import shutil
import subprocess
import sys
+import tempfile
import time
import tokenize
import traceback
STR_FORMAT_RE,
formatSeconds,
GeoRestrictedError,
+ HEADRequest,
int_or_none,
iri_to_uri,
ISO3166Utils,
preferredencoding,
prepend_extension,
process_communicate_or_kill,
- random_uuidv4,
register_socks_protocols,
RejectedVideoReached,
render_table,
str_or_none,
strftime_or_none,
subtitles_filename,
+ ThrottledDownload,
to_high_limit_path,
traverse_obj,
+ try_get,
UnavailableVideoError,
url_basename,
version_tuple,
)
from .downloader.rtmp import rtmpdump_version
from .postprocessor import (
+ get_postprocessor,
+ FFmpegFixupDurationPP,
FFmpegFixupM3u8PP,
FFmpegFixupM4aPP,
FFmpegFixupStretchedPP,
+ FFmpegFixupTimestampPP,
FFmpegMergerPP,
FFmpegPostProcessor,
- # FFmpegSubtitlesConvertorPP,
- get_postprocessor,
MoveFilesAfterDownloadPP,
)
from .version import __version__
into a single file
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)
+ 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)
outtmpl: Dictionary of templates for output names. Allowed keys
if True, otherwise use ffmpeg/avconv if False, otherwise
use downloader suggested by extractor if None.
compat_opts: Compatibility options. See "Differences in default behavior".
- Note that only format-sort, format-spec, no-live-chat,
- no-attach-info-json, playlist-index, list-formats,
- no-direct-merge, no-youtube-channel-redirect,
- and no-youtube-unavailable-videos works when used via the API
+ The following options do not work when used through the API:
+ filename, abort-on-error, multistreams, no-live-chat,
+ no-playlist-metafiles. Refer __init__.py for their implementation
The following parameters are not used by YoutubeDL itself, they are used by
the downloader (see yt_dlp/downloader/common.py):
- nopart, updatetime, buffersize, ratelimit, min_filesize, max_filesize, test,
- noresizebuffer, retries, continuedl, noprogress, consoletitle,
- xattr_set_filesize, external_downloader_args, hls_use_mpegts,
- http_chunk_size.
+ nopart, updatetime, buffersize, ratelimit, throttledratelimit, min_filesize,
+ max_filesize, test, noresizebuffer, retries, continuedl, noprogress, consoletitle,
+ xattr_set_filesize, external_downloader_args, hls_use_mpegts, http_chunk_size.
The following options are used by the post processors:
prefer_ffmpeg: If False, use avconv instead of ffmpeg if both are available,
dynamic_mpd: Whether to process dynamic DASH manifests (default: True)
hls_split_discontinuity: Split HLS playlists to different formats at
discontinuities such as ad breaks (default: False)
- youtube_include_dash_manifest: If True (default), DASH manifests and related
+ extractor_args: A dictionary of arguments to be passed to the extractors.
+ See "EXTRACTOR ARGUMENTS" for details.
+ Eg: {'youtube': {'skip': ['dash', 'hls']}}
+ youtube_include_dash_manifest: Deprecated - Use extractor_args instead.
+ If True (default), DASH manifests and related
data will be downloaded and processed by extractor.
You can reduce network I/O by disabling it if you don't
care about DASH. (only for youtube)
- youtube_include_hls_manifest: If True (default), HLS manifests and related
+ youtube_include_hls_manifest: Deprecated - Use extractor_args instead.
+ If True (default), HLS manifests and related
data will be downloaded and processed by extractor.
You can reduce network I/O by disabling it if you don't
care about HLS. (only for youtube)
params = None
_ies = []
_pps = {'pre_process': [], 'before_dl': [], 'after_move': [], 'post_process': []}
- __prepare_filename_warned = False
+ _reported_warnings = set()
_first_webpage_request = True
_download_retcode = None
_num_downloads = None
self._ies = []
self._ies_instances = {}
self._pps = {'pre_process': [], 'before_dl': [], 'after_move': [], 'post_process': []}
- self.__prepare_filename_warned = False
+ self._reported_warnings = set()
self._first_webpage_request = True
self._post_hooks = []
self._progress_hooks = []
self.outtmpl_dict = self.parse_outtmpl()
+ # 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.build_format_selector(self.params['format']))
+
self._setup_opener()
"""Preload the archive, if any is specified"""
self.add_default_info_extractors()
for pp_def_raw in self.params.get('postprocessors', []):
- pp_class = get_postprocessor(pp_def_raw['key'])
pp_def = dict(pp_def_raw)
- del pp_def['key']
- if 'when' in pp_def:
- when = pp_def['when']
- del pp_def['when']
- else:
- when = 'post_process'
+ 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)
self.to_stdout(
message, skip_eol, quiet=self.params.get('quiet', False))
- def report_warning(self, message):
+ def report_warning(self, message, only_once=False):
'''
Print the message to stderr, it will be prefixed with 'WARNING:'
If stderr is a tty file the 'WARNING:' will be colored
'''
+ if only_once:
+ if message in self._reported_warnings:
+ return
+ self._reported_warnings.add(message)
if self.params.get('logger') is not None:
self.params['logger'].warning(message)
else:
'Put from __future__ import unicode_literals at the top of your code file or consider switching to Python 3.x.')
return outtmpl_dict
+ def get_output_path(self, dir_type='', filename=None):
+ paths = self.params.get('paths', {})
+ assert isinstance(paths, dict)
+ path = os.path.join(
+ expand_path(paths.get('home', '').strip()),
+ expand_path(paths.get(dir_type, '').strip()) if dir_type else '',
+ filename or '')
+
+ # Temporary fix for #4787
+ # 'Treat' all problem characters by passing filename through preferredencoding
+ # to workaround encoding issues with subprocess on python2 @ Windows
+ if sys.version_info < (3, 0) and sys.platform == 'win32':
+ path = encodeFilename(path, True).decode(preferredencoding())
+ return sanitize_path(path, force=self.params.get('windowsfilenames'))
+
@staticmethod
def validate_outtmpl(tmpl):
''' @return None or Exception object '''
fmt = outer_mobj.group('format')
mobj = re.match(INTERNAL_FORMAT_RE, key)
if mobj is None:
- value, default = None, na
+ value, default, mobj = None, na, {'fields': ''}
else:
mobj = mobj.groupdict()
default = mobj['default'] if mobj['default'] is not None else na
fmt = '0{:d}d'.format(field_size_compat_map[key])
value = default if value is None else value
- key += '\0%s' % fmt
if fmt == 'c':
value = compat_str(value)
# So we convert it to repr first
value, fmt = repr(value), '%ss' % fmt[:-1]
if fmt[-1] in 'csr':
- value = sanitize(key, value)
+ value = sanitize(mobj['fields'].split('.')[-1], value)
+ key += '\0%s' % fmt
TMPL_DICT[key] = value
return '%({key}){fmt}'.format(key=key, fmt=fmt)
def prepare_filename(self, info_dict, dir_type='', warn=False):
"""Generate the output filename."""
- paths = self.params.get('paths', {})
- assert isinstance(paths, dict)
+
filename = self._prepare_filename(info_dict, dir_type or 'default')
- if warn and not self.__prepare_filename_warned:
- if not paths:
+ if warn:
+ if not self.params.get('paths'):
pass
elif filename == '-':
- self.report_warning('--paths is ignored when an outputting to stdout')
+ self.report_warning('--paths is ignored when an outputting to stdout', only_once=True)
elif os.path.isabs(filename):
- self.report_warning('--paths is ignored since an absolute path is given in output template')
+ self.report_warning('--paths is ignored since an absolute path is given in output template', only_once=True)
self.__prepare_filename_warned = True
if filename == '-' or not filename:
return filename
- homepath = expand_path(paths.get('home', '').strip())
- assert isinstance(homepath, compat_str)
- subdir = expand_path(paths.get(dir_type, '').strip()) if dir_type else ''
- assert isinstance(subdir, compat_str)
- path = os.path.join(homepath, subdir, filename)
-
- # Temporary fix for #4787
- # 'Treat' all problem characters by passing filename through preferredencoding
- # to workaround encoding issues with subprocess on python2 @ Windows
- if sys.version_info < (3, 0) and sys.platform == 'win32':
- path = encodeFilename(path, True).decode(preferredencoding())
- return sanitize_path(path, force=self.params.get('windowsfilenames'))
+ return self.get_output_path(dir_type, filename)
def _match_entry(self, info_dict, incomplete=False, silent=False):
""" Returns None if the file should be downloaded """
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')
+ return wrapper(self, *args, **kwargs)
except (MaxDownloadsReached, ExistingVideoReached, RejectedVideoReached):
raise
except Exception as e:
return ie_result
def add_default_extra_info(self, ie_result, ie, url):
- self.add_extra_info(ie_result, {
- 'extractor': ie.IE_NAME,
- 'webpage_url': url,
- 'original_url': url,
- 'webpage_url_basename': url_basename(url),
- 'extractor_key': ie.ie_key(),
- })
+ if url is not None:
+ self.add_extra_info(ie_result, {
+ 'webpage_url': url,
+ 'original_url': url,
+ 'webpage_url_basename': url_basename(url),
+ })
+ if ie is not None:
+ self.add_extra_info(ie_result, {
+ 'extractor': ie.IE_NAME,
+ 'extractor_key': ie.ie_key(),
+ })
def process_ie_result(self, ie_result, download=True, extra_info={}):
"""
or extract_flat is True):
info_copy = ie_result.copy()
self.add_extra_info(info_copy, extra_info)
- self.add_default_extra_info(
- info_copy, self.get_info_extractor(ie_result.get('ie_key')), ie_result['url'])
+ ie = try_get(ie_result.get('ie_key'), self.get_info_extractor)
+ self.add_default_extra_info(info_copy, ie, ie_result['url'])
self.__forced_printings(info_copy, self.prepare_filename(info_copy), incomplete=True)
return ie_result
if not isinstance(ie_entries, (list, PagedList)):
ie_entries = LazyList(ie_entries)
+ def get_entry(i):
+ return YoutubeDL.__handle_extraction_exceptions(
+ lambda self, i: ie_entries[i - 1]
+ )(self, i)
+
entries = []
for i in playlistitems or itertools.count(playliststart):
if playlistitems is None and playlistend is not None and playlistend < i:
break
entry = None
try:
- entry = ie_entries[i - 1]
+ entry = get_entry(i)
if entry is None:
raise EntryNotInPlaylist()
except (IndexError, EntryNotInPlaylist):
'!=': operator.ne,
}
operator_rex = re.compile(r'''(?x)\s*
- (?P<key>width|height|tbr|abr|vbr|asr|filesize|filesize_approx|fps)
- \s*(?P<op>%s)(?P<none_inclusive>\s*\?)?\s*
- (?P<value>[0-9.]+(?:[kKmMgGtTpPeEzZyY]i?[Bb]?)?)
- $
+ (?P<key>width|height|tbr|abr|vbr|asr|filesize|filesize_approx|fps)\s*
+ (?P<op>%s)(?P<none_inclusive>\s*\?)?\s*
+ (?P<value>[0-9.]+(?:[kKmMgGtTpPeEzZyY]i?[Bb]?)?)\s*
''' % '|'.join(map(re.escape, OPERATORS.keys())))
- m = operator_rex.search(filter_spec)
+ m = operator_rex.fullmatch(filter_spec)
if m:
try:
comparison_value = int(m.group('value'))
'$=': lambda attr, value: attr.endswith(value),
'*=': lambda attr, value: value in attr,
}
- str_operator_rex = re.compile(r'''(?x)
- \s*(?P<key>[a-zA-Z0-9._-]+)
- \s*(?P<negation>!\s*)?(?P<op>%s)(?P<none_inclusive>\s*\?)?
- \s*(?P<value>[a-zA-Z0-9._-]+)
- \s*$
+ str_operator_rex = re.compile(r'''(?x)\s*
+ (?P<key>[a-zA-Z0-9._-]+)\s*
+ (?P<negation>!\s*)?(?P<op>%s)(?P<none_inclusive>\s*\?)?\s*
+ (?P<value>[a-zA-Z0-9._-]+)\s*
''' % '|'.join(map(re.escape, STR_OPERATORS.keys())))
- m = str_operator_rex.search(filter_spec)
+ m = str_operator_rex.fullmatch(filter_spec)
if m:
comparison_value = m.group('value')
str_op = STR_OPERATORS[m.group('op')]
op = str_op
if not m:
- raise ValueError('Invalid filter specification %r' % filter_spec)
+ raise SyntaxError('Invalid filter specification %r' % filter_spec)
def _filter(f):
actual_value = f.get(m.group('key'))
formats_info.extend(format_2.get('requested_formats', (format_2,)))
if not allow_multiple_streams['video'] or not allow_multiple_streams['audio']:
- get_no_more = {"video": False, "audio": False}
+ get_no_more = {'video': False, 'audio': False}
for (i, fmt_info) in enumerate(formats_info):
- for aud_vid in ["audio", "video"]:
+ if fmt_info.get('acodec') == fmt_info.get('vcodec') == 'none':
+ formats_info.pop(i)
+ continue
+ for aud_vid in ['audio', 'video']:
if not allow_multiple_streams[aud_vid] and fmt_info.get(aud_vid[0] + 'codec') != 'none':
if get_no_more[aud_vid]:
formats_info.pop(i)
return new_dict
def _check_formats(formats):
+ if not check_formats:
+ yield from formats
+ return
for f in formats:
self.to_screen('[info] Testing format %s' % f['format_id'])
- paths = self.params.get('paths', {})
- temp_file = os.path.join(
- expand_path(paths.get('home', '').strip()),
- expand_path(paths.get('temp', '').strip()),
- 'ytdl.%s.f%s.check-format' % (random_uuidv4(), f['format_id']))
+ temp_file = tempfile.NamedTemporaryFile(
+ suffix='.tmp', delete=False,
+ dir=self.get_output_path('temp') or None)
+ temp_file.close()
try:
- dl, _ = self.dl(temp_file, f, test=True)
- except (ExtractorError, IOError, OSError, ValueError) + network_exceptions:
- dl = False
+ 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):
- os.remove(temp_file)
- if dl:
+ 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 selector_function(ctx):
for f in fs:
- for format in f(ctx):
- yield format
+ yield from f(ctx)
return selector_function
elif selector.type == GROUP: # ()
return picked_formats
return []
+ elif selector.type == MERGE: # +
+ 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))):
+ yield _merge(pair)
+
elif selector.type == SINGLE: # atom
format_spec = selector.selector or 'best'
# TODO: Add allvideo, allaudio etc by generalizing the code with best/worst selector
if format_spec == 'all':
def selector_function(ctx):
- formats = list(ctx['formats'])
- if check_formats:
- formats = _check_formats(formats)
- for f in formats:
- yield f
+ yield from _check_formats(ctx['formats'])
elif format_spec == 'mergeall':
def selector_function(ctx):
formats = list(_check_formats(ctx['formats']))
format_modified = mobj.group('mod') is not None
format_fallback = not format_type and not format_modified # for b, w
- filter_f = (
+ _filter_f = (
(lambda f: f.get('%scodec' % format_type) != 'none')
if format_type and format_modified # bv*, ba*, wv*, wa*
else (lambda f: f.get('%scodec' % not_format_type) == 'none')
if format_type # bv, ba, wv, wa
else (lambda f: f.get('vcodec') != 'none' and f.get('acodec') != 'none')
if not format_modified # b, w
- else None) # b*, w*
+ else lambda f: True) # b*, w*
+ filter_f = lambda f: _filter_f(f) and (
+ f.get('vcodec') != 'none' or f.get('acodec') != 'none')
else:
filter_f = ((lambda f: f.get('ext') == format_spec)
if format_spec in ['mp4', 'flv', 'webm', '3gp', 'm4a', 'mp3', 'ogg', 'aac', 'wav'] # extension
def selector_function(ctx):
formats = list(ctx['formats'])
- if not formats:
- return
matches = list(filter(filter_f, formats)) if filter_f is not None else formats
if format_fallback and ctx['incomplete_formats'] and not matches:
# for extractors with incomplete formats (audio only (soundcloud)
# or video only (imgur)) best/worst will fallback to
# best/worst {video,audio}-only format
matches = formats
- if format_reverse:
- matches = matches[::-1]
- if check_formats:
- matches = list(itertools.islice(_check_formats(matches), format_idx))
- n = len(matches)
- if -n <= format_idx - 1 < n:
+ matches = LazyList(_check_formats(matches[::-1 if format_reverse else 1]))
+ try:
yield matches[format_idx - 1]
-
- elif selector.type == MERGE: # +
- 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))):
- yield _merge(pair)
+ except IndexError:
+ return
filters = [self._build_format_filter(f) for f in selector.filters]
self.cookiejar.add_cookie_header(pr)
return pr.get_header('Cookie')
- @staticmethod
- def _sanitize_thumbnails(info_dict):
+ def _sanitize_thumbnails(self, info_dict):
thumbnails = info_dict.get('thumbnails')
if thumbnails is None:
thumbnail = info_dict.get('thumbnail')
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):
- t['url'] = sanitize_url(t['url'])
- if t.get('width') and t.get('height'):
- t['resolution'] = '%dx%d' % (t['width'], t['height'])
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 not False:
+ info_dict['thumbnails'] = LazyList(filter(thumbnail_tester(), thumbnails[::-1])).reverse()
+ else:
+ info_dict['thumbnails'] = thumbnails
def process_video_result(self, info_dict, download=True):
assert info_dict.get('_type', 'video') == 'video'
self._sanitize_thumbnails(info_dict)
- if self.params.get('list_thumbnails'):
- self.list_thumbnails(info_dict)
- return
-
thumbnail = info_dict.get('thumbnail')
thumbnails = info_dict.get('thumbnails')
if thumbnail:
automatic_captions = info_dict.get('automatic_captions')
subtitles = info_dict.get('subtitles')
- if self.params.get('listsubtitles', False):
- if 'automatic_captions' in info_dict:
- self.list_subtitles(
- info_dict['id'], automatic_captions, 'automatic captions')
- self.list_subtitles(info_dict['id'], subtitles, 'subtitles')
- return
-
info_dict['requested_subtitles'] = self.process_subtitles(
info_dict['id'], subtitles, automatic_captions)
info_dict, _ = self.pre_process(info_dict)
- if self.params.get('listformats'):
- if not info_dict.get('formats'):
- raise ExtractorError('No video formats found', expected=True)
- self.list_formats(info_dict)
+ list_only = self.params.get('list_thumbnails') or self.params.get('listformats') or self.params.get('listsubtitles')
+ if list_only:
+ self.__forced_printings(info_dict, self.prepare_filename(info_dict), incomplete=True)
+ if self.params.get('list_thumbnails'):
+ self.list_thumbnails(info_dict)
+ if self.params.get('listformats'):
+ if not info_dict.get('formats'):
+ raise ExtractorError('No video formats found', expected=True)
+ 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')
return
- req_format = self.params.get('format')
- if req_format is None:
+ format_selector = self.format_selector
+ if format_selector is None:
req_format = self._default_format_spec(info_dict, download=download)
self.write_debug('Default format spec: %s' % req_format)
-
- format_selector = self.build_format_selector(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.
raise ExtractorError('Requested format is not available', expected=True)
else:
self.report_warning('Requested format is not available')
+ # Process what we can, even without any available formats.
+ self.process_info(dict(info_dict))
elif download:
self.to_screen(
'[info] %s: Downloading %d format(s): %s' % (
# TODO: backward compatibility, to be removed
info_dict['fulltitle'] = info_dict['title']
- if 'format' not in info_dict:
+ if 'format' not in info_dict and 'ext' in info_dict:
info_dict['format'] = info_dict['ext']
if self._match_entry(info_dict) is not None:
files_to_move = {}
# Forced printings
- self.__forced_printings(info_dict, full_filename, incomplete=False)
+ self.__forced_printings(info_dict, full_filename, incomplete=('format' not in info_dict))
if self.params.get('simulate', False):
if self.params.get('force_write_download_archive', False):
requested_formats = info_dict['requested_formats']
old_ext = info_dict['ext']
- if self.params.get('merge_output_format') is None:
- if not compatible_formats(requested_formats):
- info_dict['ext'] = 'mkv'
- self.report_warning(
- 'Requested formats are incompatible for merge and will be merged into mkv.')
- if (info_dict['ext'] == 'webm'
- and self.params.get('writethumbnail', False)
- and info_dict.get('thumbnails')):
- info_dict['ext'] = 'mkv'
- self.report_warning(
- 'webm doesn\'t support embedding a thumbnail, mkv will be used.')
+ if self.params.get('merge_output_format') is None and not compatible_formats(requested_formats):
+ info_dict['ext'] = 'mkv'
+ self.report_warning(
+ 'Requested formats are incompatible for merge and will be merged into mkv.')
def correct_ext(filename):
filename_real_ext = os.path.splitext(filename)[1][1:]
return
if success and full_filename != '-':
- # Fixup content
- fixup_policy = self.params.get('fixup')
- if fixup_policy is None:
- fixup_policy = 'detect_or_warn'
-
- INSTALL_FFMPEG_MESSAGE = 'Install ffmpeg to fix this automatically.'
-
- stretched_ratio = info_dict.get('stretched_ratio')
- if stretched_ratio is not None and stretched_ratio != 1:
- if fixup_policy == 'warn':
- self.report_warning('%s: Non-uniform pixel ratio (%s)' % (
- info_dict['id'], stretched_ratio))
- elif fixup_policy == 'detect_or_warn':
- stretched_pp = FFmpegFixupStretchedPP(self)
- if stretched_pp.available:
- info_dict['__postprocessors'].append(stretched_pp)
- else:
- self.report_warning(
- '%s: Non-uniform pixel ratio (%s). %s'
- % (info_dict['id'], stretched_ratio, INSTALL_FFMPEG_MESSAGE))
- else:
- assert fixup_policy in ('ignore', 'never')
- if (info_dict.get('requested_formats') is None
- and info_dict.get('container') == 'm4a_dash'
- and info_dict.get('ext') == 'm4a'):
- if fixup_policy == 'warn':
- self.report_warning(
- '%s: writing DASH m4a. '
- 'Only some players support this container.'
- % info_dict['id'])
- elif fixup_policy == 'detect_or_warn':
- fixup_pp = FFmpegFixupM4aPP(self)
- if fixup_pp.available:
- info_dict['__postprocessors'].append(fixup_pp)
- else:
- self.report_warning(
- '%s: writing DASH m4a. '
- 'Only some players support this container. %s'
- % (info_dict['id'], INSTALL_FFMPEG_MESSAGE))
- else:
- assert fixup_policy in ('ignore', 'never')
-
- if ('protocol' in info_dict
- and get_suitable_downloader(info_dict, self.params).__name__ == 'HlsFD'):
- if fixup_policy == 'warn':
- self.report_warning('%s: malformed AAC bitstream detected.' % (
- info_dict['id']))
- elif fixup_policy == 'detect_or_warn':
- fixup_pp = FFmpegFixupM3u8PP(self)
- if fixup_pp.available:
- info_dict['__postprocessors'].append(fixup_pp)
- else:
- self.report_warning(
- '%s: malformed AAC bitstream detected. %s'
- % (info_dict['id'], INSTALL_FFMPEG_MESSAGE))
- else:
- assert fixup_policy in ('ignore', 'never')
+ def fixup():
+ do_fixup = True
+ fixup_policy = self.params.get('fixup')
+ vid = info_dict['id']
+ if fixup_policy in ('ignore', 'never'):
+ return
+ elif fixup_policy == 'warn':
+ do_fixup = False
+ elif fixup_policy != 'force':
+ assert fixup_policy in ('detect_or_warn', None)
+ if not info_dict.get('__real_download'):
+ do_fixup = False
+
+ def ffmpeg_fixup(cndn, msg, cls):
+ if not cndn:
+ return
+ if not do_fixup:
+ self.report_warning(f'{vid}: {msg}')
+ return
+ pp = cls(self)
+ if pp.available:
+ info_dict['__postprocessors'].append(pp)
+ else:
+ self.report_warning(f'{vid}: {msg}. Install ffmpeg to fix this automatically')
+
+ stretched_ratio = info_dict.get('stretched_ratio')
+ ffmpeg_fixup(
+ stretched_ratio not in (1, None),
+ f'Non-uniform pixel ratio {stretched_ratio}',
+ FFmpegFixupStretchedPP)
+
+ ffmpeg_fixup(
+ (info_dict.get('requested_formats') is None
+ and info_dict.get('container') == 'm4a_dash'
+ and info_dict.get('ext') == 'm4a'),
+ 'writing DASH m4a. Only some players support this container',
+ FFmpegFixupM4aPP)
+
+ downloader = (get_suitable_downloader(info_dict, self.params).__name__
+ if 'protocol' in info_dict else None)
+ ffmpeg_fixup(downloader == 'HlsFD', 'malformed AAC bitstream detected', FFmpegFixupM3u8PP)
+ ffmpeg_fixup(downloader == 'WebSocketFragmentFD', 'malformed timestamps detected', FFmpegFixupTimestampPP)
+ ffmpeg_fixup(downloader == 'WebSocketFragmentFD', 'malformed duration detected', FFmpegFixupDurationPP)
+
+ fixup()
try:
info_dict = self.post_process(dl_filename, info_dict, files_to_move)
except PostProcessingError as err:
info = self.filter_requested_info(json.loads('\n'.join(f)), self.params.get('clean_infojson', True))
try:
self.process_ie_result(info, download=True)
- except (DownloadError, EntryNotInPlaylist):
+ except (DownloadError, EntryNotInPlaylist, ThrottledDownload):
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)
info_dict['epoch'] = int(time.time())
reject = lambda k, v: k in remove_keys
filter_fn = lambda obj: (
- list(map(filter_fn, obj)) if isinstance(obj, (list, tuple, set))
+ 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)))
return filter_fn(info_dict)
@staticmethod
def format_resolution(format, default='unknown'):
if format.get('vcodec') == 'none':
+ if format.get('acodec') == 'none':
+ return 'images'
return 'audio only'
if format.get('resolution') is not None:
return format['resolution']
res += '~' + format_bytes(fdict['filesize_approx'])
return res
- def _format_note_table(self, f):
- def join_fields(*vargs):
- return ', '.join((val for val in vargs if val != ''))
-
- return join_fields(
- 'UNSUPPORTED' if f.get('ext') in ('f4f', 'f4m') else '',
- format_field(f, 'language', '[%s]'),
- format_field(f, 'format_note'),
- format_field(f, 'container', ignore=(None, f.get('ext'))),
- format_field(f, 'asr', '%5dHz'))
-
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('list_formats_as_table', True) is not False)
+ and self.params.get('listformats_table', True) is not False)
if new_format:
table = [
[
format_field(f, 'acodec', default='unknown').replace('none', ''),
format_field(f, 'abr', '%3dk'),
format_field(f, 'asr', '%5dHz'),
- self._format_note_table(f)]
- for f in formats
- if f.get('preference') is None or f['preference'] >= -1000]
+ ', '.join(filter(None, (
+ 'UNSUPPORTED' if f.get('ext') in ('f4f', 'f4m') else '',
+ format_field(f, 'language', '[%s]'),
+ format_field(f, 'format_note'),
+ format_field(f, 'container', ignore=(None, f.get('ext'))),
+ format_field(f, 'asr', '%5dHz')))),
+ ] 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', 'NOTE']
+ '|', 'VCODEC', ' VBR', 'ACODEC', ' ABR', ' ASR', 'MORE INFO']
else:
table = [
[
header_line = ['format code', 'extension', 'resolution', 'note']
self.to_screen(
- '[info] Available formats for %s:\n%s' % (info_dict['id'], render_table(
- header_line,
- table,
- delim=new_format,
- extraGap=(0 if new_format else 1),
- hideEmpty=new_format)))
+ '[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))
def list_thumbnails(self, info_dict):
- thumbnails = info_dict.get('thumbnails')
+ thumbnails = list(info_dict.get('thumbnails'))
if not thumbnails:
self.to_screen('[info] No thumbnails present for %s' % info_dict['id'])
return
self.to_screen(
'[info] Thumbnails for %s:' % info_dict['id'])
- self.to_screen(render_table(
+ self.to_stdout(render_table(
['ID', 'width', 'height', 'URL'],
[[t['id'], t.get('width', 'unknown'), t.get('height', 'unknown'), t['url']] for t in thumbnails]))
'Available %s for %s:' % (name, video_id))
def _row(lang, formats):
- exts, names = zip(*((f['ext'], f.get('name', 'unknown')) for f in reversed(formats)))
+ exts, names = zip(*((f['ext'], f.get('name') or 'unknown') for f in reversed(formats)))
if len(set(names)) == 1:
names = [] if names[0] == 'unknown' else names[:1]
return [lang, ', '.join(names), ', '.join(exts)]
- self.to_screen(render_table(
+ self.to_stdout(render_table(
['Language', 'Name', 'Formats'],
[_row(lang, formats) for lang, formats in subtitles.items()],
hideEmpty=True))
multiple = write_all and len(thumbnails) > 1
ret = []
- for t in thumbnails[::1 if write_all else -1]:
+ for t in thumbnails[::-1]:
thumb_ext = determine_ext(t['url'], 'jpg')
suffix = '%s.' % t['id'] if multiple else ''
thumb_display_id = '%s ' % t['id'] if multiple else ''
if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(thumb_filename)):
ret.append(suffix + thumb_ext)
+ t['filepath'] = thumb_filename
self.to_screen('[%s] %s: Thumbnail %sis already present' %
(info_dict['extractor'], info_dict['id'], thumb_display_id))
else:
projects / yt-dlp.git / blobdiff