25 from .cache
import Cache
26 from .compat
import urllib
# isort: split
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 .plugins
import directories
as plugin_directories
36 from .postprocessor
import _PLUGIN_CLASSES
as plugin_pps
37 from .postprocessor
import (
39 FFmpegFixupDuplicateMoovPP
,
40 FFmpegFixupDurationPP
,
43 FFmpegFixupStretchedPP
,
44 FFmpegFixupTimestampPP
,
47 FFmpegVideoConvertorPP
,
48 MoveFilesAfterDownloadPP
,
51 from .postprocessor
.ffmpeg
import resolve_mapping
as resolve_recode_mapping
52 from .update
import REPOSITORY
, current_git_head
, detect_variant
79 PerRequestProxyHandler
,
86 UnavailableVideoError
,
88 YoutubeDLCookieProcessor
,
90 YoutubeDLRedirectHandler
,
106 format_decimal_suffix
,
123 orderedSet_from_options
,
127 remove_terminal_sequences
,
138 supports_terminal_sequences
,
148 windows_enable_vt_mode
,
152 from .version
import CHANNEL
, RELEASE_GIT_HEAD
, VARIANT
, __version__
154 if compat_os_name
== 'nt':
161 YoutubeDL objects are the ones responsible of downloading the
162 actual video file and writing it to disk if the user has requested
163 it, among some other tasks. In most cases there should be one per
164 program. As, given a video URL, the downloader doesn't know how to
165 extract all the needed information, task that InfoExtractors do, it
166 has to pass the URL to one of them.
168 For this, YoutubeDL objects have a method that allows
169 InfoExtractors to be registered in a given order. When it is passed
170 a URL, the YoutubeDL object handles it to the first InfoExtractor it
171 finds that reports being able to handle it. The InfoExtractor extracts
172 all the information about the video or videos the URL refers to, and
173 YoutubeDL process the extracted information, possibly using a File
174 Downloader to download the video.
176 YoutubeDL objects accept a lot of parameters. In order not to saturate
177 the object constructor with arguments, it receives a dictionary of
178 options instead. These options are available through the params
179 attribute for the InfoExtractors to use. The YoutubeDL also
180 registers itself as the downloader in charge for the InfoExtractors
181 that are added to it, so this is a "mutual registration".
185 username: Username for authentication purposes.
186 password: Password for authentication purposes.
187 videopassword: Password for accessing a video.
188 ap_mso: Adobe Pass multiple-system operator identifier.
189 ap_username: Multiple-system operator account username.
190 ap_password: Multiple-system operator account password.
191 usenetrc: Use netrc for authentication instead.
192 netrc_location: Location of the netrc file. Defaults to ~/.netrc.
193 netrc_cmd: Use a shell command to get credentials
194 verbose: Print additional info to stdout.
195 quiet: Do not print messages to stdout.
196 no_warnings: Do not print out anything for warnings.
197 forceprint: A dict with keys WHEN mapped to a list of templates to
198 print to stdout. The allowed keys are video or any of the
199 items in utils.POSTPROCESS_WHEN.
200 For compatibility, a single list is also accepted
201 print_to_file: A dict with keys WHEN (same as forceprint) mapped to
202 a list of tuples with (template, filename)
203 forcejson: Force printing info_dict as JSON.
204 dump_single_json: Force printing the info_dict of the whole playlist
205 (or video) as a single JSON line.
206 force_write_download_archive: Force writing download archive regardless
207 of 'skip_download' or 'simulate'.
208 simulate: Do not download the video files. If unset (or None),
209 simulate only if listsubtitles, listformats or list_thumbnails is used
210 format: Video format code. see "FORMAT SELECTION" for more details.
211 You can also pass a function. The function takes 'ctx' as
212 argument and returns the formats to download.
213 See "build_format_selector" for an implementation
214 allow_unplayable_formats: Allow unplayable formats to be extracted and downloaded.
215 ignore_no_formats_error: Ignore "No video formats" error. Usefull for
216 extracting metadata even if the video is not actually
217 available for download (experimental)
218 format_sort: A list of fields by which to sort the video formats.
219 See "Sorting Formats" for more details.
220 format_sort_force: Force the given format_sort. see "Sorting Formats"
222 prefer_free_formats: Whether to prefer video formats with free containers
223 over non-free ones of same quality.
224 allow_multiple_video_streams: Allow multiple video streams to be merged
226 allow_multiple_audio_streams: Allow multiple audio streams to be merged
228 check_formats Whether to test if the formats are downloadable.
229 Can be True (check all), False (check none),
230 'selected' (check selected formats),
231 or None (check only if requested by extractor)
232 paths: Dictionary of output paths. The allowed keys are 'home'
233 'temp' and the keys of OUTTMPL_TYPES (in utils.py)
234 outtmpl: Dictionary of templates for output names. Allowed keys
235 are 'default' and the keys of OUTTMPL_TYPES (in utils.py).
236 For compatibility with youtube-dl, a single string can also be used
237 outtmpl_na_placeholder: Placeholder for unavailable meta fields.
238 restrictfilenames: Do not allow "&" and spaces in file names
239 trim_file_name: Limit length of filename (extension excluded)
240 windowsfilenames: Force the filenames to be windows compatible
241 ignoreerrors: Do not stop on download/postprocessing errors.
242 Can be 'only_download' to ignore only download errors.
243 Default is 'only_download' for CLI, but False for API
244 skip_playlist_after_errors: Number of allowed failures until the rest of
245 the playlist is skipped
246 allowed_extractors: List of regexes to match against extractor names that are allowed
247 overwrites: Overwrite all video and metadata files if True,
248 overwrite only non-video files if None
249 and don't overwrite any file if False
250 For compatibility with youtube-dl,
251 "nooverwrites" may also be used instead
252 playlist_items: Specific indices of playlist to download.
253 playlistrandom: Download playlist items in random order.
254 lazy_playlist: Process playlist entries as they are received.
255 matchtitle: Download only matching titles.
256 rejecttitle: Reject downloads for matching titles.
257 logger: Log messages to a logging.Logger instance.
258 logtostderr: Print everything to stderr instead of stdout.
259 consoletitle: Display progress in console window's titlebar.
260 writedescription: Write the video description to a .description file
261 writeinfojson: Write the video description to a .info.json file
262 clean_infojson: Remove internal metadata from the infojson
263 getcomments: Extract video comments. This will not be written to disk
264 unless writeinfojson is also given
265 writeannotations: Write the video annotations to a .annotations.xml file
266 writethumbnail: Write the thumbnail image to a file
267 allow_playlist_files: Whether to write playlists' description, infojson etc
268 also to disk when using the 'write*' options
269 write_all_thumbnails: Write all thumbnail formats to files
270 writelink: Write an internet shortcut file, depending on the
271 current platform (.url/.webloc/.desktop)
272 writeurllink: Write a Windows internet shortcut file (.url)
273 writewebloclink: Write a macOS internet shortcut file (.webloc)
274 writedesktoplink: Write a Linux internet shortcut file (.desktop)
275 writesubtitles: Write the video subtitles to a file
276 writeautomaticsub: Write the automatically generated subtitles to a file
277 listsubtitles: Lists all available subtitles for the video
278 subtitlesformat: The format code for subtitles
279 subtitleslangs: List of languages of the subtitles to download (can be regex).
280 The list may contain "all" to refer to all the available
281 subtitles. The language can be prefixed with a "-" to
282 exclude it from the requested languages, e.g. ['all', '-live_chat']
283 keepvideo: Keep the video file after post-processing
284 daterange: A utils.DateRange object, download only if the upload_date is in the range.
285 skip_download: Skip the actual download of the video file
286 cachedir: Location of the cache files in the filesystem.
287 False to disable filesystem cache.
288 noplaylist: Download single video instead of a playlist if in doubt.
289 age_limit: An integer representing the user's age in years.
290 Unsuitable videos for the given age are skipped.
291 min_views: An integer representing the minimum view count the video
292 must have in order to not be skipped.
293 Videos without view count information are always
294 downloaded. None for no limit.
295 max_views: An integer representing the maximum view count.
296 Videos that are more popular than that are not
298 Videos without view count information are always
299 downloaded. None for no limit.
300 download_archive: A set, or the name of a file where all downloads are recorded.
301 Videos already present in the file are not downloaded again.
302 break_on_existing: Stop the download process after attempting to download a
303 file that is in the archive.
304 break_per_url: Whether break_on_reject and break_on_existing
305 should act on each input URL as opposed to for the entire queue
306 cookiefile: File name or text stream from where cookies should be read and dumped to
307 cookiesfrombrowser: A tuple containing the name of the browser, the profile
308 name/path from where cookies are loaded, the name of the keyring,
309 and the container name, e.g. ('chrome', ) or
310 ('vivaldi', 'default', 'BASICTEXT') or ('firefox', 'default', None, 'Meta')
311 legacyserverconnect: Explicitly allow HTTPS connection to servers that do not
312 support RFC 5746 secure renegotiation
313 nocheckcertificate: Do not verify SSL certificates
314 client_certificate: Path to client certificate file in PEM format. May include the private key
315 client_certificate_key: Path to private key file for client certificate
316 client_certificate_password: Password for client certificate private key, if encrypted.
317 If not provided and the key is encrypted, yt-dlp will ask interactively
318 prefer_insecure: Use HTTP instead of HTTPS to retrieve information.
319 (Only supported by some extractors)
320 enable_file_urls: Enable file:// URLs. This is disabled by default for security reasons.
321 http_headers: A dictionary of custom headers to be used for all requests
322 proxy: URL of the proxy server to use
323 geo_verification_proxy: URL of the proxy to use for IP address verification
324 on geo-restricted sites.
325 socket_timeout: Time to wait for unresponsive hosts, in seconds
326 bidi_workaround: Work around buggy terminals without bidirectional text
327 support, using fridibi
328 debug_printtraffic:Print out sent and received HTTP traffic
329 default_search: Prepend this string if an input url is not valid.
330 'auto' for elaborate guessing
331 encoding: Use this encoding instead of the system-specified.
332 extract_flat: Whether to resolve and process url_results further
333 * False: Always process. Default for API
334 * True: Never process
335 * 'in_playlist': Do not process inside playlist/multi_video
336 * 'discard': Always process, but don't return the result
337 from inside playlist/multi_video
338 * 'discard_in_playlist': Same as "discard", but only for
339 playlists (not multi_video). Default for CLI
340 wait_for_video: If given, wait for scheduled streams to become available.
341 The value should be a tuple containing the range
342 (min_secs, max_secs) to wait between retries
343 postprocessors: A list of dictionaries, each with an entry
344 * key: The name of the postprocessor. See
345 yt_dlp/postprocessor/__init__.py for a list.
346 * when: When to run the postprocessor. Allowed values are
347 the entries of utils.POSTPROCESS_WHEN
348 Assumed to be 'post_process' if not given
349 progress_hooks: A list of functions that get called on download
350 progress, with a dictionary with the entries
351 * status: One of "downloading", "error", or "finished".
352 Check this first and ignore unknown values.
353 * info_dict: The extracted info_dict
355 If status is one of "downloading", or "finished", the
356 following properties may also be present:
357 * filename: The final filename (always present)
358 * tmpfilename: The filename we're currently writing to
359 * downloaded_bytes: Bytes on disk
360 * total_bytes: Size of the whole file, None if unknown
361 * total_bytes_estimate: Guess of the eventual file size,
363 * elapsed: The number of seconds since download started.
364 * eta: The estimated time in seconds, None if unknown
365 * speed: The download speed in bytes/second, None if
367 * fragment_index: The counter of the currently
368 downloaded video fragment.
369 * fragment_count: The number of fragments (= individual
370 files that will be merged)
372 Progress hooks are guaranteed to be called at least once
373 (with status "finished") if the download is successful.
374 postprocessor_hooks: A list of functions that get called on postprocessing
375 progress, with a dictionary with the entries
376 * status: One of "started", "processing", or "finished".
377 Check this first and ignore unknown values.
378 * postprocessor: Name of the postprocessor
379 * info_dict: The extracted info_dict
381 Progress hooks are guaranteed to be called at least twice
382 (with status "started" and "finished") if the processing is successful.
383 merge_output_format: "/" separated list of extensions to use when merging formats.
384 final_ext: Expected final extension; used to detect when the file was
385 already downloaded and converted
386 fixup: Automatically correct known faults of the file.
388 - "never": do nothing
389 - "warn": only emit a warning
390 - "detect_or_warn": check whether we can do anything
391 about it, warn otherwise (default)
392 source_address: Client-side IP address to bind to.
393 sleep_interval_requests: Number of seconds to sleep between requests
395 sleep_interval: Number of seconds to sleep before each download when
396 used alone or a lower bound of a range for randomized
397 sleep before each download (minimum possible number
398 of seconds to sleep) when used along with
400 max_sleep_interval:Upper bound of a range for randomized sleep before each
401 download (maximum possible number of seconds to sleep).
402 Must only be used along with sleep_interval.
403 Actual sleep time will be a random float from range
404 [sleep_interval; max_sleep_interval].
405 sleep_interval_subtitles: Number of seconds to sleep before each subtitle download
406 listformats: Print an overview of available video formats and exit.
407 list_thumbnails: Print a table of all thumbnails and exit.
408 match_filter: A function that gets called for every video with the signature
409 (info_dict, *, incomplete: bool) -> Optional[str]
410 For backward compatibility with youtube-dl, the signature
411 (info_dict) -> Optional[str] is also allowed.
412 - If it returns a message, the video is ignored.
413 - If it returns None, the video is downloaded.
414 - If it returns utils.NO_DEFAULT, the user is interactively
415 asked whether to download the video.
416 - Raise utils.DownloadCancelled(msg) to abort remaining
417 downloads when a video is rejected.
418 match_filter_func in utils.py is one example for this.
419 color: A Dictionary with output stream names as keys
420 and their respective color policy as values.
421 Can also just be a single color policy,
422 in which case it applies to all outputs.
423 Valid stream names are 'stdout' and 'stderr'.
424 Valid color policies are one of 'always', 'auto', 'no_color' or 'never'.
425 geo_bypass: Bypass geographic restriction via faking X-Forwarded-For
428 Two-letter ISO 3166-2 country code that will be used for
429 explicit geographic restriction bypassing via faking
430 X-Forwarded-For HTTP header
432 IP range in CIDR notation that will be used similarly to
434 external_downloader: A dictionary of protocol keys and the executable of the
435 external downloader to use for it. The allowed protocols
436 are default|http|ftp|m3u8|dash|rtsp|rtmp|mms.
437 Set the value to 'native' to use the native downloader
438 compat_opts: Compatibility options. See "Differences in default behavior".
439 The following options do not work when used through the API:
440 filename, abort-on-error, multistreams, no-live-chat, format-sort
441 no-clean-infojson, no-playlist-metafiles, no-keep-subs, no-attach-info-json.
442 Refer __init__.py for their implementation
443 progress_template: Dictionary of templates for progress outputs.
444 Allowed keys are 'download', 'postprocess',
445 'download-title' (console title) and 'postprocess-title'.
446 The template is mapped on a dictionary with keys 'progress' and 'info'
447 retry_sleep_functions: Dictionary of functions that takes the number of attempts
448 as argument and returns the time to sleep in seconds.
449 Allowed keys are 'http', 'fragment', 'file_access'
450 download_ranges: A callback function that gets called for every video with
451 the signature (info_dict, ydl) -> Iterable[Section].
452 Only the returned sections will be downloaded.
453 Each Section is a dict with the following keys:
454 * start_time: Start time of the section in seconds
455 * end_time: End time of the section in seconds
456 * title: Section title (Optional)
457 * index: Section number (Optional)
458 force_keyframes_at_cuts: Re-encode the video when downloading ranges to get precise cuts
459 noprogress: Do not print the progress bar
460 live_from_start: Whether to download livestreams videos from the start
462 The following parameters are not used by YoutubeDL itself, they are used by
463 the downloader (see yt_dlp/downloader/common.py):
464 nopart, updatetime, buffersize, ratelimit, throttledratelimit, min_filesize,
465 max_filesize, test, noresizebuffer, retries, file_access_retries, fragment_retries,
466 continuedl, xattr_set_filesize, hls_use_mpegts, http_chunk_size,
467 external_downloader_args, concurrent_fragment_downloads.
469 The following options are used by the post processors:
470 ffmpeg_location: Location of the ffmpeg/avconv binary; either the path
471 to the binary or its containing directory.
472 postprocessor_args: A dictionary of postprocessor/executable keys (in lower case)
473 and a list of additional command-line arguments for the
474 postprocessor/executable. The dict can also have "PP+EXE" keys
475 which are used when the given exe is used by the given PP.
476 Use 'default' as the name for arguments to passed to all PP
477 For compatibility with youtube-dl, a single list of args
480 The following options are used by the extractors:
481 extractor_retries: Number of times to retry for known errors (default: 3)
482 dynamic_mpd: Whether to process dynamic DASH manifests (default: True)
483 hls_split_discontinuity: Split HLS playlists to different formats at
484 discontinuities such as ad breaks (default: False)
485 extractor_args: A dictionary of arguments to be passed to the extractors.
486 See "EXTRACTOR ARGUMENTS" for details.
487 E.g. {'youtube': {'skip': ['dash', 'hls']}}
488 mark_watched: Mark videos watched (even with --simulate). Only for YouTube
490 The following options are deprecated and may be removed in the future:
492 break_on_reject: Stop the download process when encountering a video that
493 has been filtered out.
494 - `raise DownloadCancelled(msg)` in match_filter instead
495 force_generic_extractor: Force downloader to use the generic extractor
496 - Use allowed_extractors = ['generic', 'default']
497 playliststart: - Use playlist_items
498 Playlist item to start at.
499 playlistend: - Use playlist_items
500 Playlist item to end at.
501 playlistreverse: - Use playlist_items
502 Download playlist items in reverse order.
503 forceurl: - Use forceprint
504 Force printing final URL.
505 forcetitle: - Use forceprint
506 Force printing title.
507 forceid: - Use forceprint
509 forcethumbnail: - Use forceprint
510 Force printing thumbnail URL.
511 forcedescription: - Use forceprint
512 Force printing description.
513 forcefilename: - Use forceprint
514 Force printing final filename.
515 forceduration: - Use forceprint
516 Force printing duration.
517 allsubtitles: - Use subtitleslangs = ['all']
518 Downloads all the subtitles of the video
519 (requires writesubtitles or writeautomaticsub)
520 include_ads: - Doesn't work
522 call_home: - Not implemented
523 Boolean, true iff we are allowed to contact the
524 yt-dlp servers for debugging.
525 post_hooks: - Register a custom postprocessor
526 A list of functions that get called as the final step
527 for each video file, after all postprocessors have been
528 called. The filename will be passed as the only argument.
529 hls_prefer_native: - Use external_downloader = {'m3u8': 'native'} or {'m3u8': 'ffmpeg'}.
530 Use the native HLS downloader instead of ffmpeg/avconv
531 if True, otherwise use ffmpeg/avconv if False, otherwise
532 use downloader suggested by extractor if None.
533 prefer_ffmpeg: - avconv support is deprecated
534 If False, use avconv instead of ffmpeg if both are available,
535 otherwise prefer ffmpeg.
536 youtube_include_dash_manifest: - Use extractor_args
537 If True (default), DASH manifests and related
538 data will be downloaded and processed by extractor.
539 You can reduce network I/O by disabling it if you don't
540 care about DASH. (only for youtube)
541 youtube_include_hls_manifest: - Use extractor_args
542 If True (default), HLS manifests and related
543 data will be downloaded and processed by extractor.
544 You can reduce network I/O by disabling it if you don't
545 care about HLS. (only for youtube)
546 no_color: Same as `color='no_color'`
550 'width', 'height', 'asr', 'audio_channels', 'fps',
551 'tbr', 'abr', 'vbr', 'filesize', 'filesize_approx',
552 'timestamp', 'release_timestamp',
553 'duration', 'view_count', 'like_count', 'dislike_count', 'repost_count',
554 'average_rating', 'comment_count', 'age_limit',
555 'start_time', 'end_time',
556 'chapter_number', 'season_number', 'episode_number',
557 'track_number', 'disc_number', 'release_year',
561 # NB: Keep in sync with the docstring of extractor/common.py
562 'url', 'manifest_url', 'manifest_stream_number', 'ext', 'format', 'format_id', 'format_note',
563 'width', 'height', 'aspect_ratio', 'resolution', 'dynamic_range', 'tbr', 'abr', 'acodec', 'asr', 'audio_channels',
564 'vbr', 'fps', 'vcodec', 'container', 'filesize', 'filesize_approx', 'rows', 'columns',
565 'player_url', 'protocol', 'fragment_base_url', 'fragments', 'is_from_start',
566 'preference', 'language', 'language_preference', 'quality', 'source_preference',
567 'http_headers', 'stretched_ratio', 'no_resume', 'has_drm', 'extra_param_to_segment_url', 'hls_aes', 'downloader_options',
568 'page_url', 'app', 'play_path', 'tc_url', 'flash_version', 'rtmp_live', 'rtmp_conn', 'rtmp_protocol', 'rtmp_real_time'
570 _format_selection_exts
= {
571 'audio': set(MEDIA_EXTENSIONS
.common_audio
),
572 'video': set(MEDIA_EXTENSIONS
.common_video
+ ('3gp', )),
573 'storyboards': set(MEDIA_EXTENSIONS
.storyboards
),
576 def __init__(self
, params
=None, auto_init
=True):
577 """Create a FileDownloader object with the given options.
578 @param auto_init Whether to load the default extractors and print header (if verbose).
579 Set to 'no_verbose_header' to not print the header
585 self
._ies
_instances
= {}
586 self
._pps
= {k: [] for k in POSTPROCESS_WHEN}
587 self
._printed
_messages
= set()
588 self
._first
_webpage
_request
= True
589 self
._post
_hooks
= []
590 self
._progress
_hooks
= []
591 self
._postprocessor
_hooks
= []
592 self
._download
_retcode
= 0
593 self
._num
_downloads
= 0
595 self
._playlist
_level
= 0
596 self
._playlist
_urls
= set()
597 self
.cache
= Cache(self
)
599 stdout
= sys
.stderr
if self
.params
.get('logtostderr') else sys
.stdout
600 self
._out
_files
= Namespace(
603 screen
=sys
.stderr
if self
.params
.get('quiet') else stdout
,
604 console
=None if compat_os_name
== 'nt' else next(
605 filter(supports_terminal_sequences
, (sys
.stderr
, sys
.stdout
)), None)
609 windows_enable_vt_mode()
610 except Exception as e
:
611 self
.write_debug(f
'Failed to enable VT mode: {e}')
613 if self
.params
.get('no_color'):
614 if self
.params
.get('color') is not None:
615 self
.report_warning('Overwriting params from "color" with "no_color"')
616 self
.params
['color'] = 'no_color'
618 term_allow_color
= os
.environ
.get('TERM', '').lower() != 'dumb'
620 def process_color_policy(stream
):
621 stream_name
= {sys.stdout: 'stdout', sys.stderr: 'stderr'}
[stream
]
622 policy
= traverse_obj(self
.params
, ('color', (stream_name
, None), {str}
), get_all
=False)
623 if policy
in ('auto', None):
624 return term_allow_color
and supports_terminal_sequences(stream
)
625 assert policy
in ('always', 'never', 'no_color')
626 return {'always': True, 'never': False}
.get(policy
, policy
)
628 self
._allow
_colors
= Namespace(**{
629 name
: process_color_policy(stream
)
630 for name
, stream
in self
._out
_files
.items_
if name
!= 'console'
633 # The code is left like this to be reused for future deprecations
634 MIN_SUPPORTED
, MIN_RECOMMENDED
= (3, 7), (3, 7)
635 current_version
= sys
.version_info
[:2]
636 if current_version
< MIN_RECOMMENDED
:
637 msg
= ('Support for Python version %d.%d has been deprecated. '
638 'See https://github.com/yt-dlp/yt-dlp/issues/3764 for more details.'
639 '\n You will no longer receive updates on this version')
640 if current_version
< MIN_SUPPORTED
:
641 msg
= 'Python version %d.%d is no longer supported'
642 self
.deprecated_feature(
643 f
'{msg}! Please update to Python %d.%d or above' % (*current_version
, *MIN_RECOMMENDED
))
645 if self
.params
.get('allow_unplayable_formats'):
647 f
'You have asked for {self._format_err("UNPLAYABLE", self.Styles.EMPHASIS)} formats to be listed/downloaded. '
648 'This is a developer option intended for debugging. \n'
649 ' If you experience any issues while using this option, '
650 f
'{self._format_err("DO NOT", self.Styles.ERROR)} open a bug report')
652 if self
.params
.get('bidi_workaround', False):
655 master
, slave
= pty
.openpty()
656 width
= shutil
.get_terminal_size().columns
657 width_args
= [] if width
is None else ['-w', str(width
)]
658 sp_kwargs
= {'stdin': subprocess.PIPE, 'stdout': slave, 'stderr': self._out_files.error}
660 self
._output
_process
= Popen(['bidiv'] + width_args
, **sp_kwargs
)
662 self
._output
_process
= Popen(['fribidi', '-c', 'UTF-8'] + width_args
, **sp_kwargs
)
663 self
._output
_channel
= os
.fdopen(master
, 'rb')
664 except OSError as ose
:
665 if ose
.errno
== errno
.ENOENT
:
667 'Could not find fribidi executable, ignoring --bidi-workaround. '
668 'Make sure that fribidi is an executable file in one of the directories in your $PATH.')
672 self
.params
['compat_opts'] = set(self
.params
.get('compat_opts', ()))
673 if auto_init
and auto_init
!= 'no_verbose_header':
674 self
.print_debug_header()
676 def check_deprecated(param
, option
, suggestion
):
677 if self
.params
.get(param
) is not None:
678 self
.report_warning(f
'{option} is deprecated. Use {suggestion} instead')
682 if check_deprecated('cn_verification_proxy', '--cn-verification-proxy', '--geo-verification-proxy'):
683 if self
.params
.get('geo_verification_proxy') is None:
684 self
.params
['geo_verification_proxy'] = self
.params
['cn_verification_proxy']
686 check_deprecated('autonumber', '--auto-number', '-o "%(autonumber)s-%(title)s.%(ext)s"')
687 check_deprecated('usetitle', '--title', '-o "%(title)s-%(id)s.%(ext)s"')
688 check_deprecated('useid', '--id', '-o "%(id)s.%(ext)s"')
690 for msg
in self
.params
.get('_warnings', []):
691 self
.report_warning(msg
)
692 for msg
in self
.params
.get('_deprecation_warnings', []):
693 self
.deprecated_feature(msg
)
695 if 'list-formats' in self
.params
['compat_opts']:
696 self
.params
['listformats_table'] = False
698 if 'overwrites' not in self
.params
and self
.params
.get('nooverwrites') is not None:
699 # nooverwrites was unnecessarily changed to overwrites
700 # in 0c3d0f51778b153f65c21906031c2e091fcfb641
701 # This ensures compatibility with both keys
702 self
.params
['overwrites'] = not self
.params
['nooverwrites']
703 elif self
.params
.get('overwrites') is None:
704 self
.params
.pop('overwrites', None)
706 self
.params
['nooverwrites'] = not self
.params
['overwrites']
708 if self
.params
.get('simulate') is None and any((
709 self
.params
.get('list_thumbnails'),
710 self
.params
.get('listformats'),
711 self
.params
.get('listsubtitles'),
713 self
.params
['simulate'] = 'list_only'
715 self
.params
.setdefault('forceprint', {})
716 self
.params
.setdefault('print_to_file', {})
718 # Compatibility with older syntax
719 if not isinstance(params
['forceprint'], dict):
720 self
.params
['forceprint'] = {'video': params['forceprint']}
723 self
.add_default_info_extractors()
725 if (sys
.platform
!= 'win32'
726 and sys
.getfilesystemencoding() in ['ascii', 'ANSI_X3.4-1968']
727 and not self
.params
.get('restrictfilenames', False)):
728 # Unicode filesystem API will throw errors (#1474, #13027)
730 'Assuming --restrict-filenames since file system encoding '
731 'cannot encode all characters. '
732 'Set the LC_ALL environment variable to fix this.')
733 self
.params
['restrictfilenames'] = True
735 self
._parse
_outtmpl
()
737 # Creating format selector here allows us to catch syntax errors before the extraction
738 self
.format_selector
= (
739 self
.params
.get('format') if self
.params
.get('format') in (None, '-')
740 else self
.params
['format'] if callable(self
.params
['format'])
741 else self
.build_format_selector(self
.params
['format']))
743 # Set http_headers defaults according to std_headers
744 self
.params
['http_headers'] = merge_headers(std_headers
, self
.params
.get('http_headers', {}))
747 'post_hooks': self
.add_post_hook
,
748 'progress_hooks': self
.add_progress_hook
,
749 'postprocessor_hooks': self
.add_postprocessor_hook
,
751 for opt
, fn
in hooks
.items():
752 for ph
in self
.params
.get(opt
, []):
755 for pp_def_raw
in self
.params
.get('postprocessors', []):
756 pp_def
= dict(pp_def_raw
)
757 when
= pp_def
.pop('when', 'post_process')
758 self
.add_post_processor(
759 get_postprocessor(pp_def
.pop('key'))(self
, **pp_def
),
764 def preload_download_archive(fn
):
765 """Preload the archive, if any is specified"""
769 elif not is_path_like(fn
):
772 self
.write_debug(f
'Loading archive file {fn!r}')
774 with locked_file(fn
, 'r', encoding
='utf-8') as archive_file
:
775 for line
in archive_file
:
776 archive
.add(line
.strip())
777 except OSError as ioe
:
778 if ioe
.errno
!= errno
.ENOENT
:
782 self
.archive
= preload_download_archive(self
.params
.get('download_archive'))
784 def warn_if_short_id(self
, argv
):
785 # short YouTube ID starting with dash?
787 i
for i
, a
in enumerate(argv
)
788 if re
.match(r
'^-[0-9A-Za-z_-]{10}$', a
)]
792 + [a
for i
, a
in enumerate(argv
) if i
not in idxs
]
793 + ['--'] + [argv
[i
] for i
in idxs
]
796 'Long argument string detected. '
797 'Use -- to separate parameters and URLs, like this:\n%s' %
798 args_to_str(correct_argv
))
800 def add_info_extractor(self
, ie
):
801 """Add an InfoExtractor object to the end of the list."""
803 self
._ies
[ie_key
] = ie
804 if not isinstance(ie
, type):
805 self
._ies
_instances
[ie_key
] = ie
806 ie
.set_downloader(self
)
808 def get_info_extractor(self
, ie_key
):
810 Get an instance of an IE with name ie_key, it will try to get one from
811 the _ies list, if there's no instance it will create a new one and add
812 it to the extractor list.
814 ie
= self
._ies
_instances
.get(ie_key
)
816 ie
= get_info_extractor(ie_key
)()
817 self
.add_info_extractor(ie
)
820 def add_default_info_extractors(self
):
822 Add the InfoExtractors returned by gen_extractors to the end of the list
824 all_ies
= {ie.IE_NAME.lower(): ie for ie in gen_extractor_classes()}
825 all_ies
['end'] = UnsupportedURLIE()
827 ie_names
= orderedSet_from_options(
828 self
.params
.get('allowed_extractors', ['default']), {
829 'all': list(all_ies
),
830 'default': [name
for name
, ie
in all_ies
.items() if ie
._ENABLED
],
832 except re
.error
as e
:
833 raise ValueError(f
'Wrong regex for allowed_extractors: {e.pattern}')
834 for name
in ie_names
:
835 self
.add_info_extractor(all_ies
[name
])
836 self
.write_debug(f
'Loaded {len(ie_names)} extractors')
838 def add_post_processor(self
, pp
, when
='post_process'):
839 """Add a PostProcessor object to the end of the chain."""
840 assert when
in POSTPROCESS_WHEN
, f
'Invalid when={when}'
841 self
._pps
[when
].append(pp
)
842 pp
.set_downloader(self
)
844 def add_post_hook(self
, ph
):
845 """Add the post hook"""
846 self
._post
_hooks
.append(ph
)
848 def add_progress_hook(self
, ph
):
849 """Add the download progress hook"""
850 self
._progress
_hooks
.append(ph
)
852 def add_postprocessor_hook(self
, ph
):
853 """Add the postprocessing progress hook"""
854 self
._postprocessor
_hooks
.append(ph
)
855 for pps
in self
._pps
.values():
857 pp
.add_progress_hook(ph
)
859 def _bidi_workaround(self
, message
):
860 if not hasattr(self
, '_output_channel'):
863 assert hasattr(self
, '_output_process')
864 assert isinstance(message
, str)
865 line_count
= message
.count('\n') + 1
866 self
._output
_process
.stdin
.write((message
+ '\n').encode())
867 self
._output
_process
.stdin
.flush()
868 res
= ''.join(self
._output
_channel
.readline().decode()
869 for _
in range(line_count
))
870 return res
[:-len('\n')]
872 def _write_string(self
, message
, out
=None, only_once
=False):
874 if message
in self
._printed
_messages
:
876 self
._printed
_messages
.add(message
)
877 write_string(message
, out
=out
, encoding
=self
.params
.get('encoding'))
879 def to_stdout(self
, message
, skip_eol
=False, quiet
=None):
880 """Print message to stdout"""
881 if quiet
is not None:
882 self
.deprecation_warning('"YoutubeDL.to_stdout" no longer accepts the argument quiet. '
883 'Use "YoutubeDL.to_screen" instead')
884 if skip_eol
is not False:
885 self
.deprecation_warning('"YoutubeDL.to_stdout" no longer accepts the argument skip_eol. '
886 'Use "YoutubeDL.to_screen" instead')
887 self
._write
_string
(f
'{self._bidi_workaround(message)}\n', self
._out
_files
.out
)
889 def to_screen(self
, message
, skip_eol
=False, quiet
=None, only_once
=False):
890 """Print message to screen if not in quiet mode"""
891 if self
.params
.get('logger'):
892 self
.params
['logger'].debug(message
)
894 if (self
.params
.get('quiet') if quiet
is None else quiet
) and not self
.params
.get('verbose'):
897 '%s%s' % (self
._bidi
_workaround
(message
), ('' if skip_eol
else '\n')),
898 self
._out
_files
.screen
, only_once
=only_once
)
900 def to_stderr(self
, message
, only_once
=False):
901 """Print message to stderr"""
902 assert isinstance(message
, str)
903 if self
.params
.get('logger'):
904 self
.params
['logger'].error(message
)
906 self
._write
_string
(f
'{self._bidi_workaround(message)}\n', self
._out
_files
.error
, only_once
=only_once
)
908 def _send_console_code(self
, code
):
909 if compat_os_name
== 'nt' or not self
._out
_files
.console
:
911 self
._write
_string
(code
, self
._out
_files
.console
)
913 def to_console_title(self
, message
):
914 if not self
.params
.get('consoletitle', False):
916 message
= remove_terminal_sequences(message
)
917 if compat_os_name
== 'nt':
918 if ctypes
.windll
.kernel32
.GetConsoleWindow():
919 # c_wchar_p() might not be necessary if `message` is
920 # already of type unicode()
921 ctypes
.windll
.kernel32
.SetConsoleTitleW(ctypes
.c_wchar_p(message
))
923 self
._send
_console
_code
(f
'\033]0;{message}\007')
925 def save_console_title(self
):
926 if not self
.params
.get('consoletitle') or self
.params
.get('simulate'):
928 self
._send
_console
_code
('\033[22;0t') # Save the title on stack
930 def restore_console_title(self
):
931 if not self
.params
.get('consoletitle') or self
.params
.get('simulate'):
933 self
._send
_console
_code
('\033[23;0t') # Restore the title from stack
936 self
.save_console_title()
939 def __exit__(self
, *args
):
940 self
.restore_console_title()
942 if self
.params
.get('cookiefile') is not None:
943 self
.cookiejar
.save(ignore_discard
=True, ignore_expires
=True)
945 def trouble(self
, message
=None, tb
=None, is_error
=True):
946 """Determine action to take when a download problem appears.
948 Depending on if the downloader has been configured to ignore
949 download errors or not, this method may throw an exception or
950 not when errors are found, after printing the message.
952 @param tb If given, is additional traceback information
953 @param is_error Whether to raise error according to ignorerrors
955 if message
is not None:
956 self
.to_stderr(message
)
957 if self
.params
.get('verbose'):
959 if sys
.exc_info()[0]: # if .trouble has been called from an except block
961 if hasattr(sys
.exc_info()[1], 'exc_info') and sys
.exc_info()[1].exc_info
[0]:
962 tb
+= ''.join(traceback
.format_exception(*sys
.exc_info()[1].exc_info
))
963 tb
+= encode_compat_str(traceback
.format_exc())
965 tb_data
= traceback
.format_list(traceback
.extract_stack())
966 tb
= ''.join(tb_data
)
971 if not self
.params
.get('ignoreerrors'):
972 if sys
.exc_info()[0] and hasattr(sys
.exc_info()[1], 'exc_info') and sys
.exc_info()[1].exc_info
[0]:
973 exc_info
= sys
.exc_info()[1].exc_info
975 exc_info
= sys
.exc_info()
976 raise DownloadError(message
, exc_info
)
977 self
._download
_retcode
= 1
981 EMPHASIS
='light blue',
987 SUPPRESS
='light black',
990 def _format_text(self
, handle
, allow_colors
, text
, f
, fallback
=None, *, test_encoding
=False):
994 # handle.encoding can be None. See https://github.com/yt-dlp/yt-dlp/issues/2711
995 encoding
= self
.params
.get('encoding') or getattr(handle
, 'encoding', None) or 'ascii'
996 text
= text
.encode(encoding
, 'ignore').decode(encoding
)
997 if fallback
is not None and text
!= original_text
:
999 return format_text(text
, f
) if allow_colors
is True else text
if fallback
is None else fallback
1001 def _format_out(self
, *args
, **kwargs
):
1002 return self
._format
_text
(self
._out
_files
.out
, self
._allow
_colors
.out
, *args
, **kwargs
)
1004 def _format_screen(self
, *args
, **kwargs
):
1005 return self
._format
_text
(self
._out
_files
.screen
, self
._allow
_colors
.screen
, *args
, **kwargs
)
1007 def _format_err(self
, *args
, **kwargs
):
1008 return self
._format
_text
(self
._out
_files
.error
, self
._allow
_colors
.error
, *args
, **kwargs
)
1010 def report_warning(self
, message
, only_once
=False):
1012 Print the message to stderr, it will be prefixed with 'WARNING:'
1013 If stderr is a tty file the 'WARNING:' will be colored
1015 if self
.params
.get('logger') is not None:
1016 self
.params
['logger'].warning(message
)
1018 if self
.params
.get('no_warnings'):
1020 self
.to_stderr(f
'{self._format_err("WARNING:", self.Styles.WARNING)} {message}', only_once
)
1022 def deprecation_warning(self
, message
, *, stacklevel
=0):
1023 deprecation_warning(
1024 message
, stacklevel
=stacklevel
+ 1, printer
=self
.report_error
, is_error
=False)
1026 def deprecated_feature(self
, message
):
1027 if self
.params
.get('logger') is not None:
1028 self
.params
['logger'].warning(f
'Deprecated Feature: {message}')
1029 self
.to_stderr(f
'{self._format_err("Deprecated Feature:", self.Styles.ERROR)} {message}', True)
1031 def report_error(self
, message
, *args
, **kwargs
):
1033 Do the same as trouble, but prefixes the message with 'ERROR:', colored
1034 in red if stderr is a tty file.
1036 self
.trouble(f
'{self._format_err("ERROR:", self.Styles.ERROR)} {message}', *args
, **kwargs
)
1038 def write_debug(self
, message
, only_once
=False):
1039 '''Log debug message or Print message to stderr'''
1040 if not self
.params
.get('verbose', False):
1042 message
= f
'[debug] {message}'
1043 if self
.params
.get('logger'):
1044 self
.params
['logger'].debug(message
)
1046 self
.to_stderr(message
, only_once
)
1048 def report_file_already_downloaded(self
, file_name
):
1049 """Report file has already been fully downloaded."""
1051 self
.to_screen('[download] %s has already been downloaded' % file_name
)
1052 except UnicodeEncodeError:
1053 self
.to_screen('[download] The file has already been downloaded')
1055 def report_file_delete(self
, file_name
):
1056 """Report that existing file will be deleted."""
1058 self
.to_screen('Deleting existing file %s' % file_name
)
1059 except UnicodeEncodeError:
1060 self
.to_screen('Deleting existing file')
1062 def raise_no_formats(self
, info
, forced
=False, *, msg
=None):
1063 has_drm
= info
.get('_has_drm')
1064 ignored
, expected
= self
.params
.get('ignore_no_formats_error'), bool(msg
)
1065 msg
= msg
or has_drm
and 'This video is DRM protected' or 'No video formats found!'
1066 if forced
or not ignored
:
1067 raise ExtractorError(msg
, video_id
=info
['id'], ie
=info
['extractor'],
1068 expected
=has_drm
or ignored
or expected
)
1070 self
.report_warning(msg
)
1072 def parse_outtmpl(self
):
1073 self
.deprecation_warning('"YoutubeDL.parse_outtmpl" is deprecated and may be removed in a future version')
1074 self
._parse
_outtmpl
()
1075 return self
.params
['outtmpl']
1077 def _parse_outtmpl(self
):
1079 if self
.params
.get('restrictfilenames'): # Remove spaces in the default template
1080 sanitize
= lambda x
: x
.replace(' - ', ' ').replace(' ', '-')
1082 outtmpl
= self
.params
.setdefault('outtmpl', {})
1083 if not isinstance(outtmpl
, dict):
1084 self
.params
['outtmpl'] = outtmpl
= {'default': outtmpl}
1085 outtmpl
.update({k: sanitize(v) for k, v in DEFAULT_OUTTMPL.items() if outtmpl.get(k) is None}
)
1087 def get_output_path(self
, dir_type
='', filename
=None):
1088 paths
= self
.params
.get('paths', {})
1089 assert isinstance(paths
, dict), '"paths" parameter must be a dictionary'
1090 path
= os
.path
.join(
1091 expand_path(paths
.get('home', '').strip()),
1092 expand_path(paths
.get(dir_type
, '').strip()) if dir_type
else '',
1094 return sanitize_path(path
, force
=self
.params
.get('windowsfilenames'))
1097 def _outtmpl_expandpath(outtmpl
):
1098 # expand_path translates '%%' into '%' and '$$' into '$'
1099 # correspondingly that is not what we want since we need to keep
1100 # '%%' intact for template dict substitution step. Working around
1101 # with boundary-alike separator hack.
1102 sep
= ''.join(random
.choices(string
.ascii_letters
, k
=32))
1103 outtmpl
= outtmpl
.replace('%%', f
'%{sep}%').replace('$$', f
'${sep}$')
1105 # outtmpl should be expand_path'ed before template dict substitution
1106 # because meta fields may contain env variables we don't want to
1107 # be expanded. E.g. for outtmpl "%(title)s.%(ext)s" and
1108 # title "Hello $PATH", we don't want `$PATH` to be expanded.
1109 return expand_path(outtmpl
).replace(sep
, '')
1112 def escape_outtmpl(outtmpl
):
1113 ''' Escape any remaining strings like %s, %abc% etc. '''
1115 STR_FORMAT_RE_TMPL
.format('', '(?![%(\0])'),
1116 lambda mobj
: ('' if mobj
.group('has_key') else '%') + mobj
.group(0),
1120 def validate_outtmpl(cls
, outtmpl
):
1121 ''' @return None or Exception object '''
1123 STR_FORMAT_RE_TMPL
.format('[^)]*', '[ljhqBUDS]'),
1124 lambda mobj
: f
'{mobj.group(0)[:-1]}s',
1125 cls
._outtmpl
_expandpath
(outtmpl
))
1127 cls
.escape_outtmpl(outtmpl
) % collections
.defaultdict(int)
1129 except ValueError as err
:
1133 def _copy_infodict(info_dict
):
1134 info_dict
= dict(info_dict
)
1135 info_dict
.pop('__postprocessors', None)
1136 info_dict
.pop('__pending_error', None)
1139 def prepare_outtmpl(self
, outtmpl
, info_dict
, sanitize
=False):
1140 """ Make the outtmpl and info_dict suitable for substitution: ydl.escape_outtmpl(outtmpl) % info_dict
1141 @param sanitize Whether to sanitize the output as a filename.
1142 For backward compatibility, a function can also be passed
1145 info_dict
.setdefault('epoch', int(time
.time())) # keep epoch consistent once set
1147 info_dict
= self
._copy
_infodict
(info_dict
)
1148 info_dict
['duration_string'] = ( # %(duration>%H-%M-%S)s is wrong if duration > 24hrs
1149 formatSeconds(info_dict
['duration'], '-' if sanitize
else ':')
1150 if info_dict
.get('duration', None) is not None
1152 info_dict
['autonumber'] = int(self
.params
.get('autonumber_start', 1) - 1 + self
._num
_downloads
)
1153 info_dict
['video_autonumber'] = self
._num
_videos
1154 if info_dict
.get('resolution') is None:
1155 info_dict
['resolution'] = self
.format_resolution(info_dict
, default
=None)
1157 # For fields playlist_index, playlist_autonumber and autonumber convert all occurrences
1158 # of %(field)s to %(field)0Nd for backward compatibility
1159 field_size_compat_map
= {
1160 'playlist_index': number_of_digits(info_dict
.get('__last_playlist_index') or 0),
1161 'playlist_autonumber': number_of_digits(info_dict
.get('n_entries') or 0),
1162 'autonumber': self
.params
.get('autonumber_size') or 5,
1166 EXTERNAL_FORMAT_RE
= re
.compile(STR_FORMAT_RE_TMPL
.format('[^)]*', f
'[{STR_FORMAT_TYPES}ljhqBUDS]'))
1171 # Field is of the form key1.key2...
1172 # where keys (except first) can be string, int, slice or "{field, ...}"
1173 FIELD_INNER_RE
= r
'(?:\w+|%(num)s|%(num)s?(?::%(num)s?){1,2})' % {'num': r'(?:-?\d+)'}
1174 FIELD_RE
= r
'\w*(?:\.(?:%(inner)s|{%(field)s(?:,%(field)s)*}))*' % {
1175 'inner': FIELD_INNER_RE
,
1176 'field': rf
'\w*(?:\.{FIELD_INNER_RE})*'
1178 MATH_FIELD_RE
= rf
'(?:{FIELD_RE}|-?{NUMBER_RE})'
1179 MATH_OPERATORS_RE
= r
'(?:%s)' % '|'.join(map(re
.escape
, MATH_FUNCTIONS
.keys()))
1180 INTERNAL_FORMAT_RE
= re
.compile(rf
'''(?xs)
1182 (?P<fields>{FIELD_RE})
1183 (?P<maths>(?:{MATH_OPERATORS_RE}{MATH_FIELD_RE})*)
1184 (?:>(?P<strf_format>.+?))?
1186 (?P<alternate>(?<!\\),[^|&)]+)?
1187 (?:&(?P<replacement>.*?))?
1188 (?:\|(?P<default>.*?))?
1191 def _traverse_infodict(fields
):
1192 fields
= [f
for x
in re
.split(r
'\.({.+?})\.?', fields
)
1193 for f
in ([x
] if x
.startswith('{') else x
.split('.'))]
1195 if fields
and not fields
[i
]:
1198 for i
, f
in enumerate(fields
):
1199 if not f
.startswith('{'):
1201 assert f
.endswith('}'), f
'No closing brace for {f} in {fields}'
1202 fields
[i
] = {k: k.split('.') for k in f[1:-1].split(',')}
1204 return traverse_obj(info_dict
, fields
, is_user_input
=True, traverse_string
=True)
1206 def get_value(mdict
):
1208 value
= _traverse_infodict(mdict
['fields'])
1211 value
= float_or_none(value
)
1212 if value
is not None:
1215 offset_key
= mdict
['maths']
1217 value
= float_or_none(value
)
1221 MATH_FIELD_RE
if operator
else MATH_OPERATORS_RE
,
1222 offset_key
).group(0)
1223 offset_key
= offset_key
[len(item
):]
1224 if operator
is None:
1225 operator
= MATH_FUNCTIONS
[item
]
1227 item
, multiplier
= (item
[1:], -1) if item
[0] == '-' else (item
, 1)
1228 offset
= float_or_none(item
)
1230 offset
= float_or_none(_traverse_infodict(item
))
1232 value
= operator(value
, multiplier
* offset
)
1233 except (TypeError, ZeroDivisionError):
1236 # Datetime formatting
1237 if mdict
['strf_format']:
1238 value
= strftime_or_none(value
, mdict
['strf_format'].replace('\\,', ','))
1240 # XXX: Workaround for https://github.com/yt-dlp/yt-dlp/issues/4485
1241 if sanitize
and value
== '':
1245 na
= self
.params
.get('outtmpl_na_placeholder', 'NA')
1247 def filename_sanitizer(key
, value
, restricted
=self
.params
.get('restrictfilenames')):
1248 return sanitize_filename(str(value
), restricted
=restricted
, is_id
=(
1249 bool(re
.search(r
'(^|[_.])id(\.|$)', key
))
1250 if 'filename-sanitization' in self
.params
['compat_opts']
1253 sanitizer
= sanitize
if callable(sanitize
) else filename_sanitizer
1254 sanitize
= bool(sanitize
)
1256 def _dumpjson_default(obj
):
1257 if isinstance(obj
, (set, LazyList
)):
1261 class _ReplacementFormatter(string
.Formatter
):
1262 def get_field(self
, field_name
, args
, kwargs
):
1263 if field_name
.isdigit():
1265 raise ValueError('Unsupported field')
1267 replacement_formatter
= _ReplacementFormatter()
1269 def create_key(outer_mobj
):
1270 if not outer_mobj
.group('has_key'):
1271 return outer_mobj
.group(0)
1272 key
= outer_mobj
.group('key')
1273 mobj
= re
.match(INTERNAL_FORMAT_RE
, key
)
1274 initial_field
= mobj
.group('fields') if mobj
else ''
1275 value
, replacement
, default
= None, None, na
1277 mobj
= mobj
.groupdict()
1278 default
= mobj
['default'] if mobj
['default'] is not None else default
1279 value
= get_value(mobj
)
1280 replacement
= mobj
['replacement']
1281 if value
is None and mobj
['alternate']:
1282 mobj
= re
.match(INTERNAL_FORMAT_RE
, mobj
['remaining'][1:])
1286 fmt
= outer_mobj
.group('format')
1287 if fmt
== 's' and value
is not None and key
in field_size_compat_map
.keys():
1288 fmt
= f
'0{field_size_compat_map[key]:d}d'
1290 if None not in (value
, replacement
):
1292 value
= replacement_formatter
.format(replacement
, value
)
1294 value
, default
= None, na
1296 flags
= outer_mobj
.group('conversion') or ''
1297 str_fmt
= f
'{fmt[:-1]}s'
1299 value
, fmt
= default
, 's'
1300 elif fmt
[-1] == 'l': # list
1301 delim
= '\n' if '#' in flags
else ', '
1302 value
, fmt
= delim
.join(map(str, variadic(value
, allowed_types
=(str, bytes)))), str_fmt
1303 elif fmt
[-1] == 'j': # json
1304 value
, fmt
= json
.dumps(
1305 value
, default
=_dumpjson_default
,
1306 indent
=4 if '#' in flags
else None, ensure_ascii
='+' not in flags
), str_fmt
1307 elif fmt
[-1] == 'h': # html
1308 value
, fmt
= escapeHTML(str(value
)), str_fmt
1309 elif fmt
[-1] == 'q': # quoted
1310 value
= map(str, variadic(value
) if '#' in flags
else [value
])
1311 value
, fmt
= ' '.join(map(compat_shlex_quote
, value
)), str_fmt
1312 elif fmt
[-1] == 'B': # bytes
1313 value
= f
'%{str_fmt}'.encode() % str(value
).encode()
1314 value
, fmt
= value
.decode('utf-8', 'ignore'), 's'
1315 elif fmt
[-1] == 'U': # unicode normalized
1316 value
, fmt
= unicodedata
.normalize(
1317 # "+" = compatibility equivalence, "#" = NFD
1318 'NF%s%s' % ('K' if '+' in flags
else '', 'D' if '#' in flags
else 'C'),
1320 elif fmt
[-1] == 'D': # decimal suffix
1321 num_fmt
, fmt
= fmt
[:-1].replace('#', ''), 's'
1322 value
= format_decimal_suffix(value
, f
'%{num_fmt}f%s' if num_fmt
else '%d%s',
1323 factor
=1024 if '#' in flags
else 1000)
1324 elif fmt
[-1] == 'S': # filename sanitization
1325 value
, fmt
= filename_sanitizer(initial_field
, value
, restricted
='#' in flags
), str_fmt
1326 elif fmt
[-1] == 'c':
1328 value
= str(value
)[0]
1331 elif fmt
[-1] not in 'rsa': # numeric
1332 value
= float_or_none(value
)
1334 value
, fmt
= default
, 's'
1337 # If value is an object, sanitize might convert it to a string
1338 # So we convert it to repr first
1340 value
, fmt
= repr(value
), str_fmt
1341 elif fmt
[-1] == 'a':
1342 value
, fmt
= ascii(value
), str_fmt
1343 if fmt
[-1] in 'csra':
1344 value
= sanitizer(initial_field
, value
)
1346 key
= '%s\0%s' % (key
.replace('%', '%\0'), outer_mobj
.group('format'))
1347 TMPL_DICT
[key
] = value
1348 return '{prefix}%({key}){fmt}'.format(key
=key
, fmt
=fmt
, prefix
=outer_mobj
.group('prefix'))
1350 return EXTERNAL_FORMAT_RE
.sub(create_key
, outtmpl
), TMPL_DICT
1352 def evaluate_outtmpl(self
, outtmpl
, info_dict
, *args
, **kwargs
):
1353 outtmpl
, info_dict
= self
.prepare_outtmpl(outtmpl
, info_dict
, *args
, **kwargs
)
1354 return self
.escape_outtmpl(outtmpl
) % info_dict
1356 def _prepare_filename(self
, info_dict
, *, outtmpl
=None, tmpl_type
=None):
1357 assert None in (outtmpl
, tmpl_type
), 'outtmpl and tmpl_type are mutually exclusive'
1359 outtmpl
= self
.params
['outtmpl'].get(tmpl_type
or 'default', self
.params
['outtmpl']['default'])
1361 outtmpl
= self
._outtmpl
_expandpath
(outtmpl
)
1362 filename
= self
.evaluate_outtmpl(outtmpl
, info_dict
, True)
1366 if tmpl_type
in ('', 'temp'):
1367 final_ext
, ext
= self
.params
.get('final_ext'), info_dict
.get('ext')
1368 if final_ext
and ext
and final_ext
!= ext
and filename
.endswith(f
'.{final_ext}'):
1369 filename
= replace_extension(filename
, ext
, final_ext
)
1371 force_ext
= OUTTMPL_TYPES
[tmpl_type
]
1373 filename
= replace_extension(filename
, force_ext
, info_dict
.get('ext'))
1375 # https://github.com/blackjack4494/youtube-dlc/issues/85
1376 trim_file_name
= self
.params
.get('trim_file_name', False)
1378 no_ext
, *ext
= filename
.rsplit('.', 2)
1379 filename
= join_nonempty(no_ext
[:trim_file_name
], *ext
, delim
='.')
1382 except ValueError as err
:
1383 self
.report_error('Error in output template: ' + str(err
) + ' (encoding: ' + repr(preferredencoding()) + ')')
1386 def prepare_filename(self
, info_dict
, dir_type
='', *, outtmpl
=None, warn
=False):
1387 """Generate the output filename"""
1389 assert not dir_type
, 'outtmpl and dir_type are mutually exclusive'
1391 filename
= self
._prepare
_filename
(info_dict
, tmpl_type
=dir_type
, outtmpl
=outtmpl
)
1392 if not filename
and dir_type
not in ('', 'temp'):
1396 if not self
.params
.get('paths'):
1398 elif filename
== '-':
1399 self
.report_warning('--paths is ignored when an outputting to stdout', only_once
=True)
1400 elif os
.path
.isabs(filename
):
1401 self
.report_warning('--paths is ignored since an absolute path is given in output template', only_once
=True)
1402 if filename
== '-' or not filename
:
1405 return self
.get_output_path(dir_type
, filename
)
1407 def _match_entry(self
, info_dict
, incomplete
=False, silent
=False):
1408 """Returns None if the file should be downloaded"""
1409 _type
= 'video' if 'playlist-match-filter' in self
.params
['compat_opts'] else info_dict
.get('_type', 'video')
1410 assert incomplete
or _type
== 'video', 'Only video result can be considered complete'
1412 video_title
= info_dict
.get('title', info_dict
.get('id', 'entry'))
1415 if _type
in ('playlist', 'multi_video'):
1417 elif _type
in ('url', 'url_transparent') and not try_call(
1418 lambda: self
.get_info_extractor(info_dict
['ie_key']).is_single_video(info_dict
['url'])):
1421 if 'title' in info_dict
:
1422 # This can happen when we're just evaluating the playlist
1423 title
= info_dict
['title']
1424 matchtitle
= self
.params
.get('matchtitle', False)
1426 if not re
.search(matchtitle
, title
, re
.IGNORECASE
):
1427 return '"' + title
+ '" title did not match pattern "' + matchtitle
+ '"'
1428 rejecttitle
= self
.params
.get('rejecttitle', False)
1430 if re
.search(rejecttitle
, title
, re
.IGNORECASE
):
1431 return '"' + title
+ '" title matched reject pattern "' + rejecttitle
+ '"'
1433 date
= info_dict
.get('upload_date')
1434 if date
is not None:
1435 dateRange
= self
.params
.get('daterange', DateRange())
1436 if date
not in dateRange
:
1437 return f
'{date_from_str(date).isoformat()} upload date is not in range {dateRange}'
1438 view_count
= info_dict
.get('view_count')
1439 if view_count
is not None:
1440 min_views
= self
.params
.get('min_views')
1441 if min_views
is not None and view_count
< min_views
:
1442 return 'Skipping %s, because it has not reached minimum view count (%d/%d)' % (video_title
, view_count
, min_views
)
1443 max_views
= self
.params
.get('max_views')
1444 if max_views
is not None and view_count
> max_views
:
1445 return 'Skipping %s, because it has exceeded the maximum view count (%d/%d)' % (video_title
, view_count
, max_views
)
1446 if age_restricted(info_dict
.get('age_limit'), self
.params
.get('age_limit')):
1447 return 'Skipping "%s" because it is age restricted' % video_title
1449 match_filter
= self
.params
.get('match_filter')
1450 if match_filter
is None:
1456 ret
= match_filter(info_dict
, incomplete
=incomplete
)
1458 # For backward compatibility
1459 ret
= None if incomplete
else match_filter(info_dict
)
1460 except DownloadCancelled
as err
:
1461 if err
.msg
is not NO_DEFAULT
:
1463 ret
, cancelled
= err
.msg
, err
1465 if ret
is NO_DEFAULT
:
1467 filename
= self
._format
_screen
(self
.prepare_filename(info_dict
), self
.Styles
.FILENAME
)
1468 reply
= input(self
._format
_screen
(
1469 f
'Download "{filename}"? (Y/n): ', self
.Styles
.EMPHASIS
)).lower().strip()
1470 if reply
in {'y', ''}
:
1474 raise type(cancelled
)(f
'Skipping {video_title}')
1475 return f
'Skipping {video_title}'
1478 if self
.in_download_archive(info_dict
):
1479 reason
= '%s has already been recorded in the archive' % video_title
1480 break_opt
, break_err
= 'break_on_existing', ExistingVideoReached
1483 reason
= check_filter()
1484 except DownloadCancelled
as e
:
1485 reason
, break_opt
, break_err
= e
.msg
, 'match_filter', type(e
)
1487 break_opt
, break_err
= 'break_on_reject', RejectedVideoReached
1488 if reason
is not None:
1490 self
.to_screen('[download] ' + reason
)
1491 if self
.params
.get(break_opt
, False):
1496 def add_extra_info(info_dict
, extra_info
):
1497 '''Set the keys from extra_info in info dict if they are missing'''
1498 for key
, value
in extra_info
.items():
1499 info_dict
.setdefault(key
, value
)
1501 def extract_info(self
, url
, download
=True, ie_key
=None, extra_info
=None,
1502 process
=True, force_generic_extractor
=False):
1504 Extract and return the information dictionary of the URL
1507 @param url URL to extract
1510 @param download Whether to download videos
1511 @param process Whether to resolve all unresolved references (URLs, playlist items).
1512 Must be True for download to work
1513 @param ie_key Use only the extractor with this key
1515 @param extra_info Dictionary containing the extra values to add to the info (For internal use only)
1516 @force_generic_extractor Force using the generic extractor (Deprecated; use ie_key='Generic')
1519 if extra_info
is None:
1522 if not ie_key
and force_generic_extractor
:
1526 ies
= {ie_key: self._ies[ie_key]}
if ie_key
in self
._ies
else {}
1530 for key
, ie
in ies
.items():
1531 if not ie
.suitable(url
):
1534 if not ie
.working():
1535 self
.report_warning('The program functionality for this site has been marked as broken, '
1536 'and will probably not work.')
1538 temp_id
= ie
.get_temp_id(url
)
1539 if temp_id
is not None and self
.in_download_archive({'id': temp_id, 'ie_key': key}
):
1540 self
.to_screen(f
'[{key}] {temp_id}: has already been recorded in the archive')
1541 if self
.params
.get('break_on_existing', False):
1542 raise ExistingVideoReached()
1544 return self
.__extract
_info
(url
, self
.get_info_extractor(key
), download
, extra_info
, process
)
1546 extractors_restricted
= self
.params
.get('allowed_extractors') not in (None, ['default'])
1547 self
.report_error(f
'No suitable extractor{format_field(ie_key, None, " (%s)")} found for URL {url}',
1548 tb
=False if extractors_restricted
else None)
1550 def _handle_extraction_exceptions(func
):
1551 @functools.wraps(func
)
1552 def wrapper(self
, *args
, **kwargs
):
1555 return func(self
, *args
, **kwargs
)
1556 except (DownloadCancelled
, LazyList
.IndexError, PagedList
.IndexError):
1558 except ReExtractInfo
as e
:
1560 self
.to_screen(f
'{e}; Re-extracting data')
1562 self
.to_stderr('\r')
1563 self
.report_warning(f
'{e}; Re-extracting data')
1565 except GeoRestrictedError
as e
:
1568 msg
+= '\nThis video is available in %s.' % ', '.join(
1569 map(ISO3166Utils
.short2full
, e
.countries
))
1570 msg
+= '\nYou might want to use a VPN or a proxy server (with --proxy) to workaround.'
1571 self
.report_error(msg
)
1572 except ExtractorError
as e
: # An error we somewhat expected
1573 self
.report_error(str(e
), e
.format_traceback())
1574 except Exception as e
:
1575 if self
.params
.get('ignoreerrors'):
1576 self
.report_error(str(e
), tb
=encode_compat_str(traceback
.format_exc()))
1582 def _wait_for_video(self
, ie_result
={}):
1583 if (not self
.params
.get('wait_for_video')
1584 or ie_result
.get('_type', 'video') != 'video'
1585 or ie_result
.get('formats') or ie_result
.get('url')):
1588 format_dur
= lambda dur
: '%02d:%02d:%02d' % timetuple_from_msec(dur
* 1000)[:-1]
1593 full_msg
= f
'{msg}\n'
1594 if not self
.params
.get('noprogress'):
1595 full_msg
= msg
+ ' ' * (len(last_msg
) - len(msg
)) + '\r'
1598 self
.to_screen(full_msg
, skip_eol
=True)
1601 min_wait
, max_wait
= self
.params
.get('wait_for_video')
1602 diff
= try_get(ie_result
, lambda x
: x
['release_timestamp'] - time
.time())
1603 if diff
is None and ie_result
.get('live_status') == 'is_upcoming':
1604 diff
= round(random
.uniform(min_wait
, max_wait
) if (max_wait
and min_wait
) else (max_wait
or min_wait
), 0)
1605 self
.report_warning('Release time of video is not known')
1606 elif ie_result
and (diff
or 0) <= 0:
1607 self
.report_warning('Video should already be available according to extracted info')
1608 diff
= min(max(diff
or 0, min_wait
or 0), max_wait
or float('inf'))
1609 self
.to_screen(f
'[wait] Waiting for {format_dur(diff)} - Press Ctrl+C to try now')
1611 wait_till
= time
.time() + diff
1614 diff
= wait_till
- time
.time()
1617 raise ReExtractInfo('[wait] Wait period ended', expected
=True)
1618 progress(f
'[wait] Remaining time until next attempt: {self._format_screen(format_dur(diff), self.Styles.EMPHASIS)}')
1620 except KeyboardInterrupt:
1622 raise ReExtractInfo('[wait] Interrupted by user', expected
=True)
1623 except BaseException
as e
:
1624 if not isinstance(e
, ReExtractInfo
):
1628 @_handle_extraction_exceptions
1629 def __extract_info(self
, url
, ie
, download
, extra_info
, process
):
1631 ie_result
= ie
.extract(url
)
1632 except UserNotLive
as e
:
1634 if self
.params
.get('wait_for_video'):
1635 self
.report_warning(e
)
1636 self
._wait
_for
_video
()
1638 if ie_result
is None: # Finished already (backwards compatibility; listformats and friends should be moved here)
1639 self
.report_warning(f
'Extractor {ie.IE_NAME} returned nothing{bug_reports_message()}')
1641 if isinstance(ie_result
, list):
1642 # Backwards compatibility: old IE result format
1644 '_type': 'compat_list',
1645 'entries': ie_result
,
1647 if extra_info
.get('original_url'):
1648 ie_result
.setdefault('original_url', extra_info
['original_url'])
1649 self
.add_default_extra_info(ie_result
, ie
, url
)
1651 self
._wait
_for
_video
(ie_result
)
1652 return self
.process_ie_result(ie_result
, download
, extra_info
)
1656 def add_default_extra_info(self
, ie_result
, ie
, url
):
1658 self
.add_extra_info(ie_result
, {
1660 'original_url': url
,
1662 webpage_url
= ie_result
.get('webpage_url')
1664 self
.add_extra_info(ie_result
, {
1665 'webpage_url_basename': url_basename(webpage_url
),
1666 'webpage_url_domain': get_domain(webpage_url
),
1669 self
.add_extra_info(ie_result
, {
1670 'extractor': ie
.IE_NAME
,
1671 'extractor_key': ie
.ie_key(),
1674 def process_ie_result(self
, ie_result
, download
=True, extra_info
=None):
1676 Take the result of the ie(may be modified) and resolve all unresolved
1677 references (URLs, playlist items).
1679 It will also download the videos if 'download'.
1680 Returns the resolved ie_result.
1682 if extra_info
is None:
1684 result_type
= ie_result
.get('_type', 'video')
1686 if result_type
in ('url', 'url_transparent'):
1687 ie_result
['url'] = sanitize_url(
1688 ie_result
['url'], scheme
='http' if self
.params
.get('prefer_insecure') else 'https')
1689 if ie_result
.get('original_url') and not extra_info
.get('original_url'):
1690 extra_info
= {'original_url': ie_result['original_url'], **extra_info}
1692 extract_flat
= self
.params
.get('extract_flat', False)
1693 if ((extract_flat
== 'in_playlist' and 'playlist' in extra_info
)
1694 or extract_flat
is True):
1695 info_copy
= ie_result
.copy()
1696 ie
= try_get(ie_result
.get('ie_key'), self
.get_info_extractor
)
1697 if ie
and not ie_result
.get('id'):
1698 info_copy
['id'] = ie
.get_temp_id(ie_result
['url'])
1699 self
.add_default_extra_info(info_copy
, ie
, ie_result
['url'])
1700 self
.add_extra_info(info_copy
, extra_info
)
1701 info_copy
, _
= self
.pre_process(info_copy
)
1702 self
._fill
_common
_fields
(info_copy
, False)
1703 self
.__forced
_printings
(info_copy
)
1704 self
._raise
_pending
_errors
(info_copy
)
1705 if self
.params
.get('force_write_download_archive', False):
1706 self
.record_download_archive(info_copy
)
1709 if result_type
== 'video':
1710 self
.add_extra_info(ie_result
, extra_info
)
1711 ie_result
= self
.process_video_result(ie_result
, download
=download
)
1712 self
._raise
_pending
_errors
(ie_result
)
1713 additional_urls
= (ie_result
or {}).get('additional_urls')
1715 # TODO: Improve MetadataParserPP to allow setting a list
1716 if isinstance(additional_urls
, str):
1717 additional_urls
= [additional_urls
]
1719 '[info] %s: %d additional URL(s) requested' % (ie_result
['id'], len(additional_urls
)))
1720 self
.write_debug('Additional URLs: "%s"' % '", "'.join(additional_urls
))
1721 ie_result
['additional_entries'] = [
1723 url
, download
, extra_info
=extra_info
,
1724 force_generic_extractor
=self
.params
.get('force_generic_extractor'))
1725 for url
in additional_urls
1728 elif result_type
== 'url':
1729 # We have to add extra_info to the results because it may be
1730 # contained in a playlist
1731 return self
.extract_info(
1732 ie_result
['url'], download
,
1733 ie_key
=ie_result
.get('ie_key'),
1734 extra_info
=extra_info
)
1735 elif result_type
== 'url_transparent':
1736 # Use the information from the embedding page
1737 info
= self
.extract_info(
1738 ie_result
['url'], ie_key
=ie_result
.get('ie_key'),
1739 extra_info
=extra_info
, download
=False, process
=False)
1741 # extract_info may return None when ignoreerrors is enabled and
1742 # extraction failed with an error, don't crash and return early
1747 exempted_fields
= {'_type', 'url', 'ie_key'}
1748 if not ie_result
.get('section_end') and ie_result
.get('section_start') is None:
1749 # For video clips, the id etc of the clip extractor should be used
1750 exempted_fields |
= {'id', 'extractor', 'extractor_key'}
1752 new_result
= info
.copy()
1753 new_result
.update(filter_dict(ie_result
, lambda k
, v
: v
is not None and k
not in exempted_fields
))
1755 # Extracted info may not be a video result (i.e.
1756 # info.get('_type', 'video') != video) but rather an url or
1757 # url_transparent. In such cases outer metadata (from ie_result)
1758 # should be propagated to inner one (info). For this to happen
1759 # _type of info should be overridden with url_transparent. This
1760 # fixes issue from https://github.com/ytdl-org/youtube-dl/pull/11163.
1761 if new_result
.get('_type') == 'url':
1762 new_result
['_type'] = 'url_transparent'
1764 return self
.process_ie_result(
1765 new_result
, download
=download
, extra_info
=extra_info
)
1766 elif result_type
in ('playlist', 'multi_video'):
1767 # Protect from infinite recursion due to recursively nested playlists
1768 # (see https://github.com/ytdl-org/youtube-dl/issues/27833)
1769 webpage_url
= ie_result
.get('webpage_url') # Playlists maynot have webpage_url
1770 if webpage_url
and webpage_url
in self
._playlist
_urls
:
1772 '[download] Skipping already downloaded playlist: %s'
1773 % ie_result
.get('title') or ie_result
.get('id'))
1776 self
._playlist
_level
+= 1
1777 self
._playlist
_urls
.add(webpage_url
)
1778 self
._fill
_common
_fields
(ie_result
, False)
1779 self
._sanitize
_thumbnails
(ie_result
)
1781 return self
.__process
_playlist
(ie_result
, download
)
1783 self
._playlist
_level
-= 1
1784 if not self
._playlist
_level
:
1785 self
._playlist
_urls
.clear()
1786 elif result_type
== 'compat_list':
1787 self
.report_warning(
1788 'Extractor %s returned a compat_list result. '
1789 'It needs to be updated.' % ie_result
.get('extractor'))
1792 self
.add_extra_info(r
, {
1793 'extractor': ie_result
['extractor'],
1794 'webpage_url': ie_result
['webpage_url'],
1795 'webpage_url_basename': url_basename(ie_result
['webpage_url']),
1796 'webpage_url_domain': get_domain(ie_result
['webpage_url']),
1797 'extractor_key': ie_result
['extractor_key'],
1800 ie_result
['entries'] = [
1801 self
.process_ie_result(_fixup(r
), download
, extra_info
)
1802 for r
in ie_result
['entries']
1806 raise Exception('Invalid result type: %s' % result_type
)
1808 def _ensure_dir_exists(self
, path
):
1809 return make_dir(path
, self
.report_error
)
1812 def _playlist_infodict(ie_result
, strict
=False, **kwargs
):
1814 'playlist_count': ie_result
.get('playlist_count'),
1815 'playlist': ie_result
.get('title') or ie_result
.get('id'),
1816 'playlist_id': ie_result
.get('id'),
1817 'playlist_title': ie_result
.get('title'),
1818 'playlist_uploader': ie_result
.get('uploader'),
1819 'playlist_uploader_id': ie_result
.get('uploader_id'),
1824 if ie_result
.get('webpage_url'):
1826 'webpage_url': ie_result
['webpage_url'],
1827 'webpage_url_basename': url_basename(ie_result
['webpage_url']),
1828 'webpage_url_domain': get_domain(ie_result
['webpage_url']),
1832 'playlist_index': 0,
1833 '__last_playlist_index': max(ie_result
.get('requested_entries') or (0, 0)),
1834 'extractor': ie_result
['extractor'],
1835 'extractor_key': ie_result
['extractor_key'],
1838 def __process_playlist(self
, ie_result
, download
):
1839 """Process each entry in the playlist"""
1840 assert ie_result
['_type'] in ('playlist', 'multi_video')
1842 common_info
= self
._playlist
_infodict
(ie_result
, strict
=True)
1843 title
= common_info
.get('playlist') or '<Untitled>'
1844 if self
._match
_entry
(common_info
, incomplete
=True) is not None:
1846 self
.to_screen(f
'[download] Downloading {ie_result["_type"]}: {title}')
1848 all_entries
= PlaylistEntries(self
, ie_result
)
1849 entries
= orderedSet(all_entries
.get_requested_items(), lazy
=True)
1851 lazy
= self
.params
.get('lazy_playlist')
1853 resolved_entries
, n_entries
= [], 'N/A'
1854 ie_result
['requested_entries'], ie_result
['entries'] = None, None
1856 entries
= resolved_entries
= list(entries
)
1857 n_entries
= len(resolved_entries
)
1858 ie_result
['requested_entries'], ie_result
['entries'] = tuple(zip(*resolved_entries
)) or ([], [])
1859 if not ie_result
.get('playlist_count'):
1860 # Better to do this after potentially exhausting entries
1861 ie_result
['playlist_count'] = all_entries
.get_full_count()
1863 extra
= self
._playlist
_infodict
(ie_result
, n_entries
=int_or_none(n_entries
))
1864 ie_copy
= collections
.ChainMap(ie_result
, extra
)
1866 _infojson_written
= False
1867 write_playlist_files
= self
.params
.get('allow_playlist_files', True)
1868 if write_playlist_files
and self
.params
.get('list_thumbnails'):
1869 self
.list_thumbnails(ie_result
)
1870 if write_playlist_files
and not self
.params
.get('simulate'):
1871 _infojson_written
= self
._write
_info
_json
(
1872 'playlist', ie_result
, self
.prepare_filename(ie_copy
, 'pl_infojson'))
1873 if _infojson_written
is None:
1875 if self
._write
_description
('playlist', ie_result
,
1876 self
.prepare_filename(ie_copy
, 'pl_description')) is None:
1878 # TODO: This should be passed to ThumbnailsConvertor if necessary
1879 self
._write
_thumbnails
('playlist', ie_result
, self
.prepare_filename(ie_copy
, 'pl_thumbnail'))
1882 if self
.params
.get('playlistreverse') or self
.params
.get('playlistrandom'):
1883 self
.report_warning('playlistreverse and playlistrandom are not supported with lazy_playlist', only_once
=True)
1884 elif self
.params
.get('playlistreverse'):
1886 elif self
.params
.get('playlistrandom'):
1887 random
.shuffle(entries
)
1889 self
.to_screen(f
'[{ie_result["extractor"]}] Playlist {title}: Downloading {n_entries} items'
1890 f
'{format_field(ie_result, "playlist_count", " of %s")}')
1892 keep_resolved_entries
= self
.params
.get('extract_flat') != 'discard'
1893 if self
.params
.get('extract_flat') == 'discard_in_playlist':
1894 keep_resolved_entries
= ie_result
['_type'] != 'playlist'
1895 if keep_resolved_entries
:
1896 self
.write_debug('The information of all playlist entries will be held in memory')
1899 max_failures
= self
.params
.get('skip_playlist_after_errors') or float('inf')
1900 for i
, (playlist_index
, entry
) in enumerate(entries
):
1902 resolved_entries
.append((playlist_index
, entry
))
1906 entry
['__x_forwarded_for_ip'] = ie_result
.get('__x_forwarded_for_ip')
1907 if not lazy
and 'playlist-index' in self
.params
['compat_opts']:
1908 playlist_index
= ie_result
['requested_entries'][i
]
1910 entry_copy
= collections
.ChainMap(entry
, {
1912 'n_entries': int_or_none(n_entries
),
1913 'playlist_index': playlist_index
,
1914 'playlist_autonumber': i
+ 1,
1917 if self
._match
_entry
(entry_copy
, incomplete
=True) is not None:
1918 # For compatabilty with youtube-dl. See https://github.com/yt-dlp/yt-dlp/issues/4369
1919 resolved_entries
[i
] = (playlist_index
, NO_DEFAULT
)
1922 self
.to_screen('[download] Downloading item %s of %s' % (
1923 self
._format
_screen
(i
+ 1, self
.Styles
.ID
), self
._format
_screen
(n_entries
, self
.Styles
.EMPHASIS
)))
1925 entry_result
= self
.__process
_iterable
_entry
(entry
, download
, collections
.ChainMap({
1926 'playlist_index': playlist_index
,
1927 'playlist_autonumber': i
+ 1,
1929 if not entry_result
:
1931 if failures
>= max_failures
:
1933 f
'Skipping the remaining entries in playlist "{title}" since {failures} items failed extraction')
1935 if keep_resolved_entries
:
1936 resolved_entries
[i
] = (playlist_index
, entry_result
)
1938 # Update with processed data
1939 ie_result
['entries'] = [e
for _
, e
in resolved_entries
if e
is not NO_DEFAULT
]
1940 ie_result
['requested_entries'] = [i
for i
, e
in resolved_entries
if e
is not NO_DEFAULT
]
1941 if ie_result
['requested_entries'] == try_call(lambda: list(range(1, ie_result
['playlist_count'] + 1))):
1942 # Do not set for full playlist
1943 ie_result
.pop('requested_entries')
1945 # Write the updated info to json
1946 if _infojson_written
is True and self
._write
_info
_json
(
1947 'updated playlist', ie_result
,
1948 self
.prepare_filename(ie_copy
, 'pl_infojson'), overwrite
=True) is None:
1951 ie_result
= self
.run_all_pps('playlist', ie_result
)
1952 self
.to_screen(f
'[download] Finished downloading playlist: {title}')
1955 @_handle_extraction_exceptions
1956 def __process_iterable_entry(self
, entry
, download
, extra_info
):
1957 return self
.process_ie_result(
1958 entry
, download
=download
, extra_info
=extra_info
)
1960 def _build_format_filter(self
, filter_spec
):
1961 " Returns a function to filter the formats according to the filter_spec "
1971 operator_rex
= re
.compile(r
'''(?x)\s*
1973 (?P<op>%s)(?P<none_inclusive>\s*\?)?\s*
1974 (?P<value>[0-9.]+(?:[kKmMgGtTpPeEzZyY]i?[Bb]?)?)\s*
1975 ''' % '|'.join(map(re
.escape
, OPERATORS
.keys())))
1976 m
= operator_rex
.fullmatch(filter_spec
)
1979 comparison_value
= int(m
.group('value'))
1981 comparison_value
= parse_filesize(m
.group('value'))
1982 if comparison_value
is None:
1983 comparison_value
= parse_filesize(m
.group('value') + 'B')
1984 if comparison_value
is None:
1986 'Invalid value %r in format specification %r' % (
1987 m
.group('value'), filter_spec
))
1988 op
= OPERATORS
[m
.group('op')]
1993 '^=': lambda attr
, value
: attr
.startswith(value
),
1994 '$=': lambda attr
, value
: attr
.endswith(value
),
1995 '*=': lambda attr
, value
: value
in attr
,
1996 '~=': lambda attr
, value
: value
.search(attr
) is not None
1998 str_operator_rex
= re
.compile(r
'''(?x)\s*
1999 (?P<key>[a-zA-Z0-9._-]+)\s*
2000 (?P<negation>!\s*)?(?P<op>%s)\s*(?P<none_inclusive>\?\s*)?
2002 (?P<value>(?(quote)(?:(?!(?P=quote))[^\\]|\\.)+|[\w.-]+))
2003 (?(quote)(?P=quote))\s*
2004 ''' % '|'.join(map(re
.escape
, STR_OPERATORS
.keys())))
2005 m
= str_operator_rex
.fullmatch(filter_spec
)
2007 if m
.group('op') == '~=':
2008 comparison_value
= re
.compile(m
.group('value'))
2010 comparison_value
= re
.sub(r
'''\\([\\"'])''', r
'\1', m
.group('value'))
2011 str_op
= STR_OPERATORS
[m
.group('op')]
2012 if m
.group('negation'):
2013 op
= lambda attr
, value
: not str_op(attr
, value
)
2018 raise SyntaxError('Invalid filter specification %r' % filter_spec
)
2021 actual_value
= f
.get(m
.group('key'))
2022 if actual_value
is None:
2023 return m
.group('none_inclusive')
2024 return op(actual_value
, comparison_value
)
2027 def _check_formats(self
, formats
):
2029 self
.to_screen('[info] Testing format %s' % f
['format_id'])
2030 path
= self
.get_output_path('temp')
2031 if not self
._ensure
_dir
_exists
(f
'{path}/'):
2033 temp_file
= tempfile
.NamedTemporaryFile(suffix
='.tmp', delete
=False, dir=path
or None)
2036 success
, _
= self
.dl(temp_file
.name
, f
, test
=True)
2037 except (DownloadError
, OSError, ValueError) + network_exceptions
:
2040 if os
.path
.exists(temp_file
.name
):
2042 os
.remove(temp_file
.name
)
2044 self
.report_warning('Unable to delete temporary file "%s"' % temp_file
.name
)
2048 self
.to_screen('[info] Unable to download format %s. Skipping...' % f
['format_id'])
2050 def _default_format_spec(self
, info_dict
, download
=True):
2053 merger
= FFmpegMergerPP(self
)
2054 return merger
.available
and merger
.can_merge()
2057 not self
.params
.get('simulate')
2061 or info_dict
.get('is_live') and not self
.params
.get('live_from_start')
2062 or self
.params
['outtmpl']['default'] == '-'))
2065 or self
.params
.get('allow_multiple_audio_streams', False)
2066 or 'format-spec' in self
.params
['compat_opts'])
2069 'best/bestvideo+bestaudio' if prefer_best
2070 else 'bestvideo*+bestaudio/best' if not compat
2071 else 'bestvideo+bestaudio/best')
2073 def build_format_selector(self
, format_spec
):
2074 def syntax_error(note
, start
):
2076 'Invalid format specification: '
2077 '{}\n\t{}\n\t{}^'.format(note
, format_spec
, ' ' * start
[1]))
2078 return SyntaxError(message
)
2080 PICKFIRST
= 'PICKFIRST'
2084 FormatSelector
= collections
.namedtuple('FormatSelector', ['type', 'selector', 'filters'])
2086 allow_multiple_streams
= {'audio': self
.params
.get('allow_multiple_audio_streams', False),
2087 'video': self
.params
.get('allow_multiple_video_streams', False)}
2089 check_formats
= self
.params
.get('check_formats') == 'selected'
2091 def _parse_filter(tokens
):
2093 for type, string_
, start
, _
, _
in tokens
:
2094 if type == tokenize
.OP
and string_
== ']':
2095 return ''.join(filter_parts
)
2097 filter_parts
.append(string_
)
2099 def _remove_unused_ops(tokens
):
2100 # Remove operators that we don't use and join them with the surrounding strings.
2101 # E.g. 'mp4' '-' 'baseline' '-' '16x9' is converted to 'mp4-baseline-16x9'
2102 ALLOWED_OPS
= ('/', '+', ',', '(', ')')
2103 last_string
, last_start
, last_end
, last_line
= None, None, None, None
2104 for type, string_
, start
, end
, line
in tokens
:
2105 if type == tokenize
.OP
and string_
== '[':
2107 yield tokenize
.NAME
, last_string
, last_start
, last_end
, last_line
2109 yield type, string_
, start
, end
, line
2110 # everything inside brackets will be handled by _parse_filter
2111 for type, string_
, start
, end
, line
in tokens
:
2112 yield type, string_
, start
, end
, line
2113 if type == tokenize
.OP
and string_
== ']':
2115 elif type == tokenize
.OP
and string_
in ALLOWED_OPS
:
2117 yield tokenize
.NAME
, last_string
, last_start
, last_end
, last_line
2119 yield type, string_
, start
, end
, line
2120 elif type in [tokenize
.NAME
, tokenize
.NUMBER
, tokenize
.OP
]:
2122 last_string
= string_
2126 last_string
+= string_
2128 yield tokenize
.NAME
, last_string
, last_start
, last_end
, last_line
2130 def _parse_format_selection(tokens
, inside_merge
=False, inside_choice
=False, inside_group
=False):
2132 current_selector
= None
2133 for type, string_
, start
, _
, _
in tokens
:
2134 # ENCODING is only defined in python 3.x
2135 if type == getattr(tokenize
, 'ENCODING', None):
2137 elif type in [tokenize
.NAME
, tokenize
.NUMBER
]:
2138 current_selector
= FormatSelector(SINGLE
, string_
, [])
2139 elif type == tokenize
.OP
:
2141 if not inside_group
:
2142 # ')' will be handled by the parentheses group
2143 tokens
.restore_last_token()
2145 elif inside_merge
and string_
in ['/', ',']:
2146 tokens
.restore_last_token()
2148 elif inside_choice
and string_
== ',':
2149 tokens
.restore_last_token()
2151 elif string_
== ',':
2152 if not current_selector
:
2153 raise syntax_error('"," must follow a format selector', start
)
2154 selectors
.append(current_selector
)
2155 current_selector
= None
2156 elif string_
== '/':
2157 if not current_selector
:
2158 raise syntax_error('"/" must follow a format selector', start
)
2159 first_choice
= current_selector
2160 second_choice
= _parse_format_selection(tokens
, inside_choice
=True)
2161 current_selector
= FormatSelector(PICKFIRST
, (first_choice
, second_choice
), [])
2162 elif string_
== '[':
2163 if not current_selector
:
2164 current_selector
= FormatSelector(SINGLE
, 'best', [])
2165 format_filter
= _parse_filter(tokens
)
2166 current_selector
.filters
.append(format_filter
)
2167 elif string_
== '(':
2168 if current_selector
:
2169 raise syntax_error('Unexpected "("', start
)
2170 group
= _parse_format_selection(tokens
, inside_group
=True)
2171 current_selector
= FormatSelector(GROUP
, group
, [])
2172 elif string_
== '+':
2173 if not current_selector
:
2174 raise syntax_error('Unexpected "+"', start
)
2175 selector_1
= current_selector
2176 selector_2
= _parse_format_selection(tokens
, inside_merge
=True)
2178 raise syntax_error('Expected a selector', start
)
2179 current_selector
= FormatSelector(MERGE
, (selector_1
, selector_2
), [])
2181 raise syntax_error(f
'Operator not recognized: "{string_}"', start
)
2182 elif type == tokenize
.ENDMARKER
:
2184 if current_selector
:
2185 selectors
.append(current_selector
)
2188 def _merge(formats_pair
):
2189 format_1
, format_2
= formats_pair
2192 formats_info
.extend(format_1
.get('requested_formats', (format_1
,)))
2193 formats_info
.extend(format_2
.get('requested_formats', (format_2
,)))
2195 if not allow_multiple_streams
['video'] or not allow_multiple_streams
['audio']:
2196 get_no_more
= {'video': False, 'audio': False}
2197 for (i
, fmt_info
) in enumerate(formats_info
):
2198 if fmt_info
.get('acodec') == fmt_info
.get('vcodec') == 'none':
2201 for aud_vid
in ['audio', 'video']:
2202 if not allow_multiple_streams
[aud_vid
] and fmt_info
.get(aud_vid
[0] + 'codec') != 'none':
2203 if get_no_more
[aud_vid
]:
2206 get_no_more
[aud_vid
] = True
2208 if len(formats_info
) == 1:
2209 return formats_info
[0]
2211 video_fmts
= [fmt_info
for fmt_info
in formats_info
if fmt_info
.get('vcodec') != 'none']
2212 audio_fmts
= [fmt_info
for fmt_info
in formats_info
if fmt_info
.get('acodec') != 'none']
2214 the_only_video
= video_fmts
[0] if len(video_fmts
) == 1 else None
2215 the_only_audio
= audio_fmts
[0] if len(audio_fmts
) == 1 else None
2217 output_ext
= get_compatible_ext(
2218 vcodecs
=[f
.get('vcodec') for f
in video_fmts
],
2219 acodecs
=[f
.get('acodec') for f
in audio_fmts
],
2220 vexts
=[f
['ext'] for f
in video_fmts
],
2221 aexts
=[f
['ext'] for f
in audio_fmts
],
2222 preferences
=(try_call(lambda: self
.params
['merge_output_format'].split('/'))
2223 or self
.params
.get('prefer_free_formats') and ('webm', 'mkv')))
2225 filtered
= lambda *keys
: filter(None, (traverse_obj(fmt
, *keys
) for fmt
in formats_info
))
2228 'requested_formats': formats_info
,
2229 'format': '+'.join(filtered('format')),
2230 'format_id': '+'.join(filtered('format_id')),
2232 'protocol': '+'.join(map(determine_protocol
, formats_info
)),
2233 'language': '+'.join(orderedSet(filtered('language'))) or None,
2234 'format_note': '+'.join(orderedSet(filtered('format_note'))) or None,
2235 'filesize_approx': sum(filtered('filesize', 'filesize_approx')) or None,
2236 'tbr': sum(filtered('tbr', 'vbr', 'abr')),
2241 'width': the_only_video
.get('width'),
2242 'height': the_only_video
.get('height'),
2243 'resolution': the_only_video
.get('resolution') or self
.format_resolution(the_only_video
),
2244 'fps': the_only_video
.get('fps'),
2245 'dynamic_range': the_only_video
.get('dynamic_range'),
2246 'vcodec': the_only_video
.get('vcodec'),
2247 'vbr': the_only_video
.get('vbr'),
2248 'stretched_ratio': the_only_video
.get('stretched_ratio'),
2249 'aspect_ratio': the_only_video
.get('aspect_ratio'),
2254 'acodec': the_only_audio
.get('acodec'),
2255 'abr': the_only_audio
.get('abr'),
2256 'asr': the_only_audio
.get('asr'),
2257 'audio_channels': the_only_audio
.get('audio_channels')
2262 def _check_formats(formats
):
2263 if not check_formats
:
2266 yield from self
._check
_formats
(formats
)
2268 def _build_selector_function(selector
):
2269 if isinstance(selector
, list): # ,
2270 fs
= [_build_selector_function(s
) for s
in selector
]
2272 def selector_function(ctx
):
2275 return selector_function
2277 elif selector
.type == GROUP
: # ()
2278 selector_function
= _build_selector_function(selector
.selector
)
2280 elif selector
.type == PICKFIRST
: # /
2281 fs
= [_build_selector_function(s
) for s
in selector
.selector
]
2283 def selector_function(ctx
):
2285 picked_formats
= list(f(ctx
))
2287 return picked_formats
2290 elif selector
.type == MERGE
: # +
2291 selector_1
, selector_2
= map(_build_selector_function
, selector
.selector
)
2293 def selector_function(ctx
):
2294 for pair
in itertools
.product(selector_1(ctx
), selector_2(ctx
)):
2297 elif selector
.type == SINGLE
: # atom
2298 format_spec
= selector
.selector
or 'best'
2300 # TODO: Add allvideo, allaudio etc by generalizing the code with best/worst selector
2301 if format_spec
== 'all':
2302 def selector_function(ctx
):
2303 yield from _check_formats(ctx
['formats'][::-1])
2304 elif format_spec
== 'mergeall':
2305 def selector_function(ctx
):
2306 formats
= list(_check_formats(
2307 f
for f
in ctx
['formats'] if f
.get('vcodec') != 'none' or f
.get('acodec') != 'none'))
2310 merged_format
= formats
[-1]
2311 for f
in formats
[-2::-1]:
2312 merged_format
= _merge((merged_format
, f
))
2316 format_fallback
, seperate_fallback
, format_reverse
, format_idx
= False, None, True, 1
2318 r
'(?P<bw>best|worst|b|w)(?P<type>video|audio|v|a)?(?P<mod>\*)?(?:\.(?P<n>[1-9]\d*))?$',
2320 if mobj
is not None:
2321 format_idx
= int_or_none(mobj
.group('n'), default
=1)
2322 format_reverse
= mobj
.group('bw')[0] == 'b'
2323 format_type
= (mobj
.group('type') or [None])[0]
2324 not_format_type
= {'v': 'a', 'a': 'v'}
.get(format_type
)
2325 format_modified
= mobj
.group('mod') is not None
2327 format_fallback
= not format_type
and not format_modified
# for b, w
2329 (lambda f
: f
.get('%scodec' % format_type
) != 'none')
2330 if format_type
and format_modified
# bv*, ba*, wv*, wa*
2331 else (lambda f
: f
.get('%scodec' % not_format_type
) == 'none')
2332 if format_type
# bv, ba, wv, wa
2333 else (lambda f
: f
.get('vcodec') != 'none' and f
.get('acodec') != 'none')
2334 if not format_modified
# b, w
2335 else lambda f
: True) # b*, w*
2336 filter_f
= lambda f
: _filter_f(f
) and (
2337 f
.get('vcodec') != 'none' or f
.get('acodec') != 'none')
2339 if format_spec
in self
._format
_selection
_exts
['audio']:
2340 filter_f
= lambda f
: f
.get('ext') == format_spec
and f
.get('acodec') != 'none'
2341 elif format_spec
in self
._format
_selection
_exts
['video']:
2342 filter_f
= lambda f
: f
.get('ext') == format_spec
and f
.get('acodec') != 'none' and f
.get('vcodec') != 'none'
2343 seperate_fallback
= lambda f
: f
.get('ext') == format_spec
and f
.get('vcodec') != 'none'
2344 elif format_spec
in self
._format
_selection
_exts
['storyboards']:
2345 filter_f
= lambda f
: f
.get('ext') == format_spec
and f
.get('acodec') == 'none' and f
.get('vcodec') == 'none'
2347 filter_f
= lambda f
: f
.get('format_id') == format_spec
# id
2349 def selector_function(ctx
):
2350 formats
= list(ctx
['formats'])
2351 matches
= list(filter(filter_f
, formats
)) if filter_f
is not None else formats
2353 if format_fallback
and ctx
['incomplete_formats']:
2354 # for extractors with incomplete formats (audio only (soundcloud)
2355 # or video only (imgur)) best/worst will fallback to
2356 # best/worst {video,audio}-only format
2358 elif seperate_fallback
and not ctx
['has_merged_format']:
2359 # for compatibility with youtube-dl when there is no pre-merged format
2360 matches
= list(filter(seperate_fallback
, formats
))
2361 matches
= LazyList(_check_formats(matches
[::-1 if format_reverse
else 1]))
2363 yield matches
[format_idx
- 1]
2364 except LazyList
.IndexError:
2367 filters
= [self
._build
_format
_filter
(f
) for f
in selector
.filters
]
2369 def final_selector(ctx
):
2370 ctx_copy
= dict(ctx
)
2371 for _filter
in filters
:
2372 ctx_copy
['formats'] = list(filter(_filter
, ctx_copy
['formats']))
2373 return selector_function(ctx_copy
)
2374 return final_selector
2376 stream
= io
.BytesIO(format_spec
.encode())
2378 tokens
= list(_remove_unused_ops(tokenize
.tokenize(stream
.readline
)))
2379 except tokenize
.TokenError
:
2380 raise syntax_error('Missing closing/opening brackets or parenthesis', (0, len(format_spec
)))
2382 class TokenIterator
:
2383 def __init__(self
, tokens
):
2384 self
.tokens
= tokens
2391 if self
.counter
>= len(self
.tokens
):
2392 raise StopIteration()
2393 value
= self
.tokens
[self
.counter
]
2399 def restore_last_token(self
):
2402 parsed_selector
= _parse_format_selection(iter(TokenIterator(tokens
)))
2403 return _build_selector_function(parsed_selector
)
2405 def _calc_headers(self
, info_dict
):
2406 res
= merge_headers(self
.params
['http_headers'], info_dict
.get('http_headers') or {})
2407 if 'Youtubedl-No-Compression' in res
: # deprecated
2408 res
.pop('Youtubedl-No-Compression', None)
2409 res
['Accept-Encoding'] = 'identity'
2410 cookies
= self
.cookiejar
.get_cookie_header(info_dict
['url'])
2412 res
['Cookie'] = cookies
2414 if 'X-Forwarded-For' not in res
:
2415 x_forwarded_for_ip
= info_dict
.get('__x_forwarded_for_ip')
2416 if x_forwarded_for_ip
:
2417 res
['X-Forwarded-For'] = x_forwarded_for_ip
2421 def _calc_cookies(self
, url
):
2422 self
.deprecation_warning('"YoutubeDL._calc_cookies" is deprecated and may be removed in a future version')
2423 return self
.cookiejar
.get_cookie_header(url
)
2425 def _sort_thumbnails(self
, thumbnails
):
2426 thumbnails
.sort(key
=lambda t
: (
2427 t
.get('preference') if t
.get('preference') is not None else -1,
2428 t
.get('width') if t
.get('width') is not None else -1,
2429 t
.get('height') if t
.get('height') is not None else -1,
2430 t
.get('id') if t
.get('id') is not None else '',
2433 def _sanitize_thumbnails(self
, info_dict
):
2434 thumbnails
= info_dict
.get('thumbnails')
2435 if thumbnails
is None:
2436 thumbnail
= info_dict
.get('thumbnail')
2438 info_dict
['thumbnails'] = thumbnails
= [{'url': thumbnail}
]
2442 def check_thumbnails(thumbnails
):
2443 for t
in thumbnails
:
2444 self
.to_screen(f
'[info] Testing thumbnail {t["id"]}')
2446 self
.urlopen(HEADRequest(t
['url']))
2447 except network_exceptions
as err
:
2448 self
.to_screen(f
'[info] Unable to connect to thumbnail {t["id"]} URL {t["url"]!r} - {err}. Skipping...')
2452 self
._sort
_thumbnails
(thumbnails
)
2453 for i
, t
in enumerate(thumbnails
):
2454 if t
.get('id') is None:
2456 if t
.get('width') and t
.get('height'):
2457 t
['resolution'] = '%dx%d' % (t
['width'], t
['height'])
2458 t
['url'] = sanitize_url(t
['url'])
2460 if self
.params
.get('check_formats') is True:
2461 info_dict
['thumbnails'] = LazyList(check_thumbnails(thumbnails
[::-1]), reverse
=True)
2463 info_dict
['thumbnails'] = thumbnails
2465 def _fill_common_fields(self
, info_dict
, final
=True):
2466 # TODO: move sanitization here
2468 title
= info_dict
['fulltitle'] = info_dict
.get('title')
2471 self
.write_debug('Extractor gave empty title. Creating a generic title')
2473 self
.report_warning('Extractor failed to obtain "title". Creating a generic title instead')
2474 info_dict
['title'] = f
'{info_dict["extractor"].replace(":", "-")} video #{info_dict["id"]}'
2476 if info_dict
.get('duration') is not None:
2477 info_dict
['duration_string'] = formatSeconds(info_dict
['duration'])
2479 for ts_key
, date_key
in (
2480 ('timestamp', 'upload_date'),
2481 ('release_timestamp', 'release_date'),
2482 ('modified_timestamp', 'modified_date'),
2484 if info_dict
.get(date_key
) is None and info_dict
.get(ts_key
) is not None:
2485 # Working around out-of-range timestamp values (e.g. negative ones on Windows,
2486 # see http://bugs.python.org/issue1646728)
2487 with contextlib
.suppress(ValueError, OverflowError, OSError):
2488 upload_date
= datetime
.datetime
.utcfromtimestamp(info_dict
[ts_key
])
2489 info_dict
[date_key
] = upload_date
.strftime('%Y%m%d')
2491 live_keys
= ('is_live', 'was_live')
2492 live_status
= info_dict
.get('live_status')
2493 if live_status
is None:
2494 for key
in live_keys
:
2495 if info_dict
.get(key
) is False:
2497 if info_dict
.get(key
):
2500 if all(info_dict
.get(key
) is False for key
in live_keys
):
2501 live_status
= 'not_live'
2503 info_dict
['live_status'] = live_status
2504 for key
in live_keys
:
2505 if info_dict
.get(key
) is None:
2506 info_dict
[key
] = (live_status
== key
)
2507 if live_status
== 'post_live':
2508 info_dict
['was_live'] = True
2510 # Auto generate title fields corresponding to the *_number fields when missing
2511 # in order to always have clean titles. This is very common for TV series.
2512 for field
in ('chapter', 'season', 'episode'):
2513 if final
and info_dict
.get('%s_number' % field
) is not None and not info_dict
.get(field
):
2514 info_dict
[field
] = '%s %d' % (field
.capitalize(), info_dict
['%s_number' % field
])
2516 def _raise_pending_errors(self
, info
):
2517 err
= info
.pop('__pending_error', None)
2519 self
.report_error(err
, tb
=False)
2521 def sort_formats(self
, info_dict
):
2522 formats
= self
._get
_formats
(info_dict
)
2523 formats
.sort(key
=FormatSorter(
2524 self
, info_dict
.get('_format_sort_fields') or []).calculate_preference
)
2526 def process_video_result(self
, info_dict
, download
=True):
2527 assert info_dict
.get('_type', 'video') == 'video'
2528 self
._num
_videos
+= 1
2530 if 'id' not in info_dict
:
2531 raise ExtractorError('Missing "id" field in extractor result', ie
=info_dict
['extractor'])
2532 elif not info_dict
.get('id'):
2533 raise ExtractorError('Extractor failed to obtain "id"', ie
=info_dict
['extractor'])
2535 def report_force_conversion(field
, field_not
, conversion
):
2536 self
.report_warning(
2537 '"%s" field is not %s - forcing %s conversion, there is an error in extractor'
2538 % (field
, field_not
, conversion
))
2540 def sanitize_string_field(info
, string_field
):
2541 field
= info
.get(string_field
)
2542 if field
is None or isinstance(field
, str):
2544 report_force_conversion(string_field
, 'a string', 'string')
2545 info
[string_field
] = str(field
)
2547 def sanitize_numeric_fields(info
):
2548 for numeric_field
in self
._NUMERIC
_FIELDS
:
2549 field
= info
.get(numeric_field
)
2550 if field
is None or isinstance(field
, (int, float)):
2552 report_force_conversion(numeric_field
, 'numeric', 'int')
2553 info
[numeric_field
] = int_or_none(field
)
2555 sanitize_string_field(info_dict
, 'id')
2556 sanitize_numeric_fields(info_dict
)
2557 if info_dict
.get('section_end') and info_dict
.get('section_start') is not None:
2558 info_dict
['duration'] = round(info_dict
['section_end'] - info_dict
['section_start'], 3)
2559 if (info_dict
.get('duration') or 0) <= 0 and info_dict
.pop('duration', None):
2560 self
.report_warning('"duration" field is negative, there is an error in extractor')
2562 chapters
= info_dict
.get('chapters') or []
2563 if chapters
and chapters
[0].get('start_time'):
2564 chapters
.insert(0, {'start_time': 0}
)
2566 dummy_chapter
= {'end_time': 0, 'start_time': info_dict.get('duration')}
2567 for idx
, (prev
, current
, next_
) in enumerate(zip(
2568 (dummy_chapter
, *chapters
), chapters
, (*chapters
[1:], dummy_chapter
)), 1):
2569 if current
.get('start_time') is None:
2570 current
['start_time'] = prev
.get('end_time')
2571 if not current
.get('end_time'):
2572 current
['end_time'] = next_
.get('start_time')
2573 if not current
.get('title'):
2574 current
['title'] = f
'<Untitled Chapter {idx}>'
2576 if 'playlist' not in info_dict
:
2577 # It isn't part of a playlist
2578 info_dict
['playlist'] = None
2579 info_dict
['playlist_index'] = None
2581 self
._sanitize
_thumbnails
(info_dict
)
2583 thumbnail
= info_dict
.get('thumbnail')
2584 thumbnails
= info_dict
.get('thumbnails')
2586 info_dict
['thumbnail'] = sanitize_url(thumbnail
)
2588 info_dict
['thumbnail'] = thumbnails
[-1]['url']
2590 if info_dict
.get('display_id') is None and 'id' in info_dict
:
2591 info_dict
['display_id'] = info_dict
['id']
2593 self
._fill
_common
_fields
(info_dict
)
2595 for cc_kind
in ('subtitles', 'automatic_captions'):
2596 cc
= info_dict
.get(cc_kind
)
2598 for _
, subtitle
in cc
.items():
2599 for subtitle_format
in subtitle
:
2600 if subtitle_format
.get('url'):
2601 subtitle_format
['url'] = sanitize_url(subtitle_format
['url'])
2602 if subtitle_format
.get('ext') is None:
2603 subtitle_format
['ext'] = determine_ext(subtitle_format
['url']).lower()
2605 automatic_captions
= info_dict
.get('automatic_captions')
2606 subtitles
= info_dict
.get('subtitles')
2608 info_dict
['requested_subtitles'] = self
.process_subtitles(
2609 info_dict
['id'], subtitles
, automatic_captions
)
2611 formats
= self
._get
_formats
(info_dict
)
2613 # Backward compatibility with InfoExtractor._sort_formats
2614 field_preference
= (formats
or [{}])[0].pop('__sort_fields', None)
2615 if field_preference
:
2616 info_dict
['_format_sort_fields'] = field_preference
2618 # or None ensures --clean-infojson removes it
2619 info_dict
['_has_drm'] = any(f
.get('has_drm') for f
in formats
) or None
2620 if not self
.params
.get('allow_unplayable_formats'):
2621 formats
= [f
for f
in formats
if not f
.get('has_drm')]
2623 if formats
and all(f
.get('acodec') == f
.get('vcodec') == 'none' for f
in formats
):
2624 self
.report_warning(
2625 f
'{"This video is DRM protected and " if info_dict["_has_drm"] else ""}'
2626 'only images are available for download. Use --list-formats to see them'.capitalize())
2628 get_from_start
= not info_dict
.get('is_live') or bool(self
.params
.get('live_from_start'))
2629 if not get_from_start
:
2630 info_dict
['title'] += ' ' + datetime
.datetime
.now().strftime('%Y-%m-%d %H:%M')
2631 if info_dict
.get('is_live') and formats
:
2632 formats
= [f
for f
in formats
if bool(f
.get('is_from_start')) == get_from_start
]
2633 if get_from_start
and not formats
:
2634 self
.raise_no_formats(info_dict
, msg
=(
2635 '--live-from-start is passed, but there are no formats that can be downloaded from the start. '
2636 'If you want to download from the current time, use --no-live-from-start'))
2638 def is_wellformed(f
):
2641 self
.report_warning(
2642 '"url" field is missing or empty - skipping format, '
2643 'there is an error in extractor')
2645 if isinstance(url
, bytes):
2646 sanitize_string_field(f
, 'url')
2649 # Filter out malformed formats for better extraction robustness
2650 formats
= list(filter(is_wellformed
, formats
or []))
2653 self
.raise_no_formats(info_dict
)
2655 for format
in formats
:
2656 sanitize_string_field(format
, 'format_id')
2657 sanitize_numeric_fields(format
)
2658 format
['url'] = sanitize_url(format
['url'])
2659 if format
.get('ext') is None:
2660 format
['ext'] = determine_ext(format
['url']).lower()
2661 if format
.get('protocol') is None:
2662 format
['protocol'] = determine_protocol(format
)
2663 if format
.get('resolution') is None:
2664 format
['resolution'] = self
.format_resolution(format
, default
=None)
2665 if format
.get('dynamic_range') is None and format
.get('vcodec') != 'none':
2666 format
['dynamic_range'] = 'SDR'
2667 if format
.get('aspect_ratio') is None:
2668 format
['aspect_ratio'] = try_call(lambda: round(format
['width'] / format
['height'], 2))
2669 if (not format
.get('manifest_url') # For fragmented formats, "tbr" is often max bitrate and not average
2670 and info_dict
.get('duration') and format
.get('tbr')
2671 and not format
.get('filesize') and not format
.get('filesize_approx')):
2672 format
['filesize_approx'] = int(info_dict
['duration'] * format
['tbr'] * (1024 / 8))
2673 format
['http_headers'] = self
._calc
_headers
(collections
.ChainMap(format
, info_dict
))
2675 # This is copied to http_headers by the above _calc_headers and can now be removed
2676 if '__x_forwarded_for_ip' in info_dict
:
2677 del info_dict
['__x_forwarded_for_ip']
2681 '_format_sort_fields': info_dict
.get('_format_sort_fields')
2684 # Sanitize and group by format_id
2686 for i
, format
in enumerate(formats
):
2687 if not format
.get('format_id'):
2688 format
['format_id'] = str(i
)
2690 # Sanitize format_id from characters used in format selector expression
2691 format
['format_id'] = re
.sub(r
'[\s,/+\[\]()]', '_', format
['format_id'])
2692 formats_dict
.setdefault(format
['format_id'], []).append(format
)
2694 # Make sure all formats have unique format_id
2695 common_exts
= set(itertools
.chain(*self
._format
_selection
_exts
.values()))
2696 for format_id
, ambiguous_formats
in formats_dict
.items():
2697 ambigious_id
= len(ambiguous_formats
) > 1
2698 for i
, format
in enumerate(ambiguous_formats
):
2700 format
['format_id'] = '%s-%d' % (format_id
, i
)
2701 # Ensure there is no conflict between id and ext in format selection
2702 # See https://github.com/yt-dlp/yt-dlp/issues/1282
2703 if format
['format_id'] != format
['ext'] and format
['format_id'] in common_exts
:
2704 format
['format_id'] = 'f%s' % format
['format_id']
2706 if format
.get('format') is None:
2707 format
['format'] = '{id} - {res}{note}'.format(
2708 id=format
['format_id'],
2709 res
=self
.format_resolution(format
),
2710 note
=format_field(format
, 'format_note', ' (%s)'),
2713 if self
.params
.get('check_formats') is True:
2714 formats
= LazyList(self
._check
_formats
(formats
[::-1]), reverse
=True)
2716 if not formats
or formats
[0] is not info_dict
:
2717 # only set the 'formats' fields if the original info_dict list them
2718 # otherwise we end up with a circular reference, the first (and unique)
2719 # element in the 'formats' field in info_dict is info_dict itself,
2720 # which can't be exported to json
2721 info_dict
['formats'] = formats
2723 info_dict
, _
= self
.pre_process(info_dict
)
2725 if self
._match
_entry
(info_dict
, incomplete
=self
._format
_fields
) is not None:
2728 self
.post_extract(info_dict
)
2729 info_dict
, _
= self
.pre_process(info_dict
, 'after_filter')
2731 # The pre-processors may have modified the formats
2732 formats
= self
._get
_formats
(info_dict
)
2734 list_only
= self
.params
.get('simulate') == 'list_only'
2735 interactive_format_selection
= not list_only
and self
.format_selector
== '-'
2736 if self
.params
.get('list_thumbnails'):
2737 self
.list_thumbnails(info_dict
)
2738 if self
.params
.get('listsubtitles'):
2739 if 'automatic_captions' in info_dict
:
2740 self
.list_subtitles(
2741 info_dict
['id'], automatic_captions
, 'automatic captions')
2742 self
.list_subtitles(info_dict
['id'], subtitles
, 'subtitles')
2743 if self
.params
.get('listformats') or interactive_format_selection
:
2744 self
.list_formats(info_dict
)
2746 # Without this printing, -F --print-json will not work
2747 self
.__forced
_printings
(info_dict
)
2750 format_selector
= self
.format_selector
2752 if interactive_format_selection
:
2753 req_format
= input(self
._format
_screen
('\nEnter format selector ', self
.Styles
.EMPHASIS
)
2754 + '(Press ENTER for default, or Ctrl+C to quit)'
2755 + self
._format
_screen
(': ', self
.Styles
.EMPHASIS
))
2757 format_selector
= self
.build_format_selector(req_format
) if req_format
else None
2758 except SyntaxError as err
:
2759 self
.report_error(err
, tb
=False, is_error
=False)
2762 if format_selector
is None:
2763 req_format
= self
._default
_format
_spec
(info_dict
, download
=download
)
2764 self
.write_debug(f
'Default format spec: {req_format}')
2765 format_selector
= self
.build_format_selector(req_format
)
2767 formats_to_download
= list(format_selector({
2769 'has_merged_format': any('none' not in (f
.get('acodec'), f
.get('vcodec')) for f
in formats
),
2770 'incomplete_formats': (
2771 # All formats are video-only or
2772 all(f
.get('vcodec') != 'none' and f
.get('acodec') == 'none' for f
in formats
)
2773 # all formats are audio-only
2774 or all(f
.get('vcodec') == 'none' and f
.get('acodec') != 'none' for f
in formats
)),
2776 if interactive_format_selection
and not formats_to_download
:
2777 self
.report_error('Requested format is not available', tb
=False, is_error
=False)
2781 if not formats_to_download
:
2782 if not self
.params
.get('ignore_no_formats_error'):
2783 raise ExtractorError(
2784 'Requested format is not available. Use --list-formats for a list of available formats',
2785 expected
=True, video_id
=info_dict
['id'], ie
=info_dict
['extractor'])
2786 self
.report_warning('Requested format is not available')
2787 # Process what we can, even without any available formats.
2788 formats_to_download
= [{}]
2790 requested_ranges
= tuple(self
.params
.get('download_ranges', lambda *_
: [{}])(info_dict
, self
))
2791 best_format
, downloaded_formats
= formats_to_download
[-1], []
2793 if best_format
and requested_ranges
:
2794 def to_screen(*msg
):
2795 self
.to_screen(f
'[info] {info_dict["id"]}: {" ".join(", ".join(variadic(m)) for m in msg)}')
2797 to_screen(f
'Downloading {len(formats_to_download)} format(s):',
2798 (f
['format_id'] for f
in formats_to_download
))
2799 if requested_ranges
!= ({}, ):
2800 to_screen(f
'Downloading {len(requested_ranges)} time ranges:',
2801 (f
'{c["start_time"]:.1f}-{c["end_time"]:.1f}' for c
in requested_ranges
))
2802 max_downloads_reached
= False
2804 for fmt
, chapter
in itertools
.product(formats_to_download
, requested_ranges
):
2805 new_info
= self
._copy
_infodict
(info_dict
)
2806 new_info
.update(fmt
)
2807 offset
, duration
= info_dict
.get('section_start') or 0, info_dict
.get('duration') or float('inf')
2808 end_time
= offset
+ min(chapter
.get('end_time', duration
), duration
)
2809 if chapter
or offset
:
2811 'section_start': offset
+ chapter
.get('start_time', 0),
2812 # duration may not be accurate. So allow deviations <1sec
2813 'section_end': end_time
if end_time
<= offset
+ duration
+ 1 else None,
2814 'section_title': chapter
.get('title'),
2815 'section_number': chapter
.get('index'),
2817 downloaded_formats
.append(new_info
)
2819 self
.process_info(new_info
)
2820 except MaxDownloadsReached
:
2821 max_downloads_reached
= True
2822 self
._raise
_pending
_errors
(new_info
)
2823 # Remove copied info
2824 for key
, val
in tuple(new_info
.items()):
2825 if info_dict
.get(key
) == val
:
2827 if max_downloads_reached
:
2830 write_archive
= {f.get('__write_download_archive', False) for f in downloaded_formats}
2831 assert write_archive
.issubset({True, False, 'ignore'}
)
2832 if True in write_archive
and False not in write_archive
:
2833 self
.record_download_archive(info_dict
)
2835 info_dict
['requested_downloads'] = downloaded_formats
2836 info_dict
= self
.run_all_pps('after_video', info_dict
)
2837 if max_downloads_reached
:
2838 raise MaxDownloadsReached()
2840 # We update the info dict with the selected best quality format (backwards compatibility)
2841 info_dict
.update(best_format
)
2844 def process_subtitles(self
, video_id
, normal_subtitles
, automatic_captions
):
2845 """Select the requested subtitles and their format"""
2846 available_subs
, normal_sub_langs
= {}, []
2847 if normal_subtitles
and self
.params
.get('writesubtitles'):
2848 available_subs
.update(normal_subtitles
)
2849 normal_sub_langs
= tuple(normal_subtitles
.keys())
2850 if automatic_captions
and self
.params
.get('writeautomaticsub'):
2851 for lang
, cap_info
in automatic_captions
.items():
2852 if lang
not in available_subs
:
2853 available_subs
[lang
] = cap_info
2855 if not available_subs
or (
2856 not self
.params
.get('writesubtitles')
2857 and not self
.params
.get('writeautomaticsub')):
2860 all_sub_langs
= tuple(available_subs
.keys())
2861 if self
.params
.get('allsubtitles', False):
2862 requested_langs
= all_sub_langs
2863 elif self
.params
.get('subtitleslangs', False):
2865 requested_langs
= orderedSet_from_options(
2866 self
.params
.get('subtitleslangs'), {'all': all_sub_langs}
, use_regex
=True)
2867 except re
.error
as e
:
2868 raise ValueError(f
'Wrong regex for subtitlelangs: {e.pattern}')
2870 requested_langs
= LazyList(itertools
.chain(
2871 ['en'] if 'en' in normal_sub_langs
else [],
2872 filter(lambda f
: f
.startswith('en'), normal_sub_langs
),
2873 ['en'] if 'en' in all_sub_langs
else [],
2874 filter(lambda f
: f
.startswith('en'), all_sub_langs
),
2875 normal_sub_langs
, all_sub_langs
,
2878 self
.to_screen(f
'[info] {video_id}: Downloading subtitles: {", ".join(requested_langs)}')
2880 formats_query
= self
.params
.get('subtitlesformat', 'best')
2881 formats_preference
= formats_query
.split('/') if formats_query
else []
2883 for lang
in requested_langs
:
2884 formats
= available_subs
.get(lang
)
2886 self
.report_warning(f
'{lang} subtitles not available for {video_id}')
2888 for ext
in formats_preference
:
2892 matches
= list(filter(lambda f
: f
['ext'] == ext
, formats
))
2898 self
.report_warning(
2899 'No subtitle format found matching "%s" for language %s, '
2900 'using %s' % (formats_query
, lang
, f
['ext']))
2904 def _forceprint(self
, key
, info_dict
):
2905 if info_dict
is None:
2907 info_copy
= info_dict
.copy()
2908 info_copy
.setdefault('filename', self
.prepare_filename(info_dict
))
2909 if info_dict
.get('requested_formats') is not None:
2910 # For RTMP URLs, also include the playpath
2911 info_copy
['urls'] = '\n'.join(f
['url'] + f
.get('play_path', '') for f
in info_dict
['requested_formats'])
2912 elif info_dict
.get('url'):
2913 info_copy
['urls'] = info_dict
['url'] + info_dict
.get('play_path', '')
2914 info_copy
['formats_table'] = self
.render_formats_table(info_dict
)
2915 info_copy
['thumbnails_table'] = self
.render_thumbnails_table(info_dict
)
2916 info_copy
['subtitles_table'] = self
.render_subtitles_table(info_dict
.get('id'), info_dict
.get('subtitles'))
2917 info_copy
['automatic_captions_table'] = self
.render_subtitles_table(info_dict
.get('id'), info_dict
.get('automatic_captions'))
2919 def format_tmpl(tmpl
):
2920 mobj
= re
.fullmatch(r
'([\w.:,]|-\d|(?P<dict>{([\w.:,]|-\d)+}))+=?', tmpl
)
2925 if tmpl
.startswith('{'):
2926 tmpl
, fmt
= f
'.{tmpl}', '%({})j'
2927 if tmpl
.endswith('='):
2928 tmpl
, fmt
= tmpl
[:-1], '{0} = %({0})#j'
2929 return '\n'.join(map(fmt
.format
, [tmpl
] if mobj
.group('dict') else tmpl
.split(',')))
2931 for tmpl
in self
.params
['forceprint'].get(key
, []):
2932 self
.to_stdout(self
.evaluate_outtmpl(format_tmpl(tmpl
), info_copy
))
2934 for tmpl
, file_tmpl
in self
.params
['print_to_file'].get(key
, []):
2935 filename
= self
.prepare_filename(info_dict
, outtmpl
=file_tmpl
)
2936 tmpl
= format_tmpl(tmpl
)
2937 self
.to_screen(f
'[info] Writing {tmpl!r} to: {filename}')
2938 if self
._ensure
_dir
_exists
(filename
):
2939 with open(filename
, 'a', encoding
='utf-8', newline
='') as f
:
2940 f
.write(self
.evaluate_outtmpl(tmpl
, info_copy
) + os
.linesep
)
2944 def __forced_printings(self
, info_dict
, filename
=None, incomplete
=True):
2945 if (self
.params
.get('forcejson')
2946 or self
.params
['forceprint'].get('video')
2947 or self
.params
['print_to_file'].get('video')):
2948 self
.post_extract(info_dict
)
2950 info_dict
['filename'] = filename
2951 info_copy
= self
._forceprint
('video', info_dict
)
2953 def print_field(field
, actual_field
=None, optional
=False):
2954 if actual_field
is None:
2955 actual_field
= field
2956 if self
.params
.get(f
'force{field}') and (
2957 info_copy
.get(field
) is not None or (not optional
and not incomplete
)):
2958 self
.to_stdout(info_copy
[actual_field
])
2960 print_field('title')
2962 print_field('url', 'urls')
2963 print_field('thumbnail', optional
=True)
2964 print_field('description', optional
=True)
2965 print_field('filename')
2966 if self
.params
.get('forceduration') and info_copy
.get('duration') is not None:
2967 self
.to_stdout(formatSeconds(info_copy
['duration']))
2968 print_field('format')
2970 if self
.params
.get('forcejson'):
2971 self
.to_stdout(json
.dumps(self
.sanitize_info(info_dict
)))
2973 def dl(self
, name
, info
, subtitle
=False, test
=False):
2974 if not info
.get('url'):
2975 self
.raise_no_formats(info
, True)
2978 verbose
= self
.params
.get('verbose')
2981 'quiet': self
.params
.get('quiet') or not verbose
,
2983 'noprogress': not verbose
,
2985 'skip_unavailable_fragments': False,
2986 'keep_fragments': False,
2988 '_no_ytdl_file': True,
2991 params
= self
.params
2992 fd
= get_suitable_downloader(info
, params
, to_stdout
=(name
== '-'))(self
, params
)
2994 for ph
in self
._progress
_hooks
:
2995 fd
.add_progress_hook(ph
)
2997 (f
['url'].split(',')[0] + ',<data>' if f
['url'].startswith('data:') else f
['url'])
2998 for f
in info
.get('requested_formats', []) or [info
])
2999 self
.write_debug(f
'Invoking {fd.FD_NAME} downloader on "{urls}"')
3001 # Note: Ideally info should be a deep-copied so that hooks cannot modify it.
3002 # But it may contain objects that are not deep-copyable
3003 new_info
= self
._copy
_infodict
(info
)
3004 if new_info
.get('http_headers') is None:
3005 new_info
['http_headers'] = self
._calc
_headers
(new_info
)
3006 return fd
.download(name
, new_info
, subtitle
)
3008 def existing_file(self
, filepaths
, *, default_overwrite
=True):
3009 existing_files
= list(filter(os
.path
.exists
, orderedSet(filepaths
)))
3010 if existing_files
and not self
.params
.get('overwrites', default_overwrite
):
3011 return existing_files
[0]
3013 for file in existing_files
:
3014 self
.report_file_delete(file)
3018 def process_info(self
, info_dict
):
3019 """Process a single resolved IE result. (Modifies it in-place)"""
3021 assert info_dict
.get('_type', 'video') == 'video'
3022 original_infodict
= info_dict
3024 if 'format' not in info_dict
and 'ext' in info_dict
:
3025 info_dict
['format'] = info_dict
['ext']
3027 if self
._match
_entry
(info_dict
) is not None:
3028 info_dict
['__write_download_archive'] = 'ignore'
3031 # Does nothing under normal operation - for backward compatibility of process_info
3032 self
.post_extract(info_dict
)
3034 def replace_info_dict(new_info
):
3036 if new_info
== info_dict
:
3039 info_dict
.update(new_info
)
3041 new_info
, _
= self
.pre_process(info_dict
, 'video')
3042 replace_info_dict(new_info
)
3043 self
._num
_downloads
+= 1
3045 # info_dict['_filename'] needs to be set for backward compatibility
3046 info_dict
['_filename'] = full_filename
= self
.prepare_filename(info_dict
, warn
=True)
3047 temp_filename
= self
.prepare_filename(info_dict
, 'temp')
3051 self
.__forced
_printings
(info_dict
, full_filename
, incomplete
=('format' not in info_dict
))
3053 def check_max_downloads():
3054 if self
._num
_downloads
>= float(self
.params
.get('max_downloads') or 'inf'):
3055 raise MaxDownloadsReached()
3057 if self
.params
.get('simulate'):
3058 info_dict
['__write_download_archive'] = self
.params
.get('force_write_download_archive')
3059 check_max_downloads()
3062 if full_filename
is None:
3064 if not self
._ensure
_dir
_exists
(encodeFilename(full_filename
)):
3066 if not self
._ensure
_dir
_exists
(encodeFilename(temp_filename
)):
3069 if self
._write
_description
('video', info_dict
,
3070 self
.prepare_filename(info_dict
, 'description')) is None:
3073 sub_files
= self
._write
_subtitles
(info_dict
, temp_filename
)
3074 if sub_files
is None:
3076 files_to_move
.update(dict(sub_files
))
3078 thumb_files
= self
._write
_thumbnails
(
3079 'video', info_dict
, temp_filename
, self
.prepare_filename(info_dict
, 'thumbnail'))
3080 if thumb_files
is None:
3082 files_to_move
.update(dict(thumb_files
))
3084 infofn
= self
.prepare_filename(info_dict
, 'infojson')
3085 _infojson_written
= self
._write
_info
_json
('video', info_dict
, infofn
)
3086 if _infojson_written
:
3087 info_dict
['infojson_filename'] = infofn
3088 # For backward compatibility, even though it was a private field
3089 info_dict
['__infojson_filename'] = infofn
3090 elif _infojson_written
is None:
3093 # Note: Annotations are deprecated
3095 if self
.params
.get('writeannotations', False):
3096 annofn
= self
.prepare_filename(info_dict
, 'annotation')
3098 if not self
._ensure
_dir
_exists
(encodeFilename(annofn
)):
3100 if not self
.params
.get('overwrites', True) and os
.path
.exists(encodeFilename(annofn
)):
3101 self
.to_screen('[info] Video annotations are already present')
3102 elif not info_dict
.get('annotations'):
3103 self
.report_warning('There are no annotations to write.')
3106 self
.to_screen('[info] Writing video annotations to: ' + annofn
)
3107 with open(encodeFilename(annofn
), 'w', encoding
='utf-8') as annofile
:
3108 annofile
.write(info_dict
['annotations'])
3109 except (KeyError, TypeError):
3110 self
.report_warning('There are no annotations to write.')
3112 self
.report_error('Cannot write annotations file: ' + annofn
)
3115 # Write internet shortcut files
3116 def _write_link_file(link_type
):
3117 url
= try_get(info_dict
['webpage_url'], iri_to_uri
)
3119 self
.report_warning(
3120 f
'Cannot write internet shortcut file because the actual URL of "{info_dict["webpage_url"]}" is unknown')
3122 linkfn
= replace_extension(self
.prepare_filename(info_dict
, 'link'), link_type
, info_dict
.get('ext'))
3123 if not self
._ensure
_dir
_exists
(encodeFilename(linkfn
)):
3125 if self
.params
.get('overwrites', True) and os
.path
.exists(encodeFilename(linkfn
)):
3126 self
.to_screen(f
'[info] Internet shortcut (.{link_type}) is already present')
3129 self
.to_screen(f
'[info] Writing internet shortcut (.{link_type}) to: {linkfn}')
3130 with open(encodeFilename(to_high_limit_path(linkfn
)), 'w', encoding
='utf-8',
3131 newline
='\r\n' if link_type
== 'url' else '\n') as linkfile
:
3132 template_vars
= {'url': url}
3133 if link_type
== 'desktop':
3134 template_vars
['filename'] = linkfn
[:-(len(link_type
) + 1)]
3135 linkfile
.write(LINK_TEMPLATES
[link_type
] % template_vars
)
3137 self
.report_error(f
'Cannot write internet shortcut {linkfn}')
3142 'url': self
.params
.get('writeurllink'),
3143 'webloc': self
.params
.get('writewebloclink'),
3144 'desktop': self
.params
.get('writedesktoplink'),
3146 if self
.params
.get('writelink'):
3147 link_type
= ('webloc' if sys
.platform
== 'darwin'
3148 else 'desktop' if sys
.platform
.startswith('linux')
3150 write_links
[link_type
] = True
3152 if any(should_write
and not _write_link_file(link_type
)
3153 for link_type
, should_write
in write_links
.items()):
3156 new_info
, files_to_move
= self
.pre_process(info_dict
, 'before_dl', files_to_move
)
3157 replace_info_dict(new_info
)
3159 if self
.params
.get('skip_download'):
3160 info_dict
['filepath'] = temp_filename
3161 info_dict
['__finaldir'] = os
.path
.dirname(os
.path
.abspath(encodeFilename(full_filename
)))
3162 info_dict
['__files_to_move'] = files_to_move
3163 replace_info_dict(self
.run_pp(MoveFilesAfterDownloadPP(self
, False), info_dict
))
3164 info_dict
['__write_download_archive'] = self
.params
.get('force_write_download_archive')
3167 info_dict
.setdefault('__postprocessors', [])
3170 def existing_video_file(*filepaths
):
3171 ext
= info_dict
.get('ext')
3172 converted
= lambda file: replace_extension(file, self
.params
.get('final_ext') or ext
, ext
)
3173 file = self
.existing_file(itertools
.chain(*zip(map(converted
, filepaths
), filepaths
)),
3174 default_overwrite
=False)
3176 info_dict
['ext'] = os
.path
.splitext(file)[1][1:]
3179 fd
, success
= None, True
3180 if info_dict
.get('protocol') or info_dict
.get('url'):
3181 fd
= get_suitable_downloader(info_dict
, self
.params
, to_stdout
=temp_filename
== '-')
3182 if fd
is not FFmpegFD
and 'no-direct-merge' not in self
.params
['compat_opts'] and (
3183 info_dict
.get('section_start') or info_dict
.get('section_end')):
3184 msg
= ('This format cannot be partially downloaded' if FFmpegFD
.available()
3185 else 'You have requested downloading the video partially, but ffmpeg is not installed')
3186 self
.report_error(f
'{msg}. Aborting')
3189 if info_dict
.get('requested_formats') is not None:
3190 old_ext
= info_dict
['ext']
3191 if self
.params
.get('merge_output_format') is None:
3192 if (info_dict
['ext'] == 'webm'
3193 and info_dict
.get('thumbnails')
3194 # check with type instead of pp_key, __name__, or isinstance
3195 # since we dont want any custom PPs to trigger this
3196 and any(type(pp
) == EmbedThumbnailPP
for pp
in self
._pps
['post_process'])): # noqa: E721
3197 info_dict
['ext'] = 'mkv'
3198 self
.report_warning(
3199 'webm doesn\'t support embedding a thumbnail, mkv will be used')
3200 new_ext
= info_dict
['ext']
3202 def correct_ext(filename
, ext
=new_ext
):
3205 filename_real_ext
= os
.path
.splitext(filename
)[1][1:]
3207 os
.path
.splitext(filename
)[0]
3208 if filename_real_ext
in (old_ext
, new_ext
)
3210 return f
'{filename_wo_ext}.{ext}'
3212 # Ensure filename always has a correct extension for successful merge
3213 full_filename
= correct_ext(full_filename
)
3214 temp_filename
= correct_ext(temp_filename
)
3215 dl_filename
= existing_video_file(full_filename
, temp_filename
)
3217 info_dict
['__real_download'] = False
3218 # NOTE: Copy so that original format dicts are not modified
3219 info_dict
['requested_formats'] = list(map(dict, info_dict
['requested_formats']))
3221 merger
= FFmpegMergerPP(self
)
3223 if dl_filename
is not None:
3224 self
.report_file_already_downloaded(dl_filename
)
3226 for f
in info_dict
['requested_formats'] if fd
!= FFmpegFD
else []:
3227 f
['filepath'] = fname
= prepend_extension(
3228 correct_ext(temp_filename
, info_dict
['ext']),
3229 'f%s' % f
['format_id'], info_dict
['ext'])
3230 downloaded
.append(fname
)
3231 info_dict
['url'] = '\n'.join(f
['url'] for f
in info_dict
['requested_formats'])
3232 success
, real_download
= self
.dl(temp_filename
, info_dict
)
3233 info_dict
['__real_download'] = real_download
3235 if self
.params
.get('allow_unplayable_formats'):
3236 self
.report_warning(
3237 'You have requested merging of multiple formats '
3238 'while also allowing unplayable formats to be downloaded. '
3239 'The formats won\'t be merged to prevent data corruption.')
3240 elif not merger
.available
:
3241 msg
= 'You have requested merging of multiple formats but ffmpeg is not installed'
3242 if not self
.params
.get('ignoreerrors'):
3243 self
.report_error(f
'{msg}. Aborting due to --abort-on-error')
3245 self
.report_warning(f
'{msg}. The formats won\'t be merged')
3247 if temp_filename
== '-':
3248 reason
= ('using a downloader other than ffmpeg' if FFmpegFD
.can_merge_formats(info_dict
, self
.params
)
3249 else 'but the formats are incompatible for simultaneous download' if merger
.available
3250 else 'but ffmpeg is not installed')
3251 self
.report_warning(
3252 f
'You have requested downloading multiple formats to stdout {reason}. '
3253 'The formats will be streamed one after the other')
3254 fname
= temp_filename
3255 for f
in info_dict
['requested_formats']:
3256 new_info
= dict(info_dict
)
3257 del new_info
['requested_formats']
3259 if temp_filename
!= '-':
3260 fname
= prepend_extension(
3261 correct_ext(temp_filename
, new_info
['ext']),
3262 'f%s' % f
['format_id'], new_info
['ext'])
3263 if not self
._ensure
_dir
_exists
(fname
):
3265 f
['filepath'] = fname
3266 downloaded
.append(fname
)
3267 partial_success
, real_download
= self
.dl(fname
, new_info
)
3268 info_dict
['__real_download'] = info_dict
['__real_download'] or real_download
3269 success
= success
and partial_success
3271 if downloaded
and merger
.available
and not self
.params
.get('allow_unplayable_formats'):
3272 info_dict
['__postprocessors'].append(merger
)
3273 info_dict
['__files_to_merge'] = downloaded
3274 # Even if there were no downloads, it is being merged only now
3275 info_dict
['__real_download'] = True
3277 for file in downloaded
:
3278 files_to_move
[file] = None
3280 # Just a single file
3281 dl_filename
= existing_video_file(full_filename
, temp_filename
)
3282 if dl_filename
is None or dl_filename
== temp_filename
:
3283 # dl_filename == temp_filename could mean that the file was partially downloaded with --no-part.
3284 # So we should try to resume the download
3285 success
, real_download
= self
.dl(temp_filename
, info_dict
)
3286 info_dict
['__real_download'] = real_download
3288 self
.report_file_already_downloaded(dl_filename
)
3290 dl_filename
= dl_filename
or temp_filename
3291 info_dict
['__finaldir'] = os
.path
.dirname(os
.path
.abspath(encodeFilename(full_filename
)))
3293 except network_exceptions
as err
:
3294 self
.report_error('unable to download video data: %s' % error_to_compat_str(err
))
3296 except OSError as err
:
3297 raise UnavailableVideoError(err
)
3298 except (ContentTooShortError
, ) as err
:
3299 self
.report_error(f
'content too short (expected {err.expected} bytes and served {err.downloaded})')
3302 self
._raise
_pending
_errors
(info_dict
)
3303 if success
and full_filename
!= '-':
3307 fixup_policy
= self
.params
.get('fixup')
3308 vid
= info_dict
['id']
3310 if fixup_policy
in ('ignore', 'never'):
3312 elif fixup_policy
== 'warn':
3314 elif fixup_policy
!= 'force':
3315 assert fixup_policy
in ('detect_or_warn', None)
3316 if not info_dict
.get('__real_download'):
3319 def ffmpeg_fixup(cndn
, msg
, cls
):
3320 if not (do_fixup
and cndn
):
3322 elif do_fixup
== 'warn':
3323 self
.report_warning(f
'{vid}: {msg}')
3327 info_dict
['__postprocessors'].append(pp
)
3329 self
.report_warning(f
'{vid}: {msg}. Install ffmpeg to fix this automatically')
3331 stretched_ratio
= info_dict
.get('stretched_ratio')
3332 ffmpeg_fixup(stretched_ratio
not in (1, None),
3333 f
'Non-uniform pixel ratio {stretched_ratio}',
3334 FFmpegFixupStretchedPP
)
3336 downloader
= get_suitable_downloader(info_dict
, self
.params
) if 'protocol' in info_dict
else None
3337 downloader
= downloader
.FD_NAME
if downloader
else None
3339 ext
= info_dict
.get('ext')
3340 postprocessed_by_ffmpeg
= info_dict
.get('requested_formats') or any((
3341 isinstance(pp
, FFmpegVideoConvertorPP
)
3342 and resolve_recode_mapping(ext
, pp
.mapping
)[0] not in (ext
, None)
3343 ) for pp
in self
._pps
['post_process'])
3345 if not postprocessed_by_ffmpeg
:
3346 ffmpeg_fixup(ext
== 'm4a' and info_dict
.get('container') == 'm4a_dash',
3347 'writing DASH m4a. Only some players support this container',
3349 ffmpeg_fixup(downloader
== 'hlsnative' and not self
.params
.get('hls_use_mpegts')
3350 or info_dict
.get('is_live') and self
.params
.get('hls_use_mpegts') is None,
3351 'Possible MPEG-TS in MP4 container or malformed AAC timestamps',
3353 ffmpeg_fixup(info_dict
.get('is_live') and downloader
== 'dashsegments',
3354 'Possible duplicate MOOV atoms', FFmpegFixupDuplicateMoovPP
)
3356 ffmpeg_fixup(downloader
== 'web_socket_fragment', 'Malformed timestamps detected', FFmpegFixupTimestampPP
)
3357 ffmpeg_fixup(downloader
== 'web_socket_fragment', 'Malformed duration detected', FFmpegFixupDurationPP
)
3361 replace_info_dict(self
.post_process(dl_filename
, info_dict
, files_to_move
))
3362 except PostProcessingError
as err
:
3363 self
.report_error('Postprocessing: %s' % str(err
))
3366 for ph
in self
._post
_hooks
:
3367 ph(info_dict
['filepath'])
3368 except Exception as err
:
3369 self
.report_error('post hooks: %s' % str(err
))
3371 info_dict
['__write_download_archive'] = True
3373 assert info_dict
is original_infodict
# Make sure the info_dict was modified in-place
3374 if self
.params
.get('force_write_download_archive'):
3375 info_dict
['__write_download_archive'] = True
3376 check_max_downloads()
3378 def __download_wrapper(self
, func
):
3379 @functools.wraps(func
)
3380 def wrapper(*args
, **kwargs
):
3382 res
= func(*args
, **kwargs
)
3383 except UnavailableVideoError
as e
:
3384 self
.report_error(e
)
3385 except DownloadCancelled
as e
:
3386 self
.to_screen(f
'[info] {e}')
3387 if not self
.params
.get('break_per_url'):
3389 self
._num
_downloads
= 0
3391 if self
.params
.get('dump_single_json', False):
3392 self
.post_extract(res
)
3393 self
.to_stdout(json
.dumps(self
.sanitize_info(res
)))
3396 def download(self
, url_list
):
3397 """Download a given list of URLs."""
3398 url_list
= variadic(url_list
) # Passing a single URL is a common mistake
3399 outtmpl
= self
.params
['outtmpl']['default']
3400 if (len(url_list
) > 1
3402 and '%' not in outtmpl
3403 and self
.params
.get('max_downloads') != 1):
3404 raise SameFileError(outtmpl
)
3406 for url
in url_list
:
3407 self
.__download
_wrapper
(self
.extract_info
)(
3408 url
, force_generic_extractor
=self
.params
.get('force_generic_extractor', False))
3410 return self
._download
_retcode
3412 def download_with_info_file(self
, info_filename
):
3413 with contextlib
.closing(fileinput
.FileInput(
3414 [info_filename
], mode
='r',
3415 openhook
=fileinput
.hook_encoded('utf-8'))) as f
:
3416 # FileInput doesn't have a read method, we can't call json.load
3417 infos
= [self
.sanitize_info(info
, self
.params
.get('clean_infojson', True))
3418 for info
in variadic(json
.loads('\n'.join(f
)))]
3421 self
.__download
_wrapper
(self
.process_ie_result
)(info
, download
=True)
3422 except (DownloadError
, EntryNotInPlaylist
, ReExtractInfo
) as e
:
3423 if not isinstance(e
, EntryNotInPlaylist
):
3424 self
.to_stderr('\r')
3425 webpage_url
= info
.get('webpage_url')
3426 if webpage_url
is None:
3428 self
.report_warning(f
'The info failed to download: {e}; trying with URL {webpage_url}')
3429 self
.download([webpage_url
])
3430 return self
._download
_retcode
3433 def sanitize_info(info_dict
, remove_private_keys
=False):
3434 ''' Sanitize the infodict for converting to json '''
3435 if info_dict
is None:
3437 info_dict
.setdefault('epoch', int(time
.time()))
3438 info_dict
.setdefault('_type', 'video')
3439 info_dict
.setdefault('_version', {
3440 'version': __version__
,
3441 'current_git_head': current_git_head(),
3442 'release_git_head': RELEASE_GIT_HEAD
,
3443 'repository': REPOSITORY
,
3446 if remove_private_keys
:
3447 reject
= lambda k
, v
: v
is None or k
.startswith('__') or k
in {
3448 'requested_downloads', 'requested_formats', 'requested_subtitles', 'requested_entries',
3449 'entries', 'filepath', '_filename', 'filename', 'infojson_filename', 'original_url',
3450 'playlist_autonumber', '_format_sort_fields',
3453 reject
= lambda k
, v
: False
3456 if isinstance(obj
, dict):
3457 return {k: filter_fn(v) for k, v in obj.items() if not reject(k, v)}
3458 elif isinstance(obj
, (list, tuple, set, LazyList
)):
3459 return list(map(filter_fn
, obj
))
3460 elif obj
is None or isinstance(obj
, (str, int, float, bool)):
3465 return filter_fn(info_dict
)
3468 def filter_requested_info(info_dict
, actually_filter
=True):
3469 ''' Alias of sanitize_info for backward compatibility '''
3470 return YoutubeDL
.sanitize_info(info_dict
, actually_filter
)
3472 def _delete_downloaded_files(self
, *files_to_delete
, info
={}, msg
=None):
3473 for filename
in set(filter(None, files_to_delete
)):
3475 self
.to_screen(msg
% filename
)
3479 self
.report_warning(f
'Unable to delete file {filename}')
3480 if filename
in info
.get('__files_to_move', []): # NB: Delete even if None
3481 del info
['__files_to_move'][filename
]
3484 def post_extract(info_dict
):
3485 def actual_post_extract(info_dict
):
3486 if info_dict
.get('_type') in ('playlist', 'multi_video'):
3487 for video_dict
in info_dict
.get('entries', {}):
3488 actual_post_extract(video_dict
or {})
3491 post_extractor
= info_dict
.pop('__post_extractor', None) or (lambda: {})
3492 info_dict
.update(post_extractor())
3494 actual_post_extract(info_dict
or {})
3496 def run_pp(self
, pp
, infodict
):
3497 files_to_delete
= []
3498 if '__files_to_move' not in infodict
:
3499 infodict
['__files_to_move'] = {}
3501 files_to_delete
, infodict
= pp
.run(infodict
)
3502 except PostProcessingError
as e
:
3503 # Must be True and not 'only_download'
3504 if self
.params
.get('ignoreerrors') is True:
3505 self
.report_error(e
)
3509 if not files_to_delete
:
3511 if self
.params
.get('keepvideo', False):
3512 for f
in files_to_delete
:
3513 infodict
['__files_to_move'].setdefault(f
, '')
3515 self
._delete
_downloaded
_files
(
3516 *files_to_delete
, info
=infodict
, msg
='Deleting original file %s (pass -k to keep)')
3519 def run_all_pps(self
, key
, info
, *, additional_pps
=None):
3521 self
._forceprint
(key
, info
)
3522 for pp
in (additional_pps
or []) + self
._pps
[key
]:
3523 info
= self
.run_pp(pp
, info
)
3526 def pre_process(self
, ie_info
, key
='pre_process', files_to_move
=None):
3527 info
= dict(ie_info
)
3528 info
['__files_to_move'] = files_to_move
or {}
3530 info
= self
.run_all_pps(key
, info
)
3531 except PostProcessingError
as err
:
3532 msg
= f
'Preprocessing: {err}'
3533 info
.setdefault('__pending_error', msg
)
3534 self
.report_error(msg
, is_error
=False)
3535 return info
, info
.pop('__files_to_move', None)
3537 def post_process(self
, filename
, info
, files_to_move
=None):
3538 """Run all the postprocessors on the given file."""
3539 info
['filepath'] = filename
3540 info
['__files_to_move'] = files_to_move
or {}
3541 info
= self
.run_all_pps('post_process', info
, additional_pps
=info
.get('__postprocessors'))
3542 info
= self
.run_pp(MoveFilesAfterDownloadPP(self
), info
)
3543 del info
['__files_to_move']
3544 return self
.run_all_pps('after_move', info
)
3546 def _make_archive_id(self
, info_dict
):
3547 video_id
= info_dict
.get('id')
3550 # Future-proof against any change in case
3551 # and backwards compatibility with prior versions
3552 extractor
= info_dict
.get('extractor_key') or info_dict
.get('ie_key') # key in a playlist
3553 if extractor
is None:
3554 url
= str_or_none(info_dict
.get('url'))
3557 # Try to find matching extractor for the URL and take its ie_key
3558 for ie_key
, ie
in self
._ies
.items():
3559 if ie
.suitable(url
):
3564 return make_archive_id(extractor
, video_id
)
3566 def in_download_archive(self
, info_dict
):
3567 if not self
.archive
:
3570 vid_ids
= [self
._make
_archive
_id
(info_dict
)]
3571 vid_ids
.extend(info_dict
.get('_old_archive_ids') or [])
3572 return any(id_
in self
.archive
for id_
in vid_ids
)
3574 def record_download_archive(self
, info_dict
):
3575 fn
= self
.params
.get('download_archive')
3578 vid_id
= self
._make
_archive
_id
(info_dict
)
3581 self
.write_debug(f
'Adding to archive: {vid_id}')
3582 if is_path_like(fn
):
3583 with locked_file(fn
, 'a', encoding
='utf-8') as archive_file
:
3584 archive_file
.write(vid_id
+ '\n')
3585 self
.archive
.add(vid_id
)
3588 def format_resolution(format
, default
='unknown'):
3589 if format
.get('vcodec') == 'none' and format
.get('acodec') != 'none':
3591 if format
.get('resolution') is not None:
3592 return format
['resolution']
3593 if format
.get('width') and format
.get('height'):
3594 return '%dx%d' % (format
['width'], format
['height'])
3595 elif format
.get('height'):
3596 return '%sp' % format
['height']
3597 elif format
.get('width'):
3598 return '%dx?' % format
['width']
3601 def _list_format_headers(self
, *headers
):
3602 if self
.params
.get('listformats_table', True) is not False:
3603 return [self
._format
_out
(header
, self
.Styles
.HEADERS
) for header
in headers
]
3606 def _format_note(self
, fdict
):
3608 if fdict
.get('ext') in ['f4f', 'f4m']:
3609 res
+= '(unsupported)'
3610 if fdict
.get('language'):
3613 res
+= '[%s]' % fdict
['language']
3614 if fdict
.get('format_note') is not None:
3617 res
+= fdict
['format_note']
3618 if fdict
.get('tbr') is not None:
3621 res
+= '%4dk' % fdict
['tbr']
3622 if fdict
.get('container') is not None:
3625 res
+= '%s container' % fdict
['container']
3626 if (fdict
.get('vcodec') is not None
3627 and fdict
.get('vcodec') != 'none'):
3630 res
+= fdict
['vcodec']
3631 if fdict
.get('vbr') is not None:
3633 elif fdict
.get('vbr') is not None and fdict
.get('abr') is not None:
3635 if fdict
.get('vbr') is not None:
3636 res
+= '%4dk' % fdict
['vbr']
3637 if fdict
.get('fps') is not None:
3640 res
+= '%sfps' % fdict
['fps']
3641 if fdict
.get('acodec') is not None:
3644 if fdict
['acodec'] == 'none':
3647 res
+= '%-5s' % fdict
['acodec']
3648 elif fdict
.get('abr') is not None:
3652 if fdict
.get('abr') is not None:
3653 res
+= '@%3dk' % fdict
['abr']
3654 if fdict
.get('asr') is not None:
3655 res
+= ' (%5dHz)' % fdict
['asr']
3656 if fdict
.get('filesize') is not None:
3659 res
+= format_bytes(fdict
['filesize'])
3660 elif fdict
.get('filesize_approx') is not None:
3663 res
+= '~' + format_bytes(fdict
['filesize_approx'])
3666 def _get_formats(self
, info_dict
):
3667 if info_dict
.get('formats') is None:
3668 if info_dict
.get('url') and info_dict
.get('_type', 'video') == 'video':
3671 return info_dict
['formats']
3673 def render_formats_table(self
, info_dict
):
3674 formats
= self
._get
_formats
(info_dict
)
3677 if not self
.params
.get('listformats_table', True) is not False:
3680 format_field(f
, 'format_id'),
3681 format_field(f
, 'ext'),
3682 self
.format_resolution(f
),
3683 self
._format
_note
(f
)
3684 ] for f
in formats
if (f
.get('preference') or 0) >= -1000]
3685 return render_table(['format code', 'extension', 'resolution', 'note'], table
, extra_gap
=1)
3687 def simplified_codec(f
, field
):
3688 assert field
in ('acodec', 'vcodec')
3689 codec
= f
.get(field
, 'unknown')
3692 elif codec
!= 'none':
3693 return '.'.join(codec
.split('.')[:4])
3695 if field
== 'vcodec' and f
.get('acodec') == 'none':
3697 elif field
== 'acodec' and f
.get('vcodec') == 'none':
3699 return self
._format
_out
('audio only' if field
== 'vcodec' else 'video only',
3700 self
.Styles
.SUPPRESS
)
3702 delim
= self
._format
_out
('\u2502', self
.Styles
.DELIM
, '|', test_encoding
=True)
3705 self
._format
_out
(format_field(f
, 'format_id'), self
.Styles
.ID
),
3706 format_field(f
, 'ext'),
3707 format_field(f
, func
=self
.format_resolution
, ignore
=('audio only', 'images')),
3708 format_field(f
, 'fps', '\t%d', func
=round),
3709 format_field(f
, 'dynamic_range', '%s', ignore
=(None, 'SDR')).replace('HDR', ''),
3710 format_field(f
, 'audio_channels', '\t%s'),
3712 format_field(f
, 'filesize', ' \t%s', func
=format_bytes
)
3713 or format_field(f
, 'filesize_approx', '≈\t%s', func
=format_bytes
)
3714 or format_field(try_call(lambda: format_bytes(int(info_dict
['duration'] * f
['tbr'] * (1024 / 8)))),
3715 None, self
._format
_out
('~\t%s', self
.Styles
.SUPPRESS
))),
3716 format_field(f
, 'tbr', '\t%dk', func
=round),
3717 shorten_protocol_name(f
.get('protocol', '')),
3719 simplified_codec(f
, 'vcodec'),
3720 format_field(f
, 'vbr', '\t%dk', func
=round),
3721 simplified_codec(f
, 'acodec'),
3722 format_field(f
, 'abr', '\t%dk', func
=round),
3723 format_field(f
, 'asr', '\t%s', func
=format_decimal_suffix
),
3725 self
._format
_out
('UNSUPPORTED', 'light red') if f
.get('ext') in ('f4f', 'f4m') else None,
3726 self
._format
_out
('DRM', 'light red') if f
.get('has_drm') else None,
3727 format_field(f
, 'language', '[%s]'),
3728 join_nonempty(format_field(f
, 'format_note'),
3729 format_field(f
, 'container', ignore
=(None, f
.get('ext'))),
3732 ] for f
in formats
if f
.get('preference') is None or f
['preference'] >= -1000]
3733 header_line
= self
._list
_format
_headers
(
3734 'ID', 'EXT', 'RESOLUTION', '\tFPS', 'HDR', 'CH', delim
, '\tFILESIZE', '\tTBR', 'PROTO',
3735 delim
, 'VCODEC', '\tVBR', 'ACODEC', '\tABR', '\tASR', 'MORE INFO')
3737 return render_table(
3738 header_line
, table
, hide_empty
=True,
3739 delim
=self
._format
_out
('\u2500', self
.Styles
.DELIM
, '-', test_encoding
=True))
3741 def render_thumbnails_table(self
, info_dict
):
3742 thumbnails
= list(info_dict
.get('thumbnails') or [])
3745 return render_table(
3746 self
._list
_format
_headers
('ID', 'Width', 'Height', 'URL'),
3747 [[t
.get('id'), t
.get('width') or 'unknown', t
.get('height') or 'unknown', t
['url']] for t
in thumbnails
])
3749 def render_subtitles_table(self
, video_id
, subtitles
):
3750 def _row(lang
, formats
):
3751 exts
, names
= zip(*((f
['ext'], f
.get('name') or 'unknown') for f
in reversed(formats
)))
3752 if len(set(names
)) == 1:
3753 names
= [] if names
[0] == 'unknown' else names
[:1]
3754 return [lang
, ', '.join(names
), ', '.join(exts
)]
3758 return render_table(
3759 self
._list
_format
_headers
('Language', 'Name', 'Formats'),
3760 [_row(lang
, formats
) for lang
, formats
in subtitles
.items()],
3763 def __list_table(self
, video_id
, name
, func
, *args
):
3766 self
.to_screen(f
'{video_id} has no {name}')
3768 self
.to_screen(f
'[info] Available {name} for {video_id}:')
3769 self
.to_stdout(table
)
3771 def list_formats(self
, info_dict
):
3772 self
.__list
_table
(info_dict
['id'], 'formats', self
.render_formats_table
, info_dict
)
3774 def list_thumbnails(self
, info_dict
):
3775 self
.__list
_table
(info_dict
['id'], 'thumbnails', self
.render_thumbnails_table
, info_dict
)
3777 def list_subtitles(self
, video_id
, subtitles
, name
='subtitles'):
3778 self
.__list
_table
(video_id
, name
, self
.render_subtitles_table
, video_id
, subtitles
)
3780 def urlopen(self
, req
):
3781 """ Start an HTTP download """
3782 if isinstance(req
, str):
3783 req
= sanitized_Request(req
)
3784 return self
._opener
.open(req
, timeout
=self
._socket
_timeout
)
3786 def print_debug_header(self
):
3787 if not self
.params
.get('verbose'):
3790 from . import _IN_CLI
# Must be delayed import
3792 # These imports can be slow. So import them only as needed
3793 from .extractor
.extractors
import _LAZY_LOADER
3794 from .extractor
.extractors
import (
3795 _PLUGIN_CLASSES
as plugin_ies
,
3796 _PLUGIN_OVERRIDES
as plugin_ie_overrides
3799 def get_encoding(stream
):
3800 ret
= str(getattr(stream
, 'encoding', 'missing (%s)' % type(stream
).__name
__))
3801 additional_info
= []
3802 if os
.environ
.get('TERM', '').lower() == 'dumb':
3803 additional_info
.append('dumb')
3804 if not supports_terminal_sequences(stream
):
3805 from .utils
import WINDOWS_VT_MODE
# Must be imported locally
3806 additional_info
.append('No VT' if WINDOWS_VT_MODE
is False else 'No ANSI')
3808 ret
= f
'{ret} ({",".join(additional_info)})'
3811 encoding_str
= 'Encodings: locale %s, fs %s, pref %s, %s' % (
3812 locale
.getpreferredencoding(),
3813 sys
.getfilesystemencoding(),
3814 self
.get_encoding(),
3816 f
'{key} {get_encoding(stream)}' for key
, stream
in self
._out
_files
.items_
3817 if stream
is not None and key
!= 'console')
3820 logger
= self
.params
.get('logger')
3822 write_debug
= lambda msg
: logger
.debug(f
'[debug] {msg}')
3823 write_debug(encoding_str
)
3825 write_string(f
'[debug] {encoding_str}\n', encoding
=None)
3826 write_debug
= lambda msg
: self
._write
_string
(f
'[debug] {msg}\n')
3828 source
= detect_variant()
3829 if VARIANT
not in (None, 'pip'):
3832 write_debug(join_nonempty(
3833 f
'{"yt-dlp" if REPOSITORY == "yt-dlp/yt-dlp" else REPOSITORY} version',
3834 f
'{CHANNEL}@{__version__}',
3835 f
'[{RELEASE_GIT_HEAD[:9]}]' if RELEASE_GIT_HEAD
else '',
3836 '' if source
== 'unknown' else f
'({source})',
3837 '' if _IN_CLI
else 'API' if klass
== YoutubeDL
else f
'API:{self.__module__}.{klass.__qualname__}',
3841 write_debug(f
'params: {self.params}')
3843 if not _LAZY_LOADER
:
3844 if os
.environ
.get('YTDLP_NO_LAZY_EXTRACTORS'):
3845 write_debug('Lazy loading extractors is forcibly disabled')
3847 write_debug('Lazy loading extractors is disabled')
3848 if self
.params
['compat_opts']:
3849 write_debug('Compatibility options: %s' % ', '.join(self
.params
['compat_opts']))
3851 if current_git_head():
3852 write_debug(f
'Git HEAD: {current_git_head()}')
3853 write_debug(system_identifier())
3855 exe_versions
, ffmpeg_features
= FFmpegPostProcessor
.get_versions_and_features(self
)
3856 ffmpeg_features
= {key for key, val in ffmpeg_features.items() if val}
3858 exe_versions
['ffmpeg'] += ' (%s)' % ','.join(sorted(ffmpeg_features
))
3860 exe_versions
['rtmpdump'] = rtmpdump_version()
3861 exe_versions
['phantomjs'] = PhantomJSwrapper
._version
()
3862 exe_str
= ', '.join(
3863 f
'{exe} {v}' for exe
, v
in sorted(exe_versions
.items()) if v
3865 write_debug('exe versions: %s' % exe_str
)
3867 from .compat
.compat_utils
import get_package_info
3868 from .dependencies
import available_dependencies
3870 write_debug('Optional libraries: %s' % (', '.join(sorted({
3871 join_nonempty(*get_package_info(m
)) for m
in available_dependencies
.values()
3874 self
._setup
_opener
()
3876 for handler
in self
._opener
.handlers
:
3877 if hasattr(handler
, 'proxies'):
3878 proxy_map
.update(handler
.proxies
)
3879 write_debug(f
'Proxy map: {proxy_map}')
3881 for plugin_type
, plugins
in {'Extractor': plugin_ies, 'Post-Processor': plugin_pps}
.items():
3882 display_list
= ['%s%s' % (
3883 klass
.__name
__, '' if klass
.__name
__ == name
else f
' as {name}')
3884 for name
, klass
in plugins
.items()]
3885 if plugin_type
== 'Extractor':
3886 display_list
.extend(f
'{plugins[-1].IE_NAME.partition("+")[2]} ({parent.__name__})'
3887 for parent
, plugins
in plugin_ie_overrides
.items())
3888 if not display_list
:
3890 write_debug(f
'{plugin_type} Plugins: {", ".join(sorted(display_list))}')
3892 plugin_dirs
= plugin_directories()
3894 write_debug(f
'Plugin directories: {plugin_dirs}')
3897 if False and self
.params
.get('call_home'):
3898 ipaddr
= self
.urlopen('https://yt-dl.org/ip').read().decode()
3899 write_debug('Public IP address: %s' % ipaddr
)
3900 latest_version
= self
.urlopen(
3901 'https://yt-dl.org/latest/version').read().decode()
3902 if version_tuple(latest_version
) > version_tuple(__version__
):
3903 self
.report_warning(
3904 'You are using an outdated version (newest version: %s)! '
3905 'See https://yt-dl.org/update if you need help updating.' %
3908 def _setup_opener(self
):
3909 if hasattr(self
, '_opener'):
3911 timeout_val
= self
.params
.get('socket_timeout')
3912 self
._socket
_timeout
= 20 if timeout_val
is None else float(timeout_val
)
3914 opts_cookiesfrombrowser
= self
.params
.get('cookiesfrombrowser')
3915 opts_cookiefile
= self
.params
.get('cookiefile')
3916 opts_proxy
= self
.params
.get('proxy')
3918 self
.cookiejar
= load_cookies(opts_cookiefile
, opts_cookiesfrombrowser
, self
)
3920 cookie_processor
= YoutubeDLCookieProcessor(self
.cookiejar
)
3921 if opts_proxy
is not None:
3922 if opts_proxy
== '':
3925 proxies
= {'http': opts_proxy, 'https': opts_proxy}
3927 proxies
= urllib
.request
.getproxies()
3928 # Set HTTPS proxy to HTTP one if given (https://github.com/ytdl-org/youtube-dl/issues/805)
3929 if 'http' in proxies
and 'https' not in proxies
:
3930 proxies
['https'] = proxies
['http']
3931 proxy_handler
= PerRequestProxyHandler(proxies
)
3933 debuglevel
= 1 if self
.params
.get('debug_printtraffic') else 0
3934 https_handler
= make_HTTPS_handler(self
.params
, debuglevel
=debuglevel
)
3935 ydlh
= YoutubeDLHandler(self
.params
, debuglevel
=debuglevel
)
3936 redirect_handler
= YoutubeDLRedirectHandler()
3937 data_handler
= urllib
.request
.DataHandler()
3939 # When passing our own FileHandler instance, build_opener won't add the
3940 # default FileHandler and allows us to disable the file protocol, which
3941 # can be used for malicious purposes (see
3942 # https://github.com/ytdl-org/youtube-dl/issues/8227)
3943 file_handler
= urllib
.request
.FileHandler()
3945 if not self
.params
.get('enable_file_urls'):
3946 def file_open(*args
, **kwargs
):
3947 raise urllib
.error
.URLError(
3948 'file:// URLs are explicitly disabled in yt-dlp for security reasons. '
3949 'Use --enable-file-urls to enable at your own risk.')
3950 file_handler
.file_open
= file_open
3952 opener
= urllib
.request
.build_opener(
3953 proxy_handler
, https_handler
, cookie_processor
, ydlh
, redirect_handler
, data_handler
, file_handler
)
3955 # Delete the default user-agent header, which would otherwise apply in
3956 # cases where our custom HTTP handler doesn't come into play
3957 # (See https://github.com/ytdl-org/youtube-dl/issues/1309 for details)
3958 opener
.addheaders
= []
3959 self
._opener
= opener
3961 def encode(self
, s
):
3962 if isinstance(s
, bytes):
3963 return s
# Already encoded
3966 return s
.encode(self
.get_encoding())
3967 except UnicodeEncodeError as err
:
3968 err
.reason
= err
.reason
+ '. Check your system encoding configuration or use the --encoding option.'
3971 def get_encoding(self
):
3972 encoding
= self
.params
.get('encoding')
3973 if encoding
is None:
3974 encoding
= preferredencoding()
3977 def _write_info_json(self
, label
, ie_result
, infofn
, overwrite
=None):
3978 ''' Write infojson and returns True = written, 'exists' = Already exists, False = skip, None = error '''
3979 if overwrite
is None:
3980 overwrite
= self
.params
.get('overwrites', True)
3981 if not self
.params
.get('writeinfojson'):
3984 self
.write_debug(f
'Skipping writing {label} infojson')
3986 elif not self
._ensure
_dir
_exists
(infofn
):
3988 elif not overwrite
and os
.path
.exists(infofn
):
3989 self
.to_screen(f
'[info] {label.title()} metadata is already present')
3992 self
.to_screen(f
'[info] Writing {label} metadata as JSON to: {infofn}')
3994 write_json_file(self
.sanitize_info(ie_result
, self
.params
.get('clean_infojson', True)), infofn
)
3997 self
.report_error(f
'Cannot write {label} metadata to JSON file {infofn}')
4000 def _write_description(self
, label
, ie_result
, descfn
):
4001 ''' Write description and returns True = written, False = skip, None = error '''
4002 if not self
.params
.get('writedescription'):
4005 self
.write_debug(f
'Skipping writing {label} description')
4007 elif not self
._ensure
_dir
_exists
(descfn
):
4009 elif not self
.params
.get('overwrites', True) and os
.path
.exists(descfn
):
4010 self
.to_screen(f
'[info] {label.title()} description is already present')
4011 elif ie_result
.get('description') is None:
4012 self
.to_screen(f
'[info] There\'s no {label} description to write')
4016 self
.to_screen(f
'[info] Writing {label} description to: {descfn}')
4017 with open(encodeFilename(descfn
), 'w', encoding
='utf-8') as descfile
:
4018 descfile
.write(ie_result
['description'])
4020 self
.report_error(f
'Cannot write {label} description file {descfn}')
4024 def _write_subtitles(self
, info_dict
, filename
):
4025 ''' Write subtitles to file and return list of (sub_filename, final_sub_filename); or None if error'''
4027 subtitles
= info_dict
.get('requested_subtitles')
4028 if not (self
.params
.get('writesubtitles') or self
.params
.get('writeautomaticsub')):
4029 # subtitles download errors are already managed as troubles in relevant IE
4030 # that way it will silently go on when used with unsupporting IE
4033 self
.to_screen('[info] There are no subtitles for the requested languages')
4035 sub_filename_base
= self
.prepare_filename(info_dict
, 'subtitle')
4036 if not sub_filename_base
:
4037 self
.to_screen('[info] Skipping writing video subtitles')
4040 for sub_lang
, sub_info
in subtitles
.items():
4041 sub_format
= sub_info
['ext']
4042 sub_filename
= subtitles_filename(filename
, sub_lang
, sub_format
, info_dict
.get('ext'))
4043 sub_filename_final
= subtitles_filename(sub_filename_base
, sub_lang
, sub_format
, info_dict
.get('ext'))
4044 existing_sub
= self
.existing_file((sub_filename_final
, sub_filename
))
4046 self
.to_screen(f
'[info] Video subtitle {sub_lang}.{sub_format} is already present')
4047 sub_info
['filepath'] = existing_sub
4048 ret
.append((existing_sub
, sub_filename_final
))
4051 self
.to_screen(f
'[info] Writing video subtitles to: {sub_filename}')
4052 if sub_info
.get('data') is not None:
4054 # Use newline='' to prevent conversion of newline characters
4055 # See https://github.com/ytdl-org/youtube-dl/issues/10268
4056 with open(sub_filename
, 'w', encoding
='utf-8', newline
='') as subfile
:
4057 subfile
.write(sub_info
['data'])
4058 sub_info
['filepath'] = sub_filename
4059 ret
.append((sub_filename
, sub_filename_final
))
4062 self
.report_error(f
'Cannot write video subtitles file {sub_filename}')
4066 sub_copy
= sub_info
.copy()
4067 sub_copy
.setdefault('http_headers', info_dict
.get('http_headers'))
4068 self
.dl(sub_filename
, sub_copy
, subtitle
=True)
4069 sub_info
['filepath'] = sub_filename
4070 ret
.append((sub_filename
, sub_filename_final
))
4071 except (DownloadError
, ExtractorError
, IOError, OSError, ValueError) + network_exceptions
as err
:
4072 msg
= f
'Unable to download video subtitles for {sub_lang!r}: {err}'
4073 if self
.params
.get('ignoreerrors') is not True: # False or 'only_download'
4074 if not self
.params
.get('ignoreerrors'):
4075 self
.report_error(msg
)
4076 raise DownloadError(msg
)
4077 self
.report_warning(msg
)
4080 def _write_thumbnails(self
, label
, info_dict
, filename
, thumb_filename_base
=None):
4081 ''' Write thumbnails to file and return list of (thumb_filename, final_thumb_filename) '''
4082 write_all
= self
.params
.get('write_all_thumbnails', False)
4083 thumbnails
, ret
= [], []
4084 if write_all
or self
.params
.get('writethumbnail', False):
4085 thumbnails
= info_dict
.get('thumbnails') or []
4087 self
.to_screen(f
'[info] There are no {label} thumbnails to download')
4089 multiple
= write_all
and len(thumbnails
) > 1
4091 if thumb_filename_base
is None:
4092 thumb_filename_base
= filename
4093 if thumbnails
and not thumb_filename_base
:
4094 self
.write_debug(f
'Skipping writing {label} thumbnail')
4097 for idx
, t
in list(enumerate(thumbnails
))[::-1]:
4098 thumb_ext
= (f
'{t["id"]}.' if multiple
else '') + determine_ext(t
['url'], 'jpg')
4099 thumb_display_id
= f
'{label} thumbnail {t["id"]}'
4100 thumb_filename
= replace_extension(filename
, thumb_ext
, info_dict
.get('ext'))
4101 thumb_filename_final
= replace_extension(thumb_filename_base
, thumb_ext
, info_dict
.get('ext'))
4103 existing_thumb
= self
.existing_file((thumb_filename_final
, thumb_filename
))
4105 self
.to_screen('[info] %s is already present' % (
4106 thumb_display_id
if multiple
else f
'{label} thumbnail').capitalize())
4107 t
['filepath'] = existing_thumb
4108 ret
.append((existing_thumb
, thumb_filename_final
))
4110 self
.to_screen(f
'[info] Downloading {thumb_display_id} ...')
4112 uf
= self
.urlopen(sanitized_Request(t
['url'], headers
=t
.get('http_headers', {})))
4113 self
.to_screen(f
'[info] Writing {thumb_display_id} to: {thumb_filename}')
4114 with open(encodeFilename(thumb_filename
), 'wb') as thumbf
:
4115 shutil
.copyfileobj(uf
, thumbf
)
4116 ret
.append((thumb_filename
, thumb_filename_final
))
4117 t
['filepath'] = thumb_filename
4118 except network_exceptions
as err
:
4119 if isinstance(err
, urllib
.error
.HTTPError
) and err
.code
== 404:
4120 self
.to_screen(f
'[info] {thumb_display_id.title()} does not exist')
4122 self
.report_warning(f
'Unable to download {thumb_display_id}: {err}')
4124 if ret
and not write_all
: