diff options
Diffstat (limited to 'youtube_dl/extractor')
119 files changed, 4268 insertions, 1888 deletions
diff --git a/youtube_dl/extractor/__init__.py b/youtube_dl/extractor/__init__.py index fcd9edec3..08cb93d76 100644 --- a/youtube_dl/extractor/__init__.py +++ b/youtube_dl/extractor/__init__.py @@ -45,6 +45,7 @@ from .bambuser import BambuserIE, BambuserChannelIE from .bandcamp import BandcampIE, BandcampAlbumIE from .bbc import ( BBCCoUkIE, + BBCCoUkArticleIE, BBCIE, ) from .beeg import BeegIE @@ -79,6 +80,7 @@ from .cbssports import CBSSportsIE from .ccc import CCCIE from .ceskatelevize import CeskaTelevizeIE from .channel9 import Channel9IE +from .chaturbate import ChaturbateIE from .chilloutzone import ChilloutzoneIE from .chirbit import ( ChirbitIE, @@ -91,6 +93,7 @@ from .cliphunter import CliphunterIE from .clipsyndicate import ClipsyndicateIE from .cloudy import CloudyIE from .clubic import ClubicIE +from .clyp import ClypIE from .cmt import CMTIE from .cnet import CNETIE from .cnn import ( @@ -124,6 +127,7 @@ from .dbtv import DBTVIE from .dcn import DCNIE from .dctp import DctpTvIE from .deezer import DeezerPlaylistIE +from .democracynow import DemocracynowIE from .dfb import DFBIE from .dhm import DHMIE from .dotsub import DotsubIE @@ -141,7 +145,6 @@ from .dump import DumpIE from .dumpert import DumpertIE from .defense import DefenseGouvFrIE from .discovery import DiscoveryIE -from .divxstage import DivxStageIE from .dropbox import DropboxIE from .eagleplatform import EaglePlatformIE from .ebaumsworld import EbaumsWorldIE @@ -162,6 +165,7 @@ from .eroprofile import EroProfileIE from .escapist import EscapistIE from .espn import ESPNIE from .esri import EsriVideoIE +from .europa import EuropaIE from .everyonesmixtape import EveryonesMixtapeIE from .exfm import ExfmIE from .expotv import ExpoTVIE @@ -169,14 +173,12 @@ from .extremetube import ExtremeTubeIE from .facebook import FacebookIE from .faz import FazIE from .fc2 import FC2IE +from .fczenit import FczenitIE from .firstpost import FirstpostIE from .firsttv import FirstTVIE from .fivemin import FiveMinIE from .fivetv import FiveTVIE -from .fktv import ( - FKTVIE, - FKTVPosteckeIE, -) +from .fktv import FKTVIE from .flickr import FlickrIE from .folketinget import FolketingetIE from .footyroom import FootyRoomIE @@ -213,13 +215,15 @@ from .gfycat import GfycatIE from .giantbomb import GiantBombIE from .giga import GigaIE from .glide import GlideIE -from .globo import GloboIE +from .globo import ( + GloboIE, + GloboArticleIE, +) from .godtube import GodTubeIE from .goldenmoustache import GoldenMoustacheIE from .golem import GolemIE from .googleplus import GooglePlusIE from .googlesearch import GoogleSearchIE -from .gorillavid import GorillaVidIE from .goshgay import GoshgayIE from .groupon import GrouponIE from .hark import HarkIE @@ -232,7 +236,6 @@ from .historicfilms import HistoricFilmsIE from .history import HistoryIE from .hitbox import HitboxIE, HitboxLiveIE from .hornbunny import HornBunnyIE -from .hostingbulk import HostingBulkIE from .hotnewhiphop import HotNewHipHopIE from .howcast import HowcastIE from .howstuffworks import HowStuffWorksIE @@ -302,6 +305,11 @@ from .lifenews import ( LifeNewsIE, LifeEmbedIE, ) +from .limelight import ( + LimelightMediaIE, + LimelightChannelIE, + LimelightChannelListIE, +) from .liveleak import LiveLeakIE from .livestream import ( LivestreamIE, @@ -319,7 +327,6 @@ from .macgamestore import MacGameStoreIE from .mailru import MailRuIE from .malemotion import MalemotionIE from .mdr import MDRIE -from .megavideoz import MegaVideozIE from .metacafe import MetacafeIE from .metacritic import MetacriticIE from .mgoon import MgoonIE @@ -370,6 +377,9 @@ from .nbc import ( from .ndr import ( NDRIE, NJoyIE, + NDREmbedBaseIE, + NDREmbedIE, + NJoyEmbedIE, ) from .ndtv import NDTVIE from .netzkino import NetzkinoIE @@ -405,7 +415,11 @@ from .normalboots import NormalbootsIE from .nosvideo import NosVideoIE from .nova import NovaIE from .novamov import NovaMovIE -from .nowness import NownessIE +from .nowness import ( + NownessIE, + NownessPlaylistIE, + NownessSeriesIE, +) from .nowtv import NowTVIE from .nowvideo import NowVideoIE from .npo import ( @@ -435,7 +449,6 @@ from .ooyala import ( OoyalaIE, OoyalaExternalIE, ) -from .openfilm import OpenFilmIE from .orf import ( ORFTVthekIE, ORFOE1IE, @@ -581,6 +594,7 @@ from .spankwire import SpankwireIE from .spiegel import SpiegelIE, SpiegelArticleIE from .spiegeltv import SpiegeltvIE from .spike import SpikeIE +from .stitcher import StitcherIE from .sport5 import Sport5IE from .sportbox import ( SportBoxIE, @@ -685,7 +699,7 @@ from .twitch import ( TwitchBookmarksIE, TwitchStreamIE, ) -from .twitter import TwitterCardIE +from .twitter import TwitterCardIE, TwitterIE from .ubu import UbuIE from .udemy import ( UdemyIE, @@ -712,7 +726,6 @@ from .vh1 import VH1IE from .vice import ViceIE from .viddler import ViddlerIE from .videodetective import VideoDetectiveIE -from .videolecturesnet import VideoLecturesNetIE from .videofyme import VideofyMeIE from .videomega import VideoMegaIE from .videopremium import VideoPremiumIE @@ -722,6 +735,7 @@ from .vidme import VidmeIE from .vidzi import VidziIE from .vier import VierIE, VierVideosIE from .viewster import ViewsterIE +from .viidea import ViideaIE from .vimeo import ( VimeoIE, VimeoAlbumIE, @@ -774,6 +788,7 @@ from .wrzuta import WrzutaIE from .wsj import WSJIE from .xbef import XBefIE from .xboxclips import XboxClipsIE +from .xfileshare import XFileShareIE from .xhamster import ( XHamsterIE, XHamsterEmbedIE, diff --git a/youtube_dl/extractor/abc.py b/youtube_dl/extractor/abc.py index f9a389f67..c0e5d1abf 100644 --- a/youtube_dl/extractor/abc.py +++ b/youtube_dl/extractor/abc.py @@ -12,7 +12,7 @@ from ..utils import ( class ABCIE(InfoExtractor): IE_NAME = 'abc.net.au' - _VALID_URL = r'http://www\.abc\.net\.au/news/[^/]+/[^/]+/(?P<id>\d+)' + _VALID_URL = r'http://www\.abc\.net\.au/news/(?:[^/]+/){1,2}(?P<id>\d+)' _TESTS = [{ 'url': 'http://www.abc.net.au/news/2014-11-05/australia-to-staff-ebola-treatment-centre-in-sierra-leone/5868334', @@ -36,6 +36,18 @@ class ABCIE(InfoExtractor): 'title': 'Marriage Equality: Warren Entsch introduces same sex marriage bill', }, 'add_ie': ['Youtube'], + }, { + 'url': 'http://www.abc.net.au/news/2015-10-23/nab-lifts-interest-rates-following-westpac-and-cba/6880080', + 'md5': 'b96eee7c9edf4fc5a358a0252881cc1f', + 'info_dict': { + 'id': '6880080', + 'ext': 'mp3', + 'title': 'NAB lifts interest rates, following Westpac and CBA', + 'description': 'md5:f13d8edc81e462fce4a0437c7dc04728', + }, + }, { + 'url': 'http://www.abc.net.au/news/2015-10-19/6866214', + 'only_matching': True, }] def _real_extract(self, url): @@ -43,7 +55,7 @@ class ABCIE(InfoExtractor): webpage = self._download_webpage(url, video_id) mobj = re.search( - r'inline(?P<type>Video|YouTube)Data\.push\((?P<json_data>[^)]+)\);', + r'inline(?P<type>Video|Audio|YouTube)Data\.push\((?P<json_data>[^)]+)\);', webpage) if mobj is None: raise ExtractorError('Unable to extract video urls') @@ -60,11 +72,13 @@ class ABCIE(InfoExtractor): formats = [{ 'url': url_info['url'], + 'vcodec': url_info.get('codec') if mobj.group('type') == 'Video' else 'none', 'width': int_or_none(url_info.get('width')), 'height': int_or_none(url_info.get('height')), 'tbr': int_or_none(url_info.get('bitrate')), 'filesize': int_or_none(url_info.get('filesize')), } for url_info in urls_info] + self._sort_formats(formats) return { diff --git a/youtube_dl/extractor/adultswim.py b/youtube_dl/extractor/adultswim.py index 4327c2f61..3ae618e71 100644 --- a/youtube_dl/extractor/adultswim.py +++ b/youtube_dl/extractor/adultswim.py @@ -5,6 +5,7 @@ import re from .common import InfoExtractor from ..utils import ( + determine_ext, ExtractorError, float_or_none, xpath_text, @@ -40,7 +41,8 @@ class AdultSwimIE(InfoExtractor): 'id': 'rQxZvXQ4ROaSOqq-or2Mow', 'title': 'Rick and Morty - Pilot', 'description': "Rick moves in with his daughter's family and establishes himself as a bad influence on his grandson, Morty. " - } + }, + 'skip': 'This video is only available for registered users', }, { 'url': 'http://www.adultswim.com/videos/playlists/american-parenting/putting-francine-out-of-business/', 'playlist': [ @@ -123,7 +125,6 @@ class AdultSwimIE(InfoExtractor): else: collections = bootstrapped_data['show']['collections'] collection, video_info = self.find_collection_containing_video(collections, episode_path) - # Video wasn't found in the collections, let's try `slugged_video`. if video_info is None: if bootstrapped_data.get('slugged_video', {}).get('slug') == episode_path: @@ -133,7 +134,15 @@ class AdultSwimIE(InfoExtractor): show = bootstrapped_data['show'] show_title = show['title'] - segment_ids = [clip['videoPlaybackID'] for clip in video_info['clips']] + stream = video_info.get('stream') + clips = [stream] if stream else video_info.get('clips') + if not clips: + raise ExtractorError( + 'This video is only available via cable service provider subscription that' + ' is not currently supported. You may want to use --cookies.' + if video_info.get('auth') is True else 'Unable to find stream or clips', + expected=True) + segment_ids = [clip['videoPlaybackID'] for clip in clips] episode_id = video_info['id'] episode_title = video_info['title'] @@ -142,7 +151,7 @@ class AdultSwimIE(InfoExtractor): entries = [] for part_num, segment_id in enumerate(segment_ids): - segment_url = 'http://www.adultswim.com/videos/api/v0/assets?id=%s&platform=mobile' % segment_id + segment_url = 'http://www.adultswim.com/videos/api/v0/assets?id=%s&platform=desktop' % segment_id segment_title = '%s - %s' % (show_title, episode_title) if len(segment_ids) > 1: @@ -158,17 +167,30 @@ class AdultSwimIE(InfoExtractor): formats = [] file_els = idoc.findall('.//files/file') or idoc.findall('./files/file') + unique_urls = [] + unique_file_els = [] for file_el in file_els: + media_url = file_el.text + if not media_url or determine_ext(media_url) == 'f4m': + continue + if file_el.text not in unique_urls: + unique_urls.append(file_el.text) + unique_file_els.append(file_el) + + for file_el in unique_file_els: bitrate = file_el.attrib.get('bitrate') ftype = file_el.attrib.get('type') - - formats.append({ - 'format_id': '%s_%s' % (bitrate, ftype), - 'url': file_el.text.strip(), - # The bitrate may not be a number (for example: 'iphone') - 'tbr': int(bitrate) if bitrate.isdigit() else None, - 'quality': 1 if ftype == 'hd' else -1 - }) + media_url = file_el.text + if determine_ext(media_url) == 'm3u8': + formats.extend(self._extract_m3u8_formats( + media_url, segment_title, 'mp4', preference=0, m3u8_id='hls')) + else: + formats.append({ + 'format_id': '%s_%s' % (bitrate, ftype), + 'url': file_el.text.strip(), + # The bitrate may not be a number (for example: 'iphone') + 'tbr': int(bitrate) if bitrate.isdigit() else None, + }) self._sort_formats(formats) diff --git a/youtube_dl/extractor/anitube.py b/youtube_dl/extractor/anitube.py index 31f0d417c..23f942ae2 100644 --- a/youtube_dl/extractor/anitube.py +++ b/youtube_dl/extractor/anitube.py @@ -26,8 +26,8 @@ class AnitubeIE(InfoExtractor): video_id = mobj.group('id') webpage = self._download_webpage(url, video_id) - key = self._html_search_regex( - r'http://www\.anitube\.se/embed/([A-Za-z0-9_-]*)', webpage, 'key') + key = self._search_regex( + r'src=["\']https?://[^/]+/embed/([A-Za-z0-9_-]+)', webpage, 'key') config_xml = self._download_xml( 'http://www.anitube.se/nuevo/econfig.php?key=%s' % key, key) diff --git a/youtube_dl/extractor/appletrailers.py b/youtube_dl/extractor/appletrailers.py index 576f03b5b..f68dc3236 100644 --- a/youtube_dl/extractor/appletrailers.py +++ b/youtube_dl/extractor/appletrailers.py @@ -13,53 +13,53 @@ from ..utils import ( class AppleTrailersIE(InfoExtractor): _VALID_URL = r'https?://(?:www\.)?trailers\.apple\.com/(?:trailers|ca)/(?P<company>[^/]+)/(?P<movie>[^/]+)' _TESTS = [{ - "url": "http://trailers.apple.com/trailers/wb/manofsteel/", + 'url': 'http://trailers.apple.com/trailers/wb/manofsteel/', 'info_dict': { 'id': 'manofsteel', }, - "playlist": [ + 'playlist': [ { - "md5": "d97a8e575432dbcb81b7c3acb741f8a8", - "info_dict": { - "id": "manofsteel-trailer4", - "ext": "mov", - "duration": 111, - "title": "Trailer 4", - "upload_date": "20130523", - "uploader_id": "wb", + 'md5': 'd97a8e575432dbcb81b7c3acb741f8a8', + 'info_dict': { + 'id': 'manofsteel-trailer4', + 'ext': 'mov', + 'duration': 111, + 'title': 'Trailer 4', + 'upload_date': '20130523', + 'uploader_id': 'wb', }, }, { - "md5": "b8017b7131b721fb4e8d6f49e1df908c", - "info_dict": { - "id": "manofsteel-trailer3", - "ext": "mov", - "duration": 182, - "title": "Trailer 3", - "upload_date": "20130417", - "uploader_id": "wb", + 'md5': 'b8017b7131b721fb4e8d6f49e1df908c', + 'info_dict': { + 'id': 'manofsteel-trailer3', + 'ext': 'mov', + 'duration': 182, + 'title': 'Trailer 3', + 'upload_date': '20130417', + 'uploader_id': 'wb', }, }, { - "md5": "d0f1e1150989b9924679b441f3404d48", - "info_dict": { - "id": "manofsteel-trailer", - "ext": "mov", - "duration": 148, - "title": "Trailer", - "upload_date": "20121212", - "uploader_id": "wb", + 'md5': 'd0f1e1150989b9924679b441f3404d48', + 'info_dict': { + 'id': 'manofsteel-trailer', + 'ext': 'mov', + 'duration': 148, + 'title': 'Trailer', + 'upload_date': '20121212', + 'uploader_id': 'wb', }, }, { - "md5": "5fe08795b943eb2e757fa95cb6def1cb", - "info_dict": { - "id": "manofsteel-teaser", - "ext": "mov", - "duration": 93, - "title": "Teaser", - "upload_date": "20120721", - "uploader_id": "wb", + 'md5': '5fe08795b943eb2e757fa95cb6def1cb', + 'info_dict': { + 'id': 'manofsteel-teaser', + 'ext': 'mov', + 'duration': 93, + 'title': 'Teaser', + 'upload_date': '20120721', + 'uploader_id': 'wb', }, }, ] diff --git a/youtube_dl/extractor/ard.py b/youtube_dl/extractor/ard.py index 6f465789b..73be6d204 100644 --- a/youtube_dl/extractor/ard.py +++ b/youtube_dl/extractor/ard.py @@ -14,8 +14,8 @@ from ..utils import ( parse_duration, unified_strdate, xpath_text, - parse_xml, ) +from ..compat import compat_etree_fromstring class ARDMediathekIE(InfoExtractor): @@ -161,7 +161,7 @@ class ARDMediathekIE(InfoExtractor): raise ExtractorError('This program is only suitable for those aged 12 and older. Video %s is therefore only available between 20 pm and 6 am.' % video_id, expected=True) if re.search(r'[\?&]rss($|[=&])', url): - doc = parse_xml(webpage) + doc = compat_etree_fromstring(webpage.encode('utf-8')) if doc.tag == 'rss': return GenericIE()._extract_rss(url, video_id, doc) diff --git a/youtube_dl/extractor/arte.py b/youtube_dl/extractor/arte.py index 76de24477..2a00da3ee 100644 --- a/youtube_dl/extractor/arte.py +++ b/youtube_dl/extractor/arte.py @@ -4,6 +4,10 @@ from __future__ import unicode_literals import re from .common import InfoExtractor +from ..compat import ( + compat_parse_qs, + compat_urllib_parse_urlparse, +) from ..utils import ( find_xpath_attr, unified_strdate, @@ -77,7 +81,13 @@ class ArteTVPlus7IE(InfoExtractor): def _extract_from_webpage(self, webpage, video_id, lang): json_url = self._html_search_regex( [r'arte_vp_url=["\'](.*?)["\']', r'data-url=["\']([^"]+)["\']'], - webpage, 'json vp url') + webpage, 'json vp url', default=None) + if not json_url: + iframe_url = self._html_search_regex( + r'<iframe[^>]+src=(["\'])(?P<url>.+\bjson_url=.+?)\1', + webpage, 'iframe url', group='url') + json_url = compat_parse_qs( + compat_urllib_parse_urlparse(iframe_url).query)['json_url'][0] return self._extract_from_json_url(json_url, video_id, lang) def _extract_from_json_url(self, json_url, video_id, lang): diff --git a/youtube_dl/extractor/bandcamp.py b/youtube_dl/extractor/bandcamp.py index 505877b77..c1ef8051d 100644 --- a/youtube_dl/extractor/bandcamp.py +++ b/youtube_dl/extractor/bandcamp.py @@ -10,6 +10,8 @@ from ..compat import ( ) from ..utils import ( ExtractorError, + float_or_none, + int_or_none, ) @@ -52,11 +54,11 @@ class BandcampIE(InfoExtractor): ext, abr_str = format_id.split('-', 1) formats.append({ 'format_id': format_id, - 'url': format_url, + 'url': self._proto_relative_url(format_url, 'http:'), 'ext': ext, 'vcodec': 'none', 'acodec': ext, - 'abr': int(abr_str), + 'abr': int_or_none(abr_str), }) self._sort_formats(formats) @@ -65,7 +67,7 @@ class BandcampIE(InfoExtractor): 'id': compat_str(data['id']), 'title': data['title'], 'formats': formats, - 'duration': float(data['duration']), + 'duration': float_or_none(data.get('duration')), } else: raise ExtractorError('No free songs found') @@ -93,8 +95,8 @@ class BandcampIE(InfoExtractor): final_url_webpage = self._download_webpage(request_url, video_id, 'Requesting download url') # If we could correctly generate the .rand field the url would be # in the "download_url" key - final_url = self._search_regex( - r'"retry_url":"(.*?)"', final_url_webpage, 'final video URL') + final_url = self._proto_relative_url(self._search_regex( + r'"retry_url":"(.+?)"', final_url_webpage, 'final video URL'), 'http:') return { 'id': video_id, diff --git a/youtube_dl/extractor/bbc.py b/youtube_dl/extractor/bbc.py index abc5a44a1..a55a6dbc9 100644 --- a/youtube_dl/extractor/bbc.py +++ b/youtube_dl/extractor/bbc.py @@ -2,7 +2,6 @@ from __future__ import unicode_literals import re -import xml.etree.ElementTree from .common import InfoExtractor from ..utils import ( @@ -11,19 +10,36 @@ from ..utils import ( int_or_none, parse_duration, parse_iso8601, + remove_end, + unescapeHTML, +) +from ..compat import ( + compat_etree_fromstring, + compat_HTTPError, ) -from ..compat import compat_HTTPError class BBCCoUkIE(InfoExtractor): IE_NAME = 'bbc.co.uk' IE_DESC = 'BBC iPlayer' - _VALID_URL = r'https?://(?:www\.)?bbc\.co\.uk/(?:(?:(?:programmes|iplayer(?:/[^/]+)?/(?:episode|playlist))/)|music/clips[/#])(?P<id>[\da-z]{8})' + _VALID_URL = r'https?://(?:www\.)?bbc\.co\.uk/(?:(?:programmes/(?!articles/)|iplayer(?:/[^/]+)?/(?:episode/|playlist/))|music/clips[/#])(?P<id>[\da-z]{8})' _MEDIASELECTOR_URLS = [ + # Provides HQ HLS streams with even better quality that pc mediaset but fails + # with geolocation in some cases when it's even not geo restricted at all (e.g. + # http://www.bbc.co.uk/programmes/b06bp7lf) + 'http://open.live.bbc.co.uk/mediaselector/5/select/version/2.0/mediaset/iptv-all/vpid/%s', 'http://open.live.bbc.co.uk/mediaselector/5/select/version/2.0/mediaset/pc/vpid/%s', ] + _MEDIASELECTION_NS = 'http://bbc.co.uk/2008/mp/mediaselection' + _EMP_PLAYLIST_NS = 'http://bbc.co.uk/2008/emp/playlist' + + _NAMESPACES = ( + _MEDIASELECTION_NS, + _EMP_PLAYLIST_NS, + ) + _TESTS = [ { 'url': 'http://www.bbc.co.uk/programmes/b039g8p7', @@ -153,6 +169,21 @@ class BBCCoUkIE(InfoExtractor): }, 'skip': 'geolocation', }, { + # iptv-all mediaset fails with geolocation however there is no geo restriction + # for this programme at all + 'url': 'http://www.bbc.co.uk/programmes/b06bp7lf', + 'info_dict': { + 'id': 'b06bp7kf', + 'ext': 'flv', + 'title': "Annie Mac's Friday Night, B.Traits sits in for Annie", + 'description': 'B.Traits sits in for Annie Mac with a Mini-Mix from Disclosure.', + 'duration': 10800, + }, + 'params': { + # rtmp download + 'skip_download': True, + }, + }, { 'url': 'http://www.bbc.co.uk/iplayer/playlist/p01dvks4', 'only_matching': True, }, { @@ -174,6 +205,7 @@ class BBCCoUkIE(InfoExtractor): def _extract_connection(self, connection, programme_id): formats = [] + kind = connection.get('kind') protocol = connection.get('protocol') supplier = connection.get('supplier') if protocol == 'http': @@ -189,11 +221,17 @@ class BBCCoUkIE(InfoExtractor): # Skip DASH until supported elif transfer_format == 'dash': pass + elif transfer_format == 'hls': + m3u8_formats = self._extract_m3u8_formats( + href, programme_id, ext='mp4', entry_protocol='m3u8_native', + m3u8_id=supplier, fatal=False) + if m3u8_formats: + formats.extend(m3u8_formats) # Direct link else: formats.append({ 'url': href, - 'format_id': supplier, + 'format_id': supplier or kind or protocol, }) elif protocol == 'rtmp': application = connection.get('application', 'ondemand') @@ -213,16 +251,24 @@ class BBCCoUkIE(InfoExtractor): return formats def _extract_items(self, playlist): - return playlist.findall('./{http://bbc.co.uk/2008/emp/playlist}item') + return playlist.findall('./{%s}item' % self._EMP_PLAYLIST_NS) + + def _findall_ns(self, element, xpath): + elements = [] + for ns in self._NAMESPACES: + elements.extend(element.findall(xpath % ns)) + return elements def _extract_medias(self, media_selection): - error = media_selection.find('./{http://bbc.co.uk/2008/mp/mediaselection}error') + error = media_selection.find('./{%s}error' % self._MEDIASELECTION_NS) + if error is None: + media_selection.find('./{%s}error' % self._EMP_PLAYLIST_NS) if error is not None: raise BBCCoUkIE.MediaSelectionError(error.get('id')) - return media_selection.findall('./{http://bbc.co.uk/2008/mp/mediaselection}media') + return self._findall_ns(media_selection, './{%s}media') def _extract_connections(self, media): - return media.findall('./{http://bbc.co.uk/2008/mp/mediaselection}connection') + return self._findall_ns(media, './{%s}connection') def _extract_video(self, media, programme_id): formats = [] @@ -236,13 +282,14 @@ class BBCCoUkIE(InfoExtractor): conn_formats = self._extract_connection(connection, programme_id) for format in conn_formats: format.update({ - 'format_id': '%s_%s' % (service, format['format_id']), 'width': width, 'height': height, 'vbr': vbr, 'vcodec': vcodec, 'filesize': file_size, }) + if service: + format['format_id'] = '%s_%s' % (service, format['format_id']) formats.extend(conn_formats) return formats @@ -287,7 +334,7 @@ class BBCCoUkIE(InfoExtractor): return self._download_media_selector_url( mediaselector_url % programme_id, programme_id) except BBCCoUkIE.MediaSelectionError as e: - if e.id == 'notukerror': + if e.id in ('notukerror', 'geolocation'): last_exception = e continue self._raise_extractor_error(e) @@ -299,7 +346,7 @@ class BBCCoUkIE(InfoExtractor): url, programme_id, 'Downloading media selection XML') except ExtractorError as ee: if isinstance(ee.cause, compat_HTTPError) and ee.cause.code == 403: - media_selection = xml.etree.ElementTree.fromstring(ee.cause.read().decode('utf-8')) + media_selection = compat_etree_fromstring(ee.cause.read().decode('utf-8')) else: raise return self._process_media_selector(media_selection, programme_id) @@ -357,7 +404,7 @@ class BBCCoUkIE(InfoExtractor): url, playlist_id, 'Downloading legacy playlist XML') def _extract_from_legacy_playlist(self, playlist, playlist_id): - no_items = playlist.find('./{http://bbc.co.uk/2008/emp/playlist}noItems') + no_items = playlist.find('./{%s}noItems' % self._EMP_PLAYLIST_NS) if no_items is not None: reason = no_items.get('reason') if reason == 'preAvailability': @@ -374,8 +421,9 @@ class BBCCoUkIE(InfoExtractor): kind = item.get('kind') if kind != 'programme' and kind != 'radioProgramme': continue - title = playlist.find('./{http://bbc.co.uk/2008/emp/playlist}title').text - description = playlist.find('./{http://bbc.co.uk/2008/emp/playlist}summary').text + title = playlist.find('./{%s}title' % self._EMP_PLAYLIST_NS).text + description_el = playlist.find('./{%s}summary' % self._EMP_PLAYLIST_NS) + description = description_el.text if description_el is not None else None def get_programme_id(item): def get_from_attributes(item): @@ -384,16 +432,18 @@ class BBCCoUkIE(InfoExtractor): if value and re.match(r'^[pb][\da-z]{7}$', value): return value get_from_attributes(item) - mediator = item.find('./{http://bbc.co.uk/2008/emp/playlist}mediator') + mediator = item.find('./{%s}mediator' % self._EMP_PLAYLIST_NS) if mediator is not None: return get_from_attributes(mediator) programme_id = get_programme_id(item) duration = int_or_none(item.get('duration')) - # TODO: programme_id can be None and media items can be incorporated right inside - # playlist's item (e.g. http://www.bbc.com/turkce/haberler/2015/06/150615_telabyad_kentin_cogu) - # as f4m and m3u8 - formats, subtitles = self._download_media_selector(programme_id) + + if programme_id: + formats, subtitles = self._download_media_selector(programme_id) + else: + formats, subtitles = self._process_media_selector(item, playlist_id) + programme_id = playlist_id return programme_id, title, description, duration, formats, subtitles @@ -445,6 +495,9 @@ class BBCIE(BBCCoUkIE): _VALID_URL = r'https?://(?:www\.)?bbc\.(?:com|co\.uk)/(?:[^/]+/)+(?P<id>[^/#?]+)' _MEDIASELECTOR_URLS = [ + # Provides HQ HLS streams but fails with geolocation in some cases when it's + # even not geo restricted at all + 'http://open.live.bbc.co.uk/mediaselector/5/select/version/2.0/mediaset/iptv-all/vpid/%s', # Provides more formats, namely direct mp4 links, but fails on some videos with # notukerror for non UK (?) users (e.g. # http://www.bbc.com/travel/story/20150625-sri-lankas-spicy-secret) @@ -454,8 +507,7 @@ class BBCIE(BBCCoUkIE): ] _TESTS = [{ - # article with multiple videos embedded with data-media-meta containing - # playlist.sxml, externalId and no direct video links + # article with multiple videos embedded with data-playable containing vpids 'url': 'http://www.bbc.com/news/world-europe-32668511', 'info_dict': { 'id': 'world-europe-32668511', @@ -464,7 +516,7 @@ class BBCIE(BBCCoUkIE): }, 'playlist_count': 2, }, { - # article with multiple videos embedded with data-media-meta (more videos) + # article with multiple videos embedded with data-playable (more videos) 'url': 'http://www.bbc.com/news/business-28299555', 'info_dict': { 'id': 'business-28299555', @@ -475,6 +527,7 @@ class BBCIE(BBCCoUkIE): 'skip': 'Save time', }, { # article with multiple videos embedded with `new SMP()` + # broken 'url': 'http://www.bbc.co.uk/blogs/adamcurtis/entries/3662a707-0af9-3149-963f-47bea720b460', 'info_dict': { 'id': '3662a707-0af9-3149-963f-47bea720b460', @@ -482,12 +535,13 @@ class BBCIE(BBCCoUkIE): }, 'playlist_count': 18, }, { - # single video embedded with mediaAssetPage.init() + # single video embedded with data-playable containing vpid 'url': 'http://www.bbc.com/news/world-europe-32041533', 'info_dict': { 'id': 'p02mprgb', 'ext': 'mp4', 'title': 'Aerial footage showed the site of the crash in the Alps - courtesy BFM TV', + 'description': 'md5:2868290467291b37feda7863f7a83f54', 'duration': 47, 'timestamp': 1427219242, 'upload_date': '20150324', @@ -497,15 +551,14 @@ class BBCIE(BBCCoUkIE): 'skip_download': True, } }, { - # article with single video embedded with data-media-meta containing - # direct video links (for now these are extracted) and playlist.xml (with - # media items as f4m and m3u8 - currently unsupported) + # article with single video embedded with data-playable containing XML playlist + # with direct video links as progressiveDownloadUrl (for now these are extracted) + # and playlist with f4m and m3u8 as streamingUrl 'url': 'http://www.bbc.com/turkce/haberler/2015/06/150615_telabyad_kentin_cogu', 'info_dict': { 'id': '150615_telabyad_kentin_cogu', 'ext': 'mp4', 'title': "YPG: Tel Abyad'ın tamamı kontrolümüzde", - 'duration': 47, 'timestamp': 1434397334, 'upload_date': '20150615', }, @@ -513,13 +566,12 @@ class BBCIE(BBCCoUkIE): 'skip_download': True, } }, { - # single video embedded with mediaAssetPage.init() (regional section) + # single video embedded with data-playable containing XML playlists (regional section) 'url': 'http://www.bbc.com/mundo/video_fotos/2015/06/150619_video_honduras_militares_hospitales_corrupcion_aw', 'info_dict': { 'id': '150619_video_honduras_militares_hospitales_corrupcion_aw', 'ext': 'mp4', 'title': 'Honduras militariza sus hospitales por nuevo escándalo de corrupción', - 'duration': 87, 'timestamp': 1434713142, 'upload_date': '20150619', }, @@ -561,21 +613,21 @@ class BBCIE(BBCCoUkIE): 'ext': 'mp4', 'title': 'Hyundai Santa Fe Sport: Rock star', 'description': 'md5:b042a26142c4154a6e472933cf20793d', - 'timestamp': 1368473503, - 'upload_date': '20130513', + 'timestamp': 1415867444, + 'upload_date': '20141113', }, 'params': { # rtmp download 'skip_download': True, } }, { - # single video with playlist.sxml URL + # single video with playlist.sxml URL in playlist param 'url': 'http://www.bbc.com/sport/0/football/33653409', 'info_dict': { 'id': 'p02xycnp', 'ext': 'mp4', 'title': 'Transfers: Cristiano Ronaldo to Man Utd, Arsenal to spend?', - 'description': 'md5:398fca0e2e701c609d726e034fa1fc89', + 'description': 'BBC Sport\'s David Ornstein has the latest transfer gossip, including rumours of a Manchester United return for Cristiano Ronaldo.', 'duration': 140, }, 'params': { @@ -583,6 +635,14 @@ class BBCIE(BBCCoUkIE): 'skip_download': True, } }, { + # article with multiple videos embedded with playlist.sxml in playlist param + 'url': 'http://www.bbc.com/sport/0/football/34475836', + 'info_dict': { + 'id': '34475836', + 'title': 'What Liverpool can expect from Klopp', + }, + 'playlist_count': 3, + }, { # single video with playlist URL from weather section 'url': 'http://www.bbc.com/weather/features/33601775', 'only_matching': True, @@ -594,7 +654,7 @@ class BBCIE(BBCCoUkIE): @classmethod def suitable(cls, url): - return False if BBCCoUkIE.suitable(url) else super(BBCIE, cls).suitable(url) + return False if BBCCoUkIE.suitable(url) or BBCCoUkArticleIE.suitable(url) else super(BBCIE, cls).suitable(url) def _extract_from_media_meta(self, media_meta, video_id): # Direct links to media in media metadata (e.g. @@ -623,40 +683,107 @@ class BBCIE(BBCCoUkIE): return [], [] + def _extract_from_playlist_sxml(self, url, playlist_id, timestamp): + programme_id, title, description, duration, formats, subtitles = \ + self._process_legacy_playlist_url(url, playlist_id) + self._sort_formats(formats) + return { + 'id': programme_id, + 'title': title, + 'description': description, + 'duration': duration, + 'timestamp': timestamp, + 'formats': formats, + 'subtitles': subtitles, + } + def _real_extract(self, url): playlist_id = self._match_id(url) webpage = self._download_webpage(url, playlist_id) - timestamp = parse_iso8601(self._search_regex( - [r'"datePublished":\s*"([^"]+)', - r'<meta[^>]+property="article:published_time"[^>]+content="([^"]+)"', - r'itemprop="datePublished"[^>]+datetime="([^"]+)"'], - webpage, 'date', default=None)) - - # single video with playlist.sxml URL (e.g. http://www.bbc.com/sport/0/football/3365340ng) - playlist = self._search_regex( - r'<param[^>]+name="playlist"[^>]+value="([^"]+)"', - webpage, 'playlist', default=None) - if playlist: - programme_id, title, description, duration, formats, subtitles = \ - self._process_legacy_playlist_url(playlist, playlist_id) - self._sort_formats(formats) - return { - 'id': programme_id, - 'title': title, - 'description': description, - 'duration': duration, - 'timestamp': timestamp, - 'formats': formats, - 'subtitles': subtitles, - } + timestamp = None + playlist_title = None + playlist_description = None + + ld = self._parse_json( + self._search_regex( + r'(?s)<script type="application/ld\+json">(.+?)</script>', + webpage, 'ld json', default='{}'), + playlist_id, fatal=False) + if ld: + timestamp = parse_iso8601(ld.get('datePublished')) + playlist_title = ld.get('headline') + playlist_description = ld.get('articleBody') + + if not timestamp: + timestamp = parse_iso8601(self._search_regex( + [r'<meta[^>]+property="article:published_time"[^>]+content="([^"]+)"', + r'itemprop="datePublished"[^>]+datetime="([^"]+)"', + r'"datePublished":\s*"([^"]+)'], + webpage, 'date', default=None)) + + entries = [] + + # article with multiple videos embedded with playlist.sxml (e.g. + # http://www.bbc.com/sport/0/football/34475836) + playlists = re.findall(r'<param[^>]+name="playlist"[^>]+value="([^"]+)"', webpage) + if playlists: + entries = [ + self._extract_from_playlist_sxml(playlist_url, playlist_id, timestamp) + for playlist_url in playlists] + + # news article with multiple videos embedded with data-playable + data_playables = re.findall(r'data-playable=(["\'])({.+?})\1', webpage) + if data_playables: + for _, data_playable_json in data_playables: + data_playable = self._parse_json( + unescapeHTML(data_playable_json), playlist_id, fatal=False) + if not data_playable: + continue + settings = data_playable.get('settings', {}) + if settings: + # data-playable with video vpid in settings.playlistObject.items (e.g. + # http://www.bbc.com/news/world-us-canada-34473351) + playlist_object = settings.get('playlistObject', {}) + if playlist_object: + items = playlist_object.get('items') + if items and isinstance(items, list): + title = playlist_object['title'] + description = playlist_object.get('summary') + duration = int_or_none(items[0].get('duration')) + programme_id = items[0].get('vpid') + formats, subtitles = self._download_media_selector(programme_id) + self._sort_formats(formats) + entries.append({ + 'id': programme_id, + 'title': title, + 'description': description, + 'timestamp': timestamp, + 'duration': duration, + 'formats': formats, + 'subtitles': subtitles, + }) + else: + # data-playable without vpid but with a playlist.sxml URLs + # in otherSettings.playlist (e.g. + # http://www.bbc.com/turkce/multimedya/2015/10/151010_vid_ankara_patlama_ani) + playlist = data_playable.get('otherSettings', {}).get('playlist', {}) + if playlist: + entries.append(self._extract_from_playlist_sxml( + playlist.get('progressiveDownloadUrl'), playlist_id, timestamp)) + + if entries: + playlist_title = playlist_title or remove_end(self._og_search_title(webpage), ' - BBC News') + playlist_description = playlist_description or self._og_search_description(webpage, default=None) + return self.playlist_result(entries, playlist_id, playlist_title, playlist_description) # single video story (e.g. http://www.bbc.com/travel/story/20150625-sri-lankas-spicy-secret) programme_id = self._search_regex( [r'data-video-player-vpid="([\da-z]{8})"', r'<param[^>]+name="externalIdentifier"[^>]+value="([\da-z]{8})"'], webpage, 'vpid', default=None) + if programme_id: formats, subtitles = self._download_media_selector(programme_id) self._sort_formats(formats) @@ -778,3 +905,33 @@ class BBCIE(BBCCoUkIE): }) return self.playlist_result(entries, playlist_id, playlist_title, playlist_description) + + +class BBCCoUkArticleIE(InfoExtractor): + _VALID_URL = 'http://www.bbc.co.uk/programmes/articles/(?P<id>[a-zA-Z0-9]+)' + IE_NAME = 'bbc.co.uk:article' + IE_DESC = 'BBC articles' + + _TEST = { + 'url': 'http://www.bbc.co.uk/programmes/articles/3jNQLTMrPlYGTBn0WV6M2MS/not-your-typical-role-model-ada-lovelace-the-19th-century-programmer', + 'info_dict': { + 'id': '3jNQLTMrPlYGTBn0WV6M2MS', + 'title': 'Calculating Ada: The Countess of Computing - Not your typical role model: Ada Lovelace the 19th century programmer - BBC Four', + 'description': 'Hannah Fry reveals some of her surprising discoveries about Ada Lovelace during filming.', + }, + 'playlist_count': 4, + 'add_ie': ['BBCCoUk'], + } + + def _real_extract(self, url): + playlist_id = self._match_id(url) + + webpage = self._download_webpage(url, playlist_id) + + title = self._og_search_title(webpage) + description = self._og_search_description(webpage).strip() + + entries = [self.url_result(programme_url) for programme_url in re.findall( + r'<div[^>]+typeof="Clip"[^>]+resource="([^"]+)"', webpage)] + + return self.playlist_result(entries, playlist_id, title, description) diff --git a/youtube_dl/extractor/beeg.py b/youtube_dl/extractor/beeg.py index b38057f2f..61bc2f744 100644 --- a/youtube_dl/extractor/beeg.py +++ b/youtube_dl/extractor/beeg.py @@ -1,65 +1,69 @@ from __future__ import unicode_literals -import re - from .common import InfoExtractor +from ..utils import ( + int_or_none, + parse_iso8601, +) class BeegIE(InfoExtractor): _VALID_URL = r'https?://(?:www\.)?beeg\.com/(?P<id>\d+)' _TEST = { 'url': 'http://beeg.com/5416503', - 'md5': '1bff67111adb785c51d1b42959ec10e5', + 'md5': '46c384def73b33dbc581262e5ee67cef', 'info_dict': { 'id': '5416503', 'ext': 'mp4', 'title': 'Sultry Striptease', - 'description': 'md5:6db3c6177972822aaba18652ff59c773', - 'categories': list, # NSFW - 'thumbnail': 're:https?://.*\.jpg$', + 'description': 'md5:d22219c09da287c14bed3d6c37ce4bc2', + 'timestamp': 1391813355, + 'upload_date': '20140207', + 'duration': 383, + 'tags': list, 'age_limit': 18, } } def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - video_id = mobj.group('id') - - webpage = self._download_webpage(url, video_id) - - quality_arr = self._search_regex( - r'(?s)var\s+qualityArr\s*=\s*{\s*(.+?)\s*}', webpage, 'quality formats') + video_id = self._match_id(url) - formats = [{ - 'url': fmt[1], - 'format_id': fmt[0], - 'height': int(fmt[0][:-1]), - } for fmt in re.findall(r"'([^']+)'\s*:\s*'([^']+)'", quality_arr)] + video = self._download_json( + 'http://beeg.com/api/v1/video/%s' % video_id, video_id) + formats = [] + for format_id, video_url in video.items(): + if not video_url: + continue + height = self._search_regex( + r'^(\d+)[pP]$', format_id, 'height', default=None) + if not height: + continue + formats.append({ + 'url': self._proto_relative_url(video_url.replace('{DATA_MARKERS}', ''), 'http:'), + 'format_id': format_id, + 'height': int(height), + }) self._sort_formats(formats) - title = self._html_search_regex( - r'<title>([^<]+)\s*-\s*beeg\.?</title>', webpage, 'title') + title = video['title'] + video_id = video.get('id') or video_id + display_id = video.get('code') + description = video.get('desc') - description = self._html_search_regex( - r'<meta name="description" content="([^"]*)"', - webpage, 'description', fatal=False) - thumbnail = self._html_search_regex( - r'\'previewer.url\'\s*:\s*"([^"]*)"', - webpage, 'thumbnail', fatal=False) + timestamp = parse_iso8601(video.get('date'), ' ') + duration = int_or_none(video.get('duration')) - categories_str = self._html_search_regex( - r'<meta name="keywords" content="([^"]+)"', webpage, 'categories', fatal=False) - categories = ( - None if categories_str is None - else categories_str.split(',')) + tags = [tag.strip() for tag in video['tags'].split(',')] if video.get('tags') else None return { 'id': video_id, + 'display_id': display_id, 'title': title, 'description': description, - 'thumbnail': thumbnail, - 'categories': categories, + 'timestamp': timestamp, + 'duration': duration, + 'tags': tags, 'formats': formats, 'age_limit': 18, } diff --git a/youtube_dl/extractor/bild.py b/youtube_dl/extractor/bild.py index 4d8cce1ef..1a0184861 100644 --- a/youtube_dl/extractor/bild.py +++ b/youtube_dl/extractor/bild.py @@ -4,7 +4,7 @@ from __future__ import unicode_literals from .common import InfoExtractor from ..utils import ( int_or_none, - fix_xml_ampersands, + unescapeHTML, ) @@ -17,26 +17,24 @@ class BildIE(InfoExtractor): 'info_dict': { 'id': '38184146', 'ext': 'mp4', - 'title': 'BILD hat sie getestet', + 'title': 'Das können die neuen iPads', + 'description': 'md5:a4058c4fa2a804ab59c00d7244bbf62f', 'thumbnail': 're:^https?://.*\.jpg$', 'duration': 196, - 'description': 'Mit dem iPad Air 2 und dem iPad Mini 3 hat Apple zwei neue Tablet-Modelle präsentiert. BILD-Reporter Sven Stein durfte die Geräte bereits testen. ', } } def _real_extract(self, url): video_id = self._match_id(url) - xml_url = url.split(".bild.html")[0] + ",view=xml.bild.xml" - doc = self._download_xml(xml_url, video_id, transform_source=fix_xml_ampersands) - - duration = int_or_none(doc.attrib.get('duration'), scale=1000) + video_data = self._download_json( + url.split('.bild.html')[0] + ',view=json.bild.html', video_id) return { 'id': video_id, - 'title': doc.attrib['ueberschrift'], - 'description': doc.attrib.get('text'), - 'url': doc.attrib['src'], - 'thumbnail': doc.attrib.get('img'), - 'duration': duration, + 'title': unescapeHTML(video_data['title']).strip(), + 'description': unescapeHTML(video_data.get('description')), + 'url': video_data['clipList'][0]['srces'][0]['src'], + 'thumbnail': video_data.get('poster'), + 'duration': int_or_none(video_data.get('durationSec')), } diff --git a/youtube_dl/extractor/bilibili.py b/youtube_dl/extractor/bilibili.py index ecc17ebeb..6c66a1236 100644 --- a/youtube_dl/extractor/bilibili.py +++ b/youtube_dl/extractor/bilibili.py @@ -4,9 +4,11 @@ from __future__ import unicode_literals import re import itertools import json -import xml.etree.ElementTree as ET from .common import InfoExtractor +from ..compat import ( + compat_etree_fromstring, +) from ..utils import ( int_or_none, unified_strdate, @@ -88,7 +90,7 @@ class BiliBiliIE(InfoExtractor): except ValueError: pass - lq_doc = ET.fromstring(lq_page) + lq_doc = compat_etree_fromstring(lq_page) lq_durls = lq_doc.findall('./durl') hq_doc = self._download_xml( diff --git a/youtube_dl/extractor/brightcove.py b/youtube_dl/extractor/brightcove.py index c6ad1d065..2c7d968a8 100644 --- a/youtube_dl/extractor/brightcove.py +++ b/youtube_dl/extractor/brightcove.py @@ -3,10 +3,10 @@ from __future__ import unicode_literals import re import json -import xml.etree.ElementTree from .common import InfoExtractor from ..compat import ( + compat_etree_fromstring, compat_parse_qs, compat_str, compat_urllib_parse, @@ -123,7 +123,7 @@ class BrightcoveIE(InfoExtractor): object_str = fix_xml_ampersands(object_str) try: - object_doc = xml.etree.ElementTree.fromstring(object_str.encode('utf-8')) + object_doc = compat_etree_fromstring(object_str.encode('utf-8')) except compat_xml_parse_error: return diff --git a/youtube_dl/extractor/canalc2.py b/youtube_dl/extractor/canalc2.py index c4fefefe4..f6a1ff381 100644 --- a/youtube_dl/extractor/canalc2.py +++ b/youtube_dl/extractor/canalc2.py @@ -4,38 +4,53 @@ from __future__ import unicode_literals import re from .common import InfoExtractor +from ..utils import parse_duration class Canalc2IE(InfoExtractor): IE_NAME = 'canalc2.tv' - _VALID_URL = r'http://.*?\.canalc2\.tv/video\.asp\?.*?idVideo=(?P<id>\d+)' + _VALID_URL = r'https?://(?:www\.)?canalc2\.tv/video/(?P<id>\d+)' _TEST = { - 'url': 'http://www.canalc2.tv/video.asp?idVideo=12163&voir=oui', + 'url': 'http://www.canalc2.tv/video/12163', 'md5': '060158428b650f896c542dfbb3d6487f', 'info_dict': { 'id': '12163', - 'ext': 'mp4', - 'title': 'Terrasses du Numérique' + 'ext': 'flv', + 'title': 'Terrasses du Numérique', + 'duration': 122, + }, + 'params': { + 'skip_download': True, # Requires rtmpdump } } def _real_extract(self, url): - video_id = re.match(self._VALID_URL, url).group('id') - # We need to set the voir field for getting the file name - url = 'http://www.canalc2.tv/video.asp?idVideo=%s&voir=oui' % video_id + video_id = self._match_id(url) webpage = self._download_webpage(url, video_id) - file_name = self._search_regex( - r"so\.addVariable\('file','(.*?)'\);", - webpage, 'file name') - video_url = 'http://vod-flash.u-strasbg.fr:8080/' + file_name + video_url = self._search_regex( + r'jwplayer\((["\'])Player\1\)\.setup\({[^}]*file\s*:\s*(["\'])(?P<file>.+?)\2', + webpage, 'video_url', group='file') + formats = [{'url': video_url}] + if video_url.startswith('rtmp://'): + rtmp = re.search(r'^(?P<url>rtmp://[^/]+/(?P<app>.+/))(?P<play_path>mp4:.+)$', video_url) + formats[0].update({ + 'url': rtmp.group('url'), + 'ext': 'flv', + 'app': rtmp.group('app'), + 'play_path': rtmp.group('play_path'), + 'page_url': url, + }) title = self._html_search_regex( - r'class="evenement8">(.*?)</a>', webpage, 'title') + r'(?s)class="[^"]*col_description[^"]*">.*?<h3>(.*?)</h3>', webpage, 'title') + duration = parse_duration(self._search_regex( + r'id=["\']video_duree["\'][^>]*>([^<]+)', + webpage, 'duration', fatal=False)) return { 'id': video_id, - 'ext': 'mp4', - 'url': video_url, 'title': title, + 'duration': duration, + 'formats': formats, } diff --git a/youtube_dl/extractor/canalplus.py b/youtube_dl/extractor/canalplus.py index 57e0cda2c..004372f8d 100644 --- a/youtube_dl/extractor/canalplus.py +++ b/youtube_dl/extractor/canalplus.py @@ -78,7 +78,8 @@ class CanalplusIE(InfoExtractor): if video_id is None: webpage = self._download_webpage(url, display_id) video_id = self._search_regex( - r'<canal:player[^>]+?videoId="(\d+)"', webpage, 'video id') + [r'<canal:player[^>]+?videoId=(["\'])(?P<id>\d+)', r'id=["\']canal_video_player(?P<id>\d+)'], + webpage, 'video id', group='id') info_url = self._VIDEO_INFO_TEMPLATE % (site_id, video_id) doc = self._download_xml(info_url, video_id, 'Downloading video XML') diff --git a/youtube_dl/extractor/cbs.py b/youtube_dl/extractor/cbs.py index 75fffb156..43f05d278 100644 --- a/youtube_dl/extractor/cbs.py +++ b/youtube_dl/extractor/cbs.py @@ -1,6 +1,8 @@ from __future__ import unicode_literals from .common import InfoExtractor +from ..compat import compat_urllib_request +from ..utils import smuggle_url class CBSIE(InfoExtractor): @@ -46,13 +48,19 @@ class CBSIE(InfoExtractor): def _real_extract(self, url): display_id = self._match_id(url) - webpage = self._download_webpage(url, display_id) + request = compat_urllib_request.Request(url) + # Android UA is served with higher quality (720p) streams (see + # https://github.com/rg3/youtube-dl/issues/7490) + request.add_header('User-Agent', 'Mozilla/5.0 (Linux; Android 4.4; Nexus 5)') + webpage = self._download_webpage(request, display_id) real_id = self._search_regex( [r"video\.settings\.pid\s*=\s*'([^']+)';", r"cbsplayer\.pid\s*=\s*'([^']+)';"], webpage, 'real video ID') return { '_type': 'url_transparent', 'ie_key': 'ThePlatform', - 'url': 'theplatform:%s' % real_id, + 'url': smuggle_url( + 'http://link.theplatform.com/s/dJ5BDC/%s?mbr=true&manifest=m3u' % real_id, + {'force_smil_url': True}), 'display_id': display_id, } diff --git a/youtube_dl/extractor/cbsnews.py b/youtube_dl/extractor/cbsnews.py index 52e61d85b..f9a64a0a2 100644 --- a/youtube_dl/extractor/cbsnews.py +++ b/youtube_dl/extractor/cbsnews.py @@ -67,9 +67,12 @@ class CBSNewsIE(InfoExtractor): 'format_id': format_id, } if uri.startswith('rtmp'): + play_path = re.sub( + r'{slistFilePath}', '', + uri.split('<break>')[-1].split('{break}')[-1]) fmt.update({ 'app': 'ondemand?auth=cbs', - 'play_path': 'mp4:' + uri.split('<break>')[-1], + 'play_path': 'mp4:' + play_path, 'player_url': 'http://www.cbsnews.com/[[IMPORT]]/vidtech.cbsinteractive.com/player/3_3_0/CBSI_PLAYER_HD.swf', 'page_url': 'http://www.cbsnews.com', 'ext': 'flv', diff --git a/youtube_dl/extractor/channel9.py b/youtube_dl/extractor/channel9.py index 3dfc24f5b..c74553dcf 100644 --- a/youtube_dl/extractor/channel9.py +++ b/youtube_dl/extractor/channel9.py @@ -3,7 +3,11 @@ from __future__ import unicode_literals import re from .common import InfoExtractor -from ..utils import ExtractorError +from ..utils import ( + ExtractorError, + parse_filesize, + qualities, +) class Channel9IE(InfoExtractor): @@ -28,7 +32,7 @@ class Channel9IE(InfoExtractor): 'title': 'Developer Kick-Off Session: Stuff We Love', 'description': 'md5:c08d72240b7c87fcecafe2692f80e35f', 'duration': 4576, - 'thumbnail': 'http://video.ch9.ms/ch9/9d51/03902f2d-fc97-4d3c-b195-0bfe15a19d51/KOS002_220.jpg', + 'thumbnail': 're:http://.*\.jpg', 'session_code': 'KOS002', 'session_day': 'Day 1', 'session_room': 'Arena 1A', @@ -44,31 +48,29 @@ class Channel9IE(InfoExtractor): 'title': 'Self-service BI with Power BI - nuclear testing', 'description': 'md5:d1e6ecaafa7fb52a2cacdf9599829f5b', 'duration': 1540, - 'thumbnail': 'http://video.ch9.ms/ch9/87e1/0300391f-a455-4c72-bec3-4422f19287e1/selfservicenuk_512.jpg', + 'thumbnail': 're:http://.*\.jpg', 'authors': ['Mike Wilmot'], }, + }, + { + # low quality mp4 is best + 'url': 'https://channel9.msdn.com/Events/CPP/CppCon-2015/Ranges-for-the-Standard-Library', + 'info_dict': { + 'id': 'Events/CPP/CppCon-2015/Ranges-for-the-Standard-Library', + 'ext': 'mp4', + 'title': 'Ranges for the Standard Library', + 'description': 'md5:2e6b4917677af3728c5f6d63784c4c5d', + 'duration': 5646, + 'thumbnail': 're:http://.*\.jpg', + }, + 'params': { + 'skip_download': True, + }, } ] _RSS_URL = 'http://channel9.msdn.com/%s/RSS' - # Sorted by quality - _known_formats = ['MP3', 'MP4', 'Mid Quality WMV', 'Mid Quality MP4', 'High Quality WMV', 'High Quality MP4'] - - def _restore_bytes(self, formatted_size): - if not formatted_size: - return 0 - m = re.match(r'^(?P<size>\d+(?:\.\d+)?)\s+(?P<units>[a-zA-Z]+)', formatted_size) - if not m: - return 0 - units = m.group('units') - try: - exponent = ['B', 'KB', 'MB', 'GB', 'TB', 'PB', 'EB', 'ZB', 'YB'].index(units.upper()) - except ValueError: - return 0 - size = float(m.group('size')) - return int(size * (1024 ** exponent)) - def _formats_from_html(self, html): FORMAT_REGEX = r''' (?x) @@ -78,16 +80,20 @@ class Channel9IE(InfoExtractor): <h3>File\s+size</h3>\s*(?P<filesize>.*?)\s* </div>)? # File size part may be missing ''' - # Extract known formats + quality = qualities(( + 'MP3', 'MP4', + 'Low Quality WMV', 'Low Quality MP4', + 'Mid Quality WMV', 'Mid Quality MP4', + 'High Quality WMV', 'High Quality MP4')) formats = [{ 'url': x.group('url'), 'format_id': x.group('quality'), 'format_note': x.group('note'), 'format': '%s (%s)' % (x.group('quality'), x.group('note')), - 'filesize': self._restore_bytes(x.group('filesize')), # File size is approximate - 'preference': self._known_formats.index(x.group('quality')), + 'filesize_approx': parse_filesize(x.group('filesize')), + 'quality': quality(x.group('quality')), 'vcodec': 'none' if x.group('note') == 'Audio only' else None, - } for x in list(re.finditer(FORMAT_REGEX, html)) if x.group('quality') in self._known_formats] + } for x in list(re.finditer(FORMAT_REGEX, html))] self._sort_formats(formats) @@ -158,7 +164,7 @@ class Channel9IE(InfoExtractor): def _extract_session_day(self, html): m = re.search(r'<li class="day">\s*<a href="/Events/[^"]+">(?P<day>[^<]+)</a>\s*</li>', html) - return m.group('day') if m is not None else None + return m.group('day').strip() if m is not None else None def _extract_session_room(self, html): m = re.search(r'<li class="room">\s*(?P<room>.+?)\s*</li>', html) @@ -224,12 +230,12 @@ class Channel9IE(InfoExtractor): if contents is None: return contents - authors = self._extract_authors(html) + if len(contents) > 1: + raise ExtractorError('Got more than one entry') + result = contents[0] + result['authors'] = self._extract_authors(html) - for content in contents: - content['authors'] = authors - - return contents + return result def _extract_session(self, html, content_path): contents = self._extract_content(html, content_path) diff --git a/youtube_dl/extractor/chaturbate.py b/youtube_dl/extractor/chaturbate.py new file mode 100644 index 000000000..0b67ba67d --- /dev/null +++ b/youtube_dl/extractor/chaturbate.py @@ -0,0 +1,50 @@ +from __future__ import unicode_literals + +from .common import InfoExtractor +from ..utils import ExtractorError + + +class ChaturbateIE(InfoExtractor): + _VALID_URL = r'https?://(?:[^/]+\.)?chaturbate\.com/(?P<id>[^/?#]+)' + _TESTS = [{ + 'url': 'https://www.chaturbate.com/siswet19/', + 'info_dict': { + 'id': 'siswet19', + 'ext': 'mp4', + 'title': 're:^siswet19 [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$', + 'age_limit': 18, + 'is_live': True, + }, + 'params': { + 'skip_download': True, + } + }, { + 'url': 'https://en.chaturbate.com/siswet19/', + 'only_matching': True, + }] + + def _real_extract(self, url): + video_id = self._match_id(url) + + webpage = self._download_webpage(url, video_id) + + m3u8_url = self._search_regex( + r'src=(["\'])(?P<url>http.+?\.m3u8.*?)\1', webpage, + 'playlist', default=None, group='url') + + if not m3u8_url: + error = self._search_regex( + r'<span[^>]+class=(["\'])desc_span\1[^>]*>(?P<error>[^<]+)</span>', + webpage, 'error', group='error') + raise ExtractorError(error, expected=True) + + formats = self._extract_m3u8_formats(m3u8_url, video_id, ext='mp4') + + return { + 'id': video_id, + 'title': self._live_title(video_id), + 'thumbnail': 'https://cdn-s.highwebmedia.com/uHK3McUtGCG3SMFcd4ZJsRv8/roomimage/%s.jpg' % video_id, + 'age_limit': self._rta_search(webpage), + 'is_live': True, + 'formats': formats, + } diff --git a/youtube_dl/extractor/clubic.py b/youtube_dl/extractor/clubic.py index 14f215c5c..1dfa7c12e 100644 --- a/youtube_dl/extractor/clubic.py +++ b/youtube_dl/extractor/clubic.py @@ -12,9 +12,9 @@ from ..utils import ( class ClubicIE(InfoExtractor): - _VALID_URL = r'http://(?:www\.)?clubic\.com/video/[^/]+/video.*-(?P<id>[0-9]+)\.html' + _VALID_URL = r'http://(?:www\.)?clubic\.com/video/(?:[^/]+/)*video.*-(?P<id>[0-9]+)\.html' - _TEST = { + _TESTS = [{ 'url': 'http://www.clubic.com/video/clubic-week/video-clubic-week-2-0-le-fbi-se-lance-dans-la-photo-d-identite-448474.html', 'md5': '1592b694ba586036efac1776b0b43cd3', 'info_dict': { @@ -24,7 +24,10 @@ class ClubicIE(InfoExtractor): 'description': 're:Gueule de bois chez Nokia. Le constructeur a indiqué cette.*', 'thumbnail': 're:^http://img\.clubic\.com/.*\.jpg$', } - } + }, { + 'url': 'http://www.clubic.com/video/video-clubic-week-2-0-apple-iphone-6s-et-plus-mais-surtout-le-pencil-469792.html', + 'only_matching': True, + }] def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) diff --git a/youtube_dl/extractor/clyp.py b/youtube_dl/extractor/clyp.py new file mode 100644 index 000000000..57e643799 --- /dev/null +++ b/youtube_dl/extractor/clyp.py @@ -0,0 +1,57 @@ +from __future__ import unicode_literals + +from .common import InfoExtractor +from ..utils import ( + float_or_none, + parse_iso8601, +) + + +class ClypIE(InfoExtractor): + _VALID_URL = r'https?://(?:www\.)?clyp\.it/(?P<id>[a-z0-9]+)' + _TEST = { + 'url': 'https://clyp.it/ojz2wfah', + 'md5': '1d4961036c41247ecfdcc439c0cddcbb', + 'info_dict': { + 'id': 'ojz2wfah', + 'ext': 'mp3', + 'title': 'Krisson80 - bits wip wip', + 'description': '#Krisson80BitsWipWip #chiptune\n#wip', + 'duration': 263.21, + 'timestamp': 1443515251, + 'upload_date': '20150929', + }, + } + + def _real_extract(self, url): + audio_id = self._match_id(url) + + metadata = self._download_json( + 'https://api.clyp.it/%s' % audio_id, audio_id) + + formats = [] + for secure in ('', 'Secure'): + for ext in ('Ogg', 'Mp3'): + format_id = '%s%s' % (secure, ext) + format_url = metadata.get('%sUrl' % format_id) + if format_url: + formats.append({ + 'url': format_url, + 'format_id': format_id, + 'vcodec': 'none', + }) + self._sort_formats(formats) + + title = metadata['Title'] + description = metadata.get('Description') + duration = float_or_none(metadata.get('Duration')) + timestamp = parse_iso8601(metadata.get('DateCreated')) + + return { + 'id': audio_id, + 'title': title, + 'description': description, + 'duration': duration, + 'timestamp': timestamp, + 'formats': formats, + } diff --git a/youtube_dl/extractor/cmt.py b/youtube_dl/extractor/cmt.py index e96c59f71..f1311b14f 100644 --- a/youtube_dl/extractor/cmt.py +++ b/youtube_dl/extractor/cmt.py @@ -4,7 +4,7 @@ from .mtv import MTVIE class CMTIE(MTVIE): IE_NAME = 'cmt.com' - _VALID_URL = r'https?://www\.cmt\.com/videos/.+?/(?P<videoid>[^/]+)\.jhtml' + _VALID_URL = r'https?://www\.cmt\.com/(?:videos|shows)/(?:[^/]+/)*(?P<videoid>\d+)' _FEED_URL = 'http://www.cmt.com/sitewide/apps/player/embed/rss/' _TESTS = [{ @@ -16,4 +16,7 @@ class CMTIE(MTVIE): 'title': 'Garth Brooks - "The Call (featuring Trisha Yearwood)"', 'description': 'Blame It All On My Roots', }, + }, { + 'url': 'http://www.cmt.com/shows/party-down-south/party-down-south-ep-407-gone-girl/1738172/playlist/#id=1738172', + 'only_matching': True, }] diff --git a/youtube_dl/extractor/comedycentral.py b/youtube_dl/extractor/comedycentral.py index 91ebb0ce5..3e4bd10b6 100644 --- a/youtube_dl/extractor/comedycentral.py +++ b/youtube_dl/extractor/comedycentral.py @@ -151,12 +151,7 @@ class ComedyCentralShowsIE(MTVServicesInfoExtractor): mobj = re.match(self._VALID_URL, url) if mobj.group('shortname'): - if mobj.group('shortname') in ('tds', 'thedailyshow'): - url = 'http://thedailyshow.cc.com/full-episodes/' - else: - url = 'http://thecolbertreport.cc.com/full-episodes/' - mobj = re.match(self._VALID_URL, url, re.VERBOSE) - assert mobj is not None + return self.url_result('http://www.cc.com/shows/the-daily-show-with-trevor-noah/full-episodes') if mobj.group('clip'): if mobj.group('videotitle'): diff --git a/youtube_dl/extractor/common.py b/youtube_dl/extractor/common.py index 5eeeda08d..5e263f8b5 100644 --- a/youtube_dl/extractor/common.py +++ b/youtube_dl/extractor/common.py @@ -10,13 +10,11 @@ import re import socket import sys import time -import xml.etree.ElementTree from ..compat import ( compat_cookiejar, compat_cookies, compat_getpass, - compat_HTTPError, compat_http_client, compat_urllib_error, compat_urllib_parse, @@ -24,6 +22,7 @@ from ..compat import ( compat_urllib_request, compat_urlparse, compat_str, + compat_etree_fromstring, ) from ..utils import ( NO_DEFAULT, @@ -39,6 +38,7 @@ from ..utils import ( RegexNotFoundError, sanitize_filename, unescapeHTML, + unified_strdate, url_basename, xpath_text, xpath_with_ns, @@ -152,6 +152,7 @@ class InfoExtractor(object): description: Full video description. uploader: Full name of the video uploader. creator: The main artist who created the video. + release_date: The date (YYYYMMDD) when the video was released. timestamp: UNIX timestamp of the moment the video became available. upload_date: Video upload date (YYYYMMDD). If not explicitly set, calculated from timestamp. @@ -163,12 +164,14 @@ class InfoExtractor(object): with the "ext" entry and one of: * "data": The subtitles file contents * "url": A URL pointing to the subtitles file + "ext" will be calculated from URL if missing automatic_captions: Like 'subtitles', used by the YoutubeIE for automatically generated captions duration: Length of the video in seconds, as an integer. view_count: How many users have watched the video on the platform. like_count: Number of positive ratings of the video dislike_count: Number of negative ratings of the video + repost_count: Number of reposts of the video average_rating: Average rating give by users, the scale used depends on the webpage comment_count: Number of comments on the video comments: A list of comments, each with one or more of the following @@ -307,11 +310,11 @@ class InfoExtractor(object): @classmethod def ie_key(cls): """A string for getting the InfoExtractor with get_info_extractor""" - return cls.__name__[:-2] + return compat_str(cls.__name__[:-2]) @property def IE_NAME(self): - return type(self).__name__[:-2] + return compat_str(type(self).__name__[:-2]) def _request_webpage(self, url_or_request, video_id, note=None, errnote=None, fatal=True): """ Returns the response handle """ @@ -458,7 +461,7 @@ class InfoExtractor(object): return xml_string if transform_source: xml_string = transform_source(xml_string) - return xml.etree.ElementTree.fromstring(xml_string.encode('utf-8')) + return compat_etree_fromstring(xml_string.encode('utf-8')) def _download_json(self, url_or_request, video_id, note='Downloading JSON metadata', @@ -516,6 +519,12 @@ class InfoExtractor(object): '%s. Use --username and --password or --netrc to provide account credentials.' % msg, expected=True) + @staticmethod + def raise_geo_restricted(msg='This video is not available from your location due to geo restriction'): + raise ExtractorError( + '%s. You might want to use --proxy to workaround.' % msg, + expected=True) + # Methods for following #608 @staticmethod def url_result(url, ie=None, video_id=None, video_title=None): @@ -636,8 +645,9 @@ class InfoExtractor(object): # Helper functions for extracting OpenGraph info @staticmethod def _og_regexes(prop): - content_re = r'content=(?:"([^>]+?)"|\'([^>]+?)\')' - property_re = r'(?:name|property)=[\'"]og:%s[\'"]' % re.escape(prop) + content_re = r'content=(?:"([^"]+?)"|\'([^\']+?)\'|\s*([^\s"\'=<>`]+?))' + property_re = (r'(?:name|property)=(?:\'og:%(prop)s\'|"og:%(prop)s"|\s*og:%(prop)s\b)' + % {'prop': re.escape(prop)}) template = r'<meta[^>]+?%s[^>]+?%s' return [ template % (property_re, content_re), @@ -731,8 +741,9 @@ class InfoExtractor(object): @staticmethod def _hidden_inputs(html): + html = re.sub(r'<!--(?:(?!<!--).)*-->', '', html) hidden_inputs = {} - for input in re.findall(r'<input([^>]+)>', html): + for input in re.findall(r'(?i)<input([^>]+)>', html): if not re.search(r'type=(["\'])(?:hidden|submit)\1', input): continue name = re.search(r'name=(["\'])(?P<value>.+?)\1', input) @@ -746,7 +757,7 @@ class InfoExtractor(object): def _form_hidden_inputs(self, form_id, html): form = self._search_regex( - r'(?s)<form[^>]+?id=(["\'])%s\1[^>]*>(?P<form>.+?)</form>' % form_id, + r'(?is)<form[^>]+?id=(["\'])%s\1[^>]*>(?P<form>.+?)</form>' % form_id, html, '%s form' % form_id, group='form') return self._hidden_inputs(form) @@ -830,7 +841,7 @@ class InfoExtractor(object): self._request_webpage(url, video_id, 'Checking %s URL' % item) return True except ExtractorError as e: - if isinstance(e.cause, compat_HTTPError): + if isinstance(e.cause, compat_urllib_error.URLError): self.to_screen( '%s: %s URL is invalid, skipping' % (video_id, item)) return False @@ -861,13 +872,18 @@ class InfoExtractor(object): time.sleep(timeout) def _extract_f4m_formats(self, manifest_url, video_id, preference=None, f4m_id=None, - transform_source=lambda s: fix_xml_ampersands(s).strip()): + transform_source=lambda s: fix_xml_ampersands(s).strip(), + fatal=True): manifest = self._download_xml( manifest_url, video_id, 'Downloading f4m manifest', 'Unable to download f4m manifest', # Some manifests may be malformed, e.g. prosiebensat1 generated manifests # (see https://github.com/rg3/youtube-dl/issues/6215#issuecomment-121704244) - transform_source=transform_source) + transform_source=transform_source, + fatal=fatal) + + if manifest is False: + return manifest formats = [] manifest_version = '1.0' @@ -888,7 +904,10 @@ class InfoExtractor(object): # may differ leading to inability to resolve the format by requested # bitrate in f4m downloader if determine_ext(manifest_url) == 'f4m': - formats.extend(self._extract_f4m_formats(manifest_url, video_id, preference, f4m_id)) + f4m_formats = self._extract_f4m_formats( + manifest_url, video_id, preference, f4m_id, fatal=fatal) + if f4m_formats: + formats.extend(f4m_formats) continue tbr = int_or_none(media_el.attrib.get('bitrate')) formats.append({ @@ -924,13 +943,15 @@ class InfoExtractor(object): if re.match(r'^https?://', u) else compat_urlparse.urljoin(m3u8_url, u)) - m3u8_doc = self._download_webpage( + res = self._download_webpage_handle( m3u8_url, video_id, note=note or 'Downloading m3u8 information', errnote=errnote or 'Failed to download m3u8 information', fatal=fatal) - if m3u8_doc is False: - return m3u8_doc + if res is False: + return res + m3u8_doc, urlh = res + m3u8_url = urlh.geturl() last_info = None last_media = None kv_rex = re.compile( @@ -1036,6 +1057,7 @@ class InfoExtractor(object): video_id = os.path.splitext(url_basename(smil_url))[0] title = None description = None + upload_date = None for meta in smil.findall(self._xpath_ns('./head/meta', namespace)): name = meta.attrib.get('name') content = meta.attrib.get('content') @@ -1045,11 +1067,22 @@ class InfoExtractor(object): title = content elif not description and name in ('description', 'abstract'): description = content + elif not upload_date and name == 'date': + upload_date = unified_strdate(content) + + thumbnails = [{ + 'id': image.get('type'), + 'url': image.get('src'), + 'width': int_or_none(image.get('width')), + 'height': int_or_none(image.get('height')), + } for image in smil.findall(self._xpath_ns('.//image', namespace)) if image.get('src')] return { 'id': video_id, 'title': title or video_id, 'description': description, + 'upload_date': upload_date, + 'thumbnails': thumbnails, 'formats': formats, 'subtitles': subtitles, } @@ -1076,7 +1109,7 @@ class InfoExtractor(object): if not src: continue - bitrate = int_or_none(video.get('system-bitrate') or video.get('systemBitrate'), 1000) + bitrate = float_or_none(video.get('system-bitrate') or video.get('systemBitrate'), 1000) filesize = int_or_none(video.get('size') or video.get('fileSize')) width = int_or_none(video.get('width')) height = int_or_none(video.get('height')) @@ -1108,8 +1141,10 @@ class InfoExtractor(object): src_url = src if src.startswith('http') else compat_urlparse.urljoin(base, src) if proto == 'm3u8' or src_ext == 'm3u8': - formats.extend(self._extract_m3u8_formats( - src_url, video_id, ext or 'mp4', m3u8_id='hls')) + m3u8_formats = self._extract_m3u8_formats( + src_url, video_id, ext or 'mp4', m3u8_id='hls', fatal=False) + if m3u8_formats: + formats.extend(m3u8_formats) continue if src_ext == 'f4m': @@ -1121,10 +1156,12 @@ class InfoExtractor(object): } f4m_url += '&' if '?' in f4m_url else '?' f4m_url += compat_urllib_parse.urlencode(f4m_params) - formats.extend(self._extract_f4m_formats(f4m_url, video_id, f4m_id='hds')) + f4m_formats = self._extract_f4m_formats(f4m_url, video_id, f4m_id='hds', fatal=False) + if f4m_formats: + formats.extend(f4m_formats) continue - if src_url.startswith('http'): + if src_url.startswith('http') and self._is_valid_url(src, video_id): http_count += 1 formats.append({ 'url': src_url, diff --git a/youtube_dl/extractor/condenast.py b/youtube_dl/extractor/condenast.py index 3db4db4e4..6f92ae2ed 100644 --- a/youtube_dl/extractor/condenast.py +++ b/youtube_dl/extractor/condenast.py @@ -2,7 +2,6 @@ from __future__ import unicode_literals import re -import json from .common import InfoExtractor from ..compat import ( @@ -12,6 +11,7 @@ from ..compat import ( ) from ..utils import ( orderedSet, + remove_end, ) @@ -24,21 +24,33 @@ class CondeNastIE(InfoExtractor): # The keys are the supported sites and the values are the name to be shown # to the user and in the extractor description. _SITES = { - 'wired': 'WIRED', + 'allure': 'Allure', + 'architecturaldigest': 'Architectural Digest', + 'arstechnica': 'Ars Technica', + 'bonappetit': 'Bon Appétit', + 'brides': 'Brides', + 'cnevids': 'Condé Nast', + 'cntraveler': 'Condé Nast Traveler', + 'details': 'Details', + 'epicurious': 'Epicurious', + 'glamour': 'Glamour', + 'golfdigest': 'Golf Digest', 'gq': 'GQ', + 'newyorker': 'The New Yorker', + 'self': 'SELF', + 'teenvogue': 'Teen Vogue', + 'vanityfair': 'Vanity Fair', 'vogue': 'Vogue', - 'glamour': 'Glamour', + 'wired': 'WIRED', 'wmagazine': 'W Magazine', - 'vanityfair': 'Vanity Fair', - 'cnevids': 'Condé Nast', } - _VALID_URL = r'http://(video|www|player)\.(?P<site>%s)\.com/(?P<type>watch|series|video|embed)/(?P<id>[^/?#]+)' % '|'.join(_SITES.keys()) + _VALID_URL = r'http://(?:video|www|player)\.(?P<site>%s)\.com/(?P<type>watch|series|video|embed(?:js)?)/(?P<id>[^/?#]+)' % '|'.join(_SITES.keys()) IE_DESC = 'Condé Nast media group: %s' % ', '.join(sorted(_SITES.values())) - EMBED_URL = r'(?:https?:)?//player\.(?P<site>%s)\.com/(?P<type>embed)/.+?' % '|'.join(_SITES.keys()) + EMBED_URL = r'(?:https?:)?//player\.(?P<site>%s)\.com/(?P<type>embed(?:js)?)/.+?' % '|'.join(_SITES.keys()) - _TEST = { + _TESTS = [{ 'url': 'http://video.wired.com/watch/3d-printed-speakers-lit-with-led', 'md5': '1921f713ed48aabd715691f774c451f7', 'info_dict': { @@ -47,7 +59,16 @@ class CondeNastIE(InfoExtractor): 'title': '3D Printed Speakers Lit With LED', 'description': 'Check out these beautiful 3D printed LED speakers. You can\'t actually buy them, but LumiGeek is working on a board that will let you make you\'re own.', } - } + }, { + # JS embed + 'url': 'http://player.cnevids.com/embedjs/55f9cf8b61646d1acf00000c/5511d76261646d5566020000.js', + 'md5': 'f1a6f9cafb7083bab74a710f65d08999', + 'info_dict': { + 'id': '55f9cf8b61646d1acf00000c', + 'ext': 'mp4', + 'title': '3D printed TSA Travel Sentry keys really do open TSA locks', + } + }] def _extract_series(self, url, webpage): title = self._html_search_regex(r'<div class="cne-series-info">.*?<h1>(.+?)</h1>', @@ -86,8 +107,8 @@ class CondeNastIE(InfoExtractor): info_url = base_info_url + data info_page = self._download_webpage(info_url, video_id, 'Downloading video info') - video_info = self._search_regex(r'var video = ({.+?});', info_page, 'video info') - video_info = json.loads(video_info) + video_info = self._search_regex(r'var\s+video\s*=\s*({.+?});', info_page, 'video info') + video_info = self._parse_json(video_info, video_id) formats = [{ 'format_id': '%s-%s' % (fdata['type'].split('/')[-1], fdata['quality']), @@ -111,6 +132,13 @@ class CondeNastIE(InfoExtractor): url_type = mobj.group('type') item_id = mobj.group('id') + # Convert JS embed to regular embed + if url_type == 'embedjs': + parsed_url = compat_urlparse.urlparse(url) + url = compat_urlparse.urlunparse(parsed_url._replace( + path=remove_end(parsed_url.path, '.js').replace('/embedjs/', '/embed/'))) + url_type = 'embed' + self.to_screen('Extracting from %s with the Condé Nast extractor' % self._SITES[site]) webpage = self._download_webpage(url, item_id) diff --git a/youtube_dl/extractor/criterion.py b/youtube_dl/extractor/criterion.py index 4fb178165..dedb810a0 100644 --- a/youtube_dl/extractor/criterion.py +++ b/youtube_dl/extractor/criterion.py @@ -27,9 +27,7 @@ class CriterionIE(InfoExtractor): final_url = self._search_regex( r'so.addVariable\("videoURL", "(.+?)"\)\;', webpage, 'video url') title = self._og_search_title(webpage) - description = self._html_search_regex( - r'<meta name="description" content="(.+?)" />', - webpage, 'video description') + description = self._html_search_meta('description', webpage) thumbnail = self._search_regex( r'so.addVariable\("thumbnailURL", "(.+?)"\)\;', webpage, 'thumbnail url') diff --git a/youtube_dl/extractor/crunchyroll.py b/youtube_dl/extractor/crunchyroll.py index 95952bc29..6e5999c72 100644 --- a/youtube_dl/extractor/crunchyroll.py +++ b/youtube_dl/extractor/crunchyroll.py @@ -5,12 +5,12 @@ import re import json import base64 import zlib -import xml.etree.ElementTree from hashlib import sha1 from math import pow, sqrt, floor from .common import InfoExtractor from ..compat import ( + compat_etree_fromstring, compat_urllib_parse, compat_urllib_parse_unquote, compat_urllib_request, @@ -21,6 +21,7 @@ from ..utils import ( bytes_to_intlist, intlist_to_bytes, int_or_none, + lowercase_escape, remove_end, unified_strdate, urlencode_postdata, @@ -32,6 +33,26 @@ from ..aes import ( class CrunchyrollBaseIE(InfoExtractor): + _NETRC_MACHINE = 'crunchyroll' + + def _login(self): + (username, password) = self._get_login_info() + if username is None: + return + self.report_login() + login_url = 'https://www.crunchyroll.com/?a=formhandler' + data = urlencode_postdata({ + 'formname': 'RpcApiUser_Login', + 'name': username, + 'password': password, + }) + login_request = compat_urllib_request.Request(login_url, data) + login_request.add_header('Content-Type', 'application/x-www-form-urlencoded') + self._download_webpage(login_request, None, False, 'Wrong login info') + + def _real_initialize(self): + self._login() + def _download_webpage(self, url_or_request, video_id, note=None, errnote=None, fatal=True, tries=1, timeout=5, encoding=None): request = (url_or_request if isinstance(url_or_request, compat_urllib_request.Request) else compat_urllib_request.Request(url_or_request)) @@ -46,10 +67,22 @@ class CrunchyrollBaseIE(InfoExtractor): return super(CrunchyrollBaseIE, self)._download_webpage( request, video_id, note, errnote, fatal, tries, timeout, encoding) + @staticmethod + def _add_skip_wall(url): + parsed_url = compat_urlparse.urlparse(url) + qs = compat_urlparse.parse_qs(parsed_url.query) + # Always force skip_wall to bypass maturity wall, namely 18+ confirmation message: + # > This content may be inappropriate for some people. + # > Are you sure you want to continue? + # since it's not disabled by default in crunchyroll account's settings. + # See https://github.com/rg3/youtube-dl/issues/7202. + qs['skip_wall'] = ['1'] + return compat_urlparse.urlunparse( + parsed_url._replace(query=compat_urllib_parse.urlencode(qs, True))) + class CrunchyrollIE(CrunchyrollBaseIE): _VALID_URL = r'https?://(?:(?P<prefix>www|m)\.)?(?P<url>crunchyroll\.(?:com|fr)/(?:media(?:-|/\?id=)|[^/]*/[^/?&]*?)(?P<video_id>[0-9]+))(?:[/?&]|$)' - _NETRC_MACHINE = 'crunchyroll' _TESTS = [{ 'url': 'http://www.crunchyroll.com/wanna-be-the-strongest-in-the-world/episode-1-an-idol-wrestler-is-born-645513', 'info_dict': { @@ -72,7 +105,7 @@ class CrunchyrollIE(CrunchyrollBaseIE): 'id': '589804', 'ext': 'flv', 'title': 'Culture Japan Episode 1 – Rebuilding Japan after the 3.11', - 'description': 'md5:fe2743efedb49d279552926d0bd0cd9e', + 'description': 'md5:2fbc01f90b87e8e9137296f37b461c12', 'thumbnail': 're:^https?://.*\.jpg$', 'uploader': 'Danny Choo Network', 'upload_date': '20120213', @@ -81,10 +114,13 @@ class CrunchyrollIE(CrunchyrollBaseIE): # rtmp 'skip_download': True, }, - }, { 'url': 'http://www.crunchyroll.fr/girl-friend-beta/episode-11-goodbye-la-mode-661697', 'only_matching': True, + }, { + # geo-restricted (US), 18+ maturity wall, non-premium available + 'url': 'http://www.crunchyroll.com/cosplay-complex-ova/episode-1-the-birth-of-the-cosplay-club-565617', + 'only_matching': True, }] _FORMAT_IDS = { @@ -94,24 +130,6 @@ class CrunchyrollIE(CrunchyrollBaseIE): '1080': ('80', '108'), } - def _login(self): - (username, password) = self._get_login_info() - if username is None: - return - self.report_login() - login_url = 'https://www.crunchyroll.com/?a=formhandler' - data = urlencode_postdata({ - 'formname': 'RpcApiUser_Login', - 'name': username, - 'password': password, - }) - login_request = compat_urllib_request.Request(login_url, data) - login_request.add_header('Content-Type', 'application/x-www-form-urlencoded') - self._download_webpage(login_request, None, False, 'Wrong login info') - - def _real_initialize(self): - self._login() - def _decrypt_subtitles(self, data, iv, id): data = bytes_to_intlist(base64.b64decode(data.encode('utf-8'))) iv = bytes_to_intlist(base64.b64decode(iv.encode('utf-8'))) @@ -217,7 +235,7 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text return output def _extract_subtitles(self, subtitle): - sub_root = xml.etree.ElementTree.fromstring(subtitle) + sub_root = compat_etree_fromstring(subtitle) return [{ 'ext': 'srt', 'data': self._convert_subtitles_to_srt(sub_root), @@ -228,7 +246,7 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text def _get_subtitles(self, video_id, webpage): subtitles = {} - for sub_id, sub_name in re.findall(r'\?ssid=([0-9]+)" title="([^"]+)', webpage): + for sub_id, sub_name in re.findall(r'\bssid=([0-9]+)"[^>]+?\btitle="([^"]+)', webpage): sub_page = self._download_webpage( 'http://www.crunchyroll.com/xml/?req=RpcApiSubtitle_GetXml&subtitle_script_id=' + sub_id, video_id, note='Downloading subtitles for ' + sub_name) @@ -254,7 +272,7 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text else: webpage_url = 'http://www.' + mobj.group('url') - webpage = self._download_webpage(webpage_url, video_id, 'Downloading webpage') + webpage = self._download_webpage(self._add_skip_wall(webpage_url), video_id, 'Downloading webpage') note_m = self._html_search_regex( r'<div class="showmedia-trailer-notice">(.+?)</div>', webpage, 'trailer-notice', default='') @@ -270,11 +288,15 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text if 'To view this, please log in to verify you are 18 or older.' in webpage: self.raise_login_required() - video_title = self._html_search_regex(r'<h1[^>]*>(.+?)</h1>', webpage, 'video_title', flags=re.DOTALL) + video_title = self._html_search_regex( + r'(?s)<h1[^>]*>((?:(?!<h1).)*?<span[^>]+itemprop=["\']title["\'][^>]*>(?:(?!<h1).)+?)</h1>', + webpage, 'video_title') video_title = re.sub(r' {2,}', ' ', video_title) - video_description = self._html_search_regex(r'"description":"([^"]+)', webpage, 'video_description', default='') - if not video_description: - video_description = None + video_description = self._html_search_regex( + r'<script[^>]*>\s*.+?\[media_id=%s\].+?"description"\s*:\s*"([^"]+)' % video_id, + webpage, 'description', default=None) + if video_description: + video_description = lowercase_escape(video_description.replace(r'\r\n', '\n')) video_upload_date = self._html_search_regex( [r'<div>Availability for free users:(.+?)</div>', r'<div>[^<>]+<span>\s*(.+?\d{4})\s*</span></div>'], webpage, 'video_upload_date', fatal=False, flags=re.DOTALL) @@ -352,7 +374,7 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text class CrunchyrollShowPlaylistIE(CrunchyrollBaseIE): IE_NAME = "crunchyroll:playlist" - _VALID_URL = r'https?://(?:(?P<prefix>www|m)\.)?(?P<url>crunchyroll\.com/(?!(?:news|anime-news|library|forum|launchcalendar|lineup|store|comics|freetrial|login))(?P<id>[\w\-]+))/?$' + _VALID_URL = r'https?://(?:(?P<prefix>www|m)\.)?(?P<url>crunchyroll\.com/(?!(?:news|anime-news|library|forum|launchcalendar|lineup|store|comics|freetrial|login))(?P<id>[\w\-]+))/?(?:\?|$)' _TESTS = [{ 'url': 'http://www.crunchyroll.com/a-bridge-to-the-starry-skies-hoshizora-e-kakaru-hashi', @@ -361,12 +383,25 @@ class CrunchyrollShowPlaylistIE(CrunchyrollBaseIE): 'title': 'A Bridge to the Starry Skies - Hoshizora e Kakaru Hashi' }, 'playlist_count': 13, + }, { + # geo-restricted (US), 18+ maturity wall, non-premium available + 'url': 'http://www.crunchyroll.com/cosplay-complex-ova', + 'info_dict': { + 'id': 'cosplay-complex-ova', + 'title': 'Cosplay Complex OVA' + }, + 'playlist_count': 3, + 'skip': 'Georestricted', + }, { + # geo-restricted (US), 18+ maturity wall, non-premium will be available since 2015.11.14 + 'url': 'http://www.crunchyroll.com/ladies-versus-butlers?skip_wall=1', + 'only_matching': True, }] def _real_extract(self, url): show_id = self._match_id(url) - webpage = self._download_webpage(url, show_id) + webpage = self._download_webpage(self._add_skip_wall(url), show_id) title = self._html_search_regex( r'(?s)<h1[^>]*>\s*<span itemprop="name">(.*?)</span>', webpage, 'title') diff --git a/youtube_dl/extractor/dailymotion.py b/youtube_dl/extractor/dailymotion.py index 2d90b2224..bc7823931 100644 --- a/youtube_dl/extractor/dailymotion.py +++ b/youtube_dl/extractor/dailymotion.py @@ -96,6 +96,11 @@ class DailymotionIE(DailymotionBaseInfoExtractor): 'uploader': 'HotWaves1012', 'age_limit': 18, } + }, + # geo-restricted, player v5 + { + 'url': 'http://www.dailymotion.com/video/xhza0o', + 'only_matching': True, } ] @@ -119,11 +124,14 @@ class DailymotionIE(DailymotionBaseInfoExtractor): webpage, 'comment count', fatal=False)) player_v5 = self._search_regex( - r'playerV5\s*=\s*dmp\.create\([^,]+?,\s*({.+?})\);', + [r'buildPlayer\(({.+?})\);', r'playerV5\s*=\s*dmp\.create\([^,]+?,\s*({.+?})\);'], webpage, 'player v5', default=None) if player_v5: player = self._parse_json(player_v5, video_id) metadata = player['metadata'] + + self._check_error(metadata) + formats = [] for quality, media_list in metadata['qualities'].items(): for media in media_list: @@ -133,9 +141,17 @@ class DailymotionIE(DailymotionBaseInfoExtractor): type_ = media.get('type') if type_ == 'application/vnd.lumberjack.manifest': continue - if type_ == 'application/x-mpegURL' or determine_ext(media_url) == 'm3u8': - formats.extend(self._extract_m3u8_formats( - media_url, video_id, 'mp4', m3u8_id='hls')) + ext = determine_ext(media_url) + if type_ == 'application/x-mpegURL' or ext == 'm3u8': + m3u8_formats = self._extract_m3u8_formats( + media_url, video_id, 'mp4', m3u8_id='hls', fatal=False) + if m3u8_formats: + formats.extend(m3u8_formats) + elif type_ == 'application/f4m' or ext == 'f4m': + f4m_formats = self._extract_f4m_formats( + media_url, video_id, preference=-1, f4m_id='hds', fatal=False) + if f4m_formats: + formats.extend(f4m_formats) else: f = { 'url': media_url, @@ -201,9 +217,7 @@ class DailymotionIE(DailymotionBaseInfoExtractor): 'video info', flags=re.MULTILINE), video_id) - if info.get('error') is not None: - msg = 'Couldn\'t get video, Dailymotion says: %s' % info['error']['title'] - raise ExtractorError(msg, expected=True) + self._check_error(info) formats = [] for (key, format_id) in self._FORMATS: @@ -246,6 +260,11 @@ class DailymotionIE(DailymotionBaseInfoExtractor): 'duration': info['duration'] } + def _check_error(self, info): + if info.get('error') is not None: + raise ExtractorError( + '%s said: %s' % (self.IE_NAME, info['error']['title']), expected=True) + def _get_subtitles(self, video_id, webpage): try: sub_list = self._download_webpage( diff --git a/youtube_dl/extractor/democracynow.py b/youtube_dl/extractor/democracynow.py new file mode 100644 index 000000000..6cd395e11 --- /dev/null +++ b/youtube_dl/extractor/democracynow.py @@ -0,0 +1,88 @@ +# coding: utf-8 +from __future__ import unicode_literals + +import re +import os.path + +from .common import InfoExtractor +from ..compat import compat_urlparse +from ..utils import ( + url_basename, + remove_start, +) + + +class DemocracynowIE(InfoExtractor): + _VALID_URL = r'https?://(?:www\.)?democracynow.org/(?P<id>[^\?]*)' + IE_NAME = 'democracynow' + _TESTS = [{ + 'url': 'http://www.democracynow.org/shows/2015/7/3', + 'md5': 'fbb8fe3d7a56a5e12431ce2f9b2fab0d', + 'info_dict': { + 'id': '2015-0703-001', + 'ext': 'mp4', + 'title': 'July 03, 2015 - Democracy Now!', + 'description': 'A daily independent global news hour with Amy Goodman & Juan González "What to the Slave is 4th of July?": James Earl Jones Reads Frederick Douglass\u2019 Historic Speech : "This Flag Comes Down Today": Bree Newsome Scales SC Capitol Flagpole, Takes Down Confederate Flag : "We Shall Overcome": Remembering Folk Icon, Activist Pete Seeger in His Own Words & Songs', + }, + }, { + 'url': 'http://www.democracynow.org/2015/7/3/this_flag_comes_down_today_bree', + 'md5': 'fbb8fe3d7a56a5e12431ce2f9b2fab0d', + 'info_dict': { + 'id': '2015-0703-001', + 'ext': 'mp4', + 'title': '"This Flag Comes Down Today": Bree Newsome Scales SC Capitol Flagpole, Takes Down Confederate Flag', + 'description': 'md5:4d2bc4f0d29f5553c2210a4bc7761a21', + }, + }] + + def _real_extract(self, url): + display_id = self._match_id(url) + webpage = self._download_webpage(url, display_id) + description = self._og_search_description(webpage) + + json_data = self._parse_json(self._search_regex( + r'<script[^>]+type="text/json"[^>]*>\s*({[^>]+})', webpage, 'json'), + display_id) + video_id = None + formats = [] + + default_lang = 'en' + + subtitles = {} + + def add_subtitle_item(lang, info_dict): + if lang not in subtitles: + subtitles[lang] = [] + subtitles[lang].append(info_dict) + + # chapter_file are not subtitles + if 'caption_file' in json_data: + add_subtitle_item(default_lang, { + 'url': compat_urlparse.urljoin(url, json_data['caption_file']), + }) + + for subtitle_item in json_data.get('captions', []): + lang = subtitle_item.get('language', '').lower() or default_lang + add_subtitle_item(lang, { + 'url': compat_urlparse.urljoin(url, subtitle_item['url']), + }) + + for key in ('file', 'audio', 'video'): + media_url = json_data.get(key, '') + if not media_url: + continue + media_url = re.sub(r'\?.*', '', compat_urlparse.urljoin(url, media_url)) + video_id = video_id or remove_start(os.path.splitext(url_basename(media_url))[0], 'dn') + formats.append({ + 'url': media_url, + }) + + self._sort_formats(formats) + + return { + 'id': video_id or display_id, + 'title': json_data['title'], + 'description': description, + 'subtitles': subtitles, + 'formats': formats, + } diff --git a/youtube_dl/extractor/divxstage.py b/youtube_dl/extractor/divxstage.py deleted file mode 100644 index b88379e06..000000000 --- a/youtube_dl/extractor/divxstage.py +++ /dev/null @@ -1,27 +0,0 @@ -from __future__ import unicode_literals - -from .novamov import NovaMovIE - - -class DivxStageIE(NovaMovIE): - IE_NAME = 'divxstage' - IE_DESC = 'DivxStage' - - _VALID_URL = NovaMovIE._VALID_URL_TEMPLATE % {'host': 'divxstage\.(?:eu|net|ch|co|at|ag|to)'} - - _HOST = 'www.divxstage.eu' - - _FILE_DELETED_REGEX = r'>This file no longer exists on our servers.<' - _TITLE_REGEX = r'<div class="video_det">\s*<strong>([^<]+)</strong>' - _DESCRIPTION_REGEX = r'<div class="video_det">\s*<strong>[^<]+</strong>\s*<p>([^<]+)</p>' - - _TEST = { - 'url': 'http://www.divxstage.eu/video/57f238e2e5e01', - 'md5': '63969f6eb26533a1968c4d325be63e72', - 'info_dict': { - 'id': '57f238e2e5e01', - 'ext': 'flv', - 'title': 'youtubedl test video', - 'description': 'This is a test video for youtubedl.', - } - } diff --git a/youtube_dl/extractor/eagleplatform.py b/youtube_dl/extractor/eagleplatform.py index a1ee51568..7bbf617d4 100644 --- a/youtube_dl/extractor/eagleplatform.py +++ b/youtube_dl/extractor/eagleplatform.py @@ -21,7 +21,7 @@ class EaglePlatformIE(InfoExtractor): _TESTS = [{ # http://lenta.ru/news/2015/03/06/navalny/ 'url': 'http://lentaru.media.eagleplatform.com/index/player?player=new&record_id=227304&player_template_id=5201', - 'md5': '0b7994faa2bd5c0f69a3db6db28d078d', + 'md5': '70f5187fb620f2c1d503b3b22fd4efe3', 'info_dict': { 'id': '227304', 'ext': 'mp4', @@ -36,7 +36,7 @@ class EaglePlatformIE(InfoExtractor): # http://muz-tv.ru/play/7129/ # http://media.clipyou.ru/index/player?record_id=12820&width=730&height=415&autoplay=true 'url': 'eagleplatform:media.clipyou.ru:12820', - 'md5': '6c2ebeab03b739597ce8d86339d5a905', + 'md5': '90b26344ba442c8e44aa4cf8f301164a', 'info_dict': { 'id': '12820', 'ext': 'mp4', @@ -48,7 +48,8 @@ class EaglePlatformIE(InfoExtractor): 'skip': 'Georestricted', }] - def _handle_error(self, response): + @staticmethod + def _handle_error(response): status = int_or_none(response.get('status', 200)) if status != 200: raise ExtractorError(' '.join(response['errors']), expected=True) @@ -58,6 +59,9 @@ class EaglePlatformIE(InfoExtractor): self._handle_error(response) return response + def _get_video_url(self, url_or_request, video_id, note='Downloading JSON metadata'): + return self._download_json(url_or_request, video_id, note)['data'][0] + def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) host, video_id = mobj.group('custom_host') or mobj.group('host'), mobj.group('id') @@ -69,7 +73,7 @@ class EaglePlatformIE(InfoExtractor): title = media['title'] description = media.get('description') - thumbnail = media.get('snapshot') + thumbnail = self._proto_relative_url(media.get('snapshot'), 'http:') duration = int_or_none(media.get('duration')) view_count = int_or_none(media.get('views')) @@ -78,13 +82,20 @@ class EaglePlatformIE(InfoExtractor): if age_restriction: age_limit = 0 if age_restriction == 'allow_all' else 18 - m3u8_data = self._download_json( - self._proto_relative_url(media['sources']['secure_m3u8']['auto'], 'http:'), - video_id, 'Downloading m3u8 JSON') + secure_m3u8 = self._proto_relative_url(media['sources']['secure_m3u8']['auto'], 'http:') + m3u8_url = self._get_video_url(secure_m3u8, video_id, 'Downloading m3u8 JSON') formats = self._extract_m3u8_formats( - m3u8_data['data'][0], video_id, - 'mp4', entry_protocol='m3u8_native') + m3u8_url, video_id, + 'mp4', entry_protocol='m3u8_native', m3u8_id='hls') + + mp4_url = self._get_video_url( + # Secure mp4 URL is constructed according to Player.prototype.mp4 from + # http://lentaru.media.eagleplatform.com/player/player.js + re.sub(r'm3u8|hlsvod|hls|f4m', 'mp4', secure_m3u8), + video_id, 'Downloading mp4 JSON') + formats.append({'url': mp4_url, 'format_id': 'mp4'}) + self._sort_formats(formats) return { diff --git a/youtube_dl/extractor/eitb.py b/youtube_dl/extractor/eitb.py index 2cba82532..357a2196c 100644 --- a/youtube_dl/extractor/eitb.py +++ b/youtube_dl/extractor/eitb.py @@ -1,39 +1,92 @@ # encoding: utf-8 from __future__ import unicode_literals -import re - from .common import InfoExtractor -from .brightcove import BrightcoveIE -from ..utils import ExtractorError +from ..compat import compat_urllib_request +from ..utils import ( + float_or_none, + int_or_none, + parse_iso8601, +) class EitbIE(InfoExtractor): IE_NAME = 'eitb.tv' - _VALID_URL = r'https?://www\.eitb\.tv/(eu/bideoa|es/video)/[^/]+/(?P<playlist_id>\d+)/(?P<chapter_id>\d+)' + _VALID_URL = r'https?://(?:www\.)?eitb\.tv/(?:eu/bideoa|es/video)/[^/]+/\d+/(?P<id>\d+)' _TEST = { - 'add_ie': ['Brightcove'], - 'url': 'http://www.eitb.tv/es/video/60-minutos-60-minutos-2013-2014/2677100210001/2743577154001/lasa-y-zabala-30-anos/', + 'url': 'http://www.eitb.tv/es/video/60-minutos-60-minutos-2013-2014/4104995148001/4090227752001/lasa-y-zabala-30-anos/', 'md5': 'edf4436247185adee3ea18ce64c47998', 'info_dict': { - 'id': '2743577154001', + 'id': '4090227752001', 'ext': 'mp4', 'title': '60 minutos (Lasa y Zabala, 30 años)', - # All videos from eitb has this description in the brightcove info - 'description': '.', - 'uploader': 'Euskal Telebista', + 'description': 'Programa de reportajes de actualidad.', + 'duration': 3996.76, + 'timestamp': 1381789200, + 'upload_date': '20131014', + 'tags': list, }, } def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - chapter_id = mobj.group('chapter_id') - webpage = self._download_webpage(url, chapter_id) - bc_url = BrightcoveIE._extract_brightcove_url(webpage) - if bc_url is None: - raise ExtractorError('Could not extract the Brightcove url') - # The BrightcoveExperience object doesn't contain the video id, we set - # it manually - bc_url += '&%40videoPlayer={0}'.format(chapter_id) - return self.url_result(bc_url, BrightcoveIE.ie_key()) + video_id = self._match_id(url) + + video = self._download_json( + 'http://mam.eitb.eus/mam/REST/ServiceMultiweb/Video/MULTIWEBTV/%s/' % video_id, + video_id, 'Downloading video JSON') + + media = video['web_media'][0] + + formats = [] + for rendition in media['RENDITIONS']: + video_url = rendition.get('PMD_URL') + if not video_url: + continue + tbr = float_or_none(rendition.get('ENCODING_RATE'), 1000) + format_id = 'http' + if tbr: + format_id += '-%d' % int(tbr) + formats.append({ + 'url': rendition['PMD_URL'], + 'format_id': format_id, + 'width': int_or_none(rendition.get('FRAME_WIDTH')), + 'height': int_or_none(rendition.get('FRAME_HEIGHT')), + 'tbr': tbr, + }) + + hls_url = media.get('HLS_SURL') + if hls_url: + request = compat_urllib_request.Request( + 'http://mam.eitb.eus/mam/REST/ServiceMultiweb/DomainRestrictedSecurity/TokenAuth/', + headers={'Referer': url}) + token_data = self._download_json( + request, video_id, 'Downloading auth token', fatal=False) + if token_data: + token = token_data.get('token') + if token: + m3u8_formats = self._extract_m3u8_formats( + '%s?hdnts=%s' % (hls_url, token), video_id, m3u8_id='hls', fatal=False) + if m3u8_formats: + formats.extend(m3u8_formats) + + hds_url = media.get('HDS_SURL') + if hds_url: + f4m_formats = self._extract_f4m_formats( + '%s?hdcore=3.7.0' % hds_url.replace('euskalsvod', 'euskalvod'), + video_id, f4m_id='hds', fatal=False) + if f4m_formats: + formats.extend(f4m_formats) + + self._sort_formats(formats) + + return { + 'id': video_id, + 'title': media.get('NAME_ES') or media.get('name') or media['NAME_EU'], + 'description': media.get('SHORT_DESC_ES') or video.get('desc_group') or media.get('SHORT_DESC_EU'), + 'thumbnail': media.get('STILL_URL') or media.get('THUMBNAIL_URL'), + 'duration': float_or_none(media.get('LENGTH'), 1000), + 'timestamp': parse_iso8601(media.get('BROADCST_DATE'), ' '), + 'tags': media.get('TAGS'), + 'formats': formats, + } diff --git a/youtube_dl/extractor/engadget.py b/youtube_dl/extractor/engadget.py index 4ea37ebd9..e4180701d 100644 --- a/youtube_dl/extractor/engadget.py +++ b/youtube_dl/extractor/engadget.py @@ -10,7 +10,7 @@ from ..utils import ( class EngadgetIE(InfoExtractor): _VALID_URL = r'''(?x)https?://www.engadget.com/ - (?:video/5min/(?P<id>\d+)| + (?:video(?:/5min)?/(?P<id>\d+)| [\d/]+/.*?) ''' diff --git a/youtube_dl/extractor/europa.py b/youtube_dl/extractor/europa.py new file mode 100644 index 000000000..adc43919e --- /dev/null +++ b/youtube_dl/extractor/europa.py @@ -0,0 +1,93 @@ +# coding: utf-8 +from __future__ import unicode_literals + +from .common import InfoExtractor +from ..compat import compat_urlparse +from ..utils import ( + int_or_none, + orderedSet, + parse_duration, + qualities, + unified_strdate, + xpath_text +) + + +class EuropaIE(InfoExtractor): + _VALID_URL = r'https?://ec\.europa\.eu/avservices/(?:video/player|audio/audioDetails)\.cfm\?.*?\bref=(?P<id>[A-Za-z0-9-]+)' + _TESTS = [{ + 'url': 'http://ec.europa.eu/avservices/video/player.cfm?ref=I107758', + 'md5': '574f080699ddd1e19a675b0ddf010371', + 'info_dict': { + 'id': 'I107758', + 'ext': 'mp4', + 'title': 'TRADE - Wikileaks on TTIP', + 'description': 'NEW LIVE EC Midday press briefing of 11/08/2015', + 'thumbnail': 're:^https?://.*\.jpg$', + 'upload_date': '20150811', + 'duration': 34, + 'view_count': int, + 'formats': 'mincount:3', + } + }, { + 'url': 'http://ec.europa.eu/avservices/video/player.cfm?sitelang=en&ref=I107786', + 'only_matching': True, + }, { + 'url': 'http://ec.europa.eu/avservices/audio/audioDetails.cfm?ref=I-109295&sitelang=en', + 'only_matching': True, + }] + + def _real_extract(self, url): + video_id = self._match_id(url) + + playlist = self._download_xml( + 'http://ec.europa.eu/avservices/video/player/playlist.cfm?ID=%s' % video_id, video_id) + + def get_item(type_, preference): + items = {} + for item in playlist.findall('./info/%s/item' % type_): + lang, label = xpath_text(item, 'lg', default=None), xpath_text(item, 'label', default=None) + if lang and label: + items[lang] = label.strip() + for p in preference: + if items.get(p): + return items[p] + + query = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query) + preferred_lang = query.get('sitelang', ('en', ))[0] + + preferred_langs = orderedSet((preferred_lang, 'en', 'int')) + + title = get_item('title', preferred_langs) or video_id + description = get_item('description', preferred_langs) + thumbnmail = xpath_text(playlist, './info/thumburl', 'thumbnail') + upload_date = unified_strdate(xpath_text(playlist, './info/date', 'upload date')) + duration = parse_duration(xpath_text(playlist, './info/duration', 'duration')) + view_count = int_or_none(xpath_text(playlist, './info/views', 'views')) + + language_preference = qualities(preferred_langs[::-1]) + + formats = [] + for file_ in playlist.findall('./files/file'): + video_url = xpath_text(file_, './url') + if not video_url: + continue + lang = xpath_text(file_, './lg') + formats.append({ + 'url': video_url, + 'format_id': lang, + 'format_note': xpath_text(file_, './lglabel'), + 'language_preference': language_preference(lang) + }) + self._sort_formats(formats) + + return { + 'id': video_id, + 'title': title, + 'description': description, + 'thumbnail': thumbnmail, + 'upload_date': upload_date, + 'duration': duration, + 'view_count': view_count, + 'formats': formats + } diff --git a/youtube_dl/extractor/expotv.py b/youtube_dl/extractor/expotv.py index a38b773e8..1585a03bb 100644 --- a/youtube_dl/extractor/expotv.py +++ b/youtube_dl/extractor/expotv.py @@ -33,20 +33,27 @@ class ExpoTVIE(InfoExtractor): webpage = self._download_webpage(url, video_id) player_key = self._search_regex( r'<param name="playerKey" value="([^"]+)"', webpage, 'player key') - config_url = 'http://client.expotv.com/video/config/%s/%s' % ( - video_id, player_key) config = self._download_json( - config_url, video_id, - note='Downloading video configuration') + 'http://client.expotv.com/video/config/%s/%s' % (video_id, player_key), + video_id, 'Downloading video configuration') - formats = [{ - 'url': fcfg['file'], - 'height': int_or_none(fcfg.get('height')), - 'format_note': fcfg.get('label'), - 'ext': self._search_regex( - r'filename=.*\.([a-z0-9_A-Z]+)&', fcfg['file'], - 'file extension', default=None), - } for fcfg in config['sources']] + formats = [] + for fcfg in config['sources']: + media_url = fcfg.get('file') + if not media_url: + continue + if fcfg.get('type') == 'm3u8': + formats.extend(self._extract_m3u8_formats( + media_url, video_id, 'mp4', entry_protocol='m3u8_native', m3u8_id='hls')) + else: + formats.append({ + 'url': media_url, + 'height': int_or_none(fcfg.get('height')), + 'format_id': fcfg.get('label'), + 'ext': self._search_regex( + r'filename=.*\.([a-z0-9_A-Z]+)&', media_url, + 'file extension', default=None) or fcfg.get('type'), + }) self._sort_formats(formats) title = self._og_search_title(webpage) diff --git a/youtube_dl/extractor/extremetube.py b/youtube_dl/extractor/extremetube.py index c826a5404..c5677c82b 100644 --- a/youtube_dl/extractor/extremetube.py +++ b/youtube_dl/extractor/extremetube.py @@ -3,23 +3,20 @@ from __future__ import unicode_literals import re from .common import InfoExtractor -from ..compat import ( - compat_parse_qs, - compat_urllib_request, -) +from ..compat import compat_urllib_request from ..utils import ( - qualities, + int_or_none, str_to_int, ) class ExtremeTubeIE(InfoExtractor): - _VALID_URL = r'https?://(?:www\.)?(?P<url>extremetube\.com/.*?video/.+?(?P<id>[0-9]+))(?:[/?&]|$)' + _VALID_URL = r'https?://(?:www\.)?extremetube\.com/(?:[^/]+/)?video/(?P<id>[^/#?&]+)' _TESTS = [{ 'url': 'http://www.extremetube.com/video/music-video-14-british-euro-brit-european-cumshots-swallow-652431', 'md5': '344d0c6d50e2f16b06e49ca011d8ac69', 'info_dict': { - 'id': '652431', + 'id': 'music-video-14-british-euro-brit-european-cumshots-swallow-652431', 'ext': 'mp4', 'title': 'Music Video 14 british euro brit european cumshots swallow', 'uploader': 'unknown', @@ -29,12 +26,16 @@ class ExtremeTubeIE(InfoExtractor): }, { 'url': 'http://www.extremetube.com/gay/video/abcde-1234', 'only_matching': True, + }, { + 'url': 'http://www.extremetube.com/video/latina-slut-fucked-by-fat-black-dick', + 'only_matching': True, + }, { + 'url': 'http://www.extremetube.com/video/652431', + 'only_matching': True, }] def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - video_id = mobj.group('id') - url = 'http://www.' + mobj.group('url') + video_id = self._match_id(url) req = compat_urllib_request.Request(url) req.add_header('Cookie', 'age_verified=1') @@ -49,20 +50,36 @@ class ExtremeTubeIE(InfoExtractor): r'Views:\s*</strong>\s*<span>([\d,\.]+)</span>', webpage, 'view count', fatal=False)) - flash_vars = compat_parse_qs(self._search_regex( - r'<param[^>]+?name="flashvars"[^>]+?value="([^"]+)"', webpage, 'flash vars')) + flash_vars = self._parse_json( + self._search_regex( + r'var\s+flashvars\s*=\s*({.+?});', webpage, 'flash vars'), + video_id) formats = [] - quality = qualities(['180p', '240p', '360p', '480p', '720p', '1080p']) - for k, vals in flash_vars.items(): - m = re.match(r'quality_(?P<quality>[0-9]+p)$', k) - if m is not None: - formats.append({ - 'format_id': m.group('quality'), - 'quality': quality(m.group('quality')), - 'url': vals[0], + for quality_key, video_url in flash_vars.items(): + height = int_or_none(self._search_regex( + r'quality_(\d+)[pP]$', quality_key, 'height', default=None)) + if not height: + continue + f = { + 'url': video_url, + } + mobj = re.search( + r'/(?P<height>\d{3,4})[pP]_(?P<bitrate>\d+)[kK]_\d+', video_url) + if mobj: + height = int(mobj.group('height')) + bitrate = int(mobj.group('bitrate')) + f.update({ + 'format_id': '%dp-%dk' % (height, bitrate), + 'height': height, + 'tbr': bitrate, }) - + else: + f.update({ + 'format_id': '%dp' % height, + 'height': height, + }) + formats.append(f) self._sort_formats(formats) return { diff --git a/youtube_dl/extractor/facebook.py b/youtube_dl/extractor/facebook.py index 178a7ca4c..f53c51615 100644 --- a/youtube_dl/extractor/facebook.py +++ b/youtube_dl/extractor/facebook.py @@ -14,7 +14,6 @@ from ..compat import ( ) from ..utils import ( ExtractorError, - int_or_none, limit_length, urlencode_postdata, get_element_by_id, @@ -142,16 +141,20 @@ class FacebookIE(InfoExtractor): data = dict(json.loads(m.group(1))) params_raw = compat_urllib_parse_unquote(data['params']) params = json.loads(params_raw) - video_data = params['video_data'][0] formats = [] - for quality in ['sd', 'hd']: - src = video_data.get('%s_src' % quality) - if src is not None: - formats.append({ - 'format_id': quality, - 'url': src, - }) + for format_id, f in params['video_data'].items(): + if not f or not isinstance(f, list): + continue + for quality in ('sd', 'hd'): + for src_type in ('src', 'src_no_ratelimit'): + src = f[0].get('%s_%s' % (quality, src_type)) + if src: + formats.append({ + 'format_id': '%s_%s_%s' % (format_id, quality, src_type), + 'url': src, + 'preference': -10 if format_id == 'progressive' else 0, + }) if not formats: raise ExtractorError('Cannot find video formats') @@ -171,7 +174,5 @@ class FacebookIE(InfoExtractor): 'id': video_id, 'title': video_title, 'formats': formats, - 'duration': int_or_none(video_data.get('video_duration')), - 'thumbnail': video_data.get('thumbnail_src'), 'uploader': uploader, } diff --git a/youtube_dl/extractor/fczenit.py b/youtube_dl/extractor/fczenit.py new file mode 100644 index 000000000..f1f150ef2 --- /dev/null +++ b/youtube_dl/extractor/fczenit.py @@ -0,0 +1,41 @@ +# coding: utf-8 +from __future__ import unicode_literals + +import re + +from .common import InfoExtractor + + +class FczenitIE(InfoExtractor): + _VALID_URL = r'https?://(?:www\.)?fc-zenit\.ru/video/gl(?P<id>[0-9]+)' + _TEST = { + 'url': 'http://fc-zenit.ru/video/gl6785/', + 'md5': '458bacc24549173fe5a5aa29174a5606', + 'info_dict': { + 'id': '6785', + 'ext': 'mp4', + 'title': '«Зенит-ТВ»: как Олег Шатов играл против «Урала»', + }, + } + + def _real_extract(self, url): + video_id = self._match_id(url) + webpage = self._download_webpage(url, video_id) + + video_title = self._html_search_regex(r'<div class=\"photoalbum__title\">([^<]+)', webpage, 'title') + + bitrates_raw = self._html_search_regex(r'bitrates:.*\n(.*)\]', webpage, 'video URL') + bitrates = re.findall(r'url:.?\'(.+?)\'.*?bitrate:.?([0-9]{3}?)', bitrates_raw) + + formats = [{ + 'url': furl, + 'tbr': tbr, + } for furl, tbr in bitrates] + + self._sort_formats(formats) + + return { + 'id': video_id, + 'title': video_title, + 'formats': formats, + } diff --git a/youtube_dl/extractor/fivemin.py b/youtube_dl/extractor/fivemin.py index 157094e8c..2955965d9 100644 --- a/youtube_dl/extractor/fivemin.py +++ b/youtube_dl/extractor/fivemin.py @@ -2,11 +2,15 @@ from __future__ import unicode_literals from .common import InfoExtractor from ..compat import ( - compat_str, compat_urllib_parse, + compat_parse_qs, + compat_urllib_parse_urlparse, + compat_urlparse, ) from ..utils import ( ExtractorError, + parse_duration, + replace_extension, ) @@ -28,6 +32,7 @@ class FiveMinIE(InfoExtractor): 'id': '518013791', 'ext': 'mp4', 'title': 'iPad Mini with Retina Display Review', + 'duration': 177, }, }, { @@ -38,9 +43,52 @@ class FiveMinIE(InfoExtractor): 'id': '518086247', 'ext': 'mp4', 'title': 'How to Make a Next-Level Fruit Salad', + 'duration': 184, }, }, ] + _ERRORS = { + 'ErrorVideoNotExist': 'We\'re sorry, but the video you are trying to watch does not exist.', + 'ErrorVideoNoLongerAvailable': 'We\'re sorry, but the video you are trying to watch is no longer available.', + 'ErrorVideoRejected': 'We\'re sorry, but the video you are trying to watch has been removed.', + 'ErrorVideoUserNotGeo': 'We\'re sorry, but the video you are trying to watch cannot be viewed from your current location.', + 'ErrorVideoLibraryRestriction': 'We\'re sorry, but the video you are trying to watch is currently unavailable for viewing at this domain.', + 'ErrorExposurePermission': 'We\'re sorry, but the video you are trying to watch is currently unavailable for viewing at this domain.', + } + _QUALITIES = { + 1: { + 'width': 640, + 'height': 360, + }, + 2: { + 'width': 854, + 'height': 480, + }, + 4: { + 'width': 1280, + 'height': 720, + }, + 8: { + 'width': 1920, + 'height': 1080, + }, + 16: { + 'width': 640, + 'height': 360, + }, + 32: { + 'width': 854, + 'height': 480, + }, + 64: { + 'width': 1280, + 'height': 720, + }, + 128: { + 'width': 640, + 'height': 360, + }, + } def _real_extract(self, url): video_id = self._match_id(url) @@ -59,26 +107,36 @@ class FiveMinIE(InfoExtractor): 'https://syn.5min.com/handlers/SenseHandler.ashx?' + query, video_id) if not response['success']: - err_msg = response['errorMessage'] - if err_msg == 'ErrorVideoUserNotGeo': - msg = 'Video not available from your location' - else: - msg = 'Aol said: %s' % err_msg - raise ExtractorError(msg, expected=True, video_id=video_id) + raise ExtractorError( + '%s said: %s' % ( + self.IE_NAME, + self._ERRORS.get(response['errorMessage'], response['errorMessage'])), + expected=True) info = response['binding'][0] - second_id = compat_str(int(video_id[:-2]) + 1) formats = [] - for quality, height in [(1, 320), (2, 480), (4, 720), (8, 1080)]: - if any(r['ID'] == quality for r in info['Renditions']): + parsed_video_url = compat_urllib_parse_urlparse(compat_parse_qs( + compat_urllib_parse_urlparse(info['EmbededURL']).query)['videoUrl'][0]) + for rendition in info['Renditions']: + if rendition['RenditionType'] == 'm3u8': + formats.extend(self._extract_m3u8_formats(rendition['Url'], video_id, m3u8_id='hls')) + elif rendition['RenditionType'] == 'aac': + continue + else: + rendition_url = compat_urlparse.urlunparse(parsed_video_url._replace(path=replace_extension(parsed_video_url.path.replace('//', '/%s/' % rendition['ID']), rendition['RenditionType']))) + quality = self._QUALITIES.get(rendition['ID'], {}) formats.append({ - 'format_id': compat_str(quality), - 'url': 'http://avideos.5min.com/%s/%s/%s_%s.mp4' % (second_id[-3:], second_id, video_id, quality), - 'height': height, + 'format_id': '%s-%d' % (rendition['RenditionType'], rendition['ID']), + 'url': rendition_url, + 'width': quality.get('width'), + 'height': quality.get('height'), }) + self._sort_formats(formats) return { 'id': video_id, 'title': info['Title'], + 'thumbnail': info.get('ThumbURL'), + 'duration': parse_duration(info.get('Duration')), 'formats': formats, } diff --git a/youtube_dl/extractor/fktv.py b/youtube_dl/extractor/fktv.py index 190d9f9ad..40ea27895 100644 --- a/youtube_dl/extractor/fktv.py +++ b/youtube_dl/extractor/fktv.py @@ -1,13 +1,12 @@ from __future__ import unicode_literals import re -import random -import json from .common import InfoExtractor from ..utils import ( - get_element_by_id, clean_html, + determine_ext, + ExtractorError, ) @@ -17,66 +16,40 @@ class FKTVIE(InfoExtractor): _TEST = { 'url': 'http://fernsehkritik.tv/folge-1', + 'md5': '21f0b0c99bce7d5b524eb1b17b1c6d79', 'info_dict': { - 'id': '00011', - 'ext': 'flv', + 'id': '1', + 'ext': 'mp4', 'title': 'Folge 1 vom 10. April 2007', - 'description': 'md5:fb4818139c7cfe6907d4b83412a6864f', + 'thumbnail': 're:^https?://.*\.jpg$', }, } def _real_extract(self, url): - episode = int(self._match_id(url)) - - video_thumbnail = 'http://fernsehkritik.tv/images/magazin/folge%s.jpg' % episode - start_webpage = self._download_webpage('http://fernsehkritik.tv/folge-%s/Start' % episode, - episode) - playlist = self._search_regex(r'playlist = (\[.*?\]);', start_webpage, - 'playlist', flags=re.DOTALL) - files = json.loads(re.sub('{[^{}]*?}', '{}', playlist)) - - videos = [] - for i, _ in enumerate(files, 1): - video_id = '%04d%d' % (episode, i) - video_url = 'http://fernsehkritik.tv/js/directme.php?file=%s%s.flv' % (episode, '' if i == 1 else '-%d' % i) - videos.append({ - 'ext': 'flv', - 'id': video_id, - 'url': video_url, - 'title': clean_html(get_element_by_id('eptitle', start_webpage)), - 'description': clean_html(get_element_by_id('contentlist', start_webpage)), - 'thumbnail': video_thumbnail - }) - return { - '_type': 'multi_video', - 'entries': videos, - 'id': 'folge-%s' % episode, - } - - -class FKTVPosteckeIE(InfoExtractor): - IE_NAME = 'fernsehkritik.tv:postecke' - _VALID_URL = r'http://(?:www\.)?fernsehkritik\.tv/inline-video/postecke\.php\?(.*&)?ep=(?P<ep>[0-9]+)(&|$)' - _TEST = { - 'url': 'http://fernsehkritik.tv/inline-video/postecke.php?iframe=true&width=625&height=440&ep=120', - 'md5': '262f0adbac80317412f7e57b4808e5c4', - 'info_dict': { - 'id': '0120', - 'ext': 'flv', - 'title': 'Postecke 120', - } - } - - def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - episode = int(mobj.group('ep')) - - server = random.randint(2, 4) - video_id = '%04d' % episode - video_url = 'http://dl%d.fernsehkritik.tv/postecke/postecke%d.flv' % (server, episode) - video_title = 'Postecke %d' % episode + episode = self._match_id(url) + + webpage = self._download_webpage( + 'http://fernsehkritik.tv/folge-%s/play' % episode, episode) + title = clean_html(self._html_search_regex( + '<h3>([^<]+)</h3>', webpage, 'title')) + matches = re.search( + r'(?s)<video(?:(?!poster)[^>])+(?:poster="([^"]+)")?[^>]*>(.*)</video>', + webpage) + if matches is None: + raise ExtractorError('Unable to extract the video') + + poster, sources = matches.groups() + if poster is None: + self.report_warning('unable to extract thumbnail') + + urls = re.findall(r'<source[^>]+src="([^"]+)"', sources) + formats = [{ + 'url': furl, + 'format_id': determine_ext(furl), + } for furl in urls] return { - 'id': video_id, - 'url': video_url, - 'title': video_title, + 'id': episode, + 'title': title, + 'formats': formats, + 'thumbnail': poster, } diff --git a/youtube_dl/extractor/fourtube.py b/youtube_dl/extractor/fourtube.py index 3bb4f6239..fb6d108c0 100644 --- a/youtube_dl/extractor/fourtube.py +++ b/youtube_dl/extractor/fourtube.py @@ -46,10 +46,10 @@ class FourTubeIE(InfoExtractor): thumbnail = self._html_search_meta('thumbnailUrl', webpage) uploader_id = self._html_search_regex( r'<a class="img-avatar" href="[^"]+/channels/([^/"]+)" title="Go to [^"]+ page">', - webpage, 'uploader id') + webpage, 'uploader id', fatal=False) uploader = self._html_search_regex( r'<a class="img-avatar" href="[^"]+/channels/[^/"]+" title="Go to ([^"]+) page">', - webpage, 'uploader') + webpage, 'uploader', fatal=False) categories_html = self._search_regex( r'(?s)><i class="icon icon-tag"></i>\s*Categories / Tags\s*.*?<ul class="list">(.*?)</ul>', @@ -68,13 +68,24 @@ class FourTubeIE(InfoExtractor): webpage, 'like count', fatal=False)) duration = parse_duration(self._html_search_meta('duration', webpage)) - params_js = self._search_regex( - r'\$\.ajax\(url,\ opts\);\s*\}\s*\}\)\(([0-9,\[\] ]+)\)', - webpage, 'initialization parameters' - ) - params = self._parse_json('[%s]' % params_js, video_id) - media_id = params[0] - sources = ['%s' % p for p in params[2]] + media_id = self._search_regex( + r'<button[^>]+data-id=(["\'])(?P<id>\d+)\1[^>]+data-quality=', webpage, + 'media id', default=None, group='id') + sources = [ + quality + for _, quality in re.findall(r'<button[^>]+data-quality=(["\'])(.+?)\1', webpage)] + if not (media_id and sources): + player_js = self._download_webpage( + self._search_regex( + r'<script[^>]id=(["\'])playerembed\1[^>]+src=(["\'])(?P<url>.+?)\2', + webpage, 'player JS', group='url'), + video_id, 'Downloading player JS') + params_js = self._search_regex( + r'\$\.ajax\(url,\ opts\);\s*\}\s*\}\)\(([0-9,\[\] ]+)\)', + player_js, 'initialization parameters') + params = self._parse_json('[%s]' % params_js, video_id) + media_id = params[0] + sources = ['%s' % p for p in params[2]] token_url = 'http://tkn.4tube.com/{0}/desktop/{1}'.format( media_id, '+'.join(sources)) diff --git a/youtube_dl/extractor/francetv.py b/youtube_dl/extractor/francetv.py index 129984a5f..8e60cf60f 100644 --- a/youtube_dl/extractor/francetv.py +++ b/youtube_dl/extractor/francetv.py @@ -83,6 +83,14 @@ class FranceTVBaseInfoExtractor(InfoExtractor): if subtitle: title += ' - %s' % subtitle + subtitles = {} + subtitles_list = [{ + 'url': subformat['url'], + 'ext': subformat.get('format'), + } for subformat in info.get('subtitles', []) if subformat.get('url')] + if subtitles_list: + subtitles['fr'] = subtitles_list + return { 'id': video_id, 'title': title, @@ -91,20 +99,27 @@ class FranceTVBaseInfoExtractor(InfoExtractor): 'duration': int_or_none(info.get('real_duration')) or parse_duration(info['duree']), 'timestamp': int_or_none(info['diffusion']['timestamp']), 'formats': formats, + 'subtitles': subtitles, } class PluzzIE(FranceTVBaseInfoExtractor): IE_NAME = 'pluzz.francetv.fr' - _VALID_URL = r'https?://pluzz\.francetv\.fr/videos/(.*?)\.html' + _VALID_URL = r'https?://(?:m\.)?pluzz\.francetv\.fr/videos/(?P<id>.+?)\.html' # Can't use tests, videos expire in 7 days def _real_extract(self, url): - title = re.match(self._VALID_URL, url).group(1) - webpage = self._download_webpage(url, title) - video_id = self._search_regex( - r'data-diffusion="(\d+)"', webpage, 'ID') + display_id = self._match_id(url) + + webpage = self._download_webpage(url, display_id) + + video_id = self._html_search_meta( + 'id_video', webpage, 'video id', default=None) + if not video_id: + video_id = self._search_regex( + r'data-diffusion=["\'](\d+)', webpage, 'video id') + return self._extract_video(video_id, 'Pluzz') @@ -120,6 +135,9 @@ class FranceTvInfoIE(FranceTVBaseInfoExtractor): 'title': 'Soir 3', 'upload_date': '20130826', 'timestamp': 1377548400, + 'subtitles': { + 'fr': 'mincount:2', + }, }, }, { 'url': 'http://www.francetvinfo.fr/elections/europeennes/direct-europeennes-regardez-le-debat-entre-les-candidats-a-la-presidence-de-la-commission_600639.html', diff --git a/youtube_dl/extractor/funnyordie.py b/youtube_dl/extractor/funnyordie.py index f5f13689c..7f21d7410 100644 --- a/youtube_dl/extractor/funnyordie.py +++ b/youtube_dl/extractor/funnyordie.py @@ -45,11 +45,20 @@ class FunnyOrDieIE(InfoExtractor): links.sort(key=lambda link: 1 if link[1] == 'mp4' else 0) - bitrates = self._html_search_regex(r'<source src="[^"]+/v,((?:\d+,)+)\.mp4\.csmil', webpage, 'video bitrates') - bitrates = [int(b) for b in bitrates.rstrip(',').split(',')] - bitrates.sort() + m3u8_url = self._search_regex( + r'<source[^>]+src=(["\'])(?P<url>.+?/master\.m3u8)\1', + webpage, 'm3u8 url', default=None, group='url') formats = [] + + m3u8_formats = self._extract_m3u8_formats( + m3u8_url, video_id, 'mp4', 'm3u8_native', m3u8_id='hls', fatal=False) + if m3u8_formats: + formats.extend(m3u8_formats) + + bitrates = [int(bitrate) for bitrate in re.findall(r'[,/]v(\d+)[,/]', m3u8_url)] + bitrates.sort() + for bitrate in bitrates: for link in links: formats.append({ diff --git a/youtube_dl/extractor/generic.py b/youtube_dl/extractor/generic.py index 7a3a7f66b..34d930a2d 100644 --- a/youtube_dl/extractor/generic.py +++ b/youtube_dl/extractor/generic.py @@ -4,10 +4,12 @@ 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, @@ -20,7 +22,6 @@ from ..utils import ( HEADRequest, is_html, orderedSet, - parse_xml, smuggle_url, unescapeHTML, unified_strdate, @@ -52,6 +53,7 @@ from .dailymotion import DailymotionCloudIE from .onionstudios import OnionStudiosIE from .snagfilms import SnagFilmsEmbedIE from .screenwavemedia import ScreenwaveMediaIE +from .mtv import MTVServicesEmbeddedIE class GenericIE(InfoExtractor): @@ -142,6 +144,7 @@ class GenericIE(InfoExtractor): 'ext': 'mp4', 'title': 'Automatics, robotics and biocybernetics', 'description': 'md5:815fc1deb6b3a2bff99de2d5325be482', + 'upload_date': '20130627', 'formats': 'mincount:16', 'subtitles': 'mincount:1', }, @@ -234,6 +237,22 @@ class GenericIE(InfoExtractor): } }, { + # 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', 'info_dict': { @@ -1233,7 +1252,7 @@ class GenericIE(InfoExtractor): # Is it an RSS feed, a SMIL file or a XSPF playlist? 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): @@ -1613,12 +1632,9 @@ class GenericIE(InfoExtractor): return self.url_result(url, ie='Vulture') # Look for embedded mtvservices player - mobj = re.search( - r'<iframe src="(?P<url>https?://media\.mtvnservices\.com/embed/[^"]+)"', - webpage) - if mobj is not None: - url = unescapeHTML(mobj.group('url')) - return self.url_result(url, ie='MTVServicesEmbedded') + mtvservices_url = MTVServicesEmbeddedIE._extract_url(webpage) + if mtvservices_url: + return self.url_result(mtvservices_url, ie='MTVServicesEmbedded') # Look for embedded yahoo player mobj = re.search( @@ -1657,7 +1673,7 @@ class GenericIE(InfoExtractor): return self.url_result(mobj.group('url'), 'MLB') mobj = re.search( - r'<iframe[^>]+?src=(["\'])(?P<url>%s)\1' % CondeNastIE.EMBED_URL, + r'<(?:iframe|script)[^>]+?src=(["\'])(?P<url>%s)\1' % CondeNastIE.EMBED_URL, webpage) if mobj is not None: return self.url_result(self._proto_relative_url(mobj.group('url'), scheme='http:'), 'CondeNast') @@ -1675,8 +1691,8 @@ class GenericIE(InfoExtractor): return self.url_result(mobj.group('url'), 'Zapiks') # Look for Kaltura embeds - mobj = (re.search(r"(?s)kWidget\.(?:thumb)?[Ee]mbed\(\{.*?'wid'\s*:\s*'_?(?P<partner_id>[^']+)',.*?'entry_id'\s*:\s*'(?P<id>[^']+)',", webpage) or - re.search(r'(?s)(["\'])(?:https?:)?//cdnapisec\.kaltura\.com/.*?(?:p|partner_id)/(?P<partner_id>\d+).*?\1.*?entry_id\s*:\s*(["\'])(?P<id>[^\2]+?)\2', webpage)) + mobj = (re.search(r"(?s)kWidget\.(?:thumb)?[Ee]mbed\(\{.*?'wid'\s*:\s*'_?(?P<partner_id>[^']+)',.*?'entry_?[Ii]d'\s*:\s*'(?P<id>[^']+)',", webpage) or + re.search(r'(?s)(?P<q1>["\'])(?:https?:)?//cdnapi(?:sec)?\.kaltura\.com/.*?(?:p|partner_id)/(?P<partner_id>\d+).*?(?P=q1).*?entry_?[Ii]d\s*:\s*(?P<q2>["\'])(?P<id>.+?)(?P=q2)', webpage)) if mobj is not None: return self.url_result('kaltura:%(partner_id)s:%(id)s' % mobj.groupdict(), 'Kaltura') @@ -1827,6 +1843,9 @@ class GenericIE(InfoExtractor): # Look also in Refresh HTTP header refresh_header = head_response.headers.get('Refresh') if refresh_header: + # In python 2 response HTTP headers are bytestrings + if sys.version_info < (3, 0) and isinstance(refresh_header, str): + refresh_header = refresh_header.decode('iso-8859-1') found = re.search(REDIRECT_REGEX, refresh_header) if found: new_url = compat_urlparse.urljoin(url, unescapeHTML(found.group(1))) diff --git a/youtube_dl/extractor/globo.py b/youtube_dl/extractor/globo.py index 33d6432a6..c65ef6bcf 100644 --- a/youtube_dl/extractor/globo.py +++ b/youtube_dl/extractor/globo.py @@ -14,79 +14,58 @@ from ..utils import ( ExtractorError, float_or_none, int_or_none, + str_or_none, ) class GloboIE(InfoExtractor): - _VALID_URL = 'https?://.+?\.globo\.com/(?P<id>.+)' + _VALID_URL = '(?:globo:|https?://.+?\.globo\.com/(?:[^/]+/)*(?:v/(?:[^/]+/)?|videos/))(?P<id>\d{7,})' _API_URL_TEMPLATE = 'http://api.globovideos.com/videos/%s/playlist' _SECURITY_URL_TEMPLATE = 'http://security.video.globo.com/videos/%s/hash?player=flash&version=17.0.0.132&resource_id=%s' - _VIDEOID_REGEXES = [ - r'\bdata-video-id="(\d+)"', - r'\bdata-player-videosids="(\d+)"', - r'<div[^>]+\bid="(\d+)"', - ] - _RESIGN_EXPIRATION = 86400 - _TESTS = [ - { - 'url': 'http://globotv.globo.com/sportv/futebol-nacional/v/os-gols-de-atletico-mg-3-x-2-santos-pela-24a-rodada-do-brasileirao/3654973/', - 'md5': '03ebf41cb7ade43581608b7d9b71fab0', - 'info_dict': { - 'id': '3654973', - 'ext': 'mp4', - 'title': 'Os gols de Atlético-MG 3 x 2 Santos pela 24ª rodada do Brasileirão', - 'duration': 251.585, - 'uploader': 'SporTV', - 'uploader_id': 698, - 'like_count': int, - } - }, - { - 'url': 'http://g1.globo.com/carros/autoesporte/videos/t/exclusivos-do-g1/v/mercedes-benz-gla-passa-por-teste-de-colisao-na-europa/3607726/', - 'md5': 'b3ccc801f75cd04a914d51dadb83a78d', - 'info_dict': { - 'id': '3607726', - 'ext': 'mp4', - 'title': 'Mercedes-Benz GLA passa por teste de colisão na Europa', - 'duration': 103.204, - 'uploader': 'Globo.com', - 'uploader_id': 265, - 'like_count': int, - } - }, - { - 'url': 'http://g1.globo.com/jornal-nacional/noticia/2014/09/novidade-na-fiscalizacao-de-bagagem-pela-receita-provoca-discussoes.html', - 'md5': '307fdeae4390ccfe6ba1aa198cf6e72b', - 'info_dict': { - 'id': '3652183', - 'ext': 'mp4', - 'title': 'Receita Federal explica como vai fiscalizar bagagens de quem retorna ao Brasil de avião', - 'duration': 110.711, - 'uploader': 'Rede Globo', - 'uploader_id': 196, - 'like_count': int, - } + _TESTS = [{ + 'url': 'http://g1.globo.com/carros/autoesporte/videos/t/exclusivos-do-g1/v/mercedes-benz-gla-passa-por-teste-de-colisao-na-europa/3607726/', + 'md5': 'b3ccc801f75cd04a914d51dadb83a78d', + 'info_dict': { + 'id': '3607726', + 'ext': 'mp4', + 'title': 'Mercedes-Benz GLA passa por teste de colisão na Europa', + 'duration': 103.204, + 'uploader': 'Globo.com', + 'uploader_id': '265', }, - { - 'url': 'http://globotv.globo.com/canal-brasil/sangue-latino/t/todos-os-videos/v/ator-e-diretor-argentino-ricado-darin-fala-sobre-utopias-e-suas-perdas/3928201/', - 'md5': 'c1defca721ce25b2354e927d3e4b3dec', - 'info_dict': { - 'id': '3928201', - 'ext': 'mp4', - 'title': 'Ator e diretor argentino, Ricado Darín fala sobre utopias e suas perdas', - 'duration': 1472.906, - 'uploader': 'Canal Brasil', - 'uploader_id': 705, - 'like_count': int, - } + }, { + 'url': 'http://globoplay.globo.com/v/4581987/', + 'md5': 'f36a1ecd6a50da1577eee6dd17f67eff', + 'info_dict': { + 'id': '4581987', + 'ext': 'mp4', + 'title': 'Acidentes de trânsito estão entre as maiores causas de queda de energia em SP', + 'duration': 137.973, + 'uploader': 'Rede Globo', + 'uploader_id': '196', }, - ] - - class MD5(): + }, { + 'url': 'http://canalbrasil.globo.com/programas/sangue-latino/videos/3928201.html', + 'only_matching': True, + }, { + 'url': 'http://globosatplay.globo.com/globonews/v/4472924/', + 'only_matching': True, + }, { + 'url': 'http://globotv.globo.com/t/programa/v/clipe-sexo-e-as-negas-adeus/3836166/', + 'only_matching': True, + }, { + 'url': 'http://globotv.globo.com/canal-brasil/sangue-latino/t/todos-os-videos/v/ator-e-diretor-argentino-ricado-darin-fala-sobre-utopias-e-suas-perdas/3928201/', + 'only_matching': True, + }, { + 'url': 'http://canaloff.globo.com/programas/desejar-profundo/videos/4518560.html', + 'only_matching': True, + }] + + class MD5: HEX_FORMAT_LOWERCASE = 0 HEX_FORMAT_UPPERCASE = 1 BASE64_PAD_CHARACTER_DEFAULT_COMPLIANCE = '' @@ -353,9 +332,6 @@ class GloboIE(InfoExtractor): def _real_extract(self, url): video_id = self._match_id(url) - webpage = self._download_webpage(url, video_id) - video_id = self._search_regex(self._VIDEOID_REGEXES, webpage, 'video id') - video = self._download_json( self._API_URL_TEMPLATE % video_id, video_id)['videos'][0] @@ -364,7 +340,7 @@ class GloboIE(InfoExtractor): formats = [] for resource in video['resources']: resource_id = resource.get('_id') - if not resource_id: + if not resource_id or resource_id.endswith('manifest'): continue security = self._download_json( @@ -393,20 +369,23 @@ class GloboIE(InfoExtractor): resource_url = resource['url'] signed_url = '%s?h=%s&k=%s' % (resource_url, signed_hash, 'flash') if resource_id.endswith('m3u8') or resource_url.endswith('.m3u8'): - formats.extend(self._extract_m3u8_formats(signed_url, resource_id, 'mp4')) + m3u8_formats = self._extract_m3u8_formats( + signed_url, resource_id, 'mp4', entry_protocol='m3u8_native', + m3u8_id='hls', fatal=False) + if m3u8_formats: + formats.extend(m3u8_formats) else: formats.append({ 'url': signed_url, - 'format_id': resource_id, - 'height': resource.get('height'), + 'format_id': 'http-%s' % resource_id, + 'height': int_or_none(resource.get('height')), }) self._sort_formats(formats) duration = float_or_none(video.get('duration'), 1000) - like_count = int_or_none(video.get('likes')) uploader = video.get('channel') - uploader_id = video.get('channel_id') + uploader_id = str_or_none(video.get('channel_id')) return { 'id': video_id, @@ -414,6 +393,46 @@ class GloboIE(InfoExtractor): 'duration': duration, 'uploader': uploader, 'uploader_id': uploader_id, - 'like_count': like_count, 'formats': formats } + + +class GloboArticleIE(InfoExtractor): + _VALID_URL = 'https?://.+?\.globo\.com/(?:[^/]+/)*(?P<id>[^/]+)\.html' + + _VIDEOID_REGEXES = [ + r'\bdata-video-id=["\'](\d{7,})', + r'\bdata-player-videosids=["\'](\d{7,})', + r'\bvideosIDs\s*:\s*["\'](\d{7,})', + r'\bdata-id=["\'](\d{7,})', + r'<div[^>]+\bid=["\'](\d{7,})', + ] + + _TESTS = [{ + 'url': 'http://g1.globo.com/jornal-nacional/noticia/2014/09/novidade-na-fiscalizacao-de-bagagem-pela-receita-provoca-discussoes.html', + 'md5': '307fdeae4390ccfe6ba1aa198cf6e72b', + 'info_dict': { + 'id': '3652183', + 'ext': 'mp4', + 'title': 'Receita Federal explica como vai fiscalizar bagagens de quem retorna ao Brasil de avião', + 'duration': 110.711, + 'uploader': 'Rede Globo', + 'uploader_id': '196', + } + }, { + 'url': 'http://gq.globo.com/Prazeres/Poder/noticia/2015/10/all-o-desafio-assista-ao-segundo-capitulo-da-serie.html', + 'only_matching': True, + }, { + 'url': 'http://gshow.globo.com/programas/tv-xuxa/O-Programa/noticia/2014/01/xuxa-e-junno-namoram-muuuito-em-luau-de-zeze-di-camargo-e-luciano.html', + 'only_matching': True, + }] + + @classmethod + def suitable(cls, url): + return False if GloboIE.suitable(url) else super(GloboArticleIE, cls).suitable(url) + + def _real_extract(self, url): + display_id = self._match_id(url) + webpage = self._download_webpage(url, display_id) + video_id = self._search_regex(self._VIDEOID_REGEXES, webpage, 'video id') + return self.url_result('globo:%s' % video_id, 'Globo') diff --git a/youtube_dl/extractor/googleplus.py b/youtube_dl/extractor/googleplus.py index fcefe54cd..731bacd67 100644 --- a/youtube_dl/extractor/googleplus.py +++ b/youtube_dl/extractor/googleplus.py @@ -61,7 +61,7 @@ class GooglePlusIE(InfoExtractor): 'width': int(width), 'height': int(height), } for width, height, video_url in re.findall( - r'\d+,(\d+),(\d+),"(https?://redirector\.googlevideo\.com.*?)"', webpage)] + r'\d+,(\d+),(\d+),"(https?://[^.]+\.googleusercontent.com.*?)"', webpage)] self._sort_formats(formats) return { diff --git a/youtube_dl/extractor/hostingbulk.py b/youtube_dl/extractor/hostingbulk.py deleted file mode 100644 index a3154cfde..000000000 --- a/youtube_dl/extractor/hostingbulk.py +++ /dev/null @@ -1,80 +0,0 @@ -# coding: utf-8 -from __future__ import unicode_literals - -import re - -from .common import InfoExtractor -from ..compat import ( - compat_urllib_request, -) -from ..utils import ( - ExtractorError, - int_or_none, - urlencode_postdata, -) - - -class HostingBulkIE(InfoExtractor): - _VALID_URL = r'''(?x) - https?://(?:www\.)?hostingbulk\.com/ - (?:embed-)?(?P<id>[A-Za-z0-9]{12})(?:-\d+x\d+)?\.html''' - _FILE_DELETED_REGEX = r'<b>File Not Found</b>' - _TEST = { - 'url': 'http://hostingbulk.com/n0ulw1hv20fm.html', - 'md5': '6c8653c8ecf7ebfa83b76e24b7b2fe3f', - 'info_dict': { - 'id': 'n0ulw1hv20fm', - 'ext': 'mp4', - 'title': 'md5:5afeba33f48ec87219c269e054afd622', - 'filesize': 6816081, - 'thumbnail': 're:^http://.*\.jpg$', - } - } - - def _real_extract(self, url): - video_id = self._match_id(url) - url = 'http://hostingbulk.com/{0:}.html'.format(video_id) - - # Custom request with cookie to set language to English, so our file - # deleted regex would work. - request = compat_urllib_request.Request( - url, headers={'Cookie': 'lang=english'}) - webpage = self._download_webpage(request, video_id) - - if re.search(self._FILE_DELETED_REGEX, webpage) is not None: - raise ExtractorError('Video %s does not exist' % video_id, - expected=True) - - title = self._html_search_regex(r'<h3>(.*?)</h3>', webpage, 'title') - filesize = int_or_none( - self._search_regex( - r'<small>\((\d+)\sbytes?\)</small>', - webpage, - 'filesize', - fatal=False - ) - ) - thumbnail = self._search_regex( - r'<img src="([^"]+)".+?class="pic"', - webpage, 'thumbnail', fatal=False) - - fields = self._hidden_inputs(webpage) - - request = compat_urllib_request.Request(url, urlencode_postdata(fields)) - request.add_header('Content-type', 'application/x-www-form-urlencoded') - response = self._request_webpage(request, video_id, - 'Submiting download request') - video_url = response.geturl() - - formats = [{ - 'format_id': 'sd', - 'filesize': filesize, - 'url': video_url, - }] - - return { - 'id': video_id, - 'title': title, - 'thumbnail': thumbnail, - 'formats': formats, - } diff --git a/youtube_dl/extractor/iconosquare.py b/youtube_dl/extractor/iconosquare.py index 70e4c0d41..a39f422e9 100644 --- a/youtube_dl/extractor/iconosquare.py +++ b/youtube_dl/extractor/iconosquare.py @@ -1,7 +1,11 @@ from __future__ import unicode_literals from .common import InfoExtractor -from ..utils import int_or_none +from ..utils import ( + int_or_none, + get_element_by_id, + remove_end, +) class IconosquareIE(InfoExtractor): @@ -12,7 +16,7 @@ class IconosquareIE(InfoExtractor): 'info_dict': { 'id': '522207370455279102_24101272', 'ext': 'mp4', - 'title': 'Instagram media by @aguynamedpatrick (Patrick Janelle)', + 'title': 'Instagram photo by @aguynamedpatrick (Patrick Janelle)', 'description': 'md5:644406a9ec27457ed7aa7a9ebcd4ce3d', 'timestamp': 1376471991, 'upload_date': '20130814', @@ -29,8 +33,7 @@ class IconosquareIE(InfoExtractor): webpage = self._download_webpage(url, video_id) media = self._parse_json( - self._search_regex( - r'window\.media\s*=\s*({.+?});\n', webpage, 'media'), + get_element_by_id('mediaJson', webpage), video_id) formats = [{ @@ -41,9 +44,7 @@ class IconosquareIE(InfoExtractor): } for format_id, f in media['videos'].items()] self._sort_formats(formats) - title = self._html_search_regex( - r'<title>(.+?)(?: *\(Videos?\))? \| (?:Iconosquare|Statigram)</title>', - webpage, 'title') + title = remove_end(self._og_search_title(webpage), ' - via Iconosquare') timestamp = int_or_none(media.get('created_time') or media.get('caption', {}).get('created_time')) description = media.get('caption', {}).get('text') @@ -61,6 +62,14 @@ class IconosquareIE(InfoExtractor): 'height': int_or_none(t.get('height')) } for thumbnail_id, t in media.get('images', {}).items()] + comments = [{ + 'id': comment.get('id'), + 'text': comment['text'], + 'timestamp': int_or_none(comment.get('created_time')), + 'author': comment.get('from', {}).get('full_name'), + 'author_id': comment.get('from', {}).get('username'), + } for comment in media.get('comments', {}).get('data', []) if 'text' in comment] + return { 'id': video_id, 'title': title, @@ -72,4 +81,5 @@ class IconosquareIE(InfoExtractor): 'comment_count': comment_count, 'like_count': like_count, 'formats': formats, + 'comments': comments, } diff --git a/youtube_dl/extractor/imdb.py b/youtube_dl/extractor/imdb.py index 4bb574cf3..02e1e428e 100644 --- a/youtube_dl/extractor/imdb.py +++ b/youtube_dl/extractor/imdb.py @@ -4,8 +4,8 @@ import re import json from .common import InfoExtractor -from ..compat import ( - compat_urlparse, +from ..utils import ( + qualities, ) @@ -30,24 +30,33 @@ class ImdbIE(InfoExtractor): descr = self._html_search_regex( r'(?s)<span itemprop="description">(.*?)</span>', webpage, 'description', fatal=False) - available_formats = re.findall( - r'case \'(?P<f_id>.*?)\' :$\s+url = \'(?P<path>.*?)\'', webpage, - flags=re.MULTILINE) + player_url = 'http://www.imdb.com/video/imdb/vi%s/imdb/single' % video_id + player_page = self._download_webpage( + player_url, video_id, 'Downloading player page') + # the player page contains the info for the default format, we have to + # fetch other pages for the rest of the formats + extra_formats = re.findall(r'href="(?P<url>%s.*?)".*?>(?P<name>.*?)<' % re.escape(player_url), player_page) + format_pages = [ + self._download_webpage( + f_url, video_id, 'Downloading info for %s format' % f_name) + for f_url, f_name in extra_formats] + format_pages.append(player_page) + + quality = qualities(['SD', '480p', '720p']) formats = [] - for f_id, f_path in available_formats: - f_path = f_path.strip() - format_page = self._download_webpage( - compat_urlparse.urljoin(url, f_path), - 'Downloading info for %s format' % f_id) + for format_page in format_pages: json_data = self._search_regex( r'<script[^>]+class="imdb-player-data"[^>]*?>(.*?)</script>', format_page, 'json data', flags=re.DOTALL) info = json.loads(json_data) format_info = info['videoPlayerObject']['video'] + f_id = format_info['ffname'] formats.append({ 'format_id': f_id, 'url': format_info['videoInfoList'][0]['videoUrl'], + 'quality': quality(f_id), }) + self._sort_formats(formats) return { 'id': video_id, diff --git a/youtube_dl/extractor/iqiyi.py b/youtube_dl/extractor/iqiyi.py index 393e67e35..2df1da3f0 100644 --- a/youtube_dl/extractor/iqiyi.py +++ b/youtube_dl/extractor/iqiyi.py @@ -95,6 +95,10 @@ class IqiyiIE(InfoExtractor): ('10', 'h1'), ] + @staticmethod + def md5_text(text): + return hashlib.md5(text.encode('utf-8')).hexdigest() + def construct_video_urls(self, data, video_id, _uuid): def do_xor(x, y): a = y % 3 @@ -121,7 +125,7 @@ class IqiyiIE(InfoExtractor): note='Download path key of segment %d for format %s' % (segment_index + 1, format_id) )['t'] t = str(int(math.floor(int(tm) / (600.0)))) - return hashlib.md5((t + mg + x).encode('utf8')).hexdigest() + return self.md5_text(t + mg + x) video_urls_dict = {} for format_item in data['vp']['tkl'][0]['vs']: @@ -179,20 +183,19 @@ class IqiyiIE(InfoExtractor): def get_raw_data(self, tvid, video_id, enc_key, _uuid): tm = str(int(time.time())) + tail = tm + tvid param = { 'key': 'fvip', - 'src': hashlib.md5(b'youtube-dl').hexdigest(), + 'src': self.md5_text('youtube-dl'), 'tvId': tvid, 'vid': video_id, 'vinfo': 1, 'tm': tm, - 'enc': hashlib.md5( - (enc_key + tm + tvid).encode('utf8')).hexdigest(), + 'enc': self.md5_text(enc_key + tail), 'qyid': _uuid, 'tn': random.random(), 'um': 0, - 'authkey': hashlib.md5( - (tm + tvid).encode('utf8')).hexdigest() + 'authkey': self.md5_text(self.md5_text('') + tail), } api_url = 'http://cache.video.qiyi.com/vms' + '?' + \ @@ -201,7 +204,10 @@ class IqiyiIE(InfoExtractor): return raw_data def get_enc_key(self, swf_url, video_id): - enc_key = '3601ba290e4f4662848c710e2122007e' # last update at 2015-08-10 for Zombie + # TODO: automatic key extraction + # last update at 2015-10-22 for Zombie::bite + # '7223c67061dbea1259d0ceb44f44b6d62288f4f80c972170de5201d2321060270e05'[2:66][0::2] + enc_key = '2c76de15dcb44bd28ff0927d50d31620' return enc_key def _real_extract(self, url): diff --git a/youtube_dl/extractor/jeuxvideo.py b/youtube_dl/extractor/jeuxvideo.py index 1df084d87..eef7daa29 100644 --- a/youtube_dl/extractor/jeuxvideo.py +++ b/youtube_dl/extractor/jeuxvideo.py @@ -28,7 +28,7 @@ class JeuxVideoIE(InfoExtractor): mobj = re.match(self._VALID_URL, url) title = mobj.group(1) webpage = self._download_webpage(url, title) - title = self._html_search_meta('name', webpage) + title = self._html_search_meta('name', webpage) or self._og_search_title(webpage) config_url = self._html_search_regex( r'data-src="(/contenu/medias/video.php.*?)"', webpage, 'config URL') diff --git a/youtube_dl/extractor/kaltura.py b/youtube_dl/extractor/kaltura.py index 3dca0e566..0dcd6cd05 100644 --- a/youtube_dl/extractor/kaltura.py +++ b/youtube_dl/extractor/kaltura.py @@ -16,7 +16,7 @@ class KalturaIE(InfoExtractor): (?: kaltura:(?P<partner_id_s>\d+):(?P<id_s>[0-9a-z_]+)| https?:// - (:?(?:www|cdnapisec)\.)?kaltura\.com/ + (:?(?:www|cdnapi(?:sec)?)\.)?kaltura\.com/ (?: (?: # flash player diff --git a/youtube_dl/extractor/keek.py b/youtube_dl/extractor/keek.py index c0956ba09..94a03d277 100644 --- a/youtube_dl/extractor/keek.py +++ b/youtube_dl/extractor/keek.py @@ -1,46 +1,39 @@ +# coding: utf-8 from __future__ import unicode_literals from .common import InfoExtractor class KeekIE(InfoExtractor): - _VALID_URL = r'https?://(?:www\.)?keek\.com/(?:!|\w+/keeks/)(?P<id>\w+)' + _VALID_URL = r'https?://(?:www\.)?keek\.com/keek/(?P<id>\w+)' IE_NAME = 'keek' _TEST = { - 'url': 'https://www.keek.com/ytdl/keeks/NODfbab', - 'md5': '09c5c109067536c1cec8bac8c21fea05', + 'url': 'https://www.keek.com/keek/NODfbab', + 'md5': '9b0636f8c0f7614afa4ea5e4c6e57e83', 'info_dict': { 'id': 'NODfbab', 'ext': 'mp4', - 'uploader': 'youtube-dl project', - 'uploader_id': 'ytdl', - 'title': 'test chars: "\'/\\\u00e4<>This is a test video for youtube-dl.For more information, contact phihag@phihag.de .', + 'title': 'md5:35d42050a3ece241d5ddd7fdcc6fd896', + 'uploader': 'ytdl', + 'uploader_id': 'eGT5bab', }, } def _real_extract(self, url): video_id = self._match_id(url) - video_url = 'http://cdn.keek.com/keek/video/%s' % video_id - thumbnail = 'http://cdn.keek.com/keek/thumbnail/%s/w100/h75' % video_id webpage = self._download_webpage(url, video_id) - raw_desc = self._html_search_meta('description', webpage) - if raw_desc: - uploader = self._html_search_regex( - r'Watch (.*?)\s+\(', raw_desc, 'uploader', fatal=False) - uploader_id = self._html_search_regex( - r'Watch .*?\(@(.+?)\)', raw_desc, 'uploader_id', fatal=False) - else: - uploader = None - uploader_id = None - return { 'id': video_id, - 'url': video_url, + 'url': self._og_search_video_url(webpage), 'ext': 'mp4', - 'title': self._og_search_title(webpage), - 'thumbnail': thumbnail, - 'uploader': uploader, - 'uploader_id': uploader_id, + 'title': self._og_search_description(webpage).strip(), + 'thumbnail': self._og_search_thumbnail(webpage), + 'uploader': self._search_regex( + r'data-username=(["\'])(?P<uploader>.+?)\1', webpage, + 'uploader', fatal=False, group='uploader'), + 'uploader_id': self._search_regex( + r'data-user-id=(["\'])(?P<uploader_id>.+?)\1', webpage, + 'uploader id', fatal=False, group='uploader_id'), } diff --git a/youtube_dl/extractor/kuwo.py b/youtube_dl/extractor/kuwo.py index fa233377d..0c8ed5d07 100644 --- a/youtube_dl/extractor/kuwo.py +++ b/youtube_dl/extractor/kuwo.py @@ -57,6 +57,7 @@ class KuwoIE(KuwoBaseIE): 'upload_date': '20080122', 'description': 'md5:ed13f58e3c3bf3f7fd9fbc4e5a7aa75c' }, + 'skip': 'this song has been offline because of copyright issues', }, { 'url': 'http://www.kuwo.cn/yinyue/6446136/', 'info_dict': { @@ -76,9 +77,11 @@ class KuwoIE(KuwoBaseIE): webpage = self._download_webpage( url, song_id, note='Download song detail info', errnote='Unable to get song detail info') + if '对不起,该歌曲由于版权问题已被下线,将返回网站首页' in webpage: + raise ExtractorError('this song has been offline because of copyright issues', expected=True) song_name = self._html_search_regex( - r'<h1[^>]+title="([^"]+)">', webpage, 'song name') + r'(?s)class="(?:[^"\s]+\s+)*title(?:\s+[^"\s]+)*".*?<h1[^>]+title="([^"]+)"', webpage, 'song name') singer_name = self._html_search_regex( r'<div[^>]+class="s_img">\s*<a[^>]+title="([^>]+)"', webpage, 'singer name', fatal=False) diff --git a/youtube_dl/extractor/letv.py b/youtube_dl/extractor/letv.py index a28abb0f0..effd9eb92 100644 --- a/youtube_dl/extractor/letv.py +++ b/youtube_dl/extractor/letv.py @@ -9,13 +9,14 @@ from .common import InfoExtractor from ..compat import ( compat_urllib_parse, compat_urllib_request, - compat_urlparse, + compat_ord, ) from ..utils import ( determine_ext, ExtractorError, parse_iso8601, int_or_none, + encode_data_uri, ) @@ -25,15 +26,16 @@ class LetvIE(InfoExtractor): _TESTS = [{ 'url': 'http://www.letv.com/ptv/vplay/22005890.html', - 'md5': 'cab23bd68d5a8db9be31c9a222c1e8df', + 'md5': 'edadcfe5406976f42f9f266057ee5e40', 'info_dict': { 'id': '22005890', 'ext': 'mp4', 'title': '第87届奥斯卡颁奖礼完美落幕 《鸟人》成最大赢家', - 'timestamp': 1424747397, - 'upload_date': '20150224', 'description': 'md5:a9cb175fd753e2962176b7beca21a47c', - } + }, + 'params': { + 'hls_prefer_native': True, + }, }, { 'url': 'http://www.letv.com/ptv/vplay/1415246.html', 'info_dict': { @@ -42,16 +44,22 @@ class LetvIE(InfoExtractor): 'title': '美人天下01', 'description': 'md5:f88573d9d7225ada1359eaf0dbf8bcda', }, + 'params': { + 'hls_prefer_native': True, + }, }, { 'note': 'This video is available only in Mainland China, thus a proxy is needed', 'url': 'http://www.letv.com/ptv/vplay/1118082.html', - 'md5': 'f80936fbe20fb2f58648e81386ff7927', + 'md5': '2424c74948a62e5f31988438979c5ad1', 'info_dict': { 'id': '1118082', 'ext': 'mp4', 'title': '与龙共舞 完整版', 'description': 'md5:7506a5eeb1722bb9d4068f85024e3986', }, + 'params': { + 'hls_prefer_native': True, + }, 'skip': 'Only available in China', }] @@ -74,6 +82,27 @@ class LetvIE(InfoExtractor): _loc3_ = self.ror(_loc3_, _loc2_ % 17) return _loc3_ + # see M3U8Encryption class in KLetvPlayer.swf + @staticmethod + def decrypt_m3u8(encrypted_data): + if encrypted_data[:5].decode('utf-8').lower() != 'vc_01': + return encrypted_data + encrypted_data = encrypted_data[5:] + + _loc4_ = bytearray() + while encrypted_data: + b = compat_ord(encrypted_data[0]) + _loc4_.extend([b // 16, b & 0x0f]) + encrypted_data = encrypted_data[1:] + idx = len(_loc4_) - 11 + _loc4_ = _loc4_[idx:] + _loc4_[:idx] + _loc7_ = bytearray() + while _loc4_: + _loc7_.append(_loc4_[0] * 16 + _loc4_[1]) + _loc4_ = _loc4_[2:] + + return bytes(_loc7_) + def _real_extract(self, url): media_id = self._match_id(url) page = self._download_webpage(url, media_id) @@ -115,23 +144,28 @@ class LetvIE(InfoExtractor): for format_id in formats: if format_id in dispatch: media_url = playurl['domain'][0] + dispatch[format_id][0] - - # Mimic what flvxz.com do - url_parts = list(compat_urlparse.urlparse(media_url)) - qs = dict(compat_urlparse.parse_qs(url_parts[4])) - qs.update({ - 'platid': '14', - 'splatid': '1401', - 'tss': 'no', - 'retry': 1 + media_url += '&' + compat_urllib_parse.urlencode({ + 'm3v': 1, + 'format': 1, + 'expect': 3, + 'rateid': format_id, }) - url_parts[4] = compat_urllib_parse.urlencode(qs) - media_url = compat_urlparse.urlunparse(url_parts) + + nodes_data = self._download_json( + media_url, media_id, + 'Download JSON metadata for format %s' % format_id) + + req = self._request_webpage( + nodes_data['nodelist'][0]['location'], media_id, + note='Downloading m3u8 information for format %s' % format_id) + + m3u8_data = self.decrypt_m3u8(req.read()) url_info_dict = { - 'url': media_url, + 'url': encode_data_uri(m3u8_data, 'application/vnd.apple.mpegurl'), 'ext': determine_ext(dispatch[format_id][1]), 'format_id': format_id, + 'protocol': 'm3u8', } if format_id[-1:] == 'p': diff --git a/youtube_dl/extractor/limelight.py b/youtube_dl/extractor/limelight.py new file mode 100644 index 000000000..fb03dd527 --- /dev/null +++ b/youtube_dl/extractor/limelight.py @@ -0,0 +1,229 @@ +# coding: utf-8 +from __future__ import unicode_literals + +import re + +from .common import InfoExtractor +from ..utils import ( + determine_ext, + float_or_none, + int_or_none, +) + + +class LimelightBaseIE(InfoExtractor): + _PLAYLIST_SERVICE_URL = 'http://production-ps.lvp.llnw.net/r/PlaylistService/%s/%s/%s' + _API_URL = 'http://api.video.limelight.com/rest/organizations/%s/%s/%s/%s.json' + + def _call_playlist_service(self, item_id, method, fatal=True): + return self._download_json( + self._PLAYLIST_SERVICE_URL % (self._PLAYLIST_SERVICE_PATH, item_id, method), + item_id, 'Downloading PlaylistService %s JSON' % method, fatal=fatal) + + def _call_api(self, organization_id, item_id, method): + return self._download_json( + self._API_URL % (organization_id, self._API_PATH, item_id, method), + item_id, 'Downloading API %s JSON' % method) + + def _extract(self, item_id, pc_method, mobile_method, meta_method): + pc = self._call_playlist_service(item_id, pc_method) + metadata = self._call_api(pc['orgId'], item_id, meta_method) + mobile = self._call_playlist_service(item_id, mobile_method, fatal=False) + return pc, mobile, metadata + + def _extract_info(self, streams, mobile_urls, properties): + video_id = properties['media_id'] + formats = [] + + for stream in streams: + stream_url = stream.get('url') + if not stream_url: + continue + if '.f4m' in stream_url: + formats.extend(self._extract_f4m_formats(stream_url, video_id)) + else: + fmt = { + 'url': stream_url, + 'abr': float_or_none(stream.get('audioBitRate')), + 'vbr': float_or_none(stream.get('videoBitRate')), + 'fps': float_or_none(stream.get('videoFrameRate')), + 'width': int_or_none(stream.get('videoWidthInPixels')), + 'height': int_or_none(stream.get('videoHeightInPixels')), + 'ext': determine_ext(stream_url) + } + rtmp = re.search(r'^(?P<url>rtmpe?://[^/]+/(?P<app>.+))/(?P<playpath>mp4:.+)$', stream_url) + if rtmp: + format_id = 'rtmp' + if stream.get('videoBitRate'): + format_id += '-%d' % int_or_none(stream['videoBitRate']) + fmt.update({ + 'url': rtmp.group('url'), + 'play_path': rtmp.group('playpath'), + 'app': rtmp.group('app'), + 'ext': 'flv', + 'format_id': format_id, + }) + formats.append(fmt) + + for mobile_url in mobile_urls: + media_url = mobile_url.get('mobileUrl') + if not media_url: + continue + format_id = mobile_url.get('targetMediaPlatform') + if determine_ext(media_url) == 'm3u8': + formats.extend(self._extract_m3u8_formats( + media_url, video_id, 'mp4', entry_protocol='m3u8_native', + preference=-1, m3u8_id=format_id)) + else: + formats.append({ + 'url': media_url, + 'format_id': format_id, + 'preference': -1, + }) + + self._sort_formats(formats) + + title = properties['title'] + description = properties.get('description') + timestamp = int_or_none(properties.get('publish_date') or properties.get('create_date')) + duration = float_or_none(properties.get('duration_in_milliseconds'), 1000) + filesize = int_or_none(properties.get('total_storage_in_bytes')) + categories = [properties.get('category')] + tags = properties.get('tags', []) + thumbnails = [{ + 'url': thumbnail['url'], + 'width': int_or_none(thumbnail.get('width')), + 'height': int_or_none(thumbnail.get('height')), + } for thumbnail in properties.get('thumbnails', []) if thumbnail.get('url')] + + subtitles = {} + for caption in properties.get('captions', {}): + lang = caption.get('language_code') + subtitles_url = caption.get('url') + if lang and subtitles_url: + subtitles[lang] = [{ + 'url': subtitles_url, + }] + + return { + 'id': video_id, + 'title': title, + 'description': description, + 'formats': formats, + 'timestamp': timestamp, + 'duration': duration, + 'filesize': filesize, + 'categories': categories, + 'tags': tags, + 'thumbnails': thumbnails, + 'subtitles': subtitles, + } + + +class LimelightMediaIE(LimelightBaseIE): + IE_NAME = 'limelight' + _VALID_URL = r'(?:limelight:media:|http://link\.videoplatform\.limelight\.com/media/\??\bmediaId=)(?P<id>[a-z0-9]{32})' + _TESTS = [{ + 'url': 'http://link.videoplatform.limelight.com/media/?mediaId=3ffd040b522b4485b6d84effc750cd86', + 'info_dict': { + 'id': '3ffd040b522b4485b6d84effc750cd86', + 'ext': 'flv', + 'title': 'HaP and the HB Prince Trailer', + 'description': 'md5:8005b944181778e313d95c1237ddb640', + 'thumbnail': 're:^https?://.*\.jpeg$', + 'duration': 144.23, + 'timestamp': 1244136834, + 'upload_date': '20090604', + }, + 'params': { + # rtmp download + 'skip_download': True, + }, + }, { + # video with subtitles + 'url': 'limelight:media:a3e00274d4564ec4a9b29b9466432335', + 'info_dict': { + 'id': 'a3e00274d4564ec4a9b29b9466432335', + 'ext': 'flv', + 'title': '3Play Media Overview Video', + 'description': '', + 'thumbnail': 're:^https?://.*\.jpeg$', + 'duration': 78.101, + 'timestamp': 1338929955, + 'upload_date': '20120605', + 'subtitles': 'mincount:9', + }, + 'params': { + # rtmp download + 'skip_download': True, + }, + }] + _PLAYLIST_SERVICE_PATH = 'media' + _API_PATH = 'media' + + def _real_extract(self, url): + video_id = self._match_id(url) + + pc, mobile, metadata = self._extract( + video_id, 'getPlaylistByMediaId', 'getMobilePlaylistByMediaId', 'properties') + + return self._extract_info( + pc['playlistItems'][0].get('streams', []), + mobile['mediaList'][0].get('mobileUrls', []) if mobile else [], + metadata) + + +class LimelightChannelIE(LimelightBaseIE): + IE_NAME = 'limelight:channel' + _VALID_URL = r'(?:limelight:channel:|http://link\.videoplatform\.limelight\.com/media/\??\bchannelId=)(?P<id>[a-z0-9]{32})' + _TEST = { + 'url': 'http://link.videoplatform.limelight.com/media/?channelId=ab6a524c379342f9b23642917020c082', + 'info_dict': { + 'id': 'ab6a524c379342f9b23642917020c082', + 'title': 'Javascript Sample Code', + }, + 'playlist_mincount': 3, + } + _PLAYLIST_SERVICE_PATH = 'channel' + _API_PATH = 'channels' + + def _real_extract(self, url): + channel_id = self._match_id(url) + + pc, mobile, medias = self._extract( + channel_id, 'getPlaylistByChannelId', + 'getMobilePlaylistWithNItemsByChannelId?begin=0&count=-1', 'media') + + entries = [ + self._extract_info( + pc['playlistItems'][i].get('streams', []), + mobile['mediaList'][i].get('mobileUrls', []) if mobile else [], + medias['media_list'][i]) + for i in range(len(medias['media_list']))] + + return self.playlist_result(entries, channel_id, pc['title']) + + +class LimelightChannelListIE(LimelightBaseIE): + IE_NAME = 'limelight:channel_list' + _VALID_URL = r'(?:limelight:channel_list:|http://link\.videoplatform\.limelight\.com/media/\?.*?\bchannelListId=)(?P<id>[a-z0-9]{32})' + _TEST = { + 'url': 'http://link.videoplatform.limelight.com/media/?channelListId=301b117890c4465c8179ede21fd92e2b', + 'info_dict': { + 'id': '301b117890c4465c8179ede21fd92e2b', + 'title': 'Website - Hero Player', + }, + 'playlist_mincount': 2, + } + _PLAYLIST_SERVICE_PATH = 'channel_list' + + def _real_extract(self, url): + channel_list_id = self._match_id(url) + + channel_list = self._call_playlist_service(channel_list_id, 'getMobileChannelListById') + + entries = [ + self.url_result('limelight:channel:%s' % channel['id'], 'LimelightChannel') + for channel in channel_list['channelList']] + + return self.playlist_result(entries, channel_list_id, channel_list['title']) diff --git a/youtube_dl/extractor/lynda.py b/youtube_dl/extractor/lynda.py index 378117270..9a207b2cd 100644 --- a/youtube_dl/extractor/lynda.py +++ b/youtube_dl/extractor/lynda.py @@ -82,6 +82,11 @@ class LyndaBaseIE(InfoExtractor): expected=True) raise ExtractorError('Unable to log in') + def _logout(self): + self._download_webpage( + 'http://www.lynda.com/ajax/logout.aspx', None, + 'Logging out', 'Unable to log out', fatal=False) + class LyndaIE(LyndaBaseIE): IE_NAME = 'lynda' @@ -108,50 +113,47 @@ class LyndaIE(LyndaBaseIE): def _real_extract(self, url): video_id = self._match_id(url) - page = self._download_webpage( + video = self._download_json( 'http://www.lynda.com/ajax/player?videoId=%s&type=video' % video_id, video_id, 'Downloading video JSON') - video_json = json.loads(page) - if 'Status' in video_json: + if 'Status' in video: raise ExtractorError( - 'lynda returned error: %s' % video_json['Message'], expected=True) + 'lynda returned error: %s' % video['Message'], expected=True) - if video_json['HasAccess'] is False: + if video.get('HasAccess') is False: self.raise_login_required('Video %s is only available for members' % video_id) - video_id = compat_str(video_json['ID']) - duration = video_json['DurationInSeconds'] - title = video_json['Title'] + video_id = compat_str(video.get('ID') or video_id) + duration = int_or_none(video.get('DurationInSeconds')) + title = video['Title'] formats = [] - fmts = video_json.get('Formats') + fmts = video.get('Formats') if fmts: - formats.extend([ - { - 'url': fmt['Url'], - 'ext': fmt['Extension'], - 'width': fmt['Width'], - 'height': fmt['Height'], - 'filesize': fmt['FileSize'], - 'format_id': str(fmt['Resolution']) - } for fmt in fmts]) - - prioritized_streams = video_json.get('PrioritizedStreams') + formats.extend([{ + 'url': f['Url'], + 'ext': f.get('Extension'), + 'width': int_or_none(f.get('Width')), + 'height': int_or_none(f.get('Height')), + 'filesize': int_or_none(f.get('FileSize')), + 'format_id': compat_str(f.get('Resolution')) if f.get('Resolution') else None, + } for f in fmts if f.get('Url')]) + + prioritized_streams = video.get('PrioritizedStreams') if prioritized_streams: - formats.extend([ - { + for prioritized_stream_id, prioritized_stream in prioritized_streams.items(): + formats.extend([{ 'url': video_url, 'width': int_or_none(format_id), - 'format_id': format_id, - } for format_id, video_url in prioritized_streams['0'].items() - ]) + 'format_id': '%s-%s' % (prioritized_stream_id, format_id), + } for format_id, video_url in prioritized_stream.items()]) self._check_formats(formats, video_id) self._sort_formats(formats) - subtitles = self.extract_subtitles(video_id, page) + subtitles = self.extract_subtitles(video_id) return { 'id': video_id, @@ -182,7 +184,7 @@ class LyndaIE(LyndaBaseIE): if srt: return srt - def _get_subtitles(self, video_id, webpage): + def _get_subtitles(self, video_id): url = 'http://www.lynda.com/ajax/player?videoId=%s&type=transcript' % video_id subs = self._download_json(url, None, False) if subs: @@ -204,12 +206,13 @@ class LyndaCourseIE(LyndaBaseIE): course_path = mobj.group('coursepath') course_id = mobj.group('courseid') - page = self._download_webpage( + course = self._download_json( 'http://www.lynda.com/ajax/player?courseId=%s&type=course' % course_id, course_id, 'Downloading course JSON') - course_json = json.loads(page) - if 'Status' in course_json and course_json['Status'] == 'NotFound': + self._logout() + + if course.get('Status') == 'NotFound': raise ExtractorError( 'Course %s does not exist' % course_id, expected=True) @@ -219,12 +222,13 @@ class LyndaCourseIE(LyndaBaseIE): # Might want to extract videos right here from video['Formats'] as it seems 'Formats' is not provided # by single video API anymore - for chapter in course_json['Chapters']: - for video in chapter['Videos']: - if video['HasAccess'] is False: + for chapter in course['Chapters']: + for video in chapter.get('Videos', []): + if video.get('HasAccess') is False: unaccessible_videos += 1 continue - videos.append(video['ID']) + if video.get('ID'): + videos.append(video['ID']) if unaccessible_videos > 0: self._downloader.report_warning( @@ -237,6 +241,6 @@ class LyndaCourseIE(LyndaBaseIE): 'Lynda') for video_id in videos] - course_title = course_json['Title'] + course_title = course.get('Title') return self.playlist_result(entries, course_id, course_title) diff --git a/youtube_dl/extractor/mdr.py b/youtube_dl/extractor/mdr.py index fc7499958..88334889e 100644 --- a/youtube_dl/extractor/mdr.py +++ b/youtube_dl/extractor/mdr.py @@ -1,64 +1,169 @@ +# coding: utf-8 from __future__ import unicode_literals -import re - from .common import InfoExtractor +from ..compat import compat_urlparse +from ..utils import ( + determine_ext, + int_or_none, + parse_duration, + parse_iso8601, + xpath_text, +) class MDRIE(InfoExtractor): - _VALID_URL = r'^(?P<domain>https?://(?:www\.)?mdr\.de)/(?:.*)/(?P<type>video|audio)(?P<video_id>[^/_]+)(?:_|\.html)' + IE_DESC = 'MDR.DE and KiKA' + _VALID_URL = r'https?://(?:www\.)?(?:mdr|kika)\.de/(?:.*)/[a-z]+(?P<id>\d+)(?:_.+?)?\.html' - # No tests, MDR regularily deletes its videos - _TEST = { + _TESTS = [{ + # MDR regularily deletes its videos 'url': 'http://www.mdr.de/fakt/video189002.html', 'only_matching': True, - } + }, { + # audio + 'url': 'http://www.mdr.de/kultur/audio1312272_zc-15948bad_zs-86171fdd.html', + 'md5': '64c4ee50f0a791deb9479cd7bbe9d2fa', + 'info_dict': { + 'id': '1312272', + 'ext': 'mp3', + 'title': 'Feuilleton vom 30. Oktober 2015', + 'duration': 250, + 'uploader': 'MITTELDEUTSCHER RUNDFUNK', + }, + }, { + 'url': 'http://www.kika.de/baumhaus/videos/video19636.html', + 'md5': '4930515e36b06c111213e80d1e4aad0e', + 'info_dict': { + 'id': '19636', + 'ext': 'mp4', + 'title': 'Baumhaus vom 30. Oktober 2015', + 'duration': 134, + 'uploader': 'KIKA', + }, + }, { + 'url': 'http://www.kika.de/sendungen/einzelsendungen/weihnachtsprogramm/videos/video8182.html', + 'md5': '5fe9c4dd7d71e3b238f04b8fdd588357', + 'info_dict': { + 'id': '8182', + 'ext': 'mp4', + 'title': 'Beutolomäus und der geheime Weihnachtswunsch', + 'description': 'md5:b69d32d7b2c55cbe86945ab309d39bbd', + 'timestamp': 1419047100, + 'upload_date': '20141220', + 'duration': 4628, + 'uploader': 'KIKA', + }, + }, { + 'url': 'http://www.kika.de/baumhaus/sendungen/video19636_zc-fea7f8a0_zs-4bf89c60.html', + 'only_matching': True, + }, { + 'url': 'http://www.kika.de/sendungen/einzelsendungen/weihnachtsprogramm/einzelsendung2534.html', + 'only_matching': True, + }] def _real_extract(self, url): - m = re.match(self._VALID_URL, url) - video_id = m.group('video_id') - domain = m.group('domain') + video_id = self._match_id(url) + + webpage = self._download_webpage(url, video_id) + + data_url = self._search_regex( + r'dataURL\s*:\s*(["\'])(?P<url>/.+/(?:video|audio)[0-9]+-avCustom\.xml)\1', + webpage, 'data url', group='url') - # determine title and media streams from webpage - html = self._download_webpage(url, video_id) + doc = self._download_xml( + compat_urlparse.urljoin(url, data_url), video_id) - title = self._html_search_regex(r'<h[12]>(.*?)</h[12]>', html, 'title') - xmlurl = self._search_regex( - r'dataURL:\'(/(?:.+)/(?:video|audio)[0-9]+-avCustom.xml)', html, 'XML URL') + title = xpath_text(doc, ['./title', './broadcast/broadcastName'], 'title', fatal=True) - doc = self._download_xml(domain + xmlurl, video_id) formats = [] - for a in doc.findall('./assets/asset'): - url_el = a.find('./progressiveDownloadUrl') - if url_el is None: - continue - abr = int(a.find('bitrateAudio').text) // 1000 - media_type = a.find('mediaType').text - format = { - 'abr': abr, - 'filesize': int(a.find('fileSize').text), - 'url': url_el.text, - } - - vbr_el = a.find('bitrateVideo') - if vbr_el is None: - format.update({ - 'vcodec': 'none', - 'format_id': '%s-%d' % (media_type, abr), - }) - else: - vbr = int(vbr_el.text) // 1000 - format.update({ - 'vbr': vbr, - 'width': int(a.find('frameWidth').text), - 'height': int(a.find('frameHeight').text), - 'format_id': '%s-%d' % (media_type, vbr), - }) - formats.append(format) + processed_urls = [] + for asset in doc.findall('./assets/asset'): + for source in ( + 'progressiveDownload', + 'dynamicHttpStreamingRedirector', + 'adaptiveHttpStreamingRedirector'): + url_el = asset.find('./%sUrl' % source) + if url_el is None: + continue + + video_url = url_el.text + if video_url in processed_urls: + continue + + processed_urls.append(video_url) + + vbr = int_or_none(xpath_text(asset, './bitrateVideo', 'vbr'), 1000) + abr = int_or_none(xpath_text(asset, './bitrateAudio', 'abr'), 1000) + + ext = determine_ext(url_el.text) + if ext == 'm3u8': + url_formats = self._extract_m3u8_formats( + video_url, video_id, 'mp4', entry_protocol='m3u8_native', + preference=0, m3u8_id='HLS', fatal=False) + elif ext == 'f4m': + url_formats = self._extract_f4m_formats( + video_url + '?hdcore=3.7.0&plugin=aasp-3.7.0.39.44', video_id, + preference=0, f4m_id='HDS', fatal=False) + else: + media_type = xpath_text(asset, './mediaType', 'media type', default='MP4') + vbr = int_or_none(xpath_text(asset, './bitrateVideo', 'vbr'), 1000) + abr = int_or_none(xpath_text(asset, './bitrateAudio', 'abr'), 1000) + filesize = int_or_none(xpath_text(asset, './fileSize', 'file size')) + + f = { + 'url': video_url, + 'format_id': '%s-%d' % (media_type, vbr or abr), + 'filesize': filesize, + 'abr': abr, + 'preference': 1, + } + + if vbr: + width = int_or_none(xpath_text(asset, './frameWidth', 'width')) + height = int_or_none(xpath_text(asset, './frameHeight', 'height')) + f.update({ + 'vbr': vbr, + 'width': width, + 'height': height, + }) + + url_formats = [f] + + if not url_formats: + continue + + if not vbr: + for f in url_formats: + abr = f.get('tbr') or abr + if 'tbr' in f: + del f['tbr'] + f.update({ + 'abr': abr, + 'vcodec': 'none', + }) + + formats.extend(url_formats) + self._sort_formats(formats) + description = xpath_text(doc, './broadcast/broadcastDescription', 'description') + timestamp = parse_iso8601( + xpath_text( + doc, [ + './broadcast/broadcastDate', + './broadcast/broadcastStartDate', + './broadcast/broadcastEndDate'], + 'timestamp', default=None)) + duration = parse_duration(xpath_text(doc, './duration', 'duration')) + uploader = xpath_text(doc, './rights', 'uploader') + return { 'id': video_id, 'title': title, + 'description': description, + 'timestamp': timestamp, + 'duration': duration, + 'uploader': uploader, 'formats': formats, } diff --git a/youtube_dl/extractor/megavideoz.py b/youtube_dl/extractor/megavideoz.py deleted file mode 100644 index af7ff07ea..000000000 --- a/youtube_dl/extractor/megavideoz.py +++ /dev/null @@ -1,56 +0,0 @@ -# encoding: utf-8 -from __future__ import unicode_literals - -import re - -from .common import InfoExtractor -from ..utils import ( - ExtractorError, - float_or_none, - xpath_text, -) - - -class MegaVideozIE(InfoExtractor): - _VALID_URL = r'https?://(?:www\.)?megavideoz\.eu/video/(?P<id>[^/]+)(?:/(?P<display_id>[^/]+))?' - _TEST = { - 'url': 'http://megavideoz.eu/video/WM6UB919XMXH/SMPTE-Universal-Film-Leader', - 'info_dict': { - 'id': '48723', - 'display_id': 'SMPTE-Universal-Film-Leader', - 'ext': 'mp4', - 'title': 'SMPTE Universal Film Leader', - 'thumbnail': 're:https?://.*?\.jpg', - 'duration': 10.93, - } - } - - def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - video_id = mobj.group('id') - display_id = mobj.group('display_id') or video_id - - webpage = self._download_webpage(url, display_id) - - if any(p in webpage for p in ('>Video Not Found<', '>404 Error<')): - raise ExtractorError('Video %s does not exist' % video_id, expected=True) - - config = self._download_xml( - self._search_regex( - r"var\s+cnf\s*=\s*'([^']+)'", webpage, 'cnf url'), - display_id) - - video_url = xpath_text(config, './file', 'video url', fatal=True) - title = xpath_text(config, './title', 'title', fatal=True) - thumbnail = xpath_text(config, './image', 'thumbnail') - duration = float_or_none(xpath_text(config, './duration', 'duration')) - video_id = xpath_text(config, './mediaid', 'video id') or video_id - - return { - 'id': video_id, - 'display_id': display_id, - 'url': video_url, - 'title': title, - 'thumbnail': thumbnail, - 'duration': duration - } diff --git a/youtube_dl/extractor/miomio.py b/youtube_dl/extractor/miomio.py index a784fc5fb..ce391c759 100644 --- a/youtube_dl/extractor/miomio.py +++ b/youtube_dl/extractor/miomio.py @@ -4,6 +4,7 @@ from __future__ import unicode_literals import random from .common import InfoExtractor +from ..compat import compat_urllib_request from ..utils import ( xpath_text, int_or_none, @@ -51,6 +52,8 @@ class MioMioIE(InfoExtractor): mioplayer_path = self._search_regex( r'src="(/mioplayer/[^"]+)"', webpage, 'ref_path') + http_headers = {'Referer': 'http://www.miomio.tv%s' % mioplayer_path} + xml_config = self._search_regex( r'flashvars="type=(?:sina|video)&(.+?)&', webpage, 'xml config') @@ -60,14 +63,12 @@ class MioMioIE(InfoExtractor): 'http://www.miomio.tv/mioplayer/mioplayerconfigfiles/xml.php?id=%s&r=%s' % (id, random.randint(100, 999)), video_id) - # the following xml contains the actual configuration information on the video file(s) - vid_config = self._download_xml( + vid_config_request = compat_urllib_request.Request( 'http://www.miomio.tv/mioplayer/mioplayerconfigfiles/sina.php?{0}'.format(xml_config), - video_id) + headers=http_headers) - http_headers = { - 'Referer': 'http://www.miomio.tv%s' % mioplayer_path, - } + # the following xml contains the actual configuration information on the video file(s) + vid_config = self._download_xml(vid_config_request, video_id) if not int_or_none(xpath_text(vid_config, 'timelength')): raise ExtractorError('Unable to load videos!', expected=True) diff --git a/youtube_dl/extractor/mit.py b/youtube_dl/extractor/mit.py index f088ab9e2..29ca45778 100644 --- a/youtube_dl/extractor/mit.py +++ b/youtube_dl/extractor/mit.py @@ -86,7 +86,7 @@ class MITIE(TechTVMITIE): webpage = self._download_webpage(url, page_title) embed_url = self._search_regex( r'<iframe .*?src="(.+?)"', webpage, 'embed url') - return self.url_result(embed_url, ie='TechTVMIT') + return self.url_result(embed_url) class OCWMITIE(InfoExtractor): diff --git a/youtube_dl/extractor/mitele.py b/youtube_dl/extractor/mitele.py index 852d72266..c595f2077 100644 --- a/youtube_dl/extractor/mitele.py +++ b/youtube_dl/extractor/mitele.py @@ -1,74 +1,89 @@ from __future__ import unicode_literals -import json - from .common import InfoExtractor from ..compat import ( compat_urllib_parse, - compat_urllib_parse_unquote, compat_urlparse, ) from ..utils import ( + encode_dict, get_element_by_attribute, - parse_duration, - strip_jsonp, + int_or_none, ) class MiTeleIE(InfoExtractor): - IE_NAME = 'mitele.es' + IE_DESC = 'mitele.es' _VALID_URL = r'http://www\.mitele\.es/[^/]+/[^/]+/[^/]+/(?P<id>[^/]+)/' _TESTS = [{ 'url': 'http://www.mitele.es/programas-tv/diario-de/la-redaccion/programa-144/', + 'md5': '0ff1a13aebb35d9bc14081ff633dd324', 'info_dict': { - 'id': '0fce117d', - 'ext': 'mp4', - 'title': 'Programa 144 - Tor, la web invisible', - 'description': 'md5:3b6fce7eaa41b2d97358726378d9369f', + 'id': '0NF1jJnxS1Wu3pHrmvFyw2', 'display_id': 'programa-144', + 'ext': 'flv', + 'title': 'Tor, la web invisible', + 'description': 'md5:3b6fce7eaa41b2d97358726378d9369f', + 'thumbnail': 're:(?i)^https?://.*\.jpg$', 'duration': 2913, }, - 'params': { - # m3u8 download - 'skip_download': True, - }, }] def _real_extract(self, url): - episode = self._match_id(url) - webpage = self._download_webpage(url, episode) - embed_data_json = self._search_regex( - r'(?s)MSV\.embedData\[.*?\]\s*=\s*({.*?});', webpage, 'embed data', - ).replace('\'', '"') - embed_data = json.loads(embed_data_json) + display_id = self._match_id(url) + + webpage = self._download_webpage(url, display_id) + + config_url = self._search_regex( + r'data-config\s*=\s*"([^"]+)"', webpage, 'data config url') + config_url = compat_urlparse.urljoin(url, config_url) - domain = embed_data['mediaUrl'] - if not domain.startswith('http'): - # only happens in telecinco.es videos - domain = 'http://' + domain - info_url = compat_urlparse.urljoin( - domain, - compat_urllib_parse_unquote(embed_data['flashvars']['host']) - ) - info_el = self._download_xml(info_url, episode).find('./video/info') + config = self._download_json( + config_url, display_id, 'Downloading config JSON') - video_link = info_el.find('videoUrl/link').text - token_query = compat_urllib_parse.urlencode({'id': video_link}) - token_info = self._download_json( - embed_data['flashvars']['ov_tk'] + '?' + token_query, - episode, - transform_source=strip_jsonp - ) - formats = self._extract_m3u8_formats( - token_info['tokenizedUrl'], episode, ext='mp4') + mmc = self._download_json( + config['services']['mmc'], display_id, 'Downloading mmc JSON') + + formats = [] + for location in mmc['locations']: + gat = self._proto_relative_url(location.get('gat'), 'http:') + bas = location.get('bas') + loc = location.get('loc') + ogn = location.get('ogn') + if None in (gat, bas, loc, ogn): + continue + token_data = { + 'bas': bas, + 'icd': loc, + 'ogn': ogn, + 'sta': '0', + } + media = self._download_json( + '%s/?%s' % (gat, compat_urllib_parse.urlencode(encode_dict(token_data))), + display_id, 'Downloading %s JSON' % location['loc']) + file_ = media.get('file') + if not file_: + continue + formats.extend(self._extract_f4m_formats( + file_ + '&hdcore=3.2.0&plugin=aasp-3.2.0.77.18', + display_id, f4m_id=loc)) + + title = self._search_regex( + r'class="Destacado-text"[^>]*>\s*<strong>([^<]+)</strong>', webpage, 'title') + + video_id = self._search_regex( + r'data-media-id\s*=\s*"([^"]+)"', webpage, + 'data media id', default=None) or display_id + thumbnail = config.get('poster', {}).get('imageUrl') + duration = int_or_none(mmc.get('duration')) return { - 'id': embed_data['videoId'], - 'display_id': episode, - 'title': info_el.find('title').text, - 'formats': formats, + 'id': video_id, + 'display_id': display_id, + 'title': title, 'description': get_element_by_attribute('class', 'text', webpage), - 'thumbnail': info_el.find('thumb').text, - 'duration': parse_duration(info_el.find('duration').text), + 'thumbnail': thumbnail, + 'duration': duration, + 'formats': formats, } diff --git a/youtube_dl/extractor/moniker.py b/youtube_dl/extractor/moniker.py index 69e4bcd1a..7c0c4e50e 100644 --- a/youtube_dl/extractor/moniker.py +++ b/youtube_dl/extractor/moniker.py @@ -17,7 +17,7 @@ from ..utils import ( class MonikerIE(InfoExtractor): IE_DESC = 'allmyvideos.net and vidspot.net' - _VALID_URL = r'https?://(?:www\.)?(?:allmyvideos|vidspot)\.net/(?P<id>[a-zA-Z0-9_-]+)' + _VALID_URL = r'https?://(?:www\.)?(?:allmyvideos|vidspot)\.net/(?:(?:2|v)/v-)?(?P<id>[a-zA-Z0-9_-]+)' _TESTS = [{ 'url': 'http://allmyvideos.net/jih3nce3x6wn', @@ -46,6 +46,18 @@ class MonikerIE(InfoExtractor): }, { 'url': 'https://www.vidspot.net/l2ngsmhs8ci5', 'only_matching': True, + }, { + 'url': 'http://vidspot.net/2/v-ywDf99', + 'md5': '5f8254ce12df30479428b0152fb8e7ba', + 'info_dict': { + 'id': 'ywDf99', + 'ext': 'mp4', + 'title': 'IL FAIT LE MALIN EN PORSHE CAYENNE ( mais pas pour longtemps)', + 'description': 'IL FAIT LE MALIN EN PORSHE CAYENNE.', + }, + }, { + 'url': 'http://allmyvideos.net/v/v-HXZm5t', + 'only_matching': True, }] def _real_extract(self, url): @@ -64,18 +76,30 @@ class MonikerIE(InfoExtractor): raise ExtractorError( '%s returned error: %s' % (self.IE_NAME, error), expected=True) - fields = re.findall(r'type="hidden" name="(.+?)"\s* value="?(.+?)">', orig_webpage) - data = dict(fields) + builtin_url = self._search_regex( + r'<iframe[^>]+src=(["\'])(?P<url>.+?/builtin-.+?)\1', + orig_webpage, 'builtin URL', default=None, group='url') - post = compat_urllib_parse.urlencode(data) - headers = { - b'Content-Type': b'application/x-www-form-urlencoded', - } - req = compat_urllib_request.Request(url, post, headers) - webpage = self._download_webpage( - req, video_id, note='Downloading video page ...') + if builtin_url: + req = compat_urllib_request.Request(builtin_url) + req.add_header('Referer', url) + webpage = self._download_webpage(req, video_id, 'Downloading builtin page') + title = self._og_search_title(orig_webpage).strip() + description = self._og_search_description(orig_webpage).strip() + else: + fields = re.findall(r'type="hidden" name="(.+?)"\s* value="?(.+?)">', orig_webpage) + data = dict(fields) + + post = compat_urllib_parse.urlencode(data) + headers = { + b'Content-Type': b'application/x-www-form-urlencoded', + } + req = compat_urllib_request.Request(url, post, headers) + webpage = self._download_webpage( + req, video_id, note='Downloading video page ...') - title = os.path.splitext(data['fname'])[0] + title = os.path.splitext(data['fname'])[0] + description = None # Could be several links with different quality links = re.findall(r'"file" : "?(.+?)",', webpage) @@ -89,5 +113,6 @@ class MonikerIE(InfoExtractor): return { 'id': video_id, 'title': title, + 'description': description, 'formats': formats, } diff --git a/youtube_dl/extractor/movieclips.py b/youtube_dl/extractor/movieclips.py index 04e17d055..b8c43a163 100644 --- a/youtube_dl/extractor/movieclips.py +++ b/youtube_dl/extractor/movieclips.py @@ -1,80 +1,42 @@ +# coding: utf-8 from __future__ import unicode_literals -import re - from .common import InfoExtractor from ..compat import ( - compat_str, -) -from ..utils import ( - ExtractorError, - clean_html, + compat_urllib_request, ) class MovieClipsIE(InfoExtractor): - _VALID_URL = r'https?://movieclips\.com/(?P<id>[\da-zA-Z]+)(?:-(?P<display_id>[\da-z-]+))?' + _VALID_URL = r'https?://(?:www.)?movieclips\.com/videos/(?P<id>[^/?#]+)' _TEST = { - 'url': 'http://movieclips.com/Wy7ZU-my-week-with-marilyn-movie-do-you-love-me/', + 'url': 'http://www.movieclips.com/videos/warcraft-trailer-1-561180739597?autoPlay=true&playlistId=5', 'info_dict': { - 'id': 'Wy7ZU', - 'display_id': 'my-week-with-marilyn-movie-do-you-love-me', + 'id': 'pKIGmG83AqD9', + 'display_id': 'warcraft-trailer-1-561180739597', 'ext': 'mp4', - 'title': 'My Week with Marilyn - Do You Love Me?', - 'description': 'md5:e86795bd332fe3cff461e7c8dc542acb', + 'title': 'Warcraft Trailer 1', + 'description': 'Watch Trailer 1 from Warcraft (2016). Legendary’s WARCRAFT is a 3D epic adventure of world-colliding conflict based.', 'thumbnail': 're:^https?://.*\.jpg$', }, - 'params': { - # rtmp download - 'skip_download': True, - } + 'add_ie': ['ThePlatform'], } def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - video_id = mobj.group('id') - display_id = mobj.group('display_id') - show_id = display_id or video_id - - config = self._download_xml( - 'http://config.movieclips.com/player/config/%s' % video_id, - show_id, 'Downloading player config') - - if config.find('./country-region').text == 'false': - raise ExtractorError( - '%s said: %s' % (self.IE_NAME, config.find('./region_alert').text), expected=True) - - properties = config.find('./video/properties') - smil_file = properties.attrib['smil_file'] + display_id = self._match_id(url) - smil = self._download_xml(smil_file, show_id, 'Downloading SMIL') - base_url = smil.find('./head/meta').attrib['base'] - - formats = [] - for video in smil.findall('./body/switch/video'): - vbr = int(video.attrib['system-bitrate']) / 1000 - src = video.attrib['src'] - formats.append({ - 'url': base_url, - 'play_path': src, - 'ext': src.split(':')[0], - 'vbr': vbr, - 'format_id': '%dk' % vbr, - }) - - self._sort_formats(formats) - - title = '%s - %s' % (properties.attrib['clip_movie_title'], properties.attrib['clip_title']) - description = clean_html(compat_str(properties.attrib['clip_description'])) - thumbnail = properties.attrib['image'] - categories = properties.attrib['clip_categories'].split(',') + req = compat_urllib_request.Request(url) + # it doesn't work if it thinks the browser it's too old + req.add_header('User-Agent', 'Mozilla/5.0 (X11; Linux x86_64; rv:10.0) Gecko/20150101 Firefox/43.0 (Chrome)') + webpage = self._download_webpage(req, display_id) + theplatform_link = self._html_search_regex(r'src="(http://player.theplatform.com/p/.*?)"', webpage, 'theplatform link') + title = self._html_search_regex(r'<title[^>]*>([^>]+)-\s*\d+\s*|\s*Movieclips.com</title>', webpage, 'title') + description = self._html_search_meta('description', webpage) return { - 'id': video_id, - 'display_id': display_id, + '_type': 'url_transparent', + 'url': theplatform_link, 'title': title, + 'display_id': display_id, 'description': description, - 'thumbnail': thumbnail, - 'categories': categories, - 'formats': formats, } diff --git a/youtube_dl/extractor/mtv.py b/youtube_dl/extractor/mtv.py index a597714e9..302c9bf35 100644 --- a/youtube_dl/extractor/mtv.py +++ b/youtube_dl/extractor/mtv.py @@ -200,7 +200,13 @@ class MTVServicesInfoExtractor(InfoExtractor): if mgid is None or ':' not in mgid: mgid = self._search_regex( [r'data-mgid="(.*?)"', r'swfobject.embedSWF\(".*?(mgid:.*?)"'], - webpage, 'mgid') + webpage, 'mgid', default=None) + + if not mgid: + sm4_embed = self._html_search_meta( + 'sm4:video:embed', webpage, 'sm4 embed', default='') + mgid = self._search_regex( + r'embed/(mgid:.+?)["\'&?/]', sm4_embed, 'mgid') videos_info = self._get_videos_info(mgid) return videos_info @@ -222,6 +228,13 @@ class MTVServicesEmbeddedIE(MTVServicesInfoExtractor): }, } + @staticmethod + def _extract_url(webpage): + mobj = re.search( + r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//media.mtvnservices.com/embed/.+?)\1', webpage) + if mobj: + return mobj.group('url') + def _get_feed_url(self, uri): video_id = self._id_from_uri(uri) site_id = uri.replace(video_id, '') diff --git a/youtube_dl/extractor/naver.py b/youtube_dl/extractor/naver.py index 925967753..1f5fc2145 100644 --- a/youtube_dl/extractor/naver.py +++ b/youtube_dl/extractor/naver.py @@ -10,7 +10,6 @@ from ..compat import ( ) from ..utils import ( ExtractorError, - clean_html, ) @@ -46,11 +45,11 @@ class NaverIE(InfoExtractor): m_id = re.search(r'var rmcPlayer = new nhn.rmcnmv.RMCVideoPlayer\("(.+?)", "(.+?)"', webpage) if m_id is None: - m_error = re.search( - r'(?s)<div class="(?:nation_error|nation_box)">\s*(?:<!--.*?-->)?\s*<p class="[^"]+">(?P<msg>.+?)</p>\s*</div>', - webpage) - if m_error: - raise ExtractorError(clean_html(m_error.group('msg')), expected=True) + error = self._html_search_regex( + r'(?s)<div class="(?:nation_error|nation_box|error_box)">\s*(?:<!--.*?-->)?\s*<p class="[^"]+">(?P<msg>.+?)</p>\s*</div>', + webpage, 'error', default=None) + if error: + raise ExtractorError(error, expected=True) raise ExtractorError('couldn\'t extract vid and key') vid = m_id.group(1) key = m_id.group(2) diff --git a/youtube_dl/extractor/ndr.py b/youtube_dl/extractor/ndr.py index 79a13958b..16213eed9 100644 --- a/youtube_dl/extractor/ndr.py +++ b/youtube_dl/extractor/ndr.py @@ -1,130 +1,387 @@ -# encoding: utf-8 +# coding: utf-8 from __future__ import unicode_literals import re from .common import InfoExtractor from ..utils import ( - ExtractorError, + determine_ext, int_or_none, + parse_iso8601, qualities, - parse_duration, ) class NDRBaseIE(InfoExtractor): def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) - video_id = mobj.group('id') + display_id = next(group for group in mobj.groups() if group) + webpage = self._download_webpage(url, display_id) + return self._extract_embed(webpage, display_id) - page = self._download_webpage(url, video_id, 'Downloading page') - title = self._og_search_title(page).strip() - description = self._og_search_description(page) - if description: - description = description.strip() +class NDRIE(NDRBaseIE): + IE_NAME = 'ndr' + IE_DESC = 'NDR.de - Norddeutscher Rundfunk' + _VALID_URL = r'https?://www\.ndr\.de/(?:[^/]+/)*(?P<id>[^/?#]+),[\da-z]+\.html' + _TESTS = [{ + # httpVideo, same content id + 'url': 'http://www.ndr.de/fernsehen/Party-Poette-und-Parade,hafengeburtstag988.html', + 'md5': '6515bc255dc5c5f8c85bbc38e035a659', + 'info_dict': { + 'id': 'hafengeburtstag988', + 'display_id': 'Party-Poette-und-Parade', + 'ext': 'mp4', + 'title': 'Party, Pötte und Parade', + 'description': 'md5:ad14f9d2f91d3040b6930c697e5f6b4c', + 'uploader': 'ndrtv', + 'timestamp': 1431108900, + 'upload_date': '20150510', + 'duration': 3498, + }, + 'params': { + 'skip_download': True, + }, + }, { + # httpVideo, different content id + 'url': 'http://www.ndr.de/sport/fussball/40-Osnabrueck-spielt-sich-in-einen-Rausch,osna270.html', + 'md5': '1043ff203eab307f0c51702ec49e9a71', + 'info_dict': { + 'id': 'osna272', + 'display_id': '40-Osnabrueck-spielt-sich-in-einen-Rausch', + 'ext': 'mp4', + 'title': 'Osnabrück - Wehen Wiesbaden: Die Highlights', + 'description': 'md5:32e9b800b3d2d4008103752682d5dc01', + 'uploader': 'ndrtv', + 'timestamp': 1442059200, + 'upload_date': '20150912', + 'duration': 510, + }, + 'params': { + 'skip_download': True, + }, + }, { + # httpAudio, same content id + 'url': 'http://www.ndr.de/info/La-Valette-entgeht-der-Hinrichtung,audio51535.html', + 'md5': 'bb3cd38e24fbcc866d13b50ca59307b8', + 'info_dict': { + 'id': 'audio51535', + 'display_id': 'La-Valette-entgeht-der-Hinrichtung', + 'ext': 'mp3', + 'title': 'La Valette entgeht der Hinrichtung', + 'description': 'md5:22f9541913a40fe50091d5cdd7c9f536', + 'uploader': 'ndrinfo', + 'timestamp': 1290626100, + 'upload_date': '20140729', + 'duration': 884, + }, + 'params': { + 'skip_download': True, + }, + }, { + 'url': 'https://www.ndr.de/Fettes-Brot-Ferris-MC-und-Thees-Uhlmann-live-on-stage,festivalsommer116.html', + 'only_matching': True, + }] + + def _extract_embed(self, webpage, display_id): + embed_url = self._html_search_meta( + 'embedURL', webpage, 'embed URL', fatal=True) + description = self._search_regex( + r'<p[^>]+itemprop="description">([^<]+)</p>', + webpage, 'description', fatal=False) + timestamp = parse_iso8601( + self._search_regex( + r'<span itemprop="datePublished" content="([^"]+)">', + webpage, 'upload date', fatal=False)) + return { + '_type': 'url_transparent', + 'url': embed_url, + 'display_id': display_id, + 'description': description, + 'timestamp': timestamp, + } - duration = int_or_none(self._html_search_regex(r'duration: (\d+),\n', page, 'duration', default=None)) - if not duration: - duration = parse_duration(self._html_search_regex( - r'(<span class="min">\d+</span>:<span class="sec">\d+</span>)', - page, 'duration', default=None)) - formats = [] +class NJoyIE(NDRBaseIE): + IE_NAME = 'njoy' + IE_DESC = 'N-JOY' + _VALID_URL = r'https?://www\.n-joy\.de/(?:[^/]+/)*(?:(?P<display_id>[^/?#]+),)?(?P<id>[\da-z]+)\.html' + _TESTS = [{ + # httpVideo, same content id + 'url': 'http://www.n-joy.de/entertainment/comedy/comedy_contest/Benaissa-beim-NDR-Comedy-Contest,comedycontest2480.html', + 'md5': 'cb63be60cd6f9dd75218803146d8dc67', + 'info_dict': { + 'id': 'comedycontest2480', + 'display_id': 'Benaissa-beim-NDR-Comedy-Contest', + 'ext': 'mp4', + 'title': 'Benaissa beim NDR Comedy Contest', + 'description': 'md5:f057a6c4e1c728b10d33b5ffd36ddc39', + 'uploader': 'ndrtv', + 'upload_date': '20141129', + 'duration': 654, + }, + 'params': { + 'skip_download': True, + }, + }, { + # httpVideo, different content id + 'url': 'http://www.n-joy.de/musik/Das-frueheste-DJ-Set-des-Nordens-live-mit-Felix-Jaehn-,felixjaehn168.html', + 'md5': '417660fffa90e6df2fda19f1b40a64d8', + 'info_dict': { + 'id': 'dockville882', + 'display_id': 'Das-frueheste-DJ-Set-des-Nordens-live-mit-Felix-Jaehn-', + 'ext': 'mp4', + 'title': '"Ich hab noch nie" mit Felix Jaehn', + 'description': 'md5:85dd312d53be1b99e1f998a16452a2f3', + 'uploader': 'njoy', + 'upload_date': '20150822', + 'duration': 211, + }, + 'params': { + 'skip_download': True, + }, + }, { + 'url': 'http://www.n-joy.de/radio/webradio/morningshow209.html', + 'only_matching': True, + }] + + def _extract_embed(self, webpage, display_id): + video_id = self._search_regex( + r'<iframe[^>]+id="pp_([\da-z]+)"', webpage, 'embed id') + description = self._search_regex( + r'<div[^>]+class="subline"[^>]*>[^<]+</div>\s*<p>([^<]+)</p>', + webpage, 'description', fatal=False) + return { + '_type': 'url_transparent', + 'ie_key': 'NDREmbedBase', + 'url': 'ndr:%s' % video_id, + 'display_id': display_id, + 'description': description, + } + + +class NDREmbedBaseIE(InfoExtractor): + IE_NAME = 'ndr:embed:base' + _VALID_URL = r'(?:ndr:(?P<id_s>[\da-z]+)|https?://www\.ndr\.de/(?P<id>[\da-z]+)-ppjson\.json)' + _TESTS = [{ + 'url': 'ndr:soundcheck3366', + 'only_matching': True, + }, { + 'url': 'http://www.ndr.de/soundcheck3366-ppjson.json', + 'only_matching': True, + }] + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + video_id = mobj.group('id') or mobj.group('id_s') + + ppjson = self._download_json( + 'http://www.ndr.de/%s-ppjson.json' % video_id, video_id) - mp3_url = re.search(r'''\{src:'(?P<audio>[^']+)', type:"audio/mp3"},''', page) - if mp3_url: - formats.append({ - 'url': mp3_url.group('audio'), - 'format_id': 'mp3', - }) + playlist = ppjson['playlist'] - thumbnail = None + formats = [] + quality_key = qualities(('xs', 's', 'm', 'l', 'xl')) - video_url = re.search(r'''3: \{src:'(?P<video>.+?)\.(lo|hi|hq)\.mp4', type:"video/mp4"},''', page) - if video_url: - thumbnails = re.findall(r'''\d+: \{src: "([^"]+)"(?: \|\| '[^']+')?, quality: '([^']+)'}''', page) - if thumbnails: - quality_key = qualities(['xs', 's', 'm', 'l', 'xl']) - largest = max(thumbnails, key=lambda thumb: quality_key(thumb[1])) - thumbnail = 'http://www.ndr.de' + largest[0] + for format_id, f in playlist.items(): + src = f.get('src') + if not src: + continue + ext = determine_ext(src, None) + if ext == 'f4m': + formats.extend(self._extract_f4m_formats( + src + '?hdcore=3.7.0&plugin=aasp-3.7.0.39.44', video_id, f4m_id='hds')) + elif ext == 'm3u8': + formats.extend(self._extract_m3u8_formats( + src, video_id, m3u8_id='hls', entry_protocol='m3u8_native')) + else: + quality = f.get('quality') + ff = { + 'url': src, + 'format_id': quality or format_id, + 'quality': quality_key(quality), + } + type_ = f.get('type') + if type_ and type_.split('/')[0] == 'audio': + ff['vcodec'] = 'none' + ff['ext'] = ext or 'mp3' + formats.append(ff) + self._sort_formats(formats) - for format_id in 'lo', 'hi', 'hq': - formats.append({ - 'url': '%s.%s.mp4' % (video_url.group('video'), format_id), - 'format_id': format_id, - }) + config = playlist['config'] - if not formats: - raise ExtractorError('No media links available for %s' % video_id) + live = playlist.get('config', {}).get('streamType') in ['httpVideoLive', 'httpAudioLive'] + title = config['title'] + if live: + title = self._live_title(title) + uploader = ppjson.get('config', {}).get('branding') + upload_date = ppjson.get('config', {}).get('publicationDate') + duration = int_or_none(config.get('duration')) + + thumbnails = [{ + 'id': thumbnail.get('quality') or thumbnail_id, + 'url': thumbnail['src'], + 'preference': quality_key(thumbnail.get('quality')), + } for thumbnail_id, thumbnail in config.get('poster', {}).items() if thumbnail.get('src')] return { 'id': video_id, 'title': title, - 'description': description, - 'thumbnail': thumbnail, + 'is_live': live, + 'uploader': uploader if uploader != '-' else None, + 'upload_date': upload_date[0:8] if upload_date else None, 'duration': duration, + 'thumbnails': thumbnails, 'formats': formats, } -class NDRIE(NDRBaseIE): - IE_NAME = 'ndr' - IE_DESC = 'NDR.de - Mediathek' - _VALID_URL = r'https?://www\.ndr\.de/.+?(?P<id>\d+)\.html' - - _TESTS = [ - { - 'url': 'http://www.ndr.de/fernsehen/sendungen/nordmagazin/Kartoffeltage-in-der-Lewitz,nordmagazin25866.html', - 'md5': '5bc5f5b92c82c0f8b26cddca34f8bb2c', - 'note': 'Video file', - 'info_dict': { - 'id': '25866', - 'ext': 'mp4', - 'title': 'Kartoffeltage in der Lewitz', - 'description': 'md5:48c4c04dde604c8a9971b3d4e3b9eaa8', - 'duration': 166, - }, - 'skip': '404 Not found', - }, - { - 'url': 'http://www.ndr.de/fernsehen/Party-Poette-und-Parade,hafengeburtstag988.html', - 'md5': 'dadc003c55ae12a5d2f6bd436cd73f59', - 'info_dict': { - 'id': '988', - 'ext': 'mp4', - 'title': 'Party, Pötte und Parade', - 'description': 'Hunderttausende feiern zwischen Speicherstadt und St. Pauli den 826. Hafengeburtstag. Die NDR Sondersendung zeigt die schönsten und spektakulärsten Bilder vom Auftakt.', - 'duration': 3498, - }, - }, - { - 'url': 'http://www.ndr.de/info/audio51535.html', - 'md5': 'bb3cd38e24fbcc866d13b50ca59307b8', - 'note': 'Audio file', - 'info_dict': { - 'id': '51535', - 'ext': 'mp3', - 'title': 'La Valette entgeht der Hinrichtung', - 'description': 'md5:22f9541913a40fe50091d5cdd7c9f536', - 'duration': 884, - } - } - ] - +class NDREmbedIE(NDREmbedBaseIE): + IE_NAME = 'ndr:embed' + _VALID_URL = r'https?://www\.ndr\.de/(?:[^/]+/)*(?P<id>[\da-z]+)-(?:player|externalPlayer)\.html' + _TESTS = [{ + 'url': 'http://www.ndr.de/fernsehen/sendungen/ndr_aktuell/ndraktuell28488-player.html', + 'md5': '8b9306142fe65bbdefb5ce24edb6b0a9', + 'info_dict': { + 'id': 'ndraktuell28488', + 'ext': 'mp4', + 'title': 'Norddeutschland begrüßt Flüchtlinge', + 'is_live': False, + 'uploader': 'ndrtv', + 'upload_date': '20150907', + 'duration': 132, + }, + }, { + 'url': 'http://www.ndr.de/ndr2/events/soundcheck/soundcheck3366-player.html', + 'md5': '002085c44bae38802d94ae5802a36e78', + 'info_dict': { + 'id': 'soundcheck3366', + 'ext': 'mp4', + 'title': 'Ella Henderson braucht Vergleiche nicht zu scheuen', + 'is_live': False, + 'uploader': 'ndr2', + 'upload_date': '20150912', + 'duration': 3554, + }, + 'params': { + 'skip_download': True, + }, + }, { + 'url': 'http://www.ndr.de/info/audio51535-player.html', + 'md5': 'bb3cd38e24fbcc866d13b50ca59307b8', + 'info_dict': { + 'id': 'audio51535', + 'ext': 'mp3', + 'title': 'La Valette entgeht der Hinrichtung', + 'is_live': False, + 'uploader': 'ndrinfo', + 'upload_date': '20140729', + 'duration': 884, + }, + 'params': { + 'skip_download': True, + }, + }, { + 'url': 'http://www.ndr.de/fernsehen/sendungen/visite/visite11010-externalPlayer.html', + 'md5': 'ae57f80511c1e1f2fd0d0d3d31aeae7c', + 'info_dict': { + 'id': 'visite11010', + 'ext': 'mp4', + 'title': 'Visite - die ganze Sendung', + 'is_live': False, + 'uploader': 'ndrtv', + 'upload_date': '20150902', + 'duration': 3525, + }, + 'params': { + 'skip_download': True, + }, + }, { + # httpVideoLive + 'url': 'http://www.ndr.de/fernsehen/livestream/livestream217-externalPlayer.html', + 'info_dict': { + 'id': 'livestream217', + 'ext': 'flv', + 'title': 're:^NDR Fernsehen Niedersachsen \d{4}-\d{2}-\d{2} \d{2}:\d{2}$', + 'is_live': True, + 'upload_date': '20150910', + }, + 'params': { + 'skip_download': True, + }, + }, { + 'url': 'http://www.ndr.de/ndrkultur/audio255020-player.html', + 'only_matching': True, + }, { + 'url': 'http://www.ndr.de/fernsehen/sendungen/nordtour/nordtour7124-player.html', + 'only_matching': True, + }, { + 'url': 'http://www.ndr.de/kultur/film/videos/videoimport10424-player.html', + 'only_matching': True, + }, { + 'url': 'http://www.ndr.de/fernsehen/sendungen/hamburg_journal/hamj43006-player.html', + 'only_matching': True, + }, { + 'url': 'http://www.ndr.de/fernsehen/sendungen/weltbilder/weltbilder4518-player.html', + 'only_matching': True, + }, { + 'url': 'http://www.ndr.de/fernsehen/doku952-player.html', + 'only_matching': True, + }] -class NJoyIE(NDRBaseIE): - IE_NAME = 'N-JOY' - _VALID_URL = r'https?://www\.n-joy\.de/.+?(?P<id>\d+)\.html' - _TEST = { - 'url': 'http://www.n-joy.de/entertainment/comedy/comedy_contest/Benaissa-beim-NDR-Comedy-Contest,comedycontest2480.html', - 'md5': 'cb63be60cd6f9dd75218803146d8dc67', +class NJoyEmbedIE(NDREmbedBaseIE): + IE_NAME = 'njoy:embed' + _VALID_URL = r'https?://www\.n-joy\.de/(?:[^/]+/)*(?P<id>[\da-z]+)-(?:player|externalPlayer)_[^/]+\.html' + _TESTS = [{ + # httpVideo + 'url': 'http://www.n-joy.de/events/reeperbahnfestival/doku948-player_image-bc168e87-5263-4d6d-bd27-bb643005a6de_theme-n-joy.html', + 'md5': '8483cbfe2320bd4d28a349d62d88bd74', 'info_dict': { - 'id': '2480', + 'id': 'doku948', 'ext': 'mp4', - 'title': 'Benaissa beim NDR Comedy Contest', - 'description': 'Von seinem sehr "behaarten" Leben lässt sich Benaissa trotz aller Schwierigkeiten nicht unterkriegen.', - 'duration': 654, - } - } + 'title': 'Zehn Jahre Reeperbahn Festival - die Doku', + 'is_live': False, + 'upload_date': '20150807', + 'duration': 1011, + }, + }, { + # httpAudio + 'url': 'http://www.n-joy.de/news_wissen/stefanrichter100-player_image-d5e938b1-f21a-4b9a-86b8-aaba8bca3a13_theme-n-joy.html', + 'md5': 'd989f80f28ac954430f7b8a48197188a', + 'info_dict': { + 'id': 'stefanrichter100', + 'ext': 'mp3', + 'title': 'Interview mit einem Augenzeugen', + 'is_live': False, + 'uploader': 'njoy', + 'upload_date': '20150909', + 'duration': 140, + }, + 'params': { + 'skip_download': True, + }, + }, { + # httpAudioLive, no explicit ext + 'url': 'http://www.n-joy.de/news_wissen/webradioweltweit100-player_image-3fec0484-2244-4565-8fb8-ed25fd28b173_theme-n-joy.html', + 'info_dict': { + 'id': 'webradioweltweit100', + 'ext': 'mp3', + 'title': 're:^N-JOY Weltweit \d{4}-\d{2}-\d{2} \d{2}:\d{2}$', + 'is_live': True, + 'uploader': 'njoy', + 'upload_date': '20150810', + }, + 'params': { + 'skip_download': True, + }, + }, { + 'url': 'http://www.n-joy.de/musik/dockville882-player_image-3905259e-0803-4764-ac72-8b7de077d80a_theme-n-joy.html', + 'only_matching': True, + }, { + 'url': 'http://www.n-joy.de/radio/sendungen/morningshow/urlaubsfotos190-player_image-066a5df1-5c95-49ec-a323-941d848718db_theme-n-joy.html', + 'only_matching': True, + }, { + 'url': 'http://www.n-joy.de/entertainment/comedy/krudetv290-player_image-ab261bfe-51bf-4bf3-87ba-c5122ee35b3d_theme-n-joy.html', + 'only_matching': True, + }] diff --git a/youtube_dl/extractor/nextmedia.py b/youtube_dl/extractor/nextmedia.py index c10784f6b..d1688457f 100644 --- a/youtube_dl/extractor/nextmedia.py +++ b/youtube_dl/extractor/nextmedia.py @@ -126,7 +126,8 @@ class AppleDailyIE(NextMediaIE): 'thumbnail': 're:^https?://.*\.jpg$', 'description': 'md5:23c0aac567dc08c9c16a3161a2c2e3cd', 'upload_date': '20150128', - } + }, + 'skip': 'redirect to http://www.appledaily.com.tw/animation/', }, { # No thumbnail 'url': 'http://www.appledaily.com.tw/animation/realtimenews/new/20150128/5003673/', @@ -140,10 +141,19 @@ class AppleDailyIE(NextMediaIE): }, 'expected_warnings': [ 'video thumbnail', - ] + ], + 'skip': 'redirect to http://www.appledaily.com.tw/animation/', }, { 'url': 'http://www.appledaily.com.tw/appledaily/article/supplement/20140417/35770334/', - 'only_matching': True, + 'md5': 'eaa20e6b9df418c912d7f5dec2ba734d', + 'info_dict': { + 'id': '35770334', + 'ext': 'mp4', + 'title': '咖啡占卜測 XU裝熟指數', + 'thumbnail': 're:^https?://.*\.jpg$', + 'description': 'md5:7b859991a6a4fedbdf3dd3b66545c748', + 'upload_date': '20140417', + }, }] _URL_PATTERN = r'\{url: \'(.+)\'\}' diff --git a/youtube_dl/extractor/nfl.py b/youtube_dl/extractor/nfl.py index dc54634a5..200874d68 100644 --- a/youtube_dl/extractor/nfl.py +++ b/youtube_dl/extractor/nfl.py @@ -16,53 +16,118 @@ from ..utils import ( class NFLIE(InfoExtractor): IE_NAME = 'nfl.com' - _VALID_URL = r'''(?x)https?:// - (?P<host>(?:www\.)?(?:nfl\.com|.*?\.clubs\.nfl\.com))/ - (?:.+?/)* - (?P<id>(?:[a-z0-9]{16}|\w{8}\-(?:\w{4}\-){3}\w{12}))''' - _TESTS = [ - { - 'url': 'http://www.nfl.com/videos/nfl-game-highlights/0ap3000000398478/Week-3-Redskins-vs-Eagles-highlights', - 'md5': '394ef771ddcd1354f665b471d78ec4c6', - 'info_dict': { - 'id': '0ap3000000398478', - 'ext': 'mp4', - 'title': 'Week 3: Redskins vs. Eagles highlights', - 'description': 'md5:56323bfb0ac4ee5ab24bd05fdf3bf478', - 'upload_date': '20140921', - 'timestamp': 1411337580, - 'thumbnail': 're:^https?://.*\.jpg$', - } + _VALID_URL = r'''(?x) + https?:// + (?P<host> + (?:www\.)? + (?: + (?: + nfl| + buffalobills| + miamidolphins| + patriots| + newyorkjets| + baltimoreravens| + bengals| + clevelandbrowns| + steelers| + houstontexans| + colts| + jaguars| + titansonline| + denverbroncos| + kcchiefs| + raiders| + chargers| + dallascowboys| + giants| + philadelphiaeagles| + redskins| + chicagobears| + detroitlions| + packers| + vikings| + atlantafalcons| + panthers| + neworleanssaints| + buccaneers| + azcardinals| + stlouisrams| + 49ers| + seahawks + )\.com| + .+?\.clubs\.nfl\.com + ) + )/ + (?:.+?/)* + (?P<id>[^/#?&]+) + ''' + _TESTS = [{ + 'url': 'http://www.nfl.com/videos/nfl-game-highlights/0ap3000000398478/Week-3-Redskins-vs-Eagles-highlights', + 'md5': '394ef771ddcd1354f665b471d78ec4c6', + 'info_dict': { + 'id': '0ap3000000398478', + 'ext': 'mp4', + 'title': 'Week 3: Redskins vs. Eagles highlights', + 'description': 'md5:56323bfb0ac4ee5ab24bd05fdf3bf478', + 'upload_date': '20140921', + 'timestamp': 1411337580, + 'thumbnail': 're:^https?://.*\.jpg$', + } + }, { + 'url': 'http://prod.www.steelers.clubs.nfl.com/video-and-audio/videos/LIVE_Post_Game_vs_Browns/9d72f26a-9e2b-4718-84d3-09fb4046c266', + 'md5': 'cf85bdb4bc49f6e9d3816d130c78279c', + 'info_dict': { + 'id': '9d72f26a-9e2b-4718-84d3-09fb4046c266', + 'ext': 'mp4', + 'title': 'LIVE: Post Game vs. Browns', + 'description': 'md5:6a97f7e5ebeb4c0e69a418a89e0636e8', + 'upload_date': '20131229', + 'timestamp': 1388354455, + 'thumbnail': 're:^https?://.*\.jpg$', + } + }, { + 'url': 'http://www.nfl.com/news/story/0ap3000000467586/article/patriots-seahawks-involved-in-lategame-skirmish', + 'info_dict': { + 'id': '0ap3000000467607', + 'ext': 'mp4', + 'title': 'Frustrations flare on the field', + 'description': 'Emotions ran high at the end of the Super Bowl on both sides of the ball after a dramatic finish.', + 'timestamp': 1422850320, + 'upload_date': '20150202', }, - { - 'url': 'http://prod.www.steelers.clubs.nfl.com/video-and-audio/videos/LIVE_Post_Game_vs_Browns/9d72f26a-9e2b-4718-84d3-09fb4046c266', - 'md5': 'cf85bdb4bc49f6e9d3816d130c78279c', - 'info_dict': { - 'id': '9d72f26a-9e2b-4718-84d3-09fb4046c266', - 'ext': 'mp4', - 'title': 'LIVE: Post Game vs. Browns', - 'description': 'md5:6a97f7e5ebeb4c0e69a418a89e0636e8', - 'upload_date': '20131229', - 'timestamp': 1388354455, - 'thumbnail': 're:^https?://.*\.jpg$', - } + }, { + 'url': 'http://www.patriots.com/video/2015/09/18/10-days-gillette', + 'md5': '4c319e2f625ffd0b481b4382c6fc124c', + 'info_dict': { + 'id': 'n-238346', + 'ext': 'mp4', + 'title': '10 Days at Gillette', + 'description': 'md5:8cd9cd48fac16de596eadc0b24add951', + 'timestamp': 1442618809, + 'upload_date': '20150918', }, - { - 'url': 'http://www.nfl.com/news/story/0ap3000000467586/article/patriots-seahawks-involved-in-lategame-skirmish', - 'info_dict': { - 'id': '0ap3000000467607', - 'ext': 'mp4', - 'title': 'Frustrations flare on the field', - 'description': 'Emotions ran high at the end of the Super Bowl on both sides of the ball after a dramatic finish.', - 'timestamp': 1422850320, - 'upload_date': '20150202', - }, + }, { + # lowercase data-contentid + 'url': 'http://www.steelers.com/news/article-1/Tomlin-on-Ben-getting-Vick-ready/56399c96-4160-48cf-a7ad-1d17d4a3aef7', + 'info_dict': { + 'id': '12693586-6ea9-4743-9c1c-02c59e4a5ef2', + 'ext': 'mp4', + 'title': 'Tomlin looks ahead to Ravens on a short week', + 'description': 'md5:32f3f7b139f43913181d5cbb24ecad75', + 'timestamp': 1443459651, + 'upload_date': '20150928', }, - { - 'url': 'http://www.nfl.com/videos/nfl-network-top-ten/09000d5d810a6bd4/Top-10-Gutsiest-Performances-Jack-Youngblood', - 'only_matching': True, - } - ] + 'params': { + 'skip_download': True, + }, + }, { + 'url': 'http://www.nfl.com/videos/nfl-network-top-ten/09000d5d810a6bd4/Top-10-Gutsiest-Performances-Jack-Youngblood', + 'only_matching': True, + }, { + 'url': 'http://www.buffalobills.com/video/videos/Rex_Ryan_Show_World_Wide_Rex/b1dcfab2-3190-4bb1-bfc0-d6e603d6601a', + 'only_matching': True, + }] @staticmethod def prepend_host(host, url): @@ -95,13 +160,14 @@ class NFLIE(InfoExtractor): webpage = self._download_webpage(url, video_id) config_url = NFLIE.prepend_host(host, self._search_regex( - r'(?:config|configURL)\s*:\s*"([^"]+)"', webpage, 'config URL', - default='static/content/static/config/video/config.json')) + r'(?:(?:config|configURL)\s*:\s*|<nflcs:avplayer[^>]+data-config\s*=\s*)(["\'])(?P<config>.+?)\1', + webpage, 'config URL', default='static/content/static/config/video/config.json', + group='config')) # For articles, the id in the url is not the video id video_id = self._search_regex( - r'contentId\s*:\s*"([^"]+)"', webpage, 'video id', default=video_id) - config = self._download_json(config_url, video_id, - note='Downloading player config') + r'(?:<nflcs:avplayer[^>]+data-content[Ii]d\s*=\s*|content[Ii]d\s*:\s*)(["\'])(?P<id>.+?)\1', + webpage, 'video id', default=video_id, group='id') + config = self._download_json(config_url, video_id, 'Downloading player config') url_template = NFLIE.prepend_host( host, '{contentURLTemplate:}'.format(**config)) video_data = self._download_json( diff --git a/youtube_dl/extractor/nhl.py b/youtube_dl/extractor/nhl.py index 279b18386..e98a5ef89 100644 --- a/youtube_dl/extractor/nhl.py +++ b/youtube_dl/extractor/nhl.py @@ -72,7 +72,7 @@ class NHLBaseInfoExtractor(InfoExtractor): class NHLIE(NHLBaseInfoExtractor): IE_NAME = 'nhl.com' - _VALID_URL = r'https?://video(?P<team>\.[^.]*)?\.nhl\.com/videocenter/(?:console)?(?:\?(?:.*?[?&])?)(?:id|hlg)=(?P<id>[-0-9a-zA-Z,]+)' + _VALID_URL = r'https?://video(?P<team>\.[^.]*)?\.nhl\.com/videocenter/(?:console|embed)?(?:\?(?:.*?[?&])?)(?:id|hlg|playlist)=(?P<id>[-0-9a-zA-Z,]+)' _TESTS = [{ 'url': 'http://video.canucks.nhl.com/videocenter/console?catid=6?id=453614', @@ -136,6 +136,9 @@ class NHLIE(NHLBaseInfoExtractor): 'params': { 'skip_download': True, # Requires rtmpdump } + }, { + 'url': 'http://video.nhl.com/videocenter/embed?playlist=836127', + 'only_matching': True, }] def _real_extract(self, url): @@ -146,9 +149,9 @@ class NHLIE(NHLBaseInfoExtractor): class NHLNewsIE(NHLBaseInfoExtractor): IE_NAME = 'nhl.com:news' IE_DESC = 'NHL news' - _VALID_URL = r'https?://(?:www\.)?nhl\.com/ice/news\.html?(?:\?(?:.*?[?&])?)id=(?P<id>[-0-9a-zA-Z]+)' + _VALID_URL = r'https?://(?:.+?\.)?nhl\.com/(?:ice|club)/news\.html?(?:\?(?:.*?[?&])?)id=(?P<id>[-0-9a-zA-Z]+)' - _TEST = { + _TESTS = [{ 'url': 'http://www.nhl.com/ice/news.htm?id=750727', 'md5': '4b3d1262e177687a3009937bd9ec0be8', 'info_dict': { @@ -159,13 +162,26 @@ class NHLNewsIE(NHLBaseInfoExtractor): 'duration': 37, 'upload_date': '20150128', }, - } + }, { + # iframe embed + 'url': 'http://sabres.nhl.com/club/news.htm?id=780189', + 'md5': '9f663d1c006c90ac9fb82777d4294e12', + 'info_dict': { + 'id': '836127', + 'ext': 'mp4', + 'title': 'Morning Skate: OTT vs. BUF (9/23/15)', + 'description': "Brian Duff chats with Tyler Ennis prior to Buffalo's first preseason home game.", + 'duration': 93, + 'upload_date': '20150923', + }, + }] def _real_extract(self, url): news_id = self._match_id(url) webpage = self._download_webpage(url, news_id) video_id = self._search_regex( - [r'pVid(\d+)', r"nlid\s*:\s*'(\d+)'"], + [r'pVid(\d+)', r"nlid\s*:\s*'(\d+)'", + r'<iframe[^>]+src=["\']https?://video.*?\.nhl\.com/videocenter/embed\?.*\bplaylist=(\d+)'], webpage, 'video id') return self._real_extract_video(video_id) diff --git a/youtube_dl/extractor/ninegag.py b/youtube_dl/extractor/ninegag.py index 7f842b5c2..a06d38afd 100644 --- a/youtube_dl/extractor/ninegag.py +++ b/youtube_dl/extractor/ninegag.py @@ -1,7 +1,6 @@ from __future__ import unicode_literals import re -import json from .common import InfoExtractor from ..utils import str_to_int @@ -9,61 +8,93 @@ from ..utils import str_to_int class NineGagIE(InfoExtractor): IE_NAME = '9gag' - _VALID_URL = r'''(?x)^https?://(?:www\.)?9gag\.tv/ - (?: - v/(?P<numid>[0-9]+)| - p/(?P<id>[a-zA-Z0-9]+)/(?P<display_id>[^?#/]+) - ) - ''' + _VALID_URL = r'https?://(?:www\.)?9gag(?:\.com/tv|\.tv)/(?:p|embed)/(?P<id>[a-zA-Z0-9]+)(?:/(?P<display_id>[^?#/]+))?' _TESTS = [{ - "url": "http://9gag.tv/v/1912", - "info_dict": { - "id": "1912", - "ext": "mp4", - "description": "This 3-minute video will make you smile and then make you feel untalented and insignificant. Anyway, you should share this awesomeness. (Thanks, Dino!)", - "title": "\"People Are Awesome 2013\" Is Absolutely Awesome", + 'url': 'http://9gag.com/tv/p/Kk2X5/people-are-awesome-2013-is-absolutely-awesome', + 'info_dict': { + 'id': 'Kk2X5', + 'ext': 'mp4', + 'description': 'This 3-minute video will make you smile and then make you feel untalented and insignificant. Anyway, you should share this awesomeness. (Thanks, Dino!)', + 'title': '\"People Are Awesome 2013\" Is Absolutely Awesome', 'uploader_id': 'UCdEH6EjDKwtTe-sO2f0_1XA', 'uploader': 'CompilationChannel', 'upload_date': '20131110', - "view_count": int, - "thumbnail": "re:^https?://", + 'view_count': int, }, - 'add_ie': ['Youtube'] + 'add_ie': ['Youtube'], }, { - 'url': 'http://9gag.tv/p/KklwM/alternate-banned-opening-scene-of-gravity?ref=fsidebar', + 'url': 'http://9gag.com/tv/p/aKolP3', 'info_dict': { - 'id': 'KklwM', + 'id': 'aKolP3', 'ext': 'mp4', - 'display_id': 'alternate-banned-opening-scene-of-gravity', - "description": "While Gravity was a pretty awesome movie already, YouTuber Krishna Shenoi came up with a way to improve upon it, introducing a much better solution to Sandra Bullock's seemingly endless tumble in space. The ending is priceless.", - 'title': "Banned Opening Scene Of \"Gravity\" That Changes The Whole Movie", - 'uploader': 'Krishna Shenoi', - 'upload_date': '20140401', - 'uploader_id': 'krishnashenoi93', + 'title': 'This Guy Travelled 11 countries In 44 days Just To Make This Amazing Video', + 'description': "I just saw more in 1 minute than I've seen in 1 year. This guy's video is epic!!", + 'uploader_id': 'rickmereki', + 'uploader': 'Rick Mereki', + 'upload_date': '20110803', + 'view_count': int, }, + 'add_ie': ['Vimeo'], + }, { + 'url': 'http://9gag.com/tv/p/KklwM', + 'only_matching': True, + }, { + 'url': 'http://9gag.tv/p/Kk2X5', + 'only_matching': True, + }, { + 'url': 'http://9gag.com/tv/embed/a5Dmvl', + 'only_matching': True, }] + _EXTERNAL_VIDEO_PROVIDER = { + '1': { + 'url': '%s', + 'ie_key': 'Youtube', + }, + '2': { + 'url': 'http://player.vimeo.com/video/%s', + 'ie_key': 'Vimeo', + }, + '3': { + 'url': 'http://instagram.com/p/%s', + 'ie_key': 'Instagram', + }, + '4': { + 'url': 'http://vine.co/v/%s', + 'ie_key': 'Vine', + }, + } + def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) - video_id = mobj.group('numid') or mobj.group('id') + video_id = mobj.group('id') display_id = mobj.group('display_id') or video_id webpage = self._download_webpage(url, display_id) - post_view = json.loads(self._html_search_regex( - r'var postView = new app\.PostView\({\s*post:\s*({.+?}),\s*posts:\s*prefetchedCurrentPost', webpage, 'post view')) + post_view = self._parse_json( + self._search_regex( + r'var\s+postView\s*=\s*new\s+app\.PostView\({\s*post:\s*({.+?})\s*,\s*posts:\s*prefetchedCurrentPost', + webpage, 'post view'), + display_id) - youtube_id = post_view['videoExternalId'] + ie_key = None + source_url = post_view.get('sourceUrl') + if not source_url: + external_video_id = post_view['videoExternalId'] + external_video_provider = post_view['videoExternalProvider'] + source_url = self._EXTERNAL_VIDEO_PROVIDER[external_video_provider]['url'] % external_video_id + ie_key = self._EXTERNAL_VIDEO_PROVIDER[external_video_provider]['ie_key'] title = post_view['title'] - description = post_view['description'] - view_count = str_to_int(post_view['externalView']) + description = post_view.get('description') + view_count = str_to_int(post_view.get('externalView')) thumbnail = post_view.get('thumbnail_700w') or post_view.get('ogImageUrl') or post_view.get('thumbnail_300w') return { '_type': 'url_transparent', - 'url': youtube_id, - 'ie_key': 'Youtube', + 'url': source_url, + 'ie_key': ie_key, 'id': video_id, 'display_id': display_id, 'title': title, diff --git a/youtube_dl/extractor/novamov.py b/youtube_dl/extractor/novamov.py index 04d779890..6b15fc2e5 100644 --- a/youtube_dl/extractor/novamov.py +++ b/youtube_dl/extractor/novamov.py @@ -4,10 +4,14 @@ import re from .common import InfoExtractor from ..compat import ( + compat_urllib_request, compat_urlparse, ) from ..utils import ( ExtractorError, + NO_DEFAULT, + encode_dict, + urlencode_postdata, ) @@ -38,19 +42,40 @@ class NovaMovIE(InfoExtractor): } def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - video_id = mobj.group('id') + video_id = self._match_id(url) - page = self._download_webpage( - 'http://%s/video/%s' % (self._HOST, video_id), video_id, 'Downloading video page') + url = 'http://%s/video/%s' % (self._HOST, video_id) - if re.search(self._FILE_DELETED_REGEX, page) is not None: - raise ExtractorError('Video %s does not exist' % video_id, expected=True) + webpage = self._download_webpage( + url, video_id, 'Downloading video page') - filekey = self._search_regex(self._FILEKEY_REGEX, page, 'filekey') + if re.search(self._FILE_DELETED_REGEX, webpage) is not None: + raise ExtractorError('Video %s does not exist' % video_id, expected=True) - title = self._html_search_regex(self._TITLE_REGEX, page, 'title', fatal=False) - description = self._html_search_regex(self._DESCRIPTION_REGEX, page, 'description', default='', fatal=False) + def extract_filekey(default=NO_DEFAULT): + return self._search_regex( + self._FILEKEY_REGEX, webpage, 'filekey', default=default) + + filekey = extract_filekey(default=None) + + if not filekey: + fields = self._hidden_inputs(webpage) + post_url = self._search_regex( + r'<form[^>]+action=(["\'])(?P<url>.+?)\1', webpage, + 'post url', default=url, group='url') + if not post_url.startswith('http'): + post_url = compat_urlparse.urljoin(url, post_url) + request = compat_urllib_request.Request( + post_url, urlencode_postdata(encode_dict(fields))) + request.add_header('Content-Type', 'application/x-www-form-urlencoded') + request.add_header('Referer', post_url) + webpage = self._download_webpage( + request, video_id, 'Downloading continue to the video page') + + filekey = extract_filekey() + + title = self._html_search_regex(self._TITLE_REGEX, webpage, 'title', fatal=False) + description = self._html_search_regex(self._DESCRIPTION_REGEX, webpage, 'description', default='', fatal=False) api_response = self._download_webpage( 'http://%s/api/player.api.php?key=%s&file=%s' % (self._HOST, filekey, video_id), video_id, diff --git a/youtube_dl/extractor/nowness.py b/youtube_dl/extractor/nowness.py index 6b2f3f55a..b97f62fdb 100644 --- a/youtube_dl/extractor/nowness.py +++ b/youtube_dl/extractor/nowness.py @@ -1,64 +1,134 @@ # encoding: utf-8 from __future__ import unicode_literals -import re - from .brightcove import BrightcoveIE from .common import InfoExtractor from ..utils import ExtractorError +from ..compat import ( + compat_str, + compat_urllib_request, +) + + +class NownessBaseIE(InfoExtractor): + def _extract_url_result(self, post): + if post['type'] == 'video': + for media in post['media']: + if media['type'] == 'video': + video_id = media['content'] + source = media['source'] + if source == 'brightcove': + player_code = self._download_webpage( + 'http://www.nowness.com/iframe?id=%s' % video_id, video_id, + note='Downloading player JavaScript', + errnote='Unable to download player JavaScript') + bc_url = BrightcoveIE._extract_brightcove_url(player_code) + if bc_url is None: + raise ExtractorError('Could not find player definition') + return self.url_result(bc_url, 'Brightcove') + elif source == 'vimeo': + return self.url_result('http://vimeo.com/%s' % video_id, 'Vimeo') + elif source == 'youtube': + return self.url_result(video_id, 'Youtube') + elif source == 'cinematique': + # youtube-dl currently doesn't support cinematique + # return self.url_result('http://cinematique.com/embed/%s' % video_id, 'Cinematique') + pass + def _api_request(self, url, request_path): + display_id = self._match_id(url) + request = compat_urllib_request.Request( + 'http://api.nowness.com/api/' + request_path % display_id, + headers={ + 'X-Nowness-Language': 'zh-cn' if 'cn.nowness.com' in url else 'en-us', + }) + return display_id, self._download_json(request, display_id) -class NownessIE(InfoExtractor): - _VALID_URL = r'https?://(?:(?:www|cn)\.)?nowness\.com/[^?#]*?/(?P<id>[0-9]+)/(?P<slug>[^/]+?)(?:$|[?#])' - _TESTS = [ - { - 'url': 'http://www.nowness.com/day/2013/6/27/3131/candor--the-art-of-gesticulation', - 'md5': '068bc0202558c2e391924cb8cc470676', - 'info_dict': { - 'id': '2520295746001', - 'ext': 'mp4', - 'title': 'Candor: The Art of Gesticulation', - 'description': 'Candor: The Art of Gesticulation', - 'thumbnail': 're:^https?://.*\.jpg', - 'uploader': 'Nowness', - } +class NownessIE(NownessBaseIE): + IE_NAME = 'nowness' + _VALID_URL = r'https?://(?:(?:www|cn)\.)?nowness\.com/(?:story|(?:series|category)/[^/]+)/(?P<id>[^/]+?)(?:$|[?#])' + _TESTS = [{ + 'url': 'https://www.nowness.com/story/candor-the-art-of-gesticulation', + 'md5': '068bc0202558c2e391924cb8cc470676', + 'info_dict': { + 'id': '2520295746001', + 'ext': 'mp4', + 'title': 'Candor: The Art of Gesticulation', + 'description': 'Candor: The Art of Gesticulation', + 'thumbnail': 're:^https?://.*\.jpg', + 'uploader': 'Nowness', }, - { - 'url': 'http://cn.nowness.com/day/2014/8/7/4069/kasper-bj-rke-ft-jaakko-eino-kalevi--tnr', - 'md5': 'e79cf125e387216f86b2e0a5b5c63aa3', - 'info_dict': { - 'id': '3716354522001', - 'ext': 'mp4', - 'title': 'Kasper Bjørke ft. Jaakko Eino Kalevi: TNR', - 'description': 'Kasper Bjørke ft. Jaakko Eino Kalevi: TNR', - 'thumbnail': 're:^https?://.*\.jpg', - 'uploader': 'Nowness', - } + }, { + 'url': 'https://cn.nowness.com/story/kasper-bjorke-ft-jaakko-eino-kalevi-tnr', + 'md5': 'e79cf125e387216f86b2e0a5b5c63aa3', + 'info_dict': { + 'id': '3716354522001', + 'ext': 'mp4', + 'title': 'Kasper Bjørke ft. Jaakko Eino Kalevi: TNR', + 'description': 'Kasper Bjørke ft. Jaakko Eino Kalevi: TNR', + 'thumbnail': 're:^https?://.*\.jpg', + 'uploader': 'Nowness', }, - ] + }, { + # vimeo + 'url': 'https://www.nowness.com/series/nowness-picks/jean-luc-godard-supercut', + 'md5': '9a5a6a8edf806407e411296ab6bc2a49', + 'info_dict': { + 'id': '130020913', + 'ext': 'mp4', + 'title': 'Bleu, Blanc, Rouge - A Godard Supercut', + 'description': 'md5:f0ea5f1857dffca02dbd37875d742cec', + 'thumbnail': 're:^https?://.*\.jpg', + 'upload_date': '20150607', + 'uploader': 'Cinema Sem Lei', + 'uploader_id': 'cinemasemlei', + }, + }] def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - video_id = mobj.group('slug') + _, post = self._api_request(url, 'post/getBySlug/%s') + return self._extract_url_result(post) - webpage = self._download_webpage(url, video_id) - player_url = self._search_regex( - r'"([^"]+/content/issue-[0-9.]+.js)"', webpage, 'player URL') - real_id = self._search_regex( - r'\sdata-videoId="([0-9]+)"', webpage, 'internal video ID') - player_code = self._download_webpage( - player_url, video_id, - note='Downloading player JavaScript', - errnote='Player download failed') - player_code = player_code.replace("'+d+'", real_id) +class NownessPlaylistIE(NownessBaseIE): + IE_NAME = 'nowness:playlist' + _VALID_URL = r'https?://(?:(?:www|cn)\.)?nowness\.com/playlist/(?P<id>\d+)' + _TEST = { + 'url': 'https://www.nowness.com/playlist/3286/i-guess-thats-why-they-call-it-the-blues', + 'info_dict': { + 'id': '3286', + }, + 'playlist_mincount': 8, + } - bc_url = BrightcoveIE._extract_brightcove_url(player_code) - if bc_url is None: - raise ExtractorError('Could not find player definition') - return { - '_type': 'url', - 'url': bc_url, - 'ie_key': 'Brightcove', - } + def _real_extract(self, url): + playlist_id, playlist = self._api_request(url, 'post?PlaylistId=%s') + entries = [self._extract_url_result(item) for item in playlist['items']] + return self.playlist_result(entries, playlist_id) + + +class NownessSeriesIE(NownessBaseIE): + IE_NAME = 'nowness:series' + _VALID_URL = r'https?://(?:(?:www|cn)\.)?nowness\.com/series/(?P<id>[^/]+?)(?:$|[?#])' + _TEST = { + 'url': 'https://www.nowness.com/series/60-seconds', + 'info_dict': { + 'id': '60', + 'title': '60 Seconds', + 'description': 'One-minute wisdom in a new NOWNESS series', + }, + 'playlist_mincount': 4, + } + + def _real_extract(self, url): + display_id, series = self._api_request(url, 'series/getBySlug/%s') + entries = [self._extract_url_result(post) for post in series['posts']] + series_title = None + series_description = None + translations = series.get('translations', []) + if translations: + series_title = translations[0].get('title') or translations[0]['seoTitle'] + series_description = translations[0].get('seoDescription') + return self.playlist_result( + entries, compat_str(series['id']), series_title, series_description) diff --git a/youtube_dl/extractor/nowtv.py b/youtube_dl/extractor/nowtv.py index c8257719f..b0bdffc4e 100644 --- a/youtube_dl/extractor/nowtv.py +++ b/youtube_dl/extractor/nowtv.py @@ -167,8 +167,8 @@ class NowTVIE(InfoExtractor): 'app': app, 'play_path': 'mp4:%s' % play_path, 'ext': 'flv', - 'page_url': url, - 'player_url': 'http://rtl-now.rtl.de/includes/nc_player.swf', + 'page_url': 'http://rtlnow.rtl.de', + 'player_url': 'http://cdn.static-fra.de/now/vodplayer.swf', 'tbr': int_or_none(item.get('bitrate')), }) self._sort_formats(formats) diff --git a/youtube_dl/extractor/nowvideo.py b/youtube_dl/extractor/nowvideo.py index 17baa9679..57ee3d366 100644 --- a/youtube_dl/extractor/nowvideo.py +++ b/youtube_dl/extractor/nowvideo.py @@ -7,9 +7,9 @@ class NowVideoIE(NovaMovIE): IE_NAME = 'nowvideo' IE_DESC = 'NowVideo' - _VALID_URL = NovaMovIE._VALID_URL_TEMPLATE % {'host': 'nowvideo\.(?:ch|ec|sx|eu|at|ag|co|li)'} + _VALID_URL = NovaMovIE._VALID_URL_TEMPLATE % {'host': 'nowvideo\.(?:to|ch|ec|sx|eu|at|ag|co|li)'} - _HOST = 'www.nowvideo.ch' + _HOST = 'www.nowvideo.to' _FILE_DELETED_REGEX = r'>This file no longer exists on our servers.<' _FILEKEY_REGEX = r'var fkzd="([^"]+)";' diff --git a/youtube_dl/extractor/nrk.py b/youtube_dl/extractor/nrk.py index d066a96db..8ac38a174 100644 --- a/youtube_dl/extractor/nrk.py +++ b/youtube_dl/extractor/nrk.py @@ -4,6 +4,7 @@ from __future__ import unicode_literals import re from .common import InfoExtractor +from ..compat import compat_urlparse from ..utils import ( ExtractorError, float_or_none, @@ -49,7 +50,7 @@ class NRKIE(InfoExtractor): if data['usageRights']['isGeoBlocked']: raise ExtractorError( - 'NRK har ikke rettig-heter til å vise dette programmet utenfor Norge', + 'NRK har ikke rettigheter til å vise dette programmet utenfor Norge', expected=True) video_url = data['mediaUrl'] + '?hdcore=3.5.0&plugin=aasp-3.5.0.151.81' @@ -196,20 +197,6 @@ class NRKTVIE(InfoExtractor): } ] - def _debug_print(self, txt): - if self._downloader.params.get('verbose', False): - self.to_screen('[debug] %s' % txt) - - def _get_subtitles(self, subtitlesurl, video_id, baseurl): - url = "%s%s" % (baseurl, subtitlesurl) - self._debug_print('%s: Subtitle url: %s' % (video_id, url)) - captions = self._download_xml( - url, video_id, 'Downloading subtitles') - lang = captions.get('lang', 'no') - return {lang: [ - {'ext': 'ttml', 'url': url}, - ]} - def _extract_f4m(self, manifest_url, video_id): return self._extract_f4m_formats( manifest_url + '?hdcore=3.1.1&plugin=aasp-3.1.1.69.124', video_id, f4m_id='hds') @@ -218,7 +205,7 @@ class NRKTVIE(InfoExtractor): mobj = re.match(self._VALID_URL, url) video_id = mobj.group('id') part_id = mobj.group('part_id') - baseurl = mobj.group('baseurl') + base_url = mobj.group('baseurl') webpage = self._download_webpage(url, video_id) @@ -278,11 +265,14 @@ class NRKTVIE(InfoExtractor): self._sort_formats(formats) subtitles_url = self._html_search_regex( - r'data-subtitlesurl[ ]*=[ ]*"([^"]+)"', - webpage, 'subtitle URL', default=None) - subtitles = None + r'data-subtitlesurl\s*=\s*(["\'])(?P<url>.+?)\1', + webpage, 'subtitle URL', default=None, group='url') + subtitles = {} if subtitles_url: - subtitles = self.extract_subtitles(subtitles_url, video_id, baseurl) + subtitles['no'] = [{ + 'ext': 'ttml', + 'url': compat_urlparse.urljoin(base_url, subtitles_url), + }] return { 'id': video_id, diff --git a/youtube_dl/extractor/odnoklassniki.py b/youtube_dl/extractor/odnoklassniki.py index 66520c2c5..184c7a323 100644 --- a/youtube_dl/extractor/odnoklassniki.py +++ b/youtube_dl/extractor/odnoklassniki.py @@ -4,6 +4,7 @@ from __future__ import unicode_literals from .common import InfoExtractor from ..compat import compat_urllib_parse_unquote from ..utils import ( + ExtractorError, unified_strdate, int_or_none, qualities, @@ -12,7 +13,7 @@ from ..utils import ( class OdnoklassnikiIE(InfoExtractor): - _VALID_URL = r'https?://(?:www\.)?(?:odnoklassniki|ok)\.ru/(?:video|web-api/video/moviePlayer)/(?P<id>[\d-]+)' + _VALID_URL = r'https?://(?:www\.)?(?:odnoklassniki|ok)\.ru/(?:video(?:embed)?|web-api/video/moviePlayer)/(?P<id>[\d-]+)' _TESTS = [{ # metadata in JSON 'url': 'http://ok.ru/video/20079905452', @@ -28,6 +29,7 @@ class OdnoklassnikiIE(InfoExtractor): 'like_count': int, 'age_limit': 0, }, + 'skip': 'Video has been blocked', }, { # metadataUrl 'url': 'http://ok.ru/video/63567059965189-0', @@ -64,6 +66,9 @@ class OdnoklassnikiIE(InfoExtractor): }, { 'url': 'http://www.ok.ru/video/20648036891', 'only_matching': True, + }, { + 'url': 'http://www.ok.ru/videoembed/20648036891', + 'only_matching': True, }] def _real_extract(self, url): @@ -72,6 +77,12 @@ class OdnoklassnikiIE(InfoExtractor): webpage = self._download_webpage( 'http://ok.ru/video/%s' % video_id, video_id) + error = self._search_regex( + r'[^>]+class="vp_video_stub_txt"[^>]*>([^<]+)<', + webpage, 'error', default=None) + if error: + raise ExtractorError(error, expected=True) + player = self._parse_json( unescapeHTML(self._search_regex( r'data-options=(?P<quote>["\'])(?P<player>{.+?%s.+?})(?P=quote)' % video_id, diff --git a/youtube_dl/extractor/openfilm.py b/youtube_dl/extractor/openfilm.py deleted file mode 100644 index d2ceedd01..000000000 --- a/youtube_dl/extractor/openfilm.py +++ /dev/null @@ -1,70 +0,0 @@ -from __future__ import unicode_literals - -import json - -from .common import InfoExtractor -from ..compat import compat_urllib_parse_unquote_plus -from ..utils import ( - parse_iso8601, - parse_age_limit, - int_or_none, -) - - -class OpenFilmIE(InfoExtractor): - _VALID_URL = r'http://(?:www\.)openfilm\.com/videos/(?P<id>.+)' - _TEST = { - 'url': 'http://www.openfilm.com/videos/human-resources-remastered', - 'md5': '42bcd88c2f3ec13b65edf0f8ad1cac37', - 'info_dict': { - 'id': '32736', - 'display_id': 'human-resources-remastered', - 'ext': 'mp4', - 'title': 'Human Resources (Remastered)', - 'description': 'Social Engineering in the 20th Century.', - 'thumbnail': 're:^https?://.*\.jpg$', - 'duration': 7164, - 'timestamp': 1334756988, - 'upload_date': '20120418', - 'uploader_id': '41117', - 'view_count': int, - 'age_limit': 0, - }, - } - - def _real_extract(self, url): - display_id = self._match_id(url) - - webpage = self._download_webpage(url, display_id) - - player = compat_urllib_parse_unquote_plus( - self._og_search_video_url(webpage)) - - video = json.loads(self._search_regex( - r'\bp=({.+?})(?:&|$)', player, 'video JSON')) - - video_url = '%s1.mp4' % video['location'] - video_id = video.get('video_id') - display_id = video.get('alias') or display_id - title = video.get('title') - description = video.get('description') - thumbnail = video.get('main_thumb') - duration = int_or_none(video.get('duration')) - timestamp = parse_iso8601(video.get('dt_published'), ' ') - uploader_id = video.get('user_id') - view_count = int_or_none(video.get('views_count')) - age_limit = parse_age_limit(video.get('age_limit')) - - return { - 'id': video_id, - 'display_id': display_id, - 'url': video_url, - 'title': title, - 'description': description, - 'thumbnail': thumbnail, - 'duration': duration, - 'timestamp': timestamp, - 'uploader_id': uploader_id, - 'view_count': view_count, - 'age_limit': age_limit, - } diff --git a/youtube_dl/extractor/pbs.py b/youtube_dl/extractor/pbs.py index 683c81de3..8fb9b1849 100644 --- a/youtube_dl/extractor/pbs.py +++ b/youtube_dl/extractor/pbs.py @@ -8,6 +8,7 @@ from ..utils import ( ExtractorError, determine_ext, int_or_none, + strip_jsonp, unified_strdate, US_RATINGS, ) @@ -108,12 +109,12 @@ class PBSIE(InfoExtractor): { 'url': 'http://www.pbs.org/wgbh/americanexperience/films/death/player/', 'info_dict': { - 'id': '2280706814', + 'id': '2276541483', 'display_id': 'player', 'ext': 'mp4', - 'title': 'American Experience - Death and the Civil War', + 'title': 'American Experience - Death and the Civil War, Chapter 1', 'description': 'American Experience, TV’s most-watched history series, brings to life the compelling stories from our past that inform our understanding of the world today.', - 'duration': 6705, + 'duration': 682, 'thumbnail': 're:^https?://.*\.jpg$', }, 'params': { @@ -134,8 +135,49 @@ class PBSIE(InfoExtractor): 'params': { 'skip_download': True, # requires ffmpeg }, + 'skip': 'Expired', + }, + { + # Video embedded in iframe containing angle brackets as attribute's value (e.g. + # "<iframe style='position: absolute;<br />\ntop: 0; left: 0;' ...", see + # https://github.com/rg3/youtube-dl/issues/7059) + 'url': 'http://www.pbs.org/food/features/a-chefs-life-season-3-episode-5-prickly-business/', + 'info_dict': { + 'id': '2365546844', + 'display_id': 'a-chefs-life-season-3-episode-5-prickly-business', + 'ext': 'mp4', + 'title': "A Chef's Life - Season 3, Ep. 5: Prickly Business", + 'description': 'md5:61db2ddf27c9912f09c241014b118ed1', + 'duration': 1480, + 'thumbnail': 're:^https?://.*\.jpg$', + }, + 'params': { + 'skip_download': True, # requires ffmpeg + }, + }, + { + # Frontline video embedded via flp2012.js + 'url': 'http://www.pbs.org/wgbh/pages/frontline/the-atomic-artists', + 'info_dict': { + 'id': '2070868960', + 'display_id': 'the-atomic-artists', + 'ext': 'mp4', + 'title': 'FRONTLINE - The Atomic Artists', + 'description': 'md5:f5bfbefadf421e8bb8647602011caf8e', + 'duration': 723, + 'thumbnail': 're:^https?://.*\.jpg$', + }, + 'params': { + 'skip_download': True, # requires ffmpeg + }, } ] + _ERRORS = { + 101: 'We\'re sorry, but this video is not yet available.', + 403: 'We\'re sorry, but this video is not available in your region due to right restrictions.', + 404: 'We are experiencing technical difficulties that are preventing us from playing the video at this time. Please check back again soon.', + 410: 'This video has expired and is no longer available for online streaming.', + } def _extract_webpage(self, url): mobj = re.match(self._VALID_URL, url) @@ -166,9 +208,30 @@ class PBSIE(InfoExtractor): if media_id: return media_id, presumptive_id, upload_date - url = self._search_regex( - r'<iframe\s+[^>]*\s+src=["\']([^\'"]+partnerplayer[^\'"]+)["\']', - webpage, 'player URL') + # Fronline video embedded via flp + video_id = self._search_regex( + r'videoid\s*:\s*"([\d+a-z]{7,})"', webpage, 'videoid', default=None) + if video_id: + # pkg_id calculation is reverse engineered from + # http://www.pbs.org/wgbh/pages/frontline/js/flp2012.js + prg_id = self._search_regex( + r'videoid\s*:\s*"([\d+a-z]{7,})"', webpage, 'videoid')[7:] + if 'q' in prg_id: + prg_id = prg_id.split('q')[1] + prg_id = int(prg_id, 16) + getdir = self._download_json( + 'http://www.pbs.org/wgbh/pages/frontline/.json/getdir/getdir%d.json' % prg_id, + presumptive_id, 'Downloading getdir JSON', + transform_source=strip_jsonp) + return getdir['mid'], presumptive_id, upload_date + + for iframe in re.findall(r'(?s)<iframe(.+?)></iframe>', webpage): + url = self._search_regex( + r'src=(["\'])(?P<url>.+?partnerplayer.+?)\1', iframe, + 'player URL', default=None, group='url') + if url: + break + mobj = re.match(self._VALID_URL, url) player_id = mobj.group('player_id') @@ -213,13 +276,11 @@ class PBSIE(InfoExtractor): 'Downloading %s video url info' % encoding_name) if redirect_info['status'] == 'error': - if redirect_info['http_code'] == 403: - message = ( - 'The video is not available in your region due to ' - 'right restrictions') - else: - message = redirect_info['message'] - raise ExtractorError(message, expected=True) + raise ExtractorError( + '%s said: %s' % ( + self.IE_NAME, + self._ERRORS.get(redirect_info['http_code'], redirect_info['message'])), + expected=True) format_url = redirect_info.get('url') if not format_url: diff --git a/youtube_dl/extractor/periscope.py b/youtube_dl/extractor/periscope.py index 8ad936758..887c8020d 100644 --- a/youtube_dl/extractor/periscope.py +++ b/youtube_dl/extractor/periscope.py @@ -12,7 +12,8 @@ from ..utils import parse_iso8601 class PeriscopeIE(InfoExtractor): IE_DESC = 'Periscope' _VALID_URL = r'https?://(?:www\.)?periscope\.tv/w/(?P<id>[^/?#]+)' - _TEST = { + # Alive example URLs can be found here http://onperiscope.com/ + _TESTS = [{ 'url': 'https://www.periscope.tv/w/aJUQnjY3MjA3ODF8NTYxMDIyMDl2zCg2pECBgwTqRpQuQD352EMPTKQjT4uqlM3cgWFA-g==', 'md5': '65b57957972e503fcbbaeed8f4fa04ca', 'info_dict': { @@ -25,11 +26,15 @@ class PeriscopeIE(InfoExtractor): 'uploader_id': '1465763', }, 'skip': 'Expires in 24 hours', - } + }, { + 'url': 'https://www.periscope.tv/w/1ZkKzPbMVggJv', + 'only_matching': True, + }] - def _call_api(self, method, token): + def _call_api(self, method, value): + attribute = 'token' if len(value) > 13 else 'broadcast_id' return self._download_json( - 'https://api.periscope.tv/api/v2/%s?token=%s' % (method, token), token) + 'https://api.periscope.tv/api/v2/%s?%s=%s' % (method, attribute, value), value) def _real_extract(self, url): token = self._match_id(url) diff --git a/youtube_dl/extractor/playwire.py b/youtube_dl/extractor/playwire.py index bdc71017b..6d138ef25 100644 --- a/youtube_dl/extractor/playwire.py +++ b/youtube_dl/extractor/playwire.py @@ -19,7 +19,7 @@ class PlaywireIE(InfoExtractor): 'id': '3353705', 'ext': 'mp4', 'title': 'S04_RM_UCL_Rus', - 'thumbnail': 're:^http://.*\.png$', + 'thumbnail': 're:^https?://.*\.png$', 'duration': 145.94, }, }, { diff --git a/youtube_dl/extractor/pornhub.py b/youtube_dl/extractor/pornhub.py index 7b0cdc41a..a656ad85a 100644 --- a/youtube_dl/extractor/pornhub.py +++ b/youtube_dl/extractor/pornhub.py @@ -20,7 +20,7 @@ from ..aes import ( class PornHubIE(InfoExtractor): - _VALID_URL = r'https?://(?:www\.)?pornhub\.com/(?:view_video\.php\?viewkey=|embed/)(?P<id>[0-9a-z]+)' + _VALID_URL = r'https?://(?:[a-z]+\.)?pornhub\.com/(?:view_video\.php\?viewkey=|embed/)(?P<id>[0-9a-z]+)' _TESTS = [{ 'url': 'http://www.pornhub.com/view_video.php?viewkey=648719015', 'md5': '882f488fa1f0026f023f33576004a2ed', @@ -34,6 +34,9 @@ class PornHubIE(InfoExtractor): }, { 'url': 'http://www.pornhub.com/view_video.php?viewkey=ph557bbb6676d2d', 'only_matching': True, + }, { + 'url': 'http://fr.pornhub.com/view_video.php?viewkey=ph55ca2f9760862', + 'only_matching': True, }] @classmethod diff --git a/youtube_dl/extractor/prosiebensat1.py b/youtube_dl/extractor/prosiebensat1.py index effcf1db3..baa54a3af 100644 --- a/youtube_dl/extractor/prosiebensat1.py +++ b/youtube_dl/extractor/prosiebensat1.py @@ -20,7 +20,7 @@ from ..utils import ( class ProSiebenSat1IE(InfoExtractor): IE_NAME = 'prosiebensat1' IE_DESC = 'ProSiebenSat.1 Digital' - _VALID_URL = r'https?://(?:www\.)?(?:(?:prosieben|prosiebenmaxx|sixx|sat1|kabeleins|the-voice-of-germany)\.(?:de|at)|ran\.de|fem\.com)/(?P<id>.+)' + _VALID_URL = r'https?://(?:www\.)?(?:(?:prosieben|prosiebenmaxx|sixx|sat1|kabeleins|the-voice-of-germany)\.(?:de|at|ch)|ran\.de|fem\.com)/(?P<id>.+)' _TESTS = [ { diff --git a/youtube_dl/extractor/qqmusic.py b/youtube_dl/extractor/qqmusic.py index 1654a641f..c98539f6a 100644 --- a/youtube_dl/extractor/qqmusic.py +++ b/youtube_dl/extractor/qqmusic.py @@ -25,7 +25,7 @@ class QQMusicIE(InfoExtractor): 'id': '004295Et37taLD', 'ext': 'mp3', 'title': '可惜没如果', - 'upload_date': '20141227', + 'release_date': '20141227', 'creator': '林俊杰', 'description': 'md5:d327722d0361576fde558f1ac68a7065', 'thumbnail': 're:^https?://.*\.jpg$', @@ -38,11 +38,26 @@ class QQMusicIE(InfoExtractor): 'id': '004MsGEo3DdNxV', 'ext': 'mp3', 'title': '如果', - 'upload_date': '20050626', + 'release_date': '20050626', 'creator': '李季美', 'description': 'md5:46857d5ed62bc4ba84607a805dccf437', 'thumbnail': 're:^https?://.*\.jpg$', } + }, { + 'note': 'lyrics not in .lrc format', + 'url': 'http://y.qq.com/#type=song&mid=001JyApY11tIp6', + 'info_dict': { + 'id': '001JyApY11tIp6', + 'ext': 'mp3', + 'title': 'Shadows Over Transylvania', + 'release_date': '19970225', + 'creator': 'Dark Funeral', + 'description': 'md5:ed14d5bd7ecec19609108052c25b2c11', + 'thumbnail': 're:^https?://.*\.jpg$', + }, + 'params': { + 'skip_download': True, + }, }] _FORMATS = { @@ -112,15 +127,27 @@ class QQMusicIE(InfoExtractor): self._check_formats(formats, mid) self._sort_formats(formats) - return { + actual_lrc_lyrics = ''.join( + line + '\n' for line in re.findall( + r'(?m)^(\[[0-9]{2}:[0-9]{2}(?:\.[0-9]{2,})?\][^\n]*|\[[^\]]*\])', lrc_content)) + + info_dict = { 'id': mid, 'formats': formats, 'title': song_name, - 'upload_date': publish_time, + 'release_date': publish_time, 'creator': singer, 'description': lrc_content, - 'thumbnail': thumbnail_url, + 'thumbnail': thumbnail_url } + if actual_lrc_lyrics: + info_dict['subtitles'] = { + 'origin': [{ + 'ext': 'lrc', + 'data': actual_lrc_lyrics, + }] + } + return info_dict class QQPlaylistBaseIE(InfoExtractor): diff --git a/youtube_dl/extractor/rai.py b/youtube_dl/extractor/rai.py index 1631faf29..7ff1d06c4 100644 --- a/youtube_dl/extractor/rai.py +++ b/youtube_dl/extractor/rai.py @@ -5,6 +5,7 @@ import re from .common import InfoExtractor from ..compat import ( compat_urllib_parse, + compat_urlparse, ) from ..utils import ( parse_duration, @@ -72,6 +73,18 @@ class RaiIE(InfoExtractor): 'description': 'Primo appuntamento con "Il candidato" con Filippo Timi, alias Piero Zucca presidente!', 'uploader': 'RaiTre', } + }, + { + 'url': 'http://www.report.rai.it/dl/Report/puntata/ContentItem-0c7a664b-d0f4-4b2c-8835-3f82e46f433e.html', + 'md5': '037104d2c14132887e5e4cf114569214', + 'info_dict': { + 'id': '0c7a664b-d0f4-4b2c-8835-3f82e46f433e', + 'ext': 'flv', + 'title': 'Il pacco', + 'description': 'md5:4b1afae1364115ce5d78ed83cd2e5b3a', + 'uploader': 'RaiTre', + 'upload_date': '20141221', + }, } ] @@ -90,11 +103,14 @@ class RaiIE(InfoExtractor): relinker_url = self._extract_relinker_url(webpage) if not relinker_url: - iframe_path = self._search_regex( - r'<iframe[^>]+src="/?(dl/[^"]+\?iframe\b[^"]*)"', + iframe_url = self._search_regex( + [r'<iframe[^>]+src="([^"]*/dl/[^"]+\?iframe\b[^"]*)"', + r'drawMediaRaiTV\(["\'](.+?)["\']'], webpage, 'iframe') + if not iframe_url.startswith('http'): + iframe_url = compat_urlparse.urljoin(url, iframe_url) webpage = self._download_webpage( - '%s/%s' % (host, iframe_path), video_id) + iframe_url, video_id) relinker_url = self._extract_relinker_url(webpage) relinker = self._download_json( diff --git a/youtube_dl/extractor/rtbf.py b/youtube_dl/extractor/rtbf.py index e4215d546..e42b319a3 100644 --- a/youtube_dl/extractor/rtbf.py +++ b/youtube_dl/extractor/rtbf.py @@ -9,8 +9,8 @@ from ..utils import ( class RTBFIE(InfoExtractor): - _VALID_URL = r'https?://www.rtbf.be/video/[^\?]+\?id=(?P<id>\d+)' - _TEST = { + _VALID_URL = r'https?://(?:www\.)?rtbf\.be/(?:video/[^?]+\?.*\bid=|ouftivi/(?:[^/]+/)*[^?]+\?.*\bvideoId=)(?P<id>\d+)' + _TESTS = [{ 'url': 'https://www.rtbf.be/video/detail_les-diables-au-coeur-episode-2?id=1921274', 'md5': '799f334ddf2c0a582ba80c44655be570', 'info_dict': { @@ -19,7 +19,14 @@ class RTBFIE(InfoExtractor): 'title': 'Les Diables au coeur (épisode 2)', 'duration': 3099, } - } + }, { + # geo restricted + 'url': 'http://www.rtbf.be/ouftivi/heros/detail_scooby-doo-mysteres-associes?id=1097&videoId=2057442', + 'only_matching': True, + }, { + 'url': 'http://www.rtbf.be/ouftivi/niouzz?videoId=2055858', + 'only_matching': True, + }] _QUALITIES = [ ('mobile', 'mobile'), @@ -36,7 +43,7 @@ class RTBFIE(InfoExtractor): data = self._parse_json( unescapeHTML(self._search_regex( - r'data-video="([^"]+)"', webpage, 'data video')), + r'data-media="([^"]+)"', webpage, 'data video')), video_id) if data.get('provider').lower() == 'youtube': diff --git a/youtube_dl/extractor/rte.py b/youtube_dl/extractor/rte.py index 04158b993..d9cfbf180 100644 --- a/youtube_dl/extractor/rte.py +++ b/youtube_dl/extractor/rte.py @@ -9,16 +9,16 @@ from ..utils import ( class RteIE(InfoExtractor): - _VALID_URL = r'http?://(?:www\.)?rte\.ie/player/[^/]{2,3}/show/(?P<id>[0-9]+)/' + _VALID_URL = r'https?://(?:www\.)?rte\.ie/player/[^/]{2,3}/show/[^/]+/(?P<id>[0-9]+)' _TEST = { - 'url': 'http://www.rte.ie/player/de/show/10363114/', + 'url': 'http://www.rte.ie/player/ie/show/iwitness-862/10478715/', 'info_dict': { - 'id': '10363114', + 'id': '10478715', 'ext': 'mp4', - 'title': 'One News', + 'title': 'Watch iWitness online', 'thumbnail': 're:^https?://.*\.jpg$', - 'description': 'The One O\'Clock News followed by Weather.', - 'duration': 436.844, + 'description': 'iWitness : The spirit of Ireland, one voice and one minute at a time.', + 'duration': 60.046, }, 'params': { 'skip_download': 'f4m fails with --test atm' diff --git a/youtube_dl/extractor/ruutu.py b/youtube_dl/extractor/ruutu.py index c67ad25ce..e417bf661 100644 --- a/youtube_dl/extractor/ruutu.py +++ b/youtube_dl/extractor/ruutu.py @@ -57,16 +57,21 @@ class RuutuIE(InfoExtractor): extract_formats(child) elif child.tag.endswith('File'): video_url = child.text - if not video_url or video_url in processed_urls or 'NOT_USED' in video_url: + if (not video_url or video_url in processed_urls or + any(p in video_url for p in ('NOT_USED', 'NOT-USED'))): return processed_urls.append(video_url) ext = determine_ext(video_url) if ext == 'm3u8': - formats.extend(self._extract_m3u8_formats( - video_url, video_id, 'mp4', m3u8_id='hls')) + m3u8_formats = self._extract_m3u8_formats( + video_url, video_id, 'mp4', m3u8_id='hls', fatal=False) + if m3u8_formats: + formats.extend(m3u8_formats) elif ext == 'f4m': - formats.extend(self._extract_f4m_formats( - video_url, video_id, f4m_id='hds')) + f4m_formats = self._extract_f4m_formats( + video_url, video_id, f4m_id='hds', fatal=False) + if f4m_formats: + formats.extend(f4m_formats) else: proto = compat_urllib_parse_urlparse(video_url).scheme if not child.tag.startswith('HTTP') and proto != 'rtmp': @@ -74,7 +79,7 @@ class RuutuIE(InfoExtractor): preference = -1 if proto == 'rtmp' else 1 label = child.get('label') tbr = int_or_none(child.get('bitrate')) - width, height = [int_or_none(x) for x in child.get('resolution', '').split('x')] + width, height = [int_or_none(x) for x in child.get('resolution', 'x').split('x')[:2]] formats.append({ 'format_id': '%s-%s' % (proto, label if label else tbr), 'url': video_url, diff --git a/youtube_dl/extractor/senateisvp.py b/youtube_dl/extractor/senateisvp.py index 9c53704ea..474ebb49b 100644 --- a/youtube_dl/extractor/senateisvp.py +++ b/youtube_dl/extractor/senateisvp.py @@ -121,9 +121,9 @@ class SenateISVPIE(InfoExtractor): 'url': compat_urlparse.urljoin(domain, filename) + '?v=3.1.0&fp=&r=&g=', }] else: - hdcore_sign = '?hdcore=3.1.0' + hdcore_sign = 'hdcore=3.1.0' url_params = (domain, video_id, stream_num) - f4m_url = '%s/z/%s_1@%s/manifest.f4m' % url_params + hdcore_sign + f4m_url = '%s/z/%s_1@%s/manifest.f4m?' % url_params + hdcore_sign m3u8_url = '%s/i/%s_1@%s/master.m3u8' % url_params for entry in self._extract_f4m_formats(f4m_url, video_id, f4m_id='f4m'): # URLs without the extra param induce an 404 error diff --git a/youtube_dl/extractor/shahid.py b/youtube_dl/extractor/shahid.py index 6e9903d5e..f76fb12c0 100644 --- a/youtube_dl/extractor/shahid.py +++ b/youtube_dl/extractor/shahid.py @@ -16,7 +16,7 @@ class ShahidIE(InfoExtractor): 'url': 'https://shahid.mbc.net/ar/episode/90574/%D8%A7%D9%84%D9%85%D9%84%D9%83-%D8%B9%D8%A8%D8%AF%D8%A7%D9%84%D9%84%D9%87-%D8%A7%D9%84%D8%A5%D9%86%D8%B3%D8%A7%D9%86-%D8%A7%D9%84%D9%85%D9%88%D8%B3%D9%85-1-%D9%83%D9%84%D9%8A%D8%A8-3.html', 'info_dict': { 'id': '90574', - 'ext': 'm3u8', + 'ext': 'mp4', 'title': 'الملك عبدالله الإنسان الموسم 1 كليب 3', 'description': 'الفيلم الوثائقي - الملك عبد الله الإنسان', 'duration': 2972, @@ -81,7 +81,7 @@ class ShahidIE(InfoExtractor): compat_urllib_parse.urlencode({ 'apiKey': 'sh@hid0nlin3', 'hash': 'b2wMCTHpSmyxGqQjJFOycRmLSex+BpTK/ooxy6vHaqs=', - }).encode('utf-8')), + })), video_id, 'Downloading video JSON') video = video[api_vars['playerType']] diff --git a/youtube_dl/extractor/soundcloud.py b/youtube_dl/extractor/soundcloud.py index ed5dcc0d3..2b60d354a 100644 --- a/youtube_dl/extractor/soundcloud.py +++ b/youtube_dl/extractor/soundcloud.py @@ -113,7 +113,7 @@ class SoundcloudIE(InfoExtractor): }, ] - _CLIENT_ID = 'b45b1aa10f1ac2941910a7f0d10f8e28' + _CLIENT_ID = '02gUJC0hH2ct1EGOcYXQIzRFU91c72Ea' _IPHONE_CLIENT_ID = '376f225bf427445fc4bfb6b99b72e0bf' def report_resolve(self, video_id): diff --git a/youtube_dl/extractor/spiegeltv.py b/youtube_dl/extractor/spiegeltv.py index 27f4033c5..034bd47ff 100644 --- a/youtube_dl/extractor/spiegeltv.py +++ b/youtube_dl/extractor/spiegeltv.py @@ -77,17 +77,21 @@ class SpiegeltvIE(InfoExtractor): 'rtmp_live': True, }) elif determine_ext(endpoint) == 'm3u8': - m3u8_formats = self._extract_m3u8_formats( - endpoint.replace('[video]', play_path), - video_id, 'm4v', - preference=1, # Prefer hls since it allows to workaround georestriction - m3u8_id='hls', fatal=False) - if m3u8_formats is not False: - formats.extend(m3u8_formats) + formats.append({ + 'url': endpoint.replace('[video]', play_path), + 'ext': 'm4v', + 'format_id': 'hls', # Prefer hls since it allows to workaround georestriction + 'protocol': 'm3u8', + 'preference': 1, + 'http_headers': { + 'Accept-Encoding': 'deflate', # gzip causes trouble on the server side + }, + }) else: formats.append({ 'url': endpoint, }) + self._check_formats(formats, video_id) thumbnails = [] for image in media_json['images']: diff --git a/youtube_dl/extractor/stitcher.py b/youtube_dl/extractor/stitcher.py new file mode 100644 index 000000000..d5c852f52 --- /dev/null +++ b/youtube_dl/extractor/stitcher.py @@ -0,0 +1,81 @@ +from __future__ import unicode_literals + +import re + +from .common import InfoExtractor +from ..utils import ( + determine_ext, + int_or_none, + js_to_json, + unescapeHTML, +) + + +class StitcherIE(InfoExtractor): + _VALID_URL = r'https?://(?:www\.)?stitcher\.com/podcast/(?:[^/]+/)+e/(?:(?P<display_id>[^/#?&]+?)-)?(?P<id>\d+)(?:[/#?&]|$)' + _TESTS = [{ + 'url': 'http://www.stitcher.com/podcast/the-talking-machines/e/40789481?autoplay=true', + 'md5': '391dd4e021e6edeb7b8e68fbf2e9e940', + 'info_dict': { + 'id': '40789481', + 'ext': 'mp3', + 'title': 'Machine Learning Mastery and Cancer Clusters', + 'description': 'md5:55163197a44e915a14a1ac3a1de0f2d3', + 'duration': 1604, + 'thumbnail': 're:^https?://.*\.jpg', + }, + }, { + 'url': 'http://www.stitcher.com/podcast/panoply/vulture-tv/e/the-rare-hourlong-comedy-plus-40846275?autoplay=true', + 'info_dict': { + 'id': '40846275', + 'display_id': 'the-rare-hourlong-comedy-plus', + 'ext': 'mp3', + 'title': "The CW's 'Crazy Ex-Girlfriend'", + 'description': 'md5:04f1e2f98eb3f5cbb094cea0f9e19b17', + 'duration': 2235, + 'thumbnail': 're:^https?://.*\.jpg', + }, + 'params': { + 'skip_download': True, + }, + }, { + # escaped title + 'url': 'http://www.stitcher.com/podcast/marketplace-on-stitcher/e/40910226?autoplay=true', + 'only_matching': True, + }, { + 'url': 'http://www.stitcher.com/podcast/panoply/getting-in/e/episode-2a-how-many-extracurriculars-should-i-have-40876278?autoplay=true', + 'only_matching': True, + }] + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + audio_id = mobj.group('id') + display_id = mobj.group('display_id') or audio_id + + webpage = self._download_webpage(url, display_id) + + episode = self._parse_json( + js_to_json(self._search_regex( + r'(?s)var\s+stitcher\s*=\s*({.+?});\n', webpage, 'episode config')), + display_id)['config']['episode'] + + title = unescapeHTML(episode['title']) + formats = [{ + 'url': episode[episode_key], + 'ext': determine_ext(episode[episode_key]) or 'mp3', + 'vcodec': 'none', + } for episode_key in ('episodeURL',) if episode.get(episode_key)] + description = self._search_regex( + r'Episode Info:\s*</span>([^<]+)<', webpage, 'description', fatal=False) + duration = int_or_none(episode.get('duration')) + thumbnail = episode.get('episodeImage') + + return { + 'id': audio_id, + 'display_id': display_id, + 'title': title, + 'description': description, + 'duration': duration, + 'thumbnail': thumbnail, + 'formats': formats, + } diff --git a/youtube_dl/extractor/tapely.py b/youtube_dl/extractor/tapely.py index f1f43d0a7..744f9db38 100644 --- a/youtube_dl/extractor/tapely.py +++ b/youtube_dl/extractor/tapely.py @@ -16,7 +16,7 @@ from ..utils import ( class TapelyIE(InfoExtractor): - _VALID_URL = r'https?://(?:www\.)?tape\.ly/(?P<id>[A-Za-z0-9\-_]+)(?:/(?P<songnr>\d+))?' + _VALID_URL = r'https?://(?:www\.)?(?:tape\.ly|tapely\.com)/(?P<id>[A-Za-z0-9\-_]+)(?:/(?P<songnr>\d+))?' _API_URL = 'http://tape.ly/showtape?id={0:}' _S3_SONG_URL = 'http://mytape.s3.amazonaws.com/{0:}' _SOUNDCLOUD_SONG_URL = 'http://api.soundcloud.com{0:}' @@ -42,6 +42,10 @@ class TapelyIE(InfoExtractor): 'ext': 'm4a', }, }, + { + 'url': 'https://tapely.com/my-grief-as-told-by-water', + 'only_matching': True, + }, ] def _real_extract(self, url): diff --git a/youtube_dl/extractor/telecinco.py b/youtube_dl/extractor/telecinco.py index ae94f055c..2c8e9b941 100644 --- a/youtube_dl/extractor/telecinco.py +++ b/youtube_dl/extractor/telecinco.py @@ -1,24 +1,51 @@ # coding: utf-8 from __future__ import unicode_literals -from .mitele import MiTeleIE +import json +from .common import InfoExtractor +from ..compat import ( + compat_urllib_parse, + compat_urllib_parse_unquote, + compat_urlparse, +) +from ..utils import ( + get_element_by_attribute, + parse_duration, + strip_jsonp, +) -class TelecincoIE(MiTeleIE): - IE_NAME = 'telecinco.es' - _VALID_URL = r'https?://www\.telecinco\.es/(?:[^/]+/)+(?P<id>.+?)\.html' + +class TelecincoIE(InfoExtractor): + IE_DESC = 'telecinco.es, cuatro.com and mediaset.es' + _VALID_URL = r'https?://www\.(?:telecinco\.es|cuatro\.com|mediaset\.es)/(?:[^/]+/)+(?P<id>.+?)\.html' _TESTS = [{ 'url': 'http://www.telecinco.es/robinfood/temporada-01/t01xp14/Bacalao-cocochas-pil-pil_0_1876350223.html', + 'md5': '5cbef3ad5ef17bf0d21570332d140729', 'info_dict': { 'id': 'MDSVID20141015_0058', 'ext': 'mp4', 'title': 'Con Martín Berasategui, hacer un bacalao al ...', 'duration': 662, }, - 'params': { - # m3u8 download - 'skip_download': True, + }, { + 'url': 'http://www.cuatro.com/deportes/futbol/barcelona/Leo_Messi-Champions-Roma_2_2052780128.html', + 'md5': '0a5b9f3cc8b074f50a0578f823a12694', + 'info_dict': { + 'id': 'MDSVID20150916_0128', + 'ext': 'mp4', + 'title': '¿Quién es este ex futbolista con el que hablan ...', + 'duration': 79, + }, + }, { + 'url': 'http://www.mediaset.es/12meses/campanas/doylacara/conlatratanohaytrato/Ayudame-dar-cara-trata-trato_2_1986630220.html', + 'md5': 'ad1bfaaba922dd4a295724b05b68f86a', + 'info_dict': { + 'id': 'MDSVID20150513_0220', + 'ext': 'mp4', + 'title': '#DOYLACARA. Con la trata no hay trato', + 'duration': 50, }, }, { 'url': 'http://www.telecinco.es/informativos/nacional/Pablo_Iglesias-Informativos_Telecinco-entrevista-Pedro_Piqueras_2_1945155182.html', @@ -27,3 +54,41 @@ class TelecincoIE(MiTeleIE): 'url': 'http://www.telecinco.es/espanasinirmaslejos/Espana-gran-destino-turistico_2_1240605043.html', 'only_matching': True, }] + + def _real_extract(self, url): + episode = self._match_id(url) + webpage = self._download_webpage(url, episode) + embed_data_json = self._search_regex( + r'(?s)MSV\.embedData\[.*?\]\s*=\s*({.*?});', webpage, 'embed data', + ).replace('\'', '"') + embed_data = json.loads(embed_data_json) + + domain = embed_data['mediaUrl'] + if not domain.startswith('http'): + # only happens in telecinco.es videos + domain = 'http://' + domain + info_url = compat_urlparse.urljoin( + domain, + compat_urllib_parse_unquote(embed_data['flashvars']['host']) + ) + info_el = self._download_xml(info_url, episode).find('./video/info') + + video_link = info_el.find('videoUrl/link').text + token_query = compat_urllib_parse.urlencode({'id': video_link}) + token_info = self._download_json( + embed_data['flashvars']['ov_tk'] + '?' + token_query, + episode, + transform_source=strip_jsonp + ) + formats = self._extract_m3u8_formats( + token_info['tokenizedUrl'], episode, ext='mp4', entry_protocol='m3u8_native') + + return { + 'id': embed_data['videoId'], + 'display_id': episode, + 'title': info_el.find('title').text, + 'formats': formats, + 'description': get_element_by_attribute('class', 'text', webpage), + 'thumbnail': info_el.find('thumb').text, + 'duration': parse_duration(info_el.find('duration').text), + } diff --git a/youtube_dl/extractor/tudou.py b/youtube_dl/extractor/tudou.py index 84fe71aef..5f7ac4b35 100644 --- a/youtube_dl/extractor/tudou.py +++ b/youtube_dl/extractor/tudou.py @@ -2,14 +2,12 @@ from __future__ import unicode_literals -import re -import json - from .common import InfoExtractor +from ..compat import compat_str class TudouIE(InfoExtractor): - _VALID_URL = r'https?://(?:www\.)?tudou\.com/(?:listplay|programs(?:/view)?|albumplay)/.*?/(?P<id>[^/?#]+?)(?:\.html)?/?(?:$|[?#])' + _VALID_URL = r'https?://(?:www\.)?tudou\.com/(?:listplay|programs(?:/view)?|albumplay)/([^/]+/)*(?P<id>[^/?#]+?)(?:\.html)?/?(?:$|[?#])' _TESTS = [{ 'url': 'http://www.tudou.com/listplay/zzdE77v6Mmo/2xN2duXMxmw.html', 'md5': '140a49ed444bd22f93330985d8475fcb', @@ -27,41 +25,41 @@ class TudouIE(InfoExtractor): 'title': 'La Sylphide-Bolshoi-Ekaterina Krysanova & Vyacheslav Lopatin 2012', 'thumbnail': 're:^https?://.*\.jpg$', } + }, { + 'url': 'http://www.tudou.com/albumplay/cJAHGih4yYg.html', + 'only_matching': True, }] _PLAYER_URL = 'http://js.tudouui.com/bin/lingtong/PortalPlayer_177.swf' - def _url_for_id(self, id, quality=None): - info_url = "http://v2.tudou.com/f?id=" + str(id) + def _url_for_id(self, video_id, quality=None): + info_url = 'http://v2.tudou.com/f?id=' + compat_str(video_id) if quality: info_url += '&hd' + quality - webpage = self._download_webpage(info_url, id, "Opening the info webpage") - final_url = self._html_search_regex('>(.+?)</f>', webpage, 'video url') + xml_data = self._download_xml(info_url, video_id, "Opening the info XML page") + final_url = xml_data.text return final_url def _real_extract(self, url): video_id = self._match_id(url) webpage = self._download_webpage(url, video_id) - m = re.search(r'vcode:\s*[\'"](.+?)[\'"]', webpage) - if m and m.group(1): - return { - '_type': 'url', - 'url': 'youku:' + m.group(1), - 'ie_key': 'Youku' - } + youku_vcode = self._search_regex( + r'vcode\s*:\s*[\'"]([^\'"]*)[\'"]', webpage, 'youku vcode', default=None) + if youku_vcode: + return self.url_result('youku:' + youku_vcode, ie='Youku') title = self._search_regex( - r",kw:\s*['\"](.+?)[\"']", webpage, 'title') + r',kw\s*:\s*[\'"]([^\'"]+)[\'"]', webpage, 'title') thumbnail_url = self._search_regex( - r",pic:\s*[\"'](.+?)[\"']", webpage, 'thumbnail URL', fatal=False) + r',pic\s*:\s*[\'"]([^\'"]+)[\'"]', webpage, 'thumbnail URL', fatal=False) player_url = self._search_regex( - r"playerUrl\s*:\s*['\"](.+?\.swf)[\"']", + r'playerUrl\s*:\s*[\'"]([^\'"]+\.swf)[\'"]', webpage, 'player URL', default=self._PLAYER_URL) - segs_json = self._search_regex(r'segs: \'(.*)\'', webpage, 'segments') - segments = json.loads(segs_json) + segments = self._parse_json(self._search_regex( + r'segs: \'([^\']+)\'', webpage, 'segments'), video_id) # It looks like the keys are the arguments that have to be passed as # the hd field in the request url, we pick the higher # Also, filter non-number qualities (see issue #3643). diff --git a/youtube_dl/extractor/tumblr.py b/youtube_dl/extractor/tumblr.py index 3d3b635e4..4f844706d 100644 --- a/youtube_dl/extractor/tumblr.py +++ b/youtube_dl/extractor/tumblr.py @@ -4,6 +4,7 @@ from __future__ import unicode_literals import re from .common import InfoExtractor +from ..utils import int_or_none class TumblrIE(InfoExtractor): @@ -29,6 +30,19 @@ class TumblrIE(InfoExtractor): 'thumbnail': 're:http://.*\.jpg', } }, { + 'url': 'http://hdvideotest.tumblr.com/post/130323439814/test-description-for-my-hd-video', + 'md5': '7ae503065ad150122dc3089f8cf1546c', + 'info_dict': { + 'id': '130323439814', + 'ext': 'mp4', + 'title': 'HD Video Testing \u2014 Test description for my HD video', + 'description': 'md5:97cc3ab5fcd27ee4af6356701541319c', + 'thumbnail': 're:http://.*\.jpg', + }, + 'params': { + 'format': 'hd', + }, + }, { 'url': 'http://naked-yogi.tumblr.com/post/118312946248/naked-smoking-stretching', 'md5': 'de07e5211d60d4f3a2c3df757ea9f6ab', 'info_dict': { @@ -37,6 +51,9 @@ class TumblrIE(InfoExtractor): 'title': 'naked smoking & stretching', 'upload_date': '20150506', 'timestamp': 1430931613, + 'age_limit': 18, + 'uploader_id': '1638622', + 'uploader': 'naked-yogi', }, 'add_ie': ['Vidme'], }, { @@ -66,10 +83,38 @@ class TumblrIE(InfoExtractor): if iframe_url is None: return self.url_result(urlh.geturl(), 'Generic') - iframe = self._download_webpage(iframe_url, video_id, - 'Downloading iframe page') - video_url = self._search_regex(r'<source src="([^"]+)"', - iframe, 'video url') + iframe = self._download_webpage(iframe_url, video_id, 'Downloading iframe page') + + duration = None + sources = [] + + sd_url = self._search_regex( + r'<source[^>]+src=(["\'])(?P<url>.+?)\1', iframe, + 'sd video url', default=None, group='url') + if sd_url: + sources.append((sd_url, 'sd')) + + options = self._parse_json( + self._search_regex( + r'data-crt-options=(["\'])(?P<options>.+?)\1', iframe, + 'hd video url', default='', group='options'), + video_id, fatal=False) + if options: + duration = int_or_none(options.get('duration')) + hd_url = options.get('hdUrl') + if hd_url: + sources.append((hd_url, 'hd')) + + formats = [{ + 'url': video_url, + 'ext': 'mp4', + 'format_id': format_id, + 'height': int_or_none(self._search_regex( + r'/(\d{3,4})$', video_url, 'height', default=None)), + 'quality': quality, + } for quality, (video_url, format_id) in enumerate(sources)] + + self._sort_formats(formats) # The only place where you can get a title, it's not complete, # but searching in other places doesn't work for all videos @@ -79,9 +124,9 @@ class TumblrIE(InfoExtractor): return { 'id': video_id, - 'url': video_url, - 'ext': 'mp4', 'title': video_title, 'description': self._og_search_description(webpage, default=None), 'thumbnail': self._og_search_thumbnail(webpage, default=None), + 'duration': duration, + 'formats': formats, } diff --git a/youtube_dl/extractor/tutv.py b/youtube_dl/extractor/tutv.py index fad720b68..822372ea1 100644 --- a/youtube_dl/extractor/tutv.py +++ b/youtube_dl/extractor/tutv.py @@ -10,10 +10,10 @@ class TutvIE(InfoExtractor): _VALID_URL = r'https?://(?:www\.)?tu\.tv/videos/(?P<id>[^/?]+)' _TEST = { 'url': 'http://tu.tv/videos/robots-futbolistas', - 'md5': '627c7c124ac2a9b5ab6addb94e0e65f7', + 'md5': '0cd9e28ad270488911b0d2a72323395d', 'info_dict': { 'id': '2973058', - 'ext': 'flv', + 'ext': 'mp4', 'title': 'Robots futbolistas', }, } diff --git a/youtube_dl/extractor/twitch.py b/youtube_dl/extractor/twitch.py index 023911c41..3ec08b674 100644 --- a/youtube_dl/extractor/twitch.py +++ b/youtube_dl/extractor/twitch.py @@ -15,6 +15,7 @@ from ..compat import ( compat_urlparse, ) from ..utils import ( + encode_dict, ExtractorError, int_or_none, parse_duration, @@ -27,8 +28,7 @@ class TwitchBaseIE(InfoExtractor): _API_BASE = 'https://api.twitch.tv' _USHER_BASE = 'http://usher.twitch.tv' - _LOGIN_URL = 'https://secure.twitch.tv/login' - _LOGIN_POST_URL = 'https://passport.twitch.tv/authentications/new' + _LOGIN_URL = 'http://www.twitch.tv/login' _NETRC_MACHINE = 'twitch' def _handle_error(self, response): @@ -61,26 +61,28 @@ class TwitchBaseIE(InfoExtractor): if username is None: return - login_page = self._download_webpage( + login_page, handle = self._download_webpage_handle( self._LOGIN_URL, None, 'Downloading login page') login_form = self._hidden_inputs(login_page) login_form.update({ - 'login': username.encode('utf-8'), - 'password': password.encode('utf-8'), + 'username': username, + 'password': password, }) + redirect_url = handle.geturl() + post_url = self._search_regex( r'<form[^>]+action=(["\'])(?P<url>.+?)\1', login_page, - 'post url', default=self._LOGIN_POST_URL, group='url') + 'post url', default=redirect_url, group='url') if not post_url.startswith('http'): - post_url = compat_urlparse.urljoin(self._LOGIN_URL, post_url) + post_url = compat_urlparse.urljoin(redirect_url, post_url) request = compat_urllib_request.Request( - post_url, compat_urllib_parse.urlencode(login_form).encode('utf-8')) - request.add_header('Referer', self._LOGIN_URL) + post_url, compat_urllib_parse.urlencode(encode_dict(login_form)).encode('utf-8')) + request.add_header('Referer', redirect_url) response = self._download_webpage( request, None, 'Logging in as %s' % username) @@ -238,14 +240,24 @@ class TwitchVodIE(TwitchItemBaseIE): def _real_extract(self, url): item_id = self._match_id(url) + info = self._download_info(self._ITEM_SHORTCUT, item_id) access_token = self._download_json( '%s/api/vods/%s/access_token' % (self._API_BASE, item_id), item_id, 'Downloading %s access token' % self._ITEM_TYPE) + formats = self._extract_m3u8_formats( - '%s/vod/%s?nauth=%s&nauthsig=%s&allow_source=true' - % (self._USHER_BASE, item_id, access_token['token'], access_token['sig']), + '%s/vod/%s?%s' % ( + self._USHER_BASE, item_id, + compat_urllib_parse.urlencode({ + 'allow_source': 'true', + 'allow_spectre': 'true', + 'player': 'twitchweb', + 'nauth': access_token['token'], + 'nauthsig': access_token['sig'], + })), item_id, 'mp4') + self._prefer_source(formats) info['formats'] = formats diff --git a/youtube_dl/extractor/twitter.py b/youtube_dl/extractor/twitter.py index 1aaa06305..2bd5946ac 100644 --- a/youtube_dl/extractor/twitter.py +++ b/youtube_dl/extractor/twitter.py @@ -1,3 +1,4 @@ +# coding: utf-8 from __future__ import unicode_literals import re @@ -6,23 +7,53 @@ from .common import InfoExtractor from ..compat import compat_urllib_request from ..utils import ( float_or_none, - unescapeHTML, + xpath_text, + remove_end, + int_or_none, + ExtractorError, ) class TwitterCardIE(InfoExtractor): + IE_NAME = 'twitter:card' _VALID_URL = r'https?://(?:www\.)?twitter\.com/i/cards/tfw/v1/(?P<id>\d+)' - _TEST = { - 'url': 'https://twitter.com/i/cards/tfw/v1/560070183650213889', - 'md5': 'a74f50b310c83170319ba16de6955192', - 'info_dict': { - 'id': '560070183650213889', - 'ext': 'mp4', - 'title': 'TwitterCard', - 'thumbnail': 're:^https?://.*\.jpg$', - 'duration': 30.033, + _TESTS = [ + { + 'url': 'https://twitter.com/i/cards/tfw/v1/560070183650213889', + 'md5': '4fa26a35f9d1bf4b646590ba8e84be19', + 'info_dict': { + 'id': '560070183650213889', + 'ext': 'mp4', + 'title': 'TwitterCard', + 'thumbnail': 're:^https?://.*\.jpg$', + 'duration': 30.033, + } }, - } + { + 'url': 'https://twitter.com/i/cards/tfw/v1/623160978427936768', + 'md5': '7ee2a553b63d1bccba97fbed97d9e1c8', + 'info_dict': { + 'id': '623160978427936768', + 'ext': 'mp4', + 'title': 'TwitterCard', + 'thumbnail': 're:^https?://.*\.jpg', + 'duration': 80.155, + }, + }, + { + 'url': 'https://twitter.com/i/cards/tfw/v1/654001591733886977', + 'md5': 'b6f35e8b08a0bec6c8af77a2f4b3a814', + 'info_dict': { + 'id': 'dq4Oj5quskI', + 'ext': 'mp4', + 'title': 'Ubuntu 11.10 Overview', + 'description': 'Take a quick peek at what\'s new and improved in Ubuntu 11.10.\n\nOnce installed take a look at 10 Things to Do After Installing: http://www.omgubuntu.co.uk/2011/10/10-things-to-do-after-installing-ubuntu-11-10/', + 'upload_date': '20111013', + 'uploader': 'OMG! Ubuntu!', + 'uploader_id': 'omgubuntu', + }, + } + ] def _real_extract(self, url): video_id = self._match_id(url) @@ -40,10 +71,24 @@ class TwitterCardIE(InfoExtractor): request.add_header('User-Agent', user_agent) webpage = self._download_webpage(request, video_id) - config = self._parse_json( - unescapeHTML(self._search_regex( - r'data-player-config="([^"]+)"', webpage, 'data player config')), + youtube_url = self._html_search_regex( + r'<iframe[^>]+src="((?:https?:)?//www.youtube.com/embed/[^"]+)"', + webpage, 'youtube iframe', default=None) + if youtube_url: + return self.url_result(youtube_url, 'Youtube') + + config = self._parse_json(self._html_search_regex( + r'data-player-config="([^"]+)"', webpage, 'data player config'), video_id) + if 'playlist' not in config: + if 'vmapUrl' in config: + vmap_data = self._download_xml(config['vmapUrl'], video_id) + video_url = xpath_text(vmap_data, './/MediaFile').strip() + formats.append({ + 'url': video_url, + }) + break # same video regardless of UA + continue video_url = config['playlist'][0]['source'] @@ -70,3 +115,100 @@ class TwitterCardIE(InfoExtractor): 'duration': duration, 'formats': formats, } + + +class TwitterIE(InfoExtractor): + IE_NAME = 'twitter' + _VALID_URL = r'https?://(?:www\.|m\.|mobile\.)?twitter\.com/(?P<user_id>[^/]+)/status/(?P<id>\d+)' + _TEMPLATE_URL = 'https://twitter.com/%s/status/%s' + + _TESTS = [{ + 'url': 'https://twitter.com/freethenipple/status/643211948184596480', + 'md5': 'db6612ec5d03355953c3ca9250c97e5e', + 'info_dict': { + 'id': '643211948184596480', + 'ext': 'mp4', + 'title': 'FREE THE NIPPLE - FTN supporters on Hollywood Blvd today!', + 'thumbnail': 're:^https?://.*\.jpg', + 'duration': 12.922, + 'description': 'FREE THE NIPPLE on Twitter: "FTN supporters on Hollywood Blvd today! http://t.co/c7jHH749xJ"', + 'uploader': 'FREE THE NIPPLE', + 'uploader_id': 'freethenipple', + }, + }, { + 'url': 'https://twitter.com/giphz/status/657991469417025536/photo/1', + 'md5': 'f36dcd5fb92bf7057f155e7d927eeb42', + 'info_dict': { + 'id': '657991469417025536', + 'ext': 'mp4', + 'title': 'Gifs - tu vai cai tu vai cai tu nao eh capaz disso tu vai cai', + 'description': 'Gifs on Twitter: "tu vai cai tu vai cai tu nao eh capaz disso tu vai cai https://t.co/tM46VHFlO5"', + 'thumbnail': 're:^https?://.*\.png', + 'uploader': 'Gifs', + 'uploader_id': 'giphz', + }, + }, { + 'url': 'https://twitter.com/starwars/status/665052190608723968', + 'md5': '39b7199856dee6cd4432e72c74bc69d4', + 'info_dict': { + 'id': '665052190608723968', + 'ext': 'mp4', + 'title': 'Star Wars - A new beginning is coming December 18. Watch the official 60 second #TV spot for #StarWars: #TheForceAwakens.', + 'description': 'Star Wars on Twitter: "A new beginning is coming December 18. Watch the official 60 second #TV spot for #StarWars: #TheForceAwakens."', + 'uploader_id': 'starwars', + 'uploader': 'Star Wars', + }, + }] + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + user_id = mobj.group('user_id') + twid = mobj.group('id') + + webpage = self._download_webpage(self._TEMPLATE_URL % (user_id, twid), twid) + + username = remove_end(self._og_search_title(webpage), ' on Twitter') + + title = description = self._og_search_description(webpage).strip('').replace('\n', ' ').strip('“”') + + # strip 'https -_t.co_BJYgOjSeGA' junk from filenames + title = re.sub(r'\s+(https?://[^ ]+)', '', title) + + info = { + 'uploader_id': user_id, + 'uploader': username, + 'webpage_url': url, + 'description': '%s on Twitter: "%s"' % (username, description), + 'title': username + ' - ' + title, + } + + card_id = self._search_regex( + r'["\']/i/cards/tfw/v1/(\d+)', webpage, 'twitter card url', default=None) + if card_id: + card_url = 'https://twitter.com/i/cards/tfw/v1/' + card_id + info.update({ + '_type': 'url_transparent', + 'ie_key': 'TwitterCard', + 'url': card_url, + }) + return info + + mobj = re.search(r'''(?x) + <video[^>]+class="animated-gif"[^>]+ + (?:data-height="(?P<height>\d+)")?[^>]+ + (?:data-width="(?P<width>\d+)")?[^>]+ + (?:poster="(?P<poster>[^"]+)")?[^>]*>\s* + <source[^>]+video-src="(?P<url>[^"]+)" + ''', webpage) + + if mobj: + info.update({ + 'id': twid, + 'url': mobj.group('url'), + 'height': int_or_none(mobj.group('height')), + 'width': int_or_none(mobj.group('width')), + 'thumbnail': mobj.group('poster'), + }) + return info + + raise ExtractorError('There\'s not video in this tweet.') diff --git a/youtube_dl/extractor/ustream.py b/youtube_dl/extractor/ustream.py index c39c278ab..73b05ecab 100644 --- a/youtube_dl/extractor/ustream.py +++ b/youtube_dl/extractor/ustream.py @@ -1,17 +1,20 @@ from __future__ import unicode_literals -import json import re from .common import InfoExtractor from ..compat import ( compat_urlparse, ) -from ..utils import ExtractorError +from ..utils import ( + ExtractorError, + int_or_none, + float_or_none, +) class UstreamIE(InfoExtractor): - _VALID_URL = r'https?://www\.ustream\.tv/(?P<type>recorded|embed|embed/recorded)/(?P<videoID>\d+)' + _VALID_URL = r'https?://www\.ustream\.tv/(?P<type>recorded|embed|embed/recorded)/(?P<id>\d+)' IE_NAME = 'ustream' _TESTS = [{ 'url': 'http://www.ustream.tv/recorded/20274954', @@ -19,8 +22,12 @@ class UstreamIE(InfoExtractor): 'info_dict': { 'id': '20274954', 'ext': 'flv', - 'uploader': 'Young Americans for Liberty', 'title': 'Young Americans for Liberty February 7, 2012 2:28 AM', + 'description': 'Young Americans for Liberty February 7, 2012 2:28 AM', + 'timestamp': 1328577035, + 'upload_date': '20120207', + 'uploader': 'yaliberty', + 'uploader_id': '6780869', }, }, { # From http://sportscanada.tv/canadagames/index.php/week2/figure-skating/444 @@ -32,20 +39,21 @@ class UstreamIE(InfoExtractor): 'ext': 'flv', 'title': '-CG11- Canada Games Figure Skating', 'uploader': 'sportscanadatv', - } + }, + 'skip': 'This Pro Broadcaster has chosen to remove this video from the ustream.tv site.', }] def _real_extract(self, url): m = re.match(self._VALID_URL, url) - video_id = m.group('videoID') + video_id = m.group('id') # some sites use this embed format (see: http://github.com/rg3/youtube-dl/issues/2990) if m.group('type') == 'embed/recorded': - video_id = m.group('videoID') + video_id = m.group('id') desktop_url = 'http://www.ustream.tv/recorded/' + video_id return self.url_result(desktop_url, 'Ustream') if m.group('type') == 'embed': - video_id = m.group('videoID') + video_id = m.group('id') webpage = self._download_webpage(url, video_id) desktop_video_id = self._html_search_regex( r'ContentVideoIds=\["([^"]*?)"\]', webpage, 'desktop_video_id') @@ -53,52 +61,50 @@ class UstreamIE(InfoExtractor): return self.url_result(desktop_url, 'Ustream') params = self._download_json( - 'http://cdngw.ustream.tv/rgwjson/Viewer.getVideo/' + json.dumps({ - 'brandId': 1, - 'videoId': int(video_id), - 'autoplay': False, - }), video_id) - - if 'error' in params: - raise ExtractorError(params['error']['message'], expected=True) - - video_url = params['flv'] + 'https://api.ustream.tv/videos/%s.json' % video_id, video_id) - webpage = self._download_webpage(url, video_id) + error = params.get('error') + if error: + raise ExtractorError( + '%s returned error: %s' % (self.IE_NAME, error), expected=True) - self.report_extraction(video_id) + video = params['video'] - video_title = self._html_search_regex(r'data-title="(?P<title>.+)"', - webpage, 'title', default=None) + title = video['title'] + filesize = float_or_none(video.get('file_size')) - if not video_title: - try: - video_title = params['moduleConfig']['meta']['title'] - except KeyError: - pass - - if not video_title: - video_title = 'Ustream video ' + video_id + formats = [{ + 'id': video_id, + 'url': video_url, + 'ext': format_id, + 'filesize': filesize, + } for format_id, video_url in video['media_urls'].items()] + self._sort_formats(formats) - uploader = self._html_search_regex(r'data-content-type="channel".*?>(?P<uploader>.*?)</a>', - webpage, 'uploader', fatal=False, flags=re.DOTALL, default=None) + description = video.get('description') + timestamp = int_or_none(video.get('created_at')) + duration = float_or_none(video.get('length')) + view_count = int_or_none(video.get('views')) - if not uploader: - try: - uploader = params['moduleConfig']['meta']['userName'] - except KeyError: - uploader = None + uploader = video.get('owner', {}).get('username') + uploader_id = video.get('owner', {}).get('id') - thumbnail = self._html_search_regex(r'<link rel="image_src" href="(?P<thumb>.*?)"', - webpage, 'thumbnail', fatal=False) + thumbnails = [{ + 'id': thumbnail_id, + 'url': thumbnail_url, + } for thumbnail_id, thumbnail_url in video.get('thumbnail', {}).items()] return { 'id': video_id, - 'url': video_url, - 'ext': 'flv', - 'title': video_title, + 'title': title, + 'description': description, + 'thumbnails': thumbnails, + 'timestamp': timestamp, + 'duration': duration, + 'view_count': view_count, 'uploader': uploader, - 'thumbnail': thumbnail, + 'uploader_id': uploader_id, + 'formats': formats, } diff --git a/youtube_dl/extractor/vevo.py b/youtube_dl/extractor/vevo.py index c17094f81..4c0de354f 100644 --- a/youtube_dl/extractor/vevo.py +++ b/youtube_dl/extractor/vevo.py @@ -1,10 +1,10 @@ from __future__ import unicode_literals import re -import xml.etree.ElementTree from .common import InfoExtractor from ..compat import ( + compat_etree_fromstring, compat_urllib_request, ) from ..utils import ( @@ -97,7 +97,7 @@ class VevoIE(InfoExtractor): if last_version['version'] == -1: raise ExtractorError('Unable to extract last version of the video') - renditions = xml.etree.ElementTree.fromstring(last_version['data']) + renditions = compat_etree_fromstring(last_version['data']) formats = [] # Already sorted from worst to best quality for rend in renditions.findall('rendition'): @@ -114,7 +114,7 @@ class VevoIE(InfoExtractor): def _formats_from_smil(self, smil_xml): formats = [] - smil_doc = xml.etree.ElementTree.fromstring(smil_xml.encode('utf-8')) + smil_doc = compat_etree_fromstring(smil_xml.encode('utf-8')) els = smil_doc.findall('.//{http://www.w3.org/2001/SMIL20/Language}video') for el in els: src = el.attrib['src'] diff --git a/youtube_dl/extractor/videofyme.py b/youtube_dl/extractor/videofyme.py index 94f9e9be9..cd3f50a63 100644 --- a/youtube_dl/extractor/videofyme.py +++ b/youtube_dl/extractor/videofyme.py @@ -2,8 +2,8 @@ from __future__ import unicode_literals from .common import InfoExtractor from ..utils import ( - find_xpath_attr, int_or_none, + parse_iso8601, ) @@ -18,33 +18,35 @@ class VideofyMeIE(InfoExtractor): 'id': '1100701', 'ext': 'mp4', 'title': 'This is VideofyMe', - 'description': None, + 'description': '', + 'upload_date': '20130326', + 'timestamp': 1364288959, 'uploader': 'VideofyMe', 'uploader_id': 'thisisvideofyme', 'view_count': int, + 'likes': int, + 'comment_count': int, }, - } def _real_extract(self, url): video_id = self._match_id(url) - config = self._download_xml('http://sunshine.videofy.me/?videoId=%s' % video_id, - video_id) - video = config.find('video') - sources = video.find('sources') - url_node = next(node for node in [find_xpath_attr(sources, 'source', 'id', 'HQ %s' % key) - for key in ['on', 'av', 'off']] if node is not None) - video_url = url_node.find('url').text - view_count = int_or_none(self._search_regex( - r'([0-9]+)', video.find('views').text, 'view count', fatal=False)) + + config = self._download_json('http://vf-player-info-loader.herokuapp.com/%s.json' % video_id, video_id)['videoinfo'] + + video = config.get('video') + blog = config.get('blog', {}) return { 'id': video_id, - 'title': video.find('title').text, - 'url': video_url, - 'thumbnail': video.find('thumb').text, - 'description': video.find('description').text, - 'uploader': config.find('blog/name').text, - 'uploader_id': video.find('identifier').text, - 'view_count': view_count, + 'title': video['title'], + 'url': video['sources']['source']['url'], + 'thumbnail': video.get('thumb'), + 'description': video.get('description'), + 'timestamp': parse_iso8601(video.get('date')), + 'uploader': blog.get('name'), + 'uploader_id': blog.get('identifier'), + 'view_count': int_or_none(self._search_regex(r'([0-9]+)', video.get('views'), 'view count', fatal=False)), + 'likes': int_or_none(video.get('likes')), + 'comment_count': int_or_none(video.get('nrOfComments')), } diff --git a/youtube_dl/extractor/videolecturesnet.py b/youtube_dl/extractor/videolecturesnet.py deleted file mode 100644 index ef2da5632..000000000 --- a/youtube_dl/extractor/videolecturesnet.py +++ /dev/null @@ -1,86 +0,0 @@ -from __future__ import unicode_literals - -import re - -from .common import InfoExtractor -from ..utils import ( - find_xpath_attr, - int_or_none, - parse_duration, - unified_strdate, -) - - -class VideoLecturesNetIE(InfoExtractor): - _VALID_URL = r'http://(?:www\.)?videolectures\.net/(?P<id>[^/#?]+)/*(?:[#?].*)?$' - IE_NAME = 'videolectures.net' - - _TEST = { - 'url': 'http://videolectures.net/promogram_igor_mekjavic_eng/', - 'info_dict': { - 'id': 'promogram_igor_mekjavic_eng', - 'ext': 'mp4', - 'title': 'Automatics, robotics and biocybernetics', - 'description': 'md5:815fc1deb6b3a2bff99de2d5325be482', - 'upload_date': '20130627', - 'duration': 565, - 'thumbnail': 're:http://.*\.jpg', - }, - } - - def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - video_id = mobj.group('id') - - smil_url = 'http://videolectures.net/%s/video/1/smil.xml' % video_id - smil = self._download_xml(smil_url, video_id) - - title = find_xpath_attr(smil, './/meta', 'name', 'title').attrib['content'] - description_el = find_xpath_attr(smil, './/meta', 'name', 'abstract') - description = ( - None if description_el is None - else description_el.attrib['content']) - upload_date = unified_strdate( - find_xpath_attr(smil, './/meta', 'name', 'date').attrib['content']) - - switch = smil.find('.//switch') - duration = parse_duration(switch.attrib.get('dur')) - thumbnail_el = find_xpath_attr(switch, './image', 'type', 'thumbnail') - thumbnail = ( - None if thumbnail_el is None else thumbnail_el.attrib.get('src')) - - formats = [] - for v in switch.findall('./video'): - proto = v.attrib.get('proto') - if proto not in ['http', 'rtmp']: - continue - f = { - 'width': int_or_none(v.attrib.get('width')), - 'height': int_or_none(v.attrib.get('height')), - 'filesize': int_or_none(v.attrib.get('size')), - 'tbr': int_or_none(v.attrib.get('systemBitrate')) / 1000.0, - 'ext': v.attrib.get('ext'), - } - src = v.attrib['src'] - if proto == 'http': - if self._is_valid_url(src, video_id): - f['url'] = src - formats.append(f) - elif proto == 'rtmp': - f.update({ - 'url': v.attrib['streamer'], - 'play_path': src, - 'rtmp_real_time': True, - }) - formats.append(f) - self._sort_formats(formats) - - return { - 'id': video_id, - 'title': title, - 'description': description, - 'upload_date': upload_date, - 'duration': duration, - 'thumbnail': thumbnail, - 'formats': formats, - } diff --git a/youtube_dl/extractor/vidme.py b/youtube_dl/extractor/vidme.py index 9a794e609..3d63ed4f0 100644 --- a/youtube_dl/extractor/vidme.py +++ b/youtube_dl/extractor/vidme.py @@ -14,7 +14,7 @@ class VidmeIE(InfoExtractor): _VALID_URL = r'https?://vid\.me/(?:e/)?(?P<id>[\da-zA-Z]+)' _TESTS = [{ 'url': 'https://vid.me/QNB', - 'md5': 'c62f1156138dc3323902188c5b5a8bd6', + 'md5': 'f42d05e7149aeaec5c037b17e5d3dc82', 'info_dict': { 'id': 'QNB', 'ext': 'mp4', @@ -93,6 +93,39 @@ class VidmeIE(InfoExtractor): 'params': { 'skip_download': True, }, + }, { + # nsfw, user-disabled + 'url': 'https://vid.me/dzGJ', + 'only_matching': True, + }, { + # suspended + 'url': 'https://vid.me/Ox3G', + 'only_matching': True, + }, { + # deleted + 'url': 'https://vid.me/KTPm', + 'only_matching': True, + }, { + # no formats in the API response + 'url': 'https://vid.me/e5g', + 'info_dict': { + 'id': 'e5g', + 'ext': 'mp4', + 'title': 'Video upload (e5g)', + 'thumbnail': 're:^https?://.*\.jpg', + 'timestamp': 1401480195, + 'upload_date': '20140530', + 'uploader': None, + 'uploader_id': None, + 'age_limit': 0, + 'duration': 483, + 'view_count': int, + 'like_count': int, + 'comment_count': int, + }, + 'params': { + 'skip_download': True, + }, }] def _real_extract(self, url): @@ -114,12 +147,32 @@ class VidmeIE(InfoExtractor): video = response['video'] + if video.get('state') == 'deleted': + raise ExtractorError( + 'Vidme said: Sorry, this video has been deleted.', + expected=True) + + if video.get('state') in ('user-disabled', 'suspended'): + raise ExtractorError( + 'Vidme said: This video has been suspended either due to a copyright claim, ' + 'or for violating the terms of use.', + expected=True) + formats = [{ 'format_id': f.get('type'), 'url': f['uri'], 'width': int_or_none(f.get('width')), 'height': int_or_none(f.get('height')), + 'preference': 0 if f.get('type', '').endswith('clip') else 1, } for f in video.get('formats', []) if f.get('uri')] + + if not formats and video.get('complete_url'): + formats.append({ + 'url': video.get('complete_url'), + 'width': int_or_none(video.get('width')), + 'height': int_or_none(video.get('height')), + }) + self._sort_formats(formats) title = video['title'] @@ -136,7 +189,7 @@ class VidmeIE(InfoExtractor): return { 'id': video_id, - 'title': title, + 'title': title or 'Video upload (%s)' % video_id, 'description': description, 'thumbnail': thumbnail, 'uploader': uploader, diff --git a/youtube_dl/extractor/vidzi.py b/youtube_dl/extractor/vidzi.py index 08a5a7b8d..2ba9f31df 100644 --- a/youtube_dl/extractor/vidzi.py +++ b/youtube_dl/extractor/vidzi.py @@ -20,8 +20,14 @@ class VidziIE(InfoExtractor): video_id = self._match_id(url) webpage = self._download_webpage(url, video_id) - video_url = self._html_search_regex( - r'{\s*file\s*:\s*"([^"]+)"\s*}', webpage, 'video url') + video_host = self._html_search_regex( + r'id=\'vplayer\'><img src="http://(.*?)/i', webpage, + 'video host') + video_hash = self._html_search_regex( + r'\|([a-z0-9]+)\|hls\|type', webpage, 'video_hash') + ext = self._html_search_regex( + r'\|tracks\|([a-z0-9]+)\|', webpage, 'video ext') + video_url = 'http://' + video_host + '/' + video_hash + '/v.' + ext title = self._html_search_regex( r'(?s)<h2 class="video-title">(.*?)</h2>', webpage, 'title') diff --git a/youtube_dl/extractor/viewster.py b/youtube_dl/extractor/viewster.py index cda02ba24..7cf930d69 100644 --- a/youtube_dl/extractor/viewster.py +++ b/youtube_dl/extractor/viewster.py @@ -3,12 +3,14 @@ from __future__ import unicode_literals from .common import InfoExtractor from ..compat import ( + compat_HTTPError, compat_urllib_request, compat_urllib_parse, compat_urllib_parse_unquote, ) from ..utils import ( determine_ext, + ExtractorError, int_or_none, parse_iso8601, HEADRequest, @@ -16,14 +18,14 @@ from ..utils import ( class ViewsterIE(InfoExtractor): - _VALID_URL = r'http://(?:www\.)?viewster\.com/(?:serie|movie)/(?P<id>\d+-\d+-\d+)' + _VALID_URL = r'https?://(?:www\.)?viewster\.com/(?:serie|movie)/(?P<id>\d+-\d+-\d+)' _TESTS = [{ # movie, Type=Movie 'url': 'http://www.viewster.com/movie/1140-11855-000/the-listening-project/', - 'md5': '14d3cfffe66d57b41ae2d9c873416f01', + 'md5': 'e642d1b27fcf3a4ffa79f194f5adde36', 'info_dict': { 'id': '1140-11855-000', - 'ext': 'flv', + 'ext': 'mp4', 'title': 'The listening Project', 'description': 'md5:bac720244afd1a8ea279864e67baa071', 'timestamp': 1214870400, @@ -33,10 +35,10 @@ class ViewsterIE(InfoExtractor): }, { # series episode, Type=Episode 'url': 'http://www.viewster.com/serie/1284-19427-001/the-world-and-a-wall/', - 'md5': 'd5434c80fcfdb61651cc2199a88d6ba3', + 'md5': '9243079a8531809efe1b089db102c069', 'info_dict': { 'id': '1284-19427-001', - 'ext': 'flv', + 'ext': 'mp4', 'title': 'The World and a Wall', 'description': 'md5:24814cf74d3453fdf5bfef9716d073e3', 'timestamp': 1428192000, @@ -61,6 +63,14 @@ class ViewsterIE(InfoExtractor): 'description': 'md5:e7097a8fc97151e25f085c9eb7a1cdb1', }, 'playlist_mincount': 16, + }, { + # geo restricted series + 'url': 'https://www.viewster.com/serie/1280-18794-002/', + 'only_matching': True, + }, { + # geo restricted video + 'url': 'https://www.viewster.com/serie/1280-18794-002/what-is-extraterritoriality-lawo/', + 'only_matching': True, }] _ACCEPT_HEADER = 'application/json, text/javascript, */*; q=0.01' @@ -74,8 +84,8 @@ class ViewsterIE(InfoExtractor): def _real_extract(self, url): video_id = self._match_id(url) # Get 'api_token' cookie - self._request_webpage(HEADRequest(url), video_id) - cookies = self._get_cookies(url) + self._request_webpage(HEADRequest('http://www.viewster.com/'), video_id) + cookies = self._get_cookies('http://www.viewster.com/') self._AUTH_TOKEN = compat_urllib_parse_unquote(cookies['api_token'].value) info = self._download_json( @@ -85,10 +95,16 @@ class ViewsterIE(InfoExtractor): entry_id = info.get('Id') or info['id'] # unfinished serie has no Type - if info.get('Type') in ['Serie', None]: - episodes = self._download_json( - 'https://public-api.viewster.com/series/%s/episodes' % entry_id, - video_id, 'Downloading series JSON') + if info.get('Type') in ('Serie', None): + try: + episodes = self._download_json( + 'https://public-api.viewster.com/series/%s/episodes' % entry_id, + video_id, 'Downloading series JSON') + except ExtractorError as e: + if isinstance(e.cause, compat_HTTPError) and e.cause.code == 404: + self.raise_geo_restricted() + else: + raise entries = [ self.url_result( 'http://www.viewster.com/movie/%s' % episode['OriginId'], 'Viewster') @@ -98,7 +114,7 @@ class ViewsterIE(InfoExtractor): return self.playlist_result(entries, video_id, title, description) formats = [] - for media_type in ('application/f4m+xml', 'application/x-mpegURL'): + for media_type in ('application/f4m+xml', 'application/x-mpegURL', 'video/mp4'): media = self._download_json( 'https://public-api.viewster.com/movies/%s/video?mediaType=%s' % (entry_id, compat_urllib_parse.quote(media_type)), @@ -115,14 +131,28 @@ class ViewsterIE(InfoExtractor): formats.extend(self._extract_f4m_formats( video_url, video_id, f4m_id='hds')) elif ext == 'm3u8': - formats.extend(self._extract_m3u8_formats( + m3u8_formats = self._extract_m3u8_formats( video_url, video_id, 'mp4', m3u8_id='hls', - fatal=False # m3u8 sometimes fail - )) + fatal=False) # m3u8 sometimes fail + if m3u8_formats: + formats.extend(m3u8_formats) else: - formats.append({ + format_id = media.get('Bitrate') + f = { 'url': video_url, - }) + 'format_id': 'mp4-%s' % format_id, + 'height': int_or_none(media.get('Height')), + 'width': int_or_none(media.get('Width')), + 'preference': 1, + } + if format_id and not f['height']: + f['height'] = int_or_none(self._search_regex( + r'^(\d+)[pP]$', format_id, 'height', default=None)) + formats.append(f) + + if not formats and not info.get('LanguageSets') and not info.get('VODSettings'): + self.raise_geo_restricted() + self._sort_formats(formats) synopsis = info.get('Synopsis', {}) diff --git a/youtube_dl/extractor/viidea.py b/youtube_dl/extractor/viidea.py new file mode 100644 index 000000000..525e303d4 --- /dev/null +++ b/youtube_dl/extractor/viidea.py @@ -0,0 +1,188 @@ +from __future__ import unicode_literals + +import re + +from .common import InfoExtractor +from ..compat import ( + compat_urlparse, + compat_str, +) +from ..utils import ( + parse_duration, + js_to_json, + parse_iso8601, +) + + +class ViideaIE(InfoExtractor): + _VALID_URL = r'''(?x)http://(?:www\.)?(?: + videolectures\.net| + flexilearn\.viidea\.net| + presentations\.ocwconsortium\.org| + video\.travel-zoom\.si| + video\.pomp-forum\.si| + tv\.nil\.si| + video\.hekovnik.com| + video\.szko\.si| + kpk\.viidea\.com| + inside\.viidea\.net| + video\.kiberpipa\.org| + bvvideo\.si| + kongres\.viidea\.net| + edemokracija\.viidea\.com + )(?:/lecture)?/(?P<id>[^/]+)(?:/video/(?P<part>\d+))?/*(?:[#?].*)?$''' + + _TESTS = [{ + 'url': 'http://videolectures.net/promogram_igor_mekjavic_eng/', + 'info_dict': { + 'id': '20171', + 'display_id': 'promogram_igor_mekjavic_eng', + 'ext': 'mp4', + 'title': 'Automatics, robotics and biocybernetics', + 'description': 'md5:815fc1deb6b3a2bff99de2d5325be482', + 'thumbnail': 're:http://.*\.jpg', + 'timestamp': 1372349289, + 'upload_date': '20130627', + 'duration': 565, + }, + }, { + # video with invalid direct format links (HTTP 403) + 'url': 'http://videolectures.net/russir2010_filippova_nlp/', + 'info_dict': { + 'id': '14891', + 'display_id': 'russir2010_filippova_nlp', + 'ext': 'flv', + 'title': 'NLP at Google', + 'description': 'md5:fc7a6d9bf0302d7cc0e53f7ca23747b3', + 'thumbnail': 're:http://.*\.jpg', + 'timestamp': 1284375600, + 'upload_date': '20100913', + 'duration': 5352, + }, + 'params': { + # rtmp download + 'skip_download': True, + }, + }, { + # event playlist + 'url': 'http://videolectures.net/deeplearning2015_montreal/', + 'info_dict': { + 'id': '23181', + 'title': 'Deep Learning Summer School, Montreal 2015', + 'description': 'md5:0533a85e4bd918df52a01f0e1ebe87b7', + 'thumbnail': 're:http://.*\.jpg', + 'timestamp': 1438560000, + }, + 'playlist_count': 30, + }, { + # multi part lecture + 'url': 'http://videolectures.net/mlss09uk_bishop_ibi/', + 'info_dict': { + 'id': '9737', + 'display_id': 'mlss09uk_bishop_ibi', + 'title': 'Introduction To Bayesian Inference', + 'thumbnail': 're:http://.*\.jpg', + 'timestamp': 1251622800, + }, + 'playlist': [{ + 'info_dict': { + 'id': '9737_part1', + 'display_id': 'mlss09uk_bishop_ibi_part1', + 'ext': 'wmv', + 'title': 'Introduction To Bayesian Inference (Part 1)', + 'thumbnail': 're:http://.*\.jpg', + 'duration': 4622, + 'timestamp': 1251622800, + 'upload_date': '20090830', + }, + }, { + 'info_dict': { + 'id': '9737_part2', + 'display_id': 'mlss09uk_bishop_ibi_part2', + 'ext': 'wmv', + 'title': 'Introduction To Bayesian Inference (Part 2)', + 'thumbnail': 're:http://.*\.jpg', + 'duration': 5641, + 'timestamp': 1251622800, + 'upload_date': '20090830', + }, + }], + 'playlist_count': 2, + }] + + def _real_extract(self, url): + lecture_slug, explicit_part_id = re.match(self._VALID_URL, url).groups() + + webpage = self._download_webpage(url, lecture_slug) + + cfg = self._parse_json(self._search_regex( + [r'cfg\s*:\s*({.+?})\s*,\s*[\da-zA-Z_]+\s*:\s*\(?\s*function', + r'cfg\s*:\s*({[^}]+})'], + webpage, 'cfg'), lecture_slug, js_to_json) + + lecture_id = compat_str(cfg['obj_id']) + + base_url = self._proto_relative_url(cfg['livepipe'], 'http:') + + lecture_data = self._download_json( + '%s/site/api/lecture/%s?format=json' % (base_url, lecture_id), + lecture_id)['lecture'][0] + + lecture_info = { + 'id': lecture_id, + 'display_id': lecture_slug, + 'title': lecture_data['title'], + 'timestamp': parse_iso8601(lecture_data.get('time')), + 'description': lecture_data.get('description_wiki'), + 'thumbnail': lecture_data.get('thumb'), + } + + playlist_entries = [] + lecture_type = lecture_data.get('type') + parts = [compat_str(video) for video in cfg.get('videos', [])] + if parts: + multipart = len(parts) > 1 + + def extract_part(part_id): + smil_url = '%s/%s/video/%s/smil.xml' % (base_url, lecture_slug, part_id) + smil = self._download_smil(smil_url, lecture_id) + info = self._parse_smil(smil, smil_url, lecture_id) + info['id'] = lecture_id if not multipart else '%s_part%s' % (lecture_id, part_id) + info['display_id'] = lecture_slug if not multipart else '%s_part%s' % (lecture_slug, part_id) + if multipart: + info['title'] += ' (Part %s)' % part_id + switch = smil.find('.//switch') + if switch is not None: + info['duration'] = parse_duration(switch.attrib.get('dur')) + item_info = lecture_info.copy() + item_info.update(info) + return item_info + + if explicit_part_id or not multipart: + result = extract_part(explicit_part_id or parts[0]) + else: + result = { + '_type': 'multi_video', + 'entries': [extract_part(part) for part in parts], + } + result.update(lecture_info) + + # Immediately return explicitly requested part or non event item + if explicit_part_id or lecture_type != 'evt': + return result + + playlist_entries.append(result) + + # It's probably a playlist + if not parts or lecture_type == 'evt': + playlist_webpage = self._download_webpage( + '%s/site/ajax/drilldown/?id=%s' % (base_url, lecture_id), lecture_id) + entries = [ + self.url_result(compat_urlparse.urljoin(url, video_url), 'Viidea') + for _, video_url in re.findall( + r'<a[^>]+href=(["\'])(.+?)\1[^>]+id=["\']lec=\d+', playlist_webpage)] + playlist_entries.extend(entries) + + playlist = self.playlist_result(playlist_entries, lecture_id) + playlist.update(lecture_info) + return playlist diff --git a/youtube_dl/extractor/vimeo.py b/youtube_dl/extractor/vimeo.py index 50df79ca1..b72341a2b 100644 --- a/youtube_dl/extractor/vimeo.py +++ b/youtube_dl/extractor/vimeo.py @@ -8,11 +8,11 @@ import itertools from .common import InfoExtractor from ..compat import ( compat_HTTPError, - compat_urllib_parse, compat_urllib_request, compat_urlparse, ) from ..utils import ( + encode_dict, ExtractorError, InAdvancePagedList, int_or_none, @@ -39,23 +39,31 @@ class VimeoBaseInfoExtractor(InfoExtractor): return self.report_login() webpage = self._download_webpage(self._LOGIN_URL, None, False) - token = self._extract_xsrft(webpage) - data = urlencode_postdata({ + token, vuid = self._extract_xsrft_and_vuid(webpage) + data = urlencode_postdata(encode_dict({ 'action': 'login', 'email': username, 'password': password, 'service': 'vimeo', 'token': token, - }) + })) login_request = compat_urllib_request.Request(self._LOGIN_URL, data) login_request.add_header('Content-Type', 'application/x-www-form-urlencoded') login_request.add_header('Referer', self._LOGIN_URL) + self._set_vimeo_cookie('vuid', vuid) self._download_webpage(login_request, None, False, 'Wrong login info') - def _extract_xsrft(self, webpage): - return self._search_regex( + def _extract_xsrft_and_vuid(self, webpage): + xsrft = self._search_regex( r'xsrft\s*[=:]\s*(?P<q>["\'])(?P<xsrft>.+?)(?P=q)', webpage, 'login token', group='xsrft') + vuid = self._search_regex( + r'["\']vuid["\']\s*:\s*(["\'])(?P<vuid>.+?)\1', + webpage, 'vuid', group='vuid') + return xsrft, vuid + + def _set_vimeo_cookie(self, name, value): + self._set_cookie('vimeo.com', name, value) class VimeoIE(VimeoBaseInfoExtractor): @@ -80,12 +88,12 @@ class VimeoIE(VimeoBaseInfoExtractor): 'info_dict': { 'id': '56015672', 'ext': 'mp4', - "upload_date": "20121220", - "description": "This is a test case for youtube-dl.\nFor more information, see github.com/rg3/youtube-dl\nTest chars: \u2605 \" ' \u5e78 / \\ \u00e4 \u21ad \U0001d550", - "uploader_id": "user7108434", - "uploader": "Filippo Valsorda", - "title": "youtube-dl test video - \u2605 \" ' \u5e78 / \\ \u00e4 \u21ad \U0001d550", - "duration": 10, + 'title': "youtube-dl test video - \u2605 \" ' \u5e78 / \\ \u00e4 \u21ad \U0001d550", + 'description': 'md5:2d3305bad981a06ff79f027f19865021', + 'upload_date': '20121220', + 'uploader_id': 'user7108434', + 'uploader': 'Filippo Valsorda', + 'duration': 10, }, }, { @@ -98,7 +106,7 @@ class VimeoIE(VimeoBaseInfoExtractor): 'uploader_id': 'openstreetmapus', 'uploader': 'OpenStreetMap US', 'title': 'Andy Allan - Putting the Carto into OpenStreetMap Cartography', - 'description': 'md5:380943ec71b89736ff4bf27183233d09', + 'description': 'md5:fd69a7b8d8c34a4e1d2ec2e4afd6ec30', 'duration': 1595, }, }, @@ -128,7 +136,7 @@ class VimeoIE(VimeoBaseInfoExtractor): 'uploader_id': 'user18948128', 'uploader': 'Jaime Marquínez Ferrándiz', 'duration': 10, - 'description': 'This is "youtube-dl password protected test video" by Jaime Marquínez Ferrándiz on Vimeo, the home for high quality videos and the people who love them.', + 'description': 'This is "youtube-dl password protected test video" by Jaime Marquínez Ferrándiz on Vimeo, the home for high quality videos and the people\u2026', }, 'params': { 'videopassword': 'youtube-dl', @@ -152,7 +160,6 @@ class VimeoIE(VimeoBaseInfoExtractor): }, { 'url': 'http://vimeo.com/76979871', - 'md5': '3363dd6ffebe3784d56f4132317fd446', 'note': 'Video with subtitles', 'info_dict': { 'id': '76979871', @@ -177,6 +184,11 @@ class VimeoIE(VimeoBaseInfoExtractor): 'uploader_id': 'user28849593', }, }, + { + 'url': 'https://vimeo.com/109815029', + 'note': 'Video not completely processed, "failed" seed status', + 'only_matching': True, + }, ] @staticmethod @@ -198,17 +210,18 @@ class VimeoIE(VimeoBaseInfoExtractor): password = self._downloader.params.get('videopassword', None) if password is None: raise ExtractorError('This video is protected by a password, use the --video-password option', expected=True) - token = self._extract_xsrft(webpage) - data = urlencode_postdata({ + token, vuid = self._extract_xsrft_and_vuid(webpage) + data = urlencode_postdata(encode_dict({ 'password': password, 'token': token, - }) + })) if url.startswith('http://'): # vimeo only supports https now, but the user can give an http url url = url.replace('http://', 'https://') password_request = compat_urllib_request.Request(url + '/password', data) password_request.add_header('Content-Type', 'application/x-www-form-urlencoded') password_request.add_header('Referer', url) + self._set_vimeo_cookie('vuid', vuid) return self._download_webpage( password_request, video_id, 'Verifying the password', 'Wrong password') @@ -217,7 +230,7 @@ class VimeoIE(VimeoBaseInfoExtractor): password = self._downloader.params.get('videopassword', None) if password is None: raise ExtractorError('This video is protected by a password, use the --video-password option') - data = compat_urllib_parse.urlencode({'password': password}) + data = urlencode_postdata(encode_dict({'password': password})) pass_url = url + '/check-password' password_request = compat_urllib_request.Request(pass_url, data) password_request.add_header('Content-Type', 'application/x-www-form-urlencoded') @@ -268,20 +281,30 @@ class VimeoIE(VimeoBaseInfoExtractor): self.report_extraction(video_id) vimeo_config = self._search_regex( - r'vimeo\.config\s*=\s*({.+?});', webpage, + r'vimeo\.config\s*=\s*(?:({.+?})|_extend\([^,]+,\s+({.+?})\));', webpage, 'vimeo config', default=None) if vimeo_config: seed_status = self._parse_json(vimeo_config, video_id).get('seed_status', {}) if seed_status.get('state') == 'failed': raise ExtractorError( - '%s returned error: %s' % (self.IE_NAME, seed_status['title']), + '%s said: %s' % (self.IE_NAME, seed_status['title']), expected=True) # Extract the config JSON try: try: config_url = self._html_search_regex( - r' data-config-url="(.+?)"', webpage, 'config URL') + r' data-config-url="(.+?)"', webpage, + 'config URL', default=None) + if not config_url: + # Sometimes new react-based page is served instead of old one that require + # different config URL extraction approach (see + # https://github.com/rg3/youtube-dl/pull/7209) + vimeo_clip_page_config = self._search_regex( + r'vimeo\.clip_page_config\s*=\s*({.+?});', webpage, + 'vimeo clip page config') + config_url = self._parse_json( + vimeo_clip_page_config, video_id)['player']['config_url'] config_json = self._download_webpage(config_url, video_id) config = json.loads(config_json) except RegexNotFoundError: @@ -364,41 +387,29 @@ class VimeoIE(VimeoBaseInfoExtractor): like_count = None comment_count = None - # Vimeo specific: extract request signature and timestamp - sig = config['request']['signature'] - timestamp = config['request']['timestamp'] - - # Vimeo specific: extract video codec and quality information - # First consider quality, then codecs, then take everything - codecs = [('vp6', 'flv'), ('vp8', 'flv'), ('h264', 'mp4')] - files = {'hd': [], 'sd': [], 'other': []} - config_files = config["video"].get("files") or config["request"].get("files") - for codec_name, codec_extension in codecs: - for quality in config_files.get(codec_name, []): - format_id = '-'.join((codec_name, quality)).lower() - key = quality if quality in files else 'other' - video_url = None - if isinstance(config_files[codec_name], dict): - file_info = config_files[codec_name][quality] - video_url = file_info.get('url') - else: - file_info = {} - if video_url is None: - video_url = "http://player.vimeo.com/play_redirect?clip_id=%s&sig=%s&time=%s&quality=%s&codecs=%s&type=moogaloop_local&embed_location=" \ - % (video_id, sig, timestamp, quality, codec_name.upper()) - - files[key].append({ - 'ext': codec_extension, - 'url': video_url, - 'format_id': format_id, - 'width': file_info.get('width'), - 'height': file_info.get('height'), - }) formats = [] - for key in ('other', 'sd', 'hd'): - formats += files[key] - if len(formats) == 0: - raise ExtractorError('No known codec found') + config_files = config['video'].get('files') or config['request'].get('files', {}) + for f in config_files.get('progressive', []): + video_url = f.get('url') + if not video_url: + continue + formats.append({ + 'url': video_url, + 'format_id': 'http-%s' % f.get('quality'), + 'width': int_or_none(f.get('width')), + 'height': int_or_none(f.get('height')), + 'fps': int_or_none(f.get('fps')), + 'tbr': int_or_none(f.get('bitrate')), + }) + m3u8_url = config_files.get('hls', {}).get('url') + if m3u8_url: + m3u8_formats = self._extract_m3u8_formats( + m3u8_url, video_id, 'mp4', 'm3u8_native', 0, 'hls', fatal=False) + if m3u8_formats: + formats.extend(m3u8_formats) + # Bitrates are completely broken. Single m3u8 may contain entries in kbps and bps + # at the same time without actual units specified. This lead to wrong sorting. + self._sort_formats(formats, field_preference=('height', 'width', 'fps', 'format_id')) subtitles = {} text_tracks = config['request'].get('text_tracks') @@ -459,16 +470,17 @@ class VimeoChannelIE(VimeoBaseInfoExtractor): if password is None: raise ExtractorError('This album is protected by a password, use the --video-password option', expected=True) fields = self._hidden_inputs(login_form) - token = self._extract_xsrft(webpage) + token, vuid = self._extract_xsrft_and_vuid(webpage) fields['token'] = token fields['password'] = password - post = urlencode_postdata(fields) + post = urlencode_postdata(encode_dict(fields)) password_path = self._search_regex( r'action="([^"]+)"', login_form, 'password URL') password_url = compat_urlparse.urljoin(page_url, password_path) password_request = compat_urllib_request.Request(password_url, post) password_request.add_header('Content-type', 'application/x-www-form-urlencoded') - self._set_cookie('vimeo.com', 'xsrft', token) + self._set_vimeo_cookie('vuid', vuid) + self._set_vimeo_cookie('xsrft', token) return self._download_webpage( password_request, list_id, diff --git a/youtube_dl/extractor/vine.py b/youtube_dl/extractor/vine.py index c733a48fa..cb2a4b0b5 100644 --- a/youtube_dl/extractor/vine.py +++ b/youtube_dl/extractor/vine.py @@ -1,10 +1,14 @@ +# coding: utf-8 from __future__ import unicode_literals import re import itertools from .common import InfoExtractor -from ..utils import unified_strdate +from ..utils import ( + int_or_none, + unified_strdate, +) class VineIE(InfoExtractor): @@ -17,10 +21,12 @@ class VineIE(InfoExtractor): 'ext': 'mp4', 'title': 'Chicken.', 'alt_title': 'Vine by Jack Dorsey', - 'description': 'Chicken.', 'upload_date': '20130519', 'uploader': 'Jack Dorsey', 'uploader_id': '76', + 'like_count': int, + 'comment_count': int, + 'repost_count': int, }, }, { 'url': 'https://vine.co/v/MYxVapFvz2z', @@ -29,11 +35,13 @@ class VineIE(InfoExtractor): 'id': 'MYxVapFvz2z', 'ext': 'mp4', 'title': 'Fuck Da Police #Mikebrown #justice #ferguson #prayforferguson #protesting #NMOS14', - 'alt_title': 'Vine by Luna', - 'description': 'Fuck Da Police #Mikebrown #justice #ferguson #prayforferguson #protesting #NMOS14', + 'alt_title': 'Vine by Mars Ruiz', 'upload_date': '20140815', - 'uploader': 'Luna', + 'uploader': 'Mars Ruiz', 'uploader_id': '1102363502380728320', + 'like_count': int, + 'comment_count': int, + 'repost_count': int, }, }, { 'url': 'https://vine.co/v/bxVjBbZlPUH', @@ -43,14 +51,33 @@ class VineIE(InfoExtractor): 'ext': 'mp4', 'title': '#mw3 #ac130 #killcam #angelofdeath', 'alt_title': 'Vine by Z3k3', - 'description': '#mw3 #ac130 #killcam #angelofdeath', 'upload_date': '20130430', 'uploader': 'Z3k3', 'uploader_id': '936470460173008896', + 'like_count': int, + 'comment_count': int, + 'repost_count': int, }, }, { 'url': 'https://vine.co/oembed/MYxVapFvz2z.json', 'only_matching': True, + }, { + 'url': 'https://vine.co/v/e192BnZnZ9V', + 'info_dict': { + 'id': 'e192BnZnZ9V', + 'ext': 'mp4', + 'title': 'ยิ้ม~ เขิน~ อาย~ น่าร้ากอ้ะ >//< @n_whitewo @orlameena #lovesicktheseries #lovesickseason2', + 'alt_title': 'Vine by Pimry_zaa', + 'upload_date': '20150705', + 'uploader': 'Pimry_zaa', + 'uploader_id': '1135760698325307392', + 'like_count': int, + 'comment_count': int, + 'repost_count': int, + }, + 'params': { + 'skip_download': True, + }, }] def _real_extract(self, url): @@ -58,32 +85,33 @@ class VineIE(InfoExtractor): webpage = self._download_webpage('https://vine.co/v/' + video_id, video_id) data = self._parse_json( - self._html_search_regex( - r'window\.POST_DATA = { %s: ({.+?}) };\s*</script>' % video_id, + self._search_regex( + r'window\.POST_DATA\s*=\s*{\s*%s\s*:\s*({.+?})\s*};\s*</script>' % video_id, webpage, 'vine data'), video_id) formats = [{ 'format_id': '%(format)s-%(rate)s' % f, - 'vcodec': f['format'], - 'quality': f['rate'], + 'vcodec': f.get('format'), + 'quality': f.get('rate'), 'url': f['videoUrl'], - } for f in data['videoUrls']] + } for f in data['videoUrls'] if f.get('videoUrl')] self._sort_formats(formats) + username = data.get('username') + return { 'id': video_id, - 'title': self._og_search_title(webpage), - 'alt_title': self._og_search_description(webpage, default=None), - 'description': data['description'], - 'thumbnail': data['thumbnailUrl'], - 'upload_date': unified_strdate(data['created']), - 'uploader': data['username'], - 'uploader_id': data['userIdStr'], - 'like_count': data['likes']['count'], - 'comment_count': data['comments']['count'], - 'repost_count': data['reposts']['count'], + 'title': data.get('description') or self._og_search_title(webpage), + 'alt_title': 'Vine by %s' % username if username else self._og_search_description(webpage, default=None), + 'thumbnail': data.get('thumbnailUrl'), + 'upload_date': unified_strdate(data.get('created')), + 'uploader': username, + 'uploader_id': data.get('userIdStr'), + 'like_count': int_or_none(data.get('likes', {}).get('count')), + 'comment_count': int_or_none(data.get('comments', {}).get('count')), + 'repost_count': int_or_none(data.get('reposts', {}).get('count')), 'formats': formats, } diff --git a/youtube_dl/extractor/vk.py b/youtube_dl/extractor/vk.py index c30c5a8e5..01960b827 100644 --- a/youtube_dl/extractor/vk.py +++ b/youtube_dl/extractor/vk.py @@ -17,6 +17,7 @@ from ..utils import ( unescapeHTML, unified_strdate, ) +from .vimeo import VimeoIE class VKIE(InfoExtractor): @@ -249,6 +250,10 @@ class VKIE(InfoExtractor): if youtube_url: return self.url_result(youtube_url, 'Youtube') + vimeo_url = VimeoIE._extract_vimeo_url(url, info_page) + if vimeo_url is not None: + return self.url_result(vimeo_url) + m_rutube = re.search( r'\ssrc="((?:https?:)?//rutube\.ru\\?/video\\?/embed(?:.*?))\\?"', info_page) if m_rutube is not None: @@ -276,9 +281,13 @@ class VKIE(InfoExtractor): mobj.group(1) + ' ' + mobj.group(2) upload_date = unified_strdate(mobj.group(1) + ' ' + mobj.group(2)) - view_count = str_to_int(self._search_regex( - r'"mv_views_count_number"[^>]*>([\d,.]+) views<', - info_page, 'view count', fatal=False)) + view_count = None + views = self._html_search_regex( + r'"mv_views_count_number"[^>]*>(.+?\bviews?)<', + info_page, 'view count', fatal=False) + if views: + view_count = str_to_int(self._search_regex( + r'([\d,.]+)', views, 'view count', fatal=False)) formats = [{ 'format_id': k, diff --git a/youtube_dl/extractor/wsj.py b/youtube_dl/extractor/wsj.py index 2ddf29a69..5a897371d 100644 --- a/youtube_dl/extractor/wsj.py +++ b/youtube_dl/extractor/wsj.py @@ -84,6 +84,5 @@ class WSJIE(InfoExtractor): 'duration': duration, 'upload_date': upload_date, 'title': title, - 'formats': formats, 'categories': categories, } diff --git a/youtube_dl/extractor/gorillavid.py b/youtube_dl/extractor/xfileshare.py index d23e3eac1..7610dc627 100644 --- a/youtube_dl/extractor/gorillavid.py +++ b/youtube_dl/extractor/xfileshare.py @@ -1,4 +1,4 @@ -# -*- coding: utf-8 -*- +# coding: utf-8 from __future__ import unicode_literals import re @@ -15,11 +15,11 @@ from ..utils import ( ) -class GorillaVidIE(InfoExtractor): - IE_DESC = 'GorillaVid.in, daclips.in, movpod.in, fastvideo.in, realvid.net and filehoot.com' +class XFileShareIE(InfoExtractor): + IE_DESC = 'XFileShare based sites: GorillaVid.in, daclips.in, movpod.in, fastvideo.in, realvid.net, filehoot.com and vidto.me' _VALID_URL = r'''(?x) https?://(?P<host>(?:www\.)? - (?:daclips\.in|gorillavid\.in|movpod\.in|fastvideo\.in|realvid\.net|filehoot\.com))/ + (?:daclips\.in|gorillavid\.in|movpod\.in|fastvideo\.in|realvid\.net|filehoot\.com|vidto.\me))/ (?:embed-)?(?P<id>[0-9a-zA-Z]+)(?:-[0-9]+x[0-9]+\.html)? ''' @@ -76,6 +76,13 @@ class GorillaVidIE(InfoExtractor): 'title': 'youtube-dl test video \'äBaW_jenozKc.mp4.mp4', 'thumbnail': 're:http://.*\.jpg', } + }, { + 'url': 'http://vidto.me/ku5glz52nqe1.html', + 'info_dict': { + 'id': 'ku5glz52nqe1', + 'ext': 'mp4', + 'title': 'test' + } }] def _real_extract(self, url): @@ -104,13 +111,18 @@ class GorillaVidIE(InfoExtractor): webpage = self._download_webpage(req, video_id, 'Downloading video page') - title = self._search_regex( - [r'style="z-index: [0-9]+;">([^<]+)</span>', r'<td nowrap>([^<]+)</td>', r'>Watch (.+) '], - webpage, 'title', default=None) or self._og_search_title(webpage) + title = (self._search_regex( + [r'style="z-index: [0-9]+;">([^<]+)</span>', + r'<td nowrap>([^<]+)</td>', + r'>Watch (.+) ', + r'<h2 class="video-page-head">([^<]+)</h2>'], + webpage, 'title', default=None) or self._og_search_title(webpage)).strip() video_url = self._search_regex( - r'file\s*:\s*["\'](http[^"\']+)["\'],', webpage, 'file url') + [r'file\s*:\s*["\'](http[^"\']+)["\'],', + r'file_link\s*=\s*\'(https?:\/\/[0-9a-zA-z.\/\-_]+)'], + webpage, 'file url') thumbnail = self._search_regex( - r'image\s*:\s*["\'](http[^"\']+)["\'],', webpage, 'thumbnail', fatal=False) + r'image\s*:\s*["\'](http[^"\']+)["\'],', webpage, 'thumbnail', default=None) formats = [{ 'format_id': 'sd', diff --git a/youtube_dl/extractor/xhamster.py b/youtube_dl/extractor/xhamster.py index 97315750f..8938c0e45 100644 --- a/youtube_dl/extractor/xhamster.py +++ b/youtube_dl/extractor/xhamster.py @@ -63,7 +63,9 @@ class XHamsterIE(InfoExtractor): mrss_url = '%s://xhamster.com/movies/%s/%s.html' % (proto, video_id, seo) webpage = self._download_webpage(mrss_url, video_id) - title = self._html_search_regex(r'<title>(?P<title>.+?) - xHamster\.com</title>', webpage, 'title') + title = self._html_search_regex( + [r'<title>(?P<title>.+?)(?:, (?:[^,]+? )?Porn: xHamster| - xHamster\.com)</title>', + r'<h1>([^<]+)</h1>'], webpage, 'title') # Only a few videos have an description mobj = re.search(r'<span>Description: </span>([^<]+)', webpage) diff --git a/youtube_dl/extractor/yandexmusic.py b/youtube_dl/extractor/yandexmusic.py index 4098e4629..08dc81f3a 100644 --- a/youtube_dl/extractor/yandexmusic.py +++ b/youtube_dl/extractor/yandexmusic.py @@ -46,6 +46,12 @@ class YandexMusicTrackIE(InfoExtractor): % (data['host'], key, data['ts'] + data['path'], storage[1])) def _get_track_info(self, track): + thumbnail = None + cover_uri = track.get('albums', [{}])[0].get('coverUri') + if cover_uri: + thumbnail = cover_uri.replace('%%', 'orig') + if not thumbnail.startswith('http'): + thumbnail = 'http://' + thumbnail return { 'id': track['id'], 'ext': 'mp3', @@ -53,6 +59,7 @@ class YandexMusicTrackIE(InfoExtractor): 'title': '%s - %s' % (track['artists'][0]['name'], track['title']), 'filesize': int_or_none(track.get('fileSize')), 'duration': float_or_none(track.get('durationMs'), 1000), + 'thumbnail': thumbnail, } def _real_extract(self, url): diff --git a/youtube_dl/extractor/youporn.py b/youtube_dl/extractor/youporn.py index 4ba7c36db..9bf8d1eeb 100644 --- a/youtube_dl/extractor/youporn.py +++ b/youtube_dl/extractor/youporn.py @@ -1,121 +1,171 @@ from __future__ import unicode_literals - -import json import re -import sys from .common import InfoExtractor -from ..compat import ( - compat_urllib_parse_urlparse, - compat_urllib_request, -) +from ..compat import compat_urllib_request from ..utils import ( - ExtractorError, + int_or_none, + str_to_int, unescapeHTML, unified_strdate, ) -from ..aes import ( - aes_decrypt_text -) +from ..aes import aes_decrypt_text class YouPornIE(InfoExtractor): - _VALID_URL = r'^(?P<proto>https?://)(?:www\.)?(?P<url>youporn\.com/watch/(?P<videoid>[0-9]+)/(?P<title>[^/]+))' - _TEST = { + _VALID_URL = r'https?://(?:www\.)?youporn\.com/watch/(?P<id>\d+)/(?P<display_id>[^/?#&]+)' + _TESTS = [{ 'url': 'http://www.youporn.com/watch/505835/sex-ed-is-it-safe-to-masturbate-daily/', + 'md5': '71ec5fcfddacf80f495efa8b6a8d9a89', 'info_dict': { 'id': '505835', + 'display_id': 'sex-ed-is-it-safe-to-masturbate-daily', 'ext': 'mp4', - 'upload_date': '20101221', + 'title': 'Sex Ed: Is It Safe To Masturbate Daily?', 'description': 'Love & Sex Answers: http://bit.ly/DanAndJenn -- Is It Unhealthy To Masturbate Daily?', + 'thumbnail': 're:^https?://.*\.jpg$', 'uploader': 'Ask Dan And Jennifer', - 'title': 'Sex Ed: Is It Safe To Masturbate Daily?', + 'upload_date': '20101221', + 'average_rating': int, + 'view_count': int, + 'comment_count': int, + 'categories': list, + 'tags': list, 'age_limit': 18, - } - } + }, + }, { + # Anonymous User uploader + 'url': 'http://www.youporn.com/watch/561726/big-tits-awesome-brunette-on-amazing-webcam-show/?from=related3&al=2&from_id=561726&pos=4', + 'info_dict': { + 'id': '561726', + 'display_id': 'big-tits-awesome-brunette-on-amazing-webcam-show', + 'ext': 'mp4', + 'title': 'Big Tits Awesome Brunette On amazing webcam show', + 'description': 'http://sweetlivegirls.com Big Tits Awesome Brunette On amazing webcam show.mp4', + 'thumbnail': 're:^https?://.*\.jpg$', + 'uploader': 'Anonymous User', + 'upload_date': '20111125', + 'average_rating': int, + 'view_count': int, + 'comment_count': int, + 'categories': list, + 'tags': list, + 'age_limit': 18, + }, + 'params': { + 'skip_download': True, + }, + }] def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) - video_id = mobj.group('videoid') - url = mobj.group('proto') + 'www.' + mobj.group('url') + video_id = mobj.group('id') + display_id = mobj.group('display_id') - req = compat_urllib_request.Request(url) - req.add_header('Cookie', 'age_verified=1') - webpage = self._download_webpage(req, video_id) - age_limit = self._rta_search(webpage) + request = compat_urllib_request.Request(url) + request.add_header('Cookie', 'age_verified=1') + webpage = self._download_webpage(request, display_id) + + title = self._search_regex( + [r'(?:video_titles|videoTitle)\s*[:=]\s*(["\'])(?P<title>.+?)\1', + r'<h1[^>]+class=["\']heading\d?["\'][^>]*>([^<])<'], + webpage, 'title', group='title') - # Get JSON parameters - json_params = self._search_regex( - [r'videoJa?son\s*=\s*({.+})', - r'var\s+currentVideo\s*=\s*new\s+Video\((.+?)\)[,;]'], - webpage, 'JSON parameters') - try: - params = json.loads(json_params) - except ValueError: - raise ExtractorError('Invalid JSON') - - self.report_extraction(video_id) - try: - video_title = params['title'] - upload_date = unified_strdate(params['release_date_f']) - video_description = params['description'] - video_uploader = params['submitted_by'] - thumbnail = params['thumbnails'][0]['image'] - except KeyError: - raise ExtractorError('Missing JSON parameter: ' + sys.exc_info()[1]) - - # Get all of the links from the page - DOWNLOAD_LIST_RE = r'(?s)<ul class="downloadList">(?P<download_list>.*?)</ul>' - download_list_html = self._search_regex(DOWNLOAD_LIST_RE, - webpage, 'download list').strip() - LINK_RE = r'<a href="([^"]+)">' - links = re.findall(LINK_RE, download_list_html) - - # Get all encrypted links - encrypted_links = re.findall(r'var encryptedQuality[0-9]{3}URL = \'([a-zA-Z0-9+/]+={0,2})\';', webpage) - for encrypted_link in encrypted_links: - link = aes_decrypt_text(encrypted_link, video_title, 32).decode('utf-8') + links = [] + + sources = self._search_regex( + r'sources\s*:\s*({.+?})', webpage, 'sources', default=None) + if sources: + for _, link in re.findall(r'[^:]+\s*:\s*(["\'])(http.+?)\1', sources): + links.append(link) + + # Fallback #1 + for _, link in re.findall( + r'(?:videoUrl|videoSrc|videoIpadUrl|html5PlayerSrc)\s*[:=]\s*(["\'])(http.+?)\1', webpage): + links.append(link) + + # Fallback #2, this also contains extra low quality 180p format + for _, link in re.findall(r'<a[^>]+href=(["\'])(http.+?)\1[^>]+title=["\']Download [Vv]ideo', webpage): links.append(link) + # Fallback #3, encrypted links + for _, encrypted_link in re.findall( + r'encryptedQuality\d{3,4}URL\s*=\s*(["\'])([\da-zA-Z+/=]+)\1', webpage): + links.append(aes_decrypt_text(encrypted_link, title, 32).decode('utf-8')) + formats = [] - for link in links: - # A link looks like this: - # http://cdn1.download.youporn.phncdn.com/201210/31/8004515/480p_370k_8004515/YouPorn%20-%20Nubile%20Films%20The%20Pillow%20Fight.mp4?nvb=20121113051249&nva=20121114051249&ir=1200&sr=1200&hash=014b882080310e95fb6a0 - # A path looks like this: - # /201210/31/8004515/480p_370k_8004515/YouPorn%20-%20Nubile%20Films%20The%20Pillow%20Fight.mp4 - video_url = unescapeHTML(link) - path = compat_urllib_parse_urlparse(video_url).path - format_parts = path.split('/')[4].split('_')[:2] - - dn = compat_urllib_parse_urlparse(video_url).netloc.partition('.')[0] - - resolution = format_parts[0] - height = int(resolution[:-len('p')]) - bitrate = int(format_parts[1][:-len('k')]) - format = '-'.join(format_parts) + '-' + dn - - formats.append({ + for video_url in set(unescapeHTML(link) for link in links): + f = { 'url': video_url, - 'format': format, - 'format_id': format, - 'height': height, - 'tbr': bitrate, - 'resolution': resolution, - }) - + } + # Video URL's path looks like this: + # /201012/17/505835/720p_1500k_505835/YouPorn%20-%20Sex%20Ed%20Is%20It%20Safe%20To%20Masturbate%20Daily.mp4 + # We will benefit from it by extracting some metadata + mobj = re.search(r'/(?P<height>\d{3,4})[pP]_(?P<bitrate>\d+)[kK]_\d+/', video_url) + if mobj: + height = int(mobj.group('height')) + bitrate = int(mobj.group('bitrate')) + f.update({ + 'format_id': '%dp-%dk' % (height, bitrate), + 'height': height, + 'tbr': bitrate, + }) + formats.append(f) self._sort_formats(formats) - if not formats: - raise ExtractorError('ERROR: no known formats available for video') + description = self._html_search_regex( + r'(?s)<div[^>]+class=["\']video-description["\'][^>]*>(.+?)</div>', + webpage, 'description', default=None) + thumbnail = self._search_regex( + r'(?:imageurl\s*=|poster\s*:)\s*(["\'])(?P<thumbnail>.+?)\1', + webpage, 'thumbnail', fatal=False, group='thumbnail') + + uploader = self._html_search_regex( + r'(?s)<div[^>]+class=["\']videoInfoBy["\'][^>]*>\s*By:\s*</div>(.+?)</(?:a|div)>', + webpage, 'uploader', fatal=False) + upload_date = unified_strdate(self._html_search_regex( + r'(?s)<div[^>]+class=["\']videoInfoTime["\'][^>]*>(.+?)</div>', + webpage, 'upload date', fatal=False)) + + age_limit = self._rta_search(webpage) + + average_rating = int_or_none(self._search_regex( + r'<div[^>]+class=["\']videoInfoRating["\'][^>]*>\s*<div[^>]+class=["\']videoRatingPercentage["\'][^>]*>(\d+)%</div>', + webpage, 'average rating', fatal=False)) + + view_count = str_to_int(self._search_regex( + r'(?s)<div[^>]+class=["\']videoInfoViews["\'][^>]*>.*?([\d,.]+)\s*</div>', + webpage, 'view count', fatal=False)) + comment_count = str_to_int(self._search_regex( + r'>All [Cc]omments? \(([\d,.]+)\)', + webpage, 'comment count', fatal=False)) + + def extract_tag_box(title): + tag_box = self._search_regex( + (r'<div[^>]+class=["\']tagBoxTitle["\'][^>]*>\s*%s\b.*?</div>\s*' + '<div[^>]+class=["\']tagBoxContent["\']>(.+?)</div>') % re.escape(title), + webpage, '%s tag box' % title, default=None) + if not tag_box: + return [] + return re.findall(r'<a[^>]+href=[^>]+>([^<]+)', tag_box) + + categories = extract_tag_box('Category') + tags = extract_tag_box('Tags') return { 'id': video_id, - 'uploader': video_uploader, - 'upload_date': upload_date, - 'title': video_title, + 'display_id': display_id, + 'title': title, + 'description': description, 'thumbnail': thumbnail, - 'description': video_description, + 'uploader': uploader, + 'upload_date': upload_date, + 'average_rating': average_rating, + 'view_count': view_count, + 'comment_count': comment_count, + 'categories': categories, + 'tags': tags, 'age_limit': age_limit, 'formats': formats, } diff --git a/youtube_dl/extractor/youtube.py b/youtube_dl/extractor/youtube.py index 97ce36550..687e0b4db 100644 --- a/youtube_dl/extractor/youtube.py +++ b/youtube_dl/extractor/youtube.py @@ -178,6 +178,52 @@ class YoutubeBaseInfoExtractor(InfoExtractor): return +class YoutubePlaylistBaseInfoExtractor(InfoExtractor): + # Extract the video ids from the playlist pages + def _entries(self, page, playlist_id): + more_widget_html = content_html = page + for page_num in itertools.count(1): + for video_id, video_title in self.extract_videos_from_page(content_html): + yield self.url_result( + video_id, 'Youtube', video_id=video_id, + video_title=video_title) + + mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html) + if not mobj: + break + + more = self._download_json( + 'https://youtube.com/%s' % mobj.group('more'), playlist_id, + 'Downloading page #%s' % page_num, + transform_source=uppercase_escape) + content_html = more['content_html'] + if not content_html.strip(): + # Some webpages show a "Load more" button but they don't + # have more videos + break + more_widget_html = more['load_more_widget_html'] + + def extract_videos_from_page(self, page): + ids_in_page = [] + titles_in_page = [] + for mobj in re.finditer(self._VIDEO_RE, page): + # The link with index 0 is not the first video of the playlist (not sure if still actual) + if 'index' in mobj.groupdict() and mobj.group('id') == '0': + continue + video_id = mobj.group('id') + video_title = unescapeHTML(mobj.group('title')) + if video_title: + video_title = video_title.strip() + try: + idx = ids_in_page.index(video_id) + if video_title and not titles_in_page[idx]: + titles_in_page[idx] = video_title + except ValueError: + ids_in_page.append(video_id) + titles_in_page.append(video_title) + return zip(ids_in_page, titles_in_page) + + class YoutubeIE(YoutubeBaseInfoExtractor): IE_DESC = 'YouTube.com' _VALID_URL = r"""(?x)^ @@ -657,7 +703,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): def _extract_signature_function(self, video_id, player_url, example_sig): id_m = re.match( - r'.*?-(?P<id>[a-zA-Z0-9_-]+)(?:/watch_as3|/html5player(?:-new)?)?\.(?P<ext>[a-z]+)$', + r'.*?-(?P<id>[a-zA-Z0-9_-]+)(?:/watch_as3|/html5player(?:-new)?|/base)?\.(?P<ext>[a-z]+)$', player_url) if not id_m: raise ExtractorError('Cannot identify player %r' % player_url) @@ -1061,6 +1107,17 @@ class YoutubeIE(YoutubeBaseInfoExtractor): if not video_info: video_info = get_video_info if 'token' in get_video_info: + # Different get_video_info requests may report different results, e.g. + # some may report video unavailability, but some may serve it without + # any complaint (see https://github.com/rg3/youtube-dl/issues/7362, + # the original webpage as well as el=info and el=embedded get_video_info + # requests report video unavailability due to geo restriction while + # el=detailpage succeeds and returns valid data). This is probably + # due to YouTube measures against IP ranges of hosting providers. + # Working around by preferring the first succeeded video_info containing + # the token if no such video_info yet was found. + if 'token' not in video_info: + video_info = get_video_info break if 'token' not in video_info: if 'reason' in video_info: @@ -1286,7 +1343,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): player_desc = 'flash player %s' % player_version else: player_version = self._search_regex( - r'html5player-([^/]+?)(?:/html5player(?:-new)?)?\.js', + [r'html5player-([^/]+?)(?:/html5player(?:-new)?)?\.js', r'(?:www|player)-([^/]+)/base\.js'], player_url, 'html5 player', fatal=False) player_desc = 'html5 player %s' % player_version @@ -1419,7 +1476,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): } -class YoutubePlaylistIE(YoutubeBaseInfoExtractor): +class YoutubePlaylistIE(YoutubeBaseInfoExtractor, YoutubePlaylistBaseInfoExtractor): IE_DESC = 'YouTube.com playlists' _VALID_URL = r"""(?x)(?: (?:https?://)? @@ -1440,7 +1497,7 @@ class YoutubePlaylistIE(YoutubeBaseInfoExtractor): ((?:PL|LL|EC|UU|FL|RD|UL)[0-9A-Za-z-_]{10,}) )""" _TEMPLATE_URL = 'https://www.youtube.com/playlist?list=%s' - _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})&[^"]*?index=(?P<index>\d+)' + _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})&[^"]*?index=(?P<index>\d+)(?:[^>]+>(?P<title>[^<]+))?' IE_NAME = 'youtube:playlist' _TESTS = [{ 'url': 'https://www.youtube.com/playlist?list=PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re', @@ -1557,37 +1614,11 @@ class YoutubePlaylistIE(YoutubeBaseInfoExtractor): else: self.report_warning('Youtube gives an alert message: ' + match) - # Extract the video ids from the playlist pages - def _entries(): - more_widget_html = content_html = page - for page_num in itertools.count(1): - matches = re.finditer(self._VIDEO_RE, content_html) - # We remove the duplicates and the link with index 0 - # (it's not the first video of the playlist) - new_ids = orderedSet(m.group('id') for m in matches if m.group('index') != '0') - for vid_id in new_ids: - yield self.url_result(vid_id, 'Youtube', video_id=vid_id) - - mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html) - if not mobj: - break - - more = self._download_json( - 'https://youtube.com/%s' % mobj.group('more'), playlist_id, - 'Downloading page #%s' % page_num, - transform_source=uppercase_escape) - content_html = more['content_html'] - if not content_html.strip(): - # Some webpages show a "Load more" button but they don't - # have more videos - break - more_widget_html = more['load_more_widget_html'] - playlist_title = self._html_search_regex( r'(?s)<h1 class="pl-header-title[^"]*">\s*(.*?)\s*</h1>', page, 'title') - return self.playlist_result(_entries(), playlist_id, playlist_title) + return self.playlist_result(self._entries(page, playlist_id), playlist_id, playlist_title) def _real_extract(self, url): # Extract playlist id @@ -1613,36 +1644,31 @@ class YoutubePlaylistIE(YoutubeBaseInfoExtractor): return self._extract_playlist(playlist_id) -class YoutubeChannelIE(InfoExtractor): +class YoutubeChannelIE(YoutubePlaylistBaseInfoExtractor): IE_DESC = 'YouTube.com channels' _VALID_URL = r'https?://(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/(?P<id>[0-9A-Za-z_-]+)' _TEMPLATE_URL = 'https://www.youtube.com/channel/%s/videos' + _VIDEO_RE = r'(?:title="(?P<title>[^"]+)"[^>]+)?href="/watch\?v=(?P<id>[0-9A-Za-z_-]+)&?' IE_NAME = 'youtube:channel' _TESTS = [{ 'note': 'paginated channel', 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w', 'playlist_mincount': 91, 'info_dict': { - 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w', + 'id': 'UUKfVa3S1e4PHvxWcwyMMg8w', + 'title': 'Uploads from lex will', } + }, { + 'note': 'Age restricted channel', + # from https://www.youtube.com/user/DeusExOfficial + 'url': 'https://www.youtube.com/channel/UCs0ifCMCm1icqRbqhUINa0w', + 'playlist_mincount': 64, + 'info_dict': { + 'id': 'UUs0ifCMCm1icqRbqhUINa0w', + 'title': 'Uploads from Deus Ex', + }, }] - @staticmethod - def extract_videos_from_page(page): - ids_in_page = [] - titles_in_page = [] - for mobj in re.finditer(r'(?:title="(?P<title>[^"]+)"[^>]+)?href="/watch\?v=(?P<id>[0-9A-Za-z_-]+)&?', page): - video_id = mobj.group('id') - video_title = unescapeHTML(mobj.group('title')) - try: - idx = ids_in_page.index(video_id) - if video_title and not titles_in_page[idx]: - titles_in_page[idx] = video_title - except ValueError: - ids_in_page.append(video_id) - titles_in_page.append(video_title) - return zip(ids_in_page, titles_in_page) - def _real_extract(self, url): channel_id = self._match_id(url) @@ -1654,12 +1680,15 @@ class YoutubeChannelIE(InfoExtractor): channel_page = self._download_webpage( url + '?view=57', channel_id, 'Downloading channel page', fatal=False) - channel_playlist_id = self._html_search_meta( - 'channelId', channel_page, 'channel id', default=None) - if not channel_playlist_id: - channel_playlist_id = self._search_regex( - r'data-channel-external-id="([^"]+)"', - channel_page, 'channel id', default=None) + if channel_page is False: + channel_playlist_id = False + else: + channel_playlist_id = self._html_search_meta( + 'channelId', channel_page, 'channel id', default=None) + if not channel_playlist_id: + channel_playlist_id = self._search_regex( + r'data-(?:channel-external-|yt)id="([^"]+)"', + channel_page, 'channel id', default=None) if channel_playlist_id and channel_playlist_id.startswith('UC'): playlist_id = 'UU' + channel_playlist_id[2:] return self.url_result( @@ -1682,29 +1711,7 @@ class YoutubeChannelIE(InfoExtractor): for video_id, video_title in self.extract_videos_from_page(channel_page)] return self.playlist_result(entries, channel_id) - def _entries(): - more_widget_html = content_html = channel_page - for pagenum in itertools.count(1): - - for video_id, video_title in self.extract_videos_from_page(content_html): - yield self.url_result( - video_id, 'Youtube', video_id=video_id, - video_title=video_title) - - mobj = re.search( - r'data-uix-load-more-href="/?(?P<more>[^"]+)"', - more_widget_html) - if not mobj: - break - - more = self._download_json( - 'https://youtube.com/%s' % mobj.group('more'), channel_id, - 'Downloading page #%s' % (pagenum + 1), - transform_source=uppercase_escape) - content_html = more['content_html'] - more_widget_html = more['load_more_widget_html'] - - return self.playlist_result(_entries(), channel_id) + return self.playlist_result(self._entries(channel_page, channel_id), channel_id) class YoutubeUserIE(YoutubeChannelIE): @@ -1970,6 +1977,7 @@ class YoutubeTruncatedURLIE(InfoExtractor): annotation_id=annotation_[^&]+| x-yt-cl=[0-9]+| hl=[^&]*| + t=[0-9]+ )? | attribution_link\?a=[^&]+ @@ -1992,6 +2000,9 @@ class YoutubeTruncatedURLIE(InfoExtractor): }, { 'url': 'https://www.youtube.com/watch?hl=en-GB', 'only_matching': True, + }, { + 'url': 'https://www.youtube.com/watch?t=2372', + 'only_matching': True, }] def _real_extract(self, url): diff --git a/youtube_dl/extractor/zdf.py b/youtube_dl/extractor/zdf.py index 98f15177b..a795f56b3 100644 --- a/youtube_dl/extractor/zdf.py +++ b/youtube_dl/extractor/zdf.py @@ -9,6 +9,7 @@ from ..utils import ( int_or_none, unified_strdate, OnDemandPagedList, + xpath_text, ) @@ -19,13 +20,11 @@ def extract_from_xml_url(ie, video_id, xml_url): errnote='Failed to download video info') title = doc.find('.//information/title').text - description = doc.find('.//information/detail').text - duration = int(doc.find('.//details/lengthSec').text) - uploader_node = doc.find('.//details/originChannelTitle') - uploader = None if uploader_node is None else uploader_node.text - uploader_id_node = doc.find('.//details/originChannelId') - uploader_id = None if uploader_id_node is None else uploader_id_node.text - upload_date = unified_strdate(doc.find('.//details/airtime').text) + description = xpath_text(doc, './/information/detail', 'description') + duration = int_or_none(xpath_text(doc, './/details/lengthSec', 'duration')) + uploader = xpath_text(doc, './/details/originChannelTitle', 'uploader') + uploader_id = xpath_text(doc, './/details/originChannelId', 'uploader id') + upload_date = unified_strdate(xpath_text(doc, './/details/airtime', 'upload date')) def xml_to_format(fnode): video_url = fnode.find('url').text @@ -40,15 +39,14 @@ def extract_from_xml_url(ie, video_id, xml_url): ext = format_m.group('container') proto = format_m.group('proto').lower() - quality = fnode.find('./quality').text - abr = int(fnode.find('./audioBitrate').text) // 1000 - vbr_node = fnode.find('./videoBitrate') - vbr = None if vbr_node is None else int(vbr_node.text) // 1000 + quality = xpath_text(fnode, './quality', 'quality') + abr = int_or_none(xpath_text(fnode, './audioBitrate', 'abr'), 1000) + vbr = int_or_none(xpath_text(fnode, './videoBitrate', 'vbr'), 1000) - width_node = fnode.find('./width') - width = None if width_node is None else int_or_none(width_node.text) - height_node = fnode.find('./height') - height = None if height_node is None else int_or_none(height_node.text) + width = int_or_none(xpath_text(fnode, './width', 'width')) + height = int_or_none(xpath_text(fnode, './height', 'height')) + + filesize = int_or_none(xpath_text(fnode, './filesize', 'filesize')) format_note = '' if not format_note: @@ -64,12 +62,31 @@ def extract_from_xml_url(ie, video_id, xml_url): 'vbr': vbr, 'width': width, 'height': height, - 'filesize': int_or_none(fnode.find('./filesize').text), + 'filesize': filesize, 'format_note': format_note, 'protocol': proto, '_available': is_available, } + def xml_to_thumbnails(fnode): + thumbnails = [] + for node in fnode: + thumbnail_url = node.text + if not thumbnail_url: + continue + thumbnail = { + 'url': thumbnail_url, + } + if 'key' in node.attrib: + m = re.match('^([0-9]+)x([0-9]+)$', node.attrib['key']) + if m: + thumbnail['width'] = int(m.group(1)) + thumbnail['height'] = int(m.group(2)) + thumbnails.append(thumbnail) + return thumbnails + + thumbnails = xml_to_thumbnails(doc.findall('.//teaserimages/teaserimage')) + format_nodes = doc.findall('.//formitaeten/formitaet') formats = list(filter( lambda f: f['_available'], @@ -81,6 +98,7 @@ def extract_from_xml_url(ie, video_id, xml_url): 'title': title, 'description': description, 'duration': duration, + 'thumbnails': thumbnails, 'uploader': uploader, 'uploader_id': uploader_id, 'upload_date': upload_date, diff --git a/youtube_dl/extractor/zingmp3.py b/youtube_dl/extractor/zingmp3.py index 7dc1e2f2b..437eecb67 100644 --- a/youtube_dl/extractor/zingmp3.py +++ b/youtube_dl/extractor/zingmp3.py @@ -9,9 +9,11 @@ from ..utils import ExtractorError class ZingMp3BaseInfoExtractor(InfoExtractor): - def _extract_item(self, item): + def _extract_item(self, item, fatal=True): error_message = item.find('./errormessage').text if error_message: + if not fatal: + return raise ExtractorError( '%s returned error: %s' % (self.IE_NAME, error_message), expected=True) @@ -43,7 +45,9 @@ class ZingMp3BaseInfoExtractor(InfoExtractor): entries = [] for i, item in enumerate(items, 1): - entry = self._extract_item(item) + entry = self._extract_item(item, fatal=False) + if not entry: + continue entry['id'] = '%s-%d' % (id, i) entries.append(entry) @@ -85,7 +89,7 @@ class ZingMp3SongIE(ZingMp3BaseInfoExtractor): class ZingMp3AlbumIE(ZingMp3BaseInfoExtractor): - _VALID_URL = r'https?://mp3\.zing\.vn/album/(?P<slug>[^/]+)/(?P<album_id>\w+)\.html' + _VALID_URL = r'https?://mp3\.zing\.vn/(?:album|playlist)/(?P<slug>[^/]+)/(?P<album_id>\w+)\.html' _TESTS = [{ 'url': 'http://mp3.zing.vn/album/Lau-Dai-Tinh-Ai-Bang-Kieu-Minh-Tuyet/ZWZBWDAF.html', 'info_dict': { @@ -94,6 +98,9 @@ class ZingMp3AlbumIE(ZingMp3BaseInfoExtractor): 'title': 'Lâu Đài Tình Ái - Bằng Kiều ft. Minh Tuyết | Album 320 lossless', }, 'playlist_count': 10, + }, { + 'url': 'http://mp3.zing.vn/playlist/Duong-Hong-Loan-apollobee/IWCAACCB.html', + 'only_matching': True, }] IE_NAME = 'zingmp3:album' IE_DESC = 'mp3.zing.vn albums' |