Raphaël G. Git Repositories - youtubedl/blob - 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     js_to_json,
  10 )
  11
  12
  13 class PatreonIE(InfoExtractor):
  14     _VALID_URL = r'https?://(?:www\.)?patreon\.com/creation\?hid=(.+)'
  15     _TESTS = [
  16         {
  17             'url': 'http://www.patreon.com/creation?hid=743933',
  18             'md5': 'e25505eec1053a6e6813b8ed369875cc',
  19             'info_dict': {
  20                 'id': '743933',
  21                 'ext': 'mp3',
  22                 'title': 'Episode 166: David Smalley of Dogma Debate',
  23                 'uploader': 'Cognitive Dissonance Podcast',
  24                 'thumbnail': 're:^https?://.*$',
  25             },
  26         },
  27         {
  28             'url': 'http://www.patreon.com/creation?hid=754133',
  29             'md5': '3eb09345bf44bf60451b8b0b81759d0a',
  30             'info_dict': {
  31                 'id': '754133',
  32                 'ext': 'mp3',
  33                 'title': 'CD 167 Extra',
  34                 'uploader': 'Cognitive Dissonance Podcast',
  35                 'thumbnail': 're:^https?://.*$',
  36             },
  37         },
  38     ]
  39
  40     # Currently Patreon exposes download URL via hidden CSS, so login is not
  41     # needed. Keeping this commented for when this inevitably changes.
  42     '''
  43     def _login(self):
  44         (username, password) = self._get_login_info()
  45         if username is None:
  46             return
  47
  48         login_form = {
  49             'redirectUrl': 'http://www.patreon.com/',
  50             'email': username,
  51             'password': password,
  52         }
  53
  54         request = compat_urllib_request.Request(
  55             'https://www.patreon.com/processLogin',
  56             compat_urllib_parse.urlencode(login_form).encode('utf-8')
  57         )
  58         login_page = self._download_webpage(request, None, note='Logging in as %s' % username)
  59
  60         if re.search(r'onLoginFailed', login_page):
  61             raise ExtractorError('Unable to login, incorrect username and/or password', expected=True)
  62
  63     def _real_initialize(self):
  64         self._login()
  65     '''
  66
  67     def _real_extract(self, url):
  68         mobj = re.match(self._VALID_URL, url)
  69         video_id = mobj.group(1)
  70
  71         webpage = self._download_webpage(url, video_id)
  72         title = self._og_search_title(webpage).strip()
  73
  74         attach_fn = self._html_search_regex(
  75             r'<div class="attach"><a target="_blank" href="([^"]+)">',
  76             webpage, 'attachment URL', default=None)
  77         if attach_fn is not None:
  78             video_url = 'http://www.patreon.com' + attach_fn
  79             thumbnail = self._og_search_thumbnail(webpage)
  80             uploader = self._html_search_regex(
  81                 r'<strong>(.*?)</strong> is creating', webpage, 'uploader')
  82         else:
  83             playlist_js = self._search_regex(
  84                 r'(?s)new\s+jPlayerPlaylist\(\s*\{\s*[^}]*},\s*(\[.*?,?\s*\])',
  85                 webpage, 'playlist JSON')
  86             playlist_json = js_to_json(playlist_js)
  87             playlist = json.loads(playlist_json)
  88             data = playlist[0]
  89             video_url = self._proto_relative_url(data['mp3'])
  90             thumbnail = self._proto_relative_url(data.get('cover'))
  91             uploader = data.get('artist')
  92
  93         return {
  94             'id': video_id,
  95             'url': video_url,
  96             'ext': 'mp3',
  97             'title': title,
  98             'uploader': uploader,
  99             'thumbnail': thumbnail,
 100         }