2 # -*- coding: utf-8 -*-
4 from __future__ import absolute_import, unicode_literals
33 compat_get_terminal_size,
38 compat_urllib_request,
57 PerRequestProxyHandler,
67 UnavailableVideoError,
77 from .cache import Cache
78 from .extractor import get_info_extractor, gen_extractors
79 from .downloader import get_suitable_downloader
80 from .downloader.rtmp import rtmpdump_version
81 from .postprocessor import (
83 FFmpegFixupStretchedPP,
88 from .version import __version__
91 class YoutubeDL(object):
94 YoutubeDL objects are the ones responsible of downloading the
95 actual video file and writing it to disk if the user has requested
96 it, among some other tasks. In most cases there should be one per
97 program. As, given a video URL, the downloader doesn't know how to
98 extract all the needed information, task that InfoExtractors do, it
99 has to pass the URL to one of them.
101 For this, YoutubeDL objects have a method that allows
102 InfoExtractors to be registered in a given order. When it is passed
103 a URL, the YoutubeDL object handles it to the first InfoExtractor it
104 finds that reports being able to handle it. The InfoExtractor extracts
105 all the information about the video or videos the URL refers to, and
106 YoutubeDL process the extracted information, possibly using a File
107 Downloader to download the video.
109 YoutubeDL objects accept a lot of parameters. In order not to saturate
110 the object constructor with arguments, it receives a dictionary of
111 options instead. These options are available through the params
112 attribute for the InfoExtractors to use. The YoutubeDL also
113 registers itself as the downloader in charge for the InfoExtractors
114 that are added to it, so this is a "mutual registration".
118 username: Username for authentication purposes.
119 password: Password for authentication purposes.
120 videopassword: Password for acces a video.
121 usenetrc: Use netrc for authentication instead.
122 verbose: Print additional info to stdout.
123 quiet: Do not print messages to stdout.
124 no_warnings: Do not print out anything for warnings.
125 forceurl: Force printing final URL.
126 forcetitle: Force printing title.
127 forceid: Force printing ID.
128 forcethumbnail: Force printing thumbnail URL.
129 forcedescription: Force printing description.
130 forcefilename: Force printing final filename.
131 forceduration: Force printing duration.
132 forcejson: Force printing info_dict as JSON.
133 dump_single_json: Force printing the info_dict of the whole playlist
134 (or video) as a single JSON line.
135 simulate: Do not download the video files.
136 format: Video format code. See options.py for more information.
137 format_limit: Highest quality format to try.
138 outtmpl: Template for output names.
139 restrictfilenames: Do not allow "&" and spaces in file names
140 ignoreerrors: Do not stop on download errors.
141 nooverwrites: Prevent overwriting files.
142 playliststart: Playlist item to start at.
143 playlistend: Playlist item to end at.
144 playlist_items: Specific indices of playlist to download.
145 playlistreverse: Download playlist items in reverse order.
146 matchtitle: Download only matching titles.
147 rejecttitle: Reject downloads for matching titles.
148 logger: Log messages to a logging.Logger instance.
149 logtostderr: Log messages to stderr instead of stdout.
150 writedescription: Write the video description to a .description file
151 writeinfojson: Write the video description to a .info.json file
152 writeannotations: Write the video annotations to a .annotations.xml file
153 writethumbnail: Write the thumbnail image to a file
154 write_all_thumbnails: Write all thumbnail formats to files
155 writesubtitles: Write the video subtitles to a file
156 writeautomaticsub: Write the automatic subtitles to a file
157 allsubtitles: Downloads all the subtitles of the video
158 (requires writesubtitles or writeautomaticsub)
159 listsubtitles: Lists all available subtitles for the video
160 subtitlesformat: The format code for subtitles
161 subtitleslangs: List of languages of the subtitles to download
162 keepvideo: Keep the video file after post-processing
163 daterange: A DateRange object, download only if the upload_date is in the range.
164 skip_download: Skip the actual download of the video file
165 cachedir: Location of the cache files in the filesystem.
166 False to disable filesystem cache.
167 noplaylist: Download single video instead of a playlist if in doubt.
168 age_limit: An integer representing the user's age in years.
169 Unsuitable videos for the given age are skipped.
170 min_views: An integer representing the minimum view count the video
171 must have in order to not be skipped.
172 Videos without view count information are always
173 downloaded. None for no limit.
174 max_views: An integer representing the maximum view count.
175 Videos that are more popular than that are not
177 Videos without view count information are always
178 downloaded. None for no limit.
179 download_archive: File name of a file where all downloads are recorded.
180 Videos already present in the file are not downloaded
182 cookiefile: File name where cookies should be read from and dumped to.
183 nocheckcertificate:Do not verify SSL certificates
184 prefer_insecure: Use HTTP instead of HTTPS to retrieve information.
185 At the moment, this is only supported by YouTube.
186 proxy: URL of the proxy server to use
187 cn_verification_proxy: URL of the proxy to use for IP address verification
188 on Chinese sites. (Experimental)
189 socket_timeout: Time to wait for unresponsive hosts, in seconds
190 bidi_workaround: Work around buggy terminals without bidirectional text
191 support, using fridibi
192 debug_printtraffic:Print out sent and received HTTP traffic
193 include_ads: Download ads as well
194 default_search: Prepend this string if an input url is not valid.
195 'auto' for elaborate guessing
196 encoding: Use this encoding instead of the system-specified.
197 extract_flat: Do not resolve URLs, return the immediate result.
198 Pass in 'in_playlist' to only show this behavior for
200 postprocessors: A list of dictionaries, each with an entry
201 * key: The name of the postprocessor. See
202 youtube_dl/postprocessor/__init__.py for a list.
203 as well as any further keyword arguments for the
205 progress_hooks: A list of functions that get called on download
206 progress, with a dictionary with the entries
207 * status: One of "downloading", "error", or "finished".
208 Check this first and ignore unknown values.
210 If status is one of "downloading", or "finished", the
211 following properties may also be present:
212 * filename: The final filename (always present)
213 * tmpfilename: The filename we're currently writing to
214 * downloaded_bytes: Bytes on disk
215 * total_bytes: Size of the whole file, None if unknown
216 * total_bytes_estimate: Guess of the eventual file size,
218 * elapsed: The number of seconds since download started.
219 * eta: The estimated time in seconds, None if unknown
220 * speed: The download speed in bytes/second, None if
222 * fragment_index: The counter of the currently
223 downloaded video fragment.
224 * fragment_count: The number of fragments (= individual
225 files that will be merged)
227 Progress hooks are guaranteed to be called at least once
228 (with status "finished") if the download is successful.
229 merge_output_format: Extension to use when merging formats.
230 fixup: Automatically correct known faults of the file.
232 - "never": do nothing
233 - "warn": only emit a warning
234 - "detect_or_warn": check whether we can do anything
235 about it, warn otherwise (default)
236 source_address: (Experimental) Client-side IP address to bind to.
237 call_home: Boolean, true iff we are allowed to contact the
238 youtube-dl servers for debugging.
239 sleep_interval: Number of seconds to sleep before each download.
240 listformats: Print an overview of available video formats and exit.
241 list_thumbnails: Print a table of all thumbnails and exit.
242 match_filter: A function that gets called with the info_dict of
244 If it returns a message, the video is ignored.
245 If it returns None, the video is downloaded.
246 match_filter_func in utils.py is one example for this.
247 no_color: Do not emit color codes in output.
249 The following options determine which downloader is picked:
250 external_downloader: Executable of the external downloader to call.
251 None or unset for standard (built-in) downloader.
252 hls_prefer_native: Use the native HLS downloader instead of ffmpeg/avconv.
254 The following parameters are not used by YoutubeDL itself, they are used by
255 the downloader (see youtube_dl/downloader/common.py):
256 nopart, updatetime, buffersize, ratelimit, min_filesize, max_filesize, test,
257 noresizebuffer, retries, continuedl, noprogress, consoletitle,
258 xattr_set_filesize, external_downloader_args.
260 The following options are used by the post processors:
261 prefer_ffmpeg: If True, use ffmpeg instead of avconv if both are available,
262 otherwise prefer avconv.
263 exec_cmd: Arbitrary command to run after downloading
269 _download_retcode = None
270 _num_downloads = None
273 def __init__(self, params=None, auto_init=True):
274 """Create a FileDownloader object with the given options."""
278 self._ies_instances = {}
280 self._progress_hooks = []
281 self._download_retcode = 0
282 self._num_downloads = 0
283 self._screen_file = [sys.stdout, sys.stderr][params.get('logtostderr', False)]
284 self._err_file = sys.stderr
286 self.cache = Cache(self)
288 if params.get('bidi_workaround', False):
291 master, slave = pty.openpty()
292 width = compat_get_terminal_size().columns
296 width_args = ['-w', str(width)]
298 stdin=subprocess.PIPE,
300 stderr=self._err_file)
302 self._output_process = subprocess.Popen(
303 ['bidiv'] + width_args, **sp_kwargs
306 self._output_process = subprocess.Popen(
307 ['fribidi', '-c', 'UTF-8'] + width_args, **sp_kwargs)
308 self._output_channel = os.fdopen(master, 'rb')
309 except OSError as ose:
311 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 $PATH.')
315 if (sys.version_info >= (3,) and sys.platform != 'win32' and
316 sys.getfilesystemencoding() in ['ascii', 'ANSI_X3.4-1968'] and
317 not params.get('restrictfilenames', False)):
318 # On Python 3, the Unicode filesystem API will throw errors (#1474)
320 'Assuming --restrict-filenames since file system encoding '
321 'cannot encode all characters. '
322 'Set the LC_ALL environment variable to fix this.')
323 self.params['restrictfilenames'] = True
325 if '%(stitle)s' in self.params.get('outtmpl', ''):
326 self.report_warning('%(stitle)s is deprecated. Use the %(title)s and the --restrict-filenames flag(which also secures %(uploader)s et al) instead.')
331 self.print_debug_header()
332 self.add_default_info_extractors()
334 for pp_def_raw in self.params.get('postprocessors', []):
335 pp_class = get_postprocessor(pp_def_raw['key'])
336 pp_def = dict(pp_def_raw)
338 pp = pp_class(self, **compat_kwargs(pp_def))
339 self.add_post_processor(pp)
341 for ph in self.params.get('progress_hooks', []):
342 self.add_progress_hook(ph)
344 def warn_if_short_id(self, argv):
345 # short YouTube ID starting with dash?
347 i for i, a in enumerate(argv)
348 if re.match(r'^-[0-9A-Za-z_-]{10}$', a)]
352 [a for i, a in enumerate(argv) if i not in idxs] +
353 ['--'] + [argv[i] for i in idxs]
356 'Long argument string detected. '
357 'Use -- to separate parameters and URLs, like this:\n%s\n' %
358 args_to_str(correct_argv))
360 def add_info_extractor(self, ie):
361 """Add an InfoExtractor object to the end of the list."""
363 self._ies_instances[ie.ie_key()] = ie
364 ie.set_downloader(self)
366 def get_info_extractor(self, ie_key):
368 Get an instance of an IE with name ie_key, it will try to get one from
369 the _ies list, if there's no instance it will create a new one and add
370 it to the extractor list.
372 ie = self._ies_instances.get(ie_key)
374 ie = get_info_extractor(ie_key)()
375 self.add_info_extractor(ie)
378 def add_default_info_extractors(self):
380 Add the InfoExtractors returned by gen_extractors to the end of the list
382 for ie in gen_extractors():
383 self.add_info_extractor(ie)
385 def add_post_processor(self, pp):
386 """Add a PostProcessor object to the end of the chain."""
388 pp.set_downloader(self)
390 def add_progress_hook(self, ph):
391 """Add the progress hook (currently only for the file downloader)"""
392 self._progress_hooks.append(ph)
394 def _bidi_workaround(self, message):
395 if not hasattr(self, '_output_channel'):
398 assert hasattr(self, '_output_process')
399 assert isinstance(message, compat_str)
400 line_count = message.count('\n') + 1
401 self._output_process.stdin.write((message + '\n').encode('utf-8'))
402 self._output_process.stdin.flush()
403 res = ''.join(self._output_channel.readline().decode('utf-8')
404 for _ in range(line_count))
405 return res[:-len('\n')]
407 def to_screen(self, message, skip_eol=False):
408 """Print message to stdout if not in quiet mode."""
409 return self.to_stdout(message, skip_eol, check_quiet=True)
411 def _write_string(self, s, out=None):
412 write_string(s, out=out, encoding=self.params.get('encoding'))
414 def to_stdout(self, message, skip_eol=False, check_quiet=False):
415 """Print message to stdout if not in quiet mode."""
416 if self.params.get('logger'):
417 self.params['logger'].debug(message)
418 elif not check_quiet or not self.params.get('quiet', False):
419 message = self._bidi_workaround(message)
420 terminator = ['\n', ''][skip_eol]
421 output = message + terminator
423 self._write_string(output, self._screen_file)
425 def to_stderr(self, message):
426 """Print message to stderr."""
427 assert isinstance(message, compat_str)
428 if self.params.get('logger'):
429 self.params['logger'].error(message)
431 message = self._bidi_workaround(message)
432 output = message + '\n'
433 self._write_string(output, self._err_file)
435 def to_console_title(self, message):
436 if not self.params.get('consoletitle', False):
438 if os.name == 'nt' and ctypes.windll.kernel32.GetConsoleWindow():
439 # c_wchar_p() might not be necessary if `message` is
440 # already of type unicode()
441 ctypes.windll.kernel32.SetConsoleTitleW(ctypes.c_wchar_p(message))
442 elif 'TERM' in os.environ:
443 self._write_string('\033]0;%s\007' % message, self._screen_file)
445 def save_console_title(self):
446 if not self.params.get('consoletitle', False):
448 if 'TERM' in os.environ:
449 # Save the title on stack
450 self._write_string('\033[22;0t', self._screen_file)
452 def restore_console_title(self):
453 if not self.params.get('consoletitle', False):
455 if 'TERM' in os.environ:
456 # Restore the title from stack
457 self._write_string('\033[23;0t', self._screen_file)
460 self.save_console_title()
463 def __exit__(self, *args):
464 self.restore_console_title()
466 if self.params.get('cookiefile') is not None:
467 self.cookiejar.save()
469 def trouble(self, message=None, tb=None):
470 """Determine action to take when a download problem appears.
472 Depending on if the downloader has been configured to ignore
473 download errors or not, this method may throw an exception or
474 not when errors are found, after printing the message.
476 tb, if given, is additional traceback information.
478 if message is not None:
479 self.to_stderr(message)
480 if self.params.get('verbose'):
482 if sys.exc_info()[0]: # if .trouble has been called from an except block
484 if hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]:
485 tb += ''.join(traceback.format_exception(*sys.exc_info()[1].exc_info))
486 tb += compat_str(traceback.format_exc())
488 tb_data = traceback.format_list(traceback.extract_stack())
489 tb = ''.join(tb_data)
491 if not self.params.get('ignoreerrors', False):
492 if sys.exc_info()[0] and hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]:
493 exc_info = sys.exc_info()[1].exc_info
495 exc_info = sys.exc_info()
496 raise DownloadError(message, exc_info)
497 self._download_retcode = 1
499 def report_warning(self, message):
501 Print the message to stderr, it will be prefixed with 'WARNING:'
502 If stderr is a tty file the 'WARNING:' will be colored
504 if self.params.get('logger') is not None:
505 self.params['logger'].warning(message)
507 if self.params.get('no_warnings'):
509 if not self.params.get('no_color') and self._err_file.isatty() and os.name != 'nt':
510 _msg_header = '\033[0;33mWARNING:\033[0m'
512 _msg_header = 'WARNING:'
513 warning_message = '%s %s' % (_msg_header, message)
514 self.to_stderr(warning_message)
516 def report_error(self, message, tb=None):
518 Do the same as trouble, but prefixes the message with 'ERROR:', colored
519 in red if stderr is a tty file.
521 if not self.params.get('no_color') and self._err_file.isatty() and os.name != 'nt':
522 _msg_header = '\033[0;31mERROR:\033[0m'
524 _msg_header = 'ERROR:'
525 error_message = '%s %s' % (_msg_header, message)
526 self.trouble(error_message, tb)
528 def report_file_already_downloaded(self, file_name):
529 """Report file has already been fully downloaded."""
531 self.to_screen('[download] %s has already been downloaded' % file_name)
532 except UnicodeEncodeError:
533 self.to_screen('[download] The file has already been downloaded')
535 def prepare_filename(self, info_dict):
536 """Generate the output filename."""
538 template_dict = dict(info_dict)
540 template_dict['epoch'] = int(time.time())
541 autonumber_size = self.params.get('autonumber_size')
542 if autonumber_size is None:
544 autonumber_templ = '%0' + str(autonumber_size) + 'd'
545 template_dict['autonumber'] = autonumber_templ % self._num_downloads
546 if template_dict.get('playlist_index') is not None:
547 template_dict['playlist_index'] = '%0*d' % (len(str(template_dict['n_entries'])), template_dict['playlist_index'])
548 if template_dict.get('resolution') is None:
549 if template_dict.get('width') and template_dict.get('height'):
550 template_dict['resolution'] = '%dx%d' % (template_dict['width'], template_dict['height'])
551 elif template_dict.get('height'):
552 template_dict['resolution'] = '%sp' % template_dict['height']
553 elif template_dict.get('width'):
554 template_dict['resolution'] = '?x%d' % template_dict['width']
556 sanitize = lambda k, v: sanitize_filename(
558 restricted=self.params.get('restrictfilenames'),
560 template_dict = dict((k, sanitize(k, v))
561 for k, v in template_dict.items()
563 template_dict = collections.defaultdict(lambda: 'NA', template_dict)
565 outtmpl = self.params.get('outtmpl', DEFAULT_OUTTMPL)
566 tmpl = compat_expanduser(outtmpl)
567 filename = tmpl % template_dict
568 # Temporary fix for #4787
569 # 'Treat' all problem characters by passing filename through preferredencoding
570 # to workaround encoding issues with subprocess on python2 @ Windows
571 if sys.version_info < (3, 0) and sys.platform == 'win32':
572 filename = encodeFilename(filename, True).decode(preferredencoding())
574 except ValueError as err:
575 self.report_error('Error in output template: ' + str(err) + ' (encoding: ' + repr(preferredencoding()) + ')')
578 def _match_entry(self, info_dict, incomplete):
579 """ Returns None iff the file should be downloaded """
581 video_title = info_dict.get('title', info_dict.get('id', 'video'))
582 if 'title' in info_dict:
583 # This can happen when we're just evaluating the playlist
584 title = info_dict['title']
585 matchtitle = self.params.get('matchtitle', False)
587 if not re.search(matchtitle, title, re.IGNORECASE):
588 return '"' + title + '" title did not match pattern "' + matchtitle + '"'
589 rejecttitle = self.params.get('rejecttitle', False)
591 if re.search(rejecttitle, title, re.IGNORECASE):
592 return '"' + title + '" title matched reject pattern "' + rejecttitle + '"'
593 date = info_dict.get('upload_date', None)
595 dateRange = self.params.get('daterange', DateRange())
596 if date not in dateRange:
597 return '%s upload date is not in range %s' % (date_from_str(date).isoformat(), dateRange)
598 view_count = info_dict.get('view_count', None)
599 if view_count is not None:
600 min_views = self.params.get('min_views')
601 if min_views is not None and view_count < min_views:
602 return 'Skipping %s, because it has not reached minimum view count (%d/%d)' % (video_title, view_count, min_views)
603 max_views = self.params.get('max_views')
604 if max_views is not None and view_count > max_views:
605 return 'Skipping %s, because it has exceeded the maximum view count (%d/%d)' % (video_title, view_count, max_views)
606 if age_restricted(info_dict.get('age_limit'), self.params.get('age_limit')):
607 return 'Skipping "%s" because it is age restricted' % video_title
608 if self.in_download_archive(info_dict):
609 return '%s has already been recorded in archive' % video_title
612 match_filter = self.params.get('match_filter')
613 if match_filter is not None:
614 ret = match_filter(info_dict)
621 def add_extra_info(info_dict, extra_info):
622 '''Set the keys from extra_info in info dict if they are missing'''
623 for key, value in extra_info.items():
624 info_dict.setdefault(key, value)
626 def extract_info(self, url, download=True, ie_key=None, extra_info={},
629 Returns a list with a dictionary for each video we find.
630 If 'download', also downloads the videos.
631 extra_info is a dict containing the extra values to add to each result
635 ies = [self.get_info_extractor(ie_key)]
640 if not ie.suitable(url):
644 self.report_warning('The program functionality for this site has been marked as broken, '
645 'and will probably not work.')
648 ie_result = ie.extract(url)
649 if ie_result is None: # Finished already (backwards compatibility; listformats and friends should be moved here)
651 if isinstance(ie_result, list):
652 # Backwards compatibility: old IE result format
654 '_type': 'compat_list',
655 'entries': ie_result,
657 self.add_default_extra_info(ie_result, ie, url)
659 return self.process_ie_result(ie_result, download, extra_info)
662 except ExtractorError as de: # An error we somewhat expected
663 self.report_error(compat_str(de), de.format_traceback())
665 except MaxDownloadsReached:
667 except Exception as e:
668 if self.params.get('ignoreerrors', False):
669 self.report_error(compat_str(e), tb=compat_str(traceback.format_exc()))
674 self.report_error('no suitable InfoExtractor for URL %s' % url)
676 def add_default_extra_info(self, ie_result, ie, url):
677 self.add_extra_info(ie_result, {
678 'extractor': ie.IE_NAME,
680 'webpage_url_basename': url_basename(url),
681 'extractor_key': ie.ie_key(),
684 def process_ie_result(self, ie_result, download=True, extra_info={}):
686 Take the result of the ie(may be modified) and resolve all unresolved
687 references (URLs, playlist items).
689 It will also download the videos if 'download'.
690 Returns the resolved ie_result.
693 result_type = ie_result.get('_type', 'video')
695 if result_type in ('url', 'url_transparent'):
696 extract_flat = self.params.get('extract_flat', False)
697 if ((extract_flat == 'in_playlist' and 'playlist' in extra_info) or
698 extract_flat is True):
699 if self.params.get('forcejson', False):
700 self.to_stdout(json.dumps(ie_result))
703 if result_type == 'video':
704 self.add_extra_info(ie_result, extra_info)
705 return self.process_video_result(ie_result, download=download)
706 elif result_type == 'url':
707 # We have to add extra_info to the results because it may be
708 # contained in a playlist
709 return self.extract_info(ie_result['url'],
711 ie_key=ie_result.get('ie_key'),
712 extra_info=extra_info)
713 elif result_type == 'url_transparent':
714 # Use the information from the embedding page
715 info = self.extract_info(
716 ie_result['url'], ie_key=ie_result.get('ie_key'),
717 extra_info=extra_info, download=False, process=False)
719 force_properties = dict(
720 (k, v) for k, v in ie_result.items() if v is not None)
721 for f in ('_type', 'url'):
722 if f in force_properties:
723 del force_properties[f]
724 new_result = info.copy()
725 new_result.update(force_properties)
727 assert new_result.get('_type') != 'url_transparent'
729 return self.process_ie_result(
730 new_result, download=download, extra_info=extra_info)
731 elif result_type == 'playlist' or result_type == 'multi_video':
732 # We process each entry in the playlist
733 playlist = ie_result.get('title', None) or ie_result.get('id', None)
734 self.to_screen('[download] Downloading playlist: %s' % playlist)
736 playlist_results = []
738 playliststart = self.params.get('playliststart', 1) - 1
739 playlistend = self.params.get('playlistend', None)
740 # For backwards compatibility, interpret -1 as whole list
741 if playlistend == -1:
744 playlistitems_str = self.params.get('playlist_items', None)
746 if playlistitems_str is not None:
747 def iter_playlistitems(format):
748 for string_segment in format.split(','):
749 if '-' in string_segment:
750 start, end = string_segment.split('-')
751 for item in range(int(start), int(end) + 1):
754 yield int(string_segment)
755 playlistitems = iter_playlistitems(playlistitems_str)
757 ie_entries = ie_result['entries']
758 if isinstance(ie_entries, list):
759 n_all_entries = len(ie_entries)
761 entries = [ie_entries[i - 1] for i in playlistitems]
763 entries = ie_entries[playliststart:playlistend]
764 n_entries = len(entries)
766 "[%s] playlist %s: Collected %d video ids (downloading %d of them)" %
767 (ie_result['extractor'], playlist, n_all_entries, n_entries))
768 elif isinstance(ie_entries, PagedList):
771 for item in playlistitems:
772 entries.extend(ie_entries.getslice(
776 entries = ie_entries.getslice(
777 playliststart, playlistend)
778 n_entries = len(entries)
780 "[%s] playlist %s: Downloading %d videos" %
781 (ie_result['extractor'], playlist, n_entries))
784 entry_list = list(ie_entries)
785 entries = [entry_list[i - 1] for i in playlistitems]
787 entries = list(itertools.islice(
788 ie_entries, playliststart, playlistend))
789 n_entries = len(entries)
791 "[%s] playlist %s: Downloading %d videos" %
792 (ie_result['extractor'], playlist, n_entries))
794 if self.params.get('playlistreverse', False):
795 entries = entries[::-1]
797 for i, entry in enumerate(entries, 1):
798 self.to_screen('[download] Downloading video %s of %s' % (i, n_entries))
800 'n_entries': n_entries,
801 'playlist': playlist,
802 'playlist_id': ie_result.get('id'),
803 'playlist_title': ie_result.get('title'),
804 'playlist_index': i + playliststart,
805 'extractor': ie_result['extractor'],
806 'webpage_url': ie_result['webpage_url'],
807 'webpage_url_basename': url_basename(ie_result['webpage_url']),
808 'extractor_key': ie_result['extractor_key'],
811 reason = self._match_entry(entry, incomplete=True)
812 if reason is not None:
813 self.to_screen('[download] ' + reason)
816 entry_result = self.process_ie_result(entry,
819 playlist_results.append(entry_result)
820 ie_result['entries'] = playlist_results
822 elif result_type == 'compat_list':
824 'Extractor %s returned a compat_list result. '
825 'It needs to be updated.' % ie_result.get('extractor'))
831 'extractor': ie_result['extractor'],
832 'webpage_url': ie_result['webpage_url'],
833 'webpage_url_basename': url_basename(ie_result['webpage_url']),
834 'extractor_key': ie_result['extractor_key'],
838 ie_result['entries'] = [
839 self.process_ie_result(_fixup(r), download, extra_info)
840 for r in ie_result['entries']
844 raise Exception('Invalid result type: %s' % result_type)
846 def _apply_format_filter(self, format_spec, available_formats):
847 " Returns a tuple of the remaining format_spec and filtered formats "
857 operator_rex = re.compile(r'''(?x)\s*\[
858 (?P<key>width|height|tbr|abr|vbr|asr|filesize|fps)
859 \s*(?P<op>%s)(?P<none_inclusive>\s*\?)?\s*
860 (?P<value>[0-9.]+(?:[kKmMgGtTpPeEzZyY]i?[Bb]?)?)
862 ''' % '|'.join(map(re.escape, OPERATORS.keys())))
863 m = operator_rex.search(format_spec)
866 comparison_value = int(m.group('value'))
868 comparison_value = parse_filesize(m.group('value'))
869 if comparison_value is None:
870 comparison_value = parse_filesize(m.group('value') + 'B')
871 if comparison_value is None:
873 'Invalid value %r in format specification %r' % (
874 m.group('value'), format_spec))
875 op = OPERATORS[m.group('op')]
882 str_operator_rex = re.compile(r'''(?x)\s*\[
883 \s*(?P<key>ext|acodec|vcodec|container|protocol)
884 \s*(?P<op>%s)(?P<none_inclusive>\s*\?)?
885 \s*(?P<value>[a-zA-Z0-9_-]+)
887 ''' % '|'.join(map(re.escape, STR_OPERATORS.keys())))
888 m = str_operator_rex.search(format_spec)
890 comparison_value = m.group('value')
891 op = STR_OPERATORS[m.group('op')]
894 raise ValueError('Invalid format specification %r' % format_spec)
897 actual_value = f.get(m.group('key'))
898 if actual_value is None:
899 return m.group('none_inclusive')
900 return op(actual_value, comparison_value)
901 new_formats = [f for f in available_formats if _filter(f)]
903 new_format_spec = format_spec[:-len(m.group(0))]
904 if not new_format_spec:
905 new_format_spec = 'best'
907 return (new_format_spec, new_formats)
909 def select_format(self, format_spec, available_formats):
910 while format_spec.endswith(']'):
911 format_spec, available_formats = self._apply_format_filter(
912 format_spec, available_formats)
913 if not available_formats:
916 if format_spec == 'best' or format_spec is None:
917 return available_formats[-1]
918 elif format_spec == 'worst':
919 return available_formats[0]
920 elif format_spec == 'bestaudio':
922 f for f in available_formats
923 if f.get('vcodec') == 'none']
925 return audio_formats[-1]
926 elif format_spec == 'worstaudio':
928 f for f in available_formats
929 if f.get('vcodec') == 'none']
931 return audio_formats[0]
932 elif format_spec == 'bestvideo':
934 f for f in available_formats
935 if f.get('acodec') == 'none']
937 return video_formats[-1]
938 elif format_spec == 'worstvideo':
940 f for f in available_formats
941 if f.get('acodec') == 'none']
943 return video_formats[0]
945 extensions = ['mp4', 'flv', 'webm', '3gp', 'm4a', 'mp3', 'ogg', 'aac', 'wav']
946 if format_spec in extensions:
947 filter_f = lambda f: f['ext'] == format_spec
949 filter_f = lambda f: f['format_id'] == format_spec
950 matches = list(filter(filter_f, available_formats))
955 def _calc_headers(self, info_dict):
956 res = std_headers.copy()
958 add_headers = info_dict.get('http_headers')
960 res.update(add_headers)
962 cookies = self._calc_cookies(info_dict)
964 res['Cookie'] = cookies
968 def _calc_cookies(self, info_dict):
969 pr = compat_urllib_request.Request(info_dict['url'])
970 self.cookiejar.add_cookie_header(pr)
971 return pr.get_header('Cookie')
973 def process_video_result(self, info_dict, download=True):
974 assert info_dict.get('_type', 'video') == 'video'
976 if 'id' not in info_dict:
977 raise ExtractorError('Missing "id" field in extractor result')
978 if 'title' not in info_dict:
979 raise ExtractorError('Missing "title" field in extractor result')
981 if 'playlist' not in info_dict:
982 # It isn't part of a playlist
983 info_dict['playlist'] = None
984 info_dict['playlist_index'] = None
986 thumbnails = info_dict.get('thumbnails')
987 if thumbnails is None:
988 thumbnail = info_dict.get('thumbnail')
990 info_dict['thumbnails'] = thumbnails = [{'url': thumbnail}]
992 thumbnails.sort(key=lambda t: (
993 t.get('preference'), t.get('width'), t.get('height'),
994 t.get('id'), t.get('url')))
995 for i, t in enumerate(thumbnails):
996 if 'width' in t and 'height' in t:
997 t['resolution'] = '%dx%d' % (t['width'], t['height'])
998 if t.get('id') is None:
1001 if thumbnails and 'thumbnail' not in info_dict:
1002 info_dict['thumbnail'] = thumbnails[-1]['url']
1004 if 'display_id' not in info_dict and 'id' in info_dict:
1005 info_dict['display_id'] = info_dict['id']
1007 if info_dict.get('upload_date') is None and info_dict.get('timestamp') is not None:
1008 # Working around negative timestamps in Windows
1009 # (see http://bugs.python.org/issue1646728)
1010 if info_dict['timestamp'] < 0 and os.name == 'nt':
1011 info_dict['timestamp'] = 0
1012 upload_date = datetime.datetime.utcfromtimestamp(
1013 info_dict['timestamp'])
1014 info_dict['upload_date'] = upload_date.strftime('%Y%m%d')
1016 if self.params.get('listsubtitles', False):
1017 if 'automatic_captions' in info_dict:
1018 self.list_subtitles(info_dict['id'], info_dict.get('automatic_captions'), 'automatic captions')
1019 self.list_subtitles(info_dict['id'], info_dict.get('subtitles'), 'subtitles')
1021 info_dict['requested_subtitles'] = self.process_subtitles(
1022 info_dict['id'], info_dict.get('subtitles'),
1023 info_dict.get('automatic_captions'))
1025 # This extractors handle format selection themselves
1026 if info_dict['extractor'] in ['Youku']:
1028 self.process_info(info_dict)
1031 # We now pick which formats have to be downloaded
1032 if info_dict.get('formats') is None:
1033 # There's only one format available
1034 formats = [info_dict]
1036 formats = info_dict['formats']
1039 raise ExtractorError('No video formats found!')
1041 # We check that all the formats have the format and format_id fields
1042 for i, format in enumerate(formats):
1043 if 'url' not in format:
1044 raise ExtractorError('Missing "url" key in result (index %d)' % i)
1046 if format.get('format_id') is None:
1047 format['format_id'] = compat_str(i)
1048 if format.get('format') is None:
1049 format['format'] = '{id} - {res}{note}'.format(
1050 id=format['format_id'],
1051 res=self.format_resolution(format),
1052 note=' ({0})'.format(format['format_note']) if format.get('format_note') is not None else '',
1054 # Automatically determine file extension if missing
1055 if 'ext' not in format:
1056 format['ext'] = determine_ext(format['url']).lower()
1057 # Add HTTP headers, so that external programs can use them from the
1059 full_format_info = info_dict.copy()
1060 full_format_info.update(format)
1061 format['http_headers'] = self._calc_headers(full_format_info)
1063 format_limit = self.params.get('format_limit', None)
1065 formats = list(takewhile_inclusive(
1066 lambda f: f['format_id'] != format_limit, formats
1069 # TODO Central sorting goes here
1071 if formats[0] is not info_dict:
1072 # only set the 'formats' fields if the original info_dict list them
1073 # otherwise we end up with a circular reference, the first (and unique)
1074 # element in the 'formats' field in info_dict is info_dict itself,
1075 # wich can't be exported to json
1076 info_dict['formats'] = formats
1077 if self.params.get('listformats'):
1078 self.list_formats(info_dict)
1080 if self.params.get('list_thumbnails'):
1081 self.list_thumbnails(info_dict)
1084 req_format = self.params.get('format')
1085 if req_format is None:
1087 formats_to_download = []
1088 # The -1 is for supporting YoutubeIE
1089 if req_format in ('-1', 'all'):
1090 formats_to_download = formats
1092 for rfstr in req_format.split(','):
1093 # We can accept formats requested in the format: 34/5/best, we pick
1094 # the first that is available, starting from left
1095 req_formats = rfstr.split('/')
1096 for rf in req_formats:
1097 if re.match(r'.+?\+.+?', rf) is not None:
1098 # Two formats have been requested like '137+139'
1099 format_1, format_2 = rf.split('+')
1100 formats_info = (self.select_format(format_1, formats),
1101 self.select_format(format_2, formats))
1102 if all(formats_info):
1103 # The first format must contain the video and the
1105 if formats_info[0].get('vcodec') == 'none':
1106 self.report_error('The first format must '
1107 'contain the video, try using '
1108 '"-f %s+%s"' % (format_2, format_1))
1111 formats_info[0]['ext']
1112 if self.params.get('merge_output_format') is None
1113 else self.params['merge_output_format'])
1115 'requested_formats': formats_info,
1116 'format': '%s+%s' % (formats_info[0].get('format'),
1117 formats_info[1].get('format')),
1118 'format_id': '%s+%s' % (formats_info[0].get('format_id'),
1119 formats_info[1].get('format_id')),
1120 'width': formats_info[0].get('width'),
1121 'height': formats_info[0].get('height'),
1122 'resolution': formats_info[0].get('resolution'),
1123 'fps': formats_info[0].get('fps'),
1124 'vcodec': formats_info[0].get('vcodec'),
1125 'vbr': formats_info[0].get('vbr'),
1126 'stretched_ratio': formats_info[0].get('stretched_ratio'),
1127 'acodec': formats_info[1].get('acodec'),
1128 'abr': formats_info[1].get('abr'),
1132 selected_format = None
1134 selected_format = self.select_format(rf, formats)
1135 if selected_format is not None:
1136 formats_to_download.append(selected_format)
1138 if not formats_to_download:
1139 raise ExtractorError('requested format not available',
1143 if len(formats_to_download) > 1:
1144 self.to_screen('[info] %s: downloading video in %s formats' % (info_dict['id'], len(formats_to_download)))
1145 for format in formats_to_download:
1146 new_info = dict(info_dict)
1147 new_info.update(format)
1148 self.process_info(new_info)
1149 # We update the info dict with the best quality format (backwards compatibility)
1150 info_dict.update(formats_to_download[-1])
1153 def process_subtitles(self, video_id, normal_subtitles, automatic_captions):
1154 """Select the requested subtitles and their format"""
1156 if normal_subtitles and self.params.get('writesubtitles'):
1157 available_subs.update(normal_subtitles)
1158 if automatic_captions and self.params.get('writeautomaticsub'):
1159 for lang, cap_info in automatic_captions.items():
1160 if lang not in available_subs:
1161 available_subs[lang] = cap_info
1163 if (not self.params.get('writesubtitles') and not
1164 self.params.get('writeautomaticsub') or not
1168 if self.params.get('allsubtitles', False):
1169 requested_langs = available_subs.keys()
1171 if self.params.get('subtitleslangs', False):
1172 requested_langs = self.params.get('subtitleslangs')
1173 elif 'en' in available_subs:
1174 requested_langs = ['en']
1176 requested_langs = [list(available_subs.keys())[0]]
1178 formats_query = self.params.get('subtitlesformat', 'best')
1179 formats_preference = formats_query.split('/') if formats_query else []
1181 for lang in requested_langs:
1182 formats = available_subs.get(lang)
1184 self.report_warning('%s subtitles not available for %s' % (lang, video_id))
1186 for ext in formats_preference:
1190 matches = list(filter(lambda f: f['ext'] == ext, formats))
1196 self.report_warning(
1197 'No subtitle format found matching "%s" for language %s, '
1198 'using %s' % (formats_query, lang, f['ext']))
1202 def process_info(self, info_dict):
1203 """Process a single resolved IE result."""
1205 assert info_dict.get('_type', 'video') == 'video'
1207 max_downloads = self.params.get('max_downloads')
1208 if max_downloads is not None:
1209 if self._num_downloads >= int(max_downloads):
1210 raise MaxDownloadsReached()
1212 info_dict['fulltitle'] = info_dict['title']
1213 if len(info_dict['title']) > 200:
1214 info_dict['title'] = info_dict['title'][:197] + '...'
1216 # Keep for backwards compatibility
1217 info_dict['stitle'] = info_dict['title']
1219 if 'format' not in info_dict:
1220 info_dict['format'] = info_dict['ext']
1222 reason = self._match_entry(info_dict, incomplete=False)
1223 if reason is not None:
1224 self.to_screen('[download] ' + reason)
1227 self._num_downloads += 1
1229 info_dict['_filename'] = filename = self.prepare_filename(info_dict)
1232 if self.params.get('forcetitle', False):
1233 self.to_stdout(info_dict['fulltitle'])
1234 if self.params.get('forceid', False):
1235 self.to_stdout(info_dict['id'])
1236 if self.params.get('forceurl', False):
1237 if info_dict.get('requested_formats') is not None:
1238 for f in info_dict['requested_formats']:
1239 self.to_stdout(f['url'] + f.get('play_path', ''))
1241 # For RTMP URLs, also include the playpath
1242 self.to_stdout(info_dict['url'] + info_dict.get('play_path', ''))
1243 if self.params.get('forcethumbnail', False) and info_dict.get('thumbnail') is not None:
1244 self.to_stdout(info_dict['thumbnail'])
1245 if self.params.get('forcedescription', False) and info_dict.get('description') is not None:
1246 self.to_stdout(info_dict['description'])
1247 if self.params.get('forcefilename', False) and filename is not None:
1248 self.to_stdout(filename)
1249 if self.params.get('forceduration', False) and info_dict.get('duration') is not None:
1250 self.to_stdout(formatSeconds(info_dict['duration']))
1251 if self.params.get('forceformat', False):
1252 self.to_stdout(info_dict['format'])
1253 if self.params.get('forcejson', False):
1254 self.to_stdout(json.dumps(info_dict))
1256 # Do nothing else if in simulate mode
1257 if self.params.get('simulate', False):
1260 if filename is None:
1264 dn = os.path.dirname(encodeFilename(filename))
1265 if dn and not os.path.exists(dn):
1267 except (OSError, IOError) as err:
1268 self.report_error('unable to create directory ' + compat_str(err))
1271 if self.params.get('writedescription', False):
1272 descfn = filename + '.description'
1273 if self.params.get('nooverwrites', False) and os.path.exists(encodeFilename(descfn)):
1274 self.to_screen('[info] Video description is already present')
1275 elif info_dict.get('description') is None:
1276 self.report_warning('There\'s no description to write.')
1279 self.to_screen('[info] Writing video description to: ' + descfn)
1280 with io.open(encodeFilename(descfn), 'w', encoding='utf-8') as descfile:
1281 descfile.write(info_dict['description'])
1282 except (OSError, IOError):
1283 self.report_error('Cannot write description file ' + descfn)
1286 if self.params.get('writeannotations', False):
1287 annofn = filename + '.annotations.xml'
1288 if self.params.get('nooverwrites', False) and os.path.exists(encodeFilename(annofn)):
1289 self.to_screen('[info] Video annotations are already present')
1292 self.to_screen('[info] Writing video annotations to: ' + annofn)
1293 with io.open(encodeFilename(annofn), 'w', encoding='utf-8') as annofile:
1294 annofile.write(info_dict['annotations'])
1295 except (KeyError, TypeError):
1296 self.report_warning('There are no annotations to write.')
1297 except (OSError, IOError):
1298 self.report_error('Cannot write annotations file: ' + annofn)
1301 subtitles_are_requested = any([self.params.get('writesubtitles', False),
1302 self.params.get('writeautomaticsub')])
1304 if subtitles_are_requested and info_dict.get('requested_subtitles'):
1305 # subtitles download errors are already managed as troubles in relevant IE
1306 # that way it will silently go on when used with unsupporting IE
1307 subtitles = info_dict['requested_subtitles']
1308 ie = self.get_info_extractor(info_dict['extractor_key'])
1309 for sub_lang, sub_info in subtitles.items():
1310 sub_format = sub_info['ext']
1311 if sub_info.get('data') is not None:
1312 sub_data = sub_info['data']
1315 sub_data = ie._download_webpage(
1316 sub_info['url'], info_dict['id'], note=False)
1317 except ExtractorError as err:
1318 self.report_warning('Unable to download subtitle for "%s": %s' %
1319 (sub_lang, compat_str(err.cause)))
1322 sub_filename = subtitles_filename(filename, sub_lang, sub_format)
1323 if self.params.get('nooverwrites', False) and os.path.exists(encodeFilename(sub_filename)):
1324 self.to_screen('[info] Video subtitle %s.%s is already_present' % (sub_lang, sub_format))
1326 self.to_screen('[info] Writing video subtitles to: ' + sub_filename)
1327 with io.open(encodeFilename(sub_filename), 'w', encoding='utf-8') as subfile:
1328 subfile.write(sub_data)
1329 except (OSError, IOError):
1330 self.report_error('Cannot write subtitles file ' + sub_filename)
1333 if self.params.get('writeinfojson', False):
1334 infofn = os.path.splitext(filename)[0] + '.info.json'
1335 if self.params.get('nooverwrites', False) and os.path.exists(encodeFilename(infofn)):
1336 self.to_screen('[info] Video description metadata is already present')
1338 self.to_screen('[info] Writing video description metadata as JSON to: ' + infofn)
1340 write_json_file(info_dict, infofn)
1341 except (OSError, IOError):
1342 self.report_error('Cannot write metadata to JSON file ' + infofn)
1345 self._write_thumbnails(info_dict, filename)
1347 if not self.params.get('skip_download', False):
1350 fd = get_suitable_downloader(info, self.params)(self, self.params)
1351 for ph in self._progress_hooks:
1352 fd.add_progress_hook(ph)
1353 if self.params.get('verbose'):
1354 self.to_stdout('[debug] Invoking downloader on %r' % info.get('url'))
1355 return fd.download(name, info)
1357 if info_dict.get('requested_formats') is not None:
1360 merger = FFmpegMergerPP(self, not self.params.get('keepvideo'))
1361 if not merger.available:
1363 self.report_warning('You have requested multiple '
1364 'formats but ffmpeg or avconv are not installed.'
1365 ' The formats won\'t be merged')
1367 postprocessors = [merger]
1368 for f in info_dict['requested_formats']:
1369 new_info = dict(info_dict)
1371 fname = self.prepare_filename(new_info)
1372 fname = prepend_extension(fname, 'f%s' % f['format_id'])
1373 downloaded.append(fname)
1374 partial_success = dl(fname, new_info)
1375 success = success and partial_success
1376 info_dict['__postprocessors'] = postprocessors
1377 info_dict['__files_to_merge'] = downloaded
1379 # Just a single file
1380 success = dl(filename, info_dict)
1381 except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
1382 self.report_error('unable to download video data: %s' % str(err))
1384 except (OSError, IOError) as err:
1385 raise UnavailableVideoError(err)
1386 except (ContentTooShortError, ) as err:
1387 self.report_error('content too short (expected %s bytes and served %s)' % (err.expected, err.downloaded))
1392 fixup_policy = self.params.get('fixup')
1393 if fixup_policy is None:
1394 fixup_policy = 'detect_or_warn'
1396 stretched_ratio = info_dict.get('stretched_ratio')
1397 if stretched_ratio is not None and stretched_ratio != 1:
1398 if fixup_policy == 'warn':
1399 self.report_warning('%s: Non-uniform pixel ratio (%s)' % (
1400 info_dict['id'], stretched_ratio))
1401 elif fixup_policy == 'detect_or_warn':
1402 stretched_pp = FFmpegFixupStretchedPP(self)
1403 if stretched_pp.available:
1404 info_dict.setdefault('__postprocessors', [])
1405 info_dict['__postprocessors'].append(stretched_pp)
1407 self.report_warning(
1408 '%s: Non-uniform pixel ratio (%s). Install ffmpeg or avconv to fix this automatically.' % (
1409 info_dict['id'], stretched_ratio))
1411 assert fixup_policy in ('ignore', 'never')
1413 if info_dict.get('requested_formats') is None and info_dict.get('container') == 'm4a_dash':
1414 if fixup_policy == 'warn':
1415 self.report_warning('%s: writing DASH m4a. Only some players support this container.' % (
1417 elif fixup_policy == 'detect_or_warn':
1418 fixup_pp = FFmpegFixupM4aPP(self)
1419 if fixup_pp.available:
1420 info_dict.setdefault('__postprocessors', [])
1421 info_dict['__postprocessors'].append(fixup_pp)
1423 self.report_warning(
1424 '%s: writing DASH m4a. Only some players support this container. Install ffmpeg or avconv to fix this automatically.' % (
1427 assert fixup_policy in ('ignore', 'never')
1430 self.post_process(filename, info_dict)
1431 except (PostProcessingError) as err:
1432 self.report_error('postprocessing: %s' % str(err))
1434 self.record_download_archive(info_dict)
1436 def download(self, url_list):
1437 """Download a given list of URLs."""
1438 outtmpl = self.params.get('outtmpl', DEFAULT_OUTTMPL)
1439 if (len(url_list) > 1 and
1440 '%' not in outtmpl and
1441 self.params.get('max_downloads') != 1):
1442 raise SameFileError(outtmpl)
1444 for url in url_list:
1446 # It also downloads the videos
1447 res = self.extract_info(url)
1448 except UnavailableVideoError:
1449 self.report_error('unable to download video')
1450 except MaxDownloadsReached:
1451 self.to_screen('[info] Maximum number of downloaded files reached.')
1454 if self.params.get('dump_single_json', False):
1455 self.to_stdout(json.dumps(res))
1457 return self._download_retcode
1459 def download_with_info_file(self, info_filename):
1460 with contextlib.closing(fileinput.FileInput(
1461 [info_filename], mode='r',
1462 openhook=fileinput.hook_encoded('utf-8'))) as f:
1463 # FileInput doesn't have a read method, we can't call json.load
1464 info = json.loads('\n'.join(f))
1466 self.process_ie_result(info, download=True)
1467 except DownloadError:
1468 webpage_url = info.get('webpage_url')
1469 if webpage_url is not None:
1470 self.report_warning('The info failed to download, trying with "%s"' % webpage_url)
1471 return self.download([webpage_url])
1474 return self._download_retcode
1476 def post_process(self, filename, ie_info):
1477 """Run all the postprocessors on the given file."""
1478 info = dict(ie_info)
1479 info['filepath'] = filename
1481 if ie_info.get('__postprocessors') is not None:
1482 pps_chain.extend(ie_info['__postprocessors'])
1483 pps_chain.extend(self._pps)
1484 for pp in pps_chain:
1486 old_filename = info['filepath']
1488 keep_video_wish, info = pp.run(info)
1489 if keep_video_wish is not None:
1491 keep_video = keep_video_wish
1492 elif keep_video is None:
1493 # No clear decision yet, let IE decide
1494 keep_video = keep_video_wish
1495 except PostProcessingError as e:
1496 self.report_error(e.msg)
1497 if keep_video is False and not self.params.get('keepvideo', False):
1499 self.to_screen('Deleting original file %s (pass -k to keep)' % old_filename)
1500 os.remove(encodeFilename(old_filename))
1501 except (IOError, OSError):
1502 self.report_warning('Unable to remove downloaded video file')
1504 def _make_archive_id(self, info_dict):
1505 # Future-proof against any change in case
1506 # and backwards compatibility with prior versions
1507 extractor = info_dict.get('extractor_key')
1508 if extractor is None:
1509 if 'id' in info_dict:
1510 extractor = info_dict.get('ie_key') # key in a playlist
1511 if extractor is None:
1512 return None # Incomplete video information
1513 return extractor.lower() + ' ' + info_dict['id']
1515 def in_download_archive(self, info_dict):
1516 fn = self.params.get('download_archive')
1520 vid_id = self._make_archive_id(info_dict)
1522 return False # Incomplete video information
1525 with locked_file(fn, 'r', encoding='utf-8') as archive_file:
1526 for line in archive_file:
1527 if line.strip() == vid_id:
1529 except IOError as ioe:
1530 if ioe.errno != errno.ENOENT:
1534 def record_download_archive(self, info_dict):
1535 fn = self.params.get('download_archive')
1538 vid_id = self._make_archive_id(info_dict)
1540 with locked_file(fn, 'a', encoding='utf-8') as archive_file:
1541 archive_file.write(vid_id + '\n')
1544 def format_resolution(format, default='unknown'):
1545 if format.get('vcodec') == 'none':
1547 if format.get('resolution') is not None:
1548 return format['resolution']
1549 if format.get('height') is not None:
1550 if format.get('width') is not None:
1551 res = '%sx%s' % (format['width'], format['height'])
1553 res = '%sp' % format['height']
1554 elif format.get('width') is not None:
1555 res = '?x%d' % format['width']
1560 def _format_note(self, fdict):
1562 if fdict.get('ext') in ['f4f', 'f4m']:
1563 res += '(unsupported) '
1564 if fdict.get('format_note') is not None:
1565 res += fdict['format_note'] + ' '
1566 if fdict.get('tbr') is not None:
1567 res += '%4dk ' % fdict['tbr']
1568 if fdict.get('container') is not None:
1571 res += '%s container' % fdict['container']
1572 if (fdict.get('vcodec') is not None and
1573 fdict.get('vcodec') != 'none'):
1576 res += fdict['vcodec']
1577 if fdict.get('vbr') is not None:
1579 elif fdict.get('vbr') is not None and fdict.get('abr') is not None:
1581 if fdict.get('vbr') is not None:
1582 res += '%4dk' % fdict['vbr']
1583 if fdict.get('fps') is not None:
1584 res += ', %sfps' % fdict['fps']
1585 if fdict.get('acodec') is not None:
1588 if fdict['acodec'] == 'none':
1591 res += '%-5s' % fdict['acodec']
1592 elif fdict.get('abr') is not None:
1596 if fdict.get('abr') is not None:
1597 res += '@%3dk' % fdict['abr']
1598 if fdict.get('asr') is not None:
1599 res += ' (%5dHz)' % fdict['asr']
1600 if fdict.get('filesize') is not None:
1603 res += format_bytes(fdict['filesize'])
1604 elif fdict.get('filesize_approx') is not None:
1607 res += '~' + format_bytes(fdict['filesize_approx'])
1610 def list_formats(self, info_dict):
1611 formats = info_dict.get('formats', [info_dict])
1613 [f['format_id'], f['ext'], self.format_resolution(f), self._format_note(f)]
1615 if f.get('preference') is None or f['preference'] >= -1000]
1616 if len(formats) > 1:
1617 table[-1][-1] += (' ' if table[-1][-1] else '') + '(best)'
1619 header_line = ['format code', 'extension', 'resolution', 'note']
1621 '[info] Available formats for %s:\n%s' %
1622 (info_dict['id'], render_table(header_line, table)))
1624 def list_thumbnails(self, info_dict):
1625 thumbnails = info_dict.get('thumbnails')
1627 tn_url = info_dict.get('thumbnail')
1629 thumbnails = [{'id': '0', 'url': tn_url}]
1632 '[info] No thumbnails present for %s' % info_dict['id'])
1636 '[info] Thumbnails for %s:' % info_dict['id'])
1637 self.to_screen(render_table(
1638 ['ID', 'width', 'height', 'URL'],
1639 [[t['id'], t.get('width', 'unknown'), t.get('height', 'unknown'), t['url']] for t in thumbnails]))
1641 def list_subtitles(self, video_id, subtitles, name='subtitles'):
1643 self.to_screen('%s has no %s' % (video_id, name))
1646 'Available %s for %s:' % (name, video_id))
1647 self.to_screen(render_table(
1648 ['Language', 'formats'],
1649 [[lang, ', '.join(f['ext'] for f in reversed(formats))]
1650 for lang, formats in subtitles.items()]))
1652 def urlopen(self, req):
1653 """ Start an HTTP download """
1655 # According to RFC 3986, URLs can not contain non-ASCII characters, however this is not
1656 # always respected by websites, some tend to give out URLs with non percent-encoded
1657 # non-ASCII characters (see telemb.py, ard.py [#3412])
1658 # urllib chokes on URLs with non-ASCII characters (see http://bugs.python.org/issue3991)
1659 # To work around aforementioned issue we will replace request's original URL with
1660 # percent-encoded one
1661 req_is_string = isinstance(req, compat_basestring)
1662 url = req if req_is_string else req.get_full_url()
1663 url_escaped = escape_url(url)
1665 # Substitute URL if any change after escaping
1666 if url != url_escaped:
1670 req = compat_urllib_request.Request(
1671 url_escaped, data=req.data, headers=req.headers,
1672 origin_req_host=req.origin_req_host, unverifiable=req.unverifiable)
1674 return self._opener.open(req, timeout=self._socket_timeout)
1676 def print_debug_header(self):
1677 if not self.params.get('verbose'):
1680 if type('') is not compat_str:
1681 # Python 2.6 on SLES11 SP1 (https://github.com/rg3/youtube-dl/issues/3326)
1682 self.report_warning(
1683 'Your Python is broken! Update to a newer and supported version')
1685 stdout_encoding = getattr(
1686 sys.stdout, 'encoding', 'missing (%s)' % type(sys.stdout).__name__)
1688 '[debug] Encodings: locale %s, fs %s, out %s, pref %s\n' % (
1689 locale.getpreferredencoding(),
1690 sys.getfilesystemencoding(),
1692 self.get_encoding()))
1693 write_string(encoding_str, encoding=None)
1695 self._write_string('[debug] youtube-dl version ' + __version__ + '\n')
1697 sp = subprocess.Popen(
1698 ['git', 'rev-parse', '--short', 'HEAD'],
1699 stdout=subprocess.PIPE, stderr=subprocess.PIPE,
1700 cwd=os.path.dirname(os.path.abspath(__file__)))
1701 out, err = sp.communicate()
1702 out = out.decode().strip()
1703 if re.match('[0-9a-f]+', out):
1704 self._write_string('[debug] Git HEAD: ' + out + '\n')
1710 self._write_string('[debug] Python version %s - %s\n' % (
1711 platform.python_version(), platform_name()))
1713 exe_versions = FFmpegPostProcessor.get_versions(self)
1714 exe_versions['rtmpdump'] = rtmpdump_version()
1715 exe_str = ', '.join(
1717 for exe, v in sorted(exe_versions.items())
1722 self._write_string('[debug] exe versions: %s\n' % exe_str)
1725 for handler in self._opener.handlers:
1726 if hasattr(handler, 'proxies'):
1727 proxy_map.update(handler.proxies)
1728 self._write_string('[debug] Proxy map: ' + compat_str(proxy_map) + '\n')
1730 if self.params.get('call_home', False):
1731 ipaddr = self.urlopen('https://yt-dl.org/ip').read().decode('utf-8')
1732 self._write_string('[debug] Public IP address: %s\n' % ipaddr)
1733 latest_version = self.urlopen(
1734 'https://yt-dl.org/latest/version').read().decode('utf-8')
1735 if version_tuple(latest_version) > version_tuple(__version__):
1736 self.report_warning(
1737 'You are using an outdated version (newest version: %s)! '
1738 'See https://yt-dl.org/update if you need help updating.' %
1741 def _setup_opener(self):
1742 timeout_val = self.params.get('socket_timeout')
1743 self._socket_timeout = 600 if timeout_val is None else float(timeout_val)
1745 opts_cookiefile = self.params.get('cookiefile')
1746 opts_proxy = self.params.get('proxy')
1748 if opts_cookiefile is None:
1749 self.cookiejar = compat_cookiejar.CookieJar()
1751 self.cookiejar = compat_cookiejar.MozillaCookieJar(
1753 if os.access(opts_cookiefile, os.R_OK):
1754 self.cookiejar.load()
1756 cookie_processor = compat_urllib_request.HTTPCookieProcessor(
1758 if opts_proxy is not None:
1759 if opts_proxy == '':
1762 proxies = {'http': opts_proxy, 'https': opts_proxy}
1764 proxies = compat_urllib_request.getproxies()
1765 # Set HTTPS proxy to HTTP one if given (https://github.com/rg3/youtube-dl/issues/805)
1766 if 'http' in proxies and 'https' not in proxies:
1767 proxies['https'] = proxies['http']
1768 proxy_handler = PerRequestProxyHandler(proxies)
1770 debuglevel = 1 if self.params.get('debug_printtraffic') else 0
1771 https_handler = make_HTTPS_handler(self.params, debuglevel=debuglevel)
1772 ydlh = YoutubeDLHandler(self.params, debuglevel=debuglevel)
1773 opener = compat_urllib_request.build_opener(
1774 https_handler, proxy_handler, cookie_processor, ydlh)
1775 # Delete the default user-agent header, which would otherwise apply in
1776 # cases where our custom HTTP handler doesn't come into play
1777 # (See https://github.com/rg3/youtube-dl/issues/1309 for details)
1778 opener.addheaders = []
1779 self._opener = opener
1781 def encode(self, s):
1782 if isinstance(s, bytes):
1783 return s # Already encoded
1786 return s.encode(self.get_encoding())
1787 except UnicodeEncodeError as err:
1788 err.reason = err.reason + '. Check your system encoding configuration or use the --encoding option.'
1791 def get_encoding(self):
1792 encoding = self.params.get('encoding')
1793 if encoding is None:
1794 encoding = preferredencoding()
1797 def _write_thumbnails(self, info_dict, filename):
1798 if self.params.get('writethumbnail', False):
1799 thumbnails = info_dict.get('thumbnails')
1801 thumbnails = [thumbnails[-1]]
1802 elif self.params.get('write_all_thumbnails', False):
1803 thumbnails = info_dict.get('thumbnails')
1808 # No thumbnails present, so return immediately
1811 for t in thumbnails:
1812 thumb_ext = determine_ext(t['url'], 'jpg')
1813 suffix = '_%s' % t['id'] if len(thumbnails) > 1 else ''
1814 thumb_display_id = '%s ' % t['id'] if len(thumbnails) > 1 else ''
1815 thumb_filename = os.path.splitext(filename)[0] + suffix + '.' + thumb_ext
1817 if self.params.get('nooverwrites', False) and os.path.exists(encodeFilename(thumb_filename)):
1818 self.to_screen('[%s] %s: Thumbnail %sis already present' %
1819 (info_dict['extractor'], info_dict['id'], thumb_display_id))
1821 self.to_screen('[%s] %s: Downloading thumbnail %s...' %
1822 (info_dict['extractor'], info_dict['id'], thumb_display_id))
1824 uf = self.urlopen(t['url'])
1825 with open(thumb_filename, 'wb') as thumbf:
1826 shutil.copyfileobj(uf, thumbf)
1827 self.to_screen('[%s] %s: Writing thumbnail %sto: %s' %
1828 (info_dict['extractor'], info_dict['id'], thumb_display_id, thumb_filename))
1829 except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
1830 self.report_warning('Unable to download thumbnail "%s": %s' %
1831 (t['url'], compat_str(err)))