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
.openload
import PhantomJSwrapper
33 from .minicurses
import format_text
34 from .postprocessor
import _PLUGIN_CLASSES
as plugin_postprocessors
35 from .postprocessor
import (
37 FFmpegFixupDuplicateMoovPP
,
38 FFmpegFixupDurationPP
,
41 FFmpegFixupStretchedPP
,
42 FFmpegFixupTimestampPP
,
45 FFmpegVideoConvertorPP
,
46 MoveFilesAfterDownloadPP
,
49 from .postprocessor
.ffmpeg
import resolve_mapping
as resolve_recode_mapping
50 from .update
import detect_variant
76 PerRequestProxyHandler
,
83 UnavailableVideoError
,
85 YoutubeDLCookieProcessor
,
87 YoutubeDLRedirectHandler
,
102 format_decimal_suffix
,
121 register_socks_protocols
,
122 remove_terminal_sequences
,
133 supports_terminal_sequences
,
143 windows_enable_vt_mode
,
147 from .version
import RELEASE_GIT_HEAD
, __version__
149 if compat_os_name
== 'nt':
156 YoutubeDL objects are the ones responsible of downloading the
157 actual video file and writing it to disk if the user has requested
158 it, among some other tasks. In most cases there should be one per
159 program. As, given a video URL, the downloader doesn't know how to
160 extract all the needed information, task that InfoExtractors do, it
161 has to pass the URL to one of them.
163 For this, YoutubeDL objects have a method that allows
164 InfoExtractors to be registered in a given order. When it is passed
165 a URL, the YoutubeDL object handles it to the first InfoExtractor it
166 finds that reports being able to handle it. The InfoExtractor extracts
167 all the information about the video or videos the URL refers to, and
168 YoutubeDL process the extracted information, possibly using a File
169 Downloader to download the video.
171 YoutubeDL objects accept a lot of parameters. In order not to saturate
172 the object constructor with arguments, it receives a dictionary of
173 options instead. These options are available through the params
174 attribute for the InfoExtractors to use. The YoutubeDL also
175 registers itself as the downloader in charge for the InfoExtractors
176 that are added to it, so this is a "mutual registration".
180 username: Username for authentication purposes.
181 password: Password for authentication purposes.
182 videopassword: Password for accessing a video.
183 ap_mso: Adobe Pass multiple-system operator identifier.
184 ap_username: Multiple-system operator account username.
185 ap_password: Multiple-system operator account password.
186 usenetrc: Use netrc for authentication instead.
187 verbose: Print additional info to stdout.
188 quiet: Do not print messages to stdout.
189 no_warnings: Do not print out anything for warnings.
190 forceprint: A dict with keys WHEN mapped to a list of templates to
191 print to stdout. The allowed keys are video or any of the
192 items in utils.POSTPROCESS_WHEN.
193 For compatibility, a single list is also accepted
194 print_to_file: A dict with keys WHEN (same as forceprint) mapped to
195 a list of tuples with (template, filename)
196 forcejson: Force printing info_dict as JSON.
197 dump_single_json: Force printing the info_dict of the whole playlist
198 (or video) as a single JSON line.
199 force_write_download_archive: Force writing download archive regardless
200 of 'skip_download' or 'simulate'.
201 simulate: Do not download the video files. If unset (or None),
202 simulate only if listsubtitles, listformats or list_thumbnails is used
203 format: Video format code. see "FORMAT SELECTION" for more details.
204 You can also pass a function. The function takes 'ctx' as
205 argument and returns the formats to download.
206 See "build_format_selector" for an implementation
207 allow_unplayable_formats: Allow unplayable formats to be extracted and downloaded.
208 ignore_no_formats_error: Ignore "No video formats" error. Usefull for
209 extracting metadata even if the video is not actually
210 available for download (experimental)
211 format_sort: A list of fields by which to sort the video formats.
212 See "Sorting Formats" for more details.
213 format_sort_force: Force the given format_sort. see "Sorting Formats"
215 prefer_free_formats: Whether to prefer video formats with free containers
216 over non-free ones of same quality.
217 allow_multiple_video_streams: Allow multiple video streams to be merged
219 allow_multiple_audio_streams: Allow multiple audio streams to be merged
221 check_formats Whether to test if the formats are downloadable.
222 Can be True (check all), False (check none),
223 'selected' (check selected formats),
224 or None (check only if requested by extractor)
225 paths: Dictionary of output paths. The allowed keys are 'home'
226 'temp' and the keys of OUTTMPL_TYPES (in utils.py)
227 outtmpl: Dictionary of templates for output names. Allowed keys
228 are 'default' and the keys of OUTTMPL_TYPES (in utils.py).
229 For compatibility with youtube-dl, a single string can also be used
230 outtmpl_na_placeholder: Placeholder for unavailable meta fields.
231 restrictfilenames: Do not allow "&" and spaces in file names
232 trim_file_name: Limit length of filename (extension excluded)
233 windowsfilenames: Force the filenames to be windows compatible
234 ignoreerrors: Do not stop on download/postprocessing errors.
235 Can be 'only_download' to ignore only download errors.
236 Default is 'only_download' for CLI, but False for API
237 skip_playlist_after_errors: Number of allowed failures until the rest of
238 the playlist is skipped
239 force_generic_extractor: Force downloader to use the generic extractor
240 overwrites: Overwrite all video and metadata files if True,
241 overwrite only non-video files if None
242 and don't overwrite any file if False
243 For compatibility with youtube-dl,
244 "nooverwrites" may also be used instead
245 playlist_items: Specific indices of playlist to download.
246 playlistrandom: Download playlist items in random order.
247 lazy_playlist: Process playlist entries as they are received.
248 matchtitle: Download only matching titles.
249 rejecttitle: Reject downloads for matching titles.
250 logger: Log messages to a logging.Logger instance.
251 logtostderr: Log messages to stderr instead of stdout.
252 consoletitle: Display progress in console window's titlebar.
253 writedescription: Write the video description to a .description file
254 writeinfojson: Write the video description to a .info.json file
255 clean_infojson: Remove private fields from the infojson
256 getcomments: Extract video comments. This will not be written to disk
257 unless writeinfojson is also given
258 writeannotations: Write the video annotations to a .annotations.xml file
259 writethumbnail: Write the thumbnail image to a file
260 allow_playlist_files: Whether to write playlists' description, infojson etc
261 also to disk when using the 'write*' options
262 write_all_thumbnails: Write all thumbnail formats to files
263 writelink: Write an internet shortcut file, depending on the
264 current platform (.url/.webloc/.desktop)
265 writeurllink: Write a Windows internet shortcut file (.url)
266 writewebloclink: Write a macOS internet shortcut file (.webloc)
267 writedesktoplink: Write a Linux internet shortcut file (.desktop)
268 writesubtitles: Write the video subtitles to a file
269 writeautomaticsub: Write the automatically generated subtitles to a file
270 listsubtitles: Lists all available subtitles for the video
271 subtitlesformat: The format code for subtitles
272 subtitleslangs: List of languages of the subtitles to download (can be regex).
273 The list may contain "all" to refer to all the available
274 subtitles. The language can be prefixed with a "-" to
275 exclude it from the requested languages. Eg: ['all', '-live_chat']
276 keepvideo: Keep the video file after post-processing
277 daterange: A DateRange object, download only if the upload_date is in the range.
278 skip_download: Skip the actual download of the video file
279 cachedir: Location of the cache files in the filesystem.
280 False to disable filesystem cache.
281 noplaylist: Download single video instead of a playlist if in doubt.
282 age_limit: An integer representing the user's age in years.
283 Unsuitable videos for the given age are skipped.
284 min_views: An integer representing the minimum view count the video
285 must have in order to not be skipped.
286 Videos without view count information are always
287 downloaded. None for no limit.
288 max_views: An integer representing the maximum view count.
289 Videos that are more popular than that are not
291 Videos without view count information are always
292 downloaded. None for no limit.
293 download_archive: File name of a file where all downloads are recorded.
294 Videos already present in the file are not downloaded
296 break_on_existing: Stop the download process after attempting to download a
297 file that is in the archive.
298 break_on_reject: Stop the download process when encountering a video that
299 has been filtered out.
300 break_per_url: Whether break_on_reject and break_on_existing
301 should act on each input URL as opposed to for the entire queue
302 cookiefile: File name or text stream from where cookies should be read and dumped to
303 cookiesfrombrowser: A tuple containing the name of the browser, the profile
304 name/pathfrom where cookies are loaded, and the name of the
305 keyring. Eg: ('chrome', ) or ('vivaldi', 'default', 'BASICTEXT')
306 legacyserverconnect: Explicitly allow HTTPS connection to servers that do not
307 support RFC 5746 secure renegotiation
308 nocheckcertificate: Do not verify SSL certificates
309 client_certificate: Path to client certificate file in PEM format. May include the private key
310 client_certificate_key: Path to private key file for client certificate
311 client_certificate_password: Password for client certificate private key, if encrypted.
312 If not provided and the key is encrypted, yt-dlp will ask interactively
313 prefer_insecure: Use HTTP instead of HTTPS to retrieve information.
314 (Only supported by some extractors)
315 http_headers: A dictionary of custom headers to be used for all requests
316 proxy: URL of the proxy server to use
317 geo_verification_proxy: URL of the proxy to use for IP address verification
318 on geo-restricted sites.
319 socket_timeout: Time to wait for unresponsive hosts, in seconds
320 bidi_workaround: Work around buggy terminals without bidirectional text
321 support, using fridibi
322 debug_printtraffic:Print out sent and received HTTP traffic
323 default_search: Prepend this string if an input url is not valid.
324 'auto' for elaborate guessing
325 encoding: Use this encoding instead of the system-specified.
326 extract_flat: Whether to resolve and process url_results further
327 * False: Always process (default)
328 * True: Never process
329 * 'in_playlist': Do not process inside playlist/multi_video
330 * 'discard': Always process, but don't return the result
331 from inside playlist/multi_video
332 * 'discard_in_playlist': Same as "discard", but only for
333 playlists (not multi_video)
334 wait_for_video: If given, wait for scheduled streams to become available.
335 The value should be a tuple containing the range
336 (min_secs, max_secs) to wait between retries
337 postprocessors: A list of dictionaries, each with an entry
338 * key: The name of the postprocessor. See
339 yt_dlp/postprocessor/__init__.py for a list.
340 * when: When to run the postprocessor. Allowed values are
341 the entries of utils.POSTPROCESS_WHEN
342 Assumed to be 'post_process' if not given
343 progress_hooks: A list of functions that get called on download
344 progress, with a dictionary with the entries
345 * status: One of "downloading", "error", or "finished".
346 Check this first and ignore unknown values.
347 * info_dict: The extracted info_dict
349 If status is one of "downloading", or "finished", the
350 following properties may also be present:
351 * filename: The final filename (always present)
352 * tmpfilename: The filename we're currently writing to
353 * downloaded_bytes: Bytes on disk
354 * total_bytes: Size of the whole file, None if unknown
355 * total_bytes_estimate: Guess of the eventual file size,
357 * elapsed: The number of seconds since download started.
358 * eta: The estimated time in seconds, None if unknown
359 * speed: The download speed in bytes/second, None if
361 * fragment_index: The counter of the currently
362 downloaded video fragment.
363 * fragment_count: The number of fragments (= individual
364 files that will be merged)
366 Progress hooks are guaranteed to be called at least once
367 (with status "finished") if the download is successful.
368 postprocessor_hooks: A list of functions that get called on postprocessing
369 progress, with a dictionary with the entries
370 * status: One of "started", "processing", or "finished".
371 Check this first and ignore unknown values.
372 * postprocessor: Name of the postprocessor
373 * info_dict: The extracted info_dict
375 Progress hooks are guaranteed to be called at least twice
376 (with status "started" and "finished") if the processing is successful.
377 merge_output_format: "/" separated list of extensions to use when merging formats.
378 final_ext: Expected final extension; used to detect when the file was
379 already downloaded and converted
380 fixup: Automatically correct known faults of the file.
382 - "never": do nothing
383 - "warn": only emit a warning
384 - "detect_or_warn": check whether we can do anything
385 about it, warn otherwise (default)
386 source_address: Client-side IP address to bind to.
387 sleep_interval_requests: Number of seconds to sleep between requests
389 sleep_interval: Number of seconds to sleep before each download when
390 used alone or a lower bound of a range for randomized
391 sleep before each download (minimum possible number
392 of seconds to sleep) when used along with
394 max_sleep_interval:Upper bound of a range for randomized sleep before each
395 download (maximum possible number of seconds to sleep).
396 Must only be used along with sleep_interval.
397 Actual sleep time will be a random float from range
398 [sleep_interval; max_sleep_interval].
399 sleep_interval_subtitles: Number of seconds to sleep before each subtitle download
400 listformats: Print an overview of available video formats and exit.
401 list_thumbnails: Print a table of all thumbnails and exit.
402 match_filter: A function that gets called for every video with the signature
403 (info_dict, *, incomplete: bool) -> Optional[str]
404 For backward compatibility with youtube-dl, the signature
405 (info_dict) -> Optional[str] is also allowed.
406 - If it returns a message, the video is ignored.
407 - If it returns None, the video is downloaded.
408 - If it returns utils.NO_DEFAULT, the user is interactively
409 asked whether to download the video.
410 match_filter_func in utils.py is one example for this.
411 no_color: Do not emit color codes in output.
412 geo_bypass: Bypass geographic restriction via faking X-Forwarded-For
415 Two-letter ISO 3166-2 country code that will be used for
416 explicit geographic restriction bypassing via faking
417 X-Forwarded-For HTTP header
419 IP range in CIDR notation that will be used similarly to
421 external_downloader: A dictionary of protocol keys and the executable of the
422 external downloader to use for it. The allowed protocols
423 are default|http|ftp|m3u8|dash|rtsp|rtmp|mms.
424 Set the value to 'native' to use the native downloader
425 compat_opts: Compatibility options. See "Differences in default behavior".
426 The following options do not work when used through the API:
427 filename, abort-on-error, multistreams, no-live-chat, format-sort
428 no-clean-infojson, no-playlist-metafiles, no-keep-subs, no-attach-info-json.
429 Refer __init__.py for their implementation
430 progress_template: Dictionary of templates for progress outputs.
431 Allowed keys are 'download', 'postprocess',
432 'download-title' (console title) and 'postprocess-title'.
433 The template is mapped on a dictionary with keys 'progress' and 'info'
434 retry_sleep_functions: Dictionary of functions that takes the number of attempts
435 as argument and returns the time to sleep in seconds.
436 Allowed keys are 'http', 'fragment', 'file_access'
437 download_ranges: A callback function that gets called for every video with
438 the signature (info_dict, ydl) -> Iterable[Section].
439 Only the returned sections will be downloaded.
440 Each Section is a dict with the following keys:
441 * start_time: Start time of the section in seconds
442 * end_time: End time of the section in seconds
443 * title: Section title (Optional)
444 * index: Section number (Optional)
445 force_keyframes_at_cuts: Re-encode the video when downloading ranges to get precise cuts
446 noprogress: Do not print the progress bar
448 The following parameters are not used by YoutubeDL itself, they are used by
449 the downloader (see yt_dlp/downloader/common.py):
450 nopart, updatetime, buffersize, ratelimit, throttledratelimit, min_filesize,
451 max_filesize, test, noresizebuffer, retries, file_access_retries, fragment_retries,
452 continuedl, xattr_set_filesize, hls_use_mpegts, http_chunk_size,
453 external_downloader_args, concurrent_fragment_downloads.
455 The following options are used by the post processors:
456 ffmpeg_location: Location of the ffmpeg/avconv binary; either the path
457 to the binary or its containing directory.
458 postprocessor_args: A dictionary of postprocessor/executable keys (in lower case)
459 and a list of additional command-line arguments for the
460 postprocessor/executable. The dict can also have "PP+EXE" keys
461 which are used when the given exe is used by the given PP.
462 Use 'default' as the name for arguments to passed to all PP
463 For compatibility with youtube-dl, a single list of args
466 The following options are used by the extractors:
467 extractor_retries: Number of times to retry for known errors
468 dynamic_mpd: Whether to process dynamic DASH manifests (default: True)
469 hls_split_discontinuity: Split HLS playlists to different formats at
470 discontinuities such as ad breaks (default: False)
471 extractor_args: A dictionary of arguments to be passed to the extractors.
472 See "EXTRACTOR ARGUMENTS" for details.
473 Eg: {'youtube': {'skip': ['dash', 'hls']}}
474 mark_watched: Mark videos watched (even with --simulate). Only for YouTube
476 The following options are deprecated and may be removed in the future:
478 playliststart: - Use playlist_items
479 Playlist item to start at.
480 playlistend: - Use playlist_items
481 Playlist item to end at.
482 playlistreverse: - Use playlist_items
483 Download playlist items in reverse order.
484 forceurl: - Use forceprint
485 Force printing final URL.
486 forcetitle: - Use forceprint
487 Force printing title.
488 forceid: - Use forceprint
490 forcethumbnail: - Use forceprint
491 Force printing thumbnail URL.
492 forcedescription: - Use forceprint
493 Force printing description.
494 forcefilename: - Use forceprint
495 Force printing final filename.
496 forceduration: - Use forceprint
497 Force printing duration.
498 allsubtitles: - Use subtitleslangs = ['all']
499 Downloads all the subtitles of the video
500 (requires writesubtitles or writeautomaticsub)
501 include_ads: - Doesn't work
503 call_home: - Not implemented
504 Boolean, true iff we are allowed to contact the
505 yt-dlp servers for debugging.
506 post_hooks: - Register a custom postprocessor
507 A list of functions that get called as the final step
508 for each video file, after all postprocessors have been
509 called. The filename will be passed as the only argument.
510 hls_prefer_native: - Use external_downloader = {'m3u8': 'native'} or {'m3u8': 'ffmpeg'}.
511 Use the native HLS downloader instead of ffmpeg/avconv
512 if True, otherwise use ffmpeg/avconv if False, otherwise
513 use downloader suggested by extractor if None.
514 prefer_ffmpeg: - avconv support is deprecated
515 If False, use avconv instead of ffmpeg if both are available,
516 otherwise prefer ffmpeg.
517 youtube_include_dash_manifest: - Use extractor_args
518 If True (default), DASH manifests and related
519 data will be downloaded and processed by extractor.
520 You can reduce network I/O by disabling it if you don't
521 care about DASH. (only for youtube)
522 youtube_include_hls_manifest: - Use extractor_args
523 If True (default), HLS 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 HLS. (only for youtube)
530 'width', 'height', 'tbr', 'abr', 'asr', 'vbr', 'fps', 'filesize', 'filesize_approx',
531 'timestamp', 'release_timestamp',
532 'duration', 'view_count', 'like_count', 'dislike_count', 'repost_count',
533 'average_rating', 'comment_count', 'age_limit',
534 'start_time', 'end_time',
535 'chapter_number', 'season_number', 'episode_number',
536 'track_number', 'disc_number', 'release_year',
540 # NB: Keep in sync with the docstring of extractor/common.py
541 'url', 'manifest_url', 'manifest_stream_number', 'ext', 'format', 'format_id', 'format_note',
542 'width', 'height', 'resolution', 'dynamic_range', 'tbr', 'abr', 'acodec', 'asr',
543 'vbr', 'fps', 'vcodec', 'container', 'filesize', 'filesize_approx',
544 'player_url', 'protocol', 'fragment_base_url', 'fragments', 'is_from_start',
545 'preference', 'language', 'language_preference', 'quality', 'source_preference',
546 'http_headers', 'stretched_ratio', 'no_resume', 'has_drm', 'downloader_options',
547 'page_url', 'app', 'play_path', 'tc_url', 'flash_version', 'rtmp_live', 'rtmp_conn', 'rtmp_protocol', 'rtmp_real_time'
549 _format_selection_exts
= {
550 'audio': set(MEDIA_EXTENSIONS
.common_audio
),
551 'video': set(MEDIA_EXTENSIONS
.common_video
+ ('3gp', )),
552 'storyboards': set(MEDIA_EXTENSIONS
.storyboards
),
555 def __init__(self
, params
=None, auto_init
=True):
556 """Create a FileDownloader object with the given options.
557 @param auto_init Whether to load the default extractors and print header (if verbose).
558 Set to 'no_verbose_header' to not print the header
564 self
._ies
_instances
= {}
565 self
._pps
= {k: [] for k in POSTPROCESS_WHEN}
566 self
._printed
_messages
= set()
567 self
._first
_webpage
_request
= True
568 self
._post
_hooks
= []
569 self
._progress
_hooks
= []
570 self
._postprocessor
_hooks
= []
571 self
._download
_retcode
= 0
572 self
._num
_downloads
= 0
574 self
._playlist
_level
= 0
575 self
._playlist
_urls
= set()
576 self
.cache
= Cache(self
)
578 windows_enable_vt_mode()
579 stdout
= sys
.stderr
if self
.params
.get('logtostderr') else sys
.stdout
580 self
._out
_files
= Namespace(
583 screen
=sys
.stderr
if self
.params
.get('quiet') else stdout
,
584 console
=None if compat_os_name
== 'nt' else next(
585 filter(supports_terminal_sequences
, (sys
.stderr
, sys
.stdout
)), None)
587 self
._allow
_colors
= Namespace(**{
588 type_
: not self
.params
.get('no_color') and supports_terminal_sequences(stream
)
589 for type_
, stream
in self
._out
_files
.items_
if type_
!= 'console'
592 # The code is left like this to be reused for future deprecations
593 MIN_SUPPORTED
, MIN_RECOMMENDED
= (3, 7), (3, 7)
594 current_version
= sys
.version_info
[:2]
595 if current_version
< MIN_RECOMMENDED
:
596 msg
= ('Support for Python version %d.%d has been deprecated. '
597 'See https://github.com/yt-dlp/yt-dlp/issues/3764 for more details.'
598 '\n You will no longer receive updates on this version')
599 if current_version
< MIN_SUPPORTED
:
600 msg
= 'Python version %d.%d is no longer supported'
601 self
.deprecation_warning(
602 f
'{msg}! Please update to Python %d.%d or above' % (*current_version
, *MIN_RECOMMENDED
))
604 if self
.params
.get('allow_unplayable_formats'):
606 f
'You have asked for {self._format_err("UNPLAYABLE", self.Styles.EMPHASIS)} formats to be listed/downloaded. '
607 'This is a developer option intended for debugging. \n'
608 ' If you experience any issues while using this option, '
609 f
'{self._format_err("DO NOT", self.Styles.ERROR)} open a bug report')
611 def check_deprecated(param
, option
, suggestion
):
612 if self
.params
.get(param
) is not None:
613 self
.report_warning(f
'{option} is deprecated. Use {suggestion} instead')
617 if check_deprecated('cn_verification_proxy', '--cn-verification-proxy', '--geo-verification-proxy'):
618 if self
.params
.get('geo_verification_proxy') is None:
619 self
.params
['geo_verification_proxy'] = self
.params
['cn_verification_proxy']
621 check_deprecated('autonumber', '--auto-number', '-o "%(autonumber)s-%(title)s.%(ext)s"')
622 check_deprecated('usetitle', '--title', '-o "%(title)s-%(id)s.%(ext)s"')
623 check_deprecated('useid', '--id', '-o "%(id)s.%(ext)s"')
625 for msg
in self
.params
.get('_warnings', []):
626 self
.report_warning(msg
)
627 for msg
in self
.params
.get('_deprecation_warnings', []):
628 self
.deprecation_warning(msg
)
630 self
.params
['compat_opts'] = set(self
.params
.get('compat_opts', ()))
631 if 'list-formats' in self
.params
['compat_opts']:
632 self
.params
['listformats_table'] = False
634 if 'overwrites' not in self
.params
and self
.params
.get('nooverwrites') is not None:
635 # nooverwrites was unnecessarily changed to overwrites
636 # in 0c3d0f51778b153f65c21906031c2e091fcfb641
637 # This ensures compatibility with both keys
638 self
.params
['overwrites'] = not self
.params
['nooverwrites']
639 elif self
.params
.get('overwrites') is None:
640 self
.params
.pop('overwrites', None)
642 self
.params
['nooverwrites'] = not self
.params
['overwrites']
644 self
.params
.setdefault('forceprint', {})
645 self
.params
.setdefault('print_to_file', {})
647 # Compatibility with older syntax
648 if not isinstance(params
['forceprint'], dict):
649 self
.params
['forceprint'] = {'video': params['forceprint']}
651 if self
.params
.get('bidi_workaround', False):
654 master
, slave
= pty
.openpty()
655 width
= shutil
.get_terminal_size().columns
656 width_args
= [] if width
is None else ['-w', str(width
)]
657 sp_kwargs
= {'stdin': subprocess.PIPE, 'stdout': slave, 'stderr': self._out_files.error}
659 self
._output
_process
= Popen(['bidiv'] + width_args
, **sp_kwargs
)
661 self
._output
_process
= Popen(['fribidi', '-c', 'UTF-8'] + width_args
, **sp_kwargs
)
662 self
._output
_channel
= os
.fdopen(master
, 'rb')
663 except OSError as ose
:
664 if ose
.errno
== errno
.ENOENT
:
666 'Could not find fribidi executable, ignoring --bidi-workaround. '
667 'Make sure that fribidi is an executable file in one of the directories in your $PATH.')
672 if auto_init
!= 'no_verbose_header':
673 self
.print_debug_header()
674 self
.add_default_info_extractors()
676 if (sys
.platform
!= 'win32'
677 and sys
.getfilesystemencoding() in ['ascii', 'ANSI_X3.4-1968']
678 and not self
.params
.get('restrictfilenames', False)):
679 # Unicode filesystem API will throw errors (#1474, #13027)
681 'Assuming --restrict-filenames since file system encoding '
682 'cannot encode all characters. '
683 'Set the LC_ALL environment variable to fix this.')
684 self
.params
['restrictfilenames'] = True
686 self
._parse
_outtmpl
()
688 # Creating format selector here allows us to catch syntax errors before the extraction
689 self
.format_selector
= (
690 self
.params
.get('format') if self
.params
.get('format') in (None, '-')
691 else self
.params
['format'] if callable(self
.params
['format'])
692 else self
.build_format_selector(self
.params
['format']))
694 # Set http_headers defaults according to std_headers
695 self
.params
['http_headers'] = merge_headers(std_headers
, self
.params
.get('http_headers', {}))
698 'post_hooks': self
.add_post_hook
,
699 'progress_hooks': self
.add_progress_hook
,
700 'postprocessor_hooks': self
.add_postprocessor_hook
,
702 for opt
, fn
in hooks
.items():
703 for ph
in self
.params
.get(opt
, []):
706 for pp_def_raw
in self
.params
.get('postprocessors', []):
707 pp_def
= dict(pp_def_raw
)
708 when
= pp_def
.pop('when', 'post_process')
709 self
.add_post_processor(
710 get_postprocessor(pp_def
.pop('key'))(self
, **pp_def
),
714 register_socks_protocols()
716 def preload_download_archive(fn
):
717 """Preload the archive, if any is specified"""
720 self
.write_debug(f
'Loading archive file {fn!r}')
722 with locked_file(fn
, 'r', encoding
='utf-8') as archive_file
:
723 for line
in archive_file
:
724 self
.archive
.add(line
.strip())
725 except OSError as ioe
:
726 if ioe
.errno
!= errno
.ENOENT
:
732 preload_download_archive(self
.params
.get('download_archive'))
734 def warn_if_short_id(self
, argv
):
735 # short YouTube ID starting with dash?
737 i
for i
, a
in enumerate(argv
)
738 if re
.match(r
'^-[0-9A-Za-z_-]{10}$', a
)]
742 + [a
for i
, a
in enumerate(argv
) if i
not in idxs
]
743 + ['--'] + [argv
[i
] for i
in idxs
]
746 'Long argument string detected. '
747 'Use -- to separate parameters and URLs, like this:\n%s' %
748 args_to_str(correct_argv
))
750 def add_info_extractor(self
, ie
):
751 """Add an InfoExtractor object to the end of the list."""
753 self
._ies
[ie_key
] = ie
754 if not isinstance(ie
, type):
755 self
._ies
_instances
[ie_key
] = ie
756 ie
.set_downloader(self
)
758 def _get_info_extractor_class(self
, ie_key
):
759 ie
= self
._ies
.get(ie_key
)
761 ie
= get_info_extractor(ie_key
)
762 self
.add_info_extractor(ie
)
765 def get_info_extractor(self
, ie_key
):
767 Get an instance of an IE with name ie_key, it will try to get one from
768 the _ies list, if there's no instance it will create a new one and add
769 it to the extractor list.
771 ie
= self
._ies
_instances
.get(ie_key
)
773 ie
= get_info_extractor(ie_key
)()
774 self
.add_info_extractor(ie
)
777 def add_default_info_extractors(self
):
779 Add the InfoExtractors returned by gen_extractors to the end of the list
781 for ie
in gen_extractor_classes():
782 self
.add_info_extractor(ie
)
784 def add_post_processor(self
, pp
, when
='post_process'):
785 """Add a PostProcessor object to the end of the chain."""
786 assert when
in POSTPROCESS_WHEN
, f
'Invalid when={when}'
787 self
._pps
[when
].append(pp
)
788 pp
.set_downloader(self
)
790 def add_post_hook(self
, ph
):
791 """Add the post hook"""
792 self
._post
_hooks
.append(ph
)
794 def add_progress_hook(self
, ph
):
795 """Add the download progress hook"""
796 self
._progress
_hooks
.append(ph
)
798 def add_postprocessor_hook(self
, ph
):
799 """Add the postprocessing progress hook"""
800 self
._postprocessor
_hooks
.append(ph
)
801 for pps
in self
._pps
.values():
803 pp
.add_progress_hook(ph
)
805 def _bidi_workaround(self
, message
):
806 if not hasattr(self
, '_output_channel'):
809 assert hasattr(self
, '_output_process')
810 assert isinstance(message
, str)
811 line_count
= message
.count('\n') + 1
812 self
._output
_process
.stdin
.write((message
+ '\n').encode())
813 self
._output
_process
.stdin
.flush()
814 res
= ''.join(self
._output
_channel
.readline().decode()
815 for _
in range(line_count
))
816 return res
[:-len('\n')]
818 def _write_string(self
, message
, out
=None, only_once
=False):
820 if message
in self
._printed
_messages
:
822 self
._printed
_messages
.add(message
)
823 write_string(message
, out
=out
, encoding
=self
.params
.get('encoding'))
825 def to_stdout(self
, message
, skip_eol
=False, quiet
=None):
826 """Print message to stdout"""
827 if quiet
is not None:
828 self
.deprecation_warning('"YoutubeDL.to_stdout" no longer accepts the argument quiet. Use "YoutubeDL.to_screen" instead')
829 if skip_eol
is not False:
830 self
.deprecation_warning('"YoutubeDL.to_stdout" no longer accepts the argument skip_eol. Use "YoutubeDL.to_screen" instead')
831 self
._write
_string
(f
'{self._bidi_workaround(message)}\n', self
._out
_files
.out
)
833 def to_screen(self
, message
, skip_eol
=False, quiet
=None):
834 """Print message to screen if not in quiet mode"""
835 if self
.params
.get('logger'):
836 self
.params
['logger'].debug(message
)
838 if (self
.params
.get('quiet') if quiet
is None else quiet
) and not self
.params
.get('verbose'):
841 '%s%s' % (self
._bidi
_workaround
(message
), ('' if skip_eol
else '\n')),
842 self
._out
_files
.screen
)
844 def to_stderr(self
, message
, only_once
=False):
845 """Print message to stderr"""
846 assert isinstance(message
, str)
847 if self
.params
.get('logger'):
848 self
.params
['logger'].error(message
)
850 self
._write
_string
(f
'{self._bidi_workaround(message)}\n', self
._out
_files
.error
, only_once
=only_once
)
852 def _send_console_code(self
, code
):
853 if compat_os_name
== 'nt' or not self
._out
_files
.console
:
855 self
._write
_string
(code
, self
._out
_files
.console
)
857 def to_console_title(self
, message
):
858 if not self
.params
.get('consoletitle', False):
860 message
= remove_terminal_sequences(message
)
861 if compat_os_name
== 'nt':
862 if ctypes
.windll
.kernel32
.GetConsoleWindow():
863 # c_wchar_p() might not be necessary if `message` is
864 # already of type unicode()
865 ctypes
.windll
.kernel32
.SetConsoleTitleW(ctypes
.c_wchar_p(message
))
867 self
._send
_console
_code
(f
'\033]0;{message}\007')
869 def save_console_title(self
):
870 if not self
.params
.get('consoletitle') or self
.params
.get('simulate'):
872 self
._send
_console
_code
('\033[22;0t') # Save the title on stack
874 def restore_console_title(self
):
875 if not self
.params
.get('consoletitle') or self
.params
.get('simulate'):
877 self
._send
_console
_code
('\033[23;0t') # Restore the title from stack
880 self
.save_console_title()
883 def __exit__(self
, *args
):
884 self
.restore_console_title()
886 if self
.params
.get('cookiefile') is not None:
887 self
.cookiejar
.save(ignore_discard
=True, ignore_expires
=True)
889 def trouble(self
, message
=None, tb
=None, is_error
=True):
890 """Determine action to take when a download problem appears.
892 Depending on if the downloader has been configured to ignore
893 download errors or not, this method may throw an exception or
894 not when errors are found, after printing the message.
896 @param tb If given, is additional traceback information
897 @param is_error Whether to raise error according to ignorerrors
899 if message
is not None:
900 self
.to_stderr(message
)
901 if self
.params
.get('verbose'):
903 if sys
.exc_info()[0]: # if .trouble has been called from an except block
905 if hasattr(sys
.exc_info()[1], 'exc_info') and sys
.exc_info()[1].exc_info
[0]:
906 tb
+= ''.join(traceback
.format_exception(*sys
.exc_info()[1].exc_info
))
907 tb
+= encode_compat_str(traceback
.format_exc())
909 tb_data
= traceback
.format_list(traceback
.extract_stack())
910 tb
= ''.join(tb_data
)
915 if not self
.params
.get('ignoreerrors'):
916 if sys
.exc_info()[0] and hasattr(sys
.exc_info()[1], 'exc_info') and sys
.exc_info()[1].exc_info
[0]:
917 exc_info
= sys
.exc_info()[1].exc_info
919 exc_info
= sys
.exc_info()
920 raise DownloadError(message
, exc_info
)
921 self
._download
_retcode
= 1
925 EMPHASIS
='light blue',
931 SUPPRESS
='light black',
934 def _format_text(self
, handle
, allow_colors
, text
, f
, fallback
=None, *, test_encoding
=False):
938 # handle.encoding can be None. See https://github.com/yt-dlp/yt-dlp/issues/2711
939 encoding
= self
.params
.get('encoding') or getattr(handle
, 'encoding', None) or 'ascii'
940 text
= text
.encode(encoding
, 'ignore').decode(encoding
)
941 if fallback
is not None and text
!= original_text
:
943 return format_text(text
, f
) if allow_colors
else text
if fallback
is None else fallback
945 def _format_out(self
, *args
, **kwargs
):
946 return self
._format
_text
(self
._out
_files
.out
, self
._allow
_colors
.out
, *args
, **kwargs
)
948 def _format_screen(self
, *args
, **kwargs
):
949 return self
._format
_text
(self
._out
_files
.screen
, self
._allow
_colors
.screen
, *args
, **kwargs
)
951 def _format_err(self
, *args
, **kwargs
):
952 return self
._format
_text
(self
._out
_files
.error
, self
._allow
_colors
.error
, *args
, **kwargs
)
954 def report_warning(self
, message
, only_once
=False):
956 Print the message to stderr, it will be prefixed with 'WARNING:'
957 If stderr is a tty file the 'WARNING:' will be colored
959 if self
.params
.get('logger') is not None:
960 self
.params
['logger'].warning(message
)
962 if self
.params
.get('no_warnings'):
964 self
.to_stderr(f
'{self._format_err("WARNING:", self.Styles.WARNING)} {message}', only_once
)
966 def deprecation_warning(self
, message
):
967 if self
.params
.get('logger') is not None:
968 self
.params
['logger'].warning(f
'DeprecationWarning: {message}')
970 self
.to_stderr(f
'{self._format_err("DeprecationWarning:", self.Styles.ERROR)} {message}', True)
972 def report_error(self
, message
, *args
, **kwargs
):
974 Do the same as trouble, but prefixes the message with 'ERROR:', colored
975 in red if stderr is a tty file.
977 self
.trouble(f
'{self._format_err("ERROR:", self.Styles.ERROR)} {message}', *args
, **kwargs
)
979 def write_debug(self
, message
, only_once
=False):
980 '''Log debug message or Print message to stderr'''
981 if not self
.params
.get('verbose', False):
983 message
= f
'[debug] {message}'
984 if self
.params
.get('logger'):
985 self
.params
['logger'].debug(message
)
987 self
.to_stderr(message
, only_once
)
989 def report_file_already_downloaded(self
, file_name
):
990 """Report file has already been fully downloaded."""
992 self
.to_screen('[download] %s has already been downloaded' % file_name
)
993 except UnicodeEncodeError:
994 self
.to_screen('[download] The file has already been downloaded')
996 def report_file_delete(self
, file_name
):
997 """Report that existing file will be deleted."""
999 self
.to_screen('Deleting existing file %s' % file_name
)
1000 except UnicodeEncodeError:
1001 self
.to_screen('Deleting existing file')
1003 def raise_no_formats(self
, info
, forced
=False, *, msg
=None):
1004 has_drm
= info
.get('_has_drm')
1005 ignored
, expected
= self
.params
.get('ignore_no_formats_error'), bool(msg
)
1006 msg
= msg
or has_drm
and 'This video is DRM protected' or 'No video formats found!'
1007 if forced
or not ignored
:
1008 raise ExtractorError(msg
, video_id
=info
['id'], ie
=info
['extractor'],
1009 expected
=has_drm
or ignored
or expected
)
1011 self
.report_warning(msg
)
1013 def parse_outtmpl(self
):
1014 self
.deprecation_warning('"YoutubeDL.parse_outtmpl" is deprecated and may be removed in a future version')
1015 self
._parse
_outtmpl
()
1016 return self
.params
['outtmpl']
1018 def _parse_outtmpl(self
):
1020 if self
.params
.get('restrictfilenames'): # Remove spaces in the default template
1021 sanitize
= lambda x
: x
.replace(' - ', ' ').replace(' ', '-')
1023 outtmpl
= self
.params
.setdefault('outtmpl', {})
1024 if not isinstance(outtmpl
, dict):
1025 self
.params
['outtmpl'] = outtmpl
= {'default': outtmpl}
1026 outtmpl
.update({k: sanitize(v) for k, v in DEFAULT_OUTTMPL.items() if outtmpl.get(k) is None}
)
1028 def get_output_path(self
, dir_type
='', filename
=None):
1029 paths
= self
.params
.get('paths', {})
1030 assert isinstance(paths
, dict)
1031 path
= os
.path
.join(
1032 expand_path(paths
.get('home', '').strip()),
1033 expand_path(paths
.get(dir_type
, '').strip()) if dir_type
else '',
1035 return sanitize_path(path
, force
=self
.params
.get('windowsfilenames'))
1038 def _outtmpl_expandpath(outtmpl
):
1039 # expand_path translates '%%' into '%' and '$$' into '$'
1040 # correspondingly that is not what we want since we need to keep
1041 # '%%' intact for template dict substitution step. Working around
1042 # with boundary-alike separator hack.
1043 sep
= ''.join([random
.choice(ascii_letters
) for _
in range(32)])
1044 outtmpl
= outtmpl
.replace('%%', f
'%{sep}%').replace('$$', f
'${sep}$')
1046 # outtmpl should be expand_path'ed before template dict substitution
1047 # because meta fields may contain env variables we don't want to
1048 # be expanded. For example, for outtmpl "%(title)s.%(ext)s" and
1049 # title "Hello $PATH", we don't want `$PATH` to be expanded.
1050 return expand_path(outtmpl
).replace(sep
, '')
1053 def escape_outtmpl(outtmpl
):
1054 ''' Escape any remaining strings like %s, %abc% etc. '''
1056 STR_FORMAT_RE_TMPL
.format('', '(?![%(\0])'),
1057 lambda mobj
: ('' if mobj
.group('has_key') else '%') + mobj
.group(0),
1061 def validate_outtmpl(cls
, outtmpl
):
1062 ''' @return None or Exception object '''
1064 STR_FORMAT_RE_TMPL
.format('[^)]*', '[ljhqBUDS]'),
1065 lambda mobj
: f
'{mobj.group(0)[:-1]}s',
1066 cls
._outtmpl
_expandpath
(outtmpl
))
1068 cls
.escape_outtmpl(outtmpl
) % collections
.defaultdict(int)
1070 except ValueError as err
:
1074 def _copy_infodict(info_dict
):
1075 info_dict
= dict(info_dict
)
1076 info_dict
.pop('__postprocessors', None)
1077 info_dict
.pop('__pending_error', None)
1080 def prepare_outtmpl(self
, outtmpl
, info_dict
, sanitize
=False):
1081 """ Make the outtmpl and info_dict suitable for substitution: ydl.escape_outtmpl(outtmpl) % info_dict
1082 @param sanitize Whether to sanitize the output as a filename.
1083 For backward compatibility, a function can also be passed
1086 info_dict
.setdefault('epoch', int(time
.time())) # keep epoch consistent once set
1088 info_dict
= self
._copy
_infodict
(info_dict
)
1089 info_dict
['duration_string'] = ( # %(duration>%H-%M-%S)s is wrong if duration > 24hrs
1090 formatSeconds(info_dict
['duration'], '-' if sanitize
else ':')
1091 if info_dict
.get('duration', None) is not None
1093 info_dict
['autonumber'] = int(self
.params
.get('autonumber_start', 1) - 1 + self
._num
_downloads
)
1094 info_dict
['video_autonumber'] = self
._num
_videos
1095 if info_dict
.get('resolution') is None:
1096 info_dict
['resolution'] = self
.format_resolution(info_dict
, default
=None)
1098 # For fields playlist_index, playlist_autonumber and autonumber convert all occurrences
1099 # of %(field)s to %(field)0Nd for backward compatibility
1100 field_size_compat_map
= {
1101 'playlist_index': number_of_digits(info_dict
.get('__last_playlist_index') or 0),
1102 'playlist_autonumber': number_of_digits(info_dict
.get('n_entries') or 0),
1103 'autonumber': self
.params
.get('autonumber_size') or 5,
1107 EXTERNAL_FORMAT_RE
= re
.compile(STR_FORMAT_RE_TMPL
.format('[^)]*', f
'[{STR_FORMAT_TYPES}ljhqBUDS]'))
1112 # Field is of the form key1.key2...
1113 # where keys (except first) can be string, int or slice
1114 FIELD_RE
= r
'\w*(?:\.(?:\w+|{num}|{num}?(?::{num}?){{1,2}}))*'.format(num
=r
'(?:-?\d+)')
1115 MATH_FIELD_RE
= rf
'(?:{FIELD_RE}|-?{NUMBER_RE})'
1116 MATH_OPERATORS_RE
= r
'(?:%s)' % '|'.join(map(re
.escape
, MATH_FUNCTIONS
.keys()))
1117 INTERNAL_FORMAT_RE
= re
.compile(rf
'''(?x)
1119 (?P<fields>{FIELD_RE})
1120 (?P<maths>(?:{MATH_OPERATORS_RE}{MATH_FIELD_RE})*)
1121 (?:>(?P<strf_format>.+?))?
1123 (?P<alternate>(?<!\\),[^|&)]+)?
1124 (?:&(?P<replacement>.*?))?
1125 (?:\|(?P<default>.*?))?
1128 def _traverse_infodict(k
):
1132 return traverse_obj(info_dict
, k
, is_user_input
=True, traverse_string
=True)
1134 def get_value(mdict
):
1136 value
= _traverse_infodict(mdict
['fields'])
1139 value
= float_or_none(value
)
1140 if value
is not None:
1143 offset_key
= mdict
['maths']
1145 value
= float_or_none(value
)
1149 MATH_FIELD_RE
if operator
else MATH_OPERATORS_RE
,
1150 offset_key
).group(0)
1151 offset_key
= offset_key
[len(item
):]
1152 if operator
is None:
1153 operator
= MATH_FUNCTIONS
[item
]
1155 item
, multiplier
= (item
[1:], -1) if item
[0] == '-' else (item
, 1)
1156 offset
= float_or_none(item
)
1158 offset
= float_or_none(_traverse_infodict(item
))
1160 value
= operator(value
, multiplier
* offset
)
1161 except (TypeError, ZeroDivisionError):
1164 # Datetime formatting
1165 if mdict
['strf_format']:
1166 value
= strftime_or_none(value
, mdict
['strf_format'].replace('\\,', ','))
1168 # XXX: Workaround for https://github.com/yt-dlp/yt-dlp/issues/4485
1169 if sanitize
and value
== '':
1173 na
= self
.params
.get('outtmpl_na_placeholder', 'NA')
1175 def filename_sanitizer(key
, value
, restricted
=self
.params
.get('restrictfilenames')):
1176 return sanitize_filename(str(value
), restricted
=restricted
, is_id
=(
1177 bool(re
.search(r
'(^|[_.])id(\.|$)', key
))
1178 if 'filename-sanitization' in self
.params
['compat_opts']
1181 sanitizer
= sanitize
if callable(sanitize
) else filename_sanitizer
1182 sanitize
= bool(sanitize
)
1184 def _dumpjson_default(obj
):
1185 if isinstance(obj
, (set, LazyList
)):
1189 def create_key(outer_mobj
):
1190 if not outer_mobj
.group('has_key'):
1191 return outer_mobj
.group(0)
1192 key
= outer_mobj
.group('key')
1193 mobj
= re
.match(INTERNAL_FORMAT_RE
, key
)
1194 initial_field
= mobj
.group('fields') if mobj
else ''
1195 value
, replacement
, default
= None, None, na
1197 mobj
= mobj
.groupdict()
1198 default
= mobj
['default'] if mobj
['default'] is not None else default
1199 value
= get_value(mobj
)
1200 replacement
= mobj
['replacement']
1201 if value
is None and mobj
['alternate']:
1202 mobj
= re
.match(INTERNAL_FORMAT_RE
, mobj
['remaining'][1:])
1206 fmt
= outer_mobj
.group('format')
1207 if fmt
== 's' and value
is not None and key
in field_size_compat_map
.keys():
1208 fmt
= f
'0{field_size_compat_map[key]:d}d'
1210 value
= default
if value
is None else value
if replacement
is None else replacement
1212 flags
= outer_mobj
.group('conversion') or ''
1213 str_fmt
= f
'{fmt[:-1]}s'
1214 if fmt
[-1] == 'l': # list
1215 delim
= '\n' if '#' in flags
else ', '
1216 value
, fmt
= delim
.join(map(str, variadic(value
, allowed_types
=(str, bytes)))), str_fmt
1217 elif fmt
[-1] == 'j': # json
1218 value
, fmt
= json
.dumps(value
, default
=_dumpjson_default
, indent
=4 if '#' in flags
else None), str_fmt
1219 elif fmt
[-1] == 'h': # html
1220 value
, fmt
= escapeHTML(value
), str_fmt
1221 elif fmt
[-1] == 'q': # quoted
1222 value
= map(str, variadic(value
) if '#' in flags
else [value
])
1223 value
, fmt
= ' '.join(map(compat_shlex_quote
, value
)), str_fmt
1224 elif fmt
[-1] == 'B': # bytes
1225 value
= f
'%{str_fmt}'.encode() % str(value
).encode()
1226 value
, fmt
= value
.decode('utf-8', 'ignore'), 's'
1227 elif fmt
[-1] == 'U': # unicode normalized
1228 value
, fmt
= unicodedata
.normalize(
1229 # "+" = compatibility equivalence, "#" = NFD
1230 'NF%s%s' % ('K' if '+' in flags
else '', 'D' if '#' in flags
else 'C'),
1232 elif fmt
[-1] == 'D': # decimal suffix
1233 num_fmt
, fmt
= fmt
[:-1].replace('#', ''), 's'
1234 value
= format_decimal_suffix(value
, f
'%{num_fmt}f%s' if num_fmt
else '%d%s',
1235 factor
=1024 if '#' in flags
else 1000)
1236 elif fmt
[-1] == 'S': # filename sanitization
1237 value
, fmt
= filename_sanitizer(initial_field
, value
, restricted
='#' in flags
), str_fmt
1238 elif fmt
[-1] == 'c':
1240 value
= str(value
)[0]
1243 elif fmt
[-1] not in 'rs': # numeric
1244 value
= float_or_none(value
)
1246 value
, fmt
= default
, 's'
1250 # If value is an object, sanitize might convert it to a string
1251 # So we convert it to repr first
1252 value
, fmt
= repr(value
), str_fmt
1253 if fmt
[-1] in 'csr':
1254 value
= sanitizer(initial_field
, value
)
1256 key
= '%s\0%s' % (key
.replace('%', '%\0'), outer_mobj
.group('format'))
1257 TMPL_DICT
[key
] = value
1258 return '{prefix}%({key}){fmt}'.format(key
=key
, fmt
=fmt
, prefix
=outer_mobj
.group('prefix'))
1260 return EXTERNAL_FORMAT_RE
.sub(create_key
, outtmpl
), TMPL_DICT
1262 def evaluate_outtmpl(self
, outtmpl
, info_dict
, *args
, **kwargs
):
1263 outtmpl
, info_dict
= self
.prepare_outtmpl(outtmpl
, info_dict
, *args
, **kwargs
)
1264 return self
.escape_outtmpl(outtmpl
) % info_dict
1266 def _prepare_filename(self
, info_dict
, *, outtmpl
=None, tmpl_type
=None):
1267 assert None in (outtmpl
, tmpl_type
), 'outtmpl and tmpl_type are mutually exclusive'
1269 outtmpl
= self
.params
['outtmpl'].get(tmpl_type
or 'default', self
.params
['outtmpl']['default'])
1271 outtmpl
= self
._outtmpl
_expandpath
(outtmpl
)
1272 filename
= self
.evaluate_outtmpl(outtmpl
, info_dict
, True)
1276 if tmpl_type
in ('', 'temp'):
1277 final_ext
, ext
= self
.params
.get('final_ext'), info_dict
.get('ext')
1278 if final_ext
and ext
and final_ext
!= ext
and filename
.endswith(f
'.{final_ext}'):
1279 filename
= replace_extension(filename
, ext
, final_ext
)
1281 force_ext
= OUTTMPL_TYPES
[tmpl_type
]
1283 filename
= replace_extension(filename
, force_ext
, info_dict
.get('ext'))
1285 # https://github.com/blackjack4494/youtube-dlc/issues/85
1286 trim_file_name
= self
.params
.get('trim_file_name', False)
1288 no_ext
, *ext
= filename
.rsplit('.', 2)
1289 filename
= join_nonempty(no_ext
[:trim_file_name
], *ext
, delim
='.')
1292 except ValueError as err
:
1293 self
.report_error('Error in output template: ' + str(err
) + ' (encoding: ' + repr(preferredencoding()) + ')')
1296 def prepare_filename(self
, info_dict
, dir_type
='', *, outtmpl
=None, warn
=False):
1297 """Generate the output filename"""
1299 assert not dir_type
, 'outtmpl and dir_type are mutually exclusive'
1301 filename
= self
._prepare
_filename
(info_dict
, tmpl_type
=dir_type
, outtmpl
=outtmpl
)
1302 if not filename
and dir_type
not in ('', 'temp'):
1306 if not self
.params
.get('paths'):
1308 elif filename
== '-':
1309 self
.report_warning('--paths is ignored when an outputting to stdout', only_once
=True)
1310 elif os
.path
.isabs(filename
):
1311 self
.report_warning('--paths is ignored since an absolute path is given in output template', only_once
=True)
1312 if filename
== '-' or not filename
:
1315 return self
.get_output_path(dir_type
, filename
)
1317 def _match_entry(self
, info_dict
, incomplete
=False, silent
=False):
1318 """ Returns None if the file should be downloaded """
1320 video_title
= info_dict
.get('title', info_dict
.get('id', 'entry'))
1323 if 'title' in info_dict
:
1324 # This can happen when we're just evaluating the playlist
1325 title
= info_dict
['title']
1326 matchtitle
= self
.params
.get('matchtitle', False)
1328 if not re
.search(matchtitle
, title
, re
.IGNORECASE
):
1329 return '"' + title
+ '" title did not match pattern "' + matchtitle
+ '"'
1330 rejecttitle
= self
.params
.get('rejecttitle', False)
1332 if re
.search(rejecttitle
, title
, re
.IGNORECASE
):
1333 return '"' + title
+ '" title matched reject pattern "' + rejecttitle
+ '"'
1334 date
= info_dict
.get('upload_date')
1335 if date
is not None:
1336 dateRange
= self
.params
.get('daterange', DateRange())
1337 if date
not in dateRange
:
1338 return f
'{date_from_str(date).isoformat()} upload date is not in range {dateRange}'
1339 view_count
= info_dict
.get('view_count')
1340 if view_count
is not None:
1341 min_views
= self
.params
.get('min_views')
1342 if min_views
is not None and view_count
< min_views
:
1343 return 'Skipping %s, because it has not reached minimum view count (%d/%d)' % (video_title
, view_count
, min_views
)
1344 max_views
= self
.params
.get('max_views')
1345 if max_views
is not None and view_count
> max_views
:
1346 return 'Skipping %s, because it has exceeded the maximum view count (%d/%d)' % (video_title
, view_count
, max_views
)
1347 if age_restricted(info_dict
.get('age_limit'), self
.params
.get('age_limit')):
1348 return 'Skipping "%s" because it is age restricted' % video_title
1350 match_filter
= self
.params
.get('match_filter')
1351 if match_filter
is not None:
1353 ret
= match_filter(info_dict
, incomplete
=incomplete
)
1355 # For backward compatibility
1356 ret
= None if incomplete
else match_filter(info_dict
)
1357 if ret
is NO_DEFAULT
:
1359 filename
= self
._format
_screen
(self
.prepare_filename(info_dict
), self
.Styles
.FILENAME
)
1360 reply
= input(self
._format
_screen
(
1361 f
'Download "{filename}"? (Y/n): ', self
.Styles
.EMPHASIS
)).lower().strip()
1362 if reply
in {'y', ''}
:
1365 return f
'Skipping {video_title}'
1366 elif ret
is not None:
1370 if self
.in_download_archive(info_dict
):
1371 reason
= '%s has already been recorded in the archive' % video_title
1372 break_opt
, break_err
= 'break_on_existing', ExistingVideoReached
1374 reason
= check_filter()
1375 break_opt
, break_err
= 'break_on_reject', RejectedVideoReached
1376 if reason
is not None:
1378 self
.to_screen('[download] ' + reason
)
1379 if self
.params
.get(break_opt
, False):
1384 def add_extra_info(info_dict
, extra_info
):
1385 '''Set the keys from extra_info in info dict if they are missing'''
1386 for key
, value
in extra_info
.items():
1387 info_dict
.setdefault(key
, value
)
1389 def extract_info(self
, url
, download
=True, ie_key
=None, extra_info
=None,
1390 process
=True, force_generic_extractor
=False):
1392 Return a list with a dictionary for each video extracted.
1395 url -- URL to extract
1398 download -- whether to download videos during extraction
1399 ie_key -- extractor key hint
1400 extra_info -- dictionary containing the extra values to add to each result
1401 process -- whether to resolve all unresolved references (URLs, playlist items),
1402 must be True for download to work.
1403 force_generic_extractor -- force using the generic extractor
1406 if extra_info
is None:
1409 if not ie_key
and force_generic_extractor
:
1413 ies
= {ie_key: self._get_info_extractor_class(ie_key)}
1417 for ie_key
, ie
in ies
.items():
1418 if not ie
.suitable(url
):
1421 if not ie
.working():
1422 self
.report_warning('The program functionality for this site has been marked as broken, '
1423 'and will probably not work.')
1425 temp_id
= ie
.get_temp_id(url
)
1426 if temp_id
is not None and self
.in_download_archive({'id': temp_id, 'ie_key': ie_key}
):
1427 self
.to_screen(f
'[{ie_key}] {temp_id}: has already been recorded in the archive')
1428 if self
.params
.get('break_on_existing', False):
1429 raise ExistingVideoReached()
1431 return self
.__extract
_info
(url
, self
.get_info_extractor(ie_key
), download
, extra_info
, process
)
1433 self
.report_error('no suitable InfoExtractor for URL %s' % url
)
1435 def _handle_extraction_exceptions(func
):
1436 @functools.wraps(func
)
1437 def wrapper(self
, *args
, **kwargs
):
1440 return func(self
, *args
, **kwargs
)
1441 except (DownloadCancelled
, LazyList
.IndexError, PagedList
.IndexError):
1443 except ReExtractInfo
as e
:
1445 self
.to_screen(f
'{e}; Re-extracting data')
1447 self
.to_stderr('\r')
1448 self
.report_warning(f
'{e}; Re-extracting data')
1450 except GeoRestrictedError
as e
:
1453 msg
+= '\nThis video is available in %s.' % ', '.join(
1454 map(ISO3166Utils
.short2full
, e
.countries
))
1455 msg
+= '\nYou might want to use a VPN or a proxy server (with --proxy) to workaround.'
1456 self
.report_error(msg
)
1457 except ExtractorError
as e
: # An error we somewhat expected
1458 self
.report_error(str(e
), e
.format_traceback())
1459 except Exception as e
:
1460 if self
.params
.get('ignoreerrors'):
1461 self
.report_error(str(e
), tb
=encode_compat_str(traceback
.format_exc()))
1467 def _wait_for_video(self
, ie_result
={}):
1468 if (not self
.params
.get('wait_for_video')
1469 or ie_result
.get('_type', 'video') != 'video'
1470 or ie_result
.get('formats') or ie_result
.get('url')):
1473 format_dur
= lambda dur
: '%02d:%02d:%02d' % timetuple_from_msec(dur
* 1000)[:-1]
1478 full_msg
= f
'{msg}\n'
1479 if not self
.params
.get('noprogress'):
1480 full_msg
= msg
+ ' ' * (len(last_msg
) - len(msg
)) + '\r'
1483 self
.to_screen(full_msg
, skip_eol
=True)
1486 min_wait
, max_wait
= self
.params
.get('wait_for_video')
1487 diff
= try_get(ie_result
, lambda x
: x
['release_timestamp'] - time
.time())
1488 if diff
is None and ie_result
.get('live_status') == 'is_upcoming':
1489 diff
= round(random
.uniform(min_wait
, max_wait
) if (max_wait
and min_wait
) else (max_wait
or min_wait
), 0)
1490 self
.report_warning('Release time of video is not known')
1491 elif ie_result
and (diff
or 0) <= 0:
1492 self
.report_warning('Video should already be available according to extracted info')
1493 diff
= min(max(diff
or 0, min_wait
or 0), max_wait
or float('inf'))
1494 self
.to_screen(f
'[wait] Waiting for {format_dur(diff)} - Press Ctrl+C to try now')
1496 wait_till
= time
.time() + diff
1499 diff
= wait_till
- time
.time()
1502 raise ReExtractInfo('[wait] Wait period ended', expected
=True)
1503 progress(f
'[wait] Remaining time until next attempt: {self._format_screen(format_dur(diff), self.Styles.EMPHASIS)}')
1505 except KeyboardInterrupt:
1507 raise ReExtractInfo('[wait] Interrupted by user', expected
=True)
1508 except BaseException
as e
:
1509 if not isinstance(e
, ReExtractInfo
):
1513 @_handle_extraction_exceptions
1514 def __extract_info(self
, url
, ie
, download
, extra_info
, process
):
1516 ie_result
= ie
.extract(url
)
1517 except UserNotLive
as e
:
1519 if self
.params
.get('wait_for_video'):
1520 self
.report_warning(e
)
1521 self
._wait
_for
_video
()
1523 if ie_result
is None: # Finished already (backwards compatibility; listformats and friends should be moved here)
1524 self
.report_warning(f
'Extractor {ie.IE_NAME} returned nothing{bug_reports_message()}')
1526 if isinstance(ie_result
, list):
1527 # Backwards compatibility: old IE result format
1529 '_type': 'compat_list',
1530 'entries': ie_result
,
1532 if extra_info
.get('original_url'):
1533 ie_result
.setdefault('original_url', extra_info
['original_url'])
1534 self
.add_default_extra_info(ie_result
, ie
, url
)
1536 self
._wait
_for
_video
(ie_result
)
1537 return self
.process_ie_result(ie_result
, download
, extra_info
)
1541 def add_default_extra_info(self
, ie_result
, ie
, url
):
1543 self
.add_extra_info(ie_result
, {
1545 'original_url': url
,
1547 webpage_url
= ie_result
.get('webpage_url')
1549 self
.add_extra_info(ie_result
, {
1550 'webpage_url_basename': url_basename(webpage_url
),
1551 'webpage_url_domain': get_domain(webpage_url
),
1554 self
.add_extra_info(ie_result
, {
1555 'extractor': ie
.IE_NAME
,
1556 'extractor_key': ie
.ie_key(),
1559 def process_ie_result(self
, ie_result
, download
=True, extra_info
=None):
1561 Take the result of the ie(may be modified) and resolve all unresolved
1562 references (URLs, playlist items).
1564 It will also download the videos if 'download'.
1565 Returns the resolved ie_result.
1567 if extra_info
is None:
1569 result_type
= ie_result
.get('_type', 'video')
1571 if result_type
in ('url', 'url_transparent'):
1572 ie_result
['url'] = sanitize_url(
1573 ie_result
['url'], scheme
='http' if self
.params
.get('prefer_insecure') else 'https')
1574 if ie_result
.get('original_url'):
1575 extra_info
.setdefault('original_url', ie_result
['original_url'])
1577 extract_flat
= self
.params
.get('extract_flat', False)
1578 if ((extract_flat
== 'in_playlist' and 'playlist' in extra_info
)
1579 or extract_flat
is True):
1580 info_copy
= ie_result
.copy()
1581 ie
= try_get(ie_result
.get('ie_key'), self
.get_info_extractor
)
1582 if ie
and not ie_result
.get('id'):
1583 info_copy
['id'] = ie
.get_temp_id(ie_result
['url'])
1584 self
.add_default_extra_info(info_copy
, ie
, ie_result
['url'])
1585 self
.add_extra_info(info_copy
, extra_info
)
1586 info_copy
, _
= self
.pre_process(info_copy
)
1587 self
.__forced
_printings
(info_copy
, self
.prepare_filename(info_copy
), incomplete
=True)
1588 self
._raise
_pending
_errors
(info_copy
)
1589 if self
.params
.get('force_write_download_archive', False):
1590 self
.record_download_archive(info_copy
)
1593 if result_type
== 'video':
1594 self
.add_extra_info(ie_result
, extra_info
)
1595 ie_result
= self
.process_video_result(ie_result
, download
=download
)
1596 self
._raise
_pending
_errors
(ie_result
)
1597 additional_urls
= (ie_result
or {}).get('additional_urls')
1599 # TODO: Improve MetadataParserPP to allow setting a list
1600 if isinstance(additional_urls
, str):
1601 additional_urls
= [additional_urls
]
1603 '[info] %s: %d additional URL(s) requested' % (ie_result
['id'], len(additional_urls
)))
1604 self
.write_debug('Additional URLs: "%s"' % '", "'.join(additional_urls
))
1605 ie_result
['additional_entries'] = [
1607 url
, download
, extra_info
=extra_info
,
1608 force_generic_extractor
=self
.params
.get('force_generic_extractor'))
1609 for url
in additional_urls
1612 elif result_type
== 'url':
1613 # We have to add extra_info to the results because it may be
1614 # contained in a playlist
1615 return self
.extract_info(
1616 ie_result
['url'], download
,
1617 ie_key
=ie_result
.get('ie_key'),
1618 extra_info
=extra_info
)
1619 elif result_type
== 'url_transparent':
1620 # Use the information from the embedding page
1621 info
= self
.extract_info(
1622 ie_result
['url'], ie_key
=ie_result
.get('ie_key'),
1623 extra_info
=extra_info
, download
=False, process
=False)
1625 # extract_info may return None when ignoreerrors is enabled and
1626 # extraction failed with an error, don't crash and return early
1631 exempted_fields
= {'_type', 'url', 'ie_key'}
1632 if not ie_result
.get('section_end') and ie_result
.get('section_start') is None:
1633 # For video clips, the id etc of the clip extractor should be used
1634 exempted_fields |
= {'id', 'extractor', 'extractor_key'}
1636 new_result
= info
.copy()
1637 new_result
.update(filter_dict(ie_result
, lambda k
, v
: v
is not None and k
not in exempted_fields
))
1639 # Extracted info may not be a video result (i.e.
1640 # info.get('_type', 'video') != video) but rather an url or
1641 # url_transparent. In such cases outer metadata (from ie_result)
1642 # should be propagated to inner one (info). For this to happen
1643 # _type of info should be overridden with url_transparent. This
1644 # fixes issue from https://github.com/ytdl-org/youtube-dl/pull/11163.
1645 if new_result
.get('_type') == 'url':
1646 new_result
['_type'] = 'url_transparent'
1648 return self
.process_ie_result(
1649 new_result
, download
=download
, extra_info
=extra_info
)
1650 elif result_type
in ('playlist', 'multi_video'):
1651 # Protect from infinite recursion due to recursively nested playlists
1652 # (see https://github.com/ytdl-org/youtube-dl/issues/27833)
1653 webpage_url
= ie_result
['webpage_url']
1654 if webpage_url
in self
._playlist
_urls
:
1656 '[download] Skipping already downloaded playlist: %s'
1657 % ie_result
.get('title') or ie_result
.get('id'))
1660 self
._playlist
_level
+= 1
1661 self
._playlist
_urls
.add(webpage_url
)
1662 self
._fill
_common
_fields
(ie_result
, False)
1663 self
._sanitize
_thumbnails
(ie_result
)
1665 return self
.__process
_playlist
(ie_result
, download
)
1667 self
._playlist
_level
-= 1
1668 if not self
._playlist
_level
:
1669 self
._playlist
_urls
.clear()
1670 elif result_type
== 'compat_list':
1671 self
.report_warning(
1672 'Extractor %s returned a compat_list result. '
1673 'It needs to be updated.' % ie_result
.get('extractor'))
1676 self
.add_extra_info(r
, {
1677 'extractor': ie_result
['extractor'],
1678 'webpage_url': ie_result
['webpage_url'],
1679 'webpage_url_basename': url_basename(ie_result
['webpage_url']),
1680 'webpage_url_domain': get_domain(ie_result
['webpage_url']),
1681 'extractor_key': ie_result
['extractor_key'],
1684 ie_result
['entries'] = [
1685 self
.process_ie_result(_fixup(r
), download
, extra_info
)
1686 for r
in ie_result
['entries']
1690 raise Exception('Invalid result type: %s' % result_type
)
1692 def _ensure_dir_exists(self
, path
):
1693 return make_dir(path
, self
.report_error
)
1696 def _playlist_infodict(ie_result
, strict
=False, **kwargs
):
1698 'playlist_count': ie_result
.get('playlist_count'),
1699 'playlist': ie_result
.get('title') or ie_result
.get('id'),
1700 'playlist_id': ie_result
.get('id'),
1701 'playlist_title': ie_result
.get('title'),
1702 'playlist_uploader': ie_result
.get('uploader'),
1703 'playlist_uploader_id': ie_result
.get('uploader_id'),
1710 'playlist_index': 0,
1711 '__last_playlist_index': max(ie_result
['requested_entries'] or (0, 0)),
1712 'extractor': ie_result
['extractor'],
1713 'webpage_url': ie_result
['webpage_url'],
1714 'webpage_url_basename': url_basename(ie_result
['webpage_url']),
1715 'webpage_url_domain': get_domain(ie_result
['webpage_url']),
1716 'extractor_key': ie_result
['extractor_key'],
1719 def __process_playlist(self
, ie_result
, download
):
1720 """Process each entry in the playlist"""
1721 assert ie_result
['_type'] in ('playlist', 'multi_video')
1723 common_info
= self
._playlist
_infodict
(ie_result
, strict
=True)
1724 title
= common_info
.get('playlist') or '<Untitled>'
1725 if self
._match
_entry
(common_info
, incomplete
=True) is not None:
1727 self
.to_screen(f
'[download] Downloading {ie_result["_type"]}: {title}')
1729 all_entries
= PlaylistEntries(self
, ie_result
)
1730 entries
= orderedSet(all_entries
.get_requested_items(), lazy
=True)
1732 lazy
= self
.params
.get('lazy_playlist')
1734 resolved_entries
, n_entries
= [], 'N/A'
1735 ie_result
['requested_entries'], ie_result
['entries'] = None, None
1737 entries
= resolved_entries
= list(entries
)
1738 n_entries
= len(resolved_entries
)
1739 ie_result
['requested_entries'], ie_result
['entries'] = tuple(zip(*resolved_entries
)) or ([], [])
1740 if not ie_result
.get('playlist_count'):
1741 # Better to do this after potentially exhausting entries
1742 ie_result
['playlist_count'] = all_entries
.get_full_count()
1744 extra
= self
._playlist
_infodict
(ie_result
, n_entries
=int_or_none(n_entries
))
1745 ie_copy
= collections
.ChainMap(ie_result
, extra
)
1747 _infojson_written
= False
1748 write_playlist_files
= self
.params
.get('allow_playlist_files', True)
1749 if write_playlist_files
and self
.params
.get('list_thumbnails'):
1750 self
.list_thumbnails(ie_result
)
1751 if write_playlist_files
and not self
.params
.get('simulate'):
1752 _infojson_written
= self
._write
_info
_json
(
1753 'playlist', ie_result
, self
.prepare_filename(ie_copy
, 'pl_infojson'))
1754 if _infojson_written
is None:
1756 if self
._write
_description
('playlist', ie_result
,
1757 self
.prepare_filename(ie_copy
, 'pl_description')) is None:
1759 # TODO: This should be passed to ThumbnailsConvertor if necessary
1760 self
._write
_thumbnails
('playlist', ie_result
, self
.prepare_filename(ie_copy
, 'pl_thumbnail'))
1763 if self
.params
.get('playlistreverse') or self
.params
.get('playlistrandom'):
1764 self
.report_warning('playlistreverse and playlistrandom are not supported with lazy_playlist', only_once
=True)
1765 elif self
.params
.get('playlistreverse'):
1767 elif self
.params
.get('playlistrandom'):
1768 random
.shuffle(entries
)
1770 self
.to_screen(f
'[{ie_result["extractor"]}] Playlist {title}: Downloading {n_entries} videos'
1771 f
'{format_field(ie_result, "playlist_count", " of %s")}')
1773 keep_resolved_entries
= self
.params
.get('extract_flat') != 'discard'
1774 if self
.params
.get('extract_flat') == 'discard_in_playlist':
1775 keep_resolved_entries
= ie_result
['_type'] != 'playlist'
1776 if keep_resolved_entries
:
1777 self
.write_debug('The information of all playlist entries will be held in memory')
1780 max_failures
= self
.params
.get('skip_playlist_after_errors') or float('inf')
1781 for i
, (playlist_index
, entry
) in enumerate(entries
):
1783 resolved_entries
.append((playlist_index
, entry
))
1787 entry
['__x_forwarded_for_ip'] = ie_result
.get('__x_forwarded_for_ip')
1788 if not lazy
and 'playlist-index' in self
.params
.get('compat_opts', []):
1789 playlist_index
= ie_result
['requested_entries'][i
]
1791 entry_copy
= collections
.ChainMap(entry
, {
1793 'n_entries': int_or_none(n_entries
),
1794 'playlist_index': playlist_index
,
1795 'playlist_autonumber': i
+ 1,
1798 if self
._match
_entry
(entry_copy
, incomplete
=True) is not None:
1801 self
.to_screen('[download] Downloading video %s of %s' % (
1802 self
._format
_screen
(i
+ 1, self
.Styles
.ID
), self
._format
_screen
(n_entries
, self
.Styles
.EMPHASIS
)))
1805 'playlist_index': playlist_index
,
1806 'playlist_autonumber': i
+ 1,
1808 entry_result
= self
.__process
_iterable
_entry
(entry
, download
, extra
)
1809 if not entry_result
:
1811 if failures
>= max_failures
:
1813 f
'Skipping the remaining entries in playlist "{title}" since {failures} items failed extraction')
1815 if keep_resolved_entries
:
1816 resolved_entries
[i
] = (playlist_index
, entry_result
)
1818 # Update with processed data
1819 ie_result
['requested_entries'], ie_result
['entries'] = tuple(zip(*resolved_entries
)) or ([], [])
1821 # Write the updated info to json
1822 if _infojson_written
is True and self
._write
_info
_json
(
1823 'updated playlist', ie_result
,
1824 self
.prepare_filename(ie_copy
, 'pl_infojson'), overwrite
=True) is None:
1827 ie_result
= self
.run_all_pps('playlist', ie_result
)
1828 self
.to_screen(f
'[download] Finished downloading playlist: {title}')
1831 @_handle_extraction_exceptions
1832 def __process_iterable_entry(self
, entry
, download
, extra_info
):
1833 return self
.process_ie_result(
1834 entry
, download
=download
, extra_info
=extra_info
)
1836 def _build_format_filter(self
, filter_spec
):
1837 " Returns a function to filter the formats according to the filter_spec "
1847 operator_rex
= re
.compile(r
'''(?x)\s*
1848 (?P<key>width|height|tbr|abr|vbr|asr|filesize|filesize_approx|fps)\s*
1849 (?P<op>%s)(?P<none_inclusive>\s*\?)?\s*
1850 (?P<value>[0-9.]+(?:[kKmMgGtTpPeEzZyY]i?[Bb]?)?)\s*
1851 ''' % '|'.join(map(re
.escape
, OPERATORS
.keys())))
1852 m
= operator_rex
.fullmatch(filter_spec
)
1855 comparison_value
= int(m
.group('value'))
1857 comparison_value
= parse_filesize(m
.group('value'))
1858 if comparison_value
is None:
1859 comparison_value
= parse_filesize(m
.group('value') + 'B')
1860 if comparison_value
is None:
1862 'Invalid value %r in format specification %r' % (
1863 m
.group('value'), filter_spec
))
1864 op
= OPERATORS
[m
.group('op')]
1869 '^=': lambda attr
, value
: attr
.startswith(value
),
1870 '$=': lambda attr
, value
: attr
.endswith(value
),
1871 '*=': lambda attr
, value
: value
in attr
,
1872 '~=': lambda attr
, value
: value
.search(attr
) is not None
1874 str_operator_rex
= re
.compile(r
'''(?x)\s*
1875 (?P<key>[a-zA-Z0-9._-]+)\s*
1876 (?P<negation>!\s*)?(?P<op>%s)\s*(?P<none_inclusive>\?\s*)?
1878 (?P<value>(?(quote)(?:(?!(?P=quote))[^\\]|\\.)+|[\w.-]+))
1879 (?(quote)(?P=quote))\s*
1880 ''' % '|'.join(map(re
.escape
, STR_OPERATORS
.keys())))
1881 m
= str_operator_rex
.fullmatch(filter_spec
)
1883 if m
.group('op') == '~=':
1884 comparison_value
= re
.compile(m
.group('value'))
1886 comparison_value
= re
.sub(r
'''\\([\\"'])''', r
'\1', m
.group('value'))
1887 str_op
= STR_OPERATORS
[m
.group('op')]
1888 if m
.group('negation'):
1889 op
= lambda attr
, value
: not str_op(attr
, value
)
1894 raise SyntaxError('Invalid filter specification %r' % filter_spec
)
1897 actual_value
= f
.get(m
.group('key'))
1898 if actual_value
is None:
1899 return m
.group('none_inclusive')
1900 return op(actual_value
, comparison_value
)
1903 def _check_formats(self
, formats
):
1905 self
.to_screen('[info] Testing format %s' % f
['format_id'])
1906 path
= self
.get_output_path('temp')
1907 if not self
._ensure
_dir
_exists
(f
'{path}/'):
1909 temp_file
= tempfile
.NamedTemporaryFile(suffix
='.tmp', delete
=False, dir=path
or None)
1912 success
, _
= self
.dl(temp_file
.name
, f
, test
=True)
1913 except (DownloadError
, OSError, ValueError) + network_exceptions
:
1916 if os
.path
.exists(temp_file
.name
):
1918 os
.remove(temp_file
.name
)
1920 self
.report_warning('Unable to delete temporary file "%s"' % temp_file
.name
)
1924 self
.to_screen('[info] Unable to download format %s. Skipping...' % f
['format_id'])
1926 def _default_format_spec(self
, info_dict
, download
=True):
1929 merger
= FFmpegMergerPP(self
)
1930 return merger
.available
and merger
.can_merge()
1933 not self
.params
.get('simulate')
1937 or info_dict
.get('is_live') and not self
.params
.get('live_from_start')
1938 or self
.params
['outtmpl']['default'] == '-'))
1941 or self
.params
.get('allow_multiple_audio_streams', False)
1942 or 'format-spec' in self
.params
['compat_opts'])
1945 'best/bestvideo+bestaudio' if prefer_best
1946 else 'bestvideo*+bestaudio/best' if not compat
1947 else 'bestvideo+bestaudio/best')
1949 def build_format_selector(self
, format_spec
):
1950 def syntax_error(note
, start
):
1952 'Invalid format specification: '
1953 '{}\n\t{}\n\t{}^'.format(note
, format_spec
, ' ' * start
[1]))
1954 return SyntaxError(message
)
1956 PICKFIRST
= 'PICKFIRST'
1960 FormatSelector
= collections
.namedtuple('FormatSelector', ['type', 'selector', 'filters'])
1962 allow_multiple_streams
= {'audio': self
.params
.get('allow_multiple_audio_streams', False),
1963 'video': self
.params
.get('allow_multiple_video_streams', False)}
1965 check_formats
= self
.params
.get('check_formats') == 'selected'
1967 def _parse_filter(tokens
):
1969 for type, string
, start
, _
, _
in tokens
:
1970 if type == tokenize
.OP
and string
== ']':
1971 return ''.join(filter_parts
)
1973 filter_parts
.append(string
)
1975 def _remove_unused_ops(tokens
):
1976 # Remove operators that we don't use and join them with the surrounding strings
1977 # for example: 'mp4' '-' 'baseline' '-' '16x9' is converted to 'mp4-baseline-16x9'
1978 ALLOWED_OPS
= ('/', '+', ',', '(', ')')
1979 last_string
, last_start
, last_end
, last_line
= None, None, None, None
1980 for type, string
, start
, end
, line
in tokens
:
1981 if type == tokenize
.OP
and string
== '[':
1983 yield tokenize
.NAME
, last_string
, last_start
, last_end
, last_line
1985 yield type, string
, start
, end
, line
1986 # everything inside brackets will be handled by _parse_filter
1987 for type, string
, start
, end
, line
in tokens
:
1988 yield type, string
, start
, end
, line
1989 if type == tokenize
.OP
and string
== ']':
1991 elif type == tokenize
.OP
and string
in ALLOWED_OPS
:
1993 yield tokenize
.NAME
, last_string
, last_start
, last_end
, last_line
1995 yield type, string
, start
, end
, line
1996 elif type in [tokenize
.NAME
, tokenize
.NUMBER
, tokenize
.OP
]:
1998 last_string
= string
2002 last_string
+= string
2004 yield tokenize
.NAME
, last_string
, last_start
, last_end
, last_line
2006 def _parse_format_selection(tokens
, inside_merge
=False, inside_choice
=False, inside_group
=False):
2008 current_selector
= None
2009 for type, string
, start
, _
, _
in tokens
:
2010 # ENCODING is only defined in python 3.x
2011 if type == getattr(tokenize
, 'ENCODING', None):
2013 elif type in [tokenize
.NAME
, tokenize
.NUMBER
]:
2014 current_selector
= FormatSelector(SINGLE
, string
, [])
2015 elif type == tokenize
.OP
:
2017 if not inside_group
:
2018 # ')' will be handled by the parentheses group
2019 tokens
.restore_last_token()
2021 elif inside_merge
and string
in ['/', ',']:
2022 tokens
.restore_last_token()
2024 elif inside_choice
and string
== ',':
2025 tokens
.restore_last_token()
2028 if not current_selector
:
2029 raise syntax_error('"," must follow a format selector', start
)
2030 selectors
.append(current_selector
)
2031 current_selector
= None
2033 if not current_selector
:
2034 raise syntax_error('"/" must follow a format selector', start
)
2035 first_choice
= current_selector
2036 second_choice
= _parse_format_selection(tokens
, inside_choice
=True)
2037 current_selector
= FormatSelector(PICKFIRST
, (first_choice
, second_choice
), [])
2039 if not current_selector
:
2040 current_selector
= FormatSelector(SINGLE
, 'best', [])
2041 format_filter
= _parse_filter(tokens
)
2042 current_selector
.filters
.append(format_filter
)
2044 if current_selector
:
2045 raise syntax_error('Unexpected "("', start
)
2046 group
= _parse_format_selection(tokens
, inside_group
=True)
2047 current_selector
= FormatSelector(GROUP
, group
, [])
2049 if not current_selector
:
2050 raise syntax_error('Unexpected "+"', start
)
2051 selector_1
= current_selector
2052 selector_2
= _parse_format_selection(tokens
, inside_merge
=True)
2054 raise syntax_error('Expected a selector', start
)
2055 current_selector
= FormatSelector(MERGE
, (selector_1
, selector_2
), [])
2057 raise syntax_error(f
'Operator not recognized: "{string}"', start
)
2058 elif type == tokenize
.ENDMARKER
:
2060 if current_selector
:
2061 selectors
.append(current_selector
)
2064 def _merge(formats_pair
):
2065 format_1
, format_2
= formats_pair
2068 formats_info
.extend(format_1
.get('requested_formats', (format_1
,)))
2069 formats_info
.extend(format_2
.get('requested_formats', (format_2
,)))
2071 if not allow_multiple_streams
['video'] or not allow_multiple_streams
['audio']:
2072 get_no_more
= {'video': False, 'audio': False}
2073 for (i
, fmt_info
) in enumerate(formats_info
):
2074 if fmt_info
.get('acodec') == fmt_info
.get('vcodec') == 'none':
2077 for aud_vid
in ['audio', 'video']:
2078 if not allow_multiple_streams
[aud_vid
] and fmt_info
.get(aud_vid
[0] + 'codec') != 'none':
2079 if get_no_more
[aud_vid
]:
2082 get_no_more
[aud_vid
] = True
2084 if len(formats_info
) == 1:
2085 return formats_info
[0]
2087 video_fmts
= [fmt_info
for fmt_info
in formats_info
if fmt_info
.get('vcodec') != 'none']
2088 audio_fmts
= [fmt_info
for fmt_info
in formats_info
if fmt_info
.get('acodec') != 'none']
2090 the_only_video
= video_fmts
[0] if len(video_fmts
) == 1 else None
2091 the_only_audio
= audio_fmts
[0] if len(audio_fmts
) == 1 else None
2093 output_ext
= get_compatible_ext(
2094 vcodecs
=[f
.get('vcodec') for f
in video_fmts
],
2095 acodecs
=[f
.get('acodec') for f
in audio_fmts
],
2096 vexts
=[f
['ext'] for f
in video_fmts
],
2097 aexts
=[f
['ext'] for f
in audio_fmts
],
2098 preferences
=(try_call(lambda: self
.params
['merge_output_format'].split('/'))
2099 or self
.params
.get('prefer_free_formats') and ('webm', 'mkv')))
2101 filtered
= lambda *keys
: filter(None, (traverse_obj(fmt
, *keys
) for fmt
in formats_info
))
2104 'requested_formats': formats_info
,
2105 'format': '+'.join(filtered('format')),
2106 'format_id': '+'.join(filtered('format_id')),
2108 'protocol': '+'.join(map(determine_protocol
, formats_info
)),
2109 'language': '+'.join(orderedSet(filtered('language'))) or None,
2110 'format_note': '+'.join(orderedSet(filtered('format_note'))) or None,
2111 'filesize_approx': sum(filtered('filesize', 'filesize_approx')) or None,
2112 'tbr': sum(filtered('tbr', 'vbr', 'abr')),
2117 'width': the_only_video
.get('width'),
2118 'height': the_only_video
.get('height'),
2119 'resolution': the_only_video
.get('resolution') or self
.format_resolution(the_only_video
),
2120 'fps': the_only_video
.get('fps'),
2121 'dynamic_range': the_only_video
.get('dynamic_range'),
2122 'vcodec': the_only_video
.get('vcodec'),
2123 'vbr': the_only_video
.get('vbr'),
2124 'stretched_ratio': the_only_video
.get('stretched_ratio'),
2129 'acodec': the_only_audio
.get('acodec'),
2130 'abr': the_only_audio
.get('abr'),
2131 'asr': the_only_audio
.get('asr'),
2136 def _check_formats(formats
):
2137 if not check_formats
:
2140 yield from self
._check
_formats
(formats
)
2142 def _build_selector_function(selector
):
2143 if isinstance(selector
, list): # ,
2144 fs
= [_build_selector_function(s
) for s
in selector
]
2146 def selector_function(ctx
):
2149 return selector_function
2151 elif selector
.type == GROUP
: # ()
2152 selector_function
= _build_selector_function(selector
.selector
)
2154 elif selector
.type == PICKFIRST
: # /
2155 fs
= [_build_selector_function(s
) for s
in selector
.selector
]
2157 def selector_function(ctx
):
2159 picked_formats
= list(f(ctx
))
2161 return picked_formats
2164 elif selector
.type == MERGE
: # +
2165 selector_1
, selector_2
= map(_build_selector_function
, selector
.selector
)
2167 def selector_function(ctx
):
2168 for pair
in itertools
.product(selector_1(ctx
), selector_2(ctx
)):
2171 elif selector
.type == SINGLE
: # atom
2172 format_spec
= selector
.selector
or 'best'
2174 # TODO: Add allvideo, allaudio etc by generalizing the code with best/worst selector
2175 if format_spec
== 'all':
2176 def selector_function(ctx
):
2177 yield from _check_formats(ctx
['formats'][::-1])
2178 elif format_spec
== 'mergeall':
2179 def selector_function(ctx
):
2180 formats
= list(_check_formats(
2181 f
for f
in ctx
['formats'] if f
.get('vcodec') != 'none' or f
.get('acodec') != 'none'))
2184 merged_format
= formats
[-1]
2185 for f
in formats
[-2::-1]:
2186 merged_format
= _merge((merged_format
, f
))
2190 format_fallback
, seperate_fallback
, format_reverse
, format_idx
= False, None, True, 1
2192 r
'(?P<bw>best|worst|b|w)(?P<type>video|audio|v|a)?(?P<mod>\*)?(?:\.(?P<n>[1-9]\d*))?$',
2194 if mobj
is not None:
2195 format_idx
= int_or_none(mobj
.group('n'), default
=1)
2196 format_reverse
= mobj
.group('bw')[0] == 'b'
2197 format_type
= (mobj
.group('type') or [None])[0]
2198 not_format_type
= {'v': 'a', 'a': 'v'}
.get(format_type
)
2199 format_modified
= mobj
.group('mod') is not None
2201 format_fallback
= not format_type
and not format_modified
# for b, w
2203 (lambda f
: f
.get('%scodec' % format_type
) != 'none')
2204 if format_type
and format_modified
# bv*, ba*, wv*, wa*
2205 else (lambda f
: f
.get('%scodec' % not_format_type
) == 'none')
2206 if format_type
# bv, ba, wv, wa
2207 else (lambda f
: f
.get('vcodec') != 'none' and f
.get('acodec') != 'none')
2208 if not format_modified
# b, w
2209 else lambda f
: True) # b*, w*
2210 filter_f
= lambda f
: _filter_f(f
) and (
2211 f
.get('vcodec') != 'none' or f
.get('acodec') != 'none')
2213 if format_spec
in self
._format
_selection
_exts
['audio']:
2214 filter_f
= lambda f
: f
.get('ext') == format_spec
and f
.get('acodec') != 'none'
2215 elif format_spec
in self
._format
_selection
_exts
['video']:
2216 filter_f
= lambda f
: f
.get('ext') == format_spec
and f
.get('acodec') != 'none' and f
.get('vcodec') != 'none'
2217 seperate_fallback
= lambda f
: f
.get('ext') == format_spec
and f
.get('vcodec') != 'none'
2218 elif format_spec
in self
._format
_selection
_exts
['storyboards']:
2219 filter_f
= lambda f
: f
.get('ext') == format_spec
and f
.get('acodec') == 'none' and f
.get('vcodec') == 'none'
2221 filter_f
= lambda f
: f
.get('format_id') == format_spec
# id
2223 def selector_function(ctx
):
2224 formats
= list(ctx
['formats'])
2225 matches
= list(filter(filter_f
, formats
)) if filter_f
is not None else formats
2227 if format_fallback
and ctx
['incomplete_formats']:
2228 # for extractors with incomplete formats (audio only (soundcloud)
2229 # or video only (imgur)) best/worst will fallback to
2230 # best/worst {video,audio}-only format
2232 elif seperate_fallback
and not ctx
['has_merged_format']:
2233 # for compatibility with youtube-dl when there is no pre-merged format
2234 matches
= list(filter(seperate_fallback
, formats
))
2235 matches
= LazyList(_check_formats(matches
[::-1 if format_reverse
else 1]))
2237 yield matches
[format_idx
- 1]
2238 except LazyList
.IndexError:
2241 filters
= [self
._build
_format
_filter
(f
) for f
in selector
.filters
]
2243 def final_selector(ctx
):
2244 ctx_copy
= dict(ctx
)
2245 for _filter
in filters
:
2246 ctx_copy
['formats'] = list(filter(_filter
, ctx_copy
['formats']))
2247 return selector_function(ctx_copy
)
2248 return final_selector
2250 stream
= io
.BytesIO(format_spec
.encode())
2252 tokens
= list(_remove_unused_ops(tokenize
.tokenize(stream
.readline
)))
2253 except tokenize
.TokenError
:
2254 raise syntax_error('Missing closing/opening brackets or parenthesis', (0, len(format_spec
)))
2256 class TokenIterator
:
2257 def __init__(self
, tokens
):
2258 self
.tokens
= tokens
2265 if self
.counter
>= len(self
.tokens
):
2266 raise StopIteration()
2267 value
= self
.tokens
[self
.counter
]
2273 def restore_last_token(self
):
2276 parsed_selector
= _parse_format_selection(iter(TokenIterator(tokens
)))
2277 return _build_selector_function(parsed_selector
)
2279 def _calc_headers(self
, info_dict
):
2280 res
= merge_headers(self
.params
['http_headers'], info_dict
.get('http_headers') or {})
2282 cookies
= self
._calc
_cookies
(info_dict
['url'])
2284 res
['Cookie'] = cookies
2286 if 'X-Forwarded-For' not in res
:
2287 x_forwarded_for_ip
= info_dict
.get('__x_forwarded_for_ip')
2288 if x_forwarded_for_ip
:
2289 res
['X-Forwarded-For'] = x_forwarded_for_ip
2293 def _calc_cookies(self
, url
):
2294 pr
= sanitized_Request(url
)
2295 self
.cookiejar
.add_cookie_header(pr
)
2296 return pr
.get_header('Cookie')
2298 def _sort_thumbnails(self
, thumbnails
):
2299 thumbnails
.sort(key
=lambda t
: (
2300 t
.get('preference') if t
.get('preference') is not None else -1,
2301 t
.get('width') if t
.get('width') is not None else -1,
2302 t
.get('height') if t
.get('height') is not None else -1,
2303 t
.get('id') if t
.get('id') is not None else '',
2306 def _sanitize_thumbnails(self
, info_dict
):
2307 thumbnails
= info_dict
.get('thumbnails')
2308 if thumbnails
is None:
2309 thumbnail
= info_dict
.get('thumbnail')
2311 info_dict
['thumbnails'] = thumbnails
= [{'url': thumbnail}
]
2315 def check_thumbnails(thumbnails
):
2316 for t
in thumbnails
:
2317 self
.to_screen(f
'[info] Testing thumbnail {t["id"]}')
2319 self
.urlopen(HEADRequest(t
['url']))
2320 except network_exceptions
as err
:
2321 self
.to_screen(f
'[info] Unable to connect to thumbnail {t["id"]} URL {t["url"]!r} - {err}. Skipping...')
2325 self
._sort
_thumbnails
(thumbnails
)
2326 for i
, t
in enumerate(thumbnails
):
2327 if t
.get('id') is None:
2329 if t
.get('width') and t
.get('height'):
2330 t
['resolution'] = '%dx%d' % (t
['width'], t
['height'])
2331 t
['url'] = sanitize_url(t
['url'])
2333 if self
.params
.get('check_formats') is True:
2334 info_dict
['thumbnails'] = LazyList(check_thumbnails(thumbnails
[::-1]), reverse
=True)
2336 info_dict
['thumbnails'] = thumbnails
2338 def _fill_common_fields(self
, info_dict
, is_video
=True):
2339 # TODO: move sanitization here
2341 # playlists are allowed to lack "title"
2342 title
= info_dict
.get('title', NO_DEFAULT
)
2343 if title
is NO_DEFAULT
:
2344 raise ExtractorError('Missing "title" field in extractor result',
2345 video_id
=info_dict
['id'], ie
=info_dict
['extractor'])
2346 info_dict
['fulltitle'] = title
2349 self
.write_debug('Extractor gave empty title. Creating a generic title')
2351 self
.report_warning('Extractor failed to obtain "title". Creating a generic title instead')
2352 info_dict
['title'] = f
'{info_dict["extractor"].replace(":", "-")} video #{info_dict["id"]}'
2354 if info_dict
.get('duration') is not None:
2355 info_dict
['duration_string'] = formatSeconds(info_dict
['duration'])
2357 for ts_key
, date_key
in (
2358 ('timestamp', 'upload_date'),
2359 ('release_timestamp', 'release_date'),
2360 ('modified_timestamp', 'modified_date'),
2362 if info_dict
.get(date_key
) is None and info_dict
.get(ts_key
) is not None:
2363 # Working around out-of-range timestamp values (e.g. negative ones on Windows,
2364 # see http://bugs.python.org/issue1646728)
2365 with contextlib
.suppress(ValueError, OverflowError, OSError):
2366 upload_date
= datetime
.datetime
.utcfromtimestamp(info_dict
[ts_key
])
2367 info_dict
[date_key
] = upload_date
.strftime('%Y%m%d')
2369 live_keys
= ('is_live', 'was_live')
2370 live_status
= info_dict
.get('live_status')
2371 if live_status
is None:
2372 for key
in live_keys
:
2373 if info_dict
.get(key
) is False:
2375 if info_dict
.get(key
):
2378 if all(info_dict
.get(key
) is False for key
in live_keys
):
2379 live_status
= 'not_live'
2381 info_dict
['live_status'] = live_status
2382 for key
in live_keys
:
2383 if info_dict
.get(key
) is None:
2384 info_dict
[key
] = (live_status
== key
)
2386 # Auto generate title fields corresponding to the *_number fields when missing
2387 # in order to always have clean titles. This is very common for TV series.
2388 for field
in ('chapter', 'season', 'episode'):
2389 if info_dict
.get('%s_number' % field
) is not None and not info_dict
.get(field
):
2390 info_dict
[field
] = '%s %d' % (field
.capitalize(), info_dict
['%s_number' % field
])
2392 def _raise_pending_errors(self
, info
):
2393 err
= info
.pop('__pending_error', None)
2395 self
.report_error(err
, tb
=False)
2397 def process_video_result(self
, info_dict
, download
=True):
2398 assert info_dict
.get('_type', 'video') == 'video'
2399 self
._num
_videos
+= 1
2401 if 'id' not in info_dict
:
2402 raise ExtractorError('Missing "id" field in extractor result', ie
=info_dict
['extractor'])
2403 elif not info_dict
.get('id'):
2404 raise ExtractorError('Extractor failed to obtain "id"', ie
=info_dict
['extractor'])
2406 def report_force_conversion(field
, field_not
, conversion
):
2407 self
.report_warning(
2408 '"%s" field is not %s - forcing %s conversion, there is an error in extractor'
2409 % (field
, field_not
, conversion
))
2411 def sanitize_string_field(info
, string_field
):
2412 field
= info
.get(string_field
)
2413 if field
is None or isinstance(field
, str):
2415 report_force_conversion(string_field
, 'a string', 'string')
2416 info
[string_field
] = str(field
)
2418 def sanitize_numeric_fields(info
):
2419 for numeric_field
in self
._NUMERIC
_FIELDS
:
2420 field
= info
.get(numeric_field
)
2421 if field
is None or isinstance(field
, (int, float)):
2423 report_force_conversion(numeric_field
, 'numeric', 'int')
2424 info
[numeric_field
] = int_or_none(field
)
2426 sanitize_string_field(info_dict
, 'id')
2427 sanitize_numeric_fields(info_dict
)
2428 if info_dict
.get('section_end') and info_dict
.get('section_start') is not None:
2429 info_dict
['duration'] = round(info_dict
['section_end'] - info_dict
['section_start'], 3)
2430 if (info_dict
.get('duration') or 0) <= 0 and info_dict
.pop('duration', None):
2431 self
.report_warning('"duration" field is negative, there is an error in extractor')
2433 chapters
= info_dict
.get('chapters') or []
2434 if chapters
and chapters
[0].get('start_time'):
2435 chapters
.insert(0, {'start_time': 0}
)
2437 dummy_chapter
= {'end_time': 0, 'start_time': info_dict.get('duration')}
2438 for idx
, (prev
, current
, next_
) in enumerate(zip(
2439 (dummy_chapter
, *chapters
), chapters
, (*chapters
[1:], dummy_chapter
)), 1):
2440 if current
.get('start_time') is None:
2441 current
['start_time'] = prev
.get('end_time')
2442 if not current
.get('end_time'):
2443 current
['end_time'] = next_
.get('start_time')
2444 if not current
.get('title'):
2445 current
['title'] = f
'<Untitled Chapter {idx}>'
2447 if 'playlist' not in info_dict
:
2448 # It isn't part of a playlist
2449 info_dict
['playlist'] = None
2450 info_dict
['playlist_index'] = None
2452 self
._sanitize
_thumbnails
(info_dict
)
2454 thumbnail
= info_dict
.get('thumbnail')
2455 thumbnails
= info_dict
.get('thumbnails')
2457 info_dict
['thumbnail'] = sanitize_url(thumbnail
)
2459 info_dict
['thumbnail'] = thumbnails
[-1]['url']
2461 if info_dict
.get('display_id') is None and 'id' in info_dict
:
2462 info_dict
['display_id'] = info_dict
['id']
2464 self
._fill
_common
_fields
(info_dict
)
2466 for cc_kind
in ('subtitles', 'automatic_captions'):
2467 cc
= info_dict
.get(cc_kind
)
2469 for _
, subtitle
in cc
.items():
2470 for subtitle_format
in subtitle
:
2471 if subtitle_format
.get('url'):
2472 subtitle_format
['url'] = sanitize_url(subtitle_format
['url'])
2473 if subtitle_format
.get('ext') is None:
2474 subtitle_format
['ext'] = determine_ext(subtitle_format
['url']).lower()
2476 automatic_captions
= info_dict
.get('automatic_captions')
2477 subtitles
= info_dict
.get('subtitles')
2479 info_dict
['requested_subtitles'] = self
.process_subtitles(
2480 info_dict
['id'], subtitles
, automatic_captions
)
2482 if info_dict
.get('formats') is None:
2483 # There's only one format available
2484 formats
= [info_dict
]
2486 formats
= info_dict
['formats']
2488 # or None ensures --clean-infojson removes it
2489 info_dict
['_has_drm'] = any(f
.get('has_drm') for f
in formats
) or None
2490 if not self
.params
.get('allow_unplayable_formats'):
2491 formats
= [f
for f
in formats
if not f
.get('has_drm')]
2492 if info_dict
['_has_drm'] and formats
and all(
2493 f
.get('acodec') == f
.get('vcodec') == 'none' for f
in formats
):
2494 self
.report_warning(
2495 'This video is DRM protected and only images are available for download. '
2496 'Use --list-formats to see them')
2498 get_from_start
= not info_dict
.get('is_live') or bool(self
.params
.get('live_from_start'))
2499 if not get_from_start
:
2500 info_dict
['title'] += ' ' + datetime
.datetime
.now().strftime('%Y-%m-%d %H:%M')
2501 if info_dict
.get('is_live') and formats
:
2502 formats
= [f
for f
in formats
if bool(f
.get('is_from_start')) == get_from_start
]
2503 if get_from_start
and not formats
:
2504 self
.raise_no_formats(info_dict
, msg
=(
2505 '--live-from-start is passed, but there are no formats that can be downloaded from the start. '
2506 'If you want to download from the current time, use --no-live-from-start'))
2509 self
.raise_no_formats(info_dict
)
2511 def is_wellformed(f
):
2514 self
.report_warning(
2515 '"url" field is missing or empty - skipping format, '
2516 'there is an error in extractor')
2518 if isinstance(url
, bytes):
2519 sanitize_string_field(f
, 'url')
2522 # Filter out malformed formats for better extraction robustness
2523 formats
= list(filter(is_wellformed
, formats
))
2527 # We check that all the formats have the format and format_id fields
2528 for i
, format
in enumerate(formats
):
2529 sanitize_string_field(format
, 'format_id')
2530 sanitize_numeric_fields(format
)
2531 format
['url'] = sanitize_url(format
['url'])
2532 if not format
.get('format_id'):
2533 format
['format_id'] = str(i
)
2535 # Sanitize format_id from characters used in format selector expression
2536 format
['format_id'] = re
.sub(r
'[\s,/+\[\]()]', '_', format
['format_id'])
2537 format_id
= format
['format_id']
2538 if format_id
not in formats_dict
:
2539 formats_dict
[format_id
] = []
2540 formats_dict
[format_id
].append(format
)
2542 # Make sure all formats have unique format_id
2543 common_exts
= set(itertools
.chain(*self
._format
_selection
_exts
.values()))
2544 for format_id
, ambiguous_formats
in formats_dict
.items():
2545 ambigious_id
= len(ambiguous_formats
) > 1
2546 for i
, format
in enumerate(ambiguous_formats
):
2548 format
['format_id'] = '%s-%d' % (format_id
, i
)
2549 if format
.get('ext') is None:
2550 format
['ext'] = determine_ext(format
['url']).lower()
2551 # Ensure there is no conflict between id and ext in format selection
2552 # See https://github.com/yt-dlp/yt-dlp/issues/1282
2553 if format
['format_id'] != format
['ext'] and format
['format_id'] in common_exts
:
2554 format
['format_id'] = 'f%s' % format
['format_id']
2556 for i
, format
in enumerate(formats
):
2557 if format
.get('format') is None:
2558 format
['format'] = '{id} - {res}{note}'.format(
2559 id=format
['format_id'],
2560 res
=self
.format_resolution(format
),
2561 note
=format_field(format
, 'format_note', ' (%s)'),
2563 if format
.get('protocol') is None:
2564 format
['protocol'] = determine_protocol(format
)
2565 if format
.get('resolution') is None:
2566 format
['resolution'] = self
.format_resolution(format
, default
=None)
2567 if format
.get('dynamic_range') is None and format
.get('vcodec') != 'none':
2568 format
['dynamic_range'] = 'SDR'
2569 if (info_dict
.get('duration') and format
.get('tbr')
2570 and not format
.get('filesize') and not format
.get('filesize_approx')):
2571 format
['filesize_approx'] = int(info_dict
['duration'] * format
['tbr'] * (1024 / 8))
2573 # Add HTTP headers, so that external programs can use them from the
2575 full_format_info
= info_dict
.copy()
2576 full_format_info
.update(format
)
2577 format
['http_headers'] = self
._calc
_headers
(full_format_info
)
2578 # Remove private housekeeping stuff
2579 if '__x_forwarded_for_ip' in info_dict
:
2580 del info_dict
['__x_forwarded_for_ip']
2582 if self
.params
.get('check_formats') is True:
2583 formats
= LazyList(self
._check
_formats
(formats
[::-1]), reverse
=True)
2585 if not formats
or formats
[0] is not info_dict
:
2586 # only set the 'formats' fields if the original info_dict list them
2587 # otherwise we end up with a circular reference, the first (and unique)
2588 # element in the 'formats' field in info_dict is info_dict itself,
2589 # which can't be exported to json
2590 info_dict
['formats'] = formats
2592 info_dict
, _
= self
.pre_process(info_dict
)
2594 if self
._match
_entry
(info_dict
, incomplete
=self
._format
_fields
) is not None:
2597 self
.post_extract(info_dict
)
2598 info_dict
, _
= self
.pre_process(info_dict
, 'after_filter')
2600 # The pre-processors may have modified the formats
2601 formats
= info_dict
.get('formats', [info_dict
])
2603 list_only
= self
.params
.get('simulate') is None and (
2604 self
.params
.get('list_thumbnails') or self
.params
.get('listformats') or self
.params
.get('listsubtitles'))
2605 interactive_format_selection
= not list_only
and self
.format_selector
== '-'
2606 if self
.params
.get('list_thumbnails'):
2607 self
.list_thumbnails(info_dict
)
2608 if self
.params
.get('listsubtitles'):
2609 if 'automatic_captions' in info_dict
:
2610 self
.list_subtitles(
2611 info_dict
['id'], automatic_captions
, 'automatic captions')
2612 self
.list_subtitles(info_dict
['id'], subtitles
, 'subtitles')
2613 if self
.params
.get('listformats') or interactive_format_selection
:
2614 self
.list_formats(info_dict
)
2616 # Without this printing, -F --print-json will not work
2617 self
.__forced
_printings
(info_dict
, self
.prepare_filename(info_dict
), incomplete
=True)
2620 format_selector
= self
.format_selector
2621 if format_selector
is None:
2622 req_format
= self
._default
_format
_spec
(info_dict
, download
=download
)
2623 self
.write_debug('Default format spec: %s' % req_format
)
2624 format_selector
= self
.build_format_selector(req_format
)
2627 if interactive_format_selection
:
2629 self
._format
_screen
('\nEnter format selector: ', self
.Styles
.EMPHASIS
))
2631 format_selector
= self
.build_format_selector(req_format
)
2632 except SyntaxError as err
:
2633 self
.report_error(err
, tb
=False, is_error
=False)
2636 formats_to_download
= list(format_selector({
2638 'has_merged_format': any('none' not in (f
.get('acodec'), f
.get('vcodec')) for f
in formats
),
2639 'incomplete_formats': (
2640 # All formats are video-only or
2641 all(f
.get('vcodec') != 'none' and f
.get('acodec') == 'none' for f
in formats
)
2642 # all formats are audio-only
2643 or all(f
.get('vcodec') == 'none' and f
.get('acodec') != 'none' for f
in formats
)),
2645 if interactive_format_selection
and not formats_to_download
:
2646 self
.report_error('Requested format is not available', tb
=False, is_error
=False)
2650 if not formats_to_download
:
2651 if not self
.params
.get('ignore_no_formats_error'):
2652 raise ExtractorError(
2653 'Requested format is not available. Use --list-formats for a list of available formats',
2654 expected
=True, video_id
=info_dict
['id'], ie
=info_dict
['extractor'])
2655 self
.report_warning('Requested format is not available')
2656 # Process what we can, even without any available formats.
2657 formats_to_download
= [{}]
2659 requested_ranges
= self
.params
.get('download_ranges')
2660 if requested_ranges
:
2661 requested_ranges
= tuple(requested_ranges(info_dict
, self
))
2663 best_format
, downloaded_formats
= formats_to_download
[-1], []
2666 def to_screen(*msg
):
2667 self
.to_screen(f
'[info] {info_dict["id"]}: {" ".join(", ".join(variadic(m)) for m in msg)}')
2669 to_screen(f
'Downloading {len(formats_to_download)} format(s):',
2670 (f
['format_id'] for f
in formats_to_download
))
2671 if requested_ranges
:
2672 to_screen(f
'Downloading {len(requested_ranges)} time ranges:',
2673 (f
'{int(c["start_time"])}-{int(c["end_time"])}' for c
in requested_ranges
))
2674 max_downloads_reached
= False
2676 for fmt
, chapter
in itertools
.product(formats_to_download
, requested_ranges
or [{}]):
2677 new_info
= self
._copy
_infodict
(info_dict
)
2678 new_info
.update(fmt
)
2679 offset
, duration
= info_dict
.get('section_start') or 0, info_dict
.get('duration') or float('inf')
2680 if chapter
or offset
:
2682 'section_start': offset
+ chapter
.get('start_time', 0),
2683 'section_end': offset
+ min(chapter
.get('end_time', duration
), duration
),
2684 'section_title': chapter
.get('title'),
2685 'section_number': chapter
.get('index'),
2687 downloaded_formats
.append(new_info
)
2689 self
.process_info(new_info
)
2690 except MaxDownloadsReached
:
2691 max_downloads_reached
= True
2692 self
._raise
_pending
_errors
(new_info
)
2693 # Remove copied info
2694 for key
, val
in tuple(new_info
.items()):
2695 if info_dict
.get(key
) == val
:
2697 if max_downloads_reached
:
2700 write_archive
= {f.get('__write_download_archive', False) for f in downloaded_formats}
2701 assert write_archive
.issubset({True, False, 'ignore'}
)
2702 if True in write_archive
and False not in write_archive
:
2703 self
.record_download_archive(info_dict
)
2705 info_dict
['requested_downloads'] = downloaded_formats
2706 info_dict
= self
.run_all_pps('after_video', info_dict
)
2707 if max_downloads_reached
:
2708 raise MaxDownloadsReached()
2710 # We update the info dict with the selected best quality format (backwards compatibility)
2711 info_dict
.update(best_format
)
2714 def process_subtitles(self
, video_id
, normal_subtitles
, automatic_captions
):
2715 """Select the requested subtitles and their format"""
2716 available_subs
, normal_sub_langs
= {}, []
2717 if normal_subtitles
and self
.params
.get('writesubtitles'):
2718 available_subs
.update(normal_subtitles
)
2719 normal_sub_langs
= tuple(normal_subtitles
.keys())
2720 if automatic_captions
and self
.params
.get('writeautomaticsub'):
2721 for lang
, cap_info
in automatic_captions
.items():
2722 if lang
not in available_subs
:
2723 available_subs
[lang
] = cap_info
2725 if (not self
.params
.get('writesubtitles') and not
2726 self
.params
.get('writeautomaticsub') or not
2730 all_sub_langs
= tuple(available_subs
.keys())
2731 if self
.params
.get('allsubtitles', False):
2732 requested_langs
= all_sub_langs
2733 elif self
.params
.get('subtitleslangs', False):
2734 # A list is used so that the order of languages will be the same as
2735 # given in subtitleslangs. See https://github.com/yt-dlp/yt-dlp/issues/1041
2736 requested_langs
= []
2737 for lang_re
in self
.params
.get('subtitleslangs'):
2738 discard
= lang_re
[0] == '-'
2740 lang_re
= lang_re
[1:]
2741 if lang_re
== 'all':
2743 requested_langs
= []
2745 requested_langs
.extend(all_sub_langs
)
2747 current_langs
= filter(re
.compile(lang_re
+ '$').match
, all_sub_langs
)
2749 for lang
in current_langs
:
2750 while lang
in requested_langs
:
2751 requested_langs
.remove(lang
)
2753 requested_langs
.extend(current_langs
)
2754 requested_langs
= orderedSet(requested_langs
)
2755 elif normal_sub_langs
:
2756 requested_langs
= ['en'] if 'en' in normal_sub_langs
else normal_sub_langs
[:1]
2758 requested_langs
= ['en'] if 'en' in all_sub_langs
else all_sub_langs
[:1]
2760 self
.write_debug('Downloading subtitles: %s' % ', '.join(requested_langs
))
2762 formats_query
= self
.params
.get('subtitlesformat', 'best')
2763 formats_preference
= formats_query
.split('/') if formats_query
else []
2765 for lang
in requested_langs
:
2766 formats
= available_subs
.get(lang
)
2768 self
.report_warning(f
'{lang} subtitles not available for {video_id}')
2770 for ext
in formats_preference
:
2774 matches
= list(filter(lambda f
: f
['ext'] == ext
, formats
))
2780 self
.report_warning(
2781 'No subtitle format found matching "%s" for language %s, '
2782 'using %s' % (formats_query
, lang
, f
['ext']))
2786 def _forceprint(self
, key
, info_dict
):
2787 if info_dict
is None:
2789 info_copy
= info_dict
.copy()
2790 info_copy
['formats_table'] = self
.render_formats_table(info_dict
)
2791 info_copy
['thumbnails_table'] = self
.render_thumbnails_table(info_dict
)
2792 info_copy
['subtitles_table'] = self
.render_subtitles_table(info_dict
.get('id'), info_dict
.get('subtitles'))
2793 info_copy
['automatic_captions_table'] = self
.render_subtitles_table(info_dict
.get('id'), info_dict
.get('automatic_captions'))
2795 def format_tmpl(tmpl
):
2796 mobj
= re
.match(r
'\w+(=?)$', 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', []))
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', ''),
3573 format_field(f
, 'filesize', ' \t%s', func
=format_bytes
) + format_field(f
, 'filesize_approx', '~\t%s', func
=format_bytes
),
3574 format_field(f
, 'tbr', '\t%dk', func
=round),
3575 shorten_protocol_name(f
.get('protocol', '')),
3577 simplified_codec(f
, 'vcodec'),
3578 format_field(f
, 'vbr', '\t%dk', func
=round),
3579 simplified_codec(f
, 'acodec'),
3580 format_field(f
, 'abr', '\t%dk', func
=round),
3581 format_field(f
, 'asr', '\t%s', func
=format_decimal_suffix
),
3583 self
._format
_out
('UNSUPPORTED', 'light red') if f
.get('ext') in ('f4f', 'f4m') else None,
3584 format_field(f
, 'language', '[%s]'),
3585 join_nonempty(format_field(f
, 'format_note'),
3586 format_field(f
, 'container', ignore
=(None, f
.get('ext'))),
3589 ] for f
in formats
if f
.get('preference') is None or f
['preference'] >= -1000]
3590 header_line
= self
._list
_format
_headers
(
3591 'ID', 'EXT', 'RESOLUTION', '\tFPS', 'HDR', delim
, '\tFILESIZE', '\tTBR', 'PROTO',
3592 delim
, 'VCODEC', '\tVBR', 'ACODEC', '\tABR', '\tASR', 'MORE INFO')
3594 return render_table(
3595 header_line
, table
, hide_empty
=True,
3596 delim
=self
._format
_out
('\u2500', self
.Styles
.DELIM
, '-', test_encoding
=True))
3598 def render_thumbnails_table(self
, info_dict
):
3599 thumbnails
= list(info_dict
.get('thumbnails') or [])
3602 return render_table(
3603 self
._list
_format
_headers
('ID', 'Width', 'Height', 'URL'),
3604 [[t
.get('id'), t
.get('width', 'unknown'), t
.get('height', 'unknown'), t
['url']] for t
in thumbnails
])
3606 def render_subtitles_table(self
, video_id
, subtitles
):
3607 def _row(lang
, formats
):
3608 exts
, names
= zip(*((f
['ext'], f
.get('name') or 'unknown') for f
in reversed(formats
)))
3609 if len(set(names
)) == 1:
3610 names
= [] if names
[0] == 'unknown' else names
[:1]
3611 return [lang
, ', '.join(names
), ', '.join(exts
)]
3615 return render_table(
3616 self
._list
_format
_headers
('Language', 'Name', 'Formats'),
3617 [_row(lang
, formats
) for lang
, formats
in subtitles
.items()],
3620 def __list_table(self
, video_id
, name
, func
, *args
):
3623 self
.to_screen(f
'{video_id} has no {name}')
3625 self
.to_screen(f
'[info] Available {name} for {video_id}:')
3626 self
.to_stdout(table
)
3628 def list_formats(self
, info_dict
):
3629 self
.__list
_table
(info_dict
['id'], 'formats', self
.render_formats_table
, info_dict
)
3631 def list_thumbnails(self
, info_dict
):
3632 self
.__list
_table
(info_dict
['id'], 'thumbnails', self
.render_thumbnails_table
, info_dict
)
3634 def list_subtitles(self
, video_id
, subtitles
, name
='subtitles'):
3635 self
.__list
_table
(video_id
, name
, self
.render_subtitles_table
, video_id
, subtitles
)
3637 def urlopen(self
, req
):
3638 """ Start an HTTP download """
3639 if isinstance(req
, str):
3640 req
= sanitized_Request(req
)
3641 return self
._opener
.open(req
, timeout
=self
._socket
_timeout
)
3643 def print_debug_header(self
):
3644 if not self
.params
.get('verbose'):
3647 # These imports can be slow. So import them only as needed
3648 from .extractor
.extractors
import _LAZY_LOADER
3649 from .extractor
.extractors
import _PLUGIN_CLASSES
as plugin_extractors
3651 def get_encoding(stream
):
3652 ret
= str(getattr(stream
, 'encoding', 'missing (%s)' % type(stream
).__name
__))
3653 if not supports_terminal_sequences(stream
):
3654 from .utils
import WINDOWS_VT_MODE
# Must be imported locally
3655 ret
+= ' (No VT)' if WINDOWS_VT_MODE
is False else ' (No ANSI)'
3658 encoding_str
= 'Encodings: locale %s, fs %s, pref %s, %s' % (
3659 locale
.getpreferredencoding(),
3660 sys
.getfilesystemencoding(),
3661 self
.get_encoding(),
3663 f
'{key} {get_encoding(stream)}' for key
, stream
in self
._out
_files
.items_
3664 if stream
is not None and key
!= 'console')
3667 logger
= self
.params
.get('logger')
3669 write_debug
= lambda msg
: logger
.debug(f
'[debug] {msg}')
3670 write_debug(encoding_str
)
3672 write_string(f
'[debug] {encoding_str}\n', encoding
=None)
3673 write_debug
= lambda msg
: self
._write
_string
(f
'[debug] {msg}\n')
3675 source
= detect_variant()
3676 write_debug(join_nonempty(
3677 'yt-dlp version', __version__
,
3678 f
'[{RELEASE_GIT_HEAD}]' if RELEASE_GIT_HEAD
else '',
3679 '' if source
== 'unknown' else f
'({source})',
3681 if not _LAZY_LOADER
:
3682 if os
.environ
.get('YTDLP_NO_LAZY_EXTRACTORS'):
3683 write_debug('Lazy loading extractors is forcibly disabled')
3685 write_debug('Lazy loading extractors is disabled')
3686 if plugin_extractors
or plugin_postprocessors
:
3687 write_debug('Plugins: %s' % [
3688 '%s%s' % (klass
.__name
__, '' if klass
.__name
__ == name
else f
' as {name}')
3689 for name
, klass
in itertools
.chain(plugin_extractors
.items(), plugin_postprocessors
.items())])
3690 if self
.params
['compat_opts']:
3691 write_debug('Compatibility options: %s' % ', '.join(self
.params
['compat_opts']))
3693 if source
== 'source':
3695 stdout
, _
, _
= Popen
.run(
3696 ['git', 'rev-parse', '--short', 'HEAD'],
3697 text
=True, cwd
=os
.path
.dirname(os
.path
.abspath(__file__
)),
3698 stdout
=subprocess
.PIPE
, stderr
=subprocess
.PIPE
)
3699 if re
.fullmatch('[0-9a-f]+', stdout
.strip()):
3700 write_debug(f
'Git HEAD: {stdout.strip()}')
3702 with contextlib
.suppress(Exception):
3705 write_debug(system_identifier())
3707 exe_versions
, ffmpeg_features
= FFmpegPostProcessor
.get_versions_and_features(self
)
3708 ffmpeg_features
= {key for key, val in ffmpeg_features.items() if val}
3710 exe_versions
['ffmpeg'] += ' (%s)' % ','.join(sorted(ffmpeg_features
))
3712 exe_versions
['rtmpdump'] = rtmpdump_version()
3713 exe_versions
['phantomjs'] = PhantomJSwrapper
._version
()
3714 exe_str
= ', '.join(
3715 f
'{exe} {v}' for exe
, v
in sorted(exe_versions
.items()) if v
3717 write_debug('exe versions: %s' % exe_str
)
3719 from .compat
.compat_utils
import get_package_info
3720 from .dependencies
import available_dependencies
3722 write_debug('Optional libraries: %s' % (', '.join(sorted({
3723 join_nonempty(*get_package_info(m
)) for m
in available_dependencies
.values()
3726 self
._setup
_opener
()
3728 for handler
in self
._opener
.handlers
:
3729 if hasattr(handler
, 'proxies'):
3730 proxy_map
.update(handler
.proxies
)
3731 write_debug(f
'Proxy map: {proxy_map}')
3734 if False and self
.params
.get('call_home'):
3735 ipaddr
= self
.urlopen('https://yt-dl.org/ip').read().decode()
3736 write_debug('Public IP address: %s' % ipaddr
)
3737 latest_version
= self
.urlopen(
3738 'https://yt-dl.org/latest/version').read().decode()
3739 if version_tuple(latest_version
) > version_tuple(__version__
):
3740 self
.report_warning(
3741 'You are using an outdated version (newest version: %s)! '
3742 'See https://yt-dl.org/update if you need help updating.' %
3745 def _setup_opener(self
):
3746 if hasattr(self
, '_opener'):
3748 timeout_val
= self
.params
.get('socket_timeout')
3749 self
._socket
_timeout
= 20 if timeout_val
is None else float(timeout_val
)
3751 opts_cookiesfrombrowser
= self
.params
.get('cookiesfrombrowser')
3752 opts_cookiefile
= self
.params
.get('cookiefile')
3753 opts_proxy
= self
.params
.get('proxy')
3755 self
.cookiejar
= load_cookies(opts_cookiefile
, opts_cookiesfrombrowser
, self
)
3757 cookie_processor
= YoutubeDLCookieProcessor(self
.cookiejar
)
3758 if opts_proxy
is not None:
3759 if opts_proxy
== '':
3762 proxies
= {'http': opts_proxy, 'https': opts_proxy}
3764 proxies
= urllib
.request
.getproxies()
3765 # Set HTTPS proxy to HTTP one if given (https://github.com/ytdl-org/youtube-dl/issues/805)
3766 if 'http' in proxies
and 'https' not in proxies
:
3767 proxies
['https'] = proxies
['http']
3768 proxy_handler
= PerRequestProxyHandler(proxies
)
3770 debuglevel
= 1 if self
.params
.get('debug_printtraffic') else 0
3771 https_handler
= make_HTTPS_handler(self
.params
, debuglevel
=debuglevel
)
3772 ydlh
= YoutubeDLHandler(self
.params
, debuglevel
=debuglevel
)
3773 redirect_handler
= YoutubeDLRedirectHandler()
3774 data_handler
= urllib
.request
.DataHandler()
3776 # When passing our own FileHandler instance, build_opener won't add the
3777 # default FileHandler and allows us to disable the file protocol, which
3778 # can be used for malicious purposes (see
3779 # https://github.com/ytdl-org/youtube-dl/issues/8227)
3780 file_handler
= urllib
.request
.FileHandler()
3782 def file_open(*args
, **kwargs
):
3783 raise urllib
.error
.URLError('file:// scheme is explicitly disabled in yt-dlp for security reasons')
3784 file_handler
.file_open
= file_open
3786 opener
= urllib
.request
.build_opener(
3787 proxy_handler
, https_handler
, cookie_processor
, ydlh
, redirect_handler
, data_handler
, file_handler
)
3789 # Delete the default user-agent header, which would otherwise apply in
3790 # cases where our custom HTTP handler doesn't come into play
3791 # (See https://github.com/ytdl-org/youtube-dl/issues/1309 for details)
3792 opener
.addheaders
= []
3793 self
._opener
= opener
3795 def encode(self
, s
):
3796 if isinstance(s
, bytes):
3797 return s
# Already encoded
3800 return s
.encode(self
.get_encoding())
3801 except UnicodeEncodeError as err
:
3802 err
.reason
= err
.reason
+ '. Check your system encoding configuration or use the --encoding option.'
3805 def get_encoding(self
):
3806 encoding
= self
.params
.get('encoding')
3807 if encoding
is None:
3808 encoding
= preferredencoding()
3811 def _write_info_json(self
, label
, ie_result
, infofn
, overwrite
=None):
3812 ''' Write infojson and returns True = written, 'exists' = Already exists, False = skip, None = error '''
3813 if overwrite
is None:
3814 overwrite
= self
.params
.get('overwrites', True)
3815 if not self
.params
.get('writeinfojson'):
3818 self
.write_debug(f
'Skipping writing {label} infojson')
3820 elif not self
._ensure
_dir
_exists
(infofn
):
3822 elif not overwrite
and os
.path
.exists(infofn
):
3823 self
.to_screen(f
'[info] {label.title()} metadata is already present')
3826 self
.to_screen(f
'[info] Writing {label} metadata as JSON to: {infofn}')
3828 write_json_file(self
.sanitize_info(ie_result
, self
.params
.get('clean_infojson', True)), infofn
)
3831 self
.report_error(f
'Cannot write {label} metadata to JSON file {infofn}')
3834 def _write_description(self
, label
, ie_result
, descfn
):
3835 ''' Write description and returns True = written, False = skip, None = error '''
3836 if not self
.params
.get('writedescription'):
3839 self
.write_debug(f
'Skipping writing {label} description')
3841 elif not self
._ensure
_dir
_exists
(descfn
):
3843 elif not self
.params
.get('overwrites', True) and os
.path
.exists(descfn
):
3844 self
.to_screen(f
'[info] {label.title()} description is already present')
3845 elif ie_result
.get('description') is None:
3846 self
.report_warning(f
'There\'s no {label} description to write')
3850 self
.to_screen(f
'[info] Writing {label} description to: {descfn}')
3851 with open(encodeFilename(descfn
), 'w', encoding
='utf-8') as descfile
:
3852 descfile
.write(ie_result
['description'])
3854 self
.report_error(f
'Cannot write {label} description file {descfn}')
3858 def _write_subtitles(self
, info_dict
, filename
):
3859 ''' Write subtitles to file and return list of (sub_filename, final_sub_filename); or None if error'''
3861 subtitles
= info_dict
.get('requested_subtitles')
3862 if not subtitles
or not (self
.params
.get('writesubtitles') or self
.params
.get('writeautomaticsub')):
3863 # subtitles download errors are already managed as troubles in relevant IE
3864 # that way it will silently go on when used with unsupporting IE
3867 sub_filename_base
= self
.prepare_filename(info_dict
, 'subtitle')
3868 if not sub_filename_base
:
3869 self
.to_screen('[info] Skipping writing video subtitles')
3871 for sub_lang
, sub_info
in subtitles
.items():
3872 sub_format
= sub_info
['ext']
3873 sub_filename
= subtitles_filename(filename
, sub_lang
, sub_format
, info_dict
.get('ext'))
3874 sub_filename_final
= subtitles_filename(sub_filename_base
, sub_lang
, sub_format
, info_dict
.get('ext'))
3875 existing_sub
= self
.existing_file((sub_filename_final
, sub_filename
))
3877 self
.to_screen(f
'[info] Video subtitle {sub_lang}.{sub_format} is already present')
3878 sub_info
['filepath'] = existing_sub
3879 ret
.append((existing_sub
, sub_filename_final
))
3882 self
.to_screen(f
'[info] Writing video subtitles to: {sub_filename}')
3883 if sub_info
.get('data') is not None:
3885 # Use newline='' to prevent conversion of newline characters
3886 # See https://github.com/ytdl-org/youtube-dl/issues/10268
3887 with open(sub_filename
, 'w', encoding
='utf-8', newline
='') as subfile
:
3888 subfile
.write(sub_info
['data'])
3889 sub_info
['filepath'] = sub_filename
3890 ret
.append((sub_filename
, sub_filename_final
))
3893 self
.report_error(f
'Cannot write video subtitles file {sub_filename}')
3897 sub_copy
= sub_info
.copy()
3898 sub_copy
.setdefault('http_headers', info_dict
.get('http_headers'))
3899 self
.dl(sub_filename
, sub_copy
, subtitle
=True)
3900 sub_info
['filepath'] = sub_filename
3901 ret
.append((sub_filename
, sub_filename_final
))
3902 except (DownloadError
, ExtractorError
, IOError, OSError, ValueError) + network_exceptions
as err
:
3903 msg
= f
'Unable to download video subtitles for {sub_lang!r}: {err}'
3904 if self
.params
.get('ignoreerrors') is not True: # False or 'only_download'
3905 if not self
.params
.get('ignoreerrors'):
3906 self
.report_error(msg
)
3907 raise DownloadError(msg
)
3908 self
.report_warning(msg
)
3911 def _write_thumbnails(self
, label
, info_dict
, filename
, thumb_filename_base
=None):
3912 ''' Write thumbnails to file and return list of (thumb_filename, final_thumb_filename) '''
3913 write_all
= self
.params
.get('write_all_thumbnails', False)
3914 thumbnails
, ret
= [], []
3915 if write_all
or self
.params
.get('writethumbnail', False):
3916 thumbnails
= info_dict
.get('thumbnails') or []
3917 multiple
= write_all
and len(thumbnails
) > 1
3919 if thumb_filename_base
is None:
3920 thumb_filename_base
= filename
3921 if thumbnails
and not thumb_filename_base
:
3922 self
.write_debug(f
'Skipping writing {label} thumbnail')
3925 for idx
, t
in list(enumerate(thumbnails
))[::-1]:
3926 thumb_ext
= (f
'{t["id"]}.' if multiple
else '') + determine_ext(t
['url'], 'jpg')
3927 thumb_display_id
= f
'{label} thumbnail {t["id"]}'
3928 thumb_filename
= replace_extension(filename
, thumb_ext
, info_dict
.get('ext'))
3929 thumb_filename_final
= replace_extension(thumb_filename_base
, thumb_ext
, info_dict
.get('ext'))
3931 existing_thumb
= self
.existing_file((thumb_filename_final
, thumb_filename
))
3933 self
.to_screen('[info] %s is already present' % (
3934 thumb_display_id
if multiple
else f
'{label} thumbnail').capitalize())
3935 t
['filepath'] = existing_thumb
3936 ret
.append((existing_thumb
, thumb_filename_final
))
3938 self
.to_screen(f
'[info] Downloading {thumb_display_id} ...')
3940 uf
= self
.urlopen(sanitized_Request(t
['url'], headers
=t
.get('http_headers', {})))
3941 self
.to_screen(f
'[info] Writing {thumb_display_id} to: {thumb_filename}')
3942 with open(encodeFilename(thumb_filename
), 'wb') as thumbf
:
3943 shutil
.copyfileobj(uf
, thumbf
)
3944 ret
.append((thumb_filename
, thumb_filename_final
))
3945 t
['filepath'] = thumb_filename
3946 except network_exceptions
as err
:
3948 self
.report_warning(f
'Unable to download {thumb_display_id}: {err}')
3949 if ret
and not write_all
: