24 from string
import ascii_letters
26 from .cache
import Cache
27 from .compat
import compat_os_name
, compat_shlex_quote
28 from .cookies
import load_cookies
29 from .downloader
import FFmpegFD
, get_suitable_downloader
, shorten_protocol_name
30 from .downloader
.rtmp
import rtmpdump_version
31 from .extractor
import gen_extractor_classes
, get_info_extractor
32 from .extractor
.common
import UnsupportedURLIE
33 from .extractor
.openload
import PhantomJSwrapper
34 from .minicurses
import format_text
35 from .postprocessor
import _PLUGIN_CLASSES
as plugin_postprocessors
36 from .postprocessor
import (
38 FFmpegFixupDuplicateMoovPP
,
39 FFmpegFixupDurationPP
,
42 FFmpegFixupStretchedPP
,
43 FFmpegFixupTimestampPP
,
46 FFmpegVideoConvertorPP
,
47 MoveFilesAfterDownloadPP
,
50 from .postprocessor
.ffmpeg
import resolve_mapping
as resolve_recode_mapping
51 from .update
import detect_variant
77 PerRequestProxyHandler
,
84 UnavailableVideoError
,
86 YoutubeDLCookieProcessor
,
88 YoutubeDLRedirectHandler
,
103 format_decimal_suffix
,
119 orderedSet_from_options
,
123 register_socks_protocols
,
124 remove_terminal_sequences
,
135 supports_terminal_sequences
,
145 windows_enable_vt_mode
,
149 from .version
import RELEASE_GIT_HEAD
, VARIANT
, __version__
151 if compat_os_name
== 'nt':
158 YoutubeDL objects are the ones responsible of downloading the
159 actual video file and writing it to disk if the user has requested
160 it, among some other tasks. In most cases there should be one per
161 program. As, given a video URL, the downloader doesn't know how to
162 extract all the needed information, task that InfoExtractors do, it
163 has to pass the URL to one of them.
165 For this, YoutubeDL objects have a method that allows
166 InfoExtractors to be registered in a given order. When it is passed
167 a URL, the YoutubeDL object handles it to the first InfoExtractor it
168 finds that reports being able to handle it. The InfoExtractor extracts
169 all the information about the video or videos the URL refers to, and
170 YoutubeDL process the extracted information, possibly using a File
171 Downloader to download the video.
173 YoutubeDL objects accept a lot of parameters. In order not to saturate
174 the object constructor with arguments, it receives a dictionary of
175 options instead. These options are available through the params
176 attribute for the InfoExtractors to use. The YoutubeDL also
177 registers itself as the downloader in charge for the InfoExtractors
178 that are added to it, so this is a "mutual registration".
182 username: Username for authentication purposes.
183 password: Password for authentication purposes.
184 videopassword: Password for accessing a video.
185 ap_mso: Adobe Pass multiple-system operator identifier.
186 ap_username: Multiple-system operator account username.
187 ap_password: Multiple-system operator account password.
188 usenetrc: Use netrc for authentication instead.
189 verbose: Print additional info to stdout.
190 quiet: Do not print messages to stdout.
191 no_warnings: Do not print out anything for warnings.
192 forceprint: A dict with keys WHEN mapped to a list of templates to
193 print to stdout. The allowed keys are video or any of the
194 items in utils.POSTPROCESS_WHEN.
195 For compatibility, a single list is also accepted
196 print_to_file: A dict with keys WHEN (same as forceprint) mapped to
197 a list of tuples with (template, filename)
198 forcejson: Force printing info_dict as JSON.
199 dump_single_json: Force printing the info_dict of the whole playlist
200 (or video) as a single JSON line.
201 force_write_download_archive: Force writing download archive regardless
202 of 'skip_download' or 'simulate'.
203 simulate: Do not download the video files. If unset (or None),
204 simulate only if listsubtitles, listformats or list_thumbnails is used
205 format: Video format code. see "FORMAT SELECTION" for more details.
206 You can also pass a function. The function takes 'ctx' as
207 argument and returns the formats to download.
208 See "build_format_selector" for an implementation
209 allow_unplayable_formats: Allow unplayable formats to be extracted and downloaded.
210 ignore_no_formats_error: Ignore "No video formats" error. Usefull for
211 extracting metadata even if the video is not actually
212 available for download (experimental)
213 format_sort: A list of fields by which to sort the video formats.
214 See "Sorting Formats" for more details.
215 format_sort_force: Force the given format_sort. see "Sorting Formats"
217 prefer_free_formats: Whether to prefer video formats with free containers
218 over non-free ones of same quality.
219 allow_multiple_video_streams: Allow multiple video streams to be merged
221 allow_multiple_audio_streams: Allow multiple audio streams to be merged
223 check_formats Whether to test if the formats are downloadable.
224 Can be True (check all), False (check none),
225 'selected' (check selected formats),
226 or None (check only if requested by extractor)
227 paths: Dictionary of output paths. The allowed keys are 'home'
228 'temp' and the keys of OUTTMPL_TYPES (in utils.py)
229 outtmpl: Dictionary of templates for output names. Allowed keys
230 are 'default' and the keys of OUTTMPL_TYPES (in utils.py).
231 For compatibility with youtube-dl, a single string can also be used
232 outtmpl_na_placeholder: Placeholder for unavailable meta fields.
233 restrictfilenames: Do not allow "&" and spaces in file names
234 trim_file_name: Limit length of filename (extension excluded)
235 windowsfilenames: Force the filenames to be windows compatible
236 ignoreerrors: Do not stop on download/postprocessing errors.
237 Can be 'only_download' to ignore only download errors.
238 Default is 'only_download' for CLI, but False for API
239 skip_playlist_after_errors: Number of allowed failures until the rest of
240 the playlist is skipped
241 allowed_extractors: List of regexes to match against extractor names that are allowed
242 overwrites: Overwrite all video and metadata files if True,
243 overwrite only non-video files if None
244 and don't overwrite any file if False
245 For compatibility with youtube-dl,
246 "nooverwrites" may also be used instead
247 playlist_items: Specific indices of playlist to download.
248 playlistrandom: Download playlist items in random order.
249 lazy_playlist: Process playlist entries as they are received.
250 matchtitle: Download only matching titles.
251 rejecttitle: Reject downloads for matching titles.
252 logger: Log messages to a logging.Logger instance.
253 logtostderr: Log messages to stderr instead of stdout.
254 consoletitle: Display progress in console window's titlebar.
255 writedescription: Write the video description to a .description file
256 writeinfojson: Write the video description to a .info.json file
257 clean_infojson: Remove private fields from the infojson
258 getcomments: Extract video comments. This will not be written to disk
259 unless writeinfojson is also given
260 writeannotations: Write the video annotations to a .annotations.xml file
261 writethumbnail: Write the thumbnail image to a file
262 allow_playlist_files: Whether to write playlists' description, infojson etc
263 also to disk when using the 'write*' options
264 write_all_thumbnails: Write all thumbnail formats to files
265 writelink: Write an internet shortcut file, depending on the
266 current platform (.url/.webloc/.desktop)
267 writeurllink: Write a Windows internet shortcut file (.url)
268 writewebloclink: Write a macOS internet shortcut file (.webloc)
269 writedesktoplink: Write a Linux internet shortcut file (.desktop)
270 writesubtitles: Write the video subtitles to a file
271 writeautomaticsub: Write the automatically generated subtitles to a file
272 listsubtitles: Lists all available subtitles for the video
273 subtitlesformat: The format code for subtitles
274 subtitleslangs: List of languages of the subtitles to download (can be regex).
275 The list may contain "all" to refer to all the available
276 subtitles. The language can be prefixed with a "-" to
277 exclude it from the requested languages, e.g. ['all', '-live_chat']
278 keepvideo: Keep the video file after post-processing
279 daterange: A DateRange object, download only if the upload_date is in the range.
280 skip_download: Skip the actual download of the video file
281 cachedir: Location of the cache files in the filesystem.
282 False to disable filesystem cache.
283 noplaylist: Download single video instead of a playlist if in doubt.
284 age_limit: An integer representing the user's age in years.
285 Unsuitable videos for the given age are skipped.
286 min_views: An integer representing the minimum view count the video
287 must have in order to not be skipped.
288 Videos without view count information are always
289 downloaded. None for no limit.
290 max_views: An integer representing the maximum view count.
291 Videos that are more popular than that are not
293 Videos without view count information are always
294 downloaded. None for no limit.
295 download_archive: File name of a file where all downloads are recorded.
296 Videos already present in the file are not downloaded
298 break_on_existing: Stop the download process after attempting to download a
299 file that is in the archive.
300 break_on_reject: Stop the download process when encountering a video that
301 has been filtered out.
302 break_per_url: Whether break_on_reject and break_on_existing
303 should act on each input URL as opposed to for the entire queue
304 cookiefile: File name or text stream from where cookies should be read and dumped to
305 cookiesfrombrowser: A tuple containing the name of the browser, the profile
306 name/path from where cookies are loaded, and the name of the
307 keyring, e.g. ('chrome', ) or ('vivaldi', 'default', 'BASICTEXT')
308 legacyserverconnect: Explicitly allow HTTPS connection to servers that do not
309 support RFC 5746 secure renegotiation
310 nocheckcertificate: Do not verify SSL certificates
311 client_certificate: Path to client certificate file in PEM format. May include the private key
312 client_certificate_key: Path to private key file for client certificate
313 client_certificate_password: Password for client certificate private key, if encrypted.
314 If not provided and the key is encrypted, yt-dlp will ask interactively
315 prefer_insecure: Use HTTP instead of HTTPS to retrieve information.
316 (Only supported by some extractors)
317 http_headers: A dictionary of custom headers to be used for all requests
318 proxy: URL of the proxy server to use
319 geo_verification_proxy: URL of the proxy to use for IP address verification
320 on geo-restricted sites.
321 socket_timeout: Time to wait for unresponsive hosts, in seconds
322 bidi_workaround: Work around buggy terminals without bidirectional text
323 support, using fridibi
324 debug_printtraffic:Print out sent and received HTTP traffic
325 default_search: Prepend this string if an input url is not valid.
326 'auto' for elaborate guessing
327 encoding: Use this encoding instead of the system-specified.
328 extract_flat: Whether to resolve and process url_results further
329 * False: Always process (default)
330 * True: Never process
331 * 'in_playlist': Do not process inside playlist/multi_video
332 * 'discard': Always process, but don't return the result
333 from inside playlist/multi_video
334 * 'discard_in_playlist': Same as "discard", but only for
335 playlists (not multi_video)
336 wait_for_video: If given, wait for scheduled streams to become available.
337 The value should be a tuple containing the range
338 (min_secs, max_secs) to wait between retries
339 postprocessors: A list of dictionaries, each with an entry
340 * key: The name of the postprocessor. See
341 yt_dlp/postprocessor/__init__.py for a list.
342 * when: When to run the postprocessor. Allowed values are
343 the entries of utils.POSTPROCESS_WHEN
344 Assumed to be 'post_process' if not given
345 progress_hooks: A list of functions that get called on download
346 progress, with a dictionary with the entries
347 * status: One of "downloading", "error", or "finished".
348 Check this first and ignore unknown values.
349 * info_dict: The extracted info_dict
351 If status is one of "downloading", or "finished", the
352 following properties may also be present:
353 * filename: The final filename (always present)
354 * tmpfilename: The filename we're currently writing to
355 * downloaded_bytes: Bytes on disk
356 * total_bytes: Size of the whole file, None if unknown
357 * total_bytes_estimate: Guess of the eventual file size,
359 * elapsed: The number of seconds since download started.
360 * eta: The estimated time in seconds, None if unknown
361 * speed: The download speed in bytes/second, None if
363 * fragment_index: The counter of the currently
364 downloaded video fragment.
365 * fragment_count: The number of fragments (= individual
366 files that will be merged)
368 Progress hooks are guaranteed to be called at least once
369 (with status "finished") if the download is successful.
370 postprocessor_hooks: A list of functions that get called on postprocessing
371 progress, with a dictionary with the entries
372 * status: One of "started", "processing", or "finished".
373 Check this first and ignore unknown values.
374 * postprocessor: Name of the postprocessor
375 * info_dict: The extracted info_dict
377 Progress hooks are guaranteed to be called at least twice
378 (with status "started" and "finished") if the processing is successful.
379 merge_output_format: "/" separated list of extensions to use when merging formats.
380 final_ext: Expected final extension; used to detect when the file was
381 already downloaded and converted
382 fixup: Automatically correct known faults of the file.
384 - "never": do nothing
385 - "warn": only emit a warning
386 - "detect_or_warn": check whether we can do anything
387 about it, warn otherwise (default)
388 source_address: Client-side IP address to bind to.
389 sleep_interval_requests: Number of seconds to sleep between requests
391 sleep_interval: Number of seconds to sleep before each download when
392 used alone or a lower bound of a range for randomized
393 sleep before each download (minimum possible number
394 of seconds to sleep) when used along with
396 max_sleep_interval:Upper bound of a range for randomized sleep before each
397 download (maximum possible number of seconds to sleep).
398 Must only be used along with sleep_interval.
399 Actual sleep time will be a random float from range
400 [sleep_interval; max_sleep_interval].
401 sleep_interval_subtitles: Number of seconds to sleep before each subtitle download
402 listformats: Print an overview of available video formats and exit.
403 list_thumbnails: Print a table of all thumbnails and exit.
404 match_filter: A function that gets called for every video with the signature
405 (info_dict, *, incomplete: bool) -> Optional[str]
406 For backward compatibility with youtube-dl, the signature
407 (info_dict) -> Optional[str] is also allowed.
408 - If it returns a message, the video is ignored.
409 - If it returns None, the video is downloaded.
410 - If it returns utils.NO_DEFAULT, the user is interactively
411 asked whether to download the video.
412 match_filter_func in utils.py is one example for this.
413 no_color: Do not emit color codes in output.
414 geo_bypass: Bypass geographic restriction via faking X-Forwarded-For
417 Two-letter ISO 3166-2 country code that will be used for
418 explicit geographic restriction bypassing via faking
419 X-Forwarded-For HTTP header
421 IP range in CIDR notation that will be used similarly to
423 external_downloader: A dictionary of protocol keys and the executable of the
424 external downloader to use for it. The allowed protocols
425 are default|http|ftp|m3u8|dash|rtsp|rtmp|mms.
426 Set the value to 'native' to use the native downloader
427 compat_opts: Compatibility options. See "Differences in default behavior".
428 The following options do not work when used through the API:
429 filename, abort-on-error, multistreams, no-live-chat, format-sort
430 no-clean-infojson, no-playlist-metafiles, no-keep-subs, no-attach-info-json.
431 Refer __init__.py for their implementation
432 progress_template: Dictionary of templates for progress outputs.
433 Allowed keys are 'download', 'postprocess',
434 'download-title' (console title) and 'postprocess-title'.
435 The template is mapped on a dictionary with keys 'progress' and 'info'
436 retry_sleep_functions: Dictionary of functions that takes the number of attempts
437 as argument and returns the time to sleep in seconds.
438 Allowed keys are 'http', 'fragment', 'file_access'
439 download_ranges: A callback function that gets called for every video with
440 the signature (info_dict, ydl) -> Iterable[Section].
441 Only the returned sections will be downloaded.
442 Each Section is a dict with the following keys:
443 * start_time: Start time of the section in seconds
444 * end_time: End time of the section in seconds
445 * title: Section title (Optional)
446 * index: Section number (Optional)
447 force_keyframes_at_cuts: Re-encode the video when downloading ranges to get precise cuts
448 noprogress: Do not print the progress bar
449 live_from_start: Whether to download livestreams videos from the start
451 The following parameters are not used by YoutubeDL itself, they are used by
452 the downloader (see yt_dlp/downloader/common.py):
453 nopart, updatetime, buffersize, ratelimit, throttledratelimit, min_filesize,
454 max_filesize, test, noresizebuffer, retries, file_access_retries, fragment_retries,
455 continuedl, xattr_set_filesize, hls_use_mpegts, http_chunk_size,
456 external_downloader_args, concurrent_fragment_downloads.
458 The following options are used by the post processors:
459 ffmpeg_location: Location of the ffmpeg/avconv binary; either the path
460 to the binary or its containing directory.
461 postprocessor_args: A dictionary of postprocessor/executable keys (in lower case)
462 and a list of additional command-line arguments for the
463 postprocessor/executable. The dict can also have "PP+EXE" keys
464 which are used when the given exe is used by the given PP.
465 Use 'default' as the name for arguments to passed to all PP
466 For compatibility with youtube-dl, a single list of args
469 The following options are used by the extractors:
470 extractor_retries: Number of times to retry for known errors
471 dynamic_mpd: Whether to process dynamic DASH manifests (default: True)
472 hls_split_discontinuity: Split HLS playlists to different formats at
473 discontinuities such as ad breaks (default: False)
474 extractor_args: A dictionary of arguments to be passed to the extractors.
475 See "EXTRACTOR ARGUMENTS" for details.
476 E.g. {'youtube': {'skip': ['dash', 'hls']}}
477 mark_watched: Mark videos watched (even with --simulate). Only for YouTube
479 The following options are deprecated and may be removed in the future:
481 force_generic_extractor: Force downloader to use the generic extractor
482 - Use allowed_extractors = ['generic', 'default']
483 playliststart: - Use playlist_items
484 Playlist item to start at.
485 playlistend: - Use playlist_items
486 Playlist item to end at.
487 playlistreverse: - Use playlist_items
488 Download playlist items in reverse order.
489 forceurl: - Use forceprint
490 Force printing final URL.
491 forcetitle: - Use forceprint
492 Force printing title.
493 forceid: - Use forceprint
495 forcethumbnail: - Use forceprint
496 Force printing thumbnail URL.
497 forcedescription: - Use forceprint
498 Force printing description.
499 forcefilename: - Use forceprint
500 Force printing final filename.
501 forceduration: - Use forceprint
502 Force printing duration.
503 allsubtitles: - Use subtitleslangs = ['all']
504 Downloads all the subtitles of the video
505 (requires writesubtitles or writeautomaticsub)
506 include_ads: - Doesn't work
508 call_home: - Not implemented
509 Boolean, true iff we are allowed to contact the
510 yt-dlp servers for debugging.
511 post_hooks: - Register a custom postprocessor
512 A list of functions that get called as the final step
513 for each video file, after all postprocessors have been
514 called. The filename will be passed as the only argument.
515 hls_prefer_native: - Use external_downloader = {'m3u8': 'native'} or {'m3u8': 'ffmpeg'}.
516 Use the native HLS downloader instead of ffmpeg/avconv
517 if True, otherwise use ffmpeg/avconv if False, otherwise
518 use downloader suggested by extractor if None.
519 prefer_ffmpeg: - avconv support is deprecated
520 If False, use avconv instead of ffmpeg if both are available,
521 otherwise prefer ffmpeg.
522 youtube_include_dash_manifest: - Use extractor_args
523 If True (default), DASH manifests and related
524 data will be downloaded and processed by extractor.
525 You can reduce network I/O by disabling it if you don't
526 care about DASH. (only for youtube)
527 youtube_include_hls_manifest: - Use extractor_args
528 If True (default), HLS manifests and related
529 data will be downloaded and processed by extractor.
530 You can reduce network I/O by disabling it if you don't
531 care about HLS. (only for youtube)
535 'width', 'height', 'asr', 'audio_channels', 'fps',
536 'tbr', 'abr', 'vbr', 'filesize', 'filesize_approx',
537 'timestamp', 'release_timestamp',
538 'duration', 'view_count', 'like_count', 'dislike_count', 'repost_count',
539 'average_rating', 'comment_count', 'age_limit',
540 'start_time', 'end_time',
541 'chapter_number', 'season_number', 'episode_number',
542 'track_number', 'disc_number', 'release_year',
546 # NB: Keep in sync with the docstring of extractor/common.py
547 'url', 'manifest_url', 'manifest_stream_number', 'ext', 'format', 'format_id', 'format_note',
548 'width', 'height', 'resolution', 'dynamic_range', 'tbr', 'abr', 'acodec', 'asr', 'audio_channels',
549 'vbr', 'fps', 'vcodec', 'container', 'filesize', 'filesize_approx',
550 'player_url', 'protocol', 'fragment_base_url', 'fragments', 'is_from_start',
551 'preference', 'language', 'language_preference', 'quality', 'source_preference',
552 'http_headers', 'stretched_ratio', 'no_resume', 'has_drm', 'downloader_options',
553 'page_url', 'app', 'play_path', 'tc_url', 'flash_version', 'rtmp_live', 'rtmp_conn', 'rtmp_protocol', 'rtmp_real_time'
555 _format_selection_exts
= {
556 'audio': set(MEDIA_EXTENSIONS
.common_audio
),
557 'video': set(MEDIA_EXTENSIONS
.common_video
+ ('3gp', )),
558 'storyboards': set(MEDIA_EXTENSIONS
.storyboards
),
561 def __init__(self
, params
=None, auto_init
=True):
562 """Create a FileDownloader object with the given options.
563 @param auto_init Whether to load the default extractors and print header (if verbose).
564 Set to 'no_verbose_header' to not print the header
570 self
._ies
_instances
= {}
571 self
._pps
= {k: [] for k in POSTPROCESS_WHEN}
572 self
._printed
_messages
= set()
573 self
._first
_webpage
_request
= True
574 self
._post
_hooks
= []
575 self
._progress
_hooks
= []
576 self
._postprocessor
_hooks
= []
577 self
._download
_retcode
= 0
578 self
._num
_downloads
= 0
580 self
._playlist
_level
= 0
581 self
._playlist
_urls
= set()
582 self
.cache
= Cache(self
)
584 windows_enable_vt_mode()
585 stdout
= sys
.stderr
if self
.params
.get('logtostderr') else sys
.stdout
586 self
._out
_files
= Namespace(
589 screen
=sys
.stderr
if self
.params
.get('quiet') else stdout
,
590 console
=None if compat_os_name
== 'nt' else next(
591 filter(supports_terminal_sequences
, (sys
.stderr
, sys
.stdout
)), None)
593 self
._allow
_colors
= Namespace(**{
594 type_
: not self
.params
.get('no_color') and supports_terminal_sequences(stream
)
595 for type_
, stream
in self
._out
_files
.items_
if type_
!= 'console'
598 # The code is left like this to be reused for future deprecations
599 MIN_SUPPORTED
, MIN_RECOMMENDED
= (3, 7), (3, 7)
600 current_version
= sys
.version_info
[:2]
601 if current_version
< MIN_RECOMMENDED
:
602 msg
= ('Support for Python version %d.%d has been deprecated. '
603 'See https://github.com/yt-dlp/yt-dlp/issues/3764 for more details.'
604 '\n You will no longer receive updates on this version')
605 if current_version
< MIN_SUPPORTED
:
606 msg
= 'Python version %d.%d is no longer supported'
607 self
.deprecation_warning(
608 f
'{msg}! Please update to Python %d.%d or above' % (*current_version
, *MIN_RECOMMENDED
))
610 if self
.params
.get('allow_unplayable_formats'):
612 f
'You have asked for {self._format_err("UNPLAYABLE", self.Styles.EMPHASIS)} formats to be listed/downloaded. '
613 'This is a developer option intended for debugging. \n'
614 ' If you experience any issues while using this option, '
615 f
'{self._format_err("DO NOT", self.Styles.ERROR)} open a bug report')
617 def check_deprecated(param
, option
, suggestion
):
618 if self
.params
.get(param
) is not None:
619 self
.report_warning(f
'{option} is deprecated. Use {suggestion} instead')
623 if check_deprecated('cn_verification_proxy', '--cn-verification-proxy', '--geo-verification-proxy'):
624 if self
.params
.get('geo_verification_proxy') is None:
625 self
.params
['geo_verification_proxy'] = self
.params
['cn_verification_proxy']
627 check_deprecated('autonumber', '--auto-number', '-o "%(autonumber)s-%(title)s.%(ext)s"')
628 check_deprecated('usetitle', '--title', '-o "%(title)s-%(id)s.%(ext)s"')
629 check_deprecated('useid', '--id', '-o "%(id)s.%(ext)s"')
631 for msg
in self
.params
.get('_warnings', []):
632 self
.report_warning(msg
)
633 for msg
in self
.params
.get('_deprecation_warnings', []):
634 self
.deprecation_warning(msg
)
636 self
.params
['compat_opts'] = set(self
.params
.get('compat_opts', ()))
637 if 'list-formats' in self
.params
['compat_opts']:
638 self
.params
['listformats_table'] = False
640 if 'overwrites' not in self
.params
and self
.params
.get('nooverwrites') is not None:
641 # nooverwrites was unnecessarily changed to overwrites
642 # in 0c3d0f51778b153f65c21906031c2e091fcfb641
643 # This ensures compatibility with both keys
644 self
.params
['overwrites'] = not self
.params
['nooverwrites']
645 elif self
.params
.get('overwrites') is None:
646 self
.params
.pop('overwrites', None)
648 self
.params
['nooverwrites'] = not self
.params
['overwrites']
650 self
.params
.setdefault('forceprint', {})
651 self
.params
.setdefault('print_to_file', {})
653 # Compatibility with older syntax
654 if not isinstance(params
['forceprint'], dict):
655 self
.params
['forceprint'] = {'video': params['forceprint']}
657 if self
.params
.get('bidi_workaround', False):
660 master
, slave
= pty
.openpty()
661 width
= shutil
.get_terminal_size().columns
662 width_args
= [] if width
is None else ['-w', str(width
)]
663 sp_kwargs
= {'stdin': subprocess.PIPE, 'stdout': slave, 'stderr': self._out_files.error}
665 self
._output
_process
= Popen(['bidiv'] + width_args
, **sp_kwargs
)
667 self
._output
_process
= Popen(['fribidi', '-c', 'UTF-8'] + width_args
, **sp_kwargs
)
668 self
._output
_channel
= os
.fdopen(master
, 'rb')
669 except OSError as ose
:
670 if ose
.errno
== errno
.ENOENT
:
672 'Could not find fribidi executable, ignoring --bidi-workaround. '
673 'Make sure that fribidi is an executable file in one of the directories in your $PATH.')
678 if auto_init
!= 'no_verbose_header':
679 self
.print_debug_header()
680 self
.add_default_info_extractors()
682 if (sys
.platform
!= 'win32'
683 and sys
.getfilesystemencoding() in ['ascii', 'ANSI_X3.4-1968']
684 and not self
.params
.get('restrictfilenames', False)):
685 # Unicode filesystem API will throw errors (#1474, #13027)
687 'Assuming --restrict-filenames since file system encoding '
688 'cannot encode all characters. '
689 'Set the LC_ALL environment variable to fix this.')
690 self
.params
['restrictfilenames'] = True
692 self
._parse
_outtmpl
()
694 # Creating format selector here allows us to catch syntax errors before the extraction
695 self
.format_selector
= (
696 self
.params
.get('format') if self
.params
.get('format') in (None, '-')
697 else self
.params
['format'] if callable(self
.params
['format'])
698 else self
.build_format_selector(self
.params
['format']))
700 # Set http_headers defaults according to std_headers
701 self
.params
['http_headers'] = merge_headers(std_headers
, self
.params
.get('http_headers', {}))
704 'post_hooks': self
.add_post_hook
,
705 'progress_hooks': self
.add_progress_hook
,
706 'postprocessor_hooks': self
.add_postprocessor_hook
,
708 for opt
, fn
in hooks
.items():
709 for ph
in self
.params
.get(opt
, []):
712 for pp_def_raw
in self
.params
.get('postprocessors', []):
713 pp_def
= dict(pp_def_raw
)
714 when
= pp_def
.pop('when', 'post_process')
715 self
.add_post_processor(
716 get_postprocessor(pp_def
.pop('key'))(self
, **pp_def
),
720 register_socks_protocols()
722 def preload_download_archive(fn
):
723 """Preload the archive, if any is specified"""
726 self
.write_debug(f
'Loading archive file {fn!r}')
728 with locked_file(fn
, 'r', encoding
='utf-8') as archive_file
:
729 for line
in archive_file
:
730 self
.archive
.add(line
.strip())
731 except OSError as ioe
:
732 if ioe
.errno
!= errno
.ENOENT
:
738 preload_download_archive(self
.params
.get('download_archive'))
740 def warn_if_short_id(self
, argv
):
741 # short YouTube ID starting with dash?
743 i
for i
, a
in enumerate(argv
)
744 if re
.match(r
'^-[0-9A-Za-z_-]{10}$', a
)]
748 + [a
for i
, a
in enumerate(argv
) if i
not in idxs
]
749 + ['--'] + [argv
[i
] for i
in idxs
]
752 'Long argument string detected. '
753 'Use -- to separate parameters and URLs, like this:\n%s' %
754 args_to_str(correct_argv
))
756 def add_info_extractor(self
, ie
):
757 """Add an InfoExtractor object to the end of the list."""
759 self
._ies
[ie_key
] = ie
760 if not isinstance(ie
, type):
761 self
._ies
_instances
[ie_key
] = ie
762 ie
.set_downloader(self
)
764 def get_info_extractor(self
, ie_key
):
766 Get an instance of an IE with name ie_key, it will try to get one from
767 the _ies list, if there's no instance it will create a new one and add
768 it to the extractor list.
770 ie
= self
._ies
_instances
.get(ie_key
)
772 ie
= get_info_extractor(ie_key
)()
773 self
.add_info_extractor(ie
)
776 def add_default_info_extractors(self
):
778 Add the InfoExtractors returned by gen_extractors to the end of the list
780 all_ies
= {ie.IE_NAME.lower(): ie for ie in gen_extractor_classes()}
781 all_ies
['end'] = UnsupportedURLIE()
783 ie_names
= orderedSet_from_options(
784 self
.params
.get('allowed_extractors', ['default']), {
785 'all': list(all_ies
),
786 'default': [name
for name
, ie
in all_ies
.items() if ie
._ENABLED
],
788 except re
.error
as e
:
789 raise ValueError(f
'Wrong regex for allowed_extractors: {e.pattern}')
790 for name
in ie_names
:
791 self
.add_info_extractor(all_ies
[name
])
792 self
.write_debug(f
'Loaded {len(ie_names)} extractors')
794 def add_post_processor(self
, pp
, when
='post_process'):
795 """Add a PostProcessor object to the end of the chain."""
796 assert when
in POSTPROCESS_WHEN
, f
'Invalid when={when}'
797 self
._pps
[when
].append(pp
)
798 pp
.set_downloader(self
)
800 def add_post_hook(self
, ph
):
801 """Add the post hook"""
802 self
._post
_hooks
.append(ph
)
804 def add_progress_hook(self
, ph
):
805 """Add the download progress hook"""
806 self
._progress
_hooks
.append(ph
)
808 def add_postprocessor_hook(self
, ph
):
809 """Add the postprocessing progress hook"""
810 self
._postprocessor
_hooks
.append(ph
)
811 for pps
in self
._pps
.values():
813 pp
.add_progress_hook(ph
)
815 def _bidi_workaround(self
, message
):
816 if not hasattr(self
, '_output_channel'):
819 assert hasattr(self
, '_output_process')
820 assert isinstance(message
, str)
821 line_count
= message
.count('\n') + 1
822 self
._output
_process
.stdin
.write((message
+ '\n').encode())
823 self
._output
_process
.stdin
.flush()
824 res
= ''.join(self
._output
_channel
.readline().decode()
825 for _
in range(line_count
))
826 return res
[:-len('\n')]
828 def _write_string(self
, message
, out
=None, only_once
=False):
830 if message
in self
._printed
_messages
:
832 self
._printed
_messages
.add(message
)
833 write_string(message
, out
=out
, encoding
=self
.params
.get('encoding'))
835 def to_stdout(self
, message
, skip_eol
=False, quiet
=None):
836 """Print message to stdout"""
837 if quiet
is not None:
838 self
.deprecation_warning('"YoutubeDL.to_stdout" no longer accepts the argument quiet. Use "YoutubeDL.to_screen" instead')
839 if skip_eol
is not False:
840 self
.deprecation_warning('"YoutubeDL.to_stdout" no longer accepts the argument skip_eol. Use "YoutubeDL.to_screen" instead')
841 self
._write
_string
(f
'{self._bidi_workaround(message)}\n', self
._out
_files
.out
)
843 def to_screen(self
, message
, skip_eol
=False, quiet
=None):
844 """Print message to screen if not in quiet mode"""
845 if self
.params
.get('logger'):
846 self
.params
['logger'].debug(message
)
848 if (self
.params
.get('quiet') if quiet
is None else quiet
) and not self
.params
.get('verbose'):
851 '%s%s' % (self
._bidi
_workaround
(message
), ('' if skip_eol
else '\n')),
852 self
._out
_files
.screen
)
854 def to_stderr(self
, message
, only_once
=False):
855 """Print message to stderr"""
856 assert isinstance(message
, str)
857 if self
.params
.get('logger'):
858 self
.params
['logger'].error(message
)
860 self
._write
_string
(f
'{self._bidi_workaround(message)}\n', self
._out
_files
.error
, only_once
=only_once
)
862 def _send_console_code(self
, code
):
863 if compat_os_name
== 'nt' or not self
._out
_files
.console
:
865 self
._write
_string
(code
, self
._out
_files
.console
)
867 def to_console_title(self
, message
):
868 if not self
.params
.get('consoletitle', False):
870 message
= remove_terminal_sequences(message
)
871 if compat_os_name
== 'nt':
872 if ctypes
.windll
.kernel32
.GetConsoleWindow():
873 # c_wchar_p() might not be necessary if `message` is
874 # already of type unicode()
875 ctypes
.windll
.kernel32
.SetConsoleTitleW(ctypes
.c_wchar_p(message
))
877 self
._send
_console
_code
(f
'\033]0;{message}\007')
879 def save_console_title(self
):
880 if not self
.params
.get('consoletitle') or self
.params
.get('simulate'):
882 self
._send
_console
_code
('\033[22;0t') # Save the title on stack
884 def restore_console_title(self
):
885 if not self
.params
.get('consoletitle') or self
.params
.get('simulate'):
887 self
._send
_console
_code
('\033[23;0t') # Restore the title from stack
890 self
.save_console_title()
893 def __exit__(self
, *args
):
894 self
.restore_console_title()
896 if self
.params
.get('cookiefile') is not None:
897 self
.cookiejar
.save(ignore_discard
=True, ignore_expires
=True)
899 def trouble(self
, message
=None, tb
=None, is_error
=True):
900 """Determine action to take when a download problem appears.
902 Depending on if the downloader has been configured to ignore
903 download errors or not, this method may throw an exception or
904 not when errors are found, after printing the message.
906 @param tb If given, is additional traceback information
907 @param is_error Whether to raise error according to ignorerrors
909 if message
is not None:
910 self
.to_stderr(message
)
911 if self
.params
.get('verbose'):
913 if sys
.exc_info()[0]: # if .trouble has been called from an except block
915 if hasattr(sys
.exc_info()[1], 'exc_info') and sys
.exc_info()[1].exc_info
[0]:
916 tb
+= ''.join(traceback
.format_exception(*sys
.exc_info()[1].exc_info
))
917 tb
+= encode_compat_str(traceback
.format_exc())
919 tb_data
= traceback
.format_list(traceback
.extract_stack())
920 tb
= ''.join(tb_data
)
925 if not self
.params
.get('ignoreerrors'):
926 if sys
.exc_info()[0] and hasattr(sys
.exc_info()[1], 'exc_info') and sys
.exc_info()[1].exc_info
[0]:
927 exc_info
= sys
.exc_info()[1].exc_info
929 exc_info
= sys
.exc_info()
930 raise DownloadError(message
, exc_info
)
931 self
._download
_retcode
= 1
935 EMPHASIS
='light blue',
941 SUPPRESS
='light black',
944 def _format_text(self
, handle
, allow_colors
, text
, f
, fallback
=None, *, test_encoding
=False):
948 # handle.encoding can be None. See https://github.com/yt-dlp/yt-dlp/issues/2711
949 encoding
= self
.params
.get('encoding') or getattr(handle
, 'encoding', None) or 'ascii'
950 text
= text
.encode(encoding
, 'ignore').decode(encoding
)
951 if fallback
is not None and text
!= original_text
:
953 return format_text(text
, f
) if allow_colors
else text
if fallback
is None else fallback
955 def _format_out(self
, *args
, **kwargs
):
956 return self
._format
_text
(self
._out
_files
.out
, self
._allow
_colors
.out
, *args
, **kwargs
)
958 def _format_screen(self
, *args
, **kwargs
):
959 return self
._format
_text
(self
._out
_files
.screen
, self
._allow
_colors
.screen
, *args
, **kwargs
)
961 def _format_err(self
, *args
, **kwargs
):
962 return self
._format
_text
(self
._out
_files
.error
, self
._allow
_colors
.error
, *args
, **kwargs
)
964 def report_warning(self
, message
, only_once
=False):
966 Print the message to stderr, it will be prefixed with 'WARNING:'
967 If stderr is a tty file the 'WARNING:' will be colored
969 if self
.params
.get('logger') is not None:
970 self
.params
['logger'].warning(message
)
972 if self
.params
.get('no_warnings'):
974 self
.to_stderr(f
'{self._format_err("WARNING:", self.Styles.WARNING)} {message}', only_once
)
976 def deprecation_warning(self
, message
):
977 if self
.params
.get('logger') is not None:
978 self
.params
['logger'].warning(f
'DeprecationWarning: {message}')
980 self
.to_stderr(f
'{self._format_err("DeprecationWarning:", self.Styles.ERROR)} {message}', True)
982 def report_error(self
, message
, *args
, **kwargs
):
984 Do the same as trouble, but prefixes the message with 'ERROR:', colored
985 in red if stderr is a tty file.
987 self
.trouble(f
'{self._format_err("ERROR:", self.Styles.ERROR)} {message}', *args
, **kwargs
)
989 def write_debug(self
, message
, only_once
=False):
990 '''Log debug message or Print message to stderr'''
991 if not self
.params
.get('verbose', False):
993 message
= f
'[debug] {message}'
994 if self
.params
.get('logger'):
995 self
.params
['logger'].debug(message
)
997 self
.to_stderr(message
, only_once
)
999 def report_file_already_downloaded(self
, file_name
):
1000 """Report file has already been fully downloaded."""
1002 self
.to_screen('[download] %s has already been downloaded' % file_name
)
1003 except UnicodeEncodeError:
1004 self
.to_screen('[download] The file has already been downloaded')
1006 def report_file_delete(self
, file_name
):
1007 """Report that existing file will be deleted."""
1009 self
.to_screen('Deleting existing file %s' % file_name
)
1010 except UnicodeEncodeError:
1011 self
.to_screen('Deleting existing file')
1013 def raise_no_formats(self
, info
, forced
=False, *, msg
=None):
1014 has_drm
= info
.get('_has_drm')
1015 ignored
, expected
= self
.params
.get('ignore_no_formats_error'), bool(msg
)
1016 msg
= msg
or has_drm
and 'This video is DRM protected' or 'No video formats found!'
1017 if forced
or not ignored
:
1018 raise ExtractorError(msg
, video_id
=info
['id'], ie
=info
['extractor'],
1019 expected
=has_drm
or ignored
or expected
)
1021 self
.report_warning(msg
)
1023 def parse_outtmpl(self
):
1024 self
.deprecation_warning('"YoutubeDL.parse_outtmpl" is deprecated and may be removed in a future version')
1025 self
._parse
_outtmpl
()
1026 return self
.params
['outtmpl']
1028 def _parse_outtmpl(self
):
1030 if self
.params
.get('restrictfilenames'): # Remove spaces in the default template
1031 sanitize
= lambda x
: x
.replace(' - ', ' ').replace(' ', '-')
1033 outtmpl
= self
.params
.setdefault('outtmpl', {})
1034 if not isinstance(outtmpl
, dict):
1035 self
.params
['outtmpl'] = outtmpl
= {'default': outtmpl}
1036 outtmpl
.update({k: sanitize(v) for k, v in DEFAULT_OUTTMPL.items() if outtmpl.get(k) is None}
)
1038 def get_output_path(self
, dir_type
='', filename
=None):
1039 paths
= self
.params
.get('paths', {})
1040 assert isinstance(paths
, dict)
1041 path
= os
.path
.join(
1042 expand_path(paths
.get('home', '').strip()),
1043 expand_path(paths
.get(dir_type
, '').strip()) if dir_type
else '',
1045 return sanitize_path(path
, force
=self
.params
.get('windowsfilenames'))
1048 def _outtmpl_expandpath(outtmpl
):
1049 # expand_path translates '%%' into '%' and '$$' into '$'
1050 # correspondingly that is not what we want since we need to keep
1051 # '%%' intact for template dict substitution step. Working around
1052 # with boundary-alike separator hack.
1053 sep
= ''.join([random
.choice(ascii_letters
) for _
in range(32)])
1054 outtmpl
= outtmpl
.replace('%%', f
'%{sep}%').replace('$$', f
'${sep}$')
1056 # outtmpl should be expand_path'ed before template dict substitution
1057 # because meta fields may contain env variables we don't want to
1058 # be expanded. E.g. for outtmpl "%(title)s.%(ext)s" and
1059 # title "Hello $PATH", we don't want `$PATH` to be expanded.
1060 return expand_path(outtmpl
).replace(sep
, '')
1063 def escape_outtmpl(outtmpl
):
1064 ''' Escape any remaining strings like %s, %abc% etc. '''
1066 STR_FORMAT_RE_TMPL
.format('', '(?![%(\0])'),
1067 lambda mobj
: ('' if mobj
.group('has_key') else '%') + mobj
.group(0),
1071 def validate_outtmpl(cls
, outtmpl
):
1072 ''' @return None or Exception object '''
1074 STR_FORMAT_RE_TMPL
.format('[^)]*', '[ljhqBUDS]'),
1075 lambda mobj
: f
'{mobj.group(0)[:-1]}s',
1076 cls
._outtmpl
_expandpath
(outtmpl
))
1078 cls
.escape_outtmpl(outtmpl
) % collections
.defaultdict(int)
1080 except ValueError as err
:
1084 def _copy_infodict(info_dict
):
1085 info_dict
= dict(info_dict
)
1086 info_dict
.pop('__postprocessors', None)
1087 info_dict
.pop('__pending_error', None)
1090 def prepare_outtmpl(self
, outtmpl
, info_dict
, sanitize
=False):
1091 """ Make the outtmpl and info_dict suitable for substitution: ydl.escape_outtmpl(outtmpl) % info_dict
1092 @param sanitize Whether to sanitize the output as a filename.
1093 For backward compatibility, a function can also be passed
1096 info_dict
.setdefault('epoch', int(time
.time())) # keep epoch consistent once set
1098 info_dict
= self
._copy
_infodict
(info_dict
)
1099 info_dict
['duration_string'] = ( # %(duration>%H-%M-%S)s is wrong if duration > 24hrs
1100 formatSeconds(info_dict
['duration'], '-' if sanitize
else ':')
1101 if info_dict
.get('duration', None) is not None
1103 info_dict
['autonumber'] = int(self
.params
.get('autonumber_start', 1) - 1 + self
._num
_downloads
)
1104 info_dict
['video_autonumber'] = self
._num
_videos
1105 if info_dict
.get('resolution') is None:
1106 info_dict
['resolution'] = self
.format_resolution(info_dict
, default
=None)
1108 # For fields playlist_index, playlist_autonumber and autonumber convert all occurrences
1109 # of %(field)s to %(field)0Nd for backward compatibility
1110 field_size_compat_map
= {
1111 'playlist_index': number_of_digits(info_dict
.get('__last_playlist_index') or 0),
1112 'playlist_autonumber': number_of_digits(info_dict
.get('n_entries') or 0),
1113 'autonumber': self
.params
.get('autonumber_size') or 5,
1117 EXTERNAL_FORMAT_RE
= re
.compile(STR_FORMAT_RE_TMPL
.format('[^)]*', f
'[{STR_FORMAT_TYPES}ljhqBUDS]'))
1122 # Field is of the form key1.key2...
1123 # where keys (except first) can be string, int or slice
1124 FIELD_RE
= r
'\w*(?:\.(?:\w+|{num}|{num}?(?::{num}?){{1,2}}))*'.format(num
=r
'(?:-?\d+)')
1125 MATH_FIELD_RE
= rf
'(?:{FIELD_RE}|-?{NUMBER_RE})'
1126 MATH_OPERATORS_RE
= r
'(?:%s)' % '|'.join(map(re
.escape
, MATH_FUNCTIONS
.keys()))
1127 INTERNAL_FORMAT_RE
= re
.compile(rf
'''(?x)
1129 (?P<fields>{FIELD_RE})
1130 (?P<maths>(?:{MATH_OPERATORS_RE}{MATH_FIELD_RE})*)
1131 (?:>(?P<strf_format>.+?))?
1133 (?P<alternate>(?<!\\),[^|&)]+)?
1134 (?:&(?P<replacement>.*?))?
1135 (?:\|(?P<default>.*?))?
1138 def _traverse_infodict(k
):
1142 return traverse_obj(info_dict
, k
, is_user_input
=True, traverse_string
=True)
1144 def get_value(mdict
):
1146 value
= _traverse_infodict(mdict
['fields'])
1149 value
= float_or_none(value
)
1150 if value
is not None:
1153 offset_key
= mdict
['maths']
1155 value
= float_or_none(value
)
1159 MATH_FIELD_RE
if operator
else MATH_OPERATORS_RE
,
1160 offset_key
).group(0)
1161 offset_key
= offset_key
[len(item
):]
1162 if operator
is None:
1163 operator
= MATH_FUNCTIONS
[item
]
1165 item
, multiplier
= (item
[1:], -1) if item
[0] == '-' else (item
, 1)
1166 offset
= float_or_none(item
)
1168 offset
= float_or_none(_traverse_infodict(item
))
1170 value
= operator(value
, multiplier
* offset
)
1171 except (TypeError, ZeroDivisionError):
1174 # Datetime formatting
1175 if mdict
['strf_format']:
1176 value
= strftime_or_none(value
, mdict
['strf_format'].replace('\\,', ','))
1178 # XXX: Workaround for https://github.com/yt-dlp/yt-dlp/issues/4485
1179 if sanitize
and value
== '':
1183 na
= self
.params
.get('outtmpl_na_placeholder', 'NA')
1185 def filename_sanitizer(key
, value
, restricted
=self
.params
.get('restrictfilenames')):
1186 return sanitize_filename(str(value
), restricted
=restricted
, is_id
=(
1187 bool(re
.search(r
'(^|[_.])id(\.|$)', key
))
1188 if 'filename-sanitization' in self
.params
['compat_opts']
1191 sanitizer
= sanitize
if callable(sanitize
) else filename_sanitizer
1192 sanitize
= bool(sanitize
)
1194 def _dumpjson_default(obj
):
1195 if isinstance(obj
, (set, LazyList
)):
1199 def create_key(outer_mobj
):
1200 if not outer_mobj
.group('has_key'):
1201 return outer_mobj
.group(0)
1202 key
= outer_mobj
.group('key')
1203 mobj
= re
.match(INTERNAL_FORMAT_RE
, key
)
1204 initial_field
= mobj
.group('fields') if mobj
else ''
1205 value
, replacement
, default
= None, None, na
1207 mobj
= mobj
.groupdict()
1208 default
= mobj
['default'] if mobj
['default'] is not None else default
1209 value
= get_value(mobj
)
1210 replacement
= mobj
['replacement']
1211 if value
is None and mobj
['alternate']:
1212 mobj
= re
.match(INTERNAL_FORMAT_RE
, mobj
['remaining'][1:])
1216 fmt
= outer_mobj
.group('format')
1217 if fmt
== 's' and value
is not None and key
in field_size_compat_map
.keys():
1218 fmt
= f
'0{field_size_compat_map[key]:d}d'
1220 value
= default
if value
is None else value
if replacement
is None else replacement
1222 flags
= outer_mobj
.group('conversion') or ''
1223 str_fmt
= f
'{fmt[:-1]}s'
1224 if fmt
[-1] == 'l': # list
1225 delim
= '\n' if '#' in flags
else ', '
1226 value
, fmt
= delim
.join(map(str, variadic(value
, allowed_types
=(str, bytes)))), str_fmt
1227 elif fmt
[-1] == 'j': # json
1228 value
, fmt
= json
.dumps(value
, default
=_dumpjson_default
, indent
=4 if '#' in flags
else None), str_fmt
1229 elif fmt
[-1] == 'h': # html
1230 value
, fmt
= escapeHTML(value
), str_fmt
1231 elif fmt
[-1] == 'q': # quoted
1232 value
= map(str, variadic(value
) if '#' in flags
else [value
])
1233 value
, fmt
= ' '.join(map(compat_shlex_quote
, value
)), str_fmt
1234 elif fmt
[-1] == 'B': # bytes
1235 value
= f
'%{str_fmt}'.encode() % str(value
).encode()
1236 value
, fmt
= value
.decode('utf-8', 'ignore'), 's'
1237 elif fmt
[-1] == 'U': # unicode normalized
1238 value
, fmt
= unicodedata
.normalize(
1239 # "+" = compatibility equivalence, "#" = NFD
1240 'NF%s%s' % ('K' if '+' in flags
else '', 'D' if '#' in flags
else 'C'),
1242 elif fmt
[-1] == 'D': # decimal suffix
1243 num_fmt
, fmt
= fmt
[:-1].replace('#', ''), 's'
1244 value
= format_decimal_suffix(value
, f
'%{num_fmt}f%s' if num_fmt
else '%d%s',
1245 factor
=1024 if '#' in flags
else 1000)
1246 elif fmt
[-1] == 'S': # filename sanitization
1247 value
, fmt
= filename_sanitizer(initial_field
, value
, restricted
='#' in flags
), str_fmt
1248 elif fmt
[-1] == 'c':
1250 value
= str(value
)[0]
1253 elif fmt
[-1] not in 'rs': # numeric
1254 value
= float_or_none(value
)
1256 value
, fmt
= default
, 's'
1260 # If value is an object, sanitize might convert it to a string
1261 # So we convert it to repr first
1262 value
, fmt
= repr(value
), str_fmt
1263 if fmt
[-1] in 'csr':
1264 value
= sanitizer(initial_field
, value
)
1266 key
= '%s\0%s' % (key
.replace('%', '%\0'), outer_mobj
.group('format'))
1267 TMPL_DICT
[key
] = value
1268 return '{prefix}%({key}){fmt}'.format(key
=key
, fmt
=fmt
, prefix
=outer_mobj
.group('prefix'))
1270 return EXTERNAL_FORMAT_RE
.sub(create_key
, outtmpl
), TMPL_DICT
1272 def evaluate_outtmpl(self
, outtmpl
, info_dict
, *args
, **kwargs
):
1273 outtmpl
, info_dict
= self
.prepare_outtmpl(outtmpl
, info_dict
, *args
, **kwargs
)
1274 return self
.escape_outtmpl(outtmpl
) % info_dict
1276 def _prepare_filename(self
, info_dict
, *, outtmpl
=None, tmpl_type
=None):
1277 assert None in (outtmpl
, tmpl_type
), 'outtmpl and tmpl_type are mutually exclusive'
1279 outtmpl
= self
.params
['outtmpl'].get(tmpl_type
or 'default', self
.params
['outtmpl']['default'])
1281 outtmpl
= self
._outtmpl
_expandpath
(outtmpl
)
1282 filename
= self
.evaluate_outtmpl(outtmpl
, info_dict
, True)
1286 if tmpl_type
in ('', 'temp'):
1287 final_ext
, ext
= self
.params
.get('final_ext'), info_dict
.get('ext')
1288 if final_ext
and ext
and final_ext
!= ext
and filename
.endswith(f
'.{final_ext}'):
1289 filename
= replace_extension(filename
, ext
, final_ext
)
1291 force_ext
= OUTTMPL_TYPES
[tmpl_type
]
1293 filename
= replace_extension(filename
, force_ext
, info_dict
.get('ext'))
1295 # https://github.com/blackjack4494/youtube-dlc/issues/85
1296 trim_file_name
= self
.params
.get('trim_file_name', False)
1298 no_ext
, *ext
= filename
.rsplit('.', 2)
1299 filename
= join_nonempty(no_ext
[:trim_file_name
], *ext
, delim
='.')
1302 except ValueError as err
:
1303 self
.report_error('Error in output template: ' + str(err
) + ' (encoding: ' + repr(preferredencoding()) + ')')
1306 def prepare_filename(self
, info_dict
, dir_type
='', *, outtmpl
=None, warn
=False):
1307 """Generate the output filename"""
1309 assert not dir_type
, 'outtmpl and dir_type are mutually exclusive'
1311 filename
= self
._prepare
_filename
(info_dict
, tmpl_type
=dir_type
, outtmpl
=outtmpl
)
1312 if not filename
and dir_type
not in ('', 'temp'):
1316 if not self
.params
.get('paths'):
1318 elif filename
== '-':
1319 self
.report_warning('--paths is ignored when an outputting to stdout', only_once
=True)
1320 elif os
.path
.isabs(filename
):
1321 self
.report_warning('--paths is ignored since an absolute path is given in output template', only_once
=True)
1322 if filename
== '-' or not filename
:
1325 return self
.get_output_path(dir_type
, filename
)
1327 def _match_entry(self
, info_dict
, incomplete
=False, silent
=False):
1328 """ Returns None if the file should be downloaded """
1330 video_title
= info_dict
.get('title', info_dict
.get('id', 'entry'))
1333 if 'title' in info_dict
:
1334 # This can happen when we're just evaluating the playlist
1335 title
= info_dict
['title']
1336 matchtitle
= self
.params
.get('matchtitle', False)
1338 if not re
.search(matchtitle
, title
, re
.IGNORECASE
):
1339 return '"' + title
+ '" title did not match pattern "' + matchtitle
+ '"'
1340 rejecttitle
= self
.params
.get('rejecttitle', False)
1342 if re
.search(rejecttitle
, title
, re
.IGNORECASE
):
1343 return '"' + title
+ '" title matched reject pattern "' + rejecttitle
+ '"'
1344 date
= info_dict
.get('upload_date')
1345 if date
is not None:
1346 dateRange
= self
.params
.get('daterange', DateRange())
1347 if date
not in dateRange
:
1348 return f
'{date_from_str(date).isoformat()} upload date is not in range {dateRange}'
1349 view_count
= info_dict
.get('view_count')
1350 if view_count
is not None:
1351 min_views
= self
.params
.get('min_views')
1352 if min_views
is not None and view_count
< min_views
:
1353 return 'Skipping %s, because it has not reached minimum view count (%d/%d)' % (video_title
, view_count
, min_views
)
1354 max_views
= self
.params
.get('max_views')
1355 if max_views
is not None and view_count
> max_views
:
1356 return 'Skipping %s, because it has exceeded the maximum view count (%d/%d)' % (video_title
, view_count
, max_views
)
1357 if age_restricted(info_dict
.get('age_limit'), self
.params
.get('age_limit')):
1358 return 'Skipping "%s" because it is age restricted' % video_title
1360 match_filter
= self
.params
.get('match_filter')
1361 if match_filter
is not None:
1363 ret
= match_filter(info_dict
, incomplete
=incomplete
)
1365 # For backward compatibility
1366 ret
= None if incomplete
else match_filter(info_dict
)
1367 if ret
is NO_DEFAULT
:
1369 filename
= self
._format
_screen
(self
.prepare_filename(info_dict
), self
.Styles
.FILENAME
)
1370 reply
= input(self
._format
_screen
(
1371 f
'Download "{filename}"? (Y/n): ', self
.Styles
.EMPHASIS
)).lower().strip()
1372 if reply
in {'y', ''}
:
1375 return f
'Skipping {video_title}'
1376 elif ret
is not None:
1380 if self
.in_download_archive(info_dict
):
1381 reason
= '%s has already been recorded in the archive' % video_title
1382 break_opt
, break_err
= 'break_on_existing', ExistingVideoReached
1384 reason
= check_filter()
1385 break_opt
, break_err
= 'break_on_reject', RejectedVideoReached
1386 if reason
is not None:
1388 self
.to_screen('[download] ' + reason
)
1389 if self
.params
.get(break_opt
, False):
1394 def add_extra_info(info_dict
, extra_info
):
1395 '''Set the keys from extra_info in info dict if they are missing'''
1396 for key
, value
in extra_info
.items():
1397 info_dict
.setdefault(key
, value
)
1399 def extract_info(self
, url
, download
=True, ie_key
=None, extra_info
=None,
1400 process
=True, force_generic_extractor
=False):
1402 Return a list with a dictionary for each video extracted.
1405 url -- URL to extract
1408 download -- whether to download videos during extraction
1409 ie_key -- extractor key hint
1410 extra_info -- dictionary containing the extra values to add to each result
1411 process -- whether to resolve all unresolved references (URLs, playlist items),
1412 must be True for download to work.
1413 force_generic_extractor -- force using the generic extractor
1416 if extra_info
is None:
1419 if not ie_key
and force_generic_extractor
:
1423 ies
= {ie_key: self._ies[ie_key]}
if ie_key
in self
._ies
else {}
1427 for key
, ie
in ies
.items():
1428 if not ie
.suitable(url
):
1431 if not ie
.working():
1432 self
.report_warning('The program functionality for this site has been marked as broken, '
1433 'and will probably not work.')
1435 temp_id
= ie
.get_temp_id(url
)
1436 if temp_id
is not None and self
.in_download_archive({'id': temp_id, 'ie_key': key}
):
1437 self
.to_screen(f
'[{key}] {temp_id}: has already been recorded in the archive')
1438 if self
.params
.get('break_on_existing', False):
1439 raise ExistingVideoReached()
1441 return self
.__extract
_info
(url
, self
.get_info_extractor(key
), download
, extra_info
, process
)
1443 extractors_restricted
= self
.params
.get('allowed_extractors') not in (None, ['default'])
1444 self
.report_error(f
'No suitable extractor{format_field(ie_key, None, " (%s)")} found for URL {url}',
1445 tb
=False if extractors_restricted
else None)
1447 def _handle_extraction_exceptions(func
):
1448 @functools.wraps(func
)
1449 def wrapper(self
, *args
, **kwargs
):
1452 return func(self
, *args
, **kwargs
)
1453 except (DownloadCancelled
, LazyList
.IndexError, PagedList
.IndexError):
1455 except ReExtractInfo
as e
:
1457 self
.to_screen(f
'{e}; Re-extracting data')
1459 self
.to_stderr('\r')
1460 self
.report_warning(f
'{e}; Re-extracting data')
1462 except GeoRestrictedError
as e
:
1465 msg
+= '\nThis video is available in %s.' % ', '.join(
1466 map(ISO3166Utils
.short2full
, e
.countries
))
1467 msg
+= '\nYou might want to use a VPN or a proxy server (with --proxy) to workaround.'
1468 self
.report_error(msg
)
1469 except ExtractorError
as e
: # An error we somewhat expected
1470 self
.report_error(str(e
), e
.format_traceback())
1471 except Exception as e
:
1472 if self
.params
.get('ignoreerrors'):
1473 self
.report_error(str(e
), tb
=encode_compat_str(traceback
.format_exc()))
1479 def _wait_for_video(self
, ie_result
={}):
1480 if (not self
.params
.get('wait_for_video')
1481 or ie_result
.get('_type', 'video') != 'video'
1482 or ie_result
.get('formats') or ie_result
.get('url')):
1485 format_dur
= lambda dur
: '%02d:%02d:%02d' % timetuple_from_msec(dur
* 1000)[:-1]
1490 full_msg
= f
'{msg}\n'
1491 if not self
.params
.get('noprogress'):
1492 full_msg
= msg
+ ' ' * (len(last_msg
) - len(msg
)) + '\r'
1495 self
.to_screen(full_msg
, skip_eol
=True)
1498 min_wait
, max_wait
= self
.params
.get('wait_for_video')
1499 diff
= try_get(ie_result
, lambda x
: x
['release_timestamp'] - time
.time())
1500 if diff
is None and ie_result
.get('live_status') == 'is_upcoming':
1501 diff
= round(random
.uniform(min_wait
, max_wait
) if (max_wait
and min_wait
) else (max_wait
or min_wait
), 0)
1502 self
.report_warning('Release time of video is not known')
1503 elif ie_result
and (diff
or 0) <= 0:
1504 self
.report_warning('Video should already be available according to extracted info')
1505 diff
= min(max(diff
or 0, min_wait
or 0), max_wait
or float('inf'))
1506 self
.to_screen(f
'[wait] Waiting for {format_dur(diff)} - Press Ctrl+C to try now')
1508 wait_till
= time
.time() + diff
1511 diff
= wait_till
- time
.time()
1514 raise ReExtractInfo('[wait] Wait period ended', expected
=True)
1515 progress(f
'[wait] Remaining time until next attempt: {self._format_screen(format_dur(diff), self.Styles.EMPHASIS)}')
1517 except KeyboardInterrupt:
1519 raise ReExtractInfo('[wait] Interrupted by user', expected
=True)
1520 except BaseException
as e
:
1521 if not isinstance(e
, ReExtractInfo
):
1525 @_handle_extraction_exceptions
1526 def __extract_info(self
, url
, ie
, download
, extra_info
, process
):
1528 ie_result
= ie
.extract(url
)
1529 except UserNotLive
as e
:
1531 if self
.params
.get('wait_for_video'):
1532 self
.report_warning(e
)
1533 self
._wait
_for
_video
()
1535 if ie_result
is None: # Finished already (backwards compatibility; listformats and friends should be moved here)
1536 self
.report_warning(f
'Extractor {ie.IE_NAME} returned nothing{bug_reports_message()}')
1538 if isinstance(ie_result
, list):
1539 # Backwards compatibility: old IE result format
1541 '_type': 'compat_list',
1542 'entries': ie_result
,
1544 if extra_info
.get('original_url'):
1545 ie_result
.setdefault('original_url', extra_info
['original_url'])
1546 self
.add_default_extra_info(ie_result
, ie
, url
)
1548 self
._wait
_for
_video
(ie_result
)
1549 return self
.process_ie_result(ie_result
, download
, extra_info
)
1553 def add_default_extra_info(self
, ie_result
, ie
, url
):
1555 self
.add_extra_info(ie_result
, {
1557 'original_url': url
,
1559 webpage_url
= ie_result
.get('webpage_url')
1561 self
.add_extra_info(ie_result
, {
1562 'webpage_url_basename': url_basename(webpage_url
),
1563 'webpage_url_domain': get_domain(webpage_url
),
1566 self
.add_extra_info(ie_result
, {
1567 'extractor': ie
.IE_NAME
,
1568 'extractor_key': ie
.ie_key(),
1571 def process_ie_result(self
, ie_result
, download
=True, extra_info
=None):
1573 Take the result of the ie(may be modified) and resolve all unresolved
1574 references (URLs, playlist items).
1576 It will also download the videos if 'download'.
1577 Returns the resolved ie_result.
1579 if extra_info
is None:
1581 result_type
= ie_result
.get('_type', 'video')
1583 if result_type
in ('url', 'url_transparent'):
1584 ie_result
['url'] = sanitize_url(
1585 ie_result
['url'], scheme
='http' if self
.params
.get('prefer_insecure') else 'https')
1586 if ie_result
.get('original_url'):
1587 extra_info
.setdefault('original_url', ie_result
['original_url'])
1589 extract_flat
= self
.params
.get('extract_flat', False)
1590 if ((extract_flat
== 'in_playlist' and 'playlist' in extra_info
)
1591 or extract_flat
is True):
1592 info_copy
= ie_result
.copy()
1593 ie
= try_get(ie_result
.get('ie_key'), self
.get_info_extractor
)
1594 if ie
and not ie_result
.get('id'):
1595 info_copy
['id'] = ie
.get_temp_id(ie_result
['url'])
1596 self
.add_default_extra_info(info_copy
, ie
, ie_result
['url'])
1597 self
.add_extra_info(info_copy
, extra_info
)
1598 info_copy
, _
= self
.pre_process(info_copy
)
1599 self
.__forced
_printings
(info_copy
, self
.prepare_filename(info_copy
), incomplete
=True)
1600 self
._raise
_pending
_errors
(info_copy
)
1601 if self
.params
.get('force_write_download_archive', False):
1602 self
.record_download_archive(info_copy
)
1605 if result_type
== 'video':
1606 self
.add_extra_info(ie_result
, extra_info
)
1607 ie_result
= self
.process_video_result(ie_result
, download
=download
)
1608 self
._raise
_pending
_errors
(ie_result
)
1609 additional_urls
= (ie_result
or {}).get('additional_urls')
1611 # TODO: Improve MetadataParserPP to allow setting a list
1612 if isinstance(additional_urls
, str):
1613 additional_urls
= [additional_urls
]
1615 '[info] %s: %d additional URL(s) requested' % (ie_result
['id'], len(additional_urls
)))
1616 self
.write_debug('Additional URLs: "%s"' % '", "'.join(additional_urls
))
1617 ie_result
['additional_entries'] = [
1619 url
, download
, extra_info
=extra_info
,
1620 force_generic_extractor
=self
.params
.get('force_generic_extractor'))
1621 for url
in additional_urls
1624 elif result_type
== 'url':
1625 # We have to add extra_info to the results because it may be
1626 # contained in a playlist
1627 return self
.extract_info(
1628 ie_result
['url'], download
,
1629 ie_key
=ie_result
.get('ie_key'),
1630 extra_info
=extra_info
)
1631 elif result_type
== 'url_transparent':
1632 # Use the information from the embedding page
1633 info
= self
.extract_info(
1634 ie_result
['url'], ie_key
=ie_result
.get('ie_key'),
1635 extra_info
=extra_info
, download
=False, process
=False)
1637 # extract_info may return None when ignoreerrors is enabled and
1638 # extraction failed with an error, don't crash and return early
1643 exempted_fields
= {'_type', 'url', 'ie_key'}
1644 if not ie_result
.get('section_end') and ie_result
.get('section_start') is None:
1645 # For video clips, the id etc of the clip extractor should be used
1646 exempted_fields |
= {'id', 'extractor', 'extractor_key'}
1648 new_result
= info
.copy()
1649 new_result
.update(filter_dict(ie_result
, lambda k
, v
: v
is not None and k
not in exempted_fields
))
1651 # Extracted info may not be a video result (i.e.
1652 # info.get('_type', 'video') != video) but rather an url or
1653 # url_transparent. In such cases outer metadata (from ie_result)
1654 # should be propagated to inner one (info). For this to happen
1655 # _type of info should be overridden with url_transparent. This
1656 # fixes issue from https://github.com/ytdl-org/youtube-dl/pull/11163.
1657 if new_result
.get('_type') == 'url':
1658 new_result
['_type'] = 'url_transparent'
1660 return self
.process_ie_result(
1661 new_result
, download
=download
, extra_info
=extra_info
)
1662 elif result_type
in ('playlist', 'multi_video'):
1663 # Protect from infinite recursion due to recursively nested playlists
1664 # (see https://github.com/ytdl-org/youtube-dl/issues/27833)
1665 webpage_url
= ie_result
['webpage_url']
1666 if webpage_url
in self
._playlist
_urls
:
1668 '[download] Skipping already downloaded playlist: %s'
1669 % ie_result
.get('title') or ie_result
.get('id'))
1672 self
._playlist
_level
+= 1
1673 self
._playlist
_urls
.add(webpage_url
)
1674 self
._fill
_common
_fields
(ie_result
, False)
1675 self
._sanitize
_thumbnails
(ie_result
)
1677 return self
.__process
_playlist
(ie_result
, download
)
1679 self
._playlist
_level
-= 1
1680 if not self
._playlist
_level
:
1681 self
._playlist
_urls
.clear()
1682 elif result_type
== 'compat_list':
1683 self
.report_warning(
1684 'Extractor %s returned a compat_list result. '
1685 'It needs to be updated.' % ie_result
.get('extractor'))
1688 self
.add_extra_info(r
, {
1689 'extractor': ie_result
['extractor'],
1690 'webpage_url': ie_result
['webpage_url'],
1691 'webpage_url_basename': url_basename(ie_result
['webpage_url']),
1692 'webpage_url_domain': get_domain(ie_result
['webpage_url']),
1693 'extractor_key': ie_result
['extractor_key'],
1696 ie_result
['entries'] = [
1697 self
.process_ie_result(_fixup(r
), download
, extra_info
)
1698 for r
in ie_result
['entries']
1702 raise Exception('Invalid result type: %s' % result_type
)
1704 def _ensure_dir_exists(self
, path
):
1705 return make_dir(path
, self
.report_error
)
1708 def _playlist_infodict(ie_result
, strict
=False, **kwargs
):
1710 'playlist_count': ie_result
.get('playlist_count'),
1711 'playlist': ie_result
.get('title') or ie_result
.get('id'),
1712 'playlist_id': ie_result
.get('id'),
1713 'playlist_title': ie_result
.get('title'),
1714 'playlist_uploader': ie_result
.get('uploader'),
1715 'playlist_uploader_id': ie_result
.get('uploader_id'),
1722 'playlist_index': 0,
1723 '__last_playlist_index': max(ie_result
['requested_entries'] or (0, 0)),
1724 'extractor': ie_result
['extractor'],
1725 'webpage_url': ie_result
['webpage_url'],
1726 'webpage_url_basename': url_basename(ie_result
['webpage_url']),
1727 'webpage_url_domain': get_domain(ie_result
['webpage_url']),
1728 'extractor_key': ie_result
['extractor_key'],
1731 def __process_playlist(self
, ie_result
, download
):
1732 """Process each entry in the playlist"""
1733 assert ie_result
['_type'] in ('playlist', 'multi_video')
1735 common_info
= self
._playlist
_infodict
(ie_result
, strict
=True)
1736 title
= common_info
.get('playlist') or '<Untitled>'
1737 if self
._match
_entry
(common_info
, incomplete
=True) is not None:
1739 self
.to_screen(f
'[download] Downloading {ie_result["_type"]}: {title}')
1741 all_entries
= PlaylistEntries(self
, ie_result
)
1742 entries
= orderedSet(all_entries
.get_requested_items(), lazy
=True)
1744 lazy
= self
.params
.get('lazy_playlist')
1746 resolved_entries
, n_entries
= [], 'N/A'
1747 ie_result
['requested_entries'], ie_result
['entries'] = None, None
1749 entries
= resolved_entries
= list(entries
)
1750 n_entries
= len(resolved_entries
)
1751 ie_result
['requested_entries'], ie_result
['entries'] = tuple(zip(*resolved_entries
)) or ([], [])
1752 if not ie_result
.get('playlist_count'):
1753 # Better to do this after potentially exhausting entries
1754 ie_result
['playlist_count'] = all_entries
.get_full_count()
1756 extra
= self
._playlist
_infodict
(ie_result
, n_entries
=int_or_none(n_entries
))
1757 ie_copy
= collections
.ChainMap(ie_result
, extra
)
1759 _infojson_written
= False
1760 write_playlist_files
= self
.params
.get('allow_playlist_files', True)
1761 if write_playlist_files
and self
.params
.get('list_thumbnails'):
1762 self
.list_thumbnails(ie_result
)
1763 if write_playlist_files
and not self
.params
.get('simulate'):
1764 _infojson_written
= self
._write
_info
_json
(
1765 'playlist', ie_result
, self
.prepare_filename(ie_copy
, 'pl_infojson'))
1766 if _infojson_written
is None:
1768 if self
._write
_description
('playlist', ie_result
,
1769 self
.prepare_filename(ie_copy
, 'pl_description')) is None:
1771 # TODO: This should be passed to ThumbnailsConvertor if necessary
1772 self
._write
_thumbnails
('playlist', ie_result
, self
.prepare_filename(ie_copy
, 'pl_thumbnail'))
1775 if self
.params
.get('playlistreverse') or self
.params
.get('playlistrandom'):
1776 self
.report_warning('playlistreverse and playlistrandom are not supported with lazy_playlist', only_once
=True)
1777 elif self
.params
.get('playlistreverse'):
1779 elif self
.params
.get('playlistrandom'):
1780 random
.shuffle(entries
)
1782 self
.to_screen(f
'[{ie_result["extractor"]}] Playlist {title}: Downloading {n_entries} videos'
1783 f
'{format_field(ie_result, "playlist_count", " of %s")}')
1785 keep_resolved_entries
= self
.params
.get('extract_flat') != 'discard'
1786 if self
.params
.get('extract_flat') == 'discard_in_playlist':
1787 keep_resolved_entries
= ie_result
['_type'] != 'playlist'
1788 if keep_resolved_entries
:
1789 self
.write_debug('The information of all playlist entries will be held in memory')
1792 max_failures
= self
.params
.get('skip_playlist_after_errors') or float('inf')
1793 for i
, (playlist_index
, entry
) in enumerate(entries
):
1795 resolved_entries
.append((playlist_index
, entry
))
1799 entry
['__x_forwarded_for_ip'] = ie_result
.get('__x_forwarded_for_ip')
1800 if not lazy
and 'playlist-index' in self
.params
.get('compat_opts', []):
1801 playlist_index
= ie_result
['requested_entries'][i
]
1803 entry_copy
= collections
.ChainMap(entry
, {
1805 'n_entries': int_or_none(n_entries
),
1806 'playlist_index': playlist_index
,
1807 'playlist_autonumber': i
+ 1,
1810 if self
._match
_entry
(entry_copy
, incomplete
=True) is not None:
1811 # For compatabilty with youtube-dl. See https://github.com/yt-dlp/yt-dlp/issues/4369
1812 resolved_entries
[i
] = (playlist_index
, NO_DEFAULT
)
1815 self
.to_screen('[download] Downloading video %s of %s' % (
1816 self
._format
_screen
(i
+ 1, self
.Styles
.ID
), self
._format
_screen
(n_entries
, self
.Styles
.EMPHASIS
)))
1819 'playlist_index': playlist_index
,
1820 'playlist_autonumber': i
+ 1,
1822 entry_result
= self
.__process
_iterable
_entry
(entry
, download
, extra
)
1823 if not entry_result
:
1825 if failures
>= max_failures
:
1827 f
'Skipping the remaining entries in playlist "{title}" since {failures} items failed extraction')
1829 if keep_resolved_entries
:
1830 resolved_entries
[i
] = (playlist_index
, entry_result
)
1832 # Update with processed data
1833 ie_result
['requested_entries'] = [i
for i
, e
in resolved_entries
if e
is not NO_DEFAULT
]
1834 ie_result
['entries'] = [e
for _
, e
in resolved_entries
if e
is not NO_DEFAULT
]
1836 # Write the updated info to json
1837 if _infojson_written
is True and self
._write
_info
_json
(
1838 'updated playlist', ie_result
,
1839 self
.prepare_filename(ie_copy
, 'pl_infojson'), overwrite
=True) is None:
1842 ie_result
= self
.run_all_pps('playlist', ie_result
)
1843 self
.to_screen(f
'[download] Finished downloading playlist: {title}')
1846 @_handle_extraction_exceptions
1847 def __process_iterable_entry(self
, entry
, download
, extra_info
):
1848 return self
.process_ie_result(
1849 entry
, download
=download
, extra_info
=extra_info
)
1851 def _build_format_filter(self
, filter_spec
):
1852 " Returns a function to filter the formats according to the filter_spec "
1862 operator_rex
= re
.compile(r
'''(?x)\s*
1863 (?P<key>width|height|tbr|abr|vbr|asr|filesize|filesize_approx|fps)\s*
1864 (?P<op>%s)(?P<none_inclusive>\s*\?)?\s*
1865 (?P<value>[0-9.]+(?:[kKmMgGtTpPeEzZyY]i?[Bb]?)?)\s*
1866 ''' % '|'.join(map(re
.escape
, OPERATORS
.keys())))
1867 m
= operator_rex
.fullmatch(filter_spec
)
1870 comparison_value
= int(m
.group('value'))
1872 comparison_value
= parse_filesize(m
.group('value'))
1873 if comparison_value
is None:
1874 comparison_value
= parse_filesize(m
.group('value') + 'B')
1875 if comparison_value
is None:
1877 'Invalid value %r in format specification %r' % (
1878 m
.group('value'), filter_spec
))
1879 op
= OPERATORS
[m
.group('op')]
1884 '^=': lambda attr
, value
: attr
.startswith(value
),
1885 '$=': lambda attr
, value
: attr
.endswith(value
),
1886 '*=': lambda attr
, value
: value
in attr
,
1887 '~=': lambda attr
, value
: value
.search(attr
) is not None
1889 str_operator_rex
= re
.compile(r
'''(?x)\s*
1890 (?P<key>[a-zA-Z0-9._-]+)\s*
1891 (?P<negation>!\s*)?(?P<op>%s)\s*(?P<none_inclusive>\?\s*)?
1893 (?P<value>(?(quote)(?:(?!(?P=quote))[^\\]|\\.)+|[\w.-]+))
1894 (?(quote)(?P=quote))\s*
1895 ''' % '|'.join(map(re
.escape
, STR_OPERATORS
.keys())))
1896 m
= str_operator_rex
.fullmatch(filter_spec
)
1898 if m
.group('op') == '~=':
1899 comparison_value
= re
.compile(m
.group('value'))
1901 comparison_value
= re
.sub(r
'''\\([\\"'])''', r
'\1', m
.group('value'))
1902 str_op
= STR_OPERATORS
[m
.group('op')]
1903 if m
.group('negation'):
1904 op
= lambda attr
, value
: not str_op(attr
, value
)
1909 raise SyntaxError('Invalid filter specification %r' % filter_spec
)
1912 actual_value
= f
.get(m
.group('key'))
1913 if actual_value
is None:
1914 return m
.group('none_inclusive')
1915 return op(actual_value
, comparison_value
)
1918 def _check_formats(self
, formats
):
1920 self
.to_screen('[info] Testing format %s' % f
['format_id'])
1921 path
= self
.get_output_path('temp')
1922 if not self
._ensure
_dir
_exists
(f
'{path}/'):
1924 temp_file
= tempfile
.NamedTemporaryFile(suffix
='.tmp', delete
=False, dir=path
or None)
1927 success
, _
= self
.dl(temp_file
.name
, f
, test
=True)
1928 except (DownloadError
, OSError, ValueError) + network_exceptions
:
1931 if os
.path
.exists(temp_file
.name
):
1933 os
.remove(temp_file
.name
)
1935 self
.report_warning('Unable to delete temporary file "%s"' % temp_file
.name
)
1939 self
.to_screen('[info] Unable to download format %s. Skipping...' % f
['format_id'])
1941 def _default_format_spec(self
, info_dict
, download
=True):
1944 merger
= FFmpegMergerPP(self
)
1945 return merger
.available
and merger
.can_merge()
1948 not self
.params
.get('simulate')
1952 or info_dict
.get('is_live') and not self
.params
.get('live_from_start')
1953 or self
.params
['outtmpl']['default'] == '-'))
1956 or self
.params
.get('allow_multiple_audio_streams', False)
1957 or 'format-spec' in self
.params
['compat_opts'])
1960 'best/bestvideo+bestaudio' if prefer_best
1961 else 'bestvideo*+bestaudio/best' if not compat
1962 else 'bestvideo+bestaudio/best')
1964 def build_format_selector(self
, format_spec
):
1965 def syntax_error(note
, start
):
1967 'Invalid format specification: '
1968 '{}\n\t{}\n\t{}^'.format(note
, format_spec
, ' ' * start
[1]))
1969 return SyntaxError(message
)
1971 PICKFIRST
= 'PICKFIRST'
1975 FormatSelector
= collections
.namedtuple('FormatSelector', ['type', 'selector', 'filters'])
1977 allow_multiple_streams
= {'audio': self
.params
.get('allow_multiple_audio_streams', False),
1978 'video': self
.params
.get('allow_multiple_video_streams', False)}
1980 check_formats
= self
.params
.get('check_formats') == 'selected'
1982 def _parse_filter(tokens
):
1984 for type, string
, start
, _
, _
in tokens
:
1985 if type == tokenize
.OP
and string
== ']':
1986 return ''.join(filter_parts
)
1988 filter_parts
.append(string
)
1990 def _remove_unused_ops(tokens
):
1991 # Remove operators that we don't use and join them with the surrounding strings.
1992 # E.g. 'mp4' '-' 'baseline' '-' '16x9' is converted to 'mp4-baseline-16x9'
1993 ALLOWED_OPS
= ('/', '+', ',', '(', ')')
1994 last_string
, last_start
, last_end
, last_line
= None, None, None, None
1995 for type, string
, start
, end
, line
in tokens
:
1996 if type == tokenize
.OP
and string
== '[':
1998 yield tokenize
.NAME
, last_string
, last_start
, last_end
, last_line
2000 yield type, string
, start
, end
, line
2001 # everything inside brackets will be handled by _parse_filter
2002 for type, string
, start
, end
, line
in tokens
:
2003 yield type, string
, start
, end
, line
2004 if type == tokenize
.OP
and string
== ']':
2006 elif type == tokenize
.OP
and string
in ALLOWED_OPS
:
2008 yield tokenize
.NAME
, last_string
, last_start
, last_end
, last_line
2010 yield type, string
, start
, end
, line
2011 elif type in [tokenize
.NAME
, tokenize
.NUMBER
, tokenize
.OP
]:
2013 last_string
= string
2017 last_string
+= string
2019 yield tokenize
.NAME
, last_string
, last_start
, last_end
, last_line
2021 def _parse_format_selection(tokens
, inside_merge
=False, inside_choice
=False, inside_group
=False):
2023 current_selector
= None
2024 for type, string
, start
, _
, _
in tokens
:
2025 # ENCODING is only defined in python 3.x
2026 if type == getattr(tokenize
, 'ENCODING', None):
2028 elif type in [tokenize
.NAME
, tokenize
.NUMBER
]:
2029 current_selector
= FormatSelector(SINGLE
, string
, [])
2030 elif type == tokenize
.OP
:
2032 if not inside_group
:
2033 # ')' will be handled by the parentheses group
2034 tokens
.restore_last_token()
2036 elif inside_merge
and string
in ['/', ',']:
2037 tokens
.restore_last_token()
2039 elif inside_choice
and string
== ',':
2040 tokens
.restore_last_token()
2043 if not current_selector
:
2044 raise syntax_error('"," must follow a format selector', start
)
2045 selectors
.append(current_selector
)
2046 current_selector
= None
2048 if not current_selector
:
2049 raise syntax_error('"/" must follow a format selector', start
)
2050 first_choice
= current_selector
2051 second_choice
= _parse_format_selection(tokens
, inside_choice
=True)
2052 current_selector
= FormatSelector(PICKFIRST
, (first_choice
, second_choice
), [])
2054 if not current_selector
:
2055 current_selector
= FormatSelector(SINGLE
, 'best', [])
2056 format_filter
= _parse_filter(tokens
)
2057 current_selector
.filters
.append(format_filter
)
2059 if current_selector
:
2060 raise syntax_error('Unexpected "("', start
)
2061 group
= _parse_format_selection(tokens
, inside_group
=True)
2062 current_selector
= FormatSelector(GROUP
, group
, [])
2064 if not current_selector
:
2065 raise syntax_error('Unexpected "+"', start
)
2066 selector_1
= current_selector
2067 selector_2
= _parse_format_selection(tokens
, inside_merge
=True)
2069 raise syntax_error('Expected a selector', start
)
2070 current_selector
= FormatSelector(MERGE
, (selector_1
, selector_2
), [])
2072 raise syntax_error(f
'Operator not recognized: "{string}"', start
)
2073 elif type == tokenize
.ENDMARKER
:
2075 if current_selector
:
2076 selectors
.append(current_selector
)
2079 def _merge(formats_pair
):
2080 format_1
, format_2
= formats_pair
2083 formats_info
.extend(format_1
.get('requested_formats', (format_1
,)))
2084 formats_info
.extend(format_2
.get('requested_formats', (format_2
,)))
2086 if not allow_multiple_streams
['video'] or not allow_multiple_streams
['audio']:
2087 get_no_more
= {'video': False, 'audio': False}
2088 for (i
, fmt_info
) in enumerate(formats_info
):
2089 if fmt_info
.get('acodec') == fmt_info
.get('vcodec') == 'none':
2092 for aud_vid
in ['audio', 'video']:
2093 if not allow_multiple_streams
[aud_vid
] and fmt_info
.get(aud_vid
[0] + 'codec') != 'none':
2094 if get_no_more
[aud_vid
]:
2097 get_no_more
[aud_vid
] = True
2099 if len(formats_info
) == 1:
2100 return formats_info
[0]
2102 video_fmts
= [fmt_info
for fmt_info
in formats_info
if fmt_info
.get('vcodec') != 'none']
2103 audio_fmts
= [fmt_info
for fmt_info
in formats_info
if fmt_info
.get('acodec') != 'none']
2105 the_only_video
= video_fmts
[0] if len(video_fmts
) == 1 else None
2106 the_only_audio
= audio_fmts
[0] if len(audio_fmts
) == 1 else None
2108 output_ext
= get_compatible_ext(
2109 vcodecs
=[f
.get('vcodec') for f
in video_fmts
],
2110 acodecs
=[f
.get('acodec') for f
in audio_fmts
],
2111 vexts
=[f
['ext'] for f
in video_fmts
],
2112 aexts
=[f
['ext'] for f
in audio_fmts
],
2113 preferences
=(try_call(lambda: self
.params
['merge_output_format'].split('/'))
2114 or self
.params
.get('prefer_free_formats') and ('webm', 'mkv')))
2116 filtered
= lambda *keys
: filter(None, (traverse_obj(fmt
, *keys
) for fmt
in formats_info
))
2119 'requested_formats': formats_info
,
2120 'format': '+'.join(filtered('format')),
2121 'format_id': '+'.join(filtered('format_id')),
2123 'protocol': '+'.join(map(determine_protocol
, formats_info
)),
2124 'language': '+'.join(orderedSet(filtered('language'))) or None,
2125 'format_note': '+'.join(orderedSet(filtered('format_note'))) or None,
2126 'filesize_approx': sum(filtered('filesize', 'filesize_approx')) or None,
2127 'tbr': sum(filtered('tbr', 'vbr', 'abr')),
2132 'width': the_only_video
.get('width'),
2133 'height': the_only_video
.get('height'),
2134 'resolution': the_only_video
.get('resolution') or self
.format_resolution(the_only_video
),
2135 'fps': the_only_video
.get('fps'),
2136 'dynamic_range': the_only_video
.get('dynamic_range'),
2137 'vcodec': the_only_video
.get('vcodec'),
2138 'vbr': the_only_video
.get('vbr'),
2139 'stretched_ratio': the_only_video
.get('stretched_ratio'),
2144 'acodec': the_only_audio
.get('acodec'),
2145 'abr': the_only_audio
.get('abr'),
2146 'asr': the_only_audio
.get('asr'),
2147 'audio_channels': the_only_audio
.get('audio_channels')
2152 def _check_formats(formats
):
2153 if not check_formats
:
2156 yield from self
._check
_formats
(formats
)
2158 def _build_selector_function(selector
):
2159 if isinstance(selector
, list): # ,
2160 fs
= [_build_selector_function(s
) for s
in selector
]
2162 def selector_function(ctx
):
2165 return selector_function
2167 elif selector
.type == GROUP
: # ()
2168 selector_function
= _build_selector_function(selector
.selector
)
2170 elif selector
.type == PICKFIRST
: # /
2171 fs
= [_build_selector_function(s
) for s
in selector
.selector
]
2173 def selector_function(ctx
):
2175 picked_formats
= list(f(ctx
))
2177 return picked_formats
2180 elif selector
.type == MERGE
: # +
2181 selector_1
, selector_2
= map(_build_selector_function
, selector
.selector
)
2183 def selector_function(ctx
):
2184 for pair
in itertools
.product(selector_1(ctx
), selector_2(ctx
)):
2187 elif selector
.type == SINGLE
: # atom
2188 format_spec
= selector
.selector
or 'best'
2190 # TODO: Add allvideo, allaudio etc by generalizing the code with best/worst selector
2191 if format_spec
== 'all':
2192 def selector_function(ctx
):
2193 yield from _check_formats(ctx
['formats'][::-1])
2194 elif format_spec
== 'mergeall':
2195 def selector_function(ctx
):
2196 formats
= list(_check_formats(
2197 f
for f
in ctx
['formats'] if f
.get('vcodec') != 'none' or f
.get('acodec') != 'none'))
2200 merged_format
= formats
[-1]
2201 for f
in formats
[-2::-1]:
2202 merged_format
= _merge((merged_format
, f
))
2206 format_fallback
, seperate_fallback
, format_reverse
, format_idx
= False, None, True, 1
2208 r
'(?P<bw>best|worst|b|w)(?P<type>video|audio|v|a)?(?P<mod>\*)?(?:\.(?P<n>[1-9]\d*))?$',
2210 if mobj
is not None:
2211 format_idx
= int_or_none(mobj
.group('n'), default
=1)
2212 format_reverse
= mobj
.group('bw')[0] == 'b'
2213 format_type
= (mobj
.group('type') or [None])[0]
2214 not_format_type
= {'v': 'a', 'a': 'v'}
.get(format_type
)
2215 format_modified
= mobj
.group('mod') is not None
2217 format_fallback
= not format_type
and not format_modified
# for b, w
2219 (lambda f
: f
.get('%scodec' % format_type
) != 'none')
2220 if format_type
and format_modified
# bv*, ba*, wv*, wa*
2221 else (lambda f
: f
.get('%scodec' % not_format_type
) == 'none')
2222 if format_type
# bv, ba, wv, wa
2223 else (lambda f
: f
.get('vcodec') != 'none' and f
.get('acodec') != 'none')
2224 if not format_modified
# b, w
2225 else lambda f
: True) # b*, w*
2226 filter_f
= lambda f
: _filter_f(f
) and (
2227 f
.get('vcodec') != 'none' or f
.get('acodec') != 'none')
2229 if format_spec
in self
._format
_selection
_exts
['audio']:
2230 filter_f
= lambda f
: f
.get('ext') == format_spec
and f
.get('acodec') != 'none'
2231 elif format_spec
in self
._format
_selection
_exts
['video']:
2232 filter_f
= lambda f
: f
.get('ext') == format_spec
and f
.get('acodec') != 'none' and f
.get('vcodec') != 'none'
2233 seperate_fallback
= lambda f
: f
.get('ext') == format_spec
and f
.get('vcodec') != 'none'
2234 elif format_spec
in self
._format
_selection
_exts
['storyboards']:
2235 filter_f
= lambda f
: f
.get('ext') == format_spec
and f
.get('acodec') == 'none' and f
.get('vcodec') == 'none'
2237 filter_f
= lambda f
: f
.get('format_id') == format_spec
# id
2239 def selector_function(ctx
):
2240 formats
= list(ctx
['formats'])
2241 matches
= list(filter(filter_f
, formats
)) if filter_f
is not None else formats
2243 if format_fallback
and ctx
['incomplete_formats']:
2244 # for extractors with incomplete formats (audio only (soundcloud)
2245 # or video only (imgur)) best/worst will fallback to
2246 # best/worst {video,audio}-only format
2248 elif seperate_fallback
and not ctx
['has_merged_format']:
2249 # for compatibility with youtube-dl when there is no pre-merged format
2250 matches
= list(filter(seperate_fallback
, formats
))
2251 matches
= LazyList(_check_formats(matches
[::-1 if format_reverse
else 1]))
2253 yield matches
[format_idx
- 1]
2254 except LazyList
.IndexError:
2257 filters
= [self
._build
_format
_filter
(f
) for f
in selector
.filters
]
2259 def final_selector(ctx
):
2260 ctx_copy
= dict(ctx
)
2261 for _filter
in filters
:
2262 ctx_copy
['formats'] = list(filter(_filter
, ctx_copy
['formats']))
2263 return selector_function(ctx_copy
)
2264 return final_selector
2266 stream
= io
.BytesIO(format_spec
.encode())
2268 tokens
= list(_remove_unused_ops(tokenize
.tokenize(stream
.readline
)))
2269 except tokenize
.TokenError
:
2270 raise syntax_error('Missing closing/opening brackets or parenthesis', (0, len(format_spec
)))
2272 class TokenIterator
:
2273 def __init__(self
, tokens
):
2274 self
.tokens
= tokens
2281 if self
.counter
>= len(self
.tokens
):
2282 raise StopIteration()
2283 value
= self
.tokens
[self
.counter
]
2289 def restore_last_token(self
):
2292 parsed_selector
= _parse_format_selection(iter(TokenIterator(tokens
)))
2293 return _build_selector_function(parsed_selector
)
2295 def _calc_headers(self
, info_dict
):
2296 res
= merge_headers(self
.params
['http_headers'], info_dict
.get('http_headers') or {})
2298 cookies
= self
._calc
_cookies
(info_dict
['url'])
2300 res
['Cookie'] = cookies
2302 if 'X-Forwarded-For' not in res
:
2303 x_forwarded_for_ip
= info_dict
.get('__x_forwarded_for_ip')
2304 if x_forwarded_for_ip
:
2305 res
['X-Forwarded-For'] = x_forwarded_for_ip
2309 def _calc_cookies(self
, url
):
2310 pr
= sanitized_Request(url
)
2311 self
.cookiejar
.add_cookie_header(pr
)
2312 return pr
.get_header('Cookie')
2314 def _sort_thumbnails(self
, thumbnails
):
2315 thumbnails
.sort(key
=lambda t
: (
2316 t
.get('preference') if t
.get('preference') is not None else -1,
2317 t
.get('width') if t
.get('width') is not None else -1,
2318 t
.get('height') if t
.get('height') is not None else -1,
2319 t
.get('id') if t
.get('id') is not None else '',
2322 def _sanitize_thumbnails(self
, info_dict
):
2323 thumbnails
= info_dict
.get('thumbnails')
2324 if thumbnails
is None:
2325 thumbnail
= info_dict
.get('thumbnail')
2327 info_dict
['thumbnails'] = thumbnails
= [{'url': thumbnail}
]
2331 def check_thumbnails(thumbnails
):
2332 for t
in thumbnails
:
2333 self
.to_screen(f
'[info] Testing thumbnail {t["id"]}')
2335 self
.urlopen(HEADRequest(t
['url']))
2336 except network_exceptions
as err
:
2337 self
.to_screen(f
'[info] Unable to connect to thumbnail {t["id"]} URL {t["url"]!r} - {err}. Skipping...')
2341 self
._sort
_thumbnails
(thumbnails
)
2342 for i
, t
in enumerate(thumbnails
):
2343 if t
.get('id') is None:
2345 if t
.get('width') and t
.get('height'):
2346 t
['resolution'] = '%dx%d' % (t
['width'], t
['height'])
2347 t
['url'] = sanitize_url(t
['url'])
2349 if self
.params
.get('check_formats') is True:
2350 info_dict
['thumbnails'] = LazyList(check_thumbnails(thumbnails
[::-1]), reverse
=True)
2352 info_dict
['thumbnails'] = thumbnails
2354 def _fill_common_fields(self
, info_dict
, is_video
=True):
2355 # TODO: move sanitization here
2357 # playlists are allowed to lack "title"
2358 title
= info_dict
.get('title', NO_DEFAULT
)
2359 if title
is NO_DEFAULT
:
2360 raise ExtractorError('Missing "title" field in extractor result',
2361 video_id
=info_dict
['id'], ie
=info_dict
['extractor'])
2362 info_dict
['fulltitle'] = title
2365 self
.write_debug('Extractor gave empty title. Creating a generic title')
2367 self
.report_warning('Extractor failed to obtain "title". Creating a generic title instead')
2368 info_dict
['title'] = f
'{info_dict["extractor"].replace(":", "-")} video #{info_dict["id"]}'
2370 if info_dict
.get('duration') is not None:
2371 info_dict
['duration_string'] = formatSeconds(info_dict
['duration'])
2373 for ts_key
, date_key
in (
2374 ('timestamp', 'upload_date'),
2375 ('release_timestamp', 'release_date'),
2376 ('modified_timestamp', 'modified_date'),
2378 if info_dict
.get(date_key
) is None and info_dict
.get(ts_key
) is not None:
2379 # Working around out-of-range timestamp values (e.g. negative ones on Windows,
2380 # see http://bugs.python.org/issue1646728)
2381 with contextlib
.suppress(ValueError, OverflowError, OSError):
2382 upload_date
= datetime
.datetime
.utcfromtimestamp(info_dict
[ts_key
])
2383 info_dict
[date_key
] = upload_date
.strftime('%Y%m%d')
2385 live_keys
= ('is_live', 'was_live')
2386 live_status
= info_dict
.get('live_status')
2387 if live_status
is None:
2388 for key
in live_keys
:
2389 if info_dict
.get(key
) is False:
2391 if info_dict
.get(key
):
2394 if all(info_dict
.get(key
) is False for key
in live_keys
):
2395 live_status
= 'not_live'
2397 info_dict
['live_status'] = live_status
2398 for key
in live_keys
:
2399 if info_dict
.get(key
) is None:
2400 info_dict
[key
] = (live_status
== key
)
2402 # Auto generate title fields corresponding to the *_number fields when missing
2403 # in order to always have clean titles. This is very common for TV series.
2404 for field
in ('chapter', 'season', 'episode'):
2405 if info_dict
.get('%s_number' % field
) is not None and not info_dict
.get(field
):
2406 info_dict
[field
] = '%s %d' % (field
.capitalize(), info_dict
['%s_number' % field
])
2408 def _raise_pending_errors(self
, info
):
2409 err
= info
.pop('__pending_error', None)
2411 self
.report_error(err
, tb
=False)
2413 def process_video_result(self
, info_dict
, download
=True):
2414 assert info_dict
.get('_type', 'video') == 'video'
2415 self
._num
_videos
+= 1
2417 if 'id' not in info_dict
:
2418 raise ExtractorError('Missing "id" field in extractor result', ie
=info_dict
['extractor'])
2419 elif not info_dict
.get('id'):
2420 raise ExtractorError('Extractor failed to obtain "id"', ie
=info_dict
['extractor'])
2422 def report_force_conversion(field
, field_not
, conversion
):
2423 self
.report_warning(
2424 '"%s" field is not %s - forcing %s conversion, there is an error in extractor'
2425 % (field
, field_not
, conversion
))
2427 def sanitize_string_field(info
, string_field
):
2428 field
= info
.get(string_field
)
2429 if field
is None or isinstance(field
, str):
2431 report_force_conversion(string_field
, 'a string', 'string')
2432 info
[string_field
] = str(field
)
2434 def sanitize_numeric_fields(info
):
2435 for numeric_field
in self
._NUMERIC
_FIELDS
:
2436 field
= info
.get(numeric_field
)
2437 if field
is None or isinstance(field
, (int, float)):
2439 report_force_conversion(numeric_field
, 'numeric', 'int')
2440 info
[numeric_field
] = int_or_none(field
)
2442 sanitize_string_field(info_dict
, 'id')
2443 sanitize_numeric_fields(info_dict
)
2444 if info_dict
.get('section_end') and info_dict
.get('section_start') is not None:
2445 info_dict
['duration'] = round(info_dict
['section_end'] - info_dict
['section_start'], 3)
2446 if (info_dict
.get('duration') or 0) <= 0 and info_dict
.pop('duration', None):
2447 self
.report_warning('"duration" field is negative, there is an error in extractor')
2449 chapters
= info_dict
.get('chapters') or []
2450 if chapters
and chapters
[0].get('start_time'):
2451 chapters
.insert(0, {'start_time': 0}
)
2453 dummy_chapter
= {'end_time': 0, 'start_time': info_dict.get('duration')}
2454 for idx
, (prev
, current
, next_
) in enumerate(zip(
2455 (dummy_chapter
, *chapters
), chapters
, (*chapters
[1:], dummy_chapter
)), 1):
2456 if current
.get('start_time') is None:
2457 current
['start_time'] = prev
.get('end_time')
2458 if not current
.get('end_time'):
2459 current
['end_time'] = next_
.get('start_time')
2460 if not current
.get('title'):
2461 current
['title'] = f
'<Untitled Chapter {idx}>'
2463 if 'playlist' not in info_dict
:
2464 # It isn't part of a playlist
2465 info_dict
['playlist'] = None
2466 info_dict
['playlist_index'] = None
2468 self
._sanitize
_thumbnails
(info_dict
)
2470 thumbnail
= info_dict
.get('thumbnail')
2471 thumbnails
= info_dict
.get('thumbnails')
2473 info_dict
['thumbnail'] = sanitize_url(thumbnail
)
2475 info_dict
['thumbnail'] = thumbnails
[-1]['url']
2477 if info_dict
.get('display_id') is None and 'id' in info_dict
:
2478 info_dict
['display_id'] = info_dict
['id']
2480 self
._fill
_common
_fields
(info_dict
)
2482 for cc_kind
in ('subtitles', 'automatic_captions'):
2483 cc
= info_dict
.get(cc_kind
)
2485 for _
, subtitle
in cc
.items():
2486 for subtitle_format
in subtitle
:
2487 if subtitle_format
.get('url'):
2488 subtitle_format
['url'] = sanitize_url(subtitle_format
['url'])
2489 if subtitle_format
.get('ext') is None:
2490 subtitle_format
['ext'] = determine_ext(subtitle_format
['url']).lower()
2492 automatic_captions
= info_dict
.get('automatic_captions')
2493 subtitles
= info_dict
.get('subtitles')
2495 info_dict
['requested_subtitles'] = self
.process_subtitles(
2496 info_dict
['id'], subtitles
, automatic_captions
)
2498 if info_dict
.get('formats') is None:
2499 # There's only one format available
2500 formats
= [info_dict
]
2502 formats
= info_dict
['formats']
2504 # or None ensures --clean-infojson removes it
2505 info_dict
['_has_drm'] = any(f
.get('has_drm') for f
in formats
) or None
2506 if not self
.params
.get('allow_unplayable_formats'):
2507 formats
= [f
for f
in formats
if not f
.get('has_drm')]
2508 if info_dict
['_has_drm'] and formats
and all(
2509 f
.get('acodec') == f
.get('vcodec') == 'none' for f
in formats
):
2510 self
.report_warning(
2511 'This video is DRM protected and only images are available for download. '
2512 'Use --list-formats to see them')
2514 get_from_start
= not info_dict
.get('is_live') or bool(self
.params
.get('live_from_start'))
2515 if not get_from_start
:
2516 info_dict
['title'] += ' ' + datetime
.datetime
.now().strftime('%Y-%m-%d %H:%M')
2517 if info_dict
.get('is_live') and formats
:
2518 formats
= [f
for f
in formats
if bool(f
.get('is_from_start')) == get_from_start
]
2519 if get_from_start
and not formats
:
2520 self
.raise_no_formats(info_dict
, msg
=(
2521 '--live-from-start is passed, but there are no formats that can be downloaded from the start. '
2522 'If you want to download from the current time, use --no-live-from-start'))
2525 self
.raise_no_formats(info_dict
)
2527 def is_wellformed(f
):
2530 self
.report_warning(
2531 '"url" field is missing or empty - skipping format, '
2532 'there is an error in extractor')
2534 if isinstance(url
, bytes):
2535 sanitize_string_field(f
, 'url')
2538 # Filter out malformed formats for better extraction robustness
2539 formats
= list(filter(is_wellformed
, formats
))
2543 # We check that all the formats have the format and format_id fields
2544 for i
, format
in enumerate(formats
):
2545 sanitize_string_field(format
, 'format_id')
2546 sanitize_numeric_fields(format
)
2547 format
['url'] = sanitize_url(format
['url'])
2548 if not format
.get('format_id'):
2549 format
['format_id'] = str(i
)
2551 # Sanitize format_id from characters used in format selector expression
2552 format
['format_id'] = re
.sub(r
'[\s,/+\[\]()]', '_', format
['format_id'])
2553 format_id
= format
['format_id']
2554 if format_id
not in formats_dict
:
2555 formats_dict
[format_id
] = []
2556 formats_dict
[format_id
].append(format
)
2558 # Make sure all formats have unique format_id
2559 common_exts
= set(itertools
.chain(*self
._format
_selection
_exts
.values()))
2560 for format_id
, ambiguous_formats
in formats_dict
.items():
2561 ambigious_id
= len(ambiguous_formats
) > 1
2562 for i
, format
in enumerate(ambiguous_formats
):
2564 format
['format_id'] = '%s-%d' % (format_id
, i
)
2565 if format
.get('ext') is None:
2566 format
['ext'] = determine_ext(format
['url']).lower()
2567 # Ensure there is no conflict between id and ext in format selection
2568 # See https://github.com/yt-dlp/yt-dlp/issues/1282
2569 if format
['format_id'] != format
['ext'] and format
['format_id'] in common_exts
:
2570 format
['format_id'] = 'f%s' % format
['format_id']
2572 for i
, format
in enumerate(formats
):
2573 if format
.get('format') is None:
2574 format
['format'] = '{id} - {res}{note}'.format(
2575 id=format
['format_id'],
2576 res
=self
.format_resolution(format
),
2577 note
=format_field(format
, 'format_note', ' (%s)'),
2579 if format
.get('protocol') is None:
2580 format
['protocol'] = determine_protocol(format
)
2581 if format
.get('resolution') is None:
2582 format
['resolution'] = self
.format_resolution(format
, default
=None)
2583 if format
.get('dynamic_range') is None and format
.get('vcodec') != 'none':
2584 format
['dynamic_range'] = 'SDR'
2585 if (info_dict
.get('duration') and format
.get('tbr')
2586 and not format
.get('filesize') and not format
.get('filesize_approx')):
2587 format
['filesize_approx'] = int(info_dict
['duration'] * format
['tbr'] * (1024 / 8))
2589 # Add HTTP headers, so that external programs can use them from the
2591 full_format_info
= info_dict
.copy()
2592 full_format_info
.update(format
)
2593 format
['http_headers'] = self
._calc
_headers
(full_format_info
)
2594 # Remove private housekeeping stuff
2595 if '__x_forwarded_for_ip' in info_dict
:
2596 del info_dict
['__x_forwarded_for_ip']
2598 if self
.params
.get('check_formats') is True:
2599 formats
= LazyList(self
._check
_formats
(formats
[::-1]), reverse
=True)
2601 if not formats
or formats
[0] is not info_dict
:
2602 # only set the 'formats' fields if the original info_dict list them
2603 # otherwise we end up with a circular reference, the first (and unique)
2604 # element in the 'formats' field in info_dict is info_dict itself,
2605 # which can't be exported to json
2606 info_dict
['formats'] = formats
2608 info_dict
, _
= self
.pre_process(info_dict
)
2610 if self
._match
_entry
(info_dict
, incomplete
=self
._format
_fields
) is not None:
2613 self
.post_extract(info_dict
)
2614 info_dict
, _
= self
.pre_process(info_dict
, 'after_filter')
2616 # The pre-processors may have modified the formats
2617 formats
= info_dict
.get('formats', [info_dict
])
2619 list_only
= self
.params
.get('simulate') is None and (
2620 self
.params
.get('list_thumbnails') or self
.params
.get('listformats') or self
.params
.get('listsubtitles'))
2621 interactive_format_selection
= not list_only
and self
.format_selector
== '-'
2622 if self
.params
.get('list_thumbnails'):
2623 self
.list_thumbnails(info_dict
)
2624 if self
.params
.get('listsubtitles'):
2625 if 'automatic_captions' in info_dict
:
2626 self
.list_subtitles(
2627 info_dict
['id'], automatic_captions
, 'automatic captions')
2628 self
.list_subtitles(info_dict
['id'], subtitles
, 'subtitles')
2629 if self
.params
.get('listformats') or interactive_format_selection
:
2630 self
.list_formats(info_dict
)
2632 # Without this printing, -F --print-json will not work
2633 self
.__forced
_printings
(info_dict
, self
.prepare_filename(info_dict
), incomplete
=True)
2636 format_selector
= self
.format_selector
2637 if format_selector
is None:
2638 req_format
= self
._default
_format
_spec
(info_dict
, download
=download
)
2639 self
.write_debug('Default format spec: %s' % req_format
)
2640 format_selector
= self
.build_format_selector(req_format
)
2643 if interactive_format_selection
:
2645 self
._format
_screen
('\nEnter format selector: ', self
.Styles
.EMPHASIS
))
2647 format_selector
= self
.build_format_selector(req_format
)
2648 except SyntaxError as err
:
2649 self
.report_error(err
, tb
=False, is_error
=False)
2652 formats_to_download
= list(format_selector({
2654 'has_merged_format': any('none' not in (f
.get('acodec'), f
.get('vcodec')) for f
in formats
),
2655 'incomplete_formats': (
2656 # All formats are video-only or
2657 all(f
.get('vcodec') != 'none' and f
.get('acodec') == 'none' for f
in formats
)
2658 # all formats are audio-only
2659 or all(f
.get('vcodec') == 'none' and f
.get('acodec') != 'none' for f
in formats
)),
2661 if interactive_format_selection
and not formats_to_download
:
2662 self
.report_error('Requested format is not available', tb
=False, is_error
=False)
2666 if not formats_to_download
:
2667 if not self
.params
.get('ignore_no_formats_error'):
2668 raise ExtractorError(
2669 'Requested format is not available. Use --list-formats for a list of available formats',
2670 expected
=True, video_id
=info_dict
['id'], ie
=info_dict
['extractor'])
2671 self
.report_warning('Requested format is not available')
2672 # Process what we can, even without any available formats.
2673 formats_to_download
= [{}]
2675 requested_ranges
= self
.params
.get('download_ranges')
2676 if requested_ranges
:
2677 requested_ranges
= tuple(requested_ranges(info_dict
, self
))
2679 best_format
, downloaded_formats
= formats_to_download
[-1], []
2682 def to_screen(*msg
):
2683 self
.to_screen(f
'[info] {info_dict["id"]}: {" ".join(", ".join(variadic(m)) for m in msg)}')
2685 to_screen(f
'Downloading {len(formats_to_download)} format(s):',
2686 (f
['format_id'] for f
in formats_to_download
))
2687 if requested_ranges
:
2688 to_screen(f
'Downloading {len(requested_ranges)} time ranges:',
2689 (f
'{int(c["start_time"])}-{int(c["end_time"])}' for c
in requested_ranges
))
2690 max_downloads_reached
= False
2692 for fmt
, chapter
in itertools
.product(formats_to_download
, requested_ranges
or [{}]):
2693 new_info
= self
._copy
_infodict
(info_dict
)
2694 new_info
.update(fmt
)
2695 offset
, duration
= info_dict
.get('section_start') or 0, info_dict
.get('duration') or float('inf')
2696 if chapter
or offset
:
2698 'section_start': offset
+ chapter
.get('start_time', 0),
2699 'section_end': offset
+ min(chapter
.get('end_time', duration
), duration
),
2700 'section_title': chapter
.get('title'),
2701 'section_number': chapter
.get('index'),
2703 downloaded_formats
.append(new_info
)
2705 self
.process_info(new_info
)
2706 except MaxDownloadsReached
:
2707 max_downloads_reached
= True
2708 self
._raise
_pending
_errors
(new_info
)
2709 # Remove copied info
2710 for key
, val
in tuple(new_info
.items()):
2711 if info_dict
.get(key
) == val
:
2713 if max_downloads_reached
:
2716 write_archive
= {f.get('__write_download_archive', False) for f in downloaded_formats}
2717 assert write_archive
.issubset({True, False, 'ignore'}
)
2718 if True in write_archive
and False not in write_archive
:
2719 self
.record_download_archive(info_dict
)
2721 info_dict
['requested_downloads'] = downloaded_formats
2722 info_dict
= self
.run_all_pps('after_video', info_dict
)
2723 if max_downloads_reached
:
2724 raise MaxDownloadsReached()
2726 # We update the info dict with the selected best quality format (backwards compatibility)
2727 info_dict
.update(best_format
)
2730 def process_subtitles(self
, video_id
, normal_subtitles
, automatic_captions
):
2731 """Select the requested subtitles and their format"""
2732 available_subs
, normal_sub_langs
= {}, []
2733 if normal_subtitles
and self
.params
.get('writesubtitles'):
2734 available_subs
.update(normal_subtitles
)
2735 normal_sub_langs
= tuple(normal_subtitles
.keys())
2736 if automatic_captions
and self
.params
.get('writeautomaticsub'):
2737 for lang
, cap_info
in automatic_captions
.items():
2738 if lang
not in available_subs
:
2739 available_subs
[lang
] = cap_info
2741 if (not self
.params
.get('writesubtitles') and not
2742 self
.params
.get('writeautomaticsub') or not
2746 all_sub_langs
= tuple(available_subs
.keys())
2747 if self
.params
.get('allsubtitles', False):
2748 requested_langs
= all_sub_langs
2749 elif self
.params
.get('subtitleslangs', False):
2751 requested_langs
= orderedSet_from_options(
2752 self
.params
.get('subtitleslangs'), {'all': all_sub_langs}
, use_regex
=True)
2753 except re
.error
as e
:
2754 raise ValueError(f
'Wrong regex for subtitlelangs: {e.pattern}')
2755 elif normal_sub_langs
:
2756 requested_langs
= ['en'] if 'en' in normal_sub_langs
else normal_sub_langs
[:1]
2758 requested_langs
= ['en'] if 'en' in all_sub_langs
else all_sub_langs
[:1]
2760 self
.write_debug('Downloading subtitles: %s' % ', '.join(requested_langs
))
2762 formats_query
= self
.params
.get('subtitlesformat', 'best')
2763 formats_preference
= formats_query
.split('/') if formats_query
else []
2765 for lang
in requested_langs
:
2766 formats
= available_subs
.get(lang
)
2768 self
.report_warning(f
'{lang} subtitles not available for {video_id}')
2770 for ext
in formats_preference
:
2774 matches
= list(filter(lambda f
: f
['ext'] == ext
, formats
))
2780 self
.report_warning(
2781 'No subtitle format found matching "%s" for language %s, '
2782 'using %s' % (formats_query
, lang
, f
['ext']))
2786 def _forceprint(self
, key
, info_dict
):
2787 if info_dict
is None:
2789 info_copy
= info_dict
.copy()
2790 info_copy
['formats_table'] = self
.render_formats_table(info_dict
)
2791 info_copy
['thumbnails_table'] = self
.render_thumbnails_table(info_dict
)
2792 info_copy
['subtitles_table'] = self
.render_subtitles_table(info_dict
.get('id'), info_dict
.get('subtitles'))
2793 info_copy
['automatic_captions_table'] = self
.render_subtitles_table(info_dict
.get('id'), info_dict
.get('automatic_captions'))
2795 def format_tmpl(tmpl
):
2796 mobj
= re
.match(r
'\w+(=?)$', tmpl
)
2797 if mobj
and mobj
.group(1):
2798 return f
'{tmpl[:-1]} = %({tmpl[:-1]})r'
2800 return f
'%({tmpl})s'
2803 for tmpl
in self
.params
['forceprint'].get(key
, []):
2804 self
.to_stdout(self
.evaluate_outtmpl(format_tmpl(tmpl
), info_copy
))
2806 for tmpl
, file_tmpl
in self
.params
['print_to_file'].get(key
, []):
2807 filename
= self
.prepare_filename(info_dict
, outtmpl
=file_tmpl
)
2808 tmpl
= format_tmpl(tmpl
)
2809 self
.to_screen(f
'[info] Writing {tmpl!r} to: {filename}')
2810 if self
._ensure
_dir
_exists
(filename
):
2811 with open(filename
, 'a', encoding
='utf-8') as f
:
2812 f
.write(self
.evaluate_outtmpl(tmpl
, info_copy
) + '\n')
2814 def __forced_printings(self
, info_dict
, filename
, incomplete
):
2815 def print_mandatory(field
, actual_field
=None):
2816 if actual_field
is None:
2817 actual_field
= field
2818 if (self
.params
.get('force%s' % field
, False)
2819 and (not incomplete
or info_dict
.get(actual_field
) is not None)):
2820 self
.to_stdout(info_dict
[actual_field
])
2822 def print_optional(field
):
2823 if (self
.params
.get('force%s' % field
, False)
2824 and info_dict
.get(field
) is not None):
2825 self
.to_stdout(info_dict
[field
])
2827 info_dict
= info_dict
.copy()
2828 if filename
is not None:
2829 info_dict
['filename'] = filename
2830 if info_dict
.get('requested_formats') is not None:
2831 # For RTMP URLs, also include the playpath
2832 info_dict
['urls'] = '\n'.join(f
['url'] + f
.get('play_path', '') for f
in info_dict
['requested_formats'])
2833 elif info_dict
.get('url'):
2834 info_dict
['urls'] = info_dict
['url'] + info_dict
.get('play_path', '')
2836 if (self
.params
.get('forcejson')
2837 or self
.params
['forceprint'].get('video')
2838 or self
.params
['print_to_file'].get('video')):
2839 self
.post_extract(info_dict
)
2840 self
._forceprint
('video', info_dict
)
2842 print_mandatory('title')
2843 print_mandatory('id')
2844 print_mandatory('url', 'urls')
2845 print_optional('thumbnail')
2846 print_optional('description')
2847 print_optional('filename')
2848 if self
.params
.get('forceduration') and info_dict
.get('duration') is not None:
2849 self
.to_stdout(formatSeconds(info_dict
['duration']))
2850 print_mandatory('format')
2852 if self
.params
.get('forcejson'):
2853 self
.to_stdout(json
.dumps(self
.sanitize_info(info_dict
)))
2855 def dl(self
, name
, info
, subtitle
=False, test
=False):
2856 if not info
.get('url'):
2857 self
.raise_no_formats(info
, True)
2860 verbose
= self
.params
.get('verbose')
2863 'quiet': self
.params
.get('quiet') or not verbose
,
2865 'noprogress': not verbose
,
2867 'skip_unavailable_fragments': False,
2868 'keep_fragments': False,
2870 '_no_ytdl_file': True,
2873 params
= self
.params
2874 fd
= get_suitable_downloader(info
, params
, to_stdout
=(name
== '-'))(self
, params
)
2876 for ph
in self
._progress
_hooks
:
2877 fd
.add_progress_hook(ph
)
2879 (f
['url'].split(',')[0] + ',<data>' if f
['url'].startswith('data:') else f
['url'])
2880 for f
in info
.get('requested_formats', []) or [info
])
2881 self
.write_debug(f
'Invoking {fd.FD_NAME} downloader on "{urls}"')
2883 # Note: Ideally info should be a deep-copied so that hooks cannot modify it.
2884 # But it may contain objects that are not deep-copyable
2885 new_info
= self
._copy
_infodict
(info
)
2886 if new_info
.get('http_headers') is None:
2887 new_info
['http_headers'] = self
._calc
_headers
(new_info
)
2888 return fd
.download(name
, new_info
, subtitle
)
2890 def existing_file(self
, filepaths
, *, default_overwrite
=True):
2891 existing_files
= list(filter(os
.path
.exists
, orderedSet(filepaths
)))
2892 if existing_files
and not self
.params
.get('overwrites', default_overwrite
):
2893 return existing_files
[0]
2895 for file in existing_files
:
2896 self
.report_file_delete(file)
2900 def process_info(self
, info_dict
):
2901 """Process a single resolved IE result. (Modifies it in-place)"""
2903 assert info_dict
.get('_type', 'video') == 'video'
2904 original_infodict
= info_dict
2906 if 'format' not in info_dict
and 'ext' in info_dict
:
2907 info_dict
['format'] = info_dict
['ext']
2909 # This is mostly just for backward compatibility of process_info
2910 # As a side-effect, this allows for format-specific filters
2911 if self
._match
_entry
(info_dict
) is not None:
2912 info_dict
['__write_download_archive'] = 'ignore'
2915 # Does nothing under normal operation - for backward compatibility of process_info
2916 self
.post_extract(info_dict
)
2917 self
._num
_downloads
+= 1
2919 # info_dict['_filename'] needs to be set for backward compatibility
2920 info_dict
['_filename'] = full_filename
= self
.prepare_filename(info_dict
, warn
=True)
2921 temp_filename
= self
.prepare_filename(info_dict
, 'temp')
2925 self
.__forced
_printings
(info_dict
, full_filename
, incomplete
=('format' not in info_dict
))
2927 def check_max_downloads():
2928 if self
._num
_downloads
>= float(self
.params
.get('max_downloads') or 'inf'):
2929 raise MaxDownloadsReached()
2931 if self
.params
.get('simulate'):
2932 info_dict
['__write_download_archive'] = self
.params
.get('force_write_download_archive')
2933 check_max_downloads()
2936 if full_filename
is None:
2938 if not self
._ensure
_dir
_exists
(encodeFilename(full_filename
)):
2940 if not self
._ensure
_dir
_exists
(encodeFilename(temp_filename
)):
2943 if self
._write
_description
('video', info_dict
,
2944 self
.prepare_filename(info_dict
, 'description')) is None:
2947 sub_files
= self
._write
_subtitles
(info_dict
, temp_filename
)
2948 if sub_files
is None:
2950 files_to_move
.update(dict(sub_files
))
2952 thumb_files
= self
._write
_thumbnails
(
2953 'video', info_dict
, temp_filename
, self
.prepare_filename(info_dict
, 'thumbnail'))
2954 if thumb_files
is None:
2956 files_to_move
.update(dict(thumb_files
))
2958 infofn
= self
.prepare_filename(info_dict
, 'infojson')
2959 _infojson_written
= self
._write
_info
_json
('video', info_dict
, infofn
)
2960 if _infojson_written
:
2961 info_dict
['infojson_filename'] = infofn
2962 # For backward compatibility, even though it was a private field
2963 info_dict
['__infojson_filename'] = infofn
2964 elif _infojson_written
is None:
2967 # Note: Annotations are deprecated
2969 if self
.params
.get('writeannotations', False):
2970 annofn
= self
.prepare_filename(info_dict
, 'annotation')
2972 if not self
._ensure
_dir
_exists
(encodeFilename(annofn
)):
2974 if not self
.params
.get('overwrites', True) and os
.path
.exists(encodeFilename(annofn
)):
2975 self
.to_screen('[info] Video annotations are already present')
2976 elif not info_dict
.get('annotations'):
2977 self
.report_warning('There are no annotations to write.')
2980 self
.to_screen('[info] Writing video annotations to: ' + annofn
)
2981 with open(encodeFilename(annofn
), 'w', encoding
='utf-8') as annofile
:
2982 annofile
.write(info_dict
['annotations'])
2983 except (KeyError, TypeError):
2984 self
.report_warning('There are no annotations to write.')
2986 self
.report_error('Cannot write annotations file: ' + annofn
)
2989 # Write internet shortcut files
2990 def _write_link_file(link_type
):
2991 url
= try_get(info_dict
['webpage_url'], iri_to_uri
)
2993 self
.report_warning(
2994 f
'Cannot write internet shortcut file because the actual URL of "{info_dict["webpage_url"]}" is unknown')
2996 linkfn
= replace_extension(self
.prepare_filename(info_dict
, 'link'), link_type
, info_dict
.get('ext'))
2997 if not self
._ensure
_dir
_exists
(encodeFilename(linkfn
)):
2999 if self
.params
.get('overwrites', True) and os
.path
.exists(encodeFilename(linkfn
)):
3000 self
.to_screen(f
'[info] Internet shortcut (.{link_type}) is already present')
3003 self
.to_screen(f
'[info] Writing internet shortcut (.{link_type}) to: {linkfn}')
3004 with open(encodeFilename(to_high_limit_path(linkfn
)), 'w', encoding
='utf-8',
3005 newline
='\r\n' if link_type
== 'url' else '\n') as linkfile
:
3006 template_vars
= {'url': url}
3007 if link_type
== 'desktop':
3008 template_vars
['filename'] = linkfn
[:-(len(link_type
) + 1)]
3009 linkfile
.write(LINK_TEMPLATES
[link_type
] % template_vars
)
3011 self
.report_error(f
'Cannot write internet shortcut {linkfn}')
3016 'url': self
.params
.get('writeurllink'),
3017 'webloc': self
.params
.get('writewebloclink'),
3018 'desktop': self
.params
.get('writedesktoplink'),
3020 if self
.params
.get('writelink'):
3021 link_type
= ('webloc' if sys
.platform
== 'darwin'
3022 else 'desktop' if sys
.platform
.startswith('linux')
3024 write_links
[link_type
] = True
3026 if any(should_write
and not _write_link_file(link_type
)
3027 for link_type
, should_write
in write_links
.items()):
3030 def replace_info_dict(new_info
):
3032 if new_info
== info_dict
:
3035 info_dict
.update(new_info
)
3037 new_info
, files_to_move
= self
.pre_process(info_dict
, 'before_dl', files_to_move
)
3038 replace_info_dict(new_info
)
3040 if self
.params
.get('skip_download'):
3041 info_dict
['filepath'] = temp_filename
3042 info_dict
['__finaldir'] = os
.path
.dirname(os
.path
.abspath(encodeFilename(full_filename
)))
3043 info_dict
['__files_to_move'] = files_to_move
3044 replace_info_dict(self
.run_pp(MoveFilesAfterDownloadPP(self
, False), info_dict
))
3045 info_dict
['__write_download_archive'] = self
.params
.get('force_write_download_archive')
3048 info_dict
.setdefault('__postprocessors', [])
3051 def existing_video_file(*filepaths
):
3052 ext
= info_dict
.get('ext')
3053 converted
= lambda file: replace_extension(file, self
.params
.get('final_ext') or ext
, ext
)
3054 file = self
.existing_file(itertools
.chain(*zip(map(converted
, filepaths
), filepaths
)),
3055 default_overwrite
=False)
3057 info_dict
['ext'] = os
.path
.splitext(file)[1][1:]
3060 fd
, success
= None, True
3061 if info_dict
.get('protocol') or info_dict
.get('url'):
3062 fd
= get_suitable_downloader(info_dict
, self
.params
, to_stdout
=temp_filename
== '-')
3063 if fd
is not FFmpegFD
and (
3064 info_dict
.get('section_start') or info_dict
.get('section_end')):
3065 msg
= ('This format cannot be partially downloaded' if FFmpegFD
.available()
3066 else 'You have requested downloading the video partially, but ffmpeg is not installed')
3067 self
.report_error(f
'{msg}. Aborting')
3070 if info_dict
.get('requested_formats') is not None:
3071 requested_formats
= info_dict
['requested_formats']
3072 old_ext
= info_dict
['ext']
3073 if self
.params
.get('merge_output_format') is None:
3074 if (info_dict
['ext'] == 'webm'
3075 and info_dict
.get('thumbnails')
3076 # check with type instead of pp_key, __name__, or isinstance
3077 # since we dont want any custom PPs to trigger this
3078 and any(type(pp
) == EmbedThumbnailPP
for pp
in self
._pps
['post_process'])): # noqa: E721
3079 info_dict
['ext'] = 'mkv'
3080 self
.report_warning(
3081 'webm doesn\'t support embedding a thumbnail, mkv will be used')
3082 new_ext
= info_dict
['ext']
3084 def correct_ext(filename
, ext
=new_ext
):
3087 filename_real_ext
= os
.path
.splitext(filename
)[1][1:]
3089 os
.path
.splitext(filename
)[0]
3090 if filename_real_ext
in (old_ext
, new_ext
)
3092 return f
'{filename_wo_ext}.{ext}'
3094 # Ensure filename always has a correct extension for successful merge
3095 full_filename
= correct_ext(full_filename
)
3096 temp_filename
= correct_ext(temp_filename
)
3097 dl_filename
= existing_video_file(full_filename
, temp_filename
)
3098 info_dict
['__real_download'] = False
3100 merger
= FFmpegMergerPP(self
)
3102 if dl_filename
is not None:
3103 self
.report_file_already_downloaded(dl_filename
)
3105 for f
in requested_formats
if fd
!= FFmpegFD
else []:
3106 f
['filepath'] = fname
= prepend_extension(
3107 correct_ext(temp_filename
, info_dict
['ext']),
3108 'f%s' % f
['format_id'], info_dict
['ext'])
3109 downloaded
.append(fname
)
3110 info_dict
['url'] = '\n'.join(f
['url'] for f
in requested_formats
)
3111 success
, real_download
= self
.dl(temp_filename
, info_dict
)
3112 info_dict
['__real_download'] = real_download
3114 if self
.params
.get('allow_unplayable_formats'):
3115 self
.report_warning(
3116 'You have requested merging of multiple formats '
3117 'while also allowing unplayable formats to be downloaded. '
3118 'The formats won\'t be merged to prevent data corruption.')
3119 elif not merger
.available
:
3120 msg
= 'You have requested merging of multiple formats but ffmpeg is not installed'
3121 if not self
.params
.get('ignoreerrors'):
3122 self
.report_error(f
'{msg}. Aborting due to --abort-on-error')
3124 self
.report_warning(f
'{msg}. The formats won\'t be merged')
3126 if temp_filename
== '-':
3127 reason
= ('using a downloader other than ffmpeg' if FFmpegFD
.can_merge_formats(info_dict
, self
.params
)
3128 else 'but the formats are incompatible for simultaneous download' if merger
.available
3129 else 'but ffmpeg is not installed')
3130 self
.report_warning(
3131 f
'You have requested downloading multiple formats to stdout {reason}. '
3132 'The formats will be streamed one after the other')
3133 fname
= temp_filename
3134 for f
in requested_formats
:
3135 new_info
= dict(info_dict
)
3136 del new_info
['requested_formats']
3138 if temp_filename
!= '-':
3139 fname
= prepend_extension(
3140 correct_ext(temp_filename
, new_info
['ext']),
3141 'f%s' % f
['format_id'], new_info
['ext'])
3142 if not self
._ensure
_dir
_exists
(fname
):
3144 f
['filepath'] = fname
3145 downloaded
.append(fname
)
3146 partial_success
, real_download
= self
.dl(fname
, new_info
)
3147 info_dict
['__real_download'] = info_dict
['__real_download'] or real_download
3148 success
= success
and partial_success
3150 if downloaded
and merger
.available
and not self
.params
.get('allow_unplayable_formats'):
3151 info_dict
['__postprocessors'].append(merger
)
3152 info_dict
['__files_to_merge'] = downloaded
3153 # Even if there were no downloads, it is being merged only now
3154 info_dict
['__real_download'] = True
3156 for file in downloaded
:
3157 files_to_move
[file] = None
3159 # Just a single file
3160 dl_filename
= existing_video_file(full_filename
, temp_filename
)
3161 if dl_filename
is None or dl_filename
== temp_filename
:
3162 # dl_filename == temp_filename could mean that the file was partially downloaded with --no-part.
3163 # So we should try to resume the download
3164 success
, real_download
= self
.dl(temp_filename
, info_dict
)
3165 info_dict
['__real_download'] = real_download
3167 self
.report_file_already_downloaded(dl_filename
)
3169 dl_filename
= dl_filename
or temp_filename
3170 info_dict
['__finaldir'] = os
.path
.dirname(os
.path
.abspath(encodeFilename(full_filename
)))
3172 except network_exceptions
as err
:
3173 self
.report_error('unable to download video data: %s' % error_to_compat_str(err
))
3175 except OSError as err
:
3176 raise UnavailableVideoError(err
)
3177 except (ContentTooShortError
, ) as err
:
3178 self
.report_error(f
'content too short (expected {err.expected} bytes and served {err.downloaded})')
3181 self
._raise
_pending
_errors
(info_dict
)
3182 if success
and full_filename
!= '-':
3186 fixup_policy
= self
.params
.get('fixup')
3187 vid
= info_dict
['id']
3189 if fixup_policy
in ('ignore', 'never'):
3191 elif fixup_policy
== 'warn':
3193 elif fixup_policy
!= 'force':
3194 assert fixup_policy
in ('detect_or_warn', None)
3195 if not info_dict
.get('__real_download'):
3198 def ffmpeg_fixup(cndn
, msg
, cls
):
3199 if not (do_fixup
and cndn
):
3201 elif do_fixup
== 'warn':
3202 self
.report_warning(f
'{vid}: {msg}')
3206 info_dict
['__postprocessors'].append(pp
)
3208 self
.report_warning(f
'{vid}: {msg}. Install ffmpeg to fix this automatically')
3210 stretched_ratio
= info_dict
.get('stretched_ratio')
3211 ffmpeg_fixup(stretched_ratio
not in (1, None),
3212 f
'Non-uniform pixel ratio {stretched_ratio}',
3213 FFmpegFixupStretchedPP
)
3215 downloader
= get_suitable_downloader(info_dict
, self
.params
) if 'protocol' in info_dict
else None
3216 downloader
= downloader
.FD_NAME
if downloader
else None
3218 ext
= info_dict
.get('ext')
3219 postprocessed_by_ffmpeg
= info_dict
.get('requested_formats') or any((
3220 isinstance(pp
, FFmpegVideoConvertorPP
)
3221 and resolve_recode_mapping(ext
, pp
.mapping
)[0] not in (ext
, None)
3222 ) for pp
in self
._pps
['post_process'])
3224 if not postprocessed_by_ffmpeg
:
3225 ffmpeg_fixup(ext
== 'm4a' and info_dict
.get('container') == 'm4a_dash',
3226 'writing DASH m4a. Only some players support this container',
3228 ffmpeg_fixup(downloader
== 'hlsnative' and not self
.params
.get('hls_use_mpegts')
3229 or info_dict
.get('is_live') and self
.params
.get('hls_use_mpegts') is None,
3230 'Possible MPEG-TS in MP4 container or malformed AAC timestamps',
3232 ffmpeg_fixup(info_dict
.get('is_live') and downloader
== 'DashSegmentsFD',
3233 'Possible duplicate MOOV atoms', FFmpegFixupDuplicateMoovPP
)
3235 ffmpeg_fixup(downloader
== 'web_socket_fragment', 'Malformed timestamps detected', FFmpegFixupTimestampPP
)
3236 ffmpeg_fixup(downloader
== 'web_socket_fragment', 'Malformed duration detected', FFmpegFixupDurationPP
)
3240 replace_info_dict(self
.post_process(dl_filename
, info_dict
, files_to_move
))
3241 except PostProcessingError
as err
:
3242 self
.report_error('Postprocessing: %s' % str(err
))
3245 for ph
in self
._post
_hooks
:
3246 ph(info_dict
['filepath'])
3247 except Exception as err
:
3248 self
.report_error('post hooks: %s' % str(err
))
3250 info_dict
['__write_download_archive'] = True
3252 assert info_dict
is original_infodict
# Make sure the info_dict was modified in-place
3253 if self
.params
.get('force_write_download_archive'):
3254 info_dict
['__write_download_archive'] = True
3255 check_max_downloads()
3257 def __download_wrapper(self
, func
):
3258 @functools.wraps(func
)
3259 def wrapper(*args
, **kwargs
):
3261 res
= func(*args
, **kwargs
)
3262 except UnavailableVideoError
as e
:
3263 self
.report_error(e
)
3264 except DownloadCancelled
as e
:
3265 self
.to_screen(f
'[info] {e}')
3266 if not self
.params
.get('break_per_url'):
3269 if self
.params
.get('dump_single_json', False):
3270 self
.post_extract(res
)
3271 self
.to_stdout(json
.dumps(self
.sanitize_info(res
)))
3274 def download(self
, url_list
):
3275 """Download a given list of URLs."""
3276 url_list
= variadic(url_list
) # Passing a single URL is a common mistake
3277 outtmpl
= self
.params
['outtmpl']['default']
3278 if (len(url_list
) > 1
3280 and '%' not in outtmpl
3281 and self
.params
.get('max_downloads') != 1):
3282 raise SameFileError(outtmpl
)
3284 for url
in url_list
:
3285 self
.__download
_wrapper
(self
.extract_info
)(
3286 url
, force_generic_extractor
=self
.params
.get('force_generic_extractor', False))
3288 return self
._download
_retcode
3290 def download_with_info_file(self
, info_filename
):
3291 with contextlib
.closing(fileinput
.FileInput(
3292 [info_filename
], mode
='r',
3293 openhook
=fileinput
.hook_encoded('utf-8'))) as f
:
3294 # FileInput doesn't have a read method, we can't call json.load
3295 info
= self
.sanitize_info(json
.loads('\n'.join(f
)), self
.params
.get('clean_infojson', True))
3297 self
.__download
_wrapper
(self
.process_ie_result
)(info
, download
=True)
3298 except (DownloadError
, EntryNotInPlaylist
, ReExtractInfo
) as e
:
3299 if not isinstance(e
, EntryNotInPlaylist
):
3300 self
.to_stderr('\r')
3301 webpage_url
= info
.get('webpage_url')
3302 if webpage_url
is not None:
3303 self
.report_warning(f
'The info failed to download: {e}; trying with URL {webpage_url}')
3304 return self
.download([webpage_url
])
3307 return self
._download
_retcode
3310 def sanitize_info(info_dict
, remove_private_keys
=False):
3311 ''' Sanitize the infodict for converting to json '''
3312 if info_dict
is None:
3314 info_dict
.setdefault('epoch', int(time
.time()))
3315 info_dict
.setdefault('_type', 'video')
3317 if remove_private_keys
:
3318 reject
= lambda k
, v
: v
is None or k
.startswith('__') or k
in {
3319 'requested_downloads', 'requested_formats', 'requested_subtitles', 'requested_entries',
3320 'entries', 'filepath', '_filename', 'infojson_filename', 'original_url', 'playlist_autonumber',
3323 reject
= lambda k
, v
: False
3326 if isinstance(obj
, dict):
3327 return {k: filter_fn(v) for k, v in obj.items() if not reject(k, v)}
3328 elif isinstance(obj
, (list, tuple, set, LazyList
)):
3329 return list(map(filter_fn
, obj
))
3330 elif obj
is None or isinstance(obj
, (str, int, float, bool)):
3335 return filter_fn(info_dict
)
3338 def filter_requested_info(info_dict
, actually_filter
=True):
3339 ''' Alias of sanitize_info for backward compatibility '''
3340 return YoutubeDL
.sanitize_info(info_dict
, actually_filter
)
3342 def _delete_downloaded_files(self
, *files_to_delete
, info
={}, msg
=None):
3343 for filename
in set(filter(None, files_to_delete
)):
3345 self
.to_screen(msg
% filename
)
3349 self
.report_warning(f
'Unable to delete file {filename}')
3350 if filename
in info
.get('__files_to_move', []): # NB: Delete even if None
3351 del info
['__files_to_move'][filename
]
3354 def post_extract(info_dict
):
3355 def actual_post_extract(info_dict
):
3356 if info_dict
.get('_type') in ('playlist', 'multi_video'):
3357 for video_dict
in info_dict
.get('entries', {}):
3358 actual_post_extract(video_dict
or {})
3361 post_extractor
= info_dict
.pop('__post_extractor', None) or (lambda: {})
3362 info_dict
.update(post_extractor())
3364 actual_post_extract(info_dict
or {})
3366 def run_pp(self
, pp
, infodict
):
3367 files_to_delete
= []
3368 if '__files_to_move' not in infodict
:
3369 infodict
['__files_to_move'] = {}
3371 files_to_delete
, infodict
= pp
.run(infodict
)
3372 except PostProcessingError
as e
:
3373 # Must be True and not 'only_download'
3374 if self
.params
.get('ignoreerrors') is True:
3375 self
.report_error(e
)
3379 if not files_to_delete
:
3381 if self
.params
.get('keepvideo', False):
3382 for f
in files_to_delete
:
3383 infodict
['__files_to_move'].setdefault(f
, '')
3385 self
._delete
_downloaded
_files
(
3386 *files_to_delete
, info
=infodict
, msg
='Deleting original file %s (pass -k to keep)')
3389 def run_all_pps(self
, key
, info
, *, additional_pps
=None):
3390 self
._forceprint
(key
, info
)
3391 for pp
in (additional_pps
or []) + self
._pps
[key
]:
3392 info
= self
.run_pp(pp
, info
)
3395 def pre_process(self
, ie_info
, key
='pre_process', files_to_move
=None):
3396 info
= dict(ie_info
)
3397 info
['__files_to_move'] = files_to_move
or {}
3399 info
= self
.run_all_pps(key
, info
)
3400 except PostProcessingError
as err
:
3401 msg
= f
'Preprocessing: {err}'
3402 info
.setdefault('__pending_error', msg
)
3403 self
.report_error(msg
, is_error
=False)
3404 return info
, info
.pop('__files_to_move', None)
3406 def post_process(self
, filename
, info
, files_to_move
=None):
3407 """Run all the postprocessors on the given file."""
3408 info
['filepath'] = filename
3409 info
['__files_to_move'] = files_to_move
or {}
3410 info
= self
.run_all_pps('post_process', info
, additional_pps
=info
.get('__postprocessors'))
3411 info
= self
.run_pp(MoveFilesAfterDownloadPP(self
), info
)
3412 del info
['__files_to_move']
3413 return self
.run_all_pps('after_move', info
)
3415 def _make_archive_id(self
, info_dict
):
3416 video_id
= info_dict
.get('id')
3419 # Future-proof against any change in case
3420 # and backwards compatibility with prior versions
3421 extractor
= info_dict
.get('extractor_key') or info_dict
.get('ie_key') # key in a playlist
3422 if extractor
is None:
3423 url
= str_or_none(info_dict
.get('url'))
3426 # Try to find matching extractor for the URL and take its ie_key
3427 for ie_key
, ie
in self
._ies
.items():
3428 if ie
.suitable(url
):
3433 return make_archive_id(extractor
, video_id
)
3435 def in_download_archive(self
, info_dict
):
3436 fn
= self
.params
.get('download_archive')
3440 vid_ids
= [self
._make
_archive
_id
(info_dict
)]
3441 vid_ids
.extend(info_dict
.get('_old_archive_ids') or [])
3442 return any(id_
in self
.archive
for id_
in vid_ids
)
3444 def record_download_archive(self
, info_dict
):
3445 fn
= self
.params
.get('download_archive')
3448 vid_id
= self
._make
_archive
_id
(info_dict
)
3450 self
.write_debug(f
'Adding to archive: {vid_id}')
3451 with locked_file(fn
, 'a', encoding
='utf-8') as archive_file
:
3452 archive_file
.write(vid_id
+ '\n')
3453 self
.archive
.add(vid_id
)
3456 def format_resolution(format
, default
='unknown'):
3457 if format
.get('vcodec') == 'none' and format
.get('acodec') != 'none':
3459 if format
.get('resolution') is not None:
3460 return format
['resolution']
3461 if format
.get('width') and format
.get('height'):
3462 return '%dx%d' % (format
['width'], format
['height'])
3463 elif format
.get('height'):
3464 return '%sp' % format
['height']
3465 elif format
.get('width'):
3466 return '%dx?' % format
['width']
3469 def _list_format_headers(self
, *headers
):
3470 if self
.params
.get('listformats_table', True) is not False:
3471 return [self
._format
_out
(header
, self
.Styles
.HEADERS
) for header
in headers
]
3474 def _format_note(self
, fdict
):
3476 if fdict
.get('ext') in ['f4f', 'f4m']:
3477 res
+= '(unsupported)'
3478 if fdict
.get('language'):
3481 res
+= '[%s]' % fdict
['language']
3482 if fdict
.get('format_note') is not None:
3485 res
+= fdict
['format_note']
3486 if fdict
.get('tbr') is not None:
3489 res
+= '%4dk' % fdict
['tbr']
3490 if fdict
.get('container') is not None:
3493 res
+= '%s container' % fdict
['container']
3494 if (fdict
.get('vcodec') is not None
3495 and fdict
.get('vcodec') != 'none'):
3498 res
+= fdict
['vcodec']
3499 if fdict
.get('vbr') is not None:
3501 elif fdict
.get('vbr') is not None and fdict
.get('abr') is not None:
3503 if fdict
.get('vbr') is not None:
3504 res
+= '%4dk' % fdict
['vbr']
3505 if fdict
.get('fps') is not None:
3508 res
+= '%sfps' % fdict
['fps']
3509 if fdict
.get('acodec') is not None:
3512 if fdict
['acodec'] == 'none':
3515 res
+= '%-5s' % fdict
['acodec']
3516 elif fdict
.get('abr') is not None:
3520 if fdict
.get('abr') is not None:
3521 res
+= '@%3dk' % fdict
['abr']
3522 if fdict
.get('asr') is not None:
3523 res
+= ' (%5dHz)' % fdict
['asr']
3524 if fdict
.get('filesize') is not None:
3527 res
+= format_bytes(fdict
['filesize'])
3528 elif fdict
.get('filesize_approx') is not None:
3531 res
+= '~' + format_bytes(fdict
['filesize_approx'])
3534 def render_formats_table(self
, info_dict
):
3535 if not info_dict
.get('formats') and not info_dict
.get('url'):
3538 formats
= info_dict
.get('formats', [info_dict
])
3539 if not self
.params
.get('listformats_table', True) is not False:
3542 format_field(f
, 'format_id'),
3543 format_field(f
, 'ext'),
3544 self
.format_resolution(f
),
3545 self
._format
_note
(f
)
3546 ] for f
in formats
if f
.get('preference') is None or f
['preference'] >= -1000]
3547 return render_table(['format code', 'extension', 'resolution', 'note'], table
, extra_gap
=1)
3549 def simplified_codec(f
, field
):
3550 assert field
in ('acodec', 'vcodec')
3551 codec
= f
.get(field
, 'unknown')
3554 elif codec
!= 'none':
3555 return '.'.join(codec
.split('.')[:4])
3557 if field
== 'vcodec' and f
.get('acodec') == 'none':
3559 elif field
== 'acodec' and f
.get('vcodec') == 'none':
3561 return self
._format
_out
('audio only' if field
== 'vcodec' else 'video only',
3562 self
.Styles
.SUPPRESS
)
3564 delim
= self
._format
_out
('\u2502', self
.Styles
.DELIM
, '|', test_encoding
=True)
3567 self
._format
_out
(format_field(f
, 'format_id'), self
.Styles
.ID
),
3568 format_field(f
, 'ext'),
3569 format_field(f
, func
=self
.format_resolution
, ignore
=('audio only', 'images')),
3570 format_field(f
, 'fps', '\t%d', func
=round),
3571 format_field(f
, 'dynamic_range', '%s', ignore
=(None, 'SDR')).replace('HDR', ''),
3572 format_field(f
, 'audio_channels', '\t%s'),
3574 format_field(f
, 'filesize', ' \t%s', func
=format_bytes
) + format_field(f
, 'filesize_approx', '~\t%s', func
=format_bytes
),
3575 format_field(f
, 'tbr', '\t%dk', func
=round),
3576 shorten_protocol_name(f
.get('protocol', '')),
3578 simplified_codec(f
, 'vcodec'),
3579 format_field(f
, 'vbr', '\t%dk', func
=round),
3580 simplified_codec(f
, 'acodec'),
3581 format_field(f
, 'abr', '\t%dk', func
=round),
3582 format_field(f
, 'asr', '\t%s', func
=format_decimal_suffix
),
3584 self
._format
_out
('UNSUPPORTED', 'light red') if f
.get('ext') in ('f4f', 'f4m') else None,
3585 format_field(f
, 'language', '[%s]'),
3586 join_nonempty(format_field(f
, 'format_note'),
3587 format_field(f
, 'container', ignore
=(None, f
.get('ext'))),
3590 ] for f
in formats
if f
.get('preference') is None or f
['preference'] >= -1000]
3591 header_line
= self
._list
_format
_headers
(
3592 'ID', 'EXT', 'RESOLUTION', '\tFPS', 'HDR', 'CH', delim
, '\tFILESIZE', '\tTBR', 'PROTO',
3593 delim
, 'VCODEC', '\tVBR', 'ACODEC', '\tABR', '\tASR', 'MORE INFO')
3595 return render_table(
3596 header_line
, table
, hide_empty
=True,
3597 delim
=self
._format
_out
('\u2500', self
.Styles
.DELIM
, '-', test_encoding
=True))
3599 def render_thumbnails_table(self
, info_dict
):
3600 thumbnails
= list(info_dict
.get('thumbnails') or [])
3603 return render_table(
3604 self
._list
_format
_headers
('ID', 'Width', 'Height', 'URL'),
3605 [[t
.get('id'), t
.get('width', 'unknown'), t
.get('height', 'unknown'), t
['url']] for t
in thumbnails
])
3607 def render_subtitles_table(self
, video_id
, subtitles
):
3608 def _row(lang
, formats
):
3609 exts
, names
= zip(*((f
['ext'], f
.get('name') or 'unknown') for f
in reversed(formats
)))
3610 if len(set(names
)) == 1:
3611 names
= [] if names
[0] == 'unknown' else names
[:1]
3612 return [lang
, ', '.join(names
), ', '.join(exts
)]
3616 return render_table(
3617 self
._list
_format
_headers
('Language', 'Name', 'Formats'),
3618 [_row(lang
, formats
) for lang
, formats
in subtitles
.items()],
3621 def __list_table(self
, video_id
, name
, func
, *args
):
3624 self
.to_screen(f
'{video_id} has no {name}')
3626 self
.to_screen(f
'[info] Available {name} for {video_id}:')
3627 self
.to_stdout(table
)
3629 def list_formats(self
, info_dict
):
3630 self
.__list
_table
(info_dict
['id'], 'formats', self
.render_formats_table
, info_dict
)
3632 def list_thumbnails(self
, info_dict
):
3633 self
.__list
_table
(info_dict
['id'], 'thumbnails', self
.render_thumbnails_table
, info_dict
)
3635 def list_subtitles(self
, video_id
, subtitles
, name
='subtitles'):
3636 self
.__list
_table
(video_id
, name
, self
.render_subtitles_table
, video_id
, subtitles
)
3638 def urlopen(self
, req
):
3639 """ Start an HTTP download """
3640 if isinstance(req
, str):
3641 req
= sanitized_Request(req
)
3642 return self
._opener
.open(req
, timeout
=self
._socket
_timeout
)
3644 def print_debug_header(self
):
3645 if not self
.params
.get('verbose'):
3648 # These imports can be slow. So import them only as needed
3649 from .extractor
.extractors
import _LAZY_LOADER
3650 from .extractor
.extractors
import _PLUGIN_CLASSES
as plugin_extractors
3652 def get_encoding(stream
):
3653 ret
= str(getattr(stream
, 'encoding', 'missing (%s)' % type(stream
).__name
__))
3654 if not supports_terminal_sequences(stream
):
3655 from .utils
import WINDOWS_VT_MODE
# Must be imported locally
3656 ret
+= ' (No VT)' if WINDOWS_VT_MODE
is False else ' (No ANSI)'
3659 encoding_str
= 'Encodings: locale %s, fs %s, pref %s, %s' % (
3660 locale
.getpreferredencoding(),
3661 sys
.getfilesystemencoding(),
3662 self
.get_encoding(),
3664 f
'{key} {get_encoding(stream)}' for key
, stream
in self
._out
_files
.items_
3665 if stream
is not None and key
!= 'console')
3668 logger
= self
.params
.get('logger')
3670 write_debug
= lambda msg
: logger
.debug(f
'[debug] {msg}')
3671 write_debug(encoding_str
)
3673 write_string(f
'[debug] {encoding_str}\n', encoding
=None)
3674 write_debug
= lambda msg
: self
._write
_string
(f
'[debug] {msg}\n')
3676 source
= detect_variant()
3677 if VARIANT
not in (None, 'pip'):
3679 write_debug(join_nonempty(
3680 'yt-dlp version', __version__
,
3681 f
'[{RELEASE_GIT_HEAD}]' if RELEASE_GIT_HEAD
else '',
3682 '' if source
== 'unknown' else f
'({source})',
3684 if not _LAZY_LOADER
:
3685 if os
.environ
.get('YTDLP_NO_LAZY_EXTRACTORS'):
3686 write_debug('Lazy loading extractors is forcibly disabled')
3688 write_debug('Lazy loading extractors is disabled')
3689 if plugin_extractors
or plugin_postprocessors
:
3690 write_debug('Plugins: %s' % [
3691 '%s%s' % (klass
.__name
__, '' if klass
.__name
__ == name
else f
' as {name}')
3692 for name
, klass
in itertools
.chain(plugin_extractors
.items(), plugin_postprocessors
.items())])
3693 if self
.params
['compat_opts']:
3694 write_debug('Compatibility options: %s' % ', '.join(self
.params
['compat_opts']))
3696 if source
== 'source':
3698 stdout
, _
, _
= Popen
.run(
3699 ['git', 'rev-parse', '--short', 'HEAD'],
3700 text
=True, cwd
=os
.path
.dirname(os
.path
.abspath(__file__
)),
3701 stdout
=subprocess
.PIPE
, stderr
=subprocess
.PIPE
)
3702 if re
.fullmatch('[0-9a-f]+', stdout
.strip()):
3703 write_debug(f
'Git HEAD: {stdout.strip()}')
3705 with contextlib
.suppress(Exception):
3708 write_debug(system_identifier())
3710 exe_versions
, ffmpeg_features
= FFmpegPostProcessor
.get_versions_and_features(self
)
3711 ffmpeg_features
= {key for key, val in ffmpeg_features.items() if val}
3713 exe_versions
['ffmpeg'] += ' (%s)' % ','.join(sorted(ffmpeg_features
))
3715 exe_versions
['rtmpdump'] = rtmpdump_version()
3716 exe_versions
['phantomjs'] = PhantomJSwrapper
._version
()
3717 exe_str
= ', '.join(
3718 f
'{exe} {v}' for exe
, v
in sorted(exe_versions
.items()) if v
3720 write_debug('exe versions: %s' % exe_str
)
3722 from .compat
.compat_utils
import get_package_info
3723 from .dependencies
import available_dependencies
3725 write_debug('Optional libraries: %s' % (', '.join(sorted({
3726 join_nonempty(*get_package_info(m
)) for m
in available_dependencies
.values()
3729 self
._setup
_opener
()
3731 for handler
in self
._opener
.handlers
:
3732 if hasattr(handler
, 'proxies'):
3733 proxy_map
.update(handler
.proxies
)
3734 write_debug(f
'Proxy map: {proxy_map}')
3737 if False and self
.params
.get('call_home'):
3738 ipaddr
= self
.urlopen('https://yt-dl.org/ip').read().decode()
3739 write_debug('Public IP address: %s' % ipaddr
)
3740 latest_version
= self
.urlopen(
3741 'https://yt-dl.org/latest/version').read().decode()
3742 if version_tuple(latest_version
) > version_tuple(__version__
):
3743 self
.report_warning(
3744 'You are using an outdated version (newest version: %s)! '
3745 'See https://yt-dl.org/update if you need help updating.' %
3748 def _setup_opener(self
):
3749 if hasattr(self
, '_opener'):
3751 timeout_val
= self
.params
.get('socket_timeout')
3752 self
._socket
_timeout
= 20 if timeout_val
is None else float(timeout_val
)
3754 opts_cookiesfrombrowser
= self
.params
.get('cookiesfrombrowser')
3755 opts_cookiefile
= self
.params
.get('cookiefile')
3756 opts_proxy
= self
.params
.get('proxy')
3758 self
.cookiejar
= load_cookies(opts_cookiefile
, opts_cookiesfrombrowser
, self
)
3760 cookie_processor
= YoutubeDLCookieProcessor(self
.cookiejar
)
3761 if opts_proxy
is not None:
3762 if opts_proxy
== '':
3765 proxies
= {'http': opts_proxy, 'https': opts_proxy}
3767 proxies
= urllib
.request
.getproxies()
3768 # Set HTTPS proxy to HTTP one if given (https://github.com/ytdl-org/youtube-dl/issues/805)
3769 if 'http' in proxies
and 'https' not in proxies
:
3770 proxies
['https'] = proxies
['http']
3771 proxy_handler
= PerRequestProxyHandler(proxies
)
3773 debuglevel
= 1 if self
.params
.get('debug_printtraffic') else 0
3774 https_handler
= make_HTTPS_handler(self
.params
, debuglevel
=debuglevel
)
3775 ydlh
= YoutubeDLHandler(self
.params
, debuglevel
=debuglevel
)
3776 redirect_handler
= YoutubeDLRedirectHandler()
3777 data_handler
= urllib
.request
.DataHandler()
3779 # When passing our own FileHandler instance, build_opener won't add the
3780 # default FileHandler and allows us to disable the file protocol, which
3781 # can be used for malicious purposes (see
3782 # https://github.com/ytdl-org/youtube-dl/issues/8227)
3783 file_handler
= urllib
.request
.FileHandler()
3785 def file_open(*args
, **kwargs
):
3786 raise urllib
.error
.URLError('file:// scheme is explicitly disabled in yt-dlp for security reasons')
3787 file_handler
.file_open
= file_open
3789 opener
= urllib
.request
.build_opener(
3790 proxy_handler
, https_handler
, cookie_processor
, ydlh
, redirect_handler
, data_handler
, file_handler
)
3792 # Delete the default user-agent header, which would otherwise apply in
3793 # cases where our custom HTTP handler doesn't come into play
3794 # (See https://github.com/ytdl-org/youtube-dl/issues/1309 for details)
3795 opener
.addheaders
= []
3796 self
._opener
= opener
3798 def encode(self
, s
):
3799 if isinstance(s
, bytes):
3800 return s
# Already encoded
3803 return s
.encode(self
.get_encoding())
3804 except UnicodeEncodeError as err
:
3805 err
.reason
= err
.reason
+ '. Check your system encoding configuration or use the --encoding option.'
3808 def get_encoding(self
):
3809 encoding
= self
.params
.get('encoding')
3810 if encoding
is None:
3811 encoding
= preferredencoding()
3814 def _write_info_json(self
, label
, ie_result
, infofn
, overwrite
=None):
3815 ''' Write infojson and returns True = written, 'exists' = Already exists, False = skip, None = error '''
3816 if overwrite
is None:
3817 overwrite
= self
.params
.get('overwrites', True)
3818 if not self
.params
.get('writeinfojson'):
3821 self
.write_debug(f
'Skipping writing {label} infojson')
3823 elif not self
._ensure
_dir
_exists
(infofn
):
3825 elif not overwrite
and os
.path
.exists(infofn
):
3826 self
.to_screen(f
'[info] {label.title()} metadata is already present')
3829 self
.to_screen(f
'[info] Writing {label} metadata as JSON to: {infofn}')
3831 write_json_file(self
.sanitize_info(ie_result
, self
.params
.get('clean_infojson', True)), infofn
)
3834 self
.report_error(f
'Cannot write {label} metadata to JSON file {infofn}')
3837 def _write_description(self
, label
, ie_result
, descfn
):
3838 ''' Write description and returns True = written, False = skip, None = error '''
3839 if not self
.params
.get('writedescription'):
3842 self
.write_debug(f
'Skipping writing {label} description')
3844 elif not self
._ensure
_dir
_exists
(descfn
):
3846 elif not self
.params
.get('overwrites', True) and os
.path
.exists(descfn
):
3847 self
.to_screen(f
'[info] {label.title()} description is already present')
3848 elif ie_result
.get('description') is None:
3849 self
.report_warning(f
'There\'s no {label} description to write')
3853 self
.to_screen(f
'[info] Writing {label} description to: {descfn}')
3854 with open(encodeFilename(descfn
), 'w', encoding
='utf-8') as descfile
:
3855 descfile
.write(ie_result
['description'])
3857 self
.report_error(f
'Cannot write {label} description file {descfn}')
3861 def _write_subtitles(self
, info_dict
, filename
):
3862 ''' Write subtitles to file and return list of (sub_filename, final_sub_filename); or None if error'''
3864 subtitles
= info_dict
.get('requested_subtitles')
3865 if not subtitles
or not (self
.params
.get('writesubtitles') or self
.params
.get('writeautomaticsub')):
3866 # subtitles download errors are already managed as troubles in relevant IE
3867 # that way it will silently go on when used with unsupporting IE
3870 sub_filename_base
= self
.prepare_filename(info_dict
, 'subtitle')
3871 if not sub_filename_base
:
3872 self
.to_screen('[info] Skipping writing video subtitles')
3874 for sub_lang
, sub_info
in subtitles
.items():
3875 sub_format
= sub_info
['ext']
3876 sub_filename
= subtitles_filename(filename
, sub_lang
, sub_format
, info_dict
.get('ext'))
3877 sub_filename_final
= subtitles_filename(sub_filename_base
, sub_lang
, sub_format
, info_dict
.get('ext'))
3878 existing_sub
= self
.existing_file((sub_filename_final
, sub_filename
))
3880 self
.to_screen(f
'[info] Video subtitle {sub_lang}.{sub_format} is already present')
3881 sub_info
['filepath'] = existing_sub
3882 ret
.append((existing_sub
, sub_filename_final
))
3885 self
.to_screen(f
'[info] Writing video subtitles to: {sub_filename}')
3886 if sub_info
.get('data') is not None:
3888 # Use newline='' to prevent conversion of newline characters
3889 # See https://github.com/ytdl-org/youtube-dl/issues/10268
3890 with open(sub_filename
, 'w', encoding
='utf-8', newline
='') as subfile
:
3891 subfile
.write(sub_info
['data'])
3892 sub_info
['filepath'] = sub_filename
3893 ret
.append((sub_filename
, sub_filename_final
))
3896 self
.report_error(f
'Cannot write video subtitles file {sub_filename}')
3900 sub_copy
= sub_info
.copy()
3901 sub_copy
.setdefault('http_headers', info_dict
.get('http_headers'))
3902 self
.dl(sub_filename
, sub_copy
, subtitle
=True)
3903 sub_info
['filepath'] = sub_filename
3904 ret
.append((sub_filename
, sub_filename_final
))
3905 except (DownloadError
, ExtractorError
, IOError, OSError, ValueError) + network_exceptions
as err
:
3906 msg
= f
'Unable to download video subtitles for {sub_lang!r}: {err}'
3907 if self
.params
.get('ignoreerrors') is not True: # False or 'only_download'
3908 if not self
.params
.get('ignoreerrors'):
3909 self
.report_error(msg
)
3910 raise DownloadError(msg
)
3911 self
.report_warning(msg
)
3914 def _write_thumbnails(self
, label
, info_dict
, filename
, thumb_filename_base
=None):
3915 ''' Write thumbnails to file and return list of (thumb_filename, final_thumb_filename) '''
3916 write_all
= self
.params
.get('write_all_thumbnails', False)
3917 thumbnails
, ret
= [], []
3918 if write_all
or self
.params
.get('writethumbnail', False):
3919 thumbnails
= info_dict
.get('thumbnails') or []
3920 multiple
= write_all
and len(thumbnails
) > 1
3922 if thumb_filename_base
is None:
3923 thumb_filename_base
= filename
3924 if thumbnails
and not thumb_filename_base
:
3925 self
.write_debug(f
'Skipping writing {label} thumbnail')
3928 for idx
, t
in list(enumerate(thumbnails
))[::-1]:
3929 thumb_ext
= (f
'{t["id"]}.' if multiple
else '') + determine_ext(t
['url'], 'jpg')
3930 thumb_display_id
= f
'{label} thumbnail {t["id"]}'
3931 thumb_filename
= replace_extension(filename
, thumb_ext
, info_dict
.get('ext'))
3932 thumb_filename_final
= replace_extension(thumb_filename_base
, thumb_ext
, info_dict
.get('ext'))
3934 existing_thumb
= self
.existing_file((thumb_filename_final
, thumb_filename
))
3936 self
.to_screen('[info] %s is already present' % (
3937 thumb_display_id
if multiple
else f
'{label} thumbnail').capitalize())
3938 t
['filepath'] = existing_thumb
3939 ret
.append((existing_thumb
, thumb_filename_final
))
3941 self
.to_screen(f
'[info] Downloading {thumb_display_id} ...')
3943 uf
= self
.urlopen(sanitized_Request(t
['url'], headers
=t
.get('http_headers', {})))
3944 self
.to_screen(f
'[info] Writing {thumb_display_id} to: {thumb_filename}')
3945 with open(encodeFilename(thumb_filename
), 'wb') as thumbf
:
3946 shutil
.copyfileobj(uf
, thumbf
)
3947 ret
.append((thumb_filename
, thumb_filename_final
))
3948 t
['filepath'] = thumb_filename
3949 except network_exceptions
as err
:
3951 self
.report_warning(f
'Unable to download {thumb_display_id}: {err}')
3952 if ret
and not write_all
: