from .compat import (
compat_basestring,
+ compat_brotli,
compat_get_terminal_size,
compat_kwargs,
compat_numeric_types,
ExistingVideoReached,
expand_path,
ExtractorError,
+ filter_dict,
float_or_none,
format_bytes,
format_field,
formatSeconds,
GeoRestrictedError,
get_domain,
+ has_certifi,
HEADRequest,
InAdvancePagedList,
int_or_none,
make_dir,
make_HTTPS_handler,
MaxDownloadsReached,
+ merge_headers,
network_exceptions,
+ NO_DEFAULT,
number_of_digits,
orderedSet,
OUTTMPL_TYPES,
See "Sorting Formats" for more details.
format_sort_force: Force the given format_sort. see "Sorting Formats"
for more details.
+ prefer_free_formats: Whether to prefer video formats with free containers
+ over non-free ones of same quality.
allow_multiple_video_streams: Allow multiple video streams to be merged
into a single file
allow_multiple_audio_streams: Allow multiple audio streams to be merged
nocheckcertificate: Do not verify SSL certificates
prefer_insecure: Use HTTP instead of HTTPS to retrieve information.
At the moment, this is only supported by YouTube.
+ http_headers: A dictionary of custom headers to be used for all requests
proxy: URL of the proxy server to use
geo_verification_proxy: URL of the proxy to use for IP address verification
on geo-restricted sites.
'track_number', 'disc_number', 'release_year',
))
+ _format_fields = {
+ # NB: Keep in sync with the docstring of extractor/common.py
+ 'url', 'manifest_url', 'ext', 'format', 'format_id', 'format_note',
+ 'width', 'height', 'resolution', 'dynamic_range', 'tbr', 'abr', 'acodec', 'asr',
+ 'vbr', 'fps', 'vcodec', 'container', 'filesize', 'filesize_approx',
+ 'player_url', 'protocol', 'fragment_base_url', 'fragments', 'is_from_start',
+ 'preference', 'language', 'language_preference', 'quality', 'source_preference',
+ 'http_headers', 'stretched_ratio', 'no_resume', 'has_drm', 'downloader_options',
+ 'page_url', 'app', 'play_path', 'tc_url', 'flash_version', 'rtmp_live', 'rtmp_conn', 'rtmp_protocol', 'rtmp_real_time'
+ }
_format_selection_exts = {
'audio': {'m4a', 'mp3', 'ogg', 'aac'},
'video': {'mp4', 'flv', 'webm', '3gp'},
'storyboards': {'mhtml'},
}
- params = None
- _ies = {}
- _pps = {k: [] for k in POSTPROCESS_WHEN}
- _printed_messages = set()
- _first_webpage_request = True
- _download_retcode = None
- _num_downloads = None
- _playlist_level = 0
- _playlist_urls = set()
- _screen_file = None
-
def __init__(self, params=None, auto_init=True):
"""Create a FileDownloader object with the given options.
@param auto_init Whether to load the default extractors and print header (if verbose).
"""
if params is None:
params = {}
+ self.params = params
self._ies = {}
self._ies_instances = {}
self._pps = {k: [] for k in POSTPROCESS_WHEN}
self._download_retcode = 0
self._num_downloads = 0
self._num_videos = 0
- self._screen_file = [sys.stdout, sys.stderr][params.get('logtostderr', False)]
- self._err_file = sys.stderr
- self.params = params
+ self._playlist_level = 0
+ self._playlist_urls = set()
self.cache = Cache(self)
windows_enable_vt_mode()
+ self._out_files = {
+ 'error': sys.stderr,
+ 'print': sys.stderr if self.params.get('logtostderr') else sys.stdout,
+ 'console': None if compat_os_name == 'nt' else next(
+ filter(supports_terminal_sequences, (sys.stderr, sys.stdout)), None)
+ }
+ self._out_files['screen'] = sys.stderr if self.params.get('quiet') else self._out_files['print']
self._allow_colors = {
- 'screen': not self.params.get('no_color') and supports_terminal_sequences(self._screen_file),
- 'err': not self.params.get('no_color') and supports_terminal_sequences(self._err_file),
+ type_: not self.params.get('no_color') and supports_terminal_sequences(self._out_files[type_])
+ for type_ in ('screen', 'error')
}
if sys.version_info < (3, 6):
sp_kwargs = dict(
stdin=subprocess.PIPE,
stdout=slave,
- stderr=self._err_file)
+ stderr=self._out_files['error'])
try:
self._output_process = Popen(['bidiv'] + width_args, **sp_kwargs)
except OSError:
else self.params['format'] if callable(self.params['format'])
else self.build_format_selector(self.params['format']))
+ # Set http_headers defaults according to std_headers
+ self.params['http_headers'] = merge_headers(std_headers, self.params.get('http_headers', {}))
+
self._setup_opener()
if auto_init:
self._printed_messages.add(message)
write_string(message, out=out, encoding=self.params.get('encoding'))
- def to_stdout(self, message, skip_eol=False, quiet=False):
+ 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._write_string(
+ '%s%s' % (self._bidi_workaround(message), ('' if skip_eol else '\n')),
+ self._out_files['print'])
+
+ def to_screen(self, message, skip_eol=False, quiet=None):
+ """Print message to screen if not in quiet mode"""
if self.params.get('logger'):
self.params['logger'].debug(message)
- elif not quiet or self.params.get('verbose'):
- self._write_string(
- '%s%s' % (self._bidi_workaround(message), ('' if skip_eol else '\n')),
- self._err_file if quiet else self._screen_file)
+ return
+ if (self.params.get('quiet') if quiet is None else quiet) and not self.params.get('verbose'):
+ return
+ self._write_string(
+ '%s%s' % (self._bidi_workaround(message), ('' if skip_eol else '\n')),
+ self._out_files['screen'])
def to_stderr(self, message, only_once=False):
"""Print message to stderr"""
if self.params.get('logger'):
self.params['logger'].error(message)
else:
- self._write_string('%s\n' % self._bidi_workaround(message), self._err_file, only_once=only_once)
+ self._write_string('%s\n' % self._bidi_workaround(message), self._out_files['error'], only_once=only_once)
+
+ def _send_console_code(self, code):
+ if compat_os_name == 'nt' or not self._out_files['console']:
+ return
+ self._write_string(code, self._out_files['console'])
def to_console_title(self, message):
if not self.params.get('consoletitle', False):
# c_wchar_p() might not be necessary if `message` is
# already of type unicode()
ctypes.windll.kernel32.SetConsoleTitleW(ctypes.c_wchar_p(message))
- elif 'TERM' in os.environ:
- self._write_string('\033]0;%s\007' % message, self._screen_file)
+ else:
+ self._send_console_code(f'\033]0;{message}\007')
def save_console_title(self):
- if not self.params.get('consoletitle', False):
+ if not self.params.get('consoletitle') or self.params.get('simulate'):
return
- if self.params.get('simulate'):
- return
- if compat_os_name != 'nt' and 'TERM' in os.environ:
- # Save the title on stack
- self._write_string('\033[22;0t', self._screen_file)
+ self._send_console_code('\033[22;0t') # Save the title on stack
def restore_console_title(self):
- if not self.params.get('consoletitle', False):
+ if not self.params.get('consoletitle') or self.params.get('simulate'):
return
- if self.params.get('simulate'):
- return
- if compat_os_name != 'nt' and 'TERM' in os.environ:
- # Restore the title from stack
- self._write_string('\033[23;0t', self._screen_file)
+ self._send_console_code('\033[23;0t') # Restore the title from stack
def __enter__(self):
self.save_console_title()
raise DownloadError(message, exc_info)
self._download_retcode = 1
- def to_screen(self, message, skip_eol=False):
- """Print message to stdout if not in quiet mode"""
- self.to_stdout(
- message, skip_eol, quiet=self.params.get('quiet', False))
-
class Styles(Enum):
HEADERS = 'yellow'
EMPHASIS = 'light blue'
def _format_screen(self, *args, **kwargs):
return self._format_text(
- self._screen_file, self._allow_colors['screen'], *args, **kwargs)
+ self._out_files['screen'], self._allow_colors['screen'], *args, **kwargs)
def _format_err(self, *args, **kwargs):
return self._format_text(
- self._err_file, self._allow_colors['err'], *args, **kwargs)
+ self._out_files['error'], self._allow_colors['error'], *args, **kwargs)
def report_warning(self, message, only_once=False):
'''
except UnicodeEncodeError:
self.to_screen('Deleting existing file')
- def raise_no_formats(self, info, forced=False):
+ def raise_no_formats(self, info, forced=False, *, msg=None):
has_drm = info.get('__has_drm')
- msg = 'This video is DRM protected' if has_drm else 'No video formats found!'
- expected = self.params.get('ignore_no_formats_error')
- if forced or not expected:
+ ignored, expected = self.params.get('ignore_no_formats_error'), bool(msg)
+ msg = msg or has_drm and 'This video is DRM protected' or 'No video formats found!'
+ if forced or not ignored:
raise ExtractorError(msg, video_id=info['id'], ie=info['extractor'],
- expected=has_drm or expected)
+ expected=has_drm or ignored or expected)
else:
self.report_warning(msg)
@staticmethod
def _copy_infodict(info_dict):
info_dict = dict(info_dict)
- for key in ('__original_infodict', '__postprocessors'):
- info_dict.pop(key, None)
+ info_dict.pop('__postprocessors', None)
return info_dict
def prepare_outtmpl(self, outtmpl, info_dict, sanitize=False):
(?P<fields>{field})
(?P<maths>(?:{math_op}{math_field})*)
(?:>(?P<strf_format>.+?))?
- (?P<alternate>(?<!\\),[^|&)]+)?
- (?:&(?P<replacement>.*?))?
- (?:\|(?P<default>.*?))?
- $'''.format(field=FIELD_RE, math_op=MATH_OPERATORS_RE, math_field=MATH_FIELD_RE))
+ (?P<remaining>
+ (?P<alternate>(?<!\\),[^|&)]+)?
+ (?:&(?P<replacement>.*?))?
+ (?:\|(?P<default>.*?))?
+ )$'''.format(field=FIELD_RE, math_op=MATH_OPERATORS_RE, math_field=MATH_FIELD_RE))
def _traverse_infodict(k):
k = k.split('.')
na = self.params.get('outtmpl_na_placeholder', 'NA')
def filename_sanitizer(key, value, restricted=self.params.get('restrictfilenames')):
- return sanitize_filename(str(value), restricted=restricted,
- is_id=re.search(r'(^|[_.])id(\.|$)', key))
+ return sanitize_filename(str(value), restricted=restricted, is_id=(
+ bool(re.search(r'(^|[_.])id(\.|$)', key))
+ if 'filename-sanitization' in self.params.get('compat_opts', [])
+ else NO_DEFAULT))
sanitizer = sanitize if callable(sanitize) else filename_sanitizer
sanitize = bool(sanitize)
value = get_value(mobj)
replacement = mobj['replacement']
if value is None and mobj['alternate']:
- mobj = re.match(INTERNAL_FORMAT_RE, mobj['alternate'][1:])
+ mobj = re.match(INTERNAL_FORMAT_RE, mobj['remaining'][1:])
else:
break
min_wait, max_wait = self.params.get('wait_for_video')
diff = try_get(ie_result, lambda x: x['release_timestamp'] - time.time())
if diff is None and ie_result.get('live_status') == 'is_upcoming':
- diff = random.randrange(min_wait, max_wait) if (max_wait and min_wait) else (max_wait or min_wait)
+ diff = round(random.uniform(min_wait, max_wait) if (max_wait and min_wait) else (max_wait or min_wait), 0)
self.report_warning('Release time of video is not known')
elif (diff or 0) <= 0:
self.report_warning('Video should already be available according to extracted info')
if not info:
return info
- force_properties = dict(
- (k, v) for k, v in ie_result.items() if v is not None)
- for f in ('_type', 'url', 'id', 'extractor', 'extractor_key', 'ie_key'):
- if f in force_properties:
- del force_properties[f]
new_result = info.copy()
- new_result.update(force_properties)
+ new_result.update(filter_dict(ie_result, lambda k, v: (
+ v is not None and k not in {'_type', 'url', 'id', 'extractor', 'extractor_key', 'ie_key'})))
# Extracted info may not be a video result (i.e.
# info.get('_type', 'video') != video) but rather an url or
self._playlist_level += 1
self._playlist_urls.add(webpage_url)
+ self._fill_common_fields(ie_result, False)
self._sanitize_thumbnails(ie_result)
try:
return self.__process_playlist(ie_result, download)
ie_result['entries'] = playlist_results
# Write the updated info to json
- if _infojson_written and self._write_info_json(
+ if _infojson_written is True and self._write_info_json(
'updated playlist', ie_result,
self.prepare_filename(ie_copy, 'pl_infojson'), overwrite=True) is None:
return
return _build_selector_function(parsed_selector)
def _calc_headers(self, info_dict):
- res = std_headers.copy()
- res.update(info_dict.get('http_headers') or {})
+ res = merge_headers(self.params['http_headers'], info_dict.get('http_headers') or {})
cookies = self._calc_cookies(info_dict)
if cookies:
else:
info_dict['thumbnails'] = thumbnails
+ def _fill_common_fields(self, info_dict, is_video=True):
+ # TODO: move sanitization here
+ if is_video:
+ # playlists are allowed to lack "title"
+ info_dict['fulltitle'] = info_dict.get('title')
+ if 'title' not in info_dict:
+ raise ExtractorError('Missing "title" field in extractor result',
+ video_id=info_dict['id'], ie=info_dict['extractor'])
+ elif not info_dict.get('title'):
+ self.report_warning('Extractor failed to obtain "title". Creating a generic title instead')
+ info_dict['title'] = f'{info_dict["extractor"]} video #{info_dict["id"]}'
+
+ if info_dict.get('duration') is not None:
+ info_dict['duration_string'] = formatSeconds(info_dict['duration'])
+
+ for ts_key, date_key in (
+ ('timestamp', 'upload_date'),
+ ('release_timestamp', 'release_date'),
+ ('modified_timestamp', 'modified_date'),
+ ):
+ if info_dict.get(date_key) is None and info_dict.get(ts_key) is not None:
+ # Working around out-of-range timestamp values (e.g. negative ones on Windows,
+ # see http://bugs.python.org/issue1646728)
+ try:
+ upload_date = datetime.datetime.utcfromtimestamp(info_dict[ts_key])
+ info_dict[date_key] = upload_date.strftime('%Y%m%d')
+ except (ValueError, OverflowError, OSError):
+ pass
+
+ live_keys = ('is_live', 'was_live')
+ live_status = info_dict.get('live_status')
+ if live_status is None:
+ for key in live_keys:
+ if info_dict.get(key) is False:
+ continue
+ if info_dict.get(key):
+ live_status = key
+ break
+ if all(info_dict.get(key) is False for key in live_keys):
+ live_status = 'not_live'
+ if live_status:
+ info_dict['live_status'] = live_status
+ for key in live_keys:
+ if info_dict.get(key) is None:
+ info_dict[key] = (live_status == key)
+
+ # Auto generate title fields corresponding to the *_number fields when missing
+ # in order to always have clean titles. This is very common for TV series.
+ for field in ('chapter', 'season', 'episode'):
+ if info_dict.get('%s_number' % field) is not None and not info_dict.get(field):
+ info_dict[field] = '%s %d' % (field.capitalize(), info_dict['%s_number' % field])
+
def process_video_result(self, info_dict, download=True):
assert info_dict.get('_type', 'video') == 'video'
self._num_videos += 1
elif not info_dict.get('id'):
raise ExtractorError('Extractor failed to obtain "id"', ie=info_dict['extractor'])
- info_dict['fulltitle'] = info_dict.get('title')
- if 'title' not in info_dict:
- raise ExtractorError('Missing "title" field in extractor result',
- video_id=info_dict['id'], ie=info_dict['extractor'])
- elif not info_dict.get('title'):
- self.report_warning('Extractor failed to obtain "title". Creating a generic title instead')
- info_dict['title'] = f'{info_dict["extractor"]} video #{info_dict["id"]}'
-
def report_force_conversion(field, field_not, conversion):
self.report_warning(
'"%s" field is not %s - forcing %s conversion, there is an error in extractor'
sanitize_string_field(info_dict, 'id')
sanitize_numeric_fields(info_dict)
+ if (info_dict.get('duration') or 0) <= 0 and info_dict.pop('duration', None):
+ self.report_warning('"duration" field is negative, there is an error in extractor')
if 'playlist' not in info_dict:
# It isn't part of a playlist
if info_dict.get('display_id') is None and 'id' in info_dict:
info_dict['display_id'] = info_dict['id']
- if info_dict.get('duration') is not None:
- info_dict['duration_string'] = formatSeconds(info_dict['duration'])
-
- for ts_key, date_key in (
- ('timestamp', 'upload_date'),
- ('release_timestamp', 'release_date'),
- ('modified_timestamp', 'modified_date'),
- ):
- if info_dict.get(date_key) is None and info_dict.get(ts_key) is not None:
- # Working around out-of-range timestamp values (e.g. negative ones on Windows,
- # see http://bugs.python.org/issue1646728)
- try:
- upload_date = datetime.datetime.utcfromtimestamp(info_dict[ts_key])
- info_dict[date_key] = upload_date.strftime('%Y%m%d')
- except (ValueError, OverflowError, OSError):
- pass
-
- live_keys = ('is_live', 'was_live')
- live_status = info_dict.get('live_status')
- if live_status is None:
- for key in live_keys:
- if info_dict.get(key) is False:
- continue
- if info_dict.get(key):
- live_status = key
- break
- if all(info_dict.get(key) is False for key in live_keys):
- live_status = 'not_live'
- if live_status:
- info_dict['live_status'] = live_status
- for key in live_keys:
- if info_dict.get(key) is None:
- info_dict[key] = (live_status == key)
-
- # Auto generate title fields corresponding to the *_number fields when missing
- # in order to always have clean titles. This is very common for TV series.
- for field in ('chapter', 'season', 'episode'):
- if info_dict.get('%s_number' % field) is not None and not info_dict.get(field):
- info_dict[field] = '%s %d' % (field.capitalize(), info_dict['%s_number' % field])
+ self._fill_common_fields(info_dict)
for cc_kind in ('subtitles', 'automatic_captions'):
cc = info_dict.get(cc_kind)
info_dict['__has_drm'] = any(f.get('has_drm') for f in formats)
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 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 info_dict.get('is_live'):
- get_from_start = bool(self.params.get('live_from_start'))
+ get_from_start = not info_dict.get('is_live') or bool(self.params.get('live_from_start'))
+ if not get_from_start:
+ info_dict['title'] += ' ' + datetime.datetime.now().strftime('%Y-%m-%d %H:%M')
+ if info_dict.get('is_live') and formats:
formats = [f for f in formats if bool(f.get('is_from_start')) == get_from_start]
- if not get_from_start:
- info_dict['title'] += ' ' + datetime.datetime.now().strftime('%Y-%m-%d %H:%M')
+ if get_from_start and not formats:
+ self.raise_no_formats(info_dict, msg='--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, pass --no-live-from-start')
if not formats:
self.raise_no_formats(info_dict)
if '__x_forwarded_for_ip' in info_dict:
del info_dict['__x_forwarded_for_ip']
- # TODO Central sorting goes here
-
if self.params.get('check_formats') is True:
formats = LazyList(self._check_formats(formats[::-1]), reverse=True)
info_dict, _ = self.pre_process(info_dict)
+ if self._match_entry(info_dict, incomplete=self._format_fields) is not None:
+ return info_dict
+
+ self.post_extract(info_dict)
+ info_dict, _ = self.pre_process(info_dict, 'after_filter')
+
# The pre-processors may have modified the formats
formats = info_dict.get('formats', [info_dict])
if not formats_to_download:
if not self.params.get('ignore_no_formats_error'):
- raise ExtractorError('Requested format is not available', expected=True,
- video_id=info_dict['id'], ie=info_dict['extractor'])
+ raise ExtractorError(
+ 'Requested format is not available. Use --list-formats for a list of available formats',
+ expected=True, video_id=info_dict['id'], ie=info_dict['extractor'])
self.report_warning('Requested format is not available')
# Process what we can, even without any available formats.
formats_to_download = [{}]
+ ', '.join([f['format_id'] for f in formats_to_download]))
max_downloads_reached = False
for i, fmt in enumerate(formats_to_download):
- formats_to_download[i] = new_info = dict(info_dict)
- # Save a reference to the original info_dict so that it can be modified in process_info if needed
+ formats_to_download[i] = new_info = self._copy_infodict(info_dict)
new_info.update(fmt)
- new_info['__original_infodict'] = info_dict
try:
self.process_info(new_info)
except MaxDownloadsReached:
max_downloads_reached = True
- new_info.pop('__original_infodict')
# Remove copied info
for key, val in tuple(new_info.items()):
if info_dict.get(key) == val:
filename = self.evaluate_outtmpl(file_tmpl, info_dict)
tmpl = format_tmpl(tmpl)
self.to_screen(f'[info] Writing {tmpl!r} to: {filename}')
- with io.open(filename, 'a', encoding='utf-8') as f:
- f.write(self.evaluate_outtmpl(tmpl, info_copy) + '\n')
+ if self._ensure_dir_exists(filename):
+ with io.open(filename, 'a', encoding='utf-8') as f:
+ f.write(self.evaluate_outtmpl(tmpl, info_copy) + '\n')
def __forced_printings(self, info_dict, filename, incomplete):
def print_mandatory(field, actual_field=None):
if info_dict.get('requested_formats') is not None:
# For RTMP URLs, also include the playpath
info_dict['urls'] = '\n'.join(f['url'] + f.get('play_path', '') for f in info_dict['requested_formats'])
- elif 'url' in info_dict:
+ elif info_dict.get('url'):
info_dict['urls'] = info_dict['url'] + info_dict.get('play_path', '')
if (self.params.get('forcejson')
return None
def process_info(self, info_dict):
- """Process a single resolved IE result. (Modified it in-place)"""
+ """Process a single resolved IE result. (Modifies it in-place)"""
assert info_dict.get('_type', 'video') == 'video'
original_infodict = info_dict
if 'format' not in info_dict and 'ext' in info_dict:
info_dict['format'] = info_dict['ext']
+ # This is mostly just for backward compatibility of process_info
+ # As a side-effect, this allows for format-specific filters
if self._match_entry(info_dict) is not None:
info_dict['__write_download_archive'] = 'ignore'
return
+ # Does nothing under normal operation - for backward compatibility of process_info
self.post_extract(info_dict)
self._num_downloads += 1
# Write internet shortcut files
def _write_link_file(link_type):
- if 'webpage_url' not in info_dict:
- self.report_error('Cannot write internet shortcut file because the "webpage_url" field is missing in the media information')
- return False
+ url = try_get(info_dict['webpage_url'], iri_to_uri)
+ if not url:
+ self.report_warning(
+ f'Cannot write internet shortcut file because the actual URL of "{info_dict["webpage_url"]}" is unknown')
+ return True
linkfn = replace_extension(self.prepare_filename(info_dict, 'link'), link_type, info_dict.get('ext'))
if not self._ensure_dir_exists(encodeFilename(linkfn)):
return False
self.to_screen(f'[info] Writing internet shortcut (.{link_type}) to: {linkfn}')
with io.open(encodeFilename(to_high_limit_path(linkfn)), 'w', encoding='utf-8',
newline='\r\n' if link_type == 'url' else '\n') as linkfile:
- template_vars = {'url': iri_to_uri(info_dict['webpage_url'])}
+ template_vars = {'url': url}
if link_type == 'desktop':
template_vars['filename'] = linkfn[:-(len(link_type) + 1)]
linkfile.write(LINK_TEMPLATES[link_type] % template_vars)
return info_dict
info_dict.setdefault('epoch', int(time.time()))
info_dict.setdefault('_type', 'video')
- remove_keys = {'__original_infodict'} # Always remove this since this may contain a copy of the entire dict
- keep_keys = ['_type'] # Always keep this to facilitate load-info-json
+
if remove_private_keys:
- remove_keys |= {
+ reject = lambda k, v: v is None or (k.startswith('_') and k != '_type') or k in {
'requested_downloads', 'requested_formats', 'requested_subtitles', 'requested_entries',
'entries', 'filepath', 'infojson_filename', 'original_url', 'playlist_autonumber',
}
- reject = lambda k, v: k not in keep_keys and (
- k.startswith('_') or k in remove_keys or v is None)
else:
- reject = lambda k, v: k in remove_keys
+ reject = lambda k, v: False
def filter_fn(obj):
if isinstance(obj, dict):
actual_post_extract(video_dict or {})
return
- post_extractor = info_dict.get('__post_extractor') or (lambda: {})
- extra = post_extractor().items()
- info_dict.update(extra)
- info_dict.pop('__post_extractor', None)
-
- original_infodict = info_dict.get('__original_infodict') or {}
- original_infodict.update(extra)
- original_infodict.pop('__post_extractor', None)
+ post_extractor = info_dict.pop('__post_extractor', None) or (lambda: {})
+ info_dict.update(post_extractor())
actual_post_extract(info_dict or {})
return
def get_encoding(stream):
- ret = getattr(stream, 'encoding', 'missing (%s)' % type(stream).__name__)
+ ret = str(getattr(stream, 'encoding', 'missing (%s)' % type(stream).__name__))
if not supports_terminal_sequences(stream):
from .compat import WINDOWS_VT_MODE
ret += ' (No VT)' if WINDOWS_VT_MODE is False else ' (No ANSI)'
encoding_str = 'Encodings: locale %s, fs %s, out %s, err %s, pref %s' % (
locale.getpreferredencoding(),
sys.getfilesystemencoding(),
- get_encoding(self._screen_file), get_encoding(self._err_file),
+ get_encoding(self._out_files['screen']), get_encoding(self._out_files['error']),
self.get_encoding())
logger = self.params.get('logger')
from .cookies import SQLITE_AVAILABLE, SECRETSTORAGE_AVAILABLE
lib_str = join_nonempty(
+ compat_brotli and compat_brotli.__name__,
+ has_certifi and 'certifi',
compat_pycrypto_AES and compat_pycrypto_AES.__name__.split('.')[0],
SECRETSTORAGE_AVAILABLE and 'secretstorage',
has_mutagen and 'mutagen',
return encoding
def _write_info_json(self, label, ie_result, infofn, overwrite=None):
- ''' Write infojson and returns True = written, False = skip, None = error '''
+ ''' Write infojson and returns True = written, 'exists' = Already exists, False = skip, None = error '''
if overwrite is None:
overwrite = self.params.get('overwrites', True)
if not self.params.get('writeinfojson'):
return None
elif not overwrite and os.path.exists(infofn):
self.to_screen(f'[info] {label.title()} metadata is already present')
- else:
- self.to_screen(f'[info] Writing {label} metadata as JSON to: {infofn}')
- try:
- write_json_file(self.sanitize_info(ie_result, self.params.get('clean_infojson', True)), infofn)
- except (OSError, IOError):
- self.report_error(f'Cannot write {label} metadata to JSON file {infofn}')
- return None
- return True
+ return 'exists'
+
+ self.to_screen(f'[info] Writing {label} metadata as JSON to: {infofn}')
+ try:
+ write_json_file(self.sanitize_info(ie_result, self.params.get('clean_infojson', True)), infofn)
+ return True
+ except (OSError, IOError):
+ self.report_error(f'Cannot write {label} metadata to JSON file {infofn}')
+ return None
def _write_description(self, label, ie_result, descfn):
''' Write description and returns True = written, False = skip, None = error '''
sub_info['filepath'] = sub_filename
ret.append((sub_filename, sub_filename_final))
except (DownloadError, ExtractorError, IOError, OSError, ValueError) + network_exceptions as err:
+ msg = f'Unable to download video subtitles for {sub_lang!r}: {err}'
if self.params.get('ignoreerrors') is not True: # False or 'only_download'
- raise DownloadError(f'Unable to download video subtitles for {sub_lang!r}: {err}', err)
- self.report_warning(f'Unable to download video subtitles for {sub_lang!r}: {err}')
+ if not self.params.get('ignoreerrors'):
+ self.report_error(msg)
+ raise DownloadError(msg)
+ self.report_warning(msg)
return ret
def _write_thumbnails(self, label, info_dict, filename, thumb_filename_base=None):