import collections
import contextlib
-import copy
import datetime
import errno
import fileinput
format_field,
formatSeconds,
GeoRestrictedError,
+ get_domain,
HEADRequest,
int_or_none,
iri_to_uri,
PostProcessingError,
preferredencoding,
prepend_extension,
+ ReExtractInfo,
register_socks_protocols,
RejectedVideoReached,
+ remove_terminal_sequences,
render_table,
replace_extension,
SameFileError,
strftime_or_none,
subtitles_filename,
supports_terminal_sequences,
- ThrottledDownload,
+ timetuple_from_msec,
to_high_limit_path,
traverse_obj,
try_get,
from .postprocessor import (
get_postprocessor,
EmbedThumbnailPP,
+ FFmpegFixupDuplicateMoovPP,
FFmpegFixupDurationPP,
FFmpegFixupM3u8PP,
FFmpegFixupM4aPP,
_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
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.
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.
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',
self.cache = Cache(self)
windows_enable_vt_mode()
- # FIXME: This will break if we ever print color to stdout
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),
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
# Creating format selector here allows us to catch syntax errors before the extraction
self.format_selector = (
- None if self.params.get('format') is None
+ self.params.get('format') if self.params.get('format') in (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()
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
if self.params.get('cookiefile') is not None:
self.cookiejar.save(ignore_discard=True, ignore_expires=True)
- def trouble(self, message=None, tb=None):
+ def trouble(self, message=None, tb=None, is_error=True):
"""Determine action to take when a download problem appears.
Depending on if the downloader has been configured to ignore
download errors or not, this method may throw an exception or
not when errors are found, after printing the message.
- tb, if given, is additional traceback information.
+ @param tb If given, is additional traceback information
+ @param is_error Whether to raise error according to ignorerrors
"""
if message is not None:
self.to_stderr(message)
tb = ''.join(tb_data)
if tb:
self.to_stderr(tb)
+ if not is_error:
+ return
if not self.params.get('ignoreerrors'):
if sys.exc_info()[0] and hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]:
exc_info = sys.exc_info()[1].exc_info
class Styles(Enum):
HEADERS = 'yellow'
- EMPHASIS = 'blue'
+ EMPHASIS = 'light blue'
ID = 'green'
DELIM = 'blue'
ERROR = 'red'
WARNING = 'yellow'
+ SUPPRESS = 'light black'
- def __format_text(self, out, text, f, fallback=None, *, test_encoding=False):
- assert out in ('screen', 'err')
+ def _format_text(self, handle, allow_colors, text, f, fallback=None, *, test_encoding=False):
if test_encoding:
original_text = text
- handle = self._screen_file if out == 'screen' else self._err_file
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 self._allow_colors[out] else text if fallback is None else fallback
+ 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('screen', *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('err', *args, **kwargs)
+ return self._format_text(
+ self._err_file, self._allow_colors['err'], *args, **kwargs)
def report_warning(self, message, only_once=False):
'''
return
self.to_stderr(f'{self._format_err("WARNING:", self.Styles.WARNING)} {message}', only_once)
- def report_error(self, message, tb=None):
+ 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, *args, **kwargs):
'''
Do the same as trouble, but prefixes the message with 'ERROR:', colored
in red if stderr is a tty file.
'''
- self.trouble(f'{self._format_err("ERROR:", self.Styles.ERROR)} {message}', tb)
+ self.trouble(f'{self._format_err("ERROR:", self.Styles.ERROR)} {message}', *args, **kwargs)
def write_debug(self, message, only_once=False):
'''Log debug message or Print message to stderr'''
(?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))
def _dumpjson_default(obj):
if isinstance(obj, (set, LazyList)):
return list(obj)
- raise TypeError(f'Object of type {type(obj).__name__} is not JSON serializable')
+ return repr(obj)
def create_key(outer_mobj):
if not outer_mobj.group('has_key'):
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'
# 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_nonempty(fn_groups[0][:trim_file_name], sub_ext, ext, delim='.')
+ 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 as e:
- self.to_stderr('\r')
- self.report_warning(f'{e}; 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 (DownloadCancelled, 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
if 'entries' not in ie_result:
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:
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
+ _infojson_written = False
if not self.params.get('simulate') and self.params.get('allow_playlist_files', True):
ie_copy = {
'playlist': playlist,
}
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
def _check_formats(self, formats):
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)
+ 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)
'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')),
}
selector_1, selector_2 = map(_build_selector_function, selector.selector)
def selector_function(ctx):
- for pair in itertools.product(
- selector_1(copy.deepcopy(ctx)), selector_2(copy.deepcopy(ctx))):
+ for pair in itertools.product(selector_1(ctx), selector_2(ctx)):
yield _merge(pair)
elif selector.type == SINGLE: # atom
# 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']))
filters = [self._build_format_filter(f) for f in selector.filters]
def final_selector(ctx):
- ctx_copy = copy.deepcopy(ctx)
+ ctx_copy = dict(ctx)
for _filter in filters:
ctx_copy['formats'] = list(filter(_filter, ctx_copy['formats']))
return selector_function(ctx_copy)
t['url'] = sanitize_url(t['url'])
if self.params.get('check_formats') is True:
- info_dict['thumbnails'] = LazyList(check_thumbnails(thumbnails[::-1])).reverse()
+ info_dict['thumbnails'] = LazyList(check_thumbnails(thumbnails[::-1]), reverse=True)
else:
info_dict['thumbnails'] = thumbnails
if not self.params.get('allow_unplayable_formats'):
formats = [f for f in formats if not f.get('has_drm')]
+ if info_dict.get('is_live'):
+ get_from_start = bool(self.params.get('live_from_start'))
+ formats = [f for f in formats if bool(f.get('is_from_start')) == get_from_start]
+
if not formats:
self.raise_no_formats(info_dict)
# TODO Central sorting goes here
if self.params.get('check_formats') is True:
- formats = LazyList(self._check_formats(formats[::-1])).reverse()
+ 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
info_dict, _ = self.pre_process(info_dict)
+ # The pre-processors may have modified the formats
+ formats = info_dict.get('formats', [info_dict])
+
+ list_only = self.params.get('simulate') is None and (
+ self.params.get('list_thumbnails') or self.params.get('listformats') or self.params.get('listsubtitles'))
+ interactive_format_selection = not list_only and self.format_selector == '-'
if self.params.get('list_thumbnails'):
self.list_thumbnails(info_dict)
- if self.params.get('listformats'):
- if not info_dict.get('formats') and not info_dict.get('url'):
- self.to_screen('%s has no formats' % info_dict['id'])
- else:
- self.list_formats(info_dict)
if self.params.get('listsubtitles'):
if 'automatic_captions' in info_dict:
self.list_subtitles(
info_dict['id'], automatic_captions, 'automatic captions')
self.list_subtitles(info_dict['id'], subtitles, 'subtitles')
- list_only = self.params.get('simulate') is None and (
- self.params.get('list_thumbnails') or self.params.get('listformats') or self.params.get('listsubtitles'))
+ if self.params.get('listformats') or interactive_format_selection:
+ if not info_dict.get('formats') and not info_dict.get('url'):
+ self.to_screen('%s has no formats' % info_dict['id'])
+ else:
+ self.list_formats(info_dict)
if list_only:
# Without this printing, -F --print-json will not work
self.__forced_printings(info_dict, self.prepare_filename(info_dict), incomplete=True)
self.write_debug('Default format spec: %s' % req_format)
format_selector = self.build_format_selector(req_format)
- # While in format selection we may need to have an access to the original
- # format set in order to calculate some metrics or do some processing.
- # For now we need to be able to guess whether original formats provided
- # by extractor are incomplete or not (i.e. whether extractor provides only
- # video-only or audio-only formats) for proper formats selection for
- # extractors with such incomplete formats (see
- # https://github.com/ytdl-org/youtube-dl/pull/5556).
- # Since formats may be filtered during format selection and may not match
- # the original formats the results may be incorrect. Thus original formats
- # or pre-calculated metrics should be passed to format selection routines
- # as well.
- # We will pass a context object containing all necessary additional data
- # instead of just formats.
- # This fixes incorrect format selection issue (see
- # https://github.com/ytdl-org/youtube-dl/issues/10083).
- incomplete_formats = (
- # All formats are video-only or
- all(f.get('vcodec') != 'none' and f.get('acodec') == 'none' for f in formats)
- # all formats are audio-only
- or all(f.get('vcodec') == 'none' and f.get('acodec') != 'none' for f in formats))
-
- ctx = {
- 'formats': formats,
- 'incomplete_formats': incomplete_formats,
- }
+ while True:
+ if interactive_format_selection:
+ req_format = input(
+ self._format_screen('\nEnter format selector: ', self.Styles.EMPHASIS))
+ try:
+ format_selector = self.build_format_selector(req_format)
+ except SyntaxError as err:
+ self.report_error(err, tb=False, is_error=False)
+ continue
+
+ # While in format selection we may need to have an access to the original
+ # format set in order to calculate some metrics or do some processing.
+ # For now we need to be able to guess whether original formats provided
+ # by extractor are incomplete or not (i.e. whether extractor provides only
+ # video-only or audio-only formats) for proper formats selection for
+ # extractors with such incomplete formats (see
+ # https://github.com/ytdl-org/youtube-dl/pull/5556).
+ # Since formats may be filtered during format selection and may not match
+ # the original formats the results may be incorrect. Thus original formats
+ # or pre-calculated metrics should be passed to format selection routines
+ # as well.
+ # We will pass a context object containing all necessary additional data
+ # instead of just formats.
+ # This fixes incorrect format selection issue (see
+ # https://github.com/ytdl-org/youtube-dl/issues/10083).
+ incomplete_formats = (
+ # All formats are video-only or
+ all(f.get('vcodec') != 'none' and f.get('acodec') == 'none' for f in formats)
+ # all formats are audio-only
+ or all(f.get('vcodec') == 'none' and f.get('acodec') != 'none' for f in formats))
+
+ ctx = {
+ 'formats': formats,
+ 'incomplete_formats': incomplete_formats,
+ }
+
+ formats_to_download = list(format_selector(ctx))
+ if interactive_format_selection and not formats_to_download:
+ self.report_error('Requested format is not available', tb=False, is_error=False)
+ continue
+ break
- formats_to_download = list(format_selector(ctx))
if not formats_to_download:
if not self.params.get('ignore_no_formats_error'):
raise ExtractorError('Requested format is not available', expected=True,
urls = '", "'.join([f['url'] for f in info.get('requested_formats', [])] or [info['url']])
self.write_debug('Invoking downloader on "%s"' % urls)
- new_info = copy.deepcopy(self._copy_infodict(info))
+ # Note: Ideally info should be a deep-copied so that hooks cannot modify it.
+ # But it may contain objects that are not deep-copyable
+ new_info = 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') and not self.params.get('live_from_start'):
+ 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 compatibility, even though it was a private field
info_dict['__infojson_filename'] = infofn
elif _infojson_written is None:
return
self.report_error('Cannot write internet shortcut file because the "webpage_url" field is missing in the media information')
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(f'[info] Internet shortcut (.{link_type}) is already present')
return True
dl_filename = existing_file(full_filename, temp_filename)
info_dict['__real_download'] = False
+ downloaded = []
+ merger = FFmpegMergerPP(self)
+
+ fd = get_suitable_downloader(info_dict, self.params, to_stdout=temp_filename == '-')
if dl_filename is not None:
self.report_file_already_downloaded(dl_filename)
- elif get_suitable_downloader(info_dict, self.params, to_stdout=temp_filename == '-'):
+ elif fd:
+ for f in requested_formats if fd != FFmpegFD else []:
+ f['filepath'] = fname = prepend_extension(
+ correct_ext(temp_filename, info_dict['ext']),
+ 'f%s' % f['format_id'], info_dict['ext'])
+ downloaded.append(fname)
info_dict['url'] = '\n'.join(f['url'] for f in requested_formats)
success, real_download = self.dl(temp_filename, info_dict)
info_dict['__real_download'] = real_download
else:
- downloaded = []
- merger = FFmpegMergerPP(self)
if self.params.get('allow_unplayable_formats'):
self.report_warning(
'You have requested merging of multiple formats '
'The formats won\'t be merged.')
if temp_filename == '-':
- reason = ('using a downloader other than ffmpeg' if FFmpegFD.can_merge_formats(info_dict)
+ reason = ('using a downloader other than ffmpeg' if FFmpegFD.can_merge_formats(info_dict, self.params)
else 'but the formats are incompatible for simultaneous download' if merger.available
else 'but ffmpeg is not installed')
self.report_warning(
partial_success, real_download = self.dl(fname, new_info)
info_dict['__real_download'] = info_dict['__real_download'] or real_download
success = success and partial_success
- if merger.available and not self.params.get('allow_unplayable_formats'):
- info_dict['__postprocessors'].append(merger)
- info_dict['__files_to_merge'] = downloaded
- # Even if there were no downloads, it is being merged only now
- info_dict['__real_download'] = True
- else:
- for file in downloaded:
- files_to_move[file] = None
+
+ if downloaded and merger.available and not self.params.get('allow_unplayable_formats'):
+ info_dict['__postprocessors'].append(merger)
+ info_dict['__files_to_merge'] = downloaded
+ # Even if there were no downloads, it is being merged only now
+ info_dict['__real_download'] = True
+ else:
+ for file in downloaded:
+ files_to_move[file] = None
else:
# Just a single file
dl_filename = existing_file(full_filename, temp_filename)
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)
+
+ if info_dict.get('requested_formats') is None: # Not necessary if doing merger
+ ffmpeg_fixup(downloader == 'HlsFD',
+ 'Possible MPEG-TS in MP4 container or malformed AAC timestamps',
+ FFmpegFixupM3u8PP)
+ ffmpeg_fixup(info_dict.get('is_live') and downloader == 'DashSegmentsFD',
+ 'Possible duplicate MOOV atoms', FFmpegFixupDuplicateMoovPP)
+
+ ffmpeg_fixup(downloader == 'WebSocketFragmentFD', 'Malformed timestamps detected', FFmpegFixupTimestampPP)
+ ffmpeg_fixup(downloader == 'WebSocketFragmentFD', 'Malformed duration detected', FFmpegFixupDurationPP)
fixup()
try:
res = func(*args, **kwargs)
except UnavailableVideoError as e:
self.report_error(e)
- except DownloadCancelled as 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)
info = self.sanitize_info(json.loads('\n'.join(f)), self.params.get('clean_infojson', True))
try:
self.__download_wrapper(self.process_ie_result)(info, download=True)
- except (DownloadError, EntryNotInPlaylist, ThrottledDownload) as e:
- self.to_stderr('\r')
+ 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(f'The info failed to download: {e}; trying with URL {webpage_url}')
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 (
k.startswith('_') or k in remove_keys or v in empty_values)
else:
reject = lambda k, v: k in remove_keys
- filter_fn = lambda obj: (
- list(map(filter_fn, obj)) if isinstance(obj, (LazyList, list, tuple, set))
- else obj if not isinstance(obj, dict)
- else dict((k, filter_fn(v)) for k, v in obj.items() if not reject(k, v)))
+
+ def filter_fn(obj):
+ if isinstance(obj, dict):
+ return {k: filter_fn(v) for k, v in obj.items() if not reject(k, v)}
+ elif isinstance(obj, (list, tuple, set, LazyList)):
+ return list(map(filter_fn, obj))
+ elif obj is None or isinstance(obj, (str, int, float, bool)):
+ return obj
+ else:
+ return repr(obj)
+
return filter_fn(info_dict)
@staticmethod
@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 += ', '
formats = info_dict.get('formats', [info_dict])
new_format = self.params.get('listformats_table', True) is not False
if new_format:
- tbr_digits = number_of_digits(max(f.get('tbr') or 0 for f in formats))
- vbr_digits = number_of_digits(max(f.get('vbr') or 0 for f in formats))
- abr_digits = number_of_digits(max(f.get('abr') or 0 for f in formats))
delim = self._format_screen('\u2502', self.Styles.DELIM, '|', test_encoding=True)
table = [
[
self._format_screen(format_field(f, 'format_id'), self.Styles.ID),
format_field(f, 'ext'),
- self.format_resolution(f),
- format_field(f, 'fps', '%3d'),
+ 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', ' %s', func=format_bytes) + format_field(f, 'filesize_approx', '~%s', func=format_bytes),
- format_field(f, 'tbr', f'%{tbr_digits}dk'),
- shorten_protocol_name(f.get('protocol', '').replace("native", "n")),
+ 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', ''),
- format_field(f, 'vbr', f'%{vbr_digits}dk'),
- format_field(f, 'acodec', default='unknown').replace('none', ''),
- format_field(f, 'abr', f'%{abr_digits}dk'),
- format_field(f, 'asr', '%5dHz'),
+ 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'))),
- delim=', '),
+ 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 = self._list_format_headers(
- 'ID', 'EXT', 'RESOLUTION', 'FPS', 'HDR', delim, ' FILESIZE', ' TBR', 'PROTO',
- delim, 'VCODEC', ' VBR', 'ACODEC', ' ABR', ' ASR', 'MORE INFO')
+ 'ID', 'EXT', 'RESOLUTION', '\tFPS', 'HDR', delim, '\tFILESIZE', '\tTBR', 'PROTO',
+ delim, 'VCODEC', '\tVBR', 'ACODEC', '\tABR', '\tASR', 'MORE INFO')
else:
table = [
[
'[info] Available formats for %s:' % info_dict['id'])
self.to_stdout(render_table(
header_line, table,
- extraGap=(0 if new_format else 1),
- hideEmpty=new_format,
+ 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):
self.to_stdout(render_table(
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 get_encoding(stream):
ret = getattr(stream, 'encoding', 'missing (%s)' % type(stream).__name__)
if not supports_terminal_sequences(stream):
- ret += ' (No ANSI)'
+ 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' % (
write_debug = lambda msg: self._write_string(f'[debug] {msg}\n')
source = detect_variant()
- write_debug('yt-dlp version %s%s' % (__version__, '' if source == 'unknown' else f' ({source})'))
+ 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')
for name, klass in itertools.chain(plugin_extractors.items(), plugin_postprocessors.items())])
if self.params.get('compat_opts'):
write_debug('Compatibility options: %s' % ', '.join(self.params.get('compat_opts')))
- try:
- 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:
+
+ 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()
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'
+ has_websockets and 'websockets',
+ delim=', ') or 'none'
write_debug('Optional libraries: %s' % lib_str)
proxy_map = {}
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}')
self.write_debug(f'Skipping writing {label} thumbnail')
return ret
- for t in thumbnails[::-1]:
+ for idx, t in list(enumerate(thumbnails))[::-1]:
thumb_ext = (f'{t["id"]}.' if multiple else '') + determine_ext(t['url'], 'jpg')
thumb_display_id = f'{label} thumbnail {t["id"]}'
thumb_filename = replace_extension(filename, thumb_ext, info_dict.get('ext'))
ret.append((thumb_filename, thumb_filename_final))
t['filepath'] = thumb_filename
except network_exceptions as err:
+ thumbnails.pop(idx)
self.report_warning(f'Unable to download {thumb_display_id}: {err}')
if ret and not write_all:
break
projects / yt-dlp.git / blobdiff