24 from string
import ascii_letters
26 from .cache
import Cache
27 from .compat
import HAS_LEGACY
as compat_has_legacy
28 from .compat
import compat_os_name
, compat_shlex_quote
29 from .cookies
import load_cookies
30 from .downloader
import FFmpegFD
, get_suitable_downloader
, shorten_protocol_name
31 from .downloader
.rtmp
import rtmpdump_version
32 from .extractor
import gen_extractor_classes
, get_info_extractor
33 from .extractor
.openload
import PhantomJSwrapper
34 from .minicurses
import format_text
35 from .postprocessor
import _PLUGIN_CLASSES
as plugin_postprocessors
36 from .postprocessor
import (
38 FFmpegFixupDuplicateMoovPP
,
39 FFmpegFixupDurationPP
,
42 FFmpegFixupStretchedPP
,
43 FFmpegFixupTimestampPP
,
46 FFmpegVideoConvertorPP
,
47 MoveFilesAfterDownloadPP
,
50 from .postprocessor
.ffmpeg
import resolve_mapping
as resolve_recode_mapping
51 from .update
import detect_variant
76 PerRequestProxyHandler
,
83 UnavailableVideoError
,
84 YoutubeDLCookieProcessor
,
86 YoutubeDLRedirectHandler
,
100 format_decimal_suffix
,
117 register_socks_protocols
,
118 remove_terminal_sequences
,
129 supports_terminal_sequences
,
138 windows_enable_vt_mode
,
142 from .version
import RELEASE_GIT_HEAD
, __version__
144 if compat_os_name
== 'nt':
151 YoutubeDL objects are the ones responsible of downloading the
152 actual video file and writing it to disk if the user has requested
153 it, among some other tasks. In most cases there should be one per
154 program. As, given a video URL, the downloader doesn't know how to
155 extract all the needed information, task that InfoExtractors do, it
156 has to pass the URL to one of them.
158 For this, YoutubeDL objects have a method that allows
159 InfoExtractors to be registered in a given order. When it is passed
160 a URL, the YoutubeDL object handles it to the first InfoExtractor it
161 finds that reports being able to handle it. The InfoExtractor extracts
162 all the information about the video or videos the URL refers to, and
163 YoutubeDL process the extracted information, possibly using a File
164 Downloader to download the video.
166 YoutubeDL objects accept a lot of parameters. In order not to saturate
167 the object constructor with arguments, it receives a dictionary of
168 options instead. These options are available through the params
169 attribute for the InfoExtractors to use. The YoutubeDL also
170 registers itself as the downloader in charge for the InfoExtractors
171 that are added to it, so this is a "mutual registration".
175 username: Username for authentication purposes.
176 password: Password for authentication purposes.
177 videopassword: Password for accessing a video.
178 ap_mso: Adobe Pass multiple-system operator identifier.
179 ap_username: Multiple-system operator account username.
180 ap_password: Multiple-system operator account password.
181 usenetrc: Use netrc for authentication instead.
182 verbose: Print additional info to stdout.
183 quiet: Do not print messages to stdout.
184 no_warnings: Do not print out anything for warnings.
185 forceprint: A dict with keys WHEN mapped to a list of templates to
186 print to stdout. The allowed keys are video or any of the
187 items in utils.POSTPROCESS_WHEN.
188 For compatibility, a single list is also accepted
189 print_to_file: A dict with keys WHEN (same as forceprint) mapped to
190 a list of tuples with (template, filename)
191 forcejson: Force printing info_dict as JSON.
192 dump_single_json: Force printing the info_dict of the whole playlist
193 (or video) as a single JSON line.
194 force_write_download_archive: Force writing download archive regardless
195 of 'skip_download' or 'simulate'.
196 simulate: Do not download the video files. If unset (or None),
197 simulate only if listsubtitles, listformats or list_thumbnails is used
198 format: Video format code. see "FORMAT SELECTION" for more details.
199 You can also pass a function. The function takes 'ctx' as
200 argument and returns the formats to download.
201 See "build_format_selector" for an implementation
202 allow_unplayable_formats: Allow unplayable formats to be extracted and downloaded.
203 ignore_no_formats_error: Ignore "No video formats" error. Usefull for
204 extracting metadata even if the video is not actually
205 available for download (experimental)
206 format_sort: A list of fields by which to sort the video formats.
207 See "Sorting Formats" for more details.
208 format_sort_force: Force the given format_sort. see "Sorting Formats"
210 prefer_free_formats: Whether to prefer video formats with free containers
211 over non-free ones of same quality.
212 allow_multiple_video_streams: Allow multiple video streams to be merged
214 allow_multiple_audio_streams: Allow multiple audio streams to be merged
216 check_formats Whether to test if the formats are downloadable.
217 Can be True (check all), False (check none),
218 'selected' (check selected formats),
219 or None (check only if requested by extractor)
220 paths: Dictionary of output paths. The allowed keys are 'home'
221 'temp' and the keys of OUTTMPL_TYPES (in utils.py)
222 outtmpl: Dictionary of templates for output names. Allowed keys
223 are 'default' and the keys of OUTTMPL_TYPES (in utils.py).
224 For compatibility with youtube-dl, a single string can also be used
225 outtmpl_na_placeholder: Placeholder for unavailable meta fields.
226 restrictfilenames: Do not allow "&" and spaces in file names
227 trim_file_name: Limit length of filename (extension excluded)
228 windowsfilenames: Force the filenames to be windows compatible
229 ignoreerrors: Do not stop on download/postprocessing errors.
230 Can be 'only_download' to ignore only download errors.
231 Default is 'only_download' for CLI, but False for API
232 skip_playlist_after_errors: Number of allowed failures until the rest of
233 the playlist is skipped
234 force_generic_extractor: Force downloader to use the generic extractor
235 overwrites: Overwrite all video and metadata files if True,
236 overwrite only non-video files if None
237 and don't overwrite any file if False
238 For compatibility with youtube-dl,
239 "nooverwrites" may also be used instead
240 playlist_items: Specific indices of playlist to download.
241 playlistrandom: Download playlist items in random order.
242 lazy_playlist: Process playlist entries as they are received.
243 matchtitle: Download only matching titles.
244 rejecttitle: Reject downloads for matching titles.
245 logger: Log messages to a logging.Logger instance.
246 logtostderr: Log messages to stderr instead of stdout.
247 consoletitle: Display progress in console window's titlebar.
248 writedescription: Write the video description to a .description file
249 writeinfojson: Write the video description to a .info.json file
250 clean_infojson: Remove private fields from the infojson
251 getcomments: Extract video comments. This will not be written to disk
252 unless writeinfojson is also given
253 writeannotations: Write the video annotations to a .annotations.xml file
254 writethumbnail: Write the thumbnail image to a file
255 allow_playlist_files: Whether to write playlists' description, infojson etc
256 also to disk when using the 'write*' options
257 write_all_thumbnails: Write all thumbnail formats to files
258 writelink: Write an internet shortcut file, depending on the
259 current platform (.url/.webloc/.desktop)
260 writeurllink: Write a Windows internet shortcut file (.url)
261 writewebloclink: Write a macOS internet shortcut file (.webloc)
262 writedesktoplink: Write a Linux internet shortcut file (.desktop)
263 writesubtitles: Write the video subtitles to a file
264 writeautomaticsub: Write the automatically generated subtitles to a file
265 listsubtitles: Lists all available subtitles for the video
266 subtitlesformat: The format code for subtitles
267 subtitleslangs: List of languages of the subtitles to download (can be regex).
268 The list may contain "all" to refer to all the available
269 subtitles. The language can be prefixed with a "-" to
270 exclude it from the requested languages. Eg: ['all', '-live_chat']
271 keepvideo: Keep the video file after post-processing
272 daterange: A DateRange object, download only if the upload_date is in the range.
273 skip_download: Skip the actual download of the video file
274 cachedir: Location of the cache files in the filesystem.
275 False to disable filesystem cache.
276 noplaylist: Download single video instead of a playlist if in doubt.
277 age_limit: An integer representing the user's age in years.
278 Unsuitable videos for the given age are skipped.
279 min_views: An integer representing the minimum view count the video
280 must have in order to not be skipped.
281 Videos without view count information are always
282 downloaded. None for no limit.
283 max_views: An integer representing the maximum view count.
284 Videos that are more popular than that are not
286 Videos without view count information are always
287 downloaded. None for no limit.
288 download_archive: File name of a file where all downloads are recorded.
289 Videos already present in the file are not downloaded
291 break_on_existing: Stop the download process after attempting to download a
292 file that is in the archive.
293 break_on_reject: Stop the download process when encountering a video that
294 has been filtered out.
295 break_per_url: Whether break_on_reject and break_on_existing
296 should act on each input URL as opposed to for the entire queue
297 cookiefile: File name or text stream from where cookies should be read and dumped to
298 cookiesfrombrowser: A tuple containing the name of the browser, the profile
299 name/pathfrom where cookies are loaded, and the name of the
300 keyring. Eg: ('chrome', ) or ('vivaldi', 'default', 'BASICTEXT')
301 legacyserverconnect: Explicitly allow HTTPS connection to servers that do not
302 support RFC 5746 secure renegotiation
303 nocheckcertificate: Do not verify SSL certificates
304 client_certificate: Path to client certificate file in PEM format. May include the private key
305 client_certificate_key: Path to private key file for client certificate
306 client_certificate_password: Password for client certificate private key, if encrypted.
307 If not provided and the key is encrypted, yt-dlp will ask interactively
308 prefer_insecure: Use HTTP instead of HTTPS to retrieve information.
309 At the moment, this is only supported by YouTube.
310 http_headers: A dictionary of custom headers to be used for all requests
311 proxy: URL of the proxy server to use
312 geo_verification_proxy: URL of the proxy to use for IP address verification
313 on geo-restricted sites.
314 socket_timeout: Time to wait for unresponsive hosts, in seconds
315 bidi_workaround: Work around buggy terminals without bidirectional text
316 support, using fridibi
317 debug_printtraffic:Print out sent and received HTTP traffic
318 default_search: Prepend this string if an input url is not valid.
319 'auto' for elaborate guessing
320 encoding: Use this encoding instead of the system-specified.
321 extract_flat: Do not resolve URLs, return the immediate result.
322 Pass in 'in_playlist' to only show this behavior for
324 wait_for_video: If given, wait for scheduled streams to become available.
325 The value should be a tuple containing the range
326 (min_secs, max_secs) to wait between retries
327 postprocessors: A list of dictionaries, each with an entry
328 * key: The name of the postprocessor. See
329 yt_dlp/postprocessor/__init__.py for a list.
330 * when: When to run the postprocessor. Allowed values are
331 the entries of utils.POSTPROCESS_WHEN
332 Assumed to be 'post_process' if not given
333 progress_hooks: A list of functions that get called on download
334 progress, with a dictionary with the entries
335 * status: One of "downloading", "error", or "finished".
336 Check this first and ignore unknown values.
337 * info_dict: The extracted info_dict
339 If status is one of "downloading", or "finished", the
340 following properties may also be present:
341 * filename: The final filename (always present)
342 * tmpfilename: The filename we're currently writing to
343 * downloaded_bytes: Bytes on disk
344 * total_bytes: Size of the whole file, None if unknown
345 * total_bytes_estimate: Guess of the eventual file size,
347 * elapsed: The number of seconds since download started.
348 * eta: The estimated time in seconds, None if unknown
349 * speed: The download speed in bytes/second, None if
351 * fragment_index: The counter of the currently
352 downloaded video fragment.
353 * fragment_count: The number of fragments (= individual
354 files that will be merged)
356 Progress hooks are guaranteed to be called at least once
357 (with status "finished") if the download is successful.
358 postprocessor_hooks: A list of functions that get called on postprocessing
359 progress, with a dictionary with the entries
360 * status: One of "started", "processing", or "finished".
361 Check this first and ignore unknown values.
362 * postprocessor: Name of the postprocessor
363 * info_dict: The extracted info_dict
365 Progress hooks are guaranteed to be called at least twice
366 (with status "started" and "finished") if the processing is successful.
367 merge_output_format: Extension to use when merging formats.
368 final_ext: Expected final extension; used to detect when the file was
369 already downloaded and converted
370 fixup: Automatically correct known faults of the file.
372 - "never": do nothing
373 - "warn": only emit a warning
374 - "detect_or_warn": check whether we can do anything
375 about it, warn otherwise (default)
376 source_address: Client-side IP address to bind to.
377 sleep_interval_requests: Number of seconds to sleep between requests
379 sleep_interval: Number of seconds to sleep before each download when
380 used alone or a lower bound of a range for randomized
381 sleep before each download (minimum possible number
382 of seconds to sleep) when used along with
384 max_sleep_interval:Upper bound of a range for randomized sleep before each
385 download (maximum possible number of seconds to sleep).
386 Must only be used along with sleep_interval.
387 Actual sleep time will be a random float from range
388 [sleep_interval; max_sleep_interval].
389 sleep_interval_subtitles: Number of seconds to sleep before each subtitle download
390 listformats: Print an overview of available video formats and exit.
391 list_thumbnails: Print a table of all thumbnails and exit.
392 match_filter: A function that gets called for every video with the signature
393 (info_dict, *, incomplete: bool) -> Optional[str]
394 For backward compatibility with youtube-dl, the signature
395 (info_dict) -> Optional[str] is also allowed.
396 - If it returns a message, the video is ignored.
397 - If it returns None, the video is downloaded.
398 - If it returns utils.NO_DEFAULT, the user is interactively
399 asked whether to download the video.
400 match_filter_func in utils.py is one example for this.
401 no_color: Do not emit color codes in output.
402 geo_bypass: Bypass geographic restriction via faking X-Forwarded-For
405 Two-letter ISO 3166-2 country code that will be used for
406 explicit geographic restriction bypassing via faking
407 X-Forwarded-For HTTP header
409 IP range in CIDR notation that will be used similarly to
411 external_downloader: A dictionary of protocol keys and the executable of the
412 external downloader to use for it. The allowed protocols
413 are default|http|ftp|m3u8|dash|rtsp|rtmp|mms.
414 Set the value to 'native' to use the native downloader
415 compat_opts: Compatibility options. See "Differences in default behavior".
416 The following options do not work when used through the API:
417 filename, abort-on-error, multistreams, no-live-chat, format-sort
418 no-clean-infojson, no-playlist-metafiles, no-keep-subs, no-attach-info-json.
419 Refer __init__.py for their implementation
420 progress_template: Dictionary of templates for progress outputs.
421 Allowed keys are 'download', 'postprocess',
422 'download-title' (console title) and 'postprocess-title'.
423 The template is mapped on a dictionary with keys 'progress' and 'info'
424 retry_sleep_functions: Dictionary of functions that takes the number of attempts
425 as argument and returns the time to sleep in seconds.
426 Allowed keys are 'http', 'fragment', 'file_access'
427 download_ranges: A function that gets called for every video with the signature
428 (info_dict, *, ydl) -> Iterable[Section].
429 Only the returned sections will be downloaded. Each Section contains:
430 * start_time: Start time of the section in seconds
431 * end_time: End time of the section in seconds
432 * title: Section title (Optional)
433 * index: Section number (Optional)
435 The following parameters are not used by YoutubeDL itself, they are used by
436 the downloader (see yt_dlp/downloader/common.py):
437 nopart, updatetime, buffersize, ratelimit, throttledratelimit, min_filesize,
438 max_filesize, test, noresizebuffer, retries, file_access_retries, fragment_retries,
439 continuedl, noprogress, xattr_set_filesize, hls_use_mpegts, http_chunk_size,
440 external_downloader_args, concurrent_fragment_downloads.
442 The following options are used by the post processors:
443 ffmpeg_location: Location of the ffmpeg/avconv binary; either the path
444 to the binary or its containing directory.
445 postprocessor_args: A dictionary of postprocessor/executable keys (in lower case)
446 and a list of additional command-line arguments for the
447 postprocessor/executable. The dict can also have "PP+EXE" keys
448 which are used when the given exe is used by the given PP.
449 Use 'default' as the name for arguments to passed to all PP
450 For compatibility with youtube-dl, a single list of args
453 The following options are used by the extractors:
454 extractor_retries: Number of times to retry for known errors
455 dynamic_mpd: Whether to process dynamic DASH manifests (default: True)
456 hls_split_discontinuity: Split HLS playlists to different formats at
457 discontinuities such as ad breaks (default: False)
458 extractor_args: A dictionary of arguments to be passed to the extractors.
459 See "EXTRACTOR ARGUMENTS" for details.
460 Eg: {'youtube': {'skip': ['dash', 'hls']}}
461 mark_watched: Mark videos watched (even with --simulate). Only for YouTube
463 The following options are deprecated and may be removed in the future:
465 playliststart: - Use playlist_items
466 Playlist item to start at.
467 playlistend: - Use playlist_items
468 Playlist item to end at.
469 playlistreverse: - Use playlist_items
470 Download playlist items in reverse order.
471 forceurl: - Use forceprint
472 Force printing final URL.
473 forcetitle: - Use forceprint
474 Force printing title.
475 forceid: - Use forceprint
477 forcethumbnail: - Use forceprint
478 Force printing thumbnail URL.
479 forcedescription: - Use forceprint
480 Force printing description.
481 forcefilename: - Use forceprint
482 Force printing final filename.
483 forceduration: - Use forceprint
484 Force printing duration.
485 allsubtitles: - Use subtitleslangs = ['all']
486 Downloads all the subtitles of the video
487 (requires writesubtitles or writeautomaticsub)
488 include_ads: - Doesn't work
490 call_home: - Not implemented
491 Boolean, true iff we are allowed to contact the
492 yt-dlp servers for debugging.
493 post_hooks: - Register a custom postprocessor
494 A list of functions that get called as the final step
495 for each video file, after all postprocessors have been
496 called. The filename will be passed as the only argument.
497 hls_prefer_native: - Use external_downloader = {'m3u8': 'native'} or {'m3u8': 'ffmpeg'}.
498 Use the native HLS downloader instead of ffmpeg/avconv
499 if True, otherwise use ffmpeg/avconv if False, otherwise
500 use downloader suggested by extractor if None.
501 prefer_ffmpeg: - avconv support is deprecated
502 If False, use avconv instead of ffmpeg if both are available,
503 otherwise prefer ffmpeg.
504 youtube_include_dash_manifest: - Use extractor_args
505 If True (default), DASH manifests and related
506 data will be downloaded and processed by extractor.
507 You can reduce network I/O by disabling it if you don't
508 care about DASH. (only for youtube)
509 youtube_include_hls_manifest: - Use extractor_args
510 If True (default), HLS manifests and related
511 data will be downloaded and processed by extractor.
512 You can reduce network I/O by disabling it if you don't
513 care about HLS. (only for youtube)
517 'width', 'height', 'tbr', 'abr', 'asr', 'vbr', 'fps', 'filesize', 'filesize_approx',
518 'timestamp', 'release_timestamp',
519 'duration', 'view_count', 'like_count', 'dislike_count', 'repost_count',
520 'average_rating', 'comment_count', 'age_limit',
521 'start_time', 'end_time',
522 'chapter_number', 'season_number', 'episode_number',
523 'track_number', 'disc_number', 'release_year',
527 # NB: Keep in sync with the docstring of extractor/common.py
528 'url', 'manifest_url', 'manifest_stream_number', 'ext', 'format', 'format_id', 'format_note',
529 'width', 'height', 'resolution', 'dynamic_range', 'tbr', 'abr', 'acodec', 'asr',
530 'vbr', 'fps', 'vcodec', 'container', 'filesize', 'filesize_approx',
531 'player_url', 'protocol', 'fragment_base_url', 'fragments', 'is_from_start',
532 'preference', 'language', 'language_preference', 'quality', 'source_preference',
533 'http_headers', 'stretched_ratio', 'no_resume', 'has_drm', 'downloader_options',
534 'page_url', 'app', 'play_path', 'tc_url', 'flash_version', 'rtmp_live', 'rtmp_conn', 'rtmp_protocol', 'rtmp_real_time'
536 _format_selection_exts
= {
537 'audio': {'m4a', 'mp3', 'ogg', 'aac'}
,
538 'video': {'mp4', 'flv', 'webm', '3gp'}
,
539 'storyboards': {'mhtml'}
,
542 def __init__(self
, params
=None, auto_init
=True):
543 """Create a FileDownloader object with the given options.
544 @param auto_init Whether to load the default extractors and print header (if verbose).
545 Set to 'no_verbose_header' to not print the header
551 self
._ies
_instances
= {}
552 self
._pps
= {k: [] for k in POSTPROCESS_WHEN}
553 self
._printed
_messages
= set()
554 self
._first
_webpage
_request
= True
555 self
._post
_hooks
= []
556 self
._progress
_hooks
= []
557 self
._postprocessor
_hooks
= []
558 self
._download
_retcode
= 0
559 self
._num
_downloads
= 0
561 self
._playlist
_level
= 0
562 self
._playlist
_urls
= set()
563 self
.cache
= Cache(self
)
565 windows_enable_vt_mode()
566 stdout
= sys
.stderr
if self
.params
.get('logtostderr') else sys
.stdout
567 self
._out
_files
= Namespace(
570 screen
=sys
.stderr
if self
.params
.get('quiet') else stdout
,
571 console
=None if compat_os_name
== 'nt' else next(
572 filter(supports_terminal_sequences
, (sys
.stderr
, sys
.stdout
)), None)
574 self
._allow
_colors
= Namespace(**{
575 type_
: not self
.params
.get('no_color') and supports_terminal_sequences(stream
)
576 for type_
, stream
in self
._out
_files
.items_
if type_
!= 'console'
579 MIN_SUPPORTED
, MIN_RECOMMENDED
= (3, 6), (3, 7)
580 current_version
= sys
.version_info
[:2]
581 if current_version
< MIN_RECOMMENDED
:
582 msg
= ('Support for Python version %d.%d has been deprecated. '
583 'See https://github.com/yt-dlp/yt-dlp/issues/3764 for more details. '
584 'You will recieve only one more update on this version')
585 if current_version
< MIN_SUPPORTED
:
586 msg
= 'Python version %d.%d is no longer supported'
587 self
.deprecation_warning(
588 f
'{msg}! Please update to Python %d.%d or above' % (*current_version
, *MIN_RECOMMENDED
))
590 if self
.params
.get('allow_unplayable_formats'):
592 f
'You have asked for {self._format_err("UNPLAYABLE", self.Styles.EMPHASIS)} formats to be listed/downloaded. '
593 'This is a developer option intended for debugging. \n'
594 ' If you experience any issues while using this option, '
595 f
'{self._format_err("DO NOT", self.Styles.ERROR)} open a bug report')
597 def check_deprecated(param
, option
, suggestion
):
598 if self
.params
.get(param
) is not None:
599 self
.report_warning(f
'{option} is deprecated. Use {suggestion} instead')
603 if check_deprecated('cn_verification_proxy', '--cn-verification-proxy', '--geo-verification-proxy'):
604 if self
.params
.get('geo_verification_proxy') is None:
605 self
.params
['geo_verification_proxy'] = self
.params
['cn_verification_proxy']
607 check_deprecated('autonumber', '--auto-number', '-o "%(autonumber)s-%(title)s.%(ext)s"')
608 check_deprecated('usetitle', '--title', '-o "%(title)s-%(id)s.%(ext)s"')
609 check_deprecated('useid', '--id', '-o "%(id)s.%(ext)s"')
611 for msg
in self
.params
.get('_warnings', []):
612 self
.report_warning(msg
)
613 for msg
in self
.params
.get('_deprecation_warnings', []):
614 self
.deprecation_warning(msg
)
616 self
.params
['compat_opts'] = set(self
.params
.get('compat_opts', ()))
617 if not compat_has_legacy
:
618 self
.params
['compat_opts'].add('no-compat-legacy')
619 if 'list-formats' in self
.params
['compat_opts']:
620 self
.params
['listformats_table'] = False
622 if 'overwrites' not in self
.params
and self
.params
.get('nooverwrites') is not None:
623 # nooverwrites was unnecessarily changed to overwrites
624 # in 0c3d0f51778b153f65c21906031c2e091fcfb641
625 # This ensures compatibility with both keys
626 self
.params
['overwrites'] = not self
.params
['nooverwrites']
627 elif self
.params
.get('overwrites') is None:
628 self
.params
.pop('overwrites', None)
630 self
.params
['nooverwrites'] = not self
.params
['overwrites']
632 self
.params
.setdefault('forceprint', {})
633 self
.params
.setdefault('print_to_file', {})
635 # Compatibility with older syntax
636 if not isinstance(params
['forceprint'], dict):
637 self
.params
['forceprint'] = {'video': params['forceprint']}
639 if self
.params
.get('bidi_workaround', False):
642 master
, slave
= pty
.openpty()
643 width
= shutil
.get_terminal_size().columns
644 width_args
= [] if width
is None else ['-w', str(width
)]
645 sp_kwargs
= {'stdin': subprocess.PIPE, 'stdout': slave, 'stderr': self._out_files.error}
647 self
._output
_process
= Popen(['bidiv'] + width_args
, **sp_kwargs
)
649 self
._output
_process
= Popen(['fribidi', '-c', 'UTF-8'] + width_args
, **sp_kwargs
)
650 self
._output
_channel
= os
.fdopen(master
, 'rb')
651 except OSError as ose
:
652 if ose
.errno
== errno
.ENOENT
:
654 'Could not find fribidi executable, ignoring --bidi-workaround. '
655 'Make sure that fribidi is an executable file in one of the directories in your $PATH.')
660 if auto_init
!= 'no_verbose_header':
661 self
.print_debug_header()
662 self
.add_default_info_extractors()
664 if (sys
.platform
!= 'win32'
665 and sys
.getfilesystemencoding() in ['ascii', 'ANSI_X3.4-1968']
666 and not self
.params
.get('restrictfilenames', False)):
667 # Unicode filesystem API will throw errors (#1474, #13027)
669 'Assuming --restrict-filenames since file system encoding '
670 'cannot encode all characters. '
671 'Set the LC_ALL environment variable to fix this.')
672 self
.params
['restrictfilenames'] = True
674 self
._parse
_outtmpl
()
676 # Creating format selector here allows us to catch syntax errors before the extraction
677 self
.format_selector
= (
678 self
.params
.get('format') if self
.params
.get('format') in (None, '-')
679 else self
.params
['format'] if callable(self
.params
['format'])
680 else self
.build_format_selector(self
.params
['format']))
682 # Set http_headers defaults according to std_headers
683 self
.params
['http_headers'] = merge_headers(std_headers
, self
.params
.get('http_headers', {}))
686 'post_hooks': self
.add_post_hook
,
687 'progress_hooks': self
.add_progress_hook
,
688 'postprocessor_hooks': self
.add_postprocessor_hook
,
690 for opt
, fn
in hooks
.items():
691 for ph
in self
.params
.get(opt
, []):
694 for pp_def_raw
in self
.params
.get('postprocessors', []):
695 pp_def
= dict(pp_def_raw
)
696 when
= pp_def
.pop('when', 'post_process')
697 self
.add_post_processor(
698 get_postprocessor(pp_def
.pop('key'))(self
, **pp_def
),
702 register_socks_protocols()
704 def preload_download_archive(fn
):
705 """Preload the archive, if any is specified"""
708 self
.write_debug(f
'Loading archive file {fn!r}')
710 with locked_file(fn
, 'r', encoding
='utf-8') as archive_file
:
711 for line
in archive_file
:
712 self
.archive
.add(line
.strip())
713 except OSError as ioe
:
714 if ioe
.errno
!= errno
.ENOENT
:
720 preload_download_archive(self
.params
.get('download_archive'))
722 def warn_if_short_id(self
, argv
):
723 # short YouTube ID starting with dash?
725 i
for i
, a
in enumerate(argv
)
726 if re
.match(r
'^-[0-9A-Za-z_-]{10}$', a
)]
730 + [a
for i
, a
in enumerate(argv
) if i
not in idxs
]
731 + ['--'] + [argv
[i
] for i
in idxs
]
734 'Long argument string detected. '
735 'Use -- to separate parameters and URLs, like this:\n%s' %
736 args_to_str(correct_argv
))
738 def add_info_extractor(self
, ie
):
739 """Add an InfoExtractor object to the end of the list."""
741 self
._ies
[ie_key
] = ie
742 if not isinstance(ie
, type):
743 self
._ies
_instances
[ie_key
] = ie
744 ie
.set_downloader(self
)
746 def _get_info_extractor_class(self
, ie_key
):
747 ie
= self
._ies
.get(ie_key
)
749 ie
= get_info_extractor(ie_key
)
750 self
.add_info_extractor(ie
)
753 def get_info_extractor(self
, ie_key
):
755 Get an instance of an IE with name ie_key, it will try to get one from
756 the _ies list, if there's no instance it will create a new one and add
757 it to the extractor list.
759 ie
= self
._ies
_instances
.get(ie_key
)
761 ie
= get_info_extractor(ie_key
)()
762 self
.add_info_extractor(ie
)
765 def add_default_info_extractors(self
):
767 Add the InfoExtractors returned by gen_extractors to the end of the list
769 for ie
in gen_extractor_classes():
770 self
.add_info_extractor(ie
)
772 def add_post_processor(self
, pp
, when
='post_process'):
773 """Add a PostProcessor object to the end of the chain."""
774 assert when
in POSTPROCESS_WHEN
, f
'Invalid when={when}'
775 self
._pps
[when
].append(pp
)
776 pp
.set_downloader(self
)
778 def add_post_hook(self
, ph
):
779 """Add the post hook"""
780 self
._post
_hooks
.append(ph
)
782 def add_progress_hook(self
, ph
):
783 """Add the download progress hook"""
784 self
._progress
_hooks
.append(ph
)
786 def add_postprocessor_hook(self
, ph
):
787 """Add the postprocessing progress hook"""
788 self
._postprocessor
_hooks
.append(ph
)
789 for pps
in self
._pps
.values():
791 pp
.add_progress_hook(ph
)
793 def _bidi_workaround(self
, message
):
794 if not hasattr(self
, '_output_channel'):
797 assert hasattr(self
, '_output_process')
798 assert isinstance(message
, str)
799 line_count
= message
.count('\n') + 1
800 self
._output
_process
.stdin
.write((message
+ '\n').encode())
801 self
._output
_process
.stdin
.flush()
802 res
= ''.join(self
._output
_channel
.readline().decode()
803 for _
in range(line_count
))
804 return res
[:-len('\n')]
806 def _write_string(self
, message
, out
=None, only_once
=False):
808 if message
in self
._printed
_messages
:
810 self
._printed
_messages
.add(message
)
811 write_string(message
, out
=out
, encoding
=self
.params
.get('encoding'))
813 def to_stdout(self
, message
, skip_eol
=False, quiet
=None):
814 """Print message to stdout"""
815 if quiet
is not None:
816 self
.deprecation_warning('"YoutubeDL.to_stdout" no longer accepts the argument quiet. Use "YoutubeDL.to_screen" instead')
817 if skip_eol
is not False:
818 self
.deprecation_warning('"YoutubeDL.to_stdout" no longer accepts the argument skip_eol. Use "YoutubeDL.to_screen" instead')
819 self
._write
_string
(f
'{self._bidi_workaround(message)}\n', self
._out
_files
.out
)
821 def to_screen(self
, message
, skip_eol
=False, quiet
=None):
822 """Print message to screen if not in quiet mode"""
823 if self
.params
.get('logger'):
824 self
.params
['logger'].debug(message
)
826 if (self
.params
.get('quiet') if quiet
is None else quiet
) and not self
.params
.get('verbose'):
829 '%s%s' % (self
._bidi
_workaround
(message
), ('' if skip_eol
else '\n')),
830 self
._out
_files
.screen
)
832 def to_stderr(self
, message
, only_once
=False):
833 """Print message to stderr"""
834 assert isinstance(message
, str)
835 if self
.params
.get('logger'):
836 self
.params
['logger'].error(message
)
838 self
._write
_string
(f
'{self._bidi_workaround(message)}\n', self
._out
_files
.error
, only_once
=only_once
)
840 def _send_console_code(self
, code
):
841 if compat_os_name
== 'nt' or not self
._out
_files
.console
:
843 self
._write
_string
(code
, self
._out
_files
.console
)
845 def to_console_title(self
, message
):
846 if not self
.params
.get('consoletitle', False):
848 message
= remove_terminal_sequences(message
)
849 if compat_os_name
== 'nt':
850 if ctypes
.windll
.kernel32
.GetConsoleWindow():
851 # c_wchar_p() might not be necessary if `message` is
852 # already of type unicode()
853 ctypes
.windll
.kernel32
.SetConsoleTitleW(ctypes
.c_wchar_p(message
))
855 self
._send
_console
_code
(f
'\033]0;{message}\007')
857 def save_console_title(self
):
858 if not self
.params
.get('consoletitle') or self
.params
.get('simulate'):
860 self
._send
_console
_code
('\033[22;0t') # Save the title on stack
862 def restore_console_title(self
):
863 if not self
.params
.get('consoletitle') or self
.params
.get('simulate'):
865 self
._send
_console
_code
('\033[23;0t') # Restore the title from stack
868 self
.save_console_title()
871 def __exit__(self
, *args
):
872 self
.restore_console_title()
874 if self
.params
.get('cookiefile') is not None:
875 self
.cookiejar
.save(ignore_discard
=True, ignore_expires
=True)
877 def trouble(self
, message
=None, tb
=None, is_error
=True):
878 """Determine action to take when a download problem appears.
880 Depending on if the downloader has been configured to ignore
881 download errors or not, this method may throw an exception or
882 not when errors are found, after printing the message.
884 @param tb If given, is additional traceback information
885 @param is_error Whether to raise error according to ignorerrors
887 if message
is not None:
888 self
.to_stderr(message
)
889 if self
.params
.get('verbose'):
891 if sys
.exc_info()[0]: # if .trouble has been called from an except block
893 if hasattr(sys
.exc_info()[1], 'exc_info') and sys
.exc_info()[1].exc_info
[0]:
894 tb
+= ''.join(traceback
.format_exception(*sys
.exc_info()[1].exc_info
))
895 tb
+= encode_compat_str(traceback
.format_exc())
897 tb_data
= traceback
.format_list(traceback
.extract_stack())
898 tb
= ''.join(tb_data
)
903 if not self
.params
.get('ignoreerrors'):
904 if sys
.exc_info()[0] and hasattr(sys
.exc_info()[1], 'exc_info') and sys
.exc_info()[1].exc_info
[0]:
905 exc_info
= sys
.exc_info()[1].exc_info
907 exc_info
= sys
.exc_info()
908 raise DownloadError(message
, exc_info
)
909 self
._download
_retcode
= 1
913 EMPHASIS
='light blue',
919 SUPPRESS
='light black',
922 def _format_text(self
, handle
, allow_colors
, text
, f
, fallback
=None, *, test_encoding
=False):
926 # handle.encoding can be None. See https://github.com/yt-dlp/yt-dlp/issues/2711
927 encoding
= self
.params
.get('encoding') or getattr(handle
, 'encoding', None) or 'ascii'
928 text
= text
.encode(encoding
, 'ignore').decode(encoding
)
929 if fallback
is not None and text
!= original_text
:
931 return format_text(text
, f
) if allow_colors
else text
if fallback
is None else fallback
933 def _format_out(self
, *args
, **kwargs
):
934 return self
._format
_text
(self
._out
_files
.out
, self
._allow
_colors
.out
, *args
, **kwargs
)
936 def _format_screen(self
, *args
, **kwargs
):
937 return self
._format
_text
(self
._out
_files
.screen
, self
._allow
_colors
.screen
, *args
, **kwargs
)
939 def _format_err(self
, *args
, **kwargs
):
940 return self
._format
_text
(self
._out
_files
.error
, self
._allow
_colors
.error
, *args
, **kwargs
)
942 def report_warning(self
, message
, only_once
=False):
944 Print the message to stderr, it will be prefixed with 'WARNING:'
945 If stderr is a tty file the 'WARNING:' will be colored
947 if self
.params
.get('logger') is not None:
948 self
.params
['logger'].warning(message
)
950 if self
.params
.get('no_warnings'):
952 self
.to_stderr(f
'{self._format_err("WARNING:", self.Styles.WARNING)} {message}', only_once
)
954 def deprecation_warning(self
, message
):
955 if self
.params
.get('logger') is not None:
956 self
.params
['logger'].warning(f
'DeprecationWarning: {message}')
958 self
.to_stderr(f
'{self._format_err("DeprecationWarning:", self.Styles.ERROR)} {message}', True)
960 def report_error(self
, message
, *args
, **kwargs
):
962 Do the same as trouble, but prefixes the message with 'ERROR:', colored
963 in red if stderr is a tty file.
965 self
.trouble(f
'{self._format_err("ERROR:", self.Styles.ERROR)} {message}', *args
, **kwargs
)
967 def write_debug(self
, message
, only_once
=False):
968 '''Log debug message or Print message to stderr'''
969 if not self
.params
.get('verbose', False):
971 message
= f
'[debug] {message}'
972 if self
.params
.get('logger'):
973 self
.params
['logger'].debug(message
)
975 self
.to_stderr(message
, only_once
)
977 def report_file_already_downloaded(self
, file_name
):
978 """Report file has already been fully downloaded."""
980 self
.to_screen('[download] %s has already been downloaded' % file_name
)
981 except UnicodeEncodeError:
982 self
.to_screen('[download] The file has already been downloaded')
984 def report_file_delete(self
, file_name
):
985 """Report that existing file will be deleted."""
987 self
.to_screen('Deleting existing file %s' % file_name
)
988 except UnicodeEncodeError:
989 self
.to_screen('Deleting existing file')
991 def raise_no_formats(self
, info
, forced
=False, *, msg
=None):
992 has_drm
= info
.get('_has_drm')
993 ignored
, expected
= self
.params
.get('ignore_no_formats_error'), bool(msg
)
994 msg
= msg
or has_drm
and 'This video is DRM protected' or 'No video formats found!'
995 if forced
or not ignored
:
996 raise ExtractorError(msg
, video_id
=info
['id'], ie
=info
['extractor'],
997 expected
=has_drm
or ignored
or expected
)
999 self
.report_warning(msg
)
1001 def parse_outtmpl(self
):
1002 self
.deprecation_warning('"YoutubeDL.parse_outtmpl" is deprecated and may be removed in a future version')
1003 self
._parse
_outtmpl
()
1004 return self
.params
['outtmpl']
1006 def _parse_outtmpl(self
):
1008 if self
.params
.get('restrictfilenames'): # Remove spaces in the default template
1009 sanitize
= lambda x
: x
.replace(' - ', ' ').replace(' ', '-')
1011 outtmpl
= self
.params
.setdefault('outtmpl', {})
1012 if not isinstance(outtmpl
, dict):
1013 self
.params
['outtmpl'] = outtmpl
= {'default': outtmpl}
1014 outtmpl
.update({k: sanitize(v) for k, v in DEFAULT_OUTTMPL.items() if outtmpl.get(k) is None}
)
1016 def get_output_path(self
, dir_type
='', filename
=None):
1017 paths
= self
.params
.get('paths', {})
1018 assert isinstance(paths
, dict)
1019 path
= os
.path
.join(
1020 expand_path(paths
.get('home', '').strip()),
1021 expand_path(paths
.get(dir_type
, '').strip()) if dir_type
else '',
1023 return sanitize_path(path
, force
=self
.params
.get('windowsfilenames'))
1026 def _outtmpl_expandpath(outtmpl
):
1027 # expand_path translates '%%' into '%' and '$$' into '$'
1028 # correspondingly that is not what we want since we need to keep
1029 # '%%' intact for template dict substitution step. Working around
1030 # with boundary-alike separator hack.
1031 sep
= ''.join([random
.choice(ascii_letters
) for _
in range(32)])
1032 outtmpl
= outtmpl
.replace('%%', f
'%{sep}%').replace('$$', f
'${sep}$')
1034 # outtmpl should be expand_path'ed before template dict substitution
1035 # because meta fields may contain env variables we don't want to
1036 # be expanded. For example, for outtmpl "%(title)s.%(ext)s" and
1037 # title "Hello $PATH", we don't want `$PATH` to be expanded.
1038 return expand_path(outtmpl
).replace(sep
, '')
1041 def escape_outtmpl(outtmpl
):
1042 ''' Escape any remaining strings like %s, %abc% etc. '''
1044 STR_FORMAT_RE_TMPL
.format('', '(?![%(\0])'),
1045 lambda mobj
: ('' if mobj
.group('has_key') else '%') + mobj
.group(0),
1049 def validate_outtmpl(cls
, outtmpl
):
1050 ''' @return None or Exception object '''
1052 STR_FORMAT_RE_TMPL
.format('[^)]*', '[ljhqBUDS]'),
1053 lambda mobj
: f
'{mobj.group(0)[:-1]}s',
1054 cls
._outtmpl
_expandpath
(outtmpl
))
1056 cls
.escape_outtmpl(outtmpl
) % collections
.defaultdict(int)
1058 except ValueError as err
:
1062 def _copy_infodict(info_dict
):
1063 info_dict
= dict(info_dict
)
1064 info_dict
.pop('__postprocessors', None)
1065 info_dict
.pop('__pending_error', None)
1068 def prepare_outtmpl(self
, outtmpl
, info_dict
, sanitize
=False):
1069 """ Make the outtmpl and info_dict suitable for substitution: ydl.escape_outtmpl(outtmpl) % info_dict
1070 @param sanitize Whether to sanitize the output as a filename.
1071 For backward compatibility, a function can also be passed
1074 info_dict
.setdefault('epoch', int(time
.time())) # keep epoch consistent once set
1076 info_dict
= self
._copy
_infodict
(info_dict
)
1077 info_dict
['duration_string'] = ( # %(duration>%H-%M-%S)s is wrong if duration > 24hrs
1078 formatSeconds(info_dict
['duration'], '-' if sanitize
else ':')
1079 if info_dict
.get('duration', None) is not None
1081 info_dict
['autonumber'] = int(self
.params
.get('autonumber_start', 1) - 1 + self
._num
_downloads
)
1082 info_dict
['video_autonumber'] = self
._num
_videos
1083 if info_dict
.get('resolution') is None:
1084 info_dict
['resolution'] = self
.format_resolution(info_dict
, default
=None)
1086 # For fields playlist_index, playlist_autonumber and autonumber convert all occurrences
1087 # of %(field)s to %(field)0Nd for backward compatibility
1088 field_size_compat_map
= {
1089 'playlist_index': number_of_digits(info_dict
.get('__last_playlist_index') or 0),
1090 'playlist_autonumber': number_of_digits(info_dict
.get('n_entries') or 0),
1091 'autonumber': self
.params
.get('autonumber_size') or 5,
1095 EXTERNAL_FORMAT_RE
= re
.compile(STR_FORMAT_RE_TMPL
.format('[^)]*', f
'[{STR_FORMAT_TYPES}ljhqBUDS]'))
1100 # Field is of the form key1.key2...
1101 # where keys (except first) can be string, int or slice
1102 FIELD_RE
= r
'\w*(?:\.(?:\w+|{num}|{num}?(?::{num}?){{1,2}}))*'.format(num
=r
'(?:-?\d+)')
1103 MATH_FIELD_RE
= rf
'(?:{FIELD_RE}|-?{NUMBER_RE})'
1104 MATH_OPERATORS_RE
= r
'(?:%s)' % '|'.join(map(re
.escape
, MATH_FUNCTIONS
.keys()))
1105 INTERNAL_FORMAT_RE
= re
.compile(rf
'''(?x)
1107 (?P<fields>{FIELD_RE})
1108 (?P<maths>(?:{MATH_OPERATORS_RE}{MATH_FIELD_RE})*)
1109 (?:>(?P<strf_format>.+?))?
1111 (?P<alternate>(?<!\\),[^|&)]+)?
1112 (?:&(?P<replacement>.*?))?
1113 (?:\|(?P<default>.*?))?
1116 def _traverse_infodict(k
):
1120 return traverse_obj(info_dict
, k
, is_user_input
=True, traverse_string
=True)
1122 def get_value(mdict
):
1124 value
= _traverse_infodict(mdict
['fields'])
1127 value
= float_or_none(value
)
1128 if value
is not None:
1131 offset_key
= mdict
['maths']
1133 value
= float_or_none(value
)
1137 MATH_FIELD_RE
if operator
else MATH_OPERATORS_RE
,
1138 offset_key
).group(0)
1139 offset_key
= offset_key
[len(item
):]
1140 if operator
is None:
1141 operator
= MATH_FUNCTIONS
[item
]
1143 item
, multiplier
= (item
[1:], -1) if item
[0] == '-' else (item
, 1)
1144 offset
= float_or_none(item
)
1146 offset
= float_or_none(_traverse_infodict(item
))
1148 value
= operator(value
, multiplier
* offset
)
1149 except (TypeError, ZeroDivisionError):
1152 # Datetime formatting
1153 if mdict
['strf_format']:
1154 value
= strftime_or_none(value
, mdict
['strf_format'].replace('\\,', ','))
1158 na
= self
.params
.get('outtmpl_na_placeholder', 'NA')
1160 def filename_sanitizer(key
, value
, restricted
=self
.params
.get('restrictfilenames')):
1161 return sanitize_filename(str(value
), restricted
=restricted
, is_id
=(
1162 bool(re
.search(r
'(^|[_.])id(\.|$)', key
))
1163 if 'filename-sanitization' in self
.params
['compat_opts']
1166 sanitizer
= sanitize
if callable(sanitize
) else filename_sanitizer
1167 sanitize
= bool(sanitize
)
1169 def _dumpjson_default(obj
):
1170 if isinstance(obj
, (set, LazyList
)):
1174 def create_key(outer_mobj
):
1175 if not outer_mobj
.group('has_key'):
1176 return outer_mobj
.group(0)
1177 key
= outer_mobj
.group('key')
1178 mobj
= re
.match(INTERNAL_FORMAT_RE
, key
)
1179 initial_field
= mobj
.group('fields') if mobj
else ''
1180 value
, replacement
, default
= None, None, na
1182 mobj
= mobj
.groupdict()
1183 default
= mobj
['default'] if mobj
['default'] is not None else default
1184 value
= get_value(mobj
)
1185 replacement
= mobj
['replacement']
1186 if value
is None and mobj
['alternate']:
1187 mobj
= re
.match(INTERNAL_FORMAT_RE
, mobj
['remaining'][1:])
1191 fmt
= outer_mobj
.group('format')
1192 if fmt
== 's' and value
is not None and key
in field_size_compat_map
.keys():
1193 fmt
= f
'0{field_size_compat_map[key]:d}d'
1195 value
= default
if value
is None else value
if replacement
is None else replacement
1197 flags
= outer_mobj
.group('conversion') or ''
1198 str_fmt
= f
'{fmt[:-1]}s'
1199 if fmt
[-1] == 'l': # list
1200 delim
= '\n' if '#' in flags
else ', '
1201 value
, fmt
= delim
.join(map(str, variadic(value
, allowed_types
=(str, bytes)))), str_fmt
1202 elif fmt
[-1] == 'j': # json
1203 value
, fmt
= json
.dumps(value
, default
=_dumpjson_default
, indent
=4 if '#' in flags
else None), str_fmt
1204 elif fmt
[-1] == 'h': # html
1205 value
, fmt
= escapeHTML(value
), str_fmt
1206 elif fmt
[-1] == 'q': # quoted
1207 value
= map(str, variadic(value
) if '#' in flags
else [value
])
1208 value
, fmt
= ' '.join(map(compat_shlex_quote
, value
)), str_fmt
1209 elif fmt
[-1] == 'B': # bytes
1210 value
= f
'%{str_fmt}'.encode() % str(value
).encode()
1211 value
, fmt
= value
.decode('utf-8', 'ignore'), 's'
1212 elif fmt
[-1] == 'U': # unicode normalized
1213 value
, fmt
= unicodedata
.normalize(
1214 # "+" = compatibility equivalence, "#" = NFD
1215 'NF%s%s' % ('K' if '+' in flags
else '', 'D' if '#' in flags
else 'C'),
1217 elif fmt
[-1] == 'D': # decimal suffix
1218 num_fmt
, fmt
= fmt
[:-1].replace('#', ''), 's'
1219 value
= format_decimal_suffix(value
, f
'%{num_fmt}f%s' if num_fmt
else '%d%s',
1220 factor
=1024 if '#' in flags
else 1000)
1221 elif fmt
[-1] == 'S': # filename sanitization
1222 value
, fmt
= filename_sanitizer(initial_field
, value
, restricted
='#' in flags
), str_fmt
1223 elif fmt
[-1] == 'c':
1225 value
= str(value
)[0]
1228 elif fmt
[-1] not in 'rs': # numeric
1229 value
= float_or_none(value
)
1231 value
, fmt
= default
, 's'
1235 # If value is an object, sanitize might convert it to a string
1236 # So we convert it to repr first
1237 value
, fmt
= repr(value
), str_fmt
1238 if fmt
[-1] in 'csr':
1239 value
= sanitizer(initial_field
, value
)
1241 key
= '%s\0%s' % (key
.replace('%', '%\0'), outer_mobj
.group('format'))
1242 TMPL_DICT
[key
] = value
1243 return '{prefix}%({key}){fmt}'.format(key
=key
, fmt
=fmt
, prefix
=outer_mobj
.group('prefix'))
1245 return EXTERNAL_FORMAT_RE
.sub(create_key
, outtmpl
), TMPL_DICT
1247 def evaluate_outtmpl(self
, outtmpl
, info_dict
, *args
, **kwargs
):
1248 outtmpl
, info_dict
= self
.prepare_outtmpl(outtmpl
, info_dict
, *args
, **kwargs
)
1249 return self
.escape_outtmpl(outtmpl
) % info_dict
1251 def _prepare_filename(self
, info_dict
, *, outtmpl
=None, tmpl_type
=None):
1252 assert None in (outtmpl
, tmpl_type
), 'outtmpl and tmpl_type are mutually exclusive'
1254 outtmpl
= self
.params
['outtmpl'].get(tmpl_type
or 'default', self
.params
['outtmpl']['default'])
1256 outtmpl
= self
._outtmpl
_expandpath
(outtmpl
)
1257 filename
= self
.evaluate_outtmpl(outtmpl
, info_dict
, True)
1261 if tmpl_type
in ('', 'temp'):
1262 final_ext
, ext
= self
.params
.get('final_ext'), info_dict
.get('ext')
1263 if final_ext
and ext
and final_ext
!= ext
and filename
.endswith(f
'.{final_ext}'):
1264 filename
= replace_extension(filename
, ext
, final_ext
)
1266 force_ext
= OUTTMPL_TYPES
[tmpl_type
]
1268 filename
= replace_extension(filename
, force_ext
, info_dict
.get('ext'))
1270 # https://github.com/blackjack4494/youtube-dlc/issues/85
1271 trim_file_name
= self
.params
.get('trim_file_name', False)
1273 no_ext
, *ext
= filename
.rsplit('.', 2)
1274 filename
= join_nonempty(no_ext
[:trim_file_name
], *ext
, delim
='.')
1277 except ValueError as err
:
1278 self
.report_error('Error in output template: ' + str(err
) + ' (encoding: ' + repr(preferredencoding()) + ')')
1281 def prepare_filename(self
, info_dict
, dir_type
='', *, outtmpl
=None, warn
=False):
1282 """Generate the output filename"""
1284 assert not dir_type
, 'outtmpl and dir_type are mutually exclusive'
1286 filename
= self
._prepare
_filename
(info_dict
, tmpl_type
=dir_type
, outtmpl
=outtmpl
)
1287 if not filename
and dir_type
not in ('', 'temp'):
1291 if not self
.params
.get('paths'):
1293 elif filename
== '-':
1294 self
.report_warning('--paths is ignored when an outputting to stdout', only_once
=True)
1295 elif os
.path
.isabs(filename
):
1296 self
.report_warning('--paths is ignored since an absolute path is given in output template', only_once
=True)
1297 if filename
== '-' or not filename
:
1300 return self
.get_output_path(dir_type
, filename
)
1302 def _match_entry(self
, info_dict
, incomplete
=False, silent
=False):
1303 """ Returns None if the file should be downloaded """
1305 video_title
= info_dict
.get('title', info_dict
.get('id', 'video'))
1308 if 'title' in info_dict
:
1309 # This can happen when we're just evaluating the playlist
1310 title
= info_dict
['title']
1311 matchtitle
= self
.params
.get('matchtitle', False)
1313 if not re
.search(matchtitle
, title
, re
.IGNORECASE
):
1314 return '"' + title
+ '" title did not match pattern "' + matchtitle
+ '"'
1315 rejecttitle
= self
.params
.get('rejecttitle', False)
1317 if re
.search(rejecttitle
, title
, re
.IGNORECASE
):
1318 return '"' + title
+ '" title matched reject pattern "' + rejecttitle
+ '"'
1319 date
= info_dict
.get('upload_date')
1320 if date
is not None:
1321 dateRange
= self
.params
.get('daterange', DateRange())
1322 if date
not in dateRange
:
1323 return f
'{date_from_str(date).isoformat()} upload date is not in range {dateRange}'
1324 view_count
= info_dict
.get('view_count')
1325 if view_count
is not None:
1326 min_views
= self
.params
.get('min_views')
1327 if min_views
is not None and view_count
< min_views
:
1328 return 'Skipping %s, because it has not reached minimum view count (%d/%d)' % (video_title
, view_count
, min_views
)
1329 max_views
= self
.params
.get('max_views')
1330 if max_views
is not None and view_count
> max_views
:
1331 return 'Skipping %s, because it has exceeded the maximum view count (%d/%d)' % (video_title
, view_count
, max_views
)
1332 if age_restricted(info_dict
.get('age_limit'), self
.params
.get('age_limit')):
1333 return 'Skipping "%s" because it is age restricted' % video_title
1335 match_filter
= self
.params
.get('match_filter')
1336 if match_filter
is not None:
1338 ret
= match_filter(info_dict
, incomplete
=incomplete
)
1340 # For backward compatibility
1341 ret
= None if incomplete
else match_filter(info_dict
)
1342 if ret
is NO_DEFAULT
:
1344 filename
= self
._format
_screen
(self
.prepare_filename(info_dict
), self
.Styles
.FILENAME
)
1345 reply
= input(self
._format
_screen
(
1346 f
'Download "{filename}"? (Y/n): ', self
.Styles
.EMPHASIS
)).lower().strip()
1347 if reply
in {'y', ''}
:
1350 return f
'Skipping {video_title}'
1351 elif ret
is not None:
1355 if self
.in_download_archive(info_dict
):
1356 reason
= '%s has already been recorded in the archive' % video_title
1357 break_opt
, break_err
= 'break_on_existing', ExistingVideoReached
1359 reason
= check_filter()
1360 break_opt
, break_err
= 'break_on_reject', RejectedVideoReached
1361 if reason
is not None:
1363 self
.to_screen('[download] ' + reason
)
1364 if self
.params
.get(break_opt
, False):
1369 def add_extra_info(info_dict
, extra_info
):
1370 '''Set the keys from extra_info in info dict if they are missing'''
1371 for key
, value
in extra_info
.items():
1372 info_dict
.setdefault(key
, value
)
1374 def extract_info(self
, url
, download
=True, ie_key
=None, extra_info
=None,
1375 process
=True, force_generic_extractor
=False):
1377 Return a list with a dictionary for each video extracted.
1380 url -- URL to extract
1383 download -- whether to download videos during extraction
1384 ie_key -- extractor key hint
1385 extra_info -- dictionary containing the extra values to add to each result
1386 process -- whether to resolve all unresolved references (URLs, playlist items),
1387 must be True for download to work.
1388 force_generic_extractor -- force using the generic extractor
1391 if extra_info
is None:
1394 if not ie_key
and force_generic_extractor
:
1398 ies
= {ie_key: self._get_info_extractor_class(ie_key)}
1402 for ie_key
, ie
in ies
.items():
1403 if not ie
.suitable(url
):
1406 if not ie
.working():
1407 self
.report_warning('The program functionality for this site has been marked as broken, '
1408 'and will probably not work.')
1410 temp_id
= ie
.get_temp_id(url
)
1411 if temp_id
is not None and self
.in_download_archive({'id': temp_id, 'ie_key': ie_key}
):
1412 self
.to_screen(f
'[{ie_key}] {temp_id}: has already been recorded in the archive')
1413 if self
.params
.get('break_on_existing', False):
1414 raise ExistingVideoReached()
1416 return self
.__extract
_info
(url
, self
.get_info_extractor(ie_key
), download
, extra_info
, process
)
1418 self
.report_error('no suitable InfoExtractor for URL %s' % url
)
1420 def _handle_extraction_exceptions(func
):
1421 @functools.wraps(func
)
1422 def wrapper(self
, *args
, **kwargs
):
1425 return func(self
, *args
, **kwargs
)
1426 except (DownloadCancelled
, LazyList
.IndexError, PagedList
.IndexError):
1428 except ReExtractInfo
as e
:
1430 self
.to_screen(f
'{e}; Re-extracting data')
1432 self
.to_stderr('\r')
1433 self
.report_warning(f
'{e}; Re-extracting data')
1435 except GeoRestrictedError
as e
:
1438 msg
+= '\nThis video is available in %s.' % ', '.join(
1439 map(ISO3166Utils
.short2full
, e
.countries
))
1440 msg
+= '\nYou might want to use a VPN or a proxy server (with --proxy) to workaround.'
1441 self
.report_error(msg
)
1442 except ExtractorError
as e
: # An error we somewhat expected
1443 self
.report_error(str(e
), e
.format_traceback())
1444 except Exception as e
:
1445 if self
.params
.get('ignoreerrors'):
1446 self
.report_error(str(e
), tb
=encode_compat_str(traceback
.format_exc()))
1452 def _wait_for_video(self
, ie_result
):
1453 if (not self
.params
.get('wait_for_video')
1454 or ie_result
.get('_type', 'video') != 'video'
1455 or ie_result
.get('formats') or ie_result
.get('url')):
1458 format_dur
= lambda dur
: '%02d:%02d:%02d' % timetuple_from_msec(dur
* 1000)[:-1]
1463 self
.to_screen(msg
+ ' ' * (len(last_msg
) - len(msg
)) + '\r', skip_eol
=True)
1466 min_wait
, max_wait
= self
.params
.get('wait_for_video')
1467 diff
= try_get(ie_result
, lambda x
: x
['release_timestamp'] - time
.time())
1468 if diff
is None and ie_result
.get('live_status') == 'is_upcoming':
1469 diff
= round(random
.uniform(min_wait
, max_wait
) if (max_wait
and min_wait
) else (max_wait
or min_wait
), 0)
1470 self
.report_warning('Release time of video is not known')
1471 elif (diff
or 0) <= 0:
1472 self
.report_warning('Video should already be available according to extracted info')
1473 diff
= min(max(diff
or 0, min_wait
or 0), max_wait
or float('inf'))
1474 self
.to_screen(f
'[wait] Waiting for {format_dur(diff)} - Press Ctrl+C to try now')
1476 wait_till
= time
.time() + diff
1479 diff
= wait_till
- time
.time()
1482 raise ReExtractInfo('[wait] Wait period ended', expected
=True)
1483 progress(f
'[wait] Remaining time until next attempt: {self._format_screen(format_dur(diff), self.Styles.EMPHASIS)}')
1485 except KeyboardInterrupt:
1487 raise ReExtractInfo('[wait] Interrupted by user', expected
=True)
1488 except BaseException
as e
:
1489 if not isinstance(e
, ReExtractInfo
):
1493 @_handle_extraction_exceptions
1494 def __extract_info(self
, url
, ie
, download
, extra_info
, process
):
1495 ie_result
= ie
.extract(url
)
1496 if ie_result
is None: # Finished already (backwards compatibility; listformats and friends should be moved here)
1498 if isinstance(ie_result
, list):
1499 # Backwards compatibility: old IE result format
1501 '_type': 'compat_list',
1502 'entries': ie_result
,
1504 if extra_info
.get('original_url'):
1505 ie_result
.setdefault('original_url', extra_info
['original_url'])
1506 self
.add_default_extra_info(ie_result
, ie
, url
)
1508 self
._wait
_for
_video
(ie_result
)
1509 return self
.process_ie_result(ie_result
, download
, extra_info
)
1513 def add_default_extra_info(self
, ie_result
, ie
, url
):
1515 self
.add_extra_info(ie_result
, {
1517 'original_url': url
,
1519 webpage_url
= ie_result
.get('webpage_url')
1521 self
.add_extra_info(ie_result
, {
1522 'webpage_url_basename': url_basename(webpage_url
),
1523 'webpage_url_domain': get_domain(webpage_url
),
1526 self
.add_extra_info(ie_result
, {
1527 'extractor': ie
.IE_NAME
,
1528 'extractor_key': ie
.ie_key(),
1531 def process_ie_result(self
, ie_result
, download
=True, extra_info
=None):
1533 Take the result of the ie(may be modified) and resolve all unresolved
1534 references (URLs, playlist items).
1536 It will also download the videos if 'download'.
1537 Returns the resolved ie_result.
1539 if extra_info
is None:
1541 result_type
= ie_result
.get('_type', 'video')
1543 if result_type
in ('url', 'url_transparent'):
1544 ie_result
['url'] = sanitize_url(ie_result
['url'])
1545 if ie_result
.get('original_url'):
1546 extra_info
.setdefault('original_url', ie_result
['original_url'])
1548 extract_flat
= self
.params
.get('extract_flat', False)
1549 if ((extract_flat
== 'in_playlist' and 'playlist' in extra_info
)
1550 or extract_flat
is True):
1551 info_copy
= ie_result
.copy()
1552 ie
= try_get(ie_result
.get('ie_key'), self
.get_info_extractor
)
1553 if ie
and not ie_result
.get('id'):
1554 info_copy
['id'] = ie
.get_temp_id(ie_result
['url'])
1555 self
.add_default_extra_info(info_copy
, ie
, ie_result
['url'])
1556 self
.add_extra_info(info_copy
, extra_info
)
1557 info_copy
, _
= self
.pre_process(info_copy
)
1558 self
.__forced
_printings
(info_copy
, self
.prepare_filename(info_copy
), incomplete
=True)
1559 self
._raise
_pending
_errors
(info_copy
)
1560 if self
.params
.get('force_write_download_archive', False):
1561 self
.record_download_archive(info_copy
)
1564 if result_type
== 'video':
1565 self
.add_extra_info(ie_result
, extra_info
)
1566 ie_result
= self
.process_video_result(ie_result
, download
=download
)
1567 self
._raise
_pending
_errors
(ie_result
)
1568 additional_urls
= (ie_result
or {}).get('additional_urls')
1570 # TODO: Improve MetadataParserPP to allow setting a list
1571 if isinstance(additional_urls
, str):
1572 additional_urls
= [additional_urls
]
1574 '[info] %s: %d additional URL(s) requested' % (ie_result
['id'], len(additional_urls
)))
1575 self
.write_debug('Additional URLs: "%s"' % '", "'.join(additional_urls
))
1576 ie_result
['additional_entries'] = [
1578 url
, download
, extra_info
=extra_info
,
1579 force_generic_extractor
=self
.params
.get('force_generic_extractor'))
1580 for url
in additional_urls
1583 elif result_type
== 'url':
1584 # We have to add extra_info to the results because it may be
1585 # contained in a playlist
1586 return self
.extract_info(
1587 ie_result
['url'], download
,
1588 ie_key
=ie_result
.get('ie_key'),
1589 extra_info
=extra_info
)
1590 elif result_type
== 'url_transparent':
1591 # Use the information from the embedding page
1592 info
= self
.extract_info(
1593 ie_result
['url'], ie_key
=ie_result
.get('ie_key'),
1594 extra_info
=extra_info
, download
=False, process
=False)
1596 # extract_info may return None when ignoreerrors is enabled and
1597 # extraction failed with an error, don't crash and return early
1602 exempted_fields
= {'_type', 'url', 'ie_key'}
1603 if not ie_result
.get('section_end') and ie_result
.get('section_start') is None:
1604 # For video clips, the id etc of the clip extractor should be used
1605 exempted_fields |
= {'id', 'extractor', 'extractor_key'}
1607 new_result
= info
.copy()
1608 new_result
.update(filter_dict(ie_result
, lambda k
, v
: v
is not None and k
not in exempted_fields
))
1610 # Extracted info may not be a video result (i.e.
1611 # info.get('_type', 'video') != video) but rather an url or
1612 # url_transparent. In such cases outer metadata (from ie_result)
1613 # should be propagated to inner one (info). For this to happen
1614 # _type of info should be overridden with url_transparent. This
1615 # fixes issue from https://github.com/ytdl-org/youtube-dl/pull/11163.
1616 if new_result
.get('_type') == 'url':
1617 new_result
['_type'] = 'url_transparent'
1619 return self
.process_ie_result(
1620 new_result
, download
=download
, extra_info
=extra_info
)
1621 elif result_type
in ('playlist', 'multi_video'):
1622 # Protect from infinite recursion due to recursively nested playlists
1623 # (see https://github.com/ytdl-org/youtube-dl/issues/27833)
1624 webpage_url
= ie_result
['webpage_url']
1625 if webpage_url
in self
._playlist
_urls
:
1627 '[download] Skipping already downloaded playlist: %s'
1628 % ie_result
.get('title') or ie_result
.get('id'))
1631 self
._playlist
_level
+= 1
1632 self
._playlist
_urls
.add(webpage_url
)
1633 self
._fill
_common
_fields
(ie_result
, False)
1634 self
._sanitize
_thumbnails
(ie_result
)
1636 return self
.__process
_playlist
(ie_result
, download
)
1638 self
._playlist
_level
-= 1
1639 if not self
._playlist
_level
:
1640 self
._playlist
_urls
.clear()
1641 elif result_type
== 'compat_list':
1642 self
.report_warning(
1643 'Extractor %s returned a compat_list result. '
1644 'It needs to be updated.' % ie_result
.get('extractor'))
1647 self
.add_extra_info(r
, {
1648 'extractor': ie_result
['extractor'],
1649 'webpage_url': ie_result
['webpage_url'],
1650 'webpage_url_basename': url_basename(ie_result
['webpage_url']),
1651 'webpage_url_domain': get_domain(ie_result
['webpage_url']),
1652 'extractor_key': ie_result
['extractor_key'],
1655 ie_result
['entries'] = [
1656 self
.process_ie_result(_fixup(r
), download
, extra_info
)
1657 for r
in ie_result
['entries']
1661 raise Exception('Invalid result type: %s' % result_type
)
1663 def _ensure_dir_exists(self
, path
):
1664 return make_dir(path
, self
.report_error
)
1667 def _playlist_infodict(ie_result
, **kwargs
):
1670 'playlist': ie_result
.get('title') or ie_result
.get('id'),
1671 'playlist_id': ie_result
.get('id'),
1672 'playlist_title': ie_result
.get('title'),
1673 'playlist_uploader': ie_result
.get('uploader'),
1674 'playlist_uploader_id': ie_result
.get('uploader_id'),
1675 'playlist_index': 0,
1679 def __process_playlist(self
, ie_result
, download
):
1680 """Process each entry in the playlist"""
1681 title
= ie_result
.get('title') or ie_result
.get('id') or '<Untitled>'
1682 self
.to_screen(f
'[download] Downloading playlist: {title}')
1684 all_entries
= PlaylistEntries(self
, ie_result
)
1685 entries
= orderedSet(all_entries
.get_requested_items(), lazy
=True)
1687 lazy
= self
.params
.get('lazy_playlist')
1689 resolved_entries
, n_entries
= [], 'N/A'
1690 ie_result
['requested_entries'], ie_result
['entries'] = None, None
1692 entries
= resolved_entries
= list(entries
)
1693 n_entries
= len(resolved_entries
)
1694 ie_result
['requested_entries'], ie_result
['entries'] = tuple(zip(*resolved_entries
)) or ([], [])
1695 if not ie_result
.get('playlist_count'):
1696 # Better to do this after potentially exhausting entries
1697 ie_result
['playlist_count'] = all_entries
.get_full_count()
1699 _infojson_written
= False
1700 write_playlist_files
= self
.params
.get('allow_playlist_files', True)
1701 if write_playlist_files
and self
.params
.get('list_thumbnails'):
1702 self
.list_thumbnails(ie_result
)
1703 if write_playlist_files
and not self
.params
.get('simulate'):
1704 ie_copy
= self
._playlist
_infodict
(ie_result
, n_entries
=int_or_none(n_entries
))
1705 _infojson_written
= self
._write
_info
_json
(
1706 'playlist', ie_result
, self
.prepare_filename(ie_copy
, 'pl_infojson'))
1707 if _infojson_written
is None:
1709 if self
._write
_description
('playlist', ie_result
,
1710 self
.prepare_filename(ie_copy
, 'pl_description')) is None:
1712 # TODO: This should be passed to ThumbnailsConvertor if necessary
1713 self
._write
_thumbnails
('playlist', ie_copy
, self
.prepare_filename(ie_copy
, 'pl_thumbnail'))
1716 if self
.params
.get('playlistreverse') or self
.params
.get('playlistrandom'):
1717 self
.report_warning('playlistreverse and playlistrandom are not supported with lazy_playlist', only_once
=True)
1718 elif self
.params
.get('playlistreverse'):
1720 elif self
.params
.get('playlistrandom'):
1721 random
.shuffle(entries
)
1723 self
.to_screen(f
'[{ie_result["extractor"]}] Playlist {title}: Downloading {n_entries} videos'
1724 f
'{format_field(ie_result, "playlist_count", " of %s")}')
1727 max_failures
= self
.params
.get('skip_playlist_after_errors') or float('inf')
1728 for i
, (playlist_index
, entry
) in enumerate(entries
):
1730 resolved_entries
.append((playlist_index
, entry
))
1732 # TODO: Add auto-generated fields
1733 if not entry
or self
._match
_entry
(entry
, incomplete
=True) is not None:
1736 self
.to_screen('[download] Downloading video %s of %s' % (
1737 self
._format
_screen
(i
+ 1, self
.Styles
.ID
), self
._format
_screen
(n_entries
, self
.Styles
.EMPHASIS
)))
1739 entry
['__x_forwarded_for_ip'] = ie_result
.get('__x_forwarded_for_ip')
1740 if not lazy
and 'playlist-index' in self
.params
.get('compat_opts', []):
1741 playlist_index
= ie_result
['requested_entries'][i
]
1743 entry_result
= self
.__process
_iterable
_entry
(entry
, download
, {
1744 'n_entries': int_or_none(n_entries
),
1745 '__last_playlist_index': max(ie_result
['requested_entries'] or (0, 0)),
1746 'playlist_count': ie_result
.get('playlist_count'),
1747 'playlist_index': playlist_index
,
1748 'playlist_autonumber': i
+ 1,
1750 'playlist_id': ie_result
.get('id'),
1751 'playlist_title': ie_result
.get('title'),
1752 'playlist_uploader': ie_result
.get('uploader'),
1753 'playlist_uploader_id': ie_result
.get('uploader_id'),
1754 'extractor': ie_result
['extractor'],
1755 'webpage_url': ie_result
['webpage_url'],
1756 'webpage_url_basename': url_basename(ie_result
['webpage_url']),
1757 'webpage_url_domain': get_domain(ie_result
['webpage_url']),
1758 'extractor_key': ie_result
['extractor_key'],
1760 if not entry_result
:
1762 if failures
>= max_failures
:
1764 f
'Skipping the remaining entries in playlist "{title}" since {failures} items failed extraction')
1766 resolved_entries
[i
] = (playlist_index
, entry_result
)
1768 # Update with processed data
1769 ie_result
['requested_entries'], ie_result
['entries'] = tuple(zip(*resolved_entries
)) or ([], [])
1771 # Write the updated info to json
1772 if _infojson_written
is True and self
._write
_info
_json
(
1773 'updated playlist', ie_result
,
1774 self
.prepare_filename(ie_copy
, 'pl_infojson'), overwrite
=True) is None:
1777 ie_result
= self
.run_all_pps('playlist', ie_result
)
1778 self
.to_screen(f
'[download] Finished downloading playlist: {title}')
1781 @_handle_extraction_exceptions
1782 def __process_iterable_entry(self
, entry
, download
, extra_info
):
1783 return self
.process_ie_result(
1784 entry
, download
=download
, extra_info
=extra_info
)
1786 def _build_format_filter(self
, filter_spec
):
1787 " Returns a function to filter the formats according to the filter_spec "
1797 operator_rex
= re
.compile(r
'''(?x)\s*
1798 (?P<key>width|height|tbr|abr|vbr|asr|filesize|filesize_approx|fps)\s*
1799 (?P<op>%s)(?P<none_inclusive>\s*\?)?\s*
1800 (?P<value>[0-9.]+(?:[kKmMgGtTpPeEzZyY]i?[Bb]?)?)\s*
1801 ''' % '|'.join(map(re
.escape
, OPERATORS
.keys())))
1802 m
= operator_rex
.fullmatch(filter_spec
)
1805 comparison_value
= int(m
.group('value'))
1807 comparison_value
= parse_filesize(m
.group('value'))
1808 if comparison_value
is None:
1809 comparison_value
= parse_filesize(m
.group('value') + 'B')
1810 if comparison_value
is None:
1812 'Invalid value %r in format specification %r' % (
1813 m
.group('value'), filter_spec
))
1814 op
= OPERATORS
[m
.group('op')]
1819 '^=': lambda attr
, value
: attr
.startswith(value
),
1820 '$=': lambda attr
, value
: attr
.endswith(value
),
1821 '*=': lambda attr
, value
: value
in attr
,
1822 '~=': lambda attr
, value
: value
.search(attr
) is not None
1824 str_operator_rex
= re
.compile(r
'''(?x)\s*
1825 (?P<key>[a-zA-Z0-9._-]+)\s*
1826 (?P<negation>!\s*)?(?P<op>%s)\s*(?P<none_inclusive>\?\s*)?
1828 (?P<value>(?(quote)(?:(?!(?P=quote))[^\\]|\\.)+|[\w.-]+))
1829 (?(quote)(?P=quote))\s*
1830 ''' % '|'.join(map(re
.escape
, STR_OPERATORS
.keys())))
1831 m
= str_operator_rex
.fullmatch(filter_spec
)
1833 if m
.group('op') == '~=':
1834 comparison_value
= re
.compile(m
.group('value'))
1836 comparison_value
= re
.sub(r
'''\\([\\"'])''', r
'\1', m
.group('value'))
1837 str_op
= STR_OPERATORS
[m
.group('op')]
1838 if m
.group('negation'):
1839 op
= lambda attr
, value
: not str_op(attr
, value
)
1844 raise SyntaxError('Invalid filter specification %r' % filter_spec
)
1847 actual_value
= f
.get(m
.group('key'))
1848 if actual_value
is None:
1849 return m
.group('none_inclusive')
1850 return op(actual_value
, comparison_value
)
1853 def _check_formats(self
, formats
):
1855 self
.to_screen('[info] Testing format %s' % f
['format_id'])
1856 path
= self
.get_output_path('temp')
1857 if not self
._ensure
_dir
_exists
(f
'{path}/'):
1859 temp_file
= tempfile
.NamedTemporaryFile(suffix
='.tmp', delete
=False, dir=path
or None)
1862 success
, _
= self
.dl(temp_file
.name
, f
, test
=True)
1863 except (DownloadError
, OSError, ValueError) + network_exceptions
:
1866 if os
.path
.exists(temp_file
.name
):
1868 os
.remove(temp_file
.name
)
1870 self
.report_warning('Unable to delete temporary file "%s"' % temp_file
.name
)
1874 self
.to_screen('[info] Unable to download format %s. Skipping...' % f
['format_id'])
1876 def _default_format_spec(self
, info_dict
, download
=True):
1879 merger
= FFmpegMergerPP(self
)
1880 return merger
.available
and merger
.can_merge()
1883 not self
.params
.get('simulate')
1887 or info_dict
.get('is_live') and not self
.params
.get('live_from_start')
1888 or self
.params
['outtmpl']['default'] == '-'))
1891 or self
.params
.get('allow_multiple_audio_streams', False)
1892 or 'format-spec' in self
.params
['compat_opts'])
1895 'best/bestvideo+bestaudio' if prefer_best
1896 else 'bestvideo*+bestaudio/best' if not compat
1897 else 'bestvideo+bestaudio/best')
1899 def build_format_selector(self
, format_spec
):
1900 def syntax_error(note
, start
):
1902 'Invalid format specification: '
1903 '{}\n\t{}\n\t{}^'.format(note
, format_spec
, ' ' * start
[1]))
1904 return SyntaxError(message
)
1906 PICKFIRST
= 'PICKFIRST'
1910 FormatSelector
= collections
.namedtuple('FormatSelector', ['type', 'selector', 'filters'])
1912 allow_multiple_streams
= {'audio': self
.params
.get('allow_multiple_audio_streams', False),
1913 'video': self
.params
.get('allow_multiple_video_streams', False)}
1915 check_formats
= self
.params
.get('check_formats') == 'selected'
1917 def _parse_filter(tokens
):
1919 for type, string
, start
, _
, _
in tokens
:
1920 if type == tokenize
.OP
and string
== ']':
1921 return ''.join(filter_parts
)
1923 filter_parts
.append(string
)
1925 def _remove_unused_ops(tokens
):
1926 # Remove operators that we don't use and join them with the surrounding strings
1927 # for example: 'mp4' '-' 'baseline' '-' '16x9' is converted to 'mp4-baseline-16x9'
1928 ALLOWED_OPS
= ('/', '+', ',', '(', ')')
1929 last_string
, last_start
, last_end
, last_line
= None, None, None, None
1930 for type, string
, start
, end
, line
in tokens
:
1931 if type == tokenize
.OP
and string
== '[':
1933 yield tokenize
.NAME
, last_string
, last_start
, last_end
, last_line
1935 yield type, string
, start
, end
, line
1936 # everything inside brackets will be handled by _parse_filter
1937 for type, string
, start
, end
, line
in tokens
:
1938 yield type, string
, start
, end
, line
1939 if type == tokenize
.OP
and string
== ']':
1941 elif type == tokenize
.OP
and string
in ALLOWED_OPS
:
1943 yield tokenize
.NAME
, last_string
, last_start
, last_end
, last_line
1945 yield type, string
, start
, end
, line
1946 elif type in [tokenize
.NAME
, tokenize
.NUMBER
, tokenize
.OP
]:
1948 last_string
= string
1952 last_string
+= string
1954 yield tokenize
.NAME
, last_string
, last_start
, last_end
, last_line
1956 def _parse_format_selection(tokens
, inside_merge
=False, inside_choice
=False, inside_group
=False):
1958 current_selector
= None
1959 for type, string
, start
, _
, _
in tokens
:
1960 # ENCODING is only defined in python 3.x
1961 if type == getattr(tokenize
, 'ENCODING', None):
1963 elif type in [tokenize
.NAME
, tokenize
.NUMBER
]:
1964 current_selector
= FormatSelector(SINGLE
, string
, [])
1965 elif type == tokenize
.OP
:
1967 if not inside_group
:
1968 # ')' will be handled by the parentheses group
1969 tokens
.restore_last_token()
1971 elif inside_merge
and string
in ['/', ',']:
1972 tokens
.restore_last_token()
1974 elif inside_choice
and string
== ',':
1975 tokens
.restore_last_token()
1978 if not current_selector
:
1979 raise syntax_error('"," must follow a format selector', start
)
1980 selectors
.append(current_selector
)
1981 current_selector
= None
1983 if not current_selector
:
1984 raise syntax_error('"/" must follow a format selector', start
)
1985 first_choice
= current_selector
1986 second_choice
= _parse_format_selection(tokens
, inside_choice
=True)
1987 current_selector
= FormatSelector(PICKFIRST
, (first_choice
, second_choice
), [])
1989 if not current_selector
:
1990 current_selector
= FormatSelector(SINGLE
, 'best', [])
1991 format_filter
= _parse_filter(tokens
)
1992 current_selector
.filters
.append(format_filter
)
1994 if current_selector
:
1995 raise syntax_error('Unexpected "("', start
)
1996 group
= _parse_format_selection(tokens
, inside_group
=True)
1997 current_selector
= FormatSelector(GROUP
, group
, [])
1999 if not current_selector
:
2000 raise syntax_error('Unexpected "+"', start
)
2001 selector_1
= current_selector
2002 selector_2
= _parse_format_selection(tokens
, inside_merge
=True)
2004 raise syntax_error('Expected a selector', start
)
2005 current_selector
= FormatSelector(MERGE
, (selector_1
, selector_2
), [])
2007 raise syntax_error(f
'Operator not recognized: "{string}"', start
)
2008 elif type == tokenize
.ENDMARKER
:
2010 if current_selector
:
2011 selectors
.append(current_selector
)
2014 def _merge(formats_pair
):
2015 format_1
, format_2
= formats_pair
2018 formats_info
.extend(format_1
.get('requested_formats', (format_1
,)))
2019 formats_info
.extend(format_2
.get('requested_formats', (format_2
,)))
2021 if not allow_multiple_streams
['video'] or not allow_multiple_streams
['audio']:
2022 get_no_more
= {'video': False, 'audio': False}
2023 for (i
, fmt_info
) in enumerate(formats_info
):
2024 if fmt_info
.get('acodec') == fmt_info
.get('vcodec') == 'none':
2027 for aud_vid
in ['audio', 'video']:
2028 if not allow_multiple_streams
[aud_vid
] and fmt_info
.get(aud_vid
[0] + 'codec') != 'none':
2029 if get_no_more
[aud_vid
]:
2032 get_no_more
[aud_vid
] = True
2034 if len(formats_info
) == 1:
2035 return formats_info
[0]
2037 video_fmts
= [fmt_info
for fmt_info
in formats_info
if fmt_info
.get('vcodec') != 'none']
2038 audio_fmts
= [fmt_info
for fmt_info
in formats_info
if fmt_info
.get('acodec') != 'none']
2040 the_only_video
= video_fmts
[0] if len(video_fmts
) == 1 else None
2041 the_only_audio
= audio_fmts
[0] if len(audio_fmts
) == 1 else None
2043 output_ext
= self
.params
.get('merge_output_format')
2046 output_ext
= the_only_video
['ext']
2047 elif the_only_audio
and not video_fmts
:
2048 output_ext
= the_only_audio
['ext']
2052 filtered
= lambda *keys
: filter(None, (traverse_obj(fmt
, *keys
) for fmt
in formats_info
))
2055 'requested_formats': formats_info
,
2056 'format': '+'.join(filtered('format')),
2057 'format_id': '+'.join(filtered('format_id')),
2059 'protocol': '+'.join(map(determine_protocol
, formats_info
)),
2060 'language': '+'.join(orderedSet(filtered('language'))) or None,
2061 'format_note': '+'.join(orderedSet(filtered('format_note'))) or None,
2062 'filesize_approx': sum(filtered('filesize', 'filesize_approx')) or None,
2063 'tbr': sum(filtered('tbr', 'vbr', 'abr')),
2068 'width': the_only_video
.get('width'),
2069 'height': the_only_video
.get('height'),
2070 'resolution': the_only_video
.get('resolution') or self
.format_resolution(the_only_video
),
2071 'fps': the_only_video
.get('fps'),
2072 'dynamic_range': the_only_video
.get('dynamic_range'),
2073 'vcodec': the_only_video
.get('vcodec'),
2074 'vbr': the_only_video
.get('vbr'),
2075 'stretched_ratio': the_only_video
.get('stretched_ratio'),
2080 'acodec': the_only_audio
.get('acodec'),
2081 'abr': the_only_audio
.get('abr'),
2082 'asr': the_only_audio
.get('asr'),
2087 def _check_formats(formats
):
2088 if not check_formats
:
2091 yield from self
._check
_formats
(formats
)
2093 def _build_selector_function(selector
):
2094 if isinstance(selector
, list): # ,
2095 fs
= [_build_selector_function(s
) for s
in selector
]
2097 def selector_function(ctx
):
2100 return selector_function
2102 elif selector
.type == GROUP
: # ()
2103 selector_function
= _build_selector_function(selector
.selector
)
2105 elif selector
.type == PICKFIRST
: # /
2106 fs
= [_build_selector_function(s
) for s
in selector
.selector
]
2108 def selector_function(ctx
):
2110 picked_formats
= list(f(ctx
))
2112 return picked_formats
2115 elif selector
.type == MERGE
: # +
2116 selector_1
, selector_2
= map(_build_selector_function
, selector
.selector
)
2118 def selector_function(ctx
):
2119 for pair
in itertools
.product(selector_1(ctx
), selector_2(ctx
)):
2122 elif selector
.type == SINGLE
: # atom
2123 format_spec
= selector
.selector
or 'best'
2125 # TODO: Add allvideo, allaudio etc by generalizing the code with best/worst selector
2126 if format_spec
== 'all':
2127 def selector_function(ctx
):
2128 yield from _check_formats(ctx
['formats'][::-1])
2129 elif format_spec
== 'mergeall':
2130 def selector_function(ctx
):
2131 formats
= list(_check_formats(
2132 f
for f
in ctx
['formats'] if f
.get('vcodec') != 'none' or f
.get('acodec') != 'none'))
2135 merged_format
= formats
[-1]
2136 for f
in formats
[-2::-1]:
2137 merged_format
= _merge((merged_format
, f
))
2141 format_fallback
, seperate_fallback
, format_reverse
, format_idx
= False, None, True, 1
2143 r
'(?P<bw>best|worst|b|w)(?P<type>video|audio|v|a)?(?P<mod>\*)?(?:\.(?P<n>[1-9]\d*))?$',
2145 if mobj
is not None:
2146 format_idx
= int_or_none(mobj
.group('n'), default
=1)
2147 format_reverse
= mobj
.group('bw')[0] == 'b'
2148 format_type
= (mobj
.group('type') or [None])[0]
2149 not_format_type
= {'v': 'a', 'a': 'v'}
.get(format_type
)
2150 format_modified
= mobj
.group('mod') is not None
2152 format_fallback
= not format_type
and not format_modified
# for b, w
2154 (lambda f
: f
.get('%scodec' % format_type
) != 'none')
2155 if format_type
and format_modified
# bv*, ba*, wv*, wa*
2156 else (lambda f
: f
.get('%scodec' % not_format_type
) == 'none')
2157 if format_type
# bv, ba, wv, wa
2158 else (lambda f
: f
.get('vcodec') != 'none' and f
.get('acodec') != 'none')
2159 if not format_modified
# b, w
2160 else lambda f
: True) # b*, w*
2161 filter_f
= lambda f
: _filter_f(f
) and (
2162 f
.get('vcodec') != 'none' or f
.get('acodec') != 'none')
2164 if format_spec
in self
._format
_selection
_exts
['audio']:
2165 filter_f
= lambda f
: f
.get('ext') == format_spec
and f
.get('acodec') != 'none'
2166 elif format_spec
in self
._format
_selection
_exts
['video']:
2167 filter_f
= lambda f
: f
.get('ext') == format_spec
and f
.get('acodec') != 'none' and f
.get('vcodec') != 'none'
2168 seperate_fallback
= lambda f
: f
.get('ext') == format_spec
and f
.get('vcodec') != 'none'
2169 elif format_spec
in self
._format
_selection
_exts
['storyboards']:
2170 filter_f
= lambda f
: f
.get('ext') == format_spec
and f
.get('acodec') == 'none' and f
.get('vcodec') == 'none'
2172 filter_f
= lambda f
: f
.get('format_id') == format_spec
# id
2174 def selector_function(ctx
):
2175 formats
= list(ctx
['formats'])
2176 matches
= list(filter(filter_f
, formats
)) if filter_f
is not None else formats
2178 if format_fallback
and ctx
['incomplete_formats']:
2179 # for extractors with incomplete formats (audio only (soundcloud)
2180 # or video only (imgur)) best/worst will fallback to
2181 # best/worst {video,audio}-only format
2183 elif seperate_fallback
and not ctx
['has_merged_format']:
2184 # for compatibility with youtube-dl when there is no pre-merged format
2185 matches
= list(filter(seperate_fallback
, formats
))
2186 matches
= LazyList(_check_formats(matches
[::-1 if format_reverse
else 1]))
2188 yield matches
[format_idx
- 1]
2189 except LazyList
.IndexError:
2192 filters
= [self
._build
_format
_filter
(f
) for f
in selector
.filters
]
2194 def final_selector(ctx
):
2195 ctx_copy
= dict(ctx
)
2196 for _filter
in filters
:
2197 ctx_copy
['formats'] = list(filter(_filter
, ctx_copy
['formats']))
2198 return selector_function(ctx_copy
)
2199 return final_selector
2201 stream
= io
.BytesIO(format_spec
.encode())
2203 tokens
= list(_remove_unused_ops(tokenize
.tokenize(stream
.readline
)))
2204 except tokenize
.TokenError
:
2205 raise syntax_error('Missing closing/opening brackets or parenthesis', (0, len(format_spec
)))
2207 class TokenIterator
:
2208 def __init__(self
, tokens
):
2209 self
.tokens
= tokens
2216 if self
.counter
>= len(self
.tokens
):
2217 raise StopIteration()
2218 value
= self
.tokens
[self
.counter
]
2224 def restore_last_token(self
):
2227 parsed_selector
= _parse_format_selection(iter(TokenIterator(tokens
)))
2228 return _build_selector_function(parsed_selector
)
2230 def _calc_headers(self
, info_dict
):
2231 res
= merge_headers(self
.params
['http_headers'], info_dict
.get('http_headers') or {})
2233 cookies
= self
._calc
_cookies
(info_dict
['url'])
2235 res
['Cookie'] = cookies
2237 if 'X-Forwarded-For' not in res
:
2238 x_forwarded_for_ip
= info_dict
.get('__x_forwarded_for_ip')
2239 if x_forwarded_for_ip
:
2240 res
['X-Forwarded-For'] = x_forwarded_for_ip
2244 def _calc_cookies(self
, url
):
2245 pr
= sanitized_Request(url
)
2246 self
.cookiejar
.add_cookie_header(pr
)
2247 return pr
.get_header('Cookie')
2249 def _sort_thumbnails(self
, thumbnails
):
2250 thumbnails
.sort(key
=lambda t
: (
2251 t
.get('preference') if t
.get('preference') is not None else -1,
2252 t
.get('width') if t
.get('width') is not None else -1,
2253 t
.get('height') if t
.get('height') is not None else -1,
2254 t
.get('id') if t
.get('id') is not None else '',
2257 def _sanitize_thumbnails(self
, info_dict
):
2258 thumbnails
= info_dict
.get('thumbnails')
2259 if thumbnails
is None:
2260 thumbnail
= info_dict
.get('thumbnail')
2262 info_dict
['thumbnails'] = thumbnails
= [{'url': thumbnail}
]
2266 def check_thumbnails(thumbnails
):
2267 for t
in thumbnails
:
2268 self
.to_screen(f
'[info] Testing thumbnail {t["id"]}')
2270 self
.urlopen(HEADRequest(t
['url']))
2271 except network_exceptions
as err
:
2272 self
.to_screen(f
'[info] Unable to connect to thumbnail {t["id"]} URL {t["url"]!r} - {err}. Skipping...')
2276 self
._sort
_thumbnails
(thumbnails
)
2277 for i
, t
in enumerate(thumbnails
):
2278 if t
.get('id') is None:
2280 if t
.get('width') and t
.get('height'):
2281 t
['resolution'] = '%dx%d' % (t
['width'], t
['height'])
2282 t
['url'] = sanitize_url(t
['url'])
2284 if self
.params
.get('check_formats') is True:
2285 info_dict
['thumbnails'] = LazyList(check_thumbnails(thumbnails
[::-1]), reverse
=True)
2287 info_dict
['thumbnails'] = thumbnails
2289 def _fill_common_fields(self
, info_dict
, is_video
=True):
2290 # TODO: move sanitization here
2292 # playlists are allowed to lack "title"
2293 title
= info_dict
.get('title', NO_DEFAULT
)
2294 if title
is NO_DEFAULT
:
2295 raise ExtractorError('Missing "title" field in extractor result',
2296 video_id
=info_dict
['id'], ie
=info_dict
['extractor'])
2297 info_dict
['fulltitle'] = title
2300 self
.write_debug('Extractor gave empty title. Creating a generic title')
2302 self
.report_warning('Extractor failed to obtain "title". Creating a generic title instead')
2303 info_dict
['title'] = f
'{info_dict["extractor"].replace(":", "-")} video #{info_dict["id"]}'
2305 if info_dict
.get('duration') is not None:
2306 info_dict
['duration_string'] = formatSeconds(info_dict
['duration'])
2308 for ts_key
, date_key
in (
2309 ('timestamp', 'upload_date'),
2310 ('release_timestamp', 'release_date'),
2311 ('modified_timestamp', 'modified_date'),
2313 if info_dict
.get(date_key
) is None and info_dict
.get(ts_key
) is not None:
2314 # Working around out-of-range timestamp values (e.g. negative ones on Windows,
2315 # see http://bugs.python.org/issue1646728)
2316 with contextlib
.suppress(ValueError, OverflowError, OSError):
2317 upload_date
= datetime
.datetime
.utcfromtimestamp(info_dict
[ts_key
])
2318 info_dict
[date_key
] = upload_date
.strftime('%Y%m%d')
2320 live_keys
= ('is_live', 'was_live')
2321 live_status
= info_dict
.get('live_status')
2322 if live_status
is None:
2323 for key
in live_keys
:
2324 if info_dict
.get(key
) is False:
2326 if info_dict
.get(key
):
2329 if all(info_dict
.get(key
) is False for key
in live_keys
):
2330 live_status
= 'not_live'
2332 info_dict
['live_status'] = live_status
2333 for key
in live_keys
:
2334 if info_dict
.get(key
) is None:
2335 info_dict
[key
] = (live_status
== key
)
2337 # Auto generate title fields corresponding to the *_number fields when missing
2338 # in order to always have clean titles. This is very common for TV series.
2339 for field
in ('chapter', 'season', 'episode'):
2340 if info_dict
.get('%s_number' % field
) is not None and not info_dict
.get(field
):
2341 info_dict
[field
] = '%s %d' % (field
.capitalize(), info_dict
['%s_number' % field
])
2343 def _raise_pending_errors(self
, info
):
2344 err
= info
.pop('__pending_error', None)
2346 self
.report_error(err
, tb
=False)
2348 def process_video_result(self
, info_dict
, download
=True):
2349 assert info_dict
.get('_type', 'video') == 'video'
2350 self
._num
_videos
+= 1
2352 if 'id' not in info_dict
:
2353 raise ExtractorError('Missing "id" field in extractor result', ie
=info_dict
['extractor'])
2354 elif not info_dict
.get('id'):
2355 raise ExtractorError('Extractor failed to obtain "id"', ie
=info_dict
['extractor'])
2357 def report_force_conversion(field
, field_not
, conversion
):
2358 self
.report_warning(
2359 '"%s" field is not %s - forcing %s conversion, there is an error in extractor'
2360 % (field
, field_not
, conversion
))
2362 def sanitize_string_field(info
, string_field
):
2363 field
= info
.get(string_field
)
2364 if field
is None or isinstance(field
, str):
2366 report_force_conversion(string_field
, 'a string', 'string')
2367 info
[string_field
] = str(field
)
2369 def sanitize_numeric_fields(info
):
2370 for numeric_field
in self
._NUMERIC
_FIELDS
:
2371 field
= info
.get(numeric_field
)
2372 if field
is None or isinstance(field
, (int, float)):
2374 report_force_conversion(numeric_field
, 'numeric', 'int')
2375 info
[numeric_field
] = int_or_none(field
)
2377 sanitize_string_field(info_dict
, 'id')
2378 sanitize_numeric_fields(info_dict
)
2379 if info_dict
.get('section_end') and info_dict
.get('section_start') is not None:
2380 info_dict
['duration'] = round(info_dict
['section_end'] - info_dict
['section_start'], 3)
2381 if (info_dict
.get('duration') or 0) <= 0 and info_dict
.pop('duration', None):
2382 self
.report_warning('"duration" field is negative, there is an error in extractor')
2384 chapters
= info_dict
.get('chapters') or []
2385 if chapters
and chapters
[0].get('start_time'):
2386 chapters
.insert(0, {'start_time': 0}
)
2388 dummy_chapter
= {'end_time': 0, 'start_time': info_dict.get('duration')}
2389 for idx
, (prev
, current
, next_
) in enumerate(zip(
2390 (dummy_chapter
, *chapters
), chapters
, (*chapters
[1:], dummy_chapter
)), 1):
2391 if current
.get('start_time') is None:
2392 current
['start_time'] = prev
.get('end_time')
2393 if not current
.get('end_time'):
2394 current
['end_time'] = next_
.get('start_time')
2395 if not current
.get('title'):
2396 current
['title'] = f
'<Untitled Chapter {idx}>'
2398 if 'playlist' not in info_dict
:
2399 # It isn't part of a playlist
2400 info_dict
['playlist'] = None
2401 info_dict
['playlist_index'] = None
2403 self
._sanitize
_thumbnails
(info_dict
)
2405 thumbnail
= info_dict
.get('thumbnail')
2406 thumbnails
= info_dict
.get('thumbnails')
2408 info_dict
['thumbnail'] = sanitize_url(thumbnail
)
2410 info_dict
['thumbnail'] = thumbnails
[-1]['url']
2412 if info_dict
.get('display_id') is None and 'id' in info_dict
:
2413 info_dict
['display_id'] = info_dict
['id']
2415 self
._fill
_common
_fields
(info_dict
)
2417 for cc_kind
in ('subtitles', 'automatic_captions'):
2418 cc
= info_dict
.get(cc_kind
)
2420 for _
, subtitle
in cc
.items():
2421 for subtitle_format
in subtitle
:
2422 if subtitle_format
.get('url'):
2423 subtitle_format
['url'] = sanitize_url(subtitle_format
['url'])
2424 if subtitle_format
.get('ext') is None:
2425 subtitle_format
['ext'] = determine_ext(subtitle_format
['url']).lower()
2427 automatic_captions
= info_dict
.get('automatic_captions')
2428 subtitles
= info_dict
.get('subtitles')
2430 info_dict
['requested_subtitles'] = self
.process_subtitles(
2431 info_dict
['id'], subtitles
, automatic_captions
)
2433 if info_dict
.get('formats') is None:
2434 # There's only one format available
2435 formats
= [info_dict
]
2437 formats
= info_dict
['formats']
2439 # or None ensures --clean-infojson removes it
2440 info_dict
['_has_drm'] = any(f
.get('has_drm') for f
in formats
) or None
2441 if not self
.params
.get('allow_unplayable_formats'):
2442 formats
= [f
for f
in formats
if not f
.get('has_drm')]
2443 if info_dict
['_has_drm'] and all(
2444 f
.get('acodec') == f
.get('vcodec') == 'none' for f
in formats
):
2445 self
.report_warning(
2446 'This video is DRM protected and only images are available for download. '
2447 'Use --list-formats to see them')
2449 get_from_start
= not info_dict
.get('is_live') or bool(self
.params
.get('live_from_start'))
2450 if not get_from_start
:
2451 info_dict
['title'] += ' ' + datetime
.datetime
.now().strftime('%Y-%m-%d %H:%M')
2452 if info_dict
.get('is_live') and formats
:
2453 formats
= [f
for f
in formats
if bool(f
.get('is_from_start')) == get_from_start
]
2454 if get_from_start
and not formats
:
2455 self
.raise_no_formats(info_dict
, msg
=(
2456 '--live-from-start is passed, but there are no formats that can be downloaded from the start. '
2457 'If you want to download from the current time, use --no-live-from-start'))
2460 self
.raise_no_formats(info_dict
)
2462 def is_wellformed(f
):
2465 self
.report_warning(
2466 '"url" field is missing or empty - skipping format, '
2467 'there is an error in extractor')
2469 if isinstance(url
, bytes):
2470 sanitize_string_field(f
, 'url')
2473 # Filter out malformed formats for better extraction robustness
2474 formats
= list(filter(is_wellformed
, formats
))
2478 # We check that all the formats have the format and format_id fields
2479 for i
, format
in enumerate(formats
):
2480 sanitize_string_field(format
, 'format_id')
2481 sanitize_numeric_fields(format
)
2482 format
['url'] = sanitize_url(format
['url'])
2483 if not format
.get('format_id'):
2484 format
['format_id'] = str(i
)
2486 # Sanitize format_id from characters used in format selector expression
2487 format
['format_id'] = re
.sub(r
'[\s,/+\[\]()]', '_', format
['format_id'])
2488 format_id
= format
['format_id']
2489 if format_id
not in formats_dict
:
2490 formats_dict
[format_id
] = []
2491 formats_dict
[format_id
].append(format
)
2493 # Make sure all formats have unique format_id
2494 common_exts
= set(itertools
.chain(*self
._format
_selection
_exts
.values()))
2495 for format_id
, ambiguous_formats
in formats_dict
.items():
2496 ambigious_id
= len(ambiguous_formats
) > 1
2497 for i
, format
in enumerate(ambiguous_formats
):
2499 format
['format_id'] = '%s-%d' % (format_id
, i
)
2500 if format
.get('ext') is None:
2501 format
['ext'] = determine_ext(format
['url']).lower()
2502 # Ensure there is no conflict between id and ext in format selection
2503 # See https://github.com/yt-dlp/yt-dlp/issues/1282
2504 if format
['format_id'] != format
['ext'] and format
['format_id'] in common_exts
:
2505 format
['format_id'] = 'f%s' % format
['format_id']
2507 for i
, format
in enumerate(formats
):
2508 if format
.get('format') is None:
2509 format
['format'] = '{id} - {res}{note}'.format(
2510 id=format
['format_id'],
2511 res
=self
.format_resolution(format
),
2512 note
=format_field(format
, 'format_note', ' (%s)'),
2514 if format
.get('protocol') is None:
2515 format
['protocol'] = determine_protocol(format
)
2516 if format
.get('resolution') is None:
2517 format
['resolution'] = self
.format_resolution(format
, default
=None)
2518 if format
.get('dynamic_range') is None and format
.get('vcodec') != 'none':
2519 format
['dynamic_range'] = 'SDR'
2520 if (info_dict
.get('duration') and format
.get('tbr')
2521 and not format
.get('filesize') and not format
.get('filesize_approx')):
2522 format
['filesize_approx'] = int(info_dict
['duration'] * format
['tbr'] * (1024 / 8))
2524 # Add HTTP headers, so that external programs can use them from the
2526 full_format_info
= info_dict
.copy()
2527 full_format_info
.update(format
)
2528 format
['http_headers'] = self
._calc
_headers
(full_format_info
)
2529 # Remove private housekeeping stuff
2530 if '__x_forwarded_for_ip' in info_dict
:
2531 del info_dict
['__x_forwarded_for_ip']
2533 if self
.params
.get('check_formats') is True:
2534 formats
= LazyList(self
._check
_formats
(formats
[::-1]), reverse
=True)
2536 if not formats
or formats
[0] is not info_dict
:
2537 # only set the 'formats' fields if the original info_dict list them
2538 # otherwise we end up with a circular reference, the first (and unique)
2539 # element in the 'formats' field in info_dict is info_dict itself,
2540 # which can't be exported to json
2541 info_dict
['formats'] = formats
2543 info_dict
, _
= self
.pre_process(info_dict
)
2545 if self
._match
_entry
(info_dict
, incomplete
=self
._format
_fields
) is not None:
2548 self
.post_extract(info_dict
)
2549 info_dict
, _
= self
.pre_process(info_dict
, 'after_filter')
2551 # The pre-processors may have modified the formats
2552 formats
= info_dict
.get('formats', [info_dict
])
2554 list_only
= self
.params
.get('simulate') is None and (
2555 self
.params
.get('list_thumbnails') or self
.params
.get('listformats') or self
.params
.get('listsubtitles'))
2556 interactive_format_selection
= not list_only
and self
.format_selector
== '-'
2557 if self
.params
.get('list_thumbnails'):
2558 self
.list_thumbnails(info_dict
)
2559 if self
.params
.get('listsubtitles'):
2560 if 'automatic_captions' in info_dict
:
2561 self
.list_subtitles(
2562 info_dict
['id'], automatic_captions
, 'automatic captions')
2563 self
.list_subtitles(info_dict
['id'], subtitles
, 'subtitles')
2564 if self
.params
.get('listformats') or interactive_format_selection
:
2565 self
.list_formats(info_dict
)
2567 # Without this printing, -F --print-json will not work
2568 self
.__forced
_printings
(info_dict
, self
.prepare_filename(info_dict
), incomplete
=True)
2571 format_selector
= self
.format_selector
2572 if format_selector
is None:
2573 req_format
= self
._default
_format
_spec
(info_dict
, download
=download
)
2574 self
.write_debug('Default format spec: %s' % req_format
)
2575 format_selector
= self
.build_format_selector(req_format
)
2578 if interactive_format_selection
:
2580 self
._format
_screen
('\nEnter format selector: ', self
.Styles
.EMPHASIS
))
2582 format_selector
= self
.build_format_selector(req_format
)
2583 except SyntaxError as err
:
2584 self
.report_error(err
, tb
=False, is_error
=False)
2587 formats_to_download
= list(format_selector({
2589 'has_merged_format': any('none' not in (f
.get('acodec'), f
.get('vcodec')) for f
in formats
),
2590 'incomplete_formats': (
2591 # All formats are video-only or
2592 all(f
.get('vcodec') != 'none' and f
.get('acodec') == 'none' for f
in formats
)
2593 # all formats are audio-only
2594 or all(f
.get('vcodec') == 'none' and f
.get('acodec') != 'none' for f
in formats
)),
2596 if interactive_format_selection
and not formats_to_download
:
2597 self
.report_error('Requested format is not available', tb
=False, is_error
=False)
2601 if not formats_to_download
:
2602 if not self
.params
.get('ignore_no_formats_error'):
2603 raise ExtractorError(
2604 'Requested format is not available. Use --list-formats for a list of available formats',
2605 expected
=True, video_id
=info_dict
['id'], ie
=info_dict
['extractor'])
2606 self
.report_warning('Requested format is not available')
2607 # Process what we can, even without any available formats.
2608 formats_to_download
= [{}]
2610 requested_ranges
= self
.params
.get('download_ranges')
2611 if requested_ranges
:
2612 requested_ranges
= tuple(requested_ranges(info_dict
, self
))
2614 best_format
, downloaded_formats
= formats_to_download
[-1], []
2617 def to_screen(*msg
):
2618 self
.to_screen(f
'[info] {info_dict["id"]}: {" ".join(", ".join(variadic(m)) for m in msg)}')
2620 to_screen(f
'Downloading {len(formats_to_download)} format(s):',
2621 (f
['format_id'] for f
in formats_to_download
))
2622 if requested_ranges
:
2623 to_screen(f
'Downloading {len(requested_ranges)} time ranges:',
2624 (f
'{int(c["start_time"])}-{int(c["end_time"])}' for c
in requested_ranges
))
2625 max_downloads_reached
= False
2627 for fmt
, chapter
in itertools
.product(formats_to_download
, requested_ranges
or [{}]):
2628 new_info
= self
._copy
_infodict
(info_dict
)
2629 new_info
.update(fmt
)
2630 offset
, duration
= info_dict
.get('section_start') or 0, info_dict
.get('duration') or float('inf')
2631 if chapter
or offset
:
2633 'section_start': offset
+ chapter
.get('start_time', 0),
2634 'section_end': offset
+ min(chapter
.get('end_time', duration
), duration
),
2635 'section_title': chapter
.get('title'),
2636 'section_number': chapter
.get('index'),
2638 downloaded_formats
.append(new_info
)
2640 self
.process_info(new_info
)
2641 except MaxDownloadsReached
:
2642 max_downloads_reached
= True
2643 self
._raise
_pending
_errors
(new_info
)
2644 # Remove copied info
2645 for key
, val
in tuple(new_info
.items()):
2646 if info_dict
.get(key
) == val
:
2648 if max_downloads_reached
:
2651 write_archive
= {f.get('__write_download_archive', False) for f in downloaded_formats}
2652 assert write_archive
.issubset({True, False, 'ignore'}
)
2653 if True in write_archive
and False not in write_archive
:
2654 self
.record_download_archive(info_dict
)
2656 info_dict
['requested_downloads'] = downloaded_formats
2657 info_dict
= self
.run_all_pps('after_video', info_dict
)
2658 if max_downloads_reached
:
2659 raise MaxDownloadsReached()
2661 # We update the info dict with the selected best quality format (backwards compatibility)
2662 info_dict
.update(best_format
)
2665 def process_subtitles(self
, video_id
, normal_subtitles
, automatic_captions
):
2666 """Select the requested subtitles and their format"""
2667 available_subs
, normal_sub_langs
= {}, []
2668 if normal_subtitles
and self
.params
.get('writesubtitles'):
2669 available_subs
.update(normal_subtitles
)
2670 normal_sub_langs
= tuple(normal_subtitles
.keys())
2671 if automatic_captions
and self
.params
.get('writeautomaticsub'):
2672 for lang
, cap_info
in automatic_captions
.items():
2673 if lang
not in available_subs
:
2674 available_subs
[lang
] = cap_info
2676 if (not self
.params
.get('writesubtitles') and not
2677 self
.params
.get('writeautomaticsub') or not
2681 all_sub_langs
= tuple(available_subs
.keys())
2682 if self
.params
.get('allsubtitles', False):
2683 requested_langs
= all_sub_langs
2684 elif self
.params
.get('subtitleslangs', False):
2685 # A list is used so that the order of languages will be the same as
2686 # given in subtitleslangs. See https://github.com/yt-dlp/yt-dlp/issues/1041
2687 requested_langs
= []
2688 for lang_re
in self
.params
.get('subtitleslangs'):
2689 discard
= lang_re
[0] == '-'
2691 lang_re
= lang_re
[1:]
2692 if lang_re
== 'all':
2694 requested_langs
= []
2696 requested_langs
.extend(all_sub_langs
)
2698 current_langs
= filter(re
.compile(lang_re
+ '$').match
, all_sub_langs
)
2700 for lang
in current_langs
:
2701 while lang
in requested_langs
:
2702 requested_langs
.remove(lang
)
2704 requested_langs
.extend(current_langs
)
2705 requested_langs
= orderedSet(requested_langs
)
2706 elif normal_sub_langs
:
2707 requested_langs
= ['en'] if 'en' in normal_sub_langs
else normal_sub_langs
[:1]
2709 requested_langs
= ['en'] if 'en' in all_sub_langs
else all_sub_langs
[:1]
2711 self
.write_debug('Downloading subtitles: %s' % ', '.join(requested_langs
))
2713 formats_query
= self
.params
.get('subtitlesformat', 'best')
2714 formats_preference
= formats_query
.split('/') if formats_query
else []
2716 for lang
in requested_langs
:
2717 formats
= available_subs
.get(lang
)
2719 self
.report_warning(f
'{lang} subtitles not available for {video_id}')
2721 for ext
in formats_preference
:
2725 matches
= list(filter(lambda f
: f
['ext'] == ext
, formats
))
2731 self
.report_warning(
2732 'No subtitle format found matching "%s" for language %s, '
2733 'using %s' % (formats_query
, lang
, f
['ext']))
2737 def _forceprint(self
, key
, info_dict
):
2738 if info_dict
is None:
2740 info_copy
= info_dict
.copy()
2741 info_copy
['formats_table'] = self
.render_formats_table(info_dict
)
2742 info_copy
['thumbnails_table'] = self
.render_thumbnails_table(info_dict
)
2743 info_copy
['subtitles_table'] = self
.render_subtitles_table(info_dict
.get('id'), info_dict
.get('subtitles'))
2744 info_copy
['automatic_captions_table'] = self
.render_subtitles_table(info_dict
.get('id'), info_dict
.get('automatic_captions'))
2746 def format_tmpl(tmpl
):
2747 mobj
= re
.match(r
'\w+(=?)$', tmpl
)
2748 if mobj
and mobj
.group(1):
2749 return f
'{tmpl[:-1]} = %({tmpl[:-1]})r'
2751 return f
'%({tmpl})s'
2754 for tmpl
in self
.params
['forceprint'].get(key
, []):
2755 self
.to_stdout(self
.evaluate_outtmpl(format_tmpl(tmpl
), info_copy
))
2757 for tmpl
, file_tmpl
in self
.params
['print_to_file'].get(key
, []):
2758 filename
= self
.prepare_filename(info_dict
, outtmpl
=file_tmpl
)
2759 tmpl
= format_tmpl(tmpl
)
2760 self
.to_screen(f
'[info] Writing {tmpl!r} to: {filename}')
2761 if self
._ensure
_dir
_exists
(filename
):
2762 with open(filename
, 'a', encoding
='utf-8') as f
:
2763 f
.write(self
.evaluate_outtmpl(tmpl
, info_copy
) + '\n')
2765 def __forced_printings(self
, info_dict
, filename
, incomplete
):
2766 def print_mandatory(field
, actual_field
=None):
2767 if actual_field
is None:
2768 actual_field
= field
2769 if (self
.params
.get('force%s' % field
, False)
2770 and (not incomplete
or info_dict
.get(actual_field
) is not None)):
2771 self
.to_stdout(info_dict
[actual_field
])
2773 def print_optional(field
):
2774 if (self
.params
.get('force%s' % field
, False)
2775 and info_dict
.get(field
) is not None):
2776 self
.to_stdout(info_dict
[field
])
2778 info_dict
= info_dict
.copy()
2779 if filename
is not None:
2780 info_dict
['filename'] = filename
2781 if info_dict
.get('requested_formats') is not None:
2782 # For RTMP URLs, also include the playpath
2783 info_dict
['urls'] = '\n'.join(f
['url'] + f
.get('play_path', '') for f
in info_dict
['requested_formats'])
2784 elif info_dict
.get('url'):
2785 info_dict
['urls'] = info_dict
['url'] + info_dict
.get('play_path', '')
2787 if (self
.params
.get('forcejson')
2788 or self
.params
['forceprint'].get('video')
2789 or self
.params
['print_to_file'].get('video')):
2790 self
.post_extract(info_dict
)
2791 self
._forceprint
('video', info_dict
)
2793 print_mandatory('title')
2794 print_mandatory('id')
2795 print_mandatory('url', 'urls')
2796 print_optional('thumbnail')
2797 print_optional('description')
2798 print_optional('filename')
2799 if self
.params
.get('forceduration') and info_dict
.get('duration') is not None:
2800 self
.to_stdout(formatSeconds(info_dict
['duration']))
2801 print_mandatory('format')
2803 if self
.params
.get('forcejson'):
2804 self
.to_stdout(json
.dumps(self
.sanitize_info(info_dict
)))
2806 def dl(self
, name
, info
, subtitle
=False, test
=False):
2807 if not info
.get('url'):
2808 self
.raise_no_formats(info
, True)
2811 verbose
= self
.params
.get('verbose')
2814 'quiet': self
.params
.get('quiet') or not verbose
,
2816 'noprogress': not verbose
,
2818 'skip_unavailable_fragments': False,
2819 'keep_fragments': False,
2821 '_no_ytdl_file': True,
2824 params
= self
.params
2825 fd
= get_suitable_downloader(info
, params
, to_stdout
=(name
== '-'))(self
, params
)
2827 for ph
in self
._progress
_hooks
:
2828 fd
.add_progress_hook(ph
)
2830 (f
['url'].split(',')[0] + ',<data>' if f
['url'].startswith('data:') else f
['url'])
2831 for f
in info
.get('requested_formats', []) or [info
])
2832 self
.write_debug(f
'Invoking {fd.FD_NAME} downloader on "{urls}"')
2834 # Note: Ideally info should be a deep-copied so that hooks cannot modify it.
2835 # But it may contain objects that are not deep-copyable
2836 new_info
= self
._copy
_infodict
(info
)
2837 if new_info
.get('http_headers') is None:
2838 new_info
['http_headers'] = self
._calc
_headers
(new_info
)
2839 return fd
.download(name
, new_info
, subtitle
)
2841 def existing_file(self
, filepaths
, *, default_overwrite
=True):
2842 existing_files
= list(filter(os
.path
.exists
, orderedSet(filepaths
)))
2843 if existing_files
and not self
.params
.get('overwrites', default_overwrite
):
2844 return existing_files
[0]
2846 for file in existing_files
:
2847 self
.report_file_delete(file)
2851 def process_info(self
, info_dict
):
2852 """Process a single resolved IE result. (Modifies it in-place)"""
2854 assert info_dict
.get('_type', 'video') == 'video'
2855 original_infodict
= info_dict
2857 if 'format' not in info_dict
and 'ext' in info_dict
:
2858 info_dict
['format'] = info_dict
['ext']
2860 # This is mostly just for backward compatibility of process_info
2861 # As a side-effect, this allows for format-specific filters
2862 if self
._match
_entry
(info_dict
) is not None:
2863 info_dict
['__write_download_archive'] = 'ignore'
2866 # Does nothing under normal operation - for backward compatibility of process_info
2867 self
.post_extract(info_dict
)
2868 self
._num
_downloads
+= 1
2870 # info_dict['_filename'] needs to be set for backward compatibility
2871 info_dict
['_filename'] = full_filename
= self
.prepare_filename(info_dict
, warn
=True)
2872 temp_filename
= self
.prepare_filename(info_dict
, 'temp')
2876 self
.__forced
_printings
(info_dict
, full_filename
, incomplete
=('format' not in info_dict
))
2878 def check_max_downloads():
2879 if self
._num
_downloads
>= float(self
.params
.get('max_downloads') or 'inf'):
2880 raise MaxDownloadsReached()
2882 if self
.params
.get('simulate'):
2883 info_dict
['__write_download_archive'] = self
.params
.get('force_write_download_archive')
2884 check_max_downloads()
2887 if full_filename
is None:
2889 if not self
._ensure
_dir
_exists
(encodeFilename(full_filename
)):
2891 if not self
._ensure
_dir
_exists
(encodeFilename(temp_filename
)):
2894 if self
._write
_description
('video', info_dict
,
2895 self
.prepare_filename(info_dict
, 'description')) is None:
2898 sub_files
= self
._write
_subtitles
(info_dict
, temp_filename
)
2899 if sub_files
is None:
2901 files_to_move
.update(dict(sub_files
))
2903 thumb_files
= self
._write
_thumbnails
(
2904 'video', info_dict
, temp_filename
, self
.prepare_filename(info_dict
, 'thumbnail'))
2905 if thumb_files
is None:
2907 files_to_move
.update(dict(thumb_files
))
2909 infofn
= self
.prepare_filename(info_dict
, 'infojson')
2910 _infojson_written
= self
._write
_info
_json
('video', info_dict
, infofn
)
2911 if _infojson_written
:
2912 info_dict
['infojson_filename'] = infofn
2913 # For backward compatibility, even though it was a private field
2914 info_dict
['__infojson_filename'] = infofn
2915 elif _infojson_written
is None:
2918 # Note: Annotations are deprecated
2920 if self
.params
.get('writeannotations', False):
2921 annofn
= self
.prepare_filename(info_dict
, 'annotation')
2923 if not self
._ensure
_dir
_exists
(encodeFilename(annofn
)):
2925 if not self
.params
.get('overwrites', True) and os
.path
.exists(encodeFilename(annofn
)):
2926 self
.to_screen('[info] Video annotations are already present')
2927 elif not info_dict
.get('annotations'):
2928 self
.report_warning('There are no annotations to write.')
2931 self
.to_screen('[info] Writing video annotations to: ' + annofn
)
2932 with open(encodeFilename(annofn
), 'w', encoding
='utf-8') as annofile
:
2933 annofile
.write(info_dict
['annotations'])
2934 except (KeyError, TypeError):
2935 self
.report_warning('There are no annotations to write.')
2937 self
.report_error('Cannot write annotations file: ' + annofn
)
2940 # Write internet shortcut files
2941 def _write_link_file(link_type
):
2942 url
= try_get(info_dict
['webpage_url'], iri_to_uri
)
2944 self
.report_warning(
2945 f
'Cannot write internet shortcut file because the actual URL of "{info_dict["webpage_url"]}" is unknown')
2947 linkfn
= replace_extension(self
.prepare_filename(info_dict
, 'link'), link_type
, info_dict
.get('ext'))
2948 if not self
._ensure
_dir
_exists
(encodeFilename(linkfn
)):
2950 if self
.params
.get('overwrites', True) and os
.path
.exists(encodeFilename(linkfn
)):
2951 self
.to_screen(f
'[info] Internet shortcut (.{link_type}) is already present')
2954 self
.to_screen(f
'[info] Writing internet shortcut (.{link_type}) to: {linkfn}')
2955 with open(encodeFilename(to_high_limit_path(linkfn
)), 'w', encoding
='utf-8',
2956 newline
='\r\n' if link_type
== 'url' else '\n') as linkfile
:
2957 template_vars
= {'url': url}
2958 if link_type
== 'desktop':
2959 template_vars
['filename'] = linkfn
[:-(len(link_type
) + 1)]
2960 linkfile
.write(LINK_TEMPLATES
[link_type
] % template_vars
)
2962 self
.report_error(f
'Cannot write internet shortcut {linkfn}')
2967 'url': self
.params
.get('writeurllink'),
2968 'webloc': self
.params
.get('writewebloclink'),
2969 'desktop': self
.params
.get('writedesktoplink'),
2971 if self
.params
.get('writelink'):
2972 link_type
= ('webloc' if sys
.platform
== 'darwin'
2973 else 'desktop' if sys
.platform
.startswith('linux')
2975 write_links
[link_type
] = True
2977 if any(should_write
and not _write_link_file(link_type
)
2978 for link_type
, should_write
in write_links
.items()):
2981 def replace_info_dict(new_info
):
2983 if new_info
== info_dict
:
2986 info_dict
.update(new_info
)
2988 new_info
, files_to_move
= self
.pre_process(info_dict
, 'before_dl', files_to_move
)
2989 replace_info_dict(new_info
)
2991 if self
.params
.get('skip_download'):
2992 info_dict
['filepath'] = temp_filename
2993 info_dict
['__finaldir'] = os
.path
.dirname(os
.path
.abspath(encodeFilename(full_filename
)))
2994 info_dict
['__files_to_move'] = files_to_move
2995 replace_info_dict(self
.run_pp(MoveFilesAfterDownloadPP(self
, False), info_dict
))
2996 info_dict
['__write_download_archive'] = self
.params
.get('force_write_download_archive')
2999 info_dict
.setdefault('__postprocessors', [])
3002 def existing_video_file(*filepaths
):
3003 ext
= info_dict
.get('ext')
3004 converted
= lambda file: replace_extension(file, self
.params
.get('final_ext') or ext
, ext
)
3005 file = self
.existing_file(itertools
.chain(*zip(map(converted
, filepaths
), filepaths
)),
3006 default_overwrite
=False)
3008 info_dict
['ext'] = os
.path
.splitext(file)[1][1:]
3011 fd
, success
= None, True
3012 if info_dict
.get('protocol') or info_dict
.get('url'):
3013 fd
= get_suitable_downloader(info_dict
, self
.params
, to_stdout
=temp_filename
== '-')
3014 if fd
is not FFmpegFD
and (
3015 info_dict
.get('section_start') or info_dict
.get('section_end')):
3016 msg
= ('This format cannot be partially downloaded' if FFmpegFD
.available()
3017 else 'You have requested downloading the video partially, but ffmpeg is not installed')
3018 self
.report_error(f
'{msg}. Aborting')
3021 if info_dict
.get('requested_formats') is not None:
3023 def compatible_formats(formats
):
3024 # TODO: some formats actually allow this (mkv, webm, ogg, mp4), but not all of them.
3025 video_formats
= [format
for format
in formats
if format
.get('vcodec') != 'none']
3026 audio_formats
= [format
for format
in formats
if format
.get('acodec') != 'none']
3027 if len(video_formats
) > 2 or len(audio_formats
) > 2:
3031 exts
= {format.get('ext') for format in formats}
3033 {'mp3', 'mp4', 'm4a', 'm4p', 'm4b', 'm4r', 'm4v', 'ismv', 'isma'}
,
3036 for ext_sets
in COMPATIBLE_EXTS
:
3037 if ext_sets
.issuperset(exts
):
3039 # TODO: Check acodec/vcodec
3042 requested_formats
= info_dict
['requested_formats']
3043 old_ext
= info_dict
['ext']
3044 if self
.params
.get('merge_output_format') is None:
3045 if not compatible_formats(requested_formats
):
3046 info_dict
['ext'] = 'mkv'
3047 self
.report_warning(
3048 'Requested formats are incompatible for merge and will be merged into mkv')
3049 if (info_dict
['ext'] == 'webm'
3050 and info_dict
.get('thumbnails')
3051 # check with type instead of pp_key, __name__, or isinstance
3052 # since we dont want any custom PPs to trigger this
3053 and any(type(pp
) == EmbedThumbnailPP
for pp
in self
._pps
['post_process'])): # noqa: E721
3054 info_dict
['ext'] = 'mkv'
3055 self
.report_warning(
3056 'webm doesn\'t support embedding a thumbnail, mkv will be used')
3057 new_ext
= info_dict
['ext']
3059 def correct_ext(filename
, ext
=new_ext
):
3062 filename_real_ext
= os
.path
.splitext(filename
)[1][1:]
3064 os
.path
.splitext(filename
)[0]
3065 if filename_real_ext
in (old_ext
, new_ext
)
3067 return f
'{filename_wo_ext}.{ext}'
3069 # Ensure filename always has a correct extension for successful merge
3070 full_filename
= correct_ext(full_filename
)
3071 temp_filename
= correct_ext(temp_filename
)
3072 dl_filename
= existing_video_file(full_filename
, temp_filename
)
3073 info_dict
['__real_download'] = False
3075 merger
= FFmpegMergerPP(self
)
3077 if dl_filename
is not None:
3078 self
.report_file_already_downloaded(dl_filename
)
3080 for f
in requested_formats
if fd
!= FFmpegFD
else []:
3081 f
['filepath'] = fname
= prepend_extension(
3082 correct_ext(temp_filename
, info_dict
['ext']),
3083 'f%s' % f
['format_id'], info_dict
['ext'])
3084 downloaded
.append(fname
)
3085 info_dict
['url'] = '\n'.join(f
['url'] for f
in requested_formats
)
3086 success
, real_download
= self
.dl(temp_filename
, info_dict
)
3087 info_dict
['__real_download'] = real_download
3089 if self
.params
.get('allow_unplayable_formats'):
3090 self
.report_warning(
3091 'You have requested merging of multiple formats '
3092 'while also allowing unplayable formats to be downloaded. '
3093 'The formats won\'t be merged to prevent data corruption.')
3094 elif not merger
.available
:
3095 msg
= 'You have requested merging of multiple formats but ffmpeg is not installed'
3096 if not self
.params
.get('ignoreerrors'):
3097 self
.report_error(f
'{msg}. Aborting due to --abort-on-error')
3099 self
.report_warning(f
'{msg}. The formats won\'t be merged')
3101 if temp_filename
== '-':
3102 reason
= ('using a downloader other than ffmpeg' if FFmpegFD
.can_merge_formats(info_dict
, self
.params
)
3103 else 'but the formats are incompatible for simultaneous download' if merger
.available
3104 else 'but ffmpeg is not installed')
3105 self
.report_warning(
3106 f
'You have requested downloading multiple formats to stdout {reason}. '
3107 'The formats will be streamed one after the other')
3108 fname
= temp_filename
3109 for f
in requested_formats
:
3110 new_info
= dict(info_dict
)
3111 del new_info
['requested_formats']
3113 if temp_filename
!= '-':
3114 fname
= prepend_extension(
3115 correct_ext(temp_filename
, new_info
['ext']),
3116 'f%s' % f
['format_id'], new_info
['ext'])
3117 if not self
._ensure
_dir
_exists
(fname
):
3119 f
['filepath'] = fname
3120 downloaded
.append(fname
)
3121 partial_success
, real_download
= self
.dl(fname
, new_info
)
3122 info_dict
['__real_download'] = info_dict
['__real_download'] or real_download
3123 success
= success
and partial_success
3125 if downloaded
and merger
.available
and not self
.params
.get('allow_unplayable_formats'):
3126 info_dict
['__postprocessors'].append(merger
)
3127 info_dict
['__files_to_merge'] = downloaded
3128 # Even if there were no downloads, it is being merged only now
3129 info_dict
['__real_download'] = True
3131 for file in downloaded
:
3132 files_to_move
[file] = None
3134 # Just a single file
3135 dl_filename
= existing_video_file(full_filename
, temp_filename
)
3136 if dl_filename
is None or dl_filename
== temp_filename
:
3137 # dl_filename == temp_filename could mean that the file was partially downloaded with --no-part.
3138 # So we should try to resume the download
3139 success
, real_download
= self
.dl(temp_filename
, info_dict
)
3140 info_dict
['__real_download'] = real_download
3142 self
.report_file_already_downloaded(dl_filename
)
3144 dl_filename
= dl_filename
or temp_filename
3145 info_dict
['__finaldir'] = os
.path
.dirname(os
.path
.abspath(encodeFilename(full_filename
)))
3147 except network_exceptions
as err
:
3148 self
.report_error('unable to download video data: %s' % error_to_compat_str(err
))
3150 except OSError as err
:
3151 raise UnavailableVideoError(err
)
3152 except (ContentTooShortError
, ) as err
:
3153 self
.report_error(f
'content too short (expected {err.expected} bytes and served {err.downloaded})')
3156 self
._raise
_pending
_errors
(info_dict
)
3157 if success
and full_filename
!= '-':
3161 fixup_policy
= self
.params
.get('fixup')
3162 vid
= info_dict
['id']
3164 if fixup_policy
in ('ignore', 'never'):
3166 elif fixup_policy
== 'warn':
3168 elif fixup_policy
!= 'force':
3169 assert fixup_policy
in ('detect_or_warn', None)
3170 if not info_dict
.get('__real_download'):
3173 def ffmpeg_fixup(cndn
, msg
, cls
):
3174 if not (do_fixup
and cndn
):
3176 elif do_fixup
== 'warn':
3177 self
.report_warning(f
'{vid}: {msg}')
3181 info_dict
['__postprocessors'].append(pp
)
3183 self
.report_warning(f
'{vid}: {msg}. Install ffmpeg to fix this automatically')
3185 stretched_ratio
= info_dict
.get('stretched_ratio')
3186 ffmpeg_fixup(stretched_ratio
not in (1, None),
3187 f
'Non-uniform pixel ratio {stretched_ratio}',
3188 FFmpegFixupStretchedPP
)
3190 downloader
= get_suitable_downloader(info_dict
, self
.params
) if 'protocol' in info_dict
else None
3191 downloader
= downloader
.FD_NAME
if downloader
else None
3193 ext
= info_dict
.get('ext')
3194 postprocessed_by_ffmpeg
= info_dict
.get('requested_formats') or any((
3195 isinstance(pp
, FFmpegVideoConvertorPP
)
3196 and resolve_recode_mapping(ext
, pp
.mapping
)[0] not in (ext
, None)
3197 ) for pp
in self
._pps
['post_process'])
3199 if not postprocessed_by_ffmpeg
:
3200 ffmpeg_fixup(ext
== 'm4a' and info_dict
.get('container') == 'm4a_dash',
3201 'writing DASH m4a. Only some players support this container',
3203 ffmpeg_fixup(downloader
== 'hlsnative' and not self
.params
.get('hls_use_mpegts')
3204 or info_dict
.get('is_live') and self
.params
.get('hls_use_mpegts') is None,
3205 'Possible MPEG-TS in MP4 container or malformed AAC timestamps',
3207 ffmpeg_fixup(info_dict
.get('is_live') and downloader
== 'DashSegmentsFD',
3208 'Possible duplicate MOOV atoms', FFmpegFixupDuplicateMoovPP
)
3210 ffmpeg_fixup(downloader
== 'web_socket_fragment', 'Malformed timestamps detected', FFmpegFixupTimestampPP
)
3211 ffmpeg_fixup(downloader
== 'web_socket_fragment', 'Malformed duration detected', FFmpegFixupDurationPP
)
3215 replace_info_dict(self
.post_process(dl_filename
, info_dict
, files_to_move
))
3216 except PostProcessingError
as err
:
3217 self
.report_error('Postprocessing: %s' % str(err
))
3220 for ph
in self
._post
_hooks
:
3221 ph(info_dict
['filepath'])
3222 except Exception as err
:
3223 self
.report_error('post hooks: %s' % str(err
))
3225 info_dict
['__write_download_archive'] = True
3227 assert info_dict
is original_infodict
# Make sure the info_dict was modified in-place
3228 if self
.params
.get('force_write_download_archive'):
3229 info_dict
['__write_download_archive'] = True
3230 check_max_downloads()
3232 def __download_wrapper(self
, func
):
3233 @functools.wraps(func
)
3234 def wrapper(*args
, **kwargs
):
3236 res
= func(*args
, **kwargs
)
3237 except UnavailableVideoError
as e
:
3238 self
.report_error(e
)
3239 except DownloadCancelled
as e
:
3240 self
.to_screen(f
'[info] {e}')
3241 if not self
.params
.get('break_per_url'):
3244 if self
.params
.get('dump_single_json', False):
3245 self
.post_extract(res
)
3246 self
.to_stdout(json
.dumps(self
.sanitize_info(res
)))
3249 def download(self
, url_list
):
3250 """Download a given list of URLs."""
3251 url_list
= variadic(url_list
) # Passing a single URL is a common mistake
3252 outtmpl
= self
.params
['outtmpl']['default']
3253 if (len(url_list
) > 1
3255 and '%' not in outtmpl
3256 and self
.params
.get('max_downloads') != 1):
3257 raise SameFileError(outtmpl
)
3259 for url
in url_list
:
3260 self
.__download
_wrapper
(self
.extract_info
)(
3261 url
, force_generic_extractor
=self
.params
.get('force_generic_extractor', False))
3263 return self
._download
_retcode
3265 def download_with_info_file(self
, info_filename
):
3266 with contextlib
.closing(fileinput
.FileInput(
3267 [info_filename
], mode
='r',
3268 openhook
=fileinput
.hook_encoded('utf-8'))) as f
:
3269 # FileInput doesn't have a read method, we can't call json.load
3270 info
= self
.sanitize_info(json
.loads('\n'.join(f
)), self
.params
.get('clean_infojson', True))
3272 self
.__download
_wrapper
(self
.process_ie_result
)(info
, download
=True)
3273 except (DownloadError
, EntryNotInPlaylist
, ReExtractInfo
) as e
:
3274 if not isinstance(e
, EntryNotInPlaylist
):
3275 self
.to_stderr('\r')
3276 webpage_url
= info
.get('webpage_url')
3277 if webpage_url
is not None:
3278 self
.report_warning(f
'The info failed to download: {e}; trying with URL {webpage_url}')
3279 return self
.download([webpage_url
])
3282 return self
._download
_retcode
3285 def sanitize_info(info_dict
, remove_private_keys
=False):
3286 ''' Sanitize the infodict for converting to json '''
3287 if info_dict
is None:
3289 info_dict
.setdefault('epoch', int(time
.time()))
3290 info_dict
.setdefault('_type', 'video')
3292 if remove_private_keys
:
3293 reject
= lambda k
, v
: v
is None or k
.startswith('__') or k
in {
3294 'requested_downloads', 'requested_formats', 'requested_subtitles', 'requested_entries',
3295 'entries', 'filepath', '_filename', 'infojson_filename', 'original_url', 'playlist_autonumber',
3298 reject
= lambda k
, v
: False
3301 if isinstance(obj
, dict):
3302 return {k: filter_fn(v) for k, v in obj.items() if not reject(k, v)}
3303 elif isinstance(obj
, (list, tuple, set, LazyList
)):
3304 return list(map(filter_fn
, obj
))
3305 elif obj
is None or isinstance(obj
, (str, int, float, bool)):
3310 return filter_fn(info_dict
)
3313 def filter_requested_info(info_dict
, actually_filter
=True):
3314 ''' Alias of sanitize_info for backward compatibility '''
3315 return YoutubeDL
.sanitize_info(info_dict
, actually_filter
)
3317 def _delete_downloaded_files(self
, *files_to_delete
, info
={}, msg
=None):
3318 for filename
in set(filter(None, files_to_delete
)):
3320 self
.to_screen(msg
% filename
)
3324 self
.report_warning(f
'Unable to delete file {filename}')
3325 if filename
in info
.get('__files_to_move', []): # NB: Delete even if None
3326 del info
['__files_to_move'][filename
]
3329 def post_extract(info_dict
):
3330 def actual_post_extract(info_dict
):
3331 if info_dict
.get('_type') in ('playlist', 'multi_video'):
3332 for video_dict
in info_dict
.get('entries', {}):
3333 actual_post_extract(video_dict
or {})
3336 post_extractor
= info_dict
.pop('__post_extractor', None) or (lambda: {})
3337 info_dict
.update(post_extractor())
3339 actual_post_extract(info_dict
or {})
3341 def run_pp(self
, pp
, infodict
):
3342 files_to_delete
= []
3343 if '__files_to_move' not in infodict
:
3344 infodict
['__files_to_move'] = {}
3346 files_to_delete
, infodict
= pp
.run(infodict
)
3347 except PostProcessingError
as e
:
3348 # Must be True and not 'only_download'
3349 if self
.params
.get('ignoreerrors') is True:
3350 self
.report_error(e
)
3354 if not files_to_delete
:
3356 if self
.params
.get('keepvideo', False):
3357 for f
in files_to_delete
:
3358 infodict
['__files_to_move'].setdefault(f
, '')
3360 self
._delete
_downloaded
_files
(
3361 *files_to_delete
, info
=infodict
, msg
='Deleting original file %s (pass -k to keep)')
3364 def run_all_pps(self
, key
, info
, *, additional_pps
=None):
3365 self
._forceprint
(key
, info
)
3366 for pp
in (additional_pps
or []) + self
._pps
[key
]:
3367 info
= self
.run_pp(pp
, info
)
3370 def pre_process(self
, ie_info
, key
='pre_process', files_to_move
=None):
3371 info
= dict(ie_info
)
3372 info
['__files_to_move'] = files_to_move
or {}
3374 info
= self
.run_all_pps(key
, info
)
3375 except PostProcessingError
as err
:
3376 msg
= f
'Preprocessing: {err}'
3377 info
.setdefault('__pending_error', msg
)
3378 self
.report_error(msg
, is_error
=False)
3379 return info
, info
.pop('__files_to_move', None)
3381 def post_process(self
, filename
, info
, files_to_move
=None):
3382 """Run all the postprocessors on the given file."""
3383 info
['filepath'] = filename
3384 info
['__files_to_move'] = files_to_move
or {}
3385 info
= self
.run_all_pps('post_process', info
, additional_pps
=info
.get('__postprocessors'))
3386 info
= self
.run_pp(MoveFilesAfterDownloadPP(self
), info
)
3387 del info
['__files_to_move']
3388 return self
.run_all_pps('after_move', info
)
3390 def _make_archive_id(self
, info_dict
):
3391 video_id
= info_dict
.get('id')
3394 # Future-proof against any change in case
3395 # and backwards compatibility with prior versions
3396 extractor
= info_dict
.get('extractor_key') or info_dict
.get('ie_key') # key in a playlist
3397 if extractor
is None:
3398 url
= str_or_none(info_dict
.get('url'))
3401 # Try to find matching extractor for the URL and take its ie_key
3402 for ie_key
, ie
in self
._ies
.items():
3403 if ie
.suitable(url
):
3408 return f
'{extractor.lower()} {video_id}'
3410 def in_download_archive(self
, info_dict
):
3411 fn
= self
.params
.get('download_archive')
3415 vid_id
= self
._make
_archive
_id
(info_dict
)
3417 return False # Incomplete video information
3419 return vid_id
in self
.archive
3421 def record_download_archive(self
, info_dict
):
3422 fn
= self
.params
.get('download_archive')
3425 vid_id
= self
._make
_archive
_id
(info_dict
)
3427 self
.write_debug(f
'Adding to archive: {vid_id}')
3428 with locked_file(fn
, 'a', encoding
='utf-8') as archive_file
:
3429 archive_file
.write(vid_id
+ '\n')
3430 self
.archive
.add(vid_id
)
3433 def format_resolution(format
, default
='unknown'):
3434 if format
.get('vcodec') == 'none' and format
.get('acodec') != 'none':
3436 if format
.get('resolution') is not None:
3437 return format
['resolution']
3438 if format
.get('width') and format
.get('height'):
3439 return '%dx%d' % (format
['width'], format
['height'])
3440 elif format
.get('height'):
3441 return '%sp' % format
['height']
3442 elif format
.get('width'):
3443 return '%dx?' % format
['width']
3446 def _list_format_headers(self
, *headers
):
3447 if self
.params
.get('listformats_table', True) is not False:
3448 return [self
._format
_out
(header
, self
.Styles
.HEADERS
) for header
in headers
]
3451 def _format_note(self
, fdict
):
3453 if fdict
.get('ext') in ['f4f', 'f4m']:
3454 res
+= '(unsupported)'
3455 if fdict
.get('language'):
3458 res
+= '[%s]' % fdict
['language']
3459 if fdict
.get('format_note') is not None:
3462 res
+= fdict
['format_note']
3463 if fdict
.get('tbr') is not None:
3466 res
+= '%4dk' % fdict
['tbr']
3467 if fdict
.get('container') is not None:
3470 res
+= '%s container' % fdict
['container']
3471 if (fdict
.get('vcodec') is not None
3472 and fdict
.get('vcodec') != 'none'):
3475 res
+= fdict
['vcodec']
3476 if fdict
.get('vbr') is not None:
3478 elif fdict
.get('vbr') is not None and fdict
.get('abr') is not None:
3480 if fdict
.get('vbr') is not None:
3481 res
+= '%4dk' % fdict
['vbr']
3482 if fdict
.get('fps') is not None:
3485 res
+= '%sfps' % fdict
['fps']
3486 if fdict
.get('acodec') is not None:
3489 if fdict
['acodec'] == 'none':
3492 res
+= '%-5s' % fdict
['acodec']
3493 elif fdict
.get('abr') is not None:
3497 if fdict
.get('abr') is not None:
3498 res
+= '@%3dk' % fdict
['abr']
3499 if fdict
.get('asr') is not None:
3500 res
+= ' (%5dHz)' % fdict
['asr']
3501 if fdict
.get('filesize') is not None:
3504 res
+= format_bytes(fdict
['filesize'])
3505 elif fdict
.get('filesize_approx') is not None:
3508 res
+= '~' + format_bytes(fdict
['filesize_approx'])
3511 def render_formats_table(self
, info_dict
):
3512 if not info_dict
.get('formats') and not info_dict
.get('url'):
3515 formats
= info_dict
.get('formats', [info_dict
])
3516 if not self
.params
.get('listformats_table', True) is not False:
3519 format_field(f
, 'format_id'),
3520 format_field(f
, 'ext'),
3521 self
.format_resolution(f
),
3522 self
._format
_note
(f
)
3523 ] for f
in formats
if f
.get('preference') is None or f
['preference'] >= -1000]
3524 return render_table(['format code', 'extension', 'resolution', 'note'], table
, extra_gap
=1)
3526 delim
= self
._format
_out
('\u2502', self
.Styles
.DELIM
, '|', test_encoding
=True)
3529 self
._format
_out
(format_field(f
, 'format_id'), self
.Styles
.ID
),
3530 format_field(f
, 'ext'),
3531 format_field(f
, func
=self
.format_resolution
, ignore
=('audio only', 'images')),
3532 format_field(f
, 'fps', '\t%d'),
3533 format_field(f
, 'dynamic_range', '%s', ignore
=(None, 'SDR')).replace('HDR', ''),
3535 format_field(f
, 'filesize', ' \t%s', func
=format_bytes
) + format_field(f
, 'filesize_approx', '~\t%s', func
=format_bytes
),
3536 format_field(f
, 'tbr', '\t%dk'),
3537 shorten_protocol_name(f
.get('protocol', '')),
3539 format_field(f
, 'vcodec', default
='unknown').replace(
3540 'none', 'images' if f
.get('acodec') == 'none'
3541 else self
._format
_out
('audio only', self
.Styles
.SUPPRESS
)),
3542 format_field(f
, 'vbr', '\t%dk'),
3543 format_field(f
, 'acodec', default
='unknown').replace(
3544 'none', '' if f
.get('vcodec') == 'none'
3545 else self
._format
_out
('video only', self
.Styles
.SUPPRESS
)),
3546 format_field(f
, 'abr', '\t%dk'),
3547 format_field(f
, 'asr', '\t%s', func
=format_decimal_suffix
),
3549 self
._format
_out
('UNSUPPORTED', 'light red') if f
.get('ext') in ('f4f', 'f4m') else None,
3550 format_field(f
, 'language', '[%s]'),
3551 join_nonempty(format_field(f
, 'format_note'),
3552 format_field(f
, 'container', ignore
=(None, f
.get('ext'))),
3555 ] for f
in formats
if f
.get('preference') is None or f
['preference'] >= -1000]
3556 header_line
= self
._list
_format
_headers
(
3557 'ID', 'EXT', 'RESOLUTION', '\tFPS', 'HDR', delim
, '\tFILESIZE', '\tTBR', 'PROTO',
3558 delim
, 'VCODEC', '\tVBR', 'ACODEC', '\tABR', '\tASR', 'MORE INFO')
3560 return render_table(
3561 header_line
, table
, hide_empty
=True,
3562 delim
=self
._format
_out
('\u2500', self
.Styles
.DELIM
, '-', test_encoding
=True))
3564 def render_thumbnails_table(self
, info_dict
):
3565 thumbnails
= list(info_dict
.get('thumbnails') or [])
3568 return render_table(
3569 self
._list
_format
_headers
('ID', 'Width', 'Height', 'URL'),
3570 [[t
.get('id'), t
.get('width', 'unknown'), t
.get('height', 'unknown'), t
['url']] for t
in thumbnails
])
3572 def render_subtitles_table(self
, video_id
, subtitles
):
3573 def _row(lang
, formats
):
3574 exts
, names
= zip(*((f
['ext'], f
.get('name') or 'unknown') for f
in reversed(formats
)))
3575 if len(set(names
)) == 1:
3576 names
= [] if names
[0] == 'unknown' else names
[:1]
3577 return [lang
, ', '.join(names
), ', '.join(exts
)]
3581 return render_table(
3582 self
._list
_format
_headers
('Language', 'Name', 'Formats'),
3583 [_row(lang
, formats
) for lang
, formats
in subtitles
.items()],
3586 def __list_table(self
, video_id
, name
, func
, *args
):
3589 self
.to_screen(f
'{video_id} has no {name}')
3591 self
.to_screen(f
'[info] Available {name} for {video_id}:')
3592 self
.to_stdout(table
)
3594 def list_formats(self
, info_dict
):
3595 self
.__list
_table
(info_dict
['id'], 'formats', self
.render_formats_table
, info_dict
)
3597 def list_thumbnails(self
, info_dict
):
3598 self
.__list
_table
(info_dict
['id'], 'thumbnails', self
.render_thumbnails_table
, info_dict
)
3600 def list_subtitles(self
, video_id
, subtitles
, name
='subtitles'):
3601 self
.__list
_table
(video_id
, name
, self
.render_subtitles_table
, video_id
, subtitles
)
3603 def urlopen(self
, req
):
3604 """ Start an HTTP download """
3605 if isinstance(req
, str):
3606 req
= sanitized_Request(req
)
3607 return self
._opener
.open(req
, timeout
=self
._socket
_timeout
)
3609 def print_debug_header(self
):
3610 if not self
.params
.get('verbose'):
3613 # These imports can be slow. So import them only as needed
3614 from .extractor
.extractors
import _LAZY_LOADER
3615 from .extractor
.extractors
import _PLUGIN_CLASSES
as plugin_extractors
3617 def get_encoding(stream
):
3618 ret
= str(getattr(stream
, 'encoding', 'missing (%s)' % type(stream
).__name
__))
3619 if not supports_terminal_sequences(stream
):
3620 from .utils
import WINDOWS_VT_MODE
# Must be imported locally
3621 ret
+= ' (No VT)' if WINDOWS_VT_MODE
is False else ' (No ANSI)'
3624 encoding_str
= 'Encodings: locale %s, fs %s, pref %s, %s' % (
3625 locale
.getpreferredencoding(),
3626 sys
.getfilesystemencoding(),
3627 self
.get_encoding(),
3629 f
'{key} {get_encoding(stream)}' for key
, stream
in self
._out
_files
.items_
3630 if stream
is not None and key
!= 'console')
3633 logger
= self
.params
.get('logger')
3635 write_debug
= lambda msg
: logger
.debug(f
'[debug] {msg}')
3636 write_debug(encoding_str
)
3638 write_string(f
'[debug] {encoding_str}\n', encoding
=None)
3639 write_debug
= lambda msg
: self
._write
_string
(f
'[debug] {msg}\n')
3641 source
= detect_variant()
3642 write_debug(join_nonempty(
3643 'yt-dlp version', __version__
,
3644 f
'[{RELEASE_GIT_HEAD}]' if RELEASE_GIT_HEAD
else '',
3645 '' if source
== 'unknown' else f
'({source})',
3647 if not _LAZY_LOADER
:
3648 if os
.environ
.get('YTDLP_NO_LAZY_EXTRACTORS'):
3649 write_debug('Lazy loading extractors is forcibly disabled')
3651 write_debug('Lazy loading extractors is disabled')
3652 if plugin_extractors
or plugin_postprocessors
:
3653 write_debug('Plugins: %s' % [
3654 '%s%s' % (klass
.__name
__, '' if klass
.__name
__ == name
else f
' as {name}')
3655 for name
, klass
in itertools
.chain(plugin_extractors
.items(), plugin_postprocessors
.items())])
3656 if self
.params
['compat_opts']:
3657 write_debug('Compatibility options: %s' % ', '.join(self
.params
['compat_opts']))
3659 if source
== 'source':
3661 stdout
, _
, _
= Popen
.run(
3662 ['git', 'rev-parse', '--short', 'HEAD'],
3663 text
=True, cwd
=os
.path
.dirname(os
.path
.abspath(__file__
)),
3664 stdout
=subprocess
.PIPE
, stderr
=subprocess
.PIPE
)
3665 if re
.fullmatch('[0-9a-f]+', stdout
.strip()):
3666 write_debug(f
'Git HEAD: {stdout.strip()}')
3668 with contextlib
.suppress(Exception):
3671 write_debug(system_identifier())
3673 exe_versions
, ffmpeg_features
= FFmpegPostProcessor
.get_versions_and_features(self
)
3674 ffmpeg_features
= {key for key, val in ffmpeg_features.items() if val}
3676 exe_versions
['ffmpeg'] += ' (%s)' % ','.join(sorted(ffmpeg_features
))
3678 exe_versions
['rtmpdump'] = rtmpdump_version()
3679 exe_versions
['phantomjs'] = PhantomJSwrapper
._version
()
3680 exe_str
= ', '.join(
3681 f
'{exe} {v}' for exe
, v
in sorted(exe_versions
.items()) if v
3683 write_debug('exe versions: %s' % exe_str
)
3685 from .compat
.compat_utils
import get_package_info
3686 from .dependencies
import available_dependencies
3688 write_debug('Optional libraries: %s' % (', '.join(sorted({
3689 join_nonempty(*get_package_info(m
)) for m
in available_dependencies
.values()
3692 self
._setup
_opener
()
3694 for handler
in self
._opener
.handlers
:
3695 if hasattr(handler
, 'proxies'):
3696 proxy_map
.update(handler
.proxies
)
3697 write_debug(f
'Proxy map: {proxy_map}')
3700 if False and self
.params
.get('call_home'):
3701 ipaddr
= self
.urlopen('https://yt-dl.org/ip').read().decode()
3702 write_debug('Public IP address: %s' % ipaddr
)
3703 latest_version
= self
.urlopen(
3704 'https://yt-dl.org/latest/version').read().decode()
3705 if version_tuple(latest_version
) > version_tuple(__version__
):
3706 self
.report_warning(
3707 'You are using an outdated version (newest version: %s)! '
3708 'See https://yt-dl.org/update if you need help updating.' %
3711 def _setup_opener(self
):
3712 if hasattr(self
, '_opener'):
3714 timeout_val
= self
.params
.get('socket_timeout')
3715 self
._socket
_timeout
= 20 if timeout_val
is None else float(timeout_val
)
3717 opts_cookiesfrombrowser
= self
.params
.get('cookiesfrombrowser')
3718 opts_cookiefile
= self
.params
.get('cookiefile')
3719 opts_proxy
= self
.params
.get('proxy')
3721 self
.cookiejar
= load_cookies(opts_cookiefile
, opts_cookiesfrombrowser
, self
)
3723 cookie_processor
= YoutubeDLCookieProcessor(self
.cookiejar
)
3724 if opts_proxy
is not None:
3725 if opts_proxy
== '':
3728 proxies
= {'http': opts_proxy, 'https': opts_proxy}
3730 proxies
= urllib
.request
.getproxies()
3731 # Set HTTPS proxy to HTTP one if given (https://github.com/ytdl-org/youtube-dl/issues/805)
3732 if 'http' in proxies
and 'https' not in proxies
:
3733 proxies
['https'] = proxies
['http']
3734 proxy_handler
= PerRequestProxyHandler(proxies
)
3736 debuglevel
= 1 if self
.params
.get('debug_printtraffic') else 0
3737 https_handler
= make_HTTPS_handler(self
.params
, debuglevel
=debuglevel
)
3738 ydlh
= YoutubeDLHandler(self
.params
, debuglevel
=debuglevel
)
3739 redirect_handler
= YoutubeDLRedirectHandler()
3740 data_handler
= urllib
.request
.DataHandler()
3742 # When passing our own FileHandler instance, build_opener won't add the
3743 # default FileHandler and allows us to disable the file protocol, which
3744 # can be used for malicious purposes (see
3745 # https://github.com/ytdl-org/youtube-dl/issues/8227)
3746 file_handler
= urllib
.request
.FileHandler()
3748 def file_open(*args
, **kwargs
):
3749 raise urllib
.error
.URLError('file:// scheme is explicitly disabled in yt-dlp for security reasons')
3750 file_handler
.file_open
= file_open
3752 opener
= urllib
.request
.build_opener(
3753 proxy_handler
, https_handler
, cookie_processor
, ydlh
, redirect_handler
, data_handler
, file_handler
)
3755 # Delete the default user-agent header, which would otherwise apply in
3756 # cases where our custom HTTP handler doesn't come into play
3757 # (See https://github.com/ytdl-org/youtube-dl/issues/1309 for details)
3758 opener
.addheaders
= []
3759 self
._opener
= opener
3761 def encode(self
, s
):
3762 if isinstance(s
, bytes):
3763 return s
# Already encoded
3766 return s
.encode(self
.get_encoding())
3767 except UnicodeEncodeError as err
:
3768 err
.reason
= err
.reason
+ '. Check your system encoding configuration or use the --encoding option.'
3771 def get_encoding(self
):
3772 encoding
= self
.params
.get('encoding')
3773 if encoding
is None:
3774 encoding
= preferredencoding()
3777 def _write_info_json(self
, label
, ie_result
, infofn
, overwrite
=None):
3778 ''' Write infojson and returns True = written, 'exists' = Already exists, False = skip, None = error '''
3779 if overwrite
is None:
3780 overwrite
= self
.params
.get('overwrites', True)
3781 if not self
.params
.get('writeinfojson'):
3784 self
.write_debug(f
'Skipping writing {label} infojson')
3786 elif not self
._ensure
_dir
_exists
(infofn
):
3788 elif not overwrite
and os
.path
.exists(infofn
):
3789 self
.to_screen(f
'[info] {label.title()} metadata is already present')
3792 self
.to_screen(f
'[info] Writing {label} metadata as JSON to: {infofn}')
3794 write_json_file(self
.sanitize_info(ie_result
, self
.params
.get('clean_infojson', True)), infofn
)
3797 self
.report_error(f
'Cannot write {label} metadata to JSON file {infofn}')
3800 def _write_description(self
, label
, ie_result
, descfn
):
3801 ''' Write description and returns True = written, False = skip, None = error '''
3802 if not self
.params
.get('writedescription'):
3805 self
.write_debug(f
'Skipping writing {label} description')
3807 elif not self
._ensure
_dir
_exists
(descfn
):
3809 elif not self
.params
.get('overwrites', True) and os
.path
.exists(descfn
):
3810 self
.to_screen(f
'[info] {label.title()} description is already present')
3811 elif ie_result
.get('description') is None:
3812 self
.report_warning(f
'There\'s no {label} description to write')
3816 self
.to_screen(f
'[info] Writing {label} description to: {descfn}')
3817 with open(encodeFilename(descfn
), 'w', encoding
='utf-8') as descfile
:
3818 descfile
.write(ie_result
['description'])
3820 self
.report_error(f
'Cannot write {label} description file {descfn}')
3824 def _write_subtitles(self
, info_dict
, filename
):
3825 ''' Write subtitles to file and return list of (sub_filename, final_sub_filename); or None if error'''
3827 subtitles
= info_dict
.get('requested_subtitles')
3828 if not subtitles
or not (self
.params
.get('writesubtitles') or self
.params
.get('writeautomaticsub')):
3829 # subtitles download errors are already managed as troubles in relevant IE
3830 # that way it will silently go on when used with unsupporting IE
3833 sub_filename_base
= self
.prepare_filename(info_dict
, 'subtitle')
3834 if not sub_filename_base
:
3835 self
.to_screen('[info] Skipping writing video subtitles')
3837 for sub_lang
, sub_info
in subtitles
.items():
3838 sub_format
= sub_info
['ext']
3839 sub_filename
= subtitles_filename(filename
, sub_lang
, sub_format
, info_dict
.get('ext'))
3840 sub_filename_final
= subtitles_filename(sub_filename_base
, sub_lang
, sub_format
, info_dict
.get('ext'))
3841 existing_sub
= self
.existing_file((sub_filename_final
, sub_filename
))
3843 self
.to_screen(f
'[info] Video subtitle {sub_lang}.{sub_format} is already present')
3844 sub_info
['filepath'] = existing_sub
3845 ret
.append((existing_sub
, sub_filename_final
))
3848 self
.to_screen(f
'[info] Writing video subtitles to: {sub_filename}')
3849 if sub_info
.get('data') is not None:
3851 # Use newline='' to prevent conversion of newline characters
3852 # See https://github.com/ytdl-org/youtube-dl/issues/10268
3853 with open(sub_filename
, 'w', encoding
='utf-8', newline
='') as subfile
:
3854 subfile
.write(sub_info
['data'])
3855 sub_info
['filepath'] = sub_filename
3856 ret
.append((sub_filename
, sub_filename_final
))
3859 self
.report_error(f
'Cannot write video subtitles file {sub_filename}')
3863 sub_copy
= sub_info
.copy()
3864 sub_copy
.setdefault('http_headers', info_dict
.get('http_headers'))
3865 self
.dl(sub_filename
, sub_copy
, subtitle
=True)
3866 sub_info
['filepath'] = sub_filename
3867 ret
.append((sub_filename
, sub_filename_final
))
3868 except (DownloadError
, ExtractorError
, IOError, OSError, ValueError) + network_exceptions
as err
:
3869 msg
= f
'Unable to download video subtitles for {sub_lang!r}: {err}'
3870 if self
.params
.get('ignoreerrors') is not True: # False or 'only_download'
3871 if not self
.params
.get('ignoreerrors'):
3872 self
.report_error(msg
)
3873 raise DownloadError(msg
)
3874 self
.report_warning(msg
)
3877 def _write_thumbnails(self
, label
, info_dict
, filename
, thumb_filename_base
=None):
3878 ''' Write thumbnails to file and return list of (thumb_filename, final_thumb_filename) '''
3879 write_all
= self
.params
.get('write_all_thumbnails', False)
3880 thumbnails
, ret
= [], []
3881 if write_all
or self
.params
.get('writethumbnail', False):
3882 thumbnails
= info_dict
.get('thumbnails') or []
3883 multiple
= write_all
and len(thumbnails
) > 1
3885 if thumb_filename_base
is None:
3886 thumb_filename_base
= filename
3887 if thumbnails
and not thumb_filename_base
:
3888 self
.write_debug(f
'Skipping writing {label} thumbnail')
3891 for idx
, t
in list(enumerate(thumbnails
))[::-1]:
3892 thumb_ext
= (f
'{t["id"]}.' if multiple
else '') + determine_ext(t
['url'], 'jpg')
3893 thumb_display_id
= f
'{label} thumbnail {t["id"]}'
3894 thumb_filename
= replace_extension(filename
, thumb_ext
, info_dict
.get('ext'))
3895 thumb_filename_final
= replace_extension(thumb_filename_base
, thumb_ext
, info_dict
.get('ext'))
3897 existing_thumb
= self
.existing_file((thumb_filename_final
, thumb_filename
))
3899 self
.to_screen('[info] %s is already present' % (
3900 thumb_display_id
if multiple
else f
'{label} thumbnail').capitalize())
3901 t
['filepath'] = existing_thumb
3902 ret
.append((existing_thumb
, thumb_filename_final
))
3904 self
.to_screen(f
'[info] Downloading {thumb_display_id} ...')
3906 uf
= self
.urlopen(sanitized_Request(t
['url'], headers
=t
.get('http_headers', {})))
3907 self
.to_screen(f
'[info] Writing {thumb_display_id} to: {thumb_filename}')
3908 with open(encodeFilename(thumb_filename
), 'wb') as thumbf
:
3909 shutil
.copyfileobj(uf
, thumbf
)
3910 ret
.append((thumb_filename
, thumb_filename_final
))
3911 t
['filepath'] = thumb_filename
3912 except network_exceptions
as err
:
3914 self
.report_warning(f
'Unable to download {thumb_display_id}: {err}')
3915 if ret
and not write_all
: