download subtitles for mewatch.sg

youtube_dl/extractor/toggle.py

@@ -4,15 +4,10 @@ from __future__ import unicode_literals
 import json
 import re
 
+from ..utils import (ExtractorError, determine_ext, error_to_compat_str,
+                     float_or_none, int_or_none, parse_iso8601,
+                     sanitized_Request)
 from .common import InfoExtractor
-from ..utils import (
-    determine_ext,
-    ExtractorError,
-    float_or_none,
-    int_or_none,
-    parse_iso8601,
-    sanitized_Request,
-)
 
 
 class ToggleIE(InfoExtractor):
@@ -93,7 +88,29 @@ class ToggleIE(InfoExtractor):
     _API_USER = 'tvpapi_147'
     _API_PASS = '11111'
 
-    def _real_extract(self, url):
+    def _get_subtitles(self, video_id):
+        req = sanitized_Request(
+            'https://sub.toggle.sg/toggle_api/v1.0/apiService/getSubtitleFilesForMedia?mediaId={0}'.format(video_id),
+        )
+        subFiles = self._download_json(req, video_id, 'Downloading subtitles json').get('subtitleFiles', [])
+        subtitles = dict()
+
+        if len(subFiles) == 0:
+            return {}
+
+        for sub in subFiles:
+            url = sub.get('subtitleFileUrl')
+            if url is not None:
+                subtitles.setdefault(sub.get('subtitleFileLanguage'), []).append(dict(
+                    url=url,
+                    ext=re.search(
+                        r'\.(\w{3,4})$',
+                        url
+                    ).group(1)
+                ))
+        return subtitles
+
+    def _real_extract(self, url, api_url='https://tvpapi-as.ott.kaltura.com/v3_9/gateways/jsonpostgw.aspx?m=GetMediaInfo'):
         video_id = self._match_id(url)
 
         webpage = self._download_webpage(
@@ -125,9 +142,11 @@ class ToggleIE(InfoExtractor):
             'mediaType': 0,
         }
 
+        try:
             req = sanitized_Request(
-            'http://tvpapi.as.tvinci.com/v2_9/gateways/jsonpostgw.aspx?m=GetMediaInfo',
+                api_url,
-            json.dumps(params).encode('utf-8'))
+                json.dumps(params).encode('utf-8')
+            )
             info = self._download_json(req, video_id, 'Downloading video info json')
 
             title = info['MediaName']
@@ -199,6 +218,8 @@ class ToggleIE(InfoExtractor):
                     })
                 thumbnails.append(thumbnail)
 
+            subtitles = self._get_subtitles(video_id)
+
             return {
                 'id': video_id,
                 'title': title,
@@ -210,4 +231,14 @@ class ToggleIE(InfoExtractor):
                 'like_count': like_count,
                 'thumbnails': thumbnails,
                 'formats': formats,
+                'subtitles': subtitles
             }
+        except ExtractorError as err:
+            self._downloader.report_warning(
+                'Could not get metadata from tvpapi-as.ott.kaltura.com: %s \nTrying tvpapi.as.tvinci.com instead'
+                % error_to_compat_str(err)
+            )
+            self._real_extract(
+                video_id,
+                api_url='http://tvpapi.as.tvinci.com/v2_9/gateways/jsonpostgw.aspx?m=GetMediaInfo'
+            )