diff options
Diffstat (limited to 'hypervideo_dl/extractor/hotstar.py')
| -rw-r--r-- | hypervideo_dl/extractor/hotstar.py | 291 |
1 files changed, 160 insertions, 131 deletions
diff --git a/hypervideo_dl/extractor/hotstar.py b/hypervideo_dl/extractor/hotstar.py index d55a79b..61eec7b 100644 --- a/hypervideo_dl/extractor/hotstar.py +++ b/hypervideo_dl/extractor/hotstar.py @@ -1,31 +1,33 @@ -# coding: utf-8 -from __future__ import unicode_literals - import hashlib import hmac +import json import re import time import uuid -import json from .common import InfoExtractor -from ..compat import ( - compat_HTTPError, - compat_str -) +from ..compat import compat_HTTPError, compat_str from ..utils import ( - determine_ext, ExtractorError, + determine_ext, int_or_none, + join_nonempty, str_or_none, - try_get, + traverse_obj, url_or_none, ) 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' + def _call_api_v1(self, path, *args, **kwargs): + return self._download_json( + f'{self._API_URL}/o/v1/{path}', *args, **kwargs, + headers={'x-country-code': 'IN', 'x-platform-code': 'PCTV'}) + 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 @@ -36,7 +38,7 @@ class HotStarBaseIE(InfoExtractor): token = cookies.get('userUP').value else: token = self._download_json( - 'https://api.hotstar.com/um/v3/users', + f'{self._API_URL}/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={ @@ -46,58 +48,48 @@ class HotStarBaseIE(InfoExtractor): })['user_identity'] response = self._download_json( - 'https://api.hotstar.com/' + path, video_id, headers={ + f'{self._API_URL}/{path}', video_id, query=query, + 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._download_json('https://api.hotstar.com/' + path, video_id=video_id, query={ - query_name: video_id, - 'tas': 10000, - }, headers={ - 'x-country-code': 'IN', - 'x-platform-code': 'PCTV', - }) - 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={ + 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', 'device-id': cookies.get('device_id').value if cookies.get('device_id') else compat_str(uuid.uuid4()), 'os-name': 'Windows', 'os-version': '10', }) + def _playlist_entries(self, path, item_id, root=None, **kwargs): + results = self._call_api_v1(path, item_id, **kwargs)['body']['results'] + for video in traverse_obj(results, (('assets', None), 'items', ...)): + if video.get('contentId'): + yield self.url_result( + HotStarIE._video_url(video['contentId'], root=root), HotStarIE, video['contentId']) + class HotStarIE(HotStarBaseIE): IE_NAME = 'hotstar' _VALID_URL = r'''(?x) - (?: - hotstar\:| - https?://(?:www\.)?hotstar\.com(?:/in)?/(?!in/) - ) - (?: - (?P<type>movies|sports|episode|(?P<tv>tv)) - (?: - \:| - /[^/?#]+/ - (?(tv) - (?:[^/?#]+/){2}| - (?:[^/?#]+/)* - ) - )| - [^/?#]+/ - )? - (?P<id>\d{10}) - ''' + https?://(?:www\.)?hotstar\.com(?:/in)?/(?!in/) + (?: + (?P<type>movies|sports|episode|(?P<tv>tv))/ + (?(tv)(?:[^/?#]+/){2}|[^?#]*) + )? + [^/?#]+/ + (?P<id>\d{10}) + ''' + _TESTS = [{ 'url': 'https://www.hotstar.com/can-you-not-spread-rumours/1000076273', 'info_dict': { @@ -108,38 +100,9 @@ class HotStarIE(HotStarBaseIE): 'timestamp': 1447248600, 'upload_date': '20151111', '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, + 'params': {'skip_download': 'm3u8'}, }, { 'url': 'https://www.hotstar.com/tv/ek-bhram-sarvagun-sampanna/s-2116/janhvi-targets-suman/1000234847', 'info_dict': { @@ -158,12 +121,19 @@ class HotStarIE(HotStarBaseIE): 'season_id': 6771, 'episode': 'Janhvi Targets Suman', '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, }] _GEO_BYPASS = False + _TYPE = { 'movies': 'movie', 'sports': 'match', @@ -172,41 +142,54 @@ class HotStarIE(HotStarBaseIE): None: 'content', } + _IGNORE_MAP = { + 'res': 'resolution', + 'vcodec': 'video_codec', + 'dr': 'dynamic_range', + } + + @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): - mobj = self._match_valid_url(url) - video_id = mobj.group('id') - video_type = mobj.group('type') - cookies = self._get_cookies(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'] - title = video_data['title'] + cookies = self._get_cookies(url) # Cookies before any request + video_data = self._call_api_v1(f'{video_type}/detail', video_id, + query={'tas': 10000, 'contentId': video_id})['body']['results']['item'] if not self.get_param('allow_unplayable_formats') and video_data.get('drmProtected'): self.report_drm(video_id) - headers = {'Referer': 'https://www.hotstar.com/in'} - formats = [] - subs = {} + # See https://github.com/hypervideo/hypervideo/issues/396 + st = self._download_webpage_handle(f'{self._BASE_URL}/in', video_id)[1].headers.get('x-origin-date') + geo_restricted = False - _, urlh = self._download_webpage_handle('https://www.hotstar.com/in', video_id) - # Required to fix https://github.com/hypervideo/hypervideo/issues/396 - st = urlh.headers.get('x-origin-date') + formats, subs = [], {} + headers = {'Referer': f'{self._BASE_URL}/in'} + # 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 - 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')) 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 '' - 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 + format_url = re.sub(r'(?<=//staragvod)(\d)', r'web\1', format_url) + dr = re.search(r'dynamic_range:(?P<dr>[a-z]+)', playback_set.get('tagsCombination')).group('dr') ext = determine_ext(format_url) + current_formats, current_subs = [], {} try: if 'package:hls' in tags or ext == 'm3u8': @@ -218,8 +201,7 @@ class HotStarIE(HotStarBaseIE): current_formats, current_subs = self._extract_mpd_formats_and_subtitles( format_url, video_id, mpd_id=f'{dr}-dash', headers=headers) elif ext == 'f4m': - # produce broken files - pass + pass # XXX: produce broken files else: current_formats = [{ 'url': format_url, @@ -230,6 +212,7 @@ class HotStarIE(HotStarBaseIE): if isinstance(e.cause, compat_HTTPError) and e.cause.code == 403: geo_restricted = True continue + if tags and 'encryption:plain' not in tags: for f in current_formats: f['has_drm'] = True @@ -238,18 +221,18 @@ class HotStarIE(HotStarBaseIE): for f in current_formats: if not f.get('langauge'): f['language'] = lang + formats.extend(current_formats) subs = self._merge_subtitles(subs, current_subs) + 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, + 'title': video_data.get('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')), @@ -261,17 +244,51 @@ class HotStarIE(HotStarBaseIE): 'season': video_data.get('seasonName'), 'season_number': int_or_none(video_data.get('seasonNo')), 'season_id': video_data.get('seasonId'), - 'episode': title, + 'episode': video_data.get('title'), '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:1260009879', + 'info_dict': { + 'id': '1260009879', + 'ext': 'mp4', + 'title': 'Nuvvu Naaku Nachav', + 'description': 'md5:d43701b1314e6f8233ce33523c043b7d', + 'timestamp': 1567525674, + 'upload_date': '20190903', + 'duration': 10787, + 'episode': 'Nuvvu Naaku Nachav', + }, + }, { + '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): 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(?:/in)?/tv(?:/[^/]+){2}/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': { @@ -281,25 +298,49 @@ class HotStarPlaylistIE(HotStarBaseIE): }, { 'url': 'https://www.hotstar.com/tv/savdhaan-india/s-26/list/extras/t-2480', 'only_matching': True, + }, { + 'url': 'https://www.hotstar.com/in/tv/karthika-deepam/15457/list/popular-clips/t-3_2_1272', + 'only_matching': True, }] def _real_extract(self, url): - playlist_id = self._match_id(url) + id_ = self._match_id(url) + return self.playlist_result( + self._playlist_entries('tray/find', id_, query={'tas': 10000, 'uqId': id_}), id_) - collection = self._call_api('o/v1/tray/find', playlist_id, 'uqId')['body']['results'] - 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 HotStarSeasonIE(HotStarBaseIE): + IE_NAME = 'hotstar:season' + _VALID_URL = r'(?P<url>https?://(?:www\.)?hotstar\.com(?:/in)?/tv/[^/]+/\w+)/seasons/[^/]+/ss-(?P<id>\w+)' + _TESTS = [{ + 'url': 'https://www.hotstar.com/tv/radhakrishn/1260000646/seasons/season-2/ss-8028', + 'info_dict': { + 'id': '8028', + }, + 'playlist_mincount': 35, + }, { + 'url': 'https://www.hotstar.com/in/tv/ishqbaaz/9567/seasons/season-2/ss-4357', + 'info_dict': { + 'id': '4357', + }, + 'playlist_mincount': 30, + }, { + 'url': 'https://www.hotstar.com/in/tv/bigg-boss/14714/seasons/season-4/ss-8208/', + 'info_dict': { + 'id': '8208', + }, + 'playlist_mincount': 19, + }] + + def _real_extract(self, url): + url, season_id = self._match_valid_url(url).groups() + return self.playlist_result(self._playlist_entries( + 'season/asset', season_id, url, query={'tao': 0, 'tas': 0, 'size': 10000, 'id': season_id}), season_id) class HotStarSeriesIE(HotStarBaseIE): IE_NAME = 'hotstar:series' - _VALID_URL = r'(?P<url>https?://(?:www\.)?hotstar\.com(?:/in)?/tv/[^/]+/(?P<id>\d+))' + _VALID_URL = r'(?P<url>https?://(?:www\.)?hotstar\.com(?:/in)?/tv/[^/]+/(?P<id>\d+))/?(?:[#?]|$)' _TESTS = [{ 'url': 'https://www.hotstar.com/in/tv/radhakrishn/1260000646', 'info_dict': { @@ -317,25 +358,13 @@ class HotStarSeriesIE(HotStarBaseIE): 'info_dict': { 'id': '435', }, - 'playlist_mincount': 269, + 'playlist_mincount': 267, }] def _real_extract(self, url): url, series_id = self._match_valid_url(url).groups() - 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( - '%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) + id_ = self._call_api_v1( + 'show/detail', series_id, query={'contentId': series_id})['body']['results']['item']['id'] + + return self.playlist_result(self._playlist_entries( + 'tray/g/1/items', series_id, url, query={'tao': 0, 'tas': 10000, 'etid': 0, 'eid': id_}), series_id) |