diff options
Diffstat (limited to 'youtube_dl/extractor')
| -rw-r--r-- | youtube_dl/extractor/__init__.py | 4 | ||||
| -rw-r--r-- | youtube_dl/extractor/arte.py | 17 | ||||
| -rw-r--r-- | youtube_dl/extractor/bambuser.py | 80 | ||||
| -rw-r--r-- | youtube_dl/extractor/brightcove.py | 4 | ||||
| -rw-r--r-- | youtube_dl/extractor/common.py | 5 | ||||
| -rw-r--r-- | youtube_dl/extractor/dailymotion.py | 23 | ||||
| -rw-r--r-- | youtube_dl/extractor/exfm.py | 2 | ||||
| -rw-r--r-- | youtube_dl/extractor/extremetube.py | 50 | ||||
| -rw-r--r-- | youtube_dl/extractor/keezmovies.py | 2 | ||||
| -rw-r--r-- | youtube_dl/extractor/livestream.py | 10 | ||||
| -rw-r--r-- | youtube_dl/extractor/metacafe.py | 51 | ||||
| -rw-r--r-- | youtube_dl/extractor/mtv.py | 2 | ||||
| -rw-r--r-- | youtube_dl/extractor/myspace.py | 48 | ||||
| -rw-r--r-- | youtube_dl/extractor/vevo.py | 94 | ||||
| -rw-r--r-- | youtube_dl/extractor/vimeo.py | 15 | ||||
| -rw-r--r-- | youtube_dl/extractor/vk.py | 45 | ||||
| -rw-r--r-- | youtube_dl/extractor/youtube.py | 28 |
17 files changed, 401 insertions, 79 deletions
diff --git a/youtube_dl/extractor/__init__.py b/youtube_dl/extractor/__init__.py index 7efd097e4..8dad38a00 100644 --- a/youtube_dl/extractor/__init__.py +++ b/youtube_dl/extractor/__init__.py @@ -9,6 +9,7 @@ from .arte import ( ArteTVFutureIE, ) from .auengine import AUEngineIE +from .bambuser import BambuserIE, BambuserChannelIE from .bandcamp import BandcampIE from .bliptv import BlipTVIE, BlipTVUserIE from .bloomberg import BloombergIE @@ -39,6 +40,7 @@ from .ehow import EHowIE from .eighttracks import EightTracksIE from .escapist import EscapistIE from .exfm import ExfmIE +from .extremetube import ExtremeTubeIE from .facebook import FacebookIE from .faz import FazIE from .fktv import ( @@ -83,6 +85,7 @@ from .mit import TechTVMITIE, MITIE from .mixcloud import MixcloudIE from .mtv import MTVIE from .muzu import MuzuTVIE +from .myspace import MySpaceIE from .myspass import MySpassIE from .myvideo import MyVideoIE from .naver import NaverIE @@ -141,6 +144,7 @@ from .videofyme import VideofyMeIE from .videopremium import VideoPremiumIE from .vimeo import VimeoIE, VimeoChannelIE from .vine import VineIE +from .vk import VKIE from .wat import WatIE from .websurg import WeBSurgIE from .weibo import WeiboIE diff --git a/youtube_dl/extractor/arte.py b/youtube_dl/extractor/arte.py index d39b48951..e10c74c11 100644 --- a/youtube_dl/extractor/arte.py +++ b/youtube_dl/extractor/arte.py @@ -158,7 +158,9 @@ class ArteTVPlus7IE(InfoExtractor): 'thumbnail': player_info.get('programImage') or player_info.get('VTU', {}).get('IUR'), } - formats = player_info['VSR'].values() + all_formats = player_info['VSR'].values() + # Some formats use the m3u8 protocol + all_formats = list(filter(lambda f: f.get('videoFormat') != 'M3U8', all_formats)) def _match_lang(f): if f.get('versionCode') is None: return True @@ -170,11 +172,16 @@ class ArteTVPlus7IE(InfoExtractor): regexes = [r'VO?%s' % l, r'VO?.-ST%s' % l] return any(re.match(r, f['versionCode']) for r in regexes) # Some formats may not be in the same language as the url - formats = filter(_match_lang, formats) - # Some formats use the m3u8 protocol - formats = filter(lambda f: f.get('videoFormat') != 'M3U8', formats) - # We order the formats by quality + formats = filter(_match_lang, all_formats) formats = list(formats) # in python3 filter returns an iterator + if not formats: + # Some videos are only available in the 'Originalversion' + # they aren't tagged as being in French or German + if all(f['versionCode'] == 'VO' for f in all_formats): + formats = all_formats + else: + raise ExtractorError(u'The formats list is empty') + # We order the formats by quality if re.match(r'[A-Z]Q', formats[0]['quality']) is not None: sort_key = lambda f: ['HQ', 'MQ', 'EQ', 'SQ'].index(f['quality']) else: diff --git a/youtube_dl/extractor/bambuser.py b/youtube_dl/extractor/bambuser.py new file mode 100644 index 000000000..f3b36f473 --- /dev/null +++ b/youtube_dl/extractor/bambuser.py @@ -0,0 +1,80 @@ +import re +import json +import itertools + +from .common import InfoExtractor +from ..utils import ( + compat_urllib_request, +) + + +class BambuserIE(InfoExtractor): + IE_NAME = u'bambuser' + _VALID_URL = r'https?://bambuser\.com/v/(?P<id>\d+)' + _API_KEY = '005f64509e19a868399060af746a00aa' + + _TEST = { + u'url': u'http://bambuser.com/v/4050584', + u'md5': u'fba8f7693e48fd4e8641b3fd5539a641', + u'info_dict': { + u'id': u'4050584', + u'ext': u'flv', + u'title': u'Education engineering days - lightning talks', + u'duration': 3741, + u'uploader': u'pixelversity', + u'uploader_id': u'344706', + }, + } + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + video_id = mobj.group('id') + info_url = ('http://player-c.api.bambuser.com/getVideo.json?' + '&api_key=%s&vid=%s' % (self._API_KEY, video_id)) + info_json = self._download_webpage(info_url, video_id) + info = json.loads(info_json)['result'] + + return { + 'id': video_id, + 'title': info['title'], + 'url': info['url'], + 'thumbnail': info.get('preview'), + 'duration': int(info['length']), + 'view_count': int(info['views_total']), + 'uploader': info['username'], + 'uploader_id': info['uid'], + } + + +class BambuserChannelIE(InfoExtractor): + IE_NAME = u'bambuser:channel' + _VALID_URL = r'http://bambuser.com/channel/(?P<user>.*?)(?:/|#|\?|$)' + # The maximum number we can get with each request + _STEP = 50 + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + user = mobj.group('user') + urls = [] + last_id = '' + for i in itertools.count(1): + req_url = ('http://bambuser.com/xhr-api/index.php?username={user}' + '&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) + # Without setting this header, we wouldn't get any result + req.add_header('Referer', 'http://bambuser.com/channel/%s' % user) + info_json = self._download_webpage(req, user, + u'Downloading page %d' % i) + results = json.loads(info_json)['result'] + if len(results) == 0: + break + last_id = results[-1]['vid'] + urls.extend(self.url_result(v['page'], 'Bambuser') for v in results) + + return { + '_type': 'playlist', + 'title': user, + 'entries': urls, + } diff --git a/youtube_dl/extractor/brightcove.py b/youtube_dl/extractor/brightcove.py index 1392f382a..0d9b87a34 100644 --- a/youtube_dl/extractor/brightcove.py +++ b/youtube_dl/extractor/brightcove.py @@ -23,7 +23,7 @@ class BrightcoveIE(InfoExtractor): # From http://www.8tv.cat/8aldia/videos/xavier-sala-i-martin-aquesta-tarda-a-8-al-dia/ u'url': u'http://c.brightcove.com/services/viewer/htmlFederated?playerID=1654948606001&flashID=myExperience&%40videoPlayer=2371591881001', u'file': u'2371591881001.mp4', - u'md5': u'9e80619e0a94663f0bdc849b4566af19', + u'md5': u'8eccab865181d29ec2958f32a6a754f5', u'note': u'Test Brightcove downloads and detection in GenericIE', u'info_dict': { u'title': u'Xavier Sala i Martín: “Un banc que no presta és un banc zombi que no serveix per a res”', @@ -122,12 +122,10 @@ class BrightcoveIE(InfoExtractor): best_format = renditions[-1] info.update({ 'url': best_format['defaultURL'], - 'ext': 'mp4', }) elif video_info.get('FLVFullLengthURL') is not None: info.update({ 'url': video_info['FLVFullLengthURL'], - 'ext': 'flv', }) else: raise ExtractorError(u'Unable to extract video url for %s' % info['id']) diff --git a/youtube_dl/extractor/common.py b/youtube_dl/extractor/common.py index ce349fe20..e0ccba533 100644 --- a/youtube_dl/extractor/common.py +++ b/youtube_dl/extractor/common.py @@ -63,7 +63,7 @@ class InfoExtractor(object): * ext Will be calculated from url if missing * format A human-readable description of the format ("mp4 container with h264/opus"). - Calculated from the format_id, width, height + Calculated from the format_id, width, height. and format_note fields if missing. * format_id A short description of the format ("mp4_h264_opus" or "19") @@ -71,6 +71,9 @@ class InfoExtractor(object): ("3D" or "DASH video") * width Width of the video, if known * height Height of the video, if known + webpage_url: The url to the video webpage, if given to youtube-dl it + should allow to get the same result again. (It will be set + by YoutubeDL if it's missing) Unless mentioned otherwise, the fields should be Unicode strings. diff --git a/youtube_dl/extractor/dailymotion.py b/youtube_dl/extractor/dailymotion.py index 4c0488245..355b4ed0a 100644 --- a/youtube_dl/extractor/dailymotion.py +++ b/youtube_dl/extractor/dailymotion.py @@ -21,6 +21,7 @@ class DailymotionBaseInfoExtractor(InfoExtractor): """Build a request with the family filter disabled""" request = compat_urllib_request.Request(url) request.add_header('Cookie', 'family_filter=off') + request.add_header('Cookie', 'ff=off') return request class DailymotionIE(DailymotionBaseInfoExtractor, SubtitlesInfoExtractor): @@ -61,6 +62,18 @@ class DailymotionIE(DailymotionBaseInfoExtractor, SubtitlesInfoExtractor): }, u'skip': u'VEVO is only available in some countries', }, + # age-restricted video + { + u'url': u'http://www.dailymotion.com/video/xyh2zz_leanna-decker-cyber-girl-of-the-year-desires-nude-playboy-plus_redband', + u'file': u'xyh2zz.mp4', + u'md5': u'0d667a7b9cebecc3c89ee93099c4159d', + u'info_dict': { + u'title': 'Leanna Decker - Cyber Girl Of The Year Desires Nude [Playboy Plus]', + u'uploader': 'HotWaves1012', + u'age_limit': 18, + } + + } ] def _real_extract(self, url): @@ -90,7 +103,8 @@ class DailymotionIE(DailymotionBaseInfoExtractor, SubtitlesInfoExtractor): video_uploader = self._search_regex([r'(?im)<span class="owner[^\"]+?">[^<]+?<a [^>]+?>([^<]+?)</a>', # Looking for official user r'<(?:span|a) .*?rel="author".*?>([^<]+?)</'], - webpage, 'video uploader') + webpage, 'video uploader', fatal=False) + age_limit = self._rta_search(webpage) video_upload_date = None mobj = re.search(r'<div class="[^"]*uploaded_cont[^"]*" title="[^"]*">([0-9]{2})-([0-9]{2})-([0-9]{4})</div>', webpage) @@ -132,15 +146,16 @@ class DailymotionIE(DailymotionBaseInfoExtractor, SubtitlesInfoExtractor): self._list_available_subtitles(video_id) return - return [{ + return { 'id': video_id, 'formats': formats, 'uploader': video_uploader, 'upload_date': video_upload_date, 'title': self._og_search_title(webpage), 'subtitles': video_subtitles, - 'thumbnail': info['thumbnail_url'] - }] + 'thumbnail': info['thumbnail_url'], + 'age_limit': age_limit, + } def _get_available_subtitles(self, video_id): try: diff --git a/youtube_dl/extractor/exfm.py b/youtube_dl/extractor/exfm.py index c74556579..a51d79b08 100644 --- a/youtube_dl/extractor/exfm.py +++ b/youtube_dl/extractor/exfm.py @@ -21,6 +21,7 @@ class ExfmIE(InfoExtractor): u'description': u'Test House \"Love Is Not Enough\" (Extended Mix) DeadJournalist Exclusive', }, u'note': u'Soundcloud song', + u'skip': u'The site is down too often', }, { u'url': u'http://ex.fm/song/wddt8', @@ -30,6 +31,7 @@ class ExfmIE(InfoExtractor): u'title': u'Safe and Sound', u'uploader': u'Capital Cities', }, + u'skip': u'The site is down too often', }, ] diff --git a/youtube_dl/extractor/extremetube.py b/youtube_dl/extractor/extremetube.py new file mode 100644 index 000000000..0f1eec40f --- /dev/null +++ b/youtube_dl/extractor/extremetube.py @@ -0,0 +1,50 @@ +import os +import re + +from .common import InfoExtractor +from ..utils import ( + compat_urllib_parse_urlparse, + compat_urllib_request, + compat_urllib_parse, +) + +class ExtremeTubeIE(InfoExtractor): + _VALID_URL = r'^(?:https?://)?(?:www\.)?(?P<url>extremetube\.com/video/.+?(?P<videoid>[0-9]+))(?:[/?&]|$)' + _TEST = { + u'url': u'http://www.extremetube.com/video/music-video-14-british-euro-brit-european-cumshots-swallow-652431', + u'file': u'652431.mp4', + u'md5': u'1fb9228f5e3332ec8c057d6ac36f33e0', + u'info_dict': { + u"title": u"Music Video 14 british euro brit european cumshots swallow", + u"uploader": u"unknown", + u"age_limit": 18, + } + } + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + video_id = mobj.group('videoid') + url = 'http://www.' + mobj.group('url') + + req = compat_urllib_request.Request(url) + req.add_header('Cookie', 'age_verified=1') + webpage = self._download_webpage(req, video_id) + + video_title = self._html_search_regex(r'<h1 [^>]*?title="([^"]+)"[^>]*>\1<', webpage, u'title') + uploader = self._html_search_regex(r'>Posted by:(?=<)(?:\s|<[^>]*>)*(.+?)\|', webpage, u'uploader', fatal=False) + video_url = compat_urllib_parse.unquote(self._html_search_regex(r'video_url=(.+?)&', webpage, u'video_url')) + path = compat_urllib_parse_urlparse( video_url ).path + extension = os.path.splitext( path )[1][1:] + format = path.split('/')[5].split('_')[:2] + format = "-".join( format ) + + return { + 'id': video_id, + 'title': video_title, + 'uploader': uploader, + 'url': video_url, + 'ext': extension, + 'format': format, + 'format_id': format, + 'age_limit': 18, + } diff --git a/youtube_dl/extractor/keezmovies.py b/youtube_dl/extractor/keezmovies.py index 5e05900da..786924445 100644 --- a/youtube_dl/extractor/keezmovies.py +++ b/youtube_dl/extractor/keezmovies.py @@ -12,7 +12,7 @@ from ..aes import ( ) class KeezMoviesIE(InfoExtractor): - _VALID_URL = r'^(?:https?://)?(?:www\.)?(?P<url>keezmovies\.com/video/.+?(?P<videoid>[0-9]+))' + _VALID_URL = r'^(?:https?://)?(?:www\.)?(?P<url>keezmovies\.com/video/.+?(?P<videoid>[0-9]+))(?:[/?&]|$)' _TEST = { u'url': u'http://www.keezmovies.com/video/petite-asian-lady-mai-playing-in-bathtub-1214711', u'file': u'1214711.mp4', diff --git a/youtube_dl/extractor/livestream.py b/youtube_dl/extractor/livestream.py index d04da98c8..4531fd6ab 100644 --- a/youtube_dl/extractor/livestream.py +++ b/youtube_dl/extractor/livestream.py @@ -40,13 +40,9 @@ class LivestreamIE(InfoExtractor): if video_id is None: # This is an event page: - player = get_meta_content('twitter:player', webpage) - if player is None: - raise ExtractorError('Couldn\'t extract event api url') - api_url = player.replace('/player', '') - api_url = re.sub(r'^(https?://)(new\.)', r'\1api.\2', api_url) - info = json.loads(self._download_webpage(api_url, event_name, - u'Downloading event info')) + config_json = self._search_regex(r'window.config = ({.*?});', + webpage, u'window config') + info = json.loads(config_json)['event'] videos = [self._extract_video_info(video_data['data']) for video_data in info['feed']['data'] if video_data['type'] == u'video'] return self.playlist_result(videos, info['id'], info['full_name']) diff --git a/youtube_dl/extractor/metacafe.py b/youtube_dl/extractor/metacafe.py index 234b9e80f..91480ba87 100644 --- a/youtube_dl/extractor/metacafe.py +++ b/youtube_dl/extractor/metacafe.py @@ -20,7 +20,9 @@ class MetacafeIE(InfoExtractor): _DISCLAIMER = 'http://www.metacafe.com/family_filter/' _FILTER_POST = 'http://www.metacafe.com/f/index.php?inputType=filter&controllerGroup=user' IE_NAME = u'metacafe' - _TESTS = [{ + _TESTS = [ + # Youtube video + { u"add_ie": ["Youtube"], u"url": u"http://metacafe.com/watch/yt-_aUehQsCQtM/the_electric_company_short_i_pbs_kids_go/", u"file": u"_aUehQsCQtM.mp4", @@ -32,15 +34,42 @@ class MetacafeIE(InfoExtractor): u"uploader_id": u"PBS" } }, + # Normal metacafe video + { + u'url': u'http://www.metacafe.com/watch/11121940/news_stuff_you_wont_do_with_your_playstation_4/', + u'md5': u'6e0bca200eaad2552e6915ed6fd4d9ad', + u'info_dict': { + u'id': u'11121940', + u'ext': u'mp4', + u'title': u'News: Stuff You Won\'t Do with Your PlayStation 4', + u'uploader': u'ign', + u'description': u'Sony released a massive FAQ on the PlayStation Blog detailing the PS4\'s capabilities and limitations.', + }, + }, + # AnyClip video { u"url": u"http://www.metacafe.com/watch/an-dVVXnuY7Jh77J/the_andromeda_strain_1971_stop_the_bomb_part_3/", u"file": u"an-dVVXnuY7Jh77J.mp4", u"info_dict": { u"title": u"The Andromeda Strain (1971): Stop the Bomb Part 3", u"uploader": u"anyclip", - u"description": u"md5:38c711dd98f5bb87acf973d573442e67" - } - }] + u"description": u"md5:38c711dd98f5bb87acf973d573442e67", + }, + }, + # age-restricted video + { + u'url': u'http://www.metacafe.com/watch/5186653/bbc_internal_christmas_tape_79_uncensored_outtakes_etc/', + u'md5': u'98dde7c1a35d02178e8ab7560fe8bd09', + u'info_dict': { + u'id': u'5186653', + u'ext': u'mp4', + u'title': u'BBC INTERNAL Christmas Tape \'79 - UNCENSORED Outtakes, Etc.', + u'uploader': u'Dwayne Pipe', + u'description': u'md5:950bf4c581e2c059911fa3ffbe377e4b', + u'age_limit': 18, + }, + }, + ] def report_disclaimer(self): @@ -62,6 +91,7 @@ class MetacafeIE(InfoExtractor): 'submit': "Continue - I'm over 18", } request = compat_urllib_request.Request(self._FILTER_POST, compat_urllib_parse.urlencode(disclaimer_form)) + request.add_header('Content-Type', 'application/x-www-form-urlencoded') try: self.report_age_confirmation() compat_urllib_request.urlopen(request).read() @@ -83,7 +113,12 @@ class MetacafeIE(InfoExtractor): # Retrieve video webpage to extract further information req = compat_urllib_request.Request('http://www.metacafe.com/watch/%s/' % video_id) - req.headers['Cookie'] = 'flashVersion=0;' + + # AnyClip videos require the flashversion cookie so that we get the link + # to the mp4 file + mobj_an = re.match(r'^an-(.*?)$', video_id) + if mobj_an: + req.headers['Cookie'] = 'flashVersion=0;' webpage = self._download_webpage(req, video_id) # Extract URL, uploader and title from webpage @@ -125,6 +160,11 @@ class MetacafeIE(InfoExtractor): r'submitter=(.*?);|googletag\.pubads\(\)\.setTargeting\("(?:channel|submiter)","([^"]+)"\);', webpage, u'uploader nickname', fatal=False) + if re.search(r'"contentRating":"restricted"', webpage) is not None: + age_limit = 18 + else: + age_limit = 0 + return { '_type': 'video', 'id': video_id, @@ -134,4 +174,5 @@ class MetacafeIE(InfoExtractor): 'upload_date': None, 'title': video_title, 'ext': video_ext, + 'age_limit': age_limit, } diff --git a/youtube_dl/extractor/mtv.py b/youtube_dl/extractor/mtv.py index e520e2bb4..e96d3952c 100644 --- a/youtube_dl/extractor/mtv.py +++ b/youtube_dl/extractor/mtv.py @@ -80,6 +80,8 @@ class MTVIE(InfoExtractor): video_id = self._id_from_uri(uri) self.report_extraction(video_id) mediagen_url = itemdoc.find('%s/%s' % (_media_xml_tag('group'), _media_xml_tag('content'))).attrib['url'] + # Remove the templates, like &device={device} + mediagen_url = re.sub(r'&[^=]*?={.*?}(?=(&|$))', u'', mediagen_url) if 'acceptMethods' not in mediagen_url: mediagen_url += '&acceptMethods=fms' mediagen_page = self._download_webpage(mediagen_url, video_id, diff --git a/youtube_dl/extractor/myspace.py b/youtube_dl/extractor/myspace.py new file mode 100644 index 000000000..050f54a5a --- /dev/null +++ b/youtube_dl/extractor/myspace.py @@ -0,0 +1,48 @@ +import re +import json + +from .common import InfoExtractor +from ..utils import ( + compat_str, +) + + +class MySpaceIE(InfoExtractor): + _VALID_URL = r'https?://myspace\.com/([^/]+)/video/[^/]+/(?P<id>\d+)' + + _TEST = { + u'url': u'https://myspace.com/coldplay/video/viva-la-vida/100008689', + u'info_dict': { + u'id': u'100008689', + u'ext': u'flv', + u'title': u'Viva La Vida', + u'description': u'The official Viva La Vida video, directed by Hype Williams', + u'uploader': u'Coldplay', + u'uploader_id': u'coldplay', + }, + u'params': { + # rtmp download + u'skip_download': True, + }, + } + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + video_id = mobj.group('id') + webpage = self._download_webpage(url, video_id) + context = json.loads(self._search_regex(r'context = ({.*?});', webpage, + u'context')) + video = context['video'] + rtmp_url, play_path = video['streamUrl'].split(';', 1) + + return { + 'id': compat_str(video['mediaId']), + 'title': video['title'], + 'url': rtmp_url, + 'play_path': play_path, + 'ext': 'flv', + 'description': video['description'], + 'thumbnail': video['imageUrl'], + 'uploader': video['artistName'], + 'uploader_id': video['artistUsername'], + } diff --git a/youtube_dl/extractor/vevo.py b/youtube_dl/extractor/vevo.py index 1c1cc418d..3f6020f74 100644 --- a/youtube_dl/extractor/vevo.py +++ b/youtube_dl/extractor/vevo.py @@ -5,7 +5,7 @@ import datetime from .common import InfoExtractor from ..utils import ( - determine_ext, + compat_HTTPError, ExtractorError, ) @@ -16,26 +16,22 @@ class VevoIE(InfoExtractor): (currently used by MTVIE) """ _VALID_URL = r'((http://www.vevo.com/watch/.*?/.*?/)|(vevo:))(?P<id>.*?)(\?|$)' - _TEST = { + _TESTS = [{ u'url': u'http://www.vevo.com/watch/hurts/somebody-to-die-for/GB1101300280', u'file': u'GB1101300280.mp4', + u"md5": u"06bea460acb744eab74a9d7dcb4bfd61", u'info_dict': { u"upload_date": u"20130624", u"uploader": u"Hurts", u"title": u"Somebody to Die For", - u'duration': 230, + u"duration": 230, + u"width": 1920, + u"height": 1080, } - } + }] + _SMIL_BASE_URL = 'http://smil.lvl3.vevo.com/' - def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - video_id = mobj.group('id') - - json_url = 'http://videoplayer.vevo.com/VideoService/AuthenticateVideo?isrc=%s' % video_id - info_json = self._download_webpage(json_url, video_id, u'Downloading json info') - - self.report_extraction(video_id) - video_info = json.loads(info_json)['video'] + def _formats_from_json(self, video_info): last_version = {'version': -1} for version in video_info['videoVersions']: # These are the HTTP downloads, other types are for different manifests @@ -50,17 +46,74 @@ class VevoIE(InfoExtractor): # Already sorted from worst to best quality for rend in renditions.findall('rendition'): attr = rend.attrib - f_url = attr['url'] + format_note = '%(videoCodec)s@%(videoBitrate)4sk, %(audioCodec)s@%(audioBitrate)3sk' % attr formats.append({ - 'url': f_url, - 'ext': determine_ext(f_url), + 'url': attr['url'], + 'format_id': attr['name'], + 'format_note': format_note, 'height': int(attr['frameheight']), 'width': int(attr['frameWidth']), }) + return formats + + def _formats_from_smil(self, smil_xml): + formats = [] + smil_doc = xml.etree.ElementTree.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'] + m = re.match(r'''(?xi) + (?P<ext>[a-z0-9]+): + (?P<path> + [/a-z0-9]+ # The directory and main part of the URL + _(?P<cbr>[0-9]+)k + _(?P<width>[0-9]+)x(?P<height>[0-9]+) + _(?P<vcodec>[a-z0-9]+) + _(?P<vbr>[0-9]+) + _(?P<acodec>[a-z0-9]+) + _(?P<abr>[0-9]+) + \.[a-z0-9]+ # File extension + )''', src) + if not m: + continue - date_epoch = int(self._search_regex( - r'/Date\((\d+)\)/', video_info['launchDate'], u'launch date'))/1000 - upload_date = datetime.datetime.fromtimestamp(date_epoch) + format_url = self._SMIL_BASE_URL + m.group('path') + format_note = ('%(vcodec)s@%(vbr)4sk, %(acodec)s@%(abr)3sk' % + m.groupdict()) + formats.append({ + 'url': format_url, + 'format_id': u'SMIL_' + m.group('cbr'), + 'format_note': format_note, + 'ext': m.group('ext'), + 'width': int(m.group('width')), + 'height': int(m.group('height')), + }) + return formats + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + video_id = mobj.group('id') + + json_url = 'http://videoplayer.vevo.com/VideoService/AuthenticateVideo?isrc=%s' % video_id + info_json = self._download_webpage(json_url, video_id, u'Downloading json info') + video_info = json.loads(info_json)['video'] + + formats = self._formats_from_json(video_info) + try: + smil_url = '%s/Video/V2/VFILE/%s/%sr.smil' % ( + self._SMIL_BASE_URL, video_id, video_id.lower()) + smil_xml = self._download_webpage(smil_url, video_id, + u'Downloading SMIL info') + formats.extend(self._formats_from_smil(smil_xml)) + except ExtractorError as ee: + if not isinstance(ee.cause, compat_HTTPError): + raise + self._downloader.report_warning( + u'Cannot download SMIL information, falling back to JSON ..') + + timestamp_ms = int(self._search_regex( + r'/Date\((\d+)\)/', video_info['launchDate'], u'launch date')) + upload_date = datetime.datetime.fromtimestamp(timestamp_ms // 1000) info = { 'id': video_id, 'title': video_info['title'], @@ -71,7 +124,4 @@ class VevoIE(InfoExtractor): 'duration': video_info['duration'], } - # TODO: Remove when #980 has been merged - info.update(formats[-1]) - return info diff --git a/youtube_dl/extractor/vimeo.py b/youtube_dl/extractor/vimeo.py index b4dbcd2ee..62273fd33 100644 --- a/youtube_dl/extractor/vimeo.py +++ b/youtube_dl/extractor/vimeo.py @@ -20,14 +20,14 @@ class VimeoIE(InfoExtractor): """Information extractor for vimeo.com.""" # _VALID_URL matches Vimeo URLs - _VALID_URL = r'(?P<proto>https?://)?(?:(?:www|player)\.)?vimeo(?P<pro>pro)?\.com/(?:(?:(?:groups|album)/[^/]+)|(?:.*?)/)?(?P<direct_link>play_redirect_hls\?clip_id=)?(?:videos?/)?(?P<id>[0-9]+)/?(?:[?].*)?(?:#.*)?$' + _VALID_URL = r'(?P<proto>https?://)?(?:(?:www|(?P<player>player))\.)?vimeo(?P<pro>pro)?\.com/(?:(?:(?:groups|album)/[^/]+)|(?:.*?)/)?(?P<direct_link>play_redirect_hls\?clip_id=)?(?:videos?/)?(?P<id>[0-9]+)/?(?:[?].*)?(?:#.*)?$' _NETRC_MACHINE = 'vimeo' IE_NAME = u'vimeo' _TESTS = [ { u'url': u'http://vimeo.com/56015672#at=0', u'file': u'56015672.mp4', - u'md5': u'ae7a1d8b183758a0506b0622f37dfa14', + u'md5': u'8879b6cc097e987f02484baf890129e5', u'info_dict': { u"upload_date": u"20121220", u"description": u"This is a test case for youtube-dl.\nFor more information, see github.com/rg3/youtube-dl\nTest chars: \u2605 \" ' \u5e78 / \\ \u00e4 \u21ad \U0001d550", @@ -128,11 +128,9 @@ class VimeoIE(InfoExtractor): raise ExtractorError(u'Invalid URL: %s' % url) video_id = mobj.group('id') - if not mobj.group('proto'): - url = 'https://' + url - elif mobj.group('pro'): + if mobj.group('pro') or mobj.group('player'): url = 'http://player.vimeo.com/video/' + video_id - elif mobj.group('direct_link'): + else: url = 'https://vimeo.com/' + video_id # Retrieve video webpage to extract further information @@ -234,7 +232,7 @@ class VimeoIE(InfoExtractor): if len(formats) == 0: raise ExtractorError(u'No known codec found') - return [{ + return { 'id': video_id, 'uploader': video_uploader, 'uploader_id': video_uploader_id, @@ -243,7 +241,8 @@ class VimeoIE(InfoExtractor): 'thumbnail': video_thumbnail, 'description': video_description, 'formats': formats, - }] + 'webpage_url': url, + } class VimeoChannelIE(InfoExtractor): diff --git a/youtube_dl/extractor/vk.py b/youtube_dl/extractor/vk.py new file mode 100644 index 000000000..90d8a6d07 --- /dev/null +++ b/youtube_dl/extractor/vk.py @@ -0,0 +1,45 @@ +# encoding: utf-8 +import re +import json + +from .common import InfoExtractor +from ..utils import ( + compat_str, + unescapeHTML, +) + + +class VKIE(InfoExtractor): + IE_NAME = u'vk.com' + _VALID_URL = r'https?://vk\.com/(?:videos.*?\?.*?z=)?video(?P<id>.*?)(?:\?|%2F|$)' + + _TEST = { + u'url': u'http://vk.com/videos-77521?z=video-77521_162222515%2Fclub77521', + u'md5': u'0deae91935c54e00003c2a00646315f0', + u'info_dict': { + u'id': u'162222515', + u'ext': u'flv', + u'title': u'ProtivoGunz - Хуёвая песня', + u'uploader': u'Noize MC', + }, + } + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + video_id = mobj.group('id') + info_url = 'http://vk.com/al_video.php?act=show&al=1&video=%s' % video_id + info_page = self._download_webpage(info_url, video_id) + m_yt = re.search(r'src="(http://www.youtube.com/.*?)"', info_page) + if m_yt is not None: + self.to_screen(u'Youtube video detected') + return self.url_result(m_yt.group(1), 'Youtube') + vars_json = self._search_regex(r'var vars = ({.*?});', info_page, u'vars') + vars = json.loads(vars_json) + + return { + 'id': compat_str(vars['vid']), + 'url': vars['url240'], + 'title': unescapeHTML(vars['md_title']), + 'thumbnail': vars['jpg'], + 'uploader': vars['md_author'], + } diff --git a/youtube_dl/extractor/youtube.py b/youtube_dl/extractor/youtube.py index d05d0a8c1..6ddd6ef06 100644 --- a/youtube_dl/extractor/youtube.py +++ b/youtube_dl/extractor/youtube.py @@ -74,14 +74,8 @@ class YoutubeBaseInfoExtractor(InfoExtractor): self._downloader.report_warning(u'unable to fetch login page: %s' % compat_str(err)) return False - galx = None - dsh = None - match = re.search(re.compile(r'<input.+?name="GALX".+?value="(.+?)"', re.DOTALL), login_page) - if match: - galx = match.group(1) - match = re.search(re.compile(r'<input.+?name="dsh".+?value="(.+?)"', re.DOTALL), login_page) - if match: - dsh = match.group(1) + galx = self._search_regex(r'(?s)<input.+?name="GALX".+?value="(.+?)"', + login_page, u'Login GALX parameter') # Log in login_form_strs = { @@ -95,7 +89,6 @@ class YoutubeBaseInfoExtractor(InfoExtractor): u'checkConnection': u'', u'checkedDomains': u'youtube', u'dnConn': u'', - u'dsh': dsh, u'pstMsg': u'0', u'rmShown': u'1', u'secTok': u'', @@ -347,18 +340,6 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor): } }, { - u"url": u"http://www.youtube.com/watch?v=1ltcDfZMA3U", - u"file": u"1ltcDfZMA3U.mp4", - u"note": u"Test VEVO video (#897)", - u"info_dict": { - u"upload_date": u"20070518", - u"title": u"Maps - It Will Find You", - u"description": u"Music video by Maps performing It Will Find You.", - u"uploader": u"MuteUSA", - u"uploader_id": u"MuteUSA" - } - }, - { u"url": u"http://www.youtube.com/watch?v=UxxajLWwzqY", u"file": u"UxxajLWwzqY.mp4", u"note": u"Test generic use_cipher_signature video (#897)", @@ -1118,7 +1099,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor): 'lang': lang, 'v': video_id, 'fmt': self._downloader.params.get('subtitlesformat'), - 'name': l[0], + 'name': l[0].encode('utf-8'), }) url = u'http://www.youtube.com/api/timedtext?' + params sub_lang_list[lang] = url @@ -1504,7 +1485,8 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor): 'subtitles': video_subtitles, 'duration': video_duration, 'age_limit': 18 if age_gate else 0, - 'annotations': video_annotations + 'annotations': video_annotations, + 'webpage_url': 'https://www.youtube.com/watch?v=%s' % video_id, }) return results |