-#!/usr/bin/env python3
-# coding: utf-8
+import sys
-f'You are using an unsupported version of Python. Only Python versions 3.6 and above are supported by yt-dlp' # noqa: F541
+if sys.version_info < (3, 8):
+ raise ImportError(
+ f'You are using an unsupported version of Python. Only Python versions 3.8 and above are supported by yt-dlp') # noqa: F541
-__license__ = 'Public Domain'
+__license__ = 'The Unlicense'
-import codecs
-import io
+import collections
+import getpass
import itertools
+import optparse
import os
-import random
import re
-import sys
+import traceback
-from .options import (
- parseOpts,
-)
-from .compat import (
- compat_getpass,
- compat_os_name,
- compat_shlex_quote,
- workaround_optparse_bug9161,
-)
+from .compat import compat_os_name
from .cookies import SUPPORTED_BROWSERS, SUPPORTED_KEYRINGS
+from .downloader.external import get_external_downloader
+from .extractor import list_extractor_classes
+from .extractor.adobepass import MSO_INFO
+from .networking.impersonate import ImpersonateTarget
+from .options import parseOpts
+from .postprocessor import (
+ FFmpegExtractAudioPP,
+ FFmpegMergerPP,
+ FFmpegPostProcessor,
+ FFmpegSubtitlesConvertorPP,
+ FFmpegThumbnailsConvertorPP,
+ FFmpegVideoConvertorPP,
+ FFmpegVideoRemuxerPP,
+ MetadataFromFieldPP,
+ MetadataParserPP,
+)
+from .update import Updater
from .utils import (
+ NO_DEFAULT,
+ POSTPROCESS_WHEN,
DateRange,
- decodeOption,
DownloadCancelled,
DownloadError,
- error_to_compat_str,
- expand_path,
+ FormatSorter,
GeoUtils,
+ PlaylistEntries,
+ SameFileError,
+ decodeOption,
+ download_range_func,
+ expand_path,
float_or_none,
+ format_field,
int_or_none,
+ join_nonempty,
match_filter_func,
+ parse_bytes,
parse_duration,
preferredencoding,
read_batch_urls,
+ read_stdin,
render_table,
- SameFileError,
setproctitle,
- std_headers,
+ shell_quote,
+ traverse_obj,
+ variadic,
write_string,
)
-from .update import run_update
-from .downloader import (
- FileDownloader,
-)
-from .extractor import gen_extractors, list_extractors
-from .extractor.common import InfoExtractor
-from .extractor.adobepass import MSO_INFO
-from .postprocessor import (
- FFmpegExtractAudioPP,
- FFmpegSubtitlesConvertorPP,
- FFmpegThumbnailsConvertorPP,
- FFmpegVideoConvertorPP,
- FFmpegVideoRemuxerPP,
- MetadataFromFieldPP,
- MetadataParserPP,
-)
+from .utils.networking import std_headers
from .YoutubeDL import YoutubeDL
+_IN_CLI = False
-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('yt-dlp')
- parser, opts, args = parseOpts(argv)
- warnings, deprecation_warnings = [], []
+def _exit(status=0, *args):
+ for msg in args:
+ sys.stderr.write(msg)
+ raise SystemExit(status)
- # 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
- std_headers.update(opts.headers)
-
- # Dump user agent
- if opts.dump_user_agent:
- write_string(std_headers['User-Agent'] + '\n', out=sys.stdout)
- sys.exit(0)
-
- # Batch file verification
+def get_urls(urls, batchfile, verbose):
+ """
+ @param verbose -1: quiet, 0: normal, 1: verbose
+ """
batch_urls = []
- if opts.batchfile is not None:
+ if batchfile is not None:
try:
- if opts.batchfile == '-':
- write_string('Reading URLs from stdin - EOF (%s) to end:\n' % (
- 'Ctrl+Z' if compat_os_name == 'nt' else 'Ctrl+D'))
- batchfd = sys.stdin
- else:
- batchfd = io.open(
- expand_path(opts.batchfile),
- 'r', encoding='utf-8', errors='ignore')
- batch_urls = read_batch_urls(batchfd)
- if opts.verbose:
+ batch_urls = read_batch_urls(
+ read_stdin(None if verbose == -1 else 'URLs') if batchfile == '-'
+ else open(expand_path(batchfile), encoding='utf-8', errors='ignore'))
+ if verbose == 1:
write_string('[debug] Batch file urls: ' + repr(batch_urls) + '\n')
- except IOError:
- 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
+ except OSError:
+ _exit(f'ERROR: batch file {batchfile} could not be read')
_enc = preferredencoding()
- all_urls = [url.decode(_enc, 'ignore') if isinstance(url, bytes) else url for url in all_urls]
+ return [
+ url.strip().decode(_enc, 'ignore') if isinstance(url, bytes) else url.strip()
+ for url in batch_urls + urls]
- if opts.list_extractors:
- 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:
- 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 getattr(ie, 'SEARCH_KEY', None) is not None:
- _SEARCHES = ('cute kittens', 'slithering pythons', 'falling cat', 'angry poodle', 'purple fish', 'running tortoise', 'sleeping bunny', 'burping cow')
- _COUNTS = ('', '5', '10', 'all')
- desc += f'; "{ie.SEARCH_KEY}:" prefix (Example: "{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.format == 'best':
- warnings.append('.\n '.join((
- '"-f best" selects the best pre-merged format which is often not the best option',
- 'To let yt-dlp download and merge the best available formats, simply do not pass any format selection',
- 'If you know what you are doing and want only the best pre-merged format, use "-f b" instead to suppress this warning')))
- if opts.exec_cmd.get('before_dl') and opts.exec_before_dl_cmd:
- parser.error('using "--exec-before-download" conflicts with "--exec before_dl:"')
- if opts.usenetrc and (opts.username is not None or opts.password is not None):
- parser.error('using .netrc conflicts with giving username/password')
- if opts.password is not None and opts.username is None:
- 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.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.username is not None and opts.password is None:
- 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('invalid rate limit specified')
- opts.ratelimit = numeric_limit
- if opts.throttledratelimit is not None:
- numeric_limit = FileDownloader.parse_bytes(opts.throttledratelimit)
- if numeric_limit is None:
- parser.error('invalid rate limit specified')
- opts.throttledratelimit = 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.sleep_interval_subtitles is not None:
- if opts.sleep_interval_subtitles < 0:
- parser.error('subtitles sleep interval must be positive or 0')
- if opts.sleep_interval_requests is not None:
- if opts.sleep_interval_requests < 0:
- parser.error('requests sleep interval must be positive or 0')
- 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')
- if opts.overwrites: # --yes-overwrites implies --no-continue
- opts.continue_dl = False
- if opts.concurrent_fragment_downloads <= 0:
- parser.error('Concurrent fragments must be positive')
- if opts.wait_for_video is not None:
- min_wait, max_wait, *_ = map(parse_duration, opts.wait_for_video.split('-', 1) + [None])
- if min_wait is None or (max_wait is None and '-' in opts.wait_for_video):
- parser.error('Invalid time range to wait')
- elif max_wait is not None and max_wait < min_wait:
- parser.error('Minimum time range to wait must not be longer than the maximum')
- opts.wait_for_video = (min_wait, max_wait)
- def parse_retries(retries, name=''):
- if retries in ('inf', 'infinite'):
- parsed_retries = float('inf')
- else:
- try:
- parsed_retries = int(retries)
- except (TypeError, ValueError):
- parser.error('invalid %sretry count specified' % name)
- return parsed_retries
- if opts.retries is not None:
- opts.retries = parse_retries(opts.retries)
- if opts.file_access_retries is not None:
- opts.file_access_retries = parse_retries(opts.file_access_retries, 'file access ')
- if opts.fragment_retries is not None:
- opts.fragment_retries = parse_retries(opts.fragment_retries, 'fragment ')
- if opts.extractor_retries is not None:
- opts.extractor_retries = parse_retries(opts.extractor_retries, 'extractor ')
- if opts.buffersize is not None:
- numeric_buffersize = FileDownloader.parse_bytes(opts.buffersize)
- if numeric_buffersize is None:
- parser.error('invalid buffer size specified')
- opts.buffersize = numeric_buffersize
- 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 parser.error('Playlist start must be positive')
- if opts.playlistend not in (-1, None) and opts.playlistend < opts.playliststart:
- raise parser.error('Playlist end must be greater than playlist start')
- if opts.extractaudio:
- opts.audioformat = opts.audioformat.lower()
- if opts.audioformat not in ['best'] + list(FFmpegExtractAudioPP.SUPPORTED_EXTS):
- parser.error('invalid audio format specified')
- if opts.audioquality:
- opts.audioquality = opts.audioquality.strip('k').strip('K')
- audioquality = int_or_none(float_or_none(opts.audioquality)) # int_or_none prevents inf, nan
- if audioquality is None or audioquality < 0:
- parser.error('invalid audio quality specified')
- if opts.recodevideo is not None:
- opts.recodevideo = opts.recodevideo.replace(' ', '')
- if not re.match(FFmpegVideoConvertorPP.FORMAT_RE, opts.recodevideo):
- parser.error('invalid video remux format specified')
- if opts.remuxvideo is not None:
- opts.remuxvideo = opts.remuxvideo.replace(' ', '')
- if not re.match(FFmpegVideoRemuxerPP.FORMAT_RE, opts.remuxvideo):
- parser.error('invalid video remux format specified')
- if opts.convertsubtitles is not None:
- if opts.convertsubtitles not in FFmpegSubtitlesConvertorPP.SUPPORTED_EXTS:
- parser.error('invalid subtitle format specified')
- if opts.convertthumbnails is not None:
- if opts.convertthumbnails not in FFmpegThumbnailsConvertorPP.SUPPORTED_EXTS:
- parser.error('invalid thumbnail format specified')
- if opts.cookiesfrombrowser is not None:
- mobj = re.match(r'(?P<name>[^+:]+)(\s*\+\s*(?P<keyring>[^:]+))?(\s*:(?P<profile>.+))?', opts.cookiesfrombrowser)
- if mobj is None:
- parser.error(f'invalid cookies from browser arguments: {opts.cookiesfrombrowser}')
- browser_name, keyring, profile = mobj.group('name', 'keyring', 'profile')
- browser_name = browser_name.lower()
- if browser_name not in SUPPORTED_BROWSERS:
- parser.error(f'unsupported browser specified for cookies: "{browser_name}". '
- f'Supported browsers are: {", ".join(sorted(SUPPORTED_BROWSERS))}')
- if keyring is not None:
- keyring = keyring.upper()
- if keyring not in SUPPORTED_KEYRINGS:
- parser.error(f'unsupported keyring specified for cookies: "{keyring}". '
- f'Supported keyrings are: {", ".join(sorted(SUPPORTED_KEYRINGS))}')
- opts.cookiesfrombrowser = (browser_name, profile, keyring)
- geo_bypass_code = opts.geo_bypass_ip_block or opts.geo_bypass_country
- if geo_bypass_code is not None:
- try:
- GeoUtils.random_ipv4(geo_bypass_code)
- except Exception:
- parser.error('unsupported geo-bypass country or ip-block')
+def print_extractor_information(opts, urls):
+ out = ''
+ if opts.list_extractors:
+ # Importing GenericIE is currently slow since it imports YoutubeIE
+ from .extractor.generic import GenericIE
- if opts.date is not None:
- date = DateRange.day(opts.date)
+ urls = dict.fromkeys(urls, False)
+ for ie in list_extractor_classes(opts.age_limit):
+ out += ie.IE_NAME + (' (CURRENTLY BROKEN)' if not ie.working() else '') + '\n'
+ if ie == GenericIE:
+ matched_urls = [url for url, matched in urls.items() if not matched]
+ else:
+ matched_urls = tuple(filter(ie.suitable, urls.keys()))
+ urls.update(dict.fromkeys(matched_urls, True))
+ out += ''.join(f' {url}\n' for url in matched_urls)
+ elif opts.list_extractor_descriptions:
+ _SEARCHES = ('cute kittens', 'slithering pythons', 'falling cat', 'angry poodle', 'purple fish', 'running tortoise', 'sleeping bunny', 'burping cow')
+ out = '\n'.join(
+ ie.description(markdown=False, search_examples=_SEARCHES)
+ for ie in list_extractor_classes(opts.age_limit) if ie.working() and ie.IE_DESC is not False)
+ elif opts.ap_list_mso:
+ out = 'Supported TV Providers:\n{}\n'.format(render_table(
+ ['mso', 'mso name'],
+ [[mso_id, mso_info['name']] for mso_id, mso_info in MSO_INFO.items()]))
else:
- date = DateRange(opts.dateafter, opts.datebefore)
+ return False
+ write_string(out, out=sys.stdout)
+ return True
- compat_opts = opts.compat_opts
-
- def report_conflict(arg1, arg2):
- warnings.append(f'{arg2} is ignored since {arg1} was given')
+def set_compat_opts(opts):
def _unused_compat_opt(name):
- if name not in compat_opts:
+ if name not in opts.compat_opts:
return False
- compat_opts.discard(name)
- compat_opts.update(['*%s' % name])
+ opts.compat_opts.discard(name)
+ opts.compat_opts.update([f'*{name}'])
return True
def set_default_compat(compat_name, opt_name, default=True, remove_compat=True):
attr = getattr(opts, opt_name)
- if compat_name in compat_opts:
+ if compat_name in opts.compat_opts:
if attr is None:
setattr(opts, opt_name, not default)
return True
set_default_compat('abort-on-error', 'ignoreerrors', 'only_download')
set_default_compat('no-playlist-metafiles', 'allow_playlist_files')
set_default_compat('no-clean-infojson', 'clean_infojson')
- if 'no-attach-info-json' in compat_opts:
+ if 'no-attach-info-json' in opts.compat_opts:
if opts.embed_infojson:
_unused_compat_opt('no-attach-info-json')
else:
opts.embed_infojson = False
- if 'format-sort' in compat_opts:
- opts.format_sort.extend(InfoExtractor.FormatSort.ytdl_default)
+ if 'format-sort' in opts.compat_opts:
+ opts.format_sort.extend(FormatSorter.ytdl_default)
_video_multistreams_set = set_default_compat('multistreams', 'allow_multiple_video_streams', False, remove_compat=False)
_audio_multistreams_set = set_default_compat('multistreams', 'allow_multiple_audio_streams', False, remove_compat=False)
if _video_multistreams_set is False and _audio_multistreams_set is False:
_unused_compat_opt('multistreams')
- outtmpl_default = opts.outtmpl.get('default')
- if outtmpl_default == '':
- outtmpl_default, opts.skip_download = None, True
- del opts.outtmpl['default']
- if opts.useid:
- if outtmpl_default is None:
- outtmpl_default = opts.outtmpl['default'] = '%(id)s.%(ext)s'
- else:
- report_conflict('--output', '--id')
- if 'filename' in compat_opts:
- if outtmpl_default is None:
- outtmpl_default = opts.outtmpl['default'] = '%(title)s-%(id)s.%(ext)s'
+ if 'filename' in opts.compat_opts:
+ if opts.outtmpl.get('default') is None:
+ opts.outtmpl.update({'default': '%(title)s-%(id)s.%(ext)s'})
else:
_unused_compat_opt('filename')
+
+def validate_options(opts):
+ def validate(cndn, name, value=None, msg=None):
+ if cndn:
+ return True
+ raise ValueError((msg or 'invalid {name} "{value}" given').format(name=name, value=value))
+
+ def validate_in(name, value, items, msg=None):
+ return validate(value is None or value in items, name, value, msg)
+
+ def validate_regex(name, value, regex):
+ return validate(value is None or re.match(regex, value), name, value)
+
+ def validate_positive(name, value, strict=False):
+ return validate(value is None or value > 0 or (not strict and value == 0),
+ name, value, '{name} "{value}" must be positive' + ('' if strict else ' or 0'))
+
+ def validate_minmax(min_val, max_val, min_name, max_name=None):
+ if max_val is None or min_val is None or max_val >= min_val:
+ return
+ if not max_name:
+ min_name, max_name = f'min {min_name}', f'max {min_name}'
+ raise ValueError(f'{max_name} "{max_val}" must be must be greater than or equal to {min_name} "{min_val}"')
+
+ # Usernames and passwords
+ validate(sum(map(bool, (opts.usenetrc, opts.netrc_cmd, opts.username))) <= 1, '.netrc',
+ msg='{name}, netrc command and username/password are mutually exclusive options')
+ validate(opts.password is None or opts.username is not None, 'account username', msg='{name} missing')
+ validate(opts.ap_password is None or opts.ap_username is not None,
+ 'TV Provider account username', msg='{name} missing')
+ validate_in('TV Provider', opts.ap_mso, MSO_INFO,
+ 'Unsupported {name} "{value}", use --ap-list-mso to get a list of supported TV Providers')
+
+ # Numbers
+ validate_positive('autonumber start', opts.autonumber_start)
+ validate_positive('autonumber size', opts.autonumber_size, True)
+ validate_positive('concurrent fragments', opts.concurrent_fragment_downloads, True)
+ validate_positive('playlist start', opts.playliststart, True)
+ if opts.playlistend != -1:
+ validate_minmax(opts.playliststart, opts.playlistend, 'playlist start', 'playlist end')
+
+ # Time ranges
+ validate_positive('subtitles sleep interval', opts.sleep_interval_subtitles)
+ validate_positive('requests sleep interval', opts.sleep_interval_requests)
+ validate_positive('sleep interval', opts.sleep_interval)
+ validate_positive('max sleep interval', opts.max_sleep_interval)
+ if opts.sleep_interval is None:
+ validate(
+ opts.max_sleep_interval is None, 'min sleep interval',
+ msg='{name} must be specified; use --min-sleep-interval')
+ elif opts.max_sleep_interval is None:
+ opts.max_sleep_interval = opts.sleep_interval
+ else:
+ validate_minmax(opts.sleep_interval, opts.max_sleep_interval, 'sleep interval')
+
+ if opts.wait_for_video is not None:
+ min_wait, max_wait, *_ = map(parse_duration, [*opts.wait_for_video.split('-', 1), None])
+ validate(min_wait is not None and not (max_wait is None and '-' in opts.wait_for_video),
+ 'time range to wait for video', opts.wait_for_video)
+ validate_minmax(min_wait, max_wait, 'time range to wait for video')
+ opts.wait_for_video = (min_wait, max_wait)
+
+ # Format sort
+ for f in opts.format_sort:
+ validate_regex('format sorting', f, FormatSorter.regex)
+
+ # Postprocessor formats
+ validate_regex('merge output format', opts.merge_output_format,
+ r'({0})(/({0}))*'.format('|'.join(map(re.escape, FFmpegMergerPP.SUPPORTED_EXTS))))
+ validate_regex('audio format', opts.audioformat, FFmpegExtractAudioPP.FORMAT_RE)
+ validate_in('subtitle format', opts.convertsubtitles, FFmpegSubtitlesConvertorPP.SUPPORTED_EXTS)
+ validate_regex('thumbnail format', opts.convertthumbnails, FFmpegThumbnailsConvertorPP.FORMAT_RE)
+ validate_regex('recode video format', opts.recodevideo, FFmpegVideoConvertorPP.FORMAT_RE)
+ validate_regex('remux video format', opts.remuxvideo, FFmpegVideoRemuxerPP.FORMAT_RE)
+ if opts.audioquality:
+ opts.audioquality = opts.audioquality.strip('k').strip('K')
+ # int_or_none prevents inf, nan
+ validate_positive('audio quality', int_or_none(float_or_none(opts.audioquality), default=0))
+
+ # Retries
+ def parse_retries(name, value):
+ if value is None:
+ return None
+ elif value in ('inf', 'infinite'):
+ return float('inf')
+ try:
+ return int(value)
+ except (TypeError, ValueError):
+ validate(False, f'{name} retry count', value)
+
+ opts.retries = parse_retries('download', opts.retries)
+ opts.fragment_retries = parse_retries('fragment', opts.fragment_retries)
+ opts.extractor_retries = parse_retries('extractor', opts.extractor_retries)
+ opts.file_access_retries = parse_retries('file access', opts.file_access_retries)
+
+ # Retry sleep function
+ def parse_sleep_func(expr):
+ NUMBER_RE = r'\d+(?:\.\d+)?'
+ op, start, limit, step, *_ = (*tuple(re.fullmatch(
+ rf'(?:(linear|exp)=)?({NUMBER_RE})(?::({NUMBER_RE})?)?(?::({NUMBER_RE}))?',
+ expr.strip()).groups()), None, None)
+
+ if op == 'exp':
+ return lambda n: min(float(start) * (float(step or 2) ** n), float(limit or 'inf'))
+ else:
+ default_step = start if op or limit else 0
+ return lambda n: min(float(start) + float(step or default_step) * n, float(limit or 'inf'))
+
+ for key, expr in opts.retry_sleep.items():
+ if not expr:
+ del opts.retry_sleep[key]
+ continue
+ try:
+ opts.retry_sleep[key] = parse_sleep_func(expr)
+ except AttributeError:
+ raise ValueError(f'invalid {key} retry sleep expression {expr!r}')
+
+ # Bytes
+ def validate_bytes(name, value):
+ if value is None:
+ return None
+ numeric_limit = parse_bytes(value)
+ validate(numeric_limit is not None, 'rate limit', value)
+ return numeric_limit
+
+ opts.ratelimit = validate_bytes('rate limit', opts.ratelimit)
+ opts.throttledratelimit = validate_bytes('throttled rate limit', opts.throttledratelimit)
+ opts.min_filesize = validate_bytes('min filesize', opts.min_filesize)
+ opts.max_filesize = validate_bytes('max filesize', opts.max_filesize)
+ opts.buffersize = validate_bytes('buffer size', opts.buffersize)
+ opts.http_chunk_size = validate_bytes('http chunk size', opts.http_chunk_size)
+
+ # Output templates
def validate_outtmpl(tmpl, msg):
err = YoutubeDL.validate_outtmpl(tmpl)
if err:
- parser.error('invalid %s %r: %s' % (msg, tmpl, error_to_compat_str(err)))
+ raise ValueError(f'invalid {msg} "{tmpl}": {err}')
for k, tmpl in opts.outtmpl.items():
validate_outtmpl(tmpl, f'{k} output template')
validate_outtmpl(tmpl, f'{type_} print template')
for type_, tmpl_list in opts.print_to_file.items():
for tmpl, file in tmpl_list:
- validate_outtmpl(tmpl, f'{type_} print-to-file template')
- validate_outtmpl(file, f'{type_} print-to-file filename')
+ validate_outtmpl(tmpl, f'{type_} print to file template')
+ validate_outtmpl(file, f'{type_} print to file filename')
validate_outtmpl(opts.sponsorblock_chapter_title, 'SponsorBlock chapter title')
for k, tmpl in opts.progress_template.items():
k = f'{k[:-6]} console title' if '-title' in k else f'{k} progress'
validate_outtmpl(tmpl, f'{k} template')
- if opts.extractaudio and not opts.keepvideo and opts.format is None:
- opts.format = 'bestaudio/best'
+ outtmpl_default = opts.outtmpl.get('default')
+ if outtmpl_default == '':
+ opts.skip_download = None
+ del opts.outtmpl['default']
- if outtmpl_default is not None and not os.path.splitext(outtmpl_default)[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_default))
+ def parse_chapters(name, value, advanced=False):
+ parse_timestamp = lambda x: float('inf') if x in ('inf', 'infinite') else parse_duration(x)
+ TIMESTAMP_RE = r'''(?x)(?:
+ (?P<start_sign>-?)(?P<start>[^-]+)
+ )?\s*-\s*(?:
+ (?P<end_sign>-?)(?P<end>[^-]+)
+ )?'''
- for f in opts.format_sort:
- if re.match(InfoExtractor.FormatSort.regex, f) is None:
- parser.error('invalid format sort string "%s" specified' % f)
+ chapters, ranges, from_url = [], [], False
+ for regex in value or []:
+ if advanced and regex == '*from-url':
+ from_url = True
+ continue
+ elif not regex.startswith('*'):
+ try:
+ chapters.append(re.compile(regex))
+ except re.error as err:
+ raise ValueError(f'invalid {name} regex "{regex}" - {err}')
+ continue
+
+ for range_ in map(str.strip, regex[1:].split(',')):
+ mobj = range_ != '-' and re.fullmatch(TIMESTAMP_RE, range_)
+ dur = mobj and [parse_timestamp(mobj.group('start') or '0'), parse_timestamp(mobj.group('end') or 'inf')]
+ signs = mobj and (mobj.group('start_sign'), mobj.group('end_sign'))
+
+ err = None
+ if None in (dur or [None]):
+ err = 'Must be of the form "*start-end"'
+ elif not advanced and any(signs):
+ err = 'Negative timestamps are not allowed'
+ else:
+ dur[0] *= -1 if signs[0] else 1
+ dur[1] *= -1 if signs[1] else 1
+ if dur[1] == float('-inf'):
+ err = '"-inf" is not a valid end'
+ if err:
+ raise ValueError(f'invalid {name} time range "{regex}". {err}')
+ ranges.append(dur)
+
+ return chapters, ranges, from_url
+
+ opts.remove_chapters, opts.remove_ranges, _ = parse_chapters('--remove-chapters', opts.remove_chapters)
+ opts.download_ranges = download_range_func(*parse_chapters('--download-sections', opts.download_ranges, True))
+ # Cookies from browser
+ if opts.cookiesfrombrowser:
+ container = None
+ mobj = re.fullmatch(r'''(?x)
+ (?P<name>[^+:]+)
+ (?:\s*\+\s*(?P<keyring>[^:]+))?
+ (?:\s*:\s*(?!:)(?P<profile>.+?))?
+ (?:\s*::\s*(?P<container>.+))?
+ ''', opts.cookiesfrombrowser)
+ if mobj is None:
+ raise ValueError(f'invalid cookies from browser arguments: {opts.cookiesfrombrowser}')
+ browser_name, keyring, profile, container = mobj.group('name', 'keyring', 'profile', 'container')
+ browser_name = browser_name.lower()
+ if browser_name not in SUPPORTED_BROWSERS:
+ raise ValueError(f'unsupported browser specified for cookies: "{browser_name}". '
+ f'Supported browsers are: {", ".join(sorted(SUPPORTED_BROWSERS))}')
+ if keyring is not None:
+ keyring = keyring.upper()
+ if keyring not in SUPPORTED_KEYRINGS:
+ raise ValueError(f'unsupported keyring specified for cookies: "{keyring}". '
+ f'Supported keyrings are: {", ".join(sorted(SUPPORTED_KEYRINGS))}')
+ opts.cookiesfrombrowser = (browser_name, profile, keyring, container)
+
+ if opts.impersonate is not None:
+ opts.impersonate = ImpersonateTarget.from_str(opts.impersonate.lower())
+
+ # MetadataParser
def metadataparser_actions(f):
if isinstance(f, str):
- cmd = '--parse-metadata %s' % compat_shlex_quote(f)
+ cmd = f'--parse-metadata {shell_quote(f)}'
try:
actions = [MetadataFromFieldPP.to_action(f)]
except Exception as err:
- parser.error(f'{cmd} is invalid; {err}')
+ raise ValueError(f'{cmd} is invalid; {err}')
else:
- cmd = '--replace-in-metadata %s' % ' '.join(map(compat_shlex_quote, f))
+ cmd = f'--replace-in-metadata {shell_quote(f)}'
actions = ((MetadataParserPP.Actions.REPLACE, x, *f[1:]) for x in f[0].split(','))
for action in actions:
try:
MetadataParserPP.validate_action(*action)
except Exception as err:
- parser.error(f'{cmd} is invalid; {err}')
+ raise ValueError(f'{cmd} is invalid; {err}')
yield action
- if opts.parse_metadata is None:
- opts.parse_metadata = []
if opts.metafromtitle is not None:
- opts.parse_metadata.append('title:%s' % opts.metafromtitle)
- opts.parse_metadata = list(itertools.chain(*map(metadataparser_actions, opts.parse_metadata)))
+ opts.parse_metadata.setdefault('pre_process', []).append(f'title:{opts.metafromtitle}')
+ opts.parse_metadata = {
+ k: list(itertools.chain(*map(metadataparser_actions, v)))
+ for k, v in opts.parse_metadata.items()
+ }
+
+ # Other options
+ if opts.playlist_items is not None:
+ try:
+ tuple(PlaylistEntries.parse_playlist_items(opts.playlist_items))
+ except Exception as err:
+ raise ValueError(f'Invalid playlist-items {opts.playlist_items!r}: {err}')
+
+ opts.geo_bypass_country, opts.geo_bypass_ip_block = None, None
+ if opts.geo_bypass.lower() not in ('default', 'never'):
+ try:
+ GeoUtils.random_ipv4(opts.geo_bypass)
+ except Exception:
+ raise ValueError(f'Unsupported --xff "{opts.geo_bypass}"')
+ if len(opts.geo_bypass) == 2:
+ opts.geo_bypass_country = opts.geo_bypass
+ else:
+ opts.geo_bypass_ip_block = opts.geo_bypass
+ opts.geo_bypass = opts.geo_bypass.lower() != 'never'
- any_getting = (any(opts.forceprint.values()) or opts.dumpjson or opts.dump_single_json
- or opts.geturl or opts.gettitle or opts.getid or opts.getthumbnail
- or opts.getdescription or opts.getfilename or opts.getformat or opts.getduration)
+ opts.match_filter = match_filter_func(opts.match_filter, opts.breaking_match_filter)
- any_printing = opts.print_json
- download_archive_fn = expand_path(opts.download_archive) if opts.download_archive is not None else opts.download_archive
+ if opts.download_archive is not None:
+ opts.download_archive = expand_path(opts.download_archive)
- # If JSON is not printed anywhere, but comments are requested, save it to file
- printing_json = opts.dumpjson or opts.print_json or opts.dump_single_json
- if opts.getcomments and not printing_json:
- opts.writeinfojson = True
+ if opts.ffmpeg_location is not None:
+ opts.ffmpeg_location = expand_path(opts.ffmpeg_location)
+
+ if opts.user_agent is not None:
+ opts.headers.setdefault('User-Agent', opts.user_agent)
+ if opts.referer is not None:
+ opts.headers.setdefault('Referer', opts.referer)
if opts.no_sponsorblock:
- opts.sponsorblock_mark = set()
- opts.sponsorblock_remove = set()
- sponsorblock_query = opts.sponsorblock_mark | opts.sponsorblock_remove
+ opts.sponsorblock_mark = opts.sponsorblock_remove = set()
+
+ default_downloader = None
+ for proto, path in opts.external_downloader.items():
+ if path == 'native':
+ continue
+ ed = get_external_downloader(path)
+ if ed is None:
+ raise ValueError(
+ f'No such {format_field(proto, None, "%s ", ignore="default")}external downloader "{path}"')
+ elif ed and proto == 'default':
+ default_downloader = ed.get_basename()
- opts.remove_chapters = opts.remove_chapters or []
-
- if (opts.remove_chapters or sponsorblock_query) and opts.sponskrub is not False:
- if opts.sponskrub:
- if opts.remove_chapters:
- report_conflict('--remove-chapters', '--sponskrub')
- if opts.sponsorblock_mark:
- report_conflict('--sponsorblock-mark', '--sponskrub')
- if opts.sponsorblock_remove:
- report_conflict('--sponsorblock-remove', '--sponskrub')
- opts.sponskrub = False
- if opts.sponskrub_cut and opts.split_chapters and opts.sponskrub is not False:
- report_conflict('--split-chapter', '--sponskrub-cut')
- opts.sponskrub_cut = False
-
- if opts.remuxvideo and opts.recodevideo:
- report_conflict('--recode-video', '--remux-video')
- opts.remuxvideo = False
-
- if opts.allow_unplayable_formats:
- def report_unplayable_conflict(opt_name, arg, default=False, allowed=None):
- val = getattr(opts, opt_name)
- if (not allowed and val) or (allowed and not allowed(val)):
- report_conflict('--allow-unplayable-formats', arg)
- setattr(opts, opt_name, default)
-
- report_unplayable_conflict('extractaudio', '--extract-audio')
- report_unplayable_conflict('remuxvideo', '--remux-video')
- report_unplayable_conflict('recodevideo', '--recode-video')
- report_unplayable_conflict('addmetadata', '--embed-metadata')
- report_unplayable_conflict('addchapters', '--embed-chapters')
- report_unplayable_conflict('embed_infojson', '--embed-info-json')
- opts.embed_infojson = False
- report_unplayable_conflict('embedsubtitles', '--embed-subs')
- report_unplayable_conflict('embedthumbnail', '--embed-thumbnail')
- report_unplayable_conflict('xattrs', '--xattrs')
- report_unplayable_conflict('fixup', '--fixup', default='never', allowed=lambda x: x in (None, 'never', 'ignore'))
- opts.fixup = 'never'
- report_unplayable_conflict('remove_chapters', '--remove-chapters', default=[])
- report_unplayable_conflict('sponsorblock_remove', '--sponsorblock-remove', default=set())
- report_unplayable_conflict('sponskrub', '--sponskrub', default=set())
- opts.sponskrub = False
+ for policy in opts.color.values():
+ if policy not in ('always', 'auto', 'no_color', 'never'):
+ raise ValueError(f'"{policy}" is not a valid color policy')
+
+ warnings, deprecation_warnings = [], []
+
+ # Common mistake: -f best
+ if opts.format == 'best':
+ warnings.append('.\n '.join((
+ '"-f best" selects the best pre-merged format which is often not the best option',
+ 'To let yt-dlp download and merge the best available formats, simply do not pass any format selection',
+ 'If you know what you are doing and want only the best pre-merged format, use "-f b" instead to suppress this warning')))
+
+ # --(postprocessor/downloader)-args without name
+ def report_args_compat(name, value, key1, key2=None, where=None):
+ if key1 in value and key2 not in value:
+ warnings.append(f'{name.title()} arguments given without specifying name. '
+ f'The arguments will be given to {where or f"all {name}s"}')
+ return True
+ return False
+
+ if report_args_compat('external downloader', opts.external_downloader_args,
+ 'default', where=default_downloader) and default_downloader:
+ # Compat with youtube-dl's behavior. See https://github.com/ytdl-org/youtube-dl/commit/49c5293014bc11ec8c009856cd63cffa6296c1e1
+ opts.external_downloader_args.setdefault(default_downloader, opts.external_downloader_args.pop('default'))
+
+ if report_args_compat('post-processor', opts.postprocessor_args, 'default-compat', 'default'):
+ opts.postprocessor_args['default'] = opts.postprocessor_args.pop('default-compat')
+ opts.postprocessor_args.setdefault('sponskrub', [])
+
+ def report_conflict(arg1, opt1, arg2='--allow-unplayable-formats', opt2='allow_unplayable_formats',
+ val1=NO_DEFAULT, val2=NO_DEFAULT, default=False):
+ if val2 is NO_DEFAULT:
+ val2 = getattr(opts, opt2)
+ if not val2:
+ return
+
+ if val1 is NO_DEFAULT:
+ val1 = getattr(opts, opt1)
+ if val1:
+ warnings.append(f'{arg1} is ignored since {arg2} was given')
+ setattr(opts, opt1, default)
+
+ # Conflicting options
+ report_conflict('--playlist-reverse', 'playlist_reverse', '--playlist-random', 'playlist_random')
+ report_conflict('--playlist-reverse', 'playlist_reverse', '--lazy-playlist', 'lazy_playlist')
+ report_conflict('--playlist-random', 'playlist_random', '--lazy-playlist', 'lazy_playlist')
+ report_conflict('--dateafter', 'dateafter', '--date', 'date', default=None)
+ report_conflict('--datebefore', 'datebefore', '--date', 'date', default=None)
+ report_conflict('--exec-before-download', 'exec_before_dl_cmd',
+ '"--exec before_dl:"', 'exec_cmd', val2=opts.exec_cmd.get('before_dl'))
+ report_conflict('--id', 'useid', '--output', 'outtmpl', val2=opts.outtmpl.get('default'))
+ report_conflict('--remux-video', 'remuxvideo', '--recode-video', 'recodevideo')
+ report_conflict('--sponskrub', 'sponskrub', '--remove-chapters', 'remove_chapters')
+ report_conflict('--sponskrub', 'sponskrub', '--sponsorblock-mark', 'sponsorblock_mark')
+ report_conflict('--sponskrub', 'sponskrub', '--sponsorblock-remove', 'sponsorblock_remove')
+ report_conflict('--sponskrub-cut', 'sponskrub_cut', '--split-chapter', 'split_chapters',
+ val1=opts.sponskrub and opts.sponskrub_cut)
+
+ # Conflicts with --allow-unplayable-formats
+ report_conflict('--embed-metadata', 'addmetadata')
+ report_conflict('--embed-chapters', 'addchapters')
+ report_conflict('--embed-info-json', 'embed_infojson')
+ report_conflict('--embed-subs', 'embedsubtitles')
+ report_conflict('--embed-thumbnail', 'embedthumbnail')
+ report_conflict('--extract-audio', 'extractaudio')
+ report_conflict('--fixup', 'fixup', val1=opts.fixup not in (None, 'never', 'ignore'), default='never')
+ report_conflict('--recode-video', 'recodevideo')
+ report_conflict('--remove-chapters', 'remove_chapters', default=[])
+ report_conflict('--remux-video', 'remuxvideo')
+ report_conflict('--sponskrub', 'sponskrub')
+ report_conflict('--sponsorblock-remove', 'sponsorblock_remove', default=set())
+ report_conflict('--xattrs', 'xattrs')
+
+ # Fully deprecated options
+ def report_deprecation(val, old, new=None):
+ if not val:
+ return
+ deprecation_warnings.append(
+ f'{old} is deprecated and may be removed in a future version. Use {new} instead' if new
+ else f'{old} is deprecated and may not work as expected')
+
+ report_deprecation(opts.sponskrub, '--sponskrub', '--sponsorblock-mark or --sponsorblock-remove')
+ report_deprecation(not opts.prefer_ffmpeg, '--prefer-avconv', 'ffmpeg')
+ # report_deprecation(opts.include_ads, '--include-ads') # We may re-implement this in future
+ # report_deprecation(opts.call_home, '--call-home') # We may re-implement this in future
+ # report_deprecation(opts.writeannotations, '--write-annotations') # It's just that no website has it
+
+ # Dependent options
+ opts.date = DateRange.day(opts.date) if opts.date else DateRange(opts.dateafter, opts.datebefore)
+
+ if opts.exec_before_dl_cmd:
+ opts.exec_cmd['before_dl'] = opts.exec_before_dl_cmd
+
+ if opts.useid: # --id is not deprecated in youtube-dl
+ opts.outtmpl['default'] = '%(id)s.%(ext)s'
+
+ if opts.overwrites: # --force-overwrites implies --no-continue
+ opts.continue_dl = False
if (opts.addmetadata or opts.sponsorblock_mark) and opts.addchapters is None:
+ # Add chapters when adding metadata or marking sponsors
opts.addchapters = True
- # PostProcessors
- postprocessors = list(opts.add_postprocessors)
+ if opts.extractaudio and not opts.keepvideo and opts.format is None:
+ # Do not unnecessarily download audio
+ opts.format = 'bestaudio/best'
+
+ if opts.getcomments and opts.writeinfojson is None and not opts.embed_infojson:
+ # If JSON is not printed anywhere, but comments are requested, save it to file
+ if not opts.dumpjson or opts.print_json or opts.dump_single_json:
+ opts.writeinfojson = True
+
+ if opts.allsubtitles and not (opts.embedsubtitles or opts.writeautomaticsub):
+ # --all-sub automatically sets --write-sub if --write-auto-sub is not given
+ opts.writesubtitles = True
+
+ if opts.addmetadata and opts.embed_infojson is None:
+ # If embedding metadata and infojson is present, embed it
+ opts.embed_infojson = 'if_exists'
+
+ # Ask for passwords
+ if opts.username is not None and opts.password is None:
+ opts.password = getpass.getpass('Type account password and press [Return]: ')
+ if opts.ap_username is not None and opts.ap_password is None:
+ opts.ap_password = getpass.getpass('Type TV provider account password and press [Return]: ')
+
+ return warnings, deprecation_warnings
+
+
+def get_postprocessors(opts):
+ yield from opts.add_postprocessors
+
+ for when, actions in opts.parse_metadata.items():
+ yield {
+ 'key': 'MetadataParser',
+ 'actions': actions,
+ 'when': when,
+ }
+ sponsorblock_query = opts.sponsorblock_mark | opts.sponsorblock_remove
if sponsorblock_query:
- postprocessors.append({
+ yield {
'key': 'SponsorBlock',
'categories': sponsorblock_query,
'api': opts.sponsorblock_api,
- # Run this immediately after extraction is complete
- 'when': 'pre_process'
- })
- if opts.parse_metadata:
- postprocessors.append({
- 'key': 'MetadataParser',
- 'actions': opts.parse_metadata,
- # Run this immediately after extraction is complete
- 'when': 'pre_process'
- })
+ 'when': 'after_filter',
+ }
if opts.convertsubtitles:
- postprocessors.append({
+ yield {
'key': 'FFmpegSubtitlesConvertor',
'format': opts.convertsubtitles,
- # Run this before the actual video download
- 'when': 'before_dl'
- })
+ 'when': 'before_dl',
+ }
if opts.convertthumbnails:
- postprocessors.append({
+ yield {
'key': 'FFmpegThumbnailsConvertor',
'format': opts.convertthumbnails,
- # Run this before the actual video download
- 'when': 'before_dl'
- })
+ 'when': 'before_dl',
+ }
if opts.extractaudio:
- postprocessors.append({
+ yield {
'key': 'FFmpegExtractAudio',
'preferredcodec': opts.audioformat,
'preferredquality': opts.audioquality,
'nopostoverwrites': opts.nopostoverwrites,
- })
+ }
if opts.remuxvideo:
- postprocessors.append({
+ yield {
'key': 'FFmpegVideoRemuxer',
'preferedformat': opts.remuxvideo,
- })
+ }
if opts.recodevideo:
- postprocessors.append({
+ yield {
'key': 'FFmpegVideoConvertor',
'preferedformat': opts.recodevideo,
- })
+ }
# If ModifyChapters is going to remove chapters, subtitles must already be in the container.
if opts.embedsubtitles:
- already_have_subtitle = opts.writesubtitles and 'no-keep-subs' not in compat_opts
- postprocessors.append({
+ keep_subs = 'no-keep-subs' not in opts.compat_opts
+ yield {
'key': 'FFmpegEmbedSubtitle',
# already_have_subtitle = True prevents the file from being deleted after embedding
- 'already_have_subtitle': already_have_subtitle
- })
- if not opts.writeautomaticsub and 'no-keep-subs' not in compat_opts:
+ 'already_have_subtitle': opts.writesubtitles and keep_subs,
+ }
+ if not opts.writeautomaticsub and keep_subs:
opts.writesubtitles = True
- # --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
+
# ModifyChapters must run before FFmpegMetadataPP
- remove_chapters_patterns, remove_ranges = [], []
- for regex in opts.remove_chapters:
- if regex.startswith('*'):
- dur = list(map(parse_duration, regex[1:].split('-')))
- if len(dur) == 2 and all(t is not None for t in dur):
- remove_ranges.append(tuple(dur))
- continue
- parser.error(f'invalid --remove-chapters time range {regex!r}. Must be of the form *start-end')
- try:
- remove_chapters_patterns.append(re.compile(regex))
- except re.error as err:
- parser.error(f'invalid --remove-chapters regex {regex!r} - {err}')
if opts.remove_chapters or sponsorblock_query:
- postprocessors.append({
+ yield {
'key': 'ModifyChapters',
- 'remove_chapters_patterns': remove_chapters_patterns,
+ 'remove_chapters_patterns': opts.remove_chapters,
'remove_sponsor_segments': opts.sponsorblock_remove,
- 'remove_ranges': remove_ranges,
+ 'remove_ranges': opts.remove_ranges,
'sponsorblock_chapter_title': opts.sponsorblock_chapter_title,
- 'force_keyframes': opts.force_keyframes_at_cuts
- })
+ 'force_keyframes': opts.force_keyframes_at_cuts,
+ }
# FFmpegMetadataPP should be run after FFmpegVideoConvertorPP and
# FFmpegExtractAudioPP as containers before conversion may not support
# metadata (3gp, webm, etc.)
# source and target containers. From this point the container won't change,
# so metadata can be added here.
if opts.addmetadata or opts.addchapters or opts.embed_infojson:
- if opts.embed_infojson is None:
- opts.embed_infojson = 'if_exists'
- postprocessors.append({
+ yield {
'key': 'FFmpegMetadata',
'add_chapters': opts.addchapters,
'add_metadata': opts.addmetadata,
'add_infojson': opts.embed_infojson,
- })
+ }
# Deprecated
# This should be above EmbedThumbnail since sponskrub removes the thumbnail attachment
# but must be below EmbedSubtitle and FFmpegMetadata
# See https://github.com/yt-dlp/yt-dlp/issues/204 , https://github.com/faissaloo/SponSkrub/issues/29
# If opts.sponskrub is None, sponskrub is used, but it silently fails if the executable can't be found
if opts.sponskrub is not False:
- postprocessors.append({
+ yield {
'key': 'SponSkrub',
'path': opts.sponskrub_path,
'args': opts.sponskrub_args,
'force': opts.sponskrub_force,
'ignoreerror': opts.sponskrub is None,
'_from_cli': True,
- })
+ }
if opts.embedthumbnail:
- postprocessors.append({
+ yield {
'key': 'EmbedThumbnail',
# already_have_thumbnail = True prevents the file from being deleted after embedding
- 'already_have_thumbnail': opts.writethumbnail
- })
+ 'already_have_thumbnail': opts.writethumbnail,
+ }
if not opts.writethumbnail:
opts.writethumbnail = True
opts.outtmpl['pl_thumbnail'] = ''
if opts.split_chapters:
- postprocessors.append({
+ yield {
'key': 'FFmpegSplitChapters',
'force_keyframes': opts.force_keyframes_at_cuts,
- })
+ }
# XAttrMetadataPP should be run after post-processors that may change file contents
if opts.xattrs:
- postprocessors.append({'key': 'XAttrMetadata'})
+ yield {'key': 'XAttrMetadata'}
if opts.concat_playlist != 'never':
- postprocessors.append({
+ yield {
'key': 'FFmpegConcat',
'only_multi_video': opts.concat_playlist != 'always',
'when': 'playlist',
- })
+ }
# Exec must be the last PP of each category
- if opts.exec_before_dl_cmd:
- opts.exec_cmd.setdefault('before_dl', opts.exec_before_dl_cmd)
for when, exec_cmd in opts.exec_cmd.items():
- postprocessors.append({
+ yield {
'key': 'Exec',
'exec_cmd': exec_cmd,
- # Run this only after the files have been moved to their final locations
'when': when,
- })
+ }
- def report_args_compat(arg, name):
- warnings.append('%s given without specifying name. The arguments will be given to all %s' % (arg, name))
- if 'default' in opts.external_downloader_args:
- report_args_compat('--downloader-args', 'external downloaders')
+ParsedOptions = collections.namedtuple('ParsedOptions', ('parser', 'options', 'urls', 'ydl_opts'))
- if 'default-compat' in opts.postprocessor_args and 'default' not in opts.postprocessor_args:
- report_args_compat('--post-processor-args', 'post-processors')
- opts.postprocessor_args.setdefault('sponskrub', [])
- opts.postprocessor_args['default'] = opts.postprocessor_args['default-compat']
- def report_deprecation(val, old, new=None):
- if not val:
- return
- deprecation_warnings.append(
- f'{old} is deprecated and may be removed in a future version. Use {new} instead' if new
- else f'{old} is deprecated and may not work as expected')
+def parse_options(argv=None):
+ """@returns ParsedOptions(parser, opts, urls, ydl_opts)"""
+ parser, opts, urls = parseOpts(argv)
+ urls = get_urls(urls, opts.batchfile, -1 if opts.quiet and not opts.verbose else opts.verbose)
- report_deprecation(opts.sponskrub, '--sponskrub', '--sponsorblock-mark or --sponsorblock-remove')
- report_deprecation(not opts.prefer_ffmpeg, '--prefer-avconv', 'ffmpeg')
- report_deprecation(opts.include_ads, '--include-ads')
- # report_deprecation(opts.call_home, '--call-home') # We may re-implement this in future
- # report_deprecation(opts.writeannotations, '--write-annotations') # It's just that no website has it
+ set_compat_opts(opts)
+ try:
+ warnings, deprecation_warnings = validate_options(opts)
+ except ValueError as err:
+ parser.error(f'{err}\n')
+
+ postprocessors = list(get_postprocessors(opts))
+
+ print_only = bool(opts.forceprint) and all(k not in opts.forceprint for k in POSTPROCESS_WHEN[3:])
+ any_getting = any(getattr(opts, k) for k in (
+ 'dumpjson', 'dump_single_json', 'getdescription', 'getduration', 'getfilename',
+ 'getformat', 'getid', 'getthumbnail', 'gettitle', 'geturl',
+ ))
+ if opts.quiet is None:
+ opts.quiet = any_getting or opts.print_json or bool(opts.forceprint)
+
+ playlist_pps = [pp for pp in postprocessors if pp.get('when') == 'playlist']
+ write_playlist_infojson = (opts.writeinfojson and not opts.clean_infojson
+ and opts.allow_playlist_files and opts.outtmpl.get('pl_infojson') != '')
+ if not any((
+ opts.extract_flat,
+ opts.dump_single_json,
+ opts.forceprint.get('playlist'),
+ opts.print_to_file.get('playlist'),
+ write_playlist_infojson,
+ )):
+ if not playlist_pps:
+ opts.extract_flat = 'discard'
+ elif playlist_pps == [{'key': 'FFmpegConcat', 'only_multi_video': True, 'when': 'playlist'}]:
+ opts.extract_flat = 'discard_in_playlist'
final_ext = (
opts.recodevideo if opts.recodevideo in FFmpegVideoConvertorPP.SUPPORTED_EXTS
else opts.remuxvideo if opts.remuxvideo in FFmpegVideoRemuxerPP.SUPPORTED_EXTS
- else opts.audioformat if (opts.extractaudio and opts.audioformat != 'best')
+ else opts.audioformat if (opts.extractaudio and opts.audioformat in FFmpegExtractAudioPP.SUPPORTED_EXTS)
else None)
- match_filter = (
- None if opts.match_filter is None
- else match_filter_func(opts.match_filter))
-
- ydl_opts = {
+ return ParsedOptions(parser, opts, urls, {
'usenetrc': opts.usenetrc,
'netrc_location': opts.netrc_location,
+ 'netrc_cmd': opts.netrc_cmd,
'username': opts.username,
'password': opts.password,
'twofactor': opts.twofactor,
'ap_mso': opts.ap_mso,
'ap_username': opts.ap_username,
'ap_password': opts.ap_password,
- 'quiet': (opts.quiet or any_getting or any_printing),
+ 'client_certificate': opts.client_certificate,
+ 'client_certificate_key': opts.client_certificate_key,
+ 'client_certificate_password': opts.client_certificate_password,
+ 'quiet': opts.quiet,
'no_warnings': opts.no_warnings,
'forceurl': opts.geturl,
'forcetitle': opts.gettitle,
'forcejson': opts.dumpjson or opts.print_json,
'dump_single_json': opts.dump_single_json,
'force_write_download_archive': opts.force_write_download_archive,
- 'simulate': (any_getting or None) if opts.simulate is None else opts.simulate,
+ 'simulate': (print_only or any_getting or None) if opts.simulate is None else opts.simulate,
'skip_download': opts.skip_download,
'format': opts.format,
'allow_unplayable_formats': opts.allow_unplayable_formats,
'windowsfilenames': opts.windowsfilenames,
'ignoreerrors': opts.ignoreerrors,
'force_generic_extractor': opts.force_generic_extractor,
+ 'allowed_extractors': opts.allowed_extractors or ['default'],
'ratelimit': opts.ratelimit,
'throttledratelimit': opts.throttledratelimit,
'overwrites': opts.overwrites,
'file_access_retries': opts.file_access_retries,
'fragment_retries': opts.fragment_retries,
'extractor_retries': opts.extractor_retries,
+ 'retry_sleep_functions': opts.retry_sleep,
'skip_unavailable_fragments': opts.skip_unavailable_fragments,
'keep_fragments': opts.keep_fragments,
'concurrent_fragment_downloads': opts.concurrent_fragment_downloads,
'noprogress': opts.quiet if opts.noprogress is None else opts.noprogress,
'progress_with_newline': opts.progress_with_newline,
'progress_template': opts.progress_template,
+ 'progress_delta': opts.progress_delta,
'playliststart': opts.playliststart,
'playlistend': opts.playlistend,
'playlistreverse': opts.playlist_reverse,
'playlistrandom': opts.playlist_random,
+ 'lazy_playlist': opts.lazy_playlist,
'noplaylist': opts.noplaylist,
- 'logtostderr': outtmpl_default == '-',
+ 'logtostderr': opts.outtmpl.get('default') == '-',
'consoletitle': opts.consoletitle,
'nopart': opts.nopart,
'updatetime': opts.updatetime,
'verbose': opts.verbose,
'dump_intermediate_pages': opts.dump_intermediate_pages,
'write_pages': opts.write_pages,
+ 'load_pages': opts.load_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,
+ 'daterange': opts.date,
'cachedir': opts.cachedir,
'youtube_print_sig_code': opts.youtube_print_sig_code,
'age_limit': opts.age_limit,
- 'download_archive': download_archive_fn,
+ 'download_archive': opts.download_archive,
'break_on_existing': opts.break_on_existing,
'break_on_reject': opts.break_on_reject,
'break_per_url': opts.break_per_url,
'legacyserverconnect': opts.legacy_server_connect,
'nocheckcertificate': opts.no_check_certificate,
'prefer_insecure': opts.prefer_insecure,
+ 'enable_file_urls': opts.enable_file_urls,
+ 'http_headers': opts.headers,
'proxy': opts.proxy,
'socket_timeout': opts.socket_timeout,
'bidi_workaround': opts.bidi_workaround,
'postprocessors': postprocessors,
'fixup': opts.fixup,
'source_address': opts.source_address,
+ 'impersonate': opts.impersonate,
'call_home': opts.call_home,
'sleep_interval_requests': opts.sleep_interval_requests,
'sleep_interval': opts.sleep_interval,
'max_sleep_interval': opts.max_sleep_interval,
'sleep_interval_subtitles': opts.sleep_interval_subtitles,
'external_downloader': opts.external_downloader,
+ 'download_ranges': opts.download_ranges,
+ 'force_keyframes_at_cuts': opts.force_keyframes_at_cuts,
'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,
+ 'match_filter': opts.match_filter,
+ 'color': opts.color,
'ffmpeg_location': opts.ffmpeg_location,
'hls_prefer_native': opts.hls_prefer_native,
'hls_use_mpegts': opts.hls_use_mpegts,
'geo_bypass_ip_block': opts.geo_bypass_ip_block,
'_warnings': warnings,
'_deprecation_warnings': deprecation_warnings,
- 'compat_opts': compat_opts,
- }
+ 'compat_opts': opts.compat_opts,
+ })
+
+
+def _real_main(argv=None):
+ setproctitle('yt-dlp')
+
+ parser, opts, all_urls, ydl_opts = parse_options(argv)
+
+ # Dump user agent
+ if opts.dump_user_agent:
+ ua = traverse_obj(opts.headers, 'User-Agent', casesense=False, default=std_headers['User-Agent'])
+ write_string(f'{ua}\n', out=sys.stdout)
+ return
+
+ if print_extractor_information(opts, all_urls):
+ return
+
+ # We may need ffmpeg_location without having access to the YoutubeDL instance
+ # See https://github.com/yt-dlp/yt-dlp/issues/2191
+ if opts.ffmpeg_location:
+ FFmpegPostProcessor._ffmpeg_location.set(opts.ffmpeg_location)
with YoutubeDL(ydl_opts) as ydl:
+ pre_process = opts.update_self or opts.rm_cachedir
actual_use = all_urls or opts.load_info_filename
- # Remove cache dir
if opts.rm_cachedir:
ydl.cache.remove()
- # Update version
- if opts.update_self:
- # If updater returns True, exit. Required for windows
- if run_update(ydl):
- if actual_use:
- sys.exit('ERROR: The program must exit for the update to complete')
- sys.exit()
+ try:
+ updater = Updater(ydl, opts.update_self)
+ if opts.update_self and updater.update() and actual_use:
+ if updater.cmd:
+ return updater.restart()
+ # This code is reachable only for zip variant in py < 3.10
+ # It makes sense to exit here, but the old behavior is to continue
+ ydl.report_warning('Restart yt-dlp to use the updated version')
+ # return 100, 'ERROR: The program must exit for the update to complete'
+ except Exception:
+ traceback.print_exc()
+ ydl._download_retcode = 100
+
+ if opts.list_impersonate_targets:
+
+ known_targets = [
+ # List of simplified targets we know are supported,
+ # to help users know what dependencies may be required.
+ (ImpersonateTarget('chrome'), 'curl_cffi'),
+ (ImpersonateTarget('edge'), 'curl_cffi'),
+ (ImpersonateTarget('safari'), 'curl_cffi'),
+ ]
+
+ available_targets = ydl._get_available_impersonate_targets()
+
+ def make_row(target, handler):
+ return [
+ join_nonempty(target.client.title(), target.version, delim='-') or '-',
+ join_nonempty((target.os or '').title(), target.os_version, delim='-') or '-',
+ handler,
+ ]
+
+ rows = [make_row(target, handler) for target, handler in available_targets]
+
+ for known_target, known_handler in known_targets:
+ if not any(
+ known_target in target and handler == known_handler
+ for target, handler in available_targets
+ ):
+ rows.append([
+ ydl._format_out(text, ydl.Styles.SUPPRESS)
+ for text in make_row(known_target, f'{known_handler} (not available)')
+ ])
+
+ ydl.to_screen('[info] Available impersonate targets')
+ ydl.to_stdout(render_table(['Client', 'OS', 'Source'], rows, extra_gap=2, delim='-'))
+ return
- # Maybe do nothing
if not actual_use:
- if opts.update_self or opts.rm_cachedir:
- sys.exit()
+ if pre_process:
+ return ydl._download_retcode
+
+ args = sys.argv[1:] if argv is None else argv
+ ydl.warn_if_short_id(args)
+
+ # Show a useful error message and wait for keypress if not launched from shell on Windows
+ if not args and compat_os_name == 'nt' and getattr(sys, 'frozen', False):
+ import ctypes.wintypes
+ import msvcrt
+
+ kernel32 = ctypes.WinDLL('Kernel32')
- ydl.warn_if_short_id(sys.argv[1:] if argv is None else argv)
+ buffer = (1 * ctypes.wintypes.DWORD)()
+ attached_processes = kernel32.GetConsoleProcessList(buffer, 1)
+ # If we only have a single process attached, then the executable was double clicked
+ # When using `pyinstaller` with `--onefile`, two processes get attached
+ is_onefile = hasattr(sys, '_MEIPASS') and os.path.basename(sys._MEIPASS).startswith('_MEI')
+ if attached_processes == 1 or is_onefile and attached_processes == 2:
+ print(parser._generate_error_message(
+ 'Do not double-click the executable, instead call it from a command line.\n'
+ 'Please read the README for further information on how to use yt-dlp: '
+ 'https://github.com/yt-dlp/yt-dlp#readme'))
+ msvcrt.getch()
+ _exit(2)
parser.error(
'You must provide at least one URL.\n'
'Type yt-dlp --help to see a list of all options.')
+ parser.destroy()
try:
if opts.load_info_filename is not None:
- retcode = ydl.download_with_info_file(expand_path(opts.load_info_filename))
+ if all_urls:
+ ydl.report_warning('URLs are ignored due to --load-info-json')
+ return ydl.download_with_info_file(expand_path(opts.load_info_filename))
else:
- retcode = ydl.download(all_urls)
+ return ydl.download(all_urls)
except DownloadCancelled:
ydl.to_screen('Aborting remaining downloads')
- retcode = 101
-
- sys.exit(retcode)
+ return 101
def main(argv=None):
+ global _IN_CLI
+ _IN_CLI = True
try:
- _real_main(argv)
+ _exit(*variadic(_real_main(argv)))
except DownloadError:
- sys.exit(1)
+ _exit(1)
except SameFileError as e:
- sys.exit(f'ERROR: {e}')
+ _exit(f'ERROR: {e}')
except KeyboardInterrupt:
- sys.exit('\nERROR: Interrupted by user')
+ _exit('\nERROR: Interrupted by user')
except BrokenPipeError as e:
# https://docs.python.org/3/library/signal.html#note-on-sigpipe
devnull = os.open(os.devnull, os.O_WRONLY)
os.dup2(devnull, sys.stdout.fileno())
- sys.exit(f'\nERROR: {e}')
+ _exit(f'\nERROR: {e}')
+ except optparse.OptParseError as e:
+ _exit(2, f'\n{e}')
+
+from .extractor import gen_extractors, list_extractors
-__all__ = ['main', 'YoutubeDL', 'gen_extractors', 'list_extractors']
+__all__ = [
+ 'main',
+ 'YoutubeDL',
+ 'parse_options',
+ 'gen_extractors',
+ 'list_extractors',
+]
projects / yt-dlp.git / blobdiff