]> jfr.im git - yt-dlp.git/blobdiff - yt_dlp/extractor/sonyliv.py
[cleanup] Standardize `import datetime as dt` (#8978)
[yt-dlp.git] / yt_dlp / extractor / sonyliv.py
index 17d28478f9c79feb97cf08eb4e689f314a8e9001..7c914acbed231a9210dd804ea6a94fdaf8a4511e 100644 (file)
@@ -1,4 +1,5 @@
-import datetime
+import datetime as dt
+import itertools
 import json
 import math
 import random
@@ -6,12 +7,14 @@
 import uuid
 
 from .common import InfoExtractor
-from ..compat import compat_HTTPError
+from ..networking.exceptions import HTTPError
 from ..utils import (
     ExtractorError,
     int_or_none,
-    try_get,
+    jwt_decode_hs256,
+    try_call,
 )
+from ..utils.traversal import traverse_obj
 
 
 class SonyLIVIE(InfoExtractor):
@@ -77,8 +80,10 @@ def _perform_login(self, username, password):
         self._HEADERS['device_id'] = self._get_device_id()
         self._HEADERS['content-type'] = 'application/json'
 
-        if username.lower() == 'token' and len(password) > 1198:
+        if username.lower() == 'token' and try_call(lambda: jwt_decode_hs256(password)):
             self._HEADERS['authorization'] = password
+            self.report_login()
+            return
         elif len(username) != 10 or not username.isdigit():
             raise ExtractorError(f'Invalid username/password; {self._LOGIN_HINT}')
 
@@ -89,7 +94,7 @@ def _perform_login(self, username, password):
                 'mobileNumber': username,
                 'channelPartnerID': 'MSMIND',
                 'country': 'IN',
-                'timestamp': datetime.datetime.now().strftime('%Y-%m-%dT%H:%M:%S.%MZ'),
+                'timestamp': dt.datetime.now().strftime('%Y-%m-%dT%H:%M:%S.%MZ'),
                 'otpSize': 6,
                 'loginType': 'REGISTERORSIGNIN',
                 'isMobileMandatory': True,
@@ -106,7 +111,7 @@ def _perform_login(self, username, password):
                 'otp': self._get_tfa_info('OTP'),
                 'dmaId': 'IN',
                 'ageConfirmation': True,
-                'timestamp': datetime.datetime.now().strftime('%Y-%m-%dT%H:%M:%S.%MZ'),
+                'timestamp': dt.datetime.now().strftime('%Y-%m-%dT%H:%M:%S.%MZ'),
                 'isMobileMandatory': True,
             }).encode())
         if otp_verify_json['resultCode'] == 'KO':
@@ -119,12 +124,12 @@ def _call_api(self, version, path, video_id):
                 'https://apiv2.sonyliv.com/AGL/%s/A/ENG/WEB/%s' % (version, path),
                 video_id, headers=self._HEADERS)['resultObj']
         except ExtractorError as e:
-            if isinstance(e.cause, compat_HTTPError) and e.cause.code == 406 and self._parse_json(
-                    e.cause.read().decode(), video_id)['message'] == 'Please subscribe to watch this content':
+            if isinstance(e.cause, HTTPError) and e.cause.status == 406 and self._parse_json(
+                    e.cause.response.read().decode(), video_id)['message'] == 'Please subscribe to watch this content':
                 self.raise_login_required(self._LOGIN_HINT, method=None)
-            if isinstance(e.cause, compat_HTTPError) and e.cause.code == 403:
+            if isinstance(e.cause, HTTPError) and e.cause.status == 403:
                 message = self._parse_json(
-                    e.cause.read().decode(), video_id)['message']
+                    e.cause.response.read().decode(), video_id)['message']
                 if message == 'Geoblocked Country':
                     self.raise_geo_restricted(countries=self._GEO_COUNTRIES)
                 raise ExtractorError(message)
@@ -150,7 +155,6 @@ def _real_extract(self, url):
             video_id, 'mp4', m3u8_id='hls', headers=headers, fatal=False))
         for f in formats:
             f.setdefault('http_headers', {}).update(headers)
-        self._sort_formats(formats)
 
         metadata = self._call_api(
             '1.6', 'IN/DETAIL/' + video_id, video_id)['containers'][0]['metadata']
@@ -180,17 +184,21 @@ def _real_extract(self, url):
 
 
 class SonyLIVSeriesIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?sonyliv\.com/shows/[^/?#&]+-(?P<id>\d{10})$'
+    _VALID_URL = r'https?://(?:www\.)?sonyliv\.com/shows/[^/?#&]+-(?P<id>\d{10})/?(?:$|[?#])'
     _TESTS = [{
         'url': 'https://www.sonyliv.com/shows/adaalat-1700000091',
-        'playlist_mincount': 456,
+        'playlist_mincount': 452,
         'info_dict': {
             'id': '1700000091',
         },
+    }, {
+        'url': 'https://www.sonyliv.com/shows/beyhadh-1700000007/',
+        'playlist_mincount': 358,
+        'info_dict': {
+            'id': '1700000007',
+        },
     }]
-    _API_SHOW_URL = "https://apiv2.sonyliv.com/AGL/1.9/R/ENG/WEB/IN/DL/DETAIL/{}?kids_safe=false&from=0&to=49"
-    _API_EPISODES_URL = "https://apiv2.sonyliv.com/AGL/1.4/R/ENG/WEB/IN/CONTENT/DETAIL/BUNDLE/{}?from=0&to=1000&orderBy=episodeNumber&sortOrder=asc"
-    _API_SECURITY_URL = 'https://apiv2.sonyliv.com/AGL/1.4/A/ENG/WEB/ALL/GETTOKEN'
+    _API_BASE = 'https://apiv2.sonyliv.com/AGL'
 
     def _entries(self, show_id):
         headers = {
@@ -198,19 +206,34 @@ def _entries(self, show_id):
             'Referer': 'https://www.sonyliv.com',
         }
         headers['security_token'] = self._download_json(
-            self._API_SECURITY_URL, video_id=show_id, headers=headers,
-            note='Downloading security token')['resultObj']
-        seasons = try_get(
-            self._download_json(self._API_SHOW_URL.format(show_id), video_id=show_id, headers=headers),
-            lambda x: x['resultObj']['containers'][0]['containers'], list)
-        for season in seasons or []:
-            season_id = season['id']
-            episodes = try_get(
-                self._download_json(self._API_EPISODES_URL.format(season_id), video_id=season_id, headers=headers),
-                lambda x: x['resultObj']['containers'][0]['containers'], list)
-            for episode in episodes or []:
-                video_id = episode.get('id')
-                yield self.url_result('sonyliv:%s' % video_id, ie=SonyLIVIE.ie_key(), video_id=video_id)
+            f'{self._API_BASE}/1.4/A/ENG/WEB/ALL/GETTOKEN', show_id,
+            'Downloading security token', headers=headers)['resultObj']
+        seasons = traverse_obj(self._download_json(
+            f'{self._API_BASE}/1.9/R/ENG/WEB/IN/DL/DETAIL/{show_id}', show_id,
+            'Downloading series JSON', headers=headers, query={
+                'kids_safe': 'false',
+                'from': '0',
+                'to': '49',
+            }), ('resultObj', 'containers', 0, 'containers', lambda _, v: int_or_none(v['id'])))
+        for season in seasons:
+            season_id = str(season['id'])
+            note = traverse_obj(season, ('metadata', 'title', {str})) or 'season'
+            cursor = 0
+            for page_num in itertools.count(1):
+                episodes = traverse_obj(self._download_json(
+                    f'{self._API_BASE}/1.4/R/ENG/WEB/IN/CONTENT/DETAIL/BUNDLE/{season_id}',
+                    season_id, f'Downloading {note} page {page_num} JSON', headers=headers, query={
+                        'from': str(cursor),
+                        'to': str(cursor + 99),
+                        'orderBy': 'episodeNumber',
+                        'sortOrder': 'asc',
+                    }), ('resultObj', 'containers', 0, 'containers', lambda _, v: int_or_none(v['id'])))
+                if not episodes:
+                    break
+                for episode in episodes:
+                    video_id = str(episode['id'])
+                    yield self.url_result(f'sonyliv:{video_id}', SonyLIVIE, video_id)
+                cursor += 100
 
     def _real_extract(self, url):
         show_id = self._match_id(url)