[hotstar] fix extraction(closes #14694)(closes #14931)(closes #17637)

This commit is contained in:
Remita Amine 2018-09-26 08:13:16 +01:00
parent 4c89a675dd
commit 85cd69adcb

View File

@ -1,10 +1,12 @@
# coding: utf-8 # coding: utf-8
from __future__ import unicode_literals from __future__ import unicode_literals
import re import hashlib
import hmac
import time
from .common import InfoExtractor from .common import InfoExtractor
from ..compat import compat_str from ..compat import compat_HTTPError
from ..utils import ( from ..utils import (
determine_ext, determine_ext,
ExtractorError, ExtractorError,
@ -13,37 +15,40 @@ from ..utils import (
class HotStarBaseIE(InfoExtractor): class HotStarBaseIE(InfoExtractor):
_GEO_COUNTRIES = ['IN'] _AKAMAI_ENCRYPTION_KEY = b'\x05\xfc\x1a\x01\xca\xc9\x4b\xc4\x12\xfc\x53\x12\x07\x75\xf9\xee'
def _download_json(self, *args, **kwargs): def _call_api(self, path, video_id, query_name='contentId'):
response = super(HotStarBaseIE, self)._download_json(*args, **kwargs) st = int(time.time())
if response['resultCode'] != 'OK': exp = st + 6000
if kwargs.get('fatal'): auth = 'st=%d~exp=%d~acl=/*' % (st, exp)
raise ExtractorError( auth += '~hmac=' + hmac.new(self._AKAMAI_ENCRYPTION_KEY, auth.encode(), hashlib.sha256).hexdigest()
response['errorDescription'], expected=True) response = self._download_json(
return None 'https://api.hotstar.com/' + path,
return response['resultObj'] video_id, headers={
'hotstarauth': auth,
def _download_content_info(self, content_id): 'x-country-code': 'IN',
return self._download_json( 'x-platform-code': 'JIO',
'https://account.hotstar.com/AVS/besc', content_id, query={ }, query={
'action': 'GetAggregatedContentDetails', query_name: video_id,
'appVersion': '5.0.40', 'tas': 10000,
'channel': 'PCTV', })
'contentId': content_id, if response['statusCode'] != 'OK':
})['contentInfo'][0] raise ExtractorError(
response['body']['message'], expected=True)
return response['body']['results']
class HotStarIE(HotStarBaseIE): class HotStarIE(HotStarBaseIE):
IE_NAME = 'hotstar'
_VALID_URL = r'https?://(?:www\.)?hotstar\.com/(?:.+?[/-])?(?P<id>\d{10})' _VALID_URL = r'https?://(?:www\.)?hotstar\.com/(?:.+?[/-])?(?P<id>\d{10})'
_TESTS = [{ _TESTS = [{
'url': 'http://www.hotstar.com/on-air-with-aib--english-1000076273', 'url': 'https://www.hotstar.com/can-you-not-spread-rumours/1000076273',
'info_dict': { 'info_dict': {
'id': '1000076273', 'id': '1000076273',
'ext': 'mp4', 'ext': 'mp4',
'title': 'On Air With AIB', 'title': 'Can You Not Spread Rumours?',
'description': 'md5:c957d8868e9bc793ccb813691cc4c434', 'description': 'md5:c957d8868e9bc793ccb813691cc4c434',
'timestamp': 1447227000, 'timestamp': 1447248600,
'upload_date': '20151111', 'upload_date': '20151111',
'duration': 381, 'duration': 381,
}, },
@ -58,47 +63,43 @@ class HotStarIE(HotStarBaseIE):
'url': 'http://www.hotstar.com/1000000515', 'url': 'http://www.hotstar.com/1000000515',
'only_matching': True, 'only_matching': True,
}] }]
_GEO_BYPASS = False
def _real_extract(self, url): def _real_extract(self, url):
video_id = self._match_id(url) video_id = self._match_id(url)
video_data = self._download_content_info(video_id) webpage = self._download_webpage(url, video_id)
app_state = self._parse_json(self._search_regex(
r'<script>window\.APP_STATE\s*=\s*({.+?})</script>',
webpage, 'app state'), video_id)
video_data = list(app_state.values())[0]['initialState']['contentData']['content']
title = video_data['episodeTitle'] title = video_data['title']
if video_data.get('encrypted') == 'Y': if video_data.get('drmProtected'):
raise ExtractorError('This video is DRM protected.', expected=True) raise ExtractorError('This video is DRM protected.', expected=True)
formats = [] formats = []
for f in ('JIO',): format_data = self._call_api('h/v1/play', video_id)['item']
format_data = self._download_json( format_url = format_data['playbackUrl']
'http://getcdn.hotstar.com/AVS/besc', ext = determine_ext(format_url)
video_id, 'Downloading %s JSON metadata' % f, if ext == 'm3u8':
fatal=False, query={ try:
'action': 'GetCDN', formats.extend(self._extract_m3u8_formats(
'asJson': 'Y', format_url, video_id, 'mp4', m3u8_id='hls'))
'channel': f, except ExtractorError as e:
'id': video_id, if isinstance(e.cause, compat_HTTPError) and e.cause.code == 403:
'type': 'VOD', self.raise_geo_restricted(countries=['IN'])
}) raise
if format_data: elif ext == 'f4m':
format_url = format_data.get('src') # produce broken files
if not format_url: pass
continue else:
ext = determine_ext(format_url) formats.append({
if ext == 'm3u8': 'url': format_url,
formats.extend(self._extract_m3u8_formats( 'width': int_or_none(format_data.get('width')),
format_url, video_id, 'mp4', 'height': int_or_none(format_data.get('height')),
m3u8_id='hls', fatal=False)) })
elif ext == 'f4m':
# produce broken files
continue
else:
formats.append({
'url': format_url,
'width': int_or_none(format_data.get('width')),
'height': int_or_none(format_data.get('height')),
})
self._sort_formats(formats) self._sort_formats(formats)
return { return {
@ -106,57 +107,43 @@ class HotStarIE(HotStarBaseIE):
'title': title, 'title': title,
'description': video_data.get('description'), 'description': video_data.get('description'),
'duration': int_or_none(video_data.get('duration')), 'duration': int_or_none(video_data.get('duration')),
'timestamp': int_or_none(video_data.get('broadcastDate')), 'timestamp': int_or_none(video_data.get('broadcastDate') or video_data.get('startDate')),
'formats': formats, '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': title,
'episode_number': int_or_none(video_data.get('episodeNumber')), 'episode_number': int_or_none(video_data.get('episodeNo')),
'series': video_data.get('contentTitle'),
} }
class HotStarPlaylistIE(HotStarBaseIE): class HotStarPlaylistIE(HotStarBaseIE):
IE_NAME = 'hotstar:playlist' IE_NAME = 'hotstar:playlist'
_VALID_URL = r'(?P<url>https?://(?:www\.)?hotstar\.com/tv/[^/]+/(?P<content_id>\d+))/(?P<type>[^/]+)/(?P<id>\d+)' _VALID_URL = r'https?://(?:www\.)?hotstar\.com/tv/[^/]+/s-\w+/list/[^/]+/t-(?P<id>\w+)'
_TESTS = [{ _TESTS = [{
'url': 'http://www.hotstar.com/tv/pratidaan/14982/episodes/14812/9993', 'url': 'https://www.hotstar.com/tv/savdhaan-india/s-26/list/popular-clips/t-3_2_26',
'info_dict': { 'info_dict': {
'id': '14812', 'id': '3_2_26',
}, },
'playlist_mincount': 75, 'playlist_mincount': 20,
}, { }, {
'url': 'http://www.hotstar.com/tv/pratidaan/14982/popular-clips/9998/9998', 'url': 'https://www.hotstar.com/tv/savdhaan-india/s-26/list/extras/t-2480',
'only_matching': True, 'only_matching': True,
}] }]
_ITEM_TYPES = {
'episodes': 'EPISODE',
'popular-clips': 'CLIPS',
}
def _real_extract(self, url): def _real_extract(self, url):
mobj = re.match(self._VALID_URL, url) playlist_id = self._match_id(url)
base_url = mobj.group('url')
content_id = mobj.group('content_id')
playlist_type = mobj.group('type')
content_info = self._download_content_info(content_id) collection = self._call_api('o/v1/tray/find', playlist_id, 'uqId')
playlist_id = compat_str(content_info['categoryId'])
collection = self._download_json(
'https://search.hotstar.com/AVS/besc', playlist_id, query={
'action': 'SearchContents',
'appVersion': '5.0.40',
'channel': 'PCTV',
'moreFilters': 'series:%s;' % playlist_id,
'query': '*',
'searchOrder': 'last_broadcast_date desc,year desc,title asc',
'type': self._ITEM_TYPES.get(playlist_type, 'EPISODE'),
})
entries = [ entries = [
self.url_result( self.url_result(
'%s/_/%s' % (base_url, video['contentId']), 'https://www.hotstar.com/%s' % video['contentId'],
ie=HotStarIE.ie_key(), video_id=video['contentId']) ie=HotStarIE.ie_key(), video_id=video['contentId'])
for video in collection['response']['docs'] for video in collection['assets']['items']
if video.get('contentId')] if video.get('contentId')]
return self.playlist_result(entries, playlist_id) return self.playlist_result(entries, playlist_id)