X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/46113edab215c2211a604c06245c16d5d4e57dcf..f8df414a4abcde0ddd39325dac26ca071d2d15c6:/youtube_dl/extractor/patreon.py?ds=sidebyside diff --git a/youtube_dl/extractor/patreon.py b/youtube_dl/extractor/patreon.py index 5429592..a6a2c27 100644 --- a/youtube_dl/extractor/patreon.py +++ b/youtube_dl/extractor/patreon.py @@ -1,17 +1,12 @@ -# encoding: utf-8 +# coding: utf-8 from __future__ import unicode_literals -import json -import re - from .common import InfoExtractor -from ..utils import ( - js_to_json, -) +from ..utils import js_to_json class PatreonIE(InfoExtractor): - _VALID_URL = r'https?://(?:www\.)?patreon\.com/creation\?hid=(.+)' + _VALID_URL = r'https?://(?:www\.)?patreon\.com/creation\?hid=(?P[^&#]+)' _TESTS = [ { 'url': 'http://www.patreon.com/creation?hid=743933', @@ -35,6 +30,23 @@ class PatreonIE(InfoExtractor): 'thumbnail': 're:^https?://.*$', }, }, + { + 'url': 'https://www.patreon.com/creation?hid=1682498', + 'info_dict': { + 'id': 'SU4fj_aEMVw', + 'ext': 'mp4', + 'title': 'I\'m on Patreon!', + 'uploader': 'TraciJHines', + 'thumbnail': 're:^https?://.*$', + 'upload_date': '20150211', + 'description': 'md5:c5a706b1f687817a3de09db1eb93acd4', + 'uploader_id': 'TraciJHines', + }, + 'params': { + 'noplaylist': True, + 'skip_download': True, + } + } ] # Currently Patreon exposes download URL via hidden CSS, so login is not @@ -51,9 +63,9 @@ class PatreonIE(InfoExtractor): 'password': password, } - request = compat_urllib_request.Request( + request = sanitized_Request( 'https://www.patreon.com/processLogin', - compat_urllib_parse.urlencode(login_form).encode('utf-8') + compat_urllib_parse_urlencode(login_form).encode('utf-8') ) login_page = self._download_webpage(request, None, note='Logging in as %s' % username) @@ -65,26 +77,29 @@ class PatreonIE(InfoExtractor): ''' def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - video_id = mobj.group(1) - + video_id = self._match_id(url) webpage = self._download_webpage(url, video_id) title = self._og_search_title(webpage).strip() attach_fn = self._html_search_regex( r'
', webpage, 'attachment URL', default=None) + embed = self._html_search_regex( + r']+id="watchCreation"[^>]*>\s*]+src="([^"]+)"', + webpage, 'embedded URL', default=None) + if attach_fn is not None: video_url = 'http://www.patreon.com' + attach_fn thumbnail = self._og_search_thumbnail(webpage) uploader = self._html_search_regex( r'(.*?) is creating', webpage, 'uploader') + elif embed is not None: + return self.url_result(embed) else: - playlist_js = self._search_regex( + playlist = self._parse_json(self._search_regex( r'(?s)new\s+jPlayerPlaylist\(\s*\{\s*[^}]*},\s*(\[.*?,?\s*\])', - webpage, 'playlist JSON') - playlist_json = js_to_json(playlist_js) - playlist = json.loads(playlist_json) + webpage, 'playlist JSON'), + video_id, transform_source=js_to_json) data = playlist[0] video_url = self._proto_relative_url(data['mp3']) thumbnail = self._proto_relative_url(data.get('cover'))