yt_dlp/extractor/patreon.py

   1 import functools
   2 import itertools
   3 import urllib.parse
   4
   5 from .common import InfoExtractor
   6 from .sproutvideo import VidsIoIE
   7 from .vimeo import VimeoIE
   8 from ..networking.exceptions import HTTPError
   9 from ..utils import (
  10     KNOWN_EXTENSIONS,
  11     ExtractorError,
  12     clean_html,
  13     determine_ext,
  14     int_or_none,
  15     mimetype2ext,
  16     parse_iso8601,
  17     smuggle_url,
  18     str_or_none,
  19     url_or_none,
  20     urljoin,
  21 )
  22 from ..utils.traversal import traverse_obj, value
  23
  24
  25 class PatreonBaseIE(InfoExtractor):
  26     @functools.cached_property
  27     def patreon_user_agent(self):
  28         # Patreon mobile UA is needed to avoid triggering Cloudflare anti-bot protection.
  29         # Newer UA yields higher res m3u8 formats for locked posts, but gives 401 if not logged-in
  30         if self._get_cookies('https://www.patreon.com/').get('session_id'):
  31             return 'Patreon/72.2.28 (Android; Android 14; Scale/2.10)'
  32         return 'Patreon/7.6.28 (Android; Android 11; Scale/2.10)'
  33
  34     def _call_api(self, ep, item_id, query=None, headers=None, fatal=True, note=None):
  35         if headers is None:
  36             headers = {}
  37         if 'User-Agent' not in headers:
  38             headers['User-Agent'] = self.patreon_user_agent
  39         if query:
  40             query.update({'json-api-version': 1.0})
  41
  42         try:
  43             return self._download_json(
  44                 f'https://www.patreon.com/api/{ep}',
  45                 item_id, note=note if note else 'Downloading API JSON',
  46                 query=query, fatal=fatal, headers=headers)
  47         except ExtractorError as e:
  48             if not isinstance(e.cause, HTTPError) or mimetype2ext(e.cause.response.headers.get('Content-Type')) != 'json':
  49                 raise
  50             err_json = self._parse_json(self._webpage_read_content(e.cause.response, None, item_id), item_id, fatal=False)
  51             err_message = traverse_obj(err_json, ('errors', ..., 'detail'), get_all=False)
  52             if err_message:
  53                 raise ExtractorError(f'Patreon said: {err_message}', expected=True)
  54             raise
  55
  56
  57 class PatreonIE(PatreonBaseIE):
  58     IE_NAME = 'patreon'
  59     _VALID_URL = r'https?://(?:www\.)?patreon\.com/(?:creation\?hid=|posts/(?:[\w-]+-)?)(?P<id>\d+)'
  60     _TESTS = [{
  61         'url': 'http://www.patreon.com/creation?hid=743933',
  62         'md5': 'e25505eec1053a6e6813b8ed369875cc',
  63         'info_dict': {
  64             'id': '743933',
  65             'ext': 'mp3',
  66             'title': 'Episode 166: David Smalley of Dogma Debate',
  67             'description': 'md5:34d207dd29aa90e24f1b3f58841b81c7',
  68             'uploader': 'Cognitive Dissonance Podcast',
  69             'thumbnail': 're:^https?://.*$',
  70             'timestamp': 1406473987,
  71             'upload_date': '20140727',
  72             'uploader_id': '87145',
  73             'like_count': int,
  74             'comment_count': int,
  75             'uploader_url': 'https://www.patreon.com/dissonancepod',
  76             'channel_id': '80642',
  77             'channel_url': 'https://www.patreon.com/dissonancepod',
  78             'channel_follower_count': int,
  79         },
  80     }, {
  81         'url': 'http://www.patreon.com/creation?hid=754133',
  82         'md5': '3eb09345bf44bf60451b8b0b81759d0a',
  83         'info_dict': {
  84             'id': '754133',
  85             'ext': 'mp3',
  86             'title': 'CD 167 Extra',
  87             'uploader': 'Cognitive Dissonance Podcast',
  88             'thumbnail': 're:^https?://.*$',
  89             'like_count': int,
  90             'comment_count': int,
  91             'uploader_url': 'https://www.patreon.com/dissonancepod',
  92         },
  93         'skip': 'Patron-only content',
  94     }, {
  95         'url': 'https://www.patreon.com/creation?hid=1682498',
  96         'info_dict': {
  97             'id': 'SU4fj_aEMVw',
  98             'ext': 'mp4',
  99             'title': 'I\'m on Patreon!',
 100             'uploader': 'TraciJHines',
 101             'thumbnail': 're:^https?://.*$',
 102             'upload_date': '20150211',
 103             'description': 'md5:8af6425f50bd46fbf29f3db0fc3a8364',
 104             'uploader_id': '@TraciHinesMusic',
 105             'categories': ['Entertainment'],
 106             'duration': 282,
 107             'view_count': int,
 108             'tags': 'count:39',
 109             'age_limit': 0,
 110             'channel': 'TraciJHines',
 111             'channel_url': 'https://www.youtube.com/channel/UCGLim4T2loE5rwCMdpCIPVg',
 112             'live_status': 'not_live',
 113             'like_count': int,
 114             'channel_id': 'UCGLim4T2loE5rwCMdpCIPVg',
 115             'availability': 'public',
 116             'channel_follower_count': int,
 117             'playable_in_embed': True,
 118             'uploader_url': 'https://www.youtube.com/@TraciHinesMusic',
 119             'comment_count': int,
 120             'channel_is_verified': True,
 121             'chapters': 'count:4',
 122             'timestamp': 1423689666,
 123         },
 124         'params': {
 125             'noplaylist': True,
 126             'skip_download': True,
 127         },
 128     }, {
 129         'url': 'https://www.patreon.com/posts/episode-166-of-743933',
 130         'only_matching': True,
 131     }, {
 132         'url': 'https://www.patreon.com/posts/743933',
 133         'only_matching': True,
 134     }, {
 135         'url': 'https://www.patreon.com/posts/kitchen-as-seen-51706779',
 136         'md5': '96656690071f6d64895866008484251b',
 137         'info_dict': {
 138             'id': '555089736',
 139             'ext': 'mp4',
 140             'title': 'KITCHEN AS SEEN ON DEEZ NUTS EXTENDED!',
 141             'uploader': 'Cold Ones',
 142             'thumbnail': 're:^https?://.*$',
 143             'upload_date': '20210526',
 144             'description': 'md5:557a409bd79d3898689419094934ba79',
 145             'uploader_id': '14936315',
 146         },
 147         'skip': 'Patron-only content',
 148     }, {
 149         # m3u8 video (https://github.com/yt-dlp/yt-dlp/issues/2277)
 150         'url': 'https://www.patreon.com/posts/video-sketchbook-32452882',
 151         'info_dict': {
 152             'id': '32452882',
 153             'ext': 'mp4',
 154             'comment_count': int,
 155             'uploader_id': '4301314',
 156             'like_count': int,
 157             'timestamp': 1576696962,
 158             'upload_date': '20191218',
 159             'thumbnail': r're:^https?://.*$',
 160             'uploader_url': 'https://www.patreon.com/loish',
 161             'description': 'md5:e2693e97ee299c8ece47ffdb67e7d9d2',
 162             'title': 'VIDEO // sketchbook flipthrough',
 163             'uploader': 'Loish ',
 164             'tags': ['sketchbook', 'video'],
 165             'channel_id': '1641751',
 166             'channel_url': 'https://www.patreon.com/loish',
 167             'channel_follower_count': int,
 168         },
 169     }, {
 170         # bad videos under media (if media is included). Real one is under post_file
 171         'url': 'https://www.patreon.com/posts/premium-access-70282931',
 172         'info_dict': {
 173             'id': '70282931',
 174             'ext': 'mp4',
 175             'title': '[Premium Access + Uncut] The Office - 2x6 The Fight - Group Reaction',
 176             'channel_url': 'https://www.patreon.com/thenormies',
 177             'channel_id': '573397',
 178             'uploader_id': '2929435',
 179             'uploader': 'The Normies',
 180             'description': 'md5:79c9fd8778e2cef84049a94c058a5e23',
 181             'comment_count': int,
 182             'upload_date': '20220809',
 183             'thumbnail': r're:^https?://.*$',
 184             'channel_follower_count': int,
 185             'like_count': int,
 186             'timestamp': 1660052820,
 187             'tags': ['The Office', 'early access', 'uncut'],
 188             'uploader_url': 'https://www.patreon.com/thenormies',
 189         },
 190         'skip': 'Patron-only content',
 191     }, {
 192         # dead vimeo and embed URLs, need to extract post_file
 193         'url': 'https://www.patreon.com/posts/hunter-x-hunter-34007913',
 194         'info_dict': {
 195             'id': '34007913',
 196             'ext': 'mp4',
 197             'title': 'Hunter x Hunter | Kurapika DESTROYS Uvogin!!!',
 198             'like_count': int,
 199             'uploader': 'YaBoyRoshi',
 200             'timestamp': 1581636833,
 201             'channel_url': 'https://www.patreon.com/yaboyroshi',
 202             'thumbnail': r're:^https?://.*$',
 203             'tags': ['Hunter x Hunter'],
 204             'uploader_id': '14264111',
 205             'comment_count': int,
 206             'channel_follower_count': int,
 207             'description': 'Kurapika is a walking cheat code!',
 208             'upload_date': '20200213',
 209             'channel_id': '2147162',
 210             'uploader_url': 'https://www.patreon.com/yaboyroshi',
 211         },
 212     }, {
 213         # NSFW vimeo embed URL
 214         'url': 'https://www.patreon.com/posts/4k-spiderman-4k-96414599',
 215         'info_dict': {
 216             'id': '902250943',
 217             'ext': 'mp4',
 218             'title': '❤️(4K) Spiderman Girl Yeonhwa’s Gift ❤️(4K) 스파이더맨걸 연화의 선물',
 219             'description': '❤️(4K) Spiderman Girl Yeonhwa’s Gift \n❤️(4K) 스파이더맨걸 연화의 선물',
 220             'uploader': 'Npickyeonhwa',
 221             'uploader_id': '90574422',
 222             'uploader_url': 'https://www.patreon.com/Yeonhwa726',
 223             'channel_id': '10237902',
 224             'channel_url': 'https://www.patreon.com/Yeonhwa726',
 225             'duration': 70,
 226             'timestamp': 1705150153,
 227             'upload_date': '20240113',
 228             'comment_count': int,
 229             'like_count': int,
 230             'thumbnail': r're:^https?://.+',
 231         },
 232         'params': {'skip_download': 'm3u8'},
 233         'expected_warnings': ['Failed to parse XML: not well-formed'],
 234     }, {
 235         # multiple attachments/embeds
 236         'url': 'https://www.patreon.com/posts/holy-wars-solos-100601977',
 237         'playlist_count': 3,
 238         'info_dict': {
 239             'id': '100601977',
 240             'title': '"Holy Wars" (Megadeth) Solos Transcription & Lesson/Analysis',
 241             'description': 'md5:d099ab976edfce6de2a65c2b169a88d3',
 242             'uploader': 'Bradley Hall',
 243             'uploader_id': '24401883',
 244             'uploader_url': 'https://www.patreon.com/bradleyhallguitar',
 245             'channel_id': '3193932',
 246             'channel_url': 'https://www.patreon.com/bradleyhallguitar',
 247             'channel_follower_count': int,
 248             'timestamp': 1710777855,
 249             'upload_date': '20240318',
 250             'like_count': int,
 251             'comment_count': int,
 252             'thumbnail': r're:^https?://.+',
 253         },
 254         'skip': 'Patron-only content',
 255     }, {
 256         # Contains a comment reply in the 'included' section
 257         'url': 'https://www.patreon.com/posts/114721679',
 258         'info_dict': {
 259             'id': '114721679',
 260             'ext': 'mp4',
 261             'upload_date': '20241025',
 262             'uploader': 'Japanalysis',
 263             'like_count': int,
 264             'thumbnail': r're:^https?://.+',
 265             'comment_count': int,
 266             'title': 'Karasawa Part 2',
 267             'description': 'Part 2 of this video https://www.youtube.com/watch?v=Azms2-VTASk',
 268             'uploader_url': 'https://www.patreon.com/japanalysis',
 269             'uploader_id': '80504268',
 270             'channel_url': 'https://www.patreon.com/japanalysis',
 271             'channel_follower_count': int,
 272             'timestamp': 1729897015,
 273             'channel_id': '9346307',
 274         },
 275         'params': {'getcomments': True},
 276     }]
 277     _RETURN_TYPE = 'video'
 278
 279     def _real_extract(self, url):
 280         video_id = self._match_id(url)
 281         post = self._call_api(
 282             f'posts/{video_id}', video_id, query={
 283                 'fields[media]': 'download_url,mimetype,size_bytes',
 284                 'fields[post]': 'comment_count,content,embed,image,like_count,post_file,published_at,title,current_user_can_view',
 285                 'fields[user]': 'full_name,url',
 286                 'fields[post_tag]': 'value',
 287                 'fields[campaign]': 'url,name,patron_count',
 288                 'json-api-use-default-includes': 'false',
 289                 'include': 'audio,user,user_defined_tags,campaign,attachments_media',
 290             })
 291         attributes = post['data']['attributes']
 292         info = traverse_obj(attributes, {
 293             'title': ('title', {str.strip}),
 294             'description': ('content', {clean_html}),
 295             'thumbnail': ('image', ('large_url', 'url'), {url_or_none}, any),
 296             'timestamp': ('published_at', {parse_iso8601}),
 297             'like_count': ('like_count', {int_or_none}),
 298             'comment_count': ('comment_count', {int_or_none}),
 299         })
 300
 301         entries = []
 302         idx = 0
 303         for include in traverse_obj(post, ('included', lambda _, v: v['type'])):
 304             include_type = include['type']
 305             if include_type == 'media':
 306                 media_attributes = traverse_obj(include, ('attributes', {dict})) or {}
 307                 download_url = url_or_none(media_attributes.get('download_url'))
 308                 ext = mimetype2ext(media_attributes.get('mimetype'))
 309
 310                 # if size_bytes is None, this media file is likely unavailable
 311                 # See: https://github.com/yt-dlp/yt-dlp/issues/4608
 312                 size_bytes = int_or_none(media_attributes.get('size_bytes'))
 313                 if download_url and ext in KNOWN_EXTENSIONS and size_bytes is not None:
 314                     idx += 1
 315                     entries.append({
 316                         'id': f'{video_id}-{idx}',
 317                         'ext': ext,
 318                         'filesize': size_bytes,
 319                         'url': download_url,
 320                     })
 321
 322             elif include_type == 'user':
 323                 info.update(traverse_obj(include, {
 324                     'uploader': ('attributes', 'full_name', {str}),
 325                     'uploader_id': ('id', {str_or_none}),
 326                     'uploader_url': ('attributes', 'url', {url_or_none}),
 327                 }))
 328
 329             elif include_type == 'post_tag':
 330                 if post_tag := traverse_obj(include, ('attributes', 'value', {str})):
 331                     info.setdefault('tags', []).append(post_tag)
 332
 333             elif include_type == 'campaign':
 334                 info.update(traverse_obj(include, {
 335                     'channel': ('attributes', 'title', {str}),
 336                     'channel_id': ('id', {str_or_none}),
 337                     'channel_url': ('attributes', 'url', {url_or_none}),
 338                     'channel_follower_count': ('attributes', 'patron_count', {int_or_none}),
 339                 }))
 340
 341         # all-lowercase 'referer' so we can smuggle it to Generic, SproutVideo, Vimeo
 342         headers = {'referer': 'https://patreon.com/'}
 343
 344         # handle Vimeo embeds
 345         if traverse_obj(attributes, ('embed', 'provider')) == 'Vimeo':
 346             v_url = urllib.parse.unquote(self._html_search_regex(
 347                 r'(https(?:%3A%2F%2F|://)player\.vimeo\.com.+app_id(?:=|%3D)+\d+)',
 348                 traverse_obj(attributes, ('embed', 'html', {str})), 'vimeo url', fatal=False) or '')
 349             if url_or_none(v_url) and self._request_webpage(
 350                     v_url, video_id, 'Checking Vimeo embed URL', headers=headers,
 351                     fatal=False, errnote=False, expected_status=429):  # 429 is TLS fingerprint rejection
 352                 entries.append(self.url_result(
 353                     VimeoIE._smuggle_referrer(v_url, 'https://patreon.com/'),
 354                     VimeoIE, url_transparent=True))
 355
 356         embed_url = traverse_obj(attributes, ('embed', 'url', {url_or_none}))
 357         if embed_url and (urlh := self._request_webpage(
 358                 embed_url, video_id, 'Checking embed URL', headers=headers,
 359                 fatal=False, errnote=False, expected_status=403)):
 360             # Vimeo's Cloudflare anti-bot protection will return HTTP status 200 for 404, so we need
 361             # to check for "Sorry, we couldn&amp;rsquo;t find that page" in the meta description tag
 362             meta_description = clean_html(self._html_search_meta(
 363                 'description', self._webpage_read_content(urlh, embed_url, video_id, fatal=False), default=None))
 364             # Password-protected vids.io embeds return 403 errors w/o --video-password or session cookie
 365             if ((urlh.status != 403 and meta_description != 'Sorry, we couldn’t find that page')
 366                     or VidsIoIE.suitable(embed_url)):
 367                 entries.append(self.url_result(smuggle_url(embed_url, headers)))
 368
 369         post_file = traverse_obj(attributes, ('post_file', {dict}))
 370         if post_file:
 371             name = post_file.get('name')
 372             ext = determine_ext(name)
 373             if ext in KNOWN_EXTENSIONS:
 374                 entries.append({
 375                     'id': video_id,
 376                     'ext': ext,
 377                     'url': post_file['url'],
 378                 })
 379             elif name == 'video' or determine_ext(post_file.get('url')) == 'm3u8':
 380                 formats, subtitles = self._extract_m3u8_formats_and_subtitles(post_file['url'], video_id)
 381                 entries.append({
 382                     'id': video_id,
 383                     'formats': formats,
 384                     'subtitles': subtitles,
 385                 })
 386
 387         can_view_post = traverse_obj(attributes, 'current_user_can_view')
 388         comments = None
 389         if can_view_post and info.get('comment_count'):
 390             comments = self.extract_comments(video_id)
 391
 392         if not entries and can_view_post is False:
 393             self.raise_no_formats('You do not have access to this post', video_id=video_id, expected=True)
 394         elif not entries:
 395             self.raise_no_formats('No supported media found in this post', video_id=video_id, expected=True)
 396         elif len(entries) == 1:
 397             info.update(entries[0])
 398         else:
 399             for entry in entries:
 400                 entry.update(info)
 401             return self.playlist_result(entries, video_id, **info, __post_extractor=comments)
 402
 403         info['id'] = video_id
 404         info['__post_extractor'] = comments
 405         return info
 406
 407     def _get_comments(self, post_id):
 408         cursor = None
 409         count = 0
 410         params = {
 411             'page[count]': 50,
 412             'include': 'parent.commenter.campaign,parent.post.user,parent.post.campaign.creator,parent.replies.parent,parent.replies.commenter.campaign,parent.replies.post.user,parent.replies.post.campaign.creator,commenter.campaign,post.user,post.campaign.creator,replies.parent,replies.commenter.campaign,replies.post.user,replies.post.campaign.creator,on_behalf_of_campaign',
 413             'fields[comment]': 'body,created,is_by_creator',
 414             'fields[user]': 'image_url,full_name,url',
 415             'filter[flair]': 'image_tiny_url,name',
 416             'sort': '-created',
 417             'json-api-version': 1.0,
 418             'json-api-use-default-includes': 'false',
 419         }
 420
 421         for page in itertools.count(1):
 422
 423             params.update({'page[cursor]': cursor} if cursor else {})
 424             response = self._call_api(
 425                 f'posts/{post_id}/comments', post_id, query=params, note=f'Downloading comments page {page}')
 426
 427             cursor = None
 428             for comment in traverse_obj(response, (('data', 'included'), lambda _, v: v['type'] == 'comment' and v['id'])):
 429                 count += 1
 430                 author_id = traverse_obj(comment, ('relationships', 'commenter', 'data', 'id'))
 431
 432                 yield {
 433                     **traverse_obj(comment, {
 434                         'id': ('id', {str_or_none}),
 435                         'text': ('attributes', 'body', {str}),
 436                         'timestamp': ('attributes', 'created', {parse_iso8601}),
 437                         'parent': ('relationships', 'parent', 'data', ('id', {value('root')}), {str}, any),
 438                         'author_is_uploader': ('attributes', 'is_by_creator', {bool}),
 439                     }),
 440                     **traverse_obj(response, (
 441                         'included', lambda _, v: v['id'] == author_id and v['type'] == 'user', 'attributes', {
 442                             'author': ('full_name', {str}),
 443                             'author_thumbnail': ('image_url', {url_or_none}),
 444                         }), get_all=False),
 445                     'author_id': author_id,
 446                 }
 447
 448             if count < traverse_obj(response, ('meta', 'count')):
 449                 cursor = traverse_obj(response, ('data', -1, 'id'))
 450
 451             if cursor is None:
 452                 break
 453
 454
 455 class PatreonCampaignIE(PatreonBaseIE):
 456     IE_NAME = 'patreon:campaign'
 457     _VALID_URL = r'''(?x)
 458         https?://(?:www\.)?patreon\.com/(?:
 459             (?:m|api/campaigns)/(?P<campaign_id>\d+)|
 460             (?P<vanity>(?!creation[?/]|posts/|rss[?/])[\w-]+)
 461         )(?:/posts)?/?(?:$|[?#])'''
 462     _TESTS = [{
 463         'url': 'https://www.patreon.com/dissonancepod/',
 464         'info_dict': {
 465             'title': 'Cognitive Dissonance Podcast',
 466             'channel_url': 'https://www.patreon.com/dissonancepod',
 467             'id': '80642',
 468             'description': r're:(?s).*We produce a weekly news podcast focusing on stories that deal with skepticism and religion.*',
 469             'channel_id': '80642',
 470             'channel': 'Cognitive Dissonance Podcast',
 471             'age_limit': 0,
 472             'channel_follower_count': int,
 473             'uploader_id': '87145',
 474             'uploader_url': 'https://www.patreon.com/dissonancepod',
 475             'uploader': 'Cognitive Dissonance Podcast',
 476             'thumbnail': r're:^https?://.*$',
 477         },
 478         'playlist_mincount': 68,
 479     }, {
 480         'url': 'https://www.patreon.com/m/4767637/posts',
 481         'info_dict': {
 482             'title': 'Not Just Bikes',
 483             'id': '4767637',
 484             'channel_id': '4767637',
 485             'channel_url': 'https://www.patreon.com/notjustbikes',
 486             'description': r're:(?s).*Not Just Bikes started as a way to explain why we chose to live in the Netherlands.*',
 487             'age_limit': 0,
 488             'channel': 'Not Just Bikes',
 489             'uploader_url': 'https://www.patreon.com/notjustbikes',
 490             'uploader': 'Jason',
 491             'uploader_id': '37306634',
 492             'thumbnail': r're:^https?://.*$',
 493         },
 494         'playlist_mincount': 71,
 495     }, {
 496         'url': 'https://www.patreon.com/api/campaigns/4243769/posts',
 497         'info_dict': {
 498             'title': 'Second Thought',
 499             'channel_follower_count': int,
 500             'id': '4243769',
 501             'channel_id': '4243769',
 502             'channel_url': 'https://www.patreon.com/secondthought',
 503             'description': r're:(?s).*Second Thought is an educational YouTube channel.*',
 504             'age_limit': 0,
 505             'channel': 'Second Thought',
 506             'uploader_url': 'https://www.patreon.com/secondthought',
 507             'uploader': 'JT Chapman',
 508             'uploader_id': '32718287',
 509             'thumbnail': r're:^https?://.*$',
 510         },
 511         'playlist_mincount': 201,
 512     }, {
 513         'url': 'https://www.patreon.com/dissonancepod/posts',
 514         'only_matching': True,
 515     }, {
 516         'url': 'https://www.patreon.com/m/5932659',
 517         'only_matching': True,
 518     }, {
 519         'url': 'https://www.patreon.com/api/campaigns/4243769',
 520         'only_matching': True,
 521     }]
 522
 523     def _entries(self, campaign_id):
 524         cursor = None
 525         params = {
 526             'fields[post]': 'patreon_url,url',
 527             'filter[campaign_id]': campaign_id,
 528             'filter[is_draft]': 'false',
 529             'sort': '-published_at',
 530             'json-api-use-default-includes': 'false',
 531         }
 532
 533         for page in itertools.count(1):
 534
 535             params.update({'page[cursor]': cursor} if cursor else {})
 536             posts_json = self._call_api('posts', campaign_id, query=params, note=f'Downloading posts page {page}')
 537
 538             cursor = traverse_obj(posts_json, ('meta', 'pagination', 'cursors', 'next'))
 539             for post_url in traverse_obj(posts_json, ('data', ..., 'attributes', 'patreon_url')):
 540                 yield self.url_result(urljoin('https://www.patreon.com/', post_url), PatreonIE)
 541
 542             if cursor is None:
 543                 break
 544
 545     def _real_extract(self, url):
 546
 547         campaign_id, vanity = self._match_valid_url(url).group('campaign_id', 'vanity')
 548         if campaign_id is None:
 549             webpage = self._download_webpage(url, vanity, headers={'User-Agent': self.patreon_user_agent})
 550             campaign_id = self._search_nextjs_data(
 551                 webpage, vanity)['props']['pageProps']['bootstrapEnvelope']['pageBootstrap']['campaign']['data']['id']
 552
 553         params = {
 554             'json-api-use-default-includes': 'false',
 555             'fields[user]': 'full_name,url',
 556             'fields[campaign]': 'name,summary,url,patron_count,creation_count,is_nsfw,avatar_photo_url',
 557             'include': 'creator',
 558         }
 559
 560         campaign_response = self._call_api(
 561             f'campaigns/{campaign_id}', campaign_id,
 562             note='Downloading campaign info', fatal=False,
 563             query=params) or {}
 564
 565         campaign_info = campaign_response.get('data') or {}
 566         channel_name = traverse_obj(campaign_info, ('attributes', 'name'))
 567         user_info = traverse_obj(
 568             campaign_response, ('included', lambda _, v: v['type'] == 'user'),
 569             default={}, expected_type=dict, get_all=False)
 570
 571         return {
 572             '_type': 'playlist',
 573             'id': campaign_id,
 574             'title': channel_name,
 575             'entries': self._entries(campaign_id),
 576             'description': clean_html(traverse_obj(campaign_info, ('attributes', 'summary'))),
 577             'channel_url': traverse_obj(campaign_info, ('attributes', 'url')),
 578             'channel_follower_count': int_or_none(traverse_obj(campaign_info, ('attributes', 'patron_count'))),
 579             'channel_id': campaign_id,
 580             'channel': channel_name,
 581             'uploader_url': traverse_obj(user_info, ('attributes', 'url')),
 582             'uploader_id': str_or_none(user_info.get('id')),
 583             'uploader': traverse_obj(user_info, ('attributes', 'full_name')),
 584             'playlist_count': traverse_obj(campaign_info, ('attributes', 'creation_count')),
 585             'age_limit': 18 if traverse_obj(campaign_info, ('attributes', 'is_nsfw')) else 0,
 586             'thumbnail': url_or_none(traverse_obj(campaign_info, ('attributes', 'avatar_photo_url'))),
 587         }