [
                r'window\.__APP_CONFIG__\s*=\s*({.+?})(?:</script>|;)',
                r'window\.__APP_CONFIG__\s*=\s*({.+})'
            ], webpage, 'app config'), None)['cxApiParams']
        self._API_DOMAIN = self._API_PARAMS.get('apiDomain', 'https://api.vrv.co')


class VRVIE(VRVBaseIE):
    IE_NAME = 'vrv'
    _VALID_URL = r'https?://(?:www\.)?vrv\.co/watch/(?P<id>[A-Z0-9]+)'
    _TESTS = [{
        'url': 'https://vrv.co/watch/GR9PNZ396/Hidden-America-with-Jonah-Ray:BOSTON-WHERE-THE-PAST-IS-THE-PRESENT',
        'info_dict': {
            'id': 'GR9PNZ396',
            'ext': 'mp4',
            'title': 'BOSTON: WHERE THE PAST IS THE PRESENT',
            'description': 'md5:4ec8844ac262ca2df9e67c0983c6b83f',
            'uploader_id': 'seeso',
        },
        'params': {
            # m3u8 download
            'skip_download': True,
        },
    }, {
        # movie listing
        'url': 'https://vrv.co/watch/G6NQXZ1J6/Lily-CAT',
        'info_dict': {
            'id': 'G6NQXZ1J6',
            'title': 'Lily C.A.T',
            'description': 'md5:988b031e7809a6aeb60968be4af7db07',
        },
        'playlist_count': 2,
    }]
    _NETRC_MACHINE = 'vrv'

    def _real_initialize(self):
        super(VRVIE, self)._real_initialize()

        email, password = self._get_login_info()
        if email is None:
            return

        token_credentials = self._call_api(
            'authenticate/by:credentials', None, 'Token Credentials', data={
                'email': email,
                'password': password,
            })
        self._TOKEN = token_credentials['oauth_token']
        self._TOKEN_SECRET = token_credentials['oauth_token_secret']

    def _extract_vrv_formats(self, url, video_id, stream_format, audio_lang, hardsub_lang):
        if not url or stream_format not in ('hls', 'dash'):
            return []
        stream_id_list = []
        if audio_lang:
            stream_id_list.append('audio-%s' % audio_lang)
        if hardsub_lang:
            stream_id_list.append('hardsub-%s' % hardsub_lang)
        format_id = stream_format
        if stream_id_list:
            format_id += '-' + '-'.join(stream_id_list)
        if stream_format == 'hls':
            adaptive_formats = self._extract_m3u8_formats(
                url, video_id, 'mp4', m3u8_id=format_id,
                note='Downloading %s information' % format_id,
                fatal=False)
        elif stream_format == 'dash':
            adaptive_formats = self._extract_mpd_formats(
                url, video_id, mpd_id=format_id,
                note='Downloading %s information' % format_id,
                fatal=False)
        if audio_lang:
            for f in adaptive_formats:
                if f.get('acodec') != 'none':
                    f['language'] = audio_lang
        return adaptive_formats

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

        object_data = self._call_cms(self._get_cms_resource(
            'cms:/objects/' + video_id, video_id), video_id, 'object')['items'][0]
        resource_path = object_data['__links__']['resource']['href']
        video_data = self._call_cms(resource_path, video_id, 'video')
        title = video_data['title']
        description = video_data.get('description')

        if video_data.get('__class__') == 'movie_listing':
            items = self._call_cms(
                video_data['__links__']['movie_listing/movies']['href'],
                video_id, 'movie listing').get('items') or []
            if len(items) != 1:
                entries = []
                for item in items:
                    item_id = item.get('id')
                    if not item_id:
                        continue
                    entries.append(self.url_result(
                        'https://vrv.co/watch/' + item_id,
                        self.ie_key(), item_id, item.get('title')))
                return self.playlist_result(entries, video_id, title, description)
            video_data = items[0]

        streams_path = video_data['__links__'].get('streams', {}).get('href')
        if not streams_path:
            self.raise_login_required()
        streams_json = self._call_cms(streams_path, video_id, 'streams')

        audio_locale = streams_json.get('audio_locale')
        formats = []
        for stream_type, streams in streams_json.get('streams', {}).items():
            if stream_type in ('adaptive_hls', 'adaptive_dash'):
                for stream in streams.values():
                    formats.extend(self._extract_vrv_formats(
                        stream.get('url'), video_id, stream_type.split('_')[1],
                        audio_locale, stream.get('hardsub_locale')))
        self._sort_formats(formats)

        subtitles = {}
        for subtitle in streams_json.get('subtitles', {}).values():
            subtitle_url = subtitle.get('url')
            if not subtitle_url:
                continue
            subtitles.setdefault(subtitle.get('locale', 'en-US'), []).append({
                'url': subtitle_url,
                'ext': subtitle.get('format', 'ass'),
            })

        thumbnails = []
        for thumbnail in video_data.get('images', {}).get('thumbnails', []):
            thumbnail_url = thumbnail.get('source')
            if not thumbnail_url:
                continue
            thumbnails.append({
                'url': thumbnail_url,
                'width': int_or_none(thumbnail.get('width')),
                'height': int_or_none(thumbnail.get('height')),
            })

        return {
            'id': video_id,
            'title': title,
            'formats': formats,
            'subtitles': subtitles,
            'thumbnails': thumbnails,
            'description': description,
            'duration': float_or_none(video_data.get('duration_ms'), 1000),
            'uploader_id': video_data.get('channel_id'),
            'series': video_data.get('series_title'),
            'season': video_data.get('season_title'),
            'season_number': int_or_none(video_data.get('season_number')),
            'season_id': video_data.get('season_id'),
            'episode': title,
            'episode_number': int_or_none(video_data.get('episode_number')),
            'episode_id': video_data.get('production_episode_id'),
        }


class VRVSeriesIE(VRVBaseIE):
    IE_NAME = 'vrv:series'
    _VALID_URL = r'https?://(?:www\.)?vrv\.co/series/(?P<id>[A-Z0-9]+)'
    _TEST = {
        'url': 'https://vrv.co/series/G68VXG3G6/The-Perfect-Insider',
        'info_dict': {
            'id': 'G68VXG3G6',
        },
        'playlist_mincount': 11,
    }

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

        seasons_path = self._get_cms_resource(
            'cms:/seasons?series_id=' + series_id, series_id)
        seasons_data = self._call_cms(seasons_path, series_id, 'seasons')

        entries = []
        for season in seasons_data.get('items', []):
            episodes_path = season['__links__']['season/episodes']['href']
            episodes = self._call_cms(episodes_path, series_id, 'episodes')
            for episode in episodes.get('items', []):
                episode_id = episode['id']
                entries.append(self.url_result(
                    'https://vrv.co/watch/' + episode_id,
                    'VRV', episode_id, episode.get('title')))

        return self.playlist_result(entries, series_id)
Commit	Line	Data
77c8ebe6 RA	1	# coding: utf-8
	2	from __future__ import unicode_literals
	3
	4	import base64
	5	import json
	6	import hashlib
	7	import hmac
	8	import random
	9	import string
	10	import time
	11
	12	from .common import InfoExtractor
	13	from ..compat import (
4b85f0f9	14	compat_HTTPError,
77c8ebe6	15	compat_urllib_parse_urlencode,
16132cff	16	compat_urllib_parse,
77c8ebe6 RA	17	)
77c8ebe6 RA	18	from ..utils import (
4b85f0f9	19	ExtractorError,
77c8ebe6 RA	20	float_or_none,
	21	int_or_none,
	22	)
	23
	24
48ab554f	25	class VRVBaseIE(InfoExtractor):
77c8ebe6 RA	26	_API_DOMAIN = None
	27	_API_PARAMS = {}
	28	_CMS_SIGNING = {}
4b85f0f9 RA	29	_TOKEN = None
4b85f0f9 RA	30	_TOKEN_SECRET = ''
77c8ebe6 RA	31
77c8ebe6 RA	32	def _call_api(self, path, video_id, note, data=None):
4b85f0f9	33	# https://tools.ietf.org/html/rfc5849#section-3
77c8ebe6	34	base_url = self._API_DOMAIN + '/core/' + path
503b604a RA	35	query = [
	36	('oauth_consumer_key', self._API_PARAMS['oAuthKey']),
	37	('oauth_nonce', ''.join([random.choice(string.ascii_letters) for _ in range(32)])),
	38	('oauth_signature_method', 'HMAC-SHA1'),
	39	('oauth_timestamp', int(time.time())),
	40	]
4b85f0f9	41	if self._TOKEN:
503b604a	42	query.append(('oauth_token', self._TOKEN))
4b85f0f9	43	encoded_query = compat_urllib_parse_urlencode(query)
77c8ebe6 RA	44	headers = self.geo_verification_headers()
	45	if data:
	46	data = json.dumps(data).encode()
	47	headers['Content-Type'] = 'application/json'
4b85f0f9 RA	48	base_string = '&'.join([
	49	'POST' if data else 'GET',
	50	compat_urllib_parse.quote(base_url, ''),
	51	compat_urllib_parse.quote(encoded_query, '')])
77c8ebe6	52	oauth_signature = base64.b64encode(hmac.new(
4b85f0f9	53	(self._API_PARAMS['oAuthSecret'] + '&' + self._TOKEN_SECRET).encode('ascii'),
77c8ebe6	54	base_string.encode(), hashlib.sha1).digest()).decode()
16132cff	55	encoded_query += '&oauth_signature=' + compat_urllib_parse.quote(oauth_signature, '')
4b85f0f9 RA	56	try:
	57	return self._download_json(
	58	'?'.join([base_url, encoded_query]), video_id,
	59	note='Downloading %s JSON metadata' % note, headers=headers, data=data)
	60	except ExtractorError as e:
	61	if isinstance(e.cause, compat_HTTPError) and e.cause.code == 401:
	62	raise ExtractorError(json.loads(e.cause.read().decode())['message'], expected=True)
	63	raise
77c8ebe6 RA	64
77c8ebe6 RA	65	def _call_cms(self, path, video_id, note):
48ab554f RA	66	if not self._CMS_SIGNING:
48ab554f RA	67	self._CMS_SIGNING = self._call_api('index', video_id, 'CMS Signing')['cms_signing']
77c8ebe6 RA	68	return self._download_json(
	69	self._API_DOMAIN + path, video_id, query=self._CMS_SIGNING,
	70	note='Downloading %s JSON metadata' % note, headers=self.geo_verification_headers())
	71
48ab554f RA	72	def _get_cms_resource(self, resource_key, video_id):
	73	return self._call_api(
	74	'cms_resource', video_id, 'resource path', data={
	75	'resource_key': resource_key,
	76	})['__links__']['cms_resource']['href']
	77
4b85f0f9 RA	78	def _real_initialize(self):
	79	webpage = self._download_webpage(
	80	'https://vrv.co/', None, headers=self.geo_verification_headers())
	81	self._API_PARAMS = self._parse_json(self._search_regex(
	82	[
	83	r'window\.__APP_CONFIG__\s=\s({.+?})(?:</script>\|;)',
	84	r'window\.__APP_CONFIG__\s=\s({.+})'
	85	], webpage, 'app config'), None)['cxApiParams']
	86	self._API_DOMAIN = self._API_PARAMS.get('apiDomain', 'https://api.vrv.co')
	87
48ab554f RA	88
	89	class VRVIE(VRVBaseIE):
	90	IE_NAME = 'vrv'
	91	_VALID_URL = r'https?://(?:www\.)?vrv\.co/watch/(?P<id>[A-Z0-9]+)'
54a5be4d	92	_TESTS = [{
48ab554f RA	93	'url': 'https://vrv.co/watch/GR9PNZ396/Hidden-America-with-Jonah-Ray:BOSTON-WHERE-THE-PAST-IS-THE-PRESENT',
	94	'info_dict': {
	95	'id': 'GR9PNZ396',
	96	'ext': 'mp4',
	97	'title': 'BOSTON: WHERE THE PAST IS THE PRESENT',
	98	'description': 'md5:4ec8844ac262ca2df9e67c0983c6b83f',
	99	'uploader_id': 'seeso',
	100	},
	101	'params': {
	102	# m3u8 download
	103	'skip_download': True,
	104	},
1fa88937 RA	105	}, {
	106	# movie listing
	107	'url': 'https://vrv.co/watch/G6NQXZ1J6/Lily-CAT',
	108	'info_dict': {
	109	'id': 'G6NQXZ1J6',
	110	'title': 'Lily C.A.T',
	111	'description': 'md5:988b031e7809a6aeb60968be4af7db07',
	112	},
	113	'playlist_count': 2,
54a5be4d	114	}]
4b85f0f9 RA	115	_NETRC_MACHINE = 'vrv'
	116
	117	def _real_initialize(self):
	118	super(VRVIE, self)._real_initialize()
	119
	120	email, password = self._get_login_info()
	121	if email is None:
	122	return
	123
	124	token_credentials = self._call_api(
	125	'authenticate/by:credentials', None, 'Token Credentials', data={
	126	'email': email,
	127	'password': password,
	128	})
	129	self._TOKEN = token_credentials['oauth_token']
	130	self._TOKEN_SECRET = token_credentials['oauth_token_secret']
54a5be4d RA	131
	132	def _extract_vrv_formats(self, url, video_id, stream_format, audio_lang, hardsub_lang):
	133	if not url or stream_format not in ('hls', 'dash'):
	134	return []
d9b0d118 S	135	stream_id_list = []
	136	if audio_lang:
	137	stream_id_list.append('audio-%s' % audio_lang)
	138	if hardsub_lang:
	139	stream_id_list.append('hardsub-%s' % hardsub_lang)
1fa88937 RA	140	format_id = stream_format
	141	if stream_id_list:
	142	format_id += '-' + '-'.join(stream_id_list)
54a5be4d RA	143	if stream_format == 'hls':
	144	adaptive_formats = self._extract_m3u8_formats(
	145	url, video_id, 'mp4', m3u8_id=format_id,
1fa88937	146	note='Downloading %s information' % format_id,
54a5be4d RA	147	fatal=False)
	148	elif stream_format == 'dash':
	149	adaptive_formats = self._extract_mpd_formats(
	150	url, video_id, mpd_id=format_id,
1fa88937	151	note='Downloading %s information' % format_id,
54a5be4d RA	152	fatal=False)
	153	if audio_lang:
	154	for f in adaptive_formats:
	155	if f.get('acodec') != 'none':
	156	f['language'] = audio_lang
	157	return adaptive_formats
77c8ebe6 RA	158
	159	def _real_extract(self, url):
	160	video_id = self._match_id(url)
4b85f0f9	161
9f182c23 RA	162	object_data = self._call_cms(self._get_cms_resource(
	163	'cms:/objects/' + video_id, video_id), video_id, 'object')['items'][0]
	164	resource_path = object_data['__links__']['resource']['href']
	165	video_data = self._call_cms(resource_path, video_id, 'video')
77c8ebe6	166	title = video_data['title']
1fa88937 RA	167	description = video_data.get('description')
	168
	169	if video_data.get('__class__') == 'movie_listing':
	170	items = self._call_cms(
	171	video_data['__links__']['movie_listing/movies']['href'],
	172	video_id, 'movie listing').get('items') or []
	173	if len(items) != 1:
	174	entries = []
	175	for item in items:
	176	item_id = item.get('id')
	177	if not item_id:
	178	continue
	179	entries.append(self.url_result(
	180	'https://vrv.co/watch/' + item_id,
	181	self.ie_key(), item_id, item.get('title')))
	182	return self.playlist_result(entries, video_id, title, description)
	183	video_data = items[0]
77c8ebe6	184
4b85f0f9 RA	185	streams_path = video_data['__links__'].get('streams', {}).get('href')
	186	if not streams_path:
	187	self.raise_login_required()
	188	streams_json = self._call_cms(streams_path, video_id, 'streams')
77c8ebe6 RA	189
	190	audio_locale = streams_json.get('audio_locale')
	191	formats = []
afa0200b RA	192	for stream_type, streams in streams_json.get('streams', {}).items():
	193	if stream_type in ('adaptive_hls', 'adaptive_dash'):
	194	for stream in streams.values():
54a5be4d RA	195	formats.extend(self._extract_vrv_formats(
	196	stream.get('url'), video_id, stream_type.split('_')[1],
	197	audio_locale, stream.get('hardsub_locale')))
77c8ebe6 RA	198	self._sort_formats(formats)
77c8ebe6 RA	199
afa0200b RA	200	subtitles = {}
	201	for subtitle in streams_json.get('subtitles', {}).values():
	202	subtitle_url = subtitle.get('url')
	203	if not subtitle_url:
	204	continue
	205	subtitles.setdefault(subtitle.get('locale', 'en-US'), []).append({
	206	'url': subtitle_url,
	207	'ext': subtitle.get('format', 'ass'),
	208	})
	209
77c8ebe6 RA	210	thumbnails = []
	211	for thumbnail in video_data.get('images', {}).get('thumbnails', []):
	212	thumbnail_url = thumbnail.get('source')
	213	if not thumbnail_url:
	214	continue
	215	thumbnails.append({
	216	'url': thumbnail_url,
	217	'width': int_or_none(thumbnail.get('width')),
	218	'height': int_or_none(thumbnail.get('height')),
	219	})
	220
	221	return {
	222	'id': video_id,
	223	'title': title,
	224	'formats': formats,
afa0200b	225	'subtitles': subtitles,
77c8ebe6	226	'thumbnails': thumbnails,
1fa88937	227	'description': description,
77c8ebe6 RA	228	'duration': float_or_none(video_data.get('duration_ms'), 1000),
	229	'uploader_id': video_data.get('channel_id'),
	230	'series': video_data.get('series_title'),
	231	'season': video_data.get('season_title'),
	232	'season_number': int_or_none(video_data.get('season_number')),
	233	'season_id': video_data.get('season_id'),
	234	'episode': title,
	235	'episode_number': int_or_none(video_data.get('episode_number')),
	236	'episode_id': video_data.get('production_episode_id'),
	237	}
48ab554f RA	238
	239
	240	class VRVSeriesIE(VRVBaseIE):
	241	IE_NAME = 'vrv:series'
	242	_VALID_URL = r'https?://(?:www\.)?vrv\.co/series/(?P<id>[A-Z0-9]+)'
	243	_TEST = {
	244	'url': 'https://vrv.co/series/G68VXG3G6/The-Perfect-Insider',
	245	'info_dict': {
	246	'id': 'G68VXG3G6',
	247	},
	248	'playlist_mincount': 11,
	249	}
	250
	251	def _real_extract(self, url):
	252	series_id = self._match_id(url)
48ab554f	253
48ab554f RA	254	seasons_path = self._get_cms_resource(
	255	'cms:/seasons?series_id=' + series_id, series_id)
	256	seasons_data = self._call_cms(seasons_path, series_id, 'seasons')
	257
	258	entries = []
	259	for season in seasons_data.get('items', []):
	260	episodes_path = season['__links__']['season/episodes']['href']
	261	episodes = self._call_cms(episodes_path, series_id, 'episodes')
	262	for episode in episodes.get('items', []):
	263	episode_id = episode['id']
	264	entries.append(self.url_result(
	265	'https://vrv.co/watch/' + episode_id,
	266	'VRV', episode_id, episode.get('title')))
	267
	268	return self.playlist_result(entries, series_id)