X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;f=youtube_dl%2Fextractor%2Fgeneric.py;h=d224aa8e18471787bc00a515836b39fd558bac1f;hb=9e1a5b845586a0a5431fb72467142046d8571e6f;hp=9a2e54d149011f66b08d09860078f23ecf8607a9;hpb=f7300c5c90a99d234a2c7a6d70f5b5baa9d35046;p=youtube-dl diff --git a/youtube_dl/extractor/generic.py b/youtube_dl/extractor/generic.py index 9a2e54d14..d224aa8e1 100644 --- a/youtube_dl/extractor/generic.py +++ b/youtube_dl/extractor/generic.py @@ -4,26 +4,32 @@ from __future__ import unicode_literals import os import re -import xml.etree.ElementTree from .common import InfoExtractor from .youtube import YoutubeIE -from ..utils import ( - compat_urllib_error, +from ..compat import ( compat_urllib_parse, - compat_urllib_request, compat_urlparse, compat_xml_parse_error, - +) +from ..utils import ( + determine_ext, ExtractorError, + float_or_none, HEADRequest, + orderedSet, + parse_xml, smuggle_url, unescapeHTML, unified_strdate, + unsmuggle_url, url_basename, ) from .brightcove import BrightcoveIE from .ooyala import OoyalaIE +from .rutv import RUTVIE +from .smotri import SmotriIE +from .condenast import CondeNastIE class GenericIE(InfoExtractor): @@ -33,9 +39,10 @@ class GenericIE(InfoExtractor): _TESTS = [ { 'url': 'http://www.hodiho.fr/2013/02/regis-plante-sa-jeep.html', - 'file': '13601338388002.mp4', - 'md5': '6e15c93721d7ec9e9ca3fdbf07982cfd', + 'md5': '85b90ccc9d73b4acd9138d3af4c27f89', 'info_dict': { + 'id': '13601338388002', + 'ext': 'mp4', 'uploader': 'www.hodiho.fr', 'title': 'R\u00e9gis plante sa Jeep', } @@ -44,8 +51,9 @@ class GenericIE(InfoExtractor): { 'add_ie': ['Bandcamp'], 'url': 'http://bronyrock.com/track/the-pony-mash', - 'file': '3235767654.mp3', 'info_dict': { + 'id': '3235767654', + 'ext': 'mp3', 'title': 'The Pony Mash', 'uploader': 'M_Pallante', }, @@ -71,22 +79,50 @@ class GenericIE(InfoExtractor): { # https://github.com/rg3/youtube-dl/issues/2253 'url': 'http://bcove.me/i6nfkrc3', - 'file': '3101154703001.mp4', 'md5': '0ba9446db037002366bab3b3eb30c88c', 'info_dict': { + 'id': '3101154703001', + 'ext': 'mp4', 'title': 'Still no power', 'uploader': 'thestar.com', 'description': 'Mississauga resident David Farmer is still out of power as a result of the ice storm a month ago. To keep the house warm, Farmer cuts wood from his property for a wood burning stove downstairs.', }, 'add_ie': ['Brightcove'], }, + { + 'url': 'http://www.championat.com/video/football/v/87/87499.html', + 'md5': 'fb973ecf6e4a78a67453647444222983', + 'info_dict': { + 'id': '3414141473001', + 'ext': 'mp4', + 'title': 'Видео. Удаление Дзагоева (ЦСКА)', + 'description': 'Онлайн-трансляция матча ЦСКА - "Волга"', + 'uploader': 'Championat', + }, + }, + { + # https://github.com/rg3/youtube-dl/issues/3541 + 'add_ie': ['Brightcove'], + 'url': 'http://www.kijk.nl/sbs6/leermijvrouwenkennen/videos/jqMiXKAYan2S/aflevering-1', + 'info_dict': { + 'id': '3866516442001', + 'ext': 'mp4', + 'title': 'Leer mij vrouwen kennen: Aflevering 1', + 'description': 'Leer mij vrouwen kennen: Aflevering 1', + 'uploader': 'SBS Broadcasting', + }, + 'skip': 'Restricted to Netherlands', + 'params': { + 'skip_download': True, # m3u8 download + }, + }, # Direct link to a video { 'url': 'http://media.w3.org/2010/05/sintel/trailer.mp4', - 'file': 'trailer.mp4', 'md5': '67d406c2bcb6af27fa886f31aa934bbe', 'info_dict': { 'id': 'trailer', + 'ext': 'mp4', 'title': 'trailer', 'upload_date': '20100513', } @@ -94,7 +130,6 @@ class GenericIE(InfoExtractor): # ooyala video { 'url': 'http://www.rollingstone.com/music/videos/norwegian-dj-cashmere-cat-goes-spartan-on-with-me-premiere-20131219', - 'file': 'BwY2RxaTrTkslxOfcan0UCf0YqyvWysJ.mp4', 'md5': '5644c6ca5d5782c1d0d350dad9bd840c', 'info_dict': { 'id': 'BwY2RxaTrTkslxOfcan0UCf0YqyvWysJ', @@ -102,65 +137,320 @@ class GenericIE(InfoExtractor): 'title': '2cc213299525360.mov', # that's what we get }, }, + # google redirect + { + 'url': 'http://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=1&cad=rja&ved=0CCUQtwIwAA&url=http%3A%2F%2Fwww.youtube.com%2Fwatch%3Fv%3DcmQHVoWB5FY&ei=F-sNU-LLCaXk4QT52ICQBQ&usg=AFQjCNEw4hL29zgOohLXvpJ-Bdh2bils1Q&bvm=bv.61965928,d.bGE', + 'info_dict': { + 'id': 'cmQHVoWB5FY', + 'ext': 'mp4', + 'upload_date': '20130224', + 'uploader_id': 'TheVerge', + 'description': 'Chris Ziegler takes a look at the Alcatel OneTouch Fire and the ZTE Open; two of the first Firefox OS handsets to be officially announced.', + 'uploader': 'The Verge', + 'title': 'First Firefox OS phones side-by-side', + }, + 'params': { + 'skip_download': False, + } + }, + # embed.ly video + { + 'url': 'http://www.tested.com/science/weird/460206-tested-grinding-coffee-2000-frames-second/', + 'info_dict': { + 'id': '9ODmcdjQcHQ', + 'ext': 'mp4', + 'title': 'Tested: Grinding Coffee at 2000 Frames Per Second', + 'upload_date': '20140225', + 'description': 'md5:06a40fbf30b220468f1e0957c0f558ff', + 'uploader': 'Tested', + 'uploader_id': 'testedcom', + }, + # No need to test YoutubeIE here + 'params': { + 'skip_download': True, + }, + }, + # funnyordie embed + { + 'url': 'http://www.theguardian.com/world/2014/mar/11/obama-zach-galifianakis-between-two-ferns', + 'info_dict': { + 'id': '18e820ec3f', + 'ext': 'mp4', + 'title': 'Between Two Ferns with Zach Galifianakis: President Barack Obama', + 'description': 'Episode 18: President Barack Obama sits down with Zach Galifianakis for his most memorable interview yet.', + }, + }, + # RUTV embed + { + 'url': 'http://www.rg.ru/2014/03/15/reg-dfo/anklav-anons.html', + 'info_dict': { + 'id': '776940', + 'ext': 'mp4', + 'title': 'Охотское море стало целиком российским', + 'description': 'md5:5ed62483b14663e2a95ebbe115eb8f43', + }, + 'params': { + # m3u8 download + 'skip_download': True, + }, + }, + # Embedded TED video + { + 'url': 'http://en.support.wordpress.com/videos/ted-talks/', + 'md5': '65fdff94098e4a607385a60c5177c638', + 'info_dict': { + 'id': '1969', + 'ext': 'mp4', + 'title': 'Hidden miracles of the natural world', + 'uploader': 'Louie Schwartzberg', + 'description': 'md5:8145d19d320ff3e52f28401f4c4283b9', + } + }, + # Embeded Ustream video + { + 'url': 'http://www.american.edu/spa/pti/nsa-privacy-janus-2014.cfm', + 'md5': '27b99cdb639c9b12a79bca876a073417', + 'info_dict': { + 'id': '45734260', + 'ext': 'flv', + 'uploader': 'AU SPA: The NSA and Privacy', + 'title': 'NSA and Privacy Forum Debate featuring General Hayden and Barton Gellman' + } + }, + # nowvideo embed hidden behind percent encoding + { + 'url': 'http://www.waoanime.tv/the-super-dimension-fortress-macross-episode-1/', + 'md5': '2baf4ddd70f697d94b1c18cf796d5107', + 'info_dict': { + 'id': '06e53103ca9aa', + 'ext': 'flv', + 'title': 'Macross Episode 001 Watch Macross Episode 001 onl', + 'description': 'No description', + }, + }, + # arte embed + { + 'url': 'http://www.tv-replay.fr/redirection/20-03-14/x-enius-arte-10753389.html', + 'md5': '7653032cbb25bf6c80d80f217055fa43', + 'info_dict': { + 'id': '048195-004_PLUS7-F', + 'ext': 'flv', + 'title': 'X:enius', + 'description': 'md5:d5fdf32ef6613cdbfd516ae658abf168', + 'upload_date': '20140320', + }, + 'params': { + 'skip_download': 'Requires rtmpdump' + } + }, + # Condé Nast embed + { + 'url': 'http://www.wired.com/2014/04/honda-asimo/', + 'md5': 'ba0dfe966fa007657bd1443ee672db0f', + 'info_dict': { + 'id': '53501be369702d3275860000', + 'ext': 'mp4', + 'title': 'Honda’s New Asimo Robot Is More Human Than Ever', + } + }, + # Dailymotion embed + { + 'url': 'http://www.spi0n.com/zap-spi0n-com-n216/', + 'md5': '441aeeb82eb72c422c7f14ec533999cd', + 'info_dict': { + 'id': 'k2mm4bCdJ6CQ2i7c8o2', + 'ext': 'mp4', + 'title': 'Le Zap de Spi0n n°216 - Zapping du Web', + 'uploader': 'Spi0n', + }, + 'add_ie': ['Dailymotion'], + }, + # YouTube embed + { + 'url': 'http://www.badzine.de/ansicht/datum/2014/06/09/so-funktioniert-die-neue-englische-badminton-liga.html', + 'info_dict': { + 'id': 'FXRb4ykk4S0', + 'ext': 'mp4', + 'title': 'The NBL Auction 2014', + 'uploader': 'BADMINTON England', + 'uploader_id': 'BADMINTONEvents', + 'upload_date': '20140603', + 'description': 'md5:9ef128a69f1e262a700ed83edb163a73', + }, + 'add_ie': ['Youtube'], + 'params': { + 'skip_download': True, + } + }, + # MTVSercices embed + { + 'url': 'http://www.gametrailers.com/news-post/76093/north-america-europe-is-getting-that-mario-kart-8-mercedes-dlc-too', + 'md5': '35727f82f58c76d996fc188f9755b0d5', + 'info_dict': { + 'id': '0306a69b-8adf-4fb5-aace-75f8e8cbfca9', + 'ext': 'mp4', + 'title': 'Review', + 'description': 'Mario\'s life in the fast lane has never looked so good.', + }, + }, + # YouTube embed via + { + 'url': 'https://play.google.com/store/apps/details?id=com.gameloft.android.ANMP.GloftA8HM', + 'info_dict': { + 'id': '4vAffPZIT44', + 'ext': 'mp4', + 'title': 'Asphalt 8: Airborne - Update - Welcome to Dubai!', + 'uploader': 'Gameloft', + 'uploader_id': 'gameloft', + 'upload_date': '20140828', + 'description': 'md5:c80da9ed3d83ae6d1876c834de03e1c4', + }, + 'params': { + 'skip_download': True, + } + }, + # Camtasia studio + { + 'url': 'http://www.ll.mit.edu/workshops/education/videocourses/antennas/lecture1/video/', + 'playlist': [{ + 'md5': '0c5e352edabf715d762b0ad4e6d9ee67', + 'info_dict': { + 'id': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final', + 'title': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final - video1', + 'ext': 'flv', + 'duration': 2235.90, + } + }, { + 'md5': '10e4bb3aaca9fd630e273ff92d9f3c63', + 'info_dict': { + 'id': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final_PIP', + 'title': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final - pip', + 'ext': 'flv', + 'duration': 2235.93, + } + }], + 'info_dict': { + 'title': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final', + } + }, + # Flowplayer + { + 'url': 'http://www.handjobhub.com/video/busty-blonde-siri-tit-fuck-while-wank-6313.html', + 'md5': '9d65602bf31c6e20014319c7d07fba27', + 'info_dict': { + 'id': '5123ea6d5e5a7', + 'ext': 'mp4', + 'age_limit': 18, + 'uploader': 'www.handjobhub.com', + 'title': 'Busty Blonde Siri Tit Fuck While Wank at HandjobHub.com', + } + }, + # RSS feed + { + 'url': 'http://phihag.de/2014/youtube-dl/rss2.xml', + 'info_dict': { + 'id': 'http://phihag.de/2014/youtube-dl/rss2.xml', + 'title': 'Zero Punctuation', + 'description': 're:' + }, + 'playlist_mincount': 11, + }, + # Multiple brightcove videos + # https://github.com/rg3/youtube-dl/issues/2283 + { + 'url': 'http://www.newyorker.com/online/blogs/newsdesk/2014/01/always-never-nuclear-command-and-control.html', + 'info_dict': { + 'id': 'always-never', + 'title': 'Always / Never - The New Yorker', + }, + 'playlist_count': 3, + 'params': { + 'extract_flat': False, + 'skip_download': True, + } + }, + # MLB embed + { + 'url': 'http://umpire-empire.com/index.php/topic/58125-laz-decides-no-thats-low/', + 'md5': '96f09a37e44da40dd083e12d9a683327', + 'info_dict': { + 'id': '33322633', + 'ext': 'mp4', + 'title': 'Ump changes call to ball', + 'description': 'md5:71c11215384298a172a6dcb4c2e20685', + 'duration': 48, + 'timestamp': 1401537900, + 'upload_date': '20140531', + 'thumbnail': 're:^https?://.*\.jpg$', + }, + }, + # Wistia embed + { + 'url': 'http://education-portal.com/academy/lesson/north-american-exploration-failed-colonies-of-spain-france-england.html#lesson', + 'md5': '8788b683c777a5cf25621eaf286d0c23', + 'info_dict': { + 'id': '1cfaf6b7ea', + 'ext': 'mov', + 'title': 'md5:51364a8d3d009997ba99656004b5e20d', + 'duration': 643.0, + 'filesize': 182808282, + 'uploader': 'education-portal.com', + }, + }, + { + 'url': 'http://thoughtworks.wistia.com/medias/uxjb0lwrcz', + 'md5': 'baf49c2baa8a7de5f3fc145a8506dcd4', + 'info_dict': { + 'id': 'uxjb0lwrcz', + 'ext': 'mp4', + 'title': 'Conversation about Hexagonal Rails Part 1 - ThoughtWorks', + 'duration': 1715.0, + 'uploader': 'thoughtworks.wistia.com', + }, + }, + # Direct download with broken HEAD + { + 'url': 'http://ai-radio.org:8000/radio.opus', + 'info_dict': { + 'id': 'radio', + 'ext': 'opus', + 'title': 'radio', + }, + 'params': { + 'skip_download': True, # infinite live stream + }, + 'expected_warnings': [ + r'501.*Not Implemented' + ], + }, + # Soundcloud embed + { + 'url': 'http://nakedsecurity.sophos.com/2014/10/29/sscc-171-are-you-sure-that-1234-is-a-bad-password-podcast/', + 'info_dict': { + 'id': '174391317', + 'ext': 'mp3', + 'description': 'md5:ff867d6b555488ad3c52572bb33d432c', + 'uploader': 'Sophos Security', + 'title': 'Chet Chat 171 - Oct 29, 2014', + 'upload_date': '20141029', + } + }, + # Livestream embed + { + 'url': 'http://www.esa.int/Our_Activities/Space_Science/Rosetta/Philae_comet_touch-down_webcast', + 'info_dict': { + 'id': '67864563', + 'ext': 'flv', + 'upload_date': '20141112', + 'title': 'Rosetta #CometLanding webcast HL 10', + } + }, ] - def report_download_webpage(self, video_id): - """Report webpage download.""" - if not self._downloader.params.get('test', False): - self._downloader.report_warning('Falling back on generic information extractor.') - super(GenericIE, self).report_download_webpage(video_id) - def report_following_redirect(self, new_url): """Report information extraction.""" self._downloader.to_screen('[redirect] Following redirect to %s' % new_url) - def _send_head(self, url): - """Check if it is a redirect, like url shorteners, in case return the new url.""" - - class HEADRedirectHandler(compat_urllib_request.HTTPRedirectHandler): - """ - Subclass the HTTPRedirectHandler to make it use our - HEADRequest also on the redirected URL - """ - def redirect_request(self, req, fp, code, msg, headers, newurl): - if code in (301, 302, 303, 307): - newurl = newurl.replace(' ', '%20') - newheaders = dict((k,v) for k,v in req.headers.items() - if k.lower() not in ("content-length", "content-type")) - return HEADRequest(newurl, - headers=newheaders, - origin_req_host=req.get_origin_req_host(), - unverifiable=True) - else: - raise compat_urllib_error.HTTPError(req.get_full_url(), code, msg, headers, fp) - - class HTTPMethodFallback(compat_urllib_request.BaseHandler): - """ - Fallback to GET if HEAD is not allowed (405 HTTP error) - """ - def http_error_405(self, req, fp, code, msg, headers): - fp.read() - fp.close() - - newheaders = dict((k,v) for k,v in req.headers.items() - if k.lower() not in ("content-length", "content-type")) - return self.parent.open(compat_urllib_request.Request(req.get_full_url(), - headers=newheaders, - origin_req_host=req.get_origin_req_host(), - unverifiable=True)) - - # Build our opener - opener = compat_urllib_request.OpenerDirector() - for handler in [compat_urllib_request.HTTPHandler, compat_urllib_request.HTTPDefaultErrorHandler, - HTTPMethodFallback, HEADRedirectHandler, - compat_urllib_request.HTTPErrorProcessor, compat_urllib_request.HTTPSHandler]: - opener.add_handler(handler()) - - response = opener.open(HEADRequest(url)) - if response is None: - raise ExtractorError('Invalid URL protocol') - return response - def _extract_rss(self, url, video_id, doc): playlist_title = doc.find('./channel/title').text playlist_desc_el = doc.find('./channel/description') @@ -180,74 +470,158 @@ class GenericIE(InfoExtractor): 'entries': entries, } + def _extract_camtasia(self, url, video_id, webpage): + """ Returns None if no camtasia video can be found. """ + + camtasia_cfg = self._search_regex( + r'fo\.addVariable\(\s*"csConfigFile",\s*"([^"]+)"\s*\);', + webpage, 'camtasia configuration file', default=None) + if camtasia_cfg is None: + return None + + title = self._html_search_meta('DC.title', webpage, fatal=True) + + camtasia_url = compat_urlparse.urljoin(url, camtasia_cfg) + camtasia_cfg = self._download_xml( + camtasia_url, video_id, + note='Downloading camtasia configuration', + errnote='Failed to download camtasia configuration') + fileset_node = camtasia_cfg.find('./playlist/array/fileset') + + entries = [] + for n in fileset_node.getchildren(): + url_n = n.find('./uri') + if url_n is None: + continue + + entries.append({ + 'id': os.path.splitext(url_n.text.rpartition('/')[2])[0], + 'title': '%s - %s' % (title, n.tag), + 'url': compat_urlparse.urljoin(url, url_n.text), + 'duration': float_or_none(n.find('./duration').text), + }) + + return { + '_type': 'playlist', + 'entries': entries, + 'title': title, + } + def _real_extract(self, url): + if url.startswith('//'): + return { + '_type': 'url', + 'url': self.http_scheme() + url, + } + parsed_url = compat_urlparse.urlparse(url) if not parsed_url.scheme: default_search = self._downloader.params.get('default_search') if default_search is None: - default_search = 'auto' + default_search = 'fixup_error' - if default_search == 'auto': + if default_search in ('auto', 'auto_warning', 'fixup_error'): if '/' in url: self._downloader.report_warning('The url doesn\'t specify the protocol, trying with http') return self.url_result('http://' + url) - else: + elif default_search != 'fixup_error': + if default_search == 'auto_warning': + if re.match(r'^(?:url|URL)$', url): + raise ExtractorError( + 'Invalid URL: %r . Call youtube-dl like this: youtube-dl -v "https://www.youtube.com/watch?v=BaW_jenozKc" ' % url, + expected=True) + else: + self._downloader.report_warning( + 'Falling back to youtube search for %s . Set --default-search "auto" to suppress this warning.' % url) return self.url_result('ytsearch:' + url) + + if default_search in ('error', 'fixup_error'): + raise ExtractorError( + ('%r is not a valid URL. ' + 'Set --default-search "ytsearch" (or run youtube-dl "ytsearch:%s" ) to search YouTube' + ) % (url, url), expected=True) else: - assert ':' in default_search + if ':' not in default_search: + default_search += ':' return self.url_result(default_search + url) - video_id = os.path.splitext(url.split('/')[-1])[0] + + url, smuggled_data = unsmuggle_url(url) + force_videoid = None + is_intentional = smuggled_data and smuggled_data.get('to_generic') + if smuggled_data and 'force_videoid' in smuggled_data: + force_videoid = smuggled_data['force_videoid'] + video_id = force_videoid + else: + video_id = os.path.splitext(url.rstrip('/').split('/')[-1])[0] self.to_screen('%s: Requesting header' % video_id) - try: - response = self._send_head(url) + head_req = HEADRequest(url) + head_response = self._request_webpage( + head_req, video_id, + note=False, errnote='Could not send HEAD request to %s' % url, + fatal=False) + if head_response is not False: # Check for redirect - new_url = response.geturl() + new_url = head_response.geturl() if url != new_url: self.report_following_redirect(new_url) + if force_videoid: + new_url = smuggle_url( + new_url, {'force_videoid': force_videoid}) return self.url_result(new_url) - # Check for direct link to a video - content_type = response.headers.get('Content-Type', '') - m = re.match(r'^(?Paudio|video|application(?=/ogg$))/(?P.+)$', content_type) - if m: - upload_date = response.headers.get('Last-Modified') - if upload_date: - upload_date = unified_strdate(upload_date) - return { - 'id': video_id, - 'title': os.path.splitext(url_basename(url))[0], - 'formats': [{ - 'format_id': m.group('format_id'), - 'url': url, - 'vcodec': 'none' if m.group('type') == 'audio' else None - }], - 'upload_date': upload_date, - } + full_response = None + if head_response is False: + full_response = self._request_webpage(url, video_id) + head_response = full_response - except compat_urllib_error.HTTPError: - # This may be a stupid server that doesn't like HEAD, our UA, or so - pass + # Check for direct link to a video + content_type = head_response.headers.get('Content-Type', '') + m = re.match(r'^(?Paudio|video|application(?=/ogg$))/(?P.+)$', content_type) + if m: + upload_date = unified_strdate( + head_response.headers.get('Last-Modified')) + return { + 'id': video_id, + 'title': os.path.splitext(url_basename(url))[0], + 'direct': True, + 'formats': [{ + 'format_id': m.group('format_id'), + 'url': url, + 'vcodec': 'none' if m.group('type') == 'audio' else None + }], + 'upload_date': upload_date, + } - try: - webpage = self._download_webpage(url, video_id) - except ValueError: - # since this is the last-resort InfoExtractor, if - # this error is thrown, it'll be thrown here - raise ExtractorError('Failed to download URL: %s' % url) + if not self._downloader.params.get('test', False) and not is_intentional: + self._downloader.report_warning('Falling back on generic information extractor.') + if full_response: + webpage = self._webpage_read_content(full_response, url, video_id) + else: + webpage = self._download_webpage(url, video_id) self.report_extraction(video_id) # Is it an RSS feed? try: - doc = xml.etree.ElementTree.fromstring(webpage.encode('utf-8')) + doc = parse_xml(webpage) if doc.tag == 'rss': return self._extract_rss(url, video_id, doc) except compat_xml_parse_error: pass + # Is it a Camtasia project? + camtasia_res = self._extract_camtasia(url, video_id, webpage) + if camtasia_res is not None: + return camtasia_res + + # Sometimes embedded video player is hidden behind percent encoding + # (e.g. https://github.com/rg3/youtube-dl/issues/2448) + # Unescaping the whole page allows to handle those cases in a generic way + webpage = compat_urllib_parse.unquote(webpage) + # it's tempting to parse this further, but you would # have to take into account all the variations like # Video Title - Site Name @@ -258,10 +632,28 @@ class GenericIE(InfoExtractor): r'(?s)(.*?)', webpage, 'video title', default='video') + # Try to detect age limit automatically + age_limit = self._rta_search(webpage) + # And then there are the jokers who advertise that they use RTA, + # but actually don't. + AGE_LIMIT_MARKERS = [ + r'Proudly Labeled RTA', + ] + if any(re.search(marker, webpage) for marker in AGE_LIMIT_MARKERS): + age_limit = 18 + # video uploader is domain name video_uploader = self._search_regex( r'^(?:https?://)?([^/]*)/.*', url, 'video uploader') + # Helper method + def _playlist_from_matches(matches, getter, ie=None): + urlrs = orderedSet( + self.url_result(self._proto_relative_url(getter(m)), ie) + for m in matches) + return self.playlist_result( + urlrs, playlist_id=video_id, playlist_title=video_title) + # Look for BrightCove: bc_urls = BrightcoveIE._extract_brightcove_urls(webpage) if bc_urls: @@ -281,60 +673,102 @@ class GenericIE(InfoExtractor): # Look for embedded (iframe) Vimeo player mobj = re.search( - r']+?src="((?:https?:)?//player\.vimeo\.com/video/.+?)"', webpage) + r']+?src=(["\'])(?P(?:https?:)?//player\.vimeo\.com/video/.+?)\1', webpage) if mobj: - player_url = unescapeHTML(mobj.group(1)) + player_url = unescapeHTML(mobj.group('url')) surl = smuggle_url(player_url, {'Referer': url}) - return self.url_result(surl, 'Vimeo') + return self.url_result(surl) # Look for embedded (swf embed) Vimeo player mobj = re.search( - r']+?src="(https?://(?:www\.)?vimeo\.com/moogaloop\.swf.+?)"', webpage) + r']+?src="((?:https?:)?//(?:www\.)?vimeo\.com/moogaloop\.swf.+?)"', webpage) if mobj: - return self.url_result(mobj.group(1), 'Vimeo') + return self.url_result(mobj.group(1)) # Look for embedded YouTube player matches = re.findall(r'''(?x) - (?:]+?src=|embedSWF\(\s*) - (["\'])(?P(?:https?:)?//(?:www\.)?youtube\.com/ - (?:embed|v)/.+?) + (?: + ]+?src=| + data-video-url=| + ]+?src=| + embedSWF\(?:\s*| + new\s+SWFObject\( + ) + (["\']) + (?P(?:https?:)?//(?:www\.)?youtube(?:-nocookie)?\.com/ + (?:embed|v|p)/.+?) \1''', webpage) if matches: - urlrs = [self.url_result(unescapeHTML(tuppl[1]), 'Youtube') - for tuppl in matches] - return self.playlist_result( - urlrs, playlist_id=video_id, playlist_title=video_title) + return _playlist_from_matches( + matches, lambda m: unescapeHTML(m[1])) # Look for embedded Dailymotion player matches = re.findall( r']+?src=(["\'])(?P(?:https?:)?//(?:www\.)?dailymotion\.com/embed/video/.+?)\1', webpage) if matches: - urlrs = [self.url_result(unescapeHTML(tuppl[1]), 'Dailymotion') - for tuppl in matches] - return self.playlist_result( - urlrs, playlist_id=video_id, playlist_title=video_title) + return _playlist_from_matches( + matches, lambda m: unescapeHTML(m[1])) + + # Look for embedded Dailymotion playlist player (#3822) + m = re.search( + r']+?src=(["\'])(?P(?:https?:)?//(?:www\.)?dailymotion\.[a-z]{2,3}/widget/jukebox\?.+?)\1', webpage) + if m: + playlists = re.findall( + r'list\[\]=/playlist/([^/]+)/', unescapeHTML(m.group('url'))) + if playlists: + return _playlist_from_matches( + playlists, lambda p: '//dailymotion.com/playlist/%s' % p) # Look for embedded Wistia player match = re.search( - r']+?src=(["\'])(?P(?:https?:)?//(?:fast\.)?wistia\.net/embed/iframe/.+?)\1', webpage) + r'<(?:meta[^>]+?content|iframe[^>]+?src)=(["\'])(?P(?:https?:)?//(?:fast\.)?wistia\.net/embed/iframe/.+?)\1', webpage) if match: + embed_url = self._proto_relative_url( + unescapeHTML(match.group('url'))) return { '_type': 'url_transparent', - 'url': unescapeHTML(match.group('url')), + 'url': embed_url, 'ie_key': 'Wistia', 'uploader': video_uploader, 'title': video_title, 'id': video_id, } + match = re.search(r'(?:id=["\']wistia_|data-wistia-?id=["\']|Wistia\.embed\(["\'])(?P[^"\']+)', webpage) + if match: + return { + '_type': 'url_transparent', + 'url': 'http://fast.wistia.net/embed/iframe/{0:}'.format(match.group('id')), + 'ie_key': 'Wistia', + 'uploader': video_uploader, + 'title': video_title, + 'id': match.group('id') + } + # Look for embedded blip.tv player mobj = re.search(r']*https?://api\.blip\.tv/\w+/redirect/\w+/(\d+)', webpage) if mobj: - return self.url_result('http://blip.tv/a/a-'+mobj.group(1), 'BlipTV') - mobj = re.search(r'<(?:iframe|embed|object)\s[^>]*(https?://(?:\w+\.)?blip\.tv/(?:play/|api\.swf#)[a-zA-Z0-9]+)', webpage) + return self.url_result('http://blip.tv/a/a-' + mobj.group(1), 'BlipTV') + mobj = re.search(r'<(?:iframe|embed|object)\s[^>]*(https?://(?:\w+\.)?blip\.tv/(?:play/|api\.swf#)[a-zA-Z0-9_]+)', webpage) if mobj: return self.url_result(mobj.group(1), 'BlipTV') + # Look for embedded condenast player + matches = re.findall( + r']*?content="(.*?bandcamp\.com.*?)"', webpage) if mobj is not None: @@ -349,12 +783,13 @@ class GenericIE(InfoExtractor): return self.url_result(mobj.group('url')) # Look for Ooyala videos - mobj = re.search(r'player.ooyala.com/[^"?]+\?[^"]*?(?:embedCode|ec)=([^"&]+)', webpage) + mobj = (re.search(r'player.ooyala.com/[^"?]+\?[^"]*?(?:embedCode|ec)=(?P[^"&]+)', webpage) or + re.search(r'OO.Player.create\([\'"].*?[\'"],\s*[\'"](?P.{32})[\'"]', webpage)) if mobj is not None: - return OoyalaIE._build_url_result(mobj.group(1)) + return OoyalaIE._build_url_result(mobj.group('ec')) # Look for Aparat videos - mobj = re.search(r'