24 from string
import ascii_letters
26 from .cache
import Cache
27 from .compat
import compat_os_name
, compat_shlex_quote
28 from .cookies
import load_cookies
29 from .downloader
import FFmpegFD
, get_suitable_downloader
, shorten_protocol_name
30 from .downloader
.rtmp
import rtmpdump_version
31 from .extractor
import gen_extractor_classes
, get_info_extractor
32 from .extractor
.openload
import PhantomJSwrapper
33 from .minicurses
import format_text
34 from .postprocessor
import _PLUGIN_CLASSES
as plugin_postprocessors
35 from .postprocessor
import (
37 FFmpegFixupDuplicateMoovPP
,
38 FFmpegFixupDurationPP
,
41 FFmpegFixupStretchedPP
,
42 FFmpegFixupTimestampPP
,
45 FFmpegVideoConvertorPP
,
46 MoveFilesAfterDownloadPP
,
49 from .postprocessor
.ffmpeg
import resolve_mapping
as resolve_recode_mapping
50 from .update
import detect_variant
76 PerRequestProxyHandler
,
83 UnavailableVideoError
,
85 YoutubeDLCookieProcessor
,
87 YoutubeDLRedirectHandler
,
102 format_decimal_suffix
,
119 register_socks_protocols
,
120 remove_terminal_sequences
,
131 supports_terminal_sequences
,
140 windows_enable_vt_mode
,
144 from .version
import RELEASE_GIT_HEAD
, __version__
146 if compat_os_name
== 'nt':
153 YoutubeDL objects are the ones responsible of downloading the
154 actual video file and writing it to disk if the user has requested
155 it, among some other tasks. In most cases there should be one per
156 program. As, given a video URL, the downloader doesn't know how to
157 extract all the needed information, task that InfoExtractors do, it
158 has to pass the URL to one of them.
160 For this, YoutubeDL objects have a method that allows
161 InfoExtractors to be registered in a given order. When it is passed
162 a URL, the YoutubeDL object handles it to the first InfoExtractor it
163 finds that reports being able to handle it. The InfoExtractor extracts
164 all the information about the video or videos the URL refers to, and
165 YoutubeDL process the extracted information, possibly using a File
166 Downloader to download the video.
168 YoutubeDL objects accept a lot of parameters. In order not to saturate
169 the object constructor with arguments, it receives a dictionary of
170 options instead. These options are available through the params
171 attribute for the InfoExtractors to use. The YoutubeDL also
172 registers itself as the downloader in charge for the InfoExtractors
173 that are added to it, so this is a "mutual registration".
177 username: Username for authentication purposes.
178 password: Password for authentication purposes.
179 videopassword: Password for accessing a video.
180 ap_mso: Adobe Pass multiple-system operator identifier.
181 ap_username: Multiple-system operator account username.
182 ap_password: Multiple-system operator account password.
183 usenetrc: Use netrc for authentication instead.
184 verbose: Print additional info to stdout.
185 quiet: Do not print messages to stdout.
186 no_warnings: Do not print out anything for warnings.
187 forceprint: A dict with keys WHEN mapped to a list of templates to
188 print to stdout. The allowed keys are video or any of the
189 items in utils.POSTPROCESS_WHEN.
190 For compatibility, a single list is also accepted
191 print_to_file: A dict with keys WHEN (same as forceprint) mapped to
192 a list of tuples with (template, filename)
193 forcejson: Force printing info_dict as JSON.
194 dump_single_json: Force printing the info_dict of the whole playlist
195 (or video) as a single JSON line.
196 force_write_download_archive: Force writing download archive regardless
197 of 'skip_download' or 'simulate'.
198 simulate: Do not download the video files. If unset (or None),
199 simulate only if listsubtitles, listformats or list_thumbnails is used
200 format: Video format code. see "FORMAT SELECTION" for more details.
201 You can also pass a function. The function takes 'ctx' as
202 argument and returns the formats to download.
203 See "build_format_selector" for an implementation
204 allow_unplayable_formats: Allow unplayable formats to be extracted and downloaded.
205 ignore_no_formats_error: Ignore "No video formats" error. Usefull for
206 extracting metadata even if the video is not actually
207 available for download (experimental)
208 format_sort: A list of fields by which to sort the video formats.
209 See "Sorting Formats" for more details.
210 format_sort_force: Force the given format_sort. see "Sorting Formats"
212 prefer_free_formats: Whether to prefer video formats with free containers
213 over non-free ones of same quality.
214 allow_multiple_video_streams: Allow multiple video streams to be merged
216 allow_multiple_audio_streams: Allow multiple audio streams to be merged
218 check_formats Whether to test if the formats are downloadable.
219 Can be True (check all), False (check none),
220 'selected' (check selected formats),
221 or None (check only if requested by extractor)
222 paths: Dictionary of output paths. The allowed keys are 'home'
223 'temp' and the keys of OUTTMPL_TYPES (in utils.py)
224 outtmpl: Dictionary of templates for output names. Allowed keys
225 are 'default' and the keys of OUTTMPL_TYPES (in utils.py).
226 For compatibility with youtube-dl, a single string can also be used
227 outtmpl_na_placeholder: Placeholder for unavailable meta fields.
228 restrictfilenames: Do not allow "&" and spaces in file names
229 trim_file_name: Limit length of filename (extension excluded)
230 windowsfilenames: Force the filenames to be windows compatible
231 ignoreerrors: Do not stop on download/postprocessing errors.
232 Can be 'only_download' to ignore only download errors.
233 Default is 'only_download' for CLI, but False for API
234 skip_playlist_after_errors: Number of allowed failures until the rest of
235 the playlist is skipped
236 force_generic_extractor: Force downloader to use the generic extractor
237 overwrites: Overwrite all video and metadata files if True,
238 overwrite only non-video files if None
239 and don't overwrite any file if False
240 For compatibility with youtube-dl,
241 "nooverwrites" may also be used instead
242 playlist_items: Specific indices of playlist to download.
243 playlistrandom: Download playlist items in random order.
244 lazy_playlist: Process playlist entries as they are received.
245 matchtitle: Download only matching titles.
246 rejecttitle: Reject downloads for matching titles.
247 logger: Log messages to a logging.Logger instance.
248 logtostderr: Log messages to stderr instead of stdout.
249 consoletitle: Display progress in console window's titlebar.
250 writedescription: Write the video description to a .description file
251 writeinfojson: Write the video description to a .info.json file
252 clean_infojson: Remove private fields from the infojson
253 getcomments: Extract video comments. This will not be written to disk
254 unless writeinfojson is also given
255 writeannotations: Write the video annotations to a .annotations.xml file
256 writethumbnail: Write the thumbnail image to a file
257 allow_playlist_files: Whether to write playlists' description, infojson etc
258 also to disk when using the 'write*' options
259 write_all_thumbnails: Write all thumbnail formats to files
260 writelink: Write an internet shortcut file, depending on the
261 current platform (.url/.webloc/.desktop)
262 writeurllink: Write a Windows internet shortcut file (.url)
263 writewebloclink: Write a macOS internet shortcut file (.webloc)
264 writedesktoplink: Write a Linux internet shortcut file (.desktop)
265 writesubtitles: Write the video subtitles to a file
266 writeautomaticsub: Write the automatically generated subtitles to a file
267 listsubtitles: Lists all available subtitles for the video
268 subtitlesformat: The format code for subtitles
269 subtitleslangs: List of languages of the subtitles to download (can be regex).
270 The list may contain "all" to refer to all the available
271 subtitles. The language can be prefixed with a "-" to
272 exclude it from the requested languages. Eg: ['all', '-live_chat']
273 keepvideo: Keep the video file after post-processing
274 daterange: A DateRange object, download only if the upload_date is in the range.
275 skip_download: Skip the actual download of the video file
276 cachedir: Location of the cache files in the filesystem.
277 False to disable filesystem cache.
278 noplaylist: Download single video instead of a playlist if in doubt.
279 age_limit: An integer representing the user's age in years.
280 Unsuitable videos for the given age are skipped.
281 min_views: An integer representing the minimum view count the video
282 must have in order to not be skipped.
283 Videos without view count information are always
284 downloaded. None for no limit.
285 max_views: An integer representing the maximum view count.
286 Videos that are more popular than that are not
288 Videos without view count information are always
289 downloaded. None for no limit.
290 download_archive: File name of a file where all downloads are recorded.
291 Videos already present in the file are not downloaded
293 break_on_existing: Stop the download process after attempting to download a
294 file that is in the archive.
295 break_on_reject: Stop the download process when encountering a video that
296 has been filtered out.
297 break_per_url: Whether break_on_reject and break_on_existing
298 should act on each input URL as opposed to for the entire queue
299 cookiefile: File name or text stream from where cookies should be read and dumped to
300 cookiesfrombrowser: A tuple containing the name of the browser, the profile
301 name/pathfrom where cookies are loaded, and the name of the
302 keyring. Eg: ('chrome', ) or ('vivaldi', 'default', 'BASICTEXT')
303 legacyserverconnect: Explicitly allow HTTPS connection to servers that do not
304 support RFC 5746 secure renegotiation
305 nocheckcertificate: Do not verify SSL certificates
306 client_certificate: Path to client certificate file in PEM format. May include the private key
307 client_certificate_key: Path to private key file for client certificate
308 client_certificate_password: Password for client certificate private key, if encrypted.
309 If not provided and the key is encrypted, yt-dlp will ask interactively
310 prefer_insecure: Use HTTP instead of HTTPS to retrieve information.
311 (Only supported by some extractors)
312 http_headers: A dictionary of custom headers to be used for all requests
313 proxy: URL of the proxy server to use
314 geo_verification_proxy: URL of the proxy to use for IP address verification
315 on geo-restricted sites.
316 socket_timeout: Time to wait for unresponsive hosts, in seconds
317 bidi_workaround: Work around buggy terminals without bidirectional text
318 support, using fridibi
319 debug_printtraffic:Print out sent and received HTTP traffic
320 default_search: Prepend this string if an input url is not valid.
321 'auto' for elaborate guessing
322 encoding: Use this encoding instead of the system-specified.
323 extract_flat: Whether to resolve and process url_results further
324 * False: Always process (default)
325 * True: Never process
326 * 'in_playlist': Do not process inside playlist/multi_video
327 * 'discard': Always process, but don't return the result
328 from inside playlist/multi_video
329 * 'discard_in_playlist': Same as "discard", but only for
330 playlists (not multi_video)
331 wait_for_video: If given, wait for scheduled streams to become available.
332 The value should be a tuple containing the range
333 (min_secs, max_secs) to wait between retries
334 postprocessors: A list of dictionaries, each with an entry
335 * key: The name of the postprocessor. See
336 yt_dlp/postprocessor/__init__.py for a list.
337 * when: When to run the postprocessor. Allowed values are
338 the entries of utils.POSTPROCESS_WHEN
339 Assumed to be 'post_process' if not given
340 progress_hooks: A list of functions that get called on download
341 progress, with a dictionary with the entries
342 * status: One of "downloading", "error", or "finished".
343 Check this first and ignore unknown values.
344 * info_dict: The extracted info_dict
346 If status is one of "downloading", or "finished", the
347 following properties may also be present:
348 * filename: The final filename (always present)
349 * tmpfilename: The filename we're currently writing to
350 * downloaded_bytes: Bytes on disk
351 * total_bytes: Size of the whole file, None if unknown
352 * total_bytes_estimate: Guess of the eventual file size,
354 * elapsed: The number of seconds since download started.
355 * eta: The estimated time in seconds, None if unknown
356 * speed: The download speed in bytes/second, None if
358 * fragment_index: The counter of the currently
359 downloaded video fragment.
360 * fragment_count: The number of fragments (= individual
361 files that will be merged)
363 Progress hooks are guaranteed to be called at least once
364 (with status "finished") if the download is successful.
365 postprocessor_hooks: A list of functions that get called on postprocessing
366 progress, with a dictionary with the entries
367 * status: One of "started", "processing", or "finished".
368 Check this first and ignore unknown values.
369 * postprocessor: Name of the postprocessor
370 * info_dict: The extracted info_dict
372 Progress hooks are guaranteed to be called at least twice
373 (with status "started" and "finished") if the processing is successful.
374 merge_output_format: Extension to use when merging formats.
375 final_ext: Expected final extension; used to detect when the file was
376 already downloaded and converted
377 fixup: Automatically correct known faults of the file.
379 - "never": do nothing
380 - "warn": only emit a warning
381 - "detect_or_warn": check whether we can do anything
382 about it, warn otherwise (default)
383 source_address: Client-side IP address to bind to.
384 sleep_interval_requests: Number of seconds to sleep between requests
386 sleep_interval: Number of seconds to sleep before each download when
387 used alone or a lower bound of a range for randomized
388 sleep before each download (minimum possible number
389 of seconds to sleep) when used along with
391 max_sleep_interval:Upper bound of a range for randomized sleep before each
392 download (maximum possible number of seconds to sleep).
393 Must only be used along with sleep_interval.
394 Actual sleep time will be a random float from range
395 [sleep_interval; max_sleep_interval].
396 sleep_interval_subtitles: Number of seconds to sleep before each subtitle download
397 listformats: Print an overview of available video formats and exit.
398 list_thumbnails: Print a table of all thumbnails and exit.
399 match_filter: A function that gets called for every video with the signature
400 (info_dict, *, incomplete: bool) -> Optional[str]
401 For backward compatibility with youtube-dl, the signature
402 (info_dict) -> Optional[str] is also allowed.
403 - If it returns a message, the video is ignored.
404 - If it returns None, the video is downloaded.
405 - If it returns utils.NO_DEFAULT, the user is interactively
406 asked whether to download the video.
407 match_filter_func in utils.py is one example for this.
408 no_color: Do not emit color codes in output.
409 geo_bypass: Bypass geographic restriction via faking X-Forwarded-For
412 Two-letter ISO 3166-2 country code that will be used for
413 explicit geographic restriction bypassing via faking
414 X-Forwarded-For HTTP header
416 IP range in CIDR notation that will be used similarly to
418 external_downloader: A dictionary of protocol keys and the executable of the
419 external downloader to use for it. The allowed protocols
420 are default|http|ftp|m3u8|dash|rtsp|rtmp|mms.
421 Set the value to 'native' to use the native downloader
422 compat_opts: Compatibility options. See "Differences in default behavior".
423 The following options do not work when used through the API:
424 filename, abort-on-error, multistreams, no-live-chat, format-sort
425 no-clean-infojson, no-playlist-metafiles, no-keep-subs, no-attach-info-json.
426 Refer __init__.py for their implementation
427 progress_template: Dictionary of templates for progress outputs.
428 Allowed keys are 'download', 'postprocess',
429 'download-title' (console title) and 'postprocess-title'.
430 The template is mapped on a dictionary with keys 'progress' and 'info'
431 retry_sleep_functions: Dictionary of functions that takes the number of attempts
432 as argument and returns the time to sleep in seconds.
433 Allowed keys are 'http', 'fragment', 'file_access'
434 download_ranges: A callback function that gets called for every video with
435 the signature (info_dict, ydl) -> Iterable[Section].
436 Only the returned sections will be downloaded.
437 Each Section is a dict with the following keys:
438 * start_time: Start time of the section in seconds
439 * end_time: End time of the section in seconds
440 * title: Section title (Optional)
441 * index: Section number (Optional)
442 force_keyframes_at_cuts: Re-encode the video when downloading ranges to get precise cuts
443 noprogress: Do not print the progress bar
445 The following parameters are not used by YoutubeDL itself, they are used by
446 the downloader (see yt_dlp/downloader/common.py):
447 nopart, updatetime, buffersize, ratelimit, throttledratelimit, min_filesize,
448 max_filesize, test, noresizebuffer, retries, file_access_retries, fragment_retries,
449 continuedl, xattr_set_filesize, hls_use_mpegts, http_chunk_size,
450 external_downloader_args, concurrent_fragment_downloads.
452 The following options are used by the post processors:
453 ffmpeg_location: Location of the ffmpeg/avconv binary; either the path
454 to the binary or its containing directory.
455 postprocessor_args: A dictionary of postprocessor/executable keys (in lower case)
456 and a list of additional command-line arguments for the
457 postprocessor/executable. The dict can also have "PP+EXE" keys
458 which are used when the given exe is used by the given PP.
459 Use 'default' as the name for arguments to passed to all PP
460 For compatibility with youtube-dl, a single list of args
463 The following options are used by the extractors:
464 extractor_retries: Number of times to retry for known errors
465 dynamic_mpd: Whether to process dynamic DASH manifests (default: True)
466 hls_split_discontinuity: Split HLS playlists to different formats at
467 discontinuities such as ad breaks (default: False)
468 extractor_args: A dictionary of arguments to be passed to the extractors.
469 See "EXTRACTOR ARGUMENTS" for details.
470 Eg: {'youtube': {'skip': ['dash', 'hls']}}
471 mark_watched: Mark videos watched (even with --simulate). Only for YouTube
473 The following options are deprecated and may be removed in the future:
475 playliststart: - Use playlist_items
476 Playlist item to start at.
477 playlistend: - Use playlist_items
478 Playlist item to end at.
479 playlistreverse: - Use playlist_items
480 Download playlist items in reverse order.
481 forceurl: - Use forceprint
482 Force printing final URL.
483 forcetitle: - Use forceprint
484 Force printing title.
485 forceid: - Use forceprint
487 forcethumbnail: - Use forceprint
488 Force printing thumbnail URL.
489 forcedescription: - Use forceprint
490 Force printing description.
491 forcefilename: - Use forceprint
492 Force printing final filename.
493 forceduration: - Use forceprint
494 Force printing duration.
495 allsubtitles: - Use subtitleslangs = ['all']
496 Downloads all the subtitles of the video
497 (requires writesubtitles or writeautomaticsub)
498 include_ads: - Doesn't work
500 call_home: - Not implemented
501 Boolean, true iff we are allowed to contact the
502 yt-dlp servers for debugging.
503 post_hooks: - Register a custom postprocessor
504 A list of functions that get called as the final step
505 for each video file, after all postprocessors have been
506 called. The filename will be passed as the only argument.
507 hls_prefer_native: - Use external_downloader = {'m3u8': 'native'} or {'m3u8': 'ffmpeg'}.
508 Use the native HLS downloader instead of ffmpeg/avconv
509 if True, otherwise use ffmpeg/avconv if False, otherwise
510 use downloader suggested by extractor if None.
511 prefer_ffmpeg: - avconv support is deprecated
512 If False, use avconv instead of ffmpeg if both are available,
513 otherwise prefer ffmpeg.
514 youtube_include_dash_manifest: - Use extractor_args
515 If True (default), DASH manifests and related
516 data will be downloaded and processed by extractor.
517 You can reduce network I/O by disabling it if you don't
518 care about DASH. (only for youtube)
519 youtube_include_hls_manifest: - Use extractor_args
520 If True (default), HLS manifests and related
521 data will be downloaded and processed by extractor.
522 You can reduce network I/O by disabling it if you don't
523 care about HLS. (only for youtube)
527 'width', 'height', 'tbr', 'abr', 'asr', 'vbr', 'fps', 'filesize', 'filesize_approx',
528 'timestamp', 'release_timestamp',
529 'duration', 'view_count', 'like_count', 'dislike_count', 'repost_count',
530 'average_rating', 'comment_count', 'age_limit',
531 'start_time', 'end_time',
532 'chapter_number', 'season_number', 'episode_number',
533 'track_number', 'disc_number', 'release_year',
537 # NB: Keep in sync with the docstring of extractor/common.py
538 'url', 'manifest_url', 'manifest_stream_number', 'ext', 'format', 'format_id', 'format_note',
539 'width', 'height', 'resolution', 'dynamic_range', 'tbr', 'abr', 'acodec', 'asr',
540 'vbr', 'fps', 'vcodec', 'container', 'filesize', 'filesize_approx',
541 'player_url', 'protocol', 'fragment_base_url', 'fragments', 'is_from_start',
542 'preference', 'language', 'language_preference', 'quality', 'source_preference',
543 'http_headers', 'stretched_ratio', 'no_resume', 'has_drm', 'downloader_options',
544 'page_url', 'app', 'play_path', 'tc_url', 'flash_version', 'rtmp_live', 'rtmp_conn', 'rtmp_protocol', 'rtmp_real_time'
546 _format_selection_exts
= {
547 'audio': set(MEDIA_EXTENSIONS
.common_audio
),
548 'video': set(MEDIA_EXTENSIONS
.common_video
+ ('3gp', )),
549 'storyboards': set(MEDIA_EXTENSIONS
.storyboards
),
552 def __init__(self
, params
=None, auto_init
=True):
553 """Create a FileDownloader object with the given options.
554 @param auto_init Whether to load the default extractors and print header (if verbose).
555 Set to 'no_verbose_header' to not print the header
561 self
._ies
_instances
= {}
562 self
._pps
= {k: [] for k in POSTPROCESS_WHEN}
563 self
._printed
_messages
= set()
564 self
._first
_webpage
_request
= True
565 self
._post
_hooks
= []
566 self
._progress
_hooks
= []
567 self
._postprocessor
_hooks
= []
568 self
._download
_retcode
= 0
569 self
._num
_downloads
= 0
571 self
._playlist
_level
= 0
572 self
._playlist
_urls
= set()
573 self
.cache
= Cache(self
)
575 windows_enable_vt_mode()
576 stdout
= sys
.stderr
if self
.params
.get('logtostderr') else sys
.stdout
577 self
._out
_files
= Namespace(
580 screen
=sys
.stderr
if self
.params
.get('quiet') else stdout
,
581 console
=None if compat_os_name
== 'nt' else next(
582 filter(supports_terminal_sequences
, (sys
.stderr
, sys
.stdout
)), None)
584 self
._allow
_colors
= Namespace(**{
585 type_
: not self
.params
.get('no_color') and supports_terminal_sequences(stream
)
586 for type_
, stream
in self
._out
_files
.items_
if type_
!= 'console'
589 # The code is left like this to be reused for future deprecations
590 MIN_SUPPORTED
, MIN_RECOMMENDED
= (3, 7), (3, 7)
591 current_version
= sys
.version_info
[:2]
592 if current_version
< MIN_RECOMMENDED
:
593 msg
= ('Support for Python version %d.%d has been deprecated. '
594 'See https://github.com/yt-dlp/yt-dlp/issues/3764 for more details.'
595 '\n You will no longer receive updates on this version')
596 if current_version
< MIN_SUPPORTED
:
597 msg
= 'Python version %d.%d is no longer supported'
598 self
.deprecation_warning(
599 f
'{msg}! Please update to Python %d.%d or above' % (*current_version
, *MIN_RECOMMENDED
))
601 if self
.params
.get('allow_unplayable_formats'):
603 f
'You have asked for {self._format_err("UNPLAYABLE", self.Styles.EMPHASIS)} formats to be listed/downloaded. '
604 'This is a developer option intended for debugging. \n'
605 ' If you experience any issues while using this option, '
606 f
'{self._format_err("DO NOT", self.Styles.ERROR)} open a bug report')
608 def check_deprecated(param
, option
, suggestion
):
609 if self
.params
.get(param
) is not None:
610 self
.report_warning(f
'{option} is deprecated. Use {suggestion} instead')
614 if check_deprecated('cn_verification_proxy', '--cn-verification-proxy', '--geo-verification-proxy'):
615 if self
.params
.get('geo_verification_proxy') is None:
616 self
.params
['geo_verification_proxy'] = self
.params
['cn_verification_proxy']
618 check_deprecated('autonumber', '--auto-number', '-o "%(autonumber)s-%(title)s.%(ext)s"')
619 check_deprecated('usetitle', '--title', '-o "%(title)s-%(id)s.%(ext)s"')
620 check_deprecated('useid', '--id', '-o "%(id)s.%(ext)s"')
622 for msg
in self
.params
.get('_warnings', []):
623 self
.report_warning(msg
)
624 for msg
in self
.params
.get('_deprecation_warnings', []):
625 self
.deprecation_warning(msg
)
627 self
.params
['compat_opts'] = set(self
.params
.get('compat_opts', ()))
628 if 'list-formats' in self
.params
['compat_opts']:
629 self
.params
['listformats_table'] = False
631 if 'overwrites' not in self
.params
and self
.params
.get('nooverwrites') is not None:
632 # nooverwrites was unnecessarily changed to overwrites
633 # in 0c3d0f51778b153f65c21906031c2e091fcfb641
634 # This ensures compatibility with both keys
635 self
.params
['overwrites'] = not self
.params
['nooverwrites']
636 elif self
.params
.get('overwrites') is None:
637 self
.params
.pop('overwrites', None)
639 self
.params
['nooverwrites'] = not self
.params
['overwrites']
641 self
.params
.setdefault('forceprint', {})
642 self
.params
.setdefault('print_to_file', {})
644 # Compatibility with older syntax
645 if not isinstance(params
['forceprint'], dict):
646 self
.params
['forceprint'] = {'video': params['forceprint']}
648 if self
.params
.get('bidi_workaround', False):
651 master
, slave
= pty
.openpty()
652 width
= shutil
.get_terminal_size().columns
653 width_args
= [] if width
is None else ['-w', str(width
)]
654 sp_kwargs
= {'stdin': subprocess.PIPE, 'stdout': slave, 'stderr': self._out_files.error}
656 self
._output
_process
= Popen(['bidiv'] + width_args
, **sp_kwargs
)
658 self
._output
_process
= Popen(['fribidi', '-c', 'UTF-8'] + width_args
, **sp_kwargs
)
659 self
._output
_channel
= os
.fdopen(master
, 'rb')
660 except OSError as ose
:
661 if ose
.errno
== errno
.ENOENT
:
663 'Could not find fribidi executable, ignoring --bidi-workaround. '
664 'Make sure that fribidi is an executable file in one of the directories in your $PATH.')
669 if auto_init
!= 'no_verbose_header':
670 self
.print_debug_header()
671 self
.add_default_info_extractors()
673 if (sys
.platform
!= 'win32'
674 and sys
.getfilesystemencoding() in ['ascii', 'ANSI_X3.4-1968']
675 and not self
.params
.get('restrictfilenames', False)):
676 # Unicode filesystem API will throw errors (#1474, #13027)
678 'Assuming --restrict-filenames since file system encoding '
679 'cannot encode all characters. '
680 'Set the LC_ALL environment variable to fix this.')
681 self
.params
['restrictfilenames'] = True
683 self
._parse
_outtmpl
()
685 # Creating format selector here allows us to catch syntax errors before the extraction
686 self
.format_selector
= (
687 self
.params
.get('format') if self
.params
.get('format') in (None, '-')
688 else self
.params
['format'] if callable(self
.params
['format'])
689 else self
.build_format_selector(self
.params
['format']))
691 # Set http_headers defaults according to std_headers
692 self
.params
['http_headers'] = merge_headers(std_headers
, self
.params
.get('http_headers', {}))
695 'post_hooks': self
.add_post_hook
,
696 'progress_hooks': self
.add_progress_hook
,
697 'postprocessor_hooks': self
.add_postprocessor_hook
,
699 for opt
, fn
in hooks
.items():
700 for ph
in self
.params
.get(opt
, []):
703 for pp_def_raw
in self
.params
.get('postprocessors', []):
704 pp_def
= dict(pp_def_raw
)
705 when
= pp_def
.pop('when', 'post_process')
706 self
.add_post_processor(
707 get_postprocessor(pp_def
.pop('key'))(self
, **pp_def
),
711 register_socks_protocols()
713 def preload_download_archive(fn
):
714 """Preload the archive, if any is specified"""
717 self
.write_debug(f
'Loading archive file {fn!r}')
719 with locked_file(fn
, 'r', encoding
='utf-8') as archive_file
:
720 for line
in archive_file
:
721 self
.archive
.add(line
.strip())
722 except OSError as ioe
:
723 if ioe
.errno
!= errno
.ENOENT
:
729 preload_download_archive(self
.params
.get('download_archive'))
731 def warn_if_short_id(self
, argv
):
732 # short YouTube ID starting with dash?
734 i
for i
, a
in enumerate(argv
)
735 if re
.match(r
'^-[0-9A-Za-z_-]{10}$', a
)]
739 + [a
for i
, a
in enumerate(argv
) if i
not in idxs
]
740 + ['--'] + [argv
[i
] for i
in idxs
]
743 'Long argument string detected. '
744 'Use -- to separate parameters and URLs, like this:\n%s' %
745 args_to_str(correct_argv
))
747 def add_info_extractor(self
, ie
):
748 """Add an InfoExtractor object to the end of the list."""
750 self
._ies
[ie_key
] = ie
751 if not isinstance(ie
, type):
752 self
._ies
_instances
[ie_key
] = ie
753 ie
.set_downloader(self
)
755 def _get_info_extractor_class(self
, ie_key
):
756 ie
= self
._ies
.get(ie_key
)
758 ie
= get_info_extractor(ie_key
)
759 self
.add_info_extractor(ie
)
762 def get_info_extractor(self
, ie_key
):
764 Get an instance of an IE with name ie_key, it will try to get one from
765 the _ies list, if there's no instance it will create a new one and add
766 it to the extractor list.
768 ie
= self
._ies
_instances
.get(ie_key
)
770 ie
= get_info_extractor(ie_key
)()
771 self
.add_info_extractor(ie
)
774 def add_default_info_extractors(self
):
776 Add the InfoExtractors returned by gen_extractors to the end of the list
778 for ie
in gen_extractor_classes():
779 self
.add_info_extractor(ie
)
781 def add_post_processor(self
, pp
, when
='post_process'):
782 """Add a PostProcessor object to the end of the chain."""
783 assert when
in POSTPROCESS_WHEN
, f
'Invalid when={when}'
784 self
._pps
[when
].append(pp
)
785 pp
.set_downloader(self
)
787 def add_post_hook(self
, ph
):
788 """Add the post hook"""
789 self
._post
_hooks
.append(ph
)
791 def add_progress_hook(self
, ph
):
792 """Add the download progress hook"""
793 self
._progress
_hooks
.append(ph
)
795 def add_postprocessor_hook(self
, ph
):
796 """Add the postprocessing progress hook"""
797 self
._postprocessor
_hooks
.append(ph
)
798 for pps
in self
._pps
.values():
800 pp
.add_progress_hook(ph
)
802 def _bidi_workaround(self
, message
):
803 if not hasattr(self
, '_output_channel'):
806 assert hasattr(self
, '_output_process')
807 assert isinstance(message
, str)
808 line_count
= message
.count('\n') + 1
809 self
._output
_process
.stdin
.write((message
+ '\n').encode())
810 self
._output
_process
.stdin
.flush()
811 res
= ''.join(self
._output
_channel
.readline().decode()
812 for _
in range(line_count
))
813 return res
[:-len('\n')]
815 def _write_string(self
, message
, out
=None, only_once
=False):
817 if message
in self
._printed
_messages
:
819 self
._printed
_messages
.add(message
)
820 write_string(message
, out
=out
, encoding
=self
.params
.get('encoding'))
822 def to_stdout(self
, message
, skip_eol
=False, quiet
=None):
823 """Print message to stdout"""
824 if quiet
is not None:
825 self
.deprecation_warning('"YoutubeDL.to_stdout" no longer accepts the argument quiet. Use "YoutubeDL.to_screen" instead')
826 if skip_eol
is not False:
827 self
.deprecation_warning('"YoutubeDL.to_stdout" no longer accepts the argument skip_eol. Use "YoutubeDL.to_screen" instead')
828 self
._write
_string
(f
'{self._bidi_workaround(message)}\n', self
._out
_files
.out
)
830 def to_screen(self
, message
, skip_eol
=False, quiet
=None):
831 """Print message to screen if not in quiet mode"""
832 if self
.params
.get('logger'):
833 self
.params
['logger'].debug(message
)
835 if (self
.params
.get('quiet') if quiet
is None else quiet
) and not self
.params
.get('verbose'):
838 '%s%s' % (self
._bidi
_workaround
(message
), ('' if skip_eol
else '\n')),
839 self
._out
_files
.screen
)
841 def to_stderr(self
, message
, only_once
=False):
842 """Print message to stderr"""
843 assert isinstance(message
, str)
844 if self
.params
.get('logger'):
845 self
.params
['logger'].error(message
)
847 self
._write
_string
(f
'{self._bidi_workaround(message)}\n', self
._out
_files
.error
, only_once
=only_once
)
849 def _send_console_code(self
, code
):
850 if compat_os_name
== 'nt' or not self
._out
_files
.console
:
852 self
._write
_string
(code
, self
._out
_files
.console
)
854 def to_console_title(self
, message
):
855 if not self
.params
.get('consoletitle', False):
857 message
= remove_terminal_sequences(message
)
858 if compat_os_name
== 'nt':
859 if ctypes
.windll
.kernel32
.GetConsoleWindow():
860 # c_wchar_p() might not be necessary if `message` is
861 # already of type unicode()
862 ctypes
.windll
.kernel32
.SetConsoleTitleW(ctypes
.c_wchar_p(message
))
864 self
._send
_console
_code
(f
'\033]0;{message}\007')
866 def save_console_title(self
):
867 if not self
.params
.get('consoletitle') or self
.params
.get('simulate'):
869 self
._send
_console
_code
('\033[22;0t') # Save the title on stack
871 def restore_console_title(self
):
872 if not self
.params
.get('consoletitle') or self
.params
.get('simulate'):
874 self
._send
_console
_code
('\033[23;0t') # Restore the title from stack
877 self
.save_console_title()
880 def __exit__(self
, *args
):
881 self
.restore_console_title()
883 if self
.params
.get('cookiefile') is not None:
884 self
.cookiejar
.save(ignore_discard
=True, ignore_expires
=True)
886 def trouble(self
, message
=None, tb
=None, is_error
=True):
887 """Determine action to take when a download problem appears.
889 Depending on if the downloader has been configured to ignore
890 download errors or not, this method may throw an exception or
891 not when errors are found, after printing the message.
893 @param tb If given, is additional traceback information
894 @param is_error Whether to raise error according to ignorerrors
896 if message
is not None:
897 self
.to_stderr(message
)
898 if self
.params
.get('verbose'):
900 if sys
.exc_info()[0]: # if .trouble has been called from an except block
902 if hasattr(sys
.exc_info()[1], 'exc_info') and sys
.exc_info()[1].exc_info
[0]:
903 tb
+= ''.join(traceback
.format_exception(*sys
.exc_info()[1].exc_info
))
904 tb
+= encode_compat_str(traceback
.format_exc())
906 tb_data
= traceback
.format_list(traceback
.extract_stack())
907 tb
= ''.join(tb_data
)
912 if not self
.params
.get('ignoreerrors'):
913 if sys
.exc_info()[0] and hasattr(sys
.exc_info()[1], 'exc_info') and sys
.exc_info()[1].exc_info
[0]:
914 exc_info
= sys
.exc_info()[1].exc_info
916 exc_info
= sys
.exc_info()
917 raise DownloadError(message
, exc_info
)
918 self
._download
_retcode
= 1
922 EMPHASIS
='light blue',
928 SUPPRESS
='light black',
931 def _format_text(self
, handle
, allow_colors
, text
, f
, fallback
=None, *, test_encoding
=False):
935 # handle.encoding can be None. See https://github.com/yt-dlp/yt-dlp/issues/2711
936 encoding
= self
.params
.get('encoding') or getattr(handle
, 'encoding', None) or 'ascii'
937 text
= text
.encode(encoding
, 'ignore').decode(encoding
)
938 if fallback
is not None and text
!= original_text
:
940 return format_text(text
, f
) if allow_colors
else text
if fallback
is None else fallback
942 def _format_out(self
, *args
, **kwargs
):
943 return self
._format
_text
(self
._out
_files
.out
, self
._allow
_colors
.out
, *args
, **kwargs
)
945 def _format_screen(self
, *args
, **kwargs
):
946 return self
._format
_text
(self
._out
_files
.screen
, self
._allow
_colors
.screen
, *args
, **kwargs
)
948 def _format_err(self
, *args
, **kwargs
):
949 return self
._format
_text
(self
._out
_files
.error
, self
._allow
_colors
.error
, *args
, **kwargs
)
951 def report_warning(self
, message
, only_once
=False):
953 Print the message to stderr, it will be prefixed with 'WARNING:'
954 If stderr is a tty file the 'WARNING:' will be colored
956 if self
.params
.get('logger') is not None:
957 self
.params
['logger'].warning(message
)
959 if self
.params
.get('no_warnings'):
961 self
.to_stderr(f
'{self._format_err("WARNING:", self.Styles.WARNING)} {message}', only_once
)
963 def deprecation_warning(self
, message
):
964 if self
.params
.get('logger') is not None:
965 self
.params
['logger'].warning(f
'DeprecationWarning: {message}')
967 self
.to_stderr(f
'{self._format_err("DeprecationWarning:", self.Styles.ERROR)} {message}', True)
969 def report_error(self
, message
, *args
, **kwargs
):
971 Do the same as trouble, but prefixes the message with 'ERROR:', colored
972 in red if stderr is a tty file.
974 self
.trouble(f
'{self._format_err("ERROR:", self.Styles.ERROR)} {message}', *args
, **kwargs
)
976 def write_debug(self
, message
, only_once
=False):
977 '''Log debug message or Print message to stderr'''
978 if not self
.params
.get('verbose', False):
980 message
= f
'[debug] {message}'
981 if self
.params
.get('logger'):
982 self
.params
['logger'].debug(message
)
984 self
.to_stderr(message
, only_once
)
986 def report_file_already_downloaded(self
, file_name
):
987 """Report file has already been fully downloaded."""
989 self
.to_screen('[download] %s has already been downloaded' % file_name
)
990 except UnicodeEncodeError:
991 self
.to_screen('[download] The file has already been downloaded')
993 def report_file_delete(self
, file_name
):
994 """Report that existing file will be deleted."""
996 self
.to_screen('Deleting existing file %s' % file_name
)
997 except UnicodeEncodeError:
998 self
.to_screen('Deleting existing file')
1000 def raise_no_formats(self
, info
, forced
=False, *, msg
=None):
1001 has_drm
= info
.get('_has_drm')
1002 ignored
, expected
= self
.params
.get('ignore_no_formats_error'), bool(msg
)
1003 msg
= msg
or has_drm
and 'This video is DRM protected' or 'No video formats found!'
1004 if forced
or not ignored
:
1005 raise ExtractorError(msg
, video_id
=info
['id'], ie
=info
['extractor'],
1006 expected
=has_drm
or ignored
or expected
)
1008 self
.report_warning(msg
)
1010 def parse_outtmpl(self
):
1011 self
.deprecation_warning('"YoutubeDL.parse_outtmpl" is deprecated and may be removed in a future version')
1012 self
._parse
_outtmpl
()
1013 return self
.params
['outtmpl']
1015 def _parse_outtmpl(self
):
1017 if self
.params
.get('restrictfilenames'): # Remove spaces in the default template
1018 sanitize
= lambda x
: x
.replace(' - ', ' ').replace(' ', '-')
1020 outtmpl
= self
.params
.setdefault('outtmpl', {})
1021 if not isinstance(outtmpl
, dict):
1022 self
.params
['outtmpl'] = outtmpl
= {'default': outtmpl}
1023 outtmpl
.update({k: sanitize(v) for k, v in DEFAULT_OUTTMPL.items() if outtmpl.get(k) is None}
)
1025 def get_output_path(self
, dir_type
='', filename
=None):
1026 paths
= self
.params
.get('paths', {})
1027 assert isinstance(paths
, dict)
1028 path
= os
.path
.join(
1029 expand_path(paths
.get('home', '').strip()),
1030 expand_path(paths
.get(dir_type
, '').strip()) if dir_type
else '',
1032 return sanitize_path(path
, force
=self
.params
.get('windowsfilenames'))
1035 def _outtmpl_expandpath(outtmpl
):
1036 # expand_path translates '%%' into '%' and '$$' into '$'
1037 # correspondingly that is not what we want since we need to keep
1038 # '%%' intact for template dict substitution step. Working around
1039 # with boundary-alike separator hack.
1040 sep
= ''.join([random
.choice(ascii_letters
) for _
in range(32)])
1041 outtmpl
= outtmpl
.replace('%%', f
'%{sep}%').replace('$$', f
'${sep}$')
1043 # outtmpl should be expand_path'ed before template dict substitution
1044 # because meta fields may contain env variables we don't want to
1045 # be expanded. For example, for outtmpl "%(title)s.%(ext)s" and
1046 # title "Hello $PATH", we don't want `$PATH` to be expanded.
1047 return expand_path(outtmpl
).replace(sep
, '')
1050 def escape_outtmpl(outtmpl
):
1051 ''' Escape any remaining strings like %s, %abc% etc. '''
1053 STR_FORMAT_RE_TMPL
.format('', '(?![%(\0])'),
1054 lambda mobj
: ('' if mobj
.group('has_key') else '%') + mobj
.group(0),
1058 def validate_outtmpl(cls
, outtmpl
):
1059 ''' @return None or Exception object '''
1061 STR_FORMAT_RE_TMPL
.format('[^)]*', '[ljhqBUDS]'),
1062 lambda mobj
: f
'{mobj.group(0)[:-1]}s',
1063 cls
._outtmpl
_expandpath
(outtmpl
))
1065 cls
.escape_outtmpl(outtmpl
) % collections
.defaultdict(int)
1067 except ValueError as err
:
1071 def _copy_infodict(info_dict
):
1072 info_dict
= dict(info_dict
)
1073 info_dict
.pop('__postprocessors', None)
1074 info_dict
.pop('__pending_error', None)
1077 def prepare_outtmpl(self
, outtmpl
, info_dict
, sanitize
=False):
1078 """ Make the outtmpl and info_dict suitable for substitution: ydl.escape_outtmpl(outtmpl) % info_dict
1079 @param sanitize Whether to sanitize the output as a filename.
1080 For backward compatibility, a function can also be passed
1083 info_dict
.setdefault('epoch', int(time
.time())) # keep epoch consistent once set
1085 info_dict
= self
._copy
_infodict
(info_dict
)
1086 info_dict
['duration_string'] = ( # %(duration>%H-%M-%S)s is wrong if duration > 24hrs
1087 formatSeconds(info_dict
['duration'], '-' if sanitize
else ':')
1088 if info_dict
.get('duration', None) is not None
1090 info_dict
['autonumber'] = int(self
.params
.get('autonumber_start', 1) - 1 + self
._num
_downloads
)
1091 info_dict
['video_autonumber'] = self
._num
_videos
1092 if info_dict
.get('resolution') is None:
1093 info_dict
['resolution'] = self
.format_resolution(info_dict
, default
=None)
1095 # For fields playlist_index, playlist_autonumber and autonumber convert all occurrences
1096 # of %(field)s to %(field)0Nd for backward compatibility
1097 field_size_compat_map
= {
1098 'playlist_index': number_of_digits(info_dict
.get('__last_playlist_index') or 0),
1099 'playlist_autonumber': number_of_digits(info_dict
.get('n_entries') or 0),
1100 'autonumber': self
.params
.get('autonumber_size') or 5,
1104 EXTERNAL_FORMAT_RE
= re
.compile(STR_FORMAT_RE_TMPL
.format('[^)]*', f
'[{STR_FORMAT_TYPES}ljhqBUDS]'))
1109 # Field is of the form key1.key2...
1110 # where keys (except first) can be string, int or slice
1111 FIELD_RE
= r
'\w*(?:\.(?:\w+|{num}|{num}?(?::{num}?){{1,2}}))*'.format(num
=r
'(?:-?\d+)')
1112 MATH_FIELD_RE
= rf
'(?:{FIELD_RE}|-?{NUMBER_RE})'
1113 MATH_OPERATORS_RE
= r
'(?:%s)' % '|'.join(map(re
.escape
, MATH_FUNCTIONS
.keys()))
1114 INTERNAL_FORMAT_RE
= re
.compile(rf
'''(?x)
1116 (?P<fields>{FIELD_RE})
1117 (?P<maths>(?:{MATH_OPERATORS_RE}{MATH_FIELD_RE})*)
1118 (?:>(?P<strf_format>.+?))?
1120 (?P<alternate>(?<!\\),[^|&)]+)?
1121 (?:&(?P<replacement>.*?))?
1122 (?:\|(?P<default>.*?))?
1125 def _traverse_infodict(k
):
1129 return traverse_obj(info_dict
, k
, is_user_input
=True, traverse_string
=True)
1131 def get_value(mdict
):
1133 value
= _traverse_infodict(mdict
['fields'])
1136 value
= float_or_none(value
)
1137 if value
is not None:
1140 offset_key
= mdict
['maths']
1142 value
= float_or_none(value
)
1146 MATH_FIELD_RE
if operator
else MATH_OPERATORS_RE
,
1147 offset_key
).group(0)
1148 offset_key
= offset_key
[len(item
):]
1149 if operator
is None:
1150 operator
= MATH_FUNCTIONS
[item
]
1152 item
, multiplier
= (item
[1:], -1) if item
[0] == '-' else (item
, 1)
1153 offset
= float_or_none(item
)
1155 offset
= float_or_none(_traverse_infodict(item
))
1157 value
= operator(value
, multiplier
* offset
)
1158 except (TypeError, ZeroDivisionError):
1161 # Datetime formatting
1162 if mdict
['strf_format']:
1163 value
= strftime_or_none(value
, mdict
['strf_format'].replace('\\,', ','))
1167 na
= self
.params
.get('outtmpl_na_placeholder', 'NA')
1169 def filename_sanitizer(key
, value
, restricted
=self
.params
.get('restrictfilenames')):
1170 return sanitize_filename(str(value
), restricted
=restricted
, is_id
=(
1171 bool(re
.search(r
'(^|[_.])id(\.|$)', key
))
1172 if 'filename-sanitization' in self
.params
['compat_opts']
1175 sanitizer
= sanitize
if callable(sanitize
) else filename_sanitizer
1176 sanitize
= bool(sanitize
)
1178 def _dumpjson_default(obj
):
1179 if isinstance(obj
, (set, LazyList
)):
1183 def create_key(outer_mobj
):
1184 if not outer_mobj
.group('has_key'):
1185 return outer_mobj
.group(0)
1186 key
= outer_mobj
.group('key')
1187 mobj
= re
.match(INTERNAL_FORMAT_RE
, key
)
1188 initial_field
= mobj
.group('fields') if mobj
else ''
1189 value
, replacement
, default
= None, None, na
1191 mobj
= mobj
.groupdict()
1192 default
= mobj
['default'] if mobj
['default'] is not None else default
1193 value
= get_value(mobj
)
1194 replacement
= mobj
['replacement']
1195 if value
is None and mobj
['alternate']:
1196 mobj
= re
.match(INTERNAL_FORMAT_RE
, mobj
['remaining'][1:])
1200 fmt
= outer_mobj
.group('format')
1201 if fmt
== 's' and value
is not None and key
in field_size_compat_map
.keys():
1202 fmt
= f
'0{field_size_compat_map[key]:d}d'
1204 value
= default
if value
is None else value
if replacement
is None else replacement
1206 flags
= outer_mobj
.group('conversion') or ''
1207 str_fmt
= f
'{fmt[:-1]}s'
1208 if fmt
[-1] == 'l': # list
1209 delim
= '\n' if '#' in flags
else ', '
1210 value
, fmt
= delim
.join(map(str, variadic(value
, allowed_types
=(str, bytes)))), str_fmt
1211 elif fmt
[-1] == 'j': # json
1212 value
, fmt
= json
.dumps(value
, default
=_dumpjson_default
, indent
=4 if '#' in flags
else None), str_fmt
1213 elif fmt
[-1] == 'h': # html
1214 value
, fmt
= escapeHTML(value
), str_fmt
1215 elif fmt
[-1] == 'q': # quoted
1216 value
= map(str, variadic(value
) if '#' in flags
else [value
])
1217 value
, fmt
= ' '.join(map(compat_shlex_quote
, value
)), str_fmt
1218 elif fmt
[-1] == 'B': # bytes
1219 value
= f
'%{str_fmt}'.encode() % str(value
).encode()
1220 value
, fmt
= value
.decode('utf-8', 'ignore'), 's'
1221 elif fmt
[-1] == 'U': # unicode normalized
1222 value
, fmt
= unicodedata
.normalize(
1223 # "+" = compatibility equivalence, "#" = NFD
1224 'NF%s%s' % ('K' if '+' in flags
else '', 'D' if '#' in flags
else 'C'),
1226 elif fmt
[-1] == 'D': # decimal suffix
1227 num_fmt
, fmt
= fmt
[:-1].replace('#', ''), 's'
1228 value
= format_decimal_suffix(value
, f
'%{num_fmt}f%s' if num_fmt
else '%d%s',
1229 factor
=1024 if '#' in flags
else 1000)
1230 elif fmt
[-1] == 'S': # filename sanitization
1231 value
, fmt
= filename_sanitizer(initial_field
, value
, restricted
='#' in flags
), str_fmt
1232 elif fmt
[-1] == 'c':
1234 value
= str(value
)[0]
1237 elif fmt
[-1] not in 'rs': # numeric
1238 value
= float_or_none(value
)
1240 value
, fmt
= default
, 's'
1244 # If value is an object, sanitize might convert it to a string
1245 # So we convert it to repr first
1246 value
, fmt
= repr(value
), str_fmt
1247 if fmt
[-1] in 'csr':
1248 value
= sanitizer(initial_field
, value
)
1250 key
= '%s\0%s' % (key
.replace('%', '%\0'), outer_mobj
.group('format'))
1251 TMPL_DICT
[key
] = value
1252 return '{prefix}%({key}){fmt}'.format(key
=key
, fmt
=fmt
, prefix
=outer_mobj
.group('prefix'))
1254 return EXTERNAL_FORMAT_RE
.sub(create_key
, outtmpl
), TMPL_DICT
1256 def evaluate_outtmpl(self
, outtmpl
, info_dict
, *args
, **kwargs
):
1257 outtmpl
, info_dict
= self
.prepare_outtmpl(outtmpl
, info_dict
, *args
, **kwargs
)
1258 return self
.escape_outtmpl(outtmpl
) % info_dict
1260 def _prepare_filename(self
, info_dict
, *, outtmpl
=None, tmpl_type
=None):
1261 assert None in (outtmpl
, tmpl_type
), 'outtmpl and tmpl_type are mutually exclusive'
1263 outtmpl
= self
.params
['outtmpl'].get(tmpl_type
or 'default', self
.params
['outtmpl']['default'])
1265 outtmpl
= self
._outtmpl
_expandpath
(outtmpl
)
1266 filename
= self
.evaluate_outtmpl(outtmpl
, info_dict
, True)
1270 if tmpl_type
in ('', 'temp'):
1271 final_ext
, ext
= self
.params
.get('final_ext'), info_dict
.get('ext')
1272 if final_ext
and ext
and final_ext
!= ext
and filename
.endswith(f
'.{final_ext}'):
1273 filename
= replace_extension(filename
, ext
, final_ext
)
1275 force_ext
= OUTTMPL_TYPES
[tmpl_type
]
1277 filename
= replace_extension(filename
, force_ext
, info_dict
.get('ext'))
1279 # https://github.com/blackjack4494/youtube-dlc/issues/85
1280 trim_file_name
= self
.params
.get('trim_file_name', False)
1282 no_ext
, *ext
= filename
.rsplit('.', 2)
1283 filename
= join_nonempty(no_ext
[:trim_file_name
], *ext
, delim
='.')
1286 except ValueError as err
:
1287 self
.report_error('Error in output template: ' + str(err
) + ' (encoding: ' + repr(preferredencoding()) + ')')
1290 def prepare_filename(self
, info_dict
, dir_type
='', *, outtmpl
=None, warn
=False):
1291 """Generate the output filename"""
1293 assert not dir_type
, 'outtmpl and dir_type are mutually exclusive'
1295 filename
= self
._prepare
_filename
(info_dict
, tmpl_type
=dir_type
, outtmpl
=outtmpl
)
1296 if not filename
and dir_type
not in ('', 'temp'):
1300 if not self
.params
.get('paths'):
1302 elif filename
== '-':
1303 self
.report_warning('--paths is ignored when an outputting to stdout', only_once
=True)
1304 elif os
.path
.isabs(filename
):
1305 self
.report_warning('--paths is ignored since an absolute path is given in output template', only_once
=True)
1306 if filename
== '-' or not filename
:
1309 return self
.get_output_path(dir_type
, filename
)
1311 def _match_entry(self
, info_dict
, incomplete
=False, silent
=False):
1312 """ Returns None if the file should be downloaded """
1314 video_title
= info_dict
.get('title', info_dict
.get('id', 'entry'))
1317 if 'title' in info_dict
:
1318 # This can happen when we're just evaluating the playlist
1319 title
= info_dict
['title']
1320 matchtitle
= self
.params
.get('matchtitle', False)
1322 if not re
.search(matchtitle
, title
, re
.IGNORECASE
):
1323 return '"' + title
+ '" title did not match pattern "' + matchtitle
+ '"'
1324 rejecttitle
= self
.params
.get('rejecttitle', False)
1326 if re
.search(rejecttitle
, title
, re
.IGNORECASE
):
1327 return '"' + title
+ '" title matched reject pattern "' + rejecttitle
+ '"'
1328 date
= info_dict
.get('upload_date')
1329 if date
is not None:
1330 dateRange
= self
.params
.get('daterange', DateRange())
1331 if date
not in dateRange
:
1332 return f
'{date_from_str(date).isoformat()} upload date is not in range {dateRange}'
1333 view_count
= info_dict
.get('view_count')
1334 if view_count
is not None:
1335 min_views
= self
.params
.get('min_views')
1336 if min_views
is not None and view_count
< min_views
:
1337 return 'Skipping %s, because it has not reached minimum view count (%d/%d)' % (video_title
, view_count
, min_views
)
1338 max_views
= self
.params
.get('max_views')
1339 if max_views
is not None and view_count
> max_views
:
1340 return 'Skipping %s, because it has exceeded the maximum view count (%d/%d)' % (video_title
, view_count
, max_views
)
1341 if age_restricted(info_dict
.get('age_limit'), self
.params
.get('age_limit')):
1342 return 'Skipping "%s" because it is age restricted' % video_title
1344 match_filter
= self
.params
.get('match_filter')
1345 if match_filter
is not None:
1347 ret
= match_filter(info_dict
, incomplete
=incomplete
)
1349 # For backward compatibility
1350 ret
= None if incomplete
else match_filter(info_dict
)
1351 if ret
is NO_DEFAULT
:
1353 filename
= self
._format
_screen
(self
.prepare_filename(info_dict
), self
.Styles
.FILENAME
)
1354 reply
= input(self
._format
_screen
(
1355 f
'Download "{filename}"? (Y/n): ', self
.Styles
.EMPHASIS
)).lower().strip()
1356 if reply
in {'y', ''}
:
1359 return f
'Skipping {video_title}'
1360 elif ret
is not None:
1364 if self
.in_download_archive(info_dict
):
1365 reason
= '%s has already been recorded in the archive' % video_title
1366 break_opt
, break_err
= 'break_on_existing', ExistingVideoReached
1368 reason
= check_filter()
1369 break_opt
, break_err
= 'break_on_reject', RejectedVideoReached
1370 if reason
is not None:
1372 self
.to_screen('[download] ' + reason
)
1373 if self
.params
.get(break_opt
, False):
1378 def add_extra_info(info_dict
, extra_info
):
1379 '''Set the keys from extra_info in info dict if they are missing'''
1380 for key
, value
in extra_info
.items():
1381 info_dict
.setdefault(key
, value
)
1383 def extract_info(self
, url
, download
=True, ie_key
=None, extra_info
=None,
1384 process
=True, force_generic_extractor
=False):
1386 Return a list with a dictionary for each video extracted.
1389 url -- URL to extract
1392 download -- whether to download videos during extraction
1393 ie_key -- extractor key hint
1394 extra_info -- dictionary containing the extra values to add to each result
1395 process -- whether to resolve all unresolved references (URLs, playlist items),
1396 must be True for download to work.
1397 force_generic_extractor -- force using the generic extractor
1400 if extra_info
is None:
1403 if not ie_key
and force_generic_extractor
:
1407 ies
= {ie_key: self._get_info_extractor_class(ie_key)}
1411 for ie_key
, ie
in ies
.items():
1412 if not ie
.suitable(url
):
1415 if not ie
.working():
1416 self
.report_warning('The program functionality for this site has been marked as broken, '
1417 'and will probably not work.')
1419 temp_id
= ie
.get_temp_id(url
)
1420 if temp_id
is not None and self
.in_download_archive({'id': temp_id, 'ie_key': ie_key}
):
1421 self
.to_screen(f
'[{ie_key}] {temp_id}: has already been recorded in the archive')
1422 if self
.params
.get('break_on_existing', False):
1423 raise ExistingVideoReached()
1425 return self
.__extract
_info
(url
, self
.get_info_extractor(ie_key
), download
, extra_info
, process
)
1427 self
.report_error('no suitable InfoExtractor for URL %s' % url
)
1429 def _handle_extraction_exceptions(func
):
1430 @functools.wraps(func
)
1431 def wrapper(self
, *args
, **kwargs
):
1434 return func(self
, *args
, **kwargs
)
1435 except (DownloadCancelled
, LazyList
.IndexError, PagedList
.IndexError):
1437 except ReExtractInfo
as e
:
1439 self
.to_screen(f
'{e}; Re-extracting data')
1441 self
.to_stderr('\r')
1442 self
.report_warning(f
'{e}; Re-extracting data')
1444 except GeoRestrictedError
as e
:
1447 msg
+= '\nThis video is available in %s.' % ', '.join(
1448 map(ISO3166Utils
.short2full
, e
.countries
))
1449 msg
+= '\nYou might want to use a VPN or a proxy server (with --proxy) to workaround.'
1450 self
.report_error(msg
)
1451 except ExtractorError
as e
: # An error we somewhat expected
1452 self
.report_error(str(e
), e
.format_traceback())
1453 except Exception as e
:
1454 if self
.params
.get('ignoreerrors'):
1455 self
.report_error(str(e
), tb
=encode_compat_str(traceback
.format_exc()))
1461 def _wait_for_video(self
, ie_result
={}):
1462 if (not self
.params
.get('wait_for_video')
1463 or ie_result
.get('_type', 'video') != 'video'
1464 or ie_result
.get('formats') or ie_result
.get('url')):
1467 format_dur
= lambda dur
: '%02d:%02d:%02d' % timetuple_from_msec(dur
* 1000)[:-1]
1472 full_msg
= f
'{msg}\n'
1473 if not self
.params
.get('noprogress'):
1474 full_msg
= msg
+ ' ' * (len(last_msg
) - len(msg
)) + '\r'
1477 self
.to_screen(full_msg
, skip_eol
=True)
1480 min_wait
, max_wait
= self
.params
.get('wait_for_video')
1481 diff
= try_get(ie_result
, lambda x
: x
['release_timestamp'] - time
.time())
1482 if diff
is None and ie_result
.get('live_status') == 'is_upcoming':
1483 diff
= round(random
.uniform(min_wait
, max_wait
) if (max_wait
and min_wait
) else (max_wait
or min_wait
), 0)
1484 self
.report_warning('Release time of video is not known')
1485 elif ie_result
and (diff
or 0) <= 0:
1486 self
.report_warning('Video should already be available according to extracted info')
1487 diff
= min(max(diff
or 0, min_wait
or 0), max_wait
or float('inf'))
1488 self
.to_screen(f
'[wait] Waiting for {format_dur(diff)} - Press Ctrl+C to try now')
1490 wait_till
= time
.time() + diff
1493 diff
= wait_till
- time
.time()
1496 raise ReExtractInfo('[wait] Wait period ended', expected
=True)
1497 progress(f
'[wait] Remaining time until next attempt: {self._format_screen(format_dur(diff), self.Styles.EMPHASIS)}')
1499 except KeyboardInterrupt:
1501 raise ReExtractInfo('[wait] Interrupted by user', expected
=True)
1502 except BaseException
as e
:
1503 if not isinstance(e
, ReExtractInfo
):
1507 @_handle_extraction_exceptions
1508 def __extract_info(self
, url
, ie
, download
, extra_info
, process
):
1510 ie_result
= ie
.extract(url
)
1511 except UserNotLive
as e
:
1513 if self
.params
.get('wait_for_video'):
1514 self
.report_warning(e
)
1515 self
._wait
_for
_video
()
1517 if ie_result
is None: # Finished already (backwards compatibility; listformats and friends should be moved here)
1518 self
.report_warning(f
'Extractor {ie.IE_NAME} returned nothing{bug_reports_message()}')
1520 if isinstance(ie_result
, list):
1521 # Backwards compatibility: old IE result format
1523 '_type': 'compat_list',
1524 'entries': ie_result
,
1526 if extra_info
.get('original_url'):
1527 ie_result
.setdefault('original_url', extra_info
['original_url'])
1528 self
.add_default_extra_info(ie_result
, ie
, url
)
1530 self
._wait
_for
_video
(ie_result
)
1531 return self
.process_ie_result(ie_result
, download
, extra_info
)
1535 def add_default_extra_info(self
, ie_result
, ie
, url
):
1537 self
.add_extra_info(ie_result
, {
1539 'original_url': url
,
1541 webpage_url
= ie_result
.get('webpage_url')
1543 self
.add_extra_info(ie_result
, {
1544 'webpage_url_basename': url_basename(webpage_url
),
1545 'webpage_url_domain': get_domain(webpage_url
),
1548 self
.add_extra_info(ie_result
, {
1549 'extractor': ie
.IE_NAME
,
1550 'extractor_key': ie
.ie_key(),
1553 def process_ie_result(self
, ie_result
, download
=True, extra_info
=None):
1555 Take the result of the ie(may be modified) and resolve all unresolved
1556 references (URLs, playlist items).
1558 It will also download the videos if 'download'.
1559 Returns the resolved ie_result.
1561 if extra_info
is None:
1563 result_type
= ie_result
.get('_type', 'video')
1565 if result_type
in ('url', 'url_transparent'):
1566 ie_result
['url'] = sanitize_url(ie_result
['url'])
1567 if ie_result
.get('original_url'):
1568 extra_info
.setdefault('original_url', ie_result
['original_url'])
1570 extract_flat
= self
.params
.get('extract_flat', False)
1571 if ((extract_flat
== 'in_playlist' and 'playlist' in extra_info
)
1572 or extract_flat
is True):
1573 info_copy
= ie_result
.copy()
1574 ie
= try_get(ie_result
.get('ie_key'), self
.get_info_extractor
)
1575 if ie
and not ie_result
.get('id'):
1576 info_copy
['id'] = ie
.get_temp_id(ie_result
['url'])
1577 self
.add_default_extra_info(info_copy
, ie
, ie_result
['url'])
1578 self
.add_extra_info(info_copy
, extra_info
)
1579 info_copy
, _
= self
.pre_process(info_copy
)
1580 self
.__forced
_printings
(info_copy
, self
.prepare_filename(info_copy
), incomplete
=True)
1581 self
._raise
_pending
_errors
(info_copy
)
1582 if self
.params
.get('force_write_download_archive', False):
1583 self
.record_download_archive(info_copy
)
1586 if result_type
== 'video':
1587 self
.add_extra_info(ie_result
, extra_info
)
1588 ie_result
= self
.process_video_result(ie_result
, download
=download
)
1589 self
._raise
_pending
_errors
(ie_result
)
1590 additional_urls
= (ie_result
or {}).get('additional_urls')
1592 # TODO: Improve MetadataParserPP to allow setting a list
1593 if isinstance(additional_urls
, str):
1594 additional_urls
= [additional_urls
]
1596 '[info] %s: %d additional URL(s) requested' % (ie_result
['id'], len(additional_urls
)))
1597 self
.write_debug('Additional URLs: "%s"' % '", "'.join(additional_urls
))
1598 ie_result
['additional_entries'] = [
1600 url
, download
, extra_info
=extra_info
,
1601 force_generic_extractor
=self
.params
.get('force_generic_extractor'))
1602 for url
in additional_urls
1605 elif result_type
== 'url':
1606 # We have to add extra_info to the results because it may be
1607 # contained in a playlist
1608 return self
.extract_info(
1609 ie_result
['url'], download
,
1610 ie_key
=ie_result
.get('ie_key'),
1611 extra_info
=extra_info
)
1612 elif result_type
== 'url_transparent':
1613 # Use the information from the embedding page
1614 info
= self
.extract_info(
1615 ie_result
['url'], ie_key
=ie_result
.get('ie_key'),
1616 extra_info
=extra_info
, download
=False, process
=False)
1618 # extract_info may return None when ignoreerrors is enabled and
1619 # extraction failed with an error, don't crash and return early
1624 exempted_fields
= {'_type', 'url', 'ie_key'}
1625 if not ie_result
.get('section_end') and ie_result
.get('section_start') is None:
1626 # For video clips, the id etc of the clip extractor should be used
1627 exempted_fields |
= {'id', 'extractor', 'extractor_key'}
1629 new_result
= info
.copy()
1630 new_result
.update(filter_dict(ie_result
, lambda k
, v
: v
is not None and k
not in exempted_fields
))
1632 # Extracted info may not be a video result (i.e.
1633 # info.get('_type', 'video') != video) but rather an url or
1634 # url_transparent. In such cases outer metadata (from ie_result)
1635 # should be propagated to inner one (info). For this to happen
1636 # _type of info should be overridden with url_transparent. This
1637 # fixes issue from https://github.com/ytdl-org/youtube-dl/pull/11163.
1638 if new_result
.get('_type') == 'url':
1639 new_result
['_type'] = 'url_transparent'
1641 return self
.process_ie_result(
1642 new_result
, download
=download
, extra_info
=extra_info
)
1643 elif result_type
in ('playlist', 'multi_video'):
1644 # Protect from infinite recursion due to recursively nested playlists
1645 # (see https://github.com/ytdl-org/youtube-dl/issues/27833)
1646 webpage_url
= ie_result
['webpage_url']
1647 if webpage_url
in self
._playlist
_urls
:
1649 '[download] Skipping already downloaded playlist: %s'
1650 % ie_result
.get('title') or ie_result
.get('id'))
1653 self
._playlist
_level
+= 1
1654 self
._playlist
_urls
.add(webpage_url
)
1655 self
._fill
_common
_fields
(ie_result
, False)
1656 self
._sanitize
_thumbnails
(ie_result
)
1658 return self
.__process
_playlist
(ie_result
, download
)
1660 self
._playlist
_level
-= 1
1661 if not self
._playlist
_level
:
1662 self
._playlist
_urls
.clear()
1663 elif result_type
== 'compat_list':
1664 self
.report_warning(
1665 'Extractor %s returned a compat_list result. '
1666 'It needs to be updated.' % ie_result
.get('extractor'))
1669 self
.add_extra_info(r
, {
1670 'extractor': ie_result
['extractor'],
1671 'webpage_url': ie_result
['webpage_url'],
1672 'webpage_url_basename': url_basename(ie_result
['webpage_url']),
1673 'webpage_url_domain': get_domain(ie_result
['webpage_url']),
1674 'extractor_key': ie_result
['extractor_key'],
1677 ie_result
['entries'] = [
1678 self
.process_ie_result(_fixup(r
), download
, extra_info
)
1679 for r
in ie_result
['entries']
1683 raise Exception('Invalid result type: %s' % result_type
)
1685 def _ensure_dir_exists(self
, path
):
1686 return make_dir(path
, self
.report_error
)
1689 def _playlist_infodict(ie_result
, strict
=False, **kwargs
):
1691 'playlist_count': ie_result
.get('playlist_count'),
1692 'playlist': ie_result
.get('title') or ie_result
.get('id'),
1693 'playlist_id': ie_result
.get('id'),
1694 'playlist_title': ie_result
.get('title'),
1695 'playlist_uploader': ie_result
.get('uploader'),
1696 'playlist_uploader_id': ie_result
.get('uploader_id'),
1703 'playlist_index': 0,
1704 '__last_playlist_index': max(ie_result
['requested_entries'] or (0, 0)),
1705 'extractor': ie_result
['extractor'],
1706 'webpage_url': ie_result
['webpage_url'],
1707 'webpage_url_basename': url_basename(ie_result
['webpage_url']),
1708 'webpage_url_domain': get_domain(ie_result
['webpage_url']),
1709 'extractor_key': ie_result
['extractor_key'],
1712 def __process_playlist(self
, ie_result
, download
):
1713 """Process each entry in the playlist"""
1714 assert ie_result
['_type'] in ('playlist', 'multi_video')
1716 common_info
= self
._playlist
_infodict
(ie_result
, strict
=True)
1717 title
= common_info
.get('playlist') or '<Untitled>'
1718 if self
._match
_entry
(common_info
, incomplete
=True) is not None:
1720 self
.to_screen(f
'[download] Downloading {ie_result["_type"]}: {title}')
1722 all_entries
= PlaylistEntries(self
, ie_result
)
1723 entries
= orderedSet(all_entries
.get_requested_items(), lazy
=True)
1725 lazy
= self
.params
.get('lazy_playlist')
1727 resolved_entries
, n_entries
= [], 'N/A'
1728 ie_result
['requested_entries'], ie_result
['entries'] = None, None
1730 entries
= resolved_entries
= list(entries
)
1731 n_entries
= len(resolved_entries
)
1732 ie_result
['requested_entries'], ie_result
['entries'] = tuple(zip(*resolved_entries
)) or ([], [])
1733 if not ie_result
.get('playlist_count'):
1734 # Better to do this after potentially exhausting entries
1735 ie_result
['playlist_count'] = all_entries
.get_full_count()
1737 ie_copy
= collections
.ChainMap(
1738 ie_result
, self
._playlist
_infodict
(ie_result
, n_entries
=int_or_none(n_entries
)))
1740 _infojson_written
= False
1741 write_playlist_files
= self
.params
.get('allow_playlist_files', True)
1742 if write_playlist_files
and self
.params
.get('list_thumbnails'):
1743 self
.list_thumbnails(ie_result
)
1744 if write_playlist_files
and not self
.params
.get('simulate'):
1745 _infojson_written
= self
._write
_info
_json
(
1746 'playlist', ie_result
, self
.prepare_filename(ie_copy
, 'pl_infojson'))
1747 if _infojson_written
is None:
1749 if self
._write
_description
('playlist', ie_result
,
1750 self
.prepare_filename(ie_copy
, 'pl_description')) is None:
1752 # TODO: This should be passed to ThumbnailsConvertor if necessary
1753 self
._write
_thumbnails
('playlist', ie_result
, self
.prepare_filename(ie_copy
, 'pl_thumbnail'))
1756 if self
.params
.get('playlistreverse') or self
.params
.get('playlistrandom'):
1757 self
.report_warning('playlistreverse and playlistrandom are not supported with lazy_playlist', only_once
=True)
1758 elif self
.params
.get('playlistreverse'):
1760 elif self
.params
.get('playlistrandom'):
1761 random
.shuffle(entries
)
1763 self
.to_screen(f
'[{ie_result["extractor"]}] Playlist {title}: Downloading {n_entries} videos'
1764 f
'{format_field(ie_result, "playlist_count", " of %s")}')
1766 keep_resolved_entries
= self
.params
.get('extract_flat') != 'discard'
1767 if self
.params
.get('extract_flat') == 'discard_in_playlist':
1768 keep_resolved_entries
= ie_result
['_type'] != 'playlist'
1769 if keep_resolved_entries
:
1770 self
.write_debug('The information of all playlist entries will be held in memory')
1773 max_failures
= self
.params
.get('skip_playlist_after_errors') or float('inf')
1774 for i
, (playlist_index
, entry
) in enumerate(entries
):
1776 resolved_entries
.append((playlist_index
, entry
))
1780 entry
['__x_forwarded_for_ip'] = ie_result
.get('__x_forwarded_for_ip')
1781 if not lazy
and 'playlist-index' in self
.params
.get('compat_opts', []):
1782 playlist_index
= ie_result
['requested_entries'][i
]
1786 'n_entries': int_or_none(n_entries
),
1787 'playlist_index': playlist_index
,
1788 'playlist_autonumber': i
+ 1,
1791 if self
._match
_entry
(collections
.ChainMap(entry
, extra
), incomplete
=True) is not None:
1794 self
.to_screen('[download] Downloading video %s of %s' % (
1795 self
._format
_screen
(i
+ 1, self
.Styles
.ID
), self
._format
_screen
(n_entries
, self
.Styles
.EMPHASIS
)))
1797 entry_result
= self
.__process
_iterable
_entry
(entry
, download
, extra
)
1798 if not entry_result
:
1800 if failures
>= max_failures
:
1802 f
'Skipping the remaining entries in playlist "{title}" since {failures} items failed extraction')
1804 if keep_resolved_entries
:
1805 resolved_entries
[i
] = (playlist_index
, entry_result
)
1807 # Update with processed data
1808 ie_result
['requested_entries'], ie_result
['entries'] = tuple(zip(*resolved_entries
)) or ([], [])
1810 # Write the updated info to json
1811 if _infojson_written
is True and self
._write
_info
_json
(
1812 'updated playlist', ie_result
,
1813 self
.prepare_filename(ie_copy
, 'pl_infojson'), overwrite
=True) is None:
1816 ie_result
= self
.run_all_pps('playlist', ie_result
)
1817 self
.to_screen(f
'[download] Finished downloading playlist: {title}')
1820 @_handle_extraction_exceptions
1821 def __process_iterable_entry(self
, entry
, download
, extra_info
):
1822 return self
.process_ie_result(
1823 entry
, download
=download
, extra_info
=extra_info
)
1825 def _build_format_filter(self
, filter_spec
):
1826 " Returns a function to filter the formats according to the filter_spec "
1836 operator_rex
= re
.compile(r
'''(?x)\s*
1837 (?P<key>width|height|tbr|abr|vbr|asr|filesize|filesize_approx|fps)\s*
1838 (?P<op>%s)(?P<none_inclusive>\s*\?)?\s*
1839 (?P<value>[0-9.]+(?:[kKmMgGtTpPeEzZyY]i?[Bb]?)?)\s*
1840 ''' % '|'.join(map(re
.escape
, OPERATORS
.keys())))
1841 m
= operator_rex
.fullmatch(filter_spec
)
1844 comparison_value
= int(m
.group('value'))
1846 comparison_value
= parse_filesize(m
.group('value'))
1847 if comparison_value
is None:
1848 comparison_value
= parse_filesize(m
.group('value') + 'B')
1849 if comparison_value
is None:
1851 'Invalid value %r in format specification %r' % (
1852 m
.group('value'), filter_spec
))
1853 op
= OPERATORS
[m
.group('op')]
1858 '^=': lambda attr
, value
: attr
.startswith(value
),
1859 '$=': lambda attr
, value
: attr
.endswith(value
),
1860 '*=': lambda attr
, value
: value
in attr
,
1861 '~=': lambda attr
, value
: value
.search(attr
) is not None
1863 str_operator_rex
= re
.compile(r
'''(?x)\s*
1864 (?P<key>[a-zA-Z0-9._-]+)\s*
1865 (?P<negation>!\s*)?(?P<op>%s)\s*(?P<none_inclusive>\?\s*)?
1867 (?P<value>(?(quote)(?:(?!(?P=quote))[^\\]|\\.)+|[\w.-]+))
1868 (?(quote)(?P=quote))\s*
1869 ''' % '|'.join(map(re
.escape
, STR_OPERATORS
.keys())))
1870 m
= str_operator_rex
.fullmatch(filter_spec
)
1872 if m
.group('op') == '~=':
1873 comparison_value
= re
.compile(m
.group('value'))
1875 comparison_value
= re
.sub(r
'''\\([\\"'])''', r
'\1', m
.group('value'))
1876 str_op
= STR_OPERATORS
[m
.group('op')]
1877 if m
.group('negation'):
1878 op
= lambda attr
, value
: not str_op(attr
, value
)
1883 raise SyntaxError('Invalid filter specification %r' % filter_spec
)
1886 actual_value
= f
.get(m
.group('key'))
1887 if actual_value
is None:
1888 return m
.group('none_inclusive')
1889 return op(actual_value
, comparison_value
)
1892 def _check_formats(self
, formats
):
1894 self
.to_screen('[info] Testing format %s' % f
['format_id'])
1895 path
= self
.get_output_path('temp')
1896 if not self
._ensure
_dir
_exists
(f
'{path}/'):
1898 temp_file
= tempfile
.NamedTemporaryFile(suffix
='.tmp', delete
=False, dir=path
or None)
1901 success
, _
= self
.dl(temp_file
.name
, f
, test
=True)
1902 except (DownloadError
, OSError, ValueError) + network_exceptions
:
1905 if os
.path
.exists(temp_file
.name
):
1907 os
.remove(temp_file
.name
)
1909 self
.report_warning('Unable to delete temporary file "%s"' % temp_file
.name
)
1913 self
.to_screen('[info] Unable to download format %s. Skipping...' % f
['format_id'])
1915 def _default_format_spec(self
, info_dict
, download
=True):
1918 merger
= FFmpegMergerPP(self
)
1919 return merger
.available
and merger
.can_merge()
1922 not self
.params
.get('simulate')
1926 or info_dict
.get('is_live') and not self
.params
.get('live_from_start')
1927 or self
.params
['outtmpl']['default'] == '-'))
1930 or self
.params
.get('allow_multiple_audio_streams', False)
1931 or 'format-spec' in self
.params
['compat_opts'])
1934 'best/bestvideo+bestaudio' if prefer_best
1935 else 'bestvideo*+bestaudio/best' if not compat
1936 else 'bestvideo+bestaudio/best')
1938 def build_format_selector(self
, format_spec
):
1939 def syntax_error(note
, start
):
1941 'Invalid format specification: '
1942 '{}\n\t{}\n\t{}^'.format(note
, format_spec
, ' ' * start
[1]))
1943 return SyntaxError(message
)
1945 PICKFIRST
= 'PICKFIRST'
1949 FormatSelector
= collections
.namedtuple('FormatSelector', ['type', 'selector', 'filters'])
1951 allow_multiple_streams
= {'audio': self
.params
.get('allow_multiple_audio_streams', False),
1952 'video': self
.params
.get('allow_multiple_video_streams', False)}
1954 check_formats
= self
.params
.get('check_formats') == 'selected'
1956 def _parse_filter(tokens
):
1958 for type, string
, start
, _
, _
in tokens
:
1959 if type == tokenize
.OP
and string
== ']':
1960 return ''.join(filter_parts
)
1962 filter_parts
.append(string
)
1964 def _remove_unused_ops(tokens
):
1965 # Remove operators that we don't use and join them with the surrounding strings
1966 # for example: 'mp4' '-' 'baseline' '-' '16x9' is converted to 'mp4-baseline-16x9'
1967 ALLOWED_OPS
= ('/', '+', ',', '(', ')')
1968 last_string
, last_start
, last_end
, last_line
= None, None, None, None
1969 for type, string
, start
, end
, line
in tokens
:
1970 if type == tokenize
.OP
and string
== '[':
1972 yield tokenize
.NAME
, last_string
, last_start
, last_end
, last_line
1974 yield type, string
, start
, end
, line
1975 # everything inside brackets will be handled by _parse_filter
1976 for type, string
, start
, end
, line
in tokens
:
1977 yield type, string
, start
, end
, line
1978 if type == tokenize
.OP
and string
== ']':
1980 elif type == tokenize
.OP
and string
in ALLOWED_OPS
:
1982 yield tokenize
.NAME
, last_string
, last_start
, last_end
, last_line
1984 yield type, string
, start
, end
, line
1985 elif type in [tokenize
.NAME
, tokenize
.NUMBER
, tokenize
.OP
]:
1987 last_string
= string
1991 last_string
+= string
1993 yield tokenize
.NAME
, last_string
, last_start
, last_end
, last_line
1995 def _parse_format_selection(tokens
, inside_merge
=False, inside_choice
=False, inside_group
=False):
1997 current_selector
= None
1998 for type, string
, start
, _
, _
in tokens
:
1999 # ENCODING is only defined in python 3.x
2000 if type == getattr(tokenize
, 'ENCODING', None):
2002 elif type in [tokenize
.NAME
, tokenize
.NUMBER
]:
2003 current_selector
= FormatSelector(SINGLE
, string
, [])
2004 elif type == tokenize
.OP
:
2006 if not inside_group
:
2007 # ')' will be handled by the parentheses group
2008 tokens
.restore_last_token()
2010 elif inside_merge
and string
in ['/', ',']:
2011 tokens
.restore_last_token()
2013 elif inside_choice
and string
== ',':
2014 tokens
.restore_last_token()
2017 if not current_selector
:
2018 raise syntax_error('"," must follow a format selector', start
)
2019 selectors
.append(current_selector
)
2020 current_selector
= None
2022 if not current_selector
:
2023 raise syntax_error('"/" must follow a format selector', start
)
2024 first_choice
= current_selector
2025 second_choice
= _parse_format_selection(tokens
, inside_choice
=True)
2026 current_selector
= FormatSelector(PICKFIRST
, (first_choice
, second_choice
), [])
2028 if not current_selector
:
2029 current_selector
= FormatSelector(SINGLE
, 'best', [])
2030 format_filter
= _parse_filter(tokens
)
2031 current_selector
.filters
.append(format_filter
)
2033 if current_selector
:
2034 raise syntax_error('Unexpected "("', start
)
2035 group
= _parse_format_selection(tokens
, inside_group
=True)
2036 current_selector
= FormatSelector(GROUP
, group
, [])
2038 if not current_selector
:
2039 raise syntax_error('Unexpected "+"', start
)
2040 selector_1
= current_selector
2041 selector_2
= _parse_format_selection(tokens
, inside_merge
=True)
2043 raise syntax_error('Expected a selector', start
)
2044 current_selector
= FormatSelector(MERGE
, (selector_1
, selector_2
), [])
2046 raise syntax_error(f
'Operator not recognized: "{string}"', start
)
2047 elif type == tokenize
.ENDMARKER
:
2049 if current_selector
:
2050 selectors
.append(current_selector
)
2053 def _merge(formats_pair
):
2054 format_1
, format_2
= formats_pair
2057 formats_info
.extend(format_1
.get('requested_formats', (format_1
,)))
2058 formats_info
.extend(format_2
.get('requested_formats', (format_2
,)))
2060 if not allow_multiple_streams
['video'] or not allow_multiple_streams
['audio']:
2061 get_no_more
= {'video': False, 'audio': False}
2062 for (i
, fmt_info
) in enumerate(formats_info
):
2063 if fmt_info
.get('acodec') == fmt_info
.get('vcodec') == 'none':
2066 for aud_vid
in ['audio', 'video']:
2067 if not allow_multiple_streams
[aud_vid
] and fmt_info
.get(aud_vid
[0] + 'codec') != 'none':
2068 if get_no_more
[aud_vid
]:
2071 get_no_more
[aud_vid
] = True
2073 if len(formats_info
) == 1:
2074 return formats_info
[0]
2076 video_fmts
= [fmt_info
for fmt_info
in formats_info
if fmt_info
.get('vcodec') != 'none']
2077 audio_fmts
= [fmt_info
for fmt_info
in formats_info
if fmt_info
.get('acodec') != 'none']
2079 the_only_video
= video_fmts
[0] if len(video_fmts
) == 1 else None
2080 the_only_audio
= audio_fmts
[0] if len(audio_fmts
) == 1 else None
2082 output_ext
= self
.params
.get('merge_output_format')
2085 output_ext
= the_only_video
['ext']
2086 elif the_only_audio
and not video_fmts
:
2087 output_ext
= the_only_audio
['ext']
2091 filtered
= lambda *keys
: filter(None, (traverse_obj(fmt
, *keys
) for fmt
in formats_info
))
2094 'requested_formats': formats_info
,
2095 'format': '+'.join(filtered('format')),
2096 'format_id': '+'.join(filtered('format_id')),
2098 'protocol': '+'.join(map(determine_protocol
, formats_info
)),
2099 'language': '+'.join(orderedSet(filtered('language'))) or None,
2100 'format_note': '+'.join(orderedSet(filtered('format_note'))) or None,
2101 'filesize_approx': sum(filtered('filesize', 'filesize_approx')) or None,
2102 'tbr': sum(filtered('tbr', 'vbr', 'abr')),
2107 'width': the_only_video
.get('width'),
2108 'height': the_only_video
.get('height'),
2109 'resolution': the_only_video
.get('resolution') or self
.format_resolution(the_only_video
),
2110 'fps': the_only_video
.get('fps'),
2111 'dynamic_range': the_only_video
.get('dynamic_range'),
2112 'vcodec': the_only_video
.get('vcodec'),
2113 'vbr': the_only_video
.get('vbr'),
2114 'stretched_ratio': the_only_video
.get('stretched_ratio'),
2119 'acodec': the_only_audio
.get('acodec'),
2120 'abr': the_only_audio
.get('abr'),
2121 'asr': the_only_audio
.get('asr'),
2126 def _check_formats(formats
):
2127 if not check_formats
:
2130 yield from self
._check
_formats
(formats
)
2132 def _build_selector_function(selector
):
2133 if isinstance(selector
, list): # ,
2134 fs
= [_build_selector_function(s
) for s
in selector
]
2136 def selector_function(ctx
):
2139 return selector_function
2141 elif selector
.type == GROUP
: # ()
2142 selector_function
= _build_selector_function(selector
.selector
)
2144 elif selector
.type == PICKFIRST
: # /
2145 fs
= [_build_selector_function(s
) for s
in selector
.selector
]
2147 def selector_function(ctx
):
2149 picked_formats
= list(f(ctx
))
2151 return picked_formats
2154 elif selector
.type == MERGE
: # +
2155 selector_1
, selector_2
= map(_build_selector_function
, selector
.selector
)
2157 def selector_function(ctx
):
2158 for pair
in itertools
.product(selector_1(ctx
), selector_2(ctx
)):
2161 elif selector
.type == SINGLE
: # atom
2162 format_spec
= selector
.selector
or 'best'
2164 # TODO: Add allvideo, allaudio etc by generalizing the code with best/worst selector
2165 if format_spec
== 'all':
2166 def selector_function(ctx
):
2167 yield from _check_formats(ctx
['formats'][::-1])
2168 elif format_spec
== 'mergeall':
2169 def selector_function(ctx
):
2170 formats
= list(_check_formats(
2171 f
for f
in ctx
['formats'] if f
.get('vcodec') != 'none' or f
.get('acodec') != 'none'))
2174 merged_format
= formats
[-1]
2175 for f
in formats
[-2::-1]:
2176 merged_format
= _merge((merged_format
, f
))
2180 format_fallback
, seperate_fallback
, format_reverse
, format_idx
= False, None, True, 1
2182 r
'(?P<bw>best|worst|b|w)(?P<type>video|audio|v|a)?(?P<mod>\*)?(?:\.(?P<n>[1-9]\d*))?$',
2184 if mobj
is not None:
2185 format_idx
= int_or_none(mobj
.group('n'), default
=1)
2186 format_reverse
= mobj
.group('bw')[0] == 'b'
2187 format_type
= (mobj
.group('type') or [None])[0]
2188 not_format_type
= {'v': 'a', 'a': 'v'}
.get(format_type
)
2189 format_modified
= mobj
.group('mod') is not None
2191 format_fallback
= not format_type
and not format_modified
# for b, w
2193 (lambda f
: f
.get('%scodec' % format_type
) != 'none')
2194 if format_type
and format_modified
# bv*, ba*, wv*, wa*
2195 else (lambda f
: f
.get('%scodec' % not_format_type
) == 'none')
2196 if format_type
# bv, ba, wv, wa
2197 else (lambda f
: f
.get('vcodec') != 'none' and f
.get('acodec') != 'none')
2198 if not format_modified
# b, w
2199 else lambda f
: True) # b*, w*
2200 filter_f
= lambda f
: _filter_f(f
) and (
2201 f
.get('vcodec') != 'none' or f
.get('acodec') != 'none')
2203 if format_spec
in self
._format
_selection
_exts
['audio']:
2204 filter_f
= lambda f
: f
.get('ext') == format_spec
and f
.get('acodec') != 'none'
2205 elif format_spec
in self
._format
_selection
_exts
['video']:
2206 filter_f
= lambda f
: f
.get('ext') == format_spec
and f
.get('acodec') != 'none' and f
.get('vcodec') != 'none'
2207 seperate_fallback
= lambda f
: f
.get('ext') == format_spec
and f
.get('vcodec') != 'none'
2208 elif format_spec
in self
._format
_selection
_exts
['storyboards']:
2209 filter_f
= lambda f
: f
.get('ext') == format_spec
and f
.get('acodec') == 'none' and f
.get('vcodec') == 'none'
2211 filter_f
= lambda f
: f
.get('format_id') == format_spec
# id
2213 def selector_function(ctx
):
2214 formats
= list(ctx
['formats'])
2215 matches
= list(filter(filter_f
, formats
)) if filter_f
is not None else formats
2217 if format_fallback
and ctx
['incomplete_formats']:
2218 # for extractors with incomplete formats (audio only (soundcloud)
2219 # or video only (imgur)) best/worst will fallback to
2220 # best/worst {video,audio}-only format
2222 elif seperate_fallback
and not ctx
['has_merged_format']:
2223 # for compatibility with youtube-dl when there is no pre-merged format
2224 matches
= list(filter(seperate_fallback
, formats
))
2225 matches
= LazyList(_check_formats(matches
[::-1 if format_reverse
else 1]))
2227 yield matches
[format_idx
- 1]
2228 except LazyList
.IndexError:
2231 filters
= [self
._build
_format
_filter
(f
) for f
in selector
.filters
]
2233 def final_selector(ctx
):
2234 ctx_copy
= dict(ctx
)
2235 for _filter
in filters
:
2236 ctx_copy
['formats'] = list(filter(_filter
, ctx_copy
['formats']))
2237 return selector_function(ctx_copy
)
2238 return final_selector
2240 stream
= io
.BytesIO(format_spec
.encode())
2242 tokens
= list(_remove_unused_ops(tokenize
.tokenize(stream
.readline
)))
2243 except tokenize
.TokenError
:
2244 raise syntax_error('Missing closing/opening brackets or parenthesis', (0, len(format_spec
)))
2246 class TokenIterator
:
2247 def __init__(self
, tokens
):
2248 self
.tokens
= tokens
2255 if self
.counter
>= len(self
.tokens
):
2256 raise StopIteration()
2257 value
= self
.tokens
[self
.counter
]
2263 def restore_last_token(self
):
2266 parsed_selector
= _parse_format_selection(iter(TokenIterator(tokens
)))
2267 return _build_selector_function(parsed_selector
)
2269 def _calc_headers(self
, info_dict
):
2270 res
= merge_headers(self
.params
['http_headers'], info_dict
.get('http_headers') or {})
2272 cookies
= self
._calc
_cookies
(info_dict
['url'])
2274 res
['Cookie'] = cookies
2276 if 'X-Forwarded-For' not in res
:
2277 x_forwarded_for_ip
= info_dict
.get('__x_forwarded_for_ip')
2278 if x_forwarded_for_ip
:
2279 res
['X-Forwarded-For'] = x_forwarded_for_ip
2283 def _calc_cookies(self
, url
):
2284 pr
= sanitized_Request(url
)
2285 self
.cookiejar
.add_cookie_header(pr
)
2286 return pr
.get_header('Cookie')
2288 def _sort_thumbnails(self
, thumbnails
):
2289 thumbnails
.sort(key
=lambda t
: (
2290 t
.get('preference') if t
.get('preference') is not None else -1,
2291 t
.get('width') if t
.get('width') is not None else -1,
2292 t
.get('height') if t
.get('height') is not None else -1,
2293 t
.get('id') if t
.get('id') is not None else '',
2296 def _sanitize_thumbnails(self
, info_dict
):
2297 thumbnails
= info_dict
.get('thumbnails')
2298 if thumbnails
is None:
2299 thumbnail
= info_dict
.get('thumbnail')
2301 info_dict
['thumbnails'] = thumbnails
= [{'url': thumbnail}
]
2305 def check_thumbnails(thumbnails
):
2306 for t
in thumbnails
:
2307 self
.to_screen(f
'[info] Testing thumbnail {t["id"]}')
2309 self
.urlopen(HEADRequest(t
['url']))
2310 except network_exceptions
as err
:
2311 self
.to_screen(f
'[info] Unable to connect to thumbnail {t["id"]} URL {t["url"]!r} - {err}. Skipping...')
2315 self
._sort
_thumbnails
(thumbnails
)
2316 for i
, t
in enumerate(thumbnails
):
2317 if t
.get('id') is None:
2319 if t
.get('width') and t
.get('height'):
2320 t
['resolution'] = '%dx%d' % (t
['width'], t
['height'])
2321 t
['url'] = sanitize_url(t
['url'])
2323 if self
.params
.get('check_formats') is True:
2324 info_dict
['thumbnails'] = LazyList(check_thumbnails(thumbnails
[::-1]), reverse
=True)
2326 info_dict
['thumbnails'] = thumbnails
2328 def _fill_common_fields(self
, info_dict
, is_video
=True):
2329 # TODO: move sanitization here
2331 # playlists are allowed to lack "title"
2332 title
= info_dict
.get('title', NO_DEFAULT
)
2333 if title
is NO_DEFAULT
:
2334 raise ExtractorError('Missing "title" field in extractor result',
2335 video_id
=info_dict
['id'], ie
=info_dict
['extractor'])
2336 info_dict
['fulltitle'] = title
2339 self
.write_debug('Extractor gave empty title. Creating a generic title')
2341 self
.report_warning('Extractor failed to obtain "title". Creating a generic title instead')
2342 info_dict
['title'] = f
'{info_dict["extractor"].replace(":", "-")} video #{info_dict["id"]}'
2344 if info_dict
.get('duration') is not None:
2345 info_dict
['duration_string'] = formatSeconds(info_dict
['duration'])
2347 for ts_key
, date_key
in (
2348 ('timestamp', 'upload_date'),
2349 ('release_timestamp', 'release_date'),
2350 ('modified_timestamp', 'modified_date'),
2352 if info_dict
.get(date_key
) is None and info_dict
.get(ts_key
) is not None:
2353 # Working around out-of-range timestamp values (e.g. negative ones on Windows,
2354 # see http://bugs.python.org/issue1646728)
2355 with contextlib
.suppress(ValueError, OverflowError, OSError):
2356 upload_date
= datetime
.datetime
.utcfromtimestamp(info_dict
[ts_key
])
2357 info_dict
[date_key
] = upload_date
.strftime('%Y%m%d')
2359 live_keys
= ('is_live', 'was_live')
2360 live_status
= info_dict
.get('live_status')
2361 if live_status
is None:
2362 for key
in live_keys
:
2363 if info_dict
.get(key
) is False:
2365 if info_dict
.get(key
):
2368 if all(info_dict
.get(key
) is False for key
in live_keys
):
2369 live_status
= 'not_live'
2371 info_dict
['live_status'] = live_status
2372 for key
in live_keys
:
2373 if info_dict
.get(key
) is None:
2374 info_dict
[key
] = (live_status
== key
)
2376 # Auto generate title fields corresponding to the *_number fields when missing
2377 # in order to always have clean titles. This is very common for TV series.
2378 for field
in ('chapter', 'season', 'episode'):
2379 if info_dict
.get('%s_number' % field
) is not None and not info_dict
.get(field
):
2380 info_dict
[field
] = '%s %d' % (field
.capitalize(), info_dict
['%s_number' % field
])
2382 def _raise_pending_errors(self
, info
):
2383 err
= info
.pop('__pending_error', None)
2385 self
.report_error(err
, tb
=False)
2387 def process_video_result(self
, info_dict
, download
=True):
2388 assert info_dict
.get('_type', 'video') == 'video'
2389 self
._num
_videos
+= 1
2391 if 'id' not in info_dict
:
2392 raise ExtractorError('Missing "id" field in extractor result', ie
=info_dict
['extractor'])
2393 elif not info_dict
.get('id'):
2394 raise ExtractorError('Extractor failed to obtain "id"', ie
=info_dict
['extractor'])
2396 def report_force_conversion(field
, field_not
, conversion
):
2397 self
.report_warning(
2398 '"%s" field is not %s - forcing %s conversion, there is an error in extractor'
2399 % (field
, field_not
, conversion
))
2401 def sanitize_string_field(info
, string_field
):
2402 field
= info
.get(string_field
)
2403 if field
is None or isinstance(field
, str):
2405 report_force_conversion(string_field
, 'a string', 'string')
2406 info
[string_field
] = str(field
)
2408 def sanitize_numeric_fields(info
):
2409 for numeric_field
in self
._NUMERIC
_FIELDS
:
2410 field
= info
.get(numeric_field
)
2411 if field
is None or isinstance(field
, (int, float)):
2413 report_force_conversion(numeric_field
, 'numeric', 'int')
2414 info
[numeric_field
] = int_or_none(field
)
2416 sanitize_string_field(info_dict
, 'id')
2417 sanitize_numeric_fields(info_dict
)
2418 if info_dict
.get('section_end') and info_dict
.get('section_start') is not None:
2419 info_dict
['duration'] = round(info_dict
['section_end'] - info_dict
['section_start'], 3)
2420 if (info_dict
.get('duration') or 0) <= 0 and info_dict
.pop('duration', None):
2421 self
.report_warning('"duration" field is negative, there is an error in extractor')
2423 chapters
= info_dict
.get('chapters') or []
2424 if chapters
and chapters
[0].get('start_time'):
2425 chapters
.insert(0, {'start_time': 0}
)
2427 dummy_chapter
= {'end_time': 0, 'start_time': info_dict.get('duration')}
2428 for idx
, (prev
, current
, next_
) in enumerate(zip(
2429 (dummy_chapter
, *chapters
), chapters
, (*chapters
[1:], dummy_chapter
)), 1):
2430 if current
.get('start_time') is None:
2431 current
['start_time'] = prev
.get('end_time')
2432 if not current
.get('end_time'):
2433 current
['end_time'] = next_
.get('start_time')
2434 if not current
.get('title'):
2435 current
['title'] = f
'<Untitled Chapter {idx}>'
2437 if 'playlist' not in info_dict
:
2438 # It isn't part of a playlist
2439 info_dict
['playlist'] = None
2440 info_dict
['playlist_index'] = None
2442 self
._sanitize
_thumbnails
(info_dict
)
2444 thumbnail
= info_dict
.get('thumbnail')
2445 thumbnails
= info_dict
.get('thumbnails')
2447 info_dict
['thumbnail'] = sanitize_url(thumbnail
)
2449 info_dict
['thumbnail'] = thumbnails
[-1]['url']
2451 if info_dict
.get('display_id') is None and 'id' in info_dict
:
2452 info_dict
['display_id'] = info_dict
['id']
2454 self
._fill
_common
_fields
(info_dict
)
2456 for cc_kind
in ('subtitles', 'automatic_captions'):
2457 cc
= info_dict
.get(cc_kind
)
2459 for _
, subtitle
in cc
.items():
2460 for subtitle_format
in subtitle
:
2461 if subtitle_format
.get('url'):
2462 subtitle_format
['url'] = sanitize_url(subtitle_format
['url'])
2463 if subtitle_format
.get('ext') is None:
2464 subtitle_format
['ext'] = determine_ext(subtitle_format
['url']).lower()
2466 automatic_captions
= info_dict
.get('automatic_captions')
2467 subtitles
= info_dict
.get('subtitles')
2469 info_dict
['requested_subtitles'] = self
.process_subtitles(
2470 info_dict
['id'], subtitles
, automatic_captions
)
2472 if info_dict
.get('formats') is None:
2473 # There's only one format available
2474 formats
= [info_dict
]
2476 formats
= info_dict
['formats']
2478 # or None ensures --clean-infojson removes it
2479 info_dict
['_has_drm'] = any(f
.get('has_drm') for f
in formats
) or None
2480 if not self
.params
.get('allow_unplayable_formats'):
2481 formats
= [f
for f
in formats
if not f
.get('has_drm')]
2482 if info_dict
['_has_drm'] and all(
2483 f
.get('acodec') == f
.get('vcodec') == 'none' for f
in formats
):
2484 self
.report_warning(
2485 'This video is DRM protected and only images are available for download. '
2486 'Use --list-formats to see them')
2488 get_from_start
= not info_dict
.get('is_live') or bool(self
.params
.get('live_from_start'))
2489 if not get_from_start
:
2490 info_dict
['title'] += ' ' + datetime
.datetime
.now().strftime('%Y-%m-%d %H:%M')
2491 if info_dict
.get('is_live') and formats
:
2492 formats
= [f
for f
in formats
if bool(f
.get('is_from_start')) == get_from_start
]
2493 if get_from_start
and not formats
:
2494 self
.raise_no_formats(info_dict
, msg
=(
2495 '--live-from-start is passed, but there are no formats that can be downloaded from the start. '
2496 'If you want to download from the current time, use --no-live-from-start'))
2499 self
.raise_no_formats(info_dict
)
2501 def is_wellformed(f
):
2504 self
.report_warning(
2505 '"url" field is missing or empty - skipping format, '
2506 'there is an error in extractor')
2508 if isinstance(url
, bytes):
2509 sanitize_string_field(f
, 'url')
2512 # Filter out malformed formats for better extraction robustness
2513 formats
= list(filter(is_wellformed
, formats
))
2517 # We check that all the formats have the format and format_id fields
2518 for i
, format
in enumerate(formats
):
2519 sanitize_string_field(format
, 'format_id')
2520 sanitize_numeric_fields(format
)
2521 format
['url'] = sanitize_url(format
['url'])
2522 if not format
.get('format_id'):
2523 format
['format_id'] = str(i
)
2525 # Sanitize format_id from characters used in format selector expression
2526 format
['format_id'] = re
.sub(r
'[\s,/+\[\]()]', '_', format
['format_id'])
2527 format_id
= format
['format_id']
2528 if format_id
not in formats_dict
:
2529 formats_dict
[format_id
] = []
2530 formats_dict
[format_id
].append(format
)
2532 # Make sure all formats have unique format_id
2533 common_exts
= set(itertools
.chain(*self
._format
_selection
_exts
.values()))
2534 for format_id
, ambiguous_formats
in formats_dict
.items():
2535 ambigious_id
= len(ambiguous_formats
) > 1
2536 for i
, format
in enumerate(ambiguous_formats
):
2538 format
['format_id'] = '%s-%d' % (format_id
, i
)
2539 if format
.get('ext') is None:
2540 format
['ext'] = determine_ext(format
['url']).lower()
2541 # Ensure there is no conflict between id and ext in format selection
2542 # See https://github.com/yt-dlp/yt-dlp/issues/1282
2543 if format
['format_id'] != format
['ext'] and format
['format_id'] in common_exts
:
2544 format
['format_id'] = 'f%s' % format
['format_id']
2546 for i
, format
in enumerate(formats
):
2547 if format
.get('format') is None:
2548 format
['format'] = '{id} - {res}{note}'.format(
2549 id=format
['format_id'],
2550 res
=self
.format_resolution(format
),
2551 note
=format_field(format
, 'format_note', ' (%s)'),
2553 if format
.get('protocol') is None:
2554 format
['protocol'] = determine_protocol(format
)
2555 if format
.get('resolution') is None:
2556 format
['resolution'] = self
.format_resolution(format
, default
=None)
2557 if format
.get('dynamic_range') is None and format
.get('vcodec') != 'none':
2558 format
['dynamic_range'] = 'SDR'
2559 if (info_dict
.get('duration') and format
.get('tbr')
2560 and not format
.get('filesize') and not format
.get('filesize_approx')):
2561 format
['filesize_approx'] = int(info_dict
['duration'] * format
['tbr'] * (1024 / 8))
2563 # Add HTTP headers, so that external programs can use them from the
2565 full_format_info
= info_dict
.copy()
2566 full_format_info
.update(format
)
2567 format
['http_headers'] = self
._calc
_headers
(full_format_info
)
2568 # Remove private housekeeping stuff
2569 if '__x_forwarded_for_ip' in info_dict
:
2570 del info_dict
['__x_forwarded_for_ip']
2572 if self
.params
.get('check_formats') is True:
2573 formats
= LazyList(self
._check
_formats
(formats
[::-1]), reverse
=True)
2575 if not formats
or formats
[0] is not info_dict
:
2576 # only set the 'formats' fields if the original info_dict list them
2577 # otherwise we end up with a circular reference, the first (and unique)
2578 # element in the 'formats' field in info_dict is info_dict itself,
2579 # which can't be exported to json
2580 info_dict
['formats'] = formats
2582 info_dict
, _
= self
.pre_process(info_dict
)
2584 if self
._match
_entry
(info_dict
, incomplete
=self
._format
_fields
) is not None:
2587 self
.post_extract(info_dict
)
2588 info_dict
, _
= self
.pre_process(info_dict
, 'after_filter')
2590 # The pre-processors may have modified the formats
2591 formats
= info_dict
.get('formats', [info_dict
])
2593 list_only
= self
.params
.get('simulate') is None and (
2594 self
.params
.get('list_thumbnails') or self
.params
.get('listformats') or self
.params
.get('listsubtitles'))
2595 interactive_format_selection
= not list_only
and self
.format_selector
== '-'
2596 if self
.params
.get('list_thumbnails'):
2597 self
.list_thumbnails(info_dict
)
2598 if self
.params
.get('listsubtitles'):
2599 if 'automatic_captions' in info_dict
:
2600 self
.list_subtitles(
2601 info_dict
['id'], automatic_captions
, 'automatic captions')
2602 self
.list_subtitles(info_dict
['id'], subtitles
, 'subtitles')
2603 if self
.params
.get('listformats') or interactive_format_selection
:
2604 self
.list_formats(info_dict
)
2606 # Without this printing, -F --print-json will not work
2607 self
.__forced
_printings
(info_dict
, self
.prepare_filename(info_dict
), incomplete
=True)
2610 format_selector
= self
.format_selector
2611 if format_selector
is None:
2612 req_format
= self
._default
_format
_spec
(info_dict
, download
=download
)
2613 self
.write_debug('Default format spec: %s' % req_format
)
2614 format_selector
= self
.build_format_selector(req_format
)
2617 if interactive_format_selection
:
2619 self
._format
_screen
('\nEnter format selector: ', self
.Styles
.EMPHASIS
))
2621 format_selector
= self
.build_format_selector(req_format
)
2622 except SyntaxError as err
:
2623 self
.report_error(err
, tb
=False, is_error
=False)
2626 formats_to_download
= list(format_selector({
2628 'has_merged_format': any('none' not in (f
.get('acodec'), f
.get('vcodec')) for f
in formats
),
2629 'incomplete_formats': (
2630 # All formats are video-only or
2631 all(f
.get('vcodec') != 'none' and f
.get('acodec') == 'none' for f
in formats
)
2632 # all formats are audio-only
2633 or all(f
.get('vcodec') == 'none' and f
.get('acodec') != 'none' for f
in formats
)),
2635 if interactive_format_selection
and not formats_to_download
:
2636 self
.report_error('Requested format is not available', tb
=False, is_error
=False)
2640 if not formats_to_download
:
2641 if not self
.params
.get('ignore_no_formats_error'):
2642 raise ExtractorError(
2643 'Requested format is not available. Use --list-formats for a list of available formats',
2644 expected
=True, video_id
=info_dict
['id'], ie
=info_dict
['extractor'])
2645 self
.report_warning('Requested format is not available')
2646 # Process what we can, even without any available formats.
2647 formats_to_download
= [{}]
2649 requested_ranges
= self
.params
.get('download_ranges')
2650 if requested_ranges
:
2651 requested_ranges
= tuple(requested_ranges(info_dict
, self
))
2653 best_format
, downloaded_formats
= formats_to_download
[-1], []
2656 def to_screen(*msg
):
2657 self
.to_screen(f
'[info] {info_dict["id"]}: {" ".join(", ".join(variadic(m)) for m in msg)}')
2659 to_screen(f
'Downloading {len(formats_to_download)} format(s):',
2660 (f
['format_id'] for f
in formats_to_download
))
2661 if requested_ranges
:
2662 to_screen(f
'Downloading {len(requested_ranges)} time ranges:',
2663 (f
'{int(c["start_time"])}-{int(c["end_time"])}' for c
in requested_ranges
))
2664 max_downloads_reached
= False
2666 for fmt
, chapter
in itertools
.product(formats_to_download
, requested_ranges
or [{}]):
2667 new_info
= self
._copy
_infodict
(info_dict
)
2668 new_info
.update(fmt
)
2669 offset
, duration
= info_dict
.get('section_start') or 0, info_dict
.get('duration') or float('inf')
2670 if chapter
or offset
:
2672 'section_start': offset
+ chapter
.get('start_time', 0),
2673 'section_end': offset
+ min(chapter
.get('end_time', duration
), duration
),
2674 'section_title': chapter
.get('title'),
2675 'section_number': chapter
.get('index'),
2677 downloaded_formats
.append(new_info
)
2679 self
.process_info(new_info
)
2680 except MaxDownloadsReached
:
2681 max_downloads_reached
= True
2682 self
._raise
_pending
_errors
(new_info
)
2683 # Remove copied info
2684 for key
, val
in tuple(new_info
.items()):
2685 if info_dict
.get(key
) == val
:
2687 if max_downloads_reached
:
2690 write_archive
= {f.get('__write_download_archive', False) for f in downloaded_formats}
2691 assert write_archive
.issubset({True, False, 'ignore'}
)
2692 if True in write_archive
and False not in write_archive
:
2693 self
.record_download_archive(info_dict
)
2695 info_dict
['requested_downloads'] = downloaded_formats
2696 info_dict
= self
.run_all_pps('after_video', info_dict
)
2697 if max_downloads_reached
:
2698 raise MaxDownloadsReached()
2700 # We update the info dict with the selected best quality format (backwards compatibility)
2701 info_dict
.update(best_format
)
2704 def process_subtitles(self
, video_id
, normal_subtitles
, automatic_captions
):
2705 """Select the requested subtitles and their format"""
2706 available_subs
, normal_sub_langs
= {}, []
2707 if normal_subtitles
and self
.params
.get('writesubtitles'):
2708 available_subs
.update(normal_subtitles
)
2709 normal_sub_langs
= tuple(normal_subtitles
.keys())
2710 if automatic_captions
and self
.params
.get('writeautomaticsub'):
2711 for lang
, cap_info
in automatic_captions
.items():
2712 if lang
not in available_subs
:
2713 available_subs
[lang
] = cap_info
2715 if (not self
.params
.get('writesubtitles') and not
2716 self
.params
.get('writeautomaticsub') or not
2720 all_sub_langs
= tuple(available_subs
.keys())
2721 if self
.params
.get('allsubtitles', False):
2722 requested_langs
= all_sub_langs
2723 elif self
.params
.get('subtitleslangs', False):
2724 # A list is used so that the order of languages will be the same as
2725 # given in subtitleslangs. See https://github.com/yt-dlp/yt-dlp/issues/1041
2726 requested_langs
= []
2727 for lang_re
in self
.params
.get('subtitleslangs'):
2728 discard
= lang_re
[0] == '-'
2730 lang_re
= lang_re
[1:]
2731 if lang_re
== 'all':
2733 requested_langs
= []
2735 requested_langs
.extend(all_sub_langs
)
2737 current_langs
= filter(re
.compile(lang_re
+ '$').match
, all_sub_langs
)
2739 for lang
in current_langs
:
2740 while lang
in requested_langs
:
2741 requested_langs
.remove(lang
)
2743 requested_langs
.extend(current_langs
)
2744 requested_langs
= orderedSet(requested_langs
)
2745 elif normal_sub_langs
:
2746 requested_langs
= ['en'] if 'en' in normal_sub_langs
else normal_sub_langs
[:1]
2748 requested_langs
= ['en'] if 'en' in all_sub_langs
else all_sub_langs
[:1]
2750 self
.write_debug('Downloading subtitles: %s' % ', '.join(requested_langs
))
2752 formats_query
= self
.params
.get('subtitlesformat', 'best')
2753 formats_preference
= formats_query
.split('/') if formats_query
else []
2755 for lang
in requested_langs
:
2756 formats
= available_subs
.get(lang
)
2758 self
.report_warning(f
'{lang} subtitles not available for {video_id}')
2760 for ext
in formats_preference
:
2764 matches
= list(filter(lambda f
: f
['ext'] == ext
, formats
))
2770 self
.report_warning(
2771 'No subtitle format found matching "%s" for language %s, '
2772 'using %s' % (formats_query
, lang
, f
['ext']))
2776 def _forceprint(self
, key
, info_dict
):
2777 if info_dict
is None:
2779 info_copy
= info_dict
.copy()
2780 info_copy
['formats_table'] = self
.render_formats_table(info_dict
)
2781 info_copy
['thumbnails_table'] = self
.render_thumbnails_table(info_dict
)
2782 info_copy
['subtitles_table'] = self
.render_subtitles_table(info_dict
.get('id'), info_dict
.get('subtitles'))
2783 info_copy
['automatic_captions_table'] = self
.render_subtitles_table(info_dict
.get('id'), info_dict
.get('automatic_captions'))
2785 def format_tmpl(tmpl
):
2786 mobj
= re
.match(r
'\w+(=?)$', tmpl
)
2787 if mobj
and mobj
.group(1):
2788 return f
'{tmpl[:-1]} = %({tmpl[:-1]})r'
2790 return f
'%({tmpl})s'
2793 for tmpl
in self
.params
['forceprint'].get(key
, []):
2794 self
.to_stdout(self
.evaluate_outtmpl(format_tmpl(tmpl
), info_copy
))
2796 for tmpl
, file_tmpl
in self
.params
['print_to_file'].get(key
, []):
2797 filename
= self
.prepare_filename(info_dict
, outtmpl
=file_tmpl
)
2798 tmpl
= format_tmpl(tmpl
)
2799 self
.to_screen(f
'[info] Writing {tmpl!r} to: {filename}')
2800 if self
._ensure
_dir
_exists
(filename
):
2801 with open(filename
, 'a', encoding
='utf-8') as f
:
2802 f
.write(self
.evaluate_outtmpl(tmpl
, info_copy
) + '\n')
2804 def __forced_printings(self
, info_dict
, filename
, incomplete
):
2805 def print_mandatory(field
, actual_field
=None):
2806 if actual_field
is None:
2807 actual_field
= field
2808 if (self
.params
.get('force%s' % field
, False)
2809 and (not incomplete
or info_dict
.get(actual_field
) is not None)):
2810 self
.to_stdout(info_dict
[actual_field
])
2812 def print_optional(field
):
2813 if (self
.params
.get('force%s' % field
, False)
2814 and info_dict
.get(field
) is not None):
2815 self
.to_stdout(info_dict
[field
])
2817 info_dict
= info_dict
.copy()
2818 if filename
is not None:
2819 info_dict
['filename'] = filename
2820 if info_dict
.get('requested_formats') is not None:
2821 # For RTMP URLs, also include the playpath
2822 info_dict
['urls'] = '\n'.join(f
['url'] + f
.get('play_path', '') for f
in info_dict
['requested_formats'])
2823 elif info_dict
.get('url'):
2824 info_dict
['urls'] = info_dict
['url'] + info_dict
.get('play_path', '')
2826 if (self
.params
.get('forcejson')
2827 or self
.params
['forceprint'].get('video')
2828 or self
.params
['print_to_file'].get('video')):
2829 self
.post_extract(info_dict
)
2830 self
._forceprint
('video', info_dict
)
2832 print_mandatory('title')
2833 print_mandatory('id')
2834 print_mandatory('url', 'urls')
2835 print_optional('thumbnail')
2836 print_optional('description')
2837 print_optional('filename')
2838 if self
.params
.get('forceduration') and info_dict
.get('duration') is not None:
2839 self
.to_stdout(formatSeconds(info_dict
['duration']))
2840 print_mandatory('format')
2842 if self
.params
.get('forcejson'):
2843 self
.to_stdout(json
.dumps(self
.sanitize_info(info_dict
)))
2845 def dl(self
, name
, info
, subtitle
=False, test
=False):
2846 if not info
.get('url'):
2847 self
.raise_no_formats(info
, True)
2850 verbose
= self
.params
.get('verbose')
2853 'quiet': self
.params
.get('quiet') or not verbose
,
2855 'noprogress': not verbose
,
2857 'skip_unavailable_fragments': False,
2858 'keep_fragments': False,
2860 '_no_ytdl_file': True,
2863 params
= self
.params
2864 fd
= get_suitable_downloader(info
, params
, to_stdout
=(name
== '-'))(self
, params
)
2866 for ph
in self
._progress
_hooks
:
2867 fd
.add_progress_hook(ph
)
2869 (f
['url'].split(',')[0] + ',<data>' if f
['url'].startswith('data:') else f
['url'])
2870 for f
in info
.get('requested_formats', []) or [info
])
2871 self
.write_debug(f
'Invoking {fd.FD_NAME} downloader on "{urls}"')
2873 # Note: Ideally info should be a deep-copied so that hooks cannot modify it.
2874 # But it may contain objects that are not deep-copyable
2875 new_info
= self
._copy
_infodict
(info
)
2876 if new_info
.get('http_headers') is None:
2877 new_info
['http_headers'] = self
._calc
_headers
(new_info
)
2878 return fd
.download(name
, new_info
, subtitle
)
2880 def existing_file(self
, filepaths
, *, default_overwrite
=True):
2881 existing_files
= list(filter(os
.path
.exists
, orderedSet(filepaths
)))
2882 if existing_files
and not self
.params
.get('overwrites', default_overwrite
):
2883 return existing_files
[0]
2885 for file in existing_files
:
2886 self
.report_file_delete(file)
2890 def process_info(self
, info_dict
):
2891 """Process a single resolved IE result. (Modifies it in-place)"""
2893 assert info_dict
.get('_type', 'video') == 'video'
2894 original_infodict
= info_dict
2896 if 'format' not in info_dict
and 'ext' in info_dict
:
2897 info_dict
['format'] = info_dict
['ext']
2899 # This is mostly just for backward compatibility of process_info
2900 # As a side-effect, this allows for format-specific filters
2901 if self
._match
_entry
(info_dict
) is not None:
2902 info_dict
['__write_download_archive'] = 'ignore'
2905 # Does nothing under normal operation - for backward compatibility of process_info
2906 self
.post_extract(info_dict
)
2907 self
._num
_downloads
+= 1
2909 # info_dict['_filename'] needs to be set for backward compatibility
2910 info_dict
['_filename'] = full_filename
= self
.prepare_filename(info_dict
, warn
=True)
2911 temp_filename
= self
.prepare_filename(info_dict
, 'temp')
2915 self
.__forced
_printings
(info_dict
, full_filename
, incomplete
=('format' not in info_dict
))
2917 def check_max_downloads():
2918 if self
._num
_downloads
>= float(self
.params
.get('max_downloads') or 'inf'):
2919 raise MaxDownloadsReached()
2921 if self
.params
.get('simulate'):
2922 info_dict
['__write_download_archive'] = self
.params
.get('force_write_download_archive')
2923 check_max_downloads()
2926 if full_filename
is None:
2928 if not self
._ensure
_dir
_exists
(encodeFilename(full_filename
)):
2930 if not self
._ensure
_dir
_exists
(encodeFilename(temp_filename
)):
2933 if self
._write
_description
('video', info_dict
,
2934 self
.prepare_filename(info_dict
, 'description')) is None:
2937 sub_files
= self
._write
_subtitles
(info_dict
, temp_filename
)
2938 if sub_files
is None:
2940 files_to_move
.update(dict(sub_files
))
2942 thumb_files
= self
._write
_thumbnails
(
2943 'video', info_dict
, temp_filename
, self
.prepare_filename(info_dict
, 'thumbnail'))
2944 if thumb_files
is None:
2946 files_to_move
.update(dict(thumb_files
))
2948 infofn
= self
.prepare_filename(info_dict
, 'infojson')
2949 _infojson_written
= self
._write
_info
_json
('video', info_dict
, infofn
)
2950 if _infojson_written
:
2951 info_dict
['infojson_filename'] = infofn
2952 # For backward compatibility, even though it was a private field
2953 info_dict
['__infojson_filename'] = infofn
2954 elif _infojson_written
is None:
2957 # Note: Annotations are deprecated
2959 if self
.params
.get('writeannotations', False):
2960 annofn
= self
.prepare_filename(info_dict
, 'annotation')
2962 if not self
._ensure
_dir
_exists
(encodeFilename(annofn
)):
2964 if not self
.params
.get('overwrites', True) and os
.path
.exists(encodeFilename(annofn
)):
2965 self
.to_screen('[info] Video annotations are already present')
2966 elif not info_dict
.get('annotations'):
2967 self
.report_warning('There are no annotations to write.')
2970 self
.to_screen('[info] Writing video annotations to: ' + annofn
)
2971 with open(encodeFilename(annofn
), 'w', encoding
='utf-8') as annofile
:
2972 annofile
.write(info_dict
['annotations'])
2973 except (KeyError, TypeError):
2974 self
.report_warning('There are no annotations to write.')
2976 self
.report_error('Cannot write annotations file: ' + annofn
)
2979 # Write internet shortcut files
2980 def _write_link_file(link_type
):
2981 url
= try_get(info_dict
['webpage_url'], iri_to_uri
)
2983 self
.report_warning(
2984 f
'Cannot write internet shortcut file because the actual URL of "{info_dict["webpage_url"]}" is unknown')
2986 linkfn
= replace_extension(self
.prepare_filename(info_dict
, 'link'), link_type
, info_dict
.get('ext'))
2987 if not self
._ensure
_dir
_exists
(encodeFilename(linkfn
)):
2989 if self
.params
.get('overwrites', True) and os
.path
.exists(encodeFilename(linkfn
)):
2990 self
.to_screen(f
'[info] Internet shortcut (.{link_type}) is already present')
2993 self
.to_screen(f
'[info] Writing internet shortcut (.{link_type}) to: {linkfn}')
2994 with open(encodeFilename(to_high_limit_path(linkfn
)), 'w', encoding
='utf-8',
2995 newline
='\r\n' if link_type
== 'url' else '\n') as linkfile
:
2996 template_vars
= {'url': url}
2997 if link_type
== 'desktop':
2998 template_vars
['filename'] = linkfn
[:-(len(link_type
) + 1)]
2999 linkfile
.write(LINK_TEMPLATES
[link_type
] % template_vars
)
3001 self
.report_error(f
'Cannot write internet shortcut {linkfn}')
3006 'url': self
.params
.get('writeurllink'),
3007 'webloc': self
.params
.get('writewebloclink'),
3008 'desktop': self
.params
.get('writedesktoplink'),
3010 if self
.params
.get('writelink'):
3011 link_type
= ('webloc' if sys
.platform
== 'darwin'
3012 else 'desktop' if sys
.platform
.startswith('linux')
3014 write_links
[link_type
] = True
3016 if any(should_write
and not _write_link_file(link_type
)
3017 for link_type
, should_write
in write_links
.items()):
3020 def replace_info_dict(new_info
):
3022 if new_info
== info_dict
:
3025 info_dict
.update(new_info
)
3027 new_info
, files_to_move
= self
.pre_process(info_dict
, 'before_dl', files_to_move
)
3028 replace_info_dict(new_info
)
3030 if self
.params
.get('skip_download'):
3031 info_dict
['filepath'] = temp_filename
3032 info_dict
['__finaldir'] = os
.path
.dirname(os
.path
.abspath(encodeFilename(full_filename
)))
3033 info_dict
['__files_to_move'] = files_to_move
3034 replace_info_dict(self
.run_pp(MoveFilesAfterDownloadPP(self
, False), info_dict
))
3035 info_dict
['__write_download_archive'] = self
.params
.get('force_write_download_archive')
3038 info_dict
.setdefault('__postprocessors', [])
3041 def existing_video_file(*filepaths
):
3042 ext
= info_dict
.get('ext')
3043 converted
= lambda file: replace_extension(file, self
.params
.get('final_ext') or ext
, ext
)
3044 file = self
.existing_file(itertools
.chain(*zip(map(converted
, filepaths
), filepaths
)),
3045 default_overwrite
=False)
3047 info_dict
['ext'] = os
.path
.splitext(file)[1][1:]
3050 fd
, success
= None, True
3051 if info_dict
.get('protocol') or info_dict
.get('url'):
3052 fd
= get_suitable_downloader(info_dict
, self
.params
, to_stdout
=temp_filename
== '-')
3053 if fd
is not FFmpegFD
and (
3054 info_dict
.get('section_start') or info_dict
.get('section_end')):
3055 msg
= ('This format cannot be partially downloaded' if FFmpegFD
.available()
3056 else 'You have requested downloading the video partially, but ffmpeg is not installed')
3057 self
.report_error(f
'{msg}. Aborting')
3060 if info_dict
.get('requested_formats') is not None:
3062 def compatible_formats(formats
):
3063 # TODO: some formats actually allow this (mkv, webm, ogg, mp4), but not all of them.
3064 video_formats
= [format
for format
in formats
if format
.get('vcodec') != 'none']
3065 audio_formats
= [format
for format
in formats
if format
.get('acodec') != 'none']
3066 if len(video_formats
) > 2 or len(audio_formats
) > 2:
3070 exts
= {format.get('ext') for format in formats}
3072 {'mp3', 'mp4', 'm4a', 'm4p', 'm4b', 'm4r', 'm4v', 'ismv', 'isma'}
,
3075 for ext_sets
in COMPATIBLE_EXTS
:
3076 if ext_sets
.issuperset(exts
):
3078 # TODO: Check acodec/vcodec
3081 requested_formats
= info_dict
['requested_formats']
3082 old_ext
= info_dict
['ext']
3083 if self
.params
.get('merge_output_format') is None:
3084 if not compatible_formats(requested_formats
):
3085 info_dict
['ext'] = 'mkv'
3086 self
.report_warning(
3087 'Requested formats are incompatible for merge and will be merged into mkv')
3088 if (info_dict
['ext'] == 'webm'
3089 and info_dict
.get('thumbnails')
3090 # check with type instead of pp_key, __name__, or isinstance
3091 # since we dont want any custom PPs to trigger this
3092 and any(type(pp
) == EmbedThumbnailPP
for pp
in self
._pps
['post_process'])): # noqa: E721
3093 info_dict
['ext'] = 'mkv'
3094 self
.report_warning(
3095 'webm doesn\'t support embedding a thumbnail, mkv will be used')
3096 new_ext
= info_dict
['ext']
3098 def correct_ext(filename
, ext
=new_ext
):
3101 filename_real_ext
= os
.path
.splitext(filename
)[1][1:]
3103 os
.path
.splitext(filename
)[0]
3104 if filename_real_ext
in (old_ext
, new_ext
)
3106 return f
'{filename_wo_ext}.{ext}'
3108 # Ensure filename always has a correct extension for successful merge
3109 full_filename
= correct_ext(full_filename
)
3110 temp_filename
= correct_ext(temp_filename
)
3111 dl_filename
= existing_video_file(full_filename
, temp_filename
)
3112 info_dict
['__real_download'] = False
3114 merger
= FFmpegMergerPP(self
)
3116 if dl_filename
is not None:
3117 self
.report_file_already_downloaded(dl_filename
)
3119 for f
in requested_formats
if fd
!= FFmpegFD
else []:
3120 f
['filepath'] = fname
= prepend_extension(
3121 correct_ext(temp_filename
, info_dict
['ext']),
3122 'f%s' % f
['format_id'], info_dict
['ext'])
3123 downloaded
.append(fname
)
3124 info_dict
['url'] = '\n'.join(f
['url'] for f
in requested_formats
)
3125 success
, real_download
= self
.dl(temp_filename
, info_dict
)
3126 info_dict
['__real_download'] = real_download
3128 if self
.params
.get('allow_unplayable_formats'):
3129 self
.report_warning(
3130 'You have requested merging of multiple formats '
3131 'while also allowing unplayable formats to be downloaded. '
3132 'The formats won\'t be merged to prevent data corruption.')
3133 elif not merger
.available
:
3134 msg
= 'You have requested merging of multiple formats but ffmpeg is not installed'
3135 if not self
.params
.get('ignoreerrors'):
3136 self
.report_error(f
'{msg}. Aborting due to --abort-on-error')
3138 self
.report_warning(f
'{msg}. The formats won\'t be merged')
3140 if temp_filename
== '-':
3141 reason
= ('using a downloader other than ffmpeg' if FFmpegFD
.can_merge_formats(info_dict
, self
.params
)
3142 else 'but the formats are incompatible for simultaneous download' if merger
.available
3143 else 'but ffmpeg is not installed')
3144 self
.report_warning(
3145 f
'You have requested downloading multiple formats to stdout {reason}. '
3146 'The formats will be streamed one after the other')
3147 fname
= temp_filename
3148 for f
in requested_formats
:
3149 new_info
= dict(info_dict
)
3150 del new_info
['requested_formats']
3152 if temp_filename
!= '-':
3153 fname
= prepend_extension(
3154 correct_ext(temp_filename
, new_info
['ext']),
3155 'f%s' % f
['format_id'], new_info
['ext'])
3156 if not self
._ensure
_dir
_exists
(fname
):
3158 f
['filepath'] = fname
3159 downloaded
.append(fname
)
3160 partial_success
, real_download
= self
.dl(fname
, new_info
)
3161 info_dict
['__real_download'] = info_dict
['__real_download'] or real_download
3162 success
= success
and partial_success
3164 if downloaded
and merger
.available
and not self
.params
.get('allow_unplayable_formats'):
3165 info_dict
['__postprocessors'].append(merger
)
3166 info_dict
['__files_to_merge'] = downloaded
3167 # Even if there were no downloads, it is being merged only now
3168 info_dict
['__real_download'] = True
3170 for file in downloaded
:
3171 files_to_move
[file] = None
3173 # Just a single file
3174 dl_filename
= existing_video_file(full_filename
, temp_filename
)
3175 if dl_filename
is None or dl_filename
== temp_filename
:
3176 # dl_filename == temp_filename could mean that the file was partially downloaded with --no-part.
3177 # So we should try to resume the download
3178 success
, real_download
= self
.dl(temp_filename
, info_dict
)
3179 info_dict
['__real_download'] = real_download
3181 self
.report_file_already_downloaded(dl_filename
)
3183 dl_filename
= dl_filename
or temp_filename
3184 info_dict
['__finaldir'] = os
.path
.dirname(os
.path
.abspath(encodeFilename(full_filename
)))
3186 except network_exceptions
as err
:
3187 self
.report_error('unable to download video data: %s' % error_to_compat_str(err
))
3189 except OSError as err
:
3190 raise UnavailableVideoError(err
)
3191 except (ContentTooShortError
, ) as err
:
3192 self
.report_error(f
'content too short (expected {err.expected} bytes and served {err.downloaded})')
3195 self
._raise
_pending
_errors
(info_dict
)
3196 if success
and full_filename
!= '-':
3200 fixup_policy
= self
.params
.get('fixup')
3201 vid
= info_dict
['id']
3203 if fixup_policy
in ('ignore', 'never'):
3205 elif fixup_policy
== 'warn':
3207 elif fixup_policy
!= 'force':
3208 assert fixup_policy
in ('detect_or_warn', None)
3209 if not info_dict
.get('__real_download'):
3212 def ffmpeg_fixup(cndn
, msg
, cls
):
3213 if not (do_fixup
and cndn
):
3215 elif do_fixup
== 'warn':
3216 self
.report_warning(f
'{vid}: {msg}')
3220 info_dict
['__postprocessors'].append(pp
)
3222 self
.report_warning(f
'{vid}: {msg}. Install ffmpeg to fix this automatically')
3224 stretched_ratio
= info_dict
.get('stretched_ratio')
3225 ffmpeg_fixup(stretched_ratio
not in (1, None),
3226 f
'Non-uniform pixel ratio {stretched_ratio}',
3227 FFmpegFixupStretchedPP
)
3229 downloader
= get_suitable_downloader(info_dict
, self
.params
) if 'protocol' in info_dict
else None
3230 downloader
= downloader
.FD_NAME
if downloader
else None
3232 ext
= info_dict
.get('ext')
3233 postprocessed_by_ffmpeg
= info_dict
.get('requested_formats') or any((
3234 isinstance(pp
, FFmpegVideoConvertorPP
)
3235 and resolve_recode_mapping(ext
, pp
.mapping
)[0] not in (ext
, None)
3236 ) for pp
in self
._pps
['post_process'])
3238 if not postprocessed_by_ffmpeg
:
3239 ffmpeg_fixup(ext
== 'm4a' and info_dict
.get('container') == 'm4a_dash',
3240 'writing DASH m4a. Only some players support this container',
3242 ffmpeg_fixup(downloader
== 'hlsnative' and not self
.params
.get('hls_use_mpegts')
3243 or info_dict
.get('is_live') and self
.params
.get('hls_use_mpegts') is None,
3244 'Possible MPEG-TS in MP4 container or malformed AAC timestamps',
3246 ffmpeg_fixup(info_dict
.get('is_live') and downloader
== 'DashSegmentsFD',
3247 'Possible duplicate MOOV atoms', FFmpegFixupDuplicateMoovPP
)
3249 ffmpeg_fixup(downloader
== 'web_socket_fragment', 'Malformed timestamps detected', FFmpegFixupTimestampPP
)
3250 ffmpeg_fixup(downloader
== 'web_socket_fragment', 'Malformed duration detected', FFmpegFixupDurationPP
)
3254 replace_info_dict(self
.post_process(dl_filename
, info_dict
, files_to_move
))
3255 except PostProcessingError
as err
:
3256 self
.report_error('Postprocessing: %s' % str(err
))
3259 for ph
in self
._post
_hooks
:
3260 ph(info_dict
['filepath'])
3261 except Exception as err
:
3262 self
.report_error('post hooks: %s' % str(err
))
3264 info_dict
['__write_download_archive'] = True
3266 assert info_dict
is original_infodict
# Make sure the info_dict was modified in-place
3267 if self
.params
.get('force_write_download_archive'):
3268 info_dict
['__write_download_archive'] = True
3269 check_max_downloads()
3271 def __download_wrapper(self
, func
):
3272 @functools.wraps(func
)
3273 def wrapper(*args
, **kwargs
):
3275 res
= func(*args
, **kwargs
)
3276 except UnavailableVideoError
as e
:
3277 self
.report_error(e
)
3278 except DownloadCancelled
as e
:
3279 self
.to_screen(f
'[info] {e}')
3280 if not self
.params
.get('break_per_url'):
3283 if self
.params
.get('dump_single_json', False):
3284 self
.post_extract(res
)
3285 self
.to_stdout(json
.dumps(self
.sanitize_info(res
)))
3288 def download(self
, url_list
):
3289 """Download a given list of URLs."""
3290 url_list
= variadic(url_list
) # Passing a single URL is a common mistake
3291 outtmpl
= self
.params
['outtmpl']['default']
3292 if (len(url_list
) > 1
3294 and '%' not in outtmpl
3295 and self
.params
.get('max_downloads') != 1):
3296 raise SameFileError(outtmpl
)
3298 for url
in url_list
:
3299 self
.__download
_wrapper
(self
.extract_info
)(
3300 url
, force_generic_extractor
=self
.params
.get('force_generic_extractor', False))
3302 return self
._download
_retcode
3304 def download_with_info_file(self
, info_filename
):
3305 with contextlib
.closing(fileinput
.FileInput(
3306 [info_filename
], mode
='r',
3307 openhook
=fileinput
.hook_encoded('utf-8'))) as f
:
3308 # FileInput doesn't have a read method, we can't call json.load
3309 info
= self
.sanitize_info(json
.loads('\n'.join(f
)), self
.params
.get('clean_infojson', True))
3311 self
.__download
_wrapper
(self
.process_ie_result
)(info
, download
=True)
3312 except (DownloadError
, EntryNotInPlaylist
, ReExtractInfo
) as e
:
3313 if not isinstance(e
, EntryNotInPlaylist
):
3314 self
.to_stderr('\r')
3315 webpage_url
= info
.get('webpage_url')
3316 if webpage_url
is not None:
3317 self
.report_warning(f
'The info failed to download: {e}; trying with URL {webpage_url}')
3318 return self
.download([webpage_url
])
3321 return self
._download
_retcode
3324 def sanitize_info(info_dict
, remove_private_keys
=False):
3325 ''' Sanitize the infodict for converting to json '''
3326 if info_dict
is None:
3328 info_dict
.setdefault('epoch', int(time
.time()))
3329 info_dict
.setdefault('_type', 'video')
3331 if remove_private_keys
:
3332 reject
= lambda k
, v
: v
is None or k
.startswith('__') or k
in {
3333 'requested_downloads', 'requested_formats', 'requested_subtitles', 'requested_entries',
3334 'entries', 'filepath', '_filename', 'infojson_filename', 'original_url', 'playlist_autonumber',
3337 reject
= lambda k
, v
: False
3340 if isinstance(obj
, dict):
3341 return {k: filter_fn(v) for k, v in obj.items() if not reject(k, v)}
3342 elif isinstance(obj
, (list, tuple, set, LazyList
)):
3343 return list(map(filter_fn
, obj
))
3344 elif obj
is None or isinstance(obj
, (str, int, float, bool)):
3349 return filter_fn(info_dict
)
3352 def filter_requested_info(info_dict
, actually_filter
=True):
3353 ''' Alias of sanitize_info for backward compatibility '''
3354 return YoutubeDL
.sanitize_info(info_dict
, actually_filter
)
3356 def _delete_downloaded_files(self
, *files_to_delete
, info
={}, msg
=None):
3357 for filename
in set(filter(None, files_to_delete
)):
3359 self
.to_screen(msg
% filename
)
3363 self
.report_warning(f
'Unable to delete file {filename}')
3364 if filename
in info
.get('__files_to_move', []): # NB: Delete even if None
3365 del info
['__files_to_move'][filename
]
3368 def post_extract(info_dict
):
3369 def actual_post_extract(info_dict
):
3370 if info_dict
.get('_type') in ('playlist', 'multi_video'):
3371 for video_dict
in info_dict
.get('entries', {}):
3372 actual_post_extract(video_dict
or {})
3375 post_extractor
= info_dict
.pop('__post_extractor', None) or (lambda: {})
3376 info_dict
.update(post_extractor())
3378 actual_post_extract(info_dict
or {})
3380 def run_pp(self
, pp
, infodict
):
3381 files_to_delete
= []
3382 if '__files_to_move' not in infodict
:
3383 infodict
['__files_to_move'] = {}
3385 files_to_delete
, infodict
= pp
.run(infodict
)
3386 except PostProcessingError
as e
:
3387 # Must be True and not 'only_download'
3388 if self
.params
.get('ignoreerrors') is True:
3389 self
.report_error(e
)
3393 if not files_to_delete
:
3395 if self
.params
.get('keepvideo', False):
3396 for f
in files_to_delete
:
3397 infodict
['__files_to_move'].setdefault(f
, '')
3399 self
._delete
_downloaded
_files
(
3400 *files_to_delete
, info
=infodict
, msg
='Deleting original file %s (pass -k to keep)')
3403 def run_all_pps(self
, key
, info
, *, additional_pps
=None):
3404 self
._forceprint
(key
, info
)
3405 for pp
in (additional_pps
or []) + self
._pps
[key
]:
3406 info
= self
.run_pp(pp
, info
)
3409 def pre_process(self
, ie_info
, key
='pre_process', files_to_move
=None):
3410 info
= dict(ie_info
)
3411 info
['__files_to_move'] = files_to_move
or {}
3413 info
= self
.run_all_pps(key
, info
)
3414 except PostProcessingError
as err
:
3415 msg
= f
'Preprocessing: {err}'
3416 info
.setdefault('__pending_error', msg
)
3417 self
.report_error(msg
, is_error
=False)
3418 return info
, info
.pop('__files_to_move', None)
3420 def post_process(self
, filename
, info
, files_to_move
=None):
3421 """Run all the postprocessors on the given file."""
3422 info
['filepath'] = filename
3423 info
['__files_to_move'] = files_to_move
or {}
3424 info
= self
.run_all_pps('post_process', info
, additional_pps
=info
.get('__postprocessors'))
3425 info
= self
.run_pp(MoveFilesAfterDownloadPP(self
), info
)
3426 del info
['__files_to_move']
3427 return self
.run_all_pps('after_move', info
)
3429 def _make_archive_id(self
, info_dict
):
3430 video_id
= info_dict
.get('id')
3433 # Future-proof against any change in case
3434 # and backwards compatibility with prior versions
3435 extractor
= info_dict
.get('extractor_key') or info_dict
.get('ie_key') # key in a playlist
3436 if extractor
is None:
3437 url
= str_or_none(info_dict
.get('url'))
3440 # Try to find matching extractor for the URL and take its ie_key
3441 for ie_key
, ie
in self
._ies
.items():
3442 if ie
.suitable(url
):
3447 return f
'{extractor.lower()} {video_id}'
3449 def in_download_archive(self
, info_dict
):
3450 fn
= self
.params
.get('download_archive')
3454 vid_id
= self
._make
_archive
_id
(info_dict
)
3456 return False # Incomplete video information
3458 return vid_id
in self
.archive
3460 def record_download_archive(self
, info_dict
):
3461 fn
= self
.params
.get('download_archive')
3464 vid_id
= self
._make
_archive
_id
(info_dict
)
3466 self
.write_debug(f
'Adding to archive: {vid_id}')
3467 with locked_file(fn
, 'a', encoding
='utf-8') as archive_file
:
3468 archive_file
.write(vid_id
+ '\n')
3469 self
.archive
.add(vid_id
)
3472 def format_resolution(format
, default
='unknown'):
3473 if format
.get('vcodec') == 'none' and format
.get('acodec') != 'none':
3475 if format
.get('resolution') is not None:
3476 return format
['resolution']
3477 if format
.get('width') and format
.get('height'):
3478 return '%dx%d' % (format
['width'], format
['height'])
3479 elif format
.get('height'):
3480 return '%sp' % format
['height']
3481 elif format
.get('width'):
3482 return '%dx?' % format
['width']
3485 def _list_format_headers(self
, *headers
):
3486 if self
.params
.get('listformats_table', True) is not False:
3487 return [self
._format
_out
(header
, self
.Styles
.HEADERS
) for header
in headers
]
3490 def _format_note(self
, fdict
):
3492 if fdict
.get('ext') in ['f4f', 'f4m']:
3493 res
+= '(unsupported)'
3494 if fdict
.get('language'):
3497 res
+= '[%s]' % fdict
['language']
3498 if fdict
.get('format_note') is not None:
3501 res
+= fdict
['format_note']
3502 if fdict
.get('tbr') is not None:
3505 res
+= '%4dk' % fdict
['tbr']
3506 if fdict
.get('container') is not None:
3509 res
+= '%s container' % fdict
['container']
3510 if (fdict
.get('vcodec') is not None
3511 and fdict
.get('vcodec') != 'none'):
3514 res
+= fdict
['vcodec']
3515 if fdict
.get('vbr') is not None:
3517 elif fdict
.get('vbr') is not None and fdict
.get('abr') is not None:
3519 if fdict
.get('vbr') is not None:
3520 res
+= '%4dk' % fdict
['vbr']
3521 if fdict
.get('fps') is not None:
3524 res
+= '%sfps' % fdict
['fps']
3525 if fdict
.get('acodec') is not None:
3528 if fdict
['acodec'] == 'none':
3531 res
+= '%-5s' % fdict
['acodec']
3532 elif fdict
.get('abr') is not None:
3536 if fdict
.get('abr') is not None:
3537 res
+= '@%3dk' % fdict
['abr']
3538 if fdict
.get('asr') is not None:
3539 res
+= ' (%5dHz)' % fdict
['asr']
3540 if fdict
.get('filesize') is not None:
3543 res
+= format_bytes(fdict
['filesize'])
3544 elif fdict
.get('filesize_approx') is not None:
3547 res
+= '~' + format_bytes(fdict
['filesize_approx'])
3550 def render_formats_table(self
, info_dict
):
3551 if not info_dict
.get('formats') and not info_dict
.get('url'):
3554 formats
= info_dict
.get('formats', [info_dict
])
3555 if not self
.params
.get('listformats_table', True) is not False:
3558 format_field(f
, 'format_id'),
3559 format_field(f
, 'ext'),
3560 self
.format_resolution(f
),
3561 self
._format
_note
(f
)
3562 ] for f
in formats
if f
.get('preference') is None or f
['preference'] >= -1000]
3563 return render_table(['format code', 'extension', 'resolution', 'note'], table
, extra_gap
=1)
3565 def simplified_codec(f
, field
):
3566 assert field
in ('acodec', 'vcodec')
3567 codec
= f
.get(field
, 'unknown')
3570 elif codec
!= 'none':
3571 return '.'.join(codec
.split('.')[:4])
3573 if field
== 'vcodec' and f
.get('acodec') == 'none':
3575 elif field
== 'acodec' and f
.get('vcodec') == 'none':
3577 return self
._format
_out
('audio only' if field
== 'vcodec' else 'video only',
3578 self
.Styles
.SUPPRESS
)
3580 delim
= self
._format
_out
('\u2502', self
.Styles
.DELIM
, '|', test_encoding
=True)
3583 self
._format
_out
(format_field(f
, 'format_id'), self
.Styles
.ID
),
3584 format_field(f
, 'ext'),
3585 format_field(f
, func
=self
.format_resolution
, ignore
=('audio only', 'images')),
3586 format_field(f
, 'fps', '\t%d', func
=round),
3587 format_field(f
, 'dynamic_range', '%s', ignore
=(None, 'SDR')).replace('HDR', ''),
3589 format_field(f
, 'filesize', ' \t%s', func
=format_bytes
) + format_field(f
, 'filesize_approx', '~\t%s', func
=format_bytes
),
3590 format_field(f
, 'tbr', '\t%dk', func
=round),
3591 shorten_protocol_name(f
.get('protocol', '')),
3593 simplified_codec(f
, 'vcodec'),
3594 format_field(f
, 'vbr', '\t%dk', func
=round),
3595 simplified_codec(f
, 'acodec'),
3596 format_field(f
, 'abr', '\t%dk', func
=round),
3597 format_field(f
, 'asr', '\t%s', func
=format_decimal_suffix
),
3599 self
._format
_out
('UNSUPPORTED', 'light red') if f
.get('ext') in ('f4f', 'f4m') else None,
3600 format_field(f
, 'language', '[%s]'),
3601 join_nonempty(format_field(f
, 'format_note'),
3602 format_field(f
, 'container', ignore
=(None, f
.get('ext'))),
3605 ] for f
in formats
if f
.get('preference') is None or f
['preference'] >= -1000]
3606 header_line
= self
._list
_format
_headers
(
3607 'ID', 'EXT', 'RESOLUTION', '\tFPS', 'HDR', delim
, '\tFILESIZE', '\tTBR', 'PROTO',
3608 delim
, 'VCODEC', '\tVBR', 'ACODEC', '\tABR', '\tASR', 'MORE INFO')
3610 return render_table(
3611 header_line
, table
, hide_empty
=True,
3612 delim
=self
._format
_out
('\u2500', self
.Styles
.DELIM
, '-', test_encoding
=True))
3614 def render_thumbnails_table(self
, info_dict
):
3615 thumbnails
= list(info_dict
.get('thumbnails') or [])
3618 return render_table(
3619 self
._list
_format
_headers
('ID', 'Width', 'Height', 'URL'),
3620 [[t
.get('id'), t
.get('width', 'unknown'), t
.get('height', 'unknown'), t
['url']] for t
in thumbnails
])
3622 def render_subtitles_table(self
, video_id
, subtitles
):
3623 def _row(lang
, formats
):
3624 exts
, names
= zip(*((f
['ext'], f
.get('name') or 'unknown') for f
in reversed(formats
)))
3625 if len(set(names
)) == 1:
3626 names
= [] if names
[0] == 'unknown' else names
[:1]
3627 return [lang
, ', '.join(names
), ', '.join(exts
)]
3631 return render_table(
3632 self
._list
_format
_headers
('Language', 'Name', 'Formats'),
3633 [_row(lang
, formats
) for lang
, formats
in subtitles
.items()],
3636 def __list_table(self
, video_id
, name
, func
, *args
):
3639 self
.to_screen(f
'{video_id} has no {name}')
3641 self
.to_screen(f
'[info] Available {name} for {video_id}:')
3642 self
.to_stdout(table
)
3644 def list_formats(self
, info_dict
):
3645 self
.__list
_table
(info_dict
['id'], 'formats', self
.render_formats_table
, info_dict
)
3647 def list_thumbnails(self
, info_dict
):
3648 self
.__list
_table
(info_dict
['id'], 'thumbnails', self
.render_thumbnails_table
, info_dict
)
3650 def list_subtitles(self
, video_id
, subtitles
, name
='subtitles'):
3651 self
.__list
_table
(video_id
, name
, self
.render_subtitles_table
, video_id
, subtitles
)
3653 def urlopen(self
, req
):
3654 """ Start an HTTP download """
3655 if isinstance(req
, str):
3656 req
= sanitized_Request(req
)
3657 return self
._opener
.open(req
, timeout
=self
._socket
_timeout
)
3659 def print_debug_header(self
):
3660 if not self
.params
.get('verbose'):
3663 # These imports can be slow. So import them only as needed
3664 from .extractor
.extractors
import _LAZY_LOADER
3665 from .extractor
.extractors
import _PLUGIN_CLASSES
as plugin_extractors
3667 def get_encoding(stream
):
3668 ret
= str(getattr(stream
, 'encoding', 'missing (%s)' % type(stream
).__name
__))
3669 if not supports_terminal_sequences(stream
):
3670 from .utils
import WINDOWS_VT_MODE
# Must be imported locally
3671 ret
+= ' (No VT)' if WINDOWS_VT_MODE
is False else ' (No ANSI)'
3674 encoding_str
= 'Encodings: locale %s, fs %s, pref %s, %s' % (
3675 locale
.getpreferredencoding(),
3676 sys
.getfilesystemencoding(),
3677 self
.get_encoding(),
3679 f
'{key} {get_encoding(stream)}' for key
, stream
in self
._out
_files
.items_
3680 if stream
is not None and key
!= 'console')
3683 logger
= self
.params
.get('logger')
3685 write_debug
= lambda msg
: logger
.debug(f
'[debug] {msg}')
3686 write_debug(encoding_str
)
3688 write_string(f
'[debug] {encoding_str}\n', encoding
=None)
3689 write_debug
= lambda msg
: self
._write
_string
(f
'[debug] {msg}\n')
3691 source
= detect_variant()
3692 write_debug(join_nonempty(
3693 'yt-dlp version', __version__
,
3694 f
'[{RELEASE_GIT_HEAD}]' if RELEASE_GIT_HEAD
else '',
3695 '' if source
== 'unknown' else f
'({source})',
3697 if not _LAZY_LOADER
:
3698 if os
.environ
.get('YTDLP_NO_LAZY_EXTRACTORS'):
3699 write_debug('Lazy loading extractors is forcibly disabled')
3701 write_debug('Lazy loading extractors is disabled')
3702 if plugin_extractors
or plugin_postprocessors
:
3703 write_debug('Plugins: %s' % [
3704 '%s%s' % (klass
.__name
__, '' if klass
.__name
__ == name
else f
' as {name}')
3705 for name
, klass
in itertools
.chain(plugin_extractors
.items(), plugin_postprocessors
.items())])
3706 if self
.params
['compat_opts']:
3707 write_debug('Compatibility options: %s' % ', '.join(self
.params
['compat_opts']))
3709 if source
== 'source':
3711 stdout
, _
, _
= Popen
.run(
3712 ['git', 'rev-parse', '--short', 'HEAD'],
3713 text
=True, cwd
=os
.path
.dirname(os
.path
.abspath(__file__
)),
3714 stdout
=subprocess
.PIPE
, stderr
=subprocess
.PIPE
)
3715 if re
.fullmatch('[0-9a-f]+', stdout
.strip()):
3716 write_debug(f
'Git HEAD: {stdout.strip()}')
3718 with contextlib
.suppress(Exception):
3721 write_debug(system_identifier())
3723 exe_versions
, ffmpeg_features
= FFmpegPostProcessor
.get_versions_and_features(self
)
3724 ffmpeg_features
= {key for key, val in ffmpeg_features.items() if val}
3726 exe_versions
['ffmpeg'] += ' (%s)' % ','.join(sorted(ffmpeg_features
))
3728 exe_versions
['rtmpdump'] = rtmpdump_version()
3729 exe_versions
['phantomjs'] = PhantomJSwrapper
._version
()
3730 exe_str
= ', '.join(
3731 f
'{exe} {v}' for exe
, v
in sorted(exe_versions
.items()) if v
3733 write_debug('exe versions: %s' % exe_str
)
3735 from .compat
.compat_utils
import get_package_info
3736 from .dependencies
import available_dependencies
3738 write_debug('Optional libraries: %s' % (', '.join(sorted({
3739 join_nonempty(*get_package_info(m
)) for m
in available_dependencies
.values()
3742 self
._setup
_opener
()
3744 for handler
in self
._opener
.handlers
:
3745 if hasattr(handler
, 'proxies'):
3746 proxy_map
.update(handler
.proxies
)
3747 write_debug(f
'Proxy map: {proxy_map}')
3750 if False and self
.params
.get('call_home'):
3751 ipaddr
= self
.urlopen('https://yt-dl.org/ip').read().decode()
3752 write_debug('Public IP address: %s' % ipaddr
)
3753 latest_version
= self
.urlopen(
3754 'https://yt-dl.org/latest/version').read().decode()
3755 if version_tuple(latest_version
) > version_tuple(__version__
):
3756 self
.report_warning(
3757 'You are using an outdated version (newest version: %s)! '
3758 'See https://yt-dl.org/update if you need help updating.' %
3761 def _setup_opener(self
):
3762 if hasattr(self
, '_opener'):
3764 timeout_val
= self
.params
.get('socket_timeout')
3765 self
._socket
_timeout
= 20 if timeout_val
is None else float(timeout_val
)
3767 opts_cookiesfrombrowser
= self
.params
.get('cookiesfrombrowser')
3768 opts_cookiefile
= self
.params
.get('cookiefile')
3769 opts_proxy
= self
.params
.get('proxy')
3771 self
.cookiejar
= load_cookies(opts_cookiefile
, opts_cookiesfrombrowser
, self
)
3773 cookie_processor
= YoutubeDLCookieProcessor(self
.cookiejar
)
3774 if opts_proxy
is not None:
3775 if opts_proxy
== '':
3778 proxies
= {'http': opts_proxy, 'https': opts_proxy}
3780 proxies
= urllib
.request
.getproxies()
3781 # Set HTTPS proxy to HTTP one if given (https://github.com/ytdl-org/youtube-dl/issues/805)
3782 if 'http' in proxies
and 'https' not in proxies
:
3783 proxies
['https'] = proxies
['http']
3784 proxy_handler
= PerRequestProxyHandler(proxies
)
3786 debuglevel
= 1 if self
.params
.get('debug_printtraffic') else 0
3787 https_handler
= make_HTTPS_handler(self
.params
, debuglevel
=debuglevel
)
3788 ydlh
= YoutubeDLHandler(self
.params
, debuglevel
=debuglevel
)
3789 redirect_handler
= YoutubeDLRedirectHandler()
3790 data_handler
= urllib
.request
.DataHandler()
3792 # When passing our own FileHandler instance, build_opener won't add the
3793 # default FileHandler and allows us to disable the file protocol, which
3794 # can be used for malicious purposes (see
3795 # https://github.com/ytdl-org/youtube-dl/issues/8227)
3796 file_handler
= urllib
.request
.FileHandler()
3798 def file_open(*args
, **kwargs
):
3799 raise urllib
.error
.URLError('file:// scheme is explicitly disabled in yt-dlp for security reasons')
3800 file_handler
.file_open
= file_open
3802 opener
= urllib
.request
.build_opener(
3803 proxy_handler
, https_handler
, cookie_processor
, ydlh
, redirect_handler
, data_handler
, file_handler
)
3805 # Delete the default user-agent header, which would otherwise apply in
3806 # cases where our custom HTTP handler doesn't come into play
3807 # (See https://github.com/ytdl-org/youtube-dl/issues/1309 for details)
3808 opener
.addheaders
= []
3809 self
._opener
= opener
3811 def encode(self
, s
):
3812 if isinstance(s
, bytes):
3813 return s
# Already encoded
3816 return s
.encode(self
.get_encoding())
3817 except UnicodeEncodeError as err
:
3818 err
.reason
= err
.reason
+ '. Check your system encoding configuration or use the --encoding option.'
3821 def get_encoding(self
):
3822 encoding
= self
.params
.get('encoding')
3823 if encoding
is None:
3824 encoding
= preferredencoding()
3827 def _write_info_json(self
, label
, ie_result
, infofn
, overwrite
=None):
3828 ''' Write infojson and returns True = written, 'exists' = Already exists, False = skip, None = error '''
3829 if overwrite
is None:
3830 overwrite
= self
.params
.get('overwrites', True)
3831 if not self
.params
.get('writeinfojson'):
3834 self
.write_debug(f
'Skipping writing {label} infojson')
3836 elif not self
._ensure
_dir
_exists
(infofn
):
3838 elif not overwrite
and os
.path
.exists(infofn
):
3839 self
.to_screen(f
'[info] {label.title()} metadata is already present')
3842 self
.to_screen(f
'[info] Writing {label} metadata as JSON to: {infofn}')
3844 write_json_file(self
.sanitize_info(ie_result
, self
.params
.get('clean_infojson', True)), infofn
)
3847 self
.report_error(f
'Cannot write {label} metadata to JSON file {infofn}')
3850 def _write_description(self
, label
, ie_result
, descfn
):
3851 ''' Write description and returns True = written, False = skip, None = error '''
3852 if not self
.params
.get('writedescription'):
3855 self
.write_debug(f
'Skipping writing {label} description')
3857 elif not self
._ensure
_dir
_exists
(descfn
):
3859 elif not self
.params
.get('overwrites', True) and os
.path
.exists(descfn
):
3860 self
.to_screen(f
'[info] {label.title()} description is already present')
3861 elif ie_result
.get('description') is None:
3862 self
.report_warning(f
'There\'s no {label} description to write')
3866 self
.to_screen(f
'[info] Writing {label} description to: {descfn}')
3867 with open(encodeFilename(descfn
), 'w', encoding
='utf-8') as descfile
:
3868 descfile
.write(ie_result
['description'])
3870 self
.report_error(f
'Cannot write {label} description file {descfn}')
3874 def _write_subtitles(self
, info_dict
, filename
):
3875 ''' Write subtitles to file and return list of (sub_filename, final_sub_filename); or None if error'''
3877 subtitles
= info_dict
.get('requested_subtitles')
3878 if not subtitles
or not (self
.params
.get('writesubtitles') or self
.params
.get('writeautomaticsub')):
3879 # subtitles download errors are already managed as troubles in relevant IE
3880 # that way it will silently go on when used with unsupporting IE
3883 sub_filename_base
= self
.prepare_filename(info_dict
, 'subtitle')
3884 if not sub_filename_base
:
3885 self
.to_screen('[info] Skipping writing video subtitles')
3887 for sub_lang
, sub_info
in subtitles
.items():
3888 sub_format
= sub_info
['ext']
3889 sub_filename
= subtitles_filename(filename
, sub_lang
, sub_format
, info_dict
.get('ext'))
3890 sub_filename_final
= subtitles_filename(sub_filename_base
, sub_lang
, sub_format
, info_dict
.get('ext'))
3891 existing_sub
= self
.existing_file((sub_filename_final
, sub_filename
))
3893 self
.to_screen(f
'[info] Video subtitle {sub_lang}.{sub_format} is already present')
3894 sub_info
['filepath'] = existing_sub
3895 ret
.append((existing_sub
, sub_filename_final
))
3898 self
.to_screen(f
'[info] Writing video subtitles to: {sub_filename}')
3899 if sub_info
.get('data') is not None:
3901 # Use newline='' to prevent conversion of newline characters
3902 # See https://github.com/ytdl-org/youtube-dl/issues/10268
3903 with open(sub_filename
, 'w', encoding
='utf-8', newline
='') as subfile
:
3904 subfile
.write(sub_info
['data'])
3905 sub_info
['filepath'] = sub_filename
3906 ret
.append((sub_filename
, sub_filename_final
))
3909 self
.report_error(f
'Cannot write video subtitles file {sub_filename}')
3913 sub_copy
= sub_info
.copy()
3914 sub_copy
.setdefault('http_headers', info_dict
.get('http_headers'))
3915 self
.dl(sub_filename
, sub_copy
, subtitle
=True)
3916 sub_info
['filepath'] = sub_filename
3917 ret
.append((sub_filename
, sub_filename_final
))
3918 except (DownloadError
, ExtractorError
, IOError, OSError, ValueError) + network_exceptions
as err
:
3919 msg
= f
'Unable to download video subtitles for {sub_lang!r}: {err}'
3920 if self
.params
.get('ignoreerrors') is not True: # False or 'only_download'
3921 if not self
.params
.get('ignoreerrors'):
3922 self
.report_error(msg
)
3923 raise DownloadError(msg
)
3924 self
.report_warning(msg
)
3927 def _write_thumbnails(self
, label
, info_dict
, filename
, thumb_filename_base
=None):
3928 ''' Write thumbnails to file and return list of (thumb_filename, final_thumb_filename) '''
3929 write_all
= self
.params
.get('write_all_thumbnails', False)
3930 thumbnails
, ret
= [], []
3931 if write_all
or self
.params
.get('writethumbnail', False):
3932 thumbnails
= info_dict
.get('thumbnails') or []
3933 multiple
= write_all
and len(thumbnails
) > 1
3935 if thumb_filename_base
is None:
3936 thumb_filename_base
= filename
3937 if thumbnails
and not thumb_filename_base
:
3938 self
.write_debug(f
'Skipping writing {label} thumbnail')
3941 for idx
, t
in list(enumerate(thumbnails
))[::-1]:
3942 thumb_ext
= (f
'{t["id"]}.' if multiple
else '') + determine_ext(t
['url'], 'jpg')
3943 thumb_display_id
= f
'{label} thumbnail {t["id"]}'
3944 thumb_filename
= replace_extension(filename
, thumb_ext
, info_dict
.get('ext'))
3945 thumb_filename_final
= replace_extension(thumb_filename_base
, thumb_ext
, info_dict
.get('ext'))
3947 existing_thumb
= self
.existing_file((thumb_filename_final
, thumb_filename
))
3949 self
.to_screen('[info] %s is already present' % (
3950 thumb_display_id
if multiple
else f
'{label} thumbnail').capitalize())
3951 t
['filepath'] = existing_thumb
3952 ret
.append((existing_thumb
, thumb_filename_final
))
3954 self
.to_screen(f
'[info] Downloading {thumb_display_id} ...')
3956 uf
= self
.urlopen(sanitized_Request(t
['url'], headers
=t
.get('http_headers', {})))
3957 self
.to_screen(f
'[info] Writing {thumb_display_id} to: {thumb_filename}')
3958 with open(encodeFilename(thumb_filename
), 'wb') as thumbf
:
3959 shutil
.copyfileobj(uf
, thumbf
)
3960 ret
.append((thumb_filename
, thumb_filename_final
))
3961 t
['filepath'] = thumb_filename
3962 except network_exceptions
as err
:
3964 self
.report_warning(f
'Unable to download {thumb_display_id}: {err}')
3965 if ret
and not write_all
: