X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/65cc51b0a1ddf217bc2026332532e0f8019c158a..9bb07a5ee663304e100edc65967d5fc4a521bcd0:/youtube_dl/__init__.py diff --git a/youtube_dl/__init__.py b/youtube_dl/__init__.py index d7e1a91..9a659fc 100644 --- a/youtube_dl/__init__.py +++ b/youtube_dl/__init__.py @@ -1,602 +1,483 @@ #!/usr/bin/env python -# -*- coding: utf-8 -*- - -from __future__ import with_statement -from __future__ import absolute_import - -__authors__ = ( - 'Ricardo Garcia Gonzalez', - 'Danny Colligan', - 'Benjamin Johnson', - 'Vasyl\' Vavrychuk', - 'Witold Baryluk', - 'Paweł Paprota', - 'Gergely Imreh', - 'Rogério Brito', - 'Philipp Hagemeister', - 'Sören Schulze', - 'Kevin Ngo', - 'Ori Avtalion', - 'shizeeg', - 'Filippo Valsorda', - 'Christian Albrecht', - ) +# coding: utf-8 + +from __future__ import unicode_literals __license__ = 'Public Domain' -import getpass -import optparse +import codecs +import io import os -import re -import shlex -import socket -import subprocess +import random import sys -import warnings - -from .utils import * -from .version import __version__ -from .FileDownloader import * -from .InfoExtractors import * -from .PostProcessor import * - -def updateSelf(downloader, filename): - """Update the program file with the latest version from the repository""" - - # TODO: at least, check https certificates - - from zipimport import zipimporter - - API_URL = "https://api.github.com/repos/rg3/youtube-dl/downloads" - BIN_URL = "https://github.com/downloads/rg3/youtube-dl/youtube-dl" - EXE_URL = "https://github.com/downloads/rg3/youtube-dl/youtube-dl.exe" - - if hasattr(sys, "frozen"): # PY2EXE - if not os.access(filename, os.W_OK): - sys.exit('ERROR: no write permissions on %s' % filename) - - downloader.to_screen(u'Updating to latest version...') - - urla = compat_urllib_request.urlopen(API_URL) - download = filter(lambda x: x["name"] == "youtube-dl.exe", json.loads(urla.read())) - if not download: - downloader.to_screen(u'ERROR: can\'t find the current version. Please try again later.') - return - newversion = download[0]["description"].strip() - if newversion == __version__: - downloader.to_screen(u'youtube-dl is up-to-date (' + __version__ + ')') - return - urla.close() - - exe = os.path.abspath(filename) - directory = os.path.dirname(exe) - if not os.access(directory, os.W_OK): - sys.exit('ERROR: no write permissions on %s' % directory) - - try: - urlh = compat_urllib_request.urlopen(EXE_URL) - newcontent = urlh.read() - urlh.close() - with open(exe + '.new', 'wb') as outf: - outf.write(newcontent) - except (IOError, OSError) as err: - sys.exit('ERROR: unable to download latest version') - - try: - bat = os.path.join(directory, 'youtube-dl-updater.bat') - b = open(bat, 'w') - b.write(""" -echo Updating youtube-dl... -ping 127.0.0.1 -n 5 -w 1000 > NUL -move /Y "%s.new" "%s" -del "%s" - \n""" %(exe, exe, bat)) - b.close() - - os.startfile(bat) - except (IOError, OSError) as err: - sys.exit('ERROR: unable to overwrite current version') - - elif isinstance(globals().get('__loader__'), zipimporter): # UNIX ZIP - if not os.access(filename, os.W_OK): - sys.exit('ERROR: no write permissions on %s' % filename) - - downloader.to_screen(u'Updating to latest version...') - - urla = compat_urllib_request.urlopen(API_URL) - download = [x for x in json.loads(urla.read().decode('utf8')) if x["name"] == "youtube-dl"] - if not download: - downloader.to_screen(u'ERROR: can\'t find the current version. Please try again later.') - return - newversion = download[0]["description"].strip() - if newversion == __version__: - downloader.to_screen(u'youtube-dl is up-to-date (' + __version__ + ')') - return - urla.close() - - try: - urlh = compat_urllib_request.urlopen(BIN_URL) - newcontent = urlh.read() - urlh.close() - except (IOError, OSError) as err: - sys.exit('ERROR: unable to download latest version') - - try: - with open(filename, 'wb') as outf: - outf.write(newcontent) - except (IOError, OSError) as err: - sys.exit('ERROR: unable to overwrite current version') - else: - downloader.to_screen(u'It looks like you installed youtube-dl with pip or setup.py. Please use that to update.') - return - - downloader.to_screen(u'Updated youtube-dl. Restart youtube-dl to use the new version.') - -def parseOpts(): - def _readOptions(filename_bytes): - try: - optionf = open(filename_bytes) - except IOError: - return [] # silently skip if file is not present - try: - res = [] - for l in optionf: - res += shlex.split(l, comments=True) - finally: - optionf.close() - return res - - def _format_option_string(option): - ''' ('-o', '--option') -> -o, --format METAVAR''' - - opts = [] - - if option._short_opts: - opts.append(option._short_opts[0]) - if option._long_opts: - opts.append(option._long_opts[0]) - if len(opts) > 1: - opts.insert(1, ', ') - if option.takes_value(): opts.append(' %s' % option.metavar) +from .options import ( + parseOpts, +) +from .compat import ( + compat_getpass, + compat_shlex_split, + workaround_optparse_bug9161, +) +from .utils import ( + DateRange, + decodeOption, + DEFAULT_OUTTMPL, + DownloadError, + expand_path, + match_filter_func, + MaxDownloadsReached, + preferredencoding, + read_batch_urls, + SameFileError, + setproctitle, + std_headers, + write_string, + render_table, +) +from .update import update_self +from .downloader import ( + FileDownloader, +) +from .extractor import gen_extractors, list_extractors +from .extractor.adobepass import MSO_INFO +from .YoutubeDL import YoutubeDL + + +def _real_main(argv=None): + # Compatibility fixes for Windows + if sys.platform == 'win32': + # https://github.com/ytdl-org/youtube-dl/issues/820 + codecs.register(lambda name: codecs.lookup('utf-8') if name == 'cp65001' else None) + + workaround_optparse_bug9161() + + setproctitle('youtube-dl') + + parser, opts, args = parseOpts(argv) - return "".join(opts) - - def _find_term_columns(): - columns = os.environ.get('COLUMNS', None) - if columns: - return int(columns) - - try: - sp = subprocess.Popen(['stty', 'size'], stdout=subprocess.PIPE, stderr=subprocess.PIPE) - out,err = sp.communicate() - return int(out.split()[1]) - except: - pass - return None - - max_width = 80 - max_help_position = 80 - - # No need to wrap help messages if we're on a wide console - columns = _find_term_columns() - if columns: max_width = columns - - fmt = optparse.IndentedHelpFormatter(width=max_width, max_help_position=max_help_position) - fmt.format_option_strings = _format_option_string - - kw = { - 'version' : __version__, - 'formatter' : fmt, - 'usage' : '%prog [options] url [url...]', - 'conflict_handler' : 'resolve', - } - - parser = optparse.OptionParser(**kw) - - # option groups - general = optparse.OptionGroup(parser, 'General Options') - selection = optparse.OptionGroup(parser, 'Video Selection') - authentication = optparse.OptionGroup(parser, 'Authentication Options') - video_format = optparse.OptionGroup(parser, 'Video Format Options') - postproc = optparse.OptionGroup(parser, 'Post-processing Options') - filesystem = optparse.OptionGroup(parser, 'Filesystem Options') - verbosity = optparse.OptionGroup(parser, 'Verbosity / Simulation Options') - - general.add_option('-h', '--help', - action='help', help='print this help text and exit') - general.add_option('-v', '--version', - action='version', help='print program version and exit') - general.add_option('-U', '--update', - action='store_true', dest='update_self', help='update this program to latest version') - general.add_option('-i', '--ignore-errors', - action='store_true', dest='ignoreerrors', help='continue on download errors', default=False) - general.add_option('-r', '--rate-limit', - dest='ratelimit', metavar='LIMIT', help='download rate limit (e.g. 50k or 44.6m)') - general.add_option('-R', '--retries', - dest='retries', metavar='RETRIES', help='number of retries (default is %default)', default=10) - general.add_option('--buffer-size', - dest='buffersize', metavar='SIZE', help='size of download buffer (e.g. 1024 or 16k) (default is %default)', default="1024") - general.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) - general.add_option('--dump-user-agent', - action='store_true', dest='dump_user_agent', - help='display the current browser identification', default=False) - general.add_option('--user-agent', - dest='user_agent', help='specify a custom user agent', metavar='UA') - general.add_option('--list-extractors', - action='store_true', dest='list_extractors', - help='List all supported extractors and the URLs they would handle', default=False) - general.add_option('--test', action='store_true', dest='test', default=False, help=optparse.SUPPRESS_HELP) - - selection.add_option('--playlist-start', - dest='playliststart', metavar='NUMBER', help='playlist video to start at (default is %default)', default=1) - selection.add_option('--playlist-end', - dest='playlistend', metavar='NUMBER', help='playlist video to end at (default is last)', default=-1) - selection.add_option('--match-title', dest='matchtitle', metavar='REGEX',help='download only matching titles (regex or caseless sub-string)') - selection.add_option('--reject-title', dest='rejecttitle', metavar='REGEX',help='skip download for matching titles (regex or caseless sub-string)') - selection.add_option('--max-downloads', metavar='NUMBER', dest='max_downloads', help='Abort after downloading NUMBER files', default=None) - - authentication.add_option('-u', '--username', - dest='username', metavar='USERNAME', help='account username') - authentication.add_option('-p', '--password', - dest='password', metavar='PASSWORD', help='account password') - authentication.add_option('-n', '--netrc', - action='store_true', dest='usenetrc', help='use .netrc authentication data', default=False) - - - video_format.add_option('-f', '--format', - action='store', dest='format', metavar='FORMAT', help='video format code') - video_format.add_option('--all-formats', - action='store_const', dest='format', help='download all available video formats', const='all') - video_format.add_option('--prefer-free-formats', - action='store_true', dest='prefer_free_formats', default=False, help='prefer free video formats unless a specific one is requested') - video_format.add_option('--max-quality', - action='store', dest='format_limit', metavar='FORMAT', help='highest quality format to download') - video_format.add_option('-F', '--list-formats', - action='store_true', dest='listformats', help='list all available formats (currently youtube only)') - video_format.add_option('--write-srt', - action='store_true', dest='writesubtitles', - help='write video closed captions to a .srt file (currently youtube only)', default=False) - video_format.add_option('--srt-lang', - action='store', dest='subtitleslang', metavar='LANG', - help='language of the closed captions to download (optional) use IETF language tags like \'en\'') - - - verbosity.add_option('-q', '--quiet', - action='store_true', dest='quiet', help='activates quiet mode', default=False) - verbosity.add_option('-s', '--simulate', - action='store_true', dest='simulate', help='do not download the video and do not write anything to disk', default=False) - verbosity.add_option('--skip-download', - action='store_true', dest='skip_download', help='do not download the video', default=False) - verbosity.add_option('-g', '--get-url', - action='store_true', dest='geturl', help='simulate, quiet but print URL', default=False) - verbosity.add_option('-e', '--get-title', - action='store_true', dest='gettitle', help='simulate, quiet but print title', default=False) - verbosity.add_option('--get-thumbnail', - action='store_true', dest='getthumbnail', - help='simulate, quiet but print thumbnail URL', default=False) - verbosity.add_option('--get-description', - action='store_true', dest='getdescription', - help='simulate, quiet but print video description', default=False) - verbosity.add_option('--get-filename', - action='store_true', dest='getfilename', - help='simulate, quiet but print output filename', default=False) - verbosity.add_option('--get-format', - action='store_true', dest='getformat', - help='simulate, quiet but print output format', default=False) - verbosity.add_option('--no-progress', - action='store_true', dest='noprogress', help='do not print progress bar', default=False) - verbosity.add_option('--console-title', - action='store_true', dest='consoletitle', - help='display progress in console titlebar', default=False) - verbosity.add_option('-v', '--verbose', - action='store_true', dest='verbose', help='print various debugging information', default=False) - - - filesystem.add_option('-t', '--title', - action='store_true', dest='usetitle', help='use title in file name', default=False) - filesystem.add_option('--id', - action='store_true', dest='useid', help='use video ID in file name', default=False) - filesystem.add_option('-l', '--literal', - action='store_true', dest='usetitle', help='[deprecated] alias of --title', default=False) - filesystem.add_option('-A', '--auto-number', - action='store_true', dest='autonumber', - help='number downloaded files starting from 00000', default=False) - filesystem.add_option('-o', '--output', - dest='outtmpl', metavar='TEMPLATE', help='output filename template. Use %(title)s to get the title, %(uploader)s for the uploader name, %(autonumber)s to get an automatically incremented number, %(ext)s for the filename extension, %(upload_date)s for the upload date (YYYYMMDD), %(extractor)s for the provider (youtube, metacafe, etc), %(id)s for the video id and %% for a literal percent. Use - to output to stdout. Can also be used to download to a different directory, for example with -o \'/my/downloads/%(uploader)s/%(title)s-%(id)s.%(ext)s\' .') - filesystem.add_option('--restrict-filenames', - action='store_true', dest='restrictfilenames', - help='Restrict filenames to only ASCII characters, and avoid "&" and spaces in filenames', default=False) - filesystem.add_option('-a', '--batch-file', - dest='batchfile', metavar='FILE', help='file containing URLs to download (\'-\' for stdin)') - filesystem.add_option('-w', '--no-overwrites', - action='store_true', dest='nooverwrites', help='do not overwrite files', default=False) - filesystem.add_option('-c', '--continue', - action='store_true', dest='continue_dl', help='resume partially downloaded files', default=True) - filesystem.add_option('--no-continue', - action='store_false', dest='continue_dl', - help='do not resume partially downloaded files (restart from beginning)') - filesystem.add_option('--cookies', - dest='cookiefile', metavar='FILE', help='file to read cookies from and dump cookie jar in') - filesystem.add_option('--no-part', - action='store_true', dest='nopart', help='do not use .part files', default=False) - filesystem.add_option('--no-mtime', - action='store_false', dest='updatetime', - help='do not use the Last-modified header to set the file modification time', default=True) - filesystem.add_option('--write-description', - action='store_true', dest='writedescription', - help='write video description to a .description file', default=False) - filesystem.add_option('--write-info-json', - action='store_true', dest='writeinfojson', - help='write video metadata to a .info.json file', default=False) - - - postproc.add_option('-x', '--extract-audio', action='store_true', dest='extractaudio', default=False, - help='convert video files to audio-only files (requires ffmpeg or avconv and ffprobe or avprobe)') - postproc.add_option('--audio-format', metavar='FORMAT', dest='audioformat', default='best', - help='"best", "aac", "vorbis", "mp3", "m4a", or "wav"; best by default') - postproc.add_option('--audio-quality', metavar='QUALITY', dest='audioquality', default='5', - help='ffmpeg/avconv audio quality specification, insert a value between 0 (better) and 9 (worse) for VBR or a specific bitrate like 128K (default 5)') - postproc.add_option('-k', '--keep-video', action='store_true', dest='keepvideo', default=False, - help='keeps the video file on disk after the post-processing; the video is erased by default') - - - parser.add_option_group(general) - parser.add_option_group(selection) - parser.add_option_group(filesystem) - parser.add_option_group(verbosity) - parser.add_option_group(video_format) - parser.add_option_group(authentication) - parser.add_option_group(postproc) - - xdg_config_home = os.environ.get('XDG_CONFIG_HOME') - if xdg_config_home: - userConf = os.path.join(xdg_config_home, 'youtube-dl.conf') - else: - userConf = os.path.join(os.path.expanduser('~'), '.config', 'youtube-dl.conf') - argv = _readOptions('/etc/youtube-dl.conf') + _readOptions(userConf) + sys.argv[1:] - opts, args = parser.parse_args(argv) - - return parser, opts, args - -def gen_extractors(): - """ Return a list of an instance of every supported extractor. - The order does matter; the first extractor matched is the one handling the URL. - """ - return [ - YoutubePlaylistIE(), - YoutubeChannelIE(), - YoutubeUserIE(), - YoutubeSearchIE(), - YoutubeIE(), - MetacafeIE(), - DailymotionIE(), - GoogleIE(), - GoogleSearchIE(), - PhotobucketIE(), - YahooIE(), - YahooSearchIE(), - DepositFilesIE(), - FacebookIE(), - BlipTVUserIE(), - BlipTVIE(), - VimeoIE(), - MyVideoIE(), - ComedyCentralIE(), - EscapistIE(), - CollegeHumorIE(), - XVideosIE(), - SoundcloudIE(), - InfoQIE(), - MixcloudIE(), - StanfordOpenClassroomIE(), - MTVIE(), - YoukuIE(), - XNXXIE(), - GooglePlusIE(), - ArteTvIE(), - GenericIE() - ] - -def _real_main(): - parser, opts, args = parseOpts() - - # Open appropriate CookieJar - if opts.cookiefile is None: - jar = compat_cookiejar.CookieJar() - else: - try: - jar = compat_cookiejar.MozillaCookieJar(opts.cookiefile) - if os.path.isfile(opts.cookiefile) and os.access(opts.cookiefile, os.R_OK): - jar.load() - except (IOError, OSError) as err: - sys.exit(u'ERROR: unable to open cookie file') # Set user agent if opts.user_agent is not None: std_headers['User-Agent'] = opts.user_agent + # Set referer + if opts.referer is not None: + std_headers['Referer'] = opts.referer + + # Custom HTTP headers + if opts.headers is not None: + for h in opts.headers: + if ':' not in h: + parser.error('wrong header formatting, it should be key:value, not "%s"' % h) + key, value = h.split(':', 1) + if opts.verbose: + write_string('[debug] Adding header from command line option %s:%s\n' % (key, value)) + std_headers[key] = value + # Dump user agent if opts.dump_user_agent: - print(std_headers['User-Agent']) + write_string(std_headers['User-Agent'] + '\n', out=sys.stdout) sys.exit(0) # Batch file verification - batchurls = [] + batch_urls = [] if opts.batchfile is not None: try: if opts.batchfile == '-': batchfd = sys.stdin else: - batchfd = open(opts.batchfile, 'r') - batchurls = batchfd.readlines() - batchurls = [x.strip() for x in batchurls] - batchurls = [x for x in batchurls if len(x) > 0 and not re.search(r'^[#/;]', x)] + batchfd = io.open( + expand_path(opts.batchfile), + 'r', encoding='utf-8', errors='ignore') + batch_urls = read_batch_urls(batchfd) + if opts.verbose: + write_string('[debug] Batch file urls: ' + repr(batch_urls) + '\n') except IOError: - sys.exit(u'ERROR: batch file could not be read') - all_urls = batchurls + args - all_urls = [url.strip() for url in all_urls] - - # General configuration - cookie_processor = compat_urllib_request.HTTPCookieProcessor(jar) - proxy_handler = compat_urllib_request.ProxyHandler() - opener = compat_urllib_request.build_opener(proxy_handler, cookie_processor, YoutubeDLHandler()) - compat_urllib_request.install_opener(opener) - socket.setdefaulttimeout(300) # 5 minutes should be enough (famous last words) - - extractors = gen_extractors() + sys.exit('ERROR: batch file %s could not be read' % opts.batchfile) + all_urls = batch_urls + [url.strip() for url in args] # batch_urls are already striped in read_batch_urls + _enc = preferredencoding() + all_urls = [url.decode(_enc, 'ignore') if isinstance(url, bytes) else url for url in all_urls] if opts.list_extractors: - for ie in extractors: - print(ie.IE_NAME + (' (CURRENTLY BROKEN)' if not ie._WORKING else '')) - matchedUrls = filter(lambda url: ie.suitable(url), all_urls) - all_urls = filter(lambda url: url not in matchedUrls, all_urls) + for ie in list_extractors(opts.age_limit): + write_string(ie.IE_NAME + (' (CURRENTLY BROKEN)' if not ie._WORKING else '') + '\n', out=sys.stdout) + matchedUrls = [url for url in all_urls if ie.suitable(url)] for mu in matchedUrls: - print(u' ' + mu) + write_string(' ' + mu + '\n', out=sys.stdout) + sys.exit(0) + if opts.list_extractor_descriptions: + for ie in list_extractors(opts.age_limit): + if not ie._WORKING: + continue + desc = getattr(ie, 'IE_DESC', ie.IE_NAME) + if desc is False: + continue + if hasattr(ie, 'SEARCH_KEY'): + _SEARCHES = ('cute kittens', 'slithering pythons', 'falling cat', 'angry poodle', 'purple fish', 'running tortoise', 'sleeping bunny', 'burping cow') + _COUNTS = ('', '5', '10', 'all') + desc += ' (Example: "%s%s:%s" )' % (ie.SEARCH_KEY, random.choice(_COUNTS), random.choice(_SEARCHES)) + write_string(desc + '\n', out=sys.stdout) + sys.exit(0) + if opts.ap_list_mso: + table = [[mso_id, mso_info['name']] for mso_id, mso_info in MSO_INFO.items()] + write_string('Supported TV Providers:\n' + render_table(['mso', 'mso name'], table) + '\n', out=sys.stdout) sys.exit(0) # Conflicting, missing and erroneous options if opts.usenetrc and (opts.username is not None or opts.password is not None): - parser.error(u'using .netrc conflicts with giving username/password') + parser.error('using .netrc conflicts with giving username/password') if opts.password is not None and opts.username is None: - parser.error(u'account username missing') + parser.error('account username missing\n') + if opts.ap_password is not None and opts.ap_username is None: + parser.error('TV Provider account username missing\n') if opts.outtmpl is not None and (opts.usetitle or opts.autonumber or opts.useid): - parser.error(u'using output template conflicts with using title, video ID or auto number') + parser.error('using output template conflicts with using title, video ID or auto number') + if opts.autonumber_size is not None: + if opts.autonumber_size <= 0: + parser.error('auto number size must be positive') + if opts.autonumber_start is not None: + if opts.autonumber_start < 0: + parser.error('auto number start must be positive or 0') if opts.usetitle and opts.useid: - parser.error(u'using title conflicts with using video ID') + parser.error('using title conflicts with using video ID') if opts.username is not None and opts.password is None: - opts.password = getpass.getpass(u'Type account password and press return:') + opts.password = compat_getpass('Type account password and press [Return]: ') + if opts.ap_username is not None and opts.ap_password is None: + opts.ap_password = compat_getpass('Type TV provider account password and press [Return]: ') if opts.ratelimit is not None: numeric_limit = FileDownloader.parse_bytes(opts.ratelimit) if numeric_limit is None: - parser.error(u'invalid rate limit specified') + parser.error('invalid rate limit specified') opts.ratelimit = numeric_limit + if opts.min_filesize is not None: + numeric_limit = FileDownloader.parse_bytes(opts.min_filesize) + if numeric_limit is None: + parser.error('invalid min_filesize specified') + opts.min_filesize = numeric_limit + if opts.max_filesize is not None: + numeric_limit = FileDownloader.parse_bytes(opts.max_filesize) + if numeric_limit is None: + parser.error('invalid max_filesize specified') + opts.max_filesize = numeric_limit + if opts.sleep_interval is not None: + if opts.sleep_interval < 0: + parser.error('sleep interval must be positive or 0') + if opts.max_sleep_interval is not None: + if opts.max_sleep_interval < 0: + parser.error('max sleep interval must be positive or 0') + if opts.sleep_interval is None: + parser.error('min sleep interval must be specified, use --min-sleep-interval') + if opts.max_sleep_interval < opts.sleep_interval: + parser.error('max sleep interval must be greater than or equal to min sleep interval') + else: + opts.max_sleep_interval = opts.sleep_interval + if opts.ap_mso and opts.ap_mso not in MSO_INFO: + parser.error('Unsupported TV Provider, use --ap-list-mso to get a list of supported TV Providers') + + def parse_retries(retries): + if retries in ('inf', 'infinite'): + parsed_retries = float('inf') + else: + try: + parsed_retries = int(retries) + except (TypeError, ValueError): + parser.error('invalid retry count specified') + return parsed_retries if opts.retries is not None: - try: - opts.retries = int(opts.retries) - except (TypeError, ValueError) as err: - parser.error(u'invalid retry count specified') + opts.retries = parse_retries(opts.retries) + if opts.fragment_retries is not None: + opts.fragment_retries = parse_retries(opts.fragment_retries) if opts.buffersize is not None: numeric_buffersize = FileDownloader.parse_bytes(opts.buffersize) if numeric_buffersize is None: - parser.error(u'invalid buffer size specified') + parser.error('invalid buffer size specified') opts.buffersize = numeric_buffersize - try: - opts.playliststart = int(opts.playliststart) - if opts.playliststart <= 0: - raise ValueError(u'Playlist start must be positive') - except (TypeError, ValueError) as err: - parser.error(u'invalid playlist start number specified') - try: - opts.playlistend = int(opts.playlistend) - if opts.playlistend != -1 and (opts.playlistend <= 0 or opts.playlistend < opts.playliststart): - raise ValueError(u'Playlist end must be greater than playlist start') - except (TypeError, ValueError) as err: - parser.error(u'invalid playlist end number specified') + if opts.http_chunk_size is not None: + numeric_chunksize = FileDownloader.parse_bytes(opts.http_chunk_size) + if not numeric_chunksize: + parser.error('invalid http chunk size specified') + opts.http_chunk_size = numeric_chunksize + if opts.playliststart <= 0: + raise ValueError('Playlist start must be positive') + if opts.playlistend not in (-1, None) and opts.playlistend < opts.playliststart: + raise ValueError('Playlist end must be greater than playlist start') if opts.extractaudio: - if opts.audioformat not in ['best', 'aac', 'mp3', 'vorbis', 'm4a', 'wav']: - parser.error(u'invalid audio format specified') + if opts.audioformat not in ['best', 'aac', 'flac', 'mp3', 'm4a', 'opus', 'vorbis', 'wav']: + parser.error('invalid audio format specified') if opts.audioquality: opts.audioquality = opts.audioquality.strip('k').strip('K') if not opts.audioquality.isdigit(): - parser.error(u'invalid audio quality specified') + parser.error('invalid audio quality specified') + if opts.recodevideo is not None: + if opts.recodevideo not in ['mp4', 'flv', 'webm', 'ogg', 'mkv', 'avi']: + parser.error('invalid video recode format specified') + if opts.convertsubtitles is not None: + if opts.convertsubtitles not in ['srt', 'vtt', 'ass', 'lrc']: + parser.error('invalid subtitle format specified') + + if opts.date is not None: + date = DateRange.day(opts.date) + else: + date = DateRange(opts.dateafter, opts.datebefore) + + # Do not download videos when there are audio-only formats + if opts.extractaudio and not opts.keepvideo and opts.format is None: + opts.format = 'bestaudio/best' + + # --all-sub automatically sets --write-sub if --write-auto-sub is not given + # this was the old behaviour if only --all-sub was given. + if opts.allsubtitles and not opts.writeautomaticsub: + opts.writesubtitles = True + + outtmpl = ((opts.outtmpl is not None and opts.outtmpl) + or (opts.format == '-1' and opts.usetitle and '%(title)s-%(id)s-%(format)s.%(ext)s') + or (opts.format == '-1' and '%(id)s-%(format)s.%(ext)s') + or (opts.usetitle and opts.autonumber and '%(autonumber)s-%(title)s-%(id)s.%(ext)s') + or (opts.usetitle and '%(title)s-%(id)s.%(ext)s') + or (opts.useid and '%(id)s.%(ext)s') + or (opts.autonumber and '%(autonumber)s-%(id)s.%(ext)s') + or DEFAULT_OUTTMPL) + if not os.path.splitext(outtmpl)[1] and opts.extractaudio: + parser.error('Cannot download a video and extract audio into the same' + ' file! Use "{0}.%(ext)s" instead of "{0}" as the output' + ' template'.format(outtmpl)) + + any_getting = opts.geturl or opts.gettitle or opts.getid or opts.getthumbnail or opts.getdescription or opts.getfilename or opts.getformat or opts.getduration or opts.dumpjson or opts.dump_single_json + any_printing = opts.print_json + download_archive_fn = expand_path(opts.download_archive) if opts.download_archive is not None else opts.download_archive - # File downloader - fd = FileDownloader({ + # PostProcessors + postprocessors = [] + if opts.metafromtitle: + postprocessors.append({ + 'key': 'MetadataFromTitle', + 'titleformat': opts.metafromtitle + }) + if opts.extractaudio: + postprocessors.append({ + 'key': 'FFmpegExtractAudio', + 'preferredcodec': opts.audioformat, + 'preferredquality': opts.audioquality, + 'nopostoverwrites': opts.nopostoverwrites, + }) + if opts.recodevideo: + postprocessors.append({ + 'key': 'FFmpegVideoConvertor', + 'preferedformat': opts.recodevideo, + }) + # FFmpegMetadataPP should be run after FFmpegVideoConvertorPP and + # FFmpegExtractAudioPP as containers before conversion may not support + # metadata (3gp, webm, etc.) + # And this post-processor should be placed before other metadata + # manipulating post-processors (FFmpegEmbedSubtitle) to prevent loss of + # extra metadata. By default ffmpeg preserves metadata applicable for both + # source and target containers. From this point the container won't change, + # so metadata can be added here. + if opts.addmetadata: + postprocessors.append({'key': 'FFmpegMetadata'}) + if opts.convertsubtitles: + postprocessors.append({ + 'key': 'FFmpegSubtitlesConvertor', + 'format': opts.convertsubtitles, + }) + if opts.embedsubtitles: + postprocessors.append({ + 'key': 'FFmpegEmbedSubtitle', + }) + if opts.embedthumbnail: + already_have_thumbnail = opts.writethumbnail or opts.write_all_thumbnails + postprocessors.append({ + 'key': 'EmbedThumbnail', + 'already_have_thumbnail': already_have_thumbnail + }) + if not already_have_thumbnail: + opts.writethumbnail = True + # XAttrMetadataPP should be run after post-processors that may change file + # contents + if opts.xattrs: + postprocessors.append({'key': 'XAttrMetadata'}) + # Please keep ExecAfterDownload towards the bottom as it allows the user to modify the final file in any way. + # So if the user is able to remove the file before your postprocessor runs it might cause a few problems. + if opts.exec_cmd: + postprocessors.append({ + 'key': 'ExecAfterDownload', + 'exec_cmd': opts.exec_cmd, + }) + external_downloader_args = None + if opts.external_downloader_args: + external_downloader_args = compat_shlex_split(opts.external_downloader_args) + postprocessor_args = None + if opts.postprocessor_args: + postprocessor_args = compat_shlex_split(opts.postprocessor_args) + match_filter = ( + None if opts.match_filter is None + else match_filter_func(opts.match_filter)) + + ydl_opts = { 'usenetrc': opts.usenetrc, 'username': opts.username, 'password': opts.password, - 'quiet': (opts.quiet or opts.geturl or opts.gettitle or opts.getthumbnail or opts.getdescription or opts.getfilename or opts.getformat), + 'twofactor': opts.twofactor, + 'videopassword': opts.videopassword, + 'ap_mso': opts.ap_mso, + 'ap_username': opts.ap_username, + 'ap_password': opts.ap_password, + 'quiet': (opts.quiet or any_getting or any_printing), + 'no_warnings': opts.no_warnings, 'forceurl': opts.geturl, 'forcetitle': opts.gettitle, + 'forceid': opts.getid, 'forcethumbnail': opts.getthumbnail, 'forcedescription': opts.getdescription, + 'forceduration': opts.getduration, 'forcefilename': opts.getfilename, 'forceformat': opts.getformat, - 'simulate': opts.simulate, - 'skip_download': (opts.skip_download or opts.simulate or opts.geturl or opts.gettitle or opts.getthumbnail or opts.getdescription or opts.getfilename or opts.getformat), + 'forcejson': opts.dumpjson or opts.print_json, + 'dump_single_json': opts.dump_single_json, + 'simulate': opts.simulate or any_getting, + 'skip_download': opts.skip_download, 'format': opts.format, - 'format_limit': opts.format_limit, 'listformats': opts.listformats, - 'outtmpl': ((opts.outtmpl is not None and opts.outtmpl.decode(preferredencoding())) - or (opts.format == '-1' and opts.usetitle and u'%(title)s-%(id)s-%(format)s.%(ext)s') - or (opts.format == '-1' and u'%(id)s-%(format)s.%(ext)s') - or (opts.usetitle and opts.autonumber and u'%(autonumber)s-%(title)s-%(id)s.%(ext)s') - or (opts.usetitle and u'%(title)s-%(id)s.%(ext)s') - or (opts.useid and u'%(id)s.%(ext)s') - or (opts.autonumber and u'%(autonumber)s-%(id)s.%(ext)s') - or u'%(id)s.%(ext)s'), + 'outtmpl': outtmpl, + 'autonumber_size': opts.autonumber_size, + 'autonumber_start': opts.autonumber_start, 'restrictfilenames': opts.restrictfilenames, 'ignoreerrors': opts.ignoreerrors, + 'force_generic_extractor': opts.force_generic_extractor, 'ratelimit': opts.ratelimit, 'nooverwrites': opts.nooverwrites, 'retries': opts.retries, + 'fragment_retries': opts.fragment_retries, + 'skip_unavailable_fragments': opts.skip_unavailable_fragments, + 'keep_fragments': opts.keep_fragments, 'buffersize': opts.buffersize, 'noresizebuffer': opts.noresizebuffer, + 'http_chunk_size': opts.http_chunk_size, 'continuedl': opts.continue_dl, 'noprogress': opts.noprogress, + 'progress_with_newline': opts.progress_with_newline, 'playliststart': opts.playliststart, 'playlistend': opts.playlistend, + 'playlistreverse': opts.playlist_reverse, + 'playlistrandom': opts.playlist_random, + 'noplaylist': opts.noplaylist, 'logtostderr': opts.outtmpl == '-', 'consoletitle': opts.consoletitle, 'nopart': opts.nopart, 'updatetime': opts.updatetime, 'writedescription': opts.writedescription, + 'writeannotations': opts.writeannotations, 'writeinfojson': opts.writeinfojson, + 'writethumbnail': opts.writethumbnail, + 'write_all_thumbnails': opts.write_all_thumbnails, 'writesubtitles': opts.writesubtitles, - 'subtitleslang': opts.subtitleslang, - 'matchtitle': opts.matchtitle, - 'rejecttitle': opts.rejecttitle, + 'writeautomaticsub': opts.writeautomaticsub, + 'allsubtitles': opts.allsubtitles, + 'listsubtitles': opts.listsubtitles, + 'subtitlesformat': opts.subtitlesformat, + 'subtitleslangs': opts.subtitleslangs, + 'matchtitle': decodeOption(opts.matchtitle), + 'rejecttitle': decodeOption(opts.rejecttitle), 'max_downloads': opts.max_downloads, '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, + 'max_filesize': opts.max_filesize, + 'min_views': opts.min_views, + 'max_views': opts.max_views, + 'daterange': date, + 'cachedir': opts.cachedir, + 'youtube_print_sig_code': opts.youtube_print_sig_code, + 'age_limit': opts.age_limit, + 'download_archive': download_archive_fn, + 'cookiefile': opts.cookiefile, + 'nocheckcertificate': opts.no_check_certificate, + 'prefer_insecure': opts.prefer_insecure, + 'proxy': opts.proxy, + 'socket_timeout': opts.socket_timeout, + 'bidi_workaround': opts.bidi_workaround, + 'debug_printtraffic': opts.debug_printtraffic, + 'prefer_ffmpeg': opts.prefer_ffmpeg, + 'include_ads': opts.include_ads, + 'default_search': opts.default_search, + 'youtube_include_dash_manifest': opts.youtube_include_dash_manifest, + 'encoding': opts.encoding, + 'extract_flat': opts.extract_flat, + 'mark_watched': opts.mark_watched, + 'merge_output_format': opts.merge_output_format, + 'postprocessors': postprocessors, + 'fixup': opts.fixup, + 'source_address': opts.source_address, + 'call_home': opts.call_home, + 'sleep_interval': opts.sleep_interval, + 'max_sleep_interval': opts.max_sleep_interval, + 'external_downloader': opts.external_downloader, + 'list_thumbnails': opts.list_thumbnails, + 'playlist_items': opts.playlist_items, + 'xattr_set_filesize': opts.xattr_set_filesize, + 'match_filter': match_filter, + 'no_color': opts.no_color, + 'ffmpeg_location': opts.ffmpeg_location, + 'hls_prefer_native': opts.hls_prefer_native, + 'hls_use_mpegts': opts.hls_use_mpegts, + 'external_downloader_args': external_downloader_args, + 'postprocessor_args': postprocessor_args, + 'cn_verification_proxy': opts.cn_verification_proxy, + 'geo_verification_proxy': opts.geo_verification_proxy, + 'config_location': opts.config_location, + 'geo_bypass': opts.geo_bypass, + 'geo_bypass_country': opts.geo_bypass_country, + 'geo_bypass_ip_block': opts.geo_bypass_ip_block, + # just for deprecation check + 'autonumber': opts.autonumber if opts.autonumber is True else None, + 'usetitle': opts.usetitle if opts.usetitle is True else None, + } - if opts.verbose: - fd.to_screen(u'[debug] Proxy map: ' + str(proxy_handler.proxies)) + with YoutubeDL(ydl_opts) as ydl: + # Update version + if opts.update_self: + update_self(ydl.to_screen, opts.verbose, ydl._opener) - for extractor in extractors: - fd.add_info_extractor(extractor) + # Remove cache dir + if opts.rm_cachedir: + ydl.cache.remove() - # PostProcessors - if opts.extractaudio: - fd.add_post_processor(FFmpegExtractAudioPP(preferredcodec=opts.audioformat, preferredquality=opts.audioquality, keepvideo=opts.keepvideo)) + # Maybe do nothing + if (len(all_urls) < 1) and (opts.load_info_filename is None): + if opts.update_self or opts.rm_cachedir: + sys.exit() - # Update version - if opts.update_self: - updateSelf(fd, sys.argv[0]) + ydl.warn_if_short_id(sys.argv[1:] if argv is None else argv) + parser.error( + 'You must provide at least one URL.\n' + 'Type youtube-dl --help to see a list of all options.') - # Maybe do nothing - if len(all_urls) < 1: - if not opts.update_self: - parser.error(u'you must provide at least one URL') - else: - sys.exit() - - try: - retcode = fd.download(all_urls) - except MaxDownloadsReached: - fd.to_screen(u'--max-download limit reached, aborting.') - retcode = 101 - - # Dump cookie jar if requested - if opts.cookiefile is not None: try: - jar.save() - except (IOError, OSError) as err: - sys.exit(u'ERROR: unable to save cookie jar') + if opts.load_info_filename is not None: + retcode = ydl.download_with_info_file(expand_path(opts.load_info_filename)) + else: + retcode = ydl.download(all_urls) + except MaxDownloadsReached: + ydl.to_screen('--max-download limit reached, aborting.') + retcode = 101 sys.exit(retcode) -def main(): + +def main(argv=None): try: - _real_main() + _real_main(argv) except DownloadError: sys.exit(1) except SameFileError: - sys.exit(u'ERROR: fixed output name but more than one file to download') + sys.exit('ERROR: fixed output name but more than one file to download') except KeyboardInterrupt: - sys.exit(u'\nERROR: Interrupted by user') + sys.exit('\nERROR: Interrupted by user') + + +__all__ = ['main', 'YoutubeDL', 'gen_extractors', 'list_extractors']