2 from __future__ import unicode_literals
7 from .common import InfoExtractor
14 class PatreonIE(InfoExtractor):
15 _VALID_URL = r'https?://(?:www\.)?patreon\.com/creation\?hid=(.+)'
18 'url': 'http://www.patreon.com/creation?hid=743933',
19 'md5': 'e25505eec1053a6e6813b8ed369875cc',
23 'title': 'Episode 166: David Smalley of Dogma Debate',
24 'uploader': 'Cognitive Dissonance Podcast',
25 'thumbnail': 're:^https?://.*$',
29 'url': 'http://www.patreon.com/creation?hid=754133',
30 'md5': '3eb09345bf44bf60451b8b0b81759d0a',
34 'title': 'CD 167 Extra',
35 'uploader': 'Cognitive Dissonance Podcast',
36 'thumbnail': 're:^https?://.*$',
41 # Currently Patreon exposes download URL via hidden CSS, so login is not
42 # needed. Keeping this commented for when this inevitably changes.
45 (username, password) = self._get_login_info()
50 'redirectUrl': 'http://www.patreon.com/',
55 request = compat_urllib_request.Request(
56 'https://www.patreon.com/processLogin',
57 compat_urllib_parse.urlencode(login_form).encode('utf-8')
59 login_page = self._download_webpage(request, None, note='Logging in as %s' % username)
61 if re.search(r'onLoginFailed', login_page):
62 raise ExtractorError('Unable to login, incorrect username and/or password', expected=True)
64 def _real_initialize(self):
68 def _real_extract(self, url):
69 mobj = re.match(self._VALID_URL, url)
70 video_id = mobj.group(1)
72 webpage = self._download_webpage(url, video_id)
73 title = self._og_search_title(webpage).strip()
75 attach_fn = self._html_search_regex(
76 r'<div class="attach"><a target="_blank" href="([^"]+)">',
77 webpage, 'attachment URL', default=None)
78 if attach_fn is not None:
79 video_url = 'http://www.patreon.com' + attach_fn
80 thumbnail = self._og_search_thumbnail(webpage)
81 uploader = self._html_search_regex(
82 r'<strong>(.*?)</strong> is creating', webpage, 'uploader')
84 playlist_js = self._search_regex(
85 r'(?s)new\s+jPlayerPlaylist\(\s*\{\s*[^}]*},\s*(\[.*?,?\s*\])',
86 webpage, 'playlist JSON')
87 playlist_json = js_to_json(playlist_js)
88 playlist = json.loads(playlist_json)
90 video_url = self._proto_relative_url(data['mp3'])
91 thumbnail = self._proto_relative_url(data.get('cover'))
92 uploader = data.get('artist')
100 'thumbnail': thumbnail,