X-Git-Url: http://git.cielonegro.org/gitweb.cgi?a=blobdiff_plain;f=youtube_dl%2Fextractor%2Fgeneric.py;h=c63bdbd08df14e3944e57ede59f22e1b8717d5f6;hb=d6712378e73951bede475569c887a1ac73f660a9;hp=3c3066e38c5efb4734e1c564be38424d4bd4e004;hpb=974c1b2d4292308a26a47136e7fcf9b61f8b285a;p=youtube-dl.git diff --git a/youtube_dl/extractor/generic.py b/youtube_dl/extractor/generic.py index 3c3066e38..c63bdbd08 100644 --- a/youtube_dl/extractor/generic.py +++ b/youtube_dl/extractor/generic.py @@ -47,6 +47,7 @@ from .senateisvp import SenateISVPIE from .svt import SVTIE from .pornhub import PornHubIE from .xhamster import XHamsterEmbedIE +from .tnaflix import TNAFlixNetworkEmbedIE from .vimeo import VimeoIE from .dailymotion import DailymotionCloudIE from .onionstudios import OnionStudiosIE @@ -54,8 +55,12 @@ from .snagfilms import SnagFilmsEmbedIE from .screenwavemedia import ScreenwaveMediaIE from .mtv import MTVServicesEmbeddedIE from .pladform import PladformIE +from .videomore import VideomoreIE from .googledrive import GoogleDriveIE from .jwplatform import JWPlatformIE +from .digiteka import DigitekaIE +from .instagram import InstagramIE +from .liveleak import LiveLeakIE class GenericIE(InfoExtractor): @@ -100,7 +105,8 @@ class GenericIE(InfoExtractor): 'skip_download': True, # infinite live stream }, 'expected_warnings': [ - r'501.*Not Implemented' + r'501.*Not Implemented', + r'400.*Bad Request', ], }, # Direct link with incorrect MIME type @@ -222,6 +228,50 @@ class GenericIE(InfoExtractor): 'skip_download': True, }, }, + # MPD from http://dash-mse-test.appspot.com/media.html + { + 'url': 'http://yt-dash-mse-test.commondatastorage.googleapis.com/media/car-20120827-manifest.mpd', + 'md5': '4b57baab2e30d6eb3a6a09f0ba57ef53', + 'info_dict': { + 'id': 'car-20120827-manifest', + 'ext': 'mp4', + 'title': 'car-20120827-manifest', + 'formats': 'mincount:9', + 'upload_date': '20130904', + }, + 'params': { + 'format': 'bestvideo', + }, + }, + # m3u8 served with Content-Type: audio/x-mpegURL; charset=utf-8 + { + 'url': 'http://once.unicornmedia.com/now/master/playlist/bb0b18ba-64f5-4b1b-a29f-0ac252f06b68/77a785f3-5188-4806-b788-0893a61634ed/93677179-2d99-4ef4-9e17-fe70d49abfbf/content.m3u8', + 'info_dict': { + 'id': 'content', + 'ext': 'mp4', + 'title': 'content', + 'formats': 'mincount:8', + }, + 'params': { + # m3u8 downloads + 'skip_download': True, + } + }, + # m3u8 served with Content-Type: text/plain + { + 'url': 'http://www.nacentapps.com/m3u8/index.m3u8', + 'info_dict': { + 'id': 'index', + 'ext': 'mp4', + 'title': 'index', + 'upload_date': '20140720', + 'formats': 'mincount:11', + }, + 'params': { + # m3u8 downloads + 'skip_download': True, + } + }, # 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', @@ -359,19 +409,6 @@ class GenericIE(InfoExtractor): 'skip_download': True, }, }, - # multiple ooyala embeds on SBN network websites - { - 'url': 'http://www.sbnation.com/college-football-recruiting/2015/2/3/7970291/national-signing-day-rationalizations-itll-be-ok-itll-be-ok', - 'info_dict': { - 'id': 'national-signing-day-rationalizations-itll-be-ok-itll-be-ok', - 'title': '25 lies you will tell yourself on National Signing Day - SBNation.com', - }, - 'playlist_mincount': 3, - 'params': { - 'skip_download': True, - }, - 'add_ie': ['Ooyala'], - }, # embed.ly video { 'url': 'http://www.tested.com/science/weird/460206-tested-grinding-coffee-2000-frames-second/', @@ -485,7 +522,7 @@ class GenericIE(InfoExtractor): 'description': 'md5:8145d19d320ff3e52f28401f4c4283b9', } }, - # Embeded Ustream video + # Embedded Ustream video { 'url': 'http://www.american.edu/spa/pti/nsa-privacy-janus-2014.cfm', 'md5': '27b99cdb639c9b12a79bca876a073417', @@ -560,7 +597,11 @@ class GenericIE(InfoExtractor): 'id': 'k2mm4bCdJ6CQ2i7c8o2', 'ext': 'mp4', 'title': 'Le Zap de Spi0n n°216 - Zapping du Web', + 'description': 'md5:faf028e48a461b8b7fad38f1e104b119', 'uploader': 'Spi0n', + 'uploader_id': 'xgditw', + 'upload_date': '20140425', + 'timestamp': 1398441542, }, 'add_ie': ['Dailymotion'], }, @@ -693,8 +734,11 @@ class GenericIE(InfoExtractor): 'id': 'uxjb0lwrcz', 'ext': 'mp4', 'title': 'Conversation about Hexagonal Rails Part 1 - ThoughtWorks', + 'description': 'a Martin Fowler video from ThoughtWorks', 'duration': 1715.0, 'uploader': 'thoughtworks.wistia.com', + 'upload_date': '20140603', + 'timestamp': 1401832161, }, }, # Soundcloud embed @@ -945,6 +989,9 @@ class GenericIE(InfoExtractor): 'ext': 'flv', 'title': "PFT Live: New leader in the 'new-look' defense", 'description': 'md5:65a19b4bbfb3b0c0c5768bed1dfad74e', + 'uploader': 'NBCU-SPORTS', + 'upload_date': '20140107', + 'timestamp': 1389118457, }, }, # UDN embed @@ -997,6 +1044,9 @@ class GenericIE(InfoExtractor): 'title': 'SN Presents: Russell Martin, World Citizen', 'description': 'To understand why he was the Toronto Blue Jays’ top off-season priority is to appreciate his background and upbringing in Montreal, where he first developed his baseball skills. Written and narrated by Stephen Brunt.', 'uploader': 'Rogers Sportsnet', + 'uploader_id': '1704050871', + 'upload_date': '20150525', + 'timestamp': 1432570283, }, }, # Dailymotion Cloud video @@ -1077,7 +1127,50 @@ class GenericIE(InfoExtractor): # m3u8 downloads 'skip_download': True, } - } + }, + # Brightcove embed, with no valid 'renditions' but valid 'IOSRenditions' + # This video can't be played in browsers if Flash disabled and UA set to iPhone, which is actually a false alarm + { + 'url': 'https://dl.dropboxusercontent.com/u/29092637/interview.html', + 'info_dict': { + 'id': '4785848093001', + 'ext': 'mp4', + 'title': 'The Cardinal Pell Interview', + 'description': 'Sky News Contributor Andrew Bolt interviews George Pell in Rome, following the Cardinal\'s evidence before the Royal Commission into Child Abuse. ', + 'uploader': 'GlobeCast Australia - GlobeStream', + 'uploader_id': '2733773828001', + 'upload_date': '20160304', + 'timestamp': 1457083087, + }, + 'params': { + # m3u8 downloads + 'skip_download': True, + }, + }, + # Another form of arte.tv embed + { + 'url': 'http://www.tv-replay.fr/redirection/09-04-16/arte-reportage-arte-11508975.html', + 'md5': '850bfe45417ddf221288c88a0cffe2e2', + 'info_dict': { + 'id': '030273-562_PLUS7-F', + 'ext': 'mp4', + 'title': 'ARTE Reportage - Nulle part, en France', + 'description': 'md5:e3a0e8868ed7303ed509b9e3af2b870d', + 'upload_date': '20160409', + }, + }, + # LiveLeak embed + { + 'url': 'http://www.wykop.pl/link/3088787/', + 'md5': 'ace83b9ed19b21f68e1b50e844fdf95d', + 'info_dict': { + 'id': '874_1459135191', + 'ext': 'mp4', + 'title': 'Man shows poor quality of new apartment building', + 'description': 'The wall is like a sand pile.', + 'uploader': 'Lake8737', + } + }, ] def report_following_redirect(self, new_url): @@ -1225,23 +1318,31 @@ class GenericIE(InfoExtractor): full_response = self._request_webpage(request, video_id) head_response = full_response + info_dict = { + 'id': video_id, + 'title': compat_urllib_parse_unquote(os.path.splitext(url_basename(url))[0]), + 'upload_date': unified_strdate(head_response.headers.get('Last-Modified')) + } + # 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) + content_type = head_response.headers.get('Content-Type', '').lower() + m = re.match(r'^(?Paudio|video|application(?=/(?:ogg$|(?:vnd\.apple\.|x-)?mpegurl)))/(?P[^;\s]+)', content_type) if m: - upload_date = unified_strdate( - head_response.headers.get('Last-Modified')) - return { - 'id': video_id, - 'title': compat_urllib_parse_unquote(os.path.splitext(url_basename(url))[0]), - 'direct': True, - 'formats': [{ + format_id = m.group('format_id') + if format_id.endswith('mpegurl'): + formats = self._extract_m3u8_formats(url, video_id, 'mp4') + elif format_id == 'f4m': + formats = self._extract_f4m_formats(url, video_id) + else: + formats = [{ 'format_id': m.group('format_id'), 'url': url, 'vcodec': 'none' if m.group('type') == 'audio' else None - }], - 'upload_date': upload_date, - } + }] + info_dict['direct'] = True + self._sort_formats(formats) + info_dict['formats'] = formats + return info_dict if not self._downloader.params.get('test', False) and not is_intentional: force = self._downloader.params.get('force_generic_extractor', False) @@ -1261,36 +1362,50 @@ class GenericIE(InfoExtractor): request.add_header('Accept-Encoding', '*') full_response = self._request_webpage(request, video_id) + first_bytes = full_response.read(512) + + # Is it an M3U playlist? + if first_bytes.startswith(b'#EXTM3U'): + info_dict['formats'] = self._extract_m3u8_formats(url, video_id, 'mp4') + self._sort_formats(info_dict['formats']) + return info_dict + # Maybe it's a direct link to a video? # Be careful not to download the whole thing! - first_bytes = full_response.read(512) if not is_html(first_bytes): self._downloader.report_warning( 'URL could be a direct video link, returning it as such.') - upload_date = unified_strdate( - head_response.headers.get('Last-Modified')) - return { - 'id': video_id, - 'title': compat_urllib_parse_unquote(os.path.splitext(url_basename(url))[0]), + info_dict.update({ 'direct': True, 'url': url, - 'upload_date': upload_date, - } + }) + return info_dict webpage = self._webpage_read_content( full_response, url, video_id, prefix=first_bytes) self.report_extraction(video_id) - # Is it an RSS feed, a SMIL file or a XSPF playlist? + # Is it an RSS feed, a SMIL file, an XSPF playlist or a MPD manifest? try: doc = compat_etree_fromstring(webpage.encode('utf-8')) if doc.tag == 'rss': return self._extract_rss(url, video_id, doc) elif re.match(r'^(?:{[^}]+})?smil$', doc.tag): - return self._parse_smil(doc, url, video_id) + smil = self._parse_smil(doc, url, video_id) + self._sort_formats(smil['formats']) + return smil elif doc.tag == '{http://xspf.org/ns/0/}playlist': return self.playlist_result(self._parse_xspf(doc, video_id), video_id) + elif re.match(r'(?i)^(?:{[^}]+})?MPD$', doc.tag): + info_dict['formats'] = self._parse_mpd_formats( + doc, video_id, mpd_base_url=url.rpartition('/')[0]) + self._sort_formats(info_dict['formats']) + return info_dict + elif re.match(r'^{http://ns\.adobe\.com/f4m/[12]\.0}manifest$', doc.tag): + info_dict['formats'] = self._parse_f4m_formats(doc, url, video_id) + self._sort_formats(info_dict['formats']) + return info_dict except compat_xml_parse_error: pass @@ -1400,7 +1515,7 @@ class GenericIE(InfoExtractor): # Look for embedded Dailymotion player matches = re.findall( - r']+?src=(["\'])(?P(?:https?:)?//(?:www\.)?dailymotion\.com/embed/video/.+?)\1', webpage) + r'<(?:(?:embed|iframe)[^>]+?src=|input[^>]+id=[\'"]dmcloudUrlEmissionSelect[\'"][^>]+value=)(["\'])(?P(?:https?:)?//(?:www\.)?dailymotion\.com/(?:embed|swf)/video/.+?)\1', webpage) if matches: return _playlist_from_matches( matches, lambda m: unescapeHTML(m[1])) @@ -1545,6 +1660,11 @@ class GenericIE(InfoExtractor): if mobj is not None: return self.url_result(mobj.group('url'), 'VK') + # Look for embedded Odnoklassniki player + mobj = re.search(r']+?src=(["\'])(?Phttps?://(?:odnoklassniki|ok)\.ru/videoembed/.+?)\1', webpage) + if mobj is not None: + return self.url_result(mobj.group('url'), 'Odnoklassniki') + # Look for embedded ivi player mobj = re.search(r']+?src=(["\'])(?Phttps?://(?:www\.)?ivi\.ru/video/player.+?)\1', webpage) if mobj is not None: @@ -1600,6 +1720,11 @@ class GenericIE(InfoExtractor): if xhamster_urls: return _playlist_from_matches(xhamster_urls, ie='XHamsterEmbed') + # Look for embedded TNAFlixNetwork player + tnaflix_urls = TNAFlixNetworkEmbedIE._extract_urls(webpage) + if tnaflix_urls: + return _playlist_from_matches(tnaflix_urls, ie=TNAFlixNetworkEmbedIE.ie_key()) + # Look for embedded Tvigle player mobj = re.search( r']+?src=(["\'])(?P(?:https?:)?//cloud\.tvigle\.ru/video/.+?)\1', webpage) @@ -1620,7 +1745,7 @@ class GenericIE(InfoExtractor): # Look for embedded arte.tv player mobj = re.search( - r'