2 # -*- coding: utf-8 -*-
4 from __future__ import absolute_import, unicode_literals
35 compat_urllib_request,
64 UnavailableVideoError,
74 from .cache import Cache
75 from .extractor import get_info_extractor, gen_extractors
76 from .downloader import get_suitable_downloader
77 from .downloader.rtmp import rtmpdump_version
78 from .postprocessor import (
80 FFmpegFixupStretchedPP,
85 from .version import __version__
88 class YoutubeDL(object):
91 YoutubeDL objects are the ones responsible of downloading the
92 actual video file and writing it to disk if the user has requested
93 it, among some other tasks. In most cases there should be one per
94 program. As, given a video URL, the downloader doesn't know how to
95 extract all the needed information, task that InfoExtractors do, it
96 has to pass the URL to one of them.
98 For this, YoutubeDL objects have a method that allows
99 InfoExtractors to be registered in a given order. When it is passed
100 a URL, the YoutubeDL object handles it to the first InfoExtractor it
101 finds that reports being able to handle it. The InfoExtractor extracts
102 all the information about the video or videos the URL refers to, and
103 YoutubeDL process the extracted information, possibly using a File
104 Downloader to download the video.
106 YoutubeDL objects accept a lot of parameters. In order not to saturate
107 the object constructor with arguments, it receives a dictionary of
108 options instead. These options are available through the params
109 attribute for the InfoExtractors to use. The YoutubeDL also
110 registers itself as the downloader in charge for the InfoExtractors
111 that are added to it, so this is a "mutual registration".
115 username: Username for authentication purposes.
116 password: Password for authentication purposes.
117 videopassword: Password for acces a video.
118 usenetrc: Use netrc for authentication instead.
119 verbose: Print additional info to stdout.
120 quiet: Do not print messages to stdout.
121 no_warnings: Do not print out anything for warnings.
122 forceurl: Force printing final URL.
123 forcetitle: Force printing title.
124 forceid: Force printing ID.
125 forcethumbnail: Force printing thumbnail URL.
126 forcedescription: Force printing description.
127 forcefilename: Force printing final filename.
128 forceduration: Force printing duration.
129 forcejson: Force printing info_dict as JSON.
130 dump_single_json: Force printing the info_dict of the whole playlist
131 (or video) as a single JSON line.
132 simulate: Do not download the video files.
133 format: Video format code. See options.py for more information.
134 format_limit: Highest quality format to try.
135 outtmpl: Template for output names.
136 restrictfilenames: Do not allow "&" and spaces in file names
137 ignoreerrors: Do not stop on download errors.
138 nooverwrites: Prevent overwriting files.
139 playliststart: Playlist item to start at.
140 playlistend: Playlist item to end at.
141 playlist_items: Specific indices of playlist to download.
142 playlistreverse: Download playlist items in reverse order.
143 matchtitle: Download only matching titles.
144 rejecttitle: Reject downloads for matching titles.
145 logger: Log messages to a logging.Logger instance.
146 logtostderr: Log messages to stderr instead of stdout.
147 writedescription: Write the video description to a .description file
148 writeinfojson: Write the video description to a .info.json file
149 writeannotations: Write the video annotations to a .annotations.xml file
150 writethumbnail: Write the thumbnail image to a file
151 write_all_thumbnails: Write all thumbnail formats to files
152 writesubtitles: Write the video subtitles to a file
153 writeautomaticsub: Write the automatic subtitles to a file
154 allsubtitles: Downloads all the subtitles of the video
155 (requires writesubtitles or writeautomaticsub)
156 listsubtitles: Lists all available subtitles for the video
157 subtitlesformat: The format code for subtitles
158 subtitleslangs: List of languages of the subtitles to download
159 keepvideo: Keep the video file after post-processing
160 daterange: A DateRange object, download only if the upload_date is in the range.
161 skip_download: Skip the actual download of the video file
162 cachedir: Location of the cache files in the filesystem.
163 False to disable filesystem cache.
164 noplaylist: Download single video instead of a playlist if in doubt.
165 age_limit: An integer representing the user's age in years.
166 Unsuitable videos for the given age are skipped.
167 min_views: An integer representing the minimum view count the video
168 must have in order to not be skipped.
169 Videos without view count information are always
170 downloaded. None for no limit.
171 max_views: An integer representing the maximum view count.
172 Videos that are more popular than that are not
174 Videos without view count information are always
175 downloaded. None for no limit.
176 download_archive: File name of a file where all downloads are recorded.
177 Videos already present in the file are not downloaded
179 cookiefile: File name where cookies should be read from and dumped to.
180 nocheckcertificate:Do not verify SSL certificates
181 prefer_insecure: Use HTTP instead of HTTPS to retrieve information.
182 At the moment, this is only supported by YouTube.
183 proxy: URL of the proxy server to use
184 socket_timeout: Time to wait for unresponsive hosts, in seconds
185 bidi_workaround: Work around buggy terminals without bidirectional text
186 support, using fridibi
187 debug_printtraffic:Print out sent and received HTTP traffic
188 include_ads: Download ads as well
189 default_search: Prepend this string if an input url is not valid.
190 'auto' for elaborate guessing
191 encoding: Use this encoding instead of the system-specified.
192 extract_flat: Do not resolve URLs, return the immediate result.
193 Pass in 'in_playlist' to only show this behavior for
195 postprocessors: A list of dictionaries, each with an entry
196 * key: The name of the postprocessor. See
197 youtube_dl/postprocessor/__init__.py for a list.
198 as well as any further keyword arguments for the
200 progress_hooks: A list of functions that get called on download
201 progress, with a dictionary with the entries
202 * status: One of "downloading" and "finished".
203 Check this first and ignore unknown values.
205 If status is one of "downloading" or "finished", the
206 following properties may also be present:
207 * filename: The final filename (always present)
208 * downloaded_bytes: Bytes on disk
209 * total_bytes: Size of the whole file, None if unknown
210 * tmpfilename: The filename we're currently writing to
211 * eta: The estimated time in seconds, None if unknown
212 * speed: The download speed in bytes/second, None if
215 Progress hooks are guaranteed to be called at least once
216 (with status "finished") if the download is successful.
217 merge_output_format: Extension to use when merging formats.
218 fixup: Automatically correct known faults of the file.
220 - "never": do nothing
221 - "warn": only emit a warning
222 - "detect_or_warn": check whether we can do anything
223 about it, warn otherwise (default)
224 source_address: (Experimental) Client-side IP address to bind to.
225 call_home: Boolean, true iff we are allowed to contact the
226 youtube-dl servers for debugging.
227 sleep_interval: Number of seconds to sleep before each download.
228 external_downloader: Executable of the external downloader to call.
229 listformats: Print an overview of available video formats and exit.
230 list_thumbnails: Print a table of all thumbnails and exit.
231 match_filter: A function that gets called with the info_dict of
233 If it returns a message, the video is ignored.
234 If it returns None, the video is downloaded.
235 match_filter_func in utils.py is one example for this.
236 no_color: Do not emit color codes in output.
239 The following parameters are not used by YoutubeDL itself, they are used by
241 nopart, updatetime, buffersize, ratelimit, min_filesize, max_filesize, test,
242 noresizebuffer, retries, continuedl, noprogress, consoletitle,
245 The following options are used by the post processors:
246 prefer_ffmpeg: If True, use ffmpeg instead of avconv if both are available,
247 otherwise prefer avconv.
248 exec_cmd: Arbitrary command to run after downloading
254 _download_retcode = None
255 _num_downloads = None
258 def __init__(self, params=None, auto_init=True):
259 """Create a FileDownloader object with the given options."""
263 self._ies_instances = {}
265 self._progress_hooks = []
266 self._download_retcode = 0
267 self._num_downloads = 0
268 self._screen_file = [sys.stdout, sys.stderr][params.get('logtostderr', False)]
269 self._err_file = sys.stderr
271 self.cache = Cache(self)
273 if params.get('bidi_workaround', False):
276 master, slave = pty.openpty()
277 width = get_term_width()
281 width_args = ['-w', str(width)]
283 stdin=subprocess.PIPE,
285 stderr=self._err_file)
287 self._output_process = subprocess.Popen(
288 ['bidiv'] + width_args, **sp_kwargs
291 self._output_process = subprocess.Popen(
292 ['fribidi', '-c', 'UTF-8'] + width_args, **sp_kwargs)
293 self._output_channel = os.fdopen(master, 'rb')
294 except OSError as ose:
296 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 $PATH.')
300 if (sys.version_info >= (3,) and sys.platform != 'win32' and
301 sys.getfilesystemencoding() in ['ascii', 'ANSI_X3.4-1968']
302 and not params.get('restrictfilenames', False)):
303 # On Python 3, the Unicode filesystem API will throw errors (#1474)
305 'Assuming --restrict-filenames since file system encoding '
306 'cannot encode all characters. '
307 'Set the LC_ALL environment variable to fix this.')
308 self.params['restrictfilenames'] = True
310 if '%(stitle)s' in self.params.get('outtmpl', ''):
311 self.report_warning('%(stitle)s is deprecated. Use the %(title)s and the --restrict-filenames flag(which also secures %(uploader)s et al) instead.')
316 self.print_debug_header()
317 self.add_default_info_extractors()
319 for pp_def_raw in self.params.get('postprocessors', []):
320 pp_class = get_postprocessor(pp_def_raw['key'])
321 pp_def = dict(pp_def_raw)
323 pp = pp_class(self, **compat_kwargs(pp_def))
324 self.add_post_processor(pp)
326 for ph in self.params.get('progress_hooks', []):
327 self.add_progress_hook(ph)
329 def warn_if_short_id(self, argv):
330 # short YouTube ID starting with dash?
332 i for i, a in enumerate(argv)
333 if re.match(r'^-[0-9A-Za-z_-]{10}$', a)]
337 [a for i, a in enumerate(argv) if i not in idxs] +
338 ['--'] + [argv[i] for i in idxs]
341 'Long argument string detected. '
342 'Use -- to separate parameters and URLs, like this:\n%s\n' %
343 args_to_str(correct_argv))
345 def add_info_extractor(self, ie):
346 """Add an InfoExtractor object to the end of the list."""
348 self._ies_instances[ie.ie_key()] = ie
349 ie.set_downloader(self)
351 def get_info_extractor(self, ie_key):
353 Get an instance of an IE with name ie_key, it will try to get one from
354 the _ies list, if there's no instance it will create a new one and add
355 it to the extractor list.
357 ie = self._ies_instances.get(ie_key)
359 ie = get_info_extractor(ie_key)()
360 self.add_info_extractor(ie)
363 def add_default_info_extractors(self):
365 Add the InfoExtractors returned by gen_extractors to the end of the list
367 for ie in gen_extractors():
368 self.add_info_extractor(ie)
370 def add_post_processor(self, pp):
371 """Add a PostProcessor object to the end of the chain."""
373 pp.set_downloader(self)
375 def add_progress_hook(self, ph):
376 """Add the progress hook (currently only for the file downloader)"""
377 self._progress_hooks.append(ph)
379 def _bidi_workaround(self, message):
380 if not hasattr(self, '_output_channel'):
383 assert hasattr(self, '_output_process')
384 assert isinstance(message, compat_str)
385 line_count = message.count('\n') + 1
386 self._output_process.stdin.write((message + '\n').encode('utf-8'))
387 self._output_process.stdin.flush()
388 res = ''.join(self._output_channel.readline().decode('utf-8')
389 for _ in range(line_count))
390 return res[:-len('\n')]
392 def to_screen(self, message, skip_eol=False):
393 """Print message to stdout if not in quiet mode."""
394 return self.to_stdout(message, skip_eol, check_quiet=True)
396 def _write_string(self, s, out=None):
397 write_string(s, out=out, encoding=self.params.get('encoding'))
399 def to_stdout(self, message, skip_eol=False, check_quiet=False):
400 """Print message to stdout if not in quiet mode."""
401 if self.params.get('logger'):
402 self.params['logger'].debug(message)
403 elif not check_quiet or not self.params.get('quiet', False):
404 message = self._bidi_workaround(message)
405 terminator = ['\n', ''][skip_eol]
406 output = message + terminator
408 self._write_string(output, self._screen_file)
410 def to_stderr(self, message):
411 """Print message to stderr."""
412 assert isinstance(message, compat_str)
413 if self.params.get('logger'):
414 self.params['logger'].error(message)
416 message = self._bidi_workaround(message)
417 output = message + '\n'
418 self._write_string(output, self._err_file)
420 def to_console_title(self, message):
421 if not self.params.get('consoletitle', False):
423 if os.name == 'nt' and ctypes.windll.kernel32.GetConsoleWindow():
424 # c_wchar_p() might not be necessary if `message` is
425 # already of type unicode()
426 ctypes.windll.kernel32.SetConsoleTitleW(ctypes.c_wchar_p(message))
427 elif 'TERM' in os.environ:
428 self._write_string('\033]0;%s\007' % message, self._screen_file)
430 def save_console_title(self):
431 if not self.params.get('consoletitle', False):
433 if 'TERM' in os.environ:
434 # Save the title on stack
435 self._write_string('\033[22;0t', self._screen_file)
437 def restore_console_title(self):
438 if not self.params.get('consoletitle', False):
440 if 'TERM' in os.environ:
441 # Restore the title from stack
442 self._write_string('\033[23;0t', self._screen_file)
445 self.save_console_title()
448 def __exit__(self, *args):
449 self.restore_console_title()
451 if self.params.get('cookiefile') is not None:
452 self.cookiejar.save()
454 def trouble(self, message=None, tb=None):
455 """Determine action to take when a download problem appears.
457 Depending on if the downloader has been configured to ignore
458 download errors or not, this method may throw an exception or
459 not when errors are found, after printing the message.
461 tb, if given, is additional traceback information.
463 if message is not None:
464 self.to_stderr(message)
465 if self.params.get('verbose'):
467 if sys.exc_info()[0]: # if .trouble has been called from an except block
469 if hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]:
470 tb += ''.join(traceback.format_exception(*sys.exc_info()[1].exc_info))
471 tb += compat_str(traceback.format_exc())
473 tb_data = traceback.format_list(traceback.extract_stack())
474 tb = ''.join(tb_data)
476 if not self.params.get('ignoreerrors', False):
477 if sys.exc_info()[0] and hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]:
478 exc_info = sys.exc_info()[1].exc_info
480 exc_info = sys.exc_info()
481 raise DownloadError(message, exc_info)
482 self._download_retcode = 1
484 def report_warning(self, message):
486 Print the message to stderr, it will be prefixed with 'WARNING:'
487 If stderr is a tty file the 'WARNING:' will be colored
489 if self.params.get('logger') is not None:
490 self.params['logger'].warning(message)
492 if self.params.get('no_warnings'):
494 if not self.params.get('no_color') and self._err_file.isatty() and os.name != 'nt':
495 _msg_header = '\033[0;33mWARNING:\033[0m'
497 _msg_header = 'WARNING:'
498 warning_message = '%s %s' % (_msg_header, message)
499 self.to_stderr(warning_message)
501 def report_error(self, message, tb=None):
503 Do the same as trouble, but prefixes the message with 'ERROR:', colored
504 in red if stderr is a tty file.
506 if not self.params.get('no_color') and self._err_file.isatty() and os.name != 'nt':
507 _msg_header = '\033[0;31mERROR:\033[0m'
509 _msg_header = 'ERROR:'
510 error_message = '%s %s' % (_msg_header, message)
511 self.trouble(error_message, tb)
513 def report_file_already_downloaded(self, file_name):
514 """Report file has already been fully downloaded."""
516 self.to_screen('[download] %s has already been downloaded' % file_name)
517 except UnicodeEncodeError:
518 self.to_screen('[download] The file has already been downloaded')
520 def prepare_filename(self, info_dict):
521 """Generate the output filename."""
523 template_dict = dict(info_dict)
525 template_dict['epoch'] = int(time.time())
526 autonumber_size = self.params.get('autonumber_size')
527 if autonumber_size is None:
529 autonumber_templ = '%0' + str(autonumber_size) + 'd'
530 template_dict['autonumber'] = autonumber_templ % self._num_downloads
531 if template_dict.get('playlist_index') is not None:
532 template_dict['playlist_index'] = '%0*d' % (len(str(template_dict['n_entries'])), template_dict['playlist_index'])
533 if template_dict.get('resolution') is None:
534 if template_dict.get('width') and template_dict.get('height'):
535 template_dict['resolution'] = '%dx%d' % (template_dict['width'], template_dict['height'])
536 elif template_dict.get('height'):
537 template_dict['resolution'] = '%sp' % template_dict['height']
538 elif template_dict.get('width'):
539 template_dict['resolution'] = '?x%d' % template_dict['width']
541 sanitize = lambda k, v: sanitize_filename(
543 restricted=self.params.get('restrictfilenames'),
545 template_dict = dict((k, sanitize(k, v))
546 for k, v in template_dict.items()
548 template_dict = collections.defaultdict(lambda: 'NA', template_dict)
550 outtmpl = self.params.get('outtmpl', DEFAULT_OUTTMPL)
551 tmpl = compat_expanduser(outtmpl)
552 filename = tmpl % template_dict
553 # Temporary fix for #4787
554 # 'Treat' all problem characters by passing filename through preferredencoding
555 # to workaround encoding issues with subprocess on python2 @ Windows
556 if sys.version_info < (3, 0) and sys.platform == 'win32':
557 filename = encodeFilename(filename, True).decode(preferredencoding())
559 except ValueError as err:
560 self.report_error('Error in output template: ' + str(err) + ' (encoding: ' + repr(preferredencoding()) + ')')
563 def _match_entry(self, info_dict, incomplete):
564 """ Returns None iff the file should be downloaded """
566 video_title = info_dict.get('title', info_dict.get('id', 'video'))
567 if 'title' in info_dict:
568 # This can happen when we're just evaluating the playlist
569 title = info_dict['title']
570 matchtitle = self.params.get('matchtitle', False)
572 if not re.search(matchtitle, title, re.IGNORECASE):
573 return '"' + title + '" title did not match pattern "' + matchtitle + '"'
574 rejecttitle = self.params.get('rejecttitle', False)
576 if re.search(rejecttitle, title, re.IGNORECASE):
577 return '"' + title + '" title matched reject pattern "' + rejecttitle + '"'
578 date = info_dict.get('upload_date', None)
580 dateRange = self.params.get('daterange', DateRange())
581 if date not in dateRange:
582 return '%s upload date is not in range %s' % (date_from_str(date).isoformat(), dateRange)
583 view_count = info_dict.get('view_count', None)
584 if view_count is not None:
585 min_views = self.params.get('min_views')
586 if min_views is not None and view_count < min_views:
587 return 'Skipping %s, because it has not reached minimum view count (%d/%d)' % (video_title, view_count, min_views)
588 max_views = self.params.get('max_views')
589 if max_views is not None and view_count > max_views:
590 return 'Skipping %s, because it has exceeded the maximum view count (%d/%d)' % (video_title, view_count, max_views)
591 if age_restricted(info_dict.get('age_limit'), self.params.get('age_limit')):
592 return 'Skipping "%s" because it is age restricted' % video_title
593 if self.in_download_archive(info_dict):
594 return '%s has already been recorded in archive' % video_title
597 match_filter = self.params.get('match_filter')
598 if match_filter is not None:
599 ret = match_filter(info_dict)
606 def add_extra_info(info_dict, extra_info):
607 '''Set the keys from extra_info in info dict if they are missing'''
608 for key, value in extra_info.items():
609 info_dict.setdefault(key, value)
611 def extract_info(self, url, download=True, ie_key=None, extra_info={},
614 Returns a list with a dictionary for each video we find.
615 If 'download', also downloads the videos.
616 extra_info is a dict containing the extra values to add to each result
620 ies = [self.get_info_extractor(ie_key)]
625 if not ie.suitable(url):
629 self.report_warning('The program functionality for this site has been marked as broken, '
630 'and will probably not work.')
633 ie_result = ie.extract(url)
634 if ie_result is None: # Finished already (backwards compatibility; listformats and friends should be moved here)
636 if isinstance(ie_result, list):
637 # Backwards compatibility: old IE result format
639 '_type': 'compat_list',
640 'entries': ie_result,
642 self.add_default_extra_info(ie_result, ie, url)
644 return self.process_ie_result(ie_result, download, extra_info)
647 except ExtractorError as de: # An error we somewhat expected
648 self.report_error(compat_str(de), de.format_traceback())
650 except MaxDownloadsReached:
652 except Exception as e:
653 if self.params.get('ignoreerrors', False):
654 self.report_error(compat_str(e), tb=compat_str(traceback.format_exc()))
659 self.report_error('no suitable InfoExtractor for URL %s' % url)
661 def add_default_extra_info(self, ie_result, ie, url):
662 self.add_extra_info(ie_result, {
663 'extractor': ie.IE_NAME,
665 'webpage_url_basename': url_basename(url),
666 'extractor_key': ie.ie_key(),
669 def process_ie_result(self, ie_result, download=True, extra_info={}):
671 Take the result of the ie(may be modified) and resolve all unresolved
672 references (URLs, playlist items).
674 It will also download the videos if 'download'.
675 Returns the resolved ie_result.
678 result_type = ie_result.get('_type', 'video')
680 if result_type in ('url', 'url_transparent'):
681 extract_flat = self.params.get('extract_flat', False)
682 if ((extract_flat == 'in_playlist' and 'playlist' in extra_info) or
683 extract_flat is True):
684 if self.params.get('forcejson', False):
685 self.to_stdout(json.dumps(ie_result))
688 if result_type == 'video':
689 self.add_extra_info(ie_result, extra_info)
690 return self.process_video_result(ie_result, download=download)
691 elif result_type == 'url':
692 # We have to add extra_info to the results because it may be
693 # contained in a playlist
694 return self.extract_info(ie_result['url'],
696 ie_key=ie_result.get('ie_key'),
697 extra_info=extra_info)
698 elif result_type == 'url_transparent':
699 # Use the information from the embedding page
700 info = self.extract_info(
701 ie_result['url'], ie_key=ie_result.get('ie_key'),
702 extra_info=extra_info, download=False, process=False)
704 force_properties = dict(
705 (k, v) for k, v in ie_result.items() if v is not None)
706 for f in ('_type', 'url'):
707 if f in force_properties:
708 del force_properties[f]
709 new_result = info.copy()
710 new_result.update(force_properties)
712 assert new_result.get('_type') != 'url_transparent'
714 return self.process_ie_result(
715 new_result, download=download, extra_info=extra_info)
716 elif result_type == 'playlist' or result_type == 'multi_video':
717 # We process each entry in the playlist
718 playlist = ie_result.get('title', None) or ie_result.get('id', None)
719 self.to_screen('[download] Downloading playlist: %s' % playlist)
721 playlist_results = []
723 playliststart = self.params.get('playliststart', 1) - 1
724 playlistend = self.params.get('playlistend', None)
725 # For backwards compatibility, interpret -1 as whole list
726 if playlistend == -1:
729 playlistitems_str = self.params.get('playlist_items', None)
731 if playlistitems_str is not None:
732 def iter_playlistitems(format):
733 for string_segment in format.split(','):
734 if '-' in string_segment:
735 start, end = string_segment.split('-')
736 for item in range(int(start), int(end) + 1):
739 yield int(string_segment)
740 playlistitems = iter_playlistitems(playlistitems_str)
742 ie_entries = ie_result['entries']
743 if isinstance(ie_entries, list):
744 n_all_entries = len(ie_entries)
746 entries = [ie_entries[i - 1] for i in playlistitems]
748 entries = ie_entries[playliststart:playlistend]
749 n_entries = len(entries)
751 "[%s] playlist %s: Collected %d video ids (downloading %d of them)" %
752 (ie_result['extractor'], playlist, n_all_entries, n_entries))
753 elif isinstance(ie_entries, PagedList):
756 for item in playlistitems:
757 entries.extend(ie_entries.getslice(
761 entries = ie_entries.getslice(
762 playliststart, playlistend)
763 n_entries = len(entries)
765 "[%s] playlist %s: Downloading %d videos" %
766 (ie_result['extractor'], playlist, n_entries))
769 entry_list = list(ie_entries)
770 entries = [entry_list[i - 1] for i in playlistitems]
772 entries = list(itertools.islice(
773 ie_entries, playliststart, playlistend))
774 n_entries = len(entries)
776 "[%s] playlist %s: Downloading %d videos" %
777 (ie_result['extractor'], playlist, n_entries))
779 if self.params.get('playlistreverse', False):
780 entries = entries[::-1]
782 for i, entry in enumerate(entries, 1):
783 self.to_screen('[download] Downloading video %s of %s' % (i, n_entries))
785 'n_entries': n_entries,
786 'playlist': playlist,
787 'playlist_id': ie_result.get('id'),
788 'playlist_title': ie_result.get('title'),
789 'playlist_index': i + playliststart,
790 'extractor': ie_result['extractor'],
791 'webpage_url': ie_result['webpage_url'],
792 'webpage_url_basename': url_basename(ie_result['webpage_url']),
793 'extractor_key': ie_result['extractor_key'],
796 reason = self._match_entry(entry, incomplete=True)
797 if reason is not None:
798 self.to_screen('[download] ' + reason)
801 entry_result = self.process_ie_result(entry,
804 playlist_results.append(entry_result)
805 ie_result['entries'] = playlist_results
807 elif result_type == 'compat_list':
809 'Extractor %s returned a compat_list result. '
810 'It needs to be updated.' % ie_result.get('extractor'))
816 'extractor': ie_result['extractor'],
817 'webpage_url': ie_result['webpage_url'],
818 'webpage_url_basename': url_basename(ie_result['webpage_url']),
819 'extractor_key': ie_result['extractor_key'],
823 ie_result['entries'] = [
824 self.process_ie_result(_fixup(r), download, extra_info)
825 for r in ie_result['entries']
829 raise Exception('Invalid result type: %s' % result_type)
831 def _apply_format_filter(self, format_spec, available_formats):
832 " Returns a tuple of the remaining format_spec and filtered formats "
842 operator_rex = re.compile(r'''(?x)\s*\[
843 (?P<key>width|height|tbr|abr|vbr|asr|filesize|fps)
844 \s*(?P<op>%s)(?P<none_inclusive>\s*\?)?\s*
845 (?P<value>[0-9.]+(?:[kKmMgGtTpPeEzZyY]i?[Bb]?)?)
847 ''' % '|'.join(map(re.escape, OPERATORS.keys())))
848 m = operator_rex.search(format_spec)
851 comparison_value = int(m.group('value'))
853 comparison_value = parse_filesize(m.group('value'))
854 if comparison_value is None:
855 comparison_value = parse_filesize(m.group('value') + 'B')
856 if comparison_value is None:
858 'Invalid value %r in format specification %r' % (
859 m.group('value'), format_spec))
860 op = OPERATORS[m.group('op')]
867 str_operator_rex = re.compile(r'''(?x)\s*\[
868 \s*(?P<key>ext|acodec|vcodec|container|protocol)
869 \s*(?P<op>%s)(?P<none_inclusive>\s*\?)?
870 \s*(?P<value>[a-zA-Z0-9_-]+)
872 ''' % '|'.join(map(re.escape, STR_OPERATORS.keys())))
873 m = str_operator_rex.search(format_spec)
875 comparison_value = m.group('value')
876 op = STR_OPERATORS[m.group('op')]
879 raise ValueError('Invalid format specification %r' % format_spec)
882 actual_value = f.get(m.group('key'))
883 if actual_value is None:
884 return m.group('none_inclusive')
885 return op(actual_value, comparison_value)
886 new_formats = [f for f in available_formats if _filter(f)]
888 new_format_spec = format_spec[:-len(m.group(0))]
889 if not new_format_spec:
890 new_format_spec = 'best'
892 return (new_format_spec, new_formats)
894 def select_format(self, format_spec, available_formats):
895 while format_spec.endswith(']'):
896 format_spec, available_formats = self._apply_format_filter(
897 format_spec, available_formats)
898 if not available_formats:
901 if format_spec == 'best' or format_spec is None:
902 return available_formats[-1]
903 elif format_spec == 'worst':
904 return available_formats[0]
905 elif format_spec == 'bestaudio':
907 f for f in available_formats
908 if f.get('vcodec') == 'none']
910 return audio_formats[-1]
911 elif format_spec == 'worstaudio':
913 f for f in available_formats
914 if f.get('vcodec') == 'none']
916 return audio_formats[0]
917 elif format_spec == 'bestvideo':
919 f for f in available_formats
920 if f.get('acodec') == 'none']
922 return video_formats[-1]
923 elif format_spec == 'worstvideo':
925 f for f in available_formats
926 if f.get('acodec') == 'none']
928 return video_formats[0]
930 extensions = ['mp4', 'flv', 'webm', '3gp', 'm4a', 'mp3', 'ogg', 'aac', 'wav']
931 if format_spec in extensions:
932 filter_f = lambda f: f['ext'] == format_spec
934 filter_f = lambda f: f['format_id'] == format_spec
935 matches = list(filter(filter_f, available_formats))
940 def _calc_headers(self, info_dict):
941 res = std_headers.copy()
943 add_headers = info_dict.get('http_headers')
945 res.update(add_headers)
947 cookies = self._calc_cookies(info_dict)
949 res['Cookie'] = cookies
953 def _calc_cookies(self, info_dict):
954 class _PseudoRequest(object):
955 def __init__(self, url):
958 self.unverifiable = False
960 def add_unredirected_header(self, k, v):
963 def get_full_url(self):
966 def is_unverifiable(self):
967 return self.unverifiable
969 def has_header(self, h):
970 return h in self.headers
972 def get_header(self, h, default=None):
973 return self.headers.get(h, default)
975 pr = _PseudoRequest(info_dict['url'])
976 self.cookiejar.add_cookie_header(pr)
977 return pr.headers.get('Cookie')
979 def process_video_result(self, info_dict, download=True):
980 assert info_dict.get('_type', 'video') == 'video'
982 if 'id' not in info_dict:
983 raise ExtractorError('Missing "id" field in extractor result')
984 if 'title' not in info_dict:
985 raise ExtractorError('Missing "title" field in extractor result')
987 if 'playlist' not in info_dict:
988 # It isn't part of a playlist
989 info_dict['playlist'] = None
990 info_dict['playlist_index'] = None
992 thumbnails = info_dict.get('thumbnails')
993 if thumbnails is None:
994 thumbnail = info_dict.get('thumbnail')
996 info_dict['thumbnails'] = thumbnails = [{'url': thumbnail}]
998 thumbnails.sort(key=lambda t: (
999 t.get('preference'), t.get('width'), t.get('height'),
1000 t.get('id'), t.get('url')))
1001 for i, t in enumerate(thumbnails):
1002 if 'width' in t and 'height' in t:
1003 t['resolution'] = '%dx%d' % (t['width'], t['height'])
1004 if t.get('id') is None:
1007 if thumbnails and 'thumbnail' not in info_dict:
1008 info_dict['thumbnail'] = thumbnails[-1]['url']
1010 if 'display_id' not in info_dict and 'id' in info_dict:
1011 info_dict['display_id'] = info_dict['id']
1013 if info_dict.get('upload_date') is None and info_dict.get('timestamp') is not None:
1014 # Working around negative timestamps in Windows
1015 # (see http://bugs.python.org/issue1646728)
1016 if info_dict['timestamp'] < 0 and os.name == 'nt':
1017 info_dict['timestamp'] = 0
1018 upload_date = datetime.datetime.utcfromtimestamp(
1019 info_dict['timestamp'])
1020 info_dict['upload_date'] = upload_date.strftime('%Y%m%d')
1022 if self.params.get('listsubtitles', False):
1023 if 'automatic_captions' in info_dict:
1024 self.list_subtitles(info_dict['id'], info_dict.get('automatic_captions'), 'automatic captions')
1025 self.list_subtitles(info_dict['id'], info_dict.get('subtitles'), 'subtitles')
1027 info_dict['requested_subtitles'] = self.process_subtitles(
1028 info_dict['id'], info_dict.get('subtitles'),
1029 info_dict.get('automatic_captions'))
1031 # This extractors handle format selection themselves
1032 if info_dict['extractor'] in ['Youku']:
1034 self.process_info(info_dict)
1037 # We now pick which formats have to be downloaded
1038 if info_dict.get('formats') is None:
1039 # There's only one format available
1040 formats = [info_dict]
1042 formats = info_dict['formats']
1045 raise ExtractorError('No video formats found!')
1047 # We check that all the formats have the format and format_id fields
1048 for i, format in enumerate(formats):
1049 if 'url' not in format:
1050 raise ExtractorError('Missing "url" key in result (index %d)' % i)
1052 if format.get('format_id') is None:
1053 format['format_id'] = compat_str(i)
1054 if format.get('format') is None:
1055 format['format'] = '{id} - {res}{note}'.format(
1056 id=format['format_id'],
1057 res=self.format_resolution(format),
1058 note=' ({0})'.format(format['format_note']) if format.get('format_note') is not None else '',
1060 # Automatically determine file extension if missing
1061 if 'ext' not in format:
1062 format['ext'] = determine_ext(format['url']).lower()
1063 # Add HTTP headers, so that external programs can use them from the
1065 full_format_info = info_dict.copy()
1066 full_format_info.update(format)
1067 format['http_headers'] = self._calc_headers(full_format_info)
1069 format_limit = self.params.get('format_limit', None)
1071 formats = list(takewhile_inclusive(
1072 lambda f: f['format_id'] != format_limit, formats
1075 # TODO Central sorting goes here
1077 if formats[0] is not info_dict:
1078 # only set the 'formats' fields if the original info_dict list them
1079 # otherwise we end up with a circular reference, the first (and unique)
1080 # element in the 'formats' field in info_dict is info_dict itself,
1081 # wich can't be exported to json
1082 info_dict['formats'] = formats
1083 if self.params.get('listformats'):
1084 self.list_formats(info_dict)
1086 if self.params.get('list_thumbnails'):
1087 self.list_thumbnails(info_dict)
1090 req_format = self.params.get('format')
1091 if req_format is None:
1093 formats_to_download = []
1094 # The -1 is for supporting YoutubeIE
1095 if req_format in ('-1', 'all'):
1096 formats_to_download = formats
1098 for rfstr in req_format.split(','):
1099 # We can accept formats requested in the format: 34/5/best, we pick
1100 # the first that is available, starting from left
1101 req_formats = rfstr.split('/')
1102 for rf in req_formats:
1103 if re.match(r'.+?\+.+?', rf) is not None:
1104 # Two formats have been requested like '137+139'
1105 format_1, format_2 = rf.split('+')
1106 formats_info = (self.select_format(format_1, formats),
1107 self.select_format(format_2, formats))
1108 if all(formats_info):
1109 # The first format must contain the video and the
1111 if formats_info[0].get('vcodec') == 'none':
1112 self.report_error('The first format must '
1113 'contain the video, try using '
1114 '"-f %s+%s"' % (format_2, format_1))
1117 formats_info[0]['ext']
1118 if self.params.get('merge_output_format') is None
1119 else self.params['merge_output_format'])
1121 'requested_formats': formats_info,
1122 'format': '%s+%s' % (formats_info[0].get('format'),
1123 formats_info[1].get('format')),
1124 'format_id': '%s+%s' % (formats_info[0].get('format_id'),
1125 formats_info[1].get('format_id')),
1126 'width': formats_info[0].get('width'),
1127 'height': formats_info[0].get('height'),
1128 'resolution': formats_info[0].get('resolution'),
1129 'fps': formats_info[0].get('fps'),
1130 'vcodec': formats_info[0].get('vcodec'),
1131 'vbr': formats_info[0].get('vbr'),
1132 'stretched_ratio': formats_info[0].get('stretched_ratio'),
1133 'acodec': formats_info[1].get('acodec'),
1134 'abr': formats_info[1].get('abr'),
1138 selected_format = None
1140 selected_format = self.select_format(rf, formats)
1141 if selected_format is not None:
1142 formats_to_download.append(selected_format)
1144 if not formats_to_download:
1145 raise ExtractorError('requested format not available',
1149 if len(formats_to_download) > 1:
1150 self.to_screen('[info] %s: downloading video in %s formats' % (info_dict['id'], len(formats_to_download)))
1151 for format in formats_to_download:
1152 new_info = dict(info_dict)
1153 new_info.update(format)
1154 self.process_info(new_info)
1155 # We update the info dict with the best quality format (backwards compatibility)
1156 info_dict.update(formats_to_download[-1])
1159 def process_subtitles(self, video_id, normal_subtitles, automatic_captions):
1160 """Select the requested subtitles and their format"""
1162 if normal_subtitles and self.params.get('writesubtitles'):
1163 available_subs.update(normal_subtitles)
1164 if automatic_captions and self.params.get('writeautomaticsub'):
1165 for lang, cap_info in automatic_captions.items():
1166 if lang not in available_subs:
1167 available_subs[lang] = cap_info
1169 if (not self.params.get('writesubtitles') and not
1170 self.params.get('writeautomaticsub') or not
1174 if self.params.get('allsubtitles', False):
1175 requested_langs = available_subs.keys()
1177 if self.params.get('subtitleslangs', False):
1178 requested_langs = self.params.get('subtitleslangs')
1179 elif 'en' in available_subs:
1180 requested_langs = ['en']
1182 requested_langs = [list(available_subs.keys())[0]]
1184 formats_query = self.params.get('subtitlesformat', 'best')
1185 formats_preference = formats_query.split('/') if formats_query else []
1187 for lang in requested_langs:
1188 formats = available_subs.get(lang)
1190 self.report_warning('%s subtitles not available for %s' % (lang, video_id))
1192 for ext in formats_preference:
1196 matches = list(filter(lambda f: f['ext'] == ext, formats))
1202 self.report_warning(
1203 'No subtitle format found matching "%s" for language %s, '
1204 'using %s' % (formats_query, lang, f['ext']))
1208 def process_info(self, info_dict):
1209 """Process a single resolved IE result."""
1211 assert info_dict.get('_type', 'video') == 'video'
1213 max_downloads = self.params.get('max_downloads')
1214 if max_downloads is not None:
1215 if self._num_downloads >= int(max_downloads):
1216 raise MaxDownloadsReached()
1218 info_dict['fulltitle'] = info_dict['title']
1219 if len(info_dict['title']) > 200:
1220 info_dict['title'] = info_dict['title'][:197] + '...'
1222 # Keep for backwards compatibility
1223 info_dict['stitle'] = info_dict['title']
1225 if 'format' not in info_dict:
1226 info_dict['format'] = info_dict['ext']
1228 reason = self._match_entry(info_dict, incomplete=False)
1229 if reason is not None:
1230 self.to_screen('[download] ' + reason)
1233 self._num_downloads += 1
1235 info_dict['_filename'] = filename = self.prepare_filename(info_dict)
1238 if self.params.get('forcetitle', False):
1239 self.to_stdout(info_dict['fulltitle'])
1240 if self.params.get('forceid', False):
1241 self.to_stdout(info_dict['id'])
1242 if self.params.get('forceurl', False):
1243 if info_dict.get('requested_formats') is not None:
1244 for f in info_dict['requested_formats']:
1245 self.to_stdout(f['url'] + f.get('play_path', ''))
1247 # For RTMP URLs, also include the playpath
1248 self.to_stdout(info_dict['url'] + info_dict.get('play_path', ''))
1249 if self.params.get('forcethumbnail', False) and info_dict.get('thumbnail') is not None:
1250 self.to_stdout(info_dict['thumbnail'])
1251 if self.params.get('forcedescription', False) and info_dict.get('description') is not None:
1252 self.to_stdout(info_dict['description'])
1253 if self.params.get('forcefilename', False) and filename is not None:
1254 self.to_stdout(filename)
1255 if self.params.get('forceduration', False) and info_dict.get('duration') is not None:
1256 self.to_stdout(formatSeconds(info_dict['duration']))
1257 if self.params.get('forceformat', False):
1258 self.to_stdout(info_dict['format'])
1259 if self.params.get('forcejson', False):
1260 self.to_stdout(json.dumps(info_dict))
1262 # Do nothing else if in simulate mode
1263 if self.params.get('simulate', False):
1266 if filename is None:
1270 dn = os.path.dirname(encodeFilename(filename))
1271 if dn and not os.path.exists(dn):
1273 except (OSError, IOError) as err:
1274 self.report_error('unable to create directory ' + compat_str(err))
1277 if self.params.get('writedescription', False):
1278 descfn = filename + '.description'
1279 if self.params.get('nooverwrites', False) and os.path.exists(encodeFilename(descfn)):
1280 self.to_screen('[info] Video description is already present')
1281 elif info_dict.get('description') is None:
1282 self.report_warning('There\'s no description to write.')
1285 self.to_screen('[info] Writing video description to: ' + descfn)
1286 with io.open(encodeFilename(descfn), 'w', encoding='utf-8') as descfile:
1287 descfile.write(info_dict['description'])
1288 except (OSError, IOError):
1289 self.report_error('Cannot write description file ' + descfn)
1292 if self.params.get('writeannotations', False):
1293 annofn = filename + '.annotations.xml'
1294 if self.params.get('nooverwrites', False) and os.path.exists(encodeFilename(annofn)):
1295 self.to_screen('[info] Video annotations are already present')
1298 self.to_screen('[info] Writing video annotations to: ' + annofn)
1299 with io.open(encodeFilename(annofn), 'w', encoding='utf-8') as annofile:
1300 annofile.write(info_dict['annotations'])
1301 except (KeyError, TypeError):
1302 self.report_warning('There are no annotations to write.')
1303 except (OSError, IOError):
1304 self.report_error('Cannot write annotations file: ' + annofn)
1307 subtitles_are_requested = any([self.params.get('writesubtitles', False),
1308 self.params.get('writeautomaticsub')])
1310 if subtitles_are_requested and info_dict.get('requested_subtitles'):
1311 # subtitles download errors are already managed as troubles in relevant IE
1312 # that way it will silently go on when used with unsupporting IE
1313 subtitles = info_dict['requested_subtitles']
1314 for sub_lang, sub_info in subtitles.items():
1315 sub_format = sub_info['ext']
1316 if sub_info.get('data') is not None:
1317 sub_data = sub_info['data']
1320 uf = self.urlopen(sub_info['url'])
1321 sub_data = uf.read().decode('utf-8')
1322 except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
1323 self.report_warning('Unable to download subtitle for "%s": %s' %
1324 (sub_lang, compat_str(err)))
1327 sub_filename = subtitles_filename(filename, sub_lang, sub_format)
1328 if self.params.get('nooverwrites', False) and os.path.exists(encodeFilename(sub_filename)):
1329 self.to_screen('[info] Video subtitle %s.%s is already_present' % (sub_lang, sub_format))
1331 self.to_screen('[info] Writing video subtitles to: ' + sub_filename)
1332 with io.open(encodeFilename(sub_filename), 'w', encoding='utf-8') as subfile:
1333 subfile.write(sub_data)
1334 except (OSError, IOError):
1335 self.report_error('Cannot write subtitles file ' + sub_filename)
1338 if self.params.get('writeinfojson', False):
1339 infofn = os.path.splitext(filename)[0] + '.info.json'
1340 if self.params.get('nooverwrites', False) and os.path.exists(encodeFilename(infofn)):
1341 self.to_screen('[info] Video description metadata is already present')
1343 self.to_screen('[info] Writing video description metadata as JSON to: ' + infofn)
1345 write_json_file(info_dict, infofn)
1346 except (OSError, IOError):
1347 self.report_error('Cannot write metadata to JSON file ' + infofn)
1350 self._write_thumbnails(info_dict, filename)
1352 if not self.params.get('skip_download', False):
1355 fd = get_suitable_downloader(info, self.params)(self, self.params)
1356 for ph in self._progress_hooks:
1357 fd.add_progress_hook(ph)
1358 if self.params.get('verbose'):
1359 self.to_stdout('[debug] Invoking downloader on %r' % info.get('url'))
1360 return fd.download(name, info)
1362 if info_dict.get('requested_formats') is not None:
1365 merger = FFmpegMergerPP(self, not self.params.get('keepvideo'))
1366 if not merger._executable:
1368 self.report_warning('You have requested multiple '
1369 'formats but ffmpeg or avconv are not installed.'
1370 ' The formats won\'t be merged')
1372 postprocessors = [merger]
1373 for f in info_dict['requested_formats']:
1374 new_info = dict(info_dict)
1376 fname = self.prepare_filename(new_info)
1377 fname = prepend_extension(fname, 'f%s' % f['format_id'])
1378 downloaded.append(fname)
1379 partial_success = dl(fname, new_info)
1380 success = success and partial_success
1381 info_dict['__postprocessors'] = postprocessors
1382 info_dict['__files_to_merge'] = downloaded
1384 # Just a single file
1385 success = dl(filename, info_dict)
1386 except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
1387 self.report_error('unable to download video data: %s' % str(err))
1389 except (OSError, IOError) as err:
1390 raise UnavailableVideoError(err)
1391 except (ContentTooShortError, ) as err:
1392 self.report_error('content too short (expected %s bytes and served %s)' % (err.expected, err.downloaded))
1397 fixup_policy = self.params.get('fixup')
1398 if fixup_policy is None:
1399 fixup_policy = 'detect_or_warn'
1401 stretched_ratio = info_dict.get('stretched_ratio')
1402 if stretched_ratio is not None and stretched_ratio != 1:
1403 if fixup_policy == 'warn':
1404 self.report_warning('%s: Non-uniform pixel ratio (%s)' % (
1405 info_dict['id'], stretched_ratio))
1406 elif fixup_policy == 'detect_or_warn':
1407 stretched_pp = FFmpegFixupStretchedPP(self)
1408 if stretched_pp.available:
1409 info_dict.setdefault('__postprocessors', [])
1410 info_dict['__postprocessors'].append(stretched_pp)
1412 self.report_warning(
1413 '%s: Non-uniform pixel ratio (%s). Install ffmpeg or avconv to fix this automatically.' % (
1414 info_dict['id'], stretched_ratio))
1416 assert fixup_policy in ('ignore', 'never')
1418 if info_dict.get('requested_formats') is None and info_dict.get('container') == 'm4a_dash':
1419 if fixup_policy == 'warn':
1420 self.report_warning('%s: writing DASH m4a. Only some players support this container.' % (
1422 elif fixup_policy == 'detect_or_warn':
1423 fixup_pp = FFmpegFixupM4aPP(self)
1424 if fixup_pp.available:
1425 info_dict.setdefault('__postprocessors', [])
1426 info_dict['__postprocessors'].append(fixup_pp)
1428 self.report_warning(
1429 '%s: writing DASH m4a. Only some players support this container. Install ffmpeg or avconv to fix this automatically.' % (
1432 assert fixup_policy in ('ignore', 'never')
1435 self.post_process(filename, info_dict)
1436 except (PostProcessingError) as err:
1437 self.report_error('postprocessing: %s' % str(err))
1439 self.record_download_archive(info_dict)
1441 def download(self, url_list):
1442 """Download a given list of URLs."""
1443 outtmpl = self.params.get('outtmpl', DEFAULT_OUTTMPL)
1444 if (len(url_list) > 1 and
1446 and self.params.get('max_downloads') != 1):
1447 raise SameFileError(outtmpl)
1449 for url in url_list:
1451 # It also downloads the videos
1452 res = self.extract_info(url)
1453 except UnavailableVideoError:
1454 self.report_error('unable to download video')
1455 except MaxDownloadsReached:
1456 self.to_screen('[info] Maximum number of downloaded files reached.')
1459 if self.params.get('dump_single_json', False):
1460 self.to_stdout(json.dumps(res))
1462 return self._download_retcode
1464 def download_with_info_file(self, info_filename):
1465 with io.open(info_filename, 'r', encoding='utf-8') as f:
1468 self.process_ie_result(info, download=True)
1469 except DownloadError:
1470 webpage_url = info.get('webpage_url')
1471 if webpage_url is not None:
1472 self.report_warning('The info failed to download, trying with "%s"' % webpage_url)
1473 return self.download([webpage_url])
1476 return self._download_retcode
1478 def post_process(self, filename, ie_info):
1479 """Run all the postprocessors on the given file."""
1480 info = dict(ie_info)
1481 info['filepath'] = filename
1483 if ie_info.get('__postprocessors') is not None:
1484 pps_chain.extend(ie_info['__postprocessors'])
1485 pps_chain.extend(self._pps)
1486 for pp in pps_chain:
1488 old_filename = info['filepath']
1490 keep_video_wish, info = pp.run(info)
1491 if keep_video_wish is not None:
1493 keep_video = keep_video_wish
1494 elif keep_video is None:
1495 # No clear decision yet, let IE decide
1496 keep_video = keep_video_wish
1497 except PostProcessingError as e:
1498 self.report_error(e.msg)
1499 if keep_video is False and not self.params.get('keepvideo', False):
1501 self.to_screen('Deleting original file %s (pass -k to keep)' % old_filename)
1502 os.remove(encodeFilename(old_filename))
1503 except (IOError, OSError):
1504 self.report_warning('Unable to remove downloaded video file')
1506 def _make_archive_id(self, info_dict):
1507 # Future-proof against any change in case
1508 # and backwards compatibility with prior versions
1509 extractor = info_dict.get('extractor_key')
1510 if extractor is None:
1511 if 'id' in info_dict:
1512 extractor = info_dict.get('ie_key') # key in a playlist
1513 if extractor is None:
1514 return None # Incomplete video information
1515 return extractor.lower() + ' ' + info_dict['id']
1517 def in_download_archive(self, info_dict):
1518 fn = self.params.get('download_archive')
1522 vid_id = self._make_archive_id(info_dict)
1524 return False # Incomplete video information
1527 with locked_file(fn, 'r', encoding='utf-8') as archive_file:
1528 for line in archive_file:
1529 if line.strip() == vid_id:
1531 except IOError as ioe:
1532 if ioe.errno != errno.ENOENT:
1536 def record_download_archive(self, info_dict):
1537 fn = self.params.get('download_archive')
1540 vid_id = self._make_archive_id(info_dict)
1542 with locked_file(fn, 'a', encoding='utf-8') as archive_file:
1543 archive_file.write(vid_id + '\n')
1546 def format_resolution(format, default='unknown'):
1547 if format.get('vcodec') == 'none':
1549 if format.get('resolution') is not None:
1550 return format['resolution']
1551 if format.get('height') is not None:
1552 if format.get('width') is not None:
1553 res = '%sx%s' % (format['width'], format['height'])
1555 res = '%sp' % format['height']
1556 elif format.get('width') is not None:
1557 res = '?x%d' % format['width']
1562 def _format_note(self, fdict):
1564 if fdict.get('ext') in ['f4f', 'f4m']:
1565 res += '(unsupported) '
1566 if fdict.get('format_note') is not None:
1567 res += fdict['format_note'] + ' '
1568 if fdict.get('tbr') is not None:
1569 res += '%4dk ' % fdict['tbr']
1570 if fdict.get('container') is not None:
1573 res += '%s container' % fdict['container']
1574 if (fdict.get('vcodec') is not None and
1575 fdict.get('vcodec') != 'none'):
1578 res += fdict['vcodec']
1579 if fdict.get('vbr') is not None:
1581 elif fdict.get('vbr') is not None and fdict.get('abr') is not None:
1583 if fdict.get('vbr') is not None:
1584 res += '%4dk' % fdict['vbr']
1585 if fdict.get('fps') is not None:
1586 res += ', %sfps' % fdict['fps']
1587 if fdict.get('acodec') is not None:
1590 if fdict['acodec'] == 'none':
1593 res += '%-5s' % fdict['acodec']
1594 elif fdict.get('abr') is not None:
1598 if fdict.get('abr') is not None:
1599 res += '@%3dk' % fdict['abr']
1600 if fdict.get('asr') is not None:
1601 res += ' (%5dHz)' % fdict['asr']
1602 if fdict.get('filesize') is not None:
1605 res += format_bytes(fdict['filesize'])
1606 elif fdict.get('filesize_approx') is not None:
1609 res += '~' + format_bytes(fdict['filesize_approx'])
1612 def list_formats(self, info_dict):
1613 def line(format, idlen=20):
1614 return (('%-' + compat_str(idlen + 1) + 's%-10s%-12s%s') % (
1615 format['format_id'],
1617 self.format_resolution(format),
1618 self._format_note(format),
1621 formats = info_dict.get('formats', [info_dict])
1622 idlen = max(len('format code'),
1623 max(len(f['format_id']) for f in formats))
1625 line(f, idlen) for f in formats
1626 if f.get('preference') is None or f['preference'] >= -1000]
1627 if len(formats) > 1:
1628 formats_s[-1] += (' ' if self._format_note(formats[-1]) else '') + '(best)'
1630 header_line = line({
1631 'format_id': 'format code', 'ext': 'extension',
1632 'resolution': 'resolution', 'format_note': 'note'}, idlen=idlen)
1634 '[info] Available formats for %s:\n%s\n%s' %
1635 (info_dict['id'], header_line, '\n'.join(formats_s)))
1637 def list_thumbnails(self, info_dict):
1638 thumbnails = info_dict.get('thumbnails')
1640 tn_url = info_dict.get('thumbnail')
1642 thumbnails = [{'id': '0', 'url': tn_url}]
1645 '[info] No thumbnails present for %s' % info_dict['id'])
1649 '[info] Thumbnails for %s:' % info_dict['id'])
1650 self.to_screen(render_table(
1651 ['ID', 'width', 'height', 'URL'],
1652 [[t['id'], t.get('width', 'unknown'), t.get('height', 'unknown'), t['url']] for t in thumbnails]))
1654 def list_subtitles(self, video_id, subtitles, name='subtitles'):
1656 self.to_screen('%s has no %s' % (video_id, name))
1659 'Available %s for %s:' % (name, video_id))
1660 self.to_screen(render_table(
1661 ['Language', 'formats'],
1662 [[lang, ', '.join(f['ext'] for f in reversed(formats))]
1663 for lang, formats in subtitles.items()]))
1665 def urlopen(self, req):
1666 """ Start an HTTP download """
1668 # According to RFC 3986, URLs can not contain non-ASCII characters, however this is not
1669 # always respected by websites, some tend to give out URLs with non percent-encoded
1670 # non-ASCII characters (see telemb.py, ard.py [#3412])
1671 # urllib chokes on URLs with non-ASCII characters (see http://bugs.python.org/issue3991)
1672 # To work around aforementioned issue we will replace request's original URL with
1673 # percent-encoded one
1674 req_is_string = isinstance(req, compat_basestring)
1675 url = req if req_is_string else req.get_full_url()
1676 url_escaped = escape_url(url)
1678 # Substitute URL if any change after escaping
1679 if url != url_escaped:
1683 req = compat_urllib_request.Request(
1684 url_escaped, data=req.data, headers=req.headers,
1685 origin_req_host=req.origin_req_host, unverifiable=req.unverifiable)
1687 return self._opener.open(req, timeout=self._socket_timeout)
1689 def print_debug_header(self):
1690 if not self.params.get('verbose'):
1693 if type('') is not compat_str:
1694 # Python 2.6 on SLES11 SP1 (https://github.com/rg3/youtube-dl/issues/3326)
1695 self.report_warning(
1696 'Your Python is broken! Update to a newer and supported version')
1698 stdout_encoding = getattr(
1699 sys.stdout, 'encoding', 'missing (%s)' % type(sys.stdout).__name__)
1701 '[debug] Encodings: locale %s, fs %s, out %s, pref %s\n' % (
1702 locale.getpreferredencoding(),
1703 sys.getfilesystemencoding(),
1705 self.get_encoding()))
1706 write_string(encoding_str, encoding=None)
1708 self._write_string('[debug] youtube-dl version ' + __version__ + '\n')
1710 sp = subprocess.Popen(
1711 ['git', 'rev-parse', '--short', 'HEAD'],
1712 stdout=subprocess.PIPE, stderr=subprocess.PIPE,
1713 cwd=os.path.dirname(os.path.abspath(__file__)))
1714 out, err = sp.communicate()
1715 out = out.decode().strip()
1716 if re.match('[0-9a-f]+', out):
1717 self._write_string('[debug] Git HEAD: ' + out + '\n')
1723 self._write_string('[debug] Python version %s - %s\n' % (
1724 platform.python_version(), platform_name()))
1726 exe_versions = FFmpegPostProcessor.get_versions()
1727 exe_versions['rtmpdump'] = rtmpdump_version()
1728 exe_str = ', '.join(
1730 for exe, v in sorted(exe_versions.items())
1735 self._write_string('[debug] exe versions: %s\n' % exe_str)
1738 for handler in self._opener.handlers:
1739 if hasattr(handler, 'proxies'):
1740 proxy_map.update(handler.proxies)
1741 self._write_string('[debug] Proxy map: ' + compat_str(proxy_map) + '\n')
1743 if self.params.get('call_home', False):
1744 ipaddr = self.urlopen('https://yt-dl.org/ip').read().decode('utf-8')
1745 self._write_string('[debug] Public IP address: %s\n' % ipaddr)
1746 latest_version = self.urlopen(
1747 'https://yt-dl.org/latest/version').read().decode('utf-8')
1748 if version_tuple(latest_version) > version_tuple(__version__):
1749 self.report_warning(
1750 'You are using an outdated version (newest version: %s)! '
1751 'See https://yt-dl.org/update if you need help updating.' %
1754 def _setup_opener(self):
1755 timeout_val = self.params.get('socket_timeout')
1756 self._socket_timeout = 600 if timeout_val is None else float(timeout_val)
1758 opts_cookiefile = self.params.get('cookiefile')
1759 opts_proxy = self.params.get('proxy')
1761 if opts_cookiefile is None:
1762 self.cookiejar = compat_cookiejar.CookieJar()
1764 self.cookiejar = compat_cookiejar.MozillaCookieJar(
1766 if os.access(opts_cookiefile, os.R_OK):
1767 self.cookiejar.load()
1769 cookie_processor = compat_urllib_request.HTTPCookieProcessor(
1771 if opts_proxy is not None:
1772 if opts_proxy == '':
1775 proxies = {'http': opts_proxy, 'https': opts_proxy}
1777 proxies = compat_urllib_request.getproxies()
1778 # Set HTTPS proxy to HTTP one if given (https://github.com/rg3/youtube-dl/issues/805)
1779 if 'http' in proxies and 'https' not in proxies:
1780 proxies['https'] = proxies['http']
1781 proxy_handler = compat_urllib_request.ProxyHandler(proxies)
1783 debuglevel = 1 if self.params.get('debug_printtraffic') else 0
1784 https_handler = make_HTTPS_handler(self.params, debuglevel=debuglevel)
1785 ydlh = YoutubeDLHandler(self.params, debuglevel=debuglevel)
1786 opener = compat_urllib_request.build_opener(
1787 https_handler, proxy_handler, cookie_processor, ydlh)
1788 # Delete the default user-agent header, which would otherwise apply in
1789 # cases where our custom HTTP handler doesn't come into play
1790 # (See https://github.com/rg3/youtube-dl/issues/1309 for details)
1791 opener.addheaders = []
1792 self._opener = opener
1794 def encode(self, s):
1795 if isinstance(s, bytes):
1796 return s # Already encoded
1799 return s.encode(self.get_encoding())
1800 except UnicodeEncodeError as err:
1801 err.reason = err.reason + '. Check your system encoding configuration or use the --encoding option.'
1804 def get_encoding(self):
1805 encoding = self.params.get('encoding')
1806 if encoding is None:
1807 encoding = preferredencoding()
1810 def _write_thumbnails(self, info_dict, filename):
1811 if self.params.get('writethumbnail', False):
1812 thumbnails = info_dict.get('thumbnails')
1814 thumbnails = [thumbnails[-1]]
1815 elif self.params.get('write_all_thumbnails', False):
1816 thumbnails = info_dict.get('thumbnails')
1821 # No thumbnails present, so return immediately
1824 for t in thumbnails:
1825 thumb_ext = determine_ext(t['url'], 'jpg')
1826 suffix = '_%s' % t['id'] if len(thumbnails) > 1 else ''
1827 thumb_display_id = '%s ' % t['id'] if len(thumbnails) > 1 else ''
1828 thumb_filename = os.path.splitext(filename)[0] + suffix + '.' + thumb_ext
1830 if self.params.get('nooverwrites', False) and os.path.exists(encodeFilename(thumb_filename)):
1831 self.to_screen('[%s] %s: Thumbnail %sis already present' %
1832 (info_dict['extractor'], info_dict['id'], thumb_display_id))
1834 self.to_screen('[%s] %s: Downloading thumbnail %s...' %
1835 (info_dict['extractor'], info_dict['id'], thumb_display_id))
1837 uf = self.urlopen(t['url'])
1838 with open(thumb_filename, 'wb') as thumbf:
1839 shutil.copyfileobj(uf, thumbf)
1840 self.to_screen('[%s] %s: Writing thumbnail %sto: %s' %
1841 (info_dict['extractor'], info_dict['id'], thumb_display_id, thumb_filename))
1842 except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
1843 self.report_warning('Unable to download thumbnail "%s": %s' %
1844 (t['url'], compat_str(err)))