get_postprocessor,
)
from .postprocessor.ffmpeg import resolve_mapping as resolve_recode_mapping
-from .update import detect_variant
+from .update import REPOSITORY, current_git_head, detect_variant
from .utils import (
DEFAULT_OUTTMPL,
IDENTITY,
args_to_str,
bug_reports_message,
date_from_str,
+ deprecation_warning,
determine_ext,
determine_protocol,
encode_compat_str,
get_domain,
int_or_none,
iri_to_uri,
+ is_path_like,
join_nonempty,
locked_file,
make_archive_id,
matchtitle: Download only matching titles.
rejecttitle: Reject downloads for matching titles.
logger: Log messages to a logging.Logger instance.
- logtostderr: Log messages to stderr instead of stdout.
- consoletitle: Display progress in console window's titlebar.
+ logtostderr: Print everything to stderr instead of stdout.
+ consoletitle: Display progress in console window's titlebar.
writedescription: Write the video description to a .description file
writeinfojson: Write the video description to a .info.json file
clean_infojson: Remove private fields from the infojson
downloaded.
Videos without view count information are always
downloaded. None for no limit.
- download_archive: File name of a file where all downloads are recorded.
- Videos already present in the file are not downloaded
- again.
+ download_archive: A set, or the name of a file where all downloads are recorded.
+ Videos already present in the file are not downloaded again.
break_on_existing: Stop the download process after attempting to download a
file that is in the archive.
break_on_reject: Stop the download process when encountering a video that
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/path from where cookies are loaded, and the name of the
- keyring, e.g. ('chrome', ) or ('vivaldi', 'default', 'BASICTEXT')
+ name/path from where cookies are loaded, the name of the keyring,
+ and the container name, e.g. ('chrome', ) or
+ ('vivaldi', 'default', 'BASICTEXT') or ('firefox', 'default', None, 'Meta')
legacyserverconnect: Explicitly allow HTTPS connection to servers that do not
support RFC 5746 secure renegotiation
nocheckcertificate: Do not verify SSL certificates
for msg in self.params.get('_warnings', []):
self.report_warning(msg)
for msg in self.params.get('_deprecation_warnings', []):
- self.deprecation_warning(msg)
+ self.deprecated_feature(msg)
self.params['compat_opts'] = set(self.params.get('compat_opts', ()))
if 'list-formats' in self.params['compat_opts']:
def preload_download_archive(fn):
"""Preload the archive, if any is specified"""
+ archive = set()
if fn is None:
- return False
+ return archive
+ elif not is_path_like(fn):
+ return fn
+
self.write_debug(f'Loading archive file {fn!r}')
try:
with locked_file(fn, 'r', encoding='utf-8') as archive_file:
for line in archive_file:
- self.archive.add(line.strip())
+ archive.add(line.strip())
except OSError as ioe:
if ioe.errno != errno.ENOENT:
raise
- return False
- return True
+ return archive
- self.archive = set()
- preload_download_archive(self.params.get('download_archive'))
+ self.archive = preload_download_archive(self.params.get('download_archive'))
def warn_if_short_id(self, argv):
# short YouTube ID starting with dash?
def to_stdout(self, message, skip_eol=False, quiet=None):
"""Print message to stdout"""
if quiet is not None:
- self.deprecation_warning('"YoutubeDL.to_stdout" no longer accepts the argument quiet. Use "YoutubeDL.to_screen" instead')
+ self.deprecation_warning('"YoutubeDL.to_stdout" no longer accepts the argument quiet. '
+ 'Use "YoutubeDL.to_screen" instead')
if skip_eol is not False:
- self.deprecation_warning('"YoutubeDL.to_stdout" no longer accepts the argument skip_eol. Use "YoutubeDL.to_screen" instead')
+ self.deprecation_warning('"YoutubeDL.to_stdout" no longer accepts the argument skip_eol. '
+ 'Use "YoutubeDL.to_screen" instead')
self._write_string(f'{self._bidi_workaround(message)}\n', self._out_files.out)
def to_screen(self, message, skip_eol=False, quiet=None):
return
self.to_stderr(f'{self._format_err("WARNING:", self.Styles.WARNING)} {message}', only_once)
- def deprecation_warning(self, message):
+ def deprecation_warning(self, message, *, stacklevel=0):
+ deprecation_warning(
+ message, stacklevel=stacklevel + 1, printer=self.report_error, is_error=False)
+
+ def deprecated_feature(self, message):
if self.params.get('logger') is not None:
- self.params['logger'].warning(f'DeprecationWarning: {message}')
- else:
- self.to_stderr(f'{self._format_err("DeprecationWarning:", self.Styles.ERROR)} {message}', True)
+ self.params['logger'].warning(f'Deprecated Feature: {message}')
+ self.to_stderr(f'{self._format_err("Deprecated Feature:", self.Styles.ERROR)} {message}', True)
def report_error(self, message, *args, **kwargs):
'''
def get_output_path(self, dir_type='', filename=None):
paths = self.params.get('paths', {})
- assert isinstance(paths, dict)
+ assert isinstance(paths, dict), '"paths" parameter must be a dictionary'
path = os.path.join(
expand_path(paths.get('home', '').strip()),
expand_path(paths.get(dir_type, '').strip()) if dir_type else '',
'-': float.__sub__,
}
# Field is of the form key1.key2...
- # where keys (except first) can be string, int or slice
- FIELD_RE = r'\w*(?:\.(?:\w+|{num}|{num}?(?::{num}?){{1,2}}))*'.format(num=r'(?:-?\d+)')
+ # where keys (except first) can be string, int, slice or "{field, ...}"
+ FIELD_INNER_RE = r'(?:\w+|%(num)s|%(num)s?(?::%(num)s?){1,2})' % {'num': r'(?:-?\d+)'}
+ FIELD_RE = r'\w*(?:\.(?:%(inner)s|{%(field)s(?:,%(field)s)*}))*' % {
+ 'inner': FIELD_INNER_RE,
+ 'field': rf'\w*(?:\.{FIELD_INNER_RE})*'
+ }
MATH_FIELD_RE = rf'(?:{FIELD_RE}|-?{NUMBER_RE})'
MATH_OPERATORS_RE = r'(?:%s)' % '|'.join(map(re.escape, MATH_FUNCTIONS.keys()))
INTERNAL_FORMAT_RE = re.compile(rf'''(?x)
(?:\|(?P<default>.*?))?
)$''')
- def _traverse_infodict(k):
- k = k.split('.')
- if k[0] == '':
- k.pop(0)
- return traverse_obj(info_dict, k, is_user_input=True, traverse_string=True)
+ def _traverse_infodict(fields):
+ fields = [f for x in re.split(r'\.({.+?})\.?', fields)
+ for f in ([x] if x.startswith('{') else x.split('.'))]
+ for i in (0, -1):
+ if fields and not fields[i]:
+ fields.pop(i)
+
+ for i, f in enumerate(fields):
+ if not f.startswith('{'):
+ continue
+ assert f.endswith('}'), f'No closing brace for {f} in {fields}'
+ fields[i] = {k: k.split('.') for k in f[1:-1].split(',')}
+
+ return traverse_obj(info_dict, fields, is_user_input=True, traverse_string=True)
def get_value(mdict):
# Object traversal
delim = '\n' if '#' in flags else ', '
value, fmt = delim.join(map(str, variadic(value, allowed_types=(str, bytes)))), str_fmt
elif fmt[-1] == 'j': # json
- value, fmt = json.dumps(value, default=_dumpjson_default, indent=4 if '#' in flags else None), str_fmt
+ value, fmt = json.dumps(
+ value, default=_dumpjson_default,
+ indent=4 if '#' in flags else None, ensure_ascii=False), str_fmt
elif fmt[-1] == 'h': # html
- value, fmt = escapeHTML(value), str_fmt
+ value, fmt = escapeHTML(str(value)), str_fmt
elif fmt[-1] == 'q': # quoted
value = map(str, variadic(value) if '#' in flags else [value])
value, fmt = ' '.join(map(compat_shlex_quote, value)), str_fmt
def extract_info(self, url, download=True, ie_key=None, extra_info=None,
process=True, force_generic_extractor=False):
"""
- Return a list with a dictionary for each video extracted.
+ Extract and return the information dictionary of the URL
Arguments:
- url -- URL to extract
+ @param url URL to extract
Keyword arguments:
- download -- whether to download videos during extraction
- ie_key -- extractor key hint
- extra_info -- dictionary containing the extra values to add to each result
- process -- whether to resolve all unresolved references (URLs, playlist items),
- must be True for download to work.
- force_generic_extractor -- force using the generic extractor
+ @param download Whether to download videos
+ @param process Whether to resolve all unresolved references (URLs, playlist items).
+ Must be True for download to work
+ @param ie_key Use only the extractor with this key
+
+ @param extra_info Dictionary containing the extra values to add to the info (For internal use only)
+ @force_generic_extractor Force using the generic extractor (Deprecated; use ie_key='Generic')
"""
if extra_info is None:
info_dict['_has_drm'] = any(f.get('has_drm') for f in formats) or None
if not self.params.get('allow_unplayable_formats'):
formats = [f for f in formats if not f.get('has_drm')]
- if info_dict['_has_drm'] and formats and all(
- f.get('acodec') == f.get('vcodec') == 'none' for f in formats):
- self.report_warning(
- 'This video is DRM protected and only images are available for download. '
- 'Use --list-formats to see them')
+
+ if formats and all(f.get('acodec') == f.get('vcodec') == 'none' for f in formats):
+ self.report_warning(
+ f'{"This video is DRM protected and " if info_dict["_has_drm"] else ""}'
+ 'only images are available for download. Use --list-formats to see them'.capitalize())
get_from_start = not info_dict.get('is_live') or bool(self.params.get('live_from_start'))
if not get_from_start:
'--live-from-start is passed, but there are no formats that can be downloaded from the start. '
'If you want to download from the current time, use --no-live-from-start'))
- if not formats:
- self.raise_no_formats(info_dict)
-
def is_wellformed(f):
url = f.get('url')
if not url:
return True
# Filter out malformed formats for better extraction robustness
- formats = list(filter(is_wellformed, formats))
+ formats = list(filter(is_wellformed, formats or []))
+
+ if not formats:
+ self.raise_no_formats(info_dict)
formats_dict = {}
if lang not in available_subs:
available_subs[lang] = cap_info
- if (not self.params.get('writesubtitles') and not
- self.params.get('writeautomaticsub') or not
- available_subs):
+ if not available_subs or (
+ not self.params.get('writesubtitles')
+ and not self.params.get('writeautomaticsub')):
return None
all_sub_langs = tuple(available_subs.keys())
else:
requested_langs = ['en'] if 'en' in all_sub_langs else all_sub_langs[:1]
if requested_langs:
- self.write_debug('Downloading subtitles: %s' % ', '.join(requested_langs))
+ self.to_screen(f'[info] {video_id}: Downloading subtitles: {", ".join(requested_langs)}')
formats_query = self.params.get('subtitlesformat', 'best')
formats_preference = formats_query.split('/') if formats_query else []
info_copy['automatic_captions_table'] = self.render_subtitles_table(info_dict.get('id'), info_dict.get('automatic_captions'))
def format_tmpl(tmpl):
- mobj = re.match(r'\w+(=?)$', tmpl)
- if mobj and mobj.group(1):
- return f'{tmpl[:-1]} = %({tmpl[:-1]})r'
- elif mobj:
- return f'%({tmpl})s'
- return tmpl
+ mobj = re.fullmatch(r'([\w.:,]|-\d|(?P<dict>{([\w.:,]|-\d)+}))+=?', tmpl)
+ if not mobj:
+ return tmpl
+
+ fmt = '%({})s'
+ if tmpl.startswith('{'):
+ tmpl = f'.{tmpl}'
+ if tmpl.endswith('='):
+ tmpl, fmt = tmpl[:-1], '{0} = %({0})#j'
+ return '\n'.join(map(fmt.format, [tmpl] if mobj.group('dict') else tmpl.split(',')))
for tmpl in self.params['forceprint'].get(key, []):
self.to_stdout(self.evaluate_outtmpl(format_tmpl(tmpl), info_copy))
return info_dict
info_dict.setdefault('epoch', int(time.time()))
info_dict.setdefault('_type', 'video')
+ info_dict.setdefault('_version', {
+ 'version': __version__,
+ 'current_git_head': current_git_head(),
+ 'release_git_head': RELEASE_GIT_HEAD,
+ 'repository': REPOSITORY,
+ })
if remove_private_keys:
reject = lambda k, v: v is None or k.startswith('__') or k in {
return make_archive_id(extractor, video_id)
def in_download_archive(self, info_dict):
- fn = self.params.get('download_archive')
- if fn is None:
+ if not self.archive:
return False
vid_ids = [self._make_archive_id(info_dict)]
return
vid_id = self._make_archive_id(info_dict)
assert vid_id
+
self.write_debug(f'Adding to archive: {vid_id}')
- with locked_file(fn, 'a', encoding='utf-8') as archive_file:
- archive_file.write(vid_id + '\n')
+ if isinstance(fn, os.PathLike):
+ with locked_file(fn, 'a', encoding='utf-8') as archive_file:
+ archive_file.write(vid_id + '\n')
self.archive.add(vid_id)
@staticmethod
if VARIANT not in (None, 'pip'):
source += '*'
write_debug(join_nonempty(
- 'yt-dlp version', __version__,
+ f'{"yt-dlp" if REPOSITORY == "yt-dlp/yt-dlp" else REPOSITORY} version',
+ __version__,
f'[{RELEASE_GIT_HEAD}]' if RELEASE_GIT_HEAD else '',
'' if source == 'unknown' else f'({source})',
delim=' '))
if self.params['compat_opts']:
write_debug('Compatibility options: %s' % ', '.join(self.params['compat_opts']))
- if source == 'source':
- try:
- stdout, _, _ = Popen.run(
- ['git', 'rev-parse', '--short', 'HEAD'],
- text=True, cwd=os.path.dirname(os.path.abspath(__file__)),
- stdout=subprocess.PIPE, stderr=subprocess.PIPE)
- if re.fullmatch('[0-9a-f]+', stdout.strip()):
- write_debug(f'Git HEAD: {stdout.strip()}')
- except Exception:
- with contextlib.suppress(Exception):
- sys.exc_clear()
-
+ if current_git_head():
+ write_debug(f'Git HEAD: {current_git_head()}')
write_debug(system_identifier())
exe_versions, ffmpeg_features = FFmpegPostProcessor.get_versions_and_features(self)
projects / yt-dlp.git / blobdiff