4 from __future__ import absolute_import, unicode_literals
29 from string import ascii_letters
34 compat_get_terminal_size,
40 compat_tokenize_tokenize,
42 compat_urllib_request,
43 compat_urllib_request_DataHandler,
71 PerRequestProxyHandler,
76 register_socks_protocols,
87 UnavailableVideoError,
93 YoutubeDLCookieProcessor,
95 YoutubeDLRedirectHandler,
97 from .cache import Cache
98 from .extractor import get_info_extractor, gen_extractor_classes, _LAZY_LOADER
99 from .extractor.openload import PhantomJSwrapper
100 from .downloader import get_suitable_downloader
101 from .downloader.rtmp import rtmpdump_version
102 from .postprocessor import (
105 FFmpegFixupStretchedPP,
110 from .version import __version__
112 if compat_os_name == 'nt':
116 class YoutubeDL(object):
119 YoutubeDL objects are the ones responsible of downloading the
120 actual video file and writing it to disk if the user has requested
121 it, among some other tasks. In most cases there should be one per
122 program. As, given a video URL, the downloader doesn't know how to
123 extract all the needed information, task that InfoExtractors do, it
124 has to pass the URL to one of them.
126 For this, YoutubeDL objects have a method that allows
127 InfoExtractors to be registered in a given order. When it is passed
128 a URL, the YoutubeDL object handles it to the first InfoExtractor it
129 finds that reports being able to handle it. The InfoExtractor extracts
130 all the information about the video or videos the URL refers to, and
131 YoutubeDL process the extracted information, possibly using a File
132 Downloader to download the video.
134 YoutubeDL objects accept a lot of parameters. In order not to saturate
135 the object constructor with arguments, it receives a dictionary of
136 options instead. These options are available through the params
137 attribute for the InfoExtractors to use. The YoutubeDL also
138 registers itself as the downloader in charge for the InfoExtractors
139 that are added to it, so this is a "mutual registration".
143 username: Username for authentication purposes.
144 password: Password for authentication purposes.
145 videopassword: Password for accessing a video.
146 ap_mso: Adobe Pass multiple-system operator identifier.
147 ap_username: Multiple-system operator account username.
148 ap_password: Multiple-system operator account password.
149 usenetrc: Use netrc for authentication instead.
150 verbose: Print additional info to stdout.
151 quiet: Do not print messages to stdout.
152 no_warnings: Do not print out anything for warnings.
153 forceurl: Force printing final URL.
154 forcetitle: Force printing title.
155 forceid: Force printing ID.
156 forcethumbnail: Force printing thumbnail URL.
157 forcedescription: Force printing description.
158 forcefilename: Force printing final filename.
159 forceduration: Force printing duration.
160 forcejson: Force printing info_dict as JSON.
161 dump_single_json: Force printing the info_dict of the whole playlist
162 (or video) as a single JSON line.
163 simulate: Do not download the video files.
164 format: Video format code. See options.py for more information.
165 outtmpl: Template for output names.
166 outtmpl_na_placeholder: Placeholder for unavailable meta fields.
167 restrictfilenames: Do not allow "&" and spaces in file names
168 ignoreerrors: Do not stop on download errors.
169 force_generic_extractor: Force downloader to use the generic extractor
170 nooverwrites: Prevent overwriting files.
171 playliststart: Playlist item to start at.
172 playlistend: Playlist item to end at.
173 playlist_items: Specific indices of playlist to download.
174 playlistreverse: Download playlist items in reverse order.
175 playlistrandom: Download playlist items in random order.
176 matchtitle: Download only matching titles.
177 rejecttitle: Reject downloads for matching titles.
178 logger: Log messages to a logging.Logger instance.
179 logtostderr: Log messages to stderr instead of stdout.
180 writedescription: Write the video description to a .description file
181 writeinfojson: Write the video description to a .info.json file
182 writeannotations: Write the video annotations to a .annotations.xml file
183 writethumbnail: Write the thumbnail image to a file
184 write_all_thumbnails: Write all thumbnail formats to files
185 writesubtitles: Write the video subtitles to a file
186 writeautomaticsub: Write the automatically generated subtitles to a file
187 allsubtitles: Downloads all the subtitles of the video
188 (requires writesubtitles or writeautomaticsub)
189 listsubtitles: Lists all available subtitles for the video
190 subtitlesformat: The format code for subtitles
191 subtitleslangs: List of languages of the subtitles to download
192 keepvideo: Keep the video file after post-processing
193 daterange: A DateRange object, download only if the upload_date is in the range.
194 skip_download: Skip the actual download of the video file
195 cachedir: Location of the cache files in the filesystem.
196 False to disable filesystem cache.
197 noplaylist: Download single video instead of a playlist if in doubt.
198 age_limit: An integer representing the user's age in years.
199 Unsuitable videos for the given age are skipped.
200 min_views: An integer representing the minimum view count the video
201 must have in order to not be skipped.
202 Videos without view count information are always
203 downloaded. None for no limit.
204 max_views: An integer representing the maximum view count.
205 Videos that are more popular than that are not
207 Videos without view count information are always
208 downloaded. None for no limit.
209 download_archive: File name of a file where all downloads are recorded.
210 Videos already present in the file are not downloaded
212 cookiefile: File name where cookies should be read from and dumped to.
213 nocheckcertificate:Do not verify SSL certificates
214 prefer_insecure: Use HTTP instead of HTTPS to retrieve information.
215 At the moment, this is only supported by YouTube.
216 proxy: URL of the proxy server to use
217 geo_verification_proxy: URL of the proxy to use for IP address verification
218 on geo-restricted sites.
219 socket_timeout: Time to wait for unresponsive hosts, in seconds
220 bidi_workaround: Work around buggy terminals without bidirectional text
221 support, using fridibi
222 debug_printtraffic:Print out sent and received HTTP traffic
223 include_ads: Download ads as well
224 default_search: Prepend this string if an input url is not valid.
225 'auto' for elaborate guessing
226 encoding: Use this encoding instead of the system-specified.
227 extract_flat: Do not resolve URLs, return the immediate result.
228 Pass in 'in_playlist' to only show this behavior for
230 postprocessors: A list of dictionaries, each with an entry
231 * key: The name of the postprocessor. See
232 youtube_dl/postprocessor/__init__.py for a list.
233 as well as any further keyword arguments for the
235 progress_hooks: A list of functions that get called on download
236 progress, with a dictionary with the entries
237 * status: One of "downloading", "error", or "finished".
238 Check this first and ignore unknown values.
240 If status is one of "downloading", or "finished", the
241 following properties may also be present:
242 * filename: The final filename (always present)
243 * tmpfilename: The filename we're currently writing to
244 * downloaded_bytes: Bytes on disk
245 * total_bytes: Size of the whole file, None if unknown
246 * total_bytes_estimate: Guess of the eventual file size,
248 * elapsed: The number of seconds since download started.
249 * eta: The estimated time in seconds, None if unknown
250 * speed: The download speed in bytes/second, None if
252 * fragment_index: The counter of the currently
253 downloaded video fragment.
254 * fragment_count: The number of fragments (= individual
255 files that will be merged)
257 Progress hooks are guaranteed to be called at least once
258 (with status "finished") if the download is successful.
259 merge_output_format: Extension to use when merging formats.
260 fixup: Automatically correct known faults of the file.
262 - "never": do nothing
263 - "warn": only emit a warning
264 - "detect_or_warn": check whether we can do anything
265 about it, warn otherwise (default)
266 source_address: Client-side IP address to bind to.
267 call_home: Boolean, true iff we are allowed to contact the
268 youtube-dl servers for debugging.
269 sleep_interval: Number of seconds to sleep before each download when
270 used alone or a lower bound of a range for randomized
271 sleep before each download (minimum possible number
272 of seconds to sleep) when used along with
274 max_sleep_interval:Upper bound of a range for randomized sleep before each
275 download (maximum possible number of seconds to sleep).
276 Must only be used along with sleep_interval.
277 Actual sleep time will be a random float from range
278 [sleep_interval; max_sleep_interval].
279 listformats: Print an overview of available video formats and exit.
280 list_thumbnails: Print a table of all thumbnails and exit.
281 match_filter: A function that gets called with the info_dict of
283 If it returns a message, the video is ignored.
284 If it returns None, the video is downloaded.
285 match_filter_func in utils.py is one example for this.
286 no_color: Do not emit color codes in output.
287 geo_bypass: Bypass geographic restriction via faking X-Forwarded-For
290 Two-letter ISO 3166-2 country code that will be used for
291 explicit geographic restriction bypassing via faking
292 X-Forwarded-For HTTP header
294 IP range in CIDR notation that will be used similarly to
297 The following options determine which downloader is picked:
298 external_downloader: Executable of the external downloader to call.
299 None or unset for standard (built-in) downloader.
300 hls_prefer_native: Use the native HLS downloader instead of ffmpeg/avconv
301 if True, otherwise use ffmpeg/avconv if False, otherwise
302 use downloader suggested by extractor if None.
304 The following parameters are not used by YoutubeDL itself, they are used by
305 the downloader (see youtube_dl/downloader/common.py):
306 nopart, updatetime, buffersize, ratelimit, min_filesize, max_filesize, test,
307 noresizebuffer, retries, continuedl, noprogress, consoletitle,
308 xattr_set_filesize, external_downloader_args, hls_use_mpegts,
311 The following options are used by the post processors:
312 prefer_ffmpeg: If False, use avconv instead of ffmpeg if both are available,
313 otherwise prefer ffmpeg.
314 ffmpeg_location: Location of the ffmpeg/avconv binary; either the path
315 to the binary or its containing directory.
316 postprocessor_args: A list of additional command-line arguments for the
319 The following options are used by the Youtube extractor:
320 youtube_include_dash_manifest: If True (default), DASH manifests and related
321 data will be downloaded and processed by extractor.
322 You can reduce network I/O by disabling it if you don't
326 _NUMERIC_FIELDS = set((
327 'width', 'height', 'tbr', 'abr', 'asr', 'vbr', 'fps', 'filesize', 'filesize_approx',
328 'timestamp', 'upload_year', 'upload_month', 'upload_day',
329 'duration', 'view_count', 'like_count', 'dislike_count', 'repost_count',
330 'average_rating', 'comment_count', 'age_limit',
331 'start_time', 'end_time',
332 'chapter_number', 'season_number', 'episode_number',
333 'track_number', 'disc_number', 'release_year',
340 _download_retcode = None
341 _num_downloads = None
343 _playlist_urls = set()
346 def __init__(self, params=None, auto_init=True):
347 """Create a FileDownloader object with the given options."""
351 self._ies_instances = {}
353 self._progress_hooks = []
354 self._download_retcode = 0
355 self._num_downloads = 0
356 self._screen_file = [sys.stdout, sys.stderr][params.get('logtostderr', False)]
357 self._err_file = sys.stderr
360 'nocheckcertificate': False,
362 self.params.update(params)
363 self.cache = Cache(self)
365 def check_deprecated(param, option, suggestion):
366 if self.params.get(param) is not None:
368 '%s is deprecated. Use %s instead.' % (option, suggestion))
372 if check_deprecated('cn_verification_proxy', '--cn-verification-proxy', '--geo-verification-proxy'):
373 if self.params.get('geo_verification_proxy') is None:
374 self.params['geo_verification_proxy'] = self.params['cn_verification_proxy']
376 check_deprecated('autonumber_size', '--autonumber-size', 'output template with %(autonumber)0Nd, where N in the number of digits')
377 check_deprecated('autonumber', '--auto-number', '-o "%(autonumber)s-%(title)s.%(ext)s"')
378 check_deprecated('usetitle', '--title', '-o "%(title)s-%(id)s.%(ext)s"')
380 if params.get('bidi_workaround', False):
383 master, slave = pty.openpty()
384 width = compat_get_terminal_size().columns
388 width_args = ['-w', str(width)]
390 stdin=subprocess.PIPE,
392 stderr=self._err_file)
394 self._output_process = subprocess.Popen(
395 ['bidiv'] + width_args, **sp_kwargs
398 self._output_process = subprocess.Popen(
399 ['fribidi', '-c', 'UTF-8'] + width_args, **sp_kwargs)
400 self._output_channel = os.fdopen(master, 'rb')
401 except OSError as ose:
402 if ose.errno == errno.ENOENT:
403 self.report_warning('Could not find fribidi executable, ignoring --bidi-workaround . Make sure that fribidi is an executable file in one of the directories in your $PATH.')
407 if (sys.platform != 'win32'
408 and sys.getfilesystemencoding() in ['ascii', 'ANSI_X3.4-1968']
409 and not params.get('restrictfilenames', False)):
410 # Unicode filesystem API will throw errors (#1474, #13027)
412 'Assuming --restrict-filenames since file system encoding '
413 'cannot encode all characters. '
414 'Set the LC_ALL environment variable to fix this.')
415 self.params['restrictfilenames'] = True
417 if isinstance(params.get('outtmpl'), bytes):
419 'Parameter outtmpl is bytes, but should be a unicode string. '
420 'Put from __future__ import unicode_literals at the top of your code file or consider switching to Python 3.x.')
425 self.print_debug_header()
426 self.add_default_info_extractors()
428 for pp_def_raw in self.params.get('postprocessors', []):
429 pp_class = get_postprocessor(pp_def_raw['key'])
430 pp_def = dict(pp_def_raw)
432 pp = pp_class(self, **compat_kwargs(pp_def))
433 self.add_post_processor(pp)
435 for ph in self.params.get('progress_hooks', []):
436 self.add_progress_hook(ph)
438 register_socks_protocols()
440 def warn_if_short_id(self, argv):
441 # short YouTube ID starting with dash?
443 i for i, a in enumerate(argv)
444 if re.match(r'^-[0-9A-Za-z_-]{10}$', a)]
448 + [a for i, a in enumerate(argv) if i not in idxs]
449 + ['--'] + [argv[i] for i in idxs]
452 'Long argument string detected. '
453 'Use -- to separate parameters and URLs, like this:\n%s\n' %
454 args_to_str(correct_argv))
456 def add_info_extractor(self, ie):
457 """Add an InfoExtractor object to the end of the list."""
459 if not isinstance(ie, type):
460 self._ies_instances[ie.ie_key()] = ie
461 ie.set_downloader(self)
463 def get_info_extractor(self, ie_key):
465 Get an instance of an IE with name ie_key, it will try to get one from
466 the _ies list, if there's no instance it will create a new one and add
467 it to the extractor list.
469 ie = self._ies_instances.get(ie_key)
471 ie = get_info_extractor(ie_key)()
472 self.add_info_extractor(ie)
475 def add_default_info_extractors(self):
477 Add the InfoExtractors returned by gen_extractors to the end of the list
479 for ie in gen_extractor_classes():
480 self.add_info_extractor(ie)
482 def add_post_processor(self, pp):
483 """Add a PostProcessor object to the end of the chain."""
485 pp.set_downloader(self)
487 def add_progress_hook(self, ph):
488 """Add the progress hook (currently only for the file downloader)"""
489 self._progress_hooks.append(ph)
491 def _bidi_workaround(self, message):
492 if not hasattr(self, '_output_channel'):
495 assert hasattr(self, '_output_process')
496 assert isinstance(message, compat_str)
497 line_count = message.count('\n') + 1
498 self._output_process.stdin.write((message + '\n').encode('utf-8'))
499 self._output_process.stdin.flush()
500 res = ''.join(self._output_channel.readline().decode('utf-8')
501 for _ in range(line_count))
502 return res[:-len('\n')]
504 def to_screen(self, message, skip_eol=False):
505 """Print message to stdout if not in quiet mode."""
506 return self.to_stdout(message, skip_eol, check_quiet=True)
508 def _write_string(self, s, out=None):
509 write_string(s, out=out, encoding=self.params.get('encoding'))
511 def to_stdout(self, message, skip_eol=False, check_quiet=False):
512 """Print message to stdout if not in quiet mode."""
513 if self.params.get('logger'):
514 self.params['logger'].debug(message)
515 elif not check_quiet or not self.params.get('quiet', False):
516 message = self._bidi_workaround(message)
517 terminator = ['\n', ''][skip_eol]
518 output = message + terminator
520 self._write_string(output, self._screen_file)
522 def to_stderr(self, message):
523 """Print message to stderr."""
524 assert isinstance(message, compat_str)
525 if self.params.get('logger'):
526 self.params['logger'].error(message)
528 message = self._bidi_workaround(message)
529 output = message + '\n'
530 self._write_string(output, self._err_file)
532 def to_console_title(self, message):
533 if not self.params.get('consoletitle', False):
535 if compat_os_name == 'nt':
536 if ctypes.windll.kernel32.GetConsoleWindow():
537 # c_wchar_p() might not be necessary if `message` is
538 # already of type unicode()
539 ctypes.windll.kernel32.SetConsoleTitleW(ctypes.c_wchar_p(message))
540 elif 'TERM' in os.environ:
541 self._write_string('\033]0;%s\007' % message, self._screen_file)
543 def save_console_title(self):
544 if not self.params.get('consoletitle', False):
546 if self.params.get('simulate', False):
548 if compat_os_name != 'nt' and 'TERM' in os.environ:
549 # Save the title on stack
550 self._write_string('\033[22;0t', self._screen_file)
552 def restore_console_title(self):
553 if not self.params.get('consoletitle', False):
555 if self.params.get('simulate', False):
557 if compat_os_name != 'nt' and 'TERM' in os.environ:
558 # Restore the title from stack
559 self._write_string('\033[23;0t', self._screen_file)
562 self.save_console_title()
565 def __exit__(self, *args):
566 self.restore_console_title()
568 if self.params.get('cookiefile') is not None:
569 self.cookiejar.save(ignore_discard=True, ignore_expires=True)
571 def trouble(self, message=None, tb=None):
572 """Determine action to take when a download problem appears.
574 Depending on if the downloader has been configured to ignore
575 download errors or not, this method may throw an exception or
576 not when errors are found, after printing the message.
578 tb, if given, is additional traceback information.
580 if message is not None:
581 self.to_stderr(message)
582 if self.params.get('verbose'):
584 if sys.exc_info()[0]: # if .trouble has been called from an except block
586 if hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]:
587 tb += ''.join(traceback.format_exception(*sys.exc_info()[1].exc_info))
588 tb += encode_compat_str(traceback.format_exc())
590 tb_data = traceback.format_list(traceback.extract_stack())
591 tb = ''.join(tb_data)
593 if not self.params.get('ignoreerrors', False):
594 if sys.exc_info()[0] and hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]:
595 exc_info = sys.exc_info()[1].exc_info
597 exc_info = sys.exc_info()
598 raise DownloadError(message, exc_info)
599 self._download_retcode = 1
601 def report_warning(self, message):
603 Print the message to stderr, it will be prefixed with 'WARNING:'
604 If stderr is a tty file the 'WARNING:' will be colored
606 if self.params.get('logger') is not None:
607 self.params['logger'].warning(message)
609 if self.params.get('no_warnings'):
611 if not self.params.get('no_color') and self._err_file.isatty() and compat_os_name != 'nt':
612 _msg_header = '\033[0;33mWARNING:\033[0m'
614 _msg_header = 'WARNING:'
615 warning_message = '%s %s' % (_msg_header, message)
616 self.to_stderr(warning_message)
618 def report_error(self, message, tb=None):
620 Do the same as trouble, but prefixes the message with 'ERROR:', colored
621 in red if stderr is a tty file.
623 if not self.params.get('no_color') and self._err_file.isatty() and compat_os_name != 'nt':
624 _msg_header = '\033[0;31mERROR:\033[0m'
626 _msg_header = 'ERROR:'
627 error_message = '%s %s' % (_msg_header, message)
628 self.trouble(error_message, tb)
630 def report_file_already_downloaded(self, file_name):
631 """Report file has already been fully downloaded."""
633 self.to_screen('[download] %s has already been downloaded' % file_name)
634 except UnicodeEncodeError:
635 self.to_screen('[download] The file has already been downloaded')
637 def prepare_filename(self, info_dict):
638 """Generate the output filename."""
640 template_dict = dict(info_dict)
642 template_dict['epoch'] = int(time.time())
643 autonumber_size = self.params.get('autonumber_size')
644 if autonumber_size is None:
646 template_dict['autonumber'] = self.params.get('autonumber_start', 1) - 1 + self._num_downloads
647 if template_dict.get('resolution') is None:
648 if template_dict.get('width') and template_dict.get('height'):
649 template_dict['resolution'] = '%dx%d' % (template_dict['width'], template_dict['height'])
650 elif template_dict.get('height'):
651 template_dict['resolution'] = '%sp' % template_dict['height']
652 elif template_dict.get('width'):
653 template_dict['resolution'] = '%dx?' % template_dict['width']
655 sanitize = lambda k, v: sanitize_filename(
657 restricted=self.params.get('restrictfilenames'),
658 is_id=(k == 'id' or k.endswith('_id')))
659 template_dict = dict((k, v if isinstance(v, compat_numeric_types) else sanitize(k, v))
660 for k, v in template_dict.items()
661 if v is not None and not isinstance(v, (list, tuple, dict)))
662 template_dict = collections.defaultdict(lambda: self.params.get('outtmpl_na_placeholder', 'NA'), template_dict)
664 outtmpl = self.params.get('outtmpl', DEFAULT_OUTTMPL)
666 # For fields playlist_index and autonumber convert all occurrences
667 # of %(field)s to %(field)0Nd for backward compatibility
668 field_size_compat_map = {
669 'playlist_index': len(str(template_dict['n_entries'])),
670 'autonumber': autonumber_size,
672 FIELD_SIZE_COMPAT_RE = r'(?<!%)%\((?P<field>autonumber|playlist_index)\)s'
673 mobj = re.search(FIELD_SIZE_COMPAT_RE, outtmpl)
676 FIELD_SIZE_COMPAT_RE,
677 r'%%(\1)0%dd' % field_size_compat_map[mobj.group('field')],
680 # Missing numeric fields used together with integer presentation types
681 # in format specification will break the argument substitution since
682 # string NA placeholder is returned for missing fields. We will patch
683 # output template for missing fields to meet string presentation type.
684 for numeric_field in self._NUMERIC_FIELDS:
685 if numeric_field not in template_dict:
686 # As of [1] format syntax is:
687 # %[mapping_key][conversion_flags][minimum_width][.precision][length_modifier]type
688 # 1. https://docs.python.org/2/library/stdtypes.html#string-formatting
692 \({0}\) # mapping key
693 (?:[#0\-+ ]+)? # conversion flags (optional)
694 (?:\d+)? # minimum field width (optional)
695 (?:\.\d+)? # precision (optional)
696 [hlL]? # length modifier (optional)
697 [diouxXeEfFgGcrs%] # conversion type
700 FORMAT_RE.format(numeric_field),
701 r'%({0})s'.format(numeric_field), outtmpl)
703 # expand_path translates '%%' into '%' and '$$' into '$'
704 # correspondingly that is not what we want since we need to keep
705 # '%%' intact for template dict substitution step. Working around
706 # with boundary-alike separator hack.
707 sep = ''.join([random.choice(ascii_letters) for _ in range(32)])
708 outtmpl = outtmpl.replace('%%', '%{0}%'.format(sep)).replace('$$', '${0}$'.format(sep))
710 # outtmpl should be expand_path'ed before template dict substitution
711 # because meta fields may contain env variables we don't want to
712 # be expanded. For example, for outtmpl "%(title)s.%(ext)s" and
713 # title "Hello $PATH", we don't want `$PATH` to be expanded.
714 filename = expand_path(outtmpl).replace(sep, '') % template_dict
716 # Temporary fix for #4787
717 # 'Treat' all problem characters by passing filename through preferredencoding
718 # to workaround encoding issues with subprocess on python2 @ Windows
719 if sys.version_info < (3, 0) and sys.platform == 'win32':
720 filename = encodeFilename(filename, True).decode(preferredencoding())
721 return sanitize_path(filename)
722 except ValueError as err:
723 self.report_error('Error in output template: ' + str(err) + ' (encoding: ' + repr(preferredencoding()) + ')')
726 def _match_entry(self, info_dict, incomplete):
727 """ Returns None iff the file should be downloaded """
729 video_title = info_dict.get('title', info_dict.get('id', 'video'))
730 if 'title' in info_dict:
731 # This can happen when we're just evaluating the playlist
732 title = info_dict['title']
733 matchtitle = self.params.get('matchtitle', False)
735 if not re.search(matchtitle, title, re.IGNORECASE):
736 return '"' + title + '" title did not match pattern "' + matchtitle + '"'
737 rejecttitle = self.params.get('rejecttitle', False)
739 if re.search(rejecttitle, title, re.IGNORECASE):
740 return '"' + title + '" title matched reject pattern "' + rejecttitle + '"'
741 date = info_dict.get('upload_date')
743 dateRange = self.params.get('daterange', DateRange())
744 if date not in dateRange:
745 return '%s upload date is not in range %s' % (date_from_str(date).isoformat(), dateRange)
746 view_count = info_dict.get('view_count')
747 if view_count is not None:
748 min_views = self.params.get('min_views')
749 if min_views is not None and view_count < min_views:
750 return 'Skipping %s, because it has not reached minimum view count (%d/%d)' % (video_title, view_count, min_views)
751 max_views = self.params.get('max_views')
752 if max_views is not None and view_count > max_views:
753 return 'Skipping %s, because it has exceeded the maximum view count (%d/%d)' % (video_title, view_count, max_views)
754 if age_restricted(info_dict.get('age_limit'), self.params.get('age_limit')):
755 return 'Skipping "%s" because it is age restricted' % video_title
756 if self.in_download_archive(info_dict):
757 return '%s has already been recorded in archive' % video_title
760 match_filter = self.params.get('match_filter')
761 if match_filter is not None:
762 ret = match_filter(info_dict)
769 def add_extra_info(info_dict, extra_info):
770 '''Set the keys from extra_info in info dict if they are missing'''
771 for key, value in extra_info.items():
772 info_dict.setdefault(key, value)
774 def extract_info(self, url, download=True, ie_key=None, extra_info={},
775 process=True, force_generic_extractor=False):
777 Return a list with a dictionary for each video extracted.
780 url -- URL to extract
783 download -- whether to download videos during extraction
784 ie_key -- extractor key hint
785 extra_info -- dictionary containing the extra values to add to each result
786 process -- whether to resolve all unresolved references (URLs, playlist items),
787 must be True for download to work.
788 force_generic_extractor -- force using the generic extractor
791 if not ie_key and force_generic_extractor:
795 ies = [self.get_info_extractor(ie_key)]
800 if not ie.suitable(url):
803 ie = self.get_info_extractor(ie.ie_key())
805 self.report_warning('The program functionality for this site has been marked as broken, '
806 'and will probably not work.')
808 return self.__extract_info(url, ie, download, extra_info, process)
810 self.report_error('no suitable InfoExtractor for URL %s' % url)
812 def __handle_extraction_exceptions(func):
813 def wrapper(self, *args, **kwargs):
815 return func(self, *args, **kwargs)
816 except GeoRestrictedError as e:
819 msg += '\nThis video is available in %s.' % ', '.join(
820 map(ISO3166Utils.short2full, e.countries))
821 msg += '\nYou might want to use a VPN or a proxy server (with --proxy) to workaround.'
822 self.report_error(msg)
823 except ExtractorError as e: # An error we somewhat expected
824 self.report_error(compat_str(e), e.format_traceback())
825 except MaxDownloadsReached:
827 except Exception as e:
828 if self.params.get('ignoreerrors', False):
829 self.report_error(error_to_compat_str(e), tb=encode_compat_str(traceback.format_exc()))
834 @__handle_extraction_exceptions
835 def __extract_info(self, url, ie, download, extra_info, process):
836 ie_result = ie.extract(url)
837 if ie_result is None: # Finished already (backwards compatibility; listformats and friends should be moved here)
839 if isinstance(ie_result, list):
840 # Backwards compatibility: old IE result format
842 '_type': 'compat_list',
843 'entries': ie_result,
845 self.add_default_extra_info(ie_result, ie, url)
847 return self.process_ie_result(ie_result, download, extra_info)
851 def add_default_extra_info(self, ie_result, ie, url):
852 self.add_extra_info(ie_result, {
853 'extractor': ie.IE_NAME,
855 'webpage_url_basename': url_basename(url),
856 'extractor_key': ie.ie_key(),
859 def process_ie_result(self, ie_result, download=True, extra_info={}):
861 Take the result of the ie(may be modified) and resolve all unresolved
862 references (URLs, playlist items).
864 It will also download the videos if 'download'.
865 Returns the resolved ie_result.
867 result_type = ie_result.get('_type', 'video')
869 if result_type in ('url', 'url_transparent'):
870 ie_result['url'] = sanitize_url(ie_result['url'])
871 extract_flat = self.params.get('extract_flat', False)
872 if ((extract_flat == 'in_playlist' and 'playlist' in extra_info)
873 or extract_flat is True):
874 self.__forced_printings(
875 ie_result, self.prepare_filename(ie_result),
879 if result_type == 'video':
880 self.add_extra_info(ie_result, extra_info)
881 return self.process_video_result(ie_result, download=download)
882 elif result_type == 'url':
883 # We have to add extra_info to the results because it may be
884 # contained in a playlist
885 return self.extract_info(ie_result['url'],
887 ie_key=ie_result.get('ie_key'),
888 extra_info=extra_info)
889 elif result_type == 'url_transparent':
890 # Use the information from the embedding page
891 info = self.extract_info(
892 ie_result['url'], ie_key=ie_result.get('ie_key'),
893 extra_info=extra_info, download=False, process=False)
895 # extract_info may return None when ignoreerrors is enabled and
896 # extraction failed with an error, don't crash and return early
901 force_properties = dict(
902 (k, v) for k, v in ie_result.items() if v is not None)
903 for f in ('_type', 'url', 'id', 'extractor', 'extractor_key', 'ie_key'):
904 if f in force_properties:
905 del force_properties[f]
906 new_result = info.copy()
907 new_result.update(force_properties)
909 # Extracted info may not be a video result (i.e.
910 # info.get('_type', 'video') != video) but rather an url or
911 # url_transparent. In such cases outer metadata (from ie_result)
912 # should be propagated to inner one (info). For this to happen
913 # _type of info should be overridden with url_transparent. This
914 # fixes issue from https://github.com/ytdl-org/youtube-dl/pull/11163.
915 if new_result.get('_type') == 'url':
916 new_result['_type'] = 'url_transparent'
918 return self.process_ie_result(
919 new_result, download=download, extra_info=extra_info)
920 elif result_type in ('playlist', 'multi_video'):
921 # Protect from infinite recursion due to recursively nested playlists
922 # (see https://github.com/ytdl-org/youtube-dl/issues/27833)
923 webpage_url = ie_result['webpage_url']
924 if webpage_url in self._playlist_urls:
926 '[download] Skipping already downloaded playlist: %s'
927 % ie_result.get('title') or ie_result.get('id'))
930 self._playlist_level += 1
931 self._playlist_urls.add(webpage_url)
933 return self.__process_playlist(ie_result, download)
935 self._playlist_level -= 1
936 if not self._playlist_level:
937 self._playlist_urls.clear()
938 elif result_type == 'compat_list':
940 'Extractor %s returned a compat_list result. '
941 'It needs to be updated.' % ie_result.get('extractor'))
947 'extractor': ie_result['extractor'],
948 'webpage_url': ie_result['webpage_url'],
949 'webpage_url_basename': url_basename(ie_result['webpage_url']),
950 'extractor_key': ie_result['extractor_key'],
954 ie_result['entries'] = [
955 self.process_ie_result(_fixup(r), download, extra_info)
956 for r in ie_result['entries']
960 raise Exception('Invalid result type: %s' % result_type)
962 def __process_playlist(self, ie_result, download):
963 # We process each entry in the playlist
964 playlist = ie_result.get('title') or ie_result.get('id')
966 self.to_screen('[download] Downloading playlist: %s' % playlist)
968 playlist_results = []
970 playliststart = self.params.get('playliststart', 1) - 1
971 playlistend = self.params.get('playlistend')
972 # For backwards compatibility, interpret -1 as whole list
973 if playlistend == -1:
976 playlistitems_str = self.params.get('playlist_items')
978 if playlistitems_str is not None:
979 def iter_playlistitems(format):
980 for string_segment in format.split(','):
981 if '-' in string_segment:
982 start, end = string_segment.split('-')
983 for item in range(int(start), int(end) + 1):
986 yield int(string_segment)
987 playlistitems = orderedSet(iter_playlistitems(playlistitems_str))
989 ie_entries = ie_result['entries']
991 def make_playlistitems_entries(list_ie_entries):
992 num_entries = len(list_ie_entries)
994 list_ie_entries[i - 1] for i in playlistitems
995 if -num_entries <= i - 1 < num_entries]
997 def report_download(num_entries):
999 '[%s] playlist %s: Downloading %d videos' %
1000 (ie_result['extractor'], playlist, num_entries))
1002 if isinstance(ie_entries, list):
1003 n_all_entries = len(ie_entries)
1005 entries = make_playlistitems_entries(ie_entries)
1007 entries = ie_entries[playliststart:playlistend]
1008 n_entries = len(entries)
1010 '[%s] playlist %s: Collected %d video ids (downloading %d of them)' %
1011 (ie_result['extractor'], playlist, n_all_entries, n_entries))
1012 elif isinstance(ie_entries, PagedList):
1015 for item in playlistitems:
1016 entries.extend(ie_entries.getslice(
1020 entries = ie_entries.getslice(
1021 playliststart, playlistend)
1022 n_entries = len(entries)
1023 report_download(n_entries)
1026 entries = make_playlistitems_entries(list(itertools.islice(
1027 ie_entries, 0, max(playlistitems))))
1029 entries = list(itertools.islice(
1030 ie_entries, playliststart, playlistend))
1031 n_entries = len(entries)
1032 report_download(n_entries)
1034 if self.params.get('playlistreverse', False):
1035 entries = entries[::-1]
1037 if self.params.get('playlistrandom', False):
1038 random.shuffle(entries)
1040 x_forwarded_for = ie_result.get('__x_forwarded_for_ip')
1042 for i, entry in enumerate(entries, 1):
1043 self.to_screen('[download] Downloading video %s of %s' % (i, n_entries))
1044 # This __x_forwarded_for_ip thing is a bit ugly but requires
1047 entry['__x_forwarded_for_ip'] = x_forwarded_for
1049 'n_entries': n_entries,
1050 'playlist': playlist,
1051 'playlist_id': ie_result.get('id'),
1052 'playlist_title': ie_result.get('title'),
1053 'playlist_uploader': ie_result.get('uploader'),
1054 'playlist_uploader_id': ie_result.get('uploader_id'),
1055 'playlist_index': playlistitems[i - 1] if playlistitems else i + playliststart,
1056 'extractor': ie_result['extractor'],
1057 'webpage_url': ie_result['webpage_url'],
1058 'webpage_url_basename': url_basename(ie_result['webpage_url']),
1059 'extractor_key': ie_result['extractor_key'],
1062 reason = self._match_entry(entry, incomplete=True)
1063 if reason is not None:
1064 self.to_screen('[download] ' + reason)
1067 entry_result = self.__process_iterable_entry(entry, download, extra)
1068 # TODO: skip failed (empty) entries?
1069 playlist_results.append(entry_result)
1070 ie_result['entries'] = playlist_results
1071 self.to_screen('[download] Finished downloading playlist: %s' % playlist)
1074 @__handle_extraction_exceptions
1075 def __process_iterable_entry(self, entry, download, extra_info):
1076 return self.process_ie_result(
1077 entry, download=download, extra_info=extra_info)
1079 def _build_format_filter(self, filter_spec):
1080 " Returns a function to filter the formats according to the filter_spec "
1090 operator_rex = re.compile(r'''(?x)\s*
1091 (?P<key>width|height|tbr|abr|vbr|asr|filesize|filesize_approx|fps)
1092 \s*(?P<op>%s)(?P<none_inclusive>\s*\?)?\s*
1093 (?P<value>[0-9.]+(?:[kKmMgGtTpPeEzZyY]i?[Bb]?)?)
1095 ''' % '|'.join(map(re.escape, OPERATORS.keys())))
1096 m = operator_rex.search(filter_spec)
1099 comparison_value = int(m.group('value'))
1101 comparison_value = parse_filesize(m.group('value'))
1102 if comparison_value is None:
1103 comparison_value = parse_filesize(m.group('value') + 'B')
1104 if comparison_value is None:
1106 'Invalid value %r in format specification %r' % (
1107 m.group('value'), filter_spec))
1108 op = OPERATORS[m.group('op')]
1113 '^=': lambda attr, value: attr.startswith(value),
1114 '$=': lambda attr, value: attr.endswith(value),
1115 '*=': lambda attr, value: value in attr,
1117 str_operator_rex = re.compile(r'''(?x)
1118 \s*(?P<key>ext|acodec|vcodec|container|protocol|format_id|language)
1119 \s*(?P<negation>!\s*)?(?P<op>%s)(?P<none_inclusive>\s*\?)?
1120 \s*(?P<value>[a-zA-Z0-9._-]+)
1122 ''' % '|'.join(map(re.escape, STR_OPERATORS.keys())))
1123 m = str_operator_rex.search(filter_spec)
1125 comparison_value = m.group('value')
1126 str_op = STR_OPERATORS[m.group('op')]
1127 if m.group('negation'):
1128 op = lambda attr, value: not str_op(attr, value)
1133 raise ValueError('Invalid filter specification %r' % filter_spec)
1136 actual_value = f.get(m.group('key'))
1137 if actual_value is None:
1138 return m.group('none_inclusive')
1139 return op(actual_value, comparison_value)
1142 def _default_format_spec(self, info_dict, download=True):
1145 merger = FFmpegMergerPP(self)
1146 return merger.available and merger.can_merge()
1149 if self.params.get('simulate', False):
1153 if self.params.get('outtmpl', DEFAULT_OUTTMPL) == '-':
1155 if info_dict.get('is_live'):
1161 req_format_list = ['bestvideo+bestaudio', 'best']
1163 req_format_list.reverse()
1164 return '/'.join(req_format_list)
1166 def build_format_selector(self, format_spec):
1167 def syntax_error(note, start):
1169 'Invalid format specification: '
1170 '{0}\n\t{1}\n\t{2}^'.format(note, format_spec, ' ' * start[1]))
1171 return SyntaxError(message)
1173 PICKFIRST = 'PICKFIRST'
1177 FormatSelector = collections.namedtuple('FormatSelector', ['type', 'selector', 'filters'])
1179 def _parse_filter(tokens):
1181 for type, string, start, _, _ in tokens:
1182 if type == tokenize.OP and string == ']':
1183 return ''.join(filter_parts)
1185 filter_parts.append(string)
1187 def _remove_unused_ops(tokens):
1188 # Remove operators that we don't use and join them with the surrounding strings
1189 # for example: 'mp4' '-' 'baseline' '-' '16x9' is converted to 'mp4-baseline-16x9'
1190 ALLOWED_OPS = ('/', '+', ',', '(', ')')
1191 last_string, last_start, last_end, last_line = None, None, None, None
1192 for type, string, start, end, line in tokens:
1193 if type == tokenize.OP and string == '[':
1195 yield tokenize.NAME, last_string, last_start, last_end, last_line
1197 yield type, string, start, end, line
1198 # everything inside brackets will be handled by _parse_filter
1199 for type, string, start, end, line in tokens:
1200 yield type, string, start, end, line
1201 if type == tokenize.OP and string == ']':
1203 elif type == tokenize.OP and string in ALLOWED_OPS:
1205 yield tokenize.NAME, last_string, last_start, last_end, last_line
1207 yield type, string, start, end, line
1208 elif type in [tokenize.NAME, tokenize.NUMBER, tokenize.OP]:
1210 last_string = string
1214 last_string += string
1216 yield tokenize.NAME, last_string, last_start, last_end, last_line
1218 def _parse_format_selection(tokens, inside_merge=False, inside_choice=False, inside_group=False):
1220 current_selector = None
1221 for type, string, start, _, _ in tokens:
1222 # ENCODING is only defined in python 3.x
1223 if type == getattr(tokenize, 'ENCODING', None):
1225 elif type in [tokenize.NAME, tokenize.NUMBER]:
1226 current_selector = FormatSelector(SINGLE, string, [])
1227 elif type == tokenize.OP:
1229 if not inside_group:
1230 # ')' will be handled by the parentheses group
1231 tokens.restore_last_token()
1233 elif inside_merge and string in ['/', ',']:
1234 tokens.restore_last_token()
1236 elif inside_choice and string == ',':
1237 tokens.restore_last_token()
1240 if not current_selector:
1241 raise syntax_error('"," must follow a format selector', start)
1242 selectors.append(current_selector)
1243 current_selector = None
1245 if not current_selector:
1246 raise syntax_error('"/" must follow a format selector', start)
1247 first_choice = current_selector
1248 second_choice = _parse_format_selection(tokens, inside_choice=True)
1249 current_selector = FormatSelector(PICKFIRST, (first_choice, second_choice), [])
1251 if not current_selector:
1252 current_selector = FormatSelector(SINGLE, 'best', [])
1253 format_filter = _parse_filter(tokens)
1254 current_selector.filters.append(format_filter)
1256 if current_selector:
1257 raise syntax_error('Unexpected "("', start)
1258 group = _parse_format_selection(tokens, inside_group=True)
1259 current_selector = FormatSelector(GROUP, group, [])
1262 raise syntax_error('Unexpected "+"', start)
1263 video_selector = current_selector
1264 audio_selector = _parse_format_selection(tokens, inside_merge=True)
1265 if not video_selector or not audio_selector:
1266 raise syntax_error('"+" must be between two format selectors', start)
1267 current_selector = FormatSelector(MERGE, (video_selector, audio_selector), [])
1269 raise syntax_error('Operator not recognized: "{0}"'.format(string), start)
1270 elif type == tokenize.ENDMARKER:
1272 if current_selector:
1273 selectors.append(current_selector)
1276 def _build_selector_function(selector):
1277 if isinstance(selector, list):
1278 fs = [_build_selector_function(s) for s in selector]
1280 def selector_function(ctx):
1282 for format in f(ctx):
1284 return selector_function
1285 elif selector.type == GROUP:
1286 selector_function = _build_selector_function(selector.selector)
1287 elif selector.type == PICKFIRST:
1288 fs = [_build_selector_function(s) for s in selector.selector]
1290 def selector_function(ctx):
1292 picked_formats = list(f(ctx))
1294 return picked_formats
1296 elif selector.type == SINGLE:
1297 format_spec = selector.selector
1299 def selector_function(ctx):
1300 formats = list(ctx['formats'])
1303 if format_spec == 'all':
1306 elif format_spec in ['best', 'worst', None]:
1307 format_idx = 0 if format_spec == 'worst' else -1
1308 audiovideo_formats = [
1310 if f.get('vcodec') != 'none' and f.get('acodec') != 'none']
1311 if audiovideo_formats:
1312 yield audiovideo_formats[format_idx]
1313 # for extractors with incomplete formats (audio only (soundcloud)
1314 # or video only (imgur)) we will fallback to best/worst
1315 # {video,audio}-only format
1316 elif ctx['incomplete_formats']:
1317 yield formats[format_idx]
1318 elif format_spec == 'bestaudio':
1321 if f.get('vcodec') == 'none']
1323 yield audio_formats[-1]
1324 elif format_spec == 'worstaudio':
1327 if f.get('vcodec') == 'none']
1329 yield audio_formats[0]
1330 elif format_spec == 'bestvideo':
1333 if f.get('acodec') == 'none']
1335 yield video_formats[-1]
1336 elif format_spec == 'worstvideo':
1339 if f.get('acodec') == 'none']
1341 yield video_formats[0]
1343 extensions = ['mp4', 'flv', 'webm', '3gp', 'm4a', 'mp3', 'ogg', 'aac', 'wav']
1344 if format_spec in extensions:
1345 filter_f = lambda f: f['ext'] == format_spec
1347 filter_f = lambda f: f['format_id'] == format_spec
1348 matches = list(filter(filter_f, formats))
1351 elif selector.type == MERGE:
1352 def _merge(formats_info):
1353 format_1, format_2 = [f['format_id'] for f in formats_info]
1354 # The first format must contain the video and the
1356 if formats_info[0].get('vcodec') == 'none':
1357 self.report_error('The first format must '
1358 'contain the video, try using '
1359 '"-f %s+%s"' % (format_2, format_1))
1361 # Formats must be opposite (video+audio)
1362 if formats_info[0].get('acodec') == 'none' and formats_info[1].get('acodec') == 'none':
1364 'Both formats %s and %s are video-only, you must specify "-f video+audio"'
1365 % (format_1, format_2))
1368 formats_info[0]['ext']
1369 if self.params.get('merge_output_format') is None
1370 else self.params['merge_output_format'])
1372 'requested_formats': formats_info,
1373 'format': '%s+%s' % (formats_info[0].get('format'),
1374 formats_info[1].get('format')),
1375 'format_id': '%s+%s' % (formats_info[0].get('format_id'),
1376 formats_info[1].get('format_id')),
1377 'width': formats_info[0].get('width'),
1378 'height': formats_info[0].get('height'),
1379 'resolution': formats_info[0].get('resolution'),
1380 'fps': formats_info[0].get('fps'),
1381 'vcodec': formats_info[0].get('vcodec'),
1382 'vbr': formats_info[0].get('vbr'),
1383 'stretched_ratio': formats_info[0].get('stretched_ratio'),
1384 'acodec': formats_info[1].get('acodec'),
1385 'abr': formats_info[1].get('abr'),
1388 video_selector, audio_selector = map(_build_selector_function, selector.selector)
1390 def selector_function(ctx):
1391 for pair in itertools.product(
1392 video_selector(copy.deepcopy(ctx)), audio_selector(copy.deepcopy(ctx))):
1395 filters = [self._build_format_filter(f) for f in selector.filters]
1397 def final_selector(ctx):
1398 ctx_copy = copy.deepcopy(ctx)
1399 for _filter in filters:
1400 ctx_copy['formats'] = list(filter(_filter, ctx_copy['formats']))
1401 return selector_function(ctx_copy)
1402 return final_selector
1404 stream = io.BytesIO(format_spec.encode('utf-8'))
1406 tokens = list(_remove_unused_ops(compat_tokenize_tokenize(stream.readline)))
1407 except tokenize.TokenError:
1408 raise syntax_error('Missing closing/opening brackets or parenthesis', (0, len(format_spec)))
1410 class TokenIterator(object):
1411 def __init__(self, tokens):
1412 self.tokens = tokens
1419 if self.counter >= len(self.tokens):
1420 raise StopIteration()
1421 value = self.tokens[self.counter]
1427 def restore_last_token(self):
1430 parsed_selector = _parse_format_selection(iter(TokenIterator(tokens)))
1431 return _build_selector_function(parsed_selector)
1433 def _calc_headers(self, info_dict):
1434 res = std_headers.copy()
1436 add_headers = info_dict.get('http_headers')
1438 res.update(add_headers)
1440 cookies = self._calc_cookies(info_dict)
1442 res['Cookie'] = cookies
1444 if 'X-Forwarded-For' not in res:
1445 x_forwarded_for_ip = info_dict.get('__x_forwarded_for_ip')
1446 if x_forwarded_for_ip:
1447 res['X-Forwarded-For'] = x_forwarded_for_ip
1451 def _calc_cookies(self, info_dict):
1452 pr = sanitized_Request(info_dict['url'])
1453 self.cookiejar.add_cookie_header(pr)
1454 return pr.get_header('Cookie')
1456 def process_video_result(self, info_dict, download=True):
1457 assert info_dict.get('_type', 'video') == 'video'
1459 if 'id' not in info_dict:
1460 raise ExtractorError('Missing "id" field in extractor result')
1461 if 'title' not in info_dict:
1462 raise ExtractorError('Missing "title" field in extractor result')
1464 def report_force_conversion(field, field_not, conversion):
1465 self.report_warning(
1466 '"%s" field is not %s - forcing %s conversion, there is an error in extractor'
1467 % (field, field_not, conversion))
1469 def sanitize_string_field(info, string_field):
1470 field = info.get(string_field)
1471 if field is None or isinstance(field, compat_str):
1473 report_force_conversion(string_field, 'a string', 'string')
1474 info[string_field] = compat_str(field)
1476 def sanitize_numeric_fields(info):
1477 for numeric_field in self._NUMERIC_FIELDS:
1478 field = info.get(numeric_field)
1479 if field is None or isinstance(field, compat_numeric_types):
1481 report_force_conversion(numeric_field, 'numeric', 'int')
1482 info[numeric_field] = int_or_none(field)
1484 sanitize_string_field(info_dict, 'id')
1485 sanitize_numeric_fields(info_dict)
1487 if 'playlist' not in info_dict:
1488 # It isn't part of a playlist
1489 info_dict['playlist'] = None
1490 info_dict['playlist_index'] = None
1492 thumbnails = info_dict.get('thumbnails')
1493 if thumbnails is None:
1494 thumbnail = info_dict.get('thumbnail')
1496 info_dict['thumbnails'] = thumbnails = [{'url': thumbnail}]
1498 thumbnails.sort(key=lambda t: (
1499 t.get('preference') if t.get('preference') is not None else -1,
1500 t.get('width') if t.get('width') is not None else -1,
1501 t.get('height') if t.get('height') is not None else -1,
1502 t.get('id') if t.get('id') is not None else '', t.get('url')))
1503 for i, t in enumerate(thumbnails):
1504 t['url'] = sanitize_url(t['url'])
1505 if t.get('width') and t.get('height'):
1506 t['resolution'] = '%dx%d' % (t['width'], t['height'])
1507 if t.get('id') is None:
1510 if self.params.get('list_thumbnails'):
1511 self.list_thumbnails(info_dict)
1514 thumbnail = info_dict.get('thumbnail')
1516 info_dict['thumbnail'] = sanitize_url(thumbnail)
1518 info_dict['thumbnail'] = thumbnails[-1]['url']
1520 if 'display_id' not in info_dict and 'id' in info_dict:
1521 info_dict['display_id'] = info_dict['id']
1523 for ts_key, date_key in (
1524 ('timestamp', 'upload_date'),
1525 ('release_timestamp', 'release_date'),
1527 if info_dict.get(date_key) is None and info_dict.get(ts_key) is not None:
1528 # Working around out-of-range timestamp values (e.g. negative ones on Windows,
1529 # see http://bugs.python.org/issue1646728)
1531 upload_date = datetime.datetime.utcfromtimestamp(info_dict[ts_key])
1532 info_dict[date_key] = compat_str(upload_date.strftime('%Y%m%d'))
1533 except (ValueError, OverflowError, OSError):
1536 # Auto generate title fields corresponding to the *_number fields when missing
1537 # in order to always have clean titles. This is very common for TV series.
1538 for field in ('chapter', 'season', 'episode'):
1539 if info_dict.get('%s_number' % field) is not None and not info_dict.get(field):
1540 info_dict[field] = '%s %d' % (field.capitalize(), info_dict['%s_number' % field])
1542 for cc_kind in ('subtitles', 'automatic_captions'):
1543 cc = info_dict.get(cc_kind)
1545 for _, subtitle in cc.items():
1546 for subtitle_format in subtitle:
1547 if subtitle_format.get('url'):
1548 subtitle_format['url'] = sanitize_url(subtitle_format['url'])
1549 if subtitle_format.get('ext') is None:
1550 subtitle_format['ext'] = determine_ext(subtitle_format['url']).lower()
1552 automatic_captions = info_dict.get('automatic_captions')
1553 subtitles = info_dict.get('subtitles')
1555 if self.params.get('listsubtitles', False):
1556 if 'automatic_captions' in info_dict:
1557 self.list_subtitles(
1558 info_dict['id'], automatic_captions, 'automatic captions')
1559 self.list_subtitles(info_dict['id'], subtitles, 'subtitles')
1562 info_dict['requested_subtitles'] = self.process_subtitles(
1563 info_dict['id'], subtitles, automatic_captions)
1565 # We now pick which formats have to be downloaded
1566 if info_dict.get('formats') is None:
1567 # There's only one format available
1568 formats = [info_dict]
1570 formats = info_dict['formats']
1573 raise ExtractorError('No video formats found!')
1575 def is_wellformed(f):
1578 self.report_warning(
1579 '"url" field is missing or empty - skipping format, '
1580 'there is an error in extractor')
1582 if isinstance(url, bytes):
1583 sanitize_string_field(f, 'url')
1586 # Filter out malformed formats for better extraction robustness
1587 formats = list(filter(is_wellformed, formats))
1591 # We check that all the formats have the format and format_id fields
1592 for i, format in enumerate(formats):
1593 sanitize_string_field(format, 'format_id')
1594 sanitize_numeric_fields(format)
1595 format['url'] = sanitize_url(format['url'])
1596 if not format.get('format_id'):
1597 format['format_id'] = compat_str(i)
1599 # Sanitize format_id from characters used in format selector expression
1600 format['format_id'] = re.sub(r'[\s,/+\[\]()]', '_', format['format_id'])
1601 format_id = format['format_id']
1602 if format_id not in formats_dict:
1603 formats_dict[format_id] = []
1604 formats_dict[format_id].append(format)
1606 # Make sure all formats have unique format_id
1607 for format_id, ambiguous_formats in formats_dict.items():
1608 if len(ambiguous_formats) > 1:
1609 for i, format in enumerate(ambiguous_formats):
1610 format['format_id'] = '%s-%d' % (format_id, i)
1612 for i, format in enumerate(formats):
1613 if format.get('format') is None:
1614 format['format'] = '{id} - {res}{note}'.format(
1615 id=format['format_id'],
1616 res=self.format_resolution(format),
1617 note=' ({0})'.format(format['format_note']) if format.get('format_note') is not None else '',
1619 # Automatically determine file extension if missing
1620 if format.get('ext') is None:
1621 format['ext'] = determine_ext(format['url']).lower()
1622 # Automatically determine protocol if missing (useful for format
1623 # selection purposes)
1624 if format.get('protocol') is None:
1625 format['protocol'] = determine_protocol(format)
1626 # Add HTTP headers, so that external programs can use them from the
1628 full_format_info = info_dict.copy()
1629 full_format_info.update(format)
1630 format['http_headers'] = self._calc_headers(full_format_info)
1631 # Remove private housekeeping stuff
1632 if '__x_forwarded_for_ip' in info_dict:
1633 del info_dict['__x_forwarded_for_ip']
1635 # TODO Central sorting goes here
1637 if formats[0] is not info_dict:
1638 # only set the 'formats' fields if the original info_dict list them
1639 # otherwise we end up with a circular reference, the first (and unique)
1640 # element in the 'formats' field in info_dict is info_dict itself,
1641 # which can't be exported to json
1642 info_dict['formats'] = formats
1643 if self.params.get('listformats'):
1644 self.list_formats(info_dict)
1647 req_format = self.params.get('format')
1648 if req_format is None:
1649 req_format = self._default_format_spec(info_dict, download=download)
1650 if self.params.get('verbose'):
1651 self._write_string('[debug] Default format spec: %s\n' % req_format)
1653 format_selector = self.build_format_selector(req_format)
1655 # While in format selection we may need to have an access to the original
1656 # format set in order to calculate some metrics or do some processing.
1657 # For now we need to be able to guess whether original formats provided
1658 # by extractor are incomplete or not (i.e. whether extractor provides only
1659 # video-only or audio-only formats) for proper formats selection for
1660 # extractors with such incomplete formats (see
1661 # https://github.com/ytdl-org/youtube-dl/pull/5556).
1662 # Since formats may be filtered during format selection and may not match
1663 # the original formats the results may be incorrect. Thus original formats
1664 # or pre-calculated metrics should be passed to format selection routines
1666 # We will pass a context object containing all necessary additional data
1667 # instead of just formats.
1668 # This fixes incorrect format selection issue (see
1669 # https://github.com/ytdl-org/youtube-dl/issues/10083).
1670 incomplete_formats = (
1671 # All formats are video-only or
1672 all(f.get('vcodec') != 'none' and f.get('acodec') == 'none' for f in formats)
1673 # all formats are audio-only
1674 or all(f.get('vcodec') == 'none' and f.get('acodec') != 'none' for f in formats))
1678 'incomplete_formats': incomplete_formats,
1681 formats_to_download = list(format_selector(ctx))
1682 if not formats_to_download:
1683 raise ExtractorError('requested format not available',
1687 if len(formats_to_download) > 1:
1688 self.to_screen('[info] %s: downloading video in %s formats' % (info_dict['id'], len(formats_to_download)))
1689 for format in formats_to_download:
1690 new_info = dict(info_dict)
1691 new_info.update(format)
1692 self.process_info(new_info)
1693 # We update the info dict with the best quality format (backwards compatibility)
1694 info_dict.update(formats_to_download[-1])
1697 def process_subtitles(self, video_id, normal_subtitles, automatic_captions):
1698 """Select the requested subtitles and their format"""
1700 if normal_subtitles and self.params.get('writesubtitles'):
1701 available_subs.update(normal_subtitles)
1702 if automatic_captions and self.params.get('writeautomaticsub'):
1703 for lang, cap_info in automatic_captions.items():
1704 if lang not in available_subs:
1705 available_subs[lang] = cap_info
1707 if (not self.params.get('writesubtitles') and not
1708 self.params.get('writeautomaticsub') or not
1712 if self.params.get('allsubtitles', False):
1713 requested_langs = available_subs.keys()
1715 if self.params.get('subtitleslangs', False):
1716 requested_langs = self.params.get('subtitleslangs')
1717 elif 'en' in available_subs:
1718 requested_langs = ['en']
1720 requested_langs = [list(available_subs.keys())[0]]
1722 formats_query = self.params.get('subtitlesformat', 'best')
1723 formats_preference = formats_query.split('/') if formats_query else []
1725 for lang in requested_langs:
1726 formats = available_subs.get(lang)
1728 self.report_warning('%s subtitles not available for %s' % (lang, video_id))
1730 for ext in formats_preference:
1734 matches = list(filter(lambda f: f['ext'] == ext, formats))
1740 self.report_warning(
1741 'No subtitle format found matching "%s" for language %s, '
1742 'using %s' % (formats_query, lang, f['ext']))
1746 def __forced_printings(self, info_dict, filename, incomplete):
1747 def print_mandatory(field):
1748 if (self.params.get('force%s' % field, False)
1749 and (not incomplete or info_dict.get(field) is not None)):
1750 self.to_stdout(info_dict[field])
1752 def print_optional(field):
1753 if (self.params.get('force%s' % field, False)
1754 and info_dict.get(field) is not None):
1755 self.to_stdout(info_dict[field])
1757 print_mandatory('title')
1758 print_mandatory('id')
1759 if self.params.get('forceurl', False) and not incomplete:
1760 if info_dict.get('requested_formats') is not None:
1761 for f in info_dict['requested_formats']:
1762 self.to_stdout(f['url'] + f.get('play_path', ''))
1764 # For RTMP URLs, also include the playpath
1765 self.to_stdout(info_dict['url'] + info_dict.get('play_path', ''))
1766 print_optional('thumbnail')
1767 print_optional('description')
1768 if self.params.get('forcefilename', False) and filename is not None:
1769 self.to_stdout(filename)
1770 if self.params.get('forceduration', False) and info_dict.get('duration') is not None:
1771 self.to_stdout(formatSeconds(info_dict['duration']))
1772 print_mandatory('format')
1773 if self.params.get('forcejson', False):
1774 self.to_stdout(json.dumps(info_dict))
1776 def process_info(self, info_dict):
1777 """Process a single resolved IE result."""
1779 assert info_dict.get('_type', 'video') == 'video'
1781 max_downloads = self.params.get('max_downloads')
1782 if max_downloads is not None:
1783 if self._num_downloads >= int(max_downloads):
1784 raise MaxDownloadsReached()
1786 # TODO: backward compatibility, to be removed
1787 info_dict['fulltitle'] = info_dict['title']
1789 if 'format' not in info_dict:
1790 info_dict['format'] = info_dict['ext']
1792 reason = self._match_entry(info_dict, incomplete=False)
1793 if reason is not None:
1794 self.to_screen('[download] ' + reason)
1797 self._num_downloads += 1
1799 info_dict['_filename'] = filename = self.prepare_filename(info_dict)
1802 self.__forced_printings(info_dict, filename, incomplete=False)
1804 # Do nothing else if in simulate mode
1805 if self.params.get('simulate', False):
1808 if filename is None:
1811 def ensure_dir_exists(path):
1813 dn = os.path.dirname(path)
1814 if dn and not os.path.exists(dn):
1817 except (OSError, IOError) as err:
1818 if isinstance(err, OSError) and err.errno == errno.EEXIST:
1820 self.report_error('unable to create directory ' + error_to_compat_str(err))
1823 if not ensure_dir_exists(sanitize_path(encodeFilename(filename))):
1826 if self.params.get('writedescription', False):
1827 descfn = replace_extension(filename, 'description', info_dict.get('ext'))
1828 if self.params.get('nooverwrites', False) and os.path.exists(encodeFilename(descfn)):
1829 self.to_screen('[info] Video description is already present')
1830 elif info_dict.get('description') is None:
1831 self.report_warning('There\'s no description to write.')
1834 self.to_screen('[info] Writing video description to: ' + descfn)
1835 with io.open(encodeFilename(descfn), 'w', encoding='utf-8') as descfile:
1836 descfile.write(info_dict['description'])
1837 except (OSError, IOError):
1838 self.report_error('Cannot write description file ' + descfn)
1841 if self.params.get('writeannotations', False):
1842 annofn = replace_extension(filename, 'annotations.xml', info_dict.get('ext'))
1843 if self.params.get('nooverwrites', False) and os.path.exists(encodeFilename(annofn)):
1844 self.to_screen('[info] Video annotations are already present')
1845 elif not info_dict.get('annotations'):
1846 self.report_warning('There are no annotations to write.')
1849 self.to_screen('[info] Writing video annotations to: ' + annofn)
1850 with io.open(encodeFilename(annofn), 'w', encoding='utf-8') as annofile:
1851 annofile.write(info_dict['annotations'])
1852 except (KeyError, TypeError):
1853 self.report_warning('There are no annotations to write.')
1854 except (OSError, IOError):
1855 self.report_error('Cannot write annotations file: ' + annofn)
1858 subtitles_are_requested = any([self.params.get('writesubtitles', False),
1859 self.params.get('writeautomaticsub')])
1861 if subtitles_are_requested and info_dict.get('requested_subtitles'):
1862 # subtitles download errors are already managed as troubles in relevant IE
1863 # that way it will silently go on when used with unsupporting IE
1864 subtitles = info_dict['requested_subtitles']
1865 ie = self.get_info_extractor(info_dict['extractor_key'])
1866 for sub_lang, sub_info in subtitles.items():
1867 sub_format = sub_info['ext']
1868 sub_filename = subtitles_filename(filename, sub_lang, sub_format, info_dict.get('ext'))
1869 if self.params.get('nooverwrites', False) and os.path.exists(encodeFilename(sub_filename)):
1870 self.to_screen('[info] Video subtitle %s.%s is already present' % (sub_lang, sub_format))
1872 self.to_screen('[info] Writing video subtitles to: ' + sub_filename)
1873 if sub_info.get('data') is not None:
1875 # Use newline='' to prevent conversion of newline characters
1876 # See https://github.com/ytdl-org/youtube-dl/issues/10268
1877 with io.open(encodeFilename(sub_filename), 'w', encoding='utf-8', newline='') as subfile:
1878 subfile.write(sub_info['data'])
1879 except (OSError, IOError):
1880 self.report_error('Cannot write subtitles file ' + sub_filename)
1884 sub_data = ie._request_webpage(
1885 sub_info['url'], info_dict['id'], note=False).read()
1886 with io.open(encodeFilename(sub_filename), 'wb') as subfile:
1887 subfile.write(sub_data)
1888 except (ExtractorError, IOError, OSError, ValueError) as err:
1889 self.report_warning('Unable to download subtitle for "%s": %s' %
1890 (sub_lang, error_to_compat_str(err)))
1893 if self.params.get('writeinfojson', False):
1894 infofn = replace_extension(filename, 'info.json', info_dict.get('ext'))
1895 if self.params.get('nooverwrites', False) and os.path.exists(encodeFilename(infofn)):
1896 self.to_screen('[info] Video description metadata is already present')
1898 self.to_screen('[info] Writing video description metadata as JSON to: ' + infofn)
1900 write_json_file(self.filter_requested_info(info_dict), infofn)
1901 except (OSError, IOError):
1902 self.report_error('Cannot write metadata to JSON file ' + infofn)
1905 self._write_thumbnails(info_dict, filename)
1907 if not self.params.get('skip_download', False):
1909 def checked_get_suitable_downloader(info_dict, params):
1910 ed_args = params.get('external_downloader_args')
1911 dler = get_suitable_downloader(info_dict, params)
1912 if ed_args and not params.get('external_downloader_args'):
1913 # external_downloader_args was cleared because external_downloader was rejected
1914 self.report_warning('Requested external downloader cannot be used: '
1915 'ignoring --external-downloader-args.')
1919 fd = checked_get_suitable_downloader(info, self.params)(self, self.params)
1920 for ph in self._progress_hooks:
1921 fd.add_progress_hook(ph)
1922 if self.params.get('verbose'):
1923 self.to_screen('[debug] Invoking downloader on %r' % info.get('url'))
1924 return fd.download(name, info)
1926 if info_dict.get('requested_formats') is not None:
1929 merger = FFmpegMergerPP(self)
1930 if not merger.available:
1932 self.report_warning('You have requested multiple '
1933 'formats but ffmpeg or avconv are not installed.'
1934 ' The formats won\'t be merged.')
1936 postprocessors = [merger]
1938 def compatible_formats(formats):
1939 video, audio = formats
1941 video_ext, audio_ext = video.get('ext'), audio.get('ext')
1942 if video_ext and audio_ext:
1944 ('mp3', 'mp4', 'm4a', 'm4p', 'm4b', 'm4r', 'm4v', 'ismv', 'isma'),
1947 for exts in COMPATIBLE_EXTS:
1948 if video_ext in exts and audio_ext in exts:
1950 # TODO: Check acodec/vcodec
1953 filename_real_ext = os.path.splitext(filename)[1][1:]
1955 os.path.splitext(filename)[0]
1956 if filename_real_ext == info_dict['ext']
1958 requested_formats = info_dict['requested_formats']
1959 if self.params.get('merge_output_format') is None and not compatible_formats(requested_formats):
1960 info_dict['ext'] = 'mkv'
1961 self.report_warning(
1962 'Requested formats are incompatible for merge and will be merged into mkv.')
1963 # Ensure filename always has a correct extension for successful merge
1964 filename = '%s.%s' % (filename_wo_ext, info_dict['ext'])
1965 if os.path.exists(encodeFilename(filename)):
1967 '[download] %s has already been downloaded and '
1968 'merged' % filename)
1970 for f in requested_formats:
1971 new_info = dict(info_dict)
1973 fname = prepend_extension(
1974 self.prepare_filename(new_info),
1975 'f%s' % f['format_id'], new_info['ext'])
1976 if not ensure_dir_exists(fname):
1978 downloaded.append(fname)
1979 partial_success = dl(fname, new_info)
1980 success = success and partial_success
1981 info_dict['__postprocessors'] = postprocessors
1982 info_dict['__files_to_merge'] = downloaded
1984 # Just a single file
1985 success = dl(filename, info_dict)
1986 except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
1987 self.report_error('unable to download video data: %s' % error_to_compat_str(err))
1989 except (OSError, IOError) as err:
1990 raise UnavailableVideoError(err)
1991 except (ContentTooShortError, ) as err:
1992 self.report_error('content too short (expected %s bytes and served %s)' % (err.expected, err.downloaded))
1995 if success and filename != '-':
1997 fixup_policy = self.params.get('fixup')
1998 if fixup_policy is None:
1999 fixup_policy = 'detect_or_warn'
2001 INSTALL_FFMPEG_MESSAGE = 'Install ffmpeg or avconv to fix this automatically.'
2003 stretched_ratio = info_dict.get('stretched_ratio')
2004 if stretched_ratio is not None and stretched_ratio != 1:
2005 if fixup_policy == 'warn':
2006 self.report_warning('%s: Non-uniform pixel ratio (%s)' % (
2007 info_dict['id'], stretched_ratio))
2008 elif fixup_policy == 'detect_or_warn':
2009 stretched_pp = FFmpegFixupStretchedPP(self)
2010 if stretched_pp.available:
2011 info_dict.setdefault('__postprocessors', [])
2012 info_dict['__postprocessors'].append(stretched_pp)
2014 self.report_warning(
2015 '%s: Non-uniform pixel ratio (%s). %s'
2016 % (info_dict['id'], stretched_ratio, INSTALL_FFMPEG_MESSAGE))
2018 assert fixup_policy in ('ignore', 'never')
2020 if (info_dict.get('requested_formats') is None
2021 and info_dict.get('container') == 'm4a_dash'):
2022 if fixup_policy == 'warn':
2023 self.report_warning(
2024 '%s: writing DASH m4a. '
2025 'Only some players support this container.'
2027 elif fixup_policy == 'detect_or_warn':
2028 fixup_pp = FFmpegFixupM4aPP(self)
2029 if fixup_pp.available:
2030 info_dict.setdefault('__postprocessors', [])
2031 info_dict['__postprocessors'].append(fixup_pp)
2033 self.report_warning(
2034 '%s: writing DASH m4a. '
2035 'Only some players support this container. %s'
2036 % (info_dict['id'], INSTALL_FFMPEG_MESSAGE))
2038 assert fixup_policy in ('ignore', 'never')
2040 if (info_dict.get('protocol') == 'm3u8_native'
2041 or info_dict.get('protocol') == 'm3u8'
2042 and self.params.get('hls_prefer_native')):
2043 if fixup_policy == 'warn':
2044 self.report_warning('%s: malformed AAC bitstream detected.' % (
2046 elif fixup_policy == 'detect_or_warn':
2047 fixup_pp = FFmpegFixupM3u8PP(self)
2048 if fixup_pp.available:
2049 info_dict.setdefault('__postprocessors', [])
2050 info_dict['__postprocessors'].append(fixup_pp)
2052 self.report_warning(
2053 '%s: malformed AAC bitstream detected. %s'
2054 % (info_dict['id'], INSTALL_FFMPEG_MESSAGE))
2056 assert fixup_policy in ('ignore', 'never')
2059 self.post_process(filename, info_dict)
2060 except (PostProcessingError) as err:
2061 self.report_error('postprocessing: %s' % str(err))
2063 self.record_download_archive(info_dict)
2065 def download(self, url_list):
2066 """Download a given list of URLs."""
2067 outtmpl = self.params.get('outtmpl', DEFAULT_OUTTMPL)
2068 if (len(url_list) > 1
2070 and '%' not in outtmpl
2071 and self.params.get('max_downloads') != 1):
2072 raise SameFileError(outtmpl)
2074 for url in url_list:
2076 # It also downloads the videos
2077 res = self.extract_info(
2078 url, force_generic_extractor=self.params.get('force_generic_extractor', False))
2079 except UnavailableVideoError:
2080 self.report_error('unable to download video')
2081 except MaxDownloadsReached:
2082 self.to_screen('[info] Maximum number of downloaded files reached.')
2085 if self.params.get('dump_single_json', False):
2086 self.to_stdout(json.dumps(res))
2088 return self._download_retcode
2090 def download_with_info_file(self, info_filename):
2091 with contextlib.closing(fileinput.FileInput(
2092 [info_filename], mode='r',
2093 openhook=fileinput.hook_encoded('utf-8'))) as f:
2094 # FileInput doesn't have a read method, we can't call json.load
2095 info = self.filter_requested_info(json.loads('\n'.join(f)))
2097 self.process_ie_result(info, download=True)
2098 except DownloadError:
2099 webpage_url = info.get('webpage_url')
2100 if webpage_url is not None:
2101 self.report_warning('The info failed to download, trying with "%s"' % webpage_url)
2102 return self.download([webpage_url])
2105 return self._download_retcode
2108 def filter_requested_info(info_dict):
2110 (k, v) for k, v in info_dict.items()
2111 if k not in ['requested_formats', 'requested_subtitles'])
2113 def post_process(self, filename, ie_info):
2114 """Run all the postprocessors on the given file."""
2115 info = dict(ie_info)
2116 info['filepath'] = filename
2118 if ie_info.get('__postprocessors') is not None:
2119 pps_chain.extend(ie_info['__postprocessors'])
2120 pps_chain.extend(self._pps)
2121 for pp in pps_chain:
2122 files_to_delete = []
2124 files_to_delete, info = pp.run(info)
2125 except PostProcessingError as e:
2126 self.report_error(e.msg)
2127 if files_to_delete and not self.params.get('keepvideo', False):
2128 for old_filename in files_to_delete:
2129 self.to_screen('Deleting original file %s (pass -k to keep)' % old_filename)
2131 os.remove(encodeFilename(old_filename))
2132 except (IOError, OSError):
2133 self.report_warning('Unable to remove downloaded original file')
2135 def _make_archive_id(self, info_dict):
2136 video_id = info_dict.get('id')
2139 # Future-proof against any change in case
2140 # and backwards compatibility with prior versions
2141 extractor = info_dict.get('extractor_key') or info_dict.get('ie_key') # key in a playlist
2142 if extractor is None:
2143 url = str_or_none(info_dict.get('url'))
2146 # Try to find matching extractor for the URL and take its ie_key
2147 for ie in self._ies:
2148 if ie.suitable(url):
2149 extractor = ie.ie_key()
2153 return extractor.lower() + ' ' + video_id
2155 def in_download_archive(self, info_dict):
2156 fn = self.params.get('download_archive')
2160 vid_id = self._make_archive_id(info_dict)
2162 return False # Incomplete video information
2165 with locked_file(fn, 'r', encoding='utf-8') as archive_file:
2166 for line in archive_file:
2167 if line.strip() == vid_id:
2169 except IOError as ioe:
2170 if ioe.errno != errno.ENOENT:
2174 def record_download_archive(self, info_dict):
2175 fn = self.params.get('download_archive')
2178 vid_id = self._make_archive_id(info_dict)
2180 with locked_file(fn, 'a', encoding='utf-8') as archive_file:
2181 archive_file.write(vid_id + '\n')
2184 def format_resolution(format, default='unknown'):
2185 if format.get('vcodec') == 'none':
2187 if format.get('resolution') is not None:
2188 return format['resolution']
2189 if format.get('height') is not None:
2190 if format.get('width') is not None:
2191 res = '%sx%s' % (format['width'], format['height'])
2193 res = '%sp' % format['height']
2194 elif format.get('width') is not None:
2195 res = '%dx?' % format['width']
2200 def _format_note(self, fdict):
2202 if fdict.get('ext') in ['f4f', 'f4m']:
2203 res += '(unsupported) '
2204 if fdict.get('language'):
2207 res += '[%s] ' % fdict['language']
2208 if fdict.get('format_note') is not None:
2209 res += fdict['format_note'] + ' '
2210 if fdict.get('tbr') is not None:
2211 res += '%4dk ' % fdict['tbr']
2212 if fdict.get('container') is not None:
2215 res += '%s container' % fdict['container']
2216 if (fdict.get('vcodec') is not None
2217 and fdict.get('vcodec') != 'none'):
2220 res += fdict['vcodec']
2221 if fdict.get('vbr') is not None:
2223 elif fdict.get('vbr') is not None and fdict.get('abr') is not None:
2225 if fdict.get('vbr') is not None:
2226 res += '%4dk' % fdict['vbr']
2227 if fdict.get('fps') is not None:
2230 res += '%sfps' % fdict['fps']
2231 if fdict.get('acodec') is not None:
2234 if fdict['acodec'] == 'none':
2237 res += '%-5s' % fdict['acodec']
2238 elif fdict.get('abr') is not None:
2242 if fdict.get('abr') is not None:
2243 res += '@%3dk' % fdict['abr']
2244 if fdict.get('asr') is not None:
2245 res += ' (%5dHz)' % fdict['asr']
2246 if fdict.get('filesize') is not None:
2249 res += format_bytes(fdict['filesize'])
2250 elif fdict.get('filesize_approx') is not None:
2253 res += '~' + format_bytes(fdict['filesize_approx'])
2256 def list_formats(self, info_dict):
2257 formats = info_dict.get('formats', [info_dict])
2259 [f['format_id'], f['ext'], self.format_resolution(f), self._format_note(f)]
2261 if f.get('preference') is None or f['preference'] >= -1000]
2262 if len(formats) > 1:
2263 table[-1][-1] += (' ' if table[-1][-1] else '') + '(best)'
2265 header_line = ['format code', 'extension', 'resolution', 'note']
2267 '[info] Available formats for %s:\n%s' %
2268 (info_dict['id'], render_table(header_line, table)))
2270 def list_thumbnails(self, info_dict):
2271 thumbnails = info_dict.get('thumbnails')
2273 self.to_screen('[info] No thumbnails present for %s' % info_dict['id'])
2277 '[info] Thumbnails for %s:' % info_dict['id'])
2278 self.to_screen(render_table(
2279 ['ID', 'width', 'height', 'URL'],
2280 [[t['id'], t.get('width', 'unknown'), t.get('height', 'unknown'), t['url']] for t in thumbnails]))
2282 def list_subtitles(self, video_id, subtitles, name='subtitles'):
2284 self.to_screen('%s has no %s' % (video_id, name))
2287 'Available %s for %s:' % (name, video_id))
2288 self.to_screen(render_table(
2289 ['Language', 'formats'],
2290 [[lang, ', '.join(f['ext'] for f in reversed(formats))]
2291 for lang, formats in subtitles.items()]))
2293 def urlopen(self, req):
2294 """ Start an HTTP download """
2295 if isinstance(req, compat_basestring):
2296 req = sanitized_Request(req)
2297 return self._opener.open(req, timeout=self._socket_timeout)
2299 def print_debug_header(self):
2300 if not self.params.get('verbose'):
2303 if type('') is not compat_str:
2304 # Python 2.6 on SLES11 SP1 (https://github.com/ytdl-org/youtube-dl/issues/3326)
2305 self.report_warning(
2306 'Your Python is broken! Update to a newer and supported version')
2308 stdout_encoding = getattr(
2309 sys.stdout, 'encoding', 'missing (%s)' % type(sys.stdout).__name__)
2311 '[debug] Encodings: locale %s, fs %s, out %s, pref %s\n' % (
2312 locale.getpreferredencoding(),
2313 sys.getfilesystemencoding(),
2315 self.get_encoding()))
2316 write_string(encoding_str, encoding=None)
2318 self._write_string('[debug] youtube-dl version ' + __version__ + '\n')
2320 self._write_string('[debug] Lazy loading extractors enabled' + '\n')
2322 sp = subprocess.Popen(
2323 ['git', 'rev-parse', '--short', 'HEAD'],
2324 stdout=subprocess.PIPE, stderr=subprocess.PIPE,
2325 cwd=os.path.dirname(os.path.abspath(__file__)))
2326 out, err = sp.communicate()
2327 out = out.decode().strip()
2328 if re.match('[0-9a-f]+', out):
2329 self._write_string('[debug] Git HEAD: ' + out + '\n')
2336 def python_implementation():
2337 impl_name = platform.python_implementation()
2338 if impl_name == 'PyPy' and hasattr(sys, 'pypy_version_info'):
2339 return impl_name + ' version %d.%d.%d' % sys.pypy_version_info[:3]
2342 self._write_string('[debug] Python version %s (%s) - %s\n' % (
2343 platform.python_version(), python_implementation(),
2346 exe_versions = FFmpegPostProcessor.get_versions(self)
2347 exe_versions['rtmpdump'] = rtmpdump_version()
2348 exe_versions['phantomjs'] = PhantomJSwrapper._version()
2349 exe_str = ', '.join(
2351 for exe, v in sorted(exe_versions.items())
2356 self._write_string('[debug] exe versions: %s\n' % exe_str)
2359 for handler in self._opener.handlers:
2360 if hasattr(handler, 'proxies'):
2361 proxy_map.update(handler.proxies)
2362 self._write_string('[debug] Proxy map: ' + compat_str(proxy_map) + '\n')
2364 if self.params.get('call_home', False):
2365 ipaddr = self.urlopen('https://yt-dl.org/ip').read().decode('utf-8')
2366 self._write_string('[debug] Public IP address: %s\n' % ipaddr)
2367 latest_version = self.urlopen(
2368 'https://yt-dl.org/latest/version').read().decode('utf-8')
2369 if version_tuple(latest_version) > version_tuple(__version__):
2370 self.report_warning(
2371 'You are using an outdated version (newest version: %s)! '
2372 'See https://yt-dl.org/update if you need help updating.' %
2375 def _setup_opener(self):
2376 timeout_val = self.params.get('socket_timeout')
2377 self._socket_timeout = 600 if timeout_val is None else float(timeout_val)
2379 opts_cookiefile = self.params.get('cookiefile')
2380 opts_proxy = self.params.get('proxy')
2382 if opts_cookiefile is None:
2383 self.cookiejar = compat_cookiejar.CookieJar()
2385 opts_cookiefile = expand_path(opts_cookiefile)
2386 self.cookiejar = YoutubeDLCookieJar(opts_cookiefile)
2387 if os.access(opts_cookiefile, os.R_OK):
2388 self.cookiejar.load(ignore_discard=True, ignore_expires=True)
2390 cookie_processor = YoutubeDLCookieProcessor(self.cookiejar)
2391 if opts_proxy is not None:
2392 if opts_proxy == '':
2395 proxies = {'http': opts_proxy, 'https': opts_proxy}
2397 proxies = compat_urllib_request.getproxies()
2398 # Set HTTPS proxy to HTTP one if given (https://github.com/ytdl-org/youtube-dl/issues/805)
2399 if 'http' in proxies and 'https' not in proxies:
2400 proxies['https'] = proxies['http']
2401 proxy_handler = PerRequestProxyHandler(proxies)
2403 debuglevel = 1 if self.params.get('debug_printtraffic') else 0
2404 https_handler = make_HTTPS_handler(self.params, debuglevel=debuglevel)
2405 ydlh = YoutubeDLHandler(self.params, debuglevel=debuglevel)
2406 redirect_handler = YoutubeDLRedirectHandler()
2407 data_handler = compat_urllib_request_DataHandler()
2409 # When passing our own FileHandler instance, build_opener won't add the
2410 # default FileHandler and allows us to disable the file protocol, which
2411 # can be used for malicious purposes (see
2412 # https://github.com/ytdl-org/youtube-dl/issues/8227)
2413 file_handler = compat_urllib_request.FileHandler()
2415 def file_open(*args, **kwargs):
2416 raise compat_urllib_error.URLError('file:// scheme is explicitly disabled in youtube-dl for security reasons')
2417 file_handler.file_open = file_open
2419 opener = compat_urllib_request.build_opener(
2420 proxy_handler, https_handler, cookie_processor, ydlh, redirect_handler, data_handler, file_handler)
2422 # Delete the default user-agent header, which would otherwise apply in
2423 # cases where our custom HTTP handler doesn't come into play
2424 # (See https://github.com/ytdl-org/youtube-dl/issues/1309 for details)
2425 opener.addheaders = []
2426 self._opener = opener
2428 def encode(self, s):
2429 if isinstance(s, bytes):
2430 return s # Already encoded
2433 return s.encode(self.get_encoding())
2434 except UnicodeEncodeError as err:
2435 err.reason = err.reason + '. Check your system encoding configuration or use the --encoding option.'
2438 def get_encoding(self):
2439 encoding = self.params.get('encoding')
2440 if encoding is None:
2441 encoding = preferredencoding()
2444 def _write_thumbnails(self, info_dict, filename):
2445 if self.params.get('writethumbnail', False):
2446 thumbnails = info_dict.get('thumbnails')
2448 thumbnails = [thumbnails[-1]]
2449 elif self.params.get('write_all_thumbnails', False):
2450 thumbnails = info_dict.get('thumbnails')
2455 # No thumbnails present, so return immediately
2458 for t in thumbnails:
2459 thumb_ext = determine_ext(t['url'], 'jpg')
2460 suffix = '_%s' % t['id'] if len(thumbnails) > 1 else ''
2461 thumb_display_id = '%s ' % t['id'] if len(thumbnails) > 1 else ''
2462 t['filename'] = thumb_filename = replace_extension(filename + suffix, thumb_ext, info_dict.get('ext'))
2464 if self.params.get('nooverwrites', False) and os.path.exists(encodeFilename(thumb_filename)):
2465 self.to_screen('[%s] %s: Thumbnail %sis already present' %
2466 (info_dict['extractor'], info_dict['id'], thumb_display_id))
2468 self.to_screen('[%s] %s: Downloading thumbnail %s...' %
2469 (info_dict['extractor'], info_dict['id'], thumb_display_id))
2471 uf = self.urlopen(t['url'])
2472 with open(encodeFilename(thumb_filename), 'wb') as thumbf:
2473 shutil.copyfileobj(uf, thumbf)
2474 self.to_screen('[%s] %s: Writing thumbnail %sto: %s' %
2475 (info_dict['extractor'], info_dict['id'], thumb_display_id, thumb_filename))
2476 except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
2477 self.report_warning('Unable to download thumbnail "%s": %s' %
2478 (t['url'], error_to_compat_str(err)))