X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;f=youtube_dl%2Fextractor%2Fgeneric.py;h=f3de738f765819da7cbda1332e9b6b6dcada5052;hb=c9c39c22c5740c1eedcc9ce7a10f5df199ea5c78;hp=abd98e500d621c19cbb9928a35a04d5fdf40e327;hpb=5b251628e9f45c89c1becb3f62c4212874eb74ea;p=youtube-dl diff --git a/youtube_dl/extractor/generic.py b/youtube_dl/extractor/generic.py index abd98e500..f3de738f7 100644 --- a/youtube_dl/extractor/generic.py +++ b/youtube_dl/extractor/generic.py @@ -4,12 +4,13 @@ from __future__ import unicode_literals import os import re +import sys from .common import InfoExtractor from .youtube import YoutubeIE from ..compat import ( + compat_etree_fromstring, compat_urllib_parse_unquote, - compat_urllib_request, compat_urlparse, compat_xml_parse_error, ) @@ -20,7 +21,7 @@ from ..utils import ( HEADRequest, is_html, orderedSet, - parse_xml, + sanitized_Request, smuggle_url, unescapeHTML, unified_strdate, @@ -29,7 +30,10 @@ from ..utils import ( url_basename, xpath_text, ) -from .brightcove import BrightcoveIE +from .brightcove import ( + BrightcoveLegacyIE, + BrightcoveNewIE, +) from .nbc import NBCSportsVPlayerIE from .ooyala import OoyalaIE from .rutv import RUTVIE @@ -40,15 +44,22 @@ from .myvi import MyviIE from .condenast import CondeNastIE from .udn import UDNEmbedIE from .senateisvp import SenateISVPIE -from .bliptv import BlipTVIE 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 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 class GenericIE(InfoExtractor): @@ -131,6 +142,133 @@ class GenericIE(InfoExtractor): 'title': 'pdv_maddow_netcast_m4v-02-27-2015-201624', } }, + # SMIL from http://videolectures.net/promogram_igor_mekjavic_eng + { + 'url': 'http://videolectures.net/promogram_igor_mekjavic_eng/video/1/smil.xml', + 'info_dict': { + 'id': 'smil', + 'ext': 'mp4', + 'title': 'Automatics, robotics and biocybernetics', + 'description': 'md5:815fc1deb6b3a2bff99de2d5325be482', + 'upload_date': '20130627', + 'formats': 'mincount:16', + 'subtitles': 'mincount:1', + }, + 'params': { + 'force_generic_extractor': True, + 'skip_download': True, + }, + }, + # SMIL from http://www1.wdr.de/mediathek/video/livestream/index.html + { + 'url': 'http://metafilegenerator.de/WDR/WDR_FS/hds/hds.smil', + 'info_dict': { + 'id': 'hds', + 'ext': 'flv', + 'title': 'hds', + 'formats': 'mincount:1', + }, + 'params': { + 'skip_download': True, + }, + }, + # SMIL from https://www.restudy.dk/video/play/id/1637 + { + 'url': 'https://www.restudy.dk/awsmedia/SmilDirectory/video_1637.xml', + 'info_dict': { + 'id': 'video_1637', + 'ext': 'flv', + 'title': 'video_1637', + 'formats': 'mincount:3', + }, + 'params': { + 'skip_download': True, + }, + }, + # SMIL from http://adventure.howstuffworks.com/5266-cool-jobs-iditarod-musher-video.htm + { + 'url': 'http://services.media.howstuffworks.com/videos/450221/smil-service.smil', + 'info_dict': { + 'id': 'smil-service', + 'ext': 'flv', + 'title': 'smil-service', + 'formats': 'mincount:1', + }, + 'params': { + 'skip_download': True, + }, + }, + # SMIL from http://new.livestream.com/CoheedandCambria/WebsterHall/videos/4719370 + { + 'url': 'http://api.new.livestream.com/accounts/1570303/events/1585861/videos/4719370.smil', + 'info_dict': { + 'id': '4719370', + 'ext': 'mp4', + 'title': '571de1fd-47bc-48db-abf9-238872a58d1f', + 'formats': 'mincount:3', + }, + 'params': { + 'skip_download': True, + }, + }, + # XSPF playlist from http://www.telegraaf.nl/tv/nieuws/binnenland/24353229/__Tikibad_ontruimd_wegens_brand__.html + { + 'url': 'http://www.telegraaf.nl/xml/playlist/2015/8/7/mZlp2ctYIUEB.xspf', + 'info_dict': { + 'id': 'mZlp2ctYIUEB', + 'ext': 'mp4', + 'title': 'Tikibad ontruimd wegens brand', + 'description': 'md5:05ca046ff47b931f9b04855015e163a4', + 'thumbnail': 're:^https?://.*\.jpg$', + 'duration': 33, + }, + 'params': { + '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', + }, + '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', @@ -147,6 +285,22 @@ class GenericIE(InfoExtractor): 'skip_download': False, } }, + { + # redirect in Refresh HTTP header + 'url': 'https://www.facebook.com/l.php?u=https%3A%2F%2Fwww.youtube.com%2Fwatch%3Fv%3DpO8h3EaFRdo&h=TAQHsoToz&enc=AZN16h-b6o4Zq9pZkCCdOLNKMN96BbGMNtcFwHSaazus4JHT_MFYkAA-WARTX2kvsCIdlAIyHZjl6d33ILIJU7Jzwk_K3mcenAXoAzBNoZDI_Q7EXGDJnIhrGkLXo_LJ_pAa2Jzbx17UHMd3jAs--6j2zaeto5w9RTn8T_1kKg3fdC5WPX9Dbb18vzH7YFX0eSJmoa6SP114rvlkw6pkS1-T&s=1', + 'info_dict': { + 'id': 'pO8h3EaFRdo', + 'ext': 'mp4', + 'title': 'Tripeo Boiler Room x Dekmantel Festival DJ Set', + 'description': 'md5:6294cc1af09c4049e0652b51a2df10d5', + 'upload_date': '20150917', + 'uploader_id': 'brtvofficial', + 'uploader': 'Boiler Room', + }, + 'params': { + 'skip_download': False, + }, + }, { 'url': 'http://www.hodiho.fr/2013/02/regis-plante-sa-jeep.html', 'md5': '85b90ccc9d73b4acd9138d3af4c27f89', @@ -173,7 +327,7 @@ class GenericIE(InfoExtractor): # it also tests brightcove videos that need to set the 'Referer' in the # http requests { - 'add_ie': ['Brightcove'], + 'add_ie': ['BrightcoveLegacy'], 'url': 'http://www.bfmtv.com/video/bfmbusiness/cours-bourse/cours-bourse-l-analyse-technique-154522/', 'info_dict': { 'id': '2765128793001', @@ -197,7 +351,7 @@ class GenericIE(InfoExtractor): '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'], + 'add_ie': ['BrightcoveLegacy'], }, { 'url': 'http://www.championat.com/video/football/v/87/87499.html', @@ -212,7 +366,7 @@ class GenericIE(InfoExtractor): }, { # https://github.com/rg3/youtube-dl/issues/3541 - 'add_ie': ['Brightcove'], + 'add_ie': ['BrightcoveLegacy'], 'url': 'http://www.kijk.nl/sbs6/leermijvrouwenkennen/videos/jqMiXKAYan2S/aflevering-1', 'info_dict': { 'id': '3866516442001', @@ -234,9 +388,24 @@ class GenericIE(InfoExtractor): 'id': 'BwY2RxaTrTkslxOfcan0UCf0YqyvWysJ', 'ext': 'mp4', 'title': '2cc213299525360.mov', # that's what we get + 'duration': 238.231, }, 'add_ie': ['Ooyala'], }, + { + # ooyala video embedded with http://player.ooyala.com/iframe.js + 'url': 'http://www.macrumors.com/2015/07/24/steve-jobs-the-man-in-the-machine-first-trailer/', + 'info_dict': { + 'id': 'p0MGJndjoG5SOKqO_hZJuZFPB-Tr5VgB', + 'ext': 'mp4', + 'title': '"Steve Jobs: Man in the Machine" trailer', + 'description': 'The first trailer for the Alex Gibney documentary "Steve Jobs: Man in the Machine."', + 'duration': 135.427, + }, + 'params': { + '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', @@ -277,14 +446,6 @@ class GenericIE(InfoExtractor): 'description': 'Episode 18: President Barack Obama sits down with Zach Galifianakis for his most memorable interview yet.', }, }, - # BBC iPlayer embeds - { - 'url': 'http://www.bbc.co.uk/blogs/adamcurtis/posts/BUGGER', - 'info_dict': { - 'title': 'BBC - Blogs - Adam Curtis - BUGGER', - }, - 'playlist_mincount': 18, - }, # RUTV embed { 'url': 'http://www.rg.ru/2014/03/15/reg-dfo/anklav-anons.html', @@ -371,7 +532,7 @@ 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', @@ -713,6 +874,19 @@ class GenericIE(InfoExtractor): 'title': 'Os Guinness // Is It Fools Talk? // Unbelievable? Conference 2014', }, }, + # Kaltura embed protected with referrer + { + 'url': 'http://www.disney.nl/disney-channel/filmpjes/achter-de-schermen#/videoId/violetta-achter-de-schermen-ruggero', + 'info_dict': { + 'id': '1_g4fbemnq', + 'ext': 'mp4', + 'title': 'Violetta - Achter De Schermen - Ruggero', + 'description': 'Achter de schermen met Ruggero', + 'timestamp': 1435133761, + 'upload_date': '20150624', + 'uploader_id': 'echojecka', + }, + }, # Eagle.Platform embed (generic URL) { 'url': 'http://lenta.ru/news/2015/03/06/navalny/', @@ -837,8 +1011,9 @@ class GenericIE(InfoExtractor): 'info_dict': { 'id': '50YnY4czr4ms1vJ7yz3xzq0excz_pUMs', 'ext': 'mp4', - 'description': 'VIDEO: Index/Match versus VLOOKUP.', + 'description': 'VIDEO: INDEX/MATCH versus VLOOKUP.', 'title': 'This is what separates the Excel masters from the wannabes', + 'duration': 191.933, }, 'params': { # m3u8 downloads @@ -914,7 +1089,58 @@ class GenericIE(InfoExtractor): 'description': 'New experience with Acrobat DC', 'duration': 248.667, }, - } + }, + # ScreenwaveMedia embed + { + 'url': 'http://www.thecinemasnob.com/the-cinema-snob/a-nightmare-on-elm-street-2-freddys-revenge1', + 'md5': '24ace5baba0d35d55c6810b51f34e9e0', + 'info_dict': { + 'id': 'cinemasnob-55d26273809dd', + 'ext': 'mp4', + 'title': 'cinemasnob', + }, + }, + # BrightcoveInPageEmbed embed + { + 'url': 'http://www.geekandsundry.com/tabletop-bonus-wils-final-thoughts-on-dread/', + 'info_dict': { + 'id': '4238694884001', + 'ext': 'flv', + 'title': 'Tabletop: Dread, Last Thoughts', + 'description': 'Tabletop: Dread, Last Thoughts', + 'duration': 51690, + }, + }, + # JWPlayer with M3U8 + { + 'url': 'http://ren.tv/novosti/2015-09-25/sluchaynyy-prohozhiy-poymal-avtougonshchika-v-murmanske-video', + 'info_dict': { + 'id': 'playlist', + 'ext': 'mp4', + 'title': 'Случайный прохожий поймал автоугонщика в Мурманске. ВИДЕО | РЕН ТВ', + 'uploader': 'ren.tv', + }, + 'params': { + # 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', + }, + 'params': { + # m3u8 downloads + 'skip_download': True, + }, + }, ] def report_following_redirect(self, new_url): @@ -1057,28 +1283,36 @@ class GenericIE(InfoExtractor): full_response = None if head_response is False: - request = compat_urllib_request.Request(url) + request = sanitized_Request(url) request.add_header('Accept-Encoding', '*') 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) @@ -1086,7 +1320,7 @@ class GenericIE(InfoExtractor): '%s on generic information extractor.' % ('Forcing' if force else 'Falling back')) if not full_response: - request = compat_urllib_request.Request(url) + request = sanitized_Request(url) # Some webservers may serve compressed content of rather big size (e.g. gzipped flac) # making it impossible to download only chunk of the file (yet we need only 512kB to # test whether it's HTML or not). According to youtube-dl default Accept-Encoding @@ -1098,32 +1332,50 @@ 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? + # Is it an RSS feed, a SMIL file, an XSPF playlist or a MPD manifest? try: - doc = parse_xml(webpage) + 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$', doc.tag): + smil = self._parse_smil(doc, url, video_id) + self._sort_formats(smil['formats']) + return smil + elif doc.tag == '{http://xspf.org/ns/0/}playlist': + return self.playlist_result(self._parse_xspf(doc, video_id), video_id) + elif re.match(r'(?i)^(?:{[^}]+})?MPD$', 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 @@ -1169,14 +1421,14 @@ class GenericIE(InfoExtractor): return self.playlist_result( urlrs, playlist_id=video_id, playlist_title=video_title) - # Look for BrightCove: - bc_urls = BrightcoveIE._extract_brightcove_urls(webpage) + # Look for Brightcove Legacy Studio embeds + bc_urls = BrightcoveLegacyIE._extract_brightcove_urls(webpage) if bc_urls: self.to_screen('Brightcove video detected.') entries = [{ '_type': 'url', 'url': smuggle_url(bc_url, {'Referer': url}), - 'ie_key': 'Brightcove' + 'ie_key': 'BrightcoveLegacy' } for bc_url in bc_urls] return { @@ -1186,6 +1438,11 @@ class GenericIE(InfoExtractor): 'entries': entries, } + # Look for Brightcove New Studio embeds + bc_urls = BrightcoveNewIE._extract_urls(webpage) + if bc_urls: + return _playlist_from_matches(bc_urls, ie='BrightcoveNew') + # Look for embedded rtl.nl player matches = re.findall( r']+?src="((?:https?:)?//(?:www\.)?rtl\.nl/system/videoplayer/[^"]+(?:video_)?embed[^"]+)"', @@ -1228,7 +1485,7 @@ 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])) @@ -1269,11 +1526,6 @@ class GenericIE(InfoExtractor): 'id': match.group('id') } - # Look for embedded blip.tv player - bliptv_url = BlipTVIE._extract_url(webpage) - if bliptv_url: - return self.url_result(bliptv_url, 'BlipTV') - # Look for SVT player svt_url = SVTIE._extract_url(webpage) if svt_url: @@ -1329,12 +1581,12 @@ class GenericIE(InfoExtractor): return self.url_result(mobj.group('url')) # Look for Ooyala videos - mobj = (re.search(r'player\.ooyala\.com/[^"?]+\?[^"]*?(?:embedCode|ec)=(?P[^"&]+)', webpage) or + mobj = (re.search(r'player\.ooyala\.com/[^"?]+[?#][^"]*?(?:embedCode|ec)=(?P[^"&]+)', webpage) or re.search(r'OO\.Player\.create\([\'"].*?[\'"],\s*[\'"](?P.{32})[\'"]', webpage) or re.search(r'SBN\.VideoLinkset\.ooyala\([\'"](?P.{32})[\'"]\)', webpage) or re.search(r'data-ooyala-video-id\s*=\s*[\'"](?P.{32})[\'"]', webpage)) if mobj is not None: - return OoyalaIE._build_url_result(mobj.group('ec')) + return OoyalaIE._build_url_result(smuggle_url(mobj.group('ec'), {'domain': url})) # Look for multiple Ooyala embeds on SBN network websites mobj = re.search(r'SBN\.VideoLinkset\.entryGroup\((\[.*?\])', webpage) @@ -1342,7 +1594,7 @@ class GenericIE(InfoExtractor): embeds = self._parse_json(mobj.group(1), video_id, fatal=False) if embeds: return _playlist_from_matches( - embeds, getter=lambda v: OoyalaIE._url_for_embed_code(v['provider_video_id']), ie='Ooyala') + embeds, getter=lambda v: OoyalaIE._url_for_embed_code(smuggle_url(v['provider_video_id'], {'domain': url})), ie='Ooyala') # Look for Aparat videos mobj = re.search(r'