From a2751238169574a451c9d3354b3680f91e633c92 Mon Sep 17 00:00:00 2001
From: girst <tobi@isticktoit.net>
Date: Tue, 18 Aug 2020 22:05:43 +0200
Subject: [PATCH] fix learning playlists

---
 app/common/innertube.py | 30 +++++++++++++++++++++---------
 1 file changed, 21 insertions(+), 9 deletions(-)

diff --git a/app/common/innertube.py b/app/common/innertube.py
index f2c97d6..3320d55 100644
--- a/app/common/innertube.py
+++ b/app/common/innertube.py
@@ -1,6 +1,7 @@
 # functions that deal with parsing data from youtube's internal API ("innertube")
 
 from urllib.parse import parse_qs, urlparse
+import re
 
 def findall(obj, key):
     """
@@ -133,18 +134,22 @@ def parse_result_items(items):
                     ['navigationEndpoint']['browseEndpoint']['browseId'],
                 'length': content.get('lengthText',{}).get('simpleText') \
                     if not is_live else 'LIVE', # "44:07", "1:41:50"
-                'views': toInt(content.get('viewCountText',{}).get('simpleText') or # "123,456 views"
-                    listget(content.get('viewCountText',{}).get('runs'),0,{}).get('text')), # "1,234 watching"
+                'views': toInt(content.get('viewCountText',{}).get('simpleText') or # "123,456 views", ...
+                    listget(content.get('viewCountText',{}).get('runs',[]),0,{}).get('text')) or 0, # ... "1,234 watching", absent on 0 views
                 'published': age(content.get('publishedTimeText',{}).get('simpleText')),
             }})
         elif key == 'playlistRenderer':
             results.append({'type': 'PLAYLIST', 'content': {
-                'playlist_id': content['navigationEndpoint']['watchEndpoint']['playlistId'],
-                'video_id': content['navigationEndpoint']['watchEndpoint']['videoId'],
+                'playlist_id': content['navigationEndpoint'].get('watchEndpoint',{}).get('playlistId') or \
+                    content.get('playlistId'), # COURSE/"learning playlist"
+                'video_id': content['navigationEndpoint'].get('watchEndpoint',{}).get('videoId') or \
+                    videoid_from_thumbnail(content), # learning playlist
                 'title': content['title']['simpleText'],
-                'author': content['longBylineText']['runs'][0]['text'] or
-                     content['shortBylineText']['runs'][0]['text'],
-                'channel_id': content['longBylineText']['runs'][0]['navigationEndpoint']['browseEndpoint']['browseId'], # OR .shortBylineText
+                # Note: learning playlists have no author/channel_id
+                'author': listget(content.get('longBylineText',{}).get('runs',[]),0,{}).get('text') or
+                     listget(content.get('shortBylineText',{}).get('runs',[]),0,{}).get('text'),
+                'channel_id': listget(content.get('longBylineText',{}).get('runs',[]),0,{}) \
+                    .get('navigationEndpoint',{}).get('browseEndpoint',{}).get('browseId'), # OR .shortBylineText
                 'n_videos': toInt(content['videoCount']),
             }})
         elif key == 'radioRenderer': # "Mix" playlists
@@ -167,7 +172,8 @@ def parse_result_items(items):
                 'subscribers': content.get('subscriberCountText',{}).get('simpleText'), # "2.47K subscribers"
             }})
         elif key == 'shelfRenderer':
-            r, e = parse_result_items(content['content']['verticalListRenderer']['items'])
+            subkey = next(iter(content['content'].keys()), {}) #verticalListRenderer/horizontalMovieListRenderer
+            r, e = parse_result_items(content['content'][subkey]['items'])
             results.extend(r)
             extras.extend(e)
         elif key == 'movieRenderer': # movies to buy/rent
@@ -287,6 +293,12 @@ def parse_endcard(card):
         log_unknown_card(card)
         return None
 
+def videoid_from_thumbnail(content):
+    # learning playlist; example: PL96C35uN7xGJu6skU4TBYrIWxggkZBrF5 (/user/enyay/playlists)
+    return re.match(r"https?://i.ytimg.com/vi/([-_0-9a-zA-Z]{11})|()",
+        listget(listget(content.get('thumbnails',[]),0,{}).get('thumbnails',[]),0,{}).get('url','')
+    ).group(1)
+
 def parse_channel_items(items, channel_id, author):
     result = []
     extra = []
@@ -311,7 +323,7 @@ def parse_channel_items(items, channel_id, author):
         elif key == "gridPlaylistRenderer" or key == "playlistRenderer":
             result.append({'type': 'PLAYLIST', 'content': {
                 'playlist_id': content['navigationEndpoint'].get('watchEndpoint',{}).get('playlistId') or content.get('playlistId'),
-                'video_id': content['navigationEndpoint'].get('watchEndpoint',{}).get('videoId',{}),
+                'video_id': content['navigationEndpoint'].get('watchEndpoint',{}).get('videoId',{}) or videoid_from_thumbnail(content),
                 'title': (content['title'].get('simpleText') or # playlistRenderer
                     content['title']['runs'][0]['text']), # gridPlaylistRenderer
                 'author': author,
-- 
2.39.3