youtube_dl/extractor/patreon.py

   1 # encoding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import json
   5 import re
   6
   7 from .common import InfoExtractor
   8 from ..utils import (
   9     compat_urlparse,
  10     js_to_json,
  11 )
  12
  13
  14 class PatreonIE(InfoExtractor):
  15     _VALID_URL = r'https?://(?:www\.)?patreon\.com/creation\?hid=(.+)'
  16     _TESTS = [
  17         {
  18             'url': 'http://www.patreon.com/creation?hid=743933',
  19             'md5': 'e25505eec1053a6e6813b8ed369875cc',
  20             'info_dict': {
  21                 'id': '743933',
  22                 'ext': 'mp3',
  23                 'title': 'Episode 166: David Smalley of Dogma Debate',
  24                 'uploader': 'Cognitive Dissonance Podcast',
  25                 'thumbnail': 're:^https?://.*$',
  26             },
  27         },
  28         {
  29             'url': 'http://www.patreon.com/creation?hid=754133',
  30             'md5': '3eb09345bf44bf60451b8b0b81759d0a',
  31             'info_dict': {
  32                 'id': '754133',
  33                 'ext': 'mp3',
  34                 'title': 'CD 167 Extra',
  35                 'uploader': 'Cognitive Dissonance Podcast',
  36                 'thumbnail': 're:^https?://.*$',
  37             },
  38         },
  39     ]
  40
  41     # Currently Patreon exposes download URL via hidden CSS, so login is not
  42     # needed. Keeping this commented for when this inevitably changes.
  43     '''
  44     def _login(self):
  45         (username, password) = self._get_login_info()
  46         if username is None:
  47             return
  48
  49         login_form = {
  50             'redirectUrl': 'http://www.patreon.com/',
  51             'email': username,
  52             'password': password,
  53         }
  54
  55         request = compat_urllib_request.Request(
  56             'https://www.patreon.com/processLogin',
  57             compat_urllib_parse.urlencode(login_form).encode('utf-8')
  58         )
  59         login_page = self._download_webpage(request, None, note='Logging in as %s' % username)
  60
  61         if re.search(r'onLoginFailed', login_page):
  62             raise ExtractorError('Unable to login, incorrect username and/or password', expected=True)
  63
  64     def _real_initialize(self):
  65         self._login()
  66     '''
  67
  68     def _real_extract(self, url):
  69         mobj = re.match(self._VALID_URL, url)
  70         video_id = mobj.group(1)
  71
  72         webpage = self._download_webpage(url, video_id)
  73         title = self._og_search_title(webpage).strip()
  74
  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')
  83         else:
  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)
  89             data = playlist[0]
  90             video_url = self._proto_relative_url(data['mp3'])
  91             thumbnail = self._proto_relative_url(data.get('cover'))
  92             uploader = data.get('artist')
  93
  94         return {
  95             'id': video_id,
  96             'url': video_url,
  97             'ext': 'mp3',
  98             'title': title,
  99             'uploader': uploader,
 100             'thumbnail': thumbnail,
 101         }