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

import base64
import hashlib
import hmac
import json
import random
import string
import time
import urllib.parse

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


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',
            urllib.parse.quote(base_url, ''),
            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=' + 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:
            index = self._call_api('index', video_id, 'CMS Signing')
            self._CMS_SIGNING = index.get('cms_signing') or {}
            if not self._CMS_SIGNING:
                for signing_policy in index.get('signing_policies', []):
                    signing_path = signing_policy.get('path')
                    if signing_path and signing_path.startswith('/cms/'):
                        name, value = signing_policy.get('name'), signing_policy.get('value')
                        if name and value:
                            self._CMS_SIGNING[name] = value
        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 _extract_vrv_formats(self, url, video_id, stream_format, audio_lang, hardsub_lang):
        if not url or stream_format not in ('hls', 'dash', 'adaptive_hls'):
            return []
        format_id = join_nonempty(
            stream_format,
            audio_lang and 'audio-%s' % audio_lang,
            hardsub_lang and 'hardsub-%s' % hardsub_lang)
        if 'hls' in stream_format:
            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 _set_api_params(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 _perform_login(self, username, password):
        token_credentials = self._call_api(
            'authenticate/by:credentials', None, 'Token Credentials', data={
                'email': username,
                'password': password,
            })
        self._TOKEN = token_credentials['oauth_token']
        self._TOKEN_SECRET = token_credentials['oauth_token_secret']

    def _initialize_pre_login(self):
        return self._set_api_params()

    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')))

        subtitles = {}
        for k in ('captions', 'subtitles'):
            for subtitle in streams_json.get(k, {}).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 traverse_obj(video_data, ('images', 'thumbnail', ..., ...)) or []:
            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 _initialize_pre_login(self):
        return self._set_api_params()

    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	1	import base64
77c8ebe6 RA	2	import hashlib
77c8ebe6 RA	3	import hmac
ac668111	4	import json
77c8ebe6 RA	5	import random
	6	import string
	7	import time
ac668111	8	import urllib.parse
77c8ebe6 RA	9
77c8ebe6 RA	10	from .common import InfoExtractor
ac668111	11	from ..compat import compat_HTTPError, compat_urllib_parse_urlencode
77c8ebe6	12	from ..utils import (
4b85f0f9	13	ExtractorError,
77c8ebe6 RA	14	float_or_none,
77c8ebe6 RA	15	int_or_none,
34921b43	16	join_nonempty,
7bb64347	17	traverse_obj,
77c8ebe6 RA	18	)
	19
	20
48ab554f	21	class VRVBaseIE(InfoExtractor):
77c8ebe6 RA	22	_API_DOMAIN = None
	23	_API_PARAMS = {}
	24	_CMS_SIGNING = {}
4b85f0f9 RA	25	_TOKEN = None
4b85f0f9 RA	26	_TOKEN_SECRET = ''
77c8ebe6 RA	27
77c8ebe6 RA	28	def _call_api(self, path, video_id, note, data=None):
4b85f0f9	29	# https://tools.ietf.org/html/rfc5849#section-3
77c8ebe6	30	base_url = self._API_DOMAIN + '/core/' + path
503b604a RA	31	query = [
	32	('oauth_consumer_key', self._API_PARAMS['oAuthKey']),
	33	('oauth_nonce', ''.join([random.choice(string.ascii_letters) for _ in range(32)])),
	34	('oauth_signature_method', 'HMAC-SHA1'),
	35	('oauth_timestamp', int(time.time())),
	36	]
4b85f0f9	37	if self._TOKEN:
503b604a	38	query.append(('oauth_token', self._TOKEN))
4b85f0f9	39	encoded_query = compat_urllib_parse_urlencode(query)
77c8ebe6 RA	40	headers = self.geo_verification_headers()
	41	if data:
	42	data = json.dumps(data).encode()
	43	headers['Content-Type'] = 'application/json'
4b85f0f9 RA	44	base_string = '&'.join([
4b85f0f9 RA	45	'POST' if data else 'GET',
ac668111	46	urllib.parse.quote(base_url, ''),
ac668111	47	urllib.parse.quote(encoded_query, '')])
77c8ebe6	48	oauth_signature = base64.b64encode(hmac.new(
4b85f0f9	49	(self._API_PARAMS['oAuthSecret'] + '&' + self._TOKEN_SECRET).encode('ascii'),
77c8ebe6	50	base_string.encode(), hashlib.sha1).digest()).decode()
ac668111	51	encoded_query += '&oauth_signature=' + urllib.parse.quote(oauth_signature, '')
4b85f0f9 RA	52	try:
	53	return self._download_json(
	54	'?'.join([base_url, encoded_query]), video_id,
	55	note='Downloading %s JSON metadata' % note, headers=headers, data=data)
	56	except ExtractorError as e:
	57	if isinstance(e.cause, compat_HTTPError) and e.cause.code == 401:
	58	raise ExtractorError(json.loads(e.cause.read().decode())['message'], expected=True)
	59	raise
77c8ebe6 RA	60
77c8ebe6 RA	61	def _call_cms(self, path, video_id, note):
48ab554f	62	if not self._CMS_SIGNING:
1824bfdc RA	63	index = self._call_api('index', video_id, 'CMS Signing')
	64	self._CMS_SIGNING = index.get('cms_signing') or {}
	65	if not self._CMS_SIGNING:
	66	for signing_policy in index.get('signing_policies', []):
	67	signing_path = signing_policy.get('path')
	68	if signing_path and signing_path.startswith('/cms/'):
	69	name, value = signing_policy.get('name'), signing_policy.get('value')
	70	if name and value:
	71	self._CMS_SIGNING[name] = value
77c8ebe6 RA	72	return self._download_json(
	73	self._API_DOMAIN + path, video_id, query=self._CMS_SIGNING,
	74	note='Downloading %s JSON metadata' % note, headers=self.geo_verification_headers())
	75
48ab554f RA	76	def _get_cms_resource(self, resource_key, video_id):
	77	return self._call_api(
	78	'cms_resource', video_id, 'resource path', data={
	79	'resource_key': resource_key,
	80	})['__links__']['cms_resource']['href']
	81
11078c6d	82	def _extract_vrv_formats(self, url, video_id, stream_format, audio_lang, hardsub_lang):
	83	if not url or stream_format not in ('hls', 'dash', 'adaptive_hls'):
	84	return []
	85	format_id = join_nonempty(
	86	stream_format,
	87	audio_lang and 'audio-%s' % audio_lang,
	88	hardsub_lang and 'hardsub-%s' % hardsub_lang)
	89	if 'hls' in stream_format:
	90	adaptive_formats = self._extract_m3u8_formats(
	91	url, video_id, 'mp4', m3u8_id=format_id,
	92	note='Downloading %s information' % format_id,
	93	fatal=False)
	94	elif stream_format == 'dash':
	95	adaptive_formats = self._extract_mpd_formats(
	96	url, video_id, mpd_id=format_id,
	97	note='Downloading %s information' % format_id,
	98	fatal=False)
	99	if audio_lang:
	100	for f in adaptive_formats:
	101	if f.get('acodec') != 'none':
	102	f['language'] = audio_lang
	103	return adaptive_formats
	104
	105	def _set_api_params(self):
4b85f0f9 RA	106	webpage = self._download_webpage(
	107	'https://vrv.co/', None, headers=self.geo_verification_headers())
	108	self._API_PARAMS = self._parse_json(self._search_regex(
	109	[
	110	r'window\.__APP_CONFIG__\s=\s({.+?})(?:</script>\|;)',
	111	r'window\.__APP_CONFIG__\s=\s({.+})'
	112	], webpage, 'app config'), None)['cxApiParams']
	113	self._API_DOMAIN = self._API_PARAMS.get('apiDomain', 'https://api.vrv.co')
	114
48ab554f RA	115
	116	class VRVIE(VRVBaseIE):
	117	IE_NAME = 'vrv'
	118	_VALID_URL = r'https?://(?:www\.)?vrv\.co/watch/(?P<id>[A-Z0-9]+)'
54a5be4d	119	_TESTS = [{
48ab554f RA	120	'url': 'https://vrv.co/watch/GR9PNZ396/Hidden-America-with-Jonah-Ray:BOSTON-WHERE-THE-PAST-IS-THE-PRESENT',
	121	'info_dict': {
	122	'id': 'GR9PNZ396',
	123	'ext': 'mp4',
	124	'title': 'BOSTON: WHERE THE PAST IS THE PRESENT',
	125	'description': 'md5:4ec8844ac262ca2df9e67c0983c6b83f',
	126	'uploader_id': 'seeso',
	127	},
	128	'params': {
	129	# m3u8 download
	130	'skip_download': True,
	131	},
1fa88937 RA	132	}, {
	133	# movie listing
	134	'url': 'https://vrv.co/watch/G6NQXZ1J6/Lily-CAT',
	135	'info_dict': {
	136	'id': 'G6NQXZ1J6',
	137	'title': 'Lily C.A.T',
	138	'description': 'md5:988b031e7809a6aeb60968be4af7db07',
	139	},
	140	'playlist_count': 2,
54a5be4d	141	}]
4b85f0f9 RA	142	_NETRC_MACHINE = 'vrv'
4b85f0f9 RA	143
52efa4b3	144	def _perform_login(self, username, password):
4b85f0f9 RA	145	token_credentials = self._call_api(
4b85f0f9 RA	146	'authenticate/by:credentials', None, 'Token Credentials', data={
52efa4b3	147	'email': username,
4b85f0f9 RA	148	'password': password,
	149	})
	150	self._TOKEN = token_credentials['oauth_token']
	151	self._TOKEN_SECRET = token_credentials['oauth_token_secret']
54a5be4d	152
11078c6d	153	def _initialize_pre_login(self):
11078c6d	154	return self._set_api_params()
77c8ebe6 RA	155
	156	def _real_extract(self, url):
	157	video_id = self._match_id(url)
4b85f0f9	158
9f182c23 RA	159	object_data = self._call_cms(self._get_cms_resource(
	160	'cms:/objects/' + video_id, video_id), video_id, 'object')['items'][0]
	161	resource_path = object_data['__links__']['resource']['href']
	162	video_data = self._call_cms(resource_path, video_id, 'video')
77c8ebe6	163	title = video_data['title']
1fa88937 RA	164	description = video_data.get('description')
	165
	166	if video_data.get('__class__') == 'movie_listing':
	167	items = self._call_cms(
	168	video_data['__links__']['movie_listing/movies']['href'],
	169	video_id, 'movie listing').get('items') or []
	170	if len(items) != 1:
	171	entries = []
	172	for item in items:
	173	item_id = item.get('id')
	174	if not item_id:
	175	continue
	176	entries.append(self.url_result(
	177	'https://vrv.co/watch/' + item_id,
	178	self.ie_key(), item_id, item.get('title')))
	179	return self.playlist_result(entries, video_id, title, description)
	180	video_data = items[0]
77c8ebe6	181
4b85f0f9 RA	182	streams_path = video_data['__links__'].get('streams', {}).get('href')
	183	if not streams_path:
	184	self.raise_login_required()
	185	streams_json = self._call_cms(streams_path, video_id, 'streams')
77c8ebe6 RA	186
	187	audio_locale = streams_json.get('audio_locale')
	188	formats = []
afa0200b RA	189	for stream_type, streams in streams_json.get('streams', {}).items():
	190	if stream_type in ('adaptive_hls', 'adaptive_dash'):
	191	for stream in streams.values():
54a5be4d RA	192	formats.extend(self._extract_vrv_formats(
	193	stream.get('url'), video_id, stream_type.split('_')[1],
	194	audio_locale, stream.get('hardsub_locale')))
77c8ebe6	195
afa0200b	196	subtitles = {}
58317428 RA	197	for k in ('captions', 'subtitles'):
	198	for subtitle in streams_json.get(k, {}).values():
	199	subtitle_url = subtitle.get('url')
	200	if not subtitle_url:
	201	continue
	202	subtitles.setdefault(subtitle.get('locale', 'en-US'), []).append({
	203	'url': subtitle_url,
	204	'ext': subtitle.get('format', 'ass'),
	205	})
afa0200b	206
77c8ebe6	207	thumbnails = []
f7590d47	208	for thumbnail in traverse_obj(video_data, ('images', 'thumbnail', ..., ...)) or []:
77c8ebe6 RA	209	thumbnail_url = thumbnail.get('source')
	210	if not thumbnail_url:
	211	continue
	212	thumbnails.append({
	213	'url': thumbnail_url,
	214	'width': int_or_none(thumbnail.get('width')),
	215	'height': int_or_none(thumbnail.get('height')),
	216	})
	217
	218	return {
	219	'id': video_id,
	220	'title': title,
	221	'formats': formats,
afa0200b	222	'subtitles': subtitles,
77c8ebe6	223	'thumbnails': thumbnails,
1fa88937	224	'description': description,
77c8ebe6 RA	225	'duration': float_or_none(video_data.get('duration_ms'), 1000),
	226	'uploader_id': video_data.get('channel_id'),
	227	'series': video_data.get('series_title'),
	228	'season': video_data.get('season_title'),
	229	'season_number': int_or_none(video_data.get('season_number')),
	230	'season_id': video_data.get('season_id'),
	231	'episode': title,
	232	'episode_number': int_or_none(video_data.get('episode_number')),
	233	'episode_id': video_data.get('production_episode_id'),
	234	}
48ab554f RA	235
	236
	237	class VRVSeriesIE(VRVBaseIE):
	238	IE_NAME = 'vrv:series'
	239	_VALID_URL = r'https?://(?:www\.)?vrv\.co/series/(?P<id>[A-Z0-9]+)'
	240	_TEST = {
	241	'url': 'https://vrv.co/series/G68VXG3G6/The-Perfect-Insider',
	242	'info_dict': {
	243	'id': 'G68VXG3G6',
	244	},
	245	'playlist_mincount': 11,
	246	}
	247
11078c6d	248	def _initialize_pre_login(self):
	249	return self._set_api_params()
	250
48ab554f RA	251	def _real_extract(self, url):
48ab554f RA	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)