[parliamentlive.tv] Fix extractor (#1153)

Closes #1139 
Authored by: u-spec-png

yt_dlp/extractor/parliamentliveuk.py

@@ -1,6 +1,14 @@
+# coding: utf-8
 from __future__ import unicode_literals
 
+import json
+import uuid
+
 from .common import InfoExtractor
+from ..utils import (
+    unified_timestamp,
+    try_get,
+)
 
 
 class ParliamentLiveUKIE(InfoExtractor):
@@ -11,12 +19,14 @@ class ParliamentLiveUKIE(InfoExtractor):
     _TESTS = [{
         'url': 'http://parliamentlive.tv/Event/Index/c1e9d44d-fd6c-4263-b50f-97ed26cc998b',
         'info_dict': {
-            'id': '1_af9nv9ym',
+            'id': 'c1e9d44d-fd6c-4263-b50f-97ed26cc998b',
             'ext': 'mp4',
             'title': 'Home Affairs Committee',
-            'uploader_id': 'FFMPEG-01',
+            'timestamp': 1395153872,
-            'timestamp': 1422696664,
+            'upload_date': '20140318',
-            'upload_date': '20150131',
+        },
+        'params': {
+            'format': 'bestvideo',
         },
     }, {
         'url': 'http://parliamentlive.tv/event/index/3f24936f-130f-40bf-9a5d-b3d6479da6a4',
@@ -25,19 +35,49 @@ class ParliamentLiveUKIE(InfoExtractor):
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
-        webpage = self._download_webpage(
+        video_info = self._download_json(f'https://www.parliamentlive.tv/Event/GetShareVideo/{video_id}', video_id)
-            'http://vodplayer.parliamentlive.tv/?mid=' + video_id, video_id)
+        _DEVICE_ID = str(uuid.uuid4())
-        widget_config = self._parse_json(self._search_regex(
+        auth = 'Bearer ' + self._download_json(
-            r'(?s)kWidgetConfig\s*=\s*({.+});',
+            'https://exposure.api.redbee.live/v2/customer/UKParliament/businessunit/ParliamentLive/auth/anonymous',
-            webpage, 'kaltura widget config'), video_id)
+            video_id, headers={
-        kaltura_url = 'kaltura:%s:%s' % (
+                'Origin': 'https://videoplayback.parliamentlive.tv',
-            widget_config['wid'][1:], widget_config['entry_id'])
+                'Accept': 'application/json, text/plain, */*',
-        event_title = self._download_json(
+                'Content-Type': 'application/json;charset=utf-8'
-            'http://parliamentlive.tv/Event/GetShareVideo/' + video_id, video_id)['event']['title']
+            }, data=json.dumps({
-        return {
+                'deviceId': _DEVICE_ID,
-            '_type': 'url_transparent',
+                'device': {
-            'title': event_title,
+                    'deviceId': _DEVICE_ID,
-            'description': '',
+                    'width': 653,
-            'url': kaltura_url,
+                    'height': 368,
-            'ie_key': 'Kaltura',
+                    'type': 'WEB',
+                    'name': ' Mozilla Firefox 91'
+                }
+            }).encode('utf-8'))['sessionToken']
+
+        video_urls = self._download_json(
+            f'https://exposure.api.redbee.live/v2/customer/UKParliament/businessunit/ParliamentLive/entitlement/{video_id}/play',
+            video_id, headers={'Authorization': auth, 'Accept': 'application/json, text/plain, */*'})['formats']
+
+        formats = []
+        for format in video_urls:
+            if not format.get('mediaLocator'):
+                continue
+            if format.get('format') == 'DASH':
+                formats.extend(self._extract_mpd_formats(
+                    format['mediaLocator'], video_id, mpd_id='dash', fatal=False))
+            elif format.get('format') == 'SMOOTHSTREAMING':
+                formats.extend(self._extract_ism_formats(
+                    format['mediaLocator'], video_id, ism_id='ism', fatal=False))
+            elif format.get('format') == 'HLS':
+                formats.extend(self._extract_m3u8_formats(
+                    format['mediaLocator'], video_id, m3u8_id='hls', fatal=False))
+
+        self._sort_formats(formats)
+
+        return {
+            'id': video_id,
+            'formats': formats,
+            'title': video_info['event']['title'],
+            'timestamp': unified_timestamp(try_get(video_info, lambda x: x['event']['publishedStartTime'])),
+            'thumbnail': video_info.get('thumbnailUrl'),
         }