diff options
Diffstat (limited to 'youtube_dl')
36 files changed, 1090 insertions, 335 deletions
diff --git a/youtube_dl/FileDownloader.py b/youtube_dl/FileDownloader.py index 155895fe2..ea6b9d626 100644 --- a/youtube_dl/FileDownloader.py +++ b/youtube_dl/FileDownloader.py @@ -329,6 +329,35 @@ class FileDownloader(object): self.report_error(u'mplayer exited with code %d' % retval) return False + def _download_m3u8_with_ffmpeg(self, filename, url): + self.report_destination(filename) + tmpfilename = self.temp_name(filename) + + args = ['ffmpeg', '-y', '-i', url, '-f', 'mp4', tmpfilename] + # Check for ffmpeg first + try: + subprocess.call(['ffmpeg', '-h'], stdout=(open(os.path.devnull, 'w')), stderr=subprocess.STDOUT) + except (OSError, IOError): + self.report_error(u'm3u8 download detected but "%s" could not be run' % args[0] ) + return False + + retval = subprocess.call(args) + if retval == 0: + fsize = os.path.getsize(encodeFilename(tmpfilename)) + self.to_screen(u'\r[%s] %s bytes' % (args[0], fsize)) + self.try_rename(tmpfilename, filename) + self._hook_progress({ + 'downloaded_bytes': fsize, + 'total_bytes': fsize, + 'filename': filename, + 'status': 'finished', + }) + return True + else: + self.to_stderr(u"\n") + self.report_error(u'ffmpeg exited with code %d' % retval) + return False + def _do_download(self, filename, info_dict): url = info_dict['url'] @@ -354,6 +383,10 @@ class FileDownloader(object): if url.startswith('mms') or url.startswith('rtsp'): return self._download_with_mplayer(filename, url) + # m3u8 manifest are downloaded with ffmpeg + if determine_ext(url) == u'm3u8': + return self._download_m3u8_with_ffmpeg(filename, url) + tmpfilename = self.temp_name(filename) stream = None diff --git a/youtube_dl/PostProcessor.py b/youtube_dl/PostProcessor.py index 8c5e53991..fddf58606 100644 --- a/youtube_dl/PostProcessor.py +++ b/youtube_dl/PostProcessor.py @@ -100,7 +100,8 @@ class FFmpegExtractAudioPP(FFmpegPostProcessor): self._nopostoverwrites = nopostoverwrites def get_audio_codec(self, path): - if not self._exes['ffprobe'] and not self._exes['avprobe']: return None + if not self._exes['ffprobe'] and not self._exes['avprobe']: + raise PostProcessingError(u'ffprobe or avprobe not found. Please install one.') try: cmd = [self._exes['avprobe'] or self._exes['ffprobe'], '-show_streams', encodeFilename(self._ffmpeg_filename_argument(path))] handle = subprocess.Popen(cmd, stderr=compat_subprocess_get_DEVNULL(), stdout=subprocess.PIPE) @@ -208,7 +209,7 @@ class FFmpegExtractAudioPP(FFmpegPostProcessor): try: os.utime(encodeFilename(new_path), (time.time(), information['filetime'])) except: - self._downloader.to_stderr(u'WARNING: Cannot update utime of audio file') + self._downloader.report_warning(u'Cannot update utime of audio file') information['filepath'] = new_path return self._nopostoverwrites,information diff --git a/youtube_dl/YoutubeDL.py b/youtube_dl/YoutubeDL.py index c76f1118e..496866900 100644 --- a/youtube_dl/YoutubeDL.py +++ b/youtube_dl/YoutubeDL.py @@ -594,7 +594,7 @@ class YoutubeDL(object): # No clear decision yet, let IE decide keep_video = keep_video_wish except PostProcessingError as e: - self.to_stderr(u'ERROR: ' + e.msg) + self.report_error(e.msg) if keep_video is False and not self.params.get('keepvideo', False): try: self.to_screen(u'Deleting original file %s (pass -k to keep)' % filename) diff --git a/youtube_dl/__init__.py b/youtube_dl/__init__.py index 31427f7c5..bf040aacd 100644 --- a/youtube_dl/__init__.py +++ b/youtube_dl/__init__.py @@ -129,7 +129,7 @@ def parseOpts(overrideArguments=None): general.add_option('-v', '--version', action='version', help='print program version and exit') general.add_option('-U', '--update', - action='store_true', dest='update_self', help='update this program to latest version') + action='store_true', dest='update_self', help='update this program to latest version. Make sure that you have sufficient permissions (run with sudo if needed)') general.add_option('-i', '--ignore-errors', action='store_true', dest='ignoreerrors', help='continue on download errors', default=False) general.add_option('--dump-user-agent', @@ -398,6 +398,8 @@ def _real_main(argv=None): batchurls = batchfd.readlines() batchurls = [x.strip() for x in batchurls] batchurls = [x for x in batchurls if len(x) > 0 and not re.search(r'^[#/;]', x)] + if opts.verbose: + sys.stderr.write(u'[debug] Batch file urls: ' + repr(batchurls) + u'\n') except IOError: sys.exit(u'ERROR: batch file could not be read') all_urls = batchurls + args diff --git a/youtube_dl/extractor/__init__.py b/youtube_dl/extractor/__init__.py index 79868aa06..84c02c2ed 100644 --- a/youtube_dl/extractor/__init__.py +++ b/youtube_dl/extractor/__init__.py @@ -12,13 +12,14 @@ from .comedycentral import ComedyCentralIE from .condenast import CondeNastIE from .criterion import CriterionIE from .cspan import CSpanIE -from .dailymotion import DailymotionIE +from .dailymotion import DailymotionIE, DailymotionPlaylistIE from .depositfiles import DepositFilesIE from .dotsub import DotsubIE from .dreisat import DreiSatIE from .ehow import EHowIE from .eighttracks import EightTracksIE from .escapist import EscapistIE +from .exfm import ExfmIE from .facebook import FacebookIE from .flickr import FlickrIE from .freesound import FreesoundIE @@ -37,20 +38,25 @@ from .infoq import InfoQIE from .instagram import InstagramIE from .jukebox import JukeboxIE from .justintv import JustinTVIE +from .kankan import KankanIE from .keek import KeekIE from .liveleak import LiveLeakIE from .livestream import LivestreamIE from .metacafe import MetacafeIE from .mixcloud import MixcloudIE from .mtv import MTVIE +from .muzu import MuzuTVIE from .myspass import MySpassIE from .myvideo import MyVideoIE from .nba import NBAIE +from .ooyala import OoyalaIE from .photobucket import PhotobucketIE from .pornotube import PornotubeIE from .rbmaradio import RBMARadioIE from .redtube import RedTubeIE from .ringtv import RingTVIE +from .roxwel import RoxwelIE +from .sina import SinaIE from .soundcloud import SoundcloudIE, SoundcloudSetIE from .spiegel import SpiegelIE from .stanfordoc import StanfordOpenClassroomIE @@ -68,9 +74,12 @@ from .ustream import UstreamIE from .vbox7 import Vbox7IE from .veoh import VeohIE from .vevo import VevoIE -from .vimeo import VimeoIE +from .videofyme import VideofyMeIE +from .vimeo import VimeoIE, VimeoChannelIE from .vine import VineIE +from .c56 import C56IE from .wat import WatIE +from .weibo import WeiboIE from .wimp import WimpIE from .worldstarhiphop import WorldStarHipHopIE from .xhamster import XHamsterIE @@ -88,6 +97,9 @@ from .youtube import ( YoutubeChannelIE, YoutubeShowIE, YoutubeSubscriptionsIE, + YoutubeRecommendedIE, + YoutubeWatchLaterIE, + YoutubeFavouritesIE, ) from .zdf import ZDFIE diff --git a/youtube_dl/extractor/arte.py b/youtube_dl/extractor/arte.py index 993e30f7a..69b3b0ad7 100644 --- a/youtube_dl/extractor/arte.py +++ b/youtube_dl/extractor/arte.py @@ -17,13 +17,14 @@ class ArteTvIE(InfoExtractor): """ _EMISSION_URL = r'(?:http://)?www\.arte.tv/guide/(?P<lang>fr|de)/(?:(?:sendungen|emissions)/)?(?P<id>.*?)/(?P<name>.*?)(\?.*)?' _VIDEOS_URL = r'(?:http://)?videos.arte.tv/(?P<lang>fr|de)/.*-(?P<id>.*?).html' + _LIVEWEB_URL = r'(?:http://)?liveweb.arte.tv/(?P<lang>fr|de)/(?P<subpage>.+?)/(?P<name>.+)' _LIVE_URL = r'index-[0-9]+\.html$' IE_NAME = u'arte.tv' @classmethod def suitable(cls, url): - return any(re.match(regex, url) for regex in (cls._EMISSION_URL, cls._VIDEOS_URL)) + return any(re.match(regex, url) for regex in (cls._EMISSION_URL, cls._VIDEOS_URL, cls._LIVEWEB_URL)) # TODO implement Live Stream # from ..utils import compat_urllib_parse @@ -68,6 +69,12 @@ class ArteTvIE(InfoExtractor): lang = mobj.group('lang') return self._extract_video(url, id, lang) + mobj = re.match(self._LIVEWEB_URL, url) + if mobj is not None: + name = mobj.group('name') + lang = mobj.group('lang') + return self._extract_liveweb(url, name, lang) + if re.search(self._LIVE_URL, video_id) is not None: raise ExtractorError(u'Arte live streams are not yet supported, sorry') # self.extractLiveStream(url) @@ -85,7 +92,7 @@ class ArteTvIE(InfoExtractor): info_dict = {'id': player_info['VID'], 'title': player_info['VTI'], - 'description': player_info['VDE'], + 'description': player_info.get('VDE'), 'upload_date': unified_strdate(player_info['VDA'].split(' ')[0]), 'thumbnail': player_info['programImage'], 'ext': 'flv', @@ -98,12 +105,14 @@ class ArteTvIE(InfoExtractor): l = 'F' elif lang == 'de': l = 'A' - regexes = [r'VO?%s' % l, r'V%s-ST.' % l] + 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) # We order the formats by quality formats = sorted(formats, key=lambda f: int(f['height'])) + # Prefer videos without subtitles in the same language + formats = sorted(formats, key=lambda f: re.match(r'VO(F|A)-STM\1', f['versionCode']) is None) # Pick the best quality format_info = formats[-1] if format_info['mediaType'] == u'rtmp': @@ -144,3 +153,22 @@ class ArteTvIE(InfoExtractor): 'url': video_url, 'ext': 'flv', } + + def _extract_liveweb(self, url, name, lang): + """Extract form http://liveweb.arte.tv/""" + webpage = self._download_webpage(url, name) + video_id = self._search_regex(r'eventId=(\d+?)("|&)', webpage, u'event id') + config_xml = self._download_webpage('http://download.liveweb.arte.tv/o21/liveweb/events/event-%s.xml' % video_id, + video_id, u'Downloading information') + config_doc = xml.etree.ElementTree.fromstring(config_xml.encode('utf-8')) + event_doc = config_doc.find('event') + url_node = event_doc.find('video').find('urlHd') + if url_node is None: + url_node = video_doc.find('urlSd') + + return {'id': video_id, + 'title': event_doc.find('name%s' % lang.capitalize()).text, + 'url': url_node.text.replace('MP4', 'mp4'), + 'ext': 'flv', + 'thumbnail': self._og_search_thumbnail(webpage), + } diff --git a/youtube_dl/extractor/breakcom.py b/youtube_dl/extractor/breakcom.py index 34f555e89..53a898de3 100644 --- a/youtube_dl/extractor/breakcom.py +++ b/youtube_dl/extractor/breakcom.py @@ -1,6 +1,8 @@ import re +import json from .common import InfoExtractor +from ..utils import determine_ext class BreakIE(InfoExtractor): @@ -17,17 +19,20 @@ class BreakIE(InfoExtractor): def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) video_id = mobj.group(1).split("-")[-1] - webpage = self._download_webpage(url, video_id) - video_url = re.search(r"videoPath: '(.+?)',",webpage).group(1) - key = re.search(r"icon: '(.+?)',",webpage).group(1) - final_url = str(video_url)+"?"+str(key) - thumbnail_url = re.search(r"thumbnailURL: '(.+?)'",webpage).group(1) - title = re.search(r"sVidTitle: '(.+)',",webpage).group(1) - ext = video_url.split('.')[-1] + embed_url = 'http://www.break.com/embed/%s' % video_id + webpage = self._download_webpage(embed_url, video_id) + info_json = self._search_regex(r'var embedVars = ({.*?});', webpage, + u'info json', flags=re.DOTALL) + info = json.loads(info_json) + video_url = info['videoUri'] + m_youtube = re.search(r'(https?://www\.youtube\.com/watch\?v=.*)', video_url) + if m_youtube is not None: + return self.url_result(m_youtube.group(1), 'Youtube') + final_url = video_url + '?' + info['AuthToken'] return [{ 'id': video_id, 'url': final_url, - 'ext': ext, - 'title': title, - 'thumbnail': thumbnail_url, + 'ext': determine_ext(final_url), + 'title': info['contentName'], + 'thumbnail': info['thumbUri'], }] diff --git a/youtube_dl/extractor/c56.py b/youtube_dl/extractor/c56.py new file mode 100644 index 000000000..4c8a8af09 --- /dev/null +++ b/youtube_dl/extractor/c56.py @@ -0,0 +1,36 @@ +# coding: utf-8 + +import re +import json + +from .common import InfoExtractor +from ..utils import determine_ext + +class C56IE(InfoExtractor): + _VALID_URL = r'https?://((www|player)\.)?56\.com/(.+?/)?(v_|(play_album.+-))(?P<textid>.+?)\.(html|swf)' + IE_NAME = u'56.com' + + _TEST ={ + u'url': u'http://www.56.com/u39/v_OTM0NDA3MTY.html', + u'file': u'93440716.mp4', + u'md5': u'9dc07b5c8e978112a6441f9e75d2b59e', + u'info_dict': { + u'title': u'网事知多少 第32期：车怒', + }, + } + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url, flags=re.VERBOSE) + text_id = mobj.group('textid') + info_page = self._download_webpage('http://vxml.56.com/json/%s/' % text_id, + text_id, u'Downloading video info') + info = json.loads(info_page)['info'] + best_format = sorted(info['rfiles'], key=lambda f: int(f['filesize']))[-1] + video_url = best_format['url'] + + return {'id': info['vid'], + 'title': info['Subject'], + 'url': video_url, + 'ext': determine_ext(video_url), + 'thumbnail': info.get('bimg') or info.get('img'), + } diff --git a/youtube_dl/extractor/collegehumor.py b/youtube_dl/extractor/collegehumor.py index 7ae0972e5..30b9c7549 100644 --- a/youtube_dl/extractor/collegehumor.py +++ b/youtube_dl/extractor/collegehumor.py @@ -1,26 +1,26 @@ import re -import socket import xml.etree.ElementTree from .common import InfoExtractor from ..utils import ( - compat_http_client, - compat_str, - compat_urllib_error, compat_urllib_parse_urlparse, - compat_urllib_request, ExtractorError, ) class CollegeHumorIE(InfoExtractor): - _WORKING = False - _VALID_URL = r'^(?:https?://)?(?:www\.)?collegehumor\.com/video/(?P<videoid>[0-9]+)/(?P<shorttitle>.*)$' + _VALID_URL = r'^(?:https?://)?(?:www\.)?collegehumor\.com/(video|embed|e)/(?P<videoid>[0-9]+)/?(?P<shorttitle>.*)$' - def report_manifest(self, video_id): - """Report information extraction.""" - self.to_screen(u'%s: Downloading XML manifest' % video_id) + _TEST = { + u'url': u'http://www.collegehumor.com/video/6902724/comic-con-cosplay-catastrophe', + u'file': u'6902724.mp4', + u'md5': u'1264c12ad95dca142a9f0bf7968105a0', + u'info_dict': { + u'title': u'Comic-Con Cosplay Catastrophe', + u'description': u'Fans get creative this year at San Diego. Too creative. And yes, that\'s really Joss Whedon.', + }, + } def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) @@ -36,14 +36,16 @@ class CollegeHumorIE(InfoExtractor): self.report_extraction(video_id) xmlUrl = 'http://www.collegehumor.com/moogaloop/video/' + video_id - try: - metaXml = compat_urllib_request.urlopen(xmlUrl).read() - except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: - raise ExtractorError(u'Unable to download video info XML: %s' % compat_str(err)) + metaXml = self._download_webpage(xmlUrl, video_id, + u'Downloading info XML', + u'Unable to download video info XML') mdoc = xml.etree.ElementTree.fromstring(metaXml) try: videoNode = mdoc.findall('./video')[0] + youtubeIdNode = videoNode.find('./youtubeID') + if youtubeIdNode is not None: + return self.url_result(youtubeIdNode.text, 'Youtube') info['description'] = videoNode.findall('./description')[0].text info['title'] = videoNode.findall('./caption')[0].text info['thumbnail'] = videoNode.findall('./thumbnail')[0].text @@ -52,11 +54,9 @@ class CollegeHumorIE(InfoExtractor): raise ExtractorError(u'Invalid metadata XML file') manifest_url += '?hdcore=2.10.3' - self.report_manifest(video_id) - try: - manifestXml = compat_urllib_request.urlopen(manifest_url).read() - except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: - raise ExtractorError(u'Unable to download video info XML: %s' % compat_str(err)) + manifestXml = self._download_webpage(manifest_url, video_id, + u'Downloading XML manifest', + u'Unable to download video info XML') adoc = xml.etree.ElementTree.fromstring(manifestXml) try: @@ -66,9 +66,8 @@ class CollegeHumorIE(InfoExtractor): except IndexError as err: raise ExtractorError(u'Invalid manifest file') - url_pr = compat_urllib_parse_urlparse(manifest_url) - url = url_pr.scheme + '://' + url_pr.netloc + '/z' + video_id[:-2] + '/' + node_id + 'Seg1-Frag1' + url_pr = compat_urllib_parse_urlparse(info['thumbnail']) - info['url'] = url - info['ext'] = 'f4f' + info['url'] = url_pr.scheme + '://' + url_pr.netloc + video_id[:-2].replace('.csmil','').replace(',','') + info['ext'] = 'mp4' return [info] diff --git a/youtube_dl/extractor/comedycentral.py b/youtube_dl/extractor/comedycentral.py index 93d9e3d5e..bf8d711ee 100644 --- a/youtube_dl/extractor/comedycentral.py +++ b/youtube_dl/extractor/comedycentral.py @@ -24,7 +24,9 @@ class ComedyCentralIE(InfoExtractor): (full-episodes/(?P<episode>.*)| (?P<clip> (the-colbert-report-(videos|collections)/(?P<clipID>[0-9]+)/[^/]*/(?P<cntitle>.*?)) - |(watch/(?P<date>[^/]*)/(?P<tdstitle>.*))))) + |(watch/(?P<date>[^/]*)/(?P<tdstitle>.*)))| + (?P<interview> + extended-interviews/(?P<interID>[0-9]+)/playlist_tds_extended_(?P<interview_title>.*?)/.*?))) $""" _TEST = { u'url': u'http://www.thedailyshow.com/watch/thu-december-13-2012/kristen-stewart', @@ -87,6 +89,9 @@ class ComedyCentralIE(InfoExtractor): else: epTitle = mobj.group('cntitle') dlNewest = False + elif mobj.group('interview'): + epTitle = mobj.group('interview_title') + dlNewest = False else: dlNewest = not mobj.group('episode') if dlNewest: diff --git a/youtube_dl/extractor/dailymotion.py b/youtube_dl/extractor/dailymotion.py index 9bf7a28ca..fa8c630d0 100644 --- a/youtube_dl/extractor/dailymotion.py +++ b/youtube_dl/extractor/dailymotion.py @@ -1,9 +1,12 @@ import re import json +import itertools from .common import InfoExtractor from ..utils import ( compat_urllib_request, + get_element_by_attribute, + get_element_by_id, ExtractorError, ) @@ -77,3 +80,31 @@ class DailymotionIE(InfoExtractor): 'ext': video_extension, 'thumbnail': info['thumbnail_url'] }] + + +class DailymotionPlaylistIE(InfoExtractor): + _VALID_URL = r'(?:https?://)?(?:www\.)?dailymotion\.[a-z]{2,3}/playlist/(?P<id>.+?)/' + _MORE_PAGES_INDICATOR = r'<div class="next">.*?<a.*?href="/playlist/.+?".*?>.*?</a>.*?</div>' + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + playlist_id = mobj.group('id') + video_ids = [] + + for pagenum in itertools.count(1): + webpage = self._download_webpage('https://www.dailymotion.com/playlist/%s/%s' % (playlist_id, pagenum), + playlist_id, u'Downloading page %s' % pagenum) + + playlist_el = get_element_by_attribute(u'class', u'video_list', webpage) + video_ids.extend(re.findall(r'data-id="(.+?)" data-ext-id', playlist_el)) + + if re.search(self._MORE_PAGES_INDICATOR, webpage, re.DOTALL) is None: + break + + entries = [self.url_result('http://www.dailymotion.com/video/%s' % video_id, 'Dailymotion') + for video_id in video_ids] + return {'_type': 'playlist', + 'id': playlist_id, + 'title': get_element_by_id(u'playlist_name', webpage), + 'entries': entries, + } diff --git a/youtube_dl/extractor/exfm.py b/youtube_dl/extractor/exfm.py new file mode 100644 index 000000000..3443f19c5 --- /dev/null +++ b/youtube_dl/extractor/exfm.py @@ -0,0 +1,54 @@ +import re +import json + +from .common import InfoExtractor + + +class ExfmIE(InfoExtractor): + IE_NAME = u'exfm' + IE_DESC = u'ex.fm' + _VALID_URL = r'(?:http://)?(?:www\.)?ex\.fm/song/([^/]+)' + _SOUNDCLOUD_URL = r'(?:http://)?(?:www\.)?api\.soundcloud.com/tracks/([^/]+)/stream' + _TESTS = [ + { + u'url': u'http://ex.fm/song/1bgtzg', + u'file': u'95223130.mp3', + u'md5': u'8a7967a3fef10e59a1d6f86240fd41cf', + u'info_dict': { + u"title": u"We Can't Stop - Miley Cyrus", + u"uploader": u"Miley Cyrus", + u'upload_date': u'20130603', + u'description': u'Download "We Can\'t Stop" \r\niTunes: http://smarturl.it/WeCantStop?IQid=SC\r\nAmazon: http://smarturl.it/WeCantStopAMZ?IQid=SC', + }, + u'note': u'Soundcloud song', + }, + { + u'url': u'http://ex.fm/song/wddt8', + u'file': u'wddt8.mp3', + u'md5': u'966bd70741ac5b8570d8e45bfaed3643', + u'info_dict': { + u'title': u'Safe and Sound', + u'uploader': u'Capital Cities', + }, + }, + ] + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + song_id = mobj.group(1) + info_url = "http://ex.fm/api/v3/song/%s" %(song_id) + webpage = self._download_webpage(info_url, song_id) + info = json.loads(webpage) + song_url = info['song']['url'] + if re.match(self._SOUNDCLOUD_URL, song_url) is not None: + self.to_screen('Soundcloud song detected') + return self.url_result(song_url.replace('/stream',''), 'Soundcloud') + return [{ + 'id': song_id, + 'url': song_url, + 'ext': 'mp3', + 'title': info['song']['title'], + 'thumbnail': info['song']['image']['large'], + 'uploader': info['song']['artist'], + 'view_count': info['song']['loved_count'], + }] diff --git a/youtube_dl/extractor/ina.py b/youtube_dl/extractor/ina.py index 962c59214..652f19b7b 100644 --- a/youtube_dl/extractor/ina.py +++ b/youtube_dl/extractor/ina.py @@ -5,7 +5,7 @@ from .common import InfoExtractor class InaIE(InfoExtractor): """Information Extractor for Ina.fr""" - _VALID_URL = r'(?:http://)?(?:www\.)?ina\.fr/video/(?P<id>I[0-9]+)/.*' + _VALID_URL = r'(?:http://)?(?:www\.)?ina\.fr/video/(?P<id>I?[A-F0-9]+)/.*' _TEST = { u'url': u'www.ina.fr/video/I12055569/francois-hollande-je-crois-que-c-est-clair-video.html', u'file': u'I12055569.mp4', diff --git a/youtube_dl/extractor/instagram.py b/youtube_dl/extractor/instagram.py index f9ac8d5b4..ddc42882a 100644 --- a/youtube_dl/extractor/instagram.py +++ b/youtube_dl/extractor/instagram.py @@ -10,7 +10,8 @@ class InstagramIE(InfoExtractor): u'md5': u'0d2da106a9d2631273e192b372806516', u'info_dict': { u"uploader_id": u"naomipq", - u"title": u"Video by naomipq" + u"title": u"Video by naomipq", + u'description': u'md5:1f17f0ab29bd6fe2bfad705f58de3cb8', } } @@ -18,20 +19,17 @@ class InstagramIE(InfoExtractor): mobj = re.match(self._VALID_URL, url) video_id = mobj.group(1) webpage = self._download_webpage(url, video_id) - html_title = self._html_search_regex( - r'<title>(.+?)</title>', - webpage, u'title', flags=re.DOTALL) - title = re.sub(u'(?: *\(Videos?\))? \u2022 Instagram$', '', html_title).strip() - uploader_id = self._html_search_regex( - r'<div class="media-user" id="media_user">.*?<h2><a href="[^"]*">([^<]*)</a></h2>', - webpage, u'uploader id', fatal=False, flags=re.DOTALL) - ext = 'mp4' + uploader_id = self._search_regex(r'"owner":{"username":"(.+?)"', + webpage, u'uploader id', fatal=False) + desc = self._search_regex(r'"caption":"(.*?)"', webpage, u'description', + fatal=False) return [{ 'id': video_id, 'url': self._og_search_video_url(webpage), - 'ext': ext, - 'title': title, + 'ext': 'mp4', + 'title': u'Video by %s' % uploader_id, 'thumbnail': self._og_search_thumbnail(webpage), - 'uploader_id' : uploader_id + 'uploader_id' : uploader_id, + 'description': desc, }] diff --git a/youtube_dl/extractor/kankan.py b/youtube_dl/extractor/kankan.py new file mode 100644 index 000000000..8537ba584 --- /dev/null +++ b/youtube_dl/extractor/kankan.py @@ -0,0 +1,37 @@ +import re + +from .common import InfoExtractor +from ..utils import determine_ext + + +class KankanIE(InfoExtractor): + _VALID_URL = r'https?://(?:.*?\.)?kankan\.com/.+?/(?P<id>\d+)\.shtml' + + _TEST = { + u'url': u'http://yinyue.kankan.com/vod/48/48863.shtml', + u'file': u'48863.flv', + u'md5': u'29aca1e47ae68fc28804aca89f29507e', + u'info_dict': { + u'title': u'Ready To Go', + }, + } + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + video_id = mobj.group('id') + webpage = self._download_webpage(url, video_id) + + title = self._search_regex(r'G_TITLE=[\'"](.+?)[\'"]', webpage, u'video title') + gcid = self._search_regex(r'lurl:[\'"]http://.+?/.+?/(.+?)/', webpage, u'gcid') + + video_info_page = self._download_webpage('http://p2s.cl.kankan.com/getCdnresource_flv?gcid=%s' % gcid, + video_id, u'Downloading video url info') + ip = self._search_regex(r'ip:"(.+?)"', video_info_page, u'video url ip') + path = self._search_regex(r'path:"(.+?)"', video_info_page, u'video url path') + video_url = 'http://%s%s' % (ip, path) + + return {'id': video_id, + 'title': title, + 'url': video_url, + 'ext': determine_ext(video_url), + } diff --git a/youtube_dl/extractor/keek.py b/youtube_dl/extractor/keek.py index dda78743d..a7b88d2d9 100644 --- a/youtube_dl/extractor/keek.py +++ b/youtube_dl/extractor/keek.py @@ -4,10 +4,10 @@ from .common import InfoExtractor class KeekIE(InfoExtractor): - _VALID_URL = r'http://(?:www\.)?keek\.com/(?:!|\w+/keeks/)(?P<videoID>\w+)' + _VALID_URL = r'https?://(?:www\.)?keek\.com/(?:!|\w+/keeks/)(?P<videoID>\w+)' IE_NAME = u'keek' _TEST = { - u'url': u'http://www.keek.com/ytdl/keeks/NODfbab', + u'url': u'https://www.keek.com/ytdl/keeks/NODfbab', u'file': u'NODfbab.mp4', u'md5': u'9b0636f8c0f7614afa4ea5e4c6e57e83', u'info_dict': { diff --git a/youtube_dl/extractor/muzu.py b/youtube_dl/extractor/muzu.py new file mode 100644 index 000000000..03e31ea1c --- /dev/null +++ b/youtube_dl/extractor/muzu.py @@ -0,0 +1,64 @@ +import re +import json + +from .common import InfoExtractor +from ..utils import ( + compat_urllib_parse, + determine_ext, +) + + +class MuzuTVIE(InfoExtractor): + _VALID_URL = r'https?://www.muzu.tv/(.+?)/(.+?)/(?P<id>\d+)' + IE_NAME = u'muzu.tv' + + _TEST = { + u'url': u'http://www.muzu.tv/defected/marcashken-featuring-sos-cat-walk-original-mix-music-video/1981454/', + u'file': u'1981454.mp4', + u'md5': u'98f8b2c7bc50578d6a0364fff2bfb000', + u'info_dict': { + u'title': u'Cat Walk (Original Mix)', + u'description': u'md5:90e868994de201b2570e4e5854e19420', + u'uploader': u'MarcAshken featuring SOS', + }, + } + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + video_id = mobj.group('id') + + info_data = compat_urllib_parse.urlencode({'format': 'json', + 'url': url, + }) + video_info_page = self._download_webpage('http://www.muzu.tv/api/oembed/?%s' % info_data, + video_id, u'Downloading video info') + info = json.loads(video_info_page) + + player_info_page = self._download_webpage('http://player.muzu.tv/player/playerInit?ai=%s' % video_id, + video_id, u'Downloading player info') + video_info = json.loads(player_info_page)['videos'][0] + for quality in ['1080' , '720', '480', '360']: + if video_info.get('v%s' % quality): + break + + data = compat_urllib_parse.urlencode({'ai': video_id, + # Even if each time you watch a video the hash changes, + # it seems to work for different videos, and it will work + # even if you use any non empty string as a hash + 'viewhash': 'VBNff6djeV4HV5TRPW5kOHub2k', + 'device': 'web', + 'qv': quality, + }) + video_url_page = self._download_webpage('http://player.muzu.tv/player/requestVideo?%s' % data, + video_id, u'Downloading video url') + video_url_info = json.loads(video_url_page) + video_url = video_url_info['url'] + + return {'id': video_id, + 'title': info['title'], + 'url': video_url, + 'ext': determine_ext(video_url), + 'thumbnail': info['thumbnail_url'], + 'description': info['description'], + 'uploader': info['author_name'], + } diff --git a/youtube_dl/extractor/myvideo.py b/youtube_dl/extractor/myvideo.py index b2a7b1df0..0404e6e43 100644 --- a/youtube_dl/extractor/myvideo.py +++ b/youtube_dl/extractor/myvideo.py @@ -2,11 +2,13 @@ import binascii import base64 import hashlib import re +import json from .common import InfoExtractor from ..utils import ( compat_ord, compat_urllib_parse, + compat_urllib_request, ExtractorError, ) @@ -16,7 +18,7 @@ from ..utils import ( class MyVideoIE(InfoExtractor): """Information Extractor for myvideo.de.""" - _VALID_URL = r'(?:http://)?(?:www\.)?myvideo\.de/watch/([0-9]+)/([^?/]+).*' + _VALID_URL = r'(?:http://)?(?:www\.)?myvideo\.de/(?:[^/]+/)?watch/([0-9]+)/([^?/]+).*' IE_NAME = u'myvideo' _TEST = { u'url': u'http://www.myvideo.de/watch/8229274/bowling_fail_or_win', @@ -85,6 +87,20 @@ class MyVideoIE(InfoExtractor): 'ext': video_ext, }] + 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, '') + response = self._download_webpage(request, video_id, + u'Downloading video info') + info = json.loads(base64.b64decode(response).decode('utf-8')) + return {'id': video_id, + 'title': info['title'], + 'url': info['streaming_url'].replace('rtmpe', 'rtmpt'), + 'play_path': info['filename'], + 'ext': 'flv', + 'thumbnail': info['thumbnail'][0]['url'], + } + # try encxml mobj = re.search('var flashvars={(.+?)}', webpage) if mobj is None: diff --git a/youtube_dl/extractor/ooyala.py b/youtube_dl/extractor/ooyala.py new file mode 100644 index 000000000..b734722d0 --- /dev/null +++ b/youtube_dl/extractor/ooyala.py @@ -0,0 +1,52 @@ +import re +import json + +from .common import InfoExtractor +from ..utils import unescapeHTML + +class OoyalaIE(InfoExtractor): + _VALID_URL = r'https?://.+?\.ooyala\.com/.*?embedCode=(?P<id>.+?)(&|$)' + + _TEST = { + # From http://it.slashdot.org/story/13/04/25/178216/recovering-data-from-broken-hard-drives-and-ssds-video + u'url': u'http://player.ooyala.com/player.js?embedCode=pxczE2YjpfHfn1f3M-ykG_AmJRRn0PD8', + u'file': u'pxczE2YjpfHfn1f3M-ykG_AmJRRn0PD8.mp4', + u'md5': u'3f5cceb3a7bf461d6c29dc466cf8033c', + u'info_dict': { + u'title': u'Explaining Data Recovery from Hard Drives and SSDs', + u'description': u'How badly damaged does a drive have to be to defeat Russell and his crew? Apparently, smashed to bits.', + }, + } + + def _extract_result(self, info, more_info): + return {'id': info['embedCode'], + 'ext': 'mp4', + 'title': unescapeHTML(info['title']), + 'url': info['url'], + 'description': unescapeHTML(more_info['description']), + 'thumbnail': more_info['promo'], + } + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + embedCode = mobj.group('id') + player_url = 'http://player.ooyala.com/player.js?embedCode=%s' % embedCode + player = self._download_webpage(player_url, embedCode) + mobile_url = self._search_regex(r'mobile_player_url="(.+?)&device="', + player, u'mobile player url') + mobile_player = self._download_webpage(mobile_url, embedCode) + videos_info = self._search_regex(r'eval\("\((\[{.*?stream_redirect.*?}\])\)"\);', mobile_player, u'info').replace('\\"','"') + videos_more_info = self._search_regex(r'eval\("\(({.*?\\"promo\\".*?})\)"', mobile_player, u'more info').replace('\\"','"') + videos_info = json.loads(videos_info) + videos_more_info =json.loads(videos_more_info) + + if videos_more_info.get('lineup'): + videos = [self._extract_result(info, more_info) for (info, more_info) in zip(videos_info, videos_more_info['lineup'])] + return {'_type': 'playlist', + 'id': embedCode, + 'title': unescapeHTML(videos_more_info['title']), + 'entries': videos, + } + else: + return self._extract_result(videos_info[0], videos_more_info) + diff --git a/youtube_dl/extractor/roxwel.py b/youtube_dl/extractor/roxwel.py new file mode 100644 index 000000000..d339e6cb5 --- /dev/null +++ b/youtube_dl/extractor/roxwel.py @@ -0,0 +1,49 @@ +import re +import json + +from .common import InfoExtractor +from ..utils import unified_strdate, determine_ext + + +class RoxwelIE(InfoExtractor): + _VALID_URL = r'https?://www\.roxwel\.com/player/(?P<filename>.+?)(\.|\?|$)' + + _TEST = { + u'url': u'http://www.roxwel.com/player/passionpittakeawalklive.html', + u'file': u'passionpittakeawalklive.flv', + u'md5': u'd9dea8360a1e7d485d2206db7fe13035', + u'info_dict': { + u'title': u'Take A Walk (live)', + u'uploader': u'Passion Pit', + u'description': u'Passion Pit performs "Take A Walk\" live at The Backyard in Austin, Texas. ', + }, + u'skip': u'Requires rtmpdump', + } + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + filename = mobj.group('filename') + info_url = 'http://www.roxwel.com/api/videos/%s' % filename + info_page = self._download_webpage(info_url, filename, + u'Downloading video info') + + self.report_extraction(filename) + info = json.loads(info_page) + rtmp_rates = sorted([int(r.replace('flv_', '')) for r in info['media_rates'] if r.startswith('flv_')]) + best_rate = rtmp_rates[-1] + url_page_url = 'http://roxwel.com/pl_one_time.php?filename=%s&quality=%s' % (filename, best_rate) + rtmp_url = self._download_webpage(url_page_url, filename, u'Downloading video url') + ext = determine_ext(rtmp_url) + if ext == 'f4v': + rtmp_url = rtmp_url.replace(filename, 'mp4:%s' % filename) + + return {'id': filename, + 'title': info['title'], + 'url': rtmp_url, + 'ext': 'flv', + 'description': info['description'], + 'thumbnail': info.get('player_image_url') or info.get('image_url_large'), + 'uploader': info['artist'], + 'uploader_id': info['artistname'], + 'upload_date': unified_strdate(info['dbdate']), + } diff --git a/youtube_dl/extractor/sina.py b/youtube_dl/extractor/sina.py new file mode 100644 index 000000000..14b1c656c --- /dev/null +++ b/youtube_dl/extractor/sina.py @@ -0,0 +1,67 @@ +# coding: utf-8 + +import re +import xml.etree.ElementTree + +from .common import InfoExtractor +from ..utils import ( + compat_urllib_request, + compat_urllib_parse, +) + + +class SinaIE(InfoExtractor): + _VALID_URL = r'''https?://(.*?\.)?video\.sina\.com\.cn/ + ( + (.+?/(((?P<pseudo_id>\d+).html)|(.*?(\#|(vid=))(?P<id>\d+?)($|&)))) + | + # This is used by external sites like Weibo + (api/sinawebApi/outplay.php/(?P<token>.+?)\.swf) + ) + ''' + + _TEST = { + u'url': u'http://video.sina.com.cn/news/vlist/zt/chczlj2013/?opsubject_id=top12#110028898', + u'file': u'110028898.flv', + u'md5': u'd65dd22ddcf44e38ce2bf58a10c3e71f', + u'info_dict': { + u'title': u'《中国新闻》 朝鲜要求巴拿马立即释放被扣船员', + } + } + + @classmethod + def suitable(cls, url): + return re.match(cls._VALID_URL, url, flags=re.VERBOSE) is not None + + def _extract_video(self, video_id): + data = compat_urllib_parse.urlencode({'vid': video_id}) + url_page = self._download_webpage('http://v.iask.com/v_play.php?%s' % data, + video_id, u'Downloading video url') + image_page = self._download_webpage( + 'http://interface.video.sina.com.cn/interface/common/getVideoImage.php?%s' % data, + video_id, u'Downloading thumbnail info') + url_doc = xml.etree.ElementTree.fromstring(url_page.encode('utf-8')) + + return {'id': video_id, + 'url': url_doc.find('./durl/url').text, + 'ext': 'flv', + 'title': url_doc.find('./vname').text, + 'thumbnail': image_page.split('=')[1], + } + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url, flags=re.VERBOSE) + video_id = mobj.group('id') + if mobj.group('token') is not None: + # The video id is in the redirected url + self.to_screen(u'Getting video id') + request = compat_urllib_request.Request(url) + request.get_method = lambda: 'HEAD' + (_, urlh) = self._download_webpage_handle(request, 'NA', False) + return self._real_extract(urlh.geturl()) + elif video_id is None: + pseudo_id = mobj.group('pseudo_id') + webpage = self._download_webpage(url, pseudo_id) + video_id = self._search_regex(r'vid:\'(\d+?)\'', webpage, u'video id') + + return self._extract_video(video_id) diff --git a/youtube_dl/extractor/soundcloud.py b/youtube_dl/extractor/soundcloud.py index d47c49c03..7c9f1c6b6 100644 --- a/youtube_dl/extractor/soundcloud.py +++ b/youtube_dl/extractor/soundcloud.py @@ -19,7 +19,11 @@ class SoundcloudIE(InfoExtractor): of the stream token and uid """ - _VALID_URL = r'^(?:https?://)?(?:www\.)?soundcloud\.com/([\w\d-]+)/([\w\d-]+)(?:[?].*)?$' + _VALID_URL = r'''^(?:https?://)? + (?:(?:(?:www\.)?soundcloud\.com/([\w\d-]+)/([\w\d-]+)/?(?:[?].*)?$) + |(?:api\.soundcloud\.com/tracks/(?P<track_id>\d+)) + ) + ''' IE_NAME = u'soundcloud' _TEST = { u'url': u'http://soundcloud.com/ethmusic/lostin-powers-she-so-heavy', @@ -33,59 +37,65 @@ class SoundcloudIE(InfoExtractor): } } + _CLIENT_ID = 'b45b1aa10f1ac2941910a7f0d10f8e28' + + @classmethod + def suitable(cls, url): + return re.match(cls._VALID_URL, url, flags=re.VERBOSE) is not None + def report_resolve(self, video_id): """Report information extraction.""" self.to_screen(u'%s: Resolving id' % video_id) - def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - if mobj is None: - raise ExtractorError(u'Invalid URL: %s' % url) - - # extract uploader (which is in the url) - uploader = mobj.group(1) - # extract simple title (uploader + slug of song title) - slug_title = mobj.group(2) - full_title = '%s/%s' % (uploader, slug_title) - - self.report_resolve(full_title) - - url = 'http://soundcloud.com/%s/%s' % (uploader, slug_title) - resolv_url = 'http://api.soundcloud.com/resolve.json?url=' + url + '&client_id=b45b1aa10f1ac2941910a7f0d10f8e28' - info_json = self._download_webpage(resolv_url, full_title, u'Downloading info JSON') + @classmethod + def _resolv_url(cls, url): + return 'http://api.soundcloud.com/resolve.json?url=' + url + '&client_id=' + cls._CLIENT_ID - info = json.loads(info_json) + def _extract_info_dict(self, info, full_title=None): video_id = info['id'] - self.report_extraction(full_title) - - streams_url = 'https://api.sndcdn.com/i1/tracks/' + str(video_id) + '/streams?client_id=b45b1aa10f1ac2941910a7f0d10f8e28' - stream_json = self._download_webpage(streams_url, full_title, - u'Downloading stream definitions', - u'unable to download stream definitions') - - streams = json.loads(stream_json) - mediaURL = streams['http_mp3_128_url'] - upload_date = unified_strdate(info['created_at']) + name = full_title or video_id + self.report_extraction(name) - return [{ + thumbnail = info['artwork_url'] + if thumbnail is not None: + thumbnail = thumbnail.replace('-large', '-t500x500') + return { 'id': info['id'], - 'url': mediaURL, + 'url': info['stream_url'] + '?client_id=' + self._CLIENT_ID, 'uploader': info['user']['username'], - 'upload_date': upload_date, + 'upload_date': unified_strdate(info['created_at']), 'title': info['title'], 'ext': u'mp3', 'description': info['description'], - }] + 'thumbnail': thumbnail, + } -class SoundcloudSetIE(InfoExtractor): - """Information extractor for soundcloud.com sets - To access the media, the uid of the song and a stream token - must be extracted from the page source and the script must make - a request to media.soundcloud.com/crossdomain.xml. Then - the media can be grabbed by requesting from an url composed - of the stream token and uid - """ + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url, flags=re.VERBOSE) + if mobj is None: + raise ExtractorError(u'Invalid URL: %s' % url) + + track_id = mobj.group('track_id') + if track_id is not None: + info_json_url = 'http://api.soundcloud.com/tracks/' + track_id + '.json?client_id=' + self._CLIENT_ID + full_title = track_id + else: + # extract uploader (which is in the url) + uploader = mobj.group(1) + # extract simple title (uploader + slug of song title) + slug_title = mobj.group(2) + full_title = '%s/%s' % (uploader, slug_title) + + self.report_resolve(full_title) + + url = 'http://soundcloud.com/%s/%s' % (uploader, slug_title) + info_json_url = self._resolv_url(url) + info_json = self._download_webpage(info_json_url, full_title, u'Downloading info JSON') + info = json.loads(info_json) + return self._extract_info_dict(info, full_title) + +class SoundcloudSetIE(SoundcloudIE): _VALID_URL = r'^(?:https?://)?(?:www\.)?soundcloud\.com/([\w\d-]+)/sets/([\w\d-]+)(?:[?].*)?$' IE_NAME = u'soundcloud:set' _TEST = { @@ -153,10 +163,6 @@ class SoundcloudSetIE(InfoExtractor): ] } - def report_resolve(self, video_id): - """Report information extraction.""" - self.to_screen(u'%s: Resolving id' % video_id) - def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) if mobj is None: @@ -171,7 +177,7 @@ class SoundcloudSetIE(InfoExtractor): self.report_resolve(full_title) url = 'http://soundcloud.com/%s/sets/%s' % (uploader, slug_title) - resolv_url = 'http://api.soundcloud.com/resolve.json?url=' + url + '&client_id=b45b1aa10f1ac2941910a7f0d10f8e28' + resolv_url = self._resolv_url(url) info_json = self._download_webpage(resolv_url, full_title) videos = [] @@ -182,23 +188,8 @@ class SoundcloudSetIE(InfoExtractor): return self.report_extraction(full_title) - for track in info['tracks']: - video_id = track['id'] - - streams_url = 'https://api.sndcdn.com/i1/tracks/' + str(video_id) + '/streams?client_id=b45b1aa10f1ac2941910a7f0d10f8e28' - stream_json = self._download_webpage(streams_url, video_id, u'Downloading track info JSON') - - self.report_extraction(video_id) - streams = json.loads(stream_json) - mediaURL = streams['http_mp3_128_url'] - - videos.append({ - 'id': video_id, - 'url': mediaURL, - 'uploader': track['user']['username'], - 'upload_date': unified_strdate(track['created_at']), - 'title': track['title'], - 'ext': u'mp3', - 'description': track['description'], - }) - return videos + return {'_type': 'playlist', + 'entries': [self._extract_info_dict(track) for track in info['tracks']], + 'id': info['id'], + 'title': info['title'], + } diff --git a/youtube_dl/extractor/teamcoco.py b/youtube_dl/extractor/teamcoco.py index ec92e589a..c910110ca 100644 --- a/youtube_dl/extractor/teamcoco.py +++ b/youtube_dl/extractor/teamcoco.py @@ -33,7 +33,7 @@ class TeamcocoIE(InfoExtractor): data_url = 'http://teamcoco.com/cvp/2.0/%s.xml' % video_id data = self._download_webpage(data_url, video_id, 'Downloading data webpage') - video_url = self._html_search_regex(r'<file type="high".*?>(.*?)</file>', + video_url = self._html_search_regex(r'<file [^>]*type="high".*?>(.*?)</file>', data, u'video URL') return [{ diff --git a/youtube_dl/extractor/ted.py b/youtube_dl/extractor/ted.py index 8b73b8340..4c11f7a03 100644 --- a/youtube_dl/extractor/ted.py +++ b/youtube_dl/extractor/ted.py @@ -67,7 +67,7 @@ class TEDIE(InfoExtractor): webpage = self._download_webpage(url, video_id, 'Downloading \"%s\" page' % video_name) self.report_extraction(video_name) # If the url includes the language we get the title translated - title = self._html_search_regex(r'<span id="altHeadline" >(?P<title>.*)</span>', + title = self._html_search_regex(r'<span .*?id="altHeadline".+?>(?P<title>.*)</span>', webpage, 'title') json_data = self._search_regex(r'<script.*?>var talkDetails = ({.*?})</script>', webpage, 'json data') diff --git a/youtube_dl/extractor/tf1.py b/youtube_dl/extractor/tf1.py index e0ffeced5..772134a12 100644 --- a/youtube_dl/extractor/tf1.py +++ b/youtube_dl/extractor/tf1.py @@ -6,19 +6,17 @@ import re from .common import InfoExtractor class TF1IE(InfoExtractor): - """ - TF1 uses the wat.tv player, currently it can only download videos with the - html5 player enabled, it cannot download HD videos. - """ + """TF1 uses the wat.tv player.""" _VALID_URL = r'http://videos.tf1.fr/.*-(.*?).html' _TEST = { u'url': u'http://videos.tf1.fr/auto-moto/citroen-grand-c4-picasso-2013-presentation-officielle-8062060.html', u'file': u'10635995.mp4', - u'md5': u'66789d3e91278d332f75e1feb7aea327', + u'md5': u'2e378cc28b9957607d5e88f274e637d8', u'info_dict': { u'title': u'Citroën Grand C4 Picasso 2013 : présentation officielle', u'description': u'Vidéo officielle du nouveau Citroën Grand C4 Picasso, lancé à l\'automne 2013.', - } + }, + u'skip': u'Sometimes wat serves the whole file with the --test option', } def _real_extract(self, url): diff --git a/youtube_dl/extractor/traileraddict.py b/youtube_dl/extractor/traileraddict.py index 324bb6231..35f89e9ee 100644 --- a/youtube_dl/extractor/traileraddict.py +++ b/youtube_dl/extractor/traileraddict.py @@ -4,11 +4,11 @@ from .common import InfoExtractor class TrailerAddictIE(InfoExtractor): - _VALID_URL = r'(?:http://)?(?:www\.)?traileraddict\.com/trailer/([^/]+)/(?:trailer|feature-trailer)' + _VALID_URL = r'(?:http://)?(?:www\.)?traileraddict\.com/(?:trailer|clip)/(?P<movie>.+?)/(?P<trailer_name>.+)' _TEST = { u'url': u'http://www.traileraddict.com/trailer/prince-avalanche/trailer', u'file': u'76184.mp4', - u'md5': u'41365557f3c8c397d091da510e73ceb4', + u'md5': u'57e39dbcf4142ceb8e1f242ff423fd71', u'info_dict': { u"title": u"Prince Avalanche Trailer", u"description": u"Trailer for Prince Avalanche.Two highway road workers spend the summer of 1988 away from their city lives. The isolated landscape becomes a place of misadventure as the men find themselves at odds with each other and the women they left behind." @@ -17,24 +17,30 @@ class TrailerAddictIE(InfoExtractor): def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) - video_id = mobj.group(1) - webpage = self._download_webpage(url, video_id) - + name = mobj.group('movie') + '/' + mobj.group('trailer_name') + webpage = self._download_webpage(url, name) + title = self._search_regex(r'<title>(.+?)</title>', webpage, 'video title').replace(' - Trailer Addict','') view_count = self._search_regex(r'Views: (.+?)<br />', webpage, 'Views Count') video_id = self._og_search_property('video', webpage, 'Video id').split('=')[1] - info_url = "http://www.traileraddict.com/fvar.php?tid=%s" %(str(video_id)) + # Presence of (no)watchplus function indicates HD quality is available + if re.search(r'function (no)?watchplus()', webpage): + fvar = "fvarhd" + else: + fvar = "fvar" + + info_url = "http://www.traileraddict.com/%s.php?tid=%s" % (fvar, str(video_id)) info_webpage = self._download_webpage(info_url, video_id , "Downloading the info webpage") - + final_url = self._search_regex(r'&fileurl=(.+)', info_webpage, 'Download url').replace('%3F','?') thumbnail_url = self._search_regex(r'&image=(.+?)&', info_webpage, 'thumbnail url') ext = final_url.split('.')[-1].split('?')[0] - + return [{ 'id' : video_id, 'url' : final_url, diff --git a/youtube_dl/extractor/vevo.py b/youtube_dl/extractor/vevo.py index 3b16dcfbc..67537eae5 100644 --- a/youtube_dl/extractor/vevo.py +++ b/youtube_dl/extractor/vevo.py @@ -35,12 +35,12 @@ class VevoIE(InfoExtractor): self.report_extraction(video_id) video_info = json.loads(info_json) - m_urls = list(re.finditer(r'<video src="(?P<ext>.*?):(?P<url>.*?)"', links_webpage)) + m_urls = list(re.finditer(r'<video src="(?P<ext>.*?):/?(?P<url>.*?)"', links_webpage)) if m_urls is None or len(m_urls) == 0: raise ExtractorError(u'Unable to extract video url') # They are sorted from worst to best quality m_url = m_urls[-1] - video_url = base_url + m_url.group('url') + video_url = base_url + '/' + m_url.group('url') ext = m_url.group('ext') return {'url': video_url, diff --git a/youtube_dl/extractor/videofyme.py b/youtube_dl/extractor/videofyme.py new file mode 100644 index 000000000..04106672b --- /dev/null +++ b/youtube_dl/extractor/videofyme.py @@ -0,0 +1,49 @@ +import re +import xml.etree.ElementTree + +from .common import InfoExtractor +from ..utils import ( + find_xpath_attr, + determine_ext, +) + +class VideofyMeIE(InfoExtractor): + _VALID_URL = r'https?://(www.videofy.me/.+?|p.videofy.me/v)/(?P<id>\d+)(&|#|$)' + IE_NAME = u'videofy.me' + + _TEST = { + u'url': u'http://www.videofy.me/thisisvideofyme/1100701', + u'file': u'1100701.mp4', + u'md5': u'2046dd5758541d630bfa93e741e2fd79', + u'info_dict': { + u'title': u'This is VideofyMe', + u'description': None, + u'uploader': u'VideofyMe', + u'uploader_id': u'thisisvideofyme', + }, + + } + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + video_id = mobj.group('id') + config_xml = self._download_webpage('http://sunshine.videofy.me/?videoId=%s' % video_id, + video_id) + config = xml.etree.ElementTree.fromstring(config_xml.encode('utf-8')) + video = config.find('video') + sources = video.find('sources') + url_node = find_xpath_attr(sources, 'source', 'id', 'HQ on') + if url_node is None: + url_node = find_xpath_attr(sources, 'source', 'id', 'HQ off') + video_url = url_node.find('url').text + + return {'id': video_id, + 'title': video.find('title').text, + 'url': video_url, + 'ext': determine_ext(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': re.search(r'\d+', video.find('views').text).group(), + } diff --git a/youtube_dl/extractor/vimeo.py b/youtube_dl/extractor/vimeo.py index ac32043c1..cc9c8d018 100644 --- a/youtube_dl/extractor/vimeo.py +++ b/youtube_dl/extractor/vimeo.py @@ -1,5 +1,6 @@ import json import re +import itertools from .common import InfoExtractor from ..utils import ( @@ -171,3 +172,31 @@ class VimeoIE(InfoExtractor): 'thumbnail': video_thumbnail, 'description': video_description, }] + + +class VimeoChannelIE(InfoExtractor): + IE_NAME = u'vimeo:channel' + _VALID_URL = r'(?:https?://)?vimeo.\com/channels/(?P<id>[^/]+)' + _MORE_PAGES_INDICATOR = r'<a.+?rel="next"' + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + channel_id = mobj.group('id') + video_ids = [] + + for pagenum in itertools.count(1): + webpage = self._download_webpage('http://vimeo.com/channels/%s/videos/page:%d' % (channel_id, pagenum), + channel_id, u'Downloading page %s' % pagenum) + 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('http://vimeo.com/%s' % video_id, 'Vimeo') + for video_id in video_ids] + channel_title = self._html_search_regex(r'<a href="/channels/%s">(.*?)</a>' % channel_id, + webpage, u'channel title') + return {'_type': 'playlist', + 'id': channel_id, + 'title': channel_title, + 'entries': entries, + } diff --git a/youtube_dl/extractor/wat.py b/youtube_dl/extractor/wat.py index 0d1302cd2..7d228edac 100644 --- a/youtube_dl/extractor/wat.py +++ b/youtube_dl/extractor/wat.py @@ -17,11 +17,12 @@ class WatIE(InfoExtractor): _TEST = { u'url': u'http://www.wat.tv/video/world-war-philadelphia-vost-6bv55_2fjr7_.html', u'file': u'10631273.mp4', - u'md5': u'0a4fe7870f31eaeabb5e25fd8da8414a', + u'md5': u'd8b2231e1e333acd12aad94b80937e19', u'info_dict': { u'title': u'World War Z - Philadelphia VOST', u'description': u'La menace est partout. Que se passe-t-il à Philadelphia ?\r\nWORLD WAR Z, avec Brad Pitt, au cinéma le 3 juillet.\r\nhttp://www.worldwarz.fr', - } + }, + u'skip': u'Sometimes wat serves the whole file with the --test option', } def download_video_info(self, real_id): @@ -58,20 +59,8 @@ class WatIE(InfoExtractor): # Otherwise we can continue and extract just one part, we have to use # the short id for getting the video url - player_data = compat_urllib_parse.urlencode({'shortVideoId': short_id, - 'html5': '1'}) - player_info = self._download_webpage('http://www.wat.tv/player?' + player_data, - real_id, u'Downloading player info') - player = json.loads(player_info)['player'] - html5_player = self._html_search_regex(r'iframe src="(.*?)"', player, - 'html5 player') - player_webpage = self._download_webpage(html5_player, real_id, - u'Downloading player webpage') - - video_url = self._search_regex(r'urlhtml5 : "(.*?)"', player_webpage, - 'video url') info = {'id': real_id, - 'url': video_url, + 'url': 'http://wat.tv/get/android5/%s.mp4' % real_id, 'ext': 'mp4', 'title': first_chapter['title'], 'thumbnail': first_chapter['preview'], diff --git a/youtube_dl/extractor/weibo.py b/youtube_dl/extractor/weibo.py new file mode 100644 index 000000000..0757495bd --- /dev/null +++ b/youtube_dl/extractor/weibo.py @@ -0,0 +1,48 @@ +# coding: utf-8 + +import re +import json + +from .common import InfoExtractor + +class WeiboIE(InfoExtractor): + """ + The videos in Weibo come from different sites, this IE just finds the link + to the external video and returns it. + """ + _VALID_URL = r'https?://video\.weibo\.com/v/weishipin/t_(?P<id>.+?)\.htm' + + _TEST = { + u'url': u'http://video.weibo.com/v/weishipin/t_zjUw2kZ.htm', + u'file': u'98322879.flv', + u'info_dict': { + u'title': u'魔声耳机最新广告“All Eyes On Us”', + }, + u'note': u'Sina video', + u'params': { + u'skip_download': True, + }, + } + + # Additional example videos from different sites + # Youku: http://video.weibo.com/v/weishipin/t_zQGDWQ8.htm + # 56.com: http://video.weibo.com/v/weishipin/t_zQ44HxN.htm + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url, flags=re.VERBOSE) + video_id = mobj.group('id') + info_url = 'http://video.weibo.com/?s=v&a=play_list&format=json&mix_video_id=t_%s' % video_id + info_page = self._download_webpage(info_url, video_id) + info = json.loads(info_page) + + videos_urls = map(lambda v: v['play_page_url'], info['result']['data']) + #Prefer sina video since they have thumbnails + videos_urls = sorted(videos_urls, key=lambda u: u'video.sina.com' in u) + player_url = videos_urls[-1] + m_sina = re.match(r'https?://video.sina.com.cn/v/b/(\d+)-\d+.html', player_url) + if m_sina is not None: + self.to_screen('Sina video detected') + sina_id = m_sina.group(1) + player_url = 'http://you.video.sina.com.cn/swf/quotePlayer.swf?vid=%s' % sina_id + return self.url_result(player_url) + diff --git a/youtube_dl/extractor/worldstarhiphop.py b/youtube_dl/extractor/worldstarhiphop.py index 5b9779c05..3237596a3 100644 --- a/youtube_dl/extractor/worldstarhiphop.py +++ b/youtube_dl/extractor/worldstarhiphop.py @@ -21,6 +21,13 @@ class WorldStarHipHopIE(InfoExtractor): webpage_src = self._download_webpage(url, video_id) + m_vevo_id = re.search(r'videoId=(.*?)&?', + webpage_src) + + if m_vevo_id is not None: + self.to_screen(u'Vevo video detected:') + return self.url_result('vevo:%s' % m_vevo_id.group(1), ie='Vevo') + video_url = self._search_regex(r'so\.addVariable\("file","(.*?)"\)', webpage_src, u'video URL') diff --git a/youtube_dl/extractor/youku.py b/youtube_dl/extractor/youku.py index eb9829801..996d38478 100644 --- a/youtube_dl/extractor/youku.py +++ b/youtube_dl/extractor/youku.py @@ -13,7 +13,7 @@ from ..utils import ( class YoukuIE(InfoExtractor): - _VALID_URL = r'(?:http://)?v\.youku\.com/v_show/id_(?P<ID>[A-Za-z0-9]+)\.html' + _VALID_URL = r'(?:http://)?(v|player)\.youku\.com/(v_show/id_|player\.php/sid/)(?P<ID>[A-Za-z0-9]+)(\.html|/v.swf)' _TEST = { u"url": u"http://v.youku.com/v_show/id_XNDgyMDQ2NTQw.html", u"file": u"XNDgyMDQ2NTQw_part00.flv", diff --git a/youtube_dl/extractor/youtube.py b/youtube_dl/extractor/youtube.py index 7577c12f5..7df79b943 100644 --- a/youtube_dl/extractor/youtube.py +++ b/youtube_dl/extractor/youtube.py @@ -23,8 +23,114 @@ from ..utils import ( orderedSet, ) +class YoutubeBaseInfoExtractor(InfoExtractor): + """Provide base functions for Youtube extractors""" + _LOGIN_URL = 'https://accounts.google.com/ServiceLogin' + _LANG_URL = r'https://www.youtube.com/?hl=en&persist_hl=1&gl=US&persist_gl=1&opt_out_ackd=1' + _AGE_URL = 'http://www.youtube.com/verify_age?next_url=/&gl=US&hl=en' + _NETRC_MACHINE = 'youtube' + # If True it will raise an error if no login info is provided + _LOGIN_REQUIRED = False + + def report_lang(self): + """Report attempt to set language.""" + self.to_screen(u'Setting language') + + def _set_language(self): + request = compat_urllib_request.Request(self._LANG_URL) + try: + self.report_lang() + compat_urllib_request.urlopen(request).read() + except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: + self._downloader.report_warning(u'unable to set language: %s' % compat_str(err)) + return False + return True + + def _login(self): + (username, password) = self._get_login_info() + # No authentication to be performed + if username is None: + if self._LOGIN_REQUIRED: + raise ExtractorError(u'No login info available, needed for using %s.' % self.IE_NAME, expected=True) + return False + + request = compat_urllib_request.Request(self._LOGIN_URL) + try: + login_page = compat_urllib_request.urlopen(request).read().decode('utf-8') + except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: + 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) + + # Log in + login_form_strs = { + u'continue': u'https://www.youtube.com/signin?action_handle_signin=true&feature=sign_in_button&hl=en_US&nomobiletemp=1', + u'Email': username, + u'GALX': galx, + u'Passwd': password, + u'PersistentCookie': u'yes', + u'_utf8': u'霱', + u'bgresponse': u'js_disabled', + u'checkConnection': u'', + u'checkedDomains': u'youtube', + u'dnConn': u'', + u'dsh': dsh, + u'pstMsg': u'0', + u'rmShown': u'1', + u'secTok': u'', + u'signIn': u'Sign in', + u'timeStmp': u'', + u'service': u'youtube', + u'uilel': u'3', + u'hl': u'en_US', + } + # Convert to UTF-8 *before* urlencode because Python 2.x's urlencode + # chokes on unicode + login_form = dict((k.encode('utf-8'), v.encode('utf-8')) for k,v in login_form_strs.items()) + login_data = compat_urllib_parse.urlencode(login_form).encode('ascii') + request = compat_urllib_request.Request(self._LOGIN_URL, login_data) + try: + self.report_login() + login_results = compat_urllib_request.urlopen(request).read().decode('utf-8') + if re.search(r'(?i)<form[^>]* id="gaia_loginform"', login_results) is not None: + self._downloader.report_warning(u'unable to log in: bad username or password') + return False + except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: + self._downloader.report_warning(u'unable to log in: %s' % compat_str(err)) + return False + return True -class YoutubeIE(InfoExtractor): + def _confirm_age(self): + age_form = { + 'next_url': '/', + 'action_confirm': 'Confirm', + } + request = compat_urllib_request.Request(self._AGE_URL, compat_urllib_parse.urlencode(age_form)) + try: + self.report_age_confirmation() + compat_urllib_request.urlopen(request).read().decode('utf-8') + except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: + raise ExtractorError(u'Unable to confirm age: %s' % compat_str(err)) + return True + + def _real_initialize(self): + if self._downloader is None: + return + if not self._set_language(): + return + if not self._login(): + return + self._confirm_age() + +class YoutubeIE(YoutubeBaseInfoExtractor): IE_DESC = u'YouTube.com' _VALID_URL = r"""^ ( @@ -45,14 +151,16 @@ class YoutubeIE(InfoExtractor): ([0-9A-Za-z_-]+) # here is it! the YouTube video ID (?(1).+)? # if we found the ID, everything can follow $""" - _LANG_URL = r'https://www.youtube.com/?hl=en&persist_hl=1&gl=US&persist_gl=1&opt_out_ackd=1' - _LOGIN_URL = 'https://accounts.google.com/ServiceLogin' - _AGE_URL = 'http://www.youtube.com/verify_age?next_url=/&gl=US&hl=en' _NEXT_URL_RE = r'[\?&]next_url=([^&]+)' - _NETRC_MACHINE = 'youtube' # Listed in order of quality - _available_formats = ['38', '37', '46', '22', '45', '35', '44', '34', '18', '43', '6', '5', '17', '13'] - _available_formats_prefer_free = ['38', '46', '37', '45', '22', '44', '35', '43', '34', '18', '6', '5', '17', '13'] + _available_formats = ['38', '37', '46', '22', '45', '35', '44', '34', '18', '43', '6', '5', '17', '13', + '95', '94', '93', '92', '132', '151', + '85', '84', '102', '83', '101', '82', '100', + ] + _available_formats_prefer_free = ['38', '46', '37', '45', '22', '44', '35', '43', '34', '18', '6', '5', '17', '13', + '95', '94', '93', '92', '132', '151', + '85', '102', '84', '101', '83', '100', '82', + ] _video_extensions = { '13': '3gp', '17': 'mp4', @@ -64,6 +172,24 @@ class YoutubeIE(InfoExtractor): '44': 'webm', '45': 'webm', '46': 'webm', + + # 3d videos + '82': 'mp4', + '83': 'mp4', + '84': 'mp4', + '85': 'mp4', + '100': 'webm', + '101': 'webm', + '102': 'webm', + + # videos that use m3u8 + '92': 'mp4', + '93': 'mp4', + '94': 'mp4', + '95': 'mp4', + '96': 'mp4', + '132': 'mp4', + '151': 'mp4', } _video_dimensions = { '5': '240x400', @@ -80,7 +206,22 @@ class YoutubeIE(InfoExtractor): '44': '480x854', '45': '720x1280', '46': '1080x1920', + '82': '360p', + '83': '480p', + '84': '720p', + '85': '1080p', + '92': '240p', + '93': '360p', + '94': '480p', + '95': '720p', + '96': '1080p', + '100': '360p', + '101': '480p', + '102': '720p', + '132': '240p', + '151': '72p', } + _3d_itags = ['85', '84', '102', '83', '101', '82', '100'] IE_NAME = u'youtube' _TESTS = [ { @@ -130,6 +271,21 @@ class YoutubeIE(InfoExtractor): u"uploader_id": u"justintimberlakeVEVO" } }, + { + u'url': u'https://www.youtube.com/watch?v=TGi3HqYrWHE', + u'file': u'TGi3HqYrWHE.mp4', + u'note': u'm3u8 video', + u'info_dict': { + u'title': u'Triathlon - Men - London 2012 Olympic Games', + u'description': u'- Men - TR02 - Triathlon - 07 August 2012 - London 2012 Olympic Games', + u'uploader': u'olympic', + u'upload_date': u'20120807', + u'uploader_id': u'olympic', + }, + u'params': { + u'skip_download': True, + }, + }, ] @@ -139,10 +295,6 @@ class YoutubeIE(InfoExtractor): if YoutubePlaylistIE.suitable(url) or YoutubeSubscriptionsIE.suitable(url): return False return re.match(cls._VALID_URL, url, re.VERBOSE) is not None - def report_lang(self): - """Report attempt to set language.""" - self.to_screen(u'Setting language') - def report_video_webpage_download(self, video_id): """Report attempt to download video webpage.""" self.to_screen(u'%s: Downloading video webpage' % video_id) @@ -179,28 +331,42 @@ class YoutubeIE(InfoExtractor): def _decrypt_signature(self, s): """Turn the encrypted s field into a working signature""" - if len(s) == 88: + if len(s) == 92: + return s[25] + s[3:25] + s[0] + s[26:42] + s[79] + s[43:79] + s[91] + s[80:83] + elif len(s) == 90: + return s[25] + s[3:25] + s[2] + s[26:40] + s[77] + s[41:77] + s[89] + s[78:81] + elif len(s) == 88: return s[48] + s[81:67:-1] + s[82] + s[66:62:-1] + s[85] + s[61:48:-1] + s[67] + s[47:12:-1] + s[3] + s[11:3:-1] + s[2] + s[12] elif len(s) == 87: - return s[62] + s[82:62:-1] + s[83] + s[61:52:-1] + s[0] + s[51:2:-1] + return s[4:23] + s[86] + s[24:85] elif len(s) == 86: - return s[2:63] + s[82] + s[64:82] + s[63] + return s[83:85] + s[26] + s[79:46:-1] + s[85] + s[45:36:-1] + s[30] + s[35:30:-1] + s[46] + s[29:26:-1] + s[82] + s[25:1:-1] elif len(s) == 85: - return s[76] + s[82:76:-1] + s[83] + s[75:60:-1] + s[0] + s[59:50:-1] + s[1] + s[49:2:-1] + return s[2:8] + s[0] + s[9:21] + s[65] + s[22:65] + s[84] + s[66:82] + s[21] elif len(s) == 84: return s[83:27:-1] + s[0] + s[26:5:-1] + s[2:0:-1] + s[27] elif len(s) == 83: - return s[6] + s[3:6] + s[33] + s[7:24] + s[0] + s[25:33] + s[53] + s[34:53] + s[24] + s[54:] + return s[:15] + s[80] + s[16:80] + s[15] elif len(s) == 82: return s[36] + s[79:67:-1] + s[81] + s[66:40:-1] + s[33] + s[39:36:-1] + s[40] + s[35] + s[0] + s[67] + s[32:0:-1] + s[34] elif len(s) == 81: - return s[6] + s[3:6] + s[33] + s[7:24] + s[0] + s[25:33] + s[2] + s[34:53] + s[24] + s[54:81] - elif len(s) == 92: - return s[25] + s[3:25] + s[0] + s[26:42] + s[79] + s[43:79] + s[91] + s[80:83]; + return s[56] + s[79:56:-1] + s[41] + s[55:41:-1] + s[80] + s[40:34:-1] + s[0] + s[33:29:-1] + s[34] + s[28:9:-1] + s[29] + s[8:0:-1] + s[9] + elif len(s) == 79: + return s[54] + s[77:54:-1] + s[39] + s[53:39:-1] + s[78] + s[38:34:-1] + s[0] + s[33:29:-1] + s[34] + s[28:9:-1] + s[29] + s[8:0:-1] + s[9] else: raise ExtractorError(u'Unable to decrypt signature, key length %d not supported; retrying might work' % (len(s))) + def _decrypt_signature_age_gate(self, s): + # The videos with age protection use another player, so the algorithms + # can be different. + if len(s) == 86: + return s[2:63] + s[82] + s[64:82] + s[63] + else: + # Fallback to the other algortihms + return self._decrypt_signature(s) + + def _get_available_subtitles(self, video_id): self.report_video_subtitles_download(video_id) request = compat_urllib_request.Request('http://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id) @@ -302,92 +468,9 @@ class YoutubeIE(InfoExtractor): def _print_formats(self, formats): print('Available formats:') for x in formats: - print('%s\t:\t%s\t[%s]' %(x, self._video_extensions.get(x, 'flv'), self._video_dimensions.get(x, '???'))) - - def _real_initialize(self): - if self._downloader is None: - return - - # Set language - request = compat_urllib_request.Request(self._LANG_URL) - try: - self.report_lang() - compat_urllib_request.urlopen(request).read() - except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: - self._downloader.report_warning(u'unable to set language: %s' % compat_str(err)) - return - - (username, password) = self._get_login_info() - - # No authentication to be performed - if username is None: - return - - request = compat_urllib_request.Request(self._LOGIN_URL) - try: - login_page = compat_urllib_request.urlopen(request).read().decode('utf-8') - except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: - self._downloader.report_warning(u'unable to fetch login page: %s' % compat_str(err)) - return - - 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) - - # Log in - login_form_strs = { - u'continue': u'https://www.youtube.com/signin?action_handle_signin=true&feature=sign_in_button&hl=en_US&nomobiletemp=1', - u'Email': username, - u'GALX': galx, - u'Passwd': password, - u'PersistentCookie': u'yes', - u'_utf8': u'霱', - u'bgresponse': u'js_disabled', - u'checkConnection': u'', - u'checkedDomains': u'youtube', - u'dnConn': u'', - u'dsh': dsh, - u'pstMsg': u'0', - u'rmShown': u'1', - u'secTok': u'', - u'signIn': u'Sign in', - u'timeStmp': u'', - u'service': u'youtube', - u'uilel': u'3', - u'hl': u'en_US', - } - # Convert to UTF-8 *before* urlencode because Python 2.x's urlencode - # chokes on unicode - login_form = dict((k.encode('utf-8'), v.encode('utf-8')) for k,v in login_form_strs.items()) - login_data = compat_urllib_parse.urlencode(login_form).encode('ascii') - request = compat_urllib_request.Request(self._LOGIN_URL, login_data) - try: - self.report_login() - login_results = compat_urllib_request.urlopen(request).read().decode('utf-8') - if re.search(r'(?i)<form[^>]* id="gaia_loginform"', login_results) is not None: - self._downloader.report_warning(u'unable to log in: bad username or password') - return - except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: - self._downloader.report_warning(u'unable to log in: %s' % compat_str(err)) - return - - # Confirm age - age_form = { - 'next_url': '/', - 'action_confirm': 'Confirm', - } - request = compat_urllib_request.Request(self._AGE_URL, compat_urllib_parse.urlencode(age_form)) - try: - self.report_age_confirmation() - compat_urllib_request.urlopen(request).read().decode('utf-8') - except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: - raise ExtractorError(u'Unable to confirm age: %s' % compat_str(err)) + print('%s\t:\t%s\t[%s]%s' %(x, self._video_extensions.get(x, 'flv'), + self._video_dimensions.get(x, '???'), + ' (3D)' if x in self._3d_itags else '')) def _extract_id(self, url): mobj = re.match(self._VALID_URL, url, re.VERBOSE) @@ -396,6 +479,57 @@ class YoutubeIE(InfoExtractor): video_id = mobj.group(2) return video_id + def _get_video_url_list(self, url_map): + """ + Transform a dictionary in the format {itag:url} to a list of (itag, url) + with the requested formats. + """ + req_format = self._downloader.params.get('format', None) + format_limit = self._downloader.params.get('format_limit', None) + available_formats = self._available_formats_prefer_free if self._downloader.params.get('prefer_free_formats', False) else self._available_formats + if format_limit is not None and format_limit in available_formats: + format_list = available_formats[available_formats.index(format_limit):] + else: + format_list = available_formats + existing_formats = [x for x in format_list if x in url_map] + if len(existing_formats) == 0: + raise ExtractorError(u'no known formats available for video') + if self._downloader.params.get('listformats', None): + self._print_formats(existing_formats) + return + if req_format is None or req_format == 'best': + video_url_list = [(existing_formats[0], url_map[existing_formats[0]])] # Best quality + elif req_format == 'worst': + video_url_list = [(existing_formats[-1], url_map[existing_formats[-1]])] # worst quality + elif req_format in ('-1', 'all'): + video_url_list = [(f, url_map[f]) for f in existing_formats] # All formats + else: + # Specific formats. We pick the first in a slash-delimeted sequence. + # For example, if '1/2/3/4' is requested and '2' and '4' are available, we pick '2'. + req_formats = req_format.split('/') + video_url_list = None + for rf in req_formats: + if rf in url_map: + video_url_list = [(rf, url_map[rf])] + break + if video_url_list is None: + raise ExtractorError(u'requested format not available') + return video_url_list + + def _extract_from_m3u8(self, manifest_url, video_id): + url_map = {} + def _get_urls(_manifest): + lines = _manifest.split('\n') + urls = filter(lambda l: l and not l.startswith('#'), + lines) + return urls + manifest = self._download_webpage(manifest_url, video_id, u'Downloading formats manifest') + formats_urls = _get_urls(manifest) + for format_url in formats_urls: + itag = self._search_regex(r'itag/(\d+?)/', format_url, 'itag') + url_map[itag] = format_url + return url_map + def _real_extract(self, url): if re.match(r'(?:https?://)?[^/]+/watch\?feature=[a-z_]+$', url): self._downloader.report_warning(u'Did you forget to quote the URL? Remember that & is a meta-character in most shells, so you want to put the URL in quotes, like youtube-dl \'http://www.youtube.com/watch?feature=foo&v=BaW_jenozKc\' (or simply youtube-dl BaW_jenozKc ).') @@ -550,7 +684,6 @@ class YoutubeIE(InfoExtractor): video_duration = compat_urllib_parse.unquote_plus(video_info['length_seconds'][0]) # Decide which formats to download - req_format = self._downloader.params.get('format', None) try: mobj = re.search(r';ytplayer.config = ({.*?});', video_webpage) @@ -585,8 +718,8 @@ class YoutubeIE(InfoExtractor): s = url_data['s'][0] if age_gate: player_version = self._search_regex(r'ad3-(.+?)\.swf', - video_info['ad3_module'][0], 'flash player', - fatal=False) + video_info['ad3_module'][0] if 'ad3_module' in video_info else 'NOT FOUND', + 'flash player', fatal=False) player = 'flash player %s' % player_version else: player = u'html5 player %s' % self._search_regex(r'html5player-(.+?)\.js', video_webpage, @@ -594,41 +727,25 @@ class YoutubeIE(InfoExtractor): parts_sizes = u'.'.join(compat_str(len(part)) for part in s.split('.')) self.to_screen(u'encrypted signature length %d (%s), itag %s, %s' % (len(s), parts_sizes, url_data['itag'][0], player)) - signature = self._decrypt_signature(url_data['s'][0]) + encrypted_sig = url_data['s'][0] + if age_gate: + signature = self._decrypt_signature_age_gate(encrypted_sig) + else: + signature = self._decrypt_signature(encrypted_sig) url += '&signature=' + signature if 'ratebypass' not in url: url += '&ratebypass=yes' url_map[url_data['itag'][0]] = url - - format_limit = self._downloader.params.get('format_limit', None) - available_formats = self._available_formats_prefer_free if self._downloader.params.get('prefer_free_formats', False) else self._available_formats - if format_limit is not None and format_limit in available_formats: - format_list = available_formats[available_formats.index(format_limit):] - else: - format_list = available_formats - existing_formats = [x for x in format_list if x in url_map] - if len(existing_formats) == 0: - raise ExtractorError(u'no known formats available for video') - if self._downloader.params.get('listformats', None): - self._print_formats(existing_formats) + video_url_list = self._get_video_url_list(url_map) + if not video_url_list: return - if req_format is None or req_format == 'best': - video_url_list = [(existing_formats[0], url_map[existing_formats[0]])] # Best quality - elif req_format == 'worst': - video_url_list = [(existing_formats[-1], url_map[existing_formats[-1]])] # worst quality - elif req_format in ('-1', 'all'): - video_url_list = [(f, url_map[f]) for f in existing_formats] # All formats - else: - # Specific formats. We pick the first in a slash-delimeted sequence. - # For example, if '1/2/3/4' is requested and '2' and '4' are available, we pick '2'. - req_formats = req_format.split('/') - video_url_list = None - for rf in req_formats: - if rf in url_map: - video_url_list = [(rf, url_map[rf])] - break - if video_url_list is None: - raise ExtractorError(u'requested format not available') + elif video_info.get('hlsvp'): + manifest_url = video_info['hlsvp'][0] + url_map = self._extract_from_m3u8(manifest_url, video_id) + video_url_list = self._get_video_url_list(url_map) + if not video_url_list: + return + else: raise ExtractorError(u'no conn or url_encoded_fmt_stream_map information found in video info') @@ -637,8 +754,9 @@ class YoutubeIE(InfoExtractor): # Extension video_extension = self._video_extensions.get(format_param, 'flv') - video_format = '{0} - {1}'.format(format_param if format_param else video_extension, - self._video_dimensions.get(format_param, '???')) + video_format = '{0} - {1}{2}'.format(format_param if format_param else video_extension, + self._video_dimensions.get(format_param, '???'), + ' (3D)' if format_param in self._3d_itags else '') results.append({ 'id': video_id, @@ -668,10 +786,10 @@ class YoutubePlaylistIE(InfoExtractor): \? (?:.*?&)*? (?:p|a|list)= | p/ ) - ((?:PL|EC|UU)?[0-9A-Za-z-_]{10,}) + ((?:PL|EC|UU|FL)?[0-9A-Za-z-_]{10,}) .* | - ((?:PL|EC|UU)[0-9A-Za-z-_]{10,}) + ((?:PL|EC|UU|FL)[0-9A-Za-z-_]{10,}) )""" _TEMPLATE_URL = 'https://gdata.youtube.com/feeds/api/playlists/%s?max-results=%i&start-index=%i&v=2&alt=json&safeSearch=none' _MAX_RESULTS = 50 @@ -690,11 +808,14 @@ class YoutubePlaylistIE(InfoExtractor): # Download playlist videos from API playlist_id = mobj.group(1) or mobj.group(2) - page_num = 1 videos = [] - while True: - url = self._TEMPLATE_URL % (playlist_id, self._MAX_RESULTS, self._MAX_RESULTS * (page_num - 1) + 1) + for page_num in itertools.count(1): + start_index = self._MAX_RESULTS * (page_num - 1) + 1 + if start_index >= 1000: + self._downloader.report_warning(u'Max number of results reached') + break + url = self._TEMPLATE_URL % (playlist_id, self._MAX_RESULTS, start_index) page = self._download_webpage(url, playlist_id, u'Downloading page #%s' % page_num) try: @@ -714,10 +835,6 @@ class YoutubePlaylistIE(InfoExtractor): if 'media$group' in entry and 'media$player' in entry['media$group']: videos.append((index, entry['media$group']['media$player']['url'])) - if len(response['feed']['entry']) < self._MAX_RESULTS: - break - page_num += 1 - videos = [v[1] for v in sorted(videos)] url_results = [self.url_result(vurl, 'Youtube') for vurl in videos] @@ -729,7 +846,7 @@ class YoutubeChannelIE(InfoExtractor): _VALID_URL = r"^(?:https?://)?(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/([0-9A-Za-z_-]+)" _TEMPLATE_URL = 'http://www.youtube.com/channel/%s/videos?sort=da&flow=list&view=0&page=%s&gl=US&hl=en' _MORE_PAGES_INDICATOR = 'yt-uix-load-more' - _MORE_PAGES_URL = 'http://www.youtube.com/channel_ajax?action_load_more_videos=1&flow=list&paging=%s&view=0&sort=da&channel_id=%s' + _MORE_PAGES_URL = 'http://www.youtube.com/c4_browse_ajax?action_load_more_videos=1&flow=list&paging=%s&view=0&sort=da&channel_id=%s' IE_NAME = u'youtube:channel' def extract_videos_from_page(self, page): @@ -760,9 +877,7 @@ class YoutubeChannelIE(InfoExtractor): # Download any subsequent channel pages using the json-based channel_ajax query if self._MORE_PAGES_INDICATOR in page: - while True: - pagenum = pagenum + 1 - + for pagenum in itertools.count(1): url = self._MORE_PAGES_URL % (pagenum, channel_id) page = self._download_webpage(url, channel_id, u'Downloading page #%s' % pagenum) @@ -805,9 +920,8 @@ class YoutubeUserIE(InfoExtractor): # all of them. video_ids = [] - pagenum = 0 - while True: + for pagenum in itertools.count(0): start_index = pagenum * self._GDATA_PAGE_SIZE + 1 gdata_url = self._GDATA_URL % (username, self._GDATA_PAGE_SIZE, start_index) @@ -832,8 +946,6 @@ class YoutubeUserIE(InfoExtractor): if len(ids_in_page) < self._GDATA_PAGE_SIZE: break - pagenum += 1 - urls = ['http://www.youtube.com/watch?v=%s' % video_id for video_id in video_ids] url_results = [self.url_result(rurl, 'Youtube') for rurl in urls] return [self.playlist_result(url_results, playlist_title = username)] @@ -896,38 +1008,75 @@ class YoutubeShowIE(InfoExtractor): return [self.url_result('https://www.youtube.com' + season.group(1), 'YoutubePlaylist') for season in m_seasons] -class YoutubeSubscriptionsIE(YoutubeIE): - """It's a subclass of YoutubeIE because we need to login""" - IE_DESC = u'YouTube.com subscriptions feed, "ytsubs" keyword(requires authentication)' - _VALID_URL = r'https?://www\.youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?' - IE_NAME = u'youtube:subscriptions' - _FEED_TEMPLATE = 'http://www.youtube.com/feed_ajax?action_load_system_feed=1&feed_name=subscriptions&paging=%s' +class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor): + """ + Base class for extractors that fetch info from + http://www.youtube.com/feed_ajax + Subclasses must define the _FEED_NAME and _PLAYLIST_TITLE properties. + """ + _LOGIN_REQUIRED = True _PAGING_STEP = 30 + # use action_load_personal_feed instead of action_load_system_feed + _PERSONAL_FEED = False - # Overwrite YoutubeIE properties we don't want - _TESTS = [] - @classmethod - def suitable(cls, url): - return re.match(cls._VALID_URL, url) is not None + @property + def _FEED_TEMPLATE(self): + action = 'action_load_system_feed' + if self._PERSONAL_FEED: + action = 'action_load_personal_feed' + return 'http://www.youtube.com/feed_ajax?%s=1&feed_name=%s&paging=%%s' % (action, self._FEED_NAME) + + @property + def IE_NAME(self): + return u'youtube:%s' % self._FEED_NAME def _real_initialize(self): - (username, password) = self._get_login_info() - if username is None: - raise ExtractorError(u'No login info available, needed for downloading the Youtube subscriptions.', expected=True) - super(YoutubeSubscriptionsIE, self)._real_initialize() + self._login() def _real_extract(self, url): feed_entries = [] # The step argument is available only in 2.7 or higher for i in itertools.count(0): paging = i*self._PAGING_STEP - info = self._download_webpage(self._FEED_TEMPLATE % paging, 'feed', + info = self._download_webpage(self._FEED_TEMPLATE % paging, + u'%s feed' % self._FEED_NAME, u'Downloading page %s' % i) info = json.loads(info) feed_html = info['feed_html'] - m_ids = re.finditer(r'"/watch\?v=(.*?)"', feed_html) + m_ids = re.finditer(r'"/watch\?v=(.*?)["&]', feed_html) ids = orderedSet(m.group(1) for m in m_ids) feed_entries.extend(self.url_result(id, 'Youtube') for id in ids) if info['paging'] is None: break - return self.playlist_result(feed_entries, playlist_title='Youtube Subscriptions') + return self.playlist_result(feed_entries, playlist_title=self._PLAYLIST_TITLE) + +class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor): + IE_DESC = u'YouTube.com subscriptions feed, "ytsubs" keyword(requires authentication)' + _VALID_URL = r'https?://www\.youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?' + _FEED_NAME = 'subscriptions' + _PLAYLIST_TITLE = u'Youtube Subscriptions' + +class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor): + IE_DESC = u'YouTube.com recommended videos, "ytrec" keyword (requires authentication)' + _VALID_URL = r'https?://www\.youtube\.com/feed/recommended|:ytrec(?:ommended)?' + _FEED_NAME = 'recommended' + _PLAYLIST_TITLE = u'Youtube Recommended videos' + +class YoutubeWatchLaterIE(YoutubeFeedsInfoExtractor): + IE_DESC = u'Youtube watch later list, "ytwatchlater" keyword (requires authentication)' + _VALID_URL = r'https?://www\.youtube\.com/feed/watch_later|:ytwatchlater' + _FEED_NAME = 'watch_later' + _PLAYLIST_TITLE = u'Youtube Watch Later' + _PAGING_STEP = 100 + _PERSONAL_FEED = True + +class YoutubeFavouritesIE(YoutubeBaseInfoExtractor): + IE_NAME = u'youtube:favorites' + IE_DESC = u'YouTube.com favourite videos, "ytfav" keyword (requires authentication)' + _VALID_URL = r'https?://www\.youtube\.com/my_favorites|:ytfav(?:o?rites)?' + _LOGIN_REQUIRED = True + + def _real_extract(self, url): + webpage = self._download_webpage('https://www.youtube.com/my_favorites', 'Youtube Favourites videos') + playlist_id = self._search_regex(r'list=(.+?)["&]', webpage, u'favourites playlist id') + return self.url_result(playlist_id, 'YoutubePlaylist') diff --git a/youtube_dl/utils.py b/youtube_dl/utils.py index cf2ea654e..59eeaf4a8 100644 --- a/youtube_dl/utils.py +++ b/youtube_dl/utils.py @@ -207,7 +207,7 @@ if sys.version_info >= (2,7): def find_xpath_attr(node, xpath, key, val): """ Find the xpath xpath[@key=val] """ assert re.match(r'^[a-zA-Z]+$', key) - assert re.match(r'^[a-zA-Z@]*$', val) + assert re.match(r'^[a-zA-Z@\s]*$', val) expr = xpath + u"[@%s='%s']" % (key, val) return node.find(expr) else: diff --git a/youtube_dl/version.py b/youtube_dl/version.py index cba83a488..4fc85fac1 100644 --- a/youtube_dl/version.py +++ b/youtube_dl/version.py @@ -1,2 +1,2 @@ -__version__ = '2013.07.17.1' +__version__ = '2013.08.02' |