youtube-dl/youtube_dl/__init__.py

359 lines
14 KiB
Python
Raw Normal View History

#!/usr/bin/env python
# -*- coding: utf-8 -*-
__license__ = 'Public Domain'
import codecs
2014-02-25 22:31:16 +00:00
import io
import os
import random
import sys
2014-09-13 05:58:44 +00:00
from .options import (
parseOpts,
)
2013-10-15 00:07:26 +00:00
from .utils import (
compat_expanduser,
2014-03-18 13:27:42 +00:00
compat_getpass,
2013-10-15 00:07:26 +00:00
compat_print,
DateRange,
DEFAULT_OUTTMPL,
2013-10-15 00:07:26 +00:00
decodeOption,
DownloadError,
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,
)
from .extractor import gen_extractors
from .YoutubeDL import YoutubeDL
from .postprocessor import (
2014-02-22 12:55:51 +00:00
AtomicParsleyPP,
FFmpegAudioFixPP,
2013-10-15 00:07:26 +00:00
FFmpegMetadataPP,
FFmpegVideoConvertor,
FFmpegExtractAudioPP,
FFmpegEmbedSubtitlePP,
XAttrMetadataPP,
ExecAfterDownloadPP,
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)
setproctitle(u'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:
parser.error(u'wrong header formatting, it should be key:value, not "%s"'%h)
key, value = h.split(':', 2)
if opts.verbose:
write_string(u'[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:
write_string(u'[debug] Batch file urls: ' + repr(batch_urls) + u'\n')
2012-11-28 01:04:46 +00:00
except IOError:
sys.exit(u'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
extractors = gen_extractors()
if opts.list_extractors:
for ie in sorted(extractors, key=lambda ie: ie.IE_NAME.lower()):
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:
2013-06-24 13:57:53 +00:00
compat_print(u' ' + mu)
2012-11-28 01:04:46 +00:00
sys.exit(0)
if opts.list_extractor_descriptions:
for ie in sorted(extractors, key=lambda ie: ie.IE_NAME.lower()):
if not ie._WORKING:
continue
desc = getattr(ie, 'IE_DESC', ie.IE_NAME)
if desc is False:
continue
if hasattr(ie, 'SEARCH_KEY'):
_SEARCHES = (u'cute kittens', u'slithering pythons', u'falling cat', u'angry poodle', u'purple fish', u'running tortoise', u'sleeping bunny')
_COUNTS = (u'', u'5', u'10', u'all')
desc += u' (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):
parser.error(u'using .netrc conflicts with giving username/password')
if opts.password is not None and opts.username is None:
2014-01-11 23:27:00 +00:00
parser.error(u'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):
parser.error(u'using output template conflicts with using title, video ID or auto number')
if opts.usetitle and opts.useid:
parser.error(u'using title conflicts with using video ID')
if opts.username is not None and opts.password is None:
2014-03-18 13:27:42 +00:00
opts.password = compat_getpass(u'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:
parser.error(u'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(u'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(u'invalid max_filesize specified')
opts.max_filesize = numeric_limit
2012-11-28 01:04:46 +00:00
if opts.retries is not None:
try:
opts.retries = int(opts.retries)
except (TypeError, ValueError):
2012-11-28 01:04:46 +00:00
parser.error(u'invalid retry count specified')
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')
opts.buffersize = numeric_buffersize
if opts.playliststart <= 0:
raise ValueError(u'Playlist start must be positive')
if opts.playlistend not in (-1, None) and opts.playlistend < opts.playliststart:
raise ValueError(u'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']:
2012-11-28 01:04:46 +00:00
parser.error(u'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')
2013-01-12 14:07:59 +00:00
if opts.recodevideo is not None:
2014-05-10 13:09:56 +00:00
if opts.recodevideo not in ['mp4', 'flv', 'webm', 'ogg', 'mkv']:
2013-01-12 14:07:59 +00:00
parser.error(u'invalid video recode 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.
if opts.allsubtitles and (opts.writeautomaticsub == False):
opts.writesubtitles = True
if sys.version_info < (3,):
# In Python 2, sys.argv is a bytestring (also note http://bugs.python.org/issue2128 for Windows systems)
2012-12-15 22:54:44 +00:00
if opts.outtmpl is not None:
opts.outtmpl = opts.outtmpl.decode(preferredencoding())
outtmpl =((opts.outtmpl is not None and opts.outtmpl)
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 DEFAULT_OUTTMPL)
if not os.path.splitext(outtmpl)[1] and opts.extractaudio:
parser.error(u'Cannot download a video and extract audio into the same'
u' file! Use "{0}.%(ext)s" instead of "{0}" as the output'
u' template'.format(outtmpl))
2013-09-24 19:55:25 +00:00
2014-10-24 22:30:57 +00:00
any_printing = 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
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
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,
2013-12-16 03:15:10 +00:00
'quiet': (opts.quiet 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,
2013-11-19 13:59:22 +00:00
'forcejson': opts.dumpjson,
2014-10-24 22:30:57 +00:00
'dump_single_json': opts.dump_single_json,
'simulate': opts.simulate or any_printing,
'skip_download': opts.skip_download,
2012-11-28 01:04:46 +00:00
'format': opts.format,
'format_limit': opts.format_limit,
'listformats': opts.listformats,
'outtmpl': outtmpl,
'autonumber_size': opts.autonumber_size,
2012-11-28 01:04:46 +00:00
'restrictfilenames': opts.restrictfilenames,
'ignoreerrors': opts.ignoreerrors,
'ratelimit': opts.ratelimit,
'nooverwrites': opts.nooverwrites,
'retries': opts.retries,
'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,
'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,
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,
2014-08-25 08:18:01 +00:00
'exec_cmd': opts.exec_cmd,
'extract_flat': opts.extract_flat,
}
2012-11-28 01:04:46 +00:00
with YoutubeDL(ydl_opts) as ydl:
ydl.print_debug_header()
ydl.add_default_info_extractors()
# PostProcessors
# Add the metadata pp first, the other pps will copy it
if opts.addmetadata:
ydl.add_post_processor(FFmpegMetadataPP())
if opts.extractaudio:
ydl.add_post_processor(FFmpegExtractAudioPP(preferredcodec=opts.audioformat, preferredquality=opts.audioquality, nopostoverwrites=opts.nopostoverwrites))
if opts.recodevideo:
ydl.add_post_processor(FFmpegVideoConvertor(preferedformat=opts.recodevideo))
if opts.embedsubtitles:
ydl.add_post_processor(FFmpegEmbedSubtitlePP(subtitlesformat=opts.subtitlesformat))
if opts.xattrs:
ydl.add_post_processor(XAttrMetadataPP())
2014-02-22 12:55:51 +00:00
if opts.embedthumbnail:
2014-02-22 17:31:54 +00:00
if not opts.addmetadata:
ydl.add_post_processor(FFmpegAudioFixPP())
2014-02-22 12:55:51 +00:00
ydl.add_post_processor(AtomicParsleyPP())
2014-08-22 21:40:43 +00:00
# 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.
2014-08-25 08:18:01 +00:00
if opts.exec_cmd:
ydl.add_post_processor(ExecAfterDownloadPP(
verboseOutput=opts.verbose, exec_cmd=opts.exec_cmd))
2014-08-22 21:40:43 +00:00
# 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):
2014-07-24 10:16:16 +00:00
if not (opts.update_self or opts.rm_cachedir):
parser.error(u'you must provide at least one URL')
else:
sys.exit()
2012-11-28 01:04:46 +00:00
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:
ydl.to_screen(u'--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:
sys.exit(u'ERROR: fixed output name but more than one file to download')
except KeyboardInterrupt:
sys.exit(u'\nERROR: Interrupted by user')