2 # -*- coding: utf-8 -*-
4 from __future__ import absolute_import, unicode_literals
33 compat_get_terminal_size,
38 compat_urllib_request,
58 PerRequestProxyHandler,
68 UnavailableVideoError,
79 from .cache import Cache
80 from .extractor import get_info_extractor, gen_extractors
81 from .downloader import get_suitable_downloader
82 from .downloader.rtmp import rtmpdump_version
83 from .postprocessor import (
85 FFmpegFixupStretchedPP,
90 from .version import __version__
93 class YoutubeDL(object):
96 YoutubeDL objects are the ones responsible of downloading the
97 actual video file and writing it to disk if the user has requested
98 it, among some other tasks. In most cases there should be one per
99 program. As, given a video URL, the downloader doesn't know how to
100 extract all the needed information, task that InfoExtractors do, it
101 has to pass the URL to one of them.
103 For this, YoutubeDL objects have a method that allows
104 InfoExtractors to be registered in a given order. When it is passed
105 a URL, the YoutubeDL object handles it to the first InfoExtractor it
106 finds that reports being able to handle it. The InfoExtractor extracts
107 all the information about the video or videos the URL refers to, and
108 YoutubeDL process the extracted information, possibly using a File
109 Downloader to download the video.
111 YoutubeDL objects accept a lot of parameters. In order not to saturate
112 the object constructor with arguments, it receives a dictionary of
113 options instead. These options are available through the params
114 attribute for the InfoExtractors to use. The YoutubeDL also
115 registers itself as the downloader in charge for the InfoExtractors
116 that are added to it, so this is a "mutual registration".
120 username: Username for authentication purposes.
121 password: Password for authentication purposes.
122 videopassword: Password for accessing a video.
123 usenetrc: Use netrc for authentication instead.
124 verbose: Print additional info to stdout.
125 quiet: Do not print messages to stdout.
126 no_warnings: Do not print out anything for warnings.
127 forceurl: Force printing final URL.
128 forcetitle: Force printing title.
129 forceid: Force printing ID.
130 forcethumbnail: Force printing thumbnail URL.
131 forcedescription: Force printing description.
132 forcefilename: Force printing final filename.
133 forceduration: Force printing duration.
134 forcejson: Force printing info_dict as JSON.
135 dump_single_json: Force printing the info_dict of the whole playlist
136 (or video) as a single JSON line.
137 simulate: Do not download the video files.
138 format: Video format code. See options.py for more information.
139 outtmpl: Template for output names.
140 restrictfilenames: Do not allow "&" and spaces in file names
141 ignoreerrors: Do not stop on download errors.
142 force_generic_extractor: Force downloader to use the generic extractor
143 nooverwrites: Prevent overwriting files.
144 playliststart: Playlist item to start at.
145 playlistend: Playlist item to end at.
146 playlist_items: Specific indices of playlist to download.
147 playlistreverse: Download playlist items in reverse order.
148 matchtitle: Download only matching titles.
149 rejecttitle: Reject downloads for matching titles.
150 logger: Log messages to a logging.Logger instance.
151 logtostderr: Log messages to stderr instead of stdout.
152 writedescription: Write the video description to a .description file
153 writeinfojson: Write the video description to a .info.json file
154 writeannotations: Write the video annotations to a .annotations.xml file
155 writethumbnail: Write the thumbnail image to a file
156 write_all_thumbnails: Write all thumbnail formats to files
157 writesubtitles: Write the video subtitles to a file
158 writeautomaticsub: Write the automatic subtitles to a file
159 allsubtitles: Downloads all the subtitles of the video
160 (requires writesubtitles or writeautomaticsub)
161 listsubtitles: Lists all available subtitles for the video
162 subtitlesformat: The format code for subtitles
163 subtitleslangs: List of languages of the subtitles to download
164 keepvideo: Keep the video file after post-processing
165 daterange: A DateRange object, download only if the upload_date is in the range.
166 skip_download: Skip the actual download of the video file
167 cachedir: Location of the cache files in the filesystem.
168 False to disable filesystem cache.
169 noplaylist: Download single video instead of a playlist if in doubt.
170 age_limit: An integer representing the user's age in years.
171 Unsuitable videos for the given age are skipped.
172 min_views: An integer representing the minimum view count the video
173 must have in order to not be skipped.
174 Videos without view count information are always
175 downloaded. None for no limit.
176 max_views: An integer representing the maximum view count.
177 Videos that are more popular than that are not
179 Videos without view count information are always
180 downloaded. None for no limit.
181 download_archive: File name of a file where all downloads are recorded.
182 Videos already present in the file are not downloaded
184 cookiefile: File name where cookies should be read from and dumped to.
185 nocheckcertificate:Do not verify SSL certificates
186 prefer_insecure: Use HTTP instead of HTTPS to retrieve information.
187 At the moment, this is only supported by YouTube.
188 proxy: URL of the proxy server to use
189 cn_verification_proxy: URL of the proxy to use for IP address verification
190 on Chinese sites. (Experimental)
191 socket_timeout: Time to wait for unresponsive hosts, in seconds
192 bidi_workaround: Work around buggy terminals without bidirectional text
193 support, using fridibi
194 debug_printtraffic:Print out sent and received HTTP traffic
195 include_ads: Download ads as well
196 default_search: Prepend this string if an input url is not valid.
197 'auto' for elaborate guessing
198 encoding: Use this encoding instead of the system-specified.
199 extract_flat: Do not resolve URLs, return the immediate result.
200 Pass in 'in_playlist' to only show this behavior for
202 postprocessors: A list of dictionaries, each with an entry
203 * key: The name of the postprocessor. See
204 youtube_dl/postprocessor/__init__.py for a list.
205 as well as any further keyword arguments for the
207 progress_hooks: A list of functions that get called on download
208 progress, with a dictionary with the entries
209 * status: One of "downloading", "error", or "finished".
210 Check this first and ignore unknown values.
212 If status is one of "downloading", or "finished", the
213 following properties may also be present:
214 * filename: The final filename (always present)
215 * tmpfilename: The filename we're currently writing to
216 * downloaded_bytes: Bytes on disk
217 * total_bytes: Size of the whole file, None if unknown
218 * total_bytes_estimate: Guess of the eventual file size,
220 * elapsed: The number of seconds since download started.
221 * eta: The estimated time in seconds, None if unknown
222 * speed: The download speed in bytes/second, None if
224 * fragment_index: The counter of the currently
225 downloaded video fragment.
226 * fragment_count: The number of fragments (= individual
227 files that will be merged)
229 Progress hooks are guaranteed to be called at least once
230 (with status "finished") if the download is successful.
231 merge_output_format: Extension to use when merging formats.
232 fixup: Automatically correct known faults of the file.
234 - "never": do nothing
235 - "warn": only emit a warning
236 - "detect_or_warn": check whether we can do anything
237 about it, warn otherwise (default)
238 source_address: (Experimental) Client-side IP address to bind to.
239 call_home: Boolean, true iff we are allowed to contact the
240 youtube-dl servers for debugging.
241 sleep_interval: Number of seconds to sleep before each download.
242 listformats: Print an overview of available video formats and exit.
243 list_thumbnails: Print a table of all thumbnails and exit.
244 match_filter: A function that gets called with the info_dict of
246 If it returns a message, the video is ignored.
247 If it returns None, the video is downloaded.
248 match_filter_func in utils.py is one example for this.
249 no_color: Do not emit color codes in output.
251 The following options determine which downloader is picked:
252 external_downloader: Executable of the external downloader to call.
253 None or unset for standard (built-in) downloader.
254 hls_prefer_native: Use the native HLS downloader instead of ffmpeg/avconv.
256 The following parameters are not used by YoutubeDL itself, they are used by
257 the downloader (see youtube_dl/downloader/common.py):
258 nopart, updatetime, buffersize, ratelimit, min_filesize, max_filesize, test,
259 noresizebuffer, retries, continuedl, noprogress, consoletitle,
260 xattr_set_filesize, external_downloader_args.
262 The following options are used by the post processors:
263 prefer_ffmpeg: If True, use ffmpeg instead of avconv if both are available,
264 otherwise prefer avconv.
265 postprocessor_args: Extra parameters for external apps, like avconv.
271 _download_retcode = None
272 _num_downloads = None
275 def __init__(self, params=None, auto_init=True):
276 """Create a FileDownloader object with the given options."""
280 self._ies_instances = {}
282 self._progress_hooks = []
283 self._download_retcode = 0
284 self._num_downloads = 0
285 self._screen_file = [sys.stdout, sys.stderr][params.get('logtostderr', False)]
286 self._err_file = sys.stderr
288 self.cache = Cache(self)
290 if params.get('bidi_workaround', False):
293 master, slave = pty.openpty()
294 width = compat_get_terminal_size().columns
298 width_args = ['-w', str(width)]
300 stdin=subprocess.PIPE,
302 stderr=self._err_file)
304 self._output_process = subprocess.Popen(
305 ['bidiv'] + width_args, **sp_kwargs
308 self._output_process = subprocess.Popen(
309 ['fribidi', '-c', 'UTF-8'] + width_args, **sp_kwargs)
310 self._output_channel = os.fdopen(master, 'rb')
311 except OSError as ose:
313 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 $PATH.')
317 if (sys.version_info >= (3,) and sys.platform != 'win32' and
318 sys.getfilesystemencoding() in ['ascii', 'ANSI_X3.4-1968'] and
319 not params.get('restrictfilenames', False)):
320 # On Python 3, the Unicode filesystem API will throw errors (#1474)
322 'Assuming --restrict-filenames since file system encoding '
323 'cannot encode all characters. '
324 'Set the LC_ALL environment variable to fix this.')
325 self.params['restrictfilenames'] = True
327 if isinstance(params.get('outtmpl'), bytes):
329 'Parameter outtmpl is bytes, but should be a unicode string. '
330 'Put from __future__ import unicode_literals at the top of your code file or consider switching to Python 3.x.')
335 self.print_debug_header()
336 self.add_default_info_extractors()
338 for pp_def_raw in self.params.get('postprocessors', []):
339 pp_class = get_postprocessor(pp_def_raw['key'])
340 pp_def = dict(pp_def_raw)
342 pp = pp_class(self, **compat_kwargs(pp_def))
343 self.add_post_processor(pp)
345 for ph in self.params.get('progress_hooks', []):
346 self.add_progress_hook(ph)
348 def warn_if_short_id(self, argv):
349 # short YouTube ID starting with dash?
351 i for i, a in enumerate(argv)
352 if re.match(r'^-[0-9A-Za-z_-]{10}$', a)]
356 [a for i, a in enumerate(argv) if i not in idxs] +
357 ['--'] + [argv[i] for i in idxs]
360 'Long argument string detected. '
361 'Use -- to separate parameters and URLs, like this:\n%s\n' %
362 args_to_str(correct_argv))
364 def add_info_extractor(self, ie):
365 """Add an InfoExtractor object to the end of the list."""
367 self._ies_instances[ie.ie_key()] = ie
368 ie.set_downloader(self)
370 def get_info_extractor(self, ie_key):
372 Get an instance of an IE with name ie_key, it will try to get one from
373 the _ies list, if there's no instance it will create a new one and add
374 it to the extractor list.
376 ie = self._ies_instances.get(ie_key)
378 ie = get_info_extractor(ie_key)()
379 self.add_info_extractor(ie)
382 def add_default_info_extractors(self):
384 Add the InfoExtractors returned by gen_extractors to the end of the list
386 for ie in gen_extractors():
387 self.add_info_extractor(ie)
389 def add_post_processor(self, pp):
390 """Add a PostProcessor object to the end of the chain."""
392 pp.set_downloader(self)
394 def add_progress_hook(self, ph):
395 """Add the progress hook (currently only for the file downloader)"""
396 self._progress_hooks.append(ph)
398 def _bidi_workaround(self, message):
399 if not hasattr(self, '_output_channel'):
402 assert hasattr(self, '_output_process')
403 assert isinstance(message, compat_str)
404 line_count = message.count('\n') + 1
405 self._output_process.stdin.write((message + '\n').encode('utf-8'))
406 self._output_process.stdin.flush()
407 res = ''.join(self._output_channel.readline().decode('utf-8')
408 for _ in range(line_count))
409 return res[:-len('\n')]
411 def to_screen(self, message, skip_eol=False):
412 """Print message to stdout if not in quiet mode."""
413 return self.to_stdout(message, skip_eol, check_quiet=True)
415 def _write_string(self, s, out=None):
416 write_string(s, out=out, encoding=self.params.get('encoding'))
418 def to_stdout(self, message, skip_eol=False, check_quiet=False):
419 """Print message to stdout if not in quiet mode."""
420 if self.params.get('logger'):
421 self.params['logger'].debug(message)
422 elif not check_quiet or not self.params.get('quiet', False):
423 message = self._bidi_workaround(message)
424 terminator = ['\n', ''][skip_eol]
425 output = message + terminator
427 self._write_string(output, self._screen_file)
429 def to_stderr(self, message):
430 """Print message to stderr."""
431 assert isinstance(message, compat_str)
432 if self.params.get('logger'):
433 self.params['logger'].error(message)
435 message = self._bidi_workaround(message)
436 output = message + '\n'
437 self._write_string(output, self._err_file)
439 def to_console_title(self, message):
440 if not self.params.get('consoletitle', False):
442 if os.name == 'nt' and ctypes.windll.kernel32.GetConsoleWindow():
443 # c_wchar_p() might not be necessary if `message` is
444 # already of type unicode()
445 ctypes.windll.kernel32.SetConsoleTitleW(ctypes.c_wchar_p(message))
446 elif 'TERM' in os.environ:
447 self._write_string('\033]0;%s\007' % message, self._screen_file)
449 def save_console_title(self):
450 if not self.params.get('consoletitle', False):
452 if 'TERM' in os.environ:
453 # Save the title on stack
454 self._write_string('\033[22;0t', self._screen_file)
456 def restore_console_title(self):
457 if not self.params.get('consoletitle', False):
459 if 'TERM' in os.environ:
460 # Restore the title from stack
461 self._write_string('\033[23;0t', self._screen_file)
464 self.save_console_title()
467 def __exit__(self, *args):
468 self.restore_console_title()
470 if self.params.get('cookiefile') is not None:
471 self.cookiejar.save()
473 def trouble(self, message=None, tb=None):
474 """Determine action to take when a download problem appears.
476 Depending on if the downloader has been configured to ignore
477 download errors or not, this method may throw an exception or
478 not when errors are found, after printing the message.
480 tb, if given, is additional traceback information.
482 if message is not None:
483 self.to_stderr(message)
484 if self.params.get('verbose'):
486 if sys.exc_info()[0]: # if .trouble has been called from an except block
488 if hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]:
489 tb += ''.join(traceback.format_exception(*sys.exc_info()[1].exc_info))
490 tb += compat_str(traceback.format_exc())
492 tb_data = traceback.format_list(traceback.extract_stack())
493 tb = ''.join(tb_data)
495 if not self.params.get('ignoreerrors', False):
496 if sys.exc_info()[0] and hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]:
497 exc_info = sys.exc_info()[1].exc_info
499 exc_info = sys.exc_info()
500 raise DownloadError(message, exc_info)
501 self._download_retcode = 1
503 def report_warning(self, message):
505 Print the message to stderr, it will be prefixed with 'WARNING:'
506 If stderr is a tty file the 'WARNING:' will be colored
508 if self.params.get('logger') is not None:
509 self.params['logger'].warning(message)
511 if self.params.get('no_warnings'):
513 if not self.params.get('no_color') and self._err_file.isatty() and os.name != 'nt':
514 _msg_header = '\033[0;33mWARNING:\033[0m'
516 _msg_header = 'WARNING:'
517 warning_message = '%s %s' % (_msg_header, message)
518 self.to_stderr(warning_message)
520 def report_error(self, message, tb=None):
522 Do the same as trouble, but prefixes the message with 'ERROR:', colored
523 in red if stderr is a tty file.
525 if not self.params.get('no_color') and self._err_file.isatty() and os.name != 'nt':
526 _msg_header = '\033[0;31mERROR:\033[0m'
528 _msg_header = 'ERROR:'
529 error_message = '%s %s' % (_msg_header, message)
530 self.trouble(error_message, tb)
532 def report_file_already_downloaded(self, file_name):
533 """Report file has already been fully downloaded."""
535 self.to_screen('[download] %s has already been downloaded' % file_name)
536 except UnicodeEncodeError:
537 self.to_screen('[download] The file has already been downloaded')
539 def prepare_filename(self, info_dict):
540 """Generate the output filename."""
542 template_dict = dict(info_dict)
544 template_dict['epoch'] = int(time.time())
545 autonumber_size = self.params.get('autonumber_size')
546 if autonumber_size is None:
548 autonumber_templ = '%0' + str(autonumber_size) + 'd'
549 template_dict['autonumber'] = autonumber_templ % self._num_downloads
550 if template_dict.get('playlist_index') is not None:
551 template_dict['playlist_index'] = '%0*d' % (len(str(template_dict['n_entries'])), template_dict['playlist_index'])
552 if template_dict.get('resolution') is None:
553 if template_dict.get('width') and template_dict.get('height'):
554 template_dict['resolution'] = '%dx%d' % (template_dict['width'], template_dict['height'])
555 elif template_dict.get('height'):
556 template_dict['resolution'] = '%sp' % template_dict['height']
557 elif template_dict.get('width'):
558 template_dict['resolution'] = '?x%d' % template_dict['width']
560 sanitize = lambda k, v: sanitize_filename(
562 restricted=self.params.get('restrictfilenames'),
564 template_dict = dict((k, sanitize(k, v))
565 for k, v in template_dict.items()
567 template_dict = collections.defaultdict(lambda: 'NA', template_dict)
569 outtmpl = sanitize_path(self.params.get('outtmpl', DEFAULT_OUTTMPL))
570 tmpl = compat_expanduser(outtmpl)
571 filename = tmpl % template_dict
572 # Temporary fix for #4787
573 # 'Treat' all problem characters by passing filename through preferredencoding
574 # to workaround encoding issues with subprocess on python2 @ Windows
575 if sys.version_info < (3, 0) and sys.platform == 'win32':
576 filename = encodeFilename(filename, True).decode(preferredencoding())
578 except ValueError as err:
579 self.report_error('Error in output template: ' + str(err) + ' (encoding: ' + repr(preferredencoding()) + ')')
582 def _match_entry(self, info_dict, incomplete):
583 """ Returns None iff the file should be downloaded """
585 video_title = info_dict.get('title', info_dict.get('id', 'video'))
586 if 'title' in info_dict:
587 # This can happen when we're just evaluating the playlist
588 title = info_dict['title']
589 matchtitle = self.params.get('matchtitle', False)
591 if not re.search(matchtitle, title, re.IGNORECASE):
592 return '"' + title + '" title did not match pattern "' + matchtitle + '"'
593 rejecttitle = self.params.get('rejecttitle', False)
595 if re.search(rejecttitle, title, re.IGNORECASE):
596 return '"' + title + '" title matched reject pattern "' + rejecttitle + '"'
597 date = info_dict.get('upload_date', None)
599 dateRange = self.params.get('daterange', DateRange())
600 if date not in dateRange:
601 return '%s upload date is not in range %s' % (date_from_str(date).isoformat(), dateRange)
602 view_count = info_dict.get('view_count', None)
603 if view_count is not None:
604 min_views = self.params.get('min_views')
605 if min_views is not None and view_count < min_views:
606 return 'Skipping %s, because it has not reached minimum view count (%d/%d)' % (video_title, view_count, min_views)
607 max_views = self.params.get('max_views')
608 if max_views is not None and view_count > max_views:
609 return 'Skipping %s, because it has exceeded the maximum view count (%d/%d)' % (video_title, view_count, max_views)
610 if age_restricted(info_dict.get('age_limit'), self.params.get('age_limit')):
611 return 'Skipping "%s" because it is age restricted' % video_title
612 if self.in_download_archive(info_dict):
613 return '%s has already been recorded in archive' % video_title
616 match_filter = self.params.get('match_filter')
617 if match_filter is not None:
618 ret = match_filter(info_dict)
625 def add_extra_info(info_dict, extra_info):
626 '''Set the keys from extra_info in info dict if they are missing'''
627 for key, value in extra_info.items():
628 info_dict.setdefault(key, value)
630 def extract_info(self, url, download=True, ie_key=None, extra_info={},
631 process=True, force_generic_extractor=False):
633 Returns a list with a dictionary for each video we find.
634 If 'download', also downloads the videos.
635 extra_info is a dict containing the extra values to add to each result
638 if not ie_key and force_generic_extractor:
642 ies = [self.get_info_extractor(ie_key)]
647 if not ie.suitable(url):
651 self.report_warning('The program functionality for this site has been marked as broken, '
652 'and will probably not work.')
655 ie_result = ie.extract(url)
656 if ie_result is None: # Finished already (backwards compatibility; listformats and friends should be moved here)
658 if isinstance(ie_result, list):
659 # Backwards compatibility: old IE result format
661 '_type': 'compat_list',
662 'entries': ie_result,
664 self.add_default_extra_info(ie_result, ie, url)
666 return self.process_ie_result(ie_result, download, extra_info)
669 except ExtractorError as de: # An error we somewhat expected
670 self.report_error(compat_str(de), de.format_traceback())
672 except MaxDownloadsReached:
674 except Exception as e:
675 if self.params.get('ignoreerrors', False):
676 self.report_error(compat_str(e), tb=compat_str(traceback.format_exc()))
681 self.report_error('no suitable InfoExtractor for URL %s' % url)
683 def add_default_extra_info(self, ie_result, ie, url):
684 self.add_extra_info(ie_result, {
685 'extractor': ie.IE_NAME,
687 'webpage_url_basename': url_basename(url),
688 'extractor_key': ie.ie_key(),
691 def process_ie_result(self, ie_result, download=True, extra_info={}):
693 Take the result of the ie(may be modified) and resolve all unresolved
694 references (URLs, playlist items).
696 It will also download the videos if 'download'.
697 Returns the resolved ie_result.
700 result_type = ie_result.get('_type', 'video')
702 if result_type in ('url', 'url_transparent'):
703 extract_flat = self.params.get('extract_flat', False)
704 if ((extract_flat == 'in_playlist' and 'playlist' in extra_info) or
705 extract_flat is True):
706 if self.params.get('forcejson', False):
707 self.to_stdout(json.dumps(ie_result))
710 if result_type == 'video':
711 self.add_extra_info(ie_result, extra_info)
712 return self.process_video_result(ie_result, download=download)
713 elif result_type == 'url':
714 # We have to add extra_info to the results because it may be
715 # contained in a playlist
716 return self.extract_info(ie_result['url'],
718 ie_key=ie_result.get('ie_key'),
719 extra_info=extra_info)
720 elif result_type == 'url_transparent':
721 # Use the information from the embedding page
722 info = self.extract_info(
723 ie_result['url'], ie_key=ie_result.get('ie_key'),
724 extra_info=extra_info, download=False, process=False)
726 force_properties = dict(
727 (k, v) for k, v in ie_result.items() if v is not None)
728 for f in ('_type', 'url'):
729 if f in force_properties:
730 del force_properties[f]
731 new_result = info.copy()
732 new_result.update(force_properties)
734 assert new_result.get('_type') != 'url_transparent'
736 return self.process_ie_result(
737 new_result, download=download, extra_info=extra_info)
738 elif result_type == 'playlist' or result_type == 'multi_video':
739 # We process each entry in the playlist
740 playlist = ie_result.get('title', None) or ie_result.get('id', None)
741 self.to_screen('[download] Downloading playlist: %s' % playlist)
743 playlist_results = []
745 playliststart = self.params.get('playliststart', 1) - 1
746 playlistend = self.params.get('playlistend', None)
747 # For backwards compatibility, interpret -1 as whole list
748 if playlistend == -1:
751 playlistitems_str = self.params.get('playlist_items', None)
753 if playlistitems_str is not None:
754 def iter_playlistitems(format):
755 for string_segment in format.split(','):
756 if '-' in string_segment:
757 start, end = string_segment.split('-')
758 for item in range(int(start), int(end) + 1):
761 yield int(string_segment)
762 playlistitems = iter_playlistitems(playlistitems_str)
764 ie_entries = ie_result['entries']
765 if isinstance(ie_entries, list):
766 n_all_entries = len(ie_entries)
769 ie_entries[i - 1] for i in playlistitems
770 if -n_all_entries <= i - 1 < n_all_entries]
772 entries = ie_entries[playliststart:playlistend]
773 n_entries = len(entries)
775 "[%s] playlist %s: Collected %d video ids (downloading %d of them)" %
776 (ie_result['extractor'], playlist, n_all_entries, n_entries))
777 elif isinstance(ie_entries, PagedList):
780 for item in playlistitems:
781 entries.extend(ie_entries.getslice(
785 entries = ie_entries.getslice(
786 playliststart, playlistend)
787 n_entries = len(entries)
789 "[%s] playlist %s: Downloading %d videos" %
790 (ie_result['extractor'], playlist, n_entries))
793 entry_list = list(ie_entries)
794 entries = [entry_list[i - 1] for i in playlistitems]
796 entries = list(itertools.islice(
797 ie_entries, playliststart, playlistend))
798 n_entries = len(entries)
800 "[%s] playlist %s: Downloading %d videos" %
801 (ie_result['extractor'], playlist, n_entries))
803 if self.params.get('playlistreverse', False):
804 entries = entries[::-1]
806 for i, entry in enumerate(entries, 1):
807 self.to_screen('[download] Downloading video %s of %s' % (i, n_entries))
809 'n_entries': n_entries,
810 'playlist': playlist,
811 'playlist_id': ie_result.get('id'),
812 'playlist_title': ie_result.get('title'),
813 'playlist_index': i + playliststart,
814 'extractor': ie_result['extractor'],
815 'webpage_url': ie_result['webpage_url'],
816 'webpage_url_basename': url_basename(ie_result['webpage_url']),
817 'extractor_key': ie_result['extractor_key'],
820 reason = self._match_entry(entry, incomplete=True)
821 if reason is not None:
822 self.to_screen('[download] ' + reason)
825 entry_result = self.process_ie_result(entry,
828 playlist_results.append(entry_result)
829 ie_result['entries'] = playlist_results
831 elif result_type == 'compat_list':
833 'Extractor %s returned a compat_list result. '
834 'It needs to be updated.' % ie_result.get('extractor'))
840 'extractor': ie_result['extractor'],
841 'webpage_url': ie_result['webpage_url'],
842 'webpage_url_basename': url_basename(ie_result['webpage_url']),
843 'extractor_key': ie_result['extractor_key'],
847 ie_result['entries'] = [
848 self.process_ie_result(_fixup(r), download, extra_info)
849 for r in ie_result['entries']
853 raise Exception('Invalid result type: %s' % result_type)
855 def _apply_format_filter(self, format_spec, available_formats):
856 " Returns a tuple of the remaining format_spec and filtered formats "
866 operator_rex = re.compile(r'''(?x)\s*\[
867 (?P<key>width|height|tbr|abr|vbr|asr|filesize|fps)
868 \s*(?P<op>%s)(?P<none_inclusive>\s*\?)?\s*
869 (?P<value>[0-9.]+(?:[kKmMgGtTpPeEzZyY]i?[Bb]?)?)
871 ''' % '|'.join(map(re.escape, OPERATORS.keys())))
872 m = operator_rex.search(format_spec)
875 comparison_value = int(m.group('value'))
877 comparison_value = parse_filesize(m.group('value'))
878 if comparison_value is None:
879 comparison_value = parse_filesize(m.group('value') + 'B')
880 if comparison_value is None:
882 'Invalid value %r in format specification %r' % (
883 m.group('value'), format_spec))
884 op = OPERATORS[m.group('op')]
891 str_operator_rex = re.compile(r'''(?x)\s*\[
892 \s*(?P<key>ext|acodec|vcodec|container|protocol)
893 \s*(?P<op>%s)(?P<none_inclusive>\s*\?)?
894 \s*(?P<value>[a-zA-Z0-9_-]+)
896 ''' % '|'.join(map(re.escape, STR_OPERATORS.keys())))
897 m = str_operator_rex.search(format_spec)
899 comparison_value = m.group('value')
900 op = STR_OPERATORS[m.group('op')]
903 raise ValueError('Invalid format specification %r' % format_spec)
906 actual_value = f.get(m.group('key'))
907 if actual_value is None:
908 return m.group('none_inclusive')
909 return op(actual_value, comparison_value)
910 new_formats = [f for f in available_formats if _filter(f)]
912 new_format_spec = format_spec[:-len(m.group(0))]
913 if not new_format_spec:
914 new_format_spec = 'best'
916 return (new_format_spec, new_formats)
918 def select_format(self, format_spec, available_formats):
919 while format_spec.endswith(']'):
920 format_spec, available_formats = self._apply_format_filter(
921 format_spec, available_formats)
922 if not available_formats:
925 if format_spec in ['best', 'worst', None]:
926 format_idx = 0 if format_spec == 'worst' else -1
927 audiovideo_formats = [
928 f for f in available_formats
929 if f.get('vcodec') != 'none' and f.get('acodec') != 'none']
930 if audiovideo_formats:
931 return audiovideo_formats[format_idx]
932 # for audio only (soundcloud) or video only (imgur) urls, select the best/worst audio format
933 elif (all(f.get('acodec') != 'none' for f in available_formats) or
934 all(f.get('vcodec') != 'none' for f in available_formats)):
935 return available_formats[format_idx]
936 elif format_spec == 'bestaudio':
938 f for f in available_formats
939 if f.get('vcodec') == 'none']
941 return audio_formats[-1]
942 elif format_spec == 'worstaudio':
944 f for f in available_formats
945 if f.get('vcodec') == 'none']
947 return audio_formats[0]
948 elif format_spec == 'bestvideo':
950 f for f in available_formats
951 if f.get('acodec') == 'none']
953 return video_formats[-1]
954 elif format_spec == 'worstvideo':
956 f for f in available_formats
957 if f.get('acodec') == 'none']
959 return video_formats[0]
961 extensions = ['mp4', 'flv', 'webm', '3gp', 'm4a', 'mp3', 'ogg', 'aac', 'wav']
962 if format_spec in extensions:
963 filter_f = lambda f: f['ext'] == format_spec
965 filter_f = lambda f: f['format_id'] == format_spec
966 matches = list(filter(filter_f, available_formats))
971 def _calc_headers(self, info_dict):
972 res = std_headers.copy()
974 add_headers = info_dict.get('http_headers')
976 res.update(add_headers)
978 cookies = self._calc_cookies(info_dict)
980 res['Cookie'] = cookies
984 def _calc_cookies(self, info_dict):
985 pr = compat_urllib_request.Request(info_dict['url'])
986 self.cookiejar.add_cookie_header(pr)
987 return pr.get_header('Cookie')
989 def process_video_result(self, info_dict, download=True):
990 assert info_dict.get('_type', 'video') == 'video'
992 if 'id' not in info_dict:
993 raise ExtractorError('Missing "id" field in extractor result')
994 if 'title' not in info_dict:
995 raise ExtractorError('Missing "title" field in extractor result')
997 if 'playlist' not in info_dict:
998 # It isn't part of a playlist
999 info_dict['playlist'] = None
1000 info_dict['playlist_index'] = None
1002 thumbnails = info_dict.get('thumbnails')
1003 if thumbnails is None:
1004 thumbnail = info_dict.get('thumbnail')
1006 info_dict['thumbnails'] = thumbnails = [{'url': thumbnail}]
1008 thumbnails.sort(key=lambda t: (
1009 t.get('preference'), t.get('width'), t.get('height'),
1010 t.get('id'), t.get('url')))
1011 for i, t in enumerate(thumbnails):
1012 if t.get('width') and t.get('height'):
1013 t['resolution'] = '%dx%d' % (t['width'], t['height'])
1014 if t.get('id') is None:
1017 if thumbnails and 'thumbnail' not in info_dict:
1018 info_dict['thumbnail'] = thumbnails[-1]['url']
1020 if 'display_id' not in info_dict and 'id' in info_dict:
1021 info_dict['display_id'] = info_dict['id']
1023 if info_dict.get('upload_date') is None and info_dict.get('timestamp') is not None:
1024 # Working around out-of-range timestamp values (e.g. negative ones on Windows,
1025 # see http://bugs.python.org/issue1646728)
1027 upload_date = datetime.datetime.utcfromtimestamp(info_dict['timestamp'])
1028 info_dict['upload_date'] = upload_date.strftime('%Y%m%d')
1029 except (ValueError, OverflowError, OSError):
1032 if self.params.get('listsubtitles', False):
1033 if 'automatic_captions' in info_dict:
1034 self.list_subtitles(info_dict['id'], info_dict.get('automatic_captions'), 'automatic captions')
1035 self.list_subtitles(info_dict['id'], info_dict.get('subtitles'), 'subtitles')
1037 info_dict['requested_subtitles'] = self.process_subtitles(
1038 info_dict['id'], info_dict.get('subtitles'),
1039 info_dict.get('automatic_captions'))
1041 # We now pick which formats have to be downloaded
1042 if info_dict.get('formats') is None:
1043 # There's only one format available
1044 formats = [info_dict]
1046 formats = info_dict['formats']
1049 raise ExtractorError('No video formats found!')
1053 # We check that all the formats have the format and format_id fields
1054 for i, format in enumerate(formats):
1055 if 'url' not in format:
1056 raise ExtractorError('Missing "url" key in result (index %d)' % i)
1058 if format.get('format_id') is None:
1059 format['format_id'] = compat_str(i)
1060 format_id = format['format_id']
1061 if format_id not in formats_dict:
1062 formats_dict[format_id] = []
1063 formats_dict[format_id].append(format)
1065 # Make sure all formats have unique format_id
1066 for format_id, ambiguous_formats in formats_dict.items():
1067 if len(ambiguous_formats) > 1:
1068 for i, format in enumerate(ambiguous_formats):
1069 format['format_id'] = '%s-%d' % (format_id, i)
1071 for i, format in enumerate(formats):
1072 if format.get('format') is None:
1073 format['format'] = '{id} - {res}{note}'.format(
1074 id=format['format_id'],
1075 res=self.format_resolution(format),
1076 note=' ({0})'.format(format['format_note']) if format.get('format_note') is not None else '',
1078 # Automatically determine file extension if missing
1079 if 'ext' not in format:
1080 format['ext'] = determine_ext(format['url']).lower()
1081 # Add HTTP headers, so that external programs can use them from the
1083 full_format_info = info_dict.copy()
1084 full_format_info.update(format)
1085 format['http_headers'] = self._calc_headers(full_format_info)
1087 # TODO Central sorting goes here
1089 if formats[0] is not info_dict:
1090 # only set the 'formats' fields if the original info_dict list them
1091 # otherwise we end up with a circular reference, the first (and unique)
1092 # element in the 'formats' field in info_dict is info_dict itself,
1093 # wich can't be exported to json
1094 info_dict['formats'] = formats
1095 if self.params.get('listformats'):
1096 self.list_formats(info_dict)
1098 if self.params.get('list_thumbnails'):
1099 self.list_thumbnails(info_dict)
1102 req_format = self.params.get('format')
1103 if req_format is None:
1104 req_format_list = []
1105 if (self.params.get('outtmpl', DEFAULT_OUTTMPL) != '-' and
1106 info_dict['extractor'] in ['youtube', 'ted']):
1107 merger = FFmpegMergerPP(self)
1108 if merger.available and merger.can_merge():
1109 req_format_list.append('bestvideo+bestaudio')
1110 req_format_list.append('best')
1111 req_format = '/'.join(req_format_list)
1112 formats_to_download = []
1113 if req_format == 'all':
1114 formats_to_download = formats
1116 for rfstr in req_format.split(','):
1117 # We can accept formats requested in the format: 34/5/best, we pick
1118 # the first that is available, starting from left
1119 req_formats = rfstr.split('/')
1120 for rf in req_formats:
1121 if re.match(r'.+?\+.+?', rf) is not None:
1122 # Two formats have been requested like '137+139'
1123 format_1, format_2 = rf.split('+')
1124 formats_info = (self.select_format(format_1, formats),
1125 self.select_format(format_2, formats))
1126 if all(formats_info):
1127 # The first format must contain the video and the
1129 if formats_info[0].get('vcodec') == 'none':
1130 self.report_error('The first format must '
1131 'contain the video, try using '
1132 '"-f %s+%s"' % (format_2, format_1))
1135 formats_info[0]['ext']
1136 if self.params.get('merge_output_format') is None
1137 else self.params['merge_output_format'])
1139 'requested_formats': formats_info,
1140 'format': '%s+%s' % (formats_info[0].get('format'),
1141 formats_info[1].get('format')),
1142 'format_id': '%s+%s' % (formats_info[0].get('format_id'),
1143 formats_info[1].get('format_id')),
1144 'width': formats_info[0].get('width'),
1145 'height': formats_info[0].get('height'),
1146 'resolution': formats_info[0].get('resolution'),
1147 'fps': formats_info[0].get('fps'),
1148 'vcodec': formats_info[0].get('vcodec'),
1149 'vbr': formats_info[0].get('vbr'),
1150 'stretched_ratio': formats_info[0].get('stretched_ratio'),
1151 'acodec': formats_info[1].get('acodec'),
1152 'abr': formats_info[1].get('abr'),
1156 selected_format = None
1158 selected_format = self.select_format(rf, formats)
1159 if selected_format is not None:
1160 formats_to_download.append(selected_format)
1162 if not formats_to_download:
1163 raise ExtractorError('requested format not available',
1167 if len(formats_to_download) > 1:
1168 self.to_screen('[info] %s: downloading video in %s formats' % (info_dict['id'], len(formats_to_download)))
1169 for format in formats_to_download:
1170 new_info = dict(info_dict)
1171 new_info.update(format)
1172 self.process_info(new_info)
1173 # We update the info dict with the best quality format (backwards compatibility)
1174 info_dict.update(formats_to_download[-1])
1177 def process_subtitles(self, video_id, normal_subtitles, automatic_captions):
1178 """Select the requested subtitles and their format"""
1180 if normal_subtitles and self.params.get('writesubtitles'):
1181 available_subs.update(normal_subtitles)
1182 if automatic_captions and self.params.get('writeautomaticsub'):
1183 for lang, cap_info in automatic_captions.items():
1184 if lang not in available_subs:
1185 available_subs[lang] = cap_info
1187 if (not self.params.get('writesubtitles') and not
1188 self.params.get('writeautomaticsub') or not
1192 if self.params.get('allsubtitles', False):
1193 requested_langs = available_subs.keys()
1195 if self.params.get('subtitleslangs', False):
1196 requested_langs = self.params.get('subtitleslangs')
1197 elif 'en' in available_subs:
1198 requested_langs = ['en']
1200 requested_langs = [list(available_subs.keys())[0]]
1202 formats_query = self.params.get('subtitlesformat', 'best')
1203 formats_preference = formats_query.split('/') if formats_query else []
1205 for lang in requested_langs:
1206 formats = available_subs.get(lang)
1208 self.report_warning('%s subtitles not available for %s' % (lang, video_id))
1210 for ext in formats_preference:
1214 matches = list(filter(lambda f: f['ext'] == ext, formats))
1220 self.report_warning(
1221 'No subtitle format found matching "%s" for language %s, '
1222 'using %s' % (formats_query, lang, f['ext']))
1226 def process_info(self, info_dict):
1227 """Process a single resolved IE result."""
1229 assert info_dict.get('_type', 'video') == 'video'
1231 max_downloads = self.params.get('max_downloads')
1232 if max_downloads is not None:
1233 if self._num_downloads >= int(max_downloads):
1234 raise MaxDownloadsReached()
1236 info_dict['fulltitle'] = info_dict['title']
1237 if len(info_dict['title']) > 200:
1238 info_dict['title'] = info_dict['title'][:197] + '...'
1240 if 'format' not in info_dict:
1241 info_dict['format'] = info_dict['ext']
1243 reason = self._match_entry(info_dict, incomplete=False)
1244 if reason is not None:
1245 self.to_screen('[download] ' + reason)
1248 self._num_downloads += 1
1250 info_dict['_filename'] = filename = self.prepare_filename(info_dict)
1253 if self.params.get('forcetitle', False):
1254 self.to_stdout(info_dict['fulltitle'])
1255 if self.params.get('forceid', False):
1256 self.to_stdout(info_dict['id'])
1257 if self.params.get('forceurl', False):
1258 if info_dict.get('requested_formats') is not None:
1259 for f in info_dict['requested_formats']:
1260 self.to_stdout(f['url'] + f.get('play_path', ''))
1262 # For RTMP URLs, also include the playpath
1263 self.to_stdout(info_dict['url'] + info_dict.get('play_path', ''))
1264 if self.params.get('forcethumbnail', False) and info_dict.get('thumbnail') is not None:
1265 self.to_stdout(info_dict['thumbnail'])
1266 if self.params.get('forcedescription', False) and info_dict.get('description') is not None:
1267 self.to_stdout(info_dict['description'])
1268 if self.params.get('forcefilename', False) and filename is not None:
1269 self.to_stdout(filename)
1270 if self.params.get('forceduration', False) and info_dict.get('duration') is not None:
1271 self.to_stdout(formatSeconds(info_dict['duration']))
1272 if self.params.get('forceformat', False):
1273 self.to_stdout(info_dict['format'])
1274 if self.params.get('forcejson', False):
1275 self.to_stdout(json.dumps(info_dict))
1277 # Do nothing else if in simulate mode
1278 if self.params.get('simulate', False):
1281 if filename is None:
1285 dn = os.path.dirname(sanitize_path(encodeFilename(filename)))
1286 if dn and not os.path.exists(dn):
1288 except (OSError, IOError) as err:
1289 self.report_error('unable to create directory ' + compat_str(err))
1292 if self.params.get('writedescription', False):
1293 descfn = replace_extension(filename, 'description', info_dict.get('ext'))
1294 if self.params.get('nooverwrites', False) and os.path.exists(encodeFilename(descfn)):
1295 self.to_screen('[info] Video description is already present')
1296 elif info_dict.get('description') is None:
1297 self.report_warning('There\'s no description to write.')
1300 self.to_screen('[info] Writing video description to: ' + descfn)
1301 with io.open(encodeFilename(descfn), 'w', encoding='utf-8') as descfile:
1302 descfile.write(info_dict['description'])
1303 except (OSError, IOError):
1304 self.report_error('Cannot write description file ' + descfn)
1307 if self.params.get('writeannotations', False):
1308 annofn = replace_extension(filename, 'annotations.xml', info_dict.get('ext'))
1309 if self.params.get('nooverwrites', False) and os.path.exists(encodeFilename(annofn)):
1310 self.to_screen('[info] Video annotations are already present')
1313 self.to_screen('[info] Writing video annotations to: ' + annofn)
1314 with io.open(encodeFilename(annofn), 'w', encoding='utf-8') as annofile:
1315 annofile.write(info_dict['annotations'])
1316 except (KeyError, TypeError):
1317 self.report_warning('There are no annotations to write.')
1318 except (OSError, IOError):
1319 self.report_error('Cannot write annotations file: ' + annofn)
1322 subtitles_are_requested = any([self.params.get('writesubtitles', False),
1323 self.params.get('writeautomaticsub')])
1325 if subtitles_are_requested and info_dict.get('requested_subtitles'):
1326 # subtitles download errors are already managed as troubles in relevant IE
1327 # that way it will silently go on when used with unsupporting IE
1328 subtitles = info_dict['requested_subtitles']
1329 ie = self.get_info_extractor(info_dict['extractor_key'])
1330 for sub_lang, sub_info in subtitles.items():
1331 sub_format = sub_info['ext']
1332 if sub_info.get('data') is not None:
1333 sub_data = sub_info['data']
1336 sub_data = ie._download_webpage(
1337 sub_info['url'], info_dict['id'], note=False)
1338 except ExtractorError as err:
1339 self.report_warning('Unable to download subtitle for "%s": %s' %
1340 (sub_lang, compat_str(err.cause)))
1343 sub_filename = subtitles_filename(filename, sub_lang, sub_format)
1344 if self.params.get('nooverwrites', False) and os.path.exists(encodeFilename(sub_filename)):
1345 self.to_screen('[info] Video subtitle %s.%s is already_present' % (sub_lang, sub_format))
1347 self.to_screen('[info] Writing video subtitles to: ' + sub_filename)
1348 with io.open(encodeFilename(sub_filename), 'w', encoding='utf-8') as subfile:
1349 subfile.write(sub_data)
1350 except (OSError, IOError):
1351 self.report_error('Cannot write subtitles file ' + sub_filename)
1354 if self.params.get('writeinfojson', False):
1355 infofn = replace_extension(filename, 'info.json', info_dict.get('ext'))
1356 if self.params.get('nooverwrites', False) and os.path.exists(encodeFilename(infofn)):
1357 self.to_screen('[info] Video description metadata is already present')
1359 self.to_screen('[info] Writing video description metadata as JSON to: ' + infofn)
1361 write_json_file(self.filter_requested_info(info_dict), infofn)
1362 except (OSError, IOError):
1363 self.report_error('Cannot write metadata to JSON file ' + infofn)
1366 self._write_thumbnails(info_dict, filename)
1368 if not self.params.get('skip_download', False):
1371 fd = get_suitable_downloader(info, self.params)(self, self.params)
1372 for ph in self._progress_hooks:
1373 fd.add_progress_hook(ph)
1374 if self.params.get('verbose'):
1375 self.to_stdout('[debug] Invoking downloader on %r' % info.get('url'))
1376 return fd.download(name, info)
1378 if info_dict.get('requested_formats') is not None:
1381 merger = FFmpegMergerPP(self)
1382 if not merger.available:
1384 self.report_warning('You have requested multiple '
1385 'formats but ffmpeg or avconv are not installed.'
1386 ' The formats won\'t be merged.')
1388 postprocessors = [merger]
1390 def compatible_formats(formats):
1391 video, audio = formats
1393 video_ext, audio_ext = audio.get('ext'), video.get('ext')
1394 if video_ext and audio_ext:
1396 ('mp3', 'mp4', 'm4a', 'm4p', 'm4b', 'm4r', 'm4v'),
1399 for exts in COMPATIBLE_EXTS:
1400 if video_ext in exts and audio_ext in exts:
1402 # TODO: Check acodec/vcodec
1405 filename_real_ext = os.path.splitext(filename)[1][1:]
1407 os.path.splitext(filename)[0]
1408 if filename_real_ext == info_dict['ext']
1410 requested_formats = info_dict['requested_formats']
1411 if self.params.get('merge_output_format') is None and not compatible_formats(requested_formats):
1412 info_dict['ext'] = 'mkv'
1413 self.report_warning(
1414 'Requested formats are incompatible for merge and will be merged into mkv.')
1415 # Ensure filename always has a correct extension for successful merge
1416 filename = '%s.%s' % (filename_wo_ext, info_dict['ext'])
1417 if os.path.exists(encodeFilename(filename)):
1419 '[download] %s has already been downloaded and '
1420 'merged' % filename)
1422 for f in requested_formats:
1423 new_info = dict(info_dict)
1425 fname = self.prepare_filename(new_info)
1426 fname = prepend_extension(fname, 'f%s' % f['format_id'], new_info['ext'])
1427 downloaded.append(fname)
1428 partial_success = dl(fname, new_info)
1429 success = success and partial_success
1430 info_dict['__postprocessors'] = postprocessors
1431 info_dict['__files_to_merge'] = downloaded
1433 # Just a single file
1434 success = dl(filename, info_dict)
1435 except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
1436 self.report_error('unable to download video data: %s' % str(err))
1438 except (OSError, IOError) as err:
1439 raise UnavailableVideoError(err)
1440 except (ContentTooShortError, ) as err:
1441 self.report_error('content too short (expected %s bytes and served %s)' % (err.expected, err.downloaded))
1446 fixup_policy = self.params.get('fixup')
1447 if fixup_policy is None:
1448 fixup_policy = 'detect_or_warn'
1450 stretched_ratio = info_dict.get('stretched_ratio')
1451 if stretched_ratio is not None and stretched_ratio != 1:
1452 if fixup_policy == 'warn':
1453 self.report_warning('%s: Non-uniform pixel ratio (%s)' % (
1454 info_dict['id'], stretched_ratio))
1455 elif fixup_policy == 'detect_or_warn':
1456 stretched_pp = FFmpegFixupStretchedPP(self)
1457 if stretched_pp.available:
1458 info_dict.setdefault('__postprocessors', [])
1459 info_dict['__postprocessors'].append(stretched_pp)
1461 self.report_warning(
1462 '%s: Non-uniform pixel ratio (%s). Install ffmpeg or avconv to fix this automatically.' % (
1463 info_dict['id'], stretched_ratio))
1465 assert fixup_policy in ('ignore', 'never')
1467 if info_dict.get('requested_formats') is None and info_dict.get('container') == 'm4a_dash':
1468 if fixup_policy == 'warn':
1469 self.report_warning('%s: writing DASH m4a. Only some players support this container.' % (
1471 elif fixup_policy == 'detect_or_warn':
1472 fixup_pp = FFmpegFixupM4aPP(self)
1473 if fixup_pp.available:
1474 info_dict.setdefault('__postprocessors', [])
1475 info_dict['__postprocessors'].append(fixup_pp)
1477 self.report_warning(
1478 '%s: writing DASH m4a. Only some players support this container. Install ffmpeg or avconv to fix this automatically.' % (
1481 assert fixup_policy in ('ignore', 'never')
1484 self.post_process(filename, info_dict)
1485 except (PostProcessingError) as err:
1486 self.report_error('postprocessing: %s' % str(err))
1488 self.record_download_archive(info_dict)
1490 def download(self, url_list):
1491 """Download a given list of URLs."""
1492 outtmpl = self.params.get('outtmpl', DEFAULT_OUTTMPL)
1493 if (len(url_list) > 1 and
1494 '%' not in outtmpl and
1495 self.params.get('max_downloads') != 1):
1496 raise SameFileError(outtmpl)
1498 for url in url_list:
1500 # It also downloads the videos
1501 res = self.extract_info(
1502 url, force_generic_extractor=self.params.get('force_generic_extractor', False))
1503 except UnavailableVideoError:
1504 self.report_error('unable to download video')
1505 except MaxDownloadsReached:
1506 self.to_screen('[info] Maximum number of downloaded files reached.')
1509 if self.params.get('dump_single_json', False):
1510 self.to_stdout(json.dumps(res))
1512 return self._download_retcode
1514 def download_with_info_file(self, info_filename):
1515 with contextlib.closing(fileinput.FileInput(
1516 [info_filename], mode='r',
1517 openhook=fileinput.hook_encoded('utf-8'))) as f:
1518 # FileInput doesn't have a read method, we can't call json.load
1519 info = self.filter_requested_info(json.loads('\n'.join(f)))
1521 self.process_ie_result(info, download=True)
1522 except DownloadError:
1523 webpage_url = info.get('webpage_url')
1524 if webpage_url is not None:
1525 self.report_warning('The info failed to download, trying with "%s"' % webpage_url)
1526 return self.download([webpage_url])
1529 return self._download_retcode
1532 def filter_requested_info(info_dict):
1534 (k, v) for k, v in info_dict.items()
1535 if k not in ['requested_formats', 'requested_subtitles'])
1537 def post_process(self, filename, ie_info):
1538 """Run all the postprocessors on the given file."""
1539 info = dict(ie_info)
1540 info['filepath'] = filename
1542 if ie_info.get('__postprocessors') is not None:
1543 pps_chain.extend(ie_info['__postprocessors'])
1544 pps_chain.extend(self._pps)
1545 for pp in pps_chain:
1546 files_to_delete = []
1548 files_to_delete, info = pp.run(info)
1549 except PostProcessingError as e:
1550 self.report_error(e.msg)
1551 if files_to_delete and not self.params.get('keepvideo', False):
1552 for old_filename in files_to_delete:
1553 self.to_screen('Deleting original file %s (pass -k to keep)' % old_filename)
1555 os.remove(encodeFilename(old_filename))
1556 except (IOError, OSError):
1557 self.report_warning('Unable to remove downloaded original file')
1559 def _make_archive_id(self, info_dict):
1560 # Future-proof against any change in case
1561 # and backwards compatibility with prior versions
1562 extractor = info_dict.get('extractor_key')
1563 if extractor is None:
1564 if 'id' in info_dict:
1565 extractor = info_dict.get('ie_key') # key in a playlist
1566 if extractor is None:
1567 return None # Incomplete video information
1568 return extractor.lower() + ' ' + info_dict['id']
1570 def in_download_archive(self, info_dict):
1571 fn = self.params.get('download_archive')
1575 vid_id = self._make_archive_id(info_dict)
1577 return False # Incomplete video information
1580 with locked_file(fn, 'r', encoding='utf-8') as archive_file:
1581 for line in archive_file:
1582 if line.strip() == vid_id:
1584 except IOError as ioe:
1585 if ioe.errno != errno.ENOENT:
1589 def record_download_archive(self, info_dict):
1590 fn = self.params.get('download_archive')
1593 vid_id = self._make_archive_id(info_dict)
1595 with locked_file(fn, 'a', encoding='utf-8') as archive_file:
1596 archive_file.write(vid_id + '\n')
1599 def format_resolution(format, default='unknown'):
1600 if format.get('vcodec') == 'none':
1602 if format.get('resolution') is not None:
1603 return format['resolution']
1604 if format.get('height') is not None:
1605 if format.get('width') is not None:
1606 res = '%sx%s' % (format['width'], format['height'])
1608 res = '%sp' % format['height']
1609 elif format.get('width') is not None:
1610 res = '?x%d' % format['width']
1615 def _format_note(self, fdict):
1617 if fdict.get('ext') in ['f4f', 'f4m']:
1618 res += '(unsupported) '
1619 if fdict.get('format_note') is not None:
1620 res += fdict['format_note'] + ' '
1621 if fdict.get('tbr') is not None:
1622 res += '%4dk ' % fdict['tbr']
1623 if fdict.get('container') is not None:
1626 res += '%s container' % fdict['container']
1627 if (fdict.get('vcodec') is not None and
1628 fdict.get('vcodec') != 'none'):
1631 res += fdict['vcodec']
1632 if fdict.get('vbr') is not None:
1634 elif fdict.get('vbr') is not None and fdict.get('abr') is not None:
1636 if fdict.get('vbr') is not None:
1637 res += '%4dk' % fdict['vbr']
1638 if fdict.get('fps') is not None:
1639 res += ', %sfps' % fdict['fps']
1640 if fdict.get('acodec') is not None:
1643 if fdict['acodec'] == 'none':
1646 res += '%-5s' % fdict['acodec']
1647 elif fdict.get('abr') is not None:
1651 if fdict.get('abr') is not None:
1652 res += '@%3dk' % fdict['abr']
1653 if fdict.get('asr') is not None:
1654 res += ' (%5dHz)' % fdict['asr']
1655 if fdict.get('filesize') is not None:
1658 res += format_bytes(fdict['filesize'])
1659 elif fdict.get('filesize_approx') is not None:
1662 res += '~' + format_bytes(fdict['filesize_approx'])
1665 def list_formats(self, info_dict):
1666 formats = info_dict.get('formats', [info_dict])
1668 [f['format_id'], f['ext'], self.format_resolution(f), self._format_note(f)]
1670 if f.get('preference') is None or f['preference'] >= -1000]
1671 if len(formats) > 1:
1672 table[-1][-1] += (' ' if table[-1][-1] else '') + '(best)'
1674 header_line = ['format code', 'extension', 'resolution', 'note']
1676 '[info] Available formats for %s:\n%s' %
1677 (info_dict['id'], render_table(header_line, table)))
1679 def list_thumbnails(self, info_dict):
1680 thumbnails = info_dict.get('thumbnails')
1682 tn_url = info_dict.get('thumbnail')
1684 thumbnails = [{'id': '0', 'url': tn_url}]
1687 '[info] No thumbnails present for %s' % info_dict['id'])
1691 '[info] Thumbnails for %s:' % info_dict['id'])
1692 self.to_screen(render_table(
1693 ['ID', 'width', 'height', 'URL'],
1694 [[t['id'], t.get('width', 'unknown'), t.get('height', 'unknown'), t['url']] for t in thumbnails]))
1696 def list_subtitles(self, video_id, subtitles, name='subtitles'):
1698 self.to_screen('%s has no %s' % (video_id, name))
1701 'Available %s for %s:' % (name, video_id))
1702 self.to_screen(render_table(
1703 ['Language', 'formats'],
1704 [[lang, ', '.join(f['ext'] for f in reversed(formats))]
1705 for lang, formats in subtitles.items()]))
1707 def urlopen(self, req):
1708 """ Start an HTTP download """
1710 # According to RFC 3986, URLs can not contain non-ASCII characters, however this is not
1711 # always respected by websites, some tend to give out URLs with non percent-encoded
1712 # non-ASCII characters (see telemb.py, ard.py [#3412])
1713 # urllib chokes on URLs with non-ASCII characters (see http://bugs.python.org/issue3991)
1714 # To work around aforementioned issue we will replace request's original URL with
1715 # percent-encoded one
1716 req_is_string = isinstance(req, compat_basestring)
1717 url = req if req_is_string else req.get_full_url()
1718 url_escaped = escape_url(url)
1720 # Substitute URL if any change after escaping
1721 if url != url_escaped:
1725 req_type = HEADRequest if req.get_method() == 'HEAD' else compat_urllib_request.Request
1727 url_escaped, data=req.data, headers=req.headers,
1728 origin_req_host=req.origin_req_host, unverifiable=req.unverifiable)
1730 return self._opener.open(req, timeout=self._socket_timeout)
1732 def print_debug_header(self):
1733 if not self.params.get('verbose'):
1736 if type('') is not compat_str:
1737 # Python 2.6 on SLES11 SP1 (https://github.com/rg3/youtube-dl/issues/3326)
1738 self.report_warning(
1739 'Your Python is broken! Update to a newer and supported version')
1741 stdout_encoding = getattr(
1742 sys.stdout, 'encoding', 'missing (%s)' % type(sys.stdout).__name__)
1744 '[debug] Encodings: locale %s, fs %s, out %s, pref %s\n' % (
1745 locale.getpreferredencoding(),
1746 sys.getfilesystemencoding(),
1748 self.get_encoding()))
1749 write_string(encoding_str, encoding=None)
1751 self._write_string('[debug] youtube-dl version ' + __version__ + '\n')
1753 sp = subprocess.Popen(
1754 ['git', 'rev-parse', '--short', 'HEAD'],
1755 stdout=subprocess.PIPE, stderr=subprocess.PIPE,
1756 cwd=os.path.dirname(os.path.abspath(__file__)))
1757 out, err = sp.communicate()
1758 out = out.decode().strip()
1759 if re.match('[0-9a-f]+', out):
1760 self._write_string('[debug] Git HEAD: ' + out + '\n')
1766 self._write_string('[debug] Python version %s - %s\n' % (
1767 platform.python_version(), platform_name()))
1769 exe_versions = FFmpegPostProcessor.get_versions(self)
1770 exe_versions['rtmpdump'] = rtmpdump_version()
1771 exe_str = ', '.join(
1773 for exe, v in sorted(exe_versions.items())
1778 self._write_string('[debug] exe versions: %s\n' % exe_str)
1781 for handler in self._opener.handlers:
1782 if hasattr(handler, 'proxies'):
1783 proxy_map.update(handler.proxies)
1784 self._write_string('[debug] Proxy map: ' + compat_str(proxy_map) + '\n')
1786 if self.params.get('call_home', False):
1787 ipaddr = self.urlopen('https://yt-dl.org/ip').read().decode('utf-8')
1788 self._write_string('[debug] Public IP address: %s\n' % ipaddr)
1789 latest_version = self.urlopen(
1790 'https://yt-dl.org/latest/version').read().decode('utf-8')
1791 if version_tuple(latest_version) > version_tuple(__version__):
1792 self.report_warning(
1793 'You are using an outdated version (newest version: %s)! '
1794 'See https://yt-dl.org/update if you need help updating.' %
1797 def _setup_opener(self):
1798 timeout_val = self.params.get('socket_timeout')
1799 self._socket_timeout = 600 if timeout_val is None else float(timeout_val)
1801 opts_cookiefile = self.params.get('cookiefile')
1802 opts_proxy = self.params.get('proxy')
1804 if opts_cookiefile is None:
1805 self.cookiejar = compat_cookiejar.CookieJar()
1807 self.cookiejar = compat_cookiejar.MozillaCookieJar(
1809 if os.access(opts_cookiefile, os.R_OK):
1810 self.cookiejar.load()
1812 cookie_processor = compat_urllib_request.HTTPCookieProcessor(
1814 if opts_proxy is not None:
1815 if opts_proxy == '':
1818 proxies = {'http': opts_proxy, 'https': opts_proxy}
1820 proxies = compat_urllib_request.getproxies()
1821 # Set HTTPS proxy to HTTP one if given (https://github.com/rg3/youtube-dl/issues/805)
1822 if 'http' in proxies and 'https' not in proxies:
1823 proxies['https'] = proxies['http']
1824 proxy_handler = PerRequestProxyHandler(proxies)
1826 debuglevel = 1 if self.params.get('debug_printtraffic') else 0
1827 https_handler = make_HTTPS_handler(self.params, debuglevel=debuglevel)
1828 ydlh = YoutubeDLHandler(self.params, debuglevel=debuglevel)
1829 opener = compat_urllib_request.build_opener(
1830 proxy_handler, https_handler, cookie_processor, ydlh)
1832 # Delete the default user-agent header, which would otherwise apply in
1833 # cases where our custom HTTP handler doesn't come into play
1834 # (See https://github.com/rg3/youtube-dl/issues/1309 for details)
1835 opener.addheaders = []
1836 self._opener = opener
1838 def encode(self, s):
1839 if isinstance(s, bytes):
1840 return s # Already encoded
1843 return s.encode(self.get_encoding())
1844 except UnicodeEncodeError as err:
1845 err.reason = err.reason + '. Check your system encoding configuration or use the --encoding option.'
1848 def get_encoding(self):
1849 encoding = self.params.get('encoding')
1850 if encoding is None:
1851 encoding = preferredencoding()
1854 def _write_thumbnails(self, info_dict, filename):
1855 if self.params.get('writethumbnail', False):
1856 thumbnails = info_dict.get('thumbnails')
1858 thumbnails = [thumbnails[-1]]
1859 elif self.params.get('write_all_thumbnails', False):
1860 thumbnails = info_dict.get('thumbnails')
1865 # No thumbnails present, so return immediately
1868 for t in thumbnails:
1869 thumb_ext = determine_ext(t['url'], 'jpg')
1870 suffix = '_%s' % t['id'] if len(thumbnails) > 1 else ''
1871 thumb_display_id = '%s ' % t['id'] if len(thumbnails) > 1 else ''
1872 t['filename'] = thumb_filename = os.path.splitext(filename)[0] + suffix + '.' + thumb_ext
1874 if self.params.get('nooverwrites', False) and os.path.exists(encodeFilename(thumb_filename)):
1875 self.to_screen('[%s] %s: Thumbnail %sis already present' %
1876 (info_dict['extractor'], info_dict['id'], thumb_display_id))
1878 self.to_screen('[%s] %s: Downloading thumbnail %s...' %
1879 (info_dict['extractor'], info_dict['id'], thumb_display_id))
1881 uf = self.urlopen(t['url'])
1882 with open(thumb_filename, 'wb') as thumbf:
1883 shutil.copyfileobj(uf, thumbf)
1884 self.to_screen('[%s] %s: Writing thumbnail %sto: %s' %
1885 (info_dict['extractor'], info_dict['id'], thumb_display_id, thumb_filename))
1886 except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
1887 self.report_warning('Unable to download thumbnail "%s": %s' %
1888 (t['url'], compat_str(err)))