4 from __future__
import absolute_import
, unicode_literals
31 from string
import ascii_letters
35 compat_get_terminal_size
,
42 compat_tokenize_tokenize
,
44 compat_urllib_request
,
45 compat_urllib_request_DataHandler
,
46 windows_enable_vt_mode
,
48 from .cookies
import load_cookies
58 DOT_DESKTOP_LINK_TEMPLATE
,
59 DOT_URL_LINK_TEMPLATE
,
60 DOT_WEBLOC_LINK_TEMPLATE
,
88 PerRequestProxyHandler
,
94 register_socks_protocols
,
109 supports_terminal_sequences
,
115 UnavailableVideoError
,
121 YoutubeDLCookieProcessor
,
123 YoutubeDLRedirectHandler
,
125 from .cache
import Cache
126 from .extractor
import (
127 gen_extractor_classes
,
130 _PLUGIN_CLASSES
as plugin_extractors
132 from .extractor
.openload
import PhantomJSwrapper
133 from .downloader
import (
135 get_suitable_downloader
,
136 shorten_protocol_name
138 from .downloader
.rtmp
import rtmpdump_version
139 from .postprocessor
import (
142 FFmpegFixupDurationPP
,
145 FFmpegFixupStretchedPP
,
146 FFmpegFixupTimestampPP
,
149 MoveFilesAfterDownloadPP
,
150 _PLUGIN_CLASSES
as plugin_postprocessors
152 from .update
import detect_variant
153 from .version
import __version__
155 if compat_os_name
== 'nt':
159 class YoutubeDL(object):
162 YoutubeDL objects are the ones responsible of downloading the
163 actual video file and writing it to disk if the user has requested
164 it, among some other tasks. In most cases there should be one per
165 program. As, given a video URL, the downloader doesn't know how to
166 extract all the needed information, task that InfoExtractors do, it
167 has to pass the URL to one of them.
169 For this, YoutubeDL objects have a method that allows
170 InfoExtractors to be registered in a given order. When it is passed
171 a URL, the YoutubeDL object handles it to the first InfoExtractor it
172 finds that reports being able to handle it. The InfoExtractor extracts
173 all the information about the video or videos the URL refers to, and
174 YoutubeDL process the extracted information, possibly using a File
175 Downloader to download the video.
177 YoutubeDL objects accept a lot of parameters. In order not to saturate
178 the object constructor with arguments, it receives a dictionary of
179 options instead. These options are available through the params
180 attribute for the InfoExtractors to use. The YoutubeDL also
181 registers itself as the downloader in charge for the InfoExtractors
182 that are added to it, so this is a "mutual registration".
186 username: Username for authentication purposes.
187 password: Password for authentication purposes.
188 videopassword: Password for accessing a video.
189 ap_mso: Adobe Pass multiple-system operator identifier.
190 ap_username: Multiple-system operator account username.
191 ap_password: Multiple-system operator account password.
192 usenetrc: Use netrc for authentication instead.
193 verbose: Print additional info to stdout.
194 quiet: Do not print messages to stdout.
195 no_warnings: Do not print out anything for warnings.
196 forceprint: A list of templates to force print
197 forceurl: Force printing final URL. (Deprecated)
198 forcetitle: Force printing title. (Deprecated)
199 forceid: Force printing ID. (Deprecated)
200 forcethumbnail: Force printing thumbnail URL. (Deprecated)
201 forcedescription: Force printing description. (Deprecated)
202 forcefilename: Force printing final filename. (Deprecated)
203 forceduration: Force printing duration. (Deprecated)
204 forcejson: Force printing info_dict as JSON.
205 dump_single_json: Force printing the info_dict of the whole playlist
206 (or video) as a single JSON line.
207 force_write_download_archive: Force writing download archive regardless
208 of 'skip_download' or 'simulate'.
209 simulate: Do not download the video files. If unset (or None),
210 simulate only if listsubtitles, listformats or list_thumbnails is used
211 format: Video format code. see "FORMAT SELECTION" for more details.
212 allow_unplayable_formats: Allow unplayable formats to be extracted and downloaded.
213 ignore_no_formats_error: Ignore "No video formats" error. Usefull for
214 extracting metadata even if the video is not actually
215 available for download (experimental)
216 format_sort: How to sort the video formats. see "Sorting Formats"
218 format_sort_force: Force the given format_sort. see "Sorting Formats"
220 allow_multiple_video_streams: Allow multiple video streams to be merged
222 allow_multiple_audio_streams: Allow multiple audio streams to be merged
224 check_formats Whether to test if the formats are downloadable.
225 Can be True (check all), False (check none)
226 or None (check only if requested by extractor)
227 paths: Dictionary of output paths. The allowed keys are 'home'
228 'temp' and the keys of OUTTMPL_TYPES (in utils.py)
229 outtmpl: Dictionary of templates for output names. Allowed keys
230 are 'default' and the keys of OUTTMPL_TYPES (in utils.py).
231 For compatibility with youtube-dl, a single string can also be used
232 outtmpl_na_placeholder: Placeholder for unavailable meta fields.
233 restrictfilenames: Do not allow "&" and spaces in file names
234 trim_file_name: Limit length of filename (extension excluded)
235 windowsfilenames: Force the filenames to be windows compatible
236 ignoreerrors: Do not stop on download/postprocessing errors.
237 Can be 'only_download' to ignore only download errors.
238 Default is 'only_download' for CLI, but False for API
239 skip_playlist_after_errors: Number of allowed failures until the rest of
240 the playlist is skipped
241 force_generic_extractor: Force downloader to use the generic extractor
242 overwrites: Overwrite all video and metadata files if True,
243 overwrite only non-video files if None
244 and don't overwrite any file if False
245 For compatibility with youtube-dl,
246 "nooverwrites" may also be used instead
247 playliststart: Playlist item to start at.
248 playlistend: Playlist item to end at.
249 playlist_items: Specific indices of playlist to download.
250 playlistreverse: Download playlist items in reverse order.
251 playlistrandom: Download playlist items in random order.
252 matchtitle: Download only matching titles.
253 rejecttitle: Reject downloads for matching titles.
254 logger: Log messages to a logging.Logger instance.
255 logtostderr: Log messages to stderr instead of stdout.
256 consoletitle: Display progress in console window's titlebar.
257 writedescription: Write the video description to a .description file
258 writeinfojson: Write the video description to a .info.json file
259 clean_infojson: Remove private fields from the infojson
260 getcomments: Extract video comments. This will not be written to disk
261 unless writeinfojson is also given
262 writeannotations: Write the video annotations to a .annotations.xml file
263 writethumbnail: Write the thumbnail image to a file
264 allow_playlist_files: Whether to write playlists' description, infojson etc
265 also to disk when using the 'write*' options
266 write_all_thumbnails: Write all thumbnail formats to files
267 writelink: Write an internet shortcut file, depending on the
268 current platform (.url/.webloc/.desktop)
269 writeurllink: Write a Windows internet shortcut file (.url)
270 writewebloclink: Write a macOS internet shortcut file (.webloc)
271 writedesktoplink: Write a Linux internet shortcut file (.desktop)
272 writesubtitles: Write the video subtitles to a file
273 writeautomaticsub: Write the automatically generated subtitles to a file
274 allsubtitles: Deprecated - Use subtitleslangs = ['all']
275 Downloads all the subtitles of the video
276 (requires writesubtitles or writeautomaticsub)
277 listsubtitles: Lists all available subtitles for the video
278 subtitlesformat: The format code for subtitles
279 subtitleslangs: List of languages of the subtitles to download (can be regex).
280 The list may contain "all" to refer to all the available
281 subtitles. The language can be prefixed with a "-" to
282 exclude it from the requested languages. Eg: ['all', '-live_chat']
283 keepvideo: Keep the video file after post-processing
284 daterange: A DateRange object, download only if the upload_date is in the range.
285 skip_download: Skip the actual download of the video file
286 cachedir: Location of the cache files in the filesystem.
287 False to disable filesystem cache.
288 noplaylist: Download single video instead of a playlist if in doubt.
289 age_limit: An integer representing the user's age in years.
290 Unsuitable videos for the given age are skipped.
291 min_views: An integer representing the minimum view count the video
292 must have in order to not be skipped.
293 Videos without view count information are always
294 downloaded. None for no limit.
295 max_views: An integer representing the maximum view count.
296 Videos that are more popular than that are not
298 Videos without view count information are always
299 downloaded. None for no limit.
300 download_archive: File name of a file where all downloads are recorded.
301 Videos already present in the file are not downloaded
303 break_on_existing: Stop the download process after attempting to download a
304 file that is in the archive.
305 break_on_reject: Stop the download process when encountering a video that
306 has been filtered out.
307 cookiefile: File name where cookies should be read from and dumped to
308 cookiesfrombrowser: A tuple containing the name of the browser and the profile
309 name/path from where cookies are loaded.
310 Eg: ('chrome', ) or (vivaldi, 'default')
311 nocheckcertificate:Do not verify SSL certificates
312 prefer_insecure: Use HTTP instead of HTTPS to retrieve information.
313 At the moment, this is only supported by YouTube.
314 proxy: URL of the proxy server to use
315 geo_verification_proxy: URL of the proxy to use for IP address verification
316 on geo-restricted sites.
317 socket_timeout: Time to wait for unresponsive hosts, in seconds
318 bidi_workaround: Work around buggy terminals without bidirectional text
319 support, using fridibi
320 debug_printtraffic:Print out sent and received HTTP traffic
321 include_ads: Download ads as well
322 default_search: Prepend this string if an input url is not valid.
323 'auto' for elaborate guessing
324 encoding: Use this encoding instead of the system-specified.
325 extract_flat: Do not resolve URLs, return the immediate result.
326 Pass in 'in_playlist' to only show this behavior for
328 postprocessors: A list of dictionaries, each with an entry
329 * key: The name of the postprocessor. See
330 yt_dlp/postprocessor/__init__.py for a list.
331 * when: When to run the postprocessor. Can be one of
332 pre_process|before_dl|post_process|after_move.
333 Assumed to be 'post_process' if not given
334 post_hooks: Deprecated - Register a custom postprocessor instead
335 A list of functions that get called as the final step
336 for each video file, after all postprocessors have been
337 called. The filename will be passed as the only argument.
338 progress_hooks: A list of functions that get called on download
339 progress, with a dictionary with the entries
340 * status: One of "downloading", "error", or "finished".
341 Check this first and ignore unknown values.
342 * info_dict: The extracted info_dict
344 If status is one of "downloading", or "finished", the
345 following properties may also be present:
346 * filename: The final filename (always present)
347 * tmpfilename: The filename we're currently writing to
348 * downloaded_bytes: Bytes on disk
349 * total_bytes: Size of the whole file, None if unknown
350 * total_bytes_estimate: Guess of the eventual file size,
352 * elapsed: The number of seconds since download started.
353 * eta: The estimated time in seconds, None if unknown
354 * speed: The download speed in bytes/second, None if
356 * fragment_index: The counter of the currently
357 downloaded video fragment.
358 * fragment_count: The number of fragments (= individual
359 files that will be merged)
361 Progress hooks are guaranteed to be called at least once
362 (with status "finished") if the download is successful.
363 postprocessor_hooks: A list of functions that get called on postprocessing
364 progress, with a dictionary with the entries
365 * status: One of "started", "processing", or "finished".
366 Check this first and ignore unknown values.
367 * postprocessor: Name of the postprocessor
368 * info_dict: The extracted info_dict
370 Progress hooks are guaranteed to be called at least twice
371 (with status "started" and "finished") if the processing is successful.
372 merge_output_format: Extension to use when merging formats.
373 final_ext: Expected final extension; used to detect when the file was
374 already downloaded and converted. "merge_output_format" is
375 replaced by this extension when given
376 fixup: Automatically correct known faults of the file.
378 - "never": do nothing
379 - "warn": only emit a warning
380 - "detect_or_warn": check whether we can do anything
381 about it, warn otherwise (default)
382 source_address: Client-side IP address to bind to.
383 call_home: Boolean, true iff we are allowed to contact the
384 yt-dlp servers for debugging. (BROKEN)
385 sleep_interval_requests: Number of seconds to sleep between requests
387 sleep_interval: Number of seconds to sleep before each download when
388 used alone or a lower bound of a range for randomized
389 sleep before each download (minimum possible number
390 of seconds to sleep) when used along with
392 max_sleep_interval:Upper bound of a range for randomized sleep before each
393 download (maximum possible number of seconds to sleep).
394 Must only be used along with sleep_interval.
395 Actual sleep time will be a random float from range
396 [sleep_interval; max_sleep_interval].
397 sleep_interval_subtitles: Number of seconds to sleep before each subtitle download
398 listformats: Print an overview of available video formats and exit.
399 list_thumbnails: Print a table of all thumbnails and exit.
400 match_filter: A function that gets called with the info_dict of
402 If it returns a message, the video is ignored.
403 If it returns None, the video is downloaded.
404 match_filter_func in utils.py is one example for this.
405 no_color: Do not emit color codes in output.
406 geo_bypass: Bypass geographic restriction via faking X-Forwarded-For
409 Two-letter ISO 3166-2 country code that will be used for
410 explicit geographic restriction bypassing via faking
411 X-Forwarded-For HTTP header
413 IP range in CIDR notation that will be used similarly to
416 The following options determine which downloader is picked:
417 external_downloader: A dictionary of protocol keys and the executable of the
418 external downloader to use for it. The allowed protocols
419 are default|http|ftp|m3u8|dash|rtsp|rtmp|mms.
420 Set the value to 'native' to use the native downloader
421 hls_prefer_native: Deprecated - Use external_downloader = {'m3u8': 'native'}
422 or {'m3u8': 'ffmpeg'} instead.
423 Use the native HLS downloader instead of ffmpeg/avconv
424 if True, otherwise use ffmpeg/avconv if False, otherwise
425 use downloader suggested by extractor if None.
426 compat_opts: Compatibility options. See "Differences in default behavior".
427 The following options do not work when used through the API:
428 filename, abort-on-error, multistreams, no-live-chat, format-sort
429 no-clean-infojson, no-playlist-metafiles, no-keep-subs.
430 Refer __init__.py for their implementation
431 progress_template: Dictionary of templates for progress outputs.
432 Allowed keys are 'download', 'postprocess',
433 'download-title' (console title) and 'postprocess-title'.
434 The template is mapped on a dictionary with keys 'progress' and 'info'
436 The following parameters are not used by YoutubeDL itself, they are used by
437 the downloader (see yt_dlp/downloader/common.py):
438 nopart, updatetime, buffersize, ratelimit, throttledratelimit, min_filesize,
439 max_filesize, test, noresizebuffer, retries, fragment_retries, continuedl,
440 noprogress, xattr_set_filesize, hls_use_mpegts, http_chunk_size,
441 external_downloader_args.
443 The following options are used by the post processors:
444 prefer_ffmpeg: If False, use avconv instead of ffmpeg if both are available,
445 otherwise prefer ffmpeg. (avconv support is deprecated)
446 ffmpeg_location: Location of the ffmpeg/avconv binary; either the path
447 to the binary or its containing directory.
448 postprocessor_args: A dictionary of postprocessor/executable keys (in lower case)
449 and a list of additional command-line arguments for the
450 postprocessor/executable. The dict can also have "PP+EXE" keys
451 which are used when the given exe is used by the given PP.
452 Use 'default' as the name for arguments to passed to all PP
453 For compatibility with youtube-dl, a single list of args
456 The following options are used by the extractors:
457 extractor_retries: Number of times to retry for known errors
458 dynamic_mpd: Whether to process dynamic DASH manifests (default: True)
459 hls_split_discontinuity: Split HLS playlists to different formats at
460 discontinuities such as ad breaks (default: False)
461 extractor_args: A dictionary of arguments to be passed to the extractors.
462 See "EXTRACTOR ARGUMENTS" for details.
463 Eg: {'youtube': {'skip': ['dash', 'hls']}}
464 youtube_include_dash_manifest: Deprecated - Use extractor_args instead.
465 If True (default), DASH manifests and related
466 data will be downloaded and processed by extractor.
467 You can reduce network I/O by disabling it if you don't
468 care about DASH. (only for youtube)
469 youtube_include_hls_manifest: Deprecated - Use extractor_args instead.
470 If True (default), HLS manifests and related
471 data will be downloaded and processed by extractor.
472 You can reduce network I/O by disabling it if you don't
473 care about HLS. (only for youtube)
476 _NUMERIC_FIELDS
= set((
477 'width', 'height', 'tbr', 'abr', 'asr', 'vbr', 'fps', 'filesize', 'filesize_approx',
478 'timestamp', 'release_timestamp',
479 'duration', 'view_count', 'like_count', 'dislike_count', 'repost_count',
480 'average_rating', 'comment_count', 'age_limit',
481 'start_time', 'end_time',
482 'chapter_number', 'season_number', 'episode_number',
483 'track_number', 'disc_number', 'release_year',
486 _format_selection_exts
= {
487 'audio': {'m4a', 'mp3', 'ogg', 'aac'}
,
488 'video': {'mp4', 'flv', 'webm', '3gp'}
,
489 'storyboards': {'mhtml'}
,
494 _pps
= {'pre_process': [], 'before_dl': [], 'after_move': [], 'post_process': []}
495 _printed_messages
= set()
496 _first_webpage_request
= True
497 _download_retcode
= None
498 _num_downloads
= None
500 _playlist_urls
= set()
503 def __init__(self
, params
=None, auto_init
=True):
504 """Create a FileDownloader object with the given options.
505 @param auto_init Whether to load the default extractors and print header (if verbose).
506 Set to 'no_verbose_header' to not ptint the header
511 self
._ies
_instances
= {}
512 self
._pps
= {'pre_process': [], 'before_dl': [], 'after_move': [], 'post_process': []}
513 self
._printed
_messages
= set()
514 self
._first
_webpage
_request
= True
515 self
._post
_hooks
= []
516 self
._progress
_hooks
= []
517 self
._postprocessor
_hooks
= []
518 self
._download
_retcode
= 0
519 self
._num
_downloads
= 0
520 self
._screen
_file
= [sys
.stdout
, sys
.stderr
][params
.get('logtostderr', False)]
521 self
._err
_file
= sys
.stderr
523 self
.cache
= Cache(self
)
525 windows_enable_vt_mode()
526 # FIXME: This will break if we ever print color to stdout
527 self
.params
['no_color'] = self
.params
.get('no_color') or not supports_terminal_sequences(self
._err
_file
)
529 if sys
.version_info
< (3, 6):
531 'Python version %d.%d is not supported! Please update to Python 3.6 or above' % sys
.version_info
[:2])
533 if self
.params
.get('allow_unplayable_formats'):
535 f
'You have asked for {self._color_text("unplayable formats", "blue")} to be listed/downloaded. '
536 'This is a developer option intended for debugging. \n'
537 ' If you experience any issues while using this option, '
538 f
'{self._color_text("DO NOT", "red")} open a bug report')
540 def check_deprecated(param
, option
, suggestion
):
541 if self
.params
.get(param
) is not None:
542 self
.report_warning('%s is deprecated. Use %s instead' % (option
, suggestion
))
546 if check_deprecated('cn_verification_proxy', '--cn-verification-proxy', '--geo-verification-proxy'):
547 if self
.params
.get('geo_verification_proxy') is None:
548 self
.params
['geo_verification_proxy'] = self
.params
['cn_verification_proxy']
550 check_deprecated('autonumber', '--auto-number', '-o "%(autonumber)s-%(title)s.%(ext)s"')
551 check_deprecated('usetitle', '--title', '-o "%(title)s-%(id)s.%(ext)s"')
552 check_deprecated('useid', '--id', '-o "%(id)s.%(ext)s"')
554 for msg
in self
.params
.get('warnings', []):
555 self
.report_warning(msg
)
557 if 'overwrites' not in self
.params
and self
.params
.get('nooverwrites') is not None:
558 # nooverwrites was unnecessarily changed to overwrites
559 # in 0c3d0f51778b153f65c21906031c2e091fcfb641
560 # This ensures compatibility with both keys
561 self
.params
['overwrites'] = not self
.params
['nooverwrites']
562 elif self
.params
.get('overwrites') is None:
563 self
.params
.pop('overwrites', None)
565 self
.params
['nooverwrites'] = not self
.params
['overwrites']
567 if params
.get('bidi_workaround', False):
570 master
, slave
= pty
.openpty()
571 width
= compat_get_terminal_size().columns
575 width_args
= ['-w', str(width
)]
577 stdin
=subprocess
.PIPE
,
579 stderr
=self
._err
_file
)
581 self
._output
_process
= Popen(['bidiv'] + width_args
, **sp_kwargs
)
583 self
._output
_process
= Popen(['fribidi', '-c', 'UTF-8'] + width_args
, **sp_kwargs
)
584 self
._output
_channel
= os
.fdopen(master
, 'rb')
585 except OSError as ose
:
586 if ose
.errno
== errno
.ENOENT
:
587 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 $PATH.')
591 if (sys
.platform
!= 'win32'
592 and sys
.getfilesystemencoding() in ['ascii', 'ANSI_X3.4-1968']
593 and not params
.get('restrictfilenames', False)):
594 # Unicode filesystem API will throw errors (#1474, #13027)
596 'Assuming --restrict-filenames since file system encoding '
597 'cannot encode all characters. '
598 'Set the LC_ALL environment variable to fix this.')
599 self
.params
['restrictfilenames'] = True
601 self
.outtmpl_dict
= self
.parse_outtmpl()
603 # Creating format selector here allows us to catch syntax errors before the extraction
604 self
.format_selector
= (
605 None if self
.params
.get('format') is None
606 else self
.build_format_selector(self
.params
['format']))
611 if auto_init
!= 'no_verbose_header':
612 self
.print_debug_header()
613 self
.add_default_info_extractors()
615 for pp_def_raw
in self
.params
.get('postprocessors', []):
616 pp_def
= dict(pp_def_raw
)
617 when
= pp_def
.pop('when', 'post_process')
618 pp_class
= get_postprocessor(pp_def
.pop('key'))
619 pp
= pp_class(self
, **compat_kwargs(pp_def
))
620 self
.add_post_processor(pp
, when
=when
)
622 for ph
in self
.params
.get('post_hooks', []):
623 self
.add_post_hook(ph
)
625 for ph
in self
.params
.get('progress_hooks', []):
626 self
.add_progress_hook(ph
)
628 register_socks_protocols()
630 def preload_download_archive(fn
):
631 """Preload the archive, if any is specified"""
634 self
.write_debug('Loading archive file %r\n' % fn
)
636 with locked_file(fn
, 'r', encoding
='utf-8') as archive_file
:
637 for line
in archive_file
:
638 self
.archive
.add(line
.strip())
639 except IOError as ioe
:
640 if ioe
.errno
!= errno
.ENOENT
:
646 preload_download_archive(self
.params
.get('download_archive'))
648 def warn_if_short_id(self
, argv
):
649 # short YouTube ID starting with dash?
651 i
for i
, a
in enumerate(argv
)
652 if re
.match(r
'^-[0-9A-Za-z_-]{10}$', a
)]
656 + [a
for i
, a
in enumerate(argv
) if i
not in idxs
]
657 + ['--'] + [argv
[i
] for i
in idxs
]
660 'Long argument string detected. '
661 'Use -- to separate parameters and URLs, like this:\n%s\n' %
662 args_to_str(correct_argv
))
664 def add_info_extractor(self
, ie
):
665 """Add an InfoExtractor object to the end of the list."""
667 self
._ies
[ie_key
] = ie
668 if not isinstance(ie
, type):
669 self
._ies
_instances
[ie_key
] = ie
670 ie
.set_downloader(self
)
672 def _get_info_extractor_class(self
, ie_key
):
673 ie
= self
._ies
.get(ie_key
)
675 ie
= get_info_extractor(ie_key
)
676 self
.add_info_extractor(ie
)
679 def get_info_extractor(self
, ie_key
):
681 Get an instance of an IE with name ie_key, it will try to get one from
682 the _ies list, if there's no instance it will create a new one and add
683 it to the extractor list.
685 ie
= self
._ies
_instances
.get(ie_key
)
687 ie
= get_info_extractor(ie_key
)()
688 self
.add_info_extractor(ie
)
691 def add_default_info_extractors(self
):
693 Add the InfoExtractors returned by gen_extractors to the end of the list
695 for ie
in gen_extractor_classes():
696 self
.add_info_extractor(ie
)
698 def add_post_processor(self
, pp
, when
='post_process'):
699 """Add a PostProcessor object to the end of the chain."""
700 self
._pps
[when
].append(pp
)
701 pp
.set_downloader(self
)
703 def add_post_hook(self
, ph
):
704 """Add the post hook"""
705 self
._post
_hooks
.append(ph
)
707 def add_progress_hook(self
, ph
):
708 """Add the download progress hook"""
709 self
._progress
_hooks
.append(ph
)
711 def add_postprocessor_hook(self
, ph
):
712 """Add the postprocessing progress hook"""
713 self
._postprocessor
_hooks
.append(ph
)
715 def _bidi_workaround(self
, message
):
716 if not hasattr(self
, '_output_channel'):
719 assert hasattr(self
, '_output_process')
720 assert isinstance(message
, compat_str
)
721 line_count
= message
.count('\n') + 1
722 self
._output
_process
.stdin
.write((message
+ '\n').encode('utf-8'))
723 self
._output
_process
.stdin
.flush()
724 res
= ''.join(self
._output
_channel
.readline().decode('utf-8')
725 for _
in range(line_count
))
726 return res
[:-len('\n')]
728 def _write_string(self
, message
, out
=None, only_once
=False):
730 if message
in self
._printed
_messages
:
732 self
._printed
_messages
.add(message
)
733 write_string(message
, out
=out
, encoding
=self
.params
.get('encoding'))
735 def to_stdout(self
, message
, skip_eol
=False, quiet
=False):
736 """Print message to stdout"""
737 if self
.params
.get('logger'):
738 self
.params
['logger'].debug(message
)
739 elif not quiet
or self
.params
.get('verbose'):
741 '%s%s' % (self
._bidi
_workaround
(message
), ('' if skip_eol
else '\n')),
742 self
._err
_file
if quiet
else self
._screen
_file
)
744 def to_stderr(self
, message
, only_once
=False):
745 """Print message to stderr"""
746 assert isinstance(message
, compat_str
)
747 if self
.params
.get('logger'):
748 self
.params
['logger'].error(message
)
750 self
._write
_string
('%s\n' % self
._bidi
_workaround
(message
), self
._err
_file
, only_once
=only_once
)
752 def to_console_title(self
, message
):
753 if not self
.params
.get('consoletitle', False):
755 if compat_os_name
== 'nt':
756 if ctypes
.windll
.kernel32
.GetConsoleWindow():
757 # c_wchar_p() might not be necessary if `message` is
758 # already of type unicode()
759 ctypes
.windll
.kernel32
.SetConsoleTitleW(ctypes
.c_wchar_p(message
))
760 elif 'TERM' in os
.environ
:
761 self
._write
_string
('\033]0;%s\007' % message
, self
._screen
_file
)
763 def save_console_title(self
):
764 if not self
.params
.get('consoletitle', False):
766 if self
.params
.get('simulate'):
768 if compat_os_name
!= 'nt' and 'TERM' in os
.environ
:
769 # Save the title on stack
770 self
._write
_string
('\033[22;0t', self
._screen
_file
)
772 def restore_console_title(self
):
773 if not self
.params
.get('consoletitle', False):
775 if self
.params
.get('simulate'):
777 if compat_os_name
!= 'nt' and 'TERM' in os
.environ
:
778 # Restore the title from stack
779 self
._write
_string
('\033[23;0t', self
._screen
_file
)
782 self
.save_console_title()
785 def __exit__(self
, *args
):
786 self
.restore_console_title()
788 if self
.params
.get('cookiefile') is not None:
789 self
.cookiejar
.save(ignore_discard
=True, ignore_expires
=True)
791 def trouble(self
, message
=None, tb
=None):
792 """Determine action to take when a download problem appears.
794 Depending on if the downloader has been configured to ignore
795 download errors or not, this method may throw an exception or
796 not when errors are found, after printing the message.
798 tb, if given, is additional traceback information.
800 if message
is not None:
801 self
.to_stderr(message
)
802 if self
.params
.get('verbose'):
804 if sys
.exc_info()[0]: # if .trouble has been called from an except block
806 if hasattr(sys
.exc_info()[1], 'exc_info') and sys
.exc_info()[1].exc_info
[0]:
807 tb
+= ''.join(traceback
.format_exception(*sys
.exc_info()[1].exc_info
))
808 tb
+= encode_compat_str(traceback
.format_exc())
810 tb_data
= traceback
.format_list(traceback
.extract_stack())
811 tb
= ''.join(tb_data
)
814 if not self
.params
.get('ignoreerrors'):
815 if sys
.exc_info()[0] and hasattr(sys
.exc_info()[1], 'exc_info') and sys
.exc_info()[1].exc_info
[0]:
816 exc_info
= sys
.exc_info()[1].exc_info
818 exc_info
= sys
.exc_info()
819 raise DownloadError(message
, exc_info
)
820 self
._download
_retcode
= 1
822 def to_screen(self
, message
, skip_eol
=False):
823 """Print message to stdout if not in quiet mode"""
825 message
, skip_eol
, quiet
=self
.params
.get('quiet', False))
827 def _color_text(self
, text
, color
):
828 if self
.params
.get('no_color'):
830 return f
'{TERMINAL_SEQUENCES[color.upper()]}{text}{TERMINAL_SEQUENCES["RESET_STYLE"]}'
832 def report_warning(self
, message
, only_once
=False):
834 Print the message to stderr, it will be prefixed with 'WARNING:'
835 If stderr is a tty file the 'WARNING:' will be colored
837 if self
.params
.get('logger') is not None:
838 self
.params
['logger'].warning(message
)
840 if self
.params
.get('no_warnings'):
842 self
.to_stderr(f
'{self._color_text("WARNING:", "yellow")} {message}', only_once
)
844 def report_error(self
, message
, tb
=None):
846 Do the same as trouble, but prefixes the message with 'ERROR:', colored
847 in red if stderr is a tty file.
849 self
.trouble(f
'{self._color_text("ERROR:", "red")} {message}', tb
)
851 def write_debug(self
, message
, only_once
=False):
852 '''Log debug message or Print message to stderr'''
853 if not self
.params
.get('verbose', False):
855 message
= '[debug] %s' % message
856 if self
.params
.get('logger'):
857 self
.params
['logger'].debug(message
)
859 self
.to_stderr(message
, only_once
)
861 def report_file_already_downloaded(self
, file_name
):
862 """Report file has already been fully downloaded."""
864 self
.to_screen('[download] %s has already been downloaded' % file_name
)
865 except UnicodeEncodeError:
866 self
.to_screen('[download] The file has already been downloaded')
868 def report_file_delete(self
, file_name
):
869 """Report that existing file will be deleted."""
871 self
.to_screen('Deleting existing file %s' % file_name
)
872 except UnicodeEncodeError:
873 self
.to_screen('Deleting existing file')
875 def raise_no_formats(self
, info
, forced
=False):
876 has_drm
= info
.get('__has_drm')
877 msg
= 'This video is DRM protected' if has_drm
else 'No video formats found!'
878 expected
= self
.params
.get('ignore_no_formats_error')
879 if forced
or not expected
:
880 raise ExtractorError(msg
, video_id
=info
['id'], ie
=info
['extractor'],
881 expected
=has_drm
or expected
)
883 self
.report_warning(msg
)
885 def parse_outtmpl(self
):
886 outtmpl_dict
= self
.params
.get('outtmpl', {})
887 if not isinstance(outtmpl_dict
, dict):
888 outtmpl_dict
= {'default': outtmpl_dict}
889 # Remove spaces in the default template
890 if self
.params
.get('restrictfilenames'):
891 sanitize
= lambda x
: x
.replace(' - ', ' ').replace(' ', '-')
893 sanitize
= lambda x
: x
894 outtmpl_dict
.update({
895 k
: sanitize(v
) for k
, v
in DEFAULT_OUTTMPL
.items()
896 if outtmpl_dict
.get(k
) is None})
897 for key
, val
in outtmpl_dict
.items():
898 if isinstance(val
, bytes):
900 'Parameter outtmpl is bytes, but should be a unicode string. '
901 'Put from __future__ import unicode_literals at the top of your code file or consider switching to Python 3.x.')
904 def get_output_path(self
, dir_type
='', filename
=None):
905 paths
= self
.params
.get('paths', {})
906 assert isinstance(paths
, dict)
908 expand_path(paths
.get('home', '').strip()),
909 expand_path(paths
.get(dir_type
, '').strip()) if dir_type
else '',
912 # Temporary fix for #4787
913 # 'Treat' all problem characters by passing filename through preferredencoding
914 # to workaround encoding issues with subprocess on python2 @ Windows
915 if sys
.version_info
< (3, 0) and sys
.platform
== 'win32':
916 path
= encodeFilename(path
, True).decode(preferredencoding())
917 return sanitize_path(path
, force
=self
.params
.get('windowsfilenames'))
920 def _outtmpl_expandpath(outtmpl
):
921 # expand_path translates '%%' into '%' and '$$' into '$'
922 # correspondingly that is not what we want since we need to keep
923 # '%%' intact for template dict substitution step. Working around
924 # with boundary-alike separator hack.
925 sep
= ''.join([random
.choice(ascii_letters
) for _
in range(32)])
926 outtmpl
= outtmpl
.replace('%%', '%{0}%'.format(sep
)).replace('$$', '${0}$'.format(sep
))
928 # outtmpl should be expand_path'ed before template dict substitution
929 # because meta fields may contain env variables we don't want to
930 # be expanded. For example, for outtmpl "%(title)s.%(ext)s" and
931 # title "Hello $PATH", we don't want `$PATH` to be expanded.
932 return expand_path(outtmpl
).replace(sep
, '')
935 def escape_outtmpl(outtmpl
):
936 ''' Escape any remaining strings like %s, %abc% etc. '''
938 STR_FORMAT_RE_TMPL
.format('', '(?![%(\0])'),
939 lambda mobj
: ('' if mobj
.group('has_key') else '%') + mobj
.group(0),
943 def validate_outtmpl(cls
, outtmpl
):
944 ''' @return None or Exception object '''
946 STR_FORMAT_RE_TMPL
.format('[^)]*', '[ljqBU]'),
947 lambda mobj
: f
'{mobj.group(0)[:-1]}s',
948 cls
._outtmpl
_expandpath
(outtmpl
))
950 cls
.escape_outtmpl(outtmpl
) % collections
.defaultdict(int)
952 except ValueError as err
:
956 def _copy_infodict(info_dict
):
957 info_dict
= dict(info_dict
)
958 for key
in ('__original_infodict', '__postprocessors'):
959 info_dict
.pop(key
, None)
962 def prepare_outtmpl(self
, outtmpl
, info_dict
, sanitize
=None):
963 """ Make the outtmpl and info_dict suitable for substitution: ydl.escape_outtmpl(outtmpl) % info_dict """
964 info_dict
.setdefault('epoch', int(time
.time())) # keep epoch consistent once set
966 info_dict
= self
._copy
_infodict
(info_dict
)
967 info_dict
['duration_string'] = ( # %(duration>%H-%M-%S)s is wrong if duration > 24hrs
968 formatSeconds(info_dict
['duration'], '-' if sanitize
else ':')
969 if info_dict
.get('duration', None) is not None
971 info_dict
['autonumber'] = self
.params
.get('autonumber_start', 1) - 1 + self
._num
_downloads
972 if info_dict
.get('resolution') is None:
973 info_dict
['resolution'] = self
.format_resolution(info_dict
, default
=None)
975 # For fields playlist_index, playlist_autonumber and autonumber convert all occurrences
976 # of %(field)s to %(field)0Nd for backward compatibility
977 field_size_compat_map
= {
978 'playlist_index': len(str(info_dict
.get('_last_playlist_index') or '')),
979 'playlist_autonumber': len(str(info_dict
.get('n_entries') or '')),
980 'autonumber': self
.params
.get('autonumber_size') or 5,
984 EXTERNAL_FORMAT_RE
= re
.compile(STR_FORMAT_RE_TMPL
.format('[^)]*', f
'[{STR_FORMAT_TYPES}ljqBU]'))
989 # Field is of the form key1.key2...
990 # where keys (except first) can be string, int or slice
991 FIELD_RE
= r
'\w*(?:\.(?:\w+|{num}|{num}?(?::{num}?){{1,2}}))*'.format(num
=r
'(?:-?\d+)')
992 MATH_FIELD_RE
= r
'''{field}|{num}'''.format(field
=FIELD_RE
, num
=r
'-?\d+(?:.\d+)?')
993 MATH_OPERATORS_RE
= r
'(?:%s)' % '|'.join(map(re
.escape
, MATH_FUNCTIONS
.keys()))
994 INTERNAL_FORMAT_RE
= re
.compile(r
'''(?x)
997 (?P<maths>(?:{math_op}{math_field})*)
998 (?:>(?P<strf_format>.+?))?
999 (?P<alternate>(?<!\\),[^|)]+)?
1000 (?:\|(?P<default>.*?))?
1001 $'''.format(field
=FIELD_RE
, math_op
=MATH_OPERATORS_RE
, math_field
=MATH_FIELD_RE
))
1003 def _traverse_infodict(k
):
1007 return traverse_obj(info_dict
, k
, is_user_input
=True, traverse_string
=True)
1009 def get_value(mdict
):
1011 value
= _traverse_infodict(mdict
['fields'])
1014 value
= float_or_none(value
)
1015 if value
is not None:
1018 offset_key
= mdict
['maths']
1020 value
= float_or_none(value
)
1024 MATH_FIELD_RE
if operator
else MATH_OPERATORS_RE
,
1025 offset_key
).group(0)
1026 offset_key
= offset_key
[len(item
):]
1027 if operator
is None:
1028 operator
= MATH_FUNCTIONS
[item
]
1030 item
, multiplier
= (item
[1:], -1) if item
[0] == '-' else (item
, 1)
1031 offset
= float_or_none(item
)
1033 offset
= float_or_none(_traverse_infodict(item
))
1035 value
= operator(value
, multiplier
* offset
)
1036 except (TypeError, ZeroDivisionError):
1039 # Datetime formatting
1040 if mdict
['strf_format']:
1041 value
= strftime_or_none(value
, mdict
['strf_format'].replace('\\,', ','))
1045 na
= self
.params
.get('outtmpl_na_placeholder', 'NA')
1047 def _dumpjson_default(obj
):
1048 if isinstance(obj
, (set, LazyList
)):
1050 raise TypeError(f
'Object of type {type(obj).__name__} is not JSON serializable')
1052 def create_key(outer_mobj
):
1053 if not outer_mobj
.group('has_key'):
1054 return outer_mobj
.group(0)
1055 key
= outer_mobj
.group('key')
1056 mobj
= re
.match(INTERNAL_FORMAT_RE
, key
)
1057 initial_field
= mobj
.group('fields').split('.')[-1] if mobj
else ''
1058 value
, default
= None, na
1060 mobj
= mobj
.groupdict()
1061 default
= mobj
['default'] if mobj
['default'] is not None else default
1062 value
= get_value(mobj
)
1063 if value
is None and mobj
['alternate']:
1064 mobj
= re
.match(INTERNAL_FORMAT_RE
, mobj
['alternate'][1:])
1068 fmt
= outer_mobj
.group('format')
1069 if fmt
== 's' and value
is not None and key
in field_size_compat_map
.keys():
1070 fmt
= '0{:d}d'.format(field_size_compat_map
[key
])
1072 value
= default
if value
is None else value
1074 str_fmt
= f
'{fmt[:-1]}s'
1075 if fmt
[-1] == 'l': # list
1076 delim
= '\n' if '#' in (outer_mobj
.group('conversion') or '') else ', '
1077 value
, fmt
= delim
.join(variadic(value
)), str_fmt
1078 elif fmt
[-1] == 'j': # json
1079 value
, fmt
= json
.dumps(value
, default
=_dumpjson_default
), str_fmt
1080 elif fmt
[-1] == 'q': # quoted
1081 value
, fmt
= compat_shlex_quote(str(value
)), str_fmt
1082 elif fmt
[-1] == 'B': # bytes
1083 value
= f
'%{str_fmt}'.encode('utf-8') % str(value
).encode('utf-8')
1084 value
, fmt
= value
.decode('utf-8', 'ignore'), 's'
1085 elif fmt
[-1] == 'U': # unicode normalized
1086 opts
= outer_mobj
.group('conversion') or ''
1087 value
, fmt
= unicodedata
.normalize(
1088 # "+" = compatibility equivalence, "#" = NFD
1089 'NF%s%s' % ('K' if '+' in opts
else '', 'D' if '#' in opts
else 'C'),
1091 elif fmt
[-1] == 'c':
1093 value
= str(value
)[0]
1096 elif fmt
[-1] not in 'rs': # numeric
1097 value
= float_or_none(value
)
1099 value
, fmt
= default
, 's'
1103 # If value is an object, sanitize might convert it to a string
1104 # So we convert it to repr first
1105 value
, fmt
= repr(value
), str_fmt
1106 if fmt
[-1] in 'csr':
1107 value
= sanitize(initial_field
, value
)
1109 key
= '%s\0%s' % (key
.replace('%', '%\0'), outer_mobj
.group('format'))
1110 TMPL_DICT
[key
] = value
1111 return '{prefix}%({key}){fmt}'.format(key
=key
, fmt
=fmt
, prefix
=outer_mobj
.group('prefix'))
1113 return EXTERNAL_FORMAT_RE
.sub(create_key
, outtmpl
), TMPL_DICT
1115 def evaluate_outtmpl(self
, outtmpl
, info_dict
, *args
, **kwargs
):
1116 outtmpl
, info_dict
= self
.prepare_outtmpl(outtmpl
, info_dict
, *args
, **kwargs
)
1117 return self
.escape_outtmpl(outtmpl
) % info_dict
1119 def _prepare_filename(self
, info_dict
, tmpl_type
='default'):
1121 sanitize
= lambda k
, v
: sanitize_filename(
1123 restricted
=self
.params
.get('restrictfilenames'),
1124 is_id
=(k
== 'id' or k
.endswith('_id')))
1125 outtmpl
= self
._outtmpl
_expandpath
(self
.outtmpl_dict
.get(tmpl_type
, self
.outtmpl_dict
['default']))
1126 filename
= self
.evaluate_outtmpl(outtmpl
, info_dict
, sanitize
)
1128 force_ext
= OUTTMPL_TYPES
.get(tmpl_type
)
1129 if filename
and force_ext
is not None:
1130 filename
= replace_extension(filename
, force_ext
, info_dict
.get('ext'))
1132 # https://github.com/blackjack4494/youtube-dlc/issues/85
1133 trim_file_name
= self
.params
.get('trim_file_name', False)
1135 fn_groups
= filename
.rsplit('.')
1138 if len(fn_groups
) > 2:
1139 sub_ext
= fn_groups
[-2]
1140 filename
= '.'.join(filter(None, [fn_groups
[0][:trim_file_name
], sub_ext
, ext
]))
1143 except ValueError as err
:
1144 self
.report_error('Error in output template: ' + str(err
) + ' (encoding: ' + repr(preferredencoding()) + ')')
1147 def prepare_filename(self
, info_dict
, dir_type
='', warn
=False):
1148 """Generate the output filename."""
1150 filename
= self
._prepare
_filename
(info_dict
, dir_type
or 'default')
1151 if not filename
and dir_type
not in ('', 'temp'):
1155 if not self
.params
.get('paths'):
1157 elif filename
== '-':
1158 self
.report_warning('--paths is ignored when an outputting to stdout', only_once
=True)
1159 elif os
.path
.isabs(filename
):
1160 self
.report_warning('--paths is ignored since an absolute path is given in output template', only_once
=True)
1161 if filename
== '-' or not filename
:
1164 return self
.get_output_path(dir_type
, filename
)
1166 def _match_entry(self
, info_dict
, incomplete
=False, silent
=False):
1167 """ Returns None if the file should be downloaded """
1169 video_title
= info_dict
.get('title', info_dict
.get('id', 'video'))
1172 if 'title' in info_dict
:
1173 # This can happen when we're just evaluating the playlist
1174 title
= info_dict
['title']
1175 matchtitle
= self
.params
.get('matchtitle', False)
1177 if not re
.search(matchtitle
, title
, re
.IGNORECASE
):
1178 return '"' + title
+ '" title did not match pattern "' + matchtitle
+ '"'
1179 rejecttitle
= self
.params
.get('rejecttitle', False)
1181 if re
.search(rejecttitle
, title
, re
.IGNORECASE
):
1182 return '"' + title
+ '" title matched reject pattern "' + rejecttitle
+ '"'
1183 date
= info_dict
.get('upload_date')
1184 if date
is not None:
1185 dateRange
= self
.params
.get('daterange', DateRange())
1186 if date
not in dateRange
:
1187 return '%s upload date is not in range %s' % (date_from_str(date
).isoformat(), dateRange
)
1188 view_count
= info_dict
.get('view_count')
1189 if view_count
is not None:
1190 min_views
= self
.params
.get('min_views')
1191 if min_views
is not None and view_count
< min_views
:
1192 return 'Skipping %s, because it has not reached minimum view count (%d/%d)' % (video_title
, view_count
, min_views
)
1193 max_views
= self
.params
.get('max_views')
1194 if max_views
is not None and view_count
> max_views
:
1195 return 'Skipping %s, because it has exceeded the maximum view count (%d/%d)' % (video_title
, view_count
, max_views
)
1196 if age_restricted(info_dict
.get('age_limit'), self
.params
.get('age_limit')):
1197 return 'Skipping "%s" because it is age restricted' % video_title
1199 match_filter
= self
.params
.get('match_filter')
1200 if match_filter
is not None:
1202 ret
= match_filter(info_dict
, incomplete
=incomplete
)
1204 # For backward compatibility
1205 ret
= None if incomplete
else match_filter(info_dict
)
1210 if self
.in_download_archive(info_dict
):
1211 reason
= '%s has already been recorded in the archive' % video_title
1212 break_opt
, break_err
= 'break_on_existing', ExistingVideoReached
1214 reason
= check_filter()
1215 break_opt
, break_err
= 'break_on_reject', RejectedVideoReached
1216 if reason
is not None:
1218 self
.to_screen('[download] ' + reason
)
1219 if self
.params
.get(break_opt
, False):
1224 def add_extra_info(info_dict
, extra_info
):
1225 '''Set the keys from extra_info in info dict if they are missing'''
1226 for key
, value
in extra_info
.items():
1227 info_dict
.setdefault(key
, value
)
1229 def extract_info(self
, url
, download
=True, ie_key
=None, extra_info
=None,
1230 process
=True, force_generic_extractor
=False):
1232 Return a list with a dictionary for each video extracted.
1235 url -- URL to extract
1238 download -- whether to download videos during extraction
1239 ie_key -- extractor key hint
1240 extra_info -- dictionary containing the extra values to add to each result
1241 process -- whether to resolve all unresolved references (URLs, playlist items),
1242 must be True for download to work.
1243 force_generic_extractor -- force using the generic extractor
1246 if extra_info
is None:
1249 if not ie_key
and force_generic_extractor
:
1253 ies
= {ie_key: self._get_info_extractor_class(ie_key)}
1257 for ie_key
, ie
in ies
.items():
1258 if not ie
.suitable(url
):
1261 if not ie
.working():
1262 self
.report_warning('The program functionality for this site has been marked as broken, '
1263 'and will probably not work.')
1265 temp_id
= ie
.get_temp_id(url
)
1266 if temp_id
is not None and self
.in_download_archive({'id': temp_id, 'ie_key': ie_key}
):
1267 self
.to_screen("[%s] %s: has already been recorded in archive" % (
1270 return self
.__extract
_info
(url
, self
.get_info_extractor(ie_key
), download
, extra_info
, process
)
1272 self
.report_error('no suitable InfoExtractor for URL %s' % url
)
1274 def __handle_extraction_exceptions(func
):
1275 @functools.wraps(func
)
1276 def wrapper(self
, *args
, **kwargs
):
1278 return func(self
, *args
, **kwargs
)
1279 except GeoRestrictedError
as e
:
1282 msg
+= '\nThis video is available in %s.' % ', '.join(
1283 map(ISO3166Utils
.short2full
, e
.countries
))
1284 msg
+= '\nYou might want to use a VPN or a proxy server (with --proxy) to workaround.'
1285 self
.report_error(msg
)
1286 except ExtractorError
as e
: # An error we somewhat expected
1287 self
.report_error(compat_str(e
), e
.format_traceback())
1288 except ThrottledDownload
:
1289 self
.to_stderr('\r')
1290 self
.report_warning('The download speed is below throttle limit. Re-extracting data')
1291 return wrapper(self
, *args
, **kwargs
)
1292 except (MaxDownloadsReached
, ExistingVideoReached
, RejectedVideoReached
, LazyList
.IndexError):
1294 except Exception as e
:
1295 if self
.params
.get('ignoreerrors'):
1296 self
.report_error(error_to_compat_str(e
), tb
=encode_compat_str(traceback
.format_exc()))
1301 @__handle_extraction_exceptions
1302 def __extract_info(self
, url
, ie
, download
, extra_info
, process
):
1303 ie_result
= ie
.extract(url
)
1304 if ie_result
is None: # Finished already (backwards compatibility; listformats and friends should be moved here)
1306 if isinstance(ie_result
, list):
1307 # Backwards compatibility: old IE result format
1309 '_type': 'compat_list',
1310 'entries': ie_result
,
1312 if extra_info
.get('original_url'):
1313 ie_result
.setdefault('original_url', extra_info
['original_url'])
1314 self
.add_default_extra_info(ie_result
, ie
, url
)
1316 return self
.process_ie_result(ie_result
, download
, extra_info
)
1320 def add_default_extra_info(self
, ie_result
, ie
, url
):
1322 self
.add_extra_info(ie_result
, {
1324 'original_url': url
,
1325 'webpage_url_basename': url_basename(url
),
1328 self
.add_extra_info(ie_result
, {
1329 'extractor': ie
.IE_NAME
,
1330 'extractor_key': ie
.ie_key(),
1333 def process_ie_result(self
, ie_result
, download
=True, extra_info
=None):
1335 Take the result of the ie(may be modified) and resolve all unresolved
1336 references (URLs, playlist items).
1338 It will also download the videos if 'download'.
1339 Returns the resolved ie_result.
1341 if extra_info
is None:
1343 result_type
= ie_result
.get('_type', 'video')
1345 if result_type
in ('url', 'url_transparent'):
1346 ie_result
['url'] = sanitize_url(ie_result
['url'])
1347 if ie_result
.get('original_url'):
1348 extra_info
.setdefault('original_url', ie_result
['original_url'])
1350 extract_flat
= self
.params
.get('extract_flat', False)
1351 if ((extract_flat
== 'in_playlist' and 'playlist' in extra_info
)
1352 or extract_flat
is True):
1353 info_copy
= ie_result
.copy()
1354 ie
= try_get(ie_result
.get('ie_key'), self
.get_info_extractor
)
1355 if ie
and not ie_result
.get('id'):
1356 info_copy
['id'] = ie
.get_temp_id(ie_result
['url'])
1357 self
.add_default_extra_info(info_copy
, ie
, ie_result
['url'])
1358 self
.add_extra_info(info_copy
, extra_info
)
1359 self
.__forced
_printings
(info_copy
, self
.prepare_filename(info_copy
), incomplete
=True)
1360 if self
.params
.get('force_write_download_archive', False):
1361 self
.record_download_archive(info_copy
)
1364 if result_type
== 'video':
1365 self
.add_extra_info(ie_result
, extra_info
)
1366 ie_result
= self
.process_video_result(ie_result
, download
=download
)
1367 additional_urls
= (ie_result
or {}).get('additional_urls')
1369 # TODO: Improve MetadataParserPP to allow setting a list
1370 if isinstance(additional_urls
, compat_str
):
1371 additional_urls
= [additional_urls
]
1373 '[info] %s: %d additional URL(s) requested' % (ie_result
['id'], len(additional_urls
)))
1374 self
.write_debug('Additional URLs: "%s"' % '", "'.join(additional_urls
))
1375 ie_result
['additional_entries'] = [
1377 url
, download
, extra_info
,
1378 force_generic_extractor
=self
.params
.get('force_generic_extractor'))
1379 for url
in additional_urls
1382 elif result_type
== 'url':
1383 # We have to add extra_info to the results because it may be
1384 # contained in a playlist
1385 return self
.extract_info(
1386 ie_result
['url'], download
,
1387 ie_key
=ie_result
.get('ie_key'),
1388 extra_info
=extra_info
)
1389 elif result_type
== 'url_transparent':
1390 # Use the information from the embedding page
1391 info
= self
.extract_info(
1392 ie_result
['url'], ie_key
=ie_result
.get('ie_key'),
1393 extra_info
=extra_info
, download
=False, process
=False)
1395 # extract_info may return None when ignoreerrors is enabled and
1396 # extraction failed with an error, don't crash and return early
1401 force_properties
= dict(
1402 (k
, v
) for k
, v
in ie_result
.items() if v
is not None)
1403 for f
in ('_type', 'url', 'id', 'extractor', 'extractor_key', 'ie_key'):
1404 if f
in force_properties
:
1405 del force_properties
[f
]
1406 new_result
= info
.copy()
1407 new_result
.update(force_properties
)
1409 # Extracted info may not be a video result (i.e.
1410 # info.get('_type', 'video') != video) but rather an url or
1411 # url_transparent. In such cases outer metadata (from ie_result)
1412 # should be propagated to inner one (info). For this to happen
1413 # _type of info should be overridden with url_transparent. This
1414 # fixes issue from https://github.com/ytdl-org/youtube-dl/pull/11163.
1415 if new_result
.get('_type') == 'url':
1416 new_result
['_type'] = 'url_transparent'
1418 return self
.process_ie_result(
1419 new_result
, download
=download
, extra_info
=extra_info
)
1420 elif result_type
in ('playlist', 'multi_video'):
1421 # Protect from infinite recursion due to recursively nested playlists
1422 # (see https://github.com/ytdl-org/youtube-dl/issues/27833)
1423 webpage_url
= ie_result
['webpage_url']
1424 if webpage_url
in self
._playlist
_urls
:
1426 '[download] Skipping already downloaded playlist: %s'
1427 % ie_result
.get('title') or ie_result
.get('id'))
1430 self
._playlist
_level
+= 1
1431 self
._playlist
_urls
.add(webpage_url
)
1432 self
._sanitize
_thumbnails
(ie_result
)
1434 return self
.__process
_playlist
(ie_result
, download
)
1436 self
._playlist
_level
-= 1
1437 if not self
._playlist
_level
:
1438 self
._playlist
_urls
.clear()
1439 elif result_type
== 'compat_list':
1440 self
.report_warning(
1441 'Extractor %s returned a compat_list result. '
1442 'It needs to be updated.' % ie_result
.get('extractor'))
1445 self
.add_extra_info(r
, {
1446 'extractor': ie_result
['extractor'],
1447 'webpage_url': ie_result
['webpage_url'],
1448 'webpage_url_basename': url_basename(ie_result
['webpage_url']),
1449 'extractor_key': ie_result
['extractor_key'],
1452 ie_result
['entries'] = [
1453 self
.process_ie_result(_fixup(r
), download
, extra_info
)
1454 for r
in ie_result
['entries']
1458 raise Exception('Invalid result type: %s' % result_type
)
1460 def _ensure_dir_exists(self
, path
):
1461 return make_dir(path
, self
.report_error
)
1463 def __process_playlist(self
, ie_result
, download
):
1464 # We process each entry in the playlist
1465 playlist
= ie_result
.get('title') or ie_result
.get('id')
1466 self
.to_screen('[download] Downloading playlist: %s' % playlist
)
1468 if 'entries' not in ie_result
:
1469 raise EntryNotInPlaylist()
1470 incomplete_entries
= bool(ie_result
.get('requested_entries'))
1471 if incomplete_entries
:
1472 def fill_missing_entries(entries
, indexes
):
1473 ret
= [None] * max(*indexes
)
1474 for i
, entry
in zip(indexes
, entries
):
1477 ie_result
['entries'] = fill_missing_entries(ie_result
['entries'], ie_result
['requested_entries'])
1479 playlist_results
= []
1481 playliststart
= self
.params
.get('playliststart', 1)
1482 playlistend
= self
.params
.get('playlistend')
1483 # For backwards compatibility, interpret -1 as whole list
1484 if playlistend
== -1:
1487 playlistitems_str
= self
.params
.get('playlist_items')
1488 playlistitems
= None
1489 if playlistitems_str
is not None:
1490 def iter_playlistitems(format
):
1491 for string_segment
in format
.split(','):
1492 if '-' in string_segment
:
1493 start
, end
= string_segment
.split('-')
1494 for item
in range(int(start
), int(end
) + 1):
1497 yield int(string_segment
)
1498 playlistitems
= orderedSet(iter_playlistitems(playlistitems_str
))
1500 ie_entries
= ie_result
['entries']
1502 'Downloading %d videos' if not isinstance(ie_entries
, list)
1503 else 'Collected %d videos; downloading %%d of them' % len(ie_entries
))
1505 if isinstance(ie_entries
, list):
1507 return ie_entries
[i
- 1]
1509 if not isinstance(ie_entries
, PagedList
):
1510 ie_entries
= LazyList(ie_entries
)
1513 return YoutubeDL
.__handle
_extraction
_exceptions
(
1514 lambda self
, i
: ie_entries
[i
- 1]
1518 items
= playlistitems
if playlistitems
is not None else itertools
.count(playliststart
)
1522 if playlistitems
is None and playlistend
is not None and playlistend
< i
:
1526 entry
= get_entry(i
)
1528 raise EntryNotInPlaylist()
1529 except (IndexError, EntryNotInPlaylist
):
1530 if incomplete_entries
:
1531 raise EntryNotInPlaylist()
1532 elif not playlistitems
:
1534 entries
.append(entry
)
1536 if entry
is not None:
1537 self
._match
_entry
(entry
, incomplete
=True, silent
=True)
1538 except (ExistingVideoReached
, RejectedVideoReached
):
1540 ie_result
['entries'] = entries
1542 # Save playlist_index before re-ordering
1544 ((playlistitems
[i
- 1] if playlistitems
else i
+ playliststart
- 1), entry
)
1545 for i
, entry
in enumerate(entries
, 1)
1546 if entry
is not None]
1547 n_entries
= len(entries
)
1549 if not playlistitems
and (playliststart
or playlistend
):
1550 playlistitems
= list(range(playliststart
, playliststart
+ n_entries
))
1551 ie_result
['requested_entries'] = playlistitems
1553 if self
.params
.get('allow_playlist_files', True):
1555 'playlist': playlist
,
1556 'playlist_id': ie_result
.get('id'),
1557 'playlist_title': ie_result
.get('title'),
1558 'playlist_uploader': ie_result
.get('uploader'),
1559 'playlist_uploader_id': ie_result
.get('uploader_id'),
1560 'playlist_index': 0,
1562 ie_copy
.update(dict(ie_result
))
1564 if self
._write
_info
_json
('playlist', ie_result
,
1565 self
.prepare_filename(ie_copy
, 'pl_infojson')) is None:
1567 if self
._write
_description
('playlist', ie_result
,
1568 self
.prepare_filename(ie_copy
, 'pl_description')) is None:
1570 # TODO: This should be passed to ThumbnailsConvertor if necessary
1571 self
._write
_thumbnails
('playlist', ie_copy
, self
.prepare_filename(ie_copy
, 'pl_thumbnail'))
1573 if self
.params
.get('playlistreverse', False):
1574 entries
= entries
[::-1]
1575 if self
.params
.get('playlistrandom', False):
1576 random
.shuffle(entries
)
1578 x_forwarded_for
= ie_result
.get('__x_forwarded_for_ip')
1580 self
.to_screen('[%s] playlist %s: %s' % (ie_result
['extractor'], playlist
, msg
% n_entries
))
1582 max_failures
= self
.params
.get('skip_playlist_after_errors') or float('inf')
1583 for i
, entry_tuple
in enumerate(entries
, 1):
1584 playlist_index
, entry
= entry_tuple
1585 if 'playlist-index' in self
.params
.get('compat_opts', []):
1586 playlist_index
= playlistitems
[i
- 1] if playlistitems
else i
+ playliststart
- 1
1587 self
.to_screen('[download] Downloading video %s of %s' % (i
, n_entries
))
1588 # This __x_forwarded_for_ip thing is a bit ugly but requires
1591 entry
['__x_forwarded_for_ip'] = x_forwarded_for
1593 'n_entries': n_entries
,
1594 '_last_playlist_index': max(playlistitems
) if playlistitems
else (playlistend
or n_entries
),
1595 'playlist_index': playlist_index
,
1596 'playlist_autonumber': i
,
1597 'playlist': playlist
,
1598 'playlist_id': ie_result
.get('id'),
1599 'playlist_title': ie_result
.get('title'),
1600 'playlist_uploader': ie_result
.get('uploader'),
1601 'playlist_uploader_id': ie_result
.get('uploader_id'),
1602 'extractor': ie_result
['extractor'],
1603 'webpage_url': ie_result
['webpage_url'],
1604 'webpage_url_basename': url_basename(ie_result
['webpage_url']),
1605 'extractor_key': ie_result
['extractor_key'],
1608 if self
._match
_entry
(entry
, incomplete
=True) is not None:
1611 entry_result
= self
.__process
_iterable
_entry
(entry
, download
, extra
)
1612 if not entry_result
:
1614 if failures
>= max_failures
:
1616 'Skipping the remaining entries in playlist "%s" since %d items failed extraction' % (playlist
, failures
))
1618 # TODO: skip failed (empty) entries?
1619 playlist_results
.append(entry_result
)
1620 ie_result
['entries'] = playlist_results
1621 self
.to_screen('[download] Finished downloading playlist: %s' % playlist
)
1624 @__handle_extraction_exceptions
1625 def __process_iterable_entry(self
, entry
, download
, extra_info
):
1626 return self
.process_ie_result(
1627 entry
, download
=download
, extra_info
=extra_info
)
1629 def _build_format_filter(self
, filter_spec
):
1630 " Returns a function to filter the formats according to the filter_spec "
1640 operator_rex
= re
.compile(r
'''(?x)\s*
1641 (?P<key>width|height|tbr|abr|vbr|asr|filesize|filesize_approx|fps)\s*
1642 (?P<op>%s)(?P<none_inclusive>\s*\?)?\s*
1643 (?P<value>[0-9.]+(?:[kKmMgGtTpPeEzZyY]i?[Bb]?)?)\s*
1644 ''' % '|'.join(map(re
.escape
, OPERATORS
.keys())))
1645 m
= operator_rex
.fullmatch(filter_spec
)
1648 comparison_value
= int(m
.group('value'))
1650 comparison_value
= parse_filesize(m
.group('value'))
1651 if comparison_value
is None:
1652 comparison_value
= parse_filesize(m
.group('value') + 'B')
1653 if comparison_value
is None:
1655 'Invalid value %r in format specification %r' % (
1656 m
.group('value'), filter_spec
))
1657 op
= OPERATORS
[m
.group('op')]
1662 '^=': lambda attr
, value
: attr
.startswith(value
),
1663 '$=': lambda attr
, value
: attr
.endswith(value
),
1664 '*=': lambda attr
, value
: value
in attr
,
1666 str_operator_rex
= re
.compile(r
'''(?x)\s*
1667 (?P<key>[a-zA-Z0-9._-]+)\s*
1668 (?P<negation>!\s*)?(?P<op>%s)(?P<none_inclusive>\s*\?)?\s*
1669 (?P<value>[a-zA-Z0-9._-]+)\s*
1670 ''' % '|'.join(map(re
.escape
, STR_OPERATORS
.keys())))
1671 m
= str_operator_rex
.fullmatch(filter_spec
)
1673 comparison_value
= m
.group('value')
1674 str_op
= STR_OPERATORS
[m
.group('op')]
1675 if m
.group('negation'):
1676 op
= lambda attr
, value
: not str_op(attr
, value
)
1681 raise SyntaxError('Invalid filter specification %r' % filter_spec
)
1684 actual_value
= f
.get(m
.group('key'))
1685 if actual_value
is None:
1686 return m
.group('none_inclusive')
1687 return op(actual_value
, comparison_value
)
1690 def _default_format_spec(self
, info_dict
, download
=True):
1693 merger
= FFmpegMergerPP(self
)
1694 return merger
.available
and merger
.can_merge()
1697 not self
.params
.get('simulate')
1701 or info_dict
.get('is_live', False)
1702 or self
.outtmpl_dict
['default'] == '-'))
1705 or self
.params
.get('allow_multiple_audio_streams', False)
1706 or 'format-spec' in self
.params
.get('compat_opts', []))
1709 'best/bestvideo+bestaudio' if prefer_best
1710 else 'bestvideo*+bestaudio/best' if not compat
1711 else 'bestvideo+bestaudio/best')
1713 def build_format_selector(self
, format_spec
):
1714 def syntax_error(note
, start
):
1716 'Invalid format specification: '
1717 '{0}\n\t{1}\n\t{2}^'.format(note
, format_spec
, ' ' * start
[1]))
1718 return SyntaxError(message
)
1720 PICKFIRST
= 'PICKFIRST'
1724 FormatSelector
= collections
.namedtuple('FormatSelector', ['type', 'selector', 'filters'])
1726 allow_multiple_streams
= {'audio': self
.params
.get('allow_multiple_audio_streams', False),
1727 'video': self
.params
.get('allow_multiple_video_streams', False)}
1729 check_formats
= self
.params
.get('check_formats')
1731 def _parse_filter(tokens
):
1733 for type, string
, start
, _
, _
in tokens
:
1734 if type == tokenize
.OP
and string
== ']':
1735 return ''.join(filter_parts
)
1737 filter_parts
.append(string
)
1739 def _remove_unused_ops(tokens
):
1740 # Remove operators that we don't use and join them with the surrounding strings
1741 # for example: 'mp4' '-' 'baseline' '-' '16x9' is converted to 'mp4-baseline-16x9'
1742 ALLOWED_OPS
= ('/', '+', ',', '(', ')')
1743 last_string
, last_start
, last_end
, last_line
= None, None, None, None
1744 for type, string
, start
, end
, line
in tokens
:
1745 if type == tokenize
.OP
and string
== '[':
1747 yield tokenize
.NAME
, last_string
, last_start
, last_end
, last_line
1749 yield type, string
, start
, end
, line
1750 # everything inside brackets will be handled by _parse_filter
1751 for type, string
, start
, end
, line
in tokens
:
1752 yield type, string
, start
, end
, line
1753 if type == tokenize
.OP
and string
== ']':
1755 elif type == tokenize
.OP
and string
in ALLOWED_OPS
:
1757 yield tokenize
.NAME
, last_string
, last_start
, last_end
, last_line
1759 yield type, string
, start
, end
, line
1760 elif type in [tokenize
.NAME
, tokenize
.NUMBER
, tokenize
.OP
]:
1762 last_string
= string
1766 last_string
+= string
1768 yield tokenize
.NAME
, last_string
, last_start
, last_end
, last_line
1770 def _parse_format_selection(tokens
, inside_merge
=False, inside_choice
=False, inside_group
=False):
1772 current_selector
= None
1773 for type, string
, start
, _
, _
in tokens
:
1774 # ENCODING is only defined in python 3.x
1775 if type == getattr(tokenize
, 'ENCODING', None):
1777 elif type in [tokenize
.NAME
, tokenize
.NUMBER
]:
1778 current_selector
= FormatSelector(SINGLE
, string
, [])
1779 elif type == tokenize
.OP
:
1781 if not inside_group
:
1782 # ')' will be handled by the parentheses group
1783 tokens
.restore_last_token()
1785 elif inside_merge
and string
in ['/', ',']:
1786 tokens
.restore_last_token()
1788 elif inside_choice
and string
== ',':
1789 tokens
.restore_last_token()
1792 if not current_selector
:
1793 raise syntax_error('"," must follow a format selector', start
)
1794 selectors
.append(current_selector
)
1795 current_selector
= None
1797 if not current_selector
:
1798 raise syntax_error('"/" must follow a format selector', start
)
1799 first_choice
= current_selector
1800 second_choice
= _parse_format_selection(tokens
, inside_choice
=True)
1801 current_selector
= FormatSelector(PICKFIRST
, (first_choice
, second_choice
), [])
1803 if not current_selector
:
1804 current_selector
= FormatSelector(SINGLE
, 'best', [])
1805 format_filter
= _parse_filter(tokens
)
1806 current_selector
.filters
.append(format_filter
)
1808 if current_selector
:
1809 raise syntax_error('Unexpected "("', start
)
1810 group
= _parse_format_selection(tokens
, inside_group
=True)
1811 current_selector
= FormatSelector(GROUP
, group
, [])
1813 if not current_selector
:
1814 raise syntax_error('Unexpected "+"', start
)
1815 selector_1
= current_selector
1816 selector_2
= _parse_format_selection(tokens
, inside_merge
=True)
1818 raise syntax_error('Expected a selector', start
)
1819 current_selector
= FormatSelector(MERGE
, (selector_1
, selector_2
), [])
1821 raise syntax_error('Operator not recognized: "{0}"'.format(string
), start
)
1822 elif type == tokenize
.ENDMARKER
:
1824 if current_selector
:
1825 selectors
.append(current_selector
)
1828 def _merge(formats_pair
):
1829 format_1
, format_2
= formats_pair
1832 formats_info
.extend(format_1
.get('requested_formats', (format_1
,)))
1833 formats_info
.extend(format_2
.get('requested_formats', (format_2
,)))
1835 if not allow_multiple_streams
['video'] or not allow_multiple_streams
['audio']:
1836 get_no_more
= {'video': False, 'audio': False}
1837 for (i
, fmt_info
) in enumerate(formats_info
):
1838 if fmt_info
.get('acodec') == fmt_info
.get('vcodec') == 'none':
1841 for aud_vid
in ['audio', 'video']:
1842 if not allow_multiple_streams
[aud_vid
] and fmt_info
.get(aud_vid
[0] + 'codec') != 'none':
1843 if get_no_more
[aud_vid
]:
1846 get_no_more
[aud_vid
] = True
1848 if len(formats_info
) == 1:
1849 return formats_info
[0]
1851 video_fmts
= [fmt_info
for fmt_info
in formats_info
if fmt_info
.get('vcodec') != 'none']
1852 audio_fmts
= [fmt_info
for fmt_info
in formats_info
if fmt_info
.get('acodec') != 'none']
1854 the_only_video
= video_fmts
[0] if len(video_fmts
) == 1 else None
1855 the_only_audio
= audio_fmts
[0] if len(audio_fmts
) == 1 else None
1857 output_ext
= self
.params
.get('merge_output_format')
1860 output_ext
= the_only_video
['ext']
1861 elif the_only_audio
and not video_fmts
:
1862 output_ext
= the_only_audio
['ext']
1866 filtered
= lambda *keys
: filter(None, (traverse_obj(fmt
, *keys
) for fmt
in formats_info
))
1869 'requested_formats': formats_info
,
1870 'format': '+'.join(filtered('format')),
1871 'format_id': '+'.join(filtered('format_id')),
1873 'protocol': '+'.join(map(determine_protocol
, formats_info
)),
1874 'language': '+'.join(orderedSet(filtered('language'))),
1875 'format_note': '+'.join(orderedSet(filtered('format_note'))),
1876 'filesize_approx': sum(filtered('filesize', 'filesize_approx')),
1877 'tbr': sum(filtered('tbr', 'vbr', 'abr')),
1882 'width': the_only_video
.get('width'),
1883 'height': the_only_video
.get('height'),
1884 'resolution': the_only_video
.get('resolution') or self
.format_resolution(the_only_video
),
1885 'fps': the_only_video
.get('fps'),
1886 'vcodec': the_only_video
.get('vcodec'),
1887 'vbr': the_only_video
.get('vbr'),
1888 'stretched_ratio': the_only_video
.get('stretched_ratio'),
1893 'acodec': the_only_audio
.get('acodec'),
1894 'abr': the_only_audio
.get('abr'),
1895 'asr': the_only_audio
.get('asr'),
1900 def _check_formats(formats
):
1901 if not check_formats
:
1905 self
.to_screen('[info] Testing format %s' % f
['format_id'])
1906 temp_file
= tempfile
.NamedTemporaryFile(
1907 suffix
='.tmp', delete
=False,
1908 dir=self
.get_output_path('temp') or None)
1911 success
, _
= self
.dl(temp_file
.name
, f
, test
=True)
1912 except (DownloadError
, IOError, OSError, ValueError) + network_exceptions
:
1915 if os
.path
.exists(temp_file
.name
):
1917 os
.remove(temp_file
.name
)
1919 self
.report_warning('Unable to delete temporary file "%s"' % temp_file
.name
)
1923 self
.to_screen('[info] Unable to download format %s. Skipping...' % f
['format_id'])
1925 def _build_selector_function(selector
):
1926 if isinstance(selector
, list): # ,
1927 fs
= [_build_selector_function(s
) for s
in selector
]
1929 def selector_function(ctx
):
1932 return selector_function
1934 elif selector
.type == GROUP
: # ()
1935 selector_function
= _build_selector_function(selector
.selector
)
1937 elif selector
.type == PICKFIRST
: # /
1938 fs
= [_build_selector_function(s
) for s
in selector
.selector
]
1940 def selector_function(ctx
):
1942 picked_formats
= list(f(ctx
))
1944 return picked_formats
1947 elif selector
.type == MERGE
: # +
1948 selector_1
, selector_2
= map(_build_selector_function
, selector
.selector
)
1950 def selector_function(ctx
):
1951 for pair
in itertools
.product(
1952 selector_1(copy
.deepcopy(ctx
)), selector_2(copy
.deepcopy(ctx
))):
1955 elif selector
.type == SINGLE
: # atom
1956 format_spec
= selector
.selector
or 'best'
1958 # TODO: Add allvideo, allaudio etc by generalizing the code with best/worst selector
1959 if format_spec
== 'all':
1960 def selector_function(ctx
):
1961 yield from _check_formats(ctx
['formats'])
1962 elif format_spec
== 'mergeall':
1963 def selector_function(ctx
):
1964 formats
= list(_check_formats(ctx
['formats']))
1967 merged_format
= formats
[-1]
1968 for f
in formats
[-2::-1]:
1969 merged_format
= _merge((merged_format
, f
))
1973 format_fallback
, format_reverse
, format_idx
= False, True, 1
1975 r
'(?P<bw>best|worst|b|w)(?P<type>video|audio|v|a)?(?P<mod>\*)?(?:\.(?P<n>[1-9]\d*))?$',
1977 if mobj
is not None:
1978 format_idx
= int_or_none(mobj
.group('n'), default
=1)
1979 format_reverse
= mobj
.group('bw')[0] == 'b'
1980 format_type
= (mobj
.group('type') or [None])[0]
1981 not_format_type
= {'v': 'a', 'a': 'v'}
.get(format_type
)
1982 format_modified
= mobj
.group('mod') is not None
1984 format_fallback
= not format_type
and not format_modified
# for b, w
1986 (lambda f
: f
.get('%scodec' % format_type
) != 'none')
1987 if format_type
and format_modified
# bv*, ba*, wv*, wa*
1988 else (lambda f
: f
.get('%scodec' % not_format_type
) == 'none')
1989 if format_type
# bv, ba, wv, wa
1990 else (lambda f
: f
.get('vcodec') != 'none' and f
.get('acodec') != 'none')
1991 if not format_modified
# b, w
1992 else lambda f
: True) # b*, w*
1993 filter_f
= lambda f
: _filter_f(f
) and (
1994 f
.get('vcodec') != 'none' or f
.get('acodec') != 'none')
1996 if format_spec
in self
._format
_selection
_exts
['audio']:
1997 filter_f
= lambda f
: f
.get('ext') == format_spec
and f
.get('acodec') != 'none'
1998 elif format_spec
in self
._format
_selection
_exts
['video']:
1999 filter_f
= lambda f
: f
.get('ext') == format_spec
and f
.get('acodec') != 'none' and f
.get('vcodec') != 'none'
2000 elif format_spec
in self
._format
_selection
_exts
['storyboards']:
2001 filter_f
= lambda f
: f
.get('ext') == format_spec
and f
.get('acodec') == 'none' and f
.get('vcodec') == 'none'
2003 filter_f
= lambda f
: f
.get('format_id') == format_spec
# id
2005 def selector_function(ctx
):
2006 formats
= list(ctx
['formats'])
2007 matches
= list(filter(filter_f
, formats
)) if filter_f
is not None else formats
2008 if format_fallback
and ctx
['incomplete_formats'] and not matches
:
2009 # for extractors with incomplete formats (audio only (soundcloud)
2010 # or video only (imgur)) best/worst will fallback to
2011 # best/worst {video,audio}-only format
2013 matches
= LazyList(_check_formats(matches
[::-1 if format_reverse
else 1]))
2015 yield matches
[format_idx
- 1]
2019 filters
= [self
._build
_format
_filter
(f
) for f
in selector
.filters
]
2021 def final_selector(ctx
):
2022 ctx_copy
= copy
.deepcopy(ctx
)
2023 for _filter
in filters
:
2024 ctx_copy
['formats'] = list(filter(_filter
, ctx_copy
['formats']))
2025 return selector_function(ctx_copy
)
2026 return final_selector
2028 stream
= io
.BytesIO(format_spec
.encode('utf-8'))
2030 tokens
= list(_remove_unused_ops(compat_tokenize_tokenize(stream
.readline
)))
2031 except tokenize
.TokenError
:
2032 raise syntax_error('Missing closing/opening brackets or parenthesis', (0, len(format_spec
)))
2034 class TokenIterator(object):
2035 def __init__(self
, tokens
):
2036 self
.tokens
= tokens
2043 if self
.counter
>= len(self
.tokens
):
2044 raise StopIteration()
2045 value
= self
.tokens
[self
.counter
]
2051 def restore_last_token(self
):
2054 parsed_selector
= _parse_format_selection(iter(TokenIterator(tokens
)))
2055 return _build_selector_function(parsed_selector
)
2057 def _calc_headers(self
, info_dict
):
2058 res
= std_headers
.copy()
2060 add_headers
= info_dict
.get('http_headers')
2062 res
.update(add_headers
)
2064 cookies
= self
._calc
_cookies
(info_dict
)
2066 res
['Cookie'] = cookies
2068 if 'X-Forwarded-For' not in res
:
2069 x_forwarded_for_ip
= info_dict
.get('__x_forwarded_for_ip')
2070 if x_forwarded_for_ip
:
2071 res
['X-Forwarded-For'] = x_forwarded_for_ip
2075 def _calc_cookies(self
, info_dict
):
2076 pr
= sanitized_Request(info_dict
['url'])
2077 self
.cookiejar
.add_cookie_header(pr
)
2078 return pr
.get_header('Cookie')
2080 def _sanitize_thumbnails(self
, info_dict
):
2081 thumbnails
= info_dict
.get('thumbnails')
2082 if thumbnails
is None:
2083 thumbnail
= info_dict
.get('thumbnail')
2085 info_dict
['thumbnails'] = thumbnails
= [{'url': thumbnail}
]
2087 thumbnails
.sort(key
=lambda t
: (
2088 t
.get('preference') if t
.get('preference') is not None else -1,
2089 t
.get('width') if t
.get('width') is not None else -1,
2090 t
.get('height') if t
.get('height') is not None else -1,
2091 t
.get('id') if t
.get('id') is not None else '',
2094 def thumbnail_tester():
2095 def test_thumbnail(t
):
2096 self
.to_screen(f
'[info] Testing thumbnail {t["id"]}')
2098 self
.urlopen(HEADRequest(t
['url']))
2099 except network_exceptions
as err
:
2100 self
.to_screen(f
'[info] Unable to connect to thumbnail {t["id"]} URL {t["url"]!r} - {err}. Skipping...')
2103 return test_thumbnail
2105 for i
, t
in enumerate(thumbnails
):
2106 if t
.get('id') is None:
2108 if t
.get('width') and t
.get('height'):
2109 t
['resolution'] = '%dx%d' % (t
['width'], t
['height'])
2110 t
['url'] = sanitize_url(t
['url'])
2112 if self
.params
.get('check_formats'):
2113 info_dict
['thumbnails'] = LazyList(filter(thumbnail_tester(), thumbnails
[::-1])).reverse()
2115 info_dict
['thumbnails'] = thumbnails
2117 def process_video_result(self
, info_dict
, download
=True):
2118 assert info_dict
.get('_type', 'video') == 'video'
2120 if 'id' not in info_dict
:
2121 raise ExtractorError('Missing "id" field in extractor result')
2122 if 'title' not in info_dict
:
2123 raise ExtractorError('Missing "title" field in extractor result',
2124 video_id
=info_dict
['id'], ie
=info_dict
['extractor'])
2126 def report_force_conversion(field
, field_not
, conversion
):
2127 self
.report_warning(
2128 '"%s" field is not %s - forcing %s conversion, there is an error in extractor'
2129 % (field
, field_not
, conversion
))
2131 def sanitize_string_field(info
, string_field
):
2132 field
= info
.get(string_field
)
2133 if field
is None or isinstance(field
, compat_str
):
2135 report_force_conversion(string_field
, 'a string', 'string')
2136 info
[string_field
] = compat_str(field
)
2138 def sanitize_numeric_fields(info
):
2139 for numeric_field
in self
._NUMERIC
_FIELDS
:
2140 field
= info
.get(numeric_field
)
2141 if field
is None or isinstance(field
, compat_numeric_types
):
2143 report_force_conversion(numeric_field
, 'numeric', 'int')
2144 info
[numeric_field
] = int_or_none(field
)
2146 sanitize_string_field(info_dict
, 'id')
2147 sanitize_numeric_fields(info_dict
)
2149 if 'playlist' not in info_dict
:
2150 # It isn't part of a playlist
2151 info_dict
['playlist'] = None
2152 info_dict
['playlist_index'] = None
2154 self
._sanitize
_thumbnails
(info_dict
)
2156 thumbnail
= info_dict
.get('thumbnail')
2157 thumbnails
= info_dict
.get('thumbnails')
2159 info_dict
['thumbnail'] = sanitize_url(thumbnail
)
2161 info_dict
['thumbnail'] = thumbnails
[-1]['url']
2163 if info_dict
.get('display_id') is None and 'id' in info_dict
:
2164 info_dict
['display_id'] = info_dict
['id']
2166 if info_dict
.get('duration') is not None:
2167 info_dict
['duration_string'] = formatSeconds(info_dict
['duration'])
2169 for ts_key
, date_key
in (
2170 ('timestamp', 'upload_date'),
2171 ('release_timestamp', 'release_date'),
2173 if info_dict
.get(date_key
) is None and info_dict
.get(ts_key
) is not None:
2174 # Working around out-of-range timestamp values (e.g. negative ones on Windows,
2175 # see http://bugs.python.org/issue1646728)
2177 upload_date
= datetime
.datetime
.utcfromtimestamp(info_dict
[ts_key
])
2178 info_dict
[date_key
] = upload_date
.strftime('%Y%m%d')
2179 except (ValueError, OverflowError, OSError):
2182 live_keys
= ('is_live', 'was_live')
2183 live_status
= info_dict
.get('live_status')
2184 if live_status
is None:
2185 for key
in live_keys
:
2186 if info_dict
.get(key
) is False:
2188 if info_dict
.get(key
):
2191 if all(info_dict
.get(key
) is False for key
in live_keys
):
2192 live_status
= 'not_live'
2194 info_dict
['live_status'] = live_status
2195 for key
in live_keys
:
2196 if info_dict
.get(key
) is None:
2197 info_dict
[key
] = (live_status
== key
)
2199 # Auto generate title fields corresponding to the *_number fields when missing
2200 # in order to always have clean titles. This is very common for TV series.
2201 for field
in ('chapter', 'season', 'episode'):
2202 if info_dict
.get('%s_number' % field
) is not None and not info_dict
.get(field
):
2203 info_dict
[field
] = '%s %d' % (field
.capitalize(), info_dict
['%s_number' % field
])
2205 for cc_kind
in ('subtitles', 'automatic_captions'):
2206 cc
= info_dict
.get(cc_kind
)
2208 for _
, subtitle
in cc
.items():
2209 for subtitle_format
in subtitle
:
2210 if subtitle_format
.get('url'):
2211 subtitle_format
['url'] = sanitize_url(subtitle_format
['url'])
2212 if subtitle_format
.get('ext') is None:
2213 subtitle_format
['ext'] = determine_ext(subtitle_format
['url']).lower()
2215 automatic_captions
= info_dict
.get('automatic_captions')
2216 subtitles
= info_dict
.get('subtitles')
2218 info_dict
['requested_subtitles'] = self
.process_subtitles(
2219 info_dict
['id'], subtitles
, automatic_captions
)
2221 # We now pick which formats have to be downloaded
2222 if info_dict
.get('formats') is None:
2223 # There's only one format available
2224 formats
= [info_dict
]
2226 formats
= info_dict
['formats']
2228 info_dict
['__has_drm'] = any(f
.get('has_drm') for f
in formats
)
2229 if not self
.params
.get('allow_unplayable_formats'):
2230 formats
= [f
for f
in formats
if not f
.get('has_drm')]
2233 self
.raise_no_formats(info_dict
)
2235 def is_wellformed(f
):
2238 self
.report_warning(
2239 '"url" field is missing or empty - skipping format, '
2240 'there is an error in extractor')
2242 if isinstance(url
, bytes):
2243 sanitize_string_field(f
, 'url')
2246 # Filter out malformed formats for better extraction robustness
2247 formats
= list(filter(is_wellformed
, formats
))
2251 # We check that all the formats have the format and format_id fields
2252 for i
, format
in enumerate(formats
):
2253 sanitize_string_field(format
, 'format_id')
2254 sanitize_numeric_fields(format
)
2255 format
['url'] = sanitize_url(format
['url'])
2256 if not format
.get('format_id'):
2257 format
['format_id'] = compat_str(i
)
2259 # Sanitize format_id from characters used in format selector expression
2260 format
['format_id'] = re
.sub(r
'[\s,/+\[\]()]', '_', format
['format_id'])
2261 format_id
= format
['format_id']
2262 if format_id
not in formats_dict
:
2263 formats_dict
[format_id
] = []
2264 formats_dict
[format_id
].append(format
)
2266 # Make sure all formats have unique format_id
2267 common_exts
= set(itertools
.chain(*self
._format
_selection
_exts
.values()))
2268 for format_id
, ambiguous_formats
in formats_dict
.items():
2269 ambigious_id
= len(ambiguous_formats
) > 1
2270 for i
, format
in enumerate(ambiguous_formats
):
2272 format
['format_id'] = '%s-%d' % (format_id
, i
)
2273 if format
.get('ext') is None:
2274 format
['ext'] = determine_ext(format
['url']).lower()
2275 # Ensure there is no conflict between id and ext in format selection
2276 # See https://github.com/yt-dlp/yt-dlp/issues/1282
2277 if format
['format_id'] != format
['ext'] and format
['format_id'] in common_exts
:
2278 format
['format_id'] = 'f%s' % format
['format_id']
2280 for i
, format
in enumerate(formats
):
2281 if format
.get('format') is None:
2282 format
['format'] = '{id} - {res}{note}'.format(
2283 id=format
['format_id'],
2284 res
=self
.format_resolution(format
),
2285 note
=format_field(format
, 'format_note', ' (%s)'),
2287 if format
.get('protocol') is None:
2288 format
['protocol'] = determine_protocol(format
)
2289 if format
.get('resolution') is None:
2290 format
['resolution'] = self
.format_resolution(format
, default
=None)
2291 if format
.get('dynamic_range') is None and format
.get('vcodec') != 'none':
2292 format
['dynamic_range'] = 'SDR'
2293 # Add HTTP headers, so that external programs can use them from the
2295 full_format_info
= info_dict
.copy()
2296 full_format_info
.update(format
)
2297 format
['http_headers'] = self
._calc
_headers
(full_format_info
)
2298 # Remove private housekeeping stuff
2299 if '__x_forwarded_for_ip' in info_dict
:
2300 del info_dict
['__x_forwarded_for_ip']
2302 # TODO Central sorting goes here
2304 if not formats
or formats
[0] is not info_dict
:
2305 # only set the 'formats' fields if the original info_dict list them
2306 # otherwise we end up with a circular reference, the first (and unique)
2307 # element in the 'formats' field in info_dict is info_dict itself,
2308 # which can't be exported to json
2309 info_dict
['formats'] = formats
2311 info_dict
, _
= self
.pre_process(info_dict
)
2313 if self
.params
.get('list_thumbnails'):
2314 self
.list_thumbnails(info_dict
)
2315 if self
.params
.get('listformats'):
2316 if not info_dict
.get('formats') and not info_dict
.get('url'):
2317 self
.to_screen('%s has no formats' % info_dict
['id'])
2319 self
.list_formats(info_dict
)
2320 if self
.params
.get('listsubtitles'):
2321 if 'automatic_captions' in info_dict
:
2322 self
.list_subtitles(
2323 info_dict
['id'], automatic_captions
, 'automatic captions')
2324 self
.list_subtitles(info_dict
['id'], subtitles
, 'subtitles')
2325 list_only
= self
.params
.get('simulate') is None and (
2326 self
.params
.get('list_thumbnails') or self
.params
.get('listformats') or self
.params
.get('listsubtitles'))
2328 # Without this printing, -F --print-json will not work
2329 self
.__forced
_printings
(info_dict
, self
.prepare_filename(info_dict
), incomplete
=True)
2332 format_selector
= self
.format_selector
2333 if format_selector
is None:
2334 req_format
= self
._default
_format
_spec
(info_dict
, download
=download
)
2335 self
.write_debug('Default format spec: %s' % req_format
)
2336 format_selector
= self
.build_format_selector(req_format
)
2338 # While in format selection we may need to have an access to the original
2339 # format set in order to calculate some metrics or do some processing.
2340 # For now we need to be able to guess whether original formats provided
2341 # by extractor are incomplete or not (i.e. whether extractor provides only
2342 # video-only or audio-only formats) for proper formats selection for
2343 # extractors with such incomplete formats (see
2344 # https://github.com/ytdl-org/youtube-dl/pull/5556).
2345 # Since formats may be filtered during format selection and may not match
2346 # the original formats the results may be incorrect. Thus original formats
2347 # or pre-calculated metrics should be passed to format selection routines
2349 # We will pass a context object containing all necessary additional data
2350 # instead of just formats.
2351 # This fixes incorrect format selection issue (see
2352 # https://github.com/ytdl-org/youtube-dl/issues/10083).
2353 incomplete_formats
= (
2354 # All formats are video-only or
2355 all(f
.get('vcodec') != 'none' and f
.get('acodec') == 'none' for f
in formats
)
2356 # all formats are audio-only
2357 or all(f
.get('vcodec') == 'none' and f
.get('acodec') != 'none' for f
in formats
))
2361 'incomplete_formats': incomplete_formats
,
2364 formats_to_download
= list(format_selector(ctx
))
2365 if not formats_to_download
:
2366 if not self
.params
.get('ignore_no_formats_error'):
2367 raise ExtractorError('Requested format is not available', expected
=True,
2368 video_id
=info_dict
['id'], ie
=info_dict
['extractor'])
2370 self
.report_warning('Requested format is not available')
2371 # Process what we can, even without any available formats.
2372 self
.process_info(dict(info_dict
))
2375 '[info] %s: Downloading %d format(s): %s' % (
2376 info_dict
['id'], len(formats_to_download
),
2377 ", ".join([f
['format_id'] for f
in formats_to_download
])))
2378 for fmt
in formats_to_download
:
2379 new_info
= dict(info_dict
)
2380 # Save a reference to the original info_dict so that it can be modified in process_info if needed
2381 new_info
['__original_infodict'] = info_dict
2382 new_info
.update(fmt
)
2383 self
.process_info(new_info
)
2384 # We update the info dict with the best quality format (backwards compatibility)
2385 if formats_to_download
:
2386 info_dict
.update(formats_to_download
[-1])
2389 def process_subtitles(self
, video_id
, normal_subtitles
, automatic_captions
):
2390 """Select the requested subtitles and their format"""
2392 if normal_subtitles
and self
.params
.get('writesubtitles'):
2393 available_subs
.update(normal_subtitles
)
2394 if automatic_captions
and self
.params
.get('writeautomaticsub'):
2395 for lang
, cap_info
in automatic_captions
.items():
2396 if lang
not in available_subs
:
2397 available_subs
[lang
] = cap_info
2399 if (not self
.params
.get('writesubtitles') and not
2400 self
.params
.get('writeautomaticsub') or not
2404 all_sub_langs
= available_subs
.keys()
2405 if self
.params
.get('allsubtitles', False):
2406 requested_langs
= all_sub_langs
2407 elif self
.params
.get('subtitleslangs', False):
2408 # A list is used so that the order of languages will be the same as
2409 # given in subtitleslangs. See https://github.com/yt-dlp/yt-dlp/issues/1041
2410 requested_langs
= []
2411 for lang_re
in self
.params
.get('subtitleslangs'):
2412 if lang_re
== 'all':
2413 requested_langs
.extend(all_sub_langs
)
2415 discard
= lang_re
[0] == '-'
2417 lang_re
= lang_re
[1:]
2418 current_langs
= filter(re
.compile(lang_re
+ '$').match
, all_sub_langs
)
2420 for lang
in current_langs
:
2421 while lang
in requested_langs
:
2422 requested_langs
.remove(lang
)
2424 requested_langs
.extend(current_langs
)
2425 requested_langs
= orderedSet(requested_langs
)
2426 elif 'en' in available_subs
:
2427 requested_langs
= ['en']
2429 requested_langs
= [list(all_sub_langs
)[0]]
2431 self
.write_debug('Downloading subtitles: %s' % ', '.join(requested_langs
))
2433 formats_query
= self
.params
.get('subtitlesformat', 'best')
2434 formats_preference
= formats_query
.split('/') if formats_query
else []
2436 for lang
in requested_langs
:
2437 formats
= available_subs
.get(lang
)
2439 self
.report_warning('%s subtitles not available for %s' % (lang
, video_id
))
2441 for ext
in formats_preference
:
2445 matches
= list(filter(lambda f
: f
['ext'] == ext
, formats
))
2451 self
.report_warning(
2452 'No subtitle format found matching "%s" for language %s, '
2453 'using %s' % (formats_query
, lang
, f
['ext']))
2457 def __forced_printings(self
, info_dict
, filename
, incomplete
):
2458 def print_mandatory(field
, actual_field
=None):
2459 if actual_field
is None:
2460 actual_field
= field
2461 if (self
.params
.get('force%s' % field
, False)
2462 and (not incomplete
or info_dict
.get(actual_field
) is not None)):
2463 self
.to_stdout(info_dict
[actual_field
])
2465 def print_optional(field
):
2466 if (self
.params
.get('force%s' % field
, False)
2467 and info_dict
.get(field
) is not None):
2468 self
.to_stdout(info_dict
[field
])
2470 info_dict
= info_dict
.copy()
2471 if filename
is not None:
2472 info_dict
['filename'] = filename
2473 if info_dict
.get('requested_formats') is not None:
2474 # For RTMP URLs, also include the playpath
2475 info_dict
['urls'] = '\n'.join(f
['url'] + f
.get('play_path', '') for f
in info_dict
['requested_formats'])
2476 elif 'url' in info_dict
:
2477 info_dict
['urls'] = info_dict
['url'] + info_dict
.get('play_path', '')
2479 if self
.params
.get('forceprint') or self
.params
.get('forcejson'):
2480 self
.post_extract(info_dict
)
2481 for tmpl
in self
.params
.get('forceprint', []):
2482 mobj
= re
.match(r
'\w+(=?)$', tmpl
)
2483 if mobj
and mobj
.group(1):
2484 tmpl
= f
'{tmpl[:-1]} = %({tmpl[:-1]})s'
2486 tmpl
= '%({})s'.format(tmpl
)
2487 self
.to_stdout(self
.evaluate_outtmpl(tmpl
, info_dict
))
2489 print_mandatory('title')
2490 print_mandatory('id')
2491 print_mandatory('url', 'urls')
2492 print_optional('thumbnail')
2493 print_optional('description')
2494 print_optional('filename')
2495 if self
.params
.get('forceduration') and info_dict
.get('duration') is not None:
2496 self
.to_stdout(formatSeconds(info_dict
['duration']))
2497 print_mandatory('format')
2499 if self
.params
.get('forcejson'):
2500 self
.to_stdout(json
.dumps(self
.sanitize_info(info_dict
)))
2502 def dl(self
, name
, info
, subtitle
=False, test
=False):
2503 if not info
.get('url'):
2504 self
.raise_no_formats(info
, True)
2507 verbose
= self
.params
.get('verbose')
2510 'quiet': self
.params
.get('quiet') or not verbose
,
2512 'noprogress': not verbose
,
2514 'skip_unavailable_fragments': False,
2515 'keep_fragments': False,
2517 '_no_ytdl_file': True,
2520 params
= self
.params
2521 fd
= get_suitable_downloader(info
, params
, to_stdout
=(name
== '-'))(self
, params
)
2523 for ph
in self
._progress
_hooks
:
2524 fd
.add_progress_hook(ph
)
2525 urls
= '", "'.join([f
['url'] for f
in info
.get('requested_formats', [])] or [info
['url']])
2526 self
.write_debug('Invoking downloader on "%s"' % urls
)
2528 new_info
= copy
.deepcopy(self
._copy
_infodict
(info
))
2529 if new_info
.get('http_headers') is None:
2530 new_info
['http_headers'] = self
._calc
_headers
(new_info
)
2531 return fd
.download(name
, new_info
, subtitle
)
2533 def process_info(self
, info_dict
):
2534 """Process a single resolved IE result."""
2536 assert info_dict
.get('_type', 'video') == 'video'
2538 max_downloads
= self
.params
.get('max_downloads')
2539 if max_downloads
is not None:
2540 if self
._num
_downloads
>= int(max_downloads
):
2541 raise MaxDownloadsReached()
2543 # TODO: backward compatibility, to be removed
2544 info_dict
['fulltitle'] = info_dict
['title']
2546 if 'format' not in info_dict
and 'ext' in info_dict
:
2547 info_dict
['format'] = info_dict
['ext']
2549 if self
._match
_entry
(info_dict
) is not None:
2552 self
.post_extract(info_dict
)
2553 self
._num
_downloads
+= 1
2555 # info_dict['_filename'] needs to be set for backward compatibility
2556 info_dict
['_filename'] = full_filename
= self
.prepare_filename(info_dict
, warn
=True)
2557 temp_filename
= self
.prepare_filename(info_dict
, 'temp')
2561 self
.__forced
_printings
(info_dict
, full_filename
, incomplete
=('format' not in info_dict
))
2563 if self
.params
.get('simulate'):
2564 if self
.params
.get('force_write_download_archive', False):
2565 self
.record_download_archive(info_dict
)
2566 # Do nothing else if in simulate mode
2569 if full_filename
is None:
2571 if not self
._ensure
_dir
_exists
(encodeFilename(full_filename
)):
2573 if not self
._ensure
_dir
_exists
(encodeFilename(temp_filename
)):
2576 if self
._write
_description
('video', info_dict
,
2577 self
.prepare_filename(info_dict
, 'description')) is None:
2580 sub_files
= self
._write
_subtitles
(info_dict
, temp_filename
)
2581 if sub_files
is None:
2583 files_to_move
.update(dict(sub_files
))
2585 thumb_files
= self
._write
_thumbnails
(
2586 'video', info_dict
, temp_filename
, self
.prepare_filename(info_dict
, 'thumbnail'))
2587 if thumb_files
is None:
2589 files_to_move
.update(dict(thumb_files
))
2591 infofn
= self
.prepare_filename(info_dict
, 'infojson')
2592 _infojson_written
= self
._write
_info
_json
('video', info_dict
, infofn
)
2593 if _infojson_written
:
2594 info_dict
['__infojson_filename'] = infofn
2595 elif _infojson_written
is None:
2598 # Note: Annotations are deprecated
2600 if self
.params
.get('writeannotations', False):
2601 annofn
= self
.prepare_filename(info_dict
, 'annotation')
2603 if not self
._ensure
_dir
_exists
(encodeFilename(annofn
)):
2605 if not self
.params
.get('overwrites', True) and os
.path
.exists(encodeFilename(annofn
)):
2606 self
.to_screen('[info] Video annotations are already present')
2607 elif not info_dict
.get('annotations'):
2608 self
.report_warning('There are no annotations to write.')
2611 self
.to_screen('[info] Writing video annotations to: ' + annofn
)
2612 with io
.open(encodeFilename(annofn
), 'w', encoding
='utf-8') as annofile
:
2613 annofile
.write(info_dict
['annotations'])
2614 except (KeyError, TypeError):
2615 self
.report_warning('There are no annotations to write.')
2616 except (OSError, IOError):
2617 self
.report_error('Cannot write annotations file: ' + annofn
)
2620 # Write internet shortcut files
2621 url_link
= webloc_link
= desktop_link
= False
2622 if self
.params
.get('writelink', False):
2623 if sys
.platform
== "darwin": # macOS.
2625 elif sys
.platform
.startswith("linux"):
2627 else: # if sys.platform in ['win32', 'cygwin']:
2629 if self
.params
.get('writeurllink', False):
2631 if self
.params
.get('writewebloclink', False):
2633 if self
.params
.get('writedesktoplink', False):
2636 if url_link
or webloc_link
or desktop_link
:
2637 if 'webpage_url' not in info_dict
:
2638 self
.report_error('Cannot write internet shortcut file because the "webpage_url" field is missing in the media information')
2640 ascii_url
= iri_to_uri(info_dict
['webpage_url'])
2642 def _write_link_file(extension
, template
, newline
, embed_filename
):
2643 linkfn
= replace_extension(full_filename
, extension
, info_dict
.get('ext'))
2644 if self
.params
.get('overwrites', True) and os
.path
.exists(encodeFilename(linkfn
)):
2645 self
.to_screen('[info] Internet shortcut is already present')
2648 self
.to_screen('[info] Writing internet shortcut to: ' + linkfn
)
2649 with io
.open(encodeFilename(to_high_limit_path(linkfn
)), 'w', encoding
='utf-8', newline
=newline
) as linkfile
:
2650 template_vars
= {'url': ascii_url}
2652 template_vars
['filename'] = linkfn
[:-(len(extension
) + 1)]
2653 linkfile
.write(template
% template_vars
)
2654 except (OSError, IOError):
2655 self
.report_error('Cannot write internet shortcut ' + linkfn
)
2660 if not _write_link_file('url', DOT_URL_LINK_TEMPLATE
, '\r\n', embed_filename
=False):
2663 if not _write_link_file('webloc', DOT_WEBLOC_LINK_TEMPLATE
, '\n', embed_filename
=False):
2666 if not _write_link_file('desktop', DOT_DESKTOP_LINK_TEMPLATE
, '\n', embed_filename
=True):
2670 info_dict
, files_to_move
= self
.pre_process(info_dict
, 'before_dl', files_to_move
)
2671 except PostProcessingError
as err
:
2672 self
.report_error('Preprocessing: %s' % str(err
))
2675 must_record_download_archive
= False
2676 if self
.params
.get('skip_download', False):
2677 info_dict
['filepath'] = temp_filename
2678 info_dict
['__finaldir'] = os
.path
.dirname(os
.path
.abspath(encodeFilename(full_filename
)))
2679 info_dict
['__files_to_move'] = files_to_move
2680 info_dict
= self
.run_pp(MoveFilesAfterDownloadPP(self
, False), info_dict
)
2683 info_dict
.setdefault('__postprocessors', [])
2686 def existing_file(*filepaths
):
2687 ext
= info_dict
.get('ext')
2688 final_ext
= self
.params
.get('final_ext', ext
)
2690 for file in orderedSet(filepaths
):
2691 if final_ext
!= ext
:
2692 converted
= replace_extension(file, final_ext
, ext
)
2693 if os
.path
.exists(encodeFilename(converted
)):
2694 existing_files
.append(converted
)
2695 if os
.path
.exists(encodeFilename(file)):
2696 existing_files
.append(file)
2698 if not existing_files
or self
.params
.get('overwrites', False):
2699 for file in orderedSet(existing_files
):
2700 self
.report_file_delete(file)
2701 os
.remove(encodeFilename(file))
2704 info_dict
['ext'] = os
.path
.splitext(existing_files
[0])[1][1:]
2705 return existing_files
[0]
2708 if info_dict
.get('requested_formats') is not None:
2710 def compatible_formats(formats
):
2711 # TODO: some formats actually allow this (mkv, webm, ogg, mp4), but not all of them.
2712 video_formats
= [format
for format
in formats
if format
.get('vcodec') != 'none']
2713 audio_formats
= [format
for format
in formats
if format
.get('acodec') != 'none']
2714 if len(video_formats
) > 2 or len(audio_formats
) > 2:
2718 exts
= set(format
.get('ext') for format
in formats
)
2720 set(('mp3', 'mp4', 'm4a', 'm4p', 'm4b', 'm4r', 'm4v', 'ismv', 'isma')),
2723 for ext_sets
in COMPATIBLE_EXTS
:
2724 if ext_sets
.issuperset(exts
):
2726 # TODO: Check acodec/vcodec
2729 requested_formats
= info_dict
['requested_formats']
2730 old_ext
= info_dict
['ext']
2731 if self
.params
.get('merge_output_format') is None:
2732 if not compatible_formats(requested_formats
):
2733 info_dict
['ext'] = 'mkv'
2734 self
.report_warning(
2735 'Requested formats are incompatible for merge and will be merged into mkv')
2736 if (info_dict
['ext'] == 'webm'
2737 and info_dict
.get('thumbnails')
2738 # check with type instead of pp_key, __name__, or isinstance
2739 # since we dont want any custom PPs to trigger this
2740 and any(type(pp
) == EmbedThumbnailPP
for pp
in self
._pps
['post_process'])):
2741 info_dict
['ext'] = 'mkv'
2742 self
.report_warning(
2743 'webm doesn\'t support embedding a thumbnail, mkv will be used')
2744 new_ext
= info_dict
['ext']
2746 def correct_ext(filename
, ext
=new_ext
):
2749 filename_real_ext
= os
.path
.splitext(filename
)[1][1:]
2751 os
.path
.splitext(filename
)[0]
2752 if filename_real_ext
in (old_ext
, new_ext
)
2754 return '%s.%s' % (filename_wo_ext
, ext
)
2756 # Ensure filename always has a correct extension for successful merge
2757 full_filename
= correct_ext(full_filename
)
2758 temp_filename
= correct_ext(temp_filename
)
2759 dl_filename
= existing_file(full_filename
, temp_filename
)
2760 info_dict
['__real_download'] = False
2762 if dl_filename
is not None:
2763 self
.report_file_already_downloaded(dl_filename
)
2764 elif get_suitable_downloader(info_dict
, self
.params
, to_stdout
=temp_filename
== '-'):
2765 info_dict
['url'] = '\n'.join(f
['url'] for f
in requested_formats
)
2766 success
, real_download
= self
.dl(temp_filename
, info_dict
)
2767 info_dict
['__real_download'] = real_download
2770 merger
= FFmpegMergerPP(self
)
2771 if self
.params
.get('allow_unplayable_formats'):
2772 self
.report_warning(
2773 'You have requested merging of multiple formats '
2774 'while also allowing unplayable formats to be downloaded. '
2775 'The formats won\'t be merged to prevent data corruption.')
2776 elif not merger
.available
:
2777 self
.report_warning(
2778 'You have requested merging of multiple formats but ffmpeg is not installed. '
2779 'The formats won\'t be merged.')
2781 if temp_filename
== '-':
2782 reason
= ('using a downloader other than ffmpeg' if FFmpegFD
.can_merge_formats(info_dict
)
2783 else 'but the formats are incompatible for simultaneous download' if merger
.available
2784 else 'but ffmpeg is not installed')
2785 self
.report_warning(
2786 f
'You have requested downloading multiple formats to stdout {reason}. '
2787 'The formats will be streamed one after the other')
2788 fname
= temp_filename
2789 for f
in requested_formats
:
2790 new_info
= dict(info_dict
)
2791 del new_info
['requested_formats']
2793 if temp_filename
!= '-':
2794 fname
= prepend_extension(
2795 correct_ext(temp_filename
, new_info
['ext']),
2796 'f%s' % f
['format_id'], new_info
['ext'])
2797 if not self
._ensure
_dir
_exists
(fname
):
2799 f
['filepath'] = fname
2800 downloaded
.append(fname
)
2801 partial_success
, real_download
= self
.dl(fname
, new_info
)
2802 info_dict
['__real_download'] = info_dict
['__real_download'] or real_download
2803 success
= success
and partial_success
2804 if merger
.available
and not self
.params
.get('allow_unplayable_formats'):
2805 info_dict
['__postprocessors'].append(merger
)
2806 info_dict
['__files_to_merge'] = downloaded
2807 # Even if there were no downloads, it is being merged only now
2808 info_dict
['__real_download'] = True
2810 for file in downloaded
:
2811 files_to_move
[file] = None
2813 # Just a single file
2814 dl_filename
= existing_file(full_filename
, temp_filename
)
2815 if dl_filename
is None or dl_filename
== temp_filename
:
2816 # dl_filename == temp_filename could mean that the file was partially downloaded with --no-part.
2817 # So we should try to resume the download
2818 success
, real_download
= self
.dl(temp_filename
, info_dict
)
2819 info_dict
['__real_download'] = real_download
2821 self
.report_file_already_downloaded(dl_filename
)
2823 dl_filename
= dl_filename
or temp_filename
2824 info_dict
['__finaldir'] = os
.path
.dirname(os
.path
.abspath(encodeFilename(full_filename
)))
2826 except network_exceptions
as err
:
2827 self
.report_error('unable to download video data: %s' % error_to_compat_str(err
))
2829 except (OSError, IOError) as err
:
2830 raise UnavailableVideoError(err
)
2831 except (ContentTooShortError
, ) as err
:
2832 self
.report_error('content too short (expected %s bytes and served %s)' % (err
.expected
, err
.downloaded
))
2835 if success
and full_filename
!= '-':
2839 fixup_policy
= self
.params
.get('fixup')
2840 vid
= info_dict
['id']
2842 if fixup_policy
in ('ignore', 'never'):
2844 elif fixup_policy
== 'warn':
2846 elif fixup_policy
!= 'force':
2847 assert fixup_policy
in ('detect_or_warn', None)
2848 if not info_dict
.get('__real_download'):
2851 def ffmpeg_fixup(cndn
, msg
, cls
):
2855 self
.report_warning(f
'{vid}: {msg}')
2859 info_dict
['__postprocessors'].append(pp
)
2861 self
.report_warning(f
'{vid}: {msg}. Install ffmpeg to fix this automatically')
2863 stretched_ratio
= info_dict
.get('stretched_ratio')
2865 stretched_ratio
not in (1, None),
2866 f
'Non-uniform pixel ratio {stretched_ratio}',
2867 FFmpegFixupStretchedPP
)
2870 (info_dict
.get('requested_formats') is None
2871 and info_dict
.get('container') == 'm4a_dash'
2872 and info_dict
.get('ext') == 'm4a'),
2873 'writing DASH m4a. Only some players support this container',
2876 downloader
= get_suitable_downloader(info_dict
, self
.params
) if 'protocol' in info_dict
else None
2877 downloader
= downloader
.__name
__ if downloader
else None
2878 ffmpeg_fixup(info_dict
.get('requested_formats') is None and downloader
== 'HlsFD',
2879 'malformed AAC bitstream detected', FFmpegFixupM3u8PP
)
2880 ffmpeg_fixup(downloader
== 'WebSocketFragmentFD', 'malformed timestamps detected', FFmpegFixupTimestampPP
)
2881 ffmpeg_fixup(downloader
== 'WebSocketFragmentFD', 'malformed duration detected', FFmpegFixupDurationPP
)
2885 info_dict
= self
.post_process(dl_filename
, info_dict
, files_to_move
)
2886 except PostProcessingError
as err
:
2887 self
.report_error('Postprocessing: %s' % str(err
))
2890 for ph
in self
._post
_hooks
:
2891 ph(info_dict
['filepath'])
2892 except Exception as err
:
2893 self
.report_error('post hooks: %s' % str(err
))
2895 must_record_download_archive
= True
2897 if must_record_download_archive
or self
.params
.get('force_write_download_archive', False):
2898 self
.record_download_archive(info_dict
)
2899 max_downloads
= self
.params
.get('max_downloads')
2900 if max_downloads
is not None and self
._num
_downloads
>= int(max_downloads
):
2901 raise MaxDownloadsReached()
2903 def download(self
, url_list
):
2904 """Download a given list of URLs."""
2905 outtmpl
= self
.outtmpl_dict
['default']
2906 if (len(url_list
) > 1
2908 and '%' not in outtmpl
2909 and self
.params
.get('max_downloads') != 1):
2910 raise SameFileError(outtmpl
)
2912 for url
in url_list
:
2914 # It also downloads the videos
2915 res
= self
.extract_info(
2916 url
, force_generic_extractor
=self
.params
.get('force_generic_extractor', False))
2917 except UnavailableVideoError
:
2918 self
.report_error('unable to download video')
2919 except MaxDownloadsReached
:
2920 self
.to_screen('[info] Maximum number of downloads reached')
2922 except ExistingVideoReached
:
2923 self
.to_screen('[info] Encountered a video that is already in the archive, stopping due to --break-on-existing')
2925 except RejectedVideoReached
:
2926 self
.to_screen('[info] Encountered a video that did not match filter, stopping due to --break-on-reject')
2929 if self
.params
.get('dump_single_json', False):
2930 self
.post_extract(res
)
2931 self
.to_stdout(json
.dumps(self
.sanitize_info(res
)))
2933 return self
._download
_retcode
2935 def download_with_info_file(self
, info_filename
):
2936 with contextlib
.closing(fileinput
.FileInput(
2937 [info_filename
], mode
='r',
2938 openhook
=fileinput
.hook_encoded('utf-8'))) as f
:
2939 # FileInput doesn't have a read method, we can't call json.load
2940 info
= self
.sanitize_info(json
.loads('\n'.join(f
)), self
.params
.get('clean_infojson', True))
2942 self
.process_ie_result(info
, download
=True)
2943 except (DownloadError
, EntryNotInPlaylist
, ThrottledDownload
):
2944 webpage_url
= info
.get('webpage_url')
2945 if webpage_url
is not None:
2946 self
.report_warning('The info failed to download, trying with "%s"' % webpage_url
)
2947 return self
.download([webpage_url
])
2950 return self
._download
_retcode
2953 def sanitize_info(info_dict
, remove_private_keys
=False):
2954 ''' Sanitize the infodict for converting to json '''
2955 if info_dict
is None:
2957 info_dict
.setdefault('epoch', int(time
.time()))
2958 remove_keys
= {'__original_infodict'}
# Always remove this since this may contain a copy of the entire dict
2959 keep_keys
= ['_type'], # Always keep this to facilitate load-info-json
2960 if remove_private_keys
:
2962 'requested_formats', 'requested_subtitles', 'requested_entries',
2963 'filepath', 'entries', 'original_url', 'playlist_autonumber',
2965 empty_values
= (None, {}, [], set(), tuple())
2966 reject
= lambda k
, v
: k
not in keep_keys
and (
2967 k
.startswith('_') or k
in remove_keys
or v
in empty_values
)
2969 reject
= lambda k
, v
: k
in remove_keys
2970 filter_fn
= lambda obj
: (
2971 list(map(filter_fn
, obj
)) if isinstance(obj
, (LazyList
, list, tuple, set))
2972 else obj
if not isinstance(obj
, dict)
2973 else dict((k
, filter_fn(v
)) for k
, v
in obj
.items() if not reject(k
, v
)))
2974 return filter_fn(info_dict
)
2977 def filter_requested_info(info_dict
, actually_filter
=True):
2978 ''' Alias of sanitize_info for backward compatibility '''
2979 return YoutubeDL
.sanitize_info(info_dict
, actually_filter
)
2981 def run_pp(self
, pp
, infodict
):
2982 files_to_delete
= []
2983 if '__files_to_move' not in infodict
:
2984 infodict
['__files_to_move'] = {}
2986 files_to_delete
, infodict
= pp
.run(infodict
)
2987 except PostProcessingError
as e
:
2988 # Must be True and not 'only_download'
2989 if self
.params
.get('ignoreerrors') is True:
2990 self
.report_error(e
)
2994 if not files_to_delete
:
2996 if self
.params
.get('keepvideo', False):
2997 for f
in files_to_delete
:
2998 infodict
['__files_to_move'].setdefault(f
, '')
3000 for old_filename
in set(files_to_delete
):
3001 self
.to_screen('Deleting original file %s (pass -k to keep)' % old_filename
)
3003 os
.remove(encodeFilename(old_filename
))
3004 except (IOError, OSError):
3005 self
.report_warning('Unable to remove downloaded original file')
3006 if old_filename
in infodict
['__files_to_move']:
3007 del infodict
['__files_to_move'][old_filename
]
3011 def post_extract(info_dict
):
3012 def actual_post_extract(info_dict
):
3013 if info_dict
.get('_type') in ('playlist', 'multi_video'):
3014 for video_dict
in info_dict
.get('entries', {}):
3015 actual_post_extract(video_dict
or {})
3018 post_extractor
= info_dict
.get('__post_extractor') or (lambda: {})
3019 extra
= post_extractor().items()
3020 info_dict
.update(extra
)
3021 info_dict
.pop('__post_extractor', None)
3023 original_infodict
= info_dict
.get('__original_infodict') or {}
3024 original_infodict
.update(extra
)
3025 original_infodict
.pop('__post_extractor', None)
3027 actual_post_extract(info_dict
or {})
3029 def pre_process(self
, ie_info
, key
='pre_process', files_to_move
=None):
3030 info
= dict(ie_info
)
3031 info
['__files_to_move'] = files_to_move
or {}
3032 for pp
in self
._pps
[key
]:
3033 info
= self
.run_pp(pp
, info
)
3034 return info
, info
.pop('__files_to_move', None)
3036 def post_process(self
, filename
, ie_info
, files_to_move
=None):
3037 """Run all the postprocessors on the given file."""
3038 info
= dict(ie_info
)
3039 info
['filepath'] = filename
3040 info
['__files_to_move'] = files_to_move
or {}
3042 for pp
in ie_info
.get('__postprocessors', []) + self
._pps
['post_process']:
3043 info
= self
.run_pp(pp
, info
)
3044 info
= self
.run_pp(MoveFilesAfterDownloadPP(self
), info
)
3045 del info
['__files_to_move']
3046 for pp
in self
._pps
['after_move']:
3047 info
= self
.run_pp(pp
, info
)
3050 def _make_archive_id(self
, info_dict
):
3051 video_id
= info_dict
.get('id')
3054 # Future-proof against any change in case
3055 # and backwards compatibility with prior versions
3056 extractor
= info_dict
.get('extractor_key') or info_dict
.get('ie_key') # key in a playlist
3057 if extractor
is None:
3058 url
= str_or_none(info_dict
.get('url'))
3061 # Try to find matching extractor for the URL and take its ie_key
3062 for ie_key
, ie
in self
._ies
.items():
3063 if ie
.suitable(url
):
3068 return '%s %s' % (extractor
.lower(), video_id
)
3070 def in_download_archive(self
, info_dict
):
3071 fn
= self
.params
.get('download_archive')
3075 vid_id
= self
._make
_archive
_id
(info_dict
)
3077 return False # Incomplete video information
3079 return vid_id
in self
.archive
3081 def record_download_archive(self
, info_dict
):
3082 fn
= self
.params
.get('download_archive')
3085 vid_id
= self
._make
_archive
_id
(info_dict
)
3087 with locked_file(fn
, 'a', encoding
='utf-8') as archive_file
:
3088 archive_file
.write(vid_id
+ '\n')
3089 self
.archive
.add(vid_id
)
3092 def format_resolution(format
, default
='unknown'):
3093 is_images
= format
.get('vcodec') == 'none' and format
.get('acodec') == 'none'
3094 if format
.get('vcodec') == 'none' and format
.get('acodec') != 'none':
3096 if format
.get('resolution') is not None:
3097 return format
['resolution']
3098 if format
.get('width') and format
.get('height'):
3099 res
= '%dx%d' % (format
['width'], format
['height'])
3100 elif format
.get('height'):
3101 res
= '%sp' % format
['height']
3102 elif format
.get('width'):
3103 res
= '%dx?' % format
['width']
3108 return f
'{res} images' if is_images
else res
3110 def _format_note(self
, fdict
):
3112 if fdict
.get('ext') in ['f4f', 'f4m']:
3113 res
+= '(unsupported) '
3114 if fdict
.get('language'):
3117 res
+= '[%s] ' % fdict
['language']
3118 if fdict
.get('format_note') is not None:
3119 res
+= fdict
['format_note'] + ' '
3120 if fdict
.get('tbr') is not None:
3121 res
+= '%4dk ' % fdict
['tbr']
3122 if fdict
.get('container') is not None:
3125 res
+= '%s container' % fdict
['container']
3126 if (fdict
.get('vcodec') is not None
3127 and fdict
.get('vcodec') != 'none'):
3130 res
+= fdict
['vcodec']
3131 if fdict
.get('vbr') is not None:
3133 elif fdict
.get('vbr') is not None and fdict
.get('abr') is not None:
3135 if fdict
.get('vbr') is not None:
3136 res
+= '%4dk' % fdict
['vbr']
3137 if fdict
.get('fps') is not None:
3140 res
+= '%sfps' % fdict
['fps']
3141 if fdict
.get('acodec') is not None:
3144 if fdict
['acodec'] == 'none':
3147 res
+= '%-5s' % fdict
['acodec']
3148 elif fdict
.get('abr') is not None:
3152 if fdict
.get('abr') is not None:
3153 res
+= '@%3dk' % fdict
['abr']
3154 if fdict
.get('asr') is not None:
3155 res
+= ' (%5dHz)' % fdict
['asr']
3156 if fdict
.get('filesize') is not None:
3159 res
+= format_bytes(fdict
['filesize'])
3160 elif fdict
.get('filesize_approx') is not None:
3163 res
+= '~' + format_bytes(fdict
['filesize_approx'])
3166 def list_formats(self
, info_dict
):
3167 formats
= info_dict
.get('formats', [info_dict
])
3169 'list-formats' not in self
.params
.get('compat_opts', [])
3170 and self
.params
.get('listformats_table', True) is not False)
3174 format_field(f
, 'format_id'),
3175 format_field(f
, 'ext'),
3176 self
.format_resolution(f
),
3177 format_field(f
, 'fps', '%d'),
3178 format_field(f
, 'dynamic_range', '%s', ignore
=(None, 'SDR')).replace('HDR', ''),
3180 format_field(f
, 'filesize', ' %s', func
=format_bytes
) + format_field(f
, 'filesize_approx', '~%s', func
=format_bytes
),
3181 format_field(f
, 'tbr', '%4dk'),
3182 shorten_protocol_name(f
.get('protocol', '').replace("native", "n")),
3184 format_field(f
, 'vcodec', default
='unknown').replace('none', ''),
3185 format_field(f
, 'vbr', '%4dk'),
3186 format_field(f
, 'acodec', default
='unknown').replace('none', ''),
3187 format_field(f
, 'abr', '%3dk'),
3188 format_field(f
, 'asr', '%5dHz'),
3189 ', '.join(filter(None, (
3190 'UNSUPPORTED' if f
.get('ext') in ('f4f', 'f4m') else '',
3191 format_field(f
, 'language', '[%s]'),
3192 format_field(f
, 'format_note'),
3193 format_field(f
, 'container', ignore
=(None, f
.get('ext'))),
3195 ] for f
in formats
if f
.get('preference') is None or f
['preference'] >= -1000]
3196 header_line
= ['ID', 'EXT', 'RESOLUTION', 'FPS', 'HDR', '|', ' FILESIZE', ' TBR', 'PROTO',
3197 '|', 'VCODEC', ' VBR', 'ACODEC', ' ABR', ' ASR', 'MORE INFO']
3201 format_field(f
, 'format_id'),
3202 format_field(f
, 'ext'),
3203 self
.format_resolution(f
),
3204 self
._format
_note
(f
)]
3206 if f
.get('preference') is None or f
['preference'] >= -1000]
3207 header_line
= ['format code', 'extension', 'resolution', 'note']
3210 '[info] Available formats for %s:' % info_dict
['id'])
3211 self
.to_stdout(render_table(
3212 header_line
, table
, delim
=new_format
, extraGap
=(0 if new_format
else 1), hideEmpty
=new_format
))
3214 def list_thumbnails(self
, info_dict
):
3215 thumbnails
= list(info_dict
.get('thumbnails'))
3217 self
.to_screen('[info] No thumbnails present for %s' % info_dict
['id'])
3221 '[info] Thumbnails for %s:' % info_dict
['id'])
3222 self
.to_stdout(render_table(
3223 ['ID', 'width', 'height', 'URL'],
3224 [[t
['id'], t
.get('width', 'unknown'), t
.get('height', 'unknown'), t
['url']] for t
in thumbnails
]))
3226 def list_subtitles(self
, video_id
, subtitles
, name
='subtitles'):
3228 self
.to_screen('%s has no %s' % (video_id
, name
))
3231 'Available %s for %s:' % (name
, video_id
))
3233 def _row(lang
, formats
):
3234 exts
, names
= zip(*((f
['ext'], f
.get('name') or 'unknown') for f
in reversed(formats
)))
3235 if len(set(names
)) == 1:
3236 names
= [] if names
[0] == 'unknown' else names
[:1]
3237 return [lang
, ', '.join(names
), ', '.join(exts
)]
3239 self
.to_stdout(render_table(
3240 ['Language', 'Name', 'Formats'],
3241 [_row(lang
, formats
) for lang
, formats
in subtitles
.items()],
3244 def urlopen(self
, req
):
3245 """ Start an HTTP download """
3246 if isinstance(req
, compat_basestring
):
3247 req
= sanitized_Request(req
)
3248 return self
._opener
.open(req
, timeout
=self
._socket
_timeout
)
3250 def print_debug_header(self
):
3251 if not self
.params
.get('verbose'):
3253 get_encoding
= lambda stream
: getattr(stream
, 'encoding', 'missing (%s)' % type(stream
).__name
__)
3255 '[debug] Encodings: locale %s, fs %s, stdout %s, stderr %s, pref %s\n' % (
3256 locale
.getpreferredencoding(),
3257 sys
.getfilesystemencoding(),
3258 get_encoding(self
._screen
_file
), get_encoding(self
._err
_file
),
3259 self
.get_encoding()))
3261 logger
= self
.params
.get('logger')
3263 write_debug
= lambda msg
: logger
.debug(f
'[debug] {msg}')
3264 write_debug(encoding_str
)
3266 write_debug
= lambda msg
: self
._write
_string
(f
'[debug] {msg}')
3267 write_string(encoding_str
, encoding
=None)
3269 source
= detect_variant()
3270 write_debug('yt-dlp version %s%s\n' % (__version__
, '' if source
== 'unknown' else f
' ({source})'))
3271 if not _LAZY_LOADER
:
3272 if os
.environ
.get('YTDLP_NO_LAZY_EXTRACTORS'):
3273 write_debug('Lazy loading extractors is forcibly disabled\n')
3275 write_debug('Lazy loading extractors is disabled\n')
3276 if plugin_extractors
or plugin_postprocessors
:
3277 write_debug('Plugins: %s\n' % [
3278 '%s%s' % (klass
.__name
__, '' if klass
.__name
__ == name
else f
' as {name}')
3279 for name
, klass
in itertools
.chain(plugin_extractors
.items(), plugin_postprocessors
.items())])
3280 if self
.params
.get('compat_opts'):
3281 write_debug('Compatibility options: %s\n' % ', '.join(self
.params
.get('compat_opts')))
3284 ['git', 'rev-parse', '--short', 'HEAD'],
3285 stdout
=subprocess
.PIPE
, stderr
=subprocess
.PIPE
,
3286 cwd
=os
.path
.dirname(os
.path
.abspath(__file__
)))
3287 out
, err
= sp
.communicate_or_kill()
3288 out
= out
.decode().strip()
3289 if re
.match('[0-9a-f]+', out
):
3290 write_debug('Git HEAD: %s\n' % out
)
3297 def python_implementation():
3298 impl_name
= platform
.python_implementation()
3299 if impl_name
== 'PyPy' and hasattr(sys
, 'pypy_version_info'):
3300 return impl_name
+ ' version %d.%d.%d' % sys
.pypy_version_info
[:3]
3303 write_debug('Python version %s (%s %s) - %s\n' % (
3304 platform
.python_version(),
3305 python_implementation(),
3306 platform
.architecture()[0],
3309 exe_versions
= FFmpegPostProcessor
.get_versions(self
)
3310 exe_versions
['rtmpdump'] = rtmpdump_version()
3311 exe_versions
['phantomjs'] = PhantomJSwrapper
._version
()
3312 exe_str
= ', '.join(
3313 f
'{exe} {v}' for exe
, v
in sorted(exe_versions
.items()) if v
3315 write_debug('exe versions: %s\n' % exe_str
)
3317 from .downloader
.websocket
import has_websockets
3318 from .postprocessor
.embedthumbnail
import has_mutagen
3319 from .cookies
import SQLITE_AVAILABLE
, KEYRING_AVAILABLE
3321 lib_str
= ', '.join(sorted(filter(None, (
3322 compat_pycrypto_AES
and compat_pycrypto_AES
.__name
__.split('.')[0],
3323 has_websockets
and 'websockets',
3324 has_mutagen
and 'mutagen',
3325 SQLITE_AVAILABLE
and 'sqlite',
3326 KEYRING_AVAILABLE
and 'keyring',
3328 write_debug('Optional libraries: %s\n' % lib_str
)
3329 write_debug('ANSI escape support: stdout = %s, stderr = %s\n' % (
3330 supports_terminal_sequences(self
._screen
_file
),
3331 supports_terminal_sequences(self
._err
_file
)))
3334 for handler
in self
._opener
.handlers
:
3335 if hasattr(handler
, 'proxies'):
3336 proxy_map
.update(handler
.proxies
)
3337 write_debug('Proxy map: ' + compat_str(proxy_map
) + '\n')
3339 if self
.params
.get('call_home', False):
3340 ipaddr
= self
.urlopen('https://yt-dl.org/ip').read().decode('utf-8')
3341 write_debug('Public IP address: %s\n' % ipaddr
)
3343 latest_version
= self
.urlopen(
3344 'https://yt-dl.org/latest/version').read().decode('utf-8')
3345 if version_tuple(latest_version
) > version_tuple(__version__
):
3346 self
.report_warning(
3347 'You are using an outdated version (newest version: %s)! '
3348 'See https://yt-dl.org/update if you need help updating.' %
3351 def _setup_opener(self
):
3352 timeout_val
= self
.params
.get('socket_timeout')
3353 self
._socket
_timeout
= 20 if timeout_val
is None else float(timeout_val
)
3355 opts_cookiesfrombrowser
= self
.params
.get('cookiesfrombrowser')
3356 opts_cookiefile
= self
.params
.get('cookiefile')
3357 opts_proxy
= self
.params
.get('proxy')
3359 self
.cookiejar
= load_cookies(opts_cookiefile
, opts_cookiesfrombrowser
, self
)
3361 cookie_processor
= YoutubeDLCookieProcessor(self
.cookiejar
)
3362 if opts_proxy
is not None:
3363 if opts_proxy
== '':
3366 proxies
= {'http': opts_proxy, 'https': opts_proxy}
3368 proxies
= compat_urllib_request
.getproxies()
3369 # Set HTTPS proxy to HTTP one if given (https://github.com/ytdl-org/youtube-dl/issues/805)
3370 if 'http' in proxies
and 'https' not in proxies
:
3371 proxies
['https'] = proxies
['http']
3372 proxy_handler
= PerRequestProxyHandler(proxies
)
3374 debuglevel
= 1 if self
.params
.get('debug_printtraffic') else 0
3375 https_handler
= make_HTTPS_handler(self
.params
, debuglevel
=debuglevel
)
3376 ydlh
= YoutubeDLHandler(self
.params
, debuglevel
=debuglevel
)
3377 redirect_handler
= YoutubeDLRedirectHandler()
3378 data_handler
= compat_urllib_request_DataHandler()
3380 # When passing our own FileHandler instance, build_opener won't add the
3381 # default FileHandler and allows us to disable the file protocol, which
3382 # can be used for malicious purposes (see
3383 # https://github.com/ytdl-org/youtube-dl/issues/8227)
3384 file_handler
= compat_urllib_request
.FileHandler()
3386 def file_open(*args
, **kwargs
):
3387 raise compat_urllib_error
.URLError('file:// scheme is explicitly disabled in yt-dlp for security reasons')
3388 file_handler
.file_open
= file_open
3390 opener
= compat_urllib_request
.build_opener(
3391 proxy_handler
, https_handler
, cookie_processor
, ydlh
, redirect_handler
, data_handler
, file_handler
)
3393 # Delete the default user-agent header, which would otherwise apply in
3394 # cases where our custom HTTP handler doesn't come into play
3395 # (See https://github.com/ytdl-org/youtube-dl/issues/1309 for details)
3396 opener
.addheaders
= []
3397 self
._opener
= opener
3399 def encode(self
, s
):
3400 if isinstance(s
, bytes):
3401 return s
# Already encoded
3404 return s
.encode(self
.get_encoding())
3405 except UnicodeEncodeError as err
:
3406 err
.reason
= err
.reason
+ '. Check your system encoding configuration or use the --encoding option.'
3409 def get_encoding(self
):
3410 encoding
= self
.params
.get('encoding')
3411 if encoding
is None:
3412 encoding
= preferredencoding()
3415 def _write_info_json(self
, label
, ie_result
, infofn
):
3416 ''' Write infojson and returns True = written, False = skip, None = error '''
3417 if not self
.params
.get('writeinfojson'):
3420 self
.write_debug(f
'Skipping writing {label} infojson')
3422 elif not self
._ensure
_dir
_exists
(infofn
):
3424 elif not self
.params
.get('overwrites', True) and os
.path
.exists(infofn
):
3425 self
.to_screen(f
'[info] {label.title()} metadata is already present')
3427 self
.to_screen(f
'[info] Writing {label} metadata as JSON to: {infofn}')
3429 write_json_file(self
.sanitize_info(ie_result
, self
.params
.get('clean_infojson', True)), infofn
)
3430 except (OSError, IOError):
3431 self
.report_error(f
'Cannot write {label} metadata to JSON file {infofn}')
3435 def _write_description(self
, label
, ie_result
, descfn
):
3436 ''' Write description and returns True = written, False = skip, None = error '''
3437 if not self
.params
.get('writedescription'):
3440 self
.write_debug(f
'Skipping writing {label} description')
3442 elif not self
._ensure
_dir
_exists
(descfn
):
3444 elif not self
.params
.get('overwrites', True) and os
.path
.exists(descfn
):
3445 self
.to_screen(f
'[info] {label.title()} description is already present')
3446 elif ie_result
.get('description') is None:
3447 self
.report_warning(f
'There\'s no {label} description to write')
3451 self
.to_screen(f
'[info] Writing {label} description to: {descfn}')
3452 with io
.open(encodeFilename(descfn
), 'w', encoding
='utf-8') as descfile
:
3453 descfile
.write(ie_result
['description'])
3454 except (OSError, IOError):
3455 self
.report_error(f
'Cannot write {label} description file {descfn}')
3459 def _write_subtitles(self
, info_dict
, filename
):
3460 ''' Write subtitles to file and return list of (sub_filename, final_sub_filename); or None if error'''
3462 subtitles
= info_dict
.get('requested_subtitles')
3463 if not subtitles
or not (self
.params
.get('writesubtitles') or self
.params
.get('writeautomaticsub')):
3464 # subtitles download errors are already managed as troubles in relevant IE
3465 # that way it will silently go on when used with unsupporting IE
3468 sub_filename_base
= self
.prepare_filename(info_dict
, 'subtitle')
3469 if not sub_filename_base
:
3470 self
.to_screen('[info] Skipping writing video subtitles')
3472 for sub_lang
, sub_info
in subtitles
.items():
3473 sub_format
= sub_info
['ext']
3474 sub_filename
= subtitles_filename(filename
, sub_lang
, sub_format
, info_dict
.get('ext'))
3475 sub_filename_final
= subtitles_filename(sub_filename_base
, sub_lang
, sub_format
, info_dict
.get('ext'))
3476 if not self
.params
.get('overwrites', True) and os
.path
.exists(sub_filename
):
3477 self
.to_screen(f
'[info] Video subtitle {sub_lang}.{sub_format} is already present')
3478 sub_info
['filepath'] = sub_filename
3479 ret
.append((sub_filename
, sub_filename_final
))
3482 self
.to_screen(f
'[info] Writing video subtitles to: {sub_filename}')
3483 if sub_info
.get('data') is not None:
3485 # Use newline='' to prevent conversion of newline characters
3486 # See https://github.com/ytdl-org/youtube-dl/issues/10268
3487 with io
.open(sub_filename
, 'w', encoding
='utf-8', newline
='') as subfile
:
3488 subfile
.write(sub_info
['data'])
3489 sub_info
['filepath'] = sub_filename
3490 ret
.append((sub_filename
, sub_filename_final
))
3492 except (OSError, IOError):
3493 self
.report_error(f
'Cannot write video subtitles file {sub_filename}')
3497 sub_copy
= sub_info
.copy()
3498 sub_copy
.setdefault('http_headers', info_dict
.get('http_headers'))
3499 self
.dl(sub_filename
, sub_copy
, subtitle
=True)
3500 sub_info
['filepath'] = sub_filename
3501 ret
.append((sub_filename
, sub_filename_final
))
3502 except (ExtractorError
, IOError, OSError, ValueError) + network_exceptions
as err
:
3503 self
.report_warning(f
'Unable to download video subtitles for {sub_lang!r}: {err}')
3507 def _write_thumbnails(self
, label
, info_dict
, filename
, thumb_filename_base
=None):
3508 ''' Write thumbnails to file and return list of (thumb_filename, final_thumb_filename) '''
3509 write_all
= self
.params
.get('write_all_thumbnails', False)
3510 thumbnails
, ret
= [], []
3511 if write_all
or self
.params
.get('writethumbnail', False):
3512 thumbnails
= info_dict
.get('thumbnails') or []
3513 multiple
= write_all
and len(thumbnails
) > 1
3515 if thumb_filename_base
is None:
3516 thumb_filename_base
= filename
3517 if thumbnails
and not thumb_filename_base
:
3518 self
.write_debug(f
'Skipping writing {label} thumbnail')
3521 for t
in thumbnails
[::-1]:
3522 thumb_ext
= (f
'{t["id"]}.' if multiple
else '') + determine_ext(t
['url'], 'jpg')
3523 thumb_display_id
= f
'{label} thumbnail' + (f
' {t["id"]}' if multiple
else '')
3524 thumb_filename
= replace_extension(filename
, thumb_ext
, info_dict
.get('ext'))
3525 thumb_filename_final
= replace_extension(thumb_filename_base
, thumb_ext
, info_dict
.get('ext'))
3527 if not self
.params
.get('overwrites', True) and os
.path
.exists(thumb_filename
):
3528 ret
.append((thumb_filename
, thumb_filename_final
))
3529 t
['filepath'] = thumb_filename
3530 self
.to_screen(f
'[info] {thumb_display_id.title()} is already present')
3532 self
.to_screen(f
'[info] Downloading {thumb_display_id} ...')
3534 uf
= self
.urlopen(t
['url'])
3535 self
.to_screen(f
'[info] Writing {thumb_display_id} to: {thumb_filename}')
3536 with open(encodeFilename(thumb_filename
), 'wb') as thumbf
:
3537 shutil
.copyfileobj(uf
, thumbf
)
3538 ret
.append((thumb_filename
, thumb_filename_final
))
3539 t
['filepath'] = thumb_filename
3540 except network_exceptions
as err
:
3541 self
.report_warning(f
'Unable to download {thumb_display_id}: {err}')
3542 if ret
and not write_all
: