from __future__ import absolute_import, unicode_literals
import collections
+import contextlib
import datetime
import errno
+import fileinput
import io
import itertools
import json
compat_basestring,
compat_cookiejar,
compat_expanduser,
+ compat_get_terminal_size,
compat_http_client,
compat_kwargs,
compat_str,
ExtractorError,
format_bytes,
formatSeconds,
- get_term_width,
locked_file,
make_HTTPS_handler,
MaxDownloadsReached,
PagedList,
parse_filesize,
+ PerRequestProxyHandler,
PostProcessingError,
platform_name,
preferredencoding,
render_table,
SameFileError,
sanitize_filename,
+ sanitize_path,
std_headers,
subtitles_filename,
- takewhile_inclusive,
UnavailableVideoError,
url_basename,
version_tuple,
write_string,
YoutubeDLHandler,
prepend_extension,
+ replace_extension,
args_to_str,
age_restricted,
)
(or video) as a single JSON line.
simulate: Do not download the video files.
format: Video format code. See options.py for more information.
- format_limit: Highest quality format to try.
outtmpl: Template for output names.
restrictfilenames: Do not allow "&" and spaces in file names
ignoreerrors: Do not stop on download errors.
prefer_insecure: Use HTTP instead of HTTPS to retrieve information.
At the moment, this is only supported by YouTube.
proxy: URL of the proxy server to use
+ cn_verification_proxy: URL of the proxy to use for IP address verification
+ on Chinese sites. (Experimental)
socket_timeout: Time to wait for unresponsive hosts, in seconds
bidi_workaround: Work around buggy terminals without bidirectional text
support, using fridibi
postprocessor.
progress_hooks: A list of functions that get called on download
progress, with a dictionary with the entries
- * status: One of "downloading" and "finished".
+ * status: One of "downloading", "error", or "finished".
Check this first and ignore unknown values.
- If status is one of "downloading" or "finished", the
+ If status is one of "downloading", or "finished", the
following properties may also be present:
* filename: The final filename (always present)
+ * tmpfilename: The filename we're currently writing to
* downloaded_bytes: Bytes on disk
* total_bytes: Size of the whole file, None if unknown
- * tmpfilename: The filename we're currently writing to
+ * total_bytes_estimate: Guess of the eventual file size,
+ None if unavailable.
+ * elapsed: The number of seconds since download started.
* eta: The estimated time in seconds, None if unknown
* speed: The download speed in bytes/second, None if
unknown
+ * fragment_index: The counter of the currently
+ downloaded video fragment.
+ * fragment_count: The number of fragments (= individual
+ files that will be merged)
Progress hooks are guaranteed to be called at least once
(with status "finished") if the download is successful.
call_home: Boolean, true iff we are allowed to contact the
youtube-dl servers for debugging.
sleep_interval: Number of seconds to sleep before each download.
- external_downloader: Executable of the external downloader to call.
listformats: Print an overview of available video formats and exit.
list_thumbnails: Print a table of all thumbnails and exit.
match_filter: A function that gets called with the info_dict of
match_filter_func in utils.py is one example for this.
no_color: Do not emit color codes in output.
+ The following options determine which downloader is picked:
+ external_downloader: Executable of the external downloader to call.
+ None or unset for standard (built-in) downloader.
+ hls_prefer_native: Use the native HLS downloader instead of ffmpeg/avconv.
The following parameters are not used by YoutubeDL itself, they are used by
- the FileDownloader:
+ 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.
+ xattr_set_filesize, external_downloader_args.
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.
- exec_cmd: Arbitrary command to run after downloading
"""
params = None
try:
import pty
master, slave = pty.openpty()
- width = get_term_width()
+ width = compat_get_terminal_size().columns
if width is None:
width_args = []
else:
raise
if (sys.version_info >= (3,) and sys.platform != 'win32' and
- sys.getfilesystemencoding() in ['ascii', 'ANSI_X3.4-1968']
- and not params.get('restrictfilenames', False)):
+ sys.getfilesystemencoding() in ['ascii', 'ANSI_X3.4-1968'] and
+ not params.get('restrictfilenames', False)):
# On Python 3, the Unicode filesystem API will throw errors (#1474)
self.report_warning(
'Assuming --restrict-filenames since file system encoding '
'Set the LC_ALL environment variable to fix this.')
self.params['restrictfilenames'] = True
- if '%(stitle)s' in self.params.get('outtmpl', ''):
- self.report_warning('%(stitle)s is deprecated. Use the %(title)s and the --restrict-filenames flag(which also secures %(uploader)s et al) instead.')
+ if isinstance(params.get('outtmpl'), bytes):
+ self.report_warning(
+ 'Parameter outtmpl is bytes, but should be a unicode string. '
+ 'Put from __future__ import unicode_literals at the top of your code file or consider switching to Python 3.x.')
self._setup_opener()
if v is not None)
template_dict = collections.defaultdict(lambda: 'NA', template_dict)
- outtmpl = self.params.get('outtmpl', DEFAULT_OUTTMPL)
+ outtmpl = sanitize_path(self.params.get('outtmpl', DEFAULT_OUTTMPL))
tmpl = compat_expanduser(outtmpl)
filename = tmpl % template_dict
# Temporary fix for #4787
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 ie_key:
ies = [self.get_info_extractor(ie_key)]
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)
if not available_formats:
return None
- if format_spec == 'best' or format_spec is None:
- return available_formats[-1]
- elif format_spec == 'worst':
- return available_formats[0]
+ 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
return res
def _calc_cookies(self, info_dict):
- class _PseudoRequest(object):
- def __init__(self, url):
- self.url = url
- self.headers = {}
- self.unverifiable = False
-
- def add_unredirected_header(self, k, v):
- self.headers[k] = v
-
- def get_full_url(self):
- return self.url
-
- def is_unverifiable(self):
- return self.unverifiable
-
- def has_header(self, h):
- return h in self.headers
-
- def get_header(self, h, default=None):
- return self.headers.get(h, default)
-
- pr = _PseudoRequest(info_dict['url'])
+ pr = compat_urllib_request.Request(info_dict['url'])
self.cookiejar.add_cookie_header(pr)
- return pr.headers.get('Cookie')
+ return pr.get_header('Cookie')
def process_video_result(self, info_dict, download=True):
assert info_dict.get('_type', 'video') == 'video'
full_format_info.update(format)
format['http_headers'] = self._calc_headers(full_format_info)
- format_limit = self.params.get('format_limit', None)
- if format_limit:
- formats = list(takewhile_inclusive(
- lambda f: f['format_id'] != format_limit, formats
- ))
-
# TODO Central sorting goes here
if formats[0] is not info_dict:
req_format = self.params.get('format')
if req_format is None:
- req_format = 'best'
+ req_format_list = []
+ if (self.params.get('outtmpl', DEFAULT_OUTTMPL) != '-' and
+ info_dict['extractor'] in ['youtube', 'ted']):
+ 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 = []
- # The -1 is for supporting YoutubeIE
- if req_format in ('-1', 'all'):
+ if req_format == 'all':
formats_to_download = formats
else:
for rfstr in req_format.split(','):
info_dict.update(formats_to_download[-1])
return info_dict
- def process_subtitles(self, video_id, available_subs, available_autocaps):
+ def process_subtitles(self, video_id, normal_subtitles, automatic_captions):
"""Select the requested subtitles and their format"""
- if available_autocaps and self.params.get('writeautomaticsub'):
- available_subs = available_subs.copy()
- for lang, cap_info in available_autocaps.items():
+ available_subs = {}
+ if normal_subtitles and self.params.get('writesubtitles'):
+ available_subs.update(normal_subtitles)
+ if automatic_captions and self.params.get('writeautomaticsub'):
+ for lang, cap_info in automatic_captions.items():
if lang not in available_subs:
available_subs[lang] = cap_info
if formats is None:
self.report_warning('%s subtitles not available for %s' % (lang, video_id))
continue
- if isinstance(formats, compat_str):
- # TODO: convert all IE with subtitles support to the new format
- # and remove this
- subs[lang] = {
- 'ext': formats_preference[0],
- 'data': formats,
- }
- continue
for ext in formats_preference:
if ext == 'best':
f = formats[-1]
if len(info_dict['title']) > 200:
info_dict['title'] = info_dict['title'][:197] + '...'
- # Keep for backwards compatibility
- info_dict['stitle'] = info_dict['title']
-
if 'format' not in info_dict:
info_dict['format'] = info_dict['ext']
return
try:
- dn = os.path.dirname(encodeFilename(filename))
+ dn = os.path.dirname(sanitize_path(encodeFilename(filename)))
if dn and not os.path.exists(dn):
os.makedirs(dn)
except (OSError, IOError) as err:
return
if self.params.get('writedescription', False):
- descfn = filename + '.description'
+ descfn = replace_extension(filename, 'description', info_dict.get('ext'))
if self.params.get('nooverwrites', False) and os.path.exists(encodeFilename(descfn)):
self.to_screen('[info] Video description is already present')
elif info_dict.get('description') is None:
return
if self.params.get('writeannotations', False):
- annofn = filename + '.annotations.xml'
+ annofn = replace_extension(filename, 'annotations.xml', info_dict.get('ext'))
if self.params.get('nooverwrites', False) and os.path.exists(encodeFilename(annofn)):
self.to_screen('[info] Video annotations are already present')
else:
# subtitles download errors are already managed as troubles in relevant IE
# that way it will silently go on when used with unsupporting IE
subtitles = info_dict['requested_subtitles']
+ ie = self.get_info_extractor(info_dict['extractor_key'])
for sub_lang, sub_info in subtitles.items():
sub_format = sub_info['ext']
if sub_info.get('data') is not None:
sub_data = sub_info['data']
else:
try:
- uf = self.urlopen(sub_info['url'])
- sub_data = uf.read().decode('utf-8')
- except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
+ sub_data = ie._download_webpage(
+ 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)))
+ (sub_lang, compat_str(err.cause)))
continue
try:
sub_filename = subtitles_filename(filename, sub_lang, sub_format)
return
if self.params.get('writeinfojson', False):
- infofn = os.path.splitext(filename)[0] + '.info.json'
+ infofn = replace_extension(filename, 'info.json', info_dict.get('ext'))
if self.params.get('nooverwrites', False) and os.path.exists(encodeFilename(infofn)):
self.to_screen('[info] Video description metadata is already present')
else:
self.to_screen('[info] Writing video description metadata as JSON to: ' + infofn)
try:
- write_json_file(info_dict, infofn)
+ write_json_file(self.filter_requested_info(info_dict), infofn)
except (OSError, IOError):
self.report_error('Cannot write metadata to JSON file ' + infofn)
return
if info_dict.get('requested_formats') is not None:
downloaded = []
success = True
- merger = FFmpegMergerPP(self, not self.params.get('keepvideo'))
- if not merger._executable:
+ merger = FFmpegMergerPP(self)
+ if not merger.available:
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]
- for f in info_dict['requested_formats']:
- new_info = dict(info_dict)
- new_info.update(f)
- fname = self.prepare_filename(new_info)
- fname = prepend_extension(fname, 'f%s' % f['format_id'])
- downloaded.append(fname)
- partial_success = dl(fname, new_info)
- success = success and partial_success
- info_dict['__postprocessors'] = postprocessors
- info_dict['__files_to_merge'] = downloaded
+
+ def compatible_formats(formats):
+ video, audio = formats
+ # Check extension
+ video_ext, audio_ext = audio.get('ext'), video.get('ext')
+ if video_ext and audio_ext:
+ COMPATIBLE_EXTS = (
+ ('mp3', 'mp4', 'm4a', 'm4p', 'm4b', 'm4r', 'm4v'),
+ ('webm')
+ )
+ for exts in COMPATIBLE_EXTS:
+ if video_ext in exts and audio_ext in exts:
+ return True
+ # TODO: Check acodec/vcodec
+ return False
+
+ filename_real_ext = os.path.splitext(filename)[1][1:]
+ filename_wo_ext = (
+ os.path.splitext(filename)[0]
+ if filename_real_ext == info_dict['ext']
+ else filename)
+ 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(
+ '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)):
+ self.to_screen(
+ '[download] %s has already been downloaded and '
+ 'merged' % filename)
+ else:
+ for f in requested_formats:
+ new_info = dict(info_dict)
+ new_info.update(f)
+ fname = self.prepare_filename(new_info)
+ fname = prepend_extension(fname, 'f%s' % f['format_id'], new_info['ext'])
+ downloaded.append(fname)
+ partial_success = dl(fname, new_info)
+ success = success and partial_success
+ info_dict['__postprocessors'] = postprocessors
+ info_dict['__files_to_merge'] = downloaded
else:
# Just a single file
success = dl(filename, info_dict)
"""Download a given list of URLs."""
outtmpl = self.params.get('outtmpl', DEFAULT_OUTTMPL)
if (len(url_list) > 1 and
- '%' not in outtmpl
- and self.params.get('max_downloads') != 1):
+ '%' not in outtmpl and
+ self.params.get('max_downloads') != 1):
raise SameFileError(outtmpl)
for url in url_list:
return self._download_retcode
def download_with_info_file(self, info_filename):
- with io.open(info_filename, 'r', encoding='utf-8') as f:
- info = json.load(f)
+ with contextlib.closing(fileinput.FileInput(
+ [info_filename], mode='r',
+ openhook=fileinput.hook_encoded('utf-8'))) as f:
+ # FileInput doesn't have a read method, we can't call json.load
+ info = self.filter_requested_info(json.loads('\n'.join(f)))
try:
self.process_ie_result(info, download=True)
except DownloadError:
raise
return self._download_retcode
+ @staticmethod
+ def filter_requested_info(info_dict):
+ return dict(
+ (k, v) for k, v in info_dict.items()
+ if k not in ['requested_formats', 'requested_subtitles'])
+
def post_process(self, filename, ie_info):
"""Run all the postprocessors on the given file."""
info = dict(ie_info)
pps_chain.extend(ie_info['__postprocessors'])
pps_chain.extend(self._pps)
for pp in pps_chain:
- keep_video = None
- old_filename = info['filepath']
try:
- keep_video_wish, info = pp.run(info)
- if keep_video_wish is not None:
- if keep_video_wish:
- keep_video = keep_video_wish
- elif keep_video is None:
- # No clear decision yet, let IE decide
- keep_video = keep_video_wish
+ files_to_delete, info = pp.run(info)
except PostProcessingError as e:
self.report_error(e.msg)
- if keep_video is False and not self.params.get('keepvideo', False):
- try:
+ if files_to_delete and not self.params.get('keepvideo', False):
+ for old_filename in files_to_delete:
self.to_screen('Deleting original file %s (pass -k to keep)' % old_filename)
- os.remove(encodeFilename(old_filename))
- except (IOError, OSError):
- self.report_warning('Unable to remove downloaded video file')
+ try:
+ os.remove(encodeFilename(old_filename))
+ except (IOError, OSError):
+ self.report_warning('Unable to remove downloaded original file')
def _make_archive_id(self, info_dict):
# Future-proof against any change in case
return res
def list_formats(self, info_dict):
- def line(format, idlen=20):
- return (('%-' + compat_str(idlen + 1) + 's%-10s%-12s%s') % (
- format['format_id'],
- format['ext'],
- self.format_resolution(format),
- self._format_note(format),
- ))
-
formats = info_dict.get('formats', [info_dict])
- idlen = max(len('format code'),
- max(len(f['format_id']) for f in formats))
- formats_s = [
- line(f, idlen) for f in formats
+ table = [
+ [f['format_id'], f['ext'], self.format_resolution(f), self._format_note(f)]
+ for f in formats
if f.get('preference') is None or f['preference'] >= -1000]
if len(formats) > 1:
- formats_s[-1] += (' ' if self._format_note(formats[-1]) else '') + '(best)'
+ table[-1][-1] += (' ' if table[-1][-1] else '') + '(best)'
- header_line = line({
- 'format_id': 'format code', 'ext': 'extension',
- 'resolution': 'resolution', 'format_note': 'note'}, idlen=idlen)
+ header_line = ['format code', 'extension', 'resolution', 'note']
self.to_screen(
- '[info] Available formats for %s:\n%s\n%s' %
- (info_dict['id'], header_line, '\n'.join(formats_s)))
+ '[info] Available formats for %s:\n%s' %
+ (info_dict['id'], render_table(header_line, table)))
def list_thumbnails(self, info_dict):
thumbnails = info_dict.get('thumbnails')
out = out.decode().strip()
if re.match('[0-9a-f]+', out):
self._write_string('[debug] Git HEAD: ' + out + '\n')
- except:
+ except Exception:
try:
sys.exc_clear()
- except:
+ except Exception:
pass
self._write_string('[debug] Python version %s - %s\n' % (
platform.python_version(), platform_name()))
- exe_versions = FFmpegPostProcessor.get_versions()
+ exe_versions = FFmpegPostProcessor.get_versions(self)
exe_versions['rtmpdump'] = rtmpdump_version()
exe_str = ', '.join(
'%s %s' % (exe, v)
# Set HTTPS proxy to HTTP one if given (https://github.com/rg3/youtube-dl/issues/805)
if 'http' in proxies and 'https' not in proxies:
proxies['https'] = proxies['http']
- proxy_handler = compat_urllib_request.ProxyHandler(proxies)
+ proxy_handler = PerRequestProxyHandler(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)
opener = compat_urllib_request.build_opener(
- https_handler, proxy_handler, cookie_processor, ydlh)
+ proxy_handler, https_handler, cookie_processor, ydlh)
+
# Delete the default user-agent header, which would otherwise apply in
# cases where our custom HTTP handler doesn't come into play
# (See https://github.com/rg3/youtube-dl/issues/1309 for details)
thumb_ext = determine_ext(t['url'], 'jpg')
suffix = '_%s' % t['id'] if len(thumbnails) > 1 else ''
thumb_display_id = '%s ' % t['id'] if len(thumbnails) > 1 else ''
- thumb_filename = os.path.splitext(filename)[0] + suffix + '.' + thumb_ext
+ t['filename'] = thumb_filename = os.path.splitext(filename)[0] + suffix + '.' + thumb_ext
if self.params.get('nooverwrites', False) and os.path.exists(encodeFilename(thumb_filename)):
self.to_screen('[%s] %s: Thumbnail %sis already present' %
projects / youtube-dl / blobdiff