Merge ea4948068d323a081324f69a805df14ddab13f6f into 1036478d130c5f2001eca2d7d12558abe601d933

* WEB URLs are, MWEB not
* resolves #33017

youtube_dl/extractor/platzi.py

@@ -1,16 +1,23 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
+import re
+
 from .common import InfoExtractor
 from ..compat import (
-    compat_b64decode,
+    compat_kwargs,
     compat_str,
+    compat_urllib_parse_urlparse,
 )
 from ..utils import (
     clean_html,
+    dict_get,
     ExtractorError,
+    get_element_by_class,
     int_or_none,
+    parse_iso8601,
     str_or_none,
+    strip_or_none,
     try_get,
     url_or_none,
     urlencode_postdata,
@@ -22,6 +29,42 @@ class PlatziBaseIE(InfoExtractor):
     _LOGIN_URL = 'https://platzi.com/login/'
     _NETRC_MACHINE = 'platzi'
 
+    def _raise_extractor_error(self, video_id, reason, expected=True):
+        raise ExtractorError('[%s] %s: %s' % (self.IE_NAME, video_id, reason), expected=expected)
+
+    def _download_webpage(self, url_or_request, video_id, *args, **kwargs):
+        # CF likes Connection: keep-alive and so disfavours Py2
+        # retry on 403 may get in
+        kwargs['expected_status'] = 403
+        # header parameters required fpor Py3 to breach site's CF fence w/o 403
+        headers = kwargs.get('headers') or {}
+        new_hdrs = {}
+        if 'User-Agent' not in headers:
+            headers['User-Agent'] = 'Mozilla/5.0'  # (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/93.0.0.0 Safari/537.36'
+            kwargs['headers'] = new_hdrs = headers
+        if new_hdrs:
+            kwargs = compat_kwargs(kwargs)
+        for _ in range(2):
+            x = super(PlatziBaseIE, self)._download_webpage_handle(url_or_request, video_id, *args, **kwargs)
+            if x is False:
+                return x
+            if x[1].getcode() != 403:
+                break
+            kwargs.pop('expected_status', None)
+            note = kwargs.pop('note', '')
+            kwargs['note'] = (note or 'Downloading webpage') + ' - retrying'
+            kwargs = compat_kwargs(kwargs)
+        path = compat_urllib_parse_urlparse(x[1].geturl())
+        if path == '/':
+            self._raise_extractor_error(video_id, 'Redirected to home page: content expired?')
+        elif path == '/login':
+            self.raise_login_required()
+        else:
+            errs = clean_html(get_element_by_class('Errorpage-text', x[0]))
+            if errs:
+                self._raise_extractor_error(video_id, errs)
+        return x[0]
+
     def _real_initialize(self):
         self._login()
 
@@ -75,6 +118,26 @@ class PlatziIE(PlatziBaseIE):
                     '''
 
     _TESTS = [{
+        'url': 'https://platzi.com/clases/1927-intro-selenium/29383-bienvenida-al-curso',
+        'md5': '0af120f1ffd18a2246f19099d52b83e2',
+        'info_dict': {
+            'id': '29383',
+            'ext': 'mp4',
+            'title': 'Por qué aprender Selenium y qué verás',
+            'description': 'md5:bbe91d2760052ca4054a3149a6580436',
+            'timestamp': 1627400390,
+            'upload_date': '20210727',
+            'creator': 'Héctor Vega',
+            'series': 'Curso de Introducción a Selenium con Python',
+            'duration': 11700,
+            'categories': list,
+        },
+        'params': {
+            'format': 'bestvideo',
+            # 'skip_download': True,
+        },
+        'expected_warnings': ['HTTP Error 401']
+    }, {
         'url': 'https://platzi.com/clases/1311-next-js/12074-creando-nuestra-primera-pagina/',
         'md5': '8f56448241005b561c10f11a595b37e3',
         'info_dict': {
@@ -84,7 +147,7 @@ class PlatziIE(PlatziBaseIE):
             'description': 'md5:4c866e45034fc76412fbf6e60ae008bc',
             'duration': 420,
         },
-        'skip': 'Requires platzi account credentials',
+        'skip': 'Content expired',
     }, {
         'url': 'https://courses.platzi.com/classes/1367-communication-codestream/13430-background/',
         'info_dict': {
@@ -94,10 +157,7 @@ class PlatziIE(PlatziBaseIE):
             'description': 'md5:49c83c09404b15e6e71defaf87f6b305',
             'duration': 360,
         },
-        'skip': 'Requires platzi account credentials',
-        'params': {
-            'skip_download': True,
-        },
+        'skip': 'Content expired',
     }]
 
     def _real_extract(self, url):
@@ -105,50 +165,60 @@ class PlatziIE(PlatziBaseIE):
 
         webpage = self._download_webpage(url, lecture_id)
 
-        data = self._parse_json(
+        data_preloaded_state = self._parse_json(
             self._search_regex(
-                # client_data may contain "};" so that we have to try more
-                # strict regex first
-                (r'client_data\s*=\s*({.+?})\s*;\s*\n',
-                 r'client_data\s*=\s*({.+?})\s*;'),
-                webpage, 'client data'),
+                (r'window\s*.\s*__PRELOADED_STATE__\s*=\s*({.*?});?\s*</script'), webpage, 'client data'),
             lecture_id)
 
-        material = data['initialState']['material']
-        desc = material['description']
-        title = desc['title']
+        video_player = try_get(data_preloaded_state, lambda x: x['videoPlayer'], dict) or {}
+        title = strip_or_none(video_player.get('name')) or self._og_search_title(webpage)
+        servers = try_get(video_player, lambda x: x['video']['servers'], dict) or {}
+        if not servers and try_get(video_player, lambda x: x['blockedInfo']['blocked']):
+            why = video_player['blockedInfo'].get('type') or 'unspecified'
+            if why == 'unlogged':
+                self.raise_login_required()
+            self._raise_extractor_error(lecture_id, 'All video formats blocked because ' + why)
 
         formats = []
-        for server_id, server in material['videos'].items():
-            if not isinstance(server, dict):
+        headers = {'Referer': url}
+        extractions = {
+            'hls': lambda x: formats.extend(self._extract_m3u8_formats(
+                server_json[x], lecture_id, 'mp4',
+                entry_protocol='m3u8_native', m3u8_id='hls',
+                note='Downloading %s m3u8 information' % (server_json.get('id', x), ),
+                headers=headers, fatal=False)),
+            'dash': lambda x: formats.extend(self._extract_mpd_formats(
+                server_json[x], lecture_id, mpd_id='dash',
+                note='Downloading %s MPD manifest' % (server_json.get('id', x), ),
+                headers=headers, fatal=False)),
+        }
+        for server, server_json in servers.items():
+            if not isinstance(server_json, dict):
                 continue
-            for format_id in ('hls', 'dash'):
-                format_url = url_or_none(server.get(format_id))
-                if not format_url:
-                    continue
-                if format_id == 'hls':
-                    formats.extend(self._extract_m3u8_formats(
-                        format_url, lecture_id, 'mp4',
-                        entry_protocol='m3u8_native', m3u8_id=format_id,
-                        note='Downloading %s m3u8 information' % server_id,
-                        fatal=False))
-                elif format_id == 'dash':
-                    formats.extend(self._extract_mpd_formats(
-                        format_url, lecture_id, mpd_id=format_id,
-                        note='Downloading %s MPD manifest' % server_id,
-                        fatal=False))
+            for fmt in server_json.keys():
+                extraction = extractions.get(fmt)
+                if callable(extraction):
+                    extraction(fmt)
         self._sort_formats(formats)
+        for f in formats:
+            f.setdefault('http_headers', {})['Referer'] = headers['Referer']
 
-        content = str_or_none(desc.get('content'))
-        description = (clean_html(compat_b64decode(content).decode('utf-8'))
-                       if content else None)
-        duration = int_or_none(material.get('duration'), invscale=60)
+        def categories():
+            cat = strip_or_none(video_player.get('courseCategory'))
+            if cat:
+                return [cat]
 
         return {
             'id': lecture_id,
             'title': title,
-            'description': description,
-            'duration': duration,
+            'description': clean_html(video_player.get('courseDescription')) or self._og_search_description(webpage),
+            'duration': int_or_none(video_player.get('duration'), invscale=60),
+            'thumbnail': url_or_none(video_player.get('thumbnail')) or self._og_search_thumbnail(webpage),
+            'timestamp': parse_iso8601(dict_get(video_player, ('dateModified', 'datePublished'))),
+            'creator': strip_or_none(video_player.get('teacherName')) or clean_html(get_element_by_class('TeacherDetails-name', webpage)),
+            'comment_count': int_or_none(video_player.get('commentsNumber')),
+            'categories': categories(),
+            'series': strip_or_none(video_player.get('courseTitle')) or None,
             'formats': formats,
         }
 
@@ -157,17 +227,35 @@ class PlatziCourseIE(PlatziBaseIE):
     _VALID_URL = r'''(?x)
                     https?://
                         (?:
-                            platzi\.com/clases|           # es version
-                            courses\.platzi\.com/classes  # en version
+                            (?P<clas>
+                                platzi\.com/clases|           # es version
+                                courses\.platzi\.com/classes  # en version
+                            )|
+                            platzi\.com(?:/(?P<curs>cursos))?
                         )/(?P<id>[^/?\#&]+)
                     '''
     _TESTS = [{
+        'url': 'https://platzi.com/web-angular/',
+        'info_dict': {
+            'id': 'web-angular',
+            'title': 'Frontend con Angular',
+        },
+        'playlist_count': 9,
+    }, {
+        'url': 'https://platzi.com/cursos/angular/',
+        'info_dict': {
+            'id': '2478',
+            'title': 'Curso de Fundamentos de Angular',
+        },
+        'playlist_count': 21,
+    }, {
         'url': 'https://platzi.com/clases/next-js/',
         'info_dict': {
             'id': '1311',
             'title': 'Curso de Next.js',
         },
         'playlist_count': 22,
+        'skip': 'Oops (updating page)',
     }, {
         'url': 'https://courses.platzi.com/classes/communication-codestream/',
         'info_dict': {
@@ -175,23 +263,62 @@ class PlatziCourseIE(PlatziBaseIE):
             'title': 'Codestream Course',
         },
         'playlist_count': 14,
+        'skip': 'Content expired',
     }]
 
+    @classmethod
+    def _match_valid_url(cls, url):
+        return re.match(cls._VALID_URL, url)
+
     @classmethod
     def suitable(cls, url):
         return False if PlatziIE.suitable(url) else super(PlatziCourseIE, cls).suitable(url)
 
+    def __extract_things(self, webpage, thing_id, thing_pattern):
+        return self.playlist_from_matches(
+            re.finditer(thing_pattern, webpage),
+            playlist_id=thing_id,
+            playlist_title=self._og_search_title(webpage, default=None),
+            getter=lambda m: urljoin('https://platzi.com', m.group('path')))
+
+    def _extract_classes(self, webpage, course_id):
+        display_id = course_id
+        course_id = self._search_regex(
+            r'''(["'])courseId\1\s*:\s*(?P<id>\d+)''',
+            webpage, 'course id', group='id', fatal=False) or course_id
+        return self.__extract_things(
+            webpage, course_id,
+            r'''<a\b[^>]+\bhref\s*=\s*['"]?(?P<path>/clases/\d+-%s/[^/]+)'''
+            % (display_id, ))
+
+    def _extract_categories(self, webpage, cat_id):
+        return self.__extract_things(
+            webpage, cat_id,
+            r'''<a\b[^>]+\bhref\s*=\s*['"]?(?P<path>/cursos/[^/]+)''')
+
     def _real_extract(self, url):
-        course_name = self._match_id(url)
 
-        webpage = self._download_webpage(url, course_name)
+        m = self._match_valid_url(url)
+        classes, courses, this_id = m.group('clas', 'curs', 'id')
 
-        props = self._parse_json(
-            self._search_regex(r'data\s*=\s*({.+?})\s*;', webpage, 'data'),
-            course_name)['initialProps']
+        webpage = self._download_webpage(url, this_id)
 
+        if courses:
+            return self._extract_classes(webpage, this_id)
+
+        if not classes:
+            return self._extract_categories(webpage, this_id)
+
+        # this branch now seems always to give "Oops" pages
+        course_name = this_id
+
+        initialData = self._search_regex(
+            (r'window.initialData\s*=\s*({.+?})\s*;\s*\n', r'window.initialData\s*=\s*({.+?})\s*;'),
+            webpage, 'initialData')
+        props = self._parse_json(initialData, course_name, default={})
+        props = try_get(props, lambda x: x['initialProps'], dict) or {}
         entries = []
-        for chapter_num, chapter in enumerate(props['concepts'], 1):
+        for chapter_num, chapter in enumerate(props.get('concepts') or [], 1):
             if not isinstance(chapter, dict):
                 continue
             materials = chapter.get('materials')
@@ -221,4 +348,8 @@ class PlatziCourseIE(PlatziBaseIE):
         course_id = compat_str(try_get(props, lambda x: x['course']['id']))
         course_title = try_get(props, lambda x: x['course']['name'], compat_str)
 
-        return self.playlist_result(entries, course_id, course_title)
+        result = self.playlist_result(entries, course_id, course_title)
+        desc = clean_html(get_element_by_class('RouteDescription-content', webpage))
+        if desc:
+            result['description'] = desc
+        return result

youtube_dl/extractor/youtube.py

@@ -9,6 +9,7 @@ import json
 import os.path
 import random
 import re
+import string
 import time
 import traceback
 
@@ -67,6 +68,7 @@ from ..utils import (
 
 class YoutubeBaseInfoExtractor(InfoExtractor):
     """Provide base functions for Youtube extractors"""
+
     _LOGIN_URL = 'https://accounts.google.com/ServiceLogin'
     _TWOFACTOR_URL = 'https://accounts.google.com/signin/challenge'
 
@@ -138,7 +140,7 @@ class YoutubeBaseInfoExtractor(InfoExtractor):
                 [2, 1, None, 1,
                  'https://accounts.google.com/ServiceLogin?passive=true&continue=https%3A%2F%2Fwww.youtube.com%2Fsignin%3Fnext%3D%252F%26action_handle_signin%3Dtrue%26hl%3Den%26app%3Ddesktop%26feature%3Dsign_in_button&hl=en&service=youtube&uilel=3&requestPath=%2FServiceLogin&Page=PasswordSeparationSignIn',
                  None, [], 4],
-                1, [None, None, []], None, None, None, True
+                1, [None, None, []], None, None, None, True,
             ],
             username,
         ]
@@ -160,7 +162,7 @@ class YoutubeBaseInfoExtractor(InfoExtractor):
             None, 1, None, [1, None, None, None, [password, None, True]],
             [
                 None, None, [2, 1, None, 1, 'https://accounts.google.com/ServiceLogin?passive=true&continue=https%3A%2F%2Fwww.youtube.com%2Fsignin%3Fnext%3D%252F%26action_handle_signin%3Dtrue%26hl%3Den%26app%3Ddesktop%26feature%3Dsign_in_button&hl=en&service=youtube&uilel=3&requestPath=%2FServiceLogin&Page=PasswordSeparationSignIn', None, [], 4],
-                1, [None, None, []], None, None, None, True
+                1, [None, None, []], None, None, None, True,
             ]]
 
         challenge_results = req(
@@ -213,7 +215,7 @@ class YoutubeBaseInfoExtractor(InfoExtractor):
                     user_hash, None, 2, None,
                     [
                         9, None, None, None, None, None, None, None,
-                        [None, tfa_code, True, 2]
+                        [None, tfa_code, True, 2],
                     ]]
 
                 tfa_results = req(
@@ -284,7 +286,7 @@ class YoutubeBaseInfoExtractor(InfoExtractor):
             'client': {
                 'clientName': 'WEB',
                 'clientVersion': '2.20201021.03.00',
-            }
+            },
         },
     }
 
@@ -385,7 +387,7 @@ class YoutubeBaseInfoExtractor(InfoExtractor):
                 'client': {
                     'clientName': 'WEB',
                     'clientVersion': '2.20201021.03.00',
-                }
+                },
             },
             'query': query,
         }
@@ -462,7 +464,7 @@ class YoutubeBaseInfoExtractor(InfoExtractor):
             #       (HTML, videodetails, metadata, renderers)
             'name': ('content', 'author', (('ownerChannelName', None), 'title'), ['text']),
             'url': ('href', 'ownerProfileUrl', 'vanityChannelUrl',
-                    ['navigationEndpoint', 'browseEndpoint', 'canonicalBaseUrl'])
+                    ['navigationEndpoint', 'browseEndpoint', 'canonicalBaseUrl']),
         }
         if any((videodetails, metadata, renderers)):
             result = (
@@ -671,7 +673,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/8KVIDEO',
                 'description': '',
                 'uploader': '8KVIDEO',
-                'title': 'UHDTV TEST 8K VIDEO.mp4'
+                'title': 'UHDTV TEST 8K VIDEO.mp4',
             },
             'params': {
                 'youtube_include_dash_manifest': True,
@@ -711,7 +713,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                 'uploader_url': r're:https?://(?:www\.)?youtube\.com/@theamazingatheist',
                 'title': 'Burning Everyone\'s Koran',
                 'description': 'SUBSCRIBE: http://www.youtube.com/saturninefilms \r\n\r\nEven Obama has taken a stand against freedom on this issue: http://www.huffingtonpost.com/2010/09/09/obama-gma-interview-quran_n_710282.html',
-            }
+            },
         },
         # Age-gated videos
         {
@@ -839,7 +841,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
             },
             'expected_warnings': [
                 'DASH manifest missing',
-            ]
+            ],
         },
         # Olympics (https://github.com/ytdl-org/youtube-dl/issues/4431)
         {
@@ -1820,8 +1822,8 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
 
         # cpn generation algorithm is reverse engineered from base.js.
         # In fact it works even with dummy cpn.
-        CPN_ALPHABET = 'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789-_'
-        cpn = ''.join(CPN_ALPHABET[random.randint(0, 256) & 63] for _ in range(0, 16))
+        CPN_ALPHABET = string.ascii_letters + string.digits + '-_'
+        cpn = ''.join(CPN_ALPHABET[random.randint(0, 256) & 63] for _ in range(16))
 
         # more consistent results setting it to right before the end
         qs = parse_qs(playback_url)
@@ -1881,8 +1883,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
         mobj = re.match(cls._VALID_URL, url, re.VERBOSE)
         if mobj is None:
             raise ExtractorError('Invalid URL: %s' % url)
-        video_id = mobj.group(2)
-        return video_id
+        return mobj.group(2)
 
     def _extract_chapters_from_json(self, data, video_id, duration):
         chapters_list = try_get(
@@ -2035,7 +2036,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
             headers = {
                 'X-YouTube-Client-Name': '85',
                 'X-YouTube-Client-Version': '2.0',
-                'Origin': 'https://www.youtube.com'
+                'Origin': 'https://www.youtube.com',
             }
 
             video_info = self._call_api('player', query, video_id, fatal=False, headers=headers)
@@ -2064,8 +2065,8 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
             return ''.join([r['text'] for r in runs if isinstance(r.get('text'), compat_str)])
 
         search_meta = (
-            lambda x: self._html_search_meta(x, webpage, default=None)) \
-            if webpage else lambda x: None
+            (lambda x: self._html_search_meta(x, webpage, default=None))
+            if webpage else lambda _: None)
 
         video_details = player_response.get('videoDetails') or {}
         microformat = try_get(
@@ -2137,7 +2138,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
         def build_fragments(f):
             return LazyList({
                 'url': update_url_query(f['url'], {
-                    'range': '{0}-{1}'.format(range_start, min(range_start + CHUNK_SIZE - 1, f['filesize']))
+                    'range': '{0}-{1}'.format(range_start, min(range_start + CHUNK_SIZE - 1, f['filesize'])),
                 })
             } for range_start in range(0, f['filesize'], CHUNK_SIZE))
 
@@ -2236,7 +2237,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                     'protocol': 'http_dash_segments',
                     'fragments': build_fragments(dct),
                 } if dct['filesize'] else {
-                    'downloader_options': {'http_chunk_size': CHUNK_SIZE}  # No longer useful?
+                    'downloader_options': {'http_chunk_size': CHUNK_SIZE},  # No longer useful?
                 })
 
             formats.append(dct)
@@ -2414,9 +2415,9 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
             'is_live': is_live,
         }
 
-        pctr = try_get(
+        pctr = traverse_obj(
             player_response,
-            lambda x: x['captions']['playerCaptionsTracklistRenderer'], dict)
+            ('captions', 'playerCaptionsTracklistRenderer', T(dict)))
         if pctr:
             def process_language(container, base_url, lang_code, query):
                 lang_subs = []
@@ -2430,31 +2431,34 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                     })
                 container[lang_code] = lang_subs
 
-            subtitles = {}
-            for caption_track in (pctr.get('captionTracks') or []):
-                base_url = caption_track.get('baseUrl')
-                if not base_url:
-                    continue
-                if caption_track.get('kind') != 'asr':
-                    lang_code = caption_track.get('languageCode')
-                    if not lang_code:
+            def process_subtitles():
+                subtitles = {}
+                for caption_track in traverse_obj(pctr, (
+                        'captionTracks', lambda _, v: v.get('baseUrl'))):
+                    base_url = self._yt_urljoin(caption_track['baseUrl'])
+                    if not base_url:
                         continue
-                    process_language(
-                        subtitles, base_url, lang_code, {})
-                    continue
-                automatic_captions = {}
-                for translation_language in (pctr.get('translationLanguages') or []):
-                    translation_language_code = translation_language.get('languageCode')
-                    if not translation_language_code:
+                    if caption_track.get('kind') != 'asr':
+                        lang_code = caption_track.get('languageCode')
+                        if not lang_code:
+                            continue
+                        process_language(
+                            subtitles, base_url, lang_code, {})
                         continue
-                    process_language(
-                        automatic_captions, base_url, translation_language_code,
-                        {'tlang': translation_language_code})
-                info['automatic_captions'] = automatic_captions
-            info['subtitles'] = subtitles
+                    automatic_captions = {}
+                    for translation_language in traverse_obj(pctr, (
+                            'translationLanguages', lambda _, v: v.get('languageCode'))):
+                        translation_language_code = translation_language['languageCode']
+                        process_language(
+                            automatic_captions, base_url, translation_language_code,
+                            {'tlang': translation_language_code})
+                    info['automatic_captions'] = automatic_captions
+                info['subtitles'] = subtitles
+
+            process_subtitles()
 
         parsed_url = compat_urllib_parse_urlparse(url)
-        for component in [parsed_url.fragment, parsed_url.query]:
+        for component in (parsed_url.fragment, parsed_url.query):
             query = compat_parse_qs(component)
             for k, v in query.items():
                 for d_k, s_ks in [('start', ('start', 't')), ('end', ('end',))]:
@@ -2684,7 +2688,7 @@ class YoutubeTabIE(YoutubeBaseInfoExtractor):
             'title': 'Super Cooper Shorts - Shorts',
             'uploader': 'Super Cooper Shorts',
             'uploader_id': '@SuperCooperShorts',
-        }
+        },
     }, {
         # Channel that does not have a Shorts tab. Test should just download videos on Home tab instead
         'url': 'https://www.youtube.com/@emergencyawesome/shorts',
@@ -2738,7 +2742,7 @@ class YoutubeTabIE(YoutubeBaseInfoExtractor):
             'description': 'md5:609399d937ea957b0f53cbffb747a14c',
             'uploader': 'ThirstForScience',
             'uploader_id': '@ThirstForScience',
-        }
+        },
     }, {
         'url': 'https://www.youtube.com/c/ChristophLaimer/playlists',
         'only_matching': True,
@@ -3037,7 +3041,7 @@ class YoutubeTabIE(YoutubeBaseInfoExtractor):
             'uploader': '3Blue1Brown',
             'uploader_id': '@3blue1brown',
             'channel_id': 'UCYO_jab_esuFRV4b17AJtAw',
-        }
+        },
     }]
 
     @classmethod
@@ -3335,7 +3339,7 @@ class YoutubeTabIE(YoutubeBaseInfoExtractor):
             'client': {
                 'clientName': 'WEB',
                 'clientVersion': client_version,
-            }
+            },
         }
         visitor_data = try_get(context, lambda x: x['client']['visitorData'], compat_str)
 
@@ -3354,7 +3358,7 @@ class YoutubeTabIE(YoutubeBaseInfoExtractor):
                 headers['x-goog-visitor-id'] = visitor_data
             data['continuation'] = continuation['continuation']
             data['clickTracking'] = {
-                'clickTrackingParams': continuation['itct']
+                'clickTrackingParams': continuation['itct'],
             }
             count = 0
             retries = 3
@@ -3613,7 +3617,7 @@ class YoutubePlaylistIE(InfoExtractor):
             'uploader': 'milan',
             'uploader_id': '@milan5503',
             'channel_id': 'UCEI1-PVPcYXjB73Hfelbmaw',
-        }
+        },
     }, {
         'url': 'http://www.youtube.com/embed/_xDOZElKyNU?list=PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
         'playlist_mincount': 455,
@@ -3623,7 +3627,7 @@ class YoutubePlaylistIE(InfoExtractor):
             'uploader': 'LBK',
             'uploader_id': '@music_king',
             'channel_id': 'UC21nz3_MesPLqtDqwdvnoxA',
-        }
+        },
     }, {
         'url': 'TLGGrESM50VT6acwMjAyMjAxNw',
         'only_matching': True,
@@ -3734,7 +3738,7 @@ class YoutubeSearchIE(SearchInfoExtractor, YoutubeBaseInfoExtractor):
         'info_dict': {
             'id': 'youtube-dl test video',
             'title': 'youtube-dl test video',
-        }
+        },
     }]
 
     def _get_n_results(self, query, n):
@@ -3754,7 +3758,7 @@ class YoutubeSearchDateIE(YoutubeSearchIE):
         'info_dict': {
             'id': 'youtube-dl test video',
             'title': 'youtube-dl test video',
-        }
+        },
     }]
 
 
@@ -3769,7 +3773,7 @@ class YoutubeSearchURLIE(YoutubeBaseInfoExtractor):
             'id': 'youtube-dl test video',
             'title': 'youtube-dl test video',
         },
-        'params': {'playlistend': 5}
+        'params': {'playlistend': 5},
     }, {
         'url': 'https://www.youtube.com/results?q=test&sp=EgQIBBgB',
         'only_matching': True,
@@ -3785,6 +3789,7 @@ class YoutubeSearchURLIE(YoutubeBaseInfoExtractor):
 class YoutubeFeedsInfoExtractor(YoutubeTabIE):
     """
     Base class for feed extractors
+
     Subclasses must define the _FEED_NAME property.
     """
     _LOGIN_REQUIRED = True