[libsyn] Strip options from player URL
This commit is contained in:
		| @@ -8,9 +8,9 @@ from ..utils import unified_strdate | ||||
|  | ||||
|  | ||||
| class LibsynIE(InfoExtractor): | ||||
|     _VALID_URL = r'https?://html5-player\.libsyn\.com/embed/episode/id/(?P<id>[0-9]+)' | ||||
|     _VALID_URL = r'(?P<mainurl>https?://html5-player\.libsyn\.com/embed/episode/id/(?P<id>[0-9]+))' | ||||
|  | ||||
|     _TEST = { | ||||
|     _TESTS = [{ | ||||
|         'url': 'http://html5-player.libsyn.com/embed/episode/id/3377616/', | ||||
|         'md5': '443360ee1b58007bc3dcf09b41d093bb', | ||||
|         'info_dict': { | ||||
| @@ -19,12 +19,24 @@ class LibsynIE(InfoExtractor): | ||||
|             'title': "The Daily Show Podcast without Jon Stewart - Episode 12: Bassem Youssef: Egypt's Jon Stewart", | ||||
|             'description': 'md5:601cb790edd05908957dae8aaa866465', | ||||
|             'upload_date': '20150220', | ||||
|             'thumbnail': 're:^https?://.*', | ||||
|         }, | ||||
|     } | ||||
|     }, { | ||||
|         'url': 'https://html5-player.libsyn.com/embed/episode/id/3727166/height/75/width/200/theme/standard/direction/no/autoplay/no/autonext/no/thumbnail/no/preload/no/no_addthis/no/', | ||||
|         'md5': '6c5cb21acd622d754d3b1a92b582ce42', | ||||
|         'info_dict': { | ||||
|             'id': '3727166', | ||||
|             'ext': 'mp3', | ||||
|             'title': 'Clients From Hell Podcast - How a Sex Toy Company Kickstarted my Freelance Career', | ||||
|             'upload_date': '20150818', | ||||
|             'thumbnail': 're:^https?://.*', | ||||
|         } | ||||
|     }] | ||||
|  | ||||
|     def _real_extract(self, url): | ||||
|         video_id = self._match_id(url) | ||||
|  | ||||
|         m = re.match(self._VALID_URL, url) | ||||
|         video_id = m.group('id') | ||||
|         url = m.group('mainurl') | ||||
|         webpage = self._download_webpage(url, video_id) | ||||
|  | ||||
|         formats = [{ | ||||
| @@ -32,20 +44,18 @@ class LibsynIE(InfoExtractor): | ||||
|         } for media_url in set(re.findall('var\s+mediaURL(?:Libsyn)?\s*=\s*"([^"]+)"', webpage))] | ||||
|  | ||||
|         podcast_title = self._search_regex( | ||||
|             r'<h2>([^<]+)</h2>', webpage, 'title') | ||||
|             r'<h2>([^<]+)</h2>', webpage, 'podcast title', default=None) | ||||
|         episode_title = self._search_regex( | ||||
|             r'<h3>([^<]+)</h3>', webpage, 'title', default=None) | ||||
|             r'(?:<div class="episode-title">|<h3>)([^<]+)</', webpage, 'episode title') | ||||
|  | ||||
|         title = '%s - %s' % (podcast_title, episode_title) if podcast_title else episode_title | ||||
|  | ||||
|         description = self._html_search_regex( | ||||
|             r'<div id="info_text_body">(.+?)</div>', webpage, | ||||
|             'description', fatal=False) | ||||
|  | ||||
|             'description', default=None) | ||||
|         thumbnail = self._search_regex( | ||||
|             r'<img[^>]+class="info-show-icon"[^>]+src="([^"]+)"', | ||||
|             webpage, 'thumbnail', fatal=False) | ||||
|  | ||||
|         release_date = unified_strdate(self._search_regex( | ||||
|             r'<div class="release_date">Released: ([^<]+)<', webpage, 'release date', fatal=False)) | ||||
|  | ||||
|   | ||||
		Reference in New Issue
	
	Block a user