24 from string
import ascii_letters
26 from .cache
import Cache
27 from .compat
import compat_os_name
, compat_shlex_quote
28 from .cookies
import load_cookies
29 from .downloader
import FFmpegFD
, get_suitable_downloader
, shorten_protocol_name
30 from .downloader
.rtmp
import rtmpdump_version
31 from .extractor
import gen_extractor_classes
, get_info_extractor
32 from .extractor
.common
import UnsupportedURLIE
33 from .extractor
.openload
import PhantomJSwrapper
34 from .minicurses
import format_text
35 from .postprocessor
import _PLUGIN_CLASSES
as plugin_postprocessors
36 from .postprocessor
import (
38 FFmpegFixupDuplicateMoovPP
,
39 FFmpegFixupDurationPP
,
42 FFmpegFixupStretchedPP
,
43 FFmpegFixupTimestampPP
,
46 FFmpegVideoConvertorPP
,
47 MoveFilesAfterDownloadPP
,
50 from .postprocessor
.ffmpeg
import resolve_mapping
as resolve_recode_mapping
51 from .update
import REPOSITORY
, current_git_head
, detect_variant
78 PerRequestProxyHandler
,
85 UnavailableVideoError
,
87 YoutubeDLCookieProcessor
,
89 YoutubeDLRedirectHandler
,
105 format_decimal_suffix
,
122 orderedSet_from_options
,
126 register_socks_protocols
,
127 remove_terminal_sequences
,
138 supports_terminal_sequences
,
148 windows_enable_vt_mode
,
152 from .version
import RELEASE_GIT_HEAD
, VARIANT
, __version__
154 if compat_os_name
== 'nt':
161 YoutubeDL objects are the ones responsible of downloading the
162 actual video file and writing it to disk if the user has requested
163 it, among some other tasks. In most cases there should be one per
164 program. As, given a video URL, the downloader doesn't know how to
165 extract all the needed information, task that InfoExtractors do, it
166 has to pass the URL to one of them.
168 For this, YoutubeDL objects have a method that allows
169 InfoExtractors to be registered in a given order. When it is passed
170 a URL, the YoutubeDL object handles it to the first InfoExtractor it
171 finds that reports being able to handle it. The InfoExtractor extracts
172 all the information about the video or videos the URL refers to, and
173 YoutubeDL process the extracted information, possibly using a File
174 Downloader to download the video.
176 YoutubeDL objects accept a lot of parameters. In order not to saturate
177 the object constructor with arguments, it receives a dictionary of
178 options instead. These options are available through the params
179 attribute for the InfoExtractors to use. The YoutubeDL also
180 registers itself as the downloader in charge for the InfoExtractors
181 that are added to it, so this is a "mutual registration".
185 username: Username for authentication purposes.
186 password: Password for authentication purposes.
187 videopassword: Password for accessing a video.
188 ap_mso: Adobe Pass multiple-system operator identifier.
189 ap_username: Multiple-system operator account username.
190 ap_password: Multiple-system operator account password.
191 usenetrc: Use netrc for authentication instead.
192 verbose: Print additional info to stdout.
193 quiet: Do not print messages to stdout.
194 no_warnings: Do not print out anything for warnings.
195 forceprint: A dict with keys WHEN mapped to a list of templates to
196 print to stdout. The allowed keys are video or any of the
197 items in utils.POSTPROCESS_WHEN.
198 For compatibility, a single list is also accepted
199 print_to_file: A dict with keys WHEN (same as forceprint) mapped to
200 a list of tuples with (template, filename)
201 forcejson: Force printing info_dict as JSON.
202 dump_single_json: Force printing the info_dict of the whole playlist
203 (or video) as a single JSON line.
204 force_write_download_archive: Force writing download archive regardless
205 of 'skip_download' or 'simulate'.
206 simulate: Do not download the video files. If unset (or None),
207 simulate only if listsubtitles, listformats or list_thumbnails is used
208 format: Video format code. see "FORMAT SELECTION" for more details.
209 You can also pass a function. The function takes 'ctx' as
210 argument and returns the formats to download.
211 See "build_format_selector" for an implementation
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: A list of fields by which to sort the video formats.
217 See "Sorting Formats" for more details.
218 format_sort_force: Force the given format_sort. see "Sorting Formats"
220 prefer_free_formats: Whether to prefer video formats with free containers
221 over non-free ones of same quality.
222 allow_multiple_video_streams: Allow multiple video streams to be merged
224 allow_multiple_audio_streams: Allow multiple audio streams to be merged
226 check_formats Whether to test if the formats are downloadable.
227 Can be True (check all), False (check none),
228 'selected' (check selected formats),
229 or None (check only if requested by extractor)
230 paths: Dictionary of output paths. The allowed keys are 'home'
231 'temp' and the keys of OUTTMPL_TYPES (in utils.py)
232 outtmpl: Dictionary of templates for output names. Allowed keys
233 are 'default' and the keys of OUTTMPL_TYPES (in utils.py).
234 For compatibility with youtube-dl, a single string can also be used
235 outtmpl_na_placeholder: Placeholder for unavailable meta fields.
236 restrictfilenames: Do not allow "&" and spaces in file names
237 trim_file_name: Limit length of filename (extension excluded)
238 windowsfilenames: Force the filenames to be windows compatible
239 ignoreerrors: Do not stop on download/postprocessing errors.
240 Can be 'only_download' to ignore only download errors.
241 Default is 'only_download' for CLI, but False for API
242 skip_playlist_after_errors: Number of allowed failures until the rest of
243 the playlist is skipped
244 allowed_extractors: List of regexes to match against extractor names that are allowed
245 overwrites: Overwrite all video and metadata files if True,
246 overwrite only non-video files if None
247 and don't overwrite any file if False
248 For compatibility with youtube-dl,
249 "nooverwrites" may also be used instead
250 playlist_items: Specific indices of playlist to download.
251 playlistrandom: Download playlist items in random order.
252 lazy_playlist: Process playlist entries as they are received.
253 matchtitle: Download only matching titles.
254 rejecttitle: Reject downloads for matching titles.
255 logger: Log messages to a logging.Logger instance.
256 logtostderr: Print everything to stderr instead of stdout.
257 consoletitle: Display progress in console window's titlebar.
258 writedescription: Write the video description to a .description file
259 writeinfojson: Write the video description to a .info.json file
260 clean_infojson: Remove private fields from the infojson
261 getcomments: Extract video comments. This will not be written to disk
262 unless writeinfojson is also given
263 writeannotations: Write the video annotations to a .annotations.xml file
264 writethumbnail: Write the thumbnail image to a file
265 allow_playlist_files: Whether to write playlists' description, infojson etc
266 also to disk when using the 'write*' options
267 write_all_thumbnails: Write all thumbnail formats to files
268 writelink: Write an internet shortcut file, depending on the
269 current platform (.url/.webloc/.desktop)
270 writeurllink: Write a Windows internet shortcut file (.url)
271 writewebloclink: Write a macOS internet shortcut file (.webloc)
272 writedesktoplink: Write a Linux internet shortcut file (.desktop)
273 writesubtitles: Write the video subtitles to a file
274 writeautomaticsub: Write the automatically generated subtitles to a file
275 listsubtitles: Lists all available subtitles for the video
276 subtitlesformat: The format code for subtitles
277 subtitleslangs: List of languages of the subtitles to download (can be regex).
278 The list may contain "all" to refer to all the available
279 subtitles. The language can be prefixed with a "-" to
280 exclude it from the requested languages, e.g. ['all', '-live_chat']
281 keepvideo: Keep the video file after post-processing
282 daterange: A DateRange object, download only if the upload_date is in the range.
283 skip_download: Skip the actual download of the video file
284 cachedir: Location of the cache files in the filesystem.
285 False to disable filesystem cache.
286 noplaylist: Download single video instead of a playlist if in doubt.
287 age_limit: An integer representing the user's age in years.
288 Unsuitable videos for the given age are skipped.
289 min_views: An integer representing the minimum view count the video
290 must have in order to not be skipped.
291 Videos without view count information are always
292 downloaded. None for no limit.
293 max_views: An integer representing the maximum view count.
294 Videos that are more popular than that are not
296 Videos without view count information are always
297 downloaded. None for no limit.
298 download_archive: A set, or the name of a file where all downloads are recorded.
299 Videos already present in the file are not downloaded again.
300 break_on_existing: Stop the download process after attempting to download a
301 file that is in the archive.
302 break_on_reject: Stop the download process when encountering a video that
303 has been filtered out.
304 break_per_url: Whether break_on_reject and break_on_existing
305 should act on each input URL as opposed to for the entire queue
306 cookiefile: File name or text stream from where cookies should be read and dumped to
307 cookiesfrombrowser: A tuple containing the name of the browser, the profile
308 name/path from where cookies are loaded, the name of the keyring,
309 and the container name, e.g. ('chrome', ) or
310 ('vivaldi', 'default', 'BASICTEXT') or ('firefox', 'default', None, 'Meta')
311 legacyserverconnect: Explicitly allow HTTPS connection to servers that do not
312 support RFC 5746 secure renegotiation
313 nocheckcertificate: Do not verify SSL certificates
314 client_certificate: Path to client certificate file in PEM format. May include the private key
315 client_certificate_key: Path to private key file for client certificate
316 client_certificate_password: Password for client certificate private key, if encrypted.
317 If not provided and the key is encrypted, yt-dlp will ask interactively
318 prefer_insecure: Use HTTP instead of HTTPS to retrieve information.
319 (Only supported by some extractors)
320 http_headers: A dictionary of custom headers to be used for all requests
321 proxy: URL of the proxy server to use
322 geo_verification_proxy: URL of the proxy to use for IP address verification
323 on geo-restricted sites.
324 socket_timeout: Time to wait for unresponsive hosts, in seconds
325 bidi_workaround: Work around buggy terminals without bidirectional text
326 support, using fridibi
327 debug_printtraffic:Print out sent and received HTTP traffic
328 default_search: Prepend this string if an input url is not valid.
329 'auto' for elaborate guessing
330 encoding: Use this encoding instead of the system-specified.
331 extract_flat: Whether to resolve and process url_results further
332 * False: Always process (default)
333 * True: Never process
334 * 'in_playlist': Do not process inside playlist/multi_video
335 * 'discard': Always process, but don't return the result
336 from inside playlist/multi_video
337 * 'discard_in_playlist': Same as "discard", but only for
338 playlists (not multi_video)
339 wait_for_video: If given, wait for scheduled streams to become available.
340 The value should be a tuple containing the range
341 (min_secs, max_secs) to wait between retries
342 postprocessors: A list of dictionaries, each with an entry
343 * key: The name of the postprocessor. See
344 yt_dlp/postprocessor/__init__.py for a list.
345 * when: When to run the postprocessor. Allowed values are
346 the entries of utils.POSTPROCESS_WHEN
347 Assumed to be 'post_process' if not given
348 progress_hooks: A list of functions that get called on download
349 progress, with a dictionary with the entries
350 * status: One of "downloading", "error", or "finished".
351 Check this first and ignore unknown values.
352 * info_dict: The extracted info_dict
354 If status is one of "downloading", or "finished", the
355 following properties may also be present:
356 * filename: The final filename (always present)
357 * tmpfilename: The filename we're currently writing to
358 * downloaded_bytes: Bytes on disk
359 * total_bytes: Size of the whole file, None if unknown
360 * total_bytes_estimate: Guess of the eventual file size,
362 * elapsed: The number of seconds since download started.
363 * eta: The estimated time in seconds, None if unknown
364 * speed: The download speed in bytes/second, None if
366 * fragment_index: The counter of the currently
367 downloaded video fragment.
368 * fragment_count: The number of fragments (= individual
369 files that will be merged)
371 Progress hooks are guaranteed to be called at least once
372 (with status "finished") if the download is successful.
373 postprocessor_hooks: A list of functions that get called on postprocessing
374 progress, with a dictionary with the entries
375 * status: One of "started", "processing", or "finished".
376 Check this first and ignore unknown values.
377 * postprocessor: Name of the postprocessor
378 * info_dict: The extracted info_dict
380 Progress hooks are guaranteed to be called at least twice
381 (with status "started" and "finished") if the processing is successful.
382 merge_output_format: "/" separated list of extensions to use when merging formats.
383 final_ext: Expected final extension; used to detect when the file was
384 already downloaded and converted
385 fixup: Automatically correct known faults of the file.
387 - "never": do nothing
388 - "warn": only emit a warning
389 - "detect_or_warn": check whether we can do anything
390 about it, warn otherwise (default)
391 source_address: Client-side IP address to bind to.
392 sleep_interval_requests: Number of seconds to sleep between requests
394 sleep_interval: Number of seconds to sleep before each download when
395 used alone or a lower bound of a range for randomized
396 sleep before each download (minimum possible number
397 of seconds to sleep) when used along with
399 max_sleep_interval:Upper bound of a range for randomized sleep before each
400 download (maximum possible number of seconds to sleep).
401 Must only be used along with sleep_interval.
402 Actual sleep time will be a random float from range
403 [sleep_interval; max_sleep_interval].
404 sleep_interval_subtitles: Number of seconds to sleep before each subtitle download
405 listformats: Print an overview of available video formats and exit.
406 list_thumbnails: Print a table of all thumbnails and exit.
407 match_filter: A function that gets called for every video with the signature
408 (info_dict, *, incomplete: bool) -> Optional[str]
409 For backward compatibility with youtube-dl, the signature
410 (info_dict) -> Optional[str] is also allowed.
411 - If it returns a message, the video is ignored.
412 - If it returns None, the video is downloaded.
413 - If it returns utils.NO_DEFAULT, the user is interactively
414 asked whether to download the video.
415 match_filter_func in utils.py is one example for this.
416 no_color: Do not emit color codes in output.
417 geo_bypass: Bypass geographic restriction via faking X-Forwarded-For
420 Two-letter ISO 3166-2 country code that will be used for
421 explicit geographic restriction bypassing via faking
422 X-Forwarded-For HTTP header
424 IP range in CIDR notation that will be used similarly to
426 external_downloader: A dictionary of protocol keys and the executable of the
427 external downloader to use for it. The allowed protocols
428 are default|http|ftp|m3u8|dash|rtsp|rtmp|mms.
429 Set the value to 'native' to use the native downloader
430 compat_opts: Compatibility options. See "Differences in default behavior".
431 The following options do not work when used through the API:
432 filename, abort-on-error, multistreams, no-live-chat, format-sort
433 no-clean-infojson, no-playlist-metafiles, no-keep-subs, no-attach-info-json.
434 Refer __init__.py for their implementation
435 progress_template: Dictionary of templates for progress outputs.
436 Allowed keys are 'download', 'postprocess',
437 'download-title' (console title) and 'postprocess-title'.
438 The template is mapped on a dictionary with keys 'progress' and 'info'
439 retry_sleep_functions: Dictionary of functions that takes the number of attempts
440 as argument and returns the time to sleep in seconds.
441 Allowed keys are 'http', 'fragment', 'file_access'
442 download_ranges: A callback function that gets called for every video with
443 the signature (info_dict, ydl) -> Iterable[Section].
444 Only the returned sections will be downloaded.
445 Each Section is a dict with the following keys:
446 * start_time: Start time of the section in seconds
447 * end_time: End time of the section in seconds
448 * title: Section title (Optional)
449 * index: Section number (Optional)
450 force_keyframes_at_cuts: Re-encode the video when downloading ranges to get precise cuts
451 noprogress: Do not print the progress bar
452 live_from_start: Whether to download livestreams videos from the start
454 The following parameters are not used by YoutubeDL itself, they are used by
455 the downloader (see yt_dlp/downloader/common.py):
456 nopart, updatetime, buffersize, ratelimit, throttledratelimit, min_filesize,
457 max_filesize, test, noresizebuffer, retries, file_access_retries, fragment_retries,
458 continuedl, xattr_set_filesize, hls_use_mpegts, http_chunk_size,
459 external_downloader_args, concurrent_fragment_downloads.
461 The following options are used by the post processors:
462 ffmpeg_location: Location of the ffmpeg/avconv binary; either the path
463 to the binary or its containing directory.
464 postprocessor_args: A dictionary of postprocessor/executable keys (in lower case)
465 and a list of additional command-line arguments for the
466 postprocessor/executable. The dict can also have "PP+EXE" keys
467 which are used when the given exe is used by the given PP.
468 Use 'default' as the name for arguments to passed to all PP
469 For compatibility with youtube-dl, a single list of args
472 The following options are used by the extractors:
473 extractor_retries: Number of times to retry for known errors
474 dynamic_mpd: Whether to process dynamic DASH manifests (default: True)
475 hls_split_discontinuity: Split HLS playlists to different formats at
476 discontinuities such as ad breaks (default: False)
477 extractor_args: A dictionary of arguments to be passed to the extractors.
478 See "EXTRACTOR ARGUMENTS" for details.
479 E.g. {'youtube': {'skip': ['dash', 'hls']}}
480 mark_watched: Mark videos watched (even with --simulate). Only for YouTube
482 The following options are deprecated and may be removed in the future:
484 force_generic_extractor: Force downloader to use the generic extractor
485 - Use allowed_extractors = ['generic', 'default']
486 playliststart: - Use playlist_items
487 Playlist item to start at.
488 playlistend: - Use playlist_items
489 Playlist item to end at.
490 playlistreverse: - Use playlist_items
491 Download playlist items in reverse order.
492 forceurl: - Use forceprint
493 Force printing final URL.
494 forcetitle: - Use forceprint
495 Force printing title.
496 forceid: - Use forceprint
498 forcethumbnail: - Use forceprint
499 Force printing thumbnail URL.
500 forcedescription: - Use forceprint
501 Force printing description.
502 forcefilename: - Use forceprint
503 Force printing final filename.
504 forceduration: - Use forceprint
505 Force printing duration.
506 allsubtitles: - Use subtitleslangs = ['all']
507 Downloads all the subtitles of the video
508 (requires writesubtitles or writeautomaticsub)
509 include_ads: - Doesn't work
511 call_home: - Not implemented
512 Boolean, true iff we are allowed to contact the
513 yt-dlp servers for debugging.
514 post_hooks: - Register a custom postprocessor
515 A list of functions that get called as the final step
516 for each video file, after all postprocessors have been
517 called. The filename will be passed as the only argument.
518 hls_prefer_native: - Use external_downloader = {'m3u8': 'native'} or {'m3u8': 'ffmpeg'}.
519 Use the native HLS downloader instead of ffmpeg/avconv
520 if True, otherwise use ffmpeg/avconv if False, otherwise
521 use downloader suggested by extractor if None.
522 prefer_ffmpeg: - avconv support is deprecated
523 If False, use avconv instead of ffmpeg if both are available,
524 otherwise prefer ffmpeg.
525 youtube_include_dash_manifest: - Use extractor_args
526 If True (default), DASH manifests and related
527 data will be downloaded and processed by extractor.
528 You can reduce network I/O by disabling it if you don't
529 care about DASH. (only for youtube)
530 youtube_include_hls_manifest: - Use extractor_args
531 If True (default), HLS manifests and related
532 data will be downloaded and processed by extractor.
533 You can reduce network I/O by disabling it if you don't
534 care about HLS. (only for youtube)
538 'width', 'height', 'asr', 'audio_channels', 'fps',
539 'tbr', 'abr', 'vbr', 'filesize', 'filesize_approx',
540 'timestamp', 'release_timestamp',
541 'duration', 'view_count', 'like_count', 'dislike_count', 'repost_count',
542 'average_rating', 'comment_count', 'age_limit',
543 'start_time', 'end_time',
544 'chapter_number', 'season_number', 'episode_number',
545 'track_number', 'disc_number', 'release_year',
549 # NB: Keep in sync with the docstring of extractor/common.py
550 'url', 'manifest_url', 'manifest_stream_number', 'ext', 'format', 'format_id', 'format_note',
551 'width', 'height', 'aspect_ratio', 'resolution', 'dynamic_range', 'tbr', 'abr', 'acodec', 'asr', 'audio_channels',
552 'vbr', 'fps', 'vcodec', 'container', 'filesize', 'filesize_approx', 'rows', 'columns',
553 'player_url', 'protocol', 'fragment_base_url', 'fragments', 'is_from_start',
554 'preference', 'language', 'language_preference', 'quality', 'source_preference',
555 'http_headers', 'stretched_ratio', 'no_resume', 'has_drm', 'downloader_options',
556 'page_url', 'app', 'play_path', 'tc_url', 'flash_version', 'rtmp_live', 'rtmp_conn', 'rtmp_protocol', 'rtmp_real_time'
558 _format_selection_exts
= {
559 'audio': set(MEDIA_EXTENSIONS
.common_audio
),
560 'video': set(MEDIA_EXTENSIONS
.common_video
+ ('3gp', )),
561 'storyboards': set(MEDIA_EXTENSIONS
.storyboards
),
564 def __init__(self
, params
=None, auto_init
=True):
565 """Create a FileDownloader object with the given options.
566 @param auto_init Whether to load the default extractors and print header (if verbose).
567 Set to 'no_verbose_header' to not print the header
573 self
._ies
_instances
= {}
574 self
._pps
= {k: [] for k in POSTPROCESS_WHEN}
575 self
._printed
_messages
= set()
576 self
._first
_webpage
_request
= True
577 self
._post
_hooks
= []
578 self
._progress
_hooks
= []
579 self
._postprocessor
_hooks
= []
580 self
._download
_retcode
= 0
581 self
._num
_downloads
= 0
583 self
._playlist
_level
= 0
584 self
._playlist
_urls
= set()
585 self
.cache
= Cache(self
)
587 windows_enable_vt_mode()
588 stdout
= sys
.stderr
if self
.params
.get('logtostderr') else sys
.stdout
589 self
._out
_files
= Namespace(
592 screen
=sys
.stderr
if self
.params
.get('quiet') else stdout
,
593 console
=None if compat_os_name
== 'nt' else next(
594 filter(supports_terminal_sequences
, (sys
.stderr
, sys
.stdout
)), None)
596 self
._allow
_colors
= Namespace(**{
597 type_
: not self
.params
.get('no_color') and supports_terminal_sequences(stream
)
598 for type_
, stream
in self
._out
_files
.items_
if type_
!= 'console'
601 # The code is left like this to be reused for future deprecations
602 MIN_SUPPORTED
, MIN_RECOMMENDED
= (3, 7), (3, 7)
603 current_version
= sys
.version_info
[:2]
604 if current_version
< MIN_RECOMMENDED
:
605 msg
= ('Support for Python version %d.%d has been deprecated. '
606 'See https://github.com/yt-dlp/yt-dlp/issues/3764 for more details.'
607 '\n You will no longer receive updates on this version')
608 if current_version
< MIN_SUPPORTED
:
609 msg
= 'Python version %d.%d is no longer supported'
610 self
.deprecation_warning(
611 f
'{msg}! Please update to Python %d.%d or above' % (*current_version
, *MIN_RECOMMENDED
))
613 if self
.params
.get('allow_unplayable_formats'):
615 f
'You have asked for {self._format_err("UNPLAYABLE", self.Styles.EMPHASIS)} formats to be listed/downloaded. '
616 'This is a developer option intended for debugging. \n'
617 ' If you experience any issues while using this option, '
618 f
'{self._format_err("DO NOT", self.Styles.ERROR)} open a bug report')
620 if self
.params
.get('bidi_workaround', False):
623 master
, slave
= pty
.openpty()
624 width
= shutil
.get_terminal_size().columns
625 width_args
= [] if width
is None else ['-w', str(width
)]
626 sp_kwargs
= {'stdin': subprocess.PIPE, 'stdout': slave, 'stderr': self._out_files.error}
628 self
._output
_process
= Popen(['bidiv'] + width_args
, **sp_kwargs
)
630 self
._output
_process
= Popen(['fribidi', '-c', 'UTF-8'] + width_args
, **sp_kwargs
)
631 self
._output
_channel
= os
.fdopen(master
, 'rb')
632 except OSError as ose
:
633 if ose
.errno
== errno
.ENOENT
:
635 'Could not find fribidi executable, ignoring --bidi-workaround. '
636 'Make sure that fribidi is an executable file in one of the directories in your $PATH.')
640 self
.params
['compat_opts'] = set(self
.params
.get('compat_opts', ()))
641 if auto_init
and auto_init
!= 'no_verbose_header':
642 self
.print_debug_header()
644 def check_deprecated(param
, option
, suggestion
):
645 if self
.params
.get(param
) is not None:
646 self
.report_warning(f
'{option} is deprecated. Use {suggestion} instead')
650 if check_deprecated('cn_verification_proxy', '--cn-verification-proxy', '--geo-verification-proxy'):
651 if self
.params
.get('geo_verification_proxy') is None:
652 self
.params
['geo_verification_proxy'] = self
.params
['cn_verification_proxy']
654 check_deprecated('autonumber', '--auto-number', '-o "%(autonumber)s-%(title)s.%(ext)s"')
655 check_deprecated('usetitle', '--title', '-o "%(title)s-%(id)s.%(ext)s"')
656 check_deprecated('useid', '--id', '-o "%(id)s.%(ext)s"')
658 for msg
in self
.params
.get('_warnings', []):
659 self
.report_warning(msg
)
660 for msg
in self
.params
.get('_deprecation_warnings', []):
661 self
.deprecated_feature(msg
)
663 if 'list-formats' in self
.params
['compat_opts']:
664 self
.params
['listformats_table'] = False
666 if 'overwrites' not in self
.params
and self
.params
.get('nooverwrites') is not None:
667 # nooverwrites was unnecessarily changed to overwrites
668 # in 0c3d0f51778b153f65c21906031c2e091fcfb641
669 # This ensures compatibility with both keys
670 self
.params
['overwrites'] = not self
.params
['nooverwrites']
671 elif self
.params
.get('overwrites') is None:
672 self
.params
.pop('overwrites', None)
674 self
.params
['nooverwrites'] = not self
.params
['overwrites']
676 if self
.params
.get('simulate') is None and any((
677 self
.params
.get('list_thumbnails'),
678 self
.params
.get('listformats'),
679 self
.params
.get('listsubtitles'),
681 self
.params
['simulate'] = 'list_only'
683 self
.params
.setdefault('forceprint', {})
684 self
.params
.setdefault('print_to_file', {})
686 # Compatibility with older syntax
687 if not isinstance(params
['forceprint'], dict):
688 self
.params
['forceprint'] = {'video': params['forceprint']}
691 self
.add_default_info_extractors()
693 if (sys
.platform
!= 'win32'
694 and sys
.getfilesystemencoding() in ['ascii', 'ANSI_X3.4-1968']
695 and not self
.params
.get('restrictfilenames', False)):
696 # Unicode filesystem API will throw errors (#1474, #13027)
698 'Assuming --restrict-filenames since file system encoding '
699 'cannot encode all characters. '
700 'Set the LC_ALL environment variable to fix this.')
701 self
.params
['restrictfilenames'] = True
703 self
._parse
_outtmpl
()
705 # Creating format selector here allows us to catch syntax errors before the extraction
706 self
.format_selector
= (
707 self
.params
.get('format') if self
.params
.get('format') in (None, '-')
708 else self
.params
['format'] if callable(self
.params
['format'])
709 else self
.build_format_selector(self
.params
['format']))
711 # Set http_headers defaults according to std_headers
712 self
.params
['http_headers'] = merge_headers(std_headers
, self
.params
.get('http_headers', {}))
715 'post_hooks': self
.add_post_hook
,
716 'progress_hooks': self
.add_progress_hook
,
717 'postprocessor_hooks': self
.add_postprocessor_hook
,
719 for opt
, fn
in hooks
.items():
720 for ph
in self
.params
.get(opt
, []):
723 for pp_def_raw
in self
.params
.get('postprocessors', []):
724 pp_def
= dict(pp_def_raw
)
725 when
= pp_def
.pop('when', 'post_process')
726 self
.add_post_processor(
727 get_postprocessor(pp_def
.pop('key'))(self
, **pp_def
),
731 register_socks_protocols()
733 def preload_download_archive(fn
):
734 """Preload the archive, if any is specified"""
738 elif not is_path_like(fn
):
741 self
.write_debug(f
'Loading archive file {fn!r}')
743 with locked_file(fn
, 'r', encoding
='utf-8') as archive_file
:
744 for line
in archive_file
:
745 archive
.add(line
.strip())
746 except OSError as ioe
:
747 if ioe
.errno
!= errno
.ENOENT
:
751 self
.archive
= preload_download_archive(self
.params
.get('download_archive'))
753 def warn_if_short_id(self
, argv
):
754 # short YouTube ID starting with dash?
756 i
for i
, a
in enumerate(argv
)
757 if re
.match(r
'^-[0-9A-Za-z_-]{10}$', a
)]
761 + [a
for i
, a
in enumerate(argv
) if i
not in idxs
]
762 + ['--'] + [argv
[i
] for i
in idxs
]
765 'Long argument string detected. '
766 'Use -- to separate parameters and URLs, like this:\n%s' %
767 args_to_str(correct_argv
))
769 def add_info_extractor(self
, ie
):
770 """Add an InfoExtractor object to the end of the list."""
772 self
._ies
[ie_key
] = ie
773 if not isinstance(ie
, type):
774 self
._ies
_instances
[ie_key
] = ie
775 ie
.set_downloader(self
)
777 def get_info_extractor(self
, ie_key
):
779 Get an instance of an IE with name ie_key, it will try to get one from
780 the _ies list, if there's no instance it will create a new one and add
781 it to the extractor list.
783 ie
= self
._ies
_instances
.get(ie_key
)
785 ie
= get_info_extractor(ie_key
)()
786 self
.add_info_extractor(ie
)
789 def add_default_info_extractors(self
):
791 Add the InfoExtractors returned by gen_extractors to the end of the list
793 all_ies
= {ie.IE_NAME.lower(): ie for ie in gen_extractor_classes()}
794 all_ies
['end'] = UnsupportedURLIE()
796 ie_names
= orderedSet_from_options(
797 self
.params
.get('allowed_extractors', ['default']), {
798 'all': list(all_ies
),
799 'default': [name
for name
, ie
in all_ies
.items() if ie
._ENABLED
],
801 except re
.error
as e
:
802 raise ValueError(f
'Wrong regex for allowed_extractors: {e.pattern}')
803 for name
in ie_names
:
804 self
.add_info_extractor(all_ies
[name
])
805 self
.write_debug(f
'Loaded {len(ie_names)} extractors')
807 def add_post_processor(self
, pp
, when
='post_process'):
808 """Add a PostProcessor object to the end of the chain."""
809 assert when
in POSTPROCESS_WHEN
, f
'Invalid when={when}'
810 self
._pps
[when
].append(pp
)
811 pp
.set_downloader(self
)
813 def add_post_hook(self
, ph
):
814 """Add the post hook"""
815 self
._post
_hooks
.append(ph
)
817 def add_progress_hook(self
, ph
):
818 """Add the download progress hook"""
819 self
._progress
_hooks
.append(ph
)
821 def add_postprocessor_hook(self
, ph
):
822 """Add the postprocessing progress hook"""
823 self
._postprocessor
_hooks
.append(ph
)
824 for pps
in self
._pps
.values():
826 pp
.add_progress_hook(ph
)
828 def _bidi_workaround(self
, message
):
829 if not hasattr(self
, '_output_channel'):
832 assert hasattr(self
, '_output_process')
833 assert isinstance(message
, str)
834 line_count
= message
.count('\n') + 1
835 self
._output
_process
.stdin
.write((message
+ '\n').encode())
836 self
._output
_process
.stdin
.flush()
837 res
= ''.join(self
._output
_channel
.readline().decode()
838 for _
in range(line_count
))
839 return res
[:-len('\n')]
841 def _write_string(self
, message
, out
=None, only_once
=False):
843 if message
in self
._printed
_messages
:
845 self
._printed
_messages
.add(message
)
846 write_string(message
, out
=out
, encoding
=self
.params
.get('encoding'))
848 def to_stdout(self
, message
, skip_eol
=False, quiet
=None):
849 """Print message to stdout"""
850 if quiet
is not None:
851 self
.deprecation_warning('"YoutubeDL.to_stdout" no longer accepts the argument quiet. '
852 'Use "YoutubeDL.to_screen" instead')
853 if skip_eol
is not False:
854 self
.deprecation_warning('"YoutubeDL.to_stdout" no longer accepts the argument skip_eol. '
855 'Use "YoutubeDL.to_screen" instead')
856 self
._write
_string
(f
'{self._bidi_workaround(message)}\n', self
._out
_files
.out
)
858 def to_screen(self
, message
, skip_eol
=False, quiet
=None, only_once
=False):
859 """Print message to screen if not in quiet mode"""
860 if self
.params
.get('logger'):
861 self
.params
['logger'].debug(message
)
863 if (self
.params
.get('quiet') if quiet
is None else quiet
) and not self
.params
.get('verbose'):
866 '%s%s' % (self
._bidi
_workaround
(message
), ('' if skip_eol
else '\n')),
867 self
._out
_files
.screen
, only_once
=only_once
)
869 def to_stderr(self
, message
, only_once
=False):
870 """Print message to stderr"""
871 assert isinstance(message
, str)
872 if self
.params
.get('logger'):
873 self
.params
['logger'].error(message
)
875 self
._write
_string
(f
'{self._bidi_workaround(message)}\n', self
._out
_files
.error
, only_once
=only_once
)
877 def _send_console_code(self
, code
):
878 if compat_os_name
== 'nt' or not self
._out
_files
.console
:
880 self
._write
_string
(code
, self
._out
_files
.console
)
882 def to_console_title(self
, message
):
883 if not self
.params
.get('consoletitle', False):
885 message
= remove_terminal_sequences(message
)
886 if compat_os_name
== 'nt':
887 if ctypes
.windll
.kernel32
.GetConsoleWindow():
888 # c_wchar_p() might not be necessary if `message` is
889 # already of type unicode()
890 ctypes
.windll
.kernel32
.SetConsoleTitleW(ctypes
.c_wchar_p(message
))
892 self
._send
_console
_code
(f
'\033]0;{message}\007')
894 def save_console_title(self
):
895 if not self
.params
.get('consoletitle') or self
.params
.get('simulate'):
897 self
._send
_console
_code
('\033[22;0t') # Save the title on stack
899 def restore_console_title(self
):
900 if not self
.params
.get('consoletitle') or self
.params
.get('simulate'):
902 self
._send
_console
_code
('\033[23;0t') # Restore the title from stack
905 self
.save_console_title()
908 def __exit__(self
, *args
):
909 self
.restore_console_title()
911 if self
.params
.get('cookiefile') is not None:
912 self
.cookiejar
.save(ignore_discard
=True, ignore_expires
=True)
914 def trouble(self
, message
=None, tb
=None, is_error
=True):
915 """Determine action to take when a download problem appears.
917 Depending on if the downloader has been configured to ignore
918 download errors or not, this method may throw an exception or
919 not when errors are found, after printing the message.
921 @param tb If given, is additional traceback information
922 @param is_error Whether to raise error according to ignorerrors
924 if message
is not None:
925 self
.to_stderr(message
)
926 if self
.params
.get('verbose'):
928 if sys
.exc_info()[0]: # if .trouble has been called from an except block
930 if hasattr(sys
.exc_info()[1], 'exc_info') and sys
.exc_info()[1].exc_info
[0]:
931 tb
+= ''.join(traceback
.format_exception(*sys
.exc_info()[1].exc_info
))
932 tb
+= encode_compat_str(traceback
.format_exc())
934 tb_data
= traceback
.format_list(traceback
.extract_stack())
935 tb
= ''.join(tb_data
)
940 if not self
.params
.get('ignoreerrors'):
941 if sys
.exc_info()[0] and hasattr(sys
.exc_info()[1], 'exc_info') and sys
.exc_info()[1].exc_info
[0]:
942 exc_info
= sys
.exc_info()[1].exc_info
944 exc_info
= sys
.exc_info()
945 raise DownloadError(message
, exc_info
)
946 self
._download
_retcode
= 1
950 EMPHASIS
='light blue',
956 SUPPRESS
='light black',
959 def _format_text(self
, handle
, allow_colors
, text
, f
, fallback
=None, *, test_encoding
=False):
963 # handle.encoding can be None. See https://github.com/yt-dlp/yt-dlp/issues/2711
964 encoding
= self
.params
.get('encoding') or getattr(handle
, 'encoding', None) or 'ascii'
965 text
= text
.encode(encoding
, 'ignore').decode(encoding
)
966 if fallback
is not None and text
!= original_text
:
968 return format_text(text
, f
) if allow_colors
else text
if fallback
is None else fallback
970 def _format_out(self
, *args
, **kwargs
):
971 return self
._format
_text
(self
._out
_files
.out
, self
._allow
_colors
.out
, *args
, **kwargs
)
973 def _format_screen(self
, *args
, **kwargs
):
974 return self
._format
_text
(self
._out
_files
.screen
, self
._allow
_colors
.screen
, *args
, **kwargs
)
976 def _format_err(self
, *args
, **kwargs
):
977 return self
._format
_text
(self
._out
_files
.error
, self
._allow
_colors
.error
, *args
, **kwargs
)
979 def report_warning(self
, message
, only_once
=False):
981 Print the message to stderr, it will be prefixed with 'WARNING:'
982 If stderr is a tty file the 'WARNING:' will be colored
984 if self
.params
.get('logger') is not None:
985 self
.params
['logger'].warning(message
)
987 if self
.params
.get('no_warnings'):
989 self
.to_stderr(f
'{self._format_err("WARNING:", self.Styles.WARNING)} {message}', only_once
)
991 def deprecation_warning(self
, message
, *, stacklevel
=0):
993 message
, stacklevel
=stacklevel
+ 1, printer
=self
.report_error
, is_error
=False)
995 def deprecated_feature(self
, message
):
996 if self
.params
.get('logger') is not None:
997 self
.params
['logger'].warning(f
'Deprecated Feature: {message}')
998 self
.to_stderr(f
'{self._format_err("Deprecated Feature:", self.Styles.ERROR)} {message}', True)
1000 def report_error(self
, message
, *args
, **kwargs
):
1002 Do the same as trouble, but prefixes the message with 'ERROR:', colored
1003 in red if stderr is a tty file.
1005 self
.trouble(f
'{self._format_err("ERROR:", self.Styles.ERROR)} {message}', *args
, **kwargs
)
1007 def write_debug(self
, message
, only_once
=False):
1008 '''Log debug message or Print message to stderr'''
1009 if not self
.params
.get('verbose', False):
1011 message
= f
'[debug] {message}'
1012 if self
.params
.get('logger'):
1013 self
.params
['logger'].debug(message
)
1015 self
.to_stderr(message
, only_once
)
1017 def report_file_already_downloaded(self
, file_name
):
1018 """Report file has already been fully downloaded."""
1020 self
.to_screen('[download] %s has already been downloaded' % file_name
)
1021 except UnicodeEncodeError:
1022 self
.to_screen('[download] The file has already been downloaded')
1024 def report_file_delete(self
, file_name
):
1025 """Report that existing file will be deleted."""
1027 self
.to_screen('Deleting existing file %s' % file_name
)
1028 except UnicodeEncodeError:
1029 self
.to_screen('Deleting existing file')
1031 def raise_no_formats(self
, info
, forced
=False, *, msg
=None):
1032 has_drm
= info
.get('_has_drm')
1033 ignored
, expected
= self
.params
.get('ignore_no_formats_error'), bool(msg
)
1034 msg
= msg
or has_drm
and 'This video is DRM protected' or 'No video formats found!'
1035 if forced
or not ignored
:
1036 raise ExtractorError(msg
, video_id
=info
['id'], ie
=info
['extractor'],
1037 expected
=has_drm
or ignored
or expected
)
1039 self
.report_warning(msg
)
1041 def parse_outtmpl(self
):
1042 self
.deprecation_warning('"YoutubeDL.parse_outtmpl" is deprecated and may be removed in a future version')
1043 self
._parse
_outtmpl
()
1044 return self
.params
['outtmpl']
1046 def _parse_outtmpl(self
):
1048 if self
.params
.get('restrictfilenames'): # Remove spaces in the default template
1049 sanitize
= lambda x
: x
.replace(' - ', ' ').replace(' ', '-')
1051 outtmpl
= self
.params
.setdefault('outtmpl', {})
1052 if not isinstance(outtmpl
, dict):
1053 self
.params
['outtmpl'] = outtmpl
= {'default': outtmpl}
1054 outtmpl
.update({k: sanitize(v) for k, v in DEFAULT_OUTTMPL.items() if outtmpl.get(k) is None}
)
1056 def get_output_path(self
, dir_type
='', filename
=None):
1057 paths
= self
.params
.get('paths', {})
1058 assert isinstance(paths
, dict), '"paths" parameter must be a dictionary'
1059 path
= os
.path
.join(
1060 expand_path(paths
.get('home', '').strip()),
1061 expand_path(paths
.get(dir_type
, '').strip()) if dir_type
else '',
1063 return sanitize_path(path
, force
=self
.params
.get('windowsfilenames'))
1066 def _outtmpl_expandpath(outtmpl
):
1067 # expand_path translates '%%' into '%' and '$$' into '$'
1068 # correspondingly that is not what we want since we need to keep
1069 # '%%' intact for template dict substitution step. Working around
1070 # with boundary-alike separator hack.
1071 sep
= ''.join(random
.choices(ascii_letters
, k
=32))
1072 outtmpl
= outtmpl
.replace('%%', f
'%{sep}%').replace('$$', f
'${sep}$')
1074 # outtmpl should be expand_path'ed before template dict substitution
1075 # because meta fields may contain env variables we don't want to
1076 # be expanded. E.g. for outtmpl "%(title)s.%(ext)s" and
1077 # title "Hello $PATH", we don't want `$PATH` to be expanded.
1078 return expand_path(outtmpl
).replace(sep
, '')
1081 def escape_outtmpl(outtmpl
):
1082 ''' Escape any remaining strings like %s, %abc% etc. '''
1084 STR_FORMAT_RE_TMPL
.format('', '(?![%(\0])'),
1085 lambda mobj
: ('' if mobj
.group('has_key') else '%') + mobj
.group(0),
1089 def validate_outtmpl(cls
, outtmpl
):
1090 ''' @return None or Exception object '''
1092 STR_FORMAT_RE_TMPL
.format('[^)]*', '[ljhqBUDS]'),
1093 lambda mobj
: f
'{mobj.group(0)[:-1]}s',
1094 cls
._outtmpl
_expandpath
(outtmpl
))
1096 cls
.escape_outtmpl(outtmpl
) % collections
.defaultdict(int)
1098 except ValueError as err
:
1102 def _copy_infodict(info_dict
):
1103 info_dict
= dict(info_dict
)
1104 info_dict
.pop('__postprocessors', None)
1105 info_dict
.pop('__pending_error', None)
1108 def prepare_outtmpl(self
, outtmpl
, info_dict
, sanitize
=False):
1109 """ Make the outtmpl and info_dict suitable for substitution: ydl.escape_outtmpl(outtmpl) % info_dict
1110 @param sanitize Whether to sanitize the output as a filename.
1111 For backward compatibility, a function can also be passed
1114 info_dict
.setdefault('epoch', int(time
.time())) # keep epoch consistent once set
1116 info_dict
= self
._copy
_infodict
(info_dict
)
1117 info_dict
['duration_string'] = ( # %(duration>%H-%M-%S)s is wrong if duration > 24hrs
1118 formatSeconds(info_dict
['duration'], '-' if sanitize
else ':')
1119 if info_dict
.get('duration', None) is not None
1121 info_dict
['autonumber'] = int(self
.params
.get('autonumber_start', 1) - 1 + self
._num
_downloads
)
1122 info_dict
['video_autonumber'] = self
._num
_videos
1123 if info_dict
.get('resolution') is None:
1124 info_dict
['resolution'] = self
.format_resolution(info_dict
, default
=None)
1126 # For fields playlist_index, playlist_autonumber and autonumber convert all occurrences
1127 # of %(field)s to %(field)0Nd for backward compatibility
1128 field_size_compat_map
= {
1129 'playlist_index': number_of_digits(info_dict
.get('__last_playlist_index') or 0),
1130 'playlist_autonumber': number_of_digits(info_dict
.get('n_entries') or 0),
1131 'autonumber': self
.params
.get('autonumber_size') or 5,
1135 EXTERNAL_FORMAT_RE
= re
.compile(STR_FORMAT_RE_TMPL
.format('[^)]*', f
'[{STR_FORMAT_TYPES}ljhqBUDS]'))
1140 # Field is of the form key1.key2...
1141 # where keys (except first) can be string, int, slice or "{field, ...}"
1142 FIELD_INNER_RE
= r
'(?:\w+|%(num)s|%(num)s?(?::%(num)s?){1,2})' % {'num': r'(?:-?\d+)'}
1143 FIELD_RE
= r
'\w*(?:\.(?:%(inner)s|{%(field)s(?:,%(field)s)*}))*' % {
1144 'inner': FIELD_INNER_RE
,
1145 'field': rf
'\w*(?:\.{FIELD_INNER_RE})*'
1147 MATH_FIELD_RE
= rf
'(?:{FIELD_RE}|-?{NUMBER_RE})'
1148 MATH_OPERATORS_RE
= r
'(?:%s)' % '|'.join(map(re
.escape
, MATH_FUNCTIONS
.keys()))
1149 INTERNAL_FORMAT_RE
= re
.compile(rf
'''(?x)
1151 (?P<fields>{FIELD_RE})
1152 (?P<maths>(?:{MATH_OPERATORS_RE}{MATH_FIELD_RE})*)
1153 (?:>(?P<strf_format>.+?))?
1155 (?P<alternate>(?<!\\),[^|&)]+)?
1156 (?:&(?P<replacement>.*?))?
1157 (?:\|(?P<default>.*?))?
1160 def _traverse_infodict(fields
):
1161 fields
= [f
for x
in re
.split(r
'\.({.+?})\.?', fields
)
1162 for f
in ([x
] if x
.startswith('{') else x
.split('.'))]
1164 if fields
and not fields
[i
]:
1167 for i
, f
in enumerate(fields
):
1168 if not f
.startswith('{'):
1170 assert f
.endswith('}'), f
'No closing brace for {f} in {fields}'
1171 fields
[i
] = {k: k.split('.') for k in f[1:-1].split(',')}
1173 return traverse_obj(info_dict
, fields
, is_user_input
=True, traverse_string
=True)
1175 def get_value(mdict
):
1177 value
= _traverse_infodict(mdict
['fields'])
1180 value
= float_or_none(value
)
1181 if value
is not None:
1184 offset_key
= mdict
['maths']
1186 value
= float_or_none(value
)
1190 MATH_FIELD_RE
if operator
else MATH_OPERATORS_RE
,
1191 offset_key
).group(0)
1192 offset_key
= offset_key
[len(item
):]
1193 if operator
is None:
1194 operator
= MATH_FUNCTIONS
[item
]
1196 item
, multiplier
= (item
[1:], -1) if item
[0] == '-' else (item
, 1)
1197 offset
= float_or_none(item
)
1199 offset
= float_or_none(_traverse_infodict(item
))
1201 value
= operator(value
, multiplier
* offset
)
1202 except (TypeError, ZeroDivisionError):
1205 # Datetime formatting
1206 if mdict
['strf_format']:
1207 value
= strftime_or_none(value
, mdict
['strf_format'].replace('\\,', ','))
1209 # XXX: Workaround for https://github.com/yt-dlp/yt-dlp/issues/4485
1210 if sanitize
and value
== '':
1214 na
= self
.params
.get('outtmpl_na_placeholder', 'NA')
1216 def filename_sanitizer(key
, value
, restricted
=self
.params
.get('restrictfilenames')):
1217 return sanitize_filename(str(value
), restricted
=restricted
, is_id
=(
1218 bool(re
.search(r
'(^|[_.])id(\.|$)', key
))
1219 if 'filename-sanitization' in self
.params
['compat_opts']
1222 sanitizer
= sanitize
if callable(sanitize
) else filename_sanitizer
1223 sanitize
= bool(sanitize
)
1225 def _dumpjson_default(obj
):
1226 if isinstance(obj
, (set, LazyList
)):
1230 def create_key(outer_mobj
):
1231 if not outer_mobj
.group('has_key'):
1232 return outer_mobj
.group(0)
1233 key
= outer_mobj
.group('key')
1234 mobj
= re
.match(INTERNAL_FORMAT_RE
, key
)
1235 initial_field
= mobj
.group('fields') if mobj
else ''
1236 value
, replacement
, default
= None, None, na
1238 mobj
= mobj
.groupdict()
1239 default
= mobj
['default'] if mobj
['default'] is not None else default
1240 value
= get_value(mobj
)
1241 replacement
= mobj
['replacement']
1242 if value
is None and mobj
['alternate']:
1243 mobj
= re
.match(INTERNAL_FORMAT_RE
, mobj
['remaining'][1:])
1247 fmt
= outer_mobj
.group('format')
1248 if fmt
== 's' and value
is not None and key
in field_size_compat_map
.keys():
1249 fmt
= f
'0{field_size_compat_map[key]:d}d'
1251 value
= default
if value
is None else value
if replacement
is None else replacement
1253 flags
= outer_mobj
.group('conversion') or ''
1254 str_fmt
= f
'{fmt[:-1]}s'
1255 if fmt
[-1] == 'l': # list
1256 delim
= '\n' if '#' in flags
else ', '
1257 value
, fmt
= delim
.join(map(str, variadic(value
, allowed_types
=(str, bytes)))), str_fmt
1258 elif fmt
[-1] == 'j': # json
1259 value
, fmt
= json
.dumps(
1260 value
, default
=_dumpjson_default
,
1261 indent
=4 if '#' in flags
else None, ensure_ascii
='+' not in flags
), str_fmt
1262 elif fmt
[-1] == 'h': # html
1263 value
, fmt
= escapeHTML(str(value
)), str_fmt
1264 elif fmt
[-1] == 'q': # quoted
1265 value
= map(str, variadic(value
) if '#' in flags
else [value
])
1266 value
, fmt
= ' '.join(map(compat_shlex_quote
, value
)), str_fmt
1267 elif fmt
[-1] == 'B': # bytes
1268 value
= f
'%{str_fmt}'.encode() % str(value
).encode()
1269 value
, fmt
= value
.decode('utf-8', 'ignore'), 's'
1270 elif fmt
[-1] == 'U': # unicode normalized
1271 value
, fmt
= unicodedata
.normalize(
1272 # "+" = compatibility equivalence, "#" = NFD
1273 'NF%s%s' % ('K' if '+' in flags
else '', 'D' if '#' in flags
else 'C'),
1275 elif fmt
[-1] == 'D': # decimal suffix
1276 num_fmt
, fmt
= fmt
[:-1].replace('#', ''), 's'
1277 value
= format_decimal_suffix(value
, f
'%{num_fmt}f%s' if num_fmt
else '%d%s',
1278 factor
=1024 if '#' in flags
else 1000)
1279 elif fmt
[-1] == 'S': # filename sanitization
1280 value
, fmt
= filename_sanitizer(initial_field
, value
, restricted
='#' in flags
), str_fmt
1281 elif fmt
[-1] == 'c':
1283 value
= str(value
)[0]
1286 elif fmt
[-1] not in 'rs': # numeric
1287 value
= float_or_none(value
)
1289 value
, fmt
= default
, 's'
1293 # If value is an object, sanitize might convert it to a string
1294 # So we convert it to repr first
1295 value
, fmt
= repr(value
), str_fmt
1296 if fmt
[-1] in 'csr':
1297 value
= sanitizer(initial_field
, value
)
1299 key
= '%s\0%s' % (key
.replace('%', '%\0'), outer_mobj
.group('format'))
1300 TMPL_DICT
[key
] = value
1301 return '{prefix}%({key}){fmt}'.format(key
=key
, fmt
=fmt
, prefix
=outer_mobj
.group('prefix'))
1303 return EXTERNAL_FORMAT_RE
.sub(create_key
, outtmpl
), TMPL_DICT
1305 def evaluate_outtmpl(self
, outtmpl
, info_dict
, *args
, **kwargs
):
1306 outtmpl
, info_dict
= self
.prepare_outtmpl(outtmpl
, info_dict
, *args
, **kwargs
)
1307 return self
.escape_outtmpl(outtmpl
) % info_dict
1309 def _prepare_filename(self
, info_dict
, *, outtmpl
=None, tmpl_type
=None):
1310 assert None in (outtmpl
, tmpl_type
), 'outtmpl and tmpl_type are mutually exclusive'
1312 outtmpl
= self
.params
['outtmpl'].get(tmpl_type
or 'default', self
.params
['outtmpl']['default'])
1314 outtmpl
= self
._outtmpl
_expandpath
(outtmpl
)
1315 filename
= self
.evaluate_outtmpl(outtmpl
, info_dict
, True)
1319 if tmpl_type
in ('', 'temp'):
1320 final_ext
, ext
= self
.params
.get('final_ext'), info_dict
.get('ext')
1321 if final_ext
and ext
and final_ext
!= ext
and filename
.endswith(f
'.{final_ext}'):
1322 filename
= replace_extension(filename
, ext
, final_ext
)
1324 force_ext
= OUTTMPL_TYPES
[tmpl_type
]
1326 filename
= replace_extension(filename
, force_ext
, info_dict
.get('ext'))
1328 # https://github.com/blackjack4494/youtube-dlc/issues/85
1329 trim_file_name
= self
.params
.get('trim_file_name', False)
1331 no_ext
, *ext
= filename
.rsplit('.', 2)
1332 filename
= join_nonempty(no_ext
[:trim_file_name
], *ext
, delim
='.')
1335 except ValueError as err
:
1336 self
.report_error('Error in output template: ' + str(err
) + ' (encoding: ' + repr(preferredencoding()) + ')')
1339 def prepare_filename(self
, info_dict
, dir_type
='', *, outtmpl
=None, warn
=False):
1340 """Generate the output filename"""
1342 assert not dir_type
, 'outtmpl and dir_type are mutually exclusive'
1344 filename
= self
._prepare
_filename
(info_dict
, tmpl_type
=dir_type
, outtmpl
=outtmpl
)
1345 if not filename
and dir_type
not in ('', 'temp'):
1349 if not self
.params
.get('paths'):
1351 elif filename
== '-':
1352 self
.report_warning('--paths is ignored when an outputting to stdout', only_once
=True)
1353 elif os
.path
.isabs(filename
):
1354 self
.report_warning('--paths is ignored since an absolute path is given in output template', only_once
=True)
1355 if filename
== '-' or not filename
:
1358 return self
.get_output_path(dir_type
, filename
)
1360 def _match_entry(self
, info_dict
, incomplete
=False, silent
=False):
1361 """Returns None if the file should be downloaded"""
1362 _type
= info_dict
.get('_type', 'video')
1363 assert incomplete
or _type
== 'video', 'Only video result can be considered complete'
1365 video_title
= info_dict
.get('title', info_dict
.get('id', 'entry'))
1368 if _type
in ('playlist', 'multi_video'):
1370 elif _type
in ('url', 'url_transparent') and not try_call(
1371 lambda: self
.get_info_extractor(info_dict
['ie_key']).is_single_video(info_dict
['url'])):
1374 if 'title' in info_dict
:
1375 # This can happen when we're just evaluating the playlist
1376 title
= info_dict
['title']
1377 matchtitle
= self
.params
.get('matchtitle', False)
1379 if not re
.search(matchtitle
, title
, re
.IGNORECASE
):
1380 return '"' + title
+ '" title did not match pattern "' + matchtitle
+ '"'
1381 rejecttitle
= self
.params
.get('rejecttitle', False)
1383 if re
.search(rejecttitle
, title
, re
.IGNORECASE
):
1384 return '"' + title
+ '" title matched reject pattern "' + rejecttitle
+ '"'
1386 date
= info_dict
.get('upload_date')
1387 if date
is not None:
1388 dateRange
= self
.params
.get('daterange', DateRange())
1389 if date
not in dateRange
:
1390 return f
'{date_from_str(date).isoformat()} upload date is not in range {dateRange}'
1391 view_count
= info_dict
.get('view_count')
1392 if view_count
is not None:
1393 min_views
= self
.params
.get('min_views')
1394 if min_views
is not None and view_count
< min_views
:
1395 return 'Skipping %s, because it has not reached minimum view count (%d/%d)' % (video_title
, view_count
, min_views
)
1396 max_views
= self
.params
.get('max_views')
1397 if max_views
is not None and view_count
> max_views
:
1398 return 'Skipping %s, because it has exceeded the maximum view count (%d/%d)' % (video_title
, view_count
, max_views
)
1399 if age_restricted(info_dict
.get('age_limit'), self
.params
.get('age_limit')):
1400 return 'Skipping "%s" because it is age restricted' % video_title
1402 match_filter
= self
.params
.get('match_filter')
1403 if match_filter
is not None:
1405 ret
= match_filter(info_dict
, incomplete
=incomplete
)
1407 # For backward compatibility
1408 ret
= None if incomplete
else match_filter(info_dict
)
1409 if ret
is NO_DEFAULT
:
1411 filename
= self
._format
_screen
(self
.prepare_filename(info_dict
), self
.Styles
.FILENAME
)
1412 reply
= input(self
._format
_screen
(
1413 f
'Download "{filename}"? (Y/n): ', self
.Styles
.EMPHASIS
)).lower().strip()
1414 if reply
in {'y', ''}
:
1417 return f
'Skipping {video_title}'
1418 elif ret
is not None:
1422 if self
.in_download_archive(info_dict
):
1423 reason
= '%s has already been recorded in the archive' % video_title
1424 break_opt
, break_err
= 'break_on_existing', ExistingVideoReached
1426 reason
= check_filter()
1427 break_opt
, break_err
= 'break_on_reject', RejectedVideoReached
1428 if reason
is not None:
1430 self
.to_screen('[download] ' + reason
)
1431 if self
.params
.get(break_opt
, False):
1436 def add_extra_info(info_dict
, extra_info
):
1437 '''Set the keys from extra_info in info dict if they are missing'''
1438 for key
, value
in extra_info
.items():
1439 info_dict
.setdefault(key
, value
)
1441 def extract_info(self
, url
, download
=True, ie_key
=None, extra_info
=None,
1442 process
=True, force_generic_extractor
=False):
1444 Extract and return the information dictionary of the URL
1447 @param url URL to extract
1450 @param download Whether to download videos
1451 @param process Whether to resolve all unresolved references (URLs, playlist items).
1452 Must be True for download to work
1453 @param ie_key Use only the extractor with this key
1455 @param extra_info Dictionary containing the extra values to add to the info (For internal use only)
1456 @force_generic_extractor Force using the generic extractor (Deprecated; use ie_key='Generic')
1459 if extra_info
is None:
1462 if not ie_key
and force_generic_extractor
:
1466 ies
= {ie_key: self._ies[ie_key]}
if ie_key
in self
._ies
else {}
1470 for key
, ie
in ies
.items():
1471 if not ie
.suitable(url
):
1474 if not ie
.working():
1475 self
.report_warning('The program functionality for this site has been marked as broken, '
1476 'and will probably not work.')
1478 temp_id
= ie
.get_temp_id(url
)
1479 if temp_id
is not None and self
.in_download_archive({'id': temp_id, 'ie_key': key}
):
1480 self
.to_screen(f
'[{key}] {temp_id}: has already been recorded in the archive')
1481 if self
.params
.get('break_on_existing', False):
1482 raise ExistingVideoReached()
1484 return self
.__extract
_info
(url
, self
.get_info_extractor(key
), download
, extra_info
, process
)
1486 extractors_restricted
= self
.params
.get('allowed_extractors') not in (None, ['default'])
1487 self
.report_error(f
'No suitable extractor{format_field(ie_key, None, " (%s)")} found for URL {url}',
1488 tb
=False if extractors_restricted
else None)
1490 def _handle_extraction_exceptions(func
):
1491 @functools.wraps(func
)
1492 def wrapper(self
, *args
, **kwargs
):
1495 return func(self
, *args
, **kwargs
)
1496 except (DownloadCancelled
, LazyList
.IndexError, PagedList
.IndexError):
1498 except ReExtractInfo
as e
:
1500 self
.to_screen(f
'{e}; Re-extracting data')
1502 self
.to_stderr('\r')
1503 self
.report_warning(f
'{e}; Re-extracting data')
1505 except GeoRestrictedError
as e
:
1508 msg
+= '\nThis video is available in %s.' % ', '.join(
1509 map(ISO3166Utils
.short2full
, e
.countries
))
1510 msg
+= '\nYou might want to use a VPN or a proxy server (with --proxy) to workaround.'
1511 self
.report_error(msg
)
1512 except ExtractorError
as e
: # An error we somewhat expected
1513 self
.report_error(str(e
), e
.format_traceback())
1514 except Exception as e
:
1515 if self
.params
.get('ignoreerrors'):
1516 self
.report_error(str(e
), tb
=encode_compat_str(traceback
.format_exc()))
1522 def _wait_for_video(self
, ie_result
={}):
1523 if (not self
.params
.get('wait_for_video')
1524 or ie_result
.get('_type', 'video') != 'video'
1525 or ie_result
.get('formats') or ie_result
.get('url')):
1528 format_dur
= lambda dur
: '%02d:%02d:%02d' % timetuple_from_msec(dur
* 1000)[:-1]
1533 full_msg
= f
'{msg}\n'
1534 if not self
.params
.get('noprogress'):
1535 full_msg
= msg
+ ' ' * (len(last_msg
) - len(msg
)) + '\r'
1538 self
.to_screen(full_msg
, skip_eol
=True)
1541 min_wait
, max_wait
= self
.params
.get('wait_for_video')
1542 diff
= try_get(ie_result
, lambda x
: x
['release_timestamp'] - time
.time())
1543 if diff
is None and ie_result
.get('live_status') == 'is_upcoming':
1544 diff
= round(random
.uniform(min_wait
, max_wait
) if (max_wait
and min_wait
) else (max_wait
or min_wait
), 0)
1545 self
.report_warning('Release time of video is not known')
1546 elif ie_result
and (diff
or 0) <= 0:
1547 self
.report_warning('Video should already be available according to extracted info')
1548 diff
= min(max(diff
or 0, min_wait
or 0), max_wait
or float('inf'))
1549 self
.to_screen(f
'[wait] Waiting for {format_dur(diff)} - Press Ctrl+C to try now')
1551 wait_till
= time
.time() + diff
1554 diff
= wait_till
- time
.time()
1557 raise ReExtractInfo('[wait] Wait period ended', expected
=True)
1558 progress(f
'[wait] Remaining time until next attempt: {self._format_screen(format_dur(diff), self.Styles.EMPHASIS)}')
1560 except KeyboardInterrupt:
1562 raise ReExtractInfo('[wait] Interrupted by user', expected
=True)
1563 except BaseException
as e
:
1564 if not isinstance(e
, ReExtractInfo
):
1568 @_handle_extraction_exceptions
1569 def __extract_info(self
, url
, ie
, download
, extra_info
, process
):
1571 ie_result
= ie
.extract(url
)
1572 except UserNotLive
as e
:
1574 if self
.params
.get('wait_for_video'):
1575 self
.report_warning(e
)
1576 self
._wait
_for
_video
()
1578 if ie_result
is None: # Finished already (backwards compatibility; listformats and friends should be moved here)
1579 self
.report_warning(f
'Extractor {ie.IE_NAME} returned nothing{bug_reports_message()}')
1581 if isinstance(ie_result
, list):
1582 # Backwards compatibility: old IE result format
1584 '_type': 'compat_list',
1585 'entries': ie_result
,
1587 if extra_info
.get('original_url'):
1588 ie_result
.setdefault('original_url', extra_info
['original_url'])
1589 self
.add_default_extra_info(ie_result
, ie
, url
)
1591 self
._wait
_for
_video
(ie_result
)
1592 return self
.process_ie_result(ie_result
, download
, extra_info
)
1596 def add_default_extra_info(self
, ie_result
, ie
, url
):
1598 self
.add_extra_info(ie_result
, {
1600 'original_url': url
,
1602 webpage_url
= ie_result
.get('webpage_url')
1604 self
.add_extra_info(ie_result
, {
1605 'webpage_url_basename': url_basename(webpage_url
),
1606 'webpage_url_domain': get_domain(webpage_url
),
1609 self
.add_extra_info(ie_result
, {
1610 'extractor': ie
.IE_NAME
,
1611 'extractor_key': ie
.ie_key(),
1614 def process_ie_result(self
, ie_result
, download
=True, extra_info
=None):
1616 Take the result of the ie(may be modified) and resolve all unresolved
1617 references (URLs, playlist items).
1619 It will also download the videos if 'download'.
1620 Returns the resolved ie_result.
1622 if extra_info
is None:
1624 result_type
= ie_result
.get('_type', 'video')
1626 if result_type
in ('url', 'url_transparent'):
1627 ie_result
['url'] = sanitize_url(
1628 ie_result
['url'], scheme
='http' if self
.params
.get('prefer_insecure') else 'https')
1629 if ie_result
.get('original_url') and not extra_info
.get('original_url'):
1630 extra_info
= {'original_url': ie_result['original_url'], **extra_info}
1632 extract_flat
= self
.params
.get('extract_flat', False)
1633 if ((extract_flat
== 'in_playlist' and 'playlist' in extra_info
)
1634 or extract_flat
is True):
1635 info_copy
= ie_result
.copy()
1636 ie
= try_get(ie_result
.get('ie_key'), self
.get_info_extractor
)
1637 if ie
and not ie_result
.get('id'):
1638 info_copy
['id'] = ie
.get_temp_id(ie_result
['url'])
1639 self
.add_default_extra_info(info_copy
, ie
, ie_result
['url'])
1640 self
.add_extra_info(info_copy
, extra_info
)
1641 info_copy
, _
= self
.pre_process(info_copy
)
1642 self
._fill
_common
_fields
(info_copy
, False)
1643 self
.__forced
_printings
(info_copy
, self
.prepare_filename(info_copy
), incomplete
=True)
1644 self
._raise
_pending
_errors
(info_copy
)
1645 if self
.params
.get('force_write_download_archive', False):
1646 self
.record_download_archive(info_copy
)
1649 if result_type
== 'video':
1650 self
.add_extra_info(ie_result
, extra_info
)
1651 ie_result
= self
.process_video_result(ie_result
, download
=download
)
1652 self
._raise
_pending
_errors
(ie_result
)
1653 additional_urls
= (ie_result
or {}).get('additional_urls')
1655 # TODO: Improve MetadataParserPP to allow setting a list
1656 if isinstance(additional_urls
, str):
1657 additional_urls
= [additional_urls
]
1659 '[info] %s: %d additional URL(s) requested' % (ie_result
['id'], len(additional_urls
)))
1660 self
.write_debug('Additional URLs: "%s"' % '", "'.join(additional_urls
))
1661 ie_result
['additional_entries'] = [
1663 url
, download
, extra_info
=extra_info
,
1664 force_generic_extractor
=self
.params
.get('force_generic_extractor'))
1665 for url
in additional_urls
1668 elif result_type
== 'url':
1669 # We have to add extra_info to the results because it may be
1670 # contained in a playlist
1671 return self
.extract_info(
1672 ie_result
['url'], download
,
1673 ie_key
=ie_result
.get('ie_key'),
1674 extra_info
=extra_info
)
1675 elif result_type
== 'url_transparent':
1676 # Use the information from the embedding page
1677 info
= self
.extract_info(
1678 ie_result
['url'], ie_key
=ie_result
.get('ie_key'),
1679 extra_info
=extra_info
, download
=False, process
=False)
1681 # extract_info may return None when ignoreerrors is enabled and
1682 # extraction failed with an error, don't crash and return early
1687 exempted_fields
= {'_type', 'url', 'ie_key'}
1688 if not ie_result
.get('section_end') and ie_result
.get('section_start') is None:
1689 # For video clips, the id etc of the clip extractor should be used
1690 exempted_fields |
= {'id', 'extractor', 'extractor_key'}
1692 new_result
= info
.copy()
1693 new_result
.update(filter_dict(ie_result
, lambda k
, v
: v
is not None and k
not in exempted_fields
))
1695 # Extracted info may not be a video result (i.e.
1696 # info.get('_type', 'video') != video) but rather an url or
1697 # url_transparent. In such cases outer metadata (from ie_result)
1698 # should be propagated to inner one (info). For this to happen
1699 # _type of info should be overridden with url_transparent. This
1700 # fixes issue from https://github.com/ytdl-org/youtube-dl/pull/11163.
1701 if new_result
.get('_type') == 'url':
1702 new_result
['_type'] = 'url_transparent'
1704 return self
.process_ie_result(
1705 new_result
, download
=download
, extra_info
=extra_info
)
1706 elif result_type
in ('playlist', 'multi_video'):
1707 # Protect from infinite recursion due to recursively nested playlists
1708 # (see https://github.com/ytdl-org/youtube-dl/issues/27833)
1709 webpage_url
= ie_result
.get('webpage_url') # Playlists maynot have webpage_url
1710 if webpage_url
and webpage_url
in self
._playlist
_urls
:
1712 '[download] Skipping already downloaded playlist: %s'
1713 % ie_result
.get('title') or ie_result
.get('id'))
1716 self
._playlist
_level
+= 1
1717 self
._playlist
_urls
.add(webpage_url
)
1718 self
._fill
_common
_fields
(ie_result
, False)
1719 self
._sanitize
_thumbnails
(ie_result
)
1721 return self
.__process
_playlist
(ie_result
, download
)
1723 self
._playlist
_level
-= 1
1724 if not self
._playlist
_level
:
1725 self
._playlist
_urls
.clear()
1726 elif result_type
== 'compat_list':
1727 self
.report_warning(
1728 'Extractor %s returned a compat_list result. '
1729 'It needs to be updated.' % ie_result
.get('extractor'))
1732 self
.add_extra_info(r
, {
1733 'extractor': ie_result
['extractor'],
1734 'webpage_url': ie_result
['webpage_url'],
1735 'webpage_url_basename': url_basename(ie_result
['webpage_url']),
1736 'webpage_url_domain': get_domain(ie_result
['webpage_url']),
1737 'extractor_key': ie_result
['extractor_key'],
1740 ie_result
['entries'] = [
1741 self
.process_ie_result(_fixup(r
), download
, extra_info
)
1742 for r
in ie_result
['entries']
1746 raise Exception('Invalid result type: %s' % result_type
)
1748 def _ensure_dir_exists(self
, path
):
1749 return make_dir(path
, self
.report_error
)
1752 def _playlist_infodict(ie_result
, strict
=False, **kwargs
):
1754 'playlist_count': ie_result
.get('playlist_count'),
1755 'playlist': ie_result
.get('title') or ie_result
.get('id'),
1756 'playlist_id': ie_result
.get('id'),
1757 'playlist_title': ie_result
.get('title'),
1758 'playlist_uploader': ie_result
.get('uploader'),
1759 'playlist_uploader_id': ie_result
.get('uploader_id'),
1764 if ie_result
.get('webpage_url'):
1766 'webpage_url': ie_result
['webpage_url'],
1767 'webpage_url_basename': url_basename(ie_result
['webpage_url']),
1768 'webpage_url_domain': get_domain(ie_result
['webpage_url']),
1772 'playlist_index': 0,
1773 '__last_playlist_index': max(ie_result
['requested_entries'] or (0, 0)),
1774 'extractor': ie_result
['extractor'],
1775 'extractor_key': ie_result
['extractor_key'],
1778 def __process_playlist(self
, ie_result
, download
):
1779 """Process each entry in the playlist"""
1780 assert ie_result
['_type'] in ('playlist', 'multi_video')
1782 common_info
= self
._playlist
_infodict
(ie_result
, strict
=True)
1783 title
= common_info
.get('playlist') or '<Untitled>'
1784 if self
._match
_entry
(common_info
, incomplete
=True) is not None:
1786 self
.to_screen(f
'[download] Downloading {ie_result["_type"]}: {title}')
1788 all_entries
= PlaylistEntries(self
, ie_result
)
1789 entries
= orderedSet(all_entries
.get_requested_items(), lazy
=True)
1791 lazy
= self
.params
.get('lazy_playlist')
1793 resolved_entries
, n_entries
= [], 'N/A'
1794 ie_result
['requested_entries'], ie_result
['entries'] = None, None
1796 entries
= resolved_entries
= list(entries
)
1797 n_entries
= len(resolved_entries
)
1798 ie_result
['requested_entries'], ie_result
['entries'] = tuple(zip(*resolved_entries
)) or ([], [])
1799 if not ie_result
.get('playlist_count'):
1800 # Better to do this after potentially exhausting entries
1801 ie_result
['playlist_count'] = all_entries
.get_full_count()
1803 extra
= self
._playlist
_infodict
(ie_result
, n_entries
=int_or_none(n_entries
))
1804 ie_copy
= collections
.ChainMap(ie_result
, extra
)
1806 _infojson_written
= False
1807 write_playlist_files
= self
.params
.get('allow_playlist_files', True)
1808 if write_playlist_files
and self
.params
.get('list_thumbnails'):
1809 self
.list_thumbnails(ie_result
)
1810 if write_playlist_files
and not self
.params
.get('simulate'):
1811 _infojson_written
= self
._write
_info
_json
(
1812 'playlist', ie_result
, self
.prepare_filename(ie_copy
, 'pl_infojson'))
1813 if _infojson_written
is None:
1815 if self
._write
_description
('playlist', ie_result
,
1816 self
.prepare_filename(ie_copy
, 'pl_description')) is None:
1818 # TODO: This should be passed to ThumbnailsConvertor if necessary
1819 self
._write
_thumbnails
('playlist', ie_result
, self
.prepare_filename(ie_copy
, 'pl_thumbnail'))
1822 if self
.params
.get('playlistreverse') or self
.params
.get('playlistrandom'):
1823 self
.report_warning('playlistreverse and playlistrandom are not supported with lazy_playlist', only_once
=True)
1824 elif self
.params
.get('playlistreverse'):
1826 elif self
.params
.get('playlistrandom'):
1827 random
.shuffle(entries
)
1829 self
.to_screen(f
'[{ie_result["extractor"]}] Playlist {title}: Downloading {n_entries} items'
1830 f
'{format_field(ie_result, "playlist_count", " of %s")}')
1832 keep_resolved_entries
= self
.params
.get('extract_flat') != 'discard'
1833 if self
.params
.get('extract_flat') == 'discard_in_playlist':
1834 keep_resolved_entries
= ie_result
['_type'] != 'playlist'
1835 if keep_resolved_entries
:
1836 self
.write_debug('The information of all playlist entries will be held in memory')
1839 max_failures
= self
.params
.get('skip_playlist_after_errors') or float('inf')
1840 for i
, (playlist_index
, entry
) in enumerate(entries
):
1842 resolved_entries
.append((playlist_index
, entry
))
1846 entry
['__x_forwarded_for_ip'] = ie_result
.get('__x_forwarded_for_ip')
1847 if not lazy
and 'playlist-index' in self
.params
.get('compat_opts', []):
1848 playlist_index
= ie_result
['requested_entries'][i
]
1850 entry_copy
= collections
.ChainMap(entry
, {
1852 'n_entries': int_or_none(n_entries
),
1853 'playlist_index': playlist_index
,
1854 'playlist_autonumber': i
+ 1,
1857 if self
._match
_entry
(entry_copy
, incomplete
=True) is not None:
1858 # For compatabilty with youtube-dl. See https://github.com/yt-dlp/yt-dlp/issues/4369
1859 resolved_entries
[i
] = (playlist_index
, NO_DEFAULT
)
1862 self
.to_screen('[download] Downloading item %s of %s' % (
1863 self
._format
_screen
(i
+ 1, self
.Styles
.ID
), self
._format
_screen
(n_entries
, self
.Styles
.EMPHASIS
)))
1866 'playlist_index': playlist_index
,
1867 'playlist_autonumber': i
+ 1,
1869 entry_result
= self
.__process
_iterable
_entry
(entry
, download
, extra
)
1870 if not entry_result
:
1872 if failures
>= max_failures
:
1874 f
'Skipping the remaining entries in playlist "{title}" since {failures} items failed extraction')
1876 if keep_resolved_entries
:
1877 resolved_entries
[i
] = (playlist_index
, entry_result
)
1879 # Update with processed data
1880 ie_result
['entries'] = [e
for _
, e
in resolved_entries
if e
is not NO_DEFAULT
]
1881 ie_result
['requested_entries'] = [i
for i
, e
in resolved_entries
if e
is not NO_DEFAULT
]
1882 if ie_result
['requested_entries'] == try_call(lambda: list(range(1, ie_result
['playlist_count'] + 1))):
1883 # Do not set for full playlist
1884 ie_result
.pop('requested_entries')
1886 # Write the updated info to json
1887 if _infojson_written
is True and self
._write
_info
_json
(
1888 'updated playlist', ie_result
,
1889 self
.prepare_filename(ie_copy
, 'pl_infojson'), overwrite
=True) is None:
1892 ie_result
= self
.run_all_pps('playlist', ie_result
)
1893 self
.to_screen(f
'[download] Finished downloading playlist: {title}')
1896 @_handle_extraction_exceptions
1897 def __process_iterable_entry(self
, entry
, download
, extra_info
):
1898 return self
.process_ie_result(
1899 entry
, download
=download
, extra_info
=extra_info
)
1901 def _build_format_filter(self
, filter_spec
):
1902 " Returns a function to filter the formats according to the filter_spec "
1912 operator_rex
= re
.compile(r
'''(?x)\s*
1913 (?P<key>width|height|tbr|abr|vbr|asr|filesize|filesize_approx|fps)\s*
1914 (?P<op>%s)(?P<none_inclusive>\s*\?)?\s*
1915 (?P<value>[0-9.]+(?:[kKmMgGtTpPeEzZyY]i?[Bb]?)?)\s*
1916 ''' % '|'.join(map(re
.escape
, OPERATORS
.keys())))
1917 m
= operator_rex
.fullmatch(filter_spec
)
1920 comparison_value
= int(m
.group('value'))
1922 comparison_value
= parse_filesize(m
.group('value'))
1923 if comparison_value
is None:
1924 comparison_value
= parse_filesize(m
.group('value') + 'B')
1925 if comparison_value
is None:
1927 'Invalid value %r in format specification %r' % (
1928 m
.group('value'), filter_spec
))
1929 op
= OPERATORS
[m
.group('op')]
1934 '^=': lambda attr
, value
: attr
.startswith(value
),
1935 '$=': lambda attr
, value
: attr
.endswith(value
),
1936 '*=': lambda attr
, value
: value
in attr
,
1937 '~=': lambda attr
, value
: value
.search(attr
) is not None
1939 str_operator_rex
= re
.compile(r
'''(?x)\s*
1940 (?P<key>[a-zA-Z0-9._-]+)\s*
1941 (?P<negation>!\s*)?(?P<op>%s)\s*(?P<none_inclusive>\?\s*)?
1943 (?P<value>(?(quote)(?:(?!(?P=quote))[^\\]|\\.)+|[\w.-]+))
1944 (?(quote)(?P=quote))\s*
1945 ''' % '|'.join(map(re
.escape
, STR_OPERATORS
.keys())))
1946 m
= str_operator_rex
.fullmatch(filter_spec
)
1948 if m
.group('op') == '~=':
1949 comparison_value
= re
.compile(m
.group('value'))
1951 comparison_value
= re
.sub(r
'''\\([\\"'])''', r
'\1', m
.group('value'))
1952 str_op
= STR_OPERATORS
[m
.group('op')]
1953 if m
.group('negation'):
1954 op
= lambda attr
, value
: not str_op(attr
, value
)
1959 raise SyntaxError('Invalid filter specification %r' % filter_spec
)
1962 actual_value
= f
.get(m
.group('key'))
1963 if actual_value
is None:
1964 return m
.group('none_inclusive')
1965 return op(actual_value
, comparison_value
)
1968 def _check_formats(self
, formats
):
1970 self
.to_screen('[info] Testing format %s' % f
['format_id'])
1971 path
= self
.get_output_path('temp')
1972 if not self
._ensure
_dir
_exists
(f
'{path}/'):
1974 temp_file
= tempfile
.NamedTemporaryFile(suffix
='.tmp', delete
=False, dir=path
or None)
1977 success
, _
= self
.dl(temp_file
.name
, f
, test
=True)
1978 except (DownloadError
, OSError, ValueError) + network_exceptions
:
1981 if os
.path
.exists(temp_file
.name
):
1983 os
.remove(temp_file
.name
)
1985 self
.report_warning('Unable to delete temporary file "%s"' % temp_file
.name
)
1989 self
.to_screen('[info] Unable to download format %s. Skipping...' % f
['format_id'])
1991 def _default_format_spec(self
, info_dict
, download
=True):
1994 merger
= FFmpegMergerPP(self
)
1995 return merger
.available
and merger
.can_merge()
1998 not self
.params
.get('simulate')
2002 or info_dict
.get('is_live') and not self
.params
.get('live_from_start')
2003 or self
.params
['outtmpl']['default'] == '-'))
2006 or self
.params
.get('allow_multiple_audio_streams', False)
2007 or 'format-spec' in self
.params
['compat_opts'])
2010 'best/bestvideo+bestaudio' if prefer_best
2011 else 'bestvideo*+bestaudio/best' if not compat
2012 else 'bestvideo+bestaudio/best')
2014 def build_format_selector(self
, format_spec
):
2015 def syntax_error(note
, start
):
2017 'Invalid format specification: '
2018 '{}\n\t{}\n\t{}^'.format(note
, format_spec
, ' ' * start
[1]))
2019 return SyntaxError(message
)
2021 PICKFIRST
= 'PICKFIRST'
2025 FormatSelector
= collections
.namedtuple('FormatSelector', ['type', 'selector', 'filters'])
2027 allow_multiple_streams
= {'audio': self
.params
.get('allow_multiple_audio_streams', False),
2028 'video': self
.params
.get('allow_multiple_video_streams', False)}
2030 check_formats
= self
.params
.get('check_formats') == 'selected'
2032 def _parse_filter(tokens
):
2034 for type, string
, start
, _
, _
in tokens
:
2035 if type == tokenize
.OP
and string
== ']':
2036 return ''.join(filter_parts
)
2038 filter_parts
.append(string
)
2040 def _remove_unused_ops(tokens
):
2041 # Remove operators that we don't use and join them with the surrounding strings.
2042 # E.g. 'mp4' '-' 'baseline' '-' '16x9' is converted to 'mp4-baseline-16x9'
2043 ALLOWED_OPS
= ('/', '+', ',', '(', ')')
2044 last_string
, last_start
, last_end
, last_line
= None, None, None, None
2045 for type, string
, start
, end
, line
in tokens
:
2046 if type == tokenize
.OP
and string
== '[':
2048 yield tokenize
.NAME
, last_string
, last_start
, last_end
, last_line
2050 yield type, string
, start
, end
, line
2051 # everything inside brackets will be handled by _parse_filter
2052 for type, string
, start
, end
, line
in tokens
:
2053 yield type, string
, start
, end
, line
2054 if type == tokenize
.OP
and string
== ']':
2056 elif type == tokenize
.OP
and string
in ALLOWED_OPS
:
2058 yield tokenize
.NAME
, last_string
, last_start
, last_end
, last_line
2060 yield type, string
, start
, end
, line
2061 elif type in [tokenize
.NAME
, tokenize
.NUMBER
, tokenize
.OP
]:
2063 last_string
= string
2067 last_string
+= string
2069 yield tokenize
.NAME
, last_string
, last_start
, last_end
, last_line
2071 def _parse_format_selection(tokens
, inside_merge
=False, inside_choice
=False, inside_group
=False):
2073 current_selector
= None
2074 for type, string
, start
, _
, _
in tokens
:
2075 # ENCODING is only defined in python 3.x
2076 if type == getattr(tokenize
, 'ENCODING', None):
2078 elif type in [tokenize
.NAME
, tokenize
.NUMBER
]:
2079 current_selector
= FormatSelector(SINGLE
, string
, [])
2080 elif type == tokenize
.OP
:
2082 if not inside_group
:
2083 # ')' will be handled by the parentheses group
2084 tokens
.restore_last_token()
2086 elif inside_merge
and string
in ['/', ',']:
2087 tokens
.restore_last_token()
2089 elif inside_choice
and string
== ',':
2090 tokens
.restore_last_token()
2093 if not current_selector
:
2094 raise syntax_error('"," must follow a format selector', start
)
2095 selectors
.append(current_selector
)
2096 current_selector
= None
2098 if not current_selector
:
2099 raise syntax_error('"/" must follow a format selector', start
)
2100 first_choice
= current_selector
2101 second_choice
= _parse_format_selection(tokens
, inside_choice
=True)
2102 current_selector
= FormatSelector(PICKFIRST
, (first_choice
, second_choice
), [])
2104 if not current_selector
:
2105 current_selector
= FormatSelector(SINGLE
, 'best', [])
2106 format_filter
= _parse_filter(tokens
)
2107 current_selector
.filters
.append(format_filter
)
2109 if current_selector
:
2110 raise syntax_error('Unexpected "("', start
)
2111 group
= _parse_format_selection(tokens
, inside_group
=True)
2112 current_selector
= FormatSelector(GROUP
, group
, [])
2114 if not current_selector
:
2115 raise syntax_error('Unexpected "+"', start
)
2116 selector_1
= current_selector
2117 selector_2
= _parse_format_selection(tokens
, inside_merge
=True)
2119 raise syntax_error('Expected a selector', start
)
2120 current_selector
= FormatSelector(MERGE
, (selector_1
, selector_2
), [])
2122 raise syntax_error(f
'Operator not recognized: "{string}"', start
)
2123 elif type == tokenize
.ENDMARKER
:
2125 if current_selector
:
2126 selectors
.append(current_selector
)
2129 def _merge(formats_pair
):
2130 format_1
, format_2
= formats_pair
2133 formats_info
.extend(format_1
.get('requested_formats', (format_1
,)))
2134 formats_info
.extend(format_2
.get('requested_formats', (format_2
,)))
2136 if not allow_multiple_streams
['video'] or not allow_multiple_streams
['audio']:
2137 get_no_more
= {'video': False, 'audio': False}
2138 for (i
, fmt_info
) in enumerate(formats_info
):
2139 if fmt_info
.get('acodec') == fmt_info
.get('vcodec') == 'none':
2142 for aud_vid
in ['audio', 'video']:
2143 if not allow_multiple_streams
[aud_vid
] and fmt_info
.get(aud_vid
[0] + 'codec') != 'none':
2144 if get_no_more
[aud_vid
]:
2147 get_no_more
[aud_vid
] = True
2149 if len(formats_info
) == 1:
2150 return formats_info
[0]
2152 video_fmts
= [fmt_info
for fmt_info
in formats_info
if fmt_info
.get('vcodec') != 'none']
2153 audio_fmts
= [fmt_info
for fmt_info
in formats_info
if fmt_info
.get('acodec') != 'none']
2155 the_only_video
= video_fmts
[0] if len(video_fmts
) == 1 else None
2156 the_only_audio
= audio_fmts
[0] if len(audio_fmts
) == 1 else None
2158 output_ext
= get_compatible_ext(
2159 vcodecs
=[f
.get('vcodec') for f
in video_fmts
],
2160 acodecs
=[f
.get('acodec') for f
in audio_fmts
],
2161 vexts
=[f
['ext'] for f
in video_fmts
],
2162 aexts
=[f
['ext'] for f
in audio_fmts
],
2163 preferences
=(try_call(lambda: self
.params
['merge_output_format'].split('/'))
2164 or self
.params
.get('prefer_free_formats') and ('webm', 'mkv')))
2166 filtered
= lambda *keys
: filter(None, (traverse_obj(fmt
, *keys
) for fmt
in formats_info
))
2169 'requested_formats': formats_info
,
2170 'format': '+'.join(filtered('format')),
2171 'format_id': '+'.join(filtered('format_id')),
2173 'protocol': '+'.join(map(determine_protocol
, formats_info
)),
2174 'language': '+'.join(orderedSet(filtered('language'))) or None,
2175 'format_note': '+'.join(orderedSet(filtered('format_note'))) or None,
2176 'filesize_approx': sum(filtered('filesize', 'filesize_approx')) or None,
2177 'tbr': sum(filtered('tbr', 'vbr', 'abr')),
2182 'width': the_only_video
.get('width'),
2183 'height': the_only_video
.get('height'),
2184 'resolution': the_only_video
.get('resolution') or self
.format_resolution(the_only_video
),
2185 'fps': the_only_video
.get('fps'),
2186 'dynamic_range': the_only_video
.get('dynamic_range'),
2187 'vcodec': the_only_video
.get('vcodec'),
2188 'vbr': the_only_video
.get('vbr'),
2189 'stretched_ratio': the_only_video
.get('stretched_ratio'),
2190 'aspect_ratio': the_only_video
.get('aspect_ratio'),
2195 'acodec': the_only_audio
.get('acodec'),
2196 'abr': the_only_audio
.get('abr'),
2197 'asr': the_only_audio
.get('asr'),
2198 'audio_channels': the_only_audio
.get('audio_channels')
2203 def _check_formats(formats
):
2204 if not check_formats
:
2207 yield from self
._check
_formats
(formats
)
2209 def _build_selector_function(selector
):
2210 if isinstance(selector
, list): # ,
2211 fs
= [_build_selector_function(s
) for s
in selector
]
2213 def selector_function(ctx
):
2216 return selector_function
2218 elif selector
.type == GROUP
: # ()
2219 selector_function
= _build_selector_function(selector
.selector
)
2221 elif selector
.type == PICKFIRST
: # /
2222 fs
= [_build_selector_function(s
) for s
in selector
.selector
]
2224 def selector_function(ctx
):
2226 picked_formats
= list(f(ctx
))
2228 return picked_formats
2231 elif selector
.type == MERGE
: # +
2232 selector_1
, selector_2
= map(_build_selector_function
, selector
.selector
)
2234 def selector_function(ctx
):
2235 for pair
in itertools
.product(selector_1(ctx
), selector_2(ctx
)):
2238 elif selector
.type == SINGLE
: # atom
2239 format_spec
= selector
.selector
or 'best'
2241 # TODO: Add allvideo, allaudio etc by generalizing the code with best/worst selector
2242 if format_spec
== 'all':
2243 def selector_function(ctx
):
2244 yield from _check_formats(ctx
['formats'][::-1])
2245 elif format_spec
== 'mergeall':
2246 def selector_function(ctx
):
2247 formats
= list(_check_formats(
2248 f
for f
in ctx
['formats'] if f
.get('vcodec') != 'none' or f
.get('acodec') != 'none'))
2251 merged_format
= formats
[-1]
2252 for f
in formats
[-2::-1]:
2253 merged_format
= _merge((merged_format
, f
))
2257 format_fallback
, seperate_fallback
, format_reverse
, format_idx
= False, None, True, 1
2259 r
'(?P<bw>best|worst|b|w)(?P<type>video|audio|v|a)?(?P<mod>\*)?(?:\.(?P<n>[1-9]\d*))?$',
2261 if mobj
is not None:
2262 format_idx
= int_or_none(mobj
.group('n'), default
=1)
2263 format_reverse
= mobj
.group('bw')[0] == 'b'
2264 format_type
= (mobj
.group('type') or [None])[0]
2265 not_format_type
= {'v': 'a', 'a': 'v'}
.get(format_type
)
2266 format_modified
= mobj
.group('mod') is not None
2268 format_fallback
= not format_type
and not format_modified
# for b, w
2270 (lambda f
: f
.get('%scodec' % format_type
) != 'none')
2271 if format_type
and format_modified
# bv*, ba*, wv*, wa*
2272 else (lambda f
: f
.get('%scodec' % not_format_type
) == 'none')
2273 if format_type
# bv, ba, wv, wa
2274 else (lambda f
: f
.get('vcodec') != 'none' and f
.get('acodec') != 'none')
2275 if not format_modified
# b, w
2276 else lambda f
: True) # b*, w*
2277 filter_f
= lambda f
: _filter_f(f
) and (
2278 f
.get('vcodec') != 'none' or f
.get('acodec') != 'none')
2280 if format_spec
in self
._format
_selection
_exts
['audio']:
2281 filter_f
= lambda f
: f
.get('ext') == format_spec
and f
.get('acodec') != 'none'
2282 elif format_spec
in self
._format
_selection
_exts
['video']:
2283 filter_f
= lambda f
: f
.get('ext') == format_spec
and f
.get('acodec') != 'none' and f
.get('vcodec') != 'none'
2284 seperate_fallback
= lambda f
: f
.get('ext') == format_spec
and f
.get('vcodec') != 'none'
2285 elif format_spec
in self
._format
_selection
_exts
['storyboards']:
2286 filter_f
= lambda f
: f
.get('ext') == format_spec
and f
.get('acodec') == 'none' and f
.get('vcodec') == 'none'
2288 filter_f
= lambda f
: f
.get('format_id') == format_spec
# id
2290 def selector_function(ctx
):
2291 formats
= list(ctx
['formats'])
2292 matches
= list(filter(filter_f
, formats
)) if filter_f
is not None else formats
2294 if format_fallback
and ctx
['incomplete_formats']:
2295 # for extractors with incomplete formats (audio only (soundcloud)
2296 # or video only (imgur)) best/worst will fallback to
2297 # best/worst {video,audio}-only format
2299 elif seperate_fallback
and not ctx
['has_merged_format']:
2300 # for compatibility with youtube-dl when there is no pre-merged format
2301 matches
= list(filter(seperate_fallback
, formats
))
2302 matches
= LazyList(_check_formats(matches
[::-1 if format_reverse
else 1]))
2304 yield matches
[format_idx
- 1]
2305 except LazyList
.IndexError:
2308 filters
= [self
._build
_format
_filter
(f
) for f
in selector
.filters
]
2310 def final_selector(ctx
):
2311 ctx_copy
= dict(ctx
)
2312 for _filter
in filters
:
2313 ctx_copy
['formats'] = list(filter(_filter
, ctx_copy
['formats']))
2314 return selector_function(ctx_copy
)
2315 return final_selector
2317 stream
= io
.BytesIO(format_spec
.encode())
2319 tokens
= list(_remove_unused_ops(tokenize
.tokenize(stream
.readline
)))
2320 except tokenize
.TokenError
:
2321 raise syntax_error('Missing closing/opening brackets or parenthesis', (0, len(format_spec
)))
2323 class TokenIterator
:
2324 def __init__(self
, tokens
):
2325 self
.tokens
= tokens
2332 if self
.counter
>= len(self
.tokens
):
2333 raise StopIteration()
2334 value
= self
.tokens
[self
.counter
]
2340 def restore_last_token(self
):
2343 parsed_selector
= _parse_format_selection(iter(TokenIterator(tokens
)))
2344 return _build_selector_function(parsed_selector
)
2346 def _calc_headers(self
, info_dict
):
2347 res
= merge_headers(self
.params
['http_headers'], info_dict
.get('http_headers') or {})
2349 cookies
= self
._calc
_cookies
(info_dict
['url'])
2351 res
['Cookie'] = cookies
2353 if 'X-Forwarded-For' not in res
:
2354 x_forwarded_for_ip
= info_dict
.get('__x_forwarded_for_ip')
2355 if x_forwarded_for_ip
:
2356 res
['X-Forwarded-For'] = x_forwarded_for_ip
2360 def _calc_cookies(self
, url
):
2361 pr
= sanitized_Request(url
)
2362 self
.cookiejar
.add_cookie_header(pr
)
2363 return pr
.get_header('Cookie')
2365 def _sort_thumbnails(self
, thumbnails
):
2366 thumbnails
.sort(key
=lambda t
: (
2367 t
.get('preference') if t
.get('preference') is not None else -1,
2368 t
.get('width') if t
.get('width') is not None else -1,
2369 t
.get('height') if t
.get('height') is not None else -1,
2370 t
.get('id') if t
.get('id') is not None else '',
2373 def _sanitize_thumbnails(self
, info_dict
):
2374 thumbnails
= info_dict
.get('thumbnails')
2375 if thumbnails
is None:
2376 thumbnail
= info_dict
.get('thumbnail')
2378 info_dict
['thumbnails'] = thumbnails
= [{'url': thumbnail}
]
2382 def check_thumbnails(thumbnails
):
2383 for t
in thumbnails
:
2384 self
.to_screen(f
'[info] Testing thumbnail {t["id"]}')
2386 self
.urlopen(HEADRequest(t
['url']))
2387 except network_exceptions
as err
:
2388 self
.to_screen(f
'[info] Unable to connect to thumbnail {t["id"]} URL {t["url"]!r} - {err}. Skipping...')
2392 self
._sort
_thumbnails
(thumbnails
)
2393 for i
, t
in enumerate(thumbnails
):
2394 if t
.get('id') is None:
2396 if t
.get('width') and t
.get('height'):
2397 t
['resolution'] = '%dx%d' % (t
['width'], t
['height'])
2398 t
['url'] = sanitize_url(t
['url'])
2400 if self
.params
.get('check_formats') is True:
2401 info_dict
['thumbnails'] = LazyList(check_thumbnails(thumbnails
[::-1]), reverse
=True)
2403 info_dict
['thumbnails'] = thumbnails
2405 def _fill_common_fields(self
, info_dict
, final
=True):
2406 # TODO: move sanitization here
2408 title
= info_dict
.get('title', NO_DEFAULT
)
2409 if title
is NO_DEFAULT
:
2410 raise ExtractorError('Missing "title" field in extractor result',
2411 video_id
=info_dict
['id'], ie
=info_dict
['extractor'])
2412 info_dict
['fulltitle'] = title
2415 self
.write_debug('Extractor gave empty title. Creating a generic title')
2417 self
.report_warning('Extractor failed to obtain "title". Creating a generic title instead')
2418 info_dict
['title'] = f
'{info_dict["extractor"].replace(":", "-")} video #{info_dict["id"]}'
2420 if info_dict
.get('duration') is not None:
2421 info_dict
['duration_string'] = formatSeconds(info_dict
['duration'])
2423 for ts_key
, date_key
in (
2424 ('timestamp', 'upload_date'),
2425 ('release_timestamp', 'release_date'),
2426 ('modified_timestamp', 'modified_date'),
2428 if info_dict
.get(date_key
) is None and info_dict
.get(ts_key
) is not None:
2429 # Working around out-of-range timestamp values (e.g. negative ones on Windows,
2430 # see http://bugs.python.org/issue1646728)
2431 with contextlib
.suppress(ValueError, OverflowError, OSError):
2432 upload_date
= datetime
.datetime
.utcfromtimestamp(info_dict
[ts_key
])
2433 info_dict
[date_key
] = upload_date
.strftime('%Y%m%d')
2435 live_keys
= ('is_live', 'was_live')
2436 live_status
= info_dict
.get('live_status')
2437 if live_status
is None:
2438 for key
in live_keys
:
2439 if info_dict
.get(key
) is False:
2441 if info_dict
.get(key
):
2444 if all(info_dict
.get(key
) is False for key
in live_keys
):
2445 live_status
= 'not_live'
2447 info_dict
['live_status'] = live_status
2448 for key
in live_keys
:
2449 if info_dict
.get(key
) is None:
2450 info_dict
[key
] = (live_status
== key
)
2451 if live_status
== 'post_live':
2452 info_dict
['was_live'] = True
2454 # Auto generate title fields corresponding to the *_number fields when missing
2455 # in order to always have clean titles. This is very common for TV series.
2456 for field
in ('chapter', 'season', 'episode'):
2457 if final
and info_dict
.get('%s_number' % field
) is not None and not info_dict
.get(field
):
2458 info_dict
[field
] = '%s %d' % (field
.capitalize(), info_dict
['%s_number' % field
])
2460 def _raise_pending_errors(self
, info
):
2461 err
= info
.pop('__pending_error', None)
2463 self
.report_error(err
, tb
=False)
2465 def sort_formats(self
, info_dict
):
2466 formats
= self
._get
_formats
(info_dict
)
2469 # Backward compatibility with InfoExtractor._sort_formats
2470 field_preference
= formats
[0].pop('__sort_fields', None)
2471 if field_preference
:
2472 info_dict
['_format_sort_fields'] = field_preference
2474 formats
.sort(key
=FormatSorter(
2475 self
, info_dict
.get('_format_sort_fields', [])).calculate_preference
)
2477 def process_video_result(self
, info_dict
, download
=True):
2478 assert info_dict
.get('_type', 'video') == 'video'
2479 self
._num
_videos
+= 1
2481 if 'id' not in info_dict
:
2482 raise ExtractorError('Missing "id" field in extractor result', ie
=info_dict
['extractor'])
2483 elif not info_dict
.get('id'):
2484 raise ExtractorError('Extractor failed to obtain "id"', ie
=info_dict
['extractor'])
2486 def report_force_conversion(field
, field_not
, conversion
):
2487 self
.report_warning(
2488 '"%s" field is not %s - forcing %s conversion, there is an error in extractor'
2489 % (field
, field_not
, conversion
))
2491 def sanitize_string_field(info
, string_field
):
2492 field
= info
.get(string_field
)
2493 if field
is None or isinstance(field
, str):
2495 report_force_conversion(string_field
, 'a string', 'string')
2496 info
[string_field
] = str(field
)
2498 def sanitize_numeric_fields(info
):
2499 for numeric_field
in self
._NUMERIC
_FIELDS
:
2500 field
= info
.get(numeric_field
)
2501 if field
is None or isinstance(field
, (int, float)):
2503 report_force_conversion(numeric_field
, 'numeric', 'int')
2504 info
[numeric_field
] = int_or_none(field
)
2506 sanitize_string_field(info_dict
, 'id')
2507 sanitize_numeric_fields(info_dict
)
2508 if info_dict
.get('section_end') and info_dict
.get('section_start') is not None:
2509 info_dict
['duration'] = round(info_dict
['section_end'] - info_dict
['section_start'], 3)
2510 if (info_dict
.get('duration') or 0) <= 0 and info_dict
.pop('duration', None):
2511 self
.report_warning('"duration" field is negative, there is an error in extractor')
2513 chapters
= info_dict
.get('chapters') or []
2514 if chapters
and chapters
[0].get('start_time'):
2515 chapters
.insert(0, {'start_time': 0}
)
2517 dummy_chapter
= {'end_time': 0, 'start_time': info_dict.get('duration')}
2518 for idx
, (prev
, current
, next_
) in enumerate(zip(
2519 (dummy_chapter
, *chapters
), chapters
, (*chapters
[1:], dummy_chapter
)), 1):
2520 if current
.get('start_time') is None:
2521 current
['start_time'] = prev
.get('end_time')
2522 if not current
.get('end_time'):
2523 current
['end_time'] = next_
.get('start_time')
2524 if not current
.get('title'):
2525 current
['title'] = f
'<Untitled Chapter {idx}>'
2527 if 'playlist' not in info_dict
:
2528 # It isn't part of a playlist
2529 info_dict
['playlist'] = None
2530 info_dict
['playlist_index'] = None
2532 self
._sanitize
_thumbnails
(info_dict
)
2534 thumbnail
= info_dict
.get('thumbnail')
2535 thumbnails
= info_dict
.get('thumbnails')
2537 info_dict
['thumbnail'] = sanitize_url(thumbnail
)
2539 info_dict
['thumbnail'] = thumbnails
[-1]['url']
2541 if info_dict
.get('display_id') is None and 'id' in info_dict
:
2542 info_dict
['display_id'] = info_dict
['id']
2544 self
._fill
_common
_fields
(info_dict
)
2546 for cc_kind
in ('subtitles', 'automatic_captions'):
2547 cc
= info_dict
.get(cc_kind
)
2549 for _
, subtitle
in cc
.items():
2550 for subtitle_format
in subtitle
:
2551 if subtitle_format
.get('url'):
2552 subtitle_format
['url'] = sanitize_url(subtitle_format
['url'])
2553 if subtitle_format
.get('ext') is None:
2554 subtitle_format
['ext'] = determine_ext(subtitle_format
['url']).lower()
2556 automatic_captions
= info_dict
.get('automatic_captions')
2557 subtitles
= info_dict
.get('subtitles')
2559 info_dict
['requested_subtitles'] = self
.process_subtitles(
2560 info_dict
['id'], subtitles
, automatic_captions
)
2562 self
.sort_formats(info_dict
)
2563 formats
= self
._get
_formats
(info_dict
)
2565 # or None ensures --clean-infojson removes it
2566 info_dict
['_has_drm'] = any(f
.get('has_drm') for f
in formats
) or None
2567 if not self
.params
.get('allow_unplayable_formats'):
2568 formats
= [f
for f
in formats
if not f
.get('has_drm')]
2570 if formats
and all(f
.get('acodec') == f
.get('vcodec') == 'none' for f
in formats
):
2571 self
.report_warning(
2572 f
'{"This video is DRM protected and " if info_dict["_has_drm"] else ""}'
2573 'only images are available for download. Use --list-formats to see them'.capitalize())
2575 get_from_start
= not info_dict
.get('is_live') or bool(self
.params
.get('live_from_start'))
2576 if not get_from_start
:
2577 info_dict
['title'] += ' ' + datetime
.datetime
.now().strftime('%Y-%m-%d %H:%M')
2578 if info_dict
.get('is_live') and formats
:
2579 formats
= [f
for f
in formats
if bool(f
.get('is_from_start')) == get_from_start
]
2580 if get_from_start
and not formats
:
2581 self
.raise_no_formats(info_dict
, msg
=(
2582 '--live-from-start is passed, but there are no formats that can be downloaded from the start. '
2583 'If you want to download from the current time, use --no-live-from-start'))
2585 def is_wellformed(f
):
2588 self
.report_warning(
2589 '"url" field is missing or empty - skipping format, '
2590 'there is an error in extractor')
2592 if isinstance(url
, bytes):
2593 sanitize_string_field(f
, 'url')
2596 # Filter out malformed formats for better extraction robustness
2597 formats
= list(filter(is_wellformed
, formats
or []))
2600 self
.raise_no_formats(info_dict
)
2604 # We check that all the formats have the format and format_id fields
2605 for i
, format
in enumerate(formats
):
2606 sanitize_string_field(format
, 'format_id')
2607 sanitize_numeric_fields(format
)
2608 format
['url'] = sanitize_url(format
['url'])
2609 if not format
.get('format_id'):
2610 format
['format_id'] = str(i
)
2612 # Sanitize format_id from characters used in format selector expression
2613 format
['format_id'] = re
.sub(r
'[\s,/+\[\]()]', '_', format
['format_id'])
2614 format_id
= format
['format_id']
2615 if format_id
not in formats_dict
:
2616 formats_dict
[format_id
] = []
2617 formats_dict
[format_id
].append(format
)
2619 # Make sure all formats have unique format_id
2620 common_exts
= set(itertools
.chain(*self
._format
_selection
_exts
.values()))
2621 for format_id
, ambiguous_formats
in formats_dict
.items():
2622 ambigious_id
= len(ambiguous_formats
) > 1
2623 for i
, format
in enumerate(ambiguous_formats
):
2625 format
['format_id'] = '%s-%d' % (format_id
, i
)
2626 if format
.get('ext') is None:
2627 format
['ext'] = determine_ext(format
['url']).lower()
2628 # Ensure there is no conflict between id and ext in format selection
2629 # See https://github.com/yt-dlp/yt-dlp/issues/1282
2630 if format
['format_id'] != format
['ext'] and format
['format_id'] in common_exts
:
2631 format
['format_id'] = 'f%s' % format
['format_id']
2633 for i
, format
in enumerate(formats
):
2634 if format
.get('format') is None:
2635 format
['format'] = '{id} - {res}{note}'.format(
2636 id=format
['format_id'],
2637 res
=self
.format_resolution(format
),
2638 note
=format_field(format
, 'format_note', ' (%s)'),
2640 if format
.get('protocol') is None:
2641 format
['protocol'] = determine_protocol(format
)
2642 if format
.get('resolution') is None:
2643 format
['resolution'] = self
.format_resolution(format
, default
=None)
2644 if format
.get('dynamic_range') is None and format
.get('vcodec') != 'none':
2645 format
['dynamic_range'] = 'SDR'
2646 if format
.get('aspect_ratio') is None:
2647 format
['aspect_ratio'] = try_call(lambda: round(format
['width'] / format
['height'], 2))
2648 if (info_dict
.get('duration') and format
.get('tbr')
2649 and not format
.get('filesize') and not format
.get('filesize_approx')):
2650 format
['filesize_approx'] = int(info_dict
['duration'] * format
['tbr'] * (1024 / 8))
2652 # Add HTTP headers, so that external programs can use them from the
2654 full_format_info
= info_dict
.copy()
2655 full_format_info
.update(format
)
2656 format
['http_headers'] = self
._calc
_headers
(full_format_info
)
2657 # Remove private housekeeping stuff
2658 if '__x_forwarded_for_ip' in info_dict
:
2659 del info_dict
['__x_forwarded_for_ip']
2661 if self
.params
.get('check_formats') is True:
2662 formats
= LazyList(self
._check
_formats
(formats
[::-1]), reverse
=True)
2664 if not formats
or formats
[0] is not info_dict
:
2665 # only set the 'formats' fields if the original info_dict list them
2666 # otherwise we end up with a circular reference, the first (and unique)
2667 # element in the 'formats' field in info_dict is info_dict itself,
2668 # which can't be exported to json
2669 info_dict
['formats'] = formats
2671 info_dict
, _
= self
.pre_process(info_dict
)
2673 if self
._match
_entry
(info_dict
, incomplete
=self
._format
_fields
) is not None:
2676 self
.post_extract(info_dict
)
2677 info_dict
, _
= self
.pre_process(info_dict
, 'after_filter')
2679 # The pre-processors may have modified the formats
2680 formats
= self
._get
_formats
(info_dict
)
2682 list_only
= self
.params
.get('simulate') == 'list_only'
2683 interactive_format_selection
= not list_only
and self
.format_selector
== '-'
2684 if self
.params
.get('list_thumbnails'):
2685 self
.list_thumbnails(info_dict
)
2686 if self
.params
.get('listsubtitles'):
2687 if 'automatic_captions' in info_dict
:
2688 self
.list_subtitles(
2689 info_dict
['id'], automatic_captions
, 'automatic captions')
2690 self
.list_subtitles(info_dict
['id'], subtitles
, 'subtitles')
2691 if self
.params
.get('listformats') or interactive_format_selection
:
2692 self
.list_formats(info_dict
)
2694 # Without this printing, -F --print-json will not work
2695 self
.__forced
_printings
(info_dict
, self
.prepare_filename(info_dict
), incomplete
=True)
2698 format_selector
= self
.format_selector
2699 if format_selector
is None:
2700 req_format
= self
._default
_format
_spec
(info_dict
, download
=download
)
2701 self
.write_debug('Default format spec: %s' % req_format
)
2702 format_selector
= self
.build_format_selector(req_format
)
2705 if interactive_format_selection
:
2707 self
._format
_screen
('\nEnter format selector: ', self
.Styles
.EMPHASIS
))
2709 format_selector
= self
.build_format_selector(req_format
)
2710 except SyntaxError as err
:
2711 self
.report_error(err
, tb
=False, is_error
=False)
2714 formats_to_download
= list(format_selector({
2716 'has_merged_format': any('none' not in (f
.get('acodec'), f
.get('vcodec')) for f
in formats
),
2717 'incomplete_formats': (
2718 # All formats are video-only or
2719 all(f
.get('vcodec') != 'none' and f
.get('acodec') == 'none' for f
in formats
)
2720 # all formats are audio-only
2721 or all(f
.get('vcodec') == 'none' and f
.get('acodec') != 'none' for f
in formats
)),
2723 if interactive_format_selection
and not formats_to_download
:
2724 self
.report_error('Requested format is not available', tb
=False, is_error
=False)
2728 if not formats_to_download
:
2729 if not self
.params
.get('ignore_no_formats_error'):
2730 raise ExtractorError(
2731 'Requested format is not available. Use --list-formats for a list of available formats',
2732 expected
=True, video_id
=info_dict
['id'], ie
=info_dict
['extractor'])
2733 self
.report_warning('Requested format is not available')
2734 # Process what we can, even without any available formats.
2735 formats_to_download
= [{}]
2737 requested_ranges
= tuple(self
.params
.get('download_ranges', lambda *_
: [{}])(info_dict
, self
))
2738 best_format
, downloaded_formats
= formats_to_download
[-1], []
2740 if best_format
and requested_ranges
:
2741 def to_screen(*msg
):
2742 self
.to_screen(f
'[info] {info_dict["id"]}: {" ".join(", ".join(variadic(m)) for m in msg)}')
2744 to_screen(f
'Downloading {len(formats_to_download)} format(s):',
2745 (f
['format_id'] for f
in formats_to_download
))
2746 if requested_ranges
!= ({}, ):
2747 to_screen(f
'Downloading {len(requested_ranges)} time ranges:',
2748 (f
'{c["start_time"]:.1f}-{c["end_time"]:.1f}' for c
in requested_ranges
))
2749 max_downloads_reached
= False
2751 for fmt
, chapter
in itertools
.product(formats_to_download
, requested_ranges
):
2752 new_info
= self
._copy
_infodict
(info_dict
)
2753 new_info
.update(fmt
)
2754 offset
, duration
= info_dict
.get('section_start') or 0, info_dict
.get('duration') or float('inf')
2755 end_time
= offset
+ min(chapter
.get('end_time', duration
), duration
)
2756 if chapter
or offset
:
2758 'section_start': offset
+ chapter
.get('start_time', 0),
2759 # duration may not be accurate. So allow deviations <1sec
2760 'section_end': end_time
if end_time
<= offset
+ duration
+ 1 else None,
2761 'section_title': chapter
.get('title'),
2762 'section_number': chapter
.get('index'),
2764 downloaded_formats
.append(new_info
)
2766 self
.process_info(new_info
)
2767 except MaxDownloadsReached
:
2768 max_downloads_reached
= True
2769 self
._raise
_pending
_errors
(new_info
)
2770 # Remove copied info
2771 for key
, val
in tuple(new_info
.items()):
2772 if info_dict
.get(key
) == val
:
2774 if max_downloads_reached
:
2777 write_archive
= {f.get('__write_download_archive', False) for f in downloaded_formats}
2778 assert write_archive
.issubset({True, False, 'ignore'}
)
2779 if True in write_archive
and False not in write_archive
:
2780 self
.record_download_archive(info_dict
)
2782 info_dict
['requested_downloads'] = downloaded_formats
2783 info_dict
= self
.run_all_pps('after_video', info_dict
)
2784 if max_downloads_reached
:
2785 raise MaxDownloadsReached()
2787 # We update the info dict with the selected best quality format (backwards compatibility)
2788 info_dict
.update(best_format
)
2791 def process_subtitles(self
, video_id
, normal_subtitles
, automatic_captions
):
2792 """Select the requested subtitles and their format"""
2793 available_subs
, normal_sub_langs
= {}, []
2794 if normal_subtitles
and self
.params
.get('writesubtitles'):
2795 available_subs
.update(normal_subtitles
)
2796 normal_sub_langs
= tuple(normal_subtitles
.keys())
2797 if automatic_captions
and self
.params
.get('writeautomaticsub'):
2798 for lang
, cap_info
in automatic_captions
.items():
2799 if lang
not in available_subs
:
2800 available_subs
[lang
] = cap_info
2802 if not available_subs
or (
2803 not self
.params
.get('writesubtitles')
2804 and not self
.params
.get('writeautomaticsub')):
2807 all_sub_langs
= tuple(available_subs
.keys())
2808 if self
.params
.get('allsubtitles', False):
2809 requested_langs
= all_sub_langs
2810 elif self
.params
.get('subtitleslangs', False):
2812 requested_langs
= orderedSet_from_options(
2813 self
.params
.get('subtitleslangs'), {'all': all_sub_langs}
, use_regex
=True)
2814 except re
.error
as e
:
2815 raise ValueError(f
'Wrong regex for subtitlelangs: {e.pattern}')
2816 elif normal_sub_langs
:
2817 requested_langs
= ['en'] if 'en' in normal_sub_langs
else normal_sub_langs
[:1]
2819 requested_langs
= ['en'] if 'en' in all_sub_langs
else all_sub_langs
[:1]
2821 self
.to_screen(f
'[info] {video_id}: Downloading subtitles: {", ".join(requested_langs)}')
2823 formats_query
= self
.params
.get('subtitlesformat', 'best')
2824 formats_preference
= formats_query
.split('/') if formats_query
else []
2826 for lang
in requested_langs
:
2827 formats
= available_subs
.get(lang
)
2829 self
.report_warning(f
'{lang} subtitles not available for {video_id}')
2831 for ext
in formats_preference
:
2835 matches
= list(filter(lambda f
: f
['ext'] == ext
, formats
))
2841 self
.report_warning(
2842 'No subtitle format found matching "%s" for language %s, '
2843 'using %s' % (formats_query
, lang
, f
['ext']))
2847 def _forceprint(self
, key
, info_dict
):
2848 if info_dict
is None:
2850 info_copy
= info_dict
.copy()
2851 info_copy
['formats_table'] = self
.render_formats_table(info_dict
)
2852 info_copy
['thumbnails_table'] = self
.render_thumbnails_table(info_dict
)
2853 info_copy
['subtitles_table'] = self
.render_subtitles_table(info_dict
.get('id'), info_dict
.get('subtitles'))
2854 info_copy
['automatic_captions_table'] = self
.render_subtitles_table(info_dict
.get('id'), info_dict
.get('automatic_captions'))
2856 def format_tmpl(tmpl
):
2857 mobj
= re
.fullmatch(r
'([\w.:,]|-\d|(?P<dict>{([\w.:,]|-\d)+}))+=?', tmpl
)
2862 if tmpl
.startswith('{'):
2864 if tmpl
.endswith('='):
2865 tmpl
, fmt
= tmpl
[:-1], '{0} = %({0})#j'
2866 return '\n'.join(map(fmt
.format
, [tmpl
] if mobj
.group('dict') else tmpl
.split(',')))
2868 for tmpl
in self
.params
['forceprint'].get(key
, []):
2869 self
.to_stdout(self
.evaluate_outtmpl(format_tmpl(tmpl
), info_copy
))
2871 for tmpl
, file_tmpl
in self
.params
['print_to_file'].get(key
, []):
2872 filename
= self
.prepare_filename(info_dict
, outtmpl
=file_tmpl
)
2873 tmpl
= format_tmpl(tmpl
)
2874 self
.to_screen(f
'[info] Writing {tmpl!r} to: {filename}')
2875 if self
._ensure
_dir
_exists
(filename
):
2876 with open(filename
, 'a', encoding
='utf-8') as f
:
2877 f
.write(self
.evaluate_outtmpl(tmpl
, info_copy
) + '\n')
2879 def __forced_printings(self
, info_dict
, filename
, incomplete
):
2880 def print_mandatory(field
, actual_field
=None):
2881 if actual_field
is None:
2882 actual_field
= field
2883 if (self
.params
.get('force%s' % field
, False)
2884 and (not incomplete
or info_dict
.get(actual_field
) is not None)):
2885 self
.to_stdout(info_dict
[actual_field
])
2887 def print_optional(field
):
2888 if (self
.params
.get('force%s' % field
, False)
2889 and info_dict
.get(field
) is not None):
2890 self
.to_stdout(info_dict
[field
])
2892 info_dict
= info_dict
.copy()
2893 if filename
is not None:
2894 info_dict
['filename'] = filename
2895 if info_dict
.get('requested_formats') is not None:
2896 # For RTMP URLs, also include the playpath
2897 info_dict
['urls'] = '\n'.join(f
['url'] + f
.get('play_path', '') for f
in info_dict
['requested_formats'])
2898 elif info_dict
.get('url'):
2899 info_dict
['urls'] = info_dict
['url'] + info_dict
.get('play_path', '')
2901 if (self
.params
.get('forcejson')
2902 or self
.params
['forceprint'].get('video')
2903 or self
.params
['print_to_file'].get('video')):
2904 self
.post_extract(info_dict
)
2905 self
._forceprint
('video', info_dict
)
2907 print_mandatory('title')
2908 print_mandatory('id')
2909 print_mandatory('url', 'urls')
2910 print_optional('thumbnail')
2911 print_optional('description')
2912 print_optional('filename')
2913 if self
.params
.get('forceduration') and info_dict
.get('duration') is not None:
2914 self
.to_stdout(formatSeconds(info_dict
['duration']))
2915 print_mandatory('format')
2917 if self
.params
.get('forcejson'):
2918 self
.to_stdout(json
.dumps(self
.sanitize_info(info_dict
)))
2920 def dl(self
, name
, info
, subtitle
=False, test
=False):
2921 if not info
.get('url'):
2922 self
.raise_no_formats(info
, True)
2925 verbose
= self
.params
.get('verbose')
2928 'quiet': self
.params
.get('quiet') or not verbose
,
2930 'noprogress': not verbose
,
2932 'skip_unavailable_fragments': False,
2933 'keep_fragments': False,
2935 '_no_ytdl_file': True,
2938 params
= self
.params
2939 fd
= get_suitable_downloader(info
, params
, to_stdout
=(name
== '-'))(self
, params
)
2941 for ph
in self
._progress
_hooks
:
2942 fd
.add_progress_hook(ph
)
2944 (f
['url'].split(',')[0] + ',<data>' if f
['url'].startswith('data:') else f
['url'])
2945 for f
in info
.get('requested_formats', []) or [info
])
2946 self
.write_debug(f
'Invoking {fd.FD_NAME} downloader on "{urls}"')
2948 # Note: Ideally info should be a deep-copied so that hooks cannot modify it.
2949 # But it may contain objects that are not deep-copyable
2950 new_info
= self
._copy
_infodict
(info
)
2951 if new_info
.get('http_headers') is None:
2952 new_info
['http_headers'] = self
._calc
_headers
(new_info
)
2953 return fd
.download(name
, new_info
, subtitle
)
2955 def existing_file(self
, filepaths
, *, default_overwrite
=True):
2956 existing_files
= list(filter(os
.path
.exists
, orderedSet(filepaths
)))
2957 if existing_files
and not self
.params
.get('overwrites', default_overwrite
):
2958 return existing_files
[0]
2960 for file in existing_files
:
2961 self
.report_file_delete(file)
2965 def process_info(self
, info_dict
):
2966 """Process a single resolved IE result. (Modifies it in-place)"""
2968 assert info_dict
.get('_type', 'video') == 'video'
2969 original_infodict
= info_dict
2971 if 'format' not in info_dict
and 'ext' in info_dict
:
2972 info_dict
['format'] = info_dict
['ext']
2974 if self
._match
_entry
(info_dict
) is not None:
2975 info_dict
['__write_download_archive'] = 'ignore'
2978 # Does nothing under normal operation - for backward compatibility of process_info
2979 self
.post_extract(info_dict
)
2981 def replace_info_dict(new_info
):
2983 if new_info
== info_dict
:
2986 info_dict
.update(new_info
)
2988 new_info
, _
= self
.pre_process(info_dict
, 'video')
2989 replace_info_dict(new_info
)
2990 self
._num
_downloads
+= 1
2992 # info_dict['_filename'] needs to be set for backward compatibility
2993 info_dict
['_filename'] = full_filename
= self
.prepare_filename(info_dict
, warn
=True)
2994 temp_filename
= self
.prepare_filename(info_dict
, 'temp')
2998 self
.__forced
_printings
(info_dict
, full_filename
, incomplete
=('format' not in info_dict
))
3000 def check_max_downloads():
3001 if self
._num
_downloads
>= float(self
.params
.get('max_downloads') or 'inf'):
3002 raise MaxDownloadsReached()
3004 if self
.params
.get('simulate'):
3005 info_dict
['__write_download_archive'] = self
.params
.get('force_write_download_archive')
3006 check_max_downloads()
3009 if full_filename
is None:
3011 if not self
._ensure
_dir
_exists
(encodeFilename(full_filename
)):
3013 if not self
._ensure
_dir
_exists
(encodeFilename(temp_filename
)):
3016 if self
._write
_description
('video', info_dict
,
3017 self
.prepare_filename(info_dict
, 'description')) is None:
3020 sub_files
= self
._write
_subtitles
(info_dict
, temp_filename
)
3021 if sub_files
is None:
3023 files_to_move
.update(dict(sub_files
))
3025 thumb_files
= self
._write
_thumbnails
(
3026 'video', info_dict
, temp_filename
, self
.prepare_filename(info_dict
, 'thumbnail'))
3027 if thumb_files
is None:
3029 files_to_move
.update(dict(thumb_files
))
3031 infofn
= self
.prepare_filename(info_dict
, 'infojson')
3032 _infojson_written
= self
._write
_info
_json
('video', info_dict
, infofn
)
3033 if _infojson_written
:
3034 info_dict
['infojson_filename'] = infofn
3035 # For backward compatibility, even though it was a private field
3036 info_dict
['__infojson_filename'] = infofn
3037 elif _infojson_written
is None:
3040 # Note: Annotations are deprecated
3042 if self
.params
.get('writeannotations', False):
3043 annofn
= self
.prepare_filename(info_dict
, 'annotation')
3045 if not self
._ensure
_dir
_exists
(encodeFilename(annofn
)):
3047 if not self
.params
.get('overwrites', True) and os
.path
.exists(encodeFilename(annofn
)):
3048 self
.to_screen('[info] Video annotations are already present')
3049 elif not info_dict
.get('annotations'):
3050 self
.report_warning('There are no annotations to write.')
3053 self
.to_screen('[info] Writing video annotations to: ' + annofn
)
3054 with open(encodeFilename(annofn
), 'w', encoding
='utf-8') as annofile
:
3055 annofile
.write(info_dict
['annotations'])
3056 except (KeyError, TypeError):
3057 self
.report_warning('There are no annotations to write.')
3059 self
.report_error('Cannot write annotations file: ' + annofn
)
3062 # Write internet shortcut files
3063 def _write_link_file(link_type
):
3064 url
= try_get(info_dict
['webpage_url'], iri_to_uri
)
3066 self
.report_warning(
3067 f
'Cannot write internet shortcut file because the actual URL of "{info_dict["webpage_url"]}" is unknown')
3069 linkfn
= replace_extension(self
.prepare_filename(info_dict
, 'link'), link_type
, info_dict
.get('ext'))
3070 if not self
._ensure
_dir
_exists
(encodeFilename(linkfn
)):
3072 if self
.params
.get('overwrites', True) and os
.path
.exists(encodeFilename(linkfn
)):
3073 self
.to_screen(f
'[info] Internet shortcut (.{link_type}) is already present')
3076 self
.to_screen(f
'[info] Writing internet shortcut (.{link_type}) to: {linkfn}')
3077 with open(encodeFilename(to_high_limit_path(linkfn
)), 'w', encoding
='utf-8',
3078 newline
='\r\n' if link_type
== 'url' else '\n') as linkfile
:
3079 template_vars
= {'url': url}
3080 if link_type
== 'desktop':
3081 template_vars
['filename'] = linkfn
[:-(len(link_type
) + 1)]
3082 linkfile
.write(LINK_TEMPLATES
[link_type
] % template_vars
)
3084 self
.report_error(f
'Cannot write internet shortcut {linkfn}')
3089 'url': self
.params
.get('writeurllink'),
3090 'webloc': self
.params
.get('writewebloclink'),
3091 'desktop': self
.params
.get('writedesktoplink'),
3093 if self
.params
.get('writelink'):
3094 link_type
= ('webloc' if sys
.platform
== 'darwin'
3095 else 'desktop' if sys
.platform
.startswith('linux')
3097 write_links
[link_type
] = True
3099 if any(should_write
and not _write_link_file(link_type
)
3100 for link_type
, should_write
in write_links
.items()):
3103 new_info
, files_to_move
= self
.pre_process(info_dict
, 'before_dl', files_to_move
)
3104 replace_info_dict(new_info
)
3106 if self
.params
.get('skip_download'):
3107 info_dict
['filepath'] = temp_filename
3108 info_dict
['__finaldir'] = os
.path
.dirname(os
.path
.abspath(encodeFilename(full_filename
)))
3109 info_dict
['__files_to_move'] = files_to_move
3110 replace_info_dict(self
.run_pp(MoveFilesAfterDownloadPP(self
, False), info_dict
))
3111 info_dict
['__write_download_archive'] = self
.params
.get('force_write_download_archive')
3114 info_dict
.setdefault('__postprocessors', [])
3117 def existing_video_file(*filepaths
):
3118 ext
= info_dict
.get('ext')
3119 converted
= lambda file: replace_extension(file, self
.params
.get('final_ext') or ext
, ext
)
3120 file = self
.existing_file(itertools
.chain(*zip(map(converted
, filepaths
), filepaths
)),
3121 default_overwrite
=False)
3123 info_dict
['ext'] = os
.path
.splitext(file)[1][1:]
3126 fd
, success
= None, True
3127 if info_dict
.get('protocol') or info_dict
.get('url'):
3128 fd
= get_suitable_downloader(info_dict
, self
.params
, to_stdout
=temp_filename
== '-')
3129 if fd
is not FFmpegFD
and 'no-direct-merge' not in self
.params
['compat_opts'] and (
3130 info_dict
.get('section_start') or info_dict
.get('section_end')):
3131 msg
= ('This format cannot be partially downloaded' if FFmpegFD
.available()
3132 else 'You have requested downloading the video partially, but ffmpeg is not installed')
3133 self
.report_error(f
'{msg}. Aborting')
3136 if info_dict
.get('requested_formats') is not None:
3137 requested_formats
= info_dict
['requested_formats']
3138 old_ext
= info_dict
['ext']
3139 if self
.params
.get('merge_output_format') is None:
3140 if (info_dict
['ext'] == 'webm'
3141 and info_dict
.get('thumbnails')
3142 # check with type instead of pp_key, __name__, or isinstance
3143 # since we dont want any custom PPs to trigger this
3144 and any(type(pp
) == EmbedThumbnailPP
for pp
in self
._pps
['post_process'])): # noqa: E721
3145 info_dict
['ext'] = 'mkv'
3146 self
.report_warning(
3147 'webm doesn\'t support embedding a thumbnail, mkv will be used')
3148 new_ext
= info_dict
['ext']
3150 def correct_ext(filename
, ext
=new_ext
):
3153 filename_real_ext
= os
.path
.splitext(filename
)[1][1:]
3155 os
.path
.splitext(filename
)[0]
3156 if filename_real_ext
in (old_ext
, new_ext
)
3158 return f
'{filename_wo_ext}.{ext}'
3160 # Ensure filename always has a correct extension for successful merge
3161 full_filename
= correct_ext(full_filename
)
3162 temp_filename
= correct_ext(temp_filename
)
3163 dl_filename
= existing_video_file(full_filename
, temp_filename
)
3164 info_dict
['__real_download'] = False
3166 merger
= FFmpegMergerPP(self
)
3168 if dl_filename
is not None:
3169 self
.report_file_already_downloaded(dl_filename
)
3171 for f
in requested_formats
if fd
!= FFmpegFD
else []:
3172 f
['filepath'] = fname
= prepend_extension(
3173 correct_ext(temp_filename
, info_dict
['ext']),
3174 'f%s' % f
['format_id'], info_dict
['ext'])
3175 downloaded
.append(fname
)
3176 info_dict
['url'] = '\n'.join(f
['url'] for f
in requested_formats
)
3177 success
, real_download
= self
.dl(temp_filename
, info_dict
)
3178 info_dict
['__real_download'] = real_download
3180 if self
.params
.get('allow_unplayable_formats'):
3181 self
.report_warning(
3182 'You have requested merging of multiple formats '
3183 'while also allowing unplayable formats to be downloaded. '
3184 'The formats won\'t be merged to prevent data corruption.')
3185 elif not merger
.available
:
3186 msg
= 'You have requested merging of multiple formats but ffmpeg is not installed'
3187 if not self
.params
.get('ignoreerrors'):
3188 self
.report_error(f
'{msg}. Aborting due to --abort-on-error')
3190 self
.report_warning(f
'{msg}. The formats won\'t be merged')
3192 if temp_filename
== '-':
3193 reason
= ('using a downloader other than ffmpeg' if FFmpegFD
.can_merge_formats(info_dict
, self
.params
)
3194 else 'but the formats are incompatible for simultaneous download' if merger
.available
3195 else 'but ffmpeg is not installed')
3196 self
.report_warning(
3197 f
'You have requested downloading multiple formats to stdout {reason}. '
3198 'The formats will be streamed one after the other')
3199 fname
= temp_filename
3200 for f
in requested_formats
:
3201 new_info
= dict(info_dict
)
3202 del new_info
['requested_formats']
3204 if temp_filename
!= '-':
3205 fname
= prepend_extension(
3206 correct_ext(temp_filename
, new_info
['ext']),
3207 'f%s' % f
['format_id'], new_info
['ext'])
3208 if not self
._ensure
_dir
_exists
(fname
):
3210 f
['filepath'] = fname
3211 downloaded
.append(fname
)
3212 partial_success
, real_download
= self
.dl(fname
, new_info
)
3213 info_dict
['__real_download'] = info_dict
['__real_download'] or real_download
3214 success
= success
and partial_success
3216 if downloaded
and merger
.available
and not self
.params
.get('allow_unplayable_formats'):
3217 info_dict
['__postprocessors'].append(merger
)
3218 info_dict
['__files_to_merge'] = downloaded
3219 # Even if there were no downloads, it is being merged only now
3220 info_dict
['__real_download'] = True
3222 for file in downloaded
:
3223 files_to_move
[file] = None
3225 # Just a single file
3226 dl_filename
= existing_video_file(full_filename
, temp_filename
)
3227 if dl_filename
is None or dl_filename
== temp_filename
:
3228 # dl_filename == temp_filename could mean that the file was partially downloaded with --no-part.
3229 # So we should try to resume the download
3230 success
, real_download
= self
.dl(temp_filename
, info_dict
)
3231 info_dict
['__real_download'] = real_download
3233 self
.report_file_already_downloaded(dl_filename
)
3235 dl_filename
= dl_filename
or temp_filename
3236 info_dict
['__finaldir'] = os
.path
.dirname(os
.path
.abspath(encodeFilename(full_filename
)))
3238 except network_exceptions
as err
:
3239 self
.report_error('unable to download video data: %s' % error_to_compat_str(err
))
3241 except OSError as err
:
3242 raise UnavailableVideoError(err
)
3243 except (ContentTooShortError
, ) as err
:
3244 self
.report_error(f
'content too short (expected {err.expected} bytes and served {err.downloaded})')
3247 self
._raise
_pending
_errors
(info_dict
)
3248 if success
and full_filename
!= '-':
3252 fixup_policy
= self
.params
.get('fixup')
3253 vid
= info_dict
['id']
3255 if fixup_policy
in ('ignore', 'never'):
3257 elif fixup_policy
== 'warn':
3259 elif fixup_policy
!= 'force':
3260 assert fixup_policy
in ('detect_or_warn', None)
3261 if not info_dict
.get('__real_download'):
3264 def ffmpeg_fixup(cndn
, msg
, cls
):
3265 if not (do_fixup
and cndn
):
3267 elif do_fixup
== 'warn':
3268 self
.report_warning(f
'{vid}: {msg}')
3272 info_dict
['__postprocessors'].append(pp
)
3274 self
.report_warning(f
'{vid}: {msg}. Install ffmpeg to fix this automatically')
3276 stretched_ratio
= info_dict
.get('stretched_ratio')
3277 ffmpeg_fixup(stretched_ratio
not in (1, None),
3278 f
'Non-uniform pixel ratio {stretched_ratio}',
3279 FFmpegFixupStretchedPP
)
3281 downloader
= get_suitable_downloader(info_dict
, self
.params
) if 'protocol' in info_dict
else None
3282 downloader
= downloader
.FD_NAME
if downloader
else None
3284 ext
= info_dict
.get('ext')
3285 postprocessed_by_ffmpeg
= info_dict
.get('requested_formats') or any((
3286 isinstance(pp
, FFmpegVideoConvertorPP
)
3287 and resolve_recode_mapping(ext
, pp
.mapping
)[0] not in (ext
, None)
3288 ) for pp
in self
._pps
['post_process'])
3290 if not postprocessed_by_ffmpeg
:
3291 ffmpeg_fixup(ext
== 'm4a' and info_dict
.get('container') == 'm4a_dash',
3292 'writing DASH m4a. Only some players support this container',
3294 ffmpeg_fixup(downloader
== 'hlsnative' and not self
.params
.get('hls_use_mpegts')
3295 or info_dict
.get('is_live') and self
.params
.get('hls_use_mpegts') is None,
3296 'Possible MPEG-TS in MP4 container or malformed AAC timestamps',
3298 ffmpeg_fixup(info_dict
.get('is_live') and downloader
== 'DashSegmentsFD',
3299 'Possible duplicate MOOV atoms', FFmpegFixupDuplicateMoovPP
)
3301 ffmpeg_fixup(downloader
== 'web_socket_fragment', 'Malformed timestamps detected', FFmpegFixupTimestampPP
)
3302 ffmpeg_fixup(downloader
== 'web_socket_fragment', 'Malformed duration detected', FFmpegFixupDurationPP
)
3306 replace_info_dict(self
.post_process(dl_filename
, info_dict
, files_to_move
))
3307 except PostProcessingError
as err
:
3308 self
.report_error('Postprocessing: %s' % str(err
))
3311 for ph
in self
._post
_hooks
:
3312 ph(info_dict
['filepath'])
3313 except Exception as err
:
3314 self
.report_error('post hooks: %s' % str(err
))
3316 info_dict
['__write_download_archive'] = True
3318 assert info_dict
is original_infodict
# Make sure the info_dict was modified in-place
3319 if self
.params
.get('force_write_download_archive'):
3320 info_dict
['__write_download_archive'] = True
3321 check_max_downloads()
3323 def __download_wrapper(self
, func
):
3324 @functools.wraps(func
)
3325 def wrapper(*args
, **kwargs
):
3327 res
= func(*args
, **kwargs
)
3328 except UnavailableVideoError
as e
:
3329 self
.report_error(e
)
3330 except DownloadCancelled
as e
:
3331 self
.to_screen(f
'[info] {e}')
3332 if not self
.params
.get('break_per_url'):
3334 self
._num
_downloads
= 0
3336 if self
.params
.get('dump_single_json', False):
3337 self
.post_extract(res
)
3338 self
.to_stdout(json
.dumps(self
.sanitize_info(res
)))
3341 def download(self
, url_list
):
3342 """Download a given list of URLs."""
3343 url_list
= variadic(url_list
) # Passing a single URL is a common mistake
3344 outtmpl
= self
.params
['outtmpl']['default']
3345 if (len(url_list
) > 1
3347 and '%' not in outtmpl
3348 and self
.params
.get('max_downloads') != 1):
3349 raise SameFileError(outtmpl
)
3351 for url
in url_list
:
3352 self
.__download
_wrapper
(self
.extract_info
)(
3353 url
, force_generic_extractor
=self
.params
.get('force_generic_extractor', False))
3355 return self
._download
_retcode
3357 def download_with_info_file(self
, info_filename
):
3358 with contextlib
.closing(fileinput
.FileInput(
3359 [info_filename
], mode
='r',
3360 openhook
=fileinput
.hook_encoded('utf-8'))) as f
:
3361 # FileInput doesn't have a read method, we can't call json.load
3362 info
= self
.sanitize_info(json
.loads('\n'.join(f
)), self
.params
.get('clean_infojson', True))
3364 self
.__download
_wrapper
(self
.process_ie_result
)(info
, download
=True)
3365 except (DownloadError
, EntryNotInPlaylist
, ReExtractInfo
) as e
:
3366 if not isinstance(e
, EntryNotInPlaylist
):
3367 self
.to_stderr('\r')
3368 webpage_url
= info
.get('webpage_url')
3369 if webpage_url
is not None:
3370 self
.report_warning(f
'The info failed to download: {e}; trying with URL {webpage_url}')
3371 return self
.download([webpage_url
])
3374 return self
._download
_retcode
3377 def sanitize_info(info_dict
, remove_private_keys
=False):
3378 ''' Sanitize the infodict for converting to json '''
3379 if info_dict
is None:
3381 info_dict
.setdefault('epoch', int(time
.time()))
3382 info_dict
.setdefault('_type', 'video')
3383 info_dict
.setdefault('_version', {
3384 'version': __version__
,
3385 'current_git_head': current_git_head(),
3386 'release_git_head': RELEASE_GIT_HEAD
,
3387 'repository': REPOSITORY
,
3390 if remove_private_keys
:
3391 reject
= lambda k
, v
: v
is None or k
.startswith('__') or k
in {
3392 'requested_downloads', 'requested_formats', 'requested_subtitles', 'requested_entries',
3393 'entries', 'filepath', '_filename', 'infojson_filename', 'original_url', 'playlist_autonumber',
3396 reject
= lambda k
, v
: False
3399 if isinstance(obj
, dict):
3400 return {k: filter_fn(v) for k, v in obj.items() if not reject(k, v)}
3401 elif isinstance(obj
, (list, tuple, set, LazyList
)):
3402 return list(map(filter_fn
, obj
))
3403 elif obj
is None or isinstance(obj
, (str, int, float, bool)):
3408 return filter_fn(info_dict
)
3411 def filter_requested_info(info_dict
, actually_filter
=True):
3412 ''' Alias of sanitize_info for backward compatibility '''
3413 return YoutubeDL
.sanitize_info(info_dict
, actually_filter
)
3415 def _delete_downloaded_files(self
, *files_to_delete
, info
={}, msg
=None):
3416 for filename
in set(filter(None, files_to_delete
)):
3418 self
.to_screen(msg
% filename
)
3422 self
.report_warning(f
'Unable to delete file {filename}')
3423 if filename
in info
.get('__files_to_move', []): # NB: Delete even if None
3424 del info
['__files_to_move'][filename
]
3427 def post_extract(info_dict
):
3428 def actual_post_extract(info_dict
):
3429 if info_dict
.get('_type') in ('playlist', 'multi_video'):
3430 for video_dict
in info_dict
.get('entries', {}):
3431 actual_post_extract(video_dict
or {})
3434 post_extractor
= info_dict
.pop('__post_extractor', None) or (lambda: {})
3435 info_dict
.update(post_extractor())
3437 actual_post_extract(info_dict
or {})
3439 def run_pp(self
, pp
, infodict
):
3440 files_to_delete
= []
3441 if '__files_to_move' not in infodict
:
3442 infodict
['__files_to_move'] = {}
3444 files_to_delete
, infodict
= pp
.run(infodict
)
3445 except PostProcessingError
as e
:
3446 # Must be True and not 'only_download'
3447 if self
.params
.get('ignoreerrors') is True:
3448 self
.report_error(e
)
3452 if not files_to_delete
:
3454 if self
.params
.get('keepvideo', False):
3455 for f
in files_to_delete
:
3456 infodict
['__files_to_move'].setdefault(f
, '')
3458 self
._delete
_downloaded
_files
(
3459 *files_to_delete
, info
=infodict
, msg
='Deleting original file %s (pass -k to keep)')
3462 def run_all_pps(self
, key
, info
, *, additional_pps
=None):
3463 self
._forceprint
(key
, info
)
3464 for pp
in (additional_pps
or []) + self
._pps
[key
]:
3465 info
= self
.run_pp(pp
, info
)
3468 def pre_process(self
, ie_info
, key
='pre_process', files_to_move
=None):
3469 info
= dict(ie_info
)
3470 info
['__files_to_move'] = files_to_move
or {}
3472 info
= self
.run_all_pps(key
, info
)
3473 except PostProcessingError
as err
:
3474 msg
= f
'Preprocessing: {err}'
3475 info
.setdefault('__pending_error', msg
)
3476 self
.report_error(msg
, is_error
=False)
3477 return info
, info
.pop('__files_to_move', None)
3479 def post_process(self
, filename
, info
, files_to_move
=None):
3480 """Run all the postprocessors on the given file."""
3481 info
['filepath'] = filename
3482 info
['__files_to_move'] = files_to_move
or {}
3483 info
= self
.run_all_pps('post_process', info
, additional_pps
=info
.get('__postprocessors'))
3484 info
= self
.run_pp(MoveFilesAfterDownloadPP(self
), info
)
3485 del info
['__files_to_move']
3486 return self
.run_all_pps('after_move', info
)
3488 def _make_archive_id(self
, info_dict
):
3489 video_id
= info_dict
.get('id')
3492 # Future-proof against any change in case
3493 # and backwards compatibility with prior versions
3494 extractor
= info_dict
.get('extractor_key') or info_dict
.get('ie_key') # key in a playlist
3495 if extractor
is None:
3496 url
= str_or_none(info_dict
.get('url'))
3499 # Try to find matching extractor for the URL and take its ie_key
3500 for ie_key
, ie
in self
._ies
.items():
3501 if ie
.suitable(url
):
3506 return make_archive_id(extractor
, video_id
)
3508 def in_download_archive(self
, info_dict
):
3509 if not self
.archive
:
3512 vid_ids
= [self
._make
_archive
_id
(info_dict
)]
3513 vid_ids
.extend(info_dict
.get('_old_archive_ids') or [])
3514 return any(id_
in self
.archive
for id_
in vid_ids
)
3516 def record_download_archive(self
, info_dict
):
3517 fn
= self
.params
.get('download_archive')
3520 vid_id
= self
._make
_archive
_id
(info_dict
)
3523 self
.write_debug(f
'Adding to archive: {vid_id}')
3524 if is_path_like(fn
):
3525 with locked_file(fn
, 'a', encoding
='utf-8') as archive_file
:
3526 archive_file
.write(vid_id
+ '\n')
3527 self
.archive
.add(vid_id
)
3530 def format_resolution(format
, default
='unknown'):
3531 if format
.get('vcodec') == 'none' and format
.get('acodec') != 'none':
3533 if format
.get('resolution') is not None:
3534 return format
['resolution']
3535 if format
.get('width') and format
.get('height'):
3536 return '%dx%d' % (format
['width'], format
['height'])
3537 elif format
.get('height'):
3538 return '%sp' % format
['height']
3539 elif format
.get('width'):
3540 return '%dx?' % format
['width']
3543 def _list_format_headers(self
, *headers
):
3544 if self
.params
.get('listformats_table', True) is not False:
3545 return [self
._format
_out
(header
, self
.Styles
.HEADERS
) for header
in headers
]
3548 def _format_note(self
, fdict
):
3550 if fdict
.get('ext') in ['f4f', 'f4m']:
3551 res
+= '(unsupported)'
3552 if fdict
.get('language'):
3555 res
+= '[%s]' % fdict
['language']
3556 if fdict
.get('format_note') is not None:
3559 res
+= fdict
['format_note']
3560 if fdict
.get('tbr') is not None:
3563 res
+= '%4dk' % fdict
['tbr']
3564 if fdict
.get('container') is not None:
3567 res
+= '%s container' % fdict
['container']
3568 if (fdict
.get('vcodec') is not None
3569 and fdict
.get('vcodec') != 'none'):
3572 res
+= fdict
['vcodec']
3573 if fdict
.get('vbr') is not None:
3575 elif fdict
.get('vbr') is not None and fdict
.get('abr') is not None:
3577 if fdict
.get('vbr') is not None:
3578 res
+= '%4dk' % fdict
['vbr']
3579 if fdict
.get('fps') is not None:
3582 res
+= '%sfps' % fdict
['fps']
3583 if fdict
.get('acodec') is not None:
3586 if fdict
['acodec'] == 'none':
3589 res
+= '%-5s' % fdict
['acodec']
3590 elif fdict
.get('abr') is not None:
3594 if fdict
.get('abr') is not None:
3595 res
+= '@%3dk' % fdict
['abr']
3596 if fdict
.get('asr') is not None:
3597 res
+= ' (%5dHz)' % fdict
['asr']
3598 if fdict
.get('filesize') is not None:
3601 res
+= format_bytes(fdict
['filesize'])
3602 elif fdict
.get('filesize_approx') is not None:
3605 res
+= '~' + format_bytes(fdict
['filesize_approx'])
3608 def _get_formats(self
, info_dict
):
3609 if info_dict
.get('formats') is None:
3610 if info_dict
.get('url') and info_dict
.get('_type', 'video') == 'video':
3613 return info_dict
['formats']
3615 def render_formats_table(self
, info_dict
):
3616 formats
= self
._get
_formats
(info_dict
)
3619 if not self
.params
.get('listformats_table', True) is not False:
3622 format_field(f
, 'format_id'),
3623 format_field(f
, 'ext'),
3624 self
.format_resolution(f
),
3625 self
._format
_note
(f
)
3626 ] for f
in formats
if (f
.get('preference') or 0) >= -1000]
3627 return render_table(['format code', 'extension', 'resolution', 'note'], table
, extra_gap
=1)
3629 def simplified_codec(f
, field
):
3630 assert field
in ('acodec', 'vcodec')
3631 codec
= f
.get(field
, 'unknown')
3634 elif codec
!= 'none':
3635 return '.'.join(codec
.split('.')[:4])
3637 if field
== 'vcodec' and f
.get('acodec') == 'none':
3639 elif field
== 'acodec' and f
.get('vcodec') == 'none':
3641 return self
._format
_out
('audio only' if field
== 'vcodec' else 'video only',
3642 self
.Styles
.SUPPRESS
)
3644 delim
= self
._format
_out
('\u2502', self
.Styles
.DELIM
, '|', test_encoding
=True)
3647 self
._format
_out
(format_field(f
, 'format_id'), self
.Styles
.ID
),
3648 format_field(f
, 'ext'),
3649 format_field(f
, func
=self
.format_resolution
, ignore
=('audio only', 'images')),
3650 format_field(f
, 'fps', '\t%d', func
=round),
3651 format_field(f
, 'dynamic_range', '%s', ignore
=(None, 'SDR')).replace('HDR', ''),
3652 format_field(f
, 'audio_channels', '\t%s'),
3654 format_field(f
, 'filesize', ' \t%s', func
=format_bytes
) + format_field(f
, 'filesize_approx', '~\t%s', func
=format_bytes
),
3655 format_field(f
, 'tbr', '\t%dk', func
=round),
3656 shorten_protocol_name(f
.get('protocol', '')),
3658 simplified_codec(f
, 'vcodec'),
3659 format_field(f
, 'vbr', '\t%dk', func
=round),
3660 simplified_codec(f
, 'acodec'),
3661 format_field(f
, 'abr', '\t%dk', func
=round),
3662 format_field(f
, 'asr', '\t%s', func
=format_decimal_suffix
),
3664 self
._format
_out
('UNSUPPORTED', 'light red') if f
.get('ext') in ('f4f', 'f4m') else None,
3665 format_field(f
, 'language', '[%s]'),
3666 join_nonempty(format_field(f
, 'format_note'),
3667 format_field(f
, 'container', ignore
=(None, f
.get('ext'))),
3670 ] for f
in formats
if f
.get('preference') is None or f
['preference'] >= -1000]
3671 header_line
= self
._list
_format
_headers
(
3672 'ID', 'EXT', 'RESOLUTION', '\tFPS', 'HDR', 'CH', delim
, '\tFILESIZE', '\tTBR', 'PROTO',
3673 delim
, 'VCODEC', '\tVBR', 'ACODEC', '\tABR', '\tASR', 'MORE INFO')
3675 return render_table(
3676 header_line
, table
, hide_empty
=True,
3677 delim
=self
._format
_out
('\u2500', self
.Styles
.DELIM
, '-', test_encoding
=True))
3679 def render_thumbnails_table(self
, info_dict
):
3680 thumbnails
= list(info_dict
.get('thumbnails') or [])
3683 return render_table(
3684 self
._list
_format
_headers
('ID', 'Width', 'Height', 'URL'),
3685 [[t
.get('id'), t
.get('width') or 'unknown', t
.get('height') or 'unknown', t
['url']] for t
in thumbnails
])
3687 def render_subtitles_table(self
, video_id
, subtitles
):
3688 def _row(lang
, formats
):
3689 exts
, names
= zip(*((f
['ext'], f
.get('name') or 'unknown') for f
in reversed(formats
)))
3690 if len(set(names
)) == 1:
3691 names
= [] if names
[0] == 'unknown' else names
[:1]
3692 return [lang
, ', '.join(names
), ', '.join(exts
)]
3696 return render_table(
3697 self
._list
_format
_headers
('Language', 'Name', 'Formats'),
3698 [_row(lang
, formats
) for lang
, formats
in subtitles
.items()],
3701 def __list_table(self
, video_id
, name
, func
, *args
):
3704 self
.to_screen(f
'{video_id} has no {name}')
3706 self
.to_screen(f
'[info] Available {name} for {video_id}:')
3707 self
.to_stdout(table
)
3709 def list_formats(self
, info_dict
):
3710 self
.__list
_table
(info_dict
['id'], 'formats', self
.render_formats_table
, info_dict
)
3712 def list_thumbnails(self
, info_dict
):
3713 self
.__list
_table
(info_dict
['id'], 'thumbnails', self
.render_thumbnails_table
, info_dict
)
3715 def list_subtitles(self
, video_id
, subtitles
, name
='subtitles'):
3716 self
.__list
_table
(video_id
, name
, self
.render_subtitles_table
, video_id
, subtitles
)
3718 def urlopen(self
, req
):
3719 """ Start an HTTP download """
3720 if isinstance(req
, str):
3721 req
= sanitized_Request(req
)
3722 return self
._opener
.open(req
, timeout
=self
._socket
_timeout
)
3724 def print_debug_header(self
):
3725 if not self
.params
.get('verbose'):
3728 from . import _IN_CLI
# Must be delayed import
3730 # These imports can be slow. So import them only as needed
3731 from .extractor
.extractors
import _LAZY_LOADER
3732 from .extractor
.extractors
import _PLUGIN_CLASSES
as plugin_extractors
3734 def get_encoding(stream
):
3735 ret
= str(getattr(stream
, 'encoding', 'missing (%s)' % type(stream
).__name
__))
3736 if not supports_terminal_sequences(stream
):
3737 from .utils
import WINDOWS_VT_MODE
# Must be imported locally
3738 ret
+= ' (No VT)' if WINDOWS_VT_MODE
is False else ' (No ANSI)'
3741 encoding_str
= 'Encodings: locale %s, fs %s, pref %s, %s' % (
3742 locale
.getpreferredencoding(),
3743 sys
.getfilesystemencoding(),
3744 self
.get_encoding(),
3746 f
'{key} {get_encoding(stream)}' for key
, stream
in self
._out
_files
.items_
3747 if stream
is not None and key
!= 'console')
3750 logger
= self
.params
.get('logger')
3752 write_debug
= lambda msg
: logger
.debug(f
'[debug] {msg}')
3753 write_debug(encoding_str
)
3755 write_string(f
'[debug] {encoding_str}\n', encoding
=None)
3756 write_debug
= lambda msg
: self
._write
_string
(f
'[debug] {msg}\n')
3758 source
= detect_variant()
3759 if VARIANT
not in (None, 'pip'):
3761 write_debug(join_nonempty(
3762 f
'{"yt-dlp" if REPOSITORY == "yt-dlp/yt-dlp" else REPOSITORY} version',
3764 f
'[{RELEASE_GIT_HEAD}]' if RELEASE_GIT_HEAD
else '',
3765 '' if source
== 'unknown' else f
'({source})',
3766 '' if _IN_CLI
else 'API',
3770 write_debug(f
'params: {self.params}')
3772 if not _LAZY_LOADER
:
3773 if os
.environ
.get('YTDLP_NO_LAZY_EXTRACTORS'):
3774 write_debug('Lazy loading extractors is forcibly disabled')
3776 write_debug('Lazy loading extractors is disabled')
3777 if plugin_extractors
or plugin_postprocessors
:
3778 write_debug('Plugins: %s' % [
3779 '%s%s' % (klass
.__name
__, '' if klass
.__name
__ == name
else f
' as {name}')
3780 for name
, klass
in itertools
.chain(plugin_extractors
.items(), plugin_postprocessors
.items())])
3781 if self
.params
['compat_opts']:
3782 write_debug('Compatibility options: %s' % ', '.join(self
.params
['compat_opts']))
3784 if current_git_head():
3785 write_debug(f
'Git HEAD: {current_git_head()}')
3786 write_debug(system_identifier())
3788 exe_versions
, ffmpeg_features
= FFmpegPostProcessor
.get_versions_and_features(self
)
3789 ffmpeg_features
= {key for key, val in ffmpeg_features.items() if val}
3791 exe_versions
['ffmpeg'] += ' (%s)' % ','.join(sorted(ffmpeg_features
))
3793 exe_versions
['rtmpdump'] = rtmpdump_version()
3794 exe_versions
['phantomjs'] = PhantomJSwrapper
._version
()
3795 exe_str
= ', '.join(
3796 f
'{exe} {v}' for exe
, v
in sorted(exe_versions
.items()) if v
3798 write_debug('exe versions: %s' % exe_str
)
3800 from .compat
.compat_utils
import get_package_info
3801 from .dependencies
import available_dependencies
3803 write_debug('Optional libraries: %s' % (', '.join(sorted({
3804 join_nonempty(*get_package_info(m
)) for m
in available_dependencies
.values()
3807 self
._setup
_opener
()
3809 for handler
in self
._opener
.handlers
:
3810 if hasattr(handler
, 'proxies'):
3811 proxy_map
.update(handler
.proxies
)
3812 write_debug(f
'Proxy map: {proxy_map}')
3815 if False and self
.params
.get('call_home'):
3816 ipaddr
= self
.urlopen('https://yt-dl.org/ip').read().decode()
3817 write_debug('Public IP address: %s' % ipaddr
)
3818 latest_version
= self
.urlopen(
3819 'https://yt-dl.org/latest/version').read().decode()
3820 if version_tuple(latest_version
) > version_tuple(__version__
):
3821 self
.report_warning(
3822 'You are using an outdated version (newest version: %s)! '
3823 'See https://yt-dl.org/update if you need help updating.' %
3826 def _setup_opener(self
):
3827 if hasattr(self
, '_opener'):
3829 timeout_val
= self
.params
.get('socket_timeout')
3830 self
._socket
_timeout
= 20 if timeout_val
is None else float(timeout_val
)
3832 opts_cookiesfrombrowser
= self
.params
.get('cookiesfrombrowser')
3833 opts_cookiefile
= self
.params
.get('cookiefile')
3834 opts_proxy
= self
.params
.get('proxy')
3836 self
.cookiejar
= load_cookies(opts_cookiefile
, opts_cookiesfrombrowser
, self
)
3838 cookie_processor
= YoutubeDLCookieProcessor(self
.cookiejar
)
3839 if opts_proxy
is not None:
3840 if opts_proxy
== '':
3843 proxies
= {'http': opts_proxy, 'https': opts_proxy}
3845 proxies
= urllib
.request
.getproxies()
3846 # Set HTTPS proxy to HTTP one if given (https://github.com/ytdl-org/youtube-dl/issues/805)
3847 if 'http' in proxies
and 'https' not in proxies
:
3848 proxies
['https'] = proxies
['http']
3849 proxy_handler
= PerRequestProxyHandler(proxies
)
3851 debuglevel
= 1 if self
.params
.get('debug_printtraffic') else 0
3852 https_handler
= make_HTTPS_handler(self
.params
, debuglevel
=debuglevel
)
3853 ydlh
= YoutubeDLHandler(self
.params
, debuglevel
=debuglevel
)
3854 redirect_handler
= YoutubeDLRedirectHandler()
3855 data_handler
= urllib
.request
.DataHandler()
3857 # When passing our own FileHandler instance, build_opener won't add the
3858 # default FileHandler and allows us to disable the file protocol, which
3859 # can be used for malicious purposes (see
3860 # https://github.com/ytdl-org/youtube-dl/issues/8227)
3861 file_handler
= urllib
.request
.FileHandler()
3863 def file_open(*args
, **kwargs
):
3864 raise urllib
.error
.URLError('file:// scheme is explicitly disabled in yt-dlp for security reasons')
3865 file_handler
.file_open
= file_open
3867 opener
= urllib
.request
.build_opener(
3868 proxy_handler
, https_handler
, cookie_processor
, ydlh
, redirect_handler
, data_handler
, file_handler
)
3870 # Delete the default user-agent header, which would otherwise apply in
3871 # cases where our custom HTTP handler doesn't come into play
3872 # (See https://github.com/ytdl-org/youtube-dl/issues/1309 for details)
3873 opener
.addheaders
= []
3874 self
._opener
= opener
3876 def encode(self
, s
):
3877 if isinstance(s
, bytes):
3878 return s
# Already encoded
3881 return s
.encode(self
.get_encoding())
3882 except UnicodeEncodeError as err
:
3883 err
.reason
= err
.reason
+ '. Check your system encoding configuration or use the --encoding option.'
3886 def get_encoding(self
):
3887 encoding
= self
.params
.get('encoding')
3888 if encoding
is None:
3889 encoding
= preferredencoding()
3892 def _write_info_json(self
, label
, ie_result
, infofn
, overwrite
=None):
3893 ''' Write infojson and returns True = written, 'exists' = Already exists, False = skip, None = error '''
3894 if overwrite
is None:
3895 overwrite
= self
.params
.get('overwrites', True)
3896 if not self
.params
.get('writeinfojson'):
3899 self
.write_debug(f
'Skipping writing {label} infojson')
3901 elif not self
._ensure
_dir
_exists
(infofn
):
3903 elif not overwrite
and os
.path
.exists(infofn
):
3904 self
.to_screen(f
'[info] {label.title()} metadata is already present')
3907 self
.to_screen(f
'[info] Writing {label} metadata as JSON to: {infofn}')
3909 write_json_file(self
.sanitize_info(ie_result
, self
.params
.get('clean_infojson', True)), infofn
)
3912 self
.report_error(f
'Cannot write {label} metadata to JSON file {infofn}')
3915 def _write_description(self
, label
, ie_result
, descfn
):
3916 ''' Write description and returns True = written, False = skip, None = error '''
3917 if not self
.params
.get('writedescription'):
3920 self
.write_debug(f
'Skipping writing {label} description')
3922 elif not self
._ensure
_dir
_exists
(descfn
):
3924 elif not self
.params
.get('overwrites', True) and os
.path
.exists(descfn
):
3925 self
.to_screen(f
'[info] {label.title()} description is already present')
3926 elif ie_result
.get('description') is None:
3927 self
.report_warning(f
'There\'s no {label} description to write')
3931 self
.to_screen(f
'[info] Writing {label} description to: {descfn}')
3932 with open(encodeFilename(descfn
), 'w', encoding
='utf-8') as descfile
:
3933 descfile
.write(ie_result
['description'])
3935 self
.report_error(f
'Cannot write {label} description file {descfn}')
3939 def _write_subtitles(self
, info_dict
, filename
):
3940 ''' Write subtitles to file and return list of (sub_filename, final_sub_filename); or None if error'''
3942 subtitles
= info_dict
.get('requested_subtitles')
3943 if not subtitles
or not (self
.params
.get('writesubtitles') or self
.params
.get('writeautomaticsub')):
3944 # subtitles download errors are already managed as troubles in relevant IE
3945 # that way it will silently go on when used with unsupporting IE
3948 sub_filename_base
= self
.prepare_filename(info_dict
, 'subtitle')
3949 if not sub_filename_base
:
3950 self
.to_screen('[info] Skipping writing video subtitles')
3952 for sub_lang
, sub_info
in subtitles
.items():
3953 sub_format
= sub_info
['ext']
3954 sub_filename
= subtitles_filename(filename
, sub_lang
, sub_format
, info_dict
.get('ext'))
3955 sub_filename_final
= subtitles_filename(sub_filename_base
, sub_lang
, sub_format
, info_dict
.get('ext'))
3956 existing_sub
= self
.existing_file((sub_filename_final
, sub_filename
))
3958 self
.to_screen(f
'[info] Video subtitle {sub_lang}.{sub_format} is already present')
3959 sub_info
['filepath'] = existing_sub
3960 ret
.append((existing_sub
, sub_filename_final
))
3963 self
.to_screen(f
'[info] Writing video subtitles to: {sub_filename}')
3964 if sub_info
.get('data') is not None:
3966 # Use newline='' to prevent conversion of newline characters
3967 # See https://github.com/ytdl-org/youtube-dl/issues/10268
3968 with open(sub_filename
, 'w', encoding
='utf-8', newline
='') as subfile
:
3969 subfile
.write(sub_info
['data'])
3970 sub_info
['filepath'] = sub_filename
3971 ret
.append((sub_filename
, sub_filename_final
))
3974 self
.report_error(f
'Cannot write video subtitles file {sub_filename}')
3978 sub_copy
= sub_info
.copy()
3979 sub_copy
.setdefault('http_headers', info_dict
.get('http_headers'))
3980 self
.dl(sub_filename
, sub_copy
, subtitle
=True)
3981 sub_info
['filepath'] = sub_filename
3982 ret
.append((sub_filename
, sub_filename_final
))
3983 except (DownloadError
, ExtractorError
, IOError, OSError, ValueError) + network_exceptions
as err
:
3984 msg
= f
'Unable to download video subtitles for {sub_lang!r}: {err}'
3985 if self
.params
.get('ignoreerrors') is not True: # False or 'only_download'
3986 if not self
.params
.get('ignoreerrors'):
3987 self
.report_error(msg
)
3988 raise DownloadError(msg
)
3989 self
.report_warning(msg
)
3992 def _write_thumbnails(self
, label
, info_dict
, filename
, thumb_filename_base
=None):
3993 ''' Write thumbnails to file and return list of (thumb_filename, final_thumb_filename) '''
3994 write_all
= self
.params
.get('write_all_thumbnails', False)
3995 thumbnails
, ret
= [], []
3996 if write_all
or self
.params
.get('writethumbnail', False):
3997 thumbnails
= info_dict
.get('thumbnails') or []
3998 multiple
= write_all
and len(thumbnails
) > 1
4000 if thumb_filename_base
is None:
4001 thumb_filename_base
= filename
4002 if thumbnails
and not thumb_filename_base
:
4003 self
.write_debug(f
'Skipping writing {label} thumbnail')
4006 for idx
, t
in list(enumerate(thumbnails
))[::-1]:
4007 thumb_ext
= (f
'{t["id"]}.' if multiple
else '') + determine_ext(t
['url'], 'jpg')
4008 thumb_display_id
= f
'{label} thumbnail {t["id"]}'
4009 thumb_filename
= replace_extension(filename
, thumb_ext
, info_dict
.get('ext'))
4010 thumb_filename_final
= replace_extension(thumb_filename_base
, thumb_ext
, info_dict
.get('ext'))
4012 existing_thumb
= self
.existing_file((thumb_filename_final
, thumb_filename
))
4014 self
.to_screen('[info] %s is already present' % (
4015 thumb_display_id
if multiple
else f
'{label} thumbnail').capitalize())
4016 t
['filepath'] = existing_thumb
4017 ret
.append((existing_thumb
, thumb_filename_final
))
4019 self
.to_screen(f
'[info] Downloading {thumb_display_id} ...')
4021 uf
= self
.urlopen(sanitized_Request(t
['url'], headers
=t
.get('http_headers', {})))
4022 self
.to_screen(f
'[info] Writing {thumb_display_id} to: {thumb_filename}')
4023 with open(encodeFilename(thumb_filename
), 'wb') as thumbf
:
4024 shutil
.copyfileobj(uf
, thumbf
)
4025 ret
.append((thumb_filename
, thumb_filename_final
))
4026 t
['filepath'] = thumb_filename
4027 except network_exceptions
as err
:
4029 self
.report_warning(f
'Unable to download {thumb_display_id}: {err}')
4030 if ret
and not write_all
: