import socket
import sys
import time
+import tokenize
import traceback
if os.name == 'nt':
compat_http_client,
compat_kwargs,
compat_str,
+ compat_tokenize_tokenize,
compat_urllib_error,
compat_urllib_request,
+ compat_urllib_request_DataHandler,
)
from .utils import (
- escape_url,
ContentTooShortError,
date_from_str,
DateRange,
DEFAULT_OUTTMPL,
determine_ext,
+ determine_protocol,
DownloadError,
+ encode_compat_str,
encodeFilename,
+ error_to_compat_str,
ExtractorError,
format_bytes,
formatSeconds,
SameFileError,
sanitize_filename,
sanitize_path,
+ sanitized_Request,
std_headers,
subtitles_filename,
UnavailableVideoError,
version_tuple,
write_json_file,
write_string,
+ YoutubeDLCookieProcessor,
YoutubeDLHandler,
prepend_extension,
replace_extension,
username: Username for authentication purposes.
password: Password for authentication purposes.
- videopassword: Password for acces a video.
+ videopassword: Password for accessing a video.
usenetrc: Use netrc for authentication instead.
verbose: Print additional info to stdout.
quiet: Do not print messages to stdout.
outtmpl: Template for output names.
restrictfilenames: Do not allow "&" and spaces in file names
ignoreerrors: Do not stop on download errors.
+ force_generic_extractor: Force downloader to use the generic extractor
nooverwrites: Prevent overwriting files.
playliststart: Playlist item to start at.
playlistend: Playlist item to end at.
writethumbnail: Write the thumbnail image to a file
write_all_thumbnails: Write all thumbnail formats to files
writesubtitles: Write the video subtitles to a file
- writeautomaticsub: Write the automatic subtitles to a file
+ writeautomaticsub: Write the automatically generated subtitles to a file
allsubtitles: Downloads all the subtitles of the video
(requires writesubtitles or writeautomaticsub)
listsubtitles: Lists all available subtitles for the video
the downloader (see youtube_dl/downloader/common.py):
nopart, updatetime, buffersize, ratelimit, min_filesize, max_filesize, test,
noresizebuffer, retries, continuedl, noprogress, consoletitle,
- xattr_set_filesize, external_downloader_args.
+ xattr_set_filesize, external_downloader_args, hls_use_mpegts.
The following options are used by the post processors:
prefer_ffmpeg: If True, use ffmpeg instead of avconv if both are available,
otherwise prefer avconv.
+ postprocessor_args: A list of additional command-line arguments for the
+ postprocessor.
"""
params = None
self._num_downloads = 0
self._screen_file = [sys.stdout, sys.stderr][params.get('logtostderr', False)]
self._err_file = sys.stderr
- self.params = params
+ self.params = {
+ # Default parameters
+ 'nocheckcertificate': False,
+ }
+ self.params.update(params)
self.cache = Cache(self)
if params.get('bidi_workaround', False):
tb = ''
if hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]:
tb += ''.join(traceback.format_exception(*sys.exc_info()[1].exc_info))
- tb += compat_str(traceback.format_exc())
+ tb += encode_compat_str(traceback.format_exc())
else:
tb_data = traceback.format_list(traceback.extract_stack())
tb = ''.join(tb_data)
if v is not None)
template_dict = collections.defaultdict(lambda: 'NA', template_dict)
- outtmpl = sanitize_path(self.params.get('outtmpl', DEFAULT_OUTTMPL))
+ outtmpl = self.params.get('outtmpl', DEFAULT_OUTTMPL)
tmpl = compat_expanduser(outtmpl)
filename = tmpl % template_dict
# Temporary fix for #4787
# to workaround encoding issues with subprocess on python2 @ Windows
if sys.version_info < (3, 0) and sys.platform == 'win32':
filename = encodeFilename(filename, True).decode(preferredencoding())
- return filename
+ return sanitize_path(filename)
except ValueError as err:
self.report_error('Error in output template: ' + str(err) + ' (encoding: ' + repr(preferredencoding()) + ')')
return None
if rejecttitle:
if re.search(rejecttitle, title, re.IGNORECASE):
return '"' + title + '" title matched reject pattern "' + rejecttitle + '"'
- date = info_dict.get('upload_date', None)
+ date = info_dict.get('upload_date')
if date is not None:
dateRange = self.params.get('daterange', DateRange())
if date not in dateRange:
return '%s upload date is not in range %s' % (date_from_str(date).isoformat(), dateRange)
- view_count = info_dict.get('view_count', None)
+ view_count = info_dict.get('view_count')
if view_count is not None:
min_views = self.params.get('min_views')
if min_views is not None and view_count < min_views:
info_dict.setdefault(key, value)
def extract_info(self, url, download=True, ie_key=None, extra_info={},
- process=True):
+ process=True, force_generic_extractor=False):
'''
Returns a list with a dictionary for each video we find.
If 'download', also downloads the videos.
extra_info is a dict containing the extra values to add to each result
'''
+ if not ie_key and force_generic_extractor:
+ ie_key = 'Generic'
+
if ie_key:
ies = [self.get_info_extractor(ie_key)]
else:
return self.process_ie_result(ie_result, download, extra_info)
else:
return ie_result
- except ExtractorError as de: # An error we somewhat expected
- self.report_error(compat_str(de), de.format_traceback())
+ except ExtractorError as e: # An error we somewhat expected
+ self.report_error(compat_str(e), e.format_traceback())
break
except MaxDownloadsReached:
raise
except Exception as e:
if self.params.get('ignoreerrors', False):
- self.report_error(compat_str(e), tb=compat_str(traceback.format_exc()))
+ self.report_error(error_to_compat_str(e), tb=encode_compat_str(traceback.format_exc()))
break
else:
raise
It will also download the videos if 'download'.
Returns the resolved ie_result.
"""
-
result_type = ie_result.get('_type', 'video')
if result_type in ('url', 'url_transparent'):
force_properties = dict(
(k, v) for k, v in ie_result.items() if v is not None)
- for f in ('_type', 'url'):
+ for f in ('_type', 'url', 'ie_key'):
if f in force_properties:
del force_properties[f]
new_result = info.copy()
new_result, download=download, extra_info=extra_info)
elif result_type == 'playlist' or result_type == 'multi_video':
# We process each entry in the playlist
- playlist = ie_result.get('title', None) or ie_result.get('id', None)
+ playlist = ie_result.get('title') or ie_result.get('id')
self.to_screen('[download] Downloading playlist: %s' % playlist)
playlist_results = []
playliststart = self.params.get('playliststart', 1) - 1
- playlistend = self.params.get('playlistend', None)
+ playlistend = self.params.get('playlistend')
# For backwards compatibility, interpret -1 as whole list
if playlistend == -1:
playlistend = None
- playlistitems_str = self.params.get('playlist_items', None)
+ playlistitems_str = self.params.get('playlist_items')
playlistitems = None
if playlistitems_str is not None:
def iter_playlistitems(format):
if isinstance(ie_entries, list):
n_all_entries = len(ie_entries)
if playlistitems:
- entries = [ie_entries[i - 1] for i in playlistitems]
+ entries = [
+ ie_entries[i - 1] for i in playlistitems
+ if -n_all_entries <= i - 1 < n_all_entries]
else:
entries = ie_entries[playliststart:playlistend]
n_entries = len(entries)
self.to_screen(
- "[%s] playlist %s: Collected %d video ids (downloading %d of them)" %
+ '[%s] playlist %s: Collected %d video ids (downloading %d of them)' %
(ie_result['extractor'], playlist, n_all_entries, n_entries))
elif isinstance(ie_entries, PagedList):
if playlistitems:
playliststart, playlistend)
n_entries = len(entries)
self.to_screen(
- "[%s] playlist %s: Downloading %d videos" %
+ '[%s] playlist %s: Downloading %d videos' %
(ie_result['extractor'], playlist, n_entries))
else: # iterable
if playlistitems:
ie_entries, playliststart, playlistend))
n_entries = len(entries)
self.to_screen(
- "[%s] playlist %s: Downloading %d videos" %
+ '[%s] playlist %s: Downloading %d videos' %
(ie_result['extractor'], playlist, n_entries))
if self.params.get('playlistreverse', False):
extra_info=extra)
playlist_results.append(entry_result)
ie_result['entries'] = playlist_results
+ self.to_screen('[download] Finished downloading playlist: %s' % playlist)
return ie_result
elif result_type == 'compat_list':
self.report_warning(
else:
raise Exception('Invalid result type: %s' % result_type)
- def _apply_format_filter(self, format_spec, available_formats):
- " Returns a tuple of the remaining format_spec and filtered formats "
+ def _build_format_filter(self, filter_spec):
+ " Returns a function to filter the formats according to the filter_spec "
OPERATORS = {
'<': operator.lt,
'=': operator.eq,
'!=': operator.ne,
}
- operator_rex = re.compile(r'''(?x)\s*\[
+ operator_rex = re.compile(r'''(?x)\s*
(?P<key>width|height|tbr|abr|vbr|asr|filesize|fps)
\s*(?P<op>%s)(?P<none_inclusive>\s*\?)?\s*
(?P<value>[0-9.]+(?:[kKmMgGtTpPeEzZyY]i?[Bb]?)?)
- \]$
+ $
''' % '|'.join(map(re.escape, OPERATORS.keys())))
- m = operator_rex.search(format_spec)
+ m = operator_rex.search(filter_spec)
if m:
try:
comparison_value = int(m.group('value'))
if comparison_value is None:
raise ValueError(
'Invalid value %r in format specification %r' % (
- m.group('value'), format_spec))
+ m.group('value'), filter_spec))
op = OPERATORS[m.group('op')]
if not m:
STR_OPERATORS = {
'=': operator.eq,
'!=': operator.ne,
+ '^=': lambda attr, value: attr.startswith(value),
+ '$=': lambda attr, value: attr.endswith(value),
+ '*=': lambda attr, value: value in attr,
}
- str_operator_rex = re.compile(r'''(?x)\s*\[
+ str_operator_rex = re.compile(r'''(?x)
\s*(?P<key>ext|acodec|vcodec|container|protocol)
\s*(?P<op>%s)(?P<none_inclusive>\s*\?)?
- \s*(?P<value>[a-zA-Z0-9_-]+)
- \s*\]$
+ \s*(?P<value>[a-zA-Z0-9._-]+)
+ \s*$
''' % '|'.join(map(re.escape, STR_OPERATORS.keys())))
- m = str_operator_rex.search(format_spec)
+ m = str_operator_rex.search(filter_spec)
if m:
comparison_value = m.group('value')
op = STR_OPERATORS[m.group('op')]
if not m:
- raise ValueError('Invalid format specification %r' % format_spec)
+ raise ValueError('Invalid filter specification %r' % filter_spec)
def _filter(f):
actual_value = f.get(m.group('key'))
if actual_value is None:
return m.group('none_inclusive')
return op(actual_value, comparison_value)
- new_formats = [f for f in available_formats if _filter(f)]
+ return _filter
+
+ def build_format_selector(self, format_spec):
+ def syntax_error(note, start):
+ message = (
+ 'Invalid format specification: '
+ '{0}\n\t{1}\n\t{2}^'.format(note, format_spec, ' ' * start[1]))
+ return SyntaxError(message)
+
+ PICKFIRST = 'PICKFIRST'
+ MERGE = 'MERGE'
+ SINGLE = 'SINGLE'
+ GROUP = 'GROUP'
+ FormatSelector = collections.namedtuple('FormatSelector', ['type', 'selector', 'filters'])
+
+ def _parse_filter(tokens):
+ filter_parts = []
+ for type, string, start, _, _ in tokens:
+ if type == tokenize.OP and string == ']':
+ return ''.join(filter_parts)
+ else:
+ filter_parts.append(string)
+
+ def _remove_unused_ops(tokens):
+ # Remove operators that we don't use and join them with the surrounding strings
+ # for example: 'mp4' '-' 'baseline' '-' '16x9' is converted to 'mp4-baseline-16x9'
+ ALLOWED_OPS = ('/', '+', ',', '(', ')')
+ last_string, last_start, last_end, last_line = None, None, None, None
+ for type, string, start, end, line in tokens:
+ if type == tokenize.OP and string == '[':
+ if last_string:
+ yield tokenize.NAME, last_string, last_start, last_end, last_line
+ last_string = None
+ yield type, string, start, end, line
+ # everything inside brackets will be handled by _parse_filter
+ for type, string, start, end, line in tokens:
+ yield type, string, start, end, line
+ if type == tokenize.OP and string == ']':
+ break
+ elif type == tokenize.OP and string in ALLOWED_OPS:
+ if last_string:
+ yield tokenize.NAME, last_string, last_start, last_end, last_line
+ last_string = None
+ yield type, string, start, end, line
+ elif type in [tokenize.NAME, tokenize.NUMBER, tokenize.OP]:
+ if not last_string:
+ last_string = string
+ last_start = start
+ last_end = end
+ else:
+ last_string += string
+ if last_string:
+ yield tokenize.NAME, last_string, last_start, last_end, last_line
+
+ def _parse_format_selection(tokens, inside_merge=False, inside_choice=False, inside_group=False):
+ selectors = []
+ current_selector = None
+ for type, string, start, _, _ in tokens:
+ # ENCODING is only defined in python 3.x
+ if type == getattr(tokenize, 'ENCODING', None):
+ continue
+ elif type in [tokenize.NAME, tokenize.NUMBER]:
+ current_selector = FormatSelector(SINGLE, string, [])
+ elif type == tokenize.OP:
+ if string == ')':
+ if not inside_group:
+ # ')' will be handled by the parentheses group
+ tokens.restore_last_token()
+ break
+ elif inside_merge and string in ['/', ',']:
+ tokens.restore_last_token()
+ break
+ elif inside_choice and string == ',':
+ tokens.restore_last_token()
+ break
+ elif string == ',':
+ if not current_selector:
+ raise syntax_error('"," must follow a format selector', start)
+ selectors.append(current_selector)
+ current_selector = None
+ elif string == '/':
+ if not current_selector:
+ raise syntax_error('"/" must follow a format selector', start)
+ first_choice = current_selector
+ second_choice = _parse_format_selection(tokens, inside_choice=True)
+ current_selector = FormatSelector(PICKFIRST, (first_choice, second_choice), [])
+ elif string == '[':
+ if not current_selector:
+ current_selector = FormatSelector(SINGLE, 'best', [])
+ format_filter = _parse_filter(tokens)
+ current_selector.filters.append(format_filter)
+ elif string == '(':
+ if current_selector:
+ raise syntax_error('Unexpected "("', start)
+ group = _parse_format_selection(tokens, inside_group=True)
+ current_selector = FormatSelector(GROUP, group, [])
+ elif string == '+':
+ video_selector = current_selector
+ audio_selector = _parse_format_selection(tokens, inside_merge=True)
+ if not video_selector or not audio_selector:
+ raise syntax_error('"+" must be between two format selectors', start)
+ current_selector = FormatSelector(MERGE, (video_selector, audio_selector), [])
+ else:
+ raise syntax_error('Operator not recognized: "{0}"'.format(string), start)
+ elif type == tokenize.ENDMARKER:
+ break
+ if current_selector:
+ selectors.append(current_selector)
+ return selectors
+
+ def _build_selector_function(selector):
+ if isinstance(selector, list):
+ fs = [_build_selector_function(s) for s in selector]
+
+ def selector_function(formats):
+ for f in fs:
+ for format in f(formats):
+ yield format
+ return selector_function
+ elif selector.type == GROUP:
+ selector_function = _build_selector_function(selector.selector)
+ elif selector.type == PICKFIRST:
+ fs = [_build_selector_function(s) for s in selector.selector]
+
+ def selector_function(formats):
+ for f in fs:
+ picked_formats = list(f(formats))
+ if picked_formats:
+ return picked_formats
+ return []
+ elif selector.type == SINGLE:
+ format_spec = selector.selector
+
+ def selector_function(formats):
+ formats = list(formats)
+ if not formats:
+ return
+ if format_spec == 'all':
+ for f in formats:
+ yield f
+ elif format_spec in ['best', 'worst', None]:
+ format_idx = 0 if format_spec == 'worst' else -1
+ audiovideo_formats = [
+ f for f in formats
+ if f.get('vcodec') != 'none' and f.get('acodec') != 'none']
+ if audiovideo_formats:
+ yield audiovideo_formats[format_idx]
+ # for audio only (soundcloud) or video only (imgur) urls, select the best/worst audio format
+ elif (all(f.get('acodec') != 'none' for f in formats) or
+ all(f.get('vcodec') != 'none' for f in formats)):
+ yield formats[format_idx]
+ elif format_spec == 'bestaudio':
+ audio_formats = [
+ f for f in formats
+ if f.get('vcodec') == 'none']
+ if audio_formats:
+ yield audio_formats[-1]
+ elif format_spec == 'worstaudio':
+ audio_formats = [
+ f for f in formats
+ if f.get('vcodec') == 'none']
+ if audio_formats:
+ yield audio_formats[0]
+ elif format_spec == 'bestvideo':
+ video_formats = [
+ f for f in formats
+ if f.get('acodec') == 'none']
+ if video_formats:
+ yield video_formats[-1]
+ elif format_spec == 'worstvideo':
+ video_formats = [
+ f for f in formats
+ if f.get('acodec') == 'none']
+ if video_formats:
+ yield video_formats[0]
+ else:
+ extensions = ['mp4', 'flv', 'webm', '3gp', 'm4a', 'mp3', 'ogg', 'aac', 'wav']
+ if format_spec in extensions:
+ filter_f = lambda f: f['ext'] == format_spec
+ else:
+ filter_f = lambda f: f['format_id'] == format_spec
+ matches = list(filter(filter_f, formats))
+ if matches:
+ yield matches[-1]
+ elif selector.type == MERGE:
+ def _merge(formats_info):
+ format_1, format_2 = [f['format_id'] for f in formats_info]
+ # The first format must contain the video and the
+ # second the audio
+ if formats_info[0].get('vcodec') == 'none':
+ self.report_error('The first format must '
+ 'contain the video, try using '
+ '"-f %s+%s"' % (format_2, format_1))
+ return
+ # Formats must be opposite (video+audio)
+ if formats_info[0].get('acodec') == 'none' and formats_info[1].get('acodec') == 'none':
+ self.report_error(
+ 'Both formats %s and %s are video-only, you must specify "-f video+audio"'
+ % (format_1, format_2))
+ return
+ output_ext = (
+ formats_info[0]['ext']
+ if self.params.get('merge_output_format') is None
+ else self.params['merge_output_format'])
+ return {
+ 'requested_formats': formats_info,
+ 'format': '%s+%s' % (formats_info[0].get('format'),
+ formats_info[1].get('format')),
+ 'format_id': '%s+%s' % (formats_info[0].get('format_id'),
+ formats_info[1].get('format_id')),
+ 'width': formats_info[0].get('width'),
+ 'height': formats_info[0].get('height'),
+ 'resolution': formats_info[0].get('resolution'),
+ 'fps': formats_info[0].get('fps'),
+ 'vcodec': formats_info[0].get('vcodec'),
+ 'vbr': formats_info[0].get('vbr'),
+ 'stretched_ratio': formats_info[0].get('stretched_ratio'),
+ 'acodec': formats_info[1].get('acodec'),
+ 'abr': formats_info[1].get('abr'),
+ 'ext': output_ext,
+ }
+ video_selector, audio_selector = map(_build_selector_function, selector.selector)
- new_format_spec = format_spec[:-len(m.group(0))]
- if not new_format_spec:
- new_format_spec = 'best'
+ def selector_function(formats):
+ formats = list(formats)
+ for pair in itertools.product(video_selector(formats), audio_selector(formats)):
+ yield _merge(pair)
- return (new_format_spec, new_formats)
+ filters = [self._build_format_filter(f) for f in selector.filters]
- def select_format(self, format_spec, available_formats):
- while format_spec.endswith(']'):
- format_spec, available_formats = self._apply_format_filter(
- format_spec, available_formats)
- if not available_formats:
- return None
+ def final_selector(formats):
+ for _filter in filters:
+ formats = list(filter(_filter, formats))
+ return selector_function(formats)
+ return final_selector
- if format_spec in ['best', 'worst', None]:
- format_idx = 0 if format_spec == 'worst' else -1
- audiovideo_formats = [
- f for f in available_formats
- if f.get('vcodec') != 'none' and f.get('acodec') != 'none']
- if audiovideo_formats:
- return audiovideo_formats[format_idx]
- # for audio only urls, select the best/worst audio format
- elif all(f.get('acodec') != 'none' for f in available_formats):
- return available_formats[format_idx]
- elif format_spec == 'bestaudio':
- audio_formats = [
- f for f in available_formats
- if f.get('vcodec') == 'none']
- if audio_formats:
- return audio_formats[-1]
- elif format_spec == 'worstaudio':
- audio_formats = [
- f for f in available_formats
- if f.get('vcodec') == 'none']
- if audio_formats:
- return audio_formats[0]
- elif format_spec == 'bestvideo':
- video_formats = [
- f for f in available_formats
- if f.get('acodec') == 'none']
- if video_formats:
- return video_formats[-1]
- elif format_spec == 'worstvideo':
- video_formats = [
- f for f in available_formats
- if f.get('acodec') == 'none']
- if video_formats:
- return video_formats[0]
- else:
- extensions = ['mp4', 'flv', 'webm', '3gp', 'm4a', 'mp3', 'ogg', 'aac', 'wav']
- if format_spec in extensions:
- filter_f = lambda f: f['ext'] == format_spec
- else:
- filter_f = lambda f: f['format_id'] == format_spec
- matches = list(filter(filter_f, available_formats))
- if matches:
- return matches[-1]
- return None
+ stream = io.BytesIO(format_spec.encode('utf-8'))
+ try:
+ tokens = list(_remove_unused_ops(compat_tokenize_tokenize(stream.readline)))
+ except tokenize.TokenError:
+ raise syntax_error('Missing closing/opening brackets or parenthesis', (0, len(format_spec)))
+
+ class TokenIterator(object):
+ def __init__(self, tokens):
+ self.tokens = tokens
+ self.counter = 0
+
+ def __iter__(self):
+ return self
+
+ def __next__(self):
+ if self.counter >= len(self.tokens):
+ raise StopIteration()
+ value = self.tokens[self.counter]
+ self.counter += 1
+ return value
+
+ next = __next__
+
+ def restore_last_token(self):
+ self.counter -= 1
+
+ parsed_selector = _parse_format_selection(iter(TokenIterator(tokens)))
+ return _build_selector_function(parsed_selector)
def _calc_headers(self, info_dict):
res = std_headers.copy()
return res
def _calc_cookies(self, info_dict):
- pr = compat_urllib_request.Request(info_dict['url'])
+ pr = sanitized_Request(info_dict['url'])
self.cookiejar.add_cookie_header(pr)
return pr.get_header('Cookie')
t.get('preference'), t.get('width'), t.get('height'),
t.get('id'), t.get('url')))
for i, t in enumerate(thumbnails):
- if 'width' in t and 'height' in t:
+ 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
info_dict['display_id'] = info_dict['id']
if info_dict.get('upload_date') is None and info_dict.get('timestamp') is not None:
- # Working around negative timestamps in Windows
- # (see http://bugs.python.org/issue1646728)
- if info_dict['timestamp'] < 0 and os.name == 'nt':
- info_dict['timestamp'] = 0
- upload_date = datetime.datetime.utcfromtimestamp(
- info_dict['timestamp'])
- info_dict['upload_date'] = upload_date.strftime('%Y%m%d')
+ # Working around out-of-range timestamp values (e.g. negative ones on Windows,
+ # see http://bugs.python.org/issue1646728)
+ try:
+ upload_date = datetime.datetime.utcfromtimestamp(info_dict['timestamp'])
+ info_dict['upload_date'] = upload_date.strftime('%Y%m%d')
+ except (ValueError, OverflowError, OSError):
+ pass
+
+ # Auto generate title fields corresponding to the *_number fields when missing
+ # in order to always have clean titles. This is very common for TV series.
+ for field in ('chapter', 'season', 'episode'):
+ if info_dict.get('%s_number' % field) is not None and not info_dict.get(field):
+ info_dict[field] = '%s %d' % (field.capitalize(), info_dict['%s_number' % field])
+
+ subtitles = info_dict.get('subtitles')
+ if subtitles:
+ for _, subtitle in subtitles.items():
+ for subtitle_format in subtitle:
+ if 'ext' not in subtitle_format:
+ subtitle_format['ext'] = determine_ext(subtitle_format['url']).lower()
if self.params.get('listsubtitles', False):
if 'automatic_captions' in info_dict:
self.list_subtitles(info_dict['id'], info_dict.get('automatic_captions'), 'automatic captions')
- self.list_subtitles(info_dict['id'], info_dict.get('subtitles'), 'subtitles')
+ self.list_subtitles(info_dict['id'], subtitles, 'subtitles')
return
info_dict['requested_subtitles'] = self.process_subtitles(
- info_dict['id'], info_dict.get('subtitles'),
+ info_dict['id'], subtitles,
info_dict.get('automatic_captions'))
- # This extractors handle format selection themselves
- if info_dict['extractor'] in ['Youku']:
- if download:
- self.process_info(info_dict)
- return info_dict
-
# We now pick which formats have to be downloaded
if info_dict.get('formats') is None:
# There's only one format available
if not formats:
raise ExtractorError('No video formats found!')
+ formats_dict = {}
+
# We check that all the formats have the format and format_id fields
for i, format in enumerate(formats):
if 'url' not in format:
if format.get('format_id') is None:
format['format_id'] = compat_str(i)
+ else:
+ # Sanitize format_id from characters used in format selector expression
+ format['format_id'] = re.sub('[\s,/+\[\]()]', '_', format['format_id'])
+ format_id = format['format_id']
+ if format_id not in formats_dict:
+ formats_dict[format_id] = []
+ formats_dict[format_id].append(format)
+
+ # Make sure all formats have unique format_id
+ for format_id, ambiguous_formats in formats_dict.items():
+ if len(ambiguous_formats) > 1:
+ for i, format in enumerate(ambiguous_formats):
+ format['format_id'] = '%s-%d' % (format_id, i)
+
+ for i, format in enumerate(formats):
if format.get('format') is None:
format['format'] = '{id} - {res}{note}'.format(
id=format['format_id'],
# Automatically determine file extension if missing
if 'ext' not in format:
format['ext'] = determine_ext(format['url']).lower()
+ # Automatically determine protocol if missing (useful for format
+ # selection purposes)
+ if 'protocol' not in format:
+ format['protocol'] = determine_protocol(format)
# Add HTTP headers, so that external programs can use them from the
# json output
full_format_info = info_dict.copy()
# only set the 'formats' fields if the original info_dict list them
# otherwise we end up with a circular reference, the first (and unique)
# element in the 'formats' field in info_dict is info_dict itself,
- # wich can't be exported to json
+ # which can't be exported to json
info_dict['formats'] = formats
if self.params.get('listformats'):
self.list_formats(info_dict)
if req_format is None:
req_format_list = []
if (self.params.get('outtmpl', DEFAULT_OUTTMPL) != '-' and
- info_dict['extractor'] in ['youtube', 'ted']):
+ not info_dict.get('is_live')):
merger = FFmpegMergerPP(self)
if merger.available and merger.can_merge():
req_format_list.append('bestvideo+bestaudio')
req_format_list.append('best')
req_format = '/'.join(req_format_list)
- formats_to_download = []
- if req_format == 'all':
- formats_to_download = formats
- else:
- for rfstr in req_format.split(','):
- # We can accept formats requested in the format: 34/5/best, we pick
- # the first that is available, starting from left
- req_formats = rfstr.split('/')
- for rf in req_formats:
- if re.match(r'.+?\+.+?', rf) is not None:
- # Two formats have been requested like '137+139'
- format_1, format_2 = rf.split('+')
- formats_info = (self.select_format(format_1, formats),
- self.select_format(format_2, formats))
- if all(formats_info):
- # The first format must contain the video and the
- # second the audio
- if formats_info[0].get('vcodec') == 'none':
- self.report_error('The first format must '
- 'contain the video, try using '
- '"-f %s+%s"' % (format_2, format_1))
- return
- output_ext = (
- formats_info[0]['ext']
- if self.params.get('merge_output_format') is None
- else self.params['merge_output_format'])
- selected_format = {
- 'requested_formats': formats_info,
- 'format': '%s+%s' % (formats_info[0].get('format'),
- formats_info[1].get('format')),
- 'format_id': '%s+%s' % (formats_info[0].get('format_id'),
- formats_info[1].get('format_id')),
- 'width': formats_info[0].get('width'),
- 'height': formats_info[0].get('height'),
- 'resolution': formats_info[0].get('resolution'),
- 'fps': formats_info[0].get('fps'),
- 'vcodec': formats_info[0].get('vcodec'),
- 'vbr': formats_info[0].get('vbr'),
- 'stretched_ratio': formats_info[0].get('stretched_ratio'),
- 'acodec': formats_info[1].get('acodec'),
- 'abr': formats_info[1].get('abr'),
- 'ext': output_ext,
- }
- else:
- selected_format = None
- else:
- selected_format = self.select_format(rf, formats)
- if selected_format is not None:
- formats_to_download.append(selected_format)
- break
+ format_selector = self.build_format_selector(req_format)
+ formats_to_download = list(format_selector(formats))
if not formats_to_download:
raise ExtractorError('requested format not available',
expected=True)
if dn and not os.path.exists(dn):
os.makedirs(dn)
except (OSError, IOError) as err:
- self.report_error('unable to create directory ' + compat_str(err))
+ self.report_error('unable to create directory ' + error_to_compat_str(err))
return
if self.params.get('writedescription', False):
sub_info['url'], info_dict['id'], note=False)
except ExtractorError as err:
self.report_warning('Unable to download subtitle for "%s": %s' %
- (sub_lang, compat_str(err.cause)))
+ (sub_lang, error_to_compat_str(err.cause)))
continue
try:
sub_filename = subtitles_filename(filename, sub_lang, sub_format)
postprocessors = []
self.report_warning('You have requested multiple '
'formats but ffmpeg or avconv are not installed.'
- ' The formats won\'t be merged')
+ ' The formats won\'t be merged.')
else:
postprocessors = [merger]
requested_formats = info_dict['requested_formats']
if self.params.get('merge_output_format') is None and not compatible_formats(requested_formats):
info_dict['ext'] = 'mkv'
- self.report_warning('You have requested formats incompatible for merge. '
- 'The formats will be merged into mkv')
+ self.report_warning(
+ 'Requested formats are incompatible for merge and will be merged into mkv.')
# Ensure filename always has a correct extension for successful merge
filename = '%s.%s' % (filename_wo_ext, info_dict['ext'])
if os.path.exists(encodeFilename(filename)):
for url in url_list:
try:
# It also downloads the videos
- res = self.extract_info(url)
+ 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:
pps_chain.extend(ie_info['__postprocessors'])
pps_chain.extend(self._pps)
for pp in pps_chain:
+ files_to_delete = []
try:
files_to_delete, info = pp.run(info)
except PostProcessingError as e:
else:
res = '%sp' % format['height']
elif format.get('width') is not None:
- res = '?x%d' % format['width']
+ res = '%dx?' % format['width']
else:
res = default
return res
res = ''
if fdict.get('ext') in ['f4f', 'f4m']:
res += '(unsupported) '
+ if fdict.get('language'):
+ if res:
+ res += ' '
+ res += '[%s]' % fdict['language']
if fdict.get('format_note') is not None:
res += fdict['format_note'] + ' '
if fdict.get('tbr') is not None:
def urlopen(self, req):
""" Start an HTTP download """
-
- # According to RFC 3986, URLs can not contain non-ASCII characters, however this is not
- # always respected by websites, some tend to give out URLs with non percent-encoded
- # non-ASCII characters (see telemb.py, ard.py [#3412])
- # urllib chokes on URLs with non-ASCII characters (see http://bugs.python.org/issue3991)
- # To work around aforementioned issue we will replace request's original URL with
- # percent-encoded one
- req_is_string = isinstance(req, compat_basestring)
- url = req if req_is_string else req.get_full_url()
- url_escaped = escape_url(url)
-
- # Substitute URL if any change after escaping
- if url != url_escaped:
- if req_is_string:
- req = url_escaped
- else:
- req = compat_urllib_request.Request(
- url_escaped, data=req.data, headers=req.headers,
- origin_req_host=req.origin_req_host, unverifiable=req.unverifiable)
-
+ if isinstance(req, compat_basestring):
+ req = sanitized_Request(req)
return self._opener.open(req, timeout=self._socket_timeout)
def print_debug_header(self):
if os.access(opts_cookiefile, os.R_OK):
self.cookiejar.load()
- cookie_processor = compat_urllib_request.HTTPCookieProcessor(
- self.cookiejar)
+ cookie_processor = YoutubeDLCookieProcessor(self.cookiejar)
if opts_proxy is not None:
if opts_proxy == '':
proxies = {}
debuglevel = 1 if self.params.get('debug_printtraffic') else 0
https_handler = make_HTTPS_handler(self.params, debuglevel=debuglevel)
ydlh = YoutubeDLHandler(self.params, debuglevel=debuglevel)
+ data_handler = compat_urllib_request_DataHandler()
+
+ # When passing our own FileHandler instance, build_opener won't add the
+ # default FileHandler and allows us to disable the file protocol, which
+ # can be used for malicious purposes (see
+ # https://github.com/rg3/youtube-dl/issues/8227)
+ file_handler = compat_urllib_request.FileHandler()
+
+ def file_open(*args, **kwargs):
+ raise compat_urllib_error.URLError('file:// scheme is explicitly disabled in youtube-dl for security reasons')
+ file_handler.file_open = file_open
+
opener = compat_urllib_request.build_opener(
- proxy_handler, https_handler, cookie_processor, ydlh)
+ proxy_handler, https_handler, cookie_processor, ydlh, data_handler, file_handler)
# Delete the default user-agent header, which would otherwise apply in
# cases where our custom HTTP handler doesn't come into play
(info_dict['extractor'], info_dict['id'], thumb_display_id))
try:
uf = self.urlopen(t['url'])
- with open(thumb_filename, 'wb') as thumbf:
+ with open(encodeFilename(thumb_filename), 'wb') as thumbf:
shutil.copyfileobj(uf, thumbf)
self.to_screen('[%s] %s: Writing thumbnail %sto: %s' %
(info_dict['extractor'], info_dict['id'], thumb_display_id, thumb_filename))
except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
self.report_warning('Unable to download thumbnail "%s": %s' %
- (t['url'], compat_str(err)))
+ (t['url'], error_to_compat_str(err)))
Rapsys Git / youtubedl / blobdiff