-# coding: utf-8
-from __future__ import unicode_literals
-
import json
+import random
import re
+import urllib.parse
+import uuid
from .common import InfoExtractor
-from ..compat import (
- compat_kwargs,
- compat_str,
- compat_urlparse,
- compat_urllib_request,
-)
+from ..compat import compat_str
from ..utils import (
ExtractorError,
int_or_none,
- try_get,
+ remove_end,
smuggle_url,
+ strip_or_none,
+ traverse_obj,
+ try_get,
+ unified_timestamp,
unsmuggle_url,
+ url_or_none,
)
class ViuBaseIE(InfoExtractor):
- def _real_initialize(self):
- viu_auth_res = self._request_webpage(
- 'https://www.viu.com/api/apps/v2/authenticate', None,
- 'Requesting Viu auth', query={
- 'acct': 'test',
- 'appid': 'viu_desktop',
- 'fmt': 'json',
- 'iid': 'guest',
- 'languageid': 'default',
- 'platform': 'desktop',
- 'userid': 'guest',
- 'useridtype': 'guest',
- 'ver': '1.0'
- }, headers=self.geo_verification_headers())
- self._auth_token = viu_auth_res.info()['X-VIU-AUTH']
-
- def _call_api(self, path, *args, **kwargs):
- headers = self.geo_verification_headers()
- headers.update({
- 'X-VIU-AUTH': self._auth_token
- })
- headers.update(kwargs.get('headers', {}))
- kwargs['headers'] = headers
+ def _call_api(self, path, *args, headers={}, **kwargs):
response = self._download_json(
- 'https://www.viu.com/api/' + path, *args,
- **compat_kwargs(kwargs))['response']
+ f'https://www.viu.com/api/{path}', *args, **kwargs,
+ headers={**self.geo_verification_headers(), **headers})['response']
if response.get('status') != 'success':
- raise ExtractorError('%s said: %s' % (
- self.IE_NAME, response['message']), expected=True)
+ raise ExtractorError(f'{self.IE_NAME} said: {response["message"]}', expected=True)
return response
tdirforwhole = video_data.get('tdirforwhole')
# #EXT-X-BYTERANGE is not supported by native hls downloader
# and ffmpeg (#10955)
+ # FIXME: It is supported in yt-dlp
# hls_file = video_data.get('hlsfile')
hls_file = video_data.get('jwhlsfile')
if url_path and tdirforwhole and hls_file:
# r'(/hlsc_)[a-z]+(\d+\.m3u8)',
# r'\1whe\2', video_data['href'])
m3u8_url = video_data['href']
- formats = self._extract_m3u8_formats(m3u8_url, video_id, 'mp4')
- self._sort_formats(formats)
+ formats, subtitles = self._extract_m3u8_formats_and_subtitles(m3u8_url, video_id, 'mp4')
- subtitles = {}
for key, value in video_data.items():
mobj = re.match(r'^subtitle_(?P<lang>[^_]+)_(?P<ext>(vtt|srt))', key)
if not mobj:
},
'skip': 'Geo-restricted to Singapore',
}, {
- 'url': 'http://www.viu.com/ott/hk/zh-hk/vod/7123/%E5%A4%A7%E4%BA%BA%E5%A5%B3%E5%AD%90',
+ 'url': 'https://www.viu.com/ott/hk/zh-hk/vod/430078/%E7%AC%AC%E5%85%AD%E6%84%9F-3',
'info_dict': {
- 'id': '7123',
+ 'id': '430078',
'ext': 'mp4',
- 'title': '這就是我的生活之道',
- 'description': 'md5:4eb0d8b08cf04fcdc6bbbeb16043434f',
+ 'title': '大韓民國的1%',
+ 'description': 'md5:74d6db47ddd9ddb9c89a05739103ccdb',
+ 'episode_number': 1,
+ 'duration': 6614,
+ 'episode': '大韓民國的1%',
+ 'series': '第六感 3',
+ 'thumbnail': 'https://d2anahhhmp1ffz.cloudfront.net/1313295781/d2b14f48d008ef2f3a9200c98d8e9b63967b9cc2',
},
'params': {
'skip_download': 'm3u8 download',
},
'skip': 'Geo-restricted to Hong Kong',
}, {
- 'url': 'https://www.viu.com/ott/hk/zh-hk/vod/68776/%E6%99%82%E5%B0%9A%E5%AA%BD%E5%92%AA',
- 'playlist_count': 12,
+ 'url': 'https://www.viu.com/ott/hk/zh-hk/vod/444666/%E6%88%91%E7%9A%84%E5%AE%A4%E5%8F%8B%E6%98%AF%E4%B9%9D%E5%B0%BE%E7%8B%90',
+ 'playlist_count': 16,
'info_dict': {
- 'id': '3916',
- 'title': '時尚媽咪',
+ 'id': '23807',
+ 'title': '我的室友是九尾狐',
+ 'description': 'md5:b42c95f2b4a316cdd6ae14ca695f33b9',
},
'params': {
'skip_download': 'm3u8 download',
'zh-cn': 2,
'en-us': 3,
}
- _user_info = None
+
+ _user_token = None
+ _auth_codes = {}
def _detect_error(self, response):
- code = response.get('status', {}).get('code')
- if code > 0:
+ code = try_get(response, lambda x: x['status']['code'])
+ if code and code > 0:
message = try_get(response, lambda x: x['status']['message'])
- raise ExtractorError('%s said: %s (%s)' % (
- self.IE_NAME, message, code), expected=True)
- return response['data']
-
- def _raise_login_required(self):
- raise ExtractorError(
- 'This video requires login. '
- 'Specify --username and --password or --netrc (machine: %s) '
- 'to provide account credentials.' % self._NETRC_MACHINE,
- expected=True)
+ raise ExtractorError(f'{self.IE_NAME} said: {message} ({code})', expected=True)
+ return response.get('data') or {}
def _login(self, country_code, video_id):
- if not self._user_info:
+ if self._user_token is None:
username, password = self._get_login_info()
- if username is None or password is None:
+ if username is None:
return
+ headers = {
+ 'Authorization': f'Bearer {self._auth_codes[country_code]}',
+ 'Content-Type': 'application/json'
+ }
+ data = self._download_json(
+ 'https://api-gateway-global.viu.com/api/account/validate',
+ video_id, 'Validating email address', headers=headers,
+ data=json.dumps({
+ 'principal': username,
+ 'provider': 'email'
+ }).encode())
+ if not data.get('exists'):
+ raise ExtractorError('Invalid email address')
data = self._download_json(
- compat_urllib_request.Request(
- 'https://www.viu.com/ott/%s/index.php' % country_code, method='POST'),
- video_id, 'Logging in', errnote=False, fatal=False,
- query={'r': 'user/login'},
+ 'https://api-gateway-global.viu.com/api/auth/login',
+ video_id, 'Logging in', headers=headers,
data=json.dumps({
- 'username': username,
+ 'email': username,
'password': password,
- 'platform_flag_label': 'web',
+ 'provider': 'email',
}).encode())
- self._user_info = self._detect_error(data)['user']
-
- return self._user_info
+ self._detect_error(data)
+ self._user_token = data.get('identity')
+ # need to update with valid user's token else will throw an error again
+ self._auth_codes[country_code] = data.get('token')
+ return self._user_token
+
+ def _get_token(self, country_code, video_id):
+ rand = ''.join(random.choices('0123456789', k=10))
+ return self._download_json(
+ f'https://api-gateway-global.viu.com/api/auth/token?v={rand}000', video_id,
+ headers={'Content-Type': 'application/json'}, note='Getting bearer token',
+ data=json.dumps({
+ 'countryCode': country_code.upper(),
+ 'platform': 'browser',
+ 'platformFlagLabel': 'web',
+ 'language': 'en',
+ 'uuid': str(uuid.uuid4()),
+ 'carrierId': '0'
+ }).encode('utf-8'))['token']
def _real_extract(self, url):
url, idata = unsmuggle_url(url, {})
query['area_id'] = area_id
product_data = self._download_json(
- 'http://www.viu.com/ott/%s/index.php' % country_code, video_id,
+ f'http://www.viu.com/ott/{country_code}/index.php', video_id,
'Downloading video info', query=query)['data']
video_data = product_data.get('current_product')
if not video_data:
- raise ExtractorError('This video is not available in your region.', expected=True)
+ self.raise_geo_restricted()
series_id = video_data.get('series_id')
if self._yes_playlist(series_id, video_id, idata):
- series = product_data.get('series', {})
+ series = product_data.get('series') or {}
product = series.get('product')
if product:
entries = []
item_id = entry.get('product_id')
if not item_id:
continue
- item_id = compat_str(item_id)
entries.append(self.url_result(
- smuggle_url(
- 'http://www.viu.com/ott/%s/%s/vod/%s/' % (country_code, lang_code, item_id),
- {'force_noplaylist': True}), # prevent infinite recursion
- 'ViuOTT',
- item_id,
- entry.get('synopsis', '').strip()))
+ smuggle_url(f'http://www.viu.com/ott/{country_code}/{lang_code}/vod/{item_id}/',
+ {'force_noplaylist': True}),
+ ViuOTTIE, str(item_id), entry.get('synopsis', '').strip()))
return self.playlist_result(entries, series_id, series.get('name'), series.get('description'))
'ccs_product_id': video_data['ccs_product_id'],
'language_flag_id': self._LANGUAGE_FLAG.get(lang_code.lower()) or '3',
}
- headers = {
- 'Referer': url,
- 'Origin': url,
- }
- try:
+
+ def download_playback():
stream_data = self._download_json(
- 'https://d1k2us671qcoau.cloudfront.net/distribute_web_%s.php' % country_code,
- video_id, 'Downloading stream info', query=query, headers=headers)
- stream_data = self._detect_error(stream_data)['stream']
- except (ExtractorError, KeyError):
- stream_data = None
- if video_data.get('user_level', 0) > 0:
- user = self._login(country_code, video_id)
- if user:
- query['identity'] = user['identity']
- stream_data = self._download_json(
- 'https://d1k2us671qcoau.cloudfront.net/distribute_web_%s.php' % country_code,
- video_id, 'Downloading stream info', query=query, headers=headers)
- stream_data = self._detect_error(stream_data).get('stream')
- else:
- # preview is limited to 3min for non-members
- # try to bypass the duration limit
- duration_limit = True
- query['duration'] = '180'
- stream_data = self._download_json(
- 'https://d1k2us671qcoau.cloudfront.net/distribute_web_%s.php' % country_code,
- video_id, 'Downloading stream info', query=query, headers=headers)
- try:
- stream_data = self._detect_error(stream_data)['stream']
- except (ExtractorError, KeyError): # if still not working, give up
- self._raise_login_required()
+ 'https://api-gateway-global.viu.com/api/playback/distribute',
+ video_id=video_id, query=query, fatal=False, note='Downloading stream info',
+ headers={
+ 'Authorization': f'Bearer {self._auth_codes[country_code]}',
+ 'Referer': url,
+ 'Origin': url
+ })
+ return self._detect_error(stream_data).get('stream')
+
+ if not self._auth_codes.get(country_code):
+ self._auth_codes[country_code] = self._get_token(country_code, video_id)
+ stream_data = None
+ try:
+ stream_data = download_playback()
+ except (ExtractorError, KeyError):
+ token = self._login(country_code, video_id)
+ if token is not None:
+ query['identity'] = token
+ else:
+ # The content is Preview or for VIP only.
+ # We can try to bypass the duration which is limited to 3mins only
+ duration_limit, query['duration'] = True, '180'
+ try:
+ stream_data = download_playback()
+ except (ExtractorError, KeyError):
+ if token is not None:
+ raise
+ self.raise_login_required(method='password')
if not stream_data:
raise ExtractorError('Cannot get stream info', expected=True)
- stream_sizes = stream_data.get('size', {})
formats = []
- for vid_format, stream_url in stream_data.get('url', {}).items():
- height = int_or_none(self._search_regex(
- r's(\d+)p', vid_format, 'height', default=None))
+ for vid_format, stream_url in (stream_data.get('url') or {}).items():
+ height = int(self._search_regex(r's(\d+)p', vid_format, 'height', default=None))
# bypass preview duration limit
if duration_limit:
- stream_url = compat_urlparse.urlparse(stream_url)
- query = dict(compat_urlparse.parse_qsl(stream_url.query, keep_blank_values=True))
- time_duration = int_or_none(video_data.get('time_duration'))
+ old_stream_url = urllib.parse.urlparse(stream_url)
+ query = dict(urllib.parse.parse_qsl(old_stream_url.query, keep_blank_values=True))
query.update({
- 'duration': time_duration if time_duration > 0 else '9999999',
+ 'duration': video_data.get('time_duration') or '9999999',
'duration_start': '0',
})
- stream_url = stream_url._replace(query=compat_urlparse.urlencode(query)).geturl()
+ stream_url = old_stream_url._replace(query=urllib.parse.urlencode(query)).geturl()
formats.append({
'format_id': vid_format,
'url': stream_url,
'height': height,
'ext': 'mp4',
- 'filesize': int_or_none(stream_sizes.get(vid_format))
+ 'filesize': try_get(stream_data, lambda x: x['size'][vid_format], int)
})
- self._sort_formats(formats)
subtitles = {}
- for sub in video_data.get('subtitle', []):
- sub_url = sub.get('url')
- if not sub_url:
- continue
- subtitles.setdefault(sub.get('name'), []).append({
- 'url': sub_url,
- 'ext': 'srt',
- })
-
- title = video_data['synopsis'].strip()
+ for sub in video_data.get('subtitle') or []:
+ lang = sub.get('name') or 'und'
+ if sub.get('url'):
+ subtitles.setdefault(lang, []).append({
+ 'url': sub['url'],
+ 'ext': 'srt',
+ 'name': f'Spoken text for {lang}',
+ })
+ if sub.get('second_subtitle_url'):
+ subtitles.setdefault(f'{lang}_ost', []).append({
+ 'url': sub['second_subtitle_url'],
+ 'ext': 'srt',
+ 'name': f'On-screen text for {lang}',
+ })
+ title = strip_or_none(video_data.get('synopsis'))
return {
'id': video_id,
'title': title,
'description': video_data.get('description'),
- 'series': product_data.get('series', {}).get('name'),
+ 'series': try_get(product_data, lambda x: x['series']['name']),
'episode': title,
'episode_number': int_or_none(video_data.get('number')),
'duration': int_or_none(stream_data.get('duration')),
- 'thumbnail': video_data.get('cover_image_url'),
+ 'thumbnail': url_or_none(video_data.get('cover_image_url')),
+ 'formats': formats,
+ 'subtitles': subtitles,
+ }
+
+
+class ViuOTTIndonesiaBaseIE(InfoExtractor):
+ _BASE_QUERY = {
+ 'ver': 1.0,
+ 'fmt': 'json',
+ 'aver': 5.0,
+ 'appver': 2.0,
+ 'appid': 'viu_desktop',
+ 'platform': 'desktop',
+ }
+
+ _DEVICE_ID = str(uuid.uuid4())
+ _SESSION_ID = str(uuid.uuid4())
+ _TOKEN = None
+
+ _HEADERS = {
+ 'x-session-id': _SESSION_ID,
+ 'x-client': 'browser'
+ }
+
+ _AGE_RATINGS_MAPPER = {
+ 'ADULTS': 18,
+ 'teens': 13
+ }
+
+ def _real_initialize(self):
+ ViuOTTIndonesiaBaseIE._TOKEN = self._download_json(
+ 'https://um.viuapi.io/user/identity', None,
+ headers={'Content-type': 'application/json', **self._HEADERS},
+ query={**self._BASE_QUERY, 'iid': self._DEVICE_ID},
+ data=json.dumps({'deviceId': self._DEVICE_ID}).encode(),
+ note='Downloading token information')['token']
+
+
+class ViuOTTIndonesiaIE(ViuOTTIndonesiaBaseIE):
+ _VALID_URL = r'https?://www\.viu\.com/ott/\w+/\w+/all/video-[\w-]+-(?P<id>\d+)'
+ _TESTS = [{
+ 'url': 'https://www.viu.com/ott/id/id/all/video-japanese-drama-tv_shows-detective_conan_episode_793-1165863142?containerId=playlist-26271226',
+ 'info_dict': {
+ 'id': '1165863142',
+ 'ext': 'mp4',
+ 'episode_number': 793,
+ 'episode': 'Episode 793',
+ 'title': 'Detective Conan - Episode 793',
+ 'duration': 1476,
+ 'description': 'md5:b79d55345bc1e0217ece22616267c9a5',
+ 'thumbnail': 'https://vuclipi-a.akamaihd.net/p/cloudinary/h_171,w_304,dpr_1.5,f_auto,c_thumb,q_auto:low/1165863189/d-1',
+ 'upload_date': '20210101',
+ 'timestamp': 1609459200,
+ }
+ }, {
+ 'url': 'https://www.viu.com/ott/id/id/all/video-korean-reality-tv_shows-entertainment_weekly_episode_1622-1118617054',
+ 'info_dict': {
+ 'id': '1118617054',
+ 'ext': 'mp4',
+ 'episode_number': 1622,
+ 'episode': 'Episode 1622',
+ 'description': 'md5:6d68ca450004020113e9bf27ad99f0f8',
+ 'title': 'Entertainment Weekly - Episode 1622',
+ 'duration': 4729,
+ 'thumbnail': 'https://vuclipi-a.akamaihd.net/p/cloudinary/h_171,w_304,dpr_1.5,f_auto,c_thumb,q_auto:low/1120187848/d-1',
+ 'timestamp': 1420070400,
+ 'upload_date': '20150101',
+ 'cast': ['Shin Hyun-joon', 'Lee Da-Hee']
+ }
+ }, {
+ # age-limit test
+ 'url': 'https://www.viu.com/ott/id/id/all/video-japanese-trailer-tv_shows-trailer_jujutsu_kaisen_ver_01-1166044219?containerId=playlist-26273140',
+ 'info_dict': {
+ 'id': '1166044219',
+ 'ext': 'mp4',
+ 'upload_date': '20200101',
+ 'timestamp': 1577836800,
+ 'title': 'Trailer \'Jujutsu Kaisen\' Ver.01',
+ 'duration': 92,
+ 'thumbnail': 'https://vuclipi-a.akamaihd.net/p/cloudinary/h_171,w_304,dpr_1.5,f_auto,c_thumb,q_auto:low/1166044240/d-1',
+ 'description': 'Trailer \'Jujutsu Kaisen\' Ver.01',
+ 'cast': ['Junya Enoki', ' Yûichi Nakamura', ' Yuma Uchida', 'Asami Seto'],
+ 'age_limit': 13,
+ }
+ }, {
+ # json ld metadata type equal to Movie instead of TVEpisodes
+ 'url': 'https://www.viu.com/ott/id/id/all/video-japanese-animation-movies-demon_slayer_kimetsu_no_yaiba_the_movie_mugen_train-1165892707?containerId=1675060691786',
+ 'info_dict': {
+ 'id': '1165892707',
+ 'ext': 'mp4',
+ 'timestamp': 1577836800,
+ 'upload_date': '20200101',
+ 'title': 'Demon Slayer - Kimetsu no Yaiba - The Movie: Mugen Train',
+ 'age_limit': 13,
+ 'cast': 'count:9',
+ 'thumbnail': 'https://vuclipi-a.akamaihd.net/p/cloudinary/h_171,w_304,dpr_1.5,f_auto,c_thumb,q_auto:low/1165895279/d-1',
+ 'description': 'md5:1ce9c35a3aeab384085533f746c87469',
+ 'duration': 7021,
+ }
+ }]
+
+ def _real_extract(self, url):
+ display_id = self._match_id(url)
+ webpage = self._download_webpage(url, display_id)
+
+ video_data = self._download_json(
+ f'https://um.viuapi.io/drm/v1/content/{display_id}', display_id, data=b'',
+ headers={'Authorization': ViuOTTIndonesiaBaseIE._TOKEN, **self._HEADERS, 'ccode': 'ID'})
+ formats, subtitles = self._extract_m3u8_formats_and_subtitles(video_data['playUrl'], display_id)
+
+ initial_state = self._search_json(
+ r'window\.__INITIAL_STATE__\s*=', webpage, 'initial state',
+ display_id)['content']['clipDetails']
+ for key, url in initial_state.items():
+ lang, ext = self._search_regex(
+ r'^subtitle_(?P<lang>[\w-]+)_(?P<ext>\w+)$', key, 'subtitle metadata',
+ default=(None, None), group=('lang', 'ext'))
+ if lang and ext:
+ subtitles.setdefault(lang, []).append({
+ 'ext': ext,
+ 'url': url,
+ })
+
+ if ext == 'vtt':
+ subtitles[lang].append({
+ 'ext': 'srt',
+ 'url': f'{remove_end(initial_state[key], "vtt")}srt',
+ })
+
+ episode = traverse_obj(list(filter(
+ lambda x: x.get('@type') in ('TVEpisode', 'Movie'), self._yield_json_ld(webpage, display_id))), 0) or {}
+ return {
+ 'id': display_id,
+ 'title': (traverse_obj(initial_state, 'title', 'display_title')
+ or episode.get('name')),
+ 'description': initial_state.get('description') or episode.get('description'),
+ 'duration': initial_state.get('duration'),
+ 'thumbnail': traverse_obj(episode, ('image', 'url')),
+ 'timestamp': unified_timestamp(episode.get('dateCreated')),
'formats': formats,
'subtitles': subtitles,
+ 'episode_number': (traverse_obj(initial_state, 'episode_no', 'episodeno', expected_type=int_or_none)
+ or int_or_none(episode.get('episodeNumber'))),
+ 'cast': traverse_obj(episode, ('actor', ..., 'name'), default=None),
+ 'age_limit': self._AGE_RATINGS_MAPPER.get(initial_state.get('internal_age_rating'))
}
projects / yt-dlp.git / blobdiff