from .downloader import FFmpegFD, get_suitable_downloader, shorten_protocol_name
from .downloader.rtmp import rtmpdump_version
from .extractor import gen_extractor_classes, get_info_extractor
+from .extractor.common import UnsupportedURLIE
from .extractor.openload import PhantomJSwrapper
from .minicurses import format_text
from .postprocessor import _PLUGIN_CLASSES as plugin_postprocessors
network_exceptions,
number_of_digits,
orderedSet,
+ orderedSet_from_options,
parse_filesize,
preferredencoding,
prepend_extension,
Default is 'only_download' for CLI, but False for API
skip_playlist_after_errors: Number of allowed failures until the rest of
the playlist is skipped
- force_generic_extractor: Force downloader to use the generic extractor
+ allowed_extractors: List of regexes to match against extractor names that are allowed
overwrites: Overwrite all video and metadata files if True,
overwrite only non-video files if None
and don't overwrite any file if False
subtitleslangs: List of languages of the subtitles to download (can be regex).
The list may contain "all" to refer to all the available
subtitles. The language can be prefixed with a "-" to
- exclude it from the requested languages. Eg: ['all', '-live_chat']
+ exclude it from the requested languages, e.g. ['all', '-live_chat']
keepvideo: Keep the video file after post-processing
daterange: A DateRange object, download only if the upload_date is in the range.
skip_download: Skip the actual download of the video file
should act on each input URL as opposed to for the entire queue
cookiefile: File name or text stream from where cookies should be read and dumped to
cookiesfrombrowser: A tuple containing the name of the browser, the profile
- name/pathfrom where cookies are loaded, and the name of the
- keyring. Eg: ('chrome', ) or ('vivaldi', 'default', 'BASICTEXT')
+ name/path from where cookies are loaded, and the name of the
+ keyring, e.g. ('chrome', ) or ('vivaldi', 'default', 'BASICTEXT')
legacyserverconnect: Explicitly allow HTTPS connection to servers that do not
support RFC 5746 secure renegotiation
nocheckcertificate: Do not verify SSL certificates
* index: Section number (Optional)
force_keyframes_at_cuts: Re-encode the video when downloading ranges to get precise cuts
noprogress: Do not print the progress bar
+ live_from_start: Whether to download livestreams videos from the start
The following parameters are not used by YoutubeDL itself, they are used by
the downloader (see yt_dlp/downloader/common.py):
discontinuities such as ad breaks (default: False)
extractor_args: A dictionary of arguments to be passed to the extractors.
See "EXTRACTOR ARGUMENTS" for details.
- Eg: {'youtube': {'skip': ['dash', 'hls']}}
+ E.g. {'youtube': {'skip': ['dash', 'hls']}}
mark_watched: Mark videos watched (even with --simulate). Only for YouTube
The following options are deprecated and may be removed in the future:
+ force_generic_extractor: Force downloader to use the generic extractor
+ - Use allowed_extractors = ['generic', 'default']
playliststart: - Use playlist_items
Playlist item to start at.
playlistend: - Use playlist_items
self._ies_instances[ie_key] = ie
ie.set_downloader(self)
- def _get_info_extractor_class(self, ie_key):
- ie = self._ies.get(ie_key)
- if ie is None:
- ie = get_info_extractor(ie_key)
- self.add_info_extractor(ie)
- return ie
-
def get_info_extractor(self, ie_key):
"""
Get an instance of an IE with name ie_key, it will try to get one from
"""
Add the InfoExtractors returned by gen_extractors to the end of the list
"""
- for ie in gen_extractor_classes():
- self.add_info_extractor(ie)
+ all_ies = {ie.IE_NAME.lower(): ie for ie in gen_extractor_classes()}
+ all_ies['end'] = UnsupportedURLIE()
+ try:
+ ie_names = orderedSet_from_options(
+ self.params.get('allowed_extractors', ['default']), {
+ 'all': list(all_ies),
+ 'default': [name for name, ie in all_ies.items() if ie._ENABLED],
+ }, use_regex=True)
+ except re.error as e:
+ raise ValueError(f'Wrong regex for allowed_extractors: {e.pattern}')
+ for name in ie_names:
+ self.add_info_extractor(all_ies[name])
+ self.write_debug(f'Loaded {len(ie_names)} extractors')
def add_post_processor(self, pp, when='post_process'):
"""Add a PostProcessor object to the end of the chain."""
# outtmpl should be expand_path'ed before template dict substitution
# because meta fields may contain env variables we don't want to
- # be expanded. For example, for outtmpl "%(title)s.%(ext)s" and
+ # be expanded. E.g. for outtmpl "%(title)s.%(ext)s" and
# title "Hello $PATH", we don't want `$PATH` to be expanded.
return expand_path(outtmpl).replace(sep, '')
ie_key = 'Generic'
if ie_key:
- ies = {ie_key: self._get_info_extractor_class(ie_key)}
+ ies = {ie_key: self._ies[ie_key]} if ie_key in self._ies else {}
else:
ies = self._ies
- for ie_key, ie in ies.items():
+ for key, ie in ies.items():
if not ie.suitable(url):
continue
'and will probably not work.')
temp_id = ie.get_temp_id(url)
- if temp_id is not None and self.in_download_archive({'id': temp_id, 'ie_key': ie_key}):
- self.to_screen(f'[{ie_key}] {temp_id}: has already been recorded in the archive')
+ if temp_id is not None and self.in_download_archive({'id': temp_id, 'ie_key': key}):
+ self.to_screen(f'[{key}] {temp_id}: has already been recorded in the archive')
if self.params.get('break_on_existing', False):
raise ExistingVideoReached()
break
- return self.__extract_info(url, self.get_info_extractor(ie_key), download, extra_info, process)
+ return self.__extract_info(url, self.get_info_extractor(key), download, extra_info, process)
else:
- self.report_error('no suitable InfoExtractor for URL %s' % url)
+ extractors_restricted = self.params.get('allowed_extractors') not in (None, ['default'])
+ self.report_error(f'No suitable extractor{format_field(ie_key, None, " (%s)")} found for URL {url}',
+ tb=False if extractors_restricted else None)
def _handle_extraction_exceptions(func):
@functools.wraps(func)
filter_parts.append(string)
def _remove_unused_ops(tokens):
- # Remove operators that we don't use and join them with the surrounding strings
- # for example: 'mp4' '-' 'baseline' '-' '16x9' is converted to 'mp4-baseline-16x9'
+ # Remove operators that we don't use and join them with the surrounding strings.
+ # E.g. 'mp4' '-' 'baseline' '-' '16x9' is converted to 'mp4-baseline-16x9'
ALLOWED_OPS = ('/', '+', ',', '(', ')')
last_string, last_start, last_end, last_line = None, None, None, None
for type, string, start, end, line in tokens:
if self.params.get('allsubtitles', False):
requested_langs = all_sub_langs
elif self.params.get('subtitleslangs', False):
- # A list is used so that the order of languages will be the same as
- # given in subtitleslangs. See https://github.com/yt-dlp/yt-dlp/issues/1041
- requested_langs = []
- for lang_re in self.params.get('subtitleslangs'):
- discard = lang_re[0] == '-'
- if discard:
- lang_re = lang_re[1:]
- if lang_re == 'all':
- if discard:
- requested_langs = []
- else:
- requested_langs.extend(all_sub_langs)
- continue
- current_langs = filter(re.compile(lang_re + '$').match, all_sub_langs)
- if discard:
- for lang in current_langs:
- while lang in requested_langs:
- requested_langs.remove(lang)
- else:
- requested_langs.extend(current_langs)
- requested_langs = orderedSet(requested_langs)
+ try:
+ requested_langs = orderedSet_from_options(
+ self.params.get('subtitleslangs'), {'all': all_sub_langs}, use_regex=True)
+ except re.error as e:
+ raise ValueError(f'Wrong regex for subtitlelangs: {e.pattern}')
elif normal_sub_langs:
requested_langs = ['en'] if 'en' in normal_sub_langs else normal_sub_langs[:1]
else:
return False
vid_ids = [self._make_archive_id(info_dict)]
- vid_ids.extend(info_dict.get('_old_archive_ids', []))
+ vid_ids.extend(info_dict.get('_old_archive_ids') or [])
return any(id_ in self.archive for id_ in vid_ids)
def record_download_archive(self, info_dict):