youtube-dl/youtube_dl/__init__.py

416 lines
16 KiB
Python
Raw Normal View History

#!/usr/bin/env python
# -*- coding: utf-8 -*-
2014-11-16 14:08:34 +00:00
from __future__ import unicode_literals
__license__ = 'Public Domain'
import codecs
2014-02-25 22:31:16 +00:00
import io
import os
import random
import shlex
import sys
2014-09-13 05:58:44 +00:00
from .options import (
parseOpts,
)
from .compat import (
compat_expanduser,
2014-03-18 13:27:42 +00:00
compat_getpass,
2013-10-15 00:07:26 +00:00
compat_print,
workaround_optparse_bug9161,
)
from .utils import (
2013-10-15 00:07:26 +00:00
DateRange,
decodeOption,
DEFAULT_OUTTMPL,
2013-10-15 00:07:26 +00:00
DownloadError,
match_filter_func,
2013-10-15 00:07:26 +00:00
MaxDownloadsReached,
preferredencoding,
read_batch_urls,
2013-10-15 00:07:26 +00:00
SameFileError,
setproctitle,
2013-10-15 00:07:26 +00:00
std_headers,
write_string,
)
2012-12-30 18:49:14 +00:00
from .update import update_self
from .downloader import (
2013-10-15 00:07:26 +00:00
FileDownloader,
)
2015-01-07 06:41:05 +00:00
from .extractor import gen_extractors, list_extractors
from .YoutubeDL import YoutubeDL
2013-10-15 00:07:26 +00:00
def _real_main(argv=None):
# Compatibility fixes for Windows
if sys.platform == 'win32':
# https://github.com/rg3/youtube-dl/issues/820
codecs.register(lambda name: codecs.lookup('utf-8') if name == 'cp65001' else None)
workaround_optparse_bug9161()
2014-11-16 14:08:34 +00:00
setproctitle('youtube-dl')
parser, opts, args = parseOpts(argv)
2012-11-28 01:04:46 +00:00
# Set user agent
if opts.user_agent is not None:
std_headers['User-Agent'] = opts.user_agent
2013-04-24 11:56:04 +00:00
# Set referer
if opts.referer is not None:
std_headers['Referer'] = opts.referer
2012-11-28 01:04:46 +00:00
# Custom HTTP headers
if opts.headers is not None:
for h in opts.headers:
if h.find(':', 1) < 0:
2014-11-23 20:20:46 +00:00
parser.error('wrong header formatting, it should be key:value, not "%s"' % h)
key, value = h.split(':', 2)
if opts.verbose:
2014-11-23 20:20:46 +00:00
write_string('[debug] Adding header from command line option %s:%s\n' % (key, value))
std_headers[key] = value
2012-11-28 01:04:46 +00:00
# Dump user agent
if opts.dump_user_agent:
2013-06-24 13:57:53 +00:00
compat_print(std_headers['User-Agent'])
2012-11-28 01:04:46 +00:00
sys.exit(0)
# Batch file verification
batch_urls = []
2012-11-28 01:04:46 +00:00
if opts.batchfile is not None:
try:
if opts.batchfile == '-':
batchfd = sys.stdin
else:
batchfd = io.open(opts.batchfile, 'r', encoding='utf-8', errors='ignore')
batch_urls = read_batch_urls(batchfd)
if opts.verbose:
2014-11-16 14:08:34 +00:00
write_string('[debug] Batch file urls: ' + repr(batch_urls) + '\n')
2012-11-28 01:04:46 +00:00
except IOError:
2014-11-16 14:08:34 +00:00
sys.exit('ERROR: batch file could not be read')
all_urls = batch_urls + args
2012-11-28 01:04:46 +00:00
all_urls = [url.strip() for url in all_urls]
_enc = preferredencoding()
all_urls = [url.decode(_enc, 'ignore') if isinstance(url, bytes) else url for url in all_urls]
2012-11-28 01:04:46 +00:00
if opts.list_extractors:
for ie in list_extractors(opts.age_limit):
2013-06-24 13:57:53 +00:00
compat_print(ie.IE_NAME + (' (CURRENTLY BROKEN)' if not ie._WORKING else ''))
matchedUrls = [url for url in all_urls if ie.suitable(url)]
2012-11-28 01:04:46 +00:00
for mu in matchedUrls:
2014-11-16 14:08:34 +00:00
compat_print(' ' + mu)
2012-11-28 01:04:46 +00:00
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')
2014-11-16 14:08:34 +00:00
_COUNTS = ('', '5', '10', 'all')
desc += ' (Example: "%s%s:%s" )' % (ie.SEARCH_KEY, random.choice(_COUNTS), random.choice(_SEARCHES))
compat_print(desc)
sys.exit(0)
2012-11-28 01:04:46 +00:00
# Conflicting, missing and erroneous options
if opts.usenetrc and (opts.username is not None or opts.password is not None):
2014-11-16 14:08:34 +00:00
parser.error('using .netrc conflicts with giving username/password')
2012-11-28 01:04:46 +00:00
if opts.password is not None and opts.username is None:
2014-11-16 14:08:34 +00:00
parser.error('account username missing\n')
2012-11-28 01:04:46 +00:00
if opts.outtmpl is not None and (opts.usetitle or opts.autonumber or opts.useid):
2014-11-16 14:08:34 +00:00
parser.error('using output template conflicts with using title, video ID or auto number')
2012-11-28 01:04:46 +00:00
if opts.usetitle and opts.useid:
2014-11-16 14:08:34 +00:00
parser.error('using title conflicts with using video ID')
2012-11-28 01:04:46 +00:00
if opts.username is not None and opts.password is None:
2014-11-16 14:08:34 +00:00
opts.password = compat_getpass('Type account password and press [Return]: ')
2012-11-28 01:04:46 +00:00
if opts.ratelimit is not None:
numeric_limit = FileDownloader.parse_bytes(opts.ratelimit)
if numeric_limit is None:
2014-11-16 14:08:34 +00:00
parser.error('invalid rate limit specified')
2012-11-28 01:04:46 +00:00
opts.ratelimit = numeric_limit
if opts.min_filesize is not None:
numeric_limit = FileDownloader.parse_bytes(opts.min_filesize)
if numeric_limit is None:
2014-11-16 14:08:34 +00:00
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:
2014-11-16 14:08:34 +00:00
parser.error('invalid max_filesize specified')
opts.max_filesize = numeric_limit
2012-11-28 01:04:46 +00:00
if opts.retries is not None:
if opts.retries in ('inf', 'infinite'):
opts_retries = float('inf')
else:
try:
opts_retries = int(opts.retries)
except (TypeError, ValueError):
parser.error('invalid retry count specified')
2012-11-28 01:04:46 +00:00
if opts.buffersize is not None:
numeric_buffersize = FileDownloader.parse_bytes(opts.buffersize)
if numeric_buffersize is None:
2014-11-16 14:08:34 +00:00
parser.error('invalid buffer size specified')
2012-11-28 01:04:46 +00:00
opts.buffersize = numeric_buffersize
if opts.playliststart <= 0:
2014-11-16 14:08:34 +00:00
raise ValueError('Playlist start must be positive')
if opts.playlistend not in (-1, None) and opts.playlistend < opts.playliststart:
2014-11-16 14:08:34 +00:00
raise ValueError('Playlist end must be greater than playlist start')
2012-11-28 01:04:46 +00:00
if opts.extractaudio:
2013-01-10 18:15:04 +00:00
if opts.audioformat not in ['best', 'aac', 'mp3', 'm4a', 'opus', 'vorbis', 'wav']:
2014-11-16 14:08:34 +00:00
parser.error('invalid audio format specified')
2012-11-28 01:04:46 +00:00
if opts.audioquality:
opts.audioquality = opts.audioquality.strip('k').strip('K')
if not opts.audioquality.isdigit():
2014-11-16 14:08:34 +00:00
parser.error('invalid audio quality specified')
2013-01-12 14:07:59 +00:00
if opts.recodevideo is not None:
if opts.recodevideo not in ['mp4', 'flv', 'webm', 'ogg', 'mkv', 'avi']:
2014-11-16 14:08:34 +00:00
parser.error('invalid video recode format specified')
if opts.convertsubtitles is not None:
if opts.convertsubtitles not in ['srt', 'vtt', 'ass']:
parser.error('invalid subtitle format specified')
if opts.date is not None:
date = DateRange.day(opts.date)
else:
date = DateRange(opts.dateafter, opts.datebefore)
2012-11-28 01:04:46 +00:00
# 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.
2014-11-23 21:21:46 +00:00
if opts.allsubtitles and not opts.writeautomaticsub:
opts.writesubtitles = True
2015-02-21 13:55:13 +00:00
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:
2014-11-16 14:08:34 +00:00
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))
2013-09-24 19:55:25 +00:00
2015-01-08 17:01:41 +00:00
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 = compat_expanduser(opts.download_archive) if opts.download_archive is not None else opts.download_archive
2013-12-16 03:15:10 +00:00
# PostProcessors
postprocessors = []
# Add the metadata pp first, the other pps will copy it
if opts.metafromtitle:
postprocessors.append({
'key': 'MetadataFromTitle',
'titleformat': opts.metafromtitle
})
if opts.addmetadata:
postprocessors.append({'key': 'FFmpegMetadata'})
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,
})
if opts.convertsubtitles:
postprocessors.append({
'key': 'FFmpegSubtitlesConvertor',
'format': opts.convertsubtitles,
})
if opts.embedsubtitles:
postprocessors.append({
'key': 'FFmpegEmbedSubtitle',
})
if opts.xattrs:
postprocessors.append({'key': 'XAttrMetadata'})
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
# 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,
})
if opts.xattr_set_filesize:
try:
import xattr
xattr # Confuse flake8
except ImportError:
parser.error('setting filesize xattr requested but python-xattr is not available')
external_downloader_args = None
if opts.external_downloader_args:
external_downloader_args = shlex.split(opts.external_downloader_args)
match_filter = (
None if opts.match_filter is None
else match_filter_func(opts.match_filter))
ydl_opts = {
2012-11-28 01:04:46 +00:00
'usenetrc': opts.usenetrc,
'username': opts.username,
'password': opts.password,
'twofactor': opts.twofactor,
'videopassword': opts.videopassword,
2015-01-08 17:01:41 +00:00
'quiet': (opts.quiet or any_getting or any_printing),
2014-03-25 23:43:46 +00:00
'no_warnings': opts.no_warnings,
2012-11-28 01:04:46 +00:00
'forceurl': opts.geturl,
'forcetitle': opts.gettitle,
'forceid': opts.getid,
2012-11-28 01:04:46 +00:00
'forcethumbnail': opts.getthumbnail,
'forcedescription': opts.getdescription,
2013-12-16 03:15:10 +00:00
'forceduration': opts.getduration,
2012-11-28 01:04:46 +00:00
'forcefilename': opts.getfilename,
'forceformat': opts.getformat,
2015-01-08 17:01:41 +00:00
'forcejson': opts.dumpjson or opts.print_json,
2014-10-24 22:30:57 +00:00
'dump_single_json': opts.dump_single_json,
2015-01-08 17:01:41 +00:00
'simulate': opts.simulate or any_getting,
'skip_download': opts.skip_download,
2012-11-28 01:04:46 +00:00
'format': opts.format,
'listformats': opts.listformats,
'outtmpl': outtmpl,
'autonumber_size': opts.autonumber_size,
2012-11-28 01:04:46 +00:00
'restrictfilenames': opts.restrictfilenames,
'ignoreerrors': opts.ignoreerrors,
'force_generic_extractor': opts.force_generic_extractor,
2012-11-28 01:04:46 +00:00
'ratelimit': opts.ratelimit,
'nooverwrites': opts.nooverwrites,
'retries': opts_retries,
2012-11-28 01:04:46 +00:00
'buffersize': opts.buffersize,
'noresizebuffer': opts.noresizebuffer,
'continuedl': opts.continue_dl,
'noprogress': opts.noprogress,
'progress_with_newline': opts.progress_with_newline,
2012-11-28 01:04:46 +00:00
'playliststart': opts.playliststart,
'playlistend': opts.playlistend,
'playlistreverse': opts.playlist_reverse,
'noplaylist': opts.noplaylist,
2012-11-28 01:04:46 +00:00
'logtostderr': opts.outtmpl == '-',
'consoletitle': opts.consoletitle,
'nopart': opts.nopart,
'updatetime': opts.updatetime,
'writedescription': opts.writedescription,
'writeannotations': opts.writeannotations,
2012-11-28 01:04:46 +00:00
'writeinfojson': opts.writeinfojson,
'writethumbnail': opts.writethumbnail,
'write_all_thumbnails': opts.write_all_thumbnails,
2012-11-28 01:04:46 +00:00
'writesubtitles': opts.writesubtitles,
'writeautomaticsub': opts.writeautomaticsub,
'allsubtitles': opts.allsubtitles,
'listsubtitles': opts.listsubtitles,
'subtitlesformat': opts.subtitlesformat,
'subtitleslangs': opts.subtitleslangs,
'matchtitle': decodeOption(opts.matchtitle),
'rejecttitle': decodeOption(opts.rejecttitle),
2012-11-28 01:04:46 +00:00
'max_downloads': opts.max_downloads,
'prefer_free_formats': opts.prefer_free_formats,
'verbose': opts.verbose,
2013-04-11 16:31:35 +00:00
'dump_intermediate_pages': opts.dump_intermediate_pages,
2013-10-28 09:44:02 +00:00
'write_pages': opts.write_pages,
'test': opts.test,
2013-01-12 14:07:59 +00:00
'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,
2013-09-24 19:26:10 +00:00
'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,
2013-12-02 12:37:05 +00:00
'socket_timeout': opts.socket_timeout,
'bidi_workaround': opts.bidi_workaround,
2013-12-29 14:28:32 +00:00
'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,
'merge_output_format': opts.merge_output_format,
'postprocessors': postprocessors,
'postprocessor_args': shlex.split(opts.postprocessor_args or ''),
'fixup': opts.fixup,
'source_address': opts.source_address,
'call_home': opts.call_home,
'sleep_interval': opts.sleep_interval,
'external_downloader': opts.external_downloader,
2015-01-25 01:38:47 +00:00
'list_thumbnails': opts.list_thumbnails,
'playlist_items': opts.playlist_items,
'xattr_set_filesize': opts.xattr_set_filesize,
'match_filter': match_filter,
2015-02-10 03:22:10 +00:00
'no_color': opts.no_color,
2015-02-13 10:14:01 +00:00
'ffmpeg_location': opts.ffmpeg_location,
'hls_prefer_native': opts.hls_prefer_native,
'external_downloader_args': external_downloader_args,
'cn_verification_proxy': opts.cn_verification_proxy,
}
2012-11-28 01:04:46 +00:00
with YoutubeDL(ydl_opts) as ydl:
# Update version
if opts.update_self:
update_self(ydl.to_screen, opts.verbose)
2014-07-24 10:16:16 +00:00
# Remove cache dir
if opts.rm_cachedir:
ydl.cache.remove()
2014-07-24 10:16:16 +00:00
# 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()
2012-11-28 01:04:46 +00:00
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.')
try:
if opts.load_info_filename is not None:
retcode = ydl.download_with_info_file(opts.load_info_filename)
else:
retcode = ydl.download(all_urls)
except MaxDownloadsReached:
2014-11-16 14:08:34 +00:00
ydl.to_screen('--max-download limit reached, aborting.')
retcode = 101
2012-11-28 01:04:46 +00:00
sys.exit(retcode)
def main(argv=None):
2012-11-28 01:04:46 +00:00
try:
_real_main(argv)
2012-11-28 01:04:46 +00:00
except DownloadError:
sys.exit(1)
except SameFileError:
2014-11-16 14:08:34 +00:00
sys.exit('ERROR: fixed output name but more than one file to download')
2012-11-28 01:04:46 +00:00
except KeyboardInterrupt:
2014-11-16 14:08:34 +00:00
sys.exit('\nERROR: Interrupted by user')
2015-01-07 06:41:05 +00:00
__all__ = ['main', 'YoutubeDL', 'gen_extractors', 'list_extractors']