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

youtube_dl/extractor/hotstar.py

@@ -1,10 +1,12 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
-import re
+import hashlib
+import hmac
+import time
 
 from .common import InfoExtractor
-from ..compat import compat_str
+from ..compat import compat_HTTPError
 from ..utils import (
     determine_ext,
     ExtractorError,
@@ -13,37 +15,40 @@
 
 
 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)
+        auth += '~hmac=' + hmac.new(self._AKAMAI_ENCRYPTION_KEY, auth.encode(), hashlib.sha256).hexdigest()
+        response = self._download_json(
+            'https://api.hotstar.com/' + path,
+            video_id, headers={
+                'hotstarauth': auth,
+                'x-country-code': 'IN',
+                'x-platform-code': 'JIO',
+            }, query={
+                query_name: video_id,
+                'tas': 10000,
+            })
+        if response['statusCode'] != 'OK':
             raise ExtractorError(
-                    response['errorDescription'], expected=True)
+                response['body']['message'], expected=True)
-            return None
+        return response['body']['results']
-        return response['resultObj']
-
-    def _download_content_info(self, content_id):
-        return self._download_json(
-            'https://account.hotstar.com/AVS/besc', content_id, query={
-                'action': 'GetAggregatedContentDetails',
-                'appVersion': '5.0.40',
-                'channel': 'PCTV',
-                'contentId': content_id,
-            })['contentInfo'][0]
 
 
 class HotStarIE(HotStarBaseIE):
+    IE_NAME = 'hotstar'
     _VALID_URL = r'https?://(?:www\.)?hotstar\.com/(?:.+?[/-])?(?P<id>\d{10})'
     _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': {
             'id': '1000076273',
             'ext': 'mp4',
-            'title': 'On Air With AIB',
+            'title': 'Can You Not Spread Rumours?',
             'description': 'md5:c957d8868e9bc793ccb813691cc4c434',
-            'timestamp': 1447227000,
+            'timestamp': 1447248600,
             'upload_date': '20151111',
             'duration': 381,
         },
@@ -58,41 +63,37 @@ class HotStarIE(HotStarBaseIE):
         'url': 'http://www.hotstar.com/1000000515',
         'only_matching': True,
     }]
+    _GEO_BYPASS = False
 
     def _real_extract(self, 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)
 
         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',
-                video_id, 'Downloading %s JSON metadata' % f,
-                fatal=False, query={
-                    'action': 'GetCDN',
-                    'asJson': 'Y',
-                    'channel': f,
-                    'id': video_id,
-                    'type': 'VOD',
-                })
-            if format_data:
-                format_url = format_data.get('src')
-                if not format_url:
-                    continue
         ext = determine_ext(format_url)
         if ext == 'm3u8':
+            try:
                 formats.extend(self._extract_m3u8_formats(
-                        format_url, video_id, 'mp4',
+                    format_url, video_id, 'mp4', m3u8_id='hls'))
-                        m3u8_id='hls', fatal=False))
+            except ExtractorError as e:
+                if isinstance(e.cause, compat_HTTPError) and e.cause.code == 403:
+                    self.raise_geo_restricted(countries=['IN'])
+                raise
         elif ext == 'f4m':
             # produce broken files
-                    continue
+            pass
         else:
             formats.append({
                 'url': format_url,
@@ -106,57 +107,43 @@ def _real_extract(self, url):
             'title': title,
             'description': video_data.get('description'),
             '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,
+            '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_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):
     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 = [{
-        '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': {
-            '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,
     }]
-    _ITEM_TYPES = {
-        'episodes': 'EPISODE',
-        'popular-clips': 'CLIPS',
-    }
 
     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 = [
             self.url_result(
-                '%s/_/%s' % (base_url, video['contentId']),
+                'https://www.hotstar.com/%s' % 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')]
 
         return self.playlist_result(entries, playlist_id)