4 from __future__
import absolute_import
, unicode_literals
31 from string
import ascii_letters
35 compat_get_terminal_size
,
42 compat_tokenize_tokenize
,
44 compat_urllib_request
,
45 compat_urllib_request_DataHandler
,
46 windows_enable_vt_mode
,
48 from .cookies
import load_cookies
70 format_decimal_suffix
,
92 PerRequestProxyHandler
,
100 register_socks_protocols
,
101 RejectedVideoReached
,
102 remove_terminal_sequences
,
116 supports_terminal_sequences
,
121 UnavailableVideoError
,
127 YoutubeDLCookieProcessor
,
129 YoutubeDLRedirectHandler
,
131 from .cache
import Cache
132 from .minicurses
import format_text
133 from .extractor
import (
134 gen_extractor_classes
,
137 _PLUGIN_CLASSES
as plugin_extractors
139 from .extractor
.openload
import PhantomJSwrapper
140 from .downloader
import (
142 get_suitable_downloader
,
143 shorten_protocol_name
145 from .downloader
.rtmp
import rtmpdump_version
146 from .postprocessor
import (
149 FFmpegFixupDuplicateMoovPP
,
150 FFmpegFixupDurationPP
,
153 FFmpegFixupStretchedPP
,
154 FFmpegFixupTimestampPP
,
157 MoveFilesAfterDownloadPP
,
158 _PLUGIN_CLASSES
as plugin_postprocessors
160 from .update
import detect_variant
161 from .version
import __version__
, RELEASE_GIT_HEAD
163 if compat_os_name
== 'nt':
167 class YoutubeDL(object):
170 YoutubeDL objects are the ones responsible of downloading the
171 actual video file and writing it to disk if the user has requested
172 it, among some other tasks. In most cases there should be one per
173 program. As, given a video URL, the downloader doesn't know how to
174 extract all the needed information, task that InfoExtractors do, it
175 has to pass the URL to one of them.
177 For this, YoutubeDL objects have a method that allows
178 InfoExtractors to be registered in a given order. When it is passed
179 a URL, the YoutubeDL object handles it to the first InfoExtractor it
180 finds that reports being able to handle it. The InfoExtractor extracts
181 all the information about the video or videos the URL refers to, and
182 YoutubeDL process the extracted information, possibly using a File
183 Downloader to download the video.
185 YoutubeDL objects accept a lot of parameters. In order not to saturate
186 the object constructor with arguments, it receives a dictionary of
187 options instead. These options are available through the params
188 attribute for the InfoExtractors to use. The YoutubeDL also
189 registers itself as the downloader in charge for the InfoExtractors
190 that are added to it, so this is a "mutual registration".
194 username: Username for authentication purposes.
195 password: Password for authentication purposes.
196 videopassword: Password for accessing a video.
197 ap_mso: Adobe Pass multiple-system operator identifier.
198 ap_username: Multiple-system operator account username.
199 ap_password: Multiple-system operator account password.
200 usenetrc: Use netrc for authentication instead.
201 verbose: Print additional info to stdout.
202 quiet: Do not print messages to stdout.
203 no_warnings: Do not print out anything for warnings.
204 forceprint: A dict with keys WHEN mapped to a list of templates to
205 print to stdout. The allowed keys are video or any of the
206 items in utils.POSTPROCESS_WHEN.
207 For compatibility, a single list is also accepted
208 print_to_file: A dict with keys WHEN (same as forceprint) mapped to
209 a list of tuples with (template, filename)
210 forceurl: Force printing final URL. (Deprecated)
211 forcetitle: Force printing title. (Deprecated)
212 forceid: Force printing ID. (Deprecated)
213 forcethumbnail: Force printing thumbnail URL. (Deprecated)
214 forcedescription: Force printing description. (Deprecated)
215 forcefilename: Force printing final filename. (Deprecated)
216 forceduration: Force printing duration. (Deprecated)
217 forcejson: Force printing info_dict as JSON.
218 dump_single_json: Force printing the info_dict of the whole playlist
219 (or video) as a single JSON line.
220 force_write_download_archive: Force writing download archive regardless
221 of 'skip_download' or 'simulate'.
222 simulate: Do not download the video files. If unset (or None),
223 simulate only if listsubtitles, listformats or list_thumbnails is used
224 format: Video format code. see "FORMAT SELECTION" for more details.
225 You can also pass a function. The function takes 'ctx' as
226 argument and returns the formats to download.
227 See "build_format_selector" for an implementation
228 allow_unplayable_formats: Allow unplayable formats to be extracted and downloaded.
229 ignore_no_formats_error: Ignore "No video formats" error. Usefull for
230 extracting metadata even if the video is not actually
231 available for download (experimental)
232 format_sort: A list of fields by which to sort the video formats.
233 See "Sorting Formats" for more details.
234 format_sort_force: Force the given format_sort. see "Sorting Formats"
236 allow_multiple_video_streams: Allow multiple video streams to be merged
238 allow_multiple_audio_streams: Allow multiple audio streams to be merged
240 check_formats Whether to test if the formats are downloadable.
241 Can be True (check all), False (check none),
242 'selected' (check selected formats),
243 or None (check only if requested by extractor)
244 paths: Dictionary of output paths. The allowed keys are 'home'
245 'temp' and the keys of OUTTMPL_TYPES (in utils.py)
246 outtmpl: Dictionary of templates for output names. Allowed keys
247 are 'default' and the keys of OUTTMPL_TYPES (in utils.py).
248 For compatibility with youtube-dl, a single string can also be used
249 outtmpl_na_placeholder: Placeholder for unavailable meta fields.
250 restrictfilenames: Do not allow "&" and spaces in file names
251 trim_file_name: Limit length of filename (extension excluded)
252 windowsfilenames: Force the filenames to be windows compatible
253 ignoreerrors: Do not stop on download/postprocessing errors.
254 Can be 'only_download' to ignore only download errors.
255 Default is 'only_download' for CLI, but False for API
256 skip_playlist_after_errors: Number of allowed failures until the rest of
257 the playlist is skipped
258 force_generic_extractor: Force downloader to use the generic extractor
259 overwrites: Overwrite all video and metadata files if True,
260 overwrite only non-video files if None
261 and don't overwrite any file if False
262 For compatibility with youtube-dl,
263 "nooverwrites" may also be used instead
264 playliststart: Playlist item to start at.
265 playlistend: Playlist item to end at.
266 playlist_items: Specific indices of playlist to download.
267 playlistreverse: Download playlist items in reverse order.
268 playlistrandom: Download playlist items in random order.
269 matchtitle: Download only matching titles.
270 rejecttitle: Reject downloads for matching titles.
271 logger: Log messages to a logging.Logger instance.
272 logtostderr: Log messages to stderr instead of stdout.
273 consoletitle: Display progress in console window's titlebar.
274 writedescription: Write the video description to a .description file
275 writeinfojson: Write the video description to a .info.json file
276 clean_infojson: Remove private fields from the infojson
277 getcomments: Extract video comments. This will not be written to disk
278 unless writeinfojson is also given
279 writeannotations: Write the video annotations to a .annotations.xml file
280 writethumbnail: Write the thumbnail image to a file
281 allow_playlist_files: Whether to write playlists' description, infojson etc
282 also to disk when using the 'write*' options
283 write_all_thumbnails: Write all thumbnail formats to files
284 writelink: Write an internet shortcut file, depending on the
285 current platform (.url/.webloc/.desktop)
286 writeurllink: Write a Windows internet shortcut file (.url)
287 writewebloclink: Write a macOS internet shortcut file (.webloc)
288 writedesktoplink: Write a Linux internet shortcut file (.desktop)
289 writesubtitles: Write the video subtitles to a file
290 writeautomaticsub: Write the automatically generated subtitles to a file
291 allsubtitles: Deprecated - Use subtitleslangs = ['all']
292 Downloads all the subtitles of the video
293 (requires writesubtitles or writeautomaticsub)
294 listsubtitles: Lists all available subtitles for the video
295 subtitlesformat: The format code for subtitles
296 subtitleslangs: List of languages of the subtitles to download (can be regex).
297 The list may contain "all" to refer to all the available
298 subtitles. The language can be prefixed with a "-" to
299 exclude it from the requested languages. Eg: ['all', '-live_chat']
300 keepvideo: Keep the video file after post-processing
301 daterange: A DateRange object, download only if the upload_date is in the range.
302 skip_download: Skip the actual download of the video file
303 cachedir: Location of the cache files in the filesystem.
304 False to disable filesystem cache.
305 noplaylist: Download single video instead of a playlist if in doubt.
306 age_limit: An integer representing the user's age in years.
307 Unsuitable videos for the given age are skipped.
308 min_views: An integer representing the minimum view count the video
309 must have in order to not be skipped.
310 Videos without view count information are always
311 downloaded. None for no limit.
312 max_views: An integer representing the maximum view count.
313 Videos that are more popular than that are not
315 Videos without view count information are always
316 downloaded. None for no limit.
317 download_archive: File name of a file where all downloads are recorded.
318 Videos already present in the file are not downloaded
320 break_on_existing: Stop the download process after attempting to download a
321 file that is in the archive.
322 break_on_reject: Stop the download process when encountering a video that
323 has been filtered out.
324 break_per_url: Whether break_on_reject and break_on_existing
325 should act on each input URL as opposed to for the entire queue
326 cookiefile: File name where cookies should be read from and dumped to
327 cookiesfrombrowser: A tuple containing the name of the browser, the profile
328 name/pathfrom where cookies are loaded, and the name of the
329 keyring. Eg: ('chrome', ) or ('vivaldi', 'default', 'BASICTEXT')
330 legacyserverconnect: Explicitly allow HTTPS connection to servers that do not
331 support RFC 5746 secure renegotiation
332 nocheckcertificate: Do not verify SSL certificates
333 prefer_insecure: Use HTTP instead of HTTPS to retrieve information.
334 At the moment, this is only supported by YouTube.
335 proxy: URL of the proxy server to use
336 geo_verification_proxy: URL of the proxy to use for IP address verification
337 on geo-restricted sites.
338 socket_timeout: Time to wait for unresponsive hosts, in seconds
339 bidi_workaround: Work around buggy terminals without bidirectional text
340 support, using fridibi
341 debug_printtraffic:Print out sent and received HTTP traffic
342 include_ads: Download ads as well (deprecated)
343 default_search: Prepend this string if an input url is not valid.
344 'auto' for elaborate guessing
345 encoding: Use this encoding instead of the system-specified.
346 extract_flat: Do not resolve URLs, return the immediate result.
347 Pass in 'in_playlist' to only show this behavior for
349 wait_for_video: If given, wait for scheduled streams to become available.
350 The value should be a tuple containing the range
351 (min_secs, max_secs) to wait between retries
352 postprocessors: A list of dictionaries, each with an entry
353 * key: The name of the postprocessor. See
354 yt_dlp/postprocessor/__init__.py for a list.
355 * when: When to run the postprocessor. Allowed values are
356 the entries of utils.POSTPROCESS_WHEN
357 Assumed to be 'post_process' if not given
358 post_hooks: Deprecated - Register a custom postprocessor instead
359 A list of functions that get called as the final step
360 for each video file, after all postprocessors have been
361 called. The filename will be passed as the only argument.
362 progress_hooks: A list of functions that get called on download
363 progress, with a dictionary with the entries
364 * status: One of "downloading", "error", or "finished".
365 Check this first and ignore unknown values.
366 * info_dict: The extracted info_dict
368 If status is one of "downloading", or "finished", the
369 following properties may also be present:
370 * filename: The final filename (always present)
371 * tmpfilename: The filename we're currently writing to
372 * downloaded_bytes: Bytes on disk
373 * total_bytes: Size of the whole file, None if unknown
374 * total_bytes_estimate: Guess of the eventual file size,
376 * elapsed: The number of seconds since download started.
377 * eta: The estimated time in seconds, None if unknown
378 * speed: The download speed in bytes/second, None if
380 * fragment_index: The counter of the currently
381 downloaded video fragment.
382 * fragment_count: The number of fragments (= individual
383 files that will be merged)
385 Progress hooks are guaranteed to be called at least once
386 (with status "finished") if the download is successful.
387 postprocessor_hooks: A list of functions that get called on postprocessing
388 progress, with a dictionary with the entries
389 * status: One of "started", "processing", or "finished".
390 Check this first and ignore unknown values.
391 * postprocessor: Name of the postprocessor
392 * info_dict: The extracted info_dict
394 Progress hooks are guaranteed to be called at least twice
395 (with status "started" and "finished") if the processing is successful.
396 merge_output_format: Extension to use when merging formats.
397 final_ext: Expected final extension; used to detect when the file was
398 already downloaded and converted
399 fixup: Automatically correct known faults of the file.
401 - "never": do nothing
402 - "warn": only emit a warning
403 - "detect_or_warn": check whether we can do anything
404 about it, warn otherwise (default)
405 source_address: Client-side IP address to bind to.
406 call_home: Boolean, true iff we are allowed to contact the
407 yt-dlp servers for debugging. (BROKEN)
408 sleep_interval_requests: Number of seconds to sleep between requests
410 sleep_interval: Number of seconds to sleep before each download when
411 used alone or a lower bound of a range for randomized
412 sleep before each download (minimum possible number
413 of seconds to sleep) when used along with
415 max_sleep_interval:Upper bound of a range for randomized sleep before each
416 download (maximum possible number of seconds to sleep).
417 Must only be used along with sleep_interval.
418 Actual sleep time will be a random float from range
419 [sleep_interval; max_sleep_interval].
420 sleep_interval_subtitles: Number of seconds to sleep before each subtitle download
421 listformats: Print an overview of available video formats and exit.
422 list_thumbnails: Print a table of all thumbnails and exit.
423 match_filter: A function that gets called with the info_dict of
425 If it returns a message, the video is ignored.
426 If it returns None, the video is downloaded.
427 match_filter_func in utils.py is one example for this.
428 no_color: Do not emit color codes in output.
429 geo_bypass: Bypass geographic restriction via faking X-Forwarded-For
432 Two-letter ISO 3166-2 country code that will be used for
433 explicit geographic restriction bypassing via faking
434 X-Forwarded-For HTTP header
436 IP range in CIDR notation that will be used similarly to
439 The following options determine which downloader is picked:
440 external_downloader: A dictionary of protocol keys and the executable of the
441 external downloader to use for it. The allowed protocols
442 are default|http|ftp|m3u8|dash|rtsp|rtmp|mms.
443 Set the value to 'native' to use the native downloader
444 hls_prefer_native: Deprecated - Use external_downloader = {'m3u8': 'native'}
445 or {'m3u8': 'ffmpeg'} instead.
446 Use the native HLS downloader instead of ffmpeg/avconv
447 if True, otherwise use ffmpeg/avconv if False, otherwise
448 use downloader suggested by extractor if None.
449 compat_opts: Compatibility options. See "Differences in default behavior".
450 The following options do not work when used through the API:
451 filename, abort-on-error, multistreams, no-live-chat, format-sort
452 no-clean-infojson, no-playlist-metafiles, no-keep-subs, no-attach-info-json.
453 Refer __init__.py for their implementation
454 progress_template: Dictionary of templates for progress outputs.
455 Allowed keys are 'download', 'postprocess',
456 'download-title' (console title) and 'postprocess-title'.
457 The template is mapped on a dictionary with keys 'progress' and 'info'
459 The following parameters are not used by YoutubeDL itself, they are used by
460 the downloader (see yt_dlp/downloader/common.py):
461 nopart, updatetime, buffersize, ratelimit, throttledratelimit, min_filesize,
462 max_filesize, test, noresizebuffer, retries, file_access_retries, fragment_retries,
463 continuedl, noprogress, xattr_set_filesize, hls_use_mpegts, http_chunk_size,
464 external_downloader_args, concurrent_fragment_downloads.
466 The following options are used by the post processors:
467 prefer_ffmpeg: If False, use avconv instead of ffmpeg if both are available,
468 otherwise prefer ffmpeg. (avconv support is deprecated)
469 ffmpeg_location: Location of the ffmpeg/avconv binary; either the path
470 to the binary or its containing directory.
471 postprocessor_args: A dictionary of postprocessor/executable keys (in lower case)
472 and a list of additional command-line arguments for the
473 postprocessor/executable. The dict can also have "PP+EXE" keys
474 which are used when the given exe is used by the given PP.
475 Use 'default' as the name for arguments to passed to all PP
476 For compatibility with youtube-dl, a single list of args
479 The following options are used by the extractors:
480 extractor_retries: Number of times to retry for known errors
481 dynamic_mpd: Whether to process dynamic DASH manifests (default: True)
482 hls_split_discontinuity: Split HLS playlists to different formats at
483 discontinuities such as ad breaks (default: False)
484 extractor_args: A dictionary of arguments to be passed to the extractors.
485 See "EXTRACTOR ARGUMENTS" for details.
486 Eg: {'youtube': {'skip': ['dash', 'hls']}}
487 mark_watched: Mark videos watched (even with --simulate). Only for YouTube
488 youtube_include_dash_manifest: Deprecated - Use extractor_args instead.
489 If True (default), DASH manifests and related
490 data will be downloaded and processed by extractor.
491 You can reduce network I/O by disabling it if you don't
492 care about DASH. (only for youtube)
493 youtube_include_hls_manifest: Deprecated - Use extractor_args instead.
494 If True (default), HLS manifests and related
495 data will be downloaded and processed by extractor.
496 You can reduce network I/O by disabling it if you don't
497 care about HLS. (only for youtube)
500 _NUMERIC_FIELDS
= set((
501 'width', 'height', 'tbr', 'abr', 'asr', 'vbr', 'fps', 'filesize', 'filesize_approx',
502 'timestamp', 'release_timestamp',
503 'duration', 'view_count', 'like_count', 'dislike_count', 'repost_count',
504 'average_rating', 'comment_count', 'age_limit',
505 'start_time', 'end_time',
506 'chapter_number', 'season_number', 'episode_number',
507 'track_number', 'disc_number', 'release_year',
510 _format_selection_exts
= {
511 'audio': {'m4a', 'mp3', 'ogg', 'aac'}
,
512 'video': {'mp4', 'flv', 'webm', '3gp'}
,
513 'storyboards': {'mhtml'}
,
518 _pps
= {k: [] for k in POSTPROCESS_WHEN}
519 _printed_messages
= set()
520 _first_webpage_request
= True
521 _download_retcode
= None
522 _num_downloads
= None
524 _playlist_urls
= set()
527 def __init__(self
, params
=None, auto_init
=True):
528 """Create a FileDownloader object with the given options.
529 @param auto_init Whether to load the default extractors and print header (if verbose).
530 Set to 'no_verbose_header' to not print the header
535 self
._ies
_instances
= {}
536 self
._pps
= {k: [] for k in POSTPROCESS_WHEN}
537 self
._printed
_messages
= set()
538 self
._first
_webpage
_request
= True
539 self
._post
_hooks
= []
540 self
._progress
_hooks
= []
541 self
._postprocessor
_hooks
= []
542 self
._download
_retcode
= 0
543 self
._num
_downloads
= 0
545 self
._screen
_file
= [sys
.stdout
, sys
.stderr
][params
.get('logtostderr', False)]
546 self
._err
_file
= sys
.stderr
548 self
.cache
= Cache(self
)
550 windows_enable_vt_mode()
551 self
._allow
_colors
= {
552 'screen': not self
.params
.get('no_color') and supports_terminal_sequences(self
._screen
_file
),
553 'err': not self
.params
.get('no_color') and supports_terminal_sequences(self
._err
_file
),
556 if sys
.version_info
< (3, 6):
558 'Python version %d.%d is not supported! Please update to Python 3.6 or above' % sys
.version_info
[:2])
560 if self
.params
.get('allow_unplayable_formats'):
562 f
'You have asked for {self._format_err("UNPLAYABLE", self.Styles.EMPHASIS)} formats to be listed/downloaded. '
563 'This is a developer option intended for debugging. \n'
564 ' If you experience any issues while using this option, '
565 f
'{self._format_err("DO NOT", self.Styles.ERROR)} open a bug report')
567 def check_deprecated(param
, option
, suggestion
):
568 if self
.params
.get(param
) is not None:
569 self
.report_warning('%s is deprecated. Use %s instead' % (option
, suggestion
))
573 if check_deprecated('cn_verification_proxy', '--cn-verification-proxy', '--geo-verification-proxy'):
574 if self
.params
.get('geo_verification_proxy') is None:
575 self
.params
['geo_verification_proxy'] = self
.params
['cn_verification_proxy']
577 check_deprecated('autonumber', '--auto-number', '-o "%(autonumber)s-%(title)s.%(ext)s"')
578 check_deprecated('usetitle', '--title', '-o "%(title)s-%(id)s.%(ext)s"')
579 check_deprecated('useid', '--id', '-o "%(id)s.%(ext)s"')
581 for msg
in self
.params
.get('_warnings', []):
582 self
.report_warning(msg
)
583 for msg
in self
.params
.get('_deprecation_warnings', []):
584 self
.deprecation_warning(msg
)
586 if 'list-formats' in self
.params
.get('compat_opts', []):
587 self
.params
['listformats_table'] = False
589 if 'overwrites' not in self
.params
and self
.params
.get('nooverwrites') is not None:
590 # nooverwrites was unnecessarily changed to overwrites
591 # in 0c3d0f51778b153f65c21906031c2e091fcfb641
592 # This ensures compatibility with both keys
593 self
.params
['overwrites'] = not self
.params
['nooverwrites']
594 elif self
.params
.get('overwrites') is None:
595 self
.params
.pop('overwrites', None)
597 self
.params
['nooverwrites'] = not self
.params
['overwrites']
599 self
.params
.setdefault('forceprint', {})
600 self
.params
.setdefault('print_to_file', {})
602 # Compatibility with older syntax
603 if not isinstance(params
['forceprint'], dict):
604 self
.params
['forceprint'] = {'video': params['forceprint']}
606 if self
.params
.get('bidi_workaround', False):
609 master
, slave
= pty
.openpty()
610 width
= compat_get_terminal_size().columns
614 width_args
= ['-w', str(width
)]
616 stdin
=subprocess
.PIPE
,
618 stderr
=self
._err
_file
)
620 self
._output
_process
= Popen(['bidiv'] + width_args
, **sp_kwargs
)
622 self
._output
_process
= Popen(['fribidi', '-c', 'UTF-8'] + width_args
, **sp_kwargs
)
623 self
._output
_channel
= os
.fdopen(master
, 'rb')
624 except OSError as ose
:
625 if ose
.errno
== errno
.ENOENT
:
627 'Could not find fribidi executable, ignoring --bidi-workaround. '
628 'Make sure that fribidi is an executable file in one of the directories in your $PATH.')
632 if (sys
.platform
!= 'win32'
633 and sys
.getfilesystemencoding() in ['ascii', 'ANSI_X3.4-1968']
634 and not self
.params
.get('restrictfilenames', False)):
635 # Unicode filesystem API will throw errors (#1474, #13027)
637 'Assuming --restrict-filenames since file system encoding '
638 'cannot encode all characters. '
639 'Set the LC_ALL environment variable to fix this.')
640 self
.params
['restrictfilenames'] = True
642 self
.outtmpl_dict
= self
.parse_outtmpl()
644 # Creating format selector here allows us to catch syntax errors before the extraction
645 self
.format_selector
= (
646 self
.params
.get('format') if self
.params
.get('format') in (None, '-')
647 else self
.params
['format'] if callable(self
.params
['format'])
648 else self
.build_format_selector(self
.params
['format']))
653 if auto_init
!= 'no_verbose_header':
654 self
.print_debug_header()
655 self
.add_default_info_extractors()
658 'post_hooks': self
.add_post_hook
,
659 'progress_hooks': self
.add_progress_hook
,
660 'postprocessor_hooks': self
.add_postprocessor_hook
,
662 for opt
, fn
in hooks
.items():
663 for ph
in self
.params
.get(opt
, []):
666 for pp_def_raw
in self
.params
.get('postprocessors', []):
667 pp_def
= dict(pp_def_raw
)
668 when
= pp_def
.pop('when', 'post_process')
669 self
.add_post_processor(
670 get_postprocessor(pp_def
.pop('key'))(self
, **compat_kwargs(pp_def
)),
673 register_socks_protocols()
675 def preload_download_archive(fn
):
676 """Preload the archive, if any is specified"""
679 self
.write_debug(f
'Loading archive file {fn!r}')
681 with locked_file(fn
, 'r', encoding
='utf-8') as archive_file
:
682 for line
in archive_file
:
683 self
.archive
.add(line
.strip())
684 except IOError as ioe
:
685 if ioe
.errno
!= errno
.ENOENT
:
691 preload_download_archive(self
.params
.get('download_archive'))
693 def warn_if_short_id(self
, argv
):
694 # short YouTube ID starting with dash?
696 i
for i
, a
in enumerate(argv
)
697 if re
.match(r
'^-[0-9A-Za-z_-]{10}$', a
)]
701 + [a
for i
, a
in enumerate(argv
) if i
not in idxs
]
702 + ['--'] + [argv
[i
] for i
in idxs
]
705 'Long argument string detected. '
706 'Use -- to separate parameters and URLs, like this:\n%s' %
707 args_to_str(correct_argv
))
709 def add_info_extractor(self
, ie
):
710 """Add an InfoExtractor object to the end of the list."""
712 self
._ies
[ie_key
] = ie
713 if not isinstance(ie
, type):
714 self
._ies
_instances
[ie_key
] = ie
715 ie
.set_downloader(self
)
717 def _get_info_extractor_class(self
, ie_key
):
718 ie
= self
._ies
.get(ie_key
)
720 ie
= get_info_extractor(ie_key
)
721 self
.add_info_extractor(ie
)
724 def get_info_extractor(self
, ie_key
):
726 Get an instance of an IE with name ie_key, it will try to get one from
727 the _ies list, if there's no instance it will create a new one and add
728 it to the extractor list.
730 ie
= self
._ies
_instances
.get(ie_key
)
732 ie
= get_info_extractor(ie_key
)()
733 self
.add_info_extractor(ie
)
736 def add_default_info_extractors(self
):
738 Add the InfoExtractors returned by gen_extractors to the end of the list
740 for ie
in gen_extractor_classes():
741 self
.add_info_extractor(ie
)
743 def add_post_processor(self
, pp
, when
='post_process'):
744 """Add a PostProcessor object to the end of the chain."""
745 self
._pps
[when
].append(pp
)
746 pp
.set_downloader(self
)
748 def add_post_hook(self
, ph
):
749 """Add the post hook"""
750 self
._post
_hooks
.append(ph
)
752 def add_progress_hook(self
, ph
):
753 """Add the download progress hook"""
754 self
._progress
_hooks
.append(ph
)
756 def add_postprocessor_hook(self
, ph
):
757 """Add the postprocessing progress hook"""
758 self
._postprocessor
_hooks
.append(ph
)
759 for pps
in self
._pps
.values():
761 pp
.add_progress_hook(ph
)
763 def _bidi_workaround(self
, message
):
764 if not hasattr(self
, '_output_channel'):
767 assert hasattr(self
, '_output_process')
768 assert isinstance(message
, compat_str
)
769 line_count
= message
.count('\n') + 1
770 self
._output
_process
.stdin
.write((message
+ '\n').encode('utf-8'))
771 self
._output
_process
.stdin
.flush()
772 res
= ''.join(self
._output
_channel
.readline().decode('utf-8')
773 for _
in range(line_count
))
774 return res
[:-len('\n')]
776 def _write_string(self
, message
, out
=None, only_once
=False):
778 if message
in self
._printed
_messages
:
780 self
._printed
_messages
.add(message
)
781 write_string(message
, out
=out
, encoding
=self
.params
.get('encoding'))
783 def to_stdout(self
, message
, skip_eol
=False, quiet
=False):
784 """Print message to stdout"""
785 if self
.params
.get('logger'):
786 self
.params
['logger'].debug(message
)
787 elif not quiet
or self
.params
.get('verbose'):
789 '%s%s' % (self
._bidi
_workaround
(message
), ('' if skip_eol
else '\n')),
790 self
._err
_file
if quiet
else self
._screen
_file
)
792 def to_stderr(self
, message
, only_once
=False):
793 """Print message to stderr"""
794 assert isinstance(message
, compat_str
)
795 if self
.params
.get('logger'):
796 self
.params
['logger'].error(message
)
798 self
._write
_string
('%s\n' % self
._bidi
_workaround
(message
), self
._err
_file
, only_once
=only_once
)
800 def to_console_title(self
, message
):
801 if not self
.params
.get('consoletitle', False):
803 message
= remove_terminal_sequences(message
)
804 if compat_os_name
== 'nt':
805 if ctypes
.windll
.kernel32
.GetConsoleWindow():
806 # c_wchar_p() might not be necessary if `message` is
807 # already of type unicode()
808 ctypes
.windll
.kernel32
.SetConsoleTitleW(ctypes
.c_wchar_p(message
))
809 elif 'TERM' in os
.environ
:
810 self
._write
_string
('\033]0;%s\007' % message
, self
._screen
_file
)
812 def save_console_title(self
):
813 if not self
.params
.get('consoletitle', False):
815 if self
.params
.get('simulate'):
817 if compat_os_name
!= 'nt' and 'TERM' in os
.environ
:
818 # Save the title on stack
819 self
._write
_string
('\033[22;0t', self
._screen
_file
)
821 def restore_console_title(self
):
822 if not self
.params
.get('consoletitle', False):
824 if self
.params
.get('simulate'):
826 if compat_os_name
!= 'nt' and 'TERM' in os
.environ
:
827 # Restore the title from stack
828 self
._write
_string
('\033[23;0t', self
._screen
_file
)
831 self
.save_console_title()
834 def __exit__(self
, *args
):
835 self
.restore_console_title()
837 if self
.params
.get('cookiefile') is not None:
838 self
.cookiejar
.save(ignore_discard
=True, ignore_expires
=True)
840 def trouble(self
, message
=None, tb
=None, is_error
=True):
841 """Determine action to take when a download problem appears.
843 Depending on if the downloader has been configured to ignore
844 download errors or not, this method may throw an exception or
845 not when errors are found, after printing the message.
847 @param tb If given, is additional traceback information
848 @param is_error Whether to raise error according to ignorerrors
850 if message
is not None:
851 self
.to_stderr(message
)
852 if self
.params
.get('verbose'):
854 if sys
.exc_info()[0]: # if .trouble has been called from an except block
856 if hasattr(sys
.exc_info()[1], 'exc_info') and sys
.exc_info()[1].exc_info
[0]:
857 tb
+= ''.join(traceback
.format_exception(*sys
.exc_info()[1].exc_info
))
858 tb
+= encode_compat_str(traceback
.format_exc())
860 tb_data
= traceback
.format_list(traceback
.extract_stack())
861 tb
= ''.join(tb_data
)
866 if not self
.params
.get('ignoreerrors'):
867 if sys
.exc_info()[0] and hasattr(sys
.exc_info()[1], 'exc_info') and sys
.exc_info()[1].exc_info
[0]:
868 exc_info
= sys
.exc_info()[1].exc_info
870 exc_info
= sys
.exc_info()
871 raise DownloadError(message
, exc_info
)
872 self
._download
_retcode
= 1
874 def to_screen(self
, message
, skip_eol
=False):
875 """Print message to stdout if not in quiet mode"""
877 message
, skip_eol
, quiet
=self
.params
.get('quiet', False))
881 EMPHASIS
= 'light blue'
886 SUPPRESS
= 'light black'
888 def _format_text(self
, handle
, allow_colors
, text
, f
, fallback
=None, *, test_encoding
=False):
891 encoding
= self
.params
.get('encoding') or getattr(handle
, 'encoding', 'ascii')
892 text
= text
.encode(encoding
, 'ignore').decode(encoding
)
893 if fallback
is not None and text
!= original_text
:
895 if isinstance(f
, self
.Styles
):
897 return format_text(text
, f
) if allow_colors
else text
if fallback
is None else fallback
899 def _format_screen(self
, *args
, **kwargs
):
900 return self
._format
_text
(
901 self
._screen
_file
, self
._allow
_colors
['screen'], *args
, **kwargs
)
903 def _format_err(self
, *args
, **kwargs
):
904 return self
._format
_text
(
905 self
._err
_file
, self
._allow
_colors
['err'], *args
, **kwargs
)
907 def report_warning(self
, message
, only_once
=False):
909 Print the message to stderr, it will be prefixed with 'WARNING:'
910 If stderr is a tty file the 'WARNING:' will be colored
912 if self
.params
.get('logger') is not None:
913 self
.params
['logger'].warning(message
)
915 if self
.params
.get('no_warnings'):
917 self
.to_stderr(f
'{self._format_err("WARNING:", self.Styles.WARNING)} {message}', only_once
)
919 def deprecation_warning(self
, message
):
920 if self
.params
.get('logger') is not None:
921 self
.params
['logger'].warning('DeprecationWarning: {message}')
923 self
.to_stderr(f
'{self._format_err("DeprecationWarning:", self.Styles.ERROR)} {message}', True)
925 def report_error(self
, message
, *args
, **kwargs
):
927 Do the same as trouble, but prefixes the message with 'ERROR:', colored
928 in red if stderr is a tty file.
930 self
.trouble(f
'{self._format_err("ERROR:", self.Styles.ERROR)} {message}', *args
, **kwargs
)
932 def write_debug(self
, message
, only_once
=False):
933 '''Log debug message or Print message to stderr'''
934 if not self
.params
.get('verbose', False):
936 message
= '[debug] %s' % message
937 if self
.params
.get('logger'):
938 self
.params
['logger'].debug(message
)
940 self
.to_stderr(message
, only_once
)
942 def report_file_already_downloaded(self
, file_name
):
943 """Report file has already been fully downloaded."""
945 self
.to_screen('[download] %s has already been downloaded' % file_name
)
946 except UnicodeEncodeError:
947 self
.to_screen('[download] The file has already been downloaded')
949 def report_file_delete(self
, file_name
):
950 """Report that existing file will be deleted."""
952 self
.to_screen('Deleting existing file %s' % file_name
)
953 except UnicodeEncodeError:
954 self
.to_screen('Deleting existing file')
956 def raise_no_formats(self
, info
, forced
=False):
957 has_drm
= info
.get('__has_drm')
958 msg
= 'This video is DRM protected' if has_drm
else 'No video formats found!'
959 expected
= self
.params
.get('ignore_no_formats_error')
960 if forced
or not expected
:
961 raise ExtractorError(msg
, video_id
=info
['id'], ie
=info
['extractor'],
962 expected
=has_drm
or expected
)
964 self
.report_warning(msg
)
966 def parse_outtmpl(self
):
967 outtmpl_dict
= self
.params
.get('outtmpl', {})
968 if not isinstance(outtmpl_dict
, dict):
969 outtmpl_dict
= {'default': outtmpl_dict}
970 # Remove spaces in the default template
971 if self
.params
.get('restrictfilenames'):
972 sanitize
= lambda x
: x
.replace(' - ', ' ').replace(' ', '-')
974 sanitize
= lambda x
: x
975 outtmpl_dict
.update({
976 k
: sanitize(v
) for k
, v
in DEFAULT_OUTTMPL
.items()
977 if outtmpl_dict
.get(k
) is None})
978 for key
, val
in outtmpl_dict
.items():
979 if isinstance(val
, bytes):
981 'Parameter outtmpl is bytes, but should be a unicode string. '
982 'Put from __future__ import unicode_literals at the top of your code file or consider switching to Python 3.x.')
985 def get_output_path(self
, dir_type
='', filename
=None):
986 paths
= self
.params
.get('paths', {})
987 assert isinstance(paths
, dict)
989 expand_path(paths
.get('home', '').strip()),
990 expand_path(paths
.get(dir_type
, '').strip()) if dir_type
else '',
993 # Temporary fix for #4787
994 # 'Treat' all problem characters by passing filename through preferredencoding
995 # to workaround encoding issues with subprocess on python2 @ Windows
996 if sys
.version_info
< (3, 0) and sys
.platform
== 'win32':
997 path
= encodeFilename(path
, True).decode(preferredencoding())
998 return sanitize_path(path
, force
=self
.params
.get('windowsfilenames'))
1001 def _outtmpl_expandpath(outtmpl
):
1002 # expand_path translates '%%' into '%' and '$$' into '$'
1003 # correspondingly that is not what we want since we need to keep
1004 # '%%' intact for template dict substitution step. Working around
1005 # with boundary-alike separator hack.
1006 sep
= ''.join([random
.choice(ascii_letters
) for _
in range(32)])
1007 outtmpl
= outtmpl
.replace('%%', '%{0}%'.format(sep
)).replace('$$', '${0}$'.format(sep
))
1009 # outtmpl should be expand_path'ed before template dict substitution
1010 # because meta fields may contain env variables we don't want to
1011 # be expanded. For example, for outtmpl "%(title)s.%(ext)s" and
1012 # title "Hello $PATH", we don't want `$PATH` to be expanded.
1013 return expand_path(outtmpl
).replace(sep
, '')
1016 def escape_outtmpl(outtmpl
):
1017 ''' Escape any remaining strings like %s, %abc% etc. '''
1019 STR_FORMAT_RE_TMPL
.format('', '(?![%(\0])'),
1020 lambda mobj
: ('' if mobj
.group('has_key') else '%') + mobj
.group(0),
1024 def validate_outtmpl(cls
, outtmpl
):
1025 ''' @return None or Exception object '''
1027 STR_FORMAT_RE_TMPL
.format('[^)]*', '[ljqBUDS]'),
1028 lambda mobj
: f
'{mobj.group(0)[:-1]}s',
1029 cls
._outtmpl
_expandpath
(outtmpl
))
1031 cls
.escape_outtmpl(outtmpl
) % collections
.defaultdict(int)
1033 except ValueError as err
:
1037 def _copy_infodict(info_dict
):
1038 info_dict
= dict(info_dict
)
1039 for key
in ('__original_infodict', '__postprocessors'):
1040 info_dict
.pop(key
, None)
1043 def prepare_outtmpl(self
, outtmpl
, info_dict
, sanitize
=False):
1044 """ Make the outtmpl and info_dict suitable for substitution: ydl.escape_outtmpl(outtmpl) % info_dict
1045 @param sanitize Whether to sanitize the output as a filename.
1046 For backward compatibility, a function can also be passed
1049 info_dict
.setdefault('epoch', int(time
.time())) # keep epoch consistent once set
1051 info_dict
= self
._copy
_infodict
(info_dict
)
1052 info_dict
['duration_string'] = ( # %(duration>%H-%M-%S)s is wrong if duration > 24hrs
1053 formatSeconds(info_dict
['duration'], '-' if sanitize
else ':')
1054 if info_dict
.get('duration', None) is not None
1056 info_dict
['autonumber'] = self
.params
.get('autonumber_start', 1) - 1 + self
._num
_downloads
1057 info_dict
['video_autonumber'] = self
._num
_videos
1058 if info_dict
.get('resolution') is None:
1059 info_dict
['resolution'] = self
.format_resolution(info_dict
, default
=None)
1061 # For fields playlist_index, playlist_autonumber and autonumber convert all occurrences
1062 # of %(field)s to %(field)0Nd for backward compatibility
1063 field_size_compat_map
= {
1064 'playlist_index': number_of_digits(info_dict
.get('_last_playlist_index') or 0),
1065 'playlist_autonumber': number_of_digits(info_dict
.get('n_entries') or 0),
1066 'autonumber': self
.params
.get('autonumber_size') or 5,
1070 EXTERNAL_FORMAT_RE
= re
.compile(STR_FORMAT_RE_TMPL
.format('[^)]*', f
'[{STR_FORMAT_TYPES}ljqBUDS]'))
1075 # Field is of the form key1.key2...
1076 # where keys (except first) can be string, int or slice
1077 FIELD_RE
= r
'\w*(?:\.(?:\w+|{num}|{num}?(?::{num}?){{1,2}}))*'.format(num
=r
'(?:-?\d+)')
1078 MATH_FIELD_RE
= r
'''(?:{field}|{num})'''.format(field
=FIELD_RE
, num
=r
'-?\d+(?:.\d+)?')
1079 MATH_OPERATORS_RE
= r
'(?:%s)' % '|'.join(map(re
.escape
, MATH_FUNCTIONS
.keys()))
1080 INTERNAL_FORMAT_RE
= re
.compile(r
'''(?x)
1083 (?P<maths>(?:{math_op}{math_field})*)
1084 (?:>(?P<strf_format>.+?))?
1085 (?P<alternate>(?<!\\),[^|&)]+)?
1086 (?:&(?P<replacement>.*?))?
1087 (?:\|(?P<default>.*?))?
1088 $'''.format(field
=FIELD_RE
, math_op
=MATH_OPERATORS_RE
, math_field
=MATH_FIELD_RE
))
1090 def _traverse_infodict(k
):
1094 return traverse_obj(info_dict
, k
, is_user_input
=True, traverse_string
=True)
1096 def get_value(mdict
):
1098 value
= _traverse_infodict(mdict
['fields'])
1101 value
= float_or_none(value
)
1102 if value
is not None:
1105 offset_key
= mdict
['maths']
1107 value
= float_or_none(value
)
1111 MATH_FIELD_RE
if operator
else MATH_OPERATORS_RE
,
1112 offset_key
).group(0)
1113 offset_key
= offset_key
[len(item
):]
1114 if operator
is None:
1115 operator
= MATH_FUNCTIONS
[item
]
1117 item
, multiplier
= (item
[1:], -1) if item
[0] == '-' else (item
, 1)
1118 offset
= float_or_none(item
)
1120 offset
= float_or_none(_traverse_infodict(item
))
1122 value
= operator(value
, multiplier
* offset
)
1123 except (TypeError, ZeroDivisionError):
1126 # Datetime formatting
1127 if mdict
['strf_format']:
1128 value
= strftime_or_none(value
, mdict
['strf_format'].replace('\\,', ','))
1132 na
= self
.params
.get('outtmpl_na_placeholder', 'NA')
1134 def filename_sanitizer(key
, value
, restricted
=self
.params
.get('restrictfilenames')):
1135 return sanitize_filename(str(value
), restricted
=restricted
,
1136 is_id
=re
.search(r
'(^|[_.])id(\.|$)', key
))
1138 sanitizer
= sanitize
if callable(sanitize
) else filename_sanitizer
1139 sanitize
= bool(sanitize
)
1141 def _dumpjson_default(obj
):
1142 if isinstance(obj
, (set, LazyList
)):
1146 def create_key(outer_mobj
):
1147 if not outer_mobj
.group('has_key'):
1148 return outer_mobj
.group(0)
1149 key
= outer_mobj
.group('key')
1150 mobj
= re
.match(INTERNAL_FORMAT_RE
, key
)
1151 initial_field
= mobj
.group('fields') if mobj
else ''
1152 value
, replacement
, default
= None, None, na
1154 mobj
= mobj
.groupdict()
1155 default
= mobj
['default'] if mobj
['default'] is not None else default
1156 value
= get_value(mobj
)
1157 replacement
= mobj
['replacement']
1158 if value
is None and mobj
['alternate']:
1159 mobj
= re
.match(INTERNAL_FORMAT_RE
, mobj
['alternate'][1:])
1163 fmt
= outer_mobj
.group('format')
1164 if fmt
== 's' and value
is not None and key
in field_size_compat_map
.keys():
1165 fmt
= '0{:d}d'.format(field_size_compat_map
[key
])
1167 value
= default
if value
is None else value
if replacement
is None else replacement
1169 flags
= outer_mobj
.group('conversion') or ''
1170 str_fmt
= f
'{fmt[:-1]}s'
1171 if fmt
[-1] == 'l': # list
1172 delim
= '\n' if '#' in flags
else ', '
1173 value
, fmt
= delim
.join(map(str, variadic(value
, allowed_types
=(str, bytes)))), str_fmt
1174 elif fmt
[-1] == 'j': # json
1175 value
, fmt
= json
.dumps(value
, default
=_dumpjson_default
, indent
=4 if '#' in flags
else None), str_fmt
1176 elif fmt
[-1] == 'q': # quoted
1177 value
= map(str, variadic(value
) if '#' in flags
else [value
])
1178 value
, fmt
= ' '.join(map(compat_shlex_quote
, value
)), str_fmt
1179 elif fmt
[-1] == 'B': # bytes
1180 value
= f
'%{str_fmt}'.encode('utf-8') % str(value
).encode('utf-8')
1181 value
, fmt
= value
.decode('utf-8', 'ignore'), 's'
1182 elif fmt
[-1] == 'U': # unicode normalized
1183 value
, fmt
= unicodedata
.normalize(
1184 # "+" = compatibility equivalence, "#" = NFD
1185 'NF%s%s' % ('K' if '+' in flags
else '', 'D' if '#' in flags
else 'C'),
1187 elif fmt
[-1] == 'D': # decimal suffix
1188 num_fmt
, fmt
= fmt
[:-1].replace('#', ''), 's'
1189 value
= format_decimal_suffix(value
, f
'%{num_fmt}f%s' if num_fmt
else '%d%s',
1190 factor
=1024 if '#' in flags
else 1000)
1191 elif fmt
[-1] == 'S': # filename sanitization
1192 value
, fmt
= filename_sanitizer(initial_field
, value
, restricted
='#' in flags
), str_fmt
1193 elif fmt
[-1] == 'c':
1195 value
= str(value
)[0]
1198 elif fmt
[-1] not in 'rs': # numeric
1199 value
= float_or_none(value
)
1201 value
, fmt
= default
, 's'
1205 # If value is an object, sanitize might convert it to a string
1206 # So we convert it to repr first
1207 value
, fmt
= repr(value
), str_fmt
1208 if fmt
[-1] in 'csr':
1209 value
= sanitizer(initial_field
, value
)
1211 key
= '%s\0%s' % (key
.replace('%', '%\0'), outer_mobj
.group('format'))
1212 TMPL_DICT
[key
] = value
1213 return '{prefix}%({key}){fmt}'.format(key
=key
, fmt
=fmt
, prefix
=outer_mobj
.group('prefix'))
1215 return EXTERNAL_FORMAT_RE
.sub(create_key
, outtmpl
), TMPL_DICT
1217 def evaluate_outtmpl(self
, outtmpl
, info_dict
, *args
, **kwargs
):
1218 outtmpl
, info_dict
= self
.prepare_outtmpl(outtmpl
, info_dict
, *args
, **kwargs
)
1219 return self
.escape_outtmpl(outtmpl
) % info_dict
1221 def _prepare_filename(self
, info_dict
, tmpl_type
='default'):
1223 outtmpl
= self
._outtmpl
_expandpath
(self
.outtmpl_dict
.get(tmpl_type
, self
.outtmpl_dict
['default']))
1224 filename
= self
.evaluate_outtmpl(outtmpl
, info_dict
, True)
1228 if tmpl_type
in ('default', 'temp'):
1229 final_ext
, ext
= self
.params
.get('final_ext'), info_dict
.get('ext')
1230 if final_ext
and ext
and final_ext
!= ext
and filename
.endswith(f
'.{final_ext}'):
1231 filename
= replace_extension(filename
, ext
, final_ext
)
1233 force_ext
= OUTTMPL_TYPES
[tmpl_type
]
1235 filename
= replace_extension(filename
, force_ext
, info_dict
.get('ext'))
1237 # https://github.com/blackjack4494/youtube-dlc/issues/85
1238 trim_file_name
= self
.params
.get('trim_file_name', False)
1240 no_ext
, *ext
= filename
.rsplit('.', 2)
1241 filename
= join_nonempty(no_ext
[:trim_file_name
], *ext
, delim
='.')
1244 except ValueError as err
:
1245 self
.report_error('Error in output template: ' + str(err
) + ' (encoding: ' + repr(preferredencoding()) + ')')
1248 def prepare_filename(self
, info_dict
, dir_type
='', warn
=False):
1249 """Generate the output filename."""
1251 filename
= self
._prepare
_filename
(info_dict
, dir_type
or 'default')
1252 if not filename
and dir_type
not in ('', 'temp'):
1256 if not self
.params
.get('paths'):
1258 elif filename
== '-':
1259 self
.report_warning('--paths is ignored when an outputting to stdout', only_once
=True)
1260 elif os
.path
.isabs(filename
):
1261 self
.report_warning('--paths is ignored since an absolute path is given in output template', only_once
=True)
1262 if filename
== '-' or not filename
:
1265 return self
.get_output_path(dir_type
, filename
)
1267 def _match_entry(self
, info_dict
, incomplete
=False, silent
=False):
1268 """ Returns None if the file should be downloaded """
1270 video_title
= info_dict
.get('title', info_dict
.get('id', 'video'))
1273 if 'title' in info_dict
:
1274 # This can happen when we're just evaluating the playlist
1275 title
= info_dict
['title']
1276 matchtitle
= self
.params
.get('matchtitle', False)
1278 if not re
.search(matchtitle
, title
, re
.IGNORECASE
):
1279 return '"' + title
+ '" title did not match pattern "' + matchtitle
+ '"'
1280 rejecttitle
= self
.params
.get('rejecttitle', False)
1282 if re
.search(rejecttitle
, title
, re
.IGNORECASE
):
1283 return '"' + title
+ '" title matched reject pattern "' + rejecttitle
+ '"'
1284 date
= info_dict
.get('upload_date')
1285 if date
is not None:
1286 dateRange
= self
.params
.get('daterange', DateRange())
1287 if date
not in dateRange
:
1288 return '%s upload date is not in range %s' % (date_from_str(date
).isoformat(), dateRange
)
1289 view_count
= info_dict
.get('view_count')
1290 if view_count
is not None:
1291 min_views
= self
.params
.get('min_views')
1292 if min_views
is not None and view_count
< min_views
:
1293 return 'Skipping %s, because it has not reached minimum view count (%d/%d)' % (video_title
, view_count
, min_views
)
1294 max_views
= self
.params
.get('max_views')
1295 if max_views
is not None and view_count
> max_views
:
1296 return 'Skipping %s, because it has exceeded the maximum view count (%d/%d)' % (video_title
, view_count
, max_views
)
1297 if age_restricted(info_dict
.get('age_limit'), self
.params
.get('age_limit')):
1298 return 'Skipping "%s" because it is age restricted' % video_title
1300 match_filter
= self
.params
.get('match_filter')
1301 if match_filter
is not None:
1303 ret
= match_filter(info_dict
, incomplete
=incomplete
)
1305 # For backward compatibility
1306 ret
= None if incomplete
else match_filter(info_dict
)
1311 if self
.in_download_archive(info_dict
):
1312 reason
= '%s has already been recorded in the archive' % video_title
1313 break_opt
, break_err
= 'break_on_existing', ExistingVideoReached
1315 reason
= check_filter()
1316 break_opt
, break_err
= 'break_on_reject', RejectedVideoReached
1317 if reason
is not None:
1319 self
.to_screen('[download] ' + reason
)
1320 if self
.params
.get(break_opt
, False):
1325 def add_extra_info(info_dict
, extra_info
):
1326 '''Set the keys from extra_info in info dict if they are missing'''
1327 for key
, value
in extra_info
.items():
1328 info_dict
.setdefault(key
, value
)
1330 def extract_info(self
, url
, download
=True, ie_key
=None, extra_info
=None,
1331 process
=True, force_generic_extractor
=False):
1333 Return a list with a dictionary for each video extracted.
1336 url -- URL to extract
1339 download -- whether to download videos during extraction
1340 ie_key -- extractor key hint
1341 extra_info -- dictionary containing the extra values to add to each result
1342 process -- whether to resolve all unresolved references (URLs, playlist items),
1343 must be True for download to work.
1344 force_generic_extractor -- force using the generic extractor
1347 if extra_info
is None:
1350 if not ie_key
and force_generic_extractor
:
1354 ies
= {ie_key: self._get_info_extractor_class(ie_key)}
1358 for ie_key
, ie
in ies
.items():
1359 if not ie
.suitable(url
):
1362 if not ie
.working():
1363 self
.report_warning('The program functionality for this site has been marked as broken, '
1364 'and will probably not work.')
1366 temp_id
= ie
.get_temp_id(url
)
1367 if temp_id
is not None and self
.in_download_archive({'id': temp_id, 'ie_key': ie_key}
):
1368 self
.to_screen(f
'[{ie_key}] {temp_id}: has already been recorded in the archive')
1369 if self
.params
.get('break_on_existing', False):
1370 raise ExistingVideoReached()
1372 return self
.__extract
_info
(url
, self
.get_info_extractor(ie_key
), download
, extra_info
, process
)
1374 self
.report_error('no suitable InfoExtractor for URL %s' % url
)
1376 def __handle_extraction_exceptions(func
):
1377 @functools.wraps(func
)
1378 def wrapper(self
, *args
, **kwargs
):
1381 return func(self
, *args
, **kwargs
)
1382 except (DownloadCancelled
, LazyList
.IndexError, PagedList
.IndexError):
1384 except ReExtractInfo
as e
:
1386 self
.to_screen(f
'{e}; Re-extracting data')
1388 self
.to_stderr('\r')
1389 self
.report_warning(f
'{e}; Re-extracting data')
1391 except GeoRestrictedError
as e
:
1394 msg
+= '\nThis video is available in %s.' % ', '.join(
1395 map(ISO3166Utils
.short2full
, e
.countries
))
1396 msg
+= '\nYou might want to use a VPN or a proxy server (with --proxy) to workaround.'
1397 self
.report_error(msg
)
1398 except ExtractorError
as e
: # An error we somewhat expected
1399 self
.report_error(str(e
), e
.format_traceback())
1400 except Exception as e
:
1401 if self
.params
.get('ignoreerrors'):
1402 self
.report_error(str(e
), tb
=encode_compat_str(traceback
.format_exc()))
1408 def _wait_for_video(self
, ie_result
):
1409 if (not self
.params
.get('wait_for_video')
1410 or ie_result
.get('_type', 'video') != 'video'
1411 or ie_result
.get('formats') or ie_result
.get('url')):
1414 format_dur
= lambda dur
: '%02d:%02d:%02d' % timetuple_from_msec(dur
* 1000)[:-1]
1419 self
.to_screen(msg
+ ' ' * (len(last_msg
) - len(msg
)) + '\r', skip_eol
=True)
1422 min_wait
, max_wait
= self
.params
.get('wait_for_video')
1423 diff
= try_get(ie_result
, lambda x
: x
['release_timestamp'] - time
.time())
1424 if diff
is None and ie_result
.get('live_status') == 'is_upcoming':
1425 diff
= random
.randrange(min_wait
, max_wait
) if (max_wait
and min_wait
) else (max_wait
or min_wait
)
1426 self
.report_warning('Release time of video is not known')
1427 elif (diff
or 0) <= 0:
1428 self
.report_warning('Video should already be available according to extracted info')
1429 diff
= min(max(diff
or 0, min_wait
or 0), max_wait
or float('inf'))
1430 self
.to_screen(f
'[wait] Waiting for {format_dur(diff)} - Press Ctrl+C to try now')
1432 wait_till
= time
.time() + diff
1435 diff
= wait_till
- time
.time()
1438 raise ReExtractInfo('[wait] Wait period ended', expected
=True)
1439 progress(f
'[wait] Remaining time until next attempt: {self._format_screen(format_dur(diff), self.Styles.EMPHASIS)}')
1441 except KeyboardInterrupt:
1443 raise ReExtractInfo('[wait] Interrupted by user', expected
=True)
1444 except BaseException
as e
:
1445 if not isinstance(e
, ReExtractInfo
):
1449 @__handle_extraction_exceptions
1450 def __extract_info(self
, url
, ie
, download
, extra_info
, process
):
1451 ie_result
= ie
.extract(url
)
1452 if ie_result
is None: # Finished already (backwards compatibility; listformats and friends should be moved here)
1454 if isinstance(ie_result
, list):
1455 # Backwards compatibility: old IE result format
1457 '_type': 'compat_list',
1458 'entries': ie_result
,
1460 if extra_info
.get('original_url'):
1461 ie_result
.setdefault('original_url', extra_info
['original_url'])
1462 self
.add_default_extra_info(ie_result
, ie
, url
)
1464 self
._wait
_for
_video
(ie_result
)
1465 return self
.process_ie_result(ie_result
, download
, extra_info
)
1469 def add_default_extra_info(self
, ie_result
, ie
, url
):
1471 self
.add_extra_info(ie_result
, {
1473 'original_url': url
,
1474 'webpage_url_basename': url_basename(url
),
1475 'webpage_url_domain': get_domain(url
),
1478 self
.add_extra_info(ie_result
, {
1479 'extractor': ie
.IE_NAME
,
1480 'extractor_key': ie
.ie_key(),
1483 def process_ie_result(self
, ie_result
, download
=True, extra_info
=None):
1485 Take the result of the ie(may be modified) and resolve all unresolved
1486 references (URLs, playlist items).
1488 It will also download the videos if 'download'.
1489 Returns the resolved ie_result.
1491 if extra_info
is None:
1493 result_type
= ie_result
.get('_type', 'video')
1495 if result_type
in ('url', 'url_transparent'):
1496 ie_result
['url'] = sanitize_url(ie_result
['url'])
1497 if ie_result
.get('original_url'):
1498 extra_info
.setdefault('original_url', ie_result
['original_url'])
1500 extract_flat
= self
.params
.get('extract_flat', False)
1501 if ((extract_flat
== 'in_playlist' and 'playlist' in extra_info
)
1502 or extract_flat
is True):
1503 info_copy
= ie_result
.copy()
1504 ie
= try_get(ie_result
.get('ie_key'), self
.get_info_extractor
)
1505 if ie
and not ie_result
.get('id'):
1506 info_copy
['id'] = ie
.get_temp_id(ie_result
['url'])
1507 self
.add_default_extra_info(info_copy
, ie
, ie_result
['url'])
1508 self
.add_extra_info(info_copy
, extra_info
)
1509 info_copy
, _
= self
.pre_process(info_copy
)
1510 self
.__forced
_printings
(info_copy
, self
.prepare_filename(info_copy
), incomplete
=True)
1511 if self
.params
.get('force_write_download_archive', False):
1512 self
.record_download_archive(info_copy
)
1515 if result_type
== 'video':
1516 self
.add_extra_info(ie_result
, extra_info
)
1517 ie_result
= self
.process_video_result(ie_result
, download
=download
)
1518 additional_urls
= (ie_result
or {}).get('additional_urls')
1520 # TODO: Improve MetadataParserPP to allow setting a list
1521 if isinstance(additional_urls
, compat_str
):
1522 additional_urls
= [additional_urls
]
1524 '[info] %s: %d additional URL(s) requested' % (ie_result
['id'], len(additional_urls
)))
1525 self
.write_debug('Additional URLs: "%s"' % '", "'.join(additional_urls
))
1526 ie_result
['additional_entries'] = [
1528 url
, download
, extra_info
=extra_info
,
1529 force_generic_extractor
=self
.params
.get('force_generic_extractor'))
1530 for url
in additional_urls
1533 elif result_type
== 'url':
1534 # We have to add extra_info to the results because it may be
1535 # contained in a playlist
1536 return self
.extract_info(
1537 ie_result
['url'], download
,
1538 ie_key
=ie_result
.get('ie_key'),
1539 extra_info
=extra_info
)
1540 elif result_type
== 'url_transparent':
1541 # Use the information from the embedding page
1542 info
= self
.extract_info(
1543 ie_result
['url'], ie_key
=ie_result
.get('ie_key'),
1544 extra_info
=extra_info
, download
=False, process
=False)
1546 # extract_info may return None when ignoreerrors is enabled and
1547 # extraction failed with an error, don't crash and return early
1552 force_properties
= dict(
1553 (k
, v
) for k
, v
in ie_result
.items() if v
is not None)
1554 for f
in ('_type', 'url', 'id', 'extractor', 'extractor_key', 'ie_key'):
1555 if f
in force_properties
:
1556 del force_properties
[f
]
1557 new_result
= info
.copy()
1558 new_result
.update(force_properties
)
1560 # Extracted info may not be a video result (i.e.
1561 # info.get('_type', 'video') != video) but rather an url or
1562 # url_transparent. In such cases outer metadata (from ie_result)
1563 # should be propagated to inner one (info). For this to happen
1564 # _type of info should be overridden with url_transparent. This
1565 # fixes issue from https://github.com/ytdl-org/youtube-dl/pull/11163.
1566 if new_result
.get('_type') == 'url':
1567 new_result
['_type'] = 'url_transparent'
1569 return self
.process_ie_result(
1570 new_result
, download
=download
, extra_info
=extra_info
)
1571 elif result_type
in ('playlist', 'multi_video'):
1572 # Protect from infinite recursion due to recursively nested playlists
1573 # (see https://github.com/ytdl-org/youtube-dl/issues/27833)
1574 webpage_url
= ie_result
['webpage_url']
1575 if webpage_url
in self
._playlist
_urls
:
1577 '[download] Skipping already downloaded playlist: %s'
1578 % ie_result
.get('title') or ie_result
.get('id'))
1581 self
._playlist
_level
+= 1
1582 self
._playlist
_urls
.add(webpage_url
)
1583 self
._sanitize
_thumbnails
(ie_result
)
1585 return self
.__process
_playlist
(ie_result
, download
)
1587 self
._playlist
_level
-= 1
1588 if not self
._playlist
_level
:
1589 self
._playlist
_urls
.clear()
1590 elif result_type
== 'compat_list':
1591 self
.report_warning(
1592 'Extractor %s returned a compat_list result. '
1593 'It needs to be updated.' % ie_result
.get('extractor'))
1596 self
.add_extra_info(r
, {
1597 'extractor': ie_result
['extractor'],
1598 'webpage_url': ie_result
['webpage_url'],
1599 'webpage_url_basename': url_basename(ie_result
['webpage_url']),
1600 'webpage_url_domain': get_domain(ie_result
['webpage_url']),
1601 'extractor_key': ie_result
['extractor_key'],
1604 ie_result
['entries'] = [
1605 self
.process_ie_result(_fixup(r
), download
, extra_info
)
1606 for r
in ie_result
['entries']
1610 raise Exception('Invalid result type: %s' % result_type
)
1612 def _ensure_dir_exists(self
, path
):
1613 return make_dir(path
, self
.report_error
)
1616 def _playlist_infodict(ie_result
, **kwargs
):
1619 'playlist': ie_result
.get('title') or ie_result
.get('id'),
1620 'playlist_id': ie_result
.get('id'),
1621 'playlist_title': ie_result
.get('title'),
1622 'playlist_uploader': ie_result
.get('uploader'),
1623 'playlist_uploader_id': ie_result
.get('uploader_id'),
1624 'playlist_index': 0,
1628 def __process_playlist(self
, ie_result
, download
):
1629 # We process each entry in the playlist
1630 playlist
= ie_result
.get('title') or ie_result
.get('id')
1631 self
.to_screen('[download] Downloading playlist: %s' % playlist
)
1633 if 'entries' not in ie_result
:
1634 raise EntryNotInPlaylist('There are no entries')
1636 MissingEntry
= object()
1637 incomplete_entries
= bool(ie_result
.get('requested_entries'))
1638 if incomplete_entries
:
1639 def fill_missing_entries(entries
, indices
):
1640 ret
= [MissingEntry
] * max(indices
)
1641 for i
, entry
in zip(indices
, entries
):
1644 ie_result
['entries'] = fill_missing_entries(ie_result
['entries'], ie_result
['requested_entries'])
1646 playlist_results
= []
1648 playliststart
= self
.params
.get('playliststart', 1)
1649 playlistend
= self
.params
.get('playlistend')
1650 # For backwards compatibility, interpret -1 as whole list
1651 if playlistend
== -1:
1654 playlistitems_str
= self
.params
.get('playlist_items')
1655 playlistitems
= None
1656 if playlistitems_str
is not None:
1657 def iter_playlistitems(format
):
1658 for string_segment
in format
.split(','):
1659 if '-' in string_segment
:
1660 start
, end
= string_segment
.split('-')
1661 for item
in range(int(start
), int(end
) + 1):
1664 yield int(string_segment
)
1665 playlistitems
= orderedSet(iter_playlistitems(playlistitems_str
))
1667 ie_entries
= ie_result
['entries']
1668 if isinstance(ie_entries
, list):
1669 playlist_count
= len(ie_entries
)
1670 msg
= f
'Collected {playlist_count} videos; downloading %d of them'
1671 ie_result
['playlist_count'] = ie_result
.get('playlist_count') or playlist_count
1674 return ie_entries
[i
- 1]
1676 msg
= 'Downloading %d videos'
1677 if not isinstance(ie_entries
, (PagedList
, LazyList
)):
1678 ie_entries
= LazyList(ie_entries
)
1679 elif isinstance(ie_entries
, InAdvancePagedList
):
1680 if ie_entries
._pagesize
== 1:
1681 playlist_count
= ie_entries
._pagecount
1684 return YoutubeDL
.__handle
_extraction
_exceptions
(
1685 lambda self
, i
: ie_entries
[i
- 1]
1688 entries
, broken
= [], False
1689 items
= playlistitems
if playlistitems
is not None else itertools
.count(playliststart
)
1693 if playlistitems
is None and playlistend
is not None and playlistend
< i
:
1697 entry
= get_entry(i
)
1698 if entry
is MissingEntry
:
1699 raise EntryNotInPlaylist()
1700 except (IndexError, EntryNotInPlaylist
):
1701 if incomplete_entries
:
1702 raise EntryNotInPlaylist(f
'Entry {i} cannot be found')
1703 elif not playlistitems
:
1705 entries
.append(entry
)
1707 if entry
is not None:
1708 self
._match
_entry
(entry
, incomplete
=True, silent
=True)
1709 except (ExistingVideoReached
, RejectedVideoReached
):
1712 ie_result
['entries'] = entries
1714 # Save playlist_index before re-ordering
1716 ((playlistitems
[i
- 1] if playlistitems
else i
+ playliststart
- 1), entry
)
1717 for i
, entry
in enumerate(entries
, 1)
1718 if entry
is not None]
1719 n_entries
= len(entries
)
1721 if not (ie_result
.get('playlist_count') or broken
or playlistitems
or playlistend
):
1722 ie_result
['playlist_count'] = n_entries
1724 if not playlistitems
and (playliststart
!= 1 or playlistend
):
1725 playlistitems
= list(range(playliststart
, playliststart
+ n_entries
))
1726 ie_result
['requested_entries'] = playlistitems
1728 _infojson_written
= False
1729 write_playlist_files
= self
.params
.get('allow_playlist_files', True)
1730 if write_playlist_files
and self
.params
.get('list_thumbnails'):
1731 self
.list_thumbnails(ie_result
)
1732 if write_playlist_files
and not self
.params
.get('simulate'):
1733 ie_copy
= self
._playlist
_infodict
(ie_result
, n_entries
=n_entries
)
1734 _infojson_written
= self
._write
_info
_json
(
1735 'playlist', ie_result
, self
.prepare_filename(ie_copy
, 'pl_infojson'))
1736 if _infojson_written
is None:
1738 if self
._write
_description
('playlist', ie_result
,
1739 self
.prepare_filename(ie_copy
, 'pl_description')) is None:
1741 # TODO: This should be passed to ThumbnailsConvertor if necessary
1742 self
._write
_thumbnails
('playlist', ie_copy
, self
.prepare_filename(ie_copy
, 'pl_thumbnail'))
1744 if self
.params
.get('playlistreverse', False):
1745 entries
= entries
[::-1]
1746 if self
.params
.get('playlistrandom', False):
1747 random
.shuffle(entries
)
1749 x_forwarded_for
= ie_result
.get('__x_forwarded_for_ip')
1751 self
.to_screen('[%s] playlist %s: %s' % (ie_result
['extractor'], playlist
, msg
% n_entries
))
1753 max_failures
= self
.params
.get('skip_playlist_after_errors') or float('inf')
1754 for i
, entry_tuple
in enumerate(entries
, 1):
1755 playlist_index
, entry
= entry_tuple
1756 if 'playlist-index' in self
.params
.get('compat_opts', []):
1757 playlist_index
= playlistitems
[i
- 1] if playlistitems
else i
+ playliststart
- 1
1758 self
.to_screen('[download] Downloading video %s of %s' % (i
, n_entries
))
1759 # This __x_forwarded_for_ip thing is a bit ugly but requires
1762 entry
['__x_forwarded_for_ip'] = x_forwarded_for
1764 'n_entries': n_entries
,
1765 '_last_playlist_index': max(playlistitems
) if playlistitems
else (playlistend
or n_entries
),
1766 'playlist_count': ie_result
.get('playlist_count'),
1767 'playlist_index': playlist_index
,
1768 'playlist_autonumber': i
,
1769 'playlist': playlist
,
1770 'playlist_id': ie_result
.get('id'),
1771 'playlist_title': ie_result
.get('title'),
1772 'playlist_uploader': ie_result
.get('uploader'),
1773 'playlist_uploader_id': ie_result
.get('uploader_id'),
1774 'extractor': ie_result
['extractor'],
1775 'webpage_url': ie_result
['webpage_url'],
1776 'webpage_url_basename': url_basename(ie_result
['webpage_url']),
1777 'webpage_url_domain': get_domain(ie_result
['webpage_url']),
1778 'extractor_key': ie_result
['extractor_key'],
1781 if self
._match
_entry
(entry
, incomplete
=True) is not None:
1784 entry_result
= self
.__process
_iterable
_entry
(entry
, download
, extra
)
1785 if not entry_result
:
1787 if failures
>= max_failures
:
1789 'Skipping the remaining entries in playlist "%s" since %d items failed extraction' % (playlist
, failures
))
1791 playlist_results
.append(entry_result
)
1792 ie_result
['entries'] = playlist_results
1794 # Write the updated info to json
1795 if _infojson_written
and self
._write
_info
_json
(
1796 'updated playlist', ie_result
,
1797 self
.prepare_filename(ie_copy
, 'pl_infojson'), overwrite
=True) is None:
1800 ie_result
= self
.run_all_pps('playlist', ie_result
)
1801 self
.to_screen(f
'[download] Finished downloading playlist: {playlist}')
1804 @__handle_extraction_exceptions
1805 def __process_iterable_entry(self
, entry
, download
, extra_info
):
1806 return self
.process_ie_result(
1807 entry
, download
=download
, extra_info
=extra_info
)
1809 def _build_format_filter(self
, filter_spec
):
1810 " Returns a function to filter the formats according to the filter_spec "
1820 operator_rex
= re
.compile(r
'''(?x)\s*
1821 (?P<key>width|height|tbr|abr|vbr|asr|filesize|filesize_approx|fps)\s*
1822 (?P<op>%s)(?P<none_inclusive>\s*\?)?\s*
1823 (?P<value>[0-9.]+(?:[kKmMgGtTpPeEzZyY]i?[Bb]?)?)\s*
1824 ''' % '|'.join(map(re
.escape
, OPERATORS
.keys())))
1825 m
= operator_rex
.fullmatch(filter_spec
)
1828 comparison_value
= int(m
.group('value'))
1830 comparison_value
= parse_filesize(m
.group('value'))
1831 if comparison_value
is None:
1832 comparison_value
= parse_filesize(m
.group('value') + 'B')
1833 if comparison_value
is None:
1835 'Invalid value %r in format specification %r' % (
1836 m
.group('value'), filter_spec
))
1837 op
= OPERATORS
[m
.group('op')]
1842 '^=': lambda attr
, value
: attr
.startswith(value
),
1843 '$=': lambda attr
, value
: attr
.endswith(value
),
1844 '*=': lambda attr
, value
: value
in attr
,
1845 '~=': lambda attr
, value
: value
.search(attr
) is not None
1847 str_operator_rex
= re
.compile(r
'''(?x)\s*
1848 (?P<key>[a-zA-Z0-9._-]+)\s*
1849 (?P<negation>!\s*)?(?P<op>%s)\s*(?P<none_inclusive>\?\s*)?
1851 (?P<value>(?(quote)(?:(?!(?P=quote))[^\\]|\\.)+|[\w.-]+))
1852 (?(quote)(?P=quote))\s*
1853 ''' % '|'.join(map(re
.escape
, STR_OPERATORS
.keys())))
1854 m
= str_operator_rex
.fullmatch(filter_spec
)
1856 if m
.group('op') == '~=':
1857 comparison_value
= re
.compile(m
.group('value'))
1859 comparison_value
= re
.sub(r
'''\\([\\"'])''', r
'\1', m
.group('value'))
1860 str_op
= STR_OPERATORS
[m
.group('op')]
1861 if m
.group('negation'):
1862 op
= lambda attr
, value
: not str_op(attr
, value
)
1867 raise SyntaxError('Invalid filter specification %r' % filter_spec
)
1870 actual_value
= f
.get(m
.group('key'))
1871 if actual_value
is None:
1872 return m
.group('none_inclusive')
1873 return op(actual_value
, comparison_value
)
1876 def _check_formats(self
, formats
):
1878 self
.to_screen('[info] Testing format %s' % f
['format_id'])
1879 path
= self
.get_output_path('temp')
1880 if not self
._ensure
_dir
_exists
(f
'{path}/'):
1882 temp_file
= tempfile
.NamedTemporaryFile(suffix
='.tmp', delete
=False, dir=path
or None)
1885 success
, _
= self
.dl(temp_file
.name
, f
, test
=True)
1886 except (DownloadError
, IOError, OSError, ValueError) + network_exceptions
:
1889 if os
.path
.exists(temp_file
.name
):
1891 os
.remove(temp_file
.name
)
1893 self
.report_warning('Unable to delete temporary file "%s"' % temp_file
.name
)
1897 self
.to_screen('[info] Unable to download format %s. Skipping...' % f
['format_id'])
1899 def _default_format_spec(self
, info_dict
, download
=True):
1902 merger
= FFmpegMergerPP(self
)
1903 return merger
.available
and merger
.can_merge()
1906 not self
.params
.get('simulate')
1910 or info_dict
.get('is_live', False)
1911 or self
.outtmpl_dict
['default'] == '-'))
1914 or self
.params
.get('allow_multiple_audio_streams', False)
1915 or 'format-spec' in self
.params
.get('compat_opts', []))
1918 'best/bestvideo+bestaudio' if prefer_best
1919 else 'bestvideo*+bestaudio/best' if not compat
1920 else 'bestvideo+bestaudio/best')
1922 def build_format_selector(self
, format_spec
):
1923 def syntax_error(note
, start
):
1925 'Invalid format specification: '
1926 '{0}\n\t{1}\n\t{2}^'.format(note
, format_spec
, ' ' * start
[1]))
1927 return SyntaxError(message
)
1929 PICKFIRST
= 'PICKFIRST'
1933 FormatSelector
= collections
.namedtuple('FormatSelector', ['type', 'selector', 'filters'])
1935 allow_multiple_streams
= {'audio': self
.params
.get('allow_multiple_audio_streams', False),
1936 'video': self
.params
.get('allow_multiple_video_streams', False)}
1938 check_formats
= self
.params
.get('check_formats') == 'selected'
1940 def _parse_filter(tokens
):
1942 for type, string
, start
, _
, _
in tokens
:
1943 if type == tokenize
.OP
and string
== ']':
1944 return ''.join(filter_parts
)
1946 filter_parts
.append(string
)
1948 def _remove_unused_ops(tokens
):
1949 # Remove operators that we don't use and join them with the surrounding strings
1950 # for example: 'mp4' '-' 'baseline' '-' '16x9' is converted to 'mp4-baseline-16x9'
1951 ALLOWED_OPS
= ('/', '+', ',', '(', ')')
1952 last_string
, last_start
, last_end
, last_line
= None, None, None, None
1953 for type, string
, start
, end
, line
in tokens
:
1954 if type == tokenize
.OP
and string
== '[':
1956 yield tokenize
.NAME
, last_string
, last_start
, last_end
, last_line
1958 yield type, string
, start
, end
, line
1959 # everything inside brackets will be handled by _parse_filter
1960 for type, string
, start
, end
, line
in tokens
:
1961 yield type, string
, start
, end
, line
1962 if type == tokenize
.OP
and string
== ']':
1964 elif type == tokenize
.OP
and string
in ALLOWED_OPS
:
1966 yield tokenize
.NAME
, last_string
, last_start
, last_end
, last_line
1968 yield type, string
, start
, end
, line
1969 elif type in [tokenize
.NAME
, tokenize
.NUMBER
, tokenize
.OP
]:
1971 last_string
= string
1975 last_string
+= string
1977 yield tokenize
.NAME
, last_string
, last_start
, last_end
, last_line
1979 def _parse_format_selection(tokens
, inside_merge
=False, inside_choice
=False, inside_group
=False):
1981 current_selector
= None
1982 for type, string
, start
, _
, _
in tokens
:
1983 # ENCODING is only defined in python 3.x
1984 if type == getattr(tokenize
, 'ENCODING', None):
1986 elif type in [tokenize
.NAME
, tokenize
.NUMBER
]:
1987 current_selector
= FormatSelector(SINGLE
, string
, [])
1988 elif type == tokenize
.OP
:
1990 if not inside_group
:
1991 # ')' will be handled by the parentheses group
1992 tokens
.restore_last_token()
1994 elif inside_merge
and string
in ['/', ',']:
1995 tokens
.restore_last_token()
1997 elif inside_choice
and string
== ',':
1998 tokens
.restore_last_token()
2001 if not current_selector
:
2002 raise syntax_error('"," must follow a format selector', start
)
2003 selectors
.append(current_selector
)
2004 current_selector
= None
2006 if not current_selector
:
2007 raise syntax_error('"/" must follow a format selector', start
)
2008 first_choice
= current_selector
2009 second_choice
= _parse_format_selection(tokens
, inside_choice
=True)
2010 current_selector
= FormatSelector(PICKFIRST
, (first_choice
, second_choice
), [])
2012 if not current_selector
:
2013 current_selector
= FormatSelector(SINGLE
, 'best', [])
2014 format_filter
= _parse_filter(tokens
)
2015 current_selector
.filters
.append(format_filter
)
2017 if current_selector
:
2018 raise syntax_error('Unexpected "("', start
)
2019 group
= _parse_format_selection(tokens
, inside_group
=True)
2020 current_selector
= FormatSelector(GROUP
, group
, [])
2022 if not current_selector
:
2023 raise syntax_error('Unexpected "+"', start
)
2024 selector_1
= current_selector
2025 selector_2
= _parse_format_selection(tokens
, inside_merge
=True)
2027 raise syntax_error('Expected a selector', start
)
2028 current_selector
= FormatSelector(MERGE
, (selector_1
, selector_2
), [])
2030 raise syntax_error('Operator not recognized: "{0}"'.format(string
), start
)
2031 elif type == tokenize
.ENDMARKER
:
2033 if current_selector
:
2034 selectors
.append(current_selector
)
2037 def _merge(formats_pair
):
2038 format_1
, format_2
= formats_pair
2041 formats_info
.extend(format_1
.get('requested_formats', (format_1
,)))
2042 formats_info
.extend(format_2
.get('requested_formats', (format_2
,)))
2044 if not allow_multiple_streams
['video'] or not allow_multiple_streams
['audio']:
2045 get_no_more
= {'video': False, 'audio': False}
2046 for (i
, fmt_info
) in enumerate(formats_info
):
2047 if fmt_info
.get('acodec') == fmt_info
.get('vcodec') == 'none':
2050 for aud_vid
in ['audio', 'video']:
2051 if not allow_multiple_streams
[aud_vid
] and fmt_info
.get(aud_vid
[0] + 'codec') != 'none':
2052 if get_no_more
[aud_vid
]:
2055 get_no_more
[aud_vid
] = True
2057 if len(formats_info
) == 1:
2058 return formats_info
[0]
2060 video_fmts
= [fmt_info
for fmt_info
in formats_info
if fmt_info
.get('vcodec') != 'none']
2061 audio_fmts
= [fmt_info
for fmt_info
in formats_info
if fmt_info
.get('acodec') != 'none']
2063 the_only_video
= video_fmts
[0] if len(video_fmts
) == 1 else None
2064 the_only_audio
= audio_fmts
[0] if len(audio_fmts
) == 1 else None
2066 output_ext
= self
.params
.get('merge_output_format')
2069 output_ext
= the_only_video
['ext']
2070 elif the_only_audio
and not video_fmts
:
2071 output_ext
= the_only_audio
['ext']
2075 filtered
= lambda *keys
: filter(None, (traverse_obj(fmt
, *keys
) for fmt
in formats_info
))
2078 'requested_formats': formats_info
,
2079 'format': '+'.join(filtered('format')),
2080 'format_id': '+'.join(filtered('format_id')),
2082 'protocol': '+'.join(map(determine_protocol
, formats_info
)),
2083 'language': '+'.join(orderedSet(filtered('language'))) or None,
2084 'format_note': '+'.join(orderedSet(filtered('format_note'))) or None,
2085 'filesize_approx': sum(filtered('filesize', 'filesize_approx')) or None,
2086 'tbr': sum(filtered('tbr', 'vbr', 'abr')),
2091 'width': the_only_video
.get('width'),
2092 'height': the_only_video
.get('height'),
2093 'resolution': the_only_video
.get('resolution') or self
.format_resolution(the_only_video
),
2094 'fps': the_only_video
.get('fps'),
2095 'dynamic_range': the_only_video
.get('dynamic_range'),
2096 'vcodec': the_only_video
.get('vcodec'),
2097 'vbr': the_only_video
.get('vbr'),
2098 'stretched_ratio': the_only_video
.get('stretched_ratio'),
2103 'acodec': the_only_audio
.get('acodec'),
2104 'abr': the_only_audio
.get('abr'),
2105 'asr': the_only_audio
.get('asr'),
2110 def _check_formats(formats
):
2111 if not check_formats
:
2114 yield from self
._check
_formats
(formats
)
2116 def _build_selector_function(selector
):
2117 if isinstance(selector
, list): # ,
2118 fs
= [_build_selector_function(s
) for s
in selector
]
2120 def selector_function(ctx
):
2123 return selector_function
2125 elif selector
.type == GROUP
: # ()
2126 selector_function
= _build_selector_function(selector
.selector
)
2128 elif selector
.type == PICKFIRST
: # /
2129 fs
= [_build_selector_function(s
) for s
in selector
.selector
]
2131 def selector_function(ctx
):
2133 picked_formats
= list(f(ctx
))
2135 return picked_formats
2138 elif selector
.type == MERGE
: # +
2139 selector_1
, selector_2
= map(_build_selector_function
, selector
.selector
)
2141 def selector_function(ctx
):
2142 for pair
in itertools
.product(selector_1(ctx
), selector_2(ctx
)):
2145 elif selector
.type == SINGLE
: # atom
2146 format_spec
= selector
.selector
or 'best'
2148 # TODO: Add allvideo, allaudio etc by generalizing the code with best/worst selector
2149 if format_spec
== 'all':
2150 def selector_function(ctx
):
2151 yield from _check_formats(ctx
['formats'][::-1])
2152 elif format_spec
== 'mergeall':
2153 def selector_function(ctx
):
2154 formats
= list(_check_formats(ctx
['formats']))
2157 merged_format
= formats
[-1]
2158 for f
in formats
[-2::-1]:
2159 merged_format
= _merge((merged_format
, f
))
2163 format_fallback
, format_reverse
, format_idx
= False, True, 1
2165 r
'(?P<bw>best|worst|b|w)(?P<type>video|audio|v|a)?(?P<mod>\*)?(?:\.(?P<n>[1-9]\d*))?$',
2167 if mobj
is not None:
2168 format_idx
= int_or_none(mobj
.group('n'), default
=1)
2169 format_reverse
= mobj
.group('bw')[0] == 'b'
2170 format_type
= (mobj
.group('type') or [None])[0]
2171 not_format_type
= {'v': 'a', 'a': 'v'}
.get(format_type
)
2172 format_modified
= mobj
.group('mod') is not None
2174 format_fallback
= not format_type
and not format_modified
# for b, w
2176 (lambda f
: f
.get('%scodec' % format_type
) != 'none')
2177 if format_type
and format_modified
# bv*, ba*, wv*, wa*
2178 else (lambda f
: f
.get('%scodec' % not_format_type
) == 'none')
2179 if format_type
# bv, ba, wv, wa
2180 else (lambda f
: f
.get('vcodec') != 'none' and f
.get('acodec') != 'none')
2181 if not format_modified
# b, w
2182 else lambda f
: True) # b*, w*
2183 filter_f
= lambda f
: _filter_f(f
) and (
2184 f
.get('vcodec') != 'none' or f
.get('acodec') != 'none')
2186 if format_spec
in self
._format
_selection
_exts
['audio']:
2187 filter_f
= lambda f
: f
.get('ext') == format_spec
and f
.get('acodec') != 'none'
2188 elif format_spec
in self
._format
_selection
_exts
['video']:
2189 filter_f
= lambda f
: f
.get('ext') == format_spec
and f
.get('acodec') != 'none' and f
.get('vcodec') != 'none'
2190 elif format_spec
in self
._format
_selection
_exts
['storyboards']:
2191 filter_f
= lambda f
: f
.get('ext') == format_spec
and f
.get('acodec') == 'none' and f
.get('vcodec') == 'none'
2193 filter_f
= lambda f
: f
.get('format_id') == format_spec
# id
2195 def selector_function(ctx
):
2196 formats
= list(ctx
['formats'])
2197 matches
= list(filter(filter_f
, formats
)) if filter_f
is not None else formats
2198 if format_fallback
and ctx
['incomplete_formats'] and not matches
:
2199 # for extractors with incomplete formats (audio only (soundcloud)
2200 # or video only (imgur)) best/worst will fallback to
2201 # best/worst {video,audio}-only format
2203 matches
= LazyList(_check_formats(matches
[::-1 if format_reverse
else 1]))
2205 yield matches
[format_idx
- 1]
2209 filters
= [self
._build
_format
_filter
(f
) for f
in selector
.filters
]
2211 def final_selector(ctx
):
2212 ctx_copy
= dict(ctx
)
2213 for _filter
in filters
:
2214 ctx_copy
['formats'] = list(filter(_filter
, ctx_copy
['formats']))
2215 return selector_function(ctx_copy
)
2216 return final_selector
2218 stream
= io
.BytesIO(format_spec
.encode('utf-8'))
2220 tokens
= list(_remove_unused_ops(compat_tokenize_tokenize(stream
.readline
)))
2221 except tokenize
.TokenError
:
2222 raise syntax_error('Missing closing/opening brackets or parenthesis', (0, len(format_spec
)))
2224 class TokenIterator(object):
2225 def __init__(self
, tokens
):
2226 self
.tokens
= tokens
2233 if self
.counter
>= len(self
.tokens
):
2234 raise StopIteration()
2235 value
= self
.tokens
[self
.counter
]
2241 def restore_last_token(self
):
2244 parsed_selector
= _parse_format_selection(iter(TokenIterator(tokens
)))
2245 return _build_selector_function(parsed_selector
)
2247 def _calc_headers(self
, info_dict
):
2248 res
= std_headers
.copy()
2249 res
.update(info_dict
.get('http_headers') or {})
2251 cookies
= self
._calc
_cookies
(info_dict
)
2253 res
['Cookie'] = cookies
2255 if 'X-Forwarded-For' not in res
:
2256 x_forwarded_for_ip
= info_dict
.get('__x_forwarded_for_ip')
2257 if x_forwarded_for_ip
:
2258 res
['X-Forwarded-For'] = x_forwarded_for_ip
2262 def _calc_cookies(self
, info_dict
):
2263 pr
= sanitized_Request(info_dict
['url'])
2264 self
.cookiejar
.add_cookie_header(pr
)
2265 return pr
.get_header('Cookie')
2267 def _sort_thumbnails(self
, thumbnails
):
2268 thumbnails
.sort(key
=lambda t
: (
2269 t
.get('preference') if t
.get('preference') is not None else -1,
2270 t
.get('width') if t
.get('width') is not None else -1,
2271 t
.get('height') if t
.get('height') is not None else -1,
2272 t
.get('id') if t
.get('id') is not None else '',
2275 def _sanitize_thumbnails(self
, info_dict
):
2276 thumbnails
= info_dict
.get('thumbnails')
2277 if thumbnails
is None:
2278 thumbnail
= info_dict
.get('thumbnail')
2280 info_dict
['thumbnails'] = thumbnails
= [{'url': thumbnail}
]
2284 def check_thumbnails(thumbnails
):
2285 for t
in thumbnails
:
2286 self
.to_screen(f
'[info] Testing thumbnail {t["id"]}')
2288 self
.urlopen(HEADRequest(t
['url']))
2289 except network_exceptions
as err
:
2290 self
.to_screen(f
'[info] Unable to connect to thumbnail {t["id"]} URL {t["url"]!r} - {err}. Skipping...')
2294 self
._sort
_thumbnails
(thumbnails
)
2295 for i
, t
in enumerate(thumbnails
):
2296 if t
.get('id') is None:
2298 if t
.get('width') and t
.get('height'):
2299 t
['resolution'] = '%dx%d' % (t
['width'], t
['height'])
2300 t
['url'] = sanitize_url(t
['url'])
2302 if self
.params
.get('check_formats') is True:
2303 info_dict
['thumbnails'] = LazyList(check_thumbnails(thumbnails
[::-1]), reverse
=True)
2305 info_dict
['thumbnails'] = thumbnails
2307 def process_video_result(self
, info_dict
, download
=True):
2308 assert info_dict
.get('_type', 'video') == 'video'
2309 self
._num
_videos
+= 1
2311 if 'id' not in info_dict
:
2312 raise ExtractorError('Missing "id" field in extractor result', ie
=info_dict
['extractor'])
2313 elif not info_dict
.get('id'):
2314 raise ExtractorError('Extractor failed to obtain "id"', ie
=info_dict
['extractor'])
2316 info_dict
['fulltitle'] = info_dict
.get('title')
2317 if 'title' not in info_dict
:
2318 raise ExtractorError('Missing "title" field in extractor result',
2319 video_id
=info_dict
['id'], ie
=info_dict
['extractor'])
2320 elif not info_dict
.get('title'):
2321 self
.report_warning('Extractor failed to obtain "title". Creating a generic title instead')
2322 info_dict
['title'] = f
'{info_dict["extractor"]} video #{info_dict["id"]}'
2324 def report_force_conversion(field
, field_not
, conversion
):
2325 self
.report_warning(
2326 '"%s" field is not %s - forcing %s conversion, there is an error in extractor'
2327 % (field
, field_not
, conversion
))
2329 def sanitize_string_field(info
, string_field
):
2330 field
= info
.get(string_field
)
2331 if field
is None or isinstance(field
, compat_str
):
2333 report_force_conversion(string_field
, 'a string', 'string')
2334 info
[string_field
] = compat_str(field
)
2336 def sanitize_numeric_fields(info
):
2337 for numeric_field
in self
._NUMERIC
_FIELDS
:
2338 field
= info
.get(numeric_field
)
2339 if field
is None or isinstance(field
, compat_numeric_types
):
2341 report_force_conversion(numeric_field
, 'numeric', 'int')
2342 info
[numeric_field
] = int_or_none(field
)
2344 sanitize_string_field(info_dict
, 'id')
2345 sanitize_numeric_fields(info_dict
)
2347 if 'playlist' not in info_dict
:
2348 # It isn't part of a playlist
2349 info_dict
['playlist'] = None
2350 info_dict
['playlist_index'] = None
2352 self
._sanitize
_thumbnails
(info_dict
)
2354 thumbnail
= info_dict
.get('thumbnail')
2355 thumbnails
= info_dict
.get('thumbnails')
2357 info_dict
['thumbnail'] = sanitize_url(thumbnail
)
2359 info_dict
['thumbnail'] = thumbnails
[-1]['url']
2361 if info_dict
.get('display_id') is None and 'id' in info_dict
:
2362 info_dict
['display_id'] = info_dict
['id']
2364 if info_dict
.get('duration') is not None:
2365 info_dict
['duration_string'] = formatSeconds(info_dict
['duration'])
2367 for ts_key
, date_key
in (
2368 ('timestamp', 'upload_date'),
2369 ('release_timestamp', 'release_date'),
2370 ('modified_timestamp', 'modified_date'),
2372 if info_dict
.get(date_key
) is None and info_dict
.get(ts_key
) is not None:
2373 # Working around out-of-range timestamp values (e.g. negative ones on Windows,
2374 # see http://bugs.python.org/issue1646728)
2376 upload_date
= datetime
.datetime
.utcfromtimestamp(info_dict
[ts_key
])
2377 info_dict
[date_key
] = upload_date
.strftime('%Y%m%d')
2378 except (ValueError, OverflowError, OSError):
2381 live_keys
= ('is_live', 'was_live')
2382 live_status
= info_dict
.get('live_status')
2383 if live_status
is None:
2384 for key
in live_keys
:
2385 if info_dict
.get(key
) is False:
2387 if info_dict
.get(key
):
2390 if all(info_dict
.get(key
) is False for key
in live_keys
):
2391 live_status
= 'not_live'
2393 info_dict
['live_status'] = live_status
2394 for key
in live_keys
:
2395 if info_dict
.get(key
) is None:
2396 info_dict
[key
] = (live_status
== key
)
2398 # Auto generate title fields corresponding to the *_number fields when missing
2399 # in order to always have clean titles. This is very common for TV series.
2400 for field
in ('chapter', 'season', 'episode'):
2401 if info_dict
.get('%s_number' % field
) is not None and not info_dict
.get(field
):
2402 info_dict
[field
] = '%s %d' % (field
.capitalize(), info_dict
['%s_number' % field
])
2404 for cc_kind
in ('subtitles', 'automatic_captions'):
2405 cc
= info_dict
.get(cc_kind
)
2407 for _
, subtitle
in cc
.items():
2408 for subtitle_format
in subtitle
:
2409 if subtitle_format
.get('url'):
2410 subtitle_format
['url'] = sanitize_url(subtitle_format
['url'])
2411 if subtitle_format
.get('ext') is None:
2412 subtitle_format
['ext'] = determine_ext(subtitle_format
['url']).lower()
2414 automatic_captions
= info_dict
.get('automatic_captions')
2415 subtitles
= info_dict
.get('subtitles')
2417 info_dict
['requested_subtitles'] = self
.process_subtitles(
2418 info_dict
['id'], subtitles
, automatic_captions
)
2420 if info_dict
.get('formats') is None:
2421 # There's only one format available
2422 formats
= [info_dict
]
2424 formats
= info_dict
['formats']
2426 info_dict
['__has_drm'] = any(f
.get('has_drm') for f
in formats
)
2427 if not self
.params
.get('allow_unplayable_formats'):
2428 formats
= [f
for f
in formats
if not f
.get('has_drm')]
2430 if info_dict
.get('is_live'):
2431 get_from_start
= bool(self
.params
.get('live_from_start'))
2432 formats
= [f
for f
in formats
if bool(f
.get('is_from_start')) == get_from_start
]
2433 if not get_from_start
:
2434 info_dict
['title'] += ' ' + datetime
.datetime
.now().strftime('%Y-%m-%d %H:%M')
2437 self
.raise_no_formats(info_dict
)
2439 def is_wellformed(f
):
2442 self
.report_warning(
2443 '"url" field is missing or empty - skipping format, '
2444 'there is an error in extractor')
2446 if isinstance(url
, bytes):
2447 sanitize_string_field(f
, 'url')
2450 # Filter out malformed formats for better extraction robustness
2451 formats
= list(filter(is_wellformed
, formats
))
2455 # We check that all the formats have the format and format_id fields
2456 for i
, format
in enumerate(formats
):
2457 sanitize_string_field(format
, 'format_id')
2458 sanitize_numeric_fields(format
)
2459 format
['url'] = sanitize_url(format
['url'])
2460 if not format
.get('format_id'):
2461 format
['format_id'] = compat_str(i
)
2463 # Sanitize format_id from characters used in format selector expression
2464 format
['format_id'] = re
.sub(r
'[\s,/+\[\]()]', '_', format
['format_id'])
2465 format_id
= format
['format_id']
2466 if format_id
not in formats_dict
:
2467 formats_dict
[format_id
] = []
2468 formats_dict
[format_id
].append(format
)
2470 # Make sure all formats have unique format_id
2471 common_exts
= set(itertools
.chain(*self
._format
_selection
_exts
.values()))
2472 for format_id
, ambiguous_formats
in formats_dict
.items():
2473 ambigious_id
= len(ambiguous_formats
) > 1
2474 for i
, format
in enumerate(ambiguous_formats
):
2476 format
['format_id'] = '%s-%d' % (format_id
, i
)
2477 if format
.get('ext') is None:
2478 format
['ext'] = determine_ext(format
['url']).lower()
2479 # Ensure there is no conflict between id and ext in format selection
2480 # See https://github.com/yt-dlp/yt-dlp/issues/1282
2481 if format
['format_id'] != format
['ext'] and format
['format_id'] in common_exts
:
2482 format
['format_id'] = 'f%s' % format
['format_id']
2484 for i
, format
in enumerate(formats
):
2485 if format
.get('format') is None:
2486 format
['format'] = '{id} - {res}{note}'.format(
2487 id=format
['format_id'],
2488 res
=self
.format_resolution(format
),
2489 note
=format_field(format
, 'format_note', ' (%s)'),
2491 if format
.get('protocol') is None:
2492 format
['protocol'] = determine_protocol(format
)
2493 if format
.get('resolution') is None:
2494 format
['resolution'] = self
.format_resolution(format
, default
=None)
2495 if format
.get('dynamic_range') is None and format
.get('vcodec') != 'none':
2496 format
['dynamic_range'] = 'SDR'
2497 if (info_dict
.get('duration') and format
.get('tbr')
2498 and not format
.get('filesize') and not format
.get('filesize_approx')):
2499 format
['filesize_approx'] = info_dict
['duration'] * format
['tbr'] * (1024 / 8)
2501 # Add HTTP headers, so that external programs can use them from the
2503 full_format_info
= info_dict
.copy()
2504 full_format_info
.update(format
)
2505 format
['http_headers'] = self
._calc
_headers
(full_format_info
)
2506 # Remove private housekeeping stuff
2507 if '__x_forwarded_for_ip' in info_dict
:
2508 del info_dict
['__x_forwarded_for_ip']
2510 # TODO Central sorting goes here
2512 if self
.params
.get('check_formats') is True:
2513 formats
= LazyList(self
._check
_formats
(formats
[::-1]), reverse
=True)
2515 if not formats
or formats
[0] is not info_dict
:
2516 # only set the 'formats' fields if the original info_dict list them
2517 # otherwise we end up with a circular reference, the first (and unique)
2518 # element in the 'formats' field in info_dict is info_dict itself,
2519 # which can't be exported to json
2520 info_dict
['formats'] = formats
2522 info_dict
, _
= self
.pre_process(info_dict
)
2524 # The pre-processors may have modified the formats
2525 formats
= info_dict
.get('formats', [info_dict
])
2527 list_only
= self
.params
.get('simulate') is None and (
2528 self
.params
.get('list_thumbnails') or self
.params
.get('listformats') or self
.params
.get('listsubtitles'))
2529 interactive_format_selection
= not list_only
and self
.format_selector
== '-'
2530 if self
.params
.get('list_thumbnails'):
2531 self
.list_thumbnails(info_dict
)
2532 if self
.params
.get('listsubtitles'):
2533 if 'automatic_captions' in info_dict
:
2534 self
.list_subtitles(
2535 info_dict
['id'], automatic_captions
, 'automatic captions')
2536 self
.list_subtitles(info_dict
['id'], subtitles
, 'subtitles')
2537 if self
.params
.get('listformats') or interactive_format_selection
:
2538 self
.list_formats(info_dict
)
2540 # Without this printing, -F --print-json will not work
2541 self
.__forced
_printings
(info_dict
, self
.prepare_filename(info_dict
), incomplete
=True)
2544 format_selector
= self
.format_selector
2545 if format_selector
is None:
2546 req_format
= self
._default
_format
_spec
(info_dict
, download
=download
)
2547 self
.write_debug('Default format spec: %s' % req_format
)
2548 format_selector
= self
.build_format_selector(req_format
)
2551 if interactive_format_selection
:
2553 self
._format
_screen
('\nEnter format selector: ', self
.Styles
.EMPHASIS
))
2555 format_selector
= self
.build_format_selector(req_format
)
2556 except SyntaxError as err
:
2557 self
.report_error(err
, tb
=False, is_error
=False)
2560 # While in format selection we may need to have an access to the original
2561 # format set in order to calculate some metrics or do some processing.
2562 # For now we need to be able to guess whether original formats provided
2563 # by extractor are incomplete or not (i.e. whether extractor provides only
2564 # video-only or audio-only formats) for proper formats selection for
2565 # extractors with such incomplete formats (see
2566 # https://github.com/ytdl-org/youtube-dl/pull/5556).
2567 # Since formats may be filtered during format selection and may not match
2568 # the original formats the results may be incorrect. Thus original formats
2569 # or pre-calculated metrics should be passed to format selection routines
2571 # We will pass a context object containing all necessary additional data
2572 # instead of just formats.
2573 # This fixes incorrect format selection issue (see
2574 # https://github.com/ytdl-org/youtube-dl/issues/10083).
2575 incomplete_formats
= (
2576 # All formats are video-only or
2577 all(f
.get('vcodec') != 'none' and f
.get('acodec') == 'none' for f
in formats
)
2578 # all formats are audio-only
2579 or all(f
.get('vcodec') == 'none' and f
.get('acodec') != 'none' for f
in formats
))
2583 'incomplete_formats': incomplete_formats
,
2586 formats_to_download
= list(format_selector(ctx
))
2587 if interactive_format_selection
and not formats_to_download
:
2588 self
.report_error('Requested format is not available', tb
=False, is_error
=False)
2592 if not formats_to_download
:
2593 if not self
.params
.get('ignore_no_formats_error'):
2594 raise ExtractorError('Requested format is not available', expected
=True,
2595 video_id
=info_dict
['id'], ie
=info_dict
['extractor'])
2596 self
.report_warning('Requested format is not available')
2597 # Process what we can, even without any available formats.
2598 formats_to_download
= [{}]
2600 best_format
= formats_to_download
[-1]
2604 f
'[info] {info_dict["id"]}: Downloading {len(formats_to_download)} format(s): '
2605 + ', '.join([f
['format_id'] for f
in formats_to_download
]))
2606 max_downloads_reached
= False
2607 for i
, fmt
in enumerate(formats_to_download
):
2608 formats_to_download
[i
] = new_info
= dict(info_dict
)
2609 # Save a reference to the original info_dict so that it can be modified in process_info if needed
2610 new_info
.update(fmt
)
2611 new_info
['__original_infodict'] = info_dict
2613 self
.process_info(new_info
)
2614 except MaxDownloadsReached
:
2615 max_downloads_reached
= True
2616 new_info
.pop('__original_infodict')
2617 # Remove copied info
2618 for key
, val
in tuple(new_info
.items()):
2619 if info_dict
.get(key
) == val
:
2621 if max_downloads_reached
:
2624 write_archive
= set(f
.get('__write_download_archive', False) for f
in formats_to_download
)
2625 assert write_archive
.issubset({True, False, 'ignore'}
)
2626 if True in write_archive
and False not in write_archive
:
2627 self
.record_download_archive(info_dict
)
2629 info_dict
['requested_downloads'] = formats_to_download
2630 info_dict
= self
.run_all_pps('after_video', info_dict
)
2631 if max_downloads_reached
:
2632 raise MaxDownloadsReached()
2634 # We update the info dict with the selected best quality format (backwards compatibility)
2635 info_dict
.update(best_format
)
2638 def process_subtitles(self
, video_id
, normal_subtitles
, automatic_captions
):
2639 """Select the requested subtitles and their format"""
2641 if normal_subtitles
and self
.params
.get('writesubtitles'):
2642 available_subs
.update(normal_subtitles
)
2643 if automatic_captions
and self
.params
.get('writeautomaticsub'):
2644 for lang
, cap_info
in automatic_captions
.items():
2645 if lang
not in available_subs
:
2646 available_subs
[lang
] = cap_info
2648 if (not self
.params
.get('writesubtitles') and not
2649 self
.params
.get('writeautomaticsub') or not
2653 all_sub_langs
= available_subs
.keys()
2654 if self
.params
.get('allsubtitles', False):
2655 requested_langs
= all_sub_langs
2656 elif self
.params
.get('subtitleslangs', False):
2657 # A list is used so that the order of languages will be the same as
2658 # given in subtitleslangs. See https://github.com/yt-dlp/yt-dlp/issues/1041
2659 requested_langs
= []
2660 for lang_re
in self
.params
.get('subtitleslangs'):
2661 if lang_re
== 'all':
2662 requested_langs
.extend(all_sub_langs
)
2664 discard
= lang_re
[0] == '-'
2666 lang_re
= lang_re
[1:]
2667 current_langs
= filter(re
.compile(lang_re
+ '$').match
, all_sub_langs
)
2669 for lang
in current_langs
:
2670 while lang
in requested_langs
:
2671 requested_langs
.remove(lang
)
2673 requested_langs
.extend(current_langs
)
2674 requested_langs
= orderedSet(requested_langs
)
2675 elif 'en' in available_subs
:
2676 requested_langs
= ['en']
2678 requested_langs
= [list(all_sub_langs
)[0]]
2680 self
.write_debug('Downloading subtitles: %s' % ', '.join(requested_langs
))
2682 formats_query
= self
.params
.get('subtitlesformat', 'best')
2683 formats_preference
= formats_query
.split('/') if formats_query
else []
2685 for lang
in requested_langs
:
2686 formats
= available_subs
.get(lang
)
2688 self
.report_warning('%s subtitles not available for %s' % (lang
, video_id
))
2690 for ext
in formats_preference
:
2694 matches
= list(filter(lambda f
: f
['ext'] == ext
, formats
))
2700 self
.report_warning(
2701 'No subtitle format found matching "%s" for language %s, '
2702 'using %s' % (formats_query
, lang
, f
['ext']))
2706 def _forceprint(self
, key
, info_dict
):
2707 if info_dict
is None:
2709 info_copy
= info_dict
.copy()
2710 info_copy
['formats_table'] = self
.render_formats_table(info_dict
)
2711 info_copy
['thumbnails_table'] = self
.render_thumbnails_table(info_dict
)
2712 info_copy
['subtitles_table'] = self
.render_subtitles_table(info_dict
.get('id'), info_dict
.get('subtitles'))
2713 info_copy
['automatic_captions_table'] = self
.render_subtitles_table(info_dict
.get('id'), info_dict
.get('automatic_captions'))
2715 def format_tmpl(tmpl
):
2716 mobj
= re
.match(r
'\w+(=?)$', tmpl
)
2717 if mobj
and mobj
.group(1):
2718 return f
'{tmpl[:-1]} = %({tmpl[:-1]})r'
2720 return f
'%({tmpl})s'
2723 for tmpl
in self
.params
['forceprint'].get(key
, []):
2724 self
.to_stdout(self
.evaluate_outtmpl(format_tmpl(tmpl
), info_copy
))
2726 for tmpl
, file_tmpl
in self
.params
['print_to_file'].get(key
, []):
2727 filename
= self
.evaluate_outtmpl(file_tmpl
, info_dict
)
2728 tmpl
= format_tmpl(tmpl
)
2729 self
.to_screen(f
'[info] Writing {tmpl!r} to: {filename}')
2730 with io
.open(filename
, 'a', encoding
='utf-8') as f
:
2731 f
.write(self
.evaluate_outtmpl(tmpl
, info_copy
) + '\n')
2733 def __forced_printings(self
, info_dict
, filename
, incomplete
):
2734 def print_mandatory(field
, actual_field
=None):
2735 if actual_field
is None:
2736 actual_field
= field
2737 if (self
.params
.get('force%s' % field
, False)
2738 and (not incomplete
or info_dict
.get(actual_field
) is not None)):
2739 self
.to_stdout(info_dict
[actual_field
])
2741 def print_optional(field
):
2742 if (self
.params
.get('force%s' % field
, False)
2743 and info_dict
.get(field
) is not None):
2744 self
.to_stdout(info_dict
[field
])
2746 info_dict
= info_dict
.copy()
2747 if filename
is not None:
2748 info_dict
['filename'] = filename
2749 if info_dict
.get('requested_formats') is not None:
2750 # For RTMP URLs, also include the playpath
2751 info_dict
['urls'] = '\n'.join(f
['url'] + f
.get('play_path', '') for f
in info_dict
['requested_formats'])
2752 elif 'url' in info_dict
:
2753 info_dict
['urls'] = info_dict
['url'] + info_dict
.get('play_path', '')
2755 if (self
.params
.get('forcejson')
2756 or self
.params
['forceprint'].get('video')
2757 or self
.params
['print_to_file'].get('video')):
2758 self
.post_extract(info_dict
)
2759 self
._forceprint
('video', info_dict
)
2761 print_mandatory('title')
2762 print_mandatory('id')
2763 print_mandatory('url', 'urls')
2764 print_optional('thumbnail')
2765 print_optional('description')
2766 print_optional('filename')
2767 if self
.params
.get('forceduration') and info_dict
.get('duration') is not None:
2768 self
.to_stdout(formatSeconds(info_dict
['duration']))
2769 print_mandatory('format')
2771 if self
.params
.get('forcejson'):
2772 self
.to_stdout(json
.dumps(self
.sanitize_info(info_dict
)))
2774 def dl(self
, name
, info
, subtitle
=False, test
=False):
2775 if not info
.get('url'):
2776 self
.raise_no_formats(info
, True)
2779 verbose
= self
.params
.get('verbose')
2782 'quiet': self
.params
.get('quiet') or not verbose
,
2784 'noprogress': not verbose
,
2786 'skip_unavailable_fragments': False,
2787 'keep_fragments': False,
2789 '_no_ytdl_file': True,
2792 params
= self
.params
2793 fd
= get_suitable_downloader(info
, params
, to_stdout
=(name
== '-'))(self
, params
)
2795 for ph
in self
._progress
_hooks
:
2796 fd
.add_progress_hook(ph
)
2798 (f
['url'].split(',')[0] + ',<data>' if f
['url'].startswith('data:') else f
['url'])
2799 for f
in info
.get('requested_formats', []) or [info
])
2800 self
.write_debug('Invoking downloader on "%s"' % urls
)
2802 # Note: Ideally info should be a deep-copied so that hooks cannot modify it.
2803 # But it may contain objects that are not deep-copyable
2804 new_info
= self
._copy
_infodict
(info
)
2805 if new_info
.get('http_headers') is None:
2806 new_info
['http_headers'] = self
._calc
_headers
(new_info
)
2807 return fd
.download(name
, new_info
, subtitle
)
2809 def existing_file(self
, filepaths
, *, default_overwrite
=True):
2810 existing_files
= list(filter(os
.path
.exists
, orderedSet(filepaths
)))
2811 if existing_files
and not self
.params
.get('overwrites', default_overwrite
):
2812 return existing_files
[0]
2814 for file in existing_files
:
2815 self
.report_file_delete(file)
2819 def process_info(self
, info_dict
):
2820 """Process a single resolved IE result. (Modified it in-place)"""
2822 assert info_dict
.get('_type', 'video') == 'video'
2823 original_infodict
= info_dict
2825 if 'format' not in info_dict
and 'ext' in info_dict
:
2826 info_dict
['format'] = info_dict
['ext']
2828 if self
._match
_entry
(info_dict
) is not None:
2829 info_dict
['__write_download_archive'] = 'ignore'
2832 self
.post_extract(info_dict
)
2833 self
._num
_downloads
+= 1
2835 # info_dict['_filename'] needs to be set for backward compatibility
2836 info_dict
['_filename'] = full_filename
= self
.prepare_filename(info_dict
, warn
=True)
2837 temp_filename
= self
.prepare_filename(info_dict
, 'temp')
2841 self
.__forced
_printings
(info_dict
, full_filename
, incomplete
=('format' not in info_dict
))
2843 if self
.params
.get('simulate'):
2844 info_dict
['__write_download_archive'] = self
.params
.get('force_write_download_archive')
2847 if full_filename
is None:
2849 if not self
._ensure
_dir
_exists
(encodeFilename(full_filename
)):
2851 if not self
._ensure
_dir
_exists
(encodeFilename(temp_filename
)):
2854 if self
._write
_description
('video', info_dict
,
2855 self
.prepare_filename(info_dict
, 'description')) is None:
2858 sub_files
= self
._write
_subtitles
(info_dict
, temp_filename
)
2859 if sub_files
is None:
2861 files_to_move
.update(dict(sub_files
))
2863 thumb_files
= self
._write
_thumbnails
(
2864 'video', info_dict
, temp_filename
, self
.prepare_filename(info_dict
, 'thumbnail'))
2865 if thumb_files
is None:
2867 files_to_move
.update(dict(thumb_files
))
2869 infofn
= self
.prepare_filename(info_dict
, 'infojson')
2870 _infojson_written
= self
._write
_info
_json
('video', info_dict
, infofn
)
2871 if _infojson_written
:
2872 info_dict
['infojson_filename'] = infofn
2873 # For backward compatibility, even though it was a private field
2874 info_dict
['__infojson_filename'] = infofn
2875 elif _infojson_written
is None:
2878 # Note: Annotations are deprecated
2880 if self
.params
.get('writeannotations', False):
2881 annofn
= self
.prepare_filename(info_dict
, 'annotation')
2883 if not self
._ensure
_dir
_exists
(encodeFilename(annofn
)):
2885 if not self
.params
.get('overwrites', True) and os
.path
.exists(encodeFilename(annofn
)):
2886 self
.to_screen('[info] Video annotations are already present')
2887 elif not info_dict
.get('annotations'):
2888 self
.report_warning('There are no annotations to write.')
2891 self
.to_screen('[info] Writing video annotations to: ' + annofn
)
2892 with io
.open(encodeFilename(annofn
), 'w', encoding
='utf-8') as annofile
:
2893 annofile
.write(info_dict
['annotations'])
2894 except (KeyError, TypeError):
2895 self
.report_warning('There are no annotations to write.')
2896 except (OSError, IOError):
2897 self
.report_error('Cannot write annotations file: ' + annofn
)
2900 # Write internet shortcut files
2901 def _write_link_file(link_type
):
2902 if 'webpage_url' not in info_dict
:
2903 self
.report_error('Cannot write internet shortcut file because the "webpage_url" field is missing in the media information')
2905 linkfn
= replace_extension(self
.prepare_filename(info_dict
, 'link'), link_type
, info_dict
.get('ext'))
2906 if not self
._ensure
_dir
_exists
(encodeFilename(linkfn
)):
2908 if self
.params
.get('overwrites', True) and os
.path
.exists(encodeFilename(linkfn
)):
2909 self
.to_screen(f
'[info] Internet shortcut (.{link_type}) is already present')
2912 self
.to_screen(f
'[info] Writing internet shortcut (.{link_type}) to: {linkfn}')
2913 with io
.open(encodeFilename(to_high_limit_path(linkfn
)), 'w', encoding
='utf-8',
2914 newline
='\r\n' if link_type
== 'url' else '\n') as linkfile
:
2915 template_vars
= {'url': iri_to_uri(info_dict['webpage_url'])}
2916 if link_type
== 'desktop':
2917 template_vars
['filename'] = linkfn
[:-(len(link_type
) + 1)]
2918 linkfile
.write(LINK_TEMPLATES
[link_type
] % template_vars
)
2919 except (OSError, IOError):
2920 self
.report_error(f
'Cannot write internet shortcut {linkfn}')
2925 'url': self
.params
.get('writeurllink'),
2926 'webloc': self
.params
.get('writewebloclink'),
2927 'desktop': self
.params
.get('writedesktoplink'),
2929 if self
.params
.get('writelink'):
2930 link_type
= ('webloc' if sys
.platform
== 'darwin'
2931 else 'desktop' if sys
.platform
.startswith('linux')
2933 write_links
[link_type
] = True
2935 if any(should_write
and not _write_link_file(link_type
)
2936 for link_type
, should_write
in write_links
.items()):
2939 def replace_info_dict(new_info
):
2941 if new_info
== info_dict
:
2944 info_dict
.update(new_info
)
2947 new_info
, files_to_move
= self
.pre_process(info_dict
, 'before_dl', files_to_move
)
2948 replace_info_dict(new_info
)
2949 except PostProcessingError
as err
:
2950 self
.report_error('Preprocessing: %s' % str(err
))
2953 if self
.params
.get('skip_download'):
2954 info_dict
['filepath'] = temp_filename
2955 info_dict
['__finaldir'] = os
.path
.dirname(os
.path
.abspath(encodeFilename(full_filename
)))
2956 info_dict
['__files_to_move'] = files_to_move
2957 replace_info_dict(self
.run_pp(MoveFilesAfterDownloadPP(self
, False), info_dict
))
2958 info_dict
['__write_download_archive'] = self
.params
.get('force_write_download_archive')
2961 info_dict
.setdefault('__postprocessors', [])
2964 def existing_video_file(*filepaths
):
2965 ext
= info_dict
.get('ext')
2966 converted
= lambda file: replace_extension(file, self
.params
.get('final_ext') or ext
, ext
)
2967 file = self
.existing_file(itertools
.chain(*zip(map(converted
, filepaths
), filepaths
)),
2968 default_overwrite
=False)
2970 info_dict
['ext'] = os
.path
.splitext(file)[1][1:]
2974 if info_dict
.get('requested_formats') is not None:
2976 def compatible_formats(formats
):
2977 # TODO: some formats actually allow this (mkv, webm, ogg, mp4), but not all of them.
2978 video_formats
= [format
for format
in formats
if format
.get('vcodec') != 'none']
2979 audio_formats
= [format
for format
in formats
if format
.get('acodec') != 'none']
2980 if len(video_formats
) > 2 or len(audio_formats
) > 2:
2984 exts
= set(format
.get('ext') for format
in formats
)
2986 set(('mp3', 'mp4', 'm4a', 'm4p', 'm4b', 'm4r', 'm4v', 'ismv', 'isma')),
2989 for ext_sets
in COMPATIBLE_EXTS
:
2990 if ext_sets
.issuperset(exts
):
2992 # TODO: Check acodec/vcodec
2995 requested_formats
= info_dict
['requested_formats']
2996 old_ext
= info_dict
['ext']
2997 if self
.params
.get('merge_output_format') is None:
2998 if not compatible_formats(requested_formats
):
2999 info_dict
['ext'] = 'mkv'
3000 self
.report_warning(
3001 'Requested formats are incompatible for merge and will be merged into mkv')
3002 if (info_dict
['ext'] == 'webm'
3003 and info_dict
.get('thumbnails')
3004 # check with type instead of pp_key, __name__, or isinstance
3005 # since we dont want any custom PPs to trigger this
3006 and any(type(pp
) == EmbedThumbnailPP
for pp
in self
._pps
['post_process'])):
3007 info_dict
['ext'] = 'mkv'
3008 self
.report_warning(
3009 'webm doesn\'t support embedding a thumbnail, mkv will be used')
3010 new_ext
= info_dict
['ext']
3012 def correct_ext(filename
, ext
=new_ext
):
3015 filename_real_ext
= os
.path
.splitext(filename
)[1][1:]
3017 os
.path
.splitext(filename
)[0]
3018 if filename_real_ext
in (old_ext
, new_ext
)
3020 return '%s.%s' % (filename_wo_ext
, ext
)
3022 # Ensure filename always has a correct extension for successful merge
3023 full_filename
= correct_ext(full_filename
)
3024 temp_filename
= correct_ext(temp_filename
)
3025 dl_filename
= existing_video_file(full_filename
, temp_filename
)
3026 info_dict
['__real_download'] = False
3029 merger
= FFmpegMergerPP(self
)
3031 fd
= get_suitable_downloader(info_dict
, self
.params
, to_stdout
=temp_filename
== '-')
3032 if dl_filename
is not None:
3033 self
.report_file_already_downloaded(dl_filename
)
3035 for f
in requested_formats
if fd
!= FFmpegFD
else []:
3036 f
['filepath'] = fname
= prepend_extension(
3037 correct_ext(temp_filename
, info_dict
['ext']),
3038 'f%s' % f
['format_id'], info_dict
['ext'])
3039 downloaded
.append(fname
)
3040 info_dict
['url'] = '\n'.join(f
['url'] for f
in requested_formats
)
3041 success
, real_download
= self
.dl(temp_filename
, info_dict
)
3042 info_dict
['__real_download'] = real_download
3044 if self
.params
.get('allow_unplayable_formats'):
3045 self
.report_warning(
3046 'You have requested merging of multiple formats '
3047 'while also allowing unplayable formats to be downloaded. '
3048 'The formats won\'t be merged to prevent data corruption.')
3049 elif not merger
.available
:
3050 self
.report_warning(
3051 'You have requested merging of multiple formats but ffmpeg is not installed. '
3052 'The formats won\'t be merged.')
3054 if temp_filename
== '-':
3055 reason
= ('using a downloader other than ffmpeg' if FFmpegFD
.can_merge_formats(info_dict
, self
.params
)
3056 else 'but the formats are incompatible for simultaneous download' if merger
.available
3057 else 'but ffmpeg is not installed')
3058 self
.report_warning(
3059 f
'You have requested downloading multiple formats to stdout {reason}. '
3060 'The formats will be streamed one after the other')
3061 fname
= temp_filename
3062 for f
in requested_formats
:
3063 new_info
= dict(info_dict
)
3064 del new_info
['requested_formats']
3066 if temp_filename
!= '-':
3067 fname
= prepend_extension(
3068 correct_ext(temp_filename
, new_info
['ext']),
3069 'f%s' % f
['format_id'], new_info
['ext'])
3070 if not self
._ensure
_dir
_exists
(fname
):
3072 f
['filepath'] = fname
3073 downloaded
.append(fname
)
3074 partial_success
, real_download
= self
.dl(fname
, new_info
)
3075 info_dict
['__real_download'] = info_dict
['__real_download'] or real_download
3076 success
= success
and partial_success
3078 if downloaded
and merger
.available
and not self
.params
.get('allow_unplayable_formats'):
3079 info_dict
['__postprocessors'].append(merger
)
3080 info_dict
['__files_to_merge'] = downloaded
3081 # Even if there were no downloads, it is being merged only now
3082 info_dict
['__real_download'] = True
3084 for file in downloaded
:
3085 files_to_move
[file] = None
3087 # Just a single file
3088 dl_filename
= existing_video_file(full_filename
, temp_filename
)
3089 if dl_filename
is None or dl_filename
== temp_filename
:
3090 # dl_filename == temp_filename could mean that the file was partially downloaded with --no-part.
3091 # So we should try to resume the download
3092 success
, real_download
= self
.dl(temp_filename
, info_dict
)
3093 info_dict
['__real_download'] = real_download
3095 self
.report_file_already_downloaded(dl_filename
)
3097 dl_filename
= dl_filename
or temp_filename
3098 info_dict
['__finaldir'] = os
.path
.dirname(os
.path
.abspath(encodeFilename(full_filename
)))
3100 except network_exceptions
as err
:
3101 self
.report_error('unable to download video data: %s' % error_to_compat_str(err
))
3103 except (OSError, IOError) as err
:
3104 raise UnavailableVideoError(err
)
3105 except (ContentTooShortError
, ) as err
:
3106 self
.report_error('content too short (expected %s bytes and served %s)' % (err
.expected
, err
.downloaded
))
3109 if success
and full_filename
!= '-':
3113 fixup_policy
= self
.params
.get('fixup')
3114 vid
= info_dict
['id']
3116 if fixup_policy
in ('ignore', 'never'):
3118 elif fixup_policy
== 'warn':
3120 elif fixup_policy
!= 'force':
3121 assert fixup_policy
in ('detect_or_warn', None)
3122 if not info_dict
.get('__real_download'):
3125 def ffmpeg_fixup(cndn
, msg
, cls
):
3129 self
.report_warning(f
'{vid}: {msg}')
3133 info_dict
['__postprocessors'].append(pp
)
3135 self
.report_warning(f
'{vid}: {msg}. Install ffmpeg to fix this automatically')
3137 stretched_ratio
= info_dict
.get('stretched_ratio')
3139 stretched_ratio
not in (1, None),
3140 f
'Non-uniform pixel ratio {stretched_ratio}',
3141 FFmpegFixupStretchedPP
)
3144 (info_dict
.get('requested_formats') is None
3145 and info_dict
.get('container') == 'm4a_dash'
3146 and info_dict
.get('ext') == 'm4a'),
3147 'writing DASH m4a. Only some players support this container',
3150 downloader
= get_suitable_downloader(info_dict
, self
.params
) if 'protocol' in info_dict
else None
3151 downloader
= downloader
.__name
__ if downloader
else None
3153 if info_dict
.get('requested_formats') is None: # Not necessary if doing merger
3154 ffmpeg_fixup(downloader
== 'HlsFD',
3155 'Possible MPEG-TS in MP4 container or malformed AAC timestamps',
3157 ffmpeg_fixup(info_dict
.get('is_live') and downloader
== 'DashSegmentsFD',
3158 'Possible duplicate MOOV atoms', FFmpegFixupDuplicateMoovPP
)
3160 ffmpeg_fixup(downloader
== 'WebSocketFragmentFD', 'Malformed timestamps detected', FFmpegFixupTimestampPP
)
3161 ffmpeg_fixup(downloader
== 'WebSocketFragmentFD', 'Malformed duration detected', FFmpegFixupDurationPP
)
3165 replace_info_dict(self
.post_process(dl_filename
, info_dict
, files_to_move
))
3166 except PostProcessingError
as err
:
3167 self
.report_error('Postprocessing: %s' % str(err
))
3170 for ph
in self
._post
_hooks
:
3171 ph(info_dict
['filepath'])
3172 except Exception as err
:
3173 self
.report_error('post hooks: %s' % str(err
))
3175 info_dict
['__write_download_archive'] = True
3177 if self
.params
.get('force_write_download_archive'):
3178 info_dict
['__write_download_archive'] = True
3180 # Make sure the info_dict was modified in-place
3181 assert info_dict
is original_infodict
3183 max_downloads
= self
.params
.get('max_downloads')
3184 if max_downloads
is not None and self
._num
_downloads
>= int(max_downloads
):
3185 raise MaxDownloadsReached()
3187 def __download_wrapper(self
, func
):
3188 @functools.wraps(func
)
3189 def wrapper(*args
, **kwargs
):
3191 res
= func(*args
, **kwargs
)
3192 except UnavailableVideoError
as e
:
3193 self
.report_error(e
)
3194 except MaxDownloadsReached
as e
:
3195 self
.to_screen(f
'[info] {e}')
3197 except DownloadCancelled
as e
:
3198 self
.to_screen(f
'[info] {e}')
3199 if not self
.params
.get('break_per_url'):
3202 if self
.params
.get('dump_single_json', False):
3203 self
.post_extract(res
)
3204 self
.to_stdout(json
.dumps(self
.sanitize_info(res
)))
3207 def download(self
, url_list
):
3208 """Download a given list of URLs."""
3209 url_list
= variadic(url_list
) # Passing a single URL is a common mistake
3210 outtmpl
= self
.outtmpl_dict
['default']
3211 if (len(url_list
) > 1
3213 and '%' not in outtmpl
3214 and self
.params
.get('max_downloads') != 1):
3215 raise SameFileError(outtmpl
)
3217 for url
in url_list
:
3218 self
.__download
_wrapper
(self
.extract_info
)(
3219 url
, force_generic_extractor
=self
.params
.get('force_generic_extractor', False))
3221 return self
._download
_retcode
3223 def download_with_info_file(self
, info_filename
):
3224 with contextlib
.closing(fileinput
.FileInput(
3225 [info_filename
], mode
='r',
3226 openhook
=fileinput
.hook_encoded('utf-8'))) as f
:
3227 # FileInput doesn't have a read method, we can't call json.load
3228 info
= self
.sanitize_info(json
.loads('\n'.join(f
)), self
.params
.get('clean_infojson', True))
3230 self
.__download
_wrapper
(self
.process_ie_result
)(info
, download
=True)
3231 except (DownloadError
, EntryNotInPlaylist
, ReExtractInfo
) as e
:
3232 if not isinstance(e
, EntryNotInPlaylist
):
3233 self
.to_stderr('\r')
3234 webpage_url
= info
.get('webpage_url')
3235 if webpage_url
is not None:
3236 self
.report_warning(f
'The info failed to download: {e}; trying with URL {webpage_url}')
3237 return self
.download([webpage_url
])
3240 return self
._download
_retcode
3243 def sanitize_info(info_dict
, remove_private_keys
=False):
3244 ''' Sanitize the infodict for converting to json '''
3245 if info_dict
is None:
3247 info_dict
.setdefault('epoch', int(time
.time()))
3248 info_dict
.setdefault('_type', 'video')
3249 remove_keys
= {'__original_infodict'}
# Always remove this since this may contain a copy of the entire dict
3250 keep_keys
= ['_type'] # Always keep this to facilitate load-info-json
3251 if remove_private_keys
:
3253 'requested_downloads', 'requested_formats', 'requested_subtitles', 'requested_entries',
3254 'entries', 'filepath', 'infojson_filename', 'original_url', 'playlist_autonumber',
3256 reject
= lambda k
, v
: k
not in keep_keys
and (
3257 k
.startswith('_') or k
in remove_keys
or v
is None)
3259 reject
= lambda k
, v
: k
in remove_keys
3262 if isinstance(obj
, dict):
3263 return {k: filter_fn(v) for k, v in obj.items() if not reject(k, v)}
3264 elif isinstance(obj
, (list, tuple, set, LazyList
)):
3265 return list(map(filter_fn
, obj
))
3266 elif obj
is None or isinstance(obj
, (str, int, float, bool)):
3271 return filter_fn(info_dict
)
3274 def filter_requested_info(info_dict
, actually_filter
=True):
3275 ''' Alias of sanitize_info for backward compatibility '''
3276 return YoutubeDL
.sanitize_info(info_dict
, actually_filter
)
3279 def post_extract(info_dict
):
3280 def actual_post_extract(info_dict
):
3281 if info_dict
.get('_type') in ('playlist', 'multi_video'):
3282 for video_dict
in info_dict
.get('entries', {}):
3283 actual_post_extract(video_dict
or {})
3286 post_extractor
= info_dict
.get('__post_extractor') or (lambda: {})
3287 extra
= post_extractor().items()
3288 info_dict
.update(extra
)
3289 info_dict
.pop('__post_extractor', None)
3291 original_infodict
= info_dict
.get('__original_infodict') or {}
3292 original_infodict
.update(extra
)
3293 original_infodict
.pop('__post_extractor', None)
3295 actual_post_extract(info_dict
or {})
3297 def run_pp(self
, pp
, infodict
):
3298 files_to_delete
= []
3299 if '__files_to_move' not in infodict
:
3300 infodict
['__files_to_move'] = {}
3302 files_to_delete
, infodict
= pp
.run(infodict
)
3303 except PostProcessingError
as e
:
3304 # Must be True and not 'only_download'
3305 if self
.params
.get('ignoreerrors') is True:
3306 self
.report_error(e
)
3310 if not files_to_delete
:
3312 if self
.params
.get('keepvideo', False):
3313 for f
in files_to_delete
:
3314 infodict
['__files_to_move'].setdefault(f
, '')
3316 for old_filename
in set(files_to_delete
):
3317 self
.to_screen('Deleting original file %s (pass -k to keep)' % old_filename
)
3319 os
.remove(encodeFilename(old_filename
))
3320 except (IOError, OSError):
3321 self
.report_warning('Unable to remove downloaded original file')
3322 if old_filename
in infodict
['__files_to_move']:
3323 del infodict
['__files_to_move'][old_filename
]
3326 def run_all_pps(self
, key
, info
, *, additional_pps
=None):
3327 self
._forceprint
(key
, info
)
3328 for pp
in (additional_pps
or []) + self
._pps
[key
]:
3329 info
= self
.run_pp(pp
, info
)
3332 def pre_process(self
, ie_info
, key
='pre_process', files_to_move
=None):
3333 info
= dict(ie_info
)
3334 info
['__files_to_move'] = files_to_move
or {}
3335 info
= self
.run_all_pps(key
, info
)
3336 return info
, info
.pop('__files_to_move', None)
3338 def post_process(self
, filename
, info
, files_to_move
=None):
3339 """Run all the postprocessors on the given file."""
3340 info
['filepath'] = filename
3341 info
['__files_to_move'] = files_to_move
or {}
3342 info
= self
.run_all_pps('post_process', info
, additional_pps
=info
.get('__postprocessors'))
3343 info
= self
.run_pp(MoveFilesAfterDownloadPP(self
), info
)
3344 del info
['__files_to_move']
3345 return self
.run_all_pps('after_move', info
)
3347 def _make_archive_id(self
, info_dict
):
3348 video_id
= info_dict
.get('id')
3351 # Future-proof against any change in case
3352 # and backwards compatibility with prior versions
3353 extractor
= info_dict
.get('extractor_key') or info_dict
.get('ie_key') # key in a playlist
3354 if extractor
is None:
3355 url
= str_or_none(info_dict
.get('url'))
3358 # Try to find matching extractor for the URL and take its ie_key
3359 for ie_key
, ie
in self
._ies
.items():
3360 if ie
.suitable(url
):
3365 return '%s %s' % (extractor
.lower(), video_id
)
3367 def in_download_archive(self
, info_dict
):
3368 fn
= self
.params
.get('download_archive')
3372 vid_id
= self
._make
_archive
_id
(info_dict
)
3374 return False # Incomplete video information
3376 return vid_id
in self
.archive
3378 def record_download_archive(self
, info_dict
):
3379 fn
= self
.params
.get('download_archive')
3382 vid_id
= self
._make
_archive
_id
(info_dict
)
3384 self
.write_debug(f
'Adding to archive: {vid_id}')
3385 with locked_file(fn
, 'a', encoding
='utf-8') as archive_file
:
3386 archive_file
.write(vid_id
+ '\n')
3387 self
.archive
.add(vid_id
)
3390 def format_resolution(format
, default
='unknown'):
3391 if format
.get('vcodec') == 'none' and format
.get('acodec') != 'none':
3393 if format
.get('resolution') is not None:
3394 return format
['resolution']
3395 if format
.get('width') and format
.get('height'):
3396 return '%dx%d' % (format
['width'], format
['height'])
3397 elif format
.get('height'):
3398 return '%sp' % format
['height']
3399 elif format
.get('width'):
3400 return '%dx?' % format
['width']
3403 def _list_format_headers(self
, *headers
):
3404 if self
.params
.get('listformats_table', True) is not False:
3405 return [self
._format
_screen
(header
, self
.Styles
.HEADERS
) for header
in headers
]
3408 def _format_note(self
, fdict
):
3410 if fdict
.get('ext') in ['f4f', 'f4m']:
3411 res
+= '(unsupported)'
3412 if fdict
.get('language'):
3415 res
+= '[%s]' % fdict
['language']
3416 if fdict
.get('format_note') is not None:
3419 res
+= fdict
['format_note']
3420 if fdict
.get('tbr') is not None:
3423 res
+= '%4dk' % fdict
['tbr']
3424 if fdict
.get('container') is not None:
3427 res
+= '%s container' % fdict
['container']
3428 if (fdict
.get('vcodec') is not None
3429 and fdict
.get('vcodec') != 'none'):
3432 res
+= fdict
['vcodec']
3433 if fdict
.get('vbr') is not None:
3435 elif fdict
.get('vbr') is not None and fdict
.get('abr') is not None:
3437 if fdict
.get('vbr') is not None:
3438 res
+= '%4dk' % fdict
['vbr']
3439 if fdict
.get('fps') is not None:
3442 res
+= '%sfps' % fdict
['fps']
3443 if fdict
.get('acodec') is not None:
3446 if fdict
['acodec'] == 'none':
3449 res
+= '%-5s' % fdict
['acodec']
3450 elif fdict
.get('abr') is not None:
3454 if fdict
.get('abr') is not None:
3455 res
+= '@%3dk' % fdict
['abr']
3456 if fdict
.get('asr') is not None:
3457 res
+= ' (%5dHz)' % fdict
['asr']
3458 if fdict
.get('filesize') is not None:
3461 res
+= format_bytes(fdict
['filesize'])
3462 elif fdict
.get('filesize_approx') is not None:
3465 res
+= '~' + format_bytes(fdict
['filesize_approx'])
3468 def render_formats_table(self
, info_dict
):
3469 if not info_dict
.get('formats') and not info_dict
.get('url'):
3472 formats
= info_dict
.get('formats', [info_dict
])
3473 if not self
.params
.get('listformats_table', True) is not False:
3476 format_field(f
, 'format_id'),
3477 format_field(f
, 'ext'),
3478 self
.format_resolution(f
),
3479 self
._format
_note
(f
)
3480 ] for f
in formats
if f
.get('preference') is None or f
['preference'] >= -1000]
3481 return render_table(['format code', 'extension', 'resolution', 'note'], table
, extra_gap
=1)
3483 delim
= self
._format
_screen
('\u2502', self
.Styles
.DELIM
, '|', test_encoding
=True)
3486 self
._format
_screen
(format_field(f
, 'format_id'), self
.Styles
.ID
),
3487 format_field(f
, 'ext'),
3488 format_field(f
, func
=self
.format_resolution
, ignore
=('audio only', 'images')),
3489 format_field(f
, 'fps', '\t%d'),
3490 format_field(f
, 'dynamic_range', '%s', ignore
=(None, 'SDR')).replace('HDR', ''),
3492 format_field(f
, 'filesize', ' \t%s', func
=format_bytes
) + format_field(f
, 'filesize_approx', '~\t%s', func
=format_bytes
),
3493 format_field(f
, 'tbr', '\t%dk'),
3494 shorten_protocol_name(f
.get('protocol', '')),
3496 format_field(f
, 'vcodec', default
='unknown').replace(
3497 'none', 'images' if f
.get('acodec') == 'none'
3498 else self
._format
_screen
('audio only', self
.Styles
.SUPPRESS
)),
3499 format_field(f
, 'vbr', '\t%dk'),
3500 format_field(f
, 'acodec', default
='unknown').replace(
3501 'none', '' if f
.get('vcodec') == 'none'
3502 else self
._format
_screen
('video only', self
.Styles
.SUPPRESS
)),
3503 format_field(f
, 'abr', '\t%dk'),
3504 format_field(f
, 'asr', '\t%dHz'),
3506 self
._format
_screen
('UNSUPPORTED', 'light red') if f
.get('ext') in ('f4f', 'f4m') else None,
3507 format_field(f
, 'language', '[%s]'),
3508 join_nonempty(format_field(f
, 'format_note'),
3509 format_field(f
, 'container', ignore
=(None, f
.get('ext'))),
3512 ] for f
in formats
if f
.get('preference') is None or f
['preference'] >= -1000]
3513 header_line
= self
._list
_format
_headers
(
3514 'ID', 'EXT', 'RESOLUTION', '\tFPS', 'HDR', delim
, '\tFILESIZE', '\tTBR', 'PROTO',
3515 delim
, 'VCODEC', '\tVBR', 'ACODEC', '\tABR', '\tASR', 'MORE INFO')
3517 return render_table(
3518 header_line
, table
, hide_empty
=True,
3519 delim
=self
._format
_screen
('\u2500', self
.Styles
.DELIM
, '-', test_encoding
=True))
3521 def render_thumbnails_table(self
, info_dict
):
3522 thumbnails
= list(info_dict
.get('thumbnails') or [])
3525 return render_table(
3526 self
._list
_format
_headers
('ID', 'Width', 'Height', 'URL'),
3527 [[t
.get('id'), t
.get('width', 'unknown'), t
.get('height', 'unknown'), t
['url']] for t
in thumbnails
])
3529 def render_subtitles_table(self
, video_id
, subtitles
):
3530 def _row(lang
, formats
):
3531 exts
, names
= zip(*((f
['ext'], f
.get('name') or 'unknown') for f
in reversed(formats
)))
3532 if len(set(names
)) == 1:
3533 names
= [] if names
[0] == 'unknown' else names
[:1]
3534 return [lang
, ', '.join(names
), ', '.join(exts
)]
3538 return render_table(
3539 self
._list
_format
_headers
('Language', 'Name', 'Formats'),
3540 [_row(lang
, formats
) for lang
, formats
in subtitles
.items()],
3543 def __list_table(self
, video_id
, name
, func
, *args
):
3546 self
.to_screen(f
'{video_id} has no {name}')
3548 self
.to_screen(f
'[info] Available {name} for {video_id}:')
3549 self
.to_stdout(table
)
3551 def list_formats(self
, info_dict
):
3552 self
.__list
_table
(info_dict
['id'], 'formats', self
.render_formats_table
, info_dict
)
3554 def list_thumbnails(self
, info_dict
):
3555 self
.__list
_table
(info_dict
['id'], 'thumbnails', self
.render_thumbnails_table
, info_dict
)
3557 def list_subtitles(self
, video_id
, subtitles
, name
='subtitles'):
3558 self
.__list
_table
(video_id
, name
, self
.render_subtitles_table
, video_id
, subtitles
)
3560 def urlopen(self
, req
):
3561 """ Start an HTTP download """
3562 if isinstance(req
, compat_basestring
):
3563 req
= sanitized_Request(req
)
3564 return self
._opener
.open(req
, timeout
=self
._socket
_timeout
)
3566 def print_debug_header(self
):
3567 if not self
.params
.get('verbose'):
3570 def get_encoding(stream
):
3571 ret
= getattr(stream
, 'encoding', 'missing (%s)' % type(stream
).__name
__)
3572 if not supports_terminal_sequences(stream
):
3573 from .compat
import WINDOWS_VT_MODE
3574 ret
+= ' (No VT)' if WINDOWS_VT_MODE
is False else ' (No ANSI)'
3577 encoding_str
= 'Encodings: locale %s, fs %s, out %s, err %s, pref %s' % (
3578 locale
.getpreferredencoding(),
3579 sys
.getfilesystemencoding(),
3580 get_encoding(self
._screen
_file
), get_encoding(self
._err
_file
),
3581 self
.get_encoding())
3583 logger
= self
.params
.get('logger')
3585 write_debug
= lambda msg
: logger
.debug(f
'[debug] {msg}')
3586 write_debug(encoding_str
)
3588 write_string(f
'[debug] {encoding_str}\n', encoding
=None)
3589 write_debug
= lambda msg
: self
._write
_string
(f
'[debug] {msg}\n')
3591 source
= detect_variant()
3592 write_debug(join_nonempty(
3593 'yt-dlp version', __version__
,
3594 f
'[{RELEASE_GIT_HEAD}]' if RELEASE_GIT_HEAD
else '',
3595 '' if source
== 'unknown' else f
'({source})',
3597 if not _LAZY_LOADER
:
3598 if os
.environ
.get('YTDLP_NO_LAZY_EXTRACTORS'):
3599 write_debug('Lazy loading extractors is forcibly disabled')
3601 write_debug('Lazy loading extractors is disabled')
3602 if plugin_extractors
or plugin_postprocessors
:
3603 write_debug('Plugins: %s' % [
3604 '%s%s' % (klass
.__name
__, '' if klass
.__name
__ == name
else f
' as {name}')
3605 for name
, klass
in itertools
.chain(plugin_extractors
.items(), plugin_postprocessors
.items())])
3606 if self
.params
.get('compat_opts'):
3607 write_debug('Compatibility options: %s' % ', '.join(self
.params
.get('compat_opts')))
3609 if source
== 'source':
3612 ['git', 'rev-parse', '--short', 'HEAD'],
3613 stdout
=subprocess
.PIPE
, stderr
=subprocess
.PIPE
,
3614 cwd
=os
.path
.dirname(os
.path
.abspath(__file__
)))
3615 out
, err
= sp
.communicate_or_kill()
3616 out
= out
.decode().strip()
3617 if re
.match('[0-9a-f]+', out
):
3618 write_debug('Git HEAD: %s' % out
)
3625 def python_implementation():
3626 impl_name
= platform
.python_implementation()
3627 if impl_name
== 'PyPy' and hasattr(sys
, 'pypy_version_info'):
3628 return impl_name
+ ' version %d.%d.%d' % sys
.pypy_version_info
[:3]
3631 write_debug('Python version %s (%s %s) - %s' % (
3632 platform
.python_version(),
3633 python_implementation(),
3634 platform
.architecture()[0],
3637 exe_versions
, ffmpeg_features
= FFmpegPostProcessor
.get_versions_and_features(self
)
3638 ffmpeg_features
= {key for key, val in ffmpeg_features.items() if val}
3640 exe_versions
['ffmpeg'] += ' (%s)' % ','.join(ffmpeg_features
)
3642 exe_versions
['rtmpdump'] = rtmpdump_version()
3643 exe_versions
['phantomjs'] = PhantomJSwrapper
._version
()
3644 exe_str
= ', '.join(
3645 f
'{exe} {v}' for exe
, v
in sorted(exe_versions
.items()) if v
3647 write_debug('exe versions: %s' % exe_str
)
3649 from .downloader
.websocket
import has_websockets
3650 from .postprocessor
.embedthumbnail
import has_mutagen
3651 from .cookies
import SQLITE_AVAILABLE
, SECRETSTORAGE_AVAILABLE
3653 lib_str
= join_nonempty(
3654 compat_pycrypto_AES
and compat_pycrypto_AES
.__name
__.split('.')[0],
3655 SECRETSTORAGE_AVAILABLE
and 'secretstorage',
3656 has_mutagen
and 'mutagen',
3657 SQLITE_AVAILABLE
and 'sqlite',
3658 has_websockets
and 'websockets',
3659 delim
=', ') or 'none'
3660 write_debug('Optional libraries: %s' % lib_str
)
3663 for handler
in self
._opener
.handlers
:
3664 if hasattr(handler
, 'proxies'):
3665 proxy_map
.update(handler
.proxies
)
3666 write_debug(f
'Proxy map: {proxy_map}')
3669 if False and self
.params
.get('call_home'):
3670 ipaddr
= self
.urlopen('https://yt-dl.org/ip').read().decode('utf-8')
3671 write_debug('Public IP address: %s' % ipaddr
)
3672 latest_version
= self
.urlopen(
3673 'https://yt-dl.org/latest/version').read().decode('utf-8')
3674 if version_tuple(latest_version
) > version_tuple(__version__
):
3675 self
.report_warning(
3676 'You are using an outdated version (newest version: %s)! '
3677 'See https://yt-dl.org/update if you need help updating.' %
3680 def _setup_opener(self
):
3681 timeout_val
= self
.params
.get('socket_timeout')
3682 self
._socket
_timeout
= 20 if timeout_val
is None else float(timeout_val
)
3684 opts_cookiesfrombrowser
= self
.params
.get('cookiesfrombrowser')
3685 opts_cookiefile
= self
.params
.get('cookiefile')
3686 opts_proxy
= self
.params
.get('proxy')
3688 self
.cookiejar
= load_cookies(opts_cookiefile
, opts_cookiesfrombrowser
, self
)
3690 cookie_processor
= YoutubeDLCookieProcessor(self
.cookiejar
)
3691 if opts_proxy
is not None:
3692 if opts_proxy
== '':
3695 proxies
= {'http': opts_proxy, 'https': opts_proxy}
3697 proxies
= compat_urllib_request
.getproxies()
3698 # Set HTTPS proxy to HTTP one if given (https://github.com/ytdl-org/youtube-dl/issues/805)
3699 if 'http' in proxies
and 'https' not in proxies
:
3700 proxies
['https'] = proxies
['http']
3701 proxy_handler
= PerRequestProxyHandler(proxies
)
3703 debuglevel
= 1 if self
.params
.get('debug_printtraffic') else 0
3704 https_handler
= make_HTTPS_handler(self
.params
, debuglevel
=debuglevel
)
3705 ydlh
= YoutubeDLHandler(self
.params
, debuglevel
=debuglevel
)
3706 redirect_handler
= YoutubeDLRedirectHandler()
3707 data_handler
= compat_urllib_request_DataHandler()
3709 # When passing our own FileHandler instance, build_opener won't add the
3710 # default FileHandler and allows us to disable the file protocol, which
3711 # can be used for malicious purposes (see
3712 # https://github.com/ytdl-org/youtube-dl/issues/8227)
3713 file_handler
= compat_urllib_request
.FileHandler()
3715 def file_open(*args
, **kwargs
):
3716 raise compat_urllib_error
.URLError('file:// scheme is explicitly disabled in yt-dlp for security reasons')
3717 file_handler
.file_open
= file_open
3719 opener
= compat_urllib_request
.build_opener(
3720 proxy_handler
, https_handler
, cookie_processor
, ydlh
, redirect_handler
, data_handler
, file_handler
)
3722 # Delete the default user-agent header, which would otherwise apply in
3723 # cases where our custom HTTP handler doesn't come into play
3724 # (See https://github.com/ytdl-org/youtube-dl/issues/1309 for details)
3725 opener
.addheaders
= []
3726 self
._opener
= opener
3728 def encode(self
, s
):
3729 if isinstance(s
, bytes):
3730 return s
# Already encoded
3733 return s
.encode(self
.get_encoding())
3734 except UnicodeEncodeError as err
:
3735 err
.reason
= err
.reason
+ '. Check your system encoding configuration or use the --encoding option.'
3738 def get_encoding(self
):
3739 encoding
= self
.params
.get('encoding')
3740 if encoding
is None:
3741 encoding
= preferredencoding()
3744 def _write_info_json(self
, label
, ie_result
, infofn
, overwrite
=None):
3745 ''' Write infojson and returns True = written, False = skip, None = error '''
3746 if overwrite
is None:
3747 overwrite
= self
.params
.get('overwrites', True)
3748 if not self
.params
.get('writeinfojson'):
3751 self
.write_debug(f
'Skipping writing {label} infojson')
3753 elif not self
._ensure
_dir
_exists
(infofn
):
3755 elif not overwrite
and os
.path
.exists(infofn
):
3756 self
.to_screen(f
'[info] {label.title()} metadata is already present')
3758 self
.to_screen(f
'[info] Writing {label} metadata as JSON to: {infofn}')
3760 write_json_file(self
.sanitize_info(ie_result
, self
.params
.get('clean_infojson', True)), infofn
)
3761 except (OSError, IOError):
3762 self
.report_error(f
'Cannot write {label} metadata to JSON file {infofn}')
3766 def _write_description(self
, label
, ie_result
, descfn
):
3767 ''' Write description and returns True = written, False = skip, None = error '''
3768 if not self
.params
.get('writedescription'):
3771 self
.write_debug(f
'Skipping writing {label} description')
3773 elif not self
._ensure
_dir
_exists
(descfn
):
3775 elif not self
.params
.get('overwrites', True) and os
.path
.exists(descfn
):
3776 self
.to_screen(f
'[info] {label.title()} description is already present')
3777 elif ie_result
.get('description') is None:
3778 self
.report_warning(f
'There\'s no {label} description to write')
3782 self
.to_screen(f
'[info] Writing {label} description to: {descfn}')
3783 with io
.open(encodeFilename(descfn
), 'w', encoding
='utf-8') as descfile
:
3784 descfile
.write(ie_result
['description'])
3785 except (OSError, IOError):
3786 self
.report_error(f
'Cannot write {label} description file {descfn}')
3790 def _write_subtitles(self
, info_dict
, filename
):
3791 ''' Write subtitles to file and return list of (sub_filename, final_sub_filename); or None if error'''
3793 subtitles
= info_dict
.get('requested_subtitles')
3794 if not subtitles
or not (self
.params
.get('writesubtitles') or self
.params
.get('writeautomaticsub')):
3795 # subtitles download errors are already managed as troubles in relevant IE
3796 # that way it will silently go on when used with unsupporting IE
3799 sub_filename_base
= self
.prepare_filename(info_dict
, 'subtitle')
3800 if not sub_filename_base
:
3801 self
.to_screen('[info] Skipping writing video subtitles')
3803 for sub_lang
, sub_info
in subtitles
.items():
3804 sub_format
= sub_info
['ext']
3805 sub_filename
= subtitles_filename(filename
, sub_lang
, sub_format
, info_dict
.get('ext'))
3806 sub_filename_final
= subtitles_filename(sub_filename_base
, sub_lang
, sub_format
, info_dict
.get('ext'))
3807 existing_sub
= self
.existing_file((sub_filename_final
, sub_filename
))
3809 self
.to_screen(f
'[info] Video subtitle {sub_lang}.{sub_format} is already present')
3810 sub_info
['filepath'] = existing_sub
3811 ret
.append((existing_sub
, sub_filename_final
))
3814 self
.to_screen(f
'[info] Writing video subtitles to: {sub_filename}')
3815 if sub_info
.get('data') is not None:
3817 # Use newline='' to prevent conversion of newline characters
3818 # See https://github.com/ytdl-org/youtube-dl/issues/10268
3819 with io
.open(sub_filename
, 'w', encoding
='utf-8', newline
='') as subfile
:
3820 subfile
.write(sub_info
['data'])
3821 sub_info
['filepath'] = sub_filename
3822 ret
.append((sub_filename
, sub_filename_final
))
3824 except (OSError, IOError):
3825 self
.report_error(f
'Cannot write video subtitles file {sub_filename}')
3829 sub_copy
= sub_info
.copy()
3830 sub_copy
.setdefault('http_headers', info_dict
.get('http_headers'))
3831 self
.dl(sub_filename
, sub_copy
, subtitle
=True)
3832 sub_info
['filepath'] = sub_filename
3833 ret
.append((sub_filename
, sub_filename_final
))
3834 except (DownloadError
, ExtractorError
, IOError, OSError, ValueError) + network_exceptions
as err
:
3835 if self
.params
.get('ignoreerrors') is not True: # False or 'only_download'
3836 raise DownloadError(f
'Unable to download video subtitles for {sub_lang!r}: {err}', err
)
3837 self
.report_warning(f
'Unable to download video subtitles for {sub_lang!r}: {err}')
3840 def _write_thumbnails(self
, label
, info_dict
, filename
, thumb_filename_base
=None):
3841 ''' Write thumbnails to file and return list of (thumb_filename, final_thumb_filename) '''
3842 write_all
= self
.params
.get('write_all_thumbnails', False)
3843 thumbnails
, ret
= [], []
3844 if write_all
or self
.params
.get('writethumbnail', False):
3845 thumbnails
= info_dict
.get('thumbnails') or []
3846 multiple
= write_all
and len(thumbnails
) > 1
3848 if thumb_filename_base
is None:
3849 thumb_filename_base
= filename
3850 if thumbnails
and not thumb_filename_base
:
3851 self
.write_debug(f
'Skipping writing {label} thumbnail')
3854 for idx
, t
in list(enumerate(thumbnails
))[::-1]:
3855 thumb_ext
= (f
'{t["id"]}.' if multiple
else '') + determine_ext(t
['url'], 'jpg')
3856 thumb_display_id
= f
'{label} thumbnail {t["id"]}'
3857 thumb_filename
= replace_extension(filename
, thumb_ext
, info_dict
.get('ext'))
3858 thumb_filename_final
= replace_extension(thumb_filename_base
, thumb_ext
, info_dict
.get('ext'))
3860 existing_thumb
= self
.existing_file((thumb_filename_final
, thumb_filename
))
3862 self
.to_screen('[info] %s is already present' % (
3863 thumb_display_id
if multiple
else f
'{label} thumbnail').capitalize())
3864 t
['filepath'] = existing_thumb
3865 ret
.append((existing_thumb
, thumb_filename_final
))
3867 self
.to_screen(f
'[info] Downloading {thumb_display_id} ...')
3869 uf
= self
.urlopen(sanitized_Request(t
['url'], headers
=t
.get('http_headers', {})))
3870 self
.to_screen(f
'[info] Writing {thumb_display_id} to: {thumb_filename}')
3871 with open(encodeFilename(thumb_filename
), 'wb') as thumbf
:
3872 shutil
.copyfileobj(uf
, thumbf
)
3873 ret
.append((thumb_filename
, thumb_filename_final
))
3874 t
['filepath'] = thumb_filename
3875 except network_exceptions
as err
:
3877 self
.report_warning(f
'Unable to download {thumb_display_id}: {err}')
3878 if ret
and not write_all
: