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

# coding: utf-8
from __future__ import unicode_literals

import hashlib
import hmac
import re
import time
import uuid
import json

from .common import InfoExtractor
from ..compat import (
    compat_HTTPError,
    compat_str
)
from ..utils import (
    determine_ext,
    ExtractorError,
    int_or_none,
    str_or_none,
    try_get,
    url_or_none,
)


class HotStarBaseIE(InfoExtractor):
    _AKAMAI_ENCRYPTION_KEY = b'\x05\xfc\x1a\x01\xca\xc9\x4b\xc4\x12\xfc\x53\x12\x07\x75\xf9\xee'

    def _call_api_impl(self, path, video_id, query, st=None, cookies=None):
        st = int_or_none(st) or int(time.time())
        exp = st + 6000
        auth = 'st=%d~exp=%d~acl=/*' % (st, exp)
        auth += '~hmac=' + hmac.new(self._AKAMAI_ENCRYPTION_KEY, auth.encode(), hashlib.sha256).hexdigest()

        if cookies.get('userUP'):
            token = cookies.get('userUP').value
        else:
            token = self._download_json(
                'https://api.hotstar.com/um/v3/users',
                video_id, note='Downloading token',
                data=json.dumps({"device_ids": [{"id": compat_str(uuid.uuid4()), "type": "device_id"}]}).encode('utf-8'),
                headers={
                    'hotstarauth': auth,
                    'x-hs-platform': 'PCTV',  # or 'web'
                    'Content-Type': 'application/json',
                })['user_identity']

        response = self._download_json(
            'https://api.hotstar.com/' + path, video_id, headers={
                'hotstarauth': auth,
                'x-hs-appversion': '6.72.2',
                'x-hs-platform': 'web',
                'x-hs-usertoken': token,
            }, query=query)

        if response['message'] != "Playback URL's fetched successfully":
            raise ExtractorError(
                response['message'], expected=True)
        return response['data']

    def _call_api(self, path, video_id, query_name='contentId'):
        return self._call_api_impl(path, video_id, {
            query_name: video_id,
            'tas': 10000,
        })

    def _call_api_v2(self, path, video_id, st=None, cookies=None):
        return self._call_api_impl(
            '%s/content/%s' % (path, video_id), video_id, st=st, cookies=cookies, query={
                'desired-config': 'audio_channel:stereo|dynamic_range:sdr|encryption:plain|ladder:tv|package:dash|resolution:hd|subs-tag:HotstarVIP|video_codec:vp9',
                'device-id': cookies.get('device_id').value if cookies.get('device_id') else compat_str(uuid.uuid4()),
                'os-name': 'Windows',
                'os-version': '10',
            })


class HotStarIE(HotStarBaseIE):
    IE_NAME = 'hotstar'
    _VALID_URL = r'''(?x)
                           https?://(?:www\.)?hotstar\.com(?:/in)?/(?!in/)
                           (?:
                               tv/(?:[^/?#]+/){3}|
                               (?!tv/)[^?#]+/
                           )?
                           (?P<id>\d{10})
                   '''
    _TESTS = [{
        # contentData
        'url': 'https://www.hotstar.com/can-you-not-spread-rumours/1000076273',
        'info_dict': {
            'id': '1000076273',
            'ext': 'mp4',
            'title': 'Can You Not Spread Rumours?',
            'description': 'md5:c957d8868e9bc793ccb813691cc4c434',
            'timestamp': 1447248600,
            'upload_date': '20151111',
            'duration': 381,
        },
        'params': {
            # m3u8 download
            'skip_download': True,
        }
    }, {
        # contentDetail
        'url': 'https://www.hotstar.com/movies/radha-gopalam/1000057157',
        'only_matching': True,
    }, {
        'url': 'http://www.hotstar.com/sports/cricket/rajitha-sizzles-on-debut-with-329/2001477583',
        'only_matching': True,
    }, {
        'url': 'http://www.hotstar.com/1000000515',
        'only_matching': True,
    }, {
        # only available via api v2
        'url': 'https://www.hotstar.com/tv/ek-bhram-sarvagun-sampanna/s-2116/janhvi-targets-suman/1000234847',
        'only_matching': True,
    }]
    _GEO_BYPASS = False

    def _real_extract(self, url):
        video_id = self._match_id(url)
        cookies = self._get_cookies(url)
        webpage, urlh = self._download_webpage_handle(url, video_id)
        st = urlh.headers.get('x-origin-date')
        app_state = self._parse_json(self._search_regex(
            r'<script>window\.APP_STATE\s*=\s*({.+?})</script>',
            webpage, 'app state'), video_id)
        video_data = {}
        getters = list(
            lambda x, k=k: x['initialState']['content%s' % k]['content']
            for k in ('Data', 'Detail')
        )
        for v in app_state.values():
            content = try_get(v, getters, dict)
            if content and content.get('contentId') == video_id:
                video_data = content
                break

        title = video_data['title']

        if not self.get_param('allow_unplayable_formats') and video_data.get('drmProtected'):
            raise ExtractorError('This video is DRM protected.', expected=True)

        headers = {'Referer': url}
        formats = []
        geo_restricted = False
        # change to v2 in the future
        playback_sets = self._call_api_v2('play/v1/playback', video_id, st=st, cookies=cookies)['playBackSets']
        for playback_set in playback_sets:
            if not isinstance(playback_set, dict):
                continue
            format_url = url_or_none(playback_set.get('playbackUrl'))
            if not format_url:
                continue
            format_url = re.sub(
                r'(?<=//staragvod)(\d)', r'web\1', format_url)
            tags = str_or_none(playback_set.get('tagsCombination')) or ''
            if tags and 'encryption:plain' not in tags:
                continue
            ext = determine_ext(format_url)
            try:
                if 'package:hls' in tags or ext == 'm3u8':
                    formats.extend(self._extract_m3u8_formats(
                        format_url, video_id, 'mp4',
                        entry_protocol='m3u8_native',
                        m3u8_id='hls', headers=headers))
                elif 'package:dash' in tags or ext == 'mpd':
                    formats.extend(self._extract_mpd_formats(
                        format_url, video_id, mpd_id='dash', headers=headers))
                elif ext == 'f4m':
                    # produce broken files
                    pass
                else:
                    formats.append({
                        'url': format_url,
                        'width': int_or_none(playback_set.get('width')),
                        'height': int_or_none(playback_set.get('height')),
                    })
            except ExtractorError as e:
                if isinstance(e.cause, compat_HTTPError) and e.cause.code == 403:
                    geo_restricted = True
                continue
        if not formats and geo_restricted:
            self.raise_geo_restricted(countries=['IN'], metadata_available=True)
        self._sort_formats(formats)

        for f in formats:
            f.setdefault('http_headers', {}).update(headers)

        return {
            'id': video_id,
            'title': title,
            'description': video_data.get('description'),
            'duration': int_or_none(video_data.get('duration')),
            'timestamp': int_or_none(video_data.get('broadcastDate') or video_data.get('startDate')),
            'formats': formats,
            'channel': video_data.get('channelName'),
            'channel_id': video_data.get('channelId'),
            'series': video_data.get('showName'),
            'season': video_data.get('seasonName'),
            'season_number': int_or_none(video_data.get('seasonNo')),
            'season_id': video_data.get('seasonId'),
            'episode': title,
            'episode_number': int_or_none(video_data.get('episodeNo')),
        }


class HotStarPlaylistIE(HotStarBaseIE):
    IE_NAME = 'hotstar:playlist'
    _VALID_URL = r'https?://(?:www\.)?hotstar\.com/tv/[^/]+/s-\w+/list/[^/]+/t-(?P<id>\w+)'
    _TESTS = [{
        'url': 'https://www.hotstar.com/tv/savdhaan-india/s-26/list/popular-clips/t-3_2_26',
        'info_dict': {
            'id': '3_2_26',
        },
        'playlist_mincount': 20,
    }, {
        'url': 'https://www.hotstar.com/tv/savdhaan-india/s-26/list/extras/t-2480',
        'only_matching': True,
    }]

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

        collection = self._call_api('o/v1/tray/find', playlist_id, 'uqId')

        entries = [
            self.url_result(
                'https://www.hotstar.com/%s' % video['contentId'],
                ie=HotStarIE.ie_key(), video_id=video['contentId'])
            for video in collection['assets']['items']
            if video.get('contentId')]

        return self.playlist_result(entries, playlist_id)


class HotStarSeriesIE(HotStarBaseIE):
    IE_NAME = 'hotstar:series'
    _VALID_URL = r'(?:https?://)(?:www\.)?hotstar\.com(?:/in)?/tv/[^/]+/(?P<id>\d+)'
    _TESTS = [{
        'url': 'https://www.hotstar.com/in/tv/radhakrishn/1260000646',
        'info_dict': {
            'id': '1260000646',
        },
        'playlist_mincount': 690,
    }, {
        'url': 'https://www.hotstar.com/tv/dancee-/1260050431',
        'info_dict': {
            'id': '1260050431',
        },
        'playlist_mincount': 43,
    }, {
        'url': 'https://www.hotstar.com/in/tv/mahabharat/435/',
        'info_dict': {
            'id': '435',
        },
        'playlist_mincount': 269,
    }]

    def _real_extract(self, url):
        series_id = self._match_id(url)
        headers = {
            'x-country-code': 'IN',
            'x-platform-code': 'PCTV',
        }
        detail_json = self._download_json('https://api.hotstar.com/o/v1/show/detail?contentId=' + series_id,
                                          video_id=series_id, headers=headers)
        id = compat_str(try_get(detail_json, lambda x: x['body']['results']['item']['id'], int))
        item_json = self._download_json('https://api.hotstar.com/o/v1/tray/g/1/items?etid=0&tao=0&tas=10000&eid=' + id,
                                        video_id=series_id, headers=headers)
        entries = [
            self.url_result(
                'https://www.hotstar.com/%d' % video['contentId'],
                ie=HotStarIE.ie_key(), video_id=video['contentId'])
            for video in item_json['body']['results']['items']
            if video.get('contentId')]

        return self.playlist_result(entries, series_id)
Commit	Line	Data
fb8e402a	1	# coding: utf-8
	2	from __future__ import unicode_literals
	3
85cd69ad RA	4	import hashlib
85cd69ad RA	5	import hmac
1cb812d3	6	import re
85cd69ad	7	import time
2533f5b6	8	import uuid
7078ec64	9	import json
909191de	10
fb8e402a	11	from .common import InfoExtractor
2533f5b6 S	12	from ..compat import (
2533f5b6 S	13	compat_HTTPError,
6923b538	14	compat_str
2533f5b6	15	)
fb8e402a	16	from ..utils import (
fb8e402a	17	determine_ext,
909191de	18	ExtractorError,
fb8e402a	19	int_or_none,
2533f5b6	20	str_or_none,
05e7c184	21	try_get,
2533f5b6	22	url_or_none,
fb8e402a	23	)
	24
	25
909191de	26	class HotStarBaseIE(InfoExtractor):
85cd69ad RA	27	_AKAMAI_ENCRYPTION_KEY = b'\x05\xfc\x1a\x01\xca\xc9\x4b\xc4\x12\xfc\x53\x12\x07\x75\xf9\xee'
85cd69ad RA	28
fe07e2c6	29	def _call_api_impl(self, path, video_id, query, st=None, cookies=None):
9fc0de57	30	st = int_or_none(st) or int(time.time())
85cd69ad RA	31	exp = st + 6000
	32	auth = 'st=%d~exp=%d~acl=/*' % (st, exp)
	33	auth += '~hmac=' + hmac.new(self._AKAMAI_ENCRYPTION_KEY, auth.encode(), hashlib.sha256).hexdigest()
6923b538	34
fe07e2c6 A	35	if cookies.get('userUP'):
	36	token = cookies.get('userUP').value
	37	else:
	38	token = self._download_json(
	39	'https://api.hotstar.com/um/v3/users',
	40	video_id, note='Downloading token',
	41	data=json.dumps({"device_ids": [{"id": compat_str(uuid.uuid4()), "type": "device_id"}]}).encode('utf-8'),
	42	headers={
	43	'hotstarauth': auth,
	44	'x-hs-platform': 'PCTV', # or 'web'
	45	'Content-Type': 'application/json',
	46	})['user_identity']
6923b538	47
85cd69ad	48	response = self._download_json(
2533f5b6	49	'https://api.hotstar.com/' + path, video_id, headers={
85cd69ad	50	'hotstarauth': auth,
7078ec64 N	51	'x-hs-appversion': '6.72.2',
	52	'x-hs-platform': 'web',
	53	'x-hs-usertoken': token,
2533f5b6	54	}, query=query)
6923b538	55
7078ec64	56	if response['message'] != "Playback URL's fetched successfully":
85cd69ad	57	raise ExtractorError(
7078ec64 N	58	response['message'], expected=True)
7078ec64 N	59	return response['data']
909191de	60
2533f5b6 S	61	def _call_api(self, path, video_id, query_name='contentId'):
	62	return self._call_api_impl(path, video_id, {
	63	query_name: video_id,
	64	'tas': 10000,
	65	})
	66
fe07e2c6	67	def _call_api_v2(self, path, video_id, st=None, cookies=None):
2533f5b6	68	return self._call_api_impl(
fe07e2c6	69	'%s/content/%s' % (path, video_id), video_id, st=st, cookies=cookies, query={
6923b538	70	'desired-config': 'audio_channel:stereo\|dynamic_range:sdr\|encryption:plain\|ladder:tv\|package:dash\|resolution:hd\|subs-tag:HotstarVIP\|video_codec:vp9',
fe07e2c6	71	'device-id': cookies.get('device_id').value if cookies.get('device_id') else compat_str(uuid.uuid4()),
7078ec64 N	72	'os-name': 'Windows',
7078ec64 N	73	'os-version': '10',
2533f5b6 S	74	})
2533f5b6 S	75
909191de S	76
909191de S	77	class HotStarIE(HotStarBaseIE):
85cd69ad	78	IE_NAME = 'hotstar'
6e639032 A	79	_VALID_URL = r'''(?x)
	80	https?://(?:www\.)?hotstar\.com(?:/in)?/(?!in/)
	81	(?:
	82	tv/(?:[^/?#]+/){3}\|
	83	(?!tv/)[^?#]+/
	84	)?
	85	(?P<id>\d{10})
	86	'''
89d23f37	87	_TESTS = [{
adbbdefc	88	# contentData
85cd69ad	89	'url': 'https://www.hotstar.com/can-you-not-spread-rumours/1000076273',
fb8e402a	90	'info_dict': {
	91	'id': '1000076273',
	92	'ext': 'mp4',
85cd69ad	93	'title': 'Can You Not Spread Rumours?',
fb8e402a	94	'description': 'md5:c957d8868e9bc793ccb813691cc4c434',
85cd69ad	95	'timestamp': 1447248600,
fb8e402a	96	'upload_date': '20151111',
	97	'duration': 381,
	98	},
	99	'params': {
	100	# m3u8 download
	101	'skip_download': True,
	102	}
adbbdefc S	103	}, {
	104	# contentDetail
	105	'url': 'https://www.hotstar.com/movies/radha-gopalam/1000057157',
	106	'only_matching': True,
89d23f37 S	107	}, {
	108	'url': 'http://www.hotstar.com/sports/cricket/rajitha-sizzles-on-debut-with-329/2001477583',
	109	'only_matching': True,
	110	}, {
	111	'url': 'http://www.hotstar.com/1000000515',
	112	'only_matching': True,
2533f5b6 S	113	}, {
	114	# only available via api v2
	115	'url': 'https://www.hotstar.com/tv/ek-bhram-sarvagun-sampanna/s-2116/janhvi-targets-suman/1000234847',
	116	'only_matching': True,
89d23f37	117	}]
85cd69ad	118	_GEO_BYPASS = False
fb8e402a	119
fb8e402a	120	def _real_extract(self, url):
fb8e402a	121	video_id = self._match_id(url)
fe07e2c6	122	cookies = self._get_cookies(url)
9fc0de57	123	webpage, urlh = self._download_webpage_handle(url, video_id)
9fc0de57	124	st = urlh.headers.get('x-origin-date')
85cd69ad RA	125	app_state = self._parse_json(self._search_regex(
	126	r'<script>window\.APP_STATE\s=\s({.+?})</script>',
	127	webpage, 'app state'), video_id)
05e7c184	128	video_data = {}
c3c098dc	129	getters = list(
adbbdefc S	130	lambda x, k=k: x['initialState']['content%s' % k]['content']
	131	for k in ('Data', 'Detail')
	132	)
05e7c184	133	for v in app_state.values():
adbbdefc	134	content = try_get(v, getters, dict)
05e7c184 RA	135	if content and content.get('contentId') == video_id:
05e7c184 RA	136	video_data = content
c3c098dc	137	break
909191de	138
85cd69ad	139	title = video_data['title']
0dac7cbb	140
a06916d9	141	if not self.get_param('allow_unplayable_formats') and video_data.get('drmProtected'):
0dac7cbb	142	raise ExtractorError('This video is DRM protected.', expected=True)
fb8e402a	143
d7def23d	144	headers = {'Referer': url}
fb8e402a	145	formats = []
2533f5b6	146	geo_restricted = False
6923b538	147	# change to v2 in the future
fe07e2c6	148	playback_sets = self._call_api_v2('play/v1/playback', video_id, st=st, cookies=cookies)['playBackSets']
2533f5b6 S	149	for playback_set in playback_sets:
	150	if not isinstance(playback_set, dict):
	151	continue
	152	format_url = url_or_none(playback_set.get('playbackUrl'))
	153	if not format_url:
	154	continue
1cb812d3 S	155	format_url = re.sub(
1cb812d3 S	156	r'(?<=//staragvod)(\d)', r'web\1', format_url)
2533f5b6 S	157	tags = str_or_none(playback_set.get('tagsCombination')) or ''
	158	if tags and 'encryption:plain' not in tags:
	159	continue
	160	ext = determine_ext(format_url)
85cd69ad	161	try:
2533f5b6 S	162	if 'package:hls' in tags or ext == 'm3u8':
2533f5b6 S	163	formats.extend(self._extract_m3u8_formats(
d9d30986	164	format_url, video_id, 'mp4',
d7def23d RA	165	entry_protocol='m3u8_native',
d7def23d RA	166	m3u8_id='hls', headers=headers))
2533f5b6 S	167	elif 'package:dash' in tags or ext == 'mpd':
2533f5b6 S	168	formats.extend(self._extract_mpd_formats(
d7def23d	169	format_url, video_id, mpd_id='dash', headers=headers))
2533f5b6 S	170	elif ext == 'f4m':
	171	# produce broken files
	172	pass
	173	else:
	174	formats.append({
	175	'url': format_url,
	176	'width': int_or_none(playback_set.get('width')),
	177	'height': int_or_none(playback_set.get('height')),
	178	})
85cd69ad RA	179	except ExtractorError as e:
85cd69ad RA	180	if isinstance(e.cause, compat_HTTPError) and e.cause.code == 403:
2533f5b6 S	181	geo_restricted = True
	182	continue
	183	if not formats and geo_restricted:
b7da73eb	184	self.raise_geo_restricted(countries=['IN'], metadata_available=True)
fb8e402a	185	self._sort_formats(formats)
fb8e402a	186
d7def23d RA	187	for f in formats:
	188	f.setdefault('http_headers', {}).update(headers)
	189
fb8e402a	190	return {
fb8e402a	191	'id': video_id,
0dac7cbb	192	'title': title,
fb8e402a	193	'description': video_data.get('description'),
fb8e402a	194	'duration': int_or_none(video_data.get('duration')),
85cd69ad	195	'timestamp': int_or_none(video_data.get('broadcastDate') or video_data.get('startDate')),
fb8e402a	196	'formats': formats,
85cd69ad RA	197	'channel': video_data.get('channelName'),
	198	'channel_id': video_data.get('channelId'),
	199	'series': video_data.get('showName'),
	200	'season': video_data.get('seasonName'),
	201	'season_number': int_or_none(video_data.get('seasonNo')),
	202	'season_id': video_data.get('seasonId'),
0dac7cbb	203	'episode': title,
85cd69ad	204	'episode_number': int_or_none(video_data.get('episodeNo')),
fb8e402a	205	}
477c97f8 AV	206
477c97f8 AV	207
909191de	208	class HotStarPlaylistIE(HotStarBaseIE):
477c97f8	209	IE_NAME = 'hotstar:playlist'
85cd69ad	210	_VALID_URL = r'https?://(?:www\.)?hotstar\.com/tv/[^/]+/s-\w+/list/[^/]+/t-(?P<id>\w+)'
477c97f8	211	_TESTS = [{
85cd69ad	212	'url': 'https://www.hotstar.com/tv/savdhaan-india/s-26/list/popular-clips/t-3_2_26',
477c97f8	213	'info_dict': {
85cd69ad	214	'id': '3_2_26',
477c97f8	215	},
85cd69ad	216	'playlist_mincount': 20,
477c97f8	217	}, {
85cd69ad	218	'url': 'https://www.hotstar.com/tv/savdhaan-india/s-26/list/extras/t-2480',
477c97f8 AV	219	'only_matching': True,
477c97f8 AV	220	}]
477c97f8 AV	221
477c97f8 AV	222	def _real_extract(self, url):
85cd69ad RA	223	playlist_id = self._match_id(url)
	224
	225	collection = self._call_api('o/v1/tray/find', playlist_id, 'uqId')
477c97f8	226
477c97f8	227	entries = [
909191de	228	self.url_result(
85cd69ad	229	'https://www.hotstar.com/%s' % video['contentId'],
909191de	230	ie=HotStarIE.ie_key(), video_id=video['contentId'])
85cd69ad	231	for video in collection['assets']['items']
909191de S	232	if video.get('contentId')]
	233
	234	return self.playlist_result(entries, playlist_id)
6e639032 A	235
	236
	237	class HotStarSeriesIE(HotStarBaseIE):
	238	IE_NAME = 'hotstar:series'
8242bf22	239	_VALID_URL = r'(?:https?://)(?:www\.)?hotstar\.com(?:/in)?/tv/[^/]+/(?P<id>\d+)'
6e639032 A	240	_TESTS = [{
	241	'url': 'https://www.hotstar.com/in/tv/radhakrishn/1260000646',
	242	'info_dict': {
	243	'id': '1260000646',
	244	},
	245	'playlist_mincount': 690,
	246	}, {
	247	'url': 'https://www.hotstar.com/tv/dancee-/1260050431',
	248	'info_dict': {
	249	'id': '1260050431',
	250	},
	251	'playlist_mincount': 43,
8242bf22 A	252	}, {
	253	'url': 'https://www.hotstar.com/in/tv/mahabharat/435/',
	254	'info_dict': {
	255	'id': '435',
	256	},
	257	'playlist_mincount': 269,
6e639032 A	258	}]
	259
	260	def _real_extract(self, url):
	261	series_id = self._match_id(url)
	262	headers = {
	263	'x-country-code': 'IN',
	264	'x-platform-code': 'PCTV',
	265	}
	266	detail_json = self._download_json('https://api.hotstar.com/o/v1/show/detail?contentId=' + series_id,
	267	video_id=series_id, headers=headers)
	268	id = compat_str(try_get(detail_json, lambda x: x['body']['results']['item']['id'], int))
	269	item_json = self._download_json('https://api.hotstar.com/o/v1/tray/g/1/items?etid=0&tao=0&tas=10000&eid=' + id,
	270	video_id=series_id, headers=headers)
	271	entries = [
	272	self.url_result(
	273	'https://www.hotstar.com/%d' % video['contentId'],
	274	ie=HotStarIE.ie_key(), video_id=video['contentId'])
	275	for video in item_json['body']['results']['items']
	276	if video.get('contentId')]
	277
	278	return self.playlist_result(entries, series_id)