_ Git - youtube-dl/blob - youtube_dl/extractor/patreon.py

   1 # encoding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..utils import (
   8     ExtractorError,
   9     compat_html_parser,
  10     #compat_urllib_request,
  11     #compat_urllib_parse,
  12 )
  13
  14
  15 class PatreonHTMLParser(compat_html_parser.HTMLParser):
  16     _PREFIX = 'http://www.patreon.com'
  17     _ATTACH_TAGS = 5 * ['div']
  18     _ATTACH_CLASSES = [
  19         'fancyboxhidden', 'box photo double', 'boxwrapper double',
  20         'hiddendisplay shareinfo', 'attach'
  21     ]
  22     _INFO_TAGS = 4 * ['div']
  23     _INFO_CLASSES = [
  24         'fancyboxhidden', 'box photo double', 'boxwrapper double',
  25         'hiddendisplay shareinfo'
  26     ]
  27
  28     def get_creation_info(self, html_data):
  29         self.tag_stack = []
  30         self.attrs_stack = []
  31         self.creation_info = {}
  32         self.feed(html_data)
  33
  34     def handle_starttag(self, tag, attrs):
  35         self.tag_stack.append(tag.lower())
  36         self.attrs_stack.append(dict(attrs))
  37
  38     def handle_endtag(self, tag):
  39         self.tag_stack.pop()
  40         self.attrs_stack.pop()
  41
  42     def handle_data(self, data):
  43         # Check first if this is a creation attachment
  44         if self.tag_stack[-6:-1] == self._ATTACH_TAGS:
  45             attrs_classes = [
  46                 x.get('class', '').lower() for x in self.attrs_stack[-6:-1]
  47             ]
  48             if attrs_classes == self._ATTACH_CLASSES:
  49                 if self.tag_stack[-1] == 'a':
  50                     url = self._PREFIX + self.attrs_stack[-1].get('href')
  51                     self.creation_info['url'] = url
  52                     if '.' in data:
  53                         self.creation_info['ext'] = data.rsplit('.')[-1]
  54         # Next, check if this is within the div containing the creation info
  55         if self.tag_stack[-5:-1] == self._INFO_TAGS:
  56             attrs_classes = [
  57                 x.get('class', '').lower() for x in self.attrs_stack[-5:-1]
  58             ]
  59             if attrs_classes == self._INFO_CLASSES:
  60                 if self.attrs_stack[-1].get('class') == 'utitle':
  61                     self.creation_info['title'] = data.strip()
  62
  63
  64 class PatreonIE(InfoExtractor):
  65     IE_NAME = 'patreon'
  66     _VALID_URL = r'https?://(?:www\.)?patreon\.com/creation\?hid=(.+)'
  67     _TESTS = [
  68         {
  69             'url': 'http://www.patreon.com/creation?hid=743933',
  70             'md5': 'e25505eec1053a6e6813b8ed369875cc',
  71             'name': 'Patreon',
  72             'info_dict': {
  73                 'id': '743933',
  74                 'ext': 'mp3',
  75                 'title': 'Episode 166: David Smalley of Dogma Debate',
  76                 'uploader': 'Cognitive Dissonance Podcast',
  77             },
  78         },
  79     ]
  80
  81     # Currently Patreon exposes download URL via hidden CSS, so login is not
  82     # needed. Keeping this commented for when this inevitably changes.
  83     '''
  84     def _login(self):
  85         (username, password) = self._get_login_info()
  86         if username is None:
  87             return
  88
  89         login_form = {
  90             'redirectUrl': 'http://www.patreon.com/',
  91             'email': username,
  92             'password': password,
  93         }
  94
  95         request = compat_urllib_request.Request(
  96             'https://www.patreon.com/processLogin',
  97             compat_urllib_parse.urlencode(login_form).encode('utf-8')
  98         )
  99         login_page = self._download_webpage(request, None, note='Logging in as %s' % username)
 100
 101         if re.search(r'onLoginFailed', login_page):
 102             raise ExtractorError('Unable to login, incorrect username and/or password', expected=True)
 103
 104     def _real_initialize(self):
 105         self._login()
 106     '''
 107
 108     def _real_extract(self, url):
 109         mobj = re.match(self._VALID_URL, url)
 110         video_id = mobj.group(1)
 111
 112         info_page = self._download_webpage(url, video_id)
 113
 114         ret = {'id': video_id}
 115         try:
 116             ret['uploader'] = re.search(
 117                 r'<strong>(.+)</strong> is creating', info_page
 118             ).group(1)
 119         except AttributeError:
 120             pass
 121
 122         parser = PatreonHTMLParser()
 123         parser.get_creation_info(info_page)
 124         if not parser.creation_info.get('url'):
 125             raise ExtractorError('Unable to retrieve creation URL')
 126         ret.update(parser.creation_info)
 127         return ret