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
77 PerRequestProxyHandler
,
84 UnavailableVideoError
,
86 YoutubeDLCookieProcessor
,
88 YoutubeDLRedirectHandler
,
104 format_decimal_suffix
,
120 orderedSet_from_options
,
124 register_socks_protocols
,
125 remove_terminal_sequences
,
136 supports_terminal_sequences
,
146 windows_enable_vt_mode
,
150 from .version
import RELEASE_GIT_HEAD
, VARIANT
, __version__
152 if compat_os_name
== 'nt':
159 YoutubeDL objects are the ones responsible of downloading the
160 actual video file and writing it to disk if the user has requested
161 it, among some other tasks. In most cases there should be one per
162 program. As, given a video URL, the downloader doesn't know how to
163 extract all the needed information, task that InfoExtractors do, it
164 has to pass the URL to one of them.
166 For this, YoutubeDL objects have a method that allows
167 InfoExtractors to be registered in a given order. When it is passed
168 a URL, the YoutubeDL object handles it to the first InfoExtractor it
169 finds that reports being able to handle it. The InfoExtractor extracts
170 all the information about the video or videos the URL refers to, and
171 YoutubeDL process the extracted information, possibly using a File
172 Downloader to download the video.
174 YoutubeDL objects accept a lot of parameters. In order not to saturate
175 the object constructor with arguments, it receives a dictionary of
176 options instead. These options are available through the params
177 attribute for the InfoExtractors to use. The YoutubeDL also
178 registers itself as the downloader in charge for the InfoExtractors
179 that are added to it, so this is a "mutual registration".
183 username: Username for authentication purposes.
184 password: Password for authentication purposes.
185 videopassword: Password for accessing a video.
186 ap_mso: Adobe Pass multiple-system operator identifier.
187 ap_username: Multiple-system operator account username.
188 ap_password: Multiple-system operator account password.
189 usenetrc: Use netrc for authentication instead.
190 verbose: Print additional info to stdout.
191 quiet: Do not print messages to stdout.
192 no_warnings: Do not print out anything for warnings.
193 forceprint: A dict with keys WHEN mapped to a list of templates to
194 print to stdout. The allowed keys are video or any of the
195 items in utils.POSTPROCESS_WHEN.
196 For compatibility, a single list is also accepted
197 print_to_file: A dict with keys WHEN (same as forceprint) mapped to
198 a list of tuples with (template, filename)
199 forcejson: Force printing info_dict as JSON.
200 dump_single_json: Force printing the info_dict of the whole playlist
201 (or video) as a single JSON line.
202 force_write_download_archive: Force writing download archive regardless
203 of 'skip_download' or 'simulate'.
204 simulate: Do not download the video files. If unset (or None),
205 simulate only if listsubtitles, listformats or list_thumbnails is used
206 format: Video format code. see "FORMAT SELECTION" for more details.
207 You can also pass a function. The function takes 'ctx' as
208 argument and returns the formats to download.
209 See "build_format_selector" for an implementation
210 allow_unplayable_formats: Allow unplayable formats to be extracted and downloaded.
211 ignore_no_formats_error: Ignore "No video formats" error. Usefull for
212 extracting metadata even if the video is not actually
213 available for download (experimental)
214 format_sort: A list of fields by which to sort the video formats.
215 See "Sorting Formats" for more details.
216 format_sort_force: Force the given format_sort. see "Sorting Formats"
218 prefer_free_formats: Whether to prefer video formats with free containers
219 over non-free ones of same quality.
220 allow_multiple_video_streams: Allow multiple video streams to be merged
222 allow_multiple_audio_streams: Allow multiple audio streams to be merged
224 check_formats Whether to test if the formats are downloadable.
225 Can be True (check all), False (check none),
226 'selected' (check selected formats),
227 or None (check only if requested by extractor)
228 paths: Dictionary of output paths. The allowed keys are 'home'
229 'temp' and the keys of OUTTMPL_TYPES (in utils.py)
230 outtmpl: Dictionary of templates for output names. Allowed keys
231 are 'default' and the keys of OUTTMPL_TYPES (in utils.py).
232 For compatibility with youtube-dl, a single string can also be used
233 outtmpl_na_placeholder: Placeholder for unavailable meta fields.
234 restrictfilenames: Do not allow "&" and spaces in file names
235 trim_file_name: Limit length of filename (extension excluded)
236 windowsfilenames: Force the filenames to be windows compatible
237 ignoreerrors: Do not stop on download/postprocessing errors.
238 Can be 'only_download' to ignore only download errors.
239 Default is 'only_download' for CLI, but False for API
240 skip_playlist_after_errors: Number of allowed failures until the rest of
241 the playlist is skipped
242 allowed_extractors: List of regexes to match against extractor names that are allowed
243 overwrites: Overwrite all video and metadata files if True,
244 overwrite only non-video files if None
245 and don't overwrite any file if False
246 For compatibility with youtube-dl,
247 "nooverwrites" may also be used instead
248 playlist_items: Specific indices of playlist to download.
249 playlistrandom: Download playlist items in random order.
250 lazy_playlist: Process playlist entries as they are received.
251 matchtitle: Download only matching titles.
252 rejecttitle: Reject downloads for matching titles.
253 logger: Log messages to a logging.Logger instance.
254 logtostderr: Log messages to stderr instead of stdout.
255 consoletitle: Display progress in console window's titlebar.
256 writedescription: Write the video description to a .description file
257 writeinfojson: Write the video description to a .info.json file
258 clean_infojson: Remove private fields from the infojson
259 getcomments: Extract video comments. This will not be written to disk
260 unless writeinfojson is also given
261 writeannotations: Write the video annotations to a .annotations.xml file
262 writethumbnail: Write the thumbnail image to a file
263 allow_playlist_files: Whether to write playlists' description, infojson etc
264 also to disk when using the 'write*' options
265 write_all_thumbnails: Write all thumbnail formats to files
266 writelink: Write an internet shortcut file, depending on the
267 current platform (.url/.webloc/.desktop)
268 writeurllink: Write a Windows internet shortcut file (.url)
269 writewebloclink: Write a macOS internet shortcut file (.webloc)
270 writedesktoplink: Write a Linux internet shortcut file (.desktop)
271 writesubtitles: Write the video subtitles to a file
272 writeautomaticsub: Write the automatically generated subtitles to a file
273 listsubtitles: Lists all available subtitles for the video
274 subtitlesformat: The format code for subtitles
275 subtitleslangs: List of languages of the subtitles to download (can be regex).
276 The list may contain "all" to refer to all the available
277 subtitles. The language can be prefixed with a "-" to
278 exclude it from the requested languages, e.g. ['all', '-live_chat']
279 keepvideo: Keep the video file after post-processing
280 daterange: A DateRange object, download only if the upload_date is in the range.
281 skip_download: Skip the actual download of the video file
282 cachedir: Location of the cache files in the filesystem.
283 False to disable filesystem cache.
284 noplaylist: Download single video instead of a playlist if in doubt.
285 age_limit: An integer representing the user's age in years.
286 Unsuitable videos for the given age are skipped.
287 min_views: An integer representing the minimum view count the video
288 must have in order to not be skipped.
289 Videos without view count information are always
290 downloaded. None for no limit.
291 max_views: An integer representing the maximum view count.
292 Videos that are more popular than that are not
294 Videos without view count information are always
295 downloaded. None for no limit.
296 download_archive: File name of a file where all downloads are recorded.
297 Videos already present in the file are not downloaded
299 break_on_existing: Stop the download process after attempting to download a
300 file that is in the archive.
301 break_on_reject: Stop the download process when encountering a video that
302 has been filtered out.
303 break_per_url: Whether break_on_reject and break_on_existing
304 should act on each input URL as opposed to for the entire queue
305 cookiefile: File name or text stream from where cookies should be read and dumped to
306 cookiesfrombrowser: A tuple containing the name of the browser, the profile
307 name/path from where cookies are loaded, the name of the keyring,
308 and the container name, e.g. ('chrome', ) or
309 ('vivaldi', 'default', 'BASICTEXT') or ('firefox', 'default', None, 'Meta')
310 legacyserverconnect: Explicitly allow HTTPS connection to servers that do not
311 support RFC 5746 secure renegotiation
312 nocheckcertificate: Do not verify SSL certificates
313 client_certificate: Path to client certificate file in PEM format. May include the private key
314 client_certificate_key: Path to private key file for client certificate
315 client_certificate_password: Password for client certificate private key, if encrypted.
316 If not provided and the key is encrypted, yt-dlp will ask interactively
317 prefer_insecure: Use HTTP instead of HTTPS to retrieve information.
318 (Only supported by some extractors)
319 http_headers: A dictionary of custom headers to be used for all requests
320 proxy: URL of the proxy server to use
321 geo_verification_proxy: URL of the proxy to use for IP address verification
322 on geo-restricted sites.
323 socket_timeout: Time to wait for unresponsive hosts, in seconds
324 bidi_workaround: Work around buggy terminals without bidirectional text
325 support, using fridibi
326 debug_printtraffic:Print out sent and received HTTP traffic
327 default_search: Prepend this string if an input url is not valid.
328 'auto' for elaborate guessing
329 encoding: Use this encoding instead of the system-specified.
330 extract_flat: Whether to resolve and process url_results further
331 * False: Always process (default)
332 * True: Never process
333 * 'in_playlist': Do not process inside playlist/multi_video
334 * 'discard': Always process, but don't return the result
335 from inside playlist/multi_video
336 * 'discard_in_playlist': Same as "discard", but only for
337 playlists (not multi_video)
338 wait_for_video: If given, wait for scheduled streams to become available.
339 The value should be a tuple containing the range
340 (min_secs, max_secs) to wait between retries
341 postprocessors: A list of dictionaries, each with an entry
342 * key: The name of the postprocessor. See
343 yt_dlp/postprocessor/__init__.py for a list.
344 * when: When to run the postprocessor. Allowed values are
345 the entries of utils.POSTPROCESS_WHEN
346 Assumed to be 'post_process' if not given
347 progress_hooks: A list of functions that get called on download
348 progress, with a dictionary with the entries
349 * status: One of "downloading", "error", or "finished".
350 Check this first and ignore unknown values.
351 * info_dict: The extracted info_dict
353 If status is one of "downloading", or "finished", the
354 following properties may also be present:
355 * filename: The final filename (always present)
356 * tmpfilename: The filename we're currently writing to
357 * downloaded_bytes: Bytes on disk
358 * total_bytes: Size of the whole file, None if unknown
359 * total_bytes_estimate: Guess of the eventual file size,
361 * elapsed: The number of seconds since download started.
362 * eta: The estimated time in seconds, None if unknown
363 * speed: The download speed in bytes/second, None if
365 * fragment_index: The counter of the currently
366 downloaded video fragment.
367 * fragment_count: The number of fragments (= individual
368 files that will be merged)
370 Progress hooks are guaranteed to be called at least once
371 (with status "finished") if the download is successful.
372 postprocessor_hooks: A list of functions that get called on postprocessing
373 progress, with a dictionary with the entries
374 * status: One of "started", "processing", or "finished".
375 Check this first and ignore unknown values.
376 * postprocessor: Name of the postprocessor
377 * info_dict: The extracted info_dict
379 Progress hooks are guaranteed to be called at least twice
380 (with status "started" and "finished") if the processing is successful.
381 merge_output_format: "/" separated list of extensions to use when merging formats.
382 final_ext: Expected final extension; used to detect when the file was
383 already downloaded and converted
384 fixup: Automatically correct known faults of the file.
386 - "never": do nothing
387 - "warn": only emit a warning
388 - "detect_or_warn": check whether we can do anything
389 about it, warn otherwise (default)
390 source_address: Client-side IP address to bind to.
391 sleep_interval_requests: Number of seconds to sleep between requests
393 sleep_interval: Number of seconds to sleep before each download when
394 used alone or a lower bound of a range for randomized
395 sleep before each download (minimum possible number
396 of seconds to sleep) when used along with
398 max_sleep_interval:Upper bound of a range for randomized sleep before each
399 download (maximum possible number of seconds to sleep).
400 Must only be used along with sleep_interval.
401 Actual sleep time will be a random float from range
402 [sleep_interval; max_sleep_interval].
403 sleep_interval_subtitles: Number of seconds to sleep before each subtitle download
404 listformats: Print an overview of available video formats and exit.
405 list_thumbnails: Print a table of all thumbnails and exit.
406 match_filter: A function that gets called for every video with the signature
407 (info_dict, *, incomplete: bool) -> Optional[str]
408 For backward compatibility with youtube-dl, the signature
409 (info_dict) -> Optional[str] is also allowed.
410 - If it returns a message, the video is ignored.
411 - If it returns None, the video is downloaded.
412 - If it returns utils.NO_DEFAULT, the user is interactively
413 asked whether to download the video.
414 match_filter_func in utils.py is one example for this.
415 no_color: Do not emit color codes in output.
416 geo_bypass: Bypass geographic restriction via faking X-Forwarded-For
419 Two-letter ISO 3166-2 country code that will be used for
420 explicit geographic restriction bypassing via faking
421 X-Forwarded-For HTTP header
423 IP range in CIDR notation that will be used similarly to
425 external_downloader: A dictionary of protocol keys and the executable of the
426 external downloader to use for it. The allowed protocols
427 are default|http|ftp|m3u8|dash|rtsp|rtmp|mms.
428 Set the value to 'native' to use the native downloader
429 compat_opts: Compatibility options. See "Differences in default behavior".
430 The following options do not work when used through the API:
431 filename, abort-on-error, multistreams, no-live-chat, format-sort
432 no-clean-infojson, no-playlist-metafiles, no-keep-subs, no-attach-info-json.
433 Refer __init__.py for their implementation
434 progress_template: Dictionary of templates for progress outputs.
435 Allowed keys are 'download', 'postprocess',
436 'download-title' (console title) and 'postprocess-title'.
437 The template is mapped on a dictionary with keys 'progress' and 'info'
438 retry_sleep_functions: Dictionary of functions that takes the number of attempts
439 as argument and returns the time to sleep in seconds.
440 Allowed keys are 'http', 'fragment', 'file_access'
441 download_ranges: A callback function that gets called for every video with
442 the signature (info_dict, ydl) -> Iterable[Section].
443 Only the returned sections will be downloaded.
444 Each Section is a dict with the following keys:
445 * start_time: Start time of the section in seconds
446 * end_time: End time of the section in seconds
447 * title: Section title (Optional)
448 * index: Section number (Optional)
449 force_keyframes_at_cuts: Re-encode the video when downloading ranges to get precise cuts
450 noprogress: Do not print the progress bar
451 live_from_start: Whether to download livestreams videos from the start
453 The following parameters are not used by YoutubeDL itself, they are used by
454 the downloader (see yt_dlp/downloader/common.py):
455 nopart, updatetime, buffersize, ratelimit, throttledratelimit, min_filesize,
456 max_filesize, test, noresizebuffer, retries, file_access_retries, fragment_retries,
457 continuedl, xattr_set_filesize, hls_use_mpegts, http_chunk_size,
458 external_downloader_args, concurrent_fragment_downloads.
460 The following options are used by the post processors:
461 ffmpeg_location: Location of the ffmpeg/avconv binary; either the path
462 to the binary or its containing directory.
463 postprocessor_args: A dictionary of postprocessor/executable keys (in lower case)
464 and a list of additional command-line arguments for the
465 postprocessor/executable. The dict can also have "PP+EXE" keys
466 which are used when the given exe is used by the given PP.
467 Use 'default' as the name for arguments to passed to all PP
468 For compatibility with youtube-dl, a single list of args
471 The following options are used by the extractors:
472 extractor_retries: Number of times to retry for known errors
473 dynamic_mpd: Whether to process dynamic DASH manifests (default: True)
474 hls_split_discontinuity: Split HLS playlists to different formats at
475 discontinuities such as ad breaks (default: False)
476 extractor_args: A dictionary of arguments to be passed to the extractors.
477 See "EXTRACTOR ARGUMENTS" for details.
478 E.g. {'youtube': {'skip': ['dash', 'hls']}}
479 mark_watched: Mark videos watched (even with --simulate). Only for YouTube
481 The following options are deprecated and may be removed in the future:
483 force_generic_extractor: Force downloader to use the generic extractor
484 - Use allowed_extractors = ['generic', 'default']
485 playliststart: - Use playlist_items
486 Playlist item to start at.
487 playlistend: - Use playlist_items
488 Playlist item to end at.
489 playlistreverse: - Use playlist_items
490 Download playlist items in reverse order.
491 forceurl: - Use forceprint
492 Force printing final URL.
493 forcetitle: - Use forceprint
494 Force printing title.
495 forceid: - Use forceprint
497 forcethumbnail: - Use forceprint
498 Force printing thumbnail URL.
499 forcedescription: - Use forceprint
500 Force printing description.
501 forcefilename: - Use forceprint
502 Force printing final filename.
503 forceduration: - Use forceprint
504 Force printing duration.
505 allsubtitles: - Use subtitleslangs = ['all']
506 Downloads all the subtitles of the video
507 (requires writesubtitles or writeautomaticsub)
508 include_ads: - Doesn't work
510 call_home: - Not implemented
511 Boolean, true iff we are allowed to contact the
512 yt-dlp servers for debugging.
513 post_hooks: - Register a custom postprocessor
514 A list of functions that get called as the final step
515 for each video file, after all postprocessors have been
516 called. The filename will be passed as the only argument.
517 hls_prefer_native: - Use external_downloader = {'m3u8': 'native'} or {'m3u8': 'ffmpeg'}.
518 Use the native HLS downloader instead of ffmpeg/avconv
519 if True, otherwise use ffmpeg/avconv if False, otherwise
520 use downloader suggested by extractor if None.
521 prefer_ffmpeg: - avconv support is deprecated
522 If False, use avconv instead of ffmpeg if both are available,
523 otherwise prefer ffmpeg.
524 youtube_include_dash_manifest: - Use extractor_args
525 If True (default), DASH manifests and related
526 data will be downloaded and processed by extractor.
527 You can reduce network I/O by disabling it if you don't
528 care about DASH. (only for youtube)
529 youtube_include_hls_manifest: - Use extractor_args
530 If True (default), HLS manifests and related
531 data will be downloaded and processed by extractor.
532 You can reduce network I/O by disabling it if you don't
533 care about HLS. (only for youtube)
537 'width', 'height', 'asr', 'audio_channels', 'fps',
538 'tbr', 'abr', 'vbr', 'filesize', 'filesize_approx',
539 'timestamp', 'release_timestamp',
540 'duration', 'view_count', 'like_count', 'dislike_count', 'repost_count',
541 'average_rating', 'comment_count', 'age_limit',
542 'start_time', 'end_time',
543 'chapter_number', 'season_number', 'episode_number',
544 'track_number', 'disc_number', 'release_year',
548 # NB: Keep in sync with the docstring of extractor/common.py
549 'url', 'manifest_url', 'manifest_stream_number', 'ext', 'format', 'format_id', 'format_note',
550 'width', 'height', 'resolution', 'dynamic_range', 'tbr', 'abr', 'acodec', 'asr', 'audio_channels',
551 'vbr', 'fps', 'vcodec', 'container', 'filesize', 'filesize_approx',
552 'player_url', 'protocol', 'fragment_base_url', 'fragments', 'is_from_start',
553 'preference', 'language', 'language_preference', 'quality', 'source_preference',
554 'http_headers', 'stretched_ratio', 'no_resume', 'has_drm', 'downloader_options',
555 'page_url', 'app', 'play_path', 'tc_url', 'flash_version', 'rtmp_live', 'rtmp_conn', 'rtmp_protocol', 'rtmp_real_time'
557 _format_selection_exts
= {
558 'audio': set(MEDIA_EXTENSIONS
.common_audio
),
559 'video': set(MEDIA_EXTENSIONS
.common_video
+ ('3gp', )),
560 'storyboards': set(MEDIA_EXTENSIONS
.storyboards
),
563 def __init__(self
, params
=None, auto_init
=True):
564 """Create a FileDownloader object with the given options.
565 @param auto_init Whether to load the default extractors and print header (if verbose).
566 Set to 'no_verbose_header' to not print the header
572 self
._ies
_instances
= {}
573 self
._pps
= {k: [] for k in POSTPROCESS_WHEN}
574 self
._printed
_messages
= set()
575 self
._first
_webpage
_request
= True
576 self
._post
_hooks
= []
577 self
._progress
_hooks
= []
578 self
._postprocessor
_hooks
= []
579 self
._download
_retcode
= 0
580 self
._num
_downloads
= 0
582 self
._playlist
_level
= 0
583 self
._playlist
_urls
= set()
584 self
.cache
= Cache(self
)
586 windows_enable_vt_mode()
587 stdout
= sys
.stderr
if self
.params
.get('logtostderr') else sys
.stdout
588 self
._out
_files
= Namespace(
591 screen
=sys
.stderr
if self
.params
.get('quiet') else stdout
,
592 console
=None if compat_os_name
== 'nt' else next(
593 filter(supports_terminal_sequences
, (sys
.stderr
, sys
.stdout
)), None)
595 self
._allow
_colors
= Namespace(**{
596 type_
: not self
.params
.get('no_color') and supports_terminal_sequences(stream
)
597 for type_
, stream
in self
._out
_files
.items_
if type_
!= 'console'
600 # The code is left like this to be reused for future deprecations
601 MIN_SUPPORTED
, MIN_RECOMMENDED
= (3, 7), (3, 7)
602 current_version
= sys
.version_info
[:2]
603 if current_version
< MIN_RECOMMENDED
:
604 msg
= ('Support for Python version %d.%d has been deprecated. '
605 'See https://github.com/yt-dlp/yt-dlp/issues/3764 for more details.'
606 '\n You will no longer receive updates on this version')
607 if current_version
< MIN_SUPPORTED
:
608 msg
= 'Python version %d.%d is no longer supported'
609 self
.deprecation_warning(
610 f
'{msg}! Please update to Python %d.%d or above' % (*current_version
, *MIN_RECOMMENDED
))
612 if self
.params
.get('allow_unplayable_formats'):
614 f
'You have asked for {self._format_err("UNPLAYABLE", self.Styles.EMPHASIS)} formats to be listed/downloaded. '
615 'This is a developer option intended for debugging. \n'
616 ' If you experience any issues while using this option, '
617 f
'{self._format_err("DO NOT", self.Styles.ERROR)} open a bug report')
619 def check_deprecated(param
, option
, suggestion
):
620 if self
.params
.get(param
) is not None:
621 self
.report_warning(f
'{option} is deprecated. Use {suggestion} instead')
625 if check_deprecated('cn_verification_proxy', '--cn-verification-proxy', '--geo-verification-proxy'):
626 if self
.params
.get('geo_verification_proxy') is None:
627 self
.params
['geo_verification_proxy'] = self
.params
['cn_verification_proxy']
629 check_deprecated('autonumber', '--auto-number', '-o "%(autonumber)s-%(title)s.%(ext)s"')
630 check_deprecated('usetitle', '--title', '-o "%(title)s-%(id)s.%(ext)s"')
631 check_deprecated('useid', '--id', '-o "%(id)s.%(ext)s"')
633 for msg
in self
.params
.get('_warnings', []):
634 self
.report_warning(msg
)
635 for msg
in self
.params
.get('_deprecation_warnings', []):
636 self
.deprecated_feature(msg
)
638 self
.params
['compat_opts'] = set(self
.params
.get('compat_opts', ()))
639 if 'list-formats' in self
.params
['compat_opts']:
640 self
.params
['listformats_table'] = False
642 if 'overwrites' not in self
.params
and self
.params
.get('nooverwrites') is not None:
643 # nooverwrites was unnecessarily changed to overwrites
644 # in 0c3d0f51778b153f65c21906031c2e091fcfb641
645 # This ensures compatibility with both keys
646 self
.params
['overwrites'] = not self
.params
['nooverwrites']
647 elif self
.params
.get('overwrites') is None:
648 self
.params
.pop('overwrites', None)
650 self
.params
['nooverwrites'] = not self
.params
['overwrites']
652 self
.params
.setdefault('forceprint', {})
653 self
.params
.setdefault('print_to_file', {})
655 # Compatibility with older syntax
656 if not isinstance(params
['forceprint'], dict):
657 self
.params
['forceprint'] = {'video': params['forceprint']}
659 if self
.params
.get('bidi_workaround', False):
662 master
, slave
= pty
.openpty()
663 width
= shutil
.get_terminal_size().columns
664 width_args
= [] if width
is None else ['-w', str(width
)]
665 sp_kwargs
= {'stdin': subprocess.PIPE, 'stdout': slave, 'stderr': self._out_files.error}
667 self
._output
_process
= Popen(['bidiv'] + width_args
, **sp_kwargs
)
669 self
._output
_process
= Popen(['fribidi', '-c', 'UTF-8'] + width_args
, **sp_kwargs
)
670 self
._output
_channel
= os
.fdopen(master
, 'rb')
671 except OSError as ose
:
672 if ose
.errno
== errno
.ENOENT
:
674 'Could not find fribidi executable, ignoring --bidi-workaround. '
675 'Make sure that fribidi is an executable file in one of the directories in your $PATH.')
680 if auto_init
!= 'no_verbose_header':
681 self
.print_debug_header()
682 self
.add_default_info_extractors()
684 if (sys
.platform
!= 'win32'
685 and sys
.getfilesystemencoding() in ['ascii', 'ANSI_X3.4-1968']
686 and not self
.params
.get('restrictfilenames', False)):
687 # Unicode filesystem API will throw errors (#1474, #13027)
689 'Assuming --restrict-filenames since file system encoding '
690 'cannot encode all characters. '
691 'Set the LC_ALL environment variable to fix this.')
692 self
.params
['restrictfilenames'] = True
694 self
._parse
_outtmpl
()
696 # Creating format selector here allows us to catch syntax errors before the extraction
697 self
.format_selector
= (
698 self
.params
.get('format') if self
.params
.get('format') in (None, '-')
699 else self
.params
['format'] if callable(self
.params
['format'])
700 else self
.build_format_selector(self
.params
['format']))
702 # Set http_headers defaults according to std_headers
703 self
.params
['http_headers'] = merge_headers(std_headers
, self
.params
.get('http_headers', {}))
706 'post_hooks': self
.add_post_hook
,
707 'progress_hooks': self
.add_progress_hook
,
708 'postprocessor_hooks': self
.add_postprocessor_hook
,
710 for opt
, fn
in hooks
.items():
711 for ph
in self
.params
.get(opt
, []):
714 for pp_def_raw
in self
.params
.get('postprocessors', []):
715 pp_def
= dict(pp_def_raw
)
716 when
= pp_def
.pop('when', 'post_process')
717 self
.add_post_processor(
718 get_postprocessor(pp_def
.pop('key'))(self
, **pp_def
),
722 register_socks_protocols()
724 def preload_download_archive(fn
):
725 """Preload the archive, if any is specified"""
728 self
.write_debug(f
'Loading archive file {fn!r}')
730 with locked_file(fn
, 'r', encoding
='utf-8') as archive_file
:
731 for line
in archive_file
:
732 self
.archive
.add(line
.strip())
733 except OSError as ioe
:
734 if ioe
.errno
!= errno
.ENOENT
:
740 preload_download_archive(self
.params
.get('download_archive'))
742 def warn_if_short_id(self
, argv
):
743 # short YouTube ID starting with dash?
745 i
for i
, a
in enumerate(argv
)
746 if re
.match(r
'^-[0-9A-Za-z_-]{10}$', a
)]
750 + [a
for i
, a
in enumerate(argv
) if i
not in idxs
]
751 + ['--'] + [argv
[i
] for i
in idxs
]
754 'Long argument string detected. '
755 'Use -- to separate parameters and URLs, like this:\n%s' %
756 args_to_str(correct_argv
))
758 def add_info_extractor(self
, ie
):
759 """Add an InfoExtractor object to the end of the list."""
761 self
._ies
[ie_key
] = ie
762 if not isinstance(ie
, type):
763 self
._ies
_instances
[ie_key
] = ie
764 ie
.set_downloader(self
)
766 def get_info_extractor(self
, ie_key
):
768 Get an instance of an IE with name ie_key, it will try to get one from
769 the _ies list, if there's no instance it will create a new one and add
770 it to the extractor list.
772 ie
= self
._ies
_instances
.get(ie_key
)
774 ie
= get_info_extractor(ie_key
)()
775 self
.add_info_extractor(ie
)
778 def add_default_info_extractors(self
):
780 Add the InfoExtractors returned by gen_extractors to the end of the list
782 all_ies
= {ie.IE_NAME.lower(): ie for ie in gen_extractor_classes()}
783 all_ies
['end'] = UnsupportedURLIE()
785 ie_names
= orderedSet_from_options(
786 self
.params
.get('allowed_extractors', ['default']), {
787 'all': list(all_ies
),
788 'default': [name
for name
, ie
in all_ies
.items() if ie
._ENABLED
],
790 except re
.error
as e
:
791 raise ValueError(f
'Wrong regex for allowed_extractors: {e.pattern}')
792 for name
in ie_names
:
793 self
.add_info_extractor(all_ies
[name
])
794 self
.write_debug(f
'Loaded {len(ie_names)} extractors')
796 def add_post_processor(self
, pp
, when
='post_process'):
797 """Add a PostProcessor object to the end of the chain."""
798 assert when
in POSTPROCESS_WHEN
, f
'Invalid when={when}'
799 self
._pps
[when
].append(pp
)
800 pp
.set_downloader(self
)
802 def add_post_hook(self
, ph
):
803 """Add the post hook"""
804 self
._post
_hooks
.append(ph
)
806 def add_progress_hook(self
, ph
):
807 """Add the download progress hook"""
808 self
._progress
_hooks
.append(ph
)
810 def add_postprocessor_hook(self
, ph
):
811 """Add the postprocessing progress hook"""
812 self
._postprocessor
_hooks
.append(ph
)
813 for pps
in self
._pps
.values():
815 pp
.add_progress_hook(ph
)
817 def _bidi_workaround(self
, message
):
818 if not hasattr(self
, '_output_channel'):
821 assert hasattr(self
, '_output_process')
822 assert isinstance(message
, str)
823 line_count
= message
.count('\n') + 1
824 self
._output
_process
.stdin
.write((message
+ '\n').encode())
825 self
._output
_process
.stdin
.flush()
826 res
= ''.join(self
._output
_channel
.readline().decode()
827 for _
in range(line_count
))
828 return res
[:-len('\n')]
830 def _write_string(self
, message
, out
=None, only_once
=False):
832 if message
in self
._printed
_messages
:
834 self
._printed
_messages
.add(message
)
835 write_string(message
, out
=out
, encoding
=self
.params
.get('encoding'))
837 def to_stdout(self
, message
, skip_eol
=False, quiet
=None):
838 """Print message to stdout"""
839 if quiet
is not None:
840 self
.deprecation_warning('"YoutubeDL.to_stdout" no longer accepts the argument quiet. '
841 'Use "YoutubeDL.to_screen" instead')
842 if skip_eol
is not False:
843 self
.deprecation_warning('"YoutubeDL.to_stdout" no longer accepts the argument skip_eol. '
844 'Use "YoutubeDL.to_screen" instead')
845 self
._write
_string
(f
'{self._bidi_workaround(message)}\n', self
._out
_files
.out
)
847 def to_screen(self
, message
, skip_eol
=False, quiet
=None):
848 """Print message to screen if not in quiet mode"""
849 if self
.params
.get('logger'):
850 self
.params
['logger'].debug(message
)
852 if (self
.params
.get('quiet') if quiet
is None else quiet
) and not self
.params
.get('verbose'):
855 '%s%s' % (self
._bidi
_workaround
(message
), ('' if skip_eol
else '\n')),
856 self
._out
_files
.screen
)
858 def to_stderr(self
, message
, only_once
=False):
859 """Print message to stderr"""
860 assert isinstance(message
, str)
861 if self
.params
.get('logger'):
862 self
.params
['logger'].error(message
)
864 self
._write
_string
(f
'{self._bidi_workaround(message)}\n', self
._out
_files
.error
, only_once
=only_once
)
866 def _send_console_code(self
, code
):
867 if compat_os_name
== 'nt' or not self
._out
_files
.console
:
869 self
._write
_string
(code
, self
._out
_files
.console
)
871 def to_console_title(self
, message
):
872 if not self
.params
.get('consoletitle', False):
874 message
= remove_terminal_sequences(message
)
875 if compat_os_name
== 'nt':
876 if ctypes
.windll
.kernel32
.GetConsoleWindow():
877 # c_wchar_p() might not be necessary if `message` is
878 # already of type unicode()
879 ctypes
.windll
.kernel32
.SetConsoleTitleW(ctypes
.c_wchar_p(message
))
881 self
._send
_console
_code
(f
'\033]0;{message}\007')
883 def save_console_title(self
):
884 if not self
.params
.get('consoletitle') or self
.params
.get('simulate'):
886 self
._send
_console
_code
('\033[22;0t') # Save the title on stack
888 def restore_console_title(self
):
889 if not self
.params
.get('consoletitle') or self
.params
.get('simulate'):
891 self
._send
_console
_code
('\033[23;0t') # Restore the title from stack
894 self
.save_console_title()
897 def __exit__(self
, *args
):
898 self
.restore_console_title()
900 if self
.params
.get('cookiefile') is not None:
901 self
.cookiejar
.save(ignore_discard
=True, ignore_expires
=True)
903 def trouble(self
, message
=None, tb
=None, is_error
=True):
904 """Determine action to take when a download problem appears.
906 Depending on if the downloader has been configured to ignore
907 download errors or not, this method may throw an exception or
908 not when errors are found, after printing the message.
910 @param tb If given, is additional traceback information
911 @param is_error Whether to raise error according to ignorerrors
913 if message
is not None:
914 self
.to_stderr(message
)
915 if self
.params
.get('verbose'):
917 if sys
.exc_info()[0]: # if .trouble has been called from an except block
919 if hasattr(sys
.exc_info()[1], 'exc_info') and sys
.exc_info()[1].exc_info
[0]:
920 tb
+= ''.join(traceback
.format_exception(*sys
.exc_info()[1].exc_info
))
921 tb
+= encode_compat_str(traceback
.format_exc())
923 tb_data
= traceback
.format_list(traceback
.extract_stack())
924 tb
= ''.join(tb_data
)
929 if not self
.params
.get('ignoreerrors'):
930 if sys
.exc_info()[0] and hasattr(sys
.exc_info()[1], 'exc_info') and sys
.exc_info()[1].exc_info
[0]:
931 exc_info
= sys
.exc_info()[1].exc_info
933 exc_info
= sys
.exc_info()
934 raise DownloadError(message
, exc_info
)
935 self
._download
_retcode
= 1
939 EMPHASIS
='light blue',
945 SUPPRESS
='light black',
948 def _format_text(self
, handle
, allow_colors
, text
, f
, fallback
=None, *, test_encoding
=False):
952 # handle.encoding can be None. See https://github.com/yt-dlp/yt-dlp/issues/2711
953 encoding
= self
.params
.get('encoding') or getattr(handle
, 'encoding', None) or 'ascii'
954 text
= text
.encode(encoding
, 'ignore').decode(encoding
)
955 if fallback
is not None and text
!= original_text
:
957 return format_text(text
, f
) if allow_colors
else text
if fallback
is None else fallback
959 def _format_out(self
, *args
, **kwargs
):
960 return self
._format
_text
(self
._out
_files
.out
, self
._allow
_colors
.out
, *args
, **kwargs
)
962 def _format_screen(self
, *args
, **kwargs
):
963 return self
._format
_text
(self
._out
_files
.screen
, self
._allow
_colors
.screen
, *args
, **kwargs
)
965 def _format_err(self
, *args
, **kwargs
):
966 return self
._format
_text
(self
._out
_files
.error
, self
._allow
_colors
.error
, *args
, **kwargs
)
968 def report_warning(self
, message
, only_once
=False):
970 Print the message to stderr, it will be prefixed with 'WARNING:'
971 If stderr is a tty file the 'WARNING:' will be colored
973 if self
.params
.get('logger') is not None:
974 self
.params
['logger'].warning(message
)
976 if self
.params
.get('no_warnings'):
978 self
.to_stderr(f
'{self._format_err("WARNING:", self.Styles.WARNING)} {message}', only_once
)
980 def deprecation_warning(self
, message
, *, stacklevel
=0):
982 message
, stacklevel
=stacklevel
+ 1, printer
=self
.report_error
, is_error
=False)
984 def deprecated_feature(self
, message
):
985 if self
.params
.get('logger') is not None:
986 self
.params
['logger'].warning(f
'Deprecated Feature: {message}')
987 self
.to_stderr(f
'{self._format_err("Deprecated Feature:", self.Styles.ERROR)} {message}', True)
989 def report_error(self
, message
, *args
, **kwargs
):
991 Do the same as trouble, but prefixes the message with 'ERROR:', colored
992 in red if stderr is a tty file.
994 self
.trouble(f
'{self._format_err("ERROR:", self.Styles.ERROR)} {message}', *args
, **kwargs
)
996 def write_debug(self
, message
, only_once
=False):
997 '''Log debug message or Print message to stderr'''
998 if not self
.params
.get('verbose', False):
1000 message
= f
'[debug] {message}'
1001 if self
.params
.get('logger'):
1002 self
.params
['logger'].debug(message
)
1004 self
.to_stderr(message
, only_once
)
1006 def report_file_already_downloaded(self
, file_name
):
1007 """Report file has already been fully downloaded."""
1009 self
.to_screen('[download] %s has already been downloaded' % file_name
)
1010 except UnicodeEncodeError:
1011 self
.to_screen('[download] The file has already been downloaded')
1013 def report_file_delete(self
, file_name
):
1014 """Report that existing file will be deleted."""
1016 self
.to_screen('Deleting existing file %s' % file_name
)
1017 except UnicodeEncodeError:
1018 self
.to_screen('Deleting existing file')
1020 def raise_no_formats(self
, info
, forced
=False, *, msg
=None):
1021 has_drm
= info
.get('_has_drm')
1022 ignored
, expected
= self
.params
.get('ignore_no_formats_error'), bool(msg
)
1023 msg
= msg
or has_drm
and 'This video is DRM protected' or 'No video formats found!'
1024 if forced
or not ignored
:
1025 raise ExtractorError(msg
, video_id
=info
['id'], ie
=info
['extractor'],
1026 expected
=has_drm
or ignored
or expected
)
1028 self
.report_warning(msg
)
1030 def parse_outtmpl(self
):
1031 self
.deprecation_warning('"YoutubeDL.parse_outtmpl" is deprecated and may be removed in a future version')
1032 self
._parse
_outtmpl
()
1033 return self
.params
['outtmpl']
1035 def _parse_outtmpl(self
):
1037 if self
.params
.get('restrictfilenames'): # Remove spaces in the default template
1038 sanitize
= lambda x
: x
.replace(' - ', ' ').replace(' ', '-')
1040 outtmpl
= self
.params
.setdefault('outtmpl', {})
1041 if not isinstance(outtmpl
, dict):
1042 self
.params
['outtmpl'] = outtmpl
= {'default': outtmpl}
1043 outtmpl
.update({k: sanitize(v) for k, v in DEFAULT_OUTTMPL.items() if outtmpl.get(k) is None}
)
1045 def get_output_path(self
, dir_type
='', filename
=None):
1046 paths
= self
.params
.get('paths', {})
1047 assert isinstance(paths
, dict), '"paths" parameter must be a dictionary'
1048 path
= os
.path
.join(
1049 expand_path(paths
.get('home', '').strip()),
1050 expand_path(paths
.get(dir_type
, '').strip()) if dir_type
else '',
1052 return sanitize_path(path
, force
=self
.params
.get('windowsfilenames'))
1055 def _outtmpl_expandpath(outtmpl
):
1056 # expand_path translates '%%' into '%' and '$$' into '$'
1057 # correspondingly that is not what we want since we need to keep
1058 # '%%' intact for template dict substitution step. Working around
1059 # with boundary-alike separator hack.
1060 sep
= ''.join([random
.choice(ascii_letters
) for _
in range(32)])
1061 outtmpl
= outtmpl
.replace('%%', f
'%{sep}%').replace('$$', f
'${sep}$')
1063 # outtmpl should be expand_path'ed before template dict substitution
1064 # because meta fields may contain env variables we don't want to
1065 # be expanded. E.g. for outtmpl "%(title)s.%(ext)s" and
1066 # title "Hello $PATH", we don't want `$PATH` to be expanded.
1067 return expand_path(outtmpl
).replace(sep
, '')
1070 def escape_outtmpl(outtmpl
):
1071 ''' Escape any remaining strings like %s, %abc% etc. '''
1073 STR_FORMAT_RE_TMPL
.format('', '(?![%(\0])'),
1074 lambda mobj
: ('' if mobj
.group('has_key') else '%') + mobj
.group(0),
1078 def validate_outtmpl(cls
, outtmpl
):
1079 ''' @return None or Exception object '''
1081 STR_FORMAT_RE_TMPL
.format('[^)]*', '[ljhqBUDS]'),
1082 lambda mobj
: f
'{mobj.group(0)[:-1]}s',
1083 cls
._outtmpl
_expandpath
(outtmpl
))
1085 cls
.escape_outtmpl(outtmpl
) % collections
.defaultdict(int)
1087 except ValueError as err
:
1091 def _copy_infodict(info_dict
):
1092 info_dict
= dict(info_dict
)
1093 info_dict
.pop('__postprocessors', None)
1094 info_dict
.pop('__pending_error', None)
1097 def prepare_outtmpl(self
, outtmpl
, info_dict
, sanitize
=False):
1098 """ Make the outtmpl and info_dict suitable for substitution: ydl.escape_outtmpl(outtmpl) % info_dict
1099 @param sanitize Whether to sanitize the output as a filename.
1100 For backward compatibility, a function can also be passed
1103 info_dict
.setdefault('epoch', int(time
.time())) # keep epoch consistent once set
1105 info_dict
= self
._copy
_infodict
(info_dict
)
1106 info_dict
['duration_string'] = ( # %(duration>%H-%M-%S)s is wrong if duration > 24hrs
1107 formatSeconds(info_dict
['duration'], '-' if sanitize
else ':')
1108 if info_dict
.get('duration', None) is not None
1110 info_dict
['autonumber'] = int(self
.params
.get('autonumber_start', 1) - 1 + self
._num
_downloads
)
1111 info_dict
['video_autonumber'] = self
._num
_videos
1112 if info_dict
.get('resolution') is None:
1113 info_dict
['resolution'] = self
.format_resolution(info_dict
, default
=None)
1115 # For fields playlist_index, playlist_autonumber and autonumber convert all occurrences
1116 # of %(field)s to %(field)0Nd for backward compatibility
1117 field_size_compat_map
= {
1118 'playlist_index': number_of_digits(info_dict
.get('__last_playlist_index') or 0),
1119 'playlist_autonumber': number_of_digits(info_dict
.get('n_entries') or 0),
1120 'autonumber': self
.params
.get('autonumber_size') or 5,
1124 EXTERNAL_FORMAT_RE
= re
.compile(STR_FORMAT_RE_TMPL
.format('[^)]*', f
'[{STR_FORMAT_TYPES}ljhqBUDS]'))
1129 # Field is of the form key1.key2...
1130 # where keys (except first) can be string, int, slice or "{field, ...}"
1131 FIELD_INNER_RE
= r
'(?:\w+|%(num)s|%(num)s?(?::%(num)s?){1,2})' % {'num': r'(?:-?\d+)'}
1132 FIELD_RE
= r
'\w*(?:\.(?:%(inner)s|{%(field)s(?:,%(field)s)*}))*' % {
1133 'inner': FIELD_INNER_RE
,
1134 'field': rf
'\w*(?:\.{FIELD_INNER_RE})*'
1136 MATH_FIELD_RE
= rf
'(?:{FIELD_RE}|-?{NUMBER_RE})'
1137 MATH_OPERATORS_RE
= r
'(?:%s)' % '|'.join(map(re
.escape
, MATH_FUNCTIONS
.keys()))
1138 INTERNAL_FORMAT_RE
= re
.compile(rf
'''(?x)
1140 (?P<fields>{FIELD_RE})
1141 (?P<maths>(?:{MATH_OPERATORS_RE}{MATH_FIELD_RE})*)
1142 (?:>(?P<strf_format>.+?))?
1144 (?P<alternate>(?<!\\),[^|&)]+)?
1145 (?:&(?P<replacement>.*?))?
1146 (?:\|(?P<default>.*?))?
1149 def _traverse_infodict(fields
):
1150 fields
= [f
for x
in re
.split(r
'\.({.+?})\.?', fields
)
1151 for f
in ([x
] if x
.startswith('{') else x
.split('.'))]
1153 if fields
and not fields
[i
]:
1156 for i
, f
in enumerate(fields
):
1157 if not f
.startswith('{'):
1159 assert f
.endswith('}'), f
'No closing brace for {f} in {fields}'
1160 fields
[i
] = {k: k.split('.') for k in f[1:-1].split(',')}
1162 return traverse_obj(info_dict
, fields
, is_user_input
=True, traverse_string
=True)
1164 def get_value(mdict
):
1166 value
= _traverse_infodict(mdict
['fields'])
1169 value
= float_or_none(value
)
1170 if value
is not None:
1173 offset_key
= mdict
['maths']
1175 value
= float_or_none(value
)
1179 MATH_FIELD_RE
if operator
else MATH_OPERATORS_RE
,
1180 offset_key
).group(0)
1181 offset_key
= offset_key
[len(item
):]
1182 if operator
is None:
1183 operator
= MATH_FUNCTIONS
[item
]
1185 item
, multiplier
= (item
[1:], -1) if item
[0] == '-' else (item
, 1)
1186 offset
= float_or_none(item
)
1188 offset
= float_or_none(_traverse_infodict(item
))
1190 value
= operator(value
, multiplier
* offset
)
1191 except (TypeError, ZeroDivisionError):
1194 # Datetime formatting
1195 if mdict
['strf_format']:
1196 value
= strftime_or_none(value
, mdict
['strf_format'].replace('\\,', ','))
1198 # XXX: Workaround for https://github.com/yt-dlp/yt-dlp/issues/4485
1199 if sanitize
and value
== '':
1203 na
= self
.params
.get('outtmpl_na_placeholder', 'NA')
1205 def filename_sanitizer(key
, value
, restricted
=self
.params
.get('restrictfilenames')):
1206 return sanitize_filename(str(value
), restricted
=restricted
, is_id
=(
1207 bool(re
.search(r
'(^|[_.])id(\.|$)', key
))
1208 if 'filename-sanitization' in self
.params
['compat_opts']
1211 sanitizer
= sanitize
if callable(sanitize
) else filename_sanitizer
1212 sanitize
= bool(sanitize
)
1214 def _dumpjson_default(obj
):
1215 if isinstance(obj
, (set, LazyList
)):
1219 def create_key(outer_mobj
):
1220 if not outer_mobj
.group('has_key'):
1221 return outer_mobj
.group(0)
1222 key
= outer_mobj
.group('key')
1223 mobj
= re
.match(INTERNAL_FORMAT_RE
, key
)
1224 initial_field
= mobj
.group('fields') if mobj
else ''
1225 value
, replacement
, default
= None, None, na
1227 mobj
= mobj
.groupdict()
1228 default
= mobj
['default'] if mobj
['default'] is not None else default
1229 value
= get_value(mobj
)
1230 replacement
= mobj
['replacement']
1231 if value
is None and mobj
['alternate']:
1232 mobj
= re
.match(INTERNAL_FORMAT_RE
, mobj
['remaining'][1:])
1236 fmt
= outer_mobj
.group('format')
1237 if fmt
== 's' and value
is not None and key
in field_size_compat_map
.keys():
1238 fmt
= f
'0{field_size_compat_map[key]:d}d'
1240 value
= default
if value
is None else value
if replacement
is None else replacement
1242 flags
= outer_mobj
.group('conversion') or ''
1243 str_fmt
= f
'{fmt[:-1]}s'
1244 if fmt
[-1] == 'l': # list
1245 delim
= '\n' if '#' in flags
else ', '
1246 value
, fmt
= delim
.join(map(str, variadic(value
, allowed_types
=(str, bytes)))), str_fmt
1247 elif fmt
[-1] == 'j': # json
1248 value
, fmt
= json
.dumps(value
, default
=_dumpjson_default
, indent
=4 if '#' in flags
else None), str_fmt
1249 elif fmt
[-1] == 'h': # html
1250 value
, fmt
= escapeHTML(value
), str_fmt
1251 elif fmt
[-1] == 'q': # quoted
1252 value
= map(str, variadic(value
) if '#' in flags
else [value
])
1253 value
, fmt
= ' '.join(map(compat_shlex_quote
, value
)), str_fmt
1254 elif fmt
[-1] == 'B': # bytes
1255 value
= f
'%{str_fmt}'.encode() % str(value
).encode()
1256 value
, fmt
= value
.decode('utf-8', 'ignore'), 's'
1257 elif fmt
[-1] == 'U': # unicode normalized
1258 value
, fmt
= unicodedata
.normalize(
1259 # "+" = compatibility equivalence, "#" = NFD
1260 'NF%s%s' % ('K' if '+' in flags
else '', 'D' if '#' in flags
else 'C'),
1262 elif fmt
[-1] == 'D': # decimal suffix
1263 num_fmt
, fmt
= fmt
[:-1].replace('#', ''), 's'
1264 value
= format_decimal_suffix(value
, f
'%{num_fmt}f%s' if num_fmt
else '%d%s',
1265 factor
=1024 if '#' in flags
else 1000)
1266 elif fmt
[-1] == 'S': # filename sanitization
1267 value
, fmt
= filename_sanitizer(initial_field
, value
, restricted
='#' in flags
), str_fmt
1268 elif fmt
[-1] == 'c':
1270 value
= str(value
)[0]
1273 elif fmt
[-1] not in 'rs': # numeric
1274 value
= float_or_none(value
)
1276 value
, fmt
= default
, 's'
1280 # If value is an object, sanitize might convert it to a string
1281 # So we convert it to repr first
1282 value
, fmt
= repr(value
), str_fmt
1283 if fmt
[-1] in 'csr':
1284 value
= sanitizer(initial_field
, value
)
1286 key
= '%s\0%s' % (key
.replace('%', '%\0'), outer_mobj
.group('format'))
1287 TMPL_DICT
[key
] = value
1288 return '{prefix}%({key}){fmt}'.format(key
=key
, fmt
=fmt
, prefix
=outer_mobj
.group('prefix'))
1290 return EXTERNAL_FORMAT_RE
.sub(create_key
, outtmpl
), TMPL_DICT
1292 def evaluate_outtmpl(self
, outtmpl
, info_dict
, *args
, **kwargs
):
1293 outtmpl
, info_dict
= self
.prepare_outtmpl(outtmpl
, info_dict
, *args
, **kwargs
)
1294 return self
.escape_outtmpl(outtmpl
) % info_dict
1296 def _prepare_filename(self
, info_dict
, *, outtmpl
=None, tmpl_type
=None):
1297 assert None in (outtmpl
, tmpl_type
), 'outtmpl and tmpl_type are mutually exclusive'
1299 outtmpl
= self
.params
['outtmpl'].get(tmpl_type
or 'default', self
.params
['outtmpl']['default'])
1301 outtmpl
= self
._outtmpl
_expandpath
(outtmpl
)
1302 filename
= self
.evaluate_outtmpl(outtmpl
, info_dict
, True)
1306 if tmpl_type
in ('', 'temp'):
1307 final_ext
, ext
= self
.params
.get('final_ext'), info_dict
.get('ext')
1308 if final_ext
and ext
and final_ext
!= ext
and filename
.endswith(f
'.{final_ext}'):
1309 filename
= replace_extension(filename
, ext
, final_ext
)
1311 force_ext
= OUTTMPL_TYPES
[tmpl_type
]
1313 filename
= replace_extension(filename
, force_ext
, info_dict
.get('ext'))
1315 # https://github.com/blackjack4494/youtube-dlc/issues/85
1316 trim_file_name
= self
.params
.get('trim_file_name', False)
1318 no_ext
, *ext
= filename
.rsplit('.', 2)
1319 filename
= join_nonempty(no_ext
[:trim_file_name
], *ext
, delim
='.')
1322 except ValueError as err
:
1323 self
.report_error('Error in output template: ' + str(err
) + ' (encoding: ' + repr(preferredencoding()) + ')')
1326 def prepare_filename(self
, info_dict
, dir_type
='', *, outtmpl
=None, warn
=False):
1327 """Generate the output filename"""
1329 assert not dir_type
, 'outtmpl and dir_type are mutually exclusive'
1331 filename
= self
._prepare
_filename
(info_dict
, tmpl_type
=dir_type
, outtmpl
=outtmpl
)
1332 if not filename
and dir_type
not in ('', 'temp'):
1336 if not self
.params
.get('paths'):
1338 elif filename
== '-':
1339 self
.report_warning('--paths is ignored when an outputting to stdout', only_once
=True)
1340 elif os
.path
.isabs(filename
):
1341 self
.report_warning('--paths is ignored since an absolute path is given in output template', only_once
=True)
1342 if filename
== '-' or not filename
:
1345 return self
.get_output_path(dir_type
, filename
)
1347 def _match_entry(self
, info_dict
, incomplete
=False, silent
=False):
1348 """ Returns None if the file should be downloaded """
1350 video_title
= info_dict
.get('title', info_dict
.get('id', 'entry'))
1353 if 'title' in info_dict
:
1354 # This can happen when we're just evaluating the playlist
1355 title
= info_dict
['title']
1356 matchtitle
= self
.params
.get('matchtitle', False)
1358 if not re
.search(matchtitle
, title
, re
.IGNORECASE
):
1359 return '"' + title
+ '" title did not match pattern "' + matchtitle
+ '"'
1360 rejecttitle
= self
.params
.get('rejecttitle', False)
1362 if re
.search(rejecttitle
, title
, re
.IGNORECASE
):
1363 return '"' + title
+ '" title matched reject pattern "' + rejecttitle
+ '"'
1364 date
= info_dict
.get('upload_date')
1365 if date
is not None:
1366 dateRange
= self
.params
.get('daterange', DateRange())
1367 if date
not in dateRange
:
1368 return f
'{date_from_str(date).isoformat()} upload date is not in range {dateRange}'
1369 view_count
= info_dict
.get('view_count')
1370 if view_count
is not None:
1371 min_views
= self
.params
.get('min_views')
1372 if min_views
is not None and view_count
< min_views
:
1373 return 'Skipping %s, because it has not reached minimum view count (%d/%d)' % (video_title
, view_count
, min_views
)
1374 max_views
= self
.params
.get('max_views')
1375 if max_views
is not None and view_count
> max_views
:
1376 return 'Skipping %s, because it has exceeded the maximum view count (%d/%d)' % (video_title
, view_count
, max_views
)
1377 if age_restricted(info_dict
.get('age_limit'), self
.params
.get('age_limit')):
1378 return 'Skipping "%s" because it is age restricted' % video_title
1380 match_filter
= self
.params
.get('match_filter')
1381 if match_filter
is not None:
1383 ret
= match_filter(info_dict
, incomplete
=incomplete
)
1385 # For backward compatibility
1386 ret
= None if incomplete
else match_filter(info_dict
)
1387 if ret
is NO_DEFAULT
:
1389 filename
= self
._format
_screen
(self
.prepare_filename(info_dict
), self
.Styles
.FILENAME
)
1390 reply
= input(self
._format
_screen
(
1391 f
'Download "{filename}"? (Y/n): ', self
.Styles
.EMPHASIS
)).lower().strip()
1392 if reply
in {'y', ''}
:
1395 return f
'Skipping {video_title}'
1396 elif ret
is not None:
1400 if self
.in_download_archive(info_dict
):
1401 reason
= '%s has already been recorded in the archive' % video_title
1402 break_opt
, break_err
= 'break_on_existing', ExistingVideoReached
1404 reason
= check_filter()
1405 break_opt
, break_err
= 'break_on_reject', RejectedVideoReached
1406 if reason
is not None:
1408 self
.to_screen('[download] ' + reason
)
1409 if self
.params
.get(break_opt
, False):
1414 def add_extra_info(info_dict
, extra_info
):
1415 '''Set the keys from extra_info in info dict if they are missing'''
1416 for key
, value
in extra_info
.items():
1417 info_dict
.setdefault(key
, value
)
1419 def extract_info(self
, url
, download
=True, ie_key
=None, extra_info
=None,
1420 process
=True, force_generic_extractor
=False):
1422 Return a list with a dictionary for each video extracted.
1425 url -- URL to extract
1428 download -- whether to download videos during extraction
1429 ie_key -- extractor key hint
1430 extra_info -- dictionary containing the extra values to add to each result
1431 process -- whether to resolve all unresolved references (URLs, playlist items),
1432 must be True for download to work.
1433 force_generic_extractor -- force using the generic extractor
1436 if extra_info
is None:
1439 if not ie_key
and force_generic_extractor
:
1443 ies
= {ie_key: self._ies[ie_key]}
if ie_key
in self
._ies
else {}
1447 for key
, ie
in ies
.items():
1448 if not ie
.suitable(url
):
1451 if not ie
.working():
1452 self
.report_warning('The program functionality for this site has been marked as broken, '
1453 'and will probably not work.')
1455 temp_id
= ie
.get_temp_id(url
)
1456 if temp_id
is not None and self
.in_download_archive({'id': temp_id, 'ie_key': key}
):
1457 self
.to_screen(f
'[{key}] {temp_id}: has already been recorded in the archive')
1458 if self
.params
.get('break_on_existing', False):
1459 raise ExistingVideoReached()
1461 return self
.__extract
_info
(url
, self
.get_info_extractor(key
), download
, extra_info
, process
)
1463 extractors_restricted
= self
.params
.get('allowed_extractors') not in (None, ['default'])
1464 self
.report_error(f
'No suitable extractor{format_field(ie_key, None, " (%s)")} found for URL {url}',
1465 tb
=False if extractors_restricted
else None)
1467 def _handle_extraction_exceptions(func
):
1468 @functools.wraps(func
)
1469 def wrapper(self
, *args
, **kwargs
):
1472 return func(self
, *args
, **kwargs
)
1473 except (DownloadCancelled
, LazyList
.IndexError, PagedList
.IndexError):
1475 except ReExtractInfo
as e
:
1477 self
.to_screen(f
'{e}; Re-extracting data')
1479 self
.to_stderr('\r')
1480 self
.report_warning(f
'{e}; Re-extracting data')
1482 except GeoRestrictedError
as e
:
1485 msg
+= '\nThis video is available in %s.' % ', '.join(
1486 map(ISO3166Utils
.short2full
, e
.countries
))
1487 msg
+= '\nYou might want to use a VPN or a proxy server (with --proxy) to workaround.'
1488 self
.report_error(msg
)
1489 except ExtractorError
as e
: # An error we somewhat expected
1490 self
.report_error(str(e
), e
.format_traceback())
1491 except Exception as e
:
1492 if self
.params
.get('ignoreerrors'):
1493 self
.report_error(str(e
), tb
=encode_compat_str(traceback
.format_exc()))
1499 def _wait_for_video(self
, ie_result
={}):
1500 if (not self
.params
.get('wait_for_video')
1501 or ie_result
.get('_type', 'video') != 'video'
1502 or ie_result
.get('formats') or ie_result
.get('url')):
1505 format_dur
= lambda dur
: '%02d:%02d:%02d' % timetuple_from_msec(dur
* 1000)[:-1]
1510 full_msg
= f
'{msg}\n'
1511 if not self
.params
.get('noprogress'):
1512 full_msg
= msg
+ ' ' * (len(last_msg
) - len(msg
)) + '\r'
1515 self
.to_screen(full_msg
, skip_eol
=True)
1518 min_wait
, max_wait
= self
.params
.get('wait_for_video')
1519 diff
= try_get(ie_result
, lambda x
: x
['release_timestamp'] - time
.time())
1520 if diff
is None and ie_result
.get('live_status') == 'is_upcoming':
1521 diff
= round(random
.uniform(min_wait
, max_wait
) if (max_wait
and min_wait
) else (max_wait
or min_wait
), 0)
1522 self
.report_warning('Release time of video is not known')
1523 elif ie_result
and (diff
or 0) <= 0:
1524 self
.report_warning('Video should already be available according to extracted info')
1525 diff
= min(max(diff
or 0, min_wait
or 0), max_wait
or float('inf'))
1526 self
.to_screen(f
'[wait] Waiting for {format_dur(diff)} - Press Ctrl+C to try now')
1528 wait_till
= time
.time() + diff
1531 diff
= wait_till
- time
.time()
1534 raise ReExtractInfo('[wait] Wait period ended', expected
=True)
1535 progress(f
'[wait] Remaining time until next attempt: {self._format_screen(format_dur(diff), self.Styles.EMPHASIS)}')
1537 except KeyboardInterrupt:
1539 raise ReExtractInfo('[wait] Interrupted by user', expected
=True)
1540 except BaseException
as e
:
1541 if not isinstance(e
, ReExtractInfo
):
1545 @_handle_extraction_exceptions
1546 def __extract_info(self
, url
, ie
, download
, extra_info
, process
):
1548 ie_result
= ie
.extract(url
)
1549 except UserNotLive
as e
:
1551 if self
.params
.get('wait_for_video'):
1552 self
.report_warning(e
)
1553 self
._wait
_for
_video
()
1555 if ie_result
is None: # Finished already (backwards compatibility; listformats and friends should be moved here)
1556 self
.report_warning(f
'Extractor {ie.IE_NAME} returned nothing{bug_reports_message()}')
1558 if isinstance(ie_result
, list):
1559 # Backwards compatibility: old IE result format
1561 '_type': 'compat_list',
1562 'entries': ie_result
,
1564 if extra_info
.get('original_url'):
1565 ie_result
.setdefault('original_url', extra_info
['original_url'])
1566 self
.add_default_extra_info(ie_result
, ie
, url
)
1568 self
._wait
_for
_video
(ie_result
)
1569 return self
.process_ie_result(ie_result
, download
, extra_info
)
1573 def add_default_extra_info(self
, ie_result
, ie
, url
):
1575 self
.add_extra_info(ie_result
, {
1577 'original_url': url
,
1579 webpage_url
= ie_result
.get('webpage_url')
1581 self
.add_extra_info(ie_result
, {
1582 'webpage_url_basename': url_basename(webpage_url
),
1583 'webpage_url_domain': get_domain(webpage_url
),
1586 self
.add_extra_info(ie_result
, {
1587 'extractor': ie
.IE_NAME
,
1588 'extractor_key': ie
.ie_key(),
1591 def process_ie_result(self
, ie_result
, download
=True, extra_info
=None):
1593 Take the result of the ie(may be modified) and resolve all unresolved
1594 references (URLs, playlist items).
1596 It will also download the videos if 'download'.
1597 Returns the resolved ie_result.
1599 if extra_info
is None:
1601 result_type
= ie_result
.get('_type', 'video')
1603 if result_type
in ('url', 'url_transparent'):
1604 ie_result
['url'] = sanitize_url(
1605 ie_result
['url'], scheme
='http' if self
.params
.get('prefer_insecure') else 'https')
1606 if ie_result
.get('original_url'):
1607 extra_info
.setdefault('original_url', ie_result
['original_url'])
1609 extract_flat
= self
.params
.get('extract_flat', False)
1610 if ((extract_flat
== 'in_playlist' and 'playlist' in extra_info
)
1611 or extract_flat
is True):
1612 info_copy
= ie_result
.copy()
1613 ie
= try_get(ie_result
.get('ie_key'), self
.get_info_extractor
)
1614 if ie
and not ie_result
.get('id'):
1615 info_copy
['id'] = ie
.get_temp_id(ie_result
['url'])
1616 self
.add_default_extra_info(info_copy
, ie
, ie_result
['url'])
1617 self
.add_extra_info(info_copy
, extra_info
)
1618 info_copy
, _
= self
.pre_process(info_copy
)
1619 self
.__forced
_printings
(info_copy
, self
.prepare_filename(info_copy
), incomplete
=True)
1620 self
._raise
_pending
_errors
(info_copy
)
1621 if self
.params
.get('force_write_download_archive', False):
1622 self
.record_download_archive(info_copy
)
1625 if result_type
== 'video':
1626 self
.add_extra_info(ie_result
, extra_info
)
1627 ie_result
= self
.process_video_result(ie_result
, download
=download
)
1628 self
._raise
_pending
_errors
(ie_result
)
1629 additional_urls
= (ie_result
or {}).get('additional_urls')
1631 # TODO: Improve MetadataParserPP to allow setting a list
1632 if isinstance(additional_urls
, str):
1633 additional_urls
= [additional_urls
]
1635 '[info] %s: %d additional URL(s) requested' % (ie_result
['id'], len(additional_urls
)))
1636 self
.write_debug('Additional URLs: "%s"' % '", "'.join(additional_urls
))
1637 ie_result
['additional_entries'] = [
1639 url
, download
, extra_info
=extra_info
,
1640 force_generic_extractor
=self
.params
.get('force_generic_extractor'))
1641 for url
in additional_urls
1644 elif result_type
== 'url':
1645 # We have to add extra_info to the results because it may be
1646 # contained in a playlist
1647 return self
.extract_info(
1648 ie_result
['url'], download
,
1649 ie_key
=ie_result
.get('ie_key'),
1650 extra_info
=extra_info
)
1651 elif result_type
== 'url_transparent':
1652 # Use the information from the embedding page
1653 info
= self
.extract_info(
1654 ie_result
['url'], ie_key
=ie_result
.get('ie_key'),
1655 extra_info
=extra_info
, download
=False, process
=False)
1657 # extract_info may return None when ignoreerrors is enabled and
1658 # extraction failed with an error, don't crash and return early
1663 exempted_fields
= {'_type', 'url', 'ie_key'}
1664 if not ie_result
.get('section_end') and ie_result
.get('section_start') is None:
1665 # For video clips, the id etc of the clip extractor should be used
1666 exempted_fields |
= {'id', 'extractor', 'extractor_key'}
1668 new_result
= info
.copy()
1669 new_result
.update(filter_dict(ie_result
, lambda k
, v
: v
is not None and k
not in exempted_fields
))
1671 # Extracted info may not be a video result (i.e.
1672 # info.get('_type', 'video') != video) but rather an url or
1673 # url_transparent. In such cases outer metadata (from ie_result)
1674 # should be propagated to inner one (info). For this to happen
1675 # _type of info should be overridden with url_transparent. This
1676 # fixes issue from https://github.com/ytdl-org/youtube-dl/pull/11163.
1677 if new_result
.get('_type') == 'url':
1678 new_result
['_type'] = 'url_transparent'
1680 return self
.process_ie_result(
1681 new_result
, download
=download
, extra_info
=extra_info
)
1682 elif result_type
in ('playlist', 'multi_video'):
1683 # Protect from infinite recursion due to recursively nested playlists
1684 # (see https://github.com/ytdl-org/youtube-dl/issues/27833)
1685 webpage_url
= ie_result
['webpage_url']
1686 if webpage_url
in self
._playlist
_urls
:
1688 '[download] Skipping already downloaded playlist: %s'
1689 % ie_result
.get('title') or ie_result
.get('id'))
1692 self
._playlist
_level
+= 1
1693 self
._playlist
_urls
.add(webpage_url
)
1694 self
._fill
_common
_fields
(ie_result
, False)
1695 self
._sanitize
_thumbnails
(ie_result
)
1697 return self
.__process
_playlist
(ie_result
, download
)
1699 self
._playlist
_level
-= 1
1700 if not self
._playlist
_level
:
1701 self
._playlist
_urls
.clear()
1702 elif result_type
== 'compat_list':
1703 self
.report_warning(
1704 'Extractor %s returned a compat_list result. '
1705 'It needs to be updated.' % ie_result
.get('extractor'))
1708 self
.add_extra_info(r
, {
1709 'extractor': ie_result
['extractor'],
1710 'webpage_url': ie_result
['webpage_url'],
1711 'webpage_url_basename': url_basename(ie_result
['webpage_url']),
1712 'webpage_url_domain': get_domain(ie_result
['webpage_url']),
1713 'extractor_key': ie_result
['extractor_key'],
1716 ie_result
['entries'] = [
1717 self
.process_ie_result(_fixup(r
), download
, extra_info
)
1718 for r
in ie_result
['entries']
1722 raise Exception('Invalid result type: %s' % result_type
)
1724 def _ensure_dir_exists(self
, path
):
1725 return make_dir(path
, self
.report_error
)
1728 def _playlist_infodict(ie_result
, strict
=False, **kwargs
):
1730 'playlist_count': ie_result
.get('playlist_count'),
1731 'playlist': ie_result
.get('title') or ie_result
.get('id'),
1732 'playlist_id': ie_result
.get('id'),
1733 'playlist_title': ie_result
.get('title'),
1734 'playlist_uploader': ie_result
.get('uploader'),
1735 'playlist_uploader_id': ie_result
.get('uploader_id'),
1742 'playlist_index': 0,
1743 '__last_playlist_index': max(ie_result
['requested_entries'] or (0, 0)),
1744 'extractor': ie_result
['extractor'],
1745 'webpage_url': ie_result
['webpage_url'],
1746 'webpage_url_basename': url_basename(ie_result
['webpage_url']),
1747 'webpage_url_domain': get_domain(ie_result
['webpage_url']),
1748 'extractor_key': ie_result
['extractor_key'],
1751 def __process_playlist(self
, ie_result
, download
):
1752 """Process each entry in the playlist"""
1753 assert ie_result
['_type'] in ('playlist', 'multi_video')
1755 common_info
= self
._playlist
_infodict
(ie_result
, strict
=True)
1756 title
= common_info
.get('playlist') or '<Untitled>'
1757 if self
._match
_entry
(common_info
, incomplete
=True) is not None:
1759 self
.to_screen(f
'[download] Downloading {ie_result["_type"]}: {title}')
1761 all_entries
= PlaylistEntries(self
, ie_result
)
1762 entries
= orderedSet(all_entries
.get_requested_items(), lazy
=True)
1764 lazy
= self
.params
.get('lazy_playlist')
1766 resolved_entries
, n_entries
= [], 'N/A'
1767 ie_result
['requested_entries'], ie_result
['entries'] = None, None
1769 entries
= resolved_entries
= list(entries
)
1770 n_entries
= len(resolved_entries
)
1771 ie_result
['requested_entries'], ie_result
['entries'] = tuple(zip(*resolved_entries
)) or ([], [])
1772 if not ie_result
.get('playlist_count'):
1773 # Better to do this after potentially exhausting entries
1774 ie_result
['playlist_count'] = all_entries
.get_full_count()
1776 extra
= self
._playlist
_infodict
(ie_result
, n_entries
=int_or_none(n_entries
))
1777 ie_copy
= collections
.ChainMap(ie_result
, extra
)
1779 _infojson_written
= False
1780 write_playlist_files
= self
.params
.get('allow_playlist_files', True)
1781 if write_playlist_files
and self
.params
.get('list_thumbnails'):
1782 self
.list_thumbnails(ie_result
)
1783 if write_playlist_files
and not self
.params
.get('simulate'):
1784 _infojson_written
= self
._write
_info
_json
(
1785 'playlist', ie_result
, self
.prepare_filename(ie_copy
, 'pl_infojson'))
1786 if _infojson_written
is None:
1788 if self
._write
_description
('playlist', ie_result
,
1789 self
.prepare_filename(ie_copy
, 'pl_description')) is None:
1791 # TODO: This should be passed to ThumbnailsConvertor if necessary
1792 self
._write
_thumbnails
('playlist', ie_result
, self
.prepare_filename(ie_copy
, 'pl_thumbnail'))
1795 if self
.params
.get('playlistreverse') or self
.params
.get('playlistrandom'):
1796 self
.report_warning('playlistreverse and playlistrandom are not supported with lazy_playlist', only_once
=True)
1797 elif self
.params
.get('playlistreverse'):
1799 elif self
.params
.get('playlistrandom'):
1800 random
.shuffle(entries
)
1802 self
.to_screen(f
'[{ie_result["extractor"]}] Playlist {title}: Downloading {n_entries} videos'
1803 f
'{format_field(ie_result, "playlist_count", " of %s")}')
1805 keep_resolved_entries
= self
.params
.get('extract_flat') != 'discard'
1806 if self
.params
.get('extract_flat') == 'discard_in_playlist':
1807 keep_resolved_entries
= ie_result
['_type'] != 'playlist'
1808 if keep_resolved_entries
:
1809 self
.write_debug('The information of all playlist entries will be held in memory')
1812 max_failures
= self
.params
.get('skip_playlist_after_errors') or float('inf')
1813 for i
, (playlist_index
, entry
) in enumerate(entries
):
1815 resolved_entries
.append((playlist_index
, entry
))
1819 entry
['__x_forwarded_for_ip'] = ie_result
.get('__x_forwarded_for_ip')
1820 if not lazy
and 'playlist-index' in self
.params
.get('compat_opts', []):
1821 playlist_index
= ie_result
['requested_entries'][i
]
1823 entry_copy
= collections
.ChainMap(entry
, {
1825 'n_entries': int_or_none(n_entries
),
1826 'playlist_index': playlist_index
,
1827 'playlist_autonumber': i
+ 1,
1830 if self
._match
_entry
(entry_copy
, incomplete
=True) is not None:
1831 # For compatabilty with youtube-dl. See https://github.com/yt-dlp/yt-dlp/issues/4369
1832 resolved_entries
[i
] = (playlist_index
, NO_DEFAULT
)
1835 self
.to_screen('[download] Downloading video %s of %s' % (
1836 self
._format
_screen
(i
+ 1, self
.Styles
.ID
), self
._format
_screen
(n_entries
, self
.Styles
.EMPHASIS
)))
1839 'playlist_index': playlist_index
,
1840 'playlist_autonumber': i
+ 1,
1842 entry_result
= self
.__process
_iterable
_entry
(entry
, download
, extra
)
1843 if not entry_result
:
1845 if failures
>= max_failures
:
1847 f
'Skipping the remaining entries in playlist "{title}" since {failures} items failed extraction')
1849 if keep_resolved_entries
:
1850 resolved_entries
[i
] = (playlist_index
, entry_result
)
1852 # Update with processed data
1853 ie_result
['requested_entries'] = [i
for i
, e
in resolved_entries
if e
is not NO_DEFAULT
]
1854 ie_result
['entries'] = [e
for _
, e
in resolved_entries
if e
is not NO_DEFAULT
]
1856 # Write the updated info to json
1857 if _infojson_written
is True and self
._write
_info
_json
(
1858 'updated playlist', ie_result
,
1859 self
.prepare_filename(ie_copy
, 'pl_infojson'), overwrite
=True) is None:
1862 ie_result
= self
.run_all_pps('playlist', ie_result
)
1863 self
.to_screen(f
'[download] Finished downloading playlist: {title}')
1866 @_handle_extraction_exceptions
1867 def __process_iterable_entry(self
, entry
, download
, extra_info
):
1868 return self
.process_ie_result(
1869 entry
, download
=download
, extra_info
=extra_info
)
1871 def _build_format_filter(self
, filter_spec
):
1872 " Returns a function to filter the formats according to the filter_spec "
1882 operator_rex
= re
.compile(r
'''(?x)\s*
1883 (?P<key>width|height|tbr|abr|vbr|asr|filesize|filesize_approx|fps)\s*
1884 (?P<op>%s)(?P<none_inclusive>\s*\?)?\s*
1885 (?P<value>[0-9.]+(?:[kKmMgGtTpPeEzZyY]i?[Bb]?)?)\s*
1886 ''' % '|'.join(map(re
.escape
, OPERATORS
.keys())))
1887 m
= operator_rex
.fullmatch(filter_spec
)
1890 comparison_value
= int(m
.group('value'))
1892 comparison_value
= parse_filesize(m
.group('value'))
1893 if comparison_value
is None:
1894 comparison_value
= parse_filesize(m
.group('value') + 'B')
1895 if comparison_value
is None:
1897 'Invalid value %r in format specification %r' % (
1898 m
.group('value'), filter_spec
))
1899 op
= OPERATORS
[m
.group('op')]
1904 '^=': lambda attr
, value
: attr
.startswith(value
),
1905 '$=': lambda attr
, value
: attr
.endswith(value
),
1906 '*=': lambda attr
, value
: value
in attr
,
1907 '~=': lambda attr
, value
: value
.search(attr
) is not None
1909 str_operator_rex
= re
.compile(r
'''(?x)\s*
1910 (?P<key>[a-zA-Z0-9._-]+)\s*
1911 (?P<negation>!\s*)?(?P<op>%s)\s*(?P<none_inclusive>\?\s*)?
1913 (?P<value>(?(quote)(?:(?!(?P=quote))[^\\]|\\.)+|[\w.-]+))
1914 (?(quote)(?P=quote))\s*
1915 ''' % '|'.join(map(re
.escape
, STR_OPERATORS
.keys())))
1916 m
= str_operator_rex
.fullmatch(filter_spec
)
1918 if m
.group('op') == '~=':
1919 comparison_value
= re
.compile(m
.group('value'))
1921 comparison_value
= re
.sub(r
'''\\([\\"'])''', r
'\1', m
.group('value'))
1922 str_op
= STR_OPERATORS
[m
.group('op')]
1923 if m
.group('negation'):
1924 op
= lambda attr
, value
: not str_op(attr
, value
)
1929 raise SyntaxError('Invalid filter specification %r' % filter_spec
)
1932 actual_value
= f
.get(m
.group('key'))
1933 if actual_value
is None:
1934 return m
.group('none_inclusive')
1935 return op(actual_value
, comparison_value
)
1938 def _check_formats(self
, formats
):
1940 self
.to_screen('[info] Testing format %s' % f
['format_id'])
1941 path
= self
.get_output_path('temp')
1942 if not self
._ensure
_dir
_exists
(f
'{path}/'):
1944 temp_file
= tempfile
.NamedTemporaryFile(suffix
='.tmp', delete
=False, dir=path
or None)
1947 success
, _
= self
.dl(temp_file
.name
, f
, test
=True)
1948 except (DownloadError
, OSError, ValueError) + network_exceptions
:
1951 if os
.path
.exists(temp_file
.name
):
1953 os
.remove(temp_file
.name
)
1955 self
.report_warning('Unable to delete temporary file "%s"' % temp_file
.name
)
1959 self
.to_screen('[info] Unable to download format %s. Skipping...' % f
['format_id'])
1961 def _default_format_spec(self
, info_dict
, download
=True):
1964 merger
= FFmpegMergerPP(self
)
1965 return merger
.available
and merger
.can_merge()
1968 not self
.params
.get('simulate')
1972 or info_dict
.get('is_live') and not self
.params
.get('live_from_start')
1973 or self
.params
['outtmpl']['default'] == '-'))
1976 or self
.params
.get('allow_multiple_audio_streams', False)
1977 or 'format-spec' in self
.params
['compat_opts'])
1980 'best/bestvideo+bestaudio' if prefer_best
1981 else 'bestvideo*+bestaudio/best' if not compat
1982 else 'bestvideo+bestaudio/best')
1984 def build_format_selector(self
, format_spec
):
1985 def syntax_error(note
, start
):
1987 'Invalid format specification: '
1988 '{}\n\t{}\n\t{}^'.format(note
, format_spec
, ' ' * start
[1]))
1989 return SyntaxError(message
)
1991 PICKFIRST
= 'PICKFIRST'
1995 FormatSelector
= collections
.namedtuple('FormatSelector', ['type', 'selector', 'filters'])
1997 allow_multiple_streams
= {'audio': self
.params
.get('allow_multiple_audio_streams', False),
1998 'video': self
.params
.get('allow_multiple_video_streams', False)}
2000 check_formats
= self
.params
.get('check_formats') == 'selected'
2002 def _parse_filter(tokens
):
2004 for type, string
, start
, _
, _
in tokens
:
2005 if type == tokenize
.OP
and string
== ']':
2006 return ''.join(filter_parts
)
2008 filter_parts
.append(string
)
2010 def _remove_unused_ops(tokens
):
2011 # Remove operators that we don't use and join them with the surrounding strings.
2012 # E.g. 'mp4' '-' 'baseline' '-' '16x9' is converted to 'mp4-baseline-16x9'
2013 ALLOWED_OPS
= ('/', '+', ',', '(', ')')
2014 last_string
, last_start
, last_end
, last_line
= None, None, None, None
2015 for type, string
, start
, end
, line
in tokens
:
2016 if type == tokenize
.OP
and string
== '[':
2018 yield tokenize
.NAME
, last_string
, last_start
, last_end
, last_line
2020 yield type, string
, start
, end
, line
2021 # everything inside brackets will be handled by _parse_filter
2022 for type, string
, start
, end
, line
in tokens
:
2023 yield type, string
, start
, end
, line
2024 if type == tokenize
.OP
and string
== ']':
2026 elif type == tokenize
.OP
and string
in ALLOWED_OPS
:
2028 yield tokenize
.NAME
, last_string
, last_start
, last_end
, last_line
2030 yield type, string
, start
, end
, line
2031 elif type in [tokenize
.NAME
, tokenize
.NUMBER
, tokenize
.OP
]:
2033 last_string
= string
2037 last_string
+= string
2039 yield tokenize
.NAME
, last_string
, last_start
, last_end
, last_line
2041 def _parse_format_selection(tokens
, inside_merge
=False, inside_choice
=False, inside_group
=False):
2043 current_selector
= None
2044 for type, string
, start
, _
, _
in tokens
:
2045 # ENCODING is only defined in python 3.x
2046 if type == getattr(tokenize
, 'ENCODING', None):
2048 elif type in [tokenize
.NAME
, tokenize
.NUMBER
]:
2049 current_selector
= FormatSelector(SINGLE
, string
, [])
2050 elif type == tokenize
.OP
:
2052 if not inside_group
:
2053 # ')' will be handled by the parentheses group
2054 tokens
.restore_last_token()
2056 elif inside_merge
and string
in ['/', ',']:
2057 tokens
.restore_last_token()
2059 elif inside_choice
and string
== ',':
2060 tokens
.restore_last_token()
2063 if not current_selector
:
2064 raise syntax_error('"," must follow a format selector', start
)
2065 selectors
.append(current_selector
)
2066 current_selector
= None
2068 if not current_selector
:
2069 raise syntax_error('"/" must follow a format selector', start
)
2070 first_choice
= current_selector
2071 second_choice
= _parse_format_selection(tokens
, inside_choice
=True)
2072 current_selector
= FormatSelector(PICKFIRST
, (first_choice
, second_choice
), [])
2074 if not current_selector
:
2075 current_selector
= FormatSelector(SINGLE
, 'best', [])
2076 format_filter
= _parse_filter(tokens
)
2077 current_selector
.filters
.append(format_filter
)
2079 if current_selector
:
2080 raise syntax_error('Unexpected "("', start
)
2081 group
= _parse_format_selection(tokens
, inside_group
=True)
2082 current_selector
= FormatSelector(GROUP
, group
, [])
2084 if not current_selector
:
2085 raise syntax_error('Unexpected "+"', start
)
2086 selector_1
= current_selector
2087 selector_2
= _parse_format_selection(tokens
, inside_merge
=True)
2089 raise syntax_error('Expected a selector', start
)
2090 current_selector
= FormatSelector(MERGE
, (selector_1
, selector_2
), [])
2092 raise syntax_error(f
'Operator not recognized: "{string}"', start
)
2093 elif type == tokenize
.ENDMARKER
:
2095 if current_selector
:
2096 selectors
.append(current_selector
)
2099 def _merge(formats_pair
):
2100 format_1
, format_2
= formats_pair
2103 formats_info
.extend(format_1
.get('requested_formats', (format_1
,)))
2104 formats_info
.extend(format_2
.get('requested_formats', (format_2
,)))
2106 if not allow_multiple_streams
['video'] or not allow_multiple_streams
['audio']:
2107 get_no_more
= {'video': False, 'audio': False}
2108 for (i
, fmt_info
) in enumerate(formats_info
):
2109 if fmt_info
.get('acodec') == fmt_info
.get('vcodec') == 'none':
2112 for aud_vid
in ['audio', 'video']:
2113 if not allow_multiple_streams
[aud_vid
] and fmt_info
.get(aud_vid
[0] + 'codec') != 'none':
2114 if get_no_more
[aud_vid
]:
2117 get_no_more
[aud_vid
] = True
2119 if len(formats_info
) == 1:
2120 return formats_info
[0]
2122 video_fmts
= [fmt_info
for fmt_info
in formats_info
if fmt_info
.get('vcodec') != 'none']
2123 audio_fmts
= [fmt_info
for fmt_info
in formats_info
if fmt_info
.get('acodec') != 'none']
2125 the_only_video
= video_fmts
[0] if len(video_fmts
) == 1 else None
2126 the_only_audio
= audio_fmts
[0] if len(audio_fmts
) == 1 else None
2128 output_ext
= get_compatible_ext(
2129 vcodecs
=[f
.get('vcodec') for f
in video_fmts
],
2130 acodecs
=[f
.get('acodec') for f
in audio_fmts
],
2131 vexts
=[f
['ext'] for f
in video_fmts
],
2132 aexts
=[f
['ext'] for f
in audio_fmts
],
2133 preferences
=(try_call(lambda: self
.params
['merge_output_format'].split('/'))
2134 or self
.params
.get('prefer_free_formats') and ('webm', 'mkv')))
2136 filtered
= lambda *keys
: filter(None, (traverse_obj(fmt
, *keys
) for fmt
in formats_info
))
2139 'requested_formats': formats_info
,
2140 'format': '+'.join(filtered('format')),
2141 'format_id': '+'.join(filtered('format_id')),
2143 'protocol': '+'.join(map(determine_protocol
, formats_info
)),
2144 'language': '+'.join(orderedSet(filtered('language'))) or None,
2145 'format_note': '+'.join(orderedSet(filtered('format_note'))) or None,
2146 'filesize_approx': sum(filtered('filesize', 'filesize_approx')) or None,
2147 'tbr': sum(filtered('tbr', 'vbr', 'abr')),
2152 'width': the_only_video
.get('width'),
2153 'height': the_only_video
.get('height'),
2154 'resolution': the_only_video
.get('resolution') or self
.format_resolution(the_only_video
),
2155 'fps': the_only_video
.get('fps'),
2156 'dynamic_range': the_only_video
.get('dynamic_range'),
2157 'vcodec': the_only_video
.get('vcodec'),
2158 'vbr': the_only_video
.get('vbr'),
2159 'stretched_ratio': the_only_video
.get('stretched_ratio'),
2164 'acodec': the_only_audio
.get('acodec'),
2165 'abr': the_only_audio
.get('abr'),
2166 'asr': the_only_audio
.get('asr'),
2167 'audio_channels': the_only_audio
.get('audio_channels')
2172 def _check_formats(formats
):
2173 if not check_formats
:
2176 yield from self
._check
_formats
(formats
)
2178 def _build_selector_function(selector
):
2179 if isinstance(selector
, list): # ,
2180 fs
= [_build_selector_function(s
) for s
in selector
]
2182 def selector_function(ctx
):
2185 return selector_function
2187 elif selector
.type == GROUP
: # ()
2188 selector_function
= _build_selector_function(selector
.selector
)
2190 elif selector
.type == PICKFIRST
: # /
2191 fs
= [_build_selector_function(s
) for s
in selector
.selector
]
2193 def selector_function(ctx
):
2195 picked_formats
= list(f(ctx
))
2197 return picked_formats
2200 elif selector
.type == MERGE
: # +
2201 selector_1
, selector_2
= map(_build_selector_function
, selector
.selector
)
2203 def selector_function(ctx
):
2204 for pair
in itertools
.product(selector_1(ctx
), selector_2(ctx
)):
2207 elif selector
.type == SINGLE
: # atom
2208 format_spec
= selector
.selector
or 'best'
2210 # TODO: Add allvideo, allaudio etc by generalizing the code with best/worst selector
2211 if format_spec
== 'all':
2212 def selector_function(ctx
):
2213 yield from _check_formats(ctx
['formats'][::-1])
2214 elif format_spec
== 'mergeall':
2215 def selector_function(ctx
):
2216 formats
= list(_check_formats(
2217 f
for f
in ctx
['formats'] if f
.get('vcodec') != 'none' or f
.get('acodec') != 'none'))
2220 merged_format
= formats
[-1]
2221 for f
in formats
[-2::-1]:
2222 merged_format
= _merge((merged_format
, f
))
2226 format_fallback
, seperate_fallback
, format_reverse
, format_idx
= False, None, True, 1
2228 r
'(?P<bw>best|worst|b|w)(?P<type>video|audio|v|a)?(?P<mod>\*)?(?:\.(?P<n>[1-9]\d*))?$',
2230 if mobj
is not None:
2231 format_idx
= int_or_none(mobj
.group('n'), default
=1)
2232 format_reverse
= mobj
.group('bw')[0] == 'b'
2233 format_type
= (mobj
.group('type') or [None])[0]
2234 not_format_type
= {'v': 'a', 'a': 'v'}
.get(format_type
)
2235 format_modified
= mobj
.group('mod') is not None
2237 format_fallback
= not format_type
and not format_modified
# for b, w
2239 (lambda f
: f
.get('%scodec' % format_type
) != 'none')
2240 if format_type
and format_modified
# bv*, ba*, wv*, wa*
2241 else (lambda f
: f
.get('%scodec' % not_format_type
) == 'none')
2242 if format_type
# bv, ba, wv, wa
2243 else (lambda f
: f
.get('vcodec') != 'none' and f
.get('acodec') != 'none')
2244 if not format_modified
# b, w
2245 else lambda f
: True) # b*, w*
2246 filter_f
= lambda f
: _filter_f(f
) and (
2247 f
.get('vcodec') != 'none' or f
.get('acodec') != 'none')
2249 if format_spec
in self
._format
_selection
_exts
['audio']:
2250 filter_f
= lambda f
: f
.get('ext') == format_spec
and f
.get('acodec') != 'none'
2251 elif format_spec
in self
._format
_selection
_exts
['video']:
2252 filter_f
= lambda f
: f
.get('ext') == format_spec
and f
.get('acodec') != 'none' and f
.get('vcodec') != 'none'
2253 seperate_fallback
= lambda f
: f
.get('ext') == format_spec
and f
.get('vcodec') != 'none'
2254 elif format_spec
in self
._format
_selection
_exts
['storyboards']:
2255 filter_f
= lambda f
: f
.get('ext') == format_spec
and f
.get('acodec') == 'none' and f
.get('vcodec') == 'none'
2257 filter_f
= lambda f
: f
.get('format_id') == format_spec
# id
2259 def selector_function(ctx
):
2260 formats
= list(ctx
['formats'])
2261 matches
= list(filter(filter_f
, formats
)) if filter_f
is not None else formats
2263 if format_fallback
and ctx
['incomplete_formats']:
2264 # for extractors with incomplete formats (audio only (soundcloud)
2265 # or video only (imgur)) best/worst will fallback to
2266 # best/worst {video,audio}-only format
2268 elif seperate_fallback
and not ctx
['has_merged_format']:
2269 # for compatibility with youtube-dl when there is no pre-merged format
2270 matches
= list(filter(seperate_fallback
, formats
))
2271 matches
= LazyList(_check_formats(matches
[::-1 if format_reverse
else 1]))
2273 yield matches
[format_idx
- 1]
2274 except LazyList
.IndexError:
2277 filters
= [self
._build
_format
_filter
(f
) for f
in selector
.filters
]
2279 def final_selector(ctx
):
2280 ctx_copy
= dict(ctx
)
2281 for _filter
in filters
:
2282 ctx_copy
['formats'] = list(filter(_filter
, ctx_copy
['formats']))
2283 return selector_function(ctx_copy
)
2284 return final_selector
2286 stream
= io
.BytesIO(format_spec
.encode())
2288 tokens
= list(_remove_unused_ops(tokenize
.tokenize(stream
.readline
)))
2289 except tokenize
.TokenError
:
2290 raise syntax_error('Missing closing/opening brackets or parenthesis', (0, len(format_spec
)))
2292 class TokenIterator
:
2293 def __init__(self
, tokens
):
2294 self
.tokens
= tokens
2301 if self
.counter
>= len(self
.tokens
):
2302 raise StopIteration()
2303 value
= self
.tokens
[self
.counter
]
2309 def restore_last_token(self
):
2312 parsed_selector
= _parse_format_selection(iter(TokenIterator(tokens
)))
2313 return _build_selector_function(parsed_selector
)
2315 def _calc_headers(self
, info_dict
):
2316 res
= merge_headers(self
.params
['http_headers'], info_dict
.get('http_headers') or {})
2318 cookies
= self
._calc
_cookies
(info_dict
['url'])
2320 res
['Cookie'] = cookies
2322 if 'X-Forwarded-For' not in res
:
2323 x_forwarded_for_ip
= info_dict
.get('__x_forwarded_for_ip')
2324 if x_forwarded_for_ip
:
2325 res
['X-Forwarded-For'] = x_forwarded_for_ip
2329 def _calc_cookies(self
, url
):
2330 pr
= sanitized_Request(url
)
2331 self
.cookiejar
.add_cookie_header(pr
)
2332 return pr
.get_header('Cookie')
2334 def _sort_thumbnails(self
, thumbnails
):
2335 thumbnails
.sort(key
=lambda t
: (
2336 t
.get('preference') if t
.get('preference') is not None else -1,
2337 t
.get('width') if t
.get('width') is not None else -1,
2338 t
.get('height') if t
.get('height') is not None else -1,
2339 t
.get('id') if t
.get('id') is not None else '',
2342 def _sanitize_thumbnails(self
, info_dict
):
2343 thumbnails
= info_dict
.get('thumbnails')
2344 if thumbnails
is None:
2345 thumbnail
= info_dict
.get('thumbnail')
2347 info_dict
['thumbnails'] = thumbnails
= [{'url': thumbnail}
]
2351 def check_thumbnails(thumbnails
):
2352 for t
in thumbnails
:
2353 self
.to_screen(f
'[info] Testing thumbnail {t["id"]}')
2355 self
.urlopen(HEADRequest(t
['url']))
2356 except network_exceptions
as err
:
2357 self
.to_screen(f
'[info] Unable to connect to thumbnail {t["id"]} URL {t["url"]!r} - {err}. Skipping...')
2361 self
._sort
_thumbnails
(thumbnails
)
2362 for i
, t
in enumerate(thumbnails
):
2363 if t
.get('id') is None:
2365 if t
.get('width') and t
.get('height'):
2366 t
['resolution'] = '%dx%d' % (t
['width'], t
['height'])
2367 t
['url'] = sanitize_url(t
['url'])
2369 if self
.params
.get('check_formats') is True:
2370 info_dict
['thumbnails'] = LazyList(check_thumbnails(thumbnails
[::-1]), reverse
=True)
2372 info_dict
['thumbnails'] = thumbnails
2374 def _fill_common_fields(self
, info_dict
, is_video
=True):
2375 # TODO: move sanitization here
2377 # playlists are allowed to lack "title"
2378 title
= info_dict
.get('title', NO_DEFAULT
)
2379 if title
is NO_DEFAULT
:
2380 raise ExtractorError('Missing "title" field in extractor result',
2381 video_id
=info_dict
['id'], ie
=info_dict
['extractor'])
2382 info_dict
['fulltitle'] = title
2385 self
.write_debug('Extractor gave empty title. Creating a generic title')
2387 self
.report_warning('Extractor failed to obtain "title". Creating a generic title instead')
2388 info_dict
['title'] = f
'{info_dict["extractor"].replace(":", "-")} video #{info_dict["id"]}'
2390 if info_dict
.get('duration') is not None:
2391 info_dict
['duration_string'] = formatSeconds(info_dict
['duration'])
2393 for ts_key
, date_key
in (
2394 ('timestamp', 'upload_date'),
2395 ('release_timestamp', 'release_date'),
2396 ('modified_timestamp', 'modified_date'),
2398 if info_dict
.get(date_key
) is None and info_dict
.get(ts_key
) is not None:
2399 # Working around out-of-range timestamp values (e.g. negative ones on Windows,
2400 # see http://bugs.python.org/issue1646728)
2401 with contextlib
.suppress(ValueError, OverflowError, OSError):
2402 upload_date
= datetime
.datetime
.utcfromtimestamp(info_dict
[ts_key
])
2403 info_dict
[date_key
] = upload_date
.strftime('%Y%m%d')
2405 live_keys
= ('is_live', 'was_live')
2406 live_status
= info_dict
.get('live_status')
2407 if live_status
is None:
2408 for key
in live_keys
:
2409 if info_dict
.get(key
) is False:
2411 if info_dict
.get(key
):
2414 if all(info_dict
.get(key
) is False for key
in live_keys
):
2415 live_status
= 'not_live'
2417 info_dict
['live_status'] = live_status
2418 for key
in live_keys
:
2419 if info_dict
.get(key
) is None:
2420 info_dict
[key
] = (live_status
== key
)
2422 # Auto generate title fields corresponding to the *_number fields when missing
2423 # in order to always have clean titles. This is very common for TV series.
2424 for field
in ('chapter', 'season', 'episode'):
2425 if info_dict
.get('%s_number' % field
) is not None and not info_dict
.get(field
):
2426 info_dict
[field
] = '%s %d' % (field
.capitalize(), info_dict
['%s_number' % field
])
2428 def _raise_pending_errors(self
, info
):
2429 err
= info
.pop('__pending_error', None)
2431 self
.report_error(err
, tb
=False)
2433 def process_video_result(self
, info_dict
, download
=True):
2434 assert info_dict
.get('_type', 'video') == 'video'
2435 self
._num
_videos
+= 1
2437 if 'id' not in info_dict
:
2438 raise ExtractorError('Missing "id" field in extractor result', ie
=info_dict
['extractor'])
2439 elif not info_dict
.get('id'):
2440 raise ExtractorError('Extractor failed to obtain "id"', ie
=info_dict
['extractor'])
2442 def report_force_conversion(field
, field_not
, conversion
):
2443 self
.report_warning(
2444 '"%s" field is not %s - forcing %s conversion, there is an error in extractor'
2445 % (field
, field_not
, conversion
))
2447 def sanitize_string_field(info
, string_field
):
2448 field
= info
.get(string_field
)
2449 if field
is None or isinstance(field
, str):
2451 report_force_conversion(string_field
, 'a string', 'string')
2452 info
[string_field
] = str(field
)
2454 def sanitize_numeric_fields(info
):
2455 for numeric_field
in self
._NUMERIC
_FIELDS
:
2456 field
= info
.get(numeric_field
)
2457 if field
is None or isinstance(field
, (int, float)):
2459 report_force_conversion(numeric_field
, 'numeric', 'int')
2460 info
[numeric_field
] = int_or_none(field
)
2462 sanitize_string_field(info_dict
, 'id')
2463 sanitize_numeric_fields(info_dict
)
2464 if info_dict
.get('section_end') and info_dict
.get('section_start') is not None:
2465 info_dict
['duration'] = round(info_dict
['section_end'] - info_dict
['section_start'], 3)
2466 if (info_dict
.get('duration') or 0) <= 0 and info_dict
.pop('duration', None):
2467 self
.report_warning('"duration" field is negative, there is an error in extractor')
2469 chapters
= info_dict
.get('chapters') or []
2470 if chapters
and chapters
[0].get('start_time'):
2471 chapters
.insert(0, {'start_time': 0}
)
2473 dummy_chapter
= {'end_time': 0, 'start_time': info_dict.get('duration')}
2474 for idx
, (prev
, current
, next_
) in enumerate(zip(
2475 (dummy_chapter
, *chapters
), chapters
, (*chapters
[1:], dummy_chapter
)), 1):
2476 if current
.get('start_time') is None:
2477 current
['start_time'] = prev
.get('end_time')
2478 if not current
.get('end_time'):
2479 current
['end_time'] = next_
.get('start_time')
2480 if not current
.get('title'):
2481 current
['title'] = f
'<Untitled Chapter {idx}>'
2483 if 'playlist' not in info_dict
:
2484 # It isn't part of a playlist
2485 info_dict
['playlist'] = None
2486 info_dict
['playlist_index'] = None
2488 self
._sanitize
_thumbnails
(info_dict
)
2490 thumbnail
= info_dict
.get('thumbnail')
2491 thumbnails
= info_dict
.get('thumbnails')
2493 info_dict
['thumbnail'] = sanitize_url(thumbnail
)
2495 info_dict
['thumbnail'] = thumbnails
[-1]['url']
2497 if info_dict
.get('display_id') is None and 'id' in info_dict
:
2498 info_dict
['display_id'] = info_dict
['id']
2500 self
._fill
_common
_fields
(info_dict
)
2502 for cc_kind
in ('subtitles', 'automatic_captions'):
2503 cc
= info_dict
.get(cc_kind
)
2505 for _
, subtitle
in cc
.items():
2506 for subtitle_format
in subtitle
:
2507 if subtitle_format
.get('url'):
2508 subtitle_format
['url'] = sanitize_url(subtitle_format
['url'])
2509 if subtitle_format
.get('ext') is None:
2510 subtitle_format
['ext'] = determine_ext(subtitle_format
['url']).lower()
2512 automatic_captions
= info_dict
.get('automatic_captions')
2513 subtitles
= info_dict
.get('subtitles')
2515 info_dict
['requested_subtitles'] = self
.process_subtitles(
2516 info_dict
['id'], subtitles
, automatic_captions
)
2518 if info_dict
.get('formats') is None:
2519 # There's only one format available
2520 formats
= [info_dict
]
2522 formats
= info_dict
['formats']
2524 # or None ensures --clean-infojson removes it
2525 info_dict
['_has_drm'] = any(f
.get('has_drm') for f
in formats
) or None
2526 if not self
.params
.get('allow_unplayable_formats'):
2527 formats
= [f
for f
in formats
if not f
.get('has_drm')]
2528 if info_dict
['_has_drm'] and formats
and all(
2529 f
.get('acodec') == f
.get('vcodec') == 'none' for f
in formats
):
2530 self
.report_warning(
2531 'This video is DRM protected and only images are available for download. '
2532 'Use --list-formats to see them')
2534 get_from_start
= not info_dict
.get('is_live') or bool(self
.params
.get('live_from_start'))
2535 if not get_from_start
:
2536 info_dict
['title'] += ' ' + datetime
.datetime
.now().strftime('%Y-%m-%d %H:%M')
2537 if info_dict
.get('is_live') and formats
:
2538 formats
= [f
for f
in formats
if bool(f
.get('is_from_start')) == get_from_start
]
2539 if get_from_start
and not formats
:
2540 self
.raise_no_formats(info_dict
, msg
=(
2541 '--live-from-start is passed, but there are no formats that can be downloaded from the start. '
2542 'If you want to download from the current time, use --no-live-from-start'))
2544 def is_wellformed(f
):
2547 self
.report_warning(
2548 '"url" field is missing or empty - skipping format, '
2549 'there is an error in extractor')
2551 if isinstance(url
, bytes):
2552 sanitize_string_field(f
, 'url')
2555 # Filter out malformed formats for better extraction robustness
2556 formats
= list(filter(is_wellformed
, formats
or []))
2559 self
.raise_no_formats(info_dict
)
2563 # We check that all the formats have the format and format_id fields
2564 for i
, format
in enumerate(formats
):
2565 sanitize_string_field(format
, 'format_id')
2566 sanitize_numeric_fields(format
)
2567 format
['url'] = sanitize_url(format
['url'])
2568 if not format
.get('format_id'):
2569 format
['format_id'] = str(i
)
2571 # Sanitize format_id from characters used in format selector expression
2572 format
['format_id'] = re
.sub(r
'[\s,/+\[\]()]', '_', format
['format_id'])
2573 format_id
= format
['format_id']
2574 if format_id
not in formats_dict
:
2575 formats_dict
[format_id
] = []
2576 formats_dict
[format_id
].append(format
)
2578 # Make sure all formats have unique format_id
2579 common_exts
= set(itertools
.chain(*self
._format
_selection
_exts
.values()))
2580 for format_id
, ambiguous_formats
in formats_dict
.items():
2581 ambigious_id
= len(ambiguous_formats
) > 1
2582 for i
, format
in enumerate(ambiguous_formats
):
2584 format
['format_id'] = '%s-%d' % (format_id
, i
)
2585 if format
.get('ext') is None:
2586 format
['ext'] = determine_ext(format
['url']).lower()
2587 # Ensure there is no conflict between id and ext in format selection
2588 # See https://github.com/yt-dlp/yt-dlp/issues/1282
2589 if format
['format_id'] != format
['ext'] and format
['format_id'] in common_exts
:
2590 format
['format_id'] = 'f%s' % format
['format_id']
2592 for i
, format
in enumerate(formats
):
2593 if format
.get('format') is None:
2594 format
['format'] = '{id} - {res}{note}'.format(
2595 id=format
['format_id'],
2596 res
=self
.format_resolution(format
),
2597 note
=format_field(format
, 'format_note', ' (%s)'),
2599 if format
.get('protocol') is None:
2600 format
['protocol'] = determine_protocol(format
)
2601 if format
.get('resolution') is None:
2602 format
['resolution'] = self
.format_resolution(format
, default
=None)
2603 if format
.get('dynamic_range') is None and format
.get('vcodec') != 'none':
2604 format
['dynamic_range'] = 'SDR'
2605 if (info_dict
.get('duration') and format
.get('tbr')
2606 and not format
.get('filesize') and not format
.get('filesize_approx')):
2607 format
['filesize_approx'] = int(info_dict
['duration'] * format
['tbr'] * (1024 / 8))
2609 # Add HTTP headers, so that external programs can use them from the
2611 full_format_info
= info_dict
.copy()
2612 full_format_info
.update(format
)
2613 format
['http_headers'] = self
._calc
_headers
(full_format_info
)
2614 # Remove private housekeeping stuff
2615 if '__x_forwarded_for_ip' in info_dict
:
2616 del info_dict
['__x_forwarded_for_ip']
2618 if self
.params
.get('check_formats') is True:
2619 formats
= LazyList(self
._check
_formats
(formats
[::-1]), reverse
=True)
2621 if not formats
or formats
[0] is not info_dict
:
2622 # only set the 'formats' fields if the original info_dict list them
2623 # otherwise we end up with a circular reference, the first (and unique)
2624 # element in the 'formats' field in info_dict is info_dict itself,
2625 # which can't be exported to json
2626 info_dict
['formats'] = formats
2628 info_dict
, _
= self
.pre_process(info_dict
)
2630 if self
._match
_entry
(info_dict
, incomplete
=self
._format
_fields
) is not None:
2633 self
.post_extract(info_dict
)
2634 info_dict
, _
= self
.pre_process(info_dict
, 'after_filter')
2636 # The pre-processors may have modified the formats
2637 formats
= info_dict
.get('formats', [info_dict
])
2639 list_only
= self
.params
.get('simulate') is None and (
2640 self
.params
.get('list_thumbnails') or self
.params
.get('listformats') or self
.params
.get('listsubtitles'))
2641 interactive_format_selection
= not list_only
and self
.format_selector
== '-'
2642 if self
.params
.get('list_thumbnails'):
2643 self
.list_thumbnails(info_dict
)
2644 if self
.params
.get('listsubtitles'):
2645 if 'automatic_captions' in info_dict
:
2646 self
.list_subtitles(
2647 info_dict
['id'], automatic_captions
, 'automatic captions')
2648 self
.list_subtitles(info_dict
['id'], subtitles
, 'subtitles')
2649 if self
.params
.get('listformats') or interactive_format_selection
:
2650 self
.list_formats(info_dict
)
2652 # Without this printing, -F --print-json will not work
2653 self
.__forced
_printings
(info_dict
, self
.prepare_filename(info_dict
), incomplete
=True)
2656 format_selector
= self
.format_selector
2657 if format_selector
is None:
2658 req_format
= self
._default
_format
_spec
(info_dict
, download
=download
)
2659 self
.write_debug('Default format spec: %s' % req_format
)
2660 format_selector
= self
.build_format_selector(req_format
)
2663 if interactive_format_selection
:
2665 self
._format
_screen
('\nEnter format selector: ', self
.Styles
.EMPHASIS
))
2667 format_selector
= self
.build_format_selector(req_format
)
2668 except SyntaxError as err
:
2669 self
.report_error(err
, tb
=False, is_error
=False)
2672 formats_to_download
= list(format_selector({
2674 'has_merged_format': any('none' not in (f
.get('acodec'), f
.get('vcodec')) for f
in formats
),
2675 'incomplete_formats': (
2676 # All formats are video-only or
2677 all(f
.get('vcodec') != 'none' and f
.get('acodec') == 'none' for f
in formats
)
2678 # all formats are audio-only
2679 or all(f
.get('vcodec') == 'none' and f
.get('acodec') != 'none' for f
in formats
)),
2681 if interactive_format_selection
and not formats_to_download
:
2682 self
.report_error('Requested format is not available', tb
=False, is_error
=False)
2686 if not formats_to_download
:
2687 if not self
.params
.get('ignore_no_formats_error'):
2688 raise ExtractorError(
2689 'Requested format is not available. Use --list-formats for a list of available formats',
2690 expected
=True, video_id
=info_dict
['id'], ie
=info_dict
['extractor'])
2691 self
.report_warning('Requested format is not available')
2692 # Process what we can, even without any available formats.
2693 formats_to_download
= [{}]
2695 requested_ranges
= self
.params
.get('download_ranges')
2696 if requested_ranges
:
2697 requested_ranges
= tuple(requested_ranges(info_dict
, self
))
2699 best_format
, downloaded_formats
= formats_to_download
[-1], []
2702 def to_screen(*msg
):
2703 self
.to_screen(f
'[info] {info_dict["id"]}: {" ".join(", ".join(variadic(m)) for m in msg)}')
2705 to_screen(f
'Downloading {len(formats_to_download)} format(s):',
2706 (f
['format_id'] for f
in formats_to_download
))
2707 if requested_ranges
:
2708 to_screen(f
'Downloading {len(requested_ranges)} time ranges:',
2709 (f
'{int(c["start_time"])}-{int(c["end_time"])}' for c
in requested_ranges
))
2710 max_downloads_reached
= False
2712 for fmt
, chapter
in itertools
.product(formats_to_download
, requested_ranges
or [{}]):
2713 new_info
= self
._copy
_infodict
(info_dict
)
2714 new_info
.update(fmt
)
2715 offset
, duration
= info_dict
.get('section_start') or 0, info_dict
.get('duration') or float('inf')
2716 if chapter
or offset
:
2718 'section_start': offset
+ chapter
.get('start_time', 0),
2719 'section_end': offset
+ min(chapter
.get('end_time', duration
), duration
),
2720 'section_title': chapter
.get('title'),
2721 'section_number': chapter
.get('index'),
2723 downloaded_formats
.append(new_info
)
2725 self
.process_info(new_info
)
2726 except MaxDownloadsReached
:
2727 max_downloads_reached
= True
2728 self
._raise
_pending
_errors
(new_info
)
2729 # Remove copied info
2730 for key
, val
in tuple(new_info
.items()):
2731 if info_dict
.get(key
) == val
:
2733 if max_downloads_reached
:
2736 write_archive
= {f.get('__write_download_archive', False) for f in downloaded_formats}
2737 assert write_archive
.issubset({True, False, 'ignore'}
)
2738 if True in write_archive
and False not in write_archive
:
2739 self
.record_download_archive(info_dict
)
2741 info_dict
['requested_downloads'] = downloaded_formats
2742 info_dict
= self
.run_all_pps('after_video', info_dict
)
2743 if max_downloads_reached
:
2744 raise MaxDownloadsReached()
2746 # We update the info dict with the selected best quality format (backwards compatibility)
2747 info_dict
.update(best_format
)
2750 def process_subtitles(self
, video_id
, normal_subtitles
, automatic_captions
):
2751 """Select the requested subtitles and their format"""
2752 available_subs
, normal_sub_langs
= {}, []
2753 if normal_subtitles
and self
.params
.get('writesubtitles'):
2754 available_subs
.update(normal_subtitles
)
2755 normal_sub_langs
= tuple(normal_subtitles
.keys())
2756 if automatic_captions
and self
.params
.get('writeautomaticsub'):
2757 for lang
, cap_info
in automatic_captions
.items():
2758 if lang
not in available_subs
:
2759 available_subs
[lang
] = cap_info
2761 if not available_subs
or (
2762 not self
.params
.get('writesubtitles')
2763 and not self
.params
.get('writeautomaticsub')):
2766 all_sub_langs
= tuple(available_subs
.keys())
2767 if self
.params
.get('allsubtitles', False):
2768 requested_langs
= all_sub_langs
2769 elif self
.params
.get('subtitleslangs', False):
2771 requested_langs
= orderedSet_from_options(
2772 self
.params
.get('subtitleslangs'), {'all': all_sub_langs}
, use_regex
=True)
2773 except re
.error
as e
:
2774 raise ValueError(f
'Wrong regex for subtitlelangs: {e.pattern}')
2775 elif normal_sub_langs
:
2776 requested_langs
= ['en'] if 'en' in normal_sub_langs
else normal_sub_langs
[:1]
2778 requested_langs
= ['en'] if 'en' in all_sub_langs
else all_sub_langs
[:1]
2780 self
.to_screen(f
'[info] {video_id}: Downloading subtitles: {", ".join(requested_langs)}')
2782 formats_query
= self
.params
.get('subtitlesformat', 'best')
2783 formats_preference
= formats_query
.split('/') if formats_query
else []
2785 for lang
in requested_langs
:
2786 formats
= available_subs
.get(lang
)
2788 self
.report_warning(f
'{lang} subtitles not available for {video_id}')
2790 for ext
in formats_preference
:
2794 matches
= list(filter(lambda f
: f
['ext'] == ext
, formats
))
2800 self
.report_warning(
2801 'No subtitle format found matching "%s" for language %s, '
2802 'using %s' % (formats_query
, lang
, f
['ext']))
2806 def _forceprint(self
, key
, info_dict
):
2807 if info_dict
is None:
2809 info_copy
= info_dict
.copy()
2810 info_copy
['formats_table'] = self
.render_formats_table(info_dict
)
2811 info_copy
['thumbnails_table'] = self
.render_thumbnails_table(info_dict
)
2812 info_copy
['subtitles_table'] = self
.render_subtitles_table(info_dict
.get('id'), info_dict
.get('subtitles'))
2813 info_copy
['automatic_captions_table'] = self
.render_subtitles_table(info_dict
.get('id'), info_dict
.get('automatic_captions'))
2815 def format_tmpl(tmpl
):
2816 mobj
= re
.fullmatch(r
'([\w.:,]|-\d|(?P<dict>{([\w.:,]|-\d)+}))+=?', tmpl
)
2821 if tmpl
.startswith('{'):
2823 if tmpl
.endswith('='):
2824 tmpl
, fmt
= tmpl
[:-1], '{0} = %({0})#j'
2825 return '\n'.join(map(fmt
.format
, [tmpl
] if mobj
.group('dict') else tmpl
.split(',')))
2827 for tmpl
in self
.params
['forceprint'].get(key
, []):
2828 self
.to_stdout(self
.evaluate_outtmpl(format_tmpl(tmpl
), info_copy
))
2830 for tmpl
, file_tmpl
in self
.params
['print_to_file'].get(key
, []):
2831 filename
= self
.prepare_filename(info_dict
, outtmpl
=file_tmpl
)
2832 tmpl
= format_tmpl(tmpl
)
2833 self
.to_screen(f
'[info] Writing {tmpl!r} to: {filename}')
2834 if self
._ensure
_dir
_exists
(filename
):
2835 with open(filename
, 'a', encoding
='utf-8') as f
:
2836 f
.write(self
.evaluate_outtmpl(tmpl
, info_copy
) + '\n')
2838 def __forced_printings(self
, info_dict
, filename
, incomplete
):
2839 def print_mandatory(field
, actual_field
=None):
2840 if actual_field
is None:
2841 actual_field
= field
2842 if (self
.params
.get('force%s' % field
, False)
2843 and (not incomplete
or info_dict
.get(actual_field
) is not None)):
2844 self
.to_stdout(info_dict
[actual_field
])
2846 def print_optional(field
):
2847 if (self
.params
.get('force%s' % field
, False)
2848 and info_dict
.get(field
) is not None):
2849 self
.to_stdout(info_dict
[field
])
2851 info_dict
= info_dict
.copy()
2852 if filename
is not None:
2853 info_dict
['filename'] = filename
2854 if info_dict
.get('requested_formats') is not None:
2855 # For RTMP URLs, also include the playpath
2856 info_dict
['urls'] = '\n'.join(f
['url'] + f
.get('play_path', '') for f
in info_dict
['requested_formats'])
2857 elif info_dict
.get('url'):
2858 info_dict
['urls'] = info_dict
['url'] + info_dict
.get('play_path', '')
2860 if (self
.params
.get('forcejson')
2861 or self
.params
['forceprint'].get('video')
2862 or self
.params
['print_to_file'].get('video')):
2863 self
.post_extract(info_dict
)
2864 self
._forceprint
('video', info_dict
)
2866 print_mandatory('title')
2867 print_mandatory('id')
2868 print_mandatory('url', 'urls')
2869 print_optional('thumbnail')
2870 print_optional('description')
2871 print_optional('filename')
2872 if self
.params
.get('forceduration') and info_dict
.get('duration') is not None:
2873 self
.to_stdout(formatSeconds(info_dict
['duration']))
2874 print_mandatory('format')
2876 if self
.params
.get('forcejson'):
2877 self
.to_stdout(json
.dumps(self
.sanitize_info(info_dict
)))
2879 def dl(self
, name
, info
, subtitle
=False, test
=False):
2880 if not info
.get('url'):
2881 self
.raise_no_formats(info
, True)
2884 verbose
= self
.params
.get('verbose')
2887 'quiet': self
.params
.get('quiet') or not verbose
,
2889 'noprogress': not verbose
,
2891 'skip_unavailable_fragments': False,
2892 'keep_fragments': False,
2894 '_no_ytdl_file': True,
2897 params
= self
.params
2898 fd
= get_suitable_downloader(info
, params
, to_stdout
=(name
== '-'))(self
, params
)
2900 for ph
in self
._progress
_hooks
:
2901 fd
.add_progress_hook(ph
)
2903 (f
['url'].split(',')[0] + ',<data>' if f
['url'].startswith('data:') else f
['url'])
2904 for f
in info
.get('requested_formats', []) or [info
])
2905 self
.write_debug(f
'Invoking {fd.FD_NAME} downloader on "{urls}"')
2907 # Note: Ideally info should be a deep-copied so that hooks cannot modify it.
2908 # But it may contain objects that are not deep-copyable
2909 new_info
= self
._copy
_infodict
(info
)
2910 if new_info
.get('http_headers') is None:
2911 new_info
['http_headers'] = self
._calc
_headers
(new_info
)
2912 return fd
.download(name
, new_info
, subtitle
)
2914 def existing_file(self
, filepaths
, *, default_overwrite
=True):
2915 existing_files
= list(filter(os
.path
.exists
, orderedSet(filepaths
)))
2916 if existing_files
and not self
.params
.get('overwrites', default_overwrite
):
2917 return existing_files
[0]
2919 for file in existing_files
:
2920 self
.report_file_delete(file)
2924 def process_info(self
, info_dict
):
2925 """Process a single resolved IE result. (Modifies it in-place)"""
2927 assert info_dict
.get('_type', 'video') == 'video'
2928 original_infodict
= info_dict
2930 if 'format' not in info_dict
and 'ext' in info_dict
:
2931 info_dict
['format'] = info_dict
['ext']
2933 # This is mostly just for backward compatibility of process_info
2934 # As a side-effect, this allows for format-specific filters
2935 if self
._match
_entry
(info_dict
) is not None:
2936 info_dict
['__write_download_archive'] = 'ignore'
2939 # Does nothing under normal operation - for backward compatibility of process_info
2940 self
.post_extract(info_dict
)
2941 self
._num
_downloads
+= 1
2943 # info_dict['_filename'] needs to be set for backward compatibility
2944 info_dict
['_filename'] = full_filename
= self
.prepare_filename(info_dict
, warn
=True)
2945 temp_filename
= self
.prepare_filename(info_dict
, 'temp')
2949 self
.__forced
_printings
(info_dict
, full_filename
, incomplete
=('format' not in info_dict
))
2951 def check_max_downloads():
2952 if self
._num
_downloads
>= float(self
.params
.get('max_downloads') or 'inf'):
2953 raise MaxDownloadsReached()
2955 if self
.params
.get('simulate'):
2956 info_dict
['__write_download_archive'] = self
.params
.get('force_write_download_archive')
2957 check_max_downloads()
2960 if full_filename
is None:
2962 if not self
._ensure
_dir
_exists
(encodeFilename(full_filename
)):
2964 if not self
._ensure
_dir
_exists
(encodeFilename(temp_filename
)):
2967 if self
._write
_description
('video', info_dict
,
2968 self
.prepare_filename(info_dict
, 'description')) is None:
2971 sub_files
= self
._write
_subtitles
(info_dict
, temp_filename
)
2972 if sub_files
is None:
2974 files_to_move
.update(dict(sub_files
))
2976 thumb_files
= self
._write
_thumbnails
(
2977 'video', info_dict
, temp_filename
, self
.prepare_filename(info_dict
, 'thumbnail'))
2978 if thumb_files
is None:
2980 files_to_move
.update(dict(thumb_files
))
2982 infofn
= self
.prepare_filename(info_dict
, 'infojson')
2983 _infojson_written
= self
._write
_info
_json
('video', info_dict
, infofn
)
2984 if _infojson_written
:
2985 info_dict
['infojson_filename'] = infofn
2986 # For backward compatibility, even though it was a private field
2987 info_dict
['__infojson_filename'] = infofn
2988 elif _infojson_written
is None:
2991 # Note: Annotations are deprecated
2993 if self
.params
.get('writeannotations', False):
2994 annofn
= self
.prepare_filename(info_dict
, 'annotation')
2996 if not self
._ensure
_dir
_exists
(encodeFilename(annofn
)):
2998 if not self
.params
.get('overwrites', True) and os
.path
.exists(encodeFilename(annofn
)):
2999 self
.to_screen('[info] Video annotations are already present')
3000 elif not info_dict
.get('annotations'):
3001 self
.report_warning('There are no annotations to write.')
3004 self
.to_screen('[info] Writing video annotations to: ' + annofn
)
3005 with open(encodeFilename(annofn
), 'w', encoding
='utf-8') as annofile
:
3006 annofile
.write(info_dict
['annotations'])
3007 except (KeyError, TypeError):
3008 self
.report_warning('There are no annotations to write.')
3010 self
.report_error('Cannot write annotations file: ' + annofn
)
3013 # Write internet shortcut files
3014 def _write_link_file(link_type
):
3015 url
= try_get(info_dict
['webpage_url'], iri_to_uri
)
3017 self
.report_warning(
3018 f
'Cannot write internet shortcut file because the actual URL of "{info_dict["webpage_url"]}" is unknown')
3020 linkfn
= replace_extension(self
.prepare_filename(info_dict
, 'link'), link_type
, info_dict
.get('ext'))
3021 if not self
._ensure
_dir
_exists
(encodeFilename(linkfn
)):
3023 if self
.params
.get('overwrites', True) and os
.path
.exists(encodeFilename(linkfn
)):
3024 self
.to_screen(f
'[info] Internet shortcut (.{link_type}) is already present')
3027 self
.to_screen(f
'[info] Writing internet shortcut (.{link_type}) to: {linkfn}')
3028 with open(encodeFilename(to_high_limit_path(linkfn
)), 'w', encoding
='utf-8',
3029 newline
='\r\n' if link_type
== 'url' else '\n') as linkfile
:
3030 template_vars
= {'url': url}
3031 if link_type
== 'desktop':
3032 template_vars
['filename'] = linkfn
[:-(len(link_type
) + 1)]
3033 linkfile
.write(LINK_TEMPLATES
[link_type
] % template_vars
)
3035 self
.report_error(f
'Cannot write internet shortcut {linkfn}')
3040 'url': self
.params
.get('writeurllink'),
3041 'webloc': self
.params
.get('writewebloclink'),
3042 'desktop': self
.params
.get('writedesktoplink'),
3044 if self
.params
.get('writelink'):
3045 link_type
= ('webloc' if sys
.platform
== 'darwin'
3046 else 'desktop' if sys
.platform
.startswith('linux')
3048 write_links
[link_type
] = True
3050 if any(should_write
and not _write_link_file(link_type
)
3051 for link_type
, should_write
in write_links
.items()):
3054 def replace_info_dict(new_info
):
3056 if new_info
== info_dict
:
3059 info_dict
.update(new_info
)
3061 new_info
, files_to_move
= self
.pre_process(info_dict
, 'before_dl', files_to_move
)
3062 replace_info_dict(new_info
)
3064 if self
.params
.get('skip_download'):
3065 info_dict
['filepath'] = temp_filename
3066 info_dict
['__finaldir'] = os
.path
.dirname(os
.path
.abspath(encodeFilename(full_filename
)))
3067 info_dict
['__files_to_move'] = files_to_move
3068 replace_info_dict(self
.run_pp(MoveFilesAfterDownloadPP(self
, False), info_dict
))
3069 info_dict
['__write_download_archive'] = self
.params
.get('force_write_download_archive')
3072 info_dict
.setdefault('__postprocessors', [])
3075 def existing_video_file(*filepaths
):
3076 ext
= info_dict
.get('ext')
3077 converted
= lambda file: replace_extension(file, self
.params
.get('final_ext') or ext
, ext
)
3078 file = self
.existing_file(itertools
.chain(*zip(map(converted
, filepaths
), filepaths
)),
3079 default_overwrite
=False)
3081 info_dict
['ext'] = os
.path
.splitext(file)[1][1:]
3084 fd
, success
= None, True
3085 if info_dict
.get('protocol') or info_dict
.get('url'):
3086 fd
= get_suitable_downloader(info_dict
, self
.params
, to_stdout
=temp_filename
== '-')
3087 if fd
is not FFmpegFD
and (
3088 info_dict
.get('section_start') or info_dict
.get('section_end')):
3089 msg
= ('This format cannot be partially downloaded' if FFmpegFD
.available()
3090 else 'You have requested downloading the video partially, but ffmpeg is not installed')
3091 self
.report_error(f
'{msg}. Aborting')
3094 if info_dict
.get('requested_formats') is not None:
3095 requested_formats
= info_dict
['requested_formats']
3096 old_ext
= info_dict
['ext']
3097 if self
.params
.get('merge_output_format') is None:
3098 if (info_dict
['ext'] == 'webm'
3099 and info_dict
.get('thumbnails')
3100 # check with type instead of pp_key, __name__, or isinstance
3101 # since we dont want any custom PPs to trigger this
3102 and any(type(pp
) == EmbedThumbnailPP
for pp
in self
._pps
['post_process'])): # noqa: E721
3103 info_dict
['ext'] = 'mkv'
3104 self
.report_warning(
3105 'webm doesn\'t support embedding a thumbnail, mkv will be used')
3106 new_ext
= info_dict
['ext']
3108 def correct_ext(filename
, ext
=new_ext
):
3111 filename_real_ext
= os
.path
.splitext(filename
)[1][1:]
3113 os
.path
.splitext(filename
)[0]
3114 if filename_real_ext
in (old_ext
, new_ext
)
3116 return f
'{filename_wo_ext}.{ext}'
3118 # Ensure filename always has a correct extension for successful merge
3119 full_filename
= correct_ext(full_filename
)
3120 temp_filename
= correct_ext(temp_filename
)
3121 dl_filename
= existing_video_file(full_filename
, temp_filename
)
3122 info_dict
['__real_download'] = False
3124 merger
= FFmpegMergerPP(self
)
3126 if dl_filename
is not None:
3127 self
.report_file_already_downloaded(dl_filename
)
3129 for f
in requested_formats
if fd
!= FFmpegFD
else []:
3130 f
['filepath'] = fname
= prepend_extension(
3131 correct_ext(temp_filename
, info_dict
['ext']),
3132 'f%s' % f
['format_id'], info_dict
['ext'])
3133 downloaded
.append(fname
)
3134 info_dict
['url'] = '\n'.join(f
['url'] for f
in requested_formats
)
3135 success
, real_download
= self
.dl(temp_filename
, info_dict
)
3136 info_dict
['__real_download'] = real_download
3138 if self
.params
.get('allow_unplayable_formats'):
3139 self
.report_warning(
3140 'You have requested merging of multiple formats '
3141 'while also allowing unplayable formats to be downloaded. '
3142 'The formats won\'t be merged to prevent data corruption.')
3143 elif not merger
.available
:
3144 msg
= 'You have requested merging of multiple formats but ffmpeg is not installed'
3145 if not self
.params
.get('ignoreerrors'):
3146 self
.report_error(f
'{msg}. Aborting due to --abort-on-error')
3148 self
.report_warning(f
'{msg}. The formats won\'t be merged')
3150 if temp_filename
== '-':
3151 reason
= ('using a downloader other than ffmpeg' if FFmpegFD
.can_merge_formats(info_dict
, self
.params
)
3152 else 'but the formats are incompatible for simultaneous download' if merger
.available
3153 else 'but ffmpeg is not installed')
3154 self
.report_warning(
3155 f
'You have requested downloading multiple formats to stdout {reason}. '
3156 'The formats will be streamed one after the other')
3157 fname
= temp_filename
3158 for f
in requested_formats
:
3159 new_info
= dict(info_dict
)
3160 del new_info
['requested_formats']
3162 if temp_filename
!= '-':
3163 fname
= prepend_extension(
3164 correct_ext(temp_filename
, new_info
['ext']),
3165 'f%s' % f
['format_id'], new_info
['ext'])
3166 if not self
._ensure
_dir
_exists
(fname
):
3168 f
['filepath'] = fname
3169 downloaded
.append(fname
)
3170 partial_success
, real_download
= self
.dl(fname
, new_info
)
3171 info_dict
['__real_download'] = info_dict
['__real_download'] or real_download
3172 success
= success
and partial_success
3174 if downloaded
and merger
.available
and not self
.params
.get('allow_unplayable_formats'):
3175 info_dict
['__postprocessors'].append(merger
)
3176 info_dict
['__files_to_merge'] = downloaded
3177 # Even if there were no downloads, it is being merged only now
3178 info_dict
['__real_download'] = True
3180 for file in downloaded
:
3181 files_to_move
[file] = None
3183 # Just a single file
3184 dl_filename
= existing_video_file(full_filename
, temp_filename
)
3185 if dl_filename
is None or dl_filename
== temp_filename
:
3186 # dl_filename == temp_filename could mean that the file was partially downloaded with --no-part.
3187 # So we should try to resume the download
3188 success
, real_download
= self
.dl(temp_filename
, info_dict
)
3189 info_dict
['__real_download'] = real_download
3191 self
.report_file_already_downloaded(dl_filename
)
3193 dl_filename
= dl_filename
or temp_filename
3194 info_dict
['__finaldir'] = os
.path
.dirname(os
.path
.abspath(encodeFilename(full_filename
)))
3196 except network_exceptions
as err
:
3197 self
.report_error('unable to download video data: %s' % error_to_compat_str(err
))
3199 except OSError as err
:
3200 raise UnavailableVideoError(err
)
3201 except (ContentTooShortError
, ) as err
:
3202 self
.report_error(f
'content too short (expected {err.expected} bytes and served {err.downloaded})')
3205 self
._raise
_pending
_errors
(info_dict
)
3206 if success
and full_filename
!= '-':
3210 fixup_policy
= self
.params
.get('fixup')
3211 vid
= info_dict
['id']
3213 if fixup_policy
in ('ignore', 'never'):
3215 elif fixup_policy
== 'warn':
3217 elif fixup_policy
!= 'force':
3218 assert fixup_policy
in ('detect_or_warn', None)
3219 if not info_dict
.get('__real_download'):
3222 def ffmpeg_fixup(cndn
, msg
, cls
):
3223 if not (do_fixup
and cndn
):
3225 elif do_fixup
== 'warn':
3226 self
.report_warning(f
'{vid}: {msg}')
3230 info_dict
['__postprocessors'].append(pp
)
3232 self
.report_warning(f
'{vid}: {msg}. Install ffmpeg to fix this automatically')
3234 stretched_ratio
= info_dict
.get('stretched_ratio')
3235 ffmpeg_fixup(stretched_ratio
not in (1, None),
3236 f
'Non-uniform pixel ratio {stretched_ratio}',
3237 FFmpegFixupStretchedPP
)
3239 downloader
= get_suitable_downloader(info_dict
, self
.params
) if 'protocol' in info_dict
else None
3240 downloader
= downloader
.FD_NAME
if downloader
else None
3242 ext
= info_dict
.get('ext')
3243 postprocessed_by_ffmpeg
= info_dict
.get('requested_formats') or any((
3244 isinstance(pp
, FFmpegVideoConvertorPP
)
3245 and resolve_recode_mapping(ext
, pp
.mapping
)[0] not in (ext
, None)
3246 ) for pp
in self
._pps
['post_process'])
3248 if not postprocessed_by_ffmpeg
:
3249 ffmpeg_fixup(ext
== 'm4a' and info_dict
.get('container') == 'm4a_dash',
3250 'writing DASH m4a. Only some players support this container',
3252 ffmpeg_fixup(downloader
== 'hlsnative' and not self
.params
.get('hls_use_mpegts')
3253 or info_dict
.get('is_live') and self
.params
.get('hls_use_mpegts') is None,
3254 'Possible MPEG-TS in MP4 container or malformed AAC timestamps',
3256 ffmpeg_fixup(info_dict
.get('is_live') and downloader
== 'DashSegmentsFD',
3257 'Possible duplicate MOOV atoms', FFmpegFixupDuplicateMoovPP
)
3259 ffmpeg_fixup(downloader
== 'web_socket_fragment', 'Malformed timestamps detected', FFmpegFixupTimestampPP
)
3260 ffmpeg_fixup(downloader
== 'web_socket_fragment', 'Malformed duration detected', FFmpegFixupDurationPP
)
3264 replace_info_dict(self
.post_process(dl_filename
, info_dict
, files_to_move
))
3265 except PostProcessingError
as err
:
3266 self
.report_error('Postprocessing: %s' % str(err
))
3269 for ph
in self
._post
_hooks
:
3270 ph(info_dict
['filepath'])
3271 except Exception as err
:
3272 self
.report_error('post hooks: %s' % str(err
))
3274 info_dict
['__write_download_archive'] = True
3276 assert info_dict
is original_infodict
# Make sure the info_dict was modified in-place
3277 if self
.params
.get('force_write_download_archive'):
3278 info_dict
['__write_download_archive'] = True
3279 check_max_downloads()
3281 def __download_wrapper(self
, func
):
3282 @functools.wraps(func
)
3283 def wrapper(*args
, **kwargs
):
3285 res
= func(*args
, **kwargs
)
3286 except UnavailableVideoError
as e
:
3287 self
.report_error(e
)
3288 except DownloadCancelled
as e
:
3289 self
.to_screen(f
'[info] {e}')
3290 if not self
.params
.get('break_per_url'):
3292 self
._num
_downloads
= 0
3294 if self
.params
.get('dump_single_json', False):
3295 self
.post_extract(res
)
3296 self
.to_stdout(json
.dumps(self
.sanitize_info(res
)))
3299 def download(self
, url_list
):
3300 """Download a given list of URLs."""
3301 url_list
= variadic(url_list
) # Passing a single URL is a common mistake
3302 outtmpl
= self
.params
['outtmpl']['default']
3303 if (len(url_list
) > 1
3305 and '%' not in outtmpl
3306 and self
.params
.get('max_downloads') != 1):
3307 raise SameFileError(outtmpl
)
3309 for url
in url_list
:
3310 self
.__download
_wrapper
(self
.extract_info
)(
3311 url
, force_generic_extractor
=self
.params
.get('force_generic_extractor', False))
3313 return self
._download
_retcode
3315 def download_with_info_file(self
, info_filename
):
3316 with contextlib
.closing(fileinput
.FileInput(
3317 [info_filename
], mode
='r',
3318 openhook
=fileinput
.hook_encoded('utf-8'))) as f
:
3319 # FileInput doesn't have a read method, we can't call json.load
3320 info
= self
.sanitize_info(json
.loads('\n'.join(f
)), self
.params
.get('clean_infojson', True))
3322 self
.__download
_wrapper
(self
.process_ie_result
)(info
, download
=True)
3323 except (DownloadError
, EntryNotInPlaylist
, ReExtractInfo
) as e
:
3324 if not isinstance(e
, EntryNotInPlaylist
):
3325 self
.to_stderr('\r')
3326 webpage_url
= info
.get('webpage_url')
3327 if webpage_url
is not None:
3328 self
.report_warning(f
'The info failed to download: {e}; trying with URL {webpage_url}')
3329 return self
.download([webpage_url
])
3332 return self
._download
_retcode
3335 def sanitize_info(info_dict
, remove_private_keys
=False):
3336 ''' Sanitize the infodict for converting to json '''
3337 if info_dict
is None:
3339 info_dict
.setdefault('epoch', int(time
.time()))
3340 info_dict
.setdefault('_type', 'video')
3341 info_dict
.setdefault('_version', {
3342 'version': __version__
,
3343 'current_git_head': current_git_head(),
3344 'release_git_head': RELEASE_GIT_HEAD
,
3345 'repository': REPOSITORY
,
3348 if remove_private_keys
:
3349 reject
= lambda k
, v
: v
is None or k
.startswith('__') or k
in {
3350 'requested_downloads', 'requested_formats', 'requested_subtitles', 'requested_entries',
3351 'entries', 'filepath', '_filename', 'infojson_filename', 'original_url', 'playlist_autonumber',
3354 reject
= lambda k
, v
: False
3357 if isinstance(obj
, dict):
3358 return {k: filter_fn(v) for k, v in obj.items() if not reject(k, v)}
3359 elif isinstance(obj
, (list, tuple, set, LazyList
)):
3360 return list(map(filter_fn
, obj
))
3361 elif obj
is None or isinstance(obj
, (str, int, float, bool)):
3366 return filter_fn(info_dict
)
3369 def filter_requested_info(info_dict
, actually_filter
=True):
3370 ''' Alias of sanitize_info for backward compatibility '''
3371 return YoutubeDL
.sanitize_info(info_dict
, actually_filter
)
3373 def _delete_downloaded_files(self
, *files_to_delete
, info
={}, msg
=None):
3374 for filename
in set(filter(None, files_to_delete
)):
3376 self
.to_screen(msg
% filename
)
3380 self
.report_warning(f
'Unable to delete file {filename}')
3381 if filename
in info
.get('__files_to_move', []): # NB: Delete even if None
3382 del info
['__files_to_move'][filename
]
3385 def post_extract(info_dict
):
3386 def actual_post_extract(info_dict
):
3387 if info_dict
.get('_type') in ('playlist', 'multi_video'):
3388 for video_dict
in info_dict
.get('entries', {}):
3389 actual_post_extract(video_dict
or {})
3392 post_extractor
= info_dict
.pop('__post_extractor', None) or (lambda: {})
3393 info_dict
.update(post_extractor())
3395 actual_post_extract(info_dict
or {})
3397 def run_pp(self
, pp
, infodict
):
3398 files_to_delete
= []
3399 if '__files_to_move' not in infodict
:
3400 infodict
['__files_to_move'] = {}
3402 files_to_delete
, infodict
= pp
.run(infodict
)
3403 except PostProcessingError
as e
:
3404 # Must be True and not 'only_download'
3405 if self
.params
.get('ignoreerrors') is True:
3406 self
.report_error(e
)
3410 if not files_to_delete
:
3412 if self
.params
.get('keepvideo', False):
3413 for f
in files_to_delete
:
3414 infodict
['__files_to_move'].setdefault(f
, '')
3416 self
._delete
_downloaded
_files
(
3417 *files_to_delete
, info
=infodict
, msg
='Deleting original file %s (pass -k to keep)')
3420 def run_all_pps(self
, key
, info
, *, additional_pps
=None):
3421 self
._forceprint
(key
, info
)
3422 for pp
in (additional_pps
or []) + self
._pps
[key
]:
3423 info
= self
.run_pp(pp
, info
)
3426 def pre_process(self
, ie_info
, key
='pre_process', files_to_move
=None):
3427 info
= dict(ie_info
)
3428 info
['__files_to_move'] = files_to_move
or {}
3430 info
= self
.run_all_pps(key
, info
)
3431 except PostProcessingError
as err
:
3432 msg
= f
'Preprocessing: {err}'
3433 info
.setdefault('__pending_error', msg
)
3434 self
.report_error(msg
, is_error
=False)
3435 return info
, info
.pop('__files_to_move', None)
3437 def post_process(self
, filename
, info
, files_to_move
=None):
3438 """Run all the postprocessors on the given file."""
3439 info
['filepath'] = filename
3440 info
['__files_to_move'] = files_to_move
or {}
3441 info
= self
.run_all_pps('post_process', info
, additional_pps
=info
.get('__postprocessors'))
3442 info
= self
.run_pp(MoveFilesAfterDownloadPP(self
), info
)
3443 del info
['__files_to_move']
3444 return self
.run_all_pps('after_move', info
)
3446 def _make_archive_id(self
, info_dict
):
3447 video_id
= info_dict
.get('id')
3450 # Future-proof against any change in case
3451 # and backwards compatibility with prior versions
3452 extractor
= info_dict
.get('extractor_key') or info_dict
.get('ie_key') # key in a playlist
3453 if extractor
is None:
3454 url
= str_or_none(info_dict
.get('url'))
3457 # Try to find matching extractor for the URL and take its ie_key
3458 for ie_key
, ie
in self
._ies
.items():
3459 if ie
.suitable(url
):
3464 return make_archive_id(extractor
, video_id
)
3466 def in_download_archive(self
, info_dict
):
3467 fn
= self
.params
.get('download_archive')
3471 vid_ids
= [self
._make
_archive
_id
(info_dict
)]
3472 vid_ids
.extend(info_dict
.get('_old_archive_ids') or [])
3473 return any(id_
in self
.archive
for id_
in vid_ids
)
3475 def record_download_archive(self
, info_dict
):
3476 fn
= self
.params
.get('download_archive')
3479 vid_id
= self
._make
_archive
_id
(info_dict
)
3481 self
.write_debug(f
'Adding to archive: {vid_id}')
3482 with locked_file(fn
, 'a', encoding
='utf-8') as archive_file
:
3483 archive_file
.write(vid_id
+ '\n')
3484 self
.archive
.add(vid_id
)
3487 def format_resolution(format
, default
='unknown'):
3488 if format
.get('vcodec') == 'none' and format
.get('acodec') != 'none':
3490 if format
.get('resolution') is not None:
3491 return format
['resolution']
3492 if format
.get('width') and format
.get('height'):
3493 return '%dx%d' % (format
['width'], format
['height'])
3494 elif format
.get('height'):
3495 return '%sp' % format
['height']
3496 elif format
.get('width'):
3497 return '%dx?' % format
['width']
3500 def _list_format_headers(self
, *headers
):
3501 if self
.params
.get('listformats_table', True) is not False:
3502 return [self
._format
_out
(header
, self
.Styles
.HEADERS
) for header
in headers
]
3505 def _format_note(self
, fdict
):
3507 if fdict
.get('ext') in ['f4f', 'f4m']:
3508 res
+= '(unsupported)'
3509 if fdict
.get('language'):
3512 res
+= '[%s]' % fdict
['language']
3513 if fdict
.get('format_note') is not None:
3516 res
+= fdict
['format_note']
3517 if fdict
.get('tbr') is not None:
3520 res
+= '%4dk' % fdict
['tbr']
3521 if fdict
.get('container') is not None:
3524 res
+= '%s container' % fdict
['container']
3525 if (fdict
.get('vcodec') is not None
3526 and fdict
.get('vcodec') != 'none'):
3529 res
+= fdict
['vcodec']
3530 if fdict
.get('vbr') is not None:
3532 elif fdict
.get('vbr') is not None and fdict
.get('abr') is not None:
3534 if fdict
.get('vbr') is not None:
3535 res
+= '%4dk' % fdict
['vbr']
3536 if fdict
.get('fps') is not None:
3539 res
+= '%sfps' % fdict
['fps']
3540 if fdict
.get('acodec') is not None:
3543 if fdict
['acodec'] == 'none':
3546 res
+= '%-5s' % fdict
['acodec']
3547 elif fdict
.get('abr') is not None:
3551 if fdict
.get('abr') is not None:
3552 res
+= '@%3dk' % fdict
['abr']
3553 if fdict
.get('asr') is not None:
3554 res
+= ' (%5dHz)' % fdict
['asr']
3555 if fdict
.get('filesize') is not None:
3558 res
+= format_bytes(fdict
['filesize'])
3559 elif fdict
.get('filesize_approx') is not None:
3562 res
+= '~' + format_bytes(fdict
['filesize_approx'])
3565 def render_formats_table(self
, info_dict
):
3566 if not info_dict
.get('formats') and not info_dict
.get('url'):
3569 formats
= info_dict
.get('formats', [info_dict
])
3570 if not self
.params
.get('listformats_table', True) is not False:
3573 format_field(f
, 'format_id'),
3574 format_field(f
, 'ext'),
3575 self
.format_resolution(f
),
3576 self
._format
_note
(f
)
3577 ] for f
in formats
if f
.get('preference') is None or f
['preference'] >= -1000]
3578 return render_table(['format code', 'extension', 'resolution', 'note'], table
, extra_gap
=1)
3580 def simplified_codec(f
, field
):
3581 assert field
in ('acodec', 'vcodec')
3582 codec
= f
.get(field
, 'unknown')
3585 elif codec
!= 'none':
3586 return '.'.join(codec
.split('.')[:4])
3588 if field
== 'vcodec' and f
.get('acodec') == 'none':
3590 elif field
== 'acodec' and f
.get('vcodec') == 'none':
3592 return self
._format
_out
('audio only' if field
== 'vcodec' else 'video only',
3593 self
.Styles
.SUPPRESS
)
3595 delim
= self
._format
_out
('\u2502', self
.Styles
.DELIM
, '|', test_encoding
=True)
3598 self
._format
_out
(format_field(f
, 'format_id'), self
.Styles
.ID
),
3599 format_field(f
, 'ext'),
3600 format_field(f
, func
=self
.format_resolution
, ignore
=('audio only', 'images')),
3601 format_field(f
, 'fps', '\t%d', func
=round),
3602 format_field(f
, 'dynamic_range', '%s', ignore
=(None, 'SDR')).replace('HDR', ''),
3603 format_field(f
, 'audio_channels', '\t%s'),
3605 format_field(f
, 'filesize', ' \t%s', func
=format_bytes
) + format_field(f
, 'filesize_approx', '~\t%s', func
=format_bytes
),
3606 format_field(f
, 'tbr', '\t%dk', func
=round),
3607 shorten_protocol_name(f
.get('protocol', '')),
3609 simplified_codec(f
, 'vcodec'),
3610 format_field(f
, 'vbr', '\t%dk', func
=round),
3611 simplified_codec(f
, 'acodec'),
3612 format_field(f
, 'abr', '\t%dk', func
=round),
3613 format_field(f
, 'asr', '\t%s', func
=format_decimal_suffix
),
3615 self
._format
_out
('UNSUPPORTED', 'light red') if f
.get('ext') in ('f4f', 'f4m') else None,
3616 format_field(f
, 'language', '[%s]'),
3617 join_nonempty(format_field(f
, 'format_note'),
3618 format_field(f
, 'container', ignore
=(None, f
.get('ext'))),
3621 ] for f
in formats
if f
.get('preference') is None or f
['preference'] >= -1000]
3622 header_line
= self
._list
_format
_headers
(
3623 'ID', 'EXT', 'RESOLUTION', '\tFPS', 'HDR', 'CH', delim
, '\tFILESIZE', '\tTBR', 'PROTO',
3624 delim
, 'VCODEC', '\tVBR', 'ACODEC', '\tABR', '\tASR', 'MORE INFO')
3626 return render_table(
3627 header_line
, table
, hide_empty
=True,
3628 delim
=self
._format
_out
('\u2500', self
.Styles
.DELIM
, '-', test_encoding
=True))
3630 def render_thumbnails_table(self
, info_dict
):
3631 thumbnails
= list(info_dict
.get('thumbnails') or [])
3634 return render_table(
3635 self
._list
_format
_headers
('ID', 'Width', 'Height', 'URL'),
3636 [[t
.get('id'), t
.get('width', 'unknown'), t
.get('height', 'unknown'), t
['url']] for t
in thumbnails
])
3638 def render_subtitles_table(self
, video_id
, subtitles
):
3639 def _row(lang
, formats
):
3640 exts
, names
= zip(*((f
['ext'], f
.get('name') or 'unknown') for f
in reversed(formats
)))
3641 if len(set(names
)) == 1:
3642 names
= [] if names
[0] == 'unknown' else names
[:1]
3643 return [lang
, ', '.join(names
), ', '.join(exts
)]
3647 return render_table(
3648 self
._list
_format
_headers
('Language', 'Name', 'Formats'),
3649 [_row(lang
, formats
) for lang
, formats
in subtitles
.items()],
3652 def __list_table(self
, video_id
, name
, func
, *args
):
3655 self
.to_screen(f
'{video_id} has no {name}')
3657 self
.to_screen(f
'[info] Available {name} for {video_id}:')
3658 self
.to_stdout(table
)
3660 def list_formats(self
, info_dict
):
3661 self
.__list
_table
(info_dict
['id'], 'formats', self
.render_formats_table
, info_dict
)
3663 def list_thumbnails(self
, info_dict
):
3664 self
.__list
_table
(info_dict
['id'], 'thumbnails', self
.render_thumbnails_table
, info_dict
)
3666 def list_subtitles(self
, video_id
, subtitles
, name
='subtitles'):
3667 self
.__list
_table
(video_id
, name
, self
.render_subtitles_table
, video_id
, subtitles
)
3669 def urlopen(self
, req
):
3670 """ Start an HTTP download """
3671 if isinstance(req
, str):
3672 req
= sanitized_Request(req
)
3673 return self
._opener
.open(req
, timeout
=self
._socket
_timeout
)
3675 def print_debug_header(self
):
3676 if not self
.params
.get('verbose'):
3679 # These imports can be slow. So import them only as needed
3680 from .extractor
.extractors
import _LAZY_LOADER
3681 from .extractor
.extractors
import _PLUGIN_CLASSES
as plugin_extractors
3683 def get_encoding(stream
):
3684 ret
= str(getattr(stream
, 'encoding', 'missing (%s)' % type(stream
).__name
__))
3685 if not supports_terminal_sequences(stream
):
3686 from .utils
import WINDOWS_VT_MODE
# Must be imported locally
3687 ret
+= ' (No VT)' if WINDOWS_VT_MODE
is False else ' (No ANSI)'
3690 encoding_str
= 'Encodings: locale %s, fs %s, pref %s, %s' % (
3691 locale
.getpreferredencoding(),
3692 sys
.getfilesystemencoding(),
3693 self
.get_encoding(),
3695 f
'{key} {get_encoding(stream)}' for key
, stream
in self
._out
_files
.items_
3696 if stream
is not None and key
!= 'console')
3699 logger
= self
.params
.get('logger')
3701 write_debug
= lambda msg
: logger
.debug(f
'[debug] {msg}')
3702 write_debug(encoding_str
)
3704 write_string(f
'[debug] {encoding_str}\n', encoding
=None)
3705 write_debug
= lambda msg
: self
._write
_string
(f
'[debug] {msg}\n')
3707 source
= detect_variant()
3708 if VARIANT
not in (None, 'pip'):
3710 write_debug(join_nonempty(
3711 f
'{"yt-dlp" if REPOSITORY == "yt-dlp/yt-dlp" else REPOSITORY} version',
3713 f
'[{RELEASE_GIT_HEAD}]' if RELEASE_GIT_HEAD
else '',
3714 '' if source
== 'unknown' else f
'({source})',
3716 if not _LAZY_LOADER
:
3717 if os
.environ
.get('YTDLP_NO_LAZY_EXTRACTORS'):
3718 write_debug('Lazy loading extractors is forcibly disabled')
3720 write_debug('Lazy loading extractors is disabled')
3721 if plugin_extractors
or plugin_postprocessors
:
3722 write_debug('Plugins: %s' % [
3723 '%s%s' % (klass
.__name
__, '' if klass
.__name
__ == name
else f
' as {name}')
3724 for name
, klass
in itertools
.chain(plugin_extractors
.items(), plugin_postprocessors
.items())])
3725 if self
.params
['compat_opts']:
3726 write_debug('Compatibility options: %s' % ', '.join(self
.params
['compat_opts']))
3728 if current_git_head():
3729 write_debug(f
'Git HEAD: {current_git_head()}')
3730 write_debug(system_identifier())
3732 exe_versions
, ffmpeg_features
= FFmpegPostProcessor
.get_versions_and_features(self
)
3733 ffmpeg_features
= {key for key, val in ffmpeg_features.items() if val}
3735 exe_versions
['ffmpeg'] += ' (%s)' % ','.join(sorted(ffmpeg_features
))
3737 exe_versions
['rtmpdump'] = rtmpdump_version()
3738 exe_versions
['phantomjs'] = PhantomJSwrapper
._version
()
3739 exe_str
= ', '.join(
3740 f
'{exe} {v}' for exe
, v
in sorted(exe_versions
.items()) if v
3742 write_debug('exe versions: %s' % exe_str
)
3744 from .compat
.compat_utils
import get_package_info
3745 from .dependencies
import available_dependencies
3747 write_debug('Optional libraries: %s' % (', '.join(sorted({
3748 join_nonempty(*get_package_info(m
)) for m
in available_dependencies
.values()
3751 self
._setup
_opener
()
3753 for handler
in self
._opener
.handlers
:
3754 if hasattr(handler
, 'proxies'):
3755 proxy_map
.update(handler
.proxies
)
3756 write_debug(f
'Proxy map: {proxy_map}')
3759 if False and self
.params
.get('call_home'):
3760 ipaddr
= self
.urlopen('https://yt-dl.org/ip').read().decode()
3761 write_debug('Public IP address: %s' % ipaddr
)
3762 latest_version
= self
.urlopen(
3763 'https://yt-dl.org/latest/version').read().decode()
3764 if version_tuple(latest_version
) > version_tuple(__version__
):
3765 self
.report_warning(
3766 'You are using an outdated version (newest version: %s)! '
3767 'See https://yt-dl.org/update if you need help updating.' %
3770 def _setup_opener(self
):
3771 if hasattr(self
, '_opener'):
3773 timeout_val
= self
.params
.get('socket_timeout')
3774 self
._socket
_timeout
= 20 if timeout_val
is None else float(timeout_val
)
3776 opts_cookiesfrombrowser
= self
.params
.get('cookiesfrombrowser')
3777 opts_cookiefile
= self
.params
.get('cookiefile')
3778 opts_proxy
= self
.params
.get('proxy')
3780 self
.cookiejar
= load_cookies(opts_cookiefile
, opts_cookiesfrombrowser
, self
)
3782 cookie_processor
= YoutubeDLCookieProcessor(self
.cookiejar
)
3783 if opts_proxy
is not None:
3784 if opts_proxy
== '':
3787 proxies
= {'http': opts_proxy, 'https': opts_proxy}
3789 proxies
= urllib
.request
.getproxies()
3790 # Set HTTPS proxy to HTTP one if given (https://github.com/ytdl-org/youtube-dl/issues/805)
3791 if 'http' in proxies
and 'https' not in proxies
:
3792 proxies
['https'] = proxies
['http']
3793 proxy_handler
= PerRequestProxyHandler(proxies
)
3795 debuglevel
= 1 if self
.params
.get('debug_printtraffic') else 0
3796 https_handler
= make_HTTPS_handler(self
.params
, debuglevel
=debuglevel
)
3797 ydlh
= YoutubeDLHandler(self
.params
, debuglevel
=debuglevel
)
3798 redirect_handler
= YoutubeDLRedirectHandler()
3799 data_handler
= urllib
.request
.DataHandler()
3801 # When passing our own FileHandler instance, build_opener won't add the
3802 # default FileHandler and allows us to disable the file protocol, which
3803 # can be used for malicious purposes (see
3804 # https://github.com/ytdl-org/youtube-dl/issues/8227)
3805 file_handler
= urllib
.request
.FileHandler()
3807 def file_open(*args
, **kwargs
):
3808 raise urllib
.error
.URLError('file:// scheme is explicitly disabled in yt-dlp for security reasons')
3809 file_handler
.file_open
= file_open
3811 opener
= urllib
.request
.build_opener(
3812 proxy_handler
, https_handler
, cookie_processor
, ydlh
, redirect_handler
, data_handler
, file_handler
)
3814 # Delete the default user-agent header, which would otherwise apply in
3815 # cases where our custom HTTP handler doesn't come into play
3816 # (See https://github.com/ytdl-org/youtube-dl/issues/1309 for details)
3817 opener
.addheaders
= []
3818 self
._opener
= opener
3820 def encode(self
, s
):
3821 if isinstance(s
, bytes):
3822 return s
# Already encoded
3825 return s
.encode(self
.get_encoding())
3826 except UnicodeEncodeError as err
:
3827 err
.reason
= err
.reason
+ '. Check your system encoding configuration or use the --encoding option.'
3830 def get_encoding(self
):
3831 encoding
= self
.params
.get('encoding')
3832 if encoding
is None:
3833 encoding
= preferredencoding()
3836 def _write_info_json(self
, label
, ie_result
, infofn
, overwrite
=None):
3837 ''' Write infojson and returns True = written, 'exists' = Already exists, False = skip, None = error '''
3838 if overwrite
is None:
3839 overwrite
= self
.params
.get('overwrites', True)
3840 if not self
.params
.get('writeinfojson'):
3843 self
.write_debug(f
'Skipping writing {label} infojson')
3845 elif not self
._ensure
_dir
_exists
(infofn
):
3847 elif not overwrite
and os
.path
.exists(infofn
):
3848 self
.to_screen(f
'[info] {label.title()} metadata is already present')
3851 self
.to_screen(f
'[info] Writing {label} metadata as JSON to: {infofn}')
3853 write_json_file(self
.sanitize_info(ie_result
, self
.params
.get('clean_infojson', True)), infofn
)
3856 self
.report_error(f
'Cannot write {label} metadata to JSON file {infofn}')
3859 def _write_description(self
, label
, ie_result
, descfn
):
3860 ''' Write description and returns True = written, False = skip, None = error '''
3861 if not self
.params
.get('writedescription'):
3864 self
.write_debug(f
'Skipping writing {label} description')
3866 elif not self
._ensure
_dir
_exists
(descfn
):
3868 elif not self
.params
.get('overwrites', True) and os
.path
.exists(descfn
):
3869 self
.to_screen(f
'[info] {label.title()} description is already present')
3870 elif ie_result
.get('description') is None:
3871 self
.report_warning(f
'There\'s no {label} description to write')
3875 self
.to_screen(f
'[info] Writing {label} description to: {descfn}')
3876 with open(encodeFilename(descfn
), 'w', encoding
='utf-8') as descfile
:
3877 descfile
.write(ie_result
['description'])
3879 self
.report_error(f
'Cannot write {label} description file {descfn}')
3883 def _write_subtitles(self
, info_dict
, filename
):
3884 ''' Write subtitles to file and return list of (sub_filename, final_sub_filename); or None if error'''
3886 subtitles
= info_dict
.get('requested_subtitles')
3887 if not subtitles
or not (self
.params
.get('writesubtitles') or self
.params
.get('writeautomaticsub')):
3888 # subtitles download errors are already managed as troubles in relevant IE
3889 # that way it will silently go on when used with unsupporting IE
3892 sub_filename_base
= self
.prepare_filename(info_dict
, 'subtitle')
3893 if not sub_filename_base
:
3894 self
.to_screen('[info] Skipping writing video subtitles')
3896 for sub_lang
, sub_info
in subtitles
.items():
3897 sub_format
= sub_info
['ext']
3898 sub_filename
= subtitles_filename(filename
, sub_lang
, sub_format
, info_dict
.get('ext'))
3899 sub_filename_final
= subtitles_filename(sub_filename_base
, sub_lang
, sub_format
, info_dict
.get('ext'))
3900 existing_sub
= self
.existing_file((sub_filename_final
, sub_filename
))
3902 self
.to_screen(f
'[info] Video subtitle {sub_lang}.{sub_format} is already present')
3903 sub_info
['filepath'] = existing_sub
3904 ret
.append((existing_sub
, sub_filename_final
))
3907 self
.to_screen(f
'[info] Writing video subtitles to: {sub_filename}')
3908 if sub_info
.get('data') is not None:
3910 # Use newline='' to prevent conversion of newline characters
3911 # See https://github.com/ytdl-org/youtube-dl/issues/10268
3912 with open(sub_filename
, 'w', encoding
='utf-8', newline
='') as subfile
:
3913 subfile
.write(sub_info
['data'])
3914 sub_info
['filepath'] = sub_filename
3915 ret
.append((sub_filename
, sub_filename_final
))
3918 self
.report_error(f
'Cannot write video subtitles file {sub_filename}')
3922 sub_copy
= sub_info
.copy()
3923 sub_copy
.setdefault('http_headers', info_dict
.get('http_headers'))
3924 self
.dl(sub_filename
, sub_copy
, subtitle
=True)
3925 sub_info
['filepath'] = sub_filename
3926 ret
.append((sub_filename
, sub_filename_final
))
3927 except (DownloadError
, ExtractorError
, IOError, OSError, ValueError) + network_exceptions
as err
:
3928 msg
= f
'Unable to download video subtitles for {sub_lang!r}: {err}'
3929 if self
.params
.get('ignoreerrors') is not True: # False or 'only_download'
3930 if not self
.params
.get('ignoreerrors'):
3931 self
.report_error(msg
)
3932 raise DownloadError(msg
)
3933 self
.report_warning(msg
)
3936 def _write_thumbnails(self
, label
, info_dict
, filename
, thumb_filename_base
=None):
3937 ''' Write thumbnails to file and return list of (thumb_filename, final_thumb_filename) '''
3938 write_all
= self
.params
.get('write_all_thumbnails', False)
3939 thumbnails
, ret
= [], []
3940 if write_all
or self
.params
.get('writethumbnail', False):
3941 thumbnails
= info_dict
.get('thumbnails') or []
3942 multiple
= write_all
and len(thumbnails
) > 1
3944 if thumb_filename_base
is None:
3945 thumb_filename_base
= filename
3946 if thumbnails
and not thumb_filename_base
:
3947 self
.write_debug(f
'Skipping writing {label} thumbnail')
3950 for idx
, t
in list(enumerate(thumbnails
))[::-1]:
3951 thumb_ext
= (f
'{t["id"]}.' if multiple
else '') + determine_ext(t
['url'], 'jpg')
3952 thumb_display_id
= f
'{label} thumbnail {t["id"]}'
3953 thumb_filename
= replace_extension(filename
, thumb_ext
, info_dict
.get('ext'))
3954 thumb_filename_final
= replace_extension(thumb_filename_base
, thumb_ext
, info_dict
.get('ext'))
3956 existing_thumb
= self
.existing_file((thumb_filename_final
, thumb_filename
))
3958 self
.to_screen('[info] %s is already present' % (
3959 thumb_display_id
if multiple
else f
'{label} thumbnail').capitalize())
3960 t
['filepath'] = existing_thumb
3961 ret
.append((existing_thumb
, thumb_filename_final
))
3963 self
.to_screen(f
'[info] Downloading {thumb_display_id} ...')
3965 uf
= self
.urlopen(sanitized_Request(t
['url'], headers
=t
.get('http_headers', {})))
3966 self
.to_screen(f
'[info] Writing {thumb_display_id} to: {thumb_filename}')
3967 with open(encodeFilename(thumb_filename
), 'wb') as thumbf
:
3968 shutil
.copyfileobj(uf
, thumbf
)
3969 ret
.append((thumb_filename
, thumb_filename_final
))
3970 t
['filepath'] = thumb_filename
3971 except network_exceptions
as err
:
3973 self
.report_warning(f
'Unable to download {thumb_display_id}: {err}')
3974 if ret
and not write_all
: