encodeFilename,
ExtractorError,
format_bytes,
+ formatSeconds,
get_term_width,
locked_file,
make_HTTPS_handler,
subtitles_filename,
takewhile_inclusive,
UnavailableVideoError,
+ url_basename,
write_json_file,
write_string,
YoutubeDLHandler,
forcethumbnail: Force printing thumbnail URL.
forcedescription: Force printing description.
forcefilename: Force printing final filename.
+ forceduration: Force printing duration.
forcejson: Force printing info_dict as JSON.
simulate: Do not download the video files.
format: Video format code.
noplaylist: Download single video instead of a playlist if in doubt.
age_limit: An integer representing the user's age in years.
Unsuitable videos for the given age are skipped.
- download_archive: File name of a file where all downloads are recorded.
+ min_views: An integer representing the minimum view count the video
+ must have in order to not be skipped.
+ Videos without view count information are always
+ downloaded. None for no limit.
+ max_views: An integer representing the maximum view count.
+ Videos that are more popular than that are not
+ downloaded.
+ Videos without view count information are always
+ downloaded. None for no limit.
+ download_archive: File name of a file where all downloads are recorded.
Videos already present in the file are not downloaded
again.
cookiefile: File name where cookies should be read from and dumped to.
socket_timeout: Time to wait for unresponsive hosts, in seconds
bidi_workaround: Work around buggy terminals without bidirectional text
support, using fridibi
+ debug_printtraffic:Print out sent and received HTTP traffic
The following parameters are not used by YoutubeDL itself, they are used by
the FileDownloader:
def __init__(self, params=None):
"""Create a FileDownloader object with the given options."""
+ if params is None:
+ params = {}
self._ies = []
self._ies_instances = {}
self._pps = []
- self._fd_progress_hooks = []
+ self._progress_hooks = []
self._download_retcode = 0
self._num_downloads = 0
self._screen_file = [sys.stdout, sys.stderr][params.get('logtostderr', False)]
self._err_file = sys.stderr
- self.params = {} if params is None else params
+ self.params = params
if params.get('bidi_workaround', False):
try:
width_args = []
else:
width_args = ['-w', str(width)]
- self._fribidi = subprocess.Popen(
- ['fribidi', '-c', 'UTF-8'] + width_args,
+ sp_kwargs = dict(
stdin=subprocess.PIPE,
stdout=slave,
stderr=self._err_file)
- self._fribidi_channel = os.fdopen(master, 'rb')
+ try:
+ self._output_process = subprocess.Popen(
+ ['bidiv'] + width_args, **sp_kwargs
+ )
+ except OSError:
+ self._output_process = subprocess.Popen(
+ ['fribidi', '-c', 'UTF-8'] + width_args, **sp_kwargs)
+ self._output_channel = os.fdopen(master, 'rb')
except OSError as ose:
if ose.errno == 2:
self.report_warning(u'Could not find fribidi executable, ignoring --bidi-workaround . Make sure that fribidi is an executable file in one of the directories in your $PATH.')
self._pps.append(pp)
pp.set_downloader(self)
- def add_downloader_progress_hook(self, ph):
- """Add the progress hook to the file downloader"""
- self._fd_progress_hooks.append(ph)
+ def add_progress_hook(self, ph):
+ """Add the progress hook (currently only for the file downloader)"""
+ self._progress_hooks.append(ph)
def _bidi_workaround(self, message):
- if not hasattr(self, '_fribidi_channel'):
+ if not hasattr(self, '_output_channel'):
return message
+ assert hasattr(self, '_output_process')
assert type(message) == type(u'')
line_count = message.count(u'\n') + 1
- self._fribidi.stdin.write((message + u'\n').encode('utf-8'))
- self._fribidi.stdin.flush()
- res = u''.join(self._fribidi_channel.readline().decode('utf-8')
+ self._output_process.stdin.write((message + u'\n').encode('utf-8'))
+ self._output_process.stdin.flush()
+ res = u''.join(self._output_channel.readline().decode('utf-8')
for _ in range(line_count))
return res[:-len(u'\n')]
error_message = u'%s %s' % (_msg_header, message)
self.trouble(error_message, tb)
- def report_writedescription(self, descfn):
- """ Report that the description file is being written """
- self.to_screen(u'[info] Writing video description to: ' + descfn)
-
- def report_writesubtitles(self, sub_filename):
- """ Report that the subtitles file is being written """
- self.to_screen(u'[info] Writing video subtitles to: ' + sub_filename)
-
- def report_writeinfojson(self, infofn):
- """ Report that the metadata file has been written """
- self.to_screen(u'[info] Video description metadata as JSON to: ' + infofn)
-
- def report_writeannotations(self, annofn):
- """ Report that the annotations file has been written. """
- self.to_screen(u'[info] Writing video annotations to: ' + annofn)
-
def report_file_already_downloaded(self, file_name):
"""Report file has already been fully downloaded."""
try:
def _match_entry(self, info_dict):
""" Returns None iff the file should be downloaded """
+ video_title = info_dict.get('title', info_dict.get('id', u'video'))
if 'title' in info_dict:
# This can happen when we're just evaluating the playlist
title = info_dict['title']
matchtitle = self.params.get('matchtitle', False)
if matchtitle:
if not re.search(matchtitle, title, re.IGNORECASE):
- return u'[download] "' + title + '" title did not match pattern "' + matchtitle + '"'
+ return u'"' + title + '" title did not match pattern "' + matchtitle + '"'
rejecttitle = self.params.get('rejecttitle', False)
if rejecttitle:
if re.search(rejecttitle, title, re.IGNORECASE):
if date is not None:
dateRange = self.params.get('daterange', DateRange())
if date not in dateRange:
- return u'[download] %s upload date is not in range %s' % (date_from_str(date).isoformat(), dateRange)
+ return u'%s upload date is not in range %s' % (date_from_str(date).isoformat(), dateRange)
+ view_count = info_dict.get('view_count', None)
+ if view_count is not None:
+ min_views = self.params.get('min_views')
+ if min_views is not None and view_count < min_views:
+ return u'Skipping %s, because it has not reached minimum view count (%d/%d)' % (video_title, view_count, min_views)
+ max_views = self.params.get('max_views')
+ if max_views is not None and view_count > max_views:
+ return u'Skipping %s, because it has exceeded the maximum view count (%d/%d)' % (video_title, view_count, max_views)
age_limit = self.params.get('age_limit')
if age_limit is not None:
if age_limit < info_dict.get('age_limit', 0):
return u'Skipping "' + title + '" because it is age restricted'
if self.in_download_archive(info_dict):
- return (u'%s has already been recorded in archive'
- % info_dict.get('title', info_dict.get('id', u'video')))
+ return u'%s has already been recorded in archive' % video_title
return None
@staticmethod
{
'extractor': ie.IE_NAME,
'webpage_url': url,
+ 'webpage_url_basename': url_basename(url),
'extractor_key': ie.ie_key(),
})
if process:
def make_result(embedded_info):
new_result = ie_result.copy()
for f in ('_type', 'url', 'ext', 'player_url', 'formats',
- 'entries', 'urlhandle', 'ie_key', 'duration',
+ 'entries', 'ie_key', 'duration',
'subtitles', 'annotations', 'format',
'thumbnail', 'thumbnails'):
if f in new_result:
n_all_entries = len(ie_result['entries'])
playliststart = self.params.get('playliststart', 1) - 1
- playlistend = self.params.get('playlistend', -1)
-
+ playlistend = self.params.get('playlistend', None)
+ # For backwards compatibility, interpret -1 as whole list
if playlistend == -1:
- entries = ie_result['entries'][playliststart:]
- else:
- entries = ie_result['entries'][playliststart:playlistend]
+ playlistend = None
+ entries = ie_result['entries'][playliststart:playlistend]
n_entries = len(entries)
- self.to_screen(u"[%s] playlist '%s': Collected %d video ids (downloading %d of them)" %
+ self.to_screen(
+ u"[%s] playlist '%s': Collected %d video ids (downloading %d of them)" %
(ie_result['extractor'], playlist, n_all_entries, n_entries))
for i, entry in enumerate(entries, 1):
'playlist_index': i + playliststart,
'extractor': ie_result['extractor'],
'webpage_url': ie_result['webpage_url'],
+ 'webpage_url_basename': url_basename(ie_result['webpage_url']),
'extractor_key': ie_result['extractor_key'],
}
{
'extractor': ie_result['extractor'],
'webpage_url': ie_result['webpage_url'],
+ 'webpage_url_basename': url_basename(ie_result['webpage_url']),
'extractor_key': ie_result['extractor_key'],
})
return r
info_dict['playlist_index'] = None
# This extractors handle format selection themselves
- if info_dict['extractor'] in [u'youtube', u'Youku']:
+ if info_dict['extractor'] in [u'Youku']:
if download:
self.process_info(info_dict)
return info_dict
if 'ext' not in format:
format['ext'] = determine_ext(format['url'])
- if self.params.get('listformats', None):
- self.list_formats(info_dict)
- return
-
format_limit = self.params.get('format_limit', None)
if format_limit:
formats = list(takewhile_inclusive(
lambda f: f['format_id'] != format_limit, formats
))
- if self.params.get('prefer_free_formats'):
- def _free_formats_key(f):
- try:
- ext_ord = [u'flv', u'mp4', u'webm'].index(f['ext'])
- except ValueError:
- ext_ord = -1
- # We only compare the extension if they have the same height and width
- return (f.get('height'), f.get('width'), ext_ord)
- formats = sorted(formats, key=_free_formats_key)
+
+ # TODO Central sorting goes here
+
+ if formats[0] is not info_dict:
+ # only set the 'formats' fields if the original info_dict list them
+ # otherwise we end up with a circular reference, the first (and unique)
+ # element in the 'formats' field in info_dict is info_dict itself,
+ # wich can't be exported to json
+ info_dict['formats'] = formats
+ if self.params.get('listformats', None):
+ self.list_formats(info_dict)
+ return
req_format = self.params.get('format', 'best')
if req_format is None:
self.to_stdout(info_dict['description'])
if self.params.get('forcefilename', False) and filename is not None:
self.to_stdout(filename)
+ if self.params.get('forceduration', False) and info_dict.get('duration') is not None:
+ self.to_stdout(formatSeconds(info_dict['duration']))
if self.params.get('forceformat', False):
self.to_stdout(info_dict['format'])
if self.params.get('forcejson', False):
return
if self.params.get('writedescription', False):
- try:
- descfn = filename + u'.description'
- self.report_writedescription(descfn)
- with io.open(encodeFilename(descfn), 'w', encoding='utf-8') as descfile:
- descfile.write(info_dict['description'])
- except (KeyError, TypeError):
- self.report_warning(u'There\'s no description to write.')
- except (OSError, IOError):
- self.report_error(u'Cannot write description file ' + descfn)
- return
+ descfn = filename + u'.description'
+ if self.params.get('nooverwrites', False) and os.path.exists(encodeFilename(descfn)):
+ self.to_screen(u'[info] Video description is already present')
+ else:
+ try:
+ self.to_screen(u'[info] Writing video description to: ' + descfn)
+ with io.open(encodeFilename(descfn), 'w', encoding='utf-8') as descfile:
+ descfile.write(info_dict['description'])
+ except (KeyError, TypeError):
+ self.report_warning(u'There\'s no description to write.')
+ except (OSError, IOError):
+ self.report_error(u'Cannot write description file ' + descfn)
+ return
if self.params.get('writeannotations', False):
- try:
- annofn = filename + u'.annotations.xml'
- self.report_writeannotations(annofn)
- with io.open(encodeFilename(annofn), 'w', encoding='utf-8') as annofile:
- annofile.write(info_dict['annotations'])
- except (KeyError, TypeError):
- self.report_warning(u'There are no annotations to write.')
- except (OSError, IOError):
- self.report_error(u'Cannot write annotations file: ' + annofn)
- return
+ annofn = filename + u'.annotations.xml'
+ if self.params.get('nooverwrites', False) and os.path.exists(encodeFilename(annofn)):
+ self.to_screen(u'[info] Video annotations are already present')
+ else:
+ try:
+ self.to_screen(u'[info] Writing video annotations to: ' + annofn)
+ with io.open(encodeFilename(annofn), 'w', encoding='utf-8') as annofile:
+ annofile.write(info_dict['annotations'])
+ except (KeyError, TypeError):
+ self.report_warning(u'There are no annotations to write.')
+ except (OSError, IOError):
+ self.report_error(u'Cannot write annotations file: ' + annofn)
+ return
subtitles_are_requested = any([self.params.get('writesubtitles', False),
self.params.get('writeautomaticsub')])
continue
try:
sub_filename = subtitles_filename(filename, sub_lang, sub_format)
- self.report_writesubtitles(sub_filename)
- with io.open(encodeFilename(sub_filename), 'w', encoding='utf-8') as subfile:
- subfile.write(sub)
+ if self.params.get('nooverwrites', False) and os.path.exists(encodeFilename(sub_filename)):
+ self.to_screen(u'[info] Video subtitle %s.%s is already_present' % (sub_lang, sub_format))
+ else:
+ self.to_screen(u'[info] Writing video subtitles to: ' + sub_filename)
+ with io.open(encodeFilename(sub_filename), 'w', encoding='utf-8') as subfile:
+ subfile.write(sub)
except (OSError, IOError):
self.report_error(u'Cannot write subtitles file ' + descfn)
return
if self.params.get('writeinfojson', False):
infofn = os.path.splitext(filename)[0] + u'.info.json'
- self.report_writeinfojson(infofn)
- try:
- json_info_dict = dict((k, v) for k, v in info_dict.items() if not k in ['urlhandle'])
- write_json_file(json_info_dict, encodeFilename(infofn))
- except (OSError, IOError):
- self.report_error(u'Cannot write metadata to JSON file ' + infofn)
- return
+ if self.params.get('nooverwrites', False) and os.path.exists(encodeFilename(infofn)):
+ self.to_screen(u'[info] Video description metadata is already present')
+ else:
+ self.to_screen(u'[info] Writing video description metadata as JSON to: ' + infofn)
+ try:
+ write_json_file(info_dict, encodeFilename(infofn))
+ except (OSError, IOError):
+ self.report_error(u'Cannot write metadata to JSON file ' + infofn)
+ return
if self.params.get('writethumbnail', False):
if info_dict.get('thumbnail') is not None:
thumb_format = determine_ext(info_dict['thumbnail'], u'jpg')
thumb_filename = os.path.splitext(filename)[0] + u'.' + thumb_format
- self.to_screen(u'[%s] %s: Downloading thumbnail ...' %
- (info_dict['extractor'], info_dict['id']))
- try:
- uf = compat_urllib_request.urlopen(info_dict['thumbnail'])
- with open(thumb_filename, 'wb') as thumbf:
- shutil.copyfileobj(uf, thumbf)
- self.to_screen(u'[%s] %s: Writing thumbnail to: %s' %
- (info_dict['extractor'], info_dict['id'], thumb_filename))
- except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
- self.report_warning(u'Unable to download thumbnail "%s": %s' %
- (info_dict['thumbnail'], compat_str(err)))
+ if self.params.get('nooverwrites', False) and os.path.exists(encodeFilename(thumb_filename)):
+ self.to_screen(u'[%s] %s: Thumbnail is already present' %
+ (info_dict['extractor'], info_dict['id']))
+ else:
+ self.to_screen(u'[%s] %s: Downloading thumbnail ...' %
+ (info_dict['extractor'], info_dict['id']))
+ try:
+ uf = compat_urllib_request.urlopen(info_dict['thumbnail'])
+ with open(thumb_filename, 'wb') as thumbf:
+ shutil.copyfileobj(uf, thumbf)
+ self.to_screen(u'[%s] %s: Writing thumbnail to: %s' %
+ (info_dict['extractor'], info_dict['id'], thumb_filename))
+ except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
+ self.report_warning(u'Unable to download thumbnail "%s": %s' %
+ (info_dict['thumbnail'], compat_str(err)))
if not self.params.get('skip_download', False):
if self.params.get('nooverwrites', False) and os.path.exists(encodeFilename(filename)):
else:
try:
fd = get_suitable_downloader(info_dict)(self, self.params)
- for ph in self._fd_progress_hooks:
+ for ph in self._progress_hooks:
fd.add_progress_hook(ph)
success = fd.download(filename, info_dict)
except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
def format_resolution(format, default='unknown'):
if format.get('vcodec') == 'none':
return 'audio only'
- if format.get('_resolution') is not None:
- return format['_resolution']
+ if format.get('resolution') is not None:
+ return format['resolution']
if format.get('height') is not None:
if format.get('width') is not None:
res = u'%sx%s' % (format['width'], format['height'])
else:
res = u'%sp' % format['height']
+ elif format.get('width') is not None:
+ res = u'?x%d' % format['width']
else:
res = default
return res
def list_formats(self, info_dict):
def format_note(fdict):
res = u''
+ if f.get('ext') in ['f4f', 'f4m']:
+ res += u'(unsupported) '
if fdict.get('format_note') is not None:
res += fdict['format_note'] + u' '
+ if fdict.get('tbr') is not None:
+ res += u'%4dk ' % fdict['tbr']
if (fdict.get('vcodec') is not None and
fdict.get('vcodec') != 'none'):
- res += u'%-5s' % fdict['vcodec']
- elif fdict.get('vbr') is not None:
- res += u'video'
+ res += u'%-5s@' % fdict['vcodec']
+ elif fdict.get('vbr') is not None and fdict.get('abr') is not None:
+ res += u'video@'
if fdict.get('vbr') is not None:
- res += u'@%4dk' % fdict['vbr']
+ res += u'%4dk' % fdict['vbr']
if fdict.get('acodec') is not None:
if res:
res += u', '
header_line = line({
'format_id': u'format code', 'ext': u'extension',
- '_resolution': u'resolution', 'format_note': u'note'}, idlen=idlen)
+ 'resolution': u'resolution', 'format_note': u'note'}, idlen=idlen)
self.to_screen(u'[info] Available formats for %s:\n%s\n%s' %
(info_dict['id'], header_line, u"\n".join(formats_s)))
if 'http' in proxies and 'https' not in proxies:
proxies['https'] = proxies['http']
proxy_handler = compat_urllib_request.ProxyHandler(proxies)
+
+ debuglevel = 1 if self.params.get('debug_printtraffic') else 0
https_handler = make_HTTPS_handler(
- self.params.get('nocheckcertificate', False))
+ self.params.get('nocheckcertificate', False), debuglevel=debuglevel)
+ ydlh = YoutubeDLHandler(debuglevel=debuglevel)
opener = compat_urllib_request.build_opener(
- https_handler, proxy_handler, cookie_processor, YoutubeDLHandler())
+ https_handler, proxy_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)