yt_dlp/extractor/patreon.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import itertools
   5
   6 from .common import InfoExtractor
   7 from .vimeo import VimeoIE
   8
   9 from ..compat import compat_urllib_parse_unquote
  10 from ..utils import (
  11     clean_html,
  12     determine_ext,
  13     int_or_none,
  14     KNOWN_EXTENSIONS,
  15     mimetype2ext,
  16     parse_iso8601,
  17     str_or_none,
  18     try_get,
  19     url_or_none,
  20 )
  21
  22
  23 class PatreonIE(InfoExtractor):
  24     _VALID_URL = r'https?://(?:www\.)?patreon\.com/(?:creation\?hid=|posts/(?:[\w-]+-)?)(?P<id>\d+)'
  25     _TESTS = [{
  26         'url': 'http://www.patreon.com/creation?hid=743933',
  27         'md5': 'e25505eec1053a6e6813b8ed369875cc',
  28         'info_dict': {
  29             'id': '743933',
  30             'ext': 'mp3',
  31             'title': 'Episode 166: David Smalley of Dogma Debate',
  32             'description': 'md5:713b08b772cd6271b9f3906683cfacdf',
  33             'uploader': 'Cognitive Dissonance Podcast',
  34             'thumbnail': 're:^https?://.*$',
  35             'timestamp': 1406473987,
  36             'upload_date': '20140727',
  37             'uploader_id': '87145',
  38         },
  39     }, {
  40         'url': 'http://www.patreon.com/creation?hid=754133',
  41         'md5': '3eb09345bf44bf60451b8b0b81759d0a',
  42         'info_dict': {
  43             'id': '754133',
  44             'ext': 'mp3',
  45             'title': 'CD 167 Extra',
  46             'uploader': 'Cognitive Dissonance Podcast',
  47             'thumbnail': 're:^https?://.*$',
  48         },
  49         'skip': 'Patron-only content',
  50     }, {
  51         'url': 'https://www.patreon.com/creation?hid=1682498',
  52         'info_dict': {
  53             'id': 'SU4fj_aEMVw',
  54             'ext': 'mp4',
  55             'title': 'I\'m on Patreon!',
  56             'uploader': 'TraciJHines',
  57             'thumbnail': 're:^https?://.*$',
  58             'upload_date': '20150211',
  59             'description': 'md5:c5a706b1f687817a3de09db1eb93acd4',
  60             'uploader_id': 'TraciJHines',
  61         },
  62         'params': {
  63             'noplaylist': True,
  64             'skip_download': True,
  65         }
  66     }, {
  67         'url': 'https://www.patreon.com/posts/episode-166-of-743933',
  68         'only_matching': True,
  69     }, {
  70         'url': 'https://www.patreon.com/posts/743933',
  71         'only_matching': True,
  72     }, {
  73         'url': 'https://www.patreon.com/posts/kitchen-as-seen-51706779',
  74         'md5': '96656690071f6d64895866008484251b',
  75         'info_dict': {
  76             'id': '555089736',
  77             'ext': 'mp4',
  78             'title': 'KITCHEN AS SEEN ON DEEZ NUTS EXTENDED!',
  79             'uploader': 'Cold Ones',
  80             'thumbnail': 're:^https?://.*$',
  81             'upload_date': '20210526',
  82             'description': 'md5:557a409bd79d3898689419094934ba79',
  83             'uploader_id': '14936315',
  84         },
  85         'skip': 'Patron-only content'
  86     }]
  87
  88     # Currently Patreon exposes download URL via hidden CSS, so login is not
  89     # needed. Keeping this commented for when this inevitably changes.
  90     '''
  91     def _login(self):
  92         username, password = self._get_login_info()
  93         if username is None:
  94             return
  95
  96         login_form = {
  97             'redirectUrl': 'http://www.patreon.com/',
  98             'email': username,
  99             'password': password,
 100         }
 101
 102         request = sanitized_Request(
 103             'https://www.patreon.com/processLogin',
 104             compat_urllib_parse_urlencode(login_form).encode('utf-8')
 105         )
 106         login_page = self._download_webpage(request, None, note='Logging in')
 107
 108         if re.search(r'onLoginFailed', login_page):
 109             raise ExtractorError('Unable to login, incorrect username and/or password', expected=True)
 110
 111     def _real_initialize(self):
 112         self._login()
 113     '''
 114
 115     def _real_extract(self, url):
 116         video_id = self._match_id(url)
 117         post = self._download_json(
 118             'https://www.patreon.com/api/posts/' + video_id, video_id, query={
 119                 'fields[media]': 'download_url,mimetype,size_bytes',
 120                 'fields[post]': 'comment_count,content,embed,image,like_count,post_file,published_at,title',
 121                 'fields[user]': 'full_name,url',
 122                 'json-api-use-default-includes': 'false',
 123                 'include': 'media,user',
 124             })
 125         attributes = post['data']['attributes']
 126         title = attributes['title'].strip()
 127         image = attributes.get('image') or {}
 128         info = {
 129             'id': video_id,
 130             'title': title,
 131             'description': clean_html(attributes.get('content')),
 132             'thumbnail': image.get('large_url') or image.get('url'),
 133             'timestamp': parse_iso8601(attributes.get('published_at')),
 134             'like_count': int_or_none(attributes.get('like_count')),
 135             'comment_count': int_or_none(attributes.get('comment_count')),
 136         }
 137
 138         for i in post.get('included', []):
 139             i_type = i.get('type')
 140             if i_type == 'media':
 141                 media_attributes = i.get('attributes') or {}
 142                 download_url = media_attributes.get('download_url')
 143                 ext = mimetype2ext(media_attributes.get('mimetype'))
 144                 if download_url and ext in KNOWN_EXTENSIONS:
 145                     info.update({
 146                         'ext': ext,
 147                         'filesize': int_or_none(media_attributes.get('size_bytes')),
 148                         'url': download_url,
 149                     })
 150             elif i_type == 'user':
 151                 user_attributes = i.get('attributes')
 152                 if user_attributes:
 153                     info.update({
 154                         'uploader': user_attributes.get('full_name'),
 155                         'uploader_id': str_or_none(i.get('id')),
 156                         'uploader_url': user_attributes.get('url'),
 157                     })
 158
 159         if not info.get('url'):
 160             # handle Vimeo embeds
 161             if try_get(attributes, lambda x: x['embed']['provider']) == 'Vimeo':
 162                 embed_html = try_get(attributes, lambda x: x['embed']['html'])
 163                 v_url = url_or_none(compat_urllib_parse_unquote(
 164                     self._search_regex(r'(https(?:%3A%2F%2F|://)player\.vimeo\.com.+app_id(?:=|%3D)+\d+)', embed_html, 'vimeo url', fatal=False)))
 165                 if v_url:
 166                     info.update({
 167                         '_type': 'url_transparent',
 168                         'url': VimeoIE._smuggle_referrer(v_url, 'https://patreon.com'),
 169                         'ie_key': 'Vimeo',
 170                     })
 171
 172         if not info.get('url'):
 173             embed_url = try_get(attributes, lambda x: x['embed']['url'])
 174             if embed_url:
 175                 info.update({
 176                     '_type': 'url',
 177                     'url': embed_url,
 178                 })
 179
 180         if not info.get('url'):
 181             post_file = attributes['post_file']
 182             ext = determine_ext(post_file.get('name'))
 183             if ext in KNOWN_EXTENSIONS:
 184                 info.update({
 185                     'ext': ext,
 186                     'url': post_file['url'],
 187                 })
 188
 189         return info
 190
 191
 192 class PatreonUserIE(InfoExtractor):
 193
 194     _VALID_URL = r'https?://(?:www\.)?patreon\.com/(?P<id>[-_\w\d]+)/?(?:posts/?)?'
 195
 196     _TESTS = [{
 197         'url': 'https://www.patreon.com/dissonancepod/',
 198         'info_dict': {
 199             'title': 'dissonancepod',
 200         },
 201         'playlist_mincount': 68,
 202         'expected_warnings': 'Post not viewable by current user! Skipping!',
 203     }, {
 204         'url': 'https://www.patreon.com/dissonancepod/posts',
 205         'only_matching': True
 206     }, ]
 207
 208     @classmethod
 209     def suitable(cls, url):
 210         return False if PatreonIE.suitable(url) else super(PatreonUserIE, cls).suitable(url)
 211
 212     def _entries(self, campaign_id, user_id):
 213         cursor = None
 214         params = {
 215             'fields[campaign]': 'show_audio_post_download_links,name,url',
 216             'fields[post]': 'current_user_can_view,embed,image,is_paid,post_file,published_at,patreon_url,url,post_type,thumbnail_url,title',
 217             'filter[campaign_id]': campaign_id,
 218             'filter[is_draft]': 'false',
 219             'sort': '-published_at',
 220             'json-api-version': 1.0,
 221             'json-api-use-default-includes': 'false',
 222         }
 223
 224         for page in itertools.count(1):
 225
 226             params.update({'page[cursor]': cursor} if cursor else {})
 227             posts_json = self._download_json('https://www.patreon.com/api/posts', user_id, note='Downloading posts page %d' % page, query=params, headers={'Cookie': '.'})
 228
 229             cursor = try_get(posts_json, lambda x: x['meta']['pagination']['cursors']['next'])
 230
 231             for post in posts_json.get('data') or []:
 232                 yield self.url_result(url_or_none(try_get(post, lambda x: x['attributes']['patreon_url'])), 'Patreon')
 233
 234             if cursor is None:
 235                 break
 236
 237     def _real_extract(self, url):
 238
 239         user_id = self._match_id(url)
 240         webpage = self._download_webpage(url, user_id, headers={'Cookie': '.'})
 241         campaign_id = self._search_regex(r'https://www.patreon.com/api/campaigns/(\d+)/?', webpage, 'Campaign ID')
 242         return self.playlist_result(self._entries(campaign_id, user_id), playlist_title=user_id)