diff options
Diffstat (limited to 'youtube_dl/extractor')
170 files changed, 4080 insertions, 1782 deletions
diff --git a/youtube_dl/extractor/__init__.py b/youtube_dl/extractor/__init__.py index 495a18c17..947b83683 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 @@ -59,7 +60,10 @@ from .bloomberg import BloombergIE from .bpb import BpbIE from .br import BRIE from .breakcom import BreakIE -from .brightcove import BrightcoveIE +from .brightcove import ( + BrightcoveLegacyIE, + BrightcoveNewIE, +) from .buzzfeed import BuzzFeedIE from .byutv import BYUtvIE from .c56 import C56IE @@ -76,6 +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, @@ -88,6 +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 ( @@ -121,10 +127,12 @@ 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 from .douyutv import DouyuTVIE +from .dplay import DPlayIE from .dramafever import ( DramaFeverIE, DramaFeverSeriesIE, @@ -166,6 +174,7 @@ 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 @@ -207,13 +216,15 @@ 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 @@ -295,6 +306,11 @@ from .lifenews import ( LifeNewsIE, LifeEmbedIE, ) +from .limelight import ( + LimelightMediaIE, + LimelightChannelIE, + LimelightChannelListIE, +) from .liveleak import LiveLeakIE from .livestream import ( LivestreamIE, @@ -312,7 +328,6 @@ 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 @@ -406,7 +421,10 @@ from .nowness import ( NownessPlaylistIE, NownessSeriesIE, ) -from .nowtv import NowTVIE +from .nowtv import ( + NowTVIE, + NowTVListIE, +) from .nowvideo import NowVideoIE from .npo import ( NPOIE, @@ -444,10 +462,7 @@ from .orf import ( from .parliamentliveuk import ParliamentLiveUKIE from .patreon import PatreonIE from .pbs import PBSIE -from .periscope import ( - PeriscopeIE, - QuickscopeIE, -) +from .periscope import PeriscopeIE from .philharmoniedeparis import PhilharmonieDeParisIE from .phoenix import PhoenixIE from .photobucket import PhotobucketIE @@ -561,7 +576,8 @@ from .soundcloud import ( SoundcloudIE, SoundcloudSetIE, SoundcloudUserIE, - SoundcloudPlaylistIE + SoundcloudPlaylistIE, + SoundcloudSearchIE ) from .soundgasm import ( SoundgasmIE, @@ -580,6 +596,7 @@ 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, @@ -684,7 +701,7 @@ from .twitch import ( TwitchBookmarksIE, TwitchStreamIE, ) -from .twitter import TwitterCardIE +from .twitter import TwitterCardIE, TwitterIE from .ubu import UbuIE from .udemy import ( UdemyIE, @@ -711,7 +728,6 @@ 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 @@ -721,6 +737,7 @@ 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, @@ -773,6 +790,7 @@ 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, @@ -816,6 +834,7 @@ from .youtube import ( YoutubeTruncatedIDIE, YoutubeTruncatedURLIE, YoutubeUserIE, + YoutubeUserPlaylistsIE, YoutubeWatchLaterIE, ) from .zapiks import ZapiksIE 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 27de07587..3ae618e71 100644 --- a/youtube_dl/extractor/adultswim.py +++ b/youtube_dl/extractor/adultswim.py @@ -41,7 +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': [ @@ -134,7 +135,13 @@ class AdultSwimIE(InfoExtractor): show = bootstrapped_data['show'] show_title = show['title'] stream = video_info.get('stream') - clips = [stream] if stream else video_info['clips'] + 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'] @@ -176,7 +183,7 @@ class AdultSwimIE(InfoExtractor): media_url = file_el.text if determine_ext(media_url) == 'm3u8': formats.extend(self._extract_m3u8_formats( - media_url, segment_title, 'mp4', 'm3u8_native', preference=0, m3u8_id='hls')) + media_url, segment_title, 'mp4', preference=0, m3u8_id='hls')) else: formats.append({ 'format_id': '%s_%s' % (bitrate, ftype), diff --git a/youtube_dl/extractor/aljazeera.py b/youtube_dl/extractor/aljazeera.py index 184a14a4f..5b2c0dc9a 100644 --- a/youtube_dl/extractor/aljazeera.py +++ b/youtube_dl/extractor/aljazeera.py @@ -15,7 +15,7 @@ class AlJazeeraIE(InfoExtractor): 'description': 'As a birth attendant advocating for family planning, Remy is on the frontline of Tondo\'s battle with overcrowding.', 'uploader': 'Al Jazeera English', }, - 'add_ie': ['Brightcove'], + 'add_ie': ['BrightcoveLegacy'], 'skip': 'Not accessible from Travis CI server', } @@ -32,5 +32,5 @@ class AlJazeeraIE(InfoExtractor): 'playerKey=AQ~~%2CAAAAmtVJIFk~%2CTVGOQ5ZTwJbeMWnq5d_H4MOM57xfzApc' '&%40videoPlayer={0}'.format(brightcove_id) ), - 'ie_key': 'Brightcove', + 'ie_key': 'BrightcoveLegacy', } 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/atresplayer.py b/youtube_dl/extractor/atresplayer.py index 29f8795d3..50e47ba0a 100644 --- a/youtube_dl/extractor/atresplayer.py +++ b/youtube_dl/extractor/atresplayer.py @@ -7,11 +7,11 @@ from .common import InfoExtractor from ..compat import ( compat_str, compat_urllib_parse, - compat_urllib_request, ) from ..utils import ( int_or_none, float_or_none, + sanitized_Request, xpath_text, ExtractorError, ) @@ -63,7 +63,7 @@ class AtresPlayerIE(InfoExtractor): 'j_password': password, } - request = compat_urllib_request.Request( + request = sanitized_Request( self._LOGIN_URL, compat_urllib_parse.urlencode(login_form).encode('utf-8')) request.add_header('Content-Type', 'application/x-www-form-urlencoded') response = self._download_webpage( @@ -94,7 +94,7 @@ class AtresPlayerIE(InfoExtractor): formats = [] for fmt in ['windows', 'android_tablet']: - request = compat_urllib_request.Request( + request = sanitized_Request( self._URL_VIDEO_TEMPLATE.format(fmt, episode_id, timestamp_shifted, token)) request.add_header('User-Agent', self._USER_AGENT) diff --git a/youtube_dl/extractor/bambuser.py b/youtube_dl/extractor/bambuser.py index 8dff1d6e3..da986e063 100644 --- a/youtube_dl/extractor/bambuser.py +++ b/youtube_dl/extractor/bambuser.py @@ -6,13 +6,13 @@ import itertools from .common import InfoExtractor from ..compat import ( compat_urllib_parse, - compat_urllib_request, compat_str, ) from ..utils import ( ExtractorError, int_or_none, float_or_none, + sanitized_Request, ) @@ -57,7 +57,7 @@ class BambuserIE(InfoExtractor): 'pass': password, } - request = compat_urllib_request.Request( + request = sanitized_Request( self._LOGIN_URL, compat_urllib_parse.urlencode(login_form).encode('utf-8')) request.add_header('Referer', self._LOGIN_URL) response = self._download_webpage( @@ -126,7 +126,7 @@ class BambuserChannelIE(InfoExtractor): '&sort=created&access_mode=0%2C1%2C2&limit={count}' '&method=broadcast&format=json&vid_older_than={last}' ).format(user=user, count=self._STEP, last=last_id) - req = compat_urllib_request.Request(req_url) + req = sanitized_Request(req_url) # Without setting this header, we wouldn't get any result req.add_header('Referer', 'http://bambuser.com/channel/%s' % user) data = self._download_json( 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 cc2f6fed2..33b296eaf 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 ( @@ -11,23 +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://www.bbc.co.uk/programmes/b06bp7lf). Also may fail with selectionunavailable. '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', @@ -193,6 +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': @@ -218,7 +231,7 @@ class BBCCoUkIE(InfoExtractor): else: formats.append({ 'url': href, - 'format_id': supplier, + 'format_id': supplier or kind or protocol, }) elif protocol == 'rtmp': application = connection.get('application', 'ondemand') @@ -238,16 +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 = [] @@ -261,13 +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 @@ -312,7 +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 in ('notukerror', 'geolocation'): + if e.id in ('notukerror', 'geolocation', 'selectionunavailable'): last_exception = e continue self._raise_extractor_error(e) @@ -323,8 +345,8 @@ class BBCCoUkIE(InfoExtractor): media_selection = self._download_xml( 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')) + if isinstance(ee.cause, compat_HTTPError) and ee.cause.code in (403, 404): + media_selection = compat_etree_fromstring(ee.cause.read().decode('utf-8')) else: raise return self._process_media_selector(media_selection, programme_id) @@ -382,7 +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': @@ -399,8 +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): @@ -409,16 +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 @@ -470,6 +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) @@ -479,8 +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', @@ -489,7 +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', @@ -500,6 +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', @@ -507,12 +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', @@ -522,15 +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', }, @@ -538,13 +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', }, @@ -586,21 +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': { @@ -608,6 +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, @@ -619,7 +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. @@ -648,40 +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) @@ -803,3 +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/bliptv.py b/youtube_dl/extractor/bliptv.py index c3296283d..35375f7b1 100644 --- a/youtube_dl/extractor/bliptv.py +++ b/youtube_dl/extractor/bliptv.py @@ -4,14 +4,12 @@ import re from .common import InfoExtractor -from ..compat import ( - compat_urllib_request, - compat_urlparse, -) +from ..compat import compat_urlparse from ..utils import ( clean_html, int_or_none, parse_iso8601, + sanitized_Request, unescapeHTML, xpath_text, xpath_with_ns, @@ -219,7 +217,7 @@ class BlipTVIE(InfoExtractor): for lang, url in subtitles_urls.items(): # For some weird reason, blip.tv serves a video instead of subtitles # when we request with a common UA - req = compat_urllib_request.Request(url) + req = sanitized_Request(url) req.add_header('User-Agent', 'youtube-dl') subtitles[lang] = [{ # The extension is 'srt' but it's actually an 'ass' file diff --git a/youtube_dl/extractor/bloomberg.py b/youtube_dl/extractor/bloomberg.py index 0dca29b71..ebeef8f2a 100644 --- a/youtube_dl/extractor/bloomberg.py +++ b/youtube_dl/extractor/bloomberg.py @@ -6,9 +6,9 @@ from .common import InfoExtractor class BloombergIE(InfoExtractor): - _VALID_URL = r'https?://www\.bloomberg\.com/news/videos/[^/]+/(?P<id>[^/?#]+)' + _VALID_URL = r'https?://(?:www\.)?bloomberg\.com/(?:[^/]+/)*(?P<id>[^/?#]+)' - _TEST = { + _TESTS = [{ 'url': 'http://www.bloomberg.com/news/videos/b/aaeae121-5949-481e-a1ce-4562db6f5df2', # The md5 checksum changes 'info_dict': { @@ -17,22 +17,39 @@ class BloombergIE(InfoExtractor): 'title': 'Shah\'s Presentation on Foreign-Exchange Strategies', 'description': 'md5:a8ba0302912d03d246979735c17d2761', }, - } + }, { + 'url': 'http://www.bloomberg.com/news/articles/2015-11-12/five-strange-things-that-have-been-happening-in-financial-markets', + 'only_matching': True, + }, { + 'url': 'http://www.bloomberg.com/politics/videos/2015-11-25/karl-rove-on-jeb-bush-s-struggles-stopping-trump', + 'only_matching': True, + }] def _real_extract(self, url): name = self._match_id(url) webpage = self._download_webpage(url, name) - video_id = self._search_regex(r'"bmmrId":"(.+?)"', webpage, 'id') + video_id = self._search_regex( + r'["\']bmmrId["\']\s*:\s*(["\'])(?P<url>.+?)\1', + webpage, 'id', group='url') title = re.sub(': Video$', '', self._og_search_title(webpage)) embed_info = self._download_json( 'http://www.bloomberg.com/api/embed?id=%s' % video_id, video_id) formats = [] for stream in embed_info['streams']: - if stream["muxing_format"] == "TS": - formats.extend(self._extract_m3u8_formats(stream['url'], video_id)) + stream_url = stream.get('url') + if not stream_url: + continue + if stream['muxing_format'] == 'TS': + m3u8_formats = self._extract_m3u8_formats( + stream_url, video_id, 'mp4', m3u8_id='hls', fatal=False) + if m3u8_formats: + formats.extend(m3u8_formats) else: - formats.extend(self._extract_f4m_formats(stream['url'], video_id)) + f4m_formats = self._extract_f4m_formats( + stream_url, video_id, f4m_id='hds', fatal=False) + if f4m_formats: + formats.extend(f4m_formats) self._sort_formats(formats) return { diff --git a/youtube_dl/extractor/brightcove.py b/youtube_dl/extractor/brightcove.py index 4721c2293..f5ebae1e6 100644 --- a/youtube_dl/extractor/brightcove.py +++ b/youtube_dl/extractor/brightcove.py @@ -3,15 +3,14 @@ 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, compat_urllib_parse_urlparse, - compat_urllib_request, compat_urlparse, compat_xml_parse_error, ) @@ -20,12 +19,18 @@ from ..utils import ( ExtractorError, find_xpath_attr, fix_xml_ampersands, + float_or_none, + js_to_json, + int_or_none, + parse_iso8601, + sanitized_Request, unescapeHTML, unsmuggle_url, ) -class BrightcoveIE(InfoExtractor): +class BrightcoveLegacyIE(InfoExtractor): + IE_NAME = 'brightcove:legacy' _VALID_URL = r'(?:https?://.*brightcove\.com/(services|viewer).*?\?|brightcove:)(?P<query>.*)' _FEDERATED_URL_TEMPLATE = 'http://c.brightcove.com/services/viewer/htmlFederated?%s' @@ -119,7 +124,7 @@ 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 @@ -245,7 +250,7 @@ class BrightcoveIE(InfoExtractor): def _get_video_info(self, video_id, query_str, query, referer=None): request_url = self._FEDERATED_URL_TEMPLATE % query_str - req = compat_urllib_request.Request(request_url) + req = sanitized_Request(request_url) linkBase = query.get('linkBaseURL') if linkBase is not None: referer = linkBase[0] @@ -346,3 +351,172 @@ class BrightcoveIE(InfoExtractor): if 'url' not in info and not info.get('formats'): raise ExtractorError('Unable to extract video url for %s' % info['id']) return info + + +class BrightcoveNewIE(InfoExtractor): + IE_NAME = 'brightcove:new' + _VALID_URL = r'https?://players\.brightcove\.net/(?P<account_id>\d+)/(?P<player_id>[^/]+)_(?P<embed>[^/]+)/index\.html\?.*videoId=(?P<video_id>\d+)' + _TESTS = [{ + 'url': 'http://players.brightcove.net/929656772001/e41d32dc-ec74-459e-a845-6c69f7b724ea_default/index.html?videoId=4463358922001', + 'md5': 'c8100925723840d4b0d243f7025703be', + 'info_dict': { + 'id': '4463358922001', + 'ext': 'mp4', + 'title': 'Meet the man behind Popcorn Time', + 'description': 'md5:eac376a4fe366edc70279bfb681aea16', + 'duration': 165.768, + 'timestamp': 1441391203, + 'upload_date': '20150904', + 'uploader_id': '929656772001', + 'formats': 'mincount:22', + }, + }, { + # with rtmp streams + 'url': 'http://players.brightcove.net/4036320279001/5d112ed9-283f-485f-a7f9-33f42e8bc042_default/index.html?videoId=4279049078001', + 'info_dict': { + 'id': '4279049078001', + 'ext': 'mp4', + 'title': 'Titansgrave: Chapter 0', + 'description': 'Titansgrave: Chapter 0', + 'duration': 1242.058, + 'timestamp': 1433556729, + 'upload_date': '20150606', + 'uploader_id': '4036320279001', + 'formats': 'mincount:41', + }, + 'params': { + 'skip_download': True, + } + }] + + @staticmethod + def _extract_urls(webpage): + # Reference: + # 1. http://docs.brightcove.com/en/video-cloud/brightcove-player/guides/publish-video.html#setvideoiniframe + # 2. http://docs.brightcove.com/en/video-cloud/brightcove-player/guides/publish-video.html#setvideousingjavascript) + # 3. http://docs.brightcove.com/en/video-cloud/brightcove-player/guides/embed-in-page.html + + entries = [] + + # Look for iframe embeds [1] + for _, url in re.findall( + r'<iframe[^>]+src=(["\'])((?:https?:)//players\.brightcove\.net/\d+/[^/]+/index\.html.+?)\1', webpage): + entries.append(url) + + # Look for embed_in_page embeds [2] + for video_id, account_id, player_id, embed in re.findall( + # According to examples from [3] it's unclear whether video id + # may be optional and what to do when it is + r'''(?sx) + <video[^>]+ + data-video-id=["\'](\d+)["\'][^>]*>.*? + </video>.*? + <script[^>]+ + src=["\'](?:https?:)?//players\.brightcove\.net/ + (\d+)/([\da-f-]+)_([^/]+)/index\.min\.js + ''', webpage): + entries.append( + 'http://players.brightcove.net/%s/%s_%s/index.html?videoId=%s' + % (account_id, player_id, embed, video_id)) + + return entries + + def _real_extract(self, url): + account_id, player_id, embed, video_id = re.match(self._VALID_URL, url).groups() + + webpage = self._download_webpage( + 'http://players.brightcove.net/%s/%s_%s/index.min.js' + % (account_id, player_id, embed), video_id) + + policy_key = None + + catalog = self._search_regex( + r'catalog\(({.+?})\);', webpage, 'catalog', default=None) + if catalog: + catalog = self._parse_json( + js_to_json(catalog), video_id, fatal=False) + if catalog: + policy_key = catalog.get('policyKey') + + if not policy_key: + policy_key = self._search_regex( + r'policyKey\s*:\s*(["\'])(?P<pk>.+?)\1', + webpage, 'policy key', group='pk') + + req = sanitized_Request( + 'https://edge.api.brightcove.com/playback/v1/accounts/%s/videos/%s' + % (account_id, video_id), + headers={'Accept': 'application/json;pk=%s' % policy_key}) + json_data = self._download_json(req, video_id) + + title = json_data['name'] + + formats = [] + for source in json_data.get('sources', []): + source_type = source.get('type') + src = source.get('src') + if source_type == 'application/x-mpegURL': + if not src: + continue + m3u8_formats = self._extract_m3u8_formats( + src, video_id, 'mp4', entry_protocol='m3u8_native', + m3u8_id='hls', fatal=False) + if m3u8_formats: + formats.extend(m3u8_formats) + else: + streaming_src = source.get('streaming_src') + stream_name, app_name = source.get('stream_name'), source.get('app_name') + if not src and not streaming_src and (not stream_name or not app_name): + continue + tbr = float_or_none(source.get('avg_bitrate'), 1000) + height = int_or_none(source.get('height')) + f = { + 'tbr': tbr, + 'width': int_or_none(source.get('width')), + 'height': height, + 'filesize': int_or_none(source.get('size')), + 'container': source.get('container'), + 'vcodec': source.get('codec'), + 'ext': source.get('container').lower(), + } + + def build_format_id(kind): + format_id = kind + if tbr: + format_id += '-%dk' % int(tbr) + if height: + format_id += '-%dp' % height + return format_id + + if src or streaming_src: + f.update({ + 'url': src or streaming_src, + 'format_id': build_format_id('http' if src else 'http-streaming'), + 'preference': 2 if src else 1, + }) + else: + f.update({ + 'url': app_name, + 'play_path': stream_name, + 'format_id': build_format_id('rtmp'), + }) + formats.append(f) + self._sort_formats(formats) + + description = json_data.get('description') + thumbnail = json_data.get('thumbnail') + timestamp = parse_iso8601(json_data.get('published_at')) + duration = float_or_none(json_data.get('duration'), 1000) + tags = json_data.get('tags', []) + + return { + 'id': video_id, + 'title': title, + 'description': description, + 'thumbnail': thumbnail, + 'duration': duration, + 'timestamp': timestamp, + 'uploader_id': account_id, + 'formats': formats, + 'tags': tags, + } 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..40d07ab18 100644 --- a/youtube_dl/extractor/cbs.py +++ b/youtube_dl/extractor/cbs.py @@ -1,6 +1,10 @@ from __future__ import unicode_literals from .common import InfoExtractor +from ..utils import ( + sanitized_Request, + smuggle_url, +) class CBSIE(InfoExtractor): @@ -46,13 +50,19 @@ class CBSIE(InfoExtractor): def _real_extract(self, url): display_id = self._match_id(url) - webpage = self._download_webpage(url, display_id) + request = sanitized_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/ceskatelevize.py b/youtube_dl/extractor/ceskatelevize.py index e857e66f4..6f7b2a70d 100644 --- a/youtube_dl/extractor/ceskatelevize.py +++ b/youtube_dl/extractor/ceskatelevize.py @@ -5,7 +5,6 @@ import re from .common import InfoExtractor from ..compat import ( - compat_urllib_request, compat_urllib_parse, compat_urllib_parse_unquote, compat_urllib_parse_urlparse, @@ -13,6 +12,7 @@ from ..compat import ( from ..utils import ( ExtractorError, float_or_none, + sanitized_Request, ) @@ -100,7 +100,7 @@ class CeskaTelevizeIE(InfoExtractor): 'requestSource': 'iVysilani', } - req = compat_urllib_request.Request( + req = sanitized_Request( 'http://www.ceskatelevize.cz/ivysilani/ajax/get-client-playlist', data=compat_urllib_parse.urlencode(data)) @@ -115,7 +115,7 @@ class CeskaTelevizeIE(InfoExtractor): if playlist_url == 'error_region': raise ExtractorError(NOT_AVAILABLE_STRING, expected=True) - req = compat_urllib_request.Request(compat_urllib_parse_unquote(playlist_url)) + req = sanitized_Request(compat_urllib_parse_unquote(playlist_url)) req.add_header('Referer', url) playlist_title = self._og_search_title(webpage) 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/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/collegerama.py b/youtube_dl/extractor/collegerama.py index fedd48490..40667a0f1 100644 --- a/youtube_dl/extractor/collegerama.py +++ b/youtube_dl/extractor/collegerama.py @@ -3,10 +3,10 @@ from __future__ import unicode_literals import json from .common import InfoExtractor -from ..compat import compat_urllib_request from ..utils import ( float_or_none, int_or_none, + sanitized_Request, ) @@ -52,7 +52,7 @@ class CollegeRamaIE(InfoExtractor): } } - request = compat_urllib_request.Request( + request = sanitized_Request( 'http://collegerama.tudelft.nl/Mediasite/PlayerService/PlayerService.svc/json/GetPlayerOptions', json.dumps(player_options_request)) request.add_header('Content-Type', 'application/json') diff --git a/youtube_dl/extractor/common.py b/youtube_dl/extractor/common.py index 4fe2307cd..eb9bfa3d1 100644 --- a/youtube_dl/extractor/common.py +++ b/youtube_dl/extractor/common.py @@ -10,20 +10,18 @@ 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, compat_urllib_parse_urlparse, - compat_urllib_request, compat_urlparse, compat_str, + compat_etree_fromstring, ) from ..utils import ( NO_DEFAULT, @@ -38,6 +36,7 @@ from ..utils import ( int_or_none, RegexNotFoundError, sanitize_filename, + sanitized_Request, unescapeHTML, unified_strdate, url_basename, @@ -165,12 +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 @@ -309,11 +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 """ @@ -460,7 +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', @@ -644,8 +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), @@ -839,7 +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 @@ -889,6 +891,11 @@ class InfoExtractor(object): if not media_nodes: manifest_version = '2.0' media_nodes = manifest.findall('{http://ns.adobe.com/f4m/2.0}media') + base_url = xpath_text( + manifest, ['{http://ns.adobe.com/f4m/1.0}baseURL', '{http://ns.adobe.com/f4m/2.0}baseURL'], + 'base URL', default=None) + if base_url: + base_url = base_url.strip() for i, media_el in enumerate(media_nodes): if manifest_version == '2.0': media_url = media_el.attrib.get('href') or media_el.attrib.get('url') @@ -896,7 +903,7 @@ class InfoExtractor(object): continue manifest_url = ( media_url if media_url.startswith('http://') or media_url.startswith('https://') - else ('/'.join(manifest_url.split('/')[:-1]) + '/' + media_url)) + else ((base_url or '/'.join(manifest_url.split('/')[:-1])) + '/' + media_url)) # If media_url is itself a f4m manifest do the recursive extraction # since bitrates in parent manifest (this one) and media_url manifest # may differ leading to inability to resolve the format by requested @@ -941,13 +948,15 @@ 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( @@ -1276,7 +1285,7 @@ class InfoExtractor(object): def _get_cookies(self, url): """ Return a compat_cookies.SimpleCookie with the cookies for the url """ - req = compat_urllib_request.Request(url) + req = sanitized_Request(url) self._downloader.cookiejar.add_cookie_header(req) return compat_cookies.SimpleCookie(req.get_header('Cookie')) 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..00d943f77 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, @@ -21,7 +21,9 @@ from ..utils import ( bytes_to_intlist, intlist_to_bytes, int_or_none, + lowercase_escape, remove_end, + sanitized_Request, unified_strdate, urlencode_postdata, xpath_text, @@ -32,9 +34,29 @@ 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 = sanitized_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)) + else sanitized_Request(url_or_request)) # Accept-Language must be set explicitly to accept any language to avoid issues # similar to https://github.com/rg3/youtube-dl/issues/6797. # Along with IP address Crunchyroll uses Accept-Language to guess whether georestriction @@ -46,10 +68,22 @@ 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]+))(?:[/?&]|$)' - _NETRC_MACHINE = 'crunchyroll' _TESTS = [{ 'url': 'http://www.crunchyroll.com/wanna-be-the-strongest-in-the-world/episode-1-an-idol-wrestler-is-born-645513', 'info_dict': { @@ -72,7 +106,7 @@ class CrunchyrollIE(CrunchyrollBaseIE): 'id': '589804', 'ext': 'flv', 'title': 'Culture Japan Episode 1 – Rebuilding Japan after the 3.11', - 'description': 'md5:fe2743efedb49d279552926d0bd0cd9e', + 'description': 'md5:2fbc01f90b87e8e9137296f37b461c12', 'thumbnail': 're:^https?://.*\.jpg$', 'uploader': 'Danny Choo Network', 'upload_date': '20120213', @@ -81,10 +115,13 @@ 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 +131,6 @@ 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 +236,7 @@ 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 +247,7 @@ 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 +273,7 @@ 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 +289,15 @@ 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) @@ -285,7 +308,7 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text 'video_uploader', fatal=False) playerdata_url = compat_urllib_parse_unquote(self._html_search_regex(r'"config_url":"([^"]+)', webpage, 'playerdata_url')) - playerdata_req = compat_urllib_request.Request(playerdata_url) + playerdata_req = sanitized_Request(playerdata_url) playerdata_req.data = compat_urllib_parse.urlencode({'current_page': webpage_url}) playerdata_req.add_header('Content-Type', 'application/x-www-form-urlencoded') playerdata = self._download_webpage(playerdata_req, video_id, note='Downloading media info') @@ -297,7 +320,7 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text for fmt in re.findall(r'showmedia\.([0-9]{3,4})p', webpage): stream_quality, stream_format = self._FORMAT_IDS[fmt] video_format = fmt + 'p' - streamdata_req = compat_urllib_request.Request( + streamdata_req = sanitized_Request( 'http://www.crunchyroll.com/xml/?req=RpcApiVideoPlayer_GetStandardConfig&media_id=%s&video_format=%s&video_quality=%s' % (stream_id, stream_format, stream_quality), compat_urllib_parse.urlencode({'current_page': url}).encode('utf-8')) @@ -352,7 +375,7 @@ 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 +384,25 @@ 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..ab7f3aec4 100644 --- a/youtube_dl/extractor/dailymotion.py +++ b/youtube_dl/extractor/dailymotion.py @@ -7,15 +7,13 @@ import itertools from .common import InfoExtractor -from ..compat import ( - compat_str, - compat_urllib_request, -) +from ..compat import compat_str from ..utils import ( ExtractorError, determine_ext, int_or_none, parse_iso8601, + sanitized_Request, str_to_int, unescapeHTML, ) @@ -25,7 +23,7 @@ class DailymotionBaseInfoExtractor(InfoExtractor): @staticmethod def _build_request(url): """Build a request with the family filter disabled""" - request = compat_urllib_request.Request(url) + request = sanitized_Request(url) request.add_header('Cookie', 'family_filter=off; ff=off') return request @@ -96,6 +94,11 @@ class DailymotionIE(DailymotionBaseInfoExtractor): 'uploader': 'HotWaves1012', 'age_limit': 18, } + }, + # geo-restricted, player v5 + { + 'url': 'http://www.dailymotion.com/video/xhza0o', + 'only_matching': True, } ] @@ -119,11 +122,14 @@ 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 +139,17 @@ 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 +215,7 @@ 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 +258,11 @@ 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/dbtv.py b/youtube_dl/extractor/dbtv.py index 212217625..133cdc50b 100644 --- a/youtube_dl/extractor/dbtv.py +++ b/youtube_dl/extractor/dbtv.py @@ -13,8 +13,8 @@ from ..utils import ( class DBTVIE(InfoExtractor): - _VALID_URL = r'http://dbtv\.no/(?P<id>[0-9]+)#(?P<display_id>.+)' - _TEST = { + _VALID_URL = r'https?://(?:www\.)?dbtv\.no/(?:(?:lazyplayer|player)/)?(?P<id>[0-9]+)(?:#(?P<display_id>.+))?' + _TESTS = [{ 'url': 'http://dbtv.no/3649835190001#Skulle_teste_ut_fornøyelsespark,_men_kollegaen_var_bare_opptatt_av_bikinikroppen', 'md5': 'b89953ed25dacb6edb3ef6c6f430f8bc', 'info_dict': { @@ -30,12 +30,18 @@ class DBTVIE(InfoExtractor): 'view_count': int, 'categories': list, } - } + }, { + 'url': 'http://dbtv.no/3649835190001', + 'only_matching': True, + }, { + 'url': 'http://www.dbtv.no/lazyplayer/4631135248001', + 'only_matching': True, + }] def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) video_id = mobj.group('id') - display_id = mobj.group('display_id') + display_id = mobj.group('display_id') or video_id data = self._download_json( 'http://api.dbtv.no/discovery/%s' % video_id, display_id) diff --git a/youtube_dl/extractor/dcn.py b/youtube_dl/extractor/dcn.py index 6f2fea5ff..9737cff14 100644 --- a/youtube_dl/extractor/dcn.py +++ b/youtube_dl/extractor/dcn.py @@ -2,13 +2,11 @@ from __future__ import unicode_literals from .common import InfoExtractor -from ..compat import ( - compat_urllib_parse, - compat_urllib_request, -) +from ..compat import compat_urllib_parse from ..utils import ( int_or_none, parse_iso8601, + sanitized_Request, ) @@ -36,7 +34,7 @@ class DCNIE(InfoExtractor): def _real_extract(self, url): video_id = self._match_id(url) - request = compat_urllib_request.Request( + request = sanitized_Request( 'http://admin.mangomolo.com/analytics/index.php/plus/video?id=%s' % video_id, headers={'Origin': 'http://www.dcndigital.ae'}) 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/dplay.py b/youtube_dl/extractor/dplay.py new file mode 100644 index 000000000..6cda56a7f --- /dev/null +++ b/youtube_dl/extractor/dplay.py @@ -0,0 +1,51 @@ +# encoding: utf-8 +from __future__ import unicode_literals + +import time + +from .common import InfoExtractor +from ..utils import int_or_none + + +class DPlayIE(InfoExtractor): + _VALID_URL = r'http://www\.dplay\.se/[^/]+/(?P<id>[^/?#]+)' + + _TEST = { + 'url': 'http://www.dplay.se/nugammalt-77-handelser-som-format-sverige/season-1-svensken-lar-sig-njuta-av-livet/', + 'info_dict': { + 'id': '3172', + 'ext': 'mp4', + 'display_id': 'season-1-svensken-lar-sig-njuta-av-livet', + 'title': 'Svensken lär sig njuta av livet', + 'duration': 2650, + }, + } + + def _real_extract(self, url): + display_id = self._match_id(url) + webpage = self._download_webpage(url, display_id) + video_id = self._search_regex( + r'data-video-id="(\d+)"', webpage, 'video id') + + info = self._download_json( + 'http://www.dplay.se/api/v2/ajax/videos?video_id=' + video_id, + video_id)['data'][0] + + self._set_cookie( + 'secure.dplay.se', 'dsc-geo', + '{"countryCode":"NL","expiry":%d}' % ((time.time() + 20 * 60) * 1000)) + # TODO: consider adding support for 'stream_type=hds', it seems to + # require setting some cookies + manifest_url = self._download_json( + 'https://secure.dplay.se/secure/api/v2/user/authorization/stream/%s?stream_type=hls' % video_id, + video_id, 'Getting manifest url for hls stream')['hls'] + formats = self._extract_m3u8_formats( + manifest_url, video_id, ext='mp4', entry_protocol='m3u8_native') + + return { + 'id': video_id, + 'display_id': display_id, + 'title': info['title'], + 'formats': formats, + 'duration': int_or_none(info.get('video_metadata_length'), scale=1000), + } diff --git a/youtube_dl/extractor/dramafever.py b/youtube_dl/extractor/dramafever.py index 38e6597c8..d836c1a6c 100644 --- a/youtube_dl/extractor/dramafever.py +++ b/youtube_dl/extractor/dramafever.py @@ -7,7 +7,6 @@ from .common import InfoExtractor from ..compat import ( compat_HTTPError, compat_urllib_parse, - compat_urllib_request, compat_urlparse, ) from ..utils import ( @@ -16,6 +15,7 @@ from ..utils import ( determine_ext, int_or_none, parse_iso8601, + sanitized_Request, ) @@ -51,7 +51,7 @@ class DramaFeverBaseIE(InfoExtractor): 'password': password, } - request = compat_urllib_request.Request( + request = sanitized_Request( self._LOGIN_URL, compat_urllib_parse.urlencode(login_form).encode('utf-8')) response = self._download_webpage( request, None, 'Logging in as %s' % username) diff --git a/youtube_dl/extractor/dumpert.py b/youtube_dl/extractor/dumpert.py index 1f00386fe..e5aadcd25 100644 --- a/youtube_dl/extractor/dumpert.py +++ b/youtube_dl/extractor/dumpert.py @@ -2,14 +2,17 @@ from __future__ import unicode_literals import base64 +import re from .common import InfoExtractor -from ..compat import compat_urllib_request -from ..utils import qualities +from ..utils import ( + qualities, + sanitized_Request, +) class DumpertIE(InfoExtractor): - _VALID_URL = r'https?://(?:www\.)?dumpert\.nl/(?:mediabase|embed)/(?P<id>[0-9]+/[0-9a-zA-Z]+)' + _VALID_URL = r'(?P<protocol>https?)://(?:www\.)?dumpert\.nl/(?:mediabase|embed)/(?P<id>[0-9]+/[0-9a-zA-Z]+)' _TESTS = [{ 'url': 'http://www.dumpert.nl/mediabase/6646981/951bc60f/', 'md5': '1b9318d7d5054e7dcb9dc7654f21d643', @@ -26,10 +29,12 @@ class DumpertIE(InfoExtractor): }] def _real_extract(self, url): - video_id = self._match_id(url) + mobj = re.match(self._VALID_URL, url) + video_id = mobj.group('id') + protocol = mobj.group('protocol') - url = 'https://www.dumpert.nl/mediabase/' + video_id - req = compat_urllib_request.Request(url) + url = '%s://www.dumpert.nl/mediabase/%s' % (protocol, video_id) + req = sanitized_Request(url) req.add_header('Cookie', 'nsfw=1; cpc=10') webpage = self._download_webpage(req, video_id) diff --git a/youtube_dl/extractor/eagleplatform.py b/youtube_dl/extractor/eagleplatform.py index e529b9b96..7bbf617d4 100644 --- a/youtube_dl/extractor/eagleplatform.py +++ b/youtube_dl/extractor/eagleplatform.py @@ -87,7 +87,7 @@ class EaglePlatformIE(InfoExtractor): m3u8_url = self._get_video_url(secure_m3u8, video_id, 'Downloading m3u8 JSON') formats = self._extract_m3u8_formats( m3u8_url, video_id, - 'mp4', entry_protocol='m3u8_native') + '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 diff --git a/youtube_dl/extractor/eitb.py b/youtube_dl/extractor/eitb.py index 2cba82532..c83845fc2 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 ..utils import ( + float_or_none, + int_or_none, + parse_iso8601, + sanitized_Request, +) 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 = sanitized_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/escapist.py b/youtube_dl/extractor/escapist.py index c85b4c458..a3d7bbbcb 100644 --- a/youtube_dl/extractor/escapist.py +++ b/youtube_dl/extractor/escapist.py @@ -3,13 +3,12 @@ from __future__ import unicode_literals import json from .common import InfoExtractor -from ..compat import compat_urllib_request - from ..utils import ( determine_ext, clean_html, int_or_none, float_or_none, + sanitized_Request, ) @@ -75,7 +74,7 @@ class EscapistIE(InfoExtractor): video_id = ims_video['videoID'] key = ims_video['hash'] - config_req = compat_urllib_request.Request( + config_req = sanitized_Request( 'http://www.escapistmagazine.com/videos/' 'vidconfig.php?videoID=%s&hash=%s' % (video_id, key)) config_req.add_header('Referer', url) diff --git a/youtube_dl/extractor/everyonesmixtape.py b/youtube_dl/extractor/everyonesmixtape.py index d872d828f..493d38af8 100644 --- a/youtube_dl/extractor/everyonesmixtape.py +++ b/youtube_dl/extractor/everyonesmixtape.py @@ -3,11 +3,9 @@ from __future__ import unicode_literals import re from .common import InfoExtractor -from ..compat import ( - compat_urllib_request, -) from ..utils import ( ExtractorError, + sanitized_Request, ) @@ -42,7 +40,7 @@ class EveryonesMixtapeIE(InfoExtractor): playlist_id = mobj.group('id') pllist_url = 'http://everyonesmixtape.com/mixtape.php?a=getMixes&u=-1&linked=%s&explore=' % playlist_id - pllist_req = compat_urllib_request.Request(pllist_url) + pllist_req = sanitized_Request(pllist_url) pllist_req.add_header('X-Requested-With', 'XMLHttpRequest') playlist_list = self._download_json( @@ -55,7 +53,7 @@ class EveryonesMixtapeIE(InfoExtractor): raise ExtractorError('Playlist id not found') pl_url = 'http://everyonesmixtape.com/mixtape.php?a=getMix&id=%s&userId=null&code=' % playlist_no - pl_req = compat_urllib_request.Request(pl_url) + pl_req = sanitized_Request(pl_url) pl_req.add_header('X-Requested-With', 'XMLHttpRequest') playlist = self._download_json( pl_req, playlist_id, note='Downloading playlist info') 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..3403581fd 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 ..utils import ( - qualities, + int_or_none, + sanitized_Request, str_to_int, ) class ExtremeTubeIE(InfoExtractor): - _VALID_URL = r'https?://(?:www\.)?(?P<url>extremetube\.com/.*?video/.+?(?P<id>[0-9]+))(?:[/?&]|$)' + _VALID_URL = r'https?://(?:www\.)?extremetube\.com/(?:[^/]+/)?video/(?P<id>[^/#?&]+)' _TESTS = [{ 'url': 'http://www.extremetube.com/video/music-video-14-british-euro-brit-european-cumshots-swallow-652431', 'md5': '344d0c6d50e2f16b06e49ca011d8ac69', 'info_dict': { - 'id': '652431', + 'id': 'music-video-14-british-euro-brit-european-cumshots-swallow-652431', 'ext': 'mp4', 'title': 'Music Video 14 british euro brit european cumshots swallow', 'uploader': 'unknown', @@ -29,14 +26,18 @@ class ExtremeTubeIE(InfoExtractor): }, { 'url': 'http://www.extremetube.com/gay/video/abcde-1234', 'only_matching': True, + }, { + 'url': 'http://www.extremetube.com/video/latina-slut-fucked-by-fat-black-dick', + 'only_matching': True, + }, { + 'url': 'http://www.extremetube.com/video/652431', + 'only_matching': True, }] def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - video_id = mobj.group('id') - url = 'http://www.' + mobj.group('url') + video_id = self._match_id(url) - req = compat_urllib_request.Request(url) + req = sanitized_Request(url) req.add_header('Cookie', 'age_verified=1') webpage = self._download_webpage(req, video_id) @@ -49,20 +50,36 @@ class ExtremeTubeIE(InfoExtractor): r'Views:\s*</strong>\s*<span>([\d,\.]+)</span>', webpage, 'view count', fatal=False)) - flash_vars = compat_parse_qs(self._search_regex( - r'<param[^>]+?name="flashvars"[^>]+?value="([^"]+)"', webpage, 'flash vars')) + flash_vars = self._parse_json( + self._search_regex( + r'var\s+flashvars\s*=\s*({.+?});', webpage, 'flash vars'), + video_id) formats = [] - quality = qualities(['180p', '240p', '360p', '480p', '720p', '1080p']) - for k, vals in flash_vars.items(): - m = re.match(r'quality_(?P<quality>[0-9]+p)$', 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..fd854411b 100644 --- a/youtube_dl/extractor/facebook.py +++ b/youtube_dl/extractor/facebook.py @@ -10,12 +10,11 @@ from ..compat import ( compat_str, compat_urllib_error, compat_urllib_parse_unquote, - compat_urllib_request, ) from ..utils import ( ExtractorError, - int_or_none, limit_length, + sanitized_Request, urlencode_postdata, get_element_by_id, clean_html, @@ -74,7 +73,7 @@ class FacebookIE(InfoExtractor): if useremail is None: return - login_page_req = compat_urllib_request.Request(self._LOGIN_URL) + login_page_req = sanitized_Request(self._LOGIN_URL) login_page_req.add_header('Cookie', 'locale=en_US') login_page = self._download_webpage(login_page_req, None, note='Downloading login page', @@ -95,7 +94,7 @@ class FacebookIE(InfoExtractor): 'timezone': '-60', 'trynum': '1', } - request = compat_urllib_request.Request(self._LOGIN_URL, urlencode_postdata(login_form)) + request = sanitized_Request(self._LOGIN_URL, urlencode_postdata(login_form)) request.add_header('Content-Type', 'application/x-www-form-urlencoded') try: login_results = self._download_webpage(request, None, @@ -110,7 +109,7 @@ class FacebookIE(InfoExtractor): r'name="h"\s+(?:\w+="[^"]+"\s+)*?value="([^"]+)"', login_results, 'h'), 'name_action_selected': 'dont_save', } - check_req = compat_urllib_request.Request(self._CHECKPOINT_URL, urlencode_postdata(check_form)) + check_req = sanitized_Request(self._CHECKPOINT_URL, urlencode_postdata(check_form)) check_req.add_header('Content-Type', 'application/x-www-form-urlencoded') check_response = self._download_webpage(check_req, None, note='Confirming login') @@ -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/fc2.py b/youtube_dl/extractor/fc2.py index a406945e8..92e8c571f 100644 --- a/youtube_dl/extractor/fc2.py +++ b/youtube_dl/extractor/fc2.py @@ -12,6 +12,7 @@ from ..compat import ( from ..utils import ( encode_dict, ExtractorError, + sanitized_Request, ) @@ -57,7 +58,7 @@ class FC2IE(InfoExtractor): } login_data = compat_urllib_parse.urlencode(encode_dict(login_form_strs)).encode('utf-8') - request = compat_urllib_request.Request( + request = sanitized_Request( 'https://secure.id.fc2.com/index.php?mode=login&switch_language=en', login_data) login_results = self._download_webpage(request, None, note='Logging in', errnote='Unable to log in') @@ -66,7 +67,7 @@ class FC2IE(InfoExtractor): return False # this is also needed - login_redir = compat_urllib_request.Request('http://id.fc2.com/?mode=redirect&login=done') + login_redir = sanitized_Request('http://id.fc2.com/?mode=redirect&login=done') self._download_webpage( login_redir, None, note='Login redirect', errnote='Login redirect failed') 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/flickr.py b/youtube_dl/extractor/flickr.py index 2fe76d661..91cd46e76 100644 --- a/youtube_dl/extractor/flickr.py +++ b/youtube_dl/extractor/flickr.py @@ -3,10 +3,10 @@ from __future__ import unicode_literals import re from .common import InfoExtractor -from ..compat import compat_urllib_request from ..utils import ( ExtractorError, find_xpath_attr, + sanitized_Request, ) @@ -30,7 +30,7 @@ class FlickrIE(InfoExtractor): video_id = mobj.group('id') video_uploader_id = mobj.group('uploader_id') webpage_url = 'http://www.flickr.com/photos/' + video_uploader_id + '/' + video_id - req = compat_urllib_request.Request(webpage_url) + req = sanitized_Request(webpage_url) req.add_header( 'User-Agent', # it needs a more recent version diff --git a/youtube_dl/extractor/fourtube.py b/youtube_dl/extractor/fourtube.py index 3bb4f6239..fc4a5a0fb 100644 --- a/youtube_dl/extractor/fourtube.py +++ b/youtube_dl/extractor/fourtube.py @@ -3,12 +3,10 @@ from __future__ import unicode_literals import re from .common import InfoExtractor -from ..compat import ( - compat_urllib_request, -) from ..utils import ( parse_duration, parse_iso8601, + sanitized_Request, str_to_int, ) @@ -46,10 +44,10 @@ 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 +66,24 @@ 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)) @@ -82,7 +91,7 @@ class FourTubeIE(InfoExtractor): b'Content-Type': b'application/x-www-form-urlencoded', b'Origin': b'http://www.4tube.com', } - token_req = compat_urllib_request.Request(token_url, b'{}', headers) + token_req = sanitized_Request(token_url, b'{}', headers) tokens = self._download_json(token_req, video_id) formats = [{ 'url': tokens[format]['token'], 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/gdcvault.py b/youtube_dl/extractor/gdcvault.py index a6834db43..3befd3e7b 100644 --- a/youtube_dl/extractor/gdcvault.py +++ b/youtube_dl/extractor/gdcvault.py @@ -3,13 +3,11 @@ from __future__ import unicode_literals import re from .common import InfoExtractor -from ..compat import ( - compat_urllib_parse, - compat_urllib_request, -) +from ..compat import compat_urllib_parse from ..utils import ( remove_end, HEADRequest, + sanitized_Request, ) @@ -125,7 +123,7 @@ class GDCVaultIE(InfoExtractor): 'password': password, } - request = compat_urllib_request.Request(login_url, compat_urllib_parse.urlencode(login_form)) + request = sanitized_Request(login_url, compat_urllib_parse.urlencode(login_form)) request.add_header('Content-Type', 'application/x-www-form-urlencoded') self._download_webpage(request, display_id, 'Logging in') start_page = self._download_webpage(webpage_url, display_id, 'Getting authenticated video page') diff --git a/youtube_dl/extractor/generic.py b/youtube_dl/extractor/generic.py index ca5fbafb2..5075d131e 100644 --- a/youtube_dl/extractor/generic.py +++ b/youtube_dl/extractor/generic.py @@ -9,8 +9,8 @@ import sys from .common import InfoExtractor from .youtube import YoutubeIE from ..compat import ( + compat_etree_fromstring, compat_urllib_parse_unquote, - compat_urllib_request, compat_urlparse, compat_xml_parse_error, ) @@ -21,7 +21,7 @@ from ..utils import ( HEADRequest, is_html, orderedSet, - parse_xml, + sanitized_Request, smuggle_url, unescapeHTML, unified_strdate, @@ -30,7 +30,10 @@ from ..utils import ( url_basename, xpath_text, ) -from .brightcove import BrightcoveIE +from .brightcove import ( + BrightcoveLegacyIE, + BrightcoveNewIE, +) from .nbc import NBCSportsVPlayerIE from .ooyala import OoyalaIE from .rutv import RUTVIE @@ -141,6 +144,7 @@ class GenericIE(InfoExtractor): 'ext': 'mp4', 'title': 'Automatics, robotics and biocybernetics', 'description': 'md5:815fc1deb6b3a2bff99de2d5325be482', + 'upload_date': '20130627', 'formats': 'mincount:16', 'subtitles': 'mincount:1', }, @@ -274,7 +278,7 @@ class GenericIE(InfoExtractor): # it also tests brightcove videos that need to set the 'Referer' in the # http requests { - 'add_ie': ['Brightcove'], + 'add_ie': ['BrightcoveLegacy'], 'url': 'http://www.bfmtv.com/video/bfmbusiness/cours-bourse/cours-bourse-l-analyse-technique-154522/', 'info_dict': { 'id': '2765128793001', @@ -298,7 +302,7 @@ class GenericIE(InfoExtractor): 'uploader': 'thestar.com', 'description': 'Mississauga resident David Farmer is still out of power as a result of the ice storm a month ago. To keep the house warm, Farmer cuts wood from his property for a wood burning stove downstairs.', }, - 'add_ie': ['Brightcove'], + 'add_ie': ['BrightcoveLegacy'], }, { 'url': 'http://www.championat.com/video/football/v/87/87499.html', @@ -313,7 +317,7 @@ class GenericIE(InfoExtractor): }, { # https://github.com/rg3/youtube-dl/issues/3541 - 'add_ie': ['Brightcove'], + 'add_ie': ['BrightcoveLegacy'], 'url': 'http://www.kijk.nl/sbs6/leermijvrouwenkennen/videos/jqMiXKAYan2S/aflevering-1', 'info_dict': { 'id': '3866516442001', @@ -819,6 +823,19 @@ class GenericIE(InfoExtractor): 'title': 'Os Guinness // Is It Fools Talk? // Unbelievable? Conference 2014', }, }, + # Kaltura embed protected with referrer + { + 'url': 'http://www.disney.nl/disney-channel/filmpjes/achter-de-schermen#/videoId/violetta-achter-de-schermen-ruggero', + 'info_dict': { + 'id': '1_g4fbemnq', + 'ext': 'mp4', + 'title': 'Violetta - Achter De Schermen - Ruggero', + 'description': 'Achter de schermen met Ruggero', + 'timestamp': 1435133761, + 'upload_date': '20150624', + 'uploader_id': 'echojecka', + }, + }, # Eagle.Platform embed (generic URL) { 'url': 'http://lenta.ru/news/2015/03/06/navalny/', @@ -1030,6 +1047,31 @@ class GenericIE(InfoExtractor): 'ext': 'mp4', 'title': 'cinemasnob', }, + }, + # BrightcoveInPageEmbed embed + { + 'url': 'http://www.geekandsundry.com/tabletop-bonus-wils-final-thoughts-on-dread/', + 'info_dict': { + 'id': '4238694884001', + 'ext': 'flv', + 'title': 'Tabletop: Dread, Last Thoughts', + 'description': 'Tabletop: Dread, Last Thoughts', + 'duration': 51690, + }, + }, + # JWPlayer with M3U8 + { + 'url': 'http://ren.tv/novosti/2015-09-25/sluchaynyy-prohozhiy-poymal-avtougonshchika-v-murmanske-video', + 'info_dict': { + 'id': 'playlist', + 'ext': 'mp4', + 'title': 'Случайный прохожий поймал автоугонщика в Мурманске. ВИДЕО | РЕН ТВ', + 'uploader': 'ren.tv', + }, + 'params': { + # m3u8 downloads + 'skip_download': True, + } } ] @@ -1173,7 +1215,7 @@ class GenericIE(InfoExtractor): full_response = None if head_response is False: - request = compat_urllib_request.Request(url) + request = sanitized_Request(url) request.add_header('Accept-Encoding', '*') full_response = self._request_webpage(request, video_id) head_response = full_response @@ -1202,7 +1244,7 @@ class GenericIE(InfoExtractor): '%s on generic information extractor.' % ('Forcing' if force else 'Falling back')) if not full_response: - request = compat_urllib_request.Request(url) + request = sanitized_Request(url) # Some webservers may serve compressed content of rather big size (e.g. gzipped flac) # making it impossible to download only chunk of the file (yet we need only 512kB to # test whether it's HTML or not). According to youtube-dl default Accept-Encoding @@ -1237,7 +1279,7 @@ 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): @@ -1289,14 +1331,14 @@ class GenericIE(InfoExtractor): return self.playlist_result( urlrs, playlist_id=video_id, playlist_title=video_title) - # Look for BrightCove: - bc_urls = BrightcoveIE._extract_brightcove_urls(webpage) + # Look for Brightcove Legacy Studio embeds + bc_urls = BrightcoveLegacyIE._extract_brightcove_urls(webpage) if bc_urls: self.to_screen('Brightcove video detected.') entries = [{ '_type': 'url', 'url': smuggle_url(bc_url, {'Referer': url}), - 'ie_key': 'Brightcove' + 'ie_key': 'BrightcoveLegacy' } for bc_url in bc_urls] return { @@ -1306,6 +1348,11 @@ class GenericIE(InfoExtractor): 'entries': entries, } + # Look for Brightcove New Studio embeds + bc_urls = BrightcoveNewIE._extract_urls(webpage) + if bc_urls: + return _playlist_from_matches(bc_urls, ie='BrightcoveNew') + # Look for embedded rtl.nl player matches = re.findall( r'<iframe[^>]+?src="((?:https?:)?//(?:www\.)?rtl\.nl/system/videoplayer/[^"]+(?:video_)?embed[^"]+)"', @@ -1671,10 +1718,12 @@ 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') + return self.url_result(smuggle_url( + 'kaltura:%(partner_id)s:%(id)s' % mobj.groupdict(), + {'source_url': url}), 'Kaltura') # Look for Eagle.Platform embeds mobj = re.search( @@ -1719,7 +1768,7 @@ class GenericIE(InfoExtractor): # Look for UDN embeds mobj = re.search( - r'<iframe[^>]+src="(?P<url>%s)"' % UDNEmbedIE._VALID_URL, webpage) + r'<iframe[^>]+src="(?P<url>%s)"' % UDNEmbedIE._PROTOCOL_RELATIVE_VALID_URL, webpage) if mobj is not None: return self.url_result( compat_urlparse.urljoin(url, mobj.group('url')), 'UDNEmbed') @@ -1839,6 +1888,7 @@ class GenericIE(InfoExtractor): entries = [] for video_url in found: + video_url = video_url.replace('\\/', '/') video_url = compat_urlparse.urljoin(url, video_url) video_id = compat_urllib_parse_unquote(os.path.basename(video_url)) @@ -1850,25 +1900,24 @@ class GenericIE(InfoExtractor): # here's a fun little line of code for you: video_id = os.path.splitext(video_id)[0] + entry_info_dict = { + 'id': video_id, + 'uploader': video_uploader, + 'title': video_title, + 'age_limit': age_limit, + } + ext = determine_ext(video_url) if ext == 'smil': - entries.append({ - 'id': video_id, - 'formats': self._extract_smil_formats(video_url, video_id), - 'uploader': video_uploader, - 'title': video_title, - 'age_limit': age_limit, - }) + entry_info_dict['formats'] = self._extract_smil_formats(video_url, video_id) elif ext == 'xspf': return self.playlist_result(self._extract_xspf_playlist(video_url, video_id), video_id) + elif ext == 'm3u8': + entry_info_dict['formats'] = self._extract_m3u8_formats(video_url, video_id, ext='mp4') else: - entries.append({ - 'id': video_id, - 'url': video_url, - 'uploader': video_uploader, - 'title': video_title, - 'age_limit': age_limit, - }) + entry_info_dict['url'] = video_url + + entries.append(entry_info_dict) if len(entries) == 1: return entries[0] 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/hearthisat.py b/youtube_dl/extractor/hearthisat.py index a19b31ac0..7d8698655 100644 --- a/youtube_dl/extractor/hearthisat.py +++ b/youtube_dl/extractor/hearthisat.py @@ -4,12 +4,10 @@ from __future__ import unicode_literals import re from .common import InfoExtractor -from ..compat import ( - compat_urllib_request, - compat_urlparse, -) +from ..compat import compat_urlparse from ..utils import ( HEADRequest, + sanitized_Request, str_to_int, urlencode_postdata, urlhandle_detect_ext, @@ -47,7 +45,7 @@ class HearThisAtIE(InfoExtractor): r'intTrackId\s*=\s*(\d+)', webpage, 'track ID') payload = urlencode_postdata({'tracks[]': track_id}) - req = compat_urllib_request.Request(self._PLAYLIST_URL, payload) + req = sanitized_Request(self._PLAYLIST_URL, payload) req.add_header('Content-type', 'application/x-www-form-urlencoded') track = self._download_json(req, track_id, 'Downloading playlist')[0] diff --git a/youtube_dl/extractor/hotnewhiphop.py b/youtube_dl/extractor/hotnewhiphop.py index 651784b73..31e219945 100644 --- a/youtube_dl/extractor/hotnewhiphop.py +++ b/youtube_dl/extractor/hotnewhiphop.py @@ -3,13 +3,11 @@ from __future__ import unicode_literals import base64 from .common import InfoExtractor -from ..compat import ( - compat_urllib_parse, - compat_urllib_request, -) +from ..compat import compat_urllib_parse from ..utils import ( ExtractorError, HEADRequest, + sanitized_Request, ) @@ -41,7 +39,7 @@ class HotNewHipHopIE(InfoExtractor): ('mediaType', 's'), ('mediaId', video_id), ]) - r = compat_urllib_request.Request( + r = sanitized_Request( 'http://www.hotnewhiphop.com/ajax/media/getActions/', data=reqdata) r.add_header('Content-Type', 'application/x-www-form-urlencoded') mkd = self._download_json( diff --git a/youtube_dl/extractor/hypem.py b/youtube_dl/extractor/hypem.py index aa0724a02..cca3dd498 100644 --- a/youtube_dl/extractor/hypem.py +++ b/youtube_dl/extractor/hypem.py @@ -4,12 +4,10 @@ import json import time from .common import InfoExtractor -from ..compat import ( - compat_urllib_parse, - compat_urllib_request, -) +from ..compat import compat_urllib_parse from ..utils import ( ExtractorError, + sanitized_Request, ) @@ -32,7 +30,7 @@ class HypemIE(InfoExtractor): data = {'ax': 1, 'ts': time.time()} data_encoded = compat_urllib_parse.urlencode(data) complete_url = url + "?" + data_encoded - request = compat_urllib_request.Request(complete_url) + request = sanitized_Request(complete_url) response, urlh = self._download_webpage_handle( request, track_id, 'Downloading webpage with the url') cookie = urlh.headers.get('Set-Cookie', '') @@ -52,7 +50,7 @@ class HypemIE(InfoExtractor): title = track['song'] serve_url = "http://hypem.com/serve/source/%s/%s" % (track_id, key) - request = compat_urllib_request.Request( + request = sanitized_Request( serve_url, '', {'Content-Type': 'application/json'}) request.add_header('cookie', cookie) song_data = self._download_json(request, track_id, 'Downloading metadata') 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/instagram.py b/youtube_dl/extractor/instagram.py index 3d78f78c4..c158f2064 100644 --- a/youtube_dl/extractor/instagram.py +++ b/youtube_dl/extractor/instagram.py @@ -10,8 +10,8 @@ from ..utils import ( class InstagramIE(InfoExtractor): - _VALID_URL = r'https://instagram\.com/p/(?P<id>[\da-zA-Z]+)' - _TEST = { + _VALID_URL = r'https?://(?:www\.)?instagram\.com/p/(?P<id>[^/?#&]+)' + _TESTS = [{ 'url': 'https://instagram.com/p/aye83DjauH/?foo=bar#abc', 'md5': '0d2da106a9d2631273e192b372806516', 'info_dict': { @@ -21,7 +21,10 @@ class InstagramIE(InfoExtractor): 'title': 'Video by naomipq', 'description': 'md5:1f17f0ab29bd6fe2bfad705f58de3cb8', } - } + }, { + 'url': 'https://instagram.com/p/-Cmh1cukG2/', + 'only_matching': True, + }] def _real_extract(self, url): video_id = self._match_id(url) diff --git a/youtube_dl/extractor/iprima.py b/youtube_dl/extractor/iprima.py index 821c8ec10..36baf3245 100644 --- a/youtube_dl/extractor/iprima.py +++ b/youtube_dl/extractor/iprima.py @@ -6,12 +6,10 @@ from random import random from math import floor from .common import InfoExtractor -from ..compat import ( - compat_urllib_request, -) from ..utils import ( ExtractorError, remove_end, + sanitized_Request, ) @@ -61,7 +59,7 @@ class IPrimaIE(InfoExtractor): (floor(random() * 1073741824), floor(random() * 1073741824)) ) - req = compat_urllib_request.Request(player_url) + req = sanitized_Request(player_url) req.add_header('Referer', url) playerpage = self._download_webpage(req, video_id) diff --git a/youtube_dl/extractor/iqiyi.py b/youtube_dl/extractor/iqiyi.py index ce1ab3820..2df1da3f0 100644 --- a/youtube_dl/extractor/iqiyi.py +++ b/youtube_dl/extractor/iqiyi.py @@ -191,7 +191,7 @@ class IqiyiIE(InfoExtractor): 'vid': video_id, 'vinfo': 1, 'tm': tm, - 'enc': self.md5_text((enc_key + tail)[1:64:2] + tail), + 'enc': self.md5_text(enc_key + tail), 'qyid': _uuid, 'tn': random.random(), 'um': 0, @@ -205,7 +205,9 @@ class IqiyiIE(InfoExtractor): def get_enc_key(self, swf_url, video_id): # TODO: automatic key extraction - enc_key = 'eac64f22daf001da6ba9aa8da4d501508bbe90a4d4091fea3b0582a85b38c2cc' # last update at 2015-09-23-23 for Zombie::bite + # 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/ivi.py b/youtube_dl/extractor/ivi.py index e82594444..029878d24 100644 --- a/youtube_dl/extractor/ivi.py +++ b/youtube_dl/extractor/ivi.py @@ -5,11 +5,9 @@ import re import json from .common import InfoExtractor -from ..compat import ( - compat_urllib_request, -) from ..utils import ( ExtractorError, + sanitized_Request, ) @@ -78,7 +76,7 @@ class IviIE(InfoExtractor): ] } - request = compat_urllib_request.Request(api_url, json.dumps(data)) + request = sanitized_Request(api_url, json.dumps(data)) video_json_page = self._download_webpage( request, video_id, 'Downloading video JSON') 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..583b1a5ad 100644 --- a/youtube_dl/extractor/kaltura.py +++ b/youtube_dl/extractor/kaltura.py @@ -2,12 +2,18 @@ from __future__ import unicode_literals import re +import base64 from .common import InfoExtractor -from ..compat import compat_urllib_parse +from ..compat import ( + compat_urllib_parse, + compat_urlparse, +) from ..utils import ( + clean_html, ExtractorError, int_or_none, + unsmuggle_url, ) @@ -16,7 +22,7 @@ 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 @@ -121,31 +127,47 @@ class KalturaIE(InfoExtractor): video_id, actions, note='Downloading video info JSON') def _real_extract(self, url): + url, smuggled_data = unsmuggle_url(url, {}) + mobj = re.match(self._VALID_URL, url) partner_id = mobj.group('partner_id_s') or mobj.group('partner_id') or mobj.group('partner_id_html5') entry_id = mobj.group('id_s') or mobj.group('id') or mobj.group('id_html5') info, source_data = self._get_video_info(entry_id, partner_id) - formats = [{ - 'format_id': '%(fileExt)s-%(bitrate)s' % f, - 'ext': f['fileExt'], - 'tbr': f['bitrate'], - 'fps': f.get('frameRate'), - 'filesize_approx': int_or_none(f.get('size'), invscale=1024), - 'container': f.get('containerFormat'), - 'vcodec': f.get('videoCodecId'), - 'height': f.get('height'), - 'width': f.get('width'), - 'url': '%s/flavorId/%s' % (info['dataUrl'], f['id']), - } for f in source_data['flavorAssets']] + source_url = smuggled_data.get('source_url') + if source_url: + referrer = base64.b64encode( + '://'.join(compat_urlparse.urlparse(source_url)[:2]) + .encode('utf-8')).decode('utf-8') + else: + referrer = None + + formats = [] + for f in source_data['flavorAssets']: + video_url = '%s/flavorId/%s' % (info['dataUrl'], f['id']) + if referrer: + video_url += '?referrer=%s' % referrer + formats.append({ + 'format_id': '%(fileExt)s-%(bitrate)s' % f, + 'ext': f.get('fileExt'), + 'tbr': int_or_none(f['bitrate']), + 'fps': int_or_none(f.get('frameRate')), + 'filesize_approx': int_or_none(f.get('size'), invscale=1024), + 'container': f.get('containerFormat'), + 'vcodec': f.get('videoCodecId'), + 'height': int_or_none(f.get('height')), + 'width': int_or_none(f.get('width')), + 'url': video_url, + }) + self._check_formats(formats, entry_id) self._sort_formats(formats) return { 'id': entry_id, 'title': info['name'], 'formats': formats, - 'description': info.get('description'), + 'description': clean_html(info.get('description')), 'thumbnail': info.get('thumbnailUrl'), 'duration': info.get('duration'), 'timestamp': info.get('createdAt'), diff --git a/youtube_dl/extractor/keezmovies.py b/youtube_dl/extractor/keezmovies.py index 82eddec51..d79261bb5 100644 --- a/youtube_dl/extractor/keezmovies.py +++ b/youtube_dl/extractor/keezmovies.py @@ -4,10 +4,8 @@ import os import re from .common import InfoExtractor -from ..compat import ( - compat_urllib_parse_urlparse, - compat_urllib_request, -) +from ..compat import compat_urllib_parse_urlparse +from ..utils import sanitized_Request class KeezMoviesIE(InfoExtractor): @@ -26,7 +24,7 @@ class KeezMoviesIE(InfoExtractor): def _real_extract(self, url): video_id = self._match_id(url) - req = compat_urllib_request.Request(url) + req = sanitized_Request(url) req.add_header('Cookie', 'age_verified=1') webpage = self._download_webpage(req, video_id) diff --git a/youtube_dl/extractor/letv.py b/youtube_dl/extractor/letv.py index a28abb0f0..be648000e 100644 --- a/youtube_dl/extractor/letv.py +++ b/youtube_dl/extractor/letv.py @@ -8,14 +8,15 @@ import time 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, + sanitized_Request, 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) @@ -85,7 +114,7 @@ class LetvIE(InfoExtractor): 'tkey': self.calc_time_key(int(time.time())), 'domain': 'www.letv.com' } - play_json_req = compat_urllib_request.Request( + play_json_req = sanitized_Request( 'http://api.letv.com/mms/out/video/playJson?' + compat_urllib_parse.urlencode(params) ) cn_verification_proxy = self._downloader.params.get('cn_verification_proxy') @@ -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..d4e1ae99d 100644 --- a/youtube_dl/extractor/lynda.py +++ b/youtube_dl/extractor/lynda.py @@ -7,12 +7,12 @@ from .common import InfoExtractor from ..compat import ( compat_str, compat_urllib_parse, - compat_urllib_request, ) from ..utils import ( ExtractorError, clean_html, int_or_none, + sanitized_Request, ) @@ -25,7 +25,7 @@ class LyndaBaseIE(InfoExtractor): self._login() def _login(self): - (username, password) = self._get_login_info() + username, password = self._get_login_info() if username is None: return @@ -35,7 +35,7 @@ class LyndaBaseIE(InfoExtractor): 'remember': 'false', 'stayPut': 'false' } - request = compat_urllib_request.Request( + request = sanitized_Request( self._LOGIN_URL, compat_urllib_parse.urlencode(login_form).encode('utf-8')) login_page = self._download_webpage( request, None, 'Logging in as %s' % username) @@ -64,7 +64,7 @@ class LyndaBaseIE(InfoExtractor): 'remember': 'false', 'stayPut': 'false', } - request = compat_urllib_request.Request( + request = sanitized_Request( self._LOGIN_URL, compat_urllib_parse.urlencode(confirm_form).encode('utf-8')) login_page = self._download_webpage( request, None, @@ -82,6 +82,15 @@ class LyndaBaseIE(InfoExtractor): expected=True) raise ExtractorError('Unable to log in') + def _logout(self): + username, _ = self._get_login_info() + if username is None: + return + + 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 +117,47 @@ 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 +188,7 @@ 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 +210,13 @@ 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 +226,13 @@ 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 +245,6 @@ 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/metacafe.py b/youtube_dl/extractor/metacafe.py index 6e2e73a51..3c786a36d 100644 --- a/youtube_dl/extractor/metacafe.py +++ b/youtube_dl/extractor/metacafe.py @@ -7,12 +7,12 @@ from ..compat import ( compat_parse_qs, compat_urllib_parse, compat_urllib_parse_unquote, - compat_urllib_request, ) from ..utils import ( determine_ext, ExtractorError, int_or_none, + sanitized_Request, ) @@ -117,7 +117,7 @@ class MetacafeIE(InfoExtractor): 'filters': '0', 'submit': "Continue - I'm over 18", } - request = compat_urllib_request.Request(self._FILTER_POST, compat_urllib_parse.urlencode(disclaimer_form)) + request = sanitized_Request(self._FILTER_POST, compat_urllib_parse.urlencode(disclaimer_form)) request.add_header('Content-Type', 'application/x-www-form-urlencoded') self.report_age_confirmation() self._download_webpage(request, None, False, 'Unable to confirm age') @@ -142,7 +142,7 @@ class MetacafeIE(InfoExtractor): return self.url_result('theplatform:%s' % ext_id, 'ThePlatform') # Retrieve video webpage to extract further information - req = compat_urllib_request.Request('http://www.metacafe.com/watch/%s/' % video_id) + req = sanitized_Request('http://www.metacafe.com/watch/%s/' % video_id) # AnyClip videos require the flashversion cookie so that we get the link # to the mp4 file diff --git a/youtube_dl/extractor/minhateca.py b/youtube_dl/extractor/minhateca.py index 14934b7ec..e46b23a6f 100644 --- a/youtube_dl/extractor/minhateca.py +++ b/youtube_dl/extractor/minhateca.py @@ -2,14 +2,12 @@ from __future__ import unicode_literals from .common import InfoExtractor -from ..compat import ( - compat_urllib_parse, - compat_urllib_request, -) +from ..compat import compat_urllib_parse from ..utils import ( int_or_none, parse_duration, parse_filesize, + sanitized_Request, ) @@ -39,7 +37,7 @@ class MinhatecaIE(InfoExtractor): ('fileId', video_id), ('__RequestVerificationToken', token), ] - req = compat_urllib_request.Request( + req = sanitized_Request( 'http://minhateca.com.br/action/License/Download', data=compat_urllib_parse.urlencode(token_data)) req.add_header('Content-Type', 'application/x-www-form-urlencoded') diff --git a/youtube_dl/extractor/miomio.py b/youtube_dl/extractor/miomio.py index a784fc5fb..170ebd9eb 100644 --- a/youtube_dl/extractor/miomio.py +++ b/youtube_dl/extractor/miomio.py @@ -8,6 +8,7 @@ from ..utils import ( xpath_text, int_or_none, ExtractorError, + sanitized_Request, ) @@ -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 = sanitized_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 54993e2c9..c595f2077 100644 --- a/youtube_dl/extractor/mitele.py +++ b/youtube_dl/extractor/mitele.py @@ -1,7 +1,10 @@ from __future__ import unicode_literals from .common import InfoExtractor -from ..compat import compat_urllib_parse +from ..compat import ( + compat_urllib_parse, + compat_urlparse, +) from ..utils import ( encode_dict, get_element_by_attribute, @@ -15,7 +18,7 @@ class MiTeleIE(InfoExtractor): _TESTS = [{ 'url': 'http://www.mitele.es/programas-tv/diario-de/la-redaccion/programa-144/', - 'md5': 'ace7635b2a0b286aaa37d3ff192d2a8a', + 'md5': '0ff1a13aebb35d9bc14081ff633dd324', 'info_dict': { 'id': '0NF1jJnxS1Wu3pHrmvFyw2', 'display_id': 'programa-144', @@ -34,6 +37,7 @@ class MiTeleIE(InfoExtractor): config_url = self._search_regex( r'data-config\s*=\s*"([^"]+)"', webpage, 'data config url') + config_url = compat_urlparse.urljoin(url, config_url) config = self._download_json( config_url, display_id, 'Downloading config JSON') @@ -56,7 +60,7 @@ class MiTeleIE(InfoExtractor): 'sta': '0', } media = self._download_json( - '%s/?%s' % (gat, compat_urllib_parse.urlencode(encode_dict(token_data)).encode('utf-8')), + '%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_: diff --git a/youtube_dl/extractor/moevideo.py b/youtube_dl/extractor/moevideo.py index 5a66302f6..d930b9634 100644 --- a/youtube_dl/extractor/moevideo.py +++ b/youtube_dl/extractor/moevideo.py @@ -5,13 +5,11 @@ import json import re from .common import InfoExtractor -from ..compat import ( - compat_urllib_parse, - compat_urllib_request, -) +from ..compat import compat_urllib_parse from ..utils import ( ExtractorError, int_or_none, + sanitized_Request, ) @@ -80,7 +78,7 @@ class MoeVideoIE(InfoExtractor): ] r_json = json.dumps(r) post = compat_urllib_parse.urlencode({'r': r_json}) - req = compat_urllib_request.Request(self._API_URL, post) + req = sanitized_Request(self._API_URL, post) req.add_header('Content-type', 'application/x-www-form-urlencoded') response = self._download_json(req, video_id) diff --git a/youtube_dl/extractor/mofosex.py b/youtube_dl/extractor/mofosex.py index 9bf99a54a..f8226cbb2 100644 --- a/youtube_dl/extractor/mofosex.py +++ b/youtube_dl/extractor/mofosex.py @@ -7,8 +7,8 @@ from .common import InfoExtractor from ..compat import ( compat_urllib_parse_unquote, compat_urllib_parse_urlparse, - compat_urllib_request, ) +from ..utils import sanitized_Request class MofosexIE(InfoExtractor): @@ -29,7 +29,7 @@ class MofosexIE(InfoExtractor): video_id = mobj.group('id') url = 'http://www.' + mobj.group('url') - req = compat_urllib_request.Request(url) + req = sanitized_Request(url) req.add_header('Cookie', 'age_verified=1') webpage = self._download_webpage(req, video_id) diff --git a/youtube_dl/extractor/moniker.py b/youtube_dl/extractor/moniker.py index 69e4bcd1a..f6bf94f2f 100644 --- a/youtube_dl/extractor/moniker.py +++ b/youtube_dl/extractor/moniker.py @@ -5,19 +5,17 @@ import os.path import re from .common import InfoExtractor -from ..compat import ( - compat_urllib_parse, - compat_urllib_request, -) +from ..compat import compat_urllib_parse from ..utils import ( ExtractorError, remove_start, + sanitized_Request, ) 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 +44,18 @@ 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 +74,30 @@ 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 = sanitized_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 = sanitized_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 +111,6 @@ class MonikerIE(InfoExtractor): return { 'id': video_id, 'title': title, + 'description': description, 'formats': formats, } diff --git a/youtube_dl/extractor/mooshare.py b/youtube_dl/extractor/mooshare.py index 7603af5e2..7cc7f054f 100644 --- a/youtube_dl/extractor/mooshare.py +++ b/youtube_dl/extractor/mooshare.py @@ -3,12 +3,10 @@ from __future__ import unicode_literals import re from .common import InfoExtractor -from ..compat import ( - compat_urllib_request, - compat_urllib_parse, -) +from ..compat import compat_urllib_parse from ..utils import ( ExtractorError, + sanitized_Request, ) @@ -59,7 +57,7 @@ class MooshareIE(InfoExtractor): 'hash': hash_key, } - request = compat_urllib_request.Request( + request = sanitized_Request( 'http://mooshare.biz/%s' % video_id, compat_urllib_parse.urlencode(download_form)) request.add_header('Content-Type', 'application/x-www-form-urlencoded') diff --git a/youtube_dl/extractor/movieclips.py b/youtube_dl/extractor/movieclips.py index 04e17d055..1564cb71f 100644 --- a/youtube_dl/extractor/movieclips.py +++ b/youtube_dl/extractor/movieclips.py @@ -1,80 +1,40 @@ +# 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, -) +from ..utils import sanitized_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 = sanitized_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 302c9bf35..d887583e6 100644 --- a/youtube_dl/extractor/mtv.py +++ b/youtube_dl/extractor/mtv.py @@ -5,7 +5,6 @@ import re from .common import InfoExtractor from ..compat import ( compat_urllib_parse, - compat_urllib_request, compat_str, ) from ..utils import ( @@ -13,6 +12,7 @@ from ..utils import ( find_xpath_attr, fix_xml_ampersands, HEADRequest, + sanitized_Request, unescapeHTML, url_basename, RegexNotFoundError, @@ -53,7 +53,7 @@ class MTVServicesInfoExtractor(InfoExtractor): def _extract_mobile_video_formats(self, mtvn_id): webpage_url = self._MOBILE_TEMPLATE % mtvn_id - req = compat_urllib_request.Request(webpage_url) + req = sanitized_Request(webpage_url) # Otherwise we get a webpage that would execute some javascript req.add_header('User-Agent', 'curl/7') webpage = self._download_webpage(req, mtvn_id, diff --git a/youtube_dl/extractor/myvideo.py b/youtube_dl/extractor/myvideo.py index c96f472a3..36ab388b2 100644 --- a/youtube_dl/extractor/myvideo.py +++ b/youtube_dl/extractor/myvideo.py @@ -11,10 +11,10 @@ from ..compat import ( compat_ord, compat_urllib_parse, compat_urllib_parse_unquote, - compat_urllib_request, ) from ..utils import ( ExtractorError, + sanitized_Request, ) @@ -83,7 +83,7 @@ class MyVideoIE(InfoExtractor): mobj = re.search(r'data-video-service="/service/data/video/%s/config' % video_id, webpage) if mobj is not None: - request = compat_urllib_request.Request('http://www.myvideo.de/service/data/video/%s/config' % video_id, '') + request = sanitized_Request('http://www.myvideo.de/service/data/video/%s/config' % video_id, '') response = self._download_webpage(request, video_id, 'Downloading video info') info = json.loads(base64.b64decode(response).decode('utf-8')) diff --git a/youtube_dl/extractor/ndr.py b/youtube_dl/extractor/ndr.py index e3cc6fde8..16213eed9 100644 --- a/youtube_dl/extractor/ndr.py +++ b/youtube_dl/extractor/ndr.py @@ -14,7 +14,8 @@ from ..utils import ( class NDRBaseIE(InfoExtractor): def _real_extract(self, url): - display_id = self._match_id(url) + mobj = re.match(self._VALID_URL, url) + 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) @@ -22,7 +23,7 @@ class NDRBaseIE(InfoExtractor): class NDRIE(NDRBaseIE): IE_NAME = 'ndr' IE_DESC = 'NDR.de - Norddeutscher Rundfunk' - _VALID_URL = r'https?://www\.ndr\.de/(?:[^/]+/)+(?P<id>[^/?#]+),[\da-z]+\.html' + _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', @@ -77,6 +78,9 @@ class NDRIE(NDRBaseIE): '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): @@ -101,7 +105,7 @@ class NDRIE(NDRBaseIE): class NJoyIE(NDRBaseIE): IE_NAME = 'njoy' IE_DESC = 'N-JOY' - _VALID_URL = r'https?://www\.n-joy\.de/(?:[^/]+/)+(?P<id>[^/?#]+),[\da-z]+\.html' + _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', @@ -136,6 +140,9 @@ class NJoyIE(NDRBaseIE): 'params': { 'skip_download': True, }, + }, { + 'url': 'http://www.n-joy.de/radio/webradio/morningshow209.html', + 'only_matching': True, }] def _extract_embed(self, webpage, display_id): @@ -231,7 +238,7 @@ class NDREmbedBaseIE(InfoExtractor): class NDREmbedIE(NDREmbedBaseIE): IE_NAME = 'ndr:embed' - _VALID_URL = r'https?://www\.ndr\.de/(?:[^/]+/)+(?P<id>[\da-z]+)-(?:player|externalPlayer)\.html' + _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', @@ -325,7 +332,7 @@ class NDREmbedIE(NDREmbedBaseIE): class NJoyEmbedIE(NDREmbedBaseIE): IE_NAME = 'njoy:embed' - _VALID_URL = r'https?://www\.n-joy\.de/(?:[^/]+/)+(?P<id>[\da-z]+)-(?:player|externalPlayer)_[^/]+\.html' + _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', diff --git a/youtube_dl/extractor/neteasemusic.py b/youtube_dl/extractor/neteasemusic.py index a8e0a64ed..15eca825a 100644 --- a/youtube_dl/extractor/neteasemusic.py +++ b/youtube_dl/extractor/neteasemusic.py @@ -8,11 +8,11 @@ import re from .common import InfoExtractor from ..compat import ( - compat_urllib_request, compat_urllib_parse, compat_str, compat_itertools_count, ) +from ..utils import sanitized_Request class NetEaseMusicBaseIE(InfoExtractor): @@ -40,7 +40,7 @@ class NetEaseMusicBaseIE(InfoExtractor): if not details: continue formats.append({ - 'url': 'http://m1.music.126.net/%s/%s.%s' % + 'url': 'http://m5.music.126.net/%s/%s.%s' % (cls._encrypt(details['dfsId']), details['dfsId'], details['extension']), 'ext': details.get('extension'), @@ -56,7 +56,7 @@ class NetEaseMusicBaseIE(InfoExtractor): return int(round(ms / 1000.0)) def query_api(self, endpoint, video_id, note): - req = compat_urllib_request.Request('%s%s' % (self._API_BASE, endpoint)) + req = sanitized_Request('%s%s' % (self._API_BASE, endpoint)) req.add_header('Referer', self._API_BASE) return self._download_json(req, video_id, note) 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/nfb.py b/youtube_dl/extractor/nfb.py index ea077254b..5bd15f7a7 100644 --- a/youtube_dl/extractor/nfb.py +++ b/youtube_dl/extractor/nfb.py @@ -1,10 +1,8 @@ from __future__ import unicode_literals from .common import InfoExtractor -from ..compat import ( - compat_urllib_request, - compat_urllib_parse, -) +from ..compat import compat_urllib_parse +from ..utils import sanitized_Request class NFBIE(InfoExtractor): @@ -40,8 +38,9 @@ class NFBIE(InfoExtractor): uploader = self._html_search_regex(r'<em class="director-name" itemprop="name">([^<]+)</em>', page, 'director name', fatal=False) - request = compat_urllib_request.Request('https://www.nfb.ca/film/%s/player_config' % video_id, - compat_urllib_parse.urlencode({'getConfig': 'true'}).encode('ascii')) + request = sanitized_Request( + 'https://www.nfb.ca/film/%s/player_config' % video_id, + compat_urllib_parse.urlencode({'getConfig': 'true'}).encode('ascii')) request.add_header('Content-Type', 'application/x-www-form-urlencoded') request.add_header('X-NFB-Referer', 'http://www.nfb.ca/medias/flash/NFBVideoPlayer.swf') diff --git a/youtube_dl/extractor/niconico.py b/youtube_dl/extractor/niconico.py index bda1cff05..586e52a4a 100644 --- a/youtube_dl/extractor/niconico.py +++ b/youtube_dl/extractor/niconico.py @@ -8,7 +8,6 @@ import datetime from .common import InfoExtractor from ..compat import ( compat_urllib_parse, - compat_urllib_request, compat_urlparse, ) from ..utils import ( @@ -17,6 +16,7 @@ from ..utils import ( int_or_none, parse_duration, parse_iso8601, + sanitized_Request, xpath_text, determine_ext, ) @@ -102,7 +102,7 @@ class NiconicoIE(InfoExtractor): 'password': password, } login_data = compat_urllib_parse.urlencode(encode_dict(login_form_strs)).encode('utf-8') - request = compat_urllib_request.Request( + request = sanitized_Request( 'https://secure.nicovideo.jp/secure/login', login_data) login_results = self._download_webpage( request, None, note='Logging in', errnote='Unable to log in') @@ -145,7 +145,7 @@ class NiconicoIE(InfoExtractor): 'k': thumb_play_key, 'v': video_id }) - flv_info_request = compat_urllib_request.Request( + flv_info_request = sanitized_Request( 'http://ext.nicovideo.jp/thumb_watch', flv_info_data, {'Content-Type': 'application/x-www-form-urlencoded'}) flv_info_webpage = self._download_webpage( diff --git a/youtube_dl/extractor/noco.py b/youtube_dl/extractor/noco.py index a53e27b27..76bd21e6d 100644 --- a/youtube_dl/extractor/noco.py +++ b/youtube_dl/extractor/noco.py @@ -9,7 +9,6 @@ from .common import InfoExtractor from ..compat import ( compat_str, compat_urllib_parse, - compat_urllib_request, ) from ..utils import ( clean_html, @@ -17,6 +16,7 @@ from ..utils import ( int_or_none, float_or_none, parse_iso8601, + sanitized_Request, ) @@ -74,7 +74,7 @@ class NocoIE(InfoExtractor): 'username': username, 'password': password, } - request = compat_urllib_request.Request(self._LOGIN_URL, compat_urllib_parse.urlencode(login_form)) + request = sanitized_Request(self._LOGIN_URL, compat_urllib_parse.urlencode(login_form)) request.add_header('Content-Type', 'application/x-www-form-urlencoded; charset=UTF-8') login = self._download_json(request, None, 'Logging in as %s' % username) diff --git a/youtube_dl/extractor/nosvideo.py b/youtube_dl/extractor/nosvideo.py index f5ef856db..eab816e49 100644 --- a/youtube_dl/extractor/nosvideo.py +++ b/youtube_dl/extractor/nosvideo.py @@ -4,11 +4,9 @@ from __future__ import unicode_literals import re from .common import InfoExtractor -from ..compat import ( - compat_urllib_request, -) from ..utils import ( ExtractorError, + sanitized_Request, urlencode_postdata, xpath_text, xpath_with_ns, @@ -41,7 +39,7 @@ class NosVideoIE(InfoExtractor): 'op': 'download1', 'method_free': 'Continue to Video', } - req = compat_urllib_request.Request(url, urlencode_postdata(fields)) + req = sanitized_Request(url, urlencode_postdata(fields)) req.add_header('Content-type', 'application/x-www-form-urlencoded') webpage = self._download_webpage(req, video_id, 'Downloading download page') diff --git a/youtube_dl/extractor/novamov.py b/youtube_dl/extractor/novamov.py index 04d779890..6163e8855 100644 --- a/youtube_dl/extractor/novamov.py +++ b/youtube_dl/extractor/novamov.py @@ -3,11 +3,13 @@ from __future__ import unicode_literals import re from .common import InfoExtractor -from ..compat import ( - compat_urlparse, -) +from ..compat import compat_urlparse from ..utils import ( ExtractorError, + NO_DEFAULT, + encode_dict, + sanitized_Request, + urlencode_postdata, ) @@ -38,19 +40,40 @@ 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 = sanitized_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 b97f62fdb..d480fb58c 100644 --- a/youtube_dl/extractor/nowness.py +++ b/youtube_dl/extractor/nowness.py @@ -1,12 +1,12 @@ # encoding: utf-8 from __future__ import unicode_literals -from .brightcove import BrightcoveIE +from .brightcove import BrightcoveLegacyIE from .common import InfoExtractor -from ..utils import ExtractorError -from ..compat import ( - compat_str, - compat_urllib_request, +from ..compat import compat_str +from ..utils import ( + ExtractorError, + sanitized_Request, ) @@ -22,10 +22,10 @@ class NownessBaseIE(InfoExtractor): '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) + bc_url = BrightcoveLegacyIE._extract_brightcove_url(player_code) if bc_url is None: raise ExtractorError('Could not find player definition') - return self.url_result(bc_url, 'Brightcove') + return self.url_result(bc_url, 'BrightcoveLegacy') elif source == 'vimeo': return self.url_result('http://vimeo.com/%s' % video_id, 'Vimeo') elif source == 'youtube': @@ -37,7 +37,7 @@ class NownessBaseIE(InfoExtractor): def _api_request(self, url, request_path): display_id = self._match_id(url) - request = compat_urllib_request.Request( + request = sanitized_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', diff --git a/youtube_dl/extractor/nowtv.py b/youtube_dl/extractor/nowtv.py index c8257719f..67e34b294 100644 --- a/youtube_dl/extractor/nowtv.py +++ b/youtube_dl/extractor/nowtv.py @@ -1,6 +1,8 @@ # coding: utf-8 from __future__ import unicode_literals +import re + from .common import InfoExtractor from ..compat import compat_str from ..utils import ( @@ -13,8 +15,63 @@ from ..utils import ( ) -class NowTVIE(InfoExtractor): - _VALID_URL = r'https?://(?:www\.)?nowtv\.(?:de|at|ch)/(?:rtl|rtl2|rtlnitro|superrtl|ntv|vox)/(?P<id>.+?)/(?:player|preview)' +class NowTVBaseIE(InfoExtractor): + _VIDEO_FIELDS = ( + 'id', 'title', 'free', 'geoblocked', 'articleLong', 'articleShort', + 'broadcastStartDate', 'seoUrl', 'duration', 'files', + 'format.defaultImage169Format', 'format.defaultImage169Logo') + + def _extract_video(self, info, display_id=None): + video_id = compat_str(info['id']) + + files = info['files'] + if not files: + if info.get('geoblocked', False): + raise ExtractorError( + 'Video %s is not available from your location due to geo restriction' % video_id, + expected=True) + if not info.get('free', True): + raise ExtractorError( + 'Video %s is not available for free' % video_id, expected=True) + + formats = [] + for item in files['items']: + if determine_ext(item['path']) != 'f4v': + continue + app, play_path = remove_start(item['path'], '/').split('/', 1) + formats.append({ + 'url': 'rtmpe://fms.rtl.de', + 'app': app, + 'play_path': 'mp4:%s' % play_path, + 'ext': 'flv', + '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) + + title = info['title'] + description = info.get('articleLong') or info.get('articleShort') + timestamp = parse_iso8601(info.get('broadcastStartDate'), ' ') + duration = parse_duration(info.get('duration')) + + f = info.get('format', {}) + thumbnail = f.get('defaultImage169Format') or f.get('defaultImage169Logo') + + return { + 'id': video_id, + 'display_id': display_id or info.get('seoUrl'), + 'title': title, + 'description': description, + 'thumbnail': thumbnail, + 'timestamp': timestamp, + 'duration': duration, + 'formats': formats, + } + + +class NowTVIE(NowTVBaseIE): + _VALID_URL = r'https?://(?:www\.)?nowtv\.(?:de|at|ch)/(?:rtl|rtl2|rtlnitro|superrtl|ntv|vox)/(?P<show_id>[^/]+)/(?:list/[^/]+/)?(?P<id>[^/]+)/(?:player|preview)' _TESTS = [{ # rtl @@ -23,7 +80,7 @@ class NowTVIE(InfoExtractor): 'id': '203519', 'display_id': 'bauer-sucht-frau/die-neuen-bauern-und-eine-hochzeit', 'ext': 'flv', - 'title': 'Die neuen Bauern und eine Hochzeit', + 'title': 'Inka Bause stellt die neuen Bauern vor', 'description': 'md5:e234e1ed6d63cf06be5c070442612e7e', 'thumbnail': 're:^https?://.*\.jpg$', 'timestamp': 1432580700, @@ -136,58 +193,65 @@ class NowTVIE(InfoExtractor): }] def _real_extract(self, url): - display_id = self._match_id(url) - display_id_split = display_id.split('/') - if len(display_id) > 2: - display_id = '/'.join((display_id_split[0], display_id_split[-1])) + mobj = re.match(self._VALID_URL, url) + display_id = '%s/%s' % (mobj.group('show_id'), mobj.group('id')) info = self._download_json( - 'https://api.nowtv.de/v3/movies/%s?fields=id,title,free,geoblocked,articleLong,articleShort,broadcastStartDate,seoUrl,duration,format,files' % display_id, - display_id) + 'https://api.nowtv.de/v3/movies/%s?fields=%s' + % (display_id, ','.join(self._VIDEO_FIELDS)), display_id) - video_id = compat_str(info['id']) + return self._extract_video(info, display_id) - files = info['files'] - if not files: - if info.get('geoblocked', False): - raise ExtractorError( - 'Video %s is not available from your location due to geo restriction' % video_id, - expected=True) - if not info.get('free', True): - raise ExtractorError( - 'Video %s is not available for free' % video_id, expected=True) - formats = [] - for item in files['items']: - if determine_ext(item['path']) != 'f4v': - continue - app, play_path = remove_start(item['path'], '/').split('/', 1) - formats.append({ - 'url': 'rtmpe://fms.rtl.de', - 'app': app, - 'play_path': 'mp4:%s' % play_path, - 'ext': 'flv', - 'page_url': url, - 'player_url': 'http://rtl-now.rtl.de/includes/nc_player.swf', - 'tbr': int_or_none(item.get('bitrate')), - }) - self._sort_formats(formats) +class NowTVListIE(NowTVBaseIE): + _VALID_URL = r'https?://(?:www\.)?nowtv\.(?:de|at|ch)/(?:rtl|rtl2|rtlnitro|superrtl|ntv|vox)/(?P<show_id>[^/]+)/list/(?P<id>[^?/#&]+)$' - title = info['title'] - description = info.get('articleLong') or info.get('articleShort') - timestamp = parse_iso8601(info.get('broadcastStartDate'), ' ') - duration = parse_duration(info.get('duration')) + _SHOW_FIELDS = ('title', ) + _SEASON_FIELDS = ('id', 'headline', 'seoheadline', ) - f = info.get('format', {}) - thumbnail = f.get('defaultImage169Format') or f.get('defaultImage169Logo') + _TESTS = [{ + 'url': 'http://www.nowtv.at/rtl/stern-tv/list/aktuell', + 'info_dict': { + 'id': '17006', + 'title': 'stern TV - Aktuell', + }, + 'playlist_count': 1, + }, { + 'url': 'http://www.nowtv.at/rtl/das-supertalent/list/free-staffel-8', + 'info_dict': { + 'id': '20716', + 'title': 'Das Supertalent - FREE Staffel 8', + }, + 'playlist_count': 14, + }] - return { - 'id': video_id, - 'display_id': display_id, - 'title': title, - 'description': description, - 'thumbnail': thumbnail, - 'timestamp': timestamp, - 'duration': duration, - 'formats': formats, - } + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + show_id = mobj.group('show_id') + season_id = mobj.group('id') + + fields = [] + fields.extend(self._SHOW_FIELDS) + fields.extend('formatTabs.%s' % field for field in self._SEASON_FIELDS) + fields.extend( + 'formatTabs.formatTabPages.container.movies.%s' % field + for field in self._VIDEO_FIELDS) + + list_info = self._download_json( + 'https://api.nowtv.de/v3/formats/seo?fields=%s&name=%s.php' + % (','.join(fields), show_id), + season_id) + + season = next( + season for season in list_info['formatTabs']['items'] + if season.get('seoheadline') == season_id) + + title = '%s - %s' % (list_info['title'], season['headline']) + + entries = [] + for container in season['formatTabPages']['items']: + for info in ((container.get('container') or {}).get('movies') or {}).get('items') or []: + entries.append(self._extract_video(info)) + + return self.playlist_result( + entries, compat_str(season.get('id') or season_id), title) 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/nuvid.py b/youtube_dl/extractor/nuvid.py index 57928f2ae..9fa7cefad 100644 --- a/youtube_dl/extractor/nuvid.py +++ b/youtube_dl/extractor/nuvid.py @@ -3,11 +3,9 @@ from __future__ import unicode_literals import re from .common import InfoExtractor -from ..compat import ( - compat_urllib_request, -) from ..utils import ( parse_duration, + sanitized_Request, unified_strdate, ) @@ -33,7 +31,7 @@ class NuvidIE(InfoExtractor): formats = [] for dwnld_speed, format_id in [(0, '3gp'), (5, 'mp4')]: - request = compat_urllib_request.Request( + request = sanitized_Request( 'http://m.nuvid.com/play/%s' % video_id) request.add_header('Cookie', 'skip_download_page=1; dwnld_speed=%d; adv_show=1' % dwnld_speed) webpage = self._download_webpage( diff --git a/youtube_dl/extractor/odnoklassniki.py b/youtube_dl/extractor/odnoklassniki.py index ccc88cfb1..184c7a323 100644 --- a/youtube_dl/extractor/odnoklassniki.py +++ b/youtube_dl/extractor/odnoklassniki.py @@ -13,7 +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', @@ -66,6 +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): diff --git a/youtube_dl/extractor/patreon.py b/youtube_dl/extractor/patreon.py index 6cdc2638b..ec8876c28 100644 --- a/youtube_dl/extractor/patreon.py +++ b/youtube_dl/extractor/patreon.py @@ -2,9 +2,7 @@ from __future__ import unicode_literals from .common import InfoExtractor -from ..utils import ( - js_to_json, -) +from ..utils import js_to_json class PatreonIE(InfoExtractor): @@ -65,7 +63,7 @@ class PatreonIE(InfoExtractor): 'password': password, } - request = compat_urllib_request.Request( + request = sanitized_Request( 'https://www.patreon.com/processLogin', compat_urllib_parse.urlencode(login_form).encode('utf-8') ) diff --git a/youtube_dl/extractor/pbs.py b/youtube_dl/extractor/pbs.py index 683c81de3..b787e2a73 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, ) @@ -21,7 +22,7 @@ class PBSIE(InfoExtractor): # Article with embedded player (or direct video) (?:www\.)?pbs\.org/(?:[^/]+/){2,5}(?P<presumptive_id>[^/]+?)(?:\.html)?/?(?:$|[?\#]) | # Player - video\.pbs\.org/(?:widget/)?partnerplayer/(?P<player_id>[^/]+)/ + (?:video|player)\.pbs\.org/(?:widget/)?partnerplayer/(?P<player_id>[^/]+)/ ) ''' @@ -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$', }, 'params': { @@ -134,8 +135,53 @@ class PBSIE(InfoExtractor): 'params': { 'skip_download': True, # requires ffmpeg }, + 'skip': 'Expired', + }, + { + # Video embedded in iframe containing angle brackets as attribute's value (e.g. + # "<iframe style='position: absolute;<br />\ntop: 0; left: 0;' ...", see + # https://github.com/rg3/youtube-dl/issues/7059) + 'url': 'http://www.pbs.org/food/features/a-chefs-life-season-3-episode-5-prickly-business/', + 'info_dict': { + 'id': '2365546844', + 'display_id': 'a-chefs-life-season-3-episode-5-prickly-business', + 'ext': 'mp4', + 'title': "A Chef's Life - Season 3, Ep. 5: Prickly Business", + 'description': 'md5:61db2ddf27c9912f09c241014b118ed1', + 'duration': 1480, + 'thumbnail': 're:^https?://.*\.jpg$', + }, + '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 + }, + }, + { + 'url': 'http://player.pbs.org/widget/partnerplayer/2365297708/?start=0&end=0&chapterbar=false&endscreen=false&topbar=true', + 'only_matching': True, } ] + _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 +212,30 @@ 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') @@ -196,7 +263,7 @@ class PBSIE(InfoExtractor): return self.playlist_result(entries, display_id) info = self._download_json( - 'http://video.pbs.org/videoInfo/%s?format=json&type=partner' % video_id, + 'http://player.pbs.org/videoInfo/%s?format=json&type=partner' % video_id, display_id) formats = [] @@ -213,13 +280,11 @@ 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..63cc764bb 100644 --- a/youtube_dl/extractor/periscope.py +++ b/youtube_dl/extractor/periscope.py @@ -2,17 +2,14 @@ from __future__ import unicode_literals from .common import InfoExtractor -from ..compat import ( - compat_urllib_parse, - compat_urllib_request, -) from ..utils import parse_iso8601 class PeriscopeIE(InfoExtractor): IE_DESC = 'Periscope' - _VALID_URL = r'https?://(?:www\.)?periscope\.tv/w/(?P<id>[^/?#]+)' - _TEST = { + _VALID_URL = r'https?://(?:www\.)?periscope\.tv/[^/]+/(?P<id>[^/?#]+)' + # 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 +22,18 @@ class PeriscopeIE(InfoExtractor): 'uploader_id': '1465763', }, 'skip': 'Expires in 24 hours', - } + }, { + 'url': 'https://www.periscope.tv/w/1ZkKzPbMVggJv', + 'only_matching': True, + }, { + 'url': 'https://www.periscope.tv/bastaakanoggano/1OdKrlkZZjOJX', + '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) @@ -76,24 +80,3 @@ class PeriscopeIE(InfoExtractor): 'thumbnails': thumbnails, 'formats': formats, } - - -class QuickscopeIE(InfoExtractor): - IE_DESC = 'Quick Scope' - _VALID_URL = r'https?://watchonperiscope\.com/broadcast/(?P<id>\d+)' - _TEST = { - 'url': 'https://watchonperiscope.com/broadcast/56180087', - 'only_matching': True, - } - - def _real_extract(self, url): - broadcast_id = self._match_id(url) - request = compat_urllib_request.Request( - 'https://watchonperiscope.com/api/accessChannel', compat_urllib_parse.urlencode({ - 'broadcast_id': broadcast_id, - 'entry_ticket': '', - 'from_push': 'false', - 'uses_sessions': 'true', - }).encode('utf-8')) - return self.url_result( - self._download_json(request, broadcast_id)['share_url'], 'Periscope') diff --git a/youtube_dl/extractor/played.py b/youtube_dl/extractor/played.py index 8a1c296dd..2856af96f 100644 --- a/youtube_dl/extractor/played.py +++ b/youtube_dl/extractor/played.py @@ -5,12 +5,10 @@ import re import os.path from .common import InfoExtractor -from ..compat import ( - compat_urllib_parse, - compat_urllib_request, -) +from ..compat import compat_urllib_parse from ..utils import ( ExtractorError, + sanitized_Request, ) @@ -46,7 +44,7 @@ class PlayedIE(InfoExtractor): headers = { b'Content-Type': b'application/x-www-form-urlencoded', } - req = compat_urllib_request.Request(url, post, headers) + req = sanitized_Request(url, post, headers) webpage = self._download_webpage( req, video_id, note='Downloading video page ...') diff --git a/youtube_dl/extractor/pluralsight.py b/youtube_dl/extractor/pluralsight.py index fd32836cc..aa7dbcb63 100644 --- a/youtube_dl/extractor/pluralsight.py +++ b/youtube_dl/extractor/pluralsight.py @@ -1,29 +1,35 @@ from __future__ import unicode_literals -import re import json +import random +import collections from .common import InfoExtractor from ..compat import ( compat_str, compat_urllib_parse, - compat_urllib_request, compat_urlparse, ) from ..utils import ( ExtractorError, int_or_none, parse_duration, + sanitized_Request, ) -class PluralsightIE(InfoExtractor): +class PluralsightBaseIE(InfoExtractor): + _API_BASE = 'http://app.pluralsight.com' + + +class PluralsightIE(PluralsightBaseIE): IE_NAME = 'pluralsight' - _VALID_URL = r'https?://(?:www\.)?pluralsight\.com/training/player\?author=(?P<author>[^&]+)&name=(?P<name>[^&]+)(?:&mode=live)?&clip=(?P<clip>\d+)&course=(?P<course>[^&]+)' - _LOGIN_URL = 'https://www.pluralsight.com/id/' + _VALID_URL = r'https?://(?:(?:www|app)\.)?pluralsight\.com/training/player\?' + _LOGIN_URL = 'https://app.pluralsight.com/id/' + _NETRC_MACHINE = 'pluralsight' - _TEST = { + _TESTS = [{ 'url': 'http://www.pluralsight.com/training/player?author=mike-mckeown&name=hosting-sql-server-windows-azure-iaas-m7-mgmt&mode=live&clip=3&course=hosting-sql-server-windows-azure-iaas', 'md5': '4d458cf5cf4c593788672419a8dd4cf8', 'info_dict': { @@ -33,7 +39,14 @@ class PluralsightIE(InfoExtractor): 'duration': 338, }, 'skip': 'Requires pluralsight account credentials', - } + }, { + 'url': 'https://app.pluralsight.com/training/player?course=angularjs-get-started&author=scott-allen&name=angularjs-get-started-m1-introduction&clip=0&mode=live', + 'only_matching': True, + }, { + # available without pluralsight account + 'url': 'http://app.pluralsight.com/training/player?author=scott-allen&name=angularjs-get-started-m1-introduction&mode=live&clip=0&course=angularjs-get-started', + 'only_matching': True, + }] def _real_initialize(self): self._login() @@ -41,7 +54,7 @@ class PluralsightIE(InfoExtractor): def _login(self): (username, password) = self._get_login_info() if username is None: - self.raise_login_required('Pluralsight account is required') + return login_page = self._download_webpage( self._LOGIN_URL, None, 'Downloading login page') @@ -60,7 +73,7 @@ class PluralsightIE(InfoExtractor): if not post_url.startswith('http'): post_url = compat_urlparse.urljoin(self._LOGIN_URL, post_url) - request = compat_urllib_request.Request( + request = sanitized_Request( post_url, compat_urllib_parse.urlencode(login_form).encode('utf-8')) request.add_header('Content-Type', 'application/x-www-form-urlencoded') @@ -73,31 +86,48 @@ class PluralsightIE(InfoExtractor): if error: raise ExtractorError('Unable to login: %s' % error, expected=True) + if all(p not in response for p in ('__INITIAL_STATE__', '"currentUser"')): + raise ExtractorError('Unable to log in') + def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - author = mobj.group('author') - name = mobj.group('name') - clip_id = mobj.group('clip') - course = mobj.group('course') + qs = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query) + + author = qs.get('author', [None])[0] + name = qs.get('name', [None])[0] + clip_id = qs.get('clip', [None])[0] + course = qs.get('course', [None])[0] + + if any(not f for f in (author, name, clip_id, course,)): + raise ExtractorError('Invalid URL', expected=True) display_id = '%s-%s' % (name, clip_id) webpage = self._download_webpage(url, display_id) - collection = self._parse_json( - self._search_regex( - r'moduleCollection\s*:\s*new\s+ModuleCollection\((\[.+?\])\s*,\s*\$rootScope\)', - webpage, 'modules'), - display_id) + modules = self._search_regex( + r'moduleCollection\s*:\s*new\s+ModuleCollection\((\[.+?\])\s*,\s*\$rootScope\)', + webpage, 'modules', default=None) + + if modules: + collection = self._parse_json(modules, display_id) + else: + # Webpage may be served in different layout (see + # https://github.com/rg3/youtube-dl/issues/7607) + collection = self._parse_json( + self._search_regex( + r'var\s+initialState\s*=\s*({.+?});\n', webpage, 'initial state'), + display_id)['course']['modules'] module, clip = None, None for module_ in collection: - if module_.get('moduleName') == name: + if name in (module_.get('moduleName'), module_.get('name')): module = module_ for clip_ in module_.get('clips', []): clip_index = clip_.get('clipIndex') if clip_index is None: + clip_index = clip_.get('index') + if clip_index is None: continue if compat_str(clip_index) == clip_id: clip = clip_ @@ -112,13 +142,33 @@ class PluralsightIE(InfoExtractor): 'high': {'width': 1024, 'height': 768}, } + AllowedQuality = collections.namedtuple('AllowedQuality', ['ext', 'qualities']) + ALLOWED_QUALITIES = ( - ('webm', ('high',)), - ('mp4', ('low', 'medium', 'high',)), + AllowedQuality('webm', ('high',)), + AllowedQuality('mp4', ('low', 'medium', 'high',)), ) + # In order to minimize the number of calls to ViewClip API and reduce + # the probability of being throttled or banned by Pluralsight we will request + # only single format until formats listing was explicitly requested. + if self._downloader.params.get('listformats', False): + allowed_qualities = ALLOWED_QUALITIES + else: + def guess_allowed_qualities(): + req_format = self._downloader.params.get('format') or 'best' + req_format_split = req_format.split('-') + if len(req_format_split) > 1: + req_ext, req_quality = req_format_split + for allowed_quality in ALLOWED_QUALITIES: + if req_ext == allowed_quality.ext and req_quality in allowed_quality.qualities: + return (AllowedQuality(req_ext, (req_quality, )), ) + req_ext = 'webm' if self._downloader.params.get('prefer_free_formats') else 'mp4' + return (AllowedQuality(req_ext, ('high', )), ) + allowed_qualities = guess_allowed_qualities() + formats = [] - for ext, qualities in ALLOWED_QUALITIES: + for ext, qualities in allowed_qualities: for quality in qualities: f = QUALITIES[quality].copy() clip_post = { @@ -131,13 +181,24 @@ class PluralsightIE(InfoExtractor): 'mt': ext, 'q': '%dx%d' % (f['width'], f['height']), } - request = compat_urllib_request.Request( - 'http://www.pluralsight.com/training/Player/ViewClip', + request = sanitized_Request( + '%s/training/Player/ViewClip' % self._API_BASE, json.dumps(clip_post).encode('utf-8')) request.add_header('Content-Type', 'application/json;charset=utf-8') format_id = '%s-%s' % (ext, quality) clip_url = self._download_webpage( request, display_id, 'Downloading %s URL' % format_id, fatal=False) + + # Pluralsight tracks multiple sequential calls to ViewClip API and start + # to return 429 HTTP errors after some time (see + # https://github.com/rg3/youtube-dl/pull/6989). Moreover it may even lead + # to account ban (see https://github.com/rg3/youtube-dl/issues/6842). + # To somewhat reduce the probability of these consequences + # we will sleep random amount of time before each call to ViewClip. + self._sleep( + random.randint(2, 5), display_id, + '%(video_id)s: Waiting for %(timeout)s seconds to avoid throttling') + if not clip_url: continue f.update({ @@ -163,10 +224,10 @@ class PluralsightIE(InfoExtractor): } -class PluralsightCourseIE(InfoExtractor): +class PluralsightCourseIE(PluralsightBaseIE): IE_NAME = 'pluralsight:course' - _VALID_URL = r'https?://(?:www\.)?pluralsight\.com/courses/(?P<id>[^/]+)' - _TEST = { + _VALID_URL = r'https?://(?:(?:www|app)\.)?pluralsight\.com/(?:library/)?courses/(?P<id>[^/]+)' + _TESTS = [{ # Free course from Pluralsight Starter Subscription for Microsoft TechNet # https://offers.pluralsight.com/technet?loc=zTS3z&prod=zOTprodz&tech=zOttechz&prog=zOTprogz&type=zSOz&media=zOTmediaz&country=zUSz 'url': 'http://www.pluralsight.com/courses/hosting-sql-server-windows-azure-iaas', @@ -176,7 +237,14 @@ class PluralsightCourseIE(InfoExtractor): 'description': 'md5:61b37e60f21c4b2f91dc621a977d0986', }, 'playlist_count': 31, - } + }, { + # available without pluralsight account + 'url': 'https://www.pluralsight.com/courses/angularjs-get-started', + 'only_matching': True, + }, { + 'url': 'https://app.pluralsight.com/library/courses/understanding-microsoft-azure-amazon-aws/table-of-contents', + 'only_matching': True, + }] def _real_extract(self, url): course_id = self._match_id(url) @@ -184,14 +252,14 @@ class PluralsightCourseIE(InfoExtractor): # TODO: PSM cookie course = self._download_json( - 'http://www.pluralsight.com/data/course/%s' % course_id, + '%s/data/course/%s' % (self._API_BASE, course_id), course_id, 'Downloading course JSON') title = course['title'] description = course.get('description') or course.get('shortDescription') course_data = self._download_json( - 'http://www.pluralsight.com/data/course/content/%s' % course_id, + '%s/data/course/content/%s' % (self._API_BASE, course_id), course_id, 'Downloading course data JSON') entries = [] @@ -201,7 +269,7 @@ class PluralsightCourseIE(InfoExtractor): if not player_parameters: continue entries.append(self.url_result( - 'http://www.pluralsight.com/training/player?%s' % player_parameters, + '%s/training/player?%s' % (self._API_BASE, player_parameters), 'Pluralsight')) return self.playlist_result(entries, course_id, title, description) diff --git a/youtube_dl/extractor/pornhd.py b/youtube_dl/extractor/pornhd.py index dbb2c3bd9..57c78ba52 100644 --- a/youtube_dl/extractor/pornhd.py +++ b/youtube_dl/extractor/pornhd.py @@ -36,7 +36,8 @@ class PornHdIE(InfoExtractor): webpage = self._download_webpage(url, display_id or video_id) title = self._html_search_regex( - r'<title>(.+) porn HD.+?</title>', webpage, 'title') + [r'<span[^>]+class=["\']video-name["\'][^>]*>([^<]+)', + r'<title>(.+?) - .*?[Pp]ornHD.*?</title>'], webpage, 'title') description = self._html_search_regex( r'<div class="description">([^<]+)</div>', webpage, 'description', fatal=False) view_count = int_or_none(self._html_search_regex( diff --git a/youtube_dl/extractor/pornhub.py b/youtube_dl/extractor/pornhub.py index a656ad85a..965940a4b 100644 --- a/youtube_dl/extractor/pornhub.py +++ b/youtube_dl/extractor/pornhub.py @@ -8,10 +8,10 @@ from ..compat import ( compat_urllib_parse_unquote, compat_urllib_parse_unquote_plus, compat_urllib_parse_urlparse, - compat_urllib_request, ) from ..utils import ( ExtractorError, + sanitized_Request, str_to_int, ) from ..aes import ( @@ -53,7 +53,7 @@ class PornHubIE(InfoExtractor): def _real_extract(self, url): video_id = self._match_id(url) - req = compat_urllib_request.Request( + req = sanitized_Request( 'http://www.pornhub.com/view_video.php?viewkey=%s' % video_id) req.add_header('Cookie', 'age_verified=1') webpage = self._download_webpage(req, video_id) diff --git a/youtube_dl/extractor/pornotube.py b/youtube_dl/extractor/pornotube.py index 34735c51e..5398e708b 100644 --- a/youtube_dl/extractor/pornotube.py +++ b/youtube_dl/extractor/pornotube.py @@ -3,11 +3,9 @@ from __future__ import unicode_literals import json from .common import InfoExtractor -from ..compat import ( - compat_urllib_request, -) from ..utils import ( int_or_none, + sanitized_Request, ) @@ -46,7 +44,7 @@ class PornotubeIE(InfoExtractor): 'authenticationSpaceKey': originAuthenticationSpaceKey, 'credentials': 'Clip Application', } - token_req = compat_urllib_request.Request( + token_req = sanitized_Request( 'https://api.aebn.net/auth/v1/token/primal', data=json.dumps(token_req_data).encode('utf-8')) token_req.add_header('Content-Type', 'application/json') @@ -56,7 +54,7 @@ class PornotubeIE(InfoExtractor): token = token_answer['tokenKey'] # Get video URL - delivery_req = compat_urllib_request.Request( + delivery_req = sanitized_Request( 'https://api.aebn.net/delivery/v1/clips/%s/MP4' % video_id) delivery_req.add_header('Authorization', token) delivery_info = self._download_json( @@ -64,7 +62,7 @@ class PornotubeIE(InfoExtractor): video_url = delivery_info['mediaUrl'] # Get additional info (title etc.) - info_req = compat_urllib_request.Request( + info_req = sanitized_Request( 'https://api.aebn.net/content/v1/clips/%s?expand=' 'title,description,primaryImageNumber,startSecond,endSecond,' 'movie.title,movie.MovieId,movie.boxCoverFront,movie.stars,' diff --git a/youtube_dl/extractor/primesharetv.py b/youtube_dl/extractor/primesharetv.py index 304359dc5..85aae9576 100644 --- a/youtube_dl/extractor/primesharetv.py +++ b/youtube_dl/extractor/primesharetv.py @@ -1,11 +1,11 @@ from __future__ import unicode_literals from .common import InfoExtractor -from ..compat import ( - compat_urllib_parse, - compat_urllib_request, +from ..compat import compat_urllib_parse +from ..utils import ( + ExtractorError, + sanitized_Request, ) -from ..utils import ExtractorError class PrimeShareTVIE(InfoExtractor): @@ -41,7 +41,7 @@ class PrimeShareTVIE(InfoExtractor): webpage, 'wait time', default=7)) + 1 self._sleep(wait_time, video_id) - req = compat_urllib_request.Request( + req = sanitized_Request( url, compat_urllib_parse.urlencode(fields), headers) video_page = self._download_webpage( req, video_id, 'Downloading video page') diff --git a/youtube_dl/extractor/promptfile.py b/youtube_dl/extractor/promptfile.py index 8190ed676..d5357283a 100644 --- a/youtube_dl/extractor/promptfile.py +++ b/youtube_dl/extractor/promptfile.py @@ -4,13 +4,11 @@ from __future__ import unicode_literals import re from .common import InfoExtractor -from ..compat import ( - compat_urllib_parse, - compat_urllib_request, -) +from ..compat import compat_urllib_parse from ..utils import ( determine_ext, ExtractorError, + sanitized_Request, ) @@ -37,7 +35,7 @@ class PromptFileIE(InfoExtractor): fields = self._hidden_inputs(webpage) post = compat_urllib_parse.urlencode(fields) - req = compat_urllib_request.Request(url, post) + req = sanitized_Request(url, post) req.add_header('Content-type', 'application/x-www-form-urlencoded') webpage = self._download_webpage( req, video_id, 'Downloading video page') 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 c98539f6a..1ba3bbddf 100644 --- a/youtube_dl/extractor/qqmusic.py +++ b/youtube_dl/extractor/qqmusic.py @@ -7,11 +7,11 @@ import re from .common import InfoExtractor from ..utils import ( + sanitized_Request, strip_jsonp, unescapeHTML, clean_html, ) -from ..compat import compat_urllib_request class QQMusicIE(InfoExtractor): @@ -201,7 +201,7 @@ class QQMusicSingerIE(QQPlaylistBaseIE): singer_desc = None if singer_id: - req = compat_urllib_request.Request( + req = sanitized_Request( 'http://s.plcloud.music.qq.com/fcgi-bin/fcg_get_singer_desc.fcg?utf8=1&outCharset=utf-8&format=xml&singerid=%s' % singer_id) req.add_header( 'Referer', 'http://s.plcloud.music.qq.com/xhr_proxy_utf8.html') 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/rtve.py b/youtube_dl/extractor/rtve.py index 5b97d33ca..603d7bd00 100644 --- a/youtube_dl/extractor/rtve.py +++ b/youtube_dl/extractor/rtve.py @@ -6,11 +6,11 @@ import re import time from .common import InfoExtractor -from ..compat import compat_urllib_request, compat_urlparse from ..utils import ( ExtractorError, float_or_none, remove_end, + sanitized_Request, std_headers, struct_unpack, ) @@ -102,20 +102,14 @@ class RTVEALaCartaIE(InfoExtractor): if info['state'] == 'DESPU': raise ExtractorError('The video is no longer available', expected=True) png_url = 'http://www.rtve.es/ztnr/movil/thumbnail/%s/videos/%s.png' % (self._manager, video_id) - png_request = compat_urllib_request.Request(png_url) + png_request = sanitized_Request(png_url) png_request.add_header('Referer', url) png = self._download_webpage(png_request, video_id, 'Downloading url information') video_url = _decrypt_url(png) if not video_url.endswith('.f4m'): - auth_url = video_url.replace( + video_url = video_url.replace( 'resources/', 'auth/resources/' ).replace('.net.rtve', '.multimedia.cdn.rtve') - video_path = self._download_webpage( - auth_url, video_id, 'Getting video url') - # Use mvod1.akcdn instead of flash.akamaihd.multimedia.cdn to get - # the right Content-Length header and the mp4 format - video_url = compat_urlparse.urljoin( - 'http://mvod1.akcdn.rtve.es/', video_path) subtitles = None if info.get('sbtFile') is not None: diff --git a/youtube_dl/extractor/rutube.py b/youtube_dl/extractor/rutube.py index d94dc7399..6b09550b0 100644 --- a/youtube_dl/extractor/rutube.py +++ b/youtube_dl/extractor/rutube.py @@ -9,7 +9,7 @@ from ..compat import ( compat_str, ) from ..utils import ( - ExtractorError, + determine_ext, unified_strdate, ) @@ -51,10 +51,25 @@ class RutubeIE(InfoExtractor): 'http://rutube.ru/api/play/options/%s/?format=json' % video_id, video_id, 'Downloading options JSON') - m3u8_url = options['video_balancer'].get('m3u8') - if m3u8_url is None: - raise ExtractorError('Couldn\'t find m3u8 manifest url') - formats = self._extract_m3u8_formats(m3u8_url, video_id, ext='mp4') + formats = [] + for format_id, format_url in options['video_balancer'].items(): + ext = determine_ext(format_url) + if ext == 'm3u8': + m3u8_formats = self._extract_m3u8_formats( + format_url, video_id, 'mp4', m3u8_id=format_id, fatal=False) + if m3u8_formats: + formats.extend(m3u8_formats) + elif ext == 'f4m': + f4m_formats = self._extract_f4m_formats( + format_url, video_id, f4m_id=format_id, fatal=False) + if f4m_formats: + formats.extend(f4m_formats) + else: + formats.append({ + 'url': format_url, + 'format_id': format_id, + }) + self._sort_formats(formats) return { 'id': video['id'], @@ -74,9 +89,9 @@ class RutubeIE(InfoExtractor): class RutubeEmbedIE(InfoExtractor): IE_NAME = 'rutube:embed' IE_DESC = 'Rutube embedded videos' - _VALID_URL = 'https?://rutube\.ru/video/embed/(?P<id>[0-9]+)' + _VALID_URL = 'https?://rutube\.ru/(?:video|play)/embed/(?P<id>[0-9]+)' - _TEST = { + _TESTS = [{ 'url': 'http://rutube.ru/video/embed/6722881?vk_puid37=&vk_puid38=', 'info_dict': { 'id': 'a10e53b86e8f349080f718582ce4c661', @@ -90,7 +105,10 @@ class RutubeEmbedIE(InfoExtractor): 'params': { 'skip_download': 'Requires ffmpeg', }, - } + }, { + 'url': 'http://rutube.ru/play/embed/8083783', + 'only_matching': True, + }] def _real_extract(self, url): embed_id = self._match_id(url) diff --git a/youtube_dl/extractor/ruutu.py b/youtube_dl/extractor/ruutu.py index a16b73ff4..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': diff --git a/youtube_dl/extractor/safari.py b/youtube_dl/extractor/safari.py index a602af692..919704261 100644 --- a/youtube_dl/extractor/safari.py +++ b/youtube_dl/extractor/safari.py @@ -4,14 +4,12 @@ from __future__ import unicode_literals import re from .common import InfoExtractor -from .brightcove import BrightcoveIE +from .brightcove import BrightcoveLegacyIE -from ..compat import ( - compat_urllib_parse, - compat_urllib_request, -) +from ..compat import compat_urllib_parse from ..utils import ( ExtractorError, + sanitized_Request, smuggle_url, std_headers, ) @@ -58,7 +56,7 @@ class SafariBaseIE(InfoExtractor): 'next': '', } - request = compat_urllib_request.Request( + request = sanitized_Request( self._LOGIN_URL, compat_urllib_parse.urlencode(login_form), headers=headers) login_page = self._download_webpage( request, None, 'Logging in as %s' % username) @@ -112,11 +110,11 @@ class SafariIE(SafariBaseIE): '%s/%s/chapter-content/%s.html' % (self._API_BASE, course_id, part), part) - bc_url = BrightcoveIE._extract_brightcove_url(webpage) + bc_url = BrightcoveLegacyIE._extract_brightcove_url(webpage) if not bc_url: raise ExtractorError('Could not extract Brightcove URL from %s' % url, expected=True) - return self.url_result(smuggle_url(bc_url, {'Referer': url}), 'Brightcove') + return self.url_result(smuggle_url(bc_url, {'Referer': url}), 'BrightcoveLegacy') class SafariCourseIE(SafariBaseIE): diff --git a/youtube_dl/extractor/sandia.py b/youtube_dl/extractor/sandia.py index 9c88167f0..759898a49 100644 --- a/youtube_dl/extractor/sandia.py +++ b/youtube_dl/extractor/sandia.py @@ -6,14 +6,12 @@ import json import re from .common import InfoExtractor -from ..compat import ( - compat_urllib_request, - compat_urlparse, -) +from ..compat import compat_urlparse from ..utils import ( int_or_none, js_to_json, mimetype2ext, + sanitized_Request, unified_strdate, ) @@ -37,7 +35,7 @@ class SandiaIE(InfoExtractor): def _real_extract(self, url): video_id = self._match_id(url) - req = compat_urllib_request.Request(url) + req = sanitized_Request(url) req.add_header('Cookie', 'MediasitePlayerCaps=ClientPlugins=4') webpage = self._download_webpage(req, video_id) 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/shared.py b/youtube_dl/extractor/shared.py index c5636e8e9..8eda3c864 100644 --- a/youtube_dl/extractor/shared.py +++ b/youtube_dl/extractor/shared.py @@ -3,13 +3,11 @@ from __future__ import unicode_literals import base64 from .common import InfoExtractor -from ..compat import ( - compat_urllib_parse, - compat_urllib_request, -) +from ..compat import compat_urllib_parse from ..utils import ( ExtractorError, int_or_none, + sanitized_Request, ) @@ -46,7 +44,7 @@ class SharedIE(InfoExtractor): 'Video %s does not exist' % video_id, expected=True) download_form = self._hidden_inputs(webpage) - request = compat_urllib_request.Request( + request = sanitized_Request( url, compat_urllib_parse.urlencode(download_form)) request.add_header('Content-Type', 'application/x-www-form-urlencoded') diff --git a/youtube_dl/extractor/sharesix.py b/youtube_dl/extractor/sharesix.py index ac3e3adf2..f1ea9bdb2 100644 --- a/youtube_dl/extractor/sharesix.py +++ b/youtube_dl/extractor/sharesix.py @@ -4,12 +4,10 @@ from __future__ import unicode_literals import re from .common import InfoExtractor -from ..compat import ( - compat_urllib_parse, - compat_urllib_request, -) +from ..compat import compat_urllib_parse from ..utils import ( parse_duration, + sanitized_Request, ) @@ -50,7 +48,7 @@ class ShareSixIE(InfoExtractor): 'method_free': 'Free' } post = compat_urllib_parse.urlencode(fields) - req = compat_urllib_request.Request(url, post) + req = sanitized_Request(url, post) req.add_header('Content-type', 'application/x-www-form-urlencoded') webpage = self._download_webpage(req, video_id, diff --git a/youtube_dl/extractor/sina.py b/youtube_dl/extractor/sina.py index 0891a441f..b2258a0f6 100644 --- a/youtube_dl/extractor/sina.py +++ b/youtube_dl/extractor/sina.py @@ -4,10 +4,8 @@ from __future__ import unicode_literals import re from .common import InfoExtractor -from ..compat import ( - compat_urllib_request, - compat_urllib_parse, -) +from ..compat import compat_urllib_parse +from ..utils import sanitized_Request class SinaIE(InfoExtractor): @@ -61,7 +59,7 @@ class SinaIE(InfoExtractor): if mobj.group('token') is not None: # The video id is in the redirected url self.to_screen('Getting video id') - request = compat_urllib_request.Request(url) + request = sanitized_Request(url) request.get_method = lambda: 'HEAD' (_, urlh) = self._download_webpage_handle(request, 'NA', False) return self._real_extract(urlh.geturl()) diff --git a/youtube_dl/extractor/smotri.py b/youtube_dl/extractor/smotri.py index 35a81ee87..30210c8a3 100644 --- a/youtube_dl/extractor/smotri.py +++ b/youtube_dl/extractor/smotri.py @@ -7,13 +7,11 @@ import hashlib import uuid from .common import InfoExtractor -from ..compat import ( - compat_urllib_parse, - compat_urllib_request, -) +from ..compat import compat_urllib_parse from ..utils import ( ExtractorError, int_or_none, + sanitized_Request, unified_strdate, ) @@ -176,7 +174,7 @@ class SmotriIE(InfoExtractor): if video_password: video_form['pass'] = hashlib.md5(video_password.encode('utf-8')).hexdigest() - request = compat_urllib_request.Request( + request = sanitized_Request( 'http://smotri.com/video/view/url/bot/', compat_urllib_parse.urlencode(video_form)) request.add_header('Content-Type', 'application/x-www-form-urlencoded') @@ -339,7 +337,7 @@ class SmotriBroadcastIE(InfoExtractor): 'password': password, } - request = compat_urllib_request.Request( + request = sanitized_Request( broadcast_url + '/?no_redirect=1', compat_urllib_parse.urlencode(login_form)) request.add_header('Content-Type', 'application/x-www-form-urlencoded') broadcast_page = self._download_webpage( diff --git a/youtube_dl/extractor/sohu.py b/youtube_dl/extractor/sohu.py index ba2d5e19b..daf6ad555 100644 --- a/youtube_dl/extractor/sohu.py +++ b/youtube_dl/extractor/sohu.py @@ -6,11 +6,11 @@ import re from .common import InfoExtractor from ..compat import ( compat_str, - compat_urllib_request, compat_urllib_parse, ) from ..utils import ( ExtractorError, + sanitized_Request, ) @@ -96,7 +96,7 @@ class SohuIE(InfoExtractor): else: base_data_url = 'http://hot.vrs.sohu.com/vrs_flash.action?vid=' - req = compat_urllib_request.Request(base_data_url + vid_id) + req = sanitized_Request(base_data_url + vid_id) cn_verification_proxy = self._downloader.params.get('cn_verification_proxy') if cn_verification_proxy: diff --git a/youtube_dl/extractor/soundcloud.py b/youtube_dl/extractor/soundcloud.py index 2b60d354a..02e64e094 100644 --- a/youtube_dl/extractor/soundcloud.py +++ b/youtube_dl/extractor/soundcloud.py @@ -4,13 +4,17 @@ from __future__ import unicode_literals import re import itertools -from .common import InfoExtractor +from .common import ( + InfoExtractor, + SearchInfoExtractor +) from ..compat import ( compat_str, compat_urlparse, compat_urllib_parse, ) from ..utils import ( + encode_dict, ExtractorError, int_or_none, unified_strdate, @@ -469,3 +473,60 @@ class SoundcloudPlaylistIE(SoundcloudIE): 'description': data.get('description'), 'entries': entries, } + + +class SoundcloudSearchIE(SearchInfoExtractor, SoundcloudIE): + IE_NAME = 'soundcloud:search' + IE_DESC = 'Soundcloud search' + _MAX_RESULTS = float('inf') + _TESTS = [{ + 'url': 'scsearch15:post-avant jazzcore', + 'info_dict': { + 'title': 'post-avant jazzcore', + }, + 'playlist_count': 15, + }] + + _SEARCH_KEY = 'scsearch' + _MAX_RESULTS_PER_PAGE = 200 + _DEFAULT_RESULTS_PER_PAGE = 50 + _API_V2_BASE = 'https://api-v2.soundcloud.com' + + def _get_collection(self, endpoint, collection_id, **query): + limit = min( + query.get('limit', self._DEFAULT_RESULTS_PER_PAGE), + self._MAX_RESULTS_PER_PAGE) + query['limit'] = limit + query['client_id'] = self._CLIENT_ID + query['linked_partitioning'] = '1' + query['offset'] = 0 + data = compat_urllib_parse.urlencode(encode_dict(query)) + next_url = '{0}{1}?{2}'.format(self._API_V2_BASE, endpoint, data) + + collected_results = 0 + + for i in itertools.count(1): + response = self._download_json( + next_url, collection_id, 'Downloading page {0}'.format(i), + 'Unable to download API page') + + collection = response.get('collection', []) + if not collection: + break + + collection = list(filter(bool, collection)) + collected_results += len(collection) + + for item in collection: + yield self.url_result(item['uri'], SoundcloudIE.ie_key()) + + if not collection or collected_results >= limit: + break + + next_url = response.get('next_href') + if not next_url: + break + + def _get_n_results(self, query, n): + tracks = self._get_collection('/search/tracks', query, limit=n, q=query) + return self.playlist_result(tracks, playlist_title=query) diff --git a/youtube_dl/extractor/space.py b/youtube_dl/extractor/space.py index c2d0d36a6..ebb5d6ec0 100644 --- a/youtube_dl/extractor/space.py +++ b/youtube_dl/extractor/space.py @@ -3,14 +3,14 @@ from __future__ import unicode_literals import re from .common import InfoExtractor -from .brightcove import BrightcoveIE +from .brightcove import BrightcoveLegacyIE from ..utils import RegexNotFoundError, ExtractorError class SpaceIE(InfoExtractor): _VALID_URL = r'https?://(?:(?:www|m)\.)?space\.com/\d+-(?P<title>[^/\.\?]*?)-video\.html' _TEST = { - 'add_ie': ['Brightcove'], + 'add_ie': ['BrightcoveLegacy'], 'url': 'http://www.space.com/23373-huge-martian-landforms-detail-revealed-by-european-probe-video.html', 'info_dict': { 'id': '2780937028001', @@ -31,8 +31,8 @@ class SpaceIE(InfoExtractor): brightcove_url = self._og_search_video_url(webpage) except RegexNotFoundError: # Other videos works fine with the info from the object - brightcove_url = BrightcoveIE._extract_brightcove_url(webpage) + brightcove_url = BrightcoveLegacyIE._extract_brightcove_url(webpage) if brightcove_url is None: raise ExtractorError( 'The webpage does not contain a video', expected=True) - return self.url_result(brightcove_url, BrightcoveIE.ie_key()) + return self.url_result(brightcove_url, BrightcoveLegacyIE.ie_key()) diff --git a/youtube_dl/extractor/spankwire.py b/youtube_dl/extractor/spankwire.py index 9e8fb35b2..692fd78e8 100644 --- a/youtube_dl/extractor/spankwire.py +++ b/youtube_dl/extractor/spankwire.py @@ -6,9 +6,9 @@ from .common import InfoExtractor from ..compat import ( compat_urllib_parse_unquote, compat_urllib_parse_urlparse, - compat_urllib_request, ) from ..utils import ( + sanitized_Request, str_to_int, unified_strdate, ) @@ -51,7 +51,7 @@ class SpankwireIE(InfoExtractor): mobj = re.match(self._VALID_URL, url) video_id = mobj.group('id') - req = compat_urllib_request.Request('http://www.' + mobj.group('url')) + req = sanitized_Request('http://www.' + mobj.group('url')) req.add_header('Cookie', 'age_verified=1') webpage = self._download_webpage(req, 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/sportdeutschland.py b/youtube_dl/extractor/sportdeutschland.py index 7ec6c613f..ebb75f059 100644 --- a/youtube_dl/extractor/sportdeutschland.py +++ b/youtube_dl/extractor/sportdeutschland.py @@ -4,11 +4,9 @@ from __future__ import unicode_literals import re from .common import InfoExtractor -from ..compat import ( - compat_urllib_request, -) from ..utils import ( parse_iso8601, + sanitized_Request, ) @@ -54,7 +52,7 @@ class SportDeutschlandIE(InfoExtractor): api_url = 'http://proxy.vidibusdynamic.net/sportdeutschland.tv/api/permalinks/%s/%s?access_token=true' % ( sport_id, video_id) - req = compat_urllib_request.Request(api_url, headers={ + req = sanitized_Request(api_url, headers={ 'Accept': 'application/vnd.vidibus.v2.html+json', 'Referer': url, }) 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/streamcloud.py b/youtube_dl/extractor/streamcloud.py index d4e134015..77841b946 100644 --- a/youtube_dl/extractor/streamcloud.py +++ b/youtube_dl/extractor/streamcloud.py @@ -4,10 +4,8 @@ from __future__ import unicode_literals import re from .common import InfoExtractor -from ..compat import ( - compat_urllib_parse, - compat_urllib_request, -) +from ..compat import compat_urllib_parse +from ..utils import sanitized_Request class StreamcloudIE(InfoExtractor): @@ -43,7 +41,7 @@ class StreamcloudIE(InfoExtractor): headers = { b'Content-Type': b'application/x-www-form-urlencoded', } - req = compat_urllib_request.Request(url, post, headers) + req = sanitized_Request(url, post, headers) webpage = self._download_webpage( req, video_id, note='Downloading video page ...') diff --git a/youtube_dl/extractor/streamcz.py b/youtube_dl/extractor/streamcz.py index e92b93285..d3d2b7eb7 100644 --- a/youtube_dl/extractor/streamcz.py +++ b/youtube_dl/extractor/streamcz.py @@ -5,11 +5,9 @@ import hashlib import time from .common import InfoExtractor -from ..compat import ( - compat_urllib_request, -) from ..utils import ( int_or_none, + sanitized_Request, ) @@ -54,7 +52,7 @@ class StreamCZIE(InfoExtractor): video_id = self._match_id(url) api_path = '/episode/%s' % video_id - req = compat_urllib_request.Request(self._API_URL + api_path) + req = sanitized_Request(self._API_URL + api_path) req.add_header('Api-Password', _get_api_key(api_path)) data = self._download_json(req, video_id) diff --git a/youtube_dl/extractor/tapely.py b/youtube_dl/extractor/tapely.py index 744f9db38..ed560bd24 100644 --- a/youtube_dl/extractor/tapely.py +++ b/youtube_dl/extractor/tapely.py @@ -4,14 +4,12 @@ from __future__ import unicode_literals import re from .common import InfoExtractor -from ..compat import ( - compat_urllib_request, -) from ..utils import ( clean_html, ExtractorError, float_or_none, parse_iso8601, + sanitized_Request, ) @@ -53,7 +51,7 @@ class TapelyIE(InfoExtractor): display_id = mobj.group('id') playlist_url = self._API_URL.format(display_id) - request = compat_urllib_request.Request(playlist_url) + request = sanitized_Request(playlist_url) request.add_header('X-Requested-With', 'XMLHttpRequest') request.add_header('Accept', 'application/json') request.add_header('Referer', url) diff --git a/youtube_dl/extractor/theplatform.py b/youtube_dl/extractor/theplatform.py index 25edc3100..1555aa77c 100644 --- a/youtube_dl/extractor/theplatform.py +++ b/youtube_dl/extractor/theplatform.py @@ -139,6 +139,11 @@ class ThePlatformIE(ThePlatformBaseIE): 'upload_date': '20150701', 'categories': ['Today/Shows/Orange Room', 'Today/Sections/Money', 'Today/Topics/Tech', "Today/Topics/Editor's picks"], }, + }, { + # From http://www.nbc.com/the-blacklist/video/sir-crispin-crandall/2928790?onid=137781#vc137781=1 + # geo-restricted (US), HLS encrypted with AES-128 + 'url': 'http://player.theplatform.com/p/NnzsPC/onsite_universal/select/media/guid/2410887629/2928790?fwsitesection=nbc_the_blacklist_video_library&autoPlay=true&carouselID=137781', + 'only_matching': True, }] @staticmethod @@ -182,8 +187,12 @@ class ThePlatformIE(ThePlatformBaseIE): # Seems there's no pattern for the interested script filename, so # I try one by one for script in reversed(scripts): - feed_script = self._download_webpage(script, video_id, 'Downloading feed script') - feed_id = self._search_regex(r'defaultFeedId\s*:\s*"([^"]+)"', feed_script, 'default feed id', default=None) + feed_script = self._download_webpage( + self._proto_relative_url(script, 'http:'), + video_id, 'Downloading feed script') + feed_id = self._search_regex( + r'defaultFeedId\s*:\s*"([^"]+)"', feed_script, + 'default feed id', default=None) if feed_id is not None: break if feed_id is None: @@ -193,6 +202,15 @@ class ThePlatformIE(ThePlatformBaseIE): if smuggled_data.get('force_smil_url', False): smil_url = url + # Explicitly specified SMIL (see https://github.com/rg3/youtube-dl/issues/7385) + elif '/guid/' in url: + webpage = self._download_webpage(url, video_id) + smil_url = self._search_regex( + r'<link[^>]+href=(["\'])(?P<url>.+?)\1[^>]+type=["\']application/smil\+xml', + webpage, 'smil url', group='url') + path = self._search_regex( + r'link\.theplatform\.com/s/((?:[^/?#&]+/)+[^/?#&]+)', smil_url, 'path') + smil_url += '?' if '?' not in smil_url else '&' + 'formats=m3u,mpeg4&format=SMIL' elif mobj.group('config'): config_url = url + '&form=json' config_url = config_url.replace('swf/', 'config/') diff --git a/youtube_dl/extractor/tlc.py b/youtube_dl/extractor/tlc.py index 13263614c..d6d038a8d 100644 --- a/youtube_dl/extractor/tlc.py +++ b/youtube_dl/extractor/tlc.py @@ -3,7 +3,7 @@ from __future__ import unicode_literals import re from .common import InfoExtractor -from .brightcove import BrightcoveIE +from .brightcove import BrightcoveLegacyIE from .discovery import DiscoveryIE from ..compat import compat_urlparse @@ -66,6 +66,6 @@ class TlcDeIE(InfoExtractor): return { '_type': 'url', - 'url': BrightcoveIE._extract_brightcove_url(iframe), - 'ie': BrightcoveIE.ie_key(), + 'url': BrightcoveLegacyIE._extract_brightcove_url(iframe), + 'ie': BrightcoveLegacyIE.ie_key(), } diff --git a/youtube_dl/extractor/tube8.py b/youtube_dl/extractor/tube8.py index c9cb69333..46ef61ff5 100644 --- a/youtube_dl/extractor/tube8.py +++ b/youtube_dl/extractor/tube8.py @@ -4,12 +4,10 @@ import json import re from .common import InfoExtractor -from ..compat import ( - compat_urllib_parse_urlparse, - compat_urllib_request, -) +from ..compat import compat_urllib_parse_urlparse from ..utils import ( int_or_none, + sanitized_Request, str_to_int, ) from ..aes import aes_decrypt_text @@ -42,7 +40,7 @@ class Tube8IE(InfoExtractor): video_id = mobj.group('id') display_id = mobj.group('display_id') - req = compat_urllib_request.Request(url) + req = sanitized_Request(url) req.add_header('Cookie', 'age_verified=1') webpage = self._download_webpage(req, display_id) diff --git a/youtube_dl/extractor/tubitv.py b/youtube_dl/extractor/tubitv.py index 4f86b3ee9..6d78b5dfe 100644 --- a/youtube_dl/extractor/tubitv.py +++ b/youtube_dl/extractor/tubitv.py @@ -5,13 +5,11 @@ import codecs import re from .common import InfoExtractor -from ..compat import ( - compat_urllib_parse, - compat_urllib_request -) +from ..compat import compat_urllib_parse from ..utils import ( ExtractorError, int_or_none, + sanitized_Request, ) @@ -44,7 +42,7 @@ class TubiTvIE(InfoExtractor): 'password': password, } payload = compat_urllib_parse.urlencode(form_data).encode('utf-8') - request = compat_urllib_request.Request(self._LOGIN_URL, payload) + request = sanitized_Request(self._LOGIN_URL, payload) request.add_header('Content-Type', 'application/x-www-form-urlencoded') login_page = self._download_webpage( request, None, False, 'Wrong login info') 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..69882da63 100644 --- a/youtube_dl/extractor/twitch.py +++ b/youtube_dl/extractor/twitch.py @@ -11,14 +11,15 @@ from ..compat import ( compat_str, compat_urllib_parse, compat_urllib_parse_urlparse, - compat_urllib_request, compat_urlparse, ) from ..utils import ( + encode_dict, ExtractorError, int_or_none, parse_duration, parse_iso8601, + sanitized_Request, ) @@ -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): @@ -48,7 +48,7 @@ class TwitchBaseIE(InfoExtractor): for cookie in self._downloader.cookiejar: if cookie.name == 'api_token': headers['Twitch-Api-Token'] = cookie.value - request = compat_urllib_request.Request(url, headers=headers) + request = sanitized_Request(url, headers=headers) response = super(TwitchBaseIE, self)._download_json(request, video_id, note) self._handle_error(response) return 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) + request = sanitized_Request( + 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..a161f046b 100644 --- a/youtube_dl/extractor/twitter.py +++ b/youtube_dl/extractor/twitter.py @@ -1,28 +1,73 @@ +# coding: utf-8 from __future__ import unicode_literals import re 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, + sanitized_Request, ) 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', + }, + 'add_ie': ['Youtube'], + }, + { + 'url': 'https://twitter.com/i/cards/tfw/v1/665289828897005568', + 'md5': 'ab2745d0b0ce53319a534fccaa986439', + 'info_dict': { + 'id': 'iBb2x00UVlv', + 'ext': 'mp4', + 'upload_date': '20151113', + 'uploader_id': '1189339351084113920', + 'uploader': '@ArsenalTerje', + 'title': 'Vine by @ArsenalTerje', + }, + 'add_ie': ['Vine'], + } + ] def _real_extract(self, url): video_id = self._match_id(url) @@ -36,14 +81,28 @@ class TwitterCardIE(InfoExtractor): config = None formats = [] for user_agent in USER_AGENTS: - request = compat_urllib_request.Request(url) + request = sanitized_Request(url) 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')), + iframe_url = self._html_search_regex( + r'<iframe[^>]+src="((?:https?:)?//(?:www.youtube.com/embed/[^"]+|(?:www\.)?vine\.co/v/\w+/card))"', + webpage, 'video iframe', default=None) + if iframe_url: + return self.url_result(iframe_url) + + 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 +129,100 @@ 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/udemy.py b/youtube_dl/extractor/udemy.py index 365d8b4bf..825172806 100644 --- a/youtube_dl/extractor/udemy.py +++ b/youtube_dl/extractor/udemy.py @@ -9,6 +9,7 @@ from ..compat import ( ) from ..utils import ( ExtractorError, + sanitized_Request, ) @@ -58,7 +59,7 @@ class UdemyIE(InfoExtractor): for header, value in headers.items(): url_or_request.add_header(header, value) else: - url_or_request = compat_urllib_request.Request(url_or_request, headers=headers) + url_or_request = sanitized_Request(url_or_request, headers=headers) response = super(UdemyIE, self)._download_json(url_or_request, video_id, note) self._handle_error(response) @@ -89,7 +90,7 @@ class UdemyIE(InfoExtractor): 'password': password.encode('utf-8'), }) - request = compat_urllib_request.Request( + request = sanitized_Request( self._LOGIN_URL, compat_urllib_parse.urlencode(login_form).encode('utf-8')) request.add_header('Referer', self._ORIGIN_URL) request.add_header('Origin', self._ORIGIN_URL) diff --git a/youtube_dl/extractor/udn.py b/youtube_dl/extractor/udn.py index 2151f8338..ee35b7227 100644 --- a/youtube_dl/extractor/udn.py +++ b/youtube_dl/extractor/udn.py @@ -12,7 +12,8 @@ from ..compat import compat_urlparse class UDNEmbedIE(InfoExtractor): IE_DESC = '聯合影音' - _VALID_URL = r'https?://video\.udn\.com/(?:embed|play)/news/(?P<id>\d+)' + _PROTOCOL_RELATIVE_VALID_URL = r'//video\.udn\.com/(?:embed|play)/news/(?P<id>\d+)' + _VALID_URL = r'https?:' + _PROTOCOL_RELATIVE_VALID_URL _TESTS = [{ 'url': 'http://video.udn.com/embed/news/300040', 'md5': 'de06b4c90b042c128395a88f0384817e', 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/vbox7.py b/youtube_dl/extractor/vbox7.py index 722eb5236..1e740fbe6 100644 --- a/youtube_dl/extractor/vbox7.py +++ b/youtube_dl/extractor/vbox7.py @@ -4,11 +4,11 @@ from __future__ import unicode_literals from .common import InfoExtractor from ..compat import ( compat_urllib_parse, - compat_urllib_request, compat_urlparse, ) from ..utils import ( ExtractorError, + sanitized_Request, ) @@ -49,7 +49,7 @@ class Vbox7IE(InfoExtractor): info_url = "http://vbox7.com/play/magare.do" data = compat_urllib_parse.urlencode({'as3': '1', 'vid': video_id}) - info_request = compat_urllib_request.Request(info_url, data) + info_request = sanitized_Request(info_url, data) info_request.add_header('Content-Type', 'application/x-www-form-urlencoded') info_response = self._download_webpage(info_request, video_id, 'Downloading info webpage') if info_response is None: diff --git a/youtube_dl/extractor/veoh.py b/youtube_dl/extractor/veoh.py index 01e258e32..9633f7ffe 100644 --- a/youtube_dl/extractor/veoh.py +++ b/youtube_dl/extractor/veoh.py @@ -4,12 +4,10 @@ import re import json from .common import InfoExtractor -from ..compat import ( - compat_urllib_request, -) from ..utils import ( int_or_none, ExtractorError, + sanitized_Request, ) @@ -110,7 +108,7 @@ class VeohIE(InfoExtractor): if 'class="adultwarning-container"' in webpage: self.report_age_confirmation() age_limit = 18 - request = compat_urllib_request.Request(url) + request = sanitized_Request(url) request.add_header('Cookie', 'confirmedAdult=true') webpage = self._download_webpage(request, video_id) diff --git a/youtube_dl/extractor/vessel.py b/youtube_dl/extractor/vessel.py index 3c8d2a943..1a0ff3395 100644 --- a/youtube_dl/extractor/vessel.py +++ b/youtube_dl/extractor/vessel.py @@ -4,10 +4,10 @@ from __future__ import unicode_literals import json from .common import InfoExtractor -from ..compat import compat_urllib_request from ..utils import ( ExtractorError, parse_iso8601, + sanitized_Request, ) @@ -33,7 +33,7 @@ class VesselIE(InfoExtractor): @staticmethod def make_json_request(url, data): payload = json.dumps(data).encode('utf-8') - req = compat_urllib_request.Request(url, payload) + req = sanitized_Request(url, payload) req.add_header('Content-Type', 'application/json; charset=utf-8') return req diff --git a/youtube_dl/extractor/vevo.py b/youtube_dl/extractor/vevo.py index c17094f81..571289421 100644 --- a/youtube_dl/extractor/vevo.py +++ b/youtube_dl/extractor/vevo.py @@ -1,15 +1,13 @@ from __future__ import unicode_literals import re -import xml.etree.ElementTree from .common import InfoExtractor -from ..compat import ( - compat_urllib_request, -) +from ..compat import compat_etree_fromstring from ..utils import ( ExtractorError, int_or_none, + sanitized_Request, ) @@ -73,7 +71,7 @@ class VevoIE(InfoExtractor): _SMIL_BASE_URL = 'http://smil.lvl3.vevo.com/' def _real_initialize(self): - req = compat_urllib_request.Request( + req = sanitized_Request( 'http://www.vevo.com/auth', data=b'') webpage = self._download_webpage( req, None, @@ -97,7 +95,7 @@ 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 +112,7 @@ 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/viddler.py b/youtube_dl/extractor/viddler.py index 8516a2940..40ffbad2a 100644 --- a/youtube_dl/extractor/viddler.py +++ b/youtube_dl/extractor/viddler.py @@ -4,9 +4,7 @@ from .common import InfoExtractor from ..utils import ( float_or_none, int_or_none, -) -from ..compat import ( - compat_urllib_request + sanitized_Request, ) @@ -65,7 +63,7 @@ class ViddlerIE(InfoExtractor): 'http://api.viddler.com/api/v2/viddler.videos.getPlaybackDetails.json?video_id=%s&key=v0vhrt7bg2xq1vyxhkct' % video_id) headers = {'Referer': 'http://static.cdn-ec.viddler.com/js/arpeggio/v2/embed.html'} - request = compat_urllib_request.Request(json_url, None, headers) + request = sanitized_Request(json_url, None, headers) data = self._download_json(request, video_id)['video'] formats = [] 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 649ac9433..000000000 --- a/youtube_dl/extractor/videolecturesnet.py +++ /dev/null @@ -1,82 +0,0 @@ -from __future__ import unicode_literals - -import re - -from .common import InfoExtractor -from ..compat import ( - compat_HTTPError, - compat_urlparse, -) -from ..utils import ( - ExtractorError, - parse_duration, -) - - -class VideoLecturesNetIE(InfoExtractor): - _VALID_URL = r'http://(?:www\.)?videolectures\.net/(?P<id>[^/#?]+)/*(?:[#?].*)?$' - IE_NAME = 'videolectures.net' - - _TESTS = [{ - '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', - }, - }, { - # video with invalid direct format links (HTTP 403) - 'url': 'http://videolectures.net/russir2010_filippova_nlp/', - 'info_dict': { - 'id': 'russir2010_filippova_nlp', - 'ext': 'flv', - 'title': 'NLP at Google', - 'description': 'md5:fc7a6d9bf0302d7cc0e53f7ca23747b3', - 'duration': 5352, - 'thumbnail': 're:http://.*\.jpg', - }, - 'params': { - # rtmp download - 'skip_download': True, - }, - }, { - 'url': 'http://videolectures.net/deeplearning2015_montreal/', - 'info_dict': { - 'id': 'deeplearning2015_montreal', - 'title': 'Deep Learning Summer School, Montreal 2015', - 'description': 'md5:90121a40cc6926df1bf04dcd8563ed3b', - }, - 'playlist_count': 30, - }] - - def _real_extract(self, url): - video_id = self._match_id(url) - - smil_url = 'http://videolectures.net/%s/video/1/smil.xml' % video_id - - try: - smil = self._download_smil(smil_url, video_id) - except ExtractorError as e: - if isinstance(e.cause, compat_HTTPError) and e.cause.code == 404: - # Probably a playlist - webpage = self._download_webpage(url, video_id) - entries = [ - self.url_result(compat_urlparse.urljoin(url, video_url), 'VideoLecturesNet') - for _, video_url in re.findall(r'<a[^>]+href=(["\'])(.+?)\1[^>]+id=["\']lec=\d+', webpage)] - playlist_title = self._html_search_meta('title', webpage, 'title', fatal=True) - playlist_description = self._html_search_meta('description', webpage, 'description') - return self.playlist_result(entries, video_id, playlist_title, playlist_description) - - info = self._parse_smil(smil, smil_url, video_id) - - info['id'] = video_id - - switch = smil.find('.//switch') - if switch is not None: - info['duration'] = parse_duration(switch.attrib.get('dur')) - - return info diff --git a/youtube_dl/extractor/videomega.py b/youtube_dl/extractor/videomega.py index 78ff6310a..87aca327b 100644 --- a/youtube_dl/extractor/videomega.py +++ b/youtube_dl/extractor/videomega.py @@ -4,7 +4,7 @@ from __future__ import unicode_literals import re from .common import InfoExtractor -from ..compat import compat_urllib_request +from ..utils import sanitized_Request class VideoMegaIE(InfoExtractor): @@ -30,7 +30,7 @@ class VideoMegaIE(InfoExtractor): video_id = self._match_id(url) iframe_url = 'http://videomega.tv/cdn.php?ref=%s' % video_id - req = compat_urllib_request.Request(iframe_url) + req = sanitized_Request(iframe_url) req.add_header('Referer', url) req.add_header('Cookie', 'noadvtday=0') webpage = self._download_webpage(req, video_id) diff --git a/youtube_dl/extractor/vidme.py b/youtube_dl/extractor/vidme.py index 078d283b2..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): @@ -114,6 +147,17 @@ 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'], @@ -121,6 +165,14 @@ class VidmeIE(InfoExtractor): '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'] @@ -137,7 +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 632e57fb4..185b1c119 100644 --- a/youtube_dl/extractor/viewster.py +++ b/youtube_dl/extractor/viewster.py @@ -4,7 +4,6 @@ from __future__ import unicode_literals from .common import InfoExtractor from ..compat import ( compat_HTTPError, - compat_urllib_request, compat_urllib_parse, compat_urllib_parse_unquote, ) @@ -13,6 +12,7 @@ from ..utils import ( ExtractorError, int_or_none, parse_iso8601, + sanitized_Request, HEADRequest, ) @@ -76,7 +76,7 @@ class ViewsterIE(InfoExtractor): _ACCEPT_HEADER = 'application/json, text/javascript, */*; q=0.01' def _download_json(self, url, video_id, note='Downloading JSON metadata', fatal=True): - request = compat_urllib_request.Request(url) + request = sanitized_Request(url) request.add_header('Accept', self._ACCEPT_HEADER) request.add_header('Auth-token', self._AUTH_TOKEN) return super(ViewsterIE, self)._download_json(request, video_id, note, fatal=fatal) @@ -131,10 +131,11 @@ 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: format_id = media.get('Bitrate') f = { 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/viki.py b/youtube_dl/extractor/viki.py index ddbd395c8..a63c23617 100644 --- a/youtube_dl/extractor/viki.py +++ b/youtube_dl/extractor/viki.py @@ -7,14 +7,14 @@ import hmac import hashlib import itertools +from .common import InfoExtractor from ..utils import ( ExtractorError, int_or_none, parse_age_limit, parse_iso8601, + sanitized_Request, ) -from ..compat import compat_urllib_request -from .common import InfoExtractor class VikiBaseIE(InfoExtractor): @@ -43,7 +43,7 @@ class VikiBaseIE(InfoExtractor): hashlib.sha1 ).hexdigest() url = self._API_URL_TEMPLATE % (query, sig) - return compat_urllib_request.Request( + return sanitized_Request( url, json.dumps(post_data).encode('utf-8')) if post_data else url def _call_api(self, path, video_id, note, timestamp=None, post_data=None): diff --git a/youtube_dl/extractor/vimeo.py b/youtube_dl/extractor/vimeo.py index 50df79ca1..f392ccf1c 100644 --- a/youtube_dl/extractor/vimeo.py +++ b/youtube_dl/extractor/vimeo.py @@ -8,15 +8,15 @@ 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, RegexNotFoundError, + sanitized_Request, smuggle_url, std_headers, unified_strdate, @@ -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 = sanitized_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 +184,15 @@ class VimeoIE(VimeoBaseInfoExtractor): 'uploader_id': 'user28849593', }, }, + { + 'url': 'https://vimeo.com/109815029', + 'note': 'Video not completely processed, "failed" seed status', + 'only_matching': True, + }, + { + 'url': 'https://vimeo.com/groups/travelhd/videos/22439234', + 'only_matching': True, + }, ] @staticmethod @@ -198,17 +214,18 @@ 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 = sanitized_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') @@ -217,9 +234,9 @@ 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 = sanitized_Request(pass_url, data) password_request.add_header('Content-Type', 'application/x-www-form-urlencoded') return self._download_json( password_request, video_id, @@ -248,7 +265,7 @@ class VimeoIE(VimeoBaseInfoExtractor): url = 'https://vimeo.com/' + video_id # Retrieve video webpage to extract further information - request = compat_urllib_request.Request(url, None, headers) + request = sanitized_Request(url, None, headers) try: webpage = self._download_webpage(request, video_id) except ExtractorError as ee: @@ -268,20 +285,30 @@ 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 +391,29 @@ 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') @@ -459,23 +474,23 @@ 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 = sanitized_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, 'Verifying the password', 'Wrong password') - def _extract_videos(self, list_id, base_url): - video_ids = [] + def _title_and_entries(self, list_id, base_url): for pagenum in itertools.count(1): page_url = self._page_url(base_url, pagenum) webpage = self._download_webpage( @@ -484,18 +499,18 @@ class VimeoChannelIE(VimeoBaseInfoExtractor): if pagenum == 1: webpage = self._login_list_password(page_url, list_id, webpage) + yield self._extract_list_title(webpage) + + for video_id in re.findall(r'id="clip_(\d+?)"', webpage): + yield self.url_result('https://vimeo.com/%s' % video_id, 'Vimeo') - video_ids.extend(re.findall(r'id="clip_(\d+?)"', webpage)) if re.search(self._MORE_PAGES_INDICATOR, webpage, re.DOTALL) is None: break - entries = [self.url_result('https://vimeo.com/%s' % video_id, 'Vimeo') - for video_id in video_ids] - return {'_type': 'playlist', - 'id': list_id, - 'title': self._extract_list_title(webpage), - 'entries': entries, - } + def _extract_videos(self, list_id, base_url): + title_and_entries = self._title_and_entries(list_id, base_url) + list_title = next(title_and_entries) + return self.playlist_result(title_and_entries, list_id, list_title) def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) @@ -556,7 +571,7 @@ class VimeoAlbumIE(VimeoChannelIE): class VimeoGroupsIE(VimeoAlbumIE): IE_NAME = 'vimeo:group' - _VALID_URL = r'https://vimeo\.com/groups/(?P<name>[^/]+)' + _VALID_URL = r'https://vimeo\.com/groups/(?P<name>[^/]+)(?:/(?!videos?/\d+)|$)' _TESTS = [{ 'url': 'https://vimeo.com/groups/rolexawards', 'info_dict': { @@ -625,7 +640,7 @@ class VimeoWatchLaterIE(VimeoChannelIE): def _page_url(self, base_url, pagenum): url = '%s/page:%d/' % (base_url, pagenum) - request = compat_urllib_request.Request(url) + request = sanitized_Request(url) # Set the header to get a partial html page with the ids, # the normal page doesn't contain them. request.add_header('X-Requested-With', 'XMLHttpRequest') 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 765e9e6fd..d99a42a9f 100644 --- a/youtube_dl/extractor/vk.py +++ b/youtube_dl/extractor/vk.py @@ -8,11 +8,11 @@ from .common import InfoExtractor from ..compat import ( compat_str, compat_urllib_parse, - compat_urllib_request, ) from ..utils import ( ExtractorError, orderedSet, + sanitized_Request, str_to_int, unescapeHTML, unified_strdate, @@ -182,7 +182,7 @@ class VKIE(InfoExtractor): 'pass': password.encode('cp1251'), }) - request = compat_urllib_request.Request( + request = sanitized_Request( 'https://login.vk.com/?act=login', compat_urllib_parse.urlencode(login_form).encode('utf-8')) login_page = self._download_webpage( @@ -281,9 +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/vodlocker.py b/youtube_dl/extractor/vodlocker.py index ccf1928b5..be0a2780f 100644 --- a/youtube_dl/extractor/vodlocker.py +++ b/youtube_dl/extractor/vodlocker.py @@ -2,10 +2,8 @@ from __future__ import unicode_literals from .common import InfoExtractor -from ..compat import ( - compat_urllib_parse, - compat_urllib_request, -) +from ..compat import compat_urllib_parse +from ..utils import sanitized_Request class VodlockerIE(InfoExtractor): @@ -31,7 +29,7 @@ class VodlockerIE(InfoExtractor): if fields['op'] == 'download1': self._sleep(3, video_id) # they do detect when requests happen too fast! post = compat_urllib_parse.urlencode(fields) - req = compat_urllib_request.Request(url, post) + req = sanitized_Request(url, post) req.add_header('Content-type', 'application/x-www-form-urlencoded') webpage = self._download_webpage( req, video_id, 'Downloading video page') diff --git a/youtube_dl/extractor/voicerepublic.py b/youtube_dl/extractor/voicerepublic.py index 254383d6c..93d15a556 100644 --- a/youtube_dl/extractor/voicerepublic.py +++ b/youtube_dl/extractor/voicerepublic.py @@ -3,14 +3,12 @@ from __future__ import unicode_literals import re from .common import InfoExtractor -from ..compat import ( - compat_urllib_request, - compat_urlparse, -) +from ..compat import compat_urlparse from ..utils import ( ExtractorError, determine_ext, int_or_none, + sanitized_Request, ) @@ -37,7 +35,7 @@ class VoiceRepublicIE(InfoExtractor): def _real_extract(self, url): display_id = self._match_id(url) - req = compat_urllib_request.Request( + req = sanitized_Request( compat_urlparse.urljoin(url, '/talks/%s' % display_id)) # Older versions of Firefox get redirected to an "upgrade browser" page req.add_header('User-Agent', 'youtube-dl') diff --git a/youtube_dl/extractor/wistia.py b/youtube_dl/extractor/wistia.py index 13a079151..fdb16d91c 100644 --- a/youtube_dl/extractor/wistia.py +++ b/youtube_dl/extractor/wistia.py @@ -1,8 +1,10 @@ from __future__ import unicode_literals from .common import InfoExtractor -from ..compat import compat_urllib_request -from ..utils import ExtractorError +from ..utils import ( + ExtractorError, + sanitized_Request, +) class WistiaIE(InfoExtractor): @@ -23,7 +25,7 @@ class WistiaIE(InfoExtractor): def _real_extract(self, url): video_id = self._match_id(url) - request = compat_urllib_request.Request(self._API_URL.format(video_id)) + request = sanitized_Request(self._API_URL.format(video_id)) request.add_header('Referer', url) # Some videos require this. data_json = self._download_json(request, video_id) if data_json.get('error'): 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..a3236e66c 100644 --- a/youtube_dl/extractor/gorillavid.py +++ b/youtube_dl/extractor/xfileshare.py @@ -1,25 +1,23 @@ -# -*- coding: utf-8 -*- +# coding: utf-8 from __future__ import unicode_literals import re from .common import InfoExtractor -from ..compat import ( - compat_urllib_parse, - compat_urllib_request, -) +from ..compat import compat_urllib_parse from ..utils import ( ExtractorError, encode_dict, int_or_none, + sanitized_Request, ) -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 +74,13 @@ 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): @@ -99,18 +104,23 @@ class GorillaVidIE(InfoExtractor): post = compat_urllib_parse.urlencode(encode_dict(fields)) - req = compat_urllib_request.Request(url, post) + req = sanitized_Request(url, post) req.add_header('Content-type', 'application/x-www-form-urlencoded') 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/xtube.py b/youtube_dl/extractor/xtube.py index 779e4f46a..a1fe24050 100644 --- a/youtube_dl/extractor/xtube.py +++ b/youtube_dl/extractor/xtube.py @@ -3,12 +3,10 @@ from __future__ import unicode_literals import re from .common import InfoExtractor -from ..compat import ( - compat_urllib_request, - compat_urllib_parse_unquote, -) +from ..compat import compat_urllib_parse_unquote from ..utils import ( parse_duration, + sanitized_Request, str_to_int, ) @@ -32,7 +30,7 @@ class XTubeIE(InfoExtractor): def _real_extract(self, url): video_id = self._match_id(url) - req = compat_urllib_request.Request(url) + req = sanitized_Request(url) req.add_header('Cookie', 'age_verified=1') webpage = self._download_webpage(req, video_id) diff --git a/youtube_dl/extractor/xvideos.py b/youtube_dl/extractor/xvideos.py index 5dcf2fdd1..710ad5041 100644 --- a/youtube_dl/extractor/xvideos.py +++ b/youtube_dl/extractor/xvideos.py @@ -3,14 +3,12 @@ from __future__ import unicode_literals import re from .common import InfoExtractor -from ..compat import ( - compat_urllib_parse_unquote, - compat_urllib_request, -) +from ..compat import compat_urllib_parse_unquote from ..utils import ( clean_html, ExtractorError, determine_ext, + sanitized_Request, ) @@ -48,7 +46,7 @@ class XVideosIE(InfoExtractor): 'url': video_url, }] - android_req = compat_urllib_request.Request(url) + android_req = sanitized_Request(url) android_req.add_header('User-Agent', self._ANDROID_USER_AGENT) android_webpage = self._download_webpage(android_req, video_id, fatal=False) diff --git a/youtube_dl/extractor/yandexmusic.py b/youtube_dl/extractor/yandexmusic.py index 4098e4629..d3cc1a29f 100644 --- a/youtube_dl/extractor/yandexmusic.py +++ b/youtube_dl/extractor/yandexmusic.py @@ -8,11 +8,11 @@ from .common import InfoExtractor from ..compat import ( compat_str, compat_urllib_parse, - compat_urllib_request, ) from ..utils import ( int_or_none, float_or_none, + sanitized_Request, ) @@ -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): @@ -147,7 +154,7 @@ class YandexMusicPlaylistIE(YandexMusicPlaylistBaseIE): if len(tracks) < len(track_ids): present_track_ids = set([compat_str(track['id']) for track in tracks if track.get('id')]) missing_track_ids = set(map(compat_str, track_ids)) - set(present_track_ids) - request = compat_urllib_request.Request( + request = sanitized_Request( 'https://music.yandex.ru/handlers/track-entries.jsx', compat_urllib_parse.urlencode({ 'entries': ','.join(missing_track_ids), diff --git a/youtube_dl/extractor/youku.py b/youtube_dl/extractor/youku.py index 2e81d9223..69ecc837a 100644 --- a/youtube_dl/extractor/youku.py +++ b/youtube_dl/extractor/youku.py @@ -4,12 +4,13 @@ from __future__ import unicode_literals import base64 from .common import InfoExtractor -from ..utils import ExtractorError - from ..compat import ( compat_urllib_parse, compat_ord, - compat_urllib_request, +) +from ..utils import ( + ExtractorError, + sanitized_Request, ) @@ -187,7 +188,7 @@ class YoukuIE(InfoExtractor): video_id = self._match_id(url) def retrieve_data(req_url, note): - req = compat_urllib_request.Request(req_url) + req = sanitized_Request(req_url) cn_verification_proxy = self._downloader.params.get('cn_verification_proxy') if cn_verification_proxy: diff --git a/youtube_dl/extractor/youporn.py b/youtube_dl/extractor/youporn.py index 4ba7c36db..dd724085a 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 ..utils import ( - ExtractorError, + int_or_none, + sanitized_Request, + 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 = sanitized_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 b252e36e1..1c2420a33 100644 --- a/youtube_dl/extractor/youtube.py +++ b/youtube_dl/extractor/youtube.py @@ -20,7 +20,6 @@ from ..compat import ( compat_urllib_parse_unquote, compat_urllib_parse_unquote_plus, compat_urllib_parse_urlparse, - compat_urllib_request, compat_urlparse, compat_str, ) @@ -35,6 +34,7 @@ from ..utils import ( orderedSet, parse_duration, remove_start, + sanitized_Request, smuggle_url, str_to_int, unescapeHTML, @@ -114,7 +114,7 @@ class YoutubeBaseInfoExtractor(InfoExtractor): login_data = compat_urllib_parse.urlencode(encode_dict(login_form_strs)).encode('ascii') - req = compat_urllib_request.Request(self._LOGIN_URL, login_data) + req = sanitized_Request(self._LOGIN_URL, login_data) login_results = self._download_webpage( req, None, note='Logging in', errnote='unable to log in', fatal=False) @@ -147,7 +147,7 @@ class YoutubeBaseInfoExtractor(InfoExtractor): tfa_data = compat_urllib_parse.urlencode(encode_dict(tfa_form_strs)).encode('ascii') - tfa_req = compat_urllib_request.Request(self._TWOFACTOR_URL, tfa_data) + tfa_req = sanitized_Request(self._TWOFACTOR_URL, tfa_data) tfa_results = self._download_webpage( tfa_req, None, note='Submitting TFA code', errnote='unable to submit tfa', fatal=False) @@ -178,6 +178,69 @@ class YoutubeBaseInfoExtractor(InfoExtractor): return +class YoutubeEntryListBaseInfoExtractor(InfoExtractor): + # Extract entries from page with "Load more" button + def _entries(self, page, playlist_id): + more_widget_html = content_html = page + for page_num in itertools.count(1): + for entry in self._process_page(content_html): + yield entry + + 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'] + + +class YoutubePlaylistBaseInfoExtractor(YoutubeEntryListBaseInfoExtractor): + def _process_page(self, content): + for video_id, video_title in self.extract_videos_from_page(content): + yield self.url_result(video_id, 'Youtube', video_id, video_title) + + 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 YoutubePlaylistsBaseInfoExtractor(YoutubeEntryListBaseInfoExtractor): + def _process_page(self, content): + for playlist_id in re.findall(r'href="/?playlist\?list=(.+?)"', content): + yield self.url_result( + 'https://www.youtube.com/playlist?list=%s' % playlist_id, 'YoutubePlaylist') + + def _real_extract(self, url): + playlist_id = self._match_id(url) + webpage = self._download_webpage(url, playlist_id) + title = self._og_search_title(webpage, fatal=False) + return self.playlist_result(self._entries(webpage, playlist_id), playlist_id, title) + + class YoutubeIE(YoutubeBaseInfoExtractor): IE_DESC = 'YouTube.com' _VALID_URL = r"""(?x)^ @@ -363,7 +426,8 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'title': 'Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012', 'description': 'md5:09b78bd971f1e3e289601dfba15ca4f7', 'uploader': 'SET India', - 'uploader_id': 'setindia' + 'uploader_id': 'setindia', + 'age_limit': 18, } }, { @@ -500,7 +564,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'info_dict': { 'id': 'lqQg6PlCWgI', 'ext': 'mp4', - 'upload_date': '20120724', + 'upload_date': '20150827', 'uploader_id': 'olympic', 'description': 'HO09 - Women - GER-AUS - Hockey - 31 July 2012 - London 2012 Olympic Games', 'uploader': 'Olympics', @@ -628,7 +692,44 @@ class YoutubeIE(YoutubeBaseInfoExtractor): { 'url': 'http://vid.plus/FlRa-iH7PGw', 'only_matching': True, - } + }, + { + # Title with JS-like syntax "};" (see https://github.com/rg3/youtube-dl/issues/7468) + 'url': 'https://www.youtube.com/watch?v=lsguqyKfVQg', + 'info_dict': { + 'id': 'lsguqyKfVQg', + 'ext': 'mp4', + 'title': '{dark walk}; Loki/AC/Dishonored; collab w/Elflover21', + 'description': 'md5:8085699c11dc3f597ce0410b0dcbb34a', + 'upload_date': '20151119', + 'uploader_id': 'IronSoulElf', + 'uploader': 'IronSoulElf', + }, + 'params': { + 'skip_download': True, + }, + }, + { + # Tags with '};' (see https://github.com/rg3/youtube-dl/issues/7468) + 'url': 'https://www.youtube.com/watch?v=Ms7iBXnlUO8', + 'only_matching': True, + }, + { + # Video with yt:stretch=17:0 + 'url': 'https://www.youtube.com/watch?v=Q39EVAstoRM', + 'info_dict': { + 'id': 'Q39EVAstoRM', + 'ext': 'mp4', + 'title': 'Clash Of Clans#14 Dicas De Ataque Para CV 4', + 'description': 'md5:ee18a25c350637c8faff806845bddee9', + 'upload_date': '20151107', + 'uploader_id': 'UCCr7TALkRbo3EtFzETQF1LA', + 'uploader': 'CH GAMER DROID', + }, + 'params': { + 'skip_download': True, + }, + }, ] def __init__(self, *args, **kwargs): @@ -657,7 +758,7 @@ 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) @@ -812,16 +913,33 @@ class YoutubeIE(YoutubeBaseInfoExtractor): return {} return sub_lang_list + def _get_ytplayer_config(self, video_id, webpage): + patterns = ( + # User data may contain arbitrary character sequences that may affect + # JSON extraction with regex, e.g. when '};' is contained the second + # regex won't capture the whole JSON. Yet working around by trying more + # concrete regex first keeping in mind proper quoted string handling + # to be implemented in future that will replace this workaround (see + # https://github.com/rg3/youtube-dl/issues/7468, + # https://github.com/rg3/youtube-dl/pull/7599) + r';ytplayer\.config\s*=\s*({.+?});ytplayer', + r';ytplayer\.config\s*=\s*({.+?});', + ) + config = self._search_regex( + patterns, webpage, 'ytplayer.config', default=None) + if config: + return self._parse_json( + uppercase_escape(config), video_id, fatal=False) + def _get_automatic_captions(self, video_id, webpage): """We need the webpage for getting the captions url, pass it as an argument to speed up the process.""" self.to_screen('%s: Looking for automatic captions' % video_id) - mobj = re.search(r';ytplayer.config = ({.*?});', webpage) + player_config = self._get_ytplayer_config(video_id, webpage) err_msg = 'Couldn\'t find automatic captions for %s' % video_id - if mobj is None: + if not player_config: self._downloader.report_warning(err_msg) return {} - player_config = json.loads(mobj.group(1)) try: args = player_config['args'] caption_url = args['ttsurl'] @@ -1028,10 +1146,8 @@ class YoutubeIE(YoutubeBaseInfoExtractor): age_gate = False video_info = None # Try looking directly into the video webpage - mobj = re.search(r';ytplayer\.config\s*=\s*({.*?});', video_webpage) - if mobj: - json_code = uppercase_escape(mobj.group(1)) - ytplayer_config = json.loads(json_code) + ytplayer_config = self._get_ytplayer_config(video_id, video_webpage) + if ytplayer_config: args = ytplayer_config['args'] if args.get('url_encoded_fmt_stream_map'): # Convert to the same format returned by compat_parse_qs @@ -1061,6 +1177,17 @@ 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 +1413,7 @@ 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 @@ -1385,10 +1512,15 @@ class YoutubeIE(YoutubeBaseInfoExtractor): r'<meta\s+property="og:video:tag".*?content="yt:stretch=(?P<w>[0-9]+):(?P<h>[0-9]+)">', video_webpage) if stretched_m: - ratio = float(stretched_m.group('w')) / float(stretched_m.group('h')) - for f in formats: - if f.get('vcodec') != 'none': - f['stretched_ratio'] = ratio + w = float(stretched_m.group('w')) + h = float(stretched_m.group('h')) + # yt:stretch may hold invalid ratio data (e.g. for Q39EVAstoRM ratio is 17:0). + # We will only process correct ratios. + if w > 0 and h > 0: + ratio = w / h + for f in formats: + if f.get('vcodec') != 'none': + f['stretched_ratio'] = ratio self._sort_formats(formats) @@ -1419,7 +1551,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): } -class YoutubePlaylistIE(YoutubeBaseInfoExtractor): +class YoutubePlaylistIE(YoutubeBaseInfoExtractor, YoutubePlaylistBaseInfoExtractor): IE_DESC = 'YouTube.com playlists' _VALID_URL = r"""(?x)(?: (?:https?://)? @@ -1440,7 +1572,7 @@ 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 +1689,11 @@ 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>', + 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 +1719,31 @@ 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) @@ -1661,7 +1762,7 @@ class YoutubeChannelIE(InfoExtractor): 'channelId', channel_page, 'channel id', default=None) if not channel_playlist_id: channel_playlist_id = self._search_regex( - r'data-channel-external-id="([^"]+)"', + 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:] @@ -1685,29 +1786,7 @@ 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): @@ -1738,6 +1817,29 @@ class YoutubeUserIE(YoutubeChannelIE): return super(YoutubeUserIE, cls).suitable(url) +class YoutubeUserPlaylistsIE(YoutubePlaylistsBaseInfoExtractor): + IE_DESC = 'YouTube.com user playlists' + _VALID_URL = r'https?://(?:\w+\.)?youtube\.com/user/(?P<id>[^/]+)/playlists' + IE_NAME = 'youtube:user:playlists' + + _TESTS = [{ + 'url': 'http://www.youtube.com/user/ThirstForScience/playlists', + 'playlist_mincount': 4, + 'info_dict': { + 'id': 'ThirstForScience', + 'title': 'Thirst for Science', + }, + }, { + # with "Load more" button + 'url': 'http://www.youtube.com/user/igorkle1/playlists?view=1&sort=dd', + 'playlist_mincount': 70, + 'info_dict': { + 'id': 'igorkle1', + 'title': 'Игорь Клейнер', + }, + }] + + class YoutubeSearchIE(SearchInfoExtractor, YoutubePlaylistIE): IE_DESC = 'YouTube.com searches' # there doesn't appear to be a real limit, for example if you search for @@ -1833,7 +1935,7 @@ class YoutubeSearchURLIE(InfoExtractor): } -class YoutubeShowIE(InfoExtractor): +class YoutubeShowIE(YoutubePlaylistsBaseInfoExtractor): IE_DESC = 'YouTube.com (multi-season) shows' _VALID_URL = r'https?://www\.youtube\.com/show/(?P<id>[^?#]*)' IE_NAME = 'youtube:show' @@ -1847,26 +1949,9 @@ class YoutubeShowIE(InfoExtractor): }] def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - playlist_id = mobj.group('id') - webpage = self._download_webpage( - 'https://www.youtube.com/show/%s/playlists' % playlist_id, playlist_id, 'Downloading show webpage') - # There's one playlist for each season of the show - m_seasons = list(re.finditer(r'href="(/playlist\?list=.*?)"', webpage)) - self.to_screen('%s: Found %s seasons' % (playlist_id, len(m_seasons))) - entries = [ - self.url_result( - 'https://www.youtube.com' + season.group(1), 'YoutubePlaylist') - for season in m_seasons - ] - title = self._og_search_title(webpage, fatal=False) - - return { - '_type': 'playlist', - 'id': playlist_id, - 'title': title, - 'entries': entries, - } + playlist_id = self._match_id(url) + return super(YoutubeShowIE, self)._real_extract( + 'https://www.youtube.com/show/%s/playlists' % playlist_id) class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor): 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' |