external/yt-dlp
1
0
Fork 0
mirror of https://github.com/yt-dlp/yt-dlp.git synced 2024-11-10 09:07:58 +01:00
Code Issues Projects Releases Packages Wiki Activity

[hotstar] Refactor extractors

Browse source 
Closes #3517
This commit is contained in:
pukkandan 2022-04-22 13:16:24 +05:30
parent 6534298b12
commit a1ddaa899c
No known key found for this signature in database
GPG key ID: 7EEE9E1E817D0A39
2 changed files with 126 additions and 110 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

1
yt_dlp/extractor/extractors.py
View file

@ -616,6 +616,7 @@
from .hotnewhiphop import HotNewHipHopIE from .hotnewhiphop import HotNewHipHopIE
from .hotstar import ( from .hotstar import (
HotStarIE, HotStarIE,
HotStarPrefixIE,
HotStarPlaylistIE, HotStarPlaylistIE,
HotStarSeriesIE, HotStarSeriesIE,
) )

235
yt_dlp/extractor/hotstar.py
View file

@ -14,6 +14,7 @@
determine_ext, determine_ext,
ExtractorError, ExtractorError,
int_or_none, int_or_none,
join_nonempty,
str_or_none, str_or_none,
try_get, try_get,
url_or_none, url_or_none,
@ -21,6 +22,8 @@
class HotStarBaseIE(InfoExtractor): class HotStarBaseIE(InfoExtractor):
_BASE_URL = 'https://www.hotstar.com'
_API_URL = 'https://api.hotstar.com'
_AKAMAI_ENCRYPTION_KEY = b'\x05\xfc\x1a\x01\xca\xc9\x4b\xc4\x12\xfc\x53\x12\x07\x75\xf9\xee' _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): def _call_api_impl(self, path, video_id, query, st=None, cookies=None):
@ -33,7 +36,7 @@ def _call_api_impl(self, path, video_id, query, st=None, cookies=None):
token = cookies.get('userUP').value token = cookies.get('userUP').value
else: else:
token = self._download_json( token = self._download_json(
'https://api.hotstar.com/um/v3/users', f'{self._API_URL}/um/v3/users',
video_id, note='Downloading token', video_id, note='Downloading token',
data=json.dumps({"device_ids": [{"id": compat_str(uuid.uuid4()), "type": "device_id"}]}).encode('utf-8'), data=json.dumps({"device_ids": [{"id": compat_str(uuid.uuid4()), "type": "device_id"}]}).encode('utf-8'),
headers={ headers={
@ -43,12 +46,13 @@ def _call_api_impl(self, path, video_id, query, st=None, cookies=None):
})['user_identity'] })['user_identity']
response = self._download_json( response = self._download_json(
'https://api.hotstar.com/' + path, video_id, headers={ f'{self._API_URL}/{path}', video_id, query=query,
headers={
'hotstarauth': auth, 'hotstarauth': auth,
'x-hs-appversion': '6.72.2', 'x-hs-appversion': '6.72.2',
'x-hs-platform': 'web', 'x-hs-platform': 'web',
'x-hs-usertoken': token, 'x-hs-usertoken': token,
}, query=query) })
if response['message'] != "Playback URL's fetched successfully": if response['message'] != "Playback URL's fetched successfully":
raise ExtractorError( raise ExtractorError(
@ -56,17 +60,20 @@ def _call_api_impl(self, path, video_id, query, st=None, cookies=None):
return response['data'] return response['data']
def _call_api(self, path, video_id, query_name='contentId'): def _call_api(self, path, video_id, query_name='contentId'):
return self._download_json('https://api.hotstar.com/' + path, video_id=video_id, query={ return self._download_json(
query_name: video_id, f'{self._API_URL}/{path}', video_id=video_id,
'tas': 10000, query={
}, headers={ query_name: video_id,
'x-country-code': 'IN', 'tas': 10000,
'x-platform-code': 'PCTV', }, headers={
}) 'x-country-code': 'IN',
'x-platform-code': 'PCTV',
})
def _call_api_v2(self, path, video_id, st=None, cookies=None): def _call_api_v2(self, path, video_id, st=None):
cookies = self._get_cookies(self._BASE_URL)
return self._call_api_impl( return self._call_api_impl(
'%s/content/%s' % (path, video_id), video_id, st=st, cookies=cookies, query={ f'{path}/content/{video_id}', video_id, st=st, cookies=cookies, query={
'desired-config': 'audio_channel:stereo|container:fmp4|dynamic_range:hdr|encryption:plain|ladder:tv|package:dash|resolution:fhd|subs-tag:HotstarVIP|video_codec:h265', 'desired-config': 'audio_channel:stereo|container:fmp4|dynamic_range:hdr|encryption:plain|ladder:tv|package:dash|resolution:fhd|subs-tag:HotstarVIP|video_codec:h265',
'device-id': cookies.get('device_id').value if cookies.get('device_id') else compat_str(uuid.uuid4()), 'device-id': cookies.get('device_id').value if cookies.get('device_id') else compat_str(uuid.uuid4()),
'os-name': 'Windows', 'os-name': 'Windows',
@ -77,24 +84,15 @@ def _call_api_v2(self, path, video_id, st=None, cookies=None):
class HotStarIE(HotStarBaseIE): class HotStarIE(HotStarBaseIE):
IE_NAME = 'hotstar' IE_NAME = 'hotstar'
_VALID_URL = r'''(?x) _VALID_URL = r'''(?x)
(?: https?://(?:www\.)?hotstar\.com(?:/in)?/(?!in/)
hotstar\:| (?:
https?://(?:www\.)?hotstar\.com(?:/in)?/(?!in/) (?P<type>movies|sports|episode|(?P<tv>tv))/
) (?(tv)(?:[^/?#]+/){2}|[^?#]*)
(?: )?
(?P<type>movies|sports|episode|(?P<tv>tv)) [^/?#]+/
(?: (?P<id>\d{10})
\:| '''
/[^/?#]+/
(?(tv)
(?:[^/?#]+/){2}|
(?:[^/?#]+/)*
)
)|
[^/?#]+/
)?
(?P<id>\d{10})
'''
_TESTS = [{ _TESTS = [{
'url': 'https://www.hotstar.com/can-you-not-spread-rumours/1000076273', 'url': 'https://www.hotstar.com/can-you-not-spread-rumours/1000076273',
'info_dict': { 'info_dict': {
@ -105,38 +103,8 @@ class HotStarIE(HotStarBaseIE):
'timestamp': 1447248600, 'timestamp': 1447248600,
'upload_date': '20151111', 'upload_date': '20151111',
'duration': 381, 'duration': 381,
'episode': 'Can You Not Spread Rumours?',
}, },
}, {
'url': 'hotstar:1000076273',
'only_matching': True,
}, {
'url': 'https://www.hotstar.com/movies/radha-gopalam/1000057157',
'info_dict': {
'id': '1000057157',
'ext': 'mp4',
'title': 'Radha Gopalam',
'description': 'md5:be3bc342cc120bbc95b3b0960e2b0d22',
'timestamp': 1140805800,
'upload_date': '20060224',
'duration': 9182,
},
}, {
'url': 'hotstar:movies:1000057157',
'only_matching': True,
}, {
'url': 'https://www.hotstar.com/in/sports/cricket/follow-the-blues-2021/recap-eng-fight-back-on-day-2/1260066104',
'only_matching': True,
}, {
'url': 'https://www.hotstar.com/in/sports/football/most-costly-pl-transfers-ft-grealish/1260065956',
'only_matching': True,
}, {
# contentData
'url': 'hotstar:sports:1260065956',
'only_matching': True,
}, {
# contentData
'url': 'hotstar:sports:1260066104',
'only_matching': True,
}, { }, {
'url': 'https://www.hotstar.com/tv/ek-bhram-sarvagun-sampanna/s-2116/janhvi-targets-suman/1000234847', 'url': 'https://www.hotstar.com/tv/ek-bhram-sarvagun-sampanna/s-2116/janhvi-targets-suman/1000234847',
'info_dict': { 'info_dict': {
@ -155,12 +123,19 @@ class HotStarIE(HotStarBaseIE):
'season_id': 6771, 'season_id': 6771,
'episode': 'Janhvi Targets Suman', 'episode': 'Janhvi Targets Suman',
'episode_number': 8, 'episode_number': 8,
}, }
}, { }, {
'url': 'hotstar:episode:1000234847', 'url': 'https://www.hotstar.com/movies/radha-gopalam/1000057157',
'only_matching': True,
}, {
'url': 'https://www.hotstar.com/in/sports/cricket/follow-the-blues-2021/recap-eng-fight-back-on-day-2/1260066104',
'only_matching': True,
}, {
'url': 'https://www.hotstar.com/in/sports/football/most-costly-pl-transfers-ft-grealish/1260065956',
'only_matching': True, 'only_matching': True,
}] }]
_GEO_BYPASS = False _GEO_BYPASS = False
_TYPE = { _TYPE = {
'movies': 'movie', 'movies': 'movie',
'sports': 'match', 'sports': 'match',
@ -169,41 +144,52 @@ class HotStarIE(HotStarBaseIE):
None: 'content', None: 'content',
} }
def _real_extract(self, url): _IGNORE_MAP = {
mobj = self._match_valid_url(url) 'res': 'resolution',
video_id = mobj.group('id') 'vcodec': 'video_codec',
video_type = mobj.group('type') 'dr': 'dynamic_range',
cookies = self._get_cookies(url) }
video_type = self._TYPE.get(video_type, video_type)
video_data = self._call_api(f'o/v1/{video_type}/detail', video_id)['body']['results']['item']
title = video_data['title']
@classmethod
def _video_url(cls, video_id, video_type=None, *, slug='ignore_me', root=None):
assert None in (video_type, root)
if not root:
root = join_nonempty(cls._BASE_URL, video_type, delim='/')
return f'{root}/{slug}/{video_id}'
def _real_extract(self, url):
video_id, video_type = self._match_valid_url(url).group('id', 'type')
video_type = self._TYPE.get(video_type, video_type)
video_data = self._call_api(f'o/v1/{video_type}/detail', video_id)['body']['results']['item']
if not self.get_param('allow_unplayable_formats') and video_data.get('drmProtected'): if not self.get_param('allow_unplayable_formats') and video_data.get('drmProtected'):
self.report_drm(video_id) self.report_drm(video_id)
headers = {'Referer': 'https://www.hotstar.com/in'} # See https://github.com/yt-dlp/yt-dlp/issues/396
formats = [] st = self._download_webpage_handle(f'{self._BASE_URL}/in', video_id)[1].headers.get('x-origin-date')
subs = {}
geo_restricted = False geo_restricted = False
_, urlh = self._download_webpage_handle('https://www.hotstar.com/in', video_id) formats, subs = [], {}
# Required to fix https://github.com/yt-dlp/yt-dlp/issues/396 headers = {'Referer': f'{self._BASE_URL}/in'}
st = urlh.headers.get('x-origin-date')
# change to v2 in the future # change to v2 in the future
playback_sets = self._call_api_v2('play/v1/playback', video_id, st=st, cookies=cookies)['playBackSets'] playback_sets = self._call_api_v2('play/v1/playback', video_id, st=st)['playBackSets']
for playback_set in playback_sets: for playback_set in playback_sets:
if not isinstance(playback_set, dict): if not isinstance(playback_set, dict):
continue continue
dr = re.search(r'dynamic_range:(?P<dr>[a-z]+)', playback_set.get('tagsCombination')).group('dr') tags = str_or_none(playback_set.get('tagsCombination')) or ''
if any(f'{prefix}:{ignore}' in tags
for key, prefix in self._IGNORE_MAP.items()
for ignore in self._configuration_arg(key)):
continue
format_url = url_or_none(playback_set.get('playbackUrl')) format_url = url_or_none(playback_set.get('playbackUrl'))
if not format_url: if not format_url:
continue continue
format_url = re.sub( format_url = re.sub(r'(?<=//staragvod)(\d)', r'web\1', format_url)
r'(?<=//staragvod)(\d)', r'web\1', format_url) dr = re.search(r'dynamic_range:(?P<dr>[a-z]+)', playback_set.get('tagsCombination')).group('dr')
tags = str_or_none(playback_set.get('tagsCombination')) or ''
ingored_res, ignored_vcodec, ignored_dr = self._configuration_arg('res'), self._configuration_arg('vcodec'), self._configuration_arg('dr')
if any(f'resolution:{ig_res}' in tags for ig_res in ingored_res) or any(f'video_codec:{ig_vc}' in tags for ig_vc in ignored_vcodec) or any(f'dynamic_range:{ig_dr}' in tags for ig_dr in ignored_dr):
continue
ext = determine_ext(format_url) ext = determine_ext(format_url)
current_formats, current_subs = [], {} current_formats, current_subs = [], {}
try: try:
if 'package:hls' in tags or ext == 'm3u8': if 'package:hls' in tags or ext == 'm3u8':
@ -215,8 +201,7 @@ def _real_extract(self, url):
current_formats, current_subs = self._extract_mpd_formats_and_subtitles( current_formats, current_subs = self._extract_mpd_formats_and_subtitles(
format_url, video_id, mpd_id=f'{dr}-dash', headers=headers) format_url, video_id, mpd_id=f'{dr}-dash', headers=headers)
elif ext == 'f4m': elif ext == 'f4m':
# produce broken files pass # XXX: produce broken files
pass
else: else:
current_formats = [{ current_formats = [{
'url': format_url, 'url': format_url,
@ -227,6 +212,7 @@ def _real_extract(self, url):
if isinstance(e.cause, compat_HTTPError) and e.cause.code == 403: if isinstance(e.cause, compat_HTTPError) and e.cause.code == 403:
geo_restricted = True geo_restricted = True
continue continue
if tags and 'encryption:plain' not in tags: if tags and 'encryption:plain' not in tags:
for f in current_formats: for f in current_formats:
f['has_drm'] = True f['has_drm'] = True
@ -235,18 +221,19 @@ def _real_extract(self, url):
for f in current_formats: for f in current_formats:
if not f.get('langauge'): if not f.get('langauge'):
f['language'] = lang f['language'] = lang
formats.extend(current_formats) formats.extend(current_formats)
subs = self._merge_subtitles(subs, current_subs) subs = self._merge_subtitles(subs, current_subs)
if not formats and geo_restricted: if not formats and geo_restricted:
self.raise_geo_restricted(countries=['IN'], metadata_available=True) self.raise_geo_restricted(countries=['IN'], metadata_available=True)
self._sort_formats(formats) self._sort_formats(formats)
for f in formats: for f in formats:
f.setdefault('http_headers', {}).update(headers) f.setdefault('http_headers', {}).update(headers)
return { return {
'id': video_id, 'id': video_id,
'title': title, 'title': video_data.get('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') or video_data.get('startDate')), 'timestamp': int_or_none(video_data.get('broadcastDate') or video_data.get('startDate')),
@ -258,14 +245,48 @@ def _real_extract(self, url):
'season': video_data.get('seasonName'), 'season': video_data.get('seasonName'),
'season_number': int_or_none(video_data.get('seasonNo')), 'season_number': int_or_none(video_data.get('seasonNo')),
'season_id': video_data.get('seasonId'), 'season_id': video_data.get('seasonId'),
'episode': title, 'episode': video_data.get('title'),
'episode_number': int_or_none(video_data.get('episodeNo')), 'episode_number': int_or_none(video_data.get('episodeNo')),
'http_headers': {
'Referer': 'https://www.hotstar.com/in',
}
} }
class HotStarPrefixIE(InfoExtractor):
""" The "hotstar:" prefix is no longer in use, but this is kept for backward compatibility """
IE_DESC = False
_VALID_URL = r'hotstar:(?:(?P<type>\w+):)?(?P<id>\d+)$'
_TESTS = [{
'url': 'hotstar:1000076273',
'only_matching': True,
}, {
'url': 'hotstar:movies:1000057157',
'info_dict': {
'id': '1000057157',
'ext': 'mp4',
'title': 'Radha Gopalam',
'description': 'md5:be3bc342cc120bbc95b3b0960e2b0d22',
'timestamp': 1140805800,
'upload_date': '20060224',
'duration': 9182,
'episode': 'Radha Gopalam',
},
}, {
'url': 'hotstar:episode:1000234847',
'only_matching': True,
}, {
# contentData
'url': 'hotstar:sports:1260065956',
'only_matching': True,
}, {
# contentData
'url': 'hotstar:sports:1260066104',
'only_matching': True,
}]
def _real_extract(self, url):
video_id, video_type = self._match_valid_url(url).group('id', 'type')
return self.url_result(HotStarIE._video_url(video_id, video_type), HotStarIE, video_id)
class HotStarPlaylistIE(HotStarBaseIE): class HotStarPlaylistIE(HotStarBaseIE):
IE_NAME = 'hotstar:playlist' IE_NAME = 'hotstar:playlist'
_VALID_URL = r'https?://(?:www\.)?hotstar\.com/tv/[^/]+/s-\w+/list/[^/]+/t-(?P<id>\w+)' _VALID_URL = r'https?://(?:www\.)?hotstar\.com/tv/[^/]+/s-\w+/list/[^/]+/t-(?P<id>\w+)'
@ -285,11 +306,8 @@ def _real_extract(self, url):
collection = self._call_api('o/v1/tray/find', playlist_id, 'uqId')['body']['results'] collection = self._call_api('o/v1/tray/find', playlist_id, 'uqId')['body']['results']
entries = [ entries = [
self.url_result( self.url_result(HotStarIE._video_url(video['contentId']), HotStarIE, video['contentId'])
'https://www.hotstar.com/%s' % video['contentId'], for video in collection['assets']['items'] if video.get('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) return self.playlist_result(entries, playlist_id)
@ -323,16 +341,13 @@ def _real_extract(self, url):
'x-country-code': 'IN', 'x-country-code': 'IN',
'x-platform-code': 'PCTV', 'x-platform-code': 'PCTV',
} }
detail_json = self._download_json('https://api.hotstar.com/o/v1/show/detail?contentId=' + series_id, detail_json = self._download_json(
video_id=series_id, headers=headers) f'{self._API_URL}/o/v1/show/detail?contentId={series_id}', series_id, headers=headers)
id = compat_str(try_get(detail_json, lambda x: x['body']['results']['item']['id'], int)) id = 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, item_json = self._download_json(
video_id=series_id, headers=headers) f'{self._API_URL}/o/v1/tray/g/1/items?etid=0&tao=0&tas=10000&eid={id}', series_id, headers=headers)
entries = [
self.url_result(
'%s/ignoreme/%d' % (url, 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) return self.playlist_result([
self.url_result(HotStarIE._video_url(video['contentId'], root=url), HotStarIE, video['contentId'])
for video in item_json['body']['results']['items'] if video.get('contentId')
], series_id)