1
0
mirror of https://github.com/yt-dlp/yt-dlp.git synced 2024-12-11 08:05:43 +01:00
yt-dlp/yt_dlp/extractor/sonyliv.py

241 lines
10 KiB
Python
Raw Normal View History

import datetime as dt
import itertools
import json
import math
import random
2021-01-01 13:26:37 +01:00
import time
import uuid
2016-08-08 01:11:00 +02:00
from .common import InfoExtractor
from ..networking.exceptions import HTTPError
2021-01-01 13:26:37 +01:00
from ..utils import (
ExtractorError,
int_or_none,
jwt_decode_hs256,
try_call,
2021-01-01 13:26:37 +01:00
)
from ..utils.traversal import traverse_obj
2016-08-08 01:11:00 +02:00
class SonyLIVIE(InfoExtractor):
_VALID_URL = r'''(?x)
(?:
sonyliv:|
https?://(?:www\.)?sonyliv\.com/(?:s(?:how|port)s/[^/]+|movies|clip|trailer|music-videos)/[^/?#&]+-
)
(?P<id>\d+)
'''
2016-08-08 21:16:28 +02:00
_TESTS = [{
2021-01-01 13:26:37 +01:00
'url': 'https://www.sonyliv.com/shows/bachelors-delight-1700000113/achaari-cheese-toast-1000022678?watch=true',
2016-08-08 01:11:00 +02:00
'info_dict': {
'title': 'Achaari Cheese Toast',
2021-01-01 13:26:37 +01:00
'id': '1000022678',
2016-08-08 01:11:00 +02:00
'ext': 'mp4',
2021-01-01 13:26:37 +01:00
'upload_date': '20200411',
'description': 'md5:3957fa31d9309bf336ceb3f37ad5b7cb',
'timestamp': 1586632091,
'duration': 185,
'season_number': 1,
'series': 'Bachelors Delight',
2021-01-01 13:26:37 +01:00
'episode_number': 1,
'release_year': 2016,
2016-08-08 01:11:00 +02:00
},
'params': {
'skip_download': True,
},
2016-08-08 21:16:28 +02:00
}, {
2021-01-01 13:26:37 +01:00
'url': 'https://www.sonyliv.com/movies/tahalka-1000050121?watch=true',
'only_matching': True,
}, {
'url': 'https://www.sonyliv.com/clip/jigarbaaz-1000098925',
'only_matching': True,
}, {
'url': 'https://www.sonyliv.com/trailer/sandwiched-forever-1000100286?watch=true',
'only_matching': True,
}, {
'url': 'https://www.sonyliv.com/sports/india-tour-of-australia-2020-21-1700000286/cricket-hls-day-3-1st-test-aus-vs-ind-19-dec-2020-1000100959?watch=true',
'only_matching': True,
}, {
'url': 'https://www.sonyliv.com/music-videos/yeh-un-dinon-ki-baat-hai-1000018779',
2016-08-08 21:16:28 +02:00
'only_matching': True,
}]
2021-01-01 13:26:37 +01:00
_GEO_COUNTRIES = ['IN']
_HEADERS = {}
_LOGIN_HINT = 'Use "--username <mobile_number>" to login using OTP or "--username token --password <auth_token>" to login using auth token.'
_NETRC_MACHINE = 'sonyliv'
def _get_device_id(self):
e = int(time.time() * 1000)
t = list('xxxxxxxxxxxx4xxxyxxxxxxxxxxxxxxx')
for i, c in enumerate(t):
n = int((e + 16 * random.random()) % 16) | 0
e = math.floor(e / 16)
if c == 'x':
t[i] = str(n)
elif c == 'y':
t[i] = f'{3 & n | 8:x}'
return ''.join(t) + '-' + str(int(time.time() * 1000))
def _perform_login(self, username, password):
self._HEADERS['device_id'] = self._get_device_id()
self._HEADERS['content-type'] = 'application/json'
if username.lower() == 'token' and try_call(lambda: jwt_decode_hs256(password)):
self._HEADERS['authorization'] = password
self.report_login()
return
elif len(username) != 10 or not username.isdigit():
raise ExtractorError(f'Invalid username/password; {self._LOGIN_HINT}')
self.report_login()
otp_request_json = self._download_json(
'https://apiv2.sonyliv.com/AGL/1.6/A/ENG/WEB/IN/HR/CREATEOTP-V2',
None, note='Sending OTP', headers=self._HEADERS, data=json.dumps({
'mobileNumber': username,
'channelPartnerID': 'MSMIND',
'country': 'IN',
'timestamp': dt.datetime.now().strftime('%Y-%m-%dT%H:%M:%S.%MZ'),
'otpSize': 6,
'loginType': 'REGISTERORSIGNIN',
'isMobileMandatory': True,
}).encode())
if otp_request_json['resultCode'] == 'KO':
raise ExtractorError(otp_request_json['message'], expected=True)
otp_verify_json = self._download_json(
'https://apiv2.sonyliv.com/AGL/2.0/A/ENG/WEB/IN/HR/CONFIRMOTP-V2',
None, note='Verifying OTP', headers=self._HEADERS, data=json.dumps({
'channelPartnerID': 'MSMIND',
'mobileNumber': username,
'country': 'IN',
'otp': self._get_tfa_info('OTP'),
'dmaId': 'IN',
'ageConfirmation': True,
'timestamp': dt.datetime.now().strftime('%Y-%m-%dT%H:%M:%S.%MZ'),
'isMobileMandatory': True,
}).encode())
if otp_verify_json['resultCode'] == 'KO':
raise ExtractorError(otp_request_json['message'], expected=True)
self._HEADERS['authorization'] = otp_verify_json['resultObj']['accessToken']
2016-08-08 21:16:28 +02:00
2021-01-01 13:26:37 +01:00
def _call_api(self, version, path, video_id):
try:
return self._download_json(
f'https://apiv2.sonyliv.com/AGL/{version}/A/ENG/WEB/{path}',
video_id, headers=self._HEADERS)['resultObj']
2021-01-01 13:26:37 +01:00
except ExtractorError as e:
if isinstance(e.cause, HTTPError) and e.cause.status == 406 and self._parse_json(
e.cause.response.read().decode(), video_id)['message'] == 'Please subscribe to watch this content':
self.raise_login_required(self._LOGIN_HINT, method=None)
if isinstance(e.cause, HTTPError) and e.cause.status == 403:
2021-01-01 13:26:37 +01:00
message = self._parse_json(
e.cause.response.read().decode(), video_id)['message']
2021-01-01 13:26:37 +01:00
if message == 'Geoblocked Country':
self.raise_geo_restricted(countries=self._GEO_COUNTRIES)
raise ExtractorError(message)
raise
def _initialize_pre_login(self):
self._HEADERS['security_token'] = self._call_api('1.4', 'ALL/GETTOKEN', None)
2016-08-08 01:11:00 +02:00
def _real_extract(self, url):
2021-01-01 13:26:37 +01:00
video_id = self._match_id(url)
content = self._call_api(
'1.5', 'IN/CONTENT/VIDEOURL/VOD/' + video_id, video_id)
if not self.get_param('allow_unplayable_formats') and content.get('isEncrypted'):
self.report_drm(video_id)
2021-01-01 13:26:37 +01:00
dash_url = content['videoURL']
headers = {
'x-playback-session-id': '%s-%d' % (uuid.uuid4().hex, time.time() * 1000),
2021-01-01 13:26:37 +01:00
}
formats = self._extract_mpd_formats(
dash_url, video_id, mpd_id='dash', headers=headers, fatal=False)
formats.extend(self._extract_m3u8_formats(
dash_url.replace('.mpd', '.m3u8').replace('/DASH/', '/HLS/'),
video_id, 'mp4', m3u8_id='hls', headers=headers, fatal=False))
for f in formats:
f.setdefault('http_headers', {}).update(headers)
metadata = self._call_api(
'1.6', 'IN/DETAIL/' + video_id, video_id)['containers'][0]['metadata']
title = metadata['episodeTitle']
subtitles = {}
for sub in content.get('subtitle', []):
sub_url = sub.get('subtitleUrl')
if not sub_url:
continue
subtitles.setdefault(sub.get('subtitleLanguageName', 'ENG'), []).append({
'url': sub_url,
})
2021-01-01 13:26:37 +01:00
return {
'id': video_id,
'title': title,
'formats': formats,
'thumbnail': content.get('posterURL'),
'description': metadata.get('longDescription') or metadata.get('shortDescription'),
'timestamp': int_or_none(metadata.get('creationDate'), 1000),
'duration': int_or_none(metadata.get('duration')),
'season_number': int_or_none(metadata.get('season')),
'series': metadata.get('title'),
2021-01-01 13:26:37 +01:00
'episode_number': int_or_none(metadata.get('episodeNumber')),
'release_year': int_or_none(metadata.get('year')),
'subtitles': subtitles,
2021-01-01 13:26:37 +01:00
}
class SonyLIVSeriesIE(InfoExtractor):
_VALID_URL = r'https?://(?:www\.)?sonyliv\.com/shows/[^/?#&]+-(?P<id>\d{10})/?(?:$|[?#])'
_TESTS = [{
'url': 'https://www.sonyliv.com/shows/adaalat-1700000091',
'playlist_mincount': 452,
'info_dict': {
'id': '1700000091',
},
}, {
'url': 'https://www.sonyliv.com/shows/beyhadh-1700000007/',
'playlist_mincount': 358,
'info_dict': {
'id': '1700000007',
},
}]
_API_BASE = 'https://apiv2.sonyliv.com/AGL'
def _entries(self, show_id):
headers = {
'Accept': 'application/json, text/plain, */*',
'Referer': 'https://www.sonyliv.com',
}
headers['security_token'] = self._download_json(
f'{self._API_BASE}/1.4/A/ENG/WEB/ALL/GETTOKEN', show_id,
'Downloading security token', headers=headers)['resultObj']
seasons = traverse_obj(self._download_json(
f'{self._API_BASE}/1.9/R/ENG/WEB/IN/DL/DETAIL/{show_id}', show_id,
'Downloading series JSON', headers=headers, query={
'kids_safe': 'false',
'from': '0',
'to': '49',
}), ('resultObj', 'containers', 0, 'containers', lambda _, v: int_or_none(v['id'])))
for season in seasons:
season_id = str(season['id'])
note = traverse_obj(season, ('metadata', 'title', {str})) or 'season'
cursor = 0
for page_num in itertools.count(1):
episodes = traverse_obj(self._download_json(
f'{self._API_BASE}/1.4/R/ENG/WEB/IN/CONTENT/DETAIL/BUNDLE/{season_id}',
season_id, f'Downloading {note} page {page_num} JSON', headers=headers, query={
'from': str(cursor),
'to': str(cursor + 99),
'orderBy': 'episodeNumber',
'sortOrder': 'asc',
}), ('resultObj', 'containers', 0, 'containers', lambda _, v: int_or_none(v['id'])))
if not episodes:
break
for episode in episodes:
video_id = str(episode['id'])
yield self.url_result(f'sonyliv:{video_id}', SonyLIVIE, video_id)
cursor += 100
def _real_extract(self, url):
show_id = self._match_id(url)
return self.playlist_result(self._entries(show_id), playlist_id=show_id)