diff options
Diffstat (limited to 'youtube_dl')
48 files changed, 1435 insertions, 374 deletions
diff --git a/youtube_dl/PostProcessor.py b/youtube_dl/PostProcessor.py index 039e01498..13b56ede5 100644 --- a/youtube_dl/PostProcessor.py +++ b/youtube_dl/PostProcessor.py @@ -2,9 +2,15 @@ import os import subprocess import sys import time -import datetime -from .utils import * + +from .utils import ( + compat_subprocess_get_DEVNULL, + encodeFilename, + PostProcessingError, + shell_quote, + subtitles_filename, +) class PostProcessor(object): diff --git a/youtube_dl/YoutubeDL.py b/youtube_dl/YoutubeDL.py index c8054544a..7f73ea360 100644 --- a/youtube_dl/YoutubeDL.py +++ b/youtube_dl/YoutubeDL.py @@ -91,7 +91,7 @@ class YoutubeDL(object): downloadarchive: File name of a file where all downloads are recorded. Videos already present in the file are not downloaded again. - + The following parameters are not used by YoutubeDL itself, they are used by the FileDownloader: nopart, updatetime, buffersize, ratelimit, min_filesize, max_filesize, test, @@ -216,10 +216,10 @@ class YoutubeDL(object): If stderr is a tty file the 'WARNING:' will be colored ''' if sys.stderr.isatty() and os.name != 'nt': - _msg_header=u'\033[0;33mWARNING:\033[0m' + _msg_header = u'\033[0;33mWARNING:\033[0m' else: - _msg_header=u'WARNING:' - warning_message=u'%s %s' % (_msg_header,message) + _msg_header = u'WARNING:' + warning_message = u'%s %s' % (_msg_header, message) self.to_stderr(warning_message) def report_error(self, message, tb=None): @@ -234,19 +234,6 @@ class YoutubeDL(object): error_message = u'%s %s' % (_msg_header, message) self.trouble(error_message, tb) - def slow_down(self, start_time, byte_counter): - """Sleep if the download speed is over the rate limit.""" - rate_limit = self.params.get('ratelimit', None) - if rate_limit is None or byte_counter == 0: - return - now = time.time() - elapsed = now - start_time - if elapsed <= 0.0: - return - speed = float(byte_counter) / elapsed - if speed > rate_limit: - time.sleep((byte_counter - rate_limit * (now - start_time)) / rate_limit) - def report_writedescription(self, descfn): """ Report that the description file is being written """ self.to_screen(u'[info] Writing video description to: ' + descfn) @@ -285,16 +272,18 @@ class YoutubeDL(object): autonumber_size = 5 autonumber_templ = u'%0' + str(autonumber_size) + u'd' template_dict['autonumber'] = autonumber_templ % self._num_downloads - if template_dict['playlist_index'] is not None: + if template_dict.get('playlist_index') is not None: template_dict['playlist_index'] = u'%05d' % template_dict['playlist_index'] - sanitize = lambda k,v: sanitize_filename( + sanitize = lambda k, v: sanitize_filename( u'NA' if v is None else compat_str(v), restricted=self.params.get('restrictfilenames'), - is_id=(k==u'id')) - template_dict = dict((k, sanitize(k, v)) for k,v in template_dict.items()) + is_id=(k == u'id')) + template_dict = dict((k, sanitize(k, v)) + for k, v in template_dict.items()) - filename = self.params['outtmpl'] % template_dict + tmpl = os.path.expanduser(self.params['outtmpl']) + filename = tmpl % template_dict return filename except KeyError as err: self.report_error(u'Erroneous output template') @@ -328,14 +317,14 @@ class YoutubeDL(object): return (u'%(title)s has already been recorded in archive' % info_dict) return None - + def extract_info(self, url, download=True, ie_key=None, extra_info={}): ''' Returns a list with a dictionary for each video we find. If 'download', also downloads the videos. extra_info is a dict containing the extra values to add to each result ''' - + if ie_key: ies = [self.get_info_extractor(ie_key)] else: @@ -377,7 +366,7 @@ class YoutubeDL(object): raise else: self.report_error(u'no suitable InfoExtractor: %s' % url) - + def process_ie_result(self, ie_result, download=True, extra_info={}): """ Take the result of the ie(may be modified) and resolve all unresolved @@ -390,13 +379,7 @@ class YoutubeDL(object): result_type = ie_result.get('_type', 'video') # If not given we suppose it's a video, support the default old system if result_type == 'video': ie_result.update(extra_info) - if 'playlist' not in ie_result: - # It isn't part of a playlist - ie_result['playlist'] = None - ie_result['playlist_index'] = None - if download: - self.process_info(ie_result) - return ie_result + return self.process_video_result(ie_result) elif result_type == 'url': # We have to add extra_info to the results because it may be # contained in a playlist @@ -407,7 +390,7 @@ class YoutubeDL(object): elif result_type == 'playlist': # We process each entry in the playlist playlist = ie_result.get('title', None) or ie_result.get('id', None) - self.to_screen(u'[download] Downloading playlist: %s' % playlist) + self.to_screen(u'[download] Downloading playlist: %s' % playlist) playlist_results = [] @@ -425,12 +408,12 @@ class YoutubeDL(object): self.to_screen(u"[%s] playlist '%s': Collected %d video ids (downloading %d of them)" % (ie_result['extractor'], playlist, n_all_entries, n_entries)) - for i,entry in enumerate(entries,1): - self.to_screen(u'[download] Downloading video #%s of %s' %(i, n_entries)) + for i, entry in enumerate(entries, 1): + self.to_screen(u'[download] Downloading video #%s of %s' % (i, n_entries)) extra = { - 'playlist': playlist, - 'playlist_index': i + playliststart, - } + 'playlist': playlist, + 'playlist_index': i + playliststart, + } if not 'extractor' in entry: # We set the extractor, if it's an url it will be set then to # the new extractor, but if it's already a video we must make @@ -454,6 +437,107 @@ class YoutubeDL(object): else: raise Exception('Invalid result type: %s' % result_type) + def select_format(self, format_spec, available_formats): + if format_spec == 'best' or format_spec is None: + return available_formats[-1] + elif format_spec == 'worst': + return available_formats[0] + else: + extensions = [u'mp4', u'flv', u'webm', u'3gp'] + if format_spec in extensions: + filter_f = lambda f: f['ext'] == format_spec + else: + filter_f = lambda f: f['format_id'] == format_spec + matches = list(filter(filter_f, available_formats)) + if matches: + return matches[-1] + return None + + def process_video_result(self, info_dict, download=True): + assert info_dict.get('_type', 'video') == 'video' + + if 'playlist' not in info_dict: + # It isn't part of a playlist + info_dict['playlist'] = None + info_dict['playlist_index'] = None + + # This extractors handle format selection themselves + if info_dict['extractor'] in [u'youtube', u'Youku']: + if download: + self.process_info(info_dict) + return info_dict + + # We now pick which formats have to be downloaded + if info_dict.get('formats') is None: + # There's only one format available + formats = [info_dict] + else: + formats = info_dict['formats'] + + # We check that all the formats have the format and format_id fields + for (i, format) in enumerate(formats): + if format.get('format_id') is None: + format['format_id'] = compat_str(i) + if format.get('format') is None: + format['format'] = u'{id} - {res}{note}'.format( + id=format['format_id'], + res=self.format_resolution(format), + note=u' ({0})'.format(format['format_note']) if format.get('format_note') is not None else '', + ) + # Automatically determine file extension if missing + if 'ext' not in format: + format['ext'] = determine_ext(format['url']) + + if self.params.get('listformats', None): + self.list_formats(info_dict) + return + + format_limit = self.params.get('format_limit', None) + if format_limit: + formats = list(takewhile_inclusive( + lambda f: f['format_id'] != format_limit, formats + )) + if self.params.get('prefer_free_formats'): + def _free_formats_key(f): + try: + ext_ord = [u'flv', u'mp4', u'webm'].index(f['ext']) + except ValueError: + ext_ord = -1 + # We only compare the extension if they have the same height and width + return (f.get('height'), f.get('width'), ext_ord) + formats = sorted(formats, key=_free_formats_key) + + req_format = self.params.get('format', 'best') + if req_format is None: + req_format = 'best' + formats_to_download = [] + # The -1 is for supporting YoutubeIE + if req_format in ('-1', 'all'): + formats_to_download = formats + else: + # We can accept formats requestd in the format: 34/5/best, we pick + # the first that is available, starting from left + req_formats = req_format.split('/') + for rf in req_formats: + selected_format = self.select_format(rf, formats) + if selected_format is not None: + formats_to_download = [selected_format] + break + if not formats_to_download: + raise ExtractorError(u'requested format not available', + expected=True) + + if download: + if len(formats_to_download) > 1: + self.to_screen(u'[info] %s: downloading video in %s formats' % (info_dict['id'], len(formats_to_download))) + for format in formats_to_download: + new_info = dict(info_dict) + new_info.update(format) + self.process_info(new_info) + # We update the info dict with the best quality format (backwards compatibility) + info_dict.update(formats_to_download[-1]) + return info_dict + def process_info(self, info_dict): """Process a single resolved IE result.""" @@ -491,9 +575,9 @@ class YoutubeDL(object): if self.params.get('forceurl', False): # For RTMP URLs, also include the playpath compat_print(info_dict['url'] + info_dict.get('play_path', u'')) - if self.params.get('forcethumbnail', False) and 'thumbnail' in info_dict: + if self.params.get('forcethumbnail', False) and info_dict.get('thumbnail') is not None: compat_print(info_dict['thumbnail']) - if self.params.get('forcedescription', False) and 'description' in info_dict: + if self.params.get('forcedescription', False) and info_dict.get('description') is not None: compat_print(info_dict['description']) if self.params.get('forcefilename', False) and filename is not None: compat_print(filename) @@ -529,20 +613,20 @@ class YoutubeDL(object): if self.params.get('writeannotations', False): try: - annofn = filename + u'.annotations.xml' - self.report_writeannotations(annofn) - with io.open(encodeFilename(annofn), 'w', encoding='utf-8') as annofile: - annofile.write(info_dict['annotations']) + annofn = filename + u'.annotations.xml' + self.report_writeannotations(annofn) + with io.open(encodeFilename(annofn), 'w', encoding='utf-8') as annofile: + annofile.write(info_dict['annotations']) except (KeyError, TypeError): self.report_warning(u'There are no annotations to write.') except (OSError, IOError): - self.report_error(u'Cannot write annotations file: ' + annofn) - return + self.report_error(u'Cannot write annotations file: ' + annofn) + return subtitles_are_requested = any([self.params.get('writesubtitles', False), self.params.get('writeautomaticsub')]) - if subtitles_are_requested and 'subtitles' in info_dict and info_dict['subtitles']: + if subtitles_are_requested and 'subtitles' in info_dict and info_dict['subtitles']: # subtitles download errors are already managed as troubles in relevant IE # that way it will silently go on when used with unsupporting IE subtitles = info_dict['subtitles'] @@ -564,7 +648,7 @@ class YoutubeDL(object): infofn = filename + u'.info.json' self.report_writeinfojson(infofn) try: - json_info_dict = dict((k, v) for k,v in info_dict.items() if not k in ['urlhandle']) + json_info_dict = dict((k, v) for k, v in info_dict.items() if not k in ['urlhandle']) write_json_file(json_info_dict, encodeFilename(infofn)) except (OSError, IOError): self.report_error(u'Cannot write metadata to JSON file ' + infofn) @@ -634,7 +718,7 @@ class YoutubeDL(object): keep_video = None for pp in self._pps: try: - keep_video_wish,new_info = pp.run(info) + keep_video_wish, new_info = pp.run(info) if keep_video_wish is not None: if keep_video_wish: keep_video = keep_video_wish @@ -672,3 +756,38 @@ class YoutubeDL(object): vid_id = info_dict['extractor'] + u' ' + info_dict['id'] with locked_file(fn, 'a', encoding='utf-8') as archive_file: archive_file.write(vid_id + u'\n') + + @staticmethod + def format_resolution(format, default='unknown'): + if format.get('_resolution') is not None: + return format['_resolution'] + if format.get('height') is not None: + if format.get('width') is not None: + res = u'%sx%s' % (format['width'], format['height']) + else: + res = u'%sp' % format['height'] + else: + res = default + return res + + def list_formats(self, info_dict): + def line(format): + return (u'%-15s%-10s%-12s%s' % ( + format['format_id'], + format['ext'], + self.format_resolution(format), + format.get('format_note', ''), + ) + ) + + formats = info_dict.get('formats', [info_dict]) + formats_s = list(map(line, formats)) + if len(formats) > 1: + formats_s[0] += (' ' if formats[0].get('format_note') else '') + '(worst)' + formats_s[-1] += (' ' if formats[-1].get('format_note') else '') + '(best)' + + header_line = line({ + 'format_id': u'format code', 'ext': u'extension', + '_resolution': u'resolution', 'format_note': u'note'}) + self.to_screen(u'[info] Available formats for %s:\n%s\n%s' % + (info_dict['id'], header_line, u"\n".join(formats_s))) diff --git a/youtube_dl/__init__.py b/youtube_dl/__init__.py index fb1270ea2..48ffcbf8e 100644 --- a/youtube_dl/__init__.py +++ b/youtube_dl/__init__.py @@ -31,6 +31,7 @@ __authors__ = ( 'Huarong Huo', 'Ismael Mejía', 'Steffan \'Ruirize\' James', + 'Andras Elso', ) __license__ = 'Public Domain' @@ -46,17 +47,43 @@ import shlex import socket import subprocess import sys -import warnings +import traceback import platform -from .utils import * +from .utils import ( + compat_cookiejar, + compat_print, + compat_str, + compat_urllib_request, + DateRange, + decodeOption, + determine_ext, + DownloadError, + get_cachedir, + make_HTTPS_handler, + MaxDownloadsReached, + platform_name, + preferredencoding, + SameFileError, + std_headers, + write_string, + YoutubeDLHandler, +) from .update import update_self from .version import __version__ -from .FileDownloader import * +from .FileDownloader import ( + FileDownloader, +) from .extractor import gen_extractors from .YoutubeDL import YoutubeDL -from .PostProcessor import * +from .PostProcessor import ( + FFmpegMetadataPP, + FFmpegVideoConvertor, + FFmpegExtractAudioPP, + FFmpegEmbedSubtitlePP, +) + def parseOpts(overrideArguments=None): def _readOptions(filename_bytes): @@ -106,7 +133,7 @@ def parseOpts(overrideArguments=None): def _hide_login_info(opts): opts = list(opts) - for private_opt in ['-p', '--password', '-u', '--username']: + for private_opt in ['-p', '--password', '-u', '--username', '--video-password']: try: i = opts.index(private_opt) opts[i+1] = '<PRIVATE>' @@ -152,6 +179,9 @@ def parseOpts(overrideArguments=None): 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, for example to to skip unavailable videos in a playlist', default=False) + general.add_option('--abort-on-error', + action='store_false', dest='ignoreerrors', + help='Abort downloading of further videos (in the playlist or the command line) if an error occurs') general.add_option('--dump-user-agent', action='store_true', dest='dump_user_agent', help='display the current browser identification', default=False) @@ -169,7 +199,7 @@ def parseOpts(overrideArguments=None): general.add_option('--proxy', dest='proxy', default=None, help='Use the specified HTTP/HTTPS proxy', metavar='URL') general.add_option('--no-check-certificate', action='store_true', dest='no_check_certificate', default=False, help='Suppress HTTPS certificate validation.') general.add_option( - '--cache-dir', dest='cachedir', default=get_cachedir(), + '--cache-dir', dest='cachedir', default=get_cachedir(), metavar='DIR', help='Location in the filesystem where youtube-dl can store downloaded information permanently. By default $XDG_CACHE_HOME/youtube-dl or ~/.cache/youtube-dl .') general.add_option( '--no-cache-dir', action='store_const', const=None, dest='cachedir', @@ -208,7 +238,7 @@ def parseOpts(overrideArguments=None): video_format.add_option('-f', '--format', - action='store', dest='format', metavar='FORMAT', + action='store', dest='format', metavar='FORMAT', default='best', help='video format code, specifiy the order of preference using slashes: "-f 22/17/18". "-f mp4" and "-f flv" are also supported') video_format.add_option('--all-formats', action='store_const', dest='format', help='download all available video formats', const='all') @@ -240,11 +270,11 @@ def parseOpts(overrideArguments=None): help='languages of the subtitles to download (optional) separated by commas, use IETF language tags like \'en,pt\'') downloader.add_option('-r', '--rate-limit', - dest='ratelimit', metavar='LIMIT', help='maximum download rate (e.g. 50k or 44.6m)') + dest='ratelimit', metavar='LIMIT', help='maximum download rate in bytes per second (e.g. 50K or 4.2M)') downloader.add_option('-R', '--retries', dest='retries', metavar='RETRIES', help='number of retries (default is %default)', default=10) downloader.add_option('--buffer-size', - dest='buffersize', metavar='SIZE', help='size of download buffer (e.g. 1024 or 16k) (default is %default)', default="1024") + dest='buffersize', metavar='SIZE', help='size of download buffer (e.g. 1024 or 16K) (default is %default)', default="1024") downloader.add_option('--no-resize-buffer', action='store_true', dest='noresizebuffer', help='do not automatically adjust the buffer size. By default, the buffer size is automatically resized from an initial value of SIZE.', default=False) @@ -286,6 +316,9 @@ def parseOpts(overrideArguments=None): verbosity.add_option('--dump-intermediate-pages', action='store_true', dest='dump_intermediate_pages', default=False, help='print downloaded pages to debug problems(very verbose)') + verbosity.add_option('--write-pages', + action='store_true', dest='write_pages', default=False, + help='Write downloaded pages to files in the current directory') verbosity.add_option('--youtube-print-sig-code', action='store_true', dest='youtube_print_sig_code', default=False, help=optparse.SUPPRESS_HELP) @@ -305,7 +338,10 @@ def parseOpts(overrideArguments=None): help=('output filename template. Use %(title)s to get the title, ' '%(uploader)s for the uploader name, %(uploader_id)s for the uploader nickname if different, ' '%(autonumber)s to get an automatically incremented number, ' - '%(ext)s for the filename extension, %(upload_date)s for the upload date (YYYYMMDD), ' + '%(ext)s for the filename extension, ' + '%(format)s for the format description (like "22 - 1280x720" or "HD"),' + '%(format_id)s for the unique id of the format (like Youtube\'s itags: "137"),' + '%(upload_date)s for the upload date (YYYYMMDD), ' '%(extractor)s for the provider (youtube, metacafe, etc), ' '%(id)s for the video id , %(playlist)s for the playlist the video is in, ' '%(playlist_index)s for the position in the playlist and %% for a literal percent. ' @@ -619,6 +655,7 @@ def _real_main(argv=None): 'prefer_free_formats': opts.prefer_free_formats, 'verbose': opts.verbose, 'dump_intermediate_pages': opts.dump_intermediate_pages, + 'write_pages': opts.write_pages, 'test': opts.test, 'keepvideo': opts.keepvideo, 'min_filesize': opts.min_filesize, @@ -688,7 +725,7 @@ def _real_main(argv=None): if opts.cookiefile is not None: try: jar.save() - except (IOError, OSError) as err: + except (IOError, OSError): sys.exit(u'ERROR: unable to save cookie jar') sys.exit(retcode) diff --git a/youtube_dl/extractor/__init__.py b/youtube_dl/extractor/__init__.py index 748f12e5a..bcf1cce7f 100644 --- a/youtube_dl/extractor/__init__.py +++ b/youtube_dl/extractor/__init__.py @@ -72,6 +72,7 @@ from .jeuxvideo import JeuxVideoIE from .jukebox import JukeboxIE from .justintv import JustinTVIE from .kankan import KankanIE +from .keezmovies import KeezMoviesIE from .kickstarter import KickStarterIE from .keek import KeekIE from .liveleak import LiveLeakIE @@ -82,6 +83,7 @@ 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 @@ -94,6 +96,7 @@ from .ooyala import OoyalaIE from .orf import ORFIE from .pbs import PBSIE from .photobucket import PhotobucketIE +from .pornhub import PornHubIE from .pornotube import PornotubeIE from .rbmaradio import RBMARadioIE from .redtube import RedTubeIE @@ -102,22 +105,27 @@ from .ro220 import Ro220IE from .rottentomatoes import RottenTomatoesIE from .roxwel import RoxwelIE from .rtlnow import RTLnowIE +from .rutube import RutubeIE from .sina import SinaIE from .slashdot import SlashdotIE from .slideshare import SlideshareIE from .sohu import SohuIE from .soundcloud import SoundcloudIE, SoundcloudSetIE, SoundcloudUserIE from .southparkstudios import SouthParkStudiosIE +from .spankwire import SpankwireIE from .spiegel import SpiegelIE from .stanfordoc import StanfordOpenClassroomIE from .statigram import StatigramIE from .steam import SteamIE +from .sztvhu import SztvHuIE from .teamcoco import TeamcocoIE +from .techtalks import TechTalksIE from .ted import TEDIE from .tf1 import TF1IE from .thisav import ThisAVIE from .traileraddict import TrailerAddictIE from .trilulilu import TriluliluIE +from .tube8 import Tube8IE from .tudou import TudouIE from .tumblr import TumblrIE from .tutv import TutvIE @@ -134,7 +142,9 @@ 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 from .wimp import WimpIE from .worldstarhiphop import WorldStarHipHopIE diff --git a/youtube_dl/extractor/addanime.py b/youtube_dl/extractor/addanime.py index 82a785a19..b99d4b966 100644 --- a/youtube_dl/extractor/addanime.py +++ b/youtube_dl/extractor/addanime.py @@ -17,8 +17,8 @@ class AddAnimeIE(InfoExtractor): IE_NAME = u'AddAnime' _TEST = { u'url': u'http://www.add-anime.net/watch_video.php?v=24MR3YO5SAS9', - u'file': u'24MR3YO5SAS9.flv', - u'md5': u'1036a0e0cd307b95bd8a8c3a5c8cfaf1', + u'file': u'24MR3YO5SAS9.mp4', + u'md5': u'72954ea10bc979ab5e2eb288b21425a0', u'info_dict': { u"description": u"One Piece 606", u"title": u"One Piece 606" @@ -31,7 +31,8 @@ class AddAnimeIE(InfoExtractor): video_id = mobj.group('video_id') webpage = self._download_webpage(url, video_id) except ExtractorError as ee: - if not isinstance(ee.cause, compat_HTTPError): + if not isinstance(ee.cause, compat_HTTPError) or \ + ee.cause.code != 503: raise redir_webpage = ee.cause.read().decode('utf-8') @@ -60,16 +61,26 @@ class AddAnimeIE(InfoExtractor): note=u'Confirming after redirect') webpage = self._download_webpage(url, video_id) - video_url = self._search_regex(r"var normal_video_file = '(.*?)';", - webpage, u'video file URL') + formats = [] + for format_id in ('normal', 'hq'): + rex = r"var %s_video_file = '(.*?)';" % re.escape(format_id) + video_url = self._search_regex(rex, webpage, u'video file URLx', + fatal=False) + if not video_url: + continue + formats.append({ + 'format_id': format_id, + 'url': video_url, + }) + if not formats: + raise ExtractorError(u'Cannot find any video format!') video_title = self._og_search_title(webpage) video_description = self._og_search_description(webpage) return { '_type': 'video', 'id': video_id, - 'url': video_url, - 'ext': 'flv', + 'formats': formats, 'title': video_title, 'description': video_description } diff --git a/youtube_dl/extractor/arte.py b/youtube_dl/extractor/arte.py index 5ee8a67b1..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 @@ -170,16 +172,36 @@ 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) + 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 - formats = sorted(formats, key=lambda f: int(f.get('height',-1))) + 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: + sort_key = lambda f: int(f.get('height',-1)) + formats = sorted(formats, key=sort_key) # Prefer videos without subtitles in the same language formats = sorted(formats, key=lambda f: re.match(r'VO(F|A)-STM\1', f.get('versionCode', '')) is None) # Pick the best quality def _format(format_info): + quality = format_info['quality'] + m_quality = re.match(r'\w*? - (\d*)p', quality) + if m_quality is not None: + quality = m_quality.group(1) + if format_info.get('versionCode') is not None: + format_id = u'%s-%s' % (quality, format_info['versionCode']) + else: + format_id = quality info = { + 'format_id': format_id, + 'format_note': format_info.get('versionLibelle'), 'width': format_info.get('width'), 'height': format_info.get('height'), } @@ -192,8 +214,6 @@ class ArteTVPlus7IE(InfoExtractor): info['ext'] = determine_ext(info['url']) return info info_dict['formats'] = [_format(f) for f in formats] - # TODO: Remove when #980 has been merged - info_dict.update(info_dict['formats'][-1]) return info_dict @@ -207,7 +227,7 @@ class ArteTVCreativeIE(ArteTVPlus7IE): u'url': u'http://creative.arte.tv/de/magazin/agentur-amateur-corporate-design', u'file': u'050489-002.mp4', u'info_dict': { - u'title': u'Agentur Amateur #2 - Corporate Design', + u'title': u'Agentur Amateur / Agence Amateur #2 : Corporate Design', }, } diff --git a/youtube_dl/extractor/brightcove.py b/youtube_dl/extractor/brightcove.py index 745212f2f..1392f382a 100644 --- a/youtube_dl/extractor/brightcove.py +++ b/youtube_dl/extractor/brightcove.py @@ -53,6 +53,8 @@ class BrightcoveIE(InfoExtractor): # Fix up some stupid HTML, see https://github.com/rg3/youtube-dl/issues/1553 object_str = re.sub(r'(<param name="[^"]+" value="[^"]+")>', lambda m: m.group(1) + '/>', object_str) + # Fix up some stupid XML, see https://github.com/rg3/youtube-dl/issues/1608 + object_str = object_str.replace(u'<--', u'<!--') object_doc = xml.etree.ElementTree.fromstring(object_str) assert u'BrightcoveExperience' in object_doc.attrib['class'] @@ -96,7 +98,10 @@ class BrightcoveIE(InfoExtractor): playlist_info = self._download_webpage(self._PLAYLIST_URL_TEMPLATE % player_key, player_key, u'Downloading playlist information') - playlist_info = json.loads(playlist_info)['videoList'] + json_data = json.loads(playlist_info) + if 'videoList' not in json_data: + raise ExtractorError(u'Empty playlist') + playlist_info = json_data['videoList'] videos = [self._extract_video_info(video_info) for video_info in playlist_info['mediaCollectionDTO']['videoDTOs']] return self.playlist_result(videos, playlist_id=playlist_info['id'], diff --git a/youtube_dl/extractor/cinemassacre.py b/youtube_dl/extractor/cinemassacre.py index 6925b96c2..2fe1033f0 100644 --- a/youtube_dl/extractor/cinemassacre.py +++ b/youtube_dl/extractor/cinemassacre.py @@ -55,30 +55,30 @@ class CinemassacreIE(InfoExtractor): video_description = None playerdata = self._download_webpage(playerdata_url, video_id) - base_url = self._html_search_regex(r'\'streamer\': \'(?P<base_url>rtmp://.*?)/(?:vod|Cinemassacre)\'', - playerdata, u'base_url') - base_url += '/Cinemassacre/' - # Important: The file names in playerdata are not used by the player and even wrong for some videos - sd_file = 'Cinemassacre-%s_high.mp4' % video_id - hd_file = 'Cinemassacre-%s.mp4' % video_id - video_thumbnail = 'http://image.screenwavemedia.com/Cinemassacre/Cinemassacre-%s_thumb_640x360.jpg' % video_id + url = self._html_search_regex(r'\'streamer\': \'(?P<url>[^\']+)\'', playerdata, u'url') + + sd_file = self._html_search_regex(r'\'file\': \'(?P<sd_file>[^\']+)\'', playerdata, u'sd_file') + hd_file = self._html_search_regex(r'\'?file\'?: "(?P<hd_file>[^"]+)"', playerdata, u'hd_file') + video_thumbnail = self._html_search_regex(r'\'image\': \'(?P<thumbnail>[^\']+)\'', playerdata, u'thumbnail', fatal=False) formats = [ { - 'url': base_url + sd_file, + 'url': url, + 'play_path': 'mp4:' + sd_file, 'ext': 'flv', 'format': 'sd', 'format_id': 'sd', }, { - 'url': base_url + hd_file, + 'url': url, + 'play_path': 'mp4:' + hd_file, 'ext': 'flv', 'format': 'hd', 'format_id': 'hd', }, ] - info = { + return { 'id': video_id, 'title': video_title, 'formats': formats, @@ -86,6 +86,3 @@ class CinemassacreIE(InfoExtractor): 'upload_date': video_date, 'thumbnail': video_thumbnail, } - # TODO: Remove when #980 has been merged - info.update(formats[-1]) - return info diff --git a/youtube_dl/extractor/common.py b/youtube_dl/extractor/common.py index 2a5a85dc6..cef4dce85 100644 --- a/youtube_dl/extractor/common.py +++ b/youtube_dl/extractor/common.py @@ -14,6 +14,8 @@ from ..utils import ( clean_html, compiled_regex_type, ExtractorError, + RegexNotFoundError, + sanitize_filename, unescapeHTML, ) @@ -61,9 +63,12 @@ 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 width and height if missing. + 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") + * format_note Additional info about the format + ("3D" or "DASH video") * width Width of the video, if known * height Height of the video, if known @@ -178,6 +183,17 @@ class InfoExtractor(object): self.to_screen(u'Dumping request to ' + url) dump = base64.b64encode(webpage_bytes).decode('ascii') self._downloader.to_screen(dump) + if self._downloader.params.get('write_pages', False): + try: + url = url_or_request.get_full_url() + except AttributeError: + url = url_or_request + raw_filename = ('%s_%s.dump' % (video_id, url)) + filename = sanitize_filename(raw_filename, restricted=True) + self.to_screen(u'Saving request to ' + filename) + with open(filename, 'wb') as outf: + outf.write(webpage_bytes) + content = webpage_bytes.decode(encoding, 'replace') return (content, urlh) @@ -228,7 +244,7 @@ class InfoExtractor(object): Perform a regex search on the given string, using a single or a list of patterns returning the first matching group. In case of failure return a default value or raise a WARNING or a - ExtractorError, depending on fatal, specifying the field name. + RegexNotFoundError, depending on fatal, specifying the field name. """ if isinstance(pattern, (str, compat_str, compiled_regex_type)): mobj = re.search(pattern, string, flags) @@ -248,7 +264,7 @@ class InfoExtractor(object): elif default is not None: return default elif fatal: - raise ExtractorError(u'Unable to extract %s' % _name) + raise RegexNotFoundError(u'Unable to extract %s' % _name) else: self._downloader.report_warning(u'unable to extract %s; ' u'please report this issue on http://yt-dl.org/bug' % _name) @@ -314,10 +330,10 @@ class InfoExtractor(object): def _og_search_title(self, html, **kargs): return self._og_search_property('title', html, **kargs) - def _og_search_video_url(self, html, name='video url', **kargs): - return self._html_search_regex([self._og_regex('video:secure_url'), - self._og_regex('video')], - html, name, **kargs) + def _og_search_video_url(self, html, name='video url', secure=True, **kargs): + regexes = [self._og_regex('video')] + if secure: regexes.insert(0, self._og_regex('video:secure_url')) + return self._html_search_regex(regexes, html, name, **kargs) def _rta_search(self, html): # See http://www.rtalabel.org/index.php?content=howtofaq#single @@ -365,7 +381,7 @@ class SearchInfoExtractor(InfoExtractor): def _get_n_results(self, query, n): """Get a specified number of results for a query""" - raise NotImplementedError("This method must be implemented by sublclasses") + raise NotImplementedError("This method must be implemented by subclasses") @property def SEARCH_KEY(self): diff --git a/youtube_dl/extractor/dailymotion.py b/youtube_dl/extractor/dailymotion.py index 3aef82bcf..e87690f9d 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): @@ -28,6 +29,15 @@ class DailymotionIE(DailymotionBaseInfoExtractor, SubtitlesInfoExtractor): _VALID_URL = r'(?i)(?:https?://)?(?:www\.)?dailymotion\.[a-z]{2,3}/(?:embed/)?video/([^/]+)' IE_NAME = u'dailymotion' + + _FORMATS = [ + (u'stream_h264_ld_url', u'ld'), + (u'stream_h264_url', u'standard'), + (u'stream_h264_hq_url', u'hq'), + (u'stream_h264_hd_url', u'hd'), + (u'stream_h264_hd1080_url', u'hd180'), + ] + _TESTS = [ { u'url': u'http://www.dailymotion.com/video/x33vw9_tutoriel-de-youtubeur-dl-des-video_tech', @@ -52,6 +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): @@ -60,7 +82,6 @@ class DailymotionIE(DailymotionBaseInfoExtractor, SubtitlesInfoExtractor): video_id = mobj.group(1).split('_')[0].split('?')[0] - video_extension = 'mp4' url = 'http://www.dailymotion.com/video/%s' % video_id # Retrieve video webpage to extract further information @@ -82,7 +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) @@ -99,18 +121,24 @@ class DailymotionIE(DailymotionBaseInfoExtractor, SubtitlesInfoExtractor): msg = 'Couldn\'t get video, Dailymotion says: %s' % info['error']['title'] raise ExtractorError(msg, expected=True) - # TODO: support choosing qualities - - for key in ['stream_h264_hd1080_url','stream_h264_hd_url', - 'stream_h264_hq_url','stream_h264_url', - 'stream_h264_ld_url']: - if info.get(key):#key in info and info[key]: - max_quality = key - self.to_screen(u'Using %s' % key) - break - else: + formats = [] + for (key, format_id) in self._FORMATS: + video_url = info.get(key) + if video_url is not None: + m_size = re.search(r'H264-(\d+)x(\d+)', video_url) + if m_size is not None: + width, height = m_size.group(1), m_size.group(2) + else: + width, height = None, None + formats.append({ + 'url': video_url, + 'ext': 'mp4', + 'format_id': format_id, + 'width': width, + 'height': height, + }) + if not formats: raise ExtractorError(u'Unable to extract video URL') - video_url = info[max_quality] # subtitles video_subtitles = self.extract_subtitles(video_id, webpage) @@ -118,16 +146,16 @@ class DailymotionIE(DailymotionBaseInfoExtractor, SubtitlesInfoExtractor): self._list_available_subtitles(video_id, webpage) return - return [{ + return { 'id': video_id, - 'url': video_url, + 'formats': formats, 'uploader': video_uploader, 'upload_date': video_upload_date, 'title': self._og_search_title(webpage), - 'ext': video_extension, 'subtitles': video_subtitles, - 'thumbnail': info['thumbnail_url'] - }] + 'thumbnail': info['thumbnail_url'], + 'age_limit': age_limit, + } def _get_available_subtitles(self, video_id, webpage): try: diff --git a/youtube_dl/extractor/eighttracks.py b/youtube_dl/extractor/eighttracks.py index cced06811..2cfbcd363 100644 --- a/youtube_dl/extractor/eighttracks.py +++ b/youtube_dl/extractor/eighttracks.py @@ -101,7 +101,7 @@ class EightTracksIE(InfoExtractor): first_url = 'http://8tracks.com/sets/%s/play?player=sm&mix_id=%s&format=jsonh' % (session, mix_id) next_url = first_url res = [] - for i in itertools.count(): + for i in range(track_count): api_json = self._download_webpage(next_url, playlist_id, note=u'Downloading song information %s/%s' % (str(i+1), track_count), errnote=u'Failed to download song information') @@ -116,7 +116,5 @@ class EightTracksIE(InfoExtractor): 'ext': 'm4a', } res.append(info) - if api_data['set']['at_last_track']: - break next_url = 'http://8tracks.com/sets/%s/next?player=sm&mix_id=%s&format=jsonh&track_id=%s' % (session, mix_id, track_data['id']) return res diff --git a/youtube_dl/extractor/exfm.py b/youtube_dl/extractor/exfm.py index 3443f19c5..c74556579 100644 --- a/youtube_dl/extractor/exfm.py +++ b/youtube_dl/extractor/exfm.py @@ -11,14 +11,14 @@ class ExfmIE(InfoExtractor): _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'url': u'http://ex.fm/song/eh359', + u'file': u'44216187.mp3', + u'md5': u'e45513df5631e6d760970b14cc0c11e7', 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"title": u"Test House \"Love Is Not Enough\" (Extended Mix) DeadJournalist Exclusive", + u"uploader": u"deadjournalist", + u'upload_date': u'20120424', + u'description': u'Test House \"Love Is Not Enough\" (Extended Mix) DeadJournalist Exclusive', }, u'note': u'Soundcloud song', }, diff --git a/youtube_dl/extractor/facebook.py b/youtube_dl/extractor/facebook.py index 9d1bc0751..f8bdfc2d3 100644 --- a/youtube_dl/extractor/facebook.py +++ b/youtube_dl/extractor/facebook.py @@ -19,7 +19,8 @@ class FacebookIE(InfoExtractor): """Information Extractor for Facebook""" _VALID_URL = r'^(?:https?://)?(?:\w+\.)?facebook\.com/(?:video/video|photo)\.php\?(?:.*?)v=(?P<ID>\d+)(?:.*)' - _LOGIN_URL = 'https://login.facebook.com/login.php?m&next=http%3A%2F%2Fm.facebook.com%2Fhome.php&' + _LOGIN_URL = 'https://www.facebook.com/login.php?next=http%3A%2F%2Ffacebook.com%2Fhome.php&login_attempt=1' + _CHECKPOINT_URL = 'https://www.facebook.com/checkpoint/?next=http%3A%2F%2Ffacebook.com%2Fhome.php&_fb_noscript=1' _NETRC_MACHINE = 'facebook' IE_NAME = u'facebook' _TEST = { @@ -36,50 +37,56 @@ class FacebookIE(InfoExtractor): """Report attempt to log in.""" self.to_screen(u'Logging in') - def _real_initialize(self): - if self._downloader is None: - return - - useremail = None - password = None - downloader_params = self._downloader.params - - # Attempt to use provided username and password or .netrc data - if downloader_params.get('username', None) is not None: - useremail = downloader_params['username'] - password = downloader_params['password'] - elif downloader_params.get('usenetrc', False): - try: - info = netrc.netrc().authenticators(self._NETRC_MACHINE) - if info is not None: - useremail = info[0] - password = info[2] - else: - raise netrc.NetrcParseError('No authenticators for %s' % self._NETRC_MACHINE) - except (IOError, netrc.NetrcParseError) as err: - self._downloader.report_warning(u'parsing .netrc: %s' % compat_str(err)) - return - + def _login(self): + (useremail, password) = self._get_login_info() if useremail is None: return - # Log in + login_page_req = compat_urllib_request.Request(self._LOGIN_URL) + login_page_req.add_header('Cookie', 'locale=en_US') + self.report_login() + login_page = self._download_webpage(login_page_req, None, note=False, + errnote=u'Unable to download login page') + lsd = self._search_regex(r'"lsd":"(\w*?)"', login_page, u'lsd') + lgnrnd = self._search_regex(r'name="lgnrnd" value="([^"]*?)"', login_page, u'lgnrnd') + login_form = { 'email': useremail, 'pass': password, - 'login': 'Log+In' + 'lsd': lsd, + 'lgnrnd': lgnrnd, + 'next': 'http://facebook.com/home.php', + 'default_persistent': '0', + 'legacy_return': '1', + 'timezone': '-60', + 'trynum': '1', } request = compat_urllib_request.Request(self._LOGIN_URL, compat_urllib_parse.urlencode(login_form)) + request.add_header('Content-Type', 'application/x-www-form-urlencoded') try: - self.report_login() login_results = compat_urllib_request.urlopen(request).read() if re.search(r'<form(.*)name="login"(.*)</form>', login_results) is not None: self._downloader.report_warning(u'unable to log in: bad username/password, or exceded login rate limit (~3/min). Check credentials or wait.') return + + check_form = { + 'fb_dtsg': self._search_regex(r'"fb_dtsg":"(.*?)"', login_results, u'fb_dtsg'), + 'nh': self._search_regex(r'name="nh" value="(\w*?)"', login_results, u'nh'), + 'name_action_selected': 'dont_save', + 'submit[Continue]': self._search_regex(r'<input value="(.*?)" name="submit\[Continue\]"', login_results, u'continue'), + } + check_req = compat_urllib_request.Request(self._CHECKPOINT_URL, compat_urllib_parse.urlencode(check_form)) + check_req.add_header('Content-Type', 'application/x-www-form-urlencoded') + check_response = compat_urllib_request.urlopen(check_req).read() + if re.search(r'id="checkpointSubmitButton"', check_response) is not None: + self._downloader.report_warning(u'Unable to confirm login, you have to login in your brower and authorize the login.') 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 + def _real_initialize(self): + self._login() + def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) if mobj is None: @@ -93,7 +100,13 @@ class FacebookIE(InfoExtractor): AFTER = '.forEach(function(variable) {swf.addVariable(variable[0], variable[1]);});' m = re.search(re.escape(BEFORE) + '(.*?)' + re.escape(AFTER), webpage) if not m: - raise ExtractorError(u'Cannot parse data') + m_msg = re.search(r'class="[^"]*uiInterstitialContent[^"]*"><div>(.*?)</div>', webpage) + if m_msg is not None: + raise ExtractorError( + u'The video is not available, Facebook said: "%s"' % m_msg.group(1), + expected=True) + else: + raise ExtractorError(u'Cannot parse data') data = dict(json.loads(m.group(1))) params_raw = compat_urllib_parse.unquote(data['params']) params = json.loads(params_raw) diff --git a/youtube_dl/extractor/faz.py b/youtube_dl/extractor/faz.py index deaa4ed2d..89ed08db4 100644 --- a/youtube_dl/extractor/faz.py +++ b/youtube_dl/extractor/faz.py @@ -5,8 +5,6 @@ import xml.etree.ElementTree from .common import InfoExtractor from ..utils import ( determine_ext, - clean_html, - get_element_by_attribute, ) @@ -47,12 +45,12 @@ class FazIE(InfoExtractor): 'format_id': code.lower(), }) - descr_html = get_element_by_attribute('class', 'Content Copy', webpage) + descr = self._html_search_regex(r'<p class="Content Copy">(.*?)</p>', webpage, u'description') info = { 'id': video_id, 'title': self._og_search_title(webpage), 'formats': formats, - 'description': clean_html(descr_html), + 'description': descr, 'thumbnail': config.find('STILL/STILL_BIG').text, } # TODO: Remove when #980 has been merged diff --git a/youtube_dl/extractor/generic.py b/youtube_dl/extractor/generic.py index d48c84f8d..2c8fcf5ae 100644 --- a/youtube_dl/extractor/generic.py +++ b/youtube_dl/extractor/generic.py @@ -11,6 +11,8 @@ from ..utils import ( compat_urlparse, ExtractorError, + smuggle_url, + unescapeHTML, ) from .brightcove import BrightcoveIE @@ -23,12 +25,33 @@ class GenericIE(InfoExtractor): { u'url': u'http://www.hodiho.fr/2013/02/regis-plante-sa-jeep.html', u'file': u'13601338388002.mp4', - u'md5': u'85b90ccc9d73b4acd9138d3af4c27f89', + u'md5': u'6e15c93721d7ec9e9ca3fdbf07982cfd', u'info_dict': { u"uploader": u"www.hodiho.fr", u"title": u"R\u00e9gis plante sa Jeep" } }, + # embedded vimeo video + { + u'url': u'http://skillsmatter.com/podcast/home/move-semanticsperfect-forwarding-and-rvalue-references', + u'file': u'22444065.mp4', + u'md5': u'2903896e23df39722c33f015af0666e2', + u'info_dict': { + u'title': u'ACCU 2011: Move Semantics,Perfect Forwarding, and Rvalue references- Scott Meyers- 13/04/2011', + u"uploader_id": u"skillsmatter", + u"uploader": u"Skills Matter", + } + }, + # bandcamp page with custom domain + { + u'url': u'http://bronyrock.com/track/the-pony-mash', + u'file': u'3235767654.mp3', + u'info_dict': { + u'title': u'The Pony Mash', + u'uploader': u'M_Pallante', + }, + u'skip': u'There is a limit of 200 free downloads / month for the test song', + }, ] def report_download_webpage(self, video_id): @@ -127,6 +150,27 @@ class GenericIE(InfoExtractor): bc_url = BrightcoveIE._build_brighcove_url(m_brightcove.group()) return self.url_result(bc_url, 'Brightcove') + # Look for embedded Vimeo player + mobj = re.search( + r'<iframe[^>]+?src="(https?://player.vimeo.com/video/.+?)"', webpage) + if mobj: + player_url = unescapeHTML(mobj.group(1)) + surl = smuggle_url(player_url, {'Referer': url}) + return self.url_result(surl, 'Vimeo') + + # Look for embedded YouTube player + mobj = re.search( + r'<iframe[^>]+?src="(https?://(?:www\.)?youtube.com/embed/.+?)"', webpage) + if mobj: + surl = unescapeHTML(mobj.group(1)) + return self.url_result(surl, 'Youtube') + + # Look for Bandcamp pages with custom domain + mobj = re.search(r'<meta property="og:url"[^>]*?content="(.*?bandcamp\.com.*?)"', webpage) + if mobj is not None: + burl = unescapeHTML(mobj.group(1)) + return self.url_result(burl, 'Bandcamp') + # Start with something easy: JW Player in SWFObject mobj = re.search(r'flashvars: [\'"](?:.*&)?file=(http[^\'"&]*)', webpage) if mobj is None: diff --git a/youtube_dl/extractor/googleplus.py b/youtube_dl/extractor/googleplus.py index ab12d7e93..2570746b2 100644 --- a/youtube_dl/extractor/googleplus.py +++ b/youtube_dl/extractor/googleplus.py @@ -41,9 +41,9 @@ class GooglePlusIE(InfoExtractor): # Extract update date upload_date = self._html_search_regex( - r'''(?x)<a.+?class="o-T-s\s[^"]+"\s+style="display:\s*none"\s*> + r'''(?x)<a.+?class="o-U-s\s[^"]+"\s+style="display:\s*none"\s*> ([0-9]{4}-[0-9]{2}-[0-9]{2})</a>''', - webpage, u'upload date', fatal=False) + webpage, u'upload date', fatal=False, flags=re.VERBOSE) if upload_date: # Convert timestring to a format suitable for filename upload_date = datetime.datetime.strptime(upload_date, "%Y-%m-%d") diff --git a/youtube_dl/extractor/instagram.py b/youtube_dl/extractor/instagram.py index ddc42882a..213aac428 100644 --- a/youtube_dl/extractor/instagram.py +++ b/youtube_dl/extractor/instagram.py @@ -26,7 +26,7 @@ class InstagramIE(InfoExtractor): return [{ 'id': video_id, - 'url': self._og_search_video_url(webpage), + 'url': self._og_search_video_url(webpage, secure=False), 'ext': 'mp4', 'title': u'Video by %s' % uploader_id, 'thumbnail': self._og_search_thumbnail(webpage), diff --git a/youtube_dl/extractor/internetvideoarchive.py b/youtube_dl/extractor/internetvideoarchive.py index 5986459d6..be8e05f53 100644 --- a/youtube_dl/extractor/internetvideoarchive.py +++ b/youtube_dl/extractor/internetvideoarchive.py @@ -19,7 +19,7 @@ class InternetVideoArchiveIE(InfoExtractor): u'info_dict': { u'title': u'SKYFALL', u'description': u'In SKYFALL, Bond\'s loyalty to M is tested as her past comes back to haunt her. As MI6 comes under attack, 007 must track down and destroy the threat, no matter how personal the cost.', - u'duration': 156, + u'duration': 153, }, } @@ -74,7 +74,7 @@ class InternetVideoArchiveIE(InfoExtractor): }) formats = sorted(formats, key=lambda f: f['bitrate']) - info = { + return { 'id': video_id, 'title': item.find('title').text, 'formats': formats, @@ -82,6 +82,3 @@ class InternetVideoArchiveIE(InfoExtractor): 'description': item.find('description').text, 'duration': int(attr['duration']), } - # TODO: Remove when #980 has been merged - info.update(formats[-1]) - return info diff --git a/youtube_dl/extractor/keezmovies.py b/youtube_dl/extractor/keezmovies.py new file mode 100644 index 000000000..5e05900da --- /dev/null +++ b/youtube_dl/extractor/keezmovies.py @@ -0,0 +1,61 @@ +import os +import re + +from .common import InfoExtractor +from ..utils import ( + compat_urllib_parse_urlparse, + compat_urllib_request, + compat_urllib_parse, +) +from ..aes import ( + aes_decrypt_text +) + +class KeezMoviesIE(InfoExtractor): + _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', + u'md5': u'6e297b7e789329923fcf83abb67c9289', + u'info_dict': { + u"title": u"Petite Asian Lady Mai Playing In Bathtub", + 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) + + # embedded video + mobj = re.search(r'href="([^"]+)"></iframe>', webpage) + if mobj: + embedded_url = mobj.group(1) + return self.url_result(embedded_url) + + video_title = self._html_search_regex(r'<h1 [^>]*>([^<]+)', webpage, u'title') + video_url = compat_urllib_parse.unquote(self._html_search_regex(r'video_url=(.+?)&', webpage, u'video_url')) + if webpage.find('encrypted=true')!=-1: + password = self._html_search_regex(r'video_title=(.+?)&', webpage, u'password') + video_url = aes_decrypt_text(video_url, password, 32).decode('utf-8') + path = compat_urllib_parse_urlparse( video_url ).path + extension = os.path.splitext( path )[1][1:] + format = path.split('/')[4].split('_')[:2] + format = "-".join( format ) + + age_limit = self._rta_search(webpage) + + return { + 'id': video_id, + 'title': video_title, + 'url': video_url, + 'ext': extension, + 'format': format, + 'format_id': format, + 'age_limit': age_limit, + } 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 e537648ff..91480ba87 100644 --- a/youtube_dl/extractor/metacafe.py +++ b/youtube_dl/extractor/metacafe.py @@ -20,10 +20,12 @@ 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.flv", + u"file": u"_aUehQsCQtM.mp4", u"info_dict": { u"upload_date": u"20090102", u"title": u"The Electric Company | \"Short I\" | PBS KIDS GO!", @@ -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/nhl.py b/youtube_dl/extractor/nhl.py index e8d43dd13..224f56ac8 100644 --- a/youtube_dl/extractor/nhl.py +++ b/youtube_dl/extractor/nhl.py @@ -90,8 +90,8 @@ class NHLVideocenterIE(NHLBaseInfoExtractor): r'{statusIndex:0,index:0,.*?id:(.*?),'], webpage, u'category id') playlist_title = self._html_search_regex( - r'\?catid=%s">(.*?)</a>' % cat_id, - webpage, u'playlist title', flags=re.DOTALL) + r'tab0"[^>]*?>(.*?)</td>', + webpage, u'playlist title', flags=re.DOTALL).lower().capitalize() data = compat_urllib_parse.urlencode({ 'cid': cat_id, diff --git a/youtube_dl/extractor/nowvideo.py b/youtube_dl/extractor/nowvideo.py index ab52ad401..241cc160b 100644 --- a/youtube_dl/extractor/nowvideo.py +++ b/youtube_dl/extractor/nowvideo.py @@ -20,7 +20,10 @@ class NowVideoIE(InfoExtractor): video_id = mobj.group('id') webpage_url = 'http://www.nowvideo.ch/video/' + video_id + embed_url = 'http://embed.nowvideo.ch/embed.php?v=' + video_id webpage = self._download_webpage(webpage_url, video_id) + embed_page = self._download_webpage(embed_url, video_id, + u'Downloading embed page') self.report_extraction(video_id) @@ -28,7 +31,7 @@ class NowVideoIE(InfoExtractor): webpage, u'video title') video_key = self._search_regex(r'var fkzd="(.*)";', - webpage, u'video key') + embed_page, u'video key') api_call = "http://www.nowvideo.ch/api/player.api.php?file={0}&numOfErrors=0&cid=1&key={1}".format(video_id, video_key) api_response = self._download_webpage(api_call, video_id, diff --git a/youtube_dl/extractor/pornhub.py b/youtube_dl/extractor/pornhub.py new file mode 100644 index 000000000..5e2454f1b --- /dev/null +++ b/youtube_dl/extractor/pornhub.py @@ -0,0 +1,69 @@ +import os +import re + +from .common import InfoExtractor +from ..utils import ( + compat_urllib_parse_urlparse, + compat_urllib_request, + compat_urllib_parse, + unescapeHTML, +) +from ..aes import ( + aes_decrypt_text +) + +class PornHubIE(InfoExtractor): + _VALID_URL = r'^(?:https?://)?(?:www\.)?(?P<url>pornhub\.com/view_video\.php\?viewkey=(?P<videoid>[0-9]+))' + _TEST = { + u'url': u'http://www.pornhub.com/view_video.php?viewkey=648719015', + u'file': u'648719015.mp4', + u'md5': u'882f488fa1f0026f023f33576004a2ed', + u'info_dict': { + u"uploader": u"BABES-COM", + u"title": u"Seductive Indian beauty strips down and fingers her pink pussy", + 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 [^>]+>([^<]+)', webpage, u'title') + video_uploader = self._html_search_regex(r'<b>From: </b>(?:\s|<[^>]*>)*(.+?)<', webpage, u'uploader', fatal=False) + thumbnail = self._html_search_regex(r'"image_url":"([^"]+)', webpage, u'thumbnail', fatal=False) + if thumbnail: + thumbnail = compat_urllib_parse.unquote(thumbnail) + + video_urls = list(map(compat_urllib_parse.unquote , re.findall(r'"quality_[0-9]{3}p":"([^"]+)', webpage))) + if webpage.find('"encrypted":true') != -1: + password = self._html_search_regex(r'"video_title":"([^"]+)', webpage, u'password').replace('+', ' ') + video_urls = list(map(lambda s: aes_decrypt_text(s, password, 32).decode('utf-8'), video_urls)) + + formats = [] + for video_url in video_urls: + path = compat_urllib_parse_urlparse( video_url ).path + extension = os.path.splitext( path )[1][1:] + format = path.split('/')[5].split('_')[:2] + format = "-".join( format ) + formats.append({ + 'url': video_url, + 'ext': extension, + 'format': format, + 'format_id': format, + }) + formats.sort(key=lambda format: list(map(lambda s: s.zfill(6), format['format'].split('-')))) + + return { + 'id': video_id, + 'uploader': video_uploader, + 'title': video_title, + 'thumbnail': thumbnail, + 'formats': formats, + 'age_limit': 18, + } diff --git a/youtube_dl/extractor/pornotube.py b/youtube_dl/extractor/pornotube.py index 5d770ec28..35dc5a9ff 100644 --- a/youtube_dl/extractor/pornotube.py +++ b/youtube_dl/extractor/pornotube.py @@ -16,7 +16,8 @@ class PornotubeIE(InfoExtractor): u'md5': u'374dd6dcedd24234453b295209aa69b6', u'info_dict': { u"upload_date": u"20090708", - u"title": u"Marilyn-Monroe-Bathing" + u"title": u"Marilyn-Monroe-Bathing", + u"age_limit": 18 } } diff --git a/youtube_dl/extractor/redtube.py b/youtube_dl/extractor/redtube.py index 365aade56..994778e16 100644 --- a/youtube_dl/extractor/redtube.py +++ b/youtube_dl/extractor/redtube.py @@ -10,7 +10,8 @@ class RedTubeIE(InfoExtractor): u'file': u'66418.mp4', u'md5': u'7b8c22b5e7098a3e1c09709df1126d2d', u'info_dict': { - u"title": u"Sucked on a toilet" + u"title": u"Sucked on a toilet", + u"age_limit": 18, } } diff --git a/youtube_dl/extractor/rtlnow.py b/youtube_dl/extractor/rtlnow.py index d1b08c9bc..9ac7c3be8 100644 --- a/youtube_dl/extractor/rtlnow.py +++ b/youtube_dl/extractor/rtlnow.py @@ -63,13 +63,12 @@ class RTLnowIE(InfoExtractor): }, }, { - u'url': u'http://www.rtlnitronow.de/recht-ordnung/lebensmittelkontrolle-erlangenordnungsamt-berlin.php?film_id=127367&player=1&season=1', - u'file': u'127367.flv', + u'url': u'http://www.rtlnitronow.de/recht-ordnung/stadtpolizei-frankfurt-gerichtsvollzieher-leipzig.php?film_id=129679&player=1&season=1', + u'file': u'129679.flv', u'info_dict': { - u'upload_date': u'20130926', - u'title': u'Recht & Ordnung - Lebensmittelkontrolle Erlangen/Ordnungsamt...', - u'description': u'Lebensmittelkontrolle Erlangen/Ordnungsamt Berlin', - u'thumbnail': u'http://autoimg.static-fra.de/nitronow/344787/1500x1500/image2.jpg', + u'upload_date': u'20131016', + u'title': u'Recht & Ordnung - Stadtpolizei Frankfurt/ Gerichtsvollzieher...', + u'description': u'Stadtpolizei Frankfurt/ Gerichtsvollzieher Leipzig', }, u'params': { u'skip_download': True, diff --git a/youtube_dl/extractor/rutube.py b/youtube_dl/extractor/rutube.py new file mode 100644 index 000000000..a18034fe2 --- /dev/null +++ b/youtube_dl/extractor/rutube.py @@ -0,0 +1,58 @@ +# encoding: utf-8 +import re +import json + +from .common import InfoExtractor +from ..utils import ( + compat_urlparse, + compat_str, + ExtractorError, +) + + +class RutubeIE(InfoExtractor): + _VALID_URL = r'https?://rutube.ru/video/(?P<long_id>\w+)' + + _TEST = { + u'url': u'http://rutube.ru/video/3eac3b4561676c17df9132a9a1e62e3e/', + u'file': u'3eac3b4561676c17df9132a9a1e62e3e.mp4', + u'info_dict': { + u'title': u'Раненный кенгуру забежал в аптеку', + u'uploader': u'NTDRussian', + u'uploader_id': u'29790', + }, + u'params': { + # It requires ffmpeg (m3u8 download) + u'skip_download': True, + }, + } + + def _get_api_response(self, short_id, subpath): + api_url = 'http://rutube.ru/api/play/%s/%s/?format=json' % (subpath, short_id) + response_json = self._download_webpage(api_url, short_id, + u'Downloading %s json' % subpath) + return json.loads(response_json) + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + long_id = mobj.group('long_id') + webpage = self._download_webpage(url, long_id) + og_video = self._og_search_video_url(webpage) + short_id = compat_urlparse.urlparse(og_video).path[1:] + options = self._get_api_response(short_id, 'options') + trackinfo = self._get_api_response(short_id, 'trackinfo') + # Some videos don't have the author field + author = trackinfo.get('author') or {} + m3u8_url = trackinfo['video_balancer'].get('m3u8') + if m3u8_url is None: + raise ExtractorError(u'Couldn\'t find m3u8 manifest url') + + return { + 'id': trackinfo['id'], + 'title': trackinfo['title'], + 'url': m3u8_url, + 'ext': 'mp4', + 'thumbnail': options['thumbnail_url'], + 'uploader': author.get('name'), + 'uploader_id': compat_str(author['id']) if author else None, + } diff --git a/youtube_dl/extractor/spankwire.py b/youtube_dl/extractor/spankwire.py new file mode 100644 index 000000000..32df0a7fb --- /dev/null +++ b/youtube_dl/extractor/spankwire.py @@ -0,0 +1,74 @@ +import os +import re + +from .common import InfoExtractor +from ..utils import ( + compat_urllib_parse_urlparse, + compat_urllib_request, + compat_urllib_parse, + unescapeHTML, +) +from ..aes import ( + aes_decrypt_text +) + +class SpankwireIE(InfoExtractor): + _VALID_URL = r'^(?:https?://)?(?:www\.)?(?P<url>spankwire\.com/[^/]*/video(?P<videoid>[0-9]+)/?)' + _TEST = { + u'url': u'http://www.spankwire.com/Buckcherry-s-X-Rated-Music-Video-Crazy-Bitch/video103545/', + u'file': u'103545.mp4', + u'md5': u'1b3f55e345500552dbc252a3e9c1af43', + u'info_dict': { + u"uploader": u"oreusz", + u"title": u"Buckcherry`s X Rated Music Video Crazy Bitch", + u"description": u"Crazy Bitch X rated music video.", + 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>([^<]+)', webpage, u'title') + video_uploader = self._html_search_regex(r'by:\s*<a [^>]*>(.+?)</a>', webpage, u'uploader', fatal=False) + thumbnail = self._html_search_regex(r'flashvars\.image_url = "([^"]+)', webpage, u'thumbnail', fatal=False) + description = self._html_search_regex(r'>\s*Description:</div>\s*<[^>]*>([^<]+)', webpage, u'description', fatal=False) + if len(description) == 0: + description = None + + video_urls = list(map(compat_urllib_parse.unquote , re.findall(r'flashvars\.quality_[0-9]{3}p = "([^"]+)', webpage))) + if webpage.find('flashvars\.encrypted = "true"') != -1: + password = self._html_search_regex(r'flashvars\.video_title = "([^"]+)', webpage, u'password').replace('+', ' ') + video_urls = list(map(lambda s: aes_decrypt_text(s, password, 32).decode('utf-8'), video_urls)) + + formats = [] + for video_url in video_urls: + path = compat_urllib_parse_urlparse( video_url ).path + extension = os.path.splitext( path )[1][1:] + format = path.split('/')[4].split('_')[:2] + format = "-".join( format ) + formats.append({ + 'url': video_url, + 'ext': extension, + 'format': format, + 'format_id': format, + }) + formats.sort(key=lambda format: list(map(lambda s: s.zfill(6), format['format'].split('-')))) + + age_limit = self._rta_search(webpage) + + return { + 'id': video_id, + 'uploader': video_uploader, + 'title': video_title, + 'thumbnail': thumbnail, + 'description': description, + 'formats': formats, + 'age_limit': age_limit, + } diff --git a/youtube_dl/extractor/sztvhu.py b/youtube_dl/extractor/sztvhu.py new file mode 100644 index 000000000..81fa35c4b --- /dev/null +++ b/youtube_dl/extractor/sztvhu.py @@ -0,0 +1,44 @@ +# -*- coding: utf-8 -*- + +import re + +from .common import InfoExtractor +from ..utils import determine_ext + + +class SztvHuIE(InfoExtractor): + _VALID_URL = r'(?:http://)?(?:(?:www\.)?sztv\.hu|www\.tvszombathely\.hu)/(?:[^/]+)/.+-(?P<id>[0-9]+)' + _TEST = { + u'url': u'http://sztv.hu/hirek/cserkeszek-nepszerusitettek-a-kornyezettudatos-eletmodot-a-savaria-teren-20130909', + u'file': u'20130909.mp4', + u'md5': u'a6df607b11fb07d0e9f2ad94613375cb', + u'info_dict': { + u"title": u"Cserkészek népszerűsítették a környezettudatos életmódot a Savaria téren", + u"description": u'A zöld nap játékos ismeretterjesztő programjait a Magyar Cserkész Szövetség szervezte, akik az ország nyolc városában adják át tudásukat az érdeklődőknek. A PET...', + } + } + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + video_id = mobj.group('id') + webpage = self._download_webpage(url, video_id) + video_file = self._search_regex( + r'file: "...:(.*?)",', webpage, 'video file') + title = self._html_search_regex( + r'<meta name="title" content="([^"]*?) - [^-]*? - [^-]*?"', + webpage, 'video title') + description = self._html_search_regex( + r'<meta name="description" content="([^"]*)"/>', + webpage, 'video description', fatal=False) + thumbnail = self._og_search_thumbnail(webpage) + + video_url = 'http://media.sztv.hu/vod/' + video_file + + return { + 'id': video_id, + 'url': video_url, + 'title': title, + 'ext': determine_ext(video_url), + 'description': description, + 'thumbnail': thumbnail, + } diff --git a/youtube_dl/extractor/techtalks.py b/youtube_dl/extractor/techtalks.py new file mode 100644 index 000000000..a55f236cb --- /dev/null +++ b/youtube_dl/extractor/techtalks.py @@ -0,0 +1,65 @@ +import re + +from .common import InfoExtractor +from ..utils import ( + get_element_by_attribute, + clean_html, +) + + +class TechTalksIE(InfoExtractor): + _VALID_URL = r'https?://techtalks\.tv/talks/[^/]*/(?P<id>\d+)/' + + _TEST = { + u'url': u'http://techtalks.tv/talks/learning-topic-models-going-beyond-svd/57758/', + u'playlist': [ + { + u'file': u'57758.flv', + u'info_dict': { + u'title': u'Learning Topic Models --- Going beyond SVD', + }, + }, + { + u'file': u'57758-slides.flv', + u'info_dict': { + u'title': u'Learning Topic Models --- Going beyond SVD', + }, + }, + ], + u'params': { + # rtmp download + u'skip_download': True, + }, + } + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + talk_id = mobj.group('id') + webpage = self._download_webpage(url, talk_id) + rtmp_url = self._search_regex(r'netConnectionUrl: \'(.*?)\'', webpage, + u'rtmp url') + play_path = self._search_regex(r'href=\'(.*?)\' [^>]*id="flowplayer_presenter"', + webpage, u'presenter play path') + title = clean_html(get_element_by_attribute('class', 'title', webpage)) + video_info = { + 'id': talk_id, + 'title': title, + 'url': rtmp_url, + 'play_path': play_path, + 'ext': 'flv', + } + m_slides = re.search(r'<a class="slides" href=\'(.*?)\'', webpage) + if m_slides is None: + return video_info + else: + return [ + video_info, + # The slides video + { + 'id': talk_id + '-slides', + 'title': title, + 'url': rtmp_url, + 'play_path': m_slides.group(1), + 'ext': 'flv', + }, + ] diff --git a/youtube_dl/extractor/tube8.py b/youtube_dl/extractor/tube8.py new file mode 100644 index 000000000..aea9d9a24 --- /dev/null +++ b/youtube_dl/extractor/tube8.py @@ -0,0 +1,65 @@ +import os +import re + +from .common import InfoExtractor +from ..utils import ( + compat_urllib_parse_urlparse, + compat_urllib_request, + compat_urllib_parse, + unescapeHTML, +) +from ..aes import ( + aes_decrypt_text +) + +class Tube8IE(InfoExtractor): + _VALID_URL = r'^(?:https?://)?(?:www\.)?(?P<url>tube8\.com/[^/]+/[^/]+/(?P<videoid>[0-9]+)/?)' + _TEST = { + u'url': u'http://www.tube8.com/teen/kasia-music-video/229795/', + u'file': u'229795.mp4', + u'md5': u'e9e0b0c86734e5e3766e653509475db0', + u'info_dict': { + u"description": u"hot teen Kasia grinding", + u"uploader": u"unknown", + u"title": u"Kasia music video", + 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'videotitle ="([^"]+)', webpage, u'title') + video_description = self._html_search_regex(r'>Description:</strong>(.+?)<', webpage, u'description', fatal=False) + video_uploader = self._html_search_regex(r'>Submitted by:</strong>(?:\s|<[^>]*>)*(.+?)<', webpage, u'uploader', fatal=False) + thumbnail = self._html_search_regex(r'"image_url":"([^"]+)', webpage, u'thumbnail', fatal=False) + if thumbnail: + thumbnail = thumbnail.replace('\\/', '/') + + video_url = self._html_search_regex(r'"video_url":"([^"]+)', webpage, u'video_url') + if webpage.find('"encrypted":true')!=-1: + password = self._html_search_regex(r'"video_title":"([^"]+)', webpage, u'password') + video_url = aes_decrypt_text(video_url, password, 32).decode('utf-8') + path = compat_urllib_parse_urlparse( video_url ).path + extension = os.path.splitext( path )[1][1:] + format = path.split('/')[4].split('_')[:2] + format = "-".join( format ) + + return { + 'id': video_id, + 'uploader': video_uploader, + 'title': video_title, + 'thumbnail': thumbnail, + 'description': video_description, + 'url': video_url, + 'ext': extension, + 'format': format, + 'format_id': format, + 'age_limit': 18, + } diff --git a/youtube_dl/extractor/tudou.py b/youtube_dl/extractor/tudou.py index 1405b73f7..7a3891b89 100644 --- a/youtube_dl/extractor/tudou.py +++ b/youtube_dl/extractor/tudou.py @@ -7,15 +7,25 @@ from .common import InfoExtractor class TudouIE(InfoExtractor): - _VALID_URL = r'(?:http://)?(?:www\.)?tudou\.com/(?:listplay|programs)/(?:view|(.+?))/(?:([^/]+)|([^/]+))(?:\.html)?' - _TEST = { + _VALID_URL = r'(?:http://)?(?:www\.)?tudou\.com/(?:listplay|programs|albumplay)/(?:view|(.+?))/(?:([^/]+)|([^/]+))(?:\.html)?' + _TESTS = [{ u'url': u'http://www.tudou.com/listplay/zzdE77v6Mmo/2xN2duXMxmw.html', u'file': u'159448201.f4v', u'md5': u'140a49ed444bd22f93330985d8475fcb', u'info_dict': { u"title": u"卡马乔国足开大脚长传冲吊集锦" } - } + }, + { + u'url': u'http://www.tudou.com/albumplay/TenTw_JgiPM/PzsAs5usU9A.html', + u'file': u'todo.mp4', + u'md5': u'todo.mp4', + u'info_dict': { + u'title': u'todo.mp4', + }, + u'add_ie': [u'Youku'], + u'skip': u'Only works from China' + }] def _url_for_id(self, id, quality = None): info_url = "http://v2.tudou.com/f?id="+str(id) @@ -29,14 +39,19 @@ class TudouIE(InfoExtractor): mobj = re.match(self._VALID_URL, url) video_id = mobj.group(2) webpage = self._download_webpage(url, video_id) - title = re.search(",kw:\"(.+)\"",webpage) - if title is None: - title = re.search(",kw: \'(.+)\'",webpage) - title = title.group(1) - thumbnail_url = re.search(",pic: \'(.+?)\'",webpage) - if thumbnail_url is None: - thumbnail_url = re.search(",pic:\"(.+?)\"",webpage) - thumbnail_url = thumbnail_url.group(1) + + m = re.search(r'vcode:\s*[\'"](.+?)[\'"]', webpage) + if m and m.group(1): + return { + '_type': 'url', + 'url': u'youku:' + m.group(1), + 'ie_key': 'Youku' + } + + title = self._search_regex( + r",kw:\s*['\"](.+?)[\"']", webpage, u'title') + thumbnail_url = self._search_regex( + r",pic:\s*[\"'](.+?)[\"']", webpage, u'thumbnail URL', fatal=False) segs_json = self._search_regex(r'segs: \'(.*)\'', webpage, 'segments') segments = json.loads(segs_json) 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/videodetective.py b/youtube_dl/extractor/videodetective.py index d89f84094..265dd5b91 100644 --- a/youtube_dl/extractor/videodetective.py +++ b/youtube_dl/extractor/videodetective.py @@ -16,7 +16,7 @@ class VideoDetectiveIE(InfoExtractor): u'info_dict': { u'title': u'KICK-ASS 2', u'description': u'md5:65ba37ad619165afac7d432eaded6013', - u'duration': 138, + u'duration': 135, }, } diff --git a/youtube_dl/extractor/vimeo.py b/youtube_dl/extractor/vimeo.py index cea29f035..c7d864a2b 100644 --- a/youtube_dl/extractor/vimeo.py +++ b/youtube_dl/extractor/vimeo.py @@ -1,3 +1,4 @@ +# encoding: utf-8 import json import re import itertools @@ -10,19 +11,21 @@ from ..utils import ( clean_html, get_element_by_attribute, ExtractorError, + RegexNotFoundError, std_headers, + unsmuggle_url, ) 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]+)/?(?:[?].*)?$' + _VALID_URL = r'(?P<proto>https?://)?(?:(?:www|player)\.)?vimeo(?P<pro>pro)?\.com/(?:(?:(?:groups|album)/[^/]+)|(?:.*?)/)?(?P<direct_link>play_redirect_hls\?clip_id=)?(?:videos?/)?(?P<id>[0-9]+)/?(?:[?].*)?(?:#.*)?$' _NETRC_MACHINE = 'vimeo' IE_NAME = u'vimeo' _TESTS = [ { - u'url': u'http://vimeo.com/56015672', + u'url': u'http://vimeo.com/56015672#at=0', u'file': u'56015672.mp4', u'md5': u'8879b6cc097e987f02484baf890129e5', u'info_dict': { @@ -54,6 +57,21 @@ class VimeoIE(InfoExtractor): u'uploader': u'The BLN & Business of Software', }, }, + { + u'url': u'http://vimeo.com/68375962', + u'file': u'68375962.mp4', + u'md5': u'aaf896bdb7ddd6476df50007a0ac0ae7', + u'note': u'Video protected with password', + u'info_dict': { + u'title': u'youtube-dl password protected test video', + u'upload_date': u'20130614', + u'uploader_id': u'user18948128', + u'uploader': u'Jaime Marquínez Ferrándiz', + }, + u'params': { + u'videopassword': u'youtube-dl', + }, + }, ] def _login(self): @@ -98,6 +116,12 @@ class VimeoIE(InfoExtractor): self._login() def _real_extract(self, url, new_video=True): + url, data = unsmuggle_url(url) + headers = std_headers + if data is not None: + headers = headers.copy() + headers.update(data) + # Extract ID from URL mobj = re.match(self._VALID_URL, url) if mobj is None: @@ -112,7 +136,7 @@ class VimeoIE(InfoExtractor): url = 'https://vimeo.com/' + video_id # Retrieve video webpage to extract further information - request = compat_urllib_request.Request(url, None, std_headers) + request = compat_urllib_request.Request(url, None, headers) webpage = self._download_webpage(request, video_id) # Now we begin extracting as much information as we can from what we @@ -122,18 +146,26 @@ class VimeoIE(InfoExtractor): # Extract the config JSON try: - config = self._search_regex([r' = {config:({.+?}),assets:', r'c=({.+?);'], - webpage, u'info section', flags=re.DOTALL) - config = json.loads(config) - except: + try: + config_url = self._html_search_regex( + r' data-config-url="(.+?)"', webpage, u'config URL') + config_json = self._download_webpage(config_url, video_id) + config = json.loads(config_json) + except RegexNotFoundError: + # For pro videos or player.vimeo.com urls + config = self._search_regex([r' = {config:({.+?}),assets:', r'c=({.+?);'], + webpage, u'info section', flags=re.DOTALL) + config = json.loads(config) + except Exception as e: if re.search('The creator of this video has not given you permission to embed it on this domain.', webpage): raise ExtractorError(u'The author has restricted the access to this video, try with the "--referer" option') - if re.search('If so please provide the correct password.', webpage): + if re.search('<form[^>]+?id="pw_form"', webpage) is not None: self._verify_video_password(url, video_id, webpage) return self._real_extract(url) else: - raise ExtractorError(u'Unable to extract info section') + raise ExtractorError(u'Unable to extract info section', + cause=e) # Extract title video_title = config["video"]["title"] @@ -172,46 +204,45 @@ class VimeoIE(InfoExtractor): # Vimeo specific: extract video codec and quality information # First consider quality, then codecs, then take everything - # TODO bind to format param - codecs = [('h264', 'mp4'), ('vp8', 'flv'), ('vp6', 'flv')] + codecs = [('vp6', 'flv'), ('vp8', 'flv'), ('h264', 'mp4')] files = { 'hd': [], 'sd': [], 'other': []} config_files = config["video"].get("files") or config["request"].get("files") for codec_name, codec_extension in codecs: - if codec_name in config_files: - if 'hd' in config_files[codec_name]: - files['hd'].append((codec_name, codec_extension, 'hd')) - elif 'sd' in config_files[codec_name]: - files['sd'].append((codec_name, codec_extension, 'sd')) + for quality in config_files.get(codec_name, []): + format_id = '-'.join((codec_name, quality)).lower() + key = quality if quality in files else 'other' + video_url = None + if isinstance(config_files[codec_name], dict): + file_info = config_files[codec_name][quality] + video_url = file_info.get('url') else: - files['other'].append((codec_name, codec_extension, config_files[codec_name][0])) - - for quality in ('hd', 'sd', 'other'): - if len(files[quality]) > 0: - video_quality = files[quality][0][2] - video_codec = files[quality][0][0] - video_extension = files[quality][0][1] - self.to_screen(u'%s: Downloading %s file at %s quality' % (video_id, video_codec.upper(), video_quality)) - break - else: - raise ExtractorError(u'No known codec found') + file_info = {} + if video_url is None: + video_url = "http://player.vimeo.com/play_redirect?clip_id=%s&sig=%s&time=%s&quality=%s&codecs=%s&type=moogaloop_local&embed_location=" \ + %(video_id, sig, timestamp, quality, codec_name.upper()) - video_url = None - if isinstance(config_files[video_codec], dict): - video_url = config_files[video_codec][video_quality].get("url") - if video_url is None: - video_url = "http://player.vimeo.com/play_redirect?clip_id=%s&sig=%s&time=%s&quality=%s&codecs=%s&type=moogaloop_local&embed_location=" \ - %(video_id, sig, timestamp, video_quality, video_codec.upper()) + files[key].append({ + 'ext': codec_extension, + 'url': video_url, + 'format_id': format_id, + 'width': file_info.get('width'), + 'height': file_info.get('height'), + }) + formats = [] + for key in ('other', 'sd', 'hd'): + formats += files[key] + if len(formats) == 0: + raise ExtractorError(u'No known codec found') return [{ 'id': video_id, - 'url': video_url, 'uploader': video_uploader, 'uploader_id': video_uploader_id, 'upload_date': video_upload_date, 'title': video_title, - 'ext': video_extension, 'thumbnail': video_thumbnail, 'description': video_description, + 'formats': formats, }] 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/websurg.py b/youtube_dl/extractor/websurg.py new file mode 100644 index 000000000..43953bfdd --- /dev/null +++ b/youtube_dl/extractor/websurg.py @@ -0,0 +1,59 @@ +# coding: utf-8 + +import re + +from ..utils import ( + compat_urllib_request, + compat_urllib_parse +) + +from .common import InfoExtractor + +class WeBSurgIE(InfoExtractor): + IE_NAME = u'websurg.com' + _VALID_URL = r'http://.*?\.websurg\.com/MEDIA/\?noheader=1&doi=(.*)' + + _TEST = { + u'url': u'http://www.websurg.com/MEDIA/?noheader=1&doi=vd01en4012', + u'file': u'vd01en4012.mp4', + u'params': { + u'skip_download': True, + }, + u'skip': u'Requires login information', + } + + _LOGIN_URL = 'http://www.websurg.com/inc/login/login_div.ajax.php?login=1' + + def _real_initialize(self): + + login_form = { + 'username': self._downloader.params['username'], + 'password': self._downloader.params['password'], + 'Submit': 1 + } + + request = compat_urllib_request.Request( + self._LOGIN_URL, compat_urllib_parse.urlencode(login_form)) + request.add_header( + 'Content-Type', 'application/x-www-form-urlencoded;charset=utf-8') + compat_urllib_request.urlopen(request).info() + webpage = self._download_webpage(self._LOGIN_URL, '', 'Logging in') + + if webpage != 'OK': + self._downloader.report_error( + u'Unable to log in: bad username/password') + + def _real_extract(self, url): + video_id = re.match(self._VALID_URL, url).group(1) + + webpage = self._download_webpage(url, video_id) + + url_info = re.search(r'streamer="(.*?)" src="(.*?)"', webpage) + + return {'id': video_id, + 'title': self._og_search_title(webpage), + 'description': self._og_search_description(webpage), + 'ext' : 'mp4', + 'url' : url_info.group(1) + '/' + url_info.group(2), + 'thumbnail': self._og_search_thumbnail(webpage) + } diff --git a/youtube_dl/extractor/xhamster.py b/youtube_dl/extractor/xhamster.py index 361619694..7444d3393 100644 --- a/youtube_dl/extractor/xhamster.py +++ b/youtube_dl/extractor/xhamster.py @@ -19,7 +19,8 @@ class XHamsterIE(InfoExtractor): u'info_dict': { u"upload_date": u"20121014", u"uploader_id": u"Ruseful2011", - u"title": u"FemaleAgent Shy beauty takes the bait" + u"title": u"FemaleAgent Shy beauty takes the bait", + u"age_limit": 18, } }, { @@ -27,28 +28,33 @@ class XHamsterIE(InfoExtractor): u'file': u'2221348.flv', u'md5': u'e767b9475de189320f691f49c679c4c7', u'info_dict': { - u"upload_date": u"20130914", - u"uploader_id": u"jojo747400", - u"title": u"Britney Spears Sexy Booty" + u"upload_date": u"20130914", + u"uploader_id": u"jojo747400", + u"title": u"Britney Spears Sexy Booty", + u"age_limit": 18, } }] def _real_extract(self,url): + def extract_video_url(webpage): + mobj = re.search(r'\'srv\': \'(?P<server>[^\']*)\',\s*\'file\': \'(?P<file>[^\']+)\',', webpage) + if mobj is None: + raise ExtractorError(u'Unable to extract media URL') + if len(mobj.group('server')) == 0: + return compat_urllib_parse.unquote(mobj.group('file')) + else: + return mobj.group('server')+'/key='+mobj.group('file') + + def is_hd(webpage): + return webpage.find('<div class=\'icon iconHD\'>') != -1 + mobj = re.match(self._VALID_URL, url) video_id = mobj.group('id') seo = mobj.group('seo') - mrss_url = 'http://xhamster.com/movies/%s/%s.html?hd' % (video_id, seo) + mrss_url = 'http://xhamster.com/movies/%s/%s.html' % (video_id, seo) webpage = self._download_webpage(mrss_url, video_id) - mobj = re.search(r'\'srv\': \'(?P<server>[^\']*)\',\s*\'file\': \'(?P<file>[^\']+)\',', webpage) - if mobj is None: - raise ExtractorError(u'Unable to extract media URL') - if len(mobj.group('server')) == 0: - video_url = compat_urllib_parse.unquote(mobj.group('file')) - else: - video_url = mobj.group('server')+'/key='+mobj.group('file') - video_title = self._html_search_regex(r'<title>(?P<title>.+?) - xHamster\.com</title>', webpage, u'title') @@ -72,13 +78,34 @@ class XHamsterIE(InfoExtractor): video_thumbnail = self._search_regex(r'\'image\':\'(?P<thumbnail>[^\']+)\'', webpage, u'thumbnail', fatal=False) - return [{ - 'id': video_id, - 'url': video_url, - 'ext': determine_ext(video_url), - 'title': video_title, + age_limit = self._rta_search(webpage) + + video_url = extract_video_url(webpage) + hd = is_hd(webpage) + formats = [{ + 'url': video_url, + 'ext': determine_ext(video_url), + 'format': 'hd' if hd else 'sd', + 'format_id': 'hd' if hd else 'sd', + }] + if not hd: + webpage = self._download_webpage(mrss_url+'?hd', video_id) + if is_hd(webpage): + video_url = extract_video_url(webpage) + formats.append({ + 'url': video_url, + 'ext': determine_ext(video_url), + 'format': 'hd', + 'format_id': 'hd', + }) + + return { + 'id': video_id, + 'title': video_title, + 'formats': formats, 'description': video_description, 'upload_date': video_upload_date, 'uploader_id': video_uploader_id, - 'thumbnail': video_thumbnail - }] + 'thumbnail': video_thumbnail, + 'age_limit': age_limit, + } diff --git a/youtube_dl/extractor/xnxx.py b/youtube_dl/extractor/xnxx.py index 40d848900..8a0eb1afd 100644 --- a/youtube_dl/extractor/xnxx.py +++ b/youtube_dl/extractor/xnxx.py @@ -18,7 +18,8 @@ class XNXXIE(InfoExtractor): u'file': u'1135332.flv', u'md5': u'0831677e2b4761795f68d417e0b7b445', u'info_dict': { - u"title": u"lida \u00bb Naked Funny Actress (5)" + u"title": u"lida \u00bb Naked Funny Actress (5)", + u"age_limit": 18, } } @@ -50,4 +51,5 @@ class XNXXIE(InfoExtractor): 'ext': 'flv', 'thumbnail': video_thumbnail, 'description': None, + 'age_limit': 18, }] diff --git a/youtube_dl/extractor/xvideos.py b/youtube_dl/extractor/xvideos.py index c3b9736d7..90138d7e5 100644 --- a/youtube_dl/extractor/xvideos.py +++ b/youtube_dl/extractor/xvideos.py @@ -13,7 +13,8 @@ class XVideosIE(InfoExtractor): u'file': u'939581.flv', u'md5': u'1d0c835822f0a71a7bf011855db929d0', u'info_dict': { - u"title": u"Funny Porns By >>>>S<<<<<< -1" + u"title": u"Funny Porns By >>>>S<<<<<< -1", + u"age_limit": 18, } } @@ -46,6 +47,7 @@ class XVideosIE(InfoExtractor): 'ext': 'flv', 'thumbnail': video_thumbnail, 'description': None, + 'age_limit': 18, } return [info] diff --git a/youtube_dl/extractor/youjizz.py b/youtube_dl/extractor/youjizz.py index 1265639e8..1fcc518ac 100644 --- a/youtube_dl/extractor/youjizz.py +++ b/youtube_dl/extractor/youjizz.py @@ -13,7 +13,8 @@ class YouJizzIE(InfoExtractor): u'file': u'2189178.flv', u'md5': u'07e15fa469ba384c7693fd246905547c', u'info_dict': { - u"title": u"Zeichentrick 1" + u"title": u"Zeichentrick 1", + u"age_limit": 18, } } @@ -25,6 +26,8 @@ class YouJizzIE(InfoExtractor): # Get webpage content webpage = self._download_webpage(url, video_id) + age_limit = self._rta_search(webpage) + # Get the video title video_title = self._html_search_regex(r'<title>(?P<title>.*)</title>', webpage, u'title').strip() @@ -60,6 +63,7 @@ class YouJizzIE(InfoExtractor): 'title': video_title, 'ext': 'flv', 'format': 'flv', - 'player_url': embed_page_url} + 'player_url': embed_page_url, + 'age_limit': age_limit} return [info] diff --git a/youtube_dl/extractor/youporn.py b/youtube_dl/extractor/youporn.py index b1f93dd1b..e46a9b4d6 100644 --- a/youtube_dl/extractor/youporn.py +++ b/youtube_dl/extractor/youporn.py @@ -17,7 +17,7 @@ from ..aes import ( ) class YouPornIE(InfoExtractor): - _VALID_URL = r'^(?:https?://)?(?:\w+\.)?youporn\.com/watch/(?P<videoid>[0-9]+)/(?P<title>[^/]+)' + _VALID_URL = r'^(?:https?://)?(?:www\.)?(?P<url>youporn\.com/watch/(?P<videoid>[0-9]+)/(?P<title>[^/]+))' _TEST = { u'url': u'http://www.youporn.com/watch/505835/sex-ed-is-it-safe-to-masturbate-daily/', u'file': u'505835.mp4', @@ -26,27 +26,15 @@ class YouPornIE(InfoExtractor): u"upload_date": u"20101221", u"description": u"Love & Sex Answers: http://bit.ly/DanAndJenn -- Is It Unhealthy To Masturbate Daily?", u"uploader": u"Ask Dan And Jennifer", - u"title": u"Sex Ed: Is It Safe To Masturbate Daily?" + u"title": u"Sex Ed: Is It Safe To Masturbate Daily?", + u"age_limit": 18, } } - def _print_formats(self, formats): - """Print all available formats""" - print(u'Available formats:') - print(u'ext\t\tformat') - print(u'---------------------------------') - for format in formats: - print(u'%s\t\t%s' % (format['ext'], format['format'])) - - def _specific(self, req_format, formats): - for x in formats: - if x["format"] == req_format: - return x - return None - 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') @@ -70,27 +58,22 @@ class YouPornIE(InfoExtractor): except KeyError: raise ExtractorError('Missing JSON parameter: ' + sys.exc_info()[1]) - # Get all of the formats available + # Get all of the links from the page DOWNLOAD_LIST_RE = r'(?s)<ul class="downloadList">(?P<download_list>.*?)</ul>' download_list_html = self._search_regex(DOWNLOAD_LIST_RE, webpage, u'download list').strip() - - # Get all of the links from the page - LINK_RE = r'(?s)<a href="(?P<url>[^"]+)">' + LINK_RE = r'<a href="([^"]+)">' links = re.findall(LINK_RE, download_list_html) - - # Get link of hd video if available - mobj = re.search(r'var encryptedQuality720URL = \'(?P<encrypted_video_url>[a-zA-Z0-9+/]+={0,2})\';', webpage) - if mobj != None: - encrypted_video_url = mobj.group(u'encrypted_video_url') - video_url = aes_decrypt_text(encrypted_video_url, video_title, 32).decode('utf-8') - links = [video_url] + links + + # Get all encrypted links + encrypted_links = re.findall(r'var encryptedQuality[0-9]{3}URL = \'([a-zA-Z0-9+/]+={0,2})\';', webpage) + for encrypted_link in encrypted_links: + link = aes_decrypt_text(encrypted_link, video_title, 32).decode('utf-8') + links.append(link) if not links: raise ExtractorError(u'ERROR: no known formats available for video') - self.to_screen(u'Links found: %d' % len(links)) - formats = [] for link in links: @@ -102,39 +85,32 @@ class YouPornIE(InfoExtractor): path = compat_urllib_parse_urlparse( video_url ).path extension = os.path.splitext( path )[1][1:] format = path.split('/')[4].split('_')[:2] + # size = format[0] # bitrate = format[1] format = "-".join( format ) # title = u'%s-%s-%s' % (video_title, size, bitrate) formats.append({ - 'id': video_id, 'url': video_url, - 'uploader': video_uploader, - 'upload_date': upload_date, - 'title': video_title, 'ext': extension, 'format': format, - 'thumbnail': thumbnail, - 'description': video_description, - 'age_limit': age_limit, + 'format_id': format, }) - if self._downloader.params.get('listformats', None): - self._print_formats(formats) - return - - req_format = self._downloader.params.get('format', 'best') - self.to_screen(u'Format: %s' % req_format) - - if req_format is None or req_format == 'best': - return [formats[0]] - elif req_format == 'worst': - return [formats[-1]] - elif req_format in ('-1', 'all'): - return formats - else: - format = self._specific( req_format, formats ) - if format is None: - raise ExtractorError(u'Requested format not available') - return [format] + # Sort and remove doubles + formats.sort(key=lambda format: list(map(lambda s: s.zfill(6), format['format'].split('-')))) + for i in range(len(formats)-1,0,-1): + if formats[i]['format_id'] == formats[i-1]['format_id']: + del formats[i] + + return { + 'id': video_id, + 'uploader': video_uploader, + 'upload_date': upload_date, + 'title': video_title, + 'thumbnail': thumbnail, + 'description': video_description, + 'age_limit': age_limit, + 'formats': formats, + } diff --git a/youtube_dl/extractor/youtube.py b/youtube_dl/extractor/youtube.py index d7c9b38f9..9053f3ead 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'', @@ -236,11 +229,13 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor): '136': 'mp4', '137': 'mp4', '138': 'mp4', - '139': 'mp4', - '140': 'mp4', - '141': 'mp4', '160': 'mp4', + # Dash mp4 audio + '139': 'm4a', + '140': 'm4a', + '141': 'm4a', + # Dash webm '171': 'webm', '172': 'webm', @@ -346,7 +341,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor): }, { u"url": u"http://www.youtube.com/watch?v=1ltcDfZMA3U", - u"file": u"1ltcDfZMA3U.flv", + u"file": u"1ltcDfZMA3U.mp4", u"note": u"Test VEVO video (#897)", u"info_dict": { u"upload_date": u"20070518", @@ -1116,7 +1111,7 @@ 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 @@ -1150,7 +1145,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor): list_page = self._download_webpage(list_url, video_id) caption_list = xml.etree.ElementTree.fromstring(list_page.encode('utf-8')) original_lang_node = caption_list.find('track') - if original_lang_node.attrib.get('kind') != 'asr' : + if original_lang_node is None or original_lang_node.attrib.get('kind') != 'asr' : self._downloader.report_warning(u'Video doesn\'t have automatic captions') return {} original_lang = original_lang_node.attrib['lang_code'] @@ -1403,32 +1398,29 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor): # this signatures are encrypted if 'url_encoded_fmt_stream_map' not in args: raise ValueError(u'No stream_map present') # caught below - m_s = re.search(r'[&,]s=', args['url_encoded_fmt_stream_map']) + re_signature = re.compile(r'[&,]s=') + m_s = re_signature.search(args['url_encoded_fmt_stream_map']) if m_s is not None: self.to_screen(u'%s: Encrypted signatures detected.' % video_id) video_info['url_encoded_fmt_stream_map'] = [args['url_encoded_fmt_stream_map']] - m_s = re.search(r'[&,]s=', args.get('adaptive_fmts', u'')) + m_s = re_signature.search(args.get('adaptive_fmts', u'')) if m_s is not None: - if 'url_encoded_fmt_stream_map' in video_info: - video_info['url_encoded_fmt_stream_map'][0] += ',' + args['adaptive_fmts'] - else: - video_info['url_encoded_fmt_stream_map'] = [args['adaptive_fmts']] - elif 'adaptive_fmts' in video_info: - if 'url_encoded_fmt_stream_map' in video_info: - video_info['url_encoded_fmt_stream_map'][0] += ',' + video_info['adaptive_fmts'][0] + if 'adaptive_fmts' in video_info: + video_info['adaptive_fmts'][0] += ',' + args['adaptive_fmts'] else: - video_info['url_encoded_fmt_stream_map'] = video_info['adaptive_fmts'] + video_info['adaptive_fmts'] = [args['adaptive_fmts']] except ValueError: pass if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'): self.report_rtmp_download() video_url_list = [(None, video_info['conn'][0])] - elif 'url_encoded_fmt_stream_map' in video_info and len(video_info['url_encoded_fmt_stream_map']) >= 1: - if 'rtmpe%3Dyes' in video_info['url_encoded_fmt_stream_map'][0]: + elif len(video_info.get('url_encoded_fmt_stream_map', [])) >= 1 or len(video_info.get('adaptive_fmts', [])) >= 1: + encoded_url_map = video_info.get('url_encoded_fmt_stream_map', [''])[0] + ',' + video_info.get('adaptive_fmts',[''])[0] + if 'rtmpe%3Dyes' in encoded_url_map: raise ExtractorError('rtmpe downloads are not supported, see https://github.com/rg3/youtube-dl/issues/343 for more information.', expected=True) url_map = {} - for url_data_str in video_info['url_encoded_fmt_stream_map'][0].split(','): + for url_data_str in encoded_url_map.split(','): url_data = compat_parse_qs(url_data_str) if 'itag' in url_data and 'url' in url_data: url = url_data['url'][0] @@ -1481,13 +1473,13 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor): raise ExtractorError(u'no conn, hlsvp or url_encoded_fmt_stream_map information found in video info') results = [] - for format_param, video_real_url in video_url_list: + for itag, video_real_url in video_url_list: # Extension - video_extension = self._video_extensions.get(format_param, 'flv') + video_extension = self._video_extensions.get(itag, 'flv') - video_format = '{0} - {1}{2}'.format(format_param if format_param else video_extension, - self._video_dimensions.get(format_param, '???'), - ' ('+self._special_itags[format_param]+')' if format_param in self._special_itags else '') + video_format = '{0} - {1}{2}'.format(itag if itag else video_extension, + self._video_dimensions.get(itag, '???'), + ' ('+self._special_itags[itag]+')' if itag in self._special_itags else '') results.append({ 'id': video_id, @@ -1498,6 +1490,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor): 'title': video_title, 'ext': video_extension, 'format': video_format, + 'format_id': itag, 'thumbnail': video_thumbnail, 'description': video_description, 'player_url': player_url, diff --git a/youtube_dl/utils.py b/youtube_dl/utils.py index 3e81c308b..1d9785341 100644 --- a/youtube_dl/utils.py +++ b/youtube_dl/utils.py @@ -572,6 +572,11 @@ class ExtractorError(Exception): return u''.join(traceback.format_tb(self.traceback)) +class RegexNotFoundError(ExtractorError): + """Error when a regex didn't match""" + pass + + class DownloadError(Exception): """Download Error exception. @@ -945,3 +950,29 @@ class locked_file(object): def shell_quote(args): return ' '.join(map(pipes.quote, args)) + + +def takewhile_inclusive(pred, seq): + """ Like itertools.takewhile, but include the latest evaluated element + (the first element so that Not pred(e)) """ + for e in seq: + yield e + if not pred(e): + return + + +def smuggle_url(url, data): + """ Pass additional data in a URL for internal use. """ + + sdata = compat_urllib_parse.urlencode( + {u'__youtubedl_smuggle': json.dumps(data)}) + return url + u'#' + sdata + + +def unsmuggle_url(smug_url): + if not '#__youtubedl_smuggle' in smug_url: + return smug_url, None + url, _, sdata = smug_url.rpartition(u'#') + jsond = compat_parse_qs(sdata)[u'__youtubedl_smuggle'][0] + data = json.loads(jsond) + return url, data diff --git a/youtube_dl/version.py b/youtube_dl/version.py index 1004af116..75a46a2d5 100644 --- a/youtube_dl/version.py +++ b/youtube_dl/version.py @@ -1,2 +1,2 @@ -__version__ = '2013.10.09' +__version__ = '2013.11.02' |