[trutv] fix extraction + downloading

youtube_dl/extractor/trutv.py

@@ -6,19 +6,22 @@ import re
 from .turner import TurnerBaseIE
 from ..utils import (
     int_or_none,
-    parse_iso8601,
 )
 
 
 class TruTVIE(TurnerBaseIE):
-    _VALID_URL = r'https?://(?:www\.)?trutv\.com/(?:shows|full-episodes)/(?P<series_slug>[0-9A-Za-z-]+)/(?:videos/(?P<clip_slug>[0-9A-Za-z-]+)|(?P<id>\d+))'
+    # https://www.trutv.com/shows/impractical-jokers/season-8/episode-2/the-closer
+    _VALID_URL = r'https?://(?:www\.)?trutv\.com/shows/[0-9A-Za-z-]+/season-\d+/episode-\d+/(?P<id>[0-9A-Za-z-]+)'
     _TEST = {
-        'url': 'https://www.trutv.com/shows/the-carbonaro-effect/videos/sunlight-activated-flower.html',
+        'url': 'https://www.trutv.com/shows/impractical-jokers/season-8/episode-2/the-closer',
         'info_dict': {
-            'id': 'f16c03beec1e84cd7d1a51f11d8fcc29124cc7f1',
+            'id': '0b90803a0d4bba757085a61cc25be505358cd8b5',
             'ext': 'mp4',
-            'title': 'Sunlight-Activated Flower',
+            'title': 'The Closer',
-            'description': "A customer is stunned when he sees Michael's sunlight-activated flower.",
+            'description': 'Q, Joe, Sal and Murr get tech help from some confused tutors, then play Hot Potato in a shoe store. Plus, the big loser wishes he could press escape during a brutal coffee shop punishment.',
+            'series': 'Impractical Jokers',
+            'season_number': 8,
+            'episode_number': 2,
         },
         'params': {
             # m3u8 download
@@ -27,49 +30,47 @@ class TruTVIE(TurnerBaseIE):
     }
 
     def _real_extract(self, url):
-        series_slug, clip_slug, video_id = re.match(self._VALID_URL, url).groups()
+        episode_slug = self._match_id(url)
+        
+        webpage = self._download_webpage(url, episode_slug)
+        
+        meta = self._parse_json(self._html_search_regex(r'<script type="application/ld\+json">(.+)</script>', webpage, episode_slug), episode_slug)
 
-        if video_id:
+        data = self._parse_json(self._html_search_regex(r'<script type="application/json" data-drupal-selector="drupal-settings-json">(.+)</script>', webpage, episode_slug), episode_slug)
-            path = 'episode'
+        
-            display_id = video_id
+        eps = data['turner_playlist']
-        else:
-            path = 'series/clip'
-            display_id = clip_slug
 
-        data = self._download_json(
+        for ep in eps:
-            'https://api.trutv.com/v2/web/%s/%s/%s' % (path, series_slug, display_id),
+            if ep['url'] in url:
-            display_id)
+                video_data = ep
-        video_data = data['episode'] if video_id else data['info']
+
-        media_id = video_data['mediaId']
+        media_id = video_data['mediaID']
         title = video_data['title'].strip()
 
         info = self._extract_ngtv_info(
             media_id, {}, {
                 'url': url,
                 'site_name': 'truTV',
-                'auth_required': video_data.get('isAuthRequired'),
+                'auth_required': video_data.get('authRequired'),
             })
 
         thumbnails = []
-        for image in video_data.get('images', []):
+        for images in meta.get('image', []):
-            image_url = image.get('srcUrl')
+            for image in images:
-            if not image_url:
+                if not image:
-                continue
+                    continue
-            thumbnails.append({
+                thumbnails.append({
-                'url': image_url,
+                    'url': image,
-                'width': int_or_none(image.get('width')),
+                })
-                'height': int_or_none(image.get('height')),
-            })
 
         info.update({
             'id': media_id,
-            'display_id': display_id,
+            'display_id': video_data.get('videoId'),
             'title': title,
-            'description': video_data.get('description'),
+            'description': video_data.get('shortDescription'),
             'thumbnails': thumbnails,
-            'timestamp': parse_iso8601(video_data.get('publicationDate')),
+            'series': meta.get('partOfSeries').get('name'),
-            'series': video_data.get('showTitle'),
+            'season_number': int_or_none(meta.get('partOfSeason').get('seasonNumber')),
-            'season_number': int_or_none(video_data.get('seasonNum')),
+            'episode_number': int_or_none(meta.get('episodeNumber')),
-            'episode_number': int_or_none(video_data.get('episodeNum')),
         })
         return info

youtube_dl/extractor/turner.py

@@ -196,8 +196,8 @@ class TurnerBaseIE(AdobePassIE):
 
     def _extract_ngtv_info(self, media_id, tokenizer_query, ap_data=None):
         streams_data = self._download_json(
-            'http://medium.ngtv.io/media/%s/tv' % media_id,
+            'http://medium.ngtv.io/media/%s/desktop' % media_id,
-            media_id)['media']['tv']
+            media_id)['media']['desktop']
         duration = None
         chapters = []
         formats = []