]> Raphaël G. Git Repositories - youtubedl/blobdiff - youtube_dl/extractor/hotstar.py
Rapsys Git / youtubedl / blobdiff
? search:
summary | shortlog | log | commit | commitdiff | tree
raw | inline | side by side
Import Debian changes 2020.01.24-0.1
[youtubedl] / youtube_dl / extractor / hotstar.py
diff --git a/youtube_dl/extractor/hotstar.py b/youtube_dl/extractor/hotstar.py
index f05d765d650f540abe7f684037cc957ff2ab0b77..f97eefa3d6789cf58b05a6e6bf91d290463cf444 100644 (file)
--- a/youtube_dl/extractor/hotstar.py
+++ b/youtube_dl/extractor/hotstar.py
@@ -1,24 +1,76 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
 # coding: utf-8
 from __future__ import unicode_literals
 
+import hashlib
+import hmac
+import re
+import time
+import uuid
+
 from .common import InfoExtractor
 from .common import InfoExtractor
+from ..compat import (
+    compat_HTTPError,
+    compat_str,
+)
 from ..utils import (
 from ..utils import (
-    ExtractorError,
     determine_ext,
     determine_ext,
+    ExtractorError,
     int_or_none,
     int_or_none,
+    str_or_none,
+    try_get,
+    url_or_none,
 )
 
 
 )
 
 
-class HotStarIE(InfoExtractor):
+class HotStarBaseIE(InfoExtractor):
+    _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 = int(time.time())
+        exp = st + 6000
+        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)
+        if response['statusCode'] != 'OK':
+            raise ExtractorError(
+                response['body']['message'], expected=True)
+        return response['body']['results']
+
+    def _call_api(self, path, video_id, query_name='contentId'):
+        return self._call_api_impl(path, video_id, {
+            query_name: video_id,
+            'tas': 10000,
+        })
+
+    def _call_api_v2(self, path, video_id):
+        return self._call_api_impl(
+            '%s/in/contents/%s' % (path, video_id), video_id, {
+                'desiredConfig': 'encryption:plain;ladder:phone,tv;package:hls,dash',
+                'client': 'mweb',
+                'clientVersion': '6.18.0',
+                'deviceId': compat_str(uuid.uuid4()),
+                'osName': 'Windows',
+                'osVersion': '10',
+            })
+
+
+class HotStarIE(HotStarBaseIE):
+    IE_NAME = 'hotstar'
     _VALID_URL = r'https?://(?:www\.)?hotstar\.com/(?:.+?[/-])?(?P<id>\d{10})'
     _TESTS = [{
     _VALID_URL = r'https?://(?:www\.)?hotstar\.com/(?:.+?[/-])?(?P<id>\d{10})'
     _TESTS = [{
-        'url': 'http://www.hotstar.com/on-air-with-aib--english-1000076273',
+        # contentData
+        'url': 'https://www.hotstar.com/can-you-not-spread-rumours/1000076273',
         'info_dict': {
             'id': '1000076273',
             'ext': 'mp4',
         'info_dict': {
             'id': '1000076273',
             'ext': 'mp4',
-            'title': 'On Air With AIB - English',
+            'title': 'Can You Not Spread Rumours?',
             'description': 'md5:c957d8868e9bc793ccb813691cc4c434',
             'description': 'md5:c957d8868e9bc793ccb813691cc4c434',
-            'timestamp': 1447227000,
+            'timestamp': 1447248600,
             'upload_date': '20151111',
             'duration': 381,
         },
             'upload_date': '20151111',
             'duration': 381,
         },
@@ -26,58 +78,133 @@ class HotStarIE(InfoExtractor):
             # m3u8 download
             'skip_download': True,
         }
             # m3u8 download
             'skip_download': True,
         }
+    }, {
+        # contentDetail
+        'url': 'https://www.hotstar.com/movies/radha-gopalam/1000057157',
+        'only_matching': True,
     }, {
         'url': 'http://www.hotstar.com/sports/cricket/rajitha-sizzles-on-debut-with-329/2001477583',
         'only_matching': True,
     }, {
         'url': 'http://www.hotstar.com/1000000515',
         'only_matching': True,
     }, {
         'url': 'http://www.hotstar.com/sports/cricket/rajitha-sizzles-on-debut-with-329/2001477583',
         'only_matching': True,
     }, {
         'url': 'http://www.hotstar.com/1000000515',
         'only_matching': True,
+    }, {
+        # only available via api v2
+        'url': 'https://www.hotstar.com/tv/ek-bhram-sarvagun-sampanna/s-2116/janhvi-targets-suman/1000234847',
+        'only_matching': True,
     }]
     }]
-
-    _GET_CONTENT_TEMPLATE = 'http://account.hotstar.com/AVS/besc?action=GetAggregatedContentDetails&channel=PCTV&contentId=%s'
-    _GET_CDN_TEMPLATE = 'http://getcdn.hotstar.com/AVS/besc?action=GetCDN&asJson=Y&channel=%s&id=%s&type=%s'
-
-    def _download_json(self, url_or_request, video_id, note='Downloading JSON metadata', fatal=True):
-        json_data = super(HotStarIE, self)._download_json(url_or_request, video_id, note, fatal=fatal)
-        if json_data['resultCode'] != 'OK':
-            if fatal:
-                raise ExtractorError(json_data['errorDescription'])
-            return None
-        return json_data['resultObj']
+    _GEO_BYPASS = False
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
-        video_data = self._download_json(
-            self._GET_CONTENT_TEMPLATE % video_id,
-            video_id)['contentInfo'][0]
 
 
+        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 = {}
+        getters = list(
+            lambda x, k=k: x['initialState']['content%s' % k]['content']
+            for k in ('Data', 'Detail')
+        )
+        for v in app_state.values():
+            content = try_get(v, getters, dict)
+            if content and content.get('contentId') == video_id:
+                video_data = content
+                break
+
+        title = video_data['title']
+
+        if video_data.get('drmProtected'):
+            raise ExtractorError('This video is DRM protected.', expected=True)
+
+        headers = {'Referer': url}
         formats = []
         formats = []
-        # PCTV for extracting f4m manifest
-        for f in ('TABLET',):
-            format_data = self._download_json(
-                self._GET_CDN_TEMPLATE % (f, video_id, 'VOD'),
-                video_id, 'Downloading %s JSON metadata' % f, fatal=False)
-            if format_data:
-                format_url = format_data['src']
-                ext = determine_ext(format_url)
-                if ext == 'm3u8':
-                    formats.extend(self._extract_m3u8_formats(format_url, video_id, 'mp4', m3u8_id='hls', fatal=False))
+        geo_restricted = False
+        playback_sets = self._call_api_v2('h/v2/play', video_id)['playBackSets']
+        for playback_set in playback_sets:
+            if not isinstance(playback_set, dict):
+                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 ''
+            if tags and 'encryption:plain' not in tags:
+                continue
+            ext = determine_ext(format_url)
+            try:
+                if 'package:hls' in tags or ext == 'm3u8':
+                    formats.extend(self._extract_m3u8_formats(
+                        format_url, video_id, 'mp4',
+                        entry_protocol='m3u8_native',
+                        m3u8_id='hls', headers=headers))
+                elif 'package:dash' in tags or ext == 'mpd':
+                    formats.extend(self._extract_mpd_formats(
+                        format_url, video_id, mpd_id='dash', headers=headers))
                 elif ext == 'f4m':
                     # produce broken files
                 elif ext == 'f4m':
                     # produce broken files
-                    continue
+                    pass
                 else:
                     formats.append({
                         'url': format_url,
                 else:
                     formats.append({
                         'url': format_url,
-                        'width': int_or_none(format_data.get('width')),
-                        'height': int_or_none(format_data.get('height')),
+                        'width': int_or_none(playback_set.get('width')),
+                        'height': int_or_none(playback_set.get('height')),
                     })
                     })
+            except ExtractorError as e:
+                if isinstance(e.cause, compat_HTTPError) and e.cause.code == 403:
+                    geo_restricted = True
+                continue
+        if not formats and geo_restricted:
+            self.raise_geo_restricted(countries=['IN'])
         self._sort_formats(formats)
 
         self._sort_formats(formats)
 
+        for f in formats:
+            f.setdefault('http_headers', {}).update(headers)
+
         return {
             'id': video_id,
         return {
             'id': video_id,
-            'title': video_data['episodeTitle'],
+            'title': title,
             'description': video_data.get('description'),
             'duration': int_or_none(video_data.get('duration')),
             '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,
             '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('episodeNo')),
         }
         }
+
+
+class HotStarPlaylistIE(HotStarBaseIE):
+    IE_NAME = 'hotstar:playlist'
+    _VALID_URL = r'https?://(?:www\.)?hotstar\.com/tv/[^/]+/s-\w+/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': {
+            'id': '3_2_26',
+        },
+        'playlist_mincount': 20,
+    }, {
+        'url': 'https://www.hotstar.com/tv/savdhaan-india/s-26/list/extras/t-2480',
+        'only_matching': True,
+    }]
+
+    def _real_extract(self, url):
+        playlist_id = self._match_id(url)
+
+        collection = self._call_api('o/v1/tray/find', playlist_id, 'uqId')
+
+        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)
Backup clone of https://github.com/ytdl-org unavailable since RIAA fucking raid...