import random
import unicodedata
+from enum import Enum
from string import ascii_letters
from .compat import (
DEFAULT_OUTTMPL,
determine_ext,
determine_protocol,
- DOT_DESKTOP_LINK_TEMPLATE,
- DOT_URL_LINK_TEMPLATE,
- DOT_WEBLOC_LINK_TEMPLATE,
+ DownloadCancelled,
DownloadError,
encode_compat_str,
encodeFilename,
format_field,
formatSeconds,
GeoRestrictedError,
+ get_domain,
HEADRequest,
int_or_none,
iri_to_uri,
ISO3166Utils,
+ join_nonempty,
LazyList,
+ LINK_TEMPLATES,
locked_file,
make_dir,
make_HTTPS_handler,
MaxDownloadsReached,
network_exceptions,
+ number_of_digits,
orderedSet,
OUTTMPL_TYPES,
PagedList,
parse_filesize,
PerRequestProxyHandler,
platform_name,
+ Popen,
PostProcessingError,
preferredencoding,
prepend_extension,
- process_communicate_or_kill,
+ ReExtractInfo,
register_socks_protocols,
RejectedVideoReached,
+ remove_terminal_sequences,
render_table,
replace_extension,
SameFileError,
strftime_or_none,
subtitles_filename,
supports_terminal_sequences,
- TERMINAL_SEQUENCES,
- ThrottledDownload,
+ timetuple_from_msec,
to_high_limit_path,
traverse_obj,
try_get,
YoutubeDLRedirectHandler,
)
from .cache import Cache
+from .minicurses import format_text
from .extractor import (
gen_extractor_classes,
get_info_extractor,
_PLUGIN_CLASSES as plugin_postprocessors
)
from .update import detect_variant
-from .version import __version__
+from .version import __version__, RELEASE_GIT_HEAD
if compat_os_name == 'nt':
import ctypes
simulate: Do not download the video files. If unset (or None),
simulate only if listsubtitles, listformats or list_thumbnails is used
format: Video format code. see "FORMAT SELECTION" for more details.
+ You can also pass a function. The function takes 'ctx' as
+ argument and returns the formats to download.
+ See "build_format_selector" for an implementation
allow_unplayable_formats: Allow unplayable formats to be extracted and downloaded.
ignore_no_formats_error: Ignore "No video formats" error. Usefull for
extracting metadata even if the video is not actually
available for download (experimental)
- format_sort: How to sort the video formats. see "Sorting Formats"
- for more details.
+ format_sort: A list of fields by which to sort the video formats.
+ See "Sorting Formats" for more details.
format_sort_force: Force the given format_sort. see "Sorting Formats"
for more details.
allow_multiple_video_streams: Allow multiple video streams to be merged
allow_multiple_audio_streams: Allow multiple audio streams to be merged
into a single file
check_formats Whether to test if the formats are downloadable.
- Can be True (check all), False (check none)
+ Can be True (check all), False (check none),
+ 'selected' (check selected formats),
or None (check only if requested by extractor)
paths: Dictionary of output paths. The allowed keys are 'home'
'temp' and the keys of OUTTMPL_TYPES (in utils.py)
file that is in the archive.
break_on_reject: Stop the download process when encountering a video that
has been filtered out.
+ break_per_url: Whether break_on_reject and break_on_existing
+ should act on each input URL as opposed to for the entire queue
cookiefile: File name where cookies should be read from and dumped to
cookiesfrombrowser: A tuple containing the name of the browser and the profile
name/path from where cookies are loaded.
- Eg: ('chrome', ) or (vivaldi, 'default')
+ Eg: ('chrome', ) or ('vivaldi', 'default')
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.
bidi_workaround: Work around buggy terminals without bidirectional text
support, using fridibi
debug_printtraffic:Print out sent and received HTTP traffic
- include_ads: Download ads as well
+ include_ads: Download ads as well (deprecated)
default_search: Prepend this string if an input url is not valid.
'auto' for elaborate guessing
encoding: Use this encoding instead of the system-specified.
extract_flat: Do not resolve URLs, return the immediate result.
Pass in 'in_playlist' to only show this behavior for
playlist items.
+ wait_for_video: If given, wait for scheduled streams to become available.
+ The value should be a tuple containing the range
+ (min_secs, max_secs) to wait between retries
postprocessors: A list of dictionaries, each with an entry
* key: The name of the postprocessor. See
yt_dlp/postprocessor/__init__.py for a list.
(with status "started" and "finished") if the processing is successful.
merge_output_format: Extension to use when merging formats.
final_ext: Expected final extension; used to detect when the file was
- already downloaded and converted. "merge_output_format" is
- replaced by this extension when given
+ already downloaded and converted
fixup: Automatically correct known faults of the file.
One of:
- "never": do nothing
compat_opts: Compatibility options. See "Differences in default behavior".
The following options do not work when used through the API:
filename, abort-on-error, multistreams, no-live-chat, format-sort
- no-clean-infojson, no-playlist-metafiles, no-keep-subs.
+ no-clean-infojson, no-playlist-metafiles, no-keep-subs, no-attach-info-json.
Refer __init__.py for their implementation
progress_template: Dictionary of templates for progress outputs.
Allowed keys are 'download', 'postprocess',
nopart, updatetime, buffersize, ratelimit, throttledratelimit, min_filesize,
max_filesize, test, noresizebuffer, retries, fragment_retries, continuedl,
noprogress, xattr_set_filesize, hls_use_mpegts, http_chunk_size,
- external_downloader_args.
+ external_downloader_args, concurrent_fragment_downloads.
The following options are used by the post processors:
prefer_ffmpeg: If False, use avconv instead of ffmpeg if both are available,
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).
- Set to 'no_verbose_header' to not ptint the header
+ Set to 'no_verbose_header' to not print the header
"""
if params is None:
params = {}
self.cache = Cache(self)
windows_enable_vt_mode()
- # FIXME: This will break if we ever print color to stdout
- self.params['no_color'] = self.params.get('no_color') or not supports_terminal_sequences(self._err_file)
+ 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),
+ }
if sys.version_info < (3, 6):
self.report_warning(
if self.params.get('allow_unplayable_formats'):
self.report_warning(
- f'You have asked for {self._color_text("unplayable formats", "blue")} to be listed/downloaded. '
+ f'You have asked for {self._format_err("UNPLAYABLE", self.Styles.EMPHASIS)} formats to be listed/downloaded. '
'This is a developer option intended for debugging. \n'
' If you experience any issues while using this option, '
- f'{self._color_text("DO NOT", "red")} open a bug report')
+ f'{self._format_err("DO NOT", self.Styles.ERROR)} open a bug report')
def check_deprecated(param, option, suggestion):
if self.params.get(param) is not None:
check_deprecated('usetitle', '--title', '-o "%(title)s-%(id)s.%(ext)s"')
check_deprecated('useid', '--id', '-o "%(id)s.%(ext)s"')
- for msg in self.params.get('warnings', []):
+ for msg in self.params.get('_warnings', []):
self.report_warning(msg)
+ for msg in self.params.get('_deprecation_warnings', []):
+ self.deprecation_warning(msg)
+
+ if 'list-formats' in self.params.get('compat_opts', []):
+ self.params['listformats_table'] = False
if 'overwrites' not in self.params and self.params.get('nooverwrites') is not None:
# nooverwrites was unnecessarily changed to overwrites
stdout=slave,
stderr=self._err_file)
try:
- self._output_process = subprocess.Popen(
- ['bidiv'] + width_args, **sp_kwargs
- )
+ self._output_process = Popen(['bidiv'] + width_args, **sp_kwargs)
except OSError:
- self._output_process = subprocess.Popen(
- ['fribidi', '-c', 'UTF-8'] + width_args, **sp_kwargs)
+ self._output_process = Popen(['fribidi', '-c', 'UTF-8'] + width_args, **sp_kwargs)
self._output_channel = os.fdopen(master, 'rb')
except OSError as ose:
if ose.errno == errno.ENOENT:
- self.report_warning('Could not find fribidi executable, ignoring --bidi-workaround . Make sure that fribidi is an executable file in one of the directories in your $PATH.')
+ self.report_warning(
+ 'Could not find fribidi executable, ignoring --bidi-workaround. '
+ 'Make sure that fribidi is an executable file in one of the directories in your $PATH.')
else:
raise
# Creating format selector here allows us to catch syntax errors before the extraction
self.format_selector = (
None if self.params.get('format') is None
+ else self.params['format'] if callable(self.params['format'])
else self.build_format_selector(self.params['format']))
self._setup_opener()
self.print_debug_header()
self.add_default_info_extractors()
+ hooks = {
+ 'post_hooks': self.add_post_hook,
+ 'progress_hooks': self.add_progress_hook,
+ 'postprocessor_hooks': self.add_postprocessor_hook,
+ }
+ for opt, fn in hooks.items():
+ for ph in self.params.get(opt, []):
+ fn(ph)
+
for pp_def_raw in self.params.get('postprocessors', []):
pp_def = dict(pp_def_raw)
when = pp_def.pop('when', 'post_process')
- pp_class = get_postprocessor(pp_def.pop('key'))
- pp = pp_class(self, **compat_kwargs(pp_def))
- self.add_post_processor(pp, when=when)
-
- for ph in self.params.get('post_hooks', []):
- self.add_post_hook(ph)
-
- for ph in self.params.get('progress_hooks', []):
- self.add_progress_hook(ph)
+ self.add_post_processor(
+ get_postprocessor(pp_def.pop('key'))(self, **compat_kwargs(pp_def)),
+ when=when)
register_socks_protocols()
"""Preload the archive, if any is specified"""
if fn is None:
return False
- self.write_debug('Loading archive file %r\n' % 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.report_warning(
'Long argument string detected. '
- 'Use -- to separate parameters and URLs, like this:\n%s\n' %
+ 'Use -- to separate parameters and URLs, like this:\n%s' %
args_to_str(correct_argv))
def add_info_extractor(self, ie):
def add_postprocessor_hook(self, ph):
"""Add the postprocessing progress hook"""
self._postprocessor_hooks.append(ph)
+ for pps in self._pps.values():
+ for pp in pps:
+ pp.add_progress_hook(ph)
def _bidi_workaround(self, message):
if not hasattr(self, '_output_channel'):
def to_console_title(self, message):
if not self.params.get('consoletitle', False):
return
+ message = remove_terminal_sequences(message)
if compat_os_name == 'nt':
if ctypes.windll.kernel32.GetConsoleWindow():
# c_wchar_p() might not be necessary if `message` is
self.to_stdout(
message, skip_eol, quiet=self.params.get('quiet', False))
- def _color_text(self, text, color):
- if self.params.get('no_color'):
- return text
- return f'{TERMINAL_SEQUENCES[color.upper()]}{text}{TERMINAL_SEQUENCES["RESET_STYLE"]}'
+ class Styles(Enum):
+ HEADERS = 'yellow'
+ EMPHASIS = 'light blue'
+ ID = 'green'
+ DELIM = 'blue'
+ ERROR = 'red'
+ WARNING = 'yellow'
+ SUPPRESS = 'light black'
+
+ def _format_text(self, handle, allow_colors, text, f, fallback=None, *, test_encoding=False):
+ if test_encoding:
+ original_text = text
+ encoding = self.params.get('encoding') or getattr(handle, 'encoding', 'ascii')
+ text = text.encode(encoding, 'ignore').decode(encoding)
+ if fallback is not None and text != original_text:
+ text = fallback
+ if isinstance(f, self.Styles):
+ f = f.value
+ return format_text(text, f) if allow_colors else text if fallback is None else fallback
+
+ def _format_screen(self, *args, **kwargs):
+ return self._format_text(
+ self._screen_file, 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)
def report_warning(self, message, only_once=False):
'''
else:
if self.params.get('no_warnings'):
return
- self.to_stderr(f'{self._color_text("WARNING:", "yellow")} {message}', only_once)
+ self.to_stderr(f'{self._format_err("WARNING:", self.Styles.WARNING)} {message}', only_once)
+
+ def deprecation_warning(self, message):
+ if self.params.get('logger') is not None:
+ self.params['logger'].warning('DeprecationWarning: {message}')
+ else:
+ self.to_stderr(f'{self._format_err("DeprecationWarning:", self.Styles.ERROR)} {message}', True)
def report_error(self, message, tb=None):
'''
Do the same as trouble, but prefixes the message with 'ERROR:', colored
in red if stderr is a tty file.
'''
- self.trouble(f'{self._color_text("ERROR:", "red")} {message}', tb)
+ self.trouble(f'{self._format_err("ERROR:", self.Styles.ERROR)} {message}', tb)
def write_debug(self, message, only_once=False):
'''Log debug message or Print message to stderr'''
outtmpl_dict = self.params.get('outtmpl', {})
if not isinstance(outtmpl_dict, dict):
outtmpl_dict = {'default': outtmpl_dict}
+ # Remove spaces in the default template
+ if self.params.get('restrictfilenames'):
+ sanitize = lambda x: x.replace(' - ', ' ').replace(' ', '-')
+ else:
+ sanitize = lambda x: x
outtmpl_dict.update({
- k: v for k, v in DEFAULT_OUTTMPL.items()
+ k: sanitize(v) for k, v in DEFAULT_OUTTMPL.items()
if outtmpl_dict.get(k) is None})
for key, val in outtmpl_dict.items():
if isinstance(val, bytes):
except ValueError as err:
return err
+ @staticmethod
+ def _copy_infodict(info_dict):
+ info_dict = dict(info_dict)
+ for key in ('__original_infodict', '__postprocessors'):
+ info_dict.pop(key, None)
+ return info_dict
+
def prepare_outtmpl(self, outtmpl, info_dict, sanitize=None):
""" Make the outtmpl and info_dict suitable for substitution: ydl.escape_outtmpl(outtmpl) % info_dict """
info_dict.setdefault('epoch', int(time.time())) # keep epoch consistent once set
- info_dict = dict(info_dict) # Do not sanitize so as not to consume LazyList
- for key in ('__original_infodict', '__postprocessors'):
- info_dict.pop(key, None)
+ info_dict = self._copy_infodict(info_dict)
info_dict['duration_string'] = ( # %(duration>%H-%M-%S)s is wrong if duration > 24hrs
formatSeconds(info_dict['duration'], '-' if sanitize else ':')
if info_dict.get('duration', None) is not None
# For fields playlist_index, playlist_autonumber and autonumber convert all occurrences
# of %(field)s to %(field)0Nd for backward compatibility
field_size_compat_map = {
- 'playlist_index': len(str(info_dict.get('_last_playlist_index') or '')),
- 'playlist_autonumber': len(str(info_dict.get('n_entries') or '')),
+ 'playlist_index': number_of_digits(info_dict.get('_last_playlist_index') or 0),
+ 'playlist_autonumber': number_of_digits(info_dict.get('n_entries') or 0),
'autonumber': self.params.get('autonumber_size') or 5,
}
(?P<fields>{field})
(?P<maths>(?:{math_op}{math_field})*)
(?:>(?P<strf_format>.+?))?
- (?P<alternate>(?<!\\),[^|)]+)?
+ (?P<alternate>(?<!\\),[^|&)]+)?
+ (?:&(?P<replacement>.*?))?
(?:\|(?P<default>.*?))?
$'''.format(field=FIELD_RE, math_op=MATH_OPERATORS_RE, math_field=MATH_FIELD_RE))
key = outer_mobj.group('key')
mobj = re.match(INTERNAL_FORMAT_RE, key)
initial_field = mobj.group('fields').split('.')[-1] if mobj else ''
- value, default = None, na
+ value, replacement, default = None, None, na
while mobj:
mobj = mobj.groupdict()
default = mobj['default'] if mobj['default'] is not None else default
value = get_value(mobj)
+ replacement = mobj['replacement']
if value is None and mobj['alternate']:
mobj = re.match(INTERNAL_FORMAT_RE, mobj['alternate'][1:])
else:
if fmt == 's' and value is not None and key in field_size_compat_map.keys():
fmt = '0{:d}d'.format(field_size_compat_map[key])
- value = default if value is None else value
+ value = default if value is None else value if replacement is None else replacement
+ flags = outer_mobj.group('conversion') or ''
str_fmt = f'{fmt[:-1]}s'
if fmt[-1] == 'l': # list
- delim = '\n' if '#' in (outer_mobj.group('conversion') or '') else ', '
+ delim = '\n' if '#' in flags else ', '
value, fmt = delim.join(variadic(value)), str_fmt
elif fmt[-1] == 'j': # json
- value, fmt = json.dumps(value, default=_dumpjson_default), str_fmt
+ value, fmt = json.dumps(value, default=_dumpjson_default, indent=4 if '#' in flags else None), str_fmt
elif fmt[-1] == 'q': # quoted
- value, fmt = compat_shlex_quote(str(value)), str_fmt
+ value = map(str, variadic(value) if '#' in flags else [value])
+ value, fmt = ' '.join(map(compat_shlex_quote, value)), str_fmt
elif fmt[-1] == 'B': # bytes
value = f'%{str_fmt}'.encode('utf-8') % str(value).encode('utf-8')
value, fmt = value.decode('utf-8', 'ignore'), 's'
elif fmt[-1] == 'U': # unicode normalized
- opts = outer_mobj.group('conversion') or ''
value, fmt = unicodedata.normalize(
# "+" = compatibility equivalence, "#" = NFD
- 'NF%s%s' % ('K' if '+' in opts else '', 'D' if '#' in opts else 'C'),
+ 'NF%s%s' % ('K' if '+' in flags else '', 'D' if '#' in flags else 'C'),
value), str_fmt
elif fmt[-1] == 'c':
if value:
# https://github.com/blackjack4494/youtube-dlc/issues/85
trim_file_name = self.params.get('trim_file_name', False)
if trim_file_name:
- fn_groups = filename.rsplit('.')
- ext = fn_groups[-1]
- sub_ext = ''
- if len(fn_groups) > 2:
- sub_ext = fn_groups[-2]
- filename = '.'.join(filter(None, [fn_groups[0][:trim_file_name], sub_ext, ext]))
+ no_ext, *ext = filename.rsplit('.', 2)
+ filename = join_nonempty(no_ext[:trim_file_name], *ext, delim='.')
return filename
except ValueError as err:
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("[%s] %s: has already been recorded in archive" % (
- ie_key, temp_id))
+ self.to_screen(f'[{ie_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)
else:
self.report_error(msg)
except ExtractorError as e: # An error we somewhat expected
self.report_error(compat_str(e), e.format_traceback())
- except ThrottledDownload:
- self.to_stderr('\r')
- self.report_warning('The download speed is below throttle limit. Re-extracting data')
+ except ReExtractInfo as e:
+ if e.expected:
+ self.to_screen(f'{e}; Re-extracting data')
+ else:
+ self.to_stderr('\r')
+ self.report_warning(f'{e}; Re-extracting data')
return wrapper(self, *args, **kwargs)
- except (MaxDownloadsReached, ExistingVideoReached, RejectedVideoReached, LazyList.IndexError):
+ except (DownloadCancelled, LazyList.IndexError, PagedList.IndexError):
raise
except Exception as e:
if self.params.get('ignoreerrors'):
raise
return wrapper
+ def _wait_for_video(self, ie_result):
+ if (not self.params.get('wait_for_video')
+ or ie_result.get('_type', 'video') != 'video'
+ or ie_result.get('formats') or ie_result.get('url')):
+ return
+
+ format_dur = lambda dur: '%02d:%02d:%02d' % timetuple_from_msec(dur * 1000)[:-1]
+ last_msg = ''
+
+ def progress(msg):
+ nonlocal last_msg
+ self.to_screen(msg + ' ' * (len(last_msg) - len(msg)) + '\r', skip_eol=True)
+ last_msg = msg
+
+ 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)
+ 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')
+ diff = min(max(diff or 0, min_wait or 0), max_wait or float('inf'))
+ self.to_screen(f'[wait] Waiting for {format_dur(diff)} - Press Ctrl+C to try now')
+
+ wait_till = time.time() + diff
+ try:
+ while True:
+ diff = wait_till - time.time()
+ if diff <= 0:
+ progress('')
+ raise ReExtractInfo('[wait] Wait period ended', expected=True)
+ progress(f'[wait] Remaining time until next attempt: {self._format_screen(format_dur(diff), self.Styles.EMPHASIS)}')
+ time.sleep(1)
+ except KeyboardInterrupt:
+ progress('')
+ raise ReExtractInfo('[wait] Interrupted by user', expected=True)
+ except BaseException as e:
+ if not isinstance(e, ReExtractInfo):
+ self.to_screen('')
+ raise
+
@__handle_extraction_exceptions
def __extract_info(self, url, ie, download, extra_info, process):
ie_result = ie.extract(url)
ie_result.setdefault('original_url', extra_info['original_url'])
self.add_default_extra_info(ie_result, ie, url)
if process:
+ self._wait_for_video(ie_result)
return self.process_ie_result(ie_result, download, extra_info)
else:
return ie_result
'webpage_url': url,
'original_url': url,
'webpage_url_basename': url_basename(url),
+ 'webpage_url_domain': get_domain(url),
})
if ie is not None:
self.add_extra_info(ie_result, {
info_copy['id'] = ie.get_temp_id(ie_result['url'])
self.add_default_extra_info(info_copy, ie, ie_result['url'])
self.add_extra_info(info_copy, extra_info)
+ info_copy, _ = self.pre_process(info_copy)
self.__forced_printings(info_copy, self.prepare_filename(info_copy), incomplete=True)
if self.params.get('force_write_download_archive', False):
self.record_download_archive(info_copy)
'extractor': ie_result['extractor'],
'webpage_url': ie_result['webpage_url'],
'webpage_url_basename': url_basename(ie_result['webpage_url']),
+ 'webpage_url_domain': get_domain(ie_result['webpage_url']),
'extractor_key': ie_result['extractor_key'],
})
return r
self.to_screen('[download] Downloading playlist: %s' % playlist)
if 'entries' not in ie_result:
- raise EntryNotInPlaylist()
+ raise EntryNotInPlaylist('There are no entries')
+
+ MissingEntry = object()
incomplete_entries = bool(ie_result.get('requested_entries'))
if incomplete_entries:
- def fill_missing_entries(entries, indexes):
- ret = [None] * max(*indexes)
- for i, entry in zip(indexes, entries):
+ def fill_missing_entries(entries, indices):
+ ret = [MissingEntry] * max(indices)
+ for i, entry in zip(indices, entries):
ret[i - 1] = entry
return ret
ie_result['entries'] = fill_missing_entries(ie_result['entries'], ie_result['requested_entries'])
def get_entry(i):
return ie_entries[i - 1]
else:
- if not isinstance(ie_entries, PagedList):
+ if not isinstance(ie_entries, (PagedList, LazyList)):
ie_entries = LazyList(ie_entries)
def get_entry(i):
entry = None
try:
entry = get_entry(i)
- if entry is None:
+ if entry is MissingEntry:
raise EntryNotInPlaylist()
except (IndexError, EntryNotInPlaylist):
if incomplete_entries:
- raise EntryNotInPlaylist()
+ raise EntryNotInPlaylist(f'Entry {i} cannot be found')
elif not playlistitems:
break
entries.append(entry)
if entry is not None]
n_entries = len(entries)
- if not playlistitems and (playliststart or playlistend):
+ if not playlistitems and (playliststart != 1 or playlistend):
playlistitems = list(range(playliststart, playliststart + n_entries))
ie_result['requested_entries'] = playlistitems
- if self.params.get('allow_playlist_files', True):
+ _infojson_written = False
+ if not self.params.get('simulate') and self.params.get('allow_playlist_files', True):
ie_copy = {
'playlist': playlist,
'playlist_id': ie_result.get('id'),
'playlist_uploader': ie_result.get('uploader'),
'playlist_uploader_id': ie_result.get('uploader_id'),
'playlist_index': 0,
+ 'n_entries': n_entries,
}
ie_copy.update(dict(ie_result))
- if self._write_info_json('playlist', ie_result,
- self.prepare_filename(ie_copy, 'pl_infojson')) is None:
+ _infojson_written = self._write_info_json(
+ 'playlist', ie_result, self.prepare_filename(ie_copy, 'pl_infojson'))
+ if _infojson_written is None:
return
if self._write_description('playlist', ie_result,
self.prepare_filename(ie_copy, 'pl_description')) is None:
'extractor': ie_result['extractor'],
'webpage_url': ie_result['webpage_url'],
'webpage_url_basename': url_basename(ie_result['webpage_url']),
+ 'webpage_url_domain': get_domain(ie_result['webpage_url']),
'extractor_key': ie_result['extractor_key'],
}
self.report_error(
'Skipping the remaining entries in playlist "%s" since %d items failed extraction' % (playlist, failures))
break
- # TODO: skip failed (empty) entries?
playlist_results.append(entry_result)
ie_result['entries'] = playlist_results
+
+ # Write the updated info to json
+ if _infojson_written and self._write_info_json(
+ 'updated playlist', ie_result,
+ self.prepare_filename(ie_copy, 'pl_infojson'), overwrite=True) is None:
+ return
self.to_screen('[download] Finished downloading playlist: %s' % playlist)
return ie_result
return op(actual_value, comparison_value)
return _filter
+ def _check_formats(self, formats):
+ for f in formats:
+ self.to_screen('[info] Testing format %s' % f['format_id'])
+ path = self.get_output_path('temp')
+ if not self._ensure_dir_exists(f'{path}/'):
+ continue
+ temp_file = tempfile.NamedTemporaryFile(suffix='.tmp', delete=False, dir=path or None)
+ temp_file.close()
+ try:
+ success, _ = self.dl(temp_file.name, f, test=True)
+ except (DownloadError, IOError, OSError, ValueError) + network_exceptions:
+ success = False
+ finally:
+ if os.path.exists(temp_file.name):
+ try:
+ os.remove(temp_file.name)
+ except OSError:
+ self.report_warning('Unable to delete temporary file "%s"' % temp_file.name)
+ if success:
+ yield f
+ else:
+ self.to_screen('[info] Unable to download format %s. Skipping...' % f['format_id'])
+
def _default_format_spec(self, info_dict, download=True):
def can_merge():
allow_multiple_streams = {'audio': self.params.get('allow_multiple_audio_streams', False),
'video': self.params.get('allow_multiple_video_streams', False)}
- check_formats = self.params.get('check_formats')
+ check_formats = self.params.get('check_formats') == 'selected'
def _parse_filter(tokens):
filter_parts = []
'format_id': '+'.join(filtered('format_id')),
'ext': output_ext,
'protocol': '+'.join(map(determine_protocol, formats_info)),
- 'language': '+'.join(orderedSet(filtered('language'))),
- 'format_note': '+'.join(orderedSet(filtered('format_note'))),
- 'filesize_approx': sum(filtered('filesize', 'filesize_approx')),
+ 'language': '+'.join(orderedSet(filtered('language'))) or None,
+ 'format_note': '+'.join(orderedSet(filtered('format_note'))) or None,
+ 'filesize_approx': sum(filtered('filesize', 'filesize_approx')) or None,
'tbr': sum(filtered('tbr', 'vbr', 'abr')),
}
'height': the_only_video.get('height'),
'resolution': the_only_video.get('resolution') or self.format_resolution(the_only_video),
'fps': the_only_video.get('fps'),
+ 'dynamic_range': the_only_video.get('dynamic_range'),
'vcodec': the_only_video.get('vcodec'),
'vbr': the_only_video.get('vbr'),
'stretched_ratio': the_only_video.get('stretched_ratio'),
if not check_formats:
yield from formats
return
- for f in formats:
- self.to_screen('[info] Testing format %s' % f['format_id'])
- temp_file = tempfile.NamedTemporaryFile(
- suffix='.tmp', delete=False,
- dir=self.get_output_path('temp') or None)
- temp_file.close()
- try:
- success, _ = self.dl(temp_file.name, f, test=True)
- except (DownloadError, IOError, OSError, ValueError) + network_exceptions:
- success = False
- finally:
- if os.path.exists(temp_file.name):
- try:
- os.remove(temp_file.name)
- except OSError:
- self.report_warning('Unable to delete temporary file "%s"' % temp_file.name)
- if success:
- yield f
- else:
- self.to_screen('[info] Unable to download format %s. Skipping...' % f['format_id'])
+ yield from self._check_formats(formats)
def _build_selector_function(selector):
if isinstance(selector, list): # ,
# TODO: Add allvideo, allaudio etc by generalizing the code with best/worst selector
if format_spec == 'all':
def selector_function(ctx):
- yield from _check_formats(ctx['formats'])
+ yield from _check_formats(ctx['formats'][::-1])
elif format_spec == 'mergeall':
def selector_function(ctx):
formats = list(_check_formats(ctx['formats']))
self.cookiejar.add_cookie_header(pr)
return pr.get_header('Cookie')
+ def _sort_thumbnails(self, thumbnails):
+ thumbnails.sort(key=lambda t: (
+ t.get('preference') if t.get('preference') is not None else -1,
+ t.get('width') if t.get('width') is not None else -1,
+ t.get('height') if t.get('height') is not None else -1,
+ t.get('id') if t.get('id') is not None else '',
+ t.get('url')))
+
def _sanitize_thumbnails(self, info_dict):
thumbnails = info_dict.get('thumbnails')
if thumbnails is None:
thumbnail = info_dict.get('thumbnail')
if thumbnail:
info_dict['thumbnails'] = thumbnails = [{'url': thumbnail}]
- if thumbnails:
- thumbnails.sort(key=lambda t: (
- t.get('preference') if t.get('preference') is not None else -1,
- t.get('width') if t.get('width') is not None else -1,
- t.get('height') if t.get('height') is not None else -1,
- t.get('id') if t.get('id') is not None else '',
- t.get('url')))
-
- def thumbnail_tester():
- if self.params.get('check_formats'):
- test_all = True
- to_screen = lambda msg: self.to_screen(f'[info] {msg}')
- else:
- test_all = False
- to_screen = self.write_debug
-
- def test_thumbnail(t):
- if not test_all and not t.get('_test_url'):
- return True
- to_screen('Testing thumbnail %s' % t['id'])
- try:
- self.urlopen(HEADRequest(t['url']))
- except network_exceptions as err:
- to_screen('Unable to connect to thumbnail %s URL "%s" - %s. Skipping...' % (
- t['id'], t['url'], error_to_compat_str(err)))
- return False
- return True
-
- return test_thumbnail
-
- for i, t in enumerate(thumbnails):
- if t.get('id') is None:
- t['id'] = '%d' % i
- if t.get('width') and t.get('height'):
- t['resolution'] = '%dx%d' % (t['width'], t['height'])
- t['url'] = sanitize_url(t['url'])
+ if not thumbnails:
+ return
- if self.params.get('check_formats') is not False:
- info_dict['thumbnails'] = LazyList(filter(thumbnail_tester(), thumbnails[::-1])).reverse()
- else:
- info_dict['thumbnails'] = thumbnails
+ def check_thumbnails(thumbnails):
+ for t in thumbnails:
+ self.to_screen(f'[info] Testing thumbnail {t["id"]}')
+ try:
+ self.urlopen(HEADRequest(t['url']))
+ except network_exceptions as err:
+ self.to_screen(f'[info] Unable to connect to thumbnail {t["id"]} URL {t["url"]!r} - {err}. Skipping...')
+ continue
+ yield t
+
+ self._sort_thumbnails(thumbnails)
+ for i, t in enumerate(thumbnails):
+ if t.get('id') is None:
+ t['id'] = '%d' % i
+ if t.get('width') and t.get('height'):
+ t['resolution'] = '%dx%d' % (t['width'], t['height'])
+ t['url'] = sanitize_url(t['url'])
+
+ if self.params.get('check_formats') is True:
+ info_dict['thumbnails'] = LazyList(check_thumbnails(thumbnails[::-1]), reverse=True)
+ else:
+ info_dict['thumbnails'] = thumbnails
def process_video_result(self, info_dict, download=True):
assert info_dict.get('_type', 'video') == 'video'
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'),
info_dict['requested_subtitles'] = self.process_subtitles(
info_dict['id'], subtitles, automatic_captions)
- # We now pick which formats have to be downloaded
if info_dict.get('formats') is None:
# There's only one format available
formats = [info_dict]
formats_dict[format_id].append(format)
# Make sure all formats have unique format_id
- common_exts = set(ext for exts in self._format_selection_exts.values() for ext in exts)
+ common_exts = set(itertools.chain(*self._format_selection_exts.values()))
for format_id, ambiguous_formats in formats_dict.items():
ambigious_id = len(ambiguous_formats) > 1
for i, format in enumerate(ambiguous_formats):
res=self.format_resolution(format),
note=format_field(format, 'format_note', ' (%s)'),
)
- # Automatically determine protocol if missing (useful for format
- # selection purposes)
if format.get('protocol') is None:
format['protocol'] = determine_protocol(format)
+ if format.get('resolution') is None:
+ format['resolution'] = self.format_resolution(format, default=None)
+ if format.get('dynamic_range') is None and format.get('vcodec') != 'none':
+ format['dynamic_range'] = 'SDR'
+ if (info_dict.get('duration') and format.get('tbr')
+ and not format.get('filesize') and not format.get('filesize_approx')):
+ format['filesize_approx'] = info_dict['duration'] * format['tbr'] * (1024 / 8)
+
# Add HTTP headers, so that external programs can use them from the
# json output
full_format_info = info_dict.copy()
# TODO Central sorting goes here
+ if self.params.get('check_formats') is True:
+ formats = LazyList(self._check_formats(formats[::-1]), reverse=True)
+
if not formats or formats[0] is not info_dict:
# only set the 'formats' fields if the original info_dict list them
# otherwise we end up with a circular reference, the first (and unique)
info_dict, _ = self.pre_process(info_dict)
+ # The pre-processors may have modified the formats
+ formats = info_dict.get('formats', [info_dict])
+
if self.params.get('list_thumbnails'):
self.list_thumbnails(info_dict)
if self.params.get('listformats'):
new_info['__original_infodict'] = info_dict
new_info.update(fmt)
self.process_info(new_info)
- # We update the info dict with the best quality format (backwards compatibility)
+ # We update the info dict with the selected best quality format (backwards compatibility)
if formats_to_download:
info_dict.update(formats_to_download[-1])
return info_dict
fd.add_progress_hook(ph)
urls = '", "'.join([f['url'] for f in info.get('requested_formats', [])] or [info['url']])
self.write_debug('Invoking downloader on "%s"' % urls)
- new_info = dict(info)
+
+ new_info = copy.deepcopy(self._copy_infodict(info))
if new_info.get('http_headers') is None:
new_info['http_headers'] = self._calc_headers(new_info)
return fd.download(name, new_info, subtitle)
if self._num_downloads >= int(max_downloads):
raise MaxDownloadsReached()
+ if info_dict.get('is_live'):
+ info_dict['title'] += ' ' + datetime.datetime.now().strftime('%Y-%m-%d %H:%M')
+
# TODO: backward compatibility, to be removed
info_dict['fulltitle'] = info_dict['title']
infofn = self.prepare_filename(info_dict, 'infojson')
_infojson_written = self._write_info_json('video', info_dict, infofn)
if _infojson_written:
+ info_dict['infojson_filename'] = infofn
+ # For backward compatability, even though it was a private field
info_dict['__infojson_filename'] = infofn
elif _infojson_written is None:
return
return
# Write internet shortcut files
- url_link = webloc_link = desktop_link = False
- if self.params.get('writelink', False):
- if sys.platform == "darwin": # macOS.
- webloc_link = True
- elif sys.platform.startswith("linux"):
- desktop_link = True
- else: # if sys.platform in ['win32', 'cygwin']:
- url_link = True
- if self.params.get('writeurllink', False):
- url_link = True
- if self.params.get('writewebloclink', False):
- webloc_link = True
- if self.params.get('writedesktoplink', False):
- desktop_link = True
-
- if url_link or webloc_link or desktop_link:
+ 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
- ascii_url = iri_to_uri(info_dict['webpage_url'])
-
- def _write_link_file(extension, template, newline, embed_filename):
- linkfn = replace_extension(full_filename, extension, info_dict.get('ext'))
+ return False
+ 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
if self.params.get('overwrites', True) and os.path.exists(encodeFilename(linkfn)):
- self.to_screen('[info] Internet shortcut is already present')
- else:
- try:
- self.to_screen('[info] Writing internet shortcut to: ' + linkfn)
- with io.open(encodeFilename(to_high_limit_path(linkfn)), 'w', encoding='utf-8', newline=newline) as linkfile:
- template_vars = {'url': ascii_url}
- if embed_filename:
- template_vars['filename'] = linkfn[:-(len(extension) + 1)]
- linkfile.write(template % template_vars)
- except (OSError, IOError):
- self.report_error('Cannot write internet shortcut ' + linkfn)
- return False
+ self.to_screen(f'[info] Internet shortcut (.{link_type}) is already present')
+ return True
+ try:
+ 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'])}
+ if link_type == 'desktop':
+ template_vars['filename'] = linkfn[:-(len(link_type) + 1)]
+ linkfile.write(LINK_TEMPLATES[link_type] % template_vars)
+ except (OSError, IOError):
+ self.report_error(f'Cannot write internet shortcut {linkfn}')
+ return False
return True
- if url_link:
- if not _write_link_file('url', DOT_URL_LINK_TEMPLATE, '\r\n', embed_filename=False):
- return
- if webloc_link:
- if not _write_link_file('webloc', DOT_WEBLOC_LINK_TEMPLATE, '\n', embed_filename=False):
- return
- if desktop_link:
- if not _write_link_file('desktop', DOT_DESKTOP_LINK_TEMPLATE, '\n', embed_filename=True):
- return
+ write_links = {
+ 'url': self.params.get('writeurllink'),
+ 'webloc': self.params.get('writewebloclink'),
+ 'desktop': self.params.get('writedesktoplink'),
+ }
+ if self.params.get('writelink'):
+ link_type = ('webloc' if sys.platform == 'darwin'
+ else 'desktop' if sys.platform.startswith('linux')
+ else 'url')
+ write_links[link_type] = True
+
+ if any(should_write and not _write_link_file(link_type)
+ for link_type, should_write in write_links.items()):
+ return
try:
info_dict, files_to_move = self.pre_process(info_dict, 'before_dl', files_to_move)
downloader = get_suitable_downloader(info_dict, self.params) if 'protocol' in info_dict else None
downloader = downloader.__name__ if downloader else None
ffmpeg_fixup(info_dict.get('requested_formats') is None and downloader == 'HlsFD',
- 'malformed AAC bitstream detected', FFmpegFixupM3u8PP)
- ffmpeg_fixup(downloader == 'WebSocketFragmentFD', 'malformed timestamps detected', FFmpegFixupTimestampPP)
- ffmpeg_fixup(downloader == 'WebSocketFragmentFD', 'malformed duration detected', FFmpegFixupDurationPP)
+ 'Possible MPEG-TS in MP4 container or malformed AAC timestamps',
+ FFmpegFixupM3u8PP)
+ ffmpeg_fixup(downloader == 'WebSocketFragmentFD', 'Malformed timestamps detected', FFmpegFixupTimestampPP)
+ ffmpeg_fixup(downloader == 'WebSocketFragmentFD', 'Malformed duration detected', FFmpegFixupDurationPP)
fixup()
try:
if max_downloads is not None and self._num_downloads >= int(max_downloads):
raise MaxDownloadsReached()
+ def __download_wrapper(self, func):
+ @functools.wraps(func)
+ def wrapper(*args, **kwargs):
+ try:
+ res = func(*args, **kwargs)
+ except UnavailableVideoError as e:
+ self.report_error(e)
+ except MaxDownloadsReached as e:
+ self.to_screen(f'[info] {e}')
+ raise
+ except DownloadCancelled as e:
+ self.to_screen(f'[info] {e}')
+ if not self.params.get('break_per_url'):
+ raise
+ else:
+ if self.params.get('dump_single_json', False):
+ self.post_extract(res)
+ self.to_stdout(json.dumps(self.sanitize_info(res)))
+ return wrapper
+
def download(self, url_list):
"""Download a given list of URLs."""
+ url_list = variadic(url_list) # Passing a single URL is a common mistake
outtmpl = self.outtmpl_dict['default']
if (len(url_list) > 1
and outtmpl != '-'
raise SameFileError(outtmpl)
for url in url_list:
- try:
- # It also downloads the videos
- res = self.extract_info(
- url, force_generic_extractor=self.params.get('force_generic_extractor', False))
- except UnavailableVideoError:
- self.report_error('unable to download video')
- except MaxDownloadsReached:
- self.to_screen('[info] Maximum number of downloads reached')
- raise
- except ExistingVideoReached:
- self.to_screen('[info] Encountered a video that is already in the archive, stopping due to --break-on-existing')
- raise
- except RejectedVideoReached:
- self.to_screen('[info] Encountered a video that did not match filter, stopping due to --break-on-reject')
- raise
- else:
- if self.params.get('dump_single_json', False):
- self.post_extract(res)
- self.to_stdout(json.dumps(self.sanitize_info(res)))
+ self.__download_wrapper(self.extract_info)(
+ url, force_generic_extractor=self.params.get('force_generic_extractor', False))
return self._download_retcode
# FileInput doesn't have a read method, we can't call json.load
info = self.sanitize_info(json.loads('\n'.join(f)), self.params.get('clean_infojson', True))
try:
- self.process_ie_result(info, download=True)
- except (DownloadError, EntryNotInPlaylist, ThrottledDownload):
+ self.__download_wrapper(self.process_ie_result)(info, download=True)
+ except (DownloadError, EntryNotInPlaylist, ReExtractInfo) as e:
+ if not isinstance(e, EntryNotInPlaylist):
+ self.to_stderr('\r')
webpage_url = info.get('webpage_url')
if webpage_url is not None:
- self.report_warning('The info failed to download, trying with "%s"' % webpage_url)
+ self.report_warning(f'The info failed to download: {e}; trying with URL {webpage_url}')
return self.download([webpage_url])
else:
raise
return info_dict
info_dict.setdefault('epoch', int(time.time()))
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
+ keep_keys = ['_type'] # Always keep this to facilitate load-info-json
if remove_private_keys:
remove_keys |= {
- 'requested_formats', 'requested_subtitles', 'requested_entries',
- 'filepath', 'entries', 'original_url', 'playlist_autonumber',
+ 'requested_formats', 'requested_subtitles', 'requested_entries', 'entries',
+ 'filepath', 'infojson_filename', 'original_url', 'playlist_autonumber',
}
empty_values = (None, {}, [], set(), tuple())
reject = lambda k, v: k not in keep_keys and (
@staticmethod
def format_resolution(format, default='unknown'):
- is_images = format.get('vcodec') == 'none' and format.get('acodec') == 'none'
if format.get('vcodec') == 'none' and format.get('acodec') != 'none':
return 'audio only'
if format.get('resolution') is not None:
return format['resolution']
if format.get('width') and format.get('height'):
- res = '%dx%d' % (format['width'], format['height'])
+ return '%dx%d' % (format['width'], format['height'])
elif format.get('height'):
- res = '%sp' % format['height']
+ return '%sp' % format['height']
elif format.get('width'):
- res = '%dx?' % format['width']
- elif is_images:
- return 'images'
- else:
- return default
- return f'{res} images' if is_images else res
+ return '%dx?' % format['width']
+ return default
def _format_note(self, fdict):
res = ''
if fdict.get('ext') in ['f4f', 'f4m']:
- res += '(unsupported) '
+ res += '(unsupported)'
if fdict.get('language'):
if res:
res += ' '
- res += '[%s] ' % fdict['language']
+ res += '[%s]' % fdict['language']
if fdict.get('format_note') is not None:
- res += fdict['format_note'] + ' '
+ if res:
+ res += ' '
+ res += fdict['format_note']
if fdict.get('tbr') is not None:
- res += '%4dk ' % fdict['tbr']
+ if res:
+ res += ', '
+ res += '%4dk' % fdict['tbr']
if fdict.get('container') is not None:
if res:
res += ', '
res += '~' + format_bytes(fdict['filesize_approx'])
return res
+ def _list_format_headers(self, *headers):
+ if self.params.get('listformats_table', True) is not False:
+ return [self._format_screen(header, self.Styles.HEADERS) for header in headers]
+ return headers
+
def list_formats(self, info_dict):
formats = info_dict.get('formats', [info_dict])
- new_format = (
- 'list-formats' not in self.params.get('compat_opts', [])
- and self.params.get('listformats_table', True) is not False)
+ new_format = self.params.get('listformats_table', True) is not False
if new_format:
+ delim = self._format_screen('\u2502', self.Styles.DELIM, '|', test_encoding=True)
table = [
[
- format_field(f, 'format_id'),
+ self._format_screen(format_field(f, 'format_id'), self.Styles.ID),
format_field(f, 'ext'),
- self.format_resolution(f),
- format_field(f, 'fps', '%d'),
- '|',
- format_field(f, 'filesize', ' %s', func=format_bytes) + format_field(f, 'filesize_approx', '~%s', func=format_bytes),
- format_field(f, 'tbr', '%4dk'),
- shorten_protocol_name(f.get('protocol', '').replace("native", "n")),
- '|',
- format_field(f, 'vcodec', default='unknown').replace('none', ''),
- format_field(f, 'vbr', '%4dk'),
- format_field(f, 'acodec', default='unknown').replace('none', ''),
- format_field(f, 'abr', '%3dk'),
- format_field(f, 'asr', '%5dHz'),
- ', '.join(filter(None, (
- 'UNSUPPORTED' if f.get('ext') in ('f4f', 'f4m') else '',
+ format_field(f, func=self.format_resolution, ignore=('audio only', 'images')),
+ format_field(f, 'fps', '\t%d'),
+ format_field(f, 'dynamic_range', '%s', ignore=(None, 'SDR')).replace('HDR', ''),
+ delim,
+ format_field(f, 'filesize', ' \t%s', func=format_bytes) + format_field(f, 'filesize_approx', '~\t%s', func=format_bytes),
+ format_field(f, 'tbr', '\t%dk'),
+ shorten_protocol_name(f.get('protocol', '').replace('native', 'n')),
+ delim,
+ format_field(f, 'vcodec', default='unknown').replace(
+ 'none',
+ 'images' if f.get('acodec') == 'none'
+ else self._format_screen('audio only', self.Styles.SUPPRESS)),
+ format_field(f, 'vbr', '\t%dk'),
+ format_field(f, 'acodec', default='unknown').replace(
+ 'none',
+ '' if f.get('vcodec') == 'none'
+ else self._format_screen('video only', self.Styles.SUPPRESS)),
+ format_field(f, 'abr', '\t%dk'),
+ format_field(f, 'asr', '\t%dHz'),
+ join_nonempty(
+ self._format_screen('UNSUPPORTED', 'light red') if f.get('ext') in ('f4f', 'f4m') else None,
format_field(f, 'language', '[%s]'),
- format_field(f, 'format_note'),
- format_field(f, 'container', ignore=(None, f.get('ext'))),
- ))),
+ join_nonempty(
+ format_field(f, 'format_note'),
+ format_field(f, 'container', ignore=(None, f.get('ext'))),
+ delim=', '),
+ delim=' '),
] for f in formats if f.get('preference') is None or f['preference'] >= -1000]
- header_line = ['ID', 'EXT', 'RESOLUTION', 'FPS', '|', ' FILESIZE', ' TBR', 'PROTO',
- '|', 'VCODEC', ' VBR', 'ACODEC', ' ABR', ' ASR', 'MORE INFO']
+ header_line = self._list_format_headers(
+ 'ID', 'EXT', 'RESOLUTION', '\tFPS', 'HDR', delim, '\tFILESIZE', '\tTBR', 'PROTO',
+ delim, 'VCODEC', '\tVBR', 'ACODEC', '\tABR', '\tASR', 'MORE INFO')
else:
table = [
[
self.to_screen(
'[info] Available formats for %s:' % info_dict['id'])
self.to_stdout(render_table(
- header_line, table, delim=new_format, extraGap=(0 if new_format else 1), hideEmpty=new_format))
+ header_line, table,
+ extra_gap=(0 if new_format else 1),
+ hide_empty=new_format,
+ delim=new_format and self._format_screen('\u2500', self.Styles.DELIM, '-', test_encoding=True)))
def list_thumbnails(self, info_dict):
thumbnails = list(info_dict.get('thumbnails'))
self.to_screen(
'[info] Thumbnails for %s:' % info_dict['id'])
self.to_stdout(render_table(
- ['ID', 'width', 'height', 'URL'],
+ self._list_format_headers('ID', 'Width', 'Height', 'URL'),
[[t['id'], t.get('width', 'unknown'), t.get('height', 'unknown'), t['url']] for t in thumbnails]))
def list_subtitles(self, video_id, subtitles, name='subtitles'):
return [lang, ', '.join(names), ', '.join(exts)]
self.to_stdout(render_table(
- ['Language', 'Name', 'Formats'],
+ self._list_format_headers('Language', 'Name', 'Formats'),
[_row(lang, formats) for lang, formats in subtitles.items()],
- hideEmpty=True))
+ hide_empty=True))
def urlopen(self, req):
""" Start an HTTP download """
def print_debug_header(self):
if not self.params.get('verbose'):
return
- get_encoding = lambda stream: getattr(stream, 'encoding', 'missing (%s)' % type(stream).__name__)
- encoding_str = (
- '[debug] Encodings: locale %s, fs %s, stdout %s, stderr %s, pref %s\n' % (
- locale.getpreferredencoding(),
- sys.getfilesystemencoding(),
- get_encoding(self._screen_file), get_encoding(self._err_file),
- self.get_encoding()))
+
+ def get_encoding(stream):
+ ret = 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)'
+ return ret
+
+ 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),
+ self.get_encoding())
logger = self.params.get('logger')
if logger:
write_debug = lambda msg: logger.debug(f'[debug] {msg}')
write_debug(encoding_str)
else:
- write_debug = lambda msg: self._write_string(f'[debug] {msg}')
- write_string(encoding_str, encoding=None)
+ write_string(f'[debug] {encoding_str}\n', encoding=None)
+ write_debug = lambda msg: self._write_string(f'[debug] {msg}\n')
source = detect_variant()
- write_debug('yt-dlp version %s%s\n' % (__version__, '' if source == 'unknown' else f' ({source})'))
- if _LAZY_LOADER:
- write_debug('Lazy loading extractors enabled\n')
+ write_debug(join_nonempty(
+ 'yt-dlp version', __version__,
+ f'[{RELEASE_GIT_HEAD}]' if RELEASE_GIT_HEAD else '',
+ '' if source == 'unknown' else f'({source})',
+ delim=' '))
+ if not _LAZY_LOADER:
+ if os.environ.get('YTDLP_NO_LAZY_EXTRACTORS'):
+ write_debug('Lazy loading extractors is forcibly disabled')
+ else:
+ write_debug('Lazy loading extractors is disabled')
if plugin_extractors or plugin_postprocessors:
- write_debug('Plugins: %s\n' % [
+ write_debug('Plugins: %s' % [
'%s%s' % (klass.__name__, '' if klass.__name__ == name else f' as {name}')
for name, klass in itertools.chain(plugin_extractors.items(), plugin_postprocessors.items())])
if self.params.get('compat_opts'):
- write_debug('Compatibility options: %s\n' % ', '.join(self.params.get('compat_opts')))
- try:
- sp = subprocess.Popen(
- ['git', 'rev-parse', '--short', 'HEAD'],
- stdout=subprocess.PIPE, stderr=subprocess.PIPE,
- cwd=os.path.dirname(os.path.abspath(__file__)))
- out, err = process_communicate_or_kill(sp)
- out = out.decode().strip()
- if re.match('[0-9a-f]+', out):
- write_debug('Git HEAD: %s\n' % out)
- except Exception:
+ write_debug('Compatibility options: %s' % ', '.join(self.params.get('compat_opts')))
+
+ if source == 'source':
try:
- sys.exc_clear()
+ sp = Popen(
+ ['git', 'rev-parse', '--short', 'HEAD'],
+ stdout=subprocess.PIPE, stderr=subprocess.PIPE,
+ cwd=os.path.dirname(os.path.abspath(__file__)))
+ out, err = sp.communicate_or_kill()
+ out = out.decode().strip()
+ if re.match('[0-9a-f]+', out):
+ write_debug('Git HEAD: %s' % out)
except Exception:
- pass
+ try:
+ sys.exc_clear()
+ except Exception:
+ pass
def python_implementation():
impl_name = platform.python_implementation()
return impl_name + ' version %d.%d.%d' % sys.pypy_version_info[:3]
return impl_name
- write_debug('Python version %s (%s %s) - %s\n' % (
+ write_debug('Python version %s (%s %s) - %s' % (
platform.python_version(),
python_implementation(),
platform.architecture()[0],
platform_name()))
- exe_versions = FFmpegPostProcessor.get_versions(self)
+ exe_versions, ffmpeg_features = FFmpegPostProcessor.get_versions_and_features(self)
+ ffmpeg_features = {key for key, val in ffmpeg_features.items() if val}
+ if ffmpeg_features:
+ exe_versions['ffmpeg'] += ' (%s)' % ','.join(ffmpeg_features)
+
exe_versions['rtmpdump'] = rtmpdump_version()
exe_versions['phantomjs'] = PhantomJSwrapper._version()
exe_str = ', '.join(
f'{exe} {v}' for exe, v in sorted(exe_versions.items()) if v
) or 'none'
- write_debug('exe versions: %s\n' % exe_str)
+ write_debug('exe versions: %s' % exe_str)
from .downloader.websocket import has_websockets
from .postprocessor.embedthumbnail import has_mutagen
from .cookies import SQLITE_AVAILABLE, KEYRING_AVAILABLE
- lib_str = ', '.join(sorted(filter(None, (
+ lib_str = join_nonempty(
compat_pycrypto_AES and compat_pycrypto_AES.__name__.split('.')[0],
- has_websockets and 'websockets',
+ KEYRING_AVAILABLE and 'keyring',
has_mutagen and 'mutagen',
SQLITE_AVAILABLE and 'sqlite',
- KEYRING_AVAILABLE and 'keyring',
- )))) or 'none'
- write_debug('Optional libraries: %s\n' % lib_str)
- write_debug('ANSI escape support: stdout = %s, stderr = %s\n' % (
- supports_terminal_sequences(self._screen_file),
- supports_terminal_sequences(self._err_file)))
+ has_websockets and 'websockets',
+ delim=', ') or 'none'
+ write_debug('Optional libraries: %s' % lib_str)
proxy_map = {}
for handler in self._opener.handlers:
if hasattr(handler, 'proxies'):
proxy_map.update(handler.proxies)
- write_debug('Proxy map: ' + compat_str(proxy_map) + '\n')
+ write_debug(f'Proxy map: {proxy_map}')
- if self.params.get('call_home', False):
+ # Not implemented
+ if False and self.params.get('call_home'):
ipaddr = self.urlopen('https://yt-dl.org/ip').read().decode('utf-8')
- write_debug('Public IP address: %s\n' % ipaddr)
- return
+ write_debug('Public IP address: %s' % ipaddr)
latest_version = self.urlopen(
'https://yt-dl.org/latest/version').read().decode('utf-8')
if version_tuple(latest_version) > version_tuple(__version__):
def _setup_opener(self):
timeout_val = self.params.get('socket_timeout')
- self._socket_timeout = 600 if timeout_val is None else float(timeout_val)
+ self._socket_timeout = 20 if timeout_val is None else float(timeout_val)
opts_cookiesfrombrowser = self.params.get('cookiesfrombrowser')
opts_cookiefile = self.params.get('cookiefile')
encoding = preferredencoding()
return encoding
- def _write_info_json(self, label, ie_result, infofn):
+ def _write_info_json(self, label, ie_result, infofn, overwrite=None):
''' Write infojson and returns True = written, False = skip, None = error '''
+ if overwrite is None:
+ overwrite = self.params.get('overwrites', True)
if not self.params.get('writeinfojson'):
return False
elif not infofn:
return False
elif not self._ensure_dir_exists(infofn):
return None
- elif not self.params.get('overwrites', True) and os.path.exists(infofn):
+ 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}')
for t in thumbnails[::-1]:
thumb_ext = (f'{t["id"]}.' if multiple else '') + determine_ext(t['url'], 'jpg')
- thumb_display_id = f'{label} thumbnail' + (f' {t["id"]}' if multiple else '')
+ thumb_display_id = f'{label} thumbnail {t["id"]}'
thumb_filename = replace_extension(filename, thumb_ext, info_dict.get('ext'))
thumb_filename_final = replace_extension(thumb_filename_base, thumb_ext, info_dict.get('ext'))
if not self.params.get('overwrites', True) and os.path.exists(thumb_filename):
ret.append((thumb_filename, thumb_filename_final))
t['filepath'] = thumb_filename
- self.to_screen(f'[info] {thumb_display_id.title()} is already present')
+ self.to_screen('[info] %s is already present' % (
+ thumb_display_id if multiple else f'{label} thumbnail').capitalize())
else:
self.to_screen(f'[info] Downloading {thumb_display_id} ...')
try:
projects / yt-dlp.git / blobdiff